Flight_Price_Model.ipynb

{
  "nbformat": 4,
  "nbformat_minor": 0,
  "metadata": {
    "colab": {
      "name": "Flight Price Model",
      "provenance": [],
      "collapsed_sections": [],
      "machine_shape": "hm",
      "include_colab_link": true
    },
    "kernelspec": {
      "name": "python3",
      "display_name": "Python 3"
    },
    "accelerator": "TPU"
  },
  "cells": [
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "view-in-github",
        "colab_type": "text"
      },
      "source": [
        "<a href=\"https://colab.research.google.com/github/ramanshsharma2806/Flight-Price/blob/master/Flight_Price_Model.ipynb\" target=\"_parent\"><img src=\"https://colab.research.google.com/assets/colab-badge.svg\" alt=\"Open In Colab\"/></a>"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "8sP5Lthf74U1",
        "colab_type": "text"
      },
      "source": [
        "\n",
        "# Flight Price\n",
        "\n",
        "---\n",
        "\n",
        "This model will predict flight prices based on [this](https://www.kaggle.com/zernach/2018-airplane-flights) data.\n",
        "\n",
        "This model is being made for [mia](http://miamarketplace.com/), a place for people to run machine learning models interactively on the web."
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "v4VLed2kKInb",
        "colab_type": "code",
        "cellView": "form",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 34
        },
        "outputId": "2ed7bf1d-1bc8-4308-a2fc-82293ddc3340"
      },
      "source": [
        "#@title Google Drive mounting\n",
        "\"\"\"\n",
        "need to mount the drive to access the data\n",
        "\"\"\"\n",
        "\n",
        "from google.colab import drive\n",
        "drive.mount('/content/drive')"
      ],
      "execution_count": 1,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Mounted at /content/drive\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "4Ij-gFA8C4H_",
        "colab_type": "code",
        "cellView": "form",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 326
        },
        "outputId": "611fd13f-1c5e-4ade-83b3-8bd95aa7deca"
      },
      "source": [
        "#@title Require libraries import cell\n",
        "\"\"\"\n",
        "required libraries imported\n",
        "\"\"\"\n",
        "\n",
        "! pip install category_encoders\n",
        "import category_encoders as ce\n",
        "\n",
        "from __future__ import print_function\n",
        "import os, io, sys, random, time, pprint\n",
        "import numpy as np\n",
        "from numpy import save, load\n",
        "import pandas as pd\n",
        "import matplotlib.pyplot as plt\n",
        "import tensorflow as tf\n",
        "from tensorflow import one_hot\n",
        "from tensorflow.keras.callbacks import LambdaCallback, Callback, ModelCheckpoint\n",
        "from tensorflow.keras.optimizers.schedules import ExponentialDecay\n",
        "from tensorflow.keras.models import load_model\n",
        "from tensorflow.keras import Model, Sequential\n",
        "from tensorflow.keras import backend\n",
        "from tensorflow.keras.layers import Activation, Reshape, Dense, Embedding, Dropout, Input, BatchNormalization, concatenate, Flatten, GlobalAveragePooling1D\n",
        "from tensorflow.keras.preprocessing.sequence import pad_sequences\n",
        "from tensorflow.keras.optimizers import Adam, RMSprop, SGD, Adadelta, Adagrad, Adamax\n",
        "from tensorflow.keras.losses import CategoricalCrossentropy\n",
        "from tensorflow.keras.utils import to_categorical, plot_model\n",
        "from sklearn.model_selection import train_test_split\n",
        "from sklearn.preprocessing import LabelEncoder\n",
        "\n",
        "# SOME CONSTANTS\n",
        "checkpoint_dir = '/content/drive/My Drive/MIA/best_model.hdf5'\n",
        "model_preds_train = '/content/drive/My Drive/MIA/train_preds.npy'\n",
        "model_preds_test = '/content/drive/My Drive/MIA/test_preds.npy'\n",
        "truth_train = '/content/drive/My Drive/MIA/truth_train.npy'\n",
        "truth_test = '/content/drive/My Drive/MIA/truth_test.npy'\n",
        "train_continuous_dir = '/content/drive/My Drive/MIA/train_continuous_dir.npy'\n",
        "test_continuous_dir = '/content/drive/My Drive/MIA/test_continuous_dir.npy'\n",
        "train_categorical_dir = '/content/drive/My Drive/MIA/train_categorical_dir.npy'\n",
        "test_categorical_dir = '/content/drive/My Drive/MIA/test_categorical_dir.npy'"
      ],
      "execution_count": 2,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Collecting category_encoders\n",
            "\u001b[?25l  Downloading https://files.pythonhosted.org/packages/44/57/fcef41c248701ee62e8325026b90c432adea35555cbc870aff9cfba23727/category_encoders-2.2.2-py2.py3-none-any.whl (80kB)\n",
            "\r\u001b[K     |████                            | 10kB 19.6MB/s eta 0:00:01\r\u001b[K     |████████▏                       | 20kB 1.9MB/s eta 0:00:01\r\u001b[K     |████████████▏                   | 30kB 2.2MB/s eta 0:00:01\r\u001b[K     |████████████████▎               | 40kB 2.6MB/s eta 0:00:01\r\u001b[K     |████████████████████▎           | 51kB 2.5MB/s eta 0:00:01\r\u001b[K     |████████████████████████▍       | 61kB 2.7MB/s eta 0:00:01\r\u001b[K     |████████████████████████████▍   | 71kB 3.0MB/s eta 0:00:01\r\u001b[K     |████████████████████████████████| 81kB 2.5MB/s \n",
            "\u001b[?25hRequirement already satisfied: scikit-learn>=0.20.0 in /usr/local/lib/python3.6/dist-packages (from category_encoders) (0.22.2.post1)\n",
            "Requirement already satisfied: pandas>=0.21.1 in /usr/local/lib/python3.6/dist-packages (from category_encoders) (1.0.4)\n",
            "Requirement already satisfied: scipy>=1.0.0 in /usr/local/lib/python3.6/dist-packages (from category_encoders) (1.4.1)\n",
            "Requirement already satisfied: patsy>=0.5.1 in /usr/local/lib/python3.6/dist-packages (from category_encoders) (0.5.1)\n",
            "Requirement already satisfied: statsmodels>=0.9.0 in /usr/local/lib/python3.6/dist-packages (from category_encoders) (0.10.2)\n",
            "Requirement already satisfied: numpy>=1.14.0 in /usr/local/lib/python3.6/dist-packages (from category_encoders) (1.18.5)\n",
            "Requirement already satisfied: joblib>=0.11 in /usr/local/lib/python3.6/dist-packages (from scikit-learn>=0.20.0->category_encoders) (0.15.1)\n",
            "Requirement already satisfied: python-dateutil>=2.6.1 in /usr/local/lib/python3.6/dist-packages (from pandas>=0.21.1->category_encoders) (2.8.1)\n",
            "Requirement already satisfied: pytz>=2017.2 in /usr/local/lib/python3.6/dist-packages (from pandas>=0.21.1->category_encoders) (2018.9)\n",
            "Requirement already satisfied: six in /usr/local/lib/python3.6/dist-packages (from patsy>=0.5.1->category_encoders) (1.12.0)\n",
            "Installing collected packages: category-encoders\n",
            "Successfully installed category-encoders-2.2.2\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "stream",
          "text": [
            "/usr/local/lib/python3.6/dist-packages/statsmodels/tools/_testing.py:19: FutureWarning: pandas.util.testing is deprecated. Use the functions in the public API at pandas.testing instead.\n",
            "  import pandas.util.testing as tm\n"
          ],
          "name": "stderr"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "fD0xB_VYOf30",
        "colab_type": "code",
        "cellView": "form",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 323
        },
        "outputId": "3b6e13f0-2d1b-47a5-cb39-6a2c9afef59e"
      },
      "source": [
        "#@title TPU/GPU test\n",
        "\"\"\"\n",
        "testing if connected to TPU and/or GPU\n",
        "\"\"\"\n",
        "\n",
        "if 'COLAB_TPU_ADDR' not in os.environ:\n",
        "  print('Not connected to a TPU runtime.')\n",
        "else:\n",
        "  tpu_address = 'grpc://' + os.environ['COLAB_TPU_ADDR']\n",
        "  print ('Connected to TPU.\\n\\nTPU address is', tpu_address)\n",
        "\n",
        "  with tf.compat.v1.Session((tpu_address)) as session:\n",
        "    devices = session.list_devices()\n",
        "    \n",
        "  print('TPU devices:')\n",
        "  pprint.pprint(devices)\n",
        "\n",
        "if tf.test.gpu_device_name() == '':\n",
        "  print('\\n\\nNot connected to a GPU runtime.')\n",
        "else:\n",
        "  print('\\n\\nConnected to GPU: ' + tf.test.gpu_device_name())"
      ],
      "execution_count": 3,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Connected to TPU.\n",
            "\n",
            "TPU address is grpc://10.10.20.82:8470\n",
            "TPU devices:\n",
            "[_DeviceAttributes(/job:tpu_worker/replica:0/task:0/device:CPU:0, CPU, -1, -7049842693166689858),\n",
            " _DeviceAttributes(/job:tpu_worker/replica:0/task:0/device:XLA_CPU:0, XLA_CPU, 17179869184, -2127539043119477044),\n",
            " _DeviceAttributes(/job:tpu_worker/replica:0/task:0/device:TPU:0, TPU, 17179869184, -4205580246040150858),\n",
            " _DeviceAttributes(/job:tpu_worker/replica:0/task:0/device:TPU:1, TPU, 17179869184, -465520314227115850),\n",
            " _DeviceAttributes(/job:tpu_worker/replica:0/task:0/device:TPU:2, TPU, 17179869184, 7567214815007854213),\n",
            " _DeviceAttributes(/job:tpu_worker/replica:0/task:0/device:TPU:3, TPU, 17179869184, -1721259273674594457),\n",
            " _DeviceAttributes(/job:tpu_worker/replica:0/task:0/device:TPU:4, TPU, 17179869184, 1534747856600781995),\n",
            " _DeviceAttributes(/job:tpu_worker/replica:0/task:0/device:TPU:5, TPU, 17179869184, -3325583748349503068),\n",
            " _DeviceAttributes(/job:tpu_worker/replica:0/task:0/device:TPU:6, TPU, 17179869184, 1315233374759963046),\n",
            " _DeviceAttributes(/job:tpu_worker/replica:0/task:0/device:TPU:7, TPU, 17179869184, -1218497948575545607),\n",
            " _DeviceAttributes(/job:tpu_worker/replica:0/task:0/device:TPU_SYSTEM:0, TPU_SYSTEM, 8589934592, 6525352579356783460)]\n",
            "\n",
            "\n",
            "Not connected to a GPU runtime.\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "mxPnoGubOh-S",
        "colab_type": "text"
      },
      "source": [
        "# Assumptions about the data\n",
        "\n",
        "\n",
        "\n",
        "## 1.   ItinID and MktID are ambiguous columns that do not seem to be significant towards predicting the flight price. The order in which the tickets are bought should not, in an obvious manner, affect how much an airline will charge for a flight. ticket. Ergo, these columns will be deleted.\n",
        "## 2. The OriginWac and DestWac columns are useless too. They are just arbirary US State/Territory World Area Code that will not affect the flight price. They shall be removed.\n",
        "## 3.   Several columns are categorical. In order to deal with categorical and contnuous data at the same time, we will use [this](https://datascience.stackexchange.com/questions/29634/how-to-combine-categorical-and-continuous-input-features-for-neural-network-trai) technique. The categorical inputs will be run through an Embedding layer and a small neural network, and then their outputs will be concantenated as inputs with the continuous data.\n",
        "\n",
        "## 4. The categorical data columns are: Origin, Dest, and AirlineCompany.\n",
        "\n",
        "## 5. ContiguousUSA is already a numerical variable with the values being [2, 1]. It does not need to be converted to numerical values, instead just needs to be normalized.\n",
        "\n"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "gq6kU9jy92xt",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 456
        },
        "outputId": "10ee9584-3cb8-42f0-d786-81526794f197"
      },
      "source": [
        "\"\"\"\n",
        "reading data\n",
        "\"\"\"\n",
        "\n",
        "path = \"/content/drive/My Drive/MIA/Cleaned_2018_Flights.csv\"\n",
        "\n",
        "df = pd.read_csv(path)\n",
        "\n",
        "df"
      ],
      "execution_count": 4,
      "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>Unnamed: 0</th>\n",
              "      <th>ItinID</th>\n",
              "      <th>MktID</th>\n",
              "      <th>MktCoupons</th>\n",
              "      <th>Quarter</th>\n",
              "      <th>Origin</th>\n",
              "      <th>OriginWac</th>\n",
              "      <th>Dest</th>\n",
              "      <th>DestWac</th>\n",
              "      <th>Miles</th>\n",
              "      <th>ContiguousUSA</th>\n",
              "      <th>NumTicketsOrdered</th>\n",
              "      <th>AirlineCompany</th>\n",
              "      <th>PricePerTicket</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>0</td>\n",
              "      <td>20181767585</td>\n",
              "      <td>2018176758501</td>\n",
              "      <td>1</td>\n",
              "      <td>1</td>\n",
              "      <td>PHL</td>\n",
              "      <td>23</td>\n",
              "      <td>LAX</td>\n",
              "      <td>91</td>\n",
              "      <td>2402.0</td>\n",
              "      <td>2</td>\n",
              "      <td>1.0</td>\n",
              "      <td>AA</td>\n",
              "      <td>672.87</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>1</td>\n",
              "      <td>20181767586</td>\n",
              "      <td>2018176758601</td>\n",
              "      <td>1</td>\n",
              "      <td>1</td>\n",
              "      <td>PHL</td>\n",
              "      <td>23</td>\n",
              "      <td>LAX</td>\n",
              "      <td>91</td>\n",
              "      <td>2402.0</td>\n",
              "      <td>2</td>\n",
              "      <td>1.0</td>\n",
              "      <td>AA</td>\n",
              "      <td>367.68</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>2</td>\n",
              "      <td>20181767587</td>\n",
              "      <td>2018176758701</td>\n",
              "      <td>1</td>\n",
              "      <td>1</td>\n",
              "      <td>PHL</td>\n",
              "      <td>23</td>\n",
              "      <td>LAX</td>\n",
              "      <td>91</td>\n",
              "      <td>2402.0</td>\n",
              "      <td>2</td>\n",
              "      <td>1.0</td>\n",
              "      <td>AA</td>\n",
              "      <td>417.94</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>3</td>\n",
              "      <td>20181767636</td>\n",
              "      <td>2018176763601</td>\n",
              "      <td>1</td>\n",
              "      <td>1</td>\n",
              "      <td>PHL</td>\n",
              "      <td>23</td>\n",
              "      <td>LAX</td>\n",
              "      <td>91</td>\n",
              "      <td>2402.0</td>\n",
              "      <td>2</td>\n",
              "      <td>1.0</td>\n",
              "      <td>AA</td>\n",
              "      <td>247.10</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>4</td>\n",
              "      <td>20181767637</td>\n",
              "      <td>2018176763701</td>\n",
              "      <td>1</td>\n",
              "      <td>1</td>\n",
              "      <td>PHL</td>\n",
              "      <td>23</td>\n",
              "      <td>LAX</td>\n",
              "      <td>91</td>\n",
              "      <td>2402.0</td>\n",
              "      <td>2</td>\n",
              "      <td>1.0</td>\n",
              "      <td>AA</td>\n",
              "      <td>276.35</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>...</th>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>9534412</th>\n",
              "      <td>9534412</td>\n",
              "      <td>201844999082</td>\n",
              "      <td>20184499908202</td>\n",
              "      <td>1</td>\n",
              "      <td>4</td>\n",
              "      <td>STL</td>\n",
              "      <td>64</td>\n",
              "      <td>AUS</td>\n",
              "      <td>74</td>\n",
              "      <td>721.0</td>\n",
              "      <td>2</td>\n",
              "      <td>1.0</td>\n",
              "      <td>WN</td>\n",
              "      <td>310.10</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>9534413</th>\n",
              "      <td>9534413</td>\n",
              "      <td>201844999092</td>\n",
              "      <td>20184499909203</td>\n",
              "      <td>1</td>\n",
              "      <td>4</td>\n",
              "      <td>BNA</td>\n",
              "      <td>54</td>\n",
              "      <td>MCI</td>\n",
              "      <td>64</td>\n",
              "      <td>491.0</td>\n",
              "      <td>2</td>\n",
              "      <td>1.0</td>\n",
              "      <td>WN</td>\n",
              "      <td>106.17</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>9534414</th>\n",
              "      <td>9534414</td>\n",
              "      <td>201844999112</td>\n",
              "      <td>20184499911203</td>\n",
              "      <td>1</td>\n",
              "      <td>4</td>\n",
              "      <td>BOS</td>\n",
              "      <td>13</td>\n",
              "      <td>LIT</td>\n",
              "      <td>71</td>\n",
              "      <td>1260.0</td>\n",
              "      <td>2</td>\n",
              "      <td>1.0</td>\n",
              "      <td>WN</td>\n",
              "      <td>161.19</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>9534415</th>\n",
              "      <td>9534415</td>\n",
              "      <td>201844999113</td>\n",
              "      <td>20184499911303</td>\n",
              "      <td>1</td>\n",
              "      <td>4</td>\n",
              "      <td>BOS</td>\n",
              "      <td>13</td>\n",
              "      <td>LIT</td>\n",
              "      <td>71</td>\n",
              "      <td>1260.0</td>\n",
              "      <td>2</td>\n",
              "      <td>1.0</td>\n",
              "      <td>WN</td>\n",
              "      <td>170.87</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>9534416</th>\n",
              "      <td>9534416</td>\n",
              "      <td>201844999114</td>\n",
              "      <td>20184499911403</td>\n",
              "      <td>1</td>\n",
              "      <td>4</td>\n",
              "      <td>BOS</td>\n",
              "      <td>13</td>\n",
              "      <td>LIT</td>\n",
              "      <td>71</td>\n",
              "      <td>1260.0</td>\n",
              "      <td>2</td>\n",
              "      <td>1.0</td>\n",
              "      <td>WN</td>\n",
              "      <td>379.02</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "<p>9534417 rows × 14 columns</p>\n",
              "</div>"
            ],
            "text/plain": [
              "         Unnamed: 0        ItinID  ...  AirlineCompany  PricePerTicket\n",
              "0                 0   20181767585  ...              AA          672.87\n",
              "1                 1   20181767586  ...              AA          367.68\n",
              "2                 2   20181767587  ...              AA          417.94\n",
              "3                 3   20181767636  ...              AA          247.10\n",
              "4                 4   20181767637  ...              AA          276.35\n",
              "...             ...           ...  ...             ...             ...\n",
              "9534412     9534412  201844999082  ...              WN          310.10\n",
              "9534413     9534413  201844999092  ...              WN          106.17\n",
              "9534414     9534414  201844999112  ...              WN          161.19\n",
              "9534415     9534415  201844999113  ...              WN          170.87\n",
              "9534416     9534416  201844999114  ...              WN          379.02\n",
              "\n",
              "[9534417 rows x 14 columns]"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 4
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "usWKXticolEl",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 456
        },
        "outputId": "c692ae1c-3fa9-42b9-e2d4-bc85bcea1ba2"
      },
      "source": [
        "\"\"\"\n",
        "scaling the target Y price values\n",
        "\"\"\"\n",
        "\n",
        "bool_scale = False\n",
        "\n",
        "if bool_scale:\n",
        "  df['PricePerTicket'] = df['PricePerTicket'] / 1000.0 # scaling the target\n",
        "\n",
        "df"
      ],
      "execution_count": 5,
      "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>Unnamed: 0</th>\n",
              "      <th>ItinID</th>\n",
              "      <th>MktID</th>\n",
              "      <th>MktCoupons</th>\n",
              "      <th>Quarter</th>\n",
              "      <th>Origin</th>\n",
              "      <th>OriginWac</th>\n",
              "      <th>Dest</th>\n",
              "      <th>DestWac</th>\n",
              "      <th>Miles</th>\n",
              "      <th>ContiguousUSA</th>\n",
              "      <th>NumTicketsOrdered</th>\n",
              "      <th>AirlineCompany</th>\n",
              "      <th>PricePerTicket</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>0</td>\n",
              "      <td>20181767585</td>\n",
              "      <td>2018176758501</td>\n",
              "      <td>1</td>\n",
              "      <td>1</td>\n",
              "      <td>PHL</td>\n",
              "      <td>23</td>\n",
              "      <td>LAX</td>\n",
              "      <td>91</td>\n",
              "      <td>2402.0</td>\n",
              "      <td>2</td>\n",
              "      <td>1.0</td>\n",
              "      <td>AA</td>\n",
              "      <td>672.87</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>1</td>\n",
              "      <td>20181767586</td>\n",
              "      <td>2018176758601</td>\n",
              "      <td>1</td>\n",
              "      <td>1</td>\n",
              "      <td>PHL</td>\n",
              "      <td>23</td>\n",
              "      <td>LAX</td>\n",
              "      <td>91</td>\n",
              "      <td>2402.0</td>\n",
              "      <td>2</td>\n",
              "      <td>1.0</td>\n",
              "      <td>AA</td>\n",
              "      <td>367.68</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>2</td>\n",
              "      <td>20181767587</td>\n",
              "      <td>2018176758701</td>\n",
              "      <td>1</td>\n",
              "      <td>1</td>\n",
              "      <td>PHL</td>\n",
              "      <td>23</td>\n",
              "      <td>LAX</td>\n",
              "      <td>91</td>\n",
              "      <td>2402.0</td>\n",
              "      <td>2</td>\n",
              "      <td>1.0</td>\n",
              "      <td>AA</td>\n",
              "      <td>417.94</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>3</td>\n",
              "      <td>20181767636</td>\n",
              "      <td>2018176763601</td>\n",
              "      <td>1</td>\n",
              "      <td>1</td>\n",
              "      <td>PHL</td>\n",
              "      <td>23</td>\n",
              "      <td>LAX</td>\n",
              "      <td>91</td>\n",
              "      <td>2402.0</td>\n",
              "      <td>2</td>\n",
              "      <td>1.0</td>\n",
              "      <td>AA</td>\n",
              "      <td>247.10</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>4</td>\n",
              "      <td>20181767637</td>\n",
              "      <td>2018176763701</td>\n",
              "      <td>1</td>\n",
              "      <td>1</td>\n",
              "      <td>PHL</td>\n",
              "      <td>23</td>\n",
              "      <td>LAX</td>\n",
              "      <td>91</td>\n",
              "      <td>2402.0</td>\n",
              "      <td>2</td>\n",
              "      <td>1.0</td>\n",
              "      <td>AA</td>\n",
              "      <td>276.35</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>...</th>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>9534412</th>\n",
              "      <td>9534412</td>\n",
              "      <td>201844999082</td>\n",
              "      <td>20184499908202</td>\n",
              "      <td>1</td>\n",
              "      <td>4</td>\n",
              "      <td>STL</td>\n",
              "      <td>64</td>\n",
              "      <td>AUS</td>\n",
              "      <td>74</td>\n",
              "      <td>721.0</td>\n",
              "      <td>2</td>\n",
              "      <td>1.0</td>\n",
              "      <td>WN</td>\n",
              "      <td>310.10</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>9534413</th>\n",
              "      <td>9534413</td>\n",
              "      <td>201844999092</td>\n",
              "      <td>20184499909203</td>\n",
              "      <td>1</td>\n",
              "      <td>4</td>\n",
              "      <td>BNA</td>\n",
              "      <td>54</td>\n",
              "      <td>MCI</td>\n",
              "      <td>64</td>\n",
              "      <td>491.0</td>\n",
              "      <td>2</td>\n",
              "      <td>1.0</td>\n",
              "      <td>WN</td>\n",
              "      <td>106.17</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>9534414</th>\n",
              "      <td>9534414</td>\n",
              "      <td>201844999112</td>\n",
              "      <td>20184499911203</td>\n",
              "      <td>1</td>\n",
              "      <td>4</td>\n",
              "      <td>BOS</td>\n",
              "      <td>13</td>\n",
              "      <td>LIT</td>\n",
              "      <td>71</td>\n",
              "      <td>1260.0</td>\n",
              "      <td>2</td>\n",
              "      <td>1.0</td>\n",
              "      <td>WN</td>\n",
              "      <td>161.19</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>9534415</th>\n",
              "      <td>9534415</td>\n",
              "      <td>201844999113</td>\n",
              "      <td>20184499911303</td>\n",
              "      <td>1</td>\n",
              "      <td>4</td>\n",
              "      <td>BOS</td>\n",
              "      <td>13</td>\n",
              "      <td>LIT</td>\n",
              "      <td>71</td>\n",
              "      <td>1260.0</td>\n",
              "      <td>2</td>\n",
              "      <td>1.0</td>\n",
              "      <td>WN</td>\n",
              "      <td>170.87</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>9534416</th>\n",
              "      <td>9534416</td>\n",
              "      <td>201844999114</td>\n",
              "      <td>20184499911403</td>\n",
              "      <td>1</td>\n",
              "      <td>4</td>\n",
              "      <td>BOS</td>\n",
              "      <td>13</td>\n",
              "      <td>LIT</td>\n",
              "      <td>71</td>\n",
              "      <td>1260.0</td>\n",
              "      <td>2</td>\n",
              "      <td>1.0</td>\n",
              "      <td>WN</td>\n",
              "      <td>379.02</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "<p>9534417 rows × 14 columns</p>\n",
              "</div>"
            ],
            "text/plain": [
              "         Unnamed: 0        ItinID  ...  AirlineCompany  PricePerTicket\n",
              "0                 0   20181767585  ...              AA          672.87\n",
              "1                 1   20181767586  ...              AA          367.68\n",
              "2                 2   20181767587  ...              AA          417.94\n",
              "3                 3   20181767636  ...              AA          247.10\n",
              "4                 4   20181767637  ...              AA          276.35\n",
              "...             ...           ...  ...             ...             ...\n",
              "9534412     9534412  201844999082  ...              WN          310.10\n",
              "9534413     9534413  201844999092  ...              WN          106.17\n",
              "9534414     9534414  201844999112  ...              WN          161.19\n",
              "9534415     9534415  201844999113  ...              WN          170.87\n",
              "9534416     9534416  201844999114  ...              WN          379.02\n",
              "\n",
              "[9534417 rows x 14 columns]"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 5
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "mtgUyuabSD9M",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 419
        },
        "outputId": "7b96ee86-a3ff-4325-ba68-308ad50fc58f"
      },
      "source": [
        "\"\"\"\n",
        "removing the first three unnecessary columns\n",
        "removing OriginWac and DestWac because they are useless\n",
        "\"\"\"\n",
        "\n",
        "df.drop(['Unnamed: 0', 'ItinID', 'MktID', 'OriginWac', 'DestWac'], axis=1, inplace=True) # dropping index_col, ItinID, and MktID, OriginWac, DestWac\n",
        "\n",
        "df"
      ],
      "execution_count": 6,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/html": [
              "<div>\n",
              "<style scoped>\n",
              "    .dataframe tbody tr th:only-of-type {\n",
              "        vertical-align: middle;\n",
              "    }\n",
              "\n",
              "    .dataframe tbody tr th {\n",
              "        vertical-align: top;\n",
              "    }\n",
              "\n",
              "    .dataframe thead th {\n",
              "        text-align: right;\n",
              "    }\n",
              "</style>\n",
              "<table border=\"1\" class=\"dataframe\">\n",
              "  <thead>\n",
              "    <tr style=\"text-align: right;\">\n",
              "      <th></th>\n",
              "      <th>MktCoupons</th>\n",
              "      <th>Quarter</th>\n",
              "      <th>Origin</th>\n",
              "      <th>Dest</th>\n",
              "      <th>Miles</th>\n",
              "      <th>ContiguousUSA</th>\n",
              "      <th>NumTicketsOrdered</th>\n",
              "      <th>AirlineCompany</th>\n",
              "      <th>PricePerTicket</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>1</td>\n",
              "      <td>1</td>\n",
              "      <td>PHL</td>\n",
              "      <td>LAX</td>\n",
              "      <td>2402.0</td>\n",
              "      <td>2</td>\n",
              "      <td>1.0</td>\n",
              "      <td>AA</td>\n",
              "      <td>672.87</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>1</td>\n",
              "      <td>1</td>\n",
              "      <td>PHL</td>\n",
              "      <td>LAX</td>\n",
              "      <td>2402.0</td>\n",
              "      <td>2</td>\n",
              "      <td>1.0</td>\n",
              "      <td>AA</td>\n",
              "      <td>367.68</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>1</td>\n",
              "      <td>1</td>\n",
              "      <td>PHL</td>\n",
              "      <td>LAX</td>\n",
              "      <td>2402.0</td>\n",
              "      <td>2</td>\n",
              "      <td>1.0</td>\n",
              "      <td>AA</td>\n",
              "      <td>417.94</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>1</td>\n",
              "      <td>1</td>\n",
              "      <td>PHL</td>\n",
              "      <td>LAX</td>\n",
              "      <td>2402.0</td>\n",
              "      <td>2</td>\n",
              "      <td>1.0</td>\n",
              "      <td>AA</td>\n",
              "      <td>247.10</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>1</td>\n",
              "      <td>1</td>\n",
              "      <td>PHL</td>\n",
              "      <td>LAX</td>\n",
              "      <td>2402.0</td>\n",
              "      <td>2</td>\n",
              "      <td>1.0</td>\n",
              "      <td>AA</td>\n",
              "      <td>276.35</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>...</th>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>9534412</th>\n",
              "      <td>1</td>\n",
              "      <td>4</td>\n",
              "      <td>STL</td>\n",
              "      <td>AUS</td>\n",
              "      <td>721.0</td>\n",
              "      <td>2</td>\n",
              "      <td>1.0</td>\n",
              "      <td>WN</td>\n",
              "      <td>310.10</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>9534413</th>\n",
              "      <td>1</td>\n",
              "      <td>4</td>\n",
              "      <td>BNA</td>\n",
              "      <td>MCI</td>\n",
              "      <td>491.0</td>\n",
              "      <td>2</td>\n",
              "      <td>1.0</td>\n",
              "      <td>WN</td>\n",
              "      <td>106.17</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>9534414</th>\n",
              "      <td>1</td>\n",
              "      <td>4</td>\n",
              "      <td>BOS</td>\n",
              "      <td>LIT</td>\n",
              "      <td>1260.0</td>\n",
              "      <td>2</td>\n",
              "      <td>1.0</td>\n",
              "      <td>WN</td>\n",
              "      <td>161.19</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>9534415</th>\n",
              "      <td>1</td>\n",
              "      <td>4</td>\n",
              "      <td>BOS</td>\n",
              "      <td>LIT</td>\n",
              "      <td>1260.0</td>\n",
              "      <td>2</td>\n",
              "      <td>1.0</td>\n",
              "      <td>WN</td>\n",
              "      <td>170.87</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>9534416</th>\n",
              "      <td>1</td>\n",
              "      <td>4</td>\n",
              "      <td>BOS</td>\n",
              "      <td>LIT</td>\n",
              "      <td>1260.0</td>\n",
              "      <td>2</td>\n",
              "      <td>1.0</td>\n",
              "      <td>WN</td>\n",
              "      <td>379.02</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "<p>9534417 rows × 9 columns</p>\n",
              "</div>"
            ],
            "text/plain": [
              "         MktCoupons  Quarter  ... AirlineCompany PricePerTicket\n",
              "0                 1        1  ...             AA         672.87\n",
              "1                 1        1  ...             AA         367.68\n",
              "2                 1        1  ...             AA         417.94\n",
              "3                 1        1  ...             AA         247.10\n",
              "4                 1        1  ...             AA         276.35\n",
              "...             ...      ...  ...            ...            ...\n",
              "9534412           1        4  ...             WN         310.10\n",
              "9534413           1        4  ...             WN         106.17\n",
              "9534414           1        4  ...             WN         161.19\n",
              "9534415           1        4  ...             WN         170.87\n",
              "9534416           1        4  ...             WN         379.02\n",
              "\n",
              "[9534417 rows x 9 columns]"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 6
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "4vEVPeqZK1Nx",
        "colab_type": "text"
      },
      "source": [
        "## Applying the different types of encodings for all the variables."
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "nDwAFyRLKuMK",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 419
        },
        "outputId": "1ffa1743-396f-408e-ee04-960fc2bba616"
      },
      "source": [
        "\"\"\"\n",
        "moving around a couple of variables\n",
        "\"\"\"\n",
        "\n",
        "# making the target variable the last one\n",
        "temp_col = df['PricePerTicket']\n",
        "df.drop('PricePerTicket', axis=1, inplace=True)\n",
        "df['PricePerTicket'] = temp_col\n",
        "\n",
        "# placing the two continuous variables in the beginning\n",
        "tickets_col = df.pop('NumTicketsOrdered')\n",
        "df.insert(0, 'NumTicketsOrdered', tickets_col)\n",
        "miles_col = df.pop('Miles')\n",
        "df.insert(0, 'Miles', miles_col)\n",
        "\n",
        "# normalizing Quarter, Miles, and NumTicketsOrdered variables================================================\n",
        "bool_norm = True\n",
        "norm_vars = ['Miles', 'NumTicketsOrdered']\n",
        "\n",
        "if bool_norm:\n",
        "  for col in norm_vars:\n",
        "    df[col] = df[col] / df[col].max()\n",
        "# ===========================================================================================================\n",
        "\n",
        "df"
      ],
      "execution_count": 7,
      "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>Miles</th>\n",
              "      <th>NumTicketsOrdered</th>\n",
              "      <th>MktCoupons</th>\n",
              "      <th>Quarter</th>\n",
              "      <th>Origin</th>\n",
              "      <th>Dest</th>\n",
              "      <th>ContiguousUSA</th>\n",
              "      <th>AirlineCompany</th>\n",
              "      <th>PricePerTicket</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>0.471443</td>\n",
              "      <td>0.05</td>\n",
              "      <td>1</td>\n",
              "      <td>1</td>\n",
              "      <td>PHL</td>\n",
              "      <td>LAX</td>\n",
              "      <td>2</td>\n",
              "      <td>AA</td>\n",
              "      <td>672.87</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>0.471443</td>\n",
              "      <td>0.05</td>\n",
              "      <td>1</td>\n",
              "      <td>1</td>\n",
              "      <td>PHL</td>\n",
              "      <td>LAX</td>\n",
              "      <td>2</td>\n",
              "      <td>AA</td>\n",
              "      <td>367.68</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>0.471443</td>\n",
              "      <td>0.05</td>\n",
              "      <td>1</td>\n",
              "      <td>1</td>\n",
              "      <td>PHL</td>\n",
              "      <td>LAX</td>\n",
              "      <td>2</td>\n",
              "      <td>AA</td>\n",
              "      <td>417.94</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>0.471443</td>\n",
              "      <td>0.05</td>\n",
              "      <td>1</td>\n",
              "      <td>1</td>\n",
              "      <td>PHL</td>\n",
              "      <td>LAX</td>\n",
              "      <td>2</td>\n",
              "      <td>AA</td>\n",
              "      <td>247.10</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>0.471443</td>\n",
              "      <td>0.05</td>\n",
              "      <td>1</td>\n",
              "      <td>1</td>\n",
              "      <td>PHL</td>\n",
              "      <td>LAX</td>\n",
              "      <td>2</td>\n",
              "      <td>AA</td>\n",
              "      <td>276.35</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>...</th>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>9534412</th>\n",
              "      <td>0.141511</td>\n",
              "      <td>0.05</td>\n",
              "      <td>1</td>\n",
              "      <td>4</td>\n",
              "      <td>STL</td>\n",
              "      <td>AUS</td>\n",
              "      <td>2</td>\n",
              "      <td>WN</td>\n",
              "      <td>310.10</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>9534413</th>\n",
              "      <td>0.096369</td>\n",
              "      <td>0.05</td>\n",
              "      <td>1</td>\n",
              "      <td>4</td>\n",
              "      <td>BNA</td>\n",
              "      <td>MCI</td>\n",
              "      <td>2</td>\n",
              "      <td>WN</td>\n",
              "      <td>106.17</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>9534414</th>\n",
              "      <td>0.247301</td>\n",
              "      <td>0.05</td>\n",
              "      <td>1</td>\n",
              "      <td>4</td>\n",
              "      <td>BOS</td>\n",
              "      <td>LIT</td>\n",
              "      <td>2</td>\n",
              "      <td>WN</td>\n",
              "      <td>161.19</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>9534415</th>\n",
              "      <td>0.247301</td>\n",
              "      <td>0.05</td>\n",
              "      <td>1</td>\n",
              "      <td>4</td>\n",
              "      <td>BOS</td>\n",
              "      <td>LIT</td>\n",
              "      <td>2</td>\n",
              "      <td>WN</td>\n",
              "      <td>170.87</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>9534416</th>\n",
              "      <td>0.247301</td>\n",
              "      <td>0.05</td>\n",
              "      <td>1</td>\n",
              "      <td>4</td>\n",
              "      <td>BOS</td>\n",
              "      <td>LIT</td>\n",
              "      <td>2</td>\n",
              "      <td>WN</td>\n",
              "      <td>379.02</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "<p>9534417 rows × 9 columns</p>\n",
              "</div>"
            ],
            "text/plain": [
              "            Miles  NumTicketsOrdered  ...  AirlineCompany  PricePerTicket\n",
              "0        0.471443               0.05  ...              AA          672.87\n",
              "1        0.471443               0.05  ...              AA          367.68\n",
              "2        0.471443               0.05  ...              AA          417.94\n",
              "3        0.471443               0.05  ...              AA          247.10\n",
              "4        0.471443               0.05  ...              AA          276.35\n",
              "...           ...                ...  ...             ...             ...\n",
              "9534412  0.141511               0.05  ...              WN          310.10\n",
              "9534413  0.096369               0.05  ...              WN          106.17\n",
              "9534414  0.247301               0.05  ...              WN          161.19\n",
              "9534415  0.247301               0.05  ...              WN          170.87\n",
              "9534416  0.247301               0.05  ...              WN          379.02\n",
              "\n",
              "[9534417 rows x 9 columns]"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 7
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "fP1e1O4MQdjZ",
        "colab_type": "text"
      },
      "source": [
        "## First we separate the X and Y into training and testing datasets."
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "MRzfBKv4Qb5T",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 102
        },
        "outputId": "9dac1171-c1eb-4ddc-a897-fc9bbf08e601"
      },
      "source": [
        "\"\"\"\n",
        "shuffling the data before separating into X and Y\n",
        "configuring training and testing datasets\n",
        "\"\"\"\n",
        "\n",
        "data = df.to_numpy()\n",
        "\n",
        "print(f\"The shape of the data is: {data.shape}\")\n",
        "\n",
        "train, test = train_test_split(data, test_size=0.1, shuffle=True)\n",
        "\n",
        "X_train, Y_train = train[:, :-1], train[:, -1]\n",
        "X_test, Y_test = test[:, :-1], test[:, -1]\n",
        "\n",
        "print(f\"\\nTraining shape: {X_train.shape}\")\n",
        "print(f\"\\nTesting shape: {X_test.shape}\")"
      ],
      "execution_count": 8,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "The shape of the data is: (9534417, 9)\n",
            "\n",
            "Training shape: (8580975, 8)\n",
            "\n",
            "Testing shape: (953442, 8)\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "r76Og-S2tp6t",
        "colab_type": "text"
      },
      "source": [
        "## Next two cells make the data into *lists* instead of *tensorflow tensors* so that the model can be run in [mia](http://miamarketplace.com/)"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "XDtTr2Sm_4p2",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 68
        },
        "outputId": "95c96109-db10-48c8-c514-75d1c6e5cdc6"
      },
      "source": [
        "\"\"\"\n",
        "preparing the inputs to the model by parsing the X_train, X_test variables\n",
        "this implementation is coming from an article\n",
        "EMBEDDING LAYERS FOR ALL 6 CATEGORICAL VARIABLES and CONTINUOUS VARIABLES\n",
        "\"\"\"\n",
        "\n",
        "def parse_vars(var_data):\n",
        "    def categorical_func(cate_data):  \n",
        "        temp = list()\n",
        "        for i in range(cate_data.shape[1]):\n",
        "            le = LabelEncoder()\n",
        "            le.fit(cate_data[:, i])\n",
        "            label_encoded_feature = le.transform(cate_data[:, i])\n",
        "            temp.append(label_encoded_feature)\n",
        "        return temp\n",
        "    \n",
        "    temp_cate = categorical_func(var_data)\n",
        "    \n",
        "    return temp_cate\n",
        "\n",
        "train_data = parse_vars(X_train)\n",
        "test_data = parse_vars(X_test)\n",
        "\n",
        "Y_train = np.asarray(Y_train, dtype=np.float32)\n",
        "Y_test = np.asarray(Y_test, dtype=np.float32)\n",
        "\n",
        "print(f\"train_data length: {len(train_data)}\")\n",
        "print(f\"\\ntest_data length: {len(test_data)}\")"
      ],
      "execution_count": 41,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "train_data length: 8\n",
            "\n",
            "test_data length: 8\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "IMJEml7VqgkB",
        "colab_type": "text"
      },
      "source": [
        "## Now we will start making the dense neural network for the model."
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "lxGTS62pRG8F",
        "colab_type": "code",
        "cellView": "form",
        "colab": {}
      },
      "source": [
        "#@title Initiating loss for training and validation, creating callback functions for stopping training and saving the model periodically.\n",
        "\"\"\"\n",
        "keras callback\n",
        "\"\"\"\n",
        "\n",
        "# loss lists so that training need not be completed to plot graphs\n",
        "training_mse = []\n",
        "validation_mse = []\n",
        "\n",
        "class callback(Callback):\n",
        "  def on_epoch_end(self, epoch, logs={}):\n",
        "    if logs.get('mse') <= 0.01:\n",
        "      print(f'\\nTraining halted here. Model has achieved {round(logs.get(\"mse\"), 2)} loss on training set.')\n",
        "      self.model.stop_training = True\n",
        "    \n",
        "    training_mse.append(logs.get('mse'))\n",
        "    validation_mse.append(logs.get('val_mse'))\n",
        "\n",
        "call = ModelCheckpoint(checkpoint_dir, monitor='mse', verbose=1,\n",
        "    save_best_only=True, mode='auto', save_freq='epoch')\n",
        "\n",
        "training_stop = callback()"
      ],
      "execution_count": 42,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "ffqOH_GJkHBt",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 1000
        },
        "outputId": "89525f15-b671-43e2-a5d7-67f7a9d77ddb"
      },
      "source": [
        "\"\"\"\n",
        "creating the keras model\n",
        "\"\"\"\n",
        "\n",
        "# ========================================================================================================\n",
        "continuous_dense = False\n",
        "if continuous_dense:\n",
        "    cont_input = Input(shape=(2,), name='continuous_input')\n",
        "    cont = BatchNormalization(name='continuous_batchnorm_1')(cont_input)    \n",
        "    for i, unit in enumerate([64, 64], start=2):\n",
        "        cont = Dense(units=unit, activation='relu', name=f'dense_cont_{i}')(cont)\n",
        "        cont = BatchNormalization(name=f'continuous_batchnorm_{i}')(cont)\n",
        "# ========================================================================================================\n",
        "in_layers = list()\n",
        "em_layers = list()\n",
        "for i in range(8):\n",
        "    n_labels = len(np.unique(train_data[i])) # unique values in each categorical column\n",
        "    in_layer = Input(shape=(1,), name=f'input_layer_{i+1}')\n",
        "    emb_shape = min(64, (n_labels + 1) // 2)\n",
        "    em_layer = Embedding(n_labels, emb_shape, name=f'embedding_layer_{i+1}')(in_layer)\n",
        "    em_layer = Flatten(name=f'input_flatten_layer_{i+1}')(em_layer)\n",
        "    em_layer = Dropout(rate=0.6, name=f'input_dropout_layer_{i+1}')(em_layer)\n",
        "    in_layers.append(in_layer)\n",
        "    em_layers.append(em_layer)\n",
        "\n",
        "cate = concatenate(inputs=em_layers, name='merge_input_layers')\n",
        "# ========================================================================================================\n",
        "categorical_dense = False\n",
        "if categorical_dense:\n",
        "  for i, unit in enumerate([64, 64], start=1):\n",
        "      cate = Dense(units=unit, activation='relu', name=f'dense_cate_{i}')(cate)\n",
        "# ========================================================================================================\n",
        "\n",
        "for i, unit in enumerate([512, 512, 512], start=1):\n",
        "    cate = Dense(units=unit, activation='relu', name=f'merged_dense_{i}')(cate)\n",
        "    cate = BatchNormalization(name=f'merged_batchnorm_{i}')(cate)\n",
        "    cate = Dropout(rate=0.6, name=f'merged_dropout_{i}')(cate)\n",
        "\n",
        "out = Dense(units=1, activation='linear', name='output_layer')(cate)\n",
        "\n",
        "# creating the Model\n",
        "model = Model(inputs=in_layers, outputs=out, name='FP_Model')\n",
        "\n",
        "# saving the model graph and seeing the architecture \n",
        "plot_model(model, show_shapes=True, to_file='/content/drive/My Drive/MIA/nn_graph.png')\n",
        "\n",
        "model.summary()"
      ],
      "execution_count": 44,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Model: \"FP_Model\"\n",
            "__________________________________________________________________________________________________\n",
            "Layer (type)                    Output Shape         Param #     Connected to                     \n",
            "==================================================================================================\n",
            "input_layer_1 (InputLayer)      [(None, 1)]          0                                            \n",
            "__________________________________________________________________________________________________\n",
            "input_layer_2 (InputLayer)      [(None, 1)]          0                                            \n",
            "__________________________________________________________________________________________________\n",
            "input_layer_3 (InputLayer)      [(None, 1)]          0                                            \n",
            "__________________________________________________________________________________________________\n",
            "input_layer_4 (InputLayer)      [(None, 1)]          0                                            \n",
            "__________________________________________________________________________________________________\n",
            "input_layer_5 (InputLayer)      [(None, 1)]          0                                            \n",
            "__________________________________________________________________________________________________\n",
            "input_layer_6 (InputLayer)      [(None, 1)]          0                                            \n",
            "__________________________________________________________________________________________________\n",
            "input_layer_7 (InputLayer)      [(None, 1)]          0                                            \n",
            "__________________________________________________________________________________________________\n",
            "input_layer_8 (InputLayer)      [(None, 1)]          0                                            \n",
            "__________________________________________________________________________________________________\n",
            "embedding_layer_1 (Embedding)   (None, 1, 64)        135232      input_layer_1[0][0]              \n",
            "__________________________________________________________________________________________________\n",
            "embedding_layer_2 (Embedding)   (None, 1, 10)        200         input_layer_2[0][0]              \n",
            "__________________________________________________________________________________________________\n",
            "embedding_layer_3 (Embedding)   (None, 1, 2)         6           input_layer_3[0][0]              \n",
            "__________________________________________________________________________________________________\n",
            "embedding_layer_4 (Embedding)   (None, 1, 2)         8           input_layer_4[0][0]              \n",
            "__________________________________________________________________________________________________\n",
            "embedding_layer_5 (Embedding)   (None, 1, 64)        16768       input_layer_5[0][0]              \n",
            "__________________________________________________________________________________________________\n",
            "embedding_layer_6 (Embedding)   (None, 1, 64)        16512       input_layer_6[0][0]              \n",
            "__________________________________________________________________________________________________\n",
            "embedding_layer_7 (Embedding)   (None, 1, 1)         2           input_layer_7[0][0]              \n",
            "__________________________________________________________________________________________________\n",
            "embedding_layer_8 (Embedding)   (None, 1, 6)         72          input_layer_8[0][0]              \n",
            "__________________________________________________________________________________________________\n",
            "input_flatten_layer_1 (Flatten) (None, 64)           0           embedding_layer_1[0][0]          \n",
            "__________________________________________________________________________________________________\n",
            "input_flatten_layer_2 (Flatten) (None, 10)           0           embedding_layer_2[0][0]          \n",
            "__________________________________________________________________________________________________\n",
            "input_flatten_layer_3 (Flatten) (None, 2)            0           embedding_layer_3[0][0]          \n",
            "__________________________________________________________________________________________________\n",
            "input_flatten_layer_4 (Flatten) (None, 2)            0           embedding_layer_4[0][0]          \n",
            "__________________________________________________________________________________________________\n",
            "input_flatten_layer_5 (Flatten) (None, 64)           0           embedding_layer_5[0][0]          \n",
            "__________________________________________________________________________________________________\n",
            "input_flatten_layer_6 (Flatten) (None, 64)           0           embedding_layer_6[0][0]          \n",
            "__________________________________________________________________________________________________\n",
            "input_flatten_layer_7 (Flatten) (None, 1)            0           embedding_layer_7[0][0]          \n",
            "__________________________________________________________________________________________________\n",
            "input_flatten_layer_8 (Flatten) (None, 6)            0           embedding_layer_8[0][0]          \n",
            "__________________________________________________________________________________________________\n",
            "input_dropout_layer_1 (Dropout) (None, 64)           0           input_flatten_layer_1[0][0]      \n",
            "__________________________________________________________________________________________________\n",
            "input_dropout_layer_2 (Dropout) (None, 10)           0           input_flatten_layer_2[0][0]      \n",
            "__________________________________________________________________________________________________\n",
            "input_dropout_layer_3 (Dropout) (None, 2)            0           input_flatten_layer_3[0][0]      \n",
            "__________________________________________________________________________________________________\n",
            "input_dropout_layer_4 (Dropout) (None, 2)            0           input_flatten_layer_4[0][0]      \n",
            "__________________________________________________________________________________________________\n",
            "input_dropout_layer_5 (Dropout) (None, 64)           0           input_flatten_layer_5[0][0]      \n",
            "__________________________________________________________________________________________________\n",
            "input_dropout_layer_6 (Dropout) (None, 64)           0           input_flatten_layer_6[0][0]      \n",
            "__________________________________________________________________________________________________\n",
            "input_dropout_layer_7 (Dropout) (None, 1)            0           input_flatten_layer_7[0][0]      \n",
            "__________________________________________________________________________________________________\n",
            "input_dropout_layer_8 (Dropout) (None, 6)            0           input_flatten_layer_8[0][0]      \n",
            "__________________________________________________________________________________________________\n",
            "merge_input_layers (Concatenate (None, 213)          0           input_dropout_layer_1[0][0]      \n",
            "                                                                 input_dropout_layer_2[0][0]      \n",
            "                                                                 input_dropout_layer_3[0][0]      \n",
            "                                                                 input_dropout_layer_4[0][0]      \n",
            "                                                                 input_dropout_layer_5[0][0]      \n",
            "                                                                 input_dropout_layer_6[0][0]      \n",
            "                                                                 input_dropout_layer_7[0][0]      \n",
            "                                                                 input_dropout_layer_8[0][0]      \n",
            "__________________________________________________________________________________________________\n",
            "merged_dense_1 (Dense)          (None, 512)          109568      merge_input_layers[0][0]         \n",
            "__________________________________________________________________________________________________\n",
            "merged_batchnorm_1 (BatchNormal (None, 512)          2048        merged_dense_1[0][0]             \n",
            "__________________________________________________________________________________________________\n",
            "merged_dropout_1 (Dropout)      (None, 512)          0           merged_batchnorm_1[0][0]         \n",
            "__________________________________________________________________________________________________\n",
            "merged_dense_2 (Dense)          (None, 512)          262656      merged_dropout_1[0][0]           \n",
            "__________________________________________________________________________________________________\n",
            "merged_batchnorm_2 (BatchNormal (None, 512)          2048        merged_dense_2[0][0]             \n",
            "__________________________________________________________________________________________________\n",
            "merged_dropout_2 (Dropout)      (None, 512)          0           merged_batchnorm_2[0][0]         \n",
            "__________________________________________________________________________________________________\n",
            "merged_dense_3 (Dense)          (None, 512)          262656      merged_dropout_2[0][0]           \n",
            "__________________________________________________________________________________________________\n",
            "merged_batchnorm_3 (BatchNormal (None, 512)          2048        merged_dense_3[0][0]             \n",
            "__________________________________________________________________________________________________\n",
            "merged_dropout_3 (Dropout)      (None, 512)          0           merged_batchnorm_3[0][0]         \n",
            "__________________________________________________________________________________________________\n",
            "output_layer (Dense)            (None, 1)            513         merged_dropout_3[0][0]           \n",
            "==================================================================================================\n",
            "Total params: 810,337\n",
            "Trainable params: 807,265\n",
            "Non-trainable params: 3,072\n",
            "__________________________________________________________________________________________________\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "C_dMTKuMy9BG",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 238
        },
        "outputId": "db586637-4f08-4bcb-f21c-f7f4fe3f18a6"
      },
      "source": [
        "\"\"\"\n",
        "compiling and fitting\n",
        "\"\"\"\n",
        "\n",
        "optimizer = Adamax(learning_rate=0.001)\n",
        "\n",
        "model.compile(optimizer=optimizer, loss='mse', metrics=['mse'])\n",
        "\n",
        "# model will begin training with previously trained weights\n",
        "try:\n",
        "    model.load_weights(checkpoint_dir)\n",
        "    print('Same architecture as before.\\n\\n')\n",
        "except:\n",
        "    print('Model architecture has been changed. No weights loaded.\\n\\n')\n",
        "\n",
        "history = model.fit(\n",
        "    [train_continuous],\n",
        "    Y_train,\n",
        "    batch_size=512,\n",
        "    epochs=500,\n",
        "    validation_data=([test_data], Y_test),\n",
        "    callbacks=[call, training_stop]\n",
        ")"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Model architecture has been changed. No weights loaded.\n",
            "\n",
            "\n",
            "Epoch 1/500\n",
            "16758/16760 [============================>.] - ETA: 0s - loss: 14811.4150 - mse: 14811.4150\n",
            "Epoch 00001: mse improved from inf to 14811.44727, saving model to /content/drive/My Drive/MIA/best_model.hdf5\n",
            "16760/16760 [==============================] - 408s 24ms/step - loss: 14811.4473 - mse: 14811.4473 - val_loss: 16904.7305 - val_mse: 16904.7305\n",
            "Epoch 2/500\n",
            "16759/16760 [============================>.] - ETA: 0s - loss: 14739.7549 - mse: 14739.7549\n",
            "Epoch 00002: mse improved from 14811.44727 to 14739.67285, saving model to /content/drive/My Drive/MIA/best_model.hdf5\n",
            "16760/16760 [==============================] - 409s 24ms/step - loss: 14739.6729 - mse: 14739.6729 - val_loss: 16905.6797 - val_mse: 16905.6797\n",
            "Epoch 3/500\n",
            " 2411/16760 [===>..........................] - ETA: 5:41 - loss: 14669.9678 - mse: 14669.9678"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "pY7yKB1qxnCG",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 433
        },
        "outputId": "e310867c-ea1b-4efe-e958-a077818ed109"
      },
      "source": [
        "\"\"\"\n",
        "plotting losses\n",
        "\"\"\"\n",
        "\n",
        "width_in_inches = 30\n",
        "height_in_inches = 10\n",
        "dots_per_inch = 50\n",
        "\n",
        "plt.figure(\n",
        "    figsize=(width_in_inches, height_in_inches),\n",
        "    dpi=dots_per_inch)\n",
        "\n",
        "plt.rcParams['axes.facecolor'] = '#3D383D'\n",
        "plt.rcParams['legend.facecolor'] = 'white'\n",
        "\n",
        "plt.plot(training_mse, 'o--b', label='training loss', mew=7, linewidth=3)\n",
        "plt.plot(validation_mse, 'o--r', label='validation loss', mew=7, linewidth=3)\n",
        "plt.legend(loc=\"upper right\", fontsize=25)\n",
        "\n",
        "plt.xlabel('Epoch', fontsize=38, color='white')\n",
        "plt.ylabel('MSE', fontsize=38, color='white')\n",
        "\n",
        "plt.xticks(range(1, len(training_mse)), fontsize=17, color='white')\n",
        "plt.yticks(fontsize=17, color='white')\n",
        "\n",
        "plt.grid(color='grey', linestyle=':', linewidth=1.7)\n",
        "plt.show()\n",
        "plt.close()"
      ],
      "execution_count": 18,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 1500x500 with 1 Axes>"
            ]
          },
          "metadata": {
            "tags": [],
            "needs_background": "dark"
          }
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "xKFfFGPJJ7b4",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "\"\"\"\n",
        "loading the best trained model (can jump directly here if training is not required)\n",
        "\"\"\"\n",
        "\n",
        "model = load_model(checkpoint_dir)"
      ],
      "execution_count": 19,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "20zIqXJcVPH1",
        "colab_type": "text"
      },
      "source": [
        "## Now that we have a trained model with its losses plotted, we can test it on example flights.\n"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "qjp20mejn1fj",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "\"\"\"\n",
        "saving the required data onto drive for later use\n",
        "\"\"\"\n",
        "\n",
        "# these go here for model predicting purposes =======\n",
        "ex_train = [train_continuous, train_categorical]\n",
        "ex_test = [test_continuous, test_categorical]\n",
        "# ===================================================\n",
        "\n",
        "# consistency with data types =====================================\n",
        "train_continuous = np.asarray(train_continuous, dtype=np.float32)\n",
        "test_continuous = np.asarray(test_continuous, dtype=np.float32)\n",
        "# =================================================================\n",
        "\n",
        "# saving model predictions on training and test sets\n",
        "preds_train = model.predict(ex_train, verbose=0)\n",
        "preds_test = model.predict(ex_test, verbose=0)\n",
        "\n",
        "save(model_preds_train, preds_train)\n",
        "save(model_preds_test, preds_test)\n",
        "\n",
        "# saving the true Y values\n",
        "save(truth_train, Y_train)\n",
        "save(truth_test, Y_test)"
      ],
      "execution_count": 20,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "QrzjQomWpVYd",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "\"\"\"\n",
        "loading the saved data from drive\n",
        "\"\"\"\n",
        "\n",
        "# loading model predictions on training and test sets\n",
        "preds_train = load(model_preds_train)\n",
        "preds_test = load(model_preds_test)\n",
        "\n",
        "# loading the true Y values\n",
        "Y_train = load(truth_train, allow_pickle=True)\n",
        "Y_test = load(truth_test, allow_pickle=True)"
      ],
      "execution_count": 21,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "xEekkyDIoJDS",
        "colab_type": "code",
        "cellView": "form",
        "colab": {}
      },
      "source": [
        "#@title enlarge(feature) util function implemented\n",
        "\"\"\"\n",
        "important util function to modify the input model shapes\n",
        "\"\"\"\n",
        "\n",
        "def enlarge(feature):\n",
        "  \n",
        "  if len(feature.shape) == 0:\n",
        "    return enlarge(np.expand_dims(feature, 0))\n",
        "\n",
        "  return np.expand_dims(feature, 0)"
      ],
      "execution_count": 22,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "FHTHwQq5ewKU",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 119
        },
        "outputId": "089fe183-4df6-4f27-c27b-60450e1e2f3e"
      },
      "source": [
        "\"\"\"\n",
        "testing the model on an example flight price\n",
        "\"\"\"\n",
        "\n",
        "rand_train = random.randint(0, len(Y_train))\n",
        "rand_test = random.randint(0, len(Y_test))\n",
        "\n",
        "truths_train = round(float(Y_train[rand_train]), 2) \n",
        "truths_test = round(float(Y_test[rand_test]), 2)\n",
        "\n",
        "pred_train = round(float(preds_train[rand_train]), 2)\n",
        "pred_test = round(float(preds_test[rand_test]), 2)\n",
        "\n",
        "print(f\"A training example\\nTrue price: ${truths_train}, model predicted: ${pred_train}\")\n",
        "print(f\"\\n\\nA testing example\\nTrue price: ${truths_test}, model predicted: ${pred_test}\")"
      ],
      "execution_count": 23,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "A training example\n",
            "True price: $134.0, model predicted: $190.28\n",
            "\n",
            "\n",
            "A testing example\n",
            "True price: $298.28, model predicted: $549.01\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "WZV0C6VGW5ft",
        "colab_type": "text"
      },
      "source": [
        "## Checking the model's accuracy on the entire training and test sets."
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "2gpk-ETdZfqI",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 68
        },
        "outputId": "b26d02c9-4a74-4bcf-9a8d-8335f1fef3d9"
      },
      "source": [
        "\"\"\"\n",
        "predicting an approximate accuracy of the model\n",
        "if the model predicts the correct price of a flight with a +- epsilon value\n",
        "then that will be a correct prediction, else incorrect\n",
        "\"\"\"\n",
        "\n",
        "def accuracy(spec='train', func_epsilon=70.):\n",
        "  preds = None\n",
        "  if spec == 'train':\n",
        "    preds = np.squeeze(preds_train)\n",
        "    diff = np.absolute(np.subtract(preds, Y_train))\n",
        "    diff = diff <= func_epsilon\n",
        "    diff = diff.astype(int)\n",
        "    diff = diff.sum()\n",
        "    return round((float(diff) / Y_train.shape[0]) * 100, 2)\n",
        "\n",
        "  else:\n",
        "    preds = np.squeeze(preds_test)\n",
        "    diff = np.absolute(np.subtract(preds, Y_test))\n",
        "    diff = diff <= func_epsilon\n",
        "    diff = diff.astype(int)\n",
        "    diff = diff.sum()\n",
        "    return round((float(diff) / Y_test.shape[0]) * 100, 2)\n",
        "\n",
        "\n",
        "epsilon = 70.0\n",
        "\n",
        "print(f\"Epsilon taken as ${int(epsilon)}\")\n",
        "print(f\"Accuracy on the training set: {accuracy('train', epsilon)}%\")\n",
        "print(f\"Accuracy on the test set: {accuracy('test', epsilon)}%\")"
      ],
      "execution_count": 24,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Epsilon taken as $70\n",
            "Accuracy on the training set: 53.41%\n",
            "Accuracy on the test set: 49.21%\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "Ii7fPY3ABP5d",
        "colab_type": "text"
      },
      "source": [
        "## Checking the effect of epsilon price on the model accuracies on training and test sets."
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "3InaHXByBVnb",
        "colab_type": "code",
        "cellView": "form",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 517
        },
        "outputId": "3098f6b0-581e-41ba-c198-233652f4afbb"
      },
      "source": [
        "#@title Code for making the graph\n",
        "\"\"\"\n",
        "checking effect of epsilon price on accuracy\n",
        "\"\"\"\n",
        "\n",
        "max_eps = 300\n",
        "\n",
        "training_accs = []\n",
        "testing_accs = []\n",
        "\n",
        "for temp_epsilon in range(0, max_eps + 1, 5):\n",
        "    training_accs.append(accuracy('train', temp_epsilon))\n",
        "    testing_accs.append(accuracy('test', temp_epsilon))\n",
        "\n",
        "width_in_inches = 35\n",
        "height_in_inches = 13\n",
        "dots_per_inch = 50\n",
        "\n",
        "plt.figure(\n",
        "    figsize=(width_in_inches, height_in_inches),\n",
        "    dpi=dots_per_inch)\n",
        "\n",
        "plt.rcParams['axes.facecolor'] = '#3D383D'\n",
        "plt.rcParams['legend.facecolor'] = 'white'\n",
        "\n",
        "plt.plot(range(0, max_eps + 1, 5), training_accs, 'o-b', label='training accuracy', mew=4, linewidth=3)\n",
        "plt.plot(range(0, max_eps + 1, 5), testing_accs, 'o-r', label='validation accuracy', mew=4, linewidth=3)\n",
        "\n",
        "plt.legend(loc=\"lower right\", fontsize=21)\n",
        "\n",
        "plt.xlabel('Epsilon price in USD ($)', fontsize=30, color='black')\n",
        "plt.ylabel('Accuracy (%)', fontsize=30, color='black')\n",
        "\n",
        "plt.xticks(range(0, max_eps + 1, 5), fontsize=15, color='black')\n",
        "plt.yticks(range(0, 101, 10), fontsize=18, color='black')\n",
        "\n",
        "plt.grid(color='grey', linestyle=':', linewidth=1.7)\n",
        "\n",
        "plt.show()\n",
        "plt.close()"
      ],
      "execution_count": 26,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "image/png": "iVBORw0KGgoAAAANSUhEUgAABYkAAAIaCAYAAABlHsJPAAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAAHsAAAB7AB1IKDYgAAADh0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uMy4yLjEsIGh0dHA6Ly9tYXRwbG90bGliLm9yZy+j8jraAAAgAElEQVR4nOy9eXgc53Wn+1ZV7419aew7CRBcABAACZIgaFvxllw7k32byYwlx/Jk8UzG2SMrcWzFSSbXGWe9dhKLjjOTZ5zcZDyZmTgTZ5krxpZkiZREyaREiRQXgOACgit6req6f1R3oQE0QKC7S6wGzvs8fLrRy69+3ae+r06d/nhKMU0TQRAEQRAEQRAEQRAEQRAEYWuiPmgDgiAIgiAIgiAIgiAIgiAIwoNDisSCIAiCIAiCIAiCIAiCIAhbGCkSC4IgCIIgCIIgCIIgCIIgbGGkSCwIgiAIgiAIgiAIgiAIgrCF8TxoA28FDXUNZsAfeNA2BEEQBEEQBEEQBEEQBEEQHjgzV2b+yjTN787+vSWKxAF/gN7u3gdt4y3D4/Gwe2Q3r7z4Crquu1JTPG4dj05oikfx6CbNcvDohKZ4FI+bXVM8ikc3aZaDRyc0xaN43Oya4lE8ukmzHDw6oemEx3Jh5srMhdy/pd3EJiRtppmfmydtpl2rKR63jkcnNMWjeHSTZjl4dEJTPIrHza4pHsWjmzTLwaMTmuJRPG52TfEoHt2kWQ4endB0wmO5opim+aA9OE57S7u5lVYSC4IgCIIgCIIgCIIgCIIgrMaxZ479J9M0P5r9W1YSb0I0TWNg5wCaprlWUzxuHY9OaIpH8egmzXLw6ISmeBSPm11TPIpHN2mWg0cnNMWjeNzsmuJRPLpJsxw8OqHphMdyRYrEgiAIgiAIgiAIgiAIgiAIW5gH0m5CUZSfAD4A7AG+Yprmd+Q8VwV8FngfEAN+zzTNT673+XxIuwlBEARBEARBEARBEARBEAQLt7SbuAw8AfxRnud+F6gDOoEp4EOKovzrDTy/5VE1la6eLlStdOEttaZ43DoendAUj+LRTZrl4NEJTfEoHje7pngUj27SLAePTmiKR/G42TXFo3h0k2Y5eHRC0wmP5coD+QZM0/wr0zS/DMzlPq4oSgj4AeBjpmneMk3zDFZR+IPreV6wUBWVmroaVKWEg7DEmuJx63h0QlM8ikc3aZaDRyc0xaN43Oya4lE8ukmzHDw6oSkexeNm1xSP4tFNmuXg0QlNJzyWKw+k3YS9cUX5ODCSbTehKMpe4ATgNU1Tzzz2LuDPTdOsvd/zq22no63D3N67HV3XUVUVVVPRUzqKouDxekglUwD4fD6SyaR9P5VKYZomXq8XwzBIp9N4PB7SZpq0kbabWhuGgaqpqIpqb0PTNFKpFIqi4PV6l+iuZxumaWIYhrUNBQy9sG14fV70lI5pmni8HtJGeu1tLP9+PJ5Vt5HrXddXbkPzaGBib0NRlDVjcL9t2N9PEdt4EHEuJgb324btvYziXFAM1thGsWPNDXEuNgbLt1HsWHtgcS5hDGRO3UAMvB7S6fuPNTfG2bEYbMY4v0X5S1nF+S3KX8o1zhuOgaaiquvLX8oxzgXFYJ35S1nF2cE5ddUYuD3Ock4o54RyTuiqOXWr5i9yTijnhIXE4B+P/aMr2k2sRgWwkC0AZ7gFVK7z+SUoivIeRVF+S9EUBncPAhBpjjC6bxSA6ppqDk4dBMAf8DP10BQejweAybdPUlFZAcD4gXHqG+sB2D2ym/aOdgC279hOX38fAB1dHewa2gVAY6SRsYkxACqrKjn0tkOAtSNNPTSFz+cD4NCRQ1RWW9ZH94/S2NQIwK6hXXR0dQDQ19/H9oHtALR1tLFnZA8A9Q31jB8cByBcEebwOw4DoHk0ph6aoq2jDVVVOXj4INW11QDsHd9LU0sTAIO7B+nq6QKgZ1sPA4MDALS2tzI8OgxAXX0d+yf329uYemgKTdNQVZWph6YIBAMATExOUFtv1eiHR4dpaWsBYMfOHXT3dQPQ1dvFjl07AGhubWbv+F4izRHq6us4cPjAkhhkJ4Wph6YIhUMA7Du0j7r6OgCGRoZo62gDoH9HP73brX7TnT2d7N23F1VViTRFGN1vxbmquopDR6wY+P3WNrxe74o4jx0Yo6GxwYrz8G46uzqJNEfYvmM72wa2AdDe2c7u4d0ANDQ2MHbAinNFZQWTb5+04uy14uz3++04V1VXWdvYP8a2gW2oqsrOPTvp6Lbi3Lu9l/4d/Xach0aG7BjsO7QPgFA4xNRDU1acNSvOwVCQSHOEA4cPUFNbY8e5ubXZisGuHXT1WnHu7utmx04rBi1tLXaca+trmZicACAQDDD10BRNLU1LtgGwf3K/HYPh0WFa21sBGBgcoGdbjxXnni57rDW1NLF3fK+1jbpaDh05hKqqBALWNjSPFefD7zhMuCIMwPjBceobrLG2Z2SPHeftA8vG2vAuIs0Razxn4lxZXWnH2efzWXH2WXE+9LZDVFZZY21sYozGyMqxtm1gG0N7h1BVlfaOdnaPWHGub6xn/MD4ijh7PB4rzgErzgenDlJdY4210X2jNLc2E2mOsHPPTjp7OoGVY21obybODXXsP7R/SZwVRVkx1g4cPkBPXw+qqjIyNmLHeWDXgD3Wunu7l4y1kbERAGrqajgweWBJnD1eD5HmyJKxtv/QfuoaMmNt75Ad5/7BnLHW3cngnvxz6qGpQ0SaIwSDQXsbsHJOtcfasjl1W7811jo6rTirqkpff1/+OTUz1nz+xTk1O9Zy59SdQzuXzKn9O/qJNEdo72hnz949dgz2HbTGWna+g8Wxlo3BwcMHl4y1ppYmKx7jI3T3di/GeacV59yxljunBkNBO86KoiwZaxOTE9Q31BNpjjAyOmLPqQM7B+jpy4y13i527LbinDvWampr7Dk1G+fcObWjqwNVVdl3cJ8d5z0je2hrX5xT+7YvjrWdQzutODdFGNtvxSA7p6qqSmt76+pz6sQYDRErzruGd9HRuTjW7Dk1Z6w1NDYwfmCcSHOEquqqxbHm9aw6p47uGyXSHLHivGcnnd3WWMvOqaqqMrBzwB5r9Q31a86pgcDiWKups+KcO9Z27LKOa5HmCD3bepbMqdmxVltXy8ThpXOqqqp2nEOhzFibtMaaqqqMT4zb813unNrZ05l3Tl0rfzn8jsNUVlUSaY5Yc2rj4px63/ylqXFxTs0Za9ltZOe73Dl1zfxlx3Y7znb+kplTVVWlo7vDzl/WmlOX5y+rzanDo8NEmiM0NDbYYy0UWn1OnTg8QW1drR3n5flL9lidHWsr5tTsWAusHGu5+Uv2uDa0d4i2zjYizREGBgeWzKk792TGWs6cmi9/WT6nqqrKxOSEHYPcOTV3rGXnVICGSIM9p+bLXwLBAJHmyIr8JdKUGWu5c+r2vsX8pX0xT82dUysqK+xxsHxOXZ6/ZOO8Y/di/tLTt3JOVVWVnm09dv6y1py6PH9ZMqdmxlp3bzcj4yOoqrpiTj142Bpr+ebUbP6yZE7da82pqqoytHdo8biWM6fmjrXcOPv8viVz6vL8pa+/D1VVl86p/duWjLXlc+ryOOfOqaqqMvm2STsGuXPq4J7BpXPqoBXn++UvqqrS1Nq0NM6Ta8ypmWNnd183A7usOOeOtbr6Og4eOWjlkPnm1PvkL/nmVFVV2TawrWTnhP2D/ewZ2WPNayU6J1RVlb379toxKPScMHtcOzh10J7XijknzB1r2Xk7G4NCzwlhMX+JNEfsbRR6TtjR3WHPqU0tTew/tB9VVQs+J2zvXDqnqqrKjl07loy1jZ4Tju4fXTKnZvfLQs8Js8fO7JyazfXsGBR4Tpg71to721FVtahzwuraantOzcY5e75W0Dlhbv4yYc1f1TXVRZ0T5uYvHZ0djE6MoqpqweeE9py6e5Cu3i4izRF6t/cWfU6Ym79sH9xu5bwFnhNmWyxMPTRFRWUFkeZISc4JYXGstbS1WMecAs8JYWn+kj1XUFW14HNCe05tb2No75CdQxZ6TmjHOXdOLfKcMN+cWopzwnB4MX9ZjhtXEh8HfDkrhd8J/EXOSuJVn19tO1ttJXEoHGJo7xDHnz0OCiX5hcPr8zI2McbxZ46TSqWK/oXD7/czPDbMiW+cwMQsya/G/oCfveN7Of7scTsmxcRZ0zT27tvLi8+/SCqVKskvWcFgkOGxYY4/exzTNIv+1djEZGz/GCdPnCQWi5Xkl6xgKGjvP4qilOSXLJ/Px+jE6JL9p5hfjX0+H8Njw7zw3Auk0+mS/GocCAQYGR+x9p+0UfSviZqa2X+Ov0gqmSrJr4mhUIihUSs2JmbRvxqbpsnYxBgnXzhJLBoryS+WwWDQ9qgoStFzqqIojE2M8dKJl4hFYyX51djr9TIyPsILz72AYRhF/2qsaRpjE2O88PwLJOKJkvxqrCgKo/tHeen4SyQSiZL8arzk2JD5notZSYyJvf9EF6IlOXb6A36GR61jA1D0L/eqolr7z/GXiMViJVn1pnk09o5njg16qujVAR6Px9p/nnuBRCJRktUBYCXFL514iUQ8UXT+4vV6GTuweg5QSI4E+fefYlZoLMktTLPoFRqqau0/Lx5/kXgsXpKVOPaxISe3KGYVSHb/OfHcCZKJZEnyVPvYsNHcYpWVxGvlAIWuxFFQbI/RaLTolTilziHTZhpNtY4NLz7/IvF4vOiVxMFQkOHRxRyyFOcj2bF94tkTJJPJkqyGyptbFLGS2OfzMbp/lOPPHkfX9ZKsuLJzi9xjQxHnhMXmkPm2sWZuUcD5yHpzyI3kL/ax4dnjpJKp0uWQ3ziOwjpyyHVsw+/3s3ffXnv/KcW5f/bYkN1/il15GAgGGBmz9p90Ol2SlcTLc4tiVzcuzyFLsbqxoNxirfrCOnLIjeYvuftPdpvFrDAtdQ55v/pCISuJ71dfKCRPzd1/ltcXCq2zrTe3WG+dze/L5JDPncBMr55DboWVxG4rEoeAm8Ah0zSPZx77aeDbTdM8cr/nV9tOe0u72dvd6+yHEQRBEARBEARBEARBEARBKAOOPXPswbebUBTFoyhKAPAAqqIoAUVRfKZpRoEvAZ9UFKVaUZTtwEeAPwa43/OChaIoVFZVoiiKazXF49bx6ISmeBSPbtIsB49OaIpH8bjZNcWjeHSTZjl4dEJTPIrHza4pHsWjmzTLwaMTmk54LFceVE/ijwEx4DHg/Zn7f5d57ieA28A08DXg86ZpfjHnvfd7fsvj9XoZPzBu91hyo6Z43DoendAUj+LRTZrl4NEJTfEoHje7pngUj27SLAePTmiKR/G42TXFo3h0k2Y5eHRC0wmP5coDbTfxViHtJgRBEARBEARBEARBEARBECxc0W5CcB4nfgEptaZ4dKdeuWiKR3fqOaEpHt2rKR7dqeeEZjl4dEJTPLpTzwlN8eheTfHoTj0nNMvBoxOa4tGdek5oikf3asoqYgspEm9CfD4fR77liH2Fczdqiset49EJTfEoHt2kWQ4endAUj+Jxs2uKR/HoJs1y8OiEpngUj5tdUzyKRzdploNHJzSd8FiuSLuJTYrP7yOZSLpaUzy6U69cNMWjO/Wc0BSP7tUUj+7Uc0KzHDw6oSke3annhKZ4dK+meHSnnhOa5eDRCU3x6E49JzTFo3s1nfBYDixvNyFFYkEQBEEQBEEQBEEQBEEQXM3c/BEuTD9MIt6CPzBLV/tRGuqeco1euSE9ibcAXp+Xt73zbXh9Jbx6ZIk1xePW8eiEpngUj27SLAePTmiKR/G42TXFo3h0k2Y5eHRCUzyKx82uKR7Fo5s0S603N3+EEy9/gWdPfJUTL3+BufkjReudPvME0WgfRjpENNrH6TNPMDc/tSEd01QxDD9Xr7+7JHqbCVlJvAlRFIWq6iru3L5DqeJbak3xuHU8OqEpHsWjmzTLwaMTmuJRPG52TfEoHt2kWQ4endAUj+Jxs2uKR/FYKHPzR7g4/TDxRAsB/yydJVhRW0q96zeO8OrrT6x4vLfrt6muepl02ks67cU0vaTN7H1fzuM+0qYHM+0lnXn86vVvRddrVmhqapSKilczr/VYmrZ2VsNrPw/amt5DobOMDT1c8GcvJ6TdhCAIgiAIgiAIgiAIgiDkwe0tDbIrapczsO2Xqat5ziq0ZgqrZtpn30+bvpyia7Yg6+PewiBXr3/bCr2a6ufweW+QTvuWFXO9S/XyFHtN01Pw53vQaGqUQ/vf+6BtvCVIkXgL4PF6GNs/xvFvHEdP6a7UFI9bx6MTmuJRPLpJsxw8OqEpHsXjZtcUj+LRTZrl4NEJTfEoHje7pngsXjO7AjaZasPnnSnJitp8BdjB/sdoqDu26vtME0zTkym++jOFWD/zNw/y5sUfX/H6lqa/IhQ8n1Ow9eUUdHP+mf5lf/uIxTowTV/Bn/HBYuLzzqGqKRQlhaomUZUUippacquqyczzKfv22vX3kNJrVyj6fFfp6fx/ULMaip6jo1vvt7V1FDWJqui8+Mof8J7YS3ycj9PDm7xJDx/n4/zv0PCWXUlcvqV9YVXSRpoL5y6QNtKu1RSPW8ejE5riUTy6SbMcPDqhKR7F42bXFI/i0U2a5eDRCU3xKB43u+ZW9Jgt6h57uhW/73JRRd3lBd1UyuopO9j/GHU1X7eKtdl/pnVrZIu4RiDznA8j5/nLs9+Td1uvvfEYM+EzSwrAi8Vb6/792hjkMnv1uwr6zOtFVRJWMVRNoSrWrVUcXXxMUa1i7Pz8JGae8qCiJOnu+GO7SLv4npz7io6S1VRT9v1XTn+aaKxnhWYodK7gAmxV5SvsODO6oqj7avfxNQv4K74b08SfNvj3VT/Jp2L/0358iJf5K76bj1R/Ny8V5LD8kZXEgiAIgiAIgiAIgiAIwhJK3SZhtT61XR2fozL8WqaI6yedDtjFXSMdsAq6pg/DLuwGMNJ+7tzdQzodzLMlE1AK9lkKVCWBqibQjQpAzfMKg/q6Y5mVtEnrVk1kbnMfy/3bKvy+ce5niCfaViiGgm+yd88HrZW0G/j4x08e5T3R0q6onZs/kr+g2792QVcxTQLpNEHDIJA2CBoGwXSagGEwdvsm/3J2ZsV7/rmmjhs+H/50Gn86TSBtWLdGGr9p4DfSBNJp/JnHffepg54NhXh4aKygz11uSLuJLYDH42Hn0E5OnTyFrpfov4OUWFM8bh2PTmiKR/HoJs1y8OiEpngUj5tdUzyKRzdploNHJzTFo3jc7Jql1itF6wXTVDGMAHM33sHrb/7ciudbm/6cUOiCVag1glYRN3PfKuoGM8Xc7GMBu9BrpIM8iOKtqsYzBdiEXYzVlMTKx7Q4c/NTeS+O5vNdpbfr95e8Plu4VdUkWk6RV1GSdpH2+MmjRKN9K/SKuThaoQXY1Ri5EOF3Zv9yxeMfafluXuq6tuLxbCE3kC3k5tzPPj589w7feXV2xXtfrKjinteTKQKnM0Vgg4CRtm7TpVtVXyhRVeO9+w89aBtvCdJuYgtgmiZ3bpXuqp5OaIrHrePRCU3xKB7dpFkOHp3QFI/icbNrikfx6CbNcvDohKZ4FI+bWdNuu/BMYW0XTFOzirRGECMd5Mb8Ic5f+jH7+Wzrhcb6/03AfzVTvM0WbTP3jYBd1DXSQdJGkLTpX3O7l69+X8GfeXXShEJvomWKudqSom4cTcu5n7lV1QQXLv0IiWTzCrVg4DzDu3/cLuRuZFVtbc2zeXsS93X/TkEF2K72o/kLuu3HN6yV5bv4K57gV+2/s20SHmOQYzQsvtA08ZlpQoa1IjeUKcjm/h0yDP7V3DN5t/Mfr/43zt8J2UVca2VvGr9ZeCF35N6ddb0uDcRVjZimElM1YppGb3Qhb0OPlKLwJ22dJFWVuKoSVzXrvmb9nVS1zOMqCVUjkXnucy+/SF8sukJvNrD2GNjMyEpiQRAEQRAEQRAEQRCENSim9YJpkllZG0Y3gtyYP7ykoJuloe4f8Puv24XfdE4R2Fh237xPMbd4VmvZYFBf+/VM8TaWuY1nCrg5j6lxVC3zmBpD1eJ889VfJxbvXqFY6KraQi8ydz9GLjTy6JVv0mXOcEFp4w+bd/Fi1/WCtI7Mz/HEmdMrHn+sf5BjdcsLuqa1sjZTzF1ym1lpGzIMfuDyNHV6aoVmXFG56vfnvMdwfGVoLFOUjWeKudn7w3du5y3oJhWFz3T3Ede0TPE3876cv+OZQu7y6v7Rk8fpi64s6hbaHmLdsdnESLuJLYCmafRt7+Ps62cxDMOVmuJx63h0QlM8ikc3aZaDRyc0xaN43Oya4lE8ukmzHDw6oSkexWOh2G0Xkm34fIW1XQBIpzWMdIjrcw9x9vxPrXg+0vAV/P6rGEYIwwhjGCH0zO3Sx0Js5KJmG0FRdDQ1umrvW0XRaWn6sl2o1dRYppAbQ7P/jmf+jtmF3hde/hzRWOnbJJS6qDs3P8XF6UdyYv1kUQXijRQOtXSaUNogrGdW5aZ1wrpB2DAIGTqhtMEPzeQv6CayBV273YKzBd2oqhLTrBW50czK3Oz98du3qDJWtju56vPzWz19mcKvZheEY9riilxzlWXapS7ogjNF3an5OR6ZvkhbMsmMz8eT7Z1bpkAM0m5ia6CA5tFK2+qn1Jri0Z165aIpHt2p54SmeHSvpnh0p54TmuXg0QlN8ehOPSc0xaN7NcWjO/VKrJkt6H7tG4X30V2ul1uITOlW24WO1j8hFDqfWc0bXizg6uGcx0JLnk+nA2tu69rct27QXRpNW8AwwuQv6KZoa/4LVC1bsF12q8VQ1ah9X1NjqKpV3Fut920weIG+7t/ZoE/o6jiat6Db1f7khrWyNNQ9xWD/Y1ZR1+6bXFxR97v4bzzMb9OmJJnBx1E6eYo1inyZFgy5xdyKzG3YMHj04vm8b/vYG69xKXCRUGY1b1g3imq74DfTdMZjeZ+Lq4stFnLbLcRUjaimcfDWPNV5eljP+vz8Wl8/0WwBOPOeuKaRXqPnxmrF19/p7uXp2vqCPt/R9q68mk+2dxWkB/BUXQOP9Q9aRd1Ukhlv8UXdY3UNfD3SRP+Ofs68egZDL80PaeWKrCQWBEEQBEEQBEEQBOEtZ7WVpTu2f4ya6hMYeiW6UYGuV9i3hv13Zd7HE8lGSr9qN3/rBUXRaY78DzRtAU2L4tGimfvL/7buq2oMRXHuYmZOrNK9MP1ITouN4gq6YBUkH56+QEs8wWzAz9H2Lp5aZ5FPMU3ChkHY0KnQdSZvzvMj0xdWvO7Z6hrueLxWMTdT/A3nFIK9Ja6DxVWVaKYou6B5iKoaUY/GgqYxcfMm1XlW6c76/Hxq20BmNa9qF4HvV9AFJ1fULsblyfauolfUOqEplBZpN7EFUDWVtvY2ZqZnSBuluTJkqTXF49bx6ISmeBSPbtIsB49OaIpH8bjZNcWjeHSTZjl4dEJTPLrTY3blbyK59gXX0mktU8CttP4ZmVu9yv776rVvRTeqiva0Pgyqq17CYxdxFzL3o3g899AyBV2rsHtvSYH3hVf+qKRFXad66WZbLyzGpvjWCw9PX6Q1meCyz8/R9s51F3Tz8Y656/zKG6+uePzLkWYuB4JU6DoVhvUvrBsr7ocMI8/668JIw5LC7kLmdte9O1TkaXlyxefn/+7dlnmP9fpopl2Doa7uyqm+t9k2CdnYlKJNQjnMZ05oOuGxXJB2E1sAVVVpiDQwe3m2dIOwxJricet4dEJTPIpHN2mWg0cnNMWjeNzsmuJRPLpJsxw8OqEpHovXzBZ0v15gK4fsBddSehW6Xs3c/CSXZh6xn9czrRwqwt9E0xJLisBGOlSUdwBVjePR7qF57uHR7uHx3MOj3c3cLn/cun3tjceIxTtXaIVC5xna+ZMF+ehqL23rBSfaLli6x2huepo9I3t4+cWX0VMrV7Cul+XFzT5d54kzp/lkXz8vV1ZTYehUZoq6S251nUpDp0I3qNR1Ko2UfX+19gzfce3KhrwlFIUFj4eaVCpv0TilKHyus3tJ8TeqadzLKfDGNC1vP93Virq/293LN2rqNuQTnGmRAFabhKebmksS6yxun8+c0nTCY7kiK4kFQRAEQRAEQRAEYZOx2mrV3u7PUBF6HV2vzhR/q3Juq9FTmVu9kpRehWn6i/Jhrcq9i8eTKe567uL13MGj3eXq3HtIpVb2PA0G32R0zwftfrsbwclVuqVuvVBq1tvKwZc2qNJ1qnSrsFupp3Lu61TpKd5543reFbVOoAN/3xDhnscq4N7TPNzzeLiXXeXr8XA3534ys3LXiYujgbRJELYO0m5iC6CqKnUNdczPzZNOl+6XmlJqiset49EJTfEoHt2kWQ4endAUj+Jxs2uKR/HoJs1y8OiE5lb0aLdySLTi96/eysE0wTDCpPQaUqnqJbd65nbuxtsw0uGiPYF1QTWP5w6pVC3kveBakv6+T+HNFoK1u3g8d/B4FlCU1QuNjrddsL/H0hR0Sxlvu5VDIsFlfwGtHEyTYDpNVabAOzk/xyMzl1a87HS4gqSqZoq/VkHYX2QdKKGo3PNo3PV4uKd5ltwueczj4a62ePsbr32TnljpirpOtXIA988VTmiWg0cnNJ3wWC5Iu4ktgKZp9O/o57mnnyvZDl5qTfG4dTw6oSkexaObNMvBoxOa4lE8bnZN8Sge3aRZDh6d0NxKHk1T5er19/D6uV+wH9OjViuH2upn0DwLpFI1OUXgakzTW+jWCPhnrNW83jt4PbczBd071mOe23g9d/B6rfsezx20+1xwLRi8RKThHzfsZLHtwgdJpdrwemfobP98SdoutDQ9w76D+3ju6edIpVJF6WUpVbxXtHKIWq0cPtPVyxvhCrvwW6XrVGfuV6dSSx6v0lP41lHsHVy4l/dxHbjr8XLX4+FOprh7J/P3Xc3D+69doSGVXPG+c8EQHxgubJXu5zu68hZ1n2zvKkgv28rhg9MXaUulmPF6+XwJWjmAe+cKJzXLwaMTmk54LFdkJbEgCIIgCIIgCIIgbJC5+SNcmH44p/3A4qpf01TQjQqrsJuqJZWqJZkp8lp/L95PpmrR9SryrdC9H6oaxeu5jdd7a8Xt5avfSTIZWWpga1AAACAASURBVPGeQi+2lv3MTqz8LQfW28pBM00q9RTVqRQ1mSJvdSqVubX+npq/QThdmlYOSUXhjsdLXSqZdw9KKgpPbBvIFIK91m2mRy95evLmfl7nLrgmrRwEwQ1Iu4ktgKIohMNhFhYWKFV8S60pHreORyc0xaN4dJNmOXh0QlM8isfNrikexaObNMvBoxOapdbLtnKIJ1oI+GcLuIibQkqvJpWs4/r827k084EVr/H7ZjFNLym9BtMszX/cVZQUPZ2fzRSAb+H13sbruYXHewtNXbnSM4vTrRwWv8fiWzmUOtbZVg4tiTiz/sDGWzlkMU3ChsG75q7x0fNnVzz9fFU1CU3LFIEzq32Nwi8gZgIzgQC3M8Xc25mVvdm/73i83PZat9m/Y6oKiuJIf16roLv4PZbigmvg/rnCCU3x6F5NJzyWC8uLxBv/qVJwPV6vl4nDE3i9hf73I+c1xePW8eiEpngUj27SLAePTmiKR/G42TXFo3h0k2Y5eHRCs5R62YLpQrQPwwixkGnlMDc/RTrtIx5v5s7dXczNT3H5yndw/tIHef3cz/DNV3+dF17+Q549/pf887P/wLPH/5oTL38hb4EYIJFsIZlqsAvEmnaXQGCaqoqXqa99iubIX9PR9kV6u36bgW2/wu7Bn2R06ANMjH4HoeDKIiRAMHiRtpa/INLwVWprnqMifAa//9qaBWJYbOVQET6Hz5egInyuJCt+G+qOcWD8wzz++G9yYPzDJVlBXMpYZ1fA9kUXCBkGfdEFnjhzmqn5Ocis8u2IRdlz5zZT83O8/+osPzxzkY+cP8vjr7/Kp0+/zOdPnuAvjz/L33/ja3zl+afzFogBxu/cZvLmPLvv3aUjHltSIL7t8XAxEOTliir+ubaO/9XYxJ+1tPMHnT1c8/ny6p0LhfihkX386O4Rfm7Hbj61bYDf7e7ji+2dfLm5lX9saOR4dS2vhyu46g8Qy1kNfHSVlg2FtnIAOFbXwIfHD/Cbjz/Oh8cPlGzFr5vnCqc0xaN7NZ3wWK7ISmJBEARBEARBEARh02EYfpKpBpLJBl5942Mkk035XgVoG1BN4/XcJqVXk2/NlarGGd7145kVv7dQ1fX3wt2qrRzW28ZhNXxpg7pkitpUkrpUiv9w/g0iyZUFdB1AUfAUUAMxgXyNGZKKwme6+7jt9XLb4+W218stj4d7Hi+GtHIQBMHlSLuJLYKmaRhGaXocOaUpHt2pVy6a4tGdek5oikf3aopHd+o5oVkOHp3QFI/u1HNCUzy6T3Otfr/ptEYyVU8y2WAXgZPJehLJxsXHkw3oRuW6t6coCXy+eXzeG/i88/h8N6z7vht4vZnHffN4PTdRVWPVC7gV0+/X+txTXJh+JOdzF9/KAdwb69WKpZ/o6+fViiq78FubSlKbSuX8nXlMTxEuwMM9TeOWx8str1XYvZm5f8vrtR+/lfPY51550aFWDs4UdLfi/CMe3alXLppOeCwHpN3EFsDn8/H2d70d3yr/hcUNmuJx63h0QlM8ikc3aZaDRyc0xaN43Oya4lE8ukmzHDyWQtM0QdfDTM9+D6fPPEE02oeRDhHNtIb4xon/yjPHv8zXvvFPPPfC/8tL3/wsp888wdnzP8mlyz/Mtbn3cuv2PqKxHrtArKoxgoFLqGos7zaDgYscHP82Jve/i/17v5+R3T/GzoGPsa3nP9HZ/kWaI/+L+tqnqaw4g983h6paRYSu9qN59branyzos2dpqDvGwfEP8/gv/SYHS9TKodSxPjI/xxdefoG/fe7rfOHlFzgyP7eu93nTaSKJOIN373B4/gbffnWWj557I+9rf+nsGf7spef5/VMn+eTrp/no+bM8PHOR77h2hSM3b7Dn3h3aE3G7QJxQFK74/JwOV7Cg5l8ZfikQ4LtG9/Mt+yf5tn2H+KG9+/ix3SP8wsAu/mNfP3/Y2cOft7Tzd41NfKOmjjMVlVzzB0iqmoOtHA7ym7/0OB8eP1iyAvFWnH/E49bx6ISmEx7LFVlJvEkJBAPEY3FXa4pHd+qVi6Z4dKeeE5ri0b2a4tGdek5oloNHJzTFozv1nNAUj8Vjr/xNtOD3L135m0XXQySSEZLJRhLJCIlExLpNRqxVwIkIRjq0ru0pSiqzuncOn28Ov28u8/f1zN838Hnn0LQFFMWZVg72ql/7M5dm1S+4d/9ZbeXvZ7p7OR8MU5dKUp9MWrepJHXJlHWbSlKtb+yCblFVY97r5abXx02vl5teL/P2fV/mOev+Qk5vXsdbOSQSzPpLt/LX7WPbCT0nNMWjO/XKRdMJj+WAtJsQBEEQBEEQBEEQSsZqBdjqqudRlLRdFDaM8DoV83eAVZUEw7t/FJ/vOl7PHRRlY+eyTrVycDMF9/s1TcKGQUMqSX0yQUMySWMyyffPTlOzwWJvLilF4YbXx7zXxw2fj6E7t6k2VuqdC4b4wHBhbRxAevMKgiCsB2k3sQXw+rxMPTSF11fCq0eWWFM8bh2PTmiKR/HoJs1y8OiEpngUj5tdUzyKRzdpllpvbv4IJ17+As+e+ConXv4Cc/NH7vseXa/g3kIfN+YPMzP73Zy78OOcOvNJXnj5Dzl95lfyvuf2nXFu3d5PNNZjF4g92l1CobPU1jxDc+Sv6Wr/Y/p7P8Xuwf/A2PC/5NC+dxMKncurFwhOUxF+A5/39oYLxGC1cjgw/ig/9/O/yoHxR0tSIHbz/pNdUdsXjRJKG/RFozxx5jTvmLtGSzzGnju3eWjuOt87O8OPXjjH46+/ym9/8yT/5cXn+dvnvs5Xnn+aP33pOJ85/QofO3uGD186v2qB2AQuBIKcqKrmq/WNfKmljT/o7OGTfQP8h8Hd/OuhUd43doBv2T/J943u59/uGeGxgZ38Zt/2vHqf7yi8jQNYrRweHT/Ar/78z/Ho+IGSFYjdPrad0BSP4nGzazrhsVyRlcSbEEVRqKmt4dbNW5QqvqXWFI9bx6MTmuJRPLpJsxw8OqEpHsXjZtcUj+LRTZql1Ftt1e/23l8lHDpPPNFMItFMItlEPNFCItFEPNGEsYGLwC36TrKt59P4/dfw+67j911H0/L3Bl6Px2JaQyx6cm9sSqEZ0nUiyQSRZIKfPfc6kWSyKC93NY05n585n48bXh8Tt25Sq6dWvK74C7hdpDUR57I/wJPtnSUp6rotNm+FnhOa4lE8bnZNJzyWC9JuQhAEQRAEQRAEYYuh62HiiVZOvfarJJLNBWl4PLcI+K/g91+1bn1XCPivcO7ijxGPd6x4fSh0lrGhhwva1lZsDQFrt4cI6TqNyQSRZNIqBCcSmb8TNGYey17I7X4kFJXrPqvlw3Wfnxs+H3NeH3M+35KicFxbehE4p/r9CoIgCG890m5iC+Dxeth3cB8er8e1muJx63h0QlM8ikc3aZaDRyc0xaN43Oya4lE8ukVzbv4IJ04e5Znjf8eJk0dXbQ2RTmvE4q3cvDXO7NVv580L/5bTZz7OCy//IU8/9z95+vmv8MLLn1+zQOz13qCy4hUa6v+B9tb/Ql/3p9k18LOMDv0wh/a9h4Pj387ePY+ys/9xert+n7aWv6S+7mv0dH4ur15X+5MFf+6GumNMjH2If//vH2di7EMlKxC7df/xGwbfOTuTtz3En594lr957uv87fNP86cnT/DpV1/h5869zsMzF3nf9avsv32LnljULhDf0Ty8EQqzoGp5t/VmMMS79h/ih/bu4yO7hvnE9h38flcvX2pt5x8aIrxUVc1MILiiQAzwVF0Dj/UPcjYUJu71cjYULkmB2O3j0ClN8Sge3aRZDh6d0HTCY7ki38AmJG2kmb44TdpIu1ZTPG4dj05oikfx6CbNcvDohKZ4FI+bXVM8ikc3aC5vu5BK9XH6zBO0Nn8Jr/cO8Xgr8UQL8XgLiWQEyF8UzOL13MRIB0mnAyueCwXPMTb8gYJ8NtQ9xWD/Y1ycfoREshW/7zKdJVj56+bYQHbV70Vanz7GZZ+fo+2deS8Kp5om9ZmVvk2JhN0OoimRoCkRJ5JMrHkxuOaclhF3NQ/X/Nbq32s5/677fdatz28Xd1db9fvHHV2grLww4Ho5VtfA1xoiNLU0cXX2Kul08d+l22PtlKZ4FI9u0iwHj05oOuGxXJF2E4IgCIIgCIIgCC7ANCGl1xCLtxOPtfPmxR8lpdeu+/2qkiAQuEzAP7t4m3Nf02KO9vvdSqxWgP3vjU3c9XoXi8GJBI2pJJ51nHebQL7SbVxV+ZE9e7nu8xPLs7p3Lax+v4vtK55s75K2EIIgCAIg7Sa2BB6Ph90ju/F4Sricv8Sa4nHreHRCUzyKRzdploNHJzTFo3jc7JriUTwWynraQ6T0Su7c3cnV6+/m/KVHePX1X7LaQjz/Nzx7/K85+c0/4My5X1yjQGzQ1Pg3dLZ/noG+TzK868eYGP1ODu1/F2PD/4ZdO36evu7foa3lL6iv+2fCoXP2BeOyq37DobN4vXHCobMlKxC7PTYFaZomFXqK7Qv3mJqf43tnZ/h358/yC2fP5H35v7h+lX91eZp33bjO8N07tCQTeEwTXVGY9fl5qbKKr9Y38p9b2/l0dx8/N7CTD+wZ5dvGD3IuGMqrORMIcDEY2nCBGKxVvx8aneCnH32UD41OlKxAvClj/QA0xaN4dJNmOXh0QtMJj+WKa78BRVH6gN8DDgBR4LdN0/yPmeeqgM8C7wNiwO+ZpvnJB+XVbZimSXQhWtKrMpZaUzxuHY9OaIpH8egmzXLw6ISmeBSPm11TPIrHQlitPURD/d+jKgaxeDuxeDu6Xr2GShq/7yrB4DT3FvrzvjYUOk9/368X7LOh7hhNjV+nq7eLC+cuYKzzYmf3w82xgZz2EM/ktIeorafS0GlJJGhOxDP/ErTk3F/vxeAAdODLza1c8/m56vfbt/NeH+n7tHg42tGVd3Xyk+1dG/2oS9hq47CcNMWjeHSTZjl4dELTCY/liivbTSiKogEngS8DHwd6ga8CP2+a5p8pivInQBPwA0AE+HvgY6ZpfjGfnrSbEARBEARBEASh1JimSiLRRDTWSSzeycXpf4NuVK3rvT7fNYKB6SX/AoFLBAOzqKrVg1ZaQ5SGgGHw/muzfOTCmyuei6sqgXX01L3h9XLFH+CK388Vf4B3X79GYyq54nVnQyEeHhor2Ku0hxAEQRDeKpa3m3DrSuKBzL9fMU0zBbymKMrngUcVRfkyVnF40jTNW8AtRVF+F/ggkLdIvNXQNI2evh7ePPtmyVYFlFpTPG4dj05oikfx6CbNcvDohKZ4FI+bXVM8bg2Pc/NHuDj9MIlkG37fDJ3tR2moe2rF6wwjQCzeYRWDY51EY13E4p3EYh2kTf99t6MoOl3tn88Ugy8RCFxG0+L3fd/SC8JlPRZ/QThwf2yyq37bkglm1rgoHACmSY2eoi0epzUepy0RozUepzURpy0eoz6VWnU72QLx8iLwbM79qz4/iWWtHk5XVDqy6vdYXQNfb2zaUuPQCb1y0RSP4tFNmuXg0QlNJzyWK24tEmd7JSvLHhvCKh77gBdznnsR+MXVxBRVwePxoOs6qqqiaip6SkdRFDxeD6mklTT4fD6SmSvH+nw+UqkUpmni9XoxDIN0Oo3H4yFtpkkbabRMomAYBqqmoiqqvQ1N00ilUiiKgtfrXaK7nm2YpolhGNY2FDD09W/D6/fiC/hAAa/Pi57SMU0Tj9dD2kivvY3l34/HY21DVQiEAnZElnvX9ZXb0DwamNjbUBTF9u7xefAFfCiqglfzkkrm/67yfj+rbcOj4Q/6QaEkcVZUBV/AZ22jiDgviUHmc6OwrhjcbxtpM40v4MPr85KOp4uPs6Is2X+KjbO9jTz7z4bG2rJteLzW96hqKqqqFjXWcrdh7z/FxtnrQVGs/cfj9dj7z7pjsNo48C3GZq04541BzjjIbsNIG9b+4/eSjqWLj7OiLPFYijnVxLT3cSNmrDsGa82pmkez95/cz1FInNNGGhTwB/1oHq00cfZ67c/t8VreNxKD1ea7JceG5XNqOv9YW2sb6XTa3n+MqFGSY2d2/1FUBZ/mW3cMVpvvAHxB63s0DGNDY221baiaunhsKDLOuq7b+4+qqfaxrJg4Lz82mAmz+PxFYc0coJAcyfa4bP8pJM7ZGGSPDYqq4FW9647BavNdNjbZ/ed+c+p6tpE9NuTmFoXGOZVKWR5Di/tPKfJU+9iw0dwic0xekb+skQMUmqea5v2PDeuJ8/W5KU6f+QRZdN1qDdHdeRRNvUks3kk80cXCQgeJZDOroShJQsFpgoEL3L47Qiq1so9wKHiBnq4vLYsB64pzQ90x2tueo3d7L6+deg0FhbSpFn8+khnbiqqAsXZusd75Lm9usc44546Dt928wa/kFGD7dJ0nzpzmM9sGuOjx0pqI05lK0hxdoCUepy0RJ3SfE/vVLgoXU1W+++AU9zJXtV9v/vLPjREeAz44c4nWZIIZr4+jnd083RCBIs8J18wtCsxT15NDbui4tmz/KVUOmS8H2NA5Ye42NNXO70t17p/v2FBoDpn9HFmPhZ4TLs9flucWRcWZlTlkMXG2Y1BIbrHGNtaTQxaSv+TWF4qKswM5ZHYbvmD++sJ6zwk3Ul8oJE9dq75QcJ1tnbnFeutsG60vlKzOVkSeWmwOmd3Gctx64brXgPPAJxRF8SuKsgt4BKgCKoAF0zT1nNffAiqXiyiK8h5FUX5L0RQGdw8CEGmOMLpvFIDqmmoOTh0EwB/wM/XQlN2oevLtk1RUVgAwfmCc+sZ6AHaP7Ka9ox2A7Tu209ffB0BHVwe7hnYB0BhpZGzC+i9GlVWVHHrbIcDakaYemsLn8wFw6MghKqst26P7R2lsagRg19AuOro6AOjr72P7wHYA2jra2DOyB4D6hnrGD44DEK4Ic/gdhwHQPBqTRyY5d+Ychm5w8PBBqmutPmZ7x/fS1NIEwODuQbp6rF+5e7b1MDA4AEBreyvDo8MA1NXXsX9yP2DtSE3NTaSNNKqqMvXQFIFgAICJyQlq661keHh0mJa2FgB27NxBd183AF29XezYtQOA5tZmhkaGOHXyFJWVlRw4fGBJDLKTwtRDU4TC1sUb9h3aR119HQBDI0O0dbQB0L+jn97tvbZ3TdUwdINIU4TR/Vacq6qrOHTEioHfb20jOxhy4zx2YIyGRmt1wu7h3bS0tXDq5Cl6+nrYNrANgPbOdnYP7wagobGBsQNWnCsqK5h8+6QVZ68VZ7/fb8e5qrrK/n7mr89j6AY79+yko9uKc+/2Xvp39NtxHhoZsmOw79A+AELhEFMPTVlx1jR7fz118hT7Du6jprbGjnNzq3Uis2PXDrp6rTh393WzY6cVg5a2FjvOtfW1TExOABAIBpg8Msmrr7xK2kgz9dAUwVAQgP2T++0YDI8O09reCsDA4AA923qsOPd02WOtqaWJveN77e+nrq4OQzcIBAJWnDMnyIffcZhwRRiA8YPj1DdYY23PyB47ztsHlo61HTt3cOrkKerq6+w4V1ZX2nH2+XxWnH1WnA+97RCVVdZYG5sYozGycqx193Zj6AaGbtDe0c7uESvO9Y31jB8YXxFnj8djxTlgxfng1EGqa6yxNrpvlPqGek6dPEX/jn46ezqBlWNtaG8mzg117D+0f0mcFUVZMdb2HdjHlZkrGLrByNiIHeeBXQP2WOvu7V4y1kbGRgCoqavhwOQBO85TD01hmianTp5i8sikPdb2H9pPXUNmrO0dsuPcP7g41jq7Oxnck39OnTg0wamTp/Bo1veTLZIvn1PtsbZsTt3Wb421js4Odg3vwtAN5q7N2Z9jyZyaGWs+/+Kcmh1ruXPqzqGdS+bU3m29nDp5iuaWZvbs3WPHYN9Ba6yFK8Irxlo2BgcPH1wy1ppamqwDv6Lan6NnWw8DO60454613Dk1GAracVYUZclYm5icoKq6ilMnT9lzEcDAzgF6+jJjrbeLHbutOOeOtZraGntOzcY5O6dOHpnk4rmLGLrBvoP77DjvGdlDW/vinNq3fXGs7RzaacW5KcLYfmu+y86phm7wxmtvMHlkMv+cOjFGQ8SK867hXXR0WjHYNrBtcU7NGWsNjQ3sHd/LqZOnCAaDi2PN61l1Th3dN0qkOWLFec9OOrutsZadUw3d4M6tO/bxub6hfs05NRCw4nzg8AFq6qw45461Hbt20N7ZzqmTp+y5KBvn7D5aW1fLxOHFOXXqoSlUVbXjHAplxtqkNdYM3SDgD9DUbB2fc+fUzp7OvHPqWvnL4XccJhAIcOrkKfaO77Xzlz0je+6fvzQ1Ls6pOWNNUzWampvQVGtfyp1T18xfdmy342znL5k51dANLr55kUNT1jbWmlOX5y+rzam7h3Zz6uQpqmuq7bEWCq0+p04cnqC2rtaO8/L8xdANYtGYnYetmFOzYy0QWDN/yR7XhvYO0dTSxKmTp+jb3rdkTt25JzPWcubUfPnL8jnV0A2qKquorbU+R+6cmjvWsnMqQEOkwc5T8+UvmqZx6uQpey4CGNs/RqQpM9Zy59TtfYv5S/tinpo7p2b3b0M3VsypBw4fWDGnAuzYvZi/9PStnFMN3eDK5SuMT1jH57Xm1OX5S0tbC6YJrZ0HuDD9EfJx/uLDnD3/US5f+R7mb+6zC8Sh0ALVVSdpjvwP3v3uv2Ns5JcZH/kBfvqnfpl3v/OnGez/Zd71zr/Kqzk+9j/tOTV3rOXG2ef3LclTc8fayNgIc9fmMHRj6Zzav23JWMudU/PmLzlzqqEb1NfXU1Fhzdu5c+rgnsGlc+qgFef75S+GbvDqN19l8sjkYpwnV59Tu3u7AStPHdg1gGqaDFVV8O9mLub9Hn/yjdf4rVdf4afffIPvm77IkfkbbI8uEDIM0sBVn5/X29r5p55e/rCjiz966Fv4xP/1fr5t/ADTNTV5Na+Gw4y+3TqX2ug54bG6Bn79B36Qn/hX/5qHh0a5MLq3JOeEhm7g0Ty0tll5WCHnhLnHtf0H93N5+jKGbhR1Tpg71sy0SVNzk72/FnJOmJu/7Du4z8ohM8eDQs8JO7o77Dm1vr6ecEXYmisLPCds71w6pxq6wd07dxncZR2fCzknHN0/as+pA4MDLNxbwNCNgs8Js8fO7Jxq6AahUMg+5ynknDA3f5k8Msn5s+cxdKOoc8Lq2moOHrbyF6/Ha+c+UNg5YW7+Mjw6zKmTpwiFQ0WdE+bmLy2tLVaRUTcKPie059Tdg7R1tHHq5Ck6uzuLOic8cPiAnb/sGdnDrRu3MHSj4HNCVbPKgtn99dTJU4ztHyvqnDCb3/n9fiaPTHLm9BkM3Sj4nBCW5i+hUIiGxgYM3Sj4nNCeU9vb2Llnp51DFnpOmI3zkjm1yHPCfHNqseeEUw9NEQ5bYy37WXNxZU9igExh+D8Bo8A08NfAh4H3AscBX7ZQrCjKO4G/ME0z7+V/O9o6zO2927fMSmJ/wE9DYwOzM7NoHq0kv3BomkZbRxvTF6et1WPFriT2emiMNHLl8hVUTS3JSmKv10tzazMzl2YAio4zCjQ1N3HtyjXS6XRJVhL7/D4iTREuT19GVdWiVxIbhkFzazNz1+dIJpIl+SXL5/fZ+0/28WJ/ycq3/xSzkljTNBqbGrk6exVFUUqyknjJ/qNQ9EpiTGuivn71OoZhlOTXRL/fT2NTo7X/aGrRK4l1XaelrWXJ/lPsL5bL959i59R0Ok1reyvXr14nkUiUZCWxqqlEmiJcvXIVTIpfSYyVyF+5fMWeL4v91TidTtPc2sz1q9fRdb0kvxovOTZoWtEriQ3dsPefRDxRkmOn1+ulIdLAlctX0DSt6F/uzbRJa3sr165cI5lMlmQlsaIoRJoj1rHBTBe9OkBRFNo625idnkXX9ZKsDrCPDdfmSKVSRecvqqrS3tm+ag5QSI5kGPn3n6JWEns8NDZlcos1VoGs97hmmiatHa1cm71mfw/FriTOHhuuXb1G2kgXvQpEURTaO9u5PH0ZXddLkqfmOzYUs5J4rRyg0JU4aeP+xwZr3zcxaeLOnU4Wot3EEr3cu9dNNNaFYVgnqflJU1fzNKHgRSoqZvD5zhEKXCIcjq3ruHZt7hDnLz5MItmK3zdLZ/vnaWr8+qr/e2U9eeqSHEBVS3I+kh3bM5dmrNWGReaQ77h9kx8+f47WZIJZf4An2zp5qq5+ZZx1g+pEgu5UkvZYlLaFBToTMdpjMVrjMXxrnJ+awIVgkBl/kCvhMNNeHzP+ANcqKpjxeElk/7dLnvzl7bfm+cSr31yh+bH+nTzd3FzwOWGxOWTelcSskVsUcD7iD/hpiDQwOz27Zg65kfxl+f5TqhwyXw5Q6Epij8dDS3sLMxdnbF/Fnvub5tLcotgVpj6fj0hLhMuXLqMoSklWEi/PLYpd3bg8hyzF6saCcou1VhKvI4fcaP7i8XpoaW2x6wvFriQudQ5pmqa1/zTnry8UspI4N4dMJlfmAIXkqbn7z/L6QqF1tvXmFuteSax5NlRf2Ewrif/x2D+WRU9iTNP8JvDu7N+KovwG8P9hrTJOAcNYxWKAEeDlVbXSpvVfOsE6Ec70nTJN0/7SAPtLW34/lVp8TVYHsA/ggBUw0nm3sZquU9vIDuyrV64u+Xx6alF31W0s/34yvhRVsRLTmcuk0+nVvedsw9AXdfNtY7nHtb6r9WzDxCTSHGH28ix6Si86zh6vx/aY9V9InHO3nU6naWpp4srslaWfY5UY3G8buR6zP/gUE+fshJbVLEWcYe39Z13jYNk2FFVZ3H9S999/1rONFftPEXHWU/n3n/XGYLVtGGkj7/6z7jgvG2v59p9i4rzW/lPofOfxeuzPfL/vJzcGa20jd/9Z7diw3jiDtSoru+/c7/tZT5xTqdSS2GTfv94YrDbfLTk2FBnnVHKpx9U+h72NdR7XjPQ69p8NHNfy7T/risEa21gyD9bjkQAAIABJREFUtjPfV6FxBmu1V6QpwuxMZv8pMs7Lx3b2PcXkFqqmri8H2MA27rv/FJK/mOkNHRvuNw48Xg9NzU1cuXwl7/dTSP6y4WPDffIXzaPZsVnz+9lA/lJwbmGk7R/O1psDFBLna3OHuTj9MMnn2vB5F3sI31sIEo3uJBrrZSHaTTTWQzTas+pF5BRFB9KYpm/Fc6HQm+za8QsrHs+xvuY4qKt5ikjj1xkeHealEy+hp3RyPkZB5yOr5QDFnI/kjm2M4nLIqRvX+XhOe4he/R5PnDnF73f2cN3noyMes/7FYrTHY1TmfiHLSCjW/x/25ykWnwuGeHh47YvCrXZc+z81dTzWP8gj0xdpSyWZ8fp4sr2TY3X1UMQ5YbE5ZL5trJlbFHA+YhgGzS3NXJ29WvS5wmr7z1ueQ65jGyjYx1g9pZfk3H95blFMDgmQNtP2sWbJ4xs4J1y+jeW5RTFxzr6nlDkkFJhbrLGN9eSQG81fgFXrC27IISGz/2ygvnC//OV+OUAheeq66gsbzF/Wm1usexvqBnPIEtXZiq2BFZNDLt9GFjevJB4CzmIVhN8HfA74FtM0TyqK8kWgAfhBIAL8PfC4aZp5L1zX3tJu9nb3vjXGBUEQBEEQBEEoOXPzRzh95okVj2vqPYz0aiuDDYKBGULB84RC5wgHzxMKvUkwcIn5W4fy6g32P1aSC8NtCUyTxmSS3z31Eq2JxLrflgau+ANcCgS5FAwyHQhyMXP/ms/P1M0beS8K91j/IMdWu3idIAiCIAgb4tgzx8pjJTHwfcCPAgHgJeA7TNM8mXnuJ7CKxtNADPi91QrEWxFFUaitr+XmjZuU6keAUmuKx63j0QlN8Sge3aRZDh6d0BSP4nGza4rHB+vRNBXiiRYWottYWOhjIdrH/K2DeV9rFYjTBPyzhELnCQfPEQqdJxR8k1DwIqqazPu+hrqnGOx/jIvTjxBPtBLwX6az/cmSFIjd8j2WSlMxTZoTcbpjUbpiMbpjUbqjUbriUcJrrAo2gZcrq6xicKYIfCkQ5HIgSFJd/fI4T9U12Kt+WxNxLvsDmVW/xReIN1tsHoSeE5rl4NEJTfEoHt2kWQ4endB0wmO54tYL12Ga5sdM06w3TTNsmuYh0zS/lvPcHdM0f9A0zUrTNCOmaX5iLa2thsfjYefunfYFF9yoKR63jkcnNMWjeHSTZjl4dEJTPIrHza4pHt86j4YR5M7dXcxe/XZeP/dRXnzlD3j6ua/w/Iv/ldNnnuDizMPcuHkE01x5FW4AVY1zaN972bf3B9k18At0d/4RkYavUhF+Y9UCcZaGumNMjD3Kz/7MJ5kYe7RkK4jdHusj83Mcfek4X/q7r3D0peMcmZ8DQEun6YxFOTI/xw/PXOTx11/lj0+e4H8/93W+9OLz/MZrp/ixi2/ybdevsnPhLmHDQAeSmRYRyzkXCvETu4b5jb5+/qytg2N1DZwPhdcsEGc5VtfAo2MTfPJnfpZHxyZKtoLY7bFxQlM8uldTPIpHN2mWg0cnNJ3wWK64tt1EKZF2E4IgCIIgCILw1jI3f4QL0w+TiLfgD8zS2XaUyvDrLET7uBfts1cJxxPted+vKAnCofOEQ28QDp3l8pXvIZ5oXfG6UOgsY0MPO/1xNg1H5ufytnK44vNRn0rhXeX8MKEoXAqGOB8McSEY5Hzm/kwgyKFb89IeQhAEQRDKjHJqNyEUiKIoBENBYtFYSZfzl1JTPG4dj05oikfx6CbNcvDohKZ4FI+bXVM8Fqd5fe7tvPrG4n/2i0b7ePX1lf1/s/i8c1YxOHzWLgqHgpdQlMVWBn7/tbw9hLvanyzYJ7j7eyxWL9smojcapTe2QE80ylRm1fBymjMXs4mqKheCIfvf+WCI86EQs/4A6VVWDOe2h2hJxJktUXuIzRybt1JTPLpXUzyKRzdploNHJzSd8FiuuLbdhFA4Xq+Xg1MH8Xrz/5c8N2iKx63j0QlN8Sge3aRZDh6d0BSP4nGza4rH9Wum0x7uLfQze/V9vH7up3jh5c/x6hsfX+XVJuHQ60Qa/paert9j9+BPMjH2fibGvovdgz9LT+fniDT8A+HQ+SUFYljsIVwRPofPl6AifK4kF5lzy/e4Gkfm5zh68gT//dg/cfTkCbs1xHJqk0lGb9/ke2dn+NmzZ/jsyy/ylUybiF87c4oPXbrAO29cx7/KCXBcVfnevfv41n2H+PCevXxq2wD/pa2Dr9XVMxMIrlogznKsroEPjx/gNx9/nA+PHyjJCmK3x8YJPSc0xaN7NcWjeHSTZjl4dELTCY/lirSb2KQoilLyX0BKrSke3alXLpri0Z16TmiKR/dqikd36jmhWQ4endAUjytJp70sRHu5tzDAvYV+7i30sxDtxTR969JW1SiT+9/rqEc36JVKc7XWEJ/t6OaOx0NvLEpv1FohXKun8mokFYULwRDnQmHeDIb4F1dnaUkmVrzubCjEw0NjRfmFrRMbJ/Wc0BSP7tUUj+7Uc0JTPLpX0wmP5YC0mxAEQRAEQRAEYQUregi3/imBwCx37/WzEO3n7r0BorGevBeQ82h3qAifsf5VvMaFS48Qi3eteF0gMPtWfJRNgWaaPHrxzbzP/dtL51c8lgYuBwKcC4Y5F8oUhUNhpgNBjJxVwNPBYN7C85PtK+MlCIIgCMLWQdpNbEJ8Ph8PvechfL71reh4EJricet4dEJTPIpHN2mWg0cnNMWjeNzsmlvN4/W5d3D6zBNEo30Y6ZDVQ/iNj/PiK5/j7Pmf4sq197MQ7cc0vXg8t6mpfo721v/Mju2Ps2/k+zkw/j727PwoPV2fpbH+n+ju/KO82ym2fzC4+3ssSNM0qUsmGb91k++/PM0vvvEaf3zyBH/7ja/RGY/nfUsa+EZ1DV9qaeNTvf18aPcI79l3iB8a2cfHBnbyZEc3/6e+kQvB0JICMSz2Dz4XriDh83EuXFGyC8xtutg8AD0nNMWjezXFo3h0k2Y5eHRC0wmP5Yq0m9iEbNXG4OLRnR6d0BSP4tFNmuXg0QlN8SgeN7vmZvZomhBPtHD33iB37+20/+VfP2JQW/08FeHXqKg4Q0X4Nfy+q9ynPS0Ac/NTXJx+hHiihYB/ls72J4vuHwzu+R5X48j8HA/nXMDtaHsnT2UKsL60QXc0Sl90gb7oAr2Z+6u1ikiTPyqlaA2xmffxt1JTPG4dj05oikfx6CbNcvDohOZWvnDd8nYTUiQWBEEQBEEQhE1MSq/g3r3BnKLwICm9dl3v1dQoh0rQQ3irsFoP4ZMVlVTrOu3xGFqe9xnAdCDIuVCYs6Fw5jbE9oV7PPH6qyteX6qVv4IgCIIgbF2WF4ml3cQmxOv1Mvn2yZJfPbKUmuJx63h0QlM8ikc3aZaDRyc0xaN43Oyabvc4N3+EEye/wLMnvsqJk19gbv4IAOm0h7v3Brh85Tt57Y1f5PkX/zPPPP83vPLqp7kw/SPM3zpESq9FUVJUhk/R0vSXDPR9kmDgYt7t+IvsIez277EUeqpp0hVd4F3Xr/HT517P+5qhe3fpyhSIb3k8HK+q5s+bW/m13u18aPcI7913iB8eGeeX+wf5Ynsn/1xXz2wgyFP1jVZriFDYag0RCpesQLwVYvNWaIrHrePRCU3xKB7dpFkOHp3QdMJjuSIriTchiqJQV1/H/I35ki7nL6WmeNw6Hp3QFI/i0U2a5eDRCU3xKB43u6abPc7NH+H0mSdWPB4IXCCRaMY0/Xmem6YyfJrKilNUVpyiIvwGqrrY4mA1zcH+x4pqEeHm77EQPS2dpjsWpX/hXubfAtui/z97Zx7eVnbe5/diJ8AVBElRXMV93zeJy8zIcWzHjhMvsVMnTqOxp3FrO5nUdZNYSTOOVSep48nipK1dV0qdpHVdL4kT723sWLFnlTQjaUQto50jUhIFcYW4APf2DxAQSIISSeCS94Lf73nmGRLL77w43znf/c7R5cEMaar6wPctKAq/VdvIRbcHv93Ous7r2CRjKnkKozCmuqcwCqORPM3AqIenHoxmkRw3IRKJRCKRSCQSmVCqamU2UM2Zc59kYXHtO0lttkky0odjNoWHsdunHuo/7h/k6sjjzM8V4nSNUpakM4SNrPD5wVcpnJtn1OXkSHFZzPnBKnsCszEbwjNUBGZxxlk/BSwWXvWkU3ovQHYwuOr5ZJwhLBKJRCKRSJRMyXETO0A2u43O3k5sdpthPYVx5zDq4SmMwmgkTzMw6uEpjMKY6p5GYAwG3dyd6ObK9cc5eeZPeObFb/LS6c+tuUGsKPN0tf0L+jp/mqa6f09ZyRG8Oc+ta4MYwOc9Sm/nE3z4wwfp7XwiKRvERujHtRQ5P7gyEMCthqgMBDh0fpinz5zkv588zrdf+DH/7fRLfPTyq/zMrTHqZ2dwahozVivHM7P4YmERv1dVyy+2dvJT3fv4UGMrf1RRHbetw8VlCbEauR/19BRGYUx1T2EURiN5moFRD089GM0q6YEUlBpSGbsxhhp68J+9baenMO4cRj08hVEYjeRpBkY9PIVRGFPdczsY5xd8TE03MzXdzORUC7OBSljxNWc26xQaVkIhz6r3p6WNkOZ6TVfG7fZLhqddVakKzPJrly/Gfb5rajL684TNxgVPOuc86Zxf+m/U6UJb47iIH3p9HKyp5/GRa+xemOeGw8nh4tKEzxA2Yj9uhacwCmOqewqjMBrJ0wyMenjqwWhWyXETIpFIJBKJRCKRzhr3D3F15ED0KIfSoiO4064xNd3C1HQTU9MtzM3vXvU+l/MGmRknycw4RWbGSdxp17hzd1CX84NTUVZNozwwS93sDHUz09QtHRlhf8AaaFFR+N3qOs570rnlcG74/GCRSCQSiUQiM0iOm9gBstqsNLQ0YLVZH/7ibfIUxp3DqIenMAqjkTzNwKiHpzAKY6p7JtMv8qVwgUAlIdVNIFDJ2QuHOH7yC7x6+d9xa/yNSxvEITzu8+ze9WXqqn+Xno63093+89RWfZLCgr/H476Komj4vD+kvuYgHvdF7PY5PO6LSdsgNnI/QviIiCMnj/PdY89y5ORxhvzj0ecUTaP0XoCfvH2TX71ykf98+iW+/fyPOXLqBL9x6QI/c2uM2tkZ7JrGhM3GrCU+07W0NP7Z6+OW07XpDWKj96NZPIVRGFPdUxiF0UieZmDUw1MPRrNKjptIRWmwML8AybxJPNmewmhMP7N4CqMx/fTwFEbjegqjMf308DQDox6eCfqpqo3pmXomp9q5fuM9a72KrMwTZGacIivjFBnpr2CzBdbl7/MepSDvx+yp2sPlVy8TCoU2B7pSBuvHWEXOEI6ocnGRQ+eHOZrjxRMKUTs7gydOP8xYrZz3pHPWk8HZ9PD/x5xOhu7eWeYXUaLnBwOG7kdTeQqjMf308DQDox6ewmhMPz08hdG4nnowmlRy3IRIJBKJRCKRSJSgVNXO9Ewdk1PtTE61MTXThKq6HvgeiyVAf88bt4jQ3EoPLvL5UyfYPT//wNfNWSycXzpD+Gx6Bmc96Yy40tY8Q3jQP87jI1cpnJtn1OXkcHFZwucHi0QikUgkEplBctzEDpDVaqWiqgKrNYm38yfZUxh3DqMensIojEbyNAOjHp7CKIyp7vkwP1W1MznVzLWRX+LUmad55sVvcPLMX3B15P1MTHWhqi5stgl83u/jsN+O6+FyjerKaATPzfhZNI2KwCxvuTnKb1w8zxdefpFvvvjsmhvEi4rCH1RU88vNHbypex8famzlM+WVfM+Xz/U095obxABHvT7e397Dv3nPL/L+9p6kbRAboR9TwVMYhTHVPYVRGI3kaQZGPTz1YDSr5LiJFJSiKHjSPShJ/JKNZHsK485h1MNTGIXRSJ5mYNTDUxiFMZU9x/1DXBs5wI+eL8Jhf43S4iN4s5956J3CdttdsjJPkJX5ElmZJ3CnXUVR7p9JvFJlxYcT4jR6P67XL3NxkYaZaRpnpmicnqZ+djrusRELioIjzl9BXktL45v5u3Rl3G5PMzDq4SmMwpjqnsIojEbyNAOjHp56MJpVctyESCQSiUQikUi0pLU2dBVlAU1zLHtsrU3h+L6DXB15nPm5QpyuUcqKDyfli+aMrCH/OAdijnI4UlzGj3Jy2ROYpXF6isaZaRpnpimZu7fqvQGLhbPpGZxOz+RMRgavpGfQOj0V9wzhgzX1ckSESCQSiUQi0QYlx03sAFksFgqLCrFYkhfeZHsK485h1MNTGIXRSJ5mYNTDUxiFMdU8NQ1mA2VcvPzkGs87sNv9+Lz/SGX503S0vJfezp+hvuYpdu/6WzzutTeIIfxFc91t7+MdbztAd9v7krJBbMR+jCjyJXOVgQBuNURlIMCh88N8+/kfcfjUCT5y5SJvHL8V3SC+6krjW758/mhPFQea23lz9z6ebGjh86Xl/Dgnl0m7gx96fRysqeei28M9m42Lbk9SNoiN3I96+ZnFUxiFMdU9hVEYjeRpBkY9PPVgNKukB1JQFquF4tJiLNYkTsIkewrjzmHUw1MYhdFInmZg1MNTGIUxFTwXFrO4Nf46zl/8TZ4/8WWOn/wrFhbjbzhaLPfo7fjZdW8KJ4txK/2S4VkwP8frb9/i31+8EPd5p6YxY7XyQlY2f1lUwkdrG3lLZx/vbevi96tq+XpBIRc96YTW6NyjXh9PdPbw2x/+VZ7oTM4ZwkbsR739zOIpjMKY6p7CKIxG8jQDox6eejCaVXLchEgkEolEIpFoR0hVHUxONzEx2c3diW5mAzXLnrco86CEUFX3qve63RfpbDmwVaimkKJplN0L0DI9RevUJC3TUxQsxP9yuYjuWSy8qXsfqpz7JxKJRCKRSLStkuMmdoAURSHHm5P0g8GT6SmMO4dRD09hFEYjeZqBUQ9PYRRGM3iGj5DYw8jouzg9/CmeefEfOD38J4zc+IXoBrHHfYHiwv9JU92/pa/7zdRWfTKuf6JfMrcWo5H8HuZpVVXqZ6Z5940RPnnuFf7u2LN84eRx/t3lV3n9ndsULMwTBM54Mpiwxf9+7BsuV8IbxGbvRyP4mcVTGIUx1T2FURiN5GkGRj089WA0q2STOAVls9tobmvGZo9fnBvBUxh3DqMensIojEbyNAOjHp7CKIxG8hz3D3Hs5BG+/X+/xIsvf4GLlz/MuVc/xvPHv8rxk/+Dy1c/xN3JXlTVhcM+Tr7v29RWfYLejp+ho+V97Cn7r+Rkv4jVsoDP+0Pqaw7i8VzC4ZjH47lEfc3BpJwhbPTYDPnHOXLyGF/6v9/myMljvO72LTomJ/jlkas8feYU33zxGT57+iU+eO0yA3f9ZAeDzFksHM/M4khRKU/WN/FT3fv4QHMbf1RRHbeNw8VlCXMavR/18DQDox6ewiiMqe4pjMJoJE8zMOrhqQejWSXHTYhEIpFIJBKJTKvb449x9tWPr/m8xTJHVsZLZGe/SE7WC7jTLm/4LOGdoMgXzT1MU1YbpzIzeTkji5MZmZz3pBNc44teBv3jPD5ylcK5eUZdTg4XlyXlDGGRSCQSiUQiUeJaedyEbJOnqFxpLubuzRnaUxiN6WcWT2E0pp8ensJoXE9hNKafHp5GY5xfyOXuRB/+iT7u+AfjvsZuu0tt9cfJyjiNxbKw5Yxb5ZmInycYpGV6kvapSd42Nhr3NUFF4fteHyczMjmZmcWVNDfaOnfZj3p9HPX6Ur4ft8rTDIx6eAqjMf308DQDox6ewmhMPz08hdG4nnowmlFy3EQKyuFw0P9IPw6Hw7CewrhzGPXwFEZhNJKnGRj18BRGYdxKT02zMDXdxJXr7+f4yc/z/PGvceHSb3DH/whrlbOq6iQn6/imN4iN8LmT7ZcWCtJ3188Hrl7ms6dO8A8vPsMfnjvDz4++hlNT475nQbHwieo6/m7Xbi67PeveIN4s43Z4CqNxPYVRGFPdUxiF0UieZmDUw1MPRrNKjptIUVmtVkKhkKE9hdGYfmbxFEZj+unhKYzG9RRGY/rp4bkdjIuLWdyd7MF/t4+7kz0Eg1kxz6pkeM6Sk/Mst27/JHPzxave73ZfpLPlgK6MRvB8kF9aKETz0p3C7ZOT1MxOr/ozwhGXi+OZ2fRO+ClYWL2hftHt5kBLp26MRvEURuN6CqMx/fTwNAOjHp7CaEw/PTyF0bieejCaQXLcxA6RHoM72Z7CaEw/s3gKozH99PAURuN6CqMx/fTwTKbfuH+IqyMHmJ8rxOkapaz4CD7vD9E0hdlANf674WMkpmcaiL1L2GadJjv7ebzZz5CT/TwO+wQAHvclhs8fWtVOWfHhhFmNHJsh/zgHYs77PVJcxvNZOTRPT9E2NUn71AR1M6s3hV9zujiRmcWJzGxeyszittMZ9Yt3JnEyvmjOyP2ol58enmZg1MNTGI3pp4enGRj18BRGY/rp4SmMxvXciRvE8STHTaSgHA4Hr3vj65J+O38yPYVx5zDq4SmMwmgkTzMw6uEpjMK4GY37hxg+f4hAoJKQ6iYQqGT4/CFOnnma545/lROnPs/VkfczPdMEWPC4L1Cy+69oafggfV1vpb764xTkfTe6QQzg8/6Q+pqDpHsu4XDMk+65RH3NQXzeowmxGjk2kQ3dykAAtxqiMhDg0PlhvvXCj/n02dO898Z1mpY2iEedTr6RV8B/rKzhne3d/Iv2bv5TZQ3fy8uPbhAD/NDr42BNPZc86cw7HFzypHOwpj7hL5ozcj/q5aeHpxkY9fAURmFMdU9hFEYjeZqBUQ9PPRjNKjluIgWlKAqedA+zM7MkK77J9hTGncOoh6cwCqORPM3AqIenMArjZnTs5BECgco1n7dYAuRkvUhO9rN4s5/D6by9LZx6+CXDU9E0KgOzfOrsaXIXF+O+ZszhXHan8JjLtaWMevvp4SmMxvUURmFMdU9hFEYjeZqBUQ9PPRjNopXHTcgmsUgkEolEIpFIN2kazAaquOMf5Npr/5J4f8imKIs01n2UrIxTWCzxNz93qnIX5umenKB74i5dkxPkBNfun3sWC2/o3gcb/HI5kUgkEolEItHO08pNYjluIgVlt9vZN7QPu91uWE9h3DmMengKozAaydMMjHp4CqMwPkiaZmFyqoVLVz7Iiy99kROnDnPttQOsVXqmpV0jJ+t4QhvERvjcyfB0hUL03vXzwSuX+MuXj/G148/zsYvnef2d2+QEFwlYLMxarHHfe8PlSniDOFX6cTv99PA0A6MensIojKnuKYzCaCRPMzDq4akHo1llfeqpp7abQXc9/emnn8rJztlujC1VIBAgMBtI6q3yyfYUxp3DqIenMAqjkTzNwKiHpzAKY6xU1c7diW5GRt/DhcsfZfTmO5ieaSIYysBimcOb/WOys15gZrZh1Xur9nwad9q1LeHcLr8h/zj/4dwZfvroUR69NcZdu52raW4UTaMmMMsbb9/kfSNXefLKRd40foummWlygouowFlPOt/KL+DzJWX8aXkVV91u9t8ZX9XGp/dUcS3NnTCrkftRL09hNK6nMApjqnsKozAaydMMjHp46sFoBl0bufbsU0899Z3I73LchEgkEolEIpFoUwqG0rg70ccd/yD+ib2EQp7oczbrNN6cH5HrPUpO1gtYrXMAjPsHuTryOPNzhThdo5QVH074S+aMrsgXza3Uy+mZlM0FyA4Glz0+5nDyQnYOL2Rlczwzm6k4d7YM+sd5fOQqhXPzjLqcHC4uS/hL5kQikUgkEolEO0emOW5CUZQiRVH+VlGUO4qijCuK8iVFUfKWnrMrivLniqLcVRTFryjKZxRFsW03s1Fks9lo727HZktelyTbUxh3DqMensIojEbyNAOjHp7CuDMYx/1DHD95hGePfZfjJ48w7h9iYTGLsVtv5pWzf8CzL36dsxc+zu07P0Eo5MFhH6ew4Gs01f1bejvfSm3VJ/F5j0Y3iAF83qP0djzBBz/4W/R2PJG0DWIjx+bx61fjPt46M0V2MEjAYuWfc7z8SXkl72nt5F3t3Xyqopof5ObF3SAGOOr18URHL7/1wQ/yREdv0jaIjdyPenkKo3E9hVEYU91TGIXRSJ5mYNTDUw9Gs8rIPfAXS/8vAxTgb4A/A/4F8NvAABD5e8VvAR8Dfm+LGQ0pVVMZvzWOqqmG9RTGncOoh6cwCqORPM3AqIenMKY+47h/iOHzh6K/Ly5WLv0eAu6fiZvmuk5uzlFyvT8kI30YRXn4X6kZ+XMny69o7h69E376Ju6y514g7msWFYUn65s5k55ByLLxezd2Qj9uhacwGtdTGIUx1T2FURiN5GkGRj089WA0qwx73ISiKCeBP9A07X8u/f4LwG9pmtakKMp14Nc1Tfvy0nM/B/yRpmll8bzkuAmRSCQSiUSijenFl/6Ke3NxSyvSPeeiG8PutCuJfldaSsihhmifnKR38i59E36K5+Ye+p6LbjcHWjq3gE4kEolEIpFIJFou0xw3ATwN/JyiKFmKomQTvoP47xVFyQGKgZdiXvsSUKooSlY8I8WiRG8bt1gs2OzhnxVFwe64/yd8Dodj2c/K0orHbrdjWbq7w2azYbGGf7ZarVit4TtpLFbLsjYi34qoKMoq3/W0EfG1Wq1YbRtrw+V2UddYh9Vmxe6wR9uw2W0Pb2Nl/yy1YbPbaGhuiL5uJXu8Nqw267I2YtmdaU7qGuuw2W3RGMTrq7j9s0YbDqeD+qZ6rDZrUuJsd9ipa6zD4XQkFOfYGDjTnNQ1hWOznhg8rA2b3UZdYx2uNFdS4qwoyrLxk2icI23EGz/riUF0rq1ow+kKjx+7w57wXIu0sWz8JBjnyLiua6zD6XJuOAZrsbvS7sfmQXF+WL6LtGG1WcPjx+1KSpwVRVnGmIycGmVMc20oBg/KqQ6nIzp+Eo2zxWLBarNS31SPw+lISpztdnt0bjtdzg3HYK18t+zasDKnrjHXHtRG7PhZ63Ns9NoZGT/R9E5yAAAgAElEQVQ2u21DMYi0sXKuRWLjdDk3FoMHtBEZPw6nI+E4K4oSZmyuj8Yt0TivvDZsNAYhdTcjo+/ipdP/Zc0NYovlHh0t/4rS4i+QnfUaVuvGa6S1xk9Em6lfItcGm922oRisbCMSg5XjJ14Miubu8Y6xG3zq3Bm+8cKzfOrcK7xz7AbFc3MEFYWXsrL5r6V7+LOyPXH78nBx2abjHBk/Dc0N0dclo06NzbsbioE1fv3yoBpgs3Xqeq4NG7muJbuGtFjjXxvWisF62lhWAyRpPRIZP7H1WiI15Jq1xTrj/LAaMtE4R9qw2qzUNdXhTIu5NiSwJky0hozXxgNri020sd4aciPXtZXjJ1k1ZLwaYCNrwtg27A77svGTjLV/vGvDZmtIAKfLGR0/m10Trpxr8a4NiaxHVtaQicQ5tt550P7CetaEG60hN1qn2h32ZdeGROKsRw0Jy8fPZteEsW08bH9hM3Xqg/YXNrvPtt7aYr37bBvdX0jWPluyrmsPi8GD2lgpI28S/wjIB+4CfiAH+H0gfen5iZjXRn7OiDVQFOUNiqI8rVgV6pvqAcjflU9HdwcAWdlZ7B3cC4Qn1+D+wWgH9z/aT3pGuKmuvi5y83IBaGprorikGIDqumoqayoBKCkrobGlEYC8/Dw6e8N3hWRkZrDvkX1AeCAN7h+MBmjf0D4yssLIHT0d5BXkAdDY0khJWQkAlTWVVNdWA1BUUkRzWzMAub5cuvZ2AeBJ9zDw2AAQHjD9Q/3hoGuwd2AvWTnhvfP2rnYKCgsAqG+qp2xPePG3p2oPtfW1AOwu3k1rRysA3lwvPf09AKSlpVFYVIiCgsViYXD/YLQw7+3vJSc3B4DWjlYKiwoBqGuoo7yyHICyijLqGusA2LV7Fy1tLaiqSlZ2Fn0DfctiEBnwg/sHcXvC39Ddva8bb64XgJa2FopKigCoqauhojp8l3hxSTE53hzQIL8gn46ecJwzszLZNxSOgdMZbiMyGWLj3NnXiS8vfJ5fU2sTRcVFqKpKRVUFVbVV4TZKi2lqbQLAl+ejsy8c5/SMdPof7Q/H2R6Os9PpjMY5MysTgLaONlwuF2jQ0NxASXk4zhXVFdTU1UTj3NLWEo1B975uANweN4P7B8NxtlqjbaiqSs/eHrJzsqNx3rV7VzgGjXWUVYTjXF5ZTl1DOAaFRYXROOfk5tDb3wuEC8j+oX5UVUVBYXD/IGnuNAB6+nuiMWjtaGV38W4Aautr2VMVXvyW7SmLzrWCwgLau9oByMrKIn9XPmjgcrnCcV5KmAOPDeBJD3/RUdfeLnJ94bnW3NYcjXN17fK5Vt9Yj6qq+PJ80ThnZGVE4+xwOMJxXkqM+x7ZR0ZmeK519naSl796rlVUVoTjpIXHUlNbOM65ebl09XWtirPNZgvHYKlA3Du4l6zs8Fzr6O4gLz8PVVWpra+ldE8psHqutbQvxdnnpWdfz7I4K8rqudbd1x3+TBq0dbZF41zbWBuda+UV5cvmWltnGwDZ3mz6+vuicR7cP4hFsaCqKv1D/dG51rOvB69vaa61t0TjXFN/f66VlpdS3xw/p/bu60VVVZyOpZxqj59To3NtRU6tqgnPtZLSEhpbG0EL54bI51iWU5fmWmQBFTvXYnNqQ0vDspxaVVOFqqrsLtpNc3tzNAbde8NzzZPuWTXXIjHYO7B32VwrKCwADbJzsqNt7KnaQ21DOM6xc21ZTnWnReOsKMvnWm9/Lzk5OaiqSnNbczSn1jbUsqdyaa5VlFHXFI5z7FzLzsmO5tRInCM5tX+oP/yzBt17u6Nxbm5rpqj4fk6trL4/1xpawqc65Rfk09kTznfRnKoBStg3bk7t7cSXH45zY2sjJaXh/qmqrbqfU2Pmmi/PR3tXO6qqkp4eM9fstjVzakd3Rzi3EM6ppeXhuRbNqVq4r5taluazL/eBOdXlCse5b6CPbG84zrFzra6xjtKyUlRVpay8bFlOjYzRHG8OvQP3c+rg/kEsFks0zm730lzrX5prGuTm5lK4eynOMTm1dE9p/Jz6gPpl4LEBPB4PqqrS0d0RrV+a25rXrF/KKvbx2ujPcersZ/nx81/k8tUPMT3TSDjIq+XzTUZz6gPrl7rqaJyj9Uskp2rhft83GJ7PD8qpK+uXtXJqU2sTqqqS482JzjW3e+2c2jvQG64bluIcW7+8w6bw+eMv8Kf/4wh/efI4Q/5xdu3eRXdrE713/Xx09Dr/+9QJ/tdLL/JrVy7Se/cOTk3llsPBsY4OPtHSzls6+/jiB36F77a28eXCYv77m9/Mtaxs5mx2Rrxe/vyx/Rz1+igtL6WheWmuxeTUePXLqpyqgS/fR27u6jo1dq5Fcyrh10fq1Hj1i8PhQFVV+vr7onOts6eT/IKluRabU6sr79cvxffr1Nic6na7w32rrc6pfQN9q3MqUNd0v37ZUxknp2rha33k+vygnLqyflmWU5fmWklZSZhDW51T9w6E59rKnDq4fzBavyzLqe1LOVULxzCSt2Nzal5BXtw61eF0LKtTY+uXts628PzQVuTUmqplcy02p8atX2Jz6tLnzcxcnVPrm+uX59T6cJwfVr+ggaZp9A/1349z/9o5tbyiHAjXqbWN4Tgvq1+ys+/XkPFy6kPql7g5dakebW0PX58TXRNWVleG46QlcU2oha8nkVp4M2vC2Otaz76e8LjSElwTxsw1i2KhsKgw2sZm1oSxc61nb0+4hnQmtiYsKS+J5tS8vLxwzLTNrwmLS1fkVC083huawm1sZk3Y0dMRzal1DXXhXKIlsCZcunZGc6oWHhsFu5Zy6ibWhLFzrX+oP7xBpSW4JszJiuZUp9NJYVFhNM6bWRNG51pBHm2dbaiqSkZGYmvC2PqlqLgo/Fm1za8Jozm1qZ6S0hJUVaW8ojyhNWHfQF+0fmlpawk/riWwJlzanB3cP4g7zY2qqnT2dCa0JozUd06nk/6hcF+hJbAmZEX9kp4evpZom18TRnNqcRGNzY2oqorXm8CakDg5NdE1YZycmuiacHD/IB7P/fplpQx53ISiKBbgEvAl4Kmlh58ChoA3Ed40rtI07eLS66uAC0C2pmmTK/1Kikq06opqgsEgFosFi9VCcDGIoijY7DYWFxaBcBJZWFiI/ry4uIimadjtdkKhEKqqYrPZUDUVNaRGOzkUCmGxWrAolmgbVquVxcXF6L8SxPqupw1N0wiFQuE2FAgFN9eG3WEnuBhE0zRsdhtqSH1wGyv7x2Zbs41Y9mBwdRtWW3gzItKGoigPjMHD2oj2TwJtbEecE4nBw9qIspsozpuKwQPaSHSuGSHOicZgZRuJzrVti3MSYyA5dQMxsNtQ1YfPNSPGWbcYpGKcV8Rgfj6Xm+MDjPv3MzUVXpyEpZKZcZJ83/ex2YOcPf9RVqq54XfJyfqB+eO8RgwenfDze2dfWfW5z7s9lM3dw6neP69uUVE4lZHJi7l5/Cgjk8tpbhxOp2HivKUxsFqwWNZXvxghzkaqX0wVZx1z6poxMHqcZU0oa0JZExoqp8qaUNaEsiZcfwz+8eg/LjtuwqibxD7gNlCiadrI0mMlwDUgDzgBPKlp2leWnnsn8LSmaaXx/HbamcRWq5WSshKuX71OKBQypKcw7hxGPTyFURiN5GkGRj08hdF8jAsLXsb9j3D7zmNMTbcQ+wdlmRkn8Xm/jy/3Bzgdd6KPj/sHuTbyOPMLu3E6blBafBif96hujEbwPPLyMSrX+KI5gJsOJ89l5/Bsdg7HsrK5Z13/90CbefwYxU8PT2E0rqcwCmOqewqjMBrJ0wyMenjqwWgWrTyTeP1V7RZK07RxRVFeBT6oKMrHlx7+IDCy9NwR4KCiKD9aeu5jwOe3g9WIUhSFzOxMlGvJ+xaZZHsK485h1MNTGIXRSJ5mYNTDUxiNyTjuH+LayAF+9HwRDvtr7N71JTTNzu07+5mcagWs0ddmpJ8mL/f7+Lw/wOm8HdfP5z3KrvxnaGhp4MzJMwSDwaRwGi026cEgPRN++u/6qVhjg3hBUXh/cztX0tygbI7b6ONHD09h3DmMengKozCmuqcwCqORPM3AqIenHoxmlSHvJAZQFKUB+GOgi/CtLieAj2iadkJRFDvwJ8B7ll7+18Cva5oWd+Wy0+4kFolEIpFItPM07h9i+PyhB74mw3MGX+738eV+H5fz1haRGVO75ubYN3GHAb+ftulJbA+piS+63Rxo6dwiOpFIJBKJRCKRSF+tvJPYsF9cp2naGU3T3qBpWq6maTmapu3XNO3E0nOLmqZ9cOnxHE3TPrzWBvFOlMVioaCwIPrth0b0FMadw6iHpzAKo5E8zcCoh6cwGotRVR1cuvKh+G0oc+wp/c90t72LtuYPULz7f29og9gM/bgeT0XTqJuZ5n3Xr3D45HG+9NILPHnlEl1TE9g0jctpbv56dzGfLSmL+/7DxfEfTybjdvvp4SmMO4dRD09hFMZU9xRGYTSSpxkY9fDUg9Gskh5IQVmtVsoryqMHfhvRUxh3DqMensIojEbyNAOjHp7CuP2MmqYwOdXKhUsf5dljf8v8wq64r1MUleLdX8TlGttyxu32dKgqfXf9fOTSBb58/Hk+d/ol/uVr16kKzBICTmRk8edle3hPaxf/srWTz5Xu4W+KSjlYU88lt4c5u51Lbg8Ha+o56vXpwmgkPz08hXHnMOrhKYzCmOqewiiMRvI0A6MennowmlWGPW4imZLjJkQikUgkEqWKAvdKuDX+Bm7dfj3zC4XRxxVlEU2zr3q9232RzpYDW4m45Rryj3Ng5CqFc/PcdDo5nplF3sICXZN3catq9HUBi5XnsnP455xcnsvOYcq+ur9EIpFIJBKJRKKdINMcNyHavBRFISs7C2WTX6qyFZ7CuHMY9fAURmE0kqcZGPXwFMatZVxYzOK1sbdz4tRnOfby33D9tV9ifqEQq3WWgrxv0Fz/q9RWfTzue8uKD28J43Z5DvnHOXR+mMpAALcaYs+9AO+4OcrQ3Tu4VZVbDgdfKyjkI3VN/HRXH79bU8/38vIfukFs9M+th58ensK4cxj18BRGYUx1T2EURiN5moFRD089GM0q2SROQdnsNtq62rDZbYb1FMadw6iHpzAKo5E8zcCoh6cw6s8YUh3cvvMYr5z9fZ4//jUuXXmSmdl6IIg3+8fUVT1Fb8fPUlP5h2RnvURe7g+przmIx3MJh2Mej+cS9TUH8XmP6sa4bZ6aRs3sDO+7foXfuXA27kv8Njvvb27nne09/PGeKl7IzmFxA2fNGfJz6+ynh6cw7hxGPTyFURhT3VMYhdFInmZg1MNTD0azSo6bEIlEIpFIJNoGjfuHuDpygPm5QpyuUcqKj5Cbc5TJ6RZu3X4D4/5HCYXSo69P9wyT7/sueb7/h8M+sY3k2yOLptE8PcmQ/w4D/jsULsw/8PUBi5U39uzbIjqRSCQSiUQikchcWnnchGyTp6gcTgcL8wuG9hRGY/qZxVMYjemnh6cwGtdTGDevcf8Qw+cPRX8PBCoZPn8Im81PMOiNPu50jJHv+y75ed/DnXZ1Sxn19Fuvp0NV6Zy8y5D/Dv1375AdDEafu2ex8Hx2DnUz0xQsrPYZdTm3hHG7PYXRmH56eJqBUQ9PYTSmnx6eZmDUw1MYjemnh6cwGtdTD0YzSo6bSEE5HA4GHxvE4XAY1lMYdw6jHp7CKIxG8jQDox6ewpiY59WR+F8kFwx6sVpnKMj7B5rrf5Xu9ndTXvr5jW0Qm7wf3cEgrxu/xVPnh/n6i8/yh+fO8ObbN8kOBpm02fiWL5+P1TTw1s4+fqemgc+UV8Zt43BxmW6MRvEURmFMdU9hFMZU9xRGYTSSpxkY9fDUg9GskuMmUlR2u53FxUVDewqjMf3M4imMxvTTw1MYjespjBuXpsHkdCunh/8YTVv9B10WZZ6+7jdjtSR2J4OR+3HIP86BkasUzs8z6nRypLiMU+mZDNy9w6D/Dp1TE9hj6tNbDgf/nJPLD70+Xs7IJBTnXOFB/ziPx3geLi7jqNeXMKvRxs9W+OnhKYzG9DOLpzAa008PTzMw6uEpjMb008NTGI3rqQejGbTyuAnZJBaJRCKRSCTSWQsLXm7efiM3b7+Ze3Mla77O7b5IZ0v8u4xTQUP+cQ6dH171uMryP2+76krjqDeXo14fZz3paPJt0yKRSCQSiUQiUVK1cpNYjptIQTkcDh59/aNJv50/mZ7CuHMY9fAURmE0kqcZGPXwFMaHe2qaFf/dPs6cO8Rzx7/Mlesf4N5cCVZLgOzMF+K+p6z48JYybrXfE9euxH3cAgx70vlcSRnvbe3kvW1dfK50D8PpGeveIDby59bLUxiFMdU9hVEYU91TGIXRSJ5mYNTDUw9Gs0ruJE5BKYpCRmYG01PTJCu+yfYUxp3DqIenMAqjkTzNwKiHpzCu7Tk3V8jY7Z/i5u03sbCQH308I/00u/L/gbzc72O13mPcP8i1kceZmy/E5RyltPgwPu/RLWHcSr+C+TkeuzPOY3duUz87E/c19ywW3tDTv62cevvp4SmMwpjqnsIojKnuKYzCaCRPMzDq4akHo1kkx02IRCKRSCQSJVmq6mDcP8DN229mYrI7+rjNNkGB7zsU5H8Dj/vK9gFusfLm53nUH94YbpqZjj6uAfHuC77odnOgpXPL+EQikUgkEolEop0uOW5iB8hut9M30IfdbjespzDuHEY9PIVRGI3kaQZGPTx3IuO4f4jjJ/+SZ499j+Mn/5Jx/xCzgT1cvPJhnjv+Fc69+lR0gzg76wXqqn+X3o53UFH+F2tuEKdSP3oXFnj72A0+88rLfOXE83z46iWaZqZRgRMZWXy6vJI/qKiO+97DxWVbxrldfnp4CqMwprqnMApjqnsKozAaydMMjHp46sFoVlmfeuqp7WbQXU9/+umncrJzthtjSzU/P8/szGxSb5VPtqcw7hxGPTyFURiN5GkGRj08dxLjuH+I4fOHWFzMQVVtLC7mMH5nP6M338b0TCOq6sLhuMXuXf+H2srfp6jwK3jcV1AUdcsY9fRcyy9rcYE3jN/iX1+7zK9ducjeibvsWpgH4GRGJv9rdxH/qaKGrxbu5lx6Bq960nnV7aH83j3SNI1rrjQ+vaeKo16frpxG8dPDUxiFMdU9hVEYU91TGIXRSJ5mYNTDUw9GM+jayLVnn3rqqe9EfpfjJkQikUgkEokeomMnjxAIVMZ5RiPX+0N25f0DOdkvrGtT2Ewa8o9zYOQqhXPzjLqcHCku40RmFkP+O+y/c5v2yQlsMa9/JT2Df8zN45+8Pm45ndvGLRKJRCKRSCQSiR4sOW5iB8hms9Ha2YrNZnv4i7fJUxh3DqMensIojEbyNAOjHp47hVFV7dwafz2BQHnc5y2WORpqfgdvznOb2iA2cj8O+cc5dH6YykAAtxqiMhDg0Plhvv7is/zGpQt0L20Qn/Wk819Ky3lXWzf/uqmN/1NY9NANYiN/br389PAURmFMdU9hFMZU9xRGYTSSpxkY9fDUg9Gskh5IQamayoR/AlVL3t1MyfYUxp3DqIenMAqjkTzNwKiHZ6ozzs/nM3rzrYzd+mkWg2sfWeVy3UgE0dD9+Pj1q3EftwIX3B6+n+vj+7l5vOZK2zZGPT2FURiN5GkGRj08hVEYU91TGIXRSJ5mYNTDUw9Gs0qOmxCJRCKRSCQCNA0mpjoYHXs7d+72E94OBYfjFlkZJ7l95ydWvae+5iA+79EtJtVPiqbRMj3JT47f5i23xlDivOaexcIbevq3nE0kEolEIpFIJBIlT3LcxA6Q1Wqltr4Wq9VqWE9h3DmMengKozAaydMMjHp4phJjMOjmxtjbOXbyC5we/hPu3B0CrGRlHqe++nfobns3ddW/R33NQTzui9htc3jcF5OyQWyUfiwPzPKvrl3mSyde4DNnTvHTa2wQA9xwubaFcas9hVEYjeRpBkY9PIVRGFPdUxiF0UieZmDUw1MPRrNKjptIUSmWtZZ2xvEURmP6mcVTGI3pp4enMBrX0+yMs4EyRm++nVu330BIdQNgtQTIz/sOhQVfw+O+suz1Pu9RCvJ+THVdNRfOXiAUCunOqKdn7sI8PzF+m9eP36ImMBt9fMZq5QdeH2MOJ+9/7dqq9x0uLtsyxu32FEZj+unhKYzG9RRGY/rp4WkGRj08hdGYfnp4CqNxPfVgNKPkuAmRSCQSiUQ7Rppm5Y5/gBs338bkVEf08TTXVXbv+hr5vm9jswW2kVBfuYNBhu7e4fXjt+icnIj+SdmiovBstpfv+vJ4JieXBUv4mUH/OI+PXKVwbp5Rl5PDxWUc9fq27wOIRCKRSCQSiUSipEiOm9gBslgtlJaXYrEmL7zJ9hTGncOoh6cwCqORPM3AqIen0RnH/UMcP3mEZ178DsdPHmHs5k9xbeSXeP7E/2b4wieWNohD5Hr/ieb6J+lsfS+7d331oRvERu/HIf84R04e5zsvPsORk8cZ8o9jVVX23r3Df7hwlr89/hwfu3ie7qUN4pMZmfzRnire1tHLwdoG/ik3L7pBDHDU6+N97d38q3f9PO9r707aBrHRx48efnp4CqMwprqnMApjqnsKozAaydMMjHp46sFoVkkPpKAsigWvz4tFSeIkTLKnMO4cRj08hVEYjeRpBkY9PI3MOO4fYvj8IWYDlQSDacwGKrlw+Te5OvJ+Fhbysdv9lBR9gZ72d9NQ8ztkZx1HWedfmBm5H4f84xw6P0xlYJa0YJDKwCyHzg/zDy8+wx+eO8NP3LmNS1W56krjv5WU8e62bj7U2MrXCwqZstu3hNFMnsIojEbyNAOjHp7CKIyp7imMwmgkTzMw6uGpB6NZJcdNiEQikUgkSikde/kIgXuVqx63WOaorvgUPu8PsFgWt4FMXx05eYzKQPw7oe/Y7fy/3Dy+68vnvCedde+Ki0QikUgkEolEopSUHDexA2SxWMgvyMdiSeK/1CTZUxh3DqMensIojEbyNAOjHp5GZAwG0xm58fME7u2J+7yCSr7vewltEBuxHx1qiNeN36J8jQ3iOYuFd3b08ufllZxPz9jwBrERY70VnsIojEbyNAOjHp7CKIyp7imMwmgkTzMw6uGpB6NZJT2QgrJarVTWVGK1Wg3rKYw7h1EPT2EURiN5moFRD08jMc7N7eLilQ/x/PEvc/nav2Gt8sbpGt02xqT7aRo1M9P8+uVX+dqx5/ndV8+xlsNrLhehBO4cNlKst9JTGIXRSJ5mYNTDUxiFMdU9hVEYjeRpBkY9PPVgNKvkuAmRSCQSiUSm1NR0Pa+Nvptx/yOwtEXqTrtEVuYJRm++Y9Xr62sO4vMe3WLK5CpzcZGfHL/FT92+SVVgNvr4mMPJmfQM9vvHV73nYE190r5wTiQSiUQikUgkEqWG5LiJHSBFUcjIzEBJ4nmDyfYUxp3DqIenMAqjkTzNwKiH53YxapqFcf8QL5/+C15+5bOM+/cDVrKznqep7iN0tPwyVXv+lPqag3jcF7FaA3jcF5O2Qbwd/WjRNHom/Dx1fpivHn+OX716iarALPOKwvdy8/j1+ibe3d7NUzX1HKyp56LbQ8Bq5aLbk5QNYjOMRz08hVEYjeRpBkY9PIVRGFPdUxiF0UieZmDUw1MPRrNKNolTUHa7na6+LuwP+Kby7fYUxp3DqIenMAqjkTzNwKiH51YzhkJp3Bh7Oy++9NcMnz/E1EwzirJIQd436Wj5ZZrr/x052S9Ej9v1eY/S1/UrHDz4h/R1/UrS7iDeyn7cPXeP91+/wpdOPM8fnX2F/f5xHJrGOU86T5dX8rbOXj5RXcexrBy0pQ9+1OvjV7r6+MODB/mVrr6k3EFshvGoh6cwCqORPM3AqIenMApjqnsKozAaydMMjHp46sFoVslxEyKRSCQSiQyr+YVcboy9g7GbbyUYygTAZpuksODv2F3wNRyOO9tMuHkN+cc5MHKVwrl5Rl1OjhSX8VxWDo/4x/mp2zfpmJqMvnbSZuN7vny+mVfAq570baQWiUQikUgkEolEqaCVx03YthNGpJ9sdhvBxaChPYXRmH5m8RRGY/rp4SmMxvVMpt+4f4irIweYny/E6RylwPctZgNV3L7zOjQt/K/6LtcIRbu+REHet7Fa57acMZmeQ/5xDp0fjv5eGQhw6PwwcxYLLlUFQAVeyMrhG/kF/Cgnl8UNfOOykWNtJk9hNKafHp7CaFxPYTSmnx6eZmDUw1MYjemnh6cwGtdTD0YzSo6bSEE5HA4eed0jOBwOw3oK485h1MNTGIXRSJ5mYNTDM5l+4/4hhs8fIhCoJBRyEwhUcvnah7g1/kY0zU5mxss01HyMrtZfZPeuv133BrGR+/HAyNW4j7tUldecLj5fXMa72nv4aH0TP8jN29AGsZFjbSZPYRRGI3magVEPT2EUxlT3FEZhNJKnGRj18NSD0ayS4yZSVE6nk/n5eUN7CqMx/cziKYzG9NPDUxiN65ksv2MvHyFwr3LV41brNM31HyEj/eymvQ3Xj5pG08wUf3bmFLY4NdicxcIbuvdFzxjeFsYt8DOLpzAa008PT2E0rqcwGtNPD08zMOrhKYzG9NPDUxiN66kHoxm08rgJ2SQWiUQikUi0LVJVB2O33sTFK79OvD9usloC7Ot549aD6SBPMMgbxm/x1pujVNwLrPm6i243B1o6t5BMJBKJRCKRSCQS7USt3CSW4yZSUHaHnUde9wh2RxK/PTLJnsK4cxj18BRGYTSSpxkY9fBMxC8USmPkxrt54cQXuXjlI6xVjjhdo9vGmBRPTaN+ZprfuHierx1/jievXKTiXoA5i4VjmVlx33K4uGxrGbfBzyyewiiMRvI0A6MensIojKnuKYzCaCRPMzDq4akHo1kldxKnoBRFISs7i8mJSZIV32R7CuPOYdTDUxiF0UieZmDUw3MzfovBdG6MvdoHMqEAACAASURBVIMbY+8kGAxvkjodN8nOep6bt3961evraw7i8x7dUsZkeKaFgrx+/DZvvTlKTWA2+vilNDdfLyjku758Zmw2Bv3jPD5yjcL5OUadLg4Xl3LU69sSxu30M4unMAqjkTzNwKiHpzAKY6p7CqMwGsnTDIx6eOrBaBaZ4rgJRVFmVjzkBIY1TWtZet4O/DHwC4AG/A3w65qmxf0qwp22SSwSiUQikZG0sJDDa6PvZvTmzxJS3QCkua5TvPtvyPd9F4slyLh/kKsjjzM/V4jTNUpZ8eGENoi3QzWzM7z15ig/MX4btxoCYF5R+H5uHl8v2MXp9ExI8KxhkUgkEolEIpFIJEqGTHHchKZp6bH/AcPAF2Ne8tvAANAANAKDwMe2ntSYstlt9OzrwWa3GdZTGHcOox6ewiiMRvI0A6Menuvxm5vP59XLv8YLJ77EyOh7CKluPO5Xqat6is7W97Ir/5tYLOF/3/V5j9Lb+QRPPvkf6O18IikbxFvRj65QiDffGuOzp07w+VMneOutMdxqiKuuND5TVsE7Onr5ZFUtpzOy4m4Qp0qsU9FTGIXRSJ5mYNTDUxiFMdU9hVEYjeRpBkY9PPVgNKsMuUkcK0VReghvBv9lzMOPA4c0TRvVNG0U+I/A+9b0sCjYbOFgWyyWaOAVRVl25ojD4Vj2s7K0mLPb7Vgs4a6y2WxYrOGfrVYrVqs17Gu1LGvDbrdH21jpu542Ir5WqxWrbWNtWK1Wrl25hhpSsTvs0TZsdtvD21jZP0ttaKrGayOvoYbUuOzx2rDarMvaiGW3WCxcu3INTdWiMYjXV3H7Z402FBRGro2ghtSkxBng2pVrKCgJxTk2BhaLhetXr6OG1HXF4GFtaKrGtSvXsFqtSYmzoijLxk+icY60oakar11fPn7WE4PoXFvRhkUJjx80Ep5rkTaWjZ8E42yz20ALjx+LYtlwDNZit1qs98fPA+L8sHwXaUMNqavGTyJxjjd+NhTnODlVDalcv3odq8W6oRg8KKcqihIeP5BwnC0WC2pIZeTaCApKUuJst9ujc9uiWDYcg7Xy3bJrw8qcusZce1AbseNn5ecI3Cvm1cu/xYsvfZHRm+9A1ZxkpJ+hofY36Wp7gl0F/4SirJ5rVos1em3YSAzixdlqs0bHj0WxbCwGa7Tx2MRdPn/seT7xZ3/KF154lk+dPc1Xjz/Hb1y6QP3sDIuKwv/LzePfNrfz3tZOvry7mDmPZ804K4qCGlJ57fpr4b+LSkKcl10bLNak1C9qSH1gDbCZNtYaP5uJc8Q3cm3QVG3dc+1B+S4ytyPj52E5dT1tRK4NkXyZSJzjjZ9k1Knxrg3rioE1fv0Sr4ZMJM42+/quDRu5riW7hrRY418b1orBetpYVgMkaT0SGT+aqm04BlF2+zpqi3XG+WE1ZKJxjrQRvTZYYq4NCawJE60h47WhhlRGrq9RW2yiDat1fTXkRuqXyLUhMn6SVUPGqwE2siaMbQONZeMnGWv/lbVFIjUkgEWxRMfPZteEK+davGtDIuuRlTVkInGOxGBTtcUD2lhPDbnR+gVg5Pr9a0MicbbZbCgo9/cXElwrRPcXlLX3F9a7JoxtI7aGTEacI69ba39hs3XqemuL9e6zbXR/IVn7bMm6rj0sBg9qY6UMv0lMePP3W5qm3QBQFCUHKAZeinnNS0CpoijLvgVGUZQ3KIrytGJVqG+qByB/Vz4d3R0AZGVnsXdwLwBOl5PB/YPRDu5/tJ/0jHQAuvq6yM3LBaCprYnikmIAquuqqaypBKCkrITGlkYA8vLz6OwNfzN5RmYG+x7ZB4QH0uD+wWiA9g3tIyMrA4COng7yCvIAaGxppKSsBIDKmkqqa6sBKCopormtGYBcXy5de7sA8KR7GHhsAAgPmIFHB5jwT6CqKnsH9pKVE+6W9q52CgoLAKhvqqdsT/jLcfZU7aG2vhaA3cW7ae1oBcCb66WnvyfaP3UNdWiahsViYXD/IK40FwC9/b3k5OYA0NrRSmFRIQB1DXWUV5YDUFZRRl1jHQC7du+itaOVsRtjZGZl0jfQtywGkQE/uH8Qtyf8Z8nd+7rx5noBaGlroaikCICauhoqqsNHiRSVFuHL86GqKvkF+XT0hOOcmZXJvqFwDJzOcBuRyRAb586+Tnx54bMgm1qb2F28m7EbY1RUV1BVWwVAcWkxTa1NAPjyfHT2heOcnpFO/6P94Tjbw3F2Op3ROGdmZQLQ1tWGpmqoqkpDcwMl5eE4V1RXUFNXE41zS1tLNAbd+7oBcHvcDO4fDMfZag1/DoedsRtj9OzrITsnOxrnXbt3hWPQWEdZRTjO5ZXl1DWEY1BYVBiNc05uDr39vQC40lwMPDrArbFbaJrG4P5B0txpAPT090Rj0NrRyu7i3QDU1teyp2pPOM57yqJzraCwgPaudgAysjKorK5EVVVcLlc4zksJc+CxATzp4U2Urr1d5PrCc625rTka5+ra5XOtvqmesRtj5OblRuOckZURjbPD4Yj2D8C+R/aRkRmea529neTlr55re6r2kJmViaqqFJcU09QWjnNuXi5dfV2r4myz2cJxdoXjvHdwL1nZ4bnW0d2BL9/H2I0xahtqKd1TGm0jdq61tC/F2eelZ1/PsjgrirJqrvXs62F+bh5VVWnrbIvGubaxNjrXyivKl821ts42ALK92fT190XjPLh/EBQYuzHGwKMD0bnWs68Hr29prrW3RONcU39/rpWWl1LfHD+n9vX3MXZjLDoPIheqlTk1OtdW5NSqmvBcKyktobG1EVVVCYVCtHcvjaXYnLrUhsN5P6dG5lpsTm1oaViWUyurKxm7MUbh7kKa25ujMejeG55rnnTPqrkWicHegb3L5lpBYQGqqpLry102lmobwnGOnWuxOTXNnRaNs6Ioy+Zab38vWTlZjN0Yo7m9OZpTaxtq2VO5NNcqyqhrCsc5dq5l52RHc2okzpGcOvDoADNTM6iqSvfe7micm9uaKSq+n1Mrq+/PtYaWhnCcC/Lp7Annu0hOVVUV/x0/A48ORHNqZd3buXDpExx7+a8ZvfkmNM1GVuYxfvZnPsNb3/L75Ob8mOq6qvs5NWau+fJ8dPR0MHZjDLfHfX+u2W1r5tSO7g7yd+WH49zcQGl5eK5Fcqqqqlht1vvz2Zf7wJzqcoXj3DfQR7Y3HOfIXBvyj/Pxs6epCMySFgxSPHeP3om7pIdCjHnS+fuBQd7e0cNn+vbhessbQVGiMbBYLNE4u91Lc60/PNdUVWVX4a778zkmp5buKY2bUx9Uvww8NkCaO42xG2N09HRE65fmtuaH1y8Fefdzasxcs9ls1DXURduIzakPrF/qqqNxjtYvSzlVVVVmpmfof+ThOXVl/bJWTm1ua2bsxhjZ3uzoXHO7186pvQO95HhzonFeWb+oqoorzUVNffj6vCqnRuaay/XA+iVyXWtpb2FX0S7GboxRVVO1LKc2NC/NtZicGq9+WZlTVVWltLw0en2OzalVtffnWiSnAvjyfdE6NV79YrPbGLsxRt9AX3SudfZ0kl+wNNdic2p15f36pfh+nRqbU9PS0qhrqAvPxxU5tW+gb1VOBahrul+/7KlcnVNVVWV+fv5+Gw/IqSvrl2U5dWmulZSVkOvLRVXVVTl170B4rq3MqYP7B6P1y7Kc2h7OqaqqkpmVuayNSE6NnWuxcXY4Hcvq1Ni51t7dTigUQlVVGlsbKSkNx6CqpmrZXIvNqXHrl5icqqoqVTVV9+dzTE6tb65fnlOX5sHD6hdVVbl18xYDjw7cj3P/6pwK4Tq1vKIcCNeptY3hOMfOtazsLCqqKqLzcVVOfUj9Ei+nqqqKpmnL2khkTVhZUxmdj8laE6qqSl5eXrQW3syaMPa61ruvl3v37qGqakJrwti5BuF8GanDNrMmjJ1rPft6wjXkUv9sdk1YUl4Szam+PB9Fpffn42bWhMWly3OqqoZvxIptY6Nrwo6ejmhOrWsM96GqqpteE0aunZGcqqoqu4t3R+fzZtaEsXNt4NEBpianUFU1oTVhVk5WNKc6HA7qGupQLOGxtJk1YWz90tbVxtiNMTwZnoTWhLH1y+6i3eQX5Idr/U2uCaM5tame4rJixm6MUVZRltCasG+gL1q/tHS0hP9hQFU3vSaMbM5G2hi7MUZXX1dCa8J9g/fn2sCjA4zfHkdV1U2vCWF5/eJJ91BdWx2ej5tcE0ZzanERja2N0Rpys2vCSJyX5dQE14Txcmqia8LB/YN4PPfrl5Uy5JnEESmK4gFuAL+kadrfLT1WAlwD8jRNG196LA+4BZRomjay0qekqESrrqgmGAyG7zSxWgguBlEUBZvdxuLCIhBOIgsLC9GfFxcX0bTw3SiRosxms6FqavRf+gBCoVD0Xx8ibVitVhYXF6P/ShDru542NE0jFAqF21AgFFx/G2nuNKpqqxg+NYxiUQguBtE0LXrnxQPbWNk/NhuLi4vY7DYaWho48/IZgsHgKvZgcHUbVpsVNKJtKIoSZXc4HVTXVXP29FlQYHEhfl/F7Z812nA4HdQ11nHm5BlUVU04zhaLhdrGWs6fOU8wGNx0nO0OezQGrjQX1XXVDJ8aBnhoDB7Whqqp1DfV8+rZV5mbm0s4zsrSpkZk/FgsloTiHGnDbrdT31zPmZP3x8+G5tqKNux2O9X11Zx75RyapiU01yJtOJ1Oahtrw+NHUxOKs81uw6IsjZ/h8wQXgxuKwVrzIC0tjaq6cGxQWDPO8WIQOw8ibaiqSn1zPa+ee5W5e3MJx1lRFFwuV5TRYrEknFMhvFC9cPYCc/fm1h2DB+VUm81GTUMN5145h6qqCcU5cmdBQ0sDZ185y8L8QsJxjixY6prquDB8gYWFhQ3FYK18t+zaoCjLc6oaf66t1YZ/8lEuX/klFhaLcDpukOv9LtPTzfgn+qPXXp/3x5QU/TXpntPrvnbGXhsURVl3DNbKdwpKePwMX2Bubm79MVjRRt5sgP968jg5wcWVJQY3nE5+oaMXxWbdcJyDwSBWqzU8fk6fZWFhIeE4OxwONE0Lj5+zF1iYX0i4frHZbDS0rl0DbKZG0jQtmn/uBe4lFOfIXHM4HFTXL9UWsO4YrJXvFEWhoaWBc2fOMT83/9Ccup42oteGmNpis3FeXFyMjp/h08MsLiwmpU6NXhs2WltYw3/5sLJ+sdvt1LfULxs/icTZYrWARpTx3r17CcVZjxoyUkc0tDRw7pVzzM/Pr4pzbAzW08bKGjIZ6xGbLVzfD58ajtb7idSQa9YW64zzw2rIUCiUUJxj725eWVsksiZMtIaM18YDa4tNrEfWW0NupH6JXhtOnonGKik15OlhLMo6ash1tOFwOKhrqouOn2Ss/RVleW2RSA2pqipOl5PahvD40TRtU2vClfXLytoikTjD6hoykThHYrCp2mKd+wtr1ZAbrV9ix0/kHzo3G+fInbK1DUs1QCiY0FohEgOXy0V1ffz9hfWuCTeyv7CZOjV2/KzcX9jsPtt6a4v17rM57Es15CtnQVu7hky0ftlsnBONwYPmwT8e/cdlZxIb/cCNnwMCwDdiHot8qV0WMB7zM8B0PBNN1QgGw2ceqmq4YyE8gSKdBkQ7beXPi4v3XxPxgXBQIlJDKipq3DbW8tWtjfkFZqZm0DSN4MJ9r+BizM9rtbGyf5a4NFVjemI6+k2Pa7LHtBEK3vdd2cbiwmL0jrbIcw/qq3W1EQwxPRlmTEacrVYrM1Mz0UUcbC7OsW1HPnckcazVPxHfh7URYVxYWIjGJqE4a9qy8ZNonGM/RyQ2K/tnXfMgThszUzOEQqF1jZ/1tBEMBu+PnwTjHCnOZqZmohvED+qfVTF4wOeIjp+YPll3nBeWtxEdP/P3x08icY74xhs/m813EcaV7HF9Y+faQ9pYOX42G2cILxamJ6ejfZRonCMbPzNTMywuLkbfv94YrJnvYq8NwfjXhvW0cfvOIMPnPx7DW8HM7Acir8SX+31Kd/8VHs+l+22s87oW79qwqn/ixGCtNmL7Mdo/64nBwiJoGp1TE7xtbJT+u3e4/4eHy5W9GCSkabDEspE4R14zPTl9P08kGueYub24cH/8JFJbaNo6a4ANtBGbf+L2zwbiHPFdXExubWG1WpmenI4+97Ccup424tUWm41z5DXTk9PRP7dMRp266doipEY5VtUAa42fTcR52fiJ6Qcj1ZAAijV8bYj8nkicYZ015Abrl+j4ieSJRGvItWqLTcY58nMkNgnHeeW1IWYOJ7ImTLSGjNfGA2uLTbaRUA0Z57oWvTaoSa4hVY2FUOJxjvwcO36SsfZfWVskUkNGfo4wbnZNuLKNlbVFInGG5NWQK6+dG64tNri/sNJ3o/VL7PhJOM7BIFYtzv7CJtcKsT9vZH/hYfXLw2qATdWp69lf2GD9st7aYr1tLLI0fkLrrCETuK5tJs4Jx2AdbURk9DuJ/xn4Z03TfnPF49eBJzVN+8rS7+8EntY0rTSeT3FhsVZRXqE7r0gkEolEqa5jLx8hcK9y1eM26xStTR/AnbbqD3pMJXcwyBvHb/G2sRuUzd2LPh6wWHAvFV6xuuh2c6ClcysRRSKRSCQSiUQikShhHX326LI7iQ17JrGiKLXAPuC/x3n6CHBQUZRdiqLsAj4GfH4r+Ywsq9VKdV31sgPXjeYpjDuHUQ9PYRRGI3magTFZnhNTrQTulcd9TtNsCW8Qb2dsyu4FePLyq3z1+PM8eeUiZXP3CFisfKWgkPe2dvLJqtq47ztcXLZljNvlaQZGPTyFURiN5GkGRj08hVEYU91TGIXRSJ5mYNTDUw9Gs8rIx028DziqadqFOM99AsgFhpd+/2vgk1sFZngpS9+grRjYUxiN6WcWT2E0pp8ensJoCM+p6SauXn+ciamuNV/jdI0mALekLY6NVdPYe/cObx8bpWtqIvr4VVcaX921m+/48gksfXnR1TQ3B2vqeXzkGkWLC7xmd3C4uJSjXp+ujIbwNAOjHp7CaEw/PTyF0biewmhMPz08zcCoh6cwGtNPD09hNK6nHowmlaGPm0iW5LgJkUgkEok2rumZOq5ef5y7k+Fvx1WURbIyjzMx2bvqtfU1B/F5j2414qaUtbjAW27d5GdujrJrYR6AEPCjnFy+uquQ45nZoEiVKBKJRCKRSCQSiVJXpjluQrR5WSwWikuLsViSF95kewrjzmHUw1MYhdFInmZg3KjnzGwVr5z7JC+d/tzSBnGQXfl/T1frL9Bc/1Hqaw7icV/EZruHx30xaRvEyfzcQ/5xjpw8zndefIYjJ48z5B+nbmaaj716jq8cf55fuX6FXQvzTNhs/NXuEt7d3s1v1zZwPCvngRvEO3H8mIFRD09hFEYjeZqBUQ9PYRTGVPcURmE0kqcZGPXw1IPRrJIeSEFZrBbyd+VjsSZxEibZUxh3DqMensIojEbyNAPjej1nA+WcOf97nDh1GP/dASBEvu/bdLX9ItUVn8LlGgPA5z1KT+cTfPDf/BY9nU8k7Q7iZH3uIf84h84PUxmYJS0YpDIwy6Hzw3zu9Eu8cfwWDk1j2JPOf6ys4Z0dvfy30nJuOV1byqiXnx6eZmDUw1MYhdFInmZg1MNTGIXx/7P33vFxXfed9nOmYTAAiN57IQqJRgAECZKgIJqJnXViJ+tUF1mSLTlO4hY7ZU3LUWy99sZF2XWRHMciHUeO10VO9t34TWJvyi4TR7EkSqQlCqTYCZESBYIdbe7Mef+4mOEAGLSZucC9g9/z+egjEjP8ngf3d8+55x5cnEn3THEURztlOsHRikwrHJ2KbDchCIIgCOuc8Ylqzo3czWuXX4f58+MwxYX/SE3V1wlknl9rvRVz4MgzNI6Pz/t6GPhhUQl/XVbBi9k5qy8mCIIgCIIgCIJgE2S7iXWAUorC4kJUCvdTTHWmOK4fRysyxVEc7ZTpBMeFMicmKzh24qM8c/gbvHb5ZwAXhQX/Qk/nPbRu/OSiC8R2/b5bb96gdmIi7muTLhefampJaoF4PZ4/TnC0IlMcxdFOmU5wtCJTHMUx3TPFURztlOkERysyrXB0KrJInIZ4PB5aN7XimflEdjtmiuP6cbQiUxzF0U6ZTnCcmzk5Vcrxk7/P0889zqXRNwBuCvL/lS0d97Kp+eNkBU6vueNKcGvN0OXX+PLzh/nq88/hWeC3pC76l7elhBWOq5VnRaYTHK3IFEdxtFOmExytyBRHcUz3THEURztlOsHRikwrHJ2KbDchCIIgCGnO6Nhuzo7cw+RkBS7XNIaRBZiToPy8J6mt2k9O9vDaSiZAthHk5y+9ylteuUDp9BQAQaV4PjuHLTeuz3v/vuY2DhYUrbamIAiCIAiCIAiC7ZDtJtYBSimysrNS/jh/KjPFcf04WpEpjuJop0y7O46O7ebF4w8xPt5IOJyJYeQCHgKZJ+ja/Fu0t/5+QgvEa/l910yM86HTJ3ji0E/4rXOnKZ2e4orHy4HKGn51Sz8f2NzFvuY2TgayGHe7ORnIStkC8Xo7f6zIc0qmOIqjnTKd4GhFpjiKY7pniqM42inTCY5WZFrh6FRkkTgN8Xq9bN+1Ha/Xa9tMcVw/jlZkiqM42inTzo5GKJMTp383/otKsSHn+YSzV/371pqtV6/wmeHnefzwM/zSqxfJDIc5Ecji0w0b+ZWefg5U13LZ5wPgYEER7+nbzmcfeID39G1P2RPE6+n8sSrPKZniKI52ynSCoxWZ4iiO6Z4pjuJop0wnOFqRaYWjU5HtJgRBEAQhzQiHPbxy6Rc49/I7CQYL4r7H7RpnR/8bVtls5WSEQvzs6CV++ZUL1E+MAxAGfpxfwHfLKnl2Qy7IT/0FQRAEQRAEQRBWxNztJmRX5jTF5XYRDoVtnSmO9sxzSqY42jPPikxxXD5aK0Yv38mZ8+9mcqoKAKWm0do3770Z/otr4rgQu8dGuWfkLOWTU1z0Z/BEaQXlU1O86dJFcg0DgHGXmx+UlPJEWQUX/Jmr7mhVpjjaN1Mc7ZlnRaY42jdTHO2ZZ0WmExytyBRHe+ZZkSmO9s20wtGJyHYTaYjP5+POn7kTn2/+ooBdMsVx/ThakSmO4minTLs4Xr3Ww3PPf4XhEw8yOVWF13uZxrrP09L4UNz311btX3XHhdg9NspDx1+kcXycQDhE4/g4v3/6BO+4cJ5cw+BChp8v1jbwlp5+vljXuOwFYrvUZjXzrMh0gqMVmeIojnbKdIKjFZniKI7pnimO4minTCc4WpFphaNTke0m0hClFP5MP5MTk6SqvqnOFMf142hFpjiKo50y19rx5q0mzpx7D1eubQPMbSSqKv6KyvLv4nZPADA6Nsi5kXuZnCrHn3GRmqr9FBUcXDXHpThw+BkaZ7aSiGXC5eKTTS38OL+QcAJbSqx1bdYiz4pMJzhakSmO4minTCc4WpEpjuKY7pniKI52ynSCoxWZVjg6hbnbTcgisSAIgiA4kMnJcs6MvIvXRn8WAKWClJf+T6orv4HPe3WN7ZaHNxzmZ0cv8XunXor7q03jLjdv6N+x6l6CIAiCIAiCIAjpztxFYtluIg3x+rwM3jmI15fCT49McaY4rh9HKzLFURztlLnajsFgLifPvI+nDz8eXSAuLvwRvV3voLHuCwsuENvp+w4YBr9+YYT/8exT/MECC8QAF/0Za+a4WpniaN9McRRHO2U6wdGKTHEUx3TPFEdxtFOmExytyLTC0am4H3zwwbV2sJyHP//wg/l5+WutsWrosOb69euM35r/q7t2yRTH9eNoRaY4iqOdMlfLMRTyM3LxNxh+6Y+5fqMbcJOX+xRtzX9ERdlf4/XcWFXPRPLyp6d5x8vn+fiJYXZdHSMrHOKSz8f/zS9k4/itee//fH0T5zIDq+q42pniaN9McRRHO2U6wdGKTHEUx3TPFEdxtFOmExytyLTC0SmcGzn35IMPPvgPkb/LdhOCIAiCYGPCYTevvvZGzo7cQzBYCEB21jHqqv+M/Lyn19hueVROTvDrF0Z4w2uvkjEz7zjrz+SvKqr4UVEJhsvF4Ngo946cpXxyiov+DPZX1XKwoGiNzQVBEARBEARBENIT2W5iHeDxeujb3ofH67FtpjiuH0crMsVRHO2Umeq80bHdHDpygCef+SE/efY7PPXsdzhx+iMEg4X4M16mpelButvvX/EC8Vp83823bvLg8Rd5/LmnefOlV8jQmuezc/hocxt3dfXydyVlGC5zKnKwoIj7erfxsfe/n/t6t6VkgXg9nj9WZDrB0YpMcRRHO2U6wdGKTHEUx3TPFEdxtFOmExytyLTC0anIdhNpitaaG9dvpPSTGVOdKY7rx9GKTHEURztlpipvdGw3Lx5/iGCwgHDYQyiUTSichdt1k4baR2hu/K9kZ51EqbX1XDRPa3quX+Ujp0/w3nNnqJ8YxwU8mZfPnzQ0s7+qlnOBLBb6JuxaGyszxdG+meIojnbKdIKjFZniKI7pnimO4minTCc4WpFphaMTkO0mBEEQBMGmPP3cXzIxWTvv64HMU/R23b36QivApTW7x0Z564URWm/dBMAA/qmomL8qr+ZUVtbaCgqCIAiCIAiCIAhRZLuJdYDH46G9qx2PJ4WP86c4UxzXj6MVmeIojnbKTEWeEcrk9Ln7mZisifv61FRZwtkRUvl97x4b5cCRQ/zwmSf5+pFD/O6pl3j88NN84qVhWm/dZNLl4onSct7avZWHmlqXvUBsx9pYnSmO9s0UR3G0U6YTHK3IFEdxTPdMcRRHO2U6wdGKTCscnYocgTREa83E+ETKH+VPZaY4rh9HKzLFURztlJlMntaKV197A2fO3x/9ULp4ZPgvJqM401Zqvu/dY6M8dPzF6N8bgkEaxm8BcN3t4YmyCr5fVs41r2/NHK3KsyJTHO2bKY7iaKdMJzhakSmO4pjumeIojnbKdIKjFZlWODoVIwmn7wAAIABJREFU2W5CEARBENaAa9c7OXX2fdy81QKA3z9CUf6/MHLx7fPe29a8j6KCg6utGJe/OPwM9RPj874+6vXxtu4+JtzuNbASBEEQBEEQBEEQVoJsN7EOcLvdNDY34k7hjXqqM8Vx/ThakSmO4minzJXmTU6W8eLxBzly9EvcvNWC232T+pov09t5F/W1X6WteR9ZgZN4PBNkBU6mbIE42e+7cHqK3z5ziro4C8QAgVAo6QXita7NWmSKo30zxVEc7ZTpBEcrMsVRHNM9UxzF0U6ZTnC0ItMKR6ci202kIUop/Jl+1AKfHG+HTHFcP45WZIqjONopc7l5oVAm5y+8jZELv4bWGUCIspK/pbb6MXzeq9H3FRUcpKzk32nZ3MKxF45hGMaqes6lbHKSt144z3967VV8i/z20UV/RrKKcv6kiPXoaEWmOIqjnTKd4GhFpjiKY7pniqM42inTCY5WZFrh6FRWvN2EUsoD5AFXtdapuXO1GNluQhAEQVgrtFZcGv1Zzpy7n+lgMQC5Gw7RWPsFsrJOrbHdwlRPjPP2l8/zM6OXoj9RPpyzgWc35HL3y+fnvX9fcxsHC4pWV1IQBEEQBEEQBEFIiBVvN6GU2qqUelAp9c9KqSvAFPAqMKWUujLz9QeVUlst9BZWgMvloqKqApcrdbuJpDpTHNePoxWZ4iiOdspcLO/6jXaee/4rHD+5j+lgMf6Ml2lr3kdH2wcXXSBey++76dZNHjz+In95+Bl+bmaB+KncPN63qZP3be5if3Ud+5rbOBnIYsLj4WQgK2ULxHL+pIb16GhFpjiKo50yneBoRaY4imO6Z4qjONop0wmOVmRa4ehU4h4BpZRbKfUepdSzwJPAx4E7gFzgJnABuDXz9ztmXn9SKXVIKXX/zNPGwhrhcs+c4O4UdsIUZ4rj+nG0IlMcxdFOmfHypqZKGH7pAQ6/8Ag3b7Xhdt+iruZRervuoqjgIEv9JtNafN+bblzn08MvsP+nz7JnbBQX8K/5BbynvZsPt3VweENu9L0HC4q4r7effb/zPu7r7U/ZE8Ry/qSG9ehoRaY4iqOdMp3gaEWmOIpjumeKozjaKdMJjlZkWuHoVOZtN6GU+mXgU0ATcAP4G+BfgJ8Aw1rrUMx7PUAr0A/cCbwZyAZOAP9Fa/2E9d/C0sh2E4IgCIIVjI7t5uzIPUxNlpPhv0h1+eNMTNYwcvE3CIf9QJiykh9QW/U1fL4ra607H63Zcv0ad718jt7r1wAIA/9UWMzjldWcCmStrZ8gCIIgCIIgCIJgCcvZbuI7mE8J/wZQqrV+p9b6gNb6hdgFYgCttaG1fl5rvV9r/Q6gBHgbMD6TI6wBSikKCgtSvjF4KjPFcf04WpEpjuJoh8zRsd28ePwhxscbCYUDjI83cuzkH3Hu5XsIh/1syHmOLR33sbHhsyteILb8+9aa7VfGeOSFw/z3F39K7/VrGErxg+JS3t7Vyyc2ti65QGzn2liVZ0WmONo3UxzF0U6ZTnC0IlMcxTHdM8VRHO2U6QRHKzKtcHQq8RaJf0lrvUVr/W2t9eRKwrTWk1rrb2mtu4H/nBpFYaV4PB42d27G40ndrh+pzhTH9eNoRaY4iqMdMs+O3BP360oFadv4AJ2b3k921ksJZaf6+949NsqBw8/w7R/9Hd859B/8j2ef4jPHXqD95g2mlOL7peX8Rncff9LYzEhmYE0crcgUx/XjaEWmOIqjnTKd4GhFpjiKY7pniqM42inTCY5WZFrh6FTmbTeRjsh2E4IgCEKq+bef/D3h8PwFVZdrnJ39b1gDo/jsHhvloeMvzvv6lFJ8v6yC75RXcdnnWwMzQRAEQRAEQRAEYa1YznYTgsNRSpEZyEz54/ypzBTH9eNoRaY4iuNaZmqteOXSGwmHM+K+7vdfTEYPSN337dKa9505Gfe1C/5MHq1tSHiB2I61sTrPikxxtG+mOIqjnTKd4GhFpjiKY7pniqM42inTCY5WZFrh6FQSWiRWSvmVUvcqpT6nlPqMUuoupZQ8hmQTvF4vO3bvwOv12jZTHNePoxWZ4iiOa5V542Yzh194hJdO/QHgjvue2qr9SRom/30rrRm6/BoHjhyidHo67ntKp6aSUbRdbVYjz4pMcbRvpjiKo50yneBoRaY4imO6Z4qjONop0wmOVmRa4ehUVrzdhFKqE/hboCrmyxo4Dfyc1jqxDRgtZD1uN+FyuQiHw7bOFEd75jklUxztmWdFph0cg0YOZ8+/m4uvvhlw4XbforZqPz7vJc5duJupyXIy/BeprdpPUcHBNXEEQGt2XbnMvSPnaBq/BUCI+MvZJwMB7unsXX3HVc4UR3vmOSVTHO2ZZ0WmONo3UxztmWdFphMcrcgUR3vmWZEpjvbNtMLRCczdbiKRXZn/DJgC7gR+AmQBbwc+B3wB+LkUeApJYsVe06nOFEd75jklUxztmWdF5lo6aq149bWf4/S538Qw8gAoLvohDTWP4vNdnvn7/0Eptbbft9Zsv3qFe0fO0nrrJgA33W6+W1bJiN/PAyePz/sn+6tqV9dxjTLF0Z55TskUR3vmWZEpjvbNFEd75lmR6QRHKzLF0Z55VmSKo30z18PntS2HedtNKKU2LfTmmS0ltgH/RWv9f7TWE1rrUa31fwP+HthtnaqwXHw+H3tevwdfCj+IKNWZ4rh+HK3IFEdxXI3Mm7c2cviFL/PSqT/EMPIIZJ6iY9P7aG16KLpAvNaOaE3f1Ss88sJhPnPsBVpv3WTc5eIbFdX82patHKiu5UfFpexrbuNUVjZTPh+nsrLZ19zGwYKi1XFcw0xxXD+OVmSKozjaKdMJjlZkiqM4pnumOIqjnTKd4GhFphWOTmXedhNKqWnMJ4If1FrfnPOaC5gAPqS1fmTOaz8BGrTWyd11zs58E/AJYCNwDfiE1vorSqkNwFeAn5/x+ZLW+pML5ay37SaUUgSyAozfGk/ZT0NSnSmO68fRikxxFEcrMw0jmzPRrSXcuF3j1FTvp6L0CVyukC0cAbqvXeVdI2fpunEdgEmXi++XVvCtikqueedPcNKhNmudZ0WmONo3UxzF0U6ZTnC0IlMcxTHdM8VRHO2U6QRHKzKtcHQKc7ebiLdI/IfAPuA68Pta62/Oef27wOuBzwBPAwHgrcB/Bh7TWt+XClGl1BuAr2FuZXEQ2ACUaq2HlVJ/AZQCvw6UAP8b+JjW+hvxstbbIrEgCIKwcrRWXBp9PafPvpegkQ9AceGPqK99hIyYJ4fXmvYb13nX+TP0Xr8GwJRS/L+l5Xyzopox+em3IAiCIAiCIAiCsAzmLhLP225Ca/1fgTbgX4G/VEr9H6VUR8xb3gP8C+YTvj8Avoe5QPw94EMpdP0k5pPD/6K1Dmmtr8wsEAcwF4c/prW+qrU+DnwReNdCQcql8HjM7ZddLhcer/lnpRRe3+1PL4x9tNzn86GUAsxPOnS5zEPl8Xhwuc0/u91u3G7z44BcbtesNiKfiqiUmpe7nDYiuW63G7dnZW0EsgLsvGMnXq8Xr88bbcPj9SzdxtzjM9OG1+eNZsZzj9eG2+Oe1Uase2ZmJjvv2InP54vWIN6xint8Fmgjw58RdUxFnTMyzDy/359UnWNrEPm+vV7vsmqwVBuRugQCgZTUOfITtIhjsnWOtOHz+eadP8upQbSvzWkjchwzMjKS7muRNvx+/+3zJ8k6e7wx50+mf8U1WMg9ELhdm8XqvNR4F2nD6505f7ICKamzUmqWYyrG1IhjZiBzRTVYbEyN1DojIyP6fYxPbOSnL36J4yc/StDIJxA4S0fbB2nd+Elysm8sWOfI+bHzjp1k+DNSUufIsdt5x04yMzOj38fmiVt8bvh5HnnhML3XrxFUir8uLedtPdt4dGMLYzMZC413s64Nc8fUBfraYudS7PmTbJ0jbWQGbl8bltvXFhvvoudP5u3zZ1k1WKSNyLXG7/cnXefIeyPn40prEO/4RI5dpN+kYv4ScVxoDpBIGwudP4nUOZI7a26xghrMbSNSg4ijP9O/aA1W0ka8uUWidY49f3wZvqTrHMmN1malcwt3/PnLYnOAROepsY7J1hlSP4d0uW9fG/x+f/T7WKgGy2kjMj5G5pCpuB+Jnj8zf092Drng3GKZdV5qDplsnSNtxL02JHFPmOwcMl4bi84tEmhjuXPIlVzXoteGmWOXsjmkb5lzyGW0ERlzo9fcFNz7zz1/Er1XiLThz7x9/iR6Tzi3r82dWyRTZ5g/h0ymzrHny4rnFoutLyxjDrnS+Uvs+ZNsna2YQ8Li6wvLvSeMbWOp9YVE5qmLrS8kOk9d7txiuets0TlkxuJzyOXUOVX3/qnqa3Pd57Yxl3mLxABa6xGt9a8BrwMKgWeUUv9dKZWrtR7TWr8JaAF+EXgz0KS1/tW521MkilIqC+gFKpVSx5VSryilvquUKp9p1wc8F/NPngM64+S8Xin1sHIr2trbACgpK6Fnaw8AuXm5DAwOAObkcHDPYPQA7xzaSXZONgB92/soLC4EoL27narqKgA2tm6ksbkRgOraajZ3bgaguKSY3m3mJ8bnbMhhxx07APNEGtwzGC3Qjt07yMnNAaCnv4fi0mIANnduprq2GoDG5kY2tmwEoLK6ko5uc72+sKiQvoE+ALKys9h15y7APGEGBgc4eeIkhmEwsGuA3PxcALb0baG0vBSAtvY2auvNDy2qb6qnpa0FgIqqCrp6ugAoKCygf2c/YJ5I/kw/oVAIl8vF4J7B6E3Ttp3byC80n7rr6umivLIcgNZNrdQ11gFQ21BL6+ZWAMoqymjvbmf46DDZOdls37V9Vg0iJ/zgnsHozePWHVspKCwAoLO7k8rqSgCaW5tp2Gg+JV5RWcHU1BSGYVBSWkJPv1nnDbkb2LHbrEFGhtlGpDPE1rl3ey9FxeZuKe1d7ZRVlDF8dJjahlqaWpoAqKqpor2rHYCi4iJ6t5t1zs7JZufQTrPOXrPOkYvyjt072JC7wXTv6eTVV17FMAw2dWyius6sc8PGBppbm6N17uzujNZg646tgDm4Du4ZNOvsdkfP1+Gjw/Ru7yUvPy9a57KKMrMGm1upbTDrXNdYR+smswblleXROucX5rNt5zbAnKQMDA5w7MVjhEIhBvcMRhfn+nf2R2vQ1dNFRVUFAC1tLdQ31Zt1rq+N9rXS8lK29G0BzHPU4/VgGAZ+v9+s88yAuevOXWRlZwHQN9BHYZHZ1zq6O6J13tgyu681tzUzfHSY/ML8aJ1zcnOidfb5fGadZwbGHXfsIGeD2dd6t/VSXDK/r9U21HLz5k0Mw6Cquor2brPOhcWF9G3vm1dnj8dj1nlm8j4wOEBuntnXerb2UFBUwPDRYTa2bqSmvgaY39c6t8zUuaiA/h39s+qslJrX13q39TJyfgTDMOju7Y7WuWVzS7Sv1TXUzepr3b3dAOQV5LF95/ZonQf3DBLWYYaPDjMwOBDta/07+ikomulrWzqjdW5uu93XaupqaOuIP6ZuHdjK8NHh2+eoN/6YGu1rc8bUpmazr1XXVLO5azOGYfDKxVei5+usMXWmr0UWR2L7WuyYuqlz06wxtb6pnuGjw5SWl9K8aSsnz7yfZw5/lWvXO3C5xmnZ+Bi/+7vfIC/3UPT7iNRgYNfArL5WWl6KYRhMTk5Gz9f6pnpaNpl1ju1rsWNqZiAzWmel1Ky+dm9RHl979ik+/rnPceDIId518xqfPvYCjx4+RP/VKxhK8S9NG/njt76NP61vwl1XE+1refl50TE1UufImDowOMCZ02cwDIOtA1ujde7o7qCy6vaY2rjxdl/b1Gl+TEFJaQm9/eZ4FxlTDcPgxEsnGBgciD+mbuulqMSs8+auzVTXmDVoamm6PabG9LWi4iK6e7sZPjps3jBF+prXs+CY2rO1h5KyErPOHZuoqTP7WmRMNQyD0ddGo+drYVHhomNqZDFn+67t5BWYdY7ta62bW6mqrmL46DBVNVWzxtRIX8svyGfbrttj6uCeQVwuV7TOkUWt/p1mXzMMA8MwKC4zz9fYMbWmvibumLrY/GXXnbvI8GcwfHSY7r7u6Pylo7tj6flLafHtMTWmrymXwp/pR7nMCWXsmLro/KV1Y7TO0fnLzJhqGAZnT5+Nnq+Ljalz5y8LjambOjcxfHSYDbkbon0tEFh4TN22axv5BfnROs+dvxiGwdWrV6Pj0rwxNdLX/P5F5y+R61rnlk5KykoYPjpMw8aGWWPqpo6ZvhYzpsabv8wdUw3DAEV0XIodU2P7WmRMBSgqKYrOU+PNX1xuF8NHh9m6Y2u0r/X291JSOtPXYsfUjY235y9Vt+epBUUFbB0w+5ovw5xDGoYxb0zdvmv7vDEVoLX99vylvnH+mGoYBi+PvBw9XxcbU+fOXyJ1btl0u69VVlcyOTmJYRiz+lpefh4Du8y+NndMHdwzGJ2/zBpTt5hjqmEY3Lp1K3q+xo6psX0tts6+DN+seWpsX+vq6eKVi69gGMbsMbW5aVZfix1T485fYsZUwzDw+rzR8zV2TG3raJs9praZdV5q/mIYBseHjzMwOHC7zjsXHlPrGuoAc57astmsc2xfy96QjdvjNueQ8cbUJeYv8cZUwzB49ZVX6dhinq/J3hPWNdZx48YNDMNI2T2hYRhMTU1Fz9dE7gljr2t9A32MnDPnkMncE8b2tXA4jD/THz1fE7knjO1rvdt7GT46HL0eJHpPWF1XHR1T8wvz0VpjGEbC94RVNbPHVMMwGLs8Fj1fE7kn7OnviY6pzW3NXBm7gmEYCd8TRq6dkTHVMAzC4XD0fE3knjC2rw0MDnD61GkMw0jqnjA3Pzc6pno8HvyZfjTmb7Ynck8YO3/p3NLJ8NFhMgOZSd0Txs5fSitKCQaDGIaR8D1hdExtb6OyqpLho8NU11UndU+4fdf26Pxlc9dmXrv0GoZhJHxPGFmcjZyvw0eH2bJ1S1L3hDsGb/e1gcEBXjr2EoZhJHxPCLPnL/6AH1+Gz7yOJXhPGB1TqyppbW9l+Ohw9H4WzDF13r3CIveEkTrPGlOTuCdcaExt2dRCfeNMX2uopbXdrPPc+ctC94SDewbJyro9f5nLvO0m5r1BKTfwAeDjwBTwB1rrry/6j5JEKVUFnAeOAG8CLmPuQVwOPAj8ndY6O+b9W4F/11p74uVVV1brjQ0bMQwDl8uFy+3CCBoopfB4PQSng4A5iExPT0f/HAwG0Vrj9XoJhUKEw2E8Hg9hHSYcCkcPcigUwuV24VKuaBtut5tgMBj9KUFs7nLa0FoTCoXMNhSEjMTa8Pq8GEEDrTUer4dwKLx4G3OPj8ezYBux7oYxvw23xw2aaBtKqUVrsFQb0eOTRBtrUedkarBUG1F3B9U5oRos0kayfc0OdU62BnPbSLavrVmdU1iDhcbUsatDnDn3TiYny/F4bhAKBTBC5mSxpOifqav+En7/6JrVefDyazx0/MV4lzJCwI9KSjlQUc2lrOzU1cDrIRxeuq/Zsc6W9bV0H1MtnL84qs6rNH9xap1XXAO3C5drefMXJ9Y5oRosc/7iqDpbOKYuWAO711nuCeWeUO4JbTWmrtf5i9wTLv+ecD2NqUvV4J8O/tPi203MZWarh4eBVuAfgMeUUj9WSm1Z6t8mQeSJ5C9orc/OPKH8R8CdQBgIKKViF4RzgRssgA6bPz0ECIfDGEHzz1rr6EEDogct8ufIAnowGCQcDgOYP5ULmX8OhUKEQuaHGIVD4VltBIPBaBtzc5fTRiQ3FAoRMlbWRliH6enviZ4UkTaMoLF0G3OPz0wbbo+b9u726BMsc93jtREyQrPaiHUH86c6bo87WoN4xyru8VmgDeVSdPZ0Rhc8kq2zy+Wip78H5VJJ1Tm2BgDdfd3Rp2qXqsFSbbg9bnr6e9Bap6TOWutZ50+ydY604fa46ejumHX+LKcG0b42pw2Uef643K6k+1qkDeVSdPV2RQfgZOpsBA1cbvP8QbHiGizkrtFs2brFPH8WqfNS412kDY/XQ09/D2EdTkmdI+dh7PmzojrHGVM9Xg/dfd3RpwyWW4N4Y+ro2G5eGP4Et8YbCYUDTE2XYoRy8Hkv0d72IVqa/oiMjNdWVOfITUpnT2f0Kctk6nzPyFniccPt5h1dvXyqoZmL/sxl1Xmue1iHo+fPvDF1gb62WBux58/c7yNeDZZz7dTo6LVhJTWItDG3r0XOnwjL7WuLtaFc6va1IYn+HOlrHq+Hzi2d0ac4VlKDeMdn1rUBnZL5i8frWXQOkEgbHq+HLVu3zDt/IiQyf4GYucUKajC3jUgNPF4PXb1dYHbtJcfU5bQRuTYopZKuc+T86djSET1/UjFPjfTtFc8tQvHnL/HmkBESnadGzp/Frg0rua6leg4ZDpnXhq6erlnXhoVqsJw2NHrWHDIV9yMer4eO7o7ob3clO4dccG6xzDovOIfcYs4hk61zpI2414Yk7gmTnUPGa2PRuUUCbWi9vDnkSq5rkWtD5PxJ1Rwy3hxgJfeEsW243C46t3RGz59U3Pt7vB7zqcyYa0Mi9wqRNlBEz59E7wnn9rW5c4tk6gzz55DJ1DlSg4TmFou0sZw55ErnLy6Xa9a1IZk6WzGHjLDQ+sJy7wlj24idQ6aizsCi6wuJzlOXO7dY7jrbStcXUrXOlqrr2lI1WKyNuSy6SKyU8iilCmcaeUVrfRewG8gEfqKUelQplb9YRiJora8C5xZ4+adAEOiK+Vr3zNcFzJP80iuXop3MjpniuH4crcgUR3FMhLPn7437dY/nBvm5zyScmyrHwukpaiYm4r7m1jCSGYj72nKxc22syhTH9eNoRaY4iqOdMp3gaEWmOIpjumeKozjaKdMJjlZkWuHoVOJuN6GU+lXgY5gfYOcCJjA/pO6jWuuTSikX8F7MD68LA/u01l9NqZhS+4BfAd4IjGFuN1Ghtf4ZpdQ3gCLgN4AS4H8DD2itvxEvq6q8SjfUNaRSTxAEQXAQV65u5fnhzxLvZ6Nu1zg7+t+w+lIzZBtB3nphhF9+5QL+cPyJyclAgHs6e1fZTBAEQRAEQRAEQUhXDj55cPHtJpRS7wW+hbn4+hjwGeBHwC8B/6GUqtRah7XWXwaagb8BHlVK/UQp1Z9C1/8K/CNwGHN/4gDwjpnXfge4BowA/wY8ttAC8XrE7XGzqWNT9Fd/7JgpjuvH0YpMcRTH5TIdzGP4xMd4fvjzLPTLMxn+i0kYJu6YEQrxtpfP8+1nn+btF0bwh8Oc9WfGfe/+qtqkHJPxXK08KzLFcf04WpEpjuJop0wnOFqRKY7imO6Z4iiOdsp0gqMVmVY4OpV4H/T2EcxF2S1a6yuRLyqlfhH4PnAv8EkArfVl4D6l1FeBLwM/XiBzxWitQ8CHZ/6b+9p1zKeIhXjomT1HFv9MwrXNFEd75jklUxztmWdFZoJ5WsOrr/0cp8/9FoaRC4TJz3uSK1d3zHtvbdX+VXV0h8O88bVXuXvkHEVBc3+oU5kBvlpTx4/zChi8cpl7R85ROT3Fy74M9lfVcLCgKDnHBDxXPc+KTHG0Z55TMsXRnnlWZIqjfTPF0Z55VmQ6wdGKTHG0Z54VmeJo30wrHB3KvO0mlFKTwA+01m+Z8/UczKd3/0xr/d64YUq9S2v9mFWyiSLbTQiCIKwfxieqOHH6I1y73gNAVuAlNjZ8lpzsYUbHBjk7ci9Tk+Vk+C9SW7WfooKDq+KltGbP5dd41/mzVE1NAnAxI4OvVdXxj0XFhJVaFQ9BEARBEARBEARBWHK7CeAF4HVKqa2RLyilFOYTvRo4ulC4HReI1yNut5v6pnrc7hQ+zp/iTHFcP45WZIqjOMYjHPZwbuQuDh05wLXrPbhck9TXPEJ3+/3kZA8DUFRwkP4t7+Ztb30v/VvenZIF4iUdtWbblTG+9tNn+aMTx6iammTM6+W/1TXytq4+flRcMm+BON1qs1aZ4rh+HK3IFEdxtFOmExytyBRHcUz3THEURztlOsHRikwrHJ3KQttN/H/Ak0qpY8AVoAEoBY5g7lMs2BilFNnZ2agUPpWW6kxxXD+OVmSKozjO5dqNdk6c+j3GJ+oByM/9D5rqH8YfZ7/h1XRsv3Gd+8+dpvvGdQBuut18q6KK75VVMrHIJCSdarOWmeK4fhytyBRHcbRTphMcrcgUR3FM90xxFEc7ZTrB0YpMKxydyrztJgCUUnXA7wFdQD5wDnPh+Kta66lV9EsJst2EIAhCemIY2Zw+dz+vXPpFALyeKzTUfYHiwn9kLa/x9eO3uO/8GXZdGQNgSim+X1bBNyuque71rp2YIAiCIAiCIAiCILC87SbQWp/RWv+21nqX1nqz1vrntNZfdOIC8XrE5XJRVlGGyxW3vLbIFMf142hFpjiKo9bw2uUhnj78l9EF4tLiv6W3++2UFC2+QJxKx91joxw4coh/ePrfOXDkEG9+5QIfPXGMA0cOsevKGCHgf5WU8bburTxa27DsBWIn18ZOmeK4fhytyBRHcbRTphMcrcgUR3FM90xxFEc7ZTrB0YpMKxydihyBNMTldlFTV4PLncJOmOJMcVw/jlZkiuP6dpycKuHosU8z/NInCAYLyfSfo6Pt/TQ3fgav58aqOe4eG+Wh4y/SOH6LTMOgcfwWHz5zkjeMXsIF/HNBEe/s6uWzDRu5lJGxomyn1sZumeK4fhytyBRHcbRTphMcrcgUR3FM90xxFEc7ZTrB0YpMKxydyrztJpRSRVrr0aSDU5STCmS7CUEQBOejtYsLr7yFM+ffRTgcQKkg1RWPU135TVyu6VX3OXDkGRrHx+d9fdzl4gObOjmWnbPqToIgCIIgCIIgCIKwHJaz3cRppdSnlVJFiTSglCpRSn0GOJ2opJAcSiny8vNSvjF4KjPFcf04WpEpjuvDcXRsN4eOHODHT/09Tz33Vzz93OOcOvs+wuEAG3I6he/vAAAgAElEQVQOs6XjXmqrD6x4gTgVjm6tqZ6YWKiFpBeI7V4bK/KsyBTH9eNoRaY4iqOdMp3gaEWmOIpjumeKozjaKdMJjlZkWuHoVOItEn8X80PrRpRSf6OU+jWlVNliIUqpcqXUbyil/hdwHvhd4Dup1xWWg8frobOnE4/XY9tMcVw/jlZkimP6O46O7ebF4w9xa7yRUCjA5GQVk1NVuFwTNNV/hs5N7ycrcHb1HbVm+5UxHjtyCF+cD34FuOhf2dYS8bBzbazKsyJTHNePoxWZ4iiOdsp0gqMVmeIojumeKY7iaKdMJzhakWmFo1OZt90EgFKqA3gIeCMQWUq/ABwDrgA3gBygAGgBymfeEwb+F/CA1voFS81XgGw3IQiC4CyeOXKA8fHGeV/P9J+hr/uuNTCCjbdu8t6zp+m7fhUwt5UIhMPz3revuY2DBQn9Mo4gCIIgCIIgCIIgrArL2W4CrfVPtdZvBuqATwCHgDJgD/AW4O6Z/98JlABPA38M1Gut/7OdFojXKxkpeJLN6kxxtGeeUzLF0Z55qcgMhTKZmKiJ+9r0dElS2RFW4lgyNcVHTxzjz3/6LH3XrzKtFN8qr+SXe/rZ19zGyUCAcbebk4FASheI7Vgbq/OsyBRHe+Y5JVMc7ZlnRaY42jdTHO2ZZ0WmExytyBRHe+ZZkSmO9s20wtGJLPrRfVrrEa31H2uttwJ5wDbgF4C3z/x/G5Cntd6mtf6E1nrEcmNhSXw+H7uGduHz+WybKY7rx9GKTHFMX8er13p45sjX0dob9/UM/8Vk9IDlOwYMg/vOneGbzz3NG0Yv4QJ+VFjMO7r6eLS2gZseLwcLinhP3wCffeAB3tM3kLIFYjvWxuo8KzLFcf04WpEpjuJop0wnOFqRKY7imO6Z4iiOdsp0gqMVmVY4OpW4202kG+txuwmPx4NhGLbOFEd75jklUxztmZdopmEEOH3uvbxy6c0zGVcwjPx572tr3kdRwUFLHd3hMG+69Ap3j5wj3wgC8FzOBh6pbWB4gQ+ks8txXO1McbRnnhWZTnC0IlMc7ZlnRaY42jdTHO2ZZ0WmExytyBRHe+ZZkSmO9s20wtEJzN1uQhaJBUEQhDXlytU+Xjr1+0xNm5+RWl7619TVfIWr1/o4O3IvU5PlZPgvUlu1PyULxAuiNbuuXOY3z52hZnICgLP+TP6spp5/zS8A+bRbQRAEQRAEQRAEIU1Y1p7EgrPx+Xzc+bN3pvxx/lRmiuP6cbQiUxzTw9Ewsnjp1O/x/PDDTE2X4c+4QEfbB2iq/1M87gmKCg4y0Pce9n3sTxjoe0/KFojjObbdvMEXjx7hU8dfpGZygiseLw/XNXJ3Zw//WlC46ALxWh/HtcoUR3FM90xxFEc7ZTrB0YpMcRTHdM8UR3G0U6YTHK3ItMLRqciTxGmIUorsnGxu3rhJquqb6kxxXD+OVmSKo/Mdx65s46XTvxf9ILqKsu9RV/1V3O7JVXUsm5jgvvNn2Hv5NQCmlItvl1fyVxVVjHs8K86za62tyBRHcUz3THEURztlOsHRikxxFMd0zxRHcbRTphMcrci0wtEpyHYTgiAIwpphGNmcOvvbvPraGwHwZ4zQ3Pgn5G44vCrt7x4b5Z6Rs5RPTjLtcpFlGHiBMPAPRSU8Vl3HpQz5ZFtBEARBEARBEAQhvZHtJtYBXq+XgcEBvF6vbTPFcf04WpEpjs50vHxlgGcO/8XMAnGYyrJv09N5z6ILxKl03D02ykPHX6RxfJxAOEzezALxicwA7+7YwqebWhJaIHZCra3IFEdxTPdMcRRHO2U6wdGKTHEUx3TPFEdxtFOmExytyLTC0am4H3zwwbV2sJyHP//wg/l5+WutsapMTk5y6+atlD4qn+pMcVw/jlZkiqNzHINGDidOfYQz599LKJxFpv8cm1s+SlnpD3C5QqvjqDWfGX6BnND89q54vRyorks8G2fU2opMcRTHdM8UR3G0U6YTHK3IFEdxTPdMcRRHO2U6wdGKTCscncC5kXNPPvjgg/8Q+btsNyEIgiBYxuWxXbx0+sMEg4VAiMry71Bb/Rhu1/SqOdSN3+J3zp6i/9rVuK+Pu9y8oX/HqvkIgiAIgiAIgiAIwlqz4u0mlFLvVEr5rdUSUonH46G7rxvPMj90aS0yxXH9OFqRKY72dBwd282hIwd48pkf8szhb3D4+S9y9PinCAYLyfSfpWvzb9NQ++iKFoiTcdwQDPLB0yfYf+QQ/deuEl7gfRf9ye1B7IRaW5EpjuKY7pniKI52ynSCoxWZ4iiO6Z4pjuJop0wnOFqRaYWjU1nOETgA/KlS6pvA17TWq/PpQkLChHWYsdExwnqhJZG1zxTH9eNoRaY42s9xdGw3Lx5/KPr3YLBu5k8hqiq+RW3V13El8PRwIo7ucJhffPUi94ycY0PIAOD/5hfyVG4eHz5zct7791fVrtgrWcd0yBRHcUz3THEURztlOsHRikxxFMd0zxRHcbRTphMcrci0wtGpLLndhFLqq8CvA9mABp4C/hz4H1rrW5YbpgDZbkIQBMFanjlygPHxxnlf92eMsHXLW1fNY9uVMX7n7ClqJycAOBHI4ku1DRzKzQNgcGyUe0fOUj45xUV/BvurajlYULRqfoIgCIIgCIIgCIJgB1a83YTW+n6gHLgfc4G4H/gqcEEp9RWlVJ9VskJiuN1uWje34na7bZspjuvH0YpMcbSf4+RkZdyvB4MFSeUu17FmYpzPDD/PZ4+9QO3kBFc8Xj5b38S7O7ZEF4gBDhYU8e4t/Xzwnnt595b+lCwQO6HWVmSKozime6Y4iqOdMp3gaEWmOIpjumeKozjaKdMJjlZkWuHoVJa14cbME8NfA76mlGrHXDB+28z/71NKHQH+DPgrrfV1q2SF5RMOp/4x+VRniqM985ySKY72yAuF/Jw+917C4fhb12f4LyaVD4s75hhB7hk5xy++ehGP1gSV4omyCv6isoZbi+wpZbfj6NRMcbRnnhWZTnC0IlMc7ZlnRaY42jdTHO2ZZ0WmExytyBRHe+ZZkSmO9s20wtGJLLndxIL/UKkM4C3AfcAdmFtRTADfBh7RWj+TKslkke0mBEEQUs/1G20cP/kxJiarF3xPW/M+igoOprxtt9a86dWL3DtyllzD3Hf4X/MLeKSmgZHMzJS3JwiCIAiCIAiCIAjpxIq3m1gEL5Az8x+AmvnaPcBPlFLfU0rlLfSPBetwuV3UNtTicidTXmszxXH9OFqRKY5r6xgOuzlz/l4Ov/BlJiar8fku0d72Qdqa95EVOInHM0FW4GRKFojjOW69eoX9Rw7xoTMnyTUMTmUG+N3Wdj7asnlZC8R2OY5OzxRHcUz3THEURztlOsHRikxxFMd0zxRHcbRTphMcrci0wtGprPgIKKW2K6UeAy4CjwCdwPeBnwU2YG5D8VPgl4AvpE5VWC4u5SIvLw+XSmEnTHGmOK4fRysyxXHtHMcnajj8wqOcf/luwENx0Q/p7byb/NxDFBUcpL/nPn7z/t+nv+e+pBeId4+N8tizT/Pn3/02jz37NL908WU+PfwCnx9+nvqJca56PHy+rpF3dfbwdF7+snPtcBzTIVMcxTHdM8VRHO2U6QRHKzLFURzTPVMcxdFOmU5wtCLTCkensqztJpRS+cA7MLeW2IT51PB54M+Br2mtX5nzfg/wLFCutV7zj42X7SYEQRCSQ2vFhVd/iTNn30tYZ+BxX6ep4fMUF/6zJe3tHhvloeMvxn3NUIonSiv4i6pqbnq8lrQvCIIgCIIgCIIgCOnMirebUEo9DrwM/CnQBvwd8CagXmv90NwFYgCttQE8BSz/0S4hZbhcLkrLSnG5UviTmhRniuP6cbQiUxxX13Fqqpjnhz/HqTMfJKwzyMv9CT1dd8ddIE6V4z0jZ+N+/ZbLzTs7e/hyXUPCC8TrsdZWZIqjOKZ7pjiKo50yneBoRaY4imO6Z4qjONop0wmOVmRa4ehUlnME3gpcBT4NNGitf15r/bda66U++u+vgU8kKyisHLfbTX1TPW6327aZ4rh+HK3IFMfVc7w0+joOHfk6V69txeWapLHuT2lv/QgZvlFLHSsnJ+N+XQHnMwNJZa/HWluRKY7imO6Z4iiOdsp0gqMVmeIojumeKY7iaKdMJzhakWmFo1NZcrsJpdRbgP8583SwI5HtJgRBEFZG0Mjh5OkP8drlvQBkZ71IS9NDBDLPW9ruhmCQ+8+f4RcuvYKK8/rJQIB7OnstdRAEQRAEQRAEQRCEdGfF201orZ9w8gLxekQpxYbcDSgVb4nFHpniuH4crcgUR2sdr1zt49Dhr88sEBvUVO2na/NvLWuBOFFHl9b8wqsX+ebhp3nTAgvEAPuraleUm0rH1cpzSqY4imO6Z4qjONop0wmOVmSKozime6Y4iqOdMp3gaEWmFY5OZTl7EvcopR5WSm1d5D39M+/pTq2ekAher5ee/h683tR9oFOqM8Vx/ThakSmO1jiGQhmcPPN+nh9+mOlgMZn+c3S3/xa1VV/H5QpZ5th68waPPv8cv3f6BLmGwclAgN/Z1Mm+5jZOZWUz5fNxKiubfc1tHCxI/rNQ12OtrcgUR3FM90xxFEc7ZTrB0YpMcRTHdM8UR3G0U6YTHK3ItMLRqSxnu4n9mPsSV2utX1vgPcXAeeAvtdb3pdwySWS7CUEQhMW5cbOVYyf2MTFpPqlbXvoE9TVfwe2esqzN3GCQ+86f4ecvvYILuOl2s7+qlr8uqyAkP8UVBEEQBEEQBEEQBMtY8XYTwCBwaKEFYoCZ1w4BdySvCEqpryulppVSN2P+G4h53auU+pJS6opSakwp9UWllCcVbacLXl/qfwKS6kxxtGeeUzLFMTlGx3bzzJED/Pipv+fJZ/6G555/lInJWnzeUdpbP0xT/X9PeIF4KUeX1rzp1Ys8PrO1hAv4+6IS3t7Vx/fKK+ctENv5OFqV55RMcbRnnhWZTnC0IlMc7ZlnRaY42jdTHO2ZZ0WmExytyBRHe+ZZkSmO9s20wtGJLGeRuBI4s4z3nQUqkrKZzSNa6+yY//495rWPAbuATcBmzIXsj6awbUfj8/nYvWc3Pp/PtpniuH4crcgUx+QyR8d28+LxhxgfbyQUChAMFgBucrKP0NP1TvLznrLMsW1ma4mPzNla4lNNLYzF+Td2Po5W5TklUxzFMd0zxVEc7ZTpBEcrMsVRHNM9UxzF0U6ZTnC0ItMKR6eynO0mrgD/rrX+T0u87wfALq11btJSSn0duKq1/uACr58HPqS1/t7M338F+JzWOu4nGq3H7SZ8Ph/T09O2zhRHe+Y5JVMcE+eZwwcYn2ic9/VA5kl6u+5JKhviO8bbWuKxqlr+ZhlbS9j1OFqZ55RMcbRnnhWZTnC0IlMc7ZlnRaY42jdTHO2ZZ0WmExytyBRHe+ZZkSmO9s20wtEJJLLdxAvALqVUwUJvmHltN3A0ecUod81sJfGCUurDSinXTFv5QBXwXMx7nwNqlFJxF6iVS+HxmLtRuFwuPF5PxHvWI+WxPzXw+XzRTzb0er24XOah8ng8uNzmn91uN26328x1u2a1EdnwWik1L3c5bURy3W43bs/K24ic3F6fN9qGx+tZuo25xyemjVjmusdrw+1xz2pjbg2mp6dn1SDe9xH3+CzSRjgcjv99JFjn6enppOs8twaGYSy7Bku1oZRieno6pXWOPX9SUedIrub2D6RW3NfitDE9PZ2Svhbbn0Mh88PZUlFnl8s8fxKtwULuwWBwXg3mtrHYeBc0spmYiPvzNKamypOus9fnnXX+RLaW+GbM1hI/LCnjri39PFFeifJ6lxxTDcNYcQ2WGlOnp6dTVmeAcDic0jpH+naiNYjn7vP5oufPvDF1kb62WBvT09Mpv3ZGrg0rrcFCfc0wjJXXYIk2oteGFNQZIKzDKa1z9NqQwvlLLKmaIwWDwZTVed7cYoU1WKivhUKhJWuw0jamp6dTVmcArXXK56kJzS3cC89fYklFncE8f1JVZyvmkAChcGjBOse6L7eN2Dlkqu5HNDqlc0i4fW1Y7pga7/jMmkPOPNSUijrHzi1SeU+YzBxyoTYWnFskOH9Zzhxypde1yNcSqcFCc8hU3BPGthHW4VltrKQGC/U1I2SkbA7p8Xii508y94Rz24idW6TifiR2DplsneOR6D1h7PFZag6ZyPwl9tqQbJ2tmEMutr6Q6PwlModMVZ0XW19IZv4SS6L3hLFtrGR9IdXrL3NrEHFPZQ0WamMuy1kkfhzIBr6nlKqa+6JSqhL4DhAAvrmMvOXwBaAFKAbeBXxg5j9mXACuxrw/8uecOW6vV0o9rNyKtvY2AErKSujZ2gNAbl4uA4PmVscZ/gwG9wxGD/DOoZ1k55hN9W3vo7C4EID27naqqs3DsLF1I43N5tN41bXVbO7cDEBxSTG923pNoQ057LhjB2CeSIN7BqMF2rF7Bzm5pnJPfw/FpcUAbO7cTHVtNQCNzY1sbNkIQGV1JR3dHQAUFhXSN9AHQFZ2Frvu3AWYJ8zgnkGG9g7h8/kY2DVAbr65dr6lbwul5aUAtLW3UVtvLhTVN9XT0tYCQEVVBV09XQAUFBbQv7MfgA25GxjcM0hGRgYul4vBPYP4M/0AbNu5jfzCfAC6eroorywHoHVTK3WNdQDUNtTSurkVgLKKMnq29jC0d4jikmK279o+qwaRE35wzyCBrAAAW3dspaDQ/DlFZ3cnldWVADS3NtOw0XxKvK6xLvorAiWlJfT090Tdd+w2a5CRYbYR6Qyxde7d3ktRcZFZ56526urrGNo7REtbC00tTQBU1VTR3tUOQFFxEb3bzTpn52Szc2inWWevN3qsInXekLvBPJe29XHH6+7A5/OxqWMT1XVmnRs2NtDc2hytc2d3Z7QGW3dsBSCQFWBwz6BZZ7dZ55ycHIb2DjGwa4C8/LxoncsqyswabG6ltqE2enxaN5k1KK8sj9Y5vzCfbTu3AeDP9EfPn8ixygxkAtC/sz9ag66eLiqqzN1lWtpaqG+qN+tcXxvta6XlpWzp2xI9VoN3mue+32+2ERkwd925i6zsLPP4DPRRWGT2tY7ujmidN7bM7mvt3e0M7R2ivLI8Wuec3JxonX0+n1nnmYFxxx07yNlg9rXebb0Ul8zva81tzex+nXn+VFVX0d5t1rmwuJC+7X3z6uzxeMw6+806DwwOkJtn9rWerT1UVFUwtHeIzZ2bqamvAeb3tc4tM3UuKqB/R/+sOiul5vW17YPbuWOvef5093ZH69yyuSXa1+oa6mb1te7ebgDyCvKoqP51nj1yAE38Ldwz/Bfp39FPQdFMX9vSGa1zc9vtvlZTV0NbR/wxdcfuHQztHSI7J5tfa67nKy8c5iOnT7DBMDiVncNvb+rkH995F65KM3fumNrUbPa16ppqNndtxufzccfr7ojWYNaYOtPXfBm3x9RIX4sdUzd1bpo1prZuamVo7xA1dTV0bOmI1mDrgNnXsrKz5vW1SA3m9rXS8lJ8PvPXkxqaGm7XeZNZ59i+FjumZgYyo3VWSs3qa9t2bqOktIShvUN093ZHx9SWTS3UN870tYZaWtvNOsf2tbz8vOiYGunPsWNqZPzZOrA1WueO7g4qq26PqY0bb/e1TZ2bzDqXltDbb453kTHV5/MxtHdo4TF1Wy9FJeaYurlrM9U1Zg2aWppuj6kxfa2ouIi+gT6G9g6RX5B/u695PQuOqT1beygpKzHr3LGJmjqzr0XG1Mj5E6lBYVHhomOq3z/T13ZtJ6/ArHNsX2vd3ErDxgaG9g5Fz6VInSN9Lb8gn227Zo+pLpcrWudAwLyu9e80+1rk/Imco7Fjak19TdwxdbH5y647d5GXn8fQ3iG2DmyNzl86ujuWnr+UFt8eU2P6WlaW2SeyssyxOnZMXXT+0roxWufo/GVmTI3UJjJ/WWxMnTt/WWhM7e7tZmjvECVlJdG+FggsPKZu27WN/IL8aJ3nzl98Ph+7X7ebTR1mP5g7pkb7mn9+X4udv0Sua51bOqmpq2Fo71D0XAJzTI20ETumxpu/RCb/kb4Wud6VlZvnaOyYGtvXImMqQFFJUXSeGm/+kp2dzdDeIXbu3hnta739vZSUzvS12DF1Y+Pt+UvV7Xlq7Jiam5cbnf/OHVO379o+b0wFaG2/PX+pb5w/pvp8Pu7Ye0e0ry02ps6dv8waU2f6WkNTQ3QOOXdMHdhl9rV4Y2pk/jJrTN1ijqmRvh3pB7Fjamxfi62zL8M3a0yN7Wt922/PIWeNqc1Ns/rarDE13vwlZkz1+XwM3jkYPUdjx9S2jrbZY2qbWeel5i+x14ZonXcuPKbWNdQB5jy1ZbNZ59i+VlxSHM2NO6bO9LWF5i/xxtTI+BO5riV7T9iyqSU6h0zVPWHk/Ikcn0TvCSPXtYHBgegcMpl7wti+FjlWGzaY40Qi94Rz+9rQ3iFycnKSuiesrquOjqnlleXRvp3oPWFVzewx1efzcceeO6JzyETuCXv6e6JjantXe9Qx0XvCyLUzMqZGzp/I8Un0njC2r0XGn2TuCXPzc6Njas4Gs86ZmeZYncg9Yez8pXdbL0N7hygoLEjqnjB2/lJbVxutTaL3hNExtb2N+qZ6hvYO0dTSlNw94a7t0fnLlr4t0dos955w+87Z9wqRxdnBPYPk5uUytHcoOk+FBO8JB2f3tcj61NwxNdrXlrgnhNnzl/yCfHbduQufz5fwPWF0TK2qpKuni6G9Q5SWlSZ8Txip86wxNcl7wnhjairuCSNz+sj3GstytpvwAP+Iue/vJPD3wMmZlxuB1wOZwL8Bd2qtjUUDE0Ap9VvAXVrr7cp8kngMaNJan5x5vQl4CcjTWl+b+++rK6v1xoaNGIaBy+XC5XZhBA2UUni8HoLT5k/F5v6EIxgMorXG6/USCoUIh8N4PB7COkw4dPsnu6FQCJfbhUu5om243W6CwWD0pwSxuctpQ2tNKDTz9IGCkLGCNjLMSdSNazfMnywFDbTWeLwewqHw4m3MPT4eD8FgEJfLRV5hHldGr6C1nuduGPPbcHvcoIm2oZS67e5xkxnI5Ob1m7g9boLT8Y9V3OOzQBsej4ec3Byujl01B9Mk66zRZGVnMX5zPHqsEqmz1+eN1sDr8xLICnD96nXcbveSNViqDcMwyN6QzeTEJMHpYNJ1jvzEKXL+RJ5cTbTOsd9HXkEeVy7fPn9W1NfmtOFyucjMyuTWjVu4XK6k+lqkDY/HQ3ZuNtfGrqFcKqk6e7wedFiTlZPFxK0JwuHwymqwQD/w+XxkZmWa54/HvWCd59Zgekpz/uW7Offy2wE3Xu8owWDRvLG2rXkf5aVPJlzn3WOXufflc5RPThBEkR0yoltLfL2uke8Xl2LAisbUUCjEhrwNTNyamPWE4JJ9bZExVbkUgawAt27eQqGSqnM4FEZrTV5BHjeu3cAwjKTrHHHP3pDNxPgEISO07BosOt5l+MgMmOdP5KmV6Jgajt/XFmvDCBrk5OYwOTHJ9FRqrp0ejwd/wM/N6zejP9FfTg0WGu/CoTC5+bncunkrep1ZVg0WaQMgkB0wrw3opOocefIjryCP61evm+7J1jn22jA+iWEYSc9flFLkF+UvOAdIZI5kGIbZt8cnZp0/idQ5UoNZc4tF+vNyr2vh8Mz5c+NWdC4UrwYraSNybRi/NY4OLzxuL3f+AuaN/bUr16JzoWTnqcFgMNq3VzS3cLtwuebPX+LNIZOps8vtImQsfW1Yyfwl1XPIsA6jw/OvDbPmwjE1WE4bs+YAbndK7keUUuQX5nN17Oq8OXZCc8iZeePc82e5dV5qDsnMHCLROkfaiMwtxm+N3742JHFPmOwcMl4bi84tErnvjJkDLDaHXMn8JXJtuHrZPH+SuVeIvQeJNwdYyT1hbBtut5sN+Ru4etl8piwV9/7xrg2JziEjxy0rJ4trV66ZTwomcE84d/4CkFeYx/Ur16NjXKJ1jpw/0fUFjyepOsdeO1c8t1ikjeXMIVc6f3F73GzI3RC9NiRTZyvmkJGaZ2Wb58/c9YXl1GDB9YXxyWh7ydR57vkzd30h0XW25c4tlrvO5na5V7S+kOj8JdE6J12DRfrBPx38p1nbTSy5SAyglApgPt37TsA95+UQ8A3gA1rrm0uGJYBS6jeBu7XW22f+fh74oNb6iZm//zLwsNa6Jt6/X497EguCIABMTJZz7MQD3LhpPllUWvwDGuu+wJVrfZwduZepyXIy/BeprdpPUcHBhNvZPTbKQ8dfnPf1Qzm5/PHGVq7E/IqLIAiCIAiCIAiCIAhrSyJ7EqO1HtdavxuoAd4G/OHMf28DarTW70rlArFS6leVUhuUSd9MW0/EvOUAsE8pVaaUKgM+CnwtVe07Ha/Xy7ad2+LuL2KXTHFcP45WZIrj8jIvjf4Mz/50PzdutuN236C16UGaG/8Et3uCooKDbO+9nw996I/Y3nt/UgvEAO85dzru13NDwaQWiO1wHFc7zymZ4iiO6Z4pjuJop0wnOFqRKY7imO6Z4iiOdsp0gqMVmVY4OpX4G1MugNb6IvAti1xi+R3gq5h+LwOPAJ+Pef2TQCEQeWztceBTq+DlCEKhEGdOnYlugG/HTHFcP45WZIrj4pmGEeDkmQ9xafT1AGzIOUxL00P4M15NuWNmKMQ9I2epmpyM+3r55FTC2SC1tnOmOIpjumeKozjaKdMJjlZkiqM4pnumOIqjnTKd4GhFphWOTmVZ2004HdluQhCE9cL1G5s4duLjTE5VAAa1VV+nuvKbKJX6C96usct84MxJSqcXXgg+GQhwT2dvytsWBEEQBEEQBEEQBCFxEtpuAsx9iZVSb1RKfVAp9YBS6uNx/nvAGm1hJXg8Hrp6uqKbo9sxUxzXj6MVmeI4P1NrF+dG7uLwC19icqqCjIyLdG1+PzVV31hwgThRx/NpP/gAACAASURBVJKpSf6fY0f51PGjlE5PcTEjg29UVMd97/6q2hVlp8pxNTOd4GhFpjiKY7pniqM42inTCY5WZIqjOKZ7pjiKo50yneBoRaYVjk5lWUdAKXU38KfAhtgvAzrO3z+ZKjkhMbTWXLt6jVQ+JZ7qTHFcP45WZIrj7MzJqRKOnXiA6ze6ACgu/BFN9Q/j8dxKqaM7HOZXXrnAPSNnyQyHMZTiW+VVfKOymim3m2PZ2dw7co6K6Sku+DLYX1XDwYKixL7ZBB3XItMJjlZkiqM4pnumOIqjnTKd4GhFpjiKY7pniqM42inTCY5WZFrh6FSW3G5CKbUX+AfgGvBl4E5gAPhN/n/23js+ruO+135mzzYUkmgECKISvZAASLABBKiS+I3zptyU1zfNtiLZiu3EuXYSx3JsyaJsxXGJFcd2EieOpcQ3vTi5zk1xwsimaTXb6oVNpASSEimJBCvqlnn/WOxiASyALWfIc3Z/z+ejj8jF4jsPzu/MnDmzhwNoBX4SaAe+CDyutf5zk8LZINtNCIKQr7xx/kZePPGbhCNrsDyTtG66j+qq/0Qpe9vpuXKZD7z0Im2TsYXnp9es5bOb2ni5uMTehgRBEARBEARBEARBME422038BrEnhG/SWt8FHAPQWn9Za/0hoBf4HHAb8Lj9ykKmWJZFR3cHlmU5NlMcC8fRRKY4WkQiRRw9fgeHj32McGQNa0qfZ2vfbdSsT3+BOB3H0nCIXz9xjD98/mnaJie45PXyyZZ2/ldPX8oFYrcdRyfmuSVTHMUx3zPFURydlOkGRxOZ4iiO+Z4pjuLopEw3OJrINOHoVtLZbmIH8KjW+ulUX9Rah5VSHwB+ArgH+Gkb/YRsUODxeGIbgDg1UxydmeeWzAJ0PDe+l5Onb+Wh79bhtc4RiQSZDVUBURrq/oLGugfweDL85XQrOWrND55/g/eOnaAiFALg36uq+cOmTVzy+bPLzIYCrLVrMsXRmXkmMt3gaCJTHJ2ZZyJTHJ2bKY7OzDOR6QZHE5ni6Mw8E5ni6NxME44uJZ3tJqaBr2mtf37u718CbgfKtNZXkt7318APaK2rDfpmhWw3IQiCmzk3vpdDR+9d8rrXe5HujrsoW5vyM7ysqZ+a4tdefpEdly4CMBYs4rOb2nhqXZmt7QiCIAiCIAiCIAiCcH3IZruJs0BF0t/PzP2/Y9H7KoCi3PQEO/BYHhqaG/BY6ZT3+mSKY+E4msgsNMex07emfN3vu5DTAvFiR180yi2nx3jgmcfZcekiM0rxp/VN3Na3Le0FYicfR1OZbnA0kSmO4pjvmeIojk7KdIOjiUxxFMd8zxRHcXRSphscTWSacHQr6RyBw8R+MV2ch4k9hP1BpWI7XyqlhoGbgSO2GwoZ41Eeqqqq8CgbO6HNmeJYOI4mMgvNcXq6LuXrMzM1WWfuHT/HV578Pl/5x7/nK09+n1tOjfHAM0/wjtMnCWjNd9eV8Yv9g3y1vpGQJ/2fwcnH0VSmGxxNZIqjOOZ7pjiKo5My3eBoIlMcxTHfM8VRHJ2U6QZHE5kmHN1KOttN/Crw+8BurfV3lVIe4AlgC/AasSeLNxPb3/gWrfVfmFXOHNluQhAEN6K1h5Onb+HkK79Iqg2SiouPM9iX+injldg7fo57jx5K+bXzPh9faGrhwcr1pP0b8ARBEARBEARBEARBcBXZbDfxVeCHiS0Io7WOAj8C/BdQDWwFJoE7nbhAXIh4PB7WV6+Pbbzt0ExxLBxHE5mF4Dg7W8mzh+7j5Cu3stwO+k3192eVfevpsZSvX/J6eVv/dh6sqs56gdhpx/FaZLrB0USmOIpjvmeKozg6KdMNjiYyxVEc8z1THMXRSZlucDSRacLRrax6BLTWl7TW39BajyW99orW+s3AWqAOqNJa/45BTyEDLMuirasNy7IcmymOheNoIjPfHS9c3M4Tz36FS5e3YXkm6Wz7GN0dH6Gk+AQ+3wwlxSfo7vgIVRUHs3LbOD2d8nVfVHPV680qM46TjuO1ynSDo4lMcRTHfM8UR3F0UqYbHE1kiqM45numOIqjkzLd4Ggi04SjW0lnu4n7gAta649fGyX7ke0mBEFwA1pbjJ3+RU698jbAQ0nxi3S1301x0Slb8pXW/PhrZ/i1l4+n/ITweHExt/YN2tKWIAiCIAiCIAiCIAjOJZvtJt4L9JlTEuxGKUXpmlKUjfuJ2p0pjoXjaCIzHx1nZit59oX7OPXKLYCHDdVfp3/zuxcsEOfiuHF6it879Cy/scwCMcD99U0Z5y7meh/H65HpBkcTmeIojvmeKY7i6KRMNziayBRHccz3THEURydlusHRRKYJR7eSziLx6TTfJzgEn8/HzuGd+Hw+x2aKY+E4msjMN8cLF3fw5DP3c+nK1rntJe6hveV3sTyzOTt6tOYtZ17hz555gm2XLzHl8fD7TS3c2d7FiZJSZvx+TpSU8pGObg5WVGX8cy5Gam0PbsgUR3HM90xxFEcnZbrB0USmOIpjvmeKozg6KdMNjiYyTTi6lXS3m3g7sElrfeWaWNmMbDchCIITiW0vcSunXnkrse0ljs1tL3HalvzGqUnuOH6MLVcvA/DE2nV8qqWdM8EiW/IFQRAEQRAEQRAEQXAn2Ww3cTdwEvg3pdRWY2aCrVhe+zfctjtTHJ2Z55ZMtzvOzFbyzAu/x6lX3k5se4n/Q//m96y6QJyOo6U1P//KKb7yzBNsuXqZCcviM5vaeH/3liULxG4/jk7JdIOjiUxxdGaeiUw3OJrIFEdn5pnIFEfnZoqjM/NMZLrB0USmODozz0SmODo304SjG0lnkfj/ADPAHuD7SqnTSqmHlVIPpvjvv83qCung9/u58QdvxO/3OzZTHAvH0USm2x3j20tcvjIwt73EPtpbPrtke4lsHFsmJ/jD557i3adeJqA1j60r55a+bfxLTS0s2mPJ7cfRKZlucDSRKY7imO+Z4iiOTsp0g6OJTHEUx3zPFEdxdFKmGxxNZJpwdCvpbDcRzSBPa60dt/xeiNtNBINBpqenHZ0pjs7Mc0umGx21thg7dSunXn07QFbbSyznaEWjvPXVU7z9lVP4tOaK5eULTS38x/rqJYvD6eTlgtS6cDLF0Zl5JjLd4GgiUxydmWciUxydmymOzswzkekGRxOZ4ujMPBOZ4ujcTBOObmDxdhPpLBJn9OvutdZjWboZoxAXiQVBcBYzs1UcPnY3l6/0A7Ch+p9pbf4inlWeHk6Hjomr3HH8KO2TEwB8p7yCz25q47w/kHO2IAiCIAiCIAiCIAj5R8Z7EmutxzL5z6y+kA4+v4+9N+/F57fxt0fanCmOheNoItNtjhcu7uCJZ+7n8pX+pO0l7st4gXixoy8a5Z2nXuZLzz5J++QEF71e7mnr5MMdPWktELvtONpFITqayBRHccz3THEURydlusHRRKY4imO+Z4qjODop0w2OJjJNOLqVVZ8kzgcK7UlipRTrytdx6cIl7Kqv3ZniWDiOJjKd7nhufC8nT9/K9MxGPGqaULgciG0v0d1+N0UZbC+xnGPX5Ut86MQxNk1NAvBgRRWf29TKRV/6+yg5/TiayDOR6QZHE5niKI75nimO4uikTDc4msgUR3HM90xxFEcnZbrB0USmCUe3kM12E42ZNKC1PpmlmzEKbZFYEITrx7nxvRw6eu+S18vXPUpP551Zby+xd/wct54eo3Z6mlmPhzXhMBZw3ufjc81tHKisytFcEARBEARBEARBEIRCIePtJoCXgZfS/O+Ezb5CFnh9XnYM7cDr8zo2UxwLx9FEppMdx07fmvL1mdD6nBaI7z16iNbJSYqjUcrmFoifWLOOW/oGs14gdvJxNJVnItMNjiYyxVEc8z1THMXRSZlucDSRKY7imO+Z4iiOTsp0g6OJTBOObsXat2/fim+45557bgLGUvx3CtDAOmKLzY8CL+3bt+/PDfpmxX2fvW9feVn59da4pkSjUa5evmrro/J2Z4pj4TiayHSio9aKE2PvBayl2ZEgDXV/kVXuPUcPUR4OLXl90mvxdxvrs8pMeDnwOJrOM5HpBkcTmeIojvmeKY7i6KRMNziayBRHccz3THEURydlusHRRKYJRzdw8vTJR/ft2/eN+N9z3pNYKdUB/CmggDdpradzU7Qf2W5CEATThMOlHDn+YcYvjKT8enHxcQb7Uj9lvBKbr1zmi88/nfKffUx6LN68czjjTEEQBEEQBEEQBEEQCptstptYEa31UeCngB7gnlzzhNzxer1sGdiC12vj4/w2Z4pj4TiayHSa49WJVp589k/mFoiXPvEL0FR/f0aZ/miUd4+9xBeWWSAGOBMMZCa6CKcdx2uRZyLTDY4mMsVRHPM9UxzF0UmZbnA0kSmO4pjvmeIojk7KdIOjiUwTjm7FliOgtT6nlHoM+FngDjsyhezRWjNxdcLWx+TtzhTHwnE0kekkx9fe+CFePPEBojpAwH+G7o6PMjNbzcnTtzEzu5GA/1Ua6++nquJg2pndV6/w4ReP0DQ9BcATa9ay7crlJe+7v74pI9fFOOk4Xqs8E5lucDSRKY7imO+Z4iiOTsp0g6OJTHEUx3zPFEdxdFKmGxxNZJpwdCs5bzeRCFLq34GbtNZBWwJtRLabEATBbqJRH8df/lXOvv4TAJSve4zOto/j8y1dzE0XXzTKLadP8guvnsICzvt8fKalnYfLKxkdP8dtp8eonZ7hTDDA/fVNHKzI7hfWCYIgCIIgCIIgCIJQ2Ni+3QSAUmorcAOxX2gnXGcsy6Ktsw3LWvrLs5ySKY6F42gi83o7Ts9U8/TzX5xbII7SWPcAvV13LFggztSxY+IqX372Sd4+t0D8X5XreXv/IA+XVwJwsKKKd27dyXvf9nbeuXWnLQvE1/s4Xo88E5lucDSRKY7imO+Z4iiOTsp0g6OJTHEUx3zPFEdxdFKmGxxNZJpwdCurbjehlProCl8uBTqAH57L+mObvIRcUOD3+2O/StCpmeLozDy3ZF5HxwsXd3D4xbsIh8vwWpfpbPs4FeWPZZ3njUZ52yuneNurp/BqzQWvj8+2tPHtVIvAeXQcr1ueiUw3OJrIFEdn5pnIdIOjiUxxdGaeiUxxdG6mODozz0SmGxxNZIqjM/NMZIqjczNNOLqUVbebUEpFAc3Kh2sS+F2t9T771OxDtpsQBCFXtFaceuVtjJ2+DfBQWnKY7vaPEgyezTqzZWKCDx8/QsfkBADfrKjivk2tXPL5bbIWBEEQBEEQBEEQBEFYSjbbTdwK3Db3/8X//QJwM1Dj1AXiQsTj8VDfWI/HY8tuIkYyxbFwHE1kXmvHUHgNzx/5JGOn3wl42FD9dfp737viAvFKeZbWvO2Vk3z5uSfpmJzgktfLvrYu7u7oXnGB2O3H0Ql5JjLd4GgiUxzFMd8zxVEcnZTpBkcTmeIojvmeKY7i6KRMNziayDTh6FZWPQJa6z9f4b+/1lp/S2s9cS1khfTwWB5qamvwWDZ2QpszxbFwHE1kXkvHqxMdPPnsn3Lh4hAeNUN7y+/Q3vK7eDyzWeU1TU7wh889xe2nxvBpzcHySm7pG+TBqvVZO2aL1LpwHE1kiqM45numOIqjkzLd4GgiUxzFMd8zxVEcnZTpBkcTmSYc3cqq203kA7LdhCAI2XD29R/hxZfej9YBgoFX6e64i9KSY1llebTmZ86c5h2nxvBrzRXLy+c2tfJfletByeZHgiAIgiAIgiAIgiBcOzLebkIptU0pdZ9SascK79k5954Bu0SF7FFKUVlVibJx4cnuTHEsHEcTmaYdo1E/R49/kGMn7kDrABVlDzOw5Z0ZLRAn5zVMTfLF55/mPSdfxq81j5SV8/b+bfxXVXVGC8RuO45OzDOR6QZHE5niKI75nimO4uikTDc4msgUR3HM90xxFEcnZbrB0USmCUe3ks6z1O8Ffhl4eYX3vDT3nl+xwUnIEa/XS9fmLrxer2MzxbFwHE1kmnQMhet5+vk/4LU3fhSI0tTwZXo6fwuf92raWXvHz/HA04/zN9/4V772+KP82dNPsPnqFa5aFr/T0s4dnb2c9weydnTDcRTH/M8UR3HM90xxFEcnZbrB0USmOIpjvmeKozg6KdMNjiYyTTi6lVW3m1BKHQPe0FoPr/K+h4EqrXWHbXJKFQHPzuWWzb22FvgS8KPAFPBFrfXHV8qR7SYEQViJc+N7GTt9K9NTdUS1D7Dwei/S1fYxysu+n1HW3vFz3Hv00JLXjxUX81udm3k9kPnisCAIgiAIgiAIgiAIgp1kvN0EUMfKTxHHGQM2Zum1HB+by03mC0AF0AiMArcrpd5uc7uuRilFcUmx7Y/z25kpjoXjaCLTzrxz43s5dPReJidbieogYAHQ3PAnGS8QA9x6avGQFcOjyXmB2MnH0VSmODo3UxzFMd8zxVEcnZTpBkcTmeIojvmeKY7i6KRMNziayDTh6FbSWSSeAcrSeN9aIJKbzjxKqUHgzcCnkl4rBn4WuFNrfVFrfZTYovE77Go3H/D5fAyNDuHz+RybKY6F42gi0868sVOph49XX/vpjLPWhEM0TU2l/FrtzEzGeYtx8nE0lSmOzs0UR3HM90xxFEcnZbrB0USmOIpjvmeKozg6KdMNjiYyTTi6lXS2m/gO0Ac0a63Hl3lPBbEnfp/TWg/lLKWUF/gu8H5iC9n/rLUuU0ptBZ4AfFrr8Nx73wT8nda6fLm8hroG3d7STjgcxuPx4LE8hENhlFJ4fV5CsyEA/H4/s7OziT+HQiG01vh8PiKRCNFoFK/XS1RHiUaiWFbsicNIJILH8uBRnkQblmURCoVQSuHz+RbkptOG1ppIJBJrQ0EknFkbiVy/j3AojNYar89LNBJduY3Fx8frnW/D72N2JrV7OLy0DctrgSbRhlJqQQ0i4dhnCvEarPhzJB+fFdrweDyEQiHb6qyjGo/Hk1OdF9dAR2PHPZ0apFPn+DG0rc4+X+KY2FHneBuLz5+M+lqKNiKRCEqpnPrabKiZxx7/M1J9XmZZkwzveHPadd4xeZU7Dj9P9dzfF3O8uJh3bt2Zdg0y6QeL67xcDVL1tcU1z7XOyeeiXWNqct9Ot6+tNqZGI1GUR+Xcn+PHx+fzEY1GM+5ryx2fcDgcOw+9VlY1WG68C4fDCd8F51J0+b6W7rmUa53jNYj/7JnWYLnxzrIslEdlVoNV2ohGo7Frgw11jrcXCUdsq3Pi2uD12jd/CfiXnQNk24bH48Hr9dpS5yVziwxrsNx4F9+bLpd+sLgNrTUey2NLneNjbbxv2zVPXTyep1UDy4PHs8z8ZZk5QC7z1LTnwmle1+yeQ0Yj0QXzvsV1TnZPt434dcbW+5GAn9BsyLY5ZMq5RZp1XnYOuSg3lzrH21hybcjxnjDhbuM94bJziyzvR+LXGbvqvPj8sWsOmWoOkOk94YK+5rUSY7Vd9/6W10KRuj9nM3+JX2dyHu+S58LJNc+1zovmkLnWOVGDTOcWK9U5jTlkNvMXyzM/Z8m1zibmkCutL2Q7j4fYPMq2Oi+as9i2zpbG3CKTdbZM1hdsXWez6bqW7fzlwYMPZrzdxF8ApcA/KKXqF39RKVUH/B1QDPxlGnnp8JvAk1rrby96vRSYiC8Qz3ERWJMqRCn1Q0qp+5Sl6N7cDUD1hmq27dgGwLqydQyNxta0A8EAozePJgbpPTfuoXRNKQDbd2+ncn0lAJsHNlPfEDsM7V3ttHa0AtDQ1EBvXy8A66vXM7hrEIA1a9cwfENsO2ef38fozaP4/X4AhvcOs2ZdTH3bzm2sr1kPQG9fLw1NDQC0drTS3tkOQF1DHVsGtgBQWVXJ9qHtAJSUljBy0wgQW1AYvXmUYFEQgKGRIdaVrwNg6/at1NTWANC9uZumTU0AbGrbRGd3JwAb6zfSv60fgIrKCnbu2QlAUXERozeNopTC4/EsaGPXnl2UV8bW6Pu39VNbVwtAV08Xza3NADS1NNHV2wXAho0b2Lp9K0opysrL2D2ye0EN4oPC6M2jFJcUA7BjeAcVlRUA9A30UddQB0BHVwct7bH9phuaG+jeMlfnmmq27YzVee26tQzvjdUgEIi1Ef+EKLnOg7sHqVpfFatz/2bqG+tRStHW2UZbZxsA9Y31bO7fDEDV+ioGd8fqXLqmlD037onV2Rerc2Bua4HhvcOsXbc2UefqDdUA9GzpoaE5VueW9hY6ujoSde4b6EvUYMfwDgCKS4oZvXk0VmcrVudAMIBSit0juykrL0vUecPGDbEa9HbR1BKrc3NrM109sRrU1tUm6lxeWc6uPbsACBYFEzVQSjF68yhFxUUA7NyzM1GD/m39bKyP7S7T2d3JprZNsTpvakr0tZraGrZu3wrAuvJ1iToHg3NteGN1HrlphJLSEgC2D22nsirW17YMbEnUub1zaV9TSrG+Zn2izmvWrUnU2e/3x+rsj9V5+IZh1qyN9bXBXYOsr17P+IVdPPnsl1huGCwruwBA5fpKtu/evqTOXq+X0ZtHKfH7eNfJl/jMM09SPTvLGz5/yrz765uW9LW+rXN1rqpg5/DOBXVO1dd2j+ymvCLW1wYGBxJ17uztTPS15pbmBX1tYHAg9vNUlLF7z+4FdfZYnkSd431t5/BOKqrm+trWvkSdO7rn+1pjc+N8X1s0pg6PDqOUSvQ1ry/1mJroa4vG1LaOWF9raGygt3+VMXWur/kD82Nqcl+Lj6k9fT1LxlSlFHX1dWzZuiVRgx1Dsb5WUlqypK8lj6nJfS0+pnZt7lo4pvbE6pzc15aMqXN1XtzX4mOqUmrBmNrZ08mm1rm+1tJE1+ZYnZP7WvKYmtyfITamxvvajqEdiTpvGdhCXf38mNraPt/Xevp6YnWuqWZwZ6wGC8bU4Apj6q5Bqqpjde7t76WhMVaDBWNqQz2bB+bH1O27t6OUWtjXfN7lx9QdC8fUxuZGYOGYurFhvq9VVlWuOKYGg/N9rawiVufkvhYfU5VSS8bUeF8rryhn18jCMdXjSeprxXN9bU/qvpY8pjZuakw9pq4wf4mPqUqpBfOXLQNbVp+/JI+pSX3N7/czelPS/CVpTF1x/tLVnqhzYv6yaEyNz1/iY2ogGKvz0OgQ68pSz18aN8XqvGT+srUfpVRsTJ3ra8XFy4+pu0Z2LRhT05m/LBhTF1/Xlpm/xK9rfVv72NiwEaXUgvlLY3MjPVvm+lrSmJpq/pJqTB3cNZhyTE3ua8ljalV1VWJMTTV/8Qf8KKUW9LXBnYNU18z1teQxtb11fv5SPz9PXTCmlpQwelPqMXXx/GXBmDo3f9nUuvyYGp+/rDSmLp6/LBhTk+Yvy42pQyOxvpb2mLp1fkxdPH+Jj6nJfS25zv6Af8GYmmr+AovG1I62BX1t8Zi6uM6Lx9Shkfm+ljymdm/pXjimds+NqavMX2Dp/cjuPcuPqc0tzUBsntrZG6tzcl8rr1g6T10wpq4yf1lpTHX6PWGq+QtkeE8YXHn+ktU9YfKYelPu94QwP39RStlyT5gYU2tWHlPTvSeEpWNqzveENfbeE8LCMTXV/AUyuydM2ddyvCdMjKnBIKM35X5PCPNjqlIq43tCWHn+Eh9TV7snTDV/WTCmzs1flFKOvCeE+b6mlLLlnhBW7muZ3hMmz1/WrJ2vcy73hEDinjA+h8zlnnDJmGrDPSGkmL/kek9YMj9/WUw6TxJ7gf8mtv/vNPAfwPG5L7cCPwQUAQ8BNy1awM0YpVTbXHtbtdbjSqkbWfgk8eOAP+lJ4h8E/l6eJJ5vo6SkhN2juzn44EE02pZPOPyB2A3iwQcPMjs7m/MnHMFgkKG9Q3znm98hqqO2PEkcLAqy54Y9HHzwYOKT/Fzq7PV62XPjHh4+8DCzs7O2PElcXFzM0N4hDj54kGg0mvNTMigYvWmURw8+yuTkpC2fZBUXFyfOH8CWT7ICgQAjN40sOH9yeZI4EAgwtHeIh771EJFIJKO+5vX6ePnUW3hp7JcAD0XBMaamm5aMG5u7Pkp52bdWrEFLOMRvHXqOzomrAPxL9Qa+0NTC7quX+cWTL1MXmuUVn58HGpo4UFaR86eJyX07fmxyeZIYYheJ5PMn108sF58/uY6p8UnRIwcfYXJiMu2+ttKY6vf7Gb5hmIe+9RDhcDjnp0Dik8OHDjzE9NS0LZ8aezweRm4a4ZFvP8L09LQtnxovuDbMHY9cniRWqMT5MzExYcu1M1gUZGg0dm3QWuf8yb3liU3oHvn2I0xOTqZfgxXa8Pq87Llh7toQms356YD4xPahbz3E9PS0LU8HKKVi58/BR5iems55/hK/yVpuDpDNHCk+EV58/uTyhMaCuUU0mvMTGvEbgoe//TBTk1O2PEnstZbOLXJ5CiR+/nznW99hZnrGlnlqqmtDLk8SrzQHyPZJHI/yLDh/cn0Sx+45ZFRH8Vqxa8PDBx5mamoq5yeJi0uKGRqdn0PacT8S79vf+eZ3mJmZseVpqJRzixyeJE4+f+LnfK5PXCXmFsnXhhzuCYuKihi+YTh2/kTCttwTrji3yOJ+JN05ZCbzl8S14ZsHmZ2ZtW8O+c2DoNOYQ6bRRjAYZM+NexLnjx33/vFrQ/z8yfUJ06LiIob3xs6fSCRiy5PEi+cWuT7duHgOacfTjVnNLVZaX0hjDpnp/CX5/Ik/6Z3LE6Z2zyFXW1/I5kni1dYXspmnJp8/QM5zyEzmFumuswUDc3PIb32HaGT5OWQhPEm86iIxJPYC/jxwC/Hf6jRPBPgq8D6t9dVVw1Zv6xeBLwHxLB+xJ4XHgZ8G/hMY1lo/Pvf+DwA/rrXeu1xmfW29bmluyVXNNSilCBYFmZ6aJp36Xo9McSwcRxOZ2eZFIgGOnfggb5x/EwAbqr9Oa/PnGL84xMnTtzE9U0swcIbG+vupqji4fJDW/NjrZ/nVsRMEo1Eueb18uqWdgxVVOTuuhFOOxNA9XgAAIABJREFU47XMFEfnZoqjOOZ7pjiKo5My3eBoIlMcxTHfM8VRHJ2U6QZHE5kmHN3CwUcPZr5InHizUrXAjUDD3EungG9prc/YJTi3IF2R9NIQ8KdAL/D63J+rgJ8DqoH9wF1a668ul1loi8SCICxleqaaF458gonJDpQK09r8OWprvp5xzrpQiA+eOMbohfMAfG9dGZ9o7eC8P2C3siAIgiAIgiAIgiAIghEWLxKnsydxAq31Ga31X2utPz3331/buUA818ak1vp0/D/gjdjL+rTWehZ4L3AJOE1si4uvrLRAXIj4fD5GbhxJ/JMvJ2aKY+E4msjMNO/S5T6eevbLTEx24PNeYEv3ry1ZIE4nc8fFC/zZM48zeuE8s0rxxaZNfKBrc8oF4nw8jtcjUxydmymO4pjvmeIojk7KdIOjiUxxFMd8zxRHcXRSphscTWSacHQr6exJ7CH2C+OmtNahZd7jI7Yv8VWtddR2yxwptCeJlVKUV5Zz4fwFWx/ntzNTHAvH0URmJnlnXvtxjr/8frT2UlJ8lJ7OjxAMvJZRpj8a5faTL/MzZ18B4OWiIj7W1sWLJaW2OKZLIZ4/4ujcTHEUx3zPFEdxdFKmGxxNZIqjOOZ7pjiKo5My3eBoItOEo1vIeLsJpdRvAJ8GfkhrvX+Z9/wg8A3g17XWv2+jry0U2iKxIAgQjXo5/vL7OPv6/wCgqvK/6Wj5JJY1k1FO8+QEH33xCG2TsV+s9LWaWv6ocRMz1uLt2QVBEARBEARBEARBENxBNttN/CRwarkFYoC5r50m9ovlhOuM1+dlcNcgXp/XsZniWDiOJjJXy5udLefZQ783t0AcpbnhS3S13bPiAvGSTK35qbOv8uVnn6JtcoILXh93dPbwuU1taS0Q58NxdEKmODo3UxzFMd8zxVEcnZTpBkcTmeIojvmeKY7i6KRMNziayDTh6FbSOQLtwPfTeN9zwLbcdAQ7iEainH31LNGIfTt/2J0pjoXjaCJzpbyrEx28cOS3mZmtwbKu0NX2cSrKH80os3x2lg+dOMrQxQsAPFpWzu+0dHDB77fFMVsK8fwRR+dmiqM45numOIqjkzLd4GgiUxzFMd8zxVEcnZTpBkcTmSYc3Uo6201MA1/TWv/8Ku/7K+CntNZBG/1sQbabEITC4PVzP8Cx4x8iqgMUBcfo6fwwxUWnVv2+vePnuPX0GLXTM1zyeimNhFkTiTCjFH/U1MLXampBqWvwEwiCIAiCIAiCIAiCIJgnm+0mzgB9abyvD3g9WzHBPiyvRW9fL5bXvj1T7c4Ux8JxNJG5OE9rDy+NvZsjL95NVAeoKHuYgc3vTnuB+N6jh2idnKQ4GqF2doY1kQhn/X5+actWvrZhY1YLxG48jk7MFEfnZoqjOOZ7pjiKo5My3eBoIlMcxTHfM8VRHJ2U6QZHE5kmHN1KOovEDwLdSqmfWe4NSqn/CfQA37RLTMgBDdPT02DnL2W0O1McnZnnlsykvFC4lOcPf5LTZ2L/2KFh4/+mp/PDeL0TaUXdenos5euTlpeXiktscbSNQjx/xNG5meLozDwTmW5wNJEpjs7MM5Epjs7NFEdn5pnIdIOjiUxxdGaeiUxxdG6mCUeXks52E13Ak8QWlL8M/AlwfO7LrcAvAbcTO5zbtdbPGbPNEtluQhDyh3Pjexk7fSsz07X4/eeIRIqZDVXh8UzT0fJJ1lc9mH6Y1vzXdx8ikGIcnPRYvHnnsI3mgiAIgiAIgiAIgiAIziDj7Sa01oeBtwMR4D3EFowvz/33JPDLc1+7xYkLxIWIZVm0tLdgWTY+zm9zpjgWjqOdmefG93Lo6L1MTrYSiRYzNd3IbKgKr/cC/b2/nNECcVEkzF0vHkm5QAxwJhjIydXJx9FUnolMcXRupjiKY75niqM4OinTDY4mMsVRHPM9UxzF0UmZbnA0kWnC0a1403mT1vrvlVJPAr8O/ADQMPelU8B+4HNa62NmFIVMUUpRXFyMsvEXbdmdKY6F42hn5tjpW1O+7vNdpLTkxbRz2iaucs+xwzRMTxEBUl0K7q9vyk5yDicfR1N5JjLF0bmZ4iiO+Z4pjuLopEw3OJrIFEdxzPdMcRRHJ2W6wdFEpglHt7LqdhMZhSnl0VpHbQu0CdluQhDyg4e++x9Eo8VLXrc8kwzvfPPqAVrzE6+d4VfGThDQmtf8Ae5p76I8NMttp8eonZ7hTDDA/fVNHKyoMvATCIIgCIIgCIIgCIIgXH8y3m4iHZRSW5VS9wGn7cgTcsPj8bCxfiMejy3lNZIpjoXjaFdmKLQOSP3JXiB4ZtXvLwmHuefYYX795eMEtOY75RXc1reV59as5WBFFe8Y2MFtP/X/8Y6BHbYsEDv1OJrMM5Epjs7NFEdxzPdMcRRHJ2W6wdFEpjiKY75niqM4OinTDY4mMk04upWsj4BSqkEp9SGl1HPA94H3AzW2mQlZ47E81DXU4bFs7IQ2Z4pj4TjakTk51chTz32JaLQo5deb6u9f8fu7rl7hK88+yU3j5wgpxReaWvhwRw9XvD7bHBfjxONoOs9Epjg6N1McxTHfM8VRHJ2U6QZHE5niKI75nimO4uikTDc4msg04ehWMtpuQim1BngL8FZgL7HH+hTwCvC3wF9rrR834JkTst2EILiXC5cGOXT0Y0QiawgGXmVj7d9x9vUfY2a6lkDwDE3191NVcTD1N2vNW86+yrtPvoRPa84EAuxr7+ZQ6Zpr+0MIgiAIgiAIgiAIgiA4iIy3m1BKWUqpH1FK/Q1wFvgycGPS994INGqtP+DEBeJCRClFeUW57RuD25kpjoXjmEvmmdd+jOcOfYZIZA1r1zxD/+Z3Ubfha2zvv40fefPPsb3/tmUXiNeEQ3zi6Av86tgJfFpzoKKSd2zZtuwCcSHWRhwLx9FEpjiKY75niqM4OinTDY4mMsVRHPM9UxzF0UmZbnA0kWnC0a0su0islNqhlPo88CrwdeB/At65P78F+B6A1vrb2s7ffifkjNfnZfPAZrw+r2MzxbFwHLPJ1NrDibFf4cWXfhPwUl31DbZ0/xp+36W08nqvXOYrzzzJyIVxZpXi95pbuau9m6ve5dsvxNqIY+E4msgUR3HM90xxFEcnZbrB0USmOIpjvmeKozg6KdMNjiYyTTi6lSXbTSil7gR+Aehg/rdEPQz8BfB3WuvxufcdBIa11ta1080O2W5CENxDJFLE4RfvYvzCCABN9X9KQ91XSedDPaU1P3vmNLeffBkvcDoQZF9HN0dLSs1KC4IgCIIgCIIgCIIguIh0tpv4GLEF4rPAXUCL1npEa/2l+AKx4GyUUgSLgrY/zm9npjgWjmMmmTMz1Tz9/BcZvzCCUjN0te2jsX7pAnGqvHWhEJ868jzvmVsg/u/KKt65ZWvaC8SFWBtxLBxHE5niKI75nimO4uikTDc4msgUR3HM90xxFEcnZbrB0USmCUe3stx2EwrYAPwQ8CalVNm1UxJyxefzseeGPfh8PsdmimPhOKabeeVqJ0899yUmJtvx+cbp63kf66seTCuv7/IlvvLsE+y+eIEZ5eEzm9q4p62LyRW2l8jGMRPcUBtxLBxHE5niKI75nimO4uikTDc4msgUR3HM90xxFEcnZbrB0USmCUe3kmq7iR3A24GfAaoADcwC/wb8JfAvWuuQbDfhbCzLIhKJODpTHJ2Zdz0yz43v5ciLdxKNBikuOkFv1x0EA6+tmhcNh3nrq6e47dQYFjAWLOLu9m5OlJTY7uiEPBOZ4ujMPLdkiqMz80xkusHRRKY4OjPPRKY4OjdTHJ2ZZyLTDY4mMsXRmXkmMsXRuZkmHN3A4u0mliwSJ76glBf4YeCtwI8BQWILxheAfwRuAlplkVgQhGzRGk6/+gu8fOpdAJSXPUpX2z683sllv2fv+DluPT3GxulpNFAcjQLwH1XV/N6mNqYsxw9JgiAIgiAIgiAIgiAI15V09iQGQGsd1lr/i9b6Z4htPXE7cBAon/tzK4BS6pNKqQGz2kIm+P1+fuDNP4Df73dspjgWjuNymdGol2Mn7kgsEG/c8A/0dv7WqgvE9x49ROvkJEXRaGKB+B9qavlEW2dOC8SFWBtxLBxHE5niKI75nimO4uikTDc4msgUR3HM90xxFEcnZbrB0USmCUe3suyTxMt+g1INxJ4ufivQTezpYoCjwF9prT9uq6ENFNqTxEopikuKmZyYJNP6XqtMcSwcx1SZodBaXjh6L5evDAARWps/z8YN/7RqzgNPP07r1NJF5OPFxdzaN2irY664oTbiWDiOJjLFURzzPVMcxdFJmW5wNJEpjuKY75niKI5OynSDo4lME45uIe3tJtJBKbUNeBvws0ANoJ24/UShLRILgpOZnKrn+SOfZnq6HsuaoKv9birKvrvq95WGw3z9+4+Q6lfRTXos3rxz2H5ZQRAEQRAEQRAEQRCEPCTt7SbSQWv9hNb614A64EeAv8nRT7ABn8/H8A3Dtv/2SDszxbFwHJMzr07s4OnnvsT0dD2BwBn6e385rQXilskJ/uS5J1MuEAOcCQZscyyk2ohj4TiayBRHccz3THEURydlusHRRKY4imO+Z4qjODop0w2OJjJNOLoVa9++fTmH7Nu3T+/bt+/Fffv2fS13Jfu577P37SsvK7/eGteUyYlJ2x+VtztTHAvD8dz4Xl448lEOfPsnOPPaDxPVQdaUPseWnl+jKHh21e9/07nX+Z0jL1ARCnHB66Vobh/iZD67qY2TRcU5uxZabUzkmcgUR+dmiqM45numOIqjkzLd4GgiUxzFMd8zxVEcnZTpBkcTmSYc3cDJ0ycf3bdv3zfif89puwm3INtNCML14dz4Xg4dvXfJ611td7O+6psrfq83GuW9Yyf4qdfOAPDddWV8vK2LviuXuO30GLXTM5wJBri/vomDFVVG/AVBEARBEARBEARBEPIRW7ebEJyJ1+dl245teH3L/eP8658pjoXhOHbqtpSvn3z17St+3/qZGT7/wjOJBeI/r2vgg12bueTzcbCiitsHd/HhX3kvtw/usm2BuNBqYyLPRKY4OjdTHMUx3zPFURydlOkGRxOZ4iiO+Z4pjuLopEw3OJrINOHoVuQI5CHRSJQ3Xn+DaGTpP8t3SqY45r9jKLSOyammlF+bma5d9vu2XbrI3ccOUx4OccXycm9bB4+UVxpxNJkpjuKY75niKI75nimO4uikTDc4msgUR3HM90xxFEcnZbrB0USmCUe3IttNCIJgO1PTG3n+8GeYmm5I+fXi4uMM9t268EWt+flXT3P7qZexgGPFJdzZ0c2ZYJF5YUEQBEEQBEEQBEEQhAJCtpsoACyvRffmbiyv5dhMccxfx8tXunn6uT9iaroBr/dCyvc01d+/4O8l4TD3Hj3Eu+cWiP+9qpr3bO5fdoHYiT+36TwTmeJYOI4mMsVRHPM9UxzF0UmZbnA0kSmO4pjvmeIojk7KdIOjiUwTjm5FtpvIRzSEw2Gw8yFxuzPF0Zl5OWaeHx/h8IsfJRoNUlpyhN7OD3H5ag8nT9/GzGwdAf8rNNbfT1XFwcT3tExO8PGjh2iYnmJWKT7f3MrXqzeAUkYcr1mmODozz0SmGxxNZIqjM/NMZLrB0USmODozz0SmODo3UxydmWci0w2OJjLF0Zl5JjLF0bmZJhxdimw3IQiCLbx69qc4/vL/AjyUlz1Cd/s+LGtqxe9507nX+cCJYxRFo7zmD3BXRzeHS9dcG2FBEARBEARBEARBEIQCRbabKAAsy6K5tRnLsvFxfpszxTF/HLVWnBj7ZY6//H7Aw4bqr9Pb+eEFC8SLM73RKO976UXuevEIRdEo31tXxju3bE17gdgJP/e1zjORKY6F42giUxzFMd8zxVEcnZTpBkcTmeIojvmeKY7i6KRMNziayDTh6FZku4k8RCnF2rVrUSv9c/3rnCmO+eEYjfo58uKHOTd+MwDNDX9M/ca/XLJTRHLm+pkZ7jl2iM1XrwDw53UNPFDfRDSDn+F6/9zXI89EpjgWjqOJTHEUx3zPFEdxdFKmGxxNZIqjOOZ7pjiKo5My3eBoItOEo1uR7SYEQciKUGgtLxz9BJev9KFUiI7WT1Jd9V8rfs/WSxfZd+ww5eEQVyyL327r5OHyymtkLAiCIAiCIAiCIAiCIIBsN1EQeDweNtRuwOOxr7x2Z4qjux2npmt5+vk/4PKVPizrCpu7PrDsAvHe8XM88MwT/Pd3H+Jzh56lPBziWHEJt2/ZmvUCsdRGHJ2U6QZHE5niKI75nimO4uikTDc4msgUR3HM90xxFEcnZbrB0USmCUe34tgjoJT6glLqlFLqslLqFaXU55RS/rmvrVVK/dXc115TSt11vX2dhMfy0NjSiMeysRPanCmO7nW8crWLp5//I6ammwj4X6O/91coW/dkyvfuHT/HvUcP0To5gS8aJf6PN/5yYz2vBouMOTohUxzFMd8zxVEc8z1THMXRSZlucDSRKY7imO+Z4iiOTsp0g6OJTBOObsWx200opbqBk1rrCaVUFfD3wH9rre9VSv05UAP8LFAN7Afu1Fp/NVWWbDchCPZw/sIwh4/dTTRaREnxUXq77iDgP7/s+//iqe/TOD215PXjxcXc2jdoUlUQBEEQBEEQBEEQBEFYBtdsN6G1PqS1npj7qwKiQLtSqpjY4vCdWuuLWuujwBeAd1wnVcehlGJd2TrbNwa3M1Mc3ef46tn/wQtHfptotIjydY/R1/urKy4Q774wTkOKBWKA2ukZI45OyhRHccz3THEUx3zPFEdxdFKmGxxNZIqjOOZ7pjiKo5My3eBoItOEo1tx7CIxgFLqQ0qpq8DrQD+xxeBOwA88lfTWp4C+ZXM8Cq/XC8T2GvH6vPF8fH5f4n1+v3/Bn+MniM/nS+xN4vV6E4+gW5aFZVmxXMuzoA2fz5doY3FuOm3Ecy3LwvJm1kZRcRED2wfw+rz4/L5EG16fd/U2Fh+fuTa8fm8iM5V7qjYsr7WgjWT3YFGQge0D+Py+RA1SHauUx2eZNvxBf8LRjjr7A7G8QDCQU52TaxD/ub0+b1o1WK2NeF2KiotsqbNSasH5E2vPw0tj7+b4y78BWNSs/7/0bb4Tv28mdZ0ti7e+cpJPHnme5YbYs0XzW02k1dcWtRE/jv6AP+e+Fm8jEAzMnz851tnrSzp/igIZ12A59+TarFTn1ca7eBte39LzJ5v+nNzG0vMngzqnGFPjjsHiYEY1WGlMjdfaH/DnXOf4zz+wfQB/0G9LnX2+2Lg4sH2AYFEw4xosN94tuDYsHlOX6WsrtZF8/iz3c2R67QwWz18bMqlBvI3FfS1x/hTNnz9p1WCFNuLXmkAwkHOdlVLz50/Ab0ud/X7//PlTHMy4Bqnc447LzQGyaWO58yebOsdzF8wtMqjB4jbiNYg7BooCK9YgkzZSzS2yrXPy+eML+HKuczw31bUhrRpYqecv8fMx+fzJpc7xa2Dy+ZNLncH+OWT8WMRrHf85lqtBOm3Ex8f4HNKO+5HE+TPXfq5zyGXnFmnWOVU/SD5/cq1zvI2U14Yc7glznUOmamPFuUUWbaQ7h8zkupa4Nvi9GdcgVRsJR3+ac8g02oiPufHzx457/8XnTy5zSIBA0fz5k+094eK+tnhukUudYekcMpc6J483Gc8tVlpfSGMOmen8Jfn8ybXOJuaQsPL6Qrr3hMltrLa+kM08dfH9oR3rbOnOLdJdZ0vMIQMrzyHTqbNd9/529bXF7ovbWIyjF4m11p/UWpcCPcCXgLNAKTChtQ4nvfUisGbx9yulfkgpdZ+yFN2buwGo3lDNth3bAFhXto6h0SEgdnEfvXk0cYD33LiH0jWlAGzfvZ3K9bFfsLV5YDP1DfUAtHe109rRCkBDUwO9fb0ArK9ez+Cu2D+lX7N2DcM3DAOxE2n05tFEgYb3DrNmXUx7285trK9ZD0BvXy8NTQ0AtHa00t7ZDkBdQx1bBrYAUFlVyfah7QCUlJYwctMIEDthhkaHeOw7jxGaDTE0MsS68nUAbN2+lZraGgC6N3fTtKkJgE1tm+js7gRgY/1G+rf1A1BRWcHOPTtj7l4fXq+XcCiMx+Nh9ObRxIVx155dlFeWA9C/rZ/auloAunq6aG5tBqCppYmu3i4ANmzcwOb+zRzYf4CSkhJ2j+xeUIP4CT968yjFJcUA7BjeQUVlBQB9A33UNdQB0NHVQUt7bCuR2o21XDh/gdBsiOqaarbtjNV57bq1DO+N1SAQiLUR7wzJdR7cPUjV+qpYnfs3U1Nbw4H9B2ja1ERbZxsA9Y31bO7fDEDV+ioGd8fqXLqmlD037okdK1+szoFAIFHntevWArBl6xaOPH+E0GyIni09NDTH6tzS3kJHV0eizn0DfYka7BjeAUBxSTGjN4/G6mxZsWPlsTiw/wCDuwYpKy9L1HnDxg2xGvR20dQSq3NzazNdPbEa1NbVJupcXlnOrj27gNhFZmh0iIMPHiQcCrN79EaOndjH6TM/HztvOv+S9pZPs3X7ZjbWbwSgs7uTTW2bYvWo38hnz5zkl06N4QEONcXaXsz3fvRHEgPmyE0jlJSWALB9aDuVVbG+tmVgS6LO7Z0L+1pHVwcH9h+grLwsUec169Yk6uz3+2N1nhsYh28YZs3aWF8b3DXI+uqlfa2xuZHXzrxGaDZEfUM9mwdida5cX8n23duX1Nnr9cbqPHfzNzQ6xLqyWF/btmMb5RXlHNh/gPaOdho3NQJL+1rf1rk6V1Wwc3jngjorpZb0tcGdgzz75LOEZkMMDA4k6tzZ25noa80tzQv62sDgAABlFWXs3rM7UefRm0eJRCIc2H+AodGhRF/bObyTiqq5vra1L1Hnju75vtbY3Ej3ltRj6o7dOziw/wAeFXOPX6gWj6mJvrZoTG3riPW1hsYGevt7Cc2GOPTcIfq3xs7XBWPqXF+LT4CT+1rymNrT17NgTG1uaebA/gNU11SzZeuWRA12DMX6WklpyZK+Fq/B0MjQgr5WU1tDaDbE+Llx6urq5uvcE6tzcl9LHlOLiosSdVZKMXrzaGKxY9eeXZSuKeXA/gP0bOlJjKmdPZ1sao31taaWJro2x+pcU1vD1u1bY3UuL0uMqfE6x8fUodEhHn/0cUKzIXYM7UjUecvAFurq58fU1vb5vtbT1xOrc001gztj4118TA3Nhnjk4CMMjQ6lHlN3DVJVHatzb38vDY2xGrR1ts2PqUl9rWp9FQPbBjiw/wCBQGC+r/m8y46p23Zso3pDdazOW3pobI71tfiYGpoNcfzY8cQcoLKqcsUxNRiM1Xn3yG7KKmJ1Tu5rXb1d1NXXcWD/Aeoa6haMqfG+Vl5Rzq6R+TF19OZRPB5Pos7FxXN9bU+sr4VmQ1y9fDUxLiWPqY2bGhPuyXVeaf4yctMIfr+fA/sPMLBtIDF/2TKwZfX5S836+TE1qa8pYh+2q7mP/pLH1BXnL13tiTon5i9zY2poNsTjjz2eOF9XGlMXz1+WG1N7NvdwYP8B1qxdk+hrxcXLj6m7RnZRXlGeqPPi+UtoNsSpsVO0tcfO1yVjaryvBYMrzl/i17W+rX2sr1nPgf0H2NS6acGY2rNlrq8ljamp5i+Lx9TQbIjZmVnK1sXO1+QxNbmvxcdUgKrqqsQ8NdX8RSnFgf0H2DG0I9HXBncOUl0z19eSx9T21vn5S/38PDV5TPX7Yh/IhWZDS8bU3SO7l4ypAF2b5+cvm1qXjqmh2RDPPvVs4litNKbu3LMzMYdMnqd29sz3tbq6OsbPjROaDS0ZU4dGYn1t8Zg6evNoYv6yYEzdGhtTQ7MhXjvzWmJcSh5Tk/tacp39Af+CeWpyX+vf2s+h5w4Rmg0tHFM72hb0teQxNeX8JWlMDc2GQJMYl5LH1O4t3QvH1O5YnVebv4RmQxz85kGGRofm67xn+TG1uaUZiM1TO3tjdU7ua6WlpUQjUUKzodRj6irzl1Rjamg2xJEXjiTO11zvCZtbmjnzyhlCsyHb7glDsyEujl+ktjZ2vmZzT5h8Xdu+aztPP/E0odlQTveEyX0tEo7EFt2t2LHK5p4wua8N7hrkwP4DWB4rp3vChuaGxJhaXl7O1NQUodlQ1veE9Y0Lx9TQbIiXjr+UOFbZ3BNu27ktMaZ2dHYw9tIYodlQ1veE8WtnfEwNzYaYnJhMXH+yuSdM7mtDo0N875HvEZoNLRlTU/W1pk1Nqecv5esSY6rliS1uRaNRILt7wuT5S99AHwf2HyBYFMzpnjB5/lKzoYZLFy8Rmg1lfU+YGFM3d7OxbiMH9h+gvrE+p3vC3SO7E/OX3r5ejh0+Rmg2lPU9YXxxdvTmUfy+2Bxy6/atOd0TDo/O97Wh0SEeOvAQodlQ1veEsHD+EgwGUUoRmg1lfU+YGFPr6+jq7eLA/gOsXbs263vCeJ0XjKk53hOmGlNzvSccvXmUkpL5+ctiHLsn8WKUUm8B3gX8JvA44I8vFCulfhD4e611earvbahr0O0t7YTDsQVOj+UhHAonPmULzYaA2CAyOzub+HMoFEJrjc/nIxKJEI1GYwOZjhKNRBMHORKJ4LE8eJQn0YZlWYRCocSnBMm56bShtSYSicTaUBAJZ9aGUoqZmRl8fh/hUBitNV6fl2gkunIbi4+P15too6SkhKtXr6Z0D4eXtmF5LdAk2lBKLaiB5bGYnZ1N1CDVz5Hy+KzQRiAYYHJi0rY6+7yxny2XOi+ugc/rY2pqKq0arNZGKBTC7/cT1VHb6hz/xOrqhJ/Dxz7BxUv9KBWmq/13WV/578vWuWbiKp84eoiWyQlCSvG5lnb+fWMdQ6+d5bbTJ6mdmeZMIMD99U08tmFjZn0txbkUPy659rXk/hwIBJicnLSlztFINNFWNjVYzt2yLKanp1es8+K+trgfJLcRCATQWufcn5Pb8CgPMzMzto2pwaJUkV9zAAAgAElEQVQgkXAkoxqsNqZ6LS+hcMiWOkejUYpLipmZnrGtzuFwGL/fTyQayaoGy413Ho+H6enppWNqdPm+tlIb8fPHzmunx+NhdnY24xosN94VFRURCocyq8EqbXi93ti1wYY6a60pKSlhamrKtjonrg3RqG3zl9LS0mXnANm2EQwGiUajttR5ydwiwxosN94VFRcRmg2lPaam04bP5yMcCdtSZ601JaUlTE1O2TpPDQQCmc8tLA8eT+r5y+I5ZK51DofCBINBIpHlrw2ZXtfsnkNGI1GKi4uZmZlJWefk45NuG5bXYnpq2tb7kdLSUiYmJmybQ6acW6RZ5+X6QXFJMRNXJ2ypc7yNJdeGHO8Jc5lDLtfGsnOLLOeplmf1OWSm17Xk88euOWSqOUCm94TJbRQVFTExMWHrvX9RURGhUMiWOaTXG/vXv1OTUzndEy7ua8lzi1zr7Pf7UR7FzPSMLXWO1yDjucUKbaQzh8xm/hIMBhPXhlzrbGIOudL6Qib3hMltxOeQdtU5eX3KznW2dOYWmayzZbK+YOc6m13XtWznLw8efNAdexKnwAe0A0eAELHtJ+IMAM8u9406qhMLfdFobMILxCYvcwcNSBy0+J/jC+ihUCjxCVc4HCYaif05EokQiURiuZHogjZCoVCijcW56bQRz41EIkTCmbUB808ShWZDiTbCofDqbSw+PnNt+Hw+do3sSnzivdg9VRuRcGRBG8nuHuVh5KYRfD5fogapjlXK47NMG5YVe4o6MajlWGev5WXkphEsy8qpzsk18CgPwzcM4/f706rBam34fD5GbhpBoWyp8xvnR3ns8T/hwMP/ymOP/xMXL/VjWVfp7foAVRX/tmydt50/x588+xQtkxOc9/l4X08f/7K+hnAozMGKKt61fTefuesu3rV9iIMVVZn3tUXnkscTO3/ig+tqNUinDcuyGNo7d/7kWOdwKIzXGzt/PB5PRjVYqR8opdhz457Y+bNCnVcb7+Jt+P3+xBMnufTn5DYUKjH+2DGm+v1+9tywJ/FPYtKtwUpjquWxYueP5c25ztFoFL/fz9Do0IIJTS51DoVCib7tUZ6Ma7DceAckzp8lY+oyfW2lNpLPn+V+jkyvnUqpxLUhkxrE21jc1/x+P8M3DONRnsxqsEIblmXNXxtyrLPWGr/fz+7R3Ymn1nKt84Jrg1K2zF/8fv+Kc4Bs2vD7/YmncFL9HNnMXxbMLTKoweI24jXw+/0M7x1O/PO+1cbUdNqIXxssj5VznRPnz8j8+WPHPDXetzOeW0RSz19SzSFzqXM4FE6cPytdGzK5rtk9h4xG5q4NexdeG5arQTptKKXYc8OexBzSjvuReN+OP02ZyxxyxblFmnVO1Q98Ph+7R3bj9/tzrnO8jZTXhhzuCXOdQ6ZqY8W5RRZtKNKbQ2ZyXVt8/tg1h0w1B8jknjC5Da/Xy+7R+fPHjnv/xPmTdG3Idg4JsX8iPrw3dn+Y7T3h4r62eG6RS53jjNw4smBxKd0aLDemZjW3WKGNdOaQmc5fvJZ3wbUhlzqbmEPCyusL6d4TJreRPIe0o85xku8P7VhnS3duke46W6brC3ats9l1XVutBiu1sRhHPkmslCoF3gL8E3AJ2Az8LfAdrfUvKaW+ClQBPwdUA/uBu7TWX02VV19br1uaW66Ju1NIvkFyaqY4Oivv3PheDh29d8nrLU2fp672H1J/k9b87JlXeNfJl7CA50vXcGdHN+f9AWOepvJMZIqjM/NMZLrB0USmODozz0SmGxxNZIqjM/NMZIqjczPF0Zl5JjLd4GgiUxydmWciUxydm2nC0Q0cfPTggieJnbpIXAL8M7ANCBD7xXX/CNyttZ5USq0F/hj4UWAK+KLW+mPL5RXiIrEgZMrjzzzA5GTrkteLi48z2HfrktcDkQgfPHGMN51/A4B/XV/DfZvaCHnc9A8UBEEQBEEQBEEQBEEQCo/Fi8SOXM3RWk9ord+kta7UWpdqrVu01r+ptZ6c+/plrfXPaa3XaK2rV1ogLkT8fj83/T83LfgNhk7LFEfnOU5N1ad8fWa6dslrNTPT/MHzT/Om828QVor7mlv5VEv7sgvETv65TWWKozjme6Y4imO+Z4qjODop0w2OJjLFURzzPVMcxdFJmW5wNJFpwtGtOPJJYrsptCeJlVKUrinl6pWr2FVfuzPF0TmOWsPpMz/Hyyffk/Lri58k3nrpIvccO0RZOMwFr4+PdnTz9Np1xj1N5pnIFEdxzPdMcRTHfM8UR3F0UqYbHE1kiqM45numOIqjkzLd4Ggi04SjW3DFdhN2U2iLxIKQLlorToy9l1fPvmXZ93R3fISqioOgNT999lV+ZewEXuBwSSl3dvTwemDp/sOCIAiCIAiCIAiCIAiCc3HFdhNCbvh8PnaP7k78plknZorj9XeMRn0cPnZ3YoF4U9MX6e74CCXFJ/D5ZigpPpFYIPZHo/zW8aO8b26B+BtV1by3ty/tBWIn/dzXKlMcxTHfM8VRHPM9UxzF0UmZbnA0kSmO4pjvmeIojk7KdIOjiUwTjm7F2rdv3/V2MM59n71vX3lZ+fXWuKbMTM8wcXXC1kfl7c4Ux+vnGA6X8PyRT3Lh4jBKhehsu5famv9LcdFJ6mq/ztaB/6Bs7V9TFByjemaGzxx+jqFLFwgDf9DUwpcam4lk+AvqnPBzX+tMcRTHfM8UR3HM90xxFEcnZbrB0USmOIpjvmeKozg6KdMNjiYyTTi6gZOnTz66b9++b8T/LttNCEKBMTtbyXOHP83EZDuWZ5Luzo9Qvu7xlO/dcvkSHz92iIpQiIteL/vau3hiXWF94CIIgiAIgiAIgiAIgpBvyHYTBYDX62VgcACv1+vYTHG8Po6TUw089dwfMjHZjs93nr7eX12wQLx3/BwPPPME//n4o/zjE4/x+ReeoSIU4lhxCe/avDXrBeLr/XNfj0xxFMd8zxRHccz3THEURydlusHRRKY4imO+Z4qjODop0w2OJjJNOLoVOQJ5SFRHGR8fJ6qjjs0Ux2vvePlKN88f+RThcBlFwVP0dn2AouCZxNf3jp/j3qOHEn8Pzv3/mdI1fKB7C9OWdU08r0eeiUxxFMd8zxRHccz3THEURydlusHRRKY4imO+Z4qjODop0w2OJjJNOLoV2W5CEAqA8Qu7OXTsY0SjQUpLDtHb9UH8vksL3vPAM4/TOjm55HuPFxVza//gtVIVBEEQBEEQBEEQBEEQDCPbTRQAlmXR2dOJlcOTn6YzxfHaOZ59/f/l+SOfIBoNUl72KH0971+yQAxQNz2d8vtrZ2auief1zDORKY7imO+Z4iiO+Z4pjuLopEw3OJrIFEdxzPdMcRRHJ2W6wdFEpglHtyLbTQhCnqI1nHr1bYyduh2A6qp/p73l03g8kSXvveH8OfzR1P+04kwwYNRTEARBEARBEARBEARBuL7IdhOCkIdo7eH4y+/jzGs/CUDDxv9NU8OXUWrJG7nllZO84/TJZbM+0tHNwYoqg7aCIAiCIAiCIAiCIAjCtUS2mygAPJaHxk2NeCz7ymt3pjiac4xG/Rw6ds/cAnGUlubP0dy4dIE4EImw79jhxALx326o4672Lo4XlzDl9XK8uMS2BWKpjT2Ioz0UoqOJTHEUx3zPFEdxdFKmGxxNZIqjOOZ7pjiKo5My3eBoItOEo1uRI5CHeJSHiooKPMrGTmhzpjiacQyHS3n20O9yfvwGlJqlq30fdRu+tuT71s/M8MUXnuHm8XOElOJTLe38QXMLByrXc/u2ndzxrvdw+7adtj1BLLWxB3G0h0J0NJEpjuKY75niKI5OynSDo4lMcRTHfM8UR3F0UqYbHE1kmnB0K7LdhCDkCTMz63nu8GeYnGrBsq7S0/FhytY9teR93Vcu84mjL1AZCnHR6+XOjh6eWbvuOhgLgiAIgiAIgiAIgiAI1wPZbqIA8Hg8VNdU4/HY+EmNzZnimHvmufG9PPHMAzzyvW/w/af+kiee/QqTUy34fefo73lvygXiN73xOp9/4RkqQyGOFxfzS5u3LlkgdvrPbSLPRKY4imO+Z4qjOOZ7pjiKo5My3eBoIlMcxTHfM8VRHJ2U6QZHE5kmHN2KHIE8xLIsWjpasCzLsZnimFvmufG9HDp6LxOTrYQjRUxNNxAOl+H3vU7/5vdQUnJiwfuV1rzr5EvcdfwIAa05WF7JL/f2czYYNOZoMlMcxdFJmW5wNJEpjuKY75niKI5OynSDo4lMcRTHfM8UR3F0UqYbHE1kmnB0K7LdhCC4kMefeYDJydYlrxcXvcRg/y0LXiuKhPnoi0fYc2EcgK9ubOArDU3oxb/JThAEQRAEQRAEQRAEQSgIZLuJAkApxZq1a1A2LgLanSmOuWXOTNemfn2mZsHfa6en+aPnnmbPhXFmlOJjbZ38aWPzigvETv65TeWZyBRHccz3THEUx3zPFEdxdFKmGxxNZIqjOOZ7pjiKo5My3eBoItOEo1uRReI8xOfzsX33dnw+n2MzxTH7TK3BY02l/FogeCbx5/7LF/nj556kZWqScz4//6u3n/1V1dfE0XSmOIqjkzLd4GgiUxzFMd8zxVEcnZTpBkcTmeIojvmeKY7i6KRMNziayPz/2TvzMLmqOn+/99bS+169r+lO7/uadCcdIIDijCi4wU/GBR1BBXcBAdEguIGArI7iGHV0RkRRQBFRmBkZEYFgQhMCCQlk3zp70unuWu7vj0rdVPWW7q463fdWvp/n4Xk6XdXveau+95x77qnLKRWOdo1sNyGR2CiGofHG5k+yfddFEz5eX3M9nuynOX/3Tj735kachsG6lFSur21g0J0wx7YSiUQikUgkEolEIpFIJBIrRrabOE2i4hOQWDPFcWYJBBys33iduUCcl/sYyckbcTiGSE7eSH3N9eRn/YVPv7mRq954Hadh8GSOh083tMx4gdhKr3uueCqY4mhNngqmHRxVMMXRmjwVTDs4qmCKozV5KpjiaF2mOFqTp4JpB0cVTHG0Jk8FUxyty5S7iIORReI4jNvtZtnZy3C73ZZliuPMmH5/Iq+s/yZ7Bt+KpvmoqbqZ2qpv0dt1OTfccCu9XZdTkf4Ut7z6Mu/ZtQOAH5aUc+PCOkZm+A2dVnrdc8VTwRRHcYx3pjiKY7wzxVEcrcS0g6MKpjiKY7wzxVEcrcS0g6MKpgpHu0a2m4jTJCQkMDIyYmmmOE4vXm86a1/7FkeONqHrw9RXf4XsrGdZtn+QS7dtpnBkhEGXm2S/H493lOO6ztcX1vKXbM+cOc4HUxytyVPBFEfrMsXRmjwVTDs4qmCKozV5KpjiaF2mOFqTp4JpB0cVTHG0Jk8FUxyty1ThaIeM3W5CFoklEgtnZCSPgVe/w/HjFTgdh2msu4b0tLUs2z/IzevXjXv+QaeTz9c383pK6jzYSiQSiUQikUgkEolEIpFI7BDZk/g0iMvt4oxzzsDljuG3R8aYKY6nZg4dL2f12vs4frwCt3sPLY1Xkp62FoBLt22e8G8OOl1RLxDP9+ueD54KpjiKY7wzxVEc450pjuJoJaYdHFUwxVEc450pjuJoJaYdHFUwVTjaNXIncRxG0zTSM9I5fOgwsapvrJniODXz8JF61r52Cz5fBkmJm2mq/wKJCXvMxx9/7hmSA/5xfzekOzivp29OHOeTKY7iaCWmHRxVMMVRHOOdKY7iaCWmHRxVMMVRHOOdKY7iaCWmHRxVMFU42iWy3YREYvHsP9jDuvU3EQgkkZbyCo111+ByHTIfT/L7eGjVc6RMsEi8MTmZS1s651JXIpFIJBKJRCKRSCQSiURis8h2E6dBnC4nPUt6cLqclmWK48TMPYPn8spr3yIQSCIr4+80N3wuYoE4b2SEe9a+NOECMcCPSsqVO1qBKY7iaCWmHRxVMMVRHOOdKY7iaCWmHRxVMMVRHOOdKY7iaCWmHRxVMFU42jWOFStWzLeD8tx+2+0rsjKz5ltjTuPz+jh65GhMb5WPNVMcI5nbd76H19+4GtDJzfkTddUrcDhGzcdrjh7hu+sGKB0ZZp/LxU9Kysj2ekkyDLYkJnHbgoU8ne1R6mglpjiKo5WYdnBUwRRHcYx3pjiKo5WYdnBUwRRHcYx3pjiKo5WYdnBUwVThaIds2bbl2RUrVvwx9G/ZbkIimecYBmze+jG27vgAAEUFv6Ky/G407WTfXLp/Hze8/ipJgQCbkpK5pq6R3QmJ86UskUgkEolEIpFIJBKJRCKxcWS7idMgTqeTlvYWnM4Y3s4fY6Y4BpmG4WDDpqvNBeLy0h9QWX7XyQViw+B9O7dx8/pXSAoE+HtGFlc0tpoLxHZ4H1UwxVEcrcS0g6MKpjiKY7wzxVEcrcS0g6MKpjiKY7wzxVEcrcS0g6MKpgpHu0begTiMYRgcPhzbb2WMNVMcDfwBN69t+Cr7DvQDfqorv0NB3u/N5zoCAT775kbeuWcXAL/JL+Suiir8mjZnjlZliqM4WolpB0cVTHEUx3hniqM4WolpB0cVTHEUx3hniqM4WolpB0cVTBWOdo1sNyGRzEN8vlTWvvZNDh9pRdNGqKu+EU/2/5mPp/h83LhhHT2HDhIA7i2v5MGCIghbIJZIJBKJRCKRSCQSiUQikUhmE9lu4jSIw+Gguq4ah8NhWebp6Di4fxkvvrSSv73wBM+u+g2Hj7TicBylqf6LEQvEBcPD3Lt2DT2HDnJc1/lyTQMPFhZPuEBsh/dRBVMcxdFKTDs4qmCKozjGO1McxdFKTDs4qmCKozjGO1McxdFKTDs4qmCqcLRrZLuJeIwW3FOFWN50GmvmaeY4uH8Z69bfPO73ZcU/JjN9jfnv+iOH+eb6V8j2etnrcnNtXSPrU1LnxNFWTHG0Jk8FUxytyxRHa/JUMO3gqIIpjtbkqWCKo3WZ4mhNngqmHRxVMMXRmjwVTHG0LlOFo00j201IJHOQVS+tZGioatzvk5M30tlyKQBn7NvLl19fT4IRYENyCl+qbWRvQsJcq0okEolEIpFIJBKJRCKRSOI8tthuQtO0BE3T7tc07Q1N045omvaqpmkfCXs8XdO0/9Q07bCmabs1TbthPn2tFl3XKS0vRddjV95YM083x+Hh4gl/PzJcCIbBJdu3ctOGV0kwAvw1M5srG1untUBsh/dRBVMcxdFKTDs4qmCKozjGO1McxdFKTDs4qmCKozjGO1McxdFKTDs4qmCqcLRrrPoOOIGdwDlAOvBh4DZN095y4vG7gWygDOgHPqZp2gfnwdOS0R06uXm56I4YdsIYM08nx337lxIIuCd8LDVhM9ds2sDlW98E4MGCIq6vbeD4NPfCscP7qIIpjuJoJaYdHFUwxVEc450pjuJoJaYdHFUwxVEc450pjuJoJaYdHFUwVTjaNbbZbkLTtIeAl4FvAQeAJYZhvHDisauAtxuGccZEf1taXGpUV1bj8/nQdR3doePz+tA0DafLiXfUC4Db7WZ0dNT82ev1YhgGLpcLv99PIBDA6XQSMAIE/AFzU2u/34/u0NE13WzD4XDg9XrRNA2XyxXBnU4bhmHg9/uDbWjg982uDZfbhc/rwzAMnC4nAX9g6jbGvj9O56RthLv7fOPbcDgdYGC2oWnalDU4VRvm+xNFG3Nd51173saGTVcD4xd9MznAH5I7WDz0Jn7gnspqfp1XEOF+qjZMdxvVeVY1mKKNaPtarPrzfNZgbBvR9rV5q3MMayBj6gxq4HISCJy6r1mxzspqEI91nqP5i63qPEfzF7vWecY1cOjo+vTmL3as86xqMM35i63qrHBMnbQGVq+zXBPKNaHiGsg1oVwTyjWhXBOqqsFTTz9l/e0mxkbTtESgB3gJqAXcwOqwp6wGWib4u7dqmna75tCob6oHIK8gj47uDgAyMjPo7e8FICExgf7l/cHNqoElZy4hNS34hWFdi7vIyc0BoKmtiZLSEgCq66qpqgnuM1taXkpjSyMAuXm5dC7qBCAtPY2+M/qA4IHUv7wftzt4V2nfsj7SMtIA6OjpIDc/F4DGlkZKy0sBqKqporq2GoDi0mKa25oByPHk0NXbBUBKagpLz1oKgMPpoH95P4XFhei6Tu/SXjKyMgBo72onvzAfgPqmesoXlAOwYOECautrASgqKaK1oxWA7Jxsepb0mG30L+/H4XCg6zr9y/tJTEoEYNGSRWTlZAHQ2tFKYXEhAHUNdVRUVQBQXllOXWMdAAVFBbR3tePJ9ZCdk83ipYsjahAaFPqX95OckgxAd1832TnZALS0tVBcGty+oaauhsrq4H7TZQvKaOtqQ9d18vLz6OgJ1jk9I52+ZcEaJCQE23C5XOPq3Lm4E0+uJ1jn1ibKysvw5HqorqtmYe1CAErKSmhqbQLAk+uhc3GwzqlpqSw5c0mwzq5gnXftuYQNm64FHJQU/5m66hvweHbidA7TlPgUL6XUs3joTYZ0B3efcy6vLD/LrHNLW4tZg+6+bgCSU5LpX94frLMjWOek5CQ8uR4WL11MZlamWeeCooJgDRrrKK8M1rmiqoK6hmANCosLzTpn5WSxaMkiABKTEulf3k9uXm5EGwA9S3rMGrR2tFJUUgRAbX0tCxYuCNZ5QbnZ1/IL82nvag+2kZ1F37I+dF0nMTHYhsMZrPPSs5aSkpoCQFdvFzmeYF9rbms261xdO6avtTbiyfUE+/OJOqdlpJl1drvdwTq7g3XuO6OPtPRgX+tc1Elu3vi+trB2Ic1tzei6TklpCU1twTrn5ObQtbhrXJ2dTif9y/tJSAxuC9Lb30tGZrCvdXR3UFBUgCfXQ0NzA2ULyoDxfa2l/USdPdn09PVE1FnTtHF9bfHSxZQvKEfXddo628w61zbWmn2torIioq+1dbYBkJmdyeIliyPq7HQ58eR6IvpaT18P2Z4Tfa29xaxzTX1YX6soo7554jG1r78PT66HpKQksw0YP6aafW3MmLqwJtjXSsuCddZ1ncqFlROPqSf6mjvh5JianpEerEHYmNrQ0hAxptbU1eDJ9VBSWkJze7NZg+7eYF8LjXdwsq+FatC7tDeir+UX5qPrOq2drVRUVpysc0OwzuF9LXxMTUpOMuusaVpEX1u0ZBE5nhw8uR7aOtrMMbW2oZYFVSf6WmU5dU3BOof3tcysTHNMDdU5fEwtLi1G13W6e7vNOje3NVNccnJMrao+2dcaWhqCdc7Po7MnWIPQmKrrOgVFBZOPqYs68eQF69zY2khp2cm+Zo6pYX3Nk+sxj430jPSTfc11oq+d2IInos7dHeQV5AXr3NxAWUWwr1VWV1JTV4Ou69TU1Zh9LceTM+WYmph4sq9lZgfrHN7X6hqD5zVProcFCxdEjKmhvpaVncWipZFjqq7rZp2Tk0/0tSXBvqbrOp09neZ4Fz6mli0om3BMnWr+svSspaSlpwXfz96T85fmtuZTz1/yc0+OqWF9LdRGaLwLH1OnnL/UVZt1NucvJ8bU0Fgbmr9MNaaOnb9MNqa2drTiyfXgyfWYfS05efIxddHSRWRlZ5l1Hjt/0XWd+qZ6s6+NG1NDfS1xfF8Ln7+Ezmst7S0UlxXjyfVQW18bMaY2NJ/oa2Fj6kTzl7Fjqq7rLFqyyKxB+Jga3tdCYyqAJ89jjqkTzV8SkxLx5Hoi+lpnTyd5+Sf6WviYWh0cUwGKS07OU8PH1NS0VLMfjB1Tx85fQnWuazo5f1lQNX5M1XWdisoKc/4y1Zg6dv4SMaae6GsVlRW0dga5Y8fU3qXBvjbRmBqav0SMqe3BMVXXdZrbmk+e18LG1PC+Fl5nd4I7YkwdO3+pXFiJruuRY2rNwoi+NnZMHVvn8DFV13WWnLHErEH4mFrfXB85ptYH63yq+UtoHh5R5yVTjKknzp0VVRXUNgbrHN7XsnOy6V3WG5xDTjSmnmL+MtGYqus6VdVVMbsmrKmvoam1yfzflWNxTajrOu1d7WYNZntNGDqv9fb3UlZRZnJne00Y3tdC43aoBrO9JoST8xdPrsdsY7bXhKUVpeaYml+YT09vD7quz/qasKQsckzVdZ3a+tqIvjbVNeGE85eejogxtbahNjjnPTF/CdV5uteEoXNnaEzVdZ3uxd0nazDLa8LwvlZUUoSu61FdE2ZkZZhjaqjOoeu1WV0Ths9fFnXgyfUEr0eiuCYMn7+UlpXS0dOBruuzviY0x9Smesory/Hkeqisroz6mjB8/mIel7O8JgzdPdu/vJ/UtNTgPCoG14Rwsq/lFwSvmWZ7TQiR85fQtYKu67O+JjTH1JJiWtpbzDnkbK8JzTqHj6lRXhNONKbG4powJeXk/GVsLH8nsaZpGvAfQDFwNrAE+INhGKlhz+kG/mYYhnMixul2J3FySjJtXW08/8zzoBGTTzhcbhfdvd08/8zzeL3eqD/hSEhIoL2nnRf+9gIGRkw+NU5ITKBzUSfPP/O8WZNo6uxwOOhc3MmLf38Rr9c77TobBmzd8Uk2b70YgLKSBygvuY9l+/fyke1bKRkZxuHz4QB2uxO4praRzenps/rU2MCgu7eb1c+v5vjx4zH5JCspOck8fjRNi8knWW63m67erojjJ5pPjd1uN+097ax6dhWBQCAmnxonJibSsagjePwE/FF/mujQTxw/z72Id9Qbk08Tk5OTaesO1sbAiPpTY8Mw6O7rZvULqzk+dDwmn1gmJSWZjpqmRT2mappGd183/3j+HxwfOh6TT41dLhcdizpY9ewq/H5/1J8aOxwOuvu6WfX3VYwMj8TkU2NN0+jq7eIfz/2DkZGRmHxqHHFuOPE+R3MnMQbm8TN0bCgm586ExATau4PnBiDqT+51TQ8eP8/9g+PHj8fkrjeH00HnohPnBp836rsDnE5n8Ph5dhUjIyMxuTsAghdY/3j+H4wMj0Q9f3G5XHT3TT4HmM0cCSY+fqK5QyNibmEYUd+hoevB4+fF515k+PhwTO7EMZn+/+QAACAASURBVM8NYXOLaO4CCR0/Lzz7AqMjozGZp5rnhpnOLSa5k3iqOcBs78TR0EzHoaGhqO/EifUcMmAEcOjBc8OLf3+R4eHhqO8kTkpOor27PTgHMIyYXI+E+vYLf3uB0dHRmNwNNeHcIoo7icOPH5/PF5M7rsy5Rfi5IYprwmjnkBO1MeXcYhbXndOdQ85k/mKeG/72PN5Rb+zmkH97Ho1pzCGn0UZCQgKdizvN4ycW1/6hc0Po+In2zsPEpEQ6eoLHTyAQiMmdxGPnFtHe3Th2DhmLuxtnNbeYan1hGnPImc5fwo+fUJvR3GEa6zmkYRgkJSXR3nPy3BDtncSnWl+YzTw1/PgZu74w23W26c4tprvOluA+MYd89gWMwORzyNPhTmJLLxKfWCC+D+gCzjEM45Cmae3AKsBtGIbvxPPOAR40DCNrIk5JYYlRWVE5V9qS0ziG4WDDpi+we+/bAago+x6lRf/Fsv2D3Lx+3bjnf6uymsfyCuZaUyKRSCQSiUQikUgkEolEchrn6Weftsd2EycWiO8FFgFvMQzj0ImHXgO8QGvY09uAgbk1tG40TSMlNYXgW2hNZjw6BgJu1q2/8cQCsZ/qym9TWvRfAFy6dfOEf/PeXdvn1DFemOIojlZi2sFRBVMcxTHemeIojlZi2sFRBVMcxTHemeIojlZi2sFRBVOFo11j2UVi4B6CW0ucaxjGgdAvDcMYAh4AbtI0LUPTtGrgU8AP50fTenG5XCxeutjcY8mKzHhz9PmSefnVW9h3YBmaNkp9zVcpyPs9AO5AgPLh4xP+XeHwyJw5xhNTHMXRSkw7OKpgiqM4xjtTHMXRSkw7OKpgiqM4xjtTHMXRSkw7OKpgqnC0ayy53YSmaeXAm8AI4At76GeGYXxc07R04PvA24HjwD2GYXxtMp5sNyFRmVFvBmtfvZWjx+pw6EM01F5HZsaLAGR4R/n6+nW0HDk84d9uTE7m0pbOudSVSCQSiUQikUgkEolEIpGc5rHFdhOGYWw2DEMzDCPRMIzUsP8+fuLxw4Zh/D/DMNIMw8ibaoH4dE1os28rM+PBcXgkj5fW3svRY3U4nQdpbviMuUBcfnyI77+8hpYjhzmuT9zVflRSrtwxXpniaE2eCqY4WpcpjtbkqWDawVEFUxytyVPBFEfrMsXRmjwVTDs4qmCKozV5KpjiaF2mCkc7xpKLxJLo4na7OfPcM81vOLciMx4cjw2Vs2btfRwfLiPBvZvWxitIS30NgM5DB7jv5dUUjQyzPSGRjzW3c31NPZtSUhlxu9mUksr1NfU8ne1R6hivTHEURysx7eCogimO4hjvTHEURysx7eCogimO4hjvTHEURysx7eCogqnC0a6x5HYTsc7ptt2EpmkkJCYwMjxCrOoba6bdHQ8fqWfta7fg82WQlPQmzXVfJCFhDwDn797J5954HSewJi2dL9c0cOjE3jan4/uogimO4mglph0cVTDFURzjnSmO4mglph0cVTDFURzjnSmO4mglph0cVTBVONolY7ebkEViiWSGOXCwi1fW30wgkExayis01l2Ny3UY3TD4xJY3uGjndgD+6MnjlspqvJNsNSGRSCQSiUQikUgkEolEIpHMR2yxJ7EkurjcLvqX9+Nyx/DbI2PMtKvj3n1nsva1bxMIJJOZ8TzNDZ/D5TpMkt/PzetfMReIf1hSzterasYtEJ+O76MKpjiKo5WYdnBUwRRHcYx3pjiKo5WYdnBUwRRHcYx3pjiKo5WYdnBUwVThaNc4VqxYMd8OynP7bbevyMrMmm+NOYsRMDh86DBDx4Ysy7Sj487d72D9xmsBJ57sp6iv+SoOxwi5IyPctm6A9iOHGdE0bl5Yx28LikDT5tzxdGGKozhaiWkHRxVMcRTHeGeKozhaiWkHRxVMcRTHeGeKozhaiWkHRxVMFY52yZZtW55dsWLFH0P/lu0mJJJTxDBg644PsHnrxwAoyHuYhQvuQNMC1Bw7yjdfXUuud5QDThfX1TawNi19no0lEolEIpFIJBKJRCKRSCSSySPbTZwGcbqcdPd243Q5Lcu0uuPg/mW8+NJKnl31BH9f9Yi5QFxa/BMWLrgNTQuwdP8gd69dQ653lE1JyVze1HbKBeLT7X1UxRRHcbQS0w6OKpjiKI7xzhRHcbQS0w6OKpjiKI7xzhRHcbQS0w6OKpgqHO0a2W4iTmMYBkcOH4npNzPGmmlVx8H9y1i3/ma83mwCASeBQCIA+bm/o6riHjQMLt65nas3bcBtGDyXkclVdU0ccLvnzFElzy5McRRHKzHt4KiCKY7iGO9McRRHKzHt4KiCKY7iGO9McRRHKzHt4KiCqcLRDpHtJiSSU2TVSysZGqoa9/vk5I30NH2Iz7+5kfP37ALgN/mF3FVRhX+C/YclEolEIpFIJBKJRCKRSCQSK0a2mzgN4nQ6aWptwumM4e38MWZa2XF4uHDC3ycdT+TWV9dy/p5d+IE7yyu5Y4YLxKfT+6iSKY7iaCWmHRxVMMVRHOOdKY7iaCWmHRxVMMVRHOOdKY7iaCWmHRxVMFU42jXyDsRhDMNgaGgo5rfyx5JpVcfR0WwMwzHu95Vs5DHeQu3hgwzpOjdW1/G3rJx5cVTJswtTHMXRSkw7OKpgiqM4xjtTHMXRSkw7OKpgiqM4xjtTHMXRSkw7OKpgqnC0a2S7CYnkRIaHCxlYdxvDIyVcyEOsYAULeIPd5JHHHtI5yh63m2tqG9mYkjrfuhKJRCKRSCQSiUQikUgkEsmsIttNnAZxOBxU1VThcIy/I9YqTKs5HhuqZM3aexkeKeF9zn/nId5NCwOkcZSFbCKdo2xLSOTypvaoFojj/X2cK6Y4iqOVmHZwVMEUR3GMd6Y4iqOVmHZwVMEUR3GMd6Y4iqOVmHZwVMFU4WjXyHYT8RgNEhMTIZbfpRZrpoUcDx9pYu2r38bnTyM5aRM3G58H3/jnjeo6+9zueXGcM55dmOJoTZ4KpjhalymO1uSpYNrBUQVTHK3JU8EUR+syxdGaPBVMOziqYIqjNXkqmOJoXaYKR5tGtpuQnNbZf7CHdetvJhBIJC31ZRrrrubJF/9IciAw7rlDuoPzevrmwVIikUgkEolEIpFIJBKJRCKJXWS7idMguq5TXFqMrseuvLFmWsFxz+ByXnntWwQCiWRlPEtz/efJ5sCkHx7tTEyYc8e55tmFKY7iaCWmHRxVMMVRHOOdKY7iaCWmHRxVMMVRHOOdKY7iaCWmHRxVMFU42jXyDsRhdIdOQVEBuiOGnTDGzPl23LHrAl57/SsYhpPcnD/RUHsdJd4D3PfyGpImuIsY4Ecl5XPqOB88uzDFURytxLSDowqmOIpjvDPFURytxLSDowqmOIpjvDPFURytxLSDowqmCke7RrabkJxWMQzYsv1DbNn2UQAK839DVcV3aTx6iG++9gpZPi+73Qk8UFjMP+/dReHwCDsTE/hRSTlPZ3vm2V4ikUgkEolEIpFIJBKJRCKJPrLdxGkQTdPI9mSjabHbdTvWzPlwNAyNTZs/bS4QlxWvpKriDs7cv5c7Xxkgy+dlfXIKH29q5VeFxXyktYv3/9Pb+UhrV8wWiOPhfbQCUxzF0UpMOziqYIqjOMY7UxzF0UpMOziqYIqjOMY7UxzF0UpMOziqYKpwtGtkkTgO43Q6aWhqwOl0WpY5146BgIP1G69jx673AFBZfiflJT/i4p3buHHDOhKMAM9kZvOpxlb2uRPmxdEKPLswxVEcrcS0g6MKpjiKY7wzxVEcrcS0g6MKpjiKY7wzxVEcrcS0g6MKpgpHu0a2m5DEffz+BF7dcCP7D/YBPmoXfpPCnCf49JsbuXD3TgAeyi/k7ooq/PLJkUQikUgkEolEIpFIJBKJJM4j202cBtE0jaTkpJjfzh9L5lw5+nypvPzqd9h/sA9dG6Gx9jrKsx7nG6+t5cLdOwkAd5dX8t0JFojlfbQuUxzF0UpMOziqYIqjOMY7UxzF0UpMOziqYIqjOMY7UxzF0UpMOziqYKpwtGtkkTgO43K56FvWh8vlsixzLhxHR7N56ZU7OXykFYfjCE31n6cm5X+5e+0aeg8eYFjXuaGmngcLi2GCwUDeR+syxVEcrcS0g6MKpjiKY7wzxVEcrcS0g6MKpjiKY7wzxVEcrcS0g6MKpgpHu0a2m4jT6LpOIBCwNFOl4/BwIQPrbmN4pASXax9NdV+khZe45bWXyRsd5YDTxZdqG1iXlj5vjlbl2YUpjtbkqWCKo3WZ4mhNngqmHRxVMMXRmjwVTHG0LlMcrclTwbSDowqmOFqTp4IpjtZlqnC0Q8ZuNyG7MsdpVCz+x5qpyvHYUCUvr/sOo14PiQk7aKr/PMuG1/K1DetI8fvZnJjE1XWN7ExMmjdHK/PswhRHa/JUMMXRukxxtCZPBdMOjiqY4mhNngqmOFqXKY7W5Klg2sFRBVMcrclTwRRH6zJPhxtopxPZbiIO43a7Wf7W5bjdbssyY80b3L+Mfwz8mGee/xMvvvTvjHo9JCdtoqXxCt53aBXffvVlUvx+XkzP4BNNrdNaID4d30e7MMVRHK3EtIOjCqY4imO8M8VRHK3EtIOjCqY4imO8M8VRHK3EtIOjCqYKR7tGtpuIw4Q23T4+dDxmn4bEmhlL3uD+Zaxbf/O439dW3sgNwz/kkh3bAHjck8ctldX49Ol9NnK6vY92YoqjOFqJaQdHFUxxFMd4Z4qjOFqJaQdHFUxxFMd4Z4qjOFqJaQdHFUwVjnbJ2O0mZJFYYvusemklQ0NVEb9LYJifOd7Ne/yPAbCyuIyVJWUTfkGdRCKRSCQSiUQikUgkEolEcjpl7CKxbDcRh3G5XCw5c0nMvz0ylsxY8o4fL434dw6DPMnZvMf/GF5N4xtVNawsLZ/xAvHp9j7aiSmO4mglph0cVTDFURzjnSmO4mglph0cVTDFURzjnSmO4mglph0cVTBVONo1jhUrVsy3g/LcftvtK7Iys+ZbY85iGAbHjh7j2LFjlmXGgmcYsHX7Bzl4uJsLeYgHuIhbuYqruJUFvMlB0riuvob/y/bMm6Nqph0cVTDFURytxLSDowqmOIpjvDPFURytxLSDowqmOIpjvDPFURytxLSDowqmCke7ZMu2Lc+uWLHij6F/y3YTElvGMDQ2bb6CHbvex4U8xEO8e9xzvp57Ln+sGp4HO4lEIpFIJBKJRCKRSCQSicS6ke0mToM4XU46F3XidDkty4yGZxgO1m+8lh273gfAN1xXTvi8i4/9dd4c54ppB0cVTHEURysx7eCogimO4hjvTHEURysx7eCogimO4hjvTHEURysx7eCogqnC0a6R7SbiNLquc/jQ4Zh+M2OsmbPh+QNu1m34GoP7lwM+aiq/zvX7f4djgucm+gP8rLh0gkfUOs410w6OKpjiKI5WYtrBUQVTHMUx3pniKI5WYtrBUQVTHMUx3pniKI5WYtrBUQVThaMdIttNSGwbny+FV177BoeOtKNpIzRXf5mvH/gP3r5394TP35iczKUtnXNsKZFIJBKJRCKRSCQSiUQikVg7st3EaRCH00FDcwMO50T311qDOVPeqDeTl165k0NH2nE4jrK45lP8cPcPefve3fgn+ZsflZTPqeN8MO3gqIIpjuJoJaYdHFUwxVEc450pjuJoJaYdHFUwxVEc450pjuJoJaYdHFUwVTjaNbLhRjzGgNHRUYjlTeKxZs6ANzySz8vrbuP4cBku5wHOqfo492z5HZXHhziu66yorsNpGHxk2xaKR0fY7k7gRyVlPJ3tmTPHeWPawVEFUxytyVPBFEfrMsXRmjwVTDs4qmCKozV5KpjiaF2mOFqTp4JpB0cVTHG0Jk8FUxyty1ThaNNYcrsJTdOuBD4MNAN/MAzjgrDH0oF/A94OHAfuMQzjpql4st2EfTN0vJyBdbcxOppHgnsX7y7/MHe++d94vKMMutx8qa6R9Smp860pkUgkEolEIpFIJBKJRCKR2CZ22W5iB3AzcP8Ej90NZANlQD/wMU3TPjiHbpaPw+GgcmElDkcMb+ePMXM6vCNH61iz9m5GR/NISnqTT5a8i/s3/gmPd5RNScl8vKk1YoF4Phznm2kHRxVMcRRHKzHt4KiCKY7iGO9McRRHKzHt4KiCKY7iGO9McRRHKzHt4KiCqcLRrrHkIrFhGA8ZhvFbYDD895qmJQMXA182DOOgYRjrCS4af3QqnqZrOJ3BnTV0XcfpcoZ4uNwu83lutzviZ03TAHC5XOh68K1yOp3ojuDPDofDPIh0hx7RhsvlMtsYy51OGyGuw+Ew90WZdhsJblJSU8zXF2rD6XKeuo2x70+oDV0jNT3VZI11n6gNh9MR0UaEu9tFSmqK+fPY13HgUAcD676Lz5dJaso6bs5/J7du+hvJgQAvZGbxqeZ29iQkRrbhdJCaFnSMRZ0dDgcpqSkR79Vs6hxeg9Dr1jRtWjU4VRuarpGSmhLxOqKqs6ZFHD/R1jn8dYw9fqZTA7OvjWnD5Qq+j6GfT1WD6bThdDpPHj9R1tnpCjt+XLOowSTubvfJ2kxV51ONd+G1TUlNwZ0QmzqHuOHHz4zqPMGYGnKcaJyYqgZTjalOp9M8fqKts67raJpGalqqWdto6xxyT0lNiXgd063BpONd+Lkh/P1xTd7Xpmoj/PiZ6nWEajCtc2fYuWEmNQi1MbavmcdP2Ps+rRpM1YYz7NwQZZ01TTt5/ISdy6Kqs9t98vhxx2b+omlTzwFm08Zkx89s6hzONecWM6jB2DZCNQjVJvxcNtWYOp02JppbzLbO4cdP+Lks2nmqWZuZzi0cE89fppoDzHaeGu4YbZ1DP8dyDqk7Tp4bQu/7VOecabUxZg4Zi+uRUN+OOB9EM4ecbG4xzTpPOoc8UZuo63yijYnmFtFcE0Y7h5yojSnnFrNsYzpzyJmc18xzgx7jOaQ+zTnkNNpwOBwRx08srv01TSMlLSVijJvtHDL0c8hx1teEY/ra2LlFNHWG8XPIaOoc3u9mPLeY5vpCNNeE4WNq+PETdZ0VzCFDP0+2vjDda8KINk6xvjCreeoU6wuznadOd24x3XU2cw7piEGdY3TtH6u+NtZ9bBtjY8lF4ilSC7iB1WG/Ww20TPRkTdPeqmna7ZpDo76pHoC8gjw6ujsAyMjMoLe/F4CExAT6l/ebb/CSM5eQmha8S7VrcRc5uTkANLU1UVJaAkB1XTVVNVUAlJaX0tjSCEBuXi6dizoBSEtPo++MPiB4IPUv7zcL1Lesj7SMNAA6ejrIzc8FoLGlkdLyUgCqaqqorq0GoLi0mOa2ZgByPDl09XYBkJKawtKzlgLBA6ZvWR8bXt2Az+ejd2kvGVkZALR3tZNfmA9AfVM95QuCX+y2YOECautrASgqKaK1oxWA7Jxsepb0AMEDKTcvF7/fj67r9C/vJzEpEYBFSxaRlZMFQGtHK4XFhQDUNdRRUVUBQHllOXWNdQAUFBXQ3NbMwOoBUtNSWbx0cUQN9h88k7Wv3oLfn0xu1gvcn/o2Pv/mSziAR/MK+M/3X0LmCfeauhoqqytNdwCfz0defh4dPcE6p2ek07csWIOEhGAboc4QXufOxZ14coP7GDe1NlFQXMDA6gEqqipYWLsQgJKyEppamwDw5HroXBysc2paKkvOXBKssytY54SEBLPO6RnpALR0tLB39158Ph8NzQ2UVgTrXFldSU1djVnnlrYWswbdfd0AJKck07+8P1hnh8M8XgdWD9DV20VmVqZZ54KigmANGusorwy+VxVVFdQ1BGtQWFxo1jkrJ4tFSxYBkJiUSN+yPta+tBa/30//8n6SkpMA6FnSQ3ZOtlnn0PtdW1/LgoULgnVeUG72tfzCfNq72gFISUshMzMTn89HYmIi/cv7zQFz6VlLSUlNAaCrt4scT7CvNbc1U1xaDEB1bWRfq22oZWD1AFk5WWad0zLSzDq73e5gnU8MjH1n9JGWHuxrnYs6yc0b39fKK8sZHR3F5/NRUlpCU1uwzjm5OXQt7hpXZ6fTGaxzYrDOvf29ZGQG+1pHdwfZnmwGVg9QU1dD2YIyYHxfa2k/UWdPNj19PRF1Dk1Gwvta1+Iutm/djs/no62zzaxzbWOt2dcqKisi+lpbZxsAmdmZLF6y2Kxz//J+AkaAgdUD9C3rIzklOVjnvh6yPcE6t7S3mHWuqT/Z18oqyqhvnnhM7enrYWD1wMlj1DXxmGr2tTFj6sKaYF8rLSulsbURn8/H7l27zdcRMaae6GuhRabwvhY+pja0NESMqQsWLmBg9QD5hfk0tzebNejuDfa1lNSUcX0tVIPepb0RfS2/MB+fz4dhGObrWLBwAbUNwTqH97XwMTUpOcmss6ZpEX1t0ZJFpGWkMbB6gMbWRnNMrW2oZUHVib5WWU5dU7DO4X0tMyvTHFNDdQ5NIvqW9fHG62/g8/no7u0269zc1kxxSbCv1dTVUFV9sq81tDQE65yfR2dPcLwLjak+n4/X1r1G37K+icfURZ148oJ1bmxtpLQsWIOFtQtPjqlhfc2T66G9q52B1QMkJiWe7Gsu56Rjakd3B3kFecE6NzdQVhHsa6Ex1efzceDAAfN15HhyphxTExODdV68dDGZ2cE6h/e1usY6SspKGFg9QEl5ScSYGjpGs7KzWLT05Jjav7zfvODvX95PcvKJvrYk2Nd8Ph8ul8t8HeFjatmCsgnH1KnmL0vPWkpCYgIDqwdo72o35y/Nbc2nnr/k554cU8P6mq7r5OblmpPT8DF1yvlLXbVZZ3P+cmJM9fl8vLHxDfN1TDWmjp2/TDamNrY0MrB6gPTMdLOvJSdPPqYuWrqIrOwss85j5y8+n4+jR4+a87BxY2qoryVG9rX+5f3mmNrd122e11raW8grzGNg9QCV1ZURY2pD84m+FjamTjR/GTum+nw+UlJTyMoKvo7wMTW8r4XGVABPnsecp040f9EdOgOrB+hZ0mP2tc6eTvLyT/S18DG1uurk/KXk5Dw1fExNSEggNy8Xn883bkxdvHTxuDEVoK7p5PxlQdX4MdXn87F923bzdUw1po6dv0SMqSf6WklpCYZh4PP5xo2pvUuDx+jYMbV/eb85f4kYU9uDY6rP52N0dNR8HeFjanhfC6+zO8EdMU8N72ttnW3s3rUbn88XOabWLIzoa+Fj6oTzl7Ax1efzkZWVZb6O8DG1vrk+ckytD9b5VPMXn8/H2oG19C3rO1nnJZOPqRWVFUBwnlrbGKxzeF9LS08jIzMjOIecaEw9xfxlojHV5/Oxd89e83VEe024YOEChoeH8fl8Mbsm9Pl8aJpGUXHwdczmmjD8vNbd283WLVvx+XxRXROG97VAIEBuXq55vM7mmjC8r3X1djGwesA8H0w2pob6c0tbi3mtEH5NWFpRao6p2TnZJCYn4vP5Zn1NWFIWOab6fD4OHTxkvo7ZXBN29HSYY2ptfS2HDx/G5/PN+powdO4Mjak+n4+ExARzHjaba8Lwvta3rI+NGzbi8/miuibMyMowx1Sn00luXi7GiU1gZ3NNGD5/ae1oZWD1AEkpSVFdE4bPXwqKCtAdOj6fb9bXhOaY2lRPcWkxA6sHKK0ojeqacPHSxeb8pamtiX2D+/D5fLO+JgwtzoaO14HVA3T0dER1TdjXf7Kv9S3rY93adfh8vllfE0Lk/CUpOYnsnJPz6NlcE5pjakkx9c31DKweICMzY9bXhKE6R4ypUV4TTjSmRntN2L+8n5SUk/OXsbHknsShaJq2AmgL7UmsaVo/wT2KU8Oe0w38zTCMSb+Er7S41KiurMbn86HrerCje33mJ6neUS8QHERGR0fNn71eL4Zh4HK58Pv9BAIBnE4nASNAwB8w32S/34/u0NE13WzD4XDg9XrNTwnCudNpwzAM/H5/sA0N/L7pt5GQmEB2Tja7d+7G4XTg8wYXMJwuJwF/YOo2xr4/TiderxeHw0FhSSE7tu4gEAiMcw8tkoS34XA6wMBsQ9M0093pcpLjyWHPrj3oDh3vaPB17N13PuvWfw5wUJbzJ34auIQzDuwF4AdlC/hZYTFOt2vCNkIX2Tu37wSIus5owcnd4J5BAoHArOvscrvMGrgT3HhyPezasQtd109Zg1O14ff7ySvIY/++/YyOjEZdZ00L3kkcOn5Cv59tnUNtOBwOCosL2bHt5PEzo742pg2Hw0FObg57d+9F07So+lqojYjjRyOqOjtdTjCCE5Z9e/fh9/tnVIOxbYTcExISyMnNCR4/Dn3SOk9Ug/B+EGojdDEcfvxEU+eJjp9ox9RAIEBBUQH79u5jZGRk2jWYakzVHTqeXA979+wFg6jqHPAHgOCJf8+uPeZ4GU2dXS4XgUCAvII89u0NTvpmUoPJxruIc4PDcfL9cTkJBCbua1O14ff5zeNnZHgkJudOl8tFtiebPbv24HA4pl2DycY7I2BQUFTA4J5BRkdHp1+DKdrQNA1Pnid4bjACUdU5tAhQWFLI7h3BxZ9o6xxxbhjcj9frjXr+ous6RaVFk84BZjNH8vsnPn5mU+dQDZxOJzm5J+YWuj7tGkw23hmGQUFxAYO7B833YaoxdTpthM4Ng3sHCfgDUdU59PqKSorYtWMXPp8vJvPUic4N06qBQ0fXx89fJppDRlPn0PnvVOeGmcxfYj2HDBgBMMafG8LrHF6D6bQRMQfQ9Zhcj+i6TlFJETu378Tv90c9h5x0bjHNOp9qDhl6r2db51AbobnF4N5BRkdGo74mjHYOOVEbU84tZnE9kpCYQLYnm907dk85h5zJ/CV0bti5bafZj2Ixh5xoDjBhDabRhtPppKCowDx+YnHtbxiRc4to5pChMdGT72HX9l3Bu/tmcU04dv4y0blhtnU2j5+wOWQ0dQ7VYFZziynamM4ccqbzF6fLSX5BvnluiKbOKuaQhmEEj5+8idcXpntNONkc4HFhtQAAIABJREFUcnR0/BxgNvPU8ONn7PrCbNfZpju3mO46m9PhnNH6wmznL7Otc7Q1mKofPPX0U7bYk3iyHAWSNU0LXxDOAI5M9UdGIHhnAEAgEMDnDf5sGIb5pgHmmxb6ObSAHrrAguBiQmhRILTwAxDwByLa8Hq9ZhtjudNpI8T1+/34fTNrw+/3U1JWYi6+htrweX2nbmPs+3OiDU3XKCwqND9lGus+URt+nz+ijXD3QCBASVkJmq6ZNdi6/SLWrf8i4KAx5z/4w/C7OePAXkY1jRULa/lZUQlo2qRtGBgUlRQFJz4xqDME7xw2MKKqc3gNAoEAxaXF5ieSp6rBqdrQdI2SshL8fn9M6hwa0ELHT7R1DrWh6RqFxZHHz3RqYPa1MW0EjODxg0bUfS3URsTxE2WdfV4faMHjJ2AEZlyDydz9Af/J42eKOp9qvAu1oTv0ccdPNHWe6PiZUZ0nGFN1h05xaTH+gH9GNZhqTDUMw7wrJNo6BwIBdEfwIjt0J0S0dfZ6vWbfDhiBGddgsvEu4twQ/v54J+9rU7URfvxM9jpmeu70B/zmuWEmNQi1MbavhY6fgBGYWQ2maMPAOHluiLLOhmEEj5/iItCISZ0jzg0Bf0zmL7pDn3IOMJs2Jjt+ZlPnEDd0btB0bUY1GNtGqAa6Q6e45OTxc6oxdTpthM4NofEymjqHjp/C4kLz+InFPHWic8O0auCfeP4y0Rwymjr7vNM7N8zkvBbrOWTAP/G5YbIaTKeNiDlAjK5HQsdPaLuAaOeQk84tplnnU80ho61zqA3z3BAIOzdEcU0Y7RxyojamnFvMog2/309Jackp55AzOa+Fzg2h4ydWc8iJ5gAT1mAabaARcfzE4tp/7NwimjkkQMAIUFwS7NuzvSYc29cmOjdEcz0ydg4ZTZ1DNZjV3GKKNqYzh5zp/AWIODdEU2cVc0g4cfxMsr4w3WvC8DbC55CxqHPoeZOtL8x2njrducV019lmur4Qq3W2WJ3XTlWDqdoYG7vdSZwMHAD6DMNYdeJ3XwTeYRjGssk4JYUlRmVF5RwYS2YTw4A3t17Oth2XAHCm51Z+fuQGikZGOOR0cl1NAwPpGfNsKZFIJBKJRCKRSCQSiUQikcRHnn72aevfSaxpmlPTtETACeiapiVqmuY2DGMIeAC4SdO0DE3TqoFPAT+cT1+rRdM0MrMzzc2orcgM8cDB629cZS4QX5z3WR45cB1FIyNsS0zkE41t014gVuVoh/fRyo4qmOIojlZi2sFRBVMcxTHemeIojlZi2sFRBVMcxTHemeIojlZi2sFRBVOFo11jyUVi4MvAceB64PwTPz9x4rErgUPANuCvwL8bhvHT+ZC0apwuJy1tLeaXm1iNObh/GateWskf//Qgzzz/KLv2nA/4uSr3ffx0712k+X0MpKbzicY2tiUlzYujCp4Kph0cVTDFURytxLSDowqmOIpjvDPFURytxLSDowqmOIpjvDPFURytxLSDowqmCke7xtLbTcQqst2EdTK4fxnr1t885rcGt6W+h88ffQiAJ3M8fLOqllHdqp9hSCQSiUQikUgkEolEIpFIJPaNLbabkESfxMRESzI3b7s04t8uRvkJHzIXiH9WVMLXFtbNeoE41q/bqu+jSp5dmOJoTZ4KpjhalymO1uSpYNrBUQVTHK3JU8EUR+syxdGaPBVMOziqYIqjNXkqmOJoXaYKRztGFonjMG63myVnLsHtdluOOTxcxIU8xBpaOEwa+8nmg/wHPhzcsmAhPyhbgDHLfWBi/bqt/D6q4tmFKY7iaCWmHRxVMMVRHOOdKY7iaCWmHRxVMMVRHOOdKY7iaCWmHRxVMFU42jWy3UScxuF04Pf5LcU8PlxE2ep38WsuGvfYN51X8Ieul6LRA2L/uq34Pqrm2YUpjtbkqWCKo3WZ4mhNngqmHRxVMMXRmjwVTHG0LlMcrclTwbSDowqmOFqTp4IpjtZlqnC0Q2S7idMkKg7uaJjHjlWyZu29fJWx+xEH8y79l7NmhyfWr9tq7+Nc8OzCFEdr8lQwxdG6THG0Jk8F0w6OKpjiaE2eCqY4WpcpjtbkqWDawVEFUxytyVPBFEfrMk/HBeKJIovEcRi3283yty6P+e38s2UePtLES6/cjdebQzXrJ3xOqW9/tIoxf91Wex/ngmcXpjiKo5WYdnBUwRRHcYx3pjiKo5WYdnBUwRRHcYx3pjiKo5WYdnBUwVThaNfIdhNxGE3TSElN4djRY8SqvrNlHjjYzSvrbyYQSOTL7i9w0+gdEz5vY3Iyl7Z0zovjXPFUMO3gqIIpjuJoJaYdHFUwxVEc450pjuJoJaYdHFUwxVEc450pjuoc8/LzuOBdF5Cckow2i+890tDQHToBfwCDGDjGmKeCaQdHFUwVjvMdwzBY/9p6Hnv0sSmfN3a7CVkklijL3n1n8trrN+Aw4H73e/nw6MOTPvf6mnqezvbMoZ1EIpFIJBKJRCKRSCSSeEtefh7/+vF/pbysfFYLxBJJPOTgwYP8ZOVPWDuwdtLnyJ7Ep0FcLhe9y3pxuVzzxty5++28umEFqcZxnnD18uHRhwkAd5ZXcn1NPZuSUxhxudiUnBKzBeJYv24rvI9zzbMLUxzF0UpMOziqYIqjOMY7UxzF0UpMOziqYIqjOMY7UxzVOF7wrguiXiDWNI3UtNSYLTLHmqeCaQdHFUwVjlZIRkYGZ5x1xoz+xrFixQo1NhbK7bfdviIrM2u+NeY0w0PDHDsWu/8dZCbMrTv+H5s2f5YStvG/jh56fOs4rut8tbqeP+blsyUpmUcKi/lDaxu/SM9ic2LSnDvOF08F0w6OKpjiKI5WYtrBUQVTHMUx3pniKI5WYtrBUQVTHMUx3pniGHvH/jP7ycvLi4qnaRqBQAAjELstNmLJU8G0g6MKpgpHK0TTNA7sP8Czzzw76XO2bNvy7IoVK/5o/o1sNyGJVQwD3tx6Odt2XEIHq3hMP4f8wEH2uVxcU9vI+tS0+VaUSCQSiUQikUgkEolEEsf53FWfo76+fr41JJJ5z7p167jj1om/Gwxku4nTIk6nk/audpxO55wxDUPn9Te+yLYdl3A+j/C01kd+4CAbk5O5vKlt3ALxfDjON08F0w6OKpjiKI5WYtrBUQVTHMUx3pniKI5WYtrBUQVTHMUx3pniaF1HTdNm/aV3c8FTwbSDowqmCke7RhaJ4zABI8Dg4CABIzAnzEDAyasbvsKuPe/gU3yX33IBycYoz2VkckVDK3sSEufd0Qo8FUw7OKpgiqM4WolpB0cVTHEUx3hniqM4WolpB0cVTHEUx3hniqN1HQF8Pt+88P785z/z61//OqbMD37og1x99dUz4j333HPUN9SzefPmabUxHWasMl+1iffIdhOSqOL3J7Ju/U0cOtTFd/kMn+JeAB7JK+COiir8unwOIZFIJBKJRCKRSCQSiWRuYvftJq697lq2bNnCz3/285gxX3/9ddxuN2VlZdP+m6NHj/L6xtdpqG/A7XbHzEUyd5HtJiQ4nA7qGutwOB1KmV5fKgPrbmf0UCMPc765QHxf2QK+s2DhlAvEc+VoJZ4Kph0cVTDFURytxLSDowqmOIpjvDPFURytxLSDowqmOIpjvDPF0bqOmqaRmJQY0y0NYskDGBkZmTZz4cKFp1wgHuuYmppKW2tbVAvEKl73XNcmEAgwOjoak7asHlkkjscYwYOYWN4kPoY5OprDS6/cTerRbJ5mCW/nD4xoOjdU1/GLohI4VWedA0fL8VQw7eCogimO1uSpYIqjdZniaE2eCqYdHFUwxdGaPBVMcbQuUxytyVPBtIOjCqY4WpM3j7n2umv57W9/y4svvkh9Qz31DfX85je/AeDsc87mjjvu4I477qC/v59/fvs/A/DUU0/xwQ99kMWLF7O4dzGXXX7ZuC0ixm43ce1113LJv1zCk08+ydv+6W10dnVy2WWXsWfPHvM5Y7eb2L59O/UN9Tzxpye4+uqr6ezs5Jxzz+GXv/xlRFvHjh3jS9d+ic7OTvqX9bNy5UquuvoqPvihD0752n/9619z0cUX0d3TTX9/P5/7/OcYHByMeI7X6+Wuu++iv7+f5pZmzn3LuXzv374X8fjdd9/NOeeeQ0try7jH6xvqefBXD0Ywr7r6Ki6++GLz3/fccw9nnnUmzzzzDBdceAGtba2sXr2a3bt3c82XrmH58uW0d7TzzgveySOPPDLudbz88st89F8/SmdXJz2Lerj00kvZtGkTmzZtor6hnhdffDHi+Zs3b6a+oZ7nnntuyvdnLiLbTUhmnOHhQgbW3UbNyD5+xz9Ryg4OOF1cW9vAK2np860nkUgkEolEIpFIJBKJ5DRNrLabeOJPqdx7r4dt21yUlHi54opB3nLu0RgYTp4tW7bw9W98nd27d7NixQoAykrLyM7O5uxzzmZ0ZJSm5iYuvii4qHnGGWfws5/9DIfTQUV5BSOjIzzwwAO8/PLL/PHxP5KcnAwEF4kL8gu45ZZbgOAi8f89/X/kF+Rz2WWX4fP5+MY3vkFHRwd33XkXEFwk/tCHP8Tjf3ic8vJytm/fzjnnnkNRURHnn38+3d3dPPbYYzz00EM8/NuHqampAeDLN3yZxx9/nC9+4Yvk5+dz/w/vZ8eOHZSVlfHTn/x00td+7733UlhYSElJCYcOH2LlypUMDQ3xm4d+Y97le82XruHxxx/nEx//BK2trezatYt169Zx3XXXTevx+oZ6vva1r/He97zXbPfqq69m1+5dpts999zDyh+vxOPx8PHLP05+fj6VlZUcOHCAxx9/nLa2NpKTk1mzZg333HMP37ntO7zl3LcAsH7Dei666CIaGxr5l3/5F5KTk3nhhRdYunQpPT09XPz/Lqa2ppYbb7zRbP+uu+/i4Ycf5s9/+nPMvzxvpttNxO6rHyWWicPhoLS8lK2bt+L3+2PKfHWdgzVrb2W593l+yXtJ4xibE5O4uq6JnYnjv6BuPhxjxRRH6zLFURytxLSDowqmOIpjvDPFURytxLSDowqmOIpjvDPFcW4cDQOOH5/Z4ttT/53GVVcVmf9ev97BZz5Twq23bmf5WTNbKE5KMtB1DbfbzejoKFPdrFlWFlwQPnr0KG2tbeMedye4ufuuu3E6nWhakPmBD3zAZPr9fhYvWkzfkj6efvpp3vrWt07a1qFDh/jlL39JYWEhALt37+b222/HMIwpFyv7l/bz2c98FoCe7h6efPJJ/vzkn6mpqWH//v088sgjXHftdVx88cVomkZPTw9Lli455XYXV1xxhfmz3++nsaGRs885m7Vr19LU1MTrr7/OI488ws033cz73/9+87288MILAczHb7rpJt7z7veYrNDjk0YDXdfRNM18H4eGhvjqV75KX1+f+bSCggLzwwfDMOjs7GTL1i089NBD5iLx9+77Hrm5uaxcuRKXywXAsmXLTMa7LnwXt91+G9dffz1utxvDMHj00Ud55zvfGfMF4tlEtpuIw2iaRkZmRswOsMH9y3j+Hz/kv375Q57/xw+51PsrHuV80jjGi+kZfKKpdUYLxCocVTDF0bpMcRRHKzHt4KiCKY7iGO9McRRHKzHt4KiCKY7iGO9McZwbx+PHNTq7amf0X/gCcXiuuqp4xqzQArXDEf2eyX19fTidJ+/3dDgcbNu2jc9+9rMs7V9KU3MT7R3tHD9+nC1btkzJKq8oNxeIAaoqq/D5fOzbt2/Kv+vt7TV/drlclJWVsWd3cJuK9evX4/V6OeOMM8zn5Obm0traesrX9uqrr/Kxyz7G4t7FNDU3cfY5ZwOweUtwu4vnX3geXdd5xzveMeF7aT5+/jtO2dbYaEQeOy6XK+J1QnDh+vvf/z5vfetbaW1tpbmlmV//+tcR7/PzLzzPP73tn8wF4rF529vexujoKE899RQAL6x6gW3btvHOd7xzxs4qIncSx2F8Ph9rXlwTE9bg/mXUre/gF1zIAt5gmERyCe4J81huPt9ZsBDfFF9QNxeOqpjiaF2mOFqTp4IpjtZliqM1eSqYdnBUwRRHa/JUMMXRukxxtCZPBdMOjiqY4mhNnooYhsHQ0FDUnJzsnAjmkSNH+MQnPwHANVdfQ3FxMW63m3/92L8yMjIyJSs9PXK70NDi86n+Li0tLeLfLpeLkdHg3+zbH1xgzszMNB2HhobIysxi/4H9kzKPHDnCxy77GOVl5dy44kby8vPAgPdf8n5GR4JfGnfw4EHS09NxuVwTvpehx2f8RXsG+AP+iDu8s7Kyxn3osPLHK7n3vnu58ooraW5uJi0tjZ//588j9hI+ePAgubm5kzaVlpbGueecy8OPPMx5553Hww8/THt7O+Xl5TNzVhRZJI7D6LpObn4ue3fvDW7eHkVa3yziv3i3+e80gv9bxS8c53Nf5YFTf0HdHDiqYoqjdZniKI5WYtrBUQVTHMUx3pniKI5WYtrBUQVTHMUx3pniODeOSUkGq154bUbMiy6u4PXXE8b9fuHCER74xZszYiUlBbdvcDqd+Hy+KbebOGXClmA0TWPHjh1s2LCBn/70p3R3dQPBL287cuTIzNGh9Z0obvAOLWIfOHCApKQk83UfOHhgyr9bs2YNg4ODPPCLBygqCt7FvXXr1ojnZGZmcvjwYbxeL8nJyePey9Djo6Ojky4Uu1wuvF5vxO8OHz6MpmkR201MdFf6n//8Z85/+/lcdtll5u/GHm+ZmZns3bt3ytd64YUXctnll7Fjxw6eeOIJvviFL075/LmMbDcRh3E4HFRUVsTkf2W4dvR7E/6+wb9p1gvEEFtHVUxxtC5THMXRSkw7OKpgiqM4xjtTHMXRSkw7OKpgiqM4xjtTHOfGUdMgOdmY0X+f/tTghKxPf3rvjFmhpRN3wvTucHW5XIyOjk7ruf6A3/ybUJ740xMxW9CfaWpqanC5XPzlL38xf3ds6Bhr1kx9p/fw8DAQ+Toef/zxiOf0dPcQCAR49HePTvhehh7/3e9+N2k7BQUFvPnGmyfdjh1jzUtr0Kfxf8iPDI9E+A0NDUW8ToCenh7+8Ic/jFuIDs/ixYvJy8vj6quvZnR0lPPOO++Ubc9V5E7iOIzX6+Xvf/171JxtOy6miqsmfGwBbwLZs2bHylElUxytyxRHa/JUMMXRukxxtCZPBdMOjiqY4mhNngqmOFqXKY7W5Klg2sFRBVMcrckDOPfcI9x11zbuucfDtm0uSkq8XHnlIOeeM7MvrQvFMAyOHT02recuqFjAo48+ypNPPkl+fj7FJcVkZWZNyCwsKCQ3N5dbbrmFT37ik2zfvp37f3g/qamps3IM/jDjPzWTnZ3NO97xDr5z23fQNI28/Dzuv/9+MjIyplyIbW1tJSkpiRUrVnDJJZfwyrpXeOg3D0U8p6qqigsvuJCbbrqJwb2DtLS0sGfPHl5e+zLXXXvdycdvvonBwfGPA5y9/Gx++eAvqamtwePx8OMf/5jExET8fv8p7/BetHgRDz74IM3NzWRnZ/PvP/p3kpOTI57zyU98kvdd9D4+8tGP8IF/+QDJycm8sOoF+nr76OnpAYJ3KV/wzgu473v3cd55543b9mM+I3cSx2E0TSM9M33Wm8AbBry55WNUbqkhkeEJn7PDnTzh76ebaB3ngimO1mWKozhaiWkHRxVMcRTHeGeKozhaiWkHRxVMcRTHeGeKo3UdAc5763Ee/u2brHphAw//9s1ZLxCHMt07p9/97nfT39/Ptdddy3vf917+57//Z9LnJicnc+d37+T40HGu/NSVPPirB7n11ltJSUmJyjWaXPulazn77LP59i3f5is3fIVzzzmXxsZGUlMmX7jOzc3l1ltuZeOmjXzyik/yv//7v9x5553jnve1r32ND3/4w/zigV9w2eWXcc+990Ts02w+/ouJH7/iiitYftZybr31Vm688Ubedt7b6O7qntaxc+UVV3LWWWfxrW9/i6+u+Cr9S/s5//zzI56zcOFCfvqTn+J0OLnmS9fwhS9+gTVr1ozbp/iss84C4IJ3XnDKducyWlR7odgkJYUlRmVF5XxrzFncbjd9y/p45i/PTPt/UQjFMHRef+NzvG3PPr7P5bjwTfi862vqeTrbMy+Oc8UUR+syxVEcrcS0g6MKpjiKY7wzxVEcrcS0g6MKpjiKY7wzxVGN4+eu+hz19fVRMTVNIzUtlaNHjka3h7AingqmKkeH00Ffbx/vfs+7+cynPxMTptVf96ly//3389P/+Cn/89//E9OtXMZm3bp13HHrHZM+/vSzT99hGMbnQ/+WRWKJmUDAyfoN1/HZA09xHd8E4E85uTydlcOHdmyhcHiEnYkJ/KikPKoFYolEIpFIJBKJRCKRSCQSFYnFIrFk9vnrX//K+vXrqaur4+ixo/z85z9n9erVPPrIo5SWls633rxm+/btbNy4kWuvu5aLLrqIT3/q00rbm+kisexJHKdxu90z+kTO709k02tf5q7D93ExDwCwsriMlSVloGn8jyd3xsxYO84HUxytyxRHa/JUMMXRukxxtCZPBdMOjiqY4mhNngqmOFqXKY7W5Klg2sFRBVMcrcmD4B2msbyxMtY8FcxY8JKSkvj9Y7/nnnvvwe/3U1dXxw9+8IOYLhBb8XVPJ/fcew+///3v6e3t5aMf+ajy9mYa2ZM4DuN2u+lf3o/bPb1vzvT6Utnxypf51eFruZgH8KLzjaoaVpaWE/oazpkyY+04H0xxtC5THMXRSkw7OKpgiqM4xjtTHMXRSkw7OKpgiqM4xjtTHK3rqGkaaelpMd03OZY8FcxY8To6OvjVg79i1QurWLN6DY899hiLehbFxBGs+7qnk29+45u8tOYlvv9v35/XfaMni2w3EadxuV14R72nfN7oaA7Dr3ySXw9/nCo2cVh3c0NtLf/IyJw1M9aO88kUR+syxdGaPBVMcbQuUxytyVPBtIOjCqY4WpOngimO1mWKozV5Kph2cFTBFMfY82K13YQd7lY9HR1VMOfqTuK5jmw3IQGY1mB7fLiQzLXv5THvRWRzgG3uVL5UX8uWpORZM2PtON9McbQuUxytyfv/7Z13nFNV2se/T5IpdAaGOnSBoXcYYEAQxPK6rGV1xYaubbHruqBrW3R3XQW7YkF8xdUVRdeCrwpWBGHtiquiuOtasGEBlT7lvH+cm0ySSTLJvYmTzDzfzyefJDf3/s4v57mn5NyTczOhqR6zV1M9ZqdeJjRzwWMmNNVjduplQlM9Zq+mesxOvUxo5oLHTGiqx+zUA9I+aJiJQcjG6DETmg1xgNgNutxEAyQ/P5/J0yYn/KvFtu29GPSvSTxecSRt2MzbTdty6uCBcQeIk9FMt8f61lSP2aupHtVjNmnmgsdMaKpH9djQNdWjeswmzVzwmAlN9ageG7qmesxej7mwpEFj9JgJzZ9zuYlsR5ebaIAET/Cffvwp5tWQH38cwCHri7jEXAHAM607c0XfHuz2+V1rpttjNmiqx+zVVI/qMZs0c8FjJjTVo3ps6JrqUT1mk2YueMyEpnpUjw1dUz1mxmO6lpvw+/1UVVV51smUXiY0c8FjJjQz4TEbSHW5CR0kbmRs/X44v9vwDUezBIC/dezFHd07Y/SKiaIoiqIoiqIoiqIoOU66BokVJddJdZBYl5togOTl5TF2wljy8vIitu/eVMa1G97kaJZQiZ8revRnUY+SpAaI42mm22M2aarH7NVUj+oxmzRzwWMmNNWjemzomupRPWaTZi54zISmelSPDV1TPWavRxGhefPmaV3SIJ16mdDMBY+Z0MyEx1zFP3fu3Pr2kHGuufqauUWti+rbxs/K7l272bp1a+ivG3kbx7Lok2WM5E1+lKb8obQ/q9qllifRmun2mI2a6jF7NdWjeswmzVzwmAlN9ageG7qmelSP2aSZCx4zoake1WND11SP6fc4rnwc7dq186QnIhhjqK6uTou/dOtlQjMXPGZCMxMes4Vvv/2Wl9a+FPfzTzd++tLcuXNXBN/rchONgPYflXHLpsW041s2+tpxwcDufNysSX3bUhRFURRFURRFURRFSSu63ISl/4D+XHbZZRx26GEATN17KtOnT+fss86Oe8ycOXP46uuv+Ntdf0s6nVdeeYX33nuP4447LmL7zGNn0rFDR+bNm+fKv+KdRrPchIjkichNIrJZRL4XkRtFJFDfvrKBQCDA0BFD8fsDDHp/KPdsupF2fMu6vJ6cPrSXqwHioGYgkJ4sTrdeJjTVY/Zqqkf1mE2aueAxE5rqUT02dE31qB6zSTMXPGZCUz2qx4auqR6z16OI0LRp07QuaZBOPTeaN95wIzMOn5FQzx/wp+zjlVdeYfFdi2ttv+TiSzjjjDPq/Xv/3Hq5TC4Pql4ETAAGOO+fBC4ALqs3R1nAsE/ac/JX79D9xVfZRlM68jwATzUZwlWDmrPTn3qBB6g21WzZsoVqk57p9+nWy4SmesxeTfWoHrNJMxc8ZkJTParHhq6pHtVjNmnmgsdMaKpH9djQNdVj9noEqKyqzGq9VDUHDBhQ5z7pXG2gd+/eiEi9f+/60AtijGH37t0UFBRkRD/d5OxyEyLyGXCOMeZB5/1hwFXGmO7R+zaW5SaGfdKeG778R63ty/NHccXwQqr1qoiiKIqiKIqiKIqiKA2YdC030fzppyhesIC8jRup6NKFb087ja3T9kmDw/gsfWApV15xJWvWrKGwsDC0/cknn+S8885j1apVtG7dmtsX3c4Tjz/Bp59+SstWLdlz4p7MmTOHZs2ahY6pa7mJHTt28Oc//5nly5fTrHkzTjjhBN59592I5SY+/PBDblpwE2+++Sbbtm6j1x69OOvMs5gwYQIAN910EwtuXhDxHU479TROP/30mMtN3Hnnndy75F6+/vprunXtxqxZs/jFL34R+jx4zKjRo7j1llvZum0re+65J3P/OJfmzZvHzbfnnnuOxXd2z7IQAAAgAElEQVQtZsMHG0BgyJAhXHjBhXTvHjlEePfdd7PkviVs3LiRNm3aMGXKFC65+JKkPo+1XMd111/HY489xrPPPAvAww8/zAUXXsDS+5fyl8v/wvr165n7x7nss88+zL9qPmvXrmXTpk107tyZww49jOOOOy5iBvMnn3zC1ddczcsvv0xlRSV9+vZhzuw59O7dmz0n7cmfLvsT06dPD+2/bds2Ju45kYsvupiDDz64Vr40iuUmRKQI6AK8Fbb5LaCbiLSqtb9PQn878Pl8BPICQR3y8mvufpmfnx/xOhiovLw8fD6bVYFAAJ/fvvb7/fidmbk+vy8ijeBdNUWklm4yaQR1/X5/aLp/XWmc/NU7MfOrS8VPVIsQyAvUnUZ0/jhpBAIB+g3sFzom2nvwdXga/oA/Io1w7wWFBfTt35dAIBCKQay8ipk/cdLIz8+ndEApfr8/LXHOy8ujb/++5Ofne4pzXn5N/hQUFoQ8JhODutIIBAL07d+XwsLCmDFINc4iQmGTQvr274vf7/cc52Aasc6fZGIQKmtRaRQU2PMnLy/Pc1kLphFx/niMcyCv5vwpKChIOQbxvBcW1sQmUZzrqu+Cafj9fnv+NClMS5xFJMJjOurUkEeng5RsDBLVqfkF+aHzx2ucfT4ffr+f0gGlof28xjnoK3j+pBqDePVdeNmuVafGKWuJ0gg/f+J9j1TbzuD5EwgEUopBMI3oshaMTfDKfdIxSJBGfn5+TdvgMc4igt/vp9/AfqFjvMY5um1IR/8l6DFeH8BNGvHOHzdxDuoG24ZgG55sDKLTCMYg+vypq05NJo1Q36Ig33OcY50/6einhte7KcXAH7v/kqgP4CbOgbzk2oZU2rV09yF9/thtQ7wYJJNGRB8gTb9HgudP8Bivfci4fYsk41xXH9JrnMP736UDSikoDGsbPPwm9NqHjJWG3++ndGCcvoWLNJLtQ6bSrkWfP+nqQ8bqA6TymzA8jby8vIjzJx2//WO1DW77kAAFBTW/D93+Jowua7HaBi+/R6L7kF7iHIyBq75FgjTinT+hwTdj8O3YgWzfjmzfjm/HjtD7eK9bPv5/dDnrLAo3bMC/fTuFGzbQ5ayzaPH4/9XSivc69J6a39oiEvIV/jr4ftre09i1exerV6+O2L5ixQrGjh1LUVERYG9IdsKJJ3DH/97B73//e9544w1mz5ldS1eISoOa1/Pnz2f58uX8fvbvufTSS3nyiSd55dVXIvbZ9M0mBvQfwJ8u+xMLFiygbEwZs06ZxTvvvIOIcOihh3Lorw6lTZs2LFmyhCVLlnDYYYcRlmDoe//93r8z/6r5HHDAAdx8882MHDWS2XNms3r16giPr772Kk888QSXXHIJc2bP4fnnnmfBzQsi8i28nRERvvjiC/bff3+uu+46rrjiCvx+P8cccww7duwI7XPjjTdyxZVXsNdee3HLLbcwe/Zstm/fbs+hJoU1n0/ei1tuvoU5s+ewffv2yHMp+jVR2523s+fMZvr06Sy8bSHDhg9jx44d5AXymDN7DrcvvJ0jjziSW269hbv+dlfo+G+//Zajjj6Kjz76iIsvupgbbryBSZMm8fXXX9OqVSumTp3Ko8sejfjeK1asQBD23XffuB4T1XfR5OQgMRC8fLAlbFvwdYvgBhHZV0SuEb/Qf5C9itS+Y3tGjB4BQKvWrRg3cRxgB+8mTpkYqmjKJ5fTvIVNZtTYUbRt1xaAQcMG0aVrFwD69OvDHn33AKBr964MHDIQgHbt2zGybKQ107IF4yeNB2ylPXHKxFCAxu85nhatrN0RY0bQroO9++bAIQPp2r0rAHv03YM+pX0AKOlawuBhgwFoW9yWUeNGAdCseTMm7DWB7ubzmJkV3D581HA6dOoAQP9B/ene015R6dm7J6X9SwHo3KUzQ0cMBaBN2zaMKR8DQJOmTSjpWoKI4PP5mDhlYuiHXVl5GUVtbUU1dMRQOpV0AqDfgH702KOH9dCrO/0G9gOgY+eODBk+BJ/PR6uiVoydMDYiBsGKf+KUiTRt1hSA0eNH06ZtGwCGDBtCSdcSAPr260uvPnaWeJduXUJxat+hPSPG2Di3bNWS8XvaGBQU2DSChSE8ziPHjqS4XbGN89BBlHQtwefz0atPL3qX9g6lMWjoIACK2xUzcqyNc/MWzSmfXG7jnGfjHOw4jN9zPC1btQRg2Mhhoe80YPAAuvawce7Vpxd9+/UNxXnIsCGhGIwePxqAps2aMnHKRMA2ihOnTKSg0A4gjRk/htZFrUNx7ti5o43BwH5072Xj3GOPHvQbYGPQqaRTKM5FbYsoKy8DbOVdPqkcn8+HiDBxykSaNLVrWI8pHxOKwdARQ+ncpTMApf1L6dm7p41zz+6hstahUweGjxoO2LLWqbM9LwoLC22cnc7JhL0m0Ky5vdo5atwo2hbbGA4eNjgU5z6lkWWt/6D++Hw+itsXh+LcolWLUJzz8/NtnJ1OyPhJ42nR0pa1kWUjade+dlnr1bsXRW3sedylaxcGDbNxbtuuLaPGjqoV50AgEIoBwLiJ42jV2l6jGjHalmefz0fpgFK69ewG1C5rQ4Y7cS5uw5jxYyLiHKusjR43miZNmoTOpWCcSweWhspaj149IsrasJHDAGjdpjVjy8eG4jxxykR8Ph8+n4/ySeWh83LM+DG0KXbK2vAhoTj37V9T1rr16Eb/wbHr1LLystBgzcQpE0Odwug6NVTWourU3n1tWevarSsDh9o6tVmzZgwbZb9HRJ3qlLX8gpo6NVjWwuvUAUMGRNSpvfv2xufz0blLZwYPHxyKwehxtqw1a96sVlkLxmDchHERZS1Yp7Ytbhsqzz1796R0gI1zeFmLrlODcY4ua2XlZRS1KcLn8zF4+OBQnVo6oJSeezhlrVd3+g2ycQ4va62LWofq1GCcg3Vq+aTyUF6NHjc6FOfBwwZT0qWmTt2jT01ZGzDE/iWtfYf2jBxj67vwOjUQCFA+qTx2nVo2kuL2Ns4Dhw6kazebP71Le9fUqWFlrbhdMcNHD8fn80WWtbxA3Dp1xOgRtO/Y3sZ58AC69bBlLbxObd6ieajeblvcNmGdGhxwGjthLK3b2DiHl7V+A/vRrUc3fD4f3Xt2j6hTg2WtqE0RZRNq6tRgWQvGuWlTp6yV15S14vbFNXEOq1O79ewWt06N138J1qk+n48RY0aE2sXBwwbX3X/p0K6mTo3qv5R0LYlZpybsv/TrE4pzqP8SVqcWFBREnEvx6tTo/ku8OnXQsEH4fD6K2haFylrTpvHr1LIJZaF6f9jIYTH7L61at6Jvf3su1apTg2WtsDBh/yXYrg0ZPoROJZ3w+Xz07ts7ok4dMNgpa2F1aqz+S6w6tUPHDqE0wuvU8LIWXqcWty8O9VNj9V/yC/Lx+XyMnTA2VNZGjhlJ+w5OWQuvU/vsUdN/6VLTTw2vU5s2axpqz6Pr1LETxsasU/sNqum/9Nwjdp3apGmTUF84UZ0a3X+JqFOdsta1R9dQHkbXqeMm2LIWXadOnDIx1H+JqFOH19SpRW2KatIIq1PDy1p4nPML8iP6qeFlbdioYaHZYRF1at/eEWUtvE6N2X+JqlM7lXSKWaf2H9w/sk51ykFd/ZdgnMsnldfEuTx+ndqjVw/A9lNLB9o4R5S1otah/WPWqXX0X+LVqc2aNQudS15/E+7Rd4/QeZzO34TFxcWUdCsJpZHsb0Kwg2nh7dqY8jGheHj9TRhep5Z0LfH0mzC8rI0ZPyY0mcjLb8KuPbqG6tR27duF6i63vwm7dKtdp7Zo2SKUhpvfhCPGjAj56jewX2i7l9+EEFmntuvQrqZOdfGbMLyslU8qD52vnn4TFrUK1akFhQWUdC3x9JswvP8ybNQwfD6fLWtOnEWEFi1b2LZhxw76jhxB6aiRlI4aSd+RI0Lv473uPNsOuEZTMnt2La14r4Pvm2BC/fCCwoJQ2czLywu1V4FAgGbNm1FUVERZWRnPPvds6HsE8gKsWr2Kfffdl+bNmxMIBPjD+X/g8MMPZ/z48ey/3/5c/tfLWblyJd9//z35Bfmhsunz+0JlKPwCwJYftvDQww9xzu/O4YgZR3DggQdy2223sXnz5oiLUdOmTeOMM85g0qRJTJo0ifPPP58xY8aw7LFlFDYppGPHjnTq3In8/HyGDR3GyJEj6d3HlhXBXnwQhKqqKhYuXMiRRx7J2WedzX777cflf7mciRMmcvMtN4f6mT6fj127dnHzgpvZZ999mHnsTA4++GCeffbZUDvo8/tC3yMY52NmHsMRM45g2j7TmDxpMtdecy1bt21lzZo1AOzatYtFdyzi5JNP5uKLL2bq1Kkc8D8HcM3V11DYpJAtW7aw6I5FnH766cyePZs999yTGUfM4Iq/XlFzLjkjwM1bNK+5GBV2AaiwSSEBv90+a9YsTjzxRMrKyijtW0r37t258MIL2W+//Zi812RmzJjB8ccfzyOPPBI6t+5dci+7d+/mnrvvYcYRM5i05yROmXUKhx56KHl5eRxy8CG89NJL/PDDD4DtNyx7bBl7T9ubVq1bheIcCARCdVqw/xvRf2lW03+JJieXmxA7k/h7oLcx5j/Ott7Ah0BrY8wP4ft3Lelq+vTqQ2VlpR0U8fuorKgMFbaK3RWAHVjavXt36HVFRQXGGPLy8qiqqqK6uppAIEC1qaa6qjqUyVVVVfj8PnziC6Xh9/upqKgIXZEL100mDWMMVVVVNg2Bqsq607hu9UYGmQ9q5dc7UsqpZe0J5AWorqpOnEZ0/gQCcb9HuPfKykqMMRFp+AN+MITSEJGEMagrjVD+eEijPuKcl59HZUXt/EkmBnWlEfLuIY2fO86uYpAgDa9lLRvi7DUG0Wl4LWv1Fuc0xiAddWq2xTljMcgLUF1dd1nLxjhnLAYNMc4/U/8lp+L8M/VfcjXOKcfAby88JtN/ycU4u4pBkv2XnIpzBuvUuDHI9jhn2W/CXC9r+ptQfxOm4zfhaWedRv/+/RERjDHI9u2UjrKD+fXBhtffwDRtar04A4rRr4GQ36UPLOXKK69k7Zq1FBQUsGLFCmbPns2qVasoKirCGMMrr77CDTfcwAcffMDWrVtDaS1ZsoThw+zAfL/+/fjTZX/isMMOwxjD1L2n8svpv+Sss87i5Zdf5rjfHMczTz9DSUlJKO2Zx84ECC03sWPHDm686UaWL1/O119/TXW1Xfd5woQJLLp9EcYYblpwEw8++CArn18Z8T1mHjuTjh07Mu/KeXz++efsPW1vFt+5mLKystB3f/DBB7n0skt568238Pl8zDx2Jk2bNuXWW24N7fP3e//OlVdeydvr3q6VV8HXn332GVdddRWvvf4a3333XSg/fve733HSiSexevVqTv7tyTz5xJP07NmzVgxeeOEFfjvrtyx/cnloiYroNKZMncL06dM55+xzQtuvv/56lj22jGefeRYR4aGHH+KCCy7ghZUv0KFDh4g07rvvPu6+524+++yziHN/3VvrADh8xuF069qN+fPn10ob7Lk/de+pHHP0MRx//PF88cUX7D1tbxbdvojy8vKY59L69etZcP2CuGXtudXP5f5yE8aYzcBGYFjY5mHAZ9EDxACm2lBZaRehrq6uprKiMqgTqqCAUKYFXwczt6KiIlQQKisrqa6yr6uqqqiqqrK6VdURaVRUVITSiNZNJo2gblVVFVWVyaWxsOOgmPl1m7O9sqKy7jSi88dJQ3xCx84dQ3/1iPYefB2eRlVlVUQa4d6rq6vp1qMb4pNQDGLlVcz8iZOGwdC5S2d8fl9a4gx2do/BeIpzxe6a/KmurqZLty62oUwiBnWlIT6hW49uVFVVxYxBqnE2xlBZVWlnyvl9nuMcTEN8QqfOnSLOn2RiECprUWlUG3v+IHgua8E0DMbOHvf7PMe5sqISxJ4/1aY65RjE815VXUXX7l1DHa14ca6rvgum4fP76NajG5VVlWmJc7DzGX7+pBTnGHWqz++jS7cuVFVXpRSDRHWqMSY0O8prnKurq/H57axkg0lLnCsqKkJlO1hfphKDePVdZVVl6PypVafGKWuJ0gg/f+J9j1TbzqrqqlDbkEoMgmlEl7Xg+RPcJ+kYJEjDYGraBo9xNsbY86ekc+ivaV7jHNE2VFelpf/i8/sS9gHcpOHz++javWut88dNnIO6EX2LFGIQnUYwBj6/nSUXvAlOXXVqMmkE24ZgfeklzsHzp1NJp9D5k45+arBsp9y3qIrdf4nVh/QS58qKytD5k6htSKVdS3cfsrrKtg0lXUoi2oZ4MUgmjarqqog+ZDp+j/j8Pjp17oT4JOUYhLxXJNG3SDLOcfuQJbYP6TXOwTRitg0efhN67UPGSiNh38JFGlVVyfUhU2nXgm1D8PxJVx8yVh8gld+E4Wkg0Lmkc+j8Scdvf5/fR5euXSLaBrd9SLA3XAueP25/E0aXtei+hZc4Q+0+pJc4B2Pgqm+RII14fcigpmnShA9eez2lx67evYnFrt69U9aqdmYOB2cTh3wZE3odvn3a3tPYtWsXq1avAmD58uWMHTuW1q1bY4xh48aNzJo1i1atWnH11Vdz/333c92119m83LU7QtcQlYZTr3z73bcAtGnTJpS2iFBcXBzxfefNn8c//vEPjv/N8fzvHf/LA0sfYPz48ezatatGN2ruaXh6we/97be10zPG0La4LRUVFXz//feh/Vu2aBmxT/CiRVBXRCJm8FZWVnLqaafy0X8/4rw55/H3e/7OA0sfoFWrVuzauQuAzZs3A9CuXbvI/HGet26zA+3h3z9WbGq9Jmq78zY4mB/c/sQTTzD30rlMmjSJmxfczNL7l3LSiSdFnK9bNm+hXbt2MdMzxuDz+TjwwAN5dNmjADy67FE6dOjA2LFj455LkLi+iyYnB4kd7gQuFJGOItIRuABYVM+e6pW3um/ijE6/4h0p5Sea846UckanX7Gu+ybP2j7x0aa4DT5JzymTbr1MaKrH7NVUj+oxmzRzwWMmNNWjemzomupRPWaTZi54zISmelSPDV1TPf5MHkXsTN4UHt+ecWZMqW/OPDNlLZzBzOASBXVRVFTEmDFjWLFiBTt37mTVKrvURJA1a9ZQVVXFdddex3777cfQoUNp1arW7bkSUtzWDoaGD85CzWBqkGeeeYaZM2dy9NFHU1ZWxqBBg9i1a1dKaQUCgdDgZ3R63337HXmBvNBay6loBvn000/58MMPufjii5k+fTojRoygtLSUn376KbRP69Z2KZZvvvkmpl7btm0Tfg41s3DD+fHHH2PuGxzADvLMs89QVlbG7N/Ppry8nMGDB4eWEQv3mCh9gIMPOpgNGzbw3nvvsWzZMn45/ZehNbzTQS4PEv8J+Cew3nmsAS6vV0dZwLrumzi1rD0HjB3OqWXt0zJADPbKzFuvvVVzVTbL9DKhqR6zV1M9qsds0swFj5nQVI/qsaFrqkf1mE2aueAxE5rqUT02dE31mL0ef5o2jY033MDOvn2patqUnX37svGGG9i69zRXesYYtm/bXmuWbTz2228/Vq5cydNPP01FRQVTp04NfbZz104CfnvjwKDm8uXLU/LTt6+9EfPzzz8f2vbdd9/xxhtvROy3a9cu8vNqbnz25Zdf8va6tyP2CV+epBYGtm/bTocOHSguLubpp5+O+HjFUyvoP6B/aEmXZIjOy507d4Z8BHnq6adCM9IBhg4dSkFBAY899lhMvX6l/eJ+HqRjh458/N+PQ++rq6t55eVXkvK8a+euCH/V1dU8teKpiH3GlI3hhVUvhNYcjkX37t0ZOXIkl19+OR9//DEHHXRQUuknS84OEhtjKowxpxljipzHGcaY9NUIOYzP5wvdNCtbNdVj4/GYCU31qB6zSTMXPGZCUz2qx4auqR7VYzZp5oLHTGiqR/XY0DXVY/Z6FBF27r8/nzy6jA9fe52PH3nU9QBxUC+QF6g1wzQe0/aexs6dO5l/1fzQUhNBxowew46dO7j00kt5+ZWXuebaa1izdk1KfoqKijj4oIO59tprWXLfEp5//nlOPfXUiHQAysrKuPueu3nyySd56qmnOPm3J4duBB+kZ8+ebN68mUceeYR33nmHTZvCJiuKvWlqIBBg1m9nseS+Jdxw4w2sWbOGuZfOZfXq1Zx+2ukpeY/Oy169etGuXTvmzZvHiy++yP3338+1115L8+bNQ8e0atWKk048idsW3sZVV1/FmrVrWL58ORdccAEiQtvitpx0UuzPg0ydOpVVq1exePFiXnzxRc4991y2bd+WlOeysjLWrl3LXX+7i9WrV3PmWWeyY+eOiH2OnXksBfkFzDx2Jo8//jhr167l9ttv54knn4jY75CDD+H1N15nyOAhofWV00XODhIr8fH7/fTu2zulKzE/t6Z6bDweM6GpHtVjNmnmgsdMaKpH9djQNdWjeswmzVzwmAlN9ageG7qmesxejwCFzlrC9aEXXHLim2++iVhqAqB///7M/eNcXlzzIqeccgr//ve/+evlf03Zz3nnnce+++7LVfOv4pJLLmHf/fYN3QAtyMUXXUy/fv246KKLmDd/HjNnzmTkiMibAO61114ccsghXDnvSg779WEsXbo04vPg9z7qqKM493fnsmzZMk455RReffVV5l05j4kTJ6bsPTwvCwoKuP6669mxfQenn3E6Dzz4APPnz6dZs2YRx5x22mnMnj2bp59+mlNOOYUr510Z2qewsJDTTzs97ucAv/71r5lx+AwW3r6Q884/j959enPggQcm5XfGjBnMmDGDW265hTnnzaGkpIQTTzwxYp/i4mLuvvtuupR04Y9z/8gZZ57Bc88/R6eOnSL222uvvQA48KDk0k4FSXaqey7TpVMX06tHr/q2oSiKoiiKoiiKoiiKomSQc2afQ//+/evbhqJkhMefeJw//OEPrHphVa2Z39GsX7+ea+dfG/fz1S+tvtYY87vge51J3AAREVq0bJH03xjqQ1M9Nh6PmdBUj+oxmzRzwWMmNNWjemzomupRPWaTZi54zISmelSPDV1TPWavRyDtM5PTrZcJzVzwmAnNTHhMN5s2beLll1/mpptuYv/9969zgNgNOkjcAMnLy2PU2FERi2Jnm6Z6bDweM6GpHtVjNmnmgsdMaKpH9djQNdWjeswmzVzwmAlN9ageG7qmesxejyJC02ZN0zqQnU69TGjmgsdMaGbCYyZYunQpJ5x4AkVFRZx77rkZSUOXm1AURVEURVEURVEURVEaBLrchKJYdLkJBYBAIJD1muoxO/VyRVM9ZqdeJjTVY/Zqqsfs1MuEZi54zISmesxOvUxoqsfs1VSP2amXCc1c8JgJTfWYnXpA2meWZmKmamP0mAnNbJ9F/HOhg8QNkPz8fCbtPYn8/Pys1VSPjcdjJjTVo3rMJs1c8JgJTfWoHhu6pnpUj9mkmQseM6GpHtVjQ9dUj5nxmI5/zOfCusmN0WMmNDO1pnVWkGJR0OUmGigFhQXs2rkrqzXVY3bq5YqmesxOvUxoqsfs1VSP2amXCc1c8JgJTfWYnXqZ0FSP2aupHrNTLxOaueAxE5rqMf16J59yMiNHjfQ86Ofz+aiurk6HvYzoZUIzFzxmQjMTHusbYwzr1q3j5htujrtP9HITOkisKIqiKIqiKIqiKIqiNAjad2jPibNOpHu37g1zdqiiJMGWLVu46867ePdf78bdJ3qQOP2Ltij1Tl5+HuMmjuOfq/9Jxe6KrNRUj43HYyY01aN6zCbNXPCYCU31qB4buqZ6VI/ZpJkLHjOhqR7VY0PXVI+Z8bjp600sunURBx1yEE2bNXU1UOzz+WjVuhU/bPkhLTNM062XCc1c8JgJzUx4rG+MMWz4YEPCAeJY6EziBoiIhE7wdMU33ZrqsfF4zISmelSP2aSZCx4zoake1WND11SP6jGbNHPBYyY01aN6bOia6lE9ZpNmLnjMhGYmPOYKutyEoiiKoiiKoiiKoiiKoihKIyZ6kNhXn2aUzBDICzBm/BgCeelbTSTdmuqx8XjMhKZ6VI/ZpJkLHjOhqR7VY0PXVI/qMZs0c8FjJjTVo3ps6JrqUT1mk2YueMyEZiY85ir+uXPn1reHjHPN1dfMLWpdVN82flYqKyvZ+tPWtE6VT7ememw8HjOhqR7VYzZp5oLHTGiqR/XY0DXVo3rMJs1c8JgJTfWoHhu6pnpUj9mkmQseM6GZCY+5wKcbP31p7ty5K4LvdbkJRVEURVEURVEURVEURVGURoQuN9EICAQCDB42mEAgjdP506ypHhuPx0xoqkf1mE2aueAxE5rqUT02dE31qB6zSTMXPGZCUz2qx4auqR7VYzZp5oLHTGhmwmOu0ihmEovIP4BP6tvHz0x30v+d062pHrNTL1c01WN26mVCUz1mr6Z6zE69TGjmgsdMaKrH7NTLhKZ6zF5N9ZidepnQzAWPmdBUj9mplwlN9Zi9mpnwmAt0N8b8KvTOGKOPBvgArsl2TfWYnXq5oqkes1NPPTYuTfWYnXqN1WNj/d7qMTv1GqvHxvq91WN26uWKpnrMTj312Lg0M+ExFx+63ETDZUXdu9S7pnrMTr1c0VSP2amXCU31mL2a6jE79TKhmQseM6GpHrNTLxOa6jF7NdVjduplQjMXPGZCUz1mp14mNNVj9mpmwmPO0SiWm1AURVEURVEURVEURVEURVFiozOJFUVRFEVRFEVRFEVRFEVRGjE6SKwoiqIoiqIoiqIoiqIoitKYqe9FkfWR/gcwGXgX+DewCPB71PvY0XvLeQx2oXE9sBGojNp+heNzA/ArL3rO9/4pzOfDKXrsCjwLrHe+71+9+IynlwafTznH/Qt4EGjpMS9r6Xn1GKa9ICpGrjzG0ktDPsY8rz3kYy29NHhsBtwFfAC8D/zWo8dael48Au3DjnsL+Cp4vAePMTU9+jwaeNs5bjVQ6gHviX0AABq/SURBVPV8jKXp0eMs4B3nHLoZp9726LGWZqoeSbHuBgYBrwMfAo8AzT3qrXS0gn73TcYj0A1YBWwDFkXtX+J8tsHR7+RRbzH2TshBj79J0uNRwDrnPHoNmOLRYyI9tx7HhR3zLnCaR4+J9Fx5DPusNfB5eHzceKxDz20+TiZOuXOZj4n0XOcjUAa84sTmXaBzMh5daib0Ged7nxi2/1vAbuBALx7r0HTjsQD4G7ZPtR64CfB59JhI043HlsD9jt7bwC+TPR9datblMeX+N3W3NW40VxKnvYmnh7e2xo1m3LxMoOelrXGj6cajl7bGjWbKHutoG1x5rEPTTT5Oxn1b40YzrsckymHK7Y1LPVcecdneuNRz69FVe+NSz61HV+2NS72EHp19UhpHoY62pqE+6t2APtIcUDs7/N/AAOf9UuBYj5ofA108akwAOhLZodzbqRj8TiXxabIFL47eZOAZDx47AaOc1/nYgZ8D3fpMoOfVZ6uw19cBcz3mZSw9Tx4drYnYAcngoK5rj3H0vOZjrfPaYz7G0vPq8VZgjvNasAOoXjzG0vMc6zD9Z4AjvcY6jqYrn0BT4Dug2Hk/C3jAYz7G03TrcSC23m7lvL8emOnRYzzNlDySYt0NvAjs47yeB/zRo95KYIILj0XAeCc20T+y7wFOdl6fCtzpUW8xcLQLj+OBts7rwcDX1HTG3XhMpOfWY1Mg4Lxuie189/DgMZGeK49hny0E7ibyR3bKHuvQc5uPk4lT7lzmYyI9tx5bYC8g9nfetwKaJOPRpWZCn4li43zeFdgMFHrxWIdmyh6Bk4B7ndd+bJ14gMd8TKTpxuOVwNXO69bYH9rNPXpMpFmXx5T739Td1rjRXEmc9iaBnpe2xo1m3LxMoOelrXGj6cajl7bGjWbKHsM+d9PWuNF0k4+Tcd/WuNGM67EOTVftjUs9Vx6j9km6vXGp5zYfXbU3LvXcenTV3rjUS+gxeG6Eva5zHIU62pqG+tDlJhoeo4EvjDHvOe/vAH5Vj34AMMa8aIz5Kmrzr4DFxpgqY8znwBpgHw96Xj1+aYx5zXm9G3gTezXflc8Eel59/gAgIj6gEDBuPSbQ84SIFGCvyP0+bLNrj3H0MoFrj+lGRFoAvwSuATCWTW49JtBLl9/OwCjsVda05GOUplt82AHx5s77VsCXHj3G03TLAODlYFnE3ln3cI8e42mmRCp1t4h0ALoZY55y9qvV/qS7LYinaYzZbIxZC+yMccgvsDMlwF54OtCjnluPa40x3zlv38HO4GjmwWMiPbcetxtjKp23BdgOdBA3HhPpufIIICJ7OXrPRn2Ussc69Fx7TIArj16Io3kk8IQxZr2zzw/GmB3JeHSp6cZjtPZDxphgmXTrMZGmW4/NRCSAPYcKsP+I8eoxnqYbj4OA5c7nW7D/LNrfo8dEmnV5TKn/nWRb87P06T22NW403Xj00ta40XTj0Utb40YzZY/gqa1xo+nKYwJce3RLAk1X7Y1LPbcew0m6vXGp58Vjyu2NSz23Hl21Ny71kvGZ9DhKMm1NQ0UHiRseXYDPwt5/ir1a5ZXHROQtEfmLiOSlQQ8y43WkiLwpIqtEZF+3IiLSBjgIeJo0+IzS8+xTRB4GNmH/4n61V48x9Lx6vAS4wxjzTdg2Lx5j6Xn1CLXPa6+xjlVO3HrshZ2ZcZOIvCEiD4tIdw8e4+l58RjOEcAjxpjtHjwm0nTl0xizFTgdeEdEPgeOBS724jGBpiuP2KvgE0Sks4j4gcMcL17yMZ6mW4/hxPPl1m9dx90mIutE5CbnYodrRKQtsC3YMTfGbAMqRKSVF11groi8LSJ3i0hHF8fPAP5ljPkpTR5Del49ishgEfkXNi7zjDEfe/EYS8+LRxFpAvyVqIuIbj3G0/Pi0aFWufMY60Tl2I3HfkCBiDzj6F4qFi8eY2p69BnkKOzMu3SW65CmB49/w/6N+Gvn8ZQx5nWPHmNqevD4JvBrEfGJSAl2ZnBXjx5jaqbqMcn+d0ptTYp9+jrbmxh9+lj7pJSXyWiGUWdeJtBz3dYko+nFYzrammQ03XpMV1uTjKZbj84mz21NMpqpeIyh6bm9SUbPo8dwXLU3yeh59Oi5vUlGz6NHz+1NMnqpeJTkx1EyNa6W9eggccND6t4lZSYaY4YD5djClK7ZnOn2+gbQ3fF6CrBIRHqmbEokH7tGzfXGmPfx6DOGnmefxpiDgc7YNeEO9eoxhp5rjyIyBLsu1J3RH7nxlkDPaz7GOq+95GMsPS8eA8Aw4EFjzAjgMeB/PXiMp5eWcoNdo/ce53W6yna4piufYgfrTwVGG2NKsGXxSi8eE2i68miM+QC4EFiG/VvTZ0CVF48JNNMR73i+3PpNdNzRxpiB2BnlVcBVLtNIJi23XAj0BYZiB+fvSsmQyHDsPyWOD27yYiaGniePxph/GWMGY2ennygi7b14jKPnxeNcYGGMi4huPcbT8+IxXrlz6zFROXbrMYBd1ulQ7FqeY7F1sJfzMZ6mF5+IyFDs30xfCG7y4DGepluPY7DfuzP2x+QBIjLeo8d4mm49/hWoxp5HC7Hf2VObk0AzaY8p9L+T9plin77O9iaGXtykPXhMRJ15GU/PS1uTgqZrj17bmhQ03Xqci8e2JgVNtx49tzUp/m51W7Y9tTcp6HnxGNzuqr1JQc+LR0/tTQp6Xjx6am9S0EvaYwrjKJn4zZAT6CBxw+MzIq9wdMMWANcYYz5znrdhb4Q3PvERSZNWr8aYH40xPzqv38X+VWBEKhpiZ9zdC7xljAnOqHXtM5ZeOnw6x+4G7gMO9uIxlp5Hj+XYzth/ReRjwO88f+PSYzw9vORjnPPadT7G0vOYjxuB74wxzzjv7wNGevAYUy9N5WYA0A54ztnk+XyM1vTgcxggxvkLGvZ7e4p1PE0veWmMuccYM8oYMw57lfx9jx5jaqap/onna6NLv3G/pzEm+FwB3IL39uc77F/pCgFEpBmQb2qW5UgZY8znxphqY4zB3uQjaY8i0hf4BzDDGPOhV49x9Dx5DNP4L3bNv3IvHuPoefE4HrjEaReuAg4XkYUePMbTc+0xQblz5TFROfaQj59hZw1tMXZmz6PYNsdLrONpej0njwb+7hyLR4/xNN16nIH92/MuY/8K+xR2wMKLx3iarjwaY7YZY35rjBlmjDkAu3br+148JtBMymOK/e+k2ppU+/R1tTdx9OKRVF6mqFlnXsbT89LWpKjp2mPY8Sm3NSlquvXoqa1JUdOVR69tTaq/Wz2WbVftTYp6XjwGSbm9SVHPi0fX7U2Keq49emlvUtRLqT00yY2juP1dk/PoIHHD4zWgizPAAnAC8JBbMRFpJiItndd+7Dosb3t2aXkIOE5E/FLzd4Gn6jgmLiLSScT+tcTRG4e9G2YqLMTewfXcNPmspefFp4i0EJFOzmsfdo3Zd916jKfnxaMx5hZjTGdjTA9jTA+gynm+143HBHrNPORjvPPabT7G1POYj19jYxEcwJuGh1jH00tTuTkGe6ODaud9Osp2hKYHnxuBUucYsN/7PY8eY2p6LNsdnOeWwHnAjR49xtRMU7xj+jJ2jcrPRCS4vmOy7U9MPREJSM2sH4Bf47H9cTqOj2Nv4gd2qZBlXjTFrp0dJGmPItLF8XKKMeafXj3G0/PosZfYtemC51MZsN6Dx5h6XjwaYyaGtQ+/B+43xpzs1mM8PS8e45U7D/kYtxy79Ygth+UiUuD0B6Z68ZhI04tPR+cIwv6q67Vcx9L04PETYKrYv8IWYGe3vefRY0xNtx5FpJWjg4iUAz2Bp714jKeZgsek+98ptDVJaybZ3sTSi0kKeZm0JiSVl7F+d3hta5LW9ODRa1uTtKZbj2loa5LWdOsxDW1NSr9bPZZtt+1N0noePXppb5LW8+jRS3uTtJ4Xjx7bm6T1kvEoKY6jePhdk/uYLLh7nj7S+wCmYAv0f7B/Jw940OoFvIUtZO9iZ0g2daFzG3ZgxTjPC5zt8xyfHwKHedHDrhH6ruP3LeCYFD2WO3r/CtM4063PeHpefGL/GvGqE493sAuoN/XgMaae17yMSiP87teu4h1Lz2M+xj2vXeZjTL00nJMDgH86uquouWuv23JTSy8NHgXbqRgStd11rGNpeoz3idg6cR3wDPYmBF491tL06PF5R289cHya8rGWZqoeSbHuBoZgZy1/iO3stXCrh70JzmvO+fov4AGgfTIeseVvI/A9sM15PdnZvyv2LskfOuWgxKPeM46/ddgbBPZJ0uPtwA9hsXiLmjuvu/GYSM+tx+Oc82WdozczbH83HhPpufIY9flxRN4dPmWPdei5zce45c5lPibSc52PwGnYOuMd4GbAn4xHl5oJfSbQmwK8ESN9Lx7jaabsEVtv3UdN3ftXrx7r0HTjcYzjYT22nShNg8dEmnV5TLn/Td1tTUqa1NHexNPDW1vjRjNuXibQ89LWuNF04/E43Lc1bjRT9uixrXGj6SYfvbQ1bjS9lO2U2xuXel48ptzeuNRz5RGX7Y1LPbceXbU3LvXq8pjyOAp1tDUN9SHOl1cURVEURVEURVEURVEURVEaIbrchKIoiqIoiqIoiqIoiqIoSiNGB4kVRVEURVEURVEURVEURVEaMTpIrCiKoiiKoiiKoiiKoiiK0ojRQWJFURRFUZQsQERMEo/FP7OnlU66PaK2fywijebGFiKy2MmHyfXtJUi82KQ5jclOGh8n6eW4GJ/1EZFbReRDEdkpIltF5CMReUpELhKR3lH7z41x3v8kIp85x8z1+p1FZKaje0AS32tyHftcJyI7RKSbF0+KoiiKoij1TaC+DSiKoiiKoigR3JXgsxd/NheK4hER2Qd4BGgCfAk8B2wBugATgGlAJXBFjMPXYe9oDlAIdADKnGMuFpHrgfONMbtT9FQI/Bl41RjzeKrfKQZXAr91NGemQU9RFEVRFKVe0EFiRVEURVGULMIYc1x9ewhjJtAU+Ly+jdQzf8AOZH5a30bCyOrYiEgT4G7sAPHlwFxjTEXY582AA4FtcSQeMcbMjdIMADOA64BzsAPHR6Vo7RSgK3BWHN9TgAuAUUArYIWIfA2sBx42xtwavr8x5ksRuQs4WUSuMMa8l6IfRVEURVGUrECXm1AURVEURVFiYoz51BjzfvjgXmPEGPOlkw/b69tLkByIzQSgPfC5MebCaJ/GmG3GmHuNMY8mK2iMqTTG3ONobwWOFJFfpujrFOB74P+iPxCR/wGeBqYCHwBfA6uB/wKTgNPjaN4DCDArRS+KoiiKoihZgw4SK4qiKIqi5CjB9WJFJF9ELhWR/zjrvn4kIpc5f62PPqa5iPxBRNaJyA/OGrH/EZEHRGTfqH1TXvdWRMaJyKMi8o2I7HL83SwinWPse5yjP1dEuonIvc5xO0TkNRGZnmJ+hNYOFpH9ReRF5/ttFpGHRKRfHR76ish9IvK1iFSLyEHRujGObyYi5zl+fxSRbSLyvogsEJG+MfYvc/L6SxHZLSIbRWRRqmvaJlgvOnhO+B1fG5w4fCYiV4pIQSrpeKCd8/xNuoWNMe9jZxMDnJnscSIyCegD/CPO4Pql2N9HxxtjyoD3gT8bYyYBJcCf4kivwc4yPzpWmVMURVEURckFdJBYURRFURQltxHgH8Bs4D3gcaANcDHwfyLiD+1oXz+D/ft/Z2Cls/9XwP8AR3gyInI0dublL7EzMR8CdmFnb74Ra5DWoQfwKjAGeBZ4ExgJPOKsa5sqh2G/Vz7wGPAFcDDwkogMjXNMaZiH57EzShPO0hWRTsDL2KUoemHz8wnsEgqzsHkavv+pwFrgEOAT7Hq93wEnAK+JSP/UvmZC7gUuwsbhKaAFMAe4I41pJOIz53mwiEzMgP59zvN4EclP8phfOM8r43w+BNhsjLkz+gNjzHfGmPtjHWSMMcALQBEwPkkviqIoiqIoWYWuSawoiqIoipLbdMNe+B9kjPkIQETaYW8SNhU4g5pZl3tib/71KrCnMWZnUEREWmJnWbpCRLoCC523BxpjljnbfcDVwNnYNWpHxzj8WGefOcaYaue4s4FrsQOdT6Vo51TgZGPM7Y6WAH8FzgMWA8NjHDMDuAk42xhTlWQ6dwMDgaXACcaYrcEPnBm+LcPejwVuwN7A7UBjzOthn50ALALuBMYmmXYiugPbgT7GmK+cNHoCbwBHicgfjTH/SUM6iViLvWgxAHheRJZjLwC8DryWhqU71mMvQDTBXmTYkMQxwcHqV+N8vhVoISJFxpjNKfp5BTgGuyzFcykeqyiKoiiKUu/oTGJFURRFUZQswlkuIN7joDiHXRYcIAYwxnyDnVkMkeuoBpcAWBM+QOwc82P4wKULTsQO2C0NDhA7utXA+djZvKNEpDzGsf8FLggOEDvcBGwGxqYwUzTI2uAAsePBYGdWbwSGiciEGMd8A5yX7ACxiIzBDsJvAk4MHyB20vzYGPN22KbzAT8wKzqfjTF3AMuAMhGJNYDthjODA8ROGv/Frp0LNYOlGcPJxwOwg8V+5/U12Bm3W0RkmYiM8qBfjT0/wM7gTYYhQDXw7zifPwrkActF5BdAKktHvO88D0vhGEVRFEVRlKxBB4kVRVEURVGyi7sSPD6Nc8x90RuMMcuxg2h7OMsiALyFHST7jYicJCJt0+g7OPD49xhedgEPRO0XzkpjzO6oYyqxg8d5QKo+Y+VHBfBgAg/PpDi7dW/neYkx5qdEOzqzqadiZ/euiLPbaud5TAoe4lGBXTIjmuBs204xPks7zkB5OVAOXIkdIN6Kjel04J8i8msPSUgwqTp3FGmOvYixxbloEItzsMuxjMEuU1IG3C4iC0VkzzqS+N55bpdwL0VRFEVRlCxFl5tQFEVRFEXJIowxx6V4yOYEg5SfYGdZdga+NMZsEJE52KUXFgK3isg72GUAFkfNfE2V4I3pPo7zeXB7SYzPNsY5Jvi9Ur3Z2id1eKh1Ez3iD8DHo6vznMyyDcVAc+f1brv6RcJ9vfJVnBnRqeZnnYOvDgkHa40xa7EzinFmhU8D5mGXorhNRJ6InoldZ4J24D04g/j7RPs6tHKe4w7oG2N+AKY5aygfhF0GZQ+gN3CSiNwN/CZO3v7oPLdOwouiKIqiKErWoYPEiqIoiqIojQhjzNUishQ7CDYNO6v2HOBsETnHGHN9ppJO8Fl1gs9+LnbWvYtrgv/e24q9yWAi3k1DeunKz+DM6qZ17Bf8fFtdgs6M8cdF5HXsAHtr7M3eUl13eiD2xoTbiX9hIpwfnOcWSXhcDawWkZHYwewW2PWxjwFepGbt7XCCg9BbkvCiKIqiKIqSdeggsaIoiqIoSm5TJCIt4swm7uY8fxG+0RjzGXAjcKOIBLA3bbsTmCcif3Nx065gGqXYm6bFGujs4Tx/7kI7VbrXsf2LOJ+nwmfO8x5J7PstdhC6GjsTNdkZuvVNcIZ3cYJzDKBX1P51Yoz5SkTWAyNxN3v6cOf5RWdpkrrS2yoiO4DWIuKLWv86EduNMU84xz4K7E/sQeLgrOZvktRVFEVRFEXJKnRNYkVRFEVRlNyn1rquIrIP0Ab4yBjzZbwDjTGVxph7gFexMzP7uPQQXFP3iBhe8oHDovbLJLHyIwD8ynn7YhrSeMZ5PsJZ7zYuziDmSqAldm3inMA5b/6NXU7if2Lt49yIsA12lvRbYdsTrqkhIn6gp/M2pQsHItIPONt5m8rM93XY3z+9U0nPIbgcSbylOvo7z2/F+VxRFEVRFCWr0UFiRVEURVGU3OePItIj+EZEioH5ztsFYdv3EpG9nfVcCdveEzvIZUhhNmgUdwA7gBkickCYtg+4HLsW8evGmDUu9VNhgogcH7XtUuzM6red5QQ8YYx5BXtzuPbAQhFpFv65iPQQkcFhm/6CnUl8p4hMjtYTkeYicryINPHqLc0EB2HniUjf8A+cGyLe7Ly91RgTvmTHdBFZKiLjowWdvLoFO7j8BfDPZIyISEBEjsJeaGgG/M0Y80QK3yUY99Fx9K92yk70dj81g9KvxdEO3nDwhRT8KIqiKIqiZA263ISiKIqiKEoWISKLE3z8qTHmkuhtwNvAuyLyLFABTMGu9fo8cEPYvkOxa6t+46wJ+x3QDpiEnSF5ozHG1VIMxphPReS3wGLgMRFZg12SYQR2GYqvgaPdaLvgFmCR4+c/wBDsGrY/AselMZ1jsDf9OwLYV0ReBHZhl6AYBpwL/AvAGPOiiJwG3AQ879wwcAM2Xj2c/QuAh7CD7dnCAqAcuyTJO05cP8eeNxOBJtiB0ejz0oedPX6YiHwFvAlsdo4bhV2eYRtwjLNOcTQHhV34KAw7riV2sP1q4A8pfpfHgdnAZODvMT7/HXCaU47WYS8qnAPcBvTFns83Rh/kzJqehF2PeG2KnhRFURRFUbICHSRWFEVRFEXJLo5N8Nk6ag/GGeBQZ/uRQGfgS+zg3l+i1mv9P6AtsBd2wLgtdg3VF7EzQh/2YtwYc7eI/Ac4H3szsjLHyy2Ol59jPWKApcATwAXAgdiB2EeBC4wx76UrEWPM5yIyGjvL9FDsjQCrsLOxb8bmd/j+t4rIS87+k4FfYG+89jl20PIham6wlhUYY4yIHInNv99gB/0nAD9hZ9XeC9xhjKmIOnQ5cACwLzAOOwjeDjsA/l/sGtg3GGM+iZP0UOcBdjB5C/AysAZYnOC4RN/lBRHZAPxKRE6LMTg9CTsYPhY7M7gY6Io9hxcCc40xsdYcnuDsd2PUbGpFURRFUZScQXLnvhmKoiiKoihKOCJigE+MMT3q20s24MzCPhbYyxizsn7dKNmIiJwFXAccaoz5Rx37rsQODK+sY7/bgJOAwcaYWDdtVBRFURRFyXp0TWJFURRFURRFURoLt2GXaDk/HWLOuswzgXt0gFhRFEVRlFxGB4kVRVEURVEURWkUOMtBXAyMEpFfpEHyPOf5ojRoKYqiKIqi1Bu63ISiKIqiKEqOostNRKLLTSiKoiiKoiiKO3SQWFEURVEURVEURVEURVEUpRGjy00oiqIoiqIoiqIoiqIoiqI0YnSQWFEURVEURVEURVEURVEUpRGjg8SKoiiKoiiKoiiKoiiKoiiNGB0kVhRFURRFURRFURRFURRFacT8Pz9rCVwFj9x3AAAAAElFTkSuQmCC\n",
            "text/plain": [
              "<Figure size 1750x650 with 1 Axes>"
            ]
          },
          "metadata": {
            "tags": [],
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "rZPkQdeeVdPs",
        "colab_type": "text"
      },
      "source": [
        "## Once the model is satisfiably good, it will be exported as a tf model to the drive to be uploaded to [mia](http://miamarketplace.com/)."
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "l8LfhIpIG-Sz",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 139
        },
        "outputId": "c292a370-a7fa-4577-979f-1b90921e3c59"
      },
      "source": [
        "\"\"\"\n",
        "saving the model and the files for uploading to mia\n",
        "checking the tensorflow and python versions\n",
        "\"\"\"\n",
        "\n",
        "model.save('/content/drive/My Drive/MIA/', save_format='tf')\n",
        "\n",
        "print(tf.__version__)\n",
        "! python --version"
      ],
      "execution_count": 27,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "WARNING:tensorflow:From /usr/local/lib/python3.6/dist-packages/tensorflow/python/ops/resource_variable_ops.py:1817: calling BaseResourceVariable.__init__ (from tensorflow.python.ops.resource_variable_ops) with constraint is deprecated and will be removed in a future version.\n",
            "Instructions for updating:\n",
            "If using Keras pass *_constraint arguments to layers.\n",
            "INFO:tensorflow:Assets written to: /content/drive/My Drive/MIA/assets\n",
            "2.2.0\n",
            "Python 3.6.9\n"
          ],
          "name": "stdout"
        }
      ]
    }
  ]
}