Machine_learning_Eindhoven.ipynb

{
  "nbformat": 4,
  "nbformat_minor": 0,
  "metadata": {
    "colab": {
      "name": "Machine_learning_Eindhoven",
      "provenance": [],
      "collapsed_sections": [],
      "toc_visible": true,
      "mount_file_id": "1zYETJnh52r_PcfkrKan9_R4mNjUamfoC",
      "authorship_tag": "ABX9TyM5eVS6Q283CUhPtbP/+5WS",
      "include_colab_link": true
    },
    "kernelspec": {
      "name": "python3",
      "display_name": "Python 3"
    }
  },
  "cells": [
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "view-in-github",
        "colab_type": "text"
      },
      "source": [
        "<a href=\"https://colab.research.google.com/github/mostafahadizadeh1991/Machine_learning_Eindhoven/blob/main/Machine_learning_Eindhoven.ipynb\" target=\"_parent\"><img src=\"https://colab.research.google.com/assets/colab-badge.svg\" alt=\"Open In Colab\"/></a>"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "NPf2qm_YpkBq"
      },
      "source": [
        "<img src=\"https://upload.wikimedia.org/wikipedia/commons/thumb/d/d3/Eindhoven_University_of_Technology_logo.svg/800px-Eindhoven_University_of_Technology_logo.svg.png\" width=\"400\"/>\r\n",
        "\r\n",
        "\r\n",
        "# **PREDICTION OF EURUSD END-OF-DAY TRADED VOLUME USING DEEP LEARNING ON 2007-2020** "
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "A35WRY7tpnOA"
      },
      "source": [
        "\r\n",
        "# **Introduction:**\r\n",
        "\r\n",
        "---\r\n",
        "\r\n",
        "---\r\n",
        "\r\n",
        "\r\n",
        "The prediction of the end-of-day traded volume is an important quantity in finance. In particular, this is used by execution algorithms in high-frequency trading. The traded volume allows us to estimate the characteristic size of the asset that needs to be either bought or liquidated, and the related market impact ([inter alia, Almgren and Chriss, 2001, Gatheral, 2010](#Almgren)).\r\n",
        "\r\n",
        "The ECB publishes a daily fixing for selected EUR end-of-day traded volume: these reference mid-rates are based on a daily concertation procedure between central banks within and outside the European System of Central Banks, which normally takes place at 2.15 p.m. ECB time. The reference end-of-day traded volume are published both by electronic market information providers and on the ECB’s website shortly after the concertation procedure has been completed. Although only a reference rate, many financial institutions are ready to trade at the EUR fixing and it is therefore possible to leave orders with a bank for business to be transacted at this level. The ECB daily fixing of the EUR/USD is therefore a tradable level that makes using it a more realistic alternative to, say, London closing prices and this is the series that we investigate in this notebook. We examined the ECB daily fixing of the EUR/USD since its first trading day on 4 January 2007 until 31 December 2020.\r\n",
        "\r\n",
        "In financial applications, [Kamijo and Tanigawa (1990)](#Kamijo) applied them successfully to the recognition of stock patterns of the Tokyo stock exchange whereas [Tenti (1996)](#Tenti) achieved remarkable results using RNNs to forecast the exchange rate of the Deutsche Mark. [Tino, Schittenkopf and Doffner (2001)](#Tino) use them to trade successfully the volatility of the DAX and the FTSE 100 using straddles whereas [Dunis and Huang (2002)](#Dunis), using continuous implied volatility data from the currency options market, obtain remarkable results for their GBP/USD and USD/JPY exchange rate volatility trading simulation\r\n",
        "\r\n",
        "\r\n",
        "The motivation of this notebook is to investigate the use of a Deep learning when applied to the task of forecasting and trading the Euro/Dollar (EUR/USD) end-of-day traded volume using the European Central Bank (ECB) fixing series.\r\n",
        "\r\n",
        "A model that predicts the total end-of-day volume and updates the prediction as new intraday information is observed and proposed. This model is a time-varying coefficients model which is estimated using the framework of functional data. \r\n",
        "\r\n",
        "\r\n",
        "\r\n",
        "\r\n",
        "\r\n",
        "\r\n",
        " \r\n",
        "\r\n",
        "\r\n",
        "\r\n"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "k7Vk0bxnPx9O"
      },
      "source": [
        "# **Data and Methods:**"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "PgMWCbYLPyDG"
      },
      "source": [
        "This section presents some statistical properties of the EURUSD series both as a whole and through time.\r\n",
        "\r\n",
        "For this porpuse I received EURUSD daily data from the https://forexsb.com/historical-forex-data and as shown in the following figure I plotted all historical parameters of EURUSD.\r\n"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 17
        },
        "id": "OGmESTZMALG8",
        "outputId": "6b0ed840-7b97-43ae-9300-c0f8c5a28b94"
      },
      "source": [
        "import pandas as pd\r\n",
        "import csv\r\n",
        "import datetime\r\n",
        "\r\n",
        "EURUSD1440csv = \"/content/drive/MyDrive/Machine_learning_Eindhoven/EURUSD1440.csv\"\r\n",
        "EURUSD1440json = \"/content/drive/MyDrive/Machine_learning_Eindhoven/EURUSD1440.json\"\r\n",
        "\r\n",
        "data = []\r\n",
        "\r\n",
        "with open(EURUSD1440csv, 'r') as f:\r\n",
        "    reader = csv.reader(f)\r\n",
        "    next(f)\r\n",
        "    for row in reader:\r\n",
        "      dt = datetime.datetime.strptime(row[0], \"%Y-%m-%d %H:%M\")\r\n",
        "      data.append('['+str(dt.timestamp()*1000)+',' +row[1]+','+row[2]+','+row[3]+','+row[4]+','+row[5]+']')\r\n",
        "\r\n",
        "with open(EURUSD1440json, 'w') as f:\r\n",
        "    f.write('[%s]' % ', '.join(map(str, data)))\r\n"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/html": [
              "<script src=\"https://cdnjs.cloudflare.com/ajax/libs/proj4js/2.3.6/proj4.js\"></script>\n",
              "<script src=\"https://code.highcharts.com/stock/highstock.js\"></script>\n",
              "<script src=\"https://code.highcharts.com/maps/modules/map.js\"></script>\n",
              "<script src=\"https://code.highcharts.com/modules/data.js\"></script>\n",
              "<script src=\"https://code.highcharts.com/mapdata/countries/in/in-all.js\"></script>\n",
              "\n"
            ],
            "text/plain": [
              "<IPython.core.display.HTML object>"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 90
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 526
        },
        "id": "d9uvXoaUOWqs",
        "outputId": "51ad3f7f-9ae2-4d4b-9b47-0bbb760909ba"
      },
      "source": [
        "%%html \r\n",
        "<body>  \r\n",
        "<div class=\"chart-wrapper\"><div class=\"chart-inner\">\r\n",
        "<div id=\"container\" style=\"width:100%; height: 100%;\"></div></div></div>\r\n",
        "\r\n",
        " </body>  \r\n",
        "<html>  \r\n",
        " <head>  \r\n",
        "  <meta http-equiv=\"Content-Type\" content=\"text/html; charset=utf-8\">  \r\n",
        "  <title>Highcharts Example</title>  \r\n",
        "  \r\n",
        "<script src=\"https://cdnjs.cloudflare.com/ajax/libs/proj4js/2.3.6/proj4.js\"></script>\r\n",
        "<script src=\"https://code.highcharts.com/stock/highstock.js\"></script>\r\n",
        "<script src=\"https://code.highcharts.com/maps/modules/map.js\"></script>\r\n",
        "<script src=\"https://code.highcharts.com/modules/data.js\"></script>\r\n",
        "<script src=\"https://code.highcharts.com/mapdata/countries/in/in-all.js\"></script>\r\n",
        "<style>\r\n",
        "  .chart-wrapper {\r\n",
        " position: relative;\r\n",
        "    padding-bottom: 50%;\r\n",
        "    width:80%;\r\n",
        "    float:left;\r\n",
        "}\r\n",
        "\r\n",
        ".chart-inner {\r\n",
        "position: absolute;\r\n",
        "    width: 100%; height: 100%;\r\n",
        "}\r\n",
        "</style>\r\n",
        "\r\n",
        "\r\n",
        "<script>\r\n",
        "Highcharts.getJSON('https://raw.githubusercontent.com/mostafahadizadeh1991/Machine_learning_Eindhoven/main/EURUSD1440.json', function (data) {\r\n",
        "\r\n",
        "    // split the data set into ohlc and volume\r\n",
        "    var ohlc = [],\r\n",
        "        volume = [],\r\n",
        "        dataLength = data.length,\r\n",
        "        // set the allowed units for data grouping\r\n",
        "        groupingUnits = [[\r\n",
        "            'week',                         // unit name\r\n",
        "            [1]                             // allowed multiples\r\n",
        "        ], [\r\n",
        "            'month',\r\n",
        "            [1, 2, 3, 4, 6]\r\n",
        "        ]],\r\n",
        "\r\n",
        "        i = 0;\r\n",
        "\r\n",
        "    for (i; i < dataLength; i += 1) {\r\n",
        "        ohlc.push([\r\n",
        "            data[i][0], // the date\r\n",
        "            data[i][1], // open\r\n",
        "            data[i][2], // high\r\n",
        "            data[i][3], // low\r\n",
        "            data[i][4] // close\r\n",
        "        ]);\r\n",
        "\r\n",
        "        volume.push([\r\n",
        "            data[i][0], // the date\r\n",
        "            data[i][5] // the volume\r\n",
        "        ]);\r\n",
        "    }\r\n",
        "\r\n",
        "\r\n",
        "    // create the chart\r\n",
        "    Highcharts.stockChart('container', {\r\n",
        "\r\n",
        "        rangeSelector: {\r\n",
        "            selected: 1\r\n",
        "        },\r\n",
        "\r\n",
        "        title: {\r\n",
        "            text: 'EURUSD Historical'\r\n",
        "        },\r\n",
        "\r\n",
        "        yAxis: [{\r\n",
        "            labels: {\r\n",
        "                align: 'right',\r\n",
        "                x: -3\r\n",
        "            },\r\n",
        "            title: {\r\n",
        "                text: 'OHLC'\r\n",
        "            },\r\n",
        "            height: '60%',\r\n",
        "            lineWidth: 2,\r\n",
        "            resize: {\r\n",
        "                enabled: true\r\n",
        "            }\r\n",
        "        }, {\r\n",
        "            labels: {\r\n",
        "                align: 'right',\r\n",
        "                x: -3\r\n",
        "            },\r\n",
        "            title: {\r\n",
        "                text: 'Volume'\r\n",
        "            },\r\n",
        "            top: '65%',\r\n",
        "            height: '35%',\r\n",
        "            offset: 0,\r\n",
        "            lineWidth: 2\r\n",
        "        }],\r\n",
        "\r\n",
        "        tooltip: {\r\n",
        "            split: true\r\n",
        "        },\r\n",
        "        plotOptions: {\r\n",
        "            candlestick: {\r\n",
        "                color: 'blue',\r\n",
        "                upColor: 'red'\r\n",
        "            }\r\n",
        "        },\r\n",
        "        series: [{\r\n",
        "            type: 'candlestick',\r\n",
        "            name: 'AAPL',\r\n",
        "            data: ohlc,\r\n",
        "            dataGrouping: {\r\n",
        "                units: groupingUnits\r\n",
        "            }\r\n",
        "        }, {\r\n",
        "            type: 'column',\r\n",
        "            name: 'Volume',\r\n",
        "            data: volume,\r\n",
        "            yAxis: 1,\r\n",
        "            dataGrouping: {\r\n",
        "                units: groupingUnits\r\n",
        "            }\r\n",
        "        }]\r\n",
        "    });\r\n",
        "});\r\n",
        "\r\n",
        "</script>\r\n",
        "\r\n",
        " </head>  \r\n",
        "\r\n",
        "</html> "
      ],
      "execution_count": 43,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "text/html": [
              "<body>  \n",
              "<div class=\"chart-wrapper\"><div class=\"chart-inner\">\n",
              "<div id=\"container\" style=\"width:100%; height: 100%;\"></div></div></div>\n",
              "\n",
              " </body>  \n",
              "<html>  \n",
              " <head>  \n",
              "  <meta http-equiv=\"Content-Type\" content=\"text/html; charset=utf-8\">  \n",
              "  <title>Highcharts Example</title>  \n",
              "  \n",
              "<script src=\"https://cdnjs.cloudflare.com/ajax/libs/proj4js/2.3.6/proj4.js\"></script>\n",
              "<script src=\"https://code.highcharts.com/stock/highstock.js\"></script>\n",
              "<script src=\"https://code.highcharts.com/maps/modules/map.js\"></script>\n",
              "<script src=\"https://code.highcharts.com/modules/data.js\"></script>\n",
              "<script src=\"https://code.highcharts.com/mapdata/countries/in/in-all.js\"></script>\n",
              "<style>\n",
              "  .chart-wrapper {\n",
              " position: relative;\n",
              "    padding-bottom: 50%;\n",
              "    width:80%;\n",
              "    float:left;\n",
              "}\n",
              "\n",
              ".chart-inner {\n",
              "position: absolute;\n",
              "    width: 100%; height: 100%;\n",
              "}\n",
              "</style>\n",
              "\n",
              "\n",
              "<script>\n",
              "Highcharts.getJSON('https://raw.githubusercontent.com/mostafahadizadeh1991/Machine_learning_Eindhoven/main/EURUSD1440.json', function (data) {\n",
              "\n",
              "    // split the data set into ohlc and volume\n",
              "    var ohlc = [],\n",
              "        volume = [],\n",
              "        dataLength = data.length,\n",
              "        // set the allowed units for data grouping\n",
              "        groupingUnits = [[\n",
              "            'week',                         // unit name\n",
              "            [1]                             // allowed multiples\n",
              "        ], [\n",
              "            'month',\n",
              "            [1, 2, 3, 4, 6]\n",
              "        ]],\n",
              "\n",
              "        i = 0;\n",
              "\n",
              "    for (i; i < dataLength; i += 1) {\n",
              "        ohlc.push([\n",
              "            data[i][0], // the date\n",
              "            data[i][1], // open\n",
              "            data[i][2], // high\n",
              "            data[i][3], // low\n",
              "            data[i][4] // close\n",
              "        ]);\n",
              "\n",
              "        volume.push([\n",
              "            data[i][0], // the date\n",
              "            data[i][5] // the volume\n",
              "        ]);\n",
              "    }\n",
              "\n",
              "\n",
              "    // create the chart\n",
              "    Highcharts.stockChart('container', {\n",
              "\n",
              "        rangeSelector: {\n",
              "            selected: 1\n",
              "        },\n",
              "\n",
              "        title: {\n",
              "            text: 'EURUSD Historical'\n",
              "        },\n",
              "\n",
              "        yAxis: [{\n",
              "            labels: {\n",
              "                align: 'right',\n",
              "                x: -3\n",
              "            },\n",
              "            title: {\n",
              "                text: 'OHLC'\n",
              "            },\n",
              "            height: '60%',\n",
              "            lineWidth: 2,\n",
              "            resize: {\n",
              "                enabled: true\n",
              "            }\n",
              "        }, {\n",
              "            labels: {\n",
              "                align: 'right',\n",
              "                x: -3\n",
              "            },\n",
              "            title: {\n",
              "                text: 'Volume'\n",
              "            },\n",
              "            top: '65%',\n",
              "            height: '35%',\n",
              "            offset: 0,\n",
              "            lineWidth: 2\n",
              "        }],\n",
              "\n",
              "        tooltip: {\n",
              "            split: true\n",
              "        },\n",
              "        plotOptions: {\n",
              "            candlestick: {\n",
              "                color: 'blue',\n",
              "                upColor: 'red'\n",
              "            }\n",
              "        },\n",
              "        series: [{\n",
              "            type: 'candlestick',\n",
              "            name: 'AAPL',\n",
              "            data: ohlc,\n",
              "            dataGrouping: {\n",
              "                units: groupingUnits\n",
              "            }\n",
              "        }, {\n",
              "            type: 'column',\n",
              "            name: 'Volume',\n",
              "            data: volume,\n",
              "            yAxis: 1,\n",
              "            dataGrouping: {\n",
              "                units: groupingUnits\n",
              "            }\n",
              "        }]\n",
              "    });\n",
              "});\n",
              "\n",
              "</script>\n",
              "\n",
              " </head>  \n",
              "\n",
              "</html> "
            ],
            "text/plain": [
              "<IPython.core.display.HTML object>"
            ]
          },
          "metadata": {
            "tags": []
          }
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "IM7PvhjH47RN"
      },
      "source": [
        "# **Main Section:**\r\n",
        "\r\n",
        "\r\n",
        "\r\n",
        "---\r\n",
        "\r\n",
        "---\r\n",
        "\r\n",
        "In this section I designed a deep learning method that predicted end-of-day traded volume of EURUSD. For this purpose I considered 30 percent of data as validation set and 70 percent as training set.\r\n",
        "\r\n"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "dd2yZ9_urSQY"
      },
      "source": [
        "import tensorflow as tf\r\n",
        "from keras.layers.core import Masking, RepeatVector\r\n",
        "from keras.layers.wrappers import TimeDistributed\r\n",
        "from tensorflow.keras.callbacks import LearningRateScheduler\r\n",
        "from tensorflow.keras.layers import Conv1D, Lambda, LSTM, Bidirectional, SimpleRNN, Dense\r\n",
        "from tensorflow.keras.losses import Huber\r\n",
        "from tensorflow.keras.optimizers import SGD\r\n",
        "from tensorflow.keras.metrics import mean_absolute_error\r\n",
        "import numpy as np\r\n",
        "%matplotlib inline\r\n",
        "import matplotlib.pyplot as plt\r\n",
        "import os\r\n",
        "import csv"
      ],
      "execution_count": 1,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 762
        },
        "id": "OCrRGhRFveA8",
        "outputId": "68fa5943-2768-4aab-fad1-dd379b8a1b7c"
      },
      "source": [
        "import pandas as pd\r\n",
        "import matplotlib.dates as mdates\r\n",
        "import matplotlib.pyplot as plt\r\n",
        "\r\n",
        "fig = plt.subplots(figsize = (21,15))\r\n",
        "\r\n",
        "ax = plt.subplot(221)\r\n",
        "\r\n",
        "\r\n",
        "data = pd.read_csv(\"/content/drive/MyDrive/Machine_learning_Eindhoven/EURUSD1440.csv\",sep=\",\")\r\n",
        "data['date'] = pd.to_datetime(data['date'])\r\n",
        "data = data.set_index('date')\r\n",
        "\r\n",
        "data_columns = ['vol']\r\n",
        "\r\n",
        "ax.plot(data[data_columns], linewidth=2, color='0.2', label='End-of-day traded volume')\r\n",
        "\r\n",
        "ax.legend()\r\n",
        "ax.set_xlabel('time')\r\n",
        "ax.set_ylabel('volume')\r\n",
        "ax.set_title('EURUSD End-of-day traded volume from 2007-2020')\r\n",
        "\r\n",
        "# Convert to numpy arrays\r\n",
        "vol = data[data_columns].values.tolist()\r\n",
        "#remove 'nan' data\r\n",
        "vol = [x for x in vol if str(x) != 'nan']\r\n",
        "vols_final = np.array(vol)\r\n",
        "id = len(vols_final)\r\n",
        "\r\n",
        "#0.7 is training set section\r\n",
        "criteria = int(id*0.7)\r\n",
        "training_vol = vols_final[: criteria]\r\n",
        "validation_vol = vols_final[criteria:]\r\n",
        "\r\n",
        "\r\n",
        "ax1 = plt.subplot(222)\r\n",
        "ax1.plot(training_vol, linewidth=2, color='0.2', label='End-of-day traded volume')\r\n",
        "ax1.legend()\r\n",
        "ax1.set_xlabel('time')\r\n",
        "ax1.set_ylabel('volume')\r\n",
        "ax1.set_title('EURUSD End-of-day traded volume')\r\n",
        "\r\n",
        "ax2 = plt.subplot(223)\r\n",
        "ax2.plot(validation_vol, linewidth=2, color='0.2', label='End-of-day traded volume')\r\n",
        "ax2.legend()\r\n",
        "ax2.set_xlabel('time')\r\n",
        "ax2.set_ylabel('volume')\r\n",
        "ax2.set_title('EURUSD End-of-day traded volume')\r\n",
        "\r\n"
      ],
      "execution_count": 2,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "Text(0.5, 1.0, 'EURUSD End-of-day traded volume')"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 2
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 1512x1080 with 3 Axes>"
            ]
          },
          "metadata": {
            "tags": [],
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "5qkdmWbqwncE"
      },
      "source": [
        "def get_windowed_dataset(series, window_size, batch_size, shuffle_buffer_size):\r\n",
        "    series = tf.expand_dims(series, axis=-1)\r\n",
        "    ds = tf.data.Dataset.from_tensor_slices(series)\r\n",
        "    ds = ds.window(size=window_size + 1, shift=1, drop_remainder=True)\r\n",
        "    ds = ds.flat_map(lambda win: win.batch(window_size + 1))\r\n",
        "    ds = ds.shuffle(shuffle_buffer_size)\\\r\n",
        "        .map(lambda win: (win[: -1], win[-1]))\r\n",
        "    ds = ds.batch(batch_size).prefetch(1)\r\n",
        "    return ds\r\n",
        "\r\n",
        "def get_forecast(model, series, window_size, batch_size, shuffle_buffer_size):\r\n",
        "    ds = tf.data.Dataset.from_tensor_slices(series)\r\n",
        "    ds = ds.window(size=window_size, shift=1, drop_remainder=True)\r\n",
        "    ds = ds.flat_map(lambda win: win.batch(window_size))\r\n",
        "    ds = ds.batch(batch_size).prefetch(1)\r\n",
        "    forecast = model.predict(ds)\r\n",
        "    return forecast"
      ],
      "execution_count": 3,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "pqv6NCb-woBT",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "b1f281b8-fb5a-49c1-ad96-6b1cc8ec86ad"
      },
      "source": [
        "WINDOW_SIZE = 60\r\n",
        "BATCH_SIZE = 128\r\n",
        "BUFFER_SIZE = 1000\r\n",
        "\r\n",
        "ds_vol = get_windowed_dataset(training_vol, WINDOW_SIZE, BATCH_SIZE, BUFFER_SIZE)\r\n",
        "\r\n",
        "model = tf.keras.Sequential([\r\n",
        "    Masking(input_shape=[None, 1]),\r\n",
        "    LSTM(128, return_sequences=True),\r\n",
        "    LSTM(128),\r\n",
        "    RepeatVector(16),\r\n",
        "    Dense(64, activation='relu'),\r\n",
        "    Dense(64, activation='relu'),\r\n",
        "    TimeDistributed(Dense(64, activation='relu')),\r\n",
        "    Dense(1),\r\n",
        "    Lambda(lambda x: x * 25.0)\r\n",
        "])\r\n",
        "\r\n",
        "\r\n",
        "\r\n",
        "\r\n",
        "lr_scheduler = LearningRateScheduler(lambda epoch: 1e-8 * 10 ** (epoch / 20))\r\n",
        "\r\n",
        "model.compile(loss=Huber(), optimizer=SGD(lr=1e-8, momentum=0.9), metrics=['mae'])\r\n",
        "hist_vol = model.fit(ds_vol, epochs=20, callbacks=[lr_scheduler], verbose=2)\r\n"
      ],
      "execution_count": 4,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Epoch 1/20\n",
            "24/24 - 18s - loss: 291124.7500 - mae: 291125.0625\n",
            "Epoch 2/20\n",
            "24/24 - 11s - loss: 291124.7812 - mae: 291125.1250\n",
            "Epoch 3/20\n",
            "24/24 - 11s - loss: 291124.7500 - mae: 291125.0625\n",
            "Epoch 4/20\n",
            "24/24 - 11s - loss: 291124.7500 - mae: 291125.0625\n",
            "Epoch 5/20\n",
            "24/24 - 11s - loss: 291124.7812 - mae: 291125.1250\n",
            "Epoch 6/20\n",
            "24/24 - 11s - loss: 291124.7188 - mae: 291125.0625\n",
            "Epoch 7/20\n",
            "24/24 - 11s - loss: 291124.7188 - mae: 291125.0625\n",
            "Epoch 8/20\n",
            "24/24 - 11s - loss: 291124.7188 - mae: 291125.0625\n",
            "Epoch 9/20\n",
            "24/24 - 11s - loss: 291124.7188 - mae: 291125.0625\n",
            "Epoch 10/20\n",
            "24/24 - 11s - loss: 291124.3438 - mae: 291124.8750\n",
            "Epoch 11/20\n",
            "24/24 - 11s - loss: 291124.1875 - mae: 291124.8125\n",
            "Epoch 12/20\n",
            "24/24 - 11s - loss: 291124.2188 - mae: 291124.8750\n",
            "Epoch 13/20\n",
            "24/24 - 11s - loss: 291124.1875 - mae: 291124.8125\n",
            "Epoch 14/20\n",
            "24/24 - 11s - loss: 291124.1875 - mae: 291124.7500\n",
            "Epoch 15/20\n",
            "24/24 - 11s - loss: 291124.1250 - mae: 291124.7500\n",
            "Epoch 16/20\n",
            "24/24 - 11s - loss: 291124.0625 - mae: 291124.7500\n",
            "Epoch 17/20\n",
            "24/24 - 11s - loss: 291124.0938 - mae: 291124.6875\n",
            "Epoch 18/20\n",
            "24/24 - 11s - loss: 291123.8750 - mae: 291124.2812\n",
            "Epoch 19/20\n",
            "24/24 - 11s - loss: 291123.8438 - mae: 291124.1562\n",
            "Epoch 20/20\n",
            "24/24 - 11s - loss: 291123.8125 - mae: 291124.1562\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 357
        },
        "id": "mlICNkvqwrJj",
        "outputId": "05585e18-cc89-49c6-a35d-f6b5424d4814"
      },
      "source": [
        "fig = plt.subplots(figsize = (10,5))\r\n",
        "\r\n",
        "ax = plt.subplot(111)\r\n",
        "ax.semilogx(hist_vol.history['lr'], hist_vol.history['loss'])\r\n",
        "ax.set_title('Learning Rate End-of-day traded volume Training set')\r\n"
      ],
      "execution_count": 5,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "Text(0.5, 1.0, 'Learning Rate End-of-day traded volume Training set')"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 5
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 720x360 with 1 Axes>"
            ]
          },
          "metadata": {
            "tags": [],
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "iHgBtOa8wt0K",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "bf9833ba-1bd5-400b-80f4-292a8279853e"
      },
      "source": [
        "model.compile(loss=Huber(), optimizer=SGD(lr=3e-5, momentum=0.9), metrics=['mae'])\r\n",
        "history_vol = model.fit(ds_vol, epochs=30, verbose=2)\r\n"
      ],
      "execution_count": 7,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Epoch 1/30\n",
            "24/24 - 18s - loss: 291111.7812 - mae: 291112.2500\n",
            "Epoch 2/30\n",
            "24/24 - 11s - loss: 290960.3125 - mae: 290960.8438\n",
            "Epoch 3/30\n",
            "24/24 - 11s - loss: 257124.4531 - mae: 257124.9531\n",
            "Epoch 4/30\n",
            "24/24 - 14s - loss: 210134.8281 - mae: 210135.3125\n",
            "Epoch 5/30\n",
            "24/24 - 14s - loss: 203742.1094 - mae: 203742.5781\n",
            "Epoch 6/30\n",
            "24/24 - 11s - loss: 207200.0625 - mae: 207200.5469\n",
            "Epoch 7/30\n",
            "24/24 - 11s - loss: 219841.6875 - mae: 219842.1875\n",
            "Epoch 8/30\n",
            "24/24 - 11s - loss: 250564.6406 - mae: 250565.1250\n",
            "Epoch 9/30\n",
            "24/24 - 11s - loss: 206909.4219 - mae: 206909.8906\n",
            "Epoch 10/30\n",
            "24/24 - 11s - loss: 202686.8125 - mae: 202687.3125\n",
            "Epoch 11/30\n",
            "24/24 - 11s - loss: 202735.0156 - mae: 202735.4844\n",
            "Epoch 12/30\n",
            "24/24 - 11s - loss: 208112.2500 - mae: 208112.7344\n",
            "Epoch 13/30\n",
            "24/24 - 11s - loss: 227357.2344 - mae: 227357.7656\n",
            "Epoch 14/30\n",
            "24/24 - 11s - loss: 208472.2656 - mae: 208472.7656\n",
            "Epoch 15/30\n",
            "24/24 - 11s - loss: 210469.2812 - mae: 210469.7500\n",
            "Epoch 16/30\n",
            "24/24 - 11s - loss: 233821.7500 - mae: 233822.1875\n",
            "Epoch 17/30\n",
            "24/24 - 11s - loss: 218563.6875 - mae: 218564.2656\n",
            "Epoch 18/30\n",
            "24/24 - 11s - loss: 205313.2344 - mae: 205313.7500\n",
            "Epoch 19/30\n",
            "24/24 - 11s - loss: 202554.8906 - mae: 202555.4062\n",
            "Epoch 20/30\n",
            "24/24 - 11s - loss: 203013.5156 - mae: 203014.0000\n",
            "Epoch 21/30\n",
            "24/24 - 11s - loss: 202141.8750 - mae: 202142.3594\n",
            "Epoch 22/30\n",
            "24/24 - 11s - loss: 216314.0156 - mae: 216314.5469\n",
            "Epoch 23/30\n",
            "24/24 - 11s - loss: 202649.2656 - mae: 202649.7812\n",
            "Epoch 24/30\n",
            "24/24 - 11s - loss: 207406.6875 - mae: 207407.1406\n",
            "Epoch 25/30\n",
            "24/24 - 11s - loss: 213646.7812 - mae: 213647.2812\n",
            "Epoch 26/30\n",
            "24/24 - 11s - loss: 213752.9531 - mae: 213753.4219\n",
            "Epoch 27/30\n",
            "24/24 - 11s - loss: 206949.5625 - mae: 206950.0938\n",
            "Epoch 28/30\n",
            "24/24 - 11s - loss: 213737.1719 - mae: 213737.6406\n",
            "Epoch 29/30\n",
            "24/24 - 11s - loss: 203755.6875 - mae: 203756.1719\n",
            "Epoch 30/30\n",
            "24/24 - 11s - loss: 206808.9375 - mae: 206809.4688\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "_TdEVxSVwwkq"
      },
      "source": [
        "forecast_vol = get_forecast(model, vols_final[..., np.newaxis], WINDOW_SIZE, BATCH_SIZE, BUFFER_SIZE)\r\n"
      ],
      "execution_count": 11,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "nANkGkU8wz0C"
      },
      "source": [
        "result_vol = forecast_vol[criteria - WINDOW_SIZE: -1, 0]\r\n"
      ],
      "execution_count": 12,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 536
        },
        "id": "RlqvF4FFw5tb",
        "outputId": "c613798c-21fd-464d-96d5-f5d90f1e7b41"
      },
      "source": [
        "fig = plt.subplots(figsize = (21,10))\r\n",
        "\r\n",
        "ax = plt.subplot(111)\r\n",
        "ax.plot(validation_vol, linewidth=2, label='Actual')\r\n",
        "ax.plot(result_vol, linewidth=2, label='Predicted')\r\n",
        "\r\n",
        "ax.legend()\r\n",
        "ax.set_xlabel('time')\r\n",
        "ax.set_ylabel('volume ')\r\n",
        "ax.set_title('Comparison EURUSD End-of-day traded volume Validation set and predicted')\r\n"
      ],
      "execution_count": 13,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "Text(0.5, 1.0, 'Comparison EURUSD End-of-day traded volume Validation set and predicted')"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 13
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "iVBORw0KGgoAAAANSUhEUgAABMcAAAJcCAYAAAD94sPtAAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADh0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uMy4yLjIsIGh0dHA6Ly9tYXRwbG90bGliLm9yZy+WH4yJAAAgAElEQVR4nOzde7zldV0v/td7LjAoiHLJVFQoLwkeQAWvaFpeKG+pmaVpltfKbqad8vgz7ZTH4zmlR+yEmKWeEk1CzQQzA8JLqKCkBpYYKAjKMMAAM8z98/vj+90za+/Ze8+ey9p7Md/n8/HYj1lrfS/r813ru/bMes378/5Way0AAAAAMETLlnoAAAAAALBUhGMAAAAADJZwDAAAAIDBEo4BAAAAMFjCMQAAAAAGSzgGAAAAwGAJxwAYpKp6QVV9aqnHwXRVdVVVPXEPtz2oqj5eVWur6sMLWP/FVfXZPXmupVBVrarut9jbzrPPO9TrN5uqemNV/VV/+z5VdVtVLd/Vunv4XP9WVY/f0+3v6MZxDu7Gc19QVS/tby/K7/6qOro/5hXjfi4A9p5wDIC9UlXPr6qL+y+V11XVuVV1ylKPa1daa3/dWnvyYj5nH/zc3r9WUz/v7JfN+sV79Atl/wVvQ7/dDVV1dlXdY2Td91bVH87YftoXtKo6pao+3wdIN1bV56rq5H7Zi6tq68jYrqyqv6yqB8xzTI+vqm0zjum2qnrUvnnVdstPJ7l7ksNba89dgudPsvchCgtXVY+sqnVVdfAsy75SVa9a6L5aa99prR3cWtu6D8a102extXZca+2Cvd33vrKUYdVSWujvfp9jgGERjgGwx6rq1UnenuTN6UKJ+yT5v0meuZTj2pUl/p/8p/dfwKd+Fvzlvfeq1trBSe6X5OAk/3uhG1bVXZL8fZLTkhyW5F5J3pRk48hq/9Lv/9AkT0xye5JLqurB8+z62hnHdHBr7V9266j2jfsm+Y/W2pYleO4Fq45/g+0DrbWLklyTLhjdrj9fj01y5lKMi/FRiQXAOPiHGQB7pKoOTfIHSX61tXZ2a21da21za+3jrbXX9uscWFVvr6pr+5+3V9WB/bLHV9U1VfU7VXV9X3X2U1X1k1X1H31V0+tGnu+NVXVWVX2oqm6tqi9X1Qkjy3+3qr7VL7usqp41suzFfYXU26pqTZI3jk4J68OKt/XjuKWqvjYVBlXVoVX1/qpaXVXfrqrXTwUbU/uoqv9dVTf1lVY/sQgvf1prNyf5aJITd2OzB/Tbntla29pau7219qnW2ldn2f/W1tq3Wmu/kuSfk7xxT8bZV7v99/71v7WqPlVVR4wsf2H/uq6pqv+2gP09qN/nzdVNU3tG//ibkrwhyfP6yrWXzLLt4VX1d/17/MUkPzxj+f+pqqv75ZdU1WP7x3+wqtZX1eEj6z60PydWztjHqUleNzKOfx15Hf6oqj6XZH2SH6qqX6yqy/vX5T+r6hUz9vXa/nNxbVX90oxlB/bn3Xeq6vtVdXpVHbSQbWfs53lVdfGMx36rqv6uvz3n+T9jm52mkNX0qWyjn8Gb++N9dP/41f1n7xcWenwzvC/Ji2Y89qIk57TW1sz1vu7qGKrqmKr65/79+cckR8xY/8NV9b3qqjAvrKrj+sdfnuQFSX6nPwc+3j++fcpwLex342/Xjt+NvzjHsU+9tv/Zj/PKqnrByLJf6s+xm6rqH6rqvv3jF/ar/Gs/xufNst8frqrzqvts3lBVf11Vdx1ZflVVvaaqvtq/Bh+qqlUjyxd0DvbrXlBV/6Oqvti/Tx+rqsNmvC8vqarvJDlvvmPrlz2pqr7Rj+udSWrG6/XZkfvHVdU/Vvd3zver6nU19+f40Kp6T39c362qP6x+Gm5VLe/P2Ruq6j+TPHW+YwZgsgjHANhTj0qyKslH5lnnvyV5ZLoA54QkD0/y+pHlP9jv417pgo13J/n5JA9L8tgk/19VHTOy/jOTfDhd1dMHkny0doQT3+q3OTRdNdRf1ciUwySPSPKf6Src/mjGOJ+c5HHpwqNDk/xMkjX9stP6x34oyY+m+9I9+kX1EUn+Pd0X57cmeU9VVcasuqDm2Umu2I3N/iPJ1qp6X1X9RFXdbYHbnZ3utd1Tz0/3mv1AkgOSvCZJqurYJH+W5IVJ7pnk8CRHzbWT/r3+eJJP9fv6tSR/XVUPbK39froKxg/1lWvvmWUXf5pkQ5J7JPml/mfUl9Kdq1Pn14eralVr7XtJLkh3Xkx5YZIPttY2j+6gtfbJGeM4YcY2L09ySJJvJ7k+ydOS3KV/fd5WVQ/tj/XU/nV6UpL7p6viG/WWdOfriemqCKc+QwvZdtTHkzywqu4/8tjz++NPdn3+745HJPlquvf5A0k+mOTkfvw/n+SdtWN65JzHN4v/l+RxVXXvJKkuvHt+utAsmeN9XcB4P5DkknSf7f+e5BdmLD833ev7A0m+nOSvk6S1dkZ/+639OfD0Wfa9kN+Nh/bH/ZIkfzrb57Wq7pzkHUl+orV2SJJHJ7m0X/bMdAHPs5McmeQz6SvpWmuP63dxQj/GD80yxkryP9J9Nh+U5N7ZOST/mSSnJjkmyfFJXtw/9+6cg1NelO4zeY8kW/rjGvWj/TieMt+xVRe+n53u9Twi3d8Nj5ntCavqkCSfTvLJ/jjvl+Sf5vkcv7cf2/2SPCTd3x0v7Ze9LN3n+SFJTsqMakYAJlxr7Q73k+Qv0v2D8usLXP9nklyW5N+SfGCpx+/Hjx8/+8NPusqI7+1inW8l+cmR+09JclV/+/Hppuwt7+8fkqQlecTI+pck+an+9huTXDSybFmS65I8do7nvjTJM/vbL07ynRnLX5zks/3tH0sXHD0yybKRdZYn2ZTk2JHHXpHkgpF9XDGy7E79MfzgHGO6KsltSW4e+XnZyPH91SzbtCT3629fkK7qaG3/+KVJ7jOy7nuT/OGM7Y/u113R339Qv9416b7k/V2Su898TWbs49Qkm+c4pscn2TbjmG5OcueRMb9+ZP1fSfLJ/vYb0gVMU8vu3L/eT5zjuR6b5Hsz3qMzk7xxvtdw5L3cnORHRh5782zHO7L8pnThQZI8L8nnRvb1vSQPn2O7ncbRvw5/sIvPy0eT/EZ/+y+SvGVk2QOmzoV0ocW6JD88svxRSa7c1bZzPO9fJXlDf/v+SW5Ndy4v5Pyf+gxNO89GjvmlI+t+c2TZf+nXv/vIY2vShUXzHt8cx/DpJK/rbz8pyeokKxfwvm5/r0aPId0U8S3pz+N++QfmOb/u2m976DyfxavSn9tZ2O/G0dfy+iSPnOV575zu8/acJAfNWHZukpeM3F+W7vfHffv7c54TcxzjTyX5yozj+fmR+29NcvoenoMXzFj/2P7cWz7yvvzQQo4tXcg2+ndFpft9N3ouTp23Pzd6TPN9jtP9x8rG0de53/78/vZ5SV45suzJmfGZ8OPHjx8/k/tzR60ce2+6f6jvUv8/ob+X5DGtteOS/OYYxwUwJGuSHFHz93+5Z7oKmSnf7h/bvo+2o/n17f2f3x9Zfnu6vlpTrp660Vrblu4Lzz2TpKpeVFWXVjdl6+YkD870aVBXZw6ttfOSvDNdZdH1VXVGdf25jkiycpZjuNfI/e+N7Gd9f3On5uAjfqq1dteRn3f3j2/pn2u7kaq40eqkX2+tHZquSuNumV5ptdM++vvb+p+01i5vrb24tXZUutfonun6xs3nXklunGf5tTOO6a6ttXUjy783cnt9drw+98z093RddlTspaY3+L/P1Pr9ez9l5vsxte3rRrY9PV11yYpMPw++PWOb1/TTtNb259Ch2XEOfSzJsX0l45OSrG2tfXGe12Q2087Bvnrvon46181JfnLk+e6Zucd6ZLrw6pKR8/2T/eO72nY2H0j3JT/pKq4+2p/LCzn/d8fMz3Zaa7N93nd1fLN5X7rKvGRGVd8u3te53DPJTTPO4+2vQz+F7i3VTeW+JV1QlAXsd3T/u/rdONo7b/Rzs10/vucleWWS66rqE1X1I/3i+yb5PyOv4Y3pgqIFvX9Vdfeq+mA/ffCWdCHqzONb0Gc7uz4HM8v6KzP37/D5jm3m75WWuX//3ztdULkQ9+3HdN3I874rXeVgZj5vFnbMAEyIO2Q41lq7MDP+kV5dX4RPVtdL4jMj/zB4WZI/ba3d1G97/SIPF2B/9S/p/hf9p+ZZ59p0Xyim3Kd/bE/de+pGP3XqqCTX9r1m3p3kVemuVHjXJF/PSJ+ZdP+DP6fW2jtaaw9LV7HwgCSvTXJDumBq5jF8dy+OYS7fSVchMeqYdIHXTs/XWvtakj9MN91q6jjn2sfMQGlqH99I9x9O8zXbT5JnpZu2tK9dl+nv6Z3STbmbGt9og//vpDt37l3Te17N+n601t48su0r01USbRl9vn7bqed+bJLfSVdtfrf+HFqb/hxqrW1I8jfppv+9MN1UvrnMda5tf7y6/lJ/m+6CCnfvn++c7Dhnp702o2NNd17enuS4kTDy0NZdSGFX287mH5McWVUnpgvJpqZU7s75PxUi3WnksR/cxfPOZVfHN5uzkxxVVU9IN9Xufcmu39d5XJfkbv20xSmjr+Pz003zfmK6sO3o/vGp/c77+yb78Hdja+0fWmtPSjcd8RvpfhcmXVDzihmh9UGttc8vcNdvTncc/6W1dpd05/5Cp4zv7jmYWdbfnO5cmDL6ms53bDN/r9SMfWfGfn5ojmUz38Or0/2dd8TIc96l/8/3zHzeLOyYAZgQd8hwbA5nJPm1/ovNa9JdLS3pvuA8oLomsBf1PRAA2EuttbXppsX9aXWN9O9UVSv7api39qudmeT1VXVk3wfmDemqD/bUw6rq2X212m+m+6JyUbqpRS1dAJLqmlfvKvDZrqpOrqpH9JVa69L1pdrWV7X9TZI/qqpD+hDu1Xt5DHP5ZJIfqa5B/crqmlG/Ocnftrmvvvi+dFN9ntHf/9skT62qJ/eVLfdM13fng/1x/kh1Tb6P6u/fO10YctHMHffbH1NVp6Wb5vWmfXakO5yV5GlVdUpVHZDuAg/z/dvkC+mqU36nf40en+Tp6Y9vPv17eXa6izHcqe93NtpD6pB04dnqJCuq6g3peoGNen+6KVnPyPzh2PeTHF3zX5HygCQH9s+3pboLOTx5ZPnfJHlxVR3bh4a/P3Is29IFIG+rqh9Ikqq6V1U9ZVfbzqavsPpwkv+Vri/XP/aPL/j8b62tThea/Xx/7vxSZlzwYKEWcHyzbbMu3fn0l0m+3VqbusjAQt7X2fb37SQXJ3lTVR1QVaekO9emHJLu98+adIHgm2fs4vuZO3RJ9tHvxr6665l9iLcx3bTtqSD89CS/VzsuFHBoVT13N8Z4SL+/tVV1r3T/YbBQu3UO9n5+ZP0/SHLWSGXxTPMd2yeSHDfyd8WvZ+6g9u+T3KOqfrO6iyQcUlWP6JdN+xy31q5L1+/wj6vqLlW1rP/P+R8dOeZfr6qjqusP97sLOGYAJsR+EY5V17z10ekarF6arsR5qgnzinT9Mx6f7gvAu2vkSjsA7LnW2h+n+7L8+nRfPq9OV7310X6VP0z3BfOrSb6Wrmn1H+7FU34s3RSim9JV7zy7dVfIvCzJH6erZvt+un5Gn9uN/d4l3Zfxm9JNhVmTLihIuqbv69I18/9suqqav9iLY/h4TZ8u+JFke2XzT6Tr6XR9usq3m5P88lw7aq1tSvJ/kvx//f1/S/d33f9IV2H9L+kCpalg69Z0TdG/UFXr0oViX0/y2yO7fVRV3ZbklnR9gO6S5OS+Um0u95xxTLdV1XN29UL04/3VdK/pdele/2t2cbxPT/c63ZDuP8Je1FfALcSr0k37+l66irm/HFn2D+kCyv9Idw5syIypWK21z6ULHr7chydz+XD/55qq+vIcx3Jrui/tf5PuuJ+frv/b1PJz0013PS/dRRfOm7GL/9o/flE/5e3TSR64wG1n84F0VVAfnhHG7s75/7J0AcqaJMclWWiF0mzmPL55vC9dNdb7Rx7b5fs6j+en+7zcmC7cGd3v+/v9fTddX9uZAfN70k3DvbmqPpqd7avfjcvS/Q6+th/nj6b/ndFa+0iS/5nkg/1r+PV0n50pb0zyvn6MoxebmPKmJA9NV2n3iXTh8oLs4Tn4/9J9Lr+X7kItvz7P/uc8ttbaDUmem+6iDmvSfQ+Y9e+D/nP4pHS/V76X5JtJntAvnu1z/KJ0wfZl6T63Z2XHd453pzvf/jXd+7ng1wuApVfdNPw7nqo6Osnft9YeXF1fmH9vrd1jlvVOT/KF1tpf9vf/Kcnvtta+tJjjBWDvVNUb0zVz/vmlHgvDVVXnpbu4z58v9Vhgf1FVF6Rrfu9zBcCS2C8qx1prtyS5cqqcujpTl1z+aLqqsalLOz8g3f9+AgAsWFWdnK6S5kNLPRYAAPadO2Q4VlVnppsq8sCquqaqXpLkBUleUlX/muTf0jVJTbry5jVVdVmS85O8trW2Zrb9AgDMpqrel25q32/2U7EAANhP3GGnVQIAAADA3rpDVo4BAAAAwL6wYqkHsLuOOOKIdvTRRy/1MAAAAAC4g7jkkktuaK0dOduyO1w4dvTRR+fiiy9e6mEAAAAAcAdRVd+ea5lplQAAAAAMlnAMAAAAgMESjgEAAAAwWHe4nmMAAAAA+5vNmzfnmmuuyYYNG5Z6KHdoq1atylFHHZWVK1cueBvhGAAAAMASu+aaa3LIIYfk6KOPTlUt9XDukFprWbNmTa655pocc8wxC97OtEoAAACAJbZhw4YcfvjhgrG9UFU5/PDDd7v6TjgGAAAAMAEEY3tvT15D4RgAAAAAgyUcAwAAACBJ8tGPfjRVlW984xvzrvf2t78969ev3+Pnee9735tXvepVe7z9viQcAwAAACBJcuaZZ+aUU07JmWeeOe96exuOTRLhGAAAAAC57bbb8tnPfjbvec978sEPfjBJsnXr1rzmNa/Jgx/84Bx//PE57bTT8o53vCPXXnttnvCEJ+QJT3hCkuTggw/evp+zzjorL37xi5MkH//4x/OIRzwiD3nIQ/LEJz4x3//+9xf9uHZlxVIPAAAAAIAdjv7dT4xlv1e95anzLv/Yxz6WU089NQ94wANy+OGH55JLLskXv/jFXHXVVbn00kuzYsWK3HjjjTnssMPyJ3/yJzn//PNzxBFHzLvPU045JRdddFGqKn/+53+et771rfnjP/7jfXlYe004BgAAAEDOPPPM/MZv/EaS5Gd/9mdz5pln5sorr8wrX/nKrFjRRUiHHXbYbu3zmmuuyfOe97xcd9112bRpU4455ph9Pu69JRwDAAAAmCC7qvAahxtvvDHnnXdevva1r6WqsnXr1lRVTj755AVtX1Xbb2/YsGH77V/7tV/Lq1/96jzjGc/IBRdckDe+8Y37euh7Tc8xAAAAgIE766yz8sIXvjDf/va3c9VVV+Xqq6/OMccckxNOOCHvete7smXLliRdiJYkhxxySG699dbt29/97nfP5Zdfnm3btuUjH/nI9sfXrl2be93rXkmS973vfYt4RAsnHAMAAAAYuDPPPDPPetazpj32nOc8J9ddd13uc5/75Pjjj88JJ5yQD3zgA0mSl7/85Tn11FO3N+R/y1vekqc97Wl59KMfnXvc4x7b9/HGN74xz33uc/Owhz1sl/3Jlkq11saz46pVSS5McmC66ZtntdZ+f8Y6ByZ5f5KHJVmT5Hmttavm2+9JJ53ULr744rGMGQAAAGApXH755XnQgx601MPYL8z2WlbVJa21k2Zbf5yVYxuT/Fhr7YQkJyY5taoeOWOdlyS5qbV2vyRvS/I/xzgeAAAAAJhmbOFY69zW313Z/8wsU3tmkqkJp2cl+fEa7eAGAAAAAGM01p5jVbW8qi5Ncn2Sf2ytfWHGKvdKcnWStNa2JFmb5PBZ9vPyqrq4qi5evXr1OIcMAAAAwICMNRxrrW1trZ2Y5KgkD6+qB+/hfs5orZ3UWjvpyCOP3LeDBAAAAGCwFuVqla21m5Ocn+TUGYu+m+TeSVJVK5Icmq4xPwAAAAO2eeu2pR4CMBBjC8eq6siqumt/+6AkT0ryjRmr/V2SX+hv/3SS89q4Lp8JAADAHcIXr7wx9/9v5+ZPz79iqYcCDMA4K8fukeT8qvpqki+l6zn291X1B1X1jH6d9yQ5vKquSPLqJL87xvEAAABwB/C//uEb/Z//vsQjgWFZvnx5TjzxxDz4wQ/Oc5/73Kxfv36P9/XiF784Z511VpLkpS99aS677LI5173gggvy+c9/fref4+ijj84NN9ywx2OcsmKv9zCH1tpXkzxklsffMHJ7Q5LnjmsMAAAAACzMQQcdlEsvvTRJ8oIXvCCnn356Xv3qV29fvmXLlqxYsftR0p//+Z/Pu/yCCy7IwQcfnEc/+tG7ve99YVF6jgEAAABwx/HYxz42V1xxRS644II89rGPzTOe8Ywce+yx2bp1a1772tfm5JNPzvHHH593vetdSZLWWl71qlflgQ98YJ74xCfm+uuv376vxz/+8bn44ouTJJ/85Cfz0Ic+NCeccEJ+/Md/PFdddVVOP/30vO1tb8uJJ56Yz3zmM1m9enWe85zn5OSTT87JJ5+cz33uc0mSNWvW5MlPfnKOO+64vPSlL82+6sw1tsoxAAAAAPbAGw8d037XLmi1LVu25Nxzz82pp3bXVfzyl7+cr3/96znmmGNyxhln5NBDD82XvvSlbNy4MY95zGPy5Cc/OV/5ylfy7//+77nsssvy/e9/P8cee2x+6Zd+adp+V69enZe97GW58MILc8wxx+TGG2/MYYcdlle+8pU5+OCD85rXvCZJ8vznPz+/9Vu/lVNOOSXf+c538pSnPCWXX3553vSmN+WUU07JG97whnziE5/Ie97znn3ysgjHAAAAAMjtt9+eE088MUlXOfaSl7wkn//85/Pwhz88xxxzTJLkU5/6VL761a9u7ye2du3afPOb38yFF16Yn/u5n8vy5ctzz3veMz/2Yz+20/4vuuiiPO5xj9u+r8MOO2zWcXz605+e1qPslltuyW233ZYLL7wwZ599dpLkqU99au52t7vtk+MWjgEAAABMkgVWeO1roz3HRt35znfefru1ltNOOy1PecpTpq1zzjnn7LNxbNu2LRdddFFWrVq1z/Y5Hz3HAAAAAFiQpzzlKfmzP/uzbN68OUnyH//xH1m3bl0e97jH5UMf+lC2bt2a6667Lueff/5O2z7ykY/MhRdemCuvvDJJcuONNyZJDjnkkNx6663b13vyk5+c0047bfv9qcDucY97XD7wgQ8kSc4999zcdNNN++SYhGMAAAAALMhLX/rSHHvssXnoQx+aBz/4wXnFK16RLVu25FnPelbuf//759hjj82LXvSiPOpRj9pp2yOPPDJnnHFGnv3sZ+eEE07I8573vCTJ05/+9HzkIx/Z3pD/He94Ry6++OIcf/zxOfbYY3P66acnSX7/938/F154YY477ricffbZuc997rNPjqn2VWf/xXLSSSe1qSscAAAAsP957umfz5eu6ipCrnrLU5d4NLA4Lr/88jzoQQ9a6mHsF2Z7LavqktbaSbOtr3IMAAAAgMESjgEAAAAwWMIxAAAAgAlwR2t9NYn25DUUjgEAAAAssVWrVmXNmjUCsr3QWsuaNWuyatWq3dpuxZjGAwAAAMACHXXUUbnmmmuyevXqpR7KHdqqVaty1FFH7dY2wjEAAACAJbZy5cocc8wxSz2MQTKtEgAAgIlSqaUeAjAgwjEAAAAABks4BgAAAMBgCccAAAAAGCzhGAAAAACDJRwDAAAAYLCEYwAAAAAMlnAMAACAidLSlnoIwIAIxwAAAAAYLOEYAAAAE6VSSz0EYECEYwAAAAAMlnAMAAAAgMESjgEAAAAwWMIxAAAAAAZLOAYAAADAYAnHAAAAABgs4RgAAAAAgyUcAwAAAGCwhGMAAAAADJZwDAAAAIDBEo4BAAAAMFjCMQAAAAAGSzgGAAAAwGAJxwAAAAAYLOEYAAAAAIMlHAMAAGCy1FIPABgS4RgAAACTpS31AIAhEY4BAAAAMFjCMQAAAAAGSzgGAAAAwGAJxwAAAAAYLOEYAAAAAIMlHAMAAGCy1FIPABgS4RgAAAAAgyUcAwAAAGCwhGMAAAAADJZwDAAAAIDBEo4BAAAAMFjCMQAAAAAGSzgGAAAAwGAJxwAAAAAYLOEYAAAAAIMlHAMAAABgsIRjAAAAAAyWcAwAAACAwRKOAQAAADBYwjEAAAAABks4BgAAAMBgCccAAAAAGCzhGAAAAACDJRwDAAAAYLCEYwAAAAAMlnAMAAAAgMESjgEAAAAwWMIxAAAAAAZLOAYAAMBEqaUeADAowjEAAAAABks4BgAAAMBgCccAAAAAGCzhGAAAAACDJRwDAAAAYLDGFo5V1b2r6vyquqyq/q2qfmOWdR5fVWur6tL+5w3jGg8AAAAAzLRijPvekuS3W2tfrqpDklxSVf/YWrtsxnqfaa09bYzjAAAAAIBZja1yrLV2XWvty/3tW5NcnuRe43o+AAAAANhdi9JzrKqOTvKQJF+YZfGjqupfq+rcqjpuju1fXlUXV9XFq1evHuNIAQAAABiSsYdjVXVwkr9N8puttVtmLP5ykvu21k5IclqSj862j9baGa21k1prJx155JHjHTAAAAAAgzHWcKyqVqYLxv66tXb2zOWttVtaa7f1t89JsrKqjhjnmAAAAABgyjivVllJ3pPk8tban8yxzg/266WqHt6PZ824xgQAAAAAo8Z5tcrHJHlhkq9V1aX9Y69Lcp8kaa2dnuSnk/xyVW1JcnuSn22ttTGOCQAAAAC2G1s41lr7bJLaxTrvTPLOcY0BAAAAAOazKFerBAAAAIBJJBwDAAAAYLCEYwAAAAAMlnAMAAAAgMESjgEAAAAwWMIxAAAAAAZLOAYAAADAYAnHAAAAABgs4RgAAAATpWqpRwAMiXAMAAAAgMESjgEAAAAwWMIxAAAAAAZLOAYAAADAYAnHAAAAABgs4RgAAAAAgyUcAwAAAGCwhGMAAAAADJZwDAAAAIDBEo4BAAAAMFjCMQAAAAAGSzgGAAAAwGAJxwAAAAAYLOEYAFYUG2gAACAASURBVAAAAIMlHAMAAABgsIRjAAAAAAyWcAwAAACAwRKOAQAAADBYwjEAAAAABks4BgAAAMBgCccAAAAAGCzhGAAAAACDJRwDAABgolRqqYcADIhwDAAAAIDBEo4BAAAAMFjCMQAAACZKS1vqIQADIhwDAAAAYLCEYwAAAAAMlnAMAAAAgMESjgEAADBRKrXUQwAGRDgGAAAAwGAJxwAAAAAYLOEYAAAAAIMlHAMAAABgsIRjAAAAAAyWcAwAAACAwRKOAQAAADBYwjEAAAAABks4BgAAAMBgCccAAAAAGCzhGAAAAACDJRwDAAAAYLCEYwAAAAAMlnAMAAAAgMESjgEAAAAwWMIxAAAAJkrVUo8AGBLhGAAAABOltaUeATAkwjEAAAAABks4BgAAAMBgCccAAAAAGCzhGAAAAACDJRwDAAAAYLCEYwAAAEyUqqUeATAkwjEAAAAABks4BgAAAMBgCccAAAAAGCzhGAAAAACDJRwDAAAAYLCEYwAAAAAMlnAMAAAAgMESjgEAAAAwWMIxAAAAAAZLOAYAAADAYAnHAAAAABgs4RgAAAAAgzW2cKyq7l1V51fVZVX1b1X1G7OsU1X1jqq6oqq+WlUPHdd4AAAAAGCmFWPc95Ykv91a+3JVHZLkkqr6x9baZSPr/ESS+/c/j0jyZ/2fAAAAADB2Y6sca61d11r7cn/71iSXJ7nXjNWemeT9rXNRkrtW1T3GNSYAAAAAGLUoPceq6ugkD0nyhRmL7pXk6pH712TnAC1V9fKquriqLl69evW4hgkAAADAwIw9HKuqg5P8bZLfbK3dsif7aK2d0Vo7qbV20pFHHrlvBwgAAADAYI01HKuqlemCsb9urZ09yyrfTXLvkftH9Y8BAAAAwNiN82qVleQ9SS5vrf3JHKv9XZIX9VetfGSSta2168Y1JgAAAAAYNc6rVT4myQuTfK2qLu0fe12S+yRJa+30JOck+ckkVyRZn+QXxzgeAAAAAJhmbOFYa+2zSWoX67QkvzquMQAAAADAfBblapUAAAAAMImEYwAAAEyUmncOEsC+JRwDAAAAYLCEYwAAAAAMlnAMAAAAgMESjgEAAAAwWMIxAAAAAAZLOAYAAADAYAnHAAAAABgs4RgAAAAAgyUcAwAAAGCwhGMAAAAADJZwDAAAAIDBEo4BAAAAMFjCMQAAAAAGSzgGAAAAwGAJxwAAAAAYLOEYAAAAAIMlHAMAAABgsIRjAAAAAAyWcAwAAACAwRKOAQAAADBYwjEAAAAABks4BgAAAMBgCccAAACYKJVa6iEAAyIcAwAAAGCwhGMAAAAADJZwDAAAAIDBEo4BAAAAMFjCMQAAACZKS1vqIQADIhwDAAAAYLCEYwAAAAAMlnAMAAAAgMESjgEAADBRKrXUQwAGRDgGAAAAwGAJxwAAAAAYLOEYAAAAAIMlHAMAAABgsIRjAAAATJSWtuN2a/OsCbD3hGMAAAAADJZwDAAAgImlcAwYN+EYAAAAE0UgBiwm4RgAAAATS04GjJtwDAAAgImicgxYTMIxAAAAJparVQLjJhwDAABgojSTKYFFJBwDAABgYonJgHETjgEAAAAwWMIxAAAAJspomzEtx4BxE44BAAAAMFjCMQAAACaKYjFgMQnHAAAAmFiuXAmMm3AMAACAySIPAxaRcAwAAICJpSE/MG7CMQAAACaKqZTAYhKOAQAAADBYwjEAAAAABks4BgAAwEQZ7TOm5xgwbsIxAAAAAAZLOAYAAMBEadNuKx0Dxks4BgAAAMBgCccAAACYKG2k0ZieY8C4CccAAAAAGCzhGAAAABNL4RgwbgsKx6rqvlX1xP72QVV1yHiHBQAAwFAJxIDFtMtwrKpeluSsJO/qHzoqyUfHOSgAAABIpvcfAxiHhVSO/WqSxyS5JUlaa99M8gPjHBQAAADDJQ8DFtNCwrGNrbVNU3eqakVUuQIAALAIfPkExm0h4dg/V9XrkhxUVU9K8uEkHx/vsAAAABgqgRiwmBYSjv1uktVJvpbkFUnOSfL6cQ4KAAAAABbDil2t0FrbluTd/Q8AAAAsGv3HgHFbyNUqn1ZVX6mqG6vqlqq6tapuWYzBAQAAMEASMWAR7bJyLMnbkzw7ydeaa+gCAACwmHwLBcZsIT3Hrk7ydcEYAAAAi8GXT2AxLaRy7HeSnFNV/5xk49SDrbU/mW+jqvqLJE9Lcn1r7cGzLH98ko8lubJ/6OzW2h8scNwAAAAMQBOVAWO2kHDsj5LclmRVkgN2Y9/vTfLOJO+fZ53PtNaethv7BAAAYD9n3hKwmBYSjt1ztsqvXWmtXVhVR+/2iAAAAKAnKAPGbSE9x86pqieP6fkfVVX/WlXnVtVxc61UVS+vqour6uLVq1ePaSgAAABMAlMpgcW0kHDsl5N8sqpur6pbqurWqrplHzz3l5Pct7V2QpLTknx0rhVba2e01k5qrZ105JFH7oOnBgAA4I5ATAaM2y7DsdbaIa21Za21g1prd+nv32Vvn7i1dktr7bb+9jlJVlbVEXu7XwAAAABYqF32HKuqx832eGvtwr154qr6wSTfb621qnp4uqBuzd7sEwAAgDu+0T5jTdMxYMwW0pD/tSO3VyV5eJJLkvzYfBtV1ZlJHp/kiKq6JsnvJ1mZJK2105P8dJJfrqotSW5P8rPNbz0AAAAAFtEuw7HW2tNH71fVvZO8fQHb/dwulr8zyTt3tR8AAACGZVrl2NINAxiIhTTkn+maJA/a1wMBAAAAgMW2kJ5jp2VHWL8syYnprjQJAAAA+9xotZjmO8C4LaTn2MUjt7ckObO19rkxjQcAAAAAFs1Ceo69bzEGAgAAADM1XceAMZszHKuqr2X23oeVpLXWjh/bqAAAABisZi4lsIjmqxx72qKNAgAAAGYjJwPGbM5wrLX27anbVXX3JCf3d7/YWrt+3AMDAAAAgHFbtqsVqupnknwxyXOT/EySL1TVT497YAAAAAAwbgu5WuV/S3LyVLVYVR2Z5NNJzhrnwAAAABim0ZZjZlUC47bLyrEky2ZMo1yzwO0AAAAAJsrqWzfmnK9dl63bRK90FlI59smq+ockZ/b3n5fknPENCQAAADouXMm+9sx3fjbXrt2QNz3juPzCo49e6uEwAXZZAdZae22SdyU5vv85o7X2X8c9MAAAAIapmUzJGF27dkOS5PPfumGJR8Kk2GXlWFW9OsmHWmtnL8J4AAAAYDtBGeOykKrEa2++PQevWpG7rFo5/gGxZBbSO+yQJJ+qqs9U1auq6u7jHhQAAADDZSoli2FXp9kNt23Mo99yXh715n9alPGwdBYyrfJNrbXjkvxqknsk+eeq+vTYRwYAAMDgCcoYl12dW1+7Zm2SZN2mrYswGpbS7lx18vok30t3tcofGM9wAAAAGDp5GItj/jPtpvWbFmkcLLVdhmNV9StVdUGSf0pyeJKXtdaOH/fAAAAAQFDGuOyqcuym9ZsXZyAsuV025E9y7yS/2Vq7dNyDAQAAgGYuJRPgZpVjg7HLcKy19nuLMRAAAACYSVDGUrlxnXBsKHan5xgAAADAfmG+2PXr312bv/7CdxZtLCwt4RgAAAATZTS0UDjGuMxXlfis//u5RRwJS004BgAAAAzOfLnr5q1S2SERjgEAADBZ5BIsAlWJTBGOAQAAAIMzXzZ278MOWrRxsPSEYwAAAEwUPcdYDFu2bssbPvb1nP+N63dadrc7HbD99sEHrljMYbEEhGMAAADA4Hz+W2vy/n/5dn7xvV/aadnWbTtS2eXLajGHxRIQjgEAAACMGA3H5ruqJfsH4RgAAAATZTSMaLrzswSmhWNj2P/1t2zIr37gy/nKd24aw97ZXcIxAAAAgBGj4dg40rHXf/Tr+cRXr8uz/u/n9/3O2W3CMQAAACaKhvzDsXb95nx7zbqlHsZOtrbxVo59/9aNY9gre8olFwAAAIAlccIffCpJ8sXX/Xh+4C6rlng0O2zZOt6eY3r8TxaVYwAAAEyUNt4ZbUygb62erOqxbWMuWZSNTRbhGAAAALCklk9YKdW4G/Ivq8k63qETjgEAADBRRq9QOY4pbUyeiQ7HxnAKCscmi3AMAAAAWFIrJi0cm9aQfwzp2GQd7uAJxwAAAJhY6saGYeIqx7aOu3Js3++TPSccAwAAYKKYSTk8ExeONT3HhkQ4BgAAwMQSlO2/Rvt6TVpYtGXbeC+ZOmGHO3jCMQAAACaKQGwYNm/dtv32WPp6zWKuCzzMfHzbtvH2HJu0MHDohGMAAABMMEnZ/mq0OmvbtnlW3IdGq9Xme3zLmK9WWcKxiSIcAwAAABbdliWoHNs6R9I1Paibvk5L8tVrbs6bz7k8GzZv3SfjEI1NlhVLPQAAAACYiymW+6/NY74i5GzmqlAbDcdmBmittTzjnZ9L0l044L+e+iN7PY4Ju/7A4KkcAwAAABbdlpGkarHCsbkqx7aOBHVTUyxXLu8SrNEtLvn2Tdtvb9i8NdfefPsejUPPsckiHAMAAGCijDZHVzi2/9q8ZWQq4yKlY3P1HNs8EtRNrbO8L+8aHdrqWzduv/200z6bR7/lvFx1w7rdHodsbLIIxwAAAIBFNxpILVYIOrOf2JTR0GxqiuXKZTtHJtffsmH77Suuvy1JcuE3V+/2ODTknyzCMQAAACaKarFh2DKt59jivOtb5gjHZmvIv3z5jgBraorluk07N+S/5fbNuz0OPccmi3AMAACAiaUh//5r88jVKufIrPa5uaZvjl45cyooWz5S3XXkwQduvz0zyLtlw5bdHke5XuVEEY4BAAAwUQRiwzAaji1WveBcPcemVY71J+CykfKuA1bsiE/WrNs0bdu16/egckwaM1G8HQAAAEysZpLlfmt6ILU4zzlnOLZ1555jK5bV9sb5G7fsCPJuuG3jtG3X7sG0Sj3HJotwDAAAAFh0o5Vji1UtOOe0ypGLA0z1HFtWOyY/joZjN86oHLtlwx6EY7u9BeMkHAMAAGCijFaLmWK5/xqt1portNrX5qocG3186vaK5bW9wmvj5h2N+G9aNz0M25NwbJnKsYkiHAMAAAAW3Wi11lJXjm2eZVrl8rkqx9bPqBy7ffcb8rta5WQRjgEAADBRRvMLlWP7r01bRisEF6tybK7Hd65iWz7Sc2y0P9rN+2BapcqxySIcAwAAABbdtMqxRXrOuRvybxu5PRKOzdId7Mb1m7b3JUuSW27fvNvhnob8k0U4BgAAwERp024rHdtfTVLPsS1zVI7N1jn/pnWbdrrS5vpNW3decR6ysckiHAMAAAAW3VJcrXLrHE80Gppt7zm2bLa6seSm9ZunVb0lybqNu9d3TM+xySIcAwAAYKLoOTYMs1VrjdtclWOjQd3UOsuqZq3wumn9pmkN/JPk1t0Mx0Ztm2NMLB7hGAAAALDoplWOLdJzzhXCzdaQf8VcPcfWbZrWoyzZ/cqx0Txsi3BsyQnHAAAAgEU3Wn21eFer3HXPsaleaMuWTa8cO+zOByRJbl6/eadA67YNuxuO7dh+rjGxeIRjAAAATBhhwRBsWYKeY3NNYRztITa9cmyHw+58QKqS2zZuycbN0yvHbtvNyrHR452rDxqLRzgGAADAxJIb7L9mXvFxMcwVRI1eOXNaQ/6R0rFVK5fl4ANXJOn6jo3a3XBsWuXYVif5UhOOAQAAMFEEYsOwacto5djSTqt8y7nfyNr1m5PsqC5bPuOSkiuXL8tdVq1Mkty41+HYjtszr3zJ4hOOAQAAMLGaKZb7rdFQaFzv8rU3357/9Q/fyOpbNyaZuyH/mnWb8rPvvqgfVx+O1fRplQcsX5ZDVnWVYzfetg8rx6TBS27FUg8AAAAARokKhmHLIjTkf8n7Ls7l192Sf716bf7qpY/I1nmKtC6/7pZcc9P67dVly5dVRtOxA1aMVI6tmxGO7WZD/qYh/0QRjgEAADCxFNXsv6ZfrXI8z3H5dbckSb7ynZuSJFt3MYXx2ps3TAvHRivHDlyxLAeu6CbgzZxWuW53K8dGhrFFz7ElJxwDAAAAFt30K0SO97mm9j9f5ViSrNu0Zfs0x2UzGvIfsGJZVq1YnmTnaZW37sW0yrmmerJ49BwDAABgoixWc3aW1uatoz3HxvueTwVQu+rvtX7j1u0N+Vcsq9TotMrRnmMzG/Lv5rTK6Q35ne9LTeUYAAAAE0tssP8anVY57nxoKhPbtosnWr9py/ZqsZ0a8q9YlrscNL3nWFW373Wb9By7I1M5BgAAwEQRFQzDltHKsTFXC26vHNtlOLajcmz5LNMqpyrHburDsbvd6YAke1I5JhybJMIxAAAAJpYplvuvzdvG35B/+/77P3c5rXLT1u3THGc25D9g+fLtV6tc04djd+nDsvWbtu7WeEbzMOHY0hOOAQAAMFHkYcOwZQl6ji1kWuVUgLZ8Zs+xkWmVa2/fnCTb7+9+OLZjHHqOLT3hGAAAABNLbLD/Gr1y5LZdXEVyb01lUbuqHFt968a89dxvJOnCsYzUjq1cXtunVU6Zur9h89zh2NZtLaf/87fy9e+u3Wk8U8tZWhryAwAAAItudMrsYsVDu6oc++CXrt5+e1lNrxxbVpWVy6fXGB1y4K4rx8665Oq8pQ/crnrLU7tx6Dk2UVSOTYgbbtuYq29cv9TDAAAAWHLTQhO5wX5rNCDaNqY3ellNvz/XFMZffvwP7/TYihk9x6qSGbvbXjl2++atc/bH+9bqdTs9Nn1a5ZjL5tilsYVjVfUXVXV9VX19juVVVe+oqiuq6qtV9dBxjeWO4KQ//HQe+9bzc9vG3bvCBQAAANwRTcupxhSCLqvpcdZsVVo//8j75CH3vutOj8/sOVaZfvXKpOtDduCKLlrZuGX2kGu20Gx0GLKxpTfOyrH3Jjl1nuU/keT+/c/Lk/zZGMcy0daNBGLfW7thCUcCAACw9No899h/LE7l2PQwa7bnqVTudMDOXaeWzyg7q9q5Em3l8mU56IDlSXavKX9TOTZRxhaOtdYuTHLjPKs8M8n7W+eiJHetqnuMazyT7MobdpRYTl3xAgAAAPZnoznV3kRjrbVsmqNqa+Y8yK2zrFaV3OnA5Ts9vnHLtoxOrKx+3VErllXutLLb9vY5mvLPlvtNqxwzd3jJLWXPsXsluXrk/jX9YzupqpdX1cVVdfHq1asXZXCL6ao1O8Kxm9ZtWsKRAAAATIDR0ERusN/ato96y73kfRfnAa8/Nzev3/n79MxKr62zVGlVkjsdsHM4dtO6TdMb8s+4emWSrFi+LKv6bW/fjcqxaT3HtjrJl9odoiF/a+2M1tpJrbWTjjzyyKUezj535Uhzvhtn+TADAADA/mZfTas87xvXJ0k+e8UNOy2bOa1ytqmPVZU7zzKtcs26TTs14J9ZObZyeeWglfOHY7Md2WjlmKtVLr2lDMe+m+TeI/eP6h8bnNFplSrHAACAoVuEPu1MgNEirn3xPs+Wry0kHEtmrxy7cd2maQ34u55jMyrHli3bvu3uTKsc7Tm2VXnkklvKcOzvkryov2rlI5Osba1dt4TjWTJf++7a7bdVjgEAADAE06dV7n1ANNseZlZ6rd+0ZdZ1ZmvIP1URtn29aR3IOiuWV1atnGrIv/O+5zJ67CrHlt7YwrGqOjPJvyR5YFVdU1UvqapXVtUr+1XOSfKfSa5I8u4kvzKusUyyNbdtzDevv237/ZvXacgPAAAwRVHN/mtf9Rybz8wwa91s0ypTWbVy2U5B2puf/V+m9xyr2RvyT4VoG+aqHJslthutmtNzbOntHI3uI621n9vF8pbkV8f1/HcUX7xy+gU9VY4BAABDty+qiJh8+/qKjbOdN6PTIltrWb9x9sqxqsryqmzp9/GCR9wn9/uBg6eFYVXZqXZsxfI9m1a5zbTKiXKHaMi/P/tCH4495n6HJ9FzDAAAYJSgbP/VFqFybNtIArd5a5ujcqyzbEZ/sW7ZyGOp2RvyHzA1rXLhV6tsGvJPFOHYEnvq8ffIK370h/LTDzsqicoxAAAAUcEw7PvKseT8f78+TzvtM7mib1+0eWT+4qat27b3BVu1ckccsj3wGq0S6+/sVDm207TKZdt7jk1drfLWDZunXblytoB39Hi3CMeWnHBsiZ189GH5vZ94UB71Q0ckSW7bsPAGfgAAAHBHtS8CsZl+8S+/lK9/95a86eP/lmR6P69NW7Zl/cYutDr4wB1dpqamXo7mXjsqxzJtvZ2mVS6rHdMqN23NbRu35NFvOS9Pf+dnt68z21GOHvs24diSE45NiGX958tnAgAAGLrRzMRXpP3XPq8cm3G2tNamVWVt2rIt6/rKsWnh2NSf0yrHph6raY/NrBxbNtKQ//bNW3P1jetz64YtueL623L1jeuTTJ82OXXbtMrJIhybFNs/YD4UAAAA7P9GK6b2RRHZ6D4Ou/MB2TzjKpCjlWN3HgnHpr6PT+85Nns12bIZ6VglOeiAbl+3b96azVt3TOP81GXfTzK9em1q+bTKMX31lpxwbEJMfcB8JgAAAHbwHWn/NT0g2rf7vtudDsiWkX5jSbJp69btlWN3nlY5Nve0ysyoJptZOVaVHZVjm7Zmw+Ydz/mF/1yTJNMCs6lKtn1dNcfeEY5NiKnPlw8FAAAwdDOnx7F/Gv3+uy/e89Gv03e7086VY7ds2JJtLTlgxbKsXD7LlSmnTaGckYKlm0I589FlVVnR72vz1pbbN+9oxL/6to1JugsBTNkya+XYAg6OsRKOTYipD6HPBAAAwA6Csv3XtN5y++Bt3rBlRzB15wOXbw+ipty8flO37IDl08KvWXuO/f/snXeYHFeZ7t+q7p6oZCUHOWFjTLTJ4AuXaxN2YdfAEgxewrK7sCxrMHDZy+7CkoyxDYuxsTFGzjgHwLZsy0FWsGRljXIYSTMazYwm59Q9narq/lF9qk6qDjM9062Z7/c8emZUXV1d3dMVznve7/10gfzQO8fYIgeO0KWyd9QVx3jnGBPsKHOsvCBxrExggfxkHCMIgiAIgiAIYrZD46LZgeAcK8IffWAsKfz/jg1Nwv8HoykAQE1FOKcQZmrcZDAM8f/eumxADyQ4ga5vLAHHcQQHGyv1pG6V5UU49yrEdMBUayqrJAiCIAiCIAiC4KAh0ozFLrJzrD/qi2P72obxzN4O4fG6lkEArqtMV0Jp5Arkl/4PuMKYYbAcMdE5Fk/ZiCbFkP59bcM4bV4VlVWWGeQcKxd8HyZBEARBEARBEMSshoZFs4NiC0S8OBbLBO/zPLq9FYDbtVIXvm/yJjHpMfdx1TlmGP46DiBkjgFA32hCEMf+9cGdeGZvh/B+LTLJlBwSx8oEdhCSc4wgCIIgCIIgCMKHRkgzFyFzrAh/6f5MAD4AVFcEF8r1jCbEskr2U7NQyCYz9M4xVlbpOBC6VQJuaaXcGOC5fZ3Ce6eyytJD4liZQIH8BEEQBEEQBEEQxGyCD6Ivhj40Ek95v2fLMPvJ5W8URS5NCSUTvGS9zJSdY9zvtuOozrEx0TkGAFWRkLB/ZJIpPSSOlQkUyE8QBEEQBEEQBJGhyFlURHliF7ldZYJzbQV1gDxtXhWufPfZgsilc46psftiCSW/zOTMLnFJHOsdSyKZFsWxyrApvHcqqyw9JI6VCRTITxAEQRAEQRAEQcwmeP2qGM6xONcpMh2wQb8Lpb/MyLpMDennMQzDzxxzHE8cm1PplnUORpPKvlRFTPG9U1llySFxrEzgA/wIgiAIgiAIgiBmM3z+VDGyqIjyhC8tLMbfOZ6Hc8wXuDTdKvn1tMs0zjFuXceB162SiWO246hlleEQdassM0gcKxO8A4wOCoIgCIIgCIIgCGIWUGyBiC9plAUphs4lJj8WtJ7Bhe+Ly9zfHcfvVllTEQLgvq9UWpc55v8/SMgjpo/g9g3EtEJllQRBEARBEARBEC5FjqIiyhS7yH/nfDLHPNFLs8zUlFDKgfyyqGZyOWS24yCecl+3OiOOwXGQtHRllRTIX06Qc6xMoLJKgiAIgiAIgiAIYjbBi0LZukvmS9LKQxyDXvTif4rL/KWmKf7f344ayM+cY7Gkhb6xhPCcqohcVklKQKkhcaxMYAo1HRQEQRAEQRAEQcx2nIDfiZmF4BAs8rZzBfKLLjH2U7fMf64BQ5M5FlRW6Rbq3b3xuLIPbrdK//8BFaDENELiWJngRY7RmZ8gCIIgCIIgCIKYBfDurmJ3bAwSx7TlktoSSk0gv6GWVbrLWCC/ozjHguCdctStsvSQOFYm8AdYMeykBEEQBEEQBEEQBFHOCGWVRd62Zbt2rGULqvHXbzrVW+5njqmJ/Py43FQf1j7PMPwlDnznWHUkWByLJS2kLCqrLCdIHCsTePsmHRcEQRAEQRAEQcxmnCJnURHlCf+nLbZAlM6ITxeeNhcVYV+oMpRf9IKZN0TnxuqmoSurdLPIgIxzLJlxjlUGi2M3vHBY+L9F3/GSQ+JYGUGh/ARBEARBEARBEMRsQQzkL+62WcmmaRhSaaSmXDLzP1NXaik8F4rfzN0+yxAH4mnXscYyx/KByipLD4ljZYTJ1SkTBEEQBEEQBEHMViiQf3ZQ7G6VPGlPHBNFr3wD+eXHABa+r+lWyRldxpO5yyplSBsrPSSOlRHsEKMDgyAIgiAIgiAIgpjp8GPf4nerdB1cIdMQu1CyoH1B9BJ/Ar54lk8gP1vXth2uW2X+4hiVVZYeEsfKCL+skg4MgiAIgiAIgiBmLw5Zx2Y8slOs6GWVmcwx0zSkzpSZn9As060nZI5pAvm5JYm0K4xVhEyEpET/pXMrA/eVyipLD4ljZYTf/rXEO0IQBEEQBEEQBEEQU4isBxU9kD/zAiHDkIL2DeEnwLvJ+GXiT+9/WZxjsUxJZVXEVEo0s70/6lZZekgcKyO89q90XBAEQRAEQRAEQQCgypqZiiWpYxP9KwdllbHthyTnmJnNdpwOlQAAIABJREFUOaZbJrnJJEOY0MHSyxurCCnrpbO4wyw78CFimiBxrIzwAvnp5E8QBEEQBEEQBEHMYGS31EQD+YOexsQoWdAqJHyf/+n+rj6P72DpO8dCSldLVuapg5xjpYfEsTKCHWNUbkwQBEEQBEEQBOFCusHMRP67TvTvHPS0dMaOpZRV6gL5M//RdbCE4DqTE8dEwSyWTANwO1WaknXsrIU1ge+BxLHSQ+JYGeGXVdKBQRAEQRAEQRDE7ITGQ7MDWRCaqEAU9H1Jc2WVJqd8aMsqNc/3ssmEZbpulVxZZSrYOXbLlW8NfA9yiSkx/ZA4Vkb4ZZUEQRAEQRAEQRAEQM6xmYpaVjmx7QQ9jW3fdXCpdZW5OlMa0mPsdzNLWWUqUzpZFTEVFe2shTW46tLzs+4rUTpIHCsnMseOQ2F8BEEQBEEQBEHMUkgnmB2o3Sonuh39E5lQFTIMIXPMd46pQpg2m0xTksljAEoJZXVEDeTnt6m8B9IASg6JY2WEV1ZJ3jGCIAiCIAiCIAgAVFkzU5HLISc6Ds4lpppSKaTeEaa6ycwAh5kscOlyyKorQoqQZhqG4jpjWKQIlxwSx8oIpjZTuTFBEARBEARBEAQxk5HHvVOlD5mmGMjvi15q+L7gEtM6vwxF9NIJZlUa5xhffiljkwhQckgcKyMokJ8gCIIgCIIgiNmOPBqi8dHMRA6hn+jfOdfT5LJKP1+MW5b5aeZymEnruM8xBKENyATy57EegzLHSg+JY2UEBfITBEEQBEEQBEEQswFZDCt25hgjZBpS0L7ahVKvmBnC+myRTuBSyiojIWU9A3o3GgBYJAKUHBLHygh2oJBqTBAEQRAEQRDEbEXNoiJmIkpZ5QS3k+t5pmkEdKbklkk/AS64X8ohk/UtXQfL6khIu54u0B+gsspygMSxsoK1qyztXhAEQRAEQRAEQRDEVCKbQiZqEslVjmkaqvsL0C8Tyyr1DjF9uaS4rCpiKoKZIZV38pBBpvSQOFZGmKSNEQRBEARBEAQxy6Hx0OxAEYQm+IfP9bSQJF4x0UonhGlD+iWLmVIuqXGEyZljOhcaj5y/Rkw/JI6VEVRWSRAEQRAEQRAEIULDo5mJqo1N0DlmZ3/cNIMC+XVuMm496JYZwvpsmSx6VVeEBOcYey0K5C9fSBwrI9iBRscFQRAEQRAEQRCzFRoPzQ6UssocIlcQuUS1kNQlUi53BAI6U2ozx8T12TKlrDJcmHOMjGOlh8SxMoLKKgmCIAiCIAiCIGRohDQTUQP5J5o5lv1xOZCfoRfCNG4v4Tn68ku5rLK6QuxW6XfIDA7kP9gxjJ6RePY3Q0wZJI6VEezgoU4VBEEQBEEQBEEQxExGztma6DA4V0mi22FSJ3rpxCtwy8T13d/ZNiGsaUrKitytUv88n9aBGP721o149/Vrsr4XYuogcayMCLJYEgRBEARBEARBzBZkBxGVWc5M5C6TE/075wzkN0VRytQJVZqyR61zzPvJl2mqjrDKiJmzRJMnTQaZkkPiWBlBgfwEQRAEQRAEQRDEbEApq5zgODhnWaXUrVKbL8atm3U9TXK/Yahlm2HTFLbld8jM7YiZ6OdATA4Sx8oICuQnCIIgCIIgCGK2o3YxJGYisilkon/nnIH8piE5vXS5YWoXSr9UMndXS7lc0nWTievkSzRpFbA2USxIHCsjKJCfIAiCIAiCIAiCmA0o3SqL5ByThaqQaQjLggQu96cmhyyHw8z9XXxRI6BDppGHc2wwmsy5DlF8SBwrI7xAfrKOEQRBEARBEARBAKDKmpmK4hCcaOaY9LxwSJQ5DMMQFC59aaT4U/5dfq68nizISS/pvVZQID/P8Hgq90pE0SFxrIxgxwmd/AmCIAiCIAiCIIiZTNGcY1LtVURSoEKG5BzL/BTyxTxxTFWv+CWmqqtlhDDxeaZhaDPH8imvHIqROFYKSBwrI/xjh9QxgiAIgiAIgiAIIHemFHFyUqwGjbKmFpLFMVMql9SWVarilalziXmllqITTRa95Mwxr0OmtG9/+bdLlPczNE5llaWAxLEywi+rLPGOEARBEARBEARBlAiqpJkdFMs5Jj8vIpVVmpJzTCt6acoefRFN5zDj1oPahdI0DC7QnyvH5NZ587J5eO3Sucr7IedYaSBxrIzwAvnpYkAQBEEQBEEQBAGAxkczFVtyhRQvc0x2jhlaIUz1e0mOME0gv/+Yj2kYyjqGIW7fK8eUti+73ABgKEbOsVJA4lgZwQ4eCuQnCIIgCIIgCGK2QmWUswO5YqpY4+CwqTrH5DJI9yfUZeCXQVlP13VSJ56pgpn6mqYmyB8g51ipIHGsjDDIOUYQBEEQBEEQBCFAw6OZiSyGTZVzzJSdY9JPYZl2vTzKKiWVSw7p94P8RQubXI4JAEPUrbIkkDhWRrCDh2ZKCIIgCIIgCIKYrZBZYHbAxDHZJDI8nkJT71jB22GENd0qxRLHYPeXzmEm6VnCNth6aiC/oXWhidlnavMAAIgm0soyYuohcayMYIcFXQwIgiAIgiAIgiBcHBogzUjYnzUkmUQ+9JtX8MHfrMeJgVh+25H+Lwfyh8yAoH1dF0po1kPu9WQDmFsyqRPk+G2pQf4AkLJsZRkx9ZA4VkZQWSVBEARBEARBEAQxG2COL+aechw3pL9vzA2kb8zTPSaLp0pZpRFUVpm7XNJdpgvWF9eTRS4j6DUlwUyXOZa0SBAoBSSOlRGmpJgTBEEQBEEQBEHMNmg0NDtggfxMHLMdB+1D497jNZFQQdthaAP5tWWV6rbkcklA7yaDtD1tWWUegf+GRiBLpck5VgpIHCsj2MEjH9wEQZSe4VgKH7vlVdy36Xipd4UgCIIgCIIgTnoU5xiAhp5R7/Fk3uWFOTLHpEB+plDlCuTXucTYmnKZpiEpbaYR5ExTxTfZdUZllaWBxLEyws8cI3WMIMqN+zYfR33nCK559lCpd4UgCIIgCGJGQ+Oh2YGdcYWEPecYcLTbL6VMpPITifLrVqk6x/gOk36+mG4Z1GWSK0zNHDOk8H2NqMbtHw+JY6WBxLEygh2w5BwjiPJjPGWVehcIgiAIgiBmJaSVzUz8ssqMLOE4aO6Leo/n6xzLGcgvCVWaakpvqTakXxesL4loum3qu2H6j7NthSRljTLHSgOJY2WEf0zQwUAQ5UYqTcclQRAEQRDEdEB3XbMDv6yS/R+IcxPSiXR+k9O2HMgvObFMMx/3F/upqmi5yi91gfymGeQcUwUzJXOMnGMlgcSxMsIL5KerAUGUHXSRIgiCIAiCKA3UsGxmwspnWYC+AwdproxqomWVIU0gP1+6qHV/ST/l371lnqAlBfIrZZX6zpfyfgFUVlkukDhWRrBDgsoqCaL8SNt0kSIIgiAIgpgOyCwwO2DjXqZl2TZgcYPhfMoqd7YM4mO3vCosi0iZYyHTyCs3DIDW7QVB5FJlLsNQl7uCWa5Afn//eJLUrbIkkDhWRrCDgwIoCaL8SFJZJUEQs5ixRBoN3aO5VyQIgpgCaHg0M7EV5xiQsgpzjv3LA3XKsnBIdY5BEKWCw/G1gpa26yT/XAOGpKzIOWTa16RulWVFuNQ7QPiwg4PO/QRRftBFiiCI2cxf37wB7UPjeO7q9+PNy+aXencIgiCIk5iHt7WgeziO1546F4AvGDmOA4ur1sgnc2wsnlaWRSQnVsjUB/Ln7EypC+kPyCuT/WRut0q+9FJ9TbZMFsfIOVYayDlWRvhllSSPEUS5QWWVBEHMZtqHxgEArzb0lXhPCIKYFUjDIRoezSz++6kDuHVtIzoy1xbPOeZAyBzLRyTSlV6G5bJKw5BEqWDRCzohDPwyfbC+IQlchrx9BLvVJKOb4J4jpg8Sx8oI7+ChY4Egyg6awSEIgqBzIUEQBFE82DWFZW7ZjoM0X1Y5wWuOHMhvGHI5o/ocX7zKVUIp/mTPlTtOmpIgp9uWroMlQBUrpYLEsTKCHRQUyE8Q5UeSZnAIgiDyKnEhCIKYLHJ3SroLm5n4zik/XogP5J+oOKYN5M+RL8YWassqta4zsTxSH8jPbUtjTdN1vgRc95xNosC0Q+JYGeEF8tPpnyDKjjTN4BAEQdBsNkEQBDEpLI3oIzjHhMwx//frVh7Cdx7bnVfzurDkHHO7VaqB/KIQxh7jlgXkiynPlYQwADBM2SUm/uT3Q+5WCQApinSZdkgcKyOYYkw19QRRftCAkCCI2cJQLIk19d3aAQyVVRIEMR3I46F8BBHi5IC/p7Yyf1dPHJIyx3i38l2vHseKPR3oHI7nfA3ZOaa6uNhyfpmmrFJaP2iZKTbD9LajL9FUt6/Rxuh6WwJIHCtDKJCfIMoPCsYkCGK28JX7duCr99fh4W0tymO64GOCIAiCyBf+OsLyxfiyylyZY/mIRkogv2lIZZBqaaRW9PJUK035pSB8GZqyyqB8Mc0ynXOMxh7TzpSKY4ZhfNQwjCOGYTQahvFfmsf/0TCMXsMw9mT+fW0q96fcYQcYHQYEUX6Qc4wgiNnC3hNDAIDn93cqj000/4UgCGIy0Pho5pDiriPMJRbysrcdwbXMhDDeOZjPdUguqzTlQH72UxvSn0Mw06X5Q3V/uSH9mlJOzWuGNNukscf0E56qDRuGEQLwewAfAdAGYIdhGM84jnNIWvVxx3G+NVX7cTLhHVB09ieIsoMuUARBzDbGU4XP2CfTNgwDiMh96QmCIAqAhkMzF94RZWVytZjTy3HErC0mhPFus1gynfM1wqbs4jK0jjBd+L6uq6Uum0wu9TUkgcswgjLNVAebTnCjssrpZyrvXN4NoNFxnCbHcZIAHgPwySl8vZMedkhQWSVBlB9psjYTBDHLSKTUzpTZbtZTlo1LbliDz/5h81TuFkEQsxG6DZsx8NcRJpSZgc4xS1gPAGLJ3F2TwyE1kF+XJQbB2aUs8tbUd50UkZe6gpzY0VJZj2WfacsqSRybbqZSHFsG4AT3/7bMMpnPGIaxzzCMPxuGcZZuQ4ZhfN0wjDrDMOp6e3unYl/LAnbwPLS1BUOxZIn3hiAIHsrZIQhitjGuE8e4c+FPVhzAF+/e6g1kWgdi6I8msbdteNr2kSCImQkF8M9chMwx5hxjmWOOPnOML8WMJnznWNCEjRzIr3SD1AXyez81gpm21FJ1ivHImWN+WaXqHNOZrSlzbPoptef9WQDnOo5zEYCXAdyvW8lxnDsdx3mn4zjvXLJkybTu4HTCDs51R3rxzUd2lXZnCIIQoNkbgiBmG+Oa2Xk2ELFtBw9sacGmxn60DsQAAJ1DfgcxOmcSxMzj5UPd+MGT+0tyfDtkHZsxCN0qbSmQ33E8wQwAEpny/pRQVmlxv+tLLGUxTHVxsbJKVR3TBebL22L7yiOLZXK3SjOLIKd7HbqOTj9TKY61A+CdYGdmlnk4jtPvOE4i89+7AbxjCvfnJMA/KDY19pdwPwiCkKGySoIgZhtxjXOM3az3jSWUZScGY1mfmwvbnjnnWcdx0NwXJfcLMaP4lwfq8Oj2Vjy1qz33ypOEjpyZS0rTrdLLHAPEskpLzRyLcoLYWEIvjslllXkF8kN1dunWU+on+YekvDJdphkvyPmCmSZzjMSxaWcqxbEdAC4wDOM1hmFUALgSwDP8CoZhnM799xMA6qdwf8oeTakxQRBlAm/bpsEOQRCzgXiWQP4Tg+PestG4OzhhDrKg52Zj+/EBXHzNKjyzt2Miu1p23LqmEZfe+AqWr28q9a6cNDR0j+JTt2/ClmM0QVzuDFD8CzEJkppulXzmWNrmyyo1mWMJ3jmmn4iJSAPrkGlIopQmkF8zFtcG8uc5ZjekJgCGJ4SJ67D9k6FA/ulnysQxx3HSAL4F4CW4otcTjuMcNAzj54ZhfCKz2rcNwzhoGMZeAN8G8I9TtT8nA/keaARBTD/xtH/xtWaQu4EgCCIINmvNO7pYDlkb5xJjM/eiOFaYc+xbj+zCaCKNbz+6e8L7O53kmiS5efVRAMCNq45Mx+6c1BxoH8Zz+zrwX0/ux+7WIfz9XVtLvUtEDkoxRziVr3mkaxQfuWk91h7unroXmYUMRpO44fl6NPWOCcvFzLGMc4zLHLP4zLHMRAsvFE3IOWYa+i6Uk3CT5YPQBMAPNVOW6UwyVFY5/Uxp5pjjOM87jvM6x3HOdxznusyynziO80zm9x84jvMmx3EudhznMsdxDk/l/pQ7RkGHGkEQ04VlO8KMVZrEMYIgZiiy6DOetIRzHpulb+OcY2MZ51jbJMSxoAFOObKtqR8X/WwVns3D5UYdyHPzfx/fg289shs7WwZLvStEnkxH/td0HjrffXwPGnrG8M9/rJu+F50F/PSZg7hjQxM+f6coePP31FYmX4x1a7QdIMVljjEhLTBzLBHgHJMD+RUXl0b0yvyuFbQ4grpVAtm/t7qcM/Y7ZY6VB6UO5Cc4TPprEERZIg/yaLBDEMRMJSGVcQzEkkI48rhGHGOdw/rG/FIrXafLbASVxkyUm18+il88d6io22R89f46jCbSuDoPlxtdLnLT0DOWeyWirCiJc2wKtx09icT5k4l9bUMAgN7RhLCc7zzJhLKwZ51yhAoNzznGZ47x3Sot/bUjLA2s3bJFTZYYt062EkpeuyokCsk0VaHN1GxLV1bZ1BvN/4WIokByTBlBzjGCKE/kmn9yjhEEMVORO1RGE2mtc6xrmMscywxURuMpb1mhmWOMivDkb03jKQu3rGnA3RuPY2gKspGYy606Eir6tgmX7z2+B2lyTZQt05G9Op3dKSkuY2qwAr4nySzdKm0H+syxtN45FpTLJYtNhhzI74le2cP3dSH6hYzZ+TVNjVvN1LjVGL9YWY/9bcPa7XYMjePuV5u8z4coDiSOlROkjRFEWSLbmmdSRzWCIAge2fFl2Y6Q/zKesmDbjjC4GYun4TiOUBpZqHOMUVtRuOD0akMvbn+l0Ruwn+DKO/mumsVm2SnVgY/pXACESpAo8eTudmyiYP6ypRS3QVOpx/H3eZuP9U3dC80yrIBO7/znzX5n58y0ZcNxXCErbBqwHVcA40sxReeY/jX0ZZX5BfLryip1wfr5oBO9dCKdGXDNePlQl3b5//vTXvxiZT3+48/78t8ZIickjpURuoOHIIjSk5LuAmmGkSCImYpOHJPdsom0Da7SEmOJFOIpWxgw58oc29E8gJGM04wf6ExEVPryPdvxPy8ewY5mN7Oqud8Xx3pGiyuO8eVBc6vCgeuRNpYf2UTUEN0Xly3TUlY5jbda/H3dF+7aRvl3RUK+f2bwbi/mVq6pcM+nfpml6blzx1OWUD7JO8dSAc4xOZA/JAfys5/CaUatq9SJaIx8vqKGblua7LOga8YptRXa5Vua3MmDFXtmRofncoHEsTJiIrcAt65pwPP7O4u+LwRB+MgXXhLHCIKYqchllY4DIXMMcHMX+XKZsUQao4mUsE42cWxbUz+uWL4Fn7l9MwCgayTuPTYSn3j2T3/GJdbS7+e0yFk3k+VQ54j3+yi3rwfah/HNR3ahfcgtN6UJz/yIJYP/3imbyirLlVJkr05lmaU8AbC7lcSxYhB0vyzkh2WuOVUZIYxdb0KmgaqMkziespBMc86xJO8c058nIlLWlyEF8jOnltYlhuzOsULO73rHmbr9oMkAOQeUcfGZC7zfu7lrKDE5SBwrIwqdZazvHMFNLx/FVQ/vmpodIggCgFpWGZShQBAEcbIji1qW4yAtla1YjiOUl48lLK9jJb+dezYex+fv2KIIbnUZV0ZDzxhG4yk8vbvdeyyZtnO6zv7jz3vxueVbkLZsqbzGPVc3F1kcO9w1gh89vR9rD3dj34khbzmfsXblnVuxcl8nfvjkfgAnZ1mlbTv40t3b8IPMe5gO5O8GT1AXOqL0zDDjmJJvl+17SeRPkDiWEpxjYoaj5xwLGb5zLGmJ3Sq5c0NQR0f+HBzyHFvqeVnMF3PRO8yyd7AMghfaTI1LjP0eVFY5PJ7SLufdd22DMWxs6MOHb1ofmFFG5EewH5yYdrK1hdWRbbaNIIjikZIHhuQcIwhihqLNHJPOebYtOcfiKSFvDHAD+a/NdItcsacdV777bO8x/nZn5b5OLF9/DIbhl2qNxFOei0CmvnMET9S1AQC6RxPCIGsgmkQybWNniy9g9U4yc6xzeByfvn0zYkkLD21txYKaiPcY7xxj7/94nyvM8Xd0tu0EDnzKie7RODY2unlLP/ib12NeVSTHMyZPti6l8n2u4zgF3ysTU0QpnGNTmTkmnePiFHJeFALFMT4/LHMOqK0Uz/lh0xDKKnkRjJXk37TqCG5d26h9Db6skp1/deH4PIa27BHKcws5C+UU2pC9rDJIHOOXjydtfOmebQCArz9Yh03/+UGs3N+Jd527EKfNrypgbwlyjpURhV7u5Ra1BEFMDYpzjMSxaWU0nsJ//nkfdjQPlHpXCGLGIzsmbEfNHLNs2TmWVsQxXmQbiYs39wNjfgfJrU39SFkOzphfjfMW17rrjwdP/j26vdXfN9sRxK+BaBJ3b2xCPVf6OFnnWF3zoCDgDMX89xJLWorjxBX5HMQ50a6pb2xS+zAR2ofGsa9tKPeKHPzfftc0ZS5lE8f471A8ZeEjN2+YVlcbEcxMuw2S7+vGk1TSWwzkknwGXwopZ44xQqaJ6gq9ONYzmoBlO4HCGCAG8jPhSef+0nar1ElhmpywbGi3n8U5xjvd3rJsvvf7CCeC/WTFAXz4pvWIJdPCcv5cORhL4om6E7j60d34u99vyrmfhAipK2VEobNhvOJNg3WCmDpIHCstt61rxON1J3DF8i2l3hWCmPHoA/nV0nLeOTYaT2vLKhlJKTOlP5pUfp9fHcHcatepJItpPHu5kpGe0QT+7aGdwrYOtrvC2N9edDoAt9393hNDXifLQpFFP8At/2Fh/LrHE2lbuE58+KYNQvlnPvSMxvGRm9bjwS3NBT2P8b5frsUnbtuE5/d34uJrVmHFnvacz+GFqrrm6RHHspWvRbnSqU2NfWjsGRPEUaJ0TGX+l/ca0ktM5SvK93XkHCsO/OfKZ2Tz1wSWHyZ3Ko5IZZV8V0rLdtAzmj1nizeRhDSilCH9BPhulf4yT1iDKrZlQ1uiqc00U7PPLr1wCe77p3cBEB1iD2xpQWPPGFYd7MZoQHfoeMrG6vpuAGKeJ5EfJI6VEYU6xfmb1Ym2TCcIIjdyWWUpgmhnM22D46XeBYKYNch5X7Ymc8y2IXWrVJ1jvBgUTwWLY4Mx9/cFNRHMywhOI5oykq7hOP7mllexl8v8+vVLh9HHudD6xxLegOltZ7lhxVubBvDJ32/Cz545GPCOVVr6o7j2uUPoGY0LuWKMyojplRyOjKcF91hLfwx/2tmmPKef2898uGtDExp6xvDjFfnvt46fPnMQw+MpfOexPTkndvi/2f726cmt0UWEMOFxPKkf/BGlZ6bPEcYpc2zSpCxbuH++6uFd2Nky4D3GYLfUtZWyc8zwnWNJS5lk6RjKLvwIzjGvrFLN/xIzx9T0fX1OWQGB/BpRTVe2aUqvOd+bLHLPg/wET1Ov6EaOJy1UhH1Zp5e73vxuTQMSJPbmDYljZUShZZX8CYeCIwli6pCdY3KJEUEQxExBKau0VVeF5TjCJEE8ZSvOKEG0iorCUD9fCjnGi2PiYIDnljVHhU6RgFqSNxBNonvE3fZrMiWajPu3tCglkEF85g9bcM/G47jmmUOKIw5wXQhMwBmJp5T39+OnDyjPYU6bjQ19uGP9sZxOttFJdO3kqeay215t6M26Lv95ssHURB13+aITveZmBsn8/tB9bnkxHXOEijttGicmyTk2eXTnsK5h9/ysC9EvJHPM3VZ2cUwI5Dc1zjGtKJX5yW1noimHhmZjTCjTCXL8/hoANwHjTtDw18V90uRFLJkWzvV9XJzAb14+ij+8cmyC72L2QeJYGVFo2++Upl6bIIjiQ2WVJYY+boKYNsZTagmlPCFgSyH9tuN4JR5stpsveemTQvEHtGWVFV4Iv861MRBVnVeL51QK/++PJr3XPX1+tbL+l+/Zjid2nFCWy7D9bR2IaYU60zQwL/M+R+NpdOdRusI+wy/dsw03vHAYu09kzwPjP/NCrzl8HhyfuSaLizydw+N4nPtsbBv46YoDuPTGVwouCS0EXebY3MygMMqLY+QcKyumo6yylNC4avLoHMCn1LrHtuwCA6A0YQmHTG/ZeMoSmq8A7jkrG5GQWlapyw3Tjb51zq6JIpRommw3VDcZrwOYnHNsKJbE6kPdQpbmTqnsfTxlC+JYf1S85u5qLSx/cjZD4lgZUejBJ4QZ0k0DQUwZSlkl5bQSBDFDUcoqbUdxXNlS5ljasj2H1eI5FQCAnhH/5vxI1yge3taC7pE4HMcRSgwTmQHP/OoIwplRgqVxiLT0x5RlsmjT0h9DPGWjpsLPBOPZ0tSP//jLvrzFniVzK7WZYiHD8EpAO4bG8YnbcoceW7YjlBAOasQ+Hl7UKrSpAO964e8PWzWfIePyWzfixYNd/v46Du7f0oKW/hhePtRd0OsXgu5vMUdTVsmLaFPtZiNyMy3OsWnMHJORS8GJwtE5x5hIlbTUv2ZVJCS4p8KmgeoKV6qIc84xlk12YiD4fAYA4ZBYpghIWV/eY+DWc3/KQhUQIAjn8aXUdqY0+cfZ6/jLTAOYV+2eBwdjKXztgTpceedW7/FRTQOcqoi/Ufn7G9TxklAhcayMKDSQn88A0WU2EARRHNSySrppIghiZpJIq05ZpaxS6lZpO34wPXNz9XCCTutADP/91AHcsqYBY4m0MLnHWFAT8XJh2OvZtoO2wRjiKQsNPWrHRyaY/OzjbxSed+q8KmFgJLOxsS/wMZ7Fcyq8zDE2iw+45S+s5CVbR8j3v3YxXn/aXADuPVsj9x7kUkyZ9iHfFdGRwyEhwwfZ8/ACo+M4gsgk7w8k+6gvAAAgAElEQVT/N9cJhMVC59CZkymrZM6x1v6Y4PijaIPSM9MFSjIdTB7deYOdV3RllVWRkO/wgnueZR0s+UD+sxbWAABO5MijFcsq3Z+6skpAFa8maRYTtpErkJ+Jb/z+mqaBynBIELyyEU9ZivOOZzhWWOblbIbEsTKiUOdYipxjBDEtyBdxCuSfXvjZupl+Q04QpUYbyC+XVUrOMct2fHFsriuO6WaqR+Np7A4o71hQHfEGMOwc+9vVR/H+X63D/ZubvUHVOYtqvOewiUE5yHnJ3EphkCWz7nBP4GO8KLRoju8ck8WxRRmHXDfnkHvoq+8RtlVdEfJEOst20NDti2M9WUoxHccVBRmdOYKnZWTBiTnyWjmnxTXPHsJF16xCx5B+gMl/DlNZYhbT3L/6zjH3sY/dsgFNfVHvcV1JFjG9lEKfnM7Lv3wePBl46WAXvvnIrrIpCdW5Qtl1QyuOhU3BUcWXVcaSvnOMTUzkEu15AYpdDwyNI0znHCtmWaVOCBO6XTLnGJ85lvl1TqXqgNYxnrS85gU6hsg5ljckjpURhQfyi5ljW5v68ae63FkaBEEUhuIc09jBiYlzoH0Yf3Xzeqw/qg+LTnD28OtW1tPAiCCmENk5tv34gJLtYkndKuNpC5sb+wEAr10yJ3DbtuPg8cx9yunzq4THFtREvAHML1bW45cvHMataxsBADeuOgIAOG9xLdZ//zJcnOlEyZxjzF3AOHVelTDQkGkPEISSaRufuG2j9/+waXilQQtqRHFsSUYE7MyIXG87ewHef8Fi4X1VhEyEMqO9tG3jaM+o91hPllLJ/mhSKIvJlq2z+lA33nv9Guxs8TNoYilx0HjxWQtgGq4DLZm2kUhb+OPmZozG03hqd7t2u/x5dkTTsbNY6AbyfiC/+z6i0jp0DSg90xPIXzrk8+DJwL8+uBMr93Xi4W0tpd4VAEBUU9XERPegzDF+UoMP5I+n/G6VlRk3VSKLgHndp94siFpmoYH8ObpV5kM2oU3YN41gpnOT6eAbFlRncY4NxUgcyxcSx8qIwgP5uVm9lIUr79yK7/95H450jWZ5FkEQhSJnjunycIiJ859/2Yej3WP4yr3btY/zs4N3bzyOB7Y0T8+OEcQsRB5w3L3xOP7zL/uFZZYtdqt0HKBrJI4LT52L/33BYmHdz77jTPzy029xn2c52JQpafz025cJ682vrvAGMMm0jeXrj3GPucIUe5xVTDJnQmXYFAYRcyrDWZ1jQe7fQ50jONjhhx6nbcfLUuOdY6YBLJ3rimBdGeEq7A2+xLKgMFcq2sE5wLKF+B/tFu/jspVgfu2BOnSNxHH1I7u8ZXLIfXUkhNPnV8NxgNaBKLY1DXiPNfVGoWOUE8QKzTzLxt4TQ/jdmgZvkKyLBZmj6VbJoyvLJaaXmR7IfzI6xxhTWQZdCLrj187mHIuEhEmNkGmgOiOE8d0qeUFIx8cvPgNffM85yrkYkIUqtYRS103S26UJfuV1Qhg02Wf8NYv9FjazSzWeyzZHWSWRP/l59YhpYTJllfwJqLk/igszGRcEQUwepayS7suLSq4SAPlGjy9NIgiiuOTjmLAdNYcMAD70hqUIh8Sb+dPnV2FBjVuCaDkOYpk8LLmb5IIaP5BffT33J3ucDXTY4IgFObN9CptGVudYUPdH+Vxk247XrVIuq2TOMVZWyQY9fNZZyDS8fU1ZjiA8ZnOOyZOcdh41bOw82Tk8jh8+KYqZpmngrWctQPvQOF5t6EPXsC/M7WoVu54x+C6dxRTHPvl7t3nBqfOq8Ll3naUdQLMBH19KxUPOsdIzPYH8Ttb/TyUnc1xNJFQa78vtrzTirFNq8PGLzwAQUFaZOXTlQH7XZWtoAvkzQljS8iamq3KIY2wThrBMLavMFshvaHLIJoqpeU2dS4y/ZvnLsm97TmUYvaMJxJNW4PWTKAxyjpURkymr5Gc4dN1BCIKYOBTIP7VUhLNfiuQbLMp8I4ipg91P1GbJL5GdY4xwyFQcW/OqfNErZdlIWjYMA6iRtj+nMhwoaA1knFN++YkvOAFAVcRUgpyzlaMEiWPyucbNUnMdVHxZpWkYWJoRx/jXBMTZf9MQnWO844nv5inDnGNnnuIKiHzmW33nCH6y4oD3mTDYZ3H1I7txWBLXwqaBD79xKQBgTX2PkN91vC+KvSfUHDh+UiKbkDdRWgbcfdB2q+ScY7qunidjydtMgBenZvp1+GR2jkWyNCOZKk4MxPA/Lx7B1Y/u9paxxiDvfs1Cb5ll20hbNg53jgjPZ6WSQlkllznmOsf88z0Q3FFUJ3plF8zUbeiD+wtDF+5v6ES6LOWd2RzQAFBb6X8+uY7JfCZZCBLHyopCa5pTQrdKXhyjumKCKCZyWeVMvymcbipziGNjUuc1KqkhZiubj/XhS3dvC8zMKgbxtHu8VVcEFxcEOcfCGlGqtjLsLWNCSHXG6SU8N2TkHAgwV5b83Cppe3w5o45AcUwq8Xt+fyfiKRumIYb+h0zDK6vklwHi7D//eaRtW3A8dQyPB+YsMnHrjafP8/Z3IJrEZTe+go/d8ioe2NKC+zYdF57DJnHqWlQnmGkYuPR1rji2/fgATmSC+S85bxEA4DcvH9XuB6OYzjF+nwDRocaYW+VnjvWNqeIYOcdKA3/YBB1DxUS+1ZqqV9SJBvGUfdI2AMpVijcVJNKW8js7n1524VJ87M2nAXA/11eO9CqCOxPB5PMny5OMpyyk0mJZZZCAqROg/LJKTb6YpsRRcJNpXyUPsoheovimXteyZY7xE8psImE8ZeU8JnVNcggVEsfKiEl1q+TEsZFxco4RRDGRnWOkzRQXvgxLd3GXnQXZsnoIH8dx0NofK+oN/pr6biUPiZg+vnDXNmxs7FPK5gBXSPjR0/uxo3lA88z8YQ0w2Iy0jiBxTOfYioT8EkcmdFdFQkrOasjI7vYCggcMcuZY2DSy5rgGDSLkEu6OTPmh7UBxps2rDguDFK1zzDS8EifLdjxR593nLoTjAL9f1yi8XjSRhuM4aMyUjr+BE8fWHu7Bcc7xZTuOsL9yR1GesGnglNoKGIY7ucDEse9/9EIAQFNv9lL13rEE0kW+8LFPaUQzYKvlnGO9Y6owxw/EiemDP26mQxybLoIm3EbLJLurUCI5JhyngvGk/xkyQZvdu82p9LPEvvv4HnztgTrl+VUa51iIC+TnM8eqMq7jREHOMeY6zr6eTjBjv+q+8fkcBXoHm6Es469ZbJnumlgZ0ohjSQv8PL4unJ86VuYHiWNlRMGB/NzM2XiKnGMEMVWo4hipY8WEF7/6o+JAyLIdJVeiO0s5EuGzpr4HH/j1OvxiZX1RtnekaxRfvb8Of3XzhqJsj5g4rRlxg2f5K8fw0NZWXLF8y6S27TnHsoT7WrY+c0gnjoVDvosrm3MsFCBoLVtQLawDqPdLsnPMzFVWye38juYBT3Bn4fuym/WCpXOUWX3DMLBkjl9ayQZ1wc4xv6zyE29VM3nqmgfwpp++hGuePYTRRBrzqsJertnq+m78dMUBYZ9Mw8Dbr3058D3ysH3w/g5JC7UVIa80VJfrxWPZDjqHizspYXjOMfeelbnF3P00vTLWxh5XuLv8otPxznNOAUDOsVLBu+azibFTxVQJckFlut1F/s5PF5ESZE/xrtuezPmU7yYsu4Lfe95CfPptflOWqrB7veHPs5GQgeoK09sWO3+ydYNETV3Qvuccg3ge59cH9yx9iH5h+OKb+nzRmaYR7ryJIPFa9OE3nCpcY3jnGO+APG9JrfKauhJ2QoXEsTKi0EMvxR0EgnOMxDGCKCppuVsl3ZcXFd45wHJ4ekcTsGxH2wq8c3j8pC13mE4e3NoCALhn4/Eca+YH71qhz7+06LqRtQ0Wp9SSzcbLmWA8lu14AhN/k69zf5mGL3qxm/OqiKkMOMKmqZRC/sMl5+CMBX75oufOkp1j0vZc51jwezzQPoJrnzuEJ3acwBXLt+Bzd2wR9o8P3weAB7/6Hm3nMzmk3/0JYZmQOZYWP1t+sP/z5w4BAP64uRkAcO7iWu+5ncNxRKXg+mf3digiUVCmjO5zO2thjedqC+oKydPSrwqyhcI7vth7Z9UOfKZbyDSwOCM8Hs2UmC6eU+k59ai0vjTw4lgpnGNTJcg9v79Tu7zrJHKp89dkuSnKdMCPQ1kZNrtO1VaGhHO7YQAP/PN7PAcYwJdV+tsMmSYqw6yE0vZEzFxdGXUZXtp8MUP4ITyuC+7XvlYe+5HrMZ1LzF/mrze3KozbvvA2Yb1a3jnGHR9Xvuss5TXzOc8TJI6VFcXqVkmB/ARRXOQbcYuEgaLCW717RxPY3NiHd123Gn94pVE70xVP2ZSdkAdvPGOe9/uxHGVT+SA0fqEZyJIS03z+xRo4MucYn7Elw5dVVoTE0kJZ4AqbfskjGyxVV6jOMdMUXVenzqvEzz/5ZiE/x3NnSTdMlWFx8MWcXdkEsns2Hsd//GUfAF/4YWWf8yRx7JTaiJJpBqidKd2f4uehc47x4thNq47gA/+zDvvahoXXPGdRrbZBAROIdDlgfZoSRGF/uX0785QaJQsuGzq3Yi4Gokms2NPuiXhj3P1pNJlGMm1jPGUhZBqYW8k3PAAW1bodTo/2uOLYotoKXxwj51hJ4E8xpXCOpabg7z6WSOMHmjJ1wHepW7aDL969FTc8XxwXdrE5MRDDR3/7akn3QXCOZc5NsSQTx8RmKxctm4+KsCmcn3VllWHT8M6xacv2JujOWih2OpbRllV6mWPcMo1gZkg/3cfd/+kmBQs9CryQ/pyB/JnrCbfsU29bprikWWff8ZTfzfN7H3kdPvSGU5XX1k02EyokjpURkymr5Act5BwjiOJCZZVTRzJtC+J+fzSJH2fKh25cdTSwG5HsopgOLNvBzS8fxU5N4HU5MBBN4o71x7wudvw1YvOx/klvv2fUn0WfSLnJij3t+Oofd3glF8TE0X3/iyXas2Mue1ml360ywoljYS5fjMELRN7MfzgE2dzAi2iAPyjQC1Dic6siauYYvz4A/Nul5+PbH7og8D0l0pYnEs2rEoXBoH2TSzndx/znic4xP5CfOR8s28Gtaxu1wtO5i2q0TQU+eXGmJFPzHajv0ucB6vJ2FtZGvPfB6xx/+OLbtdtg3SUL4aqHd+I7j+3BbWsbAIiOx2gi7cWAzKsKK58lc441ZwbEc6vCnhBL4lhpEDPHpv5vIJ/SUlMgyHVmaW7Cyq3rO0ewqbEfd2xoKvrrF4MbXqjHES4LtBT3qLGEzjmmL6vUlccHBfKzc2D70Dh6RxOYVxXGaxarJYM8utywkGaZL4SpopROMCvmvLiuhFLndNOG9HPrza30GxYw5/B7z1ukjRWIJcg5lg8kjpUT0vc4l2WZn7VpG/RvrAoN5O8dTeDSX6/D3a+W50mfIEoNlVVOHbIDzLJtQRBjM49zq8L43kdeh1rmuLCmf9b6qd3tuGVNAz7zh83T/tr58Mi2FtzwwmHcmymj5LPaGooQot/DZb1NJPftR08fwJrDPfjU7eX5+Z0MZOvsWqxjIpHK7Rz7wZP7vS6+kZB48y4LOrocsuoKNZDfNKUwZk1nSl15oGG47jW+HMfUrHfJeYvwd5msLx2dQ3GMZc43snPMNKAtq5RdDvJrmobhOcnSll9WyYTHoKwjwHWOyZ/bwtoKL4dMx6GOEe1ytm9hQcg0vc/Yf80avDfTwZJxwdI5AFx3ykg8hfpO/Wvo2NrkNof4y652AGJlQzRpeZ0q51VHhEFxyDCweI7rHBuMpbz9rYxkzxoiphbeOSPfF00HubLxJkK2a1lXZhKIF+jHy7A0TS6Xy/YxvXKkBzetOlL0z1LrHPMC+cOS6OWeh/hzKru2CefUkH/+ZO/xojMXCBMyOnSOML+bMLde5nedEKYTqoqJKMixfeT2TVNqqbvGVEVCqAiZSFmONzEaMvX7TM6x/CBxrIwwJHUsnUP5528ODnI3RIMxte11Nh7Y0ozm/ljRQpsJYqbBjjV2PQrKdSEKRxbH+saSaOdmcr/z2B4AwBnzq/HtD12ApfPc/KG0bWP1oW58/o4tXue1qaalP3/nxJO72vCZP2xGf0CZUzFwHAfDMf/zY531Dna45Vm8OHYkwFFSCHzr9YlksbCb8fahcWG/iWAae0Zx29oGzx3O54DJE2jFK6vMCDhZMsf4Y1RwjmlC9cOmoYQxy6Uh7nqmIpCw53vLNI6DyrAJwzCEksGwRrzSiXQ8bYPjnHNMzL8yDEMcuGgGKbqST975YAllla7w2J7FtbJ0bqWyvyu++b6seUIdAdvT7S+/b4zKsKk4/1jHzON9MXzq95vwsVtexZ4TQ4H7oIN1nJSdYyxvcl5VRHBShEwDi+aIImDYNDznWDZRkZg6prtbpSMVrU1FWWW2axl7jJ+wK8du2fI5N5tz7B/v24Fb1zbiD68cK+o+8ALd8b4xYVlNhegU1p3H2flbngyRz1FvOXN+TrHKzw3zl3mCmeAc07jENM8tOBRc2n62feT3jT/36pxjusmhkGngnee6jUqaMi5bM6Dzsy6OgVAhcayMkL/HuS48/EWCvynuHokX1LEyV+t0gpjtMIcEs32XImtjpjI8Lor5G472Cv9nncrkmwLLdvCtR3dh2/EBfP3BndOwp6KlPtf5+XtP7MXOlkEvXHsqeGR7Ky7++Sq8cqQHADwhrr7TFcL4cvuj3aOTDtHnM44mMkDgbxRnuvMjniqOu+Dy323EjauO4o71rrM7xbk19raJAkUxSmkcR3U3ZcMwRPHK1AxmtM6xSEgRzExDLUnkfwYt88pxtC4B8bnyoOqbl52P8zNdvdoGY9pAft1Aju0n7yjRCVCmaXjurBQXyJ9NeGSEQ6qoOK8qoizjCYrVyFcsrIqElL/fxWctAODmFh7rdQdfQQHmMpVSRpiQOZZIe/s7r1osuzIMYFHGOcbvL2WOlRarxN0qJ+N2sm0HR7tHlcnNbNcyJjbzbrHrnq8vWnxN+9A4Oocn30hFFcdy/22Wry+2OOYf21ubBrD3xBAXyB8WJy9C7JzqPz8UUs+zcjk7AJw2ryqPcasqhOlKy3Xll55gBnW9Qsk3kF/nVmMLtaX70nXtw1K+WEgzIQWUJo7kZITEsTJC/h7nuvAEPZ6yHKw93JP36+ZzA1wMEmkLv1l1BHXNA96yxp7RaW8ty3KDdnD7QRDZSGduyNiNPgXyT4xk2sZX7t2OX75w2FvWNSw6q7Ydd4/LL77nbGE5u5liA7e07XhCQX3nSGA50WQZiCbxyxcOo28sgf6oL+Tx+VvZmEoR6OndbqkSc3H0jbn71zUSx2A0KdzQD8ZSnntDh+M4+MGT+/D7dY3eslgyjb+/cyvu2+SWafLv+S+72oTOc7mIJtKCk61Q10HKsvHAlma0FqFj3lRT3zmC1//4Rdz88tFJb4s5Fna2DsJxxO6tn759s3duAoBcVU6O4+QUSJkjpyKsdo7UETIMJSMmW+YYoypiKsKVYRgISeH+7jbVQH7+xr8q080snxn2sFRG+L7XLsblF7mllicGY15GDi+O6UpZdOH7uvUE55hlFyaOaQaGoZD6HnhY2eIV7zgT7z53oWZ/xX2TB1CVYfU1F8+pwBnzqwRBKsihJrOYc3/ZtoPRhC8qRBOWFwMyr0pXVik5x0KGIrYR00shE0TFfj1gcpljt7/SiL+6eQNuf6VRWM5KJ997nn+8VGdK1eo7R9A3lhAmO14+1I0fBgT4F0LKsvG+X67FJTesnfS2ZDNp0PiQv2bHklZRqyCiUp7VSwe7uED+kL4sPUfJoM7dqnOTyfgB9+Lz+Mfc37M5x3jBzEX3aU100lEn3PHnY1/MUz8j+X29h/vusueYGoVnusfbJyskjpURivKf406XDbp0wYQvHezK+tyxRBp/qjuB5r5ozpu0bz6yC1+5d/ukXQcbjvbhd2sb8dnlW7C7dRD724bx4Zs24NM58mf2tw3jC3dtLSjnIhsr9ri5QVcs31KU7REnBwPRJG5b25C3qMHDZiuZkGzNcNfLVLGjeQDrj/Zi+fpj3vmEz0vk+fIl5+D9r13s/V/nHONDs6962HWPNfWOaV0Nd21owsp9+bkdeP71wTosX38M//fxPcIMc8dQ8PeILx2K6O5QikAsmcbuVlcUG8qUKPIlnIe7RgUxCgCOdgV3rGzpj+HR7Sfw65eOeH+b5/Z2YktTP6559hAcxxEyx5p6o3h8xwnttsaTFh7Z1ip0zusfEx2CuWIDZO7ZeBw/WXEQH79tY0HPKwW/y4SP37KmoWjbjCctJNK2Mljs4z7XbM4xx3Hw+Tu34vN3bsXGhr5Ad3kixQLz1fI6HaYkfOkGLq44Jj6vWiqr1GV4ZSsr4ffN63KWo6zSNFQxKGyankNpKJbCWMJ3Min7pil5kV1zuv1l/09ZjjdorcqSHeftW0gVFcM5SkPZ33VuVQRnL6pR3oPY3EAVwqoiahZcJGTiglPnCsu6NA05XjzQhb++eQO2BDT/6B6NK90qPeeY5IgLmX7mGL+/nnOMrsElgRfECj2HF4PJlFXelpn4uXGVOGlxInMP8vGL/TzCedVhvO+1i2A7rhgmX0tfONCFtGXjoa0t2siC3a2DuG1tg/B5tQ3GcKDd70jL5+9NNv9LPo7tgPGafB2+/vl6NPZMPnIB8J1jrz/NPVcMjaeQshyvHJq/BrDzpqERzGQ3mXwd0k3AyLCH+e2bmtc0pJ88urLHQsn2LJ2DTSyrdH/KXZjdZeKEUZVkcgmKEJCz6Qg9JI6VEfLXOJXjwsMuErWVqriVK1/m/s3N+P6f9+HSG1/xWpjrSKZtrNzXifVHewsO+nccB029Y97MNj+wfPFAFzYd63P3tXsU331sd2A2zzce2onNx/rxtfvrCnr9IKbKYUJMHxOZNf7JigO4cdVR/OsESvCSUlllCXJoT1r6xhL4xoM7sa2pXzjXsEE9uzGt4AaLpgFceOpcYSDnBUpnfo7G015QMwC0DsTgOA4+cvMGXPXwLmxq7PMeO9Y7huuer8c3H9lVcLnbjma3M+XWpn50cgPCbKUQvLuJd5sVQq7JiO3HB7yBNstt42986ztHMJ4ROt6YyQw6kiWUf4DLquwbS6J7JI6tTf4gt3UghtFEGgtrK/DB1y91XzcgN+y3q4/ih0/txze4Y012rRUa5sz+nnJGXTkyFUHV4ym/k+IpNRG8eZn7N+XzcrK97kA0ie3HB7D9+AC+dM82fOMh/XkwnnEWVGrKHnW4pZB6MYjhCjri7WaVFMjvd3/knpctkF/IHAspz2WDDNmdpjixTD+rzLIdz/0gZ4657zWHMy1g9p9lsrGBSUXYzJobxtAJYbpyma9/4DyclykNZQPumoqQ8neR9zfM3Hqm+FnqXvPC00RxrFMSx5r7ovjGQztxpHsUj+1o9ZYLId0jCYwKmWOWnzlWLXarNAwD1RG5Y6iRtVvlyn2duPa5Q5OeyCWC4UWX6XCOyUxGRDp/yRxl2Z0bjuGVI26UwxkLqr3lIcPAR954GgBgY0OfEsI/pzKMHzy5Hz96+oDXXZvnU7dvxo2rjuKZve3esvf/ah0u/91Gbyw0wl3LJitaGNI5YSCawlf/uAM3v3wUv37psLf/fdJ1+O6Nx/HhmzZM6rUZrGSPTTaw+wOWCanL0+JPNWHNOTvIOZbr2qQrjfS7Vfr4wf38uUfcBr9M+1oTFc4MQ/mdf6v5Tg6ZpqFMwpqGGiEAkHMsX0gcKycKrBlnA6Ma7gaCDTDbBsez3iDwwdK7Wwe93+Uymf6ofyIttMZ+1aFufPA36/GDjP14ICoOvviyhaf3dODm1foSFOb0yRZcG0TKsvHZP2zGD5/yLdD8bE0pLu6Ei+M4+NHT+70W74ye0XhWAaNtMIa3/nwVrn++sAYSrIyWuW14ukfi2plwhldWmRHHKJA/f/7rL/vx4sEufP5O0f3ZOuCeg04MuMf1axb5DthwiAVs8zdJYlgrc5ydtdC9obUdN2uLHdMbGvzsssOdvijEBJ871h/DF+/eise2+wO5bNRWhgWBvzOLc4y9N0C9Gc1FMm3jo7/dgH95oC7rOZx3ZwzGkoinLGHgebhrBPHMzSrLDDqaZdKkm/v+tw+N46O/3YAnd/s39syN/LazFuAty+YDCC5xeflQNwCgrsW/tsi5LoXm1fADiWLleU2E5/Z14NWG3qzrTMV1ZTxlceHGYZyWaUzRxYm08us6juMtk6+fmxr17h722VZFTCUHVUdIElfyFXRk55hfLqITzPJ0juk6R0rb0wl3TIRLWrbn+pxblb2skr2+tllAwOcRS7nbrgypji0d2rJK6T3Mqwrjh3/zBrzpDPeYZPc31RUhRRhkz1eW8eKY5u8eNg2cvbBGWNYhTQ4c6PAdMW2D7mOO4whOsZRlY229H/kRTaS9+8IFNRVKZ7lISP0eZcsc++Yju3DPxuOe2EEUH37OPjUNs4TyK0ymrPI8Thxj93C7Wvz7wddzArBpGt4ERGPPmOIcGx5P4U872wC4E1VB1Heq19zjmdD0YUEcm5xokUiJx8Py9cew5nAPblnTgN+vO4ZfrDwEQMwNLTYs7H1RrVsOze5HdOce/cSH7jyungN1jloZXai+X1apil6ii0sVqrIF60+8rFL9XZwgUJf552z/uWHTULoOh0x9njg5x/KDxLEyQv4e55p9ZjMoVVxZ5PzqCE6piSCRtnGsN7iEhi/F4F1m/I0MAPSN+usVOmN/70Y3p4ZdQHhxrD+aUGYDggabc7K0k8/Fka5R1LUM4pFtrd7FsIUbuE5lJzkiO90jCTy0tRU3rjrqXVyOdo/i/b9ch+9mOhTqeGhrK2JJC3duaCro9U6pqdAut2wH77l+Dd57w5pAR5p3rGUGYRTInz/bj/uDcF4cYy4yJnKdw5UARbyBpxrgyiSGuaIAACAASURBVJaxAdjp86u9QdRz+zq89bc1+Tesh7v8133lSC/ah8ZxwwuHsamxHz94aj82H/NdZjz8jWQ8ZQnnsBOactDb1jbgmmcPopl3jhV4jmkbjOFw1yhW1/cIg7ze0QS+/ehuz/m6mRPHhmIpxaHGl1W+9Sx34JzNOcaLV22DMcGVBwAr9rif7dvPOYXrvpfJw2oZwP/59TpsbHA/R/mm7ED7MK56eJewzLIdpCw772D/Ee7a9Nafr1IGJGvqu3OKVpNlMJrEtx7ZjS/fs91b9uSuNi/7jZHP+eHEQCznd4N3SYzF054Lp7YyhNPmM3GMc45xr5tIW/jQb9bjrT9fhad2t6F9UD+5tKmxD1+5dzvW1HdnnsfyFUMTKqvUlS6GTDX/pEpypoWyuMR0ZSX84IBNWujKHkVBC9o8NLb9XS2DSKZtLFtQLTjytcKSpruYtjSU2/447xzLRxyTBoGGkdke76jNfBBse2wSs6ZCDNbXCY06AbEqHHJdHryjI2QKzl7AzYLiuwQPcuefxp4xOI6DRNoWvpMvHewSBPPxlIW2jGh7xgIxZDtIyMynrFIW7ojiYZXaOTaJskp+LPHg1hbsaxvySkOXf+ntShMO5jQ73hfN6bi5a0OTd3zzkzdjiTR2tgzir25e7y1j2+INB3JeV6HkEtcezoyBCp2sKwTZOcZKvCPSOQrwhTBDM6GhOsfEc49OMDtjfhXu/PI7vP/rHWE6dxZ7zF9mQLMs83uhOlg2Vxk/EcT2Q1wWfE2UJznk64nuOgyITl4iGBLHyghZmc514fE7Sok5G2dlZvg+fNMG7A1ot82fIPnuFWPSBYBfbyighCYIee/5wVv/WFKZiQnKPuNncAuFf83ujAONdVsCsrdwPhlJWXbRuujkIpZM4zuP7fZcIoUyxgXzsoH4TauOImnZePFgV+BsTEWWMOJs8OIY7/ziXZSsM6KMV1aZKd8JynM4GVhT340fPrV/2gKNR7gyH16Ib+l3yyCZyHXuYtE5BkCYDZNdDkycOn2+P6hq4o7tfW1D3k0qP3vb1BfFNq5c0HHcMm8dvJgXl2Zm5fLslGXjxlVHcd+mZqHjZt9Y7rLKrmG/wzAvSj3KudqufnQXntnbgX+4dzuGYknBqTE8nvKEFubwONI16p3PmXOsoXsUybSN7ccHlOOrmxMCWwdU4e9g5v1efOYC7+/DJnC+/6d9aOmP4Uv3bAMg3uClLFv7+aYsG9c8exCX3LAGz+7tUB6XGRacYzY+d4efGTmetPDV++vw5Xu2CwH1OjYf68ObfvKiJwYVAn89HI2nMBhN4ntP7MV3H9/jDYwOdgwLIp1cjgO4osKHfrMe77xuNf7rL/tw3cpD3jnpxEAMK/a0w3EcwS3XO5bwJq+qeecYlwPHn5f2tw2jqS+K0Xga1zx7yGvxzvP4jlZ88e5tWH+0F997Yi+2HOsXnGP5lFXKHSDDmoyYkGaAUx0JCYJZNpdYLkGHhbQL62nEtrCpilLhkOF9n5mo/ZE3nioK85nf5UEb/1PYX6kc3HOOceJYPs4x2XGnc9fJrjahrFLndOPPqSH1PbAJoLCU36YT865YvgXf/9NetA3GMBD1v6vD4ykc6x3De29YI6zPziFXvuss1Gbu945lrrmnzatWulVGpNLTkJm9rJIxGqcB4FTBn2OmY5JQvk5NpqySvzb89JmD+MRtmzz3W2VYLfOurQy7jSgsG0e73e/pNy87X+kMCLgdLO/Z6E7Y8i7dlv4ovnb/Du/5gH8d4aNqdNeJQsjHEdQ+NB7oHCtGNQR7D4tq3Xttdr3yOlNqncL+87XCT1DuonRtuvLdZ+NDmr8LwDmwmGCm9Wypjq1sbrF8yboFQXxTz+2+qy37JEdQzqducis2SRF2tkDiWBkh34fKF56hWBKfvG0jfrPqiPB4TYUYHMtumgHgxYBg/r6AE6R8U8HnxBTqHJMvagNRPpw5oZzM5cEno5KbsSz0AtLJXaSa+2IYiaeEi0P3yMxyjn3mD5vxjmtfDgxbLibrj/RixZ4O/MsDdWjWDLyy8eDWFiHnoGs4Dst2hJyogYCsJj67ht0o9Y8lcPWju3HnhmOBN0+82Pzu61cjnrKwqbEPH/yNP6N3kBMceNKSc8yyHYwnLXzt/jpc/ehuYQa93Pnq/XV4ZFvrhEXNQvjeE74DcE5lWCj7u2VNAzY09CGRtrGwtkLbHS6sG4xmbrSYS+mcRbVe3gJfBm47bi5Oa38MdS2+yyht2Z54dV5GkEtZNnpHE+iRxPL97er3ge3Soc4RKWzXP9fwrq5c7qC2wRjee8MafOEuV1jiHRh8rg9zSvWNJfD07nY4jp8lNhRLemH55y6uxanzKpFI294xdPp8t/Q0mrRw/fP1+NwdW3DfpmZhP3gH18EsuYzzqsOeUy9lue4v+VrFXzc6hsa1kxCj8TQe2toK2wGufnS3MjEjM6K5/rBIAH4wkq1RAgD8+xN7Ec2IaYXCv6+u4Th2cZEEzJH8t7duBP9xsCw3y3aw+VgfkmkbB9qHkbTccP3HdpzAXa8e97b1t7e+iu88tgcvHezCEPeeLdvxynFqK0I4VVNWybvNWVYe4E5s3ZNxcvP851/8uIHh8RT+/q6tXtlxVZ7OMdnhowuqD5um4hwLDOTP5RwLcBfIz9XPuqvByjrh57LXLxX2l/2uyzTTvmYezrF8hMeIaUoB+lnOi9J7qK4I60U03TLuNVh+G//+3dJTdbjQNRLHn3a24buP7cFgTLxe//qlI8qEKnMLv+H0eZ7bjy07fX5Vzny4MBfIn8gijunOFURx4EWUbA1AporJlHLqDAfsexsOiSI/+y6ev9R1j7Eg/epICJUR/dCZTa7yLt2m3qjiwmbXan5MVYijR9d1OJ+xUdKyAyfrsnWyzhf2HhZlusyy8WRYUy6py3GMaCY0IpoJjZCmjDASMqXSSP93+Vqgc4SJzjF1vYkG8mdD52ATrjGaa2lQ1qU8+aS7/gHkHMsXEsfKCPk+VO4Ec/PLR7G3bRi/W+t2XPFLvUT7Py86nXlKNWQcxwk8Qd788lH8033bvW33TUIc42f24ilLCIvuiya9+nQG30Uwmkh7ZaG8E2pjo778iWHZDu7deNzL4+ngBpct/VEhKBtwbf4zJbzVsh3saxtGynK0OQfFhv+7fO2BupyDW54fPy0GmHaPxNE+OC6IJ7zDL+h1ezJC55O72vHs3g5c//xhPLWrXfu8oXEx8+5gx7Ag3gCu4MFo6Y9633/5WEvbDtYd6cHq+m48u7cDv18ntgYvV/iZ0/AEHXj5Mp608CT3t2AD/aVzK71lN2Ry4950xjxxwCeVUALqTUHPaAKGAXzm7cu8GyV2jmH3MR1D47h25SFhkLb5WD+ezpQIvv8CtxtmMu3gPdevxruvXyOIqzrn7dK5VThjfhViSQvH+/zZ4OZ+8fu6bEE1KkImoklLe+M6HEvh2b0dWJ0RKZkQJwbji2If4E4WPJJxlH3j0vPdbY2nvNLR8xbXel1VGTUVIe/G84+bmwEAP3/uEL52f503YOA7Ue5v04vEgFjS0NAziv/9q3WC0yyWTAvCdnN/DE2Zc/mPL3+j52L7+7u2CtsNcm0ydC6F7z2xF409o+jgJ0H6swv1k7nJ5a+bncNx7ORKxI73RrVuFiZ2XvvcIXzhrm24buUhbUnu2sM92HtiyHNa1jUPKtfclZkurDUVYU/w5IVH3tVRl8lYvDDTZTBosgEA/uGSc7zfmXuqKqIGswOqCCMHLetcViETihhUJQXGazPCvGM+exdKz2mq2w9pgKMOtFQX1+I5FUruDXu+9zytCKgOSPhmBKz0qSLkdgLN9VUMSQN2OXeRrSMvA4CaSEgYaOldeGq5kza/LWR4pe46mvujyverRzMBy0TsZQuqfcEw41Q8bX6V4iKRr1G5MscYhdyLFJN4ysItqxs8EXsmwp+Gp8U5Jv0/l3PMcRx88e6t+PBN6/HJ2zZi7WF/ElCXV3Y0EzUgl+qx7z+bQGORBFWRUOCxcHom0J+fKJMbVwC+EMXfy+abOWbbDj51+2b88x93CMvzET3SliNM6PC85/o1QhXFRGCupIW1YlllNueYLnNMFsmVbpUh1TkWCRmBJYxsqd+t0n/MK7/k1/cEM0NZVkwMjUqnLbXUiGOykKtmjqnXSYAyx/KFxLEyQj6w+Vlg23awcr/vAounLK6sUhTHrrrsfO02GCPxNJKWjdqKkOAyA4A1h3uw7kgv1h12Q1MnkznG3xz1jiaEm6dk2lZuntjgzHEcfHb5FnzkpvU41jsmDGz/5YE6T0Tb2NCHu19twqqDXZ7o9aOnD+Dnzx3C9/+8F4DoHGvoGVMcPn/e2YbVXEDsyQzv/GAXyqkMrub/Lo09YxMqU2J0jcTR1CcOjo92jwouGga/jOXt8INzndsHUL+/x3qjymwLc8z0jMTxf379Ci6/dSMAf7ayigvk53OqTpby3Bbuc6oMB5/+b1ndgC/dvU1p0FEIcpYUO1994HVL8J0PXQDAzcUCgDcvmy84E3SDNt2yd52zEOcsqvUeY2LS609zHVXtQ+PYlREwfvv5twr7c9mFS/DWjFDTO5bwbvr5G8S9bUOZ7flBveGQgTdmwq95EVp2T7552TwsyQiB/Gexv20YR7pG8eMVB3D1o7vxs2cPeY+NJy0MxcTyc8dx8OKBTm9ZdUXIK9H46JtOw9zKMGwH2JrJWFM/S7dbnvxdB4DV9d146WAX4ikL+9p8IVAn3jAiXBnaqw19ynf/XsmhdLBj2BssXn7R6agM6NLXluU1gzjeF8XPn6tXylgY3SNxrDrYhX9/Yi8aMoMbPncm1/nxO4/txhfv3uoJiLzruGtEFMea+qLaa+RANIlDHSOeKHn/lhbPWXj6fP/6e/srx/DJ32/y38dATOkGyjLoaipCWDy3wtv+wY5h3LWhSRggbc+IY5dlOotm4//99YXe7+x7fLokVjDk80bI1M9sy/ky8vevKmwqHbf4n+x5ALTuKcFdEFIHDOxrJjvRdCU6cvB7RArM172m7xwzlWWyiMY+C76skr12NuS8He+zDWU/LwLud0Q3qNLlOPL76zvHxNfI1l0zEjI9Bw4rqcomxp6xoFoomVxYW6GIsabGDREOGd73L9u1qVRlld9+dDduXn0U3308ODP1ZMcSnGPl162ybXAcmxr70dgzhr1tw/jnP7oO4WgijaTmO8MqViIhQ+s8qq4QM4+rK0KBxwIr+c12/QT868iw1K1yLOFGlVz18M7A99k9GseeE0NYd6RXyEHLxznW0h/FvrZhVEfUsR8APLClxfs9Zdn43hN7cOuaBmW9INgxySJMWGSP/pqgClX+xAe3TFP6HdI4kXWZiAy/rJL9nxO9Mj9NzTJBMGPbVeTaiaNzjolllep715XHB3X0dLchvuaJgRg1ossDEsfKCPlLzH+Bj/WOCS6CtsGYN2tTXcHdPBkG/tf5i/GVzEzwT585iC/ctVVwjDyx4wQAYPHcSkTC+pszdnMxUedY2hLFr87huHfzxJwj8gWkdywB23awpr4H9Z0jsB1gc2OfYp9/YX8XdrcO4kv3bMMvVtbj6w/uxH8/vR9jibSX0dM2OA7LdoRZm3s2Hsejmfd+wVK/aw0buMZTFtYd6RFs43XNA/jvp/aXtDtavrRwrrje0QTu3HAMF/1sFVbs0TupJov8fZhIN1FG13BcmW390dMH8LZrX8a/PljnibWA6KzRiWNHAjrysf39x/91LgDX7r6EczEB/g09G/R2jbhZUOwmhM2q72gewENb/TyoqQw5LSYNXCB7UMOPlGXj5tVHsbGxD3XN+lnGfAgKWj9/yRwh7BoA3nzGfL1zLMegeE6Ve+Pqh1G7f6c3nO6KWbtbB9EfdTvjvobLNAOAL7znHO8mt034/rjC04H2YXSPJDC3KowLTvXFsUjI9DM1Mt+Llv4oruFELsAdZC7LOHfZsfHo9lZ8/LaNuPLOLXhGk7PVOTwuhPknLRvdI27JMIOJ0sxFMb/GLUfdlBFrZRcemzzhP8tLL1yCS85bBAC4f3Mzbni+Xgi8ZzeWH3jdElz/qbcI+xgOmVldJDeucrsOMxHqz3VtGIylUFsRwtK5lcLf77ILl+Cr738NALHD3Wg8hZ89c9Ar/dXlofzwb14PwBVz+DIW9vn1jiZw2Y2v4OsP7sRfdrXhH+7djtF4Sii9fTAzGIgm0tjc2Ce4iA+0D2PFng5sauz3hLs+qaySz/E63hfF8LhGzI8lhbJewBe5rvnEm/Cuc09RngMAB9uHvXPWB163RHjMdQK6392U5eDa5w7huufrvc6vgHsNP29xrdDoQodhAPOqIvjExWcA8HP7zl5Yo3XZVUquxJAU/KvtTGbmDuTXd5cUtwn4gwedW00X+CwHHBuG6NgKcc4ucX9z7JspPsY/Lpdfsv1nLqkKTUba3731DFz9wdcq+6HLW9NnjkmZbpIzL1umG/8e2D1hrkwbYT9DhnftZNfUbFEDy06pFs5HbKAuuya0zjHve+/fFx7uGsFn/7DZ+3+pxLFVmfMVu86u3NeJPQG5vycrQubYNHSrlMlVVqlrRNY/lsCbfvoSXjoYPIHLOmQz2O+ycF4VDilZeAx2fLPIhyCYOMaX/8YSFn701H6s2NOB5/d34b5Nahk8IDrR+DFUkCPo/CX+fQ+7nl564RLv3onn0e2tONA+jE2Nfbjgv1/Ak7vacdPLR7O+Fx72fZgrbTuicbxquxNncbeKpZaqcB70NwHUPC9BlDLZOvwT1GVsGwUXGmlOm7pMM20DGd05W+cw05yfddfuuZVhpG0Hv3rxsHZXD7QPY0dzcOfV2QSJY2WEHP7HW5Z3SIPU5j6/05WcOQaIN7Cbj/V75Rg9I3FclyllGhlP5TzJT1Qc6xtLKuJeynJQWxHCGRnrMRM0brziYpxSE4FlO+iPJoUspL2Z8p6lcytxy5Wu82Plvk7c8Lx4cL/a0Oc5RBg9o3GlaxHLGvrie87GZ95+JgD/Juva5w7hn+7bgVvX+jMln12+BQ9va/UGUeUMfzPaO5rAo9tPIGnZ+M5jeybUvnndkZ6sbg5WJscuvh1ZxLEdzQO48aUjgWHZ3SO+OMa7dADgpYPd+Of7d3gCleAcywgw/Hs/3DWilMrGUxbiKRth08Dbz3EHo029Y97NyY8vfyMAv+yQD2JfU9+DnlFXJDnzFHeguU3qlMc7LMuVlw52CcHoctk2g8+bkjsgFgILeF8qCZCvWVyr3Ni8/vS5OQdtfraOWmIlD6JYFhc7l7x52TxlHb5UiBdXn93bgUTa8lqfX/GOs4Qb5LBpeJkjiZSF0XgKX7l3O2QiIdMra28bjCGZtr0y0sFYCmfMV2dur3p4Fx7Z1ios29U66A0IdDeXbJbWcdxB92uXzhFmUVk3Y/5cf/bCGvzqMxdl9m3cczJddanvOgaAs06pVm505QF7EJv+64OoCJuegLTslGoYhlgqNa86InxGj25vxYU/fhFv+dkq/HFzM254wf28dNeed567EIB73uEHCfdsPI77Nh3HvrYhYcDQORzHlmP9Qmnkdc/Xo2ckjl+/dARfuHsbbn/lGAD3PPA77jrAcsz482hzf1T4/9amfjT3+fvBvoPMOabj7EU1mBfQcKZjOO59L5ctEL8rVZGQIBJsbdLf0H7gdUtyupMikjuLnVPPXlSj/TvLzjFTEr50ofRyR0tADaXXZXiFdA5SVkZoaJYJzxUf4x+XHQyy2Os6LbMLctruj9omAP533nOOed3b/A/uc+88C1/JTNz4z9U72HRuMnkQX1MRzul+02WOeQNZ6fOVz5/8RxYJmd51eWlG6AoyJ8ypDGNelbhvS+dVKvumE+SYCxYQhZkrlm8RumBOR+aqDD8ZdMl5i/BqQy+++cgufG75FsENfLJjT3O3SlmMyOUc05Xnr86jqiHoPKk0EuEiCmTGkxZODMSwv30YNRUhnFKjP7f3aJxjY4k01nKTwI9lJvLTlo1vPrzLuxZ2CVEx7vXBth2lwRkAnLuoBmv+/VK8/exMQ57MZ3PRmQuU97uwtgKxpIV/f2Ivlq8/JjyWq8kNI5W5p5TvGYJKAQF9BqJO/JfP2fKfSx7L8mMAtqpXVsmH7yN4ma7EsVCyPU3MORP3UViWpQELWyZfX3XX7v/4mDuhuP6Ivqv35b/biCuWbym4w/pMhMSxMiKbc6xOUnO/9kCdF/IoZ44B6on+upX1aOkXyz7e99rF3o2QTEP3KK56eKcQLK2bFQ9i1SGxEQArtTlrYY1Xj85m+2srQlicCXDsjyYEQYsFLs+vjuCS812nw/bmAWxvHkBtRQj//pHXAXBzVWTFu21w3Gs8IA/6zl5UgwWZCxe7yXo4Myi9JWMjHpe6eO5rG8Klv16HG56vz/tiMZ2wQGjAHcTxlmuWR5Qv9Z0j+Kf7duD9v1rnXWT6xsRtspKfN2QGgZ1ZgrCvWL4Ft61rxINbW7RZIV2cOPa6U+cqjzsO8OSuNgCic6yxZwyW7Xc8rIqYGImnvQHeQ1tbcMkNa/DFu92w8wU1ES9D4v+z991xllRV/t9X4aXu1znOdPf09Ex3T07MDEMaYIABERcxgQqICdRFUdd1zcsaVlf5KbDqyoICgqIiKqKSoYdhmJwTk3Ps7umcX/j9UXVv3XBe1etZFlic8/n4GayuV3XrhnPP/Z5zvmdvWx8H+JjxcLJnCL9cvh93v+BxiN271KlAdO6EUu1g+Iub5gJw5m0qnRlVhOHjG47gc0KVu2QqjTue3oGv/WmzFCnz3Se3/485zTYe6sQtD63lXFtAdg/sagH48wM8g+SEa8SNVXgPi+O2ZlzmRyw5eoF7C7NzjgF0ifCwZfAoMQaETB1TqBlPYdNLFRIdEU9tPY47nt6BLUecNfPJiyZIetIyDa9iWiqNZ7aekKK9vLaFUOuCqYdODWDZ7jYpOquP8PS+SkQ9Mh24sKkc08YUaN8uGqIViYh2sOeRY4o3lgFo/cNJ7GvrQyikp+CV5kfIlDM/Ly3gpFYVxmzOdwV46Vrq+DHA+XDHAJ7eelzSD2yPoNJU8iMWyhMRJNM6j8q/PbGNp4N88Ow6XOJ+FztMlOV7lWs7B0Z4yuMPnt6B/uEkHlx+QIoyYOtAdBax6NKqgigaK/JxuGOAe2Qvbi7HJZOdd3YPJDmXYUO5HL1YWxwnuf8SbtQdewfjF2NiC2CFmnopyrz6ksCxYs9RbYG6ktzAMTVVMVvkmPos2zRy4KBx/qXK29OVz3RgLejwRaZaqlxfBCDnS8ivgH48ckxJqxRfa5kE8bRJR44FAXKAm1YZEJXhgY/emFKk2BaRFiumJ4dNg+/LqjNElcqCCEKhkDTfmI4ylXFW3ynykA0LNpgaKfZ6RI5tP9YtRbuLoIxphHDnc47+GU6lcd9SOgro/6KI55LXg3NMlSBwbNcJHRzbcCg7hyYTNS2PzURVP8dsU5qXNcUxXkyofzjFC5osbCxHXkSPzgI8UEy0B7Yc7UL3YJLbFiwSc/ORLvx18zHcs2Sv4/AXbDJGJ0MBY07bDelfNl9rlMjNr1w5Cb+9eYHbphGNJy1XDr8RHjkmg4K0TnFBKaF7Kf5EEtR3+cVkgF4eJ3FmemmVTP/rfwsk6fchHTvdVUBxn1FRbVRapRoVHFL4NKmCL2+fXg0g2K6n7NC/NzkDjr2JRLVDRQCGRXO8fUa19juRc8yLpJCH9mTPEL7/1A5sEarx/fPlzVnTKh9cfgB/2ywDXKOJHLtniQMo1LspHcxjceuiiTw1je2rsbDJAb7P/mYDlgohyYyUvShuayTTb59Rjbe5i30klebRPMx4P3Sqnyt1BsgxqREOJSNKBE0m43hi1gsHrv7hJF58tRX72/txz0t7uVfnzSQHhZSavW19UlqrmHaUi4iEoluPduNkzyAu+kELbnloLb/OCO6njPH4nYJkw6FO0ovaM5jEJjdKcKoAAABO1A8A/MdTO7CvrQ+dQsn4JTtO4nj3IIZTaZQnIpxriin/B17ZLxFnF8Rs1Llz8kjHAJ/TDJztH07hG49vld7P1t55E8u0w8uE8nwkIhZGUhlccedLOO97L6B/OIm23iG8tLMVmUwGj609jPf+7BWJnwgAbvvNBvxx/RHc/4pjOP/g6R348Yu78fCKg3wuH+0cwD1L9uJHz+6UALNnth6XjPF9bX2+3p4dJ/TN7qHlB/D4hiOasSmmgYx23ojCPOlji5SDvaXz16iHZyoaguKeYH8Xn1cYszUwoTw/QkYhZCtKsGRnK++XRNSS9KRtipFjaQ6YfPyC8RwoZu1lUVFbj3ZpoexdQoGC6+bVam1gOmv9QWc86kpiPCqDPd9pj/ftnMvI1A+e4n225X17R/8IkukMxpXEpQMv4IBcKrhCRZGown4jAilUkQXbDKG2xCMxZuDVt985DYDHyaVGBTvP8fqXpRP+8iPzeeQpizpuKM/n7WFk/RPK8/lYqem/L+1s1cA2pk/EogWsrQ3lefjEhY7zhe1zRXGv37Yd6+K6bWZNEf992DKQF7FI/hqWKrvEjXS+cnq1tBZEDrken4NLPJI9BYhJtujLbGmVUSKtMsizbRo0gbJJgVkKKCM+0/m7915+Hz8E6ocDitdMjUjV9VEw5xhFyE+2zfCAIJGQX2w3ezbFhxZEvk9F1wEuNxIZdUaBhcI7yeIGMi9bKCSDY/va+jA4kkbYMrRIyOljC6XUriiR5s3eGVLAR11ne8A8c2qqFYYBaJUzX0s52jmAX7y8D2+7aykuvqMFmUwGdz23izvgAMceFYu5ZKuC/WaQ0RakEvGw16dapfNCNheC0ip3ndRtnZX72ok7ZckWOaalVdpy5FhlQRTfucbZrwZHUtwhV1ca5+tclWQqjUwmg1NCij+r1H5hs5NC3zUwgpFUmqfqAk4W5tlzfAAAIABJREFUhQhcMYc4i0gtzQvztjhtN6RvYDZHbUlcicD36BlGUp5Nw2yH7oFgcCyTyXDgNBHJEjlG6C1q75AcH4SNR4H6al+L01pzqASAXv7WzejED1STssUUAE/8rVrcxvnXf3+i4l6K4zYiloGeoaRvdO3et3BBkVzlDDj2JhK/tEpmnM8dp/OTSJxj7FBJKPq/bj6Gz/3WIaqfUVOIcaV5gYazKLmCYyOpNI50DsAIOUa9KBc3V2jewHjY4so7G2JdGLO1tlYWRPnv+oaT/FD/zlljATgHlXSGlV6WDfq8iMWVaTKV0chjv/DoRnz3Se8we7x7SKqmyaKY3kwiFh9g3ismh3MAObYf6+Zgohgh9sSmo2jZ0YreoSRe3t3G+6pTjRxzN+10OoP9bX3c6OpQKtepZa0BJ0Kia2AE40rjGjfUndfOwqJJFegdSuJP64+gZygJI+RwmxztGsQPXNChUUgnG0ll0D04ovFPFMY8kHVgJIVUOoN42EQsLM8PAGiqzJf+/9iiGGzFu2ibBspcT/muk71o7xvGrG8+i7nffg43/mIV7l26F//06Eas3t+BHz67g/9OjEBhfGqiEcR44lg6VjKd4VFumw934eaH1uLSHy4B4HErXfajl7RvYHKY4H9Ztf8UbvvNBiz49+clQEwE3Y50DqB/OInnt58geZ9EyWQy+P5Tr+IeNySfpVWqkWNUaW4HMNNBryDuCX4oFq4VxmzElfG0zZCmP2zL0KJlzh7vpOod6xzkkQkqmb1lhBA2necPp9J8XBxARAYwWFTUizta8erxHpTmhfFJJYr1O9dMx8QKea4B4NcYUFNbHEdJ3AP5PcNXAJsM2RgGvLRKmbNDj/5qrExo14ritgYcUOOnCsUZR0X5WYbB+YaOdQ7wNG42DiOpDDYf7pKKFojPY5F5TMaVxnH+xDLpWkN5Hm8H20fFvWP5HllXtuxo5evl6lkOD9fRrgE8vOKAVM2WSW1xXIs6EPcrFoE2q7aIp8AC4IUJ1L36ny9vlgz9mTWFmFiRL42NZRpZU3tEsQ0jZyBTfH48bKIkL0yCY1qERUjxbGfhSaEjx7z/T0aJMaNfSo00pPsBrw/JSCmCd0slqtf0UZaILYrgPqjamiF8u0rIrxYV0AjolbZRYKHN26HaVTQhfxBJP1XcwFL6wzYMrlcAcF7YMYVRqCnoj//jeZg/vlR4vg4Mevrea78ZcnjfxKExDU+PM+cFVYDnWNcgbvzFKl/es9OVTz+yHt/8i6ePOvtH8KPnZF6mzoERyX7vfoM40ILkD+sOY/rtz4yKZ0jiHHsdI8c82y47IJfJZMjIsb1ZKp+Lku0spF53CPnlOc5syv7hJAdmS/LCWXVv33AKt/56PY9OB4ATruPlrHHFSEQsZDJONeb/avFSHJ/acoxMq2SguwqIc+ehohtqimPafcx26BtKobN/BJYR4nyV3TmkKTPQkvE1siAIgLbT/KpVBlUK9vSx934trVKI52K/pPYY9l8UEf5rIX6PkqLEKGcO6YDR9w6/ggdyW0Kc1kiNDhQjQvcSvH1/b3IGHHsTSba0yp7BEfQMJRG1DY1AvLowyqNeAGGR5OgtzpZWKcrNCxsAOEozF2GpZTHb5Kk0TKiUnHiW6i95gvEVC1t6BINwwDvRPYThZBpNlfmYVuNUknvVPcgkohbCamqQ4RmiI6m0RuL+h/VHJKPrRPegxC+z7mBnTobXye7BnHkZ1h/swIs7Tr9y5gkBvFNTF4OiutYd7MDb7lqKWx5yKvuIQOiek70SZw6r0sjAsQllDijVNeAQ19/1/C5cdEcLPnT/agwlU5IXYuuRLt9+W9hYrgFQYdPkERcs8qM8EcF89wDNUgVvXtgg8fBsPtyl8VUURG3tIFREAK//fs10fPXtU6Rr1Ny1zJCUogXIff+kwPH1yp52Hta+4aAHRq3e34GOvmEpRYNFT4qHcTb/1CiwzUecZ53qG8461/b4eILa+4bxPZfPIplKS+042jmAzzyyAR99cA1+2uKf2rntWDd+2rIH333yVXQNjPC0yhotcowAqgxDOtxRxhRVqc0mQJjCmK3NIYsYOwo4uHXRRJTkhdEzlEQm4+hkU4masAzDixxLpvk6KIzZisHpcY4xee/cWo2A3TJDml4HPO49NqS1JTIIww7FYTGqzdKN4SiPJpMPuyq44gAw8rWIpZPvW8QhXpUwAbjwA7ACmLFIkr7hFEZSGVQkIjwt40jnAN7x45fJd4iReeLz1LFvKMvj7WEGYVl+hLeN8dEsnlIJwAHHWFXXBW7RgpYdrfimC9B98Ow66fk1xTGt34qI1OEfXTtLBk+JKL8FDSX41EUTpH6rdMFDKXWYAHspsYlxVoWKyMyPWAiFQqBeEZRWSfJ/ZUmDCY4cI4x+Kq2SiHaiwCA6cszQDhKWaZCpnFRapdw2aPeJEaoMRKKqVVKRY6YRygGQ03Ug4BCHB0W6UQdUm+hLSwHTTQEQEOXu98+W2hFzATpxDnLdoIBt6ju9lCLZMeFF/Gfw/PYTWavpvbSzlXMIZpOu/hEe4Z2rqBHgBwh7RuV4fSM40HKRz/9uI3qHktyhlYuITrLXk3NMLECSTU50D/lG0vqJun7YElbPJjHbVPY1gztXB0ZS3HlcEg9Lc/fSyZVY/uVF/P//dfMx5IVNnDfRA44Bxx5lUVxPKEV7Vuw9JdmEzJZmoHte2JL0FuWQitkmShXgzjS8fZOlaJblRzj1THcOgRGMw5Y9l8poogq3SFGrfvqN2mPEb/WtVikD/uJI82vCxdcOGvOXoKqZ7JMprktyXxPmJXuOulpYNL6aWimCzhTA/PcmZ8CxN7Ewrwwz6McUxrTQ0Qc+PF8zWgA5coAifube4ixplUwmVxfg+rOdypdBuf4bDnVi/nee42BF1Da1Dcc2Q9JhDnAMKCr8WDoIup5Lybi29PSeefUlnASTkSjnRy3tMGeb3sH4P1/Yjfffu8L32050D/I0xUqXPNYPyNp6tAs/eXE35v/78/ji7zf5PpvJNT99BR++fzU+eN8KDkDlKplMhnueRGGk0EHpcX92x2zZ7nbsON7DD/yAk7IkphkxknMGoBXl2bxPTvYM8RSCl3a24tltJySwJZnO4MHl+7O2Y0ZNoTYXbCvE5wIDLJsqE4gLm2/UNnBhUzmf94MjabLaT8RyKhKJ7yhUIn4A+lBpC1xTTMKmIYHTqoj3ZzLA++5ZjrUHOvCxX66R7mORW2zjYp5AsTBAW+8QXtndhjsFL/XgSEqKCsxWIXKf6zm9oLEMFzR6kTW1JTHkhU2s2HsKP3xmBw53DGA4leaGzavHezh4cMczO7Ombm472o233+2BGGsPnOI8dxOUqCj1oAWwqAnqwCcfjMS/sWc517z7HLBTHruwqQNhDkgnXyuOh6VKtlSUg2V682c4meZcjGqUlWWEUBCTU4xitqnNF9swNMDha2+frKWj1hbHteghsQ/E9kqHUYsdRuX71DEoiBJRYqaeAmubhgRwUcL6OmzqbdM81srz60vzSEBn+thC+R2GIUViZfuuqsIobw/bS0vzw7wdO11D8LZLG2EZIRzvHkRn/wgilsFB+WNdTiThO2eNwQ1uJWgmlYVRPeIuZmsRViVKiir7RvFb88KWVrSAGf1hSx57dT6r72O/DQIy2d9tSx8rslql4vAylLRKX+J3BYSRwRv2e/nZzjV5/Tn3BwByPgcy9T7d8RaiD2PEISWIF5ECvcJE/6pcgZah88jkyjkWsx1QKpdvF5/h/DcBjimp1JYR0qJzASddmUp1knSUpb+TXaMASbWgAnvmxkOd+OiDa3j6NJNE1MKd1zrFm/wKCgHARx5cjRt/sQqPrvUyAR5ecQAfe3AN54fbdaIHD684gBE3FU6VLUTkGgPHGBfVG1U9009EflQ1Wt9PUkTkWN9QMmvRkddK1IhBSqiUylwlm55UnUMqOGaZHlA8MJzimRLFebJdGQ+bqC6MSevqxnPrce082dkStgwOSrG5PqeuCBc0liGVzkj29OGOASRTaQ6OUVFtrI1MakvcwjiKTlWdxhUFEZ4i3T3opHhmiyDbeKiTz3FmZ0jgGNN3hC4LEXpc1ou57idy+ynOMQ8IC2l/ozjH/rdFAunc4SD3P0mPO//Sxap03a4KoxxRI8fEKNDNR7peF+D7zSxnwLE3sfxy+X68erybpxpWF0WJtK4sh0rhmppS6PxO9+xTsnhKJY+SoIjURbnuv5fjZM8Qvv6nLQDkilqsTaGQbojGs1R/kbh0iENq2NTToiZVF/B7WaW9RMTS+82i01KykcmKkWPvm+vwA734anZw7O13v4wfPO2k0T02yhTMZbvb8YF7VwbfKEhn/wiGk2ltg2ME0DtP9nDDrrVnSPMaiODXXzcd5XxigGNgipF1T2w8im1HnRRM0wghEbEQdQ9Mw8k0WoVqcE9uPo59bc7hkxlgS33KXDuVgPTDOBsrFh4/ubpAGtOwaUhz69E1h/Dc9pPagZH9Rhz7wpgelRi2CCCMingyDZQqkWOisDnDKqO29gzhffcsByBvXozfblatcxjv6B/BHU/vkADYY12D+MB9KyUw7HjXoJQG+dz2E1qxiEzGM6h+/P45mFDugT8l8TBuu7QRAHD3C7vxkQdXAwBm1xWBkl9kKS+uVjf6zapD6B5MoqogqqW+qUAVi86iDB2Kb4cEgwxxPG0SYFV1haVEqwEOeCqmOFJRDpYpRo6lOJBcFAsrB2p9DolAr/it6ry6eFIFmeIoOjMozjEvrZLSnXL/qroiTESJiWtPam8A4GIT7/RSwuTxU50edaVxTV9/713T8auPn621g+xLAohg7WGAfnl+RPrt2KIYplQXSNdqS+IozpPBTSr1NGrrOqtIiRpg7QjqD5JHLgsIqo4Bc1D4vZMSDtKJNgOR9qf+jYlGyE+R0iuee8CNzgoAuimjn0rbpCosUsCPSbWDAK804M4vlUV4PknSb9BrTWubEiVG9RmZskqkWjLgik6hDDiMZonokCpkmiGNew7Q9YWfHpAABiJ1Kls10FzShCdVO1G36gFQFRYFJkbofO1PW/Dc9hP4/VqHV/a232zA1/60BV94dCPJvaQWywK8CMG6EmfvezOCY+uF6PX8CF1VkRKRZowdoG/4+UpcefdSrNwbzO11usKyPyhwjNm2/5OIl2wRtqpzKBpWQWwD8bDDsTUwkuKFKUryZGdTmLA9w6YhZcg4fzdQFHNsSkbt8PWrpvAoZlGS6QyOdQ1yZ2pexKQjx4R2MC5T1R5Q97CKRIQ797oHkvjYg2uw4N+fxyqlUvuSna24+ifLcLPr8GXPFdOuPTDL+51J6B7yPkqPE/rYj3OM3UVFq3FwjKhWScloOfr8tJUcrab3B/svMq0yaF/IgvCNKfJoLEQZEc73XQMjEuf236OcAcfeRJJWFt3SXW244s6l+Mwj6wEAVQV65Jjq2eeGEuGhln+nHzQpCVveweJkzxBu+PlKXPL/Wsg0vcERedOK2lkMJYJzTN2Abjq3XgtdBvRIBPUQFbW8a4zsMhG1iUOfvhnMqSvC3Hqd0y0eNjE4kubf/N6zHHBs+d52rNzbjqGk47X59cqDeGVPm8SxJcpQMiUBjA+vOID73EqIlIym/DdLqawt0Tl44mETnf0juPO5XViysxUXfP8FXHHnS9xz2NU/IqWQ9gwlpbTKvuEUN/aum1eLdAb49aoDAICCqCWBUsPJNFqF6KUXXj3JAa1rFdLxGxaMw2OfPFe6Rkb4mIZ2CJ5UlZDmQtgF59jYM/6ur181BTNrPaCH8/wIz1NBDfZOP5BOvBYVIikSEQvXL/A8gSzacFJVghcWSKUzKMuPYN3XL+MH2l+vOggAvCIrAPz4xd3SmqJ47o53y+DYNx7firnfeQ5X3rUUt/56HV7e1YaRlFPm2zZDKIzbSkSmgZsXTkCpazCJ4KO40bJqnr985QDpPexTUhlY/8+uK9L7UYnYsg1Dr7RDcGf5eQvFaxQJuU3OK4pryJAqLVFksLYQvSCmVaqRY2Ei+jBMzCuLmN9hQrepgC37bzKtkjDK5ag2vc+pVD11HZgsmkXxiF8+tZKngYrvkkAenvLpD/CV5IU14IcCg8hUWaXfVNCcSVlCXvPMmy6OQ12JTqgcJgBPSmcVEmmV6p5DASSs3+R9zu0jpSiEWAURAErydHBMjHJk8uuPnY3PLJqovVO2GXSwgknElp+XNfWPOPSoXFlkZBPFOSY9H1rbqOipbJUp1Wtq2h8D6+lKmvJ3Z3unek0F4UnOsSy2GLVeaM4x71qUV3/U20uRYqvp4M41ed2rgAANjslz3CbWPBVBSvUHa6eqf4JsVtMIobrAjY7oHMjpMMv2L/HeQx0D2Ha0m6exPb7hKJ5/1dnXmisTnPh89f7sh0gW/ds7lHzTRWLsECqYB2WFiCKeU1LpDDKZDNa5QBuLMn8thb2N6fWkklb5mUfW45IfLsHAcIqPT5D8+dbz8JdPny9dy0ZFo1MNmNr5yuMc89Iqi+Nhct6L6zxsGRrfrW165PjiNdUpwaquH2jv9yLHbDlLhnJ8MFuZSi0X119FQZRHjm083IklO1vRP5zCxx5cjZ0CtQfjx2URnKxvqLRKGvz2vsmvKEsukciyCJxjCnhEAWHSVuePv79mQnGbUdxnFCE/VbSA0p+qMHC0U0mVVQvTPbf99Cl+3gpyBhx7E4m4d37+sibt72OK9PSNsCV7MZgeVzlyPqDwpHAjPCCtUk0vW7qrDXta+/DgK/ul+0Qgh/HAxML6JgJAS+WMK2mVDWV5+PKVk6TNgPRAZol+YM9i/Zkf0SODHL4h+doDH5lPpneKlS4TEQt1pXEYIQcMvPa/V+CD967Eyr3t+MofN+MD967EboLMcCSVxtvuXIprfroMmYxjUHztT1vw7b9uR2vPEGnArRQ8NL9fexhzvvUsthzpwg+f3Ymfv7yPg3KAR+Y5pigqeUXiYQufu9SZS09vPY4fv7ALgyNpdA8mecrehsOdkpdlOJlGF0Gab5sh7r3a6JbGZh4zZtwOp1JoFVLvBkZSPOqsWTg8Aw5vmMYZRERsUUBmc1VC8cDpRoBlhHDV9GqJb46KmCyM2RpQkBUcIyJVJLDJMvCNq6Zy0IsVOHA4iOQ5LpJ2t/UOozwRwXvOqkE2eWWP7pU90T3IK+Qx6ewfwbZj3fjLpmO445kdErE86xe1P4oUY+zmhQ1S/84fX4pzGkrRM5TEQ8sdYPS+pXvxL7/fhP7hJH+Hqrdm1RbpfD6GQRpw8kGRXSOA/wBwJWzq4JhF8ItRY+zoD//5YpkhHo07lExzILkwZmsRZg6YJL8zl/ntgBr+7SUjx7gBThnlOqio/lZP1dPTqcR/AecQcM8Nc/G2aV7xFdZ2Kq1S5RzT7yNARYvghTLo1Oeg1EUAKM2L0P0mXCuOh3XQ0qLmjA5A5Ucs6bcUdx05BoSXnDpUUWu5OK5Hf6jOM9sM4dyJZagrzZPu0d7J0twIAzuqpFXqwDYFaOnP06Oz2D3es+l0EQJYMvVrdJSV898ZhYkl1yi0IML/bNFeWSPHxLapTkz3/+ZOqu8PNpGRYxSYZ+rvUHWDZcjRnoAHRFMgq9w2NsdFsJ7qN7ndzn26/lTFNkIoiFmIh030DadyIsNnfLrtgmNz54kefPMvcuVqVoW9pjjGI6L9+FxL8sO8qmfvaXJh5SJdAyNaxHiQnBB40UYDjqUUW1VMx6JSu18r4Q5YIb01lc7gzxuPYm9rH770h01YtrsdiaiFgqjl9yjkRSyN5zNbkRkVjI0oNoJlGoi6RdEGRc6xPHn/4Daq4rxhNjS/ZoVQpNAxhC0dHGN0FQdP9fMU4HjYlMAmyvFBRSJTerYsP4KCmNO2X608CMAJeOgeTHL+TQBaoSb2fDmtkj1fBroBJa3SR88G7THqvixXq2T/uu8UbvUix/Rrr4WMltxf3sOcf8nIaQJozOVN/KymZIKpXH6bDnfi71nOgGNvIhEX8y0XNmjhtmKFLSaq0UKVuHUO7FN45RGAJkymhDpAAk7amgigrBeq3bEoo5iSVskWpX4QkiNypowpQMRSSC8t+jCnHVSIKIxE1NINO0NOvymK2yiI6jwx7z2rRrrGNlTx2poDHbj2vz3OssfWyhE+NcUx7Djeg71tfdh6tBt7Wvs46SXg8EoNC8YJI+xuEVLqvvDoRpzqG8a19yzH3c/vwrf+sg13PO1VP2Tk55UFUaltEcvAxZOc0tDDqTTahZTHLS5HBAufTbgGxXDSq8BXKGzSYdNAo1vBcYvLK6aO6SceXoeRVAaJqMWNQkbI31wpg2MR4qAZyQJKqQZxRSKqzA99Po8vy3N5H/znYDxiar+luPHU6BUqTZiBZSJQADhpVkFr4bwJpYiHraxcAZSs2ndKI/8VZdPhTjy5+Zj0Tiug366YWuXMI6W9N1/YAMBLQfn2X7fjt2sO4V0/fYWneuYpJbwd4nPlYKgetLJEKjjXAg5yVHg9kTIdptIqiRQdNUrR05Nye2XOMXetxG1Nz+jzQ9dP2aMU/UE0Mi2PRw7rB2WycqRyLagdVJoGB+ksfVzoNEICIJL63NTWQLaoTSqtMkx+u3xfWUIGxyLEeqSAMDoCkdBjlgoA6xFsfmB1EHBHXVMPVOx7qGqHJEhHgG9UaoYaORZS7qMOmarnHnDmRhBXGXnNBzQiQR6hOdlUKwU6S5U0fVJFg9qrVnoEvAhmKp1RlSBuGS9qQNdbdB8RelZJ9XL+Buk+NYJL9edRQDRFh+HNXX3eq+8E5H4xjeC0Sstda4y/k1X3O3SqH//6+BZyv2TAlRhB37KjFSv2nkJ5IoLvvWs6AK8YTm1JHHkR2T6/dHIFfnHTXOlacdzmdtX/Fin/ie5BzPy3Z/DB+0ZHxSHykw6PAhxTHbli5e0g6hXAcaSPBoxjrwvBm/cMkBPpQR53eXN/dv1ZgRxqVLXmbGch+swlzGcjxAGu7oEkegYduhG1+BM970Ma6GWbhuaspPYYRv+wv70PfVnSKin7hdm1pmLTiPcDzvmtICq349OLHAoOkdttr1LsiRPyS2mV+vr2A8Ko7AAyUjikfxcTmpDf/f/CfSQPGd44Ebdcpr9J+gFh/NjfcwHiuO3qgsw/eXE3ntx8TEqrBIBDAXyNb3UJjTZ/9o2WuXPnZtasWRN84/8VWfcQsO1xAE7VG1ZC9eLmCvQNJ6X87snVBciLWBLHwQWN5RhKpvh95YkIpo0pxImeQU6QWRizMaeuGDtO9Eil7KdUF2D78W6pNLAqTZUJjCmKSUANE/YMwFHQjNfICIWQzmRQkhdGZUGURyhFbBPnNpRK91qmgQsmlkntqCqIYnJ1AdYcOMW5GsaV5qGhLA/L97bzdMDJ1QWoKoiiZcdJ7gee7hK6ixWFaorjKMsPc9Jz0whhYWM5jncP8raFLQPnTSiT+igetjC3vhhrD3TwsPvCWBhz6oqwdHdbzp66eNhCTXGMhyI3VyVQmhfhpPsN5fkYWxTD0l2tMI0Q5tQVY/X+U7BNA+dOLEUIIbL/Y2ETC9wS6axP60rjONo5yNvWWJlAaV4YK/a2I2qbSKUz3DCpLoxhUlUCGw934lTfMMryI2jrHUJlQRQ9g0n0DydRGAtzwnHbNHDOhFIs3dnK+zsvYmF+fQk2HO6U0knjYQsjqTR/l20aOHdCKZbsbOX3NFbko7Iwhpd3eddm1xUjbBpYuc+LkrqouQInhfkMAOc3luNwRz/2u/OItUMcv0TUwtxxJfz7AGcuNFbkY8W+du5tqyuJY0J5vjSms2qLELVNrBA4NBY0lGIomeJcHWweies2aps4p6EUu1t7paqcc+qKsb+9j7ejND+CGWMLsWr/KT632Hi8tKtVSsGI2qZEniuKZRpIptKIhy3OO5FN2BwX119ZfgTTxxZizYEObrxXFEQxtboAy/a0cWN3fFkeKguifB4taCgl5+SkqgJOxA84QHdpXgRLhTG+oLEcvUNJDsqydnX0D/M1WpGIYuqYAhzpGsBON/KwoTwf40ri2NfWx6uWsmubj3RxI72+NA/jy/LQIlQhm1lbhJJ4WNIV5zeWIZnKSGN8fmM5jnUNYI8bjcfW2OGOfh6hV1kQRWVBFJsOd6I4HkZH/zBCoRAubCrHtqPdOOmmODdWJlBTFMPSXW28ktOkqgKU5oexbLfHuzd/fAnSaWDNgVNSOzr6hrBVmPMLm8pxtHOAp9Eyvb6ntRcH3bnG9L+4DsYUxdBcmZDWwaTqAlQXRPHy7ja+RqeMKUB5IoIlO7yxmldfAtMI8T6yTQPnTyxDz1CS70PxiIWz60twsKOf91tJXhgza4rIduw62cuJsidW5KO2OI5X9rRxpwob0yU7W3kKz/SaQpTlRSQOvoubK6QCIKFQCBc1leNo1wCPVmVza197H9cVrC93HO/hB0TWbyv2ncKAu45qiuOYWJEvzfNJVQUoS0QUnVUE0zCkfXlefQkGRlKcrNs0DCxsLMORzgG+DxTEbJxVVyy1jfWROJ9rS+KYWJ6P1QdOodfdD6eMKUBlQh6/upI4nwdMzp1QhpFUGqvdtrH9Vuw31g5xjhfGw5hTW4TW3iGNcLymOC4RnSdiNvLCJt+/z5tYhrBpSLrt4uYKAJB0ysKmchghb29jOlG0e5i+FvfqpsoExhbFpHFm80icW/PHlyIvbEp9SbXj4uYKDKfSfE2yPhKvMX0kzvsJFfmoK47jcOcAdrljysZF3BPm1BUjnclw3QY4+rS+NE+ycVi/sTluuDollcngJXffLIqHMbu2SGoH09fHugbx6vFutGaKUBY3sKjOxLGuAew47rRj2tgClOVHsPNEL492mlNXjMKYLemG2XXFKIrZ0p5wUXMFRoT+iIVN5IUtCRhhukEcl0TUxtxxxVJ/VBVGMbmqAFvXxDtDAAAgAElEQVSPdeOku/5Yf4i6bGFTOcxQSLL5Lmh0+uOV3dl5S+NhC2eP92ySGTVFjg3k7vmsvzLwHJCWaWBqdQE2EtES1YVRjC/Ll4okjS/LQ2leRNLZE8rzUZ6ISPvJhIp8HO8aRN9QEvPqS7jD8LUUsb/Z/M5FRJuN6Z1cpK13SAIRZ9UW8bktngkoGUqm8cqeNhTHw5xfNUh6h5JYvf8U8iIWBkdSSKUzfG6c6h/GRmFdMV227mCHRA2iyjkTSmEZhmSXsL5j6y8/amHeuBK09w3zKBq2x7C1Bjjzo7mqQNon2L4j6p660jgmlOVL87mpKoGiWBirBHt3Vl0RegaTfC9l7e3sH5EKNE0bW4gtR7pQGLNRnohg98le1BQ7PJmb3RRHNh7U2W/bsW6+/7E1Iu4nEyuc4hqiPTd1TCHfNy5sLkcIISxV7FW2/sRvZ2c6cT9helzU7VPHFKIiEZH0+JxxxSiM2qSuFPtyQUMpYrbJx6+6MIpJVc5cZN/FzpHJdIaPPdN3fcMpPg4XNJbDMkLSOZrNj50ne3iBM3Ztxd52HvCgziOmF8VrzNYXr7G9o3NgmJ8xmH4+eKofe1z9Oau2CMXxMHae6OF6nH272B+sHaIddXFzBf+mikQUNSUxrHPPylWFURzvGuTnjVAohIVNZTAYVFg1Dbj0dryVJBQKrc1kMnPJv50Bx95gefYbwLK73uhWnJEzckbOyBk5I2fkjJyRM3JGzsgZOSNn5Iw4Mn4h8KEn3uhWvKbiB4699m6MMzI6mXU9MM4hhnxi0xE8ts4hN3zgpvk41T+Ez/9uI7/1tksaUV0Yw5f+sIlf+/mH5uFU3zD++ffOfQvGl+ITF07AuoMduPuFXQCAaWMK8YXFzXhk1UE8ve04AOCipgrcdG49HlqxH8+7VRdvu6QRs2uL8YVHN6LNJbP/+AUNOG9CGW55aC2GUg4ifcvCBty3dB9SmQzuvXEubMPAPz+2UQtXP3t8KebVl+DHLzrtqC6M4rvXzMBTW4/hN6udSkBleRHc8d6ZeHjFfjzntmPRpArcuKAe331yOw9jf+9ZtXj79Gp85Y+bcbTLQcs/s6gRc+qK8clfreWo/ZeumISCmIWv/HELb8d1c2vRWJnAt/7q5MkXRm3cdd1srN5/Cj9p2e20rSCK775rBn675hCe3OKkoZ1VV4xPL2rEvz+5DTvdKjiz64px26JGqY/eMWMMlu9tl7yozZUJ3HLhBHz+dxtQFHOIur3IljhuWFDP2wM4xK1HOgdQFLNx57Wz8V9L9mDlvnbcsrABYcvEf7pjqQobn9uf2Ir97X346pWT8V8te3jFnI+d34CZtYX49CPrYYQcHjbbNFAYtdHWN4R3zxmLx9YdgRkK4ZMXTcCPX9wtPf/yKVV8zpTlR3DHe2bijmd3cO/PxPJ8fO3tU/CfL+zCWqW6SUUiwgnpz2koxS0LJ+DjD63hnqmPnd+AcyaU4qNuhUQA+OY/TEN+1MLnf7cBgMNv87Prz8KGQx2483mvD+6/aR6e3nIcv1njzKOGsnx846op+M3qg3hqq9PeSVUJfOmKyfhJyy5Omnv1zDG4ZnYNvvHnLdxD/Z45Nbhqxhh87nfr0eGmCn/9qikoy4vgtt+u5++867rZ6Ogbxu1PODwkBVEbd183G89tP4GHVx4AANQUxfHtd07D4xuO4I8uUSkA3H3dbPxy+X6scb00bJ1+52/beLTGFVOrcN28Otz2m/Xocj32HzmvHnUlefydAHDrxY0YGknh3pe9Yg5fvKIZHX0juFco8NBYkS/xkbE5Lq4/Ni7ff/pVTjzMdMMXH9vIx++D8+uwYEIpPv3IeuRHLHzjqqn44mOebmLy+cua8MNnd/L//7nLmjCtugAf/aXn0Lj/pnk40N7Pv6kiEcH33z0Te1p7+Zo4b0IZPn5BA17e3Yb73O+8bl4trphajb9sOorfu8UJ3j+/DpdPqcJPW3ZjlRtJce3cWrxtWjU+9eu1nKj2q1dORmNFAp94eC0GXa6+e2+ci76hJD772w28bT+/cS5e2tWGB5fvBwDUFsfxraun4aVdJ/GLZc61CxvLcX5jGb7zt+2Ih030D6d439738l687EY1fOS88VjYWC7Nq3+8aCJm1RXh40J/3PHemRhJpvHlP27m1+67cS62Hu3Gj55z+tIyQrjvxnlo2XkSD7zitGPa2EJ84bJmaa6xfhN1PZtXd72wi0frfeqiiZhfX4Iv/H4j10ufu7QRM2uK8fFfruHkrN9/90yErRDvo5J4GD983ywc7RrAV9z2ji/Lw79eNRUv7jjJ+21uXTFuXdQo6dPFU6rwgfl1eHTtIfzVTfX98Ln1uLCpAl/+wyYcc73YNywYh0smVeIff72Op4t86YpJmFRVgJseWMX76IGb5mPzkS78v2ed9PJ42MRPP3AWlu9twz0vOXOGjcvfthzD79bIe84Dr+xHy05nz7lgYhk+en4D/vXPW3DA1QvvmDEG755TI82ZWy+aiNl1RdJ8vv0dUxGzTfyLsC/f8Z6ZONkziO+7qe/FcRs/et9sLNvdxtftlOoCfPHySfjr5mN41K2Mx/roZ0v2YIXrxb5m1lhcPWustA+xve+ffrcB7a6u/+LlzZhSXSiN6U8+MAf9wyluH7B2bDzcyecWa4c4x2fWFOFzlzZh/aEO3PW8vPf8w8wx+LNQ3a++1In6YX350w/OQdy28PXHt/DUjAdumg8A+Oxv1/OUfXaNjens2mLcdkkjTvQM4l8ec/ryyunVeN9ZtVi1/xR+6u7VHzmvHgsbK7B0Vyt+7lbP/dA59bi4uQK/W3MIf3Pn2w/eMxPl+RHc9fwurD/UIb3ztt+u55ElD9w0HwMjKXzyV2ulPuofSeJTv1oHwFtXRzr78VW3GvcH59fhsilVUr/detFEzK0vwVNbjvG96fZ3TMVIKoPv/M3b79lvRf3/sw+ehaht8v6wTQP33jAX6UyGVxGePrYQ/3RZM4529XMbh83d5Xvb8MhLW/FV+2F0hyvR+N5vYdmeU3xP+KfFzZg+thCPrDzAdcPt75iK+tI83PX8Lmw41IGMcO32J7Zgf7s3foMjKXzC7aOaojgqCiJSlWu2rl7Z24b/dtcf24Of2XacF525ZFIFblhQj18s24uX3MrVTI8/tu4wntjkzK2ff2gezFAIX/3jZhxxbb6f3zgXw6kMHysA+I93z8BQMo1vPO70x7iSOP7tH6bhVysP4NntJ/D++XWIWgbud/UmANx17WyMpNL4wu+9PYxFzqty+ZQqvGtODW552FvzH5hfhxk1RZIt/skLJ6C5KiHtJ7dd0ogXd5zEpsNd+OwljZhVqxd8oiSVTmNfez82HOzA1bPH+lYHFuc8m9+5iLg/njehDB8+rx7rDnbg3pf24dZFEzGzho7sWrP/FH7c4tmJl02pxLNuAR5m/2STZ7cdx6/ceXDfjXMDiysATnrX1x/fgpqiOLoHR9A9OIK7rpuNwqiNX606wN8NAPPrS/Cpiybijmd2cOqPSVUJjKQyONE9yNNn775uNvIjFl9XzM4EPH3E5tH24934j6deBQDkhS385ANzsGp/O37a4lTovnRSBa5fUC/tV8weFfU4szPF+XzLwgZMrirAZ3/nzZnb3zEVrT1D/GwCAD9+/2zsONHLzwFMNzC7f1xJHAdO9eM9Z9VgfGkefvCMs+9c1FSOm84dL9nFFzdX4EPn1OP+ZfuwxI2e+pcrJmFyVYF0rvnIeeNRlh/mexjg7DGPrT2CPW29uGVhA99nLSPEU13rS+O4/R3TJFvoouYK3HROPTYe6cSPXPuQ2Ucr9rXzaufMBhHthm9ePQ11xXFSV375j5t4Rdq7rp2NwpjNx+/8iWX42PkOFQjT99fMHourZ47FUDLN1/NXr5zCo5PZev7Z9WchapmSLcHWlnhmZtdEW1nd19g5Qbx2YWM5PnzeeOnaLQsbcE5DGXad7OV7xdffPgUTyvOldcNsWdHG++H7ZqIkHiHbIZ65HrhpvnaWUqW+NI6IZWLHiR58YXEzpo0pdP4QL8n6m7einAHH3mgpb3L+B2D/oV1oSbsHy6bFyPQOoSXtbR4fqZ2Piop8tKTdMt0hwGy+HKmuQX5fccFYoGkWutIn0ZJ2ctLN/AqgaR4O7HkVLWlHCdUVjwOapmHfjm1oSbsGZs08oKkCayNh7Otx0jzePWY20DQGy80MukccxX/t2DnYFK9Aa88Q2qovRGHMxqNdeghzRXENmmureDsmRRJA00KcaNuPlrRzMG6I5gFNF2HfTq8d44vrgaap2PpSIV5OO8r1gsopQNN4rItE8Wq6R2rvchPoHHLe/+lx5yKWH0ZL2suVv7xmOgbGFqIl7fCFVdtRoOkSdCZPoCXthLY2u207uM/ro/yCaqBpDra+VIBlaWeDy3OvrYuEsdfto3mVzfjGO8dhxu3P8HcuXjAdaKxCSzoEyCn5aEAeLqmYytsDAHCj82vCMaBpEfZs3IiW9GFcWToD+9r7eB+qcnh7Pr7VNA2/60qjP53CnXMuw9rly3Cg1zFq3zV2NkJNFWgRMkBrCmP4p8VN+NxvN+LkkQJsS1egqiCK3rrpvD8AZ37NqmpEyxZHkTZEnLHauaIELWnHIBnMKwGazsHOteVoSXsHpm9dPRUPLj+A3V3OZlZdUgc0TccKI4Medx69a+xsmM1j8FImyYsnDDcsRCYedvoNQKFpA02L0RNq5X0QNg2Emi7H8dZ9aEk7m0h/3GnHgb3e+GXyy4Gm+di5zmvb3PImoKkRG2N52Jh2QpfPr5wMNDVgtW3hcNoxWv65/nygKC71GxoXY6hrEC1pp0BDleXMo9bOg2hJO0DBtFgB0HQBjhzdjZa0Z1TEp74NO7du5O0oLawBmmZic0sCK9LO4E8tmwA0TcIq28SxfmfDv3rMTIwZW8jfCQA3N5yNgaEUWl4q5Nduqz8XHZ0DaEk718KWgUOpGPakvcnH5rjYb2VuO7a+XIQlacdQqi+pB5qmYn0kgl3u+F1WPQ2hprFoSQPxlIlbKs+RdBOTTzecg5a0x/dxc+3ZMCaUoiXtGOGmEUKo6XIMHe/h3zQh7Myr/ngXXxOVhbVA0wy09x/h33Rx5VSgqR6Hj+1BS9oxVi9xr+0QxvhCV1esskJoG3SAg8/XnwfUFGG5CXQNO7rCbLocmYERPteMEGBOugJtvYfQknYOMTNihUDT+WjrPYyWtHOQqimuw9z6OrSko4CbkT4j37lvz5ZNaEkf4mOHphppXn2wZi6s5greHwCQnnAJksk0WtLedmw2X44es533ZdwygabFaO/x2mHmOXpdnGvVxU6/ibp+Sqkzr3asLkVL2jGmPjD2LKCpCmvDYexz58hHa+cDjeVYZqTR74JBqYmLANvkfVTr6qfhtj7e3u5oEdB0Hlq7vX5LFI4BmmZL67G5pAFomoxD+3egJe0Y/ldVO320LhLFDlevv616OtBUh5XC+H2m/lygrhgtaWGfaVqMXrMdLWmHS7PMCgNNl6Fz6BifM2zOHz2xFy3p7QA8PbZ3+xa0pB1Qe0yxo582xvKwwdULc1xdIc6Z62vmwmiuwJJMinOaDDcshB2x+L4MAOnGS9DT2sfbVmM7/dYmzOdMnqOfjgltay51+mjXBkf/A8D8imagaSK2LCnAK+4+9KFaZ+9bE7G5rv/HcecA9SVYY9vYn3auGU2XIzkwzNcqa0dPqI3PLdaOVmFuRRKVQNNcdAp2BJO55U2enQJgWrgAqaJi3pdoXAyELay2vb0aTYsBAC+HDLSlh6RrbEztPOedw6f6eV9OK5kINDXj1PBxvjddVeXMmdbew2hJOwf4K6tmAE212CfMt5GGRUBRDJuXe/sVe+cypR2hkRRaXEB4rOX0EYaTaHE3pvICR08OnOzltsUiV/ecFPqNrasjJ4W9adwFGEmlpf1+sTvHN76QwCpX/4eaFwOWyfsjahpA02IYAFrSzp4ZijtjNdTWx9tRXeTqysGj+HO6EH8bmo9zaivxUNMCd745qUUfrzkbmFiG/bu2oyW9l7cNYwqwfWUpXnR1w+D4C4CqAqy2Y9ia7vb6KOn10ZRIAcbGY16/AqgPx4Gmi9E+cJTP8VReGdB0No63ezbfhJLxQNMU7N66GS1pFzBz+/KAoBuMpsVAKIS1gs1nNF8OI5Xm7QCA1IRFGB5O8v6YGXV08cE9zneeXTQJx7oGuG4CgPXReShPROQ9zMsek2Ry2QRYk5olnX1p1TSkJlZIa/4DNWchU1/CdSUAfHL8Auw7eRAtB4/i6pKZmNVUg7beIdy7dC+uP3scry5+oL0Pv151ELdePBFHOwfx9ruXIpmOAKjC4PjJuH7BOLpxANZu8vYcNr9FGUml8b57lqMiEcE9NzhBEv3DSfxN4EBr2QV8Z1cKQAGAmTixrRB/XnS+9iwAOD54VLIT2zsKsTk9BgBQMhTGt5suA+AUkdp+rBtfuXIy55BatWsbWtIOFcjJqgsxpiiGIOk91o2WtI1mO4Gu5AiOpwfRP24RCgtjeOGV1WhJe+mMxYXO+WfrsiK0uPbMhxbPw8WTKrDojhbsHXD2ukzjYhgxm6+rOExNH02LOLZcX/gUWtJOO8vtCNB0KTpGjvN1xebzuoiJY4OOQdDP7OING9CSdhxX51ZMApomYH00hm0dzmR739g5wIRSac4MjV+I7o5+qY9DTZej2z7FdXG+bQFNizE0vgotrYdgngohlc7goqopKKlK8H2ntsg554n7cL17vhLX36117t4RtrG/x9k73jlmFvJKYvxZgLPHHDu8Dy0nj6P/QAlWpQtx2ZRKXNBYhm887qzv2THHHtgnPL/ObUd3qJXvO//g7v1tg0d5X364Zj7QVI59gv0yVL8QqExg44uerkTTYsA2scaOYFfadQA3LgbiNh+/soSjswFguWHgZHoI88qcvTQzkuLr+bY6x7YYbOvj6zkzcTEQNnFS2K/Z/Ni5zbMb2LUVpo2D7p6rzqMyw7FLxGtVBY7OFq9dUzULaBqLvlgn3ys+V3ceUFsk6U/W3n27PT2ebLgEKIhileXt/awdSzIpDDN9qZylmNSXxrkjZHasCI0V+Wg5dhiXxaZhWlN23fNWlmDY/oy8bqImuNKVsvxJlf2IackqbBKJsk6oyquJiYTJloGyfGfxtvUM80qJqsRsUyMlVttLFQYgSSTZNxAVLFViZZW4MhGVq1XyyikEcTPVjrDyfLUdEctAQdSW2huxdLLoCxqdvPP+oRTPnb98aiXeOWuM1g72nqFkCqsF3jkmLF9998levP/eFegfTqEsP4yiuEJAb+qV1IriNorjDjDBoqcqC6PafU/etpAkJ2eV+pz2MjJ779sXTarA9QvGkf0mkjlHyD43JXLubGMsvhugCzZQY+pVxwsee7WqalghBWd/pyryidecghCGtobUbwibpvQ39ryIUh0uYhkaKXbEkotfREwD18weK7efIuTPsr7Fv7H/Fknou13DWq2QpxLyhy2DJDqlyOElYmuigiVJ0k+Q+fuNu0p2SrWD0jFyZTW9/6O8PLp8n/48RtIvv5e6hxwTom3U/LOJ+U3pYnI/0arUieTDug4nn2Xo69EbZ/lbne/KTRerIpLwUroiTOhrX11PrlH5WigU0sjrVdLmiKLHqL2E7G+CMJ8iNQ9nGStA3/uD9mCb2A/ZfVS1SlUfpdP0OqUkFxoPqhy9XKlMv8+rhpZbO1ShikSQBNAEIT9ZRU14NUXIT1XozSWKhhc28KnmloTF9xWK7Nog+o1qh0a4L+o2ghSf1D2Efvaqr+p6XFTQIYooO6QXVlELZ7BnsXk6lExxTqOoq7cHR9IkJ1UiYmHfd69EbYkH2oRNp6qq2EcRoqp2xDa1ataxsCkQ8rtRsI9twj1L9krRb1fd/TLuWbIXP3lxD+5+fpdUAfJAu+JdVcSPWwsAthzpwvqDnXh66wm+b2ez2Zmk0tnXqfonkX/sVJ9Dtt81MIIvPLoRP395n8TVJnIiHncjhQdHUvjmE9tw/7J9vkT9oZC3TyTdynonemS+5Ahh41D2vqqv04ReCoH9zv9swgnoBbuYKvDinbkUu4ooKqPqWK1QkPvfUfedbLziYStwrVF2N1nV0tJtz7BpYEyhszYYL+Slkyvk7zTk9QcIJPIB1X69KuXQ7gva61R7Xa5WyXQJpH+d/9b1vR+vvVrpOFhy34eALNWPiT1Gquw7CkJ+qvDE1LGeo902DFQVOMVMgvTEW1mCd+Mz8rqJqqBVpaluyJSBzBYWfXjRNwybqFwkK2EadCjLd8CVtr4hTqoaVQ+MtlJxkihdz4AWCrgjD6lUpTNFaar9lhex6P6gDmkBwEE2cMz5BrmP1A140SSHILFvOMmJlRNRG+NKvUgbDkC5G0v3YBKbDnchFALmjvM8n81VCc04ayjLl56RrR3F8TB/PgszryqIaP1WXRSVNjgVuJPb610rz49oh/soH2fi8Cn2uRVS5rgf6EXMmSDQi7jGDSplHgVVC+QH2QAwL8+tZERXxxOMFps+tGsguWnqgJmtG0+fuHCCRI5L9RsHnQOAgogwj5LpDDrcVK6KRFRqh0o4rLadbehUZTxpjXKDSNBZZLVK3XCiAHd1DajvFn8n9gV3GCh6Rz8YUaADoRey6OIggIua35aPzgoCRKjS6tmqIlKV8XIGoS1D+23Q3uQHcqlCfjtViZHYcyjnCLkOAseGPrCTQJj0Tfrhy29MpbEnwEdeoVCpHGkH/M6rYKk/i6pWqe7zGeWd1G+YUIdQJuxX1PqWKnMRhxmL+HZ/cExuR2CFTOIgR4FNBvFbI6RXWGTzTaqIOQowT6o0R1StZc8Pci5Q38rGXh0p8TtDoZBWTd1X9xB2G+WkoUTtFcMISVVH1b2avYvtsQMjKWw/5kSezXbTGg939PMKy6I0lOchFAqRdo9qu+j7sqwrnTbo4BhLJd1yxAHsdhzvQY9rh2041IG40q8jKf/DONuLswkj9wbAi9uIBPCinNPgRHXtae3NCpClieuTqhK8Ouie1l4p7foloQjTNoFQnlVY//PGo/jFsn34tye24d6le9Hp8z1Mz7Lqmurh3c82FOeIutb8MHvJtqAcTe5/RyVwTN9jKKCKsiVs09B0rGZ7mmyOy3MlHjElYMlvrVF6Vt0D1EqaEdtAtRLt11xVQO7zogPRqwCs655QEGBG6Hb2GLmqpbIvCYPKfmpSQJhyzxstMnDn/EsBZlQV41y+gbKlmiq8KEXbCiEa9hwL9y/bhz+tP6L95q0u2XekM/K6i7rnqEpT9ZCxQ7mo+Ewf45ryZNNGeLDh7EWODfEc55piL/wWcDYLyrOvRrio35DrgYn6frWPACBum1kOqPqmF1EABvWdbDOiQJ6I4jlSx6/SReP7h1Pcg5eIWlLJYzXCav3BDgyn0miuTKCiwEvLiFomfnbDHOn55YmI1B7WdhWoKozZWuRLVUGUBmMDgDDKC8WeHQgqZpmXlEeLBNUC5nOE/9bfUKLG2Ym4kHca0whpfau1jXg+M3SC5hYZRWnpBgplmEcUL1/YciJ+rhaiEm0fQCTou9R51OaueTbnmDAgUP1OLm6XUgCGfJDTD1V+HkQ6ClS/TzWCTcLwpeaGpcxJPXKP0lmELibmrm0agQCrn74mQS8lihWQAUmLAHbpSE714JmbYU1F8ozGKFe/IXvkmP7OoOgs0pnhsyeQziJlLasRI876E77d0sfFDwQNBDzJueo8Jy0dCkJZ9lYKKNW/ncJrLGUsMpmMdJ8fyOMTkMINewrgkg8C2eeRpEN8TgqqHlCjk9Tf80OV8Ol021g7DOmaDprQ+ihI2FPId1KRf9TBM8C5wOaDb5RfJoMvXjEJCcEZ4tljIf9rPiB2rqI6mqjnsz7e39aH3qEkyvLDGFvsHOy//dft+OLvN0GVfBfIyiVbgAIOVNszaht8T2TVsVmVVCZ/ErhJd5/s4/y5TCgQj8nJ7kGsFDIL1Arqv1t9CN/8i8d196oLjj291UmHvXJ6Ff9b2DTwyM0LMKYwisGRNK8IzSSdzuC/WvZI72MyubqAV+R7fvtJ/Fn4pue3n0Qmk8E3n9iGQ6e8b2GRY2JmxPef2oFZ33wWJ3sGcbijHz9/eZ/UH6x/T/UN47bfrNd4jnPehxVnht8ypOxMas+NElkREmCdZV+3TEOzQ0RbjookV+c4k3jYlHWPT5YM9Q2qc5mKHCuIyjZezDaVfVjfX/0cm6Jq9APRxIgtrqOJ5zERx5T9hQNywt8MYp94LYGy0T6Lik6moqQpp0wur1Jtqac/uxB5ETnKj437ia5B/NsT2/DZ327Q1tpbXWiL84y8MaIYI6FQSNukKaVPRo5REVYU6GXpSigweso2eeRYe98wB8fGKh6FWFiNHCMADCJahkxXow6VRAoK5YnRI9j0A6+XHuj/7R4Qpl+Txso2tfQpJ73T4QZgZdMTEUsKx1ZBugNuHnhdSZynbrH7Fk2qRGOFlzvOQSni0BcRrhXFbelZAFBVGNMMO9Vz5G3I8sad9Z2WPs5h4pqtGNOUt1ucp6QBlGOUGAmiEamhYdPQQpSd8dTnqdxHIelfwPMoqvNUbCMgjl+AgUIZ5papGe/q88OUN5OKJvMBH9k8au2lwTHngJAd1GB/oQBQSmeZhAFnBkQm+AFJqlCHQjqtUmibQUWOmW7bxOflDn4EzQ0SwPUBloLS7znIEwA02maIXo85pmiSoBfxzuCIUnr8yG8n+o1aG1SkDd3n4tzS9wk13ZqlYFFto4E7f/3hpZ7qex8FCmjpcMSzcokYBOi0ShV0ygY0UdiKX+QYk6DIJtZMNUJOu2+UgAsT9is1Ukp7vu8173nOfFD21tMEx5hIoJcPuEm1Lah/PXDMvw1jimJ48rMXeO8kHbPscE7MN2IOUkIdLtW5Kh3EeVSNcyL88ewAACAASURBVO24G11UWRDV9k1VmF1E0T+ozi1tD3BtF7G9EcvkoMnSXa340mM6IPeMS5QOAG29Q1i9XwafDnf0qz/h8o+/Xif9/yEFeGPFDZisO9iBwZEUXt7dhlAIeNu0aq+tbjsnVjpRJPtaPXBs9f5TuPWRdfiPp17FIy4xuChR28SlUyoBOAUC1hzogBEC8sIm9rb14a+bj+Fp4TsB4IfP7sQ7/vNlPLr2sPa8h1ccxBV3LsW3/rIN9y/bz68zm+vel/bi8Q1Htd9RkV0UmK7qNT8gmFpXlK6knNwkvQFhf6n7gnS+8NkjVSddPGyRUVekQ5FwOIjrlKLvCFuGFCHH7xPPBCxyU4wc40CYqHsIHeVeo3Q7JTzbibDXRVFTJ2VniPwv4KXUvhFiGnrbyD2G6COyD9SgG0UH1pfFtXMe01ntfV4U52Pr9HX6Vhb/neKMvK5CqWfZqDdlj69PdAUdESYqZeKQQ3DkkGmVpoFSKXLM8QAxrxyTqGWQhj8Z1UAcqOV3EqBflpRP1QsZC5tZUm3EdxL9RoAfVNQBjyYLuC9qm4i7XkTmNUtE7SzgmHONeQ3LExHSYBM3Kp66KCo64tBXFAvrkWOFEa39TnqB/4bMjQUS5PHvI3bQpLxrTELEdX/QizBkyMPy6YEp1PrLBUxWr2U7ZGu/tYj0jSxea3KsJE6z3NpLggLKPGJepAoFHFMPKmrb2d5NrfcgEIaKJrOIwxcFdHjAge6AYMIMJ8kLmqUdEcJAdO7zB7lIh4YRIqO6ZIOZ0k+0YS0+Q3w/6Rwh3uHHH8W8t8GpkTqg4xdNRnNs6X2pCp3yqK8psT84GB7AV0YBzHYWUNEmdFggvxixH5KR2tTcIvZvDrhAFpkrStd/PHqIGFMqCkw7WI6Cg4VKyWLCDiQGcRAQx4/yplPt9Y0cy7nFsvhxfQFe29WILfWAx/T/aPlr2GPpaDX/AzDJOUYcUL20yuxtY3+hIxyJtRww79lvcz2SivepQLSXVun0cbvryMmPWNrBflJVgjt6gQD6B8UWN5RxpZx2UdsDEzYe7uJVolm7d5/sxZ7WPhTGbMyuc4pLqKmChzsGsP1YNxb/aAlm3P40B9O6BkZ4FW4mw8k09rf18WrijJuLcYP+Yd0RfO/JVzGcTGNSVQHGFHm0CKy/WBpovxCx9d6fLcffNsvglihR28CcOidl9UB7PzIZYGFTOb585WQAwAPL9uNo1wBMI4TvXONUs+wZTEp8ZaLc/fwuTvmxW6i4LUaOUUI5Otk8O124I8jhyuaA6HD2aBYIWyvA5lPtO8+G19eV6uSOh00putXPHqAoK1Sbj8paUNdQVIsck9efc42w3QhdSUaY+QDnTB9T9gGZVklEq3ngmA5KUULhqKPnIcsuYtuovc5L3dfv88ERuWgZaaYeqcj+v8hn+NvVh3LiDH2rSPZZd0Zed6Hmnao0qQVM8koEHV6IqCuTGxWUQpeVJkurbO8bRmt39sixoLRKv4Ns0OGFjIgwnTBlUUlErSyRY+QBUt+AAjczP4BBuBazTc7TcdLts/yoxfO7xWeJfBmAC44RPBhiKHeMiFCi2lYYs7VNryQvogF5YjvEZ1DXTgcs9OtzVahIr8C0Sh8vH2XwUB49v3ZQfUtFsEUpMM83UkV8h6kdrsImDc5Qz6KAbvlZBMDsE0GlgmNi5Fgo5DybaocqJGEucYgPSp2iwCbP+AsGV0Rhd9ARibKxTUWnqt/lB9aorZGjVOR3i98k6VPCEeJxVvkflP34v8R9KJs3lhoDcp1RAAbZXv91EDHl+c6aRQGPJEAuOmRYxCoVOZbj+KljQ6WBkN9E7XNBICgJFma/pkZnBYFNFtUOSz+4eN8n/3+/VElVcrk3iP/LiyCFdu10CflFoaY8T/kkDyn+7VXJ3IHselF6p89xPohzjO191LwMjhzT9UA2oVPcKVvL35bj0WrEO3IIhiABZrbO23vdKP2oJdku08YW4KnPLsTCxnJ+jds9hMMyaH/lwIwEmJkamMAklc7g0bUOWHbJpAokonJxm29dPRW2GULvUBIPvrIfO0/0onswiU/9ah1e3HES1/xkGQDgrHHFKMlzAL7Z33oWF93Rgvf87BX0DiUx6FYdvu9Dc3GWy1n7wCv7AQDz64s1IA/w7MhBJb3TT6K2KaVmAcCMsYWYVesAfluOdiGTAcYURVESD0v3leWHcd7E0qzPrnGd7mLkfrapSdpybM8dJeDBhCxeQqwXinOMcpYFOSdVQn7v3KRf0yPH1CJoxJ5IfgO9Z2nctqap8aFFbZpfU1xrfjyO4rhQZPN+epzpN8peT0t2jHs/BYQxp4wImGV94+hltM8K4lbLlrrvvCv4bWQQhC2PHxvjzn4PHLv3xrNyIvx/q0jwDn1GXjehUg4o7gNVxAnLHkErSOIagfgHgQkRy0BJnrORr9zbjj+4ZH01auSYms5IHXqIKm9+QBgdLeR/X1RN9aIOmuQmRRhAPD1Q9xKRPBVKO1jo9QkeOZYlrVLZgMry1cgxU/qXPV9tL2XEJaK6BzUeptPyqMgj+RoBVBHvpKLauAFPVF1RJSiyya9aZRBJvx9PhSqUQR8E0rExJ/uIWAvUIVX9Le39oeZHbuslEKRTI8dcb3yp4HW3DJ0bQzWuKDJRKuqDE/IT4D11EKKiJnKJBBSF6VGqzyxlbjg8IN5vvSpZuhFDGY5+RgbT4UFpil6FKdHwzT52JnFfLvsLJVQ7gvQwdbCg9iEvUkpfB37tyHXf8Et5IZ0exPP8uE7Y/wsuAmNov891TG2fCLNcUhdlkI46QOnfyUQFzLK9j/Km+xLys4NAAE8WSchPpOP4HapG6wFnhw6Kh0y0XbK1V4scY1F+p+mID44c09vrd6iinpdL08iI3RyBXSpajRLqwKfOI6qSJnsmczDmRyxpP4pxIEy20dT2cBuHsD0pe0atapkNHAOAX604CABYPLUSMaEdFYkIbjinnhe82XS4i39DMp3Bh+9fjb1tTtrjuRNKJRsScKpxHjrVj8GRtPtdJq6dVyvdM298CakX2bO++PtN+N6Tr+a0VqKWqUcxRTw+XdaOcSV5UvohAPzm5gW4/uxxWZ8tjj5rW7ZUUyplPqc9N+tfVCc6Rb/i2g0Bldgp/r0wRSFjGLTdRlR01gj5w5bsNCDSJT0AMXifpLIWgiLHKHCMrABMAfNEVJRfWiVFLcFEHFOq6qMqop6hKuX6iR8o5edsoUROq9T1uBdNpv8ml+ZSa15Lq3TnFSv2cda4YkwUSPv/HiR3a/iM/K8L5VX1i8LwC++UIx2YctUVtU0YSmQUl6JcWXrg0S6vlDIFjpHl50nDXzfsaC++/4GDMtCiWbjPAskyczww0YYSBQaZyIvIaZX5GueYDnoBToQOFbYtenGixMZNtTcWNrWDcEwpWkBxhOUeOeZzjYh+8zNMqPSNXMtkU6mLFFBMgUEqYSsTMvWPmruWPi5BUXPZCNHFv7H/Vg0xFZTyi2YMrNKXw7xv63E2zcKY7O1Wf6fOM/Zm6huliAbKu+mXPhTQ/tzAMb39VLqBZRgEHyTzFPvrzmwRiZRQOoYEZ8nDqPjbkNY2CoTxvPI+6VTun0Qbk+ojkieLGGfb1Oeft5aDD8+5ktmT4JhF9FHQOiDGTzOgQ/K7xXbmGg1N8a1RoDkFaKVl2iFSbEOez+K/4rOos4QWTZbJPU3R75ztRWd513Il5CdTgk/Ty+130JEjsYj2ksCdXBQB8Pr3dJNUcuUcE9vGI92Ez2O/NYg16QdkUgA+mQ7No5X1tUDp9lxFbRp16FUdjImoLdtLtm5rURHzXpEkf4coZc8YRohHjgPO/PzzrefxKtK9Q0kYIeCCxnLJDmSRYAxIYlUe7//wPO073z+/TvtWAPjgfSux3f1dzDa1KphNlQnFiWy6/3rXfrZkj5RalU2ithM9K/ZbzDY10K6uNM7PDvy+sO6sFeWZbU7xgMGRFP8GNfWUCaWzvaia7OI314PmONM5cuQYtQ9n3+vUtDk5kofa50PSe5iohPy0Q0rfh+mq30TWgmVoIKjqmKXOkTS/mK4r/a5RQyRyjvkJe5rfniBzjo1OXtu0ShEIc/71o7oARgfmycEeug60TW/+sQq7arTg34P8/X3xm1ioBUYdUH2fwSLHSINbUJAEIb8fD41aUVCtSmcaIUwdUyhdi9kmyANTjiAMmRJGHA5FA5BSzFHLpDeCgIgiKpyZ5I8i2ksBOLGwZ6AwpZOI2oiFgze98kRE2nw5aCJcoyKUKM6xvLBFeJxMpXJd9sgjknNMAnlGl1aZi+Qc2RQwptR9al8A/h4mJmytUWnIMiE/MZ+pqDM/8FEZhyAePOpZFChMHcZJT7nSDsb5kR/xwDHvwBSss0jSUUI/UVEZJH8NEYVA3ednwnh8aPr6oQ53Yt+yMSajhQgPqm+Kh9vK4JRb/RqVVkmBop4O1L81l2gWKr2fBKUCoqK8PYc42PtEAbH/RwPf/t9OAZn+umJ0ezDnrgt0yOj7kC8hf0AqHQc3cyG9J/qNSiun0yrla6M5FuRGyK+vefnwCK1t7HvE5vrp8NPmHCMi2IJ0lGGENGCCr+0cwEJKcnUaUJFuUlozkY7Dn+NXrJJxDwamcOnrj7S13GvUJ9NplcF8aOq+nq+kVVJAmF9l6QihByiHqMqrJ0ZKxWwTM2qK0FjpFVJKRG3kReSq5Wy+qPNmcnWB9M7PX9aEMUUxSdcw6g6RlytKAFXxsOIQdZ+r3rf2gMxrRglru/StYR2QG1cS167FbFOyKUvywnjoo/O1d+xr69OizmbWFuH9872IOC8TRV8Lo02rpPY10jHrrgHZea1TnPhFUaptC0t7OmVH6PMUcPqcIuSn9jVyHybmvQrMRBX7XwVFLd42GXBxfg/hPt0W8ssEoMaI4ollInOOsbOiPgmYLhH/4m+fjU5GS+5vEHtMEDjmvStYaOBfnlsqWK3qhL8Hyf10ekb+94VCxgkDMBeRct2J0FMqDc2vWpt6UKZCo1VFHVWikeiDeHYjP9eUGYojJyRtECHyMEe2LQjAIDYlCkiiyJyjtqmBimqKIxXmCgDl+RH5+SxyjNiQKUBLbEc87ICWoo6Nh60skV76nBE3Pb+IMNpDob/D7zDnHYJ1g5saF5JHjriPMhZOVyguCIq/LShazS/dSwqrNkK+wJX4LIqnjix+QQA4YZ9rw27JeJFnhB18/dJk/TxbFBcOnY6jG2JUkQ95zQebDcyIkb6bihzj80/nGKHAe0rfjTb8nfI6U8CSHwAaGKHEAMRRIwfUXkKsM24gi3NX/y0FQGUbP+oQESbeSUeOUUCmvvYoZ0MuoEZwhLQ+VmQajJn9GsXXkguvF0VYTc2PXNIqR5Oi6JtWSXC/kBUWs0RnjUZyiWCjrgWlfNKk987hUU23Oz2h3qnrwGwHW9Yer23OfdS45JSeS+hFSveQ1whnyGsRgcgeovZxfsSSnIkkOEbxmZLO2uzXUkrjooQDMz/i2YEMzBLvYxkG4rXyRAT5EYvfD3jgmfgNE4UK5vy9tqnZ7HlhS9J3PHJMuW/lPrl6JiWs38RDdMzW+dbKExGtHfGwfF/MNnFBYzn+9R1TpPvee1aNBqzdee0s1JbE+f+nUgY9Qv7R6Qj19/Lzdf0pOrm99M4AUJjpce2d3j3sb5R9LvYb4yozqTVJAdaG/g020V5VH0pFwKg9kkgz9XSld41MjyfSL/1UOxW1zkQ+F7rt8DO8Tm96vOZiEmdVkquNaHAujgTqHCIXTwiRZ/m/N/mfnQrPyGsquRgjQcIMVSoiQDq8EJ5nP46csHJoUMk342FTS+2KKWmV1AGEkyMHpMeQkUEMHIMuacJroIpatjhbO2SQx5DuF/9OAWZS2LplIB7RwTEprZKBb4RRQRF+0oT8/ofqeMRyK1HKRhv17RS/GM1jpRv+cvRedmMyFwmK8iDTKqk5QwBmuYB0vm0jDtRUugIZnRUQHaQSkTOhgB8KCCMBRAIkp1M+qb6UxywhRo6RLVXa7fMtQUT7noHl/ZZKB8glhZIS9oigNEWvGpo4xm6fSU4JXXd6xnB2S8zjjdS/XYr05RE/1BhTIIzevybR3tEK5WzgoJToUTb0PvKrzuj/Th2EYP9Fc6no+olKI6SiQ6joNN8IRNYO6vBBAoj6uNCUAvq6pVRWbqAGZQuIe6s+P5io19KZLBQPZNsCmybv18ThwCLaZhGHo9dS2NOpNEVSRykgWigUkkANLwXv9PYcmVAaWjt4WiypP3VQkdr7ctLnVDsComWoCsDZOAXF5+baNiryFtAJ+RnHVxBg5geEiW3j4LSS1iw6MJmNlifYgXECCGP3iWDQ+NI895rwW/e/xbZNKNfBMTJyLCLTjWSLHFu5t117nirMLhZ/G3fpO8TlnBexpG8KhZz3xqQxcJ5VLBD3l+WH8f33zNDaVpYfJkEjah8+XS5xm4iOpAqaRAmnMRWhO9q9js11yjEmzlM2p+TIMcI+8gEQqbaptkqESE0OE/u8lLXAIsIInsggfeTn2OSE/ITNJ+pWSn97f9PBpjeSeF7cXsnIsRCzQXQtmEuzqf5Wz3QUr9zfm5yeNXxG/lfktSgRy+4miaGpiDDpsJX9QB0Kyc+L2zLIwyKiRAWeH7VIw18+lLiADuF5oEOS9fuoLkrlYIVTRlxwNJl+jQKDKA4IyzQQV5SMk1apG83Spud61iiDLUoYFUFRRdzbaMvXQoS3JggEpKKMeNSSaLSwan6EAZ8L5xgVanw6HF5+HESnK1T0DZU6TAN3Yjv0teB574J3PRrc1J9FVg4LOgwQ8x6QI8dywhZ9PkNOT8puOJkEUEpFDJKv8mkjmQ5HzCuvVLk+dkEGZ04pHu6/wamLenv90yp1g15sxmhSiSmRvenEIYWIZvEranG6viK6WqWui3KP5NTb5idU9DIzvMnKZwEOJDJFlUfajBLUCLHf6+2g5mlu1SpfI76VHHUDRUhsjRLYHTUhPwPpqDTFHNNgREAklzlOdYdvVBuhAylCfmrNU83wa1uuaWj+h3N9Pud8JM2hbVpaZcQiASjKyUHufcReTTVDjRwTI4qY7SVmEDDQVAR+2FwRgSRW/Ebcc9l/i986gYgci1iGBKqZbvQ5zTkm99tGtxiAn1C/jdmOTSnavPkRSzo7xN17ogRAKV4rzYsgFAohJnxD2DKQH7FoB6MU7cv0sd7uXMAEGWDW7QGml8lqlQHRw9T5ShUekU+klovALptTJCerWMnW1PdhTmNB7HV+kWNesIAQwcYzTER7LiS1W7oW0q/lWgGZ6bcghyh3pr0BoNdoX0lVLlczR7L+dpQvY3aJGGRhmSGNV+5MWuUZeUOFUgijtTs5709gpI2uVChyQ26kK6mLWmh0RN8M8iMWSQJNEb/bhBefTpvTjU6yKlYO2jXXaJkIcY30VhEbiwrSxQXDJhRyjAOyWqXw7UWuBy0o/TJKgCsUPxIzuERFqm4uVC4/iyaQy5wTUVEUWEiEaLMNa7RznE0B6gAZOC4+0WSne8QLAp0pQn4/LyJ1Xy5CefFzrVYZybHfVCAxT4mEDGyjzzX5QCm3T/x7rvyBoxUeASV5PPVx4p5RYj0GpzjqB9TRCB0tSQApZASUDojQh+LRrQS+Hsm2EQfloANDDlE1VP9RxTv8IseC0ouptBJPZ2Vvmx/GSOlJKsJMBnspEC277sxl71MJoJ1n6m2jvkVPqxy9E488qFL3sfYKY0WRNo9GTwa3LfuzyDRFwhEpHahDBDj2P9xzKAmqAOx3qKIAzlwKc4jCAWAiNYuK0qQO4qQQ45HLfFP3qkTUDuQcozhlOQ8ZUY0wF8esGqEPqACXMy+klEQOmHlzJhFlgJl3LUa0V0wzZO03jJAW1aUWlWH9cToHYf5b4VzA0jPFa/kKtxoDu6gMCPE+1l8SWJgXJgrjEDqVA6/6PPLTGtTf2GMpR5P4DWyuBNMbBNsD/7+9Mw+zorj+/rdmZwYYGBiWAYZtEGTfnAFZZHHBJUAIRnHBBYLGNSYmEZdsZo9vzC+vZjGJiVGjRmPiEpNoEvNL3kRU1LiLCgqioijIKuAw9f5xu7qrb59bt3vmDjPM/X6eh2cudftWV3dXnzp16pxTruHY7rsmVFdaUMyWT1Lygs0kU23DiRSGXOx7joVD9SJt83OORZ+VqJcIpWZckOSHfd/MaZ35umIOIXG9pJuLvDNl9JlKJPV4M4eH88MVRJw7mJCftCmJV4EddUirctkSBEseZplWqEuKCkLCs0Jw7+5SVpTm8eOdW9iFrFgQ3tIkx8ZlXElfvQv9zqHEiXl5BA+RbC725rrTlU57xbBzaer+hBPtR1cCu3o7AmbzHDPKRHiyFb0Go1y5vOuMUM7mOSYbwqL3yOWdlXQSbJ67lM9INgq4y/xV4GZ7qrgVoMD1PNrfpH4kJZpuLuEVPTMpkYwC0YmVlMw/LBuiuQmALDnkEg7ehYKSm21l1LWK6E7IL8g/wShvZFU3a6dO84wlQ5ity7R0BVPKsSVOPLN410n5Gg3NTTibzfjryhknGWaavTDUHI8wwTDtMrK6m5b52cohn9HxNvbYJ5wjjieX/R4Gm/i4jSuGXKy+J91JUg6vjsrJuH0mcR+XvAHNRFloWyg0xniOSWGVCWWl04AoeGXHnVTlYmE2yLUn6JlS2owWhMXG8WpLn+B1Lg1vROQbx8SwyqhOJuWOdI8n4TqBwPBj5xwz+pid68v0FdsY1KWsOFJmDGt22yYM6IYzpg6MXFPI2OSdU1o0iBtCZXenToJhrVwwjlWkeXqZPlkmPBd7l0/Jk84sGku6p2iAimHoz0bgHRntu+JiWciDLarf+fmjHM0QDddCXr3OZYLnWELv51AoopF5aceI+VdDekl07uDrblLORrvMX/TJfM02vueYEJYdyjmG6LkOFHHPKOmGrZGQXzpnuq5PzzEax9oVyScDmQWImO9EyFdTIuawyb56AMi5FOwmZTKIhFx8hZAz1wqLhOjannD1PDAM2gaM6GTZZQySSB/YbM+xLt6AH97pxQy0QVnXsmhOikCRsY0wmV/nUM4xT8FwTqK82yB5RUnGsWwTzVKhfzQXycNHSpQaPKtoWVHMSXAcshkFOvnhrpKnUXTSnjSXgGtAlLwviwUjid02KURVeqYVXu66JEjjuuxFEpVFoueYELbT3DBZaZIZeIdEFb1u5YFxzJxTmrA7TybgekekcA45P5VgQBQ8XESvW0cDXK9v9jxnmSfFkjdgc2WFuKOgYAwXd9AT3pekuMJ3pPsRboc0DjkmM83MFSUhyYU4k8fmhFVK9Sb12Ap5jsUICQ6RMJzRhWsDEht7/E66gYALc9+k0EXJcyxuCGVzjWOy96ykZ8Zb0JDulFtGeTnH0upMbX5k60tRzzHJE0vyII0Tgm4m4tIulLYXofGKCnuORY1BnUsFzzFhh/KK0iJ8+WOjgvZ79yHUDu+c0kJTelRIJqSUHpIh0L6uzhk8zaWwSsmYZ9dlFqeyLcJKhu3gO7E5GTF6SUhWep87hfRzz0gnGKzjts3g6uvSJg5iuKTwnOMaoNLbFl5Y0V6dlnz2DS5ZPMcEOZ507PdzsQr3Npx3OvU3znubnYT6eVId2boUyePNNSYnvTxzuG3YLS5UkYWF9BzY+QCNY+2I5iYED9Xh/ZVCrKTVA2mnLPvlixv6ZoTOno/2R36bjphnSfC4CBlmEk4Qkq6e+14jWXYelPKLuUIk0o10tueYWQmUBht7kK/0lICQ27afkD86IEt33R4UTN2u8BtfaApJ3sOrqpKxMDrRLHM+v4xfifdU9lSJPiuXN6CU56C5XprSgCt5joXDuDIr4bl00RZzjgnGbynErDCLV6lRdJOM/bEVBd+YEDUcSwq9VCbhDoeLtk1S/kw7bOOY9IybH+qVuY2SkpQ1Sb9jwpALpFxDshdy9PlJuZcCryiXV02yNoY8JITcjIH8iGuUynwuV9PE1AbSbpUxPSEl4oRV2gQeP1EjnbRKnV671s03pITKHMdLxtPYhugc4G5btI9L2P3J5fnnbIejIdJ7FXdHT3minNRIbox07sUiedOQZC+0676Z7p8+oUv3HOsk5EvyQyiFcVNsh2s88b2iorqAHVZZLhh+KgSvKxNWGQrJLIkusJaXFIZ0EnM/0neSBMLjQKHQXhedshi0/NDQkqghMB1pJ/aQF5ofVhnUZfRiST+XkHpY0h0szW2VImKkxWspdFhc/HScM32DB5twWKVn8BTCzeMuBktDR5w7ZPcjU0e2sV/ymnOlypFwGcek9rVWzrGchlWGQg2iZUnDKuPszJzu5Zf+HtFzjLQpucj/4jpcmgzIW8Qmq9dmb6NDkptzS7v7CaGW0s6UElLTGl2GH4diLubvEQw/kntwnN25TFJVIHCDttnfZM4jhVUKnmNCPgsJ20hnBKgr9NSfeAveJlLujbAxSDAq5lC4SpNxd66lqLKey8FMokRQiEUPRCGsobXaIU5KHMZe+zWRQg5941humxxC9OISVm0lBSspklwwr41krKjsFLzLfo6RwmjbxHMlbZvZUUlopPQexMkdAkAUnokNHcJ5xPdRMM66J+zJ2uFa0bWfhThxcbwHSZG9olJ/Za+2qHEzW1ilIanHj2tslzz/XAZjQ6aFKNfjiztJkXYlFfOwOI0rybzrks6f7Hxopm3SLZH6enONrKHjjHFTyFMX1+NAzjmWGUkvdXmOSYsXkqeihNTsOLIhmnOsKKS7GH0p2+577pQQmSn03+/odYoJ+e3E/UIYYVc/rNIOyYx6gKW31zzbbBNs0z/ieo7ZY5HRQzsJucPCm0vJnmPS+10meJyFwyqjnmMume3y6I2LaHzw/oo5x0K6VvP0EtccMFtCfikthNsjupnWMQvT38IRMd54ksUDuNAhPyXM8WLOrEVt+AAAIABJREFUMeuzOUPSNZSWzIWbixRWGdc7OWkPk9LnFBWmdlcOedTGlAkdCRrH2hG52fVJUlqiCneQDDFurfGUp337YxjHsuR/SZxHKAcTPD8fWrbwQEcIpUS651jfyk7+5y6SccwMLNZAK3uOFUSOc+WKkIyFLgcD0y8kZSBrHjLBK8qpYCacIAQDXXSgDRkyHbvS5cIQLWGqCCe6zWykk/JFSTR3wUvMDycYEKUyuT7bjT+6+pwN1/OUyiTjrDQpjuss0+yQMyG8s7sQVinlGpKV8pYbhYxMljYzCcnYwsxyQXwP4p3ex5WPSXp+krdC0nZIK/5yknCvHWJemuiEQTK8JiXpuyrumiwsFklIOkNzw2LFsErh1BHPMUedmYgbXm2Q83pZ962Zz0pCbEfCtklIOkzid82RVDycSzLqReJqWy51KCVM7iQvMTk5d7JzSkipRYDUYmS2hPzS5jnNNZJLBihTYodVGkOYrb+VSyGJxnNMCNOUFj8NYooR4Uab5sb1ErHlTJln2JPCKu1OniSsLRyOGr1HlWJYpdFJovWJr7WrOeI9yvwDKXdwSwzABtdrGDI8SmGV0iY4Ls8xYfqW9JU07bXnJia8VHQqsK4wyNUc71zBZjJufcAPq3TcaNEQFq8ZzSBzOyTjmIQkn5ubQy/k+dcUNW4yIT9pU86ZORRdSotw6dGHNLuOuPlq/G16JWUr5kpd3Jxn6UirGOEdMqOrjc2eRMVUwuUk75m9CeIKzXTFpF+3wDgm5V+QBJNZMcwW4uhSahoFo6UzrFLqA95fqR2hyXhx9PnlNsdK5u+ybZ1tlJVcGsJCZUI7SgUl3BXqlrRtzkml9Vz8HYla4DGTntsEiA7zOfXKEPN/ScaxzJ4McZG9ZExdluHH+yuFVcY/V7K2uWSyHFZp9XnHinXcvCOu48XjzLmFRQ8xhLKVV2OlnHSS0dzGNZFLuomIVCp5tbl2OJXbESVhVKWPtKovJjBOq785sjSpEi/JgbhJiqV7n1jGOiYzcdvhSg2RC6RJsW0z9J+lY9wKl7X8HklebWJexIKobAjqTYr2zh38srykEKVFhWJuKymsUgoRF8/kuEWi/PCKbN3P9xwrji48dQql4fC8p4Rk/q4cu5KeJ12RH1bZHOOY9xtpMyFXhIIrxDCcq8yEVVrGsfLMG1VJ5CKkzlWFHcVhdGD7/XPl03LJF9diRyi3rb8LffC9FHbo3Kna4VwRF9Pf7DmduImK452P64xRJFyf9AvJ27i9Evb2TfjjpHqlUGbEhWjoziNoHGtHDKgqxzNfORoXzBnmlyVVO+Ounhc5VobFehO2Q0yMKYQGSflOEufqcQwecXcnMYdJoaehfAgJXezTB7bqLqX+Z2kgN4qE3Y4ufkL+6ApZ3IT8kvIUJ6zSRtoFSt7R0zEJznhGGel4lyIR8qJxTHhbezIeCnNwhJjFbUbS3BgSZkIoJkmP2Ta7XzYn55hsJMhcJk2qJI+RuHmW4uRfCB3veEKhsMqE+eJczzNp1xR32JU8AVupz7uN1dLKuaT4xpyx+ydN1rZsO1g6vTaFMnc/ijnmCJtOuHKzSMih/EkNd5mRzp1ef3NkqWQ4cLXNlgOS4S4Il295J08qa+N6jol9LBdxlUI7giTy1kTL0TbZAzHZ+d2LNNExxyY3ibJTSLe0qiIlq7N6jgmTQKeXb8z8kH5d3kOVPMfC+bqiYYRdyqLGIKPzudKISN+5FojjhlDZY2767nY2cTbHkgjvxB69H0HOsWi6ETmxfLRM8jj9/DHDU3+PHh6tw9le9zzBpQO73p24O8ka7yzJM1uSDXHPlRQjS+xFWCUYp8xn126c4cJokRQxItXhpzVI6lkllM08pBoAUFtVHq+OhKItnHIs849z4R0pyTY/LDYUJkzjGGljkljpk672SbvqScIi7gqLS47G3w46avlPukuYS6DHVboCz7Gowh3XqybOxgD24LBzT2P0+Kboqqe0wumHVdqrng4F5aP9yTxEJMz1lQkJ+aWdDeNWn0tPKSk/nHhOaYKeA8VA8jQyfTBkQPQUurjXnnjRM7bBQujjjvdP9hxLO1lCA5QL+32RfpvUDd9F0tw24d0qc6c8JPXOkpKV29fiNK4kKM2E01gtyNOkYaNxcRnSs+VhKU4sszJ/l7SPZ8sv5p6cC2W5eBe8v9L4mV5/ptO5N78Qzum4zlC4sj/JDOrP5QYTEmL13ulDEz5HO8ScYy1tGOA3Lu6mJBJJwyrjeniZstg5EKU6XKFQ4oJGlB6dU4uSkqe9lCtKInmYqaRbp/6Gk+oLnmNCQn5pMcrPH+vyHBMaLrXNdJm4urutU7qeqTPM23FPJU9I+3508xanWpIn8hdnHobeXUvxq7MO88vOn12Hl66eh+nDekaOj7sbrd/PYr5X7jQL8TqetCmJNL9KmqLGGfIpLs5Eyxq9eE1bPro86eJes7m+uDnHXPfZGJ+zcezoPrhjxRTcd+F0v+zkwwYAAD42riZWHS7ijmW5DKu0aRLmefnoOSZnRyTth8QGjGiZtHpuPg+t7oxpdT0wpGfnxPU6k//HHKSSCm+xba7Vu9ir+N7xgttv3N1wpFa4UrDtsIxjVRUl2LJrHw4bVBU5TsovZu6vlONIIunqneg55v3NakD0w6eSTfIlpNBT1+AhGuliDuC5QOqLomt7UeZVxKRNE40Tjve1JSFK4d2RikwDErS1BcfpyAefuPkaXbLC1eelsnBYZTKZlcvcOtl2KIqz+2Om38YhrueYUd7jGnRyYuSRDKpeWUtkVtJzuo4LGTeTGhDFROoOzzHRACx5lqQOlN6J9HetSevE9y12Qv609oTPm+iUIZxOiS0w4BtHFKn+bN4NhprKMry1bQ/qekV1Mpe3gOQxIiEvDMUrcxHbA9hhiM5NKHy05h6i55iwwOjv4hjv/G7jTrTM1Cvu9BjylPLCCK2yrmUZFqMQGB8kxJQDwnF+IvyYuvuHws70zd05PtQ2h+Fc2q1SSt/hqsOmfnAVHr38yEh5JgOhXUVFSSF27duP4X26RH5TmnCxxdXd4so56Z03RUVCuhEJcSyId3pnHaZ7ShsvxPYcEwg2sHHLAclzzWAM1SVFBXjs8rmxdr5sGNIjVPbpWXU4vK4nRtV0FY4PPpcUFmDf/iaMH1AZPU44V+LQ9oQPSzpcCqtMmjakI9CqnmNKqXlKqTVKqVeVUpcJ35cqpe7wvn9UKTWoNdvT0ZA6tj1h84/zDpR2LiooULh1+RRcvXC081xJFaW4O+8FifDjrXomzZHjcm0PlSnzV0XKJK+D6s5BeKQLl2JgJ+R/8JKZ+PXZ9Zh7aK/IcZL7v9M7S5CQHzmsdK5B1cY1GZdyTUiXfpK3wjJvVJ+M7QGAW5Y1YEy/Slx70rjId726ZL73UnJys9unTS4m3t3LS7IfBGt1S5iMZ0skaog75vWtLAMATB3aI/KdnN8j84RGoqW7VcZ1BZd/nPoTf6UxWvNl80YAAD53VDSvo3jdjo5iP38zwZInfMnucWzvJceRcZUZV948iV5dy/zPJn/iDGGF3VQieQTH9aCRysw5B/WIhjQkXW2WPBNaC5fS2pLcjEkN/7HDfoVmjOzbFTeeOTnyXOKGNIfrl8bleJhriD/pzoHB0/GdtKAmURzz2d6yvAFL6gfgulMmxm1ehAKHsUJCDN1q9tkDzBWHPMcc90hazHTdNfu7Pp5sqh8cHftMWKU0OQ9tauTwAE7qWSz1BSm1iKmjrMT2HDNhhHbOscyeLa7Fz7hhlcHmCbmdFsbYoysrpv9Iu1XaPdU1rl4wuw4AcN6soc1uh60vPXbFkVi1ci56enMByQOxJfLWID1baeyXx7DoYot5B6bXpcZt26ATV39xIcllKYWLazOluLJHcq6Q6jBzLXvO9cMlE/DFeSMwsEeFX9araxm6VwR6XaUwd5AoLFCYWNs9awTBw5+fhR+cNB6LJw2IVW8uFiqdeRGFH5j+Ftqtksax3KGUKgRwPYBjAYwEsEQpNTLtsGUAtmqt6wBcC+A7rdWeg5VR/VIWZjtX1U9Pn4SenUvwf0+Z4Jf98qzDMG9Un1C+MkNva0JjcIZfCoatft07RcpyMTkr8o0EQshnTGVSevm7eBN3OwG+QbqWHhXxDB2GOSN64YLZdfj12fV+2ci+0VWDmd7E8bBB3f2yO1ZMwYxhPXHVCcHr0LNzKWYeUi0OREagK6Xw3cVj8a1FY/z7Jd2hxRP7AwA+4f019WdCWinsIRzfX+wDKSQPrz6V0X532KAqPHbFXPzo1EDxv2BOSmlZMXOIXzZ9WE/cd+F01PXq4pfdsqwBn5jYP3ScwfRxSfkd3LMCXzphJG44fZL/3cmHDUCfrmW4yDs3ABw1sjcAYGz/YFVn8sDUc5tY2y3UjpmHVIeen6GvcM29u6buZZFgZJ0xrBozhvXEF+YF+S2kXUzjKij3XTgdP1s6OfTsDdJ7Ul6a+T2VjJD2qqNRTsd4Msp4OrjGc0nZMDLOZqBl/Fg+fTAWjq/xDdKSAiIpWJKOf+a0wXj08rl+nwOCazpE8NToLDwLo6Db12IUCZfh1mbRhH4AUnLEYPrO0Op4HiM1gmwzil11l1JcftwI/J8To8blbPWavn5I76Add593OI4a2RvfXjTGL3vg4hn43aenYtbw4BpOnzIQQ6srfOOstCPduP7dMG9Un9DGM5LnnelX9n2+6ex6zB9XgxtOn+y8LoO9O3B6OyTvOgnXgpOELWuNXDRyxMYYV8ObPaQ+S/lMThjbFwBwtCenAODjXj86wsuDAgDHjOodOW5JfUoZXzZ9cKTe4X2CcevwoT3QpbTIH8tsmfXAxTMwZ0Tv0LsJAFcvHI1zZw1Fj4qSkBwz/bhcyGHUWZA7ktyTkHbdNRjDSJU1np8wNhXqYo/BxpAiySPp0Urv9RDv+mwjiLldQ3pWRI6XJnA13aLjxZDqzvjWorEh3cWMSccIi0rpzwMIZJ807o8f0N1rT3Cl5t7Yff0np01C9/Ji/NQaNy8/LrW4sNL7a9Ora/Rchf7GAFFDtDQeLZ06CMN6dcYVxx3ql0k5sMxkzX5+91wwDdecOE7s45J+5xvJVbRtEkknqnLOU6Feb8S0J6DmnbE9foyMlGSPK+dY3ATwuUhWLkdPRK1jZ0wdCABYPiP6rKRWmGckea5L75U0mR83oBteunoevjAv2nfj0tvq4xWlRSEdN7xbpSejJG9R4S4tnpTS146yZLbRp0b3i84rSgQdSDLGdu2UukeSwaqmWyc8ddVRuOf8aX5ZXC9Dg3QtUle036v7L5yOP5w/TcyPapD6ohyVYeqXZGsgP790wkh8b/FYjOgTzCfmj6vBp7MYSmcM64lPzRiMn5yWfKHCjBn2vLBft05YOKFf6PrMXHX8gG5Ix+ifPTpH5Zc0rkrGLpP+RM6DF223v6FCnu9W2ZphlfUAXtVarwMApdTtABYAeME6ZgGAr3if7wJwnVJK6VzENnQQvjZ/FPp37+THNAMpBenokb1DSvXs4b0we3jY6+iOFVPw7JvbMNVyAT1+TF+8u2OP0/Np0cT+uP2xN0Lx0587ejj2ftSEk+qDdhzatyue2vCBqAjaL+Ls4dV4eM1mzB8f1HfhnDo8vXEbJg00iloBTpzUH0WFyr+uqUN7oH5QFQ4bHCi1kwdV4Wf/ei1kjLlobh3OveVJnGMZTW5bMQXXPvQyrjg+ULLuPu9w3P3kRlw0NzAg3njmZPzn1fdx9MhA6Txr2iC89cGHvjHIGCZtIV5QoHDpMeGEnctnDEZJYUHI++trC0dj4sDuOH5MX7+sYUiPiFuuxBXHHYonN2zFtLrAM+OTk8MrDhMHdsfY/pUhwVrboxwvXT0v9AyuOmEk9jY24VOWMnLLsgZcdvcz+P4nx/tlNy+rx6/+/TouOzZQIH5/3uFY/frW0MA9ra4H1r67C4O8VZey4kI8dMnMkOGiYXAVrjz+UH+Sa+jVJTwhOLVhII44pFo0ZNpMH9Yzkgfit+dMxUubtvv9CAAWjK/Bjj2NvmIAAGenKczdK0rwyMo5oXfotCkDMbhnBcZbhrAfnToRN69aj9OmDHS2485zp+JfL2/G/HH9/LLfn3c43tm+F/27pyYv5SWFOH5s31CoRHFhAW5e1hCq65wjhuLZN7dhSX2tX3bl8Ydi2U2rsdJ6Llccdyi+8cCLuPKEoI/37Fwaek5AaoD/1yvvYcawYAJ96/IG7Nzb6O+EKhlajhvTF8+9uQ2HW/1vjxVKMd+TD9edMhE/+d+1/qRk5bEjsOLmJ/zEtkBKFn3rTy/hu4vH+mX/uHQW/vnK5tB1/vGi6VizaQcmW+HFV6YZIWt7lOPcI4aGjNwLx/fDj/+xFgssGbPy2EPx3JvbcfHc8IJB+mLBHy+ajjsefwMXWgsL158yEff8902cPmWQX/a9xWPx0qYdfn8uKy7EBbPrUKCCyd+hfbvi24vGYLAlE5dPH4zfP/Umzjw8qOvs6YMxtn8lxvYP+tqtyxtwwz/X4UKrvV9fOBo//Nsr+MIxwXO/+7zD8dSGDzDLMoj89bMzsWNPY2iyuGJmWPE7fmwNfvSPtTh2dCCLLj16OFa/vhXnzw6Mhd9dPA43/ed138sTACbWdsfPloYNUpWdijFpYDgM/OqFo6G1Dr1X150yAY37dbChQoHCT6wJN5Ayqv3jpc34+MTg/fn8McNR3aXUNwoBqQnDD5dMCP3224vG4JoHX8YVxwf95LfnTMUja9/HsaMDub5i5hCsfXcnDvEM7lUVJeiTtlI8e0QvVHYqDr1D58+uwyvv7sQpVj/94ZIJWPHr1fi65XF917lT8YO/vhIq++tnj8AT67eG2nHjmZPx9BvbMGVIcO+WTh2Yen6ecaKyvBj/uWyO70ECpDbsef6rx4SU4kUT+2F4ny6hELzvfGIsZg3vhXnWOb/58TG45MhDQp5/f//cEXh64wc40hqvbl3egMam4FmVFRfiJ6dNQom14+mIPikPstqqctR06+R7t6y+8sjQc//mojHo3bUMS6cGsvM3yxtw9R9fxDUnBnLgZ0sn4+ZV6/GZIwNj6YqZQ/CX5zeFruG7i8di2+6PfMPXgKpynHPEkNC4MbKmKx66ZCZ6W5PWBeNrMKS6AsOshZYblk7GNx94EZdaSbe/vWgMvnrfC/jaguD5/WZ5A/7y/Cac0hBcw+orj8TWXfv8iXFhgcI5RwyBQqC7DKgqx70XTAstai6dOgg//3+v4expwVh09YLRKCsuDJVJ3PapKXhp03ZMrA3GufsvnI5nNm7zPUBsTDvKigvx1FVHhUI6h/fpgvsvnB6S+Uvqa9GlrBhTLL2kfnAVnrzqqNAzXTFzKJZOHRQyBNyyrAH3P/NW6BounjsMq9a9H2rbnedOReN+7S8MjfQ8VuwJa2V5MR767BGha7l6wWh8+tYnccmRgVy89VMN+Mq9z4fetd5dy3wjQzoTaqPG6Z7ehNN4lncqLvSv1chlW+drGNIDdz6xMWR4XVI/ANc8+HJo3Lnh9En46n0v4H9ODvSqL84bgZ/+cy3OmxXIWYNZFAnv+B3d/dG0LV2fAoCGwT3wjzWbQ+/ClCFVWLVuC+qFVB22UbayUzG2ffgR6gdHjzuloRa/eXQDTp8yEDevWg8gpcfs3rcfE2u74ckNH4SO71ERnVfUD67Cy+/sxJDqYEz80sdGYUlDrS+LgZQ38Ovv7/YTntsM8xZq7L5oDLm1VeVYUl8bWrydP74Gf35uU0QXam542C3LGvDn59/GKQ21GY8pLizAJyf3h4Ly+/jkgd3xsXE1GGctuM4fV4Of/u+6kMf1GVMHYdyAbiFDyq/OOgy/fmQ9llu667UnjcP1D6/FF+dFNwsYZ80Bbl5Wjy279vm6dk1lJ4zrXxmS/wBCYx+QGuvu+e9b+Iz1rn3mqENw1R+ew/mzA32ic2kRdu5txLDewfM7cVJ/3PnExtBc9eoFo7B6/VZMsbw5R6f139H9umLxpP441Lr206cOxJ+f2xR6n2cMq8bjr2/FUKsfTR5Yha5l60P6/z3nT8Ptj7+BL1o607DeXUJtjYtSKqRXJOHOc6fitsfeEBfzbR65fC627tqHAdaC2AtfOyY0Dvet7IQfnzoxNJ6cOHkAHl6zOaRbfPqIofjXK++FZPFNZ9dj5d3P4usLR/ll88fV4N6n3wrNaz42rgb3Pf0WFnoLbmZhpbBAiQ42HR3VWnYopdRiAPO01su9/58OoEFrfYF1zHPeMRu9/6/1jnkvra4VAFYAQG1t7aT169e3SpvzhfTJS3PZuHU3rn94Lc6ZOQSDvMngnavfwDceeBG3LGvwhWBTk8b2PR+hW8wwNBdaazy85l2MrgkL+i279qF7eXGrJebdtvsjFBepkJt7PqO1RpM+OLZG7ih8uG9/ZBV9x56PsiYSbWrS2NO4P2vffWf7HpQVFzpdyddt3okF1/0b582uc6667drbGNqR60CQK7nWGrSkbbm8rvZ8j1pCc6+rcX8TCpQKebbsb9Kx5FpHvZfthba4v01NOqc7J6azr7EpcdLwJCy8/t/YsGU3Vq2c26rnyRUf7tuPkqKCVtMjXnhrO556YytOqa/1+9Jzb27Dezv3hjxed+5tRFGBChlP1mzagV5dSn0DQlOTxoMvbMLE2u6+7rm/SeOpDVsxpn9l1pCq9L71+Otb8OLb27F06iC/7PsPrsH2PY34yvzURFZrjQtuewp11Z1xiZUK4C/Pb8LIvl39CfW+xibc8983MfOQan8i++6OPbh11Qac2lDrt/eJ9Vvwy3+/jq/MH+VPft/fuRfr3tsVynf7wLNv47X3duG8WUOx5p0dqKvujL+++A6KCwswqGcFbnt0A86bXYc3tuxG765lvqF4195GfPF3z2DRxH6YM6K3f29vf2wDThhbI0YTGLbs2odnNn6AmcOq/fu0dvNOvPruzpDH5CNr38fOvY0Rw9fBxHs796J7eUlO+v3GrbuxbvMu0ahoE1eeNu5vioTWvvXBh+hbWeb//tV3d+CWVRtw4Zw6P8pEa40PP8quZzaXfY1N+NNzb2NaXc+QRyzH4TBx5gSN+5uwYctu3/sZiD6/d7bvwap172POiF6xNys42FBKPaG1FsMQDgrjmM3kyZP16tWrW6XNJDdQWBHSceH7TQgh7Y+mJo0mrXOeN4oQQgjpSLiMY605gr4JwI4B6++ViccopYoAVAJ4vxXbRA4AnDgT0nHh+00IIe2PggJFwxghhBDSAlpzFH0cwDCl1GClVAmAkwHcm3bMvQDO8D4vBvB35hsjhBBCCCGEEEIIIQeKVksKo7VuVEpdAOAvAAoB3Ki1fl4p9TUAq7XW9wL4BYCblVKvAtiClAGNEEIIIYQQQgghhJADQqtmTNZaPwDggbSyL1mf9wA4sTXbQAghhBBCCCGEEEJIJpicgBBCCCGEEEIIIYTkLTSOEUIIIYQQQgghhJC8hcYxQgghhBBCCCGEEJK30DhGCCGEEEIIIYQQQvIWGscIIYQQQgghhBBCSN5C4xghhBBCCCGEEEIIyVtoHCOEEEIIIYQQQggheQuNY4QQQgghhBBCCCEkb6FxjBBCCCGEEEIIIYTkLTSOEUIIIYQQQgghhJC8hcYxQgghhBBCCCGEEJK30DhGCCGEEEIIIYQQQvIWGscIIYQQQgghhBBCSN5C4xghhBBCCCGEEEIIyVtoHCOEEEIIIYQQQggheQuNY4QQQgghhBBCCCEkb6FxjBBCCCGEEEIIIYTkLTSOEUIIIYQQQgghhJC8hcYxQgghhBBCCCGEEJK3KK11W7chEUqpzQDWt3U7ckxPAO+1dSMIAfsiaR+wH5L2APshaS+wL5L2APshaQ+wH5KWMlBrXS19cdAZxzoiSqnVWuvJbd0OQtgXSXuA/ZC0B9gPSXuBfZG0B9gPSXuA/ZC0JgyrJIQQQgghhBBCCCF5C41jhBBCCCGEEEIIISRvoXGsfXBDWzeAEA/2RdIeYD8k7QH2Q9JeYF8k7QH2Q9IeYD8krQZzjhFCCCGEEEIIIYSQvIWeY4QQQgghhBBCCCEkb6FxjBBCCCGEEEIIIYTkLTSOtTFKqXlKqTVKqVeVUpe1dXtIx0UpNUAp9bBS6gWl1PNKqYu98iql1ENKqVe8v929cqWU+qHXN59RSk1s2ysgHQmlVKFS6iml1P3e/wcrpR71+tsdSqkSr7zU+/+r3veD2rLdpGOhlOqmlLpLKfWSUupFpdRUykRyoFFKXeKNy88ppW5TSpVRJpLWRil1o1LqXaXUc1ZZYvmnlDrDO/4VpdQZbXEt5OAmQ1/8njc2P6OU+r1Sqpv13UqvL65RSh1jlXNeTVoEjWNtiFKqEMD1AI4FMBLAEqXUyLZtFenANAL4nNZ6JIApAM73+ttlAP6mtR4G4G/e/4FUvxzm/VsB4McHvsmkA3MxgBet/38HwLVa6zoAWwEs88qXAdjqlV/rHUdIrvgfAH/WWo8AMA6pPkmZSA4YSql+AC4CMFlrPRpAIYCTQZlIWp9fAZiXVpZI/imlqgB8GUADgHoAXzYGNUIS8CtE++JDAEZrrccCeBnASgDw5i4nAxjl/eZH3oIr59WkxdA41rbUA3hVa71Oa70PwO0AFrRxm0gHRWv9ttb6Se/zDqQmgf2Q6nM3eYfdBGCh93kBgF/rFKsAdFNK9T3AzSYdEKVUfwDHA/i5938FYA6Au7xD0vuh6Z93AZjrHU9Ii1BKVQKYCeAXAKC13qe1/gCUieTAUwSgk1KqCEA5gLdBmUhaGa31PwFsSStOKv+OAfCQ1nqL1norUgaNdCMHIU6kvqi1flBr3ej9dxWA/t7nBQBu11rv1Vq/BuBVpObUnFeTFkPjWNvSD8Ab1v83emWEtCpeGMYEAI8C6K21ftv7ahOA3t5n9k/SWvwAwBcANHn/7wEM5y3gAAAE+UlEQVTgA0sJsvua3w+977d5xxPSUgYD2Azgl16I78+VUhWgTCQHEK31mwCuAbABKaPYNgBPgDKRtA1J5R/lIjkQnA3gT95n9kXSatA4RkieoZTqDOB3AD6jtd5uf6e11gB0mzSM5AVKqRMAvKu1fqKt20LyniIAEwH8WGs9AcAuBCFEACgTSevjhaAtQMpYWwOgAvS8Ie0Ayj/SHlBKXYFUaphb27otpOND41jb8iaAAdb/+3tlhLQKSqlipAxjt2qt7/aK3zGhQd7fd71y9k/SGkwDMF8p9TpSLu9zkMr71M0LKQLCfc3vh973lQDeP5ANJh2WjQA2aq0f9f5/F1LGMspEciA5EsBrWuvNWuuPANyNlJykTCRtQVL5R7lIWg2l1JkATgBwqmesBdgXSStC41jb8jiAYd6ORCVIJRe8t43bRDooXk6SXwB4UWv9feurewGY3YXOAHCPVb7U26FoCoBtlqs9Ic1Ca71Sa91faz0IKZn3d631qQAeBrDYOyy9H5r+udg7nivZpMVorTcBeEMpNdwrmgvgBVAmkgPLBgBTlFLl3jht+iFlImkLksq/vwA4WinV3fOCPNorI6RFKKXmIZWCY77Werf11b0ATvZ27h2M1CYRj4HzapIDFMfTtkUpdRxS+XcKAdyotf5GGzeJdFCUUtMB/AvAswhyPV2OVN6x3wKoBbAewCe11ls8Jf06pMI7dgM4S2u9+oA3nHRYlFKzAFyqtT5BKTUEKU+yKgBPAThNa71XKVUG4GakcuRtAXCy1npdW7WZdCyUUuOR2hiiBMA6AGchtXBImUgOGEqprwI4CanQoacALEcqVw5lImk1lFK3AZgFoCeAd5DadfIPSCj/lFJnI6VPAsA3tNa/PJDXQQ5+MvTFlQBKEXjGrtJan+sdfwVSecgakUoT8yevnPNq0iJoHCOEEEIIIYQQQggheQvDKgkhhBBCCCGEEEJI3kLjGCGEEEIIIYQQQgjJW2gcI4QQQgghhBBCCCF5C41jhBBCCCGEEEIIISRvoXGMEEIIIYQQQgghhOQtNI4RQgghhLRjlFLdlFLneZ9rlFJ3tXWbCCGEEEI6Ekpr3dZtIIQQQgghGVBKDQJwv9Z6dBs3hRBCCCGkQ1LU1g0ghBBCCCFOvg1gqFLqvwBeAXCo1nq0UupMAAsBVAAYBuAaACUATgewF8BxWustSqmhAK4HUA1gN4BPaa1fOvCXQQghhBDSPmFYJSGEEEJI++YyAGu11uMBfD7tu9EAFgE4DMA3AOzWWk8A8AiApd4xNwC4UGs9CcClAH50QFpNCCGEEHKQQM8xQgghhJCDl4e11jsA7FBKbQNwn1f+LICxSqnOAA4HcKdSyvym9MA3kxBCCCGk/ULjGCGEEELIwcte63OT9f8mpPS8AgAfeF5nhBBCCCFEgGGVhBBCCCHtmx0AujTnh1rr7QBeU0qdCAAqxbhcNo4QQggh5GCHxjFCCCGEkHaM1vp9AP9WSj0H4HvNqOJUAMuUUk8DeB7Agly2jxBCCCHkYEdprdu6DYQQQgghhBBCCCGEtAn0HCOEEEIIIYQQQggheQuNY4QQQgghhBBCCCEkb6FxjBBCCCGEEEIIIYTkLTSOEUIIIYQQQgghhJC8hcYxQgghhBBCCCGEEJK30DhGCCGEEEIIIYQQQvIWGscIIYQQQgghhBBCSN7y/wFNNj4q0etyPAAAAABJRU5ErkJggg==\n",
            "text/plain": [
              "<Figure size 1512x720 with 1 Axes>"
            ]
          },
          "metadata": {
            "tags": [],
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 367
        },
        "id": "LYRHPphow8Fb",
        "outputId": "f2c90126-6ebd-4da9-dcc1-a0f1514f437f"
      },
      "source": [
        "loss_vol = history_vol.history['loss']\r\n",
        "mae_vol = history_vol.history['mae']\r\n",
        "\r\n",
        "\r\n",
        "fig = plt.subplots(figsize = (15,5))\r\n",
        "epochs = range(30)\r\n",
        "\r\n",
        "ax = plt.subplot(111)\r\n",
        "ax.plot(epochs, loss_vol, linewidth=2, label='Loss')\r\n",
        "ax.plot(epochs, mae_vol, linewidth=2, label='MAE')\r\n",
        "\r\n",
        "ax.legend()\r\n",
        "ax.set_xlabel('time')\r\n",
        "ax.set_ylabel('volume ')\r\n",
        "ax.set_title('Comparison EURUSD End-of-day traded volume Loss and MAE')\r\n",
        "\r\n"
      ],
      "execution_count": 15,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "Text(0.5, 1.0, 'Comparison EURUSD End-of-day traded volume Loss and MAE')"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 15
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 1080x360 with 1 Axes>"
            ]
          },
          "metadata": {
            "tags": [],
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "kjXct6PuxI8f"
      },
      "source": [
        "# **References:**\r\n",
        "\r\n",
        "\r\n",
        "\r\n",
        "---\r\n",
        "\r\n",
        "---\r\n",
        "<a name=\"Almgren\"></a>\r\n",
        "1. Almgren, R., and N. Chriss. 2001. Optimal Execution of Portfolio Transactions. The Journal of Risk 3: 5–40\r\n",
        "\r\n",
        "2. Gatheral, J. 2010. No-Dynamic-Arbitrage and Market Impact. Quantitative Finance 10: 749–759.\r\n",
        "<a name=\"Kamijo\"></a>\r\n",
        "3. Kamijo, K. and Tanigawa, T., Stock price pattern recognition: a recurrent neural network approach, In Proceedings of the International Joint Conference on Neural Networks, 1990, pp. 1215–1221.\r\n",
        "<a name=\"Tenti\"></a>\r\n",
        "4. Tenti, P., Forecasting foreign exchange rates using recurrent neural networks. Appl. Artif. Intell., 1996, 10, 567–581.\r\n",
        "<a name=\"Tino\"></a>\r\n",
        "5. Tino, P., Schittenkopf, C. and Doffner, G., Financial volatility trading using recurrent networks. IEEE Trans. Neural Netw., 2001, 12, 865–874.\r\n",
        "<a name=\"Dunis\"></a>\r\n",
        "6. Dunis, C. and Huang, X., Forecasting and trading currency volatility: an application of recurrent neural regression and model combination. J. Forecast., 2002, 21, 317–354.\r\n"
      ]
    }
  ]
}