team_8_assignement_1.ipynb

{
  "nbformat": 4,
  "nbformat_minor": 0,
  "metadata": {
    "colab": {
      "name": "team_8_assignement_1.ipynb",
      "provenance": [],
      "collapsed_sections": [],
      "toc_visible": true
    },
    "kernelspec": {
      "name": "python3",
      "display_name": "Python 3"
    },
    "language_info": {
      "name": "python"
    },
    "accelerator": "GPU"
  },
  "cells": [
    {
      "cell_type": "markdown",
      "source": [
        "# Team Information & Contents \n",
        "\n",
        "|Team-number :| 8|\n",
        "|:----:|:----:|\n",
        "\n",
        "\n",
        "|Name|    E-Mail        |matriculation-nr.|\n",
        "|:----:|:----:|:----:|\n",
        "|Altuğ Demirşah| demira02@ads.uni-passau.de|109776|\n",
        "|Elif Günay| guenay01@ads.uni-passau.de|109806|\n",
        "|Evren Can| can05@ads.uni-passau.de|109484|\n",
        "\n",
        "**\n",
        "\n",
        "----------------------------------------------------------------------\n",
        "\n",
        "**\n",
        "\n",
        "\n",
        "This report is provided for the sub-assignment of the Deep Learning and Text Analysis in Finance course at the University of Passau on WiSe 21/22.\n",
        "\n",
        "**\n",
        "\n",
        "**Aim**: This report contains our own preprocessing and modelling approaches for a [competition](https://https://www.kaggle.com/c/g-research-crypto-forecasting) based on Kaggle and this competition aims to predict some of the most popular cryptocurrencies' future changes.\n",
        "\n",
        "**In summary**, we prefered working on BTC and ETH cryptocurrencies, built two different models for the purpose of predicting their *Target* and those models are built-on LSTM and GRU, fine-tuned hyperparameters to get better performance, evaluated the outputs and made some final comments.\n",
        "\n",
        "**\n",
        "\n",
        "Contents:\n",
        "\n",
        "1.   **Import needed libraries**\n",
        "2.   **Data Preprocessing**\n",
        "    \n",
        "    2.1. Reading datasets\n",
        "\n",
        "    2.2. Dealing with missing data\n",
        "    \n",
        "    2.3. Picking two coins to work with\n",
        "\n",
        "    2.4. Plot data\n",
        "\n",
        "    2.5. Split & Scale data\n",
        "\n",
        "    2.6. Observation & Next Price\n",
        "\n",
        "    2.7. Oversee preprocessed data\n",
        "\n",
        "3. **Modelling**\n",
        "\n",
        "    3.1. Modelling 1\n",
        "    \n",
        "      - 3.1.1. train\n",
        "\n",
        "      - 3.1.2. performence\n",
        "\n",
        "      - 3.1.3. predict\n",
        "\n",
        "  3.2. Modelling 2\n",
        "    \n",
        "      - 3.2.1. train\n",
        "\n",
        "      - 3.2.2. performence\n",
        "\n",
        "      - 3.2.3. predict\n",
        "\n",
        "4. **Evaluation**\n",
        "\n",
        "    4.1. Evaluation on model 1\n",
        "\n",
        "    4.2. Evaluation on model 2\n",
        "\n",
        "5. Comments\n",
        "\n",
        "    5.1. regarding to model 1\n",
        "\n",
        "    5.2. regarding to model 2\n",
        "\n",
        "    5.3. regarding to hyper parameters\n",
        "\n",
        "    5.4. regarding to technical analysis\n",
        "\n",
        "    5.5. final comments\n",
        "\n",
        "\n",
        "In order to run this code file, these should be considered:\n",
        "*   This code file was created on Google Colab and data was uploaded from Google Drive.\n",
        "*   Since we run this code file on the basic Colab version, limited memory and computational sources are provided by the platform. Running this code file on a different device might require/take a long time depending on the sources that the device has.\n",
        "\n"
      ],
      "metadata": {
        "id": "gn-N183PRZpm"
      }
    },
    {
      "cell_type": "markdown",
      "source": [
        "# Import needed libraries"
      ],
      "metadata": {
        "id": "PSg0BaaWLBkb"
      }
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "12i-PhuFU3dV"
      },
      "outputs": [],
      "source": [
        "\"\"\"\n",
        "Needed libraries:\n",
        "\n",
        "  For data proprocessing:\n",
        "\n",
        "    pandas: we use pandas library to manipulate data and analyse inside of it. In our case, manipulating numerical tables and time series.\n",
        "    numpy:  we use numpy since it offers a large collection of high-level mathematical functions to operate on the arrays.\n",
        "    matplotlib: matplotlib is a plotting library. In our case, we plot change of a cryptocurrency by time\n",
        "    datetime: datetime module supplies classes for manipulating dates and times. In our case, converting timestamps into the UTC datetime corresponding to the POSIX timestamp\n",
        "    sklearn:  to change raw feature vectors into a representation that is more suitable for the downstream estimators. In our case, scaling data with StandardScaler or MinMaxScaler\n",
        "\n",
        "  For modelling:\n",
        "\n",
        "    tensorflow:  used to design, build, and train deep learning models. \n",
        "      ->  layers: this is the class from which all layers inherit such as LSTM, GRU, Dense and so on\n",
        "\n",
        "\"\"\"\n",
        "# data proprocessing\n",
        "import pandas as pd                                                 \n",
        "import numpy as np                                                  \n",
        "import matplotlib.pyplot as plt                                    \n",
        "from datetime import datetime as dt                                 \n",
        "from sklearn.preprocessing import StandardScaler, MinMaxScaler     \n",
        "\n",
        "# modelling\n",
        "import tensorflow as tf                                             \n",
        "from tensorflow import keras\n",
        "from keras import layers"
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "## Connecting to Google Drive"
      ],
      "metadata": {
        "id": "hYAibHO9NX2a"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "# to download data, first we need to connect to google drive\n",
        "from google.colab import drive\n",
        "drive.mount('/gdrive')"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "638O3OLlL1tJ",
        "outputId": "ab2d32a7-94a8-4343-efe8-109229c4aba4"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Mounted at /gdrive\n"
          ]
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "# Data preprocessing"
      ],
      "metadata": {
        "id": "-h-wAAkqLVsn"
      }
    },
    {
      "cell_type": "markdown",
      "source": [
        "## Reading datasets"
      ],
      "metadata": {
        "id": "CwVtABcjNp_d"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "%%time\n",
        "'''\n",
        "    timestamp - a timestamp for the minute covered by the row.\n",
        "    Asset_ID - an ID code for the cryptoasset.\n",
        "    Count - the number of trades that took place this minute. \n",
        "    Open - the USD price at the beginning of the minute. \n",
        "    High - the highest USD price during the minute. \n",
        "    Low - the lowest USD price during the minute. \n",
        "    Close - the USD price at the end of the minute. \n",
        "    Volume - the number of cryptoasset units traded during the minute. \n",
        "    VWAP - the volume weighted average price for the minute. \n",
        "    Target - is derived from log returns over 15 minutes\n",
        "\n",
        "'''\n",
        "train_dtypes={\n",
        "    'timestamp': np.int64,\n",
        "    'Asset_ID': np.int8,\n",
        "    'Count': np.int64,\n",
        "    'Open': np.float64,\n",
        "    'High': np.float64,\n",
        "    'Low': np.float64,\n",
        "    'Close': np.float64,\n",
        "    'Volume': np.float64,\n",
        "    'VWAP': np.float64,\n",
        "    'Target': np.float64,\n",
        "}\n",
        "\n",
        "#############################################################################################################################################\n",
        "\n",
        "\"\"\"\n",
        "    Asset_ID - an ID code for the cryptoasset\n",
        "    Weight - each cryptoasset receives in the metric\n",
        "    Asset_Name - the real name and of the cryptoasset\n",
        "\n",
        "\"\"\"\n",
        "asset_dtypes= {\n",
        "    'Asset_ID': np.int8,\n",
        "    'Weight': np.float64,\n",
        "    'Asset_Name': np.object\n",
        "}\n",
        "\n",
        "#############################################################################################################################################\n",
        "\n",
        "# load train data\n",
        "train_PATH = '/gdrive/MyDrive/DLTA/Data/train.csv'\n",
        "df = pd.read_csv(train_PATH, sep=',', dtype=train_dtypes, usecols=list(train_dtypes.keys()))\n",
        "\n",
        "# convert timestamps into Time column that contains the UTC datetime corresponding to the POSIX timestamp\n",
        "df['Time'] = [dt.utcfromtimestamp(d).strftime('%Y-%m-%d %H:%M:%S') for d in df.timestamp]\n",
        "\n",
        "# load asset details\n",
        "asset_PATH = '/gdrive/MyDrive/DLTA/Data/asset_details.csv'\n",
        "asset_details = pd.read_csv(asset_PATH, sep=',', dtype=asset_dtypes, usecols=list(asset_dtypes.keys()))\n",
        "\n",
        "# merge df and asset details to get all columns in a table\n",
        "df = pd.merge(df, asset_details,\n",
        "                on =\"Asset_ID\",\n",
        "                how = 'left')\n",
        "\n",
        "# print a concise summary of the dataframe\n",
        "df.info(null_counts=True)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "StTwFsPCVDcy",
        "outputId": "6508661e-3589-458f-c09f-9294379139b3"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "<class 'pandas.core.frame.DataFrame'>\n",
            "Int64Index: 24236806 entries, 0 to 24236805\n",
            "Data columns (total 13 columns):\n",
            " #   Column      Non-Null Count     Dtype  \n",
            "---  ------      --------------     -----  \n",
            " 0   timestamp   24236806 non-null  int64  \n",
            " 1   Asset_ID    24236806 non-null  int8   \n",
            " 2   Count       24236806 non-null  int64  \n",
            " 3   Open        24236806 non-null  float64\n",
            " 4   High        24236806 non-null  float64\n",
            " 5   Low         24236806 non-null  float64\n",
            " 6   Close       24236806 non-null  float64\n",
            " 7   Volume      24236806 non-null  float64\n",
            " 8   VWAP        24236797 non-null  float64\n",
            " 9   Target      23486468 non-null  float64\n",
            " 10  Time        24236806 non-null  object \n",
            " 11  Weight      24236806 non-null  float64\n",
            " 12  Asset_Name  24236806 non-null  object \n",
            "dtypes: float64(8), int64(2), int8(1), object(2)\n",
            "memory usage: 2.4+ GB\n",
            "CPU times: user 1min 50s, sys: 8.35 s, total: 1min 58s\n",
            "Wall time: 2min 13s\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "# tabular data\n",
        "df"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 424
        },
        "id": "QopK8TyLXN9z",
        "outputId": "f11cdd19-240a-4cf6-8161-0e0b820ea806"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/html": [
              "\n",
              "  <div id=\"df-64ad7ae3-cdfa-4acc-86d9-99ce96d516c6\">\n",
              "    <div class=\"colab-df-container\">\n",
              "      <div>\n",
              "<style scoped>\n",
              "    .dataframe tbody tr th:only-of-type {\n",
              "        vertical-align: middle;\n",
              "    }\n",
              "\n",
              "    .dataframe tbody tr th {\n",
              "        vertical-align: top;\n",
              "    }\n",
              "\n",
              "    .dataframe thead th {\n",
              "        text-align: right;\n",
              "    }\n",
              "</style>\n",
              "<table border=\"1\" class=\"dataframe\">\n",
              "  <thead>\n",
              "    <tr style=\"text-align: right;\">\n",
              "      <th></th>\n",
              "      <th>timestamp</th>\n",
              "      <th>Asset_ID</th>\n",
              "      <th>Count</th>\n",
              "      <th>Open</th>\n",
              "      <th>High</th>\n",
              "      <th>Low</th>\n",
              "      <th>Close</th>\n",
              "      <th>Volume</th>\n",
              "      <th>VWAP</th>\n",
              "      <th>Target</th>\n",
              "      <th>Time</th>\n",
              "      <th>Weight</th>\n",
              "      <th>Asset_Name</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>1514764860</td>\n",
              "      <td>2</td>\n",
              "      <td>40</td>\n",
              "      <td>2376.580000</td>\n",
              "      <td>2399.500000</td>\n",
              "      <td>2357.140000</td>\n",
              "      <td>2374.590000</td>\n",
              "      <td>1.923301e+01</td>\n",
              "      <td>2373.116392</td>\n",
              "      <td>-0.004218</td>\n",
              "      <td>2018-01-01 00:01:00</td>\n",
              "      <td>2.397895</td>\n",
              "      <td>Bitcoin Cash</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>1514764860</td>\n",
              "      <td>0</td>\n",
              "      <td>5</td>\n",
              "      <td>8.530000</td>\n",
              "      <td>8.530000</td>\n",
              "      <td>8.530000</td>\n",
              "      <td>8.530000</td>\n",
              "      <td>7.838000e+01</td>\n",
              "      <td>8.530000</td>\n",
              "      <td>-0.014399</td>\n",
              "      <td>2018-01-01 00:01:00</td>\n",
              "      <td>4.304065</td>\n",
              "      <td>Binance Coin</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>1514764860</td>\n",
              "      <td>1</td>\n",
              "      <td>229</td>\n",
              "      <td>13835.194000</td>\n",
              "      <td>14013.800000</td>\n",
              "      <td>13666.110000</td>\n",
              "      <td>13850.176000</td>\n",
              "      <td>3.155006e+01</td>\n",
              "      <td>13827.062093</td>\n",
              "      <td>-0.014643</td>\n",
              "      <td>2018-01-01 00:01:00</td>\n",
              "      <td>6.779922</td>\n",
              "      <td>Bitcoin</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>1514764860</td>\n",
              "      <td>5</td>\n",
              "      <td>32</td>\n",
              "      <td>7.659600</td>\n",
              "      <td>7.659600</td>\n",
              "      <td>7.656700</td>\n",
              "      <td>7.657600</td>\n",
              "      <td>6.626713e+03</td>\n",
              "      <td>7.657713</td>\n",
              "      <td>-0.013922</td>\n",
              "      <td>2018-01-01 00:01:00</td>\n",
              "      <td>1.386294</td>\n",
              "      <td>EOS.IO</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>1514764860</td>\n",
              "      <td>7</td>\n",
              "      <td>5</td>\n",
              "      <td>25.920000</td>\n",
              "      <td>25.920000</td>\n",
              "      <td>25.874000</td>\n",
              "      <td>25.877000</td>\n",
              "      <td>1.210873e+02</td>\n",
              "      <td>25.891363</td>\n",
              "      <td>-0.008264</td>\n",
              "      <td>2018-01-01 00:01:00</td>\n",
              "      <td>2.079442</td>\n",
              "      <td>Ethereum Classic</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>...</th>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>24236801</th>\n",
              "      <td>1632182400</td>\n",
              "      <td>9</td>\n",
              "      <td>775</td>\n",
              "      <td>157.181571</td>\n",
              "      <td>157.250000</td>\n",
              "      <td>156.700000</td>\n",
              "      <td>156.943857</td>\n",
              "      <td>4.663725e+03</td>\n",
              "      <td>156.994319</td>\n",
              "      <td>NaN</td>\n",
              "      <td>2021-09-21 00:00:00</td>\n",
              "      <td>2.397895</td>\n",
              "      <td>Litecoin</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>24236802</th>\n",
              "      <td>1632182400</td>\n",
              "      <td>10</td>\n",
              "      <td>34</td>\n",
              "      <td>2437.065067</td>\n",
              "      <td>2438.000000</td>\n",
              "      <td>2430.226900</td>\n",
              "      <td>2432.907467</td>\n",
              "      <td>3.975460e+00</td>\n",
              "      <td>2434.818747</td>\n",
              "      <td>NaN</td>\n",
              "      <td>2021-09-21 00:00:00</td>\n",
              "      <td>1.098612</td>\n",
              "      <td>Maker</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>24236803</th>\n",
              "      <td>1632182400</td>\n",
              "      <td>13</td>\n",
              "      <td>380</td>\n",
              "      <td>0.091390</td>\n",
              "      <td>0.091527</td>\n",
              "      <td>0.091260</td>\n",
              "      <td>0.091349</td>\n",
              "      <td>2.193732e+06</td>\n",
              "      <td>0.091388</td>\n",
              "      <td>NaN</td>\n",
              "      <td>2021-09-21 00:00:00</td>\n",
              "      <td>1.791759</td>\n",
              "      <td>TRON</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>24236804</th>\n",
              "      <td>1632182400</td>\n",
              "      <td>12</td>\n",
              "      <td>177</td>\n",
              "      <td>0.282168</td>\n",
              "      <td>0.282438</td>\n",
              "      <td>0.281842</td>\n",
              "      <td>0.282051</td>\n",
              "      <td>1.828508e+05</td>\n",
              "      <td>0.282134</td>\n",
              "      <td>NaN</td>\n",
              "      <td>2021-09-21 00:00:00</td>\n",
              "      <td>2.079442</td>\n",
              "      <td>Stellar</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>24236805</th>\n",
              "      <td>1632182400</td>\n",
              "      <td>11</td>\n",
              "      <td>48</td>\n",
              "      <td>232.695000</td>\n",
              "      <td>232.800000</td>\n",
              "      <td>232.240000</td>\n",
              "      <td>232.275000</td>\n",
              "      <td>1.035123e+02</td>\n",
              "      <td>232.569697</td>\n",
              "      <td>NaN</td>\n",
              "      <td>2021-09-21 00:00:00</td>\n",
              "      <td>1.609438</td>\n",
              "      <td>Monero</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "<p>24236806 rows × 13 columns</p>\n",
              "</div>\n",
              "      <button class=\"colab-df-convert\" onclick=\"convertToInteractive('df-64ad7ae3-cdfa-4acc-86d9-99ce96d516c6')\"\n",
              "              title=\"Convert this dataframe to an interactive table.\"\n",
              "              style=\"display:none;\">\n",
              "        \n",
              "  <svg xmlns=\"http://www.w3.org/2000/svg\" height=\"24px\"viewBox=\"0 0 24 24\"\n",
              "       width=\"24px\">\n",
              "    <path d=\"M0 0h24v24H0V0z\" fill=\"none\"/>\n",
              "    <path d=\"M18.56 5.44l.94 2.06.94-2.06 2.06-.94-2.06-.94-.94-2.06-.94 2.06-2.06.94zm-11 1L8.5 8.5l.94-2.06 2.06-.94-2.06-.94L8.5 2.5l-.94 2.06-2.06.94zm10 10l.94 2.06.94-2.06 2.06-.94-2.06-.94-.94-2.06-.94 2.06-2.06.94z\"/><path d=\"M17.41 7.96l-1.37-1.37c-.4-.4-.92-.59-1.43-.59-.52 0-1.04.2-1.43.59L10.3 9.45l-7.72 7.72c-.78.78-.78 2.05 0 2.83L4 21.41c.39.39.9.59 1.41.59.51 0 1.02-.2 1.41-.59l7.78-7.78 2.81-2.81c.8-.78.8-2.07 0-2.86zM5.41 20L4 18.59l7.72-7.72 1.47 1.35L5.41 20z\"/>\n",
              "  </svg>\n",
              "      </button>\n",
              "      \n",
              "  <style>\n",
              "    .colab-df-container {\n",
              "      display:flex;\n",
              "      flex-wrap:wrap;\n",
              "      gap: 12px;\n",
              "    }\n",
              "\n",
              "    .colab-df-convert {\n",
              "      background-color: #E8F0FE;\n",
              "      border: none;\n",
              "      border-radius: 50%;\n",
              "      cursor: pointer;\n",
              "      display: none;\n",
              "      fill: #1967D2;\n",
              "      height: 32px;\n",
              "      padding: 0 0 0 0;\n",
              "      width: 32px;\n",
              "    }\n",
              "\n",
              "    .colab-df-convert:hover {\n",
              "      background-color: #E2EBFA;\n",
              "      box-shadow: 0px 1px 2px rgba(60, 64, 67, 0.3), 0px 1px 3px 1px rgba(60, 64, 67, 0.15);\n",
              "      fill: #174EA6;\n",
              "    }\n",
              "\n",
              "    [theme=dark] .colab-df-convert {\n",
              "      background-color: #3B4455;\n",
              "      fill: #D2E3FC;\n",
              "    }\n",
              "\n",
              "    [theme=dark] .colab-df-convert:hover {\n",
              "      background-color: #434B5C;\n",
              "      box-shadow: 0px 1px 3px 1px rgba(0, 0, 0, 0.15);\n",
              "      filter: drop-shadow(0px 1px 2px rgba(0, 0, 0, 0.3));\n",
              "      fill: #FFFFFF;\n",
              "    }\n",
              "  </style>\n",
              "\n",
              "      <script>\n",
              "        const buttonEl =\n",
              "          document.querySelector('#df-64ad7ae3-cdfa-4acc-86d9-99ce96d516c6 button.colab-df-convert');\n",
              "        buttonEl.style.display =\n",
              "          google.colab.kernel.accessAllowed ? 'block' : 'none';\n",
              "\n",
              "        async function convertToInteractive(key) {\n",
              "          const element = document.querySelector('#df-64ad7ae3-cdfa-4acc-86d9-99ce96d516c6');\n",
              "          const dataTable =\n",
              "            await google.colab.kernel.invokeFunction('convertToInteractive',\n",
              "                                                     [key], {});\n",
              "          if (!dataTable) return;\n",
              "\n",
              "          const docLinkHtml = 'Like what you see? Visit the ' +\n",
              "            '<a target=\"_blank\" href=https://colab.research.google.com/notebooks/data_table.ipynb>data table notebook</a>'\n",
              "            + ' to learn more about interactive tables.';\n",
              "          element.innerHTML = '';\n",
              "          dataTable['output_type'] = 'display_data';\n",
              "          await google.colab.output.renderOutput(dataTable, element);\n",
              "          const docLink = document.createElement('div');\n",
              "          docLink.innerHTML = docLinkHtml;\n",
              "          element.appendChild(docLink);\n",
              "        }\n",
              "      </script>\n",
              "    </div>\n",
              "  </div>\n",
              "  "
            ],
            "text/plain": [
              "           timestamp  Asset_ID  ...    Weight        Asset_Name\n",
              "0         1514764860         2  ...  2.397895      Bitcoin Cash\n",
              "1         1514764860         0  ...  4.304065      Binance Coin\n",
              "2         1514764860         1  ...  6.779922           Bitcoin\n",
              "3         1514764860         5  ...  1.386294            EOS.IO\n",
              "4         1514764860         7  ...  2.079442  Ethereum Classic\n",
              "...              ...       ...  ...       ...               ...\n",
              "24236801  1632182400         9  ...  2.397895          Litecoin\n",
              "24236802  1632182400        10  ...  1.098612             Maker\n",
              "24236803  1632182400        13  ...  1.791759              TRON\n",
              "24236804  1632182400        12  ...  2.079442           Stellar\n",
              "24236805  1632182400        11  ...  1.609438            Monero\n",
              "\n",
              "[24236806 rows x 13 columns]"
            ]
          },
          "metadata": {},
          "execution_count": 4
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "As it is seen, it is a big dataframe with approximately 24 million rows and 13 columns, also, including price changes of 14 cryptocurrencies per minute between 1st of January 2018 and 21st of September 2021."
      ],
      "metadata": {
        "id": "H4wu5VWaoIrt"
      }
    },
    {
      "cell_type": "markdown",
      "source": [
        "## Dealing with missing data"
      ],
      "metadata": {
        "id": "dgloW1EXTY4m"
      }
    },
    {
      "cell_type": "markdown",
      "source": [
        "Missing data is a common challenge that leads to a loss of information and hence reduced statistical power, a more insidious consequence is that this lack of data may introduce selection bias, which could potentially invalidate the entire study, especially during working with timeseries. So before we pick two different cryptocurrencies, with the intent of avoiding missing data, we first investigate each cryptocurrency in detail."
      ],
      "metadata": {
        "id": "8lYzmu4FtzKP"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "%%time\n",
        "def missing_data(df : pd.DataFrame) -> None:\n",
        "  \n",
        "  \"\"\"\n",
        "  Observing missing datas and details of each coin\n",
        "\n",
        "  Arg:\n",
        "    df (pf.DataFrame): a table that will be explored\n",
        "\n",
        "  Returns:\n",
        "    None\n",
        "  \n",
        "  \"\"\"\n",
        "  \n",
        "  # there are 14 different cryptocurrencies in the dataset\n",
        "  for i in range(14):\n",
        "    coin_table = df[df['Asset_ID'] == i]\n",
        "    coin = df[df['Asset_ID'] == i]['Asset_Name']\n",
        "    coin.reset_index(drop=True, inplace = True)\n",
        "    missing_values = df[df['Asset_Name'] == coin[0]].isna().sum()\n",
        "    print(f'Asset_ID: {i}, \\t Asset_Name: {coin[0]} \\n\\nmissing values for each column: \\n\\n{missing_values}\\n\\n')\n",
        "\n",
        "    # info\n",
        "    print('Asset info:\\n')\n",
        "    print(pd.DataFrame(coin_table).info(null_counts=True))\n",
        "    print('\\n\\n', '*'*120, '\\n', '*'*120, '\\n\\n')\n",
        "\n",
        "  \n",
        "  # del temporary variables to avoid memory problem\n",
        "  del coin, coin_table, missing_values\n",
        "  \n",
        "  return None\n",
        "\n",
        "\n",
        "missing_data(df)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "Tab8ILAZI3Ex",
        "outputId": "d455d6c3-64e1-4854-ff52-0b3fd724f75f"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Asset_ID: 0, \t Asset_Name: Binance Coin \n",
            "\n",
            "missing values for each column: \n",
            "\n",
            "timestamp         0\n",
            "Asset_ID          0\n",
            "Count             0\n",
            "Open              0\n",
            "High              0\n",
            "Low               0\n",
            "Close             0\n",
            "Volume            0\n",
            "VWAP              0\n",
            "Target        13415\n",
            "Time              0\n",
            "Weight            0\n",
            "Asset_Name        0\n",
            "dtype: int64\n",
            "\n",
            "\n",
            "Asset info:\n",
            "\n",
            "<class 'pandas.core.frame.DataFrame'>\n",
            "Int64Index: 1942619 entries, 1 to 24236794\n",
            "Data columns (total 13 columns):\n",
            " #   Column      Non-Null Count    Dtype  \n",
            "---  ------      --------------    -----  \n",
            " 0   timestamp   1942619 non-null  int64  \n",
            " 1   Asset_ID    1942619 non-null  int8   \n",
            " 2   Count       1942619 non-null  int64  \n",
            " 3   Open        1942619 non-null  float64\n",
            " 4   High        1942619 non-null  float64\n",
            " 5   Low         1942619 non-null  float64\n",
            " 6   Close       1942619 non-null  float64\n",
            " 7   Volume      1942619 non-null  float64\n",
            " 8   VWAP        1942619 non-null  float64\n",
            " 9   Target      1929204 non-null  float64\n",
            " 10  Time        1942619 non-null  object \n",
            " 11  Weight      1942619 non-null  float64\n",
            " 12  Asset_Name  1942619 non-null  object \n",
            "dtypes: float64(8), int64(2), int8(1), object(2)\n",
            "memory usage: 194.5+ MB\n",
            "None\n",
            "\n",
            "\n",
            " ************************************************************************************************************************ \n",
            " ************************************************************************************************************************ \n",
            "\n",
            "\n",
            "Asset_ID: 1, \t Asset_Name: Bitcoin \n",
            "\n",
            "missing values for each column: \n",
            "\n",
            "timestamp       0\n",
            "Asset_ID        0\n",
            "Count           0\n",
            "Open            0\n",
            "High            0\n",
            "Low             0\n",
            "Close           0\n",
            "Volume          0\n",
            "VWAP            0\n",
            "Target        304\n",
            "Time            0\n",
            "Weight          0\n",
            "Asset_Name      0\n",
            "dtype: int64\n",
            "\n",
            "\n",
            "Asset info:\n",
            "\n",
            "<class 'pandas.core.frame.DataFrame'>\n",
            "Int64Index: 1956282 entries, 2 to 24236795\n",
            "Data columns (total 13 columns):\n",
            " #   Column      Non-Null Count    Dtype  \n",
            "---  ------      --------------    -----  \n",
            " 0   timestamp   1956282 non-null  int64  \n",
            " 1   Asset_ID    1956282 non-null  int8   \n",
            " 2   Count       1956282 non-null  int64  \n",
            " 3   Open        1956282 non-null  float64\n",
            " 4   High        1956282 non-null  float64\n",
            " 5   Low         1956282 non-null  float64\n",
            " 6   Close       1956282 non-null  float64\n",
            " 7   Volume      1956282 non-null  float64\n",
            " 8   VWAP        1956282 non-null  float64\n",
            " 9   Target      1955978 non-null  float64\n",
            " 10  Time        1956282 non-null  object \n",
            " 11  Weight      1956282 non-null  float64\n",
            " 12  Asset_Name  1956282 non-null  object \n",
            "dtypes: float64(8), int64(2), int8(1), object(2)\n",
            "memory usage: 195.9+ MB\n",
            "None\n",
            "\n",
            "\n",
            " ************************************************************************************************************************ \n",
            " ************************************************************************************************************************ \n",
            "\n",
            "\n",
            "Asset_ID: 2, \t Asset_Name: Bitcoin Cash \n",
            "\n",
            "missing values for each column: \n",
            "\n",
            "timestamp        0\n",
            "Asset_ID         0\n",
            "Count            0\n",
            "Open             0\n",
            "High             0\n",
            "Low              0\n",
            "Close            0\n",
            "Volume           0\n",
            "VWAP             0\n",
            "Target        4861\n",
            "Time             0\n",
            "Weight           0\n",
            "Asset_Name       0\n",
            "dtype: int64\n",
            "\n",
            "\n",
            "Asset info:\n",
            "\n",
            "<class 'pandas.core.frame.DataFrame'>\n",
            "Int64Index: 1953537 entries, 0 to 24236793\n",
            "Data columns (total 13 columns):\n",
            " #   Column      Non-Null Count    Dtype  \n",
            "---  ------      --------------    -----  \n",
            " 0   timestamp   1953537 non-null  int64  \n",
            " 1   Asset_ID    1953537 non-null  int8   \n",
            " 2   Count       1953537 non-null  int64  \n",
            " 3   Open        1953537 non-null  float64\n",
            " 4   High        1953537 non-null  float64\n",
            " 5   Low         1953537 non-null  float64\n",
            " 6   Close       1953537 non-null  float64\n",
            " 7   Volume      1953537 non-null  float64\n",
            " 8   VWAP        1953537 non-null  float64\n",
            " 9   Target      1948676 non-null  float64\n",
            " 10  Time        1953537 non-null  object \n",
            " 11  Weight      1953537 non-null  float64\n",
            " 12  Asset_Name  1953537 non-null  object \n",
            "dtypes: float64(8), int64(2), int8(1), object(2)\n",
            "memory usage: 195.6+ MB\n",
            "None\n",
            "\n",
            "\n",
            " ************************************************************************************************************************ \n",
            " ************************************************************************************************************************ \n",
            "\n",
            "\n",
            "Asset_ID: 3, \t Asset_Name: Cardano \n",
            "\n",
            "missing values for each column: \n",
            "\n",
            "timestamp         0\n",
            "Asset_ID          0\n",
            "Count             0\n",
            "Open              0\n",
            "High              0\n",
            "Low               0\n",
            "Close             0\n",
            "Volume            0\n",
            "VWAP              0\n",
            "Target        18731\n",
            "Time              0\n",
            "Weight            0\n",
            "Asset_Name        0\n",
            "dtype: int64\n",
            "\n",
            "\n",
            "Asset info:\n",
            "\n",
            "<class 'pandas.core.frame.DataFrame'>\n",
            "Int64Index: 1791867 entries, 1294668 to 24236792\n",
            "Data columns (total 13 columns):\n",
            " #   Column      Non-Null Count    Dtype  \n",
            "---  ------      --------------    -----  \n",
            " 0   timestamp   1791867 non-null  int64  \n",
            " 1   Asset_ID    1791867 non-null  int8   \n",
            " 2   Count       1791867 non-null  int64  \n",
            " 3   Open        1791867 non-null  float64\n",
            " 4   High        1791867 non-null  float64\n",
            " 5   Low         1791867 non-null  float64\n",
            " 6   Close       1791867 non-null  float64\n",
            " 7   Volume      1791867 non-null  float64\n",
            " 8   VWAP        1791867 non-null  float64\n",
            " 9   Target      1773136 non-null  float64\n",
            " 10  Time        1791867 non-null  object \n",
            " 11  Weight      1791867 non-null  float64\n",
            " 12  Asset_Name  1791867 non-null  object \n",
            "dtypes: float64(8), int64(2), int8(1), object(2)\n",
            "memory usage: 179.4+ MB\n",
            "None\n",
            "\n",
            "\n",
            " ************************************************************************************************************************ \n",
            " ************************************************************************************************************************ \n",
            "\n",
            "\n",
            "Asset_ID: 4, \t Asset_Name: Dogecoin \n",
            "\n",
            "missing values for each column: \n",
            "\n",
            "timestamp          0\n",
            "Asset_ID           0\n",
            "Count              0\n",
            "Open               0\n",
            "High               0\n",
            "Low                0\n",
            "Close              0\n",
            "Volume             0\n",
            "VWAP               0\n",
            "Target        144974\n",
            "Time               0\n",
            "Weight             0\n",
            "Asset_Name         0\n",
            "dtype: int64\n",
            "\n",
            "\n",
            "Asset info:\n",
            "\n",
            "<class 'pandas.core.frame.DataFrame'>\n",
            "Int64Index: 1156866 entries, 7269141 to 24236796\n",
            "Data columns (total 13 columns):\n",
            " #   Column      Non-Null Count    Dtype  \n",
            "---  ------      --------------    -----  \n",
            " 0   timestamp   1156866 non-null  int64  \n",
            " 1   Asset_ID    1156866 non-null  int8   \n",
            " 2   Count       1156866 non-null  int64  \n",
            " 3   Open        1156866 non-null  float64\n",
            " 4   High        1156866 non-null  float64\n",
            " 5   Low         1156866 non-null  float64\n",
            " 6   Close       1156866 non-null  float64\n",
            " 7   Volume      1156866 non-null  float64\n",
            " 8   VWAP        1156866 non-null  float64\n",
            " 9   Target      1011892 non-null  float64\n",
            " 10  Time        1156866 non-null  object \n",
            " 11  Weight      1156866 non-null  float64\n",
            " 12  Asset_Name  1156866 non-null  object \n",
            "dtypes: float64(8), int64(2), int8(1), object(2)\n",
            "memory usage: 115.8+ MB\n",
            "None\n",
            "\n",
            "\n",
            " ************************************************************************************************************************ \n",
            " ************************************************************************************************************************ \n",
            "\n",
            "\n",
            "Asset_ID: 5, \t Asset_Name: EOS.IO \n",
            "\n",
            "missing values for each column: \n",
            "\n",
            "timestamp        0\n",
            "Asset_ID         0\n",
            "Count            0\n",
            "Open             0\n",
            "High             0\n",
            "Low              0\n",
            "Close            0\n",
            "Volume           0\n",
            "VWAP             0\n",
            "Target        2302\n",
            "Time             0\n",
            "Weight           0\n",
            "Asset_Name       0\n",
            "dtype: int64\n",
            "\n",
            "\n",
            "Asset info:\n",
            "\n",
            "<class 'pandas.core.frame.DataFrame'>\n",
            "Int64Index: 1955140 entries, 3 to 24236797\n",
            "Data columns (total 13 columns):\n",
            " #   Column      Non-Null Count    Dtype  \n",
            "---  ------      --------------    -----  \n",
            " 0   timestamp   1955140 non-null  int64  \n",
            " 1   Asset_ID    1955140 non-null  int8   \n",
            " 2   Count       1955140 non-null  int64  \n",
            " 3   Open        1955140 non-null  float64\n",
            " 4   High        1955140 non-null  float64\n",
            " 5   Low         1955140 non-null  float64\n",
            " 6   Close       1955140 non-null  float64\n",
            " 7   Volume      1955140 non-null  float64\n",
            " 8   VWAP        1955140 non-null  float64\n",
            " 9   Target      1952838 non-null  float64\n",
            " 10  Time        1955140 non-null  object \n",
            " 11  Weight      1955140 non-null  float64\n",
            " 12  Asset_Name  1955140 non-null  object \n",
            "dtypes: float64(8), int64(2), int8(1), object(2)\n",
            "memory usage: 195.8+ MB\n",
            "None\n",
            "\n",
            "\n",
            " ************************************************************************************************************************ \n",
            " ************************************************************************************************************************ \n",
            "\n",
            "\n",
            "Asset_ID: 6, \t Asset_Name: Ethereum \n",
            "\n",
            "missing values for each column: \n",
            "\n",
            "timestamp       0\n",
            "Asset_ID        0\n",
            "Count           0\n",
            "Open            0\n",
            "High            0\n",
            "Low             0\n",
            "Close           0\n",
            "Volume          0\n",
            "VWAP            0\n",
            "Target        340\n",
            "Time            0\n",
            "Weight          0\n",
            "Asset_Name      0\n",
            "dtype: int64\n",
            "\n",
            "\n",
            "Asset info:\n",
            "\n",
            "<class 'pandas.core.frame.DataFrame'>\n",
            "Int64Index: 1956200 entries, 5 to 24236799\n",
            "Data columns (total 13 columns):\n",
            " #   Column      Non-Null Count    Dtype  \n",
            "---  ------      --------------    -----  \n",
            " 0   timestamp   1956200 non-null  int64  \n",
            " 1   Asset_ID    1956200 non-null  int8   \n",
            " 2   Count       1956200 non-null  int64  \n",
            " 3   Open        1956200 non-null  float64\n",
            " 4   High        1956200 non-null  float64\n",
            " 5   Low         1956200 non-null  float64\n",
            " 6   Close       1956200 non-null  float64\n",
            " 7   Volume      1956200 non-null  float64\n",
            " 8   VWAP        1956200 non-null  float64\n",
            " 9   Target      1955860 non-null  float64\n",
            " 10  Time        1956200 non-null  object \n",
            " 11  Weight      1956200 non-null  float64\n",
            " 12  Asset_Name  1956200 non-null  object \n",
            "dtypes: float64(8), int64(2), int8(1), object(2)\n",
            "memory usage: 195.9+ MB\n",
            "None\n",
            "\n",
            "\n",
            " ************************************************************************************************************************ \n",
            " ************************************************************************************************************************ \n",
            "\n",
            "\n",
            "Asset_ID: 7, \t Asset_Name: Ethereum Classic \n",
            "\n",
            "missing values for each column: \n",
            "\n",
            "timestamp        0\n",
            "Asset_ID         0\n",
            "Count            0\n",
            "Open             0\n",
            "High             0\n",
            "Low              0\n",
            "Close            0\n",
            "Volume           0\n",
            "VWAP             0\n",
            "Target        9326\n",
            "Time             0\n",
            "Weight           0\n",
            "Asset_Name       0\n",
            "dtype: int64\n",
            "\n",
            "\n",
            "Asset info:\n",
            "\n",
            "<class 'pandas.core.frame.DataFrame'>\n",
            "Int64Index: 1951127 entries, 4 to 24236798\n",
            "Data columns (total 13 columns):\n",
            " #   Column      Non-Null Count    Dtype  \n",
            "---  ------      --------------    -----  \n",
            " 0   timestamp   1951127 non-null  int64  \n",
            " 1   Asset_ID    1951127 non-null  int8   \n",
            " 2   Count       1951127 non-null  int64  \n",
            " 3   Open        1951127 non-null  float64\n",
            " 4   High        1951127 non-null  float64\n",
            " 5   Low         1951127 non-null  float64\n",
            " 6   Close       1951127 non-null  float64\n",
            " 7   Volume      1951127 non-null  float64\n",
            " 8   VWAP        1951127 non-null  float64\n",
            " 9   Target      1941801 non-null  float64\n",
            " 10  Time        1951127 non-null  object \n",
            " 11  Weight      1951127 non-null  float64\n",
            " 12  Asset_Name  1951127 non-null  object \n",
            "dtypes: float64(8), int64(2), int8(1), object(2)\n",
            "memory usage: 195.4+ MB\n",
            "None\n",
            "\n",
            "\n",
            " ************************************************************************************************************************ \n",
            " ************************************************************************************************************************ \n",
            "\n",
            "\n",
            "Asset_ID: 8, \t Asset_Name: IOTA \n",
            "\n",
            "missing values for each column: \n",
            "\n",
            "timestamp          0\n",
            "Asset_ID           0\n",
            "Count              0\n",
            "Open               0\n",
            "High               0\n",
            "Low                0\n",
            "Close              0\n",
            "Volume             0\n",
            "VWAP               0\n",
            "Target        193106\n",
            "Time               0\n",
            "Weight             0\n",
            "Asset_Name         0\n",
            "dtype: int64\n",
            "\n",
            "\n",
            "Asset info:\n",
            "\n",
            "<class 'pandas.core.frame.DataFrame'>\n",
            "Int64Index: 1592071 entries, 1614487 to 24236800\n",
            "Data columns (total 13 columns):\n",
            " #   Column      Non-Null Count    Dtype  \n",
            "---  ------      --------------    -----  \n",
            " 0   timestamp   1592071 non-null  int64  \n",
            " 1   Asset_ID    1592071 non-null  int8   \n",
            " 2   Count       1592071 non-null  int64  \n",
            " 3   Open        1592071 non-null  float64\n",
            " 4   High        1592071 non-null  float64\n",
            " 5   Low         1592071 non-null  float64\n",
            " 6   Close       1592071 non-null  float64\n",
            " 7   Volume      1592071 non-null  float64\n",
            " 8   VWAP        1592071 non-null  float64\n",
            " 9   Target      1398965 non-null  float64\n",
            " 10  Time        1592071 non-null  object \n",
            " 11  Weight      1592071 non-null  float64\n",
            " 12  Asset_Name  1592071 non-null  object \n",
            "dtypes: float64(8), int64(2), int8(1), object(2)\n",
            "memory usage: 159.4+ MB\n",
            "None\n",
            "\n",
            "\n",
            " ************************************************************************************************************************ \n",
            " ************************************************************************************************************************ \n",
            "\n",
            "\n",
            "Asset_ID: 9, \t Asset_Name: Litecoin \n",
            "\n",
            "missing values for each column: \n",
            "\n",
            "timestamp       0\n",
            "Asset_ID        0\n",
            "Count           0\n",
            "Open            0\n",
            "High            0\n",
            "Low             0\n",
            "Close           0\n",
            "Volume          0\n",
            "VWAP            0\n",
            "Target        521\n",
            "Time            0\n",
            "Weight          0\n",
            "Asset_Name      0\n",
            "dtype: int64\n",
            "\n",
            "\n",
            "Asset info:\n",
            "\n",
            "<class 'pandas.core.frame.DataFrame'>\n",
            "Int64Index: 1956030 entries, 6 to 24236801\n",
            "Data columns (total 13 columns):\n",
            " #   Column      Non-Null Count    Dtype  \n",
            "---  ------      --------------    -----  \n",
            " 0   timestamp   1956030 non-null  int64  \n",
            " 1   Asset_ID    1956030 non-null  int8   \n",
            " 2   Count       1956030 non-null  int64  \n",
            " 3   Open        1956030 non-null  float64\n",
            " 4   High        1956030 non-null  float64\n",
            " 5   Low         1956030 non-null  float64\n",
            " 6   Close       1956030 non-null  float64\n",
            " 7   Volume      1956030 non-null  float64\n",
            " 8   VWAP        1956030 non-null  float64\n",
            " 9   Target      1955509 non-null  float64\n",
            " 10  Time        1956030 non-null  object \n",
            " 11  Weight      1956030 non-null  float64\n",
            " 12  Asset_Name  1956030 non-null  object \n",
            "dtypes: float64(8), int64(2), int8(1), object(2)\n",
            "memory usage: 195.9+ MB\n",
            "None\n",
            "\n",
            "\n",
            " ************************************************************************************************************************ \n",
            " ************************************************************************************************************************ \n",
            "\n",
            "\n",
            "Asset_ID: 10, \t Asset_Name: Maker \n",
            "\n",
            "missing values for each column: \n",
            "\n",
            "timestamp         0\n",
            "Asset_ID          0\n",
            "Count             0\n",
            "Open              0\n",
            "High              0\n",
            "Low               0\n",
            "Close             0\n",
            "Volume            0\n",
            "VWAP              9\n",
            "Target        40025\n",
            "Time              0\n",
            "Weight            0\n",
            "Asset_Name        0\n",
            "dtype: int64\n",
            "\n",
            "\n",
            "Asset info:\n",
            "\n",
            "<class 'pandas.core.frame.DataFrame'>\n",
            "Int64Index: 670497 entries, 1634867 to 24236802\n",
            "Data columns (total 13 columns):\n",
            " #   Column      Non-Null Count   Dtype  \n",
            "---  ------      --------------   -----  \n",
            " 0   timestamp   670497 non-null  int64  \n",
            " 1   Asset_ID    670497 non-null  int8   \n",
            " 2   Count       670497 non-null  int64  \n",
            " 3   Open        670497 non-null  float64\n",
            " 4   High        670497 non-null  float64\n",
            " 5   Low         670497 non-null  float64\n",
            " 6   Close       670497 non-null  float64\n",
            " 7   Volume      670497 non-null  float64\n",
            " 8   VWAP        670488 non-null  float64\n",
            " 9   Target      630472 non-null  float64\n",
            " 10  Time        670497 non-null  object \n",
            " 11  Weight      670497 non-null  float64\n",
            " 12  Asset_Name  670497 non-null  object \n",
            "dtypes: float64(8), int64(2), int8(1), object(2)\n",
            "memory usage: 67.1+ MB\n",
            "None\n",
            "\n",
            "\n",
            " ************************************************************************************************************************ \n",
            " ************************************************************************************************************************ \n",
            "\n",
            "\n",
            "Asset_ID: 11, \t Asset_Name: Monero \n",
            "\n",
            "missing values for each column: \n",
            "\n",
            "timestamp          0\n",
            "Asset_ID           0\n",
            "Count              0\n",
            "Open               0\n",
            "High               0\n",
            "Low                0\n",
            "Close              0\n",
            "Volume             0\n",
            "VWAP               0\n",
            "Target        239603\n",
            "Time               0\n",
            "Weight             0\n",
            "Asset_Name         0\n",
            "dtype: int64\n",
            "\n",
            "\n",
            "Asset info:\n",
            "\n",
            "<class 'pandas.core.frame.DataFrame'>\n",
            "Int64Index: 1701261 entries, 7 to 24236805\n",
            "Data columns (total 13 columns):\n",
            " #   Column      Non-Null Count    Dtype  \n",
            "---  ------      --------------    -----  \n",
            " 0   timestamp   1701261 non-null  int64  \n",
            " 1   Asset_ID    1701261 non-null  int8   \n",
            " 2   Count       1701261 non-null  int64  \n",
            " 3   Open        1701261 non-null  float64\n",
            " 4   High        1701261 non-null  float64\n",
            " 5   Low         1701261 non-null  float64\n",
            " 6   Close       1701261 non-null  float64\n",
            " 7   Volume      1701261 non-null  float64\n",
            " 8   VWAP        1701261 non-null  float64\n",
            " 9   Target      1461658 non-null  float64\n",
            " 10  Time        1701261 non-null  object \n",
            " 11  Weight      1701261 non-null  float64\n",
            " 12  Asset_Name  1701261 non-null  object \n",
            "dtypes: float64(8), int64(2), int8(1), object(2)\n",
            "memory usage: 170.4+ MB\n",
            "None\n",
            "\n",
            "\n",
            " ************************************************************************************************************************ \n",
            " ************************************************************************************************************************ \n",
            "\n",
            "\n",
            "Asset_ID: 12, \t Asset_Name: Stellar \n",
            "\n",
            "missing values for each column: \n",
            "\n",
            "timestamp         0\n",
            "Asset_ID          0\n",
            "Count             0\n",
            "Open              0\n",
            "High              0\n",
            "Low               0\n",
            "Close             0\n",
            "Volume            0\n",
            "VWAP              0\n",
            "Target        61816\n",
            "Time              0\n",
            "Weight            0\n",
            "Asset_Name        0\n",
            "dtype: int64\n",
            "\n",
            "\n",
            "Asset info:\n",
            "\n",
            "<class 'pandas.core.frame.DataFrame'>\n",
            "Int64Index: 1778749 entries, 539192 to 24236804\n",
            "Data columns (total 13 columns):\n",
            " #   Column      Non-Null Count    Dtype  \n",
            "---  ------      --------------    -----  \n",
            " 0   timestamp   1778749 non-null  int64  \n",
            " 1   Asset_ID    1778749 non-null  int8   \n",
            " 2   Count       1778749 non-null  int64  \n",
            " 3   Open        1778749 non-null  float64\n",
            " 4   High        1778749 non-null  float64\n",
            " 5   Low         1778749 non-null  float64\n",
            " 6   Close       1778749 non-null  float64\n",
            " 7   Volume      1778749 non-null  float64\n",
            " 8   VWAP        1778749 non-null  float64\n",
            " 9   Target      1716933 non-null  float64\n",
            " 10  Time        1778749 non-null  object \n",
            " 11  Weight      1778749 non-null  float64\n",
            " 12  Asset_Name  1778749 non-null  object \n",
            "dtypes: float64(8), int64(2), int8(1), object(2)\n",
            "memory usage: 178.1+ MB\n",
            "None\n",
            "\n",
            "\n",
            " ************************************************************************************************************************ \n",
            " ************************************************************************************************************************ \n",
            "\n",
            "\n",
            "Asset_ID: 13, \t Asset_Name: TRON \n",
            "\n",
            "missing values for each column: \n",
            "\n",
            "timestamp         0\n",
            "Asset_ID          0\n",
            "Count             0\n",
            "Open              0\n",
            "High              0\n",
            "Low               0\n",
            "Close             0\n",
            "Volume            0\n",
            "VWAP              0\n",
            "Target        21014\n",
            "Time              0\n",
            "Weight            0\n",
            "Asset_Name        0\n",
            "dtype: int64\n",
            "\n",
            "\n",
            "Asset info:\n",
            "\n",
            "<class 'pandas.core.frame.DataFrame'>\n",
            "Int64Index: 1874560 entries, 418642 to 24236803\n",
            "Data columns (total 13 columns):\n",
            " #   Column      Non-Null Count    Dtype  \n",
            "---  ------      --------------    -----  \n",
            " 0   timestamp   1874560 non-null  int64  \n",
            " 1   Asset_ID    1874560 non-null  int8   \n",
            " 2   Count       1874560 non-null  int64  \n",
            " 3   Open        1874560 non-null  float64\n",
            " 4   High        1874560 non-null  float64\n",
            " 5   Low         1874560 non-null  float64\n",
            " 6   Close       1874560 non-null  float64\n",
            " 7   Volume      1874560 non-null  float64\n",
            " 8   VWAP        1874560 non-null  float64\n",
            " 9   Target      1853546 non-null  float64\n",
            " 10  Time        1874560 non-null  object \n",
            " 11  Weight      1874560 non-null  float64\n",
            " 12  Asset_Name  1874560 non-null  object \n",
            "dtypes: float64(8), int64(2), int8(1), object(2)\n",
            "memory usage: 187.7+ MB\n",
            "None\n",
            "\n",
            "\n",
            " ************************************************************************************************************************ \n",
            " ************************************************************************************************************************ \n",
            "\n",
            "\n",
            "CPU times: user 38.4 s, sys: 345 ms, total: 38.8 s\n",
            "Wall time: 38.5 s\n"
          ]
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "- Except for the *Target* column, non of the other columns of each cryptocurrency contains any missing value. Since it is expected to use *Target* data to build a model, it is important to choose the cryptocurrencies that have the least missing value in the *Target* column. In this case, BTC and ETH have the lowest missing values. \n",
        "\n",
        "- If we look at the concise summary of each cryptocurrency, BTC and ETH have the highest amount of row numbers (almost equal to each other, approximately 2 million rows) compare to the other assets. We think that since both of the assets have fewer missing data points compared to the other assets in the specified timezone, feeding them through a model might perform better than feeding the other assets."
      ],
      "metadata": {
        "id": "tOA4E1mjxNZi"
      }
    },
    {
      "cell_type": "markdown",
      "source": [
        "## Picking two coins to work with"
      ],
      "metadata": {
        "id": "5VwnDoDzLEzn"
      }
    },
    {
      "cell_type": "markdown",
      "source": [
        "If we take the analysis above into consideration, it can be seen that BTC and ETH are the most promising coins in order to feed through a model. Also, according to [Coinmarketcap](https://https://coinmarketcap.com/), they have the biggest market cap in the cryptocurrency ecosystem and we found that predicting their future changes would be quite exciting. For these reasons, we prefer picking **BTC** and **ETH** to build two models that predict their future prices."
      ],
      "metadata": {
        "id": "RHiamVPHPL7j"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "%%time\n",
        "def create_coin_df(df : pd.DataFrame, asset_id : int) -> pd.DataFrame:\n",
        "  \"\"\"\n",
        "  This function allows us creating a new coin dataframe in order to work with the coin later to predict its future price\n",
        "  \n",
        "  Arg:\n",
        "    df (pf.DataFrame): a dataframe that will be worked on\n",
        "    asset_id: coin's asset_id that will be picked\n",
        "\n",
        "  Return:\n",
        "    pd.DataFrame\n",
        "  \"\"\"\n",
        "  \n",
        "  # we drop some columns that are not needed anymore\n",
        "  coin = df[df['Asset_ID'] == asset_id].drop(['Asset_ID', 'timestamp', 'Weight', 'Asset_Name'], axis=1)\n",
        "\n",
        "  # we prefer dropping the rows where Target has NaN data, instead of filling them, in order to avoid information leakage\n",
        "  coin = coin[coin['Target'].notna()]\n",
        "\n",
        "  # reset index to have clear view\n",
        "  coin = coin.reset_index().drop(['index'], axis=1)\n",
        "  \n",
        "  # print chosen asset's dataframe\n",
        "  with pd.option_context('display.max_columns', None):  \n",
        "    print(coin)\n",
        "  \n",
        "  return coin\n",
        "\n",
        "\n",
        "print('Bitcoin: ')\n",
        "btc = create_coin_df(df, 1)     # btc index_id -> 1\n",
        "\n",
        "print('\\n')\n",
        "print('*'*150)\n",
        "print('\\n')\n",
        "\n",
        "print('Ethereum: ')\n",
        "eth = create_coin_df(df, 6)     # eth index_id -> 6\n",
        "\n",
        "\n",
        "# we don't need the entire dataset anymore so deleting df to avoid memory problem\n",
        "del df"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "DE4AK5GAOr-y",
        "outputId": "1bde1cad-a4f0-4c13-91eb-3c8ffe6311b0"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Bitcoin: \n",
            "         Count         Open      High       Low         Close      Volume  \\\n",
            "0          229  13835.19400  14013.80  13666.11  13850.176000   31.550062   \n",
            "1          235  13835.03600  14052.30  13680.00  13828.102000   31.046432   \n",
            "2          528  13823.90000  14000.40  13601.00  13801.314000   55.061820   \n",
            "3          435  13802.51200  13999.00  13576.28  13768.040000   38.780529   \n",
            "4          742  13766.00000  13955.90  13554.44  13724.914000  108.501637   \n",
            "...        ...          ...       ...       ...           ...         ...   \n",
            "1955973   2643  42632.46500  42736.90  42607.50  42703.636250  100.797218   \n",
            "1955974   2281  42718.81500  42819.38  42690.84  42781.970571   76.339988   \n",
            "1955975   2642  42772.92125  42827.10  42690.75  42755.592500  117.429123   \n",
            "1955976   2134  42762.29000  42811.30  42694.37  42717.234286   78.049458   \n",
            "1955977   2165  42703.80250  42752.00  42636.76  42657.202500   86.518794   \n",
            "\n",
            "                 VWAP    Target                 Time  \n",
            "0        13827.062093 -0.014643  2018-01-01 00:01:00  \n",
            "1        13840.362591 -0.015037  2018-01-01 00:02:00  \n",
            "2        13806.068014 -0.010309  2018-01-01 00:03:00  \n",
            "3        13783.598101 -0.008999  2018-01-01 00:04:00  \n",
            "4        13735.586842 -0.008079  2018-01-01 00:05:00  \n",
            "...               ...       ...                  ...  \n",
            "1955973  42653.031385  0.002084  2021-09-20 23:40:00  \n",
            "1955974  42755.785162  0.003246  2021-09-20 23:41:00  \n",
            "1955975  42749.075916  0.003108  2021-09-20 23:42:00  \n",
            "1955976  42749.024591  0.002770  2021-09-20 23:43:00  \n",
            "1955977  42680.263993  0.002726  2021-09-20 23:44:00  \n",
            "\n",
            "[1955978 rows x 9 columns]\n",
            "\n",
            "\n",
            "******************************************************************************************************************************************************\n",
            "\n",
            "\n",
            "Ethereum: \n",
            "         Count         Open     High      Low        Close      Volume  \\\n",
            "0          173   738.302500   746.00   732.51   738.507500  335.987856   \n",
            "1          192   738.507500   745.14   732.49   738.260000  232.793141   \n",
            "2          120   738.332500   745.12   730.00   737.502500  174.138031   \n",
            "3          156   737.222500   744.69   728.93   737.102500  165.383926   \n",
            "4          118   736.530000   743.80   727.11   735.705000  193.078039   \n",
            "...        ...          ...      ...      ...          ...         ...   \n",
            "1955855   2291  2955.097143  2962.39  2952.39  2959.727143  717.616863   \n",
            "1955856   2534  2960.612857  2968.70  2958.39  2965.927143  776.638911   \n",
            "1955857   2321  2965.741429  2969.29  2960.60  2966.315714  976.724995   \n",
            "1955858   1971  2967.775714  2971.11  2959.22  2961.050000  674.087973   \n",
            "1955859   2023  2961.502857  2964.77  2953.66  2955.754286  738.290907   \n",
            "\n",
            "                VWAP    Target                 Time  \n",
            "0         738.839291 -0.004809  2018-01-01 00:01:00  \n",
            "1         738.268967 -0.004441  2018-01-01 00:02:00  \n",
            "2         737.994457 -0.004206  2018-01-01 00:03:00  \n",
            "3         737.303631 -0.002205  2018-01-01 00:04:00  \n",
            "4         736.163026 -0.001744  2018-01-01 00:05:00  \n",
            "...              ...       ...                  ...  \n",
            "1955855  2956.842571 -0.001309  2021-09-20 23:40:00  \n",
            "1955856  2964.322014 -0.000736  2021-09-20 23:41:00  \n",
            "1955857  2965.543538 -0.000491  2021-09-20 23:42:00  \n",
            "1955858  2964.359745 -0.000562  2021-09-20 23:43:00  \n",
            "1955859  2958.430582 -0.000346  2021-09-20 23:44:00  \n",
            "\n",
            "[1955860 rows x 9 columns]\n",
            "CPU times: user 2.14 s, sys: 56.8 ms, total: 2.2 s\n",
            "Wall time: 2.21 s\n"
          ]
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "## Plotting prices and candlesticks\n",
        "\n",
        "Let's plot coins price change by the given time in order to get familiar with how the price of each evolved"
      ],
      "metadata": {
        "id": "vDU93HoqLMO4"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "%%time\n",
        "def coin_visualization(df : pd.DataFrame, coin_name : str) -> None:\n",
        "  \"\"\"\n",
        "  Coin price change exploration and visualization by given time\n",
        "\n",
        "  Arg:\n",
        "    df (pf.DataFrame): a coin table, its price will be explored and visualized\n",
        "    coin_name: name of the coin that will be visualized\n",
        "  \n",
        "  Returns:\n",
        "    None\n",
        "  \"\"\"\n",
        "\n",
        "  coin = pd.DataFrame(df['VWAP'], columns=['VWAP'])\n",
        "  coin['Target'] = df['Target']\n",
        "  coin.index = df['Time']\n",
        "  coin.index.name = None\n",
        "  coin['Candlesticks'] = coin['VWAP'] - coin['VWAP'].shift(1)\n",
        "  coin = coin[1:]\n",
        "\n",
        "  print('\\n')\n",
        "\n",
        "  price_title = f'{coin_name} price by given time'\n",
        "  coin['VWAP'].plot(title=price_title, fontsize=11, figsize=(22,6), rot=45, grid=True)\n",
        "  plt.ylabel('PRICE($)')\n",
        "  plt.xlabel('TIME')\n",
        "  plt.legend([f'{coin_name}'], loc='upper right')\n",
        "  plt.show()\n",
        "\n",
        "  print('\\n')\n",
        "  \n",
        "  # Candlesticks show that emotion by visually representing the size of price moves per min in our case.\n",
        "  candle_title = f'{coin_name} candlesticks by given time'\n",
        "  coin['Candlesticks'].plot(title=candle_title, fontsize=11, figsize=(22,6), rot=45, grid=True)\n",
        "  plt.ylabel('PRICE($)')\n",
        "  plt.xlabel('TIME')\n",
        "  plt.legend([f'{coin_name}'], loc='upper right')\n",
        "  plt.show()\n",
        "\n",
        "  print('\\n')\n",
        "\n",
        "  price_title = f'{coin_name} Target'\n",
        "  coin['Target'].plot(title=price_title, fontsize=11, figsize=(22,6), rot=45, grid=True)\n",
        "  plt.ylabel('-')\n",
        "  plt.xlabel('TIME')\n",
        "  plt.legend([f'{coin_name} Target'], loc='upper right')\n",
        "  plt.show()\n",
        "\n",
        "  print('\\n')\n",
        "\n",
        "  # del temporary variables to avoid memory problem\n",
        "  del coin, price_title, candle_title\n",
        "\n",
        "  return None\n",
        "\n",
        "\n",
        "coin_visualization(btc, 'BTC')\n",
        "\n",
        "print('\\n\\n')\n",
        "print('X'*150)\n",
        "print('\\n')\n",
        "\n",
        "coin_visualization(eth, 'ETH')"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 1000
        },
        "id": "EYcbC7DSURuR",
        "outputId": "4725f965-2d0e-4498-e4d4-b1068cdce88b"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "\n",
            "\n"
          ]
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 1584x432 with 1 Axes>"
            ]
          },
          "metadata": {
            "needs_background": "light"
          }
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "\n",
            "\n"
          ]
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 1584x432 with 1 Axes>"
            ]
          },
          "metadata": {
            "needs_background": "light"
          }
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "\n",
            "\n"
          ]
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "iVBORw0KGgoAAAANSUhEUgAABRcAAAHTCAYAAAC9cy/0AAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADh0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uMy4yLjIsIGh0dHA6Ly9tYXRwbG90bGliLm9yZy+WH4yJAAAgAElEQVR4nOzdeXxcdb3/8feXIlYQURD5XUAtXkBFlIIgVkULgrgggorC1YuoiKiXy71e1AKC7CBgKatsQtlk3yzpmrbpXtp0T9N0SZcszZ40+zr5/v6YJSeTWc6cmcmZJK/n48GDZubMOd858z3b5/v9fr7GWisAAAAAAAAASNV+fhcAAAAAAAAAwMhEcBEAAAAAAACAJwQXAQAAAAAAAHhCcBEAAAAAAACAJwQXAQAAAAAAAHhCcBEAAAAAAACAJwQXAQAAAAAAAHhCcBEAAABxGWN2G2M6jTFtxpgmY0yeMeajofdmhV5vM8b0GmN6HH8/aoL+2xhTZIxpN8ZUGGNeNcZ8NsZ2Njs+GzDGdDn+vm6YvusEY4w1xuw/HNsDAAAYDYy11u8yAAAAIEcZY3ZLutxam2+MGS/pEUmHWmsviFpuuqQKa+2fHa89IOk7kn4laZmkcZIulPRRa+1dCbZZIOl5a+2TKZZ1f2ttXyqfifr8BEm7JL0nnfUAAACMJfRcBAAAgCvW2i5Jr0k6IdmyxpjjJP1O0iXW2gXW2m5rbYe19oVEgcUY6/l3Y8wCY0yDMabeGPOCMeaDjvd3G2P+ZIzZKKndGLO/MeZSY8ye0GduCC1zdmj5/YwxU4wxpaH3XzHGHBpa3eLQ//eFekxOcltOAACAsYrgIgAAAFwxxhwo6ceSVrpY/OsK9mRcle5mJd0p6UhJn5b0UUk3RS1ziYI9JD8o6XgFe1f+RNK/STpE0lGOZa+SdIGkr4XW2STp4dB7Xw39/4PW2vdba1ekWXYAAIBRj+AiAAAAknnLGLNPUrOkcyTd4+Izh0mqSnfD1tod1tp5oZ6PdZKmKhgYdHrAWlture2U9ENJM6y1S621PZJulOTMA3SlpOuttRXW2m4FA5U/JM8iAACAN9xEAQAAIJkLQjkXx0n6nqRFxpgTrLXVCT7ToGDPwbQYY46QdL+kMyQdrGDjeFPUYuWOfx/p/Nta22GMaXC8/3FJbxpj+h2vBSQdkW5ZAQAAxiJ6LgIAAMAVa23AWvuGgsG4ryRZfL6ko40xp6a52TsU7Hn4WWvtByT9VMGh0oOK5vh3laSjw38YY96nYC/KsHJJ37LWftDx33hrbWXUegAAAOACwUUAAAC4YoK+J+lDkrYkWtZau13B3IcvGmMmG2MOMMaMN8ZcbIyZksJmD5bUJqnZGHOUpD8kWf41Sd81xnzJGHOAgsOencHIRyXdboz5eOg7HR76TpJUJ6lf0idSKB8AAMCYRnARAAAAycwwxrRJapF0u6SfWWs3u/jcf0t6SMEJU/ZJKpV0oaQZKWz7ZkmnKJjvMU/SG4kWDpXrKkkvKdiLsU1SraTu0CL3S/qXpLnGmFYFJ6c5PfTZDgW/3zJjzD5jzBdTKCcAAMCYZKxl9AcAAABGJ2PM+xUMbB5nrd3ld3kAAABGG3ouAgAAYFQxxnzXGHOgMeYgSfdK2iRpt7+lAgAAGJ0ILgIAAGC0+Z6kvaH/jpN0sWW4DgAAQFYwLBoAAAAAAACAJ773XDTGHG+MWWGM2Rb6/3ExlvmGMabQGNNtjLk36r1xxpiHjTGlxpgdxpjL3bwHAAAAAAAAID37+10ASY9Ketha+7wx5qeSHpN0VtQyOyVdLumHksZHvfcTSccqOOTlMEnrjDH51trdSd4DAAAAAAAAkAZfh0UbYz4iaZukw6y1AWPMOEkNCs7mVxdj+Zskvd9ae43jtTxJT1trXwv9/ZCkPdbaexK9l6hcH/7wh+2ECRMy8h1zUXt7uw466CC/i4ERhnoDr6g78IJ6A6+oO/CKugMvqDfwiroDr/yqO2vWrKm31h4e6z2/ey5+VFKltTYgSaEA497Q60OCi3F8TNIex99loc8ney+uCRMmqLCw0OXmR56CggJNnjzZ72JghKHewCvqDryg3sAr6g68ou7AC+oNvKLuwCu/6o4xZk+89/wOLuYMY8wVkq6QpCOOOEIFBQX+FiiL2traRvX3Q3ZQb+AVdQdeUG/gFXUHXlF34AX1Bl5Rd+BVLtYdv4OL5ZKOMsaMcwyLPjL0ultlkj4uaXXob2dvxUTvDWKtfVzS45J06qmn2tHcgkALCbyg3sAr6g68oN7AK+oOvKLuwAvqDbyi7sCrXKw7vs4Wba2tlbRe0iWhly6RtC5WvsUEXpX0K2PMfsaYwyVdIOk1F+8BAAAAAAAASIPfPRcl6UpJzxhjbpTUJOlSSTLGzJR0o7W20BjzFUkvSfpA8C1zsaRfWmvnSHpO0umStofWd4u1dlfo34neAwAAAAAAwAjS29uriooKdXV1+V0UXxxyyCHasmVL1tY/fvx4HX300XrPe97j+jO+BxettSUKBgCjX/+2499LJR0d5/MBSb9J9T0AAAAAAACMLBUVFTr44IM1YcIEGWP8Ls6wa21t1cEHH5yVdVtr1dDQoIqKCh1zzDGuP+frsGgAAAAAAADAra6uLh122GFjMrCYbcYYHXbYYSn3CiW4CAAAAAAAgBGDwGL2eNm3BBcBAAAAAAAAl8aNG6eJEyfqpJNO0imnnKLly5dr06ZNmjhxoiZOnKhDDz1UxxxzjCZOnKizzz5bvb29mjJlio477jidcsopmjRpkmbNmjVonRdeeKEmTpyoY489VoccckhkXcuXL8/Kd5g2bZo6Ojoysi7fcy4CAAAAAAAAI8X73vc+rV+/XpI0Z84cXXvttVq0aFHktcsuu0znnXeefvjDH0qSpkyZoqqqKhUVFem9732vampqtGjRokHrfPPNNyVJBQUFuvfee/XOO++4KktfX5/23z/18N60adP005/+VAceeGDKn41Gz0UAAAAAAADAg5aWFn3oQx+K+35HR4eeeOIJPfjgg3rve98rSTriiCP0ox/9KOm6V61apUmTJunkk0/Wl770JW3dulWSNH36dJ1//vk666yz9PWvf10dHR360Y9+pBNOOEEXXnihTj/9dBUWFkqS5s6dq0mTJumUU07RRRddpLa2Nj3wwAPau3evzjzzTJ155plp7wN6LgIAAAAAAAAudXZ2auLEierq6lJVVZUWLFgQd9kdO3boYx/7mD7wgQ+kvJ1PfepTWrJkifbff3/l5+fruuuu0/Tp0yVJa9eu1caNG3XooYfq3nvv1Yc+9CEVFxerqKhIEydOlCTV19frtttuU35+vg466CD99a9/1dSpU3XjjTdq6tSpWrhwoT784Q972gdOBBcBAAAAAAAw4tw8Y7OK97ZkdJ0nHPkB/eW7n0m4jHNY9IoVK3TppZeqqKgo4xPNNDc362c/+5m2b98uY4x6e3sj751zzjk69NBDJUlLly7V1VdfLUk68cQT9bnPfU6StHLlShUXF+vLX/6yJKmnp0eTJk3KaBklhkUDAAAgi+rbunXtG5vU3RfwuygAAAAZN2nSJNXX16uuri7m+8cee6zKysrU0pJ6EPSGG27QmWeeqaKiIs2YMUNdXV2R9w466KCkn7fW6pxzztH69eu1fv16FRcX6x//+EfK5UiGnosAAADImtveKdZb6/fq9GMO1QUnH+V3cQAAwCiSrIfhcCgpKVEgENBhhx0W8/0DDzxQv/zlL3X11Vfrscce0wEHHKC6ujoVFBTooosuSrju5uZmHXVU8P4pPBw6li9/+ct65ZVXdOaZZ6q4uFibNm2SJH3xi1/U7373O+3YsUPHHnus2tvbVVlZqeOPP14HH3ywWltbMzIsmp6LAAAAyBrrdwEAAAAyLJxzceLEifrxj3+sZ555RuPGjYu7/G233abDDz9cJ5xwgk488USdd955rnIw/vGPf9S1116rk08+WX19fXGX++1vf6u6ujqdcMIJ+vOf/6zPfOYzOuSQQ3T44Ydr+vTpuuSSS/S5z31OkyZNUklJiSTpiiuu0De/+U0mdAEAAAAAAACGUyCQON1LdC/DAw44QHfffbfuvvvupOuePHmyJk+eLCk45Hrbtm2R92677Ta1trbqsssu02WXXRZ5ffz48Xr++ec1fvx4lZaW6uyzz9bHP/5xSdJZZ52l1atXD9nOVVddpauuuippedwguAgAAAAAAACMUB0dHTrzzDPV29sra60eeeQRHXDAAcO2fYKLAAAAAAAAwAh18MEHq7Cw0Lftk3MRAAAAAABkVVlDh4oqm/0uBoAsoOciAAAAAADIqq/es1CStPuu7/hcEowG1loZY/wuxqhkberT8dFzEQAAAAAAACPC+PHj1dDQ4CkIhsSstWpoaND48eNT+hw9FwEAAAAAADAiHH300aqoqFBdXZ3fRfFFV1dXysG/VIwfP15HH310Sp8huAgAAAAAAIAR4T3veY+OOeYYv4vhm4KCAp188sl+F2MQhkUDAAAAAAAA8ITgIgAAAAAAAABPCC4CAAAAAAAA8ITgIgAAAAAAAABPCC4CAAAg66ys30UAAABAFhBcBAAAQNYYvwsAAACArCK4CAAAgKwrrW33uwgAAADIAoKLAAAAyLqHFu7wuwgAAADIAoKLAAAAAAAAADwhuAgAAAAAAADAE4KLAAAAAAAAADwhuAgAAAAAAADAE4KLAAAAAAAAADwhuAgAAAAAANJSUt2i9u4+v4sBwAcEFwEAAAAAgGd9gX59c9oSXfFcod9FAeADgosAAAAAAMCzfhv8/6pdjf4WBIAvCC4CAAAAAAAA8ITgIgAAAAAAAABPfA8uGmOON8asMMZsC/3/uBjLjDPGPGyMKTXG7DDGXO5471ljzHrHf/3GmPND791kjKl1vPfwcH43AAAAAAAAYDTb3+8CSHpU0sPW2ueNMT+V9Jiks6KW+YmkYyUdJ+kwSeuMMfnW2t3W2kvDCxljTpK0QNIcx2eftdZek9VvAAAAgJiMMX4XAQAAAFnka89FY8xHJJ0i6cXQSy9KOsUYc3jUoj+W9IS1tt9aWyfpLUkXxVjlLyW9YK3tzlaZAQAAAAAAAAT5PSz6o5IqrbUBSQr9f2/odaePSdrj+LssehljzAGS/kPSU1GfvdgYs9EYM9cYMymThQcAAAAAAADGslwYFp0pF0gqs9aud7z2qKTbrbW9xphzJL1tjPm0tbYh+sPGmCskXSFJRxxxhAoKCoajzL5oa2sb1d8P2UG9gVfUHXhBvRk9qmu6Iv8ejt+UugOvqDvwgnoT1NdvJUn9/Tbp/mB/BVF34FUu1h2/g4vlko4yxoyz1gaMMeMkHRl63alM0sclrQ79Hd2TUZJ+oahei9baase/5xljyiWdKGlRdEGstY9LelySTj31VDt58mSv3ynnFRQUaDR/P2QH9QZeUXfgBfVm9Hi7Zr20t1KShuU3pe7AK+oOvKDeBPX09UtzZ8kYE39/zM6TNDzXgpGAugOvcrHu+Dos2lpbK2m9pEtCL10iaV0or6LTq5J+ZYzZL5SP8QJJr4XfNMYcLekMSS84P2SMOcrx74mSJkjamuGvAQAAAADAmMXcXcDY5nfPRUm6UtIzxpgbJTVJulSSjDEzJd1orS2U9Jyk0yVtD33mFmvtLsc6fiZphrW2KWrddxhjPi8pIKlH0n86ezMCAAAAAAAA8M734KK1tkTBwGH06992/Dsg6TcJ1nF7nNd/lokyAgAAAAAAABjK79miAQAAAAAAAIxQBBcBAAAAAAAAeEJwEQAAAAAAAIAnBBcBAAAAAAAAeEJwEQAAAAAAAIAnBBcBAAAAAEDarN8FAOALgosAAADIGuN3AQAAWce5HhjbCC4CAAAAAAAA8ITgIgAAAAAAAABPCC4CAAAAAAAA8ITgIgAAAAAAAABPCC4CAAAAAAAA8ITgIgAAAAAAAABPCC4CAAAAAAAA8ITgIgAAAAAASJu11u8iAPABwUUAAAAAAOCZMcbvIgDwEcFFAAAAAAAAAJ4QXAQAAAAAAADgCcFFAAAAAAAAAJ4QXAQAAAAAAICvAv1Wlz9TqNW7G/0uClJEcBEAAADZQ45/AADgQkNbt/K31Oi3L6z1uyhIEcFFAAAAAAAAAJ4QXAQAAEDW9AWs30UAAAwTzvjA2ERwEQAAAFnT09fvdxEAAACQRQQXAQAAAAAAAHhCcBEAAAAAAACAJwQXAQAAAAAAAHhCcBEAAABZY0nvDwAAMKoRXAQAAAAAAADgCcFFAAAAAAAAAJ4QXAQAAEDWWEZFAwAAjGoEFwEAAJA1xBYBYOygQQkYmwguAgAAAAAAAPCE4CIAAAAAAAAATwguAgAAIGuM3wVAxrR29aqtu8/vYgAAgByzv98FAAAAwOhF+q3R47M3zZUx0q47v+N3UQAAQA7xveeiMeZ4Y8wKY8y20P+Pi7HMOGPMw8aYUmPMDmPM5Y73bjLG1Bpj1of+e9jx3oHGmJdDnykxxpw3XN8LAAAAJPcfbfg9AQBAtFzoufiopIettc8bY34q6TFJZ0Ut8xNJx0o6TtJhktYZY/KttbtD7z9rrb0mxrqvkdRirT02FLRcYow51lrblpVvAgAAAAAAAIwhvvZcNMZ8RNIpkl4MvfSipFOMMYdHLfpjSU9Ya/uttXWS3pJ0kYtN/FjBYKWstdslFUr6VibKDgAAAAAAAIx1fg+L/qikSmttQJJC/98bet3pY5L2OP4ui1rmYmPMRmPMXGPMpBQ+BwAAgKxiHC0AAMBolgvDotP1qKTbrbW9xphzJL1tjPm0tbYhlZUYY66QdIUkHXHEESooKMh8SXNEW1vbqP5+yA7qDbyi7sAL6s3oUVffFfn3cPym1J3sG637l7oDL6g3Qf2OhKzJ9gf7K4i6M9S+rn5JUk9PD/smgVysO34HF8slHWWMGWetDRhjxkk6MvS6U5mkj0taHfo70iPRWlsdXshaO88YUy7pREmLHJ+rc3xuYayCWGsfl/S4JJ166ql28uTJaX+5XFVQUKDR/P2QHdQbeEXdgRfUm9Hjud2rpbpaSRqW35S6k0Wz8yQNz+/oB+oOvKDeBAX6rTRnpqQE54hRfg5JFXVnqNqWLqlgvg444AD2TQK5WHd8HRZtra2VtF7SJaGXLpG0LpRX0elVSb8yxuwXysd4gaTXJMkYc1R4IWPMREkTJG11fO7XofeOk3SapNlZ+TIAAAAAAADAGON3z0VJulLSM8aYGyU1SbpUkowxMyXdaK0tlPScpNMlbQ995hZr7a7Qv+8wxnxeUkBSj6T/dPRmvEfSdGPMjtD7V1hrW4fjSwEAAAAAAACjne/BRWttiYKBw+jXv+34d0DSb+J8/mcJ1t0ud7NKAwAAAAAAAEiR37NFAwAAYBRjrmgAAIDRjeAiAAAAAAAAAE8ILgIAAAAAAADwhOAiAAAAAAAAAE8ILiKhzp6A6tu6/S4GAAAAAAAAchDBRST0g78v16m35ftdDAAAMEJZy5QuAABkSkdPn3bXt/tdDGAQgotIqLiqxe8iAAAAAAAASZc/U6jJ9xb4XQxgEIKLAAAAAAAAI8Dy0ga/iwAMQXARAAAAAAAAgCcEFwEAAJA1ZFwEAAAY3QguAgAAIGuYzwUAAGB0I7gIAAAAAAAAwBOCiwAAAAAwyr20qkznPbjE72IAAEah/f0uAAAAAAAgu6a8scnvIgAARil6LgIAACBrSLkIAAAwuhFcHIPuLezSb19Y43cxAAAAAAAAMMIxLHoMKqoPqKi+2u9iAAAAAAAAYISj5yIAAACyxvhdAAAARiFrSTyC3EFwEQAAAFnDow8AAJnX2RvwuwhABMFFAAAAZA09KwAAAEY3gosAAAAA4FJXb0Cvr6kgcA7AV5yCkEsILgIAMIK9vqZCE6bkqbOHoTEAMBzunLlF//fqBi3ZXu93UQAAyAkEFwEAGMGmzd8mSapr7fa5JAAwNtSGzrdt3X0+l8S9fR09fhcBADCKEVwEAAAAgFHsG/ct9rsIADKMUdHIJQQXAQAAAGAUq6V3O4ARhHySIw/BRQAARgFL+zUAAMCYMSonlTJ+FwBeEVwEAGAEM9yFIceNxmcfAAAADCC4CAAAgKyhVy1GKwLnAAAEEVwEAACj0qOLSlXW0OF3MQCMMoYO48CwKKpsVm+g3+9i5CzaN5BLCC4CADAK0INmsNrWLt01q0T/+dS7fhcFAAA41Lcln2Bod327zntwqW7P2zIMJQKQLoKLAABE6Qv0q6s34HcxXKEHTWyB/mC0tbNnZPyOAACMFf9avzfpMg3tPZKkDRX7sl2cEYuGZeQSgosAAET5+fTV+tQNs/0uBtLw86dXp7R8a1ev7p5dwvCrLGDSIQAAgNGN4OIYFui36ovxEGWt1S0zirW2rMmHUgGA/5Zsr/e7CKPa8tJ6PbxwR1a3UVLdmtLyd8/eqkcKSvW2i94USA0TugAAAIxuBBfHsPMfWqpjr58V872nlu3S9x9ZPswlAgB45Xf4pjfQrx217gJ6//HEu7pnztYslyjI7bDx7r7g8OlAPz0Xc93MTVU6696CyNB3wC8EzoHsY+hvAuwb5BCCi2PY5r0tfhcBY9QfXt2gu2eX+F0MjDA/f3qVvnzXAr+LkXNyZcDpLTOKdfbUxapu7vK7KGnhISb3/eHVDdpZ367OEZIXFaNPeKg/5wvAGzcNf+SUBkYWgosAht2rayr0SEGp38XACLNwa50q93X6XQzEsWpXoyRpX2ePzyUZzG2+P/ICZg8BGIw6nC4A5AB6TyOX+B5cNMYcb4xZYYzZFvr/cTGWGWeMedgYU2qM2WGMudzx3g3GmM3GmI3GmDXGmHMd7003xlQYY9aH/rt+uL4XAAAYebhNHzksUUsAwBhGoztyie/BRUmPSnrYWnu8pIclPRZjmZ9IOlbScZImSbrJGDMh9N4qSadZaz8n6ReSXjbGvM/x2bustRND/92epe8AjDlN7T2asYGJD4BcQaAlPQy/yp5MV03Dj+WLwt2NfhcBAOBQ3tjhdxGACF+Di8aYj0g6RdKLoZdelHSKMebwqEV/LOkJa22/tbZO0luSLpIka+0ca234qNqo4ECFw7JeeGCMu+rFdbrqxXVc1ACfEWhJLNXdQ4x25OCnGl4vry73uwgAMqi2pUsLSmp82XYql2bO9fFxz4Jc4nfPxY9KqrTWBiQp9P+9odedPiZpj+PvshjLSNKlkkqttRWO135vjNlkjHnLGPPpzBUdGNv2Nge74Xf3MbNqS1evvjltsbbVuJspN5d19QbU1J5bOfOAdLi98SZGO3LwUwEYyXr6+vW3uVvV0dPnazkuemyFfjG90NcyJMK5Pjlii8gl+/tdgEwxxnxN0q2SznG8fL2kKmttvzHmUkmzjTGfCAczoz5/haQrJOmII45QQUHBMJQ6N0R/11hD68bS/kBsbW1tg+pBR0ewx+KqVatU8X5v7RSjpV6tqupTSXW3rn9xmX43cbzfxUnLrSs6Vdrcr+nfPChj64yuO+kaznozEuqo81gsO8i/NsP29mA5Vq8u1KvNAX3+iP114HsSPxok2r+ZqjfVLV2u1rN3b7ckaevWrSro3Jn2djFgnyMnVCZ+075A8IF8yZKlOihGHcv0OQdB1dXdkX+P1v3rtu7U1XZJkjYXF+vgpm1ZLlXmjdbfzy+pnnPml/XqueIe7dy1Rxd98oDsFSyJPQ3B63am6kO/4xky2Tq379ihgt49CZcp3Rd8ZG9taRm1dTbd61XR5s06sGFr5gqUA/Z1Bzuu9PT0jNrfPRNy8V7H7+BiuaSjjDHjrLUBY8w4SUeGXncqk/RxSatDfw/qyWiMmSTpeUnfs9ZGji5rbaXj388aY+6TdLQG94IMv/+4pMcl6dRTT7WTJ09O/9vloPLGDmn2wkGvRX9Xa600Z2bCZTD2FBQUDKoHB60pkNrbddppp+m4Iw5ObWWz8ySNnnrVtnGvtGGdPnL4RzR58il+F8ezzp6ASmfPlpTZ3ya67ng2nPVmBNXRAwsLpI52nfaFL+jfD3+/b+U4aN1iqa1V7/1/x+kfyzaqer/D9PB/xDkeXOzftOtNaBvJthM2p3GTVFGm4z95vL5y6kd1z9ytuuKMT+iw97/XexkgSXp02wqpMZivLxPH1P4Fc6S+Pn3lK1/RIe97z5D3M3bOwSB5dRukyuDgoNG6f93Wndf2rpWqq3TCCSdo8klHZr9gmZDiORHupXrO2bFkp1S8RUccebQmTz4hrW139gT01vpKXXzaR1NPk5Lhe53+/oFnyLjrDG3z+OOO0+QvTUi4vkPKmqSVy3XwBz6gyZO/nJEy5hrP16vQfjzhhBM0+XMj5BzkUm1rl7Rwvg444ADOVQnk4r2Or8OirbW1ktZLuiT00iWS1oXyKjq9KulXxpj9QvkYL5D0miQZY06T9LKkH1pr1zo/ZIw5yvHvcyUFJFVqDLviuTV+FwGjRPgGhu742dHa1as/vbZRbd3DN2SmcA/J+keiXBs21B4a5lXf2p1kydzSE0rxYK1UsLVOjy3aqRvf3qyOnj5d/dI61beNrO8zJnABGlbsbiD33DFzi659Y5MKtkU/PifW1TtkIF/uIrFgXOwa5BK/cy5K0pWSrjLGbJN0VehvGWNmGmNODS3znKSdkrZLWinpFmvtrtB7j0h6n6THjDHrQ/99NvTeM6F8ixsk/VnS+dZaf5Nb+KynbwRdSJDTci2gMdo8sXinXi4s11NLdyVfeAT70WMr9Mzy3X4XA1kw0nIYvr52IF1zIHS33hPo1+trKvT2+r269Z1irSEA70nGZ4vO7OoAwBeZuE42tAcbvjq6U3vG+90La5MvlEU9ff269o1NakjQcDdSJqzbUtUSM63YcCC2iFzi97BoWWtLJJ0e4/VvO/4dkPSbOJ8/LcG6z85EGQHER4tZdvbBaNutlfs6NWtTlS4/4xODXl+1q1GrdjXqZ0mGxmDkCB8PZoSEgLZWt6qiaWDWe+ex5zy2316/V2+v36uSW7+p8e8ZN3wFBACMKpm8b/S6rvkltZkrRBzfnLZYzZ29WnHt14e8986mKm0o36fuvoCm/mhi1lWIfC4AACAASURBVMuSLXM2V+vXz63R/RdP1PcmHpX8AxnmDGpurW7VcR95v/bbb2Tcf2H0yYWei8CIdOa9Bbpj5ha/i+GbcGOiHSFhsCcW79TLq8uyu5FRdC3fUduW0fX9/OlVui1vi6qbuzK6XgzwK9C/cGutTr5lrjpDQ6zCxfCjw8GczdWqau5MvqDDudMW65fPDJ4t01n06N1Kg0rqslUXRsr1B8g1m/c2e/5sW3efFpTUZLA0Y1cmT4252MmvpLpVVcnu+6zU3RcI5mscgcL3y1urW33Z/rId9ZKkDeX7dO60xXpsMRPSwT8EF8eY0rp2v4swauyqb9fjY/gEPlJ6JYXdPnOL/vT6Jr+LkdOcv+mDC7ZndN2tXcGMFP1EZjLP50Pxr7NK1NTRq7LGYO+/VIcG1bZ26ellmRn+/+vn1ujCh5dnZF1Bo6++ri1r0oQpeSqq9B5cSFXGh0V7eIruC/Sro2dMZ8ZBFvg1FDJd+zp6PX/2mlc26BfTC1XW0JF8YcQUr2HkjbUVmjAlz/eG2AlT8nTW3wqGbXuf/PNsXfPahrjvj4SjzK8yvlJYoTV7mlS5L9iwurFin08lAQguIgVzN1frew8tHbEtS8iOEXpfnfNG634dpV8rJ+Racvb9XAaAfvfCWt08o1ildZnpLVvdkuZDmbWDglej7VicuznY42jx9tSS/+eiVH6bq15cpxNunJO9wmBMGSm54LJhV32wo0JH7+gK1rd19ykwzM840dXotTXB3L+pXA+zdY3aOcwdUt5YO3TO1Vw+ylaUNvjWWzFaY3uP30UAJBFcRAw9gf6Yr1/14jptqGiO+z7Glsiw6FH24J1r/LqxYgKGkee8B5f6XQRJjpyLLn/0cA+a4X6oc8PakdszaTTzEteZVVSd+YKMMRwKA0b6NS2d8o/WdAQn/mWOrnk1fu+5XBf9m57/0FK98O4eX8rixobyoT3srnl1g3bGCKxurBi+nvZuXfLESp07bbHfxZAUlcpldB6eGCEILmIITkpAarJ5yPh1OAaydCIgUJN5ufaQG37wXLK93ueSeGM1sE+tRl9vWz8CA9na4mj7bYCRYKRN2pWKN9cN7T2XDcNxK7SxolnXv1nkevn/+uda/e6f/s4g/dqaCv3vK7EDvM1pDOUfDn7f3o6+oxEjEcFFDOH3yREjy2htwfZipCfmdm4zb2NVhted+heavmyXLn58RUbLMRplanheXWu3r9sfTk1JhhAN9MzOnfNbf79VfZu332gQPwID9IQedXLp2IAHGTiIRuCpP+fk0vXznY1VGb/3Gwu8/IRbq1v15JLM5e3PoWqEMY7gIoZINuFCuveTa8ua1N2XW7nBkLrwDRHPF9kxWvdreJipmwfTm2YUa+XOxmEoFVbtatRpt+drxoa9aa8r1+vuwpJanXzrPC2N07PSWunK59cM+jsX3Je/Tafelq+aNHNKRs/m3d9v9eiiUrV3Zy9/2khshHpxVZlmbuJBO56nl+32uwjwyUg7mq21OffcEWsfdvUGtLy0YdjL4rd3Rsl5NpXr3HceWKLb8rZ42k5PX7/WlTUNei1TwcWq5k6VNzJRU7S+QL8eW1Sac+eRXERwEdq8110ei0ycuPY0tOv7jyzXX97enP7K4CsayYZHruzn+/O3Z6yV9ZhrZ+oPr22M+d6WqpaczL032hWHrgOFu1MP5kbX0Vz/+Qr3BL/j+vKmuMv0BoJfItZX8auHQP6WWknee5hGC3+NmUVVumtWie6eXZKR9cbiNUDb05c4x3M2e89d+8Ym/fYFf4cI5rJ1MfKlDafZRdV6bqX/+eQyXQU7ewLD0is0Ez2Xc+X+JJmp87bpk3+endUGFK+c+7C4qiUr2/iDizyS5z8UO2ezm33WG+hPq1d9ovN8utfb8saOrM+enOxYau7s1RXPFg7aR31p3Cjd+k6xLnxkedwypBLk3FXfrglT8rSgJDjR26Q7F+iMuxd6Ltto9eLqct05q0R/Lyj1uyg5j+Ai1Nw5kMOiqLJZjy0eGkCYlaFWpfC2Nu/NzgUUw4cu+APGwvCw+/K3eW5ljSU8I6LTtppWfev+JZo6b2vGtjMWDFf9++xNc/TL6auHZVvDId5ui96fuXZ0p3vujf5+Xb3BB7vWrtx68N5V367j/zxLb8XIgZZLQwnHklzqgXrl82t0w1vu88llWjaqYENbtz5942w9kuMPsCPtnueVwnJJ/p3j2rr7dPOMzerqHej1lMldmOy4fHVNhTZVNKujJ/73jzdhyo7a5LNWT3l9k069LX9IkDD6Pq+rN5BegNfDMXfG3Qt1/kPLvG8zA15eXaa5xTV6bFFmjuuYnYKMt3NSuJH1X+vTH7UymnWE6m0uNlDkGoKL0NUvrVdvaAbo8x5cqgfmbx+yzKJtdcNdLIwQI+weM6tG+gNvZ0/2uvu72TVVzcHhnunOCmit1YbyfSPuAcir0rr2YdlOa1ef5pfUDnk9+rdt687tpOvhFn43tSOXZovOdDnCv5tz8pps8bLuklAvntkJZnn265fp7gsMChSMRblyXDit3t2o2UUjd4hlbahXcibSUySTiduVkXLL43dVfXD+dj29bLdeXFU29M0M7sNEv8d3H1qq/35xvav1ODuduDErdMyFnyXDomfennTnfH3mL3NSWvcguXfKGWyYyhfrWSPdapTru9ZvI+VclwsILkJ1rd3aVOn+YT6XWq7hn8iEB9SHrPBjv/7lX6MjXcGMjVX63sPL9K9heEDLNZkaMpuOhxe6b51fvmN4ZpRucAxHSnaTmMqRF+i3WjzMjW/pDmeMftD2ctP8SmG5dtW7D2p7CUQl+oTf9/nnTF2sT90w2+dSDL9VuwbSJuTilf+iR1foyudH/lB2v4Nhu+vb9eyK3XHfz2Txtla3JuxRl0l+BQjCaTacKV/8uMfb4HJ4sPOatmpXY0rXuJkJgvtNLmd7jvczLdmRnWttU3uPzrq3wFUvzVhGcuAp0tiaiyf0HMR+So7gIiS5O1gyk58l3GOEo3OkG9aZRjEs0p0owo1E55pM9YQJ3yDuHKYefbnkp0++6+lzfp2R/8NjeVNVua9zyGtT523TV5PkFkq2Xx5fvFOXPrVK87fUeC7b8yv3aMKUvKS5RsOHR9rDokP/jz6Hp3L8/fG1jfrug7FzdGWKm+8bXeRzpi7S81nKw7etplUTpuRpd327ysZowvuKpoHjqKSqRROm5I3o5P9Flc2aMCVv2BsI4hnORttEp5GLHluhG9/e7KJ3bnono56+fp07bXHWA8K58sQR6xSbiXtpt6duL42Pt8/cokufWuV6+f/657qUtxGLtVatXb2D9k93b+IcvF7N21KjnfXtejTNYcvJfgYvt7h/em3jkHzIsWpMsDfj4EDh7KIq/c9LiX+PkRwYHU4887pHcBERr8fIgZZpkRunXLnSI22xfsum9h61kZdCkjR17ladetu8lD/nx4UsmzcZfaGW+0Sz0Q/MYpt2UrnQetJbzUi0uyG9gKqXfT+S9nN9W7ceXLAj8nesINHNM4oj/04WcCtrDO7v6jQC87flBbeXbPKSsEzt78iw6EhAIzWpnOO9XPLDAZZY3zfePthe26Y/ZykP3+trg/dIsxIM084l4aB1XyA7D+ThdAznhYLM+zp6VLB1aNqEXBbuibkgRroHP+TKA2x4WOyji0p1+h35QxfI0D18uEHl3Z2je4bk51bulhQM1E2YkqeyhoHrTuHuRk2+Z2HG0ixk4njfE+M+oiiFEW6Z8PLqcn32prnaWe+tN+FwSnbUpnNcv1xYPiQHa8xrooZey698fq3ecplLkcdyd9hPyRFcRIjV8tL4F/eXVpcPW0k6ewJ6Y21FTubzcWttWZNqh6EXmJ+iL2KtXb2RVtGTb52nL94x35+C5ZgHFuxQfVuP38VwJZsPNuHgy5sxJmcYWo70xOuZNRZE33QW723RFc8WDsmFNJo0d/bqvAeXaGdd/IeQcN6+69/clPL6E1+Khq/XSaZ6Mw0ZFp3jw6JG+3G8raZVe2P0rE3XX2cFe7t0ZDk3ZDgQ9ctnCnXZ06vV0pXbOVdHguE4FntC14S+QL8mTMmL2eN3Wv521bQM7e3WnqFhzF4bShZtqxtROU/Dw6LDNlTsi/zGhXuatLuhQxsrmj2f6cK/x7T87Tr2+llp589+atnuIa+d9+BS5Rd776Gf6ozN+VuCAf9Sj0OVR4pMPetmeqK3ka6utVsl1ZmbPHYkNaL7jeDiGJJsyJXbAyfb559b84r1+1c2aMUIbsn8/iPL9Y1pi9XU3qMX3s3OEC2/RSYBCFWIs6cu0mm3D7Rw+9Fzsas3oKXb67VyZ4OaXeZ2cWNPQ7t21LbGfT8yfC9jWxwIJPxt3jb1Jzl2veoL9A+a+Ww4hmIlfPDM0ObDv8d+Y/BmIDoQ8/tX1mtucY2212TvBj1R8Ke7L5C0/ja2pxd8n7+lRkWVLYN6JFY1Dw7WPFJQqr5Av7ZWxz+O43FzXGTiupg0F2TkPJNmzsWo7zMcN83rypI/WO6obRt0fnCzT4czxcrmysw9qDh9477F+tJdC+K+v6u+3XWw4NfPFer3rwyetGG4nhlLQ8H9QCD3H1JbunqT3hO7lck66LUXsRfhmXw7Q0G6u2Y5hl4mKUA44LigJLVgU1dvQJ+5cfaQSXdS+b7balr1s6dWeZopPNOnukC/zVpaALepF6y1WrYj+LxUErq+ZSuHZbI8u4l+x5iT2bhYz/TluyP/3p7jgUZPuYU9HOyx7gGcQdhU1jnSJ6KMZ/I9C/XNaUsyvt5RFoPNCoKLY8i+jvgPcG4ePIfr/BPu8dfWlcKQqxw82vd19Orql9fr+jeLMtp6kjOiKkSs1u3hduPbRfrpP97VxY+v1C+eWS0peDObbvDia/cU6Oypi5Mul61j5BPXzUz7O8Ry9UvrU565L9YDWUl1S+RBJZlEgZFEwyCdJkzJS5jGoX8MD4v2Klun0E/+ebaufjnxDJUNobqdSu+tW2YUa87mwcNTndeBSXcODdZ09ga0uyH1B8FEE6C4CQaUN3YkPH7d7vqBtAGJlzv/oaX6yZMrk64v+qHC76vo2VMX6aK/r4j8HSlPzO87/Af30hQnH9pZ16YJU/IS9qhNpi/QrzPvLdDv/ukuJ92czTV6Y22od/hwTAMeg9/1KJm+QL8+d9NcT72YnQYaWNMv09B1Zn8vJsx/7PJX7EoxB15Vc5faewKDA5kpCveU3ZnShFKeNzdIV29g0LDjv83dqjPuXqiKptSuK8YMzX9orR10Tk6nN/P++6X3eO/mfrO5o1cTpuRp8j0L1eGi8SOV32CnI1DW4ngWzE8jt3FYUWWzvnzXgoTPxE6PLy5Nml7CdeccF8tMmJKnXz1bmHihGNubs7lGv35ujbuCxFhVrp+3U9WeZu9deEdwcQwJJDizT3ljU9Iu624uHm4lXld40hd3Xiks1zHXzhzSUyVaUWWzmrIQoEmksT1489DbF/vblDd26K5ZJTkZHE0mFy9IzpneSqqCAd2LH1+pU25NPefh2+srB+XF8ZuX75BM3qZg74EnFu/UMdfmDRm6I0lvravUhCl5kb/vn799yDLfnLZE17y6QWUNHfrB35dHbv69cnOf9sSSnXHfy1juxhHIj6+cbJszkszafdhBB0iSjvrg++IuY60ddJ58atmuyI20294+X4oRcEwm0am5tqVrYGheggXPuHuhzvhr6tuOJ9lPvLGiOdKTJZZ4RR2u69AL7+6Ju62tNUN7lib8vj5cgOIFXqqaO/XpG2ZHGhPDua7SmbU+fN+2dHvqs6oPXKOHZyeNlLNtX6iBLBKE9Wg0X1+G/ZbUU+8tD9L8yT51w2z9fPrqyN/LQumkUp0opaSqdVCPvEwbNy7+F61o6tCEKXlJr8vxlDd2qL27L3Ke89Jgl0zAWnX3DX1OzMQ16s5ZW1S5r1MrEqQCc5qWH7zn7UqQEzlZapFtMa5rYbE+Mi/J8PNYv67X0X4mFx/mchgT0iZHcHEMSXZzWtmU+Zw/0cInMWfX+p88uVIvObrLh1um3F6s3wrlcEs2M+x5Dy7VhY8sS6W4WffbF9bq0UWl2lKV+lA9v+Xi5DzOm/3wv9eXp5bnJezql9brOw9mvkt9Lrp95pa4v+PjiwcH8TYlaISYNn+b1uxp0tzN7iY8OGfqokF/Z6ou5VKdHG4j8XE33HMx0cPOdW9u0jHXzoz5ntthwq0eUzXEq05fuGO+6wBFvFb0CVPyXE/kkungX3ivhc+V2TpsZm0aPATy+jeLEuZ4Dkv0ff2M68QrVn5xjTp7AwO563w+EY3m4Fc69ovU9/i/T3t3X0ZTq7g1nMOi0wl6pyu6F3ai32LNnkZ9/tZ5kUZLb4dV5vbokhjPUqmuvbhq6Gim6HW4PXpj7Y9XC+Pnyf9daDZnt6NNop1x90L96LEVKX/nVK6/1mZmP8eSqOEtFje/Q7Lj9lUP+/rB+dsH5RW9Z05J5DdLdmrP31KjLTHqWCJ5UdfpXLSrvn1I/vD5W2p08eMrXN0frdnTqJWeg7C5nZs6lxBcHEO6kzzAJOrZ6JTp42rZjgZNeWPo8JTyFIcZOIvfF+iP2essUQubtVZT5271lJPLKzeTLKze3aib/rU57vt1rd268e0i3yZs2Fixb1DPNr/sqm8fdBOQiceqVpdD80dzS1Z0L8SE3zScEy7BnU9N68BER9H5c6yLz8fS3RfQlNc3qja0brfDqzMt3R6bmRBv3znraOW+ziE5r5L53Qvxh2W63c17GtoTNhptSjAb5Yurkk8qlo2bPivrMvdf+uLV13nFNZq+bJfrYdGpbi/bh8lvYtSd9hQeNBOdD/w48z66qDTh++H6Eh7O5/eENOHyvL2+MuXhm962l/xXaWrvyWpe5ujJPsJDOO+bty1S73sDVi/Hmazwa/cU6KRb5matfPEljy7u3depM+8tyMokQGHxNt/a1Tto1uBUa3b08m6OjWn529XQ3hNpKA5PXlK4pynFrWees/Tzy3pjNhb9dfbQIeDxvvWg19M40d88ozjuyJvtoV506TwzbN7bEvO6mKjEeRvd33N0xpusJ5MpCJyFdXONz8ANRqxV/Pt1MzVr19B7x7/N26aHFw7kkX54YamueXWDJHfHzbfuH9w5oqs3ELM3aDrXp9rWLk2YkqeFJbWe1+F6Wy1dOvPeAt0yo3jQ61c+v0YrdzZGJqhK5Ad/X6GLH0+eMiYWmurcI7g4hvQnOTGmmjslkT0N7ZowJU97Gtq1sKQ2bn60Jdvr0t5WrOvvnbNK9NV7Fqq6uUuzNlUlHTItBXuWPLBgh3746PKUtt/W3eeqe/3asqYhN7wDwZT4n7vo0RUJh0/c9K/NenbFHs2Pk4vk7fWVac8cl8jMHGjtenNdhc68t2DwzaYfw0OHaTvNnb2RQFq2NUXlpYlOubh6d2Pk32+EehGne4OZ6n6cu7lGL60u183hm44MTXyRisLdjTrp5rmDem3OLqrSK3EeXv30vYeW6srnk+dwq2vtjjyARLdqN3f2phxM/do9BYMmfYrmtadV+GOZmqDBydrMTujSmmBCo3jr+NWzhbppRnHMJIQbK/a57ik8sB0btZbwGymtRpK0qSJ+QDhhGVJYNlat8PNGP26jU1QP0PB123kOtdZmLChU29IVtzH0/IeWDjk+r35pvS54OLX7m1SkcvyefOs8fTnBJDbpui4qn+Kb64L3oPfP36775m2LvB5rGL4k1bclHjnT2N7j6mE2VW524cury7Wrvj1uYDQV8fKBxgukXP5Moc57cGnkb6/xrz2hwNfxf54V2p77zw5nB4BkwgFPa6U3tgeP8+iGk78XDG2MiPV13TynxBJv1/X2x66f4eBnKg08bqUyaYsXzu+6paol440lMa81KfTsj67Hv395/aAJtuLljFxUEfu+oKYlc/f5n7phts7460JVNXfqxreLIjkkUzmGK5o69PLqgd84fP1/dsXujJUznn2h61l0z8NkQ9K/df+ShI3jyLyUg4vGmP+XjYIg+8YNYzeer91TIEl6aMEO/Xz6av3fqxt0+TOrh8y2+J//WBV3HW6HioU5HwCXhRKvN7R36zcvrNUPHUniw+rbumPOqJbqA+pFj67QJU+sTBjsqdzXqe8/slzXRfXQzEQPq0Stj2v2NOrql9Yn7PnoVXgYfUtnZm5Q+vutZm6qUn+/1dqyJuUnyTnitNHjA+47G/emPHTAb0WVzTr9jnx94fb5g15v7uzNyk13dAty9EPHRY8OPba85hJPKeDgOGiiPxevh9edM7fo5Cz1Rgk/ZKzcORBsvfL5tfrj6xuzsr144s0S6Qy01rcNzT0bvQ97A/067fZ8/fG1oeW/7s1NOunmuTrp5tC+zNC1xcvs3je+PTBjaDaG9cTuneHt+769vlKfvWmuNu+Nfb5atC1xY1usen3+Q8t0hYck7k7RQxPzi2u0OElZwr770NLkC8XgpheIc5GFW2tjDlPNRm/Vgq21+s3zaSTGjyqTc9TI62sr9aW7FqjQ0Sjj1RfvnK9zp8WeaMx5TXQWJ1nQLBOsXE4IkcWe3qW1bXprXWXkmtjnuK97JEawJ1Wn3Dov0lBWFGcWcWutGjzub1eBDE9rHuybUb2byhs79M1pi4c0IoatSbO3YLxLhZvvEv7oxw490Pv2Uzh3b61uHdIhIBE3nQUS+d+XN2i2o6Eo3atqvHNjXxqNcM5rV6xGt1QnvUrVrvp2TbpzvrZWt+pb9y/RV/66MI21udvDXiZu6u4LaPqyXXpjXeWg1ClxZwCPs+5Y9/TWWs/5FWtbu/Wn1zfp2RV7Ul5HU3uPvvLXhfrT65sivV+9jjbKlNau3kgjT7+1mrFhr/qj6veWqpa07w2fXrZLd87aMuT1iqaOYZ/LYSTw8ggYO+kRct7494wb9m06L2L5W2pTetB+etnuyL8r93VqwpS8mAGn6JuFWkdLT9W+rsjnw/6xdJck6dTb8vWdBxwtsKH/p/qwEg5OhXsHOh+awmVbuiP4oLapslkTpuTp3jlbE36HVDi3Fi3cw2JvjBZRa63um7dN5Y6LXV+gX/nFNTEf/Kra+gclJa4NDW+M1/KfqpcLy/XbF9bqhVVl+v4jy3V5stnSknCzR//rn+uGDB1Ix6bKZhVszczwgFg3gHkbq3Teg0tj9jI+6ea5cR80vfrW/UuGHA9ujg+v9TnSoyrFj0f3xAr/HR2semzxTjVlKI9WUWXzoHNNroj33OB26P7AkMFgHZtVVDVkeNU/3029d8LaMjcPpcl/+AlT8gb1ln12xZ6s3thaWU2duy35ci4OjHDwMF6O3eib4tqWrkH5icPS/bbzo4YwRbf8X/5soS59Kn7DX7aGhw1ZJlRnmzp69POnV+tKR8Avm88ylz29WrOKUusNKjnLlDh/nJSZ66bXGEFPX3/KvV3dCAcUiyqbdcqt84bkdHtu5R49UrBj0Gtr9jRmbWTF/7y8XudOW6z+fhvpKZcNTy3bFfP1F94t0+dvy084mUO06GtYzGW83qzG0NPXr9+/siHy95NLdqokhUbK4QgqRH/N8ARgnz3qEFefr9zXGbNBLZGGtm6dO23xkA4BiUQaf9I4Qzt7ObqegThuPUhcP7zUnvAkVVLsnI3DkY+uqrnL1b1uf78d8qyVqmU76gdyRaZwzXq0YGdwtEEc0df1rjinwA0xOk7M35L6M4YzKBx9n/Fbl736nPcE33lwqX4xfXVkRKRfIwmcs84/s3yPrnpxnV5JkG/Uq5tnFEcmvHQeb1/560J98c758T42ZnkJLjLsfIQarrxwzgNv1a74rfOptKBvDPUKenVN/JOGtcFWni/cMT9ycxQrQHXrOwMn/FgPbl6Fv3asoSrPrww+jId7RYZzNoVnNy6Kk2ssXg+XWNs1Jnij+OKqssjFI9GNX1ljh+6fv12/fGZg5rtHF5Xq8mcL9eW7Fujixwf3SLt2aae+cV9mg1dO1c3BQE19irPuSXF6Fw1jS1p4+6V17brs6dWJF06isb0nmBslRgDxXxvSm90yVbF6dbppmU5/1w9tMJgY1eMw0fCV8L1TJoZFr9zZoNkxAg3nPbhUk+8tGPJ64Z70eySl6+YZmzV1bno31eE63dXbr6/eE793wNS5W7XBxaRJ338k+XBMt/Um20OvnFbubBwS6I9VTjdX15bIhATursVfuGO+znTUsfDnvDxcSMHef2+tq1RFqNd5+Hh5JkHajVSUN3boicU7XQWLOnoC2lXf7upBINzzb3dD7Ot1b6B/yLlqW1MgK8PkE3EGaePnE0wtKbyboFiy+hT9/tR523TFc2s8zUDtxvaa4H1NOGXKdW9u0oQpebrhrSLdPXvweekHf1+RtNG5vLFD61w1Tgxw/vTT5m931ZMymRveKkrpwTXc+zfZZINObu5bIvXMxfqKKpuTBjfDM9OaBOts7ugdEpTwwnlNdvayT1SHo0f4hJdMlKPXyTn83u01Jnz8us/r6Ci/i20scJGjLttpXdINBKY723o2zC6q1uOLS7Wjtk0XP7FSDy3cEXfZRHXBWqtAv9VPnnx34LUER9yMUC/mjRXN6gv0D8otPmTZDXsHXdclaV+3DX0++b3URpf13snZkcat6Oun83jr6evXgpLaSP5Tt3bUtqV8LpeCjR7hFEPVjkb9Dse9RngI+XD0zndKNp/FWLS/h888kfFSYFgM1wxHrzsuOJUJcgudetvg3Ft/fmtTJAgXLVGLoPMC4SVYuGZPo15fW6nrv/3pyGt1rd0yRvrw+9/rej3hMhZsdZ9HMnzufmfjXv3g80dHXm/u6NVl01dpXVnsC01LV68qGjt1wpEfiGz5146hcQeM208/+PzRCXtjhi8czuHn4YfOvc1d2ts8vL2y/h4KuGZqRlQvvd+eWrZb5590pA4/2P3vnkkzN1Xpty+sDyeG1gAAIABJREFU1enHHBrzgjVn8+CeuxOm5GnzzefqoPd6OZVnj5fAbnljR+QYyt9So+K9A8GCHz22QvsS9DiMBNij/45TjB/8fbk2721Wya3fSlqucPLn3Xd9Z8h7HT0B9fdb7efoIrmxollt3X16f4LfpKevXyXVLfrc0R9Mun0vwr2+f/+NT0Ze217Tps8c6a6nh+S+V8MDC+LfvKfKy7BoKTdaPMN1rryxQ9tqWvX1Tx8hSYNupPNDQcEtVa2aMCVPef/9lcHrSLaN0P9vn7lFv/rqJ1Iu48+jGj7Cx8eqUE9Qt6feeMudc98idfX2a1lpvab//AsJ1/Hk0l0qe7tI7T0B/ejUjybeTtTxPWgZWd01q0T/WLpLC6+ZHHn9jne7tLp1jX74+aN17meGJ5uP83yzryN2MCu8TKK0L7UtXfrggQfogP33c9Wg96fXN+ruH54U9/3onys8WV50Pt109DnSs4SH24UDKMl6OidrRD3j7mDjRqxzcDzOB+HHFpVm5AHwudAs4PHqa7R07mQSfTYSZHOxgXB+RLf7Lt46T7plrv7768e5WkfC9Tu+2Qk3znG8nlyy3GrZkEqnjHSHRUfr6E0v5VCy/ZSNvN3D25wzVLh3+9PLdqsq6hnmjbUV+taJ/+ZqPZ/5y5xBgStpYH+uLWvS+P3HhZ6/gsINrFuqWvSzp1clnJH6qhfXxX3v/IeWJS3b03F6SnuRaCKW++Zt0zXnfjLu+1Lqdf7sqYskpXYur2/r1m15A0OSnTmPnc+L4X+nM+zfDb/r+EiQcs9Fa+0j2SgIsi/TF+R3Nu4dkt9rRWmDbnirKM4nEosXWJQGyr6zPnbyaSl4wLs9wTlnN77o0RX657tlke7dVlan3Z4/JPiZTL+1qm/rHlQGtzcm0Uu9s2lv3MDiql2N+txNc/XtB4LDeWP9ri2hSQOic2nF4leujGjhh61UgxWldW0xJ7xJFIyKZVtNm259p1hXvehf4t/w8IR3dzW6nsmv2sXQ3OrmriGTDmVzQpj/fnGdHpy/PaXPrNzZMKilNFy/pdizvMc6zv61Ya8C/TbpcbdmT1PGJrBaFSN3Wl/Ub9fU3jOo18ct72zW+Q8t0+4M9pyOJ3x8/8/L6xMuF74xe3rZbjW19+jEv8xJuHw29NtgYvBkdT+6x8Rwn8J21bfroQWx6/e50xbrl88Ee8wv3lanC2P02JwTGpIa3VgQNruoetA1Kiz6XJ/usPzo3Zbu6IbwMVWwtS5pzuSt1S1qT9LDMfx9B3oumcg9h7OhMZzqJDo/1bzimkGNbulwk1cpfmPewAtzQ7+5c1KRsM6eYG/LL9wxX//36oYh74f95z/eHZT/6ZXC4PDEr/+tIGkZUzF/S43rSR+mOr6Pm15Zg2T5ac1NYLElzfyPE6bkxZ18IZXzk5sRz5FlMrzjWpP81nOKqodsMdF3K6psTrmxeEtVi16K6pmeaBUdPX2aXZR+rt1NFc16bsXuyN/RHRnqWrv15JKdQ7Ydq4yZuhw9tmhn8oUU//B5OkmP9JqWzPfwGo7Z6GNZvK1uUI/56MCiJP3+lQ36p6NuxfqdXl1TIWvtkMBimLVW339kub79wJKY+Tg7egIJA4uZsF8Gb3h+Pj3+SKsNLnpRDvBWpglT8vTjxwZGyZU1dHjueRg+Dqblp/bskaryxo64ow0RxGzRSNlFj67Q+vJ9+q9/rtMJN85RX6A/MmPUJU+sHDIBRCaEb6K21bTFHer0r/V7E7YIxRNe3bj90msVrWzq1Km35XvK1eRWfVu3fvTY4OHKsYo70JMr/nfyelsaHThJx7T8bZowJS+tbuyZmo03HNDI1AQ16dq8191EM1e/tC7p/jt32mJd8sTKQa/9IsFNRTKBfptg2F/Q3xwPm209yWvbH17b6Dr3ixQcPm6t1YQpeXpk4UCeosXb6yL1PZWbMGutp57PkUYJ58OFMYOGYp186zx94rqBdMXh82V0z6F9HT0xh2APt2KfJjmaV1yjG9/erCeXpNYyn0pv8Uz4xn2LdW9UHsZtNa3avLd50ENJuCd4tLizNYe8sXZoLqtYMn2tcd1z0cUyP33y3YSBhVQ6F4SXrdzXqRNunKPb8wZSmzg3Udbo7vgtb+wYkkfUKVYP2jNdBO7Cp5t+aweVq617oE6EG/7CgRzn7OEtXb2R88msBAnol2yvjxl4KE0w/DbViUVK69r0y2cK9afQkOXOnkDCocWZmCTFrd5Af9ppH6K9sS79IZ7ryvapo6dPe0LD973cS4avq3EnfpAiPeVTWb/bxsrEDdHuJjzs6g1oQUmNzntwqZ5etls1LV0DqXrinPXC3+Vb9y/RlFCew5dWlekKR2qjWI3lN7y1WVc+vzbth/3vPrRUN7w9dOLD6uYuTZiSp9Nuzx/Ue6q8sWPQUNMfP74ykjfPyv3+ziYvuZHTlUoKgEwprWvTpU+t0vVvJs+P6UyLFc8FcVK4WAXzqIY55wXIhP/8x7vJF1J2ZvkOSzV9hNvfu7mzV3fMHDohihTsTPHbF9Yo0G/11XsWptSxx3k2Wu9IzxMrl77zfsQ5sqA30K87Z20ZMsHYybfM1fdiTFi3cGtdpEd42N4EozTHIoKLY0im2ji3VLXogocHum7fl79N331oqav8gKmqaelSU3vPoPxzbV192lnXNqTV6PW1FSnPMO0UvunpcdwUtHb1anlpvaukwM6k2NHrDNsZJ3jh9sYo3CMmzFob8wHulneKtbW6NeHwmeihpG5dk6BHRTy3zCiOGfgNtzC5yTMSbfqyXZowJU8tXfEvtG+uc/eQPtx21LZpWv62jAwBL6ps0ZVJeufEmpWzMk7gw42Jt8zViX+Z46r8r6wu138tyHxrdlVzVyTosL12oEdzIGAjQTs3scVNFc269o1N+uNrG3XmvQVatasxZk6YY6+bqQsfCZ73nOeecG+V2x03TluqWmKmhLjuzU2qaemK9Ca0kv702kZNmJKnnXVtOu/Bpbry+TVxe8CkKlZ+MDez1DrzDPmhpSu1XkSxEssPt5dWlw/JaxRvmHckJ2jU+9e8uiE4zD5OxY1+8Heuf3lpfeoTY0RtZ25xzaBGh8ufKUw6g3U8q3Y36vk0H27jnV2eWLIr5rHtNkfZGXcvTJhHNFZvfjc94eNtf8aGvTFft9ZGejIOrCOoP+rcmrw3dvzj2lrp8ymOxAj35Ctv6tTOujZ9+sbZOuXWeSmtQwr2foyWKBf30u31SSebeXNtZUbTMSTT1t2n7j43x5bVL6av1tfuKRj0air3WKn0/EolQH/c9bMGpRqJJ9ElPdZEL7Hq/Gm35+sX04P3qrOLqnX6HfMjaW9SMeWNTZpbXBMpU+Q87yhjeIh/a4L7QKel2+s18Za5Q0ZdRQsHecOTOEQ74+6FgxojnUHXnzzxro67fpbnmcJTNZzDxHNReAZz532gG1c8t0YvvLtHr6wuV77jPBUvh7S1NtJwICVObeHFEpc5cNMd9puogSDWOT7Qb1UXJxd+OKdl/pYaFVU2K9BvI/fHzyzfHdmX98wp0eOL4/fEnbmpelBDm6SEDYBhznvxjY7Jb2Lli3eOYnDe5/xr/V49tminfv1cYeQ3bevuU1NHrzZUNCec3C7R9say3ErUhayKvlnNlE2VwYNqdYIbRq+uenHdkBvRk0ITO0z+5OGa/vMvZHxYr3M3/XV2SWS49jXnflLVzV16ZsVu/eEbQ/NQpNL7LrrI9W09Wl5ary/9+4e1tbo1ZovY5c+sHnLR67fxH8DOnbZYZ4fyfllZ1bR06YnFO3Xttz8d6aUZLEz8csXy1vq9mnbxyckXdAjPpvjgJbE/d+VzqQ9FDuc/SjS5w/++nHog1As3h9aczdWR3F8/eXKlalq6demkCTo0NPNhOtwnHR+Qzv1J+EY+2bFX3tiR0gzxqYo11HdrTaveDs1o2JBkhshr39ioF1cNDsBF9wwO6+u3Wle2T/39Vufctyjy+v++vGHIOsJ5GqP9892yQb0Jlu+o18uhAOBZfxtYZ01Ll474wHiVN3boyA++b/DxmoI/vrZRn/63Dwx67YePDnw/a62eWLJTF0w8KiMTHmCweEHCcCqD6GG8HT0BVbV0xT0PlzdGBawdC/7HE8GA8JI/num6fLE24xzSmr+lRgtKarTzzoH8SO/ubFCXy4eq0hQf9qLZSK/g+Ccr5zvp3Ar091sZEzyn7WckT+MvoiacSGb+ltohPcCjJ6Vy6wd/j33ekoIT+cTzkydX6nsTjxqSRzB8vzjOSL/7Z+ojQsKiz43S0F7zzq/601DvnUQ5uXpi9Ar7y9tFOnj8e7wVMiScimD8uMGvn/iXOTrxqA/E+MRg1gYngJKC9Sk/RmA1GTffwWs9d6YayZRYZXEG+sLB0qXb6zXp3w/T/2fvvuOjKPM/gH+e3fRCegGSEEICAUIJhBpK6CUioCiK3UPEXsAGWFBR9Jrnqaennp6/07Oe7azHaRThrCBWUJrYRVEpKlLm98fubGZnZ2ZnZmf75/16+TLszs7O7k555vt8n++T5raf0/Lo2s9x4vBq//e3uI7rn12PH37ah3+/8yUGVBWgtjRHc7nVm+xPdiTX+5z/j7fw0PzhttcDeBIcvtv9K6qLs0NaTyL7vTeL2U6ZgMWPmi/jdVDyb+v/cUVgaYt4INfEDWblx9/i8x9+xp0rt/ju4Ywce+draOlegsfe/gJbl7fi8ic8mcBbl7di3/7gv436fsKodJE84Y5eaRktz3/Qvuwab9mxX/Yd8AU1X928A0uffB/LZvbxmwTrZRMdrOs++9FXZxvwjD656cWN+O/5o2Om9FgkMXMxmYSpd0s+8K54Mni6uVVG2VVtG7Zj34GDpg58M7SOf2UdyBv/+zHOuX8t/tK2CWtNzI4K6PdmaN0rvf/5TkiShEk3vOybRVpphcYMoTv2/GrYYyI3biXJkxVzxytb0G3R0/jxp32Qd4i9+w4GHQ56t4MFhLVo3SwEs2ev88Pvg3nqnS9RffFTtobPKnvN9nt7w8Mxm6nyRvzX/Qd1ays6kTX5lyBD4eRC/OGilRH0W0WW8U0vbsSv+w9i9cZvNevjaN34BnPhI+8EBHmMMnGMqIfWyg69aRWueOJ9jLz+RXRTDKcGPJ0Yg5etCAhMmaH+zTdt341rnl6P0+9dE9FhjcF6gv+15jP0ufw5bNhxAF0veSqg1lU8WLPt+6B3vlq16ZY+8b7piTa0Ys5WZ9J+I0gmq/oUNfuvr+IEEz35gPkgyFPvGNdM+1SjHaC1arsZv9/v+RU1i572lWWwO0urMutQq+aX2tx73vTdgMmv1zsvhzJz7IUPqzp4FG+xauN3uPDhd/Dpjp/82hLyNcrtErYnWgKgGWALVlJDtnbb9wET40iShCWqut77DhzE3//3ieHMsFb8otG0eO/z4JkpzykyLm9pa98Wo5vLb3Z6ht3KGcId8zL8/m/E6ZqLgPaxZsbuvZ4RRWoHvPuzywUcdsvqgOGEelZ+3N6uV37Or3b+ol0KyOR3IU9UeOHD7/gml9CiHAJtl5lM0WD6XPE8WlQzC5M/uXakmWM0FMGreccXM9n4Sx5911RgUV7fY28HtsnXffqDqbaAshP9+z2/Yvde/e3resnTqF38jKnt0rNq47eov/RZv9iF3B6yWtfyRlWN+fMfXIfN2/eYLm+VaJi5mETi8bRoNNs0oJ3CbVewQI+yYLnZ846V9PVlT3+IFLe1E1rrjSvxjU66utJrGtmfY3qUAPB8x2N+14bC7DSMrS/1W+6z739CistlO3C878BBpKp6qoMNeVL68sef0TEvU/M5MxOZaFHP7mvFGfd5bj7f+/xHdC3OxvqvdqKuNDdgual/Womnzxmpux7fsNgwZBNLUvv+ufChdXhCEYCTJAn3vrYNo+pKDIeTm3Xds+tDXke4dV8SWgNELVJDcJWTFL33+Y9IcQtMvqE98+TOVzbj+ln9dIfwyPYqgqrq3e1Xb2+ynczXUATrEJJvFq593bOPOnGjF2mH3bIapwaZ0Vlrogll77qRF9Z/rZl1YVSrTW3P3v1Y48Bv/8l3ewKGggKegNhBE9fAM+5bg9a+7Zlq23ftxcsfbfe1WPSGZKlp3diY0ehtR8g1LO0mGviCSFLg7/Dt7r14bfMOw6F0EvzreIXj+gDALytDJncEbbpmKj7//mf8x7sfuoTwBRrDRe9zzrxlNXqU+V9ftY6ZuhBvMq3S67BT7n/qjqPvdu/Fpu17MLhrod/jcp2w//vfJxjdvcS3z2ekqtIn4fmeHl37ua9M0MGDEl5Y/zXG9CjF3v0H8f4XOzGwS4HdjwXAXJaOljm3v4p3PvsxIONU7jc2E2xXtsGPu7O9A0POBgU8nbT/+M0Q37/lQ+7uVVux6ZvdOG5YtY2tD489vx4IKQNSabJ3JNKCid1x4KCEFLfLl2n76iXjHHkPMjZ42X+jvQmOMjNHwRc/hF6iZ/fe/bhfVRu/+uKncMTACr/HlLdljQ7e2+vRKv/z0de78c2uX2x3qK3e+K2nY9krCZMWATC4SHHObJ0VMy6yMHwz1PPF/oOSZl2xpRaDeGYCi3peVE2CsGPPrwE9rSOuexGPnBY4rEOrh1rLggfX4UbVUOh5FmbvnPKnlbhocj0eevNTjOpegnPHd8ev+w+GFOD5y0ubcMaYWr/HjG6AteptHpQkfPDFTky9cSXOGVeHqsIsv+eDTYghX7jCkLiIX/YfQFaa59T+rCqQ+/wHXwdkfVDs08r2WPnxt5ozCqsp67yqd7dwlcogj9sMagyFSq5ppvbvIFmASn99eTO+szAcXi9DVy97/utdv5iu36lc96BlnvqACyd2111ezkBSBqb0gpCzb/sfrpzeEPD4lz/+rFmCxHZwUfG3+ubEbJH6NRr1XgF72f16jDJj1ZnSLiEcfW8tW7/7CdNvXoXHz2j2PbbxG08AdINGIDSS7n3tk4DHZt+mXfZCz0+/7sesW/+HLd/u0R3uvWbb936lKdSjI/bs3Y8/v7ARtypqF961aituX7kFN88ZgLYN3+Chtz7DygvHoFLVHokEZb0zJfn4NDPxg3rf03OsYuILOfD4/Adf4/kPvsYNKz7GgC4FuP34JlPrCje5XEWo1n+1C+u/2oWvdv6Ch9/6DOuvmux7zmx7nMgqu+ff1Rvbg+p/fkF79uaHVPdxE//4sq33ctp5D7yNS6b0tPy6D77YiTmq9o6TM3vHEw6LTiK8jzRm5abMiRPGv2JgEgI1raDYna8E3iCfco/2jS0Av4DHE6phq1aKlQOeNPtL/vUu1mz7ATes+Bi3vrQJCx9ah0UmZoXT89vnNmDO7a+i39LnfY+t+PBrXxae+qfVGmolSe2zg72rMxnP6k3f+upCqsnp/wfCcFAOX/4CNm3fjUl/fDkgS8aJLCWKDWYyQdTUGUJa5RcocTRdvUJ3MhEApgOL/3x9G7bv2qtbj1TvevjUO1+aru+ktW6tLDU1MzWXXtuyA5NuCLxxWfjQOs0i87aHRStqLmpN8BbMuk9/8Butob7Ort70rSO1UX+yMPHP/zZ/Z6sMiFXrPv0B59zfnkkz/g+Bv9f2XXvRZlA/Mhy0soOtfh+Pv/2F7zW3K/a3dZ/+4BvdsmPPr5h162q/drpyIrYLHl7nF1gE2kfGPPf+V74b9V2/7MeOPb9iyp9W+k0+ES7q9o/6uwlHG8fId3t+xX8++BoXP/KOt/yPMTmIHS/kjnXliIVdYZw9mMgOZZBNmXls5LMQJpl00rYdP5m6x5+m6vTXqmmbrMFFZi4mEcYWnTNdMVu2XeGoURkOT78bOIx5k4leaJky2DjiutBq8C1/xplhuKs3+d/wPvDGp76Z2szUyDj3gbdRX+4ZrvXm1h2atdOMeqzlC46cMXn/69scC/T88NM+jFNMDqIUzkwqin0NVzzn92+tCXEocXy7ey/O+udaTOvXKaT1XPKvd3EJ9Dt0wlWw3ExZkav+bf86+v0e7eCD+uMEC+ht2r4bNcXZvtc9uvZzW9ujzupX11l2KgtKDi7+XVF6IRY8HmRY+7F3vBb1TEY7lO2DZU9/iFNG1eDtT3/AjJtX+Y162Lx9j187YL8iY/Tjr/XbB8pOXCGAp9/9Eh9+uTMi1/un3vkSRw1qHwUzRlUf0Exdt3C4/41Pcf8bn+KlC1oMlxv/h5dx/ay+kdkoB81WTBp3qoXRQERk7NMdP5uq36yXWKKUpLFFBheTCTMXibT9VxUc/MPzgUOh1dZ7J9SwUrvw0x0/4YMvd/qyU+5evRVT+3TExf+yn4lJZJbR7HuUuMwMnw/FZY+Hp9SCmXqNodDK1L/95c0BmX1GtZ1f2/wdZv/1VUzqXYbJDeWOb2M4RbrWaqjiMbCo5df9BzHD20Gtrs05/x/tgSKXEHj87c8xqNq/TqORaNzMKusjxhqtWrBqARMeEVFSO+nuNxxZz6ubv0P3ssC6/ImOwcUkEo8TuhBFw40vODPrpJp69uQ7X9mCO18J70zcREThZGeIvhnRyLRe9rS1yYPk7IXn3v86aYdAkTVmJxmTJzXIzUhBaW66qdds2b7HV1f5vteszRxPRETOWbXxWxwfQ5NMRUrUay4KIboLIf4nhPjI+/86jWXcQoibhRCbhBAbhRBzQ30uGTFzkYiIiMgZypnM5dmmiZy065f9pkvRnHbvmjBvDRERmbF22w/BF0pAsZC5eCuAmyVJ+ocQ4lgAtwEYq1rmGAC1AOoAFAFYK4RYIUnS1hCeSzqMLRIRERERERERhUduRiyE2SIvqpmLQohSAAMA/NP70D8BDBBClKgWnQ3gdkmSDkqStB3AYwCOCPG55MPURSIiIiIiIiKisAjXZHexLtrDoisBfC5J0gEA8P7/C+/jSlUAPlH8e5tiGbvPJR2GFomIiIiIiIiIwiM5Q4uxMSw6Jggh5gGYBwBlZWVoa2uL7gaFwUef7Iv2JhARERERERERJaQ+eXvDHk/avXt3zMWsoh1c/BRAZyGEW5KkA0IIN4BO3seVtgHoAkCeG1yZkWj3OT+SJP0VwF8BoKmpSWppabH/qWLUllVbgA8/iPZmEBERERERERElnLGDGtDSt1NY36OtrQ2xFrOK6rBoSZK+AfA2gKO9Dx0NYK23PqLSQwBOEUK4vPUYZwB4OMTnkg5LLhIRERERERERhce+AwejvQlREe2aiwAwH8BZQoiPAJzl/TeEEE8LIZq8y/wfgM0APgbwKoArJUnaEuJzSYexRSIiIiIiIiKi8Hh72w/R3oSoiPawaEiStB7AEI3Hpyr+PgDgNJ3X23ouGUlMXSQiIiIiIiIiCou0lFjI4Yu85PzUREQGZjcl7aTyRERk0QnDukR7EyiO3Dd3CO47ZQjSDW4+F02tt7zezFQ3CrJSUZCVit+M6BrKJhIRUQgKstOivQlRweAiURK684Sm4AslmUum1OOSKfV4+7IJuG5W37C8x6juJb6/Zw2sCMt7EBHJLp5iPUBB1i2d3uD4Os8aW+v4Oik2DK8txvBuxXh90XgcM6Qq4PmmLgWYN6qb32MjaouDrveh+cOw9rKJWHvZRFx6SC90K8kGAGy4ejKOGhS+TtNHThuGm+cMwH1zAwai4YF5Q8P2vk54eP4w3Hh0Y7Q3w9ALC0ZHexOIyISzFdftDhmpUdyS6GFwMYlwVHRyO2xAZ9/fTV0Ko7YdW5e3ml721mMHhGUb2ha2BDx26uhuOHV0N+Rnha+n6fABnbF1eSu2Lm/F747o53t85YVjwvaeRJS85jqQvdS7Uwdbrxut6Exx0uBq565frX07IivN7dj6nGSU1QZ4giJ2DOxSoPtcc22R7++jB1fFfNAlHpw7vg5CaD+Xl5WKZTP7+D02b1QN7jppEADgr8cNBACM71mGf2gE7tSKc9L9/v3AqcPwf78ZjPQUN5YfHp5O05ribAzsUojWvh0xXBUAnd1UiSE1RTqv9Kfc95xQUZAJwBOU7VeZr7nMY2c0o6m6EJN7lzv63k6rKcnRfe6+U4LvF0Sx4j/njYr2Jmj624lNmNS7LOT1TO3b0ff3gYPJGXhhcDGJSAk2pcvyw/oEX8hBtxwzwNcoL86JbqrzyLpirDjfWk+msvGUl6XfmzJvVI3v7/mju+kuZ0eHDGtlXotUDWWnWKmDcWRTe4bh25dNwIsagUkjRlmiNx7diOfOHYXKwiw8dkazpfVa8fqicagr1W+cUnJ77tzYbOxR6FLcoTXz0lNceOrskVhxvvV9ZOmhvU0vu+HqyaaXVbZlLj2kl+5yQ2sKMaGX52ZBK1B3/7yh+PNRjfjbiYP8Hh9Z5x8gGdW9BMdHaNjzgCpPEGRQdQGOGhyY0abUpAiyvnKR+Q4qrUw52d0nDfb9fe1hfXBov06m1+ukzvmZQZeRv6tIsRuo75yfiTPHtGez/PusEbrLrr10AhZN7Ylcb8aL8EYlg9VMX3pob7ywYDTK8zL8Hi/OScfIuvAE+WXyMabFyn3HvXOtZzgqR4PInQ4XTfZka18wqQcAeAK73u/velWAtb836BjO2mjhPncM7xY8o5Uokow6CurKck2tw+g8GQ7lHTJx23FNmKZxzVs8tWfQ1+dmpGD5YX1QX97eGcvgIiW8RMtcDNbwtmrZTONhTUMVva9F2eEJesmCZYpUFWahtjTH16utJVuRjXHNzD6Y0KsMS1p7It8gsAgA3RUnfjND6vR6hLXEyj6o7N0v75BhsCRQpng+PysNXYuzg67/0dOH+/5W1twQqvSFQ/t1Qo9yz/fd38L3aOSk5mqsvHAMlrT2xLrLJuL9pZNQ2iEDl9io35QIwjkULFqG1hRi8zVTgy8IoEtRlu/vdZdNDHj+9UXjfPtgrFIHGo4bGrv17YyCXaEaVO3p3LrjeGtlLfSypgCgb0We4WvlTqzaUuv7SLWJcyUADO5aiPSU9utVsHPhxmk0AAAgAElEQVSssr2uFxRYPjIT988bhtuPb8KiqfV4UuNGZWhNEVwugaE1Rfjz0Y343yVj8e+zRuDWY/2vqzXF2Zg7oibg9WoPntqeSbj00N6aWRB/P3lwwGOy/10yFg/NH46Pl03BQ/OHB2ShfXT1FM0bHwCoKGg/zk8dZbytqYqAc7Xi/CA/99TZI3DVDOeHeVuR6+2IVAaF31s6ya+cSLM3S04O3DmV+Xb35Pb9T5kdusTEsf3WkvF48swRfkPj3C6B8yd09/27obP+Maeu0eXyHrvyLv/IaYHZqluXt+KE4dWG2W12zQuyLwHAhZP92xYvLBiNE4dXA2g/Vj+6ego+XjYl6Lqm9rGWQXix4r3l81z/ynxsXd6KvMz2tq6kWgYA/nmKfjDzkiBt36fONhf4uPHoRkudLFatv8p8pwyZd7TD95fx7ujB1trRmantiSRvLB5vuOxtGvexW5e3opvD5zM5WaauNAeF2Wk4bmgXv/s/l/dSc+NR/fG7I/phxfmj8MSZzXj3iok4RXUe3Lq81a/D9Y7jm7Dm0gkBcYmDsXLTG2EMLiaReN7F/3KM9vDYYIEyKwaphlrdMLs/AODqGQ245ZgBKMxO8/UeG92sfXil/sVePcx3en/tG4WTm417yOX3n6gzlGNCrzJcrQiWzhlSBSEE5o6swdsaAQalTnkZ+M95o/CQySFXdaU5hjdMSlb2wfPGd3csGHnVdP/GnfKG9ImzmnG/wzWBGqvah52lKW7kKguCZ2OE4vABFVg8tScqC7Mwd2QN8rJSkZ3uuciPrS/D308ejD8n2TC3sg4ZeHPJeBSpbtpiOUAVzD9PGQqXS2Dx1J5+N1BqN81pRA9FZ4FWxnJpkOB6JGllNi9p7YlVF4/1/fvDKydHPfBh5DcjuvquS8GGtdrVITMVuRkp6FeZbyrrdOWFYzTrnm1d3orTgmSnVxZmGT6vp49BAEW9DcqgHBD8uq5ssGtdiv9yzACUZ7d/9/NGdfPrNNMyrV8ndMzLREPnPN85U9YhMxVVRcG/B/lGpaIgEycMrw5oUxRmp/kNFX94/jC/83F2egrcLuEX/DtnXB0A4N65Q5CW4jJ1/i4KMrKiPkhnQu9OeRE7Px6hU3tY/g3uUmSV5qSn4DxvkC43PQVnja3DH2f3w4KJ3bHm0gm46Wj/9tWfjuqPzddMNRXU0tNkYQj+5N7lKMpJR5+KPByqaNsJ0d6xqDfxj9slML5nYDBabkvIWZMDVSVtrpgWvs6M4px03zm5tU9H3eXcLv+jsKYkx9dBLh+raSkuv/1aT4mF0SpvLRmPXoqOeOELxEre/8uPC3Qp8gSM5Q6PM8Z0w7Bu+sHouSONg6q9O+X5hl3rKclNx6H9OgV0KhsZrrFNKS7912ekxmZJh3g3tr404LGlh/b2G2FUZfPaGG2d8qy3+Q4eNL/sdYf38bW3jxlShZLc9mN6y7X+neIbl03BpN7lYQmSH9bYXgps/VWTfW2dI5oqsObSCQHtSOFtTQghMGtgBWpLc9G3It+XRS6r9Y4Cqy3N9R2vGaluzfMbMxcp4cVLAL2hc2DWnrzp6pP5asVNp1lmT2IzGj318Y4d2gVTvQ0rZWNFS7eSbGSmuXHD7P4o65COt5b499hMbvBvoOndgKW4jRsjKa72Q7e1b2Cjb1hNEWb07xzweDAd8zIwvLYYdWW5ATdGsvVXTfYLfqqDNjKthpeVXhyjOlid8zNNDZuSHTesWve50twMv6xUtWCbXJKbjmtmBg7Rv/7wvrhv7hC/LFRl0NFpKy8cg98f2c9wGOTo7iWY1q8TTmsJfbh7sBvUaFHvN/2r8lGck45nzh2Jh+YPQ0FWKgZ2KfDL6IsH71zR3ikgn39OGVWDdZfrdxZkprrxy379VqHV0grhpm7EAcAhfdtv0s8b3x2Z3t93mOKYdWJ/tkurruArF43FzXMGGHY0herdKybh8TOa0aM8F1uXtxoW/K8oyNKtezaxd7lhgFqPViBE6QRv5pKV87Rs+WH+QxfLOvgHHJRDTC3ct9t2xhhr+5feNr10QYvfv/tW5PtlImpda84aW4t7Th7sy9JTO3d8XUDgWGiGXNsph6VZCXw4Ra4nCMCvE1QZOJXjKerAVaY3mOJ2C6SluDCzsQJCCBRmpwVk/U3v3xkub7DW6oQYegFuuSTODbP7BwwH/oui81h5Tyn/HpuumYordLLYNl0zFXdolFApzE7D1uWtAbUMZf3D2KZoVAw7ry7OsjQRoHyezk4zVwpH/pmt7I9y2ZyA7GXJ//8CnjJKfzuxCYOqC7Hm0glYMKGH6ffR4wrDsTNHo2TBRtUohYUTu+PNJeODZoQlowWKDGEr1JmKE3qVBdyfnTC82m+EUbwGjgotlvXKTU/BAQv3bi4hcOm0XljS2hNXewN4D80fhmUzG3zH97tXTMSHV0723a84GSQ/ubkrti5vxW+P6IcThnXB6ovHIiPVjbysVGy+ZipOUXQcKA9hgxi+n6E17ffGwU4BcbqLhIzBxSQSzzUX5aCUHHiUh/JkmWy4KOmdxJTniOt1ZguWz696J6H/LmgB4AlMvrZoPIpy0gOWfeS04Xj7sgmGE5v0rdAeIlvj7XUtC5Jt1FxbbOum4ZWLggdrM1LdmO0daup2CSyYGNhI23D1ZLy4sAUfL5uCY4e2X7StXIw9vf3az1mpeTmz0XqQVUk+brQyOaY0lONfpw1Hh8zA/fDIQZUYbvN3sMNKhtEFGr+Z0+/ndgmc1Fwd8vtYJddZksmBk9LcDAyqLsTayybikdOG46Tmrrj12IG4XTXE1CgYsv6qydhy7VTTQ5KtqvHO7JmRGnhp7pCRamvo/Msfbff7t3w83nXiIF8PbKwoyU3Hktb22jZ3T8721RDburwV54yv8z3nROFtu+45eTBevmAMXl80TjNrOyc9Ba19O8JltrWq47zx/jdKRu17ZRDGSuDf7RJYONH/fdRBMLXFU3tqBkKU5K2xkxUuz3Ark2sYCeHZD8YpAptagTS9bDO5FttdJw7C2ksnGG7Dyxe01zBUDtlWW3f5RF/gv9QbBF2oc35VB88DLg0av2+K2+VXV07t3PHdAwLHRpecHG9G4O3HN5ke6qYeOh2KDhkpGNOjPStI77uVAzfqJoP80ax2llu9Dq+8cAzeXBIYvHn23FF46uwRmNHY2a8ja+vyVr/3UGaxCEWg1On2QLBajFYpr4c9ynL9vudxPctMT8g3paEjLpjUAxcFGV4sl3qQA4SDu3qO3UJFoNho/weAFxe24IF5QzHF23kvt03ktpsQnkzYsfVlvnVrnZv9ggaG7wjferWc7u3ssvNLq39OZd05ZSdHcU66X0YYefTq1MGvLBGg3Z5SSnEJzYzRJ88aYdgp0b3M2faTmTJUTlBeM2+eE3zSzKXTe1saJu52CeSkp2DuyBrf+W5QdSGOGdJ+D5WbkerrgDArNUjSDQD8cXY/LPKWgHK7BJZOb0AnRZveZXAONntq1jrl6sVXErEskxkMLiYRJ9sgTmWKaF0c1TcLj5w23LftAgL/PGUoHj5teMDr9MzSGXYD+BcEV55YgtV6sNI+VDeOB3Yp8M1ILJ/kjh5ciWsP64MV54/GVdN7+9WcumF2f1wzsw8emDcU47095Vr3rGePq8OAqny8v3SS7Rpq6iwBpXl9030zR8rZMpuumYq0FBe6eoecTGkox7KZDUhP8aSIp7pdfr27VvZBlxAYUFWAk5qr8dD8YRjdvaS9PpgQpn+DSQ7NAliokaH5l2MHorIwKyxZwVYy64LVjVRzuQSunB5aHSCjzNJLptTjjcXjDTNCZXJtJrPSVJmZ8s2I7IRh1fjnKUN9x/ZBnYC22yUwuaHcN2xYrm/25Fkj8OSZI/Dw/GF4Y/F4X5ZufXkuMlLdEEL4bkx6d+qAZ88dGbDue4PM6qk+7225dirOG98dd504CFuXt2L9VVO8//dkvsnDWO6dOyRo4CeYxVN74XdH9ENLj/AW+bfDJYIPR5Mps5Gt7v+hGtW9BFVFWWEfUq5X70zr3Kc8z47rGTiky5BqhfIQQjX55ls+N/WrzNcd+i0fdcGG6GpRZ1/L63JrfHDlJWvx1J545aIxujfdp7V0w7tXTMSY+tKALDc1M8OgAU/nRQdv0DAj1Y2ty1sx3caoASB8HcCtfTv6suvlzpYJvcpw7WHmZg+ePch6/TGrmYJqcoBIfZ3xDX0N81Cc3IzUgJqXgOc60buTpx1i1ARRtuGikR1qh0v4T85yrqIzJ1g2rJrbJXDGmFpfMFtP+4Q1nn9P7dMRby4Z7zcq6eLJxkGXzvmZGFJThOOHdcH7Sye1BxcVmYtm3D/P2uzreuud0qA/hDwY9V6tVZ8zXkahRYryOjm2vhSNVQV+Qdm2hcaTXfWvzNcdVaV5j+r94W88uhFtC1sCavTa5fTvqu6clCmvma19OwbtaAPgm8xUy2uLxvktpzWazgnK0iZnjqnFlmunBkwEV1mQZWkSO+UxXJ5nPMpCHkGRoyjfI58X9X67YO2MRMXgItki4CmuvfLCMX4F7G8/viloMXHZU2ePwFNnjcAD84b6hrpoGdiloP2CK4Bh3Yo0G30ydbbUtRqzSj95pqdou3IyErcr+OEwx9t7o7z5umlOI26eMwDPn6dd+yrXoHEln9jSU9w4enAVaktzfDfNz547Es+cMxIzGjtjzpAqDKkp8gVKtIZj1JXm4F+nNwfUi3JK/xK37syRVUVZeH/pJNxyzAC/3inA/6Rr5ebJJTwN1Mun9cag6kL8/eTBuHyaJyAmYC7A+9aS8Zjc4B9clOtwXDOzD0boDDWyKhxFe3MtzKx90RTrmYjHGwwVN+OUkTW4flZfvHtF4LDcbiWegsmTepdj3eUTUZCu/2MdNsDajXipaoikslj664vGweUSGNatCHO8+6FeoERWVZSFJ88cgdUXj8XW5a0ozE5Dn4o8NFUXoiQ3HZdN64Wty1vxrKq23duXTcAjpw33mxlOptUL7vcZvA3WvhV5SPH2pJ4zvi5gWzNS3Xhg3lA8dqanzk92ekrQzxNMZpobswZWRPyG9+xx7Teqvzuin+YyVoaZKTtCzNTyijXKIVdmeuSDUWbiDKuxdl4z++456f6Zd4+f0YwNV9uvZ2eW0XVP+VBD5zy/iU20aA29TzTqY/vPRzViVHf71zqrHUCAp+7eNTP74OjBVTiySb+DV8+V03vjpjmN6FeZj8fOaPbVXvTdzFlcn94+HkonW7DzjpzVHK5ZrcfWl/oCnU5Q7zcpblfYxzvJWbHK9ynOSUdGqht/nN0PNSXZ6NnRXGe5EMKv/esLLtq41mm9RD1jbLD12rnEKu8X1Med1QBvotCrbSgnjsjfSk1Jtu83aeic5+uULc/LwBNnNge8/t9njcCymQ2aE4rI5HZGjSKbfsNVU/DR1VOQm5GK6uJsTG4ox9blrX61oe0wuj962GQNfKXZqqy5NLcLVYVZAfttQXaa4eSgwShH0l17WB/DTP9QKBM8Fk7qASFEwHtZPebk7+KVi8YE7QiZM9hzT5GukZWu/uUOH1BhqYxEoom/FjnZ5kRPr9yL7BKetOfKwiz8ZkR7/b0JvcpMzS738bIp6N0pD6UdMjCkpghPn9Oe/SMfrFdN7x0wFb2Z84aynsKYHiVIdbsChvX0qcjD5IZyX7F0wFNvUKZXR/DIQZXYurwVVx7aG8U5aZjSUI7WPh3R2rejbsH4m72T0WgNVTPqha8v74CeHf0DF2O8RYaH1JgvMh5MsCEnwU64suz0FM3GlnLopTqJzGhWUO0sSvkmE/hNkElvgPaaPAB8Nzh9vNmPc4ZU4R9BMswA7R6pAosTCZ03vrul1zx+RrPpXszXF4/DzEbrN2+h2LhsCho65+HIpkrNm3VlofS8zFS/C+8YVcac2c9ZWejpVVTPTq7c5ZSZZLMGVmDr8lZTQ4f6VOTpzjqrJz8rzVdiYbmqA0N5HMjH/e+P6IdHTx+OyxSdMdfM7BNQT0ltSE0RSnNDz5AL1xDoySYzg5Wzpeplk+dYCKjHu5z0FF+QefXF4/xqasoyUl04enAlLplSjxtm9ze80ZdPl/lZqRhRpx9I0soqCEec2Tf5mQM3xHLGulb/n3L9VsplmHH3SYM0J8IJF6f6qIxG4+u1Ax8/o1mzowgwt38M6RrYJpkzpArXHtYH1x3eF2u82TFGARnlU1lpKb56q/0r831tn/Zx0drr0Jv9PBz7+OLWnobPj+5egq3LW4MGvO3624mDLF+31JTDHY2+IuX39/ricSHX+6svz8WdJzRhcav+hDQzGyvwwoIW2x1hvhrptl7dTg74qWeM1Vpv1+LskCaaVCZO6LW9o524uPywPgHlZJR1cJ2mVZu+Y15G0N/1hYUteG/pJABAQZb/tWHr8lY0dM7DMUO6oCgnXffcm5WWghtm98d9c9uvA2kpLs3jrnN+Jj64cpLfY41V+bjvlOD3GQDQySBzzsrEUrJyxT3tn49uxIdXecpVaZUrmti7HI+d0YzfzuqLv53o/9vK97Zm6ieHUglmy7VTDTtiOujUhpbvDQDrHQkTveV1Qu10VF9Xf39kP7/yLcmGwcUkYrWwqNZJYoC3eLTR8VtkYrY3dY9v1+JsX3BPXnVRTrpvSIDZwOhz547yq68gf2a5huGymf6zQ+VnpXl7WtpnsqspyfbNKKenKCcdby6ZgL8cOzDoyUweFqa1XJB2coDm2mJsuXaqX03Go71DlgZopK3fPGcA7lYUT9fSOd9c4MJu4/z4YV18M0+rM/yMZtvL1KineVDRE32iieCi3/q8gSCrH2N4N8+NujJg1rZwjF8vZbAMyHPG12FtkFm6lfpV5mser6WqQNl7Syc5EniySj3sQFmH6ZC+HQ2zZ39/ZH+/15nd9x87vRnPnzfKr0MAiI3e/KMGV+nWonr23FHYurwVhw+sQGNVAU4eYW2/dYrZUglW6jpePaPBrwGr5bzx3X2/WWufjjh7rP5NiJmOqURx9rg6PHjqMCxp7YmS3HTf0FolIQSuPawvTh3dDTMUtWO19ng5q09dNkDpzSXj8UfF8de+vugfQ3r+c94ozPeWYdHMbFU8VBdkRmirWnqU6k6EY8XIuuKAjlIgcBZYp4IGWr9msHZKTkaK7g1WqBNXCCF8HdNycPuukwbhyTM934lcS3tir+AdFfL+re54lck1xMxOJKT8ZFqTaRjJz4r/IW/K7Cat31mr6V2amxFyvT9J8tRwzE1PQVF2Gq6e3mC4/EhFh8lTZ4/wq4mq/x7tNRetUh4vay6boDnj+IWq4dqXDs3Aiwtbgk7IKNPqmFNuq3q75aClnQm4nDK5dzmO0qi/N9VgNvFQaQXyzh5Xh/kt3VBTkt0+lF+1r2akuk0nRxhlDc5o7By0nSNTzwMwZ3AVKhWdC3plRABgev9OQe/XAOuzPXcvy8G0fp3gdgm4XUK3bEf/ynwc0VSJgVWeQGZuRgrWXDrBdx+u3B976Zx/QxkRI4TAvXOH+g2zNvU6xVnc6rsvae2Ft5aMt31MxUvJi0hLnjQBcrSeg53DaUjXQry2ZQdmN2kXOPVlRbr0axgEa+Sqgy/tszv7v4fS0YOrfL23H109JaDBHyqjui/qejNmqE9mI+qKdQMbTtS+CDXjVQiBpi4FmN6/E45U/fZHNlVi5cff+j02rr4Ug7sWak7Eov4ui3PS4HYJfL1zr/e9gn+XVj/N0JpCbLh6sl/6fV5WKvIUvdNFOenYurwV73/xo+5F1yqt7/2f84YGrQcabXpDWILpV5GHdZ/9qPlca9+OKMpJR1FOOjZ+syuUzUs6Mxs749G1n/s6IfRcNb03Ln38fdSV5uDtT38wte5O+RnYtH234TLKSVjkLG61muJsbP52j2HD24jWZeHssbW48YWNttYXDiPriv3OdbkZKaguzjZdYxIwPhfL34FRJ6JeORH5tbnpKbhqhvFNfqS8v3QS0rydfjt/2QdAu+ZiPDTtu5flatZOU1/LnaojaHTDY+cdnLh/ykh1Y82lE9DBm52snNTl8TNGYP/Bg6ay8DLTPKUi6nWus75J91SrMhNAv2ZmH1wzsw/a2tqCLpso+lfmY/1Vk1F/6bN+B9NgVaaU08eZ3NHscgm8ZaLm2+3HN3m2EbAxFNz+1pd1SNcd/q4suXPR5Hp0w6eqdzV+34um1OPZ97/SfV796hOHVyMzzY2jTNZATXEJ7Hd4utrenTzHnfJclZ+VioJsT3u4tW9HPPXOlwGvmz+6G259aZOt91R/D8r7nRcWtGCztw1iVJ4o2DnMyfvjJ85sxqE3rdJ87uQRXfGXNu3vQQiBlh6B9ZKfOtu/Y2ruyBpc+e8PTG3Lh1dONqylr8U3ERK0a80DnvJEy2YOR7WqTI/ZYK6ezDR30Ile1O9xSN+OuMX7nVrt8HG7hKmEqGCinU0caxhcTCJGEzCYJfecqYuUPnFmM/buP2j42vmju+FPRzXq9gDdfdJgPPnOF3jv8x+xFv49SQdt9kC2D80yJ9QhJlrkjMwjNOoOxUanh/FGOHHSFELgT0c1Bjw+rV8nTOvXCXes3Iyrn/oQAJCR5sapo7UnDFL3RMtDc7pe8jQAzzCktg3bNV8bSg+T2RoiTtY/aqouwPqv/ANpThexH1RdgDe2fm96+SFdC4POGmfma37w1GFYtfHb4At69VMMdVN/BbFxDMWmnPQU/P6Ifpg7smvQfVO+gYr095mZ6lYMd3HuzbUmWynMTsOOPb+aXscZY7rh5hft3RCpHTOkS0BHil1GE7rYOUfIq5vSp9wvQ1JtRG0RVnz4ddDMficos5/lmovBJrKJd061P4IkeJp6XMmp71jvRtWTUWO+fWqUTZrvDXIcZrJUiNkZ3dUd1/HqD0f2w/kPrvN7TN1pu/LCMb4RN0YZXflZqfjhp322tsPqWSrDoC67nlHdSzC+ZykuPcR4+LqePx/dqDkiSLfEkfdDmT0Fq8/Vhdlp/seuRg1MdU1zw/WbXtI8rU4sSQI65mXiuXNHoWtxtmZwcXT3EkvBxd6dOqCqMAvPvPdV0JOUL0nDxDJ6RtQVI83twq8HjO9jzehbkY/DB1TgkTWfBTyntW8M7lqIIfl7fP8+dmgVGisLcNEj72D/QSmg7eYSwOqLx+K9z3/EvP97S3Mb5NOa1RmZgfaRSXqlvjzrF2isaj82zhxTi6KcNL/6i04TAP52YlPAdi2c2AMnNldj67c/GZbaCtc2AWB0UYXBxSTS34Gi0tcd3hcjaotxhCoDTTlMV+mssbX4szd7ZHT3EsOGXGVhFk5vqcUZ967RXUbr1ccMqcK9r23TXD6awwdkRTnp2HzNVMOb9nDNEmmOufd28hZuQFW+33DeuSNrfMFFoxsZediW3FumbjCYuQky2/AzkwUZTuoeQQ9nb6RH1JZYCi7+32+GBL0BNpMhMrhroW+WZ6NgyGNnNOOnvfv9Zp1WLx1LoYXHz2jGyx9pB7ejQb4JtxL01vr9BnYpwFufBO4nLiFCDjy8ctEYnPz3Nz3v7eCP2Vmj5MKRTZW4Y+VmU9kccnbECcOrMXjZf53bMC+tz/rQ/GEozknHmN+1WV6f/DvYmVzK7Pd+wvBqTO3T0dQs2XZPnVojB+SfSysDI4Fii6brPt08Z4DhLNxGX4mda1o8fcUdMlKx/qrJAVnQevuJfL7TqwMLeIJtWmULouWamX3w7ufmssvVDhtQERhchH/wvlJr9IHGF/jsOaN8WWNWhXu2b8ATkLzjhODDTPVM05jA8PXF40xnZwU7N3VSDd2/aHKPkIZ4ql0ypd7XrrZq2cwGLH70vYDH29vc7b+fPFuyUekVq/c4swZW+EZQBGtTmusgMX6+NDcDHy2bguqLnzK5hfZIkDCjfydM6l2O07z3ug+eOswvW/rqGZ4a3r07d9BsT2alpaBTfmbA/qMUSrssJz0F//jNEM1alzL1RKkLJ1mfVNKOsfWBdQxdLoHS3IyIlYhS7sntE7owuqjE4GISCfVafsmUemSnp2jW29CzYGIPX3DRbA+xTJK0/1ZT9mjKixVkpeL7n/bh4in+NVGiFSzS++w1xZ4hrnr1g8JJPUxPz+Te5fjX2s/hZFLnv04PnLlNZrSX9OrUAXee0OSrg6h0zJAqfPXjLw5sXWxQ9grKupVEtleubWELWhSBDjOZNVptGuVhp3tToXphdprbVP2/WAou9KvMD5hwJl4YnRpvmtOIYde+4PdYaW46RtaV4JUQs/GKctJ9J2Ynf8qhXQOzm4Tw1B3cu/8ghlxjLmBopcEq32RpMbOfDgpStN3MhC4HvJG4/y4YjV2/7A/+phYIIUwFFgHYii6+vnicZs1IOaioNSlRNI7/44Z2wf+9+gmOGqRd4iXcgpY78X4pMxs744wx3eByiZC+p0h+xw+eOgzf/2Q+u1iLlSw3+abQaMZ2zWBbFHlqQ1qrD2nEN5TcZM1FWXlehuladAHrtfWq6NO7HgQ7RGY3VeKBNz3Dpn93RD/07JiLjFQ3Nl0zFd0WPd2+HsWK7JaYkSk77a2a3r+zZnAxW5UJN6l3mV89ci2ZiuOxuigLW7/7yXD5TddMhUsA67/ahdc278DoHiW+LMBLVPd0gOf4PHxABU4eUa27TmWAMpQJd8wyCjbdoDGKS0t9eQfUlwfeGx5u0BEiU9/7WqU1KVx+Vio++/5nAAhrhmKsMhoREM1ElFjECV3ItIg1MLWCE4qJPEytwrucbxKPWIpAKIyoK8az547EHAsBW6co68UYuW5WX7y+eJzjtSjtGtezLCDVf/M1U3H1jAZH99Fof9qBGsNxIr0fK7NjzBRQB5z73vQ+a+BF3LPcMAcmXoiU3x3RD5N7l5ueZCUS2s+xgc9lpQb2Q149o8FyLR8l5YQL7bVx7a1PfpUyMKXVuBfw1ORRNox7deyAjbR9pu4AACAASURBVBrF+pXyMlOxJMjMsID5c6pnW0I5UgJfq65V3K0kx9LkPGoFWammh4LKdbis3rQdqSoVUpqboVkzKS8zFXefNChghlIgOpPR5HrrBsZa0ElWXeTZrv6V+agt9T/H2MmwiOR1Z3DXQkwyOQO9FXofof16Eu0rfvQcNNG549S3c+uxnrq7dso06dWMjaa/ndiEktx0HDesfbiyXOZDWTdcbrP2rcjDrIEVvtEEymtoflaaX5A73EEwo4C60oReZfjo6ilYPLUnjhnq+Zzy0FQzkzApledlBL3eul0CQgj07NgBry4ah0LFNUErAO52Cfz+yH6GIzSUL3v2nFGWttlRDgShzLS7gtVy/vdZI3DnCYHXUyN3HG8/CzhRxWpsIdqYuZhEIjEMIVzki6xWL2m8H9paPVORZfwNprpdKM3NgLnywaGzM+ucfGNtdKKf1q8j7l691W/WwVijHoq9dXkrbmnbiOuf3RCl7Wn/PquKTN5IW7zYqs9K5R0y8NXOX3T3Sq2b49UXj0VBHM3c2bNjB9x6nH6WW6iuntGA17fssPVarZ8vPTWwH1KeodFu2+rJs0bgix88veBGk15ZMbOxsy87RIvWtnbMywiY/Vxt3eWemd6NskBaepQYrkP91k63SUMZFq3lrSXBJ1mQ/XPeUHyz8xd0yEzFcXe8jg1f7woYaqn28PxhaAqSqamkVejeaP2REKxNFa0m18i6EjxxZrPfsDY5CKvepjPH1uL8B9ehPEgmypljanHTi/oTJMm/w9CaQs0SCrFKPelfMjLq3HF6F+6c72lH2Dk2/rtgNH761dls7FCNrS/DG4v9h2rmpKfgwysDh+YD0JyoUDaxl/967O6Td500KKBj6O3LJqD/lf/xe6w0NwOfe6/BaukpLtx+fBNOuedNpLgE0lJcOGVUe8CqpiQH66+abKsWptH1Vj15iVPkr7I4J912tq2191MMb1f8jrFyF97QOU9zkjEjkfje4lUch1fCgsFFMq1XR+cmqzBDeayOrS/Fn47qjykNgYGnIwdV4o5XtkRuwxJObJwVS3LTsX3XXgwIoTaoUVtsYJdC3Vm1DdcZwbuOdZdPxIED/r/H6S21mNCzDFkhzsJmh51sVa1XDCl34/lPjG8KBDyztR+UJNRf+izOndDd3PuJwNpFye7YoV1w7FDzhd8B44ymjFQ3Xl80Du998SNOvluujxjacVGYneab4CFYIMopZrPcTmqutrTety+bgKw04+NTfRNm57MaNWDlLJRBXc0H7IxYKWPSISPVV5OuX2UeNnyd2DO6nzq6G77euRcnNnc1tXw04lbqOtjttaH8HTagAocNCD7MbuGkHobBRdnZ4+o0y5ZEW7DzVTLGFuVgltwxUaFRp1bm1Lk5lPXkZaaGvY76CcO64PMfQi+vY2ciDXkfHd29BC99tN12VvYYnY4YAOiQkYKd3nIZxTlpusHFVLcLY3qU4KTmapwxplZzGTuBRSOvXjJOO4HEiX0vRg7weE7yibZYzBIMrEBKAIOLSSXUnT8zLTKj6LVOH0IITO+v3ePXvSzXV2ORJ24CgPE9S3UbRFZ59qnIXNT0isbXGczaFk52Go9a1/+j6tPwwqcHgk6kIdd0NAoCB8wWbXkLSYvW8MDMVDd+3ncAgGf25bEa2U1ONPjaMxdDW5edIZ9ar7h8Wm9L69Aayqs2sq4Yl0/rhYqCLFz79Ic6EzZ5PHFmM77ZuVf3ec3s0hQ3njt3FCoLYyPQHuxSXOLQ7LsjaiMfyMrLTMXvj+wX0jqMAjnhwPNkoPMndE/qlJNU7/U2Jz0FN81p9E2y5idM30+sfu1LpzdEexOgcSkOSYZi5IGV63WK22X5WqjFzFuuvHBMWDPjRBTDQEXZ7de6SOz3kZ4xOZnJpVE6ZDCcpsSai+SjlcIfDfJM1I0W6kWZvWDGaoMmuhLntkPugT98QIXmhChWxGIvmdOMPuLYev3eby1njfUEc7UCRC4hfHWI1KwekzyGw0u5TzxxZjMWTzWuNejkUWL3kNMutB34oHK5RVOtFzwPpQEphMBJzV0xoVcZXljYYhi471uRj/G9AmdFDKZHeW7QDMpwMxsg7mIQXLWiMDsNw4NMKBAtci1K9bXogysn4b8LRkdjkxw9fx7hnVggQ6NsQqxR75WHDeiMs8fVcVi01yF9OxlOXhWN2qbJSjJRA9Osv588GCvOj865xopw16+Vj+8g/dthkZnmDtqOclKMlMdPChdPqcctxwzAkDiq+R4JDLUmkWCNymCNq6rCyPSGjO5eYmv4qhGea41dMa2X40McoiGaDYh4ZNRTXGKxeLpc4Lt7WeCMrkp6P439wBKPbqW/2qzlqPW71JXlBmTNtvQo8QVNAGieXF3C2jEYjoCx1rA0ZY9+txLj/VTLk2eNwBtbo1dPLp5Oa0bb+n+/Gezoe915wiB8u1s/0zNahtcW43+XjEXHPP8sxagEgMNwmrxsWi889NZnGFdfFpPfv5LeZUI+J/TsGO3a15FnZpeIp3NOojhmSBes/Phb9HJgnxzd3VML+AetGdjjsO1kd1i8L28xyr3TkXj3RG0TZ8XIPapyF8pIdduaJyDRMbhIpjgd7KPok7P8XAKma0fFuva6Us5dwhO5cX3EwAoU56T56uiFYlq/TqgtzdG9SdNv7lj7htW/bWI2o+ybaHOm1XRvkfWMFOMG3N0n+QeGtDMEhaWIod1j7O6TBmHXL/tNFfmvL881LKZvRpeibMcy7kIRy/u8mfuakXXGk99YlZnmjsjMzctmNli+uVUHFsOpc34m5gypCrKUc1e03IxUrDh/NCoKMnHC314HEH8ZbiPrSvD02SPRs2N0So/EC6fiFbWlOagoyMTi1shlcsWbyQ3llu+5BlcX4vWtO1BbGrzTrLm2CKs2fqf7/HtLJ1l672BC3XWUr581MHh9WM11eHfgYGe/pYf2RkNn5zoa5Pcb3aMEy57+EFP7tLfPuhRl4ZPvfnLsvSLh4fnD8OGXOyPyXnJbPy3FhTPH1OJEi7WwnRZfV7boYnAxiTQGmSgj3hqFRqLdOxUPZjZ2xgdf7MSCieYmzgi36w7vg+ue2eCb5MEOXwPCgZ8/cY4GfUIIjK23PvxSTyjZH2a/bzmjsm9FHvbs3c/JXBwyc0BnfPr9T5g/uhv+tiqyE2T5hoEpdgIzZTrkGYQfeGNb0GVvP77Jr0df7xzh9HDVFxaMxkdf73ZmZXF0XYujTTXtmCHWJkmKtFUXj9V9LlztOzPBjFig/vzKkRq9OiVf1qJZTh/HGaluvHKR/n6aqCb1Lsfdq7diWJhKONR3zMXrW3fg2KCdC8CCiT2wauNq3TNCThQmDzRjWE2RpYnGlNozF42XO2F4ta31B7yfajO7l+UGBIwfmj8MH3zhfKBuaoO9DmYzmqoL0VTtzKRxwci/lVsInD2uLiLvSc6IzTMIhUU0h71m2Zg5zYorp/fGFU98EFDXTb6ZlIf2dQxjweB4k5HqxlUzol+8Wja2vizkQJd8PT/oQIv4d0f0wx9XfGRrxmQKn9IOGXh98TgUZ6fbbmhSoFS3Cwsm9rD8OmUj+rez+uKCh9+xHMY4fEAFlj39oS/La8X5o5GXmYr33/qfqddP798Zr2/5HhdOrseDb34W8HxuRorpzDY7w6WN1JTkoMbhdcbysKcY3jRCaMGixqp8rN32g3MbEwXFOWk4enAVThlVE+1NiZrS3HR8s2sv+lVYqGsexu1JBsO6FUVkBJiZa0O8dvyEMiLJ5Us8iJ0PX5qbgdIeod2TqgeJrLt8InJjNDhsF9sU8Sex9kCypaFzB7z3+U6kugX+OHsgyvMyMOPmVZaHtulZd/nEsAdoDunbCYf07RTwuHwhOXVUDQZ3LcDALpHpcaHocPKme0ZjZ8wIcRhlInj67JEoyLZX58YMOfNwQi/zva1GhedjRXNtEXp3yov2ZoSd8oize/zNHdkVJzZXI9U7NNtqNlRGqttw9l6tWdjZYA0vJ0tTGL6PxN/SDCe+o0dPb8biR9/Fva8FzxSONfLndwlhqxMlkTTXFuPRtZ+jKMf+KBGKf5E6bwohfCNO+nS20SZyYjt9JZMSm92alLEo0X+rRBb7U7xR2N1z8hAAwGkttZjcUI4Gh4eI5GWmIjvCPSnqm1yXSzCwmAR4j2nPvXOH4IiBFfh42RTfY3JwoFenDmGtG9YxLxPrLpuI+aMTK5Pk3rlDsShCMwR2KQp/vTk9ylOt3eNPCOELLIZqZF2xI+uJRfHR2HauNAU5L9w/S6SCylblemd6nzmAHYbyObJ7WfA6k8cMrUJ9eS5mD6oM92ZRCKycb+Vhz5UFkWs31JXl4skzR+DCyfURe08leUb7Q/sFJqFQbHJy1nSKLGYuEgqz0zhhCyWEU0fXYNXGbzGiNnEDDOHQXFuM5ih+Z3lZidPbmmyUtcxiIYPsbycOwt79B6O9GWEVA1+zrljYByiQU7Ol6r061n/3rLQUfHjlZFO1XBPdYQMqMLa+FPlZwTMXO+Zl4tlzR0Vgq8gJZo7DHuW5uO24gRhRW4wn1n0R/o3y6lMRvZEc6SlurLt8YszWk7QrkTvx5I8Wy2VgSFtiHWVElNR6d8rDW5dOiPZmkIFEbgyRRzTbgqluly8LcnZTJR5481PD5WOpBlMi4bcaW8zOlprIMsNc+zuemAksUvwxezmb1DuwDM3pLd1CmlBRj1PNgVAv1VEZMpzMJ1yHxFpoMVaz82MJu/CSXGfOtEpEEcDOx/CJ5lfrNyw6xn7jM8fWAtDerljbVjP+cGQ/TO/fKaZnt43DrzUpxMoNPhHFprPH1WHuyNgrTxOume7DKf62OPbE2rUmHtuM0cLgYpKT61AkGmajEBGFn7K9Jc/I2LU4Ozobk+BqS3Pxp6MaHatPGVa8Bsck/ixEiYsBEEoYvnHRUd0KsiEOWqhE9rFWAxElunCc56aZLHzeSzEjdqrbhbtOHIT7Thnq+PbYwUBK5Kl3RV6CY4NzvwMPKiKywKFzD888yUUefswmRPxhcDHJtfQojfYmEFEMue7wPtHeBIqy9VdNxg2z+5tadnJDOerLPbOOCgBj6ktRnJMexq2zjgEuY5WFnllD68udG24dqRtB3nBaw1EdRIknWB04ud7oyc1ddZeJ1etkrG4XhZd8qWKSUPzhhC5J7KY5jZisUdSXiIj3oMkrI9XaxAddirKw/qtdMXcT0N7zrb9h3M2B4d2K8e+zRqC3A7UcI1UfSwieo6xw6nfhd04Uu/SO8vQUN7Yub43otsjG1jOJheyLtXYlBcfgYpK6YFIPHNLX3LC3eMZeeiJrwh0c4Exrzot22ytWT7PtPd+Bz8VjkfhwauicF3whE05qrsZ/P/wakxvC23EpwMCwHfzOiChSVl08FiVBRjJ0yEjcUATbu/Z1yExFTUk2LpxUH+1NIYsS94gmQ+kpiT0inmnURLGGx2Siam8+8zdOdjUlOVh9ybhobwapOVX3TOde+ZghXfDq5h2oK8115o2IyLRY7OBLc7vQOT/TuRXG4Gek8HG7BF5Y0BLtzQgQi8darIlahEkIkSWEeEAIsVEIsV4IcYjBsqd4l9skhLhJCOHyPj5dCPGWEOI9IcT7QogFitecKIT4QQjxtve/RyPxuYiIiCIqRmJ6sdanY6UNOL5nGSb0KgvbtlB4cHSCRWH6uqb164Sty1tRkhtb9VaJkkqsXYQdEM+fiCMkEgeTlsyLZubiQgA7JUmqFULUAVgphKiVJGm3ciEhRFcAlwNoBPAdgGcAHAvgHgBfAZgmSdIXQog8AG8JIV6XJGml9+UrJEmaFakPRLGDNxxERJFx0eQe+Hb3XjTXFkd7UzSZaRLecUJT2LeDnCNYdNES+b4o1G+Mw/yIEseK80dj/B9eivj7Du5aiNe37Ij4+xJR+EVzbOxsALcBgCRJHwN4E8AUjeVmAXhMkqTtkiQdBHC797WQJOk1SZK+8P79I4APAXSJwLbHvXBH4M8ZVxfW9ZvFngYiSnTRrklTW5qLR09vRk56/FVaYXyKkoHcEmLHK1HiqSjIAoCg9Q3Vaktz4HaF5z7JqCPi/lOG4q0l4x1bHxHFjmgGF6sAfKL49zYAlXaXE0LUAxgK4AXFw6O9Q6JfFkJEZ5qsGDOspghA+BuY503oHrWZyYgodGzGxY9wT54Rrwyvc+x3iojwTxBFZjjV0crYJFHsmTeqBncc34RJve2X9ojkEF6XSyDFZS4EEc9JIgyIUjIKW5qBEGINPIFBLY4WNhJCdATwOIDT5UxGAP8G8IAkST8LIRoBPCOEGCNJ0oc665gHYB4AlJWVoa2tzclNjBm5B/YCADZt2oS2A9s0lzlw0HMyVN6Yxdv3sW/fPgDAqlWrkJsWvxemWLJ79+642w/Iug2feY6dr776Cm1t3zuyzt27d+PXXz0NydWrVyM/PbEnlIqG60dlwi1i71wdyvaEes75as9BAMDPP/8csJ53t+8HAOzYsSPmvrNIicTn3n8w8ObKkff1tk9eeuklpGhk3sTj9Sqc2/vzfs/3tf/AgdCOqa/2+v6Ot+/XLCv7zhXDMpDqFgn7XZB5oZ5zQt2HUgC89I3mLa4h+V7v5Ze1z6V2HTwoGX6mPfu856T9+w2X27DjAADgxx9/jJvjTD5Prt+wAW17Ngdd3sq+o7Xc0uEZyOB5KKw2b/4VALBt2za0tX0V5a1pF4ttnbAFFyVJGmD0vBBiGzxDmLd7H6oC8KLGovJyUCz3qWI9pQBWALhekqSHFO//reLvtUKIVQAGwzN0Wmt7/wrgrwDQ1NQktbS0GG1+3Hp51wfAJ1vQrVs3tIys0Vxm/4GDwPPP+NU0irfvI/Xl54F9+9Dc3IzC7LRob05CaGtri7v9gKz75s1PgffeQXl5OVpa+jmyzra2NqSl7Qd+3Yvhw4ajtEOGI+ulGPbsUwBCu3aEes7ZtH03sPIlZGVlBaxH2vAN8NYbKCwsREvLYNvvEZcc+G3M+nW/pz2R6hbYd8C59oTrP0/jwAEJo0aNRlpKYGdFXF2vIvB77Nm7H1jxHNxud0jv8+/t64DPPwMQf+1Cs+Jq36GYYXe/WeLejFUbv43adUg8/zQg6Z9LLfOez1wuYfh9/PjTPuC/zyMlJcVwuawtO4DX/4e8vDy0tAwPffsi4OlvPefJ+h490DJIL8+qnal9R+s64X3suEPGwhWm4e3k8b60Efh4AyqrqtDSEt1SREqxeL2KZvrIQwBOBQDvhC6DADyrsdwjAGYIIUq8s0SfAuBB7+uKAPwHwE2SJN2pfJEQorPi7y7wDJl+JwyfI64wRZuIoiGOR7ZQnOOul7jYprGGw5qJYsvckTW466TE6eB6+uyRAJw71/StyMPgroW4fFpvZ1ZIRGEVzerrvwVwtxBiI4ADAOZJkrQLAIQQVwL4QpKkWyVJ2iyEuArAq97XPQ/gH96/LwbQHcCpQohTvY/9SZKkuwCcIYSYDmC/9/FFkiStDf/Hig/xXMPCChYvJyIiPbxCxCdPfTD+emY51eRjk4qIjNSV5Ti6voxUNx48dZij6ySi8IlacFGSpD0AjtB57jLVv2+Dd2Zp1eMXALhAZx2LACwKfUuJiIgo3nA+FyIiIiKiyGBVfUpoyZKhSUREOjQuA0zAiqxwZbwxk86aUIeRH9K3o0NbQkSxIFwjvHhq5vWJkhODi0mGJzoiMtKrYwcAwMi64ihvCVFo5PrmGSlu3WXY/RReYevf4w9niXDoCxtTX+rIeogotjh1ruap2bnzLVE8imbNRYoinvaISEtD5zysu3wi8jJTnV1vpw54ccN2pBsEeoic1LU4G+eN747DB3YOvnCSKcpOi/YmUATJgQN2MBMREVG4MLhIRER+nA4sAsCf5wzAhq92IS/L+XUTaRFC4JzxdYbLJGOsZe2lE5CaEt8DVxZNqccVT36AVHd8f45IS8b9nYiIKBRlHTIAAB3zMqK8JbGPwUUiIgq7nPQUDOxSEO3NIAKQ3Nn7BQmQtXhic1ec2Nw12psRfxhdJKIwcrsEBlUX4DcjaqK9KQnlrpMGoXN+ZrQ3I2kdPqAzOmSkYHzPsmhvSsxjcJESEtvPREREFA+y08JbLoJz2xFRJAgh8ND84dHejIQzpgfr3UaTEAITe5dHezPiAoOLSWbuyK544d1tmN6/k6nlXQI4GMeROraniYiIKFa9tmic4aRDTgp1tmgiIjKHZ1tKRgwuJpmKgixc2ZyJopx0U8u/fOEYfPb9z2HeKiIiIkpUvMnSJ9dyCifOXkpEWqJ1bk5P9dTLndmYeBOuMVOckhmDi2SooiALFQVZ0d4M23hDQ0REeiROnxtWvMeKLdzdiUhLpM/VGaluvLd0EjJTI5O1TUSRweAiJSTe0BARkR7B1AJKItzdiSjW5KQnZhiCnTiUzFzR3gAiIiIiIgov3vMSEUUG+3QoGTG4SERERESUoHiTS0QUWezMoWTE4CIlJJ7QiYiIiACXd1z0nMFVUd4SIqLExjIUlMwSs9gBkRfP70REpMZrAyUTl0tgw9WTkepiTgERERGFB4OLRERERBQ2nJU7+tJTOCsrEfnjqZmInMQuTCIiIiJyHGflJiKKfTxXE5ETGFwkIiIiIiIiIoojjAtTLGFwkRIas/2JiEgPh4QRERGR09i+oGTE4CIlJHbiEBGRHvb0ExERkdPYvqBkxuAiJSR2FhERERERERERhR+Di5TQ2HlERERERERERBQ+DC4SERERUdhwNAERERFRYkuJ9gYQERERRVLHvAwAwMAuBVHeksTG0QOJ5/55Q+F28ZclSiQ8oonICQwuEhERUVKpLc3FCwtGo7ooO9qbQhRXhtYURXsTiIhinsScfUpCDC4SERFR0qkpyYn2JhAREVFCYR4oJS/WXCQiIiIiIiIiIiJbGFykhNTSvQQAkJ7KXZyIiIiIiIiIKFw4LJoS0nWz+mLBxB7ISuMuTkREREREREQULkzrooSUnuJGZWFWtDeDiIgo6UneuvacZJiIiBKZ4HWOkhjTuoiIiIjIccqbrOtn9cWAqvzobQwREWliQMw5F0zsgQMHJBw+oCLam0IUcQwuEhEREVFYHdlUGe1NICIiCquC7DRcN6tvtDeDKCo4LJqIiIiIiIiIiIhsYXCRiIiIiIiIiIiIbGFwkYiIiIiIiIiIiGxhcJGIiIiIiIiIiIhsYXCRiIiIiIiIiIiIbIlacFEIkSWEeEAIsVEIsV4IcYjBsqd4l9skhLhJCOHyPt4ihPhJCPG297/XVK+71PuaTUKIS8P9mYiIiIio3dQ+5bjn5MHR3gwiItIhhIj2JhBRAkiJ4nsvBLBTkqRaIUQdgJVCiFpJknYrFxJCdAVwOYBGAN8BeAbAsQDu8S7ygSRJTeqVCyFGATgCQIP3odeEEC9JkvRyeD4OEREREcmEELjlmIHR3gwiIiIiCrNoDoueDeA2AJAk6WMAbwKYorHcLACPSZK0XZKkgwBu977WzPrvkSTpZ0mSfoYnGGnmdURERERERERERGRCNIOLVQA+Ufx7G4BKG8t1F0KsEUK8JoQ4wcb6iYiIiIiIiIiIyIawDYsWQqyBJ8Cnpcyht1kDoFKSpB+9w6dXCCE+lyRphdUVCSHmAZgHAGVlZWhra3NoE2PP7t27DT/fgYMSAECSpIT+HsiaYPsNkR7uO8krlN+d+w3ZxX2H7OK+Q3bE+34Tz9se70Ldd9ra2lgzM0nF4nknbMFFSZIGGD0vhNgGoAuA7d6HqgC8qLGovBwUy33qfY+divfbIoR4DEAzgBVGr9PZ3r8C+CsANDU1SS0tLUabH9fa2tpg9Pn2HzgIPP8MhBCGy1FyCbbfEOnhvpOEnn0KAEL63bnfkF3cd8gu7jtkR9zuNw5cqyk0tvcdxW/H4GJyisXzTjSHRT8E4FQA8E7oMgjAsxrLPQJghhCixDtL9CkAHvS+rqPwHk1CiEIAEwG8rVj/8UKITCFEJoDj5dcRERERERERERFR6KI5W/RvAdwthNgI4ACAeZIk7QIAIcSVAL6QJOlWSZI2CyGuAvCq93XPA/iH9+/DAZwmhNgHz2f5uyRJjwOAJEltQoh/AXjfu+w9kiS9FJFPRkRERERERERElASiFlyUJGkPgCN0nrtM9e/b4J1ZWvX4TQBuMniPKwBcEcp2EhERERERERERkbZoDosmIiIiIiIiIiKiOMbgIhERETmqT+e8aG8CERERERFFSDRrLhIREVEC+ue8odi+a2+0N4OIiIiIiCKAwUUiIiJyVE56CnLS2cQgIiIiChchRLQ3gciHw6KJiIiIiIiIiIjIFgYXiYiIiIiIiIiIyBYGF4mIiIiIiIiIiMgWBheJiIiIiIiIiIjIFgYXiYiIiIiIiIiIyBYGF4mIiIiIiIiIiMgWBheJiIiIiIiIiIjIFgYXiYiIiIiIiJJIXWlOtDeBiBJISrQ3gIiIiIiIiIgi54FTh+Hjr3dFezOIKEEwc5GIiIiIiIgoiRRmp2FITVG0N4OIEgSDixRACAEAaOjUIcpbQkREREREREREsYzDoimA2yXwyGnDUFuSG+1NISIiIiIiIiKiGMbgImka2KUw2ptAREREREREREQxjsOiiYiIiIiIiIiIyBYGF4mIiIiIiIiIiMgWBheJiIiIiIiIiIjIFgYXiYiIiIiIiIiIyBYGF4mIiIiIiIiIiMgWBheJiIiIiIiIiIjIFgYXiYiIiIiIiIiIyBYGF4mIiIiIiIiIiMgWBheJiIiIiIiIiIjIFgYXiYiIiIiIiIiIyBYGF4mIiIiIiIiIiMgWBheJiIiIiIiIiIjIFgYXiYiIiIiIiIiIyBYGF4mIiIiIiIiIiMgWBheJiIiIiIiIiIjIFgYXiYiIiIiIiIiIyBYGF4mIiIiIiIiIiMgWBheJiIiIiIiIiIjIFgYXiYiIiIiIT4XDsQAAIABJREFUiIjiwNj60mhvAlGAqAUXhRBZQogHhBAbhRDrhRCHGCx7ine5TUKIm4QQLu/jZwsh3lb8t1MI8Qfvcy1CiJ8Uz70Wqc9GREREREREROS0W44ZgFcuGhPtzSDykxLF914IYKckSbVCiDoAK4UQtZIk7VYuJIToCuByAI0AvgPwDIBjAdwjSdKNAG70LpcK4HMA9yle/oEkSU3h/yhEREREREREROGVkepGRUFWtDeDyE80h0XPBnAbAEiS9DGANwFM0VhuFoDHJEnaLknSQQC3e1+rNg3Al5IkvRmm7SUiIiIiIiIiIiKFaAYXqwB8ovj3NgCVISx3MoC7VI91F0KsEUK8JoQ4IZSNJSIiIiIiIiIiIn9CkqTwrFiINfAEBrWUAfgBQI0kSdu9y98CYKMkSX9QrefPALZJkvRb778HA7hDkqS+imU6AtgEoEqSpG+9j3WA5/P96B1avQLAqZIkrdDZ3nkA5gFAWVnZwPvvv9/mJ499u3fvRk5OTrQ3g+IM9xuyi/sO2cH9huzivkN2cd8hO7jfkF3cd8iuaO07Y8aMeUuv9GDYai5KkjTA6HkhxDb8f3v3GW1JVXV9/D87kaNEQdBHARUkCJJRchQkJ0FAUQRBEBQxRwxIUESiRPFFkoAioEgwEFWC5Cg5i+TY9Hw/7H2a6sttaAroOt13/sbo0ffUqb5unmeNXbt2WAvmBR6ul+YBLhjk1t59NO67e8A92wBn9SYW6//+E42f/yPpdGA5yiTjYO09HDgcYIkllvCKK674as2fpF144YVMzv998dZI3ERbiZ1oI3ETbSV2oq3ETrSRuIm2EjvRVj/GTpfHok8GdgCoBV0+BJwzyH2nAutLmrVWif40cNKAe7YDjmpekDSnJNWfZwZWB656U/8LIiIiIiIiIiIihrAuq0X/BDhG0q3AS8BnbD8JIOm7wH22D7V9u6TvAZfWf/cn4PjeL5G0HDAt8McBv38jYEdJL1L+O4+1fcZb+l8UERERERERERExhHQ2uWj7aWCT8Xz3zQGfD6NWlh7k3ouAuQa5fhBw0BtvaURERERERERERAymy2PRERERERERERERMQnL5GJERERERERERES0ksnFiIiIiIiIiIiIaCWTixEREREREREREdFKJhcjIiIiIiIiIiKiFdnuug19R9LDwJ1dt+MtNAvwSNeNiElO4ibaSuxEG4mbaCuxE20ldqKNxE20ldiJtrqKnXltzzrYF5lcHIIk/dP2El23IyYtiZtoK7ETbSRuoq3ETrSV2Ik2EjfRVmIn2urH2Mmx6IiIiIiIiIiIiGglk4sRERERERERERHRSiYXh6bDu25ATJISN9FWYifaSNxEW4mdaCuxE20kbqKtxE601Xexk5yLERERERERERER0Up2LkZEREREREREREQrmVyMiIiIiIiIiIiIVjK5GJ2RNIekWbpuR0x6JE0hacqu2xGTnsROtCFpBknTd92OmPRkrBNt5XkVbSRu4s0gSV23ISY9ybkYnZC0EbA7MBI4H7jC9kndtiomBZLWB7YBZgMOBS61fUu3rYpJQWIn2pC0IbAjMDNwPPAv23/ttlUxKchYJ9rK8yraSNxEW5LWBD4KvAicYvuijpsUk6BMLsZEJ2kB4AxgQ2BKYAVgZeAs24d12bbob5KWBo4FPgnMA2wM3Ar81vZlXbYt+ltiJ9qQtChwIrAZMDuwOjAj8DvbZ3TZtuhvGetEW3leRRuJm2hL0mrAQcB+lInpLwB7AL+x/VyXbYtJy4iuGxBD0kzAQ7avB5B0F/AwsKmkh2yf1mnrop/9H2UV9iLgIkk3AtsDG0h60PYdnbYu+lliJ9qYE7jF9lUAkm4BNgXWrc+rSzptXfSzjHWirTyvoo3ETbS1FHC47cMBJF0DfBsYAxwnaZjtMR22LyYRybkYXbgReFjSupKG234EOBe4GPhgt02LPvcQMKWkGQBsXwn8EliQ8mCMGJ/ETrRxB/C8pIUlyfbtwMnA8+R5Fa8uY51oK8+raCNxE69LI6/iMEqcAFBPZnwH+KmkpTKxGBMqk4sxUUiau/4ZZvsx4BpgfWDxOuh+GPgrsGLvoRgBIGlqSdPWj5cCcwE/7n1fB0+nAttIGtlBE6NPJXaiDUkzNwpw3AY8DewCzFqfYbcBfwQ2kTR1V+2M/pOxTrSV51W0kbiJN8Iv58c7GlhV0pca351OOSq9Rhdti0lTJhfjLVcTmv8WOB04UNImwPeAF4DPUPIRAcxRryURaAAgaQPgNOACSbsD01CSDa8q6VBJs9VbxwCP1b8jEjvRSn1enQGcJ+m7wKLAZ4F5ge8Di9dbpwCeIHETVcY60VaeV9FG4ibakrSWpGMkHSxpI9t3A9sCG0rac8Dt077yN0QMLgVd4i0l6d2UF7WNKcfIVgPWAf4AHAnsCSwJzEpJkr91XWWLIU7Sh4BfUareTU/JG/MwZRXtAcpL3KOURZIFgI/38qLF0JbYiTYkfYCXi7dMAWwEzE2pEP034EDgbZSB9tuBrfK8CshYJ9rL8yraSNxEW5JWAQ4B9gampiycfhc4Dlik/n0h5Zj9WsCmtq/rpLExyUlBl3irTQs8Bdxse4ykEymd1baUROc/lDQN8B7gAdsPdtfU6DNzAFf2KtxJugPYCfgc8C1gPWAhSgXXf9v+T0ftjP6T2Ik2ZgTusX0NgKT7KMVbPk55bu1EmVScF7jN9j1dNTT6TsY60VaeV9FG4ibaWgg4xvaxMLZ4y88op6QPrFXHN6bME21s+4bumhqTmhyLjrfaLcCtwMaSRtl+nLIacj6wIoDtp21fncF2DPBfYApJswLYvgU4mLKTaCvbT9m+1PYZGTTFAImdaOMe4DFJH6o58+4DTqLE00dtv2D7Dtt/ycRiDJCxTrSV51W0kbiJtoYDS/Q+2P47sCvwbUnr2b7P9oG298/EYrxemVyMt5TtZ4AbgHWBpSVNUZOcnwMs1cgHEjHQZZTdIL/oXaiDp18C60masquGRd9L7EQbdwCPA18A5qoFOHoTjGtJmr7LxkX/ylgn3oA8r6KNxE20dTjwXkk/6l2oE4zfA5burFUxWcjkYrwlJDVj60eURMKfAjar1creB7xESWoeAYCK4QC2X6Ikpn6/pBMlzV1vm5Fy/OyljpoZfSqxE2004saUY2XTUZ5by9Zb5qRUjB7dSQOjb2WsE23leRVtJG6ird7zyvZTlFydy0r6ceOWaSjxE9FaCrrEW0bS7JSt+fvVh+HulESx7wamBD5l+4ou2xj9QdJ7KXmpHq2f5wJWsP2b+oJ2GuXlbCTwTkoy/CSmjldI7EQbNW42sH2QpBGU/EOz1j8zA9umAEcMJmOdaCvPq5gQNU3HmMbnxE20Uo/R/5/tyyR9kFKw7hbgTmBVYJMUb4k3IgVd4k0h6cOUrdTPAmfW/B7TUiqW9VbYflIfgv8HPGr74a7aG/1D0urAocCOwB8lCZiPUpUV2y9KWg9YuF671fadXbU3+kdNOv1eSj9zme3/kdiJ1yBpGUpC83uAa2r+xJmAMQC2R0vamZIwfy7gvno8OoY4SbPZfmjA5anJWCdeg6QlKM+nm4E7bT9CeX7NAnlexeBqTHxY0pf88o6g+UjcxGuoGzdmBW6y/VDdubgeZYf9ZbavkLQ4pWidgINt39hdi2NykJ2L8YZJWgf4IXAWpRObEdihDpya901h+/kOmhh9StIawA8oKRput71Rx02KSYSkNYF9gb8B7wd+avu0blsV/U7SusA+lEIbUwGLU3adXdNpw6LvSdoAOBXY0fZh9drwOqHYvC9jnRhHjZ19gcspGzumAr6UYgnxaiStRkm38DXb53Tdnph0SFqLMta5A3gXsJrt+yXNUAuORbwlknMx3hBJi1EGTNvZ3gv4OSVnw1QD7lsX2LmXKyRC0qrAAcAnbS8GzCZp7UHu21LSphO9gdG3JK1IObb6Gds7AlcBc0qaTdKMjfsSOzGWpLdRdkhvbvtzlGfX9JQd0ws17ttC0uc6amb0IUnvB74F/AT4qaTPwNidis37MtaJcUiahpLfbAvbWwBfBa4Ejq87i3r35XkVY0n6CPA7yhHncyTNKmkxSQtknBOvpr5f7UvZ6LMucDewQP36ycZ9q0laroMmxmQsx6LjjXoSOMD2vwBsXyVpCkq+obsb990I3DBwIB5DTz32PC2lWML2tq+WNANwO7AEZQds0wP1T0QvIfVwSuxcLOntwGbAv4DlgeGSdrd9L4mdGNezlMWveYGrbV8v6RTK8eefSdqwrugnV1UMdC9ld/Qxks4DzpIk24fVv3vHgDLWiYFGU46rfgC43PYtkvYFDPxA0vY133SeV9H0IiWP4sKSbqHsmn6Cksf1JknfqifEEjcxlqSpKUfkd6xj5LmApYDPS9oVOEfScZRj0O8Ezu2ssTFZyrHoeMMkTW/7CUkja+6P31MG4efVY6832L6r63ZGf+ltze8dK6u70U4F1rZ9WcfNi0lAnWj8IoDtfSQtQtlddJjtP3bauOgLvYmfWqRlDPAlYEHKZPRcwHuAr1N2UW9v++7x/rIYknrFFCSNsv1CvbYqcA6ws+1DJX0UuNb2HV22NfqTpK2AZYDDbV9dry0IfJ9y5PX6LtsX/UPS9MCT9bm1LPBbyqLY7raPkLQypWjUd21f3mVboz9Jmsn2/yRNBfwYeND23pJ2AjYFtrR9X++9vdvWxuQmx6LjdZO0lqQjJP1c0uq2n6hf9WaqnwHul/QxSj69HA8KoOTJk/RbgDqxOKJOLMr2hcAJwJL13sRNjCVpOUm7SNpA0pwAtXri/rb3qZ+v5uXdaREAs0Ep0lLj5QxKns5F6/cb1xd7UwsBRUCZQJS0P2V32ZKNicURtv8MrAkcUHe//pSMdaKqY51TG5dupFTy3UjSogC1IutwYP4Omhh9qKZVOIQ6hrF9MbAx8E3bR9Rr51N2Nc7UVTuj/0haRdIeAHViUZSdr/va3rteP5gyRp69fs7EYrzpMrkYr0tdrf8pcAnlGOsJknaSNLXt0fW2/wH7U3aDfKJWjo4hrD7kALYF1pf0BxhblXVU40jZ9cAe9VqOlQUwtnjLYZSk1F8AVu591+h3qHmH3k/ZlRZDXF3guqeZP9H2jbaPsL2d7T3rjrSPA+8A7u+ssdFX6qmLfYDbKEXqPtH4ekxjgvGXwErA+rZvm/gtjX4yYKyzgaSzAGz/EziZ8lK/s6Sda7+zACX/YgxxklYHvgkcZ/up3nXbfwcObNy3KaUSfXa7BipGUialt5b0LYDee1Xz5KCkzSmFV+/toq0xNCTnYrxeiwJH2z4KQNK/KQnOxwCH1nveRcmdt5TtWztpZfSrM4A/AStL+ovtj/DyjlcoMbQiJQdajtIHkpanFIrazvbfJe0NzCVpHuDFWv1uemBDyhHpzWzf2WGTow+oFGjZn5LUfM96Ovrg+t2I3qS0pO0pBRbWs/1gZw2OviHpQ5QXtU/UPmcryjNrNeC/tq+Q5LrYujKwou1ru2xz9J3mWOdvtlewfa6kmyj5z7YBHifPqwAkLUNJs7Ck7X/W0xmLUXae3WH7VkmjgM2BLwObJoVH9NSUZBcCdwLvlPQ929+gvl+pFJXaBNgV2Mr2Q501NiZ7ybkYr4ukLwOL2d68cW0VyorsZnXwtBIlv0NW1WIcddVsHeDzwK+AeYCpKZPWo20/J2kK28932MzoI5KWAobXxNRzA1cAlwJPUWLnS8DDwBbABbZv7Kyx0TckTQdsYPs4lSr0R1FyVB084L5ZgRmyEBY9NU/VorYvUSkY9RfKbug7gM9S4uoClUJkMySndAw0yFhnXmAq4AO2n605YIf1jtrH0FaLcPwHOI6STupM4B5KzEwD7EzpfzYDLs04JwaqxVqmpKRg2Iiy6WdmysL7e4FdgJ/l3TzeajkWHa/X4cDikr7du2D7PMrxoQ/Xzxek84rxuAR43vb/gIMoA+5nbT9l+zmATCzGAP+oE4sCPgrsY3s94DuUVf0FbT8GHJEBd8DYAhxPAscD2D4L2A74Zu+IdM1PNJ/thzOxGD01dp61fUm9tDglZ9Xmtvei7GhcBUre4EwsxngMHOvMAzxn+1kYmwM2E4vR20n/DPBuYEvgv8BRtjejTCreDyxcY+f4jHOiqZGK4UFgJttnADdRcnU+Xfuaa4Hd8m4eE0MmF2OC1Kqs1IHStsBqkr7buGUKYIYOmhZ9rvHgg7ISS33BPwjYCXhE0u+6aFv0v1qEo5c/5nDb+9bPN1F2L05XP48e7y+JoaaXa2gMjK0YfTZlgnFPSadRcgfn5T4GGuc4j+3f2z6scWkMkAWweIUJGOs8nLFODOIlgJpncQFgF9tH1mt3UfqbjHNiUI2c9RcAL0laDtiekqdzhKSv1e+f66J9MfQk52K8pvpiNqYeH3uH7Ysk7Q4cL+l9lMSwq1PyOUSMVXeBjKk58Wam5FEcBuwFfM727ySdSSamY4Da77geFxpl+7HeZFH9fjPgg8D3Omtk9J1ePsXa50xn+95GYvOza3+zCbBKcp3FQLXPmQmYrS5gjFX7nLUoKRgixmo8rzLWiddUi2A+I2m47ZdqnzO37WuAXzTu25SSw37vrtoa/aPXz9Sfe7EzBWVRbBRl5+uOwMfreGddasGoxiRkxFsqOxdjUJL+T9JidSW2txq7MaVYC7YvoxwXOg+4GdjQ9nWdNDb6So2bhaHsHKqTQ3sBc9bJocMpyah/Vx+UOVoWAEhauBE7rkmo9wXmbtwznaRtKFUVP+5Uox/yJM0iaWYYW4F+JPAVylHE5n1rAOtTJhavmfgtjX5Tc981P48EPk0jdiRNJekTlFQMWw+cdIyhSdJqknaCsc+rKSnFNjLWifGqEz6nADQmhz5NKWTYu2eEpC2B7wJb2r69k8ZGv5my90NjUvpoSp9zN/AFSsGWs+ttf7B9TwftjCEsBV3iFSRtQMmh+BDwGKXi3TG2H++0YdH3avL7uymD6uN6easkLTDILhBBVtOikLQWJZH5lrZvaFxf1PZVjc8zABsAF9u+eeK3NPpJfVH7PiXf0JPAN2xfL+kdHlBNs778z5kJ6QCQtDKwIHCSG5XCJc1p+/4B9y4O/C8v+QHj9Dvfsn164/p8tm8ZcG/GOgGApNWB/SgTid+wfWi9Plif8xHgvoHxFEOTpNUo6RXuB+6y/aN6fZlGjuDevelzojOZXIxxSJqHsqL2KdvXSPoUsCQlwfD+th+p960GPG778u5aG/2mTvxcBFxOiZkzbP99wFb+RYD/ZjUteuqOsn0ouYb+2jva2vi+d/xjPkouq+cyaIral5wIfAq4GjiYUl3zMOCvvYIJ9b7Hbd/RUVOjz0haETgf+Cslhk61/dAg9y1LeV5lt2IAZSKIEjN72r60ns6YdmD8ZKwTTZLWpBxv3gN4B7CE7V0HuW9Z4JEsnkaPpLUpY+S9KTUONqFUfv7TgPerOYGHbL/UXWtjqMux6BjoGUqRhF6OqiOBMyk5ZDYGqDll3kPZ2RgxVt3deiZwGjALsLqk7YCdJA2X9C7g46SQQlQ1l+t+wHl1YnEOYH9JB0o6CMYe/3gfsCcwZSYWoxoD/NP2RS4V5z8B3EY5YjYvQKPPeaa7ZkY/qTtYF6HkpzoEWBHYRNJsA+6bBViJMiaK6HkWeAK4svYvZwDHSvqTShV6SZqXjHWCsotM0tyU/NBfsH0hJZ3UpyStOuDeWSn90dMTu53RnyTNBXwd2MP2CZQNQE8Dc8HYlAyStCiliMuU4/1lERNBdi7GOOrOs58Af6McFXq+Xt8JWNf2WvXzSNsvdtfS6DeN3WXHAr+h5OM8nfJy9iXbB9X7prP9ZIdNjT4jaXtgQ0q1uy2A3wHXAd8CrqiTRkiayaVifQSSPggcBWxu+8bG9ROAEbY3qZ/T58Q46imNR20/VRfA1qDsuj/J9oOSprH9tKSpbD/bbWujX0gaBrwNOIayi2g94A5KP/QDYGFg7V5RKdtPdNTU6DOS5rJ9r14uOvZNYFbKounY0xjpc2KgelrwMuDp+p71HcpC+5cH3DfbYDvwIyam7FwMJM0vadn6QHscOAvYHVi1rvBj+2BgWknvrZ8zsRi95Pc9vZWKMyhVy94OfIBy7OydkpYCyEt+QDm+UY9wYPuXwK+BHYCjbH/b9snAdsBw1aILmVgMSbP3dpjZvoKSE/j8ujOkZ2dgSknT1fvS5wSS5m2MYe6y/VT9+WhKHC0HrCzp68DP6iRAXvKjN6mI7TG2HwYuBc6h5M471vYLtr8IDOflwoeZWBziap/zPgDb99bLY+rf1wDLAFPX3WfD633pcwJJM9eUC9g+1/YTjePOLwJT1/u2krRDvf5wB02NGMeI174lJmeS1gcOoBwlGynpTMq26ukoVcrmlHQl5YjZLMAjXbU1+ktdSVtC0q/ri1pvwPQccCTwEmVy6CrKDrQ7Omlo9J2aDH8f4D8qFaG/YPvXkv4BNAsmLEXJoTccGP3K3xRDSc1Z9V3gEUkz217a9p51EewfNXfntcDawOy8/BIXQ5ykj1HSL9xVF8V+TknF8F8A20dJegj4MTADsF4z72sMXZJWAuaQ9IfehKHt70maCtgLOAK4SNKGwIyUfNMxxE1An3OapG2AfYHtkicveupY5mvA/ZJesr1lvd47NfgscHM9Vr8r8AlIAZfoD5lcHMJUKvt+CdjA9lWSNgc+Qjna8VVKpeg1gM0oO9G26BV0iaFN0ocpq/YXA89JOtX2XfXrs4GTgbNtn1vv36N3xD6GNkkLUiYWt7d9kaQDge9IOgI4p/dCL2lbYBtgm8RO1MH2vsBuwCXA2ZL2s72H7c9Leozykn8PMB+wle3krQokvRv4NuX4/D8lfQn4KGXC6Ne9l31gGmBOYHnb13fT2ugnKhXF/0zZZTZa0jm9ndC2vyrpBWBvSTcBiwHb2n60uxZHP3gdfc7JwCq9NAwdNTf6iKS1KGPkvYB7gX0k7WL755RNG1Dez/cGbqVMTN/QSWMjBpHJxaHtCUqi8t62699IegTYCPis7Z9JOo+ya2hkBkwBIGkU8F5gc0pS4W0oR1dPqjsYLemrNZfVMMpiWiaHouc54ArKsTLqxNBXKRV/7wD+LWkZ4IvAZnnJj3q8eUvg67b/XK8dRjlSBoDtb0o6npKe4dlUaI2Gx4E7qUfGbP+k7hhaBbgLOF3SjJTdrh9OnxNQct8BK1DGxNMDOwHDJJ3VmGD8tqT5KYVbXmwcfY2h7TX7nHrf2cAFmVgMKHnFKTsRv2H7D/XaRZTcnDROiI2i5H5dwfYtXbQ1YnwyuThESRJlBeQmYBFJV9p+1vafVSpTfUbSIbZTYTPGYfsFSacBz9t+og7AN6EMuk+yfQdlQrr5IIzoGQYsRJkY+juA7R9IOoSSomEV25dI+nAWNAJKzkRJPwaekqR69OdhYMHePTU/3s2dNTL6Uh3rDAcELE554cf2sTVP5xeB020/JungHIWOHtvPSjoKeLLGx7SUvMDDekekJQ1LvxNNE9rn1GsZ48RYtv8naS9K6pdh9R3qNuBjA249FPhdFjOiH6WgyxDl4llKEvMdgPUbiWOPpewuem+HTYw+ZvvhRu6hU4FTgQ9Sjnd8Bzis7nCMGEddZT0S+L2kRRpf7Qo8XXcQZdAd47B9fW9ndL00hpcTmu8AHFBf6iLGqmOdB4FTgCPqMdfed3tT0nrMUz9nYjHGYftu24/Vn38B/Bb4NPAhSV8Gftll+6L/TGCf847OGhh9zfZVtu9pbM4YQ6mDgKQdJP2gxlgmFqMvZefiEGf7zHrk7EvAbJKuAWamFHB5oNPGRV9o7BQa5+fmZ9snS3oAOIzyEPyY7Rc6anL0iVeJlwMlTQ/8VaWo1GXAhsA8pAhHvIoBOxevkLQ28Eng081Yi2j2P7Z/VfucEyXtQknNsBgwNyU9TMRYgzy7hrlUij5I0sPAcZQiYxt21sjoO6+jz8kx6HhFPzPgu97Oxcco6YJWo6QP2n5itjHi9crk4hDS6Kh6n3sv+ifUZPgrA6sDUwAb236oq7ZGXxlJySdEzafYHDw1P89CSYa/XHJWRTUd8ETjxWzsIMr29yX9D/gRJQfR/MDWvR2xMXTVXR0PeZBcrY0Yeomy6/6DwCdtXzcRmxh9StIKwLy2jx+4KGb7F7XP2QDYApiNktc1u6QDSVPVEz2vqLpqe0xjDP0CMCUl31nGOkOcpBmAF1xSS7m3gz59Trxezff0xvv6M8AewHKU4i0Z60RfUxb6J3+SPghcZ/v58U0w1p9HUnKETGX78Y6aG32ksVJ2J3Ct7V+N576pKQOnS/PgCwBJqwAnABva/vvAvqdx3zyUHSC2ff/Ebmf0F0nrAN8B1rN936vctwiliuvytm+aWO2L/qVSUfz/AX+yvUWzzxkw1pmJMjk9yvYj3bU4+oWk1SmpOf4DXGL71+O5b0bg88BvbV87EZsYfaj2OV8F7gOutv2jxnfpc2K86nH59Sm1D263fXa9PvA9fSVKCrOFMtaJSUFyLk7mJH0U+CdwpaRRvdXX3vcDVmdH234hE4sBIGlN4MfAWZTk1EsP+L4ZR88Ax2RiMRreTxlM/0bSas2+RxWASx69+zKxGLXP+QrwtcEmFgf0OVcD82WwHTA2dr5T/ywm6SPNF7QBY53HbD+Rl/yAsRNEBwAnU46rvm/A981+5zHgR5lYjDohvT/wC8rkz7wDbmm+Y6fPibFq7PyaUqzlvcAukvaGl3dJ9+61fQEwR8Y6ManIzsXJmKRZgeOBg4BtgAWAxV2q/Q63/VKnDYy+JWl+4PfATrbPk7QWsBtwDDDG9oldti/6V2+1XtJywPKUwdPBwErAPVm8iIHqRPO7gFuBbW0fV3e0bkDZ1fpIs89pxNh48xXF0CFpCeBXwGdt/0XS94G3AzvXha+IV6j9zlS6Sj5EAAAZaklEQVTAUcCJtk+rC/K7UorUyfYhvXvT10RPPel1APB327+R9BHKROOpwDS2v1bvS9zEOOrE4beAO2wfXfNyLgjsC1xg++udNjDiDcrOxclQY0fQw5RV/D/a3piybf9fNa9MJhbjFXqxA9wObFAnFt8BHF2vvYuywrZXV22M/tYYSN8NrAacCXwZ+DvwiKR3SBreVfuivzTy4d1OKQj1zZo37zhKf/MB4FOSxiYx78VYXtqGtsbzaiQlj9lf6udLKbvPpq33Zawbr1D7nWeAe4FPSloeOAK4HpgB2FrSV3r3dtfS6BeN96sXgSeA7zR2oZ0H3AxsJOmgel/iJoBxYmcMZdF0M0lTu+QZv4xSWHXRegw6YpKVAdfkadreD7Yvbvy8BmWC8XIASRtL+tTEb170sZEAtkc3EpUL+LrtHYEfUgZRM3XUvuhjkkbUv4cDj1CSnD8HXEVJgv84MHMWN6Jh2sage0fgXOAvwKm2dwN2By4G5uiuidGnRgHYvsT2v+tuImyfCfyP8rxqJsaPGKsXL8DpwD2UHYsn2N6VsovoMGDWjpoX/WmKxs8/phyHXh843faetk+h5CmfLosaMcDIxs/HUxY1NpI0RX1GXQ88zCuP10dMUlItejJTj69+WdL1lOOrO9dj0KNqPsU1JJ0s6WngIWDdblsc/ULSqpTV+3uBm2z/EkpOPKD3s+sW/t7AyVmZjQFFo3opF56RdJmkHwObANsDswOnSFoMeDqxM7T1nlfA9XUH4462d5R0su3zoeRzlTQGmKkxCZm4GeIGPK9utn2E7Rcb/c/+wA6SZrf9YLetjX4i6f3ALXX3Gbb/BvxN0pbAMr3xsqTZgTkljezdG0NXzc25o6QbgYds7085ybMhsHndhfYM8EFgFspk0vPdtTj6hUpxzK0k3QFcAvyRUshlJWCYpFNsPybpfmD++m9ypD4mSVlVmYxIWgY4FPgZpUrrByX9rR6DfkHSVPXWUylJq9dJUuoAkLQ2cCBwNvAksJKkhQa57xPAZsCBtsfkwReS1qUUjfpHL+WCpN7q/guUicVdbP++TlgvZfupxM7QNsjzahFJl9RV/PN7R+clbQ1sCBxRjzEmboa4QZ5XK/aeV41d0VcAS1GeVxEASFqPcnrnlN5zqrfjHrgFmAb4uqSdga2B72ZiMSR9mJIe6CRKHul1JP2hfn0e8CxwhKTdgU8DX7KdicXoFRvblxInCwCr13HMTymnelYCzpb0dUp9hGMhi6gx6UpBl8lIHXAvb/urjWvnAlPaXqF+XoDyQrebS7XNGOJq4YSTgW/a/qOkWSjHhA6y/Zt6z/TAqsD3gE2dqtABSJqNEitHAMtQVuxXsP1s/X4K4D22r8sqbDRN4PNqLcoOtI3T5wRM8PNqmEvFzdWBu23f0GGTo09Imgs4gxI/76LsLvt4bxJI0ihgB8oEwMzA9xvpYWIIk7QtMKftH9ZTO9NQjkQ/aHt9SSsDq1PSUh2cuAkYWxzzNGBX23+u457tKZPUd9u+SKX46seB4cBZeV7FpC6Ti5MRSetQcsQsW4+y9q5fTKlotmf9PINTsTUqSdMCS1PynNn2aEn7Uyr77t+4713Ai7bv6aip0YfqkejbKSv3xwHvBj5i++kB92VyMcZ6jefV32x/uU4kjbZ9X1ftjP4yoc+riIHqbuglKbnN3g58g5K3c+vegljj3imy8yx66pH5b1HGNg/Ua9NTJo5O6KURkjTC9ujuWhr9RNLMwOy2b5A0N3AlcArluPw7KQVXD+mwiRFvuhyLnozY/gNwIrBfXaHt+T6N/19nYjGabD8FXGL7xcagaDQwHYCkTSRtavs/mViMgWxfYfux+iL2CcqRob9AyVEkaat6XyYWY6zXeF6NqPfclYnFaJqA59XGkrborIHRt+qR+cttP153B32XkrrjVwCSVqq7panXI3pOoexU3K3ulgZ4CjiTMlENlGKIHbQt+pTtR4Eb68fpgD1cCtd9lXIkevau2hbxVsnk4uTnaEpF6APrTjOA9wDva+RBixhHb5eZXq6eKOCueqzsy5TVtohXVScYtwGulPQ4ZWfaP7ttVfSx8T2v3itpil4Bl4im13he7UX6nBiPRk5OKDkWvwv8V9KtwDHAzfW+LIbFWLZfAM4BZgT2lDSXS4XfkcD8kkbkeRWD6fUltm+wfVz9+RngOWB6ScMSOzE5SbXoyUTvyKHtayUdQcnfcJGkM4A1gY/miEdAyStUB0rNa70jq2PqpQeBPSkVxbe1fctEbmb0qYHHmwd+tv2cpKuAtSlFo24c7PfE0DGwz8nzKt6IPK+irebzqk403ijpIcrx6LVt39ZpA6Pv9HK42v6DpDHAWpTn1anApsAa2bEYPZKGNxcxmn1O7+dGccwt6iR1xGQjORcnE40OayHgbbb/Imkl4CVKLqLbO25i9IEaEx+iJL9/pl7rxc4iwIw1dvak5JdZNC9qASBpduBRYArbT9Wk5tTCCR+grMLeRskj8ztKovwUjRriXqPPyfMqXpc8r2JCSFocuLGZ+7cROwsCd9h+WtJ8wCXAKnleRc94JoQWAMbYvkXS+pQd09emzwkASQvZvrb+PNz2SwP7nHrr2sA3gc2dQnUxGcqx6EmMpLUlbTfweu28lgbOaly7wPZf86IWMLY66y+BK3sv+TA2dpahTAj1nAoskEFTwNiqvScCJwA/kjRPXckfU/udM4FZ67XbgeXyohYT0OfkeRWDmoCxTp5XMShJa1AqiL+rcW1YjZ1lgbOBDwDUmJk/z6uQtJqkj8HYfkaNn5cG/gzMVq+dbvu09DkBIGlR4N91Nyt1YnFEY6xzNrBgXez4D2WXdCYWY7KUY9GTEEmrAb8FRkma0fYB9fowYGpgE+AzdSU/lVkDKKuuwMzAHsBOts9VqXI3EpjS9r2Uox2ftf0XgBwNih5JawL7AdtT8g2tT6m4eZektwGfp8TOxY3jIE901uDo3Ovoc3bI8yoGmsCxzg55XsVAktal5InesqZdGGn7xfKVpgJ+AOxo+9LG8+p/XbY5ulcXUE8Dhkna3vZxdWJoOKUQx+cp71cX5XkVg3iMsuA1t6Q/2l7D9miV4j8/pIyRLwewnZzAMVnLsehJRB1Uf5+SqPwG4GpgL9v7N7Zdv832f/Pgi4EkjQCOArYF5gBOomzRX4uS8+NP9b6xK7WdNDT6iqTpKEVZzrB9Yr22LzCd7R3q59ltP5h+J5peT5+TuIme1zvWgTyvoqjPq2uAq2yvL+ntlAWOqSkpO44ERtt+ssNmRp9RKQz1c0qfcxvllMYXbR/X6HMyzolB1WfW7MCPKVWgj6GkENqXstD+uO37672Jn5jsZXJxElJXXd9m+556tONC4Cu296vfz2Q7K7DxCiqVwi+kbM2fCbjZ9iGStgEOARbKccQYqL68L0SZFHqmHvXYkpLAfJt6z7S2n+qwmdGH0udEW5KmpIx17s1YJ14PSUsBxwHnUyrP/4lySmtuygv/94EXnSIK0VAnpqeuE4gbUZ5Re9o+pn4/g+3Hu2xj9J8BuTkPBg60faNKUcOFKcefz6lHpFP0J4aEHIvuc5LmB54ERtq+S9J99SjHxSoJ8C+Q9F/gKWALSVtQBk6ZNR7iGrEzyvadkr4C7AWMsP0FANvHSlqFkpg6Yhx1xf7Geqys52lgGgBJWwMfkPTVDJwifU68UXV88xxwb31xy1gnXpWkeSgTh6NsX6ZSifUPwE9t/6Tesx2wslOFPgaofc6TlGcXtk+tC6sHq1QSB1hd0hczzgkYp88ZyctpFaYG3lkXx2YErgJ2A85J3MRQksnFPibpo8A+wJXA+yX90PZJKkbV3B/vBW4F7qeskLzQZZujPwwSOz+wfbJKtbt9JX3F9g/rLrSFgWde7ffF0FFj5329lzJg4KBoGHC/SrXE3SlHXDNwGuLS50RbzT7HjQqbdXFjRMY6MT6S1qNUXr0TsKQjbZ8t6YO272rcamBEPQH0XCalhzaVoj9z2z6y9jnDertZa/9zSp1YvBB4CFgt45yAQfuco2yfRTlO/0lgKWAX27+TdKZKAcS7XuVXRkxWMrnYp+qqyAGUfFVXAmsCx9et+UdI6j3k3gfcC6xu+4ZOGht9ZZDYWQv4laSpbR8m6W5gH0nvAZagJD6/v7MGR9+oO8r+HzBtzWu2V33Bb+aJeZ4SWx8AtrJ9Y0fNjT6RPifaeq0+p/FCn7FOjKP2JwcCm1F2nS0L/D+VghynNvLlfRL4HLCN7Wc7bHL0AZUidacAI1UKRu1ne0xvgrEx1nk3cDewlu3rO2tw9I3x9Dm/rule/gHMRSnecnb9J+tnUjqGmkwu9i8DN9i+qH7+bd0pdIakR+vAaRSwImUVP4Pt6BkYO6dKego4XdITtk+TdD6lAp5tP9JZS6PfLEmpCn0+cHt9OfvygAnGBykr+TtlwB1V+pxo61X7HMpO6WHASmSsE+OaBrjO9mX18/WSngYOl/Sk7T+p5OzcijKxeG1nLY2+oFJ8Y2VKoZ9LgbNqn7Nvb4KRkrJjOLAa8LGMc6JhfH3O0ZRc5MvBOMUxM7EYQ04KuvQxSVdSqrR+u3FtM+DzwEa2HxiwoygCGG/sbE6JnY1t39dV26J/1QHRnLbvkzQf8C/gUNt71u97FRNntv1op42NvpI+J9qYgD5njox1YjB1gf1y4GDbhzeu70jZUfRpYFpgTJ5X0aNSbGwG2w+p5HM9nnHzc85i+5HmUekIeNU+53PA0pQd0k8lbmIoG9Z1A+KVJA2vP+4GfEjSTo2v/0zZpv8clC0gE7l50cdeI3bOBe4iuc5iEI1jiPfVQfUtwIeAz0r6iqRNgWMkTZkXtehJnxNtTWCfc7SkqTtuavQZlSIcLwA/BZaTtGHj68uBqSiTio/keRVNtp+3/VD9+QJga2A3SdtJWhf4gUpRjrxfxViv0edcSulznsvEYgx1ORbdh2y/VH+8Avg1sJWkuW1/lZLLan5gVFfti/6V2Im2mgsV9XjQCNs3SZqTklvmAUruoec6a2T0nfQ50dbr6HMyOR3jaPQ7fwbmANaX9HbbB1H6nDkouxYzsRjjVRc4zpe0GnAt8DCwasY5MVD6nIgJk2PRfaaujIytXCZpFkqndSBwT/15M9vXdNrQ6DuJnWhrkNiZwvbz9bt1gEOANZN7KJrS50Rb6XOirUbM9GLoHcAKwF683O9saPvfnTY0+kojbsb5u363LfBtSl7X9DkxqPQ5Ea8tk4sdk7QysDAwpe0f1Wu9B9/ywLK296nX30ZZ7M+qSCR2orUJiJ0VgANtPy3pG8BpSYYfkpakFGV5yfaF9doI26PT58SrmYDYSZ8Tg5K0KDAl8HRvoULSKNsvSFoBeJft4ySNAOYFnrD9cIdNjj7wGnGzPPB+4KjaBx0MHG77qg6bHH2m94yqP/eqz6fPiXgVybnYobo6/1PK8fQtermq6gv+EsCRwNjKiLb/mxe1gMROtDeBsXOt7afr9e/lJT9qLqojgU2Bb0n6fZ2QHp0+J17NBMZO+px4BUkbAP+PskPoh5KOBqgTRMsARwCP1Wujbd+Wl/yYwLi5vzdxZHunTCwGgKT1JJ1TJ6JH9/JK14nFJUifE/GqsnOxI5JmB04HPm/7H5J2oSS9v9P2nyVtDDxac4GkSmKMldiJthI70UaNm98CO9i+VtIiwD+A82yvVV/kHrN9QeImmhI70ZakGYGzgZ1t/0vSuymVfR+1vY5KVejbbf8xsRM9iZtoS9JHgBMouTefAT5SJ6RHAqOBPYArbZ+X2IkYXAq6dGdKYGrgAUmzUSpt/gN4n6T7ba8JL2/D7rCd0X8SO9FWYifamIISNw/VzzcCB1MSmh9p+1O9GxM3MUBiJ9oaQYmf3i6h21QKb5wr6QDbX+jdmNiJhsRNtDUt8BPbB0g6BfibpBVcqkQjaX838r1229SI/pRj0R2xfSdwMXAecBFwtO3NbS8CzCVpt3pfHnwxjsROtJXYiTZs3wVcDvxC0uLA9+pX6wKz191pEa+Q2Im2bD8C/BX4iqSZ67WngE8B75Y0X5fti/6UuIm2bP8BOKZ+3Aq4C/i7pCnqtQXrfZlYjBiPTC5ORL3OqSaAxfaOwCrAucDJjVvPpK64RUBiJ9pL7EQbjbgZXi8dDDwAfA2YBfiiS5L8qYDhg/6SGJISO9GWpEUkrVjz4kF50R8D7NqbKKLsfh0FTNNBE6MPJW6irUbsLAVg+3/17+eAbYA7gXPqcfoDGvEUEYPI5OJEImlD4AZJ89cEsb0X/buBu4FvS3p7zXm2BnBJh82NPpLYibYSO9HGgLh5CcD21bZ3Aba0/ckaT1tQXtSe77K90T8SO9GWSrGxYynpOr4saYdaZOMcYDbgQEnvBzYB5gAe6ayx0TcSN9HWgNjZS9LOze9tP2N7E2BOyq773Z1CdRGvKgVdJgJJSwK/Au4HFgKWsX2Laon7muB8N2BhykB7h7qqH0NcYifaSuxEG68SN+PkGFKpMr4DsLXtf3fT2ugniZ1oqxZSOAzY1Pa/JX2OEj9b1e8Xp1QbX4ayI203p7rvkJe4ibbGEztL2f7EgPu2BvYG1rR9fQdNjZikZHJxIpC0GLC07UMk7Q9sByxp+5YB972TUjUxRxMDSOxEe4mdaOPV4qY5SSRpTeA/tm/qsr3RPxI70YYkAVsAU9o+ql6bGzgN2LDutO/dOyPwvO1nO2ls9I3ETbT1GrGzPnC/7TH1+gbALbav7aq9EZOSTC5OJJJm7m2llrQf8EnKCsnNkpazfVG3LYx+ldiJthI70cZrxM3ytv/ebQujXyV2og1JUwHT2X5I0ihgZuAMYA3bj0laOLtcY6DETbQ1AbHzgZzmiXj9RnTdgMmVpBWA+YFngNNsPypppO0Xbe9RFk34q6SDgc0lrWr7vi7bHP0hsRNtJXaijdcZN1tIWiVxE5DYifZq7CwAPAmcbvuh+tWLwLPAi/Ul/+PAVpK2dqkEHENY4ibaahE7n7D9cEfNjZgkZefiW0DSusBPKMmE3wEsAqxo+x5Jo2y/UO+7iDIoXyUrawGJnWgvsRNtJG6ircROtDUgduah5P5d0fY99fuRwAnAdcC6lPyc13XU3OgTiZtoK7ETMXFk5+KbTNJwYCtKRamz6rVDKSv3y/dW7GsnNxuwUvI4BCR2or3ETrSRuIm2EjvR1gTGzkzAysD7gA1s39xZg6MvJG6ircROxMQzrOsGTE5qgthhwHTALL3rtj8LnAucL2mKevk+YO0MtgMSO9FeYifaSNxEW4mdaGsCY2fKelzxl8BmecmPxE20ldiJmLgyufgmcvEi8GtgV0lLNr7bAbgeWKV+/pcHVG2NoSuxE20ldqKNxE20ldiJtiYwdlaul76eSemAxE20l9iJmLhyLPpNIGk5Su6G0ZQEsb+W9G7gK5J+aPvyeutTwPRdtTP6T2In2krsRBuJm2grsRNtvc7YmRGgl7Mzhq7ETbSV2InoRgq6vEGS1gF+AFwAzEGpOPVJYF5gs/rn58AoYCfgY7Zv7aa10U8SO9FWYifaSNxEW4mdaCuxE20kbqKtxE5Ed7Jz8Q2Q9CFgP2Bz21dJWhr4FjCd7VslHQDcAGxIWTnZIp1XQGIn2kvsRBuJm2grsRNtJXaijcRNtJXYiehWdi6+AZIWAZa1fUjj2iXAV21f0Lg2Ehjt/B87qsROtJXYiTYSN9FWYifaSuxEG4mbaCuxE9Gt7Fx8A2xfLel2KJ1UTRj7BPBkvbYycIXtxzpsZvShxE60ldiJNhI30VZiJ9pK7EQbiZtoK7ET0a1Ui36DbD/Z+7H+/TzwsKSNgH1IUvMYj8ROtJXYiTYSN9FWYifaSuxEG4mbaCuxE9GdHIt+k0k6FpgaeBewje3rOm5STCISO9FWYifaSNxEW4mdaCuxE20kbqKtxE7ExJNj0W8SSao/zkfpvFZIgtiYEImdaCuxE20kbqKtxE60ldiJNhI30VZiJ2Liy87FN5mkDYBbbF/bdVti0pLYibYSO9FG4ibaSuxEW4mdaCNxE20ldiImnkwuRkRERERERERERCsp6BIRERERERERERGtZHIxIiIiIiIiIiIiWsnkYkRERERERERERLSSycWIiIiIiIiIiIhoJZOLERERERERERER0UomFyMiIiJiopH0NklX1T8PSLq38fmZes87JVnS9xv/bhZJL0o6qH7+9oB/e5WkGbv674qIiIgYqkZ03YCIiIiIGDps/xdYFMoEIfCU7X3r56cat/4HWAf4ev28CXDdgF93QO/fRkREREQ3snMxIiIiIvrRM8ANkpaonzcDTuqwPRERERExiEwuRkRERES/+g2wuaR3AC8B9w34/guNI9EXTPzmRURERESORUdEREREvzoH+B7wIHDiIN/nWHREREREx7JzMSIiIiL6ku0XgH8BewCndNyciIiIiBhEdi5GRERERD/bD/iL7Ucldd2WiIiIiBggk4sRERER0bdsX8crq0T3fEHSVo3P69u+461vVURERET0yHbXbYiIiIiIiIiIiIhJUHIuRkRERERERERERCuZXIyIiIiIiIiIiIhWMrkYERERERERERERrWRyMSIiIiIiIiIiIlrJ5GJERERERERERES0ksnFiIiIiIiIiIiIaCWTixEREREREREREdFKJhcjIiIiIiIiIiKilf8PaXvNaHJ1+C8AAAAASUVORK5CYII=\n",
            "text/plain": [
              "<Figure size 1584x432 with 1 Axes>"
            ]
          },
          "metadata": {
            "needs_background": "light"
          }
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "\n",
            "\n",
            "\n",
            "\n",
            "\n",
            "XXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXX\n",
            "\n",
            "\n",
            "\n",
            "\n"
          ]
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "iVBORw0KGgoAAAANSUhEUgAABQsAAAHTCAYAAACA4U6SAAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADh0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uMy4yLjIsIGh0dHA6Ly9tYXRwbG90bGliLm9yZy+WH4yJAAAgAElEQVR4nOzdd5RdVd3/8c93SjKTTjqkkwIhCSkEQgsMTZCIgNKbPFIeAVHBnwKK0kSjIiqCUhQekCIqvfchQBJCAgmhJCGkF9LbZPq9+/fH3NzcPrede+fOvF9rsThn73P2+c5wCIvP2vtsc84JAAAAAAAAAIryXQAAAAAAAACAloGwEAAAAAAAAIAkwkIAAAAAAAAAAYSFAAAAAAAAACQRFgIAAAAAAAAIICwEAAAAAAAAIImwEAAAoE0xs5+Z2d+zPOaFZvZuNseM85yBZlZlZsVePyuJWqrMbO981wEAAJBtJfkuAAAAoDUxs2WS+kjyhTT/n6Q1kn4WOC+RVCqpJnC+3Dk3ysycpOHOucUh490oaZhz7rxs1Oec+3U2xskH59wKSZ1y/Vwzq5T0sHMuGLI653JeBwAAQC4QFgIAAGTfSc6512O0/1pqmokn6WLn3OG5LMrMSpxzjbl8JgAAAAoLy5ABAAAKlJkNNjNnZpea2RozW2tm/y+k/0Yz+6+ZPWxm2yVdGGh7OOSaw81supltNbOVgSBTZtbezG4zsxVmts7M7jaz8sTl2J1mts3MFpjZMYHG081sTsSFV5vZM3EGGWJm08xsh5m9bmZ37ao35OctMbMzzWx2xL1XmdmzzdVvZhVmtsrMfmxm6wO/t/+JU8+tkiZLujOw9PjOQLszs2GB4/8zs7+a2UuBa94zs75m9icz2xL4fYwPGXMvM3vCzDaY2VIz+0GC3ysAAEBOERYCAAAUvqMkDZf0NUnXmNmxIX0nS/qvpG6SHgm9ycwGSXpJ0l8k9ZI0TtLcQPdUSSMCbcMk9ZP0ywQ1TJL0paSekm6Q9KSZdZf0rKQhZjYy5NrzJT0UZ5xHJc2S1EPSjYFrY3lO0j5mNjyk7ZzA/cnU31dS10D7RZLuMrM9Ih/inPu5pHckfd8518k59/049Zwh6Xo1/fx1kmZI+jBw/l9Jt0uSmRUFap8XePYxkn5kZsfHGRcAACCnCAsBAACy7+nATL1df12Swr0fht4r6dok7rnJObfTOTdf0gOSzg7pm+Gce9o553fO1UTcd46k151zjznnGpxzm5xzc83MJF0q6Srn3Gbn3A41LaE+K0EN6yX9KTDO45IWSprinKuT9Lik8yTJzEZJGizp+cgBzGygpAMl/dI5V++ce1dNYWMU51y1pGd2/ayB0HBfSc8mWX+DpJsD9b4oqUrSPgl+vuY85Zyb45yrlfSUpFrn3EPOOV/g5981s/BASb2cczcHfsYlku5T4t8tAABAzvDNQgAAgOw7Jc43C5MxIdYGJ83cszLkeLmkMXH6Ig1Q02zASL0kdZA0pyl3aypFUqJdiFc751xEHXsFjh+U9JiZXa+mmYL/DoSIkfaStDkQBIbWPyDOMx+V9AdJN6sp+HzaOVdtZr2TqH9TxPcbq5XZ5inrQo5rYpzvGnuQpL0CQfAuxWqavQgAAJB3hIUAAACFb4CkBYHjgWraeXkXF3150EpJB8Vo36imgGuUc251kjX0MzMLCQwHKjAr0Dk308zq1fTtv3MCf8WyVlJ3M+sQEhjGCwol6TVJvcxsnJpmGF6VQf2JJPodpmqlpKXOueHNXgkAAJAHLEMGAAAofL8wsw6BJb7/o6Zlr8l4RNKxZnZGYNOQHmY2zjnnV9PS2D8GZunJzPo181293pJ+YGalZna6pJGSXgzpf0jSnZIaAsuLozjnlkuaLelGM2tnZodIOineA51zDZL+I+n3krqrKTxUmvUnsk7S3mneG2mWpB1mdo2ZlZtZsZmNNrMDszQ+AABARggLAQAAsu+5wK64u/56yuPnvS1psaQ3JN3mnHs1mZuccysknSjpx5I2q2lzk7GB7msCY860pp2UX1fib/q9r6ZNVjZKulXSac65TSH9/5Q0WtLDMe4Nda6kQyRtkvQrNQWfsZYs7/KopGMl/SdiWXGq9SfyZ0mnBXY2viPNMSRJgW8YfkNNG68sVdPv6+9q2mwFAAAg7yz80zIAAAAoFGY2WE2BU2lEUNbimFm5mjZBmeCc+yKF+x6XtMA5d4NnxQEAACCImYUAAADIhcskfdBcUGhmB5rZUDMrMrMTJJ0s6emcVAgAAAA2OAEAAIC3zGyZmnYjPiWJy/tKelJSD0mrJF3mnPvIu+oAAAAQimXIAAAAAAAAACSxDBkAAAAAAABAAGEhAAAAAAAAAElt4JuFPXv2dIMHD853GZ7ZuXOnOnbsmO8yUIB4d5AO3huki3cH6eLdQTp4b5Au3h2ki3cH6cjnezNnzpyNzrlesfpafVg4ePBgzZ49O99leKayslIVFRX5LgMFiHcH6eC9Qbp4d5Au3h2kg/cG6eLdQbp4d5COfL43ZrY8Xh/LkAEAAAAAAABIIiwEAAAAAAAAEEBYCAAAAAAAAEBSG/hmIQAAAAAAABBPQ0ODVq1apdra2pw+t2vXrvr88889fUZZWZn69++v0tLSpO8hLAQAAAAAAECbtWrVKnXu3FmDBw+WmeXsuTt27FDnzp09G985p02bNmnVqlUaMmRI0vexDBkAAAAAAABtVm1trXr06JHToDAXzEw9evRIecYkYSEAAAAAAADatNYWFO6Szs9FWAgAAAAAAADkUXFxscaNGxf8a+rUqTr11FM1btw4DRs2TF27dg32TZ8+XRUVFZo9e3bw/mXLlmn06NFZqYVvFgIAAAAAAAB5VF5errlz58bsq6ys1G233abnn38+J7UwsxAAAAAAAACAJGYWAgAAAAAAAHlVU1OjcePGBc+vu+46nXnmmQnvOffcc1VeXi5Jqq+vV1FRduYEEhYCAAAAAAAAkm567lN9tmZ7Vsfcb68uuuGkUQmvSbQMOZ5HHnlEEydOlNT0zcJvfOMbadcYimXIAAAAAAAAHnpm7mo9NmtFvssAksLMQgAAAAAAAA/98F9NM8bOPmhgnitBc5qbAdgWEBYCAAAAAAB4xDmX7xJQACK/WXjCCSdo6tSpeamFsBAAAAAAAMAjfrJCJMHn88Xtq6ioUEVFRVhbZWVl2PngwYP1ySefZKUWvlkIAAAAAADgET8zC1FgCAsBAAAAAAA84mNqIQoMYSEAAAAAAIBHmFiIQkNYCAAAAAAA4BEfaWFBaK0b0aTzcxEWAgAAAAAAeIRlyC1fWVmZNm3a1OoCQ+ecNm3apLKyspTuYzdkAAAAAAAAj7S2AKo16t+/v1atWqUNGzbk9Lm1tbUpB3mpKisrU//+/VO6h7AQAAAAAADAI6EzC5dt3KnBPTvmsRrEUlpaqiFDhuT8uZWVlRo/fnzOn9scliEDAAAAAAB4JHQVcnW9L3+FAEkiLAQAAAAAAPDAtEUbdOCtr+e7DCAlhIUAAAAAAAAeuOD+WfkuAUgZYSEAAAAAAEAOnHjHO/rja4vyXQaQEGEhAAAAAABAjvz5jS/yXQKQEGEhAAAAAAAAAEmEhQAAAAAAAAACCAsBAAAAAAAASCIsBAAAAAAAABBAWAgAAAAAAABAEmEhAAAAAAAAgADCQgAAAAAAAACSchQWmtkNZubMbHTg/GAzm2dmi8zsVTPrHXJtWn0AAAAAAAAAMuN5WGhmEyQdLGl54LxI0sOSrnDOjZA0TdLUTPoAAAAAAACAUPdNW6Llm3bmu4yC42lYaGbtJd0l6bKQ5gMk1Trn3g2c3y3pjAz7AAAAAAAAAEnStuoG3fri5zrnvvfzXUrB8Xpm4c2SHnbOLQtpG6jALENJcs5tlFRkZt0z6AMAAAAAACgIfr/Ldwmtnt81/Y6r6hrzXEnhKfFqYDM7RNJESdd69YwEz75U0qWS1KdPH1VWVua6hJypqqpq1T8fvMO7g3Tw3iBdvDtIF+8O0sF7g3Tx7iBdsd6dLbX+uNe/UVmp0iLzuKq2bWdDU1i4s65BF/31FZ08tJ26tG9Zv/OW+meOZ2GhpCMljZS01Mwkqb+kVyTdIWnQrovMrKckv3Nus5mtSKcv8sHOuXsl3StJEydOdBUVFdn/6VqIyspKteafD97h3UE6eG+QLt4dpIt3B+ngvUG6eHeQrljvzvibX417/cGHHq7OZaUeV9W2bd5ZL73xmhr90hsrGvXGikZ9fvMJKm9XnO/SglrqnzmeLUN2zk11zu3lnBvsnBssaZWk4yX9XlK5mR0euPR7kv4TOJ6TZh8AAAAAAECLsaW6IW5ffWP8WYfIjkZ/9O/4xDveyUMlhcfLmYUxOef8Zna+pHvMrEzSMknnZdIHAAAAAABQKGoafPkuodWLkRVq6UZ2Rk5GzsLCwOzCXcfTJY2Jc11afQAAAAAAAIXgiTmr9cNjh+e7jFbN59hEJl1e74YMAAAAAACAEP+evTLfJbR6Ph9hYboICwEAAAAAAHJo9daafJfQ6jGzMH2EhQAAAAAAAGhVfLE+WoikEBYCAAAAAACgVfGRFaaNsBAAAAAAAACtis/PMuR0ERYCAAAAAACg1dhUVaer/z0332UULMJCAAAAAAAAtBo/f+oTLfhqR/B88vCeeaym8BAWAgAAAAAAoNWobfQFj48d2Uf/vGhSHqspPISFAAAAAAAAOeYc39TzSuXCDcHjob065rGSwkRYCAAAAAAAkGMNPsLCXOi3R3m+Syg4hIUAAAAAAAA5VheyVBbeOWFU33yXUHAICwEAAAAAAHKsvtGf7xLahPalxfkuoeAQFgIAAAAAAORYvY+wMBeKiyzfJRQcwkIAAAAAAIAs69ahNGE/Mwtzg6wwdYSFAAAAAAAAWWaSenduH7e/jrDQMwcN7h48bldM9JUqfmMAAAAAAABZ5pR4CSwzC70zqEcH9e1SpkW/+rpKCAtTxm8MAAAAAAAgy7ZWN2h7TUPcfmYWesepaflxu5LkYq81W2v05YYqb4sqIISFAAAAAAAAHthZ79M7Pz1KBw7eI6qvrtGXh4raBucks+Q/Vnjo1Dd1zB/e9rCiwkJYCAAAAAAA4JEB3TvEDK42VtXnoZq2wcnlu4SCRlgIAAAAAADgoVhz3P79wcqc19FmOCmFiYVBa7bWZL+WAkRYCAAAAAAAkEXOhc9sixVc1TawDNkrTumFhU99tDrrtRQiwkIAAAAAAIAsWbKhSs9/vLbZ62oICz3jnJPFnM+Z2Ivzm//n1haU5LsAAAAAAACA1uLoJDfK+GpbrceVtF3pziz8dM32rNdSiJhZCAAAAAAA4KGy0uKotk072eDEK87F/k4kkkNYCAAAAAAA4KHfnbZ/zPYXkliujNQ9O2+Nlm2qjtm3/42vqKquMccVFRbCQgAAAAAAgCyI3Nhkl96dy2K2X/Hoh16Wgxi21zZq9A2vBM8jN5rx+2P/M2xLCAsBAAAAAACy4IH3lsXte+enR+m9a4/OXTFIyvaahrDzE+94J0+VtBxscAIAAAAAAJChFz5eq5uf/yxu/4DuHXJYDZJlETuhLPhqR54qaTmYWQgAAAAAAJChNz5fF9X2g6OH5aESpGJLNRvNRCIsBAAAAAAAyNCTH62Oajtn0qA8VIJUvL9kU75LaHFYhgwAAAAAAOCB8nbFUW1FJrGHRv41+PxyTtpZ72v+4jaGmYUAAAAAAAAeKCuNjl1KiohivHTDM58kdd1Rt1VqxPUvqbquMaov3q7WbQVvKAAAAAAAgAfaFUfHLmSF3npwxvKkrlu1pUaS9J85q6L6fvXC51mtqdDwigIAAAAAAHggcqddSSqOaKttYBlsPq3dVhvV9o93l+ahkpaDsBAAAAAAACBHiovCw8KNVXV5qgSIjbAQAAAAAAAgR0ojlibHmn0I5BNhIQAAAAAAQI4UFREOomUjLAQAAAAAAMiyju2KY7ZHZoVtfefddDT4/Drjnhn6YNnmfJfSKhEWAgAAAAAAZNm3D+gfs31or05h535/LqppXV77bJ1mLd2s0++eEfeaf3xnYg4ral0ICwEAAAAAALLosoqhuu7rI2P23XrqmLBzPzMLU3b5Ix8Gj+sad+8mHbqz9F7dylMa8/en7Z95Ya0EYSEAAAAAAEAWfXtCP5XHWYY8sHuHsPP2pUQzsfz25QU6854ZGnztC/pqW23c61Zsqg4e//WtxcHjkXt2Sel5px3QXzOvOyb1Qlsh3kgAAAAAAIAsavTHny0Yub1JMRuexPS3yi/1/tKmbxI+M3d1sH39jvDg8K2F64PH++2VWkAYyszUt2tZ2ve3JoSFAAAAAAAAWeRLFBZGZINFkQ2I2vTlNy8tkCTNXblVB936Rljfv2atDB53KS/NWg3+BP8MW7uSfBcAAAAAAABQyFLZ0dgiwkE+WRgtVk73/Mdr9P1HP4pqH9O/q9ZsrdH97y5VhzhLv9Orwakoah5o20BYCAAAAAAAkAFfRLi1Xwrfy3MiLQw1Z/kW9ezULqo9VlAoNS35PnTqmxk983ffjt7cpA1PLCQsBAAAAAAAyITPH34eOXswoTYcSsXy7b9NT+n6VZurm7+oGadP7B88HrlnF32+dnubDnH5ZiEAAAAAAEAGGtturpRVq7fWpHzPvFXbMn5uaLj7zbF7SWrby8MJCwEAAAAAADLQkMGa1TacSUU5+rbKfJegXZtT+9twWkhYCAAAAAAAkIEGX/r3tuFMKkpdo7/5izI0pGfHhP27dqduy98sJCwEAAAAAADIwLb63cnS8N6dUrq3LX8bL1RdYwaJawqeu/LwhP3LN++UJNVmkgAXOMJCAAAAAACADPxrQX3w+OdTRqZ071Mfrc52OQXnm3e+q32ufzknzypqZu+Zh2eukCRd+8T8HFTTMhEWAgAAAAAAZGDx1t3LZyv26Z3SvQu/2pHtcgqKz+/0cRY2KQn1wIUHxu1Ldtn32m2pb7bSWhAWAgAAAAAA5En/PcrzXUJe3TttSdbHXL5pZ9w+XyAt7Ny+JOEYn67ZntWaCglhIQAAAAAAQJ7s0aFdvkvIqxWbq7M+5tfH7Bm3r1O7Eo0b0E1/Omtc1p/bWiSOUQEAAAAAAJBVZruXw/ra8ra7khp9qe2A/JPj99HvX1kYs+/DXxyn+ka/+nQpi3t/UZHp6SsOi9u/T5/OWrhuhw4f1jOluloTZhYCAAAAAADk0EPfPSh43NjWw8IUf/7R/brq9auP0Js/PjKqr3vHdurbNX5QGGnK/nvq8oqhYW3tS5uisncXb0yprtaEmYUAAAAAAAA5NHl4r+Bxo69th4Uzl2yK2/fmj4/U7GVb9NMnPg62FZk0rHfnrDz7rnMmRLWdfdBAfbyq7e6ELDGzEAAAAAAAIOfMmv7u86e2DLe1WbutNm7f3r066YwDB4S1mczTevbu2dHT8QsBYSEAAAAAAEAGjh7QtHAz1tLYeIoDaWFbX4acqsXrdwSPbzlldPC4c1l2Fs+29W9ISoSFAAAAAAAAGenTsSle6dGxfdL3FBc1hYU+RziVig9XbA0en3/woODxvedPzMr4JcVEZZ5+s9DMnpY0RJJfUpWkK51zc81shKQHJfWQtEnSBc65LwL3pNUHAAAAAACQD9UNTYFfSXHyS2RLikx1knxt/JuFkTq2K9bOel/c/qKIX/Gz3z9Mq7fU6JChPRKO+7dzJ6g+iZ2XiyMf0AZ5HZd+xzk31jk3XtJtku4PtN8t6S7n3AhJd0m6J+SedPsAAAAAAABy7pkvGyTt/g5hMnaFUixDDvfQRZMS9vfoFD57c//+3fT1MXs2O+7Xx+ypk8f1y6i2tsLTsNA5ty3ktKskv5n1ljRB0mOB9sckTTCzXun2efkzAAAAAAAAJKN9SXHS126vbZQkLdu006tyCtKIPp0S9p87aaDHFRDeeroMWZLM7O+SvibJJJ0gaYCk1c45nyQ553xmtibQbmn2bfD65wAAAAAAAEgknSWslQuJNEKVNvPNwOb6M8UnJHMQFjrnLpYkMztf0u8l/cLrZ5rZpZIulaQ+ffqosrLS60fmTVVVVav++eAd3h2kg/cG6eLdQbp4d5AO3huki3cHmUr3/bnl4dc0uX9pdospUG9UTgsex/p9zpg5U1928C4wXLRl9/cSvf7zoKX+meN5WLiLc+6fZnavpFWS+plZcWB2YLGkvSStVNPswXT6Ip91r6R7JWnixImuoqIiFz9iXlRWVqo1/3zwDu8O0sF7g3Tx7iBdvDtIB+8N0sW7g7S9/IIkpfb+BO6RpK+shyoqJmS5qAIR8nuQpBOPrdA3N87VmQcO0GHDekZdc8jBB2tA9w6elVO+ZJP0/kxJKf7zTENL/TPHsyjWzDqZ2YCQ85MkbZa0XtJcSWcHus6W9JFzboNzLq0+r34GAAAAAACARJ6ZuzrjMRzfyQsyM91x9vjdQWGEXp3bx2zPFvab8XZmYUdJ/zGzjpJ8agoKT3LOOTP7nqQHzeyXkrZIuiDkvnT7AAAAAAAAcuqH/5qb8RjzVm5r/qI27JkrDtPyzdU6YVRftSvx+puFpIWehYXOuXWSDo7Tt0BSzL2w0+0DAAAAAAAoRKu31uS7hBZt7IBuGjugW06eNSywG/Nfzh6fk+e1RN7GsQAAAAAAAEjamq01+ssbXzDDLU96dy7TsqlTdNLYvfJdSt4QFgIAAAAAALQQFz84W394bZG+3LAzqq+qrlE76xrzUBXaEsJCAAAAAACAFmJ7bYMk6ez7Zkb1jb7hFY2/+bVcl5QTN560n577/uH5LgPydoMTAAAAAACAVit0qXBZaXbmY321rVaStGFHnXx+p+IiC+uv9/m1cnO1BnTvkJXntRQXHjYk3yUggJmFAAAAAAAAadheu3tJ8JkTB6R0730XTIzZ3ujfHUDOXbkl5jWzlm6Wc071jf6Ungkkg7AQAAAAAAAgDe8t3hg8/vmU/VK6N5mZiHe/vSRm+9aaBv3mpQUacf1LWr+9NqXntiRDrntBknTYsB55rgShCAsBAAAAAADSsKmqLnjcriS1iMVkzV6zvaYhZvuWnfW6d1pTkLh4Q1VKz21Jdq3iPnBw9/wWgjCEhQAAAAAAAGkIXTKcqqLms0K5OMPf+dbi4HHfLmVp19BS+DL4PSL7CAsBAAAAAABS9K9ZK3TTc5+lfb9Z82nhuh27lxg/+v6KmNe0hpjtrYXr810CQhAWAgAAAAAApOjaJ+dndH8yMwuXb6oOHj86a3nMa56ZuyajOlqCT1Zvz3cJCEFYCAAAAAAAkGNFEWnhzrrGOFc2iReo3fHGF1mrKZc+WLY53yUgDsJCAAAAAACADHQqTf2eyJmFlQs3pLWzcaobq7QUp989I98lII7CfKMAAAAAAADypLbBF3Z+7sj2KY8R+c3CDTtqdc7f34957W9e/DzuOP26laf8bCARwkIAAAAAAIAUXPnYR2Hnk/YsTnmMooiw0O+kFSHfKAx1z7QlccdZunFnys9uaQ4d2iPfJSAEYSEAAAAAAEAKXvtsXdh5ZPCXjMg7/K417GucnjvPmZDvEhCCsBAAAAAAACBNR+/bO637omcWto2w8Dv3z9KTH64Ka+vesV2eqkEsJfkuAAAAAAAAoFAdNKS75FJfChw5GfGLdVWq9/lTHqe0OPVZjfn09qINenvRhnyXgQSYWQgAAAAAAJCmSybvndZ9kTMLq+oao67pWl4atZlKpMsrhqX1fCAewkIAAAAAAIAE/H4nF2OZ8F3nTFBxUXoz+4oiEplYS3GPH9Wn2eXJ/fdgN2RkF2EhAAAAAABAHFur67X3z17U2ffNlKSw0HBo745pj2sRW5x0KS+NuiaZzxhuqKoLO6+ub9TIX7wctQlLS7Bo3Y58l4AkEBYCAAAAAADEsXlnvSRp5pLNkqRG/+4Eb+222rTHjZyQ6PNHJ4MbquqaDQx/9/JCDb72hWCI+cW6KtU0+HTJQ7PTrs0r3/7b9HyXgCQQFgIAAAAAAMSxa5nxgO5Ny31DQ72uMWYDJssivlkYKyysXLhB22sbkhpvzvItkqSf/Hde2jV5bUdt9HcZ0fIQFgIAAAAAAMRR39i0Q3FVjKBraK9OaY8buRtyrLBQSm4psiSddvcMSdKidVVp1wRIhIUAAAAAAABxPT13tSRpS3XTDL8NO3Z/IzDNvU0kSZG3xgsL+c4fco2wEAAAAAAAIA6fP/z8yQ9XB4+TnPQXU9Qy5DhTCC984IO0nxFrB2egOYSFAAAAAAAAkpZsqNJ/56wKa9u/f9ew8z++vih43Ll9SdrPippZ6Esu2Ou/R3ncvrpGX9j5p2u2p1oWoPTfagAAAAAAgFbk6D+8LUk67YD+wbaendrHvT5ydmAqor5ZmOQswHevOVq1DT45J4385cthfXdXLgk7/8Zf3tWyqVPSrjEXDhy8R75LQATCQgAAAAAAgDjiLeU9YkSvjMY1Nb8bcqQLDhkkSSorLY7ZP3/11oxqyoeHL56U7xIQgWXIAAAAAAAAIXbUNgSP4834u6JiaFaf+c4XGxL2j+7XRTefPDrhNdtrondsbkkmD+8Z1da+JHbwifxJOiw0syIzG29mU8zsaDPr7WVhAAAAAAAA+TDmxleDx+fc937wOHSWYbzZfcmKXIa8sao+4fU/P3G/ZscsauFTwt75YmO+S0ASml2GbGZDJV0j6VhJX0jaIKlM0ggzq5Z0j6QHnXP++KMAAAAAAAC0XDO+3NTsNR+v2hY8HtC9g5flRGlf2nwSOHPJ5hxUkp6bnvs03yUgSclkzr+S9LCkoc65451z5znnTnPO7S/pm5K6SjrfyyIBAAAAAAC8dOuLn4Wdb6yqi7rm8dkrg8fdO7bL6Hmp7o2yeH1VRs/LtwfeW5bvEpCkZsNC59zZzrlpLsYXPZ1z651zf3LOPehNeQAAAAAAAN7bWecLO2/0RX+r8NH3V2TteanupLxPn85pPeeB95amdV82VdXF/pbi/v275rgSJCOt1exmdpiZnWCZ7BEOAAAAAADQQizduDPs3Ck6LOzWoTRX5UQpKU4+ghk/sFvw+KbnPktwpbemL96olZur9afXFoW1P3LxJN166pjApLwAACAASURBVGg9+/3D81QZEkkqLDSzh8xsVOD4e5LulHSlpH94WBsAAAAAAEBeTFsUvTvxjSeNytr4qc6+KkphvlacDZxz7py/v6/Jv3tLvTq3D7adOr6fDhvWU+dOGpTHypBIs2GhmQ2SNFHSjsDx/6opKLxC0sFmNtDMunhbJgAAAAAAQO5MWxS9c++z89ZIkn777TEZj99c9jd5eM+w81TCwuu+vm86JUX5a+ViTV+c+Q7GoTtH337G2IzHg7eSmVlYIamTpGMknSypm6S9JR0ZuL9C0mBPqgMAAAAAAMiDF+avjWp7c8F6SdKc5VsyHt8SzC08flQfdSkLX/JclCArvO+CiWHn72YY8M1cskkrN1frdy8v1Dl/fz+tMTbs2L1BzCPvLw8e80W7li+ZDU4elPSipHMknSHpbufcQ5L+K2m9c+4h59zH3pYJAAAAAADgDb8/9rrdGHu9SpI6tc/824WJMrPffnv/qP5EIduxI3uHnX999J6ZlKaz7p2pyb97K6MxrntyfvB40brC3sm5rSlJ8rrLJR0vqd4590agrYekn3hSFQAAAAAAQI40+P0x2xd8tSNme9+u7WO2pyLR/LpuHdpFhYMrN1drWO9OsceKuLZzWbJxT7TVW2ui2pxzKc8IrGmIvQMyWr6k3h7nnF/SSxFtKyRlb89wAAAAAACAPNhW0xCz/YL7Z8VsLy1Oar/YtIzaq2lbiMhlx6k8c9POepnt3ugklbBvxabqqLZ6n1/tS4pjXB3fe4s3pXQ9Wo5kNjh5zsxOMrOoObZmtreZ3Wxm3/WmPAAAAAAAAG99EWeZbOh390K1K8lCWBgnuztoSHdJUr9u5WHtw/vEnlUoRS+XLi8t1nvXHB0831AV++eI5ez7Zka1rd+e/P0ofMm83ZdImixpgZl9YGYvmtmbZrZE0j2S5jjn7ve0SgAAAAAAAI+s31Gb0vXZmFkYb4OT7TVNy3evOm5EWHtxjB1OdrX5Ir65OKJPJ3XrsHvO1+ot0UuLU9G+1LuZlGh5ktng5Cvn3E+dc0MlnS7pFklXSxrtnDvOOfeM10UCAAAAAAB45arH50W1HTGiV9zr22dhZmG8VcFPfLhKUnQgWVPvi7r2oe8epClj9owKEs0srK2mIfreVDT4Ym/0kqqTxu6VlXHgrWSWIe8bcrrWOTfDOTfXOVdtZgd7WBsAAAAAAEBOPXHZIerYrlgD9iiPe02sWX6pynwE6bBhPXXXuRPCvkf4yU3HS1LYNwZLipoPN9/5YkPUDMVdauqzs1nJHWeNy8o48FYyUfijIcczIvr+msVaAAAAAAAA8uqAQd1V3q5Yj7wff0/XOJmap3p1Tm4H5k7to/eyPeOeGTr29rdV1xh7huHtry7U+f+YpUsemh2z/4kPVydfaAKp7qiM/EgmLLQ4x7HOAQAAAAAACtIPjh4mSdpRm3gmXdfyqD1gU5ZKcHbhoYNVVprabsSStH//rsHjxeurYu50LEl/f3epJOnNBetj9g/vHX9zFbQ+yYSFLs5xrHMAAAAAAICCdEUgLKxr9Ce8LtbsvVSlMvvq2JF9mr3mnxcdpJu+OSqs7fKKoWHnL87/Kuo+55yqY3wPMdTTc9ckUWVis68/NuMxkBvJvN39zewONb3Hu44VOO/nWWUAAAAAAAA5FPqdv0Sy8s3CFIZI5trJw3tp8vDwTVkuf+TDsPPXP1+nHx47PKzt4ZnLmx372JG9my+gGT07JbeMGvmXTFj4k5DjyMXrsRezAwAAAAAAtFLFWfj2nsWZWzg2ZOlw8No0Hxf5bcUGX/SMyV8882mz42yqqk+vABSkZMLCxyV1ds5tCG00s16SdnhSFQAAAAAAgIf8fqd5q7aqYxpLipPYXLh5cQLAcycNin5eljYGibXb8fGj+uiVT9clvK9y0QZdddyIpJ+zaF14XPS1/ZpfRo2WI5nX+w5Jk2O0Hy7pj9ktBwAAAAAAwHv3TFuiU/86XV/747SU783GMuR4bn9tUfB4TL+mWYbpPu32M8aGnTfGCAtLkkg+563cqnE3v5r0cyN/p385Z3zS9yL/kgkLD3DOPRnZ6Jx7StIR2S8JAAAAAADAW799eUHa95ansTNxpHiTBb/aXhs8vu30sTpyRC+NHdAtrWf0iPhO4KAeHaKuOWRoj7j3TxmzZ/B4a3WDLnlotpZu3JlyHcl+CxItQzJhYfSblNr9AAAAAAAAeVdT71N1fWNGY3RuX6JBPTpmXEsyswX36dtZD373IJWlGU76XfhMwrMOHBh1zfVPfxLVtiskjAwXX/tsnY66rTKtWlA4kgn71pvZQZGNZnagpA0xrgcAAAAAAGgRdtQ26Jz7ZmpnXaPG3PiK9vvlK5KkgwZ3T3msf150kObfdHxW6rIsfYcwEX/EsuNkV0+P6d9Vfz5rnK5O4TuF8Z6JwpPsbsj/NrP/kzQn0DZR0gWSzvKoLgAAAAAAgIyd/49Zmrtyq0bd8Eqw7Y3P1+lro/po1rLNKY01eXivrNXlfVQoHTBoD3VuX6IddU2zKSNnGsbaHVmSqmobdfKR/dJ65idrtoWdL5s6Ja1xkD/Nzix0zs2SNElN7/GFgb9M0iTn3PteFgcAAAAAAJCJTyPCK0m66MHZ6lwWPn9q376d447Rrrgwv8LWrUO7sJmQkRuc1Db4Yt43IsHvojnTFrEItdAltT+4c26dpBs8rgUAAAAAACCrGnyxl8Ve88T8sPMrjx4ed4wPfn6sVm6pzmpdu1Yhty8p0ncOHax7py3J6vix+CLCwnhLoY/Zt3faz9inb5fg8eThPdMeB/nTbFhoZvMlxfo3yyQ559z+Wa8KAAAAAAAgC7qWl2pbTUPCax65eJIOGxY/2OraoVRdO3TNal0WshD5ZyeOzEtYGE/H9knNLYvpkodmS5Iurxiqn56wb9rjIH+S+af/Dc+rAAAAAAAA8EBzQeGtp45OGBR6Jd7+Jj84Jv4Mx0xFLkN++qPVnj3rmJF9PBsb3krmm4XLY/0laaWkw70vEQAAAAAAwBvf2H+vvD4/cq5f5wxm9TUncmZh5cLd3xe8/8KJCe/9ziGD9MNAkJnMDMVC/c4jkggLzayLmV1nZnea2desyZWSlkg6I8F9PczsRTNbaGbzzexJM+sV6DvYzOaZ2SIze9XMeofcl1YfAAAAAABoOTbvrC+IzS66lpcm7P/L2eM9eW5wZmFE7jawRwdPnidJ81Zu1XfunyUX2BX59c/XBfs274w9A3PK/ntKkm785ijVBDZEeWLOKs1fFb1xzLrttcFjF/OLdigEycS8/5S0j6T5ki6W9Jak0ySd4pw7OcF9TtLvnHP7OOfGSPpS0lQzK5L0sKQrnHMjJE2TNFWS0u0DAAAAAAAty4RbXtMF98/SjC835buUuBbcckLC/mVTp+iksd7OPIwM1Yb26pj1Z/zzooMkSf/6YKXeXrRBQ657UR+u2BJ2TeTu0Lv88Yxxmn39sTKz4HcVf/rExzrpznejrp306zeCx/WN/myVjxxLJizc2zl3oXPuHklnS9pP0vHOubmJbnLObXbOVYY0zZQ0SNIBkmqdc7veqru1e4Ziun0AAAAAAKAFunfal/kuIa6y0uK8PXvXBicuYgJeqQfLd/fv1y2q7Vt/nR52Pn5g9DWS1K6kSD07tW/2GZt31oed75qFiMKTzBsYnIfqnPNJWuWcq01wfZTArMDLJD0raaCk5SFjbpRUZGbdM+gDAAAAAAAt0FsLW/5S5HzYtQw5crGuF2FhcXGc3VRC9O5cpgcuPFAf/eK4uNdcE2N349oGn5xzUWHhiD6dUy8ULUIyX80ca2bbpeCe3uUh58451yWJMf4iqUrSnZJOTavSFJjZpZIulaQ+ffqosrLS60fmTVVVVav++eAd3h2kg/cG6eLdQbp4d5AO3huki3fHO/n6vfojpu19Y+9SPb9k97f5mqsr2brTeXd2bRLi97uwe2e9P0Pd2mc3MKz3Nf/9wMrKSpmkeV/Fv6ZvbfjS4pdef0uXvV6tUT2KdMY+7cL6PpszQ5/H2/IZklrunznNhoXOuYzm5JrZbZKGSzrJOec3sxVqWo68q7+nJL9zbnO6fTFqvlfSvZI0ceJEV1FRkcmP0KJVVlaqNf988A7vDtLBe4N08e4gXbw7SAfvDdLFu5NlL78QPNxj6DiNHRB7mauXvvt/H0iqliQN791Jv7/wcD3/y5eD/XH/eQdqT/Z9SOfd8fmd9OqLkgWes+uZkw9Xtw7tEt+cogafX3rtpbj9J43dSxUVzW/kUlPv01WVu39/z6ztIqlan27ya/TYCdL0pqXNT11+qMYP3CPjulu7lvpnTjK7IZeZ2Y8CuyFfamZJ7+FtZr9W07cGT3HO1QWa56hpduLhgfPvSfpPhn0AAAAAAKCF+HJDVdj5P95dmpc63lywPnj82tVHqrxd8vOh2pdkfzlwqHhz7oqLsj8br7iZGX43fXNUUuNE/v5e/nT3NMRNVXXB4y7N7DCNli2Z4O9BNX238B1JJ0oaJemHzd1kZqMkXSdpkaTp1vRiLnXOnWpm50u6x8zKJC2TdJ4kBWYeptwHAAAAAABajmP+8HbY+bPz1uiOs5ufuRbLGXfPkEz69/8eko3Sgg4YFH/m2/wbv6aiHC2hjdzgxIuwsCjOmKcf0F/PfbxG3TtmPpNx/uptweN+3cozHg/5k0xYuJ9zbowkmdk/JM1KZmDn3KeKE5Q756ZLGpPNPgAAAAAAkH8uMv3K0KxlUV8fS9m+fXdvtjF5eE+988VG3XfBxLjXdy7zfmZcvCzSi7AwloP37i4zqVt5dpY8HzOyj/7y5mJJ+d1lGplLJiwM3Q250fg4JQAAAAAAiOPZeWvyXYKk3RuISNLBe/cIHj/03YPkXPzZdrkSL19pbslwtjT6nHz+7IWTp9z1niTpkslDsjIe8ieZBfhjzWx74K8dkvbfdRzYFRkAAAAAAECS9MN/zc3aWP6QwO+JOatSurfBt3vn3m4dds8UNLO8B4WhLj48PFzL1czCBr/T3JVbVJTlTzOGLkdGYfJ8N2QAAAAAAIBULNlQpd5dyrRxx+5NM378n3n69gH9kx4jNCw8feKArNaXLcumTolqy9WKznkrt3oy7kWH7+3JuMgdb7f2AQAAAAAArdqj76/Q4Gtf0Oad9ZKkK44aKkk6YkSvtMc8+g9va/QNr6iqrjFmf1Vdo3794ueqa/TFHWNjVVM9N31zFBtu5FCPTtn5BiLyh7AQAAAAAACk7WdPzZckXfV40/LjbTVNWx/cesrotMYbfO0LweO73loc1uec05zlW/SXN77QvdOW6LZXFsYd56jbKiVJc5ZvSauO1ub4UX2yMs671xyVsH/PrmVZeQ7yh7AQAAAAAACkbOXmai38akfw/O1FG3TZw3P08MwVkqTS4swjh5c++Srs/Op/z9O3/zZd90xbIkm6752lMe/7zUufB49byoYr+XbOpEFZGaf/Hh20bOoU/frUMTH79+zKLM5Cl8xuyAAAAAAAAEFbdtZr8u/eimoPDfe8WI761Eerk7runreXZP3Zha5beWnzF6XgnEkDdc6kgZLCZ4Oi8DGzEAAAAAAApOSoP1Q2e01JkengvbtLksYP7OZZLbOXbU7YX1ZK9CFJA7p3yHcJKBD8GwMAAAAAAFKytbqh2WvMTP+69BBJ0kcrturLDVWe1HLa3TM8Gbe12bXHcrcO4TMMO7Qrzn0xaNEICwEAAAAAgOd+9K+5no3tnIvb95tvxf62Xltlkv581rjg+fRrj854zJ+fODLjMdByEBYCAAAAAIBm1Tb4tGZrjbYlMaswlvmrt2W5ot1qG/xx+44ZmZ1dgFsLJ+nkcf2C5+1KMo+G1u+ozXgMtByEhQAAAAAAoFkXPzhbh059Uzvq0gsLk/HM3OQ2MInU6I8dFn5x69fVpSy7G3u0Nu1LMl+GHG9XahQmwkIAAAAAANCsdxdvlCTNWpp4Q5FM/DCFpcp3nzcheHzmPTP16ZqmmYvzVzX9/YfHDFdpMbFHc4qLrPmL0Kbwbw0AAAAAAEja1f+eF9X26U3Hh52P6dc1q8988vJDg8c/OX4fHTh4D50wes9g22drt2vKHe/qluc/00l3vitJal9guyA/f+Xh+uOZY/NdBkBYCAAAAAAAMtOxfUnw+PopI/XclYfHvK5y4fqkx3z6isM0eXhPHTmilyYM3EOSdNQ+vXTFUcP0n+8dGvOef7y7ezlsWRaW1+bS6H5dder4/jl95qQh3bMyzl/PbZrl+d/vHZKV8ZBfJc1fAgAAAAAA2rJ/f7Aybt8FhwwKO7948t5xr73wgQ+0bOqUmH1+/+4djU8c01fjBnTTPy+aFGx779qj1aNju2RL1p9eX6TvHj4k6etbuw7tm8LTCw8dLKnpe47Flp0lyCeO2TPuP1cUHsJCAAAAAACQ0E+f+Dhu3/8eOTQrzzjz3hnB475dyqP6+3WLbktke21jxjW1Ju1LisMCPb7niHgICwEAAAAAQMpuOWW0/H6Xcog3d+VWjRvQLar9g2Vbgsdj+nfJuL6HQ2YlAkgeMTIAAAAAAEhZr07t9J3AklZJuvq4Efr5iSObve9Xz38WPF67rUZT7nhH63fUhl3zjf33SqqGcyYNjNt3yNAeSY0BIBxhIQAAAAAAiGvl5uqY7f83fVnY+Q+OGa5Ljoj/vcJdZi/fPYPwoRnL9ema7XrgvfCxkl0ie+spo3XAoD2i2pdNnaLioux8jw9oawgLAQAAAABAXNtqGmK2H7df34zH/njVVknS3yq/DLZ1bp/8F9PMTE9cFr4z8s9O3DfjuoC2jLAQAAAAAADE9epn62K2nzq+X0bj1jb49N7iTVHt8286PqNxjxzRO6P7gbaOsBAAAAAAAMTVviR2dFBWmlmkcMlDszO6P559+nb2ZFygrSAsBAAAAAAAcR05olfw+MDBu78P6PO7jMZ954uNGd0PwBuEhQAAAAAAIK6T73ovePz/vrZP8LhDu+S/LRjJn2HQCMA7hIUAAAAAACCu0BmEQ3p1DB5nstvwI+8vz6imeG45eZQn4wJtCWEhAAAAAABISu/OZVkZZ0t17B2W0/Xdw4ZIks47eFBWxwXaIsJCAAAAAADQrJ8cv0/zFyXp9tcWZW0sSfrlSftp2dQpMkt/tiOAJoSFAAAAAACgWacf0D/fJQDIAcJCAAAAAADQrN5dsrMEOZHZ1x/r+TMAJEZYCAAAAABAG/fRii2qbfAlde0Tlx2qt39SkfTYZ04ckPS1ncvS32EZQHYQFgIAAAAA0Iat3VajU/86Xdc9OT+qzx+yE/IuBwzaQ4N6dIxqj+e3p+2f9LXFfHMQyDvCQgAAAAAA2rCtgZ2Jn/podVTfmwvWe/78IT13B4/FRYSFQL4RFgIAAAAA0IZc9+R8Db72BTnXNGvQFzF7cPqXGzX42he0qapOf3lrsef1vPKjI4LH7GYM5B9hYRuwtbpeY258RYvX78h3KQAAAACAPHts1gpJ0ocrtkqS6hrDv1V4zn3vS5Je+XSdenVq53k97UqKtGzqFC2bOsXzZwFoHmFhG/DnN77QjtpGHXv7tHyXAgAAAABoIc64Z4Ykqaou9sYmP3tqvl7/3NtlyDectJ+n4wNIHWFhK9Xo8+v3ryzQ9toGVS7ckO9yAAAAAAAtzK7lxzvrGoNtm6rqYl77t3MnZOWZv/zGfmE7Hh82rGdWxgWQPYSFrdTzH6/VXW99qd++tEDLN+0Mtq/fXpvHqgAAAAAALcni9TtUXb97ZuFDM5bHvK7fHuVZed53Dx+iOdcfFzzfu2fyuyoDyA3CwlaqweeXJNU0+HT8qL7B9j+8uihfJQEAAAAAWpilG6tVHJIM/PmNL2Je176kOKPnnHfwQP3zooMkSSUhOx6XFBNLAC1NSfOXoBDt2svqyQ9Xh7UvXMcmJwAAAADQmtU2+HTa3dN1z/kT1a9b4hmBdY0+XfX4vGbHHNGnU0Y1/eqUMcHjoiJ2PAZaMiL8VuqjwK5WkeaujN0OAAAAAGgdzvv7+/pk9XYdNvXNZq8N/V5hImYEfEBbQVjYSu2obch3CQAAAACAPJi9fEvS117zxHwPKwFQiAgLW6nhvTvH7XPOxe0DAAAAABS2dnwHEEAG+BOklfrj6/E3MmnwERYCAAAAQGtT2+DTXW8tVn1gw8tIVzz6oV6cvzbHVcXWs1P7fJcAIA42OGmDlm3aqRF94s88BAAAAAAUnlue/0yPvL8ibv8LH6/VCx+nHhYO7tEhk7JieuPHR6qm3pf1cQFkjpmFbdCWnfX5LgEAAAAAkGXxgkLnnBat25HSWI9dcrDuu2CiJOmJyw7NuLZIXctL1bdrWdbHBZA5wsI26L9zViXs9/udzrhnht5auD5HFQEAAAAAvHL2fTP1tT9OS+meQ4b20HH79dGyqVPUgyXDQJtCWNgKbdhRl7C/qq4xYf/bizZo1tLN+p8HPshmWQAAAACAPJi5ZHNS1/3+tP09rgRAISAsbIUOvPX1hP0vffJVwv66Rr4bAQAAAABtzekTB+S7BAAtAGFhGzGkZ8ekr126sdrDSgAAAAAA2Zbqt+kjv0N4zQn7ZrMcAAWMsLCN+PNZ43T4sJ5JXdu36+7vUTzw3lKvSgIAAAAAZMnnX21P6foDBu0Rdj55eNP/L77z06P02c3HZ60uAIWHsLCNKDLTQUO6J3XtUx+tCR7f9NxnXpUEAAAAAMiSBWtT2+041LfG99Pofl0lSQO6d1CHdiXZKgtAASIsbCNG7dVF5aXFYW3f+ut7GnztC1HfKJy2aEPY+R1vfKGt1alNaQcAAAAA5MZTH63SwzOXJ33981ceHnZe7/NnuyQABYywsI0wM5UWW1jbhyu2SpKem7c24b23v7ZI425+zbPaAAAAAADpmbN8i656fJ6WbNwZs3/wtS+EnT9zxWHBWYS7lEVMLAHQthEWtjLOubh98f4D0K281KtyAAAAAAAeqqprTOn6sQO6RbVFrkID0LYRFrYyDb74YWHXOKHgxQ/N9qocAAAAAICH/AkmjCTrhNF9s1AJgNaCr5a2Muu218btKylOnA0P+9mLOu2A/tkuCQAAAADglcyzQh06tEfmgwBoNZhZ2Mqcctd7weM/nzUurK+kaPc3CyO/W7GtpkGNfqd/fbDS2wIBAAAAAFmTjZmFZtb8RQDaDMLCVmbTzt27Fh+8dw+NH9gtuPy4qCj+fwDG3vSq57UBAAAAALJrW01DvksA0MqwDLkVKyst1lOXHxY8L0kQFsYycdAemr18S7bLAgAAAABkyVcRn6Ia2L2DLj1ib13/9Cdh7becPEqHDeuZy9IAFCjPZhaa2W1mttTMnJmNDmkfYWYzzGxR4O/DM+1DbJHZYFGKU8vXbov//UMAAAAAQP6VFoX/b/20nx6l8w4eFHXd+YcM1t69OuWqLAAFzMtlyE9LOkLS8oj2uyXd5ZwbIekuSfdkoQ8xRH53oqQ4+bDwd6ftr+9VDM12SQAAAACALCpOcQVZqFd+dIRuOWV08xcCaFM8Cwudc+8658J2yzCz3pImSHos0PSYpAlm1ivdPq/qLzSfrtkWtWlJ5H8zUvmPyKnj++mb++8V1tbo86ddHwAAAAAg+zIJC/fp21nnx5iFCKBty/UGJwMkrXbO+SQp8Pc1gfZ0+yBpyh3vRrWZImYWpvAfkdLiInXtUKqPfnFcsG36l5tU30hgCAAAAAAtxQ3Pfhqz/YRRfXNcCYDWolVucGJml0q6VJL69OmjysrK/Bbkoaqqqrg/37R3pql9yNLjlTuig77fTi7XNe/URLXHGvOC+2eprFi6+7iOzdblnNOCzX7t270oajk0WoZE7w4QD+8N0sW7g3Tx7iAdvDdIV6G/O+eNbBesv4/bvUtyu6LY/4+H7Cn0dwf50VLfm1yHhSsl9TOzYuecz8yKJe0VaLc0+6I45+6VdK8kTZw40VVUVHj+g+VLZWWlKioqpJdfiOo7YvIRKm9XHDx3zukX770Yds1Jxx2ha955JXh++xljNahHRx0waI/dF4WMXeuTkvl97loS/aczx+mU8f2S/GmQS8F3B0gB7w3SxbuDdPHuIB28N0hXQb47If+/9q2jJmrCwMD/yy1crwc/+0CS9NZPj1a/buX5qK7NKMh3B3nXUt+bnC5Dds6tlzRX0tmBprMlfeSc25BuX+6qLxx9u5RJkiIn9EXO8Kv8fxVROyR/a0L/8KAwDc654PGPHp+rwde+oLpGX0ZjAgAAAAASCwaFCv+WIUEhgFR4Fhaa2R1mtkpSf0mvm9muDyl8T9KVZrZI0pWBc2XY16a9OH9t2PmTlx+q288Yq7LS4jh3NBncs2Oz16TqNy9+riHXvRjVfvXj87L6HAAAAABAfMN7d853CQAKlGfLkJ1zP5D0gxjtCyRNinNPWn1t3eWPfBh2vle3cn1rQv+UxxnUo0PM9kcvmaRz7ns/qTHumbYkZvsL89fqrpQrAgAAAACko2P77E4MAdB25Ho3ZHjstauOSPveMybG3ly6fUn4a+L3715mvH57rRavr0r7mQAAAACAzJVHrBrb9b9te/dsfoNKAAhFWNjKDO+T+lTzW08dLUk6ccyeMftLi8Nfk6r6xuDxQb9+Q8fe/rYkaXttg5K1bONOTV+8MdVSAQAAAAAx+EImdUhS1/JSTf3WGD1yCQv0AKQm17shI8uWb9+9ccjrVx+Z1hjnThqkcycNitu/dltt2PlL89fqzAMHhrX5/U7Pz9v97cQFt5ygFZur1bdrmfa/8VVJ0tuLNqispEgHDemuitsqg9fefd4BOmF037RqBwAAAABIN3xzv6i2sw4aGONKAEiMsLDA3TB9d5A3tJc308u314TPGHxs1kqdeeBArd+++9l7/2z3pibjBnRTWWmxRkTMcvzO/bOC/aG+9/CcsPNlU6dkpW4AAAAAaO2OEi+yRAAAIABJREFUHNFLq7ZUJ5wAAgCpYBlyK2JmnozbuSw8U567cuv/Z+++w6Sosj6Of88Mw5CT5CxZQYIgiIGsEtY1YdY1ru7qGlddVMwJdY2rrq5Zd9eEaV8RMWIEERSMSHIERAUk5wnn/aOqm+7J0zB0z8zv8zw+dldVd98Z7lTfOnXvOTw9LYuXv/ip0OOLKpQS+/riPDtjcZnaJyIiIiIiUlUtWL6Bxas2JbsZIlKJKFgoJapVveAE1Kte/YZbJs8t9Phbjtwr7vnADruV6fPGvfRVmY5PRb9t2MrWnNySDxQREREREdkBP63ZTHaul3ygiEgpKVhYSbRtVPxsvoj/nBkkty3LJMSa1dNLPihG/uDiL+u2FHrc/JtG8e8zBvDt9YcU2JeTm1emz0wl3/+ynr43vk3X8W8kuykiIiIiIiIiImWiYGElccXobqU6rlHt6gB0aVr6qskNa2WU+ti92zYosG3PFvUKPTYjPY0DOjemVvVqZE0YE5er8NTHPyv1Z6YSd+eQuz9IdjNERERERERERBKiYGEFVyM9mFU4skeLUh0fmVGY66Wfpp5ZrXQzC/99xgBeOmf/Atv/cXyfAtvmXH1wse+1csPW0jUuxRx423txz7Mr8AxJEREREREREal6FCys4GpUM/bvVPqcgNtyguDVguUbSv2a6tVK100O6Ny40O1padvXPA/u0oSsCWOoX8RsxQlhvsO5v6wvdftSydLVm+Oed75ycpJaIiIiIiIild0jHy5KdhNEpBJSsLAC25Kdy5qtzvotOaV+Te3MIJ9gqwY1S/2a6ulBN6lfs/AAX4NaGXRoXLvY95h99UEc2Lkx/zxp72KPG9Nz+wzJOSVUTa4oLnj2C9qPm1Sh8zCKiIiIiEjquXHSd8lugohUQgoWVmDdrgoKaLz25c+lfk3TupkAnLZ/+1K/JlLgpLB8hFkTxvDFVQfxzl8HF/seDWpV5+kzBhRaWTlW3RrbA5KH3f9xdCZkRfbq7GUAjLn3oyS3REREREREKrLcPGf1xm20HzeJ9uMmRbfXKmNRShGR4hQfuZFKp26NjLhCIqVRIyOdKRcOok2jmux59ZQC+60spZXLqMv4yQzr1pTf9WzBkXu3LrfP2RliZw7u074hn2Wtjtu/YsNWTn70Uz6cvxKAeTeOKvUSbxERERERkY5XvF7o9lnjD9rFLRGRykyRikrgdz1LV9xkR3RtXrfEWYHl5d25y7n4+TlJ+eyymDhrKQBH7t2K584aWGD/qo3booFCgL43vhWdOXn5S19xx5vf75qGioiIiIhIpVJTMwtFZCdSsLACO2VgOwDuOa5gteFd4ahymul34oC2hW5PhRyGnyxcSftxk1i6elOBfeNe+gqAlz7/Ka6oS1HWb8mhy/jJtB83iWdmLOYf7y5g4YrSF54REREREZGq442vS59+SkRkRyhYWIFdd1gPnhhZm/RSBKbKw9+P7lku73vTEXsVuv2w+z8mL8/L5TNL64SHPwXggFvfA4jmChn34pd0aBIUeblsZNeE33/4He/veCNFRERERKTS+dO/P497/sMtoxk/Zg9mjh+RpBaJSGWlYKEkrDxzFRbl44UrSz5oF4lNKPzsZ0tYtGIjAKfvvzsAZw/uwPWHdS9zjsjVG7ftvEaKiIiIiEiFsGlbDsvXbynVscf0a42ZceaBHWhcJ7OcWyYiVY0KnEiFsiU79asj18gI8oVcPmqPUr9mSNcmTP1+BQCfZa3i4O7Ny6VtIiIiIiKSmmKLSTarl8mnV8TPGKxfM4O1m7OpXT2dW48qn1VeIiKgmYVSAQzt2oTbxwZfhn98ambS2uFe8hLojPTCZ1vec1zv6ON3/jqYeTeO4t7jt+eafOK0/rx0zn4AbM7O3cGWioiIiIhIRfbruq1xK5ncnbWbswG4+ci9krLKS0SqDgULJaXNuGI4j5yyD3u2rJfUduTk5rH75a8Xuu+5s/aNPp5/0+hCjzmsd6vo445N6lC9WhqH9mzB3cf2Zu4NI4Hty48veHb2zmq2iIiIiIgkwUfzV7L4t4JFEYvy0udLC91+ymMzWLs5mwE3vxPdNnqvFjvcPhGR4ihYKGVyzpCOu+Rz3rtkCDcc1p2m9WqQnmbs2WJ7sLD9uElszdl1s+9Wb9xGpysnF7rvDwPb0bB29VK/V//dG0UfmxmH92kVXbbcp23D6L71W7KLfI9IUZXvfl7HluxcNmzNKfXni4iIiIjIzjPm3g+5fcrcuG3uzkmPfsqg299jz6vf4MVZhQcCY138/JxCt78/bwW9rnuT5eu3RrdVS1KBSxGpOpSzUMrkpH3b8cDUheX+Obs3rs3ujWtHn+efZt91/BtlLhySqD43vFXkvusP6wHAQyf35YBOjYt9n5La2ygm6LjXtW8C8H9/OYC9Wtcv9PhR93wY97xDk9q8+9chxX6GiIiIiIjsHBu25vDNsnV8s2wd/XffjcFdmgDErUjatC2Xv74wh1Ubt9GzdX36796oxCXEc28YSber3ihyv5Ygi0h508xCKZPILLhUsWrjNn5as3mXfd5ZgzpEH48b1S36+JDuzamdufNj74fe91H0cU5uXlzekvwWrdjIz2t33e9CRERERKQq63HN9oIkpzw2o9hjb3r9O47913R2v/x1tuVsL9oYWTUUMe/GUUVec+3Zot4umzAhIlWbgoVSJpnVUqvL7H3DW+w/4V0+XrCyXN5/Wb5A5OWjunHd77sDcHZM4LA8TZy1lPbjJhW5FDrW9EW/7YIWiYiIiIhIfjm5eSxZVXKewi7jJzMzaxULV2wosK96eL31xoUHFtj3+gUFt4mIlIfUivxIykuFmYX7dgjy/sXmLTzxkU/L5bNe/uKnuOdmxin7tSdrwphdNv3/khcKz18CcEy/1mRNGMOFIzoDcNFzRR8rIiIiIiI7x/L1Wwps+2HlRjZnly63+tgHpzH8jveL3N+teT1mX31Qwu0TEdkRChZKmUSKjdx1bK9d/tlTLxnCxD8NZPqiVUCQtzDWN8vW4u479TNvn/J99PGCm0bt1PcuTGl+r92a1yVrwhh+uGU0t40Njj93aKdiX7N5Wy5fLF7Nyg1bd/rvSERERESkqul/0zsFtq3csI2D7/og+vzuY3tHryG6NqvLD7eMLvS93vnrYLImjCmwxLhBrers1SrIX/7BpUN3VtNFREqkAidSZsma/t6+cW3aN65NtTQjJ69gwGvMvUF+v5fO2Y+9YyoLl9UH81Zwy+S5TI75OT8ZN4xq6eUfWz+iT2suem4OR/ZpxZ3H9i6QozB2ABE7szEjpm3tx01izjUHU79mRnTbnte8QWyMULlORERERER2jtvG9uSyiV8y5ZtfotvevngwnZrWAWDhzaNJL6aCcccmdYrc9+ip/fhkwW+03a3WzmuwiEgJFCyUCqd7y3rMWbq2yP1HPvBJQsGwntdOYd2WnOjz2EBdywY1y/x+iSqq7Y+ftk+p36PXdUE15ZfP2Y8ereqTfzLh2k3ZCbevMJu35ZJZLY20YgZBIiIiIiKVwYLl6+OefxVemzzxSVZ0WyRQCMQFCkd2b84bMUHFomYbRjStW4PD+7TakeaKiJSZgoVS4azfmlPiMT+t2UyrMgb4YgOFqWho16Zlfs0RD3xS6PZhd0zljgMzCt1XVsvXb4kuw9CMRRERERGp7EbcuX2p8X0n9OHlz+PznF96SNciX3v/iXuzetM2GtfJLLf2iYjsKOUslArnhbMHRh9/Mm4Y88M8ID1b149uP/vpmTvt8+ZcffBOe6+ymjV+BAAPntS3xGP377RbsfvTDM4eHFRw/m3jtmKP/W3D1ujjLdm5tB83qcCS6Ihr//dN9PGmbfEB17w8J6+QJeMiIiIiUvG4e3QWXVX0t4lf8vT0H6PPm9erwZi9WnDGAbvHHXfWoA5Fvkd6milQKCIpTzMLpcLZLebLNbI8ODKjLRLQ+vqndWV6z/xBrlMGtmP91hzOPKAD9WvtnBl4iditTmapZ+s9ffoAct3pfOXkQvc/eXp/GtWuzkPvLwLgxumbabnHero0qxt3XP6gYGa1+HyIse1Zuzmb17/avoxiz6unkDVhDBu25tDjmilx7/PmRYPo3LTOLqsiLSIiIiI71+6Xvw7Acfu0YcJRPZPcmvJ137vz6duuEcc/PJ2FN4/mi8WreW7mEp6buSR6zPQrhgPQtF6N6LZJ5x8Ql09cRKQiUrBQKpUj+7TipS+CZQAnP/opT58xoFSvO+c/n0cfT798OM3r1yjm6NSUlmakUTAQ9+Kf96NzszrUq5ERFxRdsCaPg+/6gMdP24ehXZsyfdFvzF++ocDrt+bkFfmZkdyIsYqagRhbGU7LlUVEREQqrmc/W1LpgoWRMezRXTI49Y348WzHK14vcPy5QztGH3dqWodLDu7CsG7N2LNlvfJtqIjILqBbHlKp3HFMr+jjD+evZEMp8hsCfLts+0zEihgojNWhSe3o43k3jqJvu4bUqxHMjqxVveD9gdMe/4yvf1rLcf+azlWvfF3k+47q0RyAS16Yw9pN2Tu0vPjm179L+LUiIiIismNe+3IZObl5fLtsHT/+tjGh99iSnbuTW5U8y9Zsjj5+YV7pCgFeeki3uOd/GdZZgUIRqTQ0s1AqpJfO2Y9a1dMLbM+/xLXHNVOKncX2/S/raVI3k+Xrgxx9Uy4ctHMbmgTPnrUv5/7ncx44sS/VqxW8H9C9ZT2+WRa/TPt3//iowHF1Mqtx3D5tyHO4+tA9ufvteUz++hcmzlrKxFlL445ddPPoIpdAL7hpFP94dwHnDO1I1/FvAPCvDxZxxeg9duTHFBEREZEEnPnkTN7+7te4bdMuH0aL+sUXB8y/euSQuz/g/UuH7vT2JcN+E94tcl+dzGrRCQjPnrUv+3YoPk+4iEhloGChVEh7t21Y5L6sCWM4/l/TmbboNwDe/vZXRuzZjJ/XbubvU+bFzT485O4P4l7bpG7FTzbctG4NXvjTfkXun3T+gUUuFYYgENujZf0CgcY9WhR+p/T9S4dEl0Dff8LefP/LOi4+OL4C3EUHdQHgh1tGR3PdRNpw6SFd+cPAdtStkbzckCIiIiIV3ZbsXNZtyaZp3eJXyeQPFAIMvCUIlj1x2j4M6dq0wP6NhazW+fG3TQm2tPzd8Nq3PPrRDwDcNrYnx/RrU+b3iA2gvjv3V9ZtzlGgUESqDC1DlkrpnuN7Rx+f+dRMznvmCwbe8i4vfr6U05/4DIDnPltc4HWNalffZW1MppnjR9ClYRofFHI3eO+2DQudkXhI9+bcfMReDOrSBICaGenMGj+CdrttX/Y8pmeLAoHCWIUVN7l9yvfsde2btB83iYPvej+RH0dERESkyvvDYzPof9M7fP3TWiZMnkv7cZP4Ze0WVm7YyvL1W9i8reRlw6c+/hlzlqyh/bhJ0Ru7uXlO95jCdQft2Sz62D3xtDTlxd2jgUKAyyZ+Gf153vt+eXT7tf/7hsW/bYq7if6Hge14YmRtsiaMiZtpOaxbMw7v02rX/AAiIilAMwulUsp/R/X/5iyLPn537nKen7mEv734Vdwxr59/4C5pWypoXCeTKwbUpO1utciaMIaJs5ZyyQtzuGB452Jfd8KAtpwwoO0OffbM8SMYcvvUQvNJzvu1YIEVERERESnZjB9WAfHpZfa95Z0ij//wsqFUSzeG/n0qW7K3F7Q77P6Po4/zr0Z55dz96d2mQXT77pe/XiDlT16e0+GK1/n70b04sk8r5v6yntH3fgjAXw/qwnkljDdLo8/1b7J6UzZPn9GfAzo1jt6Q3pKdS7er3ijydac9/lnc8yc+yYp7fv1hPZg6deoOt09EpKJTsFCqpMsmfhn3/Okz+lfphMRj+7ZmbN/Wu+SzGtfJ5OvrDuHZGYtpWi+T5vVq0m63WtE71hu35lA7U6cmERERkfLUplEtAL67fiR3vTWPe99dUOzxp+7Xnt5tGhTY7u6YGd/9vI5R93wY3X7JC3O45IU5ccfe8dY87nhrXrE5xXNy86iWXvgCuPzBy5MfnQFAqwY1+SmmSEnED7eM5vQnPuO971cU/YPFHCsiIgFdkUuV9sq5+9OjZb0iByRSfo7rX/gMxUjQsLhBpIhUPR8vWMkH81dw+SgVRxIRyS8vr+By4Pk3jYoWn/t9r5b8L2alzbfXHxJ9bGZcfHBXzjiwA72ue5NLD+lK4zrV41bhfHPdIUXezN398td5/9IhcYHCkixZtYk2jWqxdPUmLnh2NrN+XF3gmKwJY9i8LZc8d2pnVotbQpxfYYHCyFjy0VP2Yco3v/DCrKW8O7fw97jhsO6FpssREamqFCyUSuvNiwZx8F0fMGKPZrz93a/ccXQvGtTK4IwnZwJw1qAOhd4dFRGR1HPiI58CKFgoIpJPZNlvxF+GduKSQ4Ic0jcc1p2BHXejU9O6nDWoA5dO/JL//WV/Mgq5UV6/ZkbczdruLevTrXldsnOdmtXT446dd+MouoyfHH0++PapxbaxQ+PavHvJkOjMwANve6/En6u4gnwAD560N73aNIgWZ4FgtdDJj87gzpiChmlpxqi9WjBqrxYF3uP7X9bTuWkd0tIUKBQRiaVgoVRaXZrVjQ54tubkklktGORoxlpq+mTcMPabsH2w9/zMJRzdt7Xu8ooIW3O2J+X/LGsV+7RvlMTWiIgk34atOVz9ytecPbgjh9z9Qdy+SKAQ4OSB7aOPe7Sqz+QLSp+ju0er+gBUSy+4r7BieBHv/HUw1dPT+GThSvZp34hVG7fRLzxvX39Yd65+9ZtSt6Ew+cfyJT0vTtfmdXeoLSIilZXWXkqVkFnYKEdSSssGNcmaMIa6NYJ7GJdN/JLdL3+d1Ru3AXD/ewtKvMMsIpXLNa9+Tftxk+g6fnuy+gufnR1XpTO/yV/9XKDipYhIZeLu9LhmCi998VOBQOH1h3XfZe148vT+Bba9+OeBdGxShzaNanHsPm3p0KRONFAIcEQhFYUfOrkvWRPGRP+bNX4EALeP7cn3N44kI337jePplw8vh59ERETy08xCEUkpV47eg3Evbc+Rc/Hzs+OSUrcfN4nLR3XjmH5taFi7etxrc/Oczdm51MmsxvJ1W7j4+Tk8duo+xd79FpHUdMhdH/D9r+sLbI/NSzV90W88NS2LzdtyeezUfdj98u3L8E57/DPevGgQXZpp1oiIVC6x57pYu3r1TNeY8+sDJ+5N20a1orMRi1K3RrDU+dXZP3HBs7MBOKR787hjdquTGfezzL9JhUdERHY1BQtFJKUc179tXLCwsOp1t0yeyy2T53LOkI6cM7QTGenGtpw89rr2zQLHRvLptG5Yk4/+Nqz8Gi4i5OU5Ha98nYl/GkjfdokvFXb3QgOF+R33r+nRx4VdPI9/+Wue/9PAhNshkgpmZq3i4Q8X8eBJfcnOdW6a9C2n7NeeDk3qJLtpkgTrt2QX2PbuXwfTfrfau7wtzevX4M2LBrF749qF5kAszmG9WzG0W1Pq1cgop9aJiMiOULBQRFJO1oQx/PjbxhKTZT8wdSEPTF1YqvdcunozObl5qnwtUkbXvPo1T077ke+uH1kgwX3EIx8u4sZJ30WfH/XPadHH828aVeJF5MRZS7nkhTmF7nv13P1pUjeTlg1qljoVwfH92/LMjMXMyFpF+3GTOLpva24/ulfJLxRJIZu35bLH1duX4McGxJ+c9iPzbhzFZRPn8MrsZZw1qANXjFbxn8ru88WrOfKBT6LPUyEP947M3lagUEQkdSlYKCIpqV2+O+QHdm7M8f3bYsCf//N5ka+rXT2djduCYggPnLg3HZrUZuTdHwKwcVsu9WuWPli4JTuXNDMtY5Yq7clpPwLEBS0iZlwxnKem/ch97y0o8vWdr5wcvaDdlpMX/XvKX72zMPcc15teMVXrzx7UgZw856rf7cnfJn7JczOXAHD/CXuzdPUmurWox4GdGmMGz8xYHH3dC7OW0qphTS4c0aWUP7VI8hX2NxcrthLtvz5YxGtzlvGJ8rlVarGBwmsO3TOJLRERkcpOwUIRSXn92zfi6TMGRJ+/dt4BTJy1lCc+yeKVc/enad1MslZuZL9OjQGKnEF40J3vM+PKEaX6THen21XbL9T++8cB7Nexcdx+QNWapVKL9POi9L/5nVK9T3RG4BuTiz8wn8N6xyfCvzxm5tStY3sy4ai9gML/Drs2qxu3lPnXdVvK9NlFyctz0tL0dy+pZ9naLXEB+Vixs/VTYTZaqliwfAPnP/MF3/68rsRjHz2lH4O7NClyhcJh93/MnCVrgJJ/x+7OTZO+45GPfohua1wnk5UbtgLBuOfhU/pRv2YG7s7S1Zs58Lb3ose+dt4BJeYGFBER2REKFopIyrrj6F784935BXKO9WhVnx6t6nPt77dX/GvZoGb0cf6B/B8GtuOpaT+yfP1W/vPpj5w4oF2Bz/pkwUp6tWlAzYx00tKMBcs3xO0/4eFPWXTz6EJnQt15TC/6tWvEoNvf4/JR3bhl8lwA/jykI38b2Q2AjVtzyHOn7k5YcuPuClLKLlFUEv3C3HB4D07et100MDhr/AjOe+YLPln4W4mvrZ6exrybRpW5fcX9HUy5aBAQ5Pfa69o3eWbGEp6ZsYTPrzqIu9+ex1PhjMkX/jSQfu0asjUnjxoZhS+zdnc2bcul+zVTotsePKkvI3s0L/T4XU3nhMrj9ilzuf+9hTSotf27YuHNo5mZtYpjwxydH1w6lEG3B4GjOdccTN3MatHvptjZhkV5fuYSjunXphxav3MtW7OZNDOa169RLu+fk5vHiDvfL/XxZzw5s9THRs6DI/Zoytvfla4yeyRQCDAjaxW9riuYhxmgY5PaChSKiEi5s5JmDVR0/fr185kzS//lXtFMnTqVIUOGJLsZUgFVpb7j7gWCHpG7/k9Ny+LqV78pt88+f3hnLj6oS1yutcFdmvD4qfuQkxecf0u7zPnB9xfy9LQf46rBxnr8tH0Y0qVJuQYNqlK/qeq25eQVGnj49xkD2K/jbtHgRP4ZNGs3ZZOdl0fjOpm4O8f+azozfliFAR9cNjRudsz//rI/PVs3oLyVNtdhRMNaGdSqXq3Iv7XCPHHaPgwu57+/WEX9+8y55mDq16xcecCqwnln2sLfOP7h6QW2j9ijGY+c0g935+UvfuKgPZsVetNp7i/roik3SiPVZxfm5ObR6cr4/v3fPw5gj+b1WLZ2M2Pu/QiAZ/64LwM77lbg9bF/8y+dsx9dm9Ul152aGelkpKfx89rNDLzl3VK15dyhHbn/vdLlRy5vu+qcWdVVhXOOlA/1HUlEMvuNmc1y936F7lOwsGLTCUkSVdX6TmHBgo/+NpQDbn2vkKO3++76kdw2ZS6Pf5xVYN8fD9ydhz/8oeCLkqw8LwKrWr+pilZu2Eq/G9+O2/bmRYP4YeVGDume+Ey6SN/Jyc3jpzWbC+QlLU9lDRbuiAU3jaJaehqH3/8xs2OWJG7NyaV6etpOCSaWlO/xqdP7M6hLk1K/X26es2lbDnUyq5Wqfdm5eazeuI2m9cpnxld+lf28syU7Ny7tRaxIfyqN4vq5GSy8qeDs+FQKGq7dnM2rs38q8w28rAljePvbXznzqcTG++/8dTAdGtdm7eZsGtSqXnjbNmWDQf2aGdHf89wbRrJ41Sb+++livlm2lhsO70G35vXYvC2XjdtyCpxHm9TNZMX67bMHnzy9P4PL8Hcqu05lP+dI+VHfkUSkarBQy5BFpEqYeskQdqtTnb2u3b6sp6RAYc/W9alZPZ2rf7dnNFh4/vDOnDu0I2lmZKSnceWYwhOMz/pxVVxF2J3ttrE9uWzil4XuW7RiAx2a1Cm3z5Zd75z/zGJgx8Z0bFKbPVvUK/KCNlFZKzfSvnEQvMt/gQtBtcsdqXgZq1p62i4NFEKwZPjSiXNYvyUHgD1b1GPS+QdgZuw/4d3oDMLxY/aIq+oM0KVZHV4//8ACAZsLn/2CV2YvK/BZ+WdDQdFBnLuO7cURfVoDwQzo+cs30LZRLWpkpJOX5zw9/Ufm/rKewV2acEj3ZqzelM3eN7xV4H36796In1Zvjv4cf3hsRlwQaNO2HJ77bAnH929LZrXtActXZ//EBc/OLrRt/ds3KpACIqJzIT9jIvq3b8SMrFUA/HDL6CqzlPrMJz8rcmnqnKsPpn6txGaGTr98OOu2ZNOpSR225OSydPXmEv9uz3vmC/5xfJ+EPi9RsbP9J51/AP+cupDXvvw54ffbkZsBsX8nxZ1XY/9NYl/TpVnduJQoADWrp1OzenpKBWJFRETKSjMLKzjdvZBEVdW+0+HySeQVctrLmjAmOrvjiD6tuOHwHtTJ3LH7KV3HT2ZrTl70+ZCuTXjitP5AMJMkkh8tN89Js+D/6WmGmRW4+Bk/Zg+25eZxzpBORX5eYRdMFwzvzEUH7bwKsFWp30T+PXa1DVtz6BHmxvtk3DD2m1Bwqdy8G0dRLc246+15HNe/LfsXcgzARSO6cMGIzgW2Z+fm0fnKyZy8bzuenv5jse3ZWRe8ye47W3Ny6Tr+DQZ1acJTp/cv9ti1m7PLvJT3+H9NZ9qikvMz7kyXHNyFvwyL//ct71mU40Z1Y0KYl7U8HLV3a+44plfctmT3neLc+858+rZryImPfBrdNrZvaybOWsq9x/fh971aAjDqng/5roQiGotuHl3uxXMWLF/PiDs/iNu2M4Na+ftf7HsXlhKkOH89qAvnDQ/699pN2Tz68Q9cfFAX5v+6nk5N6xT7XgtvHs2HH7zPkCFDaD9uEmccsDtXjt6DlRu3krVyE12b1U04KCuVXyqfcyS1qe9IIlJ1ZqGChRWcTkiSqKrcd/LynPVbcuh1fTDLcEdmcqSSooIEO/NCsKR+M+WbX2i/W226Ni84m2VLdi6/v+8j5v26vXjM8f3bcu7QjrRqUDOlZhV1GT+ZbTl5An+oAAAgAElEQVR5jN6rOQ+c2HeXfvauXDJbnDcuPJBuzevttPer7Oec/DnW7jmuN4f1bsX/5izjrrfm8d4lQwD4ZtlaamakM+yO0hdWyG/a5cNoUb9mofsS7T/dmtdl7i/rSz6wlIZ3a8qDJ/flq5/WcuxD09i/U2NuPaonj3y4iLF923DI3R8U+drDerfk2kO706BWBlm/beLEB99n2Qand5sGPPPHfalZPb1AVepkFHm5/v++5bGPdywVxUd/G0qaWVyRrl0hkvPytP3bc82h3QvsT+T3uXT1phJn7BfnsytH0KRuZqmPX7F+K/94dz7TFv7GK+fuT+18N/gq+zlHyo/6jiRKfUcSkarBQi1DFpEqJy3NqF8rg8kXHMiv67ZUikAhBEHBn9dupl6NDGpnVosGDdqPm8SMK4fTtG755he76pWvo7PUsiaMwd3pc8NbrNmUzRsXHlho8v1nZizmmRmLAXj74kF0arpzlrqWVVEBlte/+oXnZy5h3eZsxvRsQeM6mWSkp5GX59z33gKmL/qNiw/qwk9rNjNmrxZxS1Vj3/OIPq144+tfOH94Z0bv1Zz6NTOoXzOD1ZuyufylL5nyza8ltjFrwhjmLFnDYfd/XOj+c4Z05Pzhnelz/Vtszs4t429gu8dO7cewbs0Sfn1VVS09rdDA/O97tYzOLgPo3jKoYvrfMwdwQsxstD8P6cilB3cFYMGKDZz86Ke8ffHgHa6g3qxeJr+u28pnV46gcZ3q/LRmMy3r12TO0jX8snYLo/ZqUeRr123JZt3mbD6av5Kj+7Wh13VvsmFrsJT75H3bccPhPUr8/L3bNmT+TaOjzyOpGwr7XUX+Zl6dvYxXC1niPXvJGva4uvDcfsUpbdGXwmYTl5QfsizS04xZ40dQr0ZGuc8gLEmksNbjH2fxt5Hd4iqB73vzO/yybgunDGzHtb/vzjEPTeOzrNXcetRetGlYi/06NY4e++yMxYx76asyf/6s8SPIzEjntMdn8JdhnRPK3dekbibXH1ZyHxQREZGy08zCCk53LyRR6juVX2EBsLuP7c3hfVoB8M53v7J/p8ZxF4klKarf7MwL6gM7N+bD+SuBYKbJluxc7njze+46tvcOzxxasmoTLerX4P++XEa/do3Ic2fw7VN3Qqt3vjuO7kWvNg3o1LTo/JN5eY5Dkcull6zaFK0+fNnIrtz2xvd8esVwmu2iwhSxdM7ZNdZuzmbj1hx+27CNLs3rkFmt9H/fpZF/Rt/Olioza8uiuNnbqzZuo1HtnZtjdGfJ/7v+9xkDOKBz4x3+N1h482jS04yZWauCIOMbwbL1+07ow1/++8UuWW4NOudI4tR3JFHqO5IIzSwUEZFd6ourDqJPvmIIFz43mwufK7ygQcT3N44sdYAh/9LLksTmoIpYtXFbXNGGSKAQYJ+bthfbKKyYRMSim0fzn09/5Pj+bcn6bRNrNm2jR6v6ZFZLY2tOXpGVRmP998wBNKpTnc5N65KeVjBvZHkb2b05XyxZzQGdmnD72J6lupgu6Zg2jWrFBTKKy3kplUNk1mp5LWst7yDP+5cOoX7NDAbe8i6bs3OZceVw5nw2jYOGDS3V67/+aS01q6fTuE4m/57+I7dP+X6ntCv2JgYElXDTzKIz9IqSqoHCwpz06KclH1SEOpnVuPOYXuzXqXH05kW/9o3o174Rfx7SMXrc73q2LOotREREJIUoWCgiUkk1rF2dF/88kOxc57h/TS/167qODwJrt4/tyZJVmzjjwA4FlvCNvPuDQvObHd+/Dc/MWBK3raSciY1qV+fswR146P1FpW5jfpFZjVe9+k1Cr3/wpL5xS+sgvt3bcvKiQYEvFq/miAc+KXRpdzLypolUJpFK2d/dMDK6LaMMAcoerepHH587tBN/Htwxen44um9rOjSpw4g9mrJ09WZOe+Kz6LE3Ht6Do/u1JrNakA9x5o+r6d6yXoE8eJVJ7zYNmL1kTaH7Gtepzqn7tefvb87j9P135+pDg+XjC5av59uf1/PNsrVcPmqPXdlcERER2YUq7whIRETo264REF/tOdas8SO48uWvueOYXnQPK/BGXDrxSwDufXdB/Ju+UfSMu1uO7MktR/YEghmDDUuZD/LyUXtELzyLC7itWL81LgH+yY9+Gjfbpzizrz6IBrUSm+UTO3uoT9uGRQZAFSgUSS1paVbo32vnZnWL/DtOSzP6796ovJuWdJ2b1mH2kjXcNrYnD3+wiPnLN3DtoXtyQOcm0fQH+attd2pal05N68bl4RQREZHKR8FCEZEqokZGeqEXxw+eHFT7jewr6/LbovJPJbr8rriAW/5KmU+fMSChzxARqerGjepGbp7zu54tOKZfm2Q3R0RERFKIgoUiIhIna8IY3vr2V/q2a0jDWhnR4F12bh7X/+cdpq+sztNnDKBZvUzNpBMRqaB2q5PJncf2TnYzREREJAUVn5U5BZlZFzObZmbzwv93LvlVIiJSFgft2YxGtavHBQMz0tMY3jaDty4eTPP6NRQoFBERERERqYQqXLAQeBC43927APcDDyW5PSIiIiIiIiIiIpVChQoWmllTYG/gmXDTM8DeZtYkea0SERERERERERGpHCpUsBBoA/zk7rkA4f+XhdtFRERERERERERkB5i7J7sNpWZmfYGn3L17zLZvgZPc/fOYbWcBZwE0a9as77PPPrvL27qrbNiwgTp16iS7GVIBqe9IItRvJFHqO5Io9R1JhPqNJEp9RxKlviOJSGa/GTp06Cx371fYvooWLGwKzAN2c/dcM0sHfgM6u/uKwl7Tr18/nzlz5q5s5i41depUhgwZkuxmSAWkviOJUL+RRKnvSKLUdyQR6jeSKPUdSZT6jiQimf3GzIoMFlaoZcjuvhyYDRwfbjoe+KKoQKGIiIiIiIiIiIiUXrVkNyABfwKeNLOrgdXAH5LcHhERERERERERkUqhwgUL3X0uMCDZ7RAREREREREREalsKtQyZBERERERERERESk/ChaKiIiIiIiIiIgIoGChiIiIiIiIiIiIhBQsFBEREREREREREQDM3ZPdhnJlZiuAH5PdjnLUGFiZ7EZIhaS+I4lQv5FEqe9IotR3JBHqN5Io9R1JlPqOJCKZ/aaduzcpbEelDxZWdmY20937JbsdUvGo70gi1G8kUeo7kij1HUmE+o0kSn1HEqW+I4lI1X6jZcgiIiIiIiIiIiICKFgoIiIiIiIiIiIiIQULK75/JbsBUmGp70gi1G8kUeo7kij1HUmE+o0kSn1HEqW+I4lIyX6jnIUiIiIiIiIiIiICaGahiIiIiIiIiIiIhBQsFBEREREREREREUDBQtmJzKy5mTVOdjuk4jGzTDOrkex2SMWifiOJMrP6ZlYv2e2Qikd9RxJhgfRkt0MqHp1zJFEaJ8uOUrBQdgozOwp4EXjdzCaY2THJbpNUDGZ2OPAs8I6ZnWxmnZPdJkl96jeSKDM7EpgIvGdmF5nZoGS3SSoG9R1JhJkdCjwJvGhmY8ysdrLbJBWDzjmSKI2TZWdQgRPZYWbWFXgVOBKoARwIDANed/eHktk2SW1mti/BAPp0oC0wFlgAvOTunyazbZK61G8kUWbWG3gOOBZoBhwMNAD+5+6vJrNtktrUdyQRZjYYeAi4FGgDnAn8F5jo7llJbJqkOJ1zJFEaJ8vOUi3ZDZBKoSGw3N2/BTCzxcAK4BgzW+7uLye1dZLKOgDT3f1j4GMzm0swkD7CzH7VQFqKoH4jiWoBzHf32QBmNh84Bjg0/L6altTWSSpT35FE7AW85u7/B2Bm3wIXA25m9wNbXTM3pHA650iiNE6WnULLkGVnmAusMLNDzSzd3VcCbwGfAHsnt2mS4pYDNcysPoC7fwE8AnQHBiSzYZLS1G8kUVnAVjPraWbm7ouAF4Ct6PtKipeF+o6UkplZ+HAD0DKy3d2nAncCpwCDFCiUYmShc44kRuNk2SkULJSEmFnr8L80d18DfAUcDvQNA4YrgA+AIZETlQiAmdUyszrh0+lAK+DWyP7wC+1F4BQzy0hCEyUFqd9IosysUUzxrYXARuA8oEn4HbYQmAIcbWa1ktVOST3qO5IIM6vG9tVbrwIHmtntkf1hwPBx4ISYoKKIzjmSMI2TpTwoWChlFhYzeQl4BbjXzI4GbgC2AWcR5C4EaB5u011TAcDMjgBeJkjUfDFQG/gdMMLMHjSzpuGhecCa8P9SxanfSKLC76tXCRJ8Xw/0Bv4EtANuBPqGh2YC61DfkZD6jiTCzH5PkGfuf2Z2hruvBoYSpOb5e8yhOcAWzSyUCJ1zJFEaJ0t5UYETKRMz60jwRTaWYBr8QcAYYBLwKHAZ0B9oQpCE9+TwToZUcWa2D/A0wdKbegS5M1YA9wG/EASgVxHcxOgKnBjJ0yJVl/qNJMrM9mJ7cvhM4CigNfBv4EPgXmA3oA7BMsGT9H0loL4jiTGzAwiW+l1I8H11GTAZeACoCbxHMONnM7APcLy7f5mc1koq0TlHEqVxspQnFTiRsqpDkH9lnrvnmdlzBHkRTiUocnKLmdUGOgG/uPuvyWuqpJjmwBeRKlxmlgWcA5wLXAP8HuhBUPHtS3f/IUntlNSifiOJagAsdfevAMxsGUFy+BMJvrfOIbjoagcsdPelyWqopBz1HUlER+Atd38DwMzmAVcDf3b3q82sF8EswwbATe4+P3lNlRSjc44kSuNkKTdahixlNZ+g9PpYM6vu7muBqcC7wBAAd9/o7nMUKJR8fgMyzawJQDhIfoDgzulJ7r7B3ae7+6v6IpMY6jeSqKXAGjPbJ8z1tAx4nqBP/c7dt7l7lru/rwsvyUd9RxKxiSCYA0A4e+d64Pdmdoq7r3H3l939cQUKJR+dcyRRGidLuVGwUMrE3TcB3wGHAvuaWWZY4OQNYEBMTgSR/D4lmJl6f2RD+IX2CMFAukayGiYpTf1GEpUFrAUuAlqFxbciF2CjzKxeMhsnKS0L9R0puxeBTmb2RGRDGDC8AxhpZrrukqJkoXOOJEbjZCk3+tKSUsk3wJlAkBz1DODYsKLSHkAuQUETEQAskA7g7rkEyXb3NLPnzKx1eFgDgqXtuUlqpqQg9RtJVEzfcYKlOHUJvrf2Cw9pQVBhMicpDZSUpb4jiYiMkd09j2CZ8UAzezLmkDrou0oKoXOOJErjZNkVVOBESs3MmhFMZ74jPEFdDPQiyNFSAzjD3T9PZhslNZhZN4IclqvC562AA9392TC4/DJBYDkDaE9QCEfJdiWO+o0kKuw7R7j7fWZWDbiHoPBWE6ARcKqSwwuAmdV2940xz1sCR6rvSGmEM8Byzaw50MfdJ5tZY4JiJlkEVWt7o2ImEgq/n9ZEzjs650hphUvU82Kea5ws5UoFTqRQZjYI2JegYttrYY6DOgRVlSJ3MW4PT0wdgFXuviJZ7ZXUYWYHAw8CfwammJkBnQmquOHu2Wb2e6BnuG2Bu/+YrPZKajCzfYFuBOeYT919Neo3UgpmNpAgefdS4Kswn1NDIA/A3XPM7C8EScBbAcvC5V1SxZnZocAYM7vI3TeHm+uhviPFMLODgIHufn0YKKwGDCToO7j7yrCYycEEY+dr3H1B8losqSIcx5wJXAF8HW7WOUdKFPadQWZ2qW+f7dUZaAwaJ0v50MxCKcDMxgC3AK8T3NVqAJzt7ivzHZfp7luT0ERJUWZ2CHAzQYqDRe5+VJKbJBWAmY0E/g58COwJ3O3uLye3VVIRhMGe2wiKbNUE+hLMgP8qqQ2TlBcGfCYAV7j7lGS3R1JfePOzOkGOsJbAE+5+WbgvboaqSH5mNpSg8MSf3X1qkpsjFUjM99WVkWrrIruCZhZKHDPrQ3DRfpK7zzKz3gQnp5r5jjsU6GJmd4ezDKWKM7MRwF0ES23mmNmHZjba3V/Pd9wJQI67P5+UhkpKMbMhBEtuTnP3T8zsHqBFWCxpW1hASf1GCjCz3QhmMB8XnnP2BCYRzGg+2N2/Do87Hmjk7vcX83ZShYSrJ6YAe7v77DDNSjeC3E4L3P2X8Dj1HYnj7lvN7BVgOTDMzP7h7ucRVEIGwMwOB7bool7y6Qvc4+5TzawNwczT6sBUd/8OdM6RgsxsMPA/oK+7fxtWPW5NcM75VeNkKU8KFkp+64G73H0WBFXczCyTIDfhkpjj5gLfKVAo4Z32OgTJmM8ML9rrA4uAfgQzVGP9Ev4nVVyYFD6doN98EubtORaYBRwApJvZxe7+E+o3UtBmoDbQDpgTDqInEizfusfMjnT3tYDy9Uh+C4FsYIiZLQBeAn4lyL+MmZ3t7ktQ35EYMUv/NgJtCKocX2VmLwG7hTdNMwiurxYmp5WSaszMwr7TDvgtHDc/A3xJcM650cwGhze45gBa9iexsgnyEPY0s/kEVdfXEfSd783smnD1n8bJstNpGbIUYGb13H2dmWWE+Q/+j2BZ4DvhMtPv3H1xstspqcXM6rv72phk30MIvtBGu/unSW6epLgwcHgJgLvfFuZ7ugZ4SEsEJSJy0RXmCMsDLgW6EwSYWwGdgPEEs5zPDAM+IlFmVt3dt5lZO4KcYTUIUq08ZmadgZuA/7j7q0ltqKSscDnpwe5+uZn9GbgV+MDdfxfuN9cFlhDfF8ysH0GqnmXAR+7+SLj978Bad78heS2VVGZmBxBcU9UCLnb3h81sGEGx0evdfUZSGyiVVlqyGyDJZ2ajzOxhM/tHuHRrXbgrMtDZBPxsZocRfMmlJ6WhknLMbGR4R50wUFgtDBRamI/lGaB/eKz6jQBgZvub2XlmdoSZtQAIq7vd6e63hc/nsH3mmEhEUwiSwId95lWCXJe9w/1j3f1bgu+v3ZLTRElFZtYaIAwUVg8Tv+8BjHP3x8J98wlmcNRIXksllZjZPmb2BzMbbEHFY4DvgNphMYFLCQLM9cxsAsTNQJQqLJxg8aGZRVI5LSbIr7sPQc7LiC27um2S2sxsiJldY2Z/MbNu7v4RcBRwlbs/DODu7xLMOmyYzLZK5aZgYRUXLpm4G5hGsGz0GTM7x8xquXtOeNhq4E6C2Rp/CCsjSxUWLqEAOBU43MwmQbSKW/WYgfK3wF/DbVqyLpFiJg8BuwMXAcMi+2LOOZjZMQTFTmbt6jZKagpvWC01s3Mj29x9rrs/7O6nuftl7p5nZicSLBH8OWmNlZQS5pBbbGbXQFzAcKm73xFz3NHAXoBmaQhmNhp4iuB76m/AiWaWARjQB/gncL6730oQNHwgWW2V1BEzRh5OkKJnqpnVdPflwCsEN7mGmdlFZnYyMBJ4ITmtlVQTBpnvIVg9MRw4DCAMGP4j5rhjgA4E11oi5ULLkKs4M7sEqObuE8Lnw4HbgX+5+4PhtikEuecGuPuCpDVWUkbMUsDjCYrfDAPauPvgyPL18Lg04FngEi1dl3AZxeMExUw+MrObgLUEfSTb3X82s3rAkQRLko9192+S12JJFWbWg+AC63ngBOBWd38g3FctEmg2szOBK4DfRwqcSNVmZl2A/wITCQI+d7v7deG+yHdZOkG/uoJgdqrOO1WcmQ0AngROcfdPzewPwAXAIHffGAaWV7v72+HxWnosccIxT1tgFME1VJcw2Nw03H4GQa74x9z9q+S1VFKFmQ0E/k1w3vnIzC4mmIX6b2CDuy8ws+rAcQTfZ8fo+0rKk4KFVZyZ/Q3o4+7HxWwbTnCH61h3fyvMzfJruLRLJMrMjgPGAOcDTxMMfmoRLAnMcfctZpbp7luT2ExJEeHFV3pYzKQ18DkwHdhA0G8uBVYAxwPvufvcpDVWUoqZ1QWOcPenwtk+jxHk6Xkg33FNgPq6sSUR4cX5WHd/JsyFOo0g2HxdvuPGEFRC/j4Z7ZTUEo59m7n7szHbpgJ/iv1uCgPNeQoUSn5mNgo4w93HmtnLBMuPtwL7ufuvkRmI6jsSYWbtgZYx4+TpBDPdlwP7EgSYvyK4uTVd42QpbwoWVnFm1pDgJPQfd782Zvs4oLa7X5WstknqCxPEX+XuZ4bLS58DFrv7XklumqQgM0sLl4kacDZQx93/bmZdgRuA/7r7K7EzxURi+k1amKcwchH2OHCDu98f3uRaHOacEwHizzkxRQZ6ElyA3eru14VLlL9TkFBihTcoarj7CtteUGkGwcz4rywoVrHE3X9NakMlZYW5Cu9399PD886HBIVM2ia5aVIBmNkJQCt3v93M6hBUX5/t7v/UOFl2FeUsrKLC5aG4+2qCvHMHmdn1MYdkAvWT0DRJcTG5WACWhtvOBe4DzgFWmtn/ktE2SW2RQE940f4vd/97+Px7gtmFdcPnGgBJLIft/ScM/EwGTgMuC2ds3E1QmEIkVqTvRAKFGe7+JcEMjYvM7C2CnMx5yWuipJrwHLPe3VeEm9IIivvlAj+Z2VjgFoLchSJR+cbIW8JtNxKkQRgLzDOz6ZHrMJGIfH0Hd/+vu98ePt5AMCs1I3yucbLsEtWS3QDZ9cJBUF64XKuNu38c5kT4t5ntAfwEHAwcndSGSsqJmaVRD2hEUNktDRgHnOvu/zOz11CgWWLE5AWrBVR39zWRwE+4/1hgb4LZhSJRkbvn4Tmnrrv/FAn8uPvk8HxzNDDcg+q2IkDc91VDoKm7f+/u2WGf+tLM7gP+BAzRjFSJyDfOiZxztoX75gNXAwOBM939l2S2VVKLmaW7e254zmnp7t+Y2SLgdOCv7v4W8JaZtY0dA4kUNtbJt/9Y4EDg3qQ0UKos3dWoIsysg5n1Ce9aRO5cjCWoSIq7fwr0Bd4B5gFHKmGqAIT9picEM3vCgM84oEU42PkXQYLd/4VBobUqZiJm1jOm37iZ1Qb+DrSOOaaumZ1CcPF1oqvSugBm1tjMGkG0wnoGcDlBTtTY4w4BDicIFCo5vGBmbcObnpHvqxrAH4npO2GfGgT8HhjmKoRT5ZlZRzPbJyZQWJtgnNMm3B8ZN+9LMHY+SeccATCz5mFuOdw9N9x8NtAqfPwycJS7vxx+l6ExskDpxjrhOPkPwLUE5x3lY5ZdSjkLqwAzOwK4jSA56hrgTeAJd1+b1IZJyjOzlsASgoDgU+4+LdzeNX9+JyVqlogwn9zNwAnu/l3M9t7uPjvmeX3gCOATd5+361sqqcbMDgVuBH4lqBJ5lbt/a2Zt3H1JvmNrENy0UJBZYs876whyLvcLt7cqZJZGdaBJ/u1S9VhQMOk2glQY5u4Dwu2d8884DVfhvK6iAgLRokg3A78RLDm+F5hCUGRrTb5jNUaWqNKOdcwsExhGUHxLM+Bll9PMwkrOzNoS3KU40t33B14C9gT+ZmaNY447yMz6J6mZkro2At8R5LA80swOCLdHAztm1svMWnsoGY2U1BHO9poAXODu34WJ4QFw99kWVI7EzDoT5Jh7UoFCgeBcAtxOkPv0SILzz3VmNoJgQB09zszau/sWBQoFwMwOJjjv/MXdBwPZZtYDIDYgaGb7mVkPd9+mQKGY2X4E55wz3H1fYJOZdQKIvTA3s97hmPluBQoFwMz6EuSt/KO7DwNWE9xYPxXIjjluPzPrpjGyRJRhrNMbaOjukxUolGRRsLDy20RwtzSS4+lR4DWCfHNjAcL8CJ0IZh6KRIWzT18jWEbRGDjYzE4DzjGzdDPbHTgRFRYQIMyDegfwjrt/YGbNgTvN7N4wPxhhPp89gMsIKk1q8CwRecBMd//Y3Te4+x+AhQTLSNsBxJxzNiWvmZJKwhnKI4HzwxzMbYGOwDgze9bMhlmgKTCY4KJeBKAp8KK7fxp+X3UHbjGzKWY2Iuw37QjOOdFq7CJAbWCqu88In19JEOg5BNgDgiXKBOec9UlpoaSqsox1dM6RpNIy5EouHETfDnwIPO/uW8Pt5wCHuvuo8HmGu2cX/U5S1dj2RM1PAs8S5LN8BRgKXOru94XH1XV3DYQEADM7k+BO6XvA8cD/gG+Aa4DPw0ERZtbQg2rsIgCY2d7AY8BxsbN3zOwZoJq7Hx0+1zlH4phZM3f/Nbz5+U+C2e+3AjcRFE86OCxuUtPdNyezrZI6zOxwgtlg9xNUV78fuIug7/QCRrv7NjOr5+7rktdSSTVh2oNHCPLmfgtcQlCptiOQ5+4nhsfpnCNxwlmpjwLH50vVo7GOpBzNLKyEzKxLOO29Zjgz7HXgYmBEmOMJd38AqGNm3cLnChQKkeTLocidhFeB6kBLYC/gA6C9mQ0A0BeZmFkLM2sB4O6PAP8hSPD9mLtf6+4vEFyIpUeWJStQKBAEecIZX7j75wQ5dd+NJIwP/QWoYWZ1w+N0zhHMrJ1tL2YSWbqVA0xw9+vCZep/JZiZ0TI8ThftVZyZRa993P0V4Dzga+Bjd7/d3XPCfpNBsOoGBQoFouecrgDuPhm4D3gaeArY193HExSiiNI5RyA6Tm4G4O6zCG6ka6wjKa9ayYdIRRLeJb2LYDpzhpm9RpBwty5wPdDCzL4gmObcGFiZrLZKajGzg4B+ZvYfd18cs9xmC8EdsFyCgM9sglliWUlpqKSUMEnzbcAPFlSQvMjd/2NmnwGLYg4dANQE0gku6KWKM7ORBN9LK82skbvv6+6XhTe1PgvzX34NjAaaoeU4EjKzwwhSHiwOb3L9A3jP3VcAX8UcdwLBeUcF3QQzGwo0N7NJkQCguz8XLl0/L1JcwMyOJhg3Kz2PAAXOOTWAO9z9FjN7gmBc83N46EigplZsSUS+cXIdgpzeV4f9aFZ4/aWxjqQkLUOuRCyoXPsCcG5YSOA4glwZm4ArgIMJcml0JZgpdkFsZVKpusxsEMGy0U8IiuC86O6Lw30G3ANMDu+kYmaZkSXtUnWZWXdgInBmmCvsXmB34GHgDXffFh53KnAucIq7f5us9krqCAOBdwAXAtOAycBn4YwezOx6gu+rpUBn4CR3/zJJzZUUYmYdCc47f3T3mWZ2KcGs91nA0+6+KrwIO57t551vktdiSWqY0rUAAB7WSURBVAVmNgx4myCYfCPBd9T6mP13A38iSIVwAHCCu3+djLZKaininNMb+BT4r7uvDI87HbgK+J3OOQJFjpPbAw+7+/+Z2bUEQcIlaKwjKUgzCyuXdQTFTGoBuPuzZrYSOAr4k7vfY2bvENwBy3D3VclrqqQKM6sOdAOOI6jIdQrBctHnwxmGbmZXuPuGcPmOK1AooS3A58B0AHc/38yuAM4gmHn6pZkNJMjlc6wChQJBHh7gBGC8u78dbnsIGBg5Jrzr/m+CdAib3X1pUhorqWgt8COwAsDdbzezU4Dh4fZXCApVjEWBQiHIGwccSDAerkdQhTTNzF6PBAzd/UIz+4igku0d7r4waQ2WVFPcOWcx8Eo4YaMxMEZjHYlR2Dj5SuAsM/vB3a81s+cIzjtbNNaRVKNgYSURzv7KBb4HepnZF+6+2d3fNrNWBCelf7q7KkhKnDB598vAVndfFw6qjyYYSD/v7lkEAWZUCVDySQN6EAR5PgJw95vN7J8E6RCGu/s0MxukmxMS4e7rzexWYIOZmQdLHFYQBHgAMLNq7j4vaY2UlBSOddIBA/oSXMDj7k+GuZ8uAV5x91lmdpxyPgkEeePM7DFgvbuvCZcCnk0wzpkUjn3M3ScmuamSYspwzllmZndp6bHkU9g4+aZwnHwfMCS2yIlIqlGBk0rCA5sJksOfDRxuZpEZhk8S3NnolsQmSgpz9xUx+XteBF4kqCA53MyuAx4KZyCKRLn7fIJ8lv9nZr1idl0AbDSzBuFxChRKHHf/NjJzOdyURzgr3szOBu4KL9JEosKxzq8Ey7oeDpeWRvbdBGwxs3bhcwUKJcrdl7j7mvDx/QQpV/4I7GNmfyNYfiwSp5TnnLbhcwUKJU4J4+R1ZlY/OS0TKR3NLKxk3P21cInXpUBTM/sKaERQ0OSXpDZOUkLMTJ64x7HP3f0FM/sFeIggyfdhkfxzUjUV01fuNbN6wAcWFFj6FDgSaIuSNEsJ8s0s/NzMRgOnE+SGUlJliYo9B7n70+F55zkzO49gmVcfoDWgIKFEFfLdlebuee5+n5mtIKhkm0PwvSUSVYZzzoYkNlNSVCnHyRrnSEpTsLACiwx4Yp5HTkrPmNkaYBhBUZNMYKy7q6qbAGQA2yC4Y5pvMBT7vDHQAthf+VeEIGi8LuZCKzrAcfcbzWw1MIEgf08X4OTIbFWp2sysDbDcC8l1GtOPcglmxe8NnO7/396dh+061vsff3+WNZnHkFJJtCkpldmWeUohc4ZQGXaD1DbVbvQru0I/Fe0k0qgIvyglRKQ0UJnKUKZltsWysCzr8/vjPO+na92etdbjwnPdq+fzOg6H+77ua63jVN/jvM7rHL7f5JkLQNIGwEttf6t/kcv2l2u/sz2lmMnSlNyo2ck8xkmav562afYx1O8zG+Pn6cBkYIOMcwKg7vSa7pLKyb0d7ulzYiQkzWf7KZi178k4OeZVqYY8D5K0BnCt7SdmN2FYP0+g5NiY3/Y/OmpuDBBJm1GKT9wKXGP7m7O5bwHKQOjXeWkPSZsA3wV2sH1Zf7/TuO8llB0atn3XaLczBo+kbYBPAG+xPWUO961OqVS6vu2/jFb7YnCpVMz+DvAz27s1+52+sc7ilMnmia5VSWPskrQ55Yjf34ArbH97NvctBrwP+KFT9TgY6nOOBKYAf7R9dOO39DkxW5JWqkeOZ5kwHOa+jJNjnpKchfMYSW8GfgdcJWlib4W093vfCuoM29MzURgAkrYE/hv4MSVZ89p9vzfjaBpwaiYKo1qVMjD+nqTNmv2OKgCXHHRTMgAKGOpzjgA+PNxEYV+f80dgpUwUBgzFzifqP6+TtGFzgaJvrPOQ7Yfz0h51suc44AfAo8Aqfb83+5yHgKMzURgwNMl8LPBlSv73l/bd0nxnTp8TQyS9BfidpG8A2H5K0nx992ScHPOk7Cych0h6AfAtSvWkvYFXAq93qWY721WMCEkrAz8CDrJ9oaStgIOBU4GZtk/vsn0xmHor6ZLWA9YHbgZOADYC7shCRAynDopXAG4C3mH7tLqavj1lRf3+Zp/TiDM5g5IxT9IbgG8CB9i+RNJRwHLAe+pCVsQsap8zP6VIyem2z6qL6++nFGyT7RN796afiaZ6Eus44DLb35O0IWXi8ExgQdsfrvcldmIWKgW1zqWcvnkV8Ljt/epveTePeV52Fs4DGqsR91FW2X9qe0fKNvnf19ws6YziaXqxA9wCbF8nCpcHTqnXVgDeK+nwrtoYg6sxKL4d2IwyIDoMuAy4X9Ly/aunMbY18sndQimQ9NGad+40Sn+zGrCfpHf2/kwvzvISNrY1nlcTKHnALqnff03ZIbZQvS9j15hF7XOmAXcC+0paHzgJuA5YFNhT0hG9e7traQySxvvVk8DDwCfqDsNvAxcCfwXeJulL9b7ETgCzPK/uoaQz+BJwDPACSSdD2WHYUfMinjPZWTgPkLSw7Uca3ye6VqaV9FNgOdurSdoRWNT2yV21NQZLM1Ya114CbG77a/VhdwDwMtuHddLIGEiSxtueUScDJwFn2N5a0uuA8ylH2Tepx0cjgPK8AqY2cjudSClc8n7bX6z5UA+lpMk4qsOmxoCRNMmNIjiSJtSXeCT9GLirt2MjoqkXK3VhYndKgbbbbR9Sn2F7AKvbPqTThsZAkTTZ9uP186LAUZSxzUzb76nX1wPeDewzXK7mGJt6Y+S+awJeS9nYc5/t/ST9O2We+ZddtDPi2crq7ICrx0V/JOmExsrWdEkT6+ctgBskPQp8jlKOPQJJmwKnSvpc3y6e22x/rX42sAiwsKRxjZWyGKMkrVFf2mf0jlDUHRu/kfTflGM57wQOB86QtFDiJuCfzyvgy3WSENsHApva/mL9Pg2YCSzezHcZY1t9Xp1Sn1fvgrLbp7Fz+VhgEUnLdNbIGDiSVm1OKtv+Ze1zzgIm1AXTp4BlgBfW46YRvfyW35N0tKRDbP/D9nsphbaWrgtbAGtQJp8TOwEMFYv8iqTD60YdYOid6s/Ax4DJkv5MOcl1RzctjXj2Mlk4wCStA3wF+L+UXAhrSPplPXY8XdL89dYzKYmct0mi5gCQtDVwPPAT4BFgI0mvHua+vYBdgONtz8wRi7FN0raUAkq/7aU3kDSp/jwd2Al4r+0f1QnntWxPTdzEMM+r1SVdUSeeL+pN+kjaE9gBOKkeHUzsjHHDPK/e1HteNY5x/QFYi/K8iugVFbiSsmg1qV4bX3++EVgQ+Iik9wB7Ap/sTSrG2FZ3e50CfJ+Si3kbSefVny8EHgNOknQI8C7gP5u7nmPskrQxJafuFcASwNt7i6MAtmfYvoqSvmcx4K22/9ZJYyOeAzmGPMDqAHp920c2rl0ATLa9Qf3+SsoL2sE5DhgwdMz4B8BHbf9U0lLA2cCXbH+v3rMIsCnwKWBnp+rxmCdpaUqcnASsQ1lN38D2Y/X3ScArbF/by0vXXWtj0IzwebUVZYfYjulzAkb8vBrnUoF9c8rR0us7bHIMAEkvAs6hxM4KlJ1fb+9N6NTTN/tTCgEuARxl+7qOmhsDRtI7gBfa/kzNgbogpQLyPba3qxNCm1PypJ6Q2AkYOmZ8KPCY7eMlLQisCHweuMn2QfW+1YFLKWPoP3XW4IjnQCYLB5ikbSgJ4te1fVvj+q8oFbsOrd8XdaqSRiVpIWBt4BLKrvgZko6lVK89tnHfCsCTtrM9PoByBJlS+OYxSkGKFYENbT/ad18mC2MWc3le/dL2YXViaIbtKV21MwbLSJ9XEU11p/KalAImywH/BUwE9uwtcDXunSUXZoSk3SlHRTe0fXe9tgjl+Pp3e6l6hstLF2ObpAMpeVG3s/1AnUB8DSWevm773Hrf0rbv7bCpEc+JHEMeYLbPA04HjqmrqD1H0fj/LhOF0WR7KnCF7Scbg5wZwMIAknaStLPtv2WiMJps/8H2Q/XFai/K8ZxLoOT3kbRHvS8ThTGLuTyvxtd7bstEYTSN4Hm1o6TdOmtgDKR6PP3KmmfueuCTlFQZ3wSQtFHdyUy9HtF0BmUn4cF1NzPAVOBcyuQzUI6UdtC2GGw/Bq4HdpS0SB0P3wzcDby8d1MmCuNfRSYLB98pwBTg+LoTDOAVwCqNXGIRs+jtBGsk8xZwWz3GdRhwVVdti3lDnTDcG7hK0j8ou8Z+122rYsDN7nn1b5ImpZhJDGcuz6vDSb8Tw2jks4SSo/CTwAOSbgJOBf5a78viVszC9nTgfEpOuUMlvcil0vEEYGVJ4/O8iuHYvhW4mrIjfndJS9VFr7uAlyvFIuNfzPi53xJd6B3zs32NpJOAtwOXSzoH2BJ4c45VBJTcPHXg07zWOyY6s166h5Jn417gHbZvHOVmxgDqP07c/93245KuBramFFC6oYt2xmDp73PyvIpnI8+raKP5vKoThzdIupdyHHlr2zd32sAYSL0cqLbPkzQT2IryvDoT2BnYIjsKYziN2DlB0rspqRDeJel8YB9gkzrpHPEvIzkLB1RvEFQrAi5p+xJJGwFPUXL53NJxE2MA1Jh4IyUZ/LR6rRc7qwOL1dg5lJJP47V58QpJywAPApNsT60JvqlFBFYDHqccq3gZ8P8oieNTQCnm1ufkeRXPSJ5XMTeSXg/c0Myd24ibVwF/t/2opJUoFUo3yfMqmpoTy43YeSUw0/aNkraj7Gi+Jn1OwFz7nZdTFrUmA9tQCuRcZPsv3bQ24vmTY8gdk7S1pH36r9fOaG1KboTetYttX5oXr4Ch6qNfA67qvbTDUOysQ5nk6TkTeGUGQVHzOJ0OfBc4WtJL6krpzNrnnAu8oF67BVgvL14BI+pz8ryKYY1grJPnVTyNpC0o1bFXaFwbV+NmXeAnwGoANV5WzvMqACRtJumtMNTPqPF5beDnwNL12tm2z0qfEzCifudS4FW2H7B9mu0TM1EY/6pyDLlDkjYDfghMlLSY7ePq9XHAAsBOwLvrSnuqjwZQVraAJYAPAgfZvkClitsEYLLtOylHKQ6wfQlAjuMEgKQtgWOAd1Jy9WxHOUZxm6QlgfdR4uZXkuarR7se7qzBMRCeQZ+zf55X0W+EY53987yKJknbUnIs715THEyw/WT5SfMDnwYOtP3rxvPqf7tscwyGuih6FjBO0jvrhI5VqmgvTBnrvNv25XleRdMI+539bV+Z2ImxIMeQO1IHyUdREndfD/wRONz2sY1tzku6lmVPZxRNksYDXwfeASwLfB/4OyX3ym62f1bvG1pJ7aShMTAkLUwpUnKO7dPrtc8DC9vev35fxvY96XOi3zPpcxI70fNMxzqQ51UMPa/+DFxteztJy1EWKxagpMg4GZhh+5EOmxkDSKVQ0hcpfc7NlJMUH7J9WqPPyVgnnib9TsTTZbKwQ3WFYknbd9Rtzb8AjrB9TP19cdtZJY2nUamE/QvKEZzFgb/aPlHS3sCJwKtz/C+a6ov4qykTPNNsPyVpd0oy773rPQu5VHWLmEX6nGhL0mTKWOfOjHVipCStBZwGXESpqv4zyomoF1Ny7h4FPOkUFIg+ddJngToh+DbKM+pQ26fW3xe1/Y8u2xiDKf1OxKxyDHmUSVoZeASYYPs2SVPq8YlfqSSEv1jSA8BUYDdJu1E6pczqjnGN2Jlo+1ZJRwCHA+NtfwDA9jckbUJJ1BwxpK6m31CPU/Q8SknMjKQ9gdUkHelUAgzS58SzV8c3jwN31l08GevEbEl6CeWFfKLt30jaCzgP+ILtz9V79gE2diqsxzBqn/MI5dmF7TPrYukJKtWyATaX9KGMdQLS70TMSSYLR5GkNwOfBa4CVpX0GdvfVzGx5s74N+Am4C5ga9vTu2xzDIZhYufTtn+gUs3t85KOsP2ZulPsNcC0Of19MTbUuFmlN9gB+gfG44C7VCoBHkI5TprBc6TPidaa/U7dwSxXksZnrBPDkfQW4KPArYAlnWz7J5LWsH1b41YD4+vpnMczwRwqBSlebPvk2ueM6+38qv3PGXWi8BfAvcBmGesEpN+JmJtMFo6SumpxHCXf01XAlsC36lb4kyT1HlqrAHcCm9u+vpPGxkAZJna2Ar4paQHb/yPpduCzkl4BvIGSlPeuzhocA6Hu9voOsFDNCXZ4fVlv5uh5ghJXqwF72L6ho+bGAEmfE23Nrd9pvKBnrBNDal9yPLALZUfYusB3VIpTnNnINbcv8B/A3rYf67DJMSBUCredAUxQKaB0jO2ZvQnDxnhnReB2YCvb13XW4BgY6Xci5i6ThaPHwPW2L6/ff1h385wj6cHaKU0E3kRZZc/gOXr6Y+dMSVOBsyU9bPssSRdRKrzZ9v2dtTQGyZqUqscXAbfUQc9hfROG91BW2Q/K4Dka0udEW3Psdyi7mccBG5GxTvzTgsC1tn9Tv18n6VHgq5Iesf0zlXyXe1Be2K/prKUxMFQKKG1MKULxa+DHtc/5fG/CkJIiYz5gM+CtGetEQ/qdiLlIgZNRJOkqSiXSjzeu7QK8D3ib7bv7dv1EALONnV0psbOj7SldtS0GU30xf6HtKZJWAn4PfMX2ofX3XjXAJWw/2GljY+Ckz4k2RtDvLJuxTvSri+VXAifY/mrj+oGU3T7vAhYCZuZ5FU0qxbcWtX2vSj7UbzFrrrmlbN/fPJocAel3IkZiXNcNGAskzVc/Hgy8UdJBjZ9/TtkW/ziULRqj3LwYYHOJnQuA20iusOjTOPI3pQ6QbwTeCBwg6QhJOwOnSpqcAVA0pc+JtkbY75wiaYGOmxoDRKUgxXTgC8B6knZo/HwlMD/lZf3+PK+in+0nbN9bP18M7AkcLGkfSdsCn1apyJ73qxiSfidiZHIMeRTYfqp+/APwbWAPSS+2fSQlF9TKwMSu2heDK7ETbTQXHepRnPG2/yLphZS8LHdT8vY83lkjYyClz4m2nkG/k8nmGNLoc34OLAtsJ2k521+i9DfLUnb35IU95qguWFwkaTPgGuA+YNOMdaJf+p2Ikckx5OdZXbkYqswlaSlKJ3Q8cEf9vIvtP3fa0Bg4iZ1oY5i4mWT7ifrbNsCJwJbJ2xP90udEW+l3oo1GvPTiZ3lgA+Bw/tnn7GD7T502NAZOI3Zm+Xf97R3Axyl5UdPnxLDS70TMXSYLn2OSNgZeA0y2fXS91nuQrQ+sa/uz9fqSlMX4rFpEYidaGUHcbAAcb/tRSf8FnJUkzQEgaU1KkZKnbP+iXhtve0b6nJiTEcRO+p14GkmvBSYDj/YWHSRNtD1d0gbACrZPkzQeeCnwsO37OmxyDIi5xM76wKrA12sfdALwVdtXd9jkGDC9Z1T93Kt0nH4nYg6Ss/A5VFfPv0A53r1bL9dTfWl/A3AyMFT5z/YDefEKSOxEOyOMm2tsP1qvfyov7AFQczmdDOwMfEzSj+ok84z0OTEnI4yd9DsxC0nbA9+h7N75jKRTAOpkzzrAScBD9doM2zfnhT1gxLFzV28iyPZBmSgMAElvkXR+nVie0cvLXCcK30D6nYg5ys7C54ikZYCzgffZ/q2k91KSwN9q++eSdgQerLk0UgUwhiR2oo3ETbRVY+eHwP62r5G0OvBb4ELbW9UXs4dsX5zYiabETrQhaTHgJ8B7bP9e0oqUqrUP2t5GpfroLbZ/mriJpsROtCVpQ+C7lNyV04AN6wTzBGAG8EHgKtsXJnYihpcCJ8+dycACwN2SlqZUkvwtsIqku2xvCf/c9txhO2PwJHaijcRNtDWJEjv31u83ACdQEnyfbHu/3o2JneiT2Ik2xlNip7eD52aVIhQXSDrO9gd6NyZuok9iJ9paCPic7eMknQH8UtIGLlWQkXSsGzlTu21qxGDKMeTniO1bgV8BFwKXA6fY3tX26sCLJB1c78uDLGaR2Ik2EjfRlu3bgCuBL0t6PfCp+tO2wDJ191jE0yR2og3b9wOXAkdIWqJemwrsB6woaaUu2xeDK7ETbdk+Dzi1ft0DuA24TNKkeu1V9b5MFEbMRiYLn4VeZ1MTomL7QGAT4ALgB41bz6WuiEVAYifaSdxEW43Yma9eOgG4G/gwsBTwIZek8fMD8w37l8SYlNiJNiStLulNNacclJf2mcD7e5M+lJ2pE4EFO2hiDKjETrTViJ21AGz/b/3348DewK3A+fX4+nGNeIqIYWSysCVJOwDXS1q5JkztvbzfDtwOfFzScjVv2BbAFR02NwZIYifaSNxEW32x8xSA7T/afi+wu+19a0ztRnnxeqLL9sbgSOxEGyrFt75BSY9xmKT9a8GJ84GlgeMlrQrsBCwL3N9ZY2OgJHairb7YOVzSe5q/255meyfghZRd8Yc4hdsi5igFTlqQtCbwTeAu4NXAOrZvVC3JXhN+Hwy8hjJw3r+uuscYl9iJNhI30dYcYmeWHD0qlbT3B/a0/aduWhuDJLETbdSiAv8D7Gz7T5L+gxI7e9TfX0+ppL0OZbfYwU7l2iCxE+3NJnbWsr1X3317Av8H2NL2dR00NWKeksnCFiS9Dljb9omSjgX2Ada0fWPffS+jVAXMccAAEjvRTuIm2ppT7DQnfSRtCfzN9l+6bG8MjsROPFOSBOwGTLb99XrtxcBZwA51J3zv3sWAJ2w/1kljY6AkdqKtucTOdsBdtmfW69sDN9q+pqv2RsxLMlnYkqQleluXJR0D7EtZwfirpPVsX95tC2NQJXaijcRNtDWX2Fnf9mXdtjAGVWInnilJ8wML275X0kRgCeAcYAvbD0l6TXagxnASO9HWCGJntZy4iXjmxnfdgHmFpA2AlYFpwFm2H5Q0wfaTtj9YFjW4VNIJwK6SNrU9pcs2x2BI7EQbiZto6xnGzm6SNknsBCR2op0aN68EHgHOtn1v/elJ4DHgyfrC/nZgD0l7ulS5jTEusRNttYidvWzf11FzI+ZJ2Vk4ApK2BT5HSa67PLA68Cbbd0iaaHt6ve9yyiB7k6x8BSR2op3ETbSV2Im2EjvRRl/cvISSO/dNtu+ov08AvgtcC2xLyW15bUfNjQGS2Im2EjsRoyM7C+dC0nzAHpSKST+u175CWVlfv7eiXjutpYGNkgchILET7SRuoq3ETrSV2Ik2Rhg3iwMbA6sA29v+a2cNjoGR2Im2EjsRo2dc1w0YZDVh6jhgYWCp3nXbBwAXABdJmlQvTwG2zuA5ILET7SRuoq3ETrSV2Ik2Rhg3k+vRwK8Bu+SFPSCxE+0ldiJGVyYL58DFk8C3gfdLWrPx2/7AdcAm9fvv3VeZNMauxE60kbiJthI70VZiJ9oYYdxsXC99JBPM0ZPYibYSOxGjK8eQhyFpPUrugxmUhKnflrQicISkz9i+st46FVikq3bG4EnsRBuJm2grsRNtJXaijWcYN4sB9PJdxtiW2Im2EjsR3UiBkz6StgE+DVwMLEupqLQv8FJgl/rPF4GJwEHAW23f1E1rY5AkdqKNxE20ldiJthI70UbiJtpK7ERbiZ2I7mRnYYOkNwLHALvavlrS2sDHgIVt3yTpOOB6YAfKysZu6YwCEjvRTuIm2krsRFuJnWgjcRNtJXaircRORLeys7BB0urAurZPbFy7AjjS9sWNaxOAGc7/eFEldqKNxE20ldiJthI70UbiJtpK7ERbiZ2IbmVnYYPtP0q6BUqnUxOoPgw8Uq9tDPzB9kMdNjMGUGIn2kjcRFuJnWgrsRNtJG6ircROtJXYiehWqiH3sf1I72P99xPAfZLeBnyWJPmO2UjsRBuJm2grsRNtJXaijcRNtJXYibYSOxHdyTHkuZD0DWABYAVgb9vXdtykmEckdqKNxE20ldiJthI70UbiJtpK7ERbiZ2I0ZNjyLMhSfXjSpTOaIMkTI2RSOxEG4mbaCuxE20ldqKNxE20ldiJthI7EaMvOwvnQtL2wI22r+m6LTFvSexEG4mbaCuxE20ldqKNxE20ldiJthI7EaMnk4UREREREREREREBpMBJREREREREREREVJksjIiIiIiIiIiICCCThREREREREREREVFlsjAiIiIiIiIiIiKATBZGRERERERERERElcnCiIiIiGhF0pKSrq7/3C3pzsb3afWel0mypKMaf24pSU9K+lL9/vG+P3u1pMW6+u+KiIiIGMvGd92AiIiIiJg32X4AeC2UCT9gqu3P1+9TG7f+DdgG+Ej9vhNwbd9fd1zvz0ZEREREd7KzMCIiIiKeb9OA6yW9oX7fBfh+h+2JiIiIiNnIZGFEREREjIbvAbtKWh54CpjS9/sHGkeQLx795kVEREQE5BhyRERERIyO84FPAfcApw/ze44hR0RERAyA7CyMiIiIiOed7enA74EPAmd03JyIiIiImI3sLIyIiIiI0XIMcIntByV13ZaIiIiIGEYmCyMiIiJiVNi+lqdXQe75gKQ9Gt+3s/33579VEREREdEk2123ISIiIiIiIiIiIgZAchZGREREREREREQEkMnCiIiIiIiIiIiIqDJZGBEREREREREREUAmCyMiIiIiIiIiIqLKZGFEREREREREREQAmSyMiIiIiIiIiIiIKpOFERERERERERERAWSyMCIiIiIiIiIiIqr/D7Rxg6ukotMvAAAAAElFTkSuQmCC\n",
            "text/plain": [
              "<Figure size 1584x432 with 1 Axes>"
            ]
          },
          "metadata": {
            "needs_background": "light"
          }
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "\n",
            "\n"
          ]
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "iVBORw0KGgoAAAANSUhEUgAABQ0AAAHTCAYAAACN/z7VAAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADh0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uMy4yLjIsIGh0dHA6Ly9tYXRwbG90bGliLm9yZy+WH4yJAAAgAElEQVR4nOzdd3iUVd7G8fskISRA6F3EUEUUpSmrIsWuqKtr2bX37rq7vq6CBctasKy7uth7Azs2EAQhdKSH3gkQAqSRkF5mzvvHTIaZZCaZhCST8v1cl5fMU88zeTKZueec8zPWWgEAAAAAAABAibBQNwAAAAAAAABA3UJoCAAAAAAAAMAHoSEAAAAAAAAAH4SGAAAAAAAAAHwQGgIAAAAAAADwQWgIAAAAAAAAwAehIQAAQANjjBlljEn0epxgjDm7Bs6TbYzpWcE2scYYa4yJOMJz1cg1+DnPW8aYx2v6PEG04xFjzHuhbgcAAGi8jujNGwAAQH1njEmQ1EmSw2vxR5KSJD3ifhwhqYmkPPfjXdba440xVlIfa+02r+M9Kam3tfa6mm157TLGxEn6zFrrCbKstS1C16KaYa29q7bPaYwZJddz282rHc/VdjsAAAC8ERoCAABIF1trZ/lZ/pwkGWNuknSbtXZ4rbYKAAAACBGGJwMAANQiY8xwY8wiY0yGMWaPO5CUMWaMMWaVMeaQe/mTXvuUDPG90Riz2xiTaox51Gt9tDHmI2PMQWPMBkknl3P+MGPMWGPMdmNMmjHmK2NMW/e6KGPMZ+7lGcaYZcaYTsaYZyWdIWmie0jyRPf21hjT26sN/zbG7DLGZBpjFhhjov2c/3L3UOMTAp2vnKfvZGPMBvd1fmiMiXIfc50x5mKvczRxP0eDAjwHDxlj9hljkowxt5W6jo+MMc+4/73RGHOR134RxpgUY8xg9+M/eP0s4909Bku2jTPG/MsYs9AYk2WM+dUY095PW5pL+kVSV/dzm22M6WqMedIY85l7m5Kf/83ue+OgMeYuY8zJxpg17vNPLHXcW9ztP2iMmWGMOaac5xUAAKAMQkMAAIBa4g5ufpH0P0kdJA2UtNq9OkfSDZJaSxoj6W5jzKWlDjFc0rGSzpI03hhznHv5E5J6uf87T9KN5TTjr5IulTRSUldJByW97l53o6RWko6W1E7SXZLyrLWPSpov6T5rbQtr7X1+jvuypCGSTpPUVtJDkpylrv9mSS9IOttauy7Q+cpp+7Xu6+slqa+kx9zLP5HkPRz8Qkn7rLWrSh/AGHO+pAcknS2pt6RR5ZxvsqSrvR6fJynVWrvSGHOUpKmSnnFf74OSvjXGdPDa/hpJN0vqKCnSvY0Pa22OpAskJbmf2xbW2qQA7RkmqY+kP0v6r6RH3ddxvKSrjDEj3df4R7mG1v9JrvtsvvtaAAAAgkZoCAAAIH3v7q1V8t/tldh3pfe+ksaWs+01kmZZaydba4ustWnW2tWSZK2Ns9autdY6rbVr5Ap5Rpba/ylrbZ61Nl5SvKST3MuvkvSstTbdWrtH0mvltOEuSY9aaxOttQWSnpR0hXEVKimSK7zrba11WGtXWGsPVfQEGGPCJN0i6W/W2r3ufRe5j1/i75L+KWmU1xyQlT3fRGvtHmttuqRndTjQ+0zShcaYlu7H10v6NMAxrpL0obV2vbU21339gUySdIkxppn78TU6HL5dJ2matXaa+2c2U9JyuQLLEh9aa7dYa/MkfSVXSHwk/mWtzbfW/ipXyDzZWptsrd0rVzBY0rPyLknPW2s3WmuL5RpmP5DehgAAoDIIDQEAAKRLrbWtvf57txL7DvbeV9KEcrY9WtJ2fyuMMcOMMXPcw18z5Qp+Sg9n3e/171xJJYVIukra47VuVzltOEbSFK+Qc6NcRWA6yRW0zZD0hXvo7ovGmCblHKtEe0lRga7N7Z+SXrfWJnotq+z5Sl9jV0ly98xbKOlyY0xruXrufR7gGKWfqz0BtpM73Nwo6WJ3cHiJXEGi5HoerywVGA+X1MXrEIF+XlV1wOvfeX4elxz/GEmverUrXZKRdNQRnh8AADQihIYAAAC1Z49cQ2v9mSTpR0lHW2tbSXpLrqAnGPvkCiRLdK+gDReUCkmj3D0Ei6y1T1lr+8s1zPgiuYZMS5It55ipkvIV+Nok6VxJjxljLi9ZUMH5/Cl9jd7DeD+Wq/fflZIWu3vf+bNPUjevx0cH2K5EyRDlP0ra4NVLco+kT0s9j82tteWFxoGU99xWxR5Jd5ZqW7S1dlE1nwcAADRghIYAAAC153NJZxtjrnIX1WhnjCkZshojKd1am2+MOUWuobDB+krSOGNMG2NMN7nmLQzkLUnPlgxVNcZ0cM+BJ2PMaGPMAGNMuKRDcg0fLpmX8ICknv4OaK11SvpA0ivuIh7hxphTjTFNvTZbL+l8Sa8bYy4J4nz+3GuM6WZchVselfSl17rvJQ2W9De55jgM5CtJNxtjjnP3Hny8nG0l6Qu5As+7dbiXoeQaEn2xMeY89/VGGWNGuZ//yjogqZ0xplUV9vXnLbnuh+MlyRjTyhhzZTUdGwAANBKEhgAAANJPXpVrs40xU2riJNba3XLNefd/cg0ZXa3D8xLeI+lpY0yWpPFyhVvBekqu4bo7Jf2qwPP5SdKrcvVo/NV9riVyFdiQpM6SvpErwNsoaa7XsV6Va+7Dg8YYf3MmPihpraRl7mt7QaXea7rnYrxI0rvGmAsqOJ8/k9zXt0OuodDPeB07T9K3knpI+i7QAay1v8g15+McSdvc1y9JBQG23ydpsVw9Ib/0Wr5Hrt6Hj0hKkat33z9LX3MwrLWb5OrRuMM9pLhrZY9R6nhT5Hr+vzDGHJK0Tq4h2wAAAEEz1lb3aAgAAACg9hljxkvqa629rsKND+9znFyhWlN30RAAAACInoYAAABoANxDlm+V9E4Q215mjGlqjGkjV4+8nwgMAQAAfBEaAgAAoF4zxtwu1/DgX6y184LY5U5JyXINcXbINV8hAAAAvDA8GQAAAAAAAIAPehoCAAAAAAAA8EFoCAAAAAAAAMBHRKgbUFvat29vY2NjQ92MGpOTk6PmzZuHuhmoh7h3UFXcO6gK7htUFfcOqoL7BlXFvYOq4t5BVYTyvlmxYkWqtbaDv3WNJjSMjY3V8uXLQ92MGhMXF6dRo0aFuhmoh7h3UFXcO6gK7htUFfcOqoL7BlXFvYOq4t5BVYTyvjHG7Aq0juHJAAAAAAAAAHwQGgIAAAAAAADwQWgIAAAAAAAAwEejmdMQAAAAAAAACKSoqEiJiYnKz8+v1fO2atVKGzdurNFzREVFqVu3bmrSpEnQ+xAaAgAAAAAAoNFLTExUTEyMYmNjZYyptfNmZWUpJiamxo5vrVVaWpoSExPVo0ePoPdjeDIAAAAAAAAavfz8fLVr165WA8PaYIxRu3btKt2DktAQAAAAAAAAkBpcYFiiKtdFaAgAAAAAAADUAeHh4Ro4cKDnvwkTJuiyyy7TwIED1bt3b7Vq1cqzbtGiRRo1apSWL1/u2T8hIUEnnHBCtbSFOQ0BAAAAAACAOiA6OlqrV6/2uy4uLk4vv/yyfv7551ppCz0NAQAAAAAAAPigpyEAAAAAAABQB+Tl5WngwIGex+PGjdOf//zncve59tprFR0dLUkqLCxUWFj19BEkNAQAAAAAAAC8PPXTem1IOlStx+zftaWeuPj4crcpb3hyIJ9//rmGDh0qyTWn4UUXXVTlNnpjeDIAAAAAAAAahY37DmnCL5tkrQ11U+o8ehoCAAAAAACgUbjyrcXKLijWfWf2VoumgWOxinoENgaEhgAAAAAAAGgUnO4ehibE7Qik9JyG559/viZMmBCSthAaAgAAAAAAAHWAw+EIuG7UqFEaNWqUz7K4uDifx7GxsVq3bl21tIU5DQEAAAAAAAD4IDQEAAAAAAAA4IPQEAAAAAAAAIAPQkMAAAAAAABAknUXSmloqnJdhIYAAAAAAABo9KKiopSWltbggkNrrdLS0hQVFVWp/aieDAAAAAAAgEavW7duSkxMVEpKSq2eNz8/v9KBXmVFRUWpW7duldqH0BAAAAAAAACNXpMmTdSjR49aP29cXJwGDRpU6+etCMOTAQAAAAAAAPggNAQAAAAAAADgg9AQAAAAAAAAjUIDq3FSowgNAQAAAAAA0KgYE+oW1H2EhgAAAAAAAAB8EBoCAAAAAAAA8EFoCAAAAAAAAMAHoSEAAAAAAAAAH4SGAAAAAAAAAHwQGgIAAAAAAADwQWgIAAAAAAAAwAehIQAAAAAAAAAfhIYAAAAAAAAAfBAaAgAAAAAAAPAR8tDQGPOyMWanMcYaY07wWt7XGLPYGLPF/f8+wawDAAAAAACoT876d5zO/c/cUDcD8BHy0FDS95JGSNpVavlbkl631vaV9Lqkt4NcBwAAAAAAUG9sT8nRlgPZoW5Go2BlQ92EeiPkoaG1doG1do/3MmNMR0mDJU12L5osabAxpkN562qrzQAAAAAAAJUxc8MBbTmQFepmwM3IhLoJdV5EqBsQwNGS9lprHZJkrXUYY5Lcy00561JC1WAAAAAAAIBAbv9kuSQpYcKYELcECE5dDQ2rhTHmDkl3SFKnTp0UFxcX2gbVoOzs7AZ9fag53DuoKu4dVAX3DaqKewdVwX2DquLeQVUFc++Ut577ruY5HU5J0rz589Q0vG70Nqyrrzl1NTTcI+koY0y4uydhuKSu7uWmnHU+rLXvSHpHkoYOHWpHjRpVaxdQ2+Li4tSQrw81h3sHVcW9g6rgvkFVce+gKrhvUFXcO6iqcu+d6VMlyf/68tahWoX99ovkdGrEGSMUHRke6uZIqruvOSGf09Afa22ypNWSrnYvulrSKmttSnnrar+lAAAAAAAAQMMT8tDQGPOaMSZRUjdJs4wx692r7pL0V2PMFkl/dT9WEOsAAAAAAAAAHIGQD0+21t4v6X4/yzdJGhZgn4DrAAAAAAAA6qrnf9mom06LVZdW0aFuClCukPc0BAAAAAAAaCzenrtDD3wZH+pmABUiNAQAAAAAAKhFxU5nqJsAVIjQEAAAAAAAAIAPQkMAAAAAAAAAPggNAQAAAAAA0ChYG+oW1B+EhgAAAAAAALWI4Cr0jAl1C+o+QkMAAAAAAAAAPggNAQAAAAAAAPggNAQAAAAAAADgg9AQAAAAAAAAgA9CQwAAAAAAAAA+CA0BAAAAAABqEcWTUR8QGgIAAAAAAADwQWgIAAAAAAAAwAehIQAAAAAAAAAfhIYAAAAAAAAAfBAaAgAAAAAA1CJrKYWCuo/QEAAAAAAAAI0CcW3wCA0BAAAAAAAA+CA0BAAAAAAAAOCD0BAAAAAAAACAD0JDAAAAAAAAAD4IDQEAAAAAAGoRxThQHxAaAgAAAAAAAPBBaAgAAAAAAADAB6EhAAAAAAAAAB+EhgAAAAAAAAB8EBoCAAAAAAAA8EFoCAAAAAAAUIss5ZNRDxAaAgAAAAAAAPBBaAgAAAAAAIDGgV6eQSM0BAAAAAAAQKNiTKhbUPcRGgIAAAAAAADwQWgIAAAAAABQixghi/qA0BAAAAAAAKAOeCNumzbvzwp1MwBJhIYAAAAAAAB1wovTN+uPry8IdTMASYSGAAAAAAAAdUZ+kTPUTQAkERoCAAAAAAAAKIXQEAAAAAAAAIAPQkMAAAAAAIDaZKmfjLqP0BAAAAAAAACAD0JDAAAAAAAAAD4IDQEAAAAAANAoWDE0PFiEhgAAAAAAAGhUjEyom1DnERoCAAAAAAAA8EFoCAAAAAAAUIsYIIv6oE6HhsaYBGPMJmPMavd/57mX/8EYE2+M2WKM+dUY0zHUbQUAAAAAAAAaijodGrpdYa0d6P5vhjEmTNJnku611vaVNE/ShNA2EQAAAAAAAGg46kNoWNoQSfnW2gXux29JuiqE7QEAAAAAAAAalPoQGn5ujFljjHnDGNNaUndJu0pWWmtTJYUZY9qGrIUAAAAAAABAA2KsrbvTbxpjjrbW7jHGNJX0X0kxkqZIusVaO8Zru1xJ3ay16aX2v0PSHZLUqVOnIV988UXtNb6WZWdnq0WLFqFuBuoh7h1UFfcOqoL7BlXFvYOq4L5BVXHvoKrKu3dump7j+XdsyzCNGxalO2fm6sq+TfT1liKfbT86v3mNtrOxcDitdh5yqnfrcM+yW2fkyGGl985tpogwE8LWHRbK15zRo0evsNYO9bcuorYbUxnW2j3u/xcYY96Q9KOkVyUdU7KNMaa9JGfpwNC93zuS3pGkoUOH2lGjRtVGs0MiLi5ODfn6UHO4d1BV3DuoCu4bVBX3DqqC+wZVxb1TMxIP5urVWVv13J8GqEl4fRj4WHnl3jvTp3r+GRMTo4EnD5Vm/qa5+8o+F9x/1eOF6Zv05pLt+um+4RrQrZUkKWzmNDkcViNGjFRkRN24D+vqa07deHb8MMY0N8a0cv/bSPqLpNWSVkiKNsYMd296l6SvQ9NKAAAAAAAQjEemrNPXKxK1cFtqqJuCRmLTvkOSpJTs/BC3pH6qyz0NO0n61hgTLilc0gZJ91hrncaY6yW9bYyJkpQg6brQNRMAAAAAAASr7k6SBsBbnQ0NrbU7JA0KsG6RpAG12yIAAAAAAFBVdWP2uLqhyOH0/DtQiLo2MVMOazXw6Na10yiglDobGgIAAAAAgAaIrobatD+rwhD14okLJEkJE8ZUsCUqow7XA65z6uychgAAAAAAoOEwdDUE6hVCQwAAAAAAUGssXQ2BeoHQEAAAAAAA1Dg6GvrHcFnUVYSGAAAAAACg1hCSuZGieoz7bq1Off63UDcDpVAIBQAAAAAA1DjDpIY+1iZmSpJSswtC3JLQm7x0d6ibAD/oaQgAAAAAAGoNPQ1dbv14ecB1ny5OqLV2AIEQGgIAAAAAgBpHP8PgPf7D+lA3ASA0BAAAAAAAtac+dzT8dkWiBjw5Q8UOZ6ibAtQ4QkMAAAAAAFDjGsKUhk/8uF5Z+cXKK3LU6nkzcgs1fd2+Wj0nQGgIAAAAAABqjWVSw0q7b9Iq3fXZSu3LzAt1U9CIEBoCAAAAAIBa0AC6GobInoO5kqSCIoZFV0Z58fQnixO0PCG9tppSL0WEugEAAAAAAKDxqM/9DOklWT8ZP4H1M1M3SpISJoyp7ebUG/Q0BAAAAAAANa4hzGlYwlRwMUOfmaXYsVOr/bxViSw/WZygP7+9uLqbUil70nNDen5vxL7BIzQEAAAAAAC1pj511sspKNautJxK75eaXVCt7Qgmb126M11XvLlIRaUqO4//Yb1+33l4GG7yoXylZFVv+8rz3cpEnfHiHC3enlZr50T1IDQEAAAAAAA17nDwVX9Sw2ve+10jX4rzPM4prN2qyZKUX+RQQlrFPfX++U28lu86qL0Hyy+Wcspzv+nkZ2dVV/MqtGp3hiRpa3JWrZ0T1YPQEAAAAAAA1LiSEb1LdtSf4hPxezJC3QSN/2Gd59+NeU7F+D0ZygtBaNuYERoCAAAAANBAZeYV6fZPlis9pzDUTfH4aFFCqJtQr2xPCW54dEPOE9OyC/TH1xfqwa/jQ92URoXQEAAAAACABuqzJbs0c8MBvTt/R6ib4reCbX1Vl6+krhWcsdUwHD2nwNXDMD4x9D0/GxNCQwAAAAAAgHqgvPitOsK5mlQdWWZNBKI3fLC0+g/aQBAaAgAAAADQQJXMgRfqzmeZeUXKyKs7Q6Trk8r+7Opzj86vlu3Ryt0HyyyvyUB03paUGjt2fUdoCAAAAABAAxfqIasnPfVrvSqAUpHlu8oGWw3hXNVl0bZUranCUOKHvl2jP72xSIu3p/ksL5mvsT4HovURoSEAAAAAAA1UQy6OEUrBVlW+8YOlSskqqOHWuFT2Z51TUFwzDZF0zXu/65KJCyVV7R68+t0lWrHrcMhccojaDr8/WZygDUmHavekdQihYSM1a8MBbdzXeG98AAAAAGgMPGELPbRqzXteRWfmbknR+wt2VtuxC4qcih07VV8t3xNwm2CDtYv/t6CaWuVrT3qu/xWVTPxSslzD2XMKilVQ7CqEUtW7uNDhLHd97Nipih07Vb/v8O3hOP6H9brwtflVPGv9R2jYSN32yXJd8Grlb/xZGw4obnNyDbQIAAAAAFDdSnp5pecGnk9wR0q25jKvW7V5ZurGGjv2QffP8d+/bi6zrrI9+nak5lRHk8o448U51XQk1wUd/8QMnf9fV35hqtjV8M5PV3j+7XAGfqJmbjhQpeM3VISGqJTbPlmumz5cFupmAAAAAACCsNw9xHPS77sDbnPmv+fqRirI1gs/r0kKdRMqraoj5KtjaH3c5sNheFJGXsXnlPTF0t21NqS8riM0BABUWfyeDHofAwAA1GEFxeUPy0TVVKa/W3XOwzd5qf9hycsS0rU3iFAsEGutsmtwjkOpeip4l3eMdXsz9fcvVgXsSVgyx2J53l+wU2O/W6s7P11exRY2LISGjURmbpEsM+ACqGZ/fH0hvY8BAADqMD4HVo+1iZmhbkK5rnxrseff5YWUgQK1d+fv0AlPzND+zPwqt2FbcpaSs6q+f2l+W1rOtd356Qp9vzopYI/C1Ozgew+m5QQezt+YEBo2AjtTc3TS07/q0yW7yqzbnRZggtIgZOQW6iC/SAAAAABQZ5UzfVuVrd6TodixU7V5f1b1H7yOunjiAh045BuILdiaGlQF4rfmbte479ZWa3vyi6rWg/TLZf57Kv6ybr8kHVFvxbNfmac/PPdblfcv7VBeUZllR9pb8cIq1HZozAgNG4GdqdmSpDmbyg4hzPTzSxisgU/P1KB/zazy/gAAAACqV5HDqS+X7ZazJpIi1EvBzONWWVPd8+o1tmlqvItkxCdm6rr3f9eDX8dXuJ+10uSlgeeUrIqqfpbPyPPt+FO2J6r/1473F+zUl8sqvgZ/Lz1V7ew61k/QuiM1R9PW7vO7fcm17DkYuHPUhn2Hgjo3HXRdCA0BAAAAoIF4M267Hv52rb5fvTfUTUEIrNubqa0HfHv/7TuC4aaBlAQq3sNgixwNb+7E2Zt8K+l6X++sja51yxLS69wIvNxChyTpqrcX69sVieVuuz3F1cmovB58N3ywVP/6eYMe/rZsiBc7dqpix07VrrSKKzFXx9yO1kr3fL7S/zr3/6959/cjPk9Nz+9YXxAaNmAJqTnqOW6qdqS4fnlzCx168sf1yi08fPPnFzuqdOzM3Kr3UAQAAABQM9Ld4UUG79cbpYv+t0Dn/GdetR3P4bSaumZfmd5oJY+MjNKyC/TZkl3q8+gvWr0nQ5JriqxvKgirHv5mjWLHTq22ttaEWz7yLYZh/ERrqdmFdW4E3rn/macRL87R0p3p+r9SPSFL96ArcvguSMkqVGZukSbO3urpsTxvS4oqEmjYs7dAPaDLC11/jA++WnR19g4sKKpaVtLQRIS6Aag5P8UnyWml/83eJkn6fWe6ft+ZrjbNIj3b/GfmFk26/Q+VPnaRs+F9iwQAAAA0FIysw3crE3VybNsq7ZuWXaB1SYe0MyVbT/60QS9efqKuOvloz3rvnoanTpitQneF5uUJ6Rp4dGtd+Op85RU5dMWQbgHP8eXyikOmuqY6qyDXtN3pwdUvKPlZZriHO9/12QpdclJX/RifpKPbNtMHCxOCOs4bcdsr3ObxH9br2mHH6Mmf1uu6Pxyjvp1iJEn//CbwEO/7J68K6vySZHnlq3b0NGwESs914PCK3yvT5fYHhjgAAAAAdVp9CjVQc6y1euCreJ3x4pwq7X/d+0t14wdLlXjQNR9iSqmqs95DkUsCQ0kKc9+AeTXYS6ug2KH35u9QcSWHQ3+zIlFP/bTeZ5nDaTVj/f4yPSkHPDnDby/ITRXMh7dq90EdzK+9DjZfLtutrPwj61VsZZVbWKyUrMM/45Kc4NuVexXv7j16ZA4/v4kH8/TJ4l265aNlklxD6mdtrPzcmKV//msTM3XgUPDVkYORkVu3hp2HAqFhAxZoYlTv9xFrKlE2/m9frPZ7DAAAAAB1S9niBgiFXWk5WpaQXuvnDfbHb63VKzO3lFm+Pdk1z52z1NyFadkF+mjhTn26ZJd7ue8nw6qE1rmFxVq3N/jPpW/M2a5npm7U1xUMfy7twa/j9WGpXnPvL9ihOz9doZ/W+BbWyMr337nm48W7yj3HZW8s0sPzq7/wTCAPf7tWA578VY99H1xl5vVJZZ/nRdvS1H/8DJ9rLnn98Dcs+ftVe8vM9VgZJb0BS+6Vie6RkZU1Y/0Bn05QF09cUGYbf9cbrJxCR9C9LBsyQsMG7L0FO/0u3xNkN2UAAAAA9Yu/OdcQOiNfitOVby0OdTN8xG1O1j++dHUImbJqr177batn3Vx3SFRYqhdXUbEr6BnyzCw9+dOGgMcuKcBRGf3Hz9BF/1ugbclZFW8s6ZC7Z11ONRSqSMpwFYlJ9eplV9XAffH2NElSFZ6CI/bZkuAqM495bUGZ63t22sYy283ZHHgOw79/uVq3fLRcB3MKlR9Ej9L0UvMV7kx11Vwoea2q6pDieyet1HXvlV/wZMxrZYPEyvD+3WisCA0bqPIqV3236siHGS/dWfvflgEAAAAoX0nvHToaNm5zyylccdOHyzRl1V4VOZzae9C3V9x783doTeLh4ahpOa4w7T+zyvZGlKTkLN/KzC/N2FzVJislK7ihoCX39vaUiqv1+pNf5Ci3Z+OzU8uGaMG4+t0lVdqvts3cUPVegt4G/Wumhr8wu8LtBpcqEnPTh65hySWvVQXFVR/OXVJ4p7z7HUeG0LCBeuGXTUFv6z13QSCJB317J94doMQ5AAAAgNAp6WdIQYDG7Wb3fHGlefcyW70nQwdKhX7GGF0ycaHn8YFDvutLe3vujiNopa/KDm2evNR/77rp6/ar57ipPnMtenvw63hd9L8FrqHWixLKrP9kSflDkOu7+EpMUVaR1Ozggt6DOWWnTtuVlqvYsVMVV06vxmBk5Bbqxg+WllleUEz14+pAaNhAra3EnBAnPztLmbmBJ0/NLijW8BeqNoEuAAAAgNpDIZTQOpRfJIczNIHt9pTsCrf51CsQM5L2pPv2NCw9h92SHbU3wszI1QvwxCdnaFty2Wv52xer/MpFFeYAACAASURBVBYnKe2uz1bIaaWkDP9zC87fmipJ2h8gEGU+0OpX7Ky54jAPfu2/8vKJT/5aY+dsTAgNG6jKvs5l5BUqu6DY73yHeaGYlAEAAABAlZF71D5X4PVrmQq91aGw2Knnpm0st1LuWf+eW+Fxxv9wuG0fLkyQs5ZulK0HKp6vMK/Iobs/W6FD+cU6+xXfaznl2Vn6YXWSJPntHehP6XkZS5QUDP3Iq8iFd4AYosy3gau5bzMCVV4+kmHPOIzQsAGZtyVFg57+VZOX7tbSSlbomrJqr054YobOeHGOT0j49tztOvnZWeXuG6jb95HKK3SUGRYNAAAAILCSarbkHrWvJKSYUg1zyJf23cpEvTNvh/79q/+5Bati6tp9nl53NSm3sFjn/Gdehdvd9OEynwIc3j02k8uZUsvhtLLW9d/7XsVAz3Wfs8jh9Bu2LnIXLpGkd+YdHmYdqp6iDZm/KsyoH4IODY0xYcaYQcaYMcaYM40xHWuyYai8l3/drIO5RRr3XXDl1r39d9bhqkAnPDlDWflFSs8p1PNBzI14/+RVlT5fMG77ZBnDogEAAGrQuO/WBDXcD3XHuO/W6J7PVwRc75nTkNyj1oXVYBGaYneQFaj3XF0y8OnDw0L7PDpNJz9TfieUQHo9Mk0ZuYXaUc6w68Jip3o9Mk0vztislbsz9K+fy1Z2vvuzlRrgZ6hq6SGzWw5kqbgePL/1UX24b+FfREUbGGN6SXpY0tmStkpKkRQlqa8xJlfS25I+ttZyF4TQ8v3FWpNYtepRpTmcVsNfmOPptl2R6ev3V8t5S1u4La3CbZIy8nTahNm6ckg3vXTlSTXSDgAAgIZq8tI9kqRih1MR4QxCqg9KfmYBhWBOw7WJmWrTvIm6tWlW+yevQ8LcvTyzC4pr7ByhDIOD/YIhw2u+/CKHVZGj6tNdDXx6Zrnr+z72iyTpzbjtejNue5n15bX5wCHf3ovn/meemoQzKSjgLZh3Bs9I+kxSL2vtedba66y1V1hrT5R0iaRWkq6vyUaiYhNXV1wBuTKCDQxLrNiVrid+WKef1yRVaj9rrd95FIN12gRXifevVyQG3GbelhTFjp2qTfsPVerYyYfylV8U+A/csoR0DXhiRrlFZPxZk5ihTxYn6D8zt+hbd7tnbjigL5f5Vv/am5GnRduPbLjA1DX7lJBaPWEyqldeoaPc+wtAYEfydwOAf4Emkq/rdqfl6o24bRQu8KM2qydfPHGBhr8wR6nZBbr1o2WVfn9c10xds0/5xYGfv6+W79FlbywsszysBqrQFDucKix2+i1wU1js5N6vZkUOnk/AW4WhobX2amvtPOvn1cham2yt/a+19uOaaR7qi8vfXKyPF+/SfZNWqcjh1P7MfE34ZZNmbjjg2SY5K1/bkrOVnlPomVPi3fk7dMaLc7R5f+CJcUvfeo9MWavYsVP131m+83mUDmDenbdDsWOneoLMZQkH9b/ftmpDkm94mF1QLKfXvBVP/bRe21Oydcpzv+m2j5d7lhcUO3zeAF351mJlFRRr8Y5UTxtfmrFJ/5m5RVsOZOmlGZuUX+TQw9+s0TXvLtGGpENKzS7QJRMXavwP6/Xqb1v1f1/H67Xftur2T5br4W99h5WfPmG2rnn3d5345AzN35qiV2dtDfhN2W0fLy+zzuG0unfSSo16Oa7cb9jun52jv3+xSvsz82Wt1cGcQq1PCq769trETKVmHw6s//fb1qAmOQ7WhqRDen3ONknStuRs7cv0XwGtPknPKdTtnyzXceOna/C/yv/mFKhrvlmRqK+XV9DDpYZ9v2qvznhxzhF/qQLUVwmpOcrMLdLutFydPmG29mf6r/5ZWd+vrtwXv3XFzR8t1YvTNwesgtoYGXdXw4qypBnr9+uLpbvL3ygIW7ze+70zb4d+25SsL5Yd+XFDZd3eTN07aaU+Xh+4U8ZD36zRqt0ZkqTJS3frtd+2un4vK/GlVnpOoX7fEXhkVUkl5HP/M8/Tm87FKiE1R7Fjp6rvY7/omakbJSno9+8AUBkVDk/2xxhzuqQYSTP8hYmoXXVtotY+j/7i83jpI2fpi2V79MpM35Dvh3tP12L35LOTl+7WzA0H9Nv/jVRUk3Cf7XqMm6bIiDAN69FWz/9pgCb97noT4j0PoyT1e3y6tj57gZq4h9Y8O831B3STO5B8/Pt1kqT/zd6mLc9eIEnKyC3UwKdn6m9n9dE/zumrtYmZ+nBhgj50V9JasC1ViQdz1bZ5pK5/f6lW7DqoxePOVFTE4Tbe9dlKSdJjY47T63NcXeJf/c3VtpLHknTha/P9Pl/ez0t+kUPrkw7pDXdQJkmH8ot1/ftLPY+LHE69Pmeb7hzRS9GR4VqwNVWzNrrC2YXbUjWoe2u9PGOLWjdr4nOeFbvSdc/nK3XpoKM07oLjDh+/0PVB4fvVSRp3QT/PPJa3Du+h9xfsVMKEMZJc99n9k1fptjN6aFD3NpJc3yq3bxGp5Y+do7WJmfr3zC36aFGCxl7QT8sS0vXiFVUbMn4wp1DRkeG6eOICV/g5urenglpJeyRpR0q22jaPVKvoJp5JvyuSll0gh7XqGBNV7na5hcXaciBbA49uXaVrCOTxH9Z5wvTcQofun7xKXVtHa0Tf9moSHqaTY9se8TlKQmLv56q0OZuTtT05W7ed0bPcY2XmFWl7SrYGu3/mwbLWKjOvSK2bRarY4dTK3Rk6pceRXxuq16dLdmlknw6KigyTw2nVpVV0uduX9ES6cujRtdE8v5a4P2D9uv6ATuvVvtqPn1tYrOgm4TLG6MChfMVERahZZJXeLoXU7zvSdOdnKzTvodFqGdWk4h0qKb/IUebvNWrHqJfj1KllU1068CjtzcjTBwt36pELj6t4xyBYa/XTmn06//jOioyoH0OVSwpPFFdT76CSjzbBvq+oiypqek5BsZKzCnTnp655Ef9ySvcqnWd9UqbenbfDJ3AOd0/qV1zJzyd70nPVrU10yJ/3gmKHDrk7N6TnB3cNJfPJl/6sM3PDAZ3Tv5PnsdNpFRZ2+PqueXeJNu3P0s7nLyxz3a/P2aaXZmzWezcM1Q73qKFHp7g+y8zamOwzRP39BTsVEWb0tlchDwCoLkG9CzbGfCLpBWvtemPMXZLulJQk6SpJt9Rg+8prU19JH0tqJylN0g3W2q3l79Uw/Rhf/dW5qtMpz/3md/kfXz/cpf+jRQmSpB0pOerftWWZbQuLnZq/NbXCwih9Hv1FU+8frpdnbPYsW5Po+61bocNZpufd18v36ObTY5VVUHYoRelznvr8bL/nLvmW70j0e3x6hduUhLL/nbVVD5/fTy9MP1ys5tr3fg+43+VvLpYkvT13h96e6/9NhXfhm5LKY5e+vlBDjmmjjjFN3RXWUvTBTSerV4cWkqTU7EJtOZCliycukCSl5RTqn9+skaQyoeGcTcnq3q6ZenVoIWut4janaFdajm46vYdnmzWJGbpkou9wj16PTPP82/sN15n/nutZPv3vZ6hf55ZaujNdzSLDdcJRrfxe4xD3RMwJE8bojBdn6+pTuuvukb30Y3ySzju+s+dD8F8nrdJvm5IlSS9feZKmrErUQ+f100nuEPGDBTv1tHui5fLCOe/rOr5rK01ds89n+Y/xrjfab831nYPliYv7Kyaqia4Y0q3CY3tbuO1w76u9GXk6qnW0cguLdeMHS/XsZQPUt1OMJOnmD5dJUoWh4U0fLtWq3Rna/tyFng8CwRj+whztzcjTr/8Yoe9X7dUbcdv1/b2nVyqEnb3pgG79eLm+uvNUdWsTrc4to+RwWubcOgLvzNuu56YFLnDVo31zXXBCZz147rEKCzM6899xunV4D1077BifXsXWWuUXORUdGa4Vu9LVr3NLNW8aodzCYvUfP0MvX3mSTuvVTjtTc3R6b1ewF7c5ucqh+JRViQozRuf276wvlrk+KH20KEFPXnK8JCklq0BRTcIU4xWOTVmVqPYtmuq7lXvVKrqJxl/UX7d8vExxm1PULDJcfz2zj16Yvkn/PO9Y3Tu6tyRpV1qORr4Up+cuG6BrhnXXMPffr/jx56pVs8oFbwXFDoUZ4/kiK1jZBcVKPpSvnu7X2Kp67Pt1ysgt0vq9hzQ0to3enrtdtw7vqejIIw/6liWk68q3FmvMiV00d3OK1j55btAf9J1Oq6z84qCfT2utthzI1rGdY46kyQ3Gyt0HJbnm4XK6w6135u3wCQ1v+3i5UrIL9Nmtp/j8TnibteGAbvtkuU7r1c5n+a8bDuj+yav0x4Fd9fQlJ2jY87P05rVDNLpf9dY/jB07VdcO667HxvRXVJMwz/2zeHuaYts3q/ALDG8RVQyp/MkrdOi48dP1wDl9df9ZfY74eDVtyqpEXTaomx76Jl6JB/M06fY/SJJnXreXZmzWJ4sTNPv/Rum7VXvVrU20Rh/bUVe8tVgb9x0eeXPv5yvVpVWUOreK8rwviNucrDP6dAj4t39/Zr7GvLagzHLvc99yeg/9FJ+kC0/sohZNfT925hc59FN8ki46satemblZ787fqXEX9NMdI3rq6+WJunTQUXr8+3U6um207hrZSxHhYdq8P0uLtqfq5tN76L35O5RdUKyeHVpo78E83T2ql+fYb83drtN7tVePDs3LnLc8JZ0JSmw+6FSxw6nNB7LUMSZKHWKaltmnvKlmbv9kuc/jnu73s51aNtXMB0Z6Ojb0GDdNb18/ROcd39l13v1Zesn9Wea2UseQXH/zSiMwBFBTTEUdBY0xx0j6RdL5ck2r+72kv0pKlDTNvTzDWlu5CeOOkDFmtqQPrLWfGWOuk3SLtfbMQNsPHTrULl9e9kW3ITj2sV8837LWdw+c01d/6NlOV729ONRNQTW4c2RPT0B5So+2Wroz3e921w7rrvScQs1Yv1/Bvue//Yweenf+Tp9lp/Vqp0Xu3qtzHhylpIw8bd6fpWembtCSR87SlJV7PcHomAFdNHXtvjLtlKRZD4zQ2a/M83venc9fqCKH9Rkm8vyfBuiPA7vqxemb1atDc1059Gg1jQjTs1M3qk3zSM8bv6rq3bGFxl3QT0OPaavmTcO1PSVHyVn5OrZzjGZvTFZkRJge+CpekeFhla5MdvZxHXX/WX10YrfDYZ61VsVOq6U70z1B9MmxbfTZbcPU1KuX7Zw5c3TzDNcwnM3PnK85m5LVrU0z/bxmnycEfeHyAXpv/k5tTfatendar3Zq36KpJzSNaRqhy4d0U8uoCJ3aq70mLd2tX9fv93ltu2NET70zb4c2P3O+8godio4M92mP5BrSfu+klfr+ntPVqlkTHcovUsuoJsordCgzr0gLt6Xq+KNaavP+LDmt1WWDuikpI087U3M0e1Oywow07oLjFBZmVFjs1BVvLdKaxEz16dhCMx8YqZdnbNbEOdu04OHRmr5uv/588tH6buVeXXRiF+109wTILihWz/YtlF1QrHYtItWpZZRyC4tV5LBqHhmuycv26Mx+HXVU6/I/FO9Jz9UZL7q+tJj1wEj17thCr8/ZppF9O6jI4dQPq5P0xMX9ZYxRUkaetqdkq1ubZkrLLtDQUuHcjpRsn5A9GPMfGu05f8eYpkr28yHlizv+oL+8s0SSKzwf+dIc7UrzHZo1/qL+OqNPe53jnlz8rhMjtde017gLjnMV0rJWrZq5egz37eT68NeldbTP8xNoioUuraK0z2to5lvXDda783eqS6so/VwqoP/z0KP1ZYCh1aOO7aC4zSk+x33ruiE+X25N//sZem7aJj164XEqLHbq2M4xyncHg+HGaMO+TA06uo0c1uqBr+L1U3ySurdtpkcu7KehsW3VvkVTZeYWqWV0hIwxOvuVudqRkq3v7jldy3am68qh3dS6WaQumbhAaxIz9cFNQ3Vmv07anZaraev26cOFO3XHiF665pTuio4MV0JqjiIjwpSeU6iL/rdATSPCdOXQbvrH2X3VrkXTgM/ZnwYfpVeuGuizLLewWPsz83XDB0v16l8G6djOMWrRNEI7U3PUpVWUopqEq6DYocSDeerVoYXenrvd50smSdryzAUKDzMKDzPalpylZpER6to6WtuSs7Vub6YuHXSUJOnF6Zv0Rtx2rR5/jlo3i5S1VgXFTp9eiwXFDn22ZLf6dY7RrI0H9OHCBL1/41Ad2rVel53v+1ZvbWKmoiPD1buj68uoQoezzOuC5HptuOfzFUpIy9XaJ88NGKZV1QvTN6lbm2iN6NNBP8YnaW1ipv5+Th/161z2i9DylMwH9+Z1Q9TV/TuwZEea5/cskHVPnafs/GIdOJTvc9/ePaqXLh14lNq1iFST8DC1inZdd6D7o1eH5tqeUnYu5MfGHKeN+7L076sOfxm4bm+mCh1ORYaHKapJmK57b6k+uOlk9e/aUl8u263MvCKddVwnz5eMxQ6nip1WUU3Cfc7/yIX91KtDC20+kKUXp29WVJMwLRp7lto2j1RWfpEembJOP8UnKf6Jc9UquokKi52KjAjT9HX79OmSXdqTnqfd6bl689rBGnVsR08wbq3Vkh3pyt+9Rr1PGiaH0yq2fXO/122t1aLtaVqflOn5YmX8Rf11y/AemrXhgNo0j9Qx7ZppqNcXjyUcTisjKSOvSN+tTNStw3sccW+5lKwCNYsMV7HDqkmE8entfDCnUIMCTG8y95+j9PXyRE30GrFS3WY9MNIz+qOqrhzSTb/vTC93KO9VQ7vpq+WH5yy/bNBRmrLK1VHi9WsG695JK322//UfI/Tp4l1asC3V8/fY28Pn99Pofh20aneGxn23ViP6dtD/ndNXLaOb6PU525SeU6jZ7i+LK/Lpraf4jAIC0DgE01HkSMTFxWnUqFE1eo5AjDErrLVD/a4LIjS8UdK/JD0h15Dkf7j/bSSNk/ScpNXW2jXV2egK2tRR0hZJ7ay1DmNMuFy9DftYa1P87dOQQ8Ngq1gBAID6r3lkuHIKKaTUIaap3x43kivkituconEX9tNP8fvK9Ob+8b7TPb3arz6lu87o014vz9isE7u18hlqeclJXfVjfJJG9O2geVv8vsXUP887tlJfDhkT2uqnqBtKf1lRkT4dW5T5Ag4AUHsIDcs/wFuSekmKlvSTtfYFY0xzSb9Ya0dUa2uDYIwZIukTa+3xXss2SLrOWrvS3z6EhgAAAAAAAKisxhoaBjvJwz2SzpNUaK0tmaCunaR/VkP7aowx5g5Jd0hSp06dFBcXF9oGAQAAAAAAoF6p6TwpOzu7TmZWQfU0rGsYnuyLnoYA0Hh5D3XsGNNUHWKaan1SrU4zXGO6t21W7pxX9cHoYztoTpBDEI9qHa2wMGlPel652/3+yFnaeiBb65Myy8wtGIwm4UZFDqtj2jXzzEHZplkTHcwtWwzsSMz752iNeOlwMbGLT+qqo9tE64247erXOUb3ndlb901aJcl1Tde//7u2HKh4+GW/zjG66MQu2pac7RlK/ON9p6tvpxil5xTq9k+Wa33SIfXrHKNPbx2mtXszdMtHvu8B1z11nsKMdPU7S7Q1OVsPnNNXaxIzNe7Cfnr8+/WatdFV5b5ZZLhyCx16bMxxPgXPuraKUn6xU+k5hZpyz2m67I1FFbb7jWsHa2hsGyUezNO9n6/0mY8ToTGoe2slZeTpwCH/w9wDee6yAXpkylq/68Ze0E8Tgvi9/MfZffXx4gSl5xT6LG/drIky3L+LJfNBXzGkmx4f01+fL92l9+fvVFqpfQAANa+x9jQMZk7DnyS9I2m6tbao1Lqekm6SlGCt/aB6mhscY0ycpPe8CqHcaq0dHWh7QkME8s71Q3THpytC3Yw6x7tQSIm7RvYqMy+UJPVs31xf3PGHgJWyJSnMSE4rrX/qPL2/YKdemblFkjT1/uHKL3Jq4uytumbYMWUqzVXGXSN7aewF/SRJOQXFMkaKDA9TRHiYpqxK1D++jC+zT++OLVRY7NSEPw3QvZNWBv2h+a3rhmjulmRNXuoqrLBw7Jnal5Gn6MhwHd/Vt3Lz9e//rvlbU/0dpkKdWjbV6GNdxUq6+imcYa2VtdLrc7ZpV3quvllxeNLwFk0jlF1Q7Hl8Rp/2ev5PA/RG3HZN+n233rl+iM4+rpOnGnWJXWk5au0uTHHgUL5W7T6o80/oUubccXFxumm6a7Lx0n9ES16Xlow7S3+dvFLLEg76rH/j2sHKyi/S7zvS1b1dMzWPjNCwnm09c4y9dvUg3T95lW4+PVbfr9qrl688SWf266g+j/6iYqfVG9cOVv8uLQNOal9ascOp8DCjzLwihYcZDXjyV0nSqsfPUUxUhNbuzVSXVtFqFd1EuYXFateiqYodTsUnZngqj79y1Ul64CvXPRQ//lxtS8lW304ttGh7mk44qpW+X7VXK3cd1Ps3nSzJVYXzzbnbdW7/TurdsYUKip06mFOo9jFNg6rmuDYx01OV/OpTuqug2KHvVu7VnSN66qf4JCVl5mvn8xfKGKN7J63U1DX7Ar6ZWbX7YFChhrcXLz9RD30b3HTFp/dup09vGaYXpm8qU8ExUGg258FRWrgtVU0jwnRcl5a6/ZPluujELp4CRyXXEuzfuLeuG6JHpqxVek6hnv7j8Rr/w3qf9eMv6u+peF7aH3q21ZId/gs1SdKzl52gPel5emfedt0+oqcuH9xN5/6nbLGkhAlj9Mvafbr7c9dMKeVVDN+fma92LSLV51FXUaWZ/xihPu7q5ou2p6p3xxbakHRIN324THeP6qVz+ndSz/bN1bpZZJljWWuVVVCs2RuTFdUkXJl5hXr4W99AY9Xj52hnWo4Gd2/js9zhtCpyHC5EUvJ8P3PpCRrcvY0ufG2+q01jz1SnllHanpLt99oXjztTD34dr4TUXD172Qlq2zzSp8BSIIXFThkjnyrTD34d73ktS5gwRln5RYqJauL3zfTyhHSlZhf4fY3yVnJdl5zUVeMu7Fdhdd70nEKlZhd4Ks5Lrr8rxU7rKSbi7/jrnzpPP6xO0iNT1mpQ99aacs/pAc+xcd8htW0eqQ4tmuq5aRt1w6mxmrM5WU/8uF4Djmqld24YolOfn62+nVr4DVOP79rS8wWBv2Jjyx49W+1buO6XSyYu1NFto/XGtUN07+crPX/fEyaM8fs79vNfhysrv1hXv+sqwPLBTUM1vHcHRUa4fk7PTt2g1s0iPZXHS8zbkqIbPliqWQ+MUNOIcB3dtpn2pOeqfYumOm78dM85b/hgqXal5Wj8Rf3126ZkTfp9t9Y9dZ5OfmaWnr3sBKXnFOpPg7tJkga7i368dd1g3fXZSp1wVEvdOryHOsZEaVdark94d8lJXXXP6F76bWOyTuvVTuv2Zmrygo364I7RysgrVI/2zeV0SkmZeZ4CLd7WJ2X6VAR+/k8DdEqPtjqrVCGp0q+1eYUOpWQVqHnTcLVr0VQrdx/Un95YpLevH6JjO8Xo8R/W6e3rh/gUNCktPadQuYXF6tIqWt+s2KPLB3eTMUZvzNmmm06P9SneE+h1ccyALnro/GM18qW4gOepjDP7ddQb1w5Wv8ddP7vTe7fTm9cN0a7UXM/fp4p4F6wa3ru9Fmzz/36oQ0xTXTusu36KTypTjGdk3w76+JZTPNd9Vr+Oun1ET903aZVSsw8HvlFNwpRf5FTChDGaumZfmWIpFWnfoqnP8Wrb57cN8xSgA1D3EBoG3rmzpAckXS4pXVKKpChJsZK2S5porf2hOhscDGNMP0kfS2oj6aCkG6y1AWehbsih4T++XO2pJlafzX9otI5u20ySlJlbpJOe/rXazxEZHqY5/xyl0yfMlnT4F3/6uv16acYmPTamv27+aJnPPp1bRmn/oXxPz4yacsWQbmrXPFLNm0Z4ArXytIpu4qkQ3CwyXJv2Z2n0sR09b8pLGzOgixxOq/X7MnXdsGPK7Z3y8S2naGTfDpq+bp/6d2mlo9pEK9wdLK3bm6mL/ud6ozjt/jPUv6urMmReocNz7vvP7K1Pl+zSVScfrbfn7tD9Z/XRA+f0Dep5yC4o1rS1+7QzNUdvxrkCyvOO76Sn/3iCOrWMUkpWgR7/fp3uP6uPMvOKtC05S6/+tk3LHj2rwmqFDqdVr0emSZIWPDxaq/dk6KITu/psc9+klfp5zT49c+kJOrVXO/0cv0/3ju6lzLwiWUlzNiVrUPfW6t3R9UEyITVHnd3VRYMRTAAy5Jg2+urOUz3PeWUMeGKGstxBYcn9XexwKsLrA3l1KS80PJhTqOyCYs/v9PPTNurteTv01zN76//OPTbgMWPHTlW3NtFa8PCZOphTqDbNfQOS/Zn5Ss7KDyqMKM+BQ/lq0TRCzYMI70p+ZgkTxii/yKGmEWFHXBkzWGO/XaMvlu3xVNItcjgVEWaUkVukA1n5nsqsTqer4nXJB3p/dqRk66UZm/XLuv0Bt/nkllMUEW40rEc7hYeZgPdr/y4ttWHfIf381+EqcrgqCTeLjNDutFyNeGmOurWJVuJBV2+5ZY+erfAwo6veXqznLhug3F1ryn1DFDt2qsac2EWvXzNYknwq9b76l4H62xerPds+euFxumzwUWrfoqkkKSu/SHlFDnWMidK6vZnq36Wlerp/53c+f6F6jJvmc64NT5+nwmLX78clExfonP6ddMOpsdqZkqPr3v9dJ3VrpfEX99eQY3yrUZe0848Du+rVvwzSXyevUtymZK196jzNWL9fd366Quf076R3b/D73su3DUmHtCstRxcMKD/0qizvn90Zfdrr01uHBb1fTNMIrX3qPM/jZpHh2vD0+Z5thj03SwcOFXgKssSPP1etmlVfNeLCYqf2HMwtE+ocyZvp9JxChRn5DV2rw+xNB9S3U4y6tWmmXWk5GvlSnG46LVZPXnJ8xTt7KSx26tXftuje0b09AVN2QbFOeGKGJOm4Li21cd8h9e/SUmcf11Gvzd6ma4Z113OXDVDvR6bp2M4xswP+owAAIABJREFUmnTbHxQZEeapIlxaQbFDxz52OMDzvlc2P3O+rJXnb9qnixPUISZK55/QubJPSRk7U3OUfChfw3q2q9R+D3y5Wt+t2qv3bxyqs47rVGZ9Sfs/vPlkDe/d3ieAlip332zaf0jn/3e+urWJ1nf3nKaOMVE+55AU9M+1pv72Sq7XuldmbtG+jHxXFXq3ynzZEigwvum0WF066CilZBXotF7t1LxphM58OU43nharG0+L9WwXqIrzskfP1oz1+3XNKd09X0pu3p+lZpHhnvcEknTKs7OUnFWgFY+drXbu13DvY0dGhCm30KHlCeka3a+jopqEKyWrQAXFDnVrc/g4Jdfw1Z2n6pQeh1+rS16LS9w3urcmztnmE7CXfr52PHeh52+GJJ3aJVx3XzBEN3zgqpT80hUn6pKBXZVf5NRJT1Xv55SS995/eWdxuV9iAQidxhoaVvhpyVq7X9JDkh4yxsRK6iIpT9IWa23IxgxZazdJCu4dcAP3t7P61PvQ8Nz+nXzeSBzJh49W0U0U/8S5yi4o1hkvzNa3d5+mM93fEG959gJJUq//Z+++w+Qqy8aPf+/dTa+kbUgljYQAoYUSEkKAEHoRKSIgiAgoWEDwxYpiAXvvvfeCBbHnVSzwij8FBRVFRCwIKiIqIOT+/XHObGY2u5vdyW5mNvv9XFeu7Jxzdube3WfOPOc+z3M/U8fwl6oVFw/fZTqH7zK9Y2XEyqi4CLjmGQcwqUxc3H7PAx3P1VnnqUs9ufCghbzve7/tWP1ycfs4Xn3Sbh37j999ZseUrk+cux+nvLO40//rlx1BSwRv+tavOWrZdCaNGd4R2/zy4qqr0YBvPW1Pjux0Qbr77Im85uu/6ug4vfDopTxmj5mMHNbacaHR1ciNXWZO4AVHL+WARVNqRmBUciiLpo3l4nWLubhMDD3niJ169TupGDuijZOXzyYzWTB1LMftPqPmAmDquBG8/Yy9Oh6vWDCZM1bs0Kvnbm0JbnjuITyayfYTRtV0OiuuPGFXDtlpGo/Zoxjh8Iy1iwA6OrQnLZ9dc3xvR7pVnLHfXD70w9/VbPvhcw7hwo/+mKOWbc8TV87r0/N19n/PX8utf7q/5mcbqIuWnmw3ZnhNwq+SAN1ccvX7lx3c0aY7JwwBpk8YyfQJI7c4vvbx9T1Hb5PD/eWqxy7jBUcv7UhuVt4LnX+/LS3B8M0kmedPHctbT9uT937vDk5ePqtm1Mp1t93LUz58I/vOn8SIto0/4w3PO4SxI9poiaC1JTjnAz/ipcfvws//eD/nf/hG5kwezfiq56mcO/aZN4mF0x5m/S/vYfyoNka0tfKNiw8EYH1t899E5w7ZeQcu6EgaHrf7zJqk4ZNXz685dtzIYR0/1y4zixG/zz1yCS+/5hc1id4zV8zluD1mMnp4G5Uc0reetaZj/4wJI3nTqXuwbuf2mt9HtdtedgSt5XO+6dQ9OravWTyVE/aYybMO6z45Xm3pjPEdN1/609HLtudLN/2Jr120uuZcvTnXPvMAJo/ZeAH/w+ccwshhteeQ0/ady2u//ivOXjWvx5sA9Rre1tLlKLAtMamL80l/OnjJxmTW3MljuObpB7Cove8/w/C2Fi49bEnNtjHDWzvKEHzhwpU88OAjjBreypu+dRsA25fns1+//Egyc7M3NUa0tXY7erFze+/t52tvzJsyhnl9/MwEuPzYnZk/dQwHLZ7W43GzJo7aJGHYV5X39Ii2lo6EYWdPO3hhl9s7G8jP3nEjh3H5MUXisqvE3/4LJvP93/x1s8/zv5eu6RiVOLythZcctzOn7D1nk+O+dcmaTbZ19RkNRT/t9P3m1mxbPH3Tc9ANz1vbbVyV5x4zoq3mhsrUcSM2Ofa2lx3B9bf/rSZhCLB2p3aesmYBb1v/G47cdXrH+/G0fedw2RFL2LBh04EALS3BHVcdxdd+/md+99d/s+DR37F6x6kd+2duN4oRba0175PKTbSufPr8FVzxpVu46a5/ALDPDpO44Y6/MXvSKD553gpWXFkMYvjZiw/rmIFw+TE788Krf7bJDA1JapTNJg0jYkmZoAP4U2beUbVvv8z84UAFp97pa9JiIFRPgzx2txnsPnsiV3zpFnZsH8u7n7A3Y0e2MW5kG488mh2j0b544Sr+cN9/OP/DN3J5F3dsrzxhV970zdv4Yzml4ZPnraAlYNdZE/jur+5l7dJ2lr7wWv5dJt5+e+WRvOM7t3PCnjM7Yvp/L1wHFLWSJlYlIr/5rDV0Ncq2ktjYb/5kPvrk/TbZX0nM7TtvEjtMHsMnflRMTa1c5FaShpcetpg/3vcfPnL9nR3f++uXHcHC532FfedN4pLDFnPJYYu57rZ7Of091zN5bG3Ha87kjQmf6jvylQ5oJZHVlcrPdcm6HTnvwAW0Rmwy/bTyvJ88bwWXve/rnHboPuw6a8Imx3TnSas2TWyNHNbK20/fiz3nbtkIsIqI4MS9ZvXLc1Wbtplk0biRwzoShgPhhccsZcHUMbzoi8U0yZ9evo4Jo4bx6afs3y/PP3JYK3t0mn7YDM7afwd+ePtfOblT0rWzrqZgN9otVxxGI8v/9mY0ZG9FRJfv31WLpnSMLKvW+aL5A2fvA8DsSaO7vNs6ddwIrn3mAcybMqbbZFs9XnzszowqE7anLJ/NJ370e648Yddefe+5qxdw7uoFtc933C49fk9EcMxuM3o8prvkxIi2Vl57yu69im0gvf6U3Xn6IYv6lDAEOkavVvRHkn4o6s9EcETw2ys3vt8qCZVKzqP6M763o6C//PQDeLSLpEkzmjBqGBce3H2/p6I/bupUfpfd/WoGepRJPc47cD6zthvNiXtu7Lts7sbYoUuLJPfcyWM4dZ/ZfOyG33PT5eu2+o2x/jCstYVVi6Zssr21Jfifw5dw0dodaWsJIoqfd7dZEzb7Plm3czGydv36oh//2yuP5Od/vL/jZlS1Lz5tFXff/yD7l7OYqi3fYRJXX7CyY5T72at24IY7/sZT1yxk+wmjOGzndr7687trSpbstP14PnX+/h3J4MfuOYvP/Lgo1zB+ZBv3P/jIJq9TXZ9W0sDpzY25bVFvrkQ+CuxZfv2Dqq8B3trpsYaoo3bdnlecuKzjjXTtz4paOYfs1F6TAKvui+w6awK7zprQbQfs1H3mcOo+c/jrAw/R1tJSM/pwbdnZWb1oase0jIjg/AMXdPlcXXWeunrDL92+6OR3dVFd8cuXHk5bSwutLcGE0cM6LmSrVWr8PHHlPNa+thiZ2Nbaws0vWldzId3bc053IwK6UunnDmtt6dUd98PnDetTwrDH5+qH6UvbumGtLZy1ch4n7z2bILqdPratmTZ+JJ/tobZXM+upBpU21Tnp1B+qp8RVLuoHS8KjUdpaW/qcMOytfcsRPSv6OM1U/esJK+by3dvu4aTlfb/RVfQRarfNGKQJ4ksPW8yrvvrLLkeh9VVlpGHn88vzj9qJ8V3UsmwGXc3oGF72/646YVfuf/C/vPyajSVpvvS0VTXJr5cctwuXHrakXxKGAz2itx7VZTu6qzH79YtWb1JHsVpEdJkwhGJ20oyJo3j24Yt55bWbVsqqvt44bOfpfOYpKzpqy77p1D35x396rqW9YNrGwSEHLJq6Sb3xV564jC/f9Cd+99d/19ReVq3qmVtSvYZiwhB6lzSMbr7u6rGaQG9XbeuN4a0tPPzohs0e94JjlgIb30iH7Ty9o+5HZz9+waF9iqFznZNq9dR868l2Y4Zv9i5yddLvuUfWdtTec+bymmmhC6fVTk2qng5YbXOjmD5x7gp+18sVRJeVCcCBmO6m/mMiSqrPcbvP4GM33Ml+8zetM6itY9/5k/nFSw4flKOStiXbTxjFl552QL8932GD9MbfBQct3GRBlnq1dnNT4pwD5nd1eNNL4NjdZvK+793Beavns3yHSZskv9paW/ot2bf+0jX98jxb26L2cR0LUfVV5drnqWsWdpk07HxsdY3c4W0t3Sa7zztwPu/439sJgs89dX/+9dCj3Pqn+2uShpU6jl/8abGKfE/XTNuynmp5Lm4fx1cvWg0U9X3rXZhQGsp6c9Wa3Xzd1WM1yDP3HMHrf1zU6Dtt3zl87sd/4Jd3/7NX33vAoim0jx/ZsVJh5UMK4NmHL+5Vnb7OK4FGxCa13yr68y7kSctnbXLHrZG6KtD99YtW19RPrFZJeWYXb6VvX7KGfz9cTEHoXL+sJ0cvm8FusybW1IiUpG3FfvMnN+UUwaHGhOG24+oLVvLO797e5xrE26LKSObNLRTZ7KoHw0yfMJIfPOeQAX29BVPHsHj6uJoat9pCVU2wUnbmF3+urZ1YqeP4xJU78N3b7mVnBwxs4m2nb5wU+dg9Z5k0lOrQm6ThrIh4I0V+o/I15eOZAxaZ+mT3aW3ccdXGgsJ9Scy94rHLmD5+JGt3auf8D9/IMctmcPSuM/jLPx/kkJ3aWbd0Oi+/5taa1dmqveDopVscf73WlEWxj2jiu+M93b3cc+52rFk8lecduWlHvZ5i4RUmDCWpa1OG6EgMqTu7zZ7YsVr5ULf9+JE8bu/ZnLFi7uYPHgS2Vu7zm1ULSQ1l40a08c+yxvt2W7iifOVPV50AroxqHDuijTeeurF27sFL2jtuplUWTWpGcyaN5s5ezpyquP3lR/Lya27l3df9tq7XrC7V1NUgjWrVfz9JG/UmaXhp1dc/6rSv82M1iYOWTOUHt29+1TTYuPDA4btM7zR6o5i+MGfyaN5+xl785Pf3cfxbvrfJ96/dqefV7Aba7S8/ste1AZvNyGGtvP+J+zQ6DEkaEq5/7iFDpo6opL5raQmueuyyRoexxQZrv3iw++YlB/LLP/+TM95zQ789Z1d1wk7ca1bNiu3VhrX0rrRUT3afPZGf/P6+jse7zZrAT8sVoOt1xn5zmTFxFK+4tucSWq947K4Ma21h7x0mMXnscFpagucfvbTbpGGlHn1/OG6PGXz4h3du/kD12Yi2Fh56ZMvapRpn86skwCeAazLzA9X/gGvKfWpCTx6A2ivVxYO/8oyNNXSiwaUtW1piyBYllST1Xvv4kU6fkzRkbG5klfrXtHEj2XlGMehiS3/zXU2Rr5Ry72n6fF/rvb/hcbtvsq3zjKX5UzfWaD93dX3XmCsXTqY3oZ2y9xxO2HMWsyeN7lX973Ejt6xG+P4LNi7otWxm1wvlDDXPXLv51er76rgu1jkYaPOn1j9rT7V6kzR8I9BVleVVwOv6Nxz1l4FOok2oWkHOfJ0kSZLULLZO53zlwqG7gvrrTtmtywW5+us3X6mvWZ0ErFzfbeghI7nT9n1b0GXJ9E1H6lXXRjxq2fY1ScqpPZT4mNjjlOxg/wVT+hRbte8++6Aut28uSbq5Vc+rv/+k5bP48tNXcf6BC/oeYJV95w3OhdpeeeIyXnL8Ljx+3zn9/twbEg5ZsnVnJ86bbNKwv/QmabhXZn6288bM/Bywuv9DUn+bMra2vuHyudtt8XOaKJQkSZKa10DXtvvAE/fh1isOH9gXaVKP2WMWHz93xSbbKyUwHr/PliVeLjxoIU9YMZfT99tYX7NjpGEP4xgvWbd4s8/9njOXd3y9ePo43vz4PWr2r1o4pSPxNXXsiJpXO3vVPN506h4ctev2Nd9z2r5zOHHPWT2+7q6zJnS5/duXrNlszN3Va1+949SOr++46ig+eHZR9mnlwsn87MWH1Qx06WyfeZN48bE78+4nLOfHLziUiGDnGRMY0dabFEn33vmE5Zs/qAmdvHw2Z+w3l2njRnLTi9b163MPxOjFzTFf0X96847oaUWFLXtHaatoHz+y5vHyHfr37se08RaVlyRJkprB1rpYbmttsU5sJyOHtfLrlx3BpYdtPnnXk3Ejh3HFcbvUrFS/d5nIO3Rp9wtAtrVu/vL8kJ1q6yEevWzj1NGbX7SOXWZO6EhWzpg4ku1GbxyA0toSHLPbjJohlfOnjuFlj9l1s6/blS9cuLLPiz++uyop191Aw5YIxo6onbq8d9U18PC2Fj553grmTx3L2qXtNYuInrFi7haNFhw1rPneEz96/trNH1SlupTLNy4+sMtj3vz4PThr/x06Hs+cOIpnHbojO7YX09l3qqo3OWu70Q04V3R/IlwyvW8jcoe63iT9/hIRm6zUEBF7A/f0f0jqb2M6nTCrOxIvPnbnup6zUsdw2rgRjGhrvhOjJEmSNJRZ0bAx2lpbBqRU1JLp47njqqM4sGp0XWdzJ/c03mfzxpXJoqOXbc87ztiLJ62a3zHF9+kHL+w4rvLTPXPtIj731JVAz+2tu1/HsllFHcEXHr2Uiw/dscfYnrhyB95++l6sXdrOOavmlXH0/vc8a7vR/PKlxcjYnupCThk7gk+ct+ko0u4snDaWJx8wr+Px8LYW3rZ2y/4O/W1KD9PKAT70pO4X5lw4bWyX23fafjzDWjf+/ttag6cdsqgjN3DVCbWJ5Mljamc/1qt69OjHnrxft8f19Ba89plOmO2L3iQNLwU+GREviohjyn8vBj5J7crKajJvPLUYan7mih1qtle/f3aZWd+KU9ExPF6SJElSs3BW3tDVeYZZvSKCw3aeXlPzr6WLYX3zp47dZArwfvMnMXPiqNrn28zrnb1qHk8/pOcprJcfszOH71KMsqz3GrSlvIitd+r+fvMn8Zg9ZtZs+8bFB/K8o5bWbBvVNrjehQcs6j4R3Z0FU8fW/B7bxxVtrzKic1KnJGF/5Q3edvqeHV/3VNJyux5rbKovNps0zMwbgH0p3utnlf8C2Dczrx/I4LRljt1tBndcdRTbjem5KG09BtdpUJIkSRpiBrqooXrlU+ev2KRuYLN59uGLec1Ju3W9s4/taO1O7XzvsoP7IarurVpYLKqyZx9r9VeuYTdswXvjpcfv0vH1bt3UaRxqdp4xviOZ95wjl/D1i1ZvUofyyKo6mPXUTKxMGa8eXdpVIhvgvWct7xg1qy3XqzXKM/Nu4PIBjkUDpdM5sXqobnfDjXv91PZFJEmSpKbhAgDNZe9+rie/JSr15jp76pqFXW6HjZeSfZkK3Fl/T9c+aMk0br3i8D7XyWttCXabPZHzVs+v63V3mz2xpvRXd0mroeb43WcyuZwCPay1hUXtRc3AQ5ZMY0Y56nS/+RtXWx9fR0Jv5cIpXP/bv7H9hI2jabv77Y8dMaxj3wUHLeBrP7+bO/76L/77aO+SF8PL2qAPP7qhz3FuizabNIyIm+l6NGkAmZnL+j0q9aue3ho9rSjVlfWXrCk6Ih3vULOGkiRJUrOxlz50TR8/kj/f/+Am21dUJW76anN5v5ULJ/Oe637LHnMm1v0afdFVwnB4ufJxd0mpiODqC1bW/ZptnZKELQ3K0H/paas4+k3X9XjM9PEjef7RO7HzjP4dDfmNiw+sSdz15D1n7d1vr3vhQQt57F6zaqa+bz9xFO85czk33PE33vG/t3dsnz1p4zETRg3j6+ViLjtc9uVevdahO7dz6x/v5/Z7/9VP0Q9uvRlpePSAR6EB1Xn49ZbcJdqhXN3qnn8+BDjSUJIkSWomW9LX1+B3w/MOYdSwVnZ90dc6ts2bMobf3vuvjqTa1Res5Nu//Euvnq+n673qBUUOXtLOLVccxujhtSmGU/eZw5rFfa+ZV499503ieUfuxMnLZ2/xc40b2cY/H3yE5x+1Ey/98q0A7DG7djp0o95p40ZuPo3zkSfvy4KpXY8sbWsJHtlQ/O0+2YdFX6BYUbvyNz5oyTTefd1v2WcLVpuuOGrX7fnyzX/qdn9LS2xSK3PmxFHMnDiKQ3Zq70ga/uIlhzNyWOvGNRjqyFe89uTduPKaX5g0LG22tWXm77raHhEtwKlAl/vVPCaOqi1C2h83RJz2IEmSJDUvb+4PTdPKBSledeIyLv30TQCctHwWr7z2lx3TaXebPZHdZvduRGCWY1arL/+6m27cOWEIcGWnVXQHUkTw5DqnHm/yXOX/J+41qyNpuHZpe80x3SXlBtrIYZuflt3d5fqnzl/BzImj2P+qbwF0m/A7eMk0TtprVs22lxy/S83feOXCKfz2yiPrmn7+08vXQcLf//0wf/nnQ+wzbxJfrhoJOH5kG/c/+Eifn7fyu9mSKfEj2lp5/lE78eTV81lZ/p6Gss0uhBIR4yPiORHx5ohYF4WnAbcDJw98iNpSu86awAfO3ocLDlrQb885afRwTthjJu86c3m/PackSZKkLePNfUHtwhMVW1aXcEuiGXyiFystt48fsZWiqX3NLVkle+8dJnXUGezJe8/amyM6taHT952zyXH1JucmjBrGhNHD2GHKmC4Tl9X1Iis1Biueffhi5k4e3flbapy8fDYjh7V0+T7ojbbWlk1GNg5VvZme/CHg78APgHOA51Ikro/PzJ8MYGzqRwfuOJWf3Hkf0D/DqFtagteesns/PJOkweyzT92fv//r4UaHIUmSOkmHGg5po3oxGq03dphclKeaU/4/VPQqF9bLhNnYEW088FDfR81VO2GPmZx34AJ2mLIxWXbBQQt4y7d/001o/Z/lHYjn7E716evrF6+u2ffUNQt7XLwHigVff/GSIwYitCFnsyMNgfmZeVZmvoNiOvJS4DAThoPPsbvPAOC4PWY2OBJJ24o952zHITu1b/5ASZK0VQyxAWHqRn/ld07caxafPn8Fxyyrb8TWYPWklfMAGD2i++Rrb3/FP3vxYVs86++Fxyxl8fRxjGgr4rnjqqO4+NDFW/Scg8XcIZawbja9SRr+t/JFZj4K3JWZmy7FpKY3b8oY7rjqqIbVXpAkSZI0sM5aOY+ZE0fVPS1PqhYRLN9hUs0os8rqueNHdb1K8bbgaYcs4o6rjupI0nWlL6snb+nA34mjh2/+oCr9efPglOWzmTymb6/flSXTx/X62L6OlP7keSv4yjMO6GtIHZbN6t9VprclvZmevFtE3M/Gdjeq6nFm5vgBi04DZsX8yf2yypEkSZKk5jFvyhi+d9nBjQ5DTeSQJe288tpfclQ/JZKftW5Hdp4xnjU7bp1VkZvVUKnz+IoTl/XL81x94UoeeXRgyibUm9t48+P3YOKo4axcOJl5z7mmn6PaNmx2pGFmtmbm+MwcV/5rq3pswnCQ+ti5+3HRoTs2OgxJkiRJ0gBaPH0cd1x1FLv202iqEW2tHLf7zK1a464ZtfThx1+xYPLABTJIjGhrZcyI3oxbg5OWzx7gaAoj2lpZtWhKt235knXmTDb7F4uIkcD5wELgJuC9mbllVTwlSZIkSdKAcB2cgTdzu42r637totU9Tgk+YFH/j8psbQleePRSrvjSLZvsG8z53F+85HCGt7bwnut+2+hQmDfF0m69qWn4AWA5cDNwJPCaAY1IkiRJkiTVrZI0Ortc0GNr++KFq7j0sG1zoY4XHr0UgGOWzejYtmP7OBa1975mX19M6qGe4NmrNv5977jqKGZPGtXtsc3um886kM9fsJKRw1pp6cswzgF05K7TGx1Cw/VmbOjSzNwVICLeA9wwsCFJkiRJkqR6RQS3v/zIho0423XWhB6nQ69cOHin6569al5Nsq633v2E5ZzzwR/1+ftufP7aPn9PDMJ11Dsv2HrUsu358k1/GtDXHNHW8zi6oT4FH/q+erLTkiVJkiRJanItLdG0SY/zVi9odAgDqrWLkXJrl7bX9VzN+jccaG983B7ccsVhA/oai3u5ovPmkovbsr6sngzFismunixJkiRJktRE3nbaXrzzO7fX1FsEWLN4Kut/eU+DoqpPa0swenjvFk4ZSB998r7MnTym0WE0zGb/ApnZujUCkSRJkiRJGuyyQSvR7DJzAm88dY9Ntr/7Cct5+NENDYho8Nt/wZRGh9BQjU/bSpIkSZIkbSP6K2X4yscu65fnaWttoa116E6xVf1sNZIkSZIkSU3m5L1nNzoEDXEmDSVJkiRJkprI8bvPaHQIkklDSZIkSZKkZjJUV01WczFpKEmSJEmStJV94cKVjQ5B6pFJQ0mSJEmSpK1s2riRDXttBzKqN1w9WZIkSZIkqZ+0tQT/fbS/1lDuf9981oGMG2k6yLzp5jnSUJIkSZIkDaiVCyc3OoStpqWXw/hGDmtMSmbB1LENHeWowcOkoSRJkiRJGlAfOWc/Dlg0pdFhNJWJo4fz4Sft2+W+Pedu16vnePvpe/GVZxzQn2Ftk150zFJeeeKyRocx6DgeVZIkSZIkqQGW79B1cvD0fef06vsP32V6f4azzTpr5bxGhzAoOdJQkiRJkiRtNc1b7a85zJw4inClEjUBk4aSJEmSJEkNMKKthcf3clShtLWZNJQkSZIkSWqAiOCFRy9tdBhSl0waSpIkSZIk9ZO+Tr8eOax1QOKQtpRJQ0mSJEmSJA0tlo3cLJOGkiRJkiRJ/WxYq1kpDW4mDSVJkiRJkvrZV55xQF3fN2Xs8H6ORKpPUyYNI+L9EXFXRPyk/Pe8qn3tEfG1iPhVRPw0IvZtZKySJEmSJEmdzdpudF3f964zl/dzJFJ92hodQA+uysw3d7H9SuA7mbkuIlYBH46IHTOzr7VGJUmSJEmSmsq0cSMbHYIENHfSsDsnAzsAZOZ1EfEQsBz4v0YGJUmSJEmS1Oflk6vsNmtC/8UhbaGmnJ5cujgibo6Iz0fETgARMRmIzLy36rg7gdkNiVCSJEmSJKkLUcc6KFdfuKr/A5Hq1JCRhhHxY2BON7vbgecBf8rMDRHxBODaiJhfx+ucC5wL0N7ezvr16+uMuPk98MAD2/TPp4Fj21G9bDuqh+1G9bLtqB62G9XLtjMw/v63BwG46ac/Jf84uCY+9rY9PPDAA2zYUGQLv/Od7zCspfvMYVfPabvber7//e8zcUTtWLpG/f6b9ZzTkHdpZu65mUP+UHXsByPidcCszPxdRBARU6pGG84Bft/N67wTeCfA8uXLc82aNVsBrKNWAAAgAElEQVQefJNav3492/LPp4Fj21G9bDuqh+1G9bLtqB62G9XLtjMw3vOb6+Gv97Jst904cMepjQ6nd679MkCv28P69etpafkPPLqB1atXM6KtdZPnqqh5zj6+jrZA+bvef//9N9aPbPDvv1nPOU05PTkiZlZ9fRjwKBsTiZ8Czi/3rQJGATdu7RglSZIkSZI0OAV1zB8fYpp1PPAHIqId2ADcDxybmY+U+y6jWDH5TOA/wBmZuaFBcUqSJEmSJEnbnKZMGmbm2h72/Rnodr8kSZIkSZKkLdOU05MlSZIkSZIa7aK1OzK8rW+pkyQHKBpp62rKkYaSJEmSJEmN9oy1i3jG2kV1fa818zTYOdJQkiRJkiRJQ8KUscMbHcKgYdJQkiRJkiRJUg2ThpIkSZIkSRoS0pKTvWbSUJIkSZIkSUNKVJWcfPrBC5m13ajGBdOkTBpKkiRJkiT1k0OWtAPQ0sM6KCvmT95K0ag3Ll63mOv+5+BGh9F0XD1ZkiRJkiSpn7zh1N356wMP09bqOC0NbrZgSZIkSZKkfjKirZUZE53qqsHPpKEkSZIkSZKkGiYNJUmSJEmSJNUwaShJkiRJkiSphklDSZIkSZIkSTVMGkqSJEmSJGlIyEYHMIiYNJQkSZIkSdKQEo0OYBAwaShJkiRJkiSphklDSZIkSZIkSTVMGkqSJEmSJEmqYdJQkiRJkiRJUg2ThpIkSZIkSVtRuoavBgGThpIkSZIkSZJqmDSUJEmSJEnSkJDpKM/eMmkoSZIkSZKkISUiGh1C0zNpKEmSJEmSJKmGSUNJkiRJkiRJNUwaSpIkSZIkSaph0lCSJEmSJElSDZOGkiRJkiRJkmqYNJQkSZIkSZJUw6ShJEmSJEmSpBomDSVJkiRJkjQkZKMDGERMGkqSJEmSJG1Faeaq4aLRAQwCJg0lSZIkSZIk1TBpKEmSJEmSJKmGSUNJkiRJkjTgjthlewDmTxnT4Egk9UZbowOQJEmSJEnbvlP3mc1j9pjJqOGtjQ5FUi840lCSJEmSJA24iDBhKA0iJg0lSZIkSZIk1TBpKEmSJEmSJKmGSUNJkiRJkiQNCZmNjmDwMGkoSZIkSZKkISWi0RE0P5OGkiRJkiRJkmqYNJQkSZIkSZJUw6ShJEmSJEmSpBomDSVJkiRJkrYi1+LQYGDSUJIkSZIkSVKNhiUNI+L0iLgpIh6JiAs77RsdEZ+IiF9HxC8i4uje7JMkSZIkSZK05doa+No/AR4HXNbFvkuA+zNzYUQsAr4bEQsz84HN7JMkSZIkSZK6lOnk8N5q2EjDzPxZZt4CbOhi9ynAO8rjbgN+BBzRi32SJEmSJElSj4JodAhNr1lrGs4Bflf1+E5gdi/2SZIkSZIkSdpCAzY9OSJ+TJHg60p7Zj46UK9dFcO5wLkA7e3trF+/fqBfsmEeeOCBbfrn08Cx7aheth3Vw3ajetl2VA/bjepl21G9ett2/nHffV0eZ7sbeI888ggA133vOsYMa47Rhs16zhmwpGFm7rkF334nMBe4p3w8B/h2L/Z1juGdwDsBli9fnmvWrNmCkJrb+vXr2ZZ/Pg0c247qZdtRPWw3qpdtR/Ww3aheth3Vq8e2c+2XO76cMHEia9as2GSf7W7gta3/KjzyCKtWrmLC6GGNDgdo3nNOs05P/hRwHkC52MnewLW92CdJkiRJkiRpCzUsaRgRp0bEXcBJwEsi4q6IWFrufhUwMSJ+DXwJODcz/9mLfZIkSZIkSZK20IBNT96czPwY8LFu9v2LIpnYp32SJEmSJEnN6KDFUxne1sJXf353o0OReqVZpydLkiRJkiRtM973xH144sp5jQ5D6jWThpIkSZIkSRoSstEBDCImDSVJkiRJkraCZbMmsHDaWC47YkmjQ1E0OoDm17CahpIkSZIkSUPJ6OFtfOPiAxsdhtQrjjSUJEmSJEmSVMOkoSRJkiRJkqQaJg0lSZIkSZIk1TBpKEmSJEmSJKmGSUNJkiRJkiRJNUwaSpIkSZIkaWjIRgcweJg0lCRJkiRJ0pAS0egImp9JQ0mSJEmSJEk1TBpKkiRJkiRJqmHSUJIkSZIkSVINk4aSJEmSJEmSapg0lCRJkiRJklTDpKEkSZIkSZKkGiYNJUmSJEmSJNVoa3QAkiRJkiRJQ9kHzt6HaHQQQ0Q2OoBBxKShJEmSJElSAx2449RGhzDkmKTdPKcnS5IkSZIkSaph0lCSJEmSJElSDZOGkiRJkiRJkmqYNJQkSZIkSZJUw6ShJEmSJEmSpBomDSVJkiRJkiTVMGkoSZIkSZKkISEzGx3CoGHSUJIkSZIkSUNKRDQ6hKZn0lCSJEmSJElSDZOGkiRJkiRJkmqYNJQkSZIkSZJUw6ShJEmSJEmSpBomDSVJkiRJkiTVMGkoSZIkSZIkqYZJQ0mSJEmSJEk1TBpKkiRJkiRpSMhGBzCImDSUJEmSJEnSkBKNDmAQMGkoSZIkSZIkqYZJQ0mSJEmSJEk1TBpKkiRJkiRJqmHSUJIkSZIkSVINk4aSJEmSJEmSapg0lCRJkiRJklTDpKEkSZIkSZKGhIMWTwOgrTUaHEnza2t0AJIkSZIkSdLW8NpTduOyI5Ywoq210aE0PUcaSpIkSZIkaUgY0dbK7EmjGx3GoNCwpGFEnB4RN0XEIxFxYad974+IuyLiJ+W/51Xta4+Ir0XEryLipxGx79aPXpIkSZIkSdp2NXJ68k+AxwGXdbP/qsx8cxfbrwS+k5nrImIV8OGI2DEzc6AClSRJkiRJkoaShiUNM/NnABGxoY/fejKwQ/kc10XEQ8By4P/6NUBJkiRJkiRpiGrmmoYXR8TNEfH5iNgJICImA5GZ91YddycwuyERSpIkSZIkSdugGKhZvRHxY2BON7vbM/PR8rj3Az+qnoocETOBP2Xmhoh4AvASYD4wEbgzM8dUHXsN8O7M/GwXMZwLnAvQ3t6+18c//vF++dma0QMPPMDYsWMbHYYGIduO6mXbUT1sN6qXbUf1sN2oXrYd1cu2o3o0st0cdNBBN2bm8q72Ddj05Mzccwu+9w9VX38wIl4HzMrM30UEETGlarThHOD33TzPO4F3AixfvjzXrFlTb0hNb/369WzLP58Gjm1H9bLtqB62G9XLtqN62G5UL9uO6mXbUT2atd005fTkcqRh5evDgEeBSiLxU8D55b5VwCjgxq0doyRJkiRJkrStathCKBFxKvAqYDvguIi4DFiXmbcAH4iIdmADcD9wbGY+Un7rZRQrJp8J/Ac4IzP7upiKJEmSJEmSpG40cvXkjwEf62bf2h6+789At/slSZIkSZIkbZmmnJ4sSZIkSZIkqXFMGkqSJEmSJEmqYdJQkiRJkiRJUo3IzEbHsFVExD3A7xodxwCaAtzb6CA0KNl2VC/bjuphu1G9bDuqh+1G9bLtqF62HdWjke1mbmZO7WrHkEkabusi4keZubzRcWjwse2oXrYd1cN2o3rZdlQP243qZdtRvWw7qkezthunJ0uSJEmSJEmqYdJQkiRJkiRJUg2ThtuOdzY6AA1ath3Vy7ajethuVC/bjuphu1G9bDuql21H9WjKdmNNQ0mSJEmSJEk1HGkoSZIkSZIkqYZJQ0mSJEmSJEk1TBqq30XE9IiY0ug4NPhExIiIGNnoODT42HZUj4iYEBHjGx2HBh/bjuoVhdZGx6HBxXOO6mUfWVvKpKH6VUQ8FvgMcE1EXBURJzc6Jg0OEXE88HHgmxFxRkQsanRMGhxsO6pHRJwAfBr4dkRcFBGrGx2TBgfbjuoVEccAHwA+ExFHRcSYRsek5uc5R/Wyj6z+4EIo6jcRsRi4GjgBGAkcABwMXJOZ72hkbGpuEbEfRSf6bGAOcCLwa+CzmXl9I2NTc7PtqB4RsTvwCeAUoB1YB0wEvpCZVzcyNjU3247qFREHAu8ALgVmA+cAHwU+nZl3NDA0NTHPOaqXfWT1l7ZGB6BtynbAXzLzFoCIuBO4Bzg5Iv6SmZ9raHRqZvOBH2bm94DvRcQvKDrTj4mIu+1Mqwe2HdVje+C2zPwJQETcBpwMHFN+Xv2godGpmdl2VK9dgS9l5hcBIuIW4GIgI+ItwEPpaA5tynOO6mUfWf3C6cnqT78A7omIYyKiNTPvBb4OfB/Ys7Ghqcn9BRgZERMAMvP/Ae8Gdgb2bWRganq2HdXjDuChiFgWEZGZtwOfAh7Czyv17A5sO+qDiIjyyweAGZXtmbkeeC1wJrDahKG6cQeec1Qf+8jqFyYNtUUiYlb5ryUz7wNuBo4H9ioTh/cA3wHWVE5YEkBEjI6IseXDHwIzgVdU9pcfbJ8BzoyIYQ0IUU3KtqN6RMSkqkW6fgP8C3gaMLX8DPsN8FXgpIgY3ag41XxsO6pXRLSxcWbX1cABEfGqyv4ycfg+4PFVyUUNcZ5zVC/7yBoIJg1Vt3LRk88CnwfeGBEnAS8BHgbOpahtCDC93OYdVAEQEY8BPkdR0PliYAxwNLA2It4eEdPKQzcA95X/S7Yd1aX8vLqaohD4FcDuwPnAXOClwF7loSOA+7HdqGTbUb0i4liKWnRfiIgnZebfgYMoyva8uurQR4AHHWko8Jyj+tlH1kBxIRTVJSIWUHygnUgxPP5Q4Cjgy8B7gGcD+wBTKYr1nlHe2dAQFxF7Ax+imI4znqK2xj3Am4E/UySi/0ZxU2MxcFqljouGNtuO6hERu7KxiPwI4LHALODDwHeBNwKTgbEUUwdP9/NKYNtR/SJiFcU0wGdSfF49G/gK8FZgFPBtilFA/wH2Bk7NzJsaE62ahecc1cs+sgaSC6GoXmMparP8KjM3RMQnKOomnEWxGMqVETEGWAj8OTPvblyoajLTgf9XWbUrIu4AngpcAFwOHAvsQrFC3E2Z+dsGxanmY9tRPSYCd2XmzQAR8UeKIvKnUXxuPZXi4msu8JvMvKtRgarp2HZUrwXA1zPzWoCI+BXwQuApmfnCiNiNYtThROBlmXlb40JVE/Gco3rZR9aAcXqy6nUbxZLtJ0bE8Mz8B7Ae+BawBiAz/5WZPzVhqE7+CoyIiKkAZUf5rRR3Uk/PzAcy84eZebUfaOrEtqN63AXcFxF7l7Wg/gh8kqI9HZ2ZD2fmHZn5v16AqRPbjur1b4rEDgDliJ4rgGMj4szMvC8zP5eZ7zNhqCqec1Qv+8gaMCYNVZfM/DdwK3AMsF9EjCgXQrkW2LeqZoLU2fUUI1XfUtlQfrC9m6IzPbJRganp2XZUjzuAfwAXATPLRboqF2JHRMT4RganpnYHth3V5zPAwoh4f2VDmTh8DXB4RHgNpq7cgecc1cc+sgaMH1jqk06dnKsoiqg+CTilXIFpJ+BRioVPJACi0AqQmY9SFOVdGhGfiIhZ5WETKaa8P9qgMNWkbDuqR1W7SYopOuMoPrf2Lw/ZnmJFykcaEqCalm1H9ar0kzNzA8X04xUR8YGqQ8biZ5U68ZyjetlH1tbgQijqs4hopxjm/JryRHUxsBtF/ZaRwJMy88eNjFHNISKWUNS4/Fv5eCZwQGZ+vEwyf44iwTwM2IFiwRyL8moTth3Vo2w3j8nMN0dEG/AGigW6pgKTgLMsIi+AiBiTmf+qejwDOMG2o94qR4U9GhHTgT0y8ysRMYVi0ZM7KFa63R0XPREdn0/3Vc47nnPUW+XU9Q1Vj+0ja0C5EIp6FBGrgf0oVnf7UlkDYSzFKkyVuxqvKk9Q84G/ZeY9jYpXzSMi1gFvB54CfDUiAlhEseobmfnfiDgWWFZu+3Vm/q5R8ap5RMR+wBKK88z1mfl3bDvajIhYQVHk+y7g5rLe03bABoDMfCQiLqQoFj4T+GM57UtDXEQcAxwVERdl5n/KzeOx7WgzIuJQYEVmXlEmDNuAFRTth8y8t1z0ZB1F//nyzPx14yJWMyj7MOcAzwV+Vm72nKPNKtvO6oi4NDeO/loETAH7yBoYjjRUtyLiKOBK4BqKu1wTgfMy895Ox43IzIcaEKKaVEQcBrycogTC7Zn52AaHpEEiIg4HXg18F1gKvD4zP9fYqNTsyqTPKykW4xoF7EUxIv7mhgamplcmfa4CnpuZX210PBocyhuhwynqiM0A3p+Zzy731YxalapFxEEUC1Q8JTPXNzgcDSJVn1fPq6zMLm0NjjRUlyJiD4oL99Mz88aI2J3iJDWq03HHADtGxOvLUYca4iJiLfA6iuk3P42I70bEkZl5TafjHg88kpmfbEigajoRsYZiOs4TM/P7EfEGYPtyYaWHy8WWbDuqERGTKUY0P6485ywFvkwxwnldZv6sPO5UYFJmvqWHp9MQUs6m+CqwZ2b+pCy/soSi9tOvM/PP5XG2HW0iMx+KiM8DfwEOjog3ZebTKFZOBiAijgce9AJfVfYC3pCZ6yNiNsUo1OHA+sy8FTznaFMRcSDwBWCvzLylXCV5FsX55m77yBpIJg3VnX8Cr8vMG6FY8S0iRlDULvx91XG/AG41YajyrvtYiqLN55QX7xOA24HlFCNWq/25/CdVise3UrSd75e1fU4BbgRWAa0RcXFm/gHbjmr9BxgDzAV+WnamP00xresNEXFCZv4DsJ6POvsN8F9gTUT8GvgscDdFfWYi4rzM/D22HXVSNS3wX8BsilWRXxARnwUmlzdQh1Fca/2mMVGqmURElO1mLvDXst/8MeAminPOSyPiwPJG108BpwOq2n8p6hQui4jbKFZov5+i7fwyIi4vZwPaR1a/c3qyuhUR4zPz/ogYVtZH+CLFVMFvltNPb83MOxsdp5pLREzIzH9UFQRfQ/HBdmRmXt/g8DQIlAnESwAy85VlPajLgXc4fVCw8eKrrB+2AbgU2JkiyTwTWAg8n2LU8zll4kfqEBHDM/PhiJhLUVNsJEUJlvdGxCLgZcBHMvPqhgaqplZONV2Xmc+JiKcArwC+k5lHl/sjvdga8qrbQUQspyjh80fgusx8d7n91cA/MvMljYtUzSwiVlFcU40GLs7Md0XEwRSLkl6RmTc0NEBts1oaHYCaR0QcERHviog3lVO67i93VTo7/wb+FBHHUXzYtTYkUDWdiDi8vLtOmTBsKxOGUdZr+RiwT3ms7UYdImJlRDwtIh4TEdsDlCvCvTYzX1k+/ikbR5NJANOgKBZftperKepg7l7uPzEzb6H4/JrcmBDVjCJiFkCZMBxeFojfCbgsM99b7ruNYkTHyMZFqmYTEXtHxBMi4sAoVkgGuBUYUy48cClFsnl8RFwFNSMSNUSVAy2+GxGVEk93UtTf3ZuiHmbFg1s7NjW3iFgTEZdHxIURsSQzrwMeC7wgM98FkJnfohiFuF0jY9W2zaShgI46dK8HfkAxnfRjEfHUiBidmY+Uh/0deC3F6I0nlCspawgrp1YAnAUcHxFfho5V34ZXdZZvAZ5VbnMqu4CORU/eAcwDLgIOruyrOu8QESdTLIpy49aOUc2nvHF1V0RcUNmWmb/IzHdl5hMz89mZuSEiTqOYNvinhgWrplLWl7szIi6HmsThXZn5mqrjTgJ2BRy1IQAi4kjggxSfU/8DnBYRw4AA9gDeBjw9M19BkTx8a6NiVXOo6iMfQlG6Z31EjMrMvwCfp7jZdXBEXBQRZwCHA59qTLRqNmWy+Q0UsykOAY4DKBOHb6o67mRgPsW1ljQgnJ4sACLiEqAtM68qHx8CvAp4Z2a+vdz2VYradPtm5q8bFqyaRtUUwVMpFsk5GJidmQdWprWXx7UAHwcucUq7oGOKxfsoFj25LiJeBvyDop38NzP/FBHjgRMopiqfkpk/b1zEagYRsQvFhdYngccDr8jMt5b72irJ5og4B3gucGxlIRQNbRGxI/BR4NMUSZ/XZ+aLy32Vz7JWinb1XIrRqp5zRETsC3wAODMzr4+IJwDPAFZn5r/KJPPfM/Mb5fFOSVaHsr8zBziC4hpqxzLhPK3c/iSKWvLvzcybGxepmkVErAA+THHOuS4iLqYYlfph4IHM/HVEDAceR/F5drKfVxpIJg0FQET8D7BHZj6uatshFHe8TsnMr5d1W+4up3xJHSLiccBRwNOBD1F0gkZTTBV8JDMfjIgRmflQA8NUEykvwlrLRU9mAT8Gfgg8QNF2LgXuAU4Fvp2Zv2hYsGoaETEOeExmfrAc+fNeijo+b+103FRggje4VFFepJ+YmR8r66T+gCLp/OJOxx1FsXLyLxsRp5pP2f9tz8yPV21bD5xf/dlUJp03mDBUtYg4AnhSZp4YEZ+jmJb8ELB/Zt5dGZFou1FFROwAzKjqI/+QYuT7X4D9KBLNN1Pc5PqhfWQNNJOGAiAitqM4GX0kM19Utf0yYExmvqBRsan5lYXkX5CZ55RTTj8B3JmZuzY4NDWpiGgpp5AGcB4wNjNfHRGLgZcAH83Mz1ePHtPQVtVmWso6hpWLsfcBL8nMt5Q3u+4sa9JJQO35pmoxgmUUF2KvyMwXl1OXbzVZqM7KmxUjM/Oe2Lj40g0UI+VvjmJhi99n5t0NDVRNqaxl+JbMPLs873yXYsGTOQ0OTYNARDwemJmZr4qIsRQrtf8kM99mH1lbizUNh7hy2iiZ+XeKunSHRsQVVYeMACY0IDQ1uapaLQB3ldsuAN4MPBW4NyK+0IjY1PwqSZ/yAv6dmfnq8vEvKUYbjisf2xlSRcLGtlMmgL4CPBF4djmC4/UUC1hI1Sptp5IwHJaZN1GM2LgoIr5OUbN5Q+NCVDMqzzP/zMx7yk0tFAsBPgr8ISJOBK6kqG0oAZv0kR8st72UojzCicCvIuKHleswqaJT2yEzP5qZryq/foBilOqw8rF9ZG0VbY0OQI1TdoQ2lNO4Zmfm98qaCR+OiJ2APwDrgJMaGqiaTtWojfHAJIqV4FqAy4ALMvMLEfElTDirk6raYaOB4Zl5XyUJVO4/BdiTYrShBGysV1iec8Zl5h8qCaDM/Ep5vjkJOCSL1XAloObzajtgWmb+MjP/W7apmyLizcD5wBpHqKpap75O5bzzcLnvNuCFwArgnMz8cyNjVfOIiNbMfLQ858zIzJ9HxO3A2cCzMvPrwNcjYk51/0fqqq/Taf8pwAHAGxsSoIYs724MMRExPyL2KO9iVO5knEixeimZeT2wF/BN4FfACRZWFUDZbpZBMdKnTPpcBmxfdnreSVGI9wtlYugfLnoiKKYBVrWdjIgxwKuBWVXHjIuIMykuwk5LV2cf8iJiSkRMgo4V2YcBz6GomVp93GHA8RQJQ4vIi4iYU978rHxejQSeTFXbKdvUauBY4OB0wRwBEbEgIvauShiOoejrzC73V/rO+1H0n0/3vKOImF7WniMzHy03nwfMLL/+HPDYzPxc+VmGfWRB7/o6ZR/5CcCLKM451mvWVmVNwyEkIh4DvJKiiOp9wNeA92fmPxoamJpeRMwAfk+RGPxgZv6g3L64c/0nCzqrWllz7uXA4zPz1qrtu2fmT6oeTwAeA3w/M3+19SNVM4mIY4CXAndTrCr5gsy8JSJmZ+bvOx07kuLmhYlmVZ9z7qeoyby83D6zi1Ebw4GpnbdraIpicaVXUpTIiMzct9y+qPMo1HJmzjUuQKAoFk96OfBXiqnIbwS+SrEY132djrWPrA697etExAjgYIpFuhwRr63OkYZDRETMobhrcUJmrgQ+CywF/iciplQdd2hE7NOgMNW8/gXcSlHj8oSIWFVu70juRMRuETErS40IUs2lHAF2FfCMzLy1LCAPQGb+JIqVJomIRRR16D5gwlBRrGz7KoraqCdQnH9eHBFrKTrWHcdFxA6Z+aAJQwFExDqKc86FmXkg8N+I2AWgOjEYEftHxC6Z+bAJQ0HRJijOO0/KzP2Af0fEQoDqi/SI2L3sN7/ehKEiYi+KmpZPzsyDgb9T3GA/C/hv1XH7R8QS+8iq6ENfZ3dgu8z8iglDNYpJw6Hj3xR3Tis1oN4DfImiHt2JAGX9hIUUIxGlDuVo1C9RTK+YAqyLiCcCT42I1oiYB5yGCxCoVNZKfQ3wzcz8TkRMB14bEW8sa4hR1vzZCXg2xcqUdqQFxWIUP8rM72XmA5n5BOA3FNNL5wJUnXP+3bgw1UzK0cqHA08vazTPARYAl0XExyPi4ChMAw6kuLiXKqYBn8nM68vPq52BKyPiqxGxtmw7cynOOx0ruGvIGwOsz8wbysfPo0j4HAbsBMXUZYpzzj8bEqGaVV/6Op5v1FBOTx4iys70q4DvAp/MzIfK7U8FjsnMI8rHwzLzv90/k4aa2FjQ+QPAxynqXX4eOAi4NDPfXB43LjPtEKlDRJxDcff028CpwBeAnwOXAz8uO0hExHZZrOAuERF7Au8FHlc9kiciPga0ZeZJ5WPPOaoREe2ZeXd5E/RtFKPhXwG8jGKBpXXlIiijMvM/jYxVzSUijqcYIfYWihXZ3wK8jqL97AYcmZkPR8T4zLy/cZGqmZTlEN5NUVf3FuASipVtFwAbMvO08jjPOapRjlJ9D3Bqp/I99nXUdBxpuA2LiB3L4fCjypFi1wAXA2vLGlBk5luBsRGxpHxswlBUijSXKncWrgaGAzOAXYHvADtExL4AfqAJICK2j4jtATLz3cBHKIqBvzczX5SZn6K4IGutTFc2YaiIaC9HgJGZP6aoufutSmH50oXAyIgYVx7nOUdExNzYuOhJZUrXI8BVmfnicvr6syhGaswoj/PiXUREx3VQZn4eeBrwM+B7mfmqzHykbDvDKGbiYMJQ5TlnMUBmfgV4M/Ah4IPAfpn5fIoFKzp4zhF09JHbATLzRoqb6fZ11PTaNn+IBqPyjunrKIY5D4uIL1EU5h0HXAFsHxH/j2L48xTg3kbFquYSEYcCyyPiI5l5Z9UUnAcp7og9SpH0+QnFiLE7GhKomk5Z0PmVwG+jWHHyosz8SET8H3B71aH7AqOAVoqLew1hEXE4xefSvRExKTP3y8xnlze3/q+sjfkz4EigHafpqBQRxxXAmW8AAB/qSURBVFGUQbizvNn1JuDbmXkPcHPVcY+nOOe48JsAiIiDgOkR8eVKIjAzP1FOa39aZSGCiDiJou9s6R51PueMBF6TmVdGxPsp+jR/Kg89HBjlDC5VdOojj6Wo9/3Csh3dWF5/2ddRU3J68jYoipVuPwVcUC428DiKWhr/Bp4LrKOotbGYYuTYM6pXMdXQFRGrKaaSfp9isZzPZOad5b4A3gB8pbyzSkSMqEx119AWETsDnwbOKeuJvRGYB7wLuDYzHy6POwu4ADgzM29pVLxqDmVC8DXAM4EfAF8B/q8c3UNEXEHxeXUXsAg4PTNvalC4aiIRsYDinPPkzPxRRFxKMQr+RuBDmfm38mLsVDaec37euIjVLCLiYOAbFInll1J8Rv2zav/rgfMpyiSsAh6fmT9rRKxqHt2cc3YHrgc+mpn3lsedDbwAONpzjqDbPvIOwLsy84sR8SKKZOHvsa+jJuRIw23T/RSLnowGyMyPR8S9wGOB8zPzDRHxTYo7YsMy82+NC1XNIiKGA0uAx1Gs4HUmxRTST5YjDjMinpuZD5RTetKEoao8CPwY+CFAZj49Ip4LPIliNOpNEbGCot7PKSYMVU69eTzw/Mz8RrntHcCKyjHlXfgPU5RJ+E9m3tWQYNWM/gH8DrgHIDNfFRFnAoeU2z9PsZjFifz/9u487tOx7v/46z1mswvZipIolZRKhNsWWVLIHomWKXdK6rZU992m0oJ+EpVkq6SIfiElRKSUqGxZs41dYmxjzPv+4zi+131el2tmrjlpzu/lej8fDw/f7/k9Zx6H+jyO8ziP5fPJhGFUkuYH1qOMiRehVC4dJ+ms3sSh7X0kXUSpfnuI7Rs7a3D0k9n1ObcCp9eNG0sCW2acEw3DjZE/Cbxf0s22PyPpZEqf83jGOtFvMmn4HFN3gz0F/B1YXdLlth+z/WtJL6B0TkfZTsXJGKQm+D4NeML2Q3VgvT1lMP1j2/+gTDSTqoExjHHAqygTPhcB2P6ipKMoqRI2tn2JpP/IQkVAydMj6cvANElyOfpwL2WiBwBJ421f11kjoy/Vsc58gIDXUV7ksX18zQ31ceB025dJ2ik5oaLH9mOSvgc8bPvBekxwCmWsc2Yd/8j2KR03NfrIXPQ5UyUdliPJMcRwY+Qv1DHyEcAGzWIoEf0mhVCeY1w8RkkiPwXYWlJvx+HxlJWOl3fYxOhjtu9t5PY5FTiVUnFyY0mfBb5ddyRGDGL7ekrOy59LWr3x00eARyQtVu/LhGEMsH11bydzvTSTukte0hTgsPqyFjGgjnXuphz3OroeN+399gXgcUkvqt8zYRiD2L7N9oP18zcp6VjeB7xB0v6UY8kRA0bY56xQv2fCMAaZwxj5IUmLdtOyiJHJTsPnKNtn1KNf/wUsJelvwOKUwid3ddq46AuNnT2DPje/2/6JpLuAb1MSgb+9l5suxq7ZxMvhkhYBLlQpxvQHYFtgBZLQOWZjyE7DP0vaAtiTkjsqyZdjQLP/sX1i7XNOlrQ35fjXa4EXApksjEGGeXaNsz3T9hGS7qVUv51BeW5FAHPV50zrsJnRp0Y4Rs44J/paJg2fA3qDnsb3Xud0kqQHgY0oxU8mAdvZTgW4AJgATIeygjpkUNT8viSwLLBO8rNEtTBlZbT3wjUw2LF9kKR/AgdTcvysAuzW28EaY5ek5YF7PEwu1EYMPUXZJb8GsKeThy4ASesBL7L9/aGLXba/WfucbShFT5ai5E3NruZA0vz1BE6zn6F+n9kYQ08HJgPrZawTdefXdJcUT+7teE+fEyMhaT7bT8Hgfidj5BitUj15FJO0BnCV7SdmNXFYP0+g5OCY3/a/Ompu9BFJm1AKVNwCXGn7xFnctwBlQPT7vLwHgKSNgZOAbW1fNLTvady3AmXHhm3fOa/bGf1F0pbAZ4G32Z46m/tWp1Q1Xdf23+dV+6J/qVTY/iHwK9s7N/ucIWOd51EmnSe6VjGNsU3SppTjfzcDl9j+wSzuWwz4MPBTp0rymFf7nE8AU4G/2D648Vv6nJglSSvXo8iDJg6HuS9j5BhVktNwlJL0VuBPwOWSJvZWS3u/D1lNnWF7eiYMA0DSZsCXgbMoSZ3XGvJ7M44eBY7LhGE0vIIySP6RpE2afY8qAJc8dVMzGIra5xwIfHK4CcMhfc5fgJUzYRgwEDufrf+8VtL6zUWKIWOdB20/lJf3gIGJn8OAnwCPAKsO+b3Z7zwIHJwJw6gTzYcC36Tkh3/RkFua787pc2KApLcBf5J0PIDtpyTNN+SejJFjVMpOw1FI0vOB71OqLe0OvAx4nUv121muakRIWgX4ObCX7XMlbQ7sAxwHzLR9cpfti/7VW12XtA6wLnAjcCSwIXB7FiViqDo4XhG4AXi37RPq6vo2lBX2+5p9TiPG5AxOxjxJrwdOBD5g+wJJBwHLAR+qC1oRT1P7nfkpxUxOtn1aXWj/CKW4m2wf1bs3fU301JNZhwEX2f6RpPUpE4inAgva/mS9L3ETg6gU3jqDchLnlcDjtt9Tf8u7eYx62Wk4ijRWJ+6lrLr/0vZ2lO3zl9W8LemU4ml6sQPcBGxTJwyXB46t11YE9pZ0QFdtjP7WGCDfBmxCGRztD1wE3Cdp+aErqjF2NfLN3UQppPQ/NS/dCZT+ZjXgPZLe2/szvRjLy9jY1nheTaDkCbugfv89ZbfYQvW+jGHjaWq/8yhwB7CnpHWBo4GrgUWB3SQd2Lu3u5ZGv2i8Xz0JPAR8tu44/AFwLnAd8A5JR9T7EjcBDHpe3U1JcXAEcAjwfEnHQNlx2FHzIp412Wk4ikha2PbDje8TXSvZSvolsJzt1SRtByxq+5iu2hr9pRkrjWsrAJva/m596H0AeLHt/TtpZPQtSeNtz6iTgpOAU2xvIem1wNmUY+4b16OlEUhaGJjWyP10FKXAyUdsf6PmS92Pkj7joA6bGn1G0iQ3iuVImlBf5pF0FnBnbwdHxFC9eKmLFLtQirndZnvf+gzbFVjd9r6dNjT6hqTJth+vnxcFDqKMa2ba/lC9vg7wfmCP4fI4x9jUGx8PuSbgNZQNPvfafo+k/6DMN/+2i3ZGPFNZpR0l6jHSn0s6srHSNV3SxPr5LcC1kh4Bvkop4x6BpDcDx0n66pBdPbfa/m79bGARYGFJ4xorZzGGSVqjvsDP6B2vqDs4/iDpy5QjO+8FDgBOkbRQYid6zyvgm3WyENsfBN5s+xv1+6PATOB5zVyYMbbV59Wx9Xn1Pii7fxq7mA8FFpG0dGeNjL4k6RXNCWbbv639zmnAhLp4+hSwNLBsPYoaY1zNffkjSQdL2tf2v2zvTSnItVRd4AJYgzIBnbgJYKCo5LckHVA37AAD71R/Az4NTJb0N8rJrtu7aWnEM5dJw1FA0trAt4D/R8mVsIak39bjyNMlzV9vPZWS7HnLJHMOAElbAIcDvwAeBjaU9Kph7nsXsCNwuO2ZOXoRkraiFFv6Yy/1gaRJ9efpwPbA3rZ/Xief32h7WmJnbBvmebW6pEvq5PN5vckfSbsB2wJH1+OEiZsxbpjn1Qa951XjeNefgTdSnlcRwEABgkspi1eT6rXx9efrgQWBT0n6ELAb8Lne5GKMXXX317HAjyl5mreUdGb9+VzgMeBoSfsC7wP+q7kLOsYuSRtRcu5eAiwOvLO3SApge4btyykpfRYD3m775k4aG/EsyPHkUaAOpNe1/YnGtXOAybbXq99fRnlR2ydHBAMGjh//BPgf27+UtCRwOnCE7R/VexYB3gx8HtjBqZIcgKSlKLFyNLA2ZYV9PduP1d8nAS+1fVUvd113rY1+MsLn1eaUHWPbpc8JGPHzapxLtfZNKcdNr+mwydEnJL0A+Bklflak7AZ7Z29yp57ImUIpGrg4cJDtqztqbvQRSe8GlrX9pZojdUFKxeS7bW9dJ4Y2peRRPTJxEzBw/Hg/4DHbh0taEFgJ+Bpwg+296n2rAxdSxs9/7azBEc+CTBqOApK2pCSSf5PtWxvXf0ep8LVf/b6oU8E0KkkLAWsBF1B2y8+QdCil0u2hjftWBJ60nW3zMUDSGpQiOY9RilesBKxv+5Eh92XSMAbM4Xn1W9v71wmiGbandtXO6C8jfV5FDFV3L69JKXSyHPDfwERgt95CV+PeQfkyY2yTtAvlCOn6tu+q1xahHGk/qZfCZ7i8dTG2SfogJWfq1rbvrxOJr6bE0/dsn1HvW8r2PR02NeJZkePJo4DtM4GTgUPqimrPQTT+P8yEYTTZngZcYvvJxmBnBrAwgKTtJe1g++ZMGMZQtv9s+8H6gvUuytGdC6DkAJK0a70vE4YxYA7Pq/H1nlszYRhNI3hebSdp584aGH2rHl2/tOaiuwb4HCWFxokAkjasu5up1yN6TqHsLNyn7m4GmAacQZmABspR0w7aFv3tLOAaYDtJi9Sx8I3AXcBLejdlwjCeKzJpOHocC0wFDq87wwBeCqzayDMWMUhvV1gj4beAW+vxrv2By7tqW4wedeJwd+BySf+i7CT7U7etij42q+fVyyVNStGTGM4cnlcHkD4nZqGR8xJKDsPPAfdLugE4Driu3pdFrhhgezpwNiXn3H6SXuBSGXkCsIqk8XlexXBs3wJcQdkhv4ukJevi153AS5SikvEcM37Ot0SXekf/bF8p6WjgncDFkn4GbAa8NUctAkrenjoAal7rHR2dWS/dTcnDcQ/wbtvXz+NmRp8aesx46Hfbj0u6AtiCUmzp2i7aGf1jaJ+T51U8E3leRVvN51WdQLxW0j2UY8pb2L6x0wZG3+nlSLV9pqSZwOaU59WpwA7AW7LDMIbTiJ0jJb2fkh7hfZLOBvYANq6TzxHPGclp2Od6A6FaQXAJ2xdI2hB4ipLr56aOmxh9oMbEGyhJ4x+t13qxszqwWI2d/Sj5Nl6TF7AAkLQ08AAwyfa0mgycWnBgNeBxypGLFwP/n5JgPsWWxrg59Dl5XsVcyfMqRkLS64Brm7l1G7HzSuAfth+RtDKlqunGeV5FT3NyuRE3LwNm2r5e0taUHc5Xps8JmGOf8xLK4tZkYEtKIZ3zbP+9m9ZG/PvkeHKfkLSFpD2GXq+d0lqU3Am9a+fbvjAvYAED1Uq/C1zee3mHgdhZmzLR03Mq8LIMhgIGqtieDJwEHCxphbp6OrP2O2cAz6/XbgLWyQtYjKDPyfMqhjWCsU6eVzEsSW+hVNResXFtXI2dNwG/AFYDqDGzSp5XIWkTSW+HgX5Gjc9rAb8GlqrXTrd9WvqcgBH1ORcCr7R9v+0TbB+VCcN4rsrx5D4gaRPgp8BESYvZPqxeHwcsAGwPvL+uvKdSaQBlpQtYHPgYsJftc1Sqvk0AJtu+g3LE4gO2LwDIEZ3okbQZcAjwXko+n60pRyxulbQE8GFK7PxO0nz1yNdDnTU4OjcXfc6UPK9iqBGOdabkeRVDSdqKkod5l5r+YILtJ8tPmh/4IvBB279vPK/+2WWbo3t1YfQ0YJyk99aJHatU3F6YMs55v+2L87yKphH2OVNsX5rYibEgx5M7VgfLB1ESfF8D/AU4wPahje3PS7iWc0+nFE2SxgPfA94NLAP8GPgHJTfLzrZ/Ve8bWFntpKHRVyQtTClm8jPbJ9drXwMWtj2lfl/a9t3pd6JpbvqcxE30zO1YB/K8iqI+r/4GXGF7a0nLURYuFqCkzjgGmGH74Q6bGX1GpaDSNyh9zo2UUxUft31Co8/JOCeeJn1OxNNl0rAP1BWLJWzfXrc7/wY40PYh9ffn2c6KaTyNSuXs31CO5TwPuM72UZJ2B44CXpVjgTFUfSl/FWWy51HbT0nahZL4e/d6z0IuleAiBqTPibYkTaaMde7IWCfmhqQ3AicA51Eqsf+KclrqhZScvAcBTzrFB6KhTv4sUCcG30F5Ru1n+7j6+6K2/9VlG6M/pc+JGCzHkzsiaRXgYWCC7VslTa1HKn6nkjj+fEn3A9OAnSXtTOmcMss7xjViZ6LtWyQdCBwAjLf9UQDbx0vamJLQOWKQusJ+bT1q0fMIJYkzknYDVpP0Cad64JiXPieeqTq+eRy4o+7qyVgnZkvSCpSX84m2/yDpXcCZwNdtf7XeswewkVOVPYaofc7DlGcXtk+tC6ZHqlTWBthU0sczzglInxMxO5k07ICktwJfAS4HXiHpS7Z/rGJiza3xcuAG4E5gC9vTu2xz9IdhYueLtn+iUv3ta5IOtP2lumvs1cCjs/v7YuyosbNqb+ADDB0kjwPuVKkeuC/lqGkG0mNc+pxoq9nn1N3MciVpfMY6MSuS3gb8D3ALYEnH2P6FpDVs39q41cD4emLn8Uw2j20qhSteaPuY2ueM6+0Eq/3PKXXC8DfAPcAmGecEpM+JmJNMGs5jdRXjMEo+qMuBzYDv1y3yR0vqPbxWBe4ANrV9TSeNjb4yTOxsDpwoaQHb35Z0G/AVSS8FXk9J3ntnZw2OvlF3gP0QWKjmDTugvrg38/g8QYmt1YBdbV/bUXOjT6TPibbm1Oc0XtQz1olBan9yOLAjZZfYm4AfqhSyOLWRj25P4D+B3W0/1mGTow+oFHc7BZigUmjpENszexOHjbHOSsBtwOa2r+6swdE30udEzFkmDec9A9fYvrh+/2nd2fMzSQ/UzmkisAFl1T2D6OgZGjunSpoGnC7pIdunSTqPUhHOtu/rrKXRb9akVEk+D7ipDoD2HzJxeDdl5X2vDKSjSp8Tbc22z6HsbB4HbEjGOjHYgsBVtv9Qv18t6RHgO5Ietv0rlZyYu1Je3q/srKXRF1QKLW1EKVbxe+Cs2ud8rTdxSEmdMR+wCfD2jHOiIX1OxBykEEoHJF1OqVr6mca1HYEPA++wfdeQHUARwCxjZydK7Gxne2pXbYv+VV/Sl7U9VdLKwGXAt2zvV3/vVRBc3PYDnTY2+kr6nGhjBH3OMhnrxHDqwvmlwJG2v9O4/kHKDqD3AQsBM/O8ih6VIl2L2r5HJV/q9xmci25J2/c1jyxHQPqciJEY13UDxhJJ89WP+wBvkLRX4+dfU7bLPw5ly8Y8bl70sTnEzjnArSSXWAyjcRxwah0sXw+8AfiApAMl7QAcJ2lyBkPRkz4n2hphn3OspAU6bmr0GZXiFdOBrwPrSNq28fOlwPyUF/f78ryKJttP2L6nfj4f2A3YR9IekrYCvqhSwT3vVzEgfU7EyOR48jxk+6n68c/AD4BdJb3Q9icouaJWASZ21b7oX4mdaKu5AFGP6Yy3/XdJy1Jyt9xFye3zeGeNjL6TPifamos+J5POMUij3/k1sAywtaTlbB9B6XOWoez4yct7zFJduDhP0ibAlcC9wJszzomh0udEjEyOJ88jdSVjoJKXpCUpndHhwO318462/9ZpQ6PvJHairWFiZ5LtJ+pvWwJHAZslt080pc+JttLnRFuNmOnF0PLAesAB/F+/s63tv3ba0OgrjbgZ9O/627uBz1DypqbPiWGlz4mYs0wa/ptI2gh4NTDZ9sH1Wu+Bti7wJttfqdeXoCzOZxUjEjvR2ghiZz3gcNuPSPpv4LQkdA5Ja1KKmTxl+zf12njbM9LnxOyMIHbS58SwJL0GmAw80luAkDTR9nRJ6wEr2j5B0njgRcBDtu/tsMnRB+YQN+sCrwC+V/ugI4Hv2L6iwyZHn+k9o+rnXmXk9DkRs5Gchv8GdTX965Tj3zv3ckHVF/fXA8cAA5UCbd+fF7CAxE60N8LYudL2I/X65/PyHjXX0zHADsCnJf28TjTPSJ8TszPC2EmfE08jaRvgh5QdPV+SdCxAnfhZGzgaeLBem2H7xry8xwjj5s7ehJDtvTJhGACS3ibp7DrBPKOXt7lOGL6e9DkRs5Wdhs8ySUsDpwMftv1HSXtTksXfYvvXkrYDHqi5NlI1MAYkdqKtxE60UePmp8AU21dKWh34I3Cu7c3rC9qDts9P3ERTYifakrQY8AvgQ7Yvk7QSpdLtA7a3VKlYepPtXyZ2oidxE21JWh84iZLb8lFg/TrRPAGYAXwMuNz2uYmdiOGlEMqzbzKwAHCXpKUolSf/CKwq6U7bm8H/bYfusJ3RfxI70VZiJ9qYRImbe+r3a4EjKYnAj7H9nt6NiZsYIrETbY2nxE9vV8+NKgUrzpF0mO2P9m5M7ERD4ibaWgj4qu3DJJ0C/FbSei5Vk5F0qBv5VLttakR/yvHkZ5ntW4DfAecCFwPH2t7J9urACyTtU+/LAy0GSexEW4mdaMP2rcClwDclvQ74fP1pK2Dpupss4mkSO9GW7fuAC4EDJS1er00D3gOsJGnlLtsX/SlxE23ZPhM4rn7dFbgVuEjSpHrtlfW+TBhGzEImDZ8FvU6nJk7F9geBjYFzgJ80bj2DukIWAYmdaC+xE2004ma+eulI4C7gk8CSwMddksvPD8w37F8SY1JiJ9qStLqkDWreOSgv8DOBj/QmgCi7VScCC3bQxOhDiZtoqxE7bwSw/c/678eB3YFbgLPrsfbDGvEUEcPIpOEzJGlb4BpJq9TEqr0X+NuA24DPSFqu5hR7C3BJh82NPpLYibYSO9HGkLh5CsD2X2zvDexie88aTztTXsCe6LK90T8SO9GWSpGu4ylpM/aXNKUWpzgbWAo4XNIrgO2BZYD7Omts9I3ETbQ1JHYOkPSh5u+2H7W9PbAsZZf8vk6Bt4jZSiGUZ0DSmsCJwJ3Aq4C1bV+vWsq9JgbfB3g1ZQA9pa7CxxiX2Im2EjvRxmziZlAOH5Wq21OA3Wz/tZvWRj9J7ERbtQDBt4EdbP9V0n9S4mfX+vvrKNW316bsINvHqXY75iVuoq1ZxM4bbb9ryH27AV8ANrN9dQdNjRhVMmn4DEh6LbCW7aMkHQrsAaxp+/oh972YUkUwRwQDSOxEe4mdaGN2cdOc/JG0GXCz7b932d7oH4mdaEOSgJ2Byba/V6+9EDgN2LbujO/duxjwhO3HOmls9I3ETbQ1h9jZGrjT9sx6fRvgettXdtXeiNEkk4bPkKTFe1uaJR0C7ElZ0bhO0jq2L+62hdGvEjvRVmIn2phD3Kxr+6JuWxj9KrETbUiaH1jY9j2SJgKLAz8D3mL7QUmvzq7UGCpxE22NIHZWy+mbiLk3vusGjDaS1gNWAR4FTrP9gKQJtp+0/bGyyMGFko4EdpL0ZttTu2xz9IfETrSV2Ik25jJudpa0ceImILET7dXYeRnwMHC67XvqT08CjwFP1pf3dwK7StrNpTJujGGJm2irRey8y/a9HTU3YlTKTsO5IGkr4KuUJLzLA6sDG9i+XdJE29PrfRdTBtsbZyUsILET7SV2oo3ETbSV2Im2hsTOCpTcuhvYvr3+PgE4CbgK2IqS//KqjpobfSJxE20ldiLmjew0HCFJ8wG7UiosnVWvfYuy0r5ub4W9dl5LARsmT0JAYifaS+xEG4mbaCuxE22NMHaeB2wErApsY/u6zhocfSFxE20ldiLmnXFdN2A0qIlVxwELA0v2rtv+AHAOcJ6kSfXyVGCLDKIDEjvRXmIn2kjcRFuJnWhrhLEzuR4b/C6wY17eI3ETbSV2IuatTBqOgIsngR8AH5G0ZuO3KcDVwMb1+2UeUsU0xq7ETrSV2Ik2EjfRVmIn2hph7GxUL30qk80BiZtoL7ETMW/lePJsSFqHkhthBiWx6g8krQQcKOlLti+tt04DFumqndF/EjvRVmIn2kjcRFuJnWhrLmNnMYBeTswYuxI30VZiJ6IbKYQyC5K2BL4InA8sQ6nAtCfwImDH+s83gInAXsDbbd/QTWujnyR2oq3ETrSRuIm2EjvRVmIn2kjcRFuJnYjuZKfhMCS9ATgE2Mn2FZLWAj4NLGz7BkmHAdcA21JWOnZOpxSQ2In2EjvRRuIm2krsRFuJnWgjcRNtJXYiupWdhsOQtDrwJttHNa5dAnzC9vmNaxOAGc7/iFEldqKtxE60kbiJthI70VZiJ9pI3ERbiZ2IbmWn4TBs/0XSTVA6n5po9SHg4XptI+DPth/ssJnRhxI70VZiJ9pI3ERbiZ1oK7ETbSRuoq3ETkS3Uj15Fmw/3PtY//0EcK+kdwBfIcnAYxYSO9FWYifaSNxEW4mdaCuxE20kbqKtxE5Ed3I8eYQkHQ8sAKwI7G77qo6bFKNEYifaSuxEG4mbaCuxE20ldqKNxE20ldiJmHdyPHkOJKl+XJnSKa2XxKoxEomdaCuxE20kbqKtxE60ldiJNhI30VZiJ2Ley07DEZK0DXC97Su7bkuMLomdaCuxE20kbqKtxE60ldiJNhI30VZiJ2LeyaRhREREREREREREDJJCKBERERERERERETFIJg0jIiIiIiIiIiJikEwaRkRERERERERExCCZNIyIiIiIiIiIiIhBMmkYERERERERERERg2TSMCIiIiKeEUlLSLqi/nOXpDsa3x+t97xYkiUd1PhzS0p6UtIR9ftnhvzZKyQt1tV/V0RERMRYNr7rBkRERETE6Gb7fuA1UCb+gGm2v1a/T2vcejOwJfCp+n174Kohf91hvT8bEREREd3JTsOIiIiImFceBa6R9Pr6fUfgxx22JyIiIiJmIZOGERERETEv/QjYSdLywFPA1CG/f7RxNPn8ed+8iIiIiIAcT46IiIiIeets4PPA3cDJw/ye48kRERERfSA7DSMiIiJinrE9HbgM+BhwSsfNiYiIiIhZyE7DiIiIiJjXDgEusP2ApK7bEhERERHDyKRhRERERMxTtq/i6VWTez4qadfG961t/+Pf36qIiIiIaJLtrtsQERERERERERERfSQ5DSMiIiIiIiIiImKQTBpGRERERERERETEIJk0jIiIiIiIiIiIiEEyaRgRERERERERERGDZNIwIiIiIiIiIiIiBsmkYURERERERERERAySScOIiIiIiIiIiIgYJJOGERERERERERERMcj/ArSdw+7ddCRoAAAAAElFTkSuQmCC\n",
            "text/plain": [
              "<Figure size 1584x432 with 1 Axes>"
            ]
          },
          "metadata": {
            "needs_background": "light"
          }
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "\n",
            "\n"
          ]
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 1584x432 with 1 Axes>"
            ]
          },
          "metadata": {
            "needs_background": "light"
          }
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "\n",
            "\n",
            "CPU times: user 6.24 s, sys: 454 ms, total: 6.69 s\n",
            "Wall time: 6.73 s\n"
          ]
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "The graphs above give interesting and enlightening pieces of knowledge about each of the coin prices change by the given timeline. \n",
        "\n",
        "From the beginning of 1st of January 2018 till the end of 2020, the market cap of each coin seems slightly constant between the same values, except the first half of 2018. During the end of 2020, both of the coin prices made a rapid jump, which caused gaining more value and becoming more volatile for each coin. \n",
        "\n",
        "Also, we can say that, since it is required to take the timeline later from beginning of June 2021 as a test set, dramatic price changes during the mentioned time make training a model that learns from the given data more challenging and bring some biases."
      ],
      "metadata": {
        "id": "SXCAtNxKr8Tn"
      }
    },
    {
      "cell_type": "markdown",
      "source": [
        "## Splitting & scaling data"
      ],
      "metadata": {
        "id": "b508w3VNUxHa"
      }
    },
    {
      "cell_type": "markdown",
      "source": [
        "Now, we split data into **train**, **validation** and **test** sets in order to feed through a model. In this sense, having a validation set helps us for fine-tuning the parameters of the model and avoiding over-fitting problems.\n",
        "\n",
        "\n",
        "Related to scaling, we will be building type of RNN models that basically contains Gradient Descent Based algorithms. On this purpose, scaling benefits optimization by:\n",
        "\n",
        "- It makes the training faster\n",
        "- It prevents the optimization from getting stuck in local optima\n",
        "- It gives a better error surface shape\n",
        "\n",
        "\n",
        "\n",
        "We preferred applying standardization on our data depending on the information we got from this page: https://machinelearningmastery.com/how-to-scale-data-for-long-short-term-memory-networks-in-python/\n",
        "\n",
        "Basically, in regression analysis, it is often crucial to standardize independent variables or we may risk obtaining misleading results.\n",
        "\n",
        "$$\n",
        "Z = \\frac{x - \\mu}{\\sigma}\n",
        "$$\n",
        "\n",
        "*Z*\t=\tstandard score\n",
        "\n",
        "*x*\t=\tobserved value\n",
        "\n",
        "$\\mu$\t=\tmean of the sample\n",
        "\n",
        "$\\sigma$\t=\tstandard deviation of the sample\n",
        "\n",
        "\n",
        "Generally speaking, the best practice is to use only the training set to figure out how to scale/normalize, then blindly apply the same transform to the valid and the test set in order to:\n",
        "- keep them identical \n",
        "- avoid leading to random skew in the data"
      ],
      "metadata": {
        "id": "qXPm_8j1zQO1"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "def split_scale_data(df : pd.DataFrame, valid_threshold : str, test_threshold : str, coin_name : str, scaler_method : StandardScaler()) -> pd.DataFrame:\n",
        "  \n",
        "  \"\"\"\n",
        "  Split data into train, validation, test sets and scale them\n",
        "\n",
        "  Arg:\n",
        "    df (pf.DataFrame): a coin table, that will be splited into train, valid and test sets\n",
        "    valid_threshold: a time point, valid dataset will be created from that point until test threshold point\n",
        "    test_threshold: a time point, test dataset will be created from that point\n",
        "    scaler_method: a scaler method, sklearn.preprocessing.StandardScaler is set as default\n",
        "    coin_name: name of the coin that will be printed\n",
        "\n",
        "  Return:\n",
        "    returns scaled train, scaled validation, scaled test sets and coin scaler\n",
        "  \"\"\"\n",
        "  \n",
        "  valid_threshold_index = df[df['Time'] == valid_threshold].index.values[0]\n",
        "  print(f'valid_threshold_index: {valid_threshold_index}')\n",
        "  test_threshold_index = df[df['Time'] == test_threshold].index.values[0]\n",
        "  print(f'test_threshold_index: {test_threshold_index}')\n",
        "  \n",
        "  # We keep VWAP column since it basically gives weighted average of Open, High, Low and Close columns\n",
        "  # That is why we drop Open, High, Low and Close columns in order to avoid having too much information and messing up\n",
        "\n",
        "  # train set\n",
        "  coin_train = pd.DataFrame(df['Count'][:valid_threshold_index], columns=['Count'])\n",
        "  coin_train['VWAP'] = df['VWAP'][:valid_threshold_index]\n",
        "  coin_train['Time'] = df['Time'][:valid_threshold_index]\n",
        "  coin_train['Target'] = df['Target'][:valid_threshold_index]                        \n",
        "  coin_train['Volume'] = df['Volume'][:valid_threshold_index]                        \n",
        "  coin_train['Gap'] = coin_train['VWAP'] - coin_train['VWAP'].shift(1)               # Gap: represents price changes between per min\n",
        "  coin_train = coin_train[1:] \n",
        "\n",
        "  # valid set\n",
        "  coin_valid = pd.DataFrame(df['Count'][valid_threshold_index:test_threshold_index], columns=['Count'])\n",
        "  coin_valid['VWAP'] = df['VWAP'][valid_threshold_index:test_threshold_index]\n",
        "  coin_valid['Time'] = df['Time'][valid_threshold_index:test_threshold_index]\n",
        "  coin_valid['Target'] = df['Target'][valid_threshold_index:test_threshold_index]\n",
        "  coin_valid['Volume'] = df['Volume'][valid_threshold_index:test_threshold_index]\n",
        "  coin_valid['Gap'] = coin_valid['VWAP'] - coin_valid['VWAP'].shift(1)               # Gap: represents price changes between per min\n",
        "  coin_valid = coin_valid[1:] \n",
        "\n",
        "  # test test\n",
        "  coin_test = pd.DataFrame(df['Count'][test_threshold_index:], columns=['Count'])\n",
        "  coin_test['VWAP'] = df['VWAP'][test_threshold_index:]\n",
        "  coin_test['Time'] = df['Time'][test_threshold_index:]\n",
        "  coin_test['Target'] = df['Target'][test_threshold_index:]\n",
        "  coin_test['Volume'] = df['Volume'][test_threshold_index:]\n",
        "  coin_test['Gap'] = coin_test['VWAP'] - coin_test['VWAP'].shift(1)                   # Gap: represents price changes between per min\n",
        "  coin_test = coin_test[1:]\n",
        "\n",
        "\n",
        "  # scaling\n",
        "  scaler = scaler_method\n",
        "\n",
        "  coin_train['Scaled_Target'] = scaler.fit_transform(coin_train['Target'].values.reshape(-1, 1))  # scaling based on train set\n",
        "  coin_valid['Scaled_Target'] = scaler.transform(coin_valid['Target'].values.reshape(-1, 1))      # valid set must use identical scaling to the training set.\n",
        "  coin_test['Scaled_Target'] = scaler.transform(coin_test['Target'].values.reshape(-1, 1))        # test set must use identical scaling to the training set.\n",
        "\n",
        "  \n",
        "  # print data sets\n",
        "\n",
        "  print(f'\\n\\n{coin_name} train: \\n')\n",
        "  with pd.option_context('display.max_columns', None):  \n",
        "    print(coin_train)\n",
        "\n",
        "  print(f'\\n\\n{coin_name} valid: \\n')\n",
        "  with pd.option_context('display.max_columns', None): \n",
        "    print(coin_valid)\n",
        "\n",
        "  print(f'\\n\\n{coin_name} test: \\n')\n",
        "  with pd.option_context('display.max_columns', None): \n",
        "    print(coin_test)\n",
        "\n",
        "\n",
        "  return coin_train, coin_valid, coin_test, scaler\n",
        "\n",
        "\n",
        "# Since prices through the timeline are not constant and there is no sufficient time period to set validation set as we observed\n",
        "# we preferred creating a validation set with the size which is equivalent to the test set size based on our intuitions\n",
        "# so this way each of them indicates around 13% of all data\n",
        "valid_threshold = '2021-02-11 00:01:00'\n",
        "\n",
        "# test set is required to pick from 2021-05-31 23:59:00 due to the task\n",
        "test_threshold = '2021-05-31 23:59:00'\n",
        "\n",
        "btc_train, btc_valid, btc_test, btc_scaler = split_scale_data(btc, valid_threshold, test_threshold, 'BTC', scaler_method=StandardScaler())\n",
        "print('\\n', '*'*100, '\\n', '*'*100)\n",
        "eth_train, eth_valid, eth_test, eth_scaler = split_scale_data(eth, valid_threshold, test_threshold, 'ETH', scaler_method=StandardScaler())\n",
        "\n",
        "# we don't need the entire BTC and ETH dataset anymore so deleting them to avoid memory problem\n",
        "del btc, eth"
      ],
      "metadata": {
        "id": "WgonO8uKcGlQ",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "c495708c-40d9-48bb-ad55-0842206ac1ac"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "valid_threshold_index: 1636321\n",
            "test_threshold_index: 1794716\n",
            "\n",
            "\n",
            "BTC train: \n",
            "\n",
            "         Count          VWAP                 Time    Target      Volume  \\\n",
            "1          235  13840.362591  2018-01-01 00:02:00 -0.015037   31.046432   \n",
            "2          528  13806.068014  2018-01-01 00:03:00 -0.010309   55.061820   \n",
            "3          435  13783.598101  2018-01-01 00:04:00 -0.008999   38.780529   \n",
            "4          742  13735.586842  2018-01-01 00:05:00 -0.008079  108.501637   \n",
            "5          554  13706.952030  2018-01-01 00:06:00 -0.004422   70.805776   \n",
            "...        ...           ...                  ...       ...         ...   \n",
            "1636316   1864  44935.449789  2021-02-10 23:56:00 -0.007227   39.538163   \n",
            "1636317   1826  44913.353225  2021-02-10 23:57:00 -0.008847   48.526747   \n",
            "1636318   2467  44912.705558  2021-02-10 23:58:00 -0.007998   92.074357   \n",
            "1636319   2466  44848.417638  2021-02-10 23:59:00 -0.007590   86.665774   \n",
            "1636320   2956  44862.019363  2021-02-11 00:00:00 -0.010984  111.507314   \n",
            "\n",
            "               Gap  Scaled_Target  \n",
            "1        13.300499      -7.494206  \n",
            "2       -34.294578      -5.138308  \n",
            "3       -22.469913      -4.485835  \n",
            "4       -48.011259      -4.027472  \n",
            "5       -28.634812      -2.205036  \n",
            "...            ...            ...  \n",
            "1636316  40.833138      -3.602531  \n",
            "1636317 -22.096564      -4.409850  \n",
            "1636318  -0.647667      -3.986812  \n",
            "1636319 -64.287919      -3.783835  \n",
            "1636320  13.601724      -5.474796  \n",
            "\n",
            "[1636320 rows x 7 columns]\n",
            "\n",
            "\n",
            "BTC valid: \n",
            "\n",
            "         Count          VWAP                 Time    Target      Volume  \\\n",
            "1636322  10837  44754.964885  2021-02-11 00:02:00 -0.001288  400.223029   \n",
            "1636323   8032  44614.092853  2021-02-11 00:03:00 -0.000645  259.376814   \n",
            "1636324   4636  44705.501653  2021-02-11 00:04:00 -0.000092  203.337976   \n",
            "1636325   4533  44664.035028  2021-02-11 00:05:00 -0.001781  196.089887   \n",
            "1636326   2698  44714.279821  2021-02-11 00:06:00  0.000589   93.771220   \n",
            "...        ...           ...                  ...       ...         ...   \n",
            "1794711   1712  37319.205839  2021-05-31 23:54:00 -0.000769   51.601578   \n",
            "1794712   2297  37328.119321  2021-05-31 23:55:00 -0.000614   59.683761   \n",
            "1794713   2080  37295.825650  2021-05-31 23:56:00 -0.000593   59.175643   \n",
            "1794714   2957  37306.378838  2021-05-31 23:57:00 -0.000735  104.229560   \n",
            "1794715   2136  37303.718998  2021-05-31 23:58:00 -0.000536   54.744085   \n",
            "\n",
            "                Gap  Scaled_Target  \n",
            "1636322 -234.526996      -0.643562  \n",
            "1636323 -140.872032      -0.322826  \n",
            "1636324   91.408800      -0.047379  \n",
            "1636325  -41.466626      -0.889201  \n",
            "1636326   50.244793       0.292061  \n",
            "...             ...            ...  \n",
            "1794711    3.082484      -0.384565  \n",
            "1794712    8.913481      -0.307381  \n",
            "1794713  -32.293671      -0.297240  \n",
            "1794714   10.553188      -0.367566  \n",
            "1794715   -2.659840      -0.268846  \n",
            "\n",
            "[158394 rows x 7 columns]\n",
            "\n",
            "\n",
            "BTC test: \n",
            "\n",
            "         Count          VWAP                 Time    Target      Volume  \\\n",
            "1794717   4467  37192.752967  2021-06-01 00:01:00  0.001795  226.663656   \n",
            "1794718   7669  37347.492196  2021-06-01 00:02:00 -0.001205  231.370492   \n",
            "1794719  14295  37559.656383  2021-06-01 00:03:00 -0.001445  829.473588   \n",
            "1794720   6329  37578.985019  2021-06-01 00:04:00  0.000694  430.111193   \n",
            "1794721   8381  37615.939720  2021-06-01 00:05:00  0.001762  553.791084   \n",
            "...        ...           ...                  ...       ...         ...   \n",
            "1955973   2643  42653.031385  2021-09-20 23:40:00  0.002084  100.797218   \n",
            "1955974   2281  42755.785162  2021-09-20 23:41:00  0.003246   76.339988   \n",
            "1955975   2642  42749.075916  2021-09-20 23:42:00  0.003108  117.429123   \n",
            "1955976   2134  42749.024591  2021-09-20 23:43:00  0.002770   78.049458   \n",
            "1955977   2165  42680.263993  2021-09-20 23:44:00  0.002726   86.518794   \n",
            "\n",
            "                Gap  Scaled_Target  \n",
            "1794717 -111.749877       0.892985  \n",
            "1794718  154.739229      -0.601842  \n",
            "1794719  212.164187      -0.721781  \n",
            "1794720   19.328636       0.344081  \n",
            "1794721   36.954701       0.876603  \n",
            "...             ...            ...  \n",
            "1955973  -57.261581       1.036806  \n",
            "1955974  102.753777       1.616077  \n",
            "1955975   -6.709245       1.547370  \n",
            "1955976   -0.051325       1.378983  \n",
            "1955977  -68.760598       1.356803  \n",
            "\n",
            "[161261 rows x 7 columns]\n",
            "\n",
            " **************************************************************************************************** \n",
            " ****************************************************************************************************\n",
            "valid_threshold_index: 1636210\n",
            "test_threshold_index: 1794598\n",
            "\n",
            "\n",
            "ETH train: \n",
            "\n",
            "         Count         VWAP                 Time    Target      Volume  \\\n",
            "1          192   738.268967  2018-01-01 00:02:00 -0.004441  232.793141   \n",
            "2          120   737.994457  2018-01-01 00:03:00 -0.004206  174.138031   \n",
            "3          156   737.303631  2018-01-01 00:04:00 -0.002205  165.383926   \n",
            "4          118   736.163026  2018-01-01 00:05:00 -0.001744  193.078039   \n",
            "5          119   734.816231  2018-01-01 00:06:00 -0.001331  366.585886   \n",
            "...        ...          ...                  ...       ...         ...   \n",
            "1636205    952  1744.819326  2021-02-10 23:56:00 -0.007403  361.585855   \n",
            "1636206   1341  1744.446378  2021-02-10 23:57:00 -0.010088  509.741571   \n",
            "1636207   1087  1744.657609  2021-02-10 23:58:00 -0.010048  573.788708   \n",
            "1636208   1128  1743.100228  2021-02-10 23:59:00 -0.008924  662.240731   \n",
            "1636209   1546  1743.167509  2021-02-11 00:00:00 -0.012347  861.938779   \n",
            "\n",
            "              Gap  Scaled_Target  \n",
            "1       -0.570324      -1.785000  \n",
            "2       -0.274510      -1.690477  \n",
            "3       -0.690826      -0.885934  \n",
            "4       -1.140605      -0.700536  \n",
            "5       -1.346795      -0.534255  \n",
            "...           ...            ...  \n",
            "1636205  0.349075      -2.976315  \n",
            "1636206 -0.372947      -4.056625  \n",
            "1636207  0.211231      -4.040318  \n",
            "1636208 -1.557381      -3.588295  \n",
            "1636209  0.067282      -4.964991  \n",
            "\n",
            "[1636209 rows x 7 columns]\n",
            "\n",
            "\n",
            "ETH valid: \n",
            "\n",
            "         Count         VWAP                 Time    Target       Volume  \\\n",
            "1636211   3993  1739.975016  2021-02-11 00:02:00 -0.002272  2840.573518   \n",
            "1636212   5510  1733.024759  2021-02-11 00:03:00 -0.003600  6089.341354   \n",
            "1636213   2467  1736.350556  2021-02-11 00:04:00 -0.004419  1663.983443   \n",
            "1636214   1716  1736.887789  2021-02-11 00:05:00 -0.005024  1465.379388   \n",
            "1636215   1347  1738.382404  2021-02-11 00:06:00 -0.003708   802.958655   \n",
            "...        ...          ...                  ...       ...          ...   \n",
            "1794593   3043  2709.249902  2021-05-31 23:54:00  0.000417  2478.497480   \n",
            "1794594   2950  2705.669506  2021-05-31 23:55:00 -0.000122  1552.162655   \n",
            "1794595   3638  2701.145405  2021-05-31 23:56:00 -0.001684  2215.333008   \n",
            "1794596   5962  2704.079620  2021-05-31 23:57:00 -0.002880  3568.405783   \n",
            "1794597   3104  2710.840698  2021-05-31 23:58:00 -0.002703  2286.300368   \n",
            "\n",
            "              Gap  Scaled_Target  \n",
            "1636211 -6.743441      -0.912690  \n",
            "1636212 -6.950257      -1.447020  \n",
            "1636213  3.325796      -1.776468  \n",
            "1636214  0.537234      -2.019814  \n",
            "1636215  1.494615      -1.490445  \n",
            "...           ...            ...  \n",
            "1794593 -0.252206       0.168642  \n",
            "1794594 -3.580396      -0.047996  \n",
            "1794595 -4.524101      -0.676109  \n",
            "1794596  2.934216      -1.157112  \n",
            "1794597  6.761078      -1.086254  \n",
            "\n",
            "[158387 rows x 7 columns]\n",
            "\n",
            "\n",
            "ETH test: \n",
            "\n",
            "         Count         VWAP                 Time    Target       Volume  \\\n",
            "1794599   4933  2696.263219  2021-06-01 00:01:00 -0.001576  3517.491488   \n",
            "1794600   8733  2718.126072  2021-06-01 00:02:00 -0.001328  5355.471191   \n",
            "1794601  11098  2725.585309  2021-06-01 00:03:00 -0.000896  6570.743206   \n",
            "1794602   4412  2722.948702  2021-06-01 00:04:00  0.000244  2335.272742   \n",
            "1794603   5315  2726.046925  2021-06-01 00:05:00 -0.000737  5145.622819   \n",
            "...        ...          ...                  ...       ...          ...   \n",
            "1955855   2291  2956.842571  2021-09-20 23:40:00 -0.001309   717.616863   \n",
            "1955856   2534  2964.322014  2021-09-20 23:41:00 -0.000736   776.638911   \n",
            "1955857   2321  2965.543538  2021-09-20 23:42:00 -0.000491   976.724995   \n",
            "1955858   1971  2964.359745  2021-09-20 23:43:00 -0.000562   674.087973   \n",
            "1955859   2023  2958.430582  2021-09-20 23:44:00 -0.000346   738.290907   \n",
            "\n",
            "               Gap  Scaled_Target  \n",
            "1794599 -14.342284      -0.632695  \n",
            "1794600  21.862853      -0.533088  \n",
            "1794601   7.459237      -0.359338  \n",
            "1794602  -2.636607       0.099192  \n",
            "1794603   3.098223      -0.295371  \n",
            "...            ...            ...  \n",
            "1955855  -5.562257      -0.525206  \n",
            "1955856   7.479443      -0.294812  \n",
            "1955857   1.221523      -0.196457  \n",
            "1955858  -1.183792      -0.224888  \n",
            "1955859  -5.929163      -0.138031  \n",
            "\n",
            "[161261 rows x 7 columns]\n"
          ]
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "## Observation & Next Price\n",
        "\n",
        "Instead of using K-Fold cross-validation, we should be aware of the fact that a time series is not randomly distributed data — It is inherently ordered by chronological order. Therefore, using K-Fold cross validation is not helpful in our case.\n",
        "\n",
        "In the case of time series data, we use techniques like forward-chaining \n",
        " where the model will learn on past data then look at forward-facing data.\n",
        "\n",
        "e.g.\n",
        "    \n",
        "    Dataset: [1 2 3 4 5 6 7 8 ...]\n",
        "    1st Observation vector: [1 2 3 4 5]   \n",
        "        next price 6\n",
        "    2nd  Observation vector: [2 3 4 5 6]\n",
        "        next price 7\n",
        "      ...\n",
        "      ..\n",
        "      .\n",
        "\n",
        "We prefer calling the training vector above as **observation** and the test set above as **next price** since it is more catchy.\n",
        "\n",
        "\n",
        "We also set the dimension of the observation vector as **60** to make the training time of the model convenient and we thought it might be beneficial since it fits into one hour, and one hour in the trading market might present much good info."
      ],
      "metadata": {
        "id": "J00u0cdXSiW7"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "%%time\n",
        "def observation_and_next_price(df : pd.DataFrame, dimension : int, df_name : str) -> np.array:\n",
        "  \"\"\"\n",
        "  Generating observation vector with the specified dimension by the user in order to predict next price\n",
        "\n",
        "  Arg:\n",
        "    df: dataframe that includes the data that we will use to generate observation vectors\n",
        "    dimension: lenght of observation vector\n",
        "    df_name: coin name that will be printed\n",
        "\n",
        "  Return:\n",
        "\n",
        "  \"\"\"\n",
        "\n",
        "  # picking data to feed into model\n",
        "  data = df['Scaled_Target'].values\n",
        "  \n",
        "  d = dimension\n",
        "  observed = []\n",
        "  next_price = []\n",
        "\n",
        "  for i in range(len(data) - d):\n",
        "    observed.append(data[ i : (i + d) ])\n",
        "    next_price.append(data[i + d])\n",
        "\n",
        "  observed = np.asarray(observed)\n",
        "  next_price = np.asarray(next_price).reshape(-1,1)\n",
        "\n",
        "  # make sure the dimensions are accurate\n",
        "  print(f'{df_name} shape : {data.shape}')\n",
        "  print(f'{df_name}_observed shape : {observed.shape}')\n",
        "  print(f'{df_name}_next_price shape : {next_price.shape}')\n",
        "  print('\\n')\n",
        "\n",
        "  return observed, next_price\n",
        "\n",
        "\n",
        "# dimension of observation vector\n",
        "d_o = 60\n",
        "\n",
        "# btc\n",
        "btc_train_observed, btc_train_next_price = observation_and_next_price(btc_train, d_o, 'btc_train')\n",
        "btc_valid_observed, btc_valid_next_price = observation_and_next_price(btc_valid, d_o, 'btc_valid')\n",
        "btc_test_observed, btc_test_next_price = observation_and_next_price(btc_test, d_o, 'btc_test')\n",
        "\n",
        "print('*'*120, '\\n')\n",
        "\n",
        "# eth\n",
        "eth_train_observed, eth_train_next_price = observation_and_next_price(eth_train, d_o, 'eth_train')\n",
        "eth_valid_observed, eth_valid_next_price = observation_and_next_price(eth_valid, d_o, 'eth_valid')\n",
        "eth_test_observed, eth_test_next_price = observation_and_next_price(eth_test, d_o, 'eth_test')\n"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "E26NGsRxAH3B",
        "outputId": "0ff89260-9210-44a6-a1f7-10ac65e22607"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "btc_train shape : (1636320,)\n",
            "btc_train_observed shape : (1636260, 60)\n",
            "btc_train_next_price shape : (1636260, 1)\n",
            "\n",
            "\n",
            "btc_valid shape : (158394,)\n",
            "btc_valid_observed shape : (158334, 60)\n",
            "btc_valid_next_price shape : (158334, 1)\n",
            "\n",
            "\n",
            "btc_test shape : (161261,)\n",
            "btc_test_observed shape : (161201, 60)\n",
            "btc_test_next_price shape : (161201, 1)\n",
            "\n",
            "\n",
            "************************************************************************************************************************ \n",
            "\n",
            "eth_train shape : (1636209,)\n",
            "eth_train_observed shape : (1636149, 60)\n",
            "eth_train_next_price shape : (1636149, 1)\n",
            "\n",
            "\n",
            "eth_valid shape : (158387,)\n",
            "eth_valid_observed shape : (158327, 60)\n",
            "eth_valid_next_price shape : (158327, 1)\n",
            "\n",
            "\n",
            "eth_test shape : (161261,)\n",
            "eth_test_observed shape : (161201, 60)\n",
            "eth_test_next_price shape : (161201, 1)\n",
            "\n",
            "\n",
            "CPU times: user 5.21 s, sys: 194 ms, total: 5.41 s\n",
            "Wall time: 5.41 s\n"
          ]
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "## Oversee preprocessed data"
      ],
      "metadata": {
        "id": "3_mlJ9FB4GHg"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "# check all data manually if all data preprocess are done correctly or is there any information leakage, before we feed them through into model \n",
        "\n",
        "print(f'\\n\\n train: \\n')\n",
        "with pd.option_context('display.max_columns', None): \n",
        "  print(btc_train)\n",
        "\n",
        "print(f'\\n\\n valid: \\n')\n",
        "with pd.option_context('display.max_columns', None): \n",
        "  print(btc_valid)\n",
        "\n",
        "print(f'\\n\\n test: \\n')\n",
        "with pd.option_context('display.max_columns', None): \n",
        "  print(btc_test)\n",
        "\n",
        "print('*'*125, '\\n', '*'*125)\n",
        "\n",
        "print(f'train_observed, dimension: {btc_train_observed.shape}')\n",
        "with pd.option_context('display.max_columns', None): \n",
        "  print(btc_train_observed)\n",
        "  print('\\n',f'train_next_price, dimension: {btc_train_next_price.shape}')\n",
        "  print(btc_train_next_price)\n",
        "\n",
        "print('*'*100,f'\\n\\nvalid_observed, dimension: {btc_valid_observed.shape}')\n",
        "with pd.option_context('display.max_columns', None): \n",
        "  print(btc_valid_observed)\n",
        "  print('\\n',f'valid_next_price, dimension: {btc_valid_next_price.shape}')\n",
        "  print(btc_valid_next_price)\n",
        "\n",
        "print('\\n', '*'*100, f'\\n\\ntest_observed, dimension: {btc_test_observed.shape}')\n",
        "with pd.option_context('display.max_columns', None): \n",
        "  print(btc_test_observed)\n",
        "  print('\\n',f'test_next_price, dimension: {btc_test_next_price.shape}')\n",
        "  print(btc_test_next_price)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "QD35qr8f0Vgz",
        "outputId": "a5b77572-ffc6-49c6-b610-7a1019783908"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "\n",
            "\n",
            " train: \n",
            "\n",
            "         Count          VWAP                 Time    Target      Volume  \\\n",
            "1          235  13840.362591  2018-01-01 00:02:00 -0.015037   31.046432   \n",
            "2          528  13806.068014  2018-01-01 00:03:00 -0.010309   55.061820   \n",
            "3          435  13783.598101  2018-01-01 00:04:00 -0.008999   38.780529   \n",
            "4          742  13735.586842  2018-01-01 00:05:00 -0.008079  108.501637   \n",
            "5          554  13706.952030  2018-01-01 00:06:00 -0.004422   70.805776   \n",
            "...        ...           ...                  ...       ...         ...   \n",
            "1636316   1864  44935.449789  2021-02-10 23:56:00 -0.007227   39.538163   \n",
            "1636317   1826  44913.353225  2021-02-10 23:57:00 -0.008847   48.526747   \n",
            "1636318   2467  44912.705558  2021-02-10 23:58:00 -0.007998   92.074357   \n",
            "1636319   2466  44848.417638  2021-02-10 23:59:00 -0.007590   86.665774   \n",
            "1636320   2956  44862.019363  2021-02-11 00:00:00 -0.010984  111.507314   \n",
            "\n",
            "               Gap  Scaled_Target  \n",
            "1        13.300499      -7.494206  \n",
            "2       -34.294578      -5.138308  \n",
            "3       -22.469913      -4.485835  \n",
            "4       -48.011259      -4.027472  \n",
            "5       -28.634812      -2.205036  \n",
            "...            ...            ...  \n",
            "1636316  40.833138      -3.602531  \n",
            "1636317 -22.096564      -4.409850  \n",
            "1636318  -0.647667      -3.986812  \n",
            "1636319 -64.287919      -3.783835  \n",
            "1636320  13.601724      -5.474796  \n",
            "\n",
            "[1636320 rows x 7 columns]\n",
            "\n",
            "\n",
            " valid: \n",
            "\n",
            "         Count          VWAP                 Time    Target      Volume  \\\n",
            "1636322  10837  44754.964885  2021-02-11 00:02:00 -0.001288  400.223029   \n",
            "1636323   8032  44614.092853  2021-02-11 00:03:00 -0.000645  259.376814   \n",
            "1636324   4636  44705.501653  2021-02-11 00:04:00 -0.000092  203.337976   \n",
            "1636325   4533  44664.035028  2021-02-11 00:05:00 -0.001781  196.089887   \n",
            "1636326   2698  44714.279821  2021-02-11 00:06:00  0.000589   93.771220   \n",
            "...        ...           ...                  ...       ...         ...   \n",
            "1794711   1712  37319.205839  2021-05-31 23:54:00 -0.000769   51.601578   \n",
            "1794712   2297  37328.119321  2021-05-31 23:55:00 -0.000614   59.683761   \n",
            "1794713   2080  37295.825650  2021-05-31 23:56:00 -0.000593   59.175643   \n",
            "1794714   2957  37306.378838  2021-05-31 23:57:00 -0.000735  104.229560   \n",
            "1794715   2136  37303.718998  2021-05-31 23:58:00 -0.000536   54.744085   \n",
            "\n",
            "                Gap  Scaled_Target  \n",
            "1636322 -234.526996      -0.643562  \n",
            "1636323 -140.872032      -0.322826  \n",
            "1636324   91.408800      -0.047379  \n",
            "1636325  -41.466626      -0.889201  \n",
            "1636326   50.244793       0.292061  \n",
            "...             ...            ...  \n",
            "1794711    3.082484      -0.384565  \n",
            "1794712    8.913481      -0.307381  \n",
            "1794713  -32.293671      -0.297240  \n",
            "1794714   10.553188      -0.367566  \n",
            "1794715   -2.659840      -0.268846  \n",
            "\n",
            "[158394 rows x 7 columns]\n",
            "\n",
            "\n",
            " test: \n",
            "\n",
            "         Count          VWAP                 Time    Target      Volume  \\\n",
            "1794717   4467  37192.752967  2021-06-01 00:01:00  0.001795  226.663656   \n",
            "1794718   7669  37347.492196  2021-06-01 00:02:00 -0.001205  231.370492   \n",
            "1794719  14295  37559.656383  2021-06-01 00:03:00 -0.001445  829.473588   \n",
            "1794720   6329  37578.985019  2021-06-01 00:04:00  0.000694  430.111193   \n",
            "1794721   8381  37615.939720  2021-06-01 00:05:00  0.001762  553.791084   \n",
            "...        ...           ...                  ...       ...         ...   \n",
            "1955973   2643  42653.031385  2021-09-20 23:40:00  0.002084  100.797218   \n",
            "1955974   2281  42755.785162  2021-09-20 23:41:00  0.003246   76.339988   \n",
            "1955975   2642  42749.075916  2021-09-20 23:42:00  0.003108  117.429123   \n",
            "1955976   2134  42749.024591  2021-09-20 23:43:00  0.002770   78.049458   \n",
            "1955977   2165  42680.263993  2021-09-20 23:44:00  0.002726   86.518794   \n",
            "\n",
            "                Gap  Scaled_Target  \n",
            "1794717 -111.749877       0.892985  \n",
            "1794718  154.739229      -0.601842  \n",
            "1794719  212.164187      -0.721781  \n",
            "1794720   19.328636       0.344081  \n",
            "1794721   36.954701       0.876603  \n",
            "...             ...            ...  \n",
            "1955973  -57.261581       1.036806  \n",
            "1955974  102.753777       1.616077  \n",
            "1955975   -6.709245       1.547370  \n",
            "1955976   -0.051325       1.378983  \n",
            "1955977  -68.760598       1.356803  \n",
            "\n",
            "[161261 rows x 7 columns]\n",
            "***************************************************************************************************************************** \n",
            " *****************************************************************************************************************************\n",
            "train_observed, dimension: (1636260, 60)\n",
            "[[-7.49420551 -5.13830834 -4.48583519 ... -2.36182858 -2.28970644\n",
            "  -1.13365392]\n",
            " [-5.13830834 -4.48583519 -4.02747151 ... -2.28970644 -1.13365392\n",
            "  -0.84417048]\n",
            " [-4.48583519 -4.02747151 -2.2050357  ... -1.13365392 -0.84417048\n",
            "   0.12868883]\n",
            " ...\n",
            " [-0.74400682 -0.5289822  -0.25305806 ... -3.94146819 -3.60253108\n",
            "  -4.40984989]\n",
            " [-0.5289822  -0.25305806 -0.2322083  ... -3.60253108 -4.40984989\n",
            "  -3.9868118 ]\n",
            " [-0.25305806 -0.2322083   1.14260109 ... -4.40984989 -3.9868118\n",
            "  -3.78383492]]\n",
            "\n",
            " train_next_price, dimension: (1636260, 1)\n",
            "[[-0.84417048]\n",
            " [ 0.12868883]\n",
            " [ 2.28007707]\n",
            " ...\n",
            " [-3.9868118 ]\n",
            " [-3.78383492]\n",
            " [-5.47479646]]\n",
            "**************************************************************************************************** \n",
            "\n",
            "valid_observed, dimension: (158334, 60)\n",
            "[[-0.64356203 -0.32282576 -0.04737893 ... -2.46808395  0.01727086\n",
            "  -0.16370147]\n",
            " [-0.32282576 -0.04737893 -0.88920114 ...  0.01727086 -0.16370147\n",
            "  -0.65125198]\n",
            " [-0.04737893 -0.88920114  0.29206124 ... -0.16370147 -0.65125198\n",
            "  -0.45664467]\n",
            " ...\n",
            " [-0.54802637 -0.98290736 -0.89119028 ... -0.10118094 -0.3845648\n",
            "  -0.30738092]\n",
            " [-0.98290736 -0.89119028 -0.61303336 ... -0.3845648  -0.30738092\n",
            "  -0.29724049]\n",
            " [-0.89119028 -0.61303336 -1.10355116 ... -0.30738092 -0.29724049\n",
            "  -0.36756551]]\n",
            "\n",
            " valid_next_price, dimension: (158334, 1)\n",
            "[[-0.65125198]\n",
            " [-0.45664467]\n",
            " [-0.9691881 ]\n",
            " ...\n",
            " [-0.29724049]\n",
            " [-0.36756551]\n",
            " [-0.26884649]]\n",
            "\n",
            " **************************************************************************************************** \n",
            "\n",
            "test_observed, dimension: (161201, 60)\n",
            "[[ 0.89298518 -0.6018416  -0.72178097 ... -0.71179544 -1.39517299\n",
            "  -1.78829194]\n",
            " [-0.6018416  -0.72178097  0.34408146 ... -1.39517299 -1.78829194\n",
            "  -1.19513225]\n",
            " [-0.72178097  0.34408146  0.8766028  ... -1.78829194 -1.19513225\n",
            "  -1.27720057]\n",
            " ...\n",
            " [-0.4824691  -0.45942066 -0.32751463 ...  0.85245268  1.03680606\n",
            "   1.61607654]\n",
            " [-0.45942066 -0.32751463 -0.34402356 ...  1.03680606  1.61607654\n",
            "   1.54737032]\n",
            " [-0.32751463 -0.34402356 -0.27144067 ...  1.61607654  1.54737032\n",
            "   1.37898283]]\n",
            "\n",
            " test_next_price, dimension: (161201, 1)\n",
            "[[-1.19513225]\n",
            " [-1.27720057]\n",
            " [-0.65670581]\n",
            " ...\n",
            " [ 1.54737032]\n",
            " [ 1.37898283]\n",
            " [ 1.35680272]]\n"
          ]
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "Eventually, there is no information leak that we observed. Now time to build the models."
      ],
      "metadata": {
        "id": "1PxCiOh0tSaI"
      }
    },
    {
      "cell_type": "markdown",
      "source": [
        "# Modelling\n",
        "\n",
        "There are many modelling methods that might be applied to time series. Since we are students, not competitors, and we want to learn about RNN, we came up with the idea which using some promising RNN modelling methods regarding to predicting time series. Basically, these techniques are LSTM and GRU.\n",
        "\n",
        "Also, we could use other techniques such as fully connecting dense layers, autoencoder or even CNN but those are mainly formed for different types of purposes like image processing and so on. On the other hand, using RNN on time series is quite popular and it has been applied successfully for decades.\n",
        "\n",
        "Intuitively, we applied LSTM on btc, and GRU on eth, since we want to get experience in using them. Since LSTM and GRU are inspired by RNN, before we use them first we should understand what is RNN.\n",
        "\n",
        "**Recurrent Neural Network**: basically,\n",
        "\n",
        "- RNNs are a robust and powerful type of neural network and are considered one of the most professional algorithms because they are the only ones with internal memory.\n",
        "\n",
        "- The algorithm performs very well for sequential data such as time series, speech, text, financial data, audio, video, weather, and more.\n",
        "\n",
        "- In an RNN, the information goes through a cycle. When making a decision, it considers the current input and also what it has learned from the inputs it has received previously.\n",
        "\n",
        "\n",
        "Currently, there are 3 types of RNN layers in the Keras library.\n",
        "\n",
        "- SimpleRNN\n",
        "- LSTM\n",
        "- GRU\n",
        "\n",
        "\n",
        "\n",
        "\n",
        "\n",
        "Now, let's dive into the details of the models and check how they perform\n"
      ],
      "metadata": {
        "id": "G-gi4CURWaox"
      }
    },
    {
      "cell_type": "markdown",
      "source": [
        "## Modelling 1: LSTM for BTC\n",
        "\n",
        "In the first section, we use LSTM (Long short-term memory ). LSTM units are units of a recurrent neural network (RNN). An RNN composed of LSTM units is often called an LSTM network (or just LSTM). A common LSTM unit is composed of a cell, an input gate, an output gate and a forget gate. The cell remembers values over arbitrary time intervals and the three gates regulate the flow of information into and out of the cell.\n",
        "\n",
        "\n",
        "Long short-term memory networks are an extension of recurrent neural networks, which basically extend the memory. Therefore it is well suited to learn from sequential data such as time series in our case.\n",
        "\n",
        "\n",
        "\n",
        "Let’s see how our architecture looks like:\n",
        "\n",
        "-- Hyper parameters --\n",
        "\n",
        "**Layers** : Due to limited computational sources and a high amount of data, having more than 2 LSTM layers causes crashes on our colab notebook. Therefore, we used two LSTM layers to keep our colab notebook sustainable and eventually it performed enough well.\n",
        "\n",
        "**num_nodes**: Based on our search, mainly 100 or 200 is chosen as the number of nodes. But eventually having a higher number of nodes doesn't really change the performance. Then intuitively, we prefered 60 for the number of nodes and it worked, giving a fairly good result. We think that it might be because of the dimension of observation vectors, which is also 60.\n",
        "\n",
        "**Activation function**: We observed that tanh is performed better than relu. We think that since the input data's scale is zero centred like tanh, tanh fits better into it.\n",
        "\n",
        "**Dropout**: is simply a layer that tries to prevent the system from memorizing.\n",
        "\n",
        "**Loss Function**: Mean Squered Error: Since we are dealing with a regression problem, MSE is the most commonly used loss function, so we only use MSE since we work with numerical data\n",
        "\n",
        "**Optimizer**: Adam: is a very recent and well-performing optimizer and proper for our task\n",
        "\n",
        "**Batch size**: We prefered trying high batch size rates, because we have +1.6m rows in train and we had to consider training time in order to try different type of hyper parameters. That is why tried size 8192, 16384 and 32768 and we found 16384 optima since it is more time friendly and for each epoch, there is 100 iteration to go through all data points.\n",
        "\n",
        "**Epochs**: In order to make the model learn better with many iterations, we picked epoch as 50, but after 30 epochs performance stayed constant."
      ],
      "metadata": {
        "id": "GJ4zBXv_ummK"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "%%time\n",
        "model_lstm = tf.keras.Sequential([\n",
        "              tf.keras.layers.InputLayer(input_shape = (btc_train_observed.shape[1], 1)),\n",
        "              tf.keras.layers.LSTM(60, return_sequences=True, activation='tanh'),\n",
        "              tf.keras.layers.Dropout(0.25),\n",
        "              tf.keras.layers.LSTM(60, activation='tanh'),\n",
        "              tf.keras.layers.Dropout(0.25),\n",
        "              tf.keras.layers.Dense(1)\n",
        "])\n",
        "\n",
        "model_lstm.summary()"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "V0MyNrDfWMve",
        "outputId": "e813aa38-391c-4a3b-a10f-36d3fc1740f8"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Model: \"sequential\"\n",
            "_________________________________________________________________\n",
            " Layer (type)                Output Shape              Param #   \n",
            "=================================================================\n",
            " lstm (LSTM)                 (None, 60, 60)            14880     \n",
            "                                                                 \n",
            " dropout (Dropout)           (None, 60, 60)            0         \n",
            "                                                                 \n",
            " lstm_1 (LSTM)               (None, 60)                29040     \n",
            "                                                                 \n",
            " dropout_1 (Dropout)         (None, 60)                0         \n",
            "                                                                 \n",
            " dense (Dense)               (None, 1)                 61        \n",
            "                                                                 \n",
            "=================================================================\n",
            "Total params: 43,981\n",
            "Trainable params: 43,981\n",
            "Non-trainable params: 0\n",
            "_________________________________________________________________\n",
            "CPU times: user 1.12 s, sys: 355 ms, total: 1.47 s\n",
            "Wall time: 3.5 s\n"
          ]
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "### LSTM model train\n",
        "Now that we have our data ready, and our model compiled, we can start training"
      ],
      "metadata": {
        "id": "VOpIqkNnhlXj"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "%%time\n",
        "model_lstm.compile(loss='mean_squared_error', optimizer='adam')\n",
        "history_lstm = model_lstm.fit(btc_train_observed, btc_train_next_price, epochs=50, batch_size=16384, validation_data=(btc_valid_observed, btc_valid_next_price))"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "kDGKpBNuYu-S",
        "outputId": "f3b27282-910a-412b-cc88-788c0d796be8"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Epoch 1/50\n",
            "100/100 [==============================] - 55s 478ms/step - loss: 0.4573 - val_loss: 0.2862\n",
            "Epoch 2/50\n",
            "100/100 [==============================] - 47s 469ms/step - loss: 0.3216 - val_loss: 0.2201\n",
            "Epoch 3/50\n",
            "100/100 [==============================] - 47s 471ms/step - loss: 0.3010 - val_loss: 0.2124\n",
            "Epoch 4/50\n",
            "100/100 [==============================] - 47s 470ms/step - loss: 0.2947 - val_loss: 0.2122\n",
            "Epoch 5/50\n",
            "100/100 [==============================] - 47s 471ms/step - loss: 0.2900 - val_loss: 0.2017\n",
            "Epoch 6/50\n",
            "100/100 [==============================] - 48s 476ms/step - loss: 0.2830 - val_loss: 0.1990\n",
            "Epoch 7/50\n",
            "100/100 [==============================] - 47s 474ms/step - loss: 0.2780 - val_loss: 0.2011\n",
            "Epoch 8/50\n",
            "100/100 [==============================] - 47s 471ms/step - loss: 0.2757 - val_loss: 0.1972\n",
            "Epoch 9/50\n",
            "100/100 [==============================] - 47s 471ms/step - loss: 0.2730 - val_loss: 0.1953\n",
            "Epoch 10/50\n",
            "100/100 [==============================] - 47s 471ms/step - loss: 0.2717 - val_loss: 0.1957\n",
            "Epoch 11/50\n",
            "100/100 [==============================] - 47s 472ms/step - loss: 0.2710 - val_loss: 0.1933\n",
            "Epoch 12/50\n",
            "100/100 [==============================] - 47s 472ms/step - loss: 0.2689 - val_loss: 0.1919\n",
            "Epoch 13/50\n",
            "100/100 [==============================] - 47s 471ms/step - loss: 0.2689 - val_loss: 0.1946\n",
            "Epoch 14/50\n",
            "100/100 [==============================] - 47s 471ms/step - loss: 0.2682 - val_loss: 0.1924\n",
            "Epoch 15/50\n",
            "100/100 [==============================] - 47s 471ms/step - loss: 0.2677 - val_loss: 0.1927\n",
            "Epoch 16/50\n",
            "100/100 [==============================] - 47s 473ms/step - loss: 0.2668 - val_loss: 0.1928\n",
            "Epoch 17/50\n",
            "100/100 [==============================] - 47s 472ms/step - loss: 0.2662 - val_loss: 0.1906\n",
            "Epoch 18/50\n",
            "100/100 [==============================] - 47s 471ms/step - loss: 0.2654 - val_loss: 0.1918\n",
            "Epoch 19/50\n",
            "100/100 [==============================] - 47s 471ms/step - loss: 0.2652 - val_loss: 0.1931\n",
            "Epoch 20/50\n",
            "100/100 [==============================] - 47s 472ms/step - loss: 0.2649 - val_loss: 0.1938\n",
            "Epoch 21/50\n",
            "100/100 [==============================] - 47s 472ms/step - loss: 0.2646 - val_loss: 0.1910\n",
            "Epoch 22/50\n",
            "100/100 [==============================] - 47s 471ms/step - loss: 0.2639 - val_loss: 0.1918\n",
            "Epoch 23/50\n",
            "100/100 [==============================] - 47s 471ms/step - loss: 0.2638 - val_loss: 0.1889\n",
            "Epoch 24/50\n",
            "100/100 [==============================] - 47s 471ms/step - loss: 0.2642 - val_loss: 0.1893\n",
            "Epoch 25/50\n",
            "100/100 [==============================] - 47s 472ms/step - loss: 0.2632 - val_loss: 0.1901\n",
            "Epoch 26/50\n",
            "100/100 [==============================] - 47s 472ms/step - loss: 0.2626 - val_loss: 0.1908\n",
            "Epoch 27/50\n",
            "100/100 [==============================] - 47s 471ms/step - loss: 0.2630 - val_loss: 0.1908\n",
            "Epoch 28/50\n",
            "100/100 [==============================] - 47s 471ms/step - loss: 0.2618 - val_loss: 0.1902\n",
            "Epoch 29/50\n",
            "100/100 [==============================] - 47s 472ms/step - loss: 0.2610 - val_loss: 0.1870\n",
            "Epoch 30/50\n",
            "100/100 [==============================] - 47s 471ms/step - loss: 0.2580 - val_loss: 0.1878\n",
            "Epoch 31/50\n",
            "100/100 [==============================] - 47s 472ms/step - loss: 0.2540 - val_loss: 0.1828\n",
            "Epoch 32/50\n",
            "100/100 [==============================] - 47s 472ms/step - loss: 0.2529 - val_loss: 0.1831\n",
            "Epoch 33/50\n",
            "100/100 [==============================] - 47s 471ms/step - loss: 0.2520 - val_loss: 0.1832\n",
            "Epoch 34/50\n",
            "100/100 [==============================] - 47s 471ms/step - loss: 0.2514 - val_loss: 0.1828\n",
            "Epoch 35/50\n",
            "100/100 [==============================] - 47s 471ms/step - loss: 0.2516 - val_loss: 0.1817\n",
            "Epoch 36/50\n",
            "100/100 [==============================] - 47s 471ms/step - loss: 0.2513 - val_loss: 0.1823\n",
            "Epoch 37/50\n",
            "100/100 [==============================] - 47s 471ms/step - loss: 0.2507 - val_loss: 0.1829\n",
            "Epoch 38/50\n",
            "100/100 [==============================] - 47s 471ms/step - loss: 0.2505 - val_loss: 0.1822\n",
            "Epoch 39/50\n",
            "100/100 [==============================] - 47s 472ms/step - loss: 0.2501 - val_loss: 0.1828\n",
            "Epoch 40/50\n",
            "100/100 [==============================] - 47s 472ms/step - loss: 0.2507 - val_loss: 0.1813\n",
            "Epoch 41/50\n",
            "100/100 [==============================] - 47s 471ms/step - loss: 0.2501 - val_loss: 0.1823\n",
            "Epoch 42/50\n",
            "100/100 [==============================] - 47s 471ms/step - loss: 0.2497 - val_loss: 0.1818\n",
            "Epoch 43/50\n",
            "100/100 [==============================] - 47s 472ms/step - loss: 0.2499 - val_loss: 0.1840\n",
            "Epoch 44/50\n",
            "100/100 [==============================] - 47s 472ms/step - loss: 0.2489 - val_loss: 0.1826\n",
            "Epoch 45/50\n",
            "100/100 [==============================] - 47s 471ms/step - loss: 0.2485 - val_loss: 0.1810\n",
            "Epoch 46/50\n",
            "100/100 [==============================] - 47s 471ms/step - loss: 0.2485 - val_loss: 0.1818\n",
            "Epoch 47/50\n",
            "100/100 [==============================] - 47s 471ms/step - loss: 0.2481 - val_loss: 0.1819\n",
            "Epoch 48/50\n",
            "100/100 [==============================] - 47s 471ms/step - loss: 0.2488 - val_loss: 0.1823\n",
            "Epoch 49/50\n",
            "100/100 [==============================] - 47s 471ms/step - loss: 0.2481 - val_loss: 0.1817\n",
            "Epoch 50/50\n",
            "100/100 [==============================] - 47s 471ms/step - loss: 0.2483 - val_loss: 0.1814\n",
            "CPU times: user 27min 6s, sys: 24.2 s, total: 27min 30s\n",
            "Wall time: 39min 24s\n"
          ]
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "### LSTM model performence\n",
        "\n",
        "Here is a chart you can use to compare loss and validation loss:"
      ],
      "metadata": {
        "id": "TZUzrQCX4hKN"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "fig = plt.figure(figsize=(18,6))\n",
        "\n",
        "# summarize history for loss\n",
        "plt.plot(history_lstm.history['val_loss'])\n",
        "plt.plot(history_lstm.history['loss'])\n",
        "plt.title('model loss')\n",
        "plt.ylabel('loss')\n",
        "plt.xlabel('epoch')\n",
        "plt.grid(True)\n",
        "plt.legend(['val_loss','loss'], loc='upper left')\n",
        "plt.show()#Our results may not be ideal, but they are enough for our purposes."
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 404
        },
        "id": "vivvmgle6DaH",
        "outputId": "7a84f8bc-fb49-4edf-a8d0-53c748ac9185"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 1296x432 with 1 Axes>"
            ]
          },
          "metadata": {
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "### LSMT model prediction\n",
        "\n",
        "Now with our trained model, we can start making some predictions and evaluating those predictions on the test data to see how well our model is doing:"
      ],
      "metadata": {
        "id": "BE0vx6OMhpYA"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "%%time\n",
        "\n",
        "# prediction on train\n",
        "btc_train_predictions = model_lstm.predict(btc_train_observed)\n",
        "btc_train_predictions_inversed = btc_scaler.inverse_transform(btc_train_predictions)\n",
        "\n",
        "# prediction on valid\n",
        "btc_valid_predictions = model_lstm.predict(btc_valid_observed)\n",
        "btc_valid_predictions_inversed = btc_scaler.inverse_transform(btc_valid_predictions)\n",
        "\n",
        "# prediction on test\n",
        "btc_test_predictions = model_lstm.predict(btc_test_observed)\n",
        "btc_test_predictions_inversed = btc_scaler.inverse_transform(btc_test_predictions)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "8au3JsPy_u25",
        "outputId": "230538b6-61c5-495f-f26f-31acc35ac7e4"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "CPU times: user 9min 26s, sys: 34.2 s, total: 10min 1s\n",
            "Wall time: 9min 28s\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "# We want to add predictions on the dataframe to analysis the model better\n",
        "# since our observation vector includes 60 elements inside and we start predicting from 61st index that is why we will fill the first 60 index with nan\n",
        "nan_array = np.empty(d_o)\n",
        "nan_array[:] = np.NaN\n",
        "\n",
        "btc_train['scaled_predictions'] = np.concatenate((nan_array, btc_train_predictions.flatten()))\n",
        "btc_train['inversed_predictions'] = np.concatenate((nan_array, btc_train_predictions_inversed.flatten()))\n",
        "\n",
        "print('*'*40,'\\t btc train: \\t','*'*40,'\\n')\n",
        "with pd.option_context('display.max_columns', None): \n",
        "  print(btc_train)\n",
        "  print('\\n\\n')\n",
        "\n",
        "btc_valid['scaled_predictions'] = np.concatenate((nan_array, btc_valid_predictions.flatten()))\n",
        "btc_valid['inversed_predictions'] = np.concatenate((nan_array, btc_valid_predictions_inversed.flatten()))\n",
        "\n",
        "print('*'*40,'\\t btc valid: \\t','*'*40,'\\n')\n",
        "with pd.option_context('display.max_columns', None): \n",
        "  print(btc_valid)\n",
        "  print('\\n\\n')\n",
        "\n",
        "print('*'*40,'\\t btc test: \\t','*'*40,'\\n')\n",
        "btc_test['scaled_predictions'] = np.concatenate((nan_array, btc_test_predictions.flatten()))\n",
        "btc_test['inversed_predictions'] = np.concatenate((nan_array, btc_test_predictions_inversed.flatten()))\n",
        "\n",
        "with pd.option_context('display.max_columns', None): \n",
        "  print(btc_test)\n",
        "  print('\\n\\n')"
      ],
      "metadata": {
        "id": "e-VIIvHC9qwn",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "e544c9fc-8ff5-4ee8-af3d-c03fbccccc8a"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "**************************************** \t btc train: \t **************************************** \n",
            "\n",
            "         Count          VWAP                 Time    Target      Volume  \\\n",
            "1          235  13840.362591  2018-01-01 00:02:00 -0.015037   31.046432   \n",
            "2          528  13806.068014  2018-01-01 00:03:00 -0.010309   55.061820   \n",
            "3          435  13783.598101  2018-01-01 00:04:00 -0.008999   38.780529   \n",
            "4          742  13735.586842  2018-01-01 00:05:00 -0.008079  108.501637   \n",
            "5          554  13706.952030  2018-01-01 00:06:00 -0.004422   70.805776   \n",
            "...        ...           ...                  ...       ...         ...   \n",
            "1636316   1864  44935.449789  2021-02-10 23:56:00 -0.007227   39.538163   \n",
            "1636317   1826  44913.353225  2021-02-10 23:57:00 -0.008847   48.526747   \n",
            "1636318   2467  44912.705558  2021-02-10 23:58:00 -0.007998   92.074357   \n",
            "1636319   2466  44848.417638  2021-02-10 23:59:00 -0.007590   86.665774   \n",
            "1636320   2956  44862.019363  2021-02-11 00:00:00 -0.010984  111.507314   \n",
            "\n",
            "               Gap  Scaled_Target  scaled_predictions  inversed_predictions  \n",
            "1        13.300499      -7.494206                 NaN                   NaN  \n",
            "2       -34.294578      -5.138308                 NaN                   NaN  \n",
            "3       -22.469913      -4.485835                 NaN                   NaN  \n",
            "4       -48.011259      -4.027472                 NaN                   NaN  \n",
            "5       -28.634812      -2.205036                 NaN                   NaN  \n",
            "...            ...            ...                 ...                   ...  \n",
            "1636316  40.833138      -3.602531           -3.756502             -0.007536  \n",
            "1636317 -22.096564      -4.409850           -3.534841             -0.007091  \n",
            "1636318  -0.647667      -3.986812           -4.009065             -0.008042  \n",
            "1636319 -64.287919      -3.783835           -3.672101             -0.007366  \n",
            "1636320  13.601724      -5.474796           -3.192489             -0.006404  \n",
            "\n",
            "[1636320 rows x 9 columns]\n",
            "\n",
            "\n",
            "\n",
            "**************************************** \t btc valid: \t **************************************** \n",
            "\n",
            "         Count          VWAP                 Time    Target      Volume  \\\n",
            "1636322  10837  44754.964885  2021-02-11 00:02:00 -0.001288  400.223029   \n",
            "1636323   8032  44614.092853  2021-02-11 00:03:00 -0.000645  259.376814   \n",
            "1636324   4636  44705.501653  2021-02-11 00:04:00 -0.000092  203.337976   \n",
            "1636325   4533  44664.035028  2021-02-11 00:05:00 -0.001781  196.089887   \n",
            "1636326   2698  44714.279821  2021-02-11 00:06:00  0.000589   93.771220   \n",
            "...        ...           ...                  ...       ...         ...   \n",
            "1794711   1712  37319.205839  2021-05-31 23:54:00 -0.000769   51.601578   \n",
            "1794712   2297  37328.119321  2021-05-31 23:55:00 -0.000614   59.683761   \n",
            "1794713   2080  37295.825650  2021-05-31 23:56:00 -0.000593   59.175643   \n",
            "1794714   2957  37306.378838  2021-05-31 23:57:00 -0.000735  104.229560   \n",
            "1794715   2136  37303.718998  2021-05-31 23:58:00 -0.000536   54.744085   \n",
            "\n",
            "                Gap  Scaled_Target  scaled_predictions  inversed_predictions  \n",
            "1636322 -234.526996      -0.643562                 NaN                   NaN  \n",
            "1636323 -140.872032      -0.322826                 NaN                   NaN  \n",
            "1636324   91.408800      -0.047379                 NaN                   NaN  \n",
            "1636325  -41.466626      -0.889201                 NaN                   NaN  \n",
            "1636326   50.244793       0.292061                 NaN                   NaN  \n",
            "...             ...            ...                 ...                   ...  \n",
            "1794711    3.082484      -0.384565           -0.176219             -0.000351  \n",
            "1794712    8.913481      -0.307381           -0.347027             -0.000693  \n",
            "1794713  -32.293671      -0.297240           -0.177368             -0.000353  \n",
            "1794714   10.553188      -0.367566           -0.073041             -0.000144  \n",
            "1794715   -2.659840      -0.268846           -0.164062             -0.000326  \n",
            "\n",
            "[158394 rows x 9 columns]\n",
            "\n",
            "\n",
            "\n",
            "**************************************** \t btc test: \t **************************************** \n",
            "\n",
            "         Count          VWAP                 Time    Target      Volume  \\\n",
            "1794717   4467  37192.752967  2021-06-01 00:01:00  0.001795  226.663656   \n",
            "1794718   7669  37347.492196  2021-06-01 00:02:00 -0.001205  231.370492   \n",
            "1794719  14295  37559.656383  2021-06-01 00:03:00 -0.001445  829.473588   \n",
            "1794720   6329  37578.985019  2021-06-01 00:04:00  0.000694  430.111193   \n",
            "1794721   8381  37615.939720  2021-06-01 00:05:00  0.001762  553.791084   \n",
            "...        ...           ...                  ...       ...         ...   \n",
            "1955973   2643  42653.031385  2021-09-20 23:40:00  0.002084  100.797218   \n",
            "1955974   2281  42755.785162  2021-09-20 23:41:00  0.003246   76.339988   \n",
            "1955975   2642  42749.075916  2021-09-20 23:42:00  0.003108  117.429123   \n",
            "1955976   2134  42749.024591  2021-09-20 23:43:00  0.002770   78.049458   \n",
            "1955977   2165  42680.263993  2021-09-20 23:44:00  0.002726   86.518794   \n",
            "\n",
            "                Gap  Scaled_Target  scaled_predictions  inversed_predictions  \n",
            "1794717 -111.749877       0.892985                 NaN                   NaN  \n",
            "1794718  154.739229      -0.601842                 NaN                   NaN  \n",
            "1794719  212.164187      -0.721781                 NaN                   NaN  \n",
            "1794720   19.328636       0.344081                 NaN                   NaN  \n",
            "1794721   36.954701       0.876603                 NaN                   NaN  \n",
            "...             ...            ...                 ...                   ...  \n",
            "1955973  -57.261581       1.036806            0.755815              0.001520  \n",
            "1955974  102.753777       1.616077            0.941650              0.001893  \n",
            "1955975   -6.709245       1.547370            1.496668              0.003007  \n",
            "1955976   -0.051325       1.378983            1.546770              0.003107  \n",
            "1955977  -68.760598       1.356803            1.368305              0.002749  \n",
            "\n",
            "[161261 rows x 9 columns]\n",
            "\n",
            "\n",
            "\n"
          ]
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "## Modelling 2: GRU for ETH"
      ],
      "metadata": {
        "id": "RaBexdQz2Sfs"
      }
    },
    {
      "cell_type": "markdown",
      "source": [
        "### GRU model train"
      ],
      "metadata": {
        "id": "mMGZfyIx2Xjt"
      }
    },
    {
      "cell_type": "markdown",
      "source": [
        "Although GRUs are very similar to LSTMs, they take fewer parameters and therefore generally run faster. However, LSTMs generally offer better performance on large datasets.\n",
        "\n",
        "Contrary to LSTM, GRUs only has two gates and these are Reset and Update gate. Basically;\n",
        "\n",
        "Reset gate - how to merge the new input with the previous memory\n",
        "\n",
        "Update gate - how much of the previous state will continue to be kept\n",
        "\n",
        "Let’s see how our architecture looks like:\n",
        "\n",
        "-- Hyper parameters --\n",
        "\n",
        "**Layers** : Due to the same reason like LSTM, we used two GRU layers to keep our colab notebook sustainable and eventually it performed enough well.\n",
        "\n",
        "**num_nodes**: In addition to LSTM, we wanted to keep the number of nodes same proportion but deeper. That is 120 is picked, which is twice of 60.\n",
        "\n",
        "**Batch Normalization**: We used it to make model faster and more stable through normalization of the layers' inputs by re-centering and re-scaling.\n",
        "\n",
        "**Activation function**: As we explained in the LSTM section, we think that since the data's scale is zero centred like tanh, tanh fits better into it.\n",
        "\n",
        "**Dropout**: is simply a layer that tries to prevent the system from memorizing. This time trying 10, instead of 25.\n",
        "\n",
        "**Loss Function**: Mean Squered Error: Since we are dealing with a regression problem, MSE is the most commonly used loss function.\n",
        "\n",
        "**Optimizer**: Adam: is a very recent and well-performing optimizer and proper for our task.\n",
        "\n",
        "**Batch size**: 4096 which is four times lower batch size compared to the one we set for LSTM, in order to make model learn better since chosen epochs is lower.\n",
        "\n",
        "**Epochs**: Since the model is deeper and slower, 20 is chosen to save time"
      ],
      "metadata": {
        "id": "yu6aFpC1YWYb"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "model_gru = tf.keras.Sequential([\n",
        "              tf.keras.layers.InputLayer(input_shape = (btc_train_observed.shape[1], 1)),\n",
        "              tf.keras.layers.BatchNormalization(),\n",
        "              tf.keras.layers.GRU(120, return_sequences=True, activation='tanh'),\n",
        "              tf.keras.layers.Dropout(0.10),\n",
        "              tf.keras.layers.BatchNormalization(),\n",
        "              tf.keras.layers.GRU(120, activation='tanh'),\n",
        "              tf.keras.layers.Dropout(0.10),\n",
        "              tf.keras.layers.Dense(1)\n",
        "])\n",
        "\n",
        "model_gru.summary()"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "gYsL5izRVuQL",
        "outputId": "0e8dcf8f-d324-4899-d9c6-8699863b377d"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Model: \"sequential_1\"\n",
            "_________________________________________________________________\n",
            " Layer (type)                Output Shape              Param #   \n",
            "=================================================================\n",
            " batch_normalization (BatchN  (None, 60, 1)            4         \n",
            " ormalization)                                                   \n",
            "                                                                 \n",
            " gru (GRU)                   (None, 60, 120)           44280     \n",
            "                                                                 \n",
            " dropout_2 (Dropout)         (None, 60, 120)           0         \n",
            "                                                                 \n",
            " batch_normalization_1 (Batc  (None, 60, 120)          480       \n",
            " hNormalization)                                                 \n",
            "                                                                 \n",
            " gru_1 (GRU)                 (None, 120)               87120     \n",
            "                                                                 \n",
            " dropout_3 (Dropout)         (None, 120)               0         \n",
            "                                                                 \n",
            " dense_1 (Dense)             (None, 1)                 121       \n",
            "                                                                 \n",
            "=================================================================\n",
            "Total params: 132,005\n",
            "Trainable params: 131,763\n",
            "Non-trainable params: 242\n",
            "_________________________________________________________________\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "%%time\n",
        "model_gru.compile(loss='mean_squared_error', optimizer='adam')\n",
        "history_gru = model_gru.fit(eth_train_observed, eth_train_next_price, epochs=20, batch_size=4096, validation_data=(eth_valid_observed, eth_valid_next_price))"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "-nIaWnRJVwEJ",
        "outputId": "b08e3c83-0f2c-4e55-fdf0-4fae4d791947"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Epoch 1/20\n",
            "400/400 [==============================] - 132s 321ms/step - loss: 0.3859 - val_loss: 0.7774\n",
            "Epoch 2/20\n",
            "400/400 [==============================] - 127s 317ms/step - loss: 0.3506 - val_loss: 0.2515\n",
            "Epoch 3/20\n",
            "400/400 [==============================] - 126s 316ms/step - loss: 0.3341 - val_loss: 0.2389\n",
            "Epoch 4/20\n",
            "400/400 [==============================] - 126s 314ms/step - loss: 0.3272 - val_loss: 0.2379\n",
            "Epoch 5/20\n",
            "400/400 [==============================] - 125s 313ms/step - loss: 0.3235 - val_loss: 0.2566\n",
            "Epoch 6/20\n",
            "400/400 [==============================] - 125s 313ms/step - loss: 0.3129 - val_loss: 0.2226\n",
            "Epoch 7/20\n",
            "400/400 [==============================] - 125s 313ms/step - loss: 0.3066 - val_loss: 0.2289\n",
            "Epoch 8/20\n",
            "400/400 [==============================] - 125s 313ms/step - loss: 0.3014 - val_loss: 0.2164\n",
            "Epoch 9/20\n",
            "400/400 [==============================] - 125s 313ms/step - loss: 0.2941 - val_loss: 0.2166\n",
            "Epoch 10/20\n",
            "400/400 [==============================] - 125s 313ms/step - loss: 0.2920 - val_loss: 0.2158\n",
            "Epoch 11/20\n",
            "400/400 [==============================] - 125s 313ms/step - loss: 0.2885 - val_loss: 0.2959\n",
            "Epoch 12/20\n",
            "400/400 [==============================] - 125s 313ms/step - loss: 0.2677 - val_loss: 0.1874\n",
            "Epoch 13/20\n",
            "400/400 [==============================] - 125s 313ms/step - loss: 0.2439 - val_loss: 0.1753\n",
            "Epoch 14/20\n",
            "400/400 [==============================] - 125s 313ms/step - loss: 0.2405 - val_loss: 0.1813\n",
            "Epoch 15/20\n",
            "400/400 [==============================] - 125s 313ms/step - loss: 0.2374 - val_loss: 0.1823\n",
            "Epoch 16/20\n",
            "400/400 [==============================] - 125s 313ms/step - loss: 0.2355 - val_loss: 0.1708\n",
            "Epoch 17/20\n",
            "400/400 [==============================] - 125s 313ms/step - loss: 0.2349 - val_loss: 0.1769\n",
            "Epoch 18/20\n",
            "400/400 [==============================] - 125s 313ms/step - loss: 0.2335 - val_loss: 0.1747\n",
            "Epoch 19/20\n",
            "400/400 [==============================] - 125s 313ms/step - loss: 0.2340 - val_loss: 0.1793\n",
            "Epoch 20/20\n",
            "400/400 [==============================] - 125s 313ms/step - loss: 0.2328 - val_loss: 0.1740\n",
            "CPU times: user 30min 48s, sys: 31 s, total: 31min 19s\n",
            "Wall time: 42min 25s\n"
          ]
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "### GRU model performence"
      ],
      "metadata": {
        "id": "JVRJOdSF2bTb"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "fig = plt.figure(figsize=(18,6))\n",
        "\n",
        "# summarize history for loss\n",
        "plt.plot(history_gru.history['val_loss'])\n",
        "plt.plot(history_gru.history['loss'])\n",
        "plt.title('model loss')\n",
        "plt.ylabel('loss')\n",
        "plt.xlabel('epoch')\n",
        "plt.grid(True)\n",
        "plt.legend(['val_loss','loss'], loc='upper left')\n",
        "plt.show()"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 404
        },
        "id": "kz3jpo3HhlJs",
        "outputId": "b535af57-1cea-4863-e80d-e6381a8f93cd"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 1296x432 with 1 Axes>"
            ]
          },
          "metadata": {
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "### GRU model prediction"
      ],
      "metadata": {
        "id": "YH5gYn1O2kFC"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "%%time\n",
        "\n",
        "# prediction on train\n",
        "eth_train_predictions = model_gru.predict(eth_train_observed)\n",
        "eth_train_predictions_inversed = eth_scaler.inverse_transform(eth_train_predictions)\n",
        "\n",
        "# prediction on valid\n",
        "eth_valid_predictions = model_gru.predict(eth_valid_observed)\n",
        "eth_valid_predictions_inversed = eth_scaler.inverse_transform(eth_valid_predictions)\n",
        "\n",
        "# prediction on test\n",
        "eth_test_predictions = model_gru.predict(eth_test_observed)\n",
        "eth_test_predictions_inversed = eth_scaler.inverse_transform(eth_test_predictions)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "lKolLgWFh36g",
        "outputId": "ab69b92f-b045-4d65-9abc-ae1593585730"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "CPU times: user 10min 49s, sys: 20.5 s, total: 11min 10s\n",
            "Wall time: 11min 36s\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "# We want to add predictions on the dataframe to analysis the model better\n",
        "# since our observation vector includes 60 elements inside and we start predicting from 61st index that is why we will fill the first 60 index with nan\n",
        "nan_array = np.empty(d_o)\n",
        "nan_array[:] = np.NaN\n",
        "\n",
        "# train\n",
        "eth_train['scaled_predictions'] = np.concatenate((nan_array, eth_train_predictions.flatten()))\n",
        "eth_train['inversed_predictions'] = np.concatenate((nan_array, eth_train_predictions_inversed.flatten()))\n",
        "\n",
        "print('*'*40,'\\t eth train: \\t','*'*40,'\\n')\n",
        "with pd.option_context('display.max_columns', None): \n",
        "  print(eth_train)\n",
        "  print('\\n\\n')\n",
        "\n",
        "# valid\n",
        "eth_valid['scaled_predictions'] = np.concatenate((nan_array, eth_valid_predictions.flatten()))\n",
        "eth_valid['inversed_predictions'] = np.concatenate((nan_array, eth_valid_predictions_inversed.flatten()))\n",
        "\n",
        "print('*'*40,'\\t eth valid: \\t','*'*40,'\\n')\n",
        "with pd.option_context('display.max_columns', None): \n",
        "  print(eth_valid)\n",
        "  print('\\n\\n')\n",
        "\n",
        "# test\n",
        "print('*'*40,'\\t eth test: \\t','*'*40,'\\n')\n",
        "eth_test['scaled_predictions'] = np.concatenate((nan_array, eth_test_predictions.flatten()))\n",
        "eth_test['inversed_predictions'] = np.concatenate((nan_array, eth_test_predictions_inversed.flatten()))\n",
        "\n",
        "with pd.option_context('display.max_columns', None): \n",
        "  print(eth_test)\n",
        "  print('\\n\\n')"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "0GfejeKsiT1R",
        "outputId": "cecd86a9-c61e-466b-80f2-77e0cbd227a5"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "**************************************** \t eth train: \t **************************************** \n",
            "\n",
            "         Count         VWAP                 Time    Target      Volume  \\\n",
            "1          192   738.268967  2018-01-01 00:02:00 -0.004441  232.793141   \n",
            "2          120   737.994457  2018-01-01 00:03:00 -0.004206  174.138031   \n",
            "3          156   737.303631  2018-01-01 00:04:00 -0.002205  165.383926   \n",
            "4          118   736.163026  2018-01-01 00:05:00 -0.001744  193.078039   \n",
            "5          119   734.816231  2018-01-01 00:06:00 -0.001331  366.585886   \n",
            "...        ...          ...                  ...       ...         ...   \n",
            "1636205    952  1744.819326  2021-02-10 23:56:00 -0.007403  361.585855   \n",
            "1636206   1341  1744.446378  2021-02-10 23:57:00 -0.010088  509.741571   \n",
            "1636207   1087  1744.657609  2021-02-10 23:58:00 -0.010048  573.788708   \n",
            "1636208   1128  1743.100228  2021-02-10 23:59:00 -0.008924  662.240731   \n",
            "1636209   1546  1743.167509  2021-02-11 00:00:00 -0.012347  861.938779   \n",
            "\n",
            "              Gap  Scaled_Target  scaled_predictions  inversed_predictions  \n",
            "1       -0.570324      -1.785000                 NaN                   NaN  \n",
            "2       -0.274510      -1.690477                 NaN                   NaN  \n",
            "3       -0.690826      -0.885934                 NaN                   NaN  \n",
            "4       -1.140605      -0.700536                 NaN                   NaN  \n",
            "5       -1.346795      -0.534255                 NaN                   NaN  \n",
            "...           ...            ...                 ...                   ...  \n",
            "1636205  0.349075      -2.976315           -2.772948             -0.006897  \n",
            "1636206 -0.372947      -4.056625           -3.222050             -0.008014  \n",
            "1636207  0.211231      -4.040318           -3.332782             -0.008289  \n",
            "1636208 -1.557381      -3.588295           -4.199104             -0.010443  \n",
            "1636209  0.067282      -4.964991           -4.370858             -0.010870  \n",
            "\n",
            "[1636209 rows x 9 columns]\n",
            "\n",
            "\n",
            "\n",
            "**************************************** \t eth valid: \t **************************************** \n",
            "\n",
            "         Count         VWAP                 Time    Target       Volume  \\\n",
            "1636211   3993  1739.975016  2021-02-11 00:02:00 -0.002272  2840.573518   \n",
            "1636212   5510  1733.024759  2021-02-11 00:03:00 -0.003600  6089.341354   \n",
            "1636213   2467  1736.350556  2021-02-11 00:04:00 -0.004419  1663.983443   \n",
            "1636214   1716  1736.887789  2021-02-11 00:05:00 -0.005024  1465.379388   \n",
            "1636215   1347  1738.382404  2021-02-11 00:06:00 -0.003708   802.958655   \n",
            "...        ...          ...                  ...       ...          ...   \n",
            "1794593   3043  2709.249902  2021-05-31 23:54:00  0.000417  2478.497480   \n",
            "1794594   2950  2705.669506  2021-05-31 23:55:00 -0.000122  1552.162655   \n",
            "1794595   3638  2701.145405  2021-05-31 23:56:00 -0.001684  2215.333008   \n",
            "1794596   5962  2704.079620  2021-05-31 23:57:00 -0.002880  3568.405783   \n",
            "1794597   3104  2710.840698  2021-05-31 23:58:00 -0.002703  2286.300368   \n",
            "\n",
            "              Gap  Scaled_Target  scaled_predictions  inversed_predictions  \n",
            "1636211 -6.743441      -0.912690                 NaN                   NaN  \n",
            "1636212 -6.950257      -1.447020                 NaN                   NaN  \n",
            "1636213  3.325796      -1.776468                 NaN                   NaN  \n",
            "1636214  0.537234      -2.019814                 NaN                   NaN  \n",
            "1636215  1.494615      -1.490445                 NaN                   NaN  \n",
            "...           ...            ...                 ...                   ...  \n",
            "1794593 -0.252206       0.168642            0.216487              0.000536  \n",
            "1794594 -3.580396      -0.047996            0.022663              0.000054  \n",
            "1794595 -4.524101      -0.676109           -0.670088             -0.001669  \n",
            "1794596  2.934216      -1.157112           -1.008528             -0.002510  \n",
            "1794597  6.761078      -1.086254           -0.761131             -0.001895  \n",
            "\n",
            "[158387 rows x 9 columns]\n",
            "\n",
            "\n",
            "\n",
            "**************************************** \t eth test: \t **************************************** \n",
            "\n",
            "         Count         VWAP                 Time    Target       Volume  \\\n",
            "1794599   4933  2696.263219  2021-06-01 00:01:00 -0.001576  3517.491488   \n",
            "1794600   8733  2718.126072  2021-06-01 00:02:00 -0.001328  5355.471191   \n",
            "1794601  11098  2725.585309  2021-06-01 00:03:00 -0.000896  6570.743206   \n",
            "1794602   4412  2722.948702  2021-06-01 00:04:00  0.000244  2335.272742   \n",
            "1794603   5315  2726.046925  2021-06-01 00:05:00 -0.000737  5145.622819   \n",
            "...        ...          ...                  ...       ...          ...   \n",
            "1955855   2291  2956.842571  2021-09-20 23:40:00 -0.001309   717.616863   \n",
            "1955856   2534  2964.322014  2021-09-20 23:41:00 -0.000736   776.638911   \n",
            "1955857   2321  2965.543538  2021-09-20 23:42:00 -0.000491   976.724995   \n",
            "1955858   1971  2964.359745  2021-09-20 23:43:00 -0.000562   674.087973   \n",
            "1955859   2023  2958.430582  2021-09-20 23:44:00 -0.000346   738.290907   \n",
            "\n",
            "               Gap  Scaled_Target  scaled_predictions  inversed_predictions  \n",
            "1794599 -14.342284      -0.632695                 NaN                   NaN  \n",
            "1794600  21.862853      -0.533088                 NaN                   NaN  \n",
            "1794601   7.459237      -0.359338                 NaN                   NaN  \n",
            "1794602  -2.636607       0.099192                 NaN                   NaN  \n",
            "1794603   3.098223      -0.295371                 NaN                   NaN  \n",
            "...            ...            ...                 ...                   ...  \n",
            "1955855  -5.562257      -0.525206           -0.540203             -0.001346  \n",
            "1955856   7.479443      -0.294812           -0.519771             -0.001295  \n",
            "1955857   1.221523      -0.196457           -0.182346             -0.000456  \n",
            "1955858  -1.183792      -0.224888           -0.216537             -0.000541  \n",
            "1955859  -5.929163      -0.138031           -0.025434             -0.000066  \n",
            "\n",
            "[161261 rows x 9 columns]\n",
            "\n",
            "\n",
            "\n"
          ]
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "# Evaluation"
      ],
      "metadata": {
        "id": "poM79lDKrn5A"
      }
    },
    {
      "cell_type": "markdown",
      "source": [
        "Pearson product-moment correlation coefficients were chosen for evaluation metrics as default by the organizers.\n",
        "\n",
        "The Pearson product-moment correlation coefficient is a measure of the linear relationship between two questions/measures/variables, X and Y. The correlation value can range from +1 to -1.\n",
        "\n",
        " \n",
        "\n",
        "A positive correlation (e.g., +0.32) means there is a positive relationship between X and Y. For example, a positive correlation between height and weight means that as height increases, so does weight.\n",
        "\n",
        "$$\n",
        "r = \\frac{\\sum(x_i - \\overline{x})(y_i - \\overline{y})}{\\sqrt{\\sum (x_i - \\overline{x})^2\\sum (y_i - \\overline{y})^2}}\n",
        "$$\n",
        "\n",
        "\n",
        "$r$\t=\tcorrelation coefficient\n",
        "\n",
        "$x_i$\t=\tvalues of the x-variable in a sample\n",
        "\n",
        "$\\overline{x}$\t=\tmean of the values of the x-variable\n",
        "\n",
        "$y_{i}$\t=\tvalues of the y-variable in a sample\n",
        "\n",
        "\\$\\overline{y}$\t=\tmean of the values of the y-variable\n",
        "\n",
        "\n",
        "\n"
      ],
      "metadata": {
        "id": "aMUt_8XNrtd1"
      }
    },
    {
      "cell_type": "markdown",
      "source": [
        "## Evaluation on model 1: LSTM on btc"
      ],
      "metadata": {
        "id": "daNsBvEp3So6"
      }
    },
    {
      "cell_type": "markdown",
      "source": [
        "### Pearson correlation coefficient"
      ],
      "metadata": {
        "id": "1u65H7JE4JZk"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "%%time\n",
        "\n",
        "# evaluation on train\n",
        "btc_train_evaluation = np.corrcoef(btc_train['Target'].values[d_o:].flatten(), \n",
        "                                   btc_train['inversed_predictions'].values[d_o:].flatten())                      \n",
        "\n",
        "btc_train_evaluation_scaled = np.corrcoef(btc_train['Scaled_Target'].values[d_o:].flatten(), \n",
        "                                          btc_train['scaled_predictions'].values[d_o:].flatten())       \n",
        "\n",
        "\n",
        "\n",
        "# evaluation on valid\n",
        "btc_valid_evaluation = np.corrcoef(btc_valid['Target'].values[d_o:].flatten(), \n",
        "                                   btc_valid['inversed_predictions'].values[d_o:].flatten())                      \n",
        "\n",
        "btc_valid_evaluation_scaled = np.corrcoef(btc_valid['Scaled_Target'].values[d_o:].flatten(), \n",
        "                                          btc_valid['scaled_predictions'].values[d_o:].flatten())    \n",
        "     \n",
        "\n",
        "\n",
        "# evaluation on test\n",
        "btc_test_evaluation = np.corrcoef(btc_test['Target'].values[d_o:].flatten(), \n",
        "                                   btc_test['inversed_predictions'].values[d_o:].flatten())                      \n",
        "\n",
        "btc_test_evaluation_scaled = np.corrcoef(btc_test['Scaled_Target'].values[d_o:].flatten(), \n",
        "                                          btc_test['scaled_predictions'].values[d_o:].flatten())    "
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "5E_BJWXF-i0E",
        "outputId": "82f50465-f49f-474e-d4f6-e6fdcc2640c1"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "CPU times: user 70.3 ms, sys: 8.21 ms, total: 78.5 ms\n",
            "Wall time: 77.9 ms\n"
          ]
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "### Plot"
      ],
      "metadata": {
        "id": "EMDsjQYy33pI"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "%%time\n",
        "\n",
        "# Visualize the data\n",
        "\n",
        "# Finally predicted values are visualized\n",
        "\n",
        "plt.figure(figsize=(22,6))\n",
        "plt.title('LSTM model on btc', fontsize=12)\n",
        "plt.xlabel('Date', fontsize=12)\n",
        "plt.ylabel('Close Price USD ($)', fontsize=12)\n",
        "plt.tick_params(labelsize=9, width=0.2, length=1)\n",
        "\n",
        "\n",
        "plt.plot(btc_train['Target'], linewidth=2.0)\n",
        "plt.plot(btc_valid['Target'], linewidth=2.0)\n",
        "plt.plot(btc_test['Target'], linewidth=2.0)\n",
        "\n",
        "plt.plot(btc_train['inversed_predictions'], linewidth=2.0)\n",
        "plt.plot(btc_valid['inversed_predictions'], linewidth=2.0)\n",
        "plt.plot(btc_test['inversed_predictions'], linewidth=2.0)\n",
        "\n",
        "\n",
        "plt.legend(['train_target', 'valid_target', 'test_target', 'train_target_prediction', 'valid_target_prediction', 'test_target_prediction'], loc='lower right', fontsize=9)\n",
        "plt.show()\n",
        "\n",
        "print('\\n\\n')\n"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 491
        },
        "id": "JnHLPnvulWGp",
        "outputId": "61d72618-db68-4294-9260-1e852ff7dfda"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 1584x432 with 1 Axes>"
            ]
          },
          "metadata": {
            "needs_background": "light"
          }
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "\n",
            "\n",
            "\n",
            "CPU times: user 903 ms, sys: 29.2 ms, total: 932 ms\n",
            "Wall time: 927 ms\n"
          ]
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "### Scatter plot"
      ],
      "metadata": {
        "id": "CEK_6ugX36_Q"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "%%time\n",
        "fig = plt.figure(figsize=(14,16))\n",
        "\n",
        "# train\n",
        "plt.subplot(3,2,1)\n",
        "plt.scatter(btc_train['Target'].values[d_o:].flatten(), \n",
        "                                   btc_train['inversed_predictions'].values[d_o:].flatten())\n",
        "plt.title(f'btc_train_evaluation: {btc_train_evaluation[0,1]:.2f}')\n",
        "\n",
        "plt.subplot(3,2,2)\n",
        "plt.scatter(btc_train['Scaled_Target'].values[d_o:].flatten(), \n",
        "                                          btc_train['scaled_predictions'].values[d_o:].flatten())\n",
        "plt.title(f'btc_train_evaluation_scaled: {btc_train_evaluation_scaled[0,1]:.2f}')\n",
        "\n",
        "\n",
        "# valid\n",
        "plt.subplot(3,2,3)\n",
        "plt.scatter(btc_valid['Target'].values[d_o:].flatten(), \n",
        "                                   btc_valid['inversed_predictions'].values[d_o:].flatten())\n",
        "plt.title(f'btc_valid_evaluation: {btc_valid_evaluation[0,1]:.2f}')\n",
        "\n",
        "\n",
        "plt.subplot(3,2,4)\n",
        "plt.scatter(btc_valid['Scaled_Target'].values[d_o:].flatten(), \n",
        "                                          btc_valid['scaled_predictions'].values[d_o:].flatten())\n",
        "plt.title(f'btc_valid_evaluation_scaled: {btc_valid_evaluation_scaled[0,1]:.2f}')\n",
        "\n",
        "\n",
        "# test\n",
        "plt.subplot(3,2,5)\n",
        "plt.scatter(btc_test['Target'].values[d_o:].flatten(), \n",
        "                                   btc_test['inversed_predictions'].values[d_o:].flatten())\n",
        "plt.title(f'btc_test_evaluation: {btc_test_evaluation[0,1]:.2f}')\n",
        "\n",
        "\n",
        "plt.subplot(3,2,6)\n",
        "plt.scatter(btc_test['Scaled_Target'].values[d_o:].flatten(), \n",
        "                                          btc_test['scaled_predictions'].values[d_o:].flatten())\n",
        "plt.title(f'btc_test_evaluation_scaled: {btc_test_evaluation_scaled[0,1]:.2f}')"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 969
        },
        "id": "iKcDFv2pJgfH",
        "outputId": "82bbf3f6-41d4-4abe-a61d-94c41e529944"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "CPU times: user 509 ms, sys: 5.67 ms, total: 514 ms\n",
            "Wall time: 508 ms\n"
          ]
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 1008x1152 with 6 Axes>"
            ]
          },
          "metadata": {
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "## Evaluation model 2: GRU on eth\n",
        "\n"
      ],
      "metadata": {
        "id": "T0XiWtqx3d58"
      }
    },
    {
      "cell_type": "markdown",
      "source": [
        "### Pearson correlation coefficient"
      ],
      "metadata": {
        "id": "iDTMVZUt4ck-"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "%%time\n",
        "\n",
        "# evaluation on train\n",
        "eth_train_evaluation = np.corrcoef(eth_train['Target'].values[d_o:].flatten(), \n",
        "                                   eth_train['inversed_predictions'].values[d_o:].flatten())                      \n",
        "\n",
        "eth_train_evaluation_scaled = np.corrcoef(eth_train['Scaled_Target'].values[d_o:].flatten(), \n",
        "                                          eth_train['scaled_predictions'].values[d_o:].flatten())       \n",
        "\n",
        "\n",
        "\n",
        "# evaluation on valid\n",
        "eth_valid_evaluation = np.corrcoef(eth_valid['Target'].values[d_o:].flatten(), \n",
        "                                   eth_valid['inversed_predictions'].values[d_o:].flatten())                      \n",
        "\n",
        "eth_valid_evaluation_scaled = np.corrcoef(eth_valid['Scaled_Target'].values[d_o:].flatten(), \n",
        "                                          eth_valid['scaled_predictions'].values[d_o:].flatten())    \n",
        "     \n",
        "\n",
        "\n",
        "# evaluation on test\n",
        "eth_test_evaluation = np.corrcoef(eth_test['Target'].values[d_o:].flatten(), \n",
        "                                  eth_test['inversed_predictions'].values[d_o:].flatten())                      \n",
        "\n",
        "eth_test_evaluation_scaled = np.corrcoef(eth_test['Scaled_Target'].values[d_o:].flatten(), \n",
        "                                         eth_test['scaled_predictions'].values[d_o:].flatten())  "
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "EZTX8eD7_0Qi",
        "outputId": "68ae233f-dc9b-42ce-a527-fa81b2731df2"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "CPU times: user 69.1 ms, sys: 5.07 ms, total: 74.2 ms\n",
            "Wall time: 74.5 ms\n"
          ]
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "### Plot"
      ],
      "metadata": {
        "id": "4k8sJxCO6C86"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "%%time\n",
        "\n",
        "# Visualize the data\n",
        "\n",
        "# Finally predicted values are visualized\n",
        "\n",
        "plt.figure(figsize=(22,6))\n",
        "plt.title('GRU model on eth', fontsize=12)\n",
        "plt.xlabel('Date', fontsize=12)\n",
        "plt.ylabel('Close Price USD ($)', fontsize=12)\n",
        "plt.tick_params(labelsize=9, width=0.2, length=1)\n",
        "\n",
        "plt.plot(eth_train['Target'], linewidth=2.0)\n",
        "plt.plot(eth_valid['Target'], linewidth=2.0)\n",
        "plt.plot(eth_test['Target'], linewidth=2.0)\n",
        "\n",
        "plt.plot(eth_train['inversed_predictions'], linewidth=2.0)\n",
        "plt.plot(eth_valid['inversed_predictions'], linewidth=2.0)\n",
        "plt.plot(eth_test['inversed_predictions'], linewidth=2.0)\n",
        "\n",
        "\n",
        "plt.legend(['train_target', 'valid_target', 'test_target', 'train_target_prediction', 'valid_target_prediction', 'test_target_prediction'], loc='lower right', fontsize=9)\n",
        "plt.show()\n",
        "\n",
        "print('\\n\\n')"
      ],
      "metadata": {
        "id": "Kcc0iBCGAXyH",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 491
        },
        "outputId": "7b4380db-37ea-44a6-95dc-df5ff48276e1"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 1584x432 with 1 Axes>"
            ]
          },
          "metadata": {
            "needs_background": "light"
          }
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "\n",
            "\n",
            "\n",
            "CPU times: user 880 ms, sys: 17 ms, total: 897 ms\n",
            "Wall time: 893 ms\n"
          ]
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "### Scatter Plot"
      ],
      "metadata": {
        "id": "FHSiOSFZ6F_u"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "%%time\n",
        "fig = plt.figure(figsize=(14,16))\n",
        "\n",
        "# train\n",
        "plt.subplot(3,2,1)\n",
        "plt.scatter(eth_train['Target'].values[d_o:].flatten(), \n",
        "            eth_train['inversed_predictions'].values[d_o:].flatten())\n",
        "\n",
        "plt.title(f'eth_train_evaluation: {eth_train_evaluation[0,1]:.2f}')\n",
        "\n",
        "plt.subplot(3,2,2)\n",
        "plt.scatter(eth_train['Scaled_Target'].values[d_o:].flatten(), \n",
        "            eth_train['scaled_predictions'].values[d_o:].flatten())\n",
        "\n",
        "plt.title(f'eth_train_evaluation_scaled: {eth_train_evaluation_scaled[0,1]:.2f}')\n",
        "\n",
        "\n",
        "# valid\n",
        "plt.subplot(3,2,3)\n",
        "plt.scatter(eth_valid['Target'].values[d_o:].flatten(), \n",
        "            eth_valid['inversed_predictions'].values[d_o:].flatten())\n",
        "\n",
        "plt.title(f'eth_valid_evaluation: {eth_valid_evaluation[0,1]:.2f}')\n",
        "\n",
        "\n",
        "plt.subplot(3,2,4)\n",
        "plt.scatter(eth_valid['Scaled_Target'].values[d_o:].flatten(), \n",
        "            eth_valid['scaled_predictions'].values[d_o:].flatten())\n",
        "\n",
        "plt.title(f'eth_valid_evaluation_scaled: {eth_valid_evaluation_scaled[0,1]:.2f}')\n",
        "\n",
        "\n",
        "# test\n",
        "plt.subplot(3,2,5)\n",
        "plt.scatter(eth_test['Target'].values[d_o:].flatten(), \n",
        "            eth_test['inversed_predictions'].values[d_o:].flatten())\n",
        "\n",
        "plt.title(f'eth_test_evaluation: {eth_test_evaluation[0,1]:.2f}')\n",
        "\n",
        "\n",
        "plt.subplot(3,2,6)\n",
        "plt.scatter(eth_test['Scaled_Target'].values[d_o:].flatten(), \n",
        "            eth_test['scaled_predictions'].values[d_o:].flatten())\n",
        "\n",
        "plt.title(f'eth_test_evaluation_scaled: {eth_test_evaluation_scaled[0,1]:.2f}')"
      ],
      "metadata": {
        "id": "DT_HOBzCAzfR",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 969
        },
        "outputId": "a9a4ef95-45e4-44fb-a13f-cda631c821e0"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "CPU times: user 525 ms, sys: 4.74 ms, total: 529 ms\n",
            "Wall time: 527 ms\n"
          ]
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 1008x1152 with 6 Axes>"
            ]
          },
          "metadata": {
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "# Comments"
      ],
      "metadata": {
        "id": "3omRlqk30XHB"
      }
    },
    {
      "cell_type": "markdown",
      "source": [
        "## Regarding to model 1: LSTM"
      ],
      "metadata": {
        "id": "zJTsGsjB7wv1"
      }
    },
    {
      "cell_type": "markdown",
      "source": [
        "Even though we did not train any deep and complex LSTM model, still, we got quite good performance on test set. We think the main reason that led us to good performance is not only related to the model, also how we split and scaled the data into train, valid and test. Additionally, there might be some information leakage that helps the model learn a good way but we couldn't find any leakage tough we over checked all steps many times. So far, with a simple LSTM model and without scaling the data, we scored the lowest score 0.79 on the test set. The best score we got, with scaled data and a more complex model, is 0.93. We think that if we were able to build more complex data including regularization, batch normalization and more, we might have gotten even higher scores.\n"
      ],
      "metadata": {
        "id": "8cw2C05NPwc3"
      }
    },
    {
      "cell_type": "markdown",
      "source": [
        "## Regarding to model 2: GRU"
      ],
      "metadata": {
        "id": "w7H_i-537zQY"
      }
    },
    {
      "cell_type": "markdown",
      "source": [
        "Contrary to our expectation, GRU performed slightly better performance on Pearson product-moment correlation coefficients compared to the LSTM model. We think that having a deeper model compared to LSTM, helped the GRU model comprehend the data much better."
      ],
      "metadata": {
        "id": "wmydsp6Iuywm"
      }
    },
    {
      "cell_type": "markdown",
      "source": [
        "## Regarding to hyper-parameters"
      ],
      "metadata": {
        "id": "ZXnKoxOw37X1"
      }
    },
    {
      "cell_type": "markdown",
      "source": [
        "The type of activation function used affects accuracy. Although sigmoid and tanh activation functions are quite successful, these two might cause vanishing gradient problem in practice. Relu is more successful in this regard. However, we got the highest accuracy from tanh activation function. Because tanh could work in high harmony with the data, since our data is zero-centered just like tanh function.\n",
        "\n",
        "We worked with several different batch sizes. Considering that our data is very large and we were working with at least 10 epochs, we needed to find a rational value in the performance/running time balance. Since a smaller number of batch size leds more iteration in each epoch, it gives higher accuracy than the larger ones. So, we got better performance with 4096 batch size than the ones we tried up to 32768.\n",
        "\n",
        "When we increase the number of nodes, it is possible to say that we generally achieve higher accuracy. Since we encountered various values in different notebooks, we intuitively used values of 30, 60, 120 and observed that larger node number increases accuracy."
      ],
      "metadata": {
        "id": "f5rIZ3LL4MPI"
      }
    },
    {
      "cell_type": "markdown",
      "source": [
        "## Regarding to technical analysis"
      ],
      "metadata": {
        "id": "h56N8EMp712M"
      }
    },
    {
      "cell_type": "markdown",
      "source": [
        "To analyse the data better and have a solid model to predict its future, we tried implementing several technical analysis methods such as ROC Indicator, RSI and so on. In fact, we did not clearly understand what does those indicator methods do since our background is not related to Economics, which is why we did not achieve implementing them as wanted. Also, if we had trained fully connected dense layers, we could have used several inputs including the ones analysed with indicators, but instead, we decided to implement some RNN models in order to learn them closer by using them."
      ],
      "metadata": {
        "id": "PGHjPRjZOjj7"
      }
    },
    {
      "cell_type": "markdown",
      "source": [
        "## Final comments"
      ],
      "metadata": {
        "id": "fHc8knYU781K"
      }
    },
    {
      "cell_type": "markdown",
      "source": [
        "\n",
        "This notebook describes our own approaches to the G-Research Crypto Competition proposed by Kaggle, whichserved as a sub-assignment for the Deep Learning and Text Analysis in Finance course at the University of Passau. \n",
        "\n",
        "For this purpose, we came up with two modelling approaches like LSTM on btc and GRU on eth, yet, both obtained similar and interesting results.\n",
        "\n",
        "However, we are afraid of information leakage in data since we always got a high score from evaluation for all the sets even though we double-checked all the processes. \n",
        "\n",
        "Also, we particularly found out that using deeper and more complex models might give interesting results, hence, there is certainly space for research.\n",
        "\n",
        "Nevertheless, it was an enlightening experience and a nice discovery of the RNN world where lots of exciting new discoveries happening in recent years."
      ],
      "metadata": {
        "id": "1PJaCkaGWAM5"
      }
    },
    {
      "cell_type": "code",
      "source": [
        ""
      ],
      "metadata": {
        "id": "Fvuhfyt7yLQN"
      },
      "execution_count": null,
      "outputs": []
    }
  ]
}