Model_Evaluation-Sentiment_Analysis_Scotref2.ipynb

model_evaluation-sentiment_analysis_scotref2.ipynb

{
  "nbformat": 4,
  "nbformat_minor": 0,
  "metadata": {
    "colab": {
      "name": "Model_Evaluation-Sentiment_Analysis_Scotref2.ipynb",
      "provenance": [],
      "collapsed_sections": [],
      "include_colab_link": true
    },
    "kernelspec": {
      "name": "python3",
      "display_name": "Python 3"
    },
    "accelerator": "GPU"
  },
  "cells": [
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "view-in-github",
        "colab_type": "text"
      },
      "source": [
        "<a href=\"https://colab.research.google.com/gist/gracecarrillo/5bce1e518154c9ce7ae836892eddb64e/model_evaluation-sentiment_analysis_scotref2.ipynb\" target=\"_parent\"><img src=\"https://colab.research.google.com/assets/colab-badge.svg\" alt=\"Open In Colab\"/></a>"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "ma5cRQJafwRY",
        "colab_type": "text"
      },
      "source": [
        "# Scottish independence: Twitter data Sentiment Analysis\n",
        "\n"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "mFiPGpFfkIUU",
        "colab_type": "text"
      },
      "source": [
        "### 6. Model Evaluation\n",
        "  - Confusion Matrix\n",
        "  - Accuracy, Recall, Precision and F1-Scores\n",
        "  - Choosing a model"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "oP1c5gDe2M6-",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "# Must be upgraded\n",
        "!pip install tqdm==4.36.1 --upgrade"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "pUU5MOCAP1j0",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "!pip install tweet_parser"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "v0Xc0nZtWke9",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "!pip install --upgrade gensim"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "VOnd_sShyvQx",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "!pip install vaderSentiment"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "rk4fAdKjRIdC",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "from google.colab import drive\n",
        "drive.mount('/content/drive')"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "cpXSoADwxTTs",
        "colab_type": "code",
        "outputId": "cea166a0-19cb-4e99-fb4b-f09e8baa41fd",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 35
        }
      },
      "source": [
        "# general\n",
        "import os\n",
        "import pandas as pd\n",
        "import numpy as np\n",
        "import csv\n",
        "import string\n",
        "import matplotlib.pyplot as plt\n",
        "import seaborn as sns\n",
        "import random\n",
        "import itertools\n",
        "import collections\n",
        "from collections import Counter\n",
        "\n",
        "# tweets\n",
        "import tweepy as tw\n",
        "import re\n",
        "from collections import Counter\n",
        "from string import punctuation\n",
        "from tweepy import OAuthHandler\n",
        "import json\n",
        "\n",
        "# text manipulation \n",
        "import nltk \n",
        "from nltk.corpus import stopwords\n",
        "from nltk.tokenize import TweetTokenizer\n",
        "from nltk.stem.wordnet import WordNetLemmatizer\n",
        "from nltk.stem.porter import PorterStemmer\n",
        "from nltk.stem.porter import * \n",
        "\n",
        "# plots\n",
        "from wordcloud import WordCloud\n",
        "import plotly\n",
        "import chart_studio.plotly as py\n",
        "import plotly.graph_objs as go \n",
        "from plotly.offline import download_plotlyjs, init_notebook_mode, plot, iplot\n",
        "import cufflinks as cf\n",
        "cf.go_offline()\n",
        "\n",
        "# Feature Engineering\n",
        "from vaderSentiment.vaderSentiment import SentimentIntensityAnalyzer\n",
        "\n",
        "# Machine Learning\n",
        "from sklearn.pipeline import Pipeline\n",
        "from sklearn.pipeline import FeatureUnion\n",
        "from sklearn.model_selection import train_test_split\n",
        "from sklearn.model_selection import KFold, cross_val_score\n",
        "from sklearn.ensemble import RandomForestClassifier\n",
        "from sklearn.naive_bayes import MultinomialNB\n",
        "from sklearn.svm import LinearSVC\n",
        "from sklearn.model_selection import GridSearchCV\n",
        "from sklearn.externals import joblib\n",
        "from sklearn.base import BaseEstimator, TransformerMixin\n",
        "from sklearn.feature_extraction.text import TfidfVectorizer, HashingVectorizer, CountVectorizer\n",
        "from sklearn.preprocessing import MinMaxScaler\n",
        "from sklearn.preprocessing import LabelEncoder\n",
        "\n",
        "# Model Evaluation\n",
        "from numpy import loadtxt\n",
        "from keras.models import load_model\n",
        "from sklearn.metrics import confusion_matrix, classification_report\n",
        "import itertools\n",
        "\n",
        "# For geoplots\n",
        "from IPython.display import IFrame\n",
        "import folium\n",
        "from folium import plugins\n",
        "from folium.plugins import MarkerCluster, FastMarkerCluster, HeatMapWithTime\n",
        "import networkx\n",
        "\n",
        "# set plot preferences\n",
        "plt.style.use(style='ggplot')\n",
        "plt.rcParams['figure.figsize'] = (10, 6)\n",
        "pd.set_option(\"display.max_colwidth\", 200) \n",
        "\n",
        "# hide warnings\n",
        "import warnings\n",
        "warnings.filterwarnings(\"ignore\")\n",
        "\n",
        "print('Libraries imported')\n",
        "%matplotlib inline"
      ],
      "execution_count": 0,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "text/html": [
              "        <script type=\"text/javascript\">\n",
              "        window.PlotlyConfig = {MathJaxConfig: 'local'};\n",
              "        if (window.MathJax) {MathJax.Hub.Config({SVG: {font: \"STIX-Web\"}});}\n",
              "        if (typeof require !== 'undefined') {\n",
              "        require.undef(\"plotly\");\n",
              "        requirejs.config({\n",
              "            paths: {\n",
              "                'plotly': ['https://cdn.plot.ly/plotly-latest.min']\n",
              "            }\n",
              "        });\n",
              "        require(['plotly'], function(Plotly) {\n",
              "            window._Plotly = Plotly;\n",
              "        });\n",
              "        }\n",
              "        </script>\n",
              "        "
            ]
          },
          "metadata": {
            "tags": []
          }
        },
        {
          "output_type": "stream",
          "text": [
            "Libraries imported\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "rIrvUtEEUxyE",
        "colab_type": "text"
      },
      "source": [
        "## 6. Evaluation\n",
        "\n",
        "To get intuition about models capacity, we will look deeper into the following metrics:\n",
        "\n",
        "- Confusion Matrix\n",
        "- Accuracy, Recall, Precision and F1-Scores\n"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "S6WvzhLFppu7",
        "colab_type": "text"
      },
      "source": [
        "First, we must import all our model's predictions and test data:"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "OOyRjymKj3ZJ",
        "colab_type": "code",
        "outputId": "037cfbc9-5cde-4274-8e98-87fb1f3a79bc",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 35
        }
      },
      "source": [
        "# Set up labels\n",
        "labels = LabelEncoder().fit(['negative', 'positive'])\n",
        "\n",
        "# Import supervised learning model predictions\n",
        "y_pred_nb = np.load('/content/drive/My Drive/Twitter_Project/y_predsNB.npy')\n",
        "y_pred_svm = np.load('/content/drive/My Drive/Twitter_Project/y_predsSVM.npy')\n",
        "\n",
        "# Import supervised learning models test data\n",
        "y_test = np.load('/content/drive/My Drive/Twitter_Project/y_test.npy')\n",
        "\n",
        "# Import deep learning predictions\n",
        "y_preds_LSTM = np.load('/content/drive/My Drive/Twitter_Project/y_predsLSTM.npy')\n",
        "y_preds_LSTMreg = np.load('/content/drive/My Drive/Twitter_Project/y_predsLSTMreg.npy')\n",
        "\n",
        "# To output vector of labels for confusion matrix\n",
        "y_preds_LSTM = np.argmax(y_preds_LSTM, axis=1)\n",
        "y_preds_LSTMreg = np.argmax(y_preds_LSTMreg, axis=1)\n",
        "\n",
        "# Import deep learning test data\n",
        "Y_test = np.load('/content/drive/My Drive/Twitter_Project/y_testLSTM.npy')\n",
        "Y_test = np.argmax(Y_test, axis = 1)\n",
        "\n",
        "print('Predictions and test data imported')"
      ],
      "execution_count": 0,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Predictions and test data imported\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "cDTPVV4Pp291",
        "colab_type": "text"
      },
      "source": [
        "\n",
        "### 6.1 Confusion Matrices\n",
        "\n",
        "With the help of the following functions (source code: [notebook](https://github.com/vtoliveira/advanced-data-science-capstone/blob/master/Model%20Evaluation%20(Deep%20Learning%20and%20Classical%20Algorithms)%20-%20Twitter%20US%20Airline%20Sentiment.ipynb)) we'll plot an analyse the Confusion Matrices."
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "BdL50Q0XNQw2",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "# Confusion Matrix helper function\n",
        "plt.style.use(style='ggplot')\n",
        "plt.rcParams.update({'font.size': 12})\n",
        "\n",
        "def plot_confusion_matrix(cm, classes,\n",
        "                          normalize=False,\n",
        "                          title='Confusion matrix',\n",
        "                          cmap=plt.cm.Reds):\n",
        "    \"\"\"\n",
        "    This function prints and plots the confusion matrix.\n",
        "    Normalization can be applied by setting `normalize=True`.\n",
        "    \"\"\"\n",
        "    if normalize:\n",
        "        cm = cm.astype('float') / cm.sum(axis=1)[:, np.newaxis]\n",
        "        #print(\"Normalized confusion matrix\")\n",
        "    else:\n",
        "        print('Confusion matrix, without normalization')\n",
        "\n",
        "\n",
        "    plt.imshow(cm, interpolation='nearest', cmap=cmap)\n",
        "    plt.title(title)\n",
        "    plt.colorbar()\n",
        "    tick_marks = np.arange(len(classes))\n",
        "    plt.xticks(tick_marks, classes, rotation=45)\n",
        "    plt.yticks(tick_marks, classes)\n",
        "\n",
        "    fmt = '.2f' if normalize else 'd'\n",
        "    thresh = cm.max() / 2.\n",
        "    for i, j in itertools.product(range(cm.shape[0]), range(cm.shape[1])):\n",
        "        plt.text(j, i, format(cm[i, j], fmt),\n",
        "                 horizontalalignment=\"center\",\n",
        "                 color=\"white\" if cm[i, j] > thresh else \"black\")\n",
        "\n",
        "    plt.ylabel('True label')\n",
        "    plt.xlabel('Predicted label')\n",
        "    plt.tight_layout()\n"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "uAS4IyfqqeAC",
        "colab_type": "code",
        "outputId": "06df34ef-045d-46c7-8cf0-229686319f3c",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 817
        }
      },
      "source": [
        "#--- Naive Bayes Classifier ---#\n",
        "\n",
        "# Compute confusion matrix\n",
        "cnf_matrix = confusion_matrix(y_test, y_pred_nb)\n",
        "np.set_printoptions(precision=2)\n",
        "\n",
        "# Plot normalized confusion matrix\n",
        "fig = plt.figure(figsize=(20,15))\n",
        "fig.add_subplot(3,3,1)\n",
        "plot_confusion_matrix(cnf_matrix, classes=['negative', 'positive'], normalize=True,\n",
        "                      title='Normalized Confusion Matrix for Naive Bayes Classifier')\n",
        "\n",
        "#--- Support Vector Machine Classifier ---#\n",
        "\n",
        "# Compute confusion matrix\n",
        "cnf_matrix = confusion_matrix(y_test, y_pred_svm)\n",
        "np.set_printoptions(precision=2)\n",
        "\n",
        "# Plot normalized confusion matrix\n",
        "fig.add_subplot(3,3,2)\n",
        "plot_confusion_matrix(cnf_matrix, classes=['negative', 'positive'], normalize=True,\n",
        "                      title='Normalized confusion matrix for SVM Classifier')\n",
        "\n",
        "\n",
        "#--- LSTM Recurrent Neural Network ---#\n",
        "\n",
        "# Compute confusion matrix\n",
        "cnf_matrix = confusion_matrix(Y_test, y_preds_LSTM)\n",
        "np.set_printoptions(precision=2)\n",
        "\n",
        "# Plot normalized confusion matrix\n",
        "fig = plt.figure(figsize=(20,15))\n",
        "fig.add_subplot(3,3,1)\n",
        "plot_confusion_matrix(cnf_matrix, classes=['negative', 'positive'], normalize=True,\n",
        "                      title='LSTM Neural Network')\n",
        "\n",
        "#--- LSTM Recurrent Neural Network with regularisation ---#\n",
        "\n",
        "# Compute confusion matrix\n",
        "\n",
        "cnf_matrix = confusion_matrix(Y_test, y_preds_LSTMreg)\n",
        "np.set_printoptions(precision=2)\n",
        "\n",
        "# Plot normalized confusion matrix\n",
        "fig.add_subplot(3,3,2)\n",
        "plot_confusion_matrix(cnf_matrix, classes=['negative', 'positive'], normalize=True,\n",
        "                      title='LSTM Neural Network with regularisation')\n",
        "\n",
        "plt.savefig('/content/drive/My Drive/Twitter_Project/ConfusionMatrices.png', bbox_inches='tight')\n",
        "plt.show()"
      ],
      "execution_count": 0,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "image/png": "iVBORw0KGgoAAAANSUhEUgAAA60AAAGQCAYAAABF3oaFAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz\nAAALEgAACxIB0t1+/AAAADh0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uMy4xLjIsIGh0\ndHA6Ly9tYXRwbG90bGliLm9yZy8li6FKAAAgAElEQVR4nOzdd1gUx/8H8DdIkyZYEAUUoxwqRYoi\niBJEbBQLqKBi70LUaFRMNLFj71EssURRRERQiiUiCFI0iV2JYrBAFBBBmkjb3x/8br8cd8AdHnKe\nn1cenifOzu7N7u3NZ2Z3dlaGYRgGhBBCCCGEEEKIBJJt6gIQQgghhBBCCCG1oU4rIYQQQgghhBCJ\nRZ1WQgghhBBCCCESizqthBBCCCGEEEIkFnVaCSGEEEIIIYRILOq0EkIIIYQQQgiRWNRprcbBwQG+\nvr7sv9PT02FoaIiQkJDPXpaQkBAYGhoiPT39s392Q927dw9jx46Fubk5DA0NkZycLPbP2L17NwwN\nDcW+XWnQWOdrXFwcRowYAVNT0y/unBTE19cXDg4OTV0MUo0kfCfJyckC6y1B9RrVQ18PaheIrry8\nHJs3b4a9vT26du2KCRMmNMrnGBoaYvfu3Y2ybWnRGHXV5/p+vxaSEE9qq9eeP3+OadOmoVevXuzy\npqyH6u20cgO5oaEhYmNj+ZZzC5+QkNAoBSTCqaioQHBwMCZOnIjevXvDyMgIffv2hbe3N/74449G\n//yysjIsWLAAmZmZWLJkCTZt2oTOnTs3+ud+btzKxdDQEPfu3ROYZ/bs2WyehsjLy8Pu3bsbpdMv\nqtzcXMyfPx8Mw2DFihXYtGkTWrZs2aifya08DQ0NcfLkSb7l3DrpzJkzjVqOxsCtL6v/9erVCyNH\njsTx48dRXl7e1EUUqxcvXuCXX37BwIEDYWpqCnNzc4wYMQLbt29HdnZ2UxevXl9LvSYqaheQupw9\nexaHDh2CnZ0dNmzYgNmzZzd1kb54iYmJ2L17N/Lz85u6KE32/SYmJmLatGmws7ODiYkJ+vXrhwkT\nJrAXLsrLy2FrawsnJ6c6t7Nx40YYGhrizp07AIAJEyawsbi0tJQvf2lpKXr37g1DQ0OROuhZWVnY\nuHEjnJycYGZmBlNTUzg5OWHdunV48eKFCHvedHx9ffHgwQN4e3tj06ZN6NWrV5OWR06UzDt37sS3\n337bWGWRODo6Orh37x7k5EQ6TJ9dYWEh5syZg5s3b8LKygrTp0+HpqYmsrOzERMTA29vb2zZsgWu\nrq6NVoaXL18iIyMDS5YswdixYxvtc+bMmYOZM2c22vaFpaioiLCwMJiamvKkv3v3DnFxcVBUVMTH\njx8btO33799jz5498PHxQe/evYVerzHO1/v376OoqAhz5szBkCFDxLZdYfn7+8Pd3R2Kiopi2+aa\nNWvAMIzYttcQM2fORJcuXQAA+fn5iI6Oxtq1a5GWloaff/65ScsmLlFRUViyZAmUlZXh6uoKAwMD\nVFRU4PHjxwgICMDFixdx6dKlpi4mq1evXrh37x7k5eXZtNrqNUmphyQBtQtITUlJSVBRUcGqVasg\nIyPTaJ9z7949NGvWrNG2L0mSkpLg7++PkSNHQl1dXej1GqOu+lzfb3XHjx/H2rVr0b17d3h5eUFT\nUxOvX7/GgwcPsH//fnz33XeQk5ODs7Mzjh07hgcPHsDY2JhvOxUVFQgPD4e+vj7MzMzYdAUFBeTn\n5+PatWsYPHgwzzqxsbHIy8uDgoKC0OW9desW5s6di7KyMri4uMDLywvNmjXD06dPERkZiVOnTuHB\ngwcNPyBiJqheKykpwe3btzF+/HhMnjyZTW/fvj2cnZ1FOh7iInSta2RkhIcPH+LKlSsYOHBgY5YJ\nxcXFUFZWbtTPEIaMjIxYG8uN5aeffsKtW7fg5+cHNzc3nmVz5sxBdHR0o5fh3bt3ACBSZdoQcnJy\nEtFY6N+/PyIiIuDr68vTyA0PD0ezZs3Qp08fXLt27bOUpbS0FLKyspCTkxP7+dpY3+uHDx/QvHnz\nOvNw65xTp07xVJifqvr31VRsbGzQp08f9t9eXl5wc3NDWFiYVHRaHz9+jMWLF8PQ0BCHDh2CpqYm\nz/IffvgBBw4caKLSCSYrK8v3+6nt/Bd3PcQwDD5+/AglJSWxbfNzoHYBESQnJweqqqqN3qGh76F2\n3N9LY7SZGuP7rd6Oqam8vBy7du2CsbExTp8+zZfn7du37P+PHDkSx44dQ1hYmMBOa2JiIrKysrBg\nwQKedE1NTbRv3x5hYWF8ndawsDCYm5vjv//+E2pfMjMz4e3tDXV1dRw9ehR6eno8yxctWoRt27YJ\nta3PRVC9xo1/ampqPOnNmjUT+8UiYdqEgAjPtLq7u6N9+/bYtWuXUHcpMjMz4evriz59+sDY2BhO\nTk44evQo37oTJkyAnZ0d/v33X8yYMQMWFhaYNWsWgP8NxUxJScHq1athY2MDCwsLzJ8/H+/fv0dF\nRQX27NmDb7/9FiYmJpg4cSJevnzJs/1//vkHP/74IwYNGoQePXqgZ8+emDZtGu7evVvvPgga411z\naF/1v+r5ysvLcfDgQTg5OcHExAS9e/fGokWLBJ708fHxcHNzg4mJCRwcHHDo0CGh7wQ9fPgQFy9e\nxPDhw/k6rFwODg48z4uVlpZix44dcHR0hLGxMezs7LB27VoUFBTwrFf9+K9duxY2Njbo0aMHpk+f\njoyMDDbfhAkT4OXlBQBYvnw5DA0N2c+r7Vk1Qc+P5eTk4Oeff0b//v1hbGwMa2treHl5ITExka9M\nNcXGxsLT0xNmZmawsLAQ+B1zv8/t27cjMjISLi4uMDExweDBgxEZGVnrMRZk+PDhyM3NRVxcHE96\naGgoBgwYAFVVVb51rl69irlz5+Lbb7+FsbExbG1tsXTpUmRmZvIcl0GDBgEA9uzZw55b3GequMPu\nYmNjsXXrVtjZ2cHU1BRv3rzhO19LS0sxfPhw2NraspUP18yZM9GjRw88e/as1n10cHDA0qVLAQBT\npkzhGxpz584dTJ06FRYWFjAzM4Onp6fAoYKGhob44YcfEBMTw57nhw4dqvP4AkC/fv1gZmaGAwcO\noLi4uM68eXl52Lx5M0aMGAFLS0uYmpqyncCaap6TI0aMgIuLi8DtLl26FKampigsLGTTXr16hSVL\nlsDW1hbGxsYYPHgwDhw4gMrKynr3qTYyMjJo1aoVXzD+888/sWjRIjg4OLD1yHfffYd///2XzVNR\nUQE7OztMmTJF4LYnTJgAe3t7nvKlpKSwd/JNTEzg6uoqcLh1YGAghg0bBnNzc5ibm2Po0KHYuXNn\nvfuzc+dOVFRUYNu2bXwdVgBo0aIFFi9eXOc2QkJCMG3aNPTt2xfGxsawt7fHmjVreL4L7v7v378f\nQ4YMYev44cOHIyAgQKQ8Neukuuq12uohYY5r9XooJCQEzs7OMDExEbkOkgTULqgiae0CruDgYLi7\nu8PMzAw9e/aEh4cHwsPDefKIWo8nJibC3d0dJiYm6N+/P44fP87mqf4byszM5DkGdT0LPGHCBL5h\nl5cuXcKYMWPQs2dPmJmZYeDAgVixYgVfmWo+05qfn4+1a9fCzs4OxsbGcHR0xM6dO/mGffr6+sLQ\n0BBv377F4sWL0bNnT5ibm2PBggXIy8ur99hW/x2Hh4fD2dkZpqamGDZsGJKSkgBUdZBGjx4NU1NT\nODg4IDQ0lGcbpaWl2L17N0aPHs3WGYJ+E76+vvD39wcADBgwgD2u1euq+n4vXLdv30b37t3x448/\n8pTl1atXMDc3x+TJk2s9z+r6foGqi29Hjx6Fk5MTjI2N0adPH/j6+vK0cYC62zGC5ObmIj8/HxYW\nFgI7ta1bt2b/v1u3buBwOIiIiBD4uE1oaChkZGQwfPhwvmXDhg3D9evXkZuby6bl5eUhJiZGYP7a\nHDx4EO/fv8e6dev4OqwA0Lx5c/z00091bkOY9iKXMHG6vjw1f5++vr7o378/gKoRb9UfeavtmVZh\n20YNbRMCItxpVVBQwNy5c7F8+XJERkbC2dm51ry5ubnw9PREdnY2xo8fDz09PVy7dg1+fn54+fIl\n352EkpISTJkyhf1SZGV5+9LLli1D69at4ePjgydPniAwMBClpaVo164dHj16hGnTpiErKwtHjhzB\nDz/8gKCgIHbd+Ph4PHnyBC4uLtDW1sbbt28RHByMCRMmICQkhB2iJ6xNmzbxpZ06dQq3b99Gq1at\nAFT9cOfPn4/Y2FiMHDkSEydORGZmJgICAnDr1i2EhoayzwXevHkTs2bNQtu2bfHdd98BAE6ePCn0\nna0rV64AQK0dVkHmzZuHa9euYciQIZgyZQr++ecfnDhxAn///TcCAwP5bvn/+OOP0NDQgLe3N7Kz\ns3H06FEsXryYfd5w9uzZsLCwgL+/P0aNGgUrKyuoqKgIXR6u+fPn49GjRxg/fjw6dOiA9+/f4969\ne3j06BFsbGxqXS8yMhILFy5Ep06d4OPjg7KyMgQGBsLLywtHjx6FpaUlT/64uDiEhITA09MT6urq\nOHPmDBYtWoRu3bqhU6dOQpW1c+fOMDIyQlhYGNuQTU1NxcOHDzFv3jy+hgFQ9RwIwzAYN24cNDU1\n8e+//yIoKAh37tzB+fPnoaioiM6dO2Pp0qXYuHEjBg4cyN696NChA8+2Nm/eDEVFRUydOhXl5eVQ\nVlbm69gpKChgy5YtcHd3x/Lly7F3714AQEBAAGJjY/Hzzz/X+Xzejz/+iOvXr+P06dPscFZucPjr\nr78wefJkaGpqYtq0aVBSUkJISAhmzZqFbdu28T1T8vDhQ1y/fh1jx46Fh4cH2rVrJ9Rxnj9/PqZM\nmYITJ07UOcTp1atXiIyMxODBgzFmzBiUlpbi8uXLWLJkCcrKyjBq1Kha13VxccHmzZuRkpKCrl27\nsukfP37ElStXYG9vz16EePHiBTw8PKCsrIxx48ahVatWuHnzJrZu3Yr09HSsXr1aqP0qLCxkLyQU\nFBTg2rVrSEhI4LujHBUVhbdv38Ld3R1aWlpIT09HUFAQxo0bh4iICLRq1QrNmjXDiBEjcPDgQWRm\nZqJt27bs+hkZGbh16xZmzZrF1qt///03pk6dCj09PUybNg0qKiq4du0ali9fjnfv3rGNnbNnz+KX\nX37BgAED4OnpCQBIS0vDn3/+Wee+lZSUID4+Hj179kTHjh2FOh6CnDhxAp06dcLkyZOhpqaGhw8f\n4vTp02x9xfXrr7/i119/hZubG6ZMmYKPHz8iNTUVf/75J8aPHy90nppErdeEPa5c0dHRyM3Nxdix\nY9GyZUuh6x5JQu2CKpLWLgCA1atXIyAgAJaWlvDx8YGSkhIePXqE2NhY9iKdqPX4P//8g4ULF2LM\nmDFwc3NDVFQU1q5diy5dusDGxgadO3fGpk2b4O/vj5ycHLZhbmFhIdLxTExMxPz589GrVy8sWLAA\ncnJyePXqVb2jl0pLSzF58mQ8evQIo0aNQrdu3fDnn39i7969ePz4Mdvxq27mzJnQ09PDwoULkZaW\nhoCAAMjLy2Pz5s1ClfX69esICwvD2LFjIS8vj99++w2zZs3Cpk2bsGrVKowdOxbDhg3DiRMn4Ovr\nC2NjY/b8KiwsxMmTJzF06FAMGzYMAHDjxg34+fnh/fv3mD9/PgDAw8MD+fn5uHr1KpYtW8ZeCKwe\nv+v7vXCZm5tj5syZ2LdvH/r374+BAweioqICixcvhry8PDZs2FDrHdT6vt81a9YgICAAtra2GDdu\nHDIyMnDixAkkJycjJCSE7wKmoHaMIK1atULz5s1x/fp1zJgxA1paWnV+JyNHjsTGjRsRHx8Pe3t7\nNr2oqAh//PEHrKys0L59e771nJycsH79ekRGRrJxISIigl22b9++Oj+X648//oCOjg6sra2Fyi+I\nMO1Fbr764nRDYrmHhwe6du0KPz8/DBgwgO/uc02ito0a2iYEU4+kpCSGw+EwQUFBTFlZGTNw4EBm\n8ODBTHl5OcMwDHP27FmGw+EwN27cYNfZuHEjw+FwmIsXL7JplZWVjLe3N8PhcJiUlBQ23cvLi+Fw\nOMy+ffv4PnvXrl0Mh8NhvL29edK/++47hsPhMKNHj2bLwTAMs2HDBobD4TCPHz9m04qLi/m2m5OT\nw1hbWzMrVqzgSe/fvz+zdOlS9t+vXr1iOBwOc/bs2VqPz8WLFxlDQ0Pml19+YdMiIiIYDofD/PHH\nHzx5Hz16xHTr1o3ZvHkzm+bm5sZYWFgwWVlZbFpWVhZjYWHBcDgc5tWrV7V+NsMwjI+PD8PhcJi8\nvLw683HFxMQwHA6HWb16NU/60aNHGQ6Hwxw/fpxN4x7/uXPn8uQ9cuQIw+FwmKdPn7Jp1c+T6pYu\nXcr079+frxzc/ElJSQzDMEx+fj7D4XCYgwcP1ll+bpm4ysrKGFtbW8bOzo55//49m/769WvGzMyM\nGTlyJJvG/T7NzMyYN2/esOnZ2dmMkZERs3Hjxjo/u/rnP3/+nDl27BhjYmLCfu6WLVsYGxsbpqys\njFm0aBFPORlG8LmYnJzMcDgc5sKFC2za8+fPGQ6Hw+zatYsvP/f35uLiwnz8+JFnWW3nK/e7DQoK\nYp49e8b06NGDmT59er37Wv3zqv++GYZh3N3dmR49ejAZGRlsWn5+PmNvb8/Y2toypaWlbDqHw2E4\nHA7z119/CfWZ3P3Ytm0bwzBVdYSVlRVTUFDAMIzgc+3jx488dQHDVNU5EydOZAYNGsSTXvOcfP36\nNWNoaMhs2rSJJ19UVBTD4XCYK1eusGnTp09nHBwcmPz8fJ683LonNTW1zn3jHk9Bf6tXr2YqKyt5\n8gs6Z/7991/GyMiI8ff3Z9PS0tIYDofD7N+/nyfvnj172POVe0ycnJyYUaNG8XxHDFNVr5qamrJ1\nydy5cxknJ6c690eQlJQUhsPhMGvWrBF6HUH1hKB9DwkJYTgcDvP333+zacOHD2dmzJhR5/aFyVOz\nTqqeVrNeq1kPiXJcuee3kZER8/LlyzrLJKmoXSDZ7YJbt24xHA6HWbhwIVNRUcGzrHodI2o93rVr\nV+bBgwds2sePHxkbGxtm3rx5PJ/h5eXF9OvXjyetruPm5eXFeHl5sf9et24dY2FhwZSVldW5nzXj\n5IkTJxgOh8P89ttvPPnWrVvHcDgcJjo6mk1bunQpw+FwmLVr1/LkXbt2LdOtWzc23tSmtvbEtWvX\nGA6Hw3Tv3p3nnHvy5AnD4XCY9evXs2nl5eV8cZxhGMbX15cxMzPjWbZt27Zav3thfi/VlZWVMe7u\n7kzv3r2ZrKwsNk5ERETUuc/VP6/m98vdv9mzZ/OcY1euXGE4HA6zYcMGNq2udkxt/P39GQ6Hwxgb\nGzNeXl7Mtm3bmBs3bvDVtwxT9Vvp1q0bs2DBAp50bvyoeQ5W3x9vb29m9OjR7LIxY8YwPj4+DMMw\nTL9+/XjOU0EKCgrY4yAsQd+RsO1FYeK0MHkE/T5rtsW4uN9f9XNRlLaRqG3C6kR65Y2cnBy8vb2R\nlpaG8+fP15ovOjoaHTt25OmZy8jIYNq0aQAg8GrZuHHjat0e98oAF/eqzqhRo3jGVXPvqL169YpN\nqz5G+sOHD+xtf1NTU9y/f7/WzxTG48eP4evrCysrKyxfvpxNj4iIQNu2bWFubo53796xf23btkXH\njh3Z4a7Z2dl48OABnJ2d0aZNG3b9Nm3aCD1pEneonKDhqIJcvXoVADBjxgye9LFjx0JVVVXg8681\nJ1aysrICAL4hV59CUVER8vLyuHnzJt9Q1ro8ePAA2dnZ8PDw4LkKra2tDVdXVzx8+JBvOIWDgwPP\n3ajWrVvjm2++4TlvhOHs7IyKigpcvHgRlZWVuHDhApydnWt9foR7LjIMw95p69KlC9TV1UU+F93d\n3YV+CH7ixImwtbXF+vXrMX/+fCgpKWH9+vUifV512dnZuH//PpydnXmuVqqpqbF3UmruT9euXUW+\n2s41f/585OXl4ejRo7XmUVBQYOuC0tJS5OXlITc3F3369MHz58/5hpRWp62tjV69eiEyMpJnWFR4\neDhatGgBOzs7AFUTZMXFxWHQoEEoKyvj+W1z83CHhdVn8eLFOHLkCI4cOYIdO3bAw8MDJ0+exMaN\nG3nyVa+/ioqKkJubixYtWqBTp048x1hfXx+WlpZ8w6HDwsJgYWHB3vH8559/kJqaCldXVxQUFPDs\nw7fffstOvABUPcf55s0b9t/CErVOqg133ysrK9my9uzZEwB49l1NTQ2pqal1DnUXJs+nEOW4ctnZ\n2QkcOvaloXYBL0loF1y8eBEAsGDBAr47bty7aA2px3v06AEjIyP23woKCujRo4fIsbM+6urq+PDh\nA+Li4kQaEh0dHQ1lZWW+0RPTp09nl9dU8xyzsrJCRUUFzyNQdanZnuCehyYmJjwjdwwMDKCmpsbT\nbmrWrBkbx8vLy/H+/Xu8e/cO1tbWKC4u5nkMRBh1/V6qk5OTw+bNm1FSUoLZs2dj7969GDZsWL2z\n7taF+/udNm0az51aR0dHdOrUSeCxF6UdM2vWLOzYsQNmZma4ffs2/P39MWXKFPTt25dvZFubNm1g\na2uL6OhontgfGhoKZWXlOu8aDh8+HHfv3sXz58/x/Plz3LlzR6ShwUVFRQDEF//qay8KE6cbGsuF\n1ZC2UUPbhCI/ne3q6or9+/fj119/rfU5sIyMDNja2vKlc4dE1BwHraGhUeewl5q38bl5a95O5j4s\nXP15hMLCQuzYsQMXL17ke8WCrq5urZ9Zn7dv32LOnDlo1aoVdu7cydNRSUtLQ2ZmZq1DWrn7w60U\nBQ0LE3aoGPeHUVRUJNTQoYyMDKioqEBbW5snXUFBAXp6egLfu1Tb8X///r1QZRSGgoICfH194efn\nh759+7Kv7HFxcalzCCv3GArKw01LT0/nCSqChoW0aNFCqOdYqmvVqhX69u2LsLAw6Onp4fXr13VW\nbmlpadi6dStu3LjBN5RX1Gnsaw4XrouMjAz8/PwwcOBAPHnyBLt27eJpDIlK2GNevUISpbw19ezZ\nE3379sXRo0drnW6eYRgcO3YMgYGBeP78OV9DJz8/v84g4urqihUrVuCvv/5Cz549UVBQgNjYWAwf\nPpwNqi9evADDMDh8+DAOHz4scDs5OTlC7VP37t15JmIaOnQomjVrhiNHjmDkyJHssyNZWVnYunUr\nrl27xvd7qznUys3NDT/99BPu378PExMT/PXXX3jx4gXbYAPANoLWrVuHdevW1bkPM2bMQFJSEjw9\nPdGuXTvY2NjA0dERDg4OdU7AUb1O+hT37t3D9u3b8ddff/HNxF3997JgwQLMnTsXTk5O0NfXR58+\nfTB48GCeoVnC5PkUohxXrk/5TUgaahdUkZR2wfPnz6GsrFznRZGG1OO1xc5//vlHqHIJa9y4cbh0\n6RJmz56NVq1awdraGv3798eQIUPqnEgvIyMDurq6fBPKaGlpQV1dXWD7RkdHh+fforZvajsPBR0r\ndXV1vu2GhYXhyJEjePLkCSoqKniW1ZxnpC71/V5q6tSpExYsWAA/Pz+0bdv2kycA5B7b2s4nQc9J\ni1oHDh06FEOHDkVpaSlSUlIQGxvLDv/X0tJib6gAVXNVXL9+HVFRURg9ejTevHmDmzdvwsXFpc7H\nPL799ltoaGggLCwMDMNAQ0OD7XgJg7vtT41/wrYXhYnTDY3lwmpI26ih8U/kTqusrCy+++47LFiw\nAGfPnhXLlMf1zQBX2yxVwsxe9f333yMpKQmTJ09G9+7doaamBllZWezfv7/BVwdLS0vh4+OD/Px8\nnD59mq/xyDAMdHV1sWbNGoHri3OaaG4FkZKSwvODFafajrMwV0Br+0EImrTGy8sLDg4OiI6ORmJi\nIo4ePYoDBw5g9erVcHd3F63QdajtWY+GGD58OBYuXIh9+/ahS5cuAmerA6oaSV5eXpCXl4ePjw86\nduyI5s2bQ0ZGBt9//73IE2yIOstocnIy2/BPSUmp9/kEcfvUWR7nz5+P0aNH49ChQ+jbty/f8kOH\nDrGvdZo9ezZatmwJOTk5xMbG4ujRo/VOkjR48GCsXr0aFy5cQM+ePXH58mWUlpby3Nngfkeenp61\nHr9PafDa2tri5MmT+PPPP2FoaIjKykpMmzYNmZmZmDRpEgwMDKCsrAxZWVmsX7+e75wZOnQo1q1b\nh7CwMJiYmCA0NBRKSkoCr577+PjwPevNxa1TvvnmG0RFRSE+Pp79CwkJQd++fXHgwIFa64WOHTtC\nXl4eKSkpDT4W6enpmDhxInR0dLB48WLo6OhASUkJFRUVmD59Os++W1pa4o8//kBMTAwSEhLwxx9/\n4OTJk3B3d2dHFAiTRxyEOa5c0jTzKbULJKtd0Fg+JXbW1TiuqKjg+d5atmyJc+fOISkpCXFxcUhI\nSEBERAQOHTqEkydPNmjOjNrUtk/CxuTa1hdmu5GRkViyZAns7Ozg5eWFNm3aQF5eHg8fPsSWLVtE\nmtxP1PqEYRhcv34dQNUFnczMTL5ZYhtbQ2dLV1BQgKmpKUxNTWFpaYkpU6YgNDSUpw3s6OgINTU1\nnD9/HqNHj8b58+dRWVmJkSNH1rvtIUOG4Pz582AYBkOHDhXpt6mqqgptbe1Pin+itBeFidMNjeXC\nakjbqKHxr0HzYA8ZMgRdu3aFv7+/wJcK6+rqChzWwB2a9SkNO1Hk5+fj+vXr8PHxYScz4BJmBsza\nrFixAnfv3sXevXthYGDAt7xjx464efMmevXqVedVQe4VvrS0NL5lgtIEcXR0xL59+xASEiJUp1VX\nVxfx8fF48+YNz93W0tJSvHr1Cj169BDqc4Ul6MoigFobBu3bt4eXlxe8vLzw/v17eHh4YPv27bV2\nWrnnkqBhf5/jfOPOFJycnIxFixbVmi8pKQlv377F77//zvPu1ZKSEr67rOJ+TcB///2H1atXw8LC\nArq6uti/fz/s7e0b/F3Xdcy5v3txH3Pu7IsnTpxA9+7d+ZZHRETAysoKW7Zs4UmvPvN0XVq0aIF+\n/frh0qVLWLFiBcLDw9lhw1x6enrsd1P9Lqm4lJWVAfjfFdonT57gyZMnAl9llZeXx9coVlFRwZAh\nQxAeHo4FCxYgKioKjo6OPINlMUoAACAASURBVHeYuVc3FRUVhdoHJSUlODo6wtHREQzDYOvWrTh4\n8CASExMFXjzgrtO3b19cv34dL1++bNAV1atXr+LDhw/Yv38/z7lU23A5NTU1uLq6wtXVFeXl5fD1\n9cXZs2cxY8YM9u6UMHkaStTjKo2oXSA57QJ9fX3ExcXh1atXtd5t/dz1eIsWLQAIHlWUnp7ON2mb\nnJwc+vbty9YzJ0+exKpVqxAeHg4PDw+Bn6Grq8uOzKjeKM7OzkZ+fv5nO8eEFRERwcbk6p1cQY9e\nibtdcPz4cdy4cQPz5s3DiRMnsHjxYgQFBTX4lXDVzyfuYxxcz54947ujLS6mpqYAwPcYmKKiIoYM\nGYLg4GBkZGQgLCwM2traQo2uGTFiBAIDA9n/F5WjoyM7AVX19p6wRGkvAsLF6YbEcmE1dtuougZd\nNpORkcG8efPw+vVrnhn5uPr3748XL16wM9sCYG8dAxD4CpTGICsrCxkZGb4rZsnJyUJNbS/IwYMH\nERoaiu+//56dDromZ2dnFBUVCXwPIcMw7DObbdq0gZGRESIiIniGKGVnZ+PChQtClcfY2BiDBg1C\naGgo33TqXDExMezzBNxjX3N66dOnT6OwsFDs303Hjh1RWFjI8xLl0tJSduZhrg8fPqCkpIQnrUWL\nFtDR0alz6KyRkRG0tLQQFBTE8+xCZmYmLly4ACMjI56hweKmqKiIlStXwsfHp87KjXslq+a5eOjQ\nIb6rqdwZ9MQx/LqyshJLly5FZWUlNm3ahF9++QXa2tpYsmRJva+RqU3r1q3ZV3RUn6K+sLAQgYGB\naNOmDUxMTD657DXNnz8fHz58wMGDB/mWycrK8h3bd+/eITg4WOjtu7q6Ijc3F6GhoUhOToazszNP\nQ6Jly5bo06cPQkNDBTa+CwsL+V6rIArub5TbKed+ds39OnfuHN+QRi53d3fk5ubi559/RkFBAd9V\n5e7du6NTp074/fffBT47Xn0IT/Vp/4Gqer9bt24A6h/O7uPjA1lZWSxatEjgsPv8/Hy+CwzV1fZ7\nEfTd1yynnJwcOBwOTzmFyfMpRDmu0oraBZLTLhgyZAgAYMeOHXzxhbvfn7seV1VVRatWrfiebYuM\njOTrcNT8vQL/qxfriov9+/dHcXExTp06xZP+22+/Afh855iwBNVzJSUlPK8R4uK2C8RRX6WmpmLL\nli2wt7eHt7c31qxZg0ePHvG9PkgU3PP+yJEjPPsTHR2NtLS0Tzr2Hz58qHWm25iYGACChyWPGDEC\nDMNgw4YNSE1NxbBhw4QaLWBubo6FCxdi0aJFMDMzE7m806dPh7q6OpYvXy7w2eiSkpI6R/iI0l4U\nJk5/SiwXRmO3japr8BuHBwwYAFNTU9y7d49v2YwZMxAVFYWFCxeyU9vHxMTg+vXrGD9+PNtYaGyq\nqqqwtrbGoUOHUFJSAn19fTx58gRnz56FgYGByGPOnzx5gm3btkFXVxdt27blm/TEwsICenp6cHFx\nwZUrV7Br1y7cvXsXNjY2UFBQQHp6OqKjo+Hi4sJe4V28eDGmTZsGDw8PdmKJwMBA6Onp4fHjx0KV\na/369Xj37h2WLl2Kc+fOwc7ODhoaGnj79i1iYmLw999/Y+vWrQCqxuvb29vj+PHjyMnJQc+ePfHk\nyRMEBQXByMgIY8aMEemY1MfFxQXbtm2Dt7c3Jk6cCBkZGZw/f55vsqLnz59j4sSJGDx4MLp06QJl\nZWXcunUL8fHxdQ4NlpOTw7Jly9hp+N3d3dlX3pSXl9f7LixxqO0ZruosLCzQsmVLLF26FOPHj4eK\nigqSk5Nx//59aGho8ORt3bo12rdvj8jISHTq1AkaGhrQ1dVt0J3RI0eO4ObNm1i/fj17xX3jxo2Y\nOHEiNmzYIPQrWmpatmwZJk2ahDFjxmDs2LFQVFRESEgI/vvvP2zbtq3BV2vr0rVrVwwePJidZKQ6\n7rv4Fi1ahN69eyMrKwuBgYFo166d0BN7OTg4QEVFBevXr0dFRQX7CoLqVq5cCU9PT7i7u2PUqFHo\n0qUL8vPzkZqaisuXL+PChQtCXc1PTExkG6QFBQW4ceMGoqOjYWVlxT73980336BTp07YuHEj/vvv\nP7Rp0wZ3797F1atXa717wn3NTEREBLS1tfmuesrKysLPzw9Tp06Fk5MTRo0ahQ4dOuDdu3d4/Pgx\nrl69yl5gmjp1Klq2bAlzc3O0bdsWb968QUBAABug6mJsbIwNGzbA19cXgwcPxrBhw2BgYICKigqk\npKQgKioKmpqa+OGHHwSu369fPygqKmLmzJnw8PCArKyswOd6gapXEVhaWsLExAStWrViX1vRqVMn\ntqErTJ5PIcpxlWbULpCMdkHPnj3h6emJwMBAvH79GgMGDICSkhJSUlLw8eNH9hU9n7seHzt2LPbs\n2YPFixfD0tIST58+RVRUFN9ojOXLlyMnJwc2NjZo3749cnNzERgYCCUlJfY1cIKMHj0aZ8+exYYN\nG/Dvv/+ia9eu+OuvvxAeHo7+/fvXekGhqQwYMACXLl3CzJkzMXDgQLx//x4hISEChz9zLyBs3boV\nrq6ukJeXh7W1NftKJWGVlZVh8eLFUFFRYZ+/d3R0xKhRo3Do0CHY29s3aIIcAwMDjB8/HgEBAZg+\nfTr69++PjIwMBAQEoH379nW+sq4+Hz58wPjx49GtWzd2ArvS0lI8ePAA4eHhaNmyJd/r4oCq34Ge\nnh4uX74MQLS7pjVfUSaKdu3aYffu3fDx8YGLiwtcXV3RvXt3yMrKIjU1FVFRUcjNzeV7Vy6XKO1F\nYeL0p8RyYYmrbVSfBndagao7H9yZ/6rT1NREYGAgtm3bhtDQUBQWFkJPTw++vr4CT6zGtHXrVmzY\nsAHnzp3Dhw8f0LVrV+zduxdhYWG4efOmSNvKzc1FZWUl0tPTsWTJEr7lfn5+7G3yHTt24NSpUzh7\n9ix27NgBWVlZthE5dOhQdh0bGxv4+/tj+/bt2LVrF7S0tNj39tV2QtekpqaGY8eO4dy5czh//jz2\n79+PoqIiaGpqwtzcnB0OyrV7927s3bsX58+fx5UrV6CpqYlx48Zh/vz5Yn+uRl1dHfv27YOfnx+2\nb9+Oli1bYsyYMbC0tOQ5F7S1tTF8+HAkJSWxs7jq6upiyZIlmDhxYp2f4eTkBBUVFezbtw+7du2C\nrKwszM3N2ZnmJEGLFi1w6NAhbNy4Ef7+/mjWrBl69+6N33//HZMmTeLLv2nTJvj5+WHDhg0oLS3F\nyJEjRe60pqSkYMeOHRg0aBBPx79Xr16YOnUqDh06hAEDBuDbb78VeX8sLS1x4sQJ7Nq1i315dLdu\n3bB///4GbU9Y8+bNw+XLl/muNs6cORPl5eUIDQ3F5cuXoauri9mzZ0NZWRnLli0TatvcBlFoaCi6\ndOnCM/MjV4cOHXDu3Dns3bsXV65cwalTp6Curo6OHTvC29tb6Amuqt9tkZeXh46ODmbPno3Zs2ez\nw2zk5ORw4MABrF+/HsePH0dZWRnMzMxw7NixOq/Surm5Yfv27Rg+fLjAq8rm5uYIDg7Gvn37EBoa\nyg417ty5M8+xGjt2LMLDwxEQEICCggK0bt0a9vb2mDt3Ll/gFMTFxQXdu3fH0aNHce3aNQQGBqJZ\ns2bQ19fH+PHj4eXlVeu6HTt2xP79+7Ft2zbs2LEDSkpKsLe3x9atW/kmspk8eTKuXbuGw4cPo7i4\nGFpaWnB3d8ecOXPYRrcweT6VsMdV2lG7gFdTtQtWrlwJQ0NDnD59Gjt37mTfA149nn7uenzWrFnI\ny8tDeHg4Ll++DFNTUxw+fJhv8rJhw4YhODgYZ86cQV5eHjQ0NGBubo45c+bUOZRfQUEBR48exc6d\nO3H58mWEhISgbdu2mDNnDubOnSv2/flUw4cPR35+Pk6cOIF169ahTZs2cHNzg4WFBaZMmcKT18bG\nBt7e3ggODsayZctQWVmJ33//XeRO665du/Do0SPs27ePfe86UPVe9uTkZCxZsgRhYWENem54+fLl\n0NPTQ1BQEDZs2AA1NTUMHToUCxcu5HucRRTq6upYt24dO7FSdnY2Kioq0K5dO7i7u2PWrFm1vudz\nxIgR2L17N4yNjeuc1FPcrK2tERERgSNHjiA2Npad2ElXVxdOTk51xj9R2ovCxOlPjeXCEFfbqD4y\njKgzwBBCCJFYhw4dwubNmxEVFYVvvvmmqYtDCCGEEPLJxDeNKiGEkCZVUVGBoKAgmJubU4eVEEII\nIVLjk4YHE0IIaXo5OTlISEhAbGwsXrx4IXCYIiGEEELIl4o6rYQQ8oVLTU3FDz/8AA0NDXh7e8PR\n0bGpi0QIIYQQIjb0TCshhBBCCCGEEIlFz7QSQgj5ZDnPXzR1EQghhJCvlrTHYbrT+gVaJi+eKaql\nhV9ZHh2TGtZnpzR1ESSKjIY2mLw3TV0MiSKjoS32bc6WURcpvz/z6S82J6QpzJZRa+oiSBR/poCO\nSQ3+RelNXQTJo9wCKOZ/3/ZXTbmFWDcnahwGvpxYTM+0EkIIEQsaukMIIYQ0HWmOw9RpJYQQIhay\nMjJNXQRCCCHkqyXNcZg6rYQQQsRCmq/wEkIIIZJOmuMwdVoJIYSIhaz0XuAlhBBCJJ40x2HqtBJC\nCBELab7CSwghhEg6aY7D1GklhBAiFtL8LA0hhBAi6aQ5DlOnlRBCiFhI8xVeQgghRNJJcxymTish\nhBCxkOZnaQghhBBJJ81xWJo75IQQQgghhBBCvnB0p5UQQohY0FVQQgghpOlIcxymTishhBCxkJHi\nCSAIIYQQSSfNcZg6rYQQQsRCmq/wEkIIIZJOmuMwdVoJIYSIhTRPAEEIIYRIOmmOw9RpJYQQIhbS\nfIWXEEIIkXTSHIep00oIIUQspPml5oQQQoikk+Y4TJ1WQgghYiHNV3gJIYQQSSfNcZg6rYQQQsRC\nmp+lIYQQQiSdNMdh6rQSQggRC2m+wksIIYRIOmmOw9RpJYQQIhaykOJLvIQQQoiEk+Y4TJ1WQggh\nYiHNw5IIIYQQSSfNcZg6rYQQQsRCmoclEUIIIZJOmuMwdVoJIYSIhTRf4SWEEEIkXWPH4bKyMpw+\nfRpxcXEoLCyEvr4+PD09YWJiUud63t7eyM7OFrhMW1sbu3btqvezqdNKCCFELBr7WZqmDJaEEEKI\npGvsOPzrr78iOTkZzs7O0NbWRkxMDNavX49Vq1aBw+HUut6kSZNQUlLCk/b27VsEBgbC1NRUqM+m\nTishhJAvQlMGS0IIIeRrlpqaioSEBEyaNAnOzs4AADs7OyxatAgBAQFYtWpVretaWVnxpZ09exYA\n0K9fP6E+nzqthBBCxKIxhyU1dbAkhBBCJF1jxuGkpCQ0a9YMAwYMYNMUFBTg4OCAwMBA5ObmQlNT\nU+jtxcfHQ0tLC4aGhkLll+bndQkhhHxGsiL+iaKuYJmSkoLc3FyRtidqsCSEEEIknahxWJRYnJaW\nBh0dHSgpKfGkd+nSBQzD4Pnz5yJtKyMjA3379hV6HbrTSgghRCwacoU3KyuLL01FRQUqKio8acIE\nS2Gv8HKDpZubm+gFJoQQQiRUQ++0ChOL8/Ly0LJlS7583NgrysXjuLg4AKBOKyGEkM+vIRNA+Pj4\n8KWNGjUKY8aM4Ulr6mBJCCGESLqGTsQkTCwuLS2FnBx/11FeXp5dLozKykokJCSgU6dO0NXVFbqM\n1GklhBAiFg25wrtnzx6+tJp3WYGmD5aEEEKIpGvonVZhYrGCggLKy8v58pWVlbHLhfHo0SO8e/eO\nnZ9CWNRpJYQQIhYNiZVaWlpC5WvqYEkIIYRIuobOwyRMLNbQ0BA4qombJuwjOvHx8ZCRkYGtra1I\nZaSJmAghhIiFrIxof6Jo6mBJCCGESDpR47AosVhfXx8ZGRl8r5B7+vQpAKBjx471bqOsrAzJyckw\nMjIS+MhPnfsmUm5CCCGkFrKQEelPFE0dLAkhhBBJJ2ocFiUWW1tbo6KiAlevXmXTysrKEBMTA0ND\nQzauvn37FhkZGQK3cfv2bRQVFTVoTgkaHkwIIUQsGvP9cNbW1rhw4QKuXr3KDu2tLVh+/PgROjo6\nfNv4lGBJCCGESLrGjMMGBgawtrbGiRMnkJubC21tbcTGxiI7Oxtz585l8+3ZswePHj1CUFAQ3zbi\n4uIgLy8Pa2trkT+fOq2EEELEojGH7jR1sCSEEEIkXWMPofXx8cHp06dx/fp1FBUVoUOHDli2bBm6\ndu1a77rFxcX4+++/YW5uDmVlZZE/mzqthBBCxKIRL/ACaNpgSQghhEi6xo7DCgoKmDBhAiZMmFBr\nnpUrVwpMV1ZWRkBAQIM/mzqthBBCxEJWpnHDZVMGS0IIIUTSNXYcbkrUaSWEECIW0hsqCSGEEMkn\nzXGYZg8mhBBCCCGEECKx6E4rIYQQsZDmK7yEEEKIpJPmOEydVkIIIWIhzcGSEEIIkXTSHIep00oI\nIUQsZKR4AghCCCFE0klzHKZOKyGEELGQ3lBJCCGESD5pjsPUaSWEECIWNLMfIYQQ0nSkOQ5Tp5UQ\nQohYSPGoJEIIIUTiSXMcpk4rIYQQsZCR6oFJhBBCiGST5jhMnVZCCCFiIb2hkhBCCJF80hyHqdNK\nCCFELKQ5WBJCCCGSTprjMHVaCSGEiIWsNEdLQgghRMJJcxymTishhBCxkOZnaQghhBBJJ81xmDqt\nhBBCxEJ6QyUhhBAi+aQ5DlOnlRBCiFhI81T7hBBCiKST5jgsze+gJYQQQgghhBDyhaM7rYQQQsRC\nii/wEkIIIRJPmuMwdVoJIYSIhaxUh0tCCCFEsklzHKZOKyGEELGQ3lBJCCGESD5pjsPUaSWEECIW\n0jwBBCGEECLppDkOU6eVEEKIWEhxrCSEEEIknjTHYeq0EkIIEQtpfqk5IYQQIumkOQ5Tp5UQQohY\nyEpvrCSEEEIknjTHYXpPK5E4zTU14HXmBFblZWBJ6n308BwlMN/kC2ewMjcdALAyNx1rirIw//YN\ndvn0Kxfw03+p+CXnJeb9FY9urk6fpfzk83iXmws3rylQ1fkG+iY9cfJMiMB8m3fthbGxMdT1uuCb\nHlbYvGsvu+zlq3So6Xbm+ZPVbIete/w/125IFRkR/wghkktZUxOzQ05iZ+EbrHv+EL3GjhaYzyfy\nLABgR8Fr7Ch4jT0fc7DiXhK7fF3aA+wqzmKXz7sU+lnKTxrfu3e5GOk5Hipt2qNjV2OcPH1GYL7N\n23fBuKcN1NTU0Km7KTZv38WzXL+bCZq30oaqlg5UtXQwyHXk5yi+VBI1Dn9JsZjutBKJM3zXFlSU\nlmKdDgftzEwwOew0Xt97gKxHKTz5jrpWBVC/sjys1NTFjD/C8ezadXZ5+EJfZD1KQWVFBfSsLDHt\nYii2du+JgjeZn3V/SOPw+eFHKCgo4M0/93Hn/gO4eExAD2MjGHUz5MnHMAx+//13mHTQwrO05xjs\n5gk9nfbwdB+BDnq6KEh/xuZNe/ESBhY2cB/m/Ll3Ryp8ScGPEFK3sb9uRXlpKZa07QxdM1P4RJxB\n+t37eF0jFu9xcoc/U4AFau0AAAuvRSIlOpYnz17XMUi5GvO5ik4+E+/vf4CCggIy057gzr37cHb3\nQA8TYxh178aTj2EY/H5wH0x72+LZgzsYNGwk9HR14Dnanc1z4UwgHB3sP+8OSCFpjsN0p1UEK1eu\nxMqVK5u6GFJNXlkZRm7DcGXlOpQWFeHFjSQ8Dr8I8/Eeda6n0bED9Pva4O8TgWzam/sPUVlRAaCq\nwpSVl0cLPZ1GLT/5PIqKinH2QgRW/7gEqqoq6GvTG8OGDsLx08F8eZfM94aFhQXk5ORgaNAFw5yG\n4EbyLYHb/T3wDOz6WEO/g15j74JUkhHxP0JERXH481BQVoa5+3CcX7EWH4uK8OxGIu6ej0TvCWPr\nXK9Vxw7o0q8Pkn8/9ZlKSppKUVERzoadx5oVP0FVVRV9+9hgmNMQHD91mi/vkoXzYWFuVhWHOQYY\n7uKEG0lJArZKPpWocfhLisXUaa0hLy8PQUFBeP78eVMX5avUmtMFleXlePv0f3e/Xt+9j7Y1rtrV\nZOHliefxich78ZInfVJoIFYXvIF3QjTSYuOR8eftRik3+byePHsGOblm4HTpzKaZGhvhUco/da7H\nMAziE5Nh1JUjcNnxwDOYOHaM2Mv7tZCREe2PEEEoDje9tv8fi7OeprJpGXcfoL1R3bG498SxSI1L\nQE6NWDw14DdszkrDvEuh0DE1bpQyk8/rydNUyMnJgWPQhU3rYWKCh48f17kewzCIu5EIo26859L4\nqTPQpmNnDHIdibv37jdKmb8GosbhLykW0/DgGvLy8hAcHAwtLS3o6+vzLFu+fHnTFOoroqiigo/5\nBTxpJfn5UFRTrXM9Cy9PRPtt4Us/NsITsnJy6DLAHlrdDMEwjFjLS5pGYWER1NXUeNJaqKuhoLCw\nzvVWbtiCyspKTBnvybcsPjEZmdnZGDXMRaxl/ZrQVVAiDhSHm56iqio+1IjFH97nQ6meWGw9cSyi\n1m7mSTs8fjpe/n0HkJGBw/y5mHcpFCu7WuLD+/diLzf5fAqLBMThFur1x+F1flVxeMJ4Ni3g8EFY\nmPUAwzDY+as/Bg93R8rtm9DQ0GiUskszaY7D0rxvYicnJwc5OernN6aPRUVQVOetBBXV1PGxoO5K\nUFVbCw/OhglcVllejieX/oCBY390cxkqtrKSpqOqqoL8At4GVX5BIdRUa29Q7TlwGMcDzyD89HEo\nKiryLT8WeAburs5QVVURe3m/FtI6+QORHBSHP4+PhYVoXiMWK6mroaSOWNzZ1gbq2m3xdzDvREvP\nEpJQVlKCsg8fcGnDVnzIe48u/fo0SrnJ56OqIiAO5+fXHYf37MHvJwMRERLEE4dtbazRvHlzKCsr\nY9nihdDQaIG4hMRGK7s0o4mYxCgoKAjBwcHYs2cPgoKCcOtW1bNlvXv3xrRp03hO4mvXruHixYtI\nT0+HkpISzM3N4eXlxXPlpbKyEsHBwbh69SqKiopgYGCAadOmwc/PD927d4e3tzcAoLCwECEhIbh7\n9y6ysrIgKysLQ0NDjBs3jr2S+/DhQ6xatQoAsHfvXuzdWzXL6Ny5c2Fvb88+R7Ny5Urk5eVh1qxZ\n8PDwgJubG88+Pn36FD/99BN8fHxgZ2cHAHj79i0CAwNx584dFBcXo3379hgxYgT69u0r/oP8BXv7\nJBWycnJo1eUb5KT+CwBo18MYmY/qHm7yMDQcpUVFdeaRlZNDy86dxFZW0nQ4nTujvLwCT5/9C4PO\n3wAA7j14iO5dDQXmP3z4MDbu3IPYiHPQ1WnPt/zDhw8IDr2AkBOHG7Xc0k7mSxpn9BWjOExxuD6Z\n/x+Ltbp0RlZq1eM6uj2M8d/D2mOxzaRxuBNyHh/ricUMw1BdIQU4Bl1QXl6Op6nPYPD/j+rcvf+A\nb9gv1+Fjx7FhwwZcvxQBXZ265xeRkZGhkXENJM2/rSa707p161aUlpZi/PjxsLGxQUxMDM6c+d9U\n2WfOnIG/vz90dHQwadIkDB48GH/++SdWrVqF0tJSNt/JkycRHByMzp07Y8KECdDW1sa6detQUlLC\n83mZmZm4desWLC0tMWnSJLi6uuLly5dYuXIl3r17BwDQ0dGBp2fVsEFHR0f4+PjAx8cH3QT8ADU0\nNNC9e3ckJvJfCUpMTIS8vDx69eoFAMjNzcVPP/2Ex48fw8nJCZMnT4ampiZ27dqFmJiYTz6W0qSs\nuBgPz13AwF9+hLyyMjr26Y3urkNxO4D/wX4AkFNSAgD8dewkT3obQwNwBjtCTkkJsnJyMBs3Bvr9\n+iDt+g1BmyFfGBUVZbi5OOGX9ZtRVFSMG0k3ERZ5CRM8+F+PFBB0Fj/++CMuhwTiG/2OArd3LjwK\nmhot0L+fbWMXnRCJQXGY4nBtSouLcTvkPFxX/wQFZWV07mONHsOdkXy89gmWLMeMROLRAJ40TT1d\ndO5jjWby8pBTVMTAH+ZDtXUrPLtBk/B86VRUVOA23BU/r1mPoqIi3EhMQlhEFCaM5Z84MyAwCD+u\nXIMrV67gm076PMtevnqFG4lJKC0tRUlJCTZv34W3OTmwtbb+PDtCvhhNNsamc+fOmDlzJvvvwsJC\nXLt2DV5eXsjKysLZs2fh5eUFV1dXNo+ZmRlWrFiBmJgYDBo0CHl5eYiIiIC1tTUWLlzI5jtz5gxP\n4AWADh06YOfOnZCV/V8/3c7ODt9//z2io6MxatQoaGhowMLCAoGBgeBwOOzV2drY2Njg0KFD+O+/\n/9C+fdXdG4ZhkJSUBHNzczRv3hwAEBgYCFlZWWzatAkqKlVDDwcNGoT169fj1KlTsLOz4ylXffzK\n8oTO+yWr/n7W7+/WHeBmXDkv1Da/uxlbfybyRdh76DCmTp2KthwTtGrVCvv8/WFs8y3i4uIwdOhQ\nFP7/czUr/LYgJycHVgP+955eLy8v+Pv/712svweHYcKkyZDVbPfZ90OaSO/1XelEcbjhcdifKag/\nk5So/n7WXx4KnnkdAJq3aIHvoyOE2ubWnBefXC7S9PbuP4ipU6dCS9+gKg7v2wejntZ8cXj5mvXI\nefeOvYgE/C8OF5SnY873i/Hs2TMoKSnBzMwMUVEX0UqPRsY1hDTH4SbrtA4cOJDn3127dsXNmzdR\nXFyMmzdvgmEY9O7dG/n5+WwebW1taGho4NGjRxg0aBAePHiAiooKDBo0iGdbQ4cO5QuW8vLy7P9X\nVlaiqKgISkpKaN++PdLS0hq0D9bW1jh8+DASEhIwalRVB+vp06d4+/YtvLy8AFQFz5s3b8LW1hYV\nFRU8+2NmZoY7d+7g9evX0KlnqER1y+TpwfTq/Mry6JjUsD47pf5MXzhNWeDcUX+eNCbvDfqaGKAg\nPRVM3hsAwL+3EyGjoc3+u3perounj/GlSTsZDW3xb1PsWySNieJww+PwbBm1+jN9RfyZAjomNfgX\npTd1ERpdS6VmCD153LIflwAAIABJREFUjDex+D36WZqiMCsDKK6abCvt4Z2qZcot2DRuXqNOuriX\nFMe/8eKvZKIu5RZi3Zw0x+Em67S2bt2a59+q///gdlFREd68eQOGYeDj4yNwXW7Ayc7OBlAVRGtu\ni3sllauyshKRkZG4fPkysrKyUFlZyffZolJXV4eRkRESExPZYJmQkABFRUVYWFiwZS0qKsLly5dx\n+fLlWvdHlGBJCCGSSJqfpZFGFId594fiMCHkSyfNcbjJOq21DcNhGAaVlZWQlZXFjz/+KDBPQ4Lb\nuXPncPr0afTv3x8eHh5QVVWFjIwMjh079kkPe9vY2ODAgQNIT0+Hjo4OkpOTYW5uDqX/f9aSu217\ne/taJ3vQ09Nr8OcTQoikkJXeWCmVKA7/D8VhQog0kOY4LJHzxmtra6OyshLa2trQ0tKqNV+bNm0A\nAG/evOG5YlxQUICiGrPXJSUlwcjICHPmzOFJLyoqgppaw4e09O7dG7/99hsSEhLQo0cP5OTkYNKk\nSexydXV1NG/eHAzDwNTUtMGfQwghkk5GmqPlV4biMCGEfHmkOQ5L5HtaraysICMjg+DgYL5llZWV\n7IPdxsbGaNasGd9wn4sXL/KtJ+iKcmJiIjtjIRf3ymzNYFsbNTU1GBsbIzExEYmJiVBSUmKHJHE/\nt1evXkhISMB///3Ht371Z2sIIeRLJiMj2h+RXBSHCSHkyyNqHP6SYrHE3mn18PBAYGAgMjMzYWlp\nCSUlJbx58wbJyclwc3PDgAEDoKGhgaFDhyI8PBybNm1Cjx498OLFC9y+fRtqamo847otLS0RHByM\nvXv3gsPh4OXLl4iPj0fbtm15PrtNmzZQVVXFlStX0Lx5cygqKsLAwKDOK802Njbw9/fHu3fv0LNn\nTygoKPAsHz9+PB4+fAhfX184Ojqiffv2KCgowLNnz5CWloZff/1VvAeQEEKawJcU/EjdKA4TQsiX\nR5rjsER2WgHAzc0N2traiIyMxJkzZyAjI4PWrVujV69ePMN7vLy8oKioiKtXr+LevXvgcDhYvnw5\nfv75Z56ZCkeOHImSkhLcuHEDCQkJ6NSpE3x9fXHyJO/7PZs1awYfHx8EBATg4MGDqKiowNy5c+sM\nllZWVjh48CA+fPgAGxsbvuWamprw8/PDmTNnkJiYiLy8PKirq6NDhw7w8OB/nxUhhHyJpHkCiK8R\nxWFCCPmySHMclmE+ZfYDCVVUVIQpU6bA09MTbm5uTV0csaPXu/CiV97w+xpeeSMKQa+8+do1xitv\nHnQS7b16xg18zQmRfNIeh+n1LrzolTf8voZX3ois5itviNhfeSNqHAa+nFgssXdahVVaWso3DCgi\nourl1t27d2+KIhFCyFepsa/wlpWV4fTp04iLi0NhYSH09fXh6ekJExMTodaPj49HREQE0tPTIScn\nhw4dOmDChAno0qVLo5Zb2lEcJoQQySDNcfiL77TGx8cjLi4O5ubmUFRUxOPHj9kZBLt27drUxSOE\nkK9GY49K+vXXX5GcnAxnZ2doa2sjJiYG69evx6pVq8DhcOpcNzAwEGFhYejXrx8cHR3x8eNHPH/+\nHHl5eY1b6K8AxWFCCJEM0hyHv/hOa8eOHREfH4+wsDAUFxdDQ0MDTk5O8PT0bOqiEULIV0W2EaNl\namoqEhISMGnSJDg7OwMA7OzssGjRIgQEBGDVqlW1rvvPP//g3LlzWLRoEaysrBqtjF8risOEECIZ\npDkOf/Gd1s6dO+Pnn39u6mIQQshXrzGv8CYlJaFZs2YYMGAAm6agoAAHBwcEBgYiNzcXmpqaAteN\njIxE586dYWVlhcrKSpSWlrKvVSGfjuIwIYRIBmmOw198p5UQQsiXKysriy9NRUUFKioqPGlpaWnQ\n0dHhC3JdunQBwzB4/vx5rcHywYMH6NOnD06ePImLFy+ipKQEbdq0gaenJ/r16ye+nSGEEEK+QMLE\n4qaOw9RpJYQQIhYNmQDCx8eHL23UqFEYM2YMT1peXh5atmzJl5cbIHNzcwVuv7CwEAUFBUhISICs\nrCzGjx8PVVVVXLp0Cbt374aioiINGSaEECIVGjoRkzCxuKnjMHVaCSGEiIWMrOjr7Nmzhy+t5l1W\noGqGWjk5/pDFfQ9oaWmpwO2XlJQAAAoKCrBu3ToYGBgAqHqv57x58xAcHEydVkIIIVKhIXEYEC4W\nN3Ucpk4rIYQQsWjIFV4tLS2h8ikoKKC8vJwvvaysjF1e23rcz+EGSqAqyPbu3RtRUVEoKSmhZ1wJ\nIYR88Rp6p1WYWNzUcbiB/XFCCCGEl4yMaH+i0NDQEDj0iJtW23M0qqqqkJeXR4sW/C9w19DQAMMw\nKC4uFq0whBBCiAQSNQ6LEoubOg5Tp5UQQohYyMjIiPQnCn19fWRkZLDDjLiePn0KoOq1K4LIyspC\nX18f796941uWk5MDWVlZqKqqilQWQgghRBKJGodFicVNHYep00oIIUQsGvNOq7W1NSoqKnD16lU2\nraysDDExMTA0NGQnh3j79i0yMjL41s3JycG9e/fYtOLiYiQmJsLQ0LDWIU2EEELIl6Qx77Q2dRym\nZ1oJIYSIRWO+1NzAwADW1tY4ceIEcnNzoa2tjdjYWGRnZ2Pu3Llsvj179uDRo0cICgpi0wYNGoTo\n6Ghs3boVzs7OUFFRQXR0NIqKijB27NhGKzMhhBDyOUlzHKZOKyGEELFozJeaA1VT8p8+fRrXr19H\nUVEROnTogGXLluH/2Lvz8KjKu//jn8kymSyQhZDEEJMAAcKSAEFDAEEFVEREBaSigFKX+ih2kW74\ne+wDtkUebZVarFat1hasILaWfTcsIUGURdn3XUhCEkIC2ef3B808hCzM4Exm5uT9uq65rnLmzMl3\n4nQ++d7nPvdJTk5u8nUBAQH6n//5H/3973/XsmXLVFFRoQ4dOujFF1+85msBAPAWRs5hk9VqtTrj\nTaD5TPUPc3cJHuXlyiJ+J1eZkbfX3SV4FFNYjKxFZ9xdhkcxhcU4/ZgnejrWAN64g88pvNPTplbu\nLsGjvG29wO/kKm+XnnR3CZ4nKFS6eN7dVXiWoPqLE30Xjuaw5D1ZzJlWAIBTuHqEFwAANM7IOUzT\nCgBwCiOHJQAAns7IOUzTCgBwCpOPgdMSAAAPZ+QcpmkFADiFkUd4AQDwdEbOYe7TCgAAAADwWJxp\nBQA4hSvvDwcAAJpm5BymaQUAOIWBsxIAAI9n5BymaQUAOIXJyGkJAICHM3IO07QCAJzCwFkJAIDH\nM3IO07QCAJzCyCO8AAB4OiPnME0rAMApDJyVAAB4PCPnME0rAMApjDzCCwCApzNyDtO0AgCcwsSd\nvwEAcBsj53CTTeuUKVPs7th/97vfOaUgAIB3MvIIrzuRxQAAexg5h5tsWvv06dNcdQAAvJ2PccPS\nnchiAIBdDJzDTTatDz/8cHPVAQDwdgYe4XUnshgAYBcD57DDM59Pnz6trVu3qqKiQpJktVqdXhQA\nwPuYTCaHHrh+ZDEA4GqO5rA3ZbHdCzGVlpZq1qxZ+vrrryVJb7zxhqKjo/WnP/1JrVq10sSJE11W\nJADACxh4WpKnIIsBAI0ycA7bfaZ1zpw5qqio0Ouvv66AgADb9n79+mnHjh0uKQ4A4EVMJscecBhZ\nDABolKM57EVZbHfTun37dk2YMEGxsbF1tsfGxio/P9/phQEAgLrIYgBAS2T39OALFy6odevW9baX\nl5c7tSAAgHcyGXhakqcgiwEAjTFyDtt9pjUhIcF2Dc2VNm7cqI4dOzq1KACAFzLolCRPQhYDABpl\n4OnBdp9pHT16tF5//XUVFRWppqZG2dnZOnnypLKysvTf//3frqwRAOAFjDzC6ynIYgBAY4ycw3af\naU1LS9Pzzz+vbdu2qbq6Wh9//LFOnTqlX/7yl+revbsrawQAeAODju56ErIYANAozrRe1rt3b/Xu\n3VvS5XvCedO9fQAALmbgEV5PQhYDABpk4Bx2qGmVpP379+vEiROSpBtvvFGdO3d2elEAAO9D89R8\nyGIAwNWMnMN2N60FBQV6/fXXtX//flksFklSWVmZOnfurB//+Mdq06aNy4oEAHgBA4/wegqyGADQ\nKAPnsN1N65///GdVVFTo1VdfVXx8vCTp+PHjeuutt/TOO+9o6tSpLisSAOAFDDzC6ynIYgBAowyc\nw3YvxLRr1y49+eSTtpCUpPj4eD3++OPatWuXS4oDAHgPk49jDziOLAYANMbRHPamLLb7TGtoaKj8\n/Orv7uvr2+CNzgEALYyBR3g9BVkMAGiUgXPY7v76wQcf1IcffqiCggLbtoKCAs2ZM0djx451SXEA\nAO9h8jE59IDjyGIAQGMczWFvyuImz7ROmTKlzipUubm5evbZZxURESHpclD6+/vr/Pnzuu2221xa\nKADAwxl4hNedyGIAgF0MnMNNNq19+vRprjoAAN7Oi0ZsvQlZDACwi4FzuMmm9eGHH26uOgAAQAPI\nYgBAS2f3QkwAADTFyDc1BwDA0xk5h+1uWmtqarRw4UJlZWUpLy9PVVVVdZ6fM2eO04sDAHgRA09L\n8hRkMQCgUQbOYbtXD16wYIGWLFmiAQMGqLKyUiNHjlRGRobMZrPGjRvnyhoBAN7AZHLsAYeRxQCA\nRjmaw16UxXafad2wYYOeeuop3XzzzfrnP/+pQYMGKSYmRosWLdLRo0ddWCIAwBsYeVqSpyCLAQCN\nMXIO232mtbCwUO3bt5ckBQQE6NKlS5Kkm2++WV9++aVrqgMAeA8fk2MPOIwsBgA0ytEc9qIstrtp\nDQ8PV3FxsSQpMjJS+/btkySdPn3a0F09AMA+JpPJoQccRxYDABrjaA57U27YPT24e/fu+vLLL9Wh\nQwfddttt+tvf/qbNmzfr8OHD6tevnytrBAB4Ay8asfVWZDEAoFEGzmG7m9Ynn3xSNTU1kqS77rpL\ngYGB2rNnj9LS0nT33Xe7rEAAgJdw8YhtZWWl5s2bpw0bNqikpESJiYl66KGHlJKS0uTr5s+frwUL\nFtTbHhoaqnfffddV5boEWQwAaJSBc9juptXX11e+vr62fw8aNEiDBg2y9+UAAIMzuXiE980339Tm\nzZt1zz33KCYmRpmZmZoxY4amT5+uzp07X/P1Tz31lMxms+3fV/5vb0EWAwAaY+QcbrJpLSkpsftA\nISEhdu8LADAgF47wHjx4UJs2bdKjjz6qe+65R9Llhm3KlCmaO3eupk+ffs1j9OvXT8HBwS6r0VXI\nYgCAXQycw002rY8//rjdB5o3b951FQAAMAgXjvDm5OTI19dXQ4YMsW0zm80aPHiwPv74YxUWFio8\nPLzJY1itVl28eFGBgYFetfgEWQwAsIuBc7jJpnXq1KkOHQwA0HJdTyOYm5tbb1twcHC9kdgjR46o\nXbt2slgsdbYnJSXJarXq6NGj1wzLZ555RmVlZQoMDFTfvn01ceJErzgzSRYDAOxxvQOy9mSxu3O4\nyaa1V69edh0Ezevl88fcXYLH4XdS19PBce4uwaO8bb2g/wrv5O4yPMrb1gvOP+h1jPBOnjy53rYx\nY8Zo7NixdbYVFRUpIiKi3r61AVlYWNjozwgJCdGwYcPUuXNn+fn5aefOnVq1apWOHDmiGTNmyN/f\n3+G6mxNZ7HneLj3p7hI8Dr+Tusjh+t62XuD3chWnZ/F1nmm1J4vdncN2L8QEAICzzZ49u962hq53\nqaiokJ9f/ciqDbqKiopGf8bw4cPr/DsjI0M33nij/vKXv2jdunUaOnSoo2UDAGAY9mSxu3PY55p7\nAABgD5PJsYekqKioeo+Gmlaz2ayqqqp62ysrK23PO+KOO+5QQECAdu7ceR1vFAAAD+RoDjuQxe7O\nYc60AgCcw4WLG4WFhTU49ah227Wuo7maj4+PIiIiHFqZFwAAj2bgHOZMKwDAOa5jdNdeiYmJOnXq\nlMrKyupsP3DggCQpISHBoeNVVVUpPz9frVu3duh1AAB4rOs802oPd+cwTSsAwDl8fBx7OCAjI0PV\n1dVas2aNbVtlZaUyMzPVpUsX2+IQ+fn5OnXqVJ3XFhcX1zvewoULVVlZqZ49e17HGwUAwAM5msMO\nZLG7c9ih6cEbN27UsmXLdPbsWc2YMUNRUVFavHixYmJidNNNNzlyKACA0bhwWlKnTp2UkZGhOXPm\nqLCwUDExMVq3bp3y8vL0zDPP2PabPXu2du/erfnz59u2PfPMM+rfv7/i4+Pl5+enXbt2afPmzUpO\nTtYtt9zisppdhSwGADTIwDlsd9OamZmpDz74QMOGDdOxY8dktVolSUFBQVq6dClBCQAtnQvDUrq8\nJP+8efO0fv16lZaWKj4+XlOnTlVycnKTr7vlllu0b98+5eTkqKqqSm3bttXo0aP1wAMPyNfX16U1\nOxtZDABolIFz2O6mdfHixXrqqac0YMAALVu2zLa9Q4cO+sc//mHvYQAARuXisDSbzZowYYImTJjQ\n6D7Tpk2rt+3pp592YVXNiywGADTKwDlsd9N65swZde7cud72wMBAXbx48TsXAgDwcg5epwrHkcUA\ngEYZOIftfmehoaHKzc2tt/3AgQNq27atU4sCAHghF64ejMvIYgBAo1y4erC72d203nLLLfr73/+u\nvLw8mUwmVVdXa+fOnZo7d65uvfVWV9YIAPAGBg1KT0IWAwAaZeCm1e7pwQ8++KDOnj2ryZMnS5Ke\nf/55Wa1WDRgwQPfdd5/LCgQAeAkvCj9vRRYDABpl4By2u2n18/PTj3/8Yz344IM6dOiQrFarOnbs\nqLi4OFfWBwDwFga+lsZTkMUAgEYZOIcduk+rJLVr107t2rVzRS0AAG9m4BFeT0MWAwDqMXAO2920\nvvfee00+/8QTT3znYgAAQOPIYgBAS2R303r8+PE6/66urtbp06clSfHx8c6tCgDgfQw8wuspyGIA\nQKMMnMN2N60vvfRSvW3l5eV688031bt3b6cWBQDwQgYOS09BFgMAGmXgHP5OV+sGBARozJgxWrBg\ngbPqAQB4KZOPj0MPOAdZDACQHM9hb8pihxdiulpFRYVKSkqcUQsAwJsZeITX05HFAAAj57DdTevm\nzZvr/NtqtaqwsFDLly9XcnKy0wsDAHgZA4elpyCLAQCNMnAO2920vvbaa/W2BQUFKSUlRY899pgz\nawIAeCMDh6WnIIsBAI0ycA7b3bTOmTOnzr9NJpP8/L7z7GIAgFF40bUx3oosBgA0ysA5bNc7q6qq\n0owZM5SXlyd/f3/5+/sTkgCAukwmxx5wCFkMAGiSoznsRVlsV9r5+fnp2LFj8jFw9w4A+I68KPy8\nEVkMAGiSgXPY7uTLyMhQTk6OK2sBAHgzg47uehKyGADQqJZ+plWSIiIi9Nlnn2nfvn3q2LGjLBZL\nnedHjBjh9OIAAF6EM4AuRxYDABpl4By2u2lduXKlzGazDh06pEOHDtV5zmQyEZQA0NJ50YittyKL\nAQCNMnAO2920vvPOO66sAwDg7Qwclp6CLAYANMrAOXzNc8iTJ0/WhQsXmqMWAIA38/Fx7AG7kcUA\ngGtyNIe9KIuvWWleXp5qamqaoxYAANAAshgA0JJxgzcAgHMYeFoSAAAez8A5bFfTWlRUpOrq6ib3\niYiIcEpBAAAvZeCw9ARkMQCgSQbOYbua1p///OfX3GfevHnfuRgAgBczcFh6ArIYANAkA+ewXU3r\nlClTFBIS4upaAADezIsWdPBGZDEAoEkGzmG7mtYuXbooNDTU1bUAALyZgUd4PQFZDABokoFzmIWY\nAADOYeCwBADA4xk4h2laAQDOYTLutCQAADyegXP4mk0rizoAAOziY9wRXncjiwEA12TgHOZMKwDA\nOQw8wgsAgMczcA7TtAIAnMPA19IAAODxDJzDNK0AAOcw8FL7AAB4PAPnME0rAMA5DDzCCwCAxzNw\nDtO0AgCcw8DX0gAA4PEMnMPGfWcAAAAAAK/HmVYAgHMYeFoSAAAez8A5TNMKAHAOAy8AAQCAxzNw\nDtO0AgCcw8AjvAAAeDwD5zBNKwDAOQy8AAQAAB7PwDlM0woAcA4f147wVlZWat68edqwYYNKSkqU\nmJiohx56SCkpKQ4d5+WXX9a2bds0fPhwPfbYY64pFgCA5mbgHDZuOw4AaF4mH8ceDnrzzTe1ZMkS\nDRw4UJMmTZLJZNKMGTO0f/9+u4+xdetW7d692+GfDQCAx3M0hx3MYnfmME0rAMA5TCbHHg44ePCg\nNm3apPHjx2v8+PEaOnSofvWrXykyMlJz58616xhVVVX68MMPdd99913PuwMAwLM5msMOZLG7c5im\nFQDgHC4c3c3JyZGvr6+GDBli22Y2mzV48GDt3btXhYWF1zzG0qVLVVFRoXvvvdfhtwYAgMdz4ZlW\nd+cw17QCAJzjOq6lyc3NrbctODhYwcHBdbYdOXJE7dq1k8ViqbM9KSlJVqtVR48eVXh4eKM/p6io\nSJ9++qkef/xxBQQEOFwnAAAez4XXtLo7h2laAQDOcR1L7U+ePLnetjFjxmjs2LF1thUVFSkiIqLe\nvrUBea0R3o8++kixsbEaOHCgwzUCAOAVrvOWN/YMILs7h2laAQDOcR2LK82ePbvetqvPskpSRUWF\n/PzqR5a/v7/t+cYcPHhQ69at0/Tp02Uy8D3sAAAt3HXe8saeAWR35zBNKwDAOa5jWlJUVJRd+5nN\nZlVVVdXbXllZaXu+IVarVR988IH69u2r5ORkh+sDAMBrXOf0YHsGkN2dwzStAADncOFNzcPCwhqc\nelS7rbHraL744gsdPHhQ48aNqzf96eLFi8rNzVVYWFijYQsAgNe4zhy2ZwDZ3TlM0woA8HiJiYla\nunSpysrK6iwCceDAAUlSQkJCg6/Lz8+X1WrVSy+9VO+5zMxMZWZm6oUXXlCvXr1cUzgAAAbg7hym\naQUAOIcLrxfNyMjQokWLtGbNGt1zzz2SLk9JyszMVJcuXWyLQ+Tn56u8vFzt2rWTJPXp00dt27at\nd7zf/e53SktL0+DBg9W+fXuX1Q0AQLMxcA7TtAIAnMOF04M7deqkjIwMzZkzR4WFhYqJidG6deuU\nl5enZ555xrbf7NmztXv3bs2fP1+SFBMTo5iYmAaPGRMTo/T0dJfVDABAszJwDrvunQHXqaCgUA88\n9IiC28YqIbmHPpr3SYP7vfr6G+pxUz+1atVK7bul6tXX36jzfGLXFAW2iVFIVDuFRLXTnfc+0Bzl\no5kEhYfr6X9+pD+UnNFvj+7SzeMebHC/yUs/lSTNuvCtZl34VrPLz+nFr3Nsz//2yE69cTHX9vwP\nV3zWLPUbko/JsYeDJk+erOHDh2v9+vX64IMPVFVVpalTp7LAEuACDmVxjx5qFR1HFrcwjuTwrAvf\nShI57GqO5rCDWezOHOZMKzzOsz/5qcxms84e2a/tX3+je0Z/Tz1Teqh7t6519rNarfrbu28pte8A\nHdq5XXeOfEA3xrXTQw+Otu2z6JOPNXTwbc37BtAsxr35e1VVVOjn0R0V1ytVk5d8opM7vtG3u/fW\n2W/28NF623pBP251gyTp+c+Xau/adXX2+dO9Y7V3TWZzlW5cLhzhlS6vTDhhwgRNmDCh0X2mTZtm\n17FqR4ABNMyhLP7b35SalKBDh4+QxS2IIzksyZbF5LALGTiHW/SZ1szMTI0dO7bBG+p+l31x/UpL\nS/Xpvxfq1y/+P4WEhOiW/v00cvgw/f0f8+rt+/Pnf6S03r3k5+enLp076b4Rw5WVk9PAUWE05qAg\n9R59nxa++BuVl5bqUFa2dixcqr4TxjX5ujYJ8Uoa2F+b//aPZqq0hTGZHHsAIos9kcNZnJZGFrcw\n5LCHcjSHvSiLW3TT2pCVK1cqMzPT3WW0WPsPHJSfn586d0qybeuZkqJde/Y0+Tqr1aoNWdnq3rXu\nCPAj339SbRM66s57H9COr79xSc1oftGdk1RTVaXcAwdt207t2KnY7l2beJXUd+I4HdywSeeOHa+z\n/ftz/6JXc4/ohys+U7vUHi6puUXw8XHsATSCLHYvshjXQg57KEdz2Iuy2HsqdYFBgwZpzpw5dVa0\nWrVqVYNB2dC+cL6S0lK1btWqzrbQ0Na6UFLS5Oum/fZl1dTUaNKER2zb5r7/ro7u+VrH9nyj2wcN\n1F33jVZRUZFL6kbzCggJ0aXiC3W2XTpfLEurkCZflzFxnLL/OrfOtvcfeUL/L7G7Xkjopn2fb9AP\nV3ymwNBQp9fcIhh0dBeuRRZ7HrIY10IOeyjOtBqTj4+PzGazTHb8B3NkX1y/kOBgFV+o+yVYXFys\nViGNfwnOnj1bf/voYy3553wFBATYtg/ol6HAwEAFBQVp6s+eV1hYqDZsynZZ7Wg+5SUlCmxd9w8q\nS+tWKrvQ+B9UHQf0U+uYaG1dUHeBh0ObclRZVqbKS5e0YubvdanovJIG9ndJ3YZn8nHsAYgs9kTX\nlcVvv0MWtyDXk8OSyGFXczSHvSiLPWohpvnz52vBggWaNWuW/vGPf2jHjh0ym8267bbb9NBDD8nP\n73K51dXV+uc//6l169apoKBAbdq00a233qpRo0bJ54rT3F9//bU++eQTnThxQtXV1YqIiFDfvn31\n8MMPS7p8bcyf/vQnzZ49W1FRUXr22WeVl5cnSRo7dqwkqVu3bpo2bVq9fV9++WWdOXNGf/jDH+q9\nj5/85CeKiIjQiy++KEmqqanR4sWL9fnnn+vs2bMKCQlRRkaGHn744To354XUuVOSqqqqdODgIXVK\n6ihJ2vHNznpTjWq9/+HfNXPmTK1fsURx/7kfVGNMJpOsVqvTa0bzO7v/oHz8/BSV1FG5Bw9JkuJ6\n9tDpXY1PXev36MPa/s+FKi8tbfLYVquVP4ivF783QyCL4XAWv/++Zv5+ltavXEoWtxDXk8OSyGFX\nM/DvzSPb69dee001NTV6+OGH1aNHDy1cuFB//etfbc+//fbb+uSTT5SUlKSJEyeqY8eOmj9/vt55\n5x3bPidOnNA5vRHaAAAgAElEQVTMmTPl5+enhx56SBMmTFCfPn20b9++Rn/uo48+qsjISLVr106T\nJ0/W5MmTNWrUqAb37devn7799lsdOXKkzvbjx4/r1KlT6tevX516P/74Y3Xr1k2TJk3SLbfcojVr\n1uiVV17hi/sqwcHBGnXfvfrVr2eotLRUWdk5+veSZZow7nv19p378Xy9MO3XWrVqlTq0T6zz3PET\nJ5SVnaOKigqVlZXp1dffUP65cxqQkdE8bwQuVXHxorb9c6Hufen/yRwUpI79M9Tzvnu0+e+NL+zQ\nZ+wD9aYkhd8Yp479M+Tr7y+/gADd8dMfKSSyjQ5lsYjIdTHodTQtFVnccjmcxS+8oFWL/kUWtyCO\n5rD/fwaGyGEXM/A1rR51prVWTEyMfvrTn0qShg0bJrPZrFWrVmnkyJG6ePGi1q1bpzvuuENPPvmk\nbZ+goCCtXr1ad999txISEvT111/L399fL774Yp0R36akp6frk08+UXBwsAYNGnTNfd955x1lZ2er\nffv2tu2bNm2Sr6+v+vbtK0nas2ePMjMz9fzzzyvjii/ppKQkzZo1Szt27FCvXr0c+v0oyNjz/P/0\n53f1/e9/X1GJndSmTRu99dZb6n5ThjZs2KC7775bJf+5pua/fz1D5woKdPPNN9teO378eL399tu6\nUHVS//WTn+nQoUOyWCzq1auXli1brjY3tm/sxxrK29YL197JIK68L9z/7NrS6H6BoaH6ydoldh3z\n9+eOfee6WiQDj/C2RGRxEwyew5KDWXzunG4eNNj2WrKYHG4MOexiBs5hj2xahw0bVu/fmZmZ2r59\nu0r/M6VgxIgRdfYZMWKEVq9erW3btikhIUHBwcEqLy/X9u3blZaW5vQag4KC1LNnT2VnZ9umOElS\nTk6OUlJS1Oo/Cxjk5OQoJCRE3bp1U3FxsW2/rl27ysfHR7t27XK8ab143invwVNFWHz12Ucf1t14\n8bwG9klVSe4p2/s/smv75eeCQuv+Ti6eV/f2cfo6Z0P9gxv8d1fr6eA4d5fgUd62XtDTplbX3rEF\ncckfVF50bQyujSxuQgvIEoey+Ooc/s++LTmLyeH6yOL6nJ7FBs5hj2xaY2JiGvx3Xl6eSktL5evr\nq+jo6Hr7+Pr62q6D6d+/v9asWaOZM2cqPDxcKSkpSk9P18033+y0efL9+/fXH//4Rx06dEgdO3bU\n0aNHdfr0ad133322fc6cOaOSkhI98cQTDR7jyvAEAK9m4BHelogsBgAvY+Ac9sim1RnMZrOmT5+u\nXbt2adu2bdq+fbvWr1+v1NRUvfDCC3ZPU2rKTTfdJH9/f2VnZ6tjx47KycmRn5+f0tPTbfvU1NQo\nPDxczz77bIPHiIiI+M51AIBHMPAIL64PWQwAzcjAOeyRTeuZM2cUGRlZ59+S1LZtWwUFBam6ulpn\nz57VDTfcUGef6urqOvdu8/HxUUpKilJSUjRx4kT9+9//1ty5c7Vr1y6lpKQ0+LMdGfkNDAxUr169\nlJ2drfHjxys7O1upqakKDg627RMdHa1du3apa9eu8vf3t/vYAAC4E1kMAPAUHtmOL1++vN6/TSaT\nevbsqd69e0uSliypeyH30qVLJcl2zcyFC/XniCcmJkqSKisrG/3ZAQEBtmt17NG/f3/l5eVp1apV\n+vbbb9W/f937SmVkZKiqqkqfffZZvddWVlbq4sWLdv8sAPBoPibHHvBoZDEAeBlHc9iLsthjz7S+\n+uqrSk1N1Z49e7Rp0yYNHTrUdu3MrbfeqpUrV6q0tFTJycnau3evsrKyNHjwYMXHx0uSPv30U+3Z\ns0e9e/dW27ZtVVxcrJUrV6pNmzZKTk5u9Gd36NBBy5cv16effqqYmBiFhoaqR48eje7fp08fmc1m\nzZkzR/7+/nVWspWkHj16aPDgwfrkk090+PBhpaSkyGQy6dtvv1V2draee+45paamOuG3BgBuZuBp\nSS0RWQwAXsbAOeyRTevzzz+vjz76SHPnzpW/v79GjBhRZ1XAp59+WlFRUcrMzFROTo4iIiI0duzY\nOvdxu+mmm5Sbm6vPP/9cFy5cUKtWrdStWzeNHTtWQUFBjf7sUaNGKTc3VwsXLtSlS5fUrVu3JoPS\nYrEoLS1NOTk5uvnmmxUYGFhvnx/84Afq0KGD1qxZY3tPUVFRGjJkiG3EGQC8noEXgGiJyGIA8DIG\nzmGT1YPuqD1//nwtWLBAH3zwQZ1rUXCVFrBUvEMaWmq/hWOp/bpYZr8+V9zypjrrnw7t7ztg1LV3\nQrMji+1A5tRFDtdDDtdHFtfn7Cx2NIcl78lijzzTCgDwPs66hQkAAHCckXOYphUA4BwGvpYGAACP\nZ+AcpmkFADiHgcMSAACPZ+Ac9qimdezYsRo7dqy7ywAAXA8vWjofjSOLAcBLGTiHPappBQB4MQOP\n8AIA4PEMnMM0rQAA5zDwAhAAAHg8A+cwTSsAwDkMPMILAIDHM3AO07QCAJzDwCO8AAB4PAPnME0r\nAMA5DDzCCwCAxzNwDhv3nQEAAAAAvB5nWgEAzmHgpfYBAPB4Bs5hmlYAgHMYeFoSAAAez8A5TNMK\nAHAOAy8AAQCAxzNwDtO0AgCcw8AjvAAAeDwD5zBNKwDAOQw8wgsAgMczcA7TtAIAnMPAI7wAAHg8\nA+cwTSsAwDl8jBuWAAB4PAPnME0rAMApTAaelgQAgKczcg7TtAIAnMPA05IAAPB4Bs5hmlYAgHMY\neIQXAACPZ+AcpmkFADiHgUd4AQDweAbOYZpWAIBzGHiEFwAAj2fgHKZpBQA4h4FXLQQAwOMZOIeN\n+84AAAAAAF6PM60AAOcw8LQkAAA8noFzmKYVAOAcBl4AAgAAj2fgHKZpBQA4h4tHeCsrKzVv3jxt\n2LBBJSUlSkxM1EMPPaSUlJQmX7dhwwatXbtWJ0+eVGlpqcLDw9W9e3c9+OCDatu2rUtrBgCg2Rg4\nh43bjgMAmpnJwYdj3nzzTS1ZskQDBw7UpEmTZDKZNGPGDO3fv7/J1x07dkwRERG699579cQTT+jW\nW2/V9u3bNXXqVBUVFTlcBwAAnsnRHHYsi92Zw5xpBQA4hwtHeA8ePKhNmzbp0Ucf1T333CNJGjRo\nkKZMmaK5c+dq+vTpjb52/Pjx9bbddNNN+uUvf6n169dr5MiRLqsbAIBmY+Ac5kwrAMA5TCbHHg7I\nycmRr6+vhgwZYttmNps1ePBg7d27V4WFhQ4dr3Y60sWLFx16HQAAHsvRHHYgi92dw5xpBQA4ieMj\nvLm5ufW2BQcHKzg4uM62I0eOqF27drJYLHW2JyUlyWq16ujRowoPD2/yZ5WUlKi6ulr5+flasGCB\nJKlHjx4O1wwAgGe6vjOt9mSxu3OYphUA4BzXMS1p8uTJ9baNGTNGY8eOrbOtqKhIERER9fatDUh7\nRnh/9KMf6cKFC5KkVq1a6fvf/z5NKwDAOK5zerA9WezuHKZpBQA4x3Vk5ezZs+ttu/osqyRVVFTI\nz69+ZPn7+9uev5YpU6aovLxcp06d0oYNG1RWVuZ4wQAAeKrrvKTVnix2dw7TtAIAnMTxtIyKirJr\nP7PZrKqqqnrbKysrbc9fS7du3SRJvXv3Vnp6uqZMmSKLxaJhw4Y5UDEAAJ7q+rpWe7LY3TnMQkwA\nAOdw4UJMYWFhDU49qt12retorhYVFaUOHTpow4YNDr0OAACP5cKFmNydwzStAADncGHTmpiYqFOn\nTtWbSnTgwAFJUkJCgsPlVlRUsHowAMA4XNi0ujuHaVoBAE7iuhuaZ2RkqLq6WmvWrLFtq6ysVGZm\nprp06WJbHCI/P1+nTp2q89ri4uJ6xzt8+LCOHj2qDh06OFQHAACey9Ectj+L3Z3DXNMKAHAOF97U\nvFOnTsrIyNCcOXNUWFiomJgYrVu3Tnl5eXrmmWds+82ePVu7d+/W/PnzbdueeeYZ9evXT/Hx8bJY\nLDpx4oQ+//xzWSwWjR492mU1AwDQrAycwzStAACvMHnyZM2bN0/r169XaWmp4uPjNXXqVCUnJzf5\nurvuukvffPONtmzZovLycoWHh6tfv34aM2aM3QtBAQDQ0rkzh01Wq9XqjDeBZnTxvLsr8CxBofxO\nrvJ0cJy7S/Aob1sv6GlTK3eX4VHetl5w+jGtZ484tL8pur3TawCaBZlTFzlcDzlcH1lcn7Oz2NEc\nlrwniznTCgBwDhdOSwIAANdg4BymaQUAOIeBwxIAAI9n4BymaQUAOIlxwxIAAM9n3BymaQUAOIXJ\nwCO8AAB4OiPnME0rAMA5DByWAAB4PAPnME0rAMBJjBuWAAB4PuPmME0rAMA5DDzCCwCAxzNwDtO0\nAgCcw8BhCQCAxzNwDtO0AgCcxLhhCQCA5zNuDtO0AgCcw8AjvAAAeDwD5zBNKwDAOYyblQAAeD4D\n5zBNKwDASQyclgAAeDzj5rCPuwsAAAAAAKAxnGkFADiHga+lAQDA4xk4h2laAQDOYeCwBADA4xk4\nh2laAQBOYtywBADA8xk3h2laAQDOYeARXgAAPJ6Bc5imFQDgHAYOSwAAPJ6Bc5imFQDgJMYNSwAA\nPJ9xc5imFQDgHAYe4QUAwOMZOIdNVqvV6u4iAAAAAABoiI+7CwAAAAAAoDE0rQAAAAAAj0XTCgAA\nAADwWDStAAAAAACPRdMKAAAAAPBYNK0AAAAAAI9F0woAAAAA8Fg0rQAAAAAAj0XTCgAAAADwWDSt\nAAAAAACPRdMKAAAAAPBYNK0AAAAAAI9F0woAAAAA8Fg0rQAAAAAAj0XTCkiyWq3uLgEeqKamxt0l\nAECLQRajIWQxJMnP3QUA7lZdXS1fX19JUllZmaxWq3x9fWU2m2W1WmUymdxcIdzhys/FunXrlJ+f\nr/LycvXv318xMTGyWCx8PgDASchiNIQsRi2TlWEttGA1NTXy8bk84eDDDz/U4cOHVVRUpDZt2mjs\n2LFKTk52c4Vwhys/FzNnztSBAwckST4+PqqsrNTtt9+u4cOHq23btu4sEwAMgSxGQ8hiXMl32rRp\n09xdBOAutSNzL7/8srZu3arExETFxMTo3Llz+uyzzxQeHq74+HjblyZahtrPxTvvvKOdO3dq4sSJ\neuihhzRu3Dht375dX375pTp06KC4uDhGdwHgOyKL0RCyGFdiejBavKVLl+ro0aN66qmn1KdPH/n7\n+yszM1NvvfWWvv32W1VXV8vPj/+rtDQFBQXat2+fhgwZovT0dFksFu3fv19HjhzRgAEDlJycLB8f\nH6YlAYATkMVoCFmMWgxZocU7fvy42rZtq5SUFPn7+2v37t167733dOutt2rYsGEKCAiw7ctiAMZ1\n9X/bixcvKi8vT0lJSbJYLNq5c6deeukl9enTRw899JDatGkjSTp9+rSqq6vdUTIAGAZZDIksRuNo\nWtGiNPSFlpubq5CQEAUHB2v37t16+eWXlZ6ernHjxikyMlKStGTJEn377bdMTTKoK6+bKS0tlSSZ\nzWaZTCZduHBBp06d0syZM5Wenq4JEyYoPDxckpSZmak333xT58+fd1vtAOBtyGI0hCxGU5hngRal\ndgW6zZs3q3fv3jKbzUpOTtbmzZu1YsUKzZkzR+np6XrkkUdsX4YHDx7UkiVL5Ovrq5iYGKafGFBt\nSL7xxhvy9fXVU089paioKPXr108ffvihKioqdMstt+h73/ue7XORl5en3bt3y9/f352lA4DXIYvR\nELIYTWGoCi3O8uXL9dprr6mkpESS1LNnT509e1bvv/++unfvrkmTJikiIkKSVFhYqI0bNyogIEAd\nOnQgJA2uqKhI+/fvt50FSE9PV5s2bWQ2m9WrVy/bNKS8vDytXbtWW7Zs0e233277vAAA7EMWozFk\nMRrC6sFoccxms7Zs2aKCggKlpqYqOjpa0dHR2rx5s8LDwxUREaHY2FgdPHhQ69at07JlyzR69Gj1\n7dvX3aXDRWpqamQymZSQkKClS5eqoqJCqampuuGGG2SxWHTs2DGtXr1aeXl52rp1qz7//HNt2rRJ\no0aN0rBhwySJRSAAwAFkMa5GFqMpNK1ocUJCQnTkyBHt2bNHaWlpat26teLj4xUTE6O1a9dq06ZN\nWrFihdauXauTJ09q1KhRuueeeyTxZWgUtcEoXf5vWjslyd/fXydOnNCBAweUnJyssLAwJSYmKj4+\nXiEhIfr666915swZRUdHa+TIkbrrrrtsx+MaKwCwH1kMshiOMFmtVqu7iwCcraamRlar1XbdTK2q\nqir5+fkpNzdXv/jFLzRw4EB9//vftz1/9OhRHTx4UEePHlVCQoLatWunbt262Y7Jl6F32r9/v5KS\nkuoti19YWGi7LqbWrl279Jvf/EYTJkzQ8OHD6zxXUlIiPz8/+fj4yGw2S+JzAQCNIYtxJbIY3wVn\nWmEY2dnZWrlypdLS0mQymWxfXmvXrlVlZaXatGlTZxTv/Pnz2rx5szp16qTIyEhZrVaFh4erQ4cO\nSktLU8eOHdW2bVtJfBl6sxMnTmjq1Kk6ceKE+vfvbwvJP/3pT3r33XcVFBQki8Wi1q1bS5KioqJ0\n5swZrVu3zjb6L10eBQ4ICLAFpclkqjMyDAAgi9EwshjfFU0rDKGiokJz587Vpk2b1LVrV0VFRUmS\n1q9fr7feektbt27V+fPnFRkZqeDgYPn5+SkgIEBr165VaGioevTo0eR0I6YheS8fHx/5+/srMjJS\nXbt2tW0vKipSQECAFi5cqJ07d+r8+fNKTEyUv7+/goKCtG7dOkVERKhLly71/lCq/TzwuQCA/0MW\nozFkMb4rmlYYgq+vr3r06KEuXbooLS3Ntj0hIUFdunSRr6+vVq5cqS1btuj48ePq0KGDEhISVFFR\nocWLF6tv374KDQ114zuAqwQEBKhz587q3r27JGnZsmXq1KmTOnbsqL59+6pTp04qKCjQ+vXrlZ2d\nrQsXLmjQoEE6efKktm3bpqFDh9ab2gYAqI8sRmPIYnxXNK0whJqaGgUFBSkuLk7S5ekm27ZtU58+\nfRQdHa2UlBQNHDhQubm52rlzp5YtW6aSkhKFhISoqKhIJSUl6tq1K1+IBmO1WmW1WuXnd/mW1F98\n8YXeeustnTt3zjZ1LSYmRikpKerXr58OHTqkL774QqtXr1Z4eLj27t2rmpoapaSkuPmdAIDnI4vR\nELIYzkDTCq909fShK6eMlJWVafXq1Tpw4ICKioqUmpoqHx8fhYSEKC0tTWlpabp48aK2bNmivXv3\n6uzZsyotLdUtt9wii8XirrcEJ6qurlZNTY18fX1tn5O8vDwlJSUpMDBQS5cuVX5+vi0szWazwsLC\nNGDAALVv317l5eXKysqSJN1+++1KTEx047sBAM9EFqMpZDGciaYVXslkMqmyslLffvutfHx8bAG3\natUqJSQk6KabbtLx48e1fft2nT9/XqmpqZIuT10KDQ3VTTfdpPj4eIWGhmrPnj0aNmxYnalM8F7V\n1dXKyspSdna2unbtKh8fH/32t7/Vjh071LdvX8XHxys4OFiLFy+uM8pbG6wxMTG6+eab1b59e/Xv\n318DBgxw91sCAI9EFqMxZDGcjaYVXmv37t364IMP5OvrqxtvvFG//vWv9cUXX9imIXXt2rXBsKyq\nqpKPj4+io6PVvXt33XnnnbaQ5N5v3s/Hx0c5OTlavHixzp49q2XLlunYsWMaM2aMYmNjZbFY1K5d\nO4WEhNQJy9qzA7X3jYuNjVW7du3qbAMA1EUWoyFkMZyNphVeKzAwULt379by5cuVnZ2tgoICPfXU\nU0pKSpKvr68sFkuDYenj46Pq6mrbF2NAQIBtdI8l073Tpk2bVF5eroiICElSjx49lJ+fr6ysLBUW\nFurxxx9X//79bf99zWZzvbDs06ePTCZTg4FISAJAw8hi1CKL4Uo0rfBaAQEBSktL0+LFi1VcXKz+\n/ftr0KBBCg4Otl30HxgYWCcsi4uLlZKSwpLpBnL+/Hm98cYbWrRoke644w7bvds2btyokydP2q6l\n6dWrl/z8/Gx/JF0ZlsuWLdPZs2dtYQkAsA9ZDIkshuvRtMKrHTp0SPv371dERIS++eYbRUZGKjo6\n2jZiW11draCgIHXt2lUnT57U5s2blZ+fr969e7u7dDhJQECAbrzxRnXt2tV23YwkhYeHq2/fvqqp\nqdHWrVt1+vRpde/eXRaLpU5YxsXFyWKxaPHixeratauio6Pd/I4AwLuQxSCL4Wo0rfAqV1/n0rp1\na/Xr10/9+/fX2bNntXz5coWGhio2NlZms9n2pVk7PenQoUPq3r27OnTo4K63ACeq/TxER0erffv2\nki7fYiE0NFRdunRRTEyM0tPTdfr0aX311Vc6c+aMunXrZlsspKysTJKUkpKiPn36KDk52W3vBQC8\nBVmMK5HFaA40rfAaV1/nUlNTIz8/P1ksFgUGBqpbt246ffq0Vq1aVScsS0tLtW/fPkVGRmrgwIFK\nSkpy47uAM109fejkyZNasGCBsrOz1b17d4WHh0uS+vTpozNnzuirr77S2bNn1atXL0nSihUr9Mkn\nn6hPnz6KiYmRxEIPANAUshhXI4vRHGha4RWuDMkFCxZo6dKlWr58uaqqqhQUFKRWrVrJYrGoR48e\nOn36tFauXKnWrVtLkjZv3qy33npL3bt3V2xsrCRWJjSKq0MtJCRESUlJ2r9/v1atWqUePXooPDxc\nJpNJffr00dmzZ/Xll18qJydH+/fv15IlSzRkyBBbcEpcTwUAjSGL0RCyGM2BphUe78qQnDlzprKz\ns+Xr6ytJysnJ0YkTJxQXF6fw8HAFBASoR48eOnv2rBYvXqwvv/xS27Zt0wMPPKDBgwfbjsmXofe7\n8nOxefNmlZWVqU2bNoqMjFRcXJz27dun1atX1wnLtLQ0lZeXKzc3V3l5eRo9erRGjhwpiT+eAKAp\nZDEaQhajuZisVqvV3UUA9vjLX/6irVu36uGHH1bPnj0VEhKi9957T59//rmSkpL06KOP1rk+ZunS\npSouLlZCQoL69esnqf60JninK/87zpo1SwcOHFBMTIx++tOfKjAwUFarVfv27dNf//pXFRYW6pe/\n/KXtOhur1aqysjKVlZXZpizxuQAA+5DFqEUWoznRtMIrHD16VK+//rpuv/123XnnnQoKCtLevXv1\n29/+Vu3atdO5c+cUGxurSZMmKTEx0fa6K78A+TI0nldffVWHDh3SqFGj1KdPH7Vp08b2XE1NjS0s\ni4qKNHXq1DqfjVqM6gKAfchiNIQsRnNgejA8UkNfXidOnNDgwYMVGRmp48ePa9q0aUpPT9fPfvYz\nlZeXa+PGjTp9+rTi4uJsN7a+8hh8GRrLmjVr9Pnnn+vRRx9VRkaGWrdubbuupqioSIGBgWrbtq0S\nEhK0f/9+LV26VN27d7d9NmrxuQCAhpHFuBayGM2FphUe58pR2IsXL8rf319ms1mpqamKjIxUcXGx\nfv/73+uGG27QhAkTFBoaqvj4eGVnZ6u0tFTffPON0tPTbUupw5g2bdqk/Px8/eAHP5C/v7+KioqU\nlZWlOXPm6F//+pfOnTunhIQExcXF6YYbbtA333yjmJgYbrEAAHYgi2EPshjNxc/dBQBXslqttpCc\nPXu2JGn8+PEKCwuzBV9BQYEKCwt111132ZZGP3bsmEwmk9LT0xUfH6/Q0FD3vAE0m4iICJ07d05f\nf/21JGnu3LkqLi5WWFiYunXrpqVLlyowMFBjx45VcnKypk2bZrtuBgDQOLIY9iKL0VxoWuExqqur\nbSsRVlVV6cSJEyorK9Nnn32mUaNG2ZbNr6ioUHl5uQoLCyVJ58+ft43cjRs3Tv7+/pK4PsIoGrv+\nKTU1VV9//bVefvll1dTUKDU1VXfffbduu+02SZfPDGzevFkjR46UxWKxhSSfCwBoHFmMhpDFcDea\nVniEmpoaW0j+4Q9/UFlZmYqLi1VWVqZly5ZJki0so6KilJiYqNWrV+vAgQOSpB07duiRRx6xhaTE\n9RFGcOUfT3v27NGlS5dUU1OjXr16KS4uTpMnT9bhw4dltVrVtWtXmc1mSVJubq5KS0vVvn37elPT\n+FwAQMPIYjSELIYn4JpWeITaL68//vGP2rFjhwYPHqyRI0dq4MCBOnfunHJyclRRUaHExESFhYUp\nOTlZubm5OnnypGpqavTAAw/o7rvvlsTonVFc+cfTrFmztGjRIq1Zs0abNm3S9u3b5e/vr8TERMXG\nxiomJsa2b15entavX6+vvvpK99xzj+Lj4935NgDAa5DFuBpZDE/BLW/gMU6dOqVf//rXuummmzRx\n4kTbSJ0kvfHGG8rOztZdd92l++67T+Hh4bp06ZJMJpPKy8tt182wlL7x1P7xNGLECCUkJKiwsFAL\nFy5UYWGhHnjgAd1zzz22Uf2cnBytW7dOO3fu1OjRo3X//fe7uXoA8C5kMRpCFsPdmB4Mj2EymVRa\nWqqIiAhbSFZVVcnPz08//OEPVVRUpBUrVthGc2uviwgICJBUd+EIGMPhw4e1e/du3XnnnRo+fLjt\nc5GRkaHp06fr3//+t6KiotS/f38dPXpUa9eu1cWLFzVp0iQNHjxYEn88AYAjyGJcjSyGJ+DTA49R\nu5z+4cOHVVlZKUny8/NTVVWVJOnuu++Wn5+ftm/friVLlqikpETS/01nYhqS8ZSXl6ugoEAxMTEy\nm82yWq2qrq5WUFCQXnzxRQUFBWnNmjWSpMTERD322GP64Q9/SEgCwHUii3E1shiegE8Qml1NTU2D\n29u2bau7775bX375pbKysiRdHrH187s8IaCyslLR0dFq166dli9frq+++sq2D7xfQ5+L2j9+cnNz\nJV3+b+3r66uqqiqFhIRoyJAh2rNnj/bt2ydJio2NVVRUlG1fQhIAGkYWoyFkMTwVnyI0q+rqatuX\n18aNG/Xvf/9b//rXv3To0CFVVVVpyJAhSklJ0VtvvaW1a9fq0qVLkqTCwkLt27dPHTp00C9+8Qtb\nWNbU1Ivsj1EAAA+RSURBVDCqawBXfi5KSkpUXFwsSUpOTlbfvn21ePFinThxQj4+PrZpapIUGBgo\nk8mk4ODgesfkcwEADSOL0RCyGJ6M1YPRbK5cge5///d/tXr1au3du1e7du3Sli1bVFhYqLS0NLVv\n314FBQVavHix9u7dqz179mj9+vXKzs7WgAED1KVLF5WXl2vdunVKTU1VZGSkm98ZvosrPxfvvvuu\nFixYoJUrV+qrr75SRESEwsLCdPz4ca1atUo9e/ZURESEpMt/PG3YsEFlZWUaNGiQQkJC3Pk2AMAr\nkMVoCFkMT0fTimZTO9r25z//WTt37tSjjz6qcePG6ZFHHtGmTZu0ZcsW3XjjjUpJSVGXLl0UHR2t\nXbt26fTp0zKZTLr//vs1YsQISdIXX3yhkydPasSIEXxBerErpw3NmjVLW7ZsUceOHdW2bVsdO3ZM\n69evV1RUlNq3b6+8vDwtWrRIBQUFOnDggDZu3KisrCw98MADSktLc/M7AQDvQBbjamQxvAGrB6NZ\n5efn6+DBg7rjjjuUnp4ui8Wiw4cP6+TJkxo8eLA6deok6fI1NcOGDVNGRob8/f1VXl5uG9U7ePCg\ndu3apbi4OELSi125MMPFixd1+vRpjR8/Xrfffrt8fX1VUlKid999V4sXL9bIkSM1ZswYbdu2TRs2\nbJAkRUdH67HHHtOwYcMkcU9AALAXWYxaZDG8BU0rmtWlS5d0+vRpJSYmymKxaOfOnZo5c6bS09M1\nZswYWxgePXpUcXFxCgsLkyTbdRJLly7Vtm3bdObMGU2fPp2g9GK1IfnKK68oMDBQZrNZaWlp8vX1\nVU1NjUJCQvSTn/xEr732mlatWqWpU6fqv/7rv3TvvffKYrHIZDKpTZs2kliZEAAcQRajFlkMb8En\nCy7T0Ap0tSsQmkwmHT161BaSEyZMsIXk+vXrNXfuXNsCALWv++qrr5SVlaXi4mJNnz5d8fHxzfZe\n4BqlpaWqqKhQTk6OTp48qfz8fNs0pdrPz8SJE2W1WrVo0SJJUlxcnCIjI22fF1YmBIDGkcW4FrIY\n3oAzrXCJ6upq2wX9+fn5slgsCgkJUXx8vFJSUjR79mxVVlbq1ltv1ejRo203J8/Pz9e2bdtUXV1d\nZ3qJyWRScnKyfvCDHyg0NFShoaFueV9wruDgYD377LP66KOPtH79em3ZskWdO3eW9H/XXUVGRqpd\nu3YqKCioM+2IewICQNPIYtiDLIY3oGmF0125At2bb76pnTt3KjAwUImJiXrmmWc0YsQI5eXl6eTJ\nkxo4cKBtxcHc3FxlZmZq+/bteuyxx2zhWSs4OLjB5dTh3cLDw/XII4+ooqJCCxcuVKtWrTRy5Ehb\nABYWFqqiokJhYWG2JfYJRwBoGlkMR5DF8HSsHgynunJ6yBtvvKGtW7cqNTVVgYGB2rZtm7766isN\nHTpU4eHhOnLkiNasWaPc3Fxt375da9asUVZWlkaPHs0F/S2MxWJR165dlZeXp2XLlqm4uFg1NTUq\nKChQZmamtmzZovvvv18dO3bk8wAA10AW43qQxfBkJqvVanV3ETCGKy/Ar6io0EsvvaTbbrtNQ4cO\nlSRt3rxZf/vb3xQSEqIpU6aoqKhImZmZ2rFjh2pqatSxY0elp6dr0KBB9Y6HlqGoqEh//etflZOT\nIx8fH3Xt2lWVlZXKyMjQ8OHDJfHHEwA0hSzGd0UWwxPRtMLpXnnlFQUFBSkvL08/+tGPbBfpV1VV\naefOnXrvvfcUGBion/3sZ4qKilJhYaH8/f3l7++vgIAASYRkS1ZUVKQ5c+YoKytLDz74oEaNGmV7\njs8FANiHLMZ3QRbD0zA9GE5VWlqqrKwsbdu2TVVVVerXr59CQkJktVrl6+urtm3bKi4uTlu2bFFO\nTo569eqlyMhImc1m+fr6ymQysQJdC2exWNSpUyfl5uZqxYoVCg4OVqdOnQhJALATWYzviiyGp6Fp\nhVOZzWalpKSoqKhIe/fuVVhYmLp37y6TyWRbFKI2LHNycrR27VoNHTpU/v7+rEAHm8DAQNt1NUuW\nLFHr1q1tN7sHADSNLIYzkMXwJDStcLrAwEB17txZ+fn5WrFihUJCQtSpU6d6YXnDDTeoS5cufAGi\nQbULQpw7d06LFi1SeHi4OnTo4O6yAMArkMVwBrIYnoKmFS5x5Qp0S5YsaTAsb7jhBrVv314SF/Sj\nYRaLRV26dNGFCxc0YMAAtW7d2t0lAYDXIIvhDGQxPAELMcGlioqK9P7772vz5s2aNGmSbfl8wBFc\nQwMA148shjOQxXAnzrTCpa6eVhIQEKAuXbq4uyx4GUb+AeD6kcVwBrIY7uTn7gJgfGFhYXrsscd0\n8eJF+fr6urscAABaHLIYgDdjejCaTXl5ue3ebwAAoPmRxQC8EU0rmh0LPQAA4F5kMQBvQtMKAAAA\nAPBYLAEGAAAAAPBYNK0AAAAAAI9F0woAAAAA8Fg0rQAAAAAAj0XTCjSz3NxcjR07Vnv37m3w381t\n/vz5eu6555rc59lnn9Wnn35q9zGd9Z7c/bsBABgPOWw/d/9ugFp+7i4AcLc333xT69atkyT5+Pgo\nIiJCvXr10rhx49SqVSuX//zIyEi98847CgkJsWv/vXv36le/+pVmz56tqKgoF1cHAIBrkcMAroWm\nFZDUtWtX/eQnP1F1dbUOHz6st99+W+fOndPUqVMb3L+qqkp+fs75v4+Pj4/CwsKcciwAALwROQyg\nKTStgCQ/Pz9bYLVp00YnTpzQvHnzVFFRoaKiIk2ePFnPPfecNm7cqN27d+uuu+7S+PHjdebMGc2Z\nM0c7d+6UyWRSx44dNXHiRMXHx9uOvWnTJv3jH/9QQUGB2rdvr/vvv7/Oz87NzdXkyZP10ksvKTk5\nWZJ0/vx5zZ07V9u2bdPFixcVGRmp++67Tz169NCvfvUrSdLkyZMlSd26ddO0adMkSVlZWfrss890\n+vRphYWFKT09Xd/73vdksVgkSRUVFfrwww+1ceNG+fj4qH///goODnb497Vx40YtXbpUp06dkp+f\nn5KSkvToo48qNja23nubP3++9u3bp/DwcI0bN04DBgywPV9UVKS5c+dq69atqqysVEJCgsaNG6du\n3bo5XBMAwHuRw44hh9HS0LQCDTCbzbJaraqurrZtmzt3rh555BE9/vjjki5/0b/44otKT0/XSy+9\nJD8/Py1fvlzTpk3TrFmz1Lp1ax05ckR/+MMfdN999+m2227TiRMn9MEHHzT5sysqKjRt2jSZzWY9\n99xzio6O1pkzZ1RSUqLIyEj9/Oc/1yuvvKIZM2YoMjLSNtKcmZmpDz/8UJMmTVKXLl107tw5vf/+\n+youLrZdK/PRRx9p8+bNmjx5smJjY7VmzRqtWLFCrVu3duj3U1lZqVGjRikuLk6XLl3S/PnzNXPm\nTL322mt1Rr7nzp2r8ePH64knntD69ev1xhtvKDY2Vu3bt1dFRYWmT5+udu3a6YUXXlBwcLA2bdqk\n3/zmN3rllVcUFxfnUE0AAOMgh5tGDqOlYSEm4ConT57UihUr1KlTJwUGBtq2Dx06VAMHDlRUVJSi\noqK0cuVKRUVF6cknn1R8fLxiY2M1adIkBQcHa8OGDZKkRYsWqXPnznr44YcVGxurvn376t57723y\n52/cuFG5ubn6+c9/rtTUVEVHR6tnz54aMGCAfP5/e/cTEkUfx3H8vZllHbZJTbemoNY/a7EHiSLo\njyAUdfDUaU2spUOHxCCiu4WHSC8RdOgkJa0iWFGgoCT9oUCKKIvaXRUJnJYtobZIy2X3OYhD+6zW\n2sMD+zz7ed1m5rczv9nDfPj+Zn4zy5bZc26cTieGYdjLPT091NfXU1NTQ2lpKdu2beP48eM8fPiQ\nr1+/MjMzw8DAAD6fj507d2KaJkePHmX9+vVL/o9qa2vZsWMHLpeLLVu20NTURCQSYXR0NK3dvn37\n2LBhAz6fj4qKCu7evQvMjXxPT09z+vRpysrKcLlcHD58GI/Hw+Dg4JL7JCIi/w/K4d9TDkuu0Z1W\nEeD169c0NjaSSCSIx+N4vV5OnDiR0qa8vDxleWxsjPHxcRobG1PW//jxg/fv3wMwOTmJ1+tN2T7/\n6NFixsfH2bhxI0VFRRn3PxaL8eHDB65du8b169fTtkciEfLz85mdncXj8aT159mzZxkfC2BiYoKe\nnh4mJib48uULyWQSgI8fP6a0q6ysTFn2eDy8evUKgNHRUT59+oTf709pE4/HWbFixZL6IyIi/23K\nYeWwyK+oaBUBKioqaGpqst9auNDLHebno8xLJpN4vV77MaWfrV69+l/r60ISiQQAfr8/LZwBCgsL\n7QD/p75//05raytVVVWcPHmSNWvWAHDmzBni8XjG+0kmk5imydmzZ9O2KSxFRHKLcjhzymHJRSpa\nRZi7OLtcriX9xu12c//+fQoLCxe9uJumSSgUSln3u2+dud1uhoaGmJqaWnCUdz7I5wMSwDAMioqK\nsCyL/fv3L7jf0tJSli9fTjAYZNOmTfb6YDD4y/783eTkJLFYDJ/PZ893CQaD9ijvz8LhMNu3b7eX\nQ6EQpmkCUFZWxoMHD1i1apUduCIikpuUw5lTDksu0pxWkT906NAhEokEbW1tvHnzhmg0ytu3bwkE\nAnYA1dXVEQqFCAQCWJbF8PCwPZdkMXv27KG4uJiLFy/y8uVLotEoIyMjPH78GJj7npzD4eD58+d8\n/vyZb9++AVBfX09fXx+9vb28e/fOPt7Vq1eBuRHqAwcO0N3dzdOnT7Esi87OTizLWtJ5FxcXk5+f\nT39/P5FIhJGRETo6OnA4HGlt7927x6NHj7Asi+7ubkKhEHV1dQDs3buXkpISLly4wIsXL4hGo4TD\nYW7evMnw8PCS+iQiIrlHOawcltyhO60if8gwDFpbWwkEArS3tzM9PY1hGFRVVbF27VpgbrT21KlT\ndHV1cefOHTZv3syxY8doa2tbdL8rV67k3LlzdHZ2cunSJWZmZli3bp39in7DMDhy5Ai3bt2io6OD\nrVu30tLSQk1NDQUFBdy+fZve3l7y8vIoKSlh165d9r4bGhqYnZ3l8uXLAOzevZuDBw/y5MmTjM/b\n6XTS3NzMjRs3GBoawjRN/H4/58+fT2vb0NDA4OAg4XAYwzBobm7G7XYDc6PqLS0tdHV1ceXKFWKx\nGE6nk/LycqqrqzPuj4iI5CblsHJYcocjudCzBCIiIiIiIiJZQI8Hi4iIiIiISNZS0SoiIiIiIiJZ\nS0WriIiIiIiIZC0VrSIiIiIiIpK1VLSKiIiIiIhI1lLRKiIiIiIiIllLRauIiIiIiIhkLRWtIiIi\nIiIikrX+AmVv8UBQURckAAAAAElFTkSuQmCC\n",
            "text/plain": [
              "<Figure size 1440x1080 with 4 Axes>"
            ]
          },
          "metadata": {
            "tags": []
          }
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "iVBORw0KGgoAAAANSUhEUgAAA6MAAAGQCAYAAABbF7Y2AAAABHNCSVQICAgIfAhkiAAAAAlwSFlz\nAAALEgAACxIB0t1+/AAAADh0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uMy4xLjIsIGh0\ndHA6Ly9tYXRwbG90bGliLm9yZy8li6FKAAAgAElEQVR4nOzdfVyN9/8H8Nc53aooUZqSSGoSkpK7\n+GJsmrmnEIaZ0W5txmaWbTKbm5mYYTb3Re6GGENEN/hh5l7kJneVim50f35/tM4cp5tz2nXO6Vy9\nnnucx2Nd57r5XMfRy/tzfa7PJZHJZDIQERERERERaZFU1w0gIiIiIiKi2ofFKBEREREREWkdi1Ei\nIiIiIiLSOhajREREREREpHUsRomIiIiIiEjrWIwSERERERGR1rEYJarFEhIS4OrqioSEBF03pcaa\nMWMGXF1dUVRUpOumEBFpBLOgarrKgp49e2LGjBkqrbt9+3a4uroiNjZWw63SHVdXVyxdulTw/ZZ9\ndsnJyYLvuyq1/e8fi1HSqLK/YFu3bq10PZlMhv3792PUqFHo0qULPDw80L17d0yYMAEbNmwA8O8v\nClVeZX+hy35+6623yj3uhQsX5Ouo8sutLIx69OiBgoICpfeDgoLg5+dX5X70Uc+ePeHq6ooPPvig\n3PddXV3x8ccfV2vfFy9exNKlS3USAkSkecwC8WAW6FZmZiaWLl1aawsXfVVSUoKlS5fizz//1HVT\nahxDXTeACADmz5+PX3/9FT4+Ppg4cSLMzc2RnJyMc+fO4bfffsPo0aPh7e2N7777TmG7efPmwdLS\nElOmTFFY7uzsLP9/ExMTnDhxAmlpaWjYsKHCejt37oSJiQny8/PVau+DBw8QERGBoKAgNc9U/+3f\nvx9XrlyBm5ubYPu8dOkSwsLC4OPjAwcHB8H2S0T6hVmgP5gF2rF//35IJBL5z0+ePEFYWBiCg4PR\nsWNHHbZMXAYMGAB/f38YGxtrZP8lJSUICwvDoEGD0Lt3b4X3vL29cf78eRgZGWnk2DUdi1HSuceP\nH2Pt2rXo3bs3li1bpvR+WloaAKBJkyZo0qSJwntLliyBtbU1BgwYUOH+O3fujNjYWOzZswfjxo2T\nLy8qKkJUVBR69uyJffv2qdVmd3d3/Pzzzxg2bBhMTU3V2lZTZDIZ8vPzNdoeJycnpKSkYMmSJfjp\np580dpyaIDc3F2ZmZrpuBlGtwSwQBrNAWLrOAk0VR+rQ9WegSWXnZmBgAAMDA520QSqVwsTERCfH\nrgk4TJd07u7duygpKYG3t3e577/Yg60uCwsL9OzZE7///rvC8piYGDx+/LjSf7xU5L333kNqaio2\nbtyo0vpXrlyR92J6eHigf//+SsPVKrtn4MV7RpKTk+Hq6orFixdj+/bt8Pf3h4eHB6KiogCUDmOb\nMGECunbtitatW6NHjx74+uuvkZ2drfa5Ps/a2hqjR4/G4cOHcf78eZW2OXjwIAICAuDp6Yl27doh\nKCgIp0+flr+/dOlSzJo1CwAwZswY+VC57du3Y/PmzXB1dcXNmzfl658+fRqurq547bXXFI4zf/58\ntG7dGs+ePZMvu3HjBt599134+PigTZs2GDRoEHbu3KnUxp49eyIwMBDnzp3DqFGj0LZtW3z11VcV\nntPjx48xaNAgdO3aFdeuXVPpcyCiyjELSjELxJcFw4cPx4gRIxSWff7553B1dcXKlSvly2QyGTp1\n6qQwzPn5P/OEhAT06dMHABAWFib/jF68p7S4uBhLly6Fn58fPDw8EBAQgCtXrlTYPnU+g7t372L6\n9Ono0qULWrdujb59+2LlypUoKSlR2E9hYSEWLlyIrl27om3btggICMC5c+cQFBSkMJKg7Du8fft2\npba8uG55MjMz8f3332PgwIHw8vJCmzZtMHjwYOzatUtp3bLh9Q8ePMCHH34Ib29vvP766wDKv2f0\n7t27mDZtGrp164bWrVujc+fOmDhxosJnefr0aUybNg09e/aEh4cHOnbsiHfffVfhu5qcnAx3d3cA\nwI4dO+R/bmXnVtHfeXW/t1euXEFQUBDatm2LLl26YPHixUp/LjURr4ySzpUNxTl48CAGDx6MevXq\nCX6MAQMGYPLkyUhMTESLFi0AlA7Lat26NZo3b672/rp27QovLy+sWrUKAQEBMDc3r3DdM2fOYPz4\n8WjSpAkmTJgAc3NzHDlyBLNmzUJ6ejrefvvtap/X4cOHkZGRgcDAQFhbW6NZs2YAgA0bNqBZs2YY\nN24c6tati4sXLyIiIgJXr16V33dVXRMmTMCmTZuwZMkS/PLLL5Wu+9tvv2HevHno3r07PvroIxQW\nFiIyMhLjxo3DmjVr4OPjg1deeQUPHz5EZGQkJk+eLP/zaN++vfxerJMnT8qXnzp1ClKpFDdv3lQY\nbnfy5El4eHigTp06AIDbt28jICAAMpkMo0aNgrW1NaKiovDpp5/i8ePHmDBhgkJbHz16hLfffhtv\nvPEGBgwYAAsLi3LP6f79+3jzzTdRVFSEzZs3K12hIaLqYRYwC8SaBd7e3vjtt98UrjCWtf/kyZOY\nNGkSACAxMRHp6ekVDr91dnbGp59+ivnz5+OVV17BK6+8AgBwdHRUWG/x4sWQSqUYP3488vPzsWbN\nGkyZMgUHDhyAoWHV//Sv6DO4ffs2RowYATMzM4wcORINGjTAyZMnsXDhQiQnJysUrbNnz8b27dvR\ns2dPdO3aFUlJSZg0aRIsLS1hZ2dXZRtUdffuXURFRaFv374YPnw4CgoKcODAAUyfPh2FhYUYOnSo\n0jZvvfUWmjdvjg8//LDce76B0mJ6/PjxyM3NRWBgIOzs7PD48WOcOnUKN27ckA9P37dvH9LS0jBk\nyBDY2toiOTkZW7ZswciRI7F37140aNAA1tbWmD9/Pj799FN06NABw4cPB1B5B5u639vU1FRMnDgR\n/fr1Q79+/XDs2DGsWLECDg4OGDZsWHU/Xq1gMUo617BhQ4wYMQIRERHo3r07vLy80K5dO3Tu3Bnt\n2rWDVPrfL+B369YN1tbW2LlzJz7++GNkZWXhyJEj+OSTT6q9z/fffx9jxozB2rVrle5TKiOTyfDF\nF1/AxcUFmzZtkt8PMGrUKLz33ntYvnw5AgICYGlpWa02JCUlYd++fUohuHHjRnkQl/H09MSMGTNw\n9uxZeHp6Vut4AGBlZYVx48YhLCwMp0+fRocOHcpd7+HDh1iwYAHGjRuHmTNnypePGDEC/fv3x4IF\nC7Blyxa4ubmhXbt2iIyMROfOnZVCuEGDBjh16hQCAgIAlAZ47969ceTIEZw8eRL9+vVDdnY2Ll++\nrDA5yaJFi5CVlYWtW7fCw8MDABAYGIhRo0ZhyZIlGDRoEKytreXr37t3DwsWLED//v0rPPebN29i\n/PjxsLCwwLp169CoUSP1P0AiKhezgFkg1izw8fHB6tWrcebMGXTt2hWPHj3C7du34e/vj+joaBQX\nF8PAwAAnT54EgEpHB/Tq1Qvz58+Hq6trhVfzJRIJwsPD5YWns7Mzpk6dihMnTqB79+6VtrWyz+Cb\nb76Bubk5du7cibp16wIAAgIC0KhRI6xZswZjx46Fs7Mzrl69iu3bt+P111/HwoUL5du7urpi1qxZ\nghajrq6u+PPPPxWG2I4dOxbjxo3DqlWryi1GPT098fXXX1e63xs3buDOnTv44YcfFK6+v9hp9PHH\nHyv9HRs4cCD69++PyMhIvP322zAzM8Prr7+OTz/9FE2aNFFpFIa639u7d+9i2bJl8vtRAwMDMWDA\nAERERNT4YpTDdKlGCAkJwVdffYUWLVrgxIkTWLp0KQIDA9GrVy+cOHHiP+/f0NAQ/v7+2L17N0pK\nSrBv3z4UFxfD39+/2vvs2LEjOnXqhF9//RVPnz4td52rV68iMTER/fv3R1ZWFtLT0+Wv7t27Iy8v\nD2fPnq12G/z8/MrtjS37xVhSUiI/btk/FP7+++9qH6/Mm2++CSsrKyxZsqTCdf744w8UFhaif//+\nCuedn5+PTp064fz58yoNFevQoYM8oIuKinD27Fn50KOy5f/3f/+H4uJieYAXFxfj2LFj8PX1lf8S\nB0rvvRk7dizy8/OVvldWVlaVfh8uXbqEUaNGoWHDhli/fj0LUSINYBZUD7OgZmeBl5eXQrF56tQp\nGBkZYdKkScjJycHFixfly21sbODk5FTlPiszfPhwhSugZZ/HnTt3VNq+vM/gyZMniImJQZ8+fVBY\nWKjwZ1k2c3R8fDwA4OjRowBKh1s/b9CgQYKPeDA2NpYXogUFBcjMzERGRgY6d+6MW7dulfvdGjly\nZJX7LbsaHBMTg5ycnArXe74QzcnJQUZGBiwtLdGsWbNq/x2rzvfWxsam3ImR7t69W602aBOvjFKN\nIJVKMWLECIwYMQK5ubm4dOkSDh48iI0bN+Kdd97Brl275MOOqmvAgAFYv349EhISsHPnTnkPeVZW\nVrX3+f777yMgIABr1qwpd5r7snsG5s6di7lz55a7j8ePH1f7+C8OzSlz/vx5LF68GP/3f/+nNDtk\nRf9YUoeFhQXGjx+PRYsWIS4uDp06dVJaJykpCQAwZMiQCveTkZFR4RCoMj4+Pvjjjz+QlJSEJ0+e\nIDc3Fx07dkRycjIOHjwIoDTADQ0N5b386enpyM3NVZhJs0zZ0LwXHx1gb29f6ZWXcePGwcbGBmvX\nrq10KB4RVR+zoHqYBTU7CywsLODm5iYvRk+ePIk2bdrAzc0NjRo1QkJCAtq0aYPTp0/Dx8dHpX1W\npnHjxgo/l11xf/LkiUrbl/cZ3L59GzKZDGvWrMGaNWvK3a7sO1z2mTZt2lThfUNDQ8FnSZbJZFi7\ndi3Cw8Nx69YtyGQyhfefPn2q9N1S5fYaBwcHTJw4EatXr8bvv/+Odu3awc/PD/3798dLL70kXy8l\nJQULFy7EkSNHlD7f+vXrV+ucqvu9fZGlpSUyMzOr1QZtYjFKNY6ZmRk6dOiADh06oEWLFpg1axai\noqIwderU/7RfDw8PODs746effsKZM2ewaNGi/9xWT09PdO/eHevWrVPqAXxecHAwvLy8yn2v7JfN\n81O3v6i4uLjc5eXNvpacnIwxY8bA3t4en3zyCezt7WFqaori4mJMnDhR6Rd1dQUFBWHt2rVYsmRJ\nuf8AKTtOWFhYhYFtY2NT5XHKgvnUqVPIzMzESy+9BEdHR/j4+GDFihXyezjc3d3/U5FY1Ux2r776\nKrZt24adO3di1KhR1T4OEamGWaCMWaC/WdCxY0esX78ez549w6lTp9C3b18ApVevTp06hV69eiE1\nNbXCIbrqqGhWWFX/zMv7DMq2DQgIkLf9RdUpNKv6vlc1w+3q1avlQ4onT54Ma2trGBoa4ujRo/jt\nt9/KncBH1ZmmP/nkEwwZMgSHDx9GXFwcli5diuXLl2Pp0qXo1q0bSkpKMGHCBDx69Ahjx46Fi4sL\nzMzMIJVKERoaKtjfMVUIcRuDrrAYpRqtTZs2AEpvphfCgAEDsGjRItStWxe9evUSZJ/vv/8+Bg8e\njNWrVyu9V9ZbbWJigs6dO1e6n7KhKy/2Vufn5yM1NVXl9hw6dAjPnj3Dzz//rBAMz8/sJgQzMzNM\nmjQJ8+bNkw/JeV5Zj6itrS3atm1b7eO4uLjAysoKCQkJePr0qTyoPT095YFz4cIFvPnmm/JtrK2t\nYWZmhhs3bijtr2yZuqE5Z84cyGQyfP311zA0NFSaGZGINIdZwCzQ9yzw9vbGmjVrcODAAdy8eVN+\nT6yPjw++++47+RDXqq6MVla8aVKTJk3kx67qO1z2md6+fRtWVlby5UVFRUhOTlZ4Nm3ZVdvyrtQn\nJycrXV190d69e+Hj44MFCxYoLI+Li6t0O1U1b94czZs3x8SJE/HgwQMMHDgQy5YtQ7du3XDt2jVc\nu3YN8+bNw+DBgxW2y8zMVLgyqs6fmya+tzWZ/pbRJBqpqakVTjkeHR0NAOUOVaiOQYMGITg4GCEh\nIYI908nd3R2vvPIKNm3aJH8OXplWrVqhWbNmWLduHdLT05W2fX5YloODAwwNDeWBVGbdunUV9oaX\np6wX8cUeuVWrVqm8D1UFBgbC1ta23PuF+vbtCyMjI/z4448oKipSev/5cy/rxS4vjCQSCby9vXHy\n5EmcOXNGHtRmZmZo3bo1Vq1ahaKiIoXeZAMDA/j5+SE+Pl5+Lw5QOjveunXrYGxsjC5duqh1rhKJ\nBHPnzkX//v3x5ZdfYtu2bWptT0SVYxaUYhaIMws6dOgAqVSKFStWwMjISD6U2NvbG9nZ2Vi/fj0a\nNmxY5Xe8bDZeVYfcCsXa2hqdO3fGzp07y+3QyM7Ols9MW3YP6bp16xTW2bFjh9KfrYWFBRo0aKD0\nfY+KilKp80kqlSp9x9PT0xEZGVn1SVUiOztb6fv60ksvwdraWv7Zl12NfPH4O3bsUOo4MjAwgImJ\niUp/bpr43tZkvDJKWhEdHV1uj26HDh1Qp04dDB06FO3bt0fnzp3x0ksvITc3F6dPn8bBgwfh5ORU\n6b0m6rC1tcW7774ryL6e99577+HQoUO4efOmwmQGUqkU8+bNw/jx49GvXz8MHToUjo6OSE9Px+XL\nl3Ho0CFcuHABQOkv5P79+2PTpk2QyWRwcXHBuXPncPr0abXuO+jWrRtMTEwwadIkjBgxAlKptNx7\nGYRgYmKCyZMnl/scNnt7e0yfPh1z587FoEGD8Nprr6Fhw4Z48OABTp06BYlEgvXr1wMAWrduDYlE\ngp9//hlPnz6Fqakp2rRpI7+vw9vbW35P0PMzLPr4+GDlypUwMDBQGvr24YcfIjY2FuPGjVOYFv3c\nuXOYPn26wix0qpJKpfj2229RVFSEWbNmwcDAAAMHDlR7P0S1FbOAWVBbs6BevXpwdXXF5cuX4eXl\nJR8q2rx5c9jY2ODmzZt49dVXqzx2w4YN0bhxY0RFRaFZs2awsrKCg4PDf7rqrKqQkBAEBARgyJAh\nGDp0KFq0aIGnT58iMTERBw4cwO7du+Hg4AA3NzcMHDgQO3fuxLNnz9ClSxckJSVh9+7d5d7fHBgY\niLCwMHzyySfw8vLC9evXsW/fvgrvhX5e7969sWTJEkybNg0dO3ZESkoKwsPD8dJLL5Xb8aOq+Ph4\nhISEoG/fvnBycoKhoSGio6Nx8+ZNvPfeewBK/+yaNWuG+fPn4/79+7CxscFff/2FQ4cOlXtfqoeH\nB+Li4vDLL7/Azs4O1tbW5Q5tBzTzva2pWIySVvz555/4888/lZZPnjwZkyZNwuzZsxETE4MdO3Yg\nLS0NEokEDg4OGD9+PN56660qJzbQtZYtW6Jfv37Ys2eP0nuenp6IjIzETz/9hJ07d8qHbjg7OytM\ncw8An332GYqKirBr1y6UlJTA19cX69atq/Khz89r2rQpfv75ZyxatAg//PADTE1N0aNHDyxcuLDC\nX3r/xbBhw/DLL7/g3r17Su+NGTMGzZo1w6+//oo1a9YgPz8fNjY28odSl3F0dMTs2bOxZs0afPHF\nFyguLsa8efPkv8zLesDt7OwUwsnb2xsrV67Eyy+/rPQdcXJyQnh4OBYvXoyNGzciLy8Pzs7OmD9/\n/n8qIA0MDPD999+jqKgIn332GQwNDeUPzSaiyjELmAW1OQu8vb1x+fJlpaG43t7eiIqKUnnyou++\n+w7z5s3Dt99+i4KCAgwaNEgrxaijoyN27NiB5cuX4+DBg9i8eTPq1auHpk2bYurUqQr3/n799ddo\n2LAhdu7ciRMnTsDd3R2rVq0qdzTC22+/jczMTOzZswcHDhxAmzZtsGbNmgon+3repEmTUFRUhJ07\nd+LAgQNwcHDA5MmTYWZmpvT3Sh2urq7o0aMHjh8/jm3btsHAwABNmzbFN998I39cjKGhIVauXInQ\n0FCsX78ehYWFaNeuHdauXYvQ0FClfYaEhGDOnDlYunQpnj17Bh8fnwr/Lmrqe1sTSWTavLuWiIiI\niIhqneLiYnTq1Al9+/at8jmfVHvwnlEiIiIiIhJMXl6e0rKtW7fiyZMn8PX11UGLqKbiMF0iIiIi\nIhLMunXrEB8fj86dO8PMzAznz5/Hzp074e7ujj59+ui6eVSDcJguEREREREJJiEhAWFhYbh27Rqy\ns7PRoEED/O9//8OHH36o8LgXIhajREREREREpHW8Z5SIiATx+NZtXTeBiIhIVMSerbwyqmcmS+rq\nugk1ygpZFj+TF6zISdZ1E2oWM0sgV7sPB6/xzCw1tuvJknpqb7NCpvyAeyJtmWagub8P+mhh8RN+\nJi9YkJGo6ybUOJJ6NpA9VX5mcG0lqWdT9Ur/gZizlRMYERGRYDjchoiISFhizlYWo0REJBipRKLr\nJhAREYmKmLOVxSgREQlGzL23REREuiDmbGUxSkREgpGKt/OWiIhIJ8ScrSxGiYhIMGLuvSUiItIF\nMWcri1EiIhKMmO9rISIi0gUxZyuLUSIiEoyYe2+JiIh0QczZymKUiIgEI+b7WoiIiHRBzNkq5kKb\niIiIiIiIaiheGSUiIsGwh5OIiEhYYs5WFqNERCQYiYgnWSAiItIFMWcri1EiIhKMmHtviYiIdEHM\n2cpilIiIBCPmSRaIiIh0QczZymKUiIgEI+beWyIiIl0Qc7ayGCUiIsGI+cHcREREuiDmbGUxSkRE\nghFz7y0REZEuiDlbWYwSEZFgxHxfCxERkS6IOVtZjBIRkWDE3HtLRESkC2LOVhajREQkGClE3H1L\nRESkA2LOVhajREQkGDEPJSIiItIFMWcri1EiIhKMmIcSERER6YKYs5XFKBERCUbMvbdERES6oOls\nLSwsREREBGJiYpCdnQ0nJycEBATAw8Oj0u2mTp2K1NTUct+zs7PDjz/+WOWxWYwSEZFgxHxfCxER\nkS5oOluXLVuGhIQE+Pv7w87ODtHR0QgNDcWcOXPQsmXLCrcbO3Ys8vLyFJalpaUhPDwcbdq0UenY\nLEaJiIiIiIhqocTERMTGxmLs2LHw9/cHAPj5+WHatGnYuHEj5syZU+G2Pj4+Ssu2bdsGAOjWrZtK\nxxfzEGQiItIyqUT9FxEREVVMk9kaHx8PAwMD9OrVS77M2NgYPXv2xJUrV5CRkaFWW48fPw5bW1u4\nurqqtD6vjBIRkWC00cOpy3tbiIiItE2T2ZqUlAR7e3uYmpoqLG/RogVkMhlu3bqF+vXrq7yve/fu\nYfDgwSofn8UoEREJRhtXOnV5bwsREZG2VTdbU1JSlJaZm5vD3Nxc/nNmZiasra2V1isrQNW5MhoT\nEwMA6Nq1q8rbsBglIiLBaHqSBV3f20JERKRt1c3W4OBgpWVDhw7F8OHD5T8XFBTA0FC5JDQyMpK/\nr4qSkhLExsaiWbNmcHBwULmNLEaJiEgwmr4yWtm9LeHh4cjIyFB5OBGg/r0tRERE2lbdbA0LC1Na\n9vxVUaA0Q4uKipTWKywslL+vikuXLiE9PV3eUawqFqNERCQYTY/S1fW9LURERNpW3Wy1tbWtch0r\nK6tyh+KWLVM1U48fPw6JRIIuXbqo1UYWo0REJBhN3tcC6P7eFiIiIm3T5KgjJycnREVFIS8vT6Gj\n9/r16wCApk2bVrmPwsJCJCQkwN3dvdyMrgyLUSIiEowm72sBdH9vCxERkbZpcj4GX19f7N69G4cO\nHZIPsS0sLER0dDRcXV3lxWVaWhry8/Nhb2+vtI+zZ88iJyenWp27LEaJiEgwmryvBdD9vS1ERETa\npskroy4uLvD19cWGDRuQkZEBOzs7HD16FKmpqZgyZYp8vbCwMFy6dAlbtmxR2kdMTAyMjIzg6+ur\n9vFZjBIRkWCq+yw0Ve5rAXR/bwsREZG2afoZ3sHBwYiIiMCxY8eQk5MDR0dHzJw5E25ublVum5ub\nizNnzsDT0xNmZmZqH5vFKBERCUbTExjp+t4WIiIibdN0thobGyMoKAhBQUEVrhMSElLucjMzM2zc\nuLHax9Z0oU1ERLWIVCJR+6UOX19fFBcX49ChQ/JlFd3bcu/evXL38V/ubSEiItI2TWerLvHKKBER\nCUbT8afre1uIiIi0TX9KS/WxGCUiIr2iy3tbiIiISDgsRomISDDa6L3V5b0tRERE2sYro0RERCoQ\nc2ASERHpgpizlcUoEREJRqJHkyYQERHpAzFnK4tRIiISjHjjkoiISDfEnK0sRomISDB8XhgREZGw\nxJytLEaJiEgwIh5JREREpBNizlYWo0REJBiJqAcTERERaZ+Ys5XFKBERCUa8cUlERKQbYs5WFqNE\nRCQYMQcmERGRLog5W1mMEhGRYKRiTkwiIiIdEHO2shglIiLBiPm+FiIiIl0Qc7ayGCUiIsGINy6J\niIh0Q8zZymKUiIgEI+bp54mIiHRBzNkq5meoEhERERERUQ3FK6NERCQYEXfeEhER6YSYs5XFKBER\nCUYq6sgkIiLSPjFnK4tRIiISjHjjkoiISDfEnK0sRomISDBinmSBiIhIF8ScrSxGiYhIMCLOSyIi\nIp0Qc7ayGCUiIsGI+cHcREREuiDmbGUxSkREgpGKNy+JiIh0QszZyueMUo1iVr8+Jm/fhCXZDzH3\n1kV4Bw4rd73gqG34IesBAOCHrAcIy3+ML87HK6zT87138M3Nv7Ek+yG+vHQati4tNN5+0o709AwM\nChgFc5vGaOrWGpsitpa73veLf0Tr1q1Rt5EDmrVqg+8X/1juekdjjkNiboVZc77RZLNrBUk1XkSk\nWXXq18e4bRsQ+vQ+Pr/5NzwDh5a73sS9kQCA0Cf3EPrkHuY/S8XH52KV1mvu1wULi5/g1a9mabTd\npD3p6RkYPGosLF5yglPr9ti0dVu5632/JAwevn6oW7cumnt0wPdLwpTWWfLTSjT36ACLl5zQyrsL\nriXe0HTzRU/M2coro1SjBC5biKKCAkxv5AyHdm0QvHcrkv/6Gw8uXVFYL6zfEADAClkWPqj7Ej46\nEoUrh4/K3+8yYSw6TxiDZf5D8eDyVTRs3gy5GZlaPRfSnKkffgxjY2M8SrqGc+f/hv+QEWjr0Rru\nrV5WWE8mk2HdunVo06IpbtxMQp83BqGJgz0Chg2Rr1NYWIj3P5mBjt4dtH0aoqRPAUhUWwwJW4Di\ngkKEvOQC+3YemLB7C+7/dZjikLIAACAASURBVAGPXsjW1f5DsbD4CT6ztAcAvHNoDxKPHFNYR2po\niIGLv8Xt+FNaaz9pXvDHM2BsZIyH1y/g3N8X8PrwUWjb2h3uL7sprCeTybB2RRjadu6OxL9Ooe+g\n4Whib4+AoYMAAKvXbsCa9RuxZ+tGvOzaEjeTbqG+lZUuTklUxJytvDKqopCQEISEhOi6GaJmbGYG\nzyED8PsX3yA/Jwc3TsThr9+j0DEosNLtGjR1RItunZGwbjMAQCKRwP/LGdj64Qw8uHwVAJB2Mwm5\nGRkaPwfSvJycHGzb9Tu+/uJzWFhYoGvnTnij36tYvzlCad3pH72P9u3bw9DQEK4tXTDg9X44Ea94\nBX3hkjD06dUTbi1dtHUKoiapxn9UOzFXtcPYzAweg9/AvtnfoCAnB0kn4nFx9z50GB1Q6Xb1mzqi\nebfOOL0+XGF5j4/exbWDR5By9Zomm01alJOTg22/78FXs2aU5monX7zxWl+sD1cedTT9g3fRvl2b\n0lx1aYE3+r2KEwknAQAlJSX4av4CLAr9Gq3cXCGRSODcvBmsretr+5RER8zZymL0OZmZmdiyZQtu\n3bql66bUSo1atkBJURFSrifKl9376wIau79cyVZAxzGBSIyJxePbdwAAVg72sG7iAPvWrRB65zK+\nufk3Xg/5DBIxz4tdi1y7nghDQ0O0fG7YdVsPD1y8fLnS7WQyGWJOxMH95X+/T7fv3MGa9Rswe+Z0\njbW3tpFI1H+ReDFXdc/mn2xNu/7vUMkHf11AI3e3SrYCOgQF4GZMLDL+yVYAqO/YBD5vjsaBr+dr\nrL2kfdcSb5bmagtn+bI2rd1x6crVSreTyWQ4HhsPdzdXAEDyvftIvncfFy5fgWOrdmju0QFfhs5H\nSUmJRttfG4g5WzlM9zmZmZmIjIyEra0tnJycFN6bNYv3RWiaiYUFnj3NUlj27MlTmNa1qHQ73zGB\n2PfN9/Kf6zuUDi96uU8vfO3hCzMrS7x3YBcyk+/j+OrfBG83aVd2Tg7q1a2rsMzSsh6ysrMr3S5k\n7jyUlJTgzaBR8mXvffyp/AorCYM9nPQ85qruGVuYI+/FbH36FCZV/N7rEBSIP0O/V1g2cMl87P9y\nLgpycgRvJ+lOaa4qfh8s66mQq/O+Q4msBG+OLh3Blnz/PgDg4OFonI89iswnT9B30HA4NG6Mt8YF\naabxtYSYs1XM5yYoQ0NDGBqydtek/Oxs1KmnWGSY1quLvKzKfxnWs2uEM5E75T8XPnsGADjw3WI8\ne/IEj2/fQczPa9C6Xx/hG01aZ2FujqdZiv+wevr0KepW8g+rsBUrsW5TOPZu3wITExMAwO6ofcjK\nysaIoYM12t7aRsyTLJCwmKvaUZCdA9MXs7VuXeRXUmg06+KLuna2+Ctyl3xZq9dfhYlFXZzbsl1j\nbSXdKM1Vxe/D06ysynM1LAzrN2/Fni2b5Llax7QOAOCT94NhZWUJp6aOmPTmGOw7+KfmGl9LiDlb\ntZoCW7ZsQWRkJMLCwrBlyxacOlV683vHjh0xYcIE+ZcZAI4cOYL9+/cjOTkZpqam8PT0xOjRo2H1\n3E3QJSUliIyMxKFDh5CTkwMXFxdMmDAB8+bNQ6tWrTB16lQAQHZ2NrZv346//voLKSkpkEqlcHV1\nxciRI+U9tRcvXsScOXMAAMuXL8fy5csBAFOmTEGPHj3k97WEhIQgMzMTb7/9NkaMGIHBgxX/IXv9\n+nV8/vnnCA4Ohp+fHwAgLS0N4eHhOHfuHHJzc9G4cWMMHDgQXbt2Ff5D1mOPriVCamgI2xbOSPln\n5jWHtq1x/2Llwy/Pbf8d+c/10j68eh2F+fmQyf5dR/b8D6TXWrq0QFFREa4n3oDLP0OK/vr7gsLw\n2+etWbMG3y78AccORMHB3l6+/NCRozh99hzsmrUEADx5+hQGBgb4++JF7NqyWfMnIlIcDq9dzFXm\nalVS/8nWhi2aIy3xJgCgcdvWeHTxSoXbdBgzEn/v2K1wBdSlZ3c06dAOX94rvVe0jmU9lBQX4yWP\nVvh10EjNngRpVMsWzUtz9cZNuDg3BwCcv3ARrf4ZfvuiNes34dtvv8fRvTvgYN9YvtzVxRnGxsYK\nOcBMEIaYP0edXBlduHAhCgoKMGrUKHTq1AnR0dHYuvXfm6S3bt2KFStWwN7eHmPHjkXfvn1x+vRp\nzJkzBwUFBfL1Nm3ahMjISDg7OyMoKAh2dnaYO3cu8vLyFI736NEjnDp1Cl5eXhg7diz69++PO3fu\nICQkBOnp6QAAe3t7BASU3szfu3dvBAcHIzg4GC+X8w9cKysrtGrVCnFxcUrvxcXFwcjICN7e3gCA\njIwMfP7557h8+TL69euHcePGoX79+vjxxx8RHR39nz9LMSnIzcXZ7b+j/1efw9jMDM6dfdF2gD8S\n1pdfGBiZmgIA4n7bqLC88Nkz/F/ENvSZ/gFMLCxgZd8Y3Sa9ifN79mv8HEjzzM3NMXhAf8z+OhQ5\nOTk4ERePXXv3IShwhNK6G8O34LPPPsPB3TvQvJmTwntfz/4c186dxrm4GJyLi8Eb/q/hrXFj8OuK\n5do5ESIBMVeZqxUpyM3F3zt249U5pdnq1Lkj3N/oh9Mbwivcpu2wgTi1dpPCsv2z5+JbNy8sat8V\ni9p3xcXd+xC/ei3Cx0/V9CmQhpmbm2Nwf398OXd+aa7GJ2BX1H4EBSg/Xm/jlkh8/tVcHDx4UClX\nzczMMGLwAHy/JAxZWdlIvncfq35bD/++HJlGFdPJ+BhnZ2dMmjRJ/nN2djaOHDmC0aNHIyUlBdu2\nbcPo0aPRv39/+Trt2rXDF198gejoaPTp0weZmZnYu3cvfH198dFHH8nX27p1q0IAA4CjoyOWLFkC\nqfTf2tvPzw8ffvghDh8+jKFDh8LKygrt27dHeHg4WrZsKe99rUinTp2wevVq3L9/H40bl/YKyWQy\nxMfHw9PTE3XqlA5VCA8Ph1QqxXfffQdzc3MAQJ8+fRAaGorNmzfDz89PoV1VWSHLqnolEXj++aJf\nXqx8+vgPD++t8L0l/zyLFADG/LIMY35Z9t8bRzq3/OdVGD9+PGydXNCgQQP89NNPcO/gi5iYGLz2\n2mvI/mf42ayvQ/H48WN4+/WUbzt69GisWLECdc0s8fzAtToW9WBuZQ1rByftnozIiLfvtmZjrlY/\nVxcWP1F5XX3nGfDv80Wn/51Q4Xp1LC3xzp+7Vdpn9w9YjIrB8lW/YPz48WjUwv2fXF2B1h27KeXq\nF3O/w+P0DHnnEPBvrgJA2M+rMWnSJNi7tYGVlRXeeustTJj6vqiv7GmDmD89nRSjr7zyisLPbm5u\nOHnyJHJzc3Hy5EnIZDJ07NgRT58+la9jZ2cHKysrXLp0CX369MGFCxdQXFyMPn0Ue1tee+01pdA0\nMjKS/39JSQlycnJgamqKxo0bIykpqVrn4OvrizVr1iA2NhZDh5b+cr9+/TrS0tIwevRoAKUhevLk\nSXTp0gXFxcUK59OuXTucO3cODx48gP1zQwerMllSt+qVapEVsix+Ji9YkZOs6yZonLWpAXZuWqu4\nMPcJunm1QXbKPSC39B+XSRfPAWaW8p+fX/dFvy1fUuF7omNmqbFdizkwazLmavVzdZqB5v4+6KOF\nxU/4mbxgQUZi1SvpufqGwI51qxSWyZ6momtbN2TdT4LsaSoA4OZfpY9xkdSzkS8rWxcA6gLYvHIp\ngKX/7igrDWK/WUpSz0az+9fo3nVLJ8Vow4YNFX4um8kyJycHDx8+hEwmQ3BwcLnblgVPamrpl97O\nzk5pX2U9pWVKSkoQFRWFAwcOICUlRWGK6erOolmvXj24u7sjLi5OHpqxsbEwMTFB+/bt5W3NycnB\ngQMHcODAgQrPR53QJCKqydj7rRvMVcXzYa4SkZiIOVt1UoxWNHxGJpOhpKQEUqkUn332WbnrVCfk\nduzYgYiICPzvf//DiBEjYGFhAYlEgrVr1/6niW06deqElStXIjk5Gfb29khISICnpydM/7mXsWzf\nPXr0qHBShSZNmlT7+ERENY1UvHlZozFX/8VcJSKxEXO21rg51e3s7FBSUgI7OzvY2tpWuJ6NTenl\n8IcPHyr0CGdlZSHnhedfxcfHw93dHe+8847C8pycHNStW/0hnh07dsQvv/yC2NhYtG3bFo8fP8bY\nsWPl79erVw916tSBTCZDmzZtqn0cIiJ9IRFzYuop5ioRkX4Tc7bWuOeM+vj4QCKRIDIyUum9kpIS\n+Q3UrVu3hoGBgdIwnf37lWdMLa/HOC4uTj7jX5myntcXQ7cidevWRevWrREXF4e4uDiYmprKhxKV\nHdfb2xuxsbG4/8+DgJ/3/L0uRERiIJGo/yLNYq4SEek3MWdrjbwyOmLECISHh+PRo0fw8vKCqakp\nHj58iISEBAwePBi9evWClZUVXnvtNezZswffffcd2rZti9u3b+Ps2bOoW7euwthqLy8vREZGYvny\n5WjZsiXu3LmD48ePo1GjRgrHtrGxgYWFBQ4ePIg6derAxMQELi4ulfYkd+rUCStWrEB6ejo6dOgA\nY2NjhfdHjRqFixcvYsaMGejduzcaN26MrKws3LhxA0lJSVi2jLO7EpF46FMA1hbMVSIi/SbmbK1x\nxSgADB48GHZ2doiKisLWrVshkUjQsGFDeHt7KwzLGT16NExMTHDo0CGcP38eLVu2xKxZszB79myF\nmf4GDRqEvLw8nDhxArGxsWjWrBlmzJiBTZsUn6FlYGCA4OBgbNy4EatWrUJxcTGmTJlSaWj6+Phg\n1apVePbsGTp16qT0fv369TFv3jxs3boVcXFxyMzMRL169eDo6IgRI5Sfi0hEpM/EPMmCPmOuEhHp\nLzFnq0T2X2YaqIFycnLw5ptvIiAgAIMHD9Z1cwTHx5go4qNdlNWGR7uopbxHu9R2Gny0y4VmzdTe\npnU1HwVC2iH2XOVjTBTx0S7KasOjXdT14qNdajtNP9pFzNlaI6+MqqqgoEBp+M7evXsBAK1atdJF\nk4iIajUx997WBsxVIqKaR9PZWlhYiIiICMTExCA7OxtOTk4ICAiAh4eHStsfP34ce/fuRXJyMgwN\nDeHo6IigoCC0aNGiym31uhg9fvw4YmJi4OnpCRMTE1y+fFk+A5+bm5uum0dEVOuwFtVvzFUioppH\n09m6bNkyJCQkwN/fH3Z2doiOjkZoaCjmzJmDli1bVrpteHg4du3ahW7duqF3797Iz8/HrVu3kJmZ\nqdKx9boYbdq0KY4fP45du3YhNzcXVlZW6NevHwICAnTdNCKiWkmqhWpUlz24YsdcJSKqeTSZrYmJ\niYiNjcXYsWPh7+8PAPDz88O0adOwceNGzJkzp8Jtr169ih07dmDatGnw8fGp1vH1uhh1dnbG7Nmz\ndd0MIiL6hzaujOqyB1fsmKtERDWPJrM1Pj4eBgYG6NWrl3yZsbExevbsifDwcGRkZKB+/frlbhsV\nFQVnZ2f4+PigpKQEBQUF8kd6qUqvi1EiIqpddN2DS0REpC9SUlKUlpmbm8Pc3Fz+c1JSEuzt7ZWK\nyBYtWkAmk+HWrVsVFqMXLlxA586dsWnTJuzfvx95eXmwsbFBQEAAunXrplIbWYwSEZFgND3Jgq57\ncImIiLStutkaHBystGzo0KEYPny4/OfMzExYW1srrVeWpRkZGeXuOzs7G1lZWYiNjYVUKsWoUaNg\nYWGBP/74A0uXLoWJiYlKHb8sRomISDASafW2U6X3FtB9Dy4REZG2VTdbw8LClJa9mKsFBQUwNFQu\nCcueLV1QUFDuvvPy8gAAWVlZmDt3LlxcXACUPiv6vffeQ2RkJItRIiLSLk323gK678ElIiLStupm\nq62tbZXrGBsbo6ioSGl5YWGh/P2Ktis7RlkhCpQWsR07dsS+ffuQl5dX5QgkFqNERCSY6o7SVaX3\nFtB9Dy4REZG2afIOGCsrq3I7csuWVTTayMLCAkZGRrC0tCx3nzKZDLm5uSxGiYhIezTZewvovgeX\niIhI2zQ5H4OTkxOioqKUMvD69esASh/5VR6pVAonJyekp6crvff48WNIpVJYWFhUefxqjkAmIiJS\nJpGo/1KHpntwiYiIahpNZquvry+Ki4tx6NAh+bLCwkJER0fD1dVVfmtMWloa7t27p7Tt48ePcf78\nefmy3NxcxMXFwdXVtcIO4ufxyigREQlGkw/mBnTfg0tERKRtmsxWFxcX+Pr6YsOGDcjIyICdnR2O\nHj2K1NRUTJkyRb5eWFgYLl26hC1btsiX9enTB4cPH8bChQvh7+8Pc3NzHD58GDk5OQgMDFTp+Lwy\nSkREgtH0lVFd9+ASERFpm6azNTg4GP369cOxY8fw66+/oqioCDNnzoSbm1ul25mYmODLL79Ehw4d\nsG/fPmzatAlmZmb44osvqty2DK+MEhGRYDT9nFFd9+ASERFpm6az1djYGEFBQQgKCqpwnZCQkHKX\nW1lZ4d133632sVmMEhGRYDSclwBKe3AjIiJw7Ngx5OTkwNHRUa0e3PXr12Pfvn0oKChA8+bN1erB\nJSIi0jZtZKuusBglIiLBaCMwddmDS0REpG0sRomIiFQgkYo4MYmIiHRAzNnKYpSIiAQj5t5bIiIi\nXRBztnI2XSIiIiIiItI6XhklIiLBaPo5o0RERLWNmLOVxSgREQlGxHlJRESkE2LOVhajREQkGE0/\nC42IiKi2EXO2shglIiLBiDgviYiIdELM2cpilIiIBCPm3lsiIiJdEHO2shglIiLBiDgviYiIdELM\n2cpilIiIBCPm3lsiIiJdEHO2shglIiLBSPj0aiIiIkGJOVsrLEanTZumchW+YMECwRpERET6S8y9\nt0JgthIRkbrEnK0VFqNeXl7abAcREYmBVLyBKQRmKxERqU3E2VphMTpy5EhttoOIiMRAxL23QmC2\nEhGR2kScrWqNQL5//z7OnDmDgoICAIBMJtNIo4iISD9JJBK1X7Uds5WIiCoj5mxVaQKjnJwc/PDD\nDzh//jwA4Mcff0SjRo2wfPly1K1bF2PGjNFoI4mISE+IeCiR0JitRESkEhFnq0pXRjds2ICCggIs\nXrwYJiYm8uWdOnXCX3/9pbHGERGRnpFI1H/VUsxWIiJSiYizVaVi9Ny5cwgKCkLjxo0Vljdu3Bhp\naWkaaRgREZGYMVuJiKi2U2mYblZWFurVq6e0PD8/X/AGERGR/pKIeCiR0JitRESkCjFnq0pXRps2\nbSq/p+V5x48fh7Ozs+CNIiIiPSXioURCY7YSEZFKRJytKl0ZHTJkCBYvXozMzEyUlJQgLi4OycnJ\nOHHiBGbNmqXpNhIRkZ4Qc++t0JitRESkCjFnq0pXRtu3b4+PPvoIZ8+eRXFxMcLDw3Hv3j3MmDED\n7u7umm4jERHpCxH33gqN2UpERCoRcbaqdGUUADw9PeHp6Qmg9Blo+vT8GiIi0hIR995qArOViIiq\nJOJsVbkYBYBr167h7t27AIAmTZqgZcuWGmkUERHpJxZT6mO2EhFRZcScrSoVo+np6Vi8eDGuXbsG\nU1NTAEBeXh5atmyJDz74AA0aNNBoI4mISE+IuPdWaMxWIiJSiYizVaVi9Oeff0ZBQQG+//57ODo6\nAgDu3LmDn376CStXrsTMmTM12kgiItITIu69FRqzlYiIVCLibFVpAqOLFy/irbfekoclADg6OmLC\nhAm4ePGixhpHRET6RSJV/1VbMVuJiEgVYs5Wla6MWlpawtBQeVUDA4NyH9hNRES1lIh7b4XGbCUi\nIpWIOFtVqpuHDRuGtWvXIj09Xb4sPT0dGzZswPDhwzXWOCIi0i8SqUTtV23FbCUiIlWIOVsrvDI6\nbdo0hZmbUlJSMHXqVFhbWwMoDUwjIyM8efIEPXr00HhDiYhID4i491YIzFYiIlKbiLO1wmLUy8tL\nm+0gIiIx0KPeWF1gthIRkdpEnK0VFqMjR47UZjuIiIhEj9lKRET0L5UmMCIiIlKFmB/MTUREpAti\nzlaVitGSkhL8/vvvOHHiBFJTU1FUVKTw/oYNGzTSOCIi0jMiHkokNGYrERGpRMTZqtJsupGRkdi7\ndy+6dOmCwsJCvPHGG/D19YWxsTECAwM13UYiItIXEon6r1qK2UpERCoRcbaqdGU0JiYGkyZNgre3\nN7Zv3w4/Pz/Y2dlh9+7duHXrloabSERE+kLMQ4mExmwlIiJViDlbVboympGRgWbNmgEATExM8OzZ\nMwCAt7c3Tp8+rbnWERGRfpFK1H/VUsxWIiJSiYizVaVitH79+nj69CkAoGHDhrh69SoA4P79+6Ku\n1ImISD0SiUTtV23FbCUiIlWIOVtVGqbr7u6O06dPo3nz5ujRowfWrVuHhIQE3Lx5E506ddJ0G4mI\nSF/oUW+srjFbiYhIJSLOVpWK0bfeegslJSUAgL59+6JOnTq4fPky2rdvj9dee02jDSQiIj2ihd7Y\nwsJCREREICYmBtnZ2XByckJAQAA8PDwq3W7Lli2IjIxUWm5paYlVq1ZpqrkVYrYSEZFKNJytusxV\nlYpRAwMDGBgYyH/28/ODn5+fSgcgIqLaQ6KF3ttly5YhISEB/v7+sLOzQ3R0NEJDQzFnzhy0bNmy\nyu0nTZoEY2Nj+c/P/782MVuJiEgVms5WXeZqhcVodna2yjuxsLBQeV0iIhIxDffeJiYmIjY2FmPH\njoW/vz+A0iJu2rRp2LhxI+bMmVPlPjp16gRzc3ONtrMizFYiIlKbBrNV17laYTE6YcIElXcSERFR\nrYMTEZHIaLj3Nj4+HgYGBujVq5d8mbGxMXr27Inw8HBkZGSgfv36le5DJpMhNzcXderU0fokD8xW\nIiJSmwazVde5WmExOnPmTLV2REREVN3iLiUlRWmZubm5Uk9rUlIS7O3tYWpqqrC8RYsWkMlkuHXr\nVpWhOWXKFOTl5aFOnTro2LEjxowZo7WrkMxWIiJSlyazVde5WmEx2q5dO5V2QNr10+Mrum5CjcPP\nRNGXVk66bkKNMqcgg5/JC+YUZGhu59XsvQ0ODlZaNnToUAwfPlxhWWZmJqytrZXWLQvKjIyKz83C\nwgKvvvoqWrZsCUNDQ1y4cAEHDx5EUlISQkNDYWRkVK22q4PZWvMszLqj6ybUOPxMFE02d9B1E2qc\nFbIsvGPZXNfNqDFWyLI0ewANZquuc1WlCYyIiIg0KSwsTGlZefefFBQUwNBQObrKAq+goKDCY/Tr\n10/hZ19fXzRp0gS//PILjh49it69e6vbbCIiohpLlWzVda5Kq1yDiIhIVRKJ+i8Atra2Sq/yilFj\nY2MUFRUpLS8sLJS/r45XXnkFJiYmuHDhQjVOloiISAs0mK26zlVeGSUiIuFoeEIgKyurcocMlS2r\n6r6WF0mlUlhbW6s1yy0REZFWaTBbdZ2rvDJKRETCqWbvraqcnJxw79495OXlKSy/fv06AKBp06Zq\n7a+oqAhpaWmoV6+eWtsRERFpjQazVde5ymKUiIiEI5Wq/1KDr68viouLcejQIfmywsJCREdHw9XV\nVT4JQ1paGu7du6ew7dOnT5X29/vvv6OwsBBt27atxskSERFpgQazVde5qvIw3ePHj2Pfvn149OgR\nQkNDYWtriz179sDOzg4dOnRQdTdERCRmGh6m6+LiAl9fX2zYsAEZGRmws7PD0aNHkZqaiilTpsjX\nCwsLw6VLl7Blyxb5silTpqBz585wdHSEoaEhLl68iISEBLi5uaFr164abXdFmK1ERFQlDWarrnNV\npWI0Ojoav/76K1599VXcvn0bMpkMAGBmZoaoqCgGJhERldJwMQqUTlUfERGBY8eOIScnB46Ojpg5\ncybc3Nwq3a5r1664evUq4uPjUVRUBBsbGwwZMgSDBg2CgYGBxtv9ImYrERGpRMPZqstcVakY3bNn\nDyZNmoQuXbpg37598uXNmzfH5s2bVToQERHVAlooRo2NjREUFISgoKAK1wkJCVFaNnnyZA22Sn3M\nViIiUomGs1WXuapSMfrw4UO0bNlSaXmdOnWQm5v7nxtBREQioeY9oLUZs5WIiFQi4mxV6cwsLS2R\nkpKitPz69euwsbERvFFERKSnNDybrpgwW4mISCUizlaVitGuXbti/fr1SE1NhUQiQXFxMS5cuICN\nGzeie/fumm4jERHpCxEHptCYrUREpBIRZ6tKw3SHDRuGR48eITg4GADw0UcfQSaToUuXLhgwYIBG\nG0hERHpEjwJQ15itRESkEhFnq0rFqKGhIT744AMMGzYMN27cgEwmg7OzMxwcHDTdPiIi0icivq9F\naMxWIiJSiYizVeXnjAKAvb097O3tNdUWIiLSdyLuvdUUZisREVVKxNmqUjG6evXqSt+fOHGiII0h\nIiKqLZitRERU26lUjN65c0fh5+LiYty/fx8A4OjoKHyriIhIP4m491ZozFYiIlKJiLNVpWL0q6++\nUlqWn5+PZcuWwdPTU/BGERGRnhJxYAqN2UpERCoRcbZW+25YExMTDB06FJGRkUK2h4iI9JhEKlX7\nRf9ithIR0YvEnK1qTWD0ooKCAmRnZwvVFiIi0nci7r3VFmYrEREpEHG2qlSMJiQkKPwsk8mQkZGB\n/fv3w83NTSMNIyIiPSTiwBQas5WIiFQi4mxVqRhdtGiR0jIzMzN4eHhg3LhxQreJiIj0lYgDU2jM\nViIiUomIs1WlYnTDhg0KP0skEhga/qcRvkREJEZ6dJ+KrjFbiYhIJSLO1irPrKioCKGhoUhNTYWR\nkRGMjIwYlkREVD6JRP1XLcRsJSIilYk4W6tMPkNDQ9y+fRtSEVfkREQkED0KQF1ithIRkcpEnK0q\npaCvry/i4+M13RYiItJ3Iu69FRqzlYiIVCLibFVpTJC1tTV27tyJq1evwtnZGaampgrvv/766xpp\nHBER6Rle6VMZs5WIiFQi4mxVqRg9cOAAjI2NcePGDdy4cUPhPYlEwsAkIqJSetQbq2vMViIiUomI\ns1WlYnTlypWabgcRcDQNpAAAIABJREFUEYmBiANTaMxWIiJSiYiztdJrvsHBwcjKytJWW4iISN9J\npeq/ahlmKxERqUXE2VppS1NTU1FSUqKtthAREYkes5WIiKgUH2pGRETCEfFQIiIiIp0QcbZWWYxm\nZmaiuLi40nWsra0FaxAREekxEQemkJitRESkMhFna5XF6PTp06vcSUREhCCNISIiPSfiwBQSs5WI\niFQm4mytshidNm0aLCwstNEWIiLSd3o0aYIuMVuJiEhlIs7WKotRV1dXWFpaaqMtRESk70Tceysk\nZisREalMxNnKCYyIiEg4Ig5MIiIinRBxtrIYJSIi4UjEO5SIiIhIJ0ScrZUWo5w8gYiI1CIVb++t\nUJitRESkFhFnK6+MEhGRcETce0tERKQTIs5WFqNERCQcEd/XQkREpBMizlYWo0REJBwRTz9PRESk\nEyLOVhajREQkHBH33hIREemEiLOVxSgREQlHxPe1EBER6YSIs1W8Z0ZEREREREQ1Fq+MEhGRcEQ8\nlIiIiEgnRJytLEaJiEg4Ip5kgYiISCdEnK0sRomISDgi7r0lIiLSCRFnK4tRIiISjhYmWSgsLERE\nRARiYmKQnZ0NJycnBAQEwMPDQ639zJs3D2fPnkW/fv0wbtw4zTSWiIjov+IERkRERCqQStR/qWnZ\nsmXYu3cvunXrhjfffBMSiQShoaG4du2ayvs4c+YMLl26pPaxiYiItE4L2aorLEaJiEg4Eqn6LzUk\nJiYiNjYWo0ePxujRo9G7d2/Mnj0bDRs2xMaNG1XaR1FREdauXYsBAwZU5wyJiIi0S8PZWlhYiA0b\nNuDtt9/GqFGj8Pnnn+Pvv/9Wu5nz5s3D8OHD8dtvv6m8DYtRIiISjkSi/ksN8fHxMDAwQK9eveTL\njI2N0bNnT1y5cgUZGRlV7iMqKgoFBQXo37+/2qdHRESkdRrOVl2OOGIxSkREwqlm721KSorSKycn\nR2n3SUlJsLe3h6mpqcLyFi1aQCaT4datW5U2LzMzE9u2bUNgYCBMTEwEO20iIiKN0eCVUV2POOIE\nRkREJJxq3qcSHBystGzo0KEYPny4wrLMzExYW1srrVu/fn0AqPLK6KZNm9C4cWN069atWu0kIiLS\nOg3eA1rZiKPw8HBkZGTIM7Yiz4842rJli1rHZzFKRETCqeb082FhYUrLzM3NlZYVFBTA0FA5uoyM\njOTvVyQxMRFHjx7FnDlzIBHxNPlERCQy1cyslJQUpWXm5uYK+arKiKPKitGyEUcTJkyo1ogjFqNE\nRCScak4/b2trq9J6xsbGKCoqUlpeWFgof788MpkMv/76Kzp27Ag3N7dqtZGIiEgnqpmtqow60vWI\nIxajREQkHA1PJ29lZVVuMJYtq6j39uTJk0hMTERgYKBST3Fubi5SUlJgZWVVYTFLRESkM9XMVlVG\nHel6xBGLUSIiEo6GH8zt5OSEqKgo5OXlKQwpun79OgCgadOm5W6XlpYGmUyGr776Sum96OhoREdH\n47PPPkO7du0003AiIqLq0uCoI12POGIxSkREesPX1xe7d+/GoUOH4O/vD6A0MKOjo+Hq6iofapSW\nlob8/HzY29sDALy8vGBjY6O0vwULFqB9+/bo2bMnmjVrpr0TISIiqgF0PeKIxSgREQlHwxMDubi4\nwNfXFxs2bEBGRgbs7Oxw9OhRpKamYsqUKfL1wsLCcOnSJfmsfnZ2drCzsyt3n3Z2dvDx8dFou4mI\niKpNg9mq6xFHLEaJiEg4Gh6mC5ROyBAREYFjx44hJycHjo6OmDlzJicmIiIicdJgtup6xBGLUapR\n0jMyMPG9j3Eg+hgaWlsj9IsZGDl0kNJ63y/9CevCI3E7+R4aWtfHO+PH4JN33wEA3Em+B/fO/1NY\nPycnF99/9QWmTX1bK+dBmlWnvhUGrFwK597/Q25aOv784iv8HR6ptN7o37cCAD5LvwsAMDA2xuNr\niVjevgsAYNyB32Hr/jIMTIyReesODs8JxdXd+7R3ImKk4QmMgNL7V4KCghAUFFThOiEhISrtS93n\noRHpo/T0DEyYEowDh46gYQNrzJvzJUaOGKa03veLf8Tazf/f3p3HR1Xf+x9/TZbJZCE7SUhiEpJA\nFpKAAQMEQUUUUEQFS9kXa22vov1Za1t6r71grVJtvWhxueq1akFlUZF937MAsu9h30JIAgkhgezz\n+4ObuYQEBDvJJDPv5+ORx8OcOTPzPeE47/l8z/f7PbM4ceIEgQH+PPPzp3jphectj0clJHMuvwBn\nZ2cA0runsXzBt812HNJ0PPz8GPs/75LwYF9KC88zb9Jktnw5p8F+Exd/TWzvdACmXTqLi9HIuYOH\n+FNKDwD+fGwPbYKDqK2pAeBo5ibe6f9Y8x2IvWrCbLX1iCMVo9KiTPztf2A0Gsnbv4Mde/YyaPg4\nOicl0ik+rt5+ZrOZz96bRufe/Ti8NZP+T4zkjrBQhg95lIjwMC6dzLHse+zESTp0u5uhjzzU3Icj\nTeThd/5KTWUlb4bHEdI5mVHfzSJv1x4K9h2ot9+MwT9hSmURr/nfAcD4FQs4tna95fElL06iYN8B\namtqCLurK+OWfss7ne6iNO9csx6PXWmGK6MicnuefeE3GI1Gzh3LYceu3Tw89Kd0Tk6iU2JCvf3M\nZjOff/45KbGRHDl6jAcHP84d4WEM/8lQyz4L5nxFv773Nu8BSJMb8e7fqK6s5LfBMYR3SWHiojmc\n3rmbs9fl6vSHrp4LH5gv8f/atOPXaxZzYPW6evu898gwDqxa21xNdwxNnK22HHHksN8a1q5dy7Bh\nwxq9Gey/sq/8eGVll/l6wWJemfQSXl6e3N0jjcEDHuCfs75usO9vn3+G1M7JuLi4ENchhsEDHyRj\n05ZGX/fzWXPpk96dqIg7mvoQpBm4eniQ8PgjrJ78GpVlZZzMzObgwiV0HvXTmz7PN/IOIu/uyc4Z\nX1m2ndu919J7i9mMk6srPuFhTdl8+2cw3P6P2A1la8tTVlbG19/N508v/zteXl7cnd6TwQ8N4J9f\nzmqw729//StSU1OvZmvHDjw66CEysrNt0GppTkYPD+4c+ijzX36VirIyjmRksXP+YrqPGXHT5wVE\nRhDbO51Nn3/ZTC11YE2crXUjjj788ENmzpzJ66+/TkpKSr19Jk+efEujiWbPns348eNv+b0dthht\nzPLly1m7dq2tm+Gwco4cxcXFmY6x0ZZtKUmJ7DuYc5NnXe3J3Zi9ucHV07rH/jnra8Y2MhxJWqeA\njjHUVldz/tARy7a8XXsISrx5713n0cM5sTGL4hOn6m0f+e1X/EfJWZ7OXMXxdRvJ3bq9SdrtMJyc\nbv9H7Jqy1bZyDh3GxcWFjh1iLds6Jyezd//+mz7PbDazISOLTgn1r56OevLntI2M4cFHHmfnrt1N\n0mZpXsEdY6mtrib/0GHLtjM79xDaKeEmz4LuY0dweEMm50+crLf9yZn/w5v5x3h+2TzCUpKapM0O\nx46z1WGH6fbp04f09HTLDV0BVqxYgaenJ/fee+8P7ivWV1pWhnebNvW2+Xi34VJp6U2fN/kvf6O2\ntpYJI4c1eGxj9mbOFRTwxOCHrdpWsR2jpxcVJZfqbau4WILRy+umz+s8ajjrX/9rg+1fPD4cJxcX\nou+/l7bxHTGbzVZtr8PRlU6HpmxteRrNVh/vH87WP79+NVvHjLJsm/nJR6R26YzZbObtdz+g/6ND\nObB9M76+vk3Sdmkebl5eXLkuV69cLMHU5ua52mPsCJa8+ma9bZ+MeoqT23aAwUDfXz3D88vmMTm+\nK1cuXrR6ux2KHWdr6ymbrczJyQmj0YjhFv5xb2df+fG8PD0puVT/w7DkUiltblJkTJ8+nX/O+pqF\nX36Gm5tbg8c/+2oOQwc9hJeXp9XbK7ZRWVaKm3f9L1Zu3m2ovMkXq4j0HniFBLHvm/mNPl5bXc3h\nZSuJ6XcfcYMGWrW9DsfgdPs/YjeUrS1Po9laUnLzbP3gQz7/4isWfTO7Xrb26tkDd3d3PDw8mPTS\nr/H19WFDZlaTtV2aR0VpKe7X5arJuw3ll27eYeEdEsy2ufPqbTuSmU1VeTlVV66wbOrfuFJ80bLg\nkfwL7DhbW8yV0dmzZzN37lymTZvGl19+yc6dOzEajdx7770MHz4cF5erTa2pqeGbb75h3bp1XLhw\ngYCAAO655x6GDBmC0zWXpHft2sWcOXM4deoUNTU1+Pv70717d0aOHAlcnavy3nvvMX36dIKCgnj2\n2WcpKCgAYNiwq1fYEhMTmTx5coN9X3/9dfLy8nj77bcbHMcLL7yAv78/L7/8MgC1tbUsXLiQNWvW\ncO7cOby8vOjRowcjR46sdy8fgY4x0VRX13DoyFE6xFwdqrtrzz4S4zo2uv8nM79i6l/+i3XzZxMe\nFtrg8StXrjD3u0V88/nHTdpuaV7nc47g5OKCf2w0Fw4fBSA4JYn86xZZuFaXMcPZP28hlWVlN31t\nJxcX/KOjrNlcx6PCokVRtkrHDrFUV1dz6PAROsTGALBz954Gw2/rfPLJJ0z92zTWL19MeNjN59Ab\nDAaNJrED53IO4+TiQlBsDPmHr06BCe+cRO7emw/l3vHNfCp+IFfNZrM6nKzBjv+GLa5sfuutt6it\nrWXkyJEkJSUxf/58Pv30U8vjH3zwAXPmzCE2NpaxY8cSExPD7Nmz+fDDDy37nDp1iqlTp+Li4sLw\n4cMZM2YMXbt25eDBgzd833HjxhEYGEhYWBgTJ05k4sSJDBkypNF9e/bsydmzZzl27Fi97SdPnuTM\nmTP07NmzXnu/+uorEhMTmTBhAnfffTerVq3ijTfe0Af4dTw9PRgyaCD/OfVvlJVdJmPTFr5bspwx\nPx3aYN+Zc77h31/9CytWrCA6qvGb8X67aCl+vj7cpx45u1J1+TL75y2k73/+AVcPD+7o2Z34Rx5i\n58yGi3HU6fTE4+z4/It62wLjOhDbvx8uJhNOLi6kjBxGZO90jq/PaOpDsG92PK+lNVO2Oi5PT0+G\nPPoIf/zTa5SVlZGRlc13i5YwZkTDRd9mfjWbP/zhD6xY8C3R7aPqPXby1CkysrKprKykvLycN//r\nHQrPn6dXjx7NcyDSZCovX2b7N/N55JV/x+jhQUx6Dzo/+jCb/tn4wkSu/9vhk/XpzHrb/e4IJya9\nB86urri4ufHAb36FV2AARzK0CNa/zI6ztcVcGa0TEhLCb37zGwAGDBiA0WhkxYoVDB48mMuXL7Nu\n3ToeeOABfv7zn1v28fDwYOXKlQwcOJDIyEh27dqFq6srL7/8cr0e3ZtJS0tjzpw5eHp60qdPnx/c\n98MPPyQrK6vezVwzMzNxdname/fuAOzfv5+1a9fy61//mh7XfFjHxsYybdo0du7cSZcuXW7r72Pw\nt++VPt/7+B88+eSTBMd3ISAggPc/+ICk9L5s2LCBgQMHUvq/QzFfnvoW5y8Ucdddd1meO3r0aD74\n4APL759/vYAx48bjFBDe7MdhS1Mqi2zdhGaTfE1HxcQdNx4qZvLxZvzyxofoXu+Xm9f98E5yY3bc\ne9uaKVtvwsPn1vdtpd7774948sknCYrqcDVb33+fTt16NMjW//jTa5w/f567+vS1PLcuWy9Vn+bf\nXniJI0eOYDKZ6NKlC0uWLCXgjh++qX1r94H50g/vZCfuGvF/Cz7+597G71JQ54XVi27pNf92/sS/\n1CbBrrO1xRWjAwYMaPD72rVr2bFjB2X/OxRg0KBB9fYZNGgQK1euZPv27URGRuLp6UlFRQU7duwg\nNTXV6m308PCgc+fOZGVlWYYmAWRnZ5OcnEyb/10oIDs7Gy8vLxITEykpKbHsl5CQgJOTE3v37r3t\nYtR84Yx1DqKF8gO+/eTdetvMF85wd6doLp08aDn+o1s3AleL82v/Jtf+99IvP2mwzRFMDtHKddea\nUlnEfxr9bN2MFqVJOyxa0TwVR6JsvYnL9r+wir/JmXlffFZ/4+WL9O6aQmn+Gcvf4NjeHVeL8+v/\nJpcv0ql9OLuyNzR8cQf4+/3S07E6tW/FB+ZL/NLQ5od3dBBN3mFhx9na4orRkJCQRn8vKCigrKwM\nZ2dngoODG+zj7OxsmZeSnp7OqlWrmDp1Kn5+fiQnJ5OWlsZdd91ltXHr6enp/P3vf+fIkSPExMRw\n/PhxcnNzefTRRy375OXlUVpaylNPPdXoa1wboiIidsGOe29bM2WriEgrZsfZ2uKKUWswGo1MmTKF\nvXv3sn37dnbs2MH69etJSUnhD3/4wy0PL7qZbt264erqSlZWFjExMWRnZ+Pi4kJaWppln9raWvz8\n/Hj22WcbfQ1/f/9/uR0iIi2KHffeOjplq4iIjdhxtra4YjQvL4/AwMB6vwO0bdsWDw8PampqOHfu\nHO3atau3T01NDW3btrVsc3JyIjk5meTkZMaOHct3333HzJkz2bt3L8nJyY2+9+307Lq7u9OlSxey\nsrIYPXo0WVlZpKSk4On5f7cQCQ4OZu/evSQkJOg+aiIiYjPKVhERaYlaXJm9dOnSBr8bDAY6d+7M\nnXfeCcCiRfUnTC9evBjAMofl0qWG47ajoqIAqKqquuF7u7m5WebO3Ir09HQKCgpYsWIFZ8+eJT29\n/qqtPXr0oLq6mnnz5jV4blVVFZcvX77l9xIRaRWcDLf/I01O2Soi0orZcba2yCujb775JikpKezf\nv5/MzEz69etnmctyzz33sHz5csrKyoiPj+fAgQNkZGTQt29fIiIiAPj666/Zv38/d955J23btqWk\npITly5cTEBBAfHz8Dd87OjqapUuX8vXXXxMSEoKPjw9JSTdeDKZr164YjUZmzJiBq6trvZVdAZKS\nkujbty9z5szh6NGjJCcnYzAYOHv2LFlZWTz33HOkpKRY4a8mItJC2PFQotZM2Soi0orZcba2uGL0\n17/+NV988QUzZ87E1dWVQYMG1VtV75e//CVBQUGsXbuW7Oxs/P39GTZsWL37lnXr1o38/HzWrFnD\npUuXaNOmDYmJiQwbNgwPD48bvveQIUPIz89n/vz5XLlyhcTExJsGpslkIjU1lezsbO666y7c3d0b\n7POLX/yC6OhoVq1aZTmmoKAg7r//fkuPsoiI3bDjRRZaM2WriEgrZsfZajC3kLtDz549m7lz5/KP\nf/yj3twQqc/RblPyQ66/tYvo1i7X061dGmrKW7vUZHxz289x7jXkh3eSH0XZegsc4NYkt6WxW7s4\nON3apSHd2qW+pr61iz1na4u7MioiIq2XtW7xISIiIlfZc7aqGBUREeux43ktIiIiNmHH2apiVERE\nrMeOA1NERMQm7DhbW0wxOmzYMIYNG2brZoiIyL+iFS0n7wiUrSIidsCOs7XFFKMiImIH7Lj3VkRE\nxCbsOFtVjIqIiPXY8SILIiIiNmHH2apiVERErMeOe29FRERswo6zVcWoiIhYjx333oqIiNiEHWer\nilEREbEeO+69FRERsQk7zlb7PTIRERERERFpsXRlVERErMeOl58XERGxCTvOVhWjIiJiPXY8lEhE\nRMQm7DhbVYyKiIj12PEiCyIiIjZhx9mqYlRERKzHjntvRUREbMKOs1XFqIiIWI8d996KiIjYhB1n\nq4pRERGxHjvuvRUREbEJO85WFaMiImI9TvYbmCIiIjZhx9mqYlRERKzGYMdDiURERGzBnrNVxaiI\niFiPHQ8lEhERsQk7zlYVoyIiYj123HsrIiJiE3acrSpGRUTEeuy491ZERMQm7DhbVYyKiIj12HHv\nrYiIiE3YcbaqGBUREetphhX/qqqqmDVrFhs2bKC0tJSoqCiGDx9OcnLyTZ+3YcMGVq9ezenTpykr\nK8PPz49OnTrxk5/8hLZt2zZ5u0VERH4UO15N136PTERE7NK7777LokWL6N27NxMmTMBgMPDaa6+R\nk5Nz0+edOHECf39/HnnkEZ566inuueceduzYwaRJkyguLm6m1ouIiEgdXRkVERHraeKhRIcPHyYz\nM5Nx48bx8MMPA9CnTx9efPFFZs6cyZQpU2743NGjRzfY1q1bN37/+9+zfv16Bg8e3GTtFhER+dHs\neJiuroyKiIj1GJxu/+c2ZGdn4+zszP3332/ZZjQa6du3LwcOHKCoqOi2Xq9ueO7ly5dv63kiIiLN\npomz1ZZ0ZVRERKznR/be5ufnN9jm6emJp6dnvW3Hjh0jLCwMk8lUb3tsbCxms5njx4/j5+d30/cq\nLS2lpqaGwsJC5s6dC0BSUtKPareIiEiTa+Iro7Zci0HFqIiIWNGPC8yJEyc22PbEE08wbNiwetuK\ni4vx9/dvsG9dAXorV0Z/9atfcenSJQDatGnDk08+qWJURERasKYtRt999102bdrEww8/TEhICGvX\nruW1115jypQpdOzY8YbPq1uL4c4778TLy4vCwkJWrlzJtm3b+Otf/4qvr+8PvreKURERsZ4f2Xs7\nffr0BtuuvyoKUFlZiYtLw+hydXW1PP5DXnzxRSoqKjhz5gwbNmygvLz8R7RYRESkmTThlVFbr8Wg\nYlRERKznRwZmUFDQLe1nNBqprq5usL2qqsry+A9JTEwE4M477yQtLY0XX3wRk8nEgAEDbqPFIiIi\nzaQJi9GbrcXw1VdfUVRU9IPTX651u2sxqBgVEREratqhRL6+vo0Oxa3bdjuBCVeL4OjoaDZs2KBi\nVEREWqimW4/B1msxqBgVERHraeJFFqKioli8eDHl5eX1gvPQoUMAREZG3vZrVlZWaqiuiIi0XD8y\nW29lPQZbr8WgYlRERKyniW+F1qNHDxYsWMCqVassc1uqqqpYu3YtcXFxlkAtLCykoqKCsLAwy3NL\nSkrw9vau93pHjx7l+PHjpKenN23DRUREfqwfma23sh6DrddiUDEqIiJW1LTVaIcOHejRowczZsyg\nqKiIkJAQ1q1bR0FBAc8884xlv+nTp7Nv3z5mz55t2fbMM8/Qs2dPIiIiMJlMnDp1ijVr1mAymRg6\ndGiTtltEROTHa7r1GGy9FoOKURERsZ4mHqYLV4cdzZo1i/Xr11NWVkZERASTJk0iPj7+ps/r378/\nu3fvZsuWLVRUVODn50fPnj154oknbnkBJRERkWbXhNlq67UYVIyKiIj1NEMxajQaGTNmDGPGjLnh\nPpMnT26w7Wb7i4iItFhNmK22XovB6bZfXURE5IYMP+JHREREbqzpsrVHjx7U1NSwatUqy7YbrcVw\n5syZes8tKSlp8Hp1azFER0ff0vvryqiIiFhPM1wZFRERcShNmK22XotBxaiIiIiIiIiDsuVaDCpG\nRUTEinRlVERExLqaNlttuRaDilEREbEeDdMVERGxLjvOVhWjIiJiPXYcmCIiIjZhx9mqYlRERKzI\nfgNTRETENuw3W1WMioiI1RjsuPdWRETEFuw5W1WMioiI9dhxYIqIiNiEHWerilEREbEi+w1MERER\n27DfbFUxKiIi1mPHvbciIiI2YcfZqmJURESsx44DU0RExCbsOFtVjIqIiBXZb2CKiIjYhv1mq4pR\nERGxHjvuvRUREbEJO85WFaMiImI99puXIiIitmHH2apiVERErMiOE1NERMQm7DdbnWzdABERERER\nEXE8ujIqIiLWY8fzWkRERGzCjrNVxaiIiFiPHQemiIiITdhxtqoYFRERK7LfwBQREbEN+81WFaMi\nImI9dtx7KyIiYhN2nK0qRkVExHrsODBFRERswo6zVcWoiIhYkf0GpoiIiG3Yb7aqGBUREeux495b\nERERm7DjbDWYzWazrRshIiIiIiIijsXJ1g0QERERERERx6NiVERERERERJqdilERERERERFpdipG\nRUREREREpNmpGBUREREREZFmp2JUREREREREmp2KUREREREREWl2KkZFRERERESk2akYFRERERER\nkWanYlRERERERESanYpRERERERERaXYqRkVERERERKTZqRgVERERERGRZqdiVERERERERJqdilFx\neGaz2dZNkBaotrbW1k0QEWmVlKtyI8pWuZ6LrRsgYks1NTU4OzsDUF5ejtlsxtnZGaPRiNlsxmAw\n2LiFYgvXnhfr1q2jsLCQiooK0tPTCQkJwWQy6fwQEWmEclVuRNkqjTGY1X0lDqq2thYnp6uDAz77\n7DOOHj1KcXExAQEBDBs2jPj4eBu3UGzh2vNi6tSpHDp0CAAnJyeqqqq47777eOihh2jbtq0tmyki\n0uIoV+VGlK1yI86TJ0+ebOtGiNhCXc/b66+/zrZt24iKiiIkJITz588zb948/Pz8iIiIsHx4imOo\nOy8+/PBD9uzZw9ixYxk+fDgjRoxgx44dfP/990RHRxMeHq7eWxGRayhX5UaUrXIjGqYrDm3x4sUc\nP36cp59+mq5du+Lq6sratWt5//33OXv2LDU1Nbi46H8TR3PhwgUOHjzI/fffT1paGiaTiZycHI4d\nO0avXr2Ij4/HyclJw4lERK6jXJUbUbZKY9Q1JQ7t5MmTtG3bluTkZFxdXdm3bx8ff/wx99xzDwMG\nDMDNzc2yrybd26/r/20vX75MQUEBsbGxmEwm9uzZwyuvvELXrl0ZPnw4AQEBAOTm5lJTU2OLJouI\ntEjKVamjbJVboWJUHEZjH2z5+fl4eXnh6enJvn37eP3110lLS2PEiBEEBgYCsGjRIs6ePathRXbq\n2nksZWVlABiNRgwGA5cuXeLMmTNMnTqVtLQ0xowZg5+fHwBr167l3Xff5eLFizZru4iILSlX5UaU\nrXKrNE5CHEbdCm6bNm3izjvvxGg0Eh8fz6ZNm1i2bBkzZswgLS2NUaNGWT4UDx8+zKJFi3B2diYk\nJETDRuxQXVi+8847ODs78/TTTxMUFETPnj357LPPqKys5O677+anP/2p5bwoKChg3759uLq62rLp\nIiI2pVyVG1G2yq1Sl5Q4lKVLl/LWW29RWloKQOfOnTl37hyffPIJnTp1YsKECfj7+wNQVFTExo0b\ncXNzIzo6WoFuVjI9AAAeFUlEQVRp54qLi8nJybH09KelpREQEIDRaKRLly6W4UMFBQWsXr2aLVu2\ncN9991nOFxERR6RclZtRtsoP0Wq64lCMRiNbtmzhwoULpKSkEBwcTHBwMJs2bcLPzw9/f39CQ0M5\nfPgw69atY8mSJQwdOpTu3bvbuunSRGprazEYDERGRrJ48WIqKytJSUmhXbt2mEwmTpw4wcqVKyko\nKGDbtm2sWbOGzMxMhgwZwoABAwC02IKIOCzlqjRG2Sq3SsWoOBQvLy+OHTvG/v37SU1Nxdvbm4iI\nCEJCQli9ejWZmZksW7aM1atXc/r0aYYMGcLDDz8M6EPRXtQFJFz9N60bSuTq6sqpU6c4dOgQ8fHx\n+Pr6EhUVRUREBF5eXuzatYu8vDyCg4MZPHgw/fv3t7ye5j2JiKNSrgooW+XHM5jNZrOtGyFiTbW1\ntZjNZstcljrV1dW4uLiQn5/P7373O3r37s2TTz5pefz48eMcPnyY48ePExkZSVhYGImJiZbX1Idi\n65STk0NsbGyD5eKLioos81Tq7N27l1dffZUxY8bw0EMP1XustLQUFxcXnJycMBqNgM4LEXEMylW5\nnrJVrEVXRsUuZGVlsXz5clJTUzEYDJYPsdWrV1NVVUVAQEC9XrqLFy+yadMmOnToQGBgIGazGT8/\nP6Kjo0lNTSUmJoa2bdsC+lBszU6dOsWkSZM4deoU6enplrB87733+Oijj/Dw8MBkMuHt7Q1AUFAQ\neXl5rFu3ztLDD1d7ed3c3CyBaTAY6vX8iojYG+Wq3IiyVaxJxai0epWVlcycOZPMzEwSEhIICgoC\nYP369bz//vts27aNixcvEhgYiKenJy4uLri5ubF69Wp8fHxISkq66VAhDSFqvZycnHB1dSUwMJCE\nhATL9uLiYtzc3Jg/fz579uzh4sWLREVF4erqioeHB+vWrcPf35+4uLgGX5rqzgedFyJir5SrcjPK\nVrEmFaPS6jk7O5OUlERcXBypqamW7ZGRkcTFxeHs7Mzy5cvZsmULJ0+eJDo6msjISCorK1m4cCHd\nu3fHx8fHhkcgTcXNzY2OHTvSqVMnAJYsWUKHDh2IiYmhe/fudOjQgQsXLrB+/XqysrK4dOkSffr0\n4fTp02zfvp1+/fo1GJYmImLvlKtyM8pWsSYVo9Lq1dbW4uHhQXh4OHB1mMj27dvp2rUrwcHBJCcn\n07t3b/Lz89mzZw9LliyhtLQULy8viouLKS0tJSEhQR+MdsZsNmM2m3FxuXo75c2bN/P+++9z/vx5\ny7CzkJAQkpOT6dmzJ0eOHGHz5s2sXLkSPz8/Dhw4QG1tLcnJyTY+EhGR5qVclRtRtoq1qRiVVuf6\noT/XDvUoLy9n5cqVHDp0iOLiYlJSUnBycsLLy4vU1FRSU1O5fPkyW7Zs4cCBA5w7d46ysjLuvvtu\nTCaTrQ5JrKimpoba2lqcnZ0t50lBQQGxsbG4u7uzePFiCgsLLaFpNBrx9fWlV69etG/fnoqKCjIy\nMgC47777iIqKsuHRiIg0PeWq/BBlqzQVFaPS6hgMBqqqqjh79ixOTk6WsFuxYgWRkZF069aNkydP\nsmPHDi5evEhKSgpwddiRj48P3bp1IyIiAh8fH/bv38+AAQPqDUOS1qumpoaMjAyysrJISEjAycmJ\nP//5z+zcuZPu3bsTERGBp6cnCxcurNeLWxewISEh3HXXXbRv35709HR69epl60MSEWlyylW5GWWr\nNCUVo9Iq7du3j3/84x84Oztzxx138Kc//YnNmzdbhhAlJCQ0GpzV1dU4OTkRHBxMp06dePDBBy2B\nqfudtX5OTk5kZ2ezcOFCzp07x5IlSzhx4gRPPPEEoaGhmEwmwsLC8PLyqheadVcA6u6TFhoaSlhY\nWL1tIiL2TLkqN6JslaakYlRaJXd3d/bt28fSpUvJysriwoULPP3008TGxuLs7IzJZGo0OJ2cnKip\nqbF8QLq5uVl677SUeOuUmZlJRUUF/v7+ACQlJVFYWEhGRgZFRUX87Gc/Iz093fLvazQaG4Rm165d\nMRgMjQajwlJEHIFyVa6lbJXmomJUWiU3NzdSU1NZuHAhJSUlpKen06dPHzw9PS2T693d3esFZ0lJ\nCcnJyVpK3I5cvHiRd955hwULFvDAAw9Y7lW2ceNGTp8+bZnb0qVLF1xcXCxfmK4NzSVLlnDu3DlL\naIqIOCLlqtRRtkpzUjEqrdaRI0fIycnB39+f3bt3ExgYSHBwsKVXtqamBg8PDxISEjh9+jSbNm2i\nsLCQO++809ZNFytxc3PjjjvuICEhwTKPBcDPz4/u3btTW1vLtm3byM3NpVOnTphMpnqhGR4ejslk\nYuHChSQkJBAcHGzjIxIRsR3lqoCyVZqXilFpNa6fe+Lt7U3Pnj1JT0/n3LlzLF26FB8fH0JDQzEa\njZYPz7qhRUeOHKFTp05ER0fb6hDEiurOh+DgYNq3bw9cvf2Aj48PcXFxhISEkJaWRm5uLlu3biUv\nL4/ExETLwhzl5eUAJCcn07VrV+Lj4212LCIitqBclespW6W5qRiVVuH6uSe1tbW4uLhgMplwd3cn\nMTGR3NxcVqxYUS84y8rKOHjwIIGBgfTu3ZvY2FgbHoVY0/XDfk6fPs3cuXPJysqiU6dO+Pn5AdC1\na1fy8vLYunUr586do0uXLgAsW7aMOXPm0LVrV0JCQgAtqCAijkO5Ko1RtkpzUzEqLd61gTl37lwW\nL17M0qVLqa6uxsPDgzZt2mAymUhKSiI3N5fly5fj7e0NwKZNm3j//ffp1KkToaGhgFb3sxfXh5uX\nlxexsbHk5OSwYsUKkpKS8PPzw2Aw0LVrV86dO8f3339PdnY2OTk5LFq0iPvvv98SoKA5TiLiGJSr\nciPKVmluKkalRbs2MKdOnUpWVhbOzs4AZGdnc+rUKcLDw/Hz88PNzY2kpCTOnTvHwoUL+f7779m+\nfTuPP/44ffv2tbymPhRbv2vPi02bNlFeXk5AQACBgYGEh4dz8OBBVq5cWS80U1NTqaioID8/n4KC\nAoYOHcrgwYMBfZESEcehXJUbUbaKLRjMZrPZ1o0Q+SH/8z//w7Zt2xg5ciSdO3fGy8uLjz/+mDVr\n1hAbG8u4cePqzVlZvHgxJSUlREZG0rNnT6DhkCRpna79d5w2bRqHDh0iJCSE3/zmN7i7u2M2mzl4\n8CCffvopRUVF/P73v7fMezGbzZSXl1NeXm4ZaqTzQkQckXJVrqVsFVtRMSot3vHjx/mv//ov7rvv\nPh588EE8PDw4cOAAf/7znwkLC+P8+fOEhoYyYcIEoqKiLM+79oNQH4r258033+TIkSMMGTKErl27\nEhAQYHmstrbWEprFxcVMmjSp3rlRR722IuKIlKtyI8pWaW4apistTmMfYqdOnaJv374EBgZy8uRJ\nJk+eTFpaGi+99BIVFRVs3LiR3NxcwsPDLTdovvY19KFoX1atWsWaNWsYN24cPXr0wNvb2zLPpbi4\nGHd3d9q2bUtkZCQ5OTksXryYTp06Wc6NOjovRMQRKFflVihbxRZUjEqLcm1P6+XLl3F1dcVoNJKS\nkkJgYCAlJSX87W9/o127dowZMwYfHx8iIiLIysqirKyM3bt3k5aWZlliXOxTZmYmhYWF/OIXv8DV\n1ZXi4mIyMjKYMWMG3377LefPnycyMpLw8HDatWvH7t27CQkJ0e0HRMThKFflVilbxRZcbN0AkTpm\ns9kSmNOnTwdg9OjR+Pr6WkLwwoULFBUV0b9/f8uS4SdOnMBgMJCWlkZERAQ+Pj62OQBpNv7+/pw/\nf55du3YBMHPmTEpKSvD19SUxMZHFixfj7u7OsGHDiI+PZ/LkyZZ5LCIijkK5KrdD2Sq2oGJUWoSa\nmhrLan7V1dWcOnWK8vJy5s2bx5AhQyxLyldWVlJRUUFRUREAFy9etPTMjRgxAldXV0DzFezFjeYk\npaSksGvXLl5//XVqa2tJSUlh4MCB3HvvvcDV3v9NmzYxePBgTCaTJSx1XoiIo1Cuyo0oW6UlUTEq\nNldbW2sJzLfffpvy8nJKSkooLy9nyZIlAJbgDAoKIioqipUrV3Lo0CEAdu7cyahRoyyBCZqvYA+u\n/SK1f/9+rly5Qm1tLV26dCE8PJyJEydy9OhRzGYzCQkJGI1GAPLz8ykrK6N9+/YNhpXpvBARR6Bc\nlRtRtkpLozmjYnN1H2J///vf2blzJ3379mXw4MH07t2b8+fPk52dTWVlJVFRUfj6+hIfH09+fj6n\nT5+mtraWxx9/nIEDBwLqnbMX136RmjZtGgsWLGDVqlVkZmayY8cOXF1diYqKIjQ0lJCQEMu+BQUF\nrF+/nq1bt/Lwww8TERFhy8MQEbEJ5ao0RtkqLZFu7SItwpkzZ/jTn/5Et27dGDt2rKUnDuCdd94h\nKyuL/v378+ijj+Ln58eVK1cwGAxUVFRY5rJomXn7U/dFatCgQURGRlJUVMT8+fMpKiri8ccf5+GH\nH7b03GdnZ7Nu3Tr27NnD0KFDeeyxx2zcehER21Guyo0oW6Ul0TBdaREMBgNlZWX4+/tbArO6uhoX\nFxeef/55iouLWbZsmaXHtm6egpubG1B/kQaxD0ePHmXfvn08+OCDPPTQQ5bzokePHkyZMoXvvvuO\noKAg0tPTOX78OKtXr+by5ctMmDCBvn37AvoiJSKOS7kqjVG2SkujM0lahLql5o8ePUpVVRUALi4u\nVFdXAzBw4EBcXFzYsWMHixYtorS0FPi/oUgaQmR/KioquHDhAiEhIRiNRsxmMzU1NXh4ePDyyy/j\n4eHBqlWrAIiKimL8+PE8//zzCksREZSr0jhlq7Q0OpukWdXW1ja6vW3btgwcOJDvv/+ejIwM4Gqv\nrIvL1Yv3VVVVBAcHExYWxtKlS9m6datlH2n9Gjsv6r4I5efnA1f/rZ2dnamursbLy4v777+f/fv3\nc/DgQQBCQ0MJCgqy7KuwFBFHoFyVG1G2SmugM0qaTU1NjeVDbOPGjXz33Xd8++23HDlyhOrqau6/\n/36Sk5N5//33Wb16NVeuXAGgqKiIgwcPEh0dze9+9ztLcNbW1qrn1g5ce16UlpZSUlICQHx8PN27\nd2fhwoWcOnUKJycnyxAzAHd3dwwGA56eng1eU+eFiDgC5arciLJVWgutpivN4toV3P7yl7+wcuVK\nDhw4wN69e9myZQtFRUWkpqbSvn17Lly4wMKFCzlw4AD79+9n/fr1ZGVl0atXL+Li4qioqGDdunWk\npKQQGBho4yOTf8W158VHH33E3LlzWb58OVu3bsXf3x9fX19OnjzJihUr6Ny5M/7+/sDVL1IbNmyg\nvLycPn364OXlZcvDEBFpdspVuRFlq7QmKkalWdT1pv33f/83e/bsYdy4cYwYMYJRo0aRmZnJli1b\nuOOOO0hOTiYuLo7g4GD27t1Lbm4uBoOBxx57jEGDBgGwefNmTp8+zaBBg/RB2YpdO9xn2rRpbNmy\nhZiYGNq2bcuJEydYv349QUFBtG/fnoKCAhYsWMCFCxc4dOgQGzduJCMjg8cff5zU1FQbH4mISPNT\nrkpjlK3S2mg1XWk2hYWFHD58mAceeIC0tDRMJhNHjx7l9OnT9O3blw4dOgBX57kMGDCAHj164Orq\nSkVFhaXX7vDhw+zdu5fw8HAFZit27QIIly9fJjc3l9GjR3Pffffh7OxMaWkpH330EQsXLmTw4ME8\n8cQTbN++nQ0bNgAQHBzM+PHjGTBgAKD74ImIY1KuyrWUrdIaqRiVZnPlyhVyc3OJiorCZDKxZ88e\npk6dSlpaGk888YQlGI8fP054eDi+vr4AlnkLixcvZvv27eTl5TFlyhSFZitWF5ZvvPEG7u7uGI1G\nUlNTcXZ2pra2Fi8vL1544QXeeustVqxYwaRJk/i3f/s3HnnkEUwmEwaDgYCAAEAr+4mI41KuyrWU\nrdIa6SyTJtHYCm51q/gZDAaOHz9uCcwxY8ZYAnP9+vXMnDnTMtG+7nlbt24lIyODkpISpkyZQkRE\nRLMdizSNsrIyKisryc7O5vTp0xQWFlqGF9WdP2PHjsVsNrNgwQIAwsPDCQwMtJwvWtlPRByFclVu\nhbJVWhtdGRWrq6mpsUycLywsxGQy4eXlRUREBMnJyUyfPp2qqiruuecehg4darnRdmFhIdu3b6em\npqbesBCDwUB8fDy/+MUv8PHxwcfHxybHJdbl6enJs88+yxdffMH69evZsmULHTt2BP5vLlRgYCBh\nYWFcuHCh3nAh3QdPRByJclVulbJVWhsVo2JV167g9u6777Jnzx7c3d2JiorimWeeYdCgQRQUFHD6\n9Gl69+5tWbUvPz+ftWvXsmPHDsaPH28J0jqenp6NLjMurZufnx+jRo2isrKS+fPn06ZNGwYPHmwJ\nwqKiIiorK/H19bUsPa+QFBFHolyV26VsldZEq+mK1Vw7rOOdd95h27ZtpKSk4O7uzvbt29m6dSv9\n+vXDz8+PY8eOsWrVKvLz89mxYwerVq0iIyODoUOHauK8gzGZTCQkJFBQUMCSJUsoKSmhtraWCxcu\nsHbtWrZs2cJjjz1GTEyMzgcRcSjKVfmxlK3SWhjMZrPZ1o2Q1u/aie6VlZW88sor3HvvvfTr1w+A\nTZs28fnnn+Pl5cWLL75IcXExa9euZefOndTW1hITE0NaWhp9+vRp8HriGIqLi/n000/Jzs7GycmJ\nhIQEqqqq6NGjBw899BCgL1Ii4jiUq2INylZp6VSMilW98cYbeHh4UFBQwK9+9SvLZPjq6mr27NnD\nxx9/jLu7Oy+99BJBQUEUFRXh6uqKq6srbm5ugALTkRUXFzNjxgwyMjL4yU9+wpAhQyyP6bwQEUek\nXJV/lbJVWjIN0xWrKSsrIyMjg+3bt1NdXU3Pnj3x8vLCbDbj7OxM27ZtCQ8PZ8uWLWRnZ9OlSxcC\nAwMxGo04OztjMBi0gpuDM5lMdOjQgfz8fJYtW4anpycdOnRQWIqIQ1KuijUoW6UlUzEqVmM0GklO\nTqa4uJgDBw7g6+tLp06dMBgMlgUY6oIzOzub1atX069fP1xdXbWCm1i4u7tb5rksWrQIb29vy43b\nRUQciXJVrEXZKi2VilGxKnd3dzp27EhhYSHLli3Dy8uLDh06NAjOdu3aERcXpw9CaVTdwgvnz59n\nwYIF+Pn5ER0dbetmiYg0O+WqWIuyVVoiFaNiddeu4LZo0aJGg7Ndu3a0b98e0MR5aZzJZCIuLo5L\nly7Rq1cvvL29bd0kERGbUK6KtShbpaXRAkbSZIqLi/nkk0/YtGkTEyZMsCwtL3I7NKdFROQq5apY\ni7JVWgpdGZUmc/1wEDc3N+Li4mzdLGll1LsvInKVclWsRdkqLYWLrRsg9s3X15fx48dz+fJlnJ2d\nbd0cERGRVk25KiL2RMN0pVlUVFRY7ncmIiIi/xrlqojYAxWj0qy0qIKIiIj1KFdFpDVTMSoiIiIi\nIiLNTstoiYiIiIiISLNTMSoiIiIiIiLNTsWoiIiIiIiINDsVoyIiIiIiItLsVIyKNKP8/HyGDRvG\ngQMHGv29uc2ePZvnnnvupvs8++yzfP3117f8mtY6Jlv/bUREpOVTrt46W/9tRBrjYusGiNjSu+++\ny7p16wBwcnLC39+fLl26MGLECNq0adPk7x8YGMiHH36Il5fXLe1/4MAB/vjHPzJ9+nSCgoKauHUi\nIiK3R7kqIrdDxag4vISEBF544QVqamo4evQoH3zwAefPn2fSpEmN7l9dXY2Li3X+13FycsLX19cq\nryUiItISKFdF5FapGBWH5+LiYgmugIAATp06xaxZs6isrKS4uJiJEyfy3HPPsXHjRvbt20f//v0Z\nPXo0eXl5zJgxgz179mAwGIiJiWHs2LFERERYXjszM5Mvv/ySCxcu0L59ex577LF6752fn8/EiRN5\n5ZVXiI+PB+DixYvMnDmT7du3c/nyZQIDA3n00UdJSkrij3/8IwATJ04EIDExkcmTJwOQkZHBvHnz\nyM3NxdfXl7S0NH76059iMpkAqKys5LPPPmPjxo04OTmRnp6Op6fnbf+9Nm7cyOLFizlz5gwuLi7E\nxsYybtw4QkNDGxzb7NmzOXjwIH5+fowYMYJevXpZHi8uLmbmzJls27aNqqoqIiMjGTFiBImJibfd\nJhERaTmUq7dHuSqOTMWoyHWMRiNms5mamhrLtpkzZzJq1Ch+9rOfAVc/8F9++WXS0tJ45ZVXcHFx\nYenSpUyePJlp06bh7e3NsWPHePvtt3n00Ue59957OXXqFP/4xz9u+t6VlZVMnjwZo9HIc889R3Bw\nMHl5eZSWlhIYGMhvf/tb3njjDV577TUCAwMtPclr167ls88+Y8KECcTFxXH+/Hk++eQTSkpKLHNX\nvvjiCzZt2sTEiRMJDQ1l1apVLFu2DG9v79v6+1RVVTFkyBDCw8O5cuUKs2fPZurUqbz11lv1erZn\nzpzJ6NGjeeqpp1i/fj3vvPMOoaGhtG/fnsrKSqZMmUJYWBh/+MMf8PT0JDMzk1dffZU33niD8PDw\n22qTiIi0XMrVm1OuiiPTAkYi1zh9+jTLli2jQ4cOuLu7W7b369eP3r17ExQURFBQEMuXLycoKIif\n//znREREEBoayoQJE/D09GTDhg0ALFiwgI4dOzJy5EhCQ0Pp3r07jzzyyE3ff+PGjeTn5/Pb3/6W\nlJQUgoOD6dy5M7169cLJyckyB8bb2xtfX1/L73PmzGHEiBH06dOH4OBgEhMTefLJJ9mwYQOlpaWU\nl5ezYsUKhg8fzl133UVYWBhjx46lXbt2t/03uu++++jWrRshISG0b9+eZ599lry8PA4fPtxgv969\nexMaGsrw4cPp0KEDCxcuBK72bF+5coUXXniBmJgYQkJCGDJkCHFxcaxcufK22yQiIi2TcvWHKVfF\nkenKqDi8vXv3MmbMGGpra6muriYpKYmnn3663j6xsbH1fj9y5AhHjx5lzJgx9bZXVlZy9uxZAM6c\nOUNSUlK9x+uGDN3I0aNHCQ8PJyAg4JbbX1JSQkFBAZ9//jn//Oc/Gzyel5eHq6srVVVVxMXFNWjP\n1q1bb/m9AI4fP86cOXM4fvw4ly5dwmw2A1BYWFhvv44dO9b7PS4ujj179gBw+PBhiouLGT9+fL19\nqqurMRqNt9UeERFpWZSrylWRW6ViVBxehw4dePbZZy2r/jW2iELd/JA6ZrOZpKQky/Cia3l4eDRZ\nWxtTW1sLwPjx4xuENIC/v78lyP9VFRUVvPrqq8THx/PMM8/g4+MDwIsvvkh1dfUtv47ZbCYsLIyX\nXnqpwWMKTRGR1k25euuUq+LoVIyKwzMajYSEhNzWc6Kjo1m3bh3+/v43/JAPCwsjJyen3rYfurdX\ndHQ0a9as4fz584324tYFel1QAvj6+hIQEEBubi79+vVr9HWDg4NxcXHh4MGD3HHHHZbtBw8evGl7\nrnfmzBlKSkoYPny4Zf7JwYMHLb241zp06BCpqamW33NycggLCwMgJiaG9evX4+7ubgleERGxD8rV\nW6dcFUenOaMiP8KAAQOora3lzTffZP/+/eTn53PgwAG+/PJLSxANGjSInJwcvvzyS3Jzc9m8ebNl\nbseN9OrVi8DAQN544w127dpFfn4+u3fvJjMzE7h6/zSDwcD27du5ePEily9fBmDEiBEsWbKEb775\nhpMnT1re78MPPwSu9kA/8MADzJo1i++//57c3FxmzJhBbm7ubR13YGAgrq6uLF26lLy8PHbv3s2n\nn36KwWBosO/q1avZuHEjubm5zJo1i5ycHAYNGgTA3XffTVBQEFOnTmXnzp3k5+dz6NAhvv32WzZv\n3nxbbRIRkdZPuapcFcekK6MiP4Kvry+vvvoqX375JX/961+5cuUKvr6+xMfH4+fnB1ztjX3++ef5\n6quvWLBgAVFRUYwbN44333zzhq/r5ubGlClTmDFjBm+//Tbl5eW0bdvWsnS9r68vI0eOZN68eXz6\n6ackJCQwefJk+vTpg8lk4rvvvuObb77B2dmZoKAgunfvbnntUaNGUVVVxd///ncA0tPT6d+/P1lZ\nWbd83N7e3jz33HN88cUXrFmzhrCwMMaPH88rr7zSYN9Ro0axcuVKDh06hK+vL8899xzR0dHA1V7z\nyZMn89VXX/Hee+9RUlKCt7c3sbGxdOnS5ZbbIyIi9kG5qlwVx2QwNzYOQERERERERKQJaZiuiIiI\niIiINDsVoyIiIiIiItLsVIyKiIiIiIhIs1MxKiIiIiIiIs1OxaiIiIiIiIg0OxWjIiIiIiIi0uxU\njIqIiIiIiEizUzEqIiIiIiIize7/A7sUolO8CpQTAAAAAElFTkSuQmCC\n",
            "text/plain": [
              "<Figure size 1440x1080 with 4 Axes>"
            ]
          },
          "metadata": {
            "tags": []
          }
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "zSGFOgYa833B",
        "colab_type": "text"
      },
      "source": [
        "### From the confusion matrices:\n",
        "\n",
        "- **Naive Bayes Classifier**: The model predicted 76% of labels correctly as negative and 73% correctly as positive. The model predicted 27% of the labels as negative, but they were positive (false negatives). The model predicted 24% of the labels as positive when they were negative (false postive).\n",
        "\n",
        "- **SVM Classifier**: The model predicted 75% of labels correctly as negative and 75% correctly as positive. The model predicted 25% of the labels as negative, but they were positive (false negatives). The model predicted 25% of the labels as positive when they were negative (false postive).\n",
        "\n",
        "- **LSTM Neural Network**: The model predicted 76% of labels correctly as negative and 73% correctly as positive. The model predicted 27% of the labels as negative, but they were positive (false negatives). The model predicted 24% of the labels as positive when they were negative (false postive).\n",
        "\n",
        "- **LSTM Neural Network with Regularisation**: The model predicted 74% of labels correctly as negative and 73% correctly as positive. The model predicted 25% of the labels as negative, but they were positive (false negatives). The model predicted 26% of the labels as positive when they were negative (false postive), which is a bit of an improvement in accuracy over the first RNN model."
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "THFbV5RfvX1u",
        "colab_type": "text"
      },
      "source": [
        "### 6.2 Accuracy, Recall, Precision and F1-Scores\n",
        "\n",
        "Now we generate a classification report. We'll save it as a pandas dataframe for better formatting, with the help of a helper function (modified from source code [here](https://gist.github.com/wassname/f3cbdc14f379ba9ec2acfafe5c1db592))."
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "Oo8SOFmVKc1k",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "# Helper function for Classification Report \n",
        "\n",
        "def classification_report_asdf(test, pred, output_dict=True):\n",
        "  ''' ------------------------------\n",
        "  Function to return a classification report \n",
        "  as dataframe, with rounded scores. \n",
        "  --- Parameters:----\n",
        "  test = test data as array\n",
        "  pred = prediction data array\n",
        "  output_dic = set to True\n",
        "  ------------------------\n",
        "  '''\n",
        "  # extract scores as dictionary\n",
        "  report = classification_report(test,pred, output_dict=output_dict)\n",
        "  df_report = pd.DataFrame(report).transpose()\n",
        "  #round\n",
        "  df_report[[\"precision\",\"recall\",\"f1-score\"]]= df_report[[\"precision\",\"recall\",\"f1-score\"]].apply(lambda x: round(x,2))\n",
        "  df_report[[\"support\"]]= df_report[[\"support\"]].apply(lambda x: x.astype(np.int))\n",
        "  # Add suport to avg\n",
        "  df_report.loc['accuracy', 'support'] = df_report.loc['weighted avg', 'support']\n",
        "\n",
        "  return df_report"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "hTJJ3RxZKmkl",
        "colab_type": "code",
        "outputId": "97f07b80-9922-4bbe-de26-8d8bc98cda9e",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 233
        }
      },
      "source": [
        "# Classification reports for Supervised ML models\n",
        "\n",
        "#--- Naive Bayes Classifier ----#\n",
        "report_nb = classification_report_asdf(y_test, y_pred_nb)\n",
        "\n",
        "#--- SVM Classifier ----#\n",
        "report_svm = classification_report_asdf(y_test, y_pred_svm)\n",
        "\n",
        "#--- Add reports together ----#\n",
        "SML_dfs = [report_nb, report_svm]\n",
        "\n",
        "SML_dfs_combined = pd.concat(\n",
        "    [df.rename(columns=lambda x: x.zfill(4)) for df in SML_dfs],\n",
        "    keys=['Naive Bayes Classifier', 'SVM Classifier'],\n",
        "    axis=1\n",
        ")\n",
        "\n",
        "# Show reports\n",
        "SML_dfs_combined"
      ],
      "execution_count": 0,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/html": [
              "<div>\n",
              "<style scoped>\n",
              "    .dataframe tbody tr th:only-of-type {\n",
              "        vertical-align: middle;\n",
              "    }\n",
              "\n",
              "    .dataframe tbody tr th {\n",
              "        vertical-align: top;\n",
              "    }\n",
              "\n",
              "    .dataframe thead tr th {\n",
              "        text-align: left;\n",
              "    }\n",
              "</style>\n",
              "<table border=\"1\" class=\"dataframe\">\n",
              "  <thead>\n",
              "    <tr>\n",
              "      <th></th>\n",
              "      <th colspan=\"4\" halign=\"left\">Naive Bayes Classifier</th>\n",
              "      <th colspan=\"4\" halign=\"left\">SVM Classifier</th>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th></th>\n",
              "      <th>precision</th>\n",
              "      <th>recall</th>\n",
              "      <th>f1-score</th>\n",
              "      <th>support</th>\n",
              "      <th>precision</th>\n",
              "      <th>recall</th>\n",
              "      <th>f1-score</th>\n",
              "      <th>support</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>0.75</td>\n",
              "      <td>0.73</td>\n",
              "      <td>0.74</td>\n",
              "      <td>7313</td>\n",
              "      <td>0.76</td>\n",
              "      <td>0.75</td>\n",
              "      <td>0.75</td>\n",
              "      <td>7313</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>0.73</td>\n",
              "      <td>0.75</td>\n",
              "      <td>0.74</td>\n",
              "      <td>7035</td>\n",
              "      <td>0.74</td>\n",
              "      <td>0.75</td>\n",
              "      <td>0.75</td>\n",
              "      <td>7035</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>accuracy</th>\n",
              "      <td>0.74</td>\n",
              "      <td>0.74</td>\n",
              "      <td>0.74</td>\n",
              "      <td>14348</td>\n",
              "      <td>0.75</td>\n",
              "      <td>0.75</td>\n",
              "      <td>0.75</td>\n",
              "      <td>14348</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>macro avg</th>\n",
              "      <td>0.74</td>\n",
              "      <td>0.74</td>\n",
              "      <td>0.74</td>\n",
              "      <td>14348</td>\n",
              "      <td>0.75</td>\n",
              "      <td>0.75</td>\n",
              "      <td>0.75</td>\n",
              "      <td>14348</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>weighted avg</th>\n",
              "      <td>0.74</td>\n",
              "      <td>0.74</td>\n",
              "      <td>0.74</td>\n",
              "      <td>14348</td>\n",
              "      <td>0.75</td>\n",
              "      <td>0.75</td>\n",
              "      <td>0.75</td>\n",
              "      <td>14348</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>"
            ],
            "text/plain": [
              "             Naive Bayes Classifier                  ... SVM Classifier                 \n",
              "                          precision recall f1-score  ...         recall f1-score support\n",
              "0                              0.75   0.73     0.74  ...           0.75     0.75    7313\n",
              "1                              0.73   0.75     0.74  ...           0.75     0.75    7035\n",
              "accuracy                       0.74   0.74     0.74  ...           0.75     0.75   14348\n",
              "macro avg                      0.74   0.74     0.74  ...           0.75     0.75   14348\n",
              "weighted avg                   0.74   0.74     0.74  ...           0.75     0.75   14348\n",
              "\n",
              "[5 rows x 8 columns]"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 98
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "xCqx16vzwpMX",
        "colab_type": "code",
        "outputId": "352a6bec-ec73-4d32-e624-0dc686da317e",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 233
        }
      },
      "source": [
        "# Classification report for RNN models\n",
        "\n",
        "#---- LSTM Network ----#\n",
        "report_LSTM = classification_report_asdf(Y_test, y_preds_LSTM)\n",
        "\n",
        "#---- LSTM Network with regularisation ----#\n",
        "report_LSTMreg = classification_report_asdf(Y_test, y_preds_LSTMreg)\n",
        "\n",
        "\n",
        "#--- Add reports together ----#\n",
        "RNN_dfs = [report_LSTM, report_LSTMreg]\n",
        "\n",
        "RNN_dfs_combined = pd.concat(\n",
        "    [df.rename(columns=lambda x: x.zfill(4)) for df in RNN_dfs],\n",
        "    keys=['LSTM Network', 'LSTM Network with regularisation'],\n",
        "    axis=1\n",
        ")\n",
        "\n",
        "# Show reports\n",
        "RNN_dfs_combined"
      ],
      "execution_count": 0,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/html": [
              "<div>\n",
              "<style scoped>\n",
              "    .dataframe tbody tr th:only-of-type {\n",
              "        vertical-align: middle;\n",
              "    }\n",
              "\n",
              "    .dataframe tbody tr th {\n",
              "        vertical-align: top;\n",
              "    }\n",
              "\n",
              "    .dataframe thead tr th {\n",
              "        text-align: left;\n",
              "    }\n",
              "</style>\n",
              "<table border=\"1\" class=\"dataframe\">\n",
              "  <thead>\n",
              "    <tr>\n",
              "      <th></th>\n",
              "      <th colspan=\"4\" halign=\"left\">LSTM Network</th>\n",
              "      <th colspan=\"4\" halign=\"left\">LSTM Network with regularisation</th>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th></th>\n",
              "      <th>precision</th>\n",
              "      <th>recall</th>\n",
              "      <th>f1-score</th>\n",
              "      <th>support</th>\n",
              "      <th>precision</th>\n",
              "      <th>recall</th>\n",
              "      <th>f1-score</th>\n",
              "      <th>support</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>0.75</td>\n",
              "      <td>0.76</td>\n",
              "      <td>0.75</td>\n",
              "      <td>4860</td>\n",
              "      <td>0.75</td>\n",
              "      <td>0.74</td>\n",
              "      <td>0.75</td>\n",
              "      <td>4860</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>0.75</td>\n",
              "      <td>0.73</td>\n",
              "      <td>0.74</td>\n",
              "      <td>4705</td>\n",
              "      <td>0.74</td>\n",
              "      <td>0.75</td>\n",
              "      <td>0.74</td>\n",
              "      <td>4705</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>accuracy</th>\n",
              "      <td>0.75</td>\n",
              "      <td>0.75</td>\n",
              "      <td>0.75</td>\n",
              "      <td>9565</td>\n",
              "      <td>0.74</td>\n",
              "      <td>0.74</td>\n",
              "      <td>0.74</td>\n",
              "      <td>9565</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>macro avg</th>\n",
              "      <td>0.75</td>\n",
              "      <td>0.75</td>\n",
              "      <td>0.75</td>\n",
              "      <td>9565</td>\n",
              "      <td>0.74</td>\n",
              "      <td>0.74</td>\n",
              "      <td>0.74</td>\n",
              "      <td>9565</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>weighted avg</th>\n",
              "      <td>0.75</td>\n",
              "      <td>0.75</td>\n",
              "      <td>0.75</td>\n",
              "      <td>9565</td>\n",
              "      <td>0.74</td>\n",
              "      <td>0.74</td>\n",
              "      <td>0.74</td>\n",
              "      <td>9565</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>"
            ],
            "text/plain": [
              "             LSTM Network                  ... LSTM Network with regularisation                 \n",
              "                precision recall f1-score  ...                           recall f1-score support\n",
              "0                    0.75   0.76     0.75  ...                             0.74     0.75    4860\n",
              "1                    0.75   0.73     0.74  ...                             0.75     0.74    4705\n",
              "accuracy             0.75   0.75     0.75  ...                             0.74     0.74    9565\n",
              "macro avg            0.75   0.75     0.75  ...                             0.74     0.74    9565\n",
              "weighted avg         0.75   0.75     0.75  ...                             0.74     0.74    9565\n",
              "\n",
              "[5 rows x 8 columns]"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 100
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "IYM7dexr-pSD",
        "colab_type": "text"
      },
      "source": [
        "### From the classification reports:\n",
        "\n",
        "- **Naive Bayes Classifier**: \n",
        "  - *Precision (Accuracy of positive predictions)*: 75% for label negative and 75% for positive. This number tells us what proportion of fraction of predicted positive events that are actually positive .\n",
        "\n",
        "  - *Recall (aka sensitivity or true positive rate)*: 76% for negative and 73% for positive label. This is the number of correct predictions out of true labels for that class.\n",
        "\n",
        "  - *F1-Score (2 x (precision x recall)/(precision + recall)*: This is the weighted average of precision and recall for that class. It generally gives the bigger picture of how the model is performing for that label. The higher this number is the better. We have  75% for label negative and 74% for label postive.\n",
        "\n",
        "  - *Accuracy*: 74% is the number of times the model predicted the correct label over all labels in the dataset. This is most common metric for classification and shows the proportion of samples predicted correctly. When the distribution of the classes in data is well balanced, accuracy can give us a good picture of how the model is performing.\n",
        "\n",
        "  - *Support*: This is just the number of samples of the true response that lie in that class. \n",
        "<br>\n",
        "\n",
        "- **SVM Classifier**: \n",
        "  - *Precision*: 76% for label negative and 74% for positive.\n",
        "\n",
        "  - *Recall*: 74% for negative and 75% for positive label.\n",
        "\n",
        "  - *F1-Score*: 75% for label negative and 74% for label postive.\n",
        "\n",
        "  - *Accuracy*: 75%\n",
        "  <br>\n",
        "\n",
        "- **LSTM Neural Network**: \n",
        "  - *Precision*: 75% for label negative and 75% for positive.\n",
        "\n",
        "  - *Recall*: 76% for negative and 73% for positive label.\n",
        "\n",
        "  - *F1-Score*: 75% for label negative and 74% for label postive.\n",
        "\n",
        "  - *Accuracy*: 75%\n",
        "  <br>\n",
        "\n",
        "- **LSTM Neural Network with regularisation**: \n",
        "  - *Precision*: 75% for label negative and 74% for positive.\n",
        "\n",
        "  - *Recall*: 74% for negative and 75% for positive label.\n",
        "\n",
        "  - *F1-Score*: 75% for label negative and 74% for label postive.\n",
        "\n",
        "  - *Accuracy*: 74%"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "ixFkf0nXTBsx",
        "colab_type": "text"
      },
      "source": [
        "### 6.3 Choosing a model.\n",
        "\n",
        "From the above metrics, our models seem to be performing relatively well.\n",
        "\n",
        "According to research [human raters typically agree 80% of the time.](http://www.mecs-press.org/ijeme/ijeme-v7-n1/IJEME-V7-N1-3.pdf).\n",
        "\n",
        "Therefore a 75% accurate program is doing almost as good as humans, even though such accuracy may not seem all that impressive. If a program were 'right' 100% of the time, humans would still disagree with it.\n",
        "\n",
        "We can also benchmark our model against [similar work](https://www-sciencedirect-com.ezproxy.napier.ac.uk/science/article/pii/S1877050917328089). The final model was based on Support Vector Machine and went on to achieve an accuracy of 79.08%, which is not far from our own models' results.\n",
        "\n",
        "Both SVM and LSTM netword classifier without regularisation outperformed the other models by 1%, achieving 75% accuracy.\n",
        "\n",
        "Thus, the performance of the LSTM network is found to be better than other models and is the one we will use for our next steps. We choose LSTM over the SVM model because generally, deep learning really shines when it comes to complex problems such as natural language processing. Another advantage is that you have we worry less about the feature engineering part when it comes to model deployment.\n",
        "\n",
        "Ultimately the true test is testing the model on unseen, real world data.\n",
        "\n",
        "We will now run sentiment predictions with our chosen model on the data we acquired during ETL via the twitter API and evaluate its performance."
      ]
    }
  ]
}