Quora_question_pairs_.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "id": "43daa04f",
   "metadata": {
    "id": "43daa04f"
   },
   "source": [
    "# Importing Libraries:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 94,
   "id": "17201d7a",
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "17201d7a",
    "outputId": "4823d9ba-aca7-4984-914c-9b0d0963a740"
   },
   "outputs": [],
   "source": [
    "import numpy as np\n",
    "import matplotlib.pyplot as plt\n",
    "import seaborn as sns\n",
    "import pandas as pd\n",
    "from fuzzywuzzy import fuzz\n",
    "import distance\n",
    "from wordcloud import WordCloud\n",
    "import nltk\n",
    "from nltk.corpus import stopwords\n",
    "from nltk.stem import WordNetLemmatizer\n",
    "from sklearn.feature_extraction.text import CountVectorizer\n",
    "from sklearn.feature_extraction.text import TfidfVectorizer\n",
    "from tqdm import tqdm\n",
    "from gensim.models import Word2Vec\n",
    "import math\n",
    "import re\n",
    "from sklearn.metrics import accuracy_score, log_loss \n",
    "from sklearn.metrics import roc_curve, auc\n",
    "from sklearn.metrics import roc_auc_score\n",
    "from sklearn.calibration import CalibratedClassifierCV\n",
    "from sklearn.model_selection import train_test_split\n",
    "from sklearn.model_selection import GridSearchCV\n",
    "from sklearn.ensemble import RandomForestClassifier\n",
    "from sklearn.model_selection import cross_val_score\n",
    "from sklearn.linear_model import SGDClassifier\n",
    "from sklearn.linear_model import LogisticRegression\n",
    "import lightgbm as ltb\n",
    "import warnings\n",
    "warnings.filterwarnings(\"ignore\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "id": "Mvg2zsU1HLW4",
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "Mvg2zsU1HLW4",
    "outputId": "e4b40d27-9ced-4d32-f960-475f6ce80e31"
   },
   "outputs": [],
   "source": [
    "#from google.colab import drive\n",
    "#drive.mount('/content/drive')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "id": "0d836e1c",
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 192
    },
    "id": "0d836e1c",
    "outputId": "1fcfc1c9-071f-4f93-f259-d9d800878272"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "(404290, 6)\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>id</th>\n",
       "      <th>qid1</th>\n",
       "      <th>qid2</th>\n",
       "      <th>question1</th>\n",
       "      <th>question2</th>\n",
       "      <th>is_duplicate</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>2</td>\n",
       "      <td>What is the step by step guide to invest in sh...</td>\n",
       "      <td>What is the step by step guide to invest in sh...</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>1</td>\n",
       "      <td>3</td>\n",
       "      <td>4</td>\n",
       "      <td>What is the story of Kohinoor (Koh-i-Noor) Dia...</td>\n",
       "      <td>What would happen if the Indian government sto...</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>2</td>\n",
       "      <td>5</td>\n",
       "      <td>6</td>\n",
       "      <td>How can I increase the speed of my internet co...</td>\n",
       "      <td>How can Internet speed be increased by hacking...</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>3</td>\n",
       "      <td>7</td>\n",
       "      <td>8</td>\n",
       "      <td>Why am I mentally very lonely? How can I solve...</td>\n",
       "      <td>Find the remainder when [math]23^{24}[/math] i...</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   id  qid1  qid2                                          question1  \\\n",
       "0   0     1     2  What is the step by step guide to invest in sh...   \n",
       "1   1     3     4  What is the story of Kohinoor (Koh-i-Noor) Dia...   \n",
       "2   2     5     6  How can I increase the speed of my internet co...   \n",
       "3   3     7     8  Why am I mentally very lonely? How can I solve...   \n",
       "\n",
       "                                           question2  is_duplicate  \n",
       "0  What is the step by step guide to invest in sh...             0  \n",
       "1  What would happen if the Indian government sto...             0  \n",
       "2  How can Internet speed be increased by hacking...             0  \n",
       "3  Find the remainder when [math]23^{24}[/math] i...             0  "
      ]
     },
     "execution_count": 6,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "path = \"train_quora.csv\"\n",
    "data = pd.read_csv(path)\n",
    "print(data.shape)\n",
    "data.head(4)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "id": "w2qKYVwUH2D5",
   "metadata": {
    "id": "w2qKYVwUH2D5"
   },
   "outputs": [],
   "source": [
    "data_ = data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "id": "d7942c79",
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "d7942c79",
    "outputId": "729a93c1-b738-4258-c287-9c45ed5afefc"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Following is the example for NON-SIMILAR questions\n",
      "****************************************************************************************************\n",
      "Method to find separation of slits using fresnel biprism?\n",
      "****************************************************************************************************\n",
      "What are some of the things technicians can tell about the durability and reliability of Laptops and its components?\n"
     ]
    }
   ],
   "source": [
    "# Example of non-similar questions:\n",
    "\n",
    "print('Following is the example for NON-SIMILAR questions')\n",
    "print('*'*100)\n",
    "print(data_['question1'].values[10])\n",
    "print('*'*100)\n",
    "print(data_['question2'].values[10])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "id": "06088db7",
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "06088db7",
    "outputId": "0b55897b-0d79-4fe6-a987-ed7f18d39c26"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Following is the example for SIMILAR questions\n",
      "****************************************************************************************************\n",
      "How do I read and find my YouTube comments?\n",
      "****************************************************************************************************\n",
      "How can I see all my Youtube comments?\n"
     ]
    }
   ],
   "source": [
    "# Example of similar questions:\n",
    "\n",
    "print('Following is the example for SIMILAR questions')\n",
    "print('*'*100)\n",
    "print(data_['question1'].values[11])\n",
    "print('*'*100)\n",
    "print(data_['question2'].values[11])"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "b528d3ed",
   "metadata": {
    "id": "b528d3ed"
   },
   "source": [
    "# Splitting Data:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "id": "cbc9f783",
   "metadata": {
    "id": "cbc9f783"
   },
   "outputs": [],
   "source": [
    "x = data_.drop(['is_duplicate'],axis = 1)\n",
    "y = data_['is_duplicate']"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "id": "0a806b6d",
   "metadata": {
    "id": "0a806b6d"
   },
   "outputs": [],
   "source": [
    "# Splitting the data into Train,CV and Test Sets:\n",
    "\n",
    "x_train,x_test,y_train,y_test = train_test_split(x,y,test_size = 0.2,stratify = y,random_state = 0)\n",
    "x_train1,x_cv,y_train1,y_cv = train_test_split(x_train,y_train,test_size = 0.2,stratify = y_train,random_state = 0)\n",
    "\n",
    "# stratify helps in fair distribution of class labels for both train and test sets."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "id": "eef2cb11",
   "metadata": {
    "id": "eef2cb11"
   },
   "outputs": [],
   "source": [
    "xy_train = pd.concat([x_train,y_train], axis = 1)\n",
    "xy_test = pd.concat([x_test,y_test], axis = 1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "id": "7c3e0123",
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "7c3e0123",
    "outputId": "a6b9c6bd-b58e-47e5-901c-dd971db63dbb"
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0    204022\n",
       "1    119410\n",
       "Name: is_duplicate, dtype: int64"
      ]
     },
     "execution_count": 13,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "labels = y_train.value_counts()\n",
    "labels"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "dd2e5074",
   "metadata": {
    "id": "dd2e5074"
   },
   "source": [
    "# Distribution of class label: 0 = Not Duplicate, 1 = Duplicate\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "id": "0f935ac4",
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 295
    },
    "id": "0f935ac4",
    "outputId": "25a53266-3fda-4de7-8fc7-df928ced8271"
   },
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "x_axis = [str(0),str(1)]\n",
    "y_axis = [labels[0],labels[1]]\n",
    "\n",
    "plt.bar(x_axis,y_axis,width = 0.4,)\n",
    "plt.xlabel('Duplicate or not')\n",
    "plt.ylabel('Number of questions')\n",
    "plt.title('Frequency of Duplicate questions')\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "e046e117",
   "metadata": {
    "id": "e046e117"
   },
   "source": [
    "- We see that this is a case of slightly imbalanced dataset. (64-36)\n",
    "- Number of non-duplicate questions are more than duplicate questions."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "id": "1c5a5cbd",
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "1c5a5cbd",
    "outputId": "757fa991-98e5-417c-e326-9c7a68046b80"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Percentage of Similar question pairs in dataset is: 36.91966162902867%\n",
      "Percentage of Not Similar questions in dataset is: 63.08033837097133%\n"
     ]
    }
   ],
   "source": [
    "# Percentage of similar questions:\n",
    "\n",
    "similar = ((labels[1])/(labels[0] + labels[1]))*100\n",
    "not_similar = ((labels[0])/(labels[0] + labels[1]))*100\n",
    "print('Percentage of Similar question pairs in dataset is: {}%'.format(similar))\n",
    "print('Percentage of Not Similar questions in dataset is: {}%'.format(not_similar))\n"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "e9f96315",
   "metadata": {
    "id": "e9f96315"
   },
   "source": [
    "# Checking null values:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "id": "274665d8",
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "274665d8",
    "outputId": "ff98e0cd-b37b-45be-fa52-210fa40328b7"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "            id    qid1    qid2                         question1  \\\n",
      "363362  363362  493340  493341                               NaN   \n",
      "105780  105780  174363  174364    How can I develop android app?   \n",
      "201841  201841  303951  174364  How can I create an Android app?   \n",
      "\n",
      "                                                question2  is_duplicate  \n",
      "363362  My Chinese name is Haichao Yu. What English na...             0  \n",
      "105780                                                NaN             0  \n",
      "201841                                                NaN             0  \n"
     ]
    }
   ],
   "source": [
    "# Train set\n",
    "\n",
    "null_value = xy_train[xy_train.isnull().any(1)]\n",
    "print(null_value)\n",
    "\n",
    "# .any(1) this returns the instance where the dataframe has null values"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "id": "7535d089",
   "metadata": {
    "id": "7535d089"
   },
   "outputs": [],
   "source": [
    "xy_train.dropna(inplace = True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "id": "8RdobsRyJAPt",
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "8RdobsRyJAPt",
    "outputId": "dcc52fcd-3d61-4d6b-994d-fd672bd979ee"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Empty DataFrame\n",
      "Columns: [id, qid1, qid2, question1, question2, is_duplicate]\n",
      "Index: []\n"
     ]
    }
   ],
   "source": [
    "null_value = xy_train[xy_train.isnull().any(1)]\n",
    "print(null_value)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "id": "72b59252",
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "72b59252",
    "outputId": "eb0b2613-0e88-4908-b9fc-46bbd12dad19"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Empty DataFrame\n",
      "Columns: [id, qid1, qid2, question1, question2, is_duplicate]\n",
      "Index: []\n"
     ]
    }
   ],
   "source": [
    "# Test set\n",
    "\n",
    "null_value = xy_test[xy_test.isnull().any(1)]\n",
    "print(null_value)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "id": "d9dd6eb2",
   "metadata": {
    "id": "d9dd6eb2"
   },
   "outputs": [],
   "source": [
    "# x_test.dropna(inplace = True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "id": "QcPSXAlUMBRP",
   "metadata": {
    "id": "QcPSXAlUMBRP"
   },
   "outputs": [],
   "source": [
    "# Storing two questions in seperate variable:\n",
    "\n",
    "q1_train = xy_train.iloc[:,3].values\n",
    "q2_train = xy_train.iloc[:,4].values\n",
    "\n",
    "q1_test = xy_test.iloc[:,3].values\n",
    "q2_test = xy_test.iloc[:,4].values"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "id": "Aves5ipAzHxb",
   "metadata": {
    "id": "Aves5ipAzHxb"
   },
   "outputs": [],
   "source": [
    "y_train = xy_train[\"is_duplicate\"]\n",
    "y_test = xy_test[\"is_duplicate\"]"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "a379e36b",
   "metadata": {
    "id": "a379e36b"
   },
   "source": [
    "# Genreal idea about data:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "id": "ffabc91c",
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "ffabc91c",
    "outputId": "ff8fa633-086f-41c3-9c24-431615e32ef3"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Question ID and their number of occurences\n",
      "2559     132\n",
      "30782     97\n",
      "4044      86\n",
      "17978     70\n",
      "2561      70\n",
      "dtype: int64\n"
     ]
    }
   ],
   "source": [
    "# Here, we see which questions are repeated multiple number of times.\n",
    "# We get to know about these questions through the \"question id\" (qid)\n",
    "\n",
    "ques = pd.Series(xy_train['qid1'].tolist() + xy_train['qid2'].tolist())\n",
    "unique_ques = len(np.unique(ques))\n",
    "repeat = np.sum(ques.value_counts() > 1)\n",
    "max_ques = max(ques.value_counts())\n",
    "values = ques.value_counts()\n",
    "\n",
    "idx = ques.value_counts().index.tolist()[0] # To get the qid value for question that repeats maximum number of times\n",
    "\n",
    "maxi = xy_train.loc[(data_['qid1'] == idx)]\n",
    "maxi2 = maxi[maxi.columns[[3]]]\n",
    "\n",
    "print('Question ID and their number of occurences')\n",
    "print(values[:5])\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "id": "61d97b18",
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 1000
    },
    "id": "61d97b18",
    "outputId": "53be5c35-77ad-4be7-cf77-4816df50ca44"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Total number of unique questions are:  450207\n",
      "***************************************************************************\n",
      "19.08% of the unique questions repeat themselves\n",
      "***************************************************************************\n",
      "The following question is repeated 132 times\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>question1</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>113625</th>\n",
       "      <td>What are the best ways to lose weight?</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>213954</th>\n",
       "      <td>What are the best ways to lose weight?</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>379679</th>\n",
       "      <td>What are the best ways to lose weight?</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>402909</th>\n",
       "      <td>What are the best ways to lose weight?</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>81363</th>\n",
       "      <td>What are the best ways to lose weight?</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>377925</th>\n",
       "      <td>What are the best ways to lose weight?</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>115816</th>\n",
       "      <td>What are the best ways to lose weight?</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>288565</th>\n",
       "      <td>What are the best ways to lose weight?</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>360010</th>\n",
       "      <td>What are the best ways to lose weight?</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>268883</th>\n",
       "      <td>What are the best ways to lose weight?</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>86631</th>\n",
       "      <td>What are the best ways to lose weight?</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>345086</th>\n",
       "      <td>What are the best ways to lose weight?</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>140355</th>\n",
       "      <td>What are the best ways to lose weight?</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>390048</th>\n",
       "      <td>What are the best ways to lose weight?</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>253672</th>\n",
       "      <td>What are the best ways to lose weight?</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>202883</th>\n",
       "      <td>What are the best ways to lose weight?</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>134629</th>\n",
       "      <td>What are the best ways to lose weight?</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>82016</th>\n",
       "      <td>What are the best ways to lose weight?</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>228265</th>\n",
       "      <td>What are the best ways to lose weight?</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>327551</th>\n",
       "      <td>What are the best ways to lose weight?</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>216562</th>\n",
       "      <td>What are the best ways to lose weight?</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>339152</th>\n",
       "      <td>What are the best ways to lose weight?</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>115228</th>\n",
       "      <td>What are the best ways to lose weight?</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>38200</th>\n",
       "      <td>What are the best ways to lose weight?</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>387610</th>\n",
       "      <td>What are the best ways to lose weight?</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>263505</th>\n",
       "      <td>What are the best ways to lose weight?</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>273689</th>\n",
       "      <td>What are the best ways to lose weight?</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>89295</th>\n",
       "      <td>What are the best ways to lose weight?</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>56239</th>\n",
       "      <td>What are the best ways to lose weight?</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>326227</th>\n",
       "      <td>What are the best ways to lose weight?</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>390396</th>\n",
       "      <td>What are the best ways to lose weight?</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>277652</th>\n",
       "      <td>What are the best ways to lose weight?</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>306135</th>\n",
       "      <td>What are the best ways to lose weight?</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>81973</th>\n",
       "      <td>What are the best ways to lose weight?</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>14712</th>\n",
       "      <td>What are the best ways to lose weight?</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                                     question1\n",
       "113625  What are the best ways to lose weight?\n",
       "213954  What are the best ways to lose weight?\n",
       "379679  What are the best ways to lose weight?\n",
       "402909  What are the best ways to lose weight?\n",
       "81363   What are the best ways to lose weight?\n",
       "377925  What are the best ways to lose weight?\n",
       "115816  What are the best ways to lose weight?\n",
       "288565  What are the best ways to lose weight?\n",
       "360010  What are the best ways to lose weight?\n",
       "268883  What are the best ways to lose weight?\n",
       "86631   What are the best ways to lose weight?\n",
       "345086  What are the best ways to lose weight?\n",
       "140355  What are the best ways to lose weight?\n",
       "390048  What are the best ways to lose weight?\n",
       "253672  What are the best ways to lose weight?\n",
       "202883  What are the best ways to lose weight?\n",
       "134629  What are the best ways to lose weight?\n",
       "82016   What are the best ways to lose weight?\n",
       "228265  What are the best ways to lose weight?\n",
       "327551  What are the best ways to lose weight?\n",
       "216562  What are the best ways to lose weight?\n",
       "339152  What are the best ways to lose weight?\n",
       "115228  What are the best ways to lose weight?\n",
       "38200   What are the best ways to lose weight?\n",
       "387610  What are the best ways to lose weight?\n",
       "263505  What are the best ways to lose weight?\n",
       "273689  What are the best ways to lose weight?\n",
       "89295   What are the best ways to lose weight?\n",
       "56239   What are the best ways to lose weight?\n",
       "326227  What are the best ways to lose weight?\n",
       "390396  What are the best ways to lose weight?\n",
       "277652  What are the best ways to lose weight?\n",
       "306135  What are the best ways to lose weight?\n",
       "81973   What are the best ways to lose weight?\n",
       "14712   What are the best ways to lose weight?"
      ]
     },
     "execution_count": 24,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "print('Total number of unique questions are: ',unique_ques)\n",
    "print('*'*75)\n",
    "print('{}% of the unique questions repeat themselves'.format(round((repeat/unique_ques)*100,2)))\n",
    "print('*'*75)\n",
    "print('The following question is repeated {} times'.format(max_ques))\n",
    "(maxi2)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "bf0a6f9e",
   "metadata": {
    "id": "bf0a6f9e"
   },
   "source": [
    "# Data Cleaning:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "id": "0f0f704e",
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "0f0f704e",
    "outputId": "d03966d7-6c20-4552-b03b-498df5271626"
   },
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "[nltk_data] Downloading package wordnet to C:\\Users\\PANCHAYAT  SAMITI\n",
      "[nltk_data]     01\\AppData\\Roaming\\nltk_data...\n",
      "[nltk_data]   Package wordnet is already up-to-date!\n",
      "[nltk_data] Downloading package omw-1.4 to C:\\Users\\PANCHAYAT  SAMITI\n",
      "[nltk_data]     01\\AppData\\Roaming\\nltk_data...\n",
      "[nltk_data]   Package omw-1.4 is already up-to-date!\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "True"
      ]
     },
     "execution_count": 25,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "lemma = WordNetLemmatizer()\n",
    "import nltk\n",
    "nltk.download('wordnet')\n",
    "nltk.download('omw-1.4')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "id": "d5d9ca82",
   "metadata": {
    "id": "d5d9ca82"
   },
   "outputs": [],
   "source": [
    "# not removing stop-words yet: \n",
    "def clean(text):\n",
    "    '''\n",
    "    This function gets rid of all punctuation marks, special characters\n",
    "    and expands the contracted words and returns words in its lemma form\n",
    "    \n",
    "    '''\n",
    "    sent = text.lower().strip() # Done for whole sentence\n",
    "    sent = re.sub('[^a-zA-Z]',' ',text) # Done for whole sentence\n",
    "    sent = sent.replace(\"′\", \"'\").replace(\"’\", \"'\")\\\n",
    "                           .replace(\"won't\", \"will not\").replace(\"cannot\", \"can not\").replace(\"can't\", \"can not\")\\\n",
    "                           .replace(\"n't\", \" not\").replace(\"what's\", \"what is\").replace(\"it's\", \"it is\")\\\n",
    "                           .replace(\"'ve\", \" have\").replace(\"i'm\", \"i am\").replace(\"'re\", \" are\")\\\n",
    "                           .replace(\"he's\", \"he is\").replace(\"she's\", \"she is\").replace(\"'s\", \" own\")\\\n",
    "                           .replace(\"%\", \" percent \").replace(\"₹\", \" rupee \").replace(\"$\", \" dollar \")\\\n",
    "                           .replace(\"€\", \" euro \").replace(\"'ll\", \" will\").replace(\"doesn't\", \"does not\")\n",
    "    sent = sent.split()\n",
    "    final = [lemma.lemmatize(word) for word in sent]\n",
    "    final_sent =  ' '.join(final)\n",
    "    return final_sent\n",
    "\n",
    "# The strip() method removes any leading (spaces at the beginning) and trailing (spaces at the end) characters"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "id": "d6a9fba8",
   "metadata": {
    "id": "d6a9fba8"
   },
   "outputs": [],
   "source": [
    "# Cleaning the train set\n",
    "\n",
    "q1_train_clean = [] # This stores all the question1 text which has been cleaned \n",
    "for z in range(len(q1_train)):\n",
    "    res = clean(str(q1_train[z]))\n",
    "    q1_train_clean.append(res)\n",
    "    \n",
    "q2_train_clean = [] # This stores all the question2 text which has been cleaned \n",
    "for v in range(len(q2_train)):\n",
    "    res2 = clean(str(q2_train[v]))\n",
    "    q2_train_clean.append(res2)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "id": "2027388e",
   "metadata": {
    "id": "2027388e"
   },
   "outputs": [],
   "source": [
    "# Cleaning the test set\n",
    "\n",
    "q1_test_clean = [] # This stores all the question1 text which has been cleaned \n",
    "for z in range(len(q1_test)):\n",
    "    res = clean(str(q1_test[z]))\n",
    "    q1_test_clean.append(res)\n",
    "    \n",
    "q2_test_clean = [] # This stores all the question2 text which has been cleaned \n",
    "for v in range(len(q2_test)):\n",
    "    res2 = clean(str(q2_test[v]))\n",
    "    q2_test_clean.append(res2)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "9d8867ee",
   "metadata": {
    "id": "9d8867ee"
   },
   "source": [
    "#  Feature Extraction: Train set"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "id": "35f78c5b",
   "metadata": {
    "id": "35f78c5b"
   },
   "outputs": [],
   "source": [
    "# Length of sentence\n",
    "q1_train_len = [len(i.split()) for i in q1_train_clean]\n",
    "q2_train_len = [len(j.split()) for j in q2_train_clean]\n",
    "\n",
    "# Common words between two sentences\n",
    "def comm(k):\n",
    "    w1 = set(map(lambda word: word.lower().strip(), q1_train_clean[k].split(\" \"))) \n",
    "    w2 = set(map(lambda word: word.lower().strip(), q2_train_clean[k].split(\" \")))    \n",
    "    return len(w1&w2)\n",
    "common = [comm(r) for r in range(len(q1_train_clean))] # storing the number of common words\n",
    "\n",
    "# Word_share: (common_word_count/total number of words)\n",
    "\n",
    "total = [a+b for a,b in zip(q1_train_len,q2_train_len)]\n",
    "word_share = [round((c/d),3) for c,d in zip(common,total)] \n",
    "\n",
    "# cwc_min & cwc_max: (Ratio of common_word_count to min & max length of word count of Q1 and Q2)\n",
    "\n",
    "min_len = [min(u,t) for u,t in zip(q1_train_len,q2_train_len)] # this returns length of either Q1 or Q2, whichever has minimum one.\n",
    "max_len = [max(u_,t_) for u_,t_ in zip(q1_train_len,q2_train_len)] # this returns length of either Q1 or Q2, whichever has maximum one.\n",
    "\n",
    "cwc_min = [round((e/f),3) if f!=0 else 0 for e,f in zip(common,min_len)] #if min_len =0, appending zero to avoid zero divison error\n",
    "cwc_max = [round((e/f),3) for e,f in zip(common,max_len)]\n",
    "\n",
    "# first word equal or not:\n",
    "\n",
    "first_word = []\n",
    "for g in range(len(q1_train_clean)):\n",
    "    if len(q1_train_clean[g].split()) != 0 and len(q2_train_clean[g].split()) != 0: # handling the condition when length of question = 0\n",
    "        s3 = q1_train_clean[g].split()[0]\n",
    "        s4 = q2_train_clean[g].split()[0]\n",
    "    \n",
    "        if s3 == s4:\n",
    "            first_word.append(1) # if first word of Q1 & Q2 is same, append 1\n",
    "        else:\n",
    "            first_word.append(0) # if first word of Q1 & Q2 is NOT same, append 0\n",
    "    else:\n",
    "        first_word.append(0)\n",
    "        \n",
    "\n",
    "# last word equal or not:\n",
    "last_word = []\n",
    "for g in range(len(q1_train_clean)):\n",
    "    if len(q1_train_clean[g].split()) != 0 and len(q2_train_clean[g].split()) != 0:\n",
    "        s3 = q1_train_clean[g].split()[-1]\n",
    "        s4 = q2_train_clean[g].split()[-1]\n",
    "        if s3 == s4: \n",
    "            last_word.append(1) # if last word of Q1 & Q2 is same, append 1. Else 0\n",
    "        else:\n",
    "            last_word.append(0)\n",
    "    else:\n",
    "        last_word.append(0)\n",
    "        \n",
    "len_diff = [abs(t1-t2) for t1,t2 in zip(q1_train_len,q2_train_len)] #this returns absolute difference between number of words in Q1 & Q2\n",
    "avg_len = [(t1+t2)/2 for t1,t2 in zip(q1_train_len,q2_train_len)] #this returns average number of words in Q1 & Q2"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "id": "26dcbb1f",
   "metadata": {
    "id": "26dcbb1f"
   },
   "outputs": [],
   "source": [
    "# Adding these new extracted features to the dataframe\n",
    "\n",
    "xy_train['q1_length'] = q1_train_len\n",
    "xy_train['q2_length'] = q2_train_len\n",
    "xy_train['common words'] = common\n",
    "xy_train['word share'] = word_share\n",
    "xy_train['cwc_min'] = cwc_min\n",
    "xy_train['cwc_max'] = cwc_max\n",
    "xy_train['first word equal'] = first_word\n",
    "xy_train['last word equal'] = last_word\n",
    "xy_train['difference in no.of words'] = len_diff\n",
    "xy_train['avg length of words'] = avg_len"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "id": "c7c6cbcb",
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 669
    },
    "id": "c7c6cbcb",
    "outputId": "2948e17a-970c-46a4-a833-5bee0d619006"
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>id</th>\n",
       "      <th>qid1</th>\n",
       "      <th>qid2</th>\n",
       "      <th>question1</th>\n",
       "      <th>question2</th>\n",
       "      <th>is_duplicate</th>\n",
       "      <th>q1_length</th>\n",
       "      <th>q2_length</th>\n",
       "      <th>common words</th>\n",
       "      <th>word share</th>\n",
       "      <th>cwc_min</th>\n",
       "      <th>cwc_max</th>\n",
       "      <th>first word equal</th>\n",
       "      <th>last word equal</th>\n",
       "      <th>difference in no.of words</th>\n",
       "      <th>avg length of words</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>95873</th>\n",
       "      <td>95873</td>\n",
       "      <td>21712</td>\n",
       "      <td>5041</td>\n",
       "      <td>Why do people often ask questions in Quora whi...</td>\n",
       "      <td>Why do people ask questions whose answer can b...</td>\n",
       "      <td>1</td>\n",
       "      <td>14</td>\n",
       "      <td>14</td>\n",
       "      <td>6</td>\n",
       "      <td>0.214</td>\n",
       "      <td>0.429</td>\n",
       "      <td>0.429</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>14.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>262013</th>\n",
       "      <td>262013</td>\n",
       "      <td>74232</td>\n",
       "      <td>33958</td>\n",
       "      <td>What are lucid dreams?</td>\n",
       "      <td>Is Lucid dreaming possible?</td>\n",
       "      <td>0</td>\n",
       "      <td>4</td>\n",
       "      <td>4</td>\n",
       "      <td>1</td>\n",
       "      <td>0.125</td>\n",
       "      <td>0.250</td>\n",
       "      <td>0.250</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>4.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>85654</th>\n",
       "      <td>85654</td>\n",
       "      <td>144574</td>\n",
       "      <td>144575</td>\n",
       "      <td>Who came first, egg or the hen?</td>\n",
       "      <td>What is first a egg or an hen?</td>\n",
       "      <td>1</td>\n",
       "      <td>7</td>\n",
       "      <td>8</td>\n",
       "      <td>4</td>\n",
       "      <td>0.267</td>\n",
       "      <td>0.571</td>\n",
       "      <td>0.500</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>7.5</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>19837</th>\n",
       "      <td>19837</td>\n",
       "      <td>37476</td>\n",
       "      <td>37477</td>\n",
       "      <td>What is the difference between reptiles, amphi...</td>\n",
       "      <td>What is the difference between amphibians and ...</td>\n",
       "      <td>0</td>\n",
       "      <td>9</td>\n",
       "      <td>8</td>\n",
       "      <td>8</td>\n",
       "      <td>0.471</td>\n",
       "      <td>1.000</td>\n",
       "      <td>0.889</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>8.5</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>105551</th>\n",
       "      <td>105551</td>\n",
       "      <td>11605</td>\n",
       "      <td>1310</td>\n",
       "      <td>What would be the best way to control anger?</td>\n",
       "      <td>How can someone control their anger?</td>\n",
       "      <td>1</td>\n",
       "      <td>9</td>\n",
       "      <td>6</td>\n",
       "      <td>2</td>\n",
       "      <td>0.133</td>\n",
       "      <td>0.333</td>\n",
       "      <td>0.222</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>3</td>\n",
       "      <td>7.5</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "            id    qid1    qid2  \\\n",
       "95873    95873   21712    5041   \n",
       "262013  262013   74232   33958   \n",
       "85654    85654  144574  144575   \n",
       "19837    19837   37476   37477   \n",
       "105551  105551   11605    1310   \n",
       "\n",
       "                                                question1  \\\n",
       "95873   Why do people often ask questions in Quora whi...   \n",
       "262013                             What are lucid dreams?   \n",
       "85654                     Who came first, egg or the hen?   \n",
       "19837   What is the difference between reptiles, amphi...   \n",
       "105551       What would be the best way to control anger?   \n",
       "\n",
       "                                                question2  is_duplicate  \\\n",
       "95873   Why do people ask questions whose answer can b...             1   \n",
       "262013                        Is Lucid dreaming possible?             0   \n",
       "85654                      What is first a egg or an hen?             1   \n",
       "19837   What is the difference between amphibians and ...             0   \n",
       "105551               How can someone control their anger?             1   \n",
       "\n",
       "        q1_length  q2_length  common words  word share  cwc_min  cwc_max  \\\n",
       "95873          14         14             6       0.214    0.429    0.429   \n",
       "262013          4          4             1       0.125    0.250    0.250   \n",
       "85654           7          8             4       0.267    0.571    0.500   \n",
       "19837           9          8             8       0.471    1.000    0.889   \n",
       "105551          9          6             2       0.133    0.333    0.222   \n",
       "\n",
       "        first word equal  last word equal  difference in no.of words  \\\n",
       "95873                  1                0                          0   \n",
       "262013                 0                0                          0   \n",
       "85654                  0                1                          1   \n",
       "19837                  1                0                          1   \n",
       "105551                 0                1                          3   \n",
       "\n",
       "        avg length of words  \n",
       "95873                  14.0  \n",
       "262013                  4.0  \n",
       "85654                   7.5  \n",
       "19837                   8.5  \n",
       "105551                  7.5  "
      ]
     },
     "execution_count": 31,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "(xy_train.head(5))"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "a9b239fc",
   "metadata": {
    "id": "a9b239fc"
   },
   "source": [
    "#  Feature Extraction: Test set"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "id": "54abe0ce",
   "metadata": {
    "id": "54abe0ce"
   },
   "outputs": [],
   "source": [
    "# Length of sentence\n",
    "q1_test_len = [len(i.split()) for i in q1_test_clean]\n",
    "q2_test_len = [len(j.split()) for j in q2_test_clean]\n",
    "\n",
    "# Common words between two sentences\n",
    "def comm(k):\n",
    "    w1 = set(map(lambda word: word.lower().strip(), q1_test_clean[k].split(\" \"))) \n",
    "    w2 = set(map(lambda word: word.lower().strip(), q2_test_clean[k].split(\" \")))    \n",
    "    return len(w1&w2)\n",
    "common_test = [comm(r) for r in range(len(q1_test_clean))] # storing the number of common words\n",
    "\n",
    "# Word_share: (common_word_count/total number of words)\n",
    "\n",
    "total_test = [a+b for a,b in zip(q1_test_len,q2_test_len)]\n",
    "word_share_test = [round((c/d),3) for c,d in zip(common_test,total_test)] \n",
    "\n",
    "# cwc_min & cwc_max: (Ratio of common_word_count to min & max length of word count of Q1 and Q2)\n",
    "\n",
    "min_len_test = [min(u,t) for u,t in zip(q1_test_len,q2_test_len)] # this returns length of either Q1 or Q2, whichever has minimum one.\n",
    "max_len_test = [max(u_,t_) for u_,t_ in zip(q1_test_len,q2_test_len)] # this returns length of either Q1 or Q2, whichever has maximum one.\n",
    "\n",
    "cwc_min_test = [round((e/f),3) if f!=0 else 0 for e,f in zip(common_test,min_len_test)] #if min_len =0, appending zero to avoid zero divison error\n",
    "cwc_max_test = [round((e/f),3) for e,f in zip(common_test,max_len_test)]\n",
    "\n",
    "# first word equal or not:\n",
    "\n",
    "first_word_test = []\n",
    "for g in range(len(q1_test_clean)):\n",
    "    if len(q1_test_clean[g].split()) != 0 and len(q2_test_clean[g].split()) != 0: # handling the condition when length of question = 0\n",
    "        s3 = q1_test_clean[g].split()[0]\n",
    "        s4 = q2_test_clean[g].split()[0]\n",
    "    \n",
    "        if s3 == s4:\n",
    "            first_word_test.append(1) # if first word of Q1 & Q2 is same, append 1\n",
    "        else:\n",
    "            first_word_test.append(0) # if first word of Q1 & Q2 is NOT same, append 0\n",
    "    else:\n",
    "        first_word_test.append(0)\n",
    "        \n",
    "\n",
    "# last word equal or not:\n",
    "last_word_test = []\n",
    "for g in range(len(q1_test_clean)):\n",
    "    if len(q1_test_clean[g].split()) != 0 and len(q2_test_clean[g].split()) != 0:\n",
    "        s3 = q1_test_clean[g].split()[-1]\n",
    "        s4 = q2_test_clean[g].split()[-1]\n",
    "        if s3 == s4: \n",
    "            last_word_test.append(1) # if last word of Q1 & Q2 is same, append 1. Else 0\n",
    "        else:\n",
    "            last_word_test.append(0)\n",
    "    else:\n",
    "        last_word_test.append(0)\n",
    "        \n",
    "len_diff_test = [abs(t1-t2) for t1,t2 in zip(q1_test_len,q2_test_len)] #this returns absolute difference between number of words in Q1 & Q2\n",
    "avg_len_test = [(t1+t2)/2 for t1,t2 in zip(q1_test_len,q2_test_len)] #this returns average number of words in Q1 & Q2"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 33,
   "id": "ebeaa076",
   "metadata": {
    "id": "ebeaa076"
   },
   "outputs": [],
   "source": [
    "# Adding these new extracted features to the test set\n",
    "\n",
    "xy_test['q1_length'] = q1_test_len\n",
    "xy_test['q2_length'] = q2_test_len\n",
    "xy_test['common words'] = common_test\n",
    "xy_test['word share'] = word_share_test\n",
    "xy_test['cwc_min'] = cwc_min_test\n",
    "xy_test['cwc_max'] = cwc_max_test\n",
    "xy_test['first word equal'] = first_word_test\n",
    "xy_test['last word equal'] = last_word_test\n",
    "xy_test['difference in no.of words'] = len_diff_test\n",
    "xy_test['avg length of words'] = avg_len_test"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 34,
   "id": "0d2d54b6",
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 721
    },
    "id": "0d2d54b6",
    "outputId": "405775b3-ee0a-4fa8-e426-9c2264ad1dbf"
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>id</th>\n",
       "      <th>qid1</th>\n",
       "      <th>qid2</th>\n",
       "      <th>question1</th>\n",
       "      <th>question2</th>\n",
       "      <th>is_duplicate</th>\n",
       "      <th>q1_length</th>\n",
       "      <th>q2_length</th>\n",
       "      <th>common words</th>\n",
       "      <th>word share</th>\n",
       "      <th>cwc_min</th>\n",
       "      <th>cwc_max</th>\n",
       "      <th>first word equal</th>\n",
       "      <th>last word equal</th>\n",
       "      <th>difference in no.of words</th>\n",
       "      <th>avg length of words</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>35865</th>\n",
       "      <td>35865</td>\n",
       "      <td>21439</td>\n",
       "      <td>44790</td>\n",
       "      <td>Any specific ideas for mechanical final year p...</td>\n",
       "      <td>What would be some final year projects for mec...</td>\n",
       "      <td>1</td>\n",
       "      <td>8</td>\n",
       "      <td>13</td>\n",
       "      <td>5</td>\n",
       "      <td>0.238</td>\n",
       "      <td>0.625</td>\n",
       "      <td>0.385</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>5</td>\n",
       "      <td>10.5</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>83858</th>\n",
       "      <td>83858</td>\n",
       "      <td>141872</td>\n",
       "      <td>141873</td>\n",
       "      <td>Can one travel to both belarus and russia with...</td>\n",
       "      <td>Can one travel to Russia on a Belarusian visa?</td>\n",
       "      <td>1</td>\n",
       "      <td>25</td>\n",
       "      <td>9</td>\n",
       "      <td>8</td>\n",
       "      <td>0.235</td>\n",
       "      <td>0.889</td>\n",
       "      <td>0.320</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>16</td>\n",
       "      <td>17.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>246204</th>\n",
       "      <td>246204</td>\n",
       "      <td>359209</td>\n",
       "      <td>359210</td>\n",
       "      <td>What was the best memory of your childhood?</td>\n",
       "      <td>What was the best memory during your childhood?</td>\n",
       "      <td>1</td>\n",
       "      <td>8</td>\n",
       "      <td>8</td>\n",
       "      <td>7</td>\n",
       "      <td>0.438</td>\n",
       "      <td>0.875</td>\n",
       "      <td>0.875</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>8.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6448</th>\n",
       "      <td>6448</td>\n",
       "      <td>12630</td>\n",
       "      <td>12631</td>\n",
       "      <td>How long do ants live?</td>\n",
       "      <td>How long does an ant typically live?</td>\n",
       "      <td>1</td>\n",
       "      <td>5</td>\n",
       "      <td>7</td>\n",
       "      <td>4</td>\n",
       "      <td>0.333</td>\n",
       "      <td>0.800</td>\n",
       "      <td>0.571</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>2</td>\n",
       "      <td>6.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>158190</th>\n",
       "      <td>158190</td>\n",
       "      <td>247180</td>\n",
       "      <td>247181</td>\n",
       "      <td>How do you stop cramps without medicine?</td>\n",
       "      <td>Hw can I stop period cramps without medicine?</td>\n",
       "      <td>1</td>\n",
       "      <td>7</td>\n",
       "      <td>8</td>\n",
       "      <td>4</td>\n",
       "      <td>0.267</td>\n",
       "      <td>0.571</td>\n",
       "      <td>0.500</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>7.5</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "            id    qid1    qid2  \\\n",
       "35865    35865   21439   44790   \n",
       "83858    83858  141872  141873   \n",
       "246204  246204  359209  359210   \n",
       "6448      6448   12630   12631   \n",
       "158190  158190  247180  247181   \n",
       "\n",
       "                                                question1  \\\n",
       "35865   Any specific ideas for mechanical final year p...   \n",
       "83858   Can one travel to both belarus and russia with...   \n",
       "246204        What was the best memory of your childhood?   \n",
       "6448                               How long do ants live?   \n",
       "158190           How do you stop cramps without medicine?   \n",
       "\n",
       "                                                question2  is_duplicate  \\\n",
       "35865   What would be some final year projects for mec...             1   \n",
       "83858      Can one travel to Russia on a Belarusian visa?             1   \n",
       "246204    What was the best memory during your childhood?             1   \n",
       "6448                 How long does an ant typically live?             1   \n",
       "158190      Hw can I stop period cramps without medicine?             1   \n",
       "\n",
       "        q1_length  q2_length  common words  word share  cwc_min  cwc_max  \\\n",
       "35865           8         13             5       0.238    0.625    0.385   \n",
       "83858          25          9             8       0.235    0.889    0.320   \n",
       "246204          8          8             7       0.438    0.875    0.875   \n",
       "6448            5          7             4       0.333    0.800    0.571   \n",
       "158190          7          8             4       0.267    0.571    0.500   \n",
       "\n",
       "        first word equal  last word equal  difference in no.of words  \\\n",
       "35865                  0                0                          5   \n",
       "83858                  1                0                         16   \n",
       "246204                 1                1                          0   \n",
       "6448                   1                1                          2   \n",
       "158190                 0                1                          1   \n",
       "\n",
       "        avg length of words  \n",
       "35865                  10.5  \n",
       "83858                  17.0  \n",
       "246204                  8.0  \n",
       "6448                    6.0  \n",
       "158190                  7.5  "
      ]
     },
     "execution_count": 34,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "xy_test.head(5)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "16015997",
   "metadata": {
    "id": "16015997"
   },
   "source": [
    "# Advanced Feature Extraction (Fuzzwuzzy) : Train set"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "f9306604",
   "metadata": {},
   "source": [
    "https://towardsdatascience.com/string-matching-with-fuzzywuzzy-e982c61f8a84"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 35,
   "id": "df10dac0",
   "metadata": {
    "id": "df10dac0"
   },
   "outputs": [],
   "source": [
    "# To get longest substring ratio\n",
    "\n",
    "def lsubstring_ratio(a, b): # abcdxyz, abcdefg ---> 4\n",
    "    strs = list(distance.lcsubstrings(a, b)) # This returns the length of longest common substring\n",
    "    if len(strs) == 0:\n",
    "        return 0\n",
    "    else:\n",
    "        return len(strs[0]) / (min(len(a), len(b)) + 1) # This gives us the LCS ratio\n",
    "\n",
    "# partial ratio:\n",
    "partial = [fuzz.partial_ratio(k.split(),l.split()) for k,l in zip(q1_train_clean,q2_train_clean)]\n",
    "\n",
    "# token sort ratio:\n",
    "token_sort = [fuzz.token_sort_ratio(k.split(),l.split()) for k,l in zip(q1_train_clean,q2_train_clean)]\n",
    "\n",
    "# token set ratio:\n",
    "token_set = [fuzz.token_set_ratio(k.split(),l.split()) for k,l in zip(q1_train_clean,q2_train_clean)]\n",
    "\n",
    "# WRatio:\n",
    "wratio = [fuzz.WRatio(k.split(),l.split()) for k,l in zip(q1_train_clean,q2_train_clean)]\n",
    "\n",
    "# LCSubstring ratio\n",
    "lcs = [lsubstring_ratio(k,l) for k,l in zip(q1_train_clean,q2_train_clean)]\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 36,
   "id": "fce70964",
   "metadata": {
    "id": "fce70964"
   },
   "outputs": [],
   "source": [
    "# Adding these new advanced features to the dataframe:\n",
    "\n",
    "xy_train[\"token_set_ratio\"] = token_set\n",
    "xy_train[\"token_sort_ratio\"] = token_sort\n",
    "xy_train[\"fuzz_WRatio\"] = wratio\n",
    "xy_train[\"fuzz_partial_ratio\"] = partial\n",
    "xy_train[\"longest_substr_ratio\"]  = lcs"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 37,
   "id": "21cf57fc",
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 699
    },
    "id": "21cf57fc",
    "outputId": "c2f0b2e8-143a-485c-ec5b-c9e163bea998"
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>id</th>\n",
       "      <th>qid1</th>\n",
       "      <th>qid2</th>\n",
       "      <th>question1</th>\n",
       "      <th>question2</th>\n",
       "      <th>is_duplicate</th>\n",
       "      <th>q1_length</th>\n",
       "      <th>q2_length</th>\n",
       "      <th>common words</th>\n",
       "      <th>word share</th>\n",
       "      <th>...</th>\n",
       "      <th>cwc_max</th>\n",
       "      <th>first word equal</th>\n",
       "      <th>last word equal</th>\n",
       "      <th>difference in no.of words</th>\n",
       "      <th>avg length of words</th>\n",
       "      <th>token_set_ratio</th>\n",
       "      <th>token_sort_ratio</th>\n",
       "      <th>fuzz_WRatio</th>\n",
       "      <th>fuzz_partial_ratio</th>\n",
       "      <th>longest_substr_ratio</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>95873</th>\n",
       "      <td>95873</td>\n",
       "      <td>21712</td>\n",
       "      <td>5041</td>\n",
       "      <td>Why do people often ask questions in Quora whi...</td>\n",
       "      <td>Why do people ask questions whose answer can b...</td>\n",
       "      <td>1</td>\n",
       "      <td>14</td>\n",
       "      <td>14</td>\n",
       "      <td>6</td>\n",
       "      <td>0.214</td>\n",
       "      <td>...</td>\n",
       "      <td>0.429</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>14.0</td>\n",
       "      <td>66</td>\n",
       "      <td>63</td>\n",
       "      <td>70</td>\n",
       "      <td>70</td>\n",
       "      <td>0.184211</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>262013</th>\n",
       "      <td>262013</td>\n",
       "      <td>74232</td>\n",
       "      <td>33958</td>\n",
       "      <td>What are lucid dreams?</td>\n",
       "      <td>Is Lucid dreaming possible?</td>\n",
       "      <td>0</td>\n",
       "      <td>4</td>\n",
       "      <td>4</td>\n",
       "      <td>1</td>\n",
       "      <td>0.125</td>\n",
       "      <td>...</td>\n",
       "      <td>0.250</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>4.0</td>\n",
       "      <td>52</td>\n",
       "      <td>52</td>\n",
       "      <td>56</td>\n",
       "      <td>61</td>\n",
       "      <td>0.476190</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>85654</th>\n",
       "      <td>85654</td>\n",
       "      <td>144574</td>\n",
       "      <td>144575</td>\n",
       "      <td>Who came first, egg or the hen?</td>\n",
       "      <td>What is first a egg or an hen?</td>\n",
       "      <td>1</td>\n",
       "      <td>7</td>\n",
       "      <td>8</td>\n",
       "      <td>4</td>\n",
       "      <td>0.267</td>\n",
       "      <td>...</td>\n",
       "      <td>0.500</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>7.5</td>\n",
       "      <td>76</td>\n",
       "      <td>72</td>\n",
       "      <td>80</td>\n",
       "      <td>78</td>\n",
       "      <td>0.266667</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>19837</th>\n",
       "      <td>19837</td>\n",
       "      <td>37476</td>\n",
       "      <td>37477</td>\n",
       "      <td>What is the difference between reptiles, amphi...</td>\n",
       "      <td>What is the difference between amphibians and ...</td>\n",
       "      <td>0</td>\n",
       "      <td>9</td>\n",
       "      <td>8</td>\n",
       "      <td>8</td>\n",
       "      <td>0.471</td>\n",
       "      <td>...</td>\n",
       "      <td>0.889</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>8.5</td>\n",
       "      <td>100</td>\n",
       "      <td>94</td>\n",
       "      <td>95</td>\n",
       "      <td>86</td>\n",
       "      <td>0.584906</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>105551</th>\n",
       "      <td>105551</td>\n",
       "      <td>11605</td>\n",
       "      <td>1310</td>\n",
       "      <td>What would be the best way to control anger?</td>\n",
       "      <td>How can someone control their anger?</td>\n",
       "      <td>1</td>\n",
       "      <td>9</td>\n",
       "      <td>6</td>\n",
       "      <td>2</td>\n",
       "      <td>0.133</td>\n",
       "      <td>...</td>\n",
       "      <td>0.222</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>3</td>\n",
       "      <td>7.5</td>\n",
       "      <td>54</td>\n",
       "      <td>51</td>\n",
       "      <td>56</td>\n",
       "      <td>61</td>\n",
       "      <td>0.250000</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>5 rows × 21 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "            id    qid1    qid2  \\\n",
       "95873    95873   21712    5041   \n",
       "262013  262013   74232   33958   \n",
       "85654    85654  144574  144575   \n",
       "19837    19837   37476   37477   \n",
       "105551  105551   11605    1310   \n",
       "\n",
       "                                                question1  \\\n",
       "95873   Why do people often ask questions in Quora whi...   \n",
       "262013                             What are lucid dreams?   \n",
       "85654                     Who came first, egg or the hen?   \n",
       "19837   What is the difference between reptiles, amphi...   \n",
       "105551       What would be the best way to control anger?   \n",
       "\n",
       "                                                question2  is_duplicate  \\\n",
       "95873   Why do people ask questions whose answer can b...             1   \n",
       "262013                        Is Lucid dreaming possible?             0   \n",
       "85654                      What is first a egg or an hen?             1   \n",
       "19837   What is the difference between amphibians and ...             0   \n",
       "105551               How can someone control their anger?             1   \n",
       "\n",
       "        q1_length  q2_length  common words  word share  ...  cwc_max  \\\n",
       "95873          14         14             6       0.214  ...    0.429   \n",
       "262013          4          4             1       0.125  ...    0.250   \n",
       "85654           7          8             4       0.267  ...    0.500   \n",
       "19837           9          8             8       0.471  ...    0.889   \n",
       "105551          9          6             2       0.133  ...    0.222   \n",
       "\n",
       "        first word equal  last word equal  difference in no.of words  \\\n",
       "95873                  1                0                          0   \n",
       "262013                 0                0                          0   \n",
       "85654                  0                1                          1   \n",
       "19837                  1                0                          1   \n",
       "105551                 0                1                          3   \n",
       "\n",
       "        avg length of words  token_set_ratio  token_sort_ratio  fuzz_WRatio  \\\n",
       "95873                  14.0               66                63           70   \n",
       "262013                  4.0               52                52           56   \n",
       "85654                   7.5               76                72           80   \n",
       "19837                   8.5              100                94           95   \n",
       "105551                  7.5               54                51           56   \n",
       "\n",
       "        fuzz_partial_ratio  longest_substr_ratio  \n",
       "95873                   70              0.184211  \n",
       "262013                  61              0.476190  \n",
       "85654                   78              0.266667  \n",
       "19837                   86              0.584906  \n",
       "105551                  61              0.250000  \n",
       "\n",
       "[5 rows x 21 columns]"
      ]
     },
     "execution_count": 37,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "xy_train.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 38,
   "id": "YHh7LMIVR6or",
   "metadata": {
    "id": "YHh7LMIVR6or"
   },
   "outputs": [],
   "source": [
    "xy_train.to_csv('train.csv')"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "c4700090",
   "metadata": {
    "id": "c4700090"
   },
   "source": [
    "# Advanced Feature Extraction (Fuzzwuzzy) : Test set"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 39,
   "id": "93031cac",
   "metadata": {
    "id": "93031cac"
   },
   "outputs": [],
   "source": [
    "# To get longest substring ratio\n",
    "\n",
    "def lsubstring_ratio(a, b):\n",
    "    strs = list(distance.lcsubstrings(a, b)) # This returns the length of longest common substring\n",
    "    if len(strs) == 0:\n",
    "        return 0\n",
    "    else:\n",
    "        return len(strs[0]) / (min(len(a), len(b)) + 1) # This gives us the LCS ratio\n",
    "\n",
    "# partial ratio:\n",
    "partial_test = [fuzz.partial_ratio(k.split(),l.split()) for k,l in zip(q1_test_clean,q2_test_clean)]\n",
    "\n",
    "# token sort ratio:\n",
    "token_sort_test = [fuzz.token_sort_ratio(k.split(),l.split()) for k,l in zip(q1_test_clean,q2_test_clean)]\n",
    "\n",
    "# token set ratio:\n",
    "token_set_test = [fuzz.token_set_ratio(k.split(),l.split()) for k,l in zip(q1_test_clean,q2_test_clean)]\n",
    "\n",
    "# WRatio:\n",
    "wratio_test = [fuzz.WRatio(k.split(),l.split()) for k,l in zip(q1_test_clean,q2_test_clean)]\n",
    "\n",
    "# LCSubstring ratio\n",
    "lcs_test = [lsubstring_ratio(k,l) for k,l in zip(q1_test_clean,q2_test_clean)]\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 40,
   "id": "b1d085d8",
   "metadata": {
    "id": "b1d085d8"
   },
   "outputs": [],
   "source": [
    "# Adding these new advanced features to the dataframe:\n",
    "\n",
    "xy_test[\"token_set_ratio\"] = token_set_test\n",
    "xy_test[\"token_sort_ratio\"] = token_sort_test\n",
    "xy_test[\"fuzz_WRatio\"] = wratio_test\n",
    "xy_test[\"fuzz_partial_ratio\"] = partial_test\n",
    "xy_test[\"longest_substr_ratio\"]  = lcs_test"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 41,
   "id": "80be74e0",
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 751
    },
    "id": "80be74e0",
    "outputId": "65569acf-cf13-4992-a82b-190079d04dd1"
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>id</th>\n",
       "      <th>qid1</th>\n",
       "      <th>qid2</th>\n",
       "      <th>question1</th>\n",
       "      <th>question2</th>\n",
       "      <th>is_duplicate</th>\n",
       "      <th>q1_length</th>\n",
       "      <th>q2_length</th>\n",
       "      <th>common words</th>\n",
       "      <th>word share</th>\n",
       "      <th>...</th>\n",
       "      <th>cwc_max</th>\n",
       "      <th>first word equal</th>\n",
       "      <th>last word equal</th>\n",
       "      <th>difference in no.of words</th>\n",
       "      <th>avg length of words</th>\n",
       "      <th>token_set_ratio</th>\n",
       "      <th>token_sort_ratio</th>\n",
       "      <th>fuzz_WRatio</th>\n",
       "      <th>fuzz_partial_ratio</th>\n",
       "      <th>longest_substr_ratio</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>35865</th>\n",
       "      <td>35865</td>\n",
       "      <td>21439</td>\n",
       "      <td>44790</td>\n",
       "      <td>Any specific ideas for mechanical final year p...</td>\n",
       "      <td>What would be some final year projects for mec...</td>\n",
       "      <td>1</td>\n",
       "      <td>8</td>\n",
       "      <td>13</td>\n",
       "      <td>5</td>\n",
       "      <td>0.238</td>\n",
       "      <td>...</td>\n",
       "      <td>0.385</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>5</td>\n",
       "      <td>10.5</td>\n",
       "      <td>79</td>\n",
       "      <td>60</td>\n",
       "      <td>86</td>\n",
       "      <td>66</td>\n",
       "      <td>0.365385</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>83858</th>\n",
       "      <td>83858</td>\n",
       "      <td>141872</td>\n",
       "      <td>141873</td>\n",
       "      <td>Can one travel to both belarus and russia with...</td>\n",
       "      <td>Can one travel to Russia on a Belarusian visa?</td>\n",
       "      <td>1</td>\n",
       "      <td>25</td>\n",
       "      <td>9</td>\n",
       "      <td>8</td>\n",
       "      <td>0.235</td>\n",
       "      <td>...</td>\n",
       "      <td>0.320</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>16</td>\n",
       "      <td>17.0</td>\n",
       "      <td>86</td>\n",
       "      <td>54</td>\n",
       "      <td>86</td>\n",
       "      <td>85</td>\n",
       "      <td>0.391304</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>246204</th>\n",
       "      <td>246204</td>\n",
       "      <td>359209</td>\n",
       "      <td>359210</td>\n",
       "      <td>What was the best memory of your childhood?</td>\n",
       "      <td>What was the best memory during your childhood?</td>\n",
       "      <td>1</td>\n",
       "      <td>8</td>\n",
       "      <td>8</td>\n",
       "      <td>7</td>\n",
       "      <td>0.438</td>\n",
       "      <td>...</td>\n",
       "      <td>0.875</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>8.0</td>\n",
       "      <td>96</td>\n",
       "      <td>88</td>\n",
       "      <td>94</td>\n",
       "      <td>91</td>\n",
       "      <td>0.571429</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6448</th>\n",
       "      <td>6448</td>\n",
       "      <td>12630</td>\n",
       "      <td>12631</td>\n",
       "      <td>How long do ants live?</td>\n",
       "      <td>How long does an ant typically live?</td>\n",
       "      <td>1</td>\n",
       "      <td>5</td>\n",
       "      <td>7</td>\n",
       "      <td>4</td>\n",
       "      <td>0.333</td>\n",
       "      <td>...</td>\n",
       "      <td>0.571</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>2</td>\n",
       "      <td>6.0</td>\n",
       "      <td>92</td>\n",
       "      <td>74</td>\n",
       "      <td>86</td>\n",
       "      <td>83</td>\n",
       "      <td>0.523810</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>158190</th>\n",
       "      <td>158190</td>\n",
       "      <td>247180</td>\n",
       "      <td>247181</td>\n",
       "      <td>How do you stop cramps without medicine?</td>\n",
       "      <td>Hw can I stop period cramps without medicine?</td>\n",
       "      <td>1</td>\n",
       "      <td>7</td>\n",
       "      <td>8</td>\n",
       "      <td>4</td>\n",
       "      <td>0.267</td>\n",
       "      <td>...</td>\n",
       "      <td>0.500</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>7.5</td>\n",
       "      <td>83</td>\n",
       "      <td>74</td>\n",
       "      <td>83</td>\n",
       "      <td>80</td>\n",
       "      <td>0.589744</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>5 rows × 21 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "            id    qid1    qid2  \\\n",
       "35865    35865   21439   44790   \n",
       "83858    83858  141872  141873   \n",
       "246204  246204  359209  359210   \n",
       "6448      6448   12630   12631   \n",
       "158190  158190  247180  247181   \n",
       "\n",
       "                                                question1  \\\n",
       "35865   Any specific ideas for mechanical final year p...   \n",
       "83858   Can one travel to both belarus and russia with...   \n",
       "246204        What was the best memory of your childhood?   \n",
       "6448                               How long do ants live?   \n",
       "158190           How do you stop cramps without medicine?   \n",
       "\n",
       "                                                question2  is_duplicate  \\\n",
       "35865   What would be some final year projects for mec...             1   \n",
       "83858      Can one travel to Russia on a Belarusian visa?             1   \n",
       "246204    What was the best memory during your childhood?             1   \n",
       "6448                 How long does an ant typically live?             1   \n",
       "158190      Hw can I stop period cramps without medicine?             1   \n",
       "\n",
       "        q1_length  q2_length  common words  word share  ...  cwc_max  \\\n",
       "35865           8         13             5       0.238  ...    0.385   \n",
       "83858          25          9             8       0.235  ...    0.320   \n",
       "246204          8          8             7       0.438  ...    0.875   \n",
       "6448            5          7             4       0.333  ...    0.571   \n",
       "158190          7          8             4       0.267  ...    0.500   \n",
       "\n",
       "        first word equal  last word equal  difference in no.of words  \\\n",
       "35865                  0                0                          5   \n",
       "83858                  1                0                         16   \n",
       "246204                 1                1                          0   \n",
       "6448                   1                1                          2   \n",
       "158190                 0                1                          1   \n",
       "\n",
       "        avg length of words  token_set_ratio  token_sort_ratio  fuzz_WRatio  \\\n",
       "35865                  10.5               79                60           86   \n",
       "83858                  17.0               86                54           86   \n",
       "246204                  8.0               96                88           94   \n",
       "6448                    6.0               92                74           86   \n",
       "158190                  7.5               83                74           83   \n",
       "\n",
       "        fuzz_partial_ratio  longest_substr_ratio  \n",
       "35865                   66              0.365385  \n",
       "83858                   85              0.391304  \n",
       "246204                  91              0.571429  \n",
       "6448                    83              0.523810  \n",
       "158190                  80              0.589744  \n",
       "\n",
       "[5 rows x 21 columns]"
      ]
     },
     "execution_count": 41,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "xy_test.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 42,
   "id": "__S7mSJoR_2i",
   "metadata": {
    "id": "__S7mSJoR_2i"
   },
   "outputs": [],
   "source": [
    "xy_test.to_csv(\"test.csv\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 44,
   "id": "FQi2FQU0r11P",
   "metadata": {
    "id": "FQi2FQU0r11P"
   },
   "outputs": [],
   "source": [
    "#xy_train = pd.read_csv(\"/content/train.csv\")\n",
    "#xy_test = pd.read_csv(\"/content/test.csv\")"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "7112b5b0",
   "metadata": {
    "id": "7112b5b0"
   },
   "source": [
    "# Data Visualization :"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "6667b4ad",
   "metadata": {
    "id": "6667b4ad"
   },
   "source": [
    "# Wordcloud"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 45,
   "id": "91ed703d",
   "metadata": {
    "id": "91ed703d"
   },
   "outputs": [],
   "source": [
    "duplicate = xy_train[xy_train['is_duplicate'] == 1] # Storing data having class label as 1 \n",
    "not_duplicate = xy_train[xy_train['is_duplicate'] == 0] # Storing data having class label as 0 \n",
    "\n",
    "similar = np.dstack([duplicate[\"question1\"],duplicate[\"question2\"]]).flatten()\n",
    "no_similar = np.dstack([not_duplicate[\"question1\"], not_duplicate[\"question2\"]]).flatten()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 46,
   "id": "6caaac4d",
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "6caaac4d",
    "outputId": "ed07a77b-f2e8-4fb1-e098-292b1d1f1fd3"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[[[1 4]\n",
      "  [2 5]\n",
      "  [3 6]]]\n",
      "[1 4 2 5 3 6]\n"
     ]
    }
   ],
   "source": [
    "# An example of using np.dstack and flatten() operation:\n",
    "gfg1 = np.array([1, 2, 3])\n",
    "gfg2 = np.array([4, 5, 6])\n",
    "\n",
    "print(np.dstack((gfg1, gfg2)))\n",
    "print(np.dstack((gfg1, gfg2)).flatten())\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 47,
   "id": "95847fb0",
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "95847fb0",
    "outputId": "b641937e-bc4a-400e-bdac-211259715426"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "['Why do people often ask questions in Quora while they can Google it themselves?'\n",
      " 'Why do people ask questions whose answer can be easily found on the internet?'\n",
      " 'Who came first, egg or the hen?' 'What is first a egg or an hen?']\n"
     ]
    }
   ],
   "source": [
    "print((similar)[:4])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 48,
   "id": "512c373d",
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "512c373d",
    "outputId": "1a2a7917-1f00-4135-d563-7fddbd32dbbf"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "['What are lucid dreams?' 'Is Lucid dreaming possible?'\n",
      " 'What is the difference between reptiles, amphibians, and mammals?'\n",
      " 'What is the difference between amphibians and reptiles?']\n"
     ]
    }
   ],
   "source": [
    "print(no_similar[:4])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 49,
   "id": "5ccd5f38",
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "5ccd5f38",
    "outputId": "e0837234-f21c-41ea-8f34-3e3206230b72"
   },
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "[nltk_data] Downloading package stopwords to C:\\Users\\PANCHAYAT\n",
      "[nltk_data]     SAMITI 01\\AppData\\Roaming\\nltk_data...\n",
      "[nltk_data]   Package stopwords is already up-to-date!\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "True"
      ]
     },
     "execution_count": 49,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "import nltk\n",
    "nltk.download('stopwords')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 50,
   "id": "a8339f82",
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 265
    },
    "id": "a8339f82",
    "outputId": "a6f2397e-0cf1-45ed-b510-66f7361ac2a2"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Word Cloud for Duplicate question\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Word Cloud for Duplicate question:\n",
    "stop_words = set(stopwords.words(\"english\")) # to remove the stopwords\n",
    "\n",
    "duplicate_words = ''\n",
    "for j in (similar): # This loop extracts words from sentences given to it as input\n",
    "    c = (str(j).split())\n",
    "    for v in c:\n",
    "        b = v.lower()\n",
    "        duplicate_words += \"\".join(b)+\" \"\n",
    "            \n",
    "wc_q = WordCloud(width = 800, height = 800,background_color ='white',stopwords = stop_words,min_font_size = 10)\n",
    "wc_q.generate(duplicate_words)\n",
    "print('Word Cloud for Duplicate question')\n",
    "plt.imshow(wc_q, interpolation='bilinear')\n",
    "plt.axis(\"off\")\n",
    "plt.show()\n",
    "    "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 51,
   "id": "4cf150aa",
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 265
    },
    "id": "4cf150aa",
    "outputId": "571d5bc9-34da-476f-f679-2b1c33ebf2d9"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Word Cloud for Non Duplicate question\n"
     ]
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAOcAAADnCAYAAADl9EEgAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjUuMSwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy/YYfK9AAAACXBIWXMAAAsTAAALEwEAmpwYAAEAAElEQVR4nOz9d3gc2Xnni38qV+dudKORcyIJ5iGHM8OJ1MxoNKM0o2DJklbyWg5ar3ed7vru2nt/3nTv+mf/vPvY6+t1zpIseZQsS5qkyYnDnEmQIIicGp1TpXP/aBAEiEBQkr1+7k9fPngIdFWfOlV13nPe84bvKwkh+CF+iB/inx7k/9Ud+CF+iB9ibfxQOH+IH+KfKH4onD/ED/FPFD8Uzh/ih/gnih8K5w/xQ/wThXqL4ytMuUIIBAJZkpf+vg5Jkn7gnfunjopb5GLuDQYjD6DK+vfdnhCCK4UjhLUESbPrB9BDcLwqR+b/AL+aYHvso8iSsnTM9SxOp7+IKvvZFn1yxbHl8ITLVOkYw4UXKTnzaLKf/vDjtPj3r3jvjlflWuFVxopvUfWy+JQ6BqMfJm72/UDu5f/FWFN4biWcK7BgLXCpcJm4Xkebv5WZygxH08fZEhpgW3jrmgKamc8jyxLVsoUkS0QTIVJTGVRdJZaMIMsSQggc22VhOotVtYnEg4SiASS51p4QgnKhQmY+j2O7BCN+IokQiiIjhCA7nwfA8OkszGSRJIm6xgiGT1/qkxACT2TxvBKqksDx0khoSJKK5+WR5TAg4XrzKHIUWYogSRKOZ5F3FnCFjSkH8KsRJGSqXpGx0jku5N6g3uxAl32EtXp02cQTLiU3R8UtoEo6QbUOVdYAqLgFPOEiSypFJ4MuGwTVOjxccvY8Z7Iv0RHYiYSMqQQJqnXf18QncJmrnCOst3LTXIsrLGYqZzDkMAIXWFs4U9VLvDb7G4S1FpLmILaoIN0kyEIIRouv89bcb9Po203C2IIjyqvO+yFq8DyPUqFKIGSu+35vSzjLboVTmdO0+luYq85TcPJ0B7oZKlymN9iDoRirvvPc519n7NIUuYUC+XSRux/fw7EXz+JYLj/zm5+gY0szmbkcX/7tZxg6PgKAZmh88Kffxb6HtyPLMhNXZviz//RVMnM5AFzH5d2fvJdHfvQgsiLznb98ldGLU2iGytTVOcrFKl3bWvjMv3+KuobI9cdBvvw8shREV9upOkMI4YBwkCQdkJAkHcedwafvxtR3Y3tVXp/7EgvWJIqkICFzX/LjRLUGrhZOci73CnPVaxxf+A664uOO2BPUGc1MlS/zzsLfIQFlt0C7f5ADiSdRJJUrhWOMFk+jLApnUKvjgeQnKTlZjqefYbx0gapbYrp8mTb/INsi9yOtPbF+39DkAPckfx5ZUpHR1jxHCMFM+TSecLmr/l8R1TuXjq0cVILx4mGCWiP3JH8enxrj/1996NWKRaVkYfh0PNfDsV0CIZNCtoyqqximSmomx/mjIzzw/j3rtnNbwqnKKl2BDgZCA7y9cBhXuGwLb2OiPIEjHAxWC2elWGVieJbP/deP8ef/+Wsce/Es//zXPsyf/aevcO7ty7T2NvCNP3yR8aFpfvq/foxQLMArXz3CX/36N2gfaKaxI0G4Lsh7Pn0fTZ1JNF3lhb95k7/7oxfZ9/AO4k1RqmWLI8+f4TP//kk+8W/ex8xoit/5xb/iyPNneORH71kcRBKG2kvZOoksm8iSiSvyCFx8+nbK1klAQZJMNLUNSZIo2TnGSud4IPlJGs1uql4ZnxICJAbCd2EofhzP4l2N/xxNNpEXV56E0cahhs/gU4JMla/wytxfsSP6LkJaHa6wmSwP8Xjzz1CnN+MJF00yiGgN3JP4CHOVUfbVPUFncBfS4r/bgRAeRWeWkrOArgQwlSg3a01CeFheEYGHJvtRpNWCWdM0bBxRpejMo0g1tb3q5ZCQ0OQAEgpCeLjCwvYqlN0FNMmPJ2wqbhYZBU32L11fCIEjKhTsaWyvjKlECGoNyJK64rq2V6pNGJJKyZmj5CygygZBtQFdCa44t+KmKTpzAATUJKYSXaYteVheCVU2cD2LvDOFJ2x8SpyAmkCSVppchBBUvRxFexZXWKiyD78ax5BDK851vAp5exrbK2EoYYJaw4pneObwcG2B2NLM6bcv49geO+7qYXYizfxUhqb2OAuzObLp4obv8raEs06Pock6by+8Q6uvhbxT4K3U25iKiSGvFszraOlO0rm1lZaeBoQQdG5tob6ljmyqtpoeef4097x3L4oiU8pV6NjaTO6PCwyfGaOxI0EoFmDHwQHyC0XKxQpNXUkqZYtSoUJ88RqtfY3c/+R+ghE/sfowrb0NTI3MruiHLAfwGXvR1Q4s5xq6aiBLAWQ5hE/fQ6n6NkLYVJ0rKHKcgBqhN7SPt1JfocHsYkvoHgJq7eVLKMiSgoSELKkoiwNMCIHlVRgpnCBrz1Jyc1TdEq6wl/qRMFppMHuW9u7XoUhKrW1JXmrvduAKm6Hstzmb+VtcYaPKJo3mTlxhrTjP8kq8NffbZK0xbK9CS2AfdyY+h7RMrS27CxxN/RFZa5ycNUbVy/Pi9H9ERsWnRrkn+YuEtEbS1lWOp/6MkpMibV1FIHhu8t8BEjGji7vr/zW6EkAIQcYa4VjqT1moXgGpNvF0BO9jZ+zjGEoIAE/YHJ7/PaJ6O7KkcTH7TSyvgBAeu+s+xZbo+xfPcxjOv8jZzN9ieQUQYChhdsY+TnvwILKkUHYzvD7zGzT6dzNTPk3WGsUVNoqksT32I/SFH1vaZ7vC5mr+Rc5lvkLJWVgUcEG9OcjB5C9gKCGEEOTscY6l/pT56sWlibMtcBc7Y5/Ep0ZrfXM9egZb8AdNUjM56puipOfypGdzZObzmH6d+uYY5VJ1w/d5WyNAkzTuid9NyS0hSzISYHs2PsWPKq/flOHTkWUJWZFRNQVZkZCk2l6zVKiQWyjwytfe4fhL54DaAA/FAkv70ZFzE3z1954jly5i+nWK2TJW2YZlalO0PoThq83ukiyhagqe6y0dlyQZQ7thmPDpO1b0UZYCeKKE5+XR1Y7aw5F1DsQ/SH/oLq4UjvDczB9xKPkZmv39S9+7WXFzhc2rc59Hl31sC99H2SswXR5ecY4iaeusiN/H3lIIZstnObbwp3QE7mVL9AN4wuFc5mnS1lVixg0Dkyb72FX3KUrOPG/P/S5Fe3ZVe7ocoC/8HlyvylDuGabKx9ld9ykMJYwiaYsrMvjVerZGn8QTNsdTf46Hw974jyFLGrocQF2ctCtulrfn/m88HO5t+CV8Sh0zldMcS/0puhxgR+xjyJKCQFCwp5kunyBuDLAv8RP4lTgld56w1rZ0r1PlExxN/SE9oUfoDj2EJzzOZ7/G4fn/m6DWSNzoQwiHjHWN2co5tkQ/wN74j+EKm1MLf83Jhb+k0beTiN5W2y8XXuftud+lxb+POxI/gSmHKbtpPOEsrv5geQXemf+fVNwcB5O/gF9NMFc5z9H5P0aVTPbEfwxZUmjrbSAU9ePzG9zz7h2UClWSLTFkWaK9r4F4Q4SJq3P07Wjb8J3elnBWvArfmXqWkltCkiSCSpB3NTyEX/Vt/MUNDBq6qeEP+XjfZx/i7vfsXv4l/CETu+rwxd/6FrIi8VP/5UeIxIMMnbjGb//8X65oR5a/v32ZJCmrBLbiFkhb0wTVGL3B/VwtnCDvLCwd1ySDqlckbU0RUKKYShBXOGTtOQYjDxDW65nOXcEWlc31AQlNMkhVx0gaHSiyjikHNmkQqhlkVMlgR+xjhPUWhBBsj36U8eLhFWfKkkJUbyeg1i+tWDdDlU0afTsRQjBdPsVc9TyNvl341fiK80wlTLN/L55wOJ/9Bq5Xpcm3d0kor2OmfIpU9RIPNv17Gn27kSSJkN7MZOkYV/Mv0h9+HJ8aWzrfFTa76z5JVO+sGfhE79IxD4eh7HcIqA3siP0IhhJeuteJ4mFGi68TN/oWn4qgzuhme/QjS6tff+QJXpr6T+TtSSJ6G46ocCn3LUJaI3fWfw6fEl9aPJZjrnKOmfIZ7mv4ZZp8e5EkibDWwnT5FCOFV9gS/SABNYE/HiBVKEO5glLnI1Tno4wg2l17dpYis/2uHlRlY2PZbQlnxsqiKzqPNT2KjIwkSWhr7FduB5F4iMG7ejny/Bl23buFWEMEu2KTXSgQCPtwHZfMXI7Bu3qJJcPYVYcTr1ygcguV4AcByytzPP0dKm4BCZkmXx/t/sGl4wmjnc7ATl6d/QIBNcrdiaeIaEl2Rx/hXO5VRouniRuttPsHl9RUUw4S0uJrXk+RNHbFHuF4+juMly/QHzrAtvB9bGZFdYVNujpCSGvGtyhAkiQR0JKrBOofA6sGdvUCAo9UZYiiPbf0ueUVKTkpql4eHzeEM6K1EdKalyam5ROU5RZZsK6gy0GuFV7j+vOxvRICQcYaRXBDa6oz+tDlwFI7hhxGkmQcrzaGqm6WrDVKR/C+JcG8+ZpCCOYrl/CEQ9oapuJml45V3Txld4GqmyWgJjh8cYwvv3oSgHLVplixiAZ9SECmWKGvJcG/+ciDxIL+DZ/hbQmnX/WRtXNcyF3Er/rRJI02f+uGAiorMopS21spy39XZWRFQtNVPvyzj/FX//Xr/Obn/hjDb+A6LvHGKJ/79Y8TigV44Kn9fPNPXmL04hQgiDVESLbFl1ZkWZFRVGXFEFZUBVn5/mIsQmqCRxt/EltYSEjosoksKXhCIITAUEzujv8ItiijSApVS6KCy5bwQbqDewGBJps4wkZdNKh0B/fSGdy55vUkSaIrsJtW/xY84aJKBptVdQUejihjKGHkZbElsqQtGXP+MTG+kOXM2Ax7u1pIhgNYbh7bKzOc/+4KAxBA3Oxb9Zki6xv4XS0cr0zFzXAh+3crjgXUJCGtccVnmuxj7edYm0BcYeMIa1Fo13/elpfHERWu5l9CvmnMx42Bpc/u2dbBzq4mLMfl8y8eY7Cjkd09zUhInBmZ4uL4HD791ovabQmnIinU6wmmKlPIkoJP8dHka0RbwwzveC4ADzy1H8d2kWRpyXIqSRLv+mf3MOmk8YRHQ3ucf/Ebn2B8aJpcqoAZ0GnsrCcUC+Dg8a6P38PAHV2kZ3NE4kGauxuYuDxNfXNtpn3oQwcoF6vIam1QyorCe3/6EG8tTDBTKtAYWFt1uxnD2QVemhjGE4KHWrvpCtUxm6lQqFo0RUPMFHKEfSbpYonR+Qy7O5u5NDWPqshsbUkynlqgMRqiYjmkixXqgj7KWMxmCyTCAWIB36K/c+MX47oqrlBQFZm8ZeHTVDwhUOSaX9fxPFxP4Nc1SraNrijIkowm+7G9Eh7ukmnHE84KY9Q/BoQQVGwHy3GxHAcAXQ7iV+q4r/GXCaoNK86XkFDlW2yNlkGWNFTJpMG3kwP1P4N0U6CbLKk3fbbxBKdIOqpkUHEzCCHWFVBdDmIoEQ42/BJhrWXde/AbOn5DZzZTIF0oc3BbJ9Fg7VjA1Hjp1DC5UhXzFgJ6W8IZUALsie3iSmEYSZLpD/ZiyiYAlmszV82hyxoRzc/h1BCKJDPY0YYQ4AiPWHcMXVbJ2iWkDp3phSKW55KxCqiqQteuNrJ2iYprYWo+Sm6V786coT/cTNe2Jnp2tC/1pX/vDQNHa9/KmVKWJVoHW7h8dIjZ8uaF069qxE0/f3T2HRr8QeoUP29fHqVYsWhPRBlPZdFUhZ7GONlSBUmSWCiUiPhNJGA6k8enawxNzSNJEufHHRRFJl0ss6ujiVjg1gNwoVzm1ZFrdEQjNIZCHB4bJ2QYeELQXRdjvlhiNJshGQjSFApxZGKCgK7xWH8fdXo3w/nvUnRmieo1o1bBnqbkzFNndG/qGfwgkK9UGZ5dwFtUbSVJIukb5GLum2StUer07iXXRC3qzFslYBtBV4IkzC2kqkM4XoWAmlzhPrl+zc3CUMJE9U6my6coOrM3tXfjHurNrYjM35KuXiVh9N/yHkxdxXZcXjx1hX19rQAcuzxB1XYwtFuL3m0JZ9Et8vLcqzSZjbiexUtzr/BY46OYisnRhWFGS/P0hproDzVxpTBDQDXoDCY5tjDMHXU9nM+NE9ODnMmMEtH8ZK0SZzLXGC7MUnFtdsc6eXn2LL2hRgpOlYOJAS7lpojqAVr9cYxlK44rPLLVCkXb5rp6EjFMQppBwbbIWhX++bZ9tATCK+6hYFXxEBiKykKlhCrLxAw/qizTGAjx7vY+vnn1PAC6qpAtVelvSiBJEq4QtEVDJEIB5rJFFFmiPhzE9Tw8Aa4nSBfKqIpCdzLG2bEZhACfptEUXdkPAMdxOXNpinDQpKutttdJl8sEdI2+eJwLc/OUHZtctYpPU5kpaFQdl9ZwhPFcDttzyVUr+DQVIaAj9ABX8t/l2PyfsCX6fjzhMpT79gpXihCCgjNF1hqn6uYoOws4coVrhdfQZB8RvX2VWni7CPtMBprqGZqeR1s0ejT6dtPiv5NjqT+h7CwQ1TvxhE3GuoahhOkNP7Zpn66MypbI+3ll5v/kjdn/Tm/4UUwlTMXNsVC9QlfoQeJG760bWoQqmWyNvJ/XZn+TN2Z/i97wuzGVCBU3Q9nN0Bd6DF0JkPQN0h68lxMLf0nVy1Gn9+DhkLVGUSSDgch7V7ijQj6DH31oD3/1wjH+7q2aJyISMPnEoT2E/eu7Hpf6tek7APJ2gZAa4q74ATzh8Z3pZyi7ZUzFpMUfZ6K8QN4u41N02vxxGn0xGswIlufgIai4NqlqDr9qcE/9AN+ePMaF3ARV10GRZYpOhbgR4t76rXxj/B0iup9Wfx07ox0EVXOpH67n8fcjF3h2dAhVlrmYnifh8/O5HXdxd2M7p1PTfGnoFFdzaX7twMPsqW9e+u4zo0NcyaZqK9vCLJqs8Mt3PEB3pG7V/WZLFWRJYmQuzX1bO0mE/ARNg0TIj+24eJ5goLme6Uwe1/NorYugqTKt8QhBU0dVFN68dA3H9RieSbG3e6UqVLUcfv+vX2Gwv5nPfep+FEmiNRxmplBgKJUiaOgkA0ESfj8egrlCkbZohFylSnMoRF8ijiYr1Pl9aIpCQuljf/1PcSb9JV6b+U10OUhX6AFCWjPKsv3rdOkUl/PP4AkXfdFaey7zNLKk0h9+fJVwBrUGEkb/qn3hSkhE9Q48YSNJEpbjMp8v4ni1lUyXA9xZ/9OcTf8tF3PfxPbKSMj4lBhboh9YEktJqvlHa9daW1hrq9gWDiZ/kTOZL3N0/g9rqrykE9XbUaRHgZr6mzAHCKrJFd/XZD/15lYMJbLUXktgP/ckf45zma9yZP4PEAgUSaPJtwcWFS9V8rEv/lnOKk9zOfcMlldcDLOM0B9+fNXkIkkSe3pa6G+pJ10oAxAN+gia+qZW9tsSzogWoeAUeG7meVzhIUsKAbVmBXOFS9KMcCU/zYFEH2HNz1B+ioQRIqiaHF8YZqw0z/54L5PlNMcWhrE8l22RVibLaRrNKM3+OqYqGWRJQpMVZGRMRed0ZpQ76rrxq7XZZqFa4vOXTvK5HQfYW9/CH5w9zHy5yP5kTXW4s6GNznCMn3vlm1RdZ8U9FG2LZ0eH+IU99/GJgd1UXYcm/9pqbzTg447uFhRZpiESQldvzIo9jTcsoL2Lv29tXTkITE1jb3cLpapNWzzCzRCA6wo8T6z4zsGOjqW/t9TXb/hO6rsCy/5S6Q4+RLNvL1Uvjyb58Kl1NculEEtqV3f4XXSGHlizvbUEsDf8bnpCDy8K+NqQkNlT9+laG2j4DY2gceN8SZLwKwn2JX6CQfcj2F4RGQVDCaPLwSUVUUbjjvhnl9pc93qSXIvhNbdQcdO4wqZYhaAWI6yF8IRAlcLc3/C/r4rvjertHGr6DysMZbKk0h64lybfXipuBg8HVTIxlcjSfUuShKHE2FX3GbZFn8LyCnhCwqdEMJQb97AcjusxOpdhfC7Dzu5mNEUmXSgTC/puKaC3JZw+xeSRhocZK4+hSAptvhuW2qgewPY8tkVaMWWNHdF24kYQQ9Y4WL+FqXKaPbEuYkaQDn89BafC7lgXUT1ASzmNKzySRoQHktswFZ1DjTswFI1DjduZr+ZRlt246wlc4eFXdXRFwadoi5E6NWOTKkmYioqyzs13hWM81NqNT914Q+43NAbbbhgvhBA4rocsS+TyNd9lJOSjajnkCxXCIRNDV5d8ZJ7nUWf6CEgqdtXF0lx0TdnwpdS+J3A9D1VRlvy3QggqVYdcoVzzrwVvXGs5JEnGp8ZWuCVAWbEIKWhrhuxdv87cZJrJ4VniTVGSLXVMjczR2tuAZdtMj6Zo6a5nYniWzFye9v4mIokQ09fmsKsO2YUCPdvbKFs2dSE/IXOlgEqoBNQEkFjz+rX3Z9y4Z9shUy4TNHQMVWWhVMZQFYK6QcGyqDouYbOekK5xbOQKMX+BXS0hhmbnGU1n2dfeQtS38jlJkowmrd7/S1LNqBOQfUuPyxXXLboenhDk7ApX8/PsibdhKjFGCgv4fH4EErbnokoyHrX3JyHxd2+f4ztHLpLKFfnZD9xLZ0OML7x0gn/1gXsJmBtb0W9LOCVJIqQF2aZtBWAof5kmXyNBNUhMr/1ch6FodAVvDOyeUOPS5to0NOJGaKnNtsCNF+VTax1OGLU9WljzE9ZW+oPiPj8Pt/Xy/zv+Cn3RBJlqhc9svQN5k0aAkG6grjHL3QrFssVf/O1btLfU8dyr56lUbT711AHeOXWN42fG2Lu9nZ/80Xvx+3SmZrN84etHuHBlmqrlIMsS3e31fOKD++luT6wpoEIICqUqX/rmUfKFKp986k4SsSCeJzh9YYIv//0xxqbSAHS3J/jRD+ynt7P+towftwpGz6YK/P2fvkz39jaOvniOhz58J2/8/Qne9ZEDpOdyXDpxjUqxyuHnTtPa18DJ1y7yoX/xCM998U1CsQDNXUkc20GWJDLFMrbrbqoPa92D5bp8/fQ5DEWlL5mgzu/jyOgEk7kcD/V1880zF2iN1rKHntq1jV0tTWiKjON5nJ+Z4/JciqZwiKjPXNX2ejiVnmSmnCOkmVRcm7Jj0RmMM1ctIISgyR9hspTFFR694SRnM5PEjQCvzlym6tp0hxKkqkXGS2l2Bdt46/w1fv6p+3n26EWgtg+dzxUplKs/GOH0hMflwhVydm7F50OFyzyiP0xQDa7zzRuwPZfjc5O8MnWV6VKemOHjnsYO7mpov+UKdjOUxZXpnqYOnujcQp3hJ2qsn3qzGt9bNJHjuBw7M8rla3N88N27efpbx/idP32Jxw9t573v2sHf/N0RHrlvC4P9zbieQNMUPvz4XurjQaZmsvzF02/zZ19+k1/52fdgGqvvuVCq8id/8wbHz47xLz/9IHWRmso6PDrHb/3RC/R1Jfncp+7Htl2++p0T/Lc/foH/8PPvpT6+OWs01FaAP7lwhLOpmVXHdieauasSZ+TCJL6Qj3y2RCFTonuwlfNHhknP5dl+Vw9XTo8zM5ZCNzXSczlK+TKSLLPv0CDt/U0ATE7MkimVKVXtpT0b1MbBsbkJXpm8yky5QN2ycWDeNA6y5Zob68MHtiNLElfmFxAI5gslMuUKUZ+PR7f28fSJMzieRzxwYxLvTcQxVJXtzSvdNp4QVBwbXVFR5dUTdMmxGM7PYyoqArivoZfzmWlc4XF/Yx+O53IhO01U9xHTfaiSguU5FJwK2yJNXCumyFoVwpoPn6wjyRIN0SDK4rVs1wOxuYi2TQvnycxpGsz6FQHZrlg9K679fcHfX7vAfz7yAvOV0tLnXxg6wc9sv5uf2HZgzQe1fnuwUClzdmGGnFVFlxV2JZp4sLUbWZI4m5rhWj5DqlLi6OwEjucxWNdAzFzfleEJj/MLc4wVssyWi5yenyasm2yN1ZP0L8+EgC09DTx4Vx9jkwt884XTPHLfFmRZ5uvPnmQuVQCgtTHKv/z0A0t+XW+bYHw6wytvD5EvVFYIpyJLFEsWf/blNzl1foJ/9WMPsXtbG7Is4Xke33rxLLqm8NOfuI9EXRAhBNGwj1/5jW9w+MQIjx/avumJyROCt2fGeGH88qpjtufySGc7zV319O/uYMsdXXRubcG2bL7wW98mGPHR3t9EMVtmYSbLtjt72HF3H+G6ILIioSzbk8dDfgaa6jGXuQw8IfjG1fP8n8e+S2rFODjJz+64h3++df+KceDXdYSAd65N0BQJcnpyGpAwNRVZYvF/ac2x49c1ZnIFri1kaI9Flp7PfKnIf37jJX5i1z52JFdbpZt8YdxYMxXXJmmGuJSdoTdcT9Gx8CsaVUnmUNMArvAYL2bI2mVGi2la/FGiho+8E2C2UkAXCrqhUB8O8Dcvn2AylcXUVU5cmaAlESGyCbfahsJ5Xf2QJZl7EnfRaDagLNtc1xsJzDVyOG9G1qrw5xeOrhBMgIJt8fmhk7ynYwudodg6316N8+lZZssFHmjpJqQb5K0qf3DmMCHdYE99M+fTc0wVc7yrrZeyY3NibpLWYISY6WMw3kCd6VulAntCcCkzz9XcAvc0dSBLEifmJqn3+VcIpyRBXbQW7xrwGYQCJgG/ge24KIqM7dyYsFxPkMkWyeYrWLZD1XJwXA9nWUD+dfz1V9/m6OlRfuGz72LHlpalmbVcsTk3NEUk7GNsMs3ETAaATLaMokgMj82v6TgXXhEkDXDBK4G8uaTtRHOMB5+6k8unRgmEfSiKTCARYvf9W4jUBTF8OoMHenEcl9GLUzR31aMoMoN39hCK3li58uUqmqJgLVNr09Uyf3Hx6ArBBMjbVf566ASPtQ/QHooufR7QNd6/YwsT2Ry6ovJAXzfjmSw7mhuoDwZoCofxayqH+nvQ5JVGn7ZYhLu6VgeWW57LhYU58ra16hhAZzBORzAOCCQkBqPNNU2Nmr5lKhp74zV/u0DwIf+eRXtHLRDBp+hMlXJ4CCQFPvmuO/ibl08wkykwly2yvbORTz60G20T0WsbCqdtOVSKVXxBk3qpHrfq4cmLTlmgU+8EB9ILWQJhH0LUwuZcx0WSJArZEtFEiOlSntFCZs1rzJULjOTStyWcL48Pk/QF+ezgPmRJpuI4nJyfYq5cxKdqfLx/17rfvSPZArSs+lyVFZ7sGVz9hVWQbgQsSzX1ZLmKIhZD+8anMnzhG+9w5docuqai6woz8/k191tvHB0mX6yQjIdI1AVX5AlYtkuhWGFsKs1/++MXlo4JAbqmYura8uQchKjgOVfAywAKktqJcK4iqd143hxC6tnw7mRZon93B/27O1Z8fvDx3Uu/66bG/ndtX3F8z/1bV/ydDAdxvTzxZfGjU8XcuuNgtpTnan6BpC+AKivYrouuKsQDNfeVqdZWyZCpY7sePlUl4jMRQtAQDlD1HEzphuFHlWW6EjEc16PiOhiKumpCtlwX23PRZAVNlpe0nMXI2lo7yx64bbs4dm2y0Q0VRZFxHQ/XddENDUmWCGsmj7VsQwC6rCD5JX7mfQcpVy0EteghRZa+f1dKIVNi+tocuYUipXwZ4QnC8SCSVFNh7KpNa28jF49dpW9XB1fPT6DpKq7tEo4HmZ9Mc/fju6k4zooZdDlc4VFy1p7F1kN/rJ4/O3+Uz188SUDTOb8wS9G22JVouq12vmfc4rlWqjZ/+IXXGJtK8zP/7AG62xMYusrT3z7O33/3zKrzY1E/P/WJ+/iLp9/mD7/wGj/344eIhmuDWpElNFXlrj1d/PiP3IN8kwrnM7WVk4M7h2e9A5IPSTLB9SNEGWGfBpEHbeM0pR8UQj6DkG+lVlVxHWxv7XHgCMFYLstLwyO0hMK8MTHKw509lG2b18ZH+UDfFt7Xt4XXx0f51pWL/Pt7HiK06Kp5+uI5rmUz/NKBe1EliZJt8+zVIZ4fuUK6UsZQVd7V0c2PbL0e0yxxanaar146x0yxQL0/wKe372FHfQMCmJrLMjGfJREJ0tVc0zimJzI8/+1TVMoW6YUiH//Mvbiux8vPn6vlbw40cs8DA6iqgq7cyO3NlSr4DZ2gzyCVKzG1kKK9Poq5hqX9Zmy4trqOW4t19elE4iFyC0UkqbZSSBJEEiEaOhI0diSwLQd3cVZp7EwQCPuWKEL8moaprD0PaLJCRN+8NQ3ggZYufmbHXeSsKmP5DH3ROP/hwMN0LFOJrkMIQbH6Do47D4DljFKqHltxjieqZEvfIV18Gs8rrWrjdpEvVrl0dYYDuzu5Y0c7ddEAhqExv1BYc+Xs70pycF8PP/3J+7hweZovfP0IlaqNEAKfT6ezLc7UbBa/T6chEaKxPkxjfZiGRIhwcKUhTJLrUIz7UIx7kbUdyEoLstqOou1G1vYhreFCWA+XF1Jcy2Y2PGemUODp82e5ltn4PKiFRxrrjANdlvGpGm9MjJKulNnX2ML/OPoW2WqVfY3N/M350+SrVdKVMpcWUjjixtZgupjnajYNCDwh+NKF0/zusbfZlkjy2d37eKJngMZAaMm1VnUcnrk6xF3NbXx21z7Kjs1vvfM6ectiai7LV14+TaliUyjfyHy6cG6CaF2Ax96/B59fpy4R5IVvn2L3vk7e/5H9nDp2jZmp7Ip7Kldt/uDbbzO1kCOVK/Ebf/sSv/Hll/j8S8dxvdVbm5ux4cpZ1xAh8tA25EU1LpoI0bG1metLh7So0vXv6UQIQUtvA7IsL83kruuhqDJNaoi+aIK56dW0DN3hOnoja/u81oOhqBxs7uRgc+emzlekENKiX6/qXKVincNv7F06LqFiqB3M5f+AkPkAMhun8tyyf7pKJORjZDzF1GwWXVM5fnaMwydH1vmGBBLsGWzjMx+5mz/64us0JsO8/+GdqIrM4w8N8uu/9yx/8fTbPPbgIH5To1i2GJ9Ms3NrywprrSQHkOSem1qPX78K3BSUsREuzM9hahr1fj+GoqLIMq7wqDouuqKgSBJhs7bnv5pJ0xGN4ng1f6DjeUuq6HU0B8L0RuKr9pwAPZEEPeE4mizzQFsXdT4fX710joc7e1BkmeevXVkM1dwY6UqZr106z0e2bOfHd96xlCwAK9017+/dwpP922qLDfBrr32XdKXM1GyG9oYoD+zpWXF+e2eCr3zhbXKZEjt2tyPLMqWiRUNTlGhdAJ9fp5Arr+hLvlJlNp0nYOq8dvYqkYDJjz92J3/07bfJFCskwgE2wobCqajKCgtcz872dc9bC9dTtkKawU9tO8BUKc9oPoMrPBRJpisc4+d33Ue9r9ZJIWxy5WcRuDjuHEHzPgy1D8u5SqHyKkgKIfNBNKWFqn2RQvWNWvvmQ+hqJ2XrJCXrCBIqYd+7UZUGCpVXKNtniPqfQpFrK7nljpMq/AWKFCbsexRZ9qOpLcjLVhVPVChUXsVyrmKovQTMe5Flmfp4kMAi40LAr1MfDyFLEoosk4yH8Pt0QgGTjzyxlz/78lv8u1//OrquEg6ZPPXYHt45eW0pbU6WJOrrgkRCtcB5WZE5dHCAhUyR51+7wNbeRga6G9gz2MZPf/I+vvTNY7x+5AqKLOEJqI8H6e9e6Sr4QcIVgpevXeXc3Cxd0RiPdvfyzPBlxnNZQrrBk1u2EdT1FVbw49OTvDk+hiLLdC9+57obIaKb/NS2u5gpFRgrZHCFQJVkusN1/Pyue0n4/Kiygl/TUGUZn6oS0HQcz0UIlgLpl+PmT9KVMtlqhV3JxqXr3qw+6opCTyy+NHH4tdr7dDyPurCf45cmuDKRwtRVmusjyIvaoqxINDbHSDZFkRWJlvY4xw8P09Qao1ysUt+wMn5aliQkuebvfev8KB+4e5CmWBhZllYYDtfD7RPVbAJCCHKZElOjKRpaYkTjQe5t7uT3H3iKd2bHmC0XSPqC3NXQTlc4tvSQhHDIlb9DyPcwhtrNQuEvqQ//LKnCXxDyPYQQ1uJn/4pM6WsYWheGOoAsBxBYZEpPEzDuRFPakBZjSX36LvKVF3C9eaAWY+t6Gfz6brLl7yBXA4R9j6zsP4Ji5Q2K1cMEjYNky99EkaME/bv5337qEXxmbRW+/84+7tzVic/UMU3Bv/2XjxFYpGR56O4BBrobmZnPoWkKbU0xQgGTh+/dQiRUG8ymofGLP/kw2rKoIU2t+UYfuW8rgcXgaFVVOHTPFvYMtjExnaFqOQT8BnUhH4onNkxz+r7eIzBY38ADHZ38+cnjDMQTvDE2ymM9fbwwcoXLCyl2N67c52cqFaKmyeO9A/zh8Xe4u7WdqFnbtkiSxAMtXfx+8CnemRljrlIk6Qtyd2MHnaEYU4U8ksSy1VZaNIAtU9vX6Gdp2YqqSDXDjuW66z4XSWJd1113S5xcscKpy5O0N8RoTkTwhOD8mQk6OuuRJHjh26eQ5V088sROjrx1hfHRFE88dQex+Ep/f8Rv0hQL8+tfepGWeITBjgYyxTKeqGWs3Ar/IMLpuR7PfOkwsUSQ6GKHFUmmP5qgP7qxCitLAfz6HShylGz5W9jOOAKLgH4nApt8+Xk8USLke5h8+VkcN0VE+QASGiHfIxQqr6Ip0+hqa81wJUeQpZXqg6kNYGo7sN0pqs7wmv0o22ex3UlK1lFkqabmyrJMXfRGW36fjt93PcpDIh6tEVlZFRvXcWlOhmltioKo0SUKzyMW8dcyXBy3xtEb9FEqlDl/9CqJpiiO5aJoCsITjF5L1bYFHQnGr8wSjPppqQsxN5EmHvRRKVYplC1i9WFGLkyiKDK+oMH8dJZYfRh/0GBqZJ6mzgT5dIlKuUpz/+ZXWlWSaAgE8Gsaiixhe15NrXUdDrZ10BxanWkDNTuCodTIz27eYyuSzEC0noHoxjHD6yGg6RRti3SlTNQwSVfKnJmbIWzUJoCE309LMMQL14bZ3dBEUNNxhcD23HXtHivuWVHYt7WdfVtvaImO4zIzlaG7r4G6RAjtwhSyLBEMmTz06PZ129JUhR97dB9Xpxdoq48SWLQ0f/LQXsL+W9tZbls4M6kC77x8Acd2ueO+fmKJEEdfvcjsZIbuLc0M7Grj2GuXOH14mDvu60dVFUqFKkdeuUC5WGXXXb00tq3vc/NECcsZQVUSSMioSgKQsNxxhKiCpCFLJrraTjz4Y2RKXyNffp664CcxtQEMtYeF4ucpVN8g4nsfAhuBgyfsGk8tYDkTON4clnMNVa5HCA8hLMDFExaKEOhKKyARC3wEIRxUZXNUH5PDs3z3i28gSRI77h1g+8EBjjx3ikvHriLLMg9+5C4i8RDPf+E18gvFGvl1OEAxX2F0aIaW7npOvj7Ene/axrWLU5gBg9GhGTLz+SVBVdQaiZRVtpgZW8D0GRx/9SKyIhMImcQbI0wMz6FqCl1bm6lWbN545hSGT0fxbT4ay6/p+FQNWZKIGiatoTB7G5sZyWQIGTrCcjlTmOHc3CymqtIcqu19T85Mka1W6I7GCGo/WBaGbYkkUdPkP77+IoOJJBP5HJbnLq2oQU3ns7v28f99+1V+/oVv0R6OUrCqtIYifG7Pnd/TNRVF5okn7+D86XHmZ7Lc966t2GrN4OO6HqoiYzvekp87YOpoWm2rF/QZ7OiqaReeEIT9BoMdDZsKNb1t4Ry5NM3UaIoDh7bhD5i18KTWOgxT57XvnKJnsJm+7a00dybYc7CfSF2A1585TTZdJJ6M8PxXj/Lxf/EuVG09ciOFYvUwQlQJ+x9HU1qI+j/AfO4rtXSiwJPIkkmu/G2q9giSJBMyH0LgkC09Q7EyjqEZmOpOHLdAKv9VHC9Fvvws4KLKMXSlmYXC55Ekmah5P447Tab0dVxR4mr6L6kPfpCI7xDp4tOk8n+CIkeIBX4EWbn1bLcwnaFSqnL/U3fS2ttILpXn5aff5u7H93Lt/ARvfvMo7VtaGB+aZv+jO3nxS2/Rf2cvut+gtSeJVbGpb4pSKVoUcmUUTaGuPkwhW6KlO7loOQ/iD5lMXJ0lm6pZgHVDI9EcRVUVmjoSlApVNF1lYTaHYWoEQiahaIBwLAATm3vXd7fW+Hs1WeZ9Hf3onsxTA9uYTGXxmzpXT07QMpDkE9t2oukqId1gLJflzpY27mvrQLYEFy9MMTi42q+8HNcn6pCu86GBQeI+P7qi8OEtg0QNE1cIPjQwSFg3CBkG/9cDj/LK2AgFy+L9fVuJGiZj+eySn/Le1g6SfpvDU9OkygpNgRAHmluRJQu/kudjW3fSEgzjennApSHg42NbtxDWrTVVYUmSSDZGSDbWbBZVy+HZN85jOR5j02lCAZNMrkQkZAISrY1R+juS5EpVvvjScbKlWpLEdSUiFvLxsft3Ew5sPJ5uWzj7d7RSzJU5/OJ57n54kHDMz5GXL+IPGpSKVRAQjPjx+XWi8QCKqjAzkSabKuLYLnX1IcSqbfwNyLKfqP/DZLJ+SgWdqakUPrMXT7TjN3VKJZ1srkQmd5BQ4D4qVYHshsnmcyRiH8Sp5FA9k/OTZdqawKk+iiw9Si5XpVoy0VWdhvp/AbiLibEq4BEPfhoBXMmmSFV1crZH0XkPhuyholEnbS5IYmBfjXHg8DMnmbgyw+BdfdgVB8d2aN/STGNnPRNXanGtVtXmnvftZduBPpAkTH9tlRnc383sRJpkc4ztd/USjvnp3NKEoikYpoayaD2vb64jHAsSqw/x4AfvwPM8DJ+OpqnE6sOoukpuoUAo6uehJ/dRKVsEIpt3pfi02iprWQ4TF2Y5MZOlf3sr50+OsuOOTsrFKsMnxtmys41krOb/bg6FqPcHyE7muTQ0gwRUKzalskVDQ4Sx0RSmT6O9Pc6VK7M0NETYsqWplmljmHxycPfS9f/Z9hts6J/afuPz3lic3thKTWZv442cXUmyaPGd5GNbdiJEBUWO4XrTCE9gyiP82I57kCQZ25nC9aZp8Dfx6e3tWM5FHNfGEyUkNDS1a00NT5ElulsTaKpCczJCNOQj6NeRJYlUtkjouq1AkWmpjxKr1AIQKpbNiSuTaKqMto4RdTluWzhTMzkUVcYwNeans7iuR7FQobGtDn1xkytJLFokaxv6bXs7uXBilKa2OPGGMOp6HZMUfNo2JCnIldEM5YqNLEv4TA1NVQgFZcan5vCZOpOzWdqbYui6SjQEw2Pz6Hoj41NlutsCuK6garlMzFTI5StomkLQD45boqE+ehNBk4Ik+RfpJkzylsVkaYG5crFmUTUDtAY3p9ZOXJ5mfmKBYCRAdj5PtD5M755OCpkSvpCJL2jSv6eTKyevYZUsVF3FH/LhD62cRRvb48TqQwQjtf3ujbISNxCrv+FCMXwr1UfNqL2L6zxLUJs0rdtwpVxHLl0inSqwMF9AN1QCQZP5mRyVkkU+V2b3gRuum62JJEIIXjh1lq6uei6cnyS1UKCpKcqFC5M0NkYoFqoceecqjuNSLlv09zeiKD84g5aEiiLXIYRFxTqJqiRw3RSS4ccTZcCl5uL38LwSrpTD9VK47jQWLo4ziaZ2oqmdrGWCUlWFwd7VAS+FUhXPE4SDtXcZMHXee+fKyKlH9+b4/W+9RdV28K2R/LDiOrd74+G6ANl0kd339NLR14gsSyiKjCRLdG9tRjc1JAkOfXAv/pCBJEls29tJpC5ALl0i0bh6kF2HhEYs8DFApq9Tq1klfQaSBKWyhaFr1IX9mKZGR3Mdfl9NaFVVYVtvI6GASUdLHXXRAJqqYOgqzckIve31OK6LIst4QmwY4BMxTPyaTtIXAElCkSR0WVmRT7oR4o0xitkSDR0Jkm0JhBC87ycOce38ZM0g1BAhEPLx/p96mOlrc4RiAYw19oGarqJtwqL3jwHDVInWBUg0hNE0hUDIJJEME0sE0TSVQq5COLrSN9zYGGFsLEUyGQYJ5ubydHYkCIZMDEMj2RBmcjJDe3v8lhkalldBWcaqfx22ZyGvyY6vYOp7kWQTRQ4jSyFcka/5u1WZ6w4YWfKhqs3Ikg9ZMtDVGrWJrg4sWftvB8GbqEeuG8Nu1hNzpQpV+9aTpHSL/L5/lEo0QggKxSrpbIl4LIDftzkah+twXBdrMfB8JlMgGQ7cktlso75YnstCpcRMucBcuUjeruJ4Look41d16kwfSV+QhBkgoOmrNvdCCAqZIq989Qg7DvbT3F0jjPI8weSVGeJNMcyAjizLeJ4HSDi2g6areG7NNaIZa4d3CSEoOTaz5QJTpTypSomyYyNRM+DUmwGaAiHqzQCGsroNy3X4F698jefXyEp5vH2A/37v+5bCz74feJ6Hs2gkuf739d+FqO3jHMe9Mbkv62fZLVJ2i8S0Wq7q26nnaff30+Rb6We/mDtBnZ6k3mzm+0WtjotH3qqyUC2TqhRJV8uUHBvbc5GR0BWFgKoTM3zUmX5iho+gZqBIa8fKZopl/vL5o2SK5aX7ns0UaI6H+YWn7l8+Rtcc7N/3W5gs5vjzi0cpO98b/WJPOM5Hu3fyzvERTENjeGSO++/uuy3hvDqbZnwhixCCd66Ms6ujicd2D2B7Ll8YOsnl7PzSuRISD7V080Bz94preEIwWy7w+tQI3524wvn0LPPl4mI8qIegtuIqsoy+GHLYHoyyP9nGu1p72RpL1twHi21OXZ1jYSaD5wlOvXaRWH2YasXm6Atn2P3AVuyqQ1NXPaMXJvGEwK46dG1r4eLRq7iOy4HHdhNe5jdzPY+xQoYXxq/w0uQVLmdTZKwKlussZeurkoyhqNSZPrZEkxxq7eGB5m4a/aEVvsMfhPDdjPFClr+8eIyy+72Ng75Igo/17cIVVU6kX2fBmmEgtJs2fx9FJ8dMZYyCk6HeaCGi1ZG25nCEjanUVuy8nSHvZCg5eYJalKTRgu1ZTFVGKDlFTMVPu793qRQj1N553qoynEtxbH6Sk/OTXMkuMFcpUrCr2J6L44klG4ksSaiSjK4ohDSDpC/I1liSg00d3FHfSoM/iMwNQdVVhS1tyVpO6yIiAZNd3c0/ePa9tZCqlPji0Emy1uZKDtyM+5o6+VDXdqqWg+24zKXynLkwyeBA89JMeyuULZvZbIFMscwH9g9y5Mo4UIv4eH5siFemrq44v+I63NvUhbqYtF12bL49epG/uHiM8+lZrHWCs8Vim47nUXJspkp5Ds+O8cXLJ3i0rZ/PbLmD3kgCCWhoT9C5tYVYMszw6VE81yMUC9LUWU9bXxNn3rpELlVgfipNIOxj6529eK7H+OVp4k0xrEotGaBWAKjC3145xReHTjGSX1gSxpvhCA/HsSgWLMYKWV6avMJAtJ7PbNnHe9oH8C+6RcLardP8rqv/m50k5ytFvnD5BDnre2Pif7C5m4/07EBIAlvU2lBlDUmSsIXFZGWEeqOZofwpHmn8KIqkMFw4i18JEtKijJeHOZl5nW3hfZzKvsWh5JOMFC9SdHOUnAKqpNLh71usYuZyJbfAixNXeHlymIuZeXJWZc0IpJufiSVcLM+lYFtMlfKcTE3xleEzdIfreLJ7O091D5IwaymFfkPnkb39G7a5Ef5JbGpkWebufd1Uqg6pdIGmxaK612G5DuezU6StWmxuwgwxGLlB1d9RH+PktSl2djQR9hl01W9sWb2aW6Ds2AQ1nflKid85/TpPXzlN8XtY/QUwXynxhaETHJ2b4N/ufYj7mjvRTY3Gjnr8IR/huiClXJm2/qZa0ZxLk7T2NHLtwiT1LXEiiSD+oIluauw4OIDnegRjtYCGkXyaXz/+Mt8dv7zupLEebM/jzMIM/8fhZzk6N8HP7TxIvS9AxNjYhO8Kj1PpUUxFo8kXRcDinluQt6vU6YElOpkfNHxKgHqjBVP20+EfWOQU0ugP7abJbOfZ8t/gCJuIHieiLzfSCdr8veyI3MVMZZyim8cWVQzZh6ppGLIJyByZG+dLl0/x6tQIs+XCLQVyM7A8lwuZOX7zxMu8NnWVX97zIIN1Dd931Nb/cuF0XI9r4ynmpnNUqw75YoWBnpU3diI9yiszl+gLNyABAdVYSn4FCPsMPnFvrTiO7bokIxvTpkyVcqSrZaqew385+gJ/N3IBV9w6S2AjCOBiZo5fffsZ/q+7HuNgUyd9ezoB2PfwjQJJDe03IqTat6zeK+15cFutPSEYys7zq28/yzuzY9/X5r/k2PzN5ZOkKyX+P/sfoc7Y2J0ihGC8tEC9GWbBKjJSmCOkmZRdi0Yziis8HmrY9g8SMgi1ygIVr4zlVdBkoxaMsmj0kZBACFzh4Alv8X8XkNAkfcnXCVBvtHAm+zadgS10B7bhCsHTV87wt1dO/4MYU2zP49WpEdLV7/Cb9zzOQPT2+J1uxvctnCFNZ3+yldlygZJjU7ItKq6D5dUSWW3PW3RRrA1FlmlKhgloOpGwj5m53FK5+euouA4HEt0cTPbCGtTDM9kC05k8pq7x7MlLPLitm50d6+d2pqtlruXTvDY1wjevrS2YhqIQ1AwCqo4my9ieR8Gukrer2Buk+4wXs/zGiZdpD0VpD9ZqeVqezcnMEI7n0h9qQ5d1QtrGmS8z5QL/+ch3bymYpqIS1mv9RJKoODYF26JoW3jLvukJwXPjQ4R0g95IHEWS1lWPVVkhaYbxKTqu8LCFS9oqElJ9BFSD2UpuxeQIteSG/fVtzFUKlGybomNRvY1xsByNZhvXihc5kn6JvdH7iOkJDNmHLCnEjRq1yIXccdLWHFW3jK6Y+JcV1Y1p9eiyQao6TdktMlEapuQUuCN2P+/pGOBb1y6Qs9dXvyVqmU9+VcevaeiyghBQdm0KVpWiY2/oqz+zMM1/O/kav3HP44RvMx1yOb5v4WwPxfid+z5A1XWouA5V16G8OECuD+ZMtcIXL5/k7MJqUilJAk1TmV8ocH5oimQitEr8QprJXw6/wYn0GLqs0BVM8HDT4NJZ05n80s+ujibOjE1vKJwl2+YPzr3NydTUEukx1FS3nnAdD7b0cGdDG22BCAFNR5VrjG45q8LlbIrvLu5V0tXymu2fWZjhC0Mn+aXd96NKEmezw1wtTlJyKuiyhsBjX922dftnuQ5/fP4dXp8eWXMISEBTIMyjrX3c39xNeyiCX9VrDn/PJVUucjY9y0sTV3hndpz84kB0heDrV8/R6A/dUp3bEm5mupIlYQSpN8PYnktE8zFZznAg0bPKQt0VruN37q+Ng+s/JcemaFvk7SoFu0q6WuYLQyc5n15dD3Q5QmqMPcZjSDJg6/Roe7FKHhXbI5rejufXSDi9JPy92LZDXAlh6PqSwOyOHcT2bA6XXuDueI1g+mj6ZWxxF3sTLexLtvLdiSsrrhnUdNqCEbbFGtgRb6I7XEfSV7PGq5ICCCquw1y5yMnUFM+NDXEyNbWKF/k6Xp4c5oXxK3ywq6ZhXH/e8qKdw0Pc0j13W8JZK0VeC2+6/nJkScKnahsy6AkhODI7vqZwekKQShcYm1xgW38z0YiPmzWB7mCCn+h7AE94qFLNUrb8lEQowAunL9MWj9CVrCNdXFtolq6J4NWpkRWfxU0/n+rfy4d7dtAcCK8T+xhhayzJu9v7eXtmjN888TKnU9OrBKhGaHaej/bupCsUo+xWaTYTjJamyTtFQur6q6YQgjdnRvny5VNrrmyKJPOu1h5+dsdBtsaSa5rxO0Mx9ta38FT3dt6YHuG3T73O2YUZBLX90XpUIcsR1nwEZRNJhrhxYxDVGau3DEIIXMfDlBX8GzjWPeFxeGbslsLpuh4XL80SDvkolubx+3TmUwU62+PMTBcQrsSFS9P4/TrVqsMduzpoa72xQimSiiTLdAcHGSqcRkFhe+QAumxiyBIf7tnBm9OjCARd4ToeaO7m/qZO+qL1xAzfuq4RqE1C+5OtfLhnB9+4eo7fPfMms+XCqvMqrsPXrp7h4dZeQrrBxYVa2cOt8STz5RIXF+a4t7Vzw+ewKeEUQpC3Kzw3eYGLuRk+2L6L7lCCS9kZ+iMNqCjYtou5jm9uI3ieYHK6luU/PpWmajnE64IrhE+SJN6ZHyZvV3myfS+zlRxNvujS8ea6MJ95cB8+vZbce2hwY56cm9HgC/Kr+97FY239qLK84T3UWL9V7mvqJGaY/OLrf8+lZa6a65gs5nhz+hpdoRhbw508M/0Wo6UZdFljb2zLuu0XbIu/vHiMzBrWbwl4d1sfv7b/Eep9K4vqOo7LfKqA36dTLFmoqszsXI4He7tpD0b51bef5cjc+KafiRDwd186TEdPkj0HNi6C5LoeX/7z19i9v4ttu9bO+b0dSLJEMhFC0xRyuTL18SCKLBEMGLQ0xwgFDLb0N+EzNRzXIxRarTrKksyW0B4GQruobYVuCNzdjR18essdbI0lOdDQRr0Z2DTnMVwvGeHjE/17COkGv3b4uTXV5DMLM4zk07QHoxydmUQIQdV1GctnyVbL3EvnhtfZlHDawuNPL7/FcH6OVLXI/kQHHYE6/mr4MD/Wew9NSoQX3rpINOyjvyNJYyKMqmw8yJc6oMhs39LMyPA8mVyJfL7MzRuaU+kxNFnFESUsz+Gd1FV2xFq5XpHTcT2Oj0xwdXaBR3f243replJyoEad8XO77uU97QO3Rc8pSRKDdY38xOCd/Orbz65Sb1wheGt6lI/27iSg+vhAywO4wsOQtUU1aTWEEBybn+DtmbE1jw9E6/ml3fevEkyAufkCr715ibbWOiYnM3R2JCiVa3VFB6L1/Ns7HuJnX/06k8Xcmm2vuj8gWhfAH7i120V4gquXZujsSd7y3M1AkWX6ehqwbAfT0EjWh2hqjAKQrK+lqV3/+zpy5QqHh8eJBXzs7WheRtalkC1VGF3IsL2lZmiM6ia/tPv+VSuk51Up2kME9c0QvdVyQh9vH+DI7DhfGDqxSoPKVitcSM/S5A8xVyqSrpSxXBdNUXiwrevW7W+mE5lqiSv5WX55x7v54vARAHRFxVA0UtUCW5INHDrQz1unRvjrvz9CV0uce/d205KMbuomAaZmsui6iqJcD3tanmBbc/56QjBaTGHI2oqV9ersAlPpPLbrkS6WOTc+S3fD5mJhH23r5/2d225LMK9DliQOtfQyED3GqdT0quND2Zr/bLh0jbgeoTu4cXaG43l869qFpT3ichiKwqe33EFneO10u1jMz97dHfh9BpGwn2R9iHy+spRlsTvexEd6dvDbp97Y0Jjhuh6nT46Qy5RRNYVAyFhqQwhBai7PyOUZZFmmq6+BaN1ifqsEpWKVE4eH8TxBz5YmwpFb1wNZDgFL5S5cz0NTFRKJUM0K77iwLEna9QSSxFLZCl1VEULw4vlh9nQ0I1EbR7brMZXN88K5K3TVF/HIIyFjqC3YbpGqO4VP7UCVoxTtCwjhAoKqM0fFuYaptqLJMcrOKK5XxNTa0JUbFndT1XiqezvfHr24ygbhCI+hbIoP9+zgk9t24wqPBv+tCdivY1PCKUkg1wrNL31WcW0KdhW/apArVHjx8CWCfoMffXwfjutyZWx+08KpKDIDvY1kciX8i0wCy7Ez1sZwYY7JcgZ5QeLJ9r0rjEbXa6R4nuDaXHrDfc9yxAwfH+/bhf82GedvbmN/sm1N4VyolkhVymiSymx1gXojCkiYio4mr370s+UC78yurXr2RRIcalltiLkO09Do6aqtXE2L8cvxuhsDQZYkHm/fwheHTjKzxh7pOoSAUqFKai7HGy9eYN89vXzoU/cghODalVn+8n++RCIZwnUFz33jBJ/+mUPE60MIT/DSM2fo6msgNZvjpWdO85O/8G6Coc1nwaSLZY5dnWBrc5Jnzwzx/j3bOH5tkta6MG9eHq1Nhlt76EhEefrIWVRZZi5f5IldA3QkYrTHo5ydrO1nPSF468ooR69O1PbajstC+SWghISKT+vB1NqwnFkK1VM0h/8ZsmSwUH6BgN7PfOnv8WsDzBe/Tcz/ILOFrxL3P7pmcaXeSJzOUGxNA+FsuYAjPPyaxvPXrjBXqj377YlG7mremAlxU8IZ1f30R5L83sVXmCnnsTyHt+auYsgKPaEEiiNz545OmpM1vhXX9WiMr50lvxYcx+PSlRnqE6GlzJblCKoG72newa5YO1HdR4s/tmJG7qyPMZbKMJ7KkitXec+egU1dd1e8ie11jTVmAuFiexbqYoC1K1wcYaPJOjIyHt5i/KWDJmtL5NqyJLG9rpY8e7MFtOTY5KwKAVPj1fnjnM+NICFxMLGLvtDqF3MxM8/EGmqnBNzb1IkkC7JWCU1WsDyXqmsT1f0Yyq0nF0mSaA9F2Rlv4rnxoXXPU1WZA/cPIIRgfvYGz64Qgue/eZLO3iQf+/H78DzBn/2PF3j5mTM8+cm78TzB4O52Pvrpe8llS/zGv/8KQ+embrlfXQ5dVbiWyqDIMtdSGc5PzjJfKHJqfJoP7Rukajt848R5Pnv/fk6PT/Pe3Vs4tK2HoLE6IKJQqfL82ct84u7dXJ5JcfTaBIpkois1JgjXy5OvHEeWTRwvi4SGqbYhSwaeqCCES9jYQ8m+iOsV0ZR6QsbuNTWBgKbTEYpxfH5ydT/sKq4QXEjNMZbPcrC5HUmSqPdvTO4FmxROVZL5VM8BvjV+lrw9zEQpy9ZIIz/ed5CwZjKZyXJtcoGWZG3GVhR5GX3HrSFJYDsuk9MZJAlam2IrLLYjxRRfGjmMqWiUHIu76nt4sGFg6UHJskRXso6Qz0BColixSIQ2vvnrA96vahSdIm/Mv4rlWbT529kW3s6JzDFmq9PossED9YcYLl5mpHgVgWB7eAet/huGj6QviKGoq+KLa75Ri92JDroCLVzPI9DWKLMnhODUOqZ5U9XYmWjkYm6KJl+UjFVkupwlqJm0+eP0hDa31zMVlTuSLbwwfnmFD3QzcGyXqfE0D71nB5pWUyE7e5OcPnoN4YlaNfGOBIoqE4n6icQCzM9kb93w8v5pGj5d4/JMiq3NSU6NT9NTX8fYQpbWWATH9ShULaqOQ9A06KmPryCtXo6y7SAh0RwNY7suZydnUeQQirw4LoSM7aSRhI4mr9wCKVIIQ21ktvh1FCmMriTQlOi6/VYkqZbFtAaucxkZqkrMMGkKhlAkeVP1gTYlnK6orRof7tzDUx27EUKgyspS7KXP0Dh3ZZpUtojP0Ni/vZ1oaPP0krIs0dYc48rIHKqqrHKljBTmuLu+l4PJXqbLWb4zcZr7G/qXDEKj8xmefvsMHYnoEq9uxy1C+Pyqzo54zaF9tXgFXdZ5IHlokVpfptXfRkgLczj1BiW3RMHOYyomd8fvRQYsN4MqB5AlDb+qo0kyNys1tQwXh4JT4q3UGQpOCUmSuLNukDb/Si4fx/O4kFnbxRA3/AxE6vHrKiHVJGeX6QgmSBghdPnWSbvXIUkSW6LJ2kRymwHqNQoUg1y2tLSa5jJlAiFzyY9XXNzj2o5LtWLj24QxaTlUWSLm9zEyn+bu3nb+7LVjvHt7H1dmU5wZn6HiOMT8Pnzaos1h2TiplQqsULZssqUKPq1Gtn1ucpZrqQwVyybufxfyMmOcwEEIB1nSsb00JWsICRVZ0oj7343r5VHkABIqCf9jG/RcWpeP9/r+3lRUTs/PcDWbRpFl7mxs5V0dG3sVNiWcaavEb555nhZ/lPsbeukNJ9HkGxkYQb/BI3cP4HoesiTfdrqW63pMz+XYvrWFkdF5BnoaViTfRnU/L0ydp+RUGSstkLFKvDh9nhZ/jK2RZiqWw9aWJA/v6K1RGG7CCBEzfDT5w0sqrSprqJKKQJC2Fngr9To9wb6liUmSZKJaFF3WqTizzJReJmpsJ2Js3ZBe3xOCM9lhbM+h5FaIaiHy9mre1pJjM1nMr9lGoz9Egz9EcDFgfW9d5yae6tpoCoQI6Trl8u0Jp6LI3P3gFp756jESyTCu43Hynat86FP3IEk1wX/71Ys0NEeYmcxSLFTpGbh9Bv5tLUkMTaUnGefe/g7a66J8eP8OXrs0gixJfHj/dkxNZXd70xJTA8BoKsOZ8RlMTeWVS1d5eFsvT94xyOHhMRJBPwd62lBk/7pjw3HygEQi8B5ARpYUZOVGtXNF2nixWR23thKdkRif3bGP2VKB3lgCQ/kBMSFEdT8/2r2fV2aG+N0LL2OqGnclurgr2UV7oI5CqQoS9LTVMzadJl+qrlnibj0oikxDfZiLl6fpbFudfJswQvSHG7GFS9IM0+SLrqD29xsar10c4eLkHLqqMNjWwEO38HXGTN8S03xnoJvX5l7iuZlv02S20OZvByRS1XnCWhhFkjFkY2mfqUg+FMmPKm/O8qbKMt3BFi4XPHRZo+CuFs6cXSFdXZttvsEfxKdo2J7LXDWHT9GJav4lC2rOLhNQDdRNrKIR3SSsm8yWVxN8bwRJkth3Tx+uWOCVV75NUO/hiY/1sn1vjWdo9/5u6hIBjr89jFV1+NRPP0iyOXDbtJ0DjfUMNNaY+T52oFbzJuQz+NG7d6847z07V9oV+hsT9DeuZHbc1pxkW/PmVH5DbcBQVxZKhppLzFssnOuKWtFm1/Nwrv8uallKt8rKmijk+NtLZ0hXyjzZt425con39azv74ZNCqcmK+yMtbAj1kzaKnMuM8ULUxf4xtgpfmHrwxRGqoxNpenryDI+k+GuXZ2baRaoWQerVYeO1joak2Emp7N4nlixcjb4wiTNEFcL88R0P3fX9xDWbpjpm2JhfuGJ+wCBEDXf6a0Q1U30xdkrrIZ5pPExKm4FQzbQZYN3Nz6OJ1wUSUWXdfpDA1zXo1xRxvZy2N7mfIaD4R4Eta3AhdwIPYHWVecUbZvSOlkx9b6ak7zq2rw1N0TOqfCprntRkPAQHF24yh11XUT0W28lTEUltEHKmOBGMaZq2aIuscziq1Tp2jtM/76tRIz9VNxhFBVsN8VDH6jDUJPsuTeIEB6aGma+9C0i5n4Mpfm2BHTNfglBqlxitlikLx5HkxVKtk3Zton7vz+G/lrMr6Bo28yUC0wWs4wVskwVc8xXSmSsMkXbxlrM7bWFi+N5SzHDjueSrW4snFOFPIOJBsbzWQSQWSf0czluK3yv6rrMlnMM5WaZqxSIGwFCuoERUSmUqvh9Onu3ti4ZhjYDz/OYnc8zei2Fpirk8hUGb8rWOJUe50hqhF2xNiZKaf5+4hQf6zwAoiaMmiJTF6yZ7ItVm6PD47dcOQOavhTbKEkSulQTyuvwKStdALKkL/vdxFASGEr8lpWiAUxFZ6gwRqqaYSDUQURbveKWFwPF10JUr/XFrxrsjnXy3ZmzwGKJ+EqONn8cn1pjRk9ZBeqNWuzsglUgboRIV4tk7RKNviiarGxojDh7fJSr56epVm2uXZnjocd3LjsqIyHj01pRFI1S5QpBfRuZyhuYaiuCCJnq6/jUDjQCWO7CIona5uF6HhXHwVTVpfxZTwh8qoonBM8OX6Y5FCZiyBSs6pLBxV7Ms1XkGq3MdaExF3NY14IQgnS1zOnUNK9Pj3B8fpKxQi16x/LcdRMDvhd0hKO8On6VS+kU8+UiT3Tf2qOwKeGsuDbPTpznpelLZO0yW8INfLrnLrZEGglpBk7IIxw0KVdsHNfDst0bZfJuAUWRaWuJ0RQPEwqapNLFVQ+z6trsirVxINHNdCXLS9MXsT0Hz4M3L47SEAny0tkrKLJMxXbQNeWWwlnbM2+qi6sgSzWjQbpymrh5a/X9bHaY8/kR2vwNHE1fwBMe2yIrXQxVz103GH09P6wA0laRr469w2d7HyKs+fm78WO8r2UvOafM4fkr3JXo5ZmpUySMEEWnyofa7lxVy3I5YvEgc2GDkOTjJ3/x3XT2JG9YxSUDXU1iqEmEcHFFBcfLIks+/FoPihzAr3VTsoYJ6FvQ5MiGAQ83w/Y8nr86TLpSpt5fyzs9OTOFLEk80NFFT6yOoK4DtVosb0+MoysKjcEQXzhzqubGEx6P9fTx4shVKo7D7sYmdjc0rli5hRDMlot8e/Qi3xg5x4X07Lpayw8KDYEg/2xwLxOFLHWmn7bQrRewTQlnwa5yKTfDI81b2RtvI2EElxz/AAvZEq8eu0J7YwxVUW4rMghq5F1DF2fo7U4yMpoiFvGz3BTnV3X+evgt3pofJlUtIEsSv3fpRQYjLezr7GIqk2dnRxM9DXEKlSqnRlcHBNwMWZJWbeKvqzd5u4pA1FTnxWMrKU1sJEkmovcvrqgblzC0PJtt4S76Q+0YskbBKVN1bXT5Riyys5hStRbWEyZZkugNNRLW/HhCYMgq7YE4Z7JjZKwSA+EmzmUnSFULJIwQVwuzZO3yUrWttdDcFqStvR1ZCiJwQLJxPQsJFYFLSN+CLGm4Xp6QvhVPFAgb22vPSbjoSj2GrxlNjhD13b3hc7kZ6UqZmUKBHxncwV+fPknc76e/LkHI0BnJpOmJ3TDQqLJMX12c8/NzgKBgWXxo6zaeHb7MsalJLi3M0xmJcWZ2hl0NjUvv0fZcXp+6xu+eeYMT85Mbpv9tBtfbvdUUdDmdIlutsL9p9ZZmPWxKOOuMAP962yEcz2WuWmC0uECdESCsmUtGCb+p09kcR1VljNtkjSsUqwyPzlOpOjQ1RFYZhPrDjfzc1keWHsD1owHVIGb4CfoM+hpr9PeO61F3G26c5ah6Dl+5doLvjJ+l0R/m13a/l+MLY3hCcG/yRtUpV1QQwsNycwS0DmD9iBsAQ9F5fuYw7yycZcHKEVYDjJSmeE/j3YS0wLJ7Wkf9Wrxz1/OoeDXCqarrYCoaVdfG9hwqro2HYGe0g78eeR1DVnmkaQdl1yJuBOkNNdAXaiSm+zfc/3mijOsOg3DwRAmQ0dRWHHcWVU5QtS8iyWbN5QS4TgbkOsr2RUx9J379RnkCU7094i1TqVnLLy+kkCQwVXWxqJFS0xIqZTKVCqlSGZ+qkSqXSFfKFCwLQ1UwFBVVlgnqBjHTR1csRmsovPRUq67Dly6f4rdPv87cLQxisiQRUHUihklUN4kZPkK6iV/V8CkaPlXFp2qLZQ01nh27xBvT19Ztr+hYzJRqzAubDbLfXPgeMFpc4I8uvc5YMY0QgpBm8vHu/dzb0IPf1FEVmRMXx9E1lbt3dd2WgCYTIe6/q498oUIwuNpYUXFtvjt9gflqHgmJbdFmHmkaXLpJx/UoVS00VeHCxBx9TbdXUvA63pm/xmszl3lP6yCvz16pcca4Li9OX+Su+i40SUEID0XyYapJVPnWUR4AW8OdJI0YjnAJqf5FOkcFn3JjH7uRmn3dKj1VSXM0NUzFs3lj/hIH4j28NncJD8Gb80P4VZ2kGaI7WE9UDxBSTfbWdZGzK5xMj9IeiCMtxiivB4GN5+VRlSSKlMT1FhDCQpUTaGoHrpdGkgxcL42yVJVNQlUaUOS6ddvdDEKGwcNdPYxkMry7pw9VklFlGUWWiPt8pEoltibqyVTLxB0fnhA0h8LkqlUOtnXg1zT2N7eSDASo8/mYKxWX3C2u8Pj61XP8xomX1+U5UiSZ1mCYffWt7Eu20h9JkPQHCWnGkuArkryKW8n1PK7mFjYUzohu8oXxU1zOpNBlhV3JJg62dKx7PtzGnvPPL7/FQKSBnxq4D1WSOZ+d5m+uHqErFCcsai+pYjlk8+VFusfNw3U9zg9Nk0oXSMZDNDVEVqhep9LjyJKEJiv0hxspOisf7tXZBSYWcpQtm7FUlomFLE/euX6BmfVwMTvDwWQPd9V38fb8CABhzaToWEsD2hVlctYFBB6eqIC28QMGmK2k+c70m8iSxP7YNgxFpz+0MrXKUFTUdZJvC3ZNbW7x1fGjXQcBFgO74f7kFh5tqtGgXGd++0DrvqUB5Fd0Hm/ehYdARqLs2utWGQdqQmg+tHgFiYpl1569qmLZLrK8D0NTKZarWK4g5NeRJQXLtsmXbAzNwm9qeEJQtZyaC8L1CPnNW9LASkBnNEZndP0Aki2JGwWQ7mvvXHW8r64W7RNJ3shKEkJwIT3H75x+Y13BbA1E+HjfLp7o2EJLIHLL1MHbRdIf4J9vv2MxHxrqzFtrd5sSzqxdIWuVeV/bTuJGbbVo9IV5c3aY8WKaO+NRHtzfh+cJjpwdpVC2SGwcoLMCsizT3ZEgHDKpWs6qdygj0RGoPfSQajJaXFhU9W6ceW0ujet5PLa7n7eGRjd/8WWI6X6uFVNLRFqOcDmfnSJhBJcsu4rkI6JvIWcNoW+yuNHV4gS7o/1cLU5gC4dCtQyhlecENA1T1dbMC1yo1MzukiQtRUUBnM/OcDw1zqGmPvJ2lbxdockfYaKUwVQ0tkQalrI4rn/Pdt1bRgdJy6JoTl2Zoli2eHBvL88duUh7MoamKbx8/Aog6Gut54E9vbx2+hrXphfIFSt89NBuDE3l888fIxr0oWsK77tnENPYnDZVcvKkrFlafd23JSBVt8J0ZZQ2fy/ysonO9jw+P3SC8XWSzLfFkvza/kfYW9/yPWUnbQaGojJfLjKaz3J/a+emOKs21RNjMYNivJjG8VxczyNVLZK2yoQ0k1LZ4q2TV3n12BUm57K3pdIKaitnW3OMHVtbaG6IrjpnR6yVnlCSnbFW3pq/wo5oy4oMma5kHd0NdRza3kvEZ7K743sjGT6Y7GailOW/n3uBC5lp/tPJb/P81EWeaNu+tJJLkowjSpSdKbLV89jurX2dDWacs7lhhguTnMkO0+RbrXYHNWPdilzXMxtuhqGoNVJjWebkwgRl1+aV6ctMl/Jczs2TtlYHNZRdh/xt0Jh2NtVx/toMswt5rkykaK6P8Ozhi/hNjeZEhFdODlOxbHpa4gx2NVK1HcZmM7ieRyZf5om7t/HU/TtvWXpgObJ2mvO5ozjCxhXukt/VFS6OZ68Ixnc8G1c4NYpTt8DZ7GE84S4j/oKxQoaXJ4bXNNokzAC/csch9idbvy/BtG/BjHgpneLY7BRXMinG8lleGR+5ZZubkqKIbvJQ0wD/7ewLtAfr0CSFsVKa7dFmekNJFCHR0hBlcjbLXTs7aagL3brRRVSrDi+8egG/piEEFEtVutoTLH9OEd1HsVRFkWQeaRpc8vtdhycEzbEwjusxmyss+TxvFw2+MP9m+yO8NnOFsUiaqO7jYLKHntDNLGoytpcnpPciSwawsXGhJ9iKLmvMVFI0+RK0+lZHrYQ0g4QZYDi3sOrYdClP0baI3sSaF9HMmpHIddBkhYxVJqgZzFeKhDQTfY20tKxVuS2O4UQkQDzs57l3LlEfDRINmjiuR9Bn0FAX4qkHdlIoVXn6pVMc3NGFpip4Xk0MQn6DoN9YKtqzGZ/wdaSqU7w2/y1kZPbXHcLyqpzMvI4tLFp9PfSFdjJSvMDV4nkUSWVH5K4lhr6im+d89ggD4T1EtQRnF2bWpBIBeLStj/3Jtu9LhfUQ5G7xTMuOTYM/WIvoqlY3VehhU8IpSzJPtG6nKxjn5MI4tvB4qGmAvfE2TEUlmy9zfniGRCzAqUsTJGJB6mObDG1TZe7a100s4MO2XbK58ipr7dnMBN8YO0HSDCNJ0BVM0OKvW7rBfLnK8ZFJbMdlfCHL9rZG3r9vfQKt9SBJEkkzxPvbdy7tMSVqxgSJ5XsQF0UyMJQ6FPnWwd2T5TmCqo/O+Pr7YFNRaQ9GOTy7mgVhppxntlxcJZwxw8/9jT2Yqsah5n6qrk26WiagZthd17omefRkMUf+NoifZUliT38r/+Pp1/jck/egyDL37ermyIUxXM+joS5EcyKM53nMpPOLPu7aszJ09TarjdyAJhvsiz3I8cxrjBQvkLbniBuNtPl7eHXu74lodZzNvsPBxGNk7BQnM2+wJ3YvVa/M26nn6QkOEtFq246h7PyanL+movJgSzeqLJMplQn7zDVLa2TLFUKmsVTG/maUHXvDHFmA/roEx2cnOZeapeLYfGzrzg3Ph9uIENJkhV11reyqu+GnydsVKq5DsWIT8hvs3dpW23OWqpsXTkUmHDQ5dnKUmbkcPR31NDdGWL6fTFULPNS4hXvqe9ec4RoiQT56d+1mx1PZ73nPWbCrfH30JCcWxnHEjZc5EG7gx/sPoi3uxSRqKppgc4av0dI0dXqEuB5Zd4ZWZJkd8UaeHl7NqZquVjifnqUvEl/xfVmSiJs3LMYhzSCoGTT7a6yBN0MIwdmFGSq3WWnMb+p0NsXoaqpdf99AK+3JKPlylUQkQDTo49PvuZNCucrBHV34DA1dVfjQAzs3FUq55jXVEAE1REiNUPZKlN0i7f5+gmoERVIpuvnFe64ZNypucTHOOI0jHIxFd48rxLpuk6Cm0x6M4rguLw9dZX9HK/WhAPlKrU5nxGeQKpZ56+ooj2ztxa+vv+0YL6ydHleybRYqZeKmn49v2cmH+rfjU9XNVdnezINaD9+ZOEd7IMaeWDt+n863Xz1HYyJMc/3mw/egppaWyxa5fIWFTHFpcBbsCi/PXGS0uMBoMcVocQFNVmjxxziQ6F6a5abSeb5z8iKO65GvVNnVfvvZEACvzV7mmclzPNG6naB6Y9VJmMEVM6osKTWD1Ca1tKgW4uW5Y1wrTiFLMoPhLlr8q1Xb7fEGQpqxyihkey6vTA7zWHv/uqlJ12EqGutFzJUc+7YJqi+MzvLiscsc3NGNuWhLkGWZhroQMc+3FEiRjAVJ3jQhJ2Ob394shwSkrVnOZt9hojzMntj9BJUwQ/mTpKrTyMg0mm2MFYc4lXmTkpun2deJIinUmy3sid7HicxrBNQwQbVu3f3g9VBGAaSKJY6OTdIei2A5Lhdn5ulvSDCTyzOTK7CeRl5jlpxgvrL2BDCcTfPW5Cjt4SgSEnsbavaQm3l/18KGbzprlUlV199PXc7NkjRDVC2brpY4B/d0Mz2fo1K1b8sopMgSA72NJOqCdLTFlwRBkWTqzTAR3c9gtGXJPRDV/SturC7o49D2XiqWjU/XaI5tnoVhOcaKaR5p3spHO+/YcA+iyVFag+9FXiNpei00mHXcHb/B+u5T1yYf6w7H6YnE18yof3NmlOHcAltj3xuJ1nUG+dNr0KmsBcdzsYVLYzzEB+4fJBb2Lar313lXPcZKc3QEGpAE2MKt+YGvF/1BQrmNXNPlCCp1bJMfxipVaBN3EHQbiGiNeFWdbD7LjvCDVPIKvco9TBRG8EthGv2dWCWZLvkO5FKEXvUAl4Zn2dIaXLdwkyu8JcE1VJWYz2Q6VyBbrlC0LBaKJcI+k0x5/f1kxqrwjZFz60Ya1Zk+TsxN8ebkGBLw5mRNq9uTbPr+qDG/M3GWL109il9de181W8lxd7KbdK7M8Pg87U0xxmcyJGJBIrfBHeO6gsnpDNu3tqAvK0fvU3XuTNyapczxPF6/OMJMpoCmKjy2q/97CkToDiY4mR6n5FiLweHLScZuOJ4FDq4oIUmBTQV2R7UQObtI2soRN6JEtLWDFyK6yUMtPZxMTa0KFJgu5vni0En+3R0P3XL1XAuW5/Lly6dY2EQ2BMCJzDB5u0RvqJkL9hjWnEOTWUdI87FgFegIJBnKT9Lsi3Nk4RJZu0RI9S1tB7qDTXQHG2+7nwCVAoxeFMQiMWRJ4szMNK7rYdkyhaIfL16iIaEyMZ3DcaNk8xUmlQmKpSqxiJ8ZYxafqVEoqnjNgvg6PsWibTFVzNMVqmNPWzNl22agoZ6JbA4ZiaZoiLGFLLtamtDXKPjseh5fv3qWI+vwPgE0BUP85M47OTYzgScE2xO1ZxK9Rb0auIVwVl2Hj3Xv576G3jWP/9WVt5GQiIX9TKfyfOX5k1Qsmy1dDWuevx5kWaJYtnjxtQu0NMbYOdh6W0HpI3NpDFXlxw/tZ3Q+w5Hh8e9JOIOawWszVxjKzdEeiC35NjuDcZ7q2L1EaWl7BaaLzxMxBokatzY8XciPcDx9kQYzzpnsMHfFtzMQXh28IEsSj7b18TeXT67iEvIQfO3qWe5saOM97QO3xbPqCo8Xxi/zzWsXNv2dqmsxEG7FkDVGi3PUmxECqsm57ChRPUidFqTGgm4xXJgmpgdxhcdEOYUE9Ic2ZhrcCJbj4nm1ZApNVaivC5LOloiEfGTzZUJBk1jYj2moeJ6gXLFQFJmqVaPS9JnaUioiArpCsSXW/uUoOTZvTF/jnsYOtjXd0Ejql1Hc1AfXnkhdz+PFiSv8zzNvbVhgyhOCqGFyqL0HIQThTQjldWwonHvj7YQ1k0bf2mpif7ih5p/z67z3/u2kcyUiIZNw4PbqQ8iyxH0HepfiDjc77hzX49z4DCNzaa7NpTkxMsl4KkvId3v0GEv9kGQebl6dAGuuSaAlbTodquiU2RMbYCDUwensZbJOcakY783qc08kzhMdW/mjc4dX8fxkrQq/fuwlDFnhgZbuDbNLrsNyXb47cZn/euyl23KhNPniBFUffsVgX10fVc+m1Z8g75RoMutI2wU8BAtWnrsTW0lZOdr9SaJ6bTDH9M1TQN6MaMjHnm1tJGJBgotUJ8l4bf/aWH9jLLaY0U21t9VNEtN9zN20LxTAN6+d57H2fnbGmzbtTik7Nt+6doH/dvI1pm9hpR3Jpnlnepz2UK2vPzDh3B7b2Jn/7patyIsDLOjXCfp1PGoVgoVXK93gCA/Ldal6a1sILc8lU63551RZRkgghIcsbtCNrPfQBDUztyxJ9DTGyZUqBEyd7uT3FuO5P9HBvng7ludScW00WcFUVucDesJCsPlyfIas853pNzmcOstcNU1ECzFanObdjXetKmikSjIf79u1WDdyblVbo4UM//tb3+ajvbt4smuQtmB0RdHe63U4SrbN1fwCXx0+y9eunl2ibZSoBS/cymK7XCXdFbuR3rav7ka9ySea71z1vRZfzX3hUauDeb2EhyM8qq5LdZ3QQctzSVfLBD0PVZNpalpkcvS8W46DW6EzFGN3fTPPja1mHRwrZPk/Dj/HL+99kL2J5jWrgUPtuRYdiwvpOb4wdIJnxi4thVVCLa2v4jirJlRXCN6eGufSQgpZguliTZhbQiG6IhuP0+/JWusKj3MLM0yV8ksVrQp2leJidana7xZFx6Lk2FRch2v59JptnUnN8NmXnl6K9A+oOgFt8Udd+X9Q0wnrBjvjTYQWhfm+Lbfek97Ofb0zd42nrx1nrprHr+gcahrgidYdKxKUVclHLZBuc4NlS7iT9sDK/ZeMhH8Nw5AkSXSEYvzM9rv5lbefWZNger5S4n+efYuvXz3LzngTfZEEcbPGj1N0LKaLeS5m5hjKzpOqlFYMl+11jdzd2M4fnz/yfZc9dD2Ps+kZpot5Cs7ycWAt/X3997JjU3HWHwcn56dWjoM13n9A0wku/h7WTXbFmwjpt9aSfKrGR3p28Ob0tRUCtXTt1BQ/++rXubepk3saO+gIxpZcUVXXIVUpcjEzz5G5cc6kplft21sDET67bT+/d+atVf7OrkiMfY3NHJ6aAARj+QxQC9fkFk6N70k4LdflN0+8ypvT1/AWZ+rNlne7GUXHWrPA0XJcD+iWpZov748f+jA7ozlkpRmBD5b8jYJaRKKEJN1+cdeL2Rn+eOh1DjUN0BGMk7FKfHPsNKqs8IG2nTdyL0UJQ0mgKptzFZiKjqlsvj+yJPFoez/DuQV+/+xblNdY5TwhmCjmmCjm+DYXl1aXjd5DSyDML+99EENR+cLQyTUF/3ZQcR1+4/grvD0z+o8zDiRpcSxAWDf5k4c+wq7Erd1mkiRxb1MX7+/cxhcvn1wzKydVKfH1q+f45sgFTEVdorBxPI/qIl/VWvfV6A/yK3cc4t6mTp4dG1olnJosc29LJ1tiSTwEHeHoLft7Hd+TcApqvrfbrbT8vcJDLBZMrVGleEIg3HFcd4yaU08gvAySpCCEi6TUo+j33fZ1jsxf40B9Fz/StW+pVFtM9/PN8dM83jK4ZJJXpQCaHF6T/fsHBVNR+ey2mtr4R+cPrznjL8etSvq1B6P86r5D3N3YznQpT53h+76FE/6Rx8HiBICorWi3U5Xar2r87I57mK8UeWH88roUJK7wKDoWxU3EaXSGYvzbvQ/xcGsvsiSxK960btpYW/jGMllxbKaKBboiG2eH/C+vbP29QlJaAAnhzYIQyEo9oICXQ5Lj3NrFu0abkoQnPGq72euJ1TXf3vLmVDlInbnne+779RhT6Wby1ZsQUDV+evAumgNhfu/sW4zk0rdF+wG1mXt/so2f23kvdyRbUCSZmO6jKRDm2iZKAf6/CY3+EP9h/yMkzABfv3qOorPxhLcedFnhnsYO/vXOg+xKNC9pLXvrW9AX2fivo+w4TBfzJHwBxnJZBIJUpcxINv0PI5wStZkovAl9/weNgKqjyDKyet2q2resV3A9bEda/Aks7lNvxlq8PPvi7fzm2efRZZWOYB0LVonnJs7z/vadS6F7sNowoUgyYd1YJTi6rKxpUU1Z45TdLK2+QSQJPOFyrXiKRrMHn3rDGilJEqaq8qGe7exONPHly6d5dnyIiUJ2zSyV5TAVlZ5InCe7Bnl/1zbqzRuVyUxVYyBaz7n0SjXSr+qsNVkIIbCqDq7rYS6rZSNRm0BuNQ6EALvqoBvq5udMAbbloKgKsrL6S0HVQJFvbwKWJIlGf4hfueMQ9zR28MXLJzmVmlo3x3PVNTWdgWg9T3Zv5/H2AWLGykJNfZEEHaEYM+Ub/MOeJ7iaTTNdLPDsyGWS/gB5q7qpvbJ0i0yBNQ96QjBayFC8haq1FmzLYW6R/nIjxOJBAot1F13Pw3Y9VFlCV1U6QtHFgbQxPCEYK2RWqoRCkM+VkSuC/oYkkWhgqcy9KzyOpUb52rWTzFULBFWDR5q3cqhpYE3HfzFfIZ0qUHZsruXThGJ+guEbwRcSEq3B8IrS47ZXIVWdwFB8RLWmRVJrm+en/5B9de+nTm+h4uZxhIUq6TjCIqDGkCUFx/OYKGY5Pj/J8blJhnMLpKslKq6zOGHqxE0//dF69idb2RFvJLFG7ckawVWB+crKlLKwbtASiKw6PzWT41tfeptkc4z737MDn99YfL4eo/nsLVegSsni2a8c4bEP70ffBCEa1MoKPv/1Y+w60E2yefUKo0gS7aHYLYtQ1RgKC2uq8GXHJmdVOTk/xtn5c8xUBHmriu0WkOUghlwlrPto9MfZEg2ys76PgWiSiGohKXUgKiCqIEfAy+MIhaG8xUh+AVcIwppBR6iO9mCUyUIOWZJoDobJVMqcnJvmgRtlANecZb6nlVOWJDpDt5FNvQzXrszye//xBbLp9cMCJUniJ3/pMXY/OshbV8d47vxlJjM5Htnax4/s20GxavHS8DC7WpuI+dePRJIXLZ/XIYRg6NwkX/mt7zI5ukB9U4Qf/7lHGdzTUUtklmT2xTvYGWvF8hwMWV3BbH8zjrwxxB/85ndwHRdZlvjk5w5x4MP7N7x/yytzufA2ITVObNFVJSHT4hvAVAIIPI5nvo3llrG8ci0dKvoISbMTVZbpCMVoD0Z5f+c2qq5DxXVwPQ8k0CQFU13sM6tXeCEExYqF39BpWGSRvxUKuTKHX75AqVCld1szuqExN5VhZiJNNB6kvSOOEDA5miKbKtDcHidWH2JuKsP8dI6mjjh6JMKxso4+5RCMaLR21eN5HmNX5nAcl/aeJIZZa3duOktTW5xoXYDzIkBvII7Pqq3WsfrQbbtTHOHxP869xjPjqwMwEmaA373nwxyIGTjdnVhSB9XKMTyaQZSRRRpD1TEUB0X2QLqMpOoI6x1QuhHuMJLSDXIIYR1BRkKV7uN/nHuFVLXEBzt2sL++jclCbjEfVTCWyyCAwcStwzD/0fecruuRTRfJLGwknFAu23z95Hl+/9XDNISCzOSLXFvILBqjPP7k9aN8dN8O3rtjY9bs5RBC8MI3T3DqyAgA87M5nvnqMbbsaEPVlKWV81vjZ4nqPn5q4D6GcrPIkkxPKLFqYFgVh0yqgOt6SJJEZRMlDgJqjGbfFjLW1I37RWZb5EEk5EX6E5eOwC7GS2eJ6c0UnTQsq4JcKwpbcxFspiDOUn9tl6+/fIb33jtIJLg5Z7jnCRzHRVHkRW4UmJ3KkE0XeeP5szz1mfuYmUhz7I0huvqbCEX92LbLc189Slt3kuNvXubRD+1jYT7P7FSGo68PcfCRQeams0yNpvD5DUYuTrPzQDfPf/UYHf0NnHjrCk/8yAEApq6lGBue5b7HbqRY5aoVhlIL1Af82K6LT9OZKRTQlFoU0EA8scQdJAENvhAtgTAlp0bcnaoWqbrOEpUmIoemJtGVCH6hISmNeM5lJKlGJSO8aZDbwJtD2GcRogIiA+hI+g6EM4wQBSSlBVcIUtUSqWqRglMlVSnxtaHz5KwK4/kc7aEI8+USD7V3894fBOP7DxLRugBPfPROZqcyFPMVSoUqxWKFueksmdQNgV0olvi7Y5f55J27eWrPIL/1wutLx4KGTmMkxIWpudsSTs8VpFMrTd3phcIi55HC+cw0v3fhFdqDdZzPTON4HsP5eY4vjPG/bX8U9QfIKbMc1yswQ21/JqOgShqqbCwW3qltAWzHZXw2QyToYy5dIBEN4C4Kz+h0msZ4mPbGGOl8iYVsiYVciY6mGIlokKGxObKFMnPpwqLRa3MIR/10DTShaSo9W5pxHBer6lDIVcimS5QKFS6fm2TPPX1sv6MTgHdeucjktRS+gEFqNkcpX6EuEeLAg1swTI3RK7NMjy1w77u3E44GePpPXyEY8dHQGuPgI9uZm3qNmck0VtXh2a8c4YmP3UV90410u8l8fomhbyiVojUc5lIqhV/T0BWFpmAIn1YrX3ExN8Ujrb18qGsHCgoFp8p/OfEcL01dXrpHWd+HcC4DAkm/A7wMivEwLAaaePbJ2kqq34nAQfJySEojQliAhqT2IiMADW5y4TUGQnxu9wGOz04yUyzwaGcflzMpzqfWLlq1HP+gwmm5Di+MDLNQLvGB/q01ysJ4kE/+9EMIAe9MjBFUdNoDYb76V2/y+T94aem76VIZSzjc19dF2GeuIPySJAm/plG0bm/Pq6gyrR0JJFlaKlvX1deIshjUfDQ1yj3JHg419fPb52t9qTeDpKpFXOGhfp+uEyEEc9WR/4e7/w6vLL3KvOHfzifno6Occ+XYXR2qOriT3W23bZyNbQyYNDDAAC8wzDDMwAzD8IIHhugA2MYGp3Zqu3MOVd2Vc1TO0pFOTjt9f+wjqVQlqVTdnvm++e7rslsl7bPT2Ws/z7PWve6b0cJpikaGofwJmj2bVzhfAaiSG0lUUEU3sqgiCc4oYFk2r54YpC4W4Mj5UXZ0NzqKg8PTbO2s5+k3LvDArb2Mz6Q5eHqIW7e0Yts2V8ZmefHIZfraEsxnbs4j5VpkUwVef/Ysd71zGxdPjWED4ZiP4UtTRON+3F6NcMxHTX2I/h0tbNrZSjDqJZspMj40x/T4Aq3dtZSKFUYHZgmE8ni8LmK1QUavzDAxPEcuU8If9CDJIgfetY1zx4dp7qghUuNMa9tCYXyqQ0ZoCYYwbYumYHBJ3T1Q9eucLqU5nRp19JQC9fQG6wlbHrzX5CsEMYSg7nZ+XuWaRSQQ/c52V39u6ScZQVlspF+ZZBMQkEWRkObimeErVCyTsUyarvCNud//W4NTFiV6ojH+5ugh3tHWiU/Vlt5+ggCn52doC0XYFKzFdY2fp6bIUHGC9NqkVb5cYXg+xW0dKxXsbgRBEHjg0Z2k5vMMX5mmvaeORz60dyn7KAqLnJ/l2z5bzuGW1Jsimtu2TaliIAigyjKlio6qyI5FuhVkU+BeRmZS2MUAthuKFR1Zct69umHS578LSVDRrDgeVcO0oFxVsve4VIYn54mHfAxNztPVFMfrUtm3pZWFbIGxmRQAPS013LrZWUufG5qhozHG3k0tnLh0fTvajRCN+5fukS/gZtst7YwPz7HtlnYCIQ+77ujmzZcv8ObLF9i8q43WrlqyqSKXz4yTaAzT3FHD7ju6Gb4yQzjmp29bM+29dRx5+SKZVIG7Ht5GLBEkly5y/OAVdtzWSU19iO4tjbR111LXGGF6YoFIjbNG1mSZxqvqhmshoLiRRZGsUSKibUzGdDUI8voO1BtBezDC+7r6GUwvcGt9M1viN24OuWFw2rbNaDbNc0MD5PUKtzU2szVey5uT45yYmSTi9nBfayfzpQKDqQXuaWnnfHKW+VKR2xtbqPH4cF21LiroOj8auEi6XGIonaIttDq/MOH3scVVy1+/cJBP3LqDVLGES5Y5OTbFD06dZy6X587O1hWfMXSTqZE53F6NfKaIy6OhuhQqJZ1ysYLH70IvVvjQp25nfiZDe38DqktZemHsjbXy2bPPkaoUmS5m+ObQUV6cvsTH2veuKVu5GrLFMs8cu0R7bRRZErk4PkvY52YhV0SRJNpqI5wfrNAY07H1FCcGJnBrKpZlYVo2NSGfI0kpCjTHwxwfmEAQ4J5tnUSDHgYnkuzsaeTg6SEeuLWX88PTnB+eYSqZpbu5hrlUfkkqBBwdoMPnRrg4MkM6t7GWsatR1xylrtnhzCqqzO33XS+3ctc7t634947bOtlx23I30/6HVspyuDwq97x7Za145+1d7Ly9a/nftzk/h2NOUDr81nlkQUMV1xfHBidbv7+md4US3/83UDFNDs+MMV8qsLumEVVa369mETcMzpJp8I8nj7G7tp6OSDNBVWMoneKHVy7w0U3bODY1wfcunaM7EuPM3Az3tLQznssymklze+P1bVGvj48wmFrg4c4eTs6s3firyTI/v38vf/XCQf7LD58nVSgiCAIvXx4i4ffxb++5ja6aldKUuXSBqaE5glEfsxMLlAplfEEP5WKF2pY4ggCXTzoaPZZp0dq7shOhKxDnM9138J2R47gllZML43y4bTf7E6vLo6wFRZaIB7zMpLKUdZNSWaeoyEiCQHdjnHS+SEM0QEsizMxCjoVcEVmSMC2Lbe31nBmeoqKb3LOjk9lUjumFLPGgF8O06G6uwefRaKoJ4XYpNNYEuXdPN5dHZ9nd10RbfRS3plDRlwvhva015IplkukC9+3twa0qTBUyXMrMoooyWyP1Kx4Wy7a5kHbMqkQEeoI1xN0ru0wmCmmuZOZwSwpbIvUrSk2OEp7O5cwc51PTTBez2DiJmb5Qgq5ADJekrHpPbdtmLJ9iMDePhEBfuJaI5jQHGHaZ4dybuOUgzd5dCLbM5cwsU8UsqiixJVy/xIk1bYtjC0NMFVP4FTd9gXpi2ltTZrgalm0xXshwIjnOcG4B07ao9wTYGqmnzR+tZslXfubM/DSvT42Qq5QJai4ylTLvar1xruSGwZmrVCjoFW5vbCHocjJ8L40MUevz0xOJIQoCXzt7ks7wcqCsVzsdyaToj9XQE4nRG42vuR1AXdDP773zbi7PJh13YsMg5vPQm4hTE/BdN9X0BT3Ut8dxe124fS7cXo1cpoDH78YX9GAZJp1bm/H4XVRKOprrmrUHAptCdbT6olQsA0kQUUQJw7ZQbJucUSZvVIhq3nXbtSzLRpZEogEv9dEAF8fmqI8GKOsGYZ8bl+Lc9sn5DB11UUrV3wsCBL0uOuqi2MCpwSm6G2Jsba9DkZwpbcjnpi7mEBVqqiqHXU1xupqW72XjNV41qiJz29blBgHbtjk3Oc2vHfwOHlnh83d+mP7wMjG/YFT4kxPPcmh2GEkQ+c2t9/CTnbtXdL58e/Akf3P+VbaG6/i7Oz60FJyWbXNmYZLPXzjI6zNDpCulJYK9JIiEVBd3JNr5md599ARrVl0uPDF2nj8//QIuSeEv9r2PO2qdrhhRkIlqrYiCjICEZVv806U3eWzoJHGXj8/f+WG6gs59EBHoDzbQ6I7gV1xEtLfewraIkmnw/eHTfPHiIYZz80t8W1kQqXH7+Im27Ryo7VhBWAFIl0t0h2IMpucp6vqGaYc3DE63rCCJIueSs7SFwiiiRMLr5cWRAlP5HBfn50h4fHgUhflikel8jnPJWbyKimXblAwd3TQpGU49rsbjYyi9wHQ+x2gmTXdk7YWxIAi4VYUtDbVsabhxV72sSNS3OfWjcI3zAEfrQiu28QbXVtq+lJnhHy6/TrpSXMG+6AvW8gu9+6lYJsfmxtAkmT3xtZXevS6Vff2tS/+ui6zeD9uScGqw8eDKB+dar5fEGlo8tm1jGham6ZggiaLgMGrWcdoG5742eIO4JJn5coHRfGpFcM6WclzJLoprm5xemHReUIvN5rbFxfQMZdOgwRtaSrBYts1r04P8wbEnGcw6tgMtvjB1Huf6pwpZRvMLfHfkNGdT0/ynnQ+yN9583bmatkXFMpfs7BdRNFKk9HG8cqQqriZgWM62FctcwdASBAFZkDi6METBKHNXop823/qDwXowLIuvDxzjz0+/QFYv45YUukJxoi4veaPCSG6evzv3GhfTM9e1jfVHavjy+aOcXZihaBp8vGf7ho55w+D0Kgof7t/K04OXeWVsmDsaW9hVV8/e+ka+dvYkXkXh0WomtikQ4CtnjpOzcvRGO5kt5PnepXOYtsX3Lp3nnR1d3NHYwsT5DN+8cIbmQJCYe+1g0U3ToepVvzzLtskUS+imRcjjWlcyP72Q50ffPkwxv3ZGVwD23NnNph1OoB2cHcSvuPhU574Va0y3rCAJIrpp0heuxSXJb7msYts2qWSeZx8/Tja9qOQOXf0N3HKgB3kVOYxrP18sVBi8OMXpY8MMX54hNZ/HNExcHpWauhA9mxvYtKOFmrrQmoEad/lIePzMlfNcycytcKAezCaZL+dp9AZJlgpcSs+S08uEq9PLgl5hOL+AAPSFEktNAiO5Bf77yWcZzCaJal5+oe927m/sIVw19U1Xijw9fpG/PvcKlzKz/PGJZ/if+95Hkze0oWWDS/Ljk+OoogcRkdX7RJZRMis0esJMFlOUrbdn8XcuNc3nzr9OVi+TcPv59c13cVddJ15FRbcshrJJ/vrcqzwzfvG6Vry428svbt1HulzCLSsoG7THvGFwCoLA5lgNfdEYlu1YrwnA/W2d3N3iKOAtljk+tWUnOaPEN0af57bGRjyyi09sXl70y9VA+/TWXVepHqz+pVRMg79/+U12tTRwS2sjNvDK5SH+5sVD5MoVHujv4tO371pTrjCbKfK9rx5ifi676t8Xry0Q9iwF5+5YCxcGpjmWHK26P1cTKi4vcZefy9lZCoZOf6gW9wbog9fCtm3S83k+/9kneeGHJ6vkBdi8s5UDD2xxCv3rwDQtzp0Y4bGvvM6pI0Nk09dnsgF+pEjUNYa59+HtPPi+XQTD3uvus0/RaPNFOLMwxeXMHJZtI1WD7MzCFBXT5EBtJy9PDTBZyDBTyi0FZ7KcZ7aYQ5NkuoLxKgXR4rGhk1xIzaCIEp/p3cfHO3chXzX9d8sKH+nYCQL88fFnOLMwxbcGT/Arm/ava0u4dP+wyRlOA3pEa+ZGhgVxLUDeKKOIMgnXzSlCXg3Ttvj+yGkmixlkQeTT3bfwaMuWJR1blwSbw3X89tZ7Gc7NczHtnKNjhGUsDTBhl5uJfIaB9Dz3Nq0u/XM11g3OolnmxZkT1TeBzYGa7RyfH6bNW4ssSlzIjOJT3AzmJtGqdnS7Iz1kjQLPzRxFExXujG/FtC1eT56haJTZGemm1hXh9eQZdNtEtwzuqtl+3bGT+SJPz42xqd5JOU9ncvzl869THwxwb2+Cbx07Q399Dff0rG6SK0kiwbCHUqmCoZtYlo1lWutyekdy81zJzCEi4paXb02bL8aOSBObQnXkjPKGeL3XwrYdAsQXPvvUisDctredX/qdh2lsvZ6BdDUqFYNnv3+cf/7b55mbWakvJFSlXRavzdBNRgfn+PLfPMf5U2N85jceoq4xvGL/siDSE6rh8dGzjOQXKBgV/KqLsmVyZmEKWRTZG29mKDfPoZlhBjNJeoI12LbNeD5NWi8RVj00eUMAzJcLPDtxCQubLn+UdzX1rwjMpeOKIg819vHY0ElOzk/w9MQFPtKxk1rPjRUTBUQSrh4Mq1xlU62PZCXH0flBfIqbsOIlrL61ckq6UuLQjNMKVu8Ncn9Dz3UC04Ig0OQLc299N5fSs9g4iaD/duSFFfzfVLnElujGhM/WDU7dMjiRuswHm+7mjflzDOYmGcxPEtMCaKLCldwEYdWHKIhcyo5T744yUZzDtC36A62cSQ9yOj1IWs+T04vEtCBPTR3mw833cDx1mQM122lwx1BWsQ3IlRyiclPYeeO9cnmIfLnCr9yzj+ZIiMHkAsdGJlcE59WjSCwR4Hf+5IPksyXy+TK5TIlcpsixg1d49dmzq16vhc1dtV18sG0XqiixOHJKgvMgOOuzBRq9oaUM4kawGJif/7MnefGJU05gigI7b+3gF3/7YeqbI+sGpmlaPPeD43zxs0+RzThTYVWT6eito39bMzV1QSRZIr2Q59LZCc4eHyGTKmAaFodevADAr/yHdxOO+q6qMwt0B2tQJZmZYpb5ShG/6iJVLjCQTRJQXXQG4nQG4rw6PciZ1BQPNDoZxoFs0llvegJLxlajuRTjhRQAW8J1K8Sur0VYdbM90sDJ+QnG82kGs/MbCs6SmWGqeB6/sjF5UFV0rOtnSxm2h2/sBrcWZopZJovOC7HDH7suc70IURDYEq5DWWwbs+GB5m56w8tr3Yl8Zk0B6mtxw2ltRA3Q4IkTyY1Tqc7bLdvGtC0sLCRBIqI60o8hxVFg88oual0RZsopMnqBuXKajJ4HQaC+qjHjk920eBKE1hCCEkUB0arWtsoVnjp7mVvammiOhJBEkbDHTba83GlQMlNMFQ5jWEVcUph67600t1//JYqiwGvPnVt1OqiIEs9OXuBwcniFd2ZvsJbP9NxBRPNQtgwCysZFmmzbJjWf5wt//tSKwNx7Rzc/91sPUde4fmA6ZP1xvvr3LywFZiwR4MM/c4A779uEP+BGqK4rbdumXNI5fXSYL/7Ppxi4MOWIHr9ykce/8SYf+dkDK9a0zd4wQcVFulJiupChxRdmLJ9ippilxRch7vbRF6pBEkTOpaapWCaKKHIx7VDP2gOxpVFhspBe6o5p8oXXrQuLgkCTL4SIQMk0GMunNnQvXVKAuKsD23aSQTdS9dZEmdvjPfgVF0Fl4y/TazFfLlCs2tLXeQKretAsIuby4ZIUKpZJX6SG3TUNuGTFMV2yLXyKSpMvtKHj3iA4BRTR8buQBBFREIlpQQ4vXEAWREzbWvq9LEqI1Z8LRonXkqcZL8yxv2Ybta4Ip9IDRBQfcZfzxcnC6kJKSxfp9eLXNf71yCkCLo2h5AI/t3/PksThXK6wQrYwWTpH2XSI8TljCss2Niz6vIhtkUb+YMfD2DhBsdjZ4ZU1JEFkoVxgLJ+6jv61dLeuuZylqeyfP8kLP3ICUxQF9t3dx8/95kPEa9e2Z1hEuazz2JdfZ2bSedt6fS5++lcfYP8Dm69bowqCgMutsuu2ThRV5k9+5xskZ7MYhsXT3z3G/vs309JxlQSky0udJ8CphUmGcvPsiTdzPj1D3qjQ5o/ikzU6AjF8sspwbp75cgG/ojGUm0cUBHqrgQuQ1ctYttMA4Fe0da/L2cZVJZ7bZPSNqgLaZPQpZEGrCqytv+ZM6wVGC3O0+xK4Jf2mpGKuRt6oLCV5fLK6bkvqoskuOFN4WZRIl0t8Z+AME/kscbeHR9s3bei46z69Hknjwdq9SILEjnAXsiChiDKjhRk0UcUtqciihCxINHlqUEXZcQd2xymaZfoDrUsjZURzRteEK4IiytxfuxuvtPYIFHBrfGr3Lv7mxUMUKjof2bONrQ0OaSBTKjOTyXGga7l2p0kBVNHHXOnMhj1MrkVc81ExDZ6ZOM9owXEZ25/ootETcmzIFZXdsSYCyvVtaoLgsGcWYds2C3M5PvdnT/DSk6cxTQtJErn9Hf185t89RLTmxu1Ptm1z5fwkRw8uk7T33NHFbff0rZs8EgSB/m1N3HKghx9+8zAAs1Npjrx6ieb2Zcc0r6LR7o9yYn6CgWwSw7Y4NT/p1HvDtSiiSL0nSI3bz2Qxw3g+Ta3Hz2Qh45QSAsv7WqpX2jeWTFm8tkVLgo0SPCRBpdm7C2f1KWFgrXssv+KmYplczk7RG6h/y8F5M3n5xSb/q3EyOYkqSnyidwdHZsY5ODXCw219N9zXusEpixK1bodeF1aXa21d/sbrtr1a4vFauUeAeneM+qt8KRf3uxYEBA50tbG1oRbDsoh43UtvJK+q8lsP7KcxvJyBC2tdYIMqBRCRkISbV2mYKKb576eeIqC6aPSEWagU+POzz/Jz3Xdya7yNomFwNjfF1kjD0lrrqhPGU9VYXZzKfv7Pn1wRmAce3MJP/+r9RDbYl2jbcPS1y0vTWVmRuP0dmxxFgRtAViS27mnnqe8ew9AdicpTR4d55MO3LL1EFpNCwjAMZedJV0oMZJO4ZYW+UKJqfeGmzR/hSnaOy5lZREEgXSkS0Tw0VpNBACHVjSSIVGzTqRNfVZq5/rps0noRqzrzutbScS045sHLyRXbsimYa5fKdMukJ1CHJspvS+doceYEJlmjvK7PSdkyrlOp8Miq01JoGhiWhWlbjGRT1Hr8S0Jiq+HHTnwvmc70xi253rZttygKRH3XB7pLkemtXVlQTpWvIAgSM8VjiIJKh5JY8UVuBK9NX6HVF+Xf9N2FKkpY2Dw5fpanJ86xO9YC2BiWhb6K9qooiLg86nJg/tkTS2tMWRa595Ht/NSv3Icv5MG0bGwsbNsZce3qty1QbRkTBGRJpFyscPbE6NLSKhjy0NpZs6H7KggCtfUhNE3GqFL5psbmKeTLBKvBKQgC3YEaNElmqphlJLfAZCFDzOWlxee8PBVRYlO4lqfHL3AhPYNHVimaBv3e4IqkWIM3hE/RmC8XGMwm0S1zXY+SgUwSG6e80nRVkDv3cpmJtJ58Z9E01vTdBKfOeTEzhSrJxDQfde7Qmtuuh6jL6WQpmjoT+TRl01iTG5ss5SkZK2uqHlnhUmqO0Wx6yepjtljgg11biLnXTpz92INzpjxHwSjS41+9xLFRWLbNeCpNWV/jjSdAjd+LT3NGq7KZIaMP45ETlMwUplVGWlWpfW1kjBI1bj9qVf1AtKHWHaBg6kvTsGQ5fx0DBJx2NM2lkM+W+PJfP7u0xlQUifsf3ckn/8078AZcHB2dYDaXR5UkFEkiUywhSyI+TSNXLqPJMn6Xxq7mBrKZIjMTqRXHuHR2gqmx1bVfr8XUxMrtCvkyxUKFYHj5gWjyhQipbubLBc6npklXiuyNNy/5iywSDVRJ5kpmDr/iQrdMugLxFQ9okzdEZyDGG7MjnFqYZKKQodW/+uxoupjjWHIccJJS127nrXrzGJbFTCm35ig8mltgJLf2vUi4gqT1IhXLoMX71tlBTrN2kLlynkuZOSYLGdr81yfyLNvieHLiulE6rLlpC0SoWCZR282eRBOboz+GrhSAklliMD9Kt7+d+UqKglGkwV3HpdwAc+V54lqUbn87aT3LUH6UZk8DgiCQM/IM5kcpGkUkUWJzoBdBEDifucx8ZQFRENkR2oxfuT5jW9J1/vCHL3Bh+irVc9tR0MYGlyrze++8e6kzJax1Ydpl4u6tZCtjyBswtb0WfcE6vnzlEN2BBHWeAFm9zHdGTtAXTCCLEj5FI6b5ruNOgmMCLADf+vJrPP2945im88Zv6UzwoZ/ejz/oTN1kUXRaxxQng+fVnHV7plTCtGwaQt7qCGqTyxTJ55aTJTNTaf7s9x/buEaWDYax/KAYhrU0ii4iqnmp9QQYzCQ5lhyjbBn0h2uXLCgEQaDFFyGkuhkvpJcU8HtDCcSrzsSvaDzcvIljyXHG8im+PniMX+nfv6IjCRxJy28PnWAwm0QSBO5v7L2uLNXkDeGWFIqmzkuTV3i4qR/vNUbARaPCt4ZOMF9eqYN0NWRRoj/41j1blq/Nxe2Jdk7OTzBVzPD9kdP8Yv8dK54D27YZzi3w7MTF617dQ9kF8kaFrdFaBEEg6r5xRw1sMDjzRpHDCyfp8LUyWZxmqjSLW3Lx6tyb7IlsX6pTuiUXFUvnSm6YVm8TaT3LU1Mvcl9iP2czF1FFFVmQGMgPU+dKcDZzgb2RHaseU5VlPrFzB+mr7Nds2yZVLPHypSFKhkF7bPmNq1s5ylaG0dyLKKKXCD0bubQV2BVtZjiX5K/Pv4BetQHYFW3mPc3blqZaPmUt1TeB5394kleeOYNeWRY9HRua5aWnTvOej+5DUSS2NtSyuUqsWGTlAEs6qovrarEqe6JfHUw21wXXTcG2rysheRWVdn+U0/OTvDE7giyIbArXrngB1Lh9NHiCnE9PkzcqeCSFDv9KI19BEHiosY+XJq/w7MRF/vnyEXTT5D0tW0i4/SDAXCnPD0bO8NUrRzBsi92xJt7XunVFkAuCQE+ohnZ/lDOpKV6cusxfnn2Z97VuJaJ5sWyLyUKG7wyf4rGhk3hkZU2BMdu2qVimsxSp8m9Lpk6hur1pW0wVM3hlFUWUUCUnualK0lIWevG7eKR5Ez8aPctgbp4vXX4TRZR4sLGXoOokna5k5vjchdcZz6dQRHGFJaAkiFxOJamYJqIgoEky9d4b13U3FJzVFUD1/53/BhU/bd5mLmQv0+N3qEguSSOihkhVlhksde4aNgW7yRl5snqOuBahbFaoWDoJVw2quPrUUxZF9rWvbKZefLAe6O/m9777FEeGx2kIBarHDhN3bcG0y0wXjmLbJgg3N61VRYkPte3i/vo+0noJj6wS07xOoqOq+i0AqUrxuubdQr7Ec4+fcNaYiiOwpesmpaLO17/4MjW1Qe64bxOSKK5qf7TaF2FZ1opg0twKiSpf9q0gFPGtyChDtXsjlODbQycZL6SJu3zX6SV5ZZXuYA1Hk2MUDJ0WX5h67/V0uJDq5re23UvFMnh1eoh/vPQG3x05TVTzIgiOqvpCdaTbHmngd7ffR637+uRY3OXjU917+aPjT5OqFPnixUN8b/g0Yc2DaVvMlfJk9RL7alrpD9Xy+YsHV73esmXw56de5MT8OCXToGw61iCLhkapcpHfeuP7eGUVTZLRRBm/ovFL/XeyO76ywbojEOVXNu3nv554mtlSnv955iW+euUoIdVF2XLWvqIg8NM9t/LsxCXOpla2QyY8PloDYUQEghuUlN1QcEqiTNmsMFtOMpwfR6kyLzYFu0lVMrw8d4hufzuyIFE2y1SsCuVqFk1asmp3wjushiiYRVySyqbgzg37jcByyj3kdtESDXNqYop3b3NS0hUrS7oyhGUbWNycrfoizKopbETzEta8zJfzS4EoiSIiAooooa1ShLZth8lT2xDmkQ/fQjFf5hv/+DLlkkEmVeAf//IZookg/duaNpwok2VpBU2spb2G/+ePP4Db/dZKAqIk4A9cn2DrDtZQ4/JRtgy6AnFnpLsKAgLbo/U8MXYOG5uuQHyJzL5iO0GgzRfhv+5+mK8NHOVHo+eYKKSXAlKTZBp9Ie6t6+Jjnbtp8YVXvReiIPDOpn5EQeBLl97kSmaOuXKemVIORZSIaB4eaNzOZ3r2MVHI8L2R01dlVJdh2TYD2SSXMisNoTRJQqtmSUumTslcTuCookS6cn1DuiiIPNjUh0dW+fzFg5xbmGammGW6mMElKbT5o3yiazcPNPYxWcgwXkjhri4NWgNhCkYjZrV2vhqtcTVsKDh9softoc0cnj+BX/YSVaMUzRLHFk5TsXT2RLajigrn0lf47tARom43Na4LaASZTNtU6ixiWhQbm7nyPLqlM5gf4UpumIfq7iaorD7Em5a1it23zUKhxMDsPD21y6UZWXDhlqMICNS4t76lUsrR5AhnU5P8ZOctnFqY4C/OPg/Az/fcyZ5YCzmjTM6o0LYGCWHnvk5++lfvp60rQbmsk0kVePwbb2KaFhOj83zu//0Rv/4H76OpbX0e7SK8fpej81otpVQqBh6vo8P0VjFVSmLrNnXVspYgCOyINvJPBz6GZdt4qoZBp1KX8cpu2n3Omq0l4OaL+z+EJqr4FA1RsCmZZVzSyvssCAIJT4Bf6d/Ph9p2MJhNMld1R49qTha41uNHXsX+8GpokswjzZu5I9HOYDbJVDGLbpn4FRet/jBN3jCqKFHj9vMP+z+KIAgrSjuL+/jd7XeTN0qISNU+0MV5oI1lGwiCyNVzGUFgqcXtWiiixD31XeyMNXIlM8dkMYtlW0Q0lc5ADQm3ozH0q5sP8FPdtxBUHe2rsmnw+uQI8+UCsiDibVeXLAHXw8ZGTkFiV3gLu8JbKBoGL4wO4MXg3pr9AKTLZXTTpjfQyb0xAd2y2BHqJl0uIZTj5PUKnb5WbGy+P/E0O8JbSLhivDhzkIJRWjU4y4bB3770BmMLK3mIlm0zMp9iOpvn07fvWvq9S47gkt+a9d8izqQmsWwLwzL51tBRDtR24ZEVnpw4y85oE6IgMF3M0OGPLnVoLEIQBHbf3kl7T+0SU+cjn7mL5Gx2iS544dQYX/zsU/zKf1zJc10L/qCHUMRLskp0T83nSc/n31JwWrbN+cwQA/lx6lwxJktJFEHGLWnMlBeIayECspvh/BgeuQm/4iWt58gZRc5lBhkqTPJQ3W34ZDeWbXEweZpUJcet0c0MF6aQBQm/4sEve5grp0jreXTLIODyoKoOT7kvUIdP3rjjuSgIRF3edXm6bllZarBeeb0mA7lDnM+8gGGVCKuN7It9DE1y9lWxirw2+2UaPVvo9O/b8DkJgkBY87A77iy5TEvnlbkvsaC3Uue5F4BaT2AFV/jCwiy9kThzxQIt/hAFY2PtaxsWVxGq7V3zpQKvjg9zem6aTKXME4OXeHr4Mt+5fBbLtoi5vUutYD5Vw1dt6XLkHwX2RLaT1jNcyFxhZ3gzcS266vFs2yZbKpMqlFb8L1uq0FUT4/ffdQ87m9f3D71ZGLaJV9EYyS8wU8pyf30ffcE6MpUSpm0TVN3EXN51bu5yC5wgCIQiXn7qV+6jd2tj9ZrgzVcu8i+ff2ljGrd+F21dyyn3bKrApXMT6ypNrIWyWWa0OEPCFWWylMS0LeYrGc5mBvHLHgbzE2T0PMlKmuHC1NIIM1GcRRIk3FeNkAICQcVPkyeBJimMF2dp8daSquRI6Vmu5MaZKM0xX8kwVJhiophEEkQmi3M3fd5vFVl9joNz/0xUbWFX5H10+vchi8szHhGJqNaCT179+dsobCzmyyNk9bWvza9qhDU3blnmYmruhqZUi7jpOmet109HKMqdja1IgsDByVF6wjHmS0WKxvprPUEQqHcnqHffuMbjUmR+/R17rn8QheW+0LdLcrgWXYEa/nXwCIdmh9gWaaTG5WcgO4daNacVEYhoXuKujY1cgiDQ0BLlM//uIf70P3yL8eEkpmnx5GNHiNcGefSjt16XoLkasiyy6/YuXnzyNHrF8Sp58clT3HZPH94NikIvQhFlVFEhWU4T10Jk9YIjuOyKENWCZIw8Y8XZquyGxFw5RUrP0ulrYrQyU9XTverFo/i4nBuj3owR00L4ZDch1fldxTKIaA5BvGRWGC3MsFDJUO92Rrih7NMYVpH2wLuukwV9u3CICzoZfRrDqtDh20tYa0RERhKUqpmvgYVJT2D/kuzoIkzbwLZtJEHCtPXq/VAQrlrPLh7DSY7e+EW5NVqLbll0BqMcn53kltqNqfnddHAKgCpJnJmbZlMsQUcoQp3Pz5Z4LZIgMJbLMFfMky6XyFUqzBXyjGUzhLRl+t1Gj6TeQBXgx41bYq3Ml/OkKyXe3bQFURCwsDmQ6EKp0sw2SjVbhCAI9Gxp5Kd/9X7+4r98j9R8nnJJ5+tffIlo3M9dD25BXIMnKwgC2/e00d5dy4XTYwCcOjzEM987xjs/sAdF2djXZ9s2IiK3R7dg2CaqqKDbBiLCUrNCTAs6SS3bRBFlypaOZVt4JBdxLYQoiGhXZdbr3DFCqh9NVNgZ7kFAoNmTIK6Fq80QTiAXjHJ1WdSzNPomS+coW1naA+9ccZ7Fss7Q9Dy2DW21EdzazWXbARYq4xxdeIy58ghZY5YXZz6PIrpp9+1lW9g53oXsy1zKvkLZzLEz8t4V09rB3JtMFS/gV+IM549h2jpNnm1sDT2EKrmxbJPB3GHOZp7Fsg3q3f0Y9voj4WguTckw6IvEaQmEGM2l12UGLWLdb9fRADKWOkGkasf7Pc3tHJuaYCqX45GOXgZTCyiiSKpcQhFE+iJxdMtEt0zuaGxFEcW3NBX7Pw23rPLe5u1L/xYEgdtq2m+KnG3bNpWKgWU5rBZRFFhIFdh+Swcf/tkD/NNfPkOxUCGbLvKlv3qWeG2QzTtb1tx/KOrj3R+5hb/8wxlKxQrlks5X//5FDMPiHe/eTiC4ekHbtm3H+mI+z6VzE8xMpnnwvbvwaM6IK19T0FnOdDoBcXWP7WrZRVEQ8F7jzC1w/e9kRWJ3pHfFtHgtDE3PM72QozayumaSZVpL5I6l865qJi3CK0fYEnqI2dIAB/U5doTfjV+J45FDS9u0+XYTURt4euovq1YXyygYKU6kfki3/062hB4kq8/w5vw3CSg1dPlvZ7Y0wCuz/0BP4AB17h5G8ieZLQ3Q7Nm+5nUFVI0nhi/yzOhlZFHk/Z3XS4uuhnWDM1MucWh8DI+ioJsWDX4/E7mq7IctcCE5h2lFKOsmmigzXyjikzX21DYiiSIxt5fWqgehbdukF/IUcmV03aBc0ikV9ep/K1y5sOwdYttw4fQY/qAbzaWguRRcbgVNU5BVGY9XXVV6Y/E4uUyRXLaEXjGplJ39Lx7LGYGWa7aXz03yyjNnnP271KXjKaqE260SjPhuqq5omhZHjg6h6yZut4qqyszNZdm1q5UH37uL+dks3/7ya47G7vgCn/uzJ/jNP3z/mkoIgiBw+739XDk/yff/5RC6bpJeyPOPf/kMrz17lt13dNPWlcAXdCMKAuWyTjZdZHJ0noGLUwxemmZ6fIHuTQ3c/57VCR9X37uB7A9xS1HqPLcwWzrFcO5Z+kMfxS1HuZT+DlFXPxGtm7nSGcbzr6NbOSJaD42+O9FEpwUup08wnH2WJt8BxvKvkKmMEFRb6Ay+B0X0XHNMi+niMaYKb4J1O+NzOhXdoC7iXzFy2rbNiz84zovfP770O1mV+Mi/uY+O/uXcgyZ5SEidmLaOLKrEtBbC2nKjhiAIuKUAkiqjiKsvDVySj93R9xNW6zEsncH8EWbKV+jy38Zg/ggeOczOyKOoopuY1sZg7o1176smyaiixGguTXsgslRiuRHWDU7DsinoOlG3h4RXI+bxMFPIo0kytg0+VWM6nyOgaZi2RY3Xj2Gtxjx1qGNf+ZvnOPjiBSoVA0M3MU1rST3OuuaN+KNvHeap7xxFkkREWUSSRGRZQtVktu1p45d/791oq9jJ2TZ8859e5fnHT6x6nBVvXhuee/wELzxxCklyjiHJIrIsoqgyPZsb+Y0/fN+S5d1GYFk22WwJr1dDlkQ8HpW62iC2ZaO5FD7wqTuZm87w/I9OYls2l86M84U/f4pf/g+PEImt3q2iuRQ+8rN3IQgCP/zmmxQLFfSKwZnjI5w5PoKsSCiKBILjB6NXR+63grKZZr58gVrPHiYLBxnJvUCdZy+y6GY0/zIRrYfx/Oucmv8C9Z59BNU2RvMvMVM6we7Yv0WV/JTNDAPZH7FQuYxXThBQmzHtClevz5y6t8Vk4U3OLHyJrsCj+MItDLlGiAQ8XNv3USkbvPT4CQ49t6xioWoK7/roxjOtG4VXCuORHJKFKIgoogvTctaYWX0GvxxHEZzAVkUP3htUCSbyGTZFE3yybydvzoxzMTXLLbU3ditYNzjdssye+gbq/U5aWAD2NTY7a7FqQdWGpX+LgkCN17e6WJNtMz+XY3ZqYxINwHIwXTOln5vJrPPwOV0hMzdxHKv6crg2fxqNr3ec1SEI0NAQxuVSaG2JIYoiVJuMBEHAF3DxiV+6l7npDCcPDy5lcL/2uRf59L+9f6ntbOU+nc99/Bfuobk9znf++XVGBmaXXjSGbq5J6xMEx4yoo7d2zbXt1QiqbcwUj1OxsmT1CWrdO0mVr+BXGrBtC00KcGr+H2n03snmyCcRkKj17Oblqd9jpnSSRu/tAFTMLLXuPbQHHkJg+R4snRcSU4XDnF34Kj3BD9Dk28/5kTmKFZ2R6RSJsJ+gd3lkW5jNMnxpbRHyHycEQVyTHCOLGmUjj42FgISNhWmvn3nvCsUYSM9zMjnF5mhiQ2JmcIPg9KoqHkUhVSyRKpUIuV1VPqhIwKUxlc3h11T8moYAzOULSKJAxONxJBx1g9l83iH7utxs2dWK5lIcJYN8nojHQzJfoGwaxLxefKqKjeOFkiwUsG2bqNeDX1v5wLZ01iDJayRREOjb2rSC3/pWYFs2ifrQmlKVpmESrw1w97u2YVWlR5rb4yiKzOZN1/a7ruSOJupD/NxvPsTj33hjqaRiGhYTI0k6rlGhv/pzbo/Kfe/ZyY5bOzjy+mUOv3KJkYFZspkilbKObdmYgo2iKcSjfhqao3RtaWDTrlY6OxNIskjZMJZepou9hOUq51MRRfxKA4ZVJFMZxrCKNPvuYqpwmExlFE0KICBRMKaJuN63pDThlRO4pAip8sBScCqil6ir96ps7MprSutDzMwdp9F7B42+OxEFmUTYT0Ufc5TqfMuJN9u2Gb44xfz0SmGzt4LF3IeNBTbV1j1H9uRGeQUBkVpXN28kv85ceYiw2shM6QoLlQlavDvX/NzlVJJvXzlDydB5oKWbkqHzjuauNbdfxLrBads2Z6Zn+MtXDiKLIpGqbs8dbS080t/LZ196jfu6O7ivuxPLtvnnoycIuV18cvcOxtMZ/ub1N0gVS9i2TWMoyM89uof3fPRWZnJ5/ujZF4glapiemkE3DPZs6uXhvh7m8gX+4pXXEQpFLNtmXhD4qQO30xz2kNEn0K0CkqAiK1A2syyUBxAEkYjWSclMUTJT3PKuCDsf8hDR2jGsEguVITTJT0htXrLauxGmR+YYOju21NhsGiZzEwvUNDmE7/OHBygv5Pl3//m9G9rf1cgu5KmpDfDLv/fuNbdJ5gpYtk3cvzKrJ4oC8dogD753F+94eDvZdJFMukCpWOHUxDSvjowQDHj40C3b0Dwqzw0M8Fp+msq8jGXbHBx27CgE4J193WRKZQ6NjCIKAg/1dlPjDyOJLmZLp1AlP1FXH8O550iWz+JXmpBEDQQRu1pyEAQBGxvbtlaURQRBQlinTKKbeVp89zKWf5U6z15q3XtYyBZxawpBr8sxBK7Ctm3OHh2iXNq49qyA4Lw8rgm4opnhZOqHZPRp0voUFzIvMl8epc7dR0/gTieJJ1zLP5aWrq3Nt5uRwnGenvoLfHIMWVAJq/UrSi3XYrKQZXdNA5dScxiW9eOpcxqWxZePHKc/EeeTu3cwkcnyWz94kj1Njo5svlKhclVLUkHXcSkypmXzlaMn8Koqv37gdkzL4s9efI3HTp/l03sdzdqB5AJdsSi/9467kEQRTXZqiUGXxqf37iLsdmFYFn/0zIu8OjRCPJDg+PyXaPHdyWzxLJ2BB1BED0UzRaoyRN6YJa/PYNglcsYMbimMZVeYLB7Hp9QyXniTJu+tJNzLmTLLskhOLBCM+ZkdmyfWGCGTzOENuFmYTpNocihuhm5y6dggz/3r69z1gVtp6q4jt5BnYmCaKydHCMb8xOrDmLrJxMAMlmXR0JHAtiE1myHeECGXylc1hESe+srLuP0u+m/ppLmnfsmC8GqkSyVMy7ouOGE5c6yoMpG4n0jcj26aPLEwymceuYPWSAhZkvjasZNsq6+j1u/nmydP0x2P0RgMcGkuSV9NnMtz87wxOsb2+jrG0xleGhjiwzs245HjTBWOUOvZjUeOIwkas8XTdAYfQRODhNQ2porHqPPcgoRGujJIyZwnonVv6KEDiLh62BT+OIro5UTyC7hqIpR0H4VyhYBHW0HbLOUrXDg+suF927aNhwR31XzmuvWgLKrUunuIa210+G4FoFioMHmmQtct0OLdSURtXmo5FBDZGXk3UtV3c2qwQI/8E2yuybCQSlOcdNO2KYK0RgMHQGcwypfPH+NKOkm6UuKDXVvX3HbFua73x1ylwvBCig9s3UzI7cajqHTGbsyoyFcqnJ2e4TO37iFaneLe0dbC4+cu8PGd2wHwqAoHOtpI+FcW9C3bZiyV5onzl8hVKoym03RUW8N8coJW750YVomiOY+AQMXKoltF8sasMzVybYHiaQJKIxl9gtnSOWc7M0tOn1oZnKbFS4+9Sd/eDr7/98/xyGfu5dwbl7nvY3cweHqU9FyWts1NWKbF7Ng8yckF0rMZaltiztv80GVcXo2BUyN8+DfezcmXzzEzlkRWZM4fHmDL7T0896+v8ZHffISzhy5TyBbpv7WL8SvT1DRGyC7ksWyb4bl5hudTlHWDLQ21uBSZ4WRqyajpymySkMdNxOPm4swctQE/RV3nwvQcEY+bvtq4o4hoWfg0FU12GhMM08KjKHhUx35QAAIuDb+m4VEVKoZJxTAp6Tot4RDN4SACIkG1ldHci/SFPowsuPAptcyVThFQWhAFhb7Qhzky9794ffq/oklBFiqXaPDeTsy1MeEqBwKioNAVfC9Fc55jyb+hzfuLdNZF0VQF+ar18cxkitEr65vNWpbN8LhD8oiGfTz/+jDb+xvJ+i0mZyZIxPwsZBy50Namzbi9TrDZlk3GKCB5kggI+OU4bjuCqdvolo6qydS4OjFNi0pJ58LJUYIRL3ub+3DJOZJahqjm9C9blkWlbCCKIooqYVk2tmVTI3v46d5dzFUKxFxeEp6NkVjWr2Lby7IZ4MwQru1ltK/6b/kahtDVMwpRWBR+Wu5bdMkrD2/bNj84d5EfnD3Px3Zuo9bvYyaXu2ofCgiCIyhsmwxknyXm6sMjx6pqbM7fBBxStSJ4CKiNRF1d1AibCaorM2SSLOEPebl4dIhYfZiLRwbwBtz4w146t7dy5JlTAKguhe6dbVw8NsgtD21HkiUuHRti061dPPiJA3z5v32HmdE5LhwZ5Cf+7UOomsI///fv0rGleWntaxgmpmGRaI7R0ltP+5Zmtt7RS6ZY4tvHznBLWxNPXBxgU30CTZYYWUhR0HUaw0HGUhlOjk/xjt5OHj99gffv2Mw3j55mU30NT58fx7AstjfWsaepke+cPkfI7eKeznb2tTbxwuVBZFFkR0PdkhW7X1NxyTJuReHernbOTc+Sr+h0x50pe71nH6KgEHH1ACKtvvvwK834lUYEQSCi9XJb4t8zUzxBxcrR7DtAzLV5qdnAI8fpD38UwQ5Uv3ODojGFLHowrDxhrQ3dTGNaRWTRx+bwTzKae5mQy6Y11n7dMzFwdpz0/Pqmv4ZpcfbSFKGAm3jEj23bBPwujp4eIR7xcfDYIMWSTmNdCHsU+rvqACiXdF55+gypZI6uTQ0YpsV3vvL6Uklu9x3d9G5t5KnHjpJJFZgeX2DfPX3ks2Ve/NEpwKZrUwO2bXPyjUHOHBtGrxjc/+guCvkyrz93FkmWEESBhz98C37vxkks66vvqQp1AR9HxyfoT9Qwm88zmFxgZ0M9kiDgVVVGUinKhsFsvsD5mVnqAn48qkJXPMYbI2PsqK/DtG3eGBlnU6JmXUEjgPMzs/TUxLi7o41kochcrkBLOIQqekm4tyAgEtbaUAQ3LjnCfOkSPqUGjxTFwsQjRYm5enBJYRTRRUhtYbp4ElGQCautK44lCALxxggvfusQu+/fyqEfHueuD9y6ekJGFDANE9MwEaseLS6fC0EUEKtemLIqUcyVME0TQXDqcHpJp1I2mJ9KoVVbvQRRoFLWsUwLSRQRBJjK5GiNhol6PXhUhYZQgHJ1ybCproavvHGcY2MTJAI+R8t2Zg6PqlCo6Ev2iLe3NbOptsYRRPO4aQgGqA/4MUyLsMdNpap9tLWu1vHvBBRJYnNtAtO2Cbud7GhAbSKgLlPMwlo3LqmdkmEiCg4ZRRLixLV78KqOsbBhWWTKZVRJwiVHSLjeweOXLnJHs4JPzVExx5BFLyVzDpFildo3i1/yo0lBOoMPr/o8WJbNmcNDmMb6TeaCAA21IQZH5rC6bPxejWyuhMetks6VcGkKqUyRQrFCIrZMStfcCrtu7+J7X30dy3ZGurGhOR76id3Yls2bL1/E63cxPbHAB396fzVwwRdwsWNfRzVAHTS0xlBUmcOvXOTimXFCES+FfJmP/vzdPPal15gYTtKz5XpxvLWwbnCqksTHdmzjr147xJnpGTyKskKT86HeLj5/6DCX5+ZRJZGo14siSciiyE/u2sb/euUgv/ujp5dakT6wbfMSgd4lr65be2dbC58/dJj/+NRziAiEPS5csowmBWjy3gJAQGlBtyvIdhCPt666WBewcZISfsGFW/JRsUpogkqDdz+yqKBbFRRLR75qfZBoieENeuja3srp1y5S2xJj+Nw4h350nOmROV757pvc8uB2AlEfvqCHb/7FExx4317cfheG7oyKoZgfb9DDbQ/v4vmvvw7Anvu3UdsaJxjz862/+BF62WDTrU6Grmt7K6989zDp2Qy3vXcPLkUBbPZ3tqJIIqlCkZlsHt00mc8XiHg8xHxenjhziZ+8ZQchj5vOeJTeRBxVluhJOGtjSRSJelcW+UPu5Te1e5E+ec3y6NrPXA3btjk9O80zg1dwywr3tXciiyLfv3geAYHeWIw7m1v55rnTpEol6nx+Hurs4sjkBE9cuUS2Uubu1iZqvAkU0YckegioXYCNKobXe/wAyGeKXDo1esPtJFGksTZEU12YoN/NrTvbMU2Tupog6UwRTZN59fAAW3obiIaXp5WCICBJK0snXp9GvDZIqVDBsmxKhQpuj4bP767Wopc/t/ixcknniW++SWNr3KFBGmY1Mx/GH/Tg8Wk3rWKxbnAKgsCupgb++F0PMJvLE/G4+ctXDi79bX97K93xGMlCgbjXiyQKVUU5gZZQiN+/7x5G02lH4TsUxKM4hqkRt5vfvmc/jcGVrWK6ZdFVG+U/PXAP+YpOvBrsqfKy1KJt20wUBxkvDqJVdW/LZpGgGkNCosnTRbI8hVcOkNHnCalxBnKnUUUXqqhR42oiri0zSmpb4nz8dx9Fc2t88vfeh+ZWMXSD9/z8fc6XLovIioyiKXzg195FpVjB5XVR0xxbSsvf97E7kBUJURJp29yIbYPb6wgrv/9XHqJcrKBq8lJGr29vJ639jQiiQLJYpGwYVEyTJ85eJFeuEFp8IckSk+ksYY+bu7vbSPi9tMfCqJLEh3Zv5eT4JIYlb9i16q3AtG2eHxrk/vYueqIxBOBfz55iZ109/bEa/vbIG+yorSNXqRD3eNlZW48qyeyua+DQ+Bgf7N9C4KpSmIuNWSksYnIkydTo/A23E0WBWGQ56AJXNQa44gqWZXH77naCfveKQSGXKXLx9BjJmQwXT43RUi052dgggCQJJBpCZFIFXnjiJJfPThCv7SU1n+PS2QnmptJcOjPhbJMuOuWqku7MrkRhSV9YlNbvX10NN2ROi4JArd9Hrd+HYVkr1pGSKNIQDNAQvL4fUxAE/C6Nftf1X4YqS3THr/flLJsGPxy+wL7aZrbWOWYvRUPn6fGLNAWDSwatquQi4WrELfmcZI9dxiP5kAQZj+xDrDhvQk10Y9sWYTWBW/LilfwoVwl/LQkbiyLFXAnLstGzRURRRHWrTkBdZaGnagrqKmRs7SplArd3JSVMUeVrxKZNECw8AWdEK8w4a2qvquJSFHyaSk8iTk9iZY9iQyhIQ2hZGqQpHFzykVkLtu2YN1UqBqZuYVkWCCCKIrIioaorr2/N/bBcpazSKVY0Y8iiyAf6NnNmdoavnTnJL+zaiyw5/jJWVbforXQQ2bbN5dNj5DLXKxPcLERRJLSKCoRt2/gCbt7x7h2IooCsiLzjPTsourKEvRHuetc2InE/7/jwFs4OX+Y9H99HKOrFtmynceGdW7EtC7dX470/eRtz0xke+ond+AJuZFmkpuoPu+/uPnyBm+skuqmuFAFI+Hy4VZnnxi5ze10rR2bHafIFOZ2cJl0pcXtdC7pp8cb0KCGXm/11rbwwMYhumnQGo/RHlnVXddPk5ckhZot5tsfq6A7FqPcG0E0T07J4eXJoyfhlMbUuCAIN7va1TxJo9a5U065xLfZT2hRyJQYuTzB8cYqhi1NMjSTJXMX5tS0bRZXx+FyEYj5au2vp3NxIa08dHp92vefCTUBRJBAnMMwpNHUbAhIdsTAf3bOJfLnEbe21RL1+bLtSZaAo2OhVhs3qNu1Xw7ZtirkyU2PzDF2YYujCJJPDSdILOYr5MkbFBMF5YXj8LsJRPw1tMdr76mnpriVWF0TVVh5HEgTubmnjySuXeHlkmHvb2rmjqYXvXzrPsakJdtc3IIsST165QLJYIO5xZjuqJFHv9/PV0yd4qLObttCNp7CL12BbtqMkMZ/n6CsXsddhg5VLOoV8eY2/3xiSLNG1yVF7UBQJRZWJtng5kz5NzBslGR5jKjVIOBZG0YokXaO43Y2MFUbxbfJjWAZpc55J3aQcLqP7M4R8UQKKQ94PVGVI65puXgjgpoJTFAQ+c+tuREHk8ZHznE5Oc2FhFgE4mZyi3uvnpYkh9te3Uu8L8NLEINuitZyZn+ajXduu62gfzaUZz6c5UN/OEyMXaAssf4Hz5SLnF2Z5sLmbbw+cvukLuxqmYTE5Mscbz53j0HNnGb40TTaVxzJvTM17ESdbW1Mfwhdwv63gfPjjt3HnwwGKlTfRjVFkKY4it+BTzuIW57HtIqVyO6LoRzeuoMitVIwBRMGN1/1OBK6n9i2OjpMjSQ49d443XzjH0IWpDV8fOJ0j4Zifri2N3PbAZnbc3k0o5l8y3t1Sk6AzEsWwLLyKI435U9t2UtYN3JLM0IVpHuzoRLcsR+oTAVEQ+HD/FgqGvsIC7+rzBqeGXMyXSc/nmZ1YYGxglpHL04wPzjIzkWL2Gu3dq6FXTL74J4/j9d9cG99auPe9u3j447fhkTzYOKWUVGWBRk8TU6UpoloURVC4mL1AnbuemdI0RbNIl7+b4fwwycocASXAfCVJYA3pnZvBzY2cgoBHdVTNt0QTfPvKGXbVNOCWnURRsy9EjdvHM6OXqfcGMC0LCwgoGlGX57rSiSA4U6RFRT/TttHNqoX4Es3qrbea2bbN/GyWp77+Bs8+doTJkeR1BPuNoFLSGRuYvfGGN8C++zYBATSlH9suYZiTiGIA3RhFFLwIohfTmseycxhmEgQFVW7HMCex7SJco4tk2zbTY/M8/a3DvPDdY0yNzb+l6zN0k9nJFLOTKd588TwdffU8+OFbuf3BLXj9jnK/R1kZYIJhc+7gIA0tMQq5EqVMmYHzk8QSQVLzOTp66whFfASvWg8vliemR+eZHE0ydmWWkUvTTI4kSc6kyaUcGuJG+cy2bTN6ef36581g6y3OjCxrZMgZWebKsxTNIjOlaWq0GgQEZFHGJblIlufwyF78SgCf7COiRvDJPkzbIKK+PbmcRbwlxXdHTMlpD+oP1xB1Od0q08UcLf4wPeE4k/ks2+P1eGSF7nDsOmU0gEZfkCZfkINToxyob2e2mGe2lKdo6rQGwvSF4xyZHacrFFtDK3ZtWKbFmSNDfOnPnuDc0SFM462ZG/24IYoBFBqw7DJIoBtDyFIdshQHJMDCMOdQlXZkMYEoRQAVgeV1rW07nSevP32Gr//NcwxdnNzwKHkj6GWD88dHGDg3yaHnzvKxX76Ptr66KoF/GaWiTqFQIZspOl0yZQPLtMlli5iriFcv4kdfO8i3PvcChVz57Wnw/m+ET/ZzR2w/NjYZI02Pvw9N1LCxl7K6pm0iVp9pR7YltPR7+Sbd7dbCW9pLxTR5dWqYTZEECY8fWRS5/yoib2tg5fri7obVrRkUUeJAw8r148/071n6uaZhOftm2zZzmTyKLBH0rL+wNk2L1548xRf++HGmx26c6fs/CVlKgLQs0+JSt2zgQ61LP9q2TTZV4NtfeIkffPlV8tmN2ufdHCplndefOs3YlRk+9ZvvZO89fSuaAAIhj+N2JkvoFQNFkQhFfaiajF4xca3Szoft6CBlFtZWaf//BUiChCRJ2LbNpsCWVe0qxTW4tNKPUXblLQWnLIrsjDfgU9QNt7+8XVi2zT+/fIytLXXcvXltHxbbtjn2ykU+90ffZ3YydcP9KqqML+jGF3BXM7FOkiGXLpLLFH9sb3enJPP2vjjbtknP5/nH//FDnvvO0bfdebMRjF6Z4a9//9tUSo9w57u2LZcGRGGpz1WtZqMXtZLX6uT5vw0Oy2z9xuiVCh/LrYE/Dryl4BQFAZ+kOpISIT+KJDKXLVAs6zREA4iCwFQqi0dT8blUplJZrkzNkymWCLg1ehtqiAccJYOybjCTzhH0uLgwMctMOkcs4GVTUwKfS3PWKaUKY8k0By+OEPK6uTTpKJ0lQn4C7pWlkfHBWf7hf/xw3cAUBIFEY5i9d/ex445u6ltieAOupZJHpWyQyxQZH5zlyEsXOPzieeYmU6to6F69U+chdXlUvAE3wbCXSE2AeH2Y2uYINfVhOje/dd8O27bJZ4r845/+kGe+fXhD03TNpRAIewmEvUvXZ5mWI5OykCc1n6eQK62TDXWQnM7w+f/2fVxelb13979lxfn/f4MjBTNGufQMun4KBA2//9eQpBoMYxTbziLLDgXSNKylNke9YlQb+9d/ib3lyXEyV+C/fPNZ/t0j++msjfK3Tx3kzMg0f/5TjxD0uPjT773EI7v76Wus4X989yXy5Qo+l8rkQhavpvKfP3QfDdEgY/Np/sPXnqIpGiRdLKHKMkMz89ze28qvvusOBAG++fopDl4a4cr0PN998yyvnBtCEOCn793DLV3LfNlK2eCxL7zE4LnJNc9b1RQOPLKd9//MARo7atY0oY3VBmntruWWe/sZvjDFN/7+eV594tSaI6nX7+Inf/UBNu9pJxD24vFraC4VSV4uPqcKRcq6gVYV5jIsi7lMnkTQ0bA1qv4s4ipv3krZ4Jufe5Fnv31k3cCUZJHG9hr23t3Llls6aGiN4w95UKo1TWwbw7Ao5svMTixw7tgIB585zcUTo+u2ZCWnM3zxjx8nlgjSsanhrY0OAmy9s5UFpYO41oBL8jFfmaRilah1tV7X4KzrJs8+dpjJ4eSa1/qO9+2mvuX6mvm1sHFEyiVRZNUzF6B3e8uGL8W2bQz9DKn072AaIwiCC7Dx+X4WqEGvHCWX/zyRyN9jVKI889VXCUR8tG9p4uCPjlPbEmffu3as2wD/loMz4HYarMfn09SGfIwn0+iWycR8GgGYSTsPXcTn5tcevoOwz4NbkRmcmec3vvQ4RwfHqY8EsCybsWSa1niYP/rIg/hcKk+duMjfPHWID+zbQltNhJ/Yt4Xbe1v4jS89zsfv3MFdm511qldbmSS5dGqMV588taaYmKxIPPKTt/Hhf/OOpSzkjSDLEu399fzif3ovXr+LJ//1jetEpgAKuTLDF6d58MO3LsmnWJbNdCZHxTCJ+b2MzKVoi0cwLIuJhQyCIHBkcJy9HY3E/V4OXRmlKRKkIxFdYcNgWTavPXWKx7/y2rrT7FhtkHd+bB/3vGcnsaqnyqqyJzgvk1htkN4dLbzjfbs4+MwZvvW5Fxm5PL3m/kcHZvja/3qGX/3jD+IPLRf0bdvGsHVM20ASZGRBoWwVl37W7QqmraOJHrbt68LoGaHJ002Du4PZ8jhn0ge5M74fy7aQxap8JSZG0eb0G1fWDk5JYv/D29lxe9fSeTiNEddjJp3jXw6e5OO3byeyiufrzaNMLvc5BFxEIl/AslJkMv9l6a+y3IFlzmEaY1hWmEKuRLgmyJHnzhCrj5BLF6iUdVzrSOC85eB0qwpNsRCD0/M0RoLIksTmplouTyWRJQlZEqkJOgXpxmiQTKHEZL6Ibjq1sFR+OZGhyhLv2NpFxOdQq3obahCAdKHsMI3cGoWyG0kU8brUVW+uZVq8+INj63Yv7Lyjmw/8/D0bDsxFCIJAIOzlo798H2MDs5w8eOW6bWzL5pUfneDAI9vZemsHgiBQMU2eO3OFXW3OdHZ0Pk084GV4LsXw3AIhj5tsqcylqSTZYpnpdI6GVVg/E8Nz/OtfP7cuU6ZzcyMf/7f34w26idYGMXQTSRKxLAtREjF1c6k7wjQsRNGhQtq2jcujcc+ju+jY1MDf/sF3OP3m4OoHseHNF8/z8g9P8NCHb3VGYkC3KxxKPoFX8lPnbkO3yowXBwCbzcHbGC9eIVmepMbVRKdvGz45BDisIb8SQqlynYfyZwiqUcpmCQuLWqGdG5m+LyojWrbNU6cu0d+YoDkaum47WRIJe93IVQ3itwvLSqMbZ/H7fglF3YWuH1txroLoA0QsO4OqKUTrQoxcmAAByoUybp/rhrIxbzk4RUGgqy7GmdFpEpNzxANetrTUcnpkCreqEA948bs15rJ5vvbycc6MTaMpMqIgMLGQWVG9VGSJkHc5YBzB6MUWs41hfjbLidcur/l3b8DNuz91B96AC9OykUSWmp9tbCzLdihntl3VQxKX3JcWic6RmgDv+eQdXDo1SjF/fTd7Nl3kue8cZdPutqXkT8Dtor0m4nhlmhYL+SKSKGBYNm01EXTLoikapGKY1XX4yn0auskPvvwqI+vo57T21PIrf/QTtPbUcubwEHOTac4eHcLjcyGKApIsksuUCEW9uDwauUyRUr6MpEhLxO76lig9W5v4xf/0Xv77r32V4YurH08vGzz+z6+z5+4+4lVqmm1b6FaZTZH7UUSVF2a+hSpqFM08GT2JKEhIgsxMaYxO37Y1ryOsJhgunMO0TXr9u9ioH1XFMJlKZfn+sfMosoQoCHhUhXC1PStVKFHUDe7b0rU021p0FBAFgWJFx+92UTEMdNMi7HWUDG3bJl+ukC2WcakKQbfrqvV2tZ9SWN03x7bzgIkgOEG49Y5eJgdnUFQZSZaINYTXFRSHtxGcgiDQXhPhpbMDnBqZors+RlddjOdOXebs2AxtNREkUeSbr5/ipXOD/D+P3kVHIophmfz6Pz6+cl9sXBd2NTjF6OkVLtDXomtzA95EgB88dxpJEmmqCzM7nyMUcJMrlKlUDJrqwqSzRfLFCnU1QVyaQr5Qpq0pir/68ti0p43mrto1O/NPvzlAcjpNojGCIonsaXdkQsu6QUcigltVGJpdQBQELk8n2dPWgEdziB1NkeB1fqQD5yZ4+Ycn10xGBcIePvHrD9K5qYF8rkQ+VyKVdPi6qkshGPEydHEKWZaolA0KuTKW5aw787kywYgXzaU47CegtaeOD/783fyv//CtVV9AAMOXpnnz+XM89JHl9jpFVBEFCQGBsFqDbVvUuduRBIXR/AXCaoKSladsFckaC0iCRMUqk9HnyRsZckaKsBrnbOYQsqDgV8JUbuAgsIgzY9N8/dBJTo1OoZsmIY+L3e2NvH+P0wX1xpVRnj59mflcgf/2oQdJBJ0S3bfeOM1UOsfYfJq2eJiybjC2kOHXHrydrtoYJ0Ym+drrJyjrBpYND27t5v4tXciSiCgEkeV2SsXHUdXdXE02tqwMxcJ3EMQAstSKUTF4+bE3CUR9KJpM++YmwjXr86LhbdrO10ecnsNLk3M8sK2L+rAf3bS4MDHLx+7cjmXZXJqco6suxrbWOmRRZHguRTK7fuPsahBFZ0StGMvaNVdj8Pwk5eLqD5MgwOY97VCVvxQEgTOXJtm5qYnLw7NUdINtvY2cH5hifDpNNOQlEQtw7vIkmiovNeaCYy7Uv7N1zeBMTqUZvjRNotF5OS0+CG5VYXNjLbZto0gS87kCNQEfYbebK5emaW2LE/C5VkzZTcPiue8cZX52dWErQYADj+xg1/4eBFHA49XYvb8HSRJp76tHkqrTzopBuajTtbkRRZWWfnfx1Bhev4uWrsTSFEsQBW65p58Xvn+cN58/t+pxTcPk5R+eYP/D2x2Ct6jS49+FJEgIgkh/YC+TpWEkQSKs1tAd2Ilt23jlALpVJqbVIwsKhlXBsHRavH2UrSJegvjkIDGtodoGuLHg3NyUIB7Yx/hChs/cvZf+hppqADnXf++mTlpiIf7wO8+v0CZaKDjdTh+/fQd/8oMX+Z1338XTpy9zfHiSeMDHF144zDu393BLRxMXJuf462dep6cuRkciCoILr++nSad+h/nkp5GkGixrjmz2LzHNCQzjCgH/byBKtc4yQhJo7W/A7XMTiPw4lBBugLDXjSJJzOVzNEaDBNwu/G6NM6PTNMdCSJJIV12MJ49f5NmTl1EVmedOXV5qIr4Z+FwarTVhHj9yHo+mOip7jXHqwgGnQXYdep2syLT11RHwaTQkgiAI1CeCjE+nqKsJOvKRfhdNdRHiUT+6btLSEGZiOkUiFljBThJEgbbeOkRJWJWVUynrjFyaYs9dvQiCQCFfZnxsnkjURyFfoVzWicX9VCoipYUSeNxMTaSIxnzkciUCATeZTJFY3M/0+LwTIGuMmpGaAA98YO/S9EiURLRr1jG2bdO9JYwgeBAEaSlZpqgy2/d1rrpfj9/FgYe3c+yVi2smoK6cnWDk0hT9u9qQBIm41sh8vogABD0umj29VXU7m1qtYynBZVoWXb7tSy9Xj7ys7j6Sv4AiatS721a/4DWgSBKu6pLJpch4tJVTTVkSUGX5uiWDiEBLLExjJEhN0EdLLEzc7yVfqTA2n+bS9ByHB32cHZ+hrBvMZvOML2ToSDiKEaq6h1D4sxTy/4yuH0MQvFQqh5HlDoLB/4LLdS+CICHKEE6EOHvoMrIq07OrHV/oxkmptxWcLlXm0b39ZAplwj4PsiTy7t19bGmupT7i1Dvff+sWdMPk+0fO4dVU7trUzoH+dqJ+5+TCXjc/sW/r0ggDEPK6ef+tW6i96ncuRebn7ruFr71ygm+8fgqfS6U27KMuHMAwzDVHF3DqfZF4gLqaILXx6nRCcJI4y65gEKs24Vq2TTpbxKUpdLTEV4zSgiAQTQRQFJmyeX3pwbZhZiKFZdlIkkChUGFkOMmVS9OUijqJuiDDQ3PVBl4Vd1VhHmB0OLmUpInF/Zw+NMDMOsTvzXvaaa66kK1td2FjmuNIUgvgwjRHEQQvoriS/3ntNfbtaCFSE2BmfPXj5zJFzhwZom9nK4IgYFo2L58fdEyVG2qwgULZUWnIlSokgj4kUWQylWFzYy0tsdB1s59GTxeNdK2pGbsR3BSJUXASRYvyO0slrGp+QJNl+utr8LmcYN/d3khP3XIrnyCIKMoWgqE/xLLS2HYRQZARhCCC4F7qP5YkkTve49hVFjJFZte4p9fibQWnJIo8uL3HuR7bpmyZ7OttYV9vC7IgolsmYZ+bT79jD9hg2haqLKFJy3SoWMDLp+/ZvWK/Ub+Hn7p75e+EagLqd99/N4bp1AOVahHX6WxYW25Q0RxygFBNNC1fwMqHYPFvIgIhv5vbdrWvEJpahOcGmbZcuuAU9iUYGpghny0hSiK6bqLrJpIoUC4baC4F3TBJpwqkFgo0Nkd44gcnuP+dW9ErBkdeXnvkkiSRPXf3gTiCrufQjYsocg82FQxjBLDQVOeBKJVfw+upw7IK5PL/gCiG8bjfjW3rlCtvIIoR3K57Ea5KbkQTARra4msGp23ZXDg+gl4xUDUFSRRoiATxuzTqwgGePnUJTZaRRIHmaIgLU7MkswV8Lo3JVJZSWSca8BL1eZaSLGtR4jaCxZE5Wyo7fccsJhbfWqDXhfyEPC7iAS+3dDiSLcWKjs+1WulDQhRDQOiq35lV9cU8uVSBs4cuUcyVKWSKJJpjtF2nbXw9fjwMXWC6mOPVySFUSSKkuWn2hRjMLKBbJgvlAk2+EAOZedyywj0NHYS05TafuUKBTKlEQyCwZLe3GgRBQJGk6zr/bcteV2NGrOr83AwkSVxT4fZGXe16ZbmrpndTA3UNYVRV5vDBK7S2xZEkkXNnxtmyvZlAwE3/5gZcLoVK2aC+IUws5md+JsOVs+NrHiMQ8dK5qQHIU668gWnOYVlpFLkTw7iEz/eziIIXJ92mY9slRDGGLHeiKluQpDoy2b9CkuqoVI6gKn3IV3F4VZdCY1ucY69cXPMcxgfnyGdKqHGnB7Q26OPi1BwN4QAuxcnYL+QLnJ+cpTMRobs2zkwmR0ssxOWJOZ49cZlowMP2tnpaa8JoyurSNRuBz6Wyo7Wef3jxMM+eucKu1gYe3NpNtlTmh8fPc3k6yUwmz5deOUpbPMI7t/UgiSKS6IzTiigtjaCSKFIT8PGR27bzlVeP8fjx80vH+OX7biPkdVdnOEXK5eeplF/HsjPXDduiFMYb+CUUNUisPkJLXwO5VJ6FmY2JY//YgnMin2GykCXm8lDvDfLSxCAdwSgnkhNENA/JUoEatw9JEMlUyiuC88p8kj999VXawxHu6+hgR10dYbd7w1lc4QbBZ1n2qsSBq6eCN/NQ6BVjXdc0WZGWRmGfz4XP58KyLPbs6yQS9aHrBrt97YSqjbgtbXFs22ZmKs2eWzuQZJGJ4eS6CufxuhCRRABJdGGakyhKHxX9FJp2K6IYQarq89h2CWwL2y4DIoKgYdtlp5lbkBAEDZe2H1G8XvI0cYMG4VQySyaVJxx31o2NkSD14QCpfAlVluitj3NlOsm25nqnTATVMpZAXchPb0MNPzxynr994iBd9TEe2NFDR23kLQWogMCD3Z3saqpHUiSao04uQZFEOhJRGiJBDvQ5a1lRMjGFMg/v6AHRwO+V+Zm7d6GpNndtakYWZcp2kXs3tdNVH2R8Po0syojuBWR18bu3KBT+mVz2s4hirDpyXqNMaReRfDZezU3PrjZSs1kEQaC5Z2OmzxsOzuErM5RLOt2bVueHdgajLJSLBFSN7mCMqUKW/nANHllhppij1R/Bqsree5WVC/bttXX8p7vv4aWhIb504jj/eOwoexoaubu9jc5IFPcaYmBLF6FIq3qMLKJS0inmru/eME2L86fH6d3ccFNk7WyqsC5Txx/0IFzTYiWKIrHqQ6xpjmPa1RAEgUS1bggwcnmKUmntqXptUwSXW0UQVFR1B4rch4CMJCZQ5GVx54p+DhuDin4KlxhHVbZTrryBJCVwu99Npfxmda10fb3OcXJjzTJOqVBZQfoQBAFJEAh5XRzobUOWRLY01SJcRUmUJQHTsnj13BAnhyZpiAT51XffSbGi8+q5QdoT4ZsOTtu2GRia5fjpEe65oxdJEh2dWd0kky7Sl3BmK+lMAbcfzuYPMlCcoMnfSbIyzYX8AnXRFi7kL9Lo72C2NMjkQoaQGifJNJ6YiwZ3G6fTA2SMKF7Zi21nKBa+i8v1LvyBX0UQAlwbnAICCC4M3eTZf3kNRVNQVJmunW1Er/qu18KGg/PYoQGSc9k1gzOguri3cTn7966WXgC2azfuUtdkmU01NfTH43xk61aOT07yjTOn+caZ02yqqeHBzi7ubm8nqGmrfnGyIhFb52LLJZ3psYWl5MX8XJZzp8dpaoly4ewE88kcHV0JZqYzmIaF169RLum43SqJuhChyLKCg23bTI4k0fXV0/yCAPH60ApyeKoyw0x5hC7frg09eLZtM3J5Zt3sRjjudzSPKiCLD2BbIEvNmAaIwj4qZSdZJbAZl+oIaRs6QAuq3IJVfbcosrP2cRwCVia4RElEEEXsNRq49YpBLn09a8mZLlZ/XuN6YwEv779tC0GPG7FK8lhU/b9Z2DbkixXcbiev8PyrF9ja38jEVIrLQ7OoVb2kfL5MuEbG3S7glQLMlMcpmQUy+gJhtYa4VkdMreVc+jBuyYthVwgpERTRqXGH1TghJQYI2HYJ2y7gct2HKK7ub7N0fpYjp7r1zl7cXg3PBpUbbnpaWyxU0HUDr1dDlERsGwr5Em63hiQ7b6xivozHq1EsVhyBK8W56YZuUirpeL3aEvXLubnOU5gplzkxNcnTV66QKpW4t72DxkCAH1y8wBvjY/z2nfsJuq7v5RQEgfa+egRRWLXDwtBNLpwYZf/D25EkgUrFeaOWSzqaSyYa83Hq+Ahuj4rHo3Hu1BjFajY1WhO4fl/HR9YMHEVTaO5MLGXqAHJGirHCBbp8u9acDl+9vaGbzK1DqAB44/lzPxZ1hvWwMJd1RMHWgGVaa9aWoZok1A1GZlIUyzrdjXEk0cnszqRz/PDIeceEV1P5wB1baa+9sZvAahBFgXjEh2GYuF0Kfp+LjpY4R0+NoBsm8aifufksoiTi82kUzBIoUOtqZrY8QVCJEFbiTh+nINMf3E2qkiSqJbCqHjAuyU2Nq4GFyixutwdB8CCKEUzrxt+BKIm4vBoHHz+G4lLYdGsXndtuTLK/qeAcGZjli3/xNAvJPO3dCd77sVuxLJvP/flTfPCTd9DQEiU5m+Urf/cCn/qle3nxqdO43QoPvGcnNjZHD13h1JFhPvELd6+gLi0Ui/zg4gWeHxzEtGzubGnhp3buoCUYQhZFHunt5XeffppLySS7G64fuQVBoGtLI/6ge81G3hMHLzM/kyFe59RfXS6ZdKpALB4gFPE5bVTZEtlMkZb2OOlUgUrZwOdf+TKYHE5y/vjwmvcoUhOgucv5Ugdyx5koOpRCGxvTNriYfZPZ8ihuyc+W0H6mioNIgkyTp5fZ8gjzlSmahK2k53NrHmPxPNYihP+fgo3TObIW0vkSf/f91zl2eQLTsvgfP/cwC9kiL58aQFRF3r9vCyGvG0EAv3vtZclGEAl7cbsVNFVmS18DggCbeurx+zT8Xhebe+uZns0QjWuUVYkmjzPLi2qJ6/aVcDWRcDVd93uffDWrx4/H8xMUCl9HkupQlH4Ewc3Kqa2AIHgcUsjWZkbOT2BbNq51lmBX46aCM58t8e5fvR9RFPi7P32C428MsnlHC7NTGSrVaZ6hm8xMprEsi55N9Xzjn15j3919eDwqrz1/nv5tTdc1HV+aT3J8aooPbd7C3sYGwq6V2qJRt5vNiZp19Vkb2+N0bGpcM7s4dnmGl35wnEc/vZ9w1Mve27vQXAq2ZSNKIom6IJbp8GoVRcKyLGy7qphXhV4xePaxw8yuM6pt2tVKrDZIRk9yNvMat0YfYTB/irJeqE6NEnjlEKdSLzJeuIRPDnE6/TJ17nYuZY9S625FLzk0u/8bYK8zsr544grpfInf+eg9fP6Hh7BtCHhdnB+Zoac9wbnRGeqjARRJoqNuZSfOzcLlUpYc4eIxPzYgKyLd7cuzmEDAjWlbSEJoiWW26DEjieJSCWZROH2xhW/xvGzbxlhqO7OxrHl04wIL8z+HJNUhCCtNf0WphlDojzEqEY6/4Jj+KprC9PAcjZ21N7ymmwrOjp5aGlucqUdHby2Dl6bZtH01h15netbWVYvbo3D+5CgNzVFmp9Js/fSd183Pt9fWsa22dkX9c8VJiiI/u2v3dQJhV8Pt1Tjw8DZOvzGwqkKAYZh8959eoWNTA1tv7cTrW0Xq5KrdX9vnaVkWbzx/jie/8eaaAlQen8aBR3YgKxKFUgZN9BDXmiiZBfJGiryR5lzmED45RMUuodtlopozExgrXCBjJNnpegfZjPF/ROXgfzdGZlLs7W2moz669GJVZQnLhkTIx3Qqy3yuUO1wCuJS3nrxYCKb4bsXzmFUSRwd4Qivjo7waE8ftzQ0cmZ2hu9ePMd8sYhXUXm0p48dtXVMZrN8/ewp6v0Bjk9POsSZ3k1sr63juxfPIwKP9vYjCgLJYoG/P3qYT2zdToPfhyB4cLveteY5iWIIAdWZdjdG8YU8XDo2tGERtpu6G1eXI0zDQnMtytHbS2s9vWr1DuByK9xyZw+HXr5IU2uM5vY4scT1koGqJDFXKHA0OUnpKrKzKAhsSSSIeq430L0WgiCw9+5+ntl2hNNvDqy6zexEir/6/cf41G88xO4DvVUV9htrwZaLOq8/fZp//NMfkU6uMd0UYM9dfWze04YgCLglH2WrwHxlktnyKKZtMl+ZpGzl2el/B3Nlx2JAFhRavJs4svAULZ5NuCQfKWN+1dLP/22oCfu4ND7Hrf3N2NgYpsnFsVmCXhe397UyNp/GMC1cirxEKHmryOs6zwxe4Wd27Oarp0+Qq1TojkZ57MJZdtfXIwoCe+oaqPP7eW10lL8+fIjPPvAu8nqF7108z4OdXXywfwsHx0f5X4cP8uf3vZN6n58vHD/CXa3thF0ujk5OMpJJE3K5ARGP9ydZn5MkACKyCjvu6ic1m8Ht1WjYwKgJNxmcl85NcvrYCJIkcOncJB/8qTvQNAWPV+PE4SFkWeK1F84vifwKgsDW3a28+ORphi7P8IlfvGdV5YG5QoHfe/YZ5goFUqUSflUlr+vU+nz8wT33EPVsrDk2FPPx/p89wPClKbKp1deeY1dm+Oz/83VuubefA4/soK23Dn/IgyyLS0kqu9q1kU0VuHJ2nOe/e5Q3nz+37lSzsS3OT/zc3bg8TkkiqMTo9u/hVPplvHKIOncHNVozE8plzqRfIe5qIaA4HM06dztH5p+kxdu/dPwbSYf834ADWzv4s2+8yH/+0jMMTc/zl4+9Sq5U5pP37+aJYxcolHXKuoEiS3TVx3Cr6+v13AgRt4fbGpt5bWyELTUJWkNhjk5OoFsWHeEIsigyk8/jVVXmS0WKupOd9msaP9G3mY5whKCm8dzgANlKhc01CRRR5PjUJHc0t/Di8CB3NrXgVRaFt6UqGSGLaU5gW1kQXEhSolo3dsgqpmFy8EfHGTk3zt4HtzF8bpwtt/fc8Ho2HJyxRIC7HtjMGy9fZH4uy/77NrFpWzOKKvHej+3jye8eY+jSNB29dey+rWtprRYMe+noreXE4SE617BUv5hMYlgWn33onXzt5El2NdQT93j5+unT+FVHR6hk6ozkk8RdAVRRxisv1+WW+bECO+/s4eGP38Y3/u75NWuRuUyRZx87wmtPnaKmPkyiKUo0EXAU3REo5sskp9NMjiSZGV+gVFjfiTgU8/HJf/cQ7VddnyhI9Pj30u3fXdXmdepet0bfDUsSiwJFM8dE8QpxVzMR1fm8IAorstmroaYhjNd/c/L+P3YITk13LSTCPn7zQ3dx6NwIIzMpfG6V3T1NtNVGuDyT5J4tHYwl02SLZfKlCqGbsMdbDYooOmUcQUSTZUTBGdcqhsmXTx3n0nySvlicgq5jWsuKyG5ZIVBVJ1t0fbOx8asq+1taeXboCs3BIKOZND+zY/fSd2zbBuXS8+TzX0DXL2DbZQRBRpLqcLvfi8f7UcDv0EtzRerba0jNZm7Yx7mIDQfnrQd6nIW15SygFUVequX1b2uiu79+KZliGtZS0scyLbLpIvsO9OD1rT41LRk6NV4ftT4ffk3DsCz6a2oQBBhcWCDsdvH9sROcT09yR00XC5U8t8W7qJgWHlkh7l4myCuqxPt+ej/p+TxPf/MN9Mra2cRivsLwpWmGL60tzXEjRBMBPvWb7+SWd/RTRkeznbe/aVsoorxkc78YateSuufKY8yVx9gRuhepqvQmy9KSGNRqkCSRD/3CPey7f/Oa2/yfgsutkp7PY9s2Hp+2wkumYpjEgl4e3te/ooSUzpfob0oQC3p59fwwFcN429naRVz7ShOAuWKB54cG+I/772ZrTS1vTozz0sjQys+s8i4UBIE7mlp4euAK3794npZQiIaAsyyzbZty+XnS6f+IovTi8/8SohjDtvPolRPk81/EslL4A7+Gqqm0bWri9cePEYz5ufdD+zZ0LRsOziUGjQTyNaxTQRBWvA1E1VEUOHdqlHMnRpkcW+DRj67uewlQ4/WxUCpS0HWaQ0GeHxwk4nYznskiiQIVyySjF7mnro+KZTBbynF6YYqJXJb99e1cbfkjCAK+oIdP/eZDBCNefvCV19ac4r4diJJA56ZGPvHrD9K8O8G5/Ai6ZRBQPETUAIZtYtqW05UgiJSsCoog45E1wqqfsqkzWUrilWqpd4cpWhKFYhK/7EZ1yasaJi3CxmEchWP+Nbe5FoVs0am3raNZ81ZQKla4cGKEfKZIe389LV3L66kn3jhPV0OcvpaapczoueFpvvPqaX76oVso6Qa39bZQG/b/2ILzWtiAR1EIu9w8M3CFgYUFXh+7saXgIur9Afpicb538Tz/Yf/dKEuZ2wL53JfQXAcIBH4PQfAuj6ieD1EsPkY2+xe4PY8iCl1k5nMkmmNIskhmIU+s4caq8D82bu1qSCXzmKbFT/7C3SuoadeiPRzm0d4+ZFFkb0MjT1+5wu8/9xzb6+rojETRJJmEK8Azk2exbZu+YB3t/ij6OkkTX8DNh3/pXrq2NPLYF1/mwvFhKuW3nwFdbBm76907eNfH9pFojDBXTlMwykiCSLKcpWIZlE2dgllCFmQWKlm8sqs6HXcTVv1kjQITxSQFo0xaz+FXPKQqefqDLbSrtetOWW3LJj2fuyn3rjeeOkkoHmD7/r4bb3wTcLlV2nrrOH9sGM81Dmtet8bnHj/IL77nNtrqohw8O8yXnj7MPTu6eO3CECOzaTTFkRX52F07l1qz3gqibjcPdXajShIHWlppDobwqxoPdnQRdXv4rdvu5KWRIVKlEh/fso2RTBpP1fflvb39S34uQU3jfb39y9NcQWB7bR1HJsfZnqi9akqbwTQn8Ho/viIwAQRBRlX3IfDXmOYEiJ1MDszgDXrQPOqG9Yv/twWnIAjsu6t3Q9tats2WRAIBiHo8/Oe772GhVCLq8eCSJXTL5K7aXroCCUzboskTIadXkEVxya15teOrmsIt9/YTCHv5q//4bQbPry2ZeaNr8fg1Glrj7Nrfw+0PbqGlK+EIZgkCIdVHoyeGS1KRBYmSWaFkVmiSnTH9VLqEKAg0eOJoonPL/bKb/kALumVQNCu4JJWKpRNQvKgoS2Ty1WDbNnPT6TWD07ZtUrNZLh4bxDJtNt3aSSlfYTg5QT5TpL6thta+BmbG57lycgRFlem/pROjYjB2ZZrcQh6330Xfng5Mw+TCkUEWZtJobpVtd/ZSLupcODqI5lbp29NOMV8mny0tUQYXcWBrO4VShb//wUF6m2t4/ewwH7p7O7dvbuVfXj7BJ+7eid+t8fVXT5LOF99WcMY8Xt7T47x47m1bFh1/pNt5BrujMbqjyxKaWxPOCO9TVT60aVl1P+Ry8+FNWwHnuSybBkcnJzjQ0lbN0i5CRhBkpxtlFTgaQobTaGA59hmyKiPL0oZ1f992cNq2zXwlyUx5ijpXPaEbmLjYts1cZYa58gyt3g7ckodLySR/++abaLJEvT9AZzRCeziCKkmcSs1weH6QTKWEXF3se2SVdzVsp2yaS9OM1WDoJm88f46vfPZJhi5cL1gVTQRpaItRzJcpFSqYhoUgOsoJLo+KP+gh0RimpbuW9v56Gttq8Ic8191cRZRp9MSv2//i9W4PdyIiEFSW37CapKJJa4lD2TfUYp0cTlIpG7jc1++jXKzw3b97hsbOWsKJYLUrx2RiYIbalhg/+OILfPLfP0opX0LRZC4cHqSQLVHbGuM7f/sMD3zsDl79/lHcXhcLM2lGLk7iC3q4cmqUzbd184MvPE9DR4Khs2PMT6XYcfcmYrVB/CGPc6yriAn37erGNC3+6anD/MK7b2P/1naqvrR87aXjeF0q58dnnNpkbZRbeppX1e29Wdi2TcU0SeYLKJJE1OOhZBiUDYOSYRDxOGqO2XKZimHiVmT813C3f3DpPC8ND1EwdP797XetWJaKYhBF3Uk+/2UkqQVF6al2/JhY5hT53N8jihFkuRNskZ5d7eRSead7aoPX97aD07ANXpl7jmZv24YdwWRB4XT6OH4lSKO7mS2JBP/5nnuYyuUYWFjg9Mw03zl3jnylwm/vv5MPtOzh+2MnuL+uH1WSeXriLBXTwCsrayr06RWDp7/5Jl/+7JOk5lbWJkVJZMftXdz/wb1s3tPG1OgCwYiXbLqAP+hhcjSJx+uitbuWbKZAIVd2Eh8LeXxBN7lMifHhJOGoj3hdkNnJNKWSTrlYoak9juZa9rgUBIGIuvG14eJnWntq15RCAZganSebKqwanNmFPKV8mTsf3b1k7CuKItv397Lz7k2cevUihWyJ2fEFxi9Pk13Ik05mqW2N0drXwK57NzE1MkdmPueUAqqymnVtNVSKOpeOD1XrvxVcHo1ivkww4kNRZc6NTPOvzx/HrJaChCqpXZElvvnSSV46OUDE7+bWza1kS2Vs26Y27CfkdRGrtpX9uDCQnOfI2ATj6Szv2dzLVDbHG8NjRLxuApqL/e0t/N3rb9IZizJXKPCxnVuJeZebHDbHE9R4fLSFwtR4vdfMUhR83p8hlf5dFuZ/HkluRBQC2HYJ0xxHEFwEgr+PKMYBgR1399/0+b+t4NStCuezZ5gsjdHoaUYWFEpmkcH8ZUzbpM3biVtyM14cJVmZo87VQI1WS1iN4FcCS71IJcNgKJXiYnKOK/MLpEpFEj4fDf4ATYEQYdWNZVsM5uZQJZm5co68USGkubFWeSFYlsVLj5/gn/7fH13HtZUkkbvfs5NP/uZDnHxjkKFLM7z4xEkOPLiV6YmFpZpyIT9DqaQzcGGSUqFCLlvC5Va49907eP25c8QSAQ69cJ77Ht3Jk98+gi/gNOCODc1x1zu3vp3bCkBzZwJfwENmYXUxtLmpNGNXZojVBq+b2ro8GpZlM3pxklA8gDfocSzQFxk4ApQKZV77wTEe+PgdK0j3sipXFSOcfSaaYxx57gzNPXX07elA0RTqWmvYde9mx+Yh6MHGeVmUikGCXhdb2uuuI/hvaVtOFLlUhfPjM1QME0UScWsqt/Y0v+1SyrUIuly0hENMZXOMptJYtk13TYx7Otv5m9cOsb2hFk2Wef/Wfr558gxXkgsrgrM97MzgVoMgCEhyO+Hw/6RUfIJK5U0sawFRDCLKtyCrD6Fp/QhvQ93hbQWngEhADuKWPETUGIqocHjhdQzbRBYkDiZfpsvfx5GFg3R4u3h57lnuTzxCQFkpC3h8cpJ//+wz9Mbi3NXWyo66OpqCQbyKuvTgvKthK4fnhzDLFu+o7cclaoiCsMSDXIRt2wxfnOarf/HUqiT4bbd18lO/9U7CcT+RuJ/zJ0eJJ4JcOjtOoj7MlXMTfOhnDzA3lebgC+fRXAp925sZGZhF02QmR5JcPDWGorSQXsgzN51BlkX27nemb0devbR0rKKuc256FmOVxJUqS/TWxNekrCWaIjS0xdYMznKxwtFXLrLtts7rgtMX8nDPB2/l8LOnkWSJ/e/dQ6I5ij/kRRQFWvsaCUR87LirjxMvn8cXdpNojuAPeWnsdIjgtS0xQjE/M2NJSvkyF48NMXx+gkd+5h7u+9jtHH3uDLZtc+tD20ESGR+apaYhTFNHDR84EFr1nBdhmBb/8vJxPJpKxOdGU+R1edNvBYZl8Y0Tp9lSV4siOt1Ttu3ImOQqFQSc3tOirpOrVCjpBqokYdk2qXzRyTOoCpliCZ9Lo1jRsW0bTZEpG47UjFdVWSj48GkfRRI/gK4XCXq8JHM6ZsnC5xFYyBWWBMhShSI+l4Zrg4oPbys4ZVGmxlWLXw5Q72pEERVGCoO4RDeKqCILMlOlcZLlWdySG9M2KVslYGVwbkok+O0793MxOcexySkOT0wQ93jpika4u62diNuNKsk0eiKYtgWCYzOorGK3ZpoWT379EBMj13dsuL0aj37qTsJx/1LWdWxojjsf2MyrT59h2952JoaTjA3NkZzJEIx4KZd0RFF0ZCYFAZdHI14XpL23ju7NjbR01nDq8CCSLF3HuZ3LF/izF15d0qK1sSnoOoWKzs7Gev7gwXvXDE6v38XWWzs5f2x4zWbnwy+e5z2fvOO6XlZBEOjZ1UbXDof3LIoS8YaQY18vwD0f2Isgitz+yE5sy6JsJZkrHaXW109tawzbtthxVx+mYfHqD45y1/v3Eoz5efIrr1DMl+jY0kT7pkZswKhKbEqSuCS7uQjbtknnSyQzhRXrUEEQyBTKNMWCeDQVda0+TuFGZAx7TZ6zKAh0xqIMJOfxuzRiXg+zuTwD8/PkTle4taUJt6pS0HW+e/ocLkWmKxZlYiHD8aEJuutiTC5kKRsGxYpOplhGNy2Cbg3Ldl6uO1rrOTc+g25aFCsVPJpKTaCEIouUqwSYVy8MIwjQW1/DufEZ7uhppSGyMdfrH2u2VkCk2dOGZVvEtBr8chCwmdTGaPa0ISIRkEPMV5Jk9QxzlVmiWg1BTeP25mZ6YjEGFuY5MTXFsclJnh8coC0cwaepfHP4MGXTwLQtNEmhwx9HXYUIvzCT5ejLF1elPNY2Reja0rj0IETifrbtbadrUwOpZB456uLOd27h5MEBVE3mtnv6Gbw4RaTGsWAXZZHahjDtt7Vw8fwE4ZCX1q4EnX31eH0apqnQ1r08fav1+/iTRx5YWhfbttOz+t1TZwm53YTca5dLRFFkz129/OhrB9ccPccGZnn1ydM88onbrjO3zRujJEsnUUQfiuilbC4QVLvJ6UMA+NUOUqWzWBjUeQ9gCwY2JsnyUXQzQ9S1E00Jc+tD2zn7xmVEQeS2h3cQjgecaW9VHM2WJWRZxBdwUyquzNYOTM7zZ994kUyhRKZQxudSyRUrbOuspyER5MpUEreq4NEUehrjeK6p7YqisG7ZwbJsCqsoXEBVfK63y+kiERzWz7OXBriluYm7O9tQJInpbI6GYICP7NiKKstIgsBYMo1XU4kHfFycnKMxEuT48CThamtbulAi4nVs6S9PJUkXSxQrhuOeVx9nJJmi1uUnmS2QyivkSmXqQgHCXjdeTWV8If2/Nzgvp5MIQEcwiiIobAvt5nJ6gfZAjN3hfQzkL1Eyi8T+P+39d3Rk2XXeDf/OzZUTCjk20Gh0DtNxck4cDmc4DBKDLMlKVLQsyZLs97Nl+ZOVgynJkizLsgJlihTTMMxwck49M51zQM6xctWN3x+3UAC6AXQPSfnV+tY8a/XqbuDWrRvOPmefvffzbK2eOt2PYk5VJoipCSQhmDNnaAt24ng2JafIudkF/uzwW7ieR8II0JmI84O7dtORiNMei+N4DoasclP9RmbKOSZKGcqORUi5OnE9OTrH7GRm1euOxIO1AAlAKGxwzyN7ALj30T08c+ESEV2n6/ZO5kolzuTmUFsDTGFhNfoKczOSSaIrwabtbUzkcgyXcogNIWYx6c/Ms3v3UgcuVZZpjl39IgxF4beefZGFUpl0OHTV7xexYXMz2/Z38dq3T676e9ty+ObnXmPnoW46ehtXrD5lewbXq6BKDRSsUWLaRqZLb2EoaQQy2cpFTDdHUGlEqg4D2ysxXTqMIacoO1MYSpLe3Z307u5c8xolWaKtpwEhiatysy+fuExDIsJPP3ozf/Ptw3zirj2c7B/HtBzu3ruRN84PUaxY7OluXbUIQZblaknl6nBsl6nR+TVTSnK1nA/8VbzdCGGEdGQP8Dyihs4tHe0oSLXtUWc6QdmymMkV2LuhleHZDLdvWWzw7DE2n2N0LsOWlgYS4QDj8wZhw1/9k+EgmiKTLVUI6iotySgHetrRVN+tbUpEaE3GAIti/vNo+k0o6tq9Zq/LOD3Po2hbDGTnCaoqY/ks0+UCZcdmQzRJQmnlcnEOgcB2JDSvBde1CMlJsqbJfFGnwdhMS8ifdXvCm+gJLyv8jRf5N4dupD4UIqLrtfTI4gN3PIk9yQ6aA3HemL6ELimrGib4nNO1CMALs3kKuVJNJvNK6IpC3jSZKRYpmCYN4TCO5zKSyVIXDFKwLHLlCpqiMFsokK+YFEy/QN92HQqmucIFXUv1wHQc8hUTa1mO1vU8PPxqokUYQY17P7Kfo69eWLPofuTyNP/wx0/zU7/+GNFEcEVwx3HLCBQMJU3eGiJpbKdgjeBiEVQaydtDCCFTsMcxnQyms0BS347rWQSUq0nIq8GxHOanc2RmC+iGVuufApDJl9nZ3URnQ4KgrlEXC3HXno38/hdfpOzZtKXjNMYjvHTqMi3JKPHwyoCQkAR1DWu3LfA8jwsnR2rynIs/y87lqRRNXztY8VdN23LwJgtE2gymR+aI1UXIzeWJZ1wujPezYXsbgZCBoSrs6lgS4GqMr4y0xwIG7ak4rSn/upoTKyffK7uuL1fwr6/qMLtugUrlOWS1G4Xv0jhdz+MbA2eJaTqt4Ti25zKSz1CxbWZLRbamGnh1fJCOSIKRfIbnRy7RFokzWsiyOVHPRDHHsyOX+LGt+68S9wKfTZAMrF1ALSFI6WH689P0xZoIyBprdFlE09U1k7xjAzM88fk3+fC/vo1g+Go9ot0tTTWCreN66IqMaTuULBtDVdjZ3IgmK5iOja4oWI6DoSjIki/9XxcMEV2ma5opV/jWmXOUl+kNWY7DG4PDpMMhYgGDuXKRodwCmqyQMyu0R+KUbIuOSBxZkthxsJsbbu3j5W8dW/WePM/j9adOEUuG+fTP31drzed4FVQpguku0Bi8BQ8XCZWo5isAZM2LaFKMijNH0thBV/QxJKEQVttwPRtplY7Onufh2C7FfLkqYiZQdYV4KoxtOUQTK72AuliIibkciixhaArHLo3RUhcjW6iQLVY4uKmDaEBnYGqeQsW82jgFtGxIryk/A3Dm3UHGB2fpWNxOePDOMydwHRdFU9ADGrn5Aon6GOF4CLNsMT4wjeu4XDgyQDAawLYcJvqn6drmqx+4bgHPzSCEgeNOIYkwktyEEDKJcJB4yMS2B8CrVH8eqU2KnlfCdcYAkOUWEEbtd647g+vOIonvYTsG23OZLhV4oKOXiKozP15kd7qZhBZgIDdPXDeIVtkj4LE5WU9vvI53pkY5Nz9N3jJZqJQwXYe1nbi1UXFtvjFyjLZQEkNWUYS9agoF/H1kKBKgUrpajd22HL70P16g/+w4tz20i66+JqLxIJqhIssSmvD/ACABrh/02tfUjCwEmiyDgKBmIMkCoa1cgVOhlRNMybJ4Z3iMXHlp1ZMkQW+6jo/v3k5QVTkzNc3l7Cyb4r48Zn92jjpj6SkZQY3HfvQ2zh0dWlP93bYdnvzHN8nMFfjEz9xNe08D6cA+HK+MLAwksfSaBdUyNa2XkNqGJDQk1BX3IS8LtPmDyiM7V+DCyRFee+okiizx4/+fD6FWlQey8wWGL03S1ddMahlf98Dmdk4PTiIJwU3bOvmzx1/Hdlxu2tZJX1c9f/vcOyiyRCoSJBW9emQIIejobSQYNiis0f5wenyBb/7D6/zrX/6Av2URsOXgRmzTRtVVbNPmzFsXcWzHv14B0UQIRVVo29SEETIwgvqKva1tnSKf+yyy3IxHBdeZIBj6QXTjfjyvSDH/F9j2eUAFJMLRn0eWO3DdaQq5/4brZgAHSUoTjvwbkBJY1hEKuc8iRNQ3+qoBr4frMk5VkumMxHly8DztkTiy8H10RZLQZYXLmTnGCzlOz08RVXUCsooiJDRZZrKUJ6CoRDS9ttp5nsdUruArn9sOUrUoOmroRFZR1JaEwMPjUm6KkKITVQPsSq5eV5tujtPV18TcGsK9lbLF60+d5O0XzhJNBImlwgRDOop6dS+Nq65DllA0mVDYIJ4K09CapGVDmpauNKn6aFWvdukkDZEwv/mBe1cUZwhETfpfCMHmZJqGYIikESRTKRNQVIq2VWNJCCHo2dbKYz92G3/9O99ak75mWw6vPHGcy2fGeOD7DnDjfdtJN8cRa+j5CiGjrjKD+7O/Lyc6P52l/9wEJ9+6zIm3LjFyeZpy0WT/HZtX3JOmqxhB/ar9YXdzig1NKYSAA5s7qE9EKJZNuptTSJJgS1uD33IvHEBbg2zd0pmmpbOO88dXL1b3XI9nvvw2jW1JHvz+g+gBjcaOpWotp7qCqrpCon5lTriBtaqwHBxnkHDkZ1G1HZSK/0S59A10424s8w0s6xTR+P8XIcLkc39EqfiPhCO/SLn0VQCi8f+C59lkF/49lcpzGIGHKRX+HlXbTyj0Q9j2eRbm31rju5dwXcYpCcG97b1MlwroskxQ8WdaSQjaIjFMx+ETvbvQZZmo5vvcqiSTNIIIAVmzwh0tGwhXXVrTdnhrYATbdbBdl3ggwNhClr0dLWxtvnqvsxjt7I02ElIWjX/1lxkI6dz3sf2cfmeAUmFtcrRl2sxOZpldR7j5ms9FEmiGXwfbu6ONQ/dsY9eNPUQTodrzWWvQLSKkajVXf/HvBCvdO1mWuOexfYwNzvLNv3sNew11e8/zGO2f5q9/91s88fk32bavix0Hu2nrbiCWChEI6T67SACeH+20Lb8DWSFfJjObZ3J4jsGLkwycm2BsYIb5mRzmOu3oAcqlCrZpXyXs7bhu9TlIKLJgY0tdVZHP4cl3z3LH9h7CAY0n3jnHwU3t1K2yekbiAfbdsZkLJ0fWdG1L+Qp//0dPMXBugvs+vp/O3kYCIb2aApNoaF8ywkVPwLEcXw2yZFLMl8lnSzS1p4hVO4DJcguKuhkhDGS5Hc97DjwLyzqJY/eTz/4+IHCcIb8KyDOxzKM4zji5zH8FPFxnBMcewvNKOM4ogeBHEVIAWdngu7zXwPW5tW4JjyLNoSS2m8PxCuhyHY5XQkgmISVGRHOwnAxC5NGkJJLwo5umM0djsA5JKHieh+XmsLwSPfVJLNvh4vQcIV3j4IY2GqOrl7kpwg8IZawSFceqraSrwZcr2cwHP30jX/vfr1C5xsD6brDYeHZRCe/1p07Su6OND/3gLey7vW9FZPi7hRHU+MRP300xV+a5r76DY6/NyHFsl9H+aUb7p3nmy28TDBtEYgGCkQC6oSIrEq7jYlsOlbLlD9BcmXLJwqxY71mFQdNVKmWLUqGygsb2jdfP0N2cYvuGpRaKc7kij792ioVKhVu3+iJqs7kCZXP19yRJErc+uJPnv/ouY4Mza15DqVDh6X86zBvPnKJjYwPtPQ2kGmM1ZQrLtCkX/fss5MrkMkVy80XyuRLFXBnbcvjF3/9+9t7mF8oLNPy9zUoINBS1l2Dok7XfCykKQkUIHU3bhxF4kEXXR5LT1eNkPG8x9uAC1+60d13GmTcvMlb4BpuS/5bh3JeYrxxhR/o3mCq+iOnM0xC8k4sLf4HtFXC9CkGllZ74Z3A9kzNzv0tP/MeJar2Ax1D2HwGJvoYfwnZdkqEgYV0jFli9FbznecwuFGmw4uxt6Fq1sdCV0AyVj3/mLiLxEF/7m5eZGV89tfK9hlmxOXm4n/6z49z/fQf42E/cSSQevG5a13oQQhCJB/nhX/4Aiirz7Jffvi4KnFOVW/nn4LQuwsP3WK4kkl8en70qyGM7LscujbO7r4V/eOkoYUOjZFokI2sHBFs2pPnAJw/xN7//xDXvObdQ5OThfk4e7gexSG5fyjOvBVVTri28JQSqfoBK5SVcN4OsdOC6C35xBzKafhvl0tfxPAdJTuA6s4CKEAEUdQuV8reRlTZs6yyOM7L+d3GdxqnJCcr2NKYzT8EewvFKVOxp8uZFovpmdDnJhtgPockJLDfDiZlfI2eeJ2HsJiA3Ml18iYjag+kuMFd+m+74j7LYlKg1ce0OvwMTc1i2Q108TOQ6yMKe51EqmbR01tGxsZHZyez/VU2eQq7M1/73KxRyZX74lz9wlZSH5zkUrQu4XhlNTlO2R1CkGKqcomwNoMq+G+Z6FSSh47g5EDK63IgcHuL7/+1m6pvjfPWvX17RDuH/LUhC0NiWRKoWJmSLZSbmcsznS4zOZjg3PFU79vzINJbt8MANfUxnC5Qti+7G1Lr6QbIscc9H93H5zBjPf+3dNauCrkK1P+h7hRAhZKWjVhcrpDCy0gFIqOoOZPlTLMx/AU2z8bwgkvwxIpE2DOM+P2BU+F94noUs1xEM/xhQRyj0ryjk/wf57O+gKBvR9TsRYv3w6HUbpyRUCtYAnucQUXvJW5eoODMElXY8oGSPMlV8AdsrYDoLOF4BSSg0hO7k0sJf0Rr5MNnKaSShE9F6V5x/rlQib1Zoiy5t2F3PYziTIR0Koat+ukJdR7oD/M3/aP80L33jKK8/fYrR/ul/Vrd2PdiWw7NffptUfZSP/+RdK5QiPBxsd5585QSa0ogsApTtQQJKF0XrAkEEBesMtrOAIkVwvDKSUAiqfdhuhnj8Jh77sU1s2NLCF/7sWc4dG163d8v3EpIsEYqu9HIUVWZmIkPnJt99nZzL8Q/Pvsvb54Y5fnmcJ988u3SsIvPhW7aTigapj19fSgF88vwP/bsHcWyXV548/s96v4q6hXD0V4Bq/ETdgaL0Ab6LPHh5K/0XIjz42A4qJXj6G2fYvnuSns1NBILfTyDwCB4OQuhQjYRLciuR2H/E88oIYeB37V5f0Oy6jFMWBrpSz0LlGJocJ6L1slA5ievZGHKakdyXmS2/RWv4QyhSlLnyO7X5Kqr1oUhB5stHmCsfJhU4gHJFlPDI+BjHJib4N4durDlGtuvyv959l0f6+pjNFLBsB8t2WY2P63ke+UyJp//pMN/8h9eZGJpdd3YVwt/LXKuVX/XseJ7PdHFdv3p6PfdoOSzT4Zv/8DrbD3az40D3UlGFm6NoXkAIufps2/BsB0WKIQkNx/NdUFWuQ5VTmPY4mtKIIsWQpRCqHAcZ9t3eR/eWZl78+lGe/tJhhi9NrbsX/Y4h/Gqqzk1NHLpnK4fu3bYi9ZBIR9hT11u7v56WOn7l++/kL7/5Jm31cQ5uXmo9oKsy0ZDxnjmbQgiS9VF+4j99iOauOp78/BvMTeW+N/d3BTxP5vL5LKND/XR21xOJBTlzfJhwxGDLzjY6NqS5cGYMISKEIhINTQnKZRPP8xi4OMlQ/wwd3fWkG1ROHbnAzHSWrTvbkWSJ/vMTNLen2NDbeE3S9XWW70mElHYmi8/RGLqbiNrDaP7rBJUWFCnEQuUESWMf9cHbKdpD2O7SQ5NFkHTgVsby38DxSrRHPnaVQdiui7ksAulVVbgXyiX6p+YJ2L4Rrbbf9DyPyZF5/ub3nuDVb59YU4w5ENZp726gZ1sLbd0NJOsj6AHtuljpflTToVyskJ0vMD2eYWxghuFLU0yNzq8rAL0wk+fJz79J366OWgdrRUqQCj2AQEISOkKo6NWGQjEpiirFiWg7QUhIqDheCQ8HRYRXOGl+8X6MR374Fm56YAfvvHiW158+xaVTo2TmC9ctXrwaFFUmmgjS0pVm694udt20kQ19zYSigaue2XKK2eL/DU3l9p3dBHSVxuR747OuBSEEsWSY7/upu7jhlk18+wtvceTV88xOZr/jexUC9IBGXWOMYLX8MLtQ5O3XLnLL3VsIRwxef/Ecre0pRoZmCYZ16hvjK65p8dbzuTLPP3GCzp4GXnjyBLv2dZFZKFLMm5SKJm++fJ7WjhQvPXWShuY4kej6FLnrMk4hBGFtA4PZ/0NY7cFQGnC9MgGlGUno1AVuZDT/OCV7BNez0ORkdZPsf7YucIDB7OeIan0E1aWOvvOlEp8/cYJ3xsaYLOT59eefr91oplJhPJdje0sD0xM5bMe9SmjZ8zymxxf477/2Fd5+8eyq+0pNV9l7ex8PfP8Bere3EY4FrhpM7xWe58tOLMzmOfnWZb761y9z8eTImivq8TcuMTYwQ1dfU/WZSGjySuUESfh7aUXyvQqJpTKw5f9eDZIk0dCS4IHvP8gdj+xhfHCWCydHuHB8mOFLU8xNZSnmK5gVyy9jcz0QolZYrukKRlAnlgxR1xijvaeBjt5G2rrrSTfFCaxSTXU9WB6l/V5CVRW23NDJxu2tTAzPcubdQU693c/IpWnmZrKUCyZmxcZ1fIE1IQlkRUbVZHRDIxwLkKiL0NiWpL2ngbaeehrbUtQ1+vEP23aQZYlUOoKQBFbFJp4KMzeTo1K2/FYdrp+SWfx7cQKvlG00XWH7ng6SdRGOvHmZlo4UqfoI5Wpjr217OtCuQx5TrNcElmXcDtstkDcvEdY2IguNrHkeXU6iy/V4WOStfmw3T0BpwfMsFCmMJidqnz0y9Uu0Rz5CffCO2ouu2DZHJyb46pkzXJid4WBbW82t1RWFQ23tbEvXM5spMDWfp7ctTXCZX1sumfzlb3ydJz//xqpubCCk87GfuIOHPn0Tocjq0eDvFj5/dILf+4XPr9mJWpIEP/tfP8q9H923TCDKo+SYFJ2lXKwAgopOQF476LVYheU3yVmn7Zzn+XSwsl3L45UKldqgFZLwq6J0FSOoEQzrBIIqesCoyXIuv9aCU8bzPMJKYM3vnS+XGMll6EulUaXvLT9zPXieh+u4lIomhWyJQrZMqVjBsR3cai8cVVMwAiqBkEEgpGNUhbZypknU0Fe42bbl8MZL58jMF9i4uRlNVzl5ZBBNV9h/cy+njw1x7Oggt96xhbr6KC9++yR6QOXmu7Zw6dw40xNZWtqT6IbK4VcuEEsE6dveRqlkMjo4Q31TnD0Hu5cLrK/6QK/bOL8TuJ6N5WaZK7/DaP5xdtT9es1gl+Po+Dhvjozwo3v3ruzRJASlisWl0Rks26GntY5ItbDY8zwOP3+G3/n5f6CQu5o2JCsSj/7rW/nkz967qpTH9xKe5/HUFw/zp//xy2u6uA996kZ+8tceXVKV9zy+PPIqXxl+pXaMEIJPdtzJ/c37Fk/MoiFSMxQLin8H6g6Etnfta6q8AO48GB+6Lja+Zw/glb+BCP4gQloZE5gqz3N0wW8QtT3WTVNg9cqal0cG+PNjb/End32QhPGdqRrkzQqj+Sy9ibpVJ4Gh7AKGolAfvP5g0lpwXJfXhoZoi8UoWBZ6lWwN/vuRPUFnKkHFcTg7OUVY13HwiCoamVKZgK76nxMyC+Uy25t9ZQXb8lfeN146R7FQwbYdGpsT7NrfhWU5SNWJcdn9rWqc11f47nq4trM4TJDV62tyWnFmuDD/p1hujs7op1Cl+KrH9aXTtMaiq7aZNzQFXVXwAF1dJlhctnjqi4dXNUyAzk1NPPzpm74rw7Rch9emLrGvrpOgsvZ5hBBsP7CBVEOUieG5VY+ZGpvHtpxaPSpA1iowUlpKrAsgZy/VkHruFFjvgn5vTZwaXDzzCEJaXVBsJd6DRIa7AOZbEPzEVb8yXQsJCRePc7khArJBXLvaOFzPw3S/8yiq53kcn57g24MX+Y8H77iq8a7ruXzuzDH2N7ZyV8d3b5zgtwKxq0TwomWhyf5qulAq0ZlI0J1OkSkUODUzTUs0iut5WKEQU6UCQVvFA6YKfjprp9TkV41V3/ENh7qZriplpBv8TMT1uLOLuOaRZtnkuf/zCqPnfVnJcCLMB370bsKJa5ewG3KaTcmfRxJqLUK7GvdOl2W0KivF8zycqnKaoShVg4VyxWL5Qj4+tE6fTAGH7t5KqjFGxbExXRuBwPVcAoqGIiQKtonp2kRVf4b3+aEaHh4F2ySkaOSsMnEtWKNxeZ5X+1xMDdTqY4Hafm0t4yxWqWzLjXMteJ4HXgnMw3iVlxDqTjxUWDG5eX5vDq8CUgRYlHRx/BVT2QwiUCsM8M9ZACGD5/rnl4LA1W6q701VwCuCiFGnx9gQbqbs+lIdynorsedxcmaSc/MzBBWV29u6aAotsTZG8lleGR1goVKmL5HmYHMbhqxQsEyeH77M1y+fYySX4bNHXkcWgvs6N7IpmebywhwvjPTzzOBFxgtZTs5OEtcNHtu4lbCm43oelzNzvD42RNGy2JFuZE9DM7rsV6YdmRqn7FjENIPXx4cRwD0dPdze1YXtur6Uni+zmwAAVFRJREFUCUsGarsugepiYLkuUV2nMRIhquuENY2oYRDVdRRJ4p2xMSq2jeU4SFIZ2y0jkFCNCI3tARQRxPHKOK6Mi4Ui/DFneyUUsXaRyjVHyvTILOOXp7j3B+9AUWVkRSawWvu8VeAXVyfIlsrEgjAxn2NsNssNG1tXHDdTLPLtixd5dPNmJCH426NHeWt0hPt6NvLo5s20NyTZ0CKt2BcMnJtYMwFvGBpb9vot5g9PD/DWzABF2ySi6myNN9MZTvHE6CkUIdEUjHFTfTdfHTrKJzccYKac54XJ83y8cy/H5kZ4efICHVvuRJcVzmeneGrM77PYFU7xQOs25OrgV1WFwDpNUW3bxfOuM6LolaD493iVZ8Cdwcv9NogQIvRjIPtBFs88DJVXwMuA1IAI/wzIafDyeMW/A+soQr0Bwj9NzWsqfR7PXfBXSXcWRBAR/gwoPVd8fwav8FcIKUFOfZQF22SmkiFnF2kLNhBW167mGS1k+fKFU2xM1HF0aowXR/r5jZvvIRUI0Z+Z5zfefIG2SIx0IMRfnXybs3PT/MiOvTieV9P1MWSFxlAYWUgEqmLPtuuiyzIekNADNIbChFUdSfhGdWJmgt87/AqbU2lCqsYfvfsaj/Rs5uObdiCAo9PjPDN4sfbdJdsia1Zoj8ZXXH8icLU73hyJEO7sJKLrPjNp2XGe57G/pQXbdfGkOc4vfBvHqwASXZH7GSu+zsbYY4wUXiSitjNbPkVr+DZst8xU6Qjd0Q+yhld7bePUAzqq7s8+ctU4x+dzzOSLKLJEMhJEV2QGpuZpiEcIBzRGZjJULJvuphST8zmePXqRW7ZtQJElTgz4+rFdjUkSYX/WvjQ3x2tDQzy6eTPvjI3xwkA/D23axNfPnWVvczMbkisV0DzPY3Rges1EdChqUNcYRwhB0TbZFGvgbGaCG1IdXMhOMZCfZV9dJ9vjzfz3cy+yJ+lrpV7MTdGfm6UlEEOTZA6mN/DWTL9PhPY8nho7hSopNAWivDR5gVsaNhLTll7SerlV+bpyqlUIA4KfRAgVzzqJiP4//qsSQfzktQPuBCL870BIeNnfAPMVPOMREFFE+Ceg8Fd4zuyK1+65s1B5yz+fXI+X/2O80tcg/POLXwxuFq/4eX+FDTyGQRDFyZDSoyS0CEF5/YlZlWR+ZPtetqTqmSjk+Mwzj/PO5Bh3d3Tz1YunSRoBfv6Gm9BkmU3JOv7wnVf5YHcfrZEYD3X3MZzLcEGe4WO921c00+1N1tEUjvD1S2e5uaWTezqXJhTbdfj82eNsq6vnJ3cdRBYSjaEwXzx3kvs7e4nr/jXPlYv815vvpSMaf0/BlEXd24Jd5pXJE5Qdk6gaZF9yE7qsEq62p5wpX0ISKnG9m5I9hyRUKm4G8DCdPJ7iEFKbmC4dx/EqRNRW1jJMuA7jlFWZ3Fyex//0SbSARiQZht2t5KurQDRoYDsOPc11PH/8Iu3pOGeHp+mojzOTKdDZkMR2XSIBjbJps1AokSmUeflkPx88uAUB5E2TRMAX+X3m0iXu7e7hI1u28u7YOJP5/FXG6breumVruqHV9ppCCAKSSlDRMGS12v7dxZAVFMlvBeDisSvZxpvT/eStCp/csH9VQ7Jcl4SmkzYifLxrb21WB5+KlsusXb8aDOvXLcMvhAQihCcMECqICKLKyfSLpyWEfnt1xXNB2QjORO1+IYAn1ljFtb1+MAkJ1B141nFqRdiejVf4C0BBhH8JISJoQhBRgkTVEOCtqRO8iKQRpLHqxiYDQRpCYYayC9iuy+m5aYazC/zyS98GoGibVByHrPnddfEu2Tbn52ewXJf+zAIAWbNMybYoWGbNODujCVoj/t7vO4nbX8iN8gdnv0TBLtMTaWZrrANdXhoDIaWRIecZAm6K5tAh/OdlYbtlKu48Hh4pfQtnM59HIGgJ3bzuhH1N44zXRfnEv3+M0QtjCEmirbeZly8Nk5IkbMdlPl+kbNrs6GxicHKeQtmkpzlFd1OKY/3jpKJB6qIhWuvi9E/O0d2UYkt7Ay+duuxHI4UgETCYKRZ4eWCAi3OzfGrXTmCJcnQVqnnGNbHsI6okISGhS76Aky6r7Em189LEBU7Mj5LQgjQEoiT1EF8fPk5bKEnKCFO0Td6c7meilOW1qUvcVN/NbQ0beXnqIrIkkdbDKyRFZicz6xbYpxpiKN9F5+arblBEq/s4gR/4ub61QEgR/3hR/dxyY3OnwKsDt+i7vXIEx3M4kblEzi6iSyobQi1VQ10drueuEDRbbOkuAEUIDjS18WjPlto7koVExxWu5XuFQKBKMje3dHJ7a1ft3Jokkw4sXatSbe/3ncDzPE5nhijYZTy8VSVoKk4Wyy2St8ZxvcO0hm7FkBNczH4Z13NQpSC6HEeTwshCR5PWF/q65mgpFys88T+fwTL9/NiZN84Tv3kjuqHhOC6yLJCFxDcPn0FRJFpSMUqmhSrLhA2tpqj22pkBmpNRwgEdWZaqgk7+k9qUqqM1GuMv3j7MQ5s20RlPMF8q4Xou9eFVonJCrNvjsFSoUCz4Udw9qQ4EsDneRFDRaA8nialBFqbKFCsWN3Z1UinZOK7LJ9sPEDMClMoWDi4BW+OHu29CVxQ0WWF3qp3WUIIFs0SdEa6Rx13X5d2Xz6/Z9VoI6Oxr+g4GhurvPz0Tj0Uj/G6xzqohNyAivwylx/HyfwKRX0WSUrQF6yk5JnE1TFJbf0BNFQucnZsmphsMZOeZKOTpTaRQJJm9jS28PjZMYyhMKhDEdl2Ktr3CAzEUhbxlUrItdFmpEtP9+5aEQJUk5ipFX39J+HTCgKKwM93IcC5DWzRGVDMwHQfTsWt7xOXw3CyeM4EQGh42QoQBxQ+Y4aerhNDxMP3idKkB07U4melft5B+tnKKemMXIbWZkcJLtIZupTf2URzPRBIKAkHBnsD1bFpCt15zm3NN45wdm0MLaHzkFx7Gc13+6Q++Tl9dnFRzsnojHoosky9XCGhqTdVAliRS0SC6qvDooW1YjkNAU2mti6MqErfv6K4N1pCm8Ys33UTONEkYBrIQRHWdX7zpZpqjVw8GSRLEU2uH0vOZEkMXJmnvaSCq+i5NpPr+Q4o/qZQyNkFNp1JyePHYJXZ2NzOdKaDXqxwZHMN2HPIlk80dDWzcUF97kA2BKA2BpWvyPI+xgVme/tLba7aKD4YNerZem1x7FdQtUPoyXu43QUojgh8FKQlCYoV7sOz/nnUezNfwKm+Cl8Mr/jWo20HdW2vYu+yD1c9W/41S2++S/29+UCj8s7QG6v0jrjGYhID6YIgvXzjFNy6fZSiXYU9DM7vSTQjgke4tXFqY45deepKUEaTs2HTHkvzcDTeiy9X0Q0MLX790ln//ytNENZ3v69vBtjqfgB9QFG5p7eTzZ49zZHKc+mCIH92xj7Cq8anNu/jdt1/mF154grhuULQt9jQ082Pb9/lRdaj97dn9eNa7eFICRAIP1496Cwk8E5D9n7kzoG5FSA3Mmjn681f321mOBmMv0+VjZMujNIjbUUQIPFAIMDkyRzwdZMG6TNrYRVhpvWaXuGsaZzgeYmEqw9tPHcWxHCzTJpYIoy9z0TLFMpMLq68aq0FXFdrqlhgoQoiaUFbRsgioCrqi0B6Pr/p5IQQtXWkUVV41KGSZNq88cZy9t/cRWIViJssSDYmIX/wuCeLhAL1taaYzBSbmcsxmC9iOS8jQ0LW194me5+99P/fZpxhapwFvR2+jLx/5XpdOZRMi9utgD4IIgYgCCiL4QyAvKkYIhPEwLOoEiQDILb4hV3+PiPh/Gx9aOg5AuxGhbsFvutqBCP8UiLCvMxT+SbDOAs51X/fmZD2/cfO9JI0A5+dnMBSVran6mnJ/KhDkPxy4g3enh8ibNjEtQE88hbasmmhrXT2/e9v99GfmUWWZlvDSRCgJiU9t3s2e+hbmykXSwZCfbhOC1kiM/3LTPZybm2GhUiKqG/TEU75Ym+fSmYii64LZSoE6tdmPVEthQMNzhnyytJQEz64G3lw/lST5+ffBwiSz5vqqGePnHOZnOoinwhQlwen+QcyyhaLITIzMkWqI4rl+AceUMcimXR21nOhquKZxxtJR7vrkLRx/6QyKKnPvv7qdYGRluPmN80P8zldfWPEzu1rX6Lfx9pULHNdFINje3shvfeoBwlWtUtfzODo+zt8dO8pQJsNPHzjAzoZGnrl8iQd7ewlrVxtY16Ym4qkwMxOr7/MOv3CWFx8/yt0f2bvU+HcZNjSnGJlaQFMVtnT6mq+bOxoYmpznht5WoiGDqfk8jcnoqoPTdT3GBqb5+//2FK88cXxNGUxFkbnlwZ3XLHJexHKnUwjZD/pcmerQdi07RoC6TGZUafP/rIblxwFCaQOqx4o4aPuWnTcJ+o3Xdc2LSAdDpIOh6r+DDBZGuFC4SFQJ4QEz5jzdoXbmGaA5Vk9XqJ5L+YssuFEMSWeyMktHsJnueB3d8dSq3xFUVfY2Xu2FCCGI6Qb7m1qv/p0naAiFKXkVRgrzpFPtiGpts796Sf6ktIri4OIxJxb6Md31id627cuERuJB8nNF9IBfFjkzkcEIaj5R3PPIZUpEE0G/i8A6uKZxCiFo39xK+2b/pofPjlLKl1cY6Na2Bn7x4Vv9GwHOjk7z7PELHOhtp68lTUjXqFg2Fydmef3cEDdv7sRYRq4dz+X47BtvsKupkbxpkq1UkCSJpy5dYmt9PVvrr9YVamhL0re7g1eeOL7qdZcKFf72D5/EcV3ueHi3z9RfZmTxcKDG0o9Vm76mokFS0aUcXlNqpUvtVVMqCzN53nz2NF//u1cZOD+xLpG7e2sLN92//RptBRYhVqVSeZ6H7TmUHbM2QBQhY8hatbX9tV3OYtlkeDpDXSxELGRcxfDJFMuEDW1F+mL59zueS8W1MF0Lt5qT1CQFfZme0/JrkIRgqDhGxsoTkHWaAg3MVuZJaXHq9ARdoTaGi2NMlKeZt7IktRhJLU5Ku7q8s6bDW7sGG9fz9RdlIVCEjC6pKJKMWKXmWJYkWoNxBvKzRNSVqSAhBIirv3P5d5ecCqczQ+s+X4CN21pr5AbbdNB0BSEJOjc14Tqur8pfrcT0OZ7rxxDWNc7JwWlCsSDTwzM1nuBbTxzh7k/dusI4W1OxmshutlTmiXfP8unb9vDYwe0o1fyeTwPzeLr9PI+/fYb7dm8iUe0qdXFujlQwwE/s28/vv/oq4AcGDEUhU149zK7pCnc/tpd3Xz63puDy/HSO//lfv87RVy9wz0f20bu9jUjC7695va7aIgMlt1Bk5PI0x964yOHnzzJ4fmL9iDEQiQX56I/fvq4w8nIIQFnm4tmuw0hphnfmznNiYYDJ8nwtWmjIGiktyqZoKzckN7Ix0oohqWve1+nBSS6OzRIPGWxoSVE0LaJBHdcFWRJcGJ/hls1dBPSlaijLcxguTHE808+ZzBDjpTnydgnLc1CEREgxSOtx+qJt7IxvoCvciF69BgmJuBYjooapOBXmzQy6rGHIOgktymhpgjo9wbyZoTFQjyxkYmpkxf17nkfWLnIxN8bpzCCX8+PMVLIU7BKm54DnoUgKQVkjqoZoCiTpCjWyIdxEa7COqBqsBZMKtokury1Gvvw7K65F3i4xU8kyXprlfG6Ui/mVUpYZq8C3xg4TXIeksBxpI8bB1OYV93ctrGuc/SeGCCdCPP23L1Lf5rsZoxcm1nThAGazRUbnsmzvaEJd5k76fEzBppY0U8++xeRCrmacqiThXNF0tWCaFE2LsLZ6TasQgp0He7j5wZ08/U+H11y9ykWTV544ztsvnqWlK03P1ha6+pqob0nU2jMoilQjVFumUxWCKjE3nWN8aJbxgRnGBmeZmVhYcyK4Erqh8uiP3Mr+O7dc56rp35NWFUKbNbN8deQ1npl4l6nywqo6vRcY5Y3ZM3x5+BX2pTbxsfbb2BhpWZHiWUQ0aJAvVTA0helsgYmFHImwX4JYqJjI1fI18CeFi/lRvjbyOm/NnmPezK2pEwyDvDh1nKgaZHeim0dab2JrrANVUtgRW+xs7uFW0w+q5NdJW66FJqnU63VIwo/ILi81zNklXp46wZPjh7mUH1/B3ln3GSIIKQbNgRS7Et3ckt7G5mg7hqxiujaVNVzTmUqG12fOcDk/znBxmqnyAvNmnpJTwfaujmtMVzL85aVvXdc1AexO9HBDYuP3zjgPfGAPC1MZHv7J+9iw3Wezv/nEu2jG2vIKiuznP8+NTrOpOY1cXaU8z09gnx+boWw5K1q+9dbVYbkOf/j6a1yem8dyHQ6PjpIMBuhKrO1y6AGV7/vMnYwNzHDy8OV1U33losmlU6NcOjWKEPj8Pt1vA76cGuU6LrbtYluLfMD1ntDqCIR0PvSDN/PID96Cosq1NJQQoGrqmsYq8POwI8Vp/vjC47w9dx73Okr+cnaJ5yaPcjY7zI90P8Ct6e1XDQJVkehtTbOxpY7R+QxhQ6OnMYVWlYCxbAddlSk5Fb419hZfGHqRyfLCdd2vh0fGKvDC1HGOL/TzWNstPNJ6I0F5bR6oXF1xruzu7XkeY6VZ/vLSt3ht5vQ193mrXUveLnE+N8L53AiX8mP85+0/gO05zJTz5IOrEyXOZof54/Nffc/f98+JdY1TCEG8PkZ8mRjvDXfvQF6n0qU+Fmbfxjb+/Kk3uDQxy9b2BoK6Rtm0ODM6xdPHLrCrs5nmZZ2WUoEAv3DjTfzDieMULJOz0zPsbGzkJ/buI6xpVBybkUIGu8p4aAhEiOt+6V9je4rP/MdH+LP//BVOvT2w7qq+CM/zOXv/HDo09c1xPvoTd3L3h/diBDUmBqZ5/VtHUDWFQFjn4IN7CK0RHJKEIGeV+JMLj3N49uyKuUYVcnV/KbA9B9O1rpqLxkqz/Mn5r6EKmZvS21bsX+tiYQYnF3j6nfPct28TPU11RAPGClWDkl3hc4PP8U/DL1N2rhavVoSEKqlICFxcTNfv+rYcc2aOv+1/mqxV4F913Uuw6kY6jotju6jaSkaT6/or6iK3cdbM8tnzX+Wt2XOr5hRVIaNIii+P6nnYnovl2qseKxDckNhIUDYIyB73Nm+plVteicV4wr8kXEeec56poWm2HNqE63qcfPUsG/dsIBwPYVkOuYUC0Xio2jDGF1H+zL0HiQcNnjtxkW++ewbH9ZAkQSxocMe2bj59254VfSmFEHQlEvzKLbdStv2ZK1BNrYBf9Dycn2cgN0/KCLFgljlQ37702c1N/Nvf+T4+98dP8eqTJ9ZURf/nRChisOeWTTz2I7fRs721NtjKxQrZ2TxNnWlUXV23EMHxXL468iqDxSk8QBUKPZFmDqT66Ik0E1f9qqScXeJyfpxXp09xNjuEtcztmjNz/OWlJ2gJ1tEVWuo8NrWQw/U87tjVQzoWuqploOXafGXkVb449BIVd0kUTRYSHaEG9iV72RxtJ6lH0CSFimMzXclwMtPPW7PnGC/N1Qyk4lp8ZeQ1YmqIj7bfhozE2VOjKKpMY1McTVMwTQtNV5mZzmFbDl3d9diuw9dGXuftufMrjM2QNbZEO7gh2UNHqIGYGkKVFBzPpWCXmK5k6M9PcDE3xnBxmgUrj+O51OlRDtVtqeY4pRX56SvRFW7kR7sfxOFqTyVjFvj66BsUnKVVN6VF+UDzAQLrUAmXo8FIvCeXFq5hnFbFYvjcGP3HB6hrSWFbDseeP0XbphbC8RCe6zJwfpKZiQzNnXX0bGnGCGikIkE+c98hPnbjDqYyBSqWjabKpKMh0tFQLb2yCM/zyFQqvDkyzMXZOYSA3lQd+1tbiWi+IvruuhbmKkVszyGkrHSr/RU0yU/+2qPsubmXb37udS6eHLkuXdfvBovFENsPdnPXozewbd+GqzqY2aaNrEhoARVNXztgA75xXi74ie6oGuT7O+7gvqa9JNRw7T7Bf177kr3c17SXr4++zheGXiRvLw2c4eIUXxx6iZ/b9ChG1W30PMiXKpwZmqIuFqoFfhbP9/bcef5x6MUVhhlWDB5tvZmHWg6Q1mNXRUI9z+OOhh0MFab4+4HneHHqGHZ1JTVdi38afpnN0XZ2xjeQz5eJRgOcPTWKWbFRNRnL8pUKUmlfY2imkuHFqeMrVuOIEuRHuu/nrobdhBRjxXNYfh0eUHYqTJTnObHQz5uzZ2kwErQE62rbqkVc+XnLtWkwEny0/dZV38tIcZpnJ4+sMM6EFuaR1htJrMJr/V5hXeMs5kocf/EUF969zEy1iU7bpmaidSsFmyRZopAtcf7ECDv2b6gFf9KxMJGAju26KJKEoa7eASxbqfCbL73E5fk52qIxPDyevXyZlwcH+KWbbyGsaQRklZQeYrZSXNU1EUIQDBvc8aE97L65l6OvXuC1p05y/vgwc1PZ75kLqxsq8bowHRsb2XGoh52HemjbkEYzVje8dFuKxs40juWsecyV0CSFT3fezSOtN6JKV7+ixXMktDAfb78dD/j7/mdqK6gHvDZzmvua9rIr4beYm8sVaamLMZ3JUzYtAssa1S5YBT4/+AIZa4lMoEsqP9B1j38NYvU26UIIZGQ6Q438VO/DVFyLV6aXeonOmTm+MvIqm6JtdHSlMQwV1/XQ6hTKJZNAUCeXLZFM+rnRkdIMU1fsc2+p38YDzfvRVnkOy69DAEHFYEO4ia5QI/c23oDtOXieS9nx28wPFidoC9ajCBlZSLieH6i6kBvBkDU6Q414nlcLUC37hjW+99oprO8G6xpnrC7KBz9zH+OXJujd679kWZVrLpHn+Sptja0JGtuSNSaI63pcmpzlG2+f4eTQBPmKSVBX2dxSz4N7+tjcWr8in3Z6epqpQp4/uP8BmiIRn5SbzfKfnnuOC7Oz7G5qYsEss2CW2JZoJKKuHb4WQpCoi3D7w7u56b7tTI8vMHBugounRhi6MMn0RIb8QpFyycQybRzbrVG9JOFPNLIqo6oKuqESihrEkmHqm+O0VcWgmjvqSKYj12VspVyZif5pYnURQvEQ5WIFYx3eJ8DOeDf3N+1d1TCvhC6rPNJyI0fmLnJ04VLt536A5hjbYp0okkxXU4rLYzN0N6WIhpZyfZ7n8ebsGU5nVhLXb05v46HmA2jS+tqqUH3maphPdNzBqcwg8+aS+uKR+YtcyI2ys8lvQBtfh6S/YOYxl63cAkFfpG1dw1zregKKTtkxeW7qCI1GkkYjyWszJ7m9fjcTpTk6Qg0MF6fI22XmzCwpLYrl2uTtEn3RdsLKdyaz8r3EtVkp9VHC8SCKppBfKKCoCkY1oV/MVxi6NIVZtvA82L6vC8/zeOfyCL/91RfwPOhpTNEVSJIvmxy+OMLLZwb4xYdv5dYtXbWB7bguTZEIDeFwrcNwUyRCfThUi1bqskLBNjk5P8FOqZm4vvbDW3RhNEOlpSvtu9xbmwmEfZGvuakspw9fIpIIkUjHsG2HofPjOJbDxqqEZQmH+mQUzVBRdIVQwHdX3Wo95PXqrmZnc6TbkpQLFeYnF3DWaEJUeyFC5s6GXe9pcETVIPc17eVkZmBF2P/o/CUWrDx1eoyJ2Sw9zXXEI4EV1152TZ6bPLZi3xpRgjzUcmBdobErIYSgO9zM9lgnL02fqP08Z5c4PHeOHfGua05kS7W/VVZLNQp8rRrUtSALibQeZ7aSZUO4mSYjRYOR4GJuhHkzx1R5Hk1WaQ2kMWSNi7kRwmrwunOX/9y4pnEuTGZ499kTbL1xE1/57LdINMR59OcexAjq6AGVLbs7GLgwgWb4pypUTP76+bfpa67nJ+8/RDoaQpIErusxmy/yV88e5m9eeIet7Q1UPAfHcwlrGvmKybOXL7Gprg7PgxOTk8hCoiuRxHFddEnm/rY+ClXGwlrwPI/JoVksy6atp7H289nxeRo70yTrYyiqzKnXHYoLBW56YCee66FrEs9/6S3u//5D6EGNFy7344kKCUlCclxmFkpIQjCWzVIXCuF5HrqiULRMQpqGJssokkRdaOXK0NrbxOVTI0wNz7Ljlj5ideuzOhJamC2x9vc0GIUQ7Ih3UadHmSgv9fCcLM8zUpyhTo/heR7PHblIOh7iwOaOmls7WpzlQm5l344N4UZ6wi3v2SA0SWFbvJOXp0+uCOiczgxRds1rGntKj2LI6oqc5ivTJ7mrcTdNRvI9X49bdVETWgRD0qg3EoyXZmkPNTJRmmNDuBnTtcnZRTqNRmYqGZoDqfV4O9cNx3MYKkyT0qNE1lEsXA/XNM58pkhuLs+R506y996dXD4+SG4uj227vPPKeTJzBZbrwM5kC4zNZfnXd+1bkS6RJb+F94cPbONXP/cEp8en+Pszx5gplhAC8hWT45MTxAyjFiBqCkeYKuRRFYnB3BxnFqbwPI90IExv3K+NnBlf4K2njuM6Lvvv3YGsSDz+P5/DMm1237aZ/ffsYOjcGKfevES6xWfSBEI6Ld0NTA754lqSLNHa3YBuaLUh5Qs9VbBcB8f1Z+5SVV+mbFtMFwq4nq9/1BiJ8NrgEFsb6q8yzmAkwL2fuhnHdpkdn6dSMlHUtVfFRiNJ6hq0rNWQ0qN0hBpWGGfJMenPT7Ar0U1rOk6+bBINGrXiEM/zOJ8bIWutJIlvirZds5JmNQghaA3UoUgy1rJ84WR5nqxVvKZxtgXTtATruJBbkhg9nxvl9858kU923uWTm9epgroSuqyyJ7HU+mN3YmPt372RlTW4s5UMUTVIR7DxPRmS6dpYru13WxfLg2xwIT/Ck+PjNAWSbI930RKoe0/Xf223Nh1ldmwOSZa58eG9DJ8bQ9UUgiGd3m2tnDk6RLopTjHvR7IWB/diHurK6N7ifxNGgF+55VYWFvLk54skGqLYpoOsytjVsjhJErRGooQUjZ5oHe3hBIoksWAuRc0Gz44yN5nhwL07CEUDaIZKR18LelBj5y19yIpEa08jbz19gtxCsWag18L2xgaoGqTtuKjVMkRdljEdh3Qo5K+W1RUzrGk0XME9PfbSGaZH5mrxhPHLU9z7qVvWzHMCNBhxdFHAMS8g5JT/lr0SSLGq7o+KkOp8PqLcAMKXY9EklbZgmjdnz64431DRn9Aujs0gSxKDk/M0pSIkI0FcPC7lx1dERwWClkDKD5asUhlzLQRkHVXIWCwZZ94ukbdKpLQAstCxvRISMrK00lgTWpi7G/YwkJ9YFtzyeHf+IhfzY+xN9nJP4x62xDqIKMH33NJhPSS1KMnke5sUXc/jVGYAVVLYGG5BX1YrKwuJuxtuYG8ixyszJ/mLi9+gLZjmYGoLuxM915VWuaZxRpJhPvZLH0JIglA0yB3ffzPRVARJlqhrjJFqiJKdL9C3y887psJB6iIhvvH2WdpScZLhYI2VMp8v8Y23zxAPBWhNxoiHDMbmTOyZecz5WRZmskSTYaLJCKOXJhCSQLq/DqELLM/l6OwoluvSGVmqGtq8rxuzbPH6k8e48cFddG9vwwjpBEI6oWr9r6op61Y1XQkhBFHDD5pE9WuvIJbjsLe1BUNZ+Tizc3na+5qXiOEetU5cayGlR/Csd3DdKSSvF4SCax0FEcFzZxGLdCYEiDBK4LHqNUO9EUewsnfpbCWL47mossypgQkKZZM3TsOtOzag6RJjy6Q5Fy/yibHDHJ47f837Xg05q7giHQN+qqJgZ8lbRXQ5zmz5NJocRRIKuhzHcgsktB4kIXNf0w2czg7y8tSJFSWDWavIc5NHeWPmDN3hZm5Mb2FfchNtwTSatHo0uXZHnkf//Dxnp6dJBoPc0NxM2bZ5Z2wM1/PY19JCybKYLBToTiY5NTXFjoYGdGV985iuZJitZEjq0auKIDw8zmSHODJ/Ecdz+EjbraT1GC9MHaPeiNMRuprMcSWuaZye69F/YojjL53mnk/fRm4+TyQZrqlo7zzQveL4SEDn07ft4fcef4lTI5P0NtURNjQKZZMLE7PkSxV+/oO3EA/5wZl4OsrCdBYtoKHqCtFkhFRzHM/zS+ckSSJvVTg6M0p/bo75Sone2JJm6+TQTK3yZGHGjxLGUmFOvXGBaCJEz84ORi5OMDk0y6XjQ6Qa4zi2w8AZf8UdvTRJXUuSC8eGmJtc4MLRQfpu6EIzVDzP4/knT3DyyCCO47JlRxv3Prz7qoGgyvKKcsRFHHxgF4qqUMiVKOVK7L1nO+HY2sp1AOHqiuAJBSEncSpvgOeAEEhyEyDjuTOAhCQvUcMEgpgaqvIXlwZK3i5he77IsaLItNcnOLilg1BAo2CXrnJpPeBsbhhyq7d5/07g4WG6JjlrAtPN4nhlFsxpdClGyZ6m4iwQVTtQhExMDfHTGz9ETA3x7MTK3CJA0alwItPPycwAX9ReYkusg1vT29md6CalR69IgfhwXJfHz56lJ5kkqPpas988fx4Jf/X71vnz3L9xI68MDPDm8DD1oRB7mq7dSiKpRVgw8+iSiuXatZwy+DnrBbPAodQW2kLpWjrK9twVx62Ha1cIjc/zztPHAMjO5Xj728dINi4pIVwJIQQ3b+4kEQ7w+OHTnB2dpmRaBDSVrW0NfHDvZra2LRGPg9EAWw5uvOo8m/YuGb3sufTE6uiOpvCA+LI8Z6opTqVk0tJdT2uPv1/Yur+bSCJUS/qrmsJtj+5FkmUkSeDJEr27O/FcX1FQANFkiAd+4NZqJHrpOtq76rBth699/k1cx+OeD+6+brkRVVeZGp7lhX96A1X3c7z3fPJmlHUkVjRJ4Lnz4Fm49hCyXk2Mi2DVOxYgFDx3BnEF1cmQtatCGYv0qnzJD7IEdBVD88vfrCoN7f8GJKEQ17pR5QhBpQFZBHA9CyEkPM9BFkuCbHV6lJ/c+EH2Jnv5ysirnM4MXrUae3jMmTlemT7JGzNn6AjVc3v9Tu5o2FkL6iyOMVmSuHPDBl4fGqJgWWxIJjk7PU1IVdEUhbpgkIiu01tXx+eOHeM37rnnulxmSQgiaoC0HiOorKSiKUKmL9rGQGGC2fksAsHGSAs94ebrfmbXNE6rYhGMBjGCOsVsCbNs1nhoju34OkKKjGM7aNUIoCxJbG9vZEtrPYWKhWU7qIpMSNdqhfCLuJ7NsSpkmoKr7weiiTDRG1bu9VRdpXvb0qrS0t1AS/dKNyJ+RSFF9/b2q84thKCnr5munkbeevm9u3lmxWJqeJbWjU1su7GX177+DqVC5Sqy+nJIQkHW9uN5WSS5GURi9QKAVdTzpatkSPyVwcPnrIYMFZaJU7lVnub/DcjCIKJV34kcX/dYIQSGrHFLehs74l0cnjvP0xPvcDozuKISahG253ApP05/YZJnJt7lgy2HuKdpD1HF31LZrk/839bQwHOXL3N7VxfbGxqoOA5t0Sgt0SizxSJnpqd5cNMmXuzv57GtW1f1hlbek1TjlfpidSuv6etjr5OzilUWjk29EV9VKX8tXNM4060pYnUR3nriCBfevcz+B3cTTfmFAqODs4wPz5Gqj2CWbbbsWerD6LgeY/M5BqfmKVSLENrq4rQmYyuoZFdidjrLwKVp0g1RXnvhLNMTGXr6mrj5zs2Eo4FaKVY+W+bNV85z7uQIuqGye/8Gtt/QWZO7d12XwUvTvPHSOaYnMzS1Jrnx9j6a2/yQfD5X5syJYZpbk7z9+kWGLk/T0p7ilrv95jTvhe85P5vn9RfPcfnCBJFIgH03bWTTthaOv3SW8+/2U8gUGTw94qdfrrH3dTwPIbetzEXaEyyU3wRkIvoWStYQshTC82wsd56Q2k1Y2+RXxFyx95EliWLZojEZIWRoTM7nKJs2QUND4spKGH8i/HDbzbQFr6fdw+qYqowTksOEFH8ClIVMa3BlfxXP85gzp9Flg7DiT7yu565Y8YQQxLUwdzfs5sa6LVzIjfLS1AnenjvPeHluRUR48fODxSn+4uI3ODJ/kR/pfoDOkO+lma5DplLmob5NxAyDu7o38NzgZU7PT7EhmaRs29zT00N7LMap6Skqjg2CmojbWjBknaHiFDE1hMzSuHaq93J7/S5mzSyykFgw89e111zENY1TkiUOfXAvW2/c5AdK6pYGbqVkMjft94Lo2rSUU8yWynz+lWN8/e3TZIqValrZI2zo3LNjI//qjhtIhlffe10+P8ln/+vXaetK09ZZh26ofOlzrzFwaYof/um70Q2VYr7C//qTpxnqn2Hb7nYKuTJ//gdP8qGPH+CBR29AkgSnjw3zZ7/3BE0tCdq70pw8MsgbL53jZ371ITo2pJmezPCnv/1NGlsSNLUkiMaCPPfEcU4eHeTf/D8PE73G3hD8AbYwV+C//863KOQr9G1vZXYqy3/7jcf51I/dwaHbNrP5QA+nXjvH5PAs3TvaCayzagKrupm2m8VxiyBkStYArmdSqFxCkSJEtC0UrMuEtU0UncpV3AxdUqmYNsMTWUzLJlMos6HJ5+YqkoR+RQWQJCT2JDeyP7mJ1eDh1YJOy/92qRoWgldmnqYt2El7cANimWKg4zm1Yzw8Sk4RWci1lXyweJGIEiep1dVyjS4uEhJBWWdnfAPb413MVDIcm7/MS1MnOJHpX1F2CGB5Dq/NnCJnl/iVLR/HdiSGzXkCAZXL5RnGxhYIKRq7mhrJWmUmKhnKjo2hqJxaKDFl58lOFynaFs2hKHXr9NuZKs8zb+bpDbeu2EuqkkJPuJmUHuWN2dPk7RKPtNy85nlWwzWNMzOd5Wv//UnKhQqlXBkjpPMDv/YxwvEQHT0NpBqi5DKlmiar7bg1w/zg3i3s7GwiqKmUTIsTQ5N87fApPDx+5sGbVnUbPDxy2RL7b9rIQ9WWeRs3N/HXf/osdz6wg01bW3j3zUucOTHCv/v1D9PZ04DruHzpc6/xjX86zP6be4nGg3zl/7xBW2cdP/0rHyAUNsjMF/n9//xVvv6Ft/jMLz4AQKlo0tldzw/91N2omsINh3r4vf/0FU6+O8ih2/uua/V85dnTzExn+dXf+Cj1TTHMis1fffZpHv/Cm+w+sIHpoRmycwW23djLyVfP0dTVQKJ+7ZB91irg4tXaPMBiZ/EmBBKOV8R2c+hyHYoUQ5OT2G7WnyjMwlW0p4gSoCkRoz4aRZElTMuulU7qkkb0itYKtuewYPpibVfev+M5nMq8S4PRwtnccTZFtpG3sqiSxqXCOQzJYFf8IK7ncDp7lEv5s/RFd9BotHBs4TDz5gz1ehNbY3uYqUxwKvsuO+P7iQGTlTFemnqShJamL7qd7vBmzuVOMF4aJqHVsT22F1VSkRE0GAnuadzDLeltXMyP8fTEO7w8dZJ5a0lkzgNOLPTzT0Mv82DTTYRVjaCioUkyU6UcmuRLcGbNCprk09hMx8bxPDRJJqIZWK5TVW9c2/VvDCT99NEVJYYSosqIEXy07TYs1yalv7dUzXWU78X4vl9+FDyPUqHC03/7Qq0ErZAv8+6rF7Eth6b2FDsPbGA2V+CZ4xf4wTv28sj+rStqaPd2t9GSjPI/nn6TXdvDNCUjqJJK3i7UtGM8zyMQ1Nmys60mzLV5u//vgYuTbNzcxNG3+5EkwcVz4wxcmgIgny0zN5NnZjKDYzv0X5jkkz96G6FqyV4sEWTvjT088eV3yGb8Tl6KIrN9T2fN1dzQ20iqPsL502Mcur2Pa8GyHI68dRlVVTh5dBBxTFR/bjMxOs/CXAFJElimRSlf9qPP10ilTFcyOJ6zQs3AUFrRlaXooefZLCrAC2Q0uR4Xj4ny3FVubb0RR0LUem4u125SJJnmwEohLcdzGSnO1DrKLYeEoGDnGCpeYrI0ii4ZBOQQFxfeoi+6k+HiZS7lz+Di0mS0ktIbOL5wmEQ6xXDxMj3hLTQF2hAIUnoDYSVKwc6BDmm9gaZAO5si22gJdLBgznIq8y674gc4tnCYBqOZlsDStmmxfnZbrJO+aBt3Nezmr/uf4tj85doz8PB4efoEDzbv59bGDbWxOJCboz4QISCrOJ5PyrBdl4vZGdpC8WpHOf8skhBMlFZvTgUwWpzB8mwszwaW0m4uHqczg2wIN3FioZ8ZM8M9jTe8p7LMa3cZq1gMnhnBsRzMssn8ZKbW4jsYNkimIzS0JGpvMlc2MW2Hvpb6q8SiJEnQ25zG9Tz6FyaIRmVM18L2HCQkn0wsBIoqrWjdpxsqmq6Qy5ZwHJfMXIGFuQIvPnVyxey+bVc7RlCjXLKolC1iidCK38fiQcplk3LJrF1PcFkRuqrKBEM62UxxUYx+XdimQzZTZHoyy/NPnljxuy0729E0hWRTI8VcidnxBfbcuZVoYv2AwERVJ0hbZkRCSCvcQ65QiRNIFOwyQ4WpK34u6AitLckpIeiJNCMLaUVg6Gx2mLJj1ojSy88YUxNcyJ+mJdjBaGmQ3fGDFJ08C9YsUTVOUk8zZ04TVmJElCiOZ6NKGjfW3cWF3CkmyiPclr4fWSjIixOMEMgoKEJBkVRkoWC6FUpOkVlzmuZAG0F59ecmhEAVCjviG/iFTY/xX059jvPLKozmzBwXciN0h5sQQmCZNh1GHFVdzI36C4AiJLYlGmvnvF5sj3eRt8tXGZ3l2rw7fwFDVjmfGyGqBbmcH2dHfMN1n/uaxlkpVjj75gXMsoUQgkMf3Eu0Wh9qmTbpphjNHUuzb1BTkSWJkdkFtrTWr7hRz/MYmfWlLFsjaTZFupGqA2O8PEmDnibHJI7truBiWlXVAiOgIUkSRlCju7eRX/6Nx2oR4toNKTJjw3OoqkypsFJ3plQ0URQZVVOoVHwVuUVDBZ+tXylbBIJXpyRWg6xI6IbKzr1d/NS/e3CF/IioXsvw+XECIYNbHtm39omWYbI8z3hpnoQWufbBV3xupLiyoCCk6HSuE4AQwmd9xNQQc8uYJJfyYwwVJ+mLtl91fFxLYbkWrYEuRooDpPVGNka2UnHK6LJBSI4gC4WL+dMMFS/RZLRhezZDBZ8xY3sWLi5T5WEmyiMUnQIJLU1SqyOmJjidOYrjOdTrTbQGOnE8G00yMOT1VxwhBK3BNHc27OZCbrTmPziey0TJL5vE85gYnKGQKdKzqxNVUygXK+gBDceujrGgjmVZuK6HXiU7rDYY/JYMkNQjq76rxUDb81PHuCGxkZxdqil5XC+uaZyaodG5pQ3L9PNMZtni3WeOEwgbNPc2c+boEJm5Ak1tKVINUeqiIfb1tPI/nn6Lkmmzta0BXVWoWDZnR6f43EtH2Nvdyq3tOzEUtRZ9DYQ6qhKLE5SKJv0XJumuBpmGB2Yoly3au9JIkmDz9ja+9PevMTm2wIbexto5HNvX6UnWhWluT3HiyCAHb92EqimYFZuTRwbp6K4nGguQz5WwLYcLZ8bZf3MvkiSYHF9gZirL/Y/csBaFbwVUTWHz9jZef/Es83N5GpriK64F4UeNL58c9tktmkKqKb5uz5S8VeKdufP0RVtXTaivhkWy9HKqFkBrME3LGl2oF9ESrGNLrGMFD3PezPPt8XfYEG66ijJWrzdxT8PDhJQI9zU+SkiJsDdxM/PWLAJBWImyJ3EI0zWxPJOEmkISMl3hXky3QkxNoAqNkBLhQPI2QGBIfo5wR3wfc+YMATmILhncVHc3C9YcmqShS9duOymEoMGIIwt5BTsnm8nzzHMv+y3nwwbj/VOMX54i3ZpkbnyBQMRgbmIBVVdp7Ewz0T+FYzscfOgG4ukoipCuEk2rOFbVlV0dapVdNFleYFusi5HS9HuecK/dK6VQ5vnPv0LLxiZc12Xg5DCbD/aSaIixYWcnfTvasKylEL4qS/zQnfswLYc//tareJ4vLmU5LpKAWzZ38aN3H6gZ5uJDVWtK5AIhCZ746juYpo1uqHzry2/Tt62F7k2+IR66bRNvvXKe//67T3DbPVsJRQymJjJ4nsdHPnUjgaDGwx/bz//6k2f4h//5It19TZw5PsyFM2N85pcerKlsS5LEGy+dJRjSSDfEeP7J4yRTEXZWe3tmF4ocfvUCmYUiw4OzqKrM1z7/BtF4kL03biQaC3Dngzs48tZl/uQ3v8mh2/vQdYXx0Xki0QAf/Nh+wvEQxVyJYy+fJRA2OHD/znWN08Xjucmj3NWwm6bAtZkYnucxWV7gqfF3VpS7CQT7kpuIrFNkD340997GG3h77nwtUuzh8czEu+yMb+DW+u0rJglFUolpfgFKXPM9JklIpPXGZceECcKK4FRSW5maiakJYupSEYXflFelXm+q3bMuGzTI15+09zyPifI8zjLDFAgiUqCmhzUzOkvv3g3MjS8wemGC3r0bOHv4IrIi072jnYvHBpkYmCLdkqzVeAdk/aqi/YxVYKq8QIOxugCdEIK2YD1twXo8z2ND6NoVR1fimsZZzJXo3N7Ow5+5D9f1+PIffYObP3yASDJCIecHOSpli/HhOeqqrbXroyF+6ZHb+ND+rVwYnyZbqhAN6PQ2p9nYVIehrlcL6RGJGNz9gZ0cf3eQ+ZkcPX1NPPJ9B2v7w0QqzE/+uwd5+utHePPl81iWQyIZ4tDtfUiyhOkW2HuoB0kSPPut45w+PkwyHeYHfu4g2/a01r5b02TufXg3g5emOPzqBeobY3ziR26jrhpNLZcthgdnMCsWu/d3ATA9mSGzUGT77g5EPEhTS4Kf+w8f5NuPH+GVZ0/jui519VE2bfUpV/G6CD27OpmbWKBtY+M1idbgS///w+Bz/HjPQ4QVY10DLToV/nHoBS5f0cejTo9yS3rbNVdfIQQ3VFMnV/Iw/+LiN5GExKG6zddF/F6E53mUHZOBwiS257At1nnNz9iew7vzF2kykjQFku/p+xa/c6w0y4uTx1eExAKyRleokenMKI5l07e/Bz2gEwwbqJrK6MUJNu/vwbZdYukovXs2+MoVtkOkSgwPyDqNRoL+wtIzzttlXpk+SV+07ZrX+p2qJVzzCUSTEWbH5vn8b3+1KhXpEYwEqoUARcaH59B0pSZo5Xoe+VKFoK6xs7OJnZ0rZ4xFiczFkL66SlNZz4Ptezq5+6FdOLaDqikrWioIIWhoivOJH7kds2JRsObw5DIBLUzeHWOucomk3s2m/UG6dx/AtEwCWogSUziihEpVscGDzu4GHnpsH2ePDGCVLTZubq5dT31jjB/8ybvwPI9LJ4Zp6kwTigYo5koUsqUa66a9K80P//TdWKbta7OqSk3wbOD8GBePDtDe18zhp08QTUWIp9cPqbt4fHv8bSquxcfab6Mj1IAqrpDwxGO8NMcXhl7gyfG3cZeF+yUEdzfuoSvc6EtyXkNEOyjrfKrzLgYKkwwVl4JK4+U5fu/sF7mzfhd3N+6hI1RPQNaRxZJQuIefYrFcm6xVYrw8y5nMUE0B4e7GPWyNdVyTI2m7Dv84+AJjpVm2xbs4kNzExmgrdXqUgKwjraLk7r9DX5H9bHaIzw08z/kruKnd4Wb6Uu003hxh84HlZaL+St6+eWVrh1TT1SuhJinsSGzgjdmzKyLBT4wfpi2Y5s6GXQRWkQFdrlQP1J7b9eLajYwSIT72iw8zct7vz9ne14IR0qmULUIRg217OwmG9ZpGT9m0+ey3XuUjh7bTnIgiSxIBXa1SyGChUOZzLx/haP8YyXCADx/cxv6e9mXaQksXr2kKrFOHKkkCI6BRlEpkzEGKFYOw2oBAZr7Sj4eL6znIsorjBRBI2K6JJ61MNyiqTGNbinPv9mOWLc4cvowkSzR21DFwZpRUY5zJ4Vmsik08HSE776dIJEniwrFBAmGDps46Lh4fJhDS2XnLpqWX4PnkAUmSKOVKXDw2SM/ODmJ1V+8/QopBb6SVEwv9WJ7DMxPvcnT+EjviXWyKtpHWYyhCrqmgvz13vpr2WHk/W2IdPNx0iHLWZGp0nkg8SCzpqyVKksBzvVoTqMVSzJ5IM5/Z+BB/ePbLTFUWaufKWkW+Ovoaz04epS2Ypi1YR0KLoMkqrudSsissWAVmKhmmKxnmzRylZYUU67XMuxKO5zJRnmdiYp7nJ48SV8M0B1J0hOppCvg816BiIAsJ07XJWAXGSj5Z/EJujLxdWvk8ZYNH226iMZGicf/qvVcAKqaNqkhXKRIuQgjBwdRmvjby2gq+bNYq8qcXHufVmVPsinfTGEiiCJ/LWnQqZK0iC2aeOTPHpmgrj7TetCJ/fS1cV6+UaCrClkMrK0bymRKvPXMKRZVp7UrT1J4i3egz7i9OzPInT7xGpljGUBUeumEzD+zpQ5ElvvHOGR4/fJob+zoYn8/xO199kd/5gQ/Q01itWlEkQhFfT9VxXDzXRVF9FXTbclBU+arZJ6ikkIWKKgWxvQoNga0IBJZbRhKSb6CShuUWkatpCEmSCEcMlGr+bzEoNzUyR7lYwbYdhi9MsGFrK5dODGObNoNnx3joh2/Hthwmh2dr3z8zNk8pX2ZhJndVT5ZEQ4x4Osr8ZIaWnkbKhQqVNaQ7O4IN/PymD/PXl7/Ni1PHcfGYrmR4dvIoz00eqwUlFrtzr4a2YJof7/kA3ozL8MwkxXyFieFZAiGd6fEFAkENI6jjeR5NHXVcOjVCKBJg72197E/18XObHuXPL36D4eL0ivPm7CKns4Oczg6u+r3faziey6yZZdbMciLTj6j2kVm+Arueu6YSvSFrfKT9Fm6q21p7H5btUKpYGJqCWyXQK7LE8OQCLfUxZMnFsl0MXcFxXMqmTUBXURWZtmCah1sO8b/7n1ohPF1yTF6fOcMbM2dXBI08vGpdczUWI1Urod6Dh/sdt1pOpCNsvaGT2aks8WSolsj38LAdh+lsgX09bWSLZf7qucPUx8Lc0N3K0f4xburr4D88dhcLxRK//Hff4tUz/bQnfFdvY18Tv/SfHyGeCDI1NENmJkvPrk48D4bPjdG2yQ8QeK6LqqnYlg22RsRoRpIlLs3OMZXPs6+tlYCy+mYdoKklwb//zY9SVx/FdT1mxuaZm8rS3tfsp2A8aOttZGJohlAsgKqppJriTA7OUCpUmJvIEImHSLckUbQsAjBLFtHkSiWERH2MWz+8/6rvX03AuC/aRmuwjp/qfRhD1nhu8miNjeHhrdoWYBESgr5oOz/e8wG2xjrJWHnmnCzJ+iiqJjMzkSGaCJGoixCKGGQXilgVm7rGOMGI4XNnhV/VUm/E+YeB53hj9iyl62yDsBqCsl6NUF57RAohfNeQ1YX7/Qnp2quwwFeT+Fj7bdzfvLfWFt51XV471s/kbI4NrSlmFgpIQrC7r5Xzg1OkYkFePz4AQEMqQiZXYno+z617umlKx1AkmYdbD5G1inxt9LUV3sHi9a33fr4TXJvP6fnE3itXK1mW2LCpic7eBqRlOrTZYgVZkvjlR+5gW3sDtuvyZ99+gzcvDLFnQwsl06KrIYkiS6TCQXZ3tXDy8hjPDGSp5Ct0bG5BC2i8+sZ5WnubGLs0SaVQoa2vhUtHB6hvS/HKV99CUWQ6trQxfnmSuYl57vi+m4nVRXi5f5AXL/ezq7mpJha2GjRdoa2zrnqPHm0bG2nqTBMIG9Q1xQGBZqiUCuVaLlWSBLbtYJsOnZub0YMasizT1tvIsZfPYQQ1hs5N0LWlFXGNSiDwo5yLHboUIfkiWAhSWpSf6f0Q2+NdfGP0DS7lx6+iTNXeg5Co1+Pc0bCTDzYfpE6NIYB4XZh43VLivnVD/TWvx78mQU+4mV/a/DHemb/A0+PvcDo7yLyZvyaDRRJ+DWyjkWBbvJMb67awJdZxXYuFJqn8SPcDdIebeGfuAqOlWYpO+bpYMwKBLik0BpLsT/VxX9NeukINV+xzBblChWjIQJFlwkEd07SpmDaKLFGx/NWwr6uBgdFZMvkyqViI2LJa6JBs8IMb7qUr3Mjjo69zOT9O2THXddwlfIZNuMqQeS+4drTWPExA3V5l4K+EkASW1Y/kaWiKn7AuW74SX13UvxhFkmiIhTkzOlULBi2/xkQowMlCGRGNEAgZzE9mUHWFufF5oqkwF4/009iZJpoMg/DbKOBBe18Ls2NzLExnSbfWEQhfOw+2FoQQK2hcy1vah64oVJcVGf2Kr/I8iR039VLIlGqr0HIs5EsMTy8ghKCzIVHrS3pH/U56ws3YjsvkfJ5t0Q1LPFfF4P6mfexP9vHC4CleGjnNnJMhFteQJAgoOg1Ggs3RdrbFOmkOpHAtjzPHhpEkiQ2bGldUP73n56Ho3Fy3lb3J3mqnrREu5caZLM+Ts0uYroWEQJf9+tw6PUpbsJ7OUAPNgZTP0hAS4ODaF5HkZoS0tiymJATd4Sa6wo18pP1WxkqzDBWmGCnNMFPJkLOKlBxf08nDQxZydWUO0xKsoyfcTFe4kZQWXTXwYjsupmVj2QJdlamYMkZIQVN9F3d6Pk9nc5J42KA+GWFmoUChbLKQKxE0lrimhqxxT+MN7Etu4sTCAJfyo4yUZhgrzpO3y0RUg4Jl0RWuo96I02AkaQ+l6Qk3rWC4OK5LybEIKdqaRnvtPKd1AdMeIaBtx/NMTHuAoLaXsnUGITRAwvPK2O4CAXUbsaABAp48ep67tvWQr1R48fRlpjN5vvj6CYZmFmiri+O6LgjBbK5APBKkpafR78HhuixMZWntbaK+rY7bv+9GzJLJeP8UpXyZycFpWnubiaWjWKbN8LkxirkS5UL5KikSz/PIliscHhmlO5WkMxHn2PiE37lMwIWZWTRZYWtDmoZwuBaBLFoWZ6ammcznSQWCbGmoJ6JrjOdyDM4vsK+tFctxeH1wmJZYlN66FBnH5GIlw750mHzFRBKCkO6/1LlckeePXeTV04P88kdvZ89GP53THqqnPVTP1EKeb7zyAgeSNnXL5gIBDI/keObJMdLxFjqi3Xz60B4iAR1JCL9nyDKvpuL5lS2KKpifya0wTtt16C+M4HgOBbvExkgHESVE3i5yMT9UJQO3E5QDDBXHEcBkZZbmQANdoUaCikpC0xF00R1ux5B1xkvTjJenqdeTdIfbWLCyzJtZBgrDBGSDjZEOZPKUsr+BEf1FFGnbumPNL+MTJLQwCS3M1lhHdUJ3sav7y8XtgKDqeUjympHclfBIJ8KYlkMsEmDzhqW87P03bV5xpKGrzCwUcF0X7YqcdMm26M/OEdMNmo0mdsS7mSzl0GWFkfwCEU1nMDfPtmQTqpCRhN+car5SZq48i+26BFWNhUqJom3W2oqshusICKnoag9l8xSKnMa0x4B3AY+IfpCSdYxc+RWSoU8CgkQoyIf2beEvn36LL7x6DNtxaU3FePTANr717jm6G5L0T87xxddPYKgKz5+8zCdv3c3WAxupuCVkIeN6DpJQyNmzeJ5ACIOEVk9r7z0rrk0zVKaGZtbs2rVQLvPfXnmd2WKRX779FjzgC8dPMpLJYigKQVVlLJsloKr85gP30ByNkilX+IOXX+Xc9AxNkQgzxQIN4TC/dNvNjGay/Onrb/HZdJqZYoFff/Z5bu3q5D/dfQevDQ7zwqV+2qIxTo5PEtV1grpGQFWYKRb4xF17ODs8jXNFq0LHdYkEdH7ioUM0JpciuJ7nYTsub50bors5xU89fBOSAL2aI/Y8D9f1cHBr7S0sy2ZyfIFEMkzf9pXdrU3X4ssjz7ArsQnTtbmQH+JDzXfwzfGXiKsRik6Ji/khHmm5k+en3kQRCl3hFkzXouKaPD76PN3hNgxZQxKCeXOB56beYFushzfnjlF2y5TsCi9OH+bm9A28OXccBPSFk75Amefg1SpqVqaF/F2mA0iAdNXvJOGiCX+4isWgmLf4GZfF+lj/Z8vdYKl6jIyqyNy8e7Gu1a5ey9XBRfDVIm67ofuqnwPMVYocnh7mUEMHl7NzVBwbWYiq1q1ERAnQGZaYKObpiaaYKuUpOzkuZGaIajoBRaNg+ZN3TFvf27umccoiQsW6iCo3YzkjyFIUTWmlYvdTNI8iSUEixh1Yzjia0oEkyXxw7xZ6Gus4NzZNUFPZs6GFhniY+3dvQpVlnjt5ib9+7jDFisW+nlbu2NZN0clwKX8ERWg4nkNUTVKwM0TUFHl7HgmJpL6yWiReH+OWxw4gxMouwZIQZMpl/sebbzNfKvErt99KczSCh/8CR7NZ/vChB+irTzM0v8DPPv5N3h0dpykS4ZmLlzgxMclvP3gvnYkE0/kC/89Tz/DF46f40NY+HNdlrlSkf26e9nicqXyBgmlxaXaOrmSCbLmC53nMFIpEHYfhcsUn7a4xs799foSn3j2P47h85qFDNCR8A+2fmOOrr53izXNDSELwO194nu1dTTxy41bwPI5dGuO5YxcpVSy2dzVx7w2bUFWFzu76WsrkSgQVgwPJHUhC4m8GvsZUZY7TmYt0R9pruUrbc5CExO7EZrZE/QHqeC5twUbGStNsjnYRlA1OZi7QGmzgUGoXYSXI6exl2oKNbIp2cWNqFyWnzExlHsJ+NZFtvolZ+hLgoQU+jKzuQgjw3EnM4pdxnX6EVI8WfAxJ9gs+PHcWq/Q4jnXKH6z6zaiBhwEPu/IaVuVp8CwU7SBq4H48N49Z+iIg8NwcirYLq/w8qnEPin4beCXM8jdxzHdB6GjGg8jaXoRYyqFnS2VeOTtA2NC5aVPHVeQNy3UIqzqm67AhmiRlBAkqGqZj0xSM0BAM00AYy3WJqBpOVW2iKRhFEoKAomK5DgXLJKyuryV07SKEwD1VipKGxx4ABCq62geeW6UuCRYV4Y68dZkLZ8bYsrOdjx7avmJmWiRYb0+k+Lkb99PR20BDLExQVyk7DnV6K5ZboewWCSlxFKGhyQFCSgxF8m9kZirL+dNjHLx1E5IkVi2FczyXP33tTcayOf7LfXfRFInUVhvwZS/76tNoskxTNEIqGGSuWMLxPN4YHGZHUyPdySSyJNEcjXBjRzsvXu7n4zu3EzV0xrI5zkxNc7C9lSNj44xlswwtLPDApl6665Kkw0F0RcF0HC5Oz1KomFjO6pG8LR0NqIrMH335JYqVpaBPfSLMw4e2UKyYKLLER2/ZSTjgr1pnhqb4m2fe4eFDW4gEdP7xxWOYlsOudB2u41EprR48slyLolNGFr4rqEkqST3GrekbiCphJCGhSSoS/t+L705C4rb0PmbMeb49/gqqpBCQdabKs9ieQ94uYUg+WcCoyl1KSLWwq+eVcO3zaIGP4linKOd+j2D890DEKOc+iySn0YIfx668QTn7OwTiv4UQOuXcH+J5ebTAo4BXjXtI2OYbVAp/jhb8NEIKYxY+h4eJoh3CKj2JHvoUZuVNPHcCWd2JWfwiirYfs/gFHOsUWvAjuM445dwfEoj9GrK6RA80VIVY0OCFU5c5uLGdKzvDNwWj3NKkEdMMdHlp7AUUldgqXQiaQ2sUnFwHc+y63FpRzQ2KZTIMAm1FhNx1BDNTWZ574ji33LWFppYErutRLlVwHQ/bdoglglimg/Cgb0MjjfVxwA/yFDOQCHQSDOmUSxau5SKVTMIRA11XcRyXuZkc05NZLp2f4OCtvawVoj87NUOhWqifr5hwRb4/puvIi3W91f2KV93PFCyTxmh4Rd1vRNcoWTaqLNEWj3FueoaB+QU+vmMbA/MLnJqcIluu0JGIY6gKRrUrmOd5RA0dx3VxndVjepGATksqepV0S9jQ2dCkEQ8H0BSZDU3J2gTzyql+ZEmgyjJl0yYdC/H66UFu+VA7ZsnCCKwuhWK5Ns9PvUXFMdkR30SdnmBHrJcXpt5CkzQ2hjvYm9xKSAmgSEsKBRW3wtOTr1F2KkhCIqHFSKhR+guj/NPwtzE9m3saDjFRmsGUfdfVkDXkxRVJqKjGw8jafmR1M1blWRzrLJLcjGMdQdZ+CM/NIOQ0Tukirj2EEEEc6xTB+O8iKT3L7sLFLj+Dou1DNe73f+SVqBQ/j6LuQkhRFP1mHPsyklSHot2IVXkR153DKj+Foh/C8wogDD9WYh5ZYZyaovhSOvKS+5yvmLx6doC5fInt7Y1sbWtgdC7DG+eHsByX/T1tdDckOTs6zVyhyNhclnQszN4NLbx2bhDbdVkolDjU20FnenVNqNXwHec5r4Rp2pw4MsTkeIbhgRma25JIkuCLf/caybowmqZw9wd2UsiXeeHpk3T1NNDUkqBSsfj2V49QyJepVCzuf2QPp48Nc/HsuE+UlgQf/sRBXn/xHEP905imvarLthydiTi/9eC9/NVb7/CHr7zGf77nLtKhZdHmtfiNQtASjTKayWK7Lpos47oeI5ksdcEgAUVlYyrF64PDlG2bzkSC7mSSt0fHkCWJ+vDKaKQQoqZ9uqh+993C8/wA03yuxNvnfflKRZY4uKUdw1DpPzuBJAmaO1LoV9DpImqIO+sPoMsaMTVMxSnTG2lhd2IzRadERPGv/5b0HoKyQd7Os2DN02A0cnPdbopOibgaQ5VkHM/hwaZbyFg5grJBSPF7jChVLuaW6AY0SQPKCGSEVPVe0BAiiOfl/D9uHsc8hisMwEM17kBIMTx3zl8YpPgVtEMXz51HUjqXJlApCW4BzzMBtfpHBmEsvWuvjOdlcOyLeK5PW5S1XUjKEoF71ecNPHn0HPmSSUc6zudfO8bPPnATspBoT8cZn8/x5TdP8PMP3cK5sWlODE/wwRs2EzF0ypbN4++c4ZF9W4gFA3z18Cl+5v4bUa4hHLaI75lxGgGNG+/o48KZMe59eDeRaIDZ6SyFfJmP/cBNhKMGsiwRDOls3tZKZsHXS50YnefEkUFuvL2PMydHOH1smHyuTPemRg7d1sff/PnzTE9mOXNimI/9q5uZGFvgrVdWKuF5nodtu6jVjtsBVaUhHOanbzrIrz39HH/6+pv84q03EdLW9/ElIbh/00Z+/dkX+Prps+xuaeLy7Dwv9w/yQ3v3oCsy3akkf/fuUTbWpUgGA2yqr+Mrp06zq7mJiPbP3wBHCGhKRqlYDj/18E1oiuyLwuNhmw6JujBmxaaQLaOnl4xTEoKkFiOqhglVicEZK8NUZYIuJcT53El0yaAt2M7l/EW6wt1Yrsml/EXCSoSsPc9wcZCu0AbGSqPk7Twbwj2MlUbpDHVRcgpcyl8grEaYrkwhEGyObqVJD+F5Fq4z6dfiugU8N4uQUggpiZDT6KFPIymLbRP8AI+LW/3cOEJapL15gIwkt+LaQ3ieH0Ry7WGElECIKwMsy/i1IoiQGlCNe1CNDyw73/oTveu6HB+cQFNkChWTkK5i2Q4XJ2Y4OzaN63rMF0rViiPY0d7Izg6fWTObK5KKBNnf00auZHJscBzH9VhH324FvmfGuRZCYZ1gSF9RuL4cruuhGQqxRIibbu+juS3FW6+cJ5YIoagyiiLhun6RlixLqKqMZTucujBBXTLEyPgC8WiAofF5bty9gbCukQr5Gq/1oRC/cOtN/NbzL/H0hUs8vKWPeCCAJi8vooe6YNBPrwC7mpv46UMH+OLxk3zp5ClUWebjO7ZzX28PorqyRg2dbQ0NaLJMVyJBWNfZXJ+uuUJX3l//5BwjMxkW8iVOD02iKjIbW+rQVYWLYzMMTs6TLZY5dnmMQtms/W41CCG4Y2c3v/+ll/jzb75OR32CmWyBHV1N7Otto29766qf0yWND7XchSEtNtP1CMhBn3fpWqS0OiQh4XkeYTXCvDlHS6CVtF6PIRvM5KcoOyXydp68nafBaEQgqNfraTKaOZE5RsWtYFZMQnIISUiUnRIQAiFhlR4HL4tjnUNIcWR1M0KEUbWbKef+CEW/CXDAs9GCn0CSW1D0mynn/gDVuKN67xHUwMOogQcpZX+LSuEvESKCVf42WuhTINaZHKUgWvAxzOLnfYMXITx3Fi34EcQyWprreliOi+P6lW66qtCSjNJRl2B/Txu24xA2dF44fZnHDmzHtG1G5zO1d3NlY2hZ+AqH3wkx5XtqnAIwAit5moaxtDf1PI+BS1OcPTlCqWhy+vgwnd31dHSluXh2DE1XaGxJoOk+C0XgS5QEgjpd3fU88ZV3qlqrHoVihWBAZXQyg6rKhAxfE/f+3h5u39CJpvgph+5kkt9+8D6/FYIQ/Nj+vYhl0VNdUfjVO28lUHU/FUni7o3dHGxvo2CafoBA12tF0fXhEH/40IOEqjIiLbEof/KhhwhrqyeTXc/jwsg0lyfm2LepjVyxwpFLozQlIyiyxKnBSSbnc9y4pZPRmQyFskl7Ou6nTIDd3c1XRQzb0nF+6SO38erpAUZnMqSiIVrrYrVnvuq7qZbHLUfWWqixStJ6PQJfEkWTNCJKlIgSQZU08naOoBzCCAQwpACu55KzsiS1FGHF31+H5BBCF4SUELJQEAhCShiBgR78BJLSgW2+jZAbMEI/gKjq8erhH8OuvIptn0KgImu7QWiAjBH+CazKKzjWaYSQkLVNgISk9BKI/ipW5QU8dxYj/JPI2g14XgE1cB9CBFC0vQgpihBxVONeBDqqcR+SVI9tvoXr5pHVjQgRW/FMLkzM8PTxCywUS3zl8Cke3NXHB2/YzNffOcPZsSmaE1E+eMNm9nW38uzJiyRDATY2+t2zE6HAinelyBKtqRiyJNBkmZbk9UuuAojVajyX4fopBfizzsJcnlgihCxL2LZDdqFEIhlCSH4wY2G+QHahCB4Ewzp19VHMis3MVBZZlqirj1IqmSiKryM0P1cgGgvgOh4zU1mMgIokSXiyQFEkKqZNOKhTrtiEgtq6mrjv47vHgrnAxfx5VEmlL7IZXf7OK7P+JcJyHMyqgJ1AVLnHYLt+UbyqSCiS72FUbAdF9qPSiizhuH7GQlkWTDJtx996ALbjoMqr5lZXtdjvqXG+j///x5Xj5TslEr+PFfiOjPN9vI/38f8Srk9B6n28j/fxfx3vG+f7eB//QvG+cb6P9/EvFO8b5/t4H/9C8b5xvo/38S8U7xvn+3gf/0Lx/wMkGY/+0jy9MAAAAABJRU5ErkJggg==\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Word Cloud for Non-Duplicate question:\n",
    "\n",
    "not_duplicate_words = ''\n",
    "for p in (no_similar):\n",
    "    d = (str(p).split())\n",
    "    for n in d:\n",
    "        s = n.lower()\n",
    "        not_duplicate_words += \"\".join(s)+\" \"\n",
    "            \n",
    "wc_q_ = WordCloud(width = 800, height = 800,background_color ='white',stopwords = stop_words,min_font_size = 10)\n",
    "wc_q_.generate(not_duplicate_words)\n",
    "print('Word Cloud for Non Duplicate question')\n",
    "plt.imshow(wc_q_, interpolation='bilinear')\n",
    "plt.axis(\"off\")\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "bf78dc78",
   "metadata": {
    "id": "bf78dc78"
   },
   "source": [
    "# Distributions for extracted features"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 52,
   "id": "3e7cc055",
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 389
    },
    "id": "3e7cc055",
    "outputId": "a1469174-f583-452f-a3a2-ab1ba2a2bb76"
   },
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 576x432 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Data visualization for \"Word Share\":\n",
    "\n",
    "plt.figure(figsize=(8, 6))\n",
    "\n",
    "plt.subplot(1,2,1)\n",
    "sns.violinplot(x = 'is_duplicate', y = 'word share', data = xy_train[0:])\n",
    "\n",
    "plt.subplot(1,2,2)\n",
    "sns.kdeplot(xy_train[xy_train['is_duplicate'] == 1.0]['word share'][0:] , label = \"1\", color = 'red')\n",
    "sns.kdeplot(xy_train[xy_train['is_duplicate'] == 0.0]['word share'][0:] , label = \"0\" , color = 'blue' )\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 53,
   "id": "a47fa4a5",
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 389
    },
    "id": "a47fa4a5",
    "outputId": "bcec4a58-d724-4dd7-f304-b7f3a7be4b72"
   },
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 576x432 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Data visualization for \"Common words\":\n",
    "\n",
    "plt.figure(figsize=(8, 6))\n",
    "\n",
    "plt.subplot(1,2,1)\n",
    "sns.violinplot(x = 'is_duplicate', y = 'common words', data = xy_train[0:])\n",
    "\n",
    "plt.subplot(1,2,2)\n",
    "sns.kdeplot(xy_train[xy_train['is_duplicate'] == 1.0]['common words'][0:] , label = \"1\", color = 'red')\n",
    "sns.kdeplot(xy_train[xy_train['is_duplicate'] == 0.0]['common words'][0:] , label = \"0\" , color = 'blue' )\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 54,
   "id": "6966c63d",
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 389
    },
    "id": "6966c63d",
    "outputId": "7156a465-77d8-47b5-a082-0d9e1980ef19"
   },
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 576x432 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Distribution of the fuzz_partial_ratio:\n",
    "\n",
    "plt.figure(figsize=(8, 6))\n",
    "\n",
    "plt.subplot(1,2,1)\n",
    "sns.violinplot(x = 'is_duplicate', y = 'fuzz_partial_ratio', data = xy_train[0:])\n",
    "\n",
    "plt.subplot(1,2,2)\n",
    "sns.kdeplot(xy_train[xy_train['is_duplicate'] == 1.0]['fuzz_partial_ratio'][0:] , label = \"1\", color = 'red',)\n",
    "sns.kdeplot(xy_train[xy_train['is_duplicate'] == 0.0]['fuzz_partial_ratio'][0:] , label = \"0\" , color = 'blue')\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 55,
   "id": "14e23056",
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 389
    },
    "id": "14e23056",
    "outputId": "34a5e19d-e496-447f-de67-5279abc4d420"
   },
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 576x432 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Distribution of the WRatio:\n",
    "\n",
    "plt.figure(figsize=(8, 6))\n",
    "\n",
    "plt.subplot(1,2,1)\n",
    "sns.violinplot(x = 'is_duplicate', y = 'fuzz_WRatio', data = xy_train[0:])\n",
    "\n",
    "plt.subplot(1,2,2)\n",
    "sns.kdeplot(xy_train[xy_train['is_duplicate'] == 1.0]['fuzz_WRatio'][0:] , label = \"1\", color = 'red',)\n",
    "sns.kdeplot(xy_train[xy_train['is_duplicate'] == 0.0]['fuzz_WRatio'][0:] , label = \"0\" , color = 'blue')\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 56,
   "id": "7a79e4c6",
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 389
    },
    "id": "7a79e4c6",
    "outputId": "479e3a5e-4b8e-410e-8a40-28bfbe5648d1"
   },
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 576x432 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Distribution of the Token Set ratio:\n",
    "\n",
    "plt.figure(figsize=(8, 6))\n",
    "\n",
    "plt.subplot(1,2,1)\n",
    "sns.violinplot(x = 'is_duplicate', y = 'token_set_ratio', data = xy_train[0:])\n",
    "\n",
    "plt.subplot(1,2,2)\n",
    "sns.kdeplot(xy_train[xy_train['is_duplicate'] == 1.0]['token_set_ratio'][0:] , label = \"1\", color = 'red',)\n",
    "sns.kdeplot(xy_train[xy_train['is_duplicate'] == 0.0]['token_set_ratio'][0:] , label = \"0\" , color = 'blue')\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "4517381c",
   "metadata": {
    "id": "4517381c"
   },
   "source": [
    "# Data Preprocessing: (Train set)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 57,
   "id": "800674b0",
   "metadata": {
    "id": "800674b0"
   },
   "outputs": [],
   "source": [
    "# Removing stop words:\n",
    "all_stopwords = stopwords.words('english')\n",
    "\n",
    "# Removing the following words from list containing stopwords\n",
    "all_stopwords.remove('not')\n",
    "all_stopwords.remove('but')\n",
    "all_stopwords.remove('because')\n",
    "all_stopwords.remove('against')\n",
    "all_stopwords.remove('between')\n",
    "all_stopwords.remove('up')\n",
    "all_stopwords.remove('down')\n",
    "all_stopwords.remove('in')\n",
    "all_stopwords.remove('out')\n",
    "all_stopwords.remove('once')\n",
    "all_stopwords.remove('before')\n",
    "all_stopwords.remove('after')\n",
    "all_stopwords.remove('few')\n",
    "all_stopwords.remove('more')\n",
    "all_stopwords.remove('most')\n",
    "all_stopwords.remove('no')\n",
    "all_stopwords.remove('nor')\n",
    "all_stopwords.remove('same')\n",
    "all_stopwords.remove('some')\n",
    "\n",
    "def remove_stopwords(texts):\n",
    "    '''\n",
    "    This function removes stopwords from the sentences\n",
    "    \n",
    "    '''\n",
    "    sentence = texts.lower().strip()\n",
    "    sentence = sentence.split()\n",
    "    final1 = [word1 for word1 in sentence if not word1 in set(all_stopwords)]\n",
    "    final1_ = ' '.join(final1)\n",
    "    return final1_\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 58,
   "id": "JfYFGJjmtxzb",
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 1000
    },
    "id": "JfYFGJjmtxzb",
    "outputId": "9108ab8a-efbe-4025-a79b-391621cd379d"
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>id</th>\n",
       "      <th>qid1</th>\n",
       "      <th>qid2</th>\n",
       "      <th>question1</th>\n",
       "      <th>question2</th>\n",
       "      <th>is_duplicate</th>\n",
       "      <th>q1_length</th>\n",
       "      <th>q2_length</th>\n",
       "      <th>common words</th>\n",
       "      <th>word share</th>\n",
       "      <th>...</th>\n",
       "      <th>cwc_max</th>\n",
       "      <th>first word equal</th>\n",
       "      <th>last word equal</th>\n",
       "      <th>difference in no.of words</th>\n",
       "      <th>avg length of words</th>\n",
       "      <th>token_set_ratio</th>\n",
       "      <th>token_sort_ratio</th>\n",
       "      <th>fuzz_WRatio</th>\n",
       "      <th>fuzz_partial_ratio</th>\n",
       "      <th>longest_substr_ratio</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>95873</th>\n",
       "      <td>95873</td>\n",
       "      <td>21712</td>\n",
       "      <td>5041</td>\n",
       "      <td>Why do people often ask questions in Quora whi...</td>\n",
       "      <td>Why do people ask questions whose answer can b...</td>\n",
       "      <td>1</td>\n",
       "      <td>14</td>\n",
       "      <td>14</td>\n",
       "      <td>6</td>\n",
       "      <td>0.214</td>\n",
       "      <td>...</td>\n",
       "      <td>0.429</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>14.0</td>\n",
       "      <td>66</td>\n",
       "      <td>63</td>\n",
       "      <td>70</td>\n",
       "      <td>70</td>\n",
       "      <td>0.184211</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>262013</th>\n",
       "      <td>262013</td>\n",
       "      <td>74232</td>\n",
       "      <td>33958</td>\n",
       "      <td>What are lucid dreams?</td>\n",
       "      <td>Is Lucid dreaming possible?</td>\n",
       "      <td>0</td>\n",
       "      <td>4</td>\n",
       "      <td>4</td>\n",
       "      <td>1</td>\n",
       "      <td>0.125</td>\n",
       "      <td>...</td>\n",
       "      <td>0.250</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>4.0</td>\n",
       "      <td>52</td>\n",
       "      <td>52</td>\n",
       "      <td>56</td>\n",
       "      <td>61</td>\n",
       "      <td>0.476190</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>85654</th>\n",
       "      <td>85654</td>\n",
       "      <td>144574</td>\n",
       "      <td>144575</td>\n",
       "      <td>Who came first, egg or the hen?</td>\n",
       "      <td>What is first a egg or an hen?</td>\n",
       "      <td>1</td>\n",
       "      <td>7</td>\n",
       "      <td>8</td>\n",
       "      <td>4</td>\n",
       "      <td>0.267</td>\n",
       "      <td>...</td>\n",
       "      <td>0.500</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>7.5</td>\n",
       "      <td>76</td>\n",
       "      <td>72</td>\n",
       "      <td>80</td>\n",
       "      <td>78</td>\n",
       "      <td>0.266667</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>19837</th>\n",
       "      <td>19837</td>\n",
       "      <td>37476</td>\n",
       "      <td>37477</td>\n",
       "      <td>What is the difference between reptiles, amphi...</td>\n",
       "      <td>What is the difference between amphibians and ...</td>\n",
       "      <td>0</td>\n",
       "      <td>9</td>\n",
       "      <td>8</td>\n",
       "      <td>8</td>\n",
       "      <td>0.471</td>\n",
       "      <td>...</td>\n",
       "      <td>0.889</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>8.5</td>\n",
       "      <td>100</td>\n",
       "      <td>94</td>\n",
       "      <td>95</td>\n",
       "      <td>86</td>\n",
       "      <td>0.584906</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>105551</th>\n",
       "      <td>105551</td>\n",
       "      <td>11605</td>\n",
       "      <td>1310</td>\n",
       "      <td>What would be the best way to control anger?</td>\n",
       "      <td>How can someone control their anger?</td>\n",
       "      <td>1</td>\n",
       "      <td>9</td>\n",
       "      <td>6</td>\n",
       "      <td>2</td>\n",
       "      <td>0.133</td>\n",
       "      <td>...</td>\n",
       "      <td>0.222</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>3</td>\n",
       "      <td>7.5</td>\n",
       "      <td>54</td>\n",
       "      <td>51</td>\n",
       "      <td>56</td>\n",
       "      <td>61</td>\n",
       "      <td>0.250000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>281943</th>\n",
       "      <td>281943</td>\n",
       "      <td>42867</td>\n",
       "      <td>66163</td>\n",
       "      <td>What is the easiest way to learn JavaScript?</td>\n",
       "      <td>What is the best way to go about learning Java...</td>\n",
       "      <td>1</td>\n",
       "      <td>8</td>\n",
       "      <td>10</td>\n",
       "      <td>6</td>\n",
       "      <td>0.333</td>\n",
       "      <td>...</td>\n",
       "      <td>0.600</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>2</td>\n",
       "      <td>9.0</td>\n",
       "      <td>84</td>\n",
       "      <td>84</td>\n",
       "      <td>84</td>\n",
       "      <td>76</td>\n",
       "      <td>0.272727</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>370194</th>\n",
       "      <td>370194</td>\n",
       "      <td>500727</td>\n",
       "      <td>500728</td>\n",
       "      <td>How and where can I buy Made Easy handwritten ...</td>\n",
       "      <td>I have three cat steaks and four hungry boys, ...</td>\n",
       "      <td>0</td>\n",
       "      <td>14</td>\n",
       "      <td>14</td>\n",
       "      <td>3</td>\n",
       "      <td>0.107</td>\n",
       "      <td>...</td>\n",
       "      <td>0.214</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>14.0</td>\n",
       "      <td>58</td>\n",
       "      <td>55</td>\n",
       "      <td>58</td>\n",
       "      <td>60</td>\n",
       "      <td>0.071429</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>370367</th>\n",
       "      <td>370367</td>\n",
       "      <td>19980</td>\n",
       "      <td>10524</td>\n",
       "      <td>Who will be USA next president, Trump or Clinton?</td>\n",
       "      <td>Who do you think will win Trump or Clinton?</td>\n",
       "      <td>1</td>\n",
       "      <td>9</td>\n",
       "      <td>9</td>\n",
       "      <td>5</td>\n",
       "      <td>0.278</td>\n",
       "      <td>...</td>\n",
       "      <td>0.556</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>9.0</td>\n",
       "      <td>75</td>\n",
       "      <td>65</td>\n",
       "      <td>74</td>\n",
       "      <td>71</td>\n",
       "      <td>0.395349</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>243389</th>\n",
       "      <td>243389</td>\n",
       "      <td>355817</td>\n",
       "      <td>355818</td>\n",
       "      <td>How old do I look to you? Honestly.</td>\n",
       "      <td>How old do I look?</td>\n",
       "      <td>0</td>\n",
       "      <td>8</td>\n",
       "      <td>5</td>\n",
       "      <td>5</td>\n",
       "      <td>0.385</td>\n",
       "      <td>...</td>\n",
       "      <td>0.625</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>3</td>\n",
       "      <td>6.5</td>\n",
       "      <td>100</td>\n",
       "      <td>68</td>\n",
       "      <td>90</td>\n",
       "      <td>97</td>\n",
       "      <td>0.944444</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>77180</th>\n",
       "      <td>77180</td>\n",
       "      <td>131824</td>\n",
       "      <td>131825</td>\n",
       "      <td>Is guava a citrus fruit?</td>\n",
       "      <td>Why is guava not considered a citrus fruit?</td>\n",
       "      <td>0</td>\n",
       "      <td>5</td>\n",
       "      <td>8</td>\n",
       "      <td>5</td>\n",
       "      <td>0.385</td>\n",
       "      <td>...</td>\n",
       "      <td>0.625</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>3</td>\n",
       "      <td>6.5</td>\n",
       "      <td>100</td>\n",
       "      <td>71</td>\n",
       "      <td>86</td>\n",
       "      <td>74</td>\n",
       "      <td>0.625000</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>323429 rows × 21 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "            id    qid1    qid2  \\\n",
       "95873    95873   21712    5041   \n",
       "262013  262013   74232   33958   \n",
       "85654    85654  144574  144575   \n",
       "19837    19837   37476   37477   \n",
       "105551  105551   11605    1310   \n",
       "...        ...     ...     ...   \n",
       "281943  281943   42867   66163   \n",
       "370194  370194  500727  500728   \n",
       "370367  370367   19980   10524   \n",
       "243389  243389  355817  355818   \n",
       "77180    77180  131824  131825   \n",
       "\n",
       "                                                question1  \\\n",
       "95873   Why do people often ask questions in Quora whi...   \n",
       "262013                             What are lucid dreams?   \n",
       "85654                     Who came first, egg or the hen?   \n",
       "19837   What is the difference between reptiles, amphi...   \n",
       "105551       What would be the best way to control anger?   \n",
       "...                                                   ...   \n",
       "281943       What is the easiest way to learn JavaScript?   \n",
       "370194  How and where can I buy Made Easy handwritten ...   \n",
       "370367  Who will be USA next president, Trump or Clinton?   \n",
       "243389                How old do I look to you? Honestly.   \n",
       "77180                            Is guava a citrus fruit?   \n",
       "\n",
       "                                                question2  is_duplicate  \\\n",
       "95873   Why do people ask questions whose answer can b...             1   \n",
       "262013                        Is Lucid dreaming possible?             0   \n",
       "85654                      What is first a egg or an hen?             1   \n",
       "19837   What is the difference between amphibians and ...             0   \n",
       "105551               How can someone control their anger?             1   \n",
       "...                                                   ...           ...   \n",
       "281943  What is the best way to go about learning Java...             1   \n",
       "370194  I have three cat steaks and four hungry boys, ...             0   \n",
       "370367        Who do you think will win Trump or Clinton?             1   \n",
       "243389                                 How old do I look?             0   \n",
       "77180         Why is guava not considered a citrus fruit?             0   \n",
       "\n",
       "        q1_length  q2_length  common words  word share  ...  cwc_max  \\\n",
       "95873          14         14             6       0.214  ...    0.429   \n",
       "262013          4          4             1       0.125  ...    0.250   \n",
       "85654           7          8             4       0.267  ...    0.500   \n",
       "19837           9          8             8       0.471  ...    0.889   \n",
       "105551          9          6             2       0.133  ...    0.222   \n",
       "...           ...        ...           ...         ...  ...      ...   \n",
       "281943          8         10             6       0.333  ...    0.600   \n",
       "370194         14         14             3       0.107  ...    0.214   \n",
       "370367          9          9             5       0.278  ...    0.556   \n",
       "243389          8          5             5       0.385  ...    0.625   \n",
       "77180           5          8             5       0.385  ...    0.625   \n",
       "\n",
       "        first word equal  last word equal  difference in no.of words  \\\n",
       "95873                  1                0                          0   \n",
       "262013                 0                0                          0   \n",
       "85654                  0                1                          1   \n",
       "19837                  1                0                          1   \n",
       "105551                 0                1                          3   \n",
       "...                  ...              ...                        ...   \n",
       "281943                 1                1                          2   \n",
       "370194                 0                0                          0   \n",
       "370367                 1                1                          0   \n",
       "243389                 1                0                          3   \n",
       "77180                  0                1                          3   \n",
       "\n",
       "        avg length of words  token_set_ratio  token_sort_ratio  fuzz_WRatio  \\\n",
       "95873                  14.0               66                63           70   \n",
       "262013                  4.0               52                52           56   \n",
       "85654                   7.5               76                72           80   \n",
       "19837                   8.5              100                94           95   \n",
       "105551                  7.5               54                51           56   \n",
       "...                     ...              ...               ...          ...   \n",
       "281943                  9.0               84                84           84   \n",
       "370194                 14.0               58                55           58   \n",
       "370367                  9.0               75                65           74   \n",
       "243389                  6.5              100                68           90   \n",
       "77180                   6.5              100                71           86   \n",
       "\n",
       "        fuzz_partial_ratio  longest_substr_ratio  \n",
       "95873                   70              0.184211  \n",
       "262013                  61              0.476190  \n",
       "85654                   78              0.266667  \n",
       "19837                   86              0.584906  \n",
       "105551                  61              0.250000  \n",
       "...                    ...                   ...  \n",
       "281943                  76              0.272727  \n",
       "370194                  60              0.071429  \n",
       "370367                  71              0.395349  \n",
       "243389                  97              0.944444  \n",
       "77180                   74              0.625000  \n",
       "\n",
       "[323429 rows x 21 columns]"
      ]
     },
     "execution_count": 58,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "xy_train"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 59,
   "id": "bd58c9b7",
   "metadata": {
    "id": "bd58c9b7"
   },
   "outputs": [],
   "source": [
    "q1_train_clean1 = [] # this contains question1 without any stopwords.\n",
    "for z in range(len(xy_train[\"question1\"].values)):\n",
    "    res = remove_stopwords(str(xy_train[\"question1\"].values[z]))\n",
    "    q1_train_clean1.append(res)\n",
    "\n",
    "q2_train_clean2 = [] # this contains question2 without any stopwords.\n",
    "for z in range(len(xy_train[\"question2\"].values)):\n",
    "    res2 = remove_stopwords(str(xy_train[\"question2\"].values[z]))\n",
    "    q2_train_clean2.append(res2)  "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 60,
   "id": "9d6fe149",
   "metadata": {
    "id": "9d6fe149"
   },
   "outputs": [],
   "source": [
    "q1_test_clean1 = [] # this contains question1 without any stopwords.\n",
    "for z in range(len(xy_test[\"question1\"].values)):\n",
    "    res = remove_stopwords(str(xy_test[\"question1\"].values[z]))\n",
    "    q1_test_clean1.append(res)\n",
    "\n",
    "q2_test_clean2 = [] # this contains question2 without any stopwords.\n",
    "for z in range(len(xy_test[\"question2\"].values)):\n",
    "    res2 = remove_stopwords(str(xy_test[\"question2\"].values[z]))\n",
    "    q2_test_clean2.append(res2)  "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 61,
   "id": "b5a8ef82",
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "b5a8ef82",
    "outputId": "9ae2f284-304b-46d4-b7c9-585b9a0c8659"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Shape of tf-idf vector is:  (323429, 1000)\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "C:\\Users\\PANCHAYAT  SAMITI 01\\AppData\\Roaming\\Python\\Python39\\site-packages\\sklearn\\utils\\deprecation.py:87: FutureWarning: Function get_feature_names is deprecated; get_feature_names is deprecated in 1.0 and will be removed in 1.2. Please use get_feature_names_out instead.\n",
      "  warnings.warn(msg, category=FutureWarning)\n"
     ]
    }
   ],
   "source": [
    "# TF-IDF Vector: Train vector\n",
    "\n",
    "total_questions = (q1_train_clean1) + (q2_train_clean2)\n",
    "tfidf = TfidfVectorizer(max_features = 1000, min_df=50)\n",
    "tfidf.fit(total_questions)\n",
    "q1_train_vector = tfidf.transform(q1_train_clean1).toarray()\n",
    "q2_train_vector = tfidf.transform(q2_train_clean2).toarray()\n",
    "\n",
    "# Creating a dictionary with word as a key, and the idf as a value\n",
    "# This is done so that we can get TF-IDF values.\n",
    "\n",
    "tfidf_values = dict(zip(tfidf.get_feature_names(), list(tfidf.idf_)))\n",
    "print('Shape of tf-idf vector is: ',q1_train_vector.shape)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 62,
   "id": "1w1KAsQAvNyd",
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "1w1KAsQAvNyd",
    "outputId": "7e862cc9-20ec-4bf0-bb75-c8dc7ec8624f"
   },
   "outputs": [],
   "source": [
    "features = tfidf.get_feature_names()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "02ab4bff",
   "metadata": {
    "id": "02ab4bff"
   },
   "source": [
    "# Creating my own W2V model on the Quora question pair corpus:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 63,
   "id": "b7ecd7d5",
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "b7ecd7d5",
    "outputId": "c5ea53fe-80ad-43a0-db50-408c6f6664b1"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "sample words  ['in', 'best', 'some', 'get', 'good', 'people', 'would', 'like', 'between', 'most', 'one', 'make', 'not', 'way', 'india?', 'after', 'difference', \"what's\", 'know', 'learn', 'new', 'much', 'many', 'think', 'find', 'money', 'india', 'use', 'indian', 'without', 'become', 'more', 'someone', 'ever', 'but', 'quora', 'time', 'start', 'quora?', 'first', 'want', 'out', 'trump', 'life?', 'why?', 'take', 'could', 'better', 'feel', 'it?']\n"
     ]
    }
   ],
   "source": [
    "# Creating w2v model on question 1:\n",
    "q1_train_list = []  \n",
    "for new in (q1_train_clean1):\n",
    "    q1_train_list.append(new.split())\n",
    "    \n",
    "w2v_q1=Word2Vec(q1_train_list,min_count=5,vector_size=100, workers=2) \n",
    "w2v_words_q1 = list(w2v_q1.wv.index_to_key) # to get the keys which are words\n",
    "print(\"sample words \", w2v_words_q1[0:50])\n",
    "\n",
    "# vector_size = 100---> each word will be represented by a 200 dimensional vector.\n",
    "# min_count = 5---> only those words are considered which atleast ocuur (5 times) in the corpus."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 64,
   "id": "c65bbb36",
   "metadata": {
    "id": "c65bbb36"
   },
   "outputs": [],
   "source": [
    "# Creating w2v model on question 2:\n",
    "q2_train_list = []\n",
    "for new2 in (q2_train_clean2):\n",
    "    q2_train_list.append(new2.split())\n",
    "    \n",
    "w2v_q2 = Word2Vec(q2_train_list,min_count=5,vector_size=100, workers=2) \n",
    "w2v_words_q2 = list(w2v_q2.wv.index_to_key)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 65,
   "id": "a2a501ae",
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "a2a501ae",
    "outputId": "1c4631e5-2cd5-4e01-ba29-ba9924a59b69"
   },
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|█████████████████████████████████████████████████████████████████████████| 323429/323429 [10:38<00:00, 506.88it/s]\n"
     ]
    }
   ],
   "source": [
    "# Getting TF-IDF*W2V values for question 1 :\n",
    "\n",
    "tfidf_w2v_q1 = [] # the tfidf-w2v for each question1 is stored in this list\n",
    "\n",
    "for sent1 in tqdm(q1_train_list): # for each question1\n",
    "    sent_vec1 = np.zeros(100) \n",
    "    tfidf_sum1 =0\n",
    "    for word1 in sent1: # for each word in question1\n",
    "        if word1 in w2v_words_q1 and word1 in features:\n",
    "            vec1 = w2v_q1.wv[word1] # w2v vector for the word\n",
    "            tf_idf_q1 = tfidf_values[word1]*(sent1.count(word1)/len(sent1)) # idf * tf = tfidf\n",
    "            sent_vec1 += (vec1 * tf_idf_q1) # w2v * tfidf\n",
    "            tfidf_sum1 += tf_idf_q1 # summation of tfidf\n",
    "    if tfidf_sum1 != 0: # handling boundary condition\n",
    "        sent_vec1 = sent_vec1/tfidf_sum1\n",
    "    tfidf_w2v_q1.append(sent_vec1)\n",
    "\n",
    "tfidf_w2v_q1_list = list(tfidf_w2v_q1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 66,
   "id": "4fe5d12b",
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "4fe5d12b",
    "outputId": "3e697552-5be7-4b63-e6b0-1ad59495f6fb"
   },
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|█████████████████████████████████████████████████████████████████████████| 323429/323429 [16:06<00:00, 334.53it/s]\n"
     ]
    }
   ],
   "source": [
    "# Getting TF-IDF*W2V values for question 2 :\n",
    "\n",
    "tfidf_w2v_q2 = [] # the tfidf-w2v for each question 2 is stored in this list\n",
    "\n",
    "for sent2 in tqdm(q2_train_list): # for each question2\n",
    "    sent_vec2 = np.zeros(100) \n",
    "    tfidf_sum2 =0\n",
    "    for word2 in sent2: # for each word in question2\n",
    "        if word2 in w2v_words_q2 and word2 in features:\n",
    "            vec2 = w2v_q2.wv[word2] \n",
    "            tf_idf_q2 = tfidf_values[word2]*(sent2.count(word2)/len(sent2)) \n",
    "            sent_vec2 += (vec2 * tf_idf_q2) \n",
    "            tfidf_sum2 += tf_idf_q2 \n",
    "    if tfidf_sum2 != 0:\n",
    "        sent_vec2 = sent_vec2/tfidf_sum2\n",
    "    tfidf_w2v_q2.append(sent_vec2)\n",
    "\n",
    "tfidf_w2v_q2_list = list(tfidf_w2v_q2)\n",
    "   "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 67,
   "id": "21edf60e",
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "21edf60e",
    "outputId": "d1e9dca9-3288-4c68-f66c-047fc4fec6a3"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "The dimensions of TF_IDF_W2V is: 100\n"
     ]
    }
   ],
   "source": [
    "print('The dimensions of TF_IDF_W2V is:',len(tfidf_w2v_q2[0]))"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "a7e01a03",
   "metadata": {
    "id": "a7e01a03"
   },
   "source": [
    "# Data Preprocessing (Test set):"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 68,
   "id": "656a005b",
   "metadata": {
    "id": "656a005b"
   },
   "outputs": [],
   "source": [
    "q1_test_list = []  \n",
    "for new in (q1_test_clean1):\n",
    "    q1_test_list.append(new.split())\n",
    "\n",
    "q2_test_list = []\n",
    "for new2 in (q2_test_clean2):\n",
    "    q2_test_list.append(new2.split())"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 69,
   "id": "6648ba68",
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "6648ba68",
    "outputId": "412aacd4-c895-4292-c957-7e56035575b4"
   },
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|███████████████████████████████████████████████████████████████████████████| 80858/80858 [04:43<00:00, 284.74it/s]\n"
     ]
    }
   ],
   "source": [
    "# Getting TF-IDF*W2V values for question 1 :\n",
    "\n",
    "tfidf_w2v_q1_test = [] # the tfidf-w2v for each question1 is stored in this list\n",
    "\n",
    "for sent1 in tqdm(q1_test_list): # for each question1\n",
    "    sent_vec1 = np.zeros(100) \n",
    "    tfidf_sum1 =0\n",
    "    for word1 in sent1: # for each word in question1\n",
    "        if word1 in w2v_words_q1 and word1 in features:\n",
    "            vec1 = w2v_q1.wv[word1] # w2v vector for the word\n",
    "            tf_idf_q1 = tfidf_values[word1]*(sent1.count(word1)/len(sent1)) # idf * tf = tfidf\n",
    "            sent_vec1 += (vec1 * tf_idf_q1) # w2v * tfidf\n",
    "            tfidf_sum1 += tf_idf_q1 # summation of tfidf\n",
    "    if tfidf_sum1 != 0: # handling boundary condition\n",
    "        sent_vec1 = sent_vec1/tfidf_sum1\n",
    "    tfidf_w2v_q1_test.append(sent_vec1)\n",
    "\n",
    "tfidf_w2v_q1_list_test = list(tfidf_w2v_q1_test)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 70,
   "id": "8513245d",
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "8513245d",
    "outputId": "62f7b6f9-f869-4fce-89da-5ab210299cb9"
   },
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|███████████████████████████████████████████████████████████████████████████| 80858/80858 [05:17<00:00, 254.84it/s]\n"
     ]
    }
   ],
   "source": [
    "# Getting TF-IDF*W2V values for question 2 :\n",
    "\n",
    "tfidf_w2v_q2_test = [] # the tfidf-w2v for each question 2 is stored in this list\n",
    "\n",
    "for sent2 in tqdm(q2_test_list): # for each question2\n",
    "    sent_vec2 = np.zeros(100) \n",
    "    tfidf_sum2 =0\n",
    "    for word2 in sent2: # for each word in question2\n",
    "        if word2 in w2v_words_q2 and word2 in features:\n",
    "            vec2 = w2v_q2.wv[word2] \n",
    "            tf_idf_q2 = tfidf_values[word2]*(sent2.count(word2)/len(sent2)) \n",
    "            sent_vec2 += (vec2 * tf_idf_q2) \n",
    "            tfidf_sum2 += tf_idf_q2 \n",
    "    if tfidf_sum2 != 0:\n",
    "        sent_vec2 = sent_vec2/tfidf_sum2\n",
    "    tfidf_w2v_q2_test.append(sent_vec2)\n",
    "\n",
    "tfidf_w2v_q2_list_test = list(tfidf_w2v_q2_test)\n",
    "   "
   ]
  },
  {
   "cell_type": "markdown",
   "id": "a2f2009c",
   "metadata": {
    "id": "a2f2009c"
   },
   "source": [
    "# Creating final dataframe: (Train set)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 71,
   "id": "4f1d6bd2",
   "metadata": {
    "id": "4f1d6bd2"
   },
   "outputs": [],
   "source": [
    "ques1_train = pd.DataFrame(tfidf_w2v_q1_list,index = xy_train.index) # Storing the q1 vector here\n",
    "ques2_train = pd.DataFrame(tfidf_w2v_q2_list,index = xy_train.index) # Storing the q2 vector here\n",
    "ques1_train['id']=xy_train['id']\n",
    "ques2_train['id']=xy_train['id']"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 72,
   "id": "92541f67",
   "metadata": {
    "id": "92541f67"
   },
   "outputs": [],
   "source": [
    "ques1_test = pd.DataFrame(tfidf_w2v_q1_list_test,index = xy_test.index) # Storing the q1 vector here\n",
    "ques2_test = pd.DataFrame(tfidf_w2v_q2_list_test,index = xy_test.index) # Storing the q2 vector here\n",
    "ques1_test['id']=xy_test['id']\n",
    "ques2_test['id']=xy_test['id']\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 73,
   "id": "0fda8cc4",
   "metadata": {
    "id": "0fda8cc4"
   },
   "outputs": [],
   "source": [
    "# Dropping the features which won't be required in Modeling\n",
    "\n",
    "X_train_ = xy_train.drop(['qid1', 'qid2', 'question1', 'question2', 'is_duplicate'], axis = 1)\n",
    "X_test_ = xy_test.drop(['qid1', 'qid2', 'question1', 'question2', 'is_duplicate'], axis = 1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 74,
   "id": "e07cccf4",
   "metadata": {
    "id": "e07cccf4"
   },
   "outputs": [],
   "source": [
    "# Merging all the dataframes:\n",
    "\n",
    "train_vector = ques1_train.merge(ques2_train,on = 'id',how ='left')\n",
    "X_train = X_train_.merge(train_vector,on = 'id',how ='left')\n",
    "\n",
    "test_vector = ques1_test.merge(ques2_test,on = 'id',how ='left')\n",
    "X_test = X_test_.merge(test_vector,on = 'id',how ='left')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 75,
   "id": "b10a725b",
   "metadata": {
    "id": "b10a725b"
   },
   "outputs": [],
   "source": [
    "X_train.drop([\"id\"],axis = 1, inplace = True)\n",
    "X_test.drop([\"id\"],axis = 1, inplace = True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 76,
   "id": "4c84ab3d",
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "4c84ab3d",
    "outputId": "f54462e8-12a1-4a03-dad0-65d78481868a"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Train data dimensions: (323429, 215)\n",
      "Test data dimensions: (80858, 215)\n"
     ]
    }
   ],
   "source": [
    "print('Train data dimensions:',X_train.shape)\n",
    "print('Test data dimensions:',X_test.shape)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 77,
   "id": "b566280f",
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "b566280f",
    "outputId": "fe72c5fc-f7f8-4b84-ad26-0357efbb70bd"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Train data dimensions: (323429,)\n",
      "Test data dimensions: (80858,)\n"
     ]
    }
   ],
   "source": [
    "print('Train data dimensions:',y_train.shape)\n",
    "print('Test data dimensions:',y_test.shape)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 78,
   "id": "T0MPdj8BcMu5",
   "metadata": {
    "id": "T0MPdj8BcMu5"
   },
   "outputs": [],
   "source": [
    "X_train.to_csv('X_train.csv')\n",
    "X_test.to_csv('X_test.csv')\n",
    "\n",
    "y_train.to_csv('y_train.csv')\n",
    "y_test.to_csv('y_test.csv')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "8934718e",
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": 79,
   "id": "2OhINQqV57W9",
   "metadata": {
    "id": "2OhINQqV57W9"
   },
   "outputs": [],
   "source": [
    "#X_train = pd.read_csv('X_train.csv')\n",
    "#X_test = pd.read_csv('X_test.csv')\n",
    "\n",
    "#y_train = pd.read_csv('y_train.csv')\n",
    "#y_test = pd.read_csv('y_test.csv')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 80,
   "id": "F0z2XE466VTh",
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 240
    },
    "id": "F0z2XE466VTh",
    "outputId": "3c01ce85-0333-4c34-aa85-67e71a7148ca"
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>q1_length</th>\n",
       "      <th>q2_length</th>\n",
       "      <th>common words</th>\n",
       "      <th>word share</th>\n",
       "      <th>cwc_min</th>\n",
       "      <th>cwc_max</th>\n",
       "      <th>first word equal</th>\n",
       "      <th>last word equal</th>\n",
       "      <th>difference in no.of words</th>\n",
       "      <th>avg length of words</th>\n",
       "      <th>...</th>\n",
       "      <th>90_y</th>\n",
       "      <th>91_y</th>\n",
       "      <th>92_y</th>\n",
       "      <th>93_y</th>\n",
       "      <th>94_y</th>\n",
       "      <th>95_y</th>\n",
       "      <th>96_y</th>\n",
       "      <th>97_y</th>\n",
       "      <th>98_y</th>\n",
       "      <th>99_y</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>14</td>\n",
       "      <td>14</td>\n",
       "      <td>6</td>\n",
       "      <td>0.214</td>\n",
       "      <td>0.429</td>\n",
       "      <td>0.429</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>14.0</td>\n",
       "      <td>...</td>\n",
       "      <td>0.999294</td>\n",
       "      <td>-0.460811</td>\n",
       "      <td>0.367546</td>\n",
       "      <td>-0.976681</td>\n",
       "      <td>2.58468</td>\n",
       "      <td>1.068159</td>\n",
       "      <td>0.369618</td>\n",
       "      <td>-0.909836</td>\n",
       "      <td>-0.65708</td>\n",
       "      <td>-1.247355</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>4</td>\n",
       "      <td>4</td>\n",
       "      <td>1</td>\n",
       "      <td>0.125</td>\n",
       "      <td>0.250</td>\n",
       "      <td>0.250</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>4.0</td>\n",
       "      <td>...</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.00000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.00000</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>2 rows × 215 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "   q1_length  q2_length  common words  word share  cwc_min  cwc_max  \\\n",
       "0         14         14             6       0.214    0.429    0.429   \n",
       "1          4          4             1       0.125    0.250    0.250   \n",
       "\n",
       "   first word equal  last word equal  difference in no.of words  \\\n",
       "0                 1                0                          0   \n",
       "1                 0                0                          0   \n",
       "\n",
       "   avg length of words  ...      90_y      91_y      92_y      93_y     94_y  \\\n",
       "0                 14.0  ...  0.999294 -0.460811  0.367546 -0.976681  2.58468   \n",
       "1                  4.0  ...  0.000000  0.000000  0.000000  0.000000  0.00000   \n",
       "\n",
       "       95_y      96_y      97_y     98_y      99_y  \n",
       "0  1.068159  0.369618 -0.909836 -0.65708 -1.247355  \n",
       "1  0.000000  0.000000  0.000000  0.00000  0.000000  \n",
       "\n",
       "[2 rows x 215 columns]"
      ]
     },
     "execution_count": 80,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "X_train.head(2)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 81,
   "id": "VqLrvXBn6rMQ",
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 240
    },
    "id": "VqLrvXBn6rMQ",
    "outputId": "3f7730f3-3445-48e3-b6b4-20b5cf40fd62"
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>q1_length</th>\n",
       "      <th>q2_length</th>\n",
       "      <th>common words</th>\n",
       "      <th>word share</th>\n",
       "      <th>cwc_min</th>\n",
       "      <th>cwc_max</th>\n",
       "      <th>first word equal</th>\n",
       "      <th>last word equal</th>\n",
       "      <th>difference in no.of words</th>\n",
       "      <th>avg length of words</th>\n",
       "      <th>...</th>\n",
       "      <th>90_y</th>\n",
       "      <th>91_y</th>\n",
       "      <th>92_y</th>\n",
       "      <th>93_y</th>\n",
       "      <th>94_y</th>\n",
       "      <th>95_y</th>\n",
       "      <th>96_y</th>\n",
       "      <th>97_y</th>\n",
       "      <th>98_y</th>\n",
       "      <th>99_y</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>8</td>\n",
       "      <td>13</td>\n",
       "      <td>5</td>\n",
       "      <td>0.238</td>\n",
       "      <td>0.625</td>\n",
       "      <td>0.385</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>5</td>\n",
       "      <td>10.5</td>\n",
       "      <td>...</td>\n",
       "      <td>-0.385447</td>\n",
       "      <td>2.043453</td>\n",
       "      <td>-0.806982</td>\n",
       "      <td>-0.715055</td>\n",
       "      <td>1.562307</td>\n",
       "      <td>0.437262</td>\n",
       "      <td>-0.104445</td>\n",
       "      <td>-0.536293</td>\n",
       "      <td>0.715435</td>\n",
       "      <td>-0.499209</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>25</td>\n",
       "      <td>9</td>\n",
       "      <td>8</td>\n",
       "      <td>0.235</td>\n",
       "      <td>0.889</td>\n",
       "      <td>0.320</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>16</td>\n",
       "      <td>17.0</td>\n",
       "      <td>...</td>\n",
       "      <td>0.677841</td>\n",
       "      <td>0.571451</td>\n",
       "      <td>0.030739</td>\n",
       "      <td>0.753049</td>\n",
       "      <td>0.504953</td>\n",
       "      <td>-0.018917</td>\n",
       "      <td>-0.302363</td>\n",
       "      <td>-0.556129</td>\n",
       "      <td>-0.893268</td>\n",
       "      <td>0.793344</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>2 rows × 215 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "   q1_length  q2_length  common words  word share  cwc_min  cwc_max  \\\n",
       "0          8         13             5       0.238    0.625    0.385   \n",
       "1         25          9             8       0.235    0.889    0.320   \n",
       "\n",
       "   first word equal  last word equal  difference in no.of words  \\\n",
       "0                 0                0                          5   \n",
       "1                 1                0                         16   \n",
       "\n",
       "   avg length of words  ...      90_y      91_y      92_y      93_y      94_y  \\\n",
       "0                 10.5  ... -0.385447  2.043453 -0.806982 -0.715055  1.562307   \n",
       "1                 17.0  ...  0.677841  0.571451  0.030739  0.753049  0.504953   \n",
       "\n",
       "       95_y      96_y      97_y      98_y      99_y  \n",
       "0  0.437262 -0.104445 -0.536293  0.715435 -0.499209  \n",
       "1 -0.018917 -0.302363 -0.556129 -0.893268  0.793344  \n",
       "\n",
       "[2 rows x 215 columns]"
      ]
     },
     "execution_count": 81,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "X_test.head(2)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 82,
   "id": "or9v2fCW6rWZ",
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 112
    },
    "id": "or9v2fCW6rWZ",
    "outputId": "eb255a07-dea1-430f-af27-a09afc80ec35"
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "95873     1\n",
       "262013    0\n",
       "Name: is_duplicate, dtype: int64"
      ]
     },
     "execution_count": 82,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "y_train.head(2)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 83,
   "id": "fC9_02oT6raV",
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 112
    },
    "id": "fC9_02oT6raV",
    "outputId": "8c3b5039-6088-4b5a-b655-70ff06e1203c"
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "35865    1\n",
       "83858    1\n",
       "Name: is_duplicate, dtype: int64"
      ]
     },
     "execution_count": 83,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "y_test.head(2)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "id": "UQK6_dKL6reS",
   "metadata": {
    "id": "UQK6_dKL6reS"
   },
   "outputs": [],
   "source": [
    "#X_train.drop([\"Unnamed: 0\"\t, \"Unnamed: 0.1\"], axis = 1, inplace = True)\n",
    "#X_test.drop([\"Unnamed: 0\"\t, \"Unnamed: 0.1\"], axis = 1, inplace = True)\n",
    "\n",
    "#y_train.drop([\"Unnamed: 0\"], axis = 1, inplace = True)\n",
    "#y_test.drop([\"Unnamed: 0\"], axis = 1, inplace = True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 108,
   "id": "f050ddb5",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<table>\n",
       "    <thead>\n",
       "        <tr>\n",
       "            <th>Model</th>\n",
       "            <th>Train-AUC</th>\n",
       "            <th>Test-AUC</th>\n",
       "        </tr>\n",
       "    </thead>\n",
       "    <tbody>\n",
       "        <tr>\n",
       "            <td>Logistic Regression</td>\n",
       "            <td>0.79</td>\n",
       "            <td>0.79</td>\n",
       "        </tr>\n",
       "        <tr>\n",
       "            <td>XGBoost</td>\n",
       "            <td>0.8418</td>\n",
       "            <td>0.8418</td>\n",
       "        </tr>\n",
       "        <tr>\n",
       "            <td>LGBM</td>\n",
       "            <td>0.875</td>\n",
       "            <td>0.869</td>\n",
       "        </tr>\n",
       "    </tbody>\n",
       "</table>"
      ],
      "text/plain": [
       "+---------------------+-----------+----------+\n",
       "|        Model        | Train-AUC | Test-AUC |\n",
       "+---------------------+-----------+----------+\n",
       "| Logistic Regression |    0.79   |   0.79   |\n",
       "|       XGBoost       |   0.8418  |  0.8418  |\n",
       "|         LGBM        |   0.875   |  0.869   |\n",
       "+---------------------+-----------+----------+"
      ]
     },
     "execution_count": 108,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "from prettytable import PrettyTable\n",
    "x = PrettyTable()\n",
    "\n",
    "x.field_names = [\"Model\",\"Train-AUC\", \"Test-AUC\"]\n",
    "x.add_row([\"Logistic Regression\",0.79 ,0.79])\n",
    "x.add_row([\"XGBoost\",0.8418,0.8418])\n",
    "x.add_row([\"LGBM\",0.875,0.869])\n",
    "(x)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "b9c64a5f",
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "accelerator": "GPU",
  "colab": {
   "collapsed_sections": [],
   "name": "Quora_question_pairs (1).ipynb",
   "provenance": []
  },
  "gpuClass": "standard",
  "kernelspec": {
   "display_name": "Python 3 (ipykernel)",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.9.12"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}