Simple Exploratory Data Analysis.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [],
   "source": [
    "import numpy as np\n",
    "import pandas as pd\n",
    "import matplotlib.pyplot as plt\n",
    "import seaborn as sns\n",
    "%matplotlib inline\n",
    "pal = sns.color_palette()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Training set"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "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",
       "    </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",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>4</td>\n",
       "      <td>9</td>\n",
       "      <td>10</td>\n",
       "      <td>Which one dissolve in water quikly sugar, salt...</td>\n",
       "      <td>Which fish would survive in salt water?</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",
       "4   4     9    10  Which one dissolve in water quikly sugar, salt...   \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  \n",
       "4            Which fish would survive in salt water?             0  "
      ]
     },
     "execution_count": 5,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_train = pd.read_csv('./input/train.csv')\n",
    "df_train.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Total number of question pairs for training: 404290\n",
      "Duplicate pairs: 36.92%\n",
      "Total number of questions in the training data: 537933\n",
      "Number of questions that appear multiple times: 111780\n",
      "\n"
     ]
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAtcAAAFNCAYAAADLm0PlAAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADl0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uIDIuMS4xLCBodHRwOi8vbWF0cGxvdGxpYi5vcmcvAOZPmwAAIABJREFUeJzt3XmcZHV57/HPl2FRQYc1LiwOOrjghtiCehNF4sI2YDBeQUIiEkZNULxXE3GJGo0R477gMhcQt0AQFRlBISKLUSMMiMgiijDIEHZwwA1EnvvHOY1F00t1T1VXFXzer9d5TZ2lznnOr05PP/2r5/xOqgpJkiRJa26tQQcgSZIk3VeYXEuSJEk9YnItSZIk9YjJtSRJktQjJteSJElSj5hcS5IkST1ici2p55L8KsmjBh3HICV5dZLr2rbYZNDxTCbJfklOHXQcknRfYnIt3UckWZnkefNwnEqyeMKydyT5wvh8VW1QVZfPsJ+dkqzqV5yDlGQd4IPAC9q2uGkIYlrUfnZrjy+rqi9W1QsGGZcGI8kZSf520HFI90Um15Luk5IsGODhHwo8ALhogDFoBgO+RiTdR5lcS/cDSQ5KclmSm5OcmOQRHetekOTSJKuTfCLJmWvao9XZu51ktyQXJ7ktydVJ3pBkfeAbwCPasolfJXlEkvWSfDjJ/7TTh5Os17Hff0xyTbvubycc5+gkn0xycpJfA89NsnuSHya5NclVSd7Rsa/xntwD2nW3JHlVkqcnuSDJL5N8fJpznDTWJI8BLm03+2WSb0/x/v2TXJnkpiRv6fzmoT2Xf+nY9h69/G1bfTnJDUmuSPLajnU7JFnRnvN1ST7YrjqrI6ZfJXlmkpcn+a+O9z4ryTnttXBOkmd1rDsjybuSfLf9LE9NsukU57ZRkq+38d3Svt5iwr7ek+TsNs6vJdl4wueytG3Xa5K8oeO9ayU5NMnP27Y7bvy97fovJbm2PYezkjyhY91cr5G/SfKLJDcmeUvH+gVJ3tzGcluSc5Ns2a57XJL/TPMzd2mS/z1ZW7XbbpzkM+353pLkhI51k/7sZpJvItLRGz3+2SZ5f7vPK5Ls2q57N/BnwMfba+HjaXwoyfVtW/w4yROnilnSNKrKycnpPjABK4HnTbJ8Z+BGYHtgPeBjwFntuk2BW4G9gbWBQ4DfA387zXEKWDxh2TuAL0y2DXAN8Gft642A7dvXOwGrJuznncB/A38CbAZ8D3hXu24X4FrgCcCDgC9MOM7RwGrgf9F0HDygPcaT2vknA9cBL2q3X9S+/1Ptti8Afgec0B5/c+B64DlTtMN0sY7ve+0p3rst8Cvg2e1n8kHgzvHPrz2Xf+nY/u62as/lXOBtwLrAo4DLgRe2678P7N++3gB4xlQxAS8H/qt9vTFwC7B/ey3s285v0q4/A/g58Bjgge38YVOc3ybAi9vP6cHAl4ATOtafAVwNPBFYH/gy7fXTEecx7bonATd0tM0hbbtv0bbdp4FjOvb9ivaY6wEfBs7vWHc0c7tG/l97zk8Bbgce367/B+DHwGOBtOs3aeO+Cjigbcun0vwMbjtFe50E/AfNz8c6tNcc0//sTvZ5nkH7s9t+tr8HDgIWAK8G/gfIxG3b+RfSXFcbtufyeODhg/5/zclpFKeBB+Dk5NSbiamT6yOBf+uY36D9pbsI+Gvg+x3r0iYFMyXXtwK/7Jh+x9TJ9S+AVwIPmbCfnbh3cv1zYLeO+RcCK9vXRwHv6Vi3mHsn15+boY0+DHyofT2enGzesf4m4KUd818GXjfFvqaL9V6Jz4T3vg04tmN+feAOukuudwR+MWF/bwI+074+C/hnYNMJ29wrJu6ZXO8PnD3hPd8HXt6+PgN4a8e6vwO+2eW1uR1wS8f8GXQk5jR/bNxBkwSOx/m4jvX/BhzZvr4E+POOdQ+nuZ7v1dY0iWIBC9fwGtmiY/3ZwD7t60uBvSbZx0uB70xY9mng7ZNs+3DgLmCjWf7sTvZ5nsE9k+vLOtY9qN3+YRO3bed3Bn4KPANYq5vP1cnJafLJshDpvu8RwJXjM1X1K5okcvN23VUd6wroLD+4KH8s2/izjn1uX1Ubjk/AYdMc/8XAbsCVaUpOntltrO3rR3Ssu6pjXefrSZcl2THJ6W15wmrgVTS99Z2u63j920nmN5hDrDOZ2O6/pvlMuvFImnKaX45PwJtp6rwBDqTpXf5JW9qxxyxiunLCsitprpNx13a8/g1TtE2SByX5dJqyl1tpEv4Nc88a587P6kqaHttNp1k/3raPBL7ace6XAH8AHtqWaRzWlmncSvMHJ9Pst9trZKrz3pLmj6yJHgnsOOEz2g942CTbbgncXFW3TLJuup/dbtwdd1X9pn056WdWVd8GPg4cDlyfZFmSh3R5HEkdTK6l+77/ofllD0CaeudNaL6Wv4bm6/Xxdemcr6onVDPaxQZV9Z25HLyqzqmqvWjKJ04AjhtfNVOswFbtMibGSpOU3OtwE+b/HTgR2LKqFtKUgGRWJzC16WKdyTV0xJ/kQTSfybhf0/Q0jutMyq4Cruj846aqHlxVuwFU1c+qal+a9n4vcHz7mU/W3tOdz/g5Xd3lOXV6PU2pxI5V9RCa8he4Z9t3fn5b0fTI3jjN+vG2vQrYdcL5P6CqrgZeBuwFPA9YSNO7O/G4vbxGrgIePcXyMyfEuEFVvXqKbTdOsuEk66b72f11u3iq62Qm97oequqjVfU0mm8SHkNT9iJplkyupfuWdZI8oGNam6Z29YAk26W5OfBfgR9U1UqaWs8nJXlRu+3fM7tf0NNKsm6asZQXVtXvacpJ7mpXXwdskmRhx1uOAd6aZLM0N8u9jaa2Gpqk/IAkj2+T0X/qIoQH0/QK/i7JDjTJV69MF+tMjgf2SPKnSdalqd/u/P/4fGC39ka3hwGv61h3NnBbkjcmeWDbW/vEJE8HSPJXSTarqrtoSnagafMb2n+nGn/8ZOAxSV6WZO0kL6VJsr7e5Tl1ejBNr/8v09xs+PZJtvmrJNu2n+U7geOr6g8d6/+p7QF/Ak3t8n+0yz8FvDvJI9vz3SzJXh3HvZ2md/dBNNd6N7HO9Ro5AnhXkm3aGwKfnGZM86/TtOX+SdZpp6cnefzEHVTVNTQ3934izY2g6yQZ/2Nkyp/dqrqBJsn+q/YaeAWTJ/pTuY6Oa6GNb8c0w0j+mqbU666p3ixpaibX0n3LyTRJzfj0jqr6Fk0i+mWaHtNHA/sAVNWNwEtoalpvokmmVtAkKL2yP7Cy/Zr+VTRfj1NVP6FJHi5vvzp/BPAv7fEvoLlR7Lx2GVX1DeCjwOnAZTQ3tTFDrH8HvDPJbTTJ73HTbDtbU8Y6k6q6iOYPmX+n+UxuoaMcB/g88COasoZT+WNiSZuA7kFTx3wFTW/vETQ9tdDc+HlRkl8BH6GpD/5tWxbwbuC7bXs/Y0JMN7X7fT3NtfCPwB7tNTJbH6a5AfBGms/pm5Ns83maGuhraW4sfO2E9WfSfM6nAe+vqvGH3XyEpqf51PZz/W+aOnSAz9GUUVwNXMwfr5HprMk18sF2+1Np/nA8EnhgVd1Gc4PsPjS9z9fSfIuw3hT72Z+m5/4nNDfRvg5gup/d1kE0vcs30dzo+71ZxP4R4C/TjCTyUeAhNDdu3kLThjcB75vF/iS1xu8aliSSrEWT5O1XVacPOp7ptL2AFwLrVdWdg45nTSVZSXOD2bcGHUu/JTmD5gbYIyZZt4jmj4Z17gufq6T7H3uupfu5JC9MsmH7tfObaepNu+nxm3dJ/iLNWNIb0fQELjcBkyQNk6FJrtM8GODdST6W5G8GHY90P/JMmhEPbgSW0Izx+9vBhjSlV9J8bf5zmhEiJrtBTJKkgelrWUiSo2hq+K6vqid2LN+Fpt5rAXBEVR2W5C+AF9HUeZ1UVaf1LTBJkiSpD/rdc300zc01d2vHOT0c2JXm5ql9k2xLM2zT96rq/2JvlCRJkkZQX5PrqjoLuHnC4h1onhp1eVXdARxLMy7pKpq7lKH5uleSJEkaKWsP4Jibc88nZK2iGUbpI8DH0jwF7qyp3pxkKbAUYP3113/a4x73uD6GKkmSJMG55557Y1VtNtN2g0iuJ9WOwXpgF9stA5YBjI2N1YoVK/odmiRJku7nklzZzXaDGC3kau75WNstmOXjdZMsSbJs9erVPQ1MkiRJWhODSK7PAbZJsnX72N99aJ621bWqWl5VSxcuXDjzxpIkSdI86WtyneQY4PvAY5OsSnJg+8CHg4FTgEuA49pHAc9mv/ZcS5IkaeiM9OPPrbmWJEnSfEhyblWNzbTd0DyhUZIkSRp1I5lcWxYiSZKkYTSSybU3NEqSJGkYjWRyLUmSJA2jkUyuLQuRJEnSMBrJ5NqyEEmSJA2joXn8+ShZdOhJs37PysN270MkkiRJGiYj2XNtWYgkSZKG0Ugm15aFSJIkaRiNZHItSZIkDSOTa0mSJKlHRjK5tuZakiRJw2gkk2trriVJkjSMRjK5liRJkoaRybUkSZLUIybXkiRJUo+YXEuSJEk9MpLJtaOFSJIkaRiNZHLtaCGSJEkaRiOZXEuSJEnDyORakiRJ6hGTa0mSJKlHTK4lSZKkHjG5liRJknpkJJNrh+KTJEnSMBrJ5Nqh+CRJkjSMRjK5liRJkoaRybUkSZLUIybXkiRJUo+YXEuSJEk9YnItSZIk9YjJtSRJktQjJteSJElSjwxNcp1kpyTfSfKpJDsNOh5JkiRptvqaXCc5Ksn1SS6csHyXJJcmuSzJoe3iAn4FPABY1c+4JEmSpH7od8/10cAunQuSLAAOB3YFtgX2TbIt8J2q2hV4I/DPfY5LkiRJ6rm+JtdVdRZw84TFOwCXVdXlVXUHcCywV1Xd1a6/BVivn3FJkiRJ/bD2AI65OXBVx/wqYMckewMvBDYEPj7Vm5MsBZYCbLXVVn0MU5IkSZqdQSTXk6qqrwBf6WK7ZcAygLGxsep3XJIkSVK3BjFayNXAlh3zW7TLupZkSZJlq1ev7mlgkiRJ0poYRHJ9DrBNkq2TrAvsA5w4mx1U1fKqWrpw4cK+BChJkiTNRb+H4jsG+D7w2CSrkhxYVXcCBwOnAJcAx1XVRf2MQ5IkSZoPfa25rqp9p1h+MnDyXPebZAmwZPHixXPdhSRJktRzQ/OExtmwLESSJEnDaCSTa29olCRJ0jAayeTanmtJkiQNo5FMriVJkqRhNJLJtWUhkiRJGkYjmVxbFiJJkqRhNJLJtSRJkjSMRjK5tixEkiRJw2gkk2vLQiRJkjSMRjK5liRJkoaRybUkSZLUIybXkiRJUo+MZHLtDY2SJEkaRiOZXHtDoyRJkobRSCbXkiRJ0jAyuZYkSZJ6xORakiRJ6pGRTK69oVGSJEnDaCSTa29olCRJ0jAayeRakiRJGkYm15IkSVKPmFxLkiRJPWJyLUmSJPWIybUkSZLUIybXkiRJUo+MZHLtONeSJEkaRiOZXDvOtSRJkobRSCbXkiRJ0jAyuZYkSZJ6xORakiRJ6hGTa0mSJKlHTK4lSZKkHjG5liRJknrE5FqSJEnqkaFKrpOsn2RFkj0GHYskSZI0W31NrpMcleT6JBdOWL5LkkuTXJbk0I5VbwSO62dMkiRJUr/MmFwnOSTJQ9I4Msl5SV7Q5f6PBnaZsL8FwOHArsC2wL5Jtk3yfOBi4PpZnYEkSZI0JNbuYptXVNVHkrwQ2AjYH/g8cOpMb6yqs5IsmrB4B+CyqrocIMmxwF7ABsD6NAn3b5OcXFV3dXsikiRJ0qB1k1yn/Xc34PNVdVGSTPeGGWwOXNUxvwrYsaoOBkjycuDGqRLrJEuBpQBbbbXVGoQxvxYdetKstl952O59ikSSJEn90k3N9blJTqVJrk9J8mCgbz3KVXV0VX19mvXLqmqsqsY222yzfoUhSZIkzVo3PdcHAtsBl1fVb5JsAhywBse8GtiyY36LdlnXkiwBlixevHgNwpAkSZJ6a8ae67Y84zpg2yTPBp4AbLgGxzwH2CbJ1knWBfYBTpzNDqpqeVUtXbhw4RqEIUmSJPXWjD3XSd4LvJRmJI8/tIsLOKuL9x4D7ARsmmQV8PaqOjLJwcApwALgqKq6aDZB23MtSZKkYdRNWciLgMdW1e2z3XlV7TvF8pOBk2e7v473LweWj42NHTTXfUiSJEm91s0NjZcD6/Q7kNlIsiTJstWrVw86FEmSJOlu3fRc/wY4P8lpwN2911X12r5FNQN7riVJkjSMukmuT2SWNxxKkiRJ90czJtdV9dl2VI/HtIsurarf9zes6XlDoyRJkobRjDXXSXYCfgYcDnwC+Gk7JN/AOBSfJEmShlE3ZSEfAF5QVZcCJHkMcAzwtH4GJkmSJI2abkYLWWc8sQaoqp8yZKOHSJIkScOgm57rFUmOAL7Qzu8HrOhfSDOz5lqSJEnDqJue61fTPJ3xte10cbtsYKy5liRJ0jDqZrSQ24EPtpMkSZKkKUyZXCc5rqr+d5IfAzVxfVU9ua+RSZIkSSNmup7rQ9p/95iPQGbDmmtJkiQNoylrrqvqmvbl31XVlZ0T8HfzE96UsVlzLUmSpKHTzQ2Nz59k2a69DkSSJEkaddPVXL+apof60Uku6Fj1YOC7/Q5MkiRJGjXT1Vz/O/AN4D3AoR3Lb6uqm/salSRJkjSCpqu5Xl1VK4G3Ate2tdZbA3+VZMN5im9SSZYkWbZ69epBhiFJkiTdQ6ruNcrePTdIzgfGgEXAycDXgCdU1W59j24GY2NjtWLF/D8sctGhJ837MWey8rDdBx2CJEnSfVaSc6tqbKbturmh8a6quhPYG/hYVf0D8PA1DVCSJEm6r+kmuf59kn2Bvwa+3i5bp38hSZIkSaOpm+T6AOCZwLur6ookWwOf729YkiRJ0uiZbrQQAKrq4iRvBLZq568A3tvvwCRJkqRRM2PPdfuo8fOBb7bz2yU5sd+BSZIkSaOmm7KQdwA7AL8EqKrzgUf1MSZJkiRpJHV1Q2NVTRxQ+q5+BNMtx7mWJEnSMOomub4oycuABUm2SfIx4Ht9jmtaVbW8qpYuXLhwkGFIkiRJ99BNcv0a4AnA7cAxwK3A6/oZlCRJkjSKuhkt5DfAW9pJkiRJ0hRmTK6TnA7c6xnpVbVzXyKSJEmSRtSMyTXwho7XDwBeDNzZn3AkSZKk0dVNWci5ExZ9N8nZfYpHc7To0JNmtf3Kw3bvUySSJEn3X92UhWzcMbsW8DTAYTokSZKkCbopCzmXpuY6NOUgVwAH9jMoSZIkaRR1Uxay9XwEkuTxwCHApsBpVfXJ+TiuJEmS1CvdlIXsPd36qvrKNO89CtgDuL6qntixfBfgI8AC4IiqOqyqLgFelWQt4HOAybUkSZJGSjdlIQcCzwK+3c4/l+YJjTfQlItMmVwDRwMfp0mWAUiyADgceD6wCjgnyYlVdXGSPYFXA5+f3WlIkiRJg9dNcr0OsG1VXQOQ5OHA0VV1wExvrKqzkiyasHgH4LKqurzd37HAXsDFVXUicGKSk4B/7/osJEmSpCHQTXK95Xhi3boO2GoNjrk5cFXH/CpgxyQ7AXsD6wEnT/XmJEuBpQBbbbUmYUiSJEm91U1yfVqSU4Bj2vmXAt/qdSBVdQZwRhfbLQOWAYyNjd3ryZGSJEnSoHQzWsjBSf4CeHa7aFlVfXUNjnk1sGXH/Bbtsq4lWQIsWbx48RqEIUmSJPVWNz3XtMn0miTUnc4BtkmyNU1SvQ/wstnsoKqWA8vHxsYO6lFMkiRJ0hrrKrmeqyTHADsBmyZZBby9qo5McjBwCs1QfEdV1UWz3K8912vIx6VLkiT1Xl+T66rad4rlJzPNTYtd7Neea0mSJA2dtaZakeS09t/3zl843UmyJMmy1atXDzoUSZIk6W5TJtfAw5M8C9gzyVOTbN85zVeAk6mq5VW1dOHChYMMQ5IkSbqH6cpC3gb8E81oHh+csK6AnfsVlCRJkjSKpkyuq+p44Pgk/1RV75rHmGbkDY2SJEkaRtOVhQBQVe9KsmeS97fTHvMR2AwxWRYiSZKkoTNjcp3kPcAhwMXtdEiSf+13YJIkSdKo6WYovt2B7arqLoAknwV+CLy5n4FpuMx2XGxwbGxJknT/M2PPdWvDjtcDr8VwKD5JkiQNo26S6/cAP0xydNtrfS7w7v6GNT1rriVJkjSMZiwLqapjkpwBPL1d9MaquravUUmSJEkjqKvHn1fVNcCJfY5FkiRJGmnd1lwPFWuuJUmSNIy66rkeNlW1HFg+NjZ20KBj0dRmO8KIo4tIkqRRN23PdZIFSX4yX8FIkiRJo2za5Lqq/gBcmmSreYpHkiRJGlndlIVsBFyU5Gzg1+MLq2rPvkUlSZIkjaBukut/6nsUs5RkCbBk8eLFgw5FkiRJutuMo4VU1ZnASmCd9vU5wHl9jmummHyIjCRJkobOjMl1koOA44FPt4s2B07oZ1CSJEnSKOqmLOTvgR2AHwBU1c+S/Elfo9L9kkP3SZKkUdfNQ2Rur6o7xmeSrA1U/0KSJEmSRlM3yfWZSd4MPDDJ84EvAcv7G5YkSZI0erpJrg8FbgB+DLwSOBl4az+DkiRJkkbRjDXXVXVXks/S1FwXcGlVWRYiSZIkTTBjcp1kd+BTwM+BAFsneWVVfaPfwU0Tk+NcyxsgJUnS0OmmLOQDwHOraqeqeg7wXOBD/Q1reo5zLUmSpGHUTXJ9W1Vd1jF/OXBbn+KRJEmSRtaUZSFJ9m5frkhyMnAcTc31S2ie0ihJkiSpw3Q110s6Xl8HPKd9fQPwwL5FJEmSJI2oKZPrqjpgPgORJEmSRl03o4VsDbwGWNS5fVXt2b+wJEmSpNEzY3INnAAcSfNUxrv6G44kSZI0urpJrn9XVR/teySSJEnSiOsmuf5IkrcDpwK3jy+sqvN6HUySFwG7Aw8BjqyqU3t9DEmSJKlfukmunwTsD+zMH8tCqp2fUZKjgD2A66vqiR3LdwE+AiwAjqiqw6rqBOCEJBsB76dJ6CVJkqSR0E1y/RLgUVV1xxyPcTTwceBz4wuSLAAOB54PrALOSXJiVV3cbvLWdr0kSZI0Mrp5QuOFwIZzPUBVnQXcPGHxDsBlVXV5m7QfC+yVxnuBb/Sj7ESSJEnqp256rjcEfpLkHO5Zc70mQ/FtDlzVMb8K2JFmyL/nAQuTLK6qT018Y5KlwFKArbbaag1C0P3NokNPmtX2Kw/bvU+RSJKk+6pukuu39z2KVjsqybQjk1TVMmAZwNjYWM1HXJIkSVI3Zkyuq+rMPhz3amDLjvkt2mVdSbIEWLJ48eJexyVJkiTN2Yw110luS3JrO/0uyR+S3LqGxz0H2CbJ1knWBfYBTuz2zVW1vKqWLly4cA3DkCRJknpnxuS6qh5cVQ+pqocADwReDHyi2wMkOQb4PvDYJKuSHFhVdwIHA6cAlwDHVdVFs9jnkiTLVq9e3e1bJEmSpL5L1ezLlpP8sKqe2od4ZmVsbKxWrFgx78ed7Y1xun/wBkhJku67kpxbVWMzbTdjzXWSvTtm1wLGgN+tQWySJEnSfVI3o4Us6Xh9J7AS2Ksv0XTJGxolSZI0jLoZLeSA+QhkNqpqObB8bGzsoEHHIkmSJI2bMrlO8rZp3ldV9a4+xCNJkiSNrOlGC/n1JBPAgcAb+xzXtBwtRJIkScNoyuS6qj4wPtE8EfGBwAHAscCj5im+qWJznGtJkiQNnWlrrpNsDPxfYD/gs8D2VXXLfAQmSZIkjZope66TvI/mSYq3AU+qqncMS2JtWYgkSZKG0XQ1168HHgG8Ffifjkeg39aDx5+vEctCJEmSNIymLAupqhkfjS5JkiTpj0ygJUmSpB4ZyeTammtJkiQNo5FMrq25liRJ0jAayeRakiRJGkbTjnMtqXuLDj2p78dYedjufT+GJEmaO5NraYTMNoE3GZckaX5ZFiJJkiT1yEgm144WIkmSpGE0ksm1o4VIkiRpGI1kci1JkiQNI5NrSZIkqUdMriVJkqQeMbmWJEmSesRxrqX7MMfFliRpfo1kz7VD8UmSJGkYjWRy7VB8kiRJGkYjmVxLkiRJw8iaa0l3s0ZbkqQ1Y8+1JEmS1CMm15IkSVKPWBYiac4sI5Ek6Z7suZYkSZJ6xORakiRJ6pGhSa6TPCrJkUmOH3QskiRJ0lz0NblOclSS65NcOGH5LkkuTXJZkkMBquryqjqwn/FIkiRJ/dTvnuujgV06FyRZABwO7ApsC+ybZNs+xyFJkiT1XV+T66o6C7h5wuIdgMvanuo7gGOBvfoZhyRJkjQfBlFzvTlwVcf8KmDzJJsk+RTw1CRvmurNSZYmWZFkxQ033NDvWCVJkqSuDc0411V1E/CqLrZbBiwDGBsbq37HJUmSJHVrED3XVwNbdsxv0S7rWpIlSZatXr26p4FJkiRJa2IQyfU5wDZJtk6yLrAPcOJsdlBVy6tq6cKFC/sSoCRJkjQX/R6K7xjg+8Bjk6xKcmBV3QkcDJwCXAIcV1UXzXK/9lxLkiRp6PS15rqq9p1i+cnAyWuw3+XA8rGxsYPmug9JkiSp14bmCY2SJEnSqBvJ5NqyEEmSJA2jkUyuvaFRkiRJw2gkk2tJkiRpGA3NQ2RmI8kSYMnixYsHHYqkWVh06Emzfs/Kw3bvQySSJPXHSPZcWxYiSZKkYTSSybUkSZI0jEYyuXa0EEmSJA2jkUyuLQuRJEnSMBrJ5FqSJEkaRibXkiRJUo+MZHJtzbUkSZKG0Ugm19ZcS5IkaRiNZHItSZIkDSOTa0mSJKlHTK4lSZKkHjG5liRJknpkJJNrRwuRJEnSMBrJ5NrRQiRJkjSMRjK5liRJkoaRybUkSZLUIybXkiRJUo+YXEuSJEk9YnItSZIk9cjagw5gLpIsAZYsXrx40KFIGjKLDj1pVtuvPGz3PkUiSbo/Gsmea4fikyRJ0jAayeRakiRJGkYm15IkSVKPmFxLkiRJPWJyLUmSJPWIybUkSZLUIybXkiRJUo+YXEuSJEk9YnItSZIk9cjQPKExyfrAJ4A7gDOq6osDDkmSJEmalb72XCc5Ksn1SS6csHyXJJcmuSzJoe3ivYHjq+ogYM9+xiVJkiT1Q7/LQo4GdulckGRv8iqbAAAMsElEQVQBcDiwK7AtsG+SbYEtgKvazf7Q57gkSZKknutrWUhVnZVk0YTFOwCXVdXlAEmOBfYCVtEk2OczTdKfZCmwFGCrrbbqfdCShsqiQ08a6f2vPGz3vu4fhu8cZhvPfLTR/dGwfQ7DFo+Gw33xuhjEDY2b88ceamiS6s2BrwAvTvJJYPlUb66qZVU1VlVjm222WX8jlSRJkmZhaG5orKpfAwd0s22SJcCSxYsX9zcoSZIkaRYG0XN9NbBlx/wW7bKuVdXyqlq6cOHCngYmSZIkrYlBJNfnANsk2TrJusA+wIkDiEOSJEnqqX4PxXcM8H3gsUlWJTmwqu4EDgZOAS4Bjquqi2a53yVJlq1evbr3QUuSJElz1O/RQvadYvnJwMlrsN/lwPKxsbGD5roPSZIkqddG8vHn9lxLkiRpGI1kcu0NjZIkSRpGI5lcS5IkScNoJJNry0IkSZI0jEYyubYsRJIkScNoJJNrSZIkaRilqgYdw5wluQG4cp4Otylw4zwd677Edpsb221ubLe5sd3mxnabG9ttbmy3uelluz2yqjabaaORTq7nU5IVVTU26DhGje02N7bb3Nhuc2O7zY3tNje229zYbnMziHazLESSJEnqEZNrSZIkqUdMrru3bNABjCjbbW5st7mx3ebGdpsb221ubLe5sd3mZt7bzZprSZIkqUfsuZYkSZJ6xOR6Bkl2SXJpksuSHDroeIZVki2TnJ7k4iQXJTmkXb5xkv9M8rP2340GHeswSrIgyQ+TfL2d3zrJD9rr7j+SrDvoGIdNkg2THJ/kJ0kuSfJMr7eZJfk/7c/ohUmOSfIAr7d7S3JUkuuTXNixbNLrK42Ptu13QZLtBxf5YE3Rbu9rf04vSPLVJBt2rHtT226XJnnhYKIeDpO1Xce61yepJJu2815zranaLclr2uvuoiT/1rG879ecyfU0kiwADgd2BbYF9k2y7WCjGlp3Aq+vqm2BZwB/37bVocBpVbUNcFo7r3s7BLikY/69wIeqajFwC3DgQKIabh8BvllVjwOeQtN+Xm/TSLI58FpgrKqeCCwA9sHrbTJHA7tMWDbV9bUrsE07LQU+OU8xDqOjuXe7/SfwxKp6MvBT4E0A7e+IfYAntO/5RPt79/7qaO7ddiTZEngB8IuOxV5zf3Q0E9otyXOBvYCnVNUTgPe3y+flmjO5nt4OwGVVdXlV3QEcS/NhaYKquqaqzmtf30aT6GxO016fbTf7LPCiwUQ4vJJsAewOHNHOB9gZOL7dxHabIMlC4NnAkQBVdUdV/RKvt26sDTwwydrAg4Br8Hq7l6o6C7h5wuKprq+9gM9V47+BDZM8fH4iHS6TtVtVnVpVd7az/w1s0b7eCzi2qm6vqiuAy2h+794vTXHNAXwI+Eeg8yY5r7nWFO32auCwqrq93eb6dvm8XHMm19PbHLiqY35Vu0zTSLIIeCrwA+ChVXVNu+pa4KEDCmuYfZjmP8672vlNgF92/DLyuru3rYEbgM+05TRHJFkfr7dpVdXVND04v6BJqlcD5+L11q2pri9/V3TvFcA32te22wyS7AVcXVU/mrDKtpveY4A/a8vdzkzy9Hb5vLSbybV6KskGwJeB11XVrZ3rqhmaxuFpOiTZA7i+qs4ddCwjZm1ge+CTVfVU4NdMKAHxeru3tkZ4L5o/Th4BrM8kX0NrZl5fs5fkLTQlhF8cdCyjIMmDgDcDbxt0LCNobWBjmjLVfwCOa78Vnhcm19O7GtiyY36LdpkmkWQdmsT6i1X1lXbxdeNfVbX/Xj/V+++n/hewZ5KVNGVHO9PUEm/Yfm0PXneTWQWsqqoftPPH0yTbXm/Tex5wRVXdUFW/B75Ccw16vXVnquvL3xUzSPJyYA9gv/rjGMC22/QeTfOH8I/a3xFbAOcleRi23UxWAV9py2bOpvlmeFPmqd1Mrqd3DrBNeyf9ujRF8CcOOKah1P5FeCRwSVV9sGPVicDftK//BvjafMc2zKrqTVW1RVUtorm+vl1V+wGnA3/Zbma7TVBV1wJXJXlsu+jPgYvxepvJL4BnJHlQ+zM73m5eb92Z6vo6EfjrdgSHZwCrO8pH7veS7EJT+rZnVf2mY9WJwD5J1kuyNc3NeWcPIsZhVFU/rqo/qapF7e+IVcD27f9/XnPTOwF4LkCSxwDrAjcyX9dcVTlNMwG70dzd/HPgLYOOZ1gn4E9pviK9ADi/nXajqR8+DfgZ8C1g40HHOqwTsBPw9fb1o9of+MuALwHrDTq+YZuA7YAV7TV3ArCR11tX7fbPwE+AC4HPA+t5vU3aTsfQ1KX/niapOXCq6wsIzchSPwd+TDMay8DPYYja7TKaOtfx3w2f6tj+LW27XQrsOuj4h63tJqxfCWzavvaam6bdaJLpL7T/z50H7Nyxfd+vOZ/QKEmSJPWIZSGSJElSj5hcS5IkST1ici1JkiT1iMm1JEmS1CMm15IkSVKPmFxLGmpJKskHOubfkOQdPdr30Un+cuYt1/g4L0lySZLT+32sUZXkfUkuSvK+eT7udkl265jfM8mh071Hkqaz9sybSNJA3Q7sneQ9VXXjoIMZl2Ttqrqzy80PBA6qqv/qZ0zTSbKgqv4wqON3YSnNuNHzHeN2wBhwMkBVnYgPC5O0Buy5ljTs7gSWAf9n4oqJPc9JftX+u1OSM5N8LcnlSQ5Lsl+Ss5P8OMmjO3bzvCQrkvw0yR7t+xe0PannJLkgySs79vudJCfSPNlwYjz7tvu/MMl722Vvo3nI0pETe2Xbp6u9r93+x0le2rHuje2yHyU5rF22OMm32mXnJXl0G9PXO9738fZR0yRZmeS9Sc4DXtJu/80k57bn8biOdvxoku+17fWXM8Qx1X5e0p7Lj5KcNUn7THq+bXtuAJzb2Qbtuk2SnNr2ah+R5MokmyZZlOTCju3u/kaj2/jSPHn3ncBLk5yf5KVJXp7k4+32i5J8u70GTkuy1UztJUn2XEsaBYcDFyT5t1m85ynA44GbgcuBI6pqhySHAK8BXtdutwjYAXg0cHqSxcBf0zxO+OlJ1gO+m+TUdvvtgSdW1RWdB0vyCOC9wNOAW4BTk7yoqt6ZZGfgDVW1YkKMe9P0nD4F2BQ4p01KtwP2Anasqt8k2bjd/ovAYVX11SQPoOkg2XKGdripqrZvYzwNeFVV/SzJjsAngJ3b7R5O80fA42h6bo9PsusUcSybYj9vA15YVVcn2XCSWCY936raM8mvqmq7Sd7zduC/2nbcneZbgJl0FV9V3dH+8TNWVQe3bfTyjv18DPhsVX02ySuAjwIvmqq9uohL0v2AybWkoVdVtyb5HPBa4Lddvu2cqroGIMnPgfHk+MfAczu2O66q7gJ+luRymmTpBcCTO3okFwLbAHcAZ09MrFtPB86oqhvaY34ReDbNo9mn8qfAMW0pxHVJzmz38xzgM1X1m/b8b07yYGDzqvpqu+x37XFmaof/aLfbAHgW8KWO96zXsd0JbTtcnOSh7bLnTRLHdPv5LnB0kuOAr8zifKcrw3g2TVJOVZ2U5JbpTnYN45vomePHpnlUfOcfd5O1lySZXEsaGR8GzgM+07HsTtrytiRrAet2rLu94/VdHfN3cc//+2rCcQoI8JqqOqVzRZKdgF/PLfy+ubsNWg+YsH483rWAX07ROwz3bK/pMvYp91NVr2p7inenKfF4WlXdNG30czfVec8qvjU4frftJel+xpprSSOhqm4GjuOeZQEracowAPYE1pnDrl+SZK00ddiPAi4FTgFenWQdgCSPSbL+DPs5G3hOWw+8ANgXOHOG93yHpt53QZLNaHppzwb+EzggyYPa429cVbcBq5K8qF22Xrv+SmDbdn5D4M8nO1BV3QpckeQl7fuT5CkzxDdZHFPuJ8mjq+oHVfU24AbuXbIy1flO5yzgZe3+dwU2apdfB/xJW5O9HrDHTOc5RXy3AQ+e4tjfA/ZpX+/Xxi9J0zK5ljRKPkBTqzvu/9EktD+i+Qp/Lr3Kv6BJ8L5BU6f7O+AImhsWz2tvmvs0M3zT15agHAqcDvwIOLeqvjbDsb8KXNBu/23gH6vq2qr6Jk2pxIok5wNvaLffH3htkgtoEr+HVdVVNH90XNj++8NpjrcfcGDbXhfR1FNPd05TxTHVft6X9obONr4fdXO+08UA/DPw7CQX0ZRo/KKN7fc0NyOO/zHyky7Oc7L4Tqf54+T8TLiZkqY2/4C2vfcHDpkhVkkiVRO/EZUkaTglWUlzA+LQDMsoSZ3suZYkSZJ6xJ5rSZIkqUfsuZYkSZJ6xORakiRJ6hGTa0mSJKlHTK4lSZKkHjG5liRJknrE5FqSJEnqkf8PYdJ6P8URuhYAAAAASUVORK5CYII=\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7fb476cee588>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "print('Total number of question pairs for training: {}'.format(len(df_train)))\n",
    "print('Duplicate pairs: {}%'.format(round(df_train['is_duplicate'].mean()*100, 2)))\n",
    "qids = pd.Series(df_train['qid1'].tolist() + df_train['qid2'].tolist())\n",
    "print('Total number of questions in the training data: {}'.format(len(np.unique(qids))))\n",
    "print('Number of questions that appear multiple times: {}'.format(np.sum(qids.value_counts() > 1)))\n",
    "\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Test Set"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "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>test_id</th>\n",
       "      <th>question1</th>\n",
       "      <th>question2</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>0</td>\n",
       "      <td>How does the Surface Pro himself 4 compare wit...</td>\n",
       "      <td>Why did Microsoft choose core m3 and not core ...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>1</td>\n",
       "      <td>Should I have a hair transplant at age 24? How...</td>\n",
       "      <td>How much cost does hair transplant require?</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>2</td>\n",
       "      <td>What but is the best way to send money from Ch...</td>\n",
       "      <td>What you send money to China?</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>3</td>\n",
       "      <td>Which food not emulsifiers?</td>\n",
       "      <td>What foods fibre?</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>4</td>\n",
       "      <td>How \"aberystwyth\" start reading?</td>\n",
       "      <td>How their can I start reading?</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   test_id                                          question1  \\\n",
       "0        0  How does the Surface Pro himself 4 compare wit...   \n",
       "1        1  Should I have a hair transplant at age 24? How...   \n",
       "2        2  What but is the best way to send money from Ch...   \n",
       "3        3                        Which food not emulsifiers?   \n",
       "4        4                   How \"aberystwyth\" start reading?   \n",
       "\n",
       "                                           question2  \n",
       "0  Why did Microsoft choose core m3 and not core ...  \n",
       "1        How much cost does hair transplant require?  \n",
       "2                      What you send money to China?  \n",
       "3                                  What foods fibre?  \n",
       "4                     How their can I start reading?  "
      ]
     },
     "execution_count": 7,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_test = pd.read_csv('./input/test.csv')\n",
    "df_test.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Total number of question pairs for testing: 2345796\n"
     ]
    }
   ],
   "source": [
    "print('Total number of question pairs for testing: {}'.format(len(df_test)))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Text analysis"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "mean-train 59.82 std-train 31.96 mean-test 60.07 std-test 31.62 max-train 1169.00 max-test 1176.00\n"
     ]
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAA40AAAJjCAYAAACobfR/AAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADl0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uIDIuMS4xLCBodHRwOi8vbWF0cGxvdGxpYi5vcmcvAOZPmwAAIABJREFUeJzs3XuYJXddJ/73hySQhFvixHBJwIlLRCCzAYmRLIiMyN0VUEAUl7DrEpyBFX6rSGaVGFjdILiALAwQ1+CARkBYIEqQcJkIKpdMMHmGS4AAgUyAEMIMJOAAge/vj6pJTp/p6stMd5++vF7P08+cqlNV51NVp8+cd3+/9a1qrQUAAACmc6tJFwAAAMDyJTQCAAAwSGgEAABgkNAIAADAIKERAACAQUIjAAAAg4RGYMFV1dlV1arq3dM895aqungCZR2Qfj+eNTJ9cVW9ZYle++tVdfYsy0ypb2CZp/XL3W6Or/sT/Tk8ah7lrnpV9diq+lRVfa+qrjqI7cx6zpaDqjp1tvffalVVT6qqp81x2RVxPuerqo7tPwfWj81/SL/PJ02mMmAShEZgMT28qn560kUssM1Jtky6iHl6Z5LTknxnjsv/RJI/TCI09qrqkCSvT3J5kp9P8vjJVrQkTk33PliLnpTkaXNc9rQkf7t4pUzMsenO//qx+R9Lt8+fW+qCgMk5dNIFAKvWN5Jck+T3kzxuoTdeVUe01v5tobc7m9baJ5f6NQ9Wa+26JNdNuo6ZVFUluU1rbe+kaxlwlyR3SHJ+a+2fJl3MkEn9XszFcq7tYLTWPjzpGpZSa+1bSdbUPgNaGoHF05L8cZJfqqoNMy1YVfetqvdV1XeqandV/XVV3Wnk+fV9d6inVNXrq2pPkr/rn7uqqv60qs6sqq9U1Ter6n9X59FV9YmquqGq3l5VR49s87ZV9cqq+nT/ul+oqldV1R1mqXVK99SqOr6q3lxVX6uqf6uqz1XV/xxb52er6h/717m+qv68qm4/tsyDq+ryqtpbVZdW1X+YwzHe55Cq+l9VdV1fx6uq6jYj296ve2pVbamqK/vXu7aq/qGq7lxVD9l3bJN8oV/vqpH1ZjxX/TJ3r6p39cfjC/3rT+mW3Hd7+3pVPaiqLkmyN8kT53pe+rr+v/5cX99v63f7506vqs9X1Z6qOq+qDp/tAPbdEXdW1Xer6uqq+uOqOnTf8Utydb/oO/rXPnuGba2rqtf278e9/b48Z2yx2c7ZXfraP98fx89U1R9V1a1Hlpnp9+KpVfVPVfWN/jxtr6pTpqn1wf1zN1b3u3NxVd2v3+f/M3Ks29j5O6mq3lnd79YNVfW3VXXnkef3dWF8RFVdUFU3JnnlgR6zqjqyql5RVV/tn7+kqh4+to2rqupPx+ZNee+P1PWQvuYb+2O8eWSdv0zyK0l+bmTfz56h9mm7sFfVr1f3O/at/vfh+KFtjKy73+dAjXVTn8t+9vN+pKrOre73e29V/UtV/czYer9ZVZ/s32Nfr+5z6j7VdUnd2S+2fd9xGDuGJ41sZy7nZ07HpQY+m2Y7dsDi0tIILKa/TfLCdK2NT55ugar60SQXJ/lUkl9PcrskL0rynqo6pbX2vZHF/zTJ/0vyxCQ/GJn/5CQfTfKfk9w/yR+l+6PYg5M8P8kR6b6wnpPkt/p1jkxySF/bdUnu1j/+2ySPmMc+vr7f/hlJ9iT58SQ/ObJ/D0zy3iRvT/KEJOv6/Tu6n05V3TXJu/p9eEKSuyb5677GufidJO9P8htJ/n2/n19M8uLpFq6qpyb5H0mel+QTfU0/n+S26bqe/W66Y/3LSb6S5Lv9erOeq6qqJBek69r6X9KFwecn+dHs353tyCTb+jo/k+TLmd95+Z10XW9/LckvJnlJVR2b5KeT/HaSuyd5Wb/tFw0dvP7L7ZvSncvn9sfwf/bH5bf61/jldO+9303yz0l2DWzriP4YHZvkBUmuSHKP/me89pnO2THpWuv/e5Ld6boMn53uOD5jbFvT/V6s7/fnc0lu3R+jD1bVfVprn+9rfUiS9yTZnuT0JN9O8sAkx/X7/L/7Ok/rt/mtfr179MdgR1//of3x+ruqOrW11kZq+4skr0vy8nTvhQM9Zn+e5JfSvW+vTPL0JO+sqo0H2PL75+nee+emOzavqqodrbWP9vty93Tv4X1hctrzPYOfSfd7/DvpPh/+rH+tRw+tsACfA6Pbuk26z52j0r2nv5ZkU5L3VtWJrbWvVtWDk7wmyVlJPpSuJf20JHdMd4yf0r/+M9N9LsxkrudnxuMyy2cTMEmtNT9+/PhZ0J90X26/3j9+Wrovsj/RT78lycUjy74oXdi6w8i8n0nXUvlr/fT6fvpt07zWVem+pBwyMu+jSW5KcsLIvBcnuXaGmg9N94W5Jbn7yPyW5Fkj0xcnecvI9I1J/uMM2/1gku1j836+3+5JI7Vdn+TIkWWe0i9z9izHuiX5wNi8tyf58Mj00/rlbtdPvzLJW2fY5i/2y68fmz+Xc/WYfvqnR5Y5Lsn3x8772f1yj51l/2Y6L9tHpm+VLuDuHqvvzUk+MstrfHiac/R7/fv2+LH34C/Osq1nJPlhkvsezDkbOA6/ni543Xq234uxdW/Vr39FkrNG5n8oXfCrgfWelaRNM/8NST69r45+3on98XpMP/2QvraXzVTbXI5Zknv1z58+tk8fT/LukXlXJfnTsXXH3/v76nrhyDKHpfsDxYtG5k35nJrD7+D4Z8Q3kxw9Mu85/XJHzLCdOX0OzHE/fzPJ95KcOPYe+lySl/TTv5vk0hnqOanf5kPG5u87hvs+v+Z6fmY9Lpnls8mPHz+T+9E9FVhsf5XkSxkePObUJBe17jqZJElr7SPpvhg9aGzZdw5s4+LW2mjL45VJrmqtfWFs3o/W1O59/6mq/rXvOvf9JPv+Iv4TM+/SFJclOafvHnb30Seq6sh0f7l/c1Uduu+nf53vp2sVTbpj8J7W2uhANW+bRw0XjU1/MslMXeEuS/LoqnpBdSNkHjLH15nLufrpJF9trV0yssw1SS6dZnstXcvKFPM4L+8beY0fJvlCui/B3xpZ5sp0oXVa/b7/VPYfyORN6b74nrbfSjP7+ST/2lq7bJblZjxn1XnOvq6D6Y7DXye5TbpWsFH7/V5U1b2q6m1VdW26MPf9JPdMfwyr6rbpAv+21lobX38Wv5Du/fnDkff0F9K9D8a7wA79zo6a7Zj9dJLKyDnqz/ffZv/PiLm6+fi31r6f5LOZ+Xdmvi5pre0emd53LfTgezEH/zkw6hfS/c59YeQcJck/5pZzdFmS+1XVy/pusbeebkNzMJ/zM9txOdDPJmCRCY3Aomqt3ZTuL+i/UVU/Ns0id0ly7TTzr03yI9PMm86esenvDcyrdF31UlWPT9d970PpuvU9ILeMiDnrNXAjfjVda83Lknyxqi6rqof2zx2drqvl1nRf2vf9fDdd68bd+uXunK772M36L443zrGG6fZ1pn04L10XsCcl+UiSa6u7Xm62L2hzOVd3zvSD7kw3b3eb2v14vudlrud9pmNxTLpzMb5f+6bH34OzWZeuxXM2s9X5nHTdTt+W5LHpAsUz++fG92dK7dVdL3tRuvfXf0/ys+m+2F8+su7R6X4f5lLruGPSdR/8/tjPj+eW9/S0tQ2Y7ZjdJcmNY2Fq37aPrJFrQedhvu+Thdh+ZnmNg/0cGHVMut+d8XP0n9Ofo9bae/vpB6drBfx6ddfWzrcr6HzOz2zH5UA/m4BF5ppGYCmcl+QP0n3RHPeVdNcyjbtT9m+dmm+LyEyemK7b4ugAGD833430rWhPq6pbpftif3aSC/pWxz3pu5YluXCa1b/c//vVjB2DvpVyTvdVPICaf5gu5L6squ6WrgvcH6e7bus1M6w6l3P11XTX3Y370ex/Tdt053NBzss8fD3dl+nx/do3uM835rm967P/9YsH4onpukH//r4ZVXXvgWXHj+Np6VrNHtZau2Jk/TuOLLM7XZfCuxxAbd9IF2b/7zTPfX2W2qYz2zH7SpLbVdWRY8HkTkm+01r7bj+9N/0fhUYcnZVjrp8Dc9nPb6T7Y9amaV5n3/FKa21bkm399cq/nO5z4YYkZ86j7rmen1kdxGcTsMi0NAKLrv/S8KfpBkYZ/5L6kSSPqJHRRKu7t+P63NItcTEckZEvT72nHOjGWms/bN3Q+y9IN3DFj7XWvp3uerl7ttZ2TPOzLzRekuRh/RfEfZbkPoCttatbay9K141zXygZahWZy7m6JMmdq+rUkWWOyy1dcWezoOdlNn235kvThbRRT0oXqj40z02+L12Xv39/kKUdzHE4ov/35vWrG413/b7p/r35kSRP7Qcvms73+nXH3wfvS3KfdF2Bx9/TV82xxvHtzXTMLkkXPp8wsj/VT49+RuxKd33dqIfnwCx0y+NczPVzYC77+b50QfxL05yjnWPLprV2XWvttemuwZ7tc2C6uudyfuZl4LMJmBAtjcBSeW26bkf/Id11Nfu8NN1fw99dVX+SW0bk3JnkrYtYz3vSjZj4++m+PD86yUNnXmWqvuXm3em6U34m3fVmv5OuxeBT/WK/l+R9VfXDdINr3JDumrTHJPn91tpn0o0s+cwkf19VL003uuCWJItyT7uqem26logPpxuYYmO6gUz2tQR/uv/3GVX1xnStBTszt3N1YbpukG+uqn378Ifpuqr9cA7lHfR5OQB/mG6fXpfkjUk2pBtB889ba/MdNfP16c7lRdXdJuHTSU5INxDUfFpv3pPkt6vqI+kGL3lK5t6C+eF0XRr/vKpenK7V8ex0900ddWa6ETbfVVXnphs99bQkO1prf59u4JwkeXZVvT/Jt1prn+639dF0o2Oel6518bgkD0vyl621i+exn8ksx6y19qmq+pskr+z/YPG5dKNz/mSmtqS9Lcn/qar/kS7I/Eq6cHsgrkjy2Kp6XLqQ9uWRP/Islrl+DsxlP1+fbuTfi6u7Pcfn03UDPjXdNccvq6oXpOt+fXG6c3i/JD+XW1oZv9S/9ulV9c0k32+t7Rgveh7nZ1Zz+GwCJkRLI7Ak+m5LL5tm/nXpvhjsTfI3SV6V7q/dDxu/3m2BvTbdLQWene52BT+WbnTK+dibLjA9O91tJrYl+U6Sh7f+JuatG27+wem6Z74h3X30fi/dff+u7Ze5Jl04OiZd+Nqc7lYG49cILZQP9TW9Ll3Ie3ySp7fW3t7X88V0Iyv+crpbK/xdP3/Wc9UPqvLYdF+6X5duSP1XpxvwYnSAmiELcV7mpbV2UbrbtpySbl+f09fwrJnWG9jW3nQDu/xdutvNvCvd+Z5v4HhhumP8R/2/30t3G5G51HBtupbTOyd5R7r9+a10LTajy30gXdA7Mt2AVW9KFxr2BeUPJnlJunPxkXTnJv0fOh6Q7v15br+PL0jXsjnlNeZY71yO2dPT/X6d1e/Tj6UbyXa0JevcdMHrt9ONmvvddMfvQGxNd13oeemC2RkHuJ05m8fnwKz72R/Tjen++PCCdPvyZ+kC2Ef7xS5J14L3mnR//NqU7g8Cfzayjaen6yXwj/3yQ+ZyfuZixs8mYHJq/oOmAcDc9S2yn0/yytbaH066HlhJqurr6X53zp50LcDapXsqAAuqqn4rXVfUz6ZrYf3v6brunjfJugCAAyM0ArDQ9qa7BunH0g2Q8dEkv9B3ewUAVhjdUwEAABhkIBwAAAAGrdnuqcccc0xbv379pMsAAACYiEsvvfTrrbUfnW25NRsa169fnx079rvdEAAAwJpQVXMab0D3VAAAAAYJjQAAAAwSGgEAABi0Zq9pBAAA1q7vf//72bVrV/bu3TvpUhbd4YcfnuOPPz6HHXbYAa0vNAIAAGvOrl27cvvb3z7r169PVU26nEXTWsv111+fXbt25YQTTjigbeieCgAArDl79+7NunXrVnVgTJKqyrp16w6qRVVoBAAA1qTVHhj3Odj9FBoBAAAY5JpGAABgzduwbcOCbm/n6TtnXWbPnj05//zzs3nz5nlt+9GPfnTOP//8HHXUUQda3rxoaQQAAJiAPXv2ZOvWrfvNv+mmm2Zc78ILL1yywJhoaQQAAJiIM888M5/73Ody3/veN4cddlgOP/zwHH300bniiivymc98Jo973ONy9dVXZ+/evXn2s5+dM844I0myfv367NixIzfeeGMe9ahH5UEPelD+5V/+Jccdd1ze8Y535IgjjljQOrU0AgAATMCLXvSi/Lt/9+9y2WWX5SUveUk+9rGP5c/+7M/ymc98Jkly3nnn5dJLL82OHTvyile8Itdff/1+2/jsZz+bZz7zmfnEJz6Ro446Km9961sXvE4tjQAAAMvAqaeeOuVeiq94xSvytre9LUly9dVX57Of/WzWrVs3ZZ0TTjgh973vfZMk97///XPVVVcteF1CIwAAwDJw29ve9ubHF198cd773vfmQx/6UI488sg85CEPmfZei7e5zW1ufnzIIYfk3/7t3xa8Lt1TAQAAJuD2t799brjhhmmf++Y3v5mjjz46Rx55ZK644op8+MMfXuLqbqGlEQAAWPPmcouMhbZu3bo88IEPzEknnZQjjjgid7rTnW5+7pGPfGRe85rX5F73ulfuec975gEPeMCS17dPtdYm9uKTdMopp7QdO3ZMugwAAGACPvWpT+Ve97rXpMtYMtPtb1Vd2lo7ZbZ1dU8FAABgkNAIAADAIKERAACAQUIjAAAAg4RGAAAABgmNAAAADHKfRgAAgO3nLOz2Nm6ZdZE9e/bk/PPPz+bNm+e9+Ze//OU544wzcuSRRx5IdfOipRGYkw3bNkz5AQDg4OzZsydbt249oHVf/vKX5zvf+c4CVzQ9LY0AAAATcOaZZ+Zzn/tc7nvf++ZhD3tYjj322Lz5zW/Od7/73Tz+8Y/PC17wgnz729/Ok570pOzatSs/+MEP8vznPz/XXnttvvzlL2fjxo055phjsn379kWtU2gEAACYgBe96EX5+Mc/nssuuywXXXRR3vKWt+SjH/1oWmv5pV/6pXzgAx/Iddddl7ve9a555zvfmST55je/mTve8Y556Utfmu3bt+eYY45Z9Dp1TwUAAJiwiy66KBdddFHud7/75ad+6qdyxRVX5LOf/Ww2bNiQ97znPXne856XD37wg7njHe+45LVpaQQAAJiw1lq2bNmSZzzjGfs997GPfSwXXnhh/uAP/iAPfehDc9ZZZy1pbVoaAQAAJuD2t799brjhhiTJIx7xiJx33nm58cYbkyTXXHNNvva1r+XLX/5yjjzyyPzGb/xGnvvc5+ZjH/vYfusuNi2NAAAAc7hFxkJbt25dHvjAB+akk07Kox71qPz6r/96TjvttCTJ7W53u/zVX/1Vrrzyyjz3uc/NrW51qxx22GF59atfnSQ544wz8shHPjJ3vetdF30gnGqtLeoLLFennHJK27Fjx6TLgBVj/DYbO0/fOaFKAAAO3qc+9anc6173mnQZS2a6/a2qS1trp8y2ru6pAAAADBIaAQAAGCQ0AgAAa9JauVTvYPfTQDjAtMavYQQAWE0OP/zwXH/99Vm3bl2qatLlLJrWWq6//vocfvjhB7wNoREAAFhzjj/++OzatSvXXXfdpEtZdIcffniOP/74A15faAQAANacww47LCeccMKky1gRXNMIAADAIKERAACAQUIjAAAAg4RGAAAABgmNAAAADBIaAQAAGOSWG8DNNmzbMOkSAABYZrQ0AgAAMEhoBAAAYJDQCAAAwCDXNAIHZPz6x52n75xQJQAALCYtjQAAAAwSGgEAABgkNAIAADBIaAQAAGCQ0AgAAMAgoREAAIBBQiMAAACDhEYAAAAGCY0AAAAMEhoBAAAYJDQCAAAwSGgEAABg0KGTLgBYOhu2bZgyvfP0nROqBACAlUJLIwAAAIO0NAILQismAMDqpKURAACAQVoaYRWZb2vf+PIAADBOSyMAAACDhEYAAAAGCY0AAAAMEhoBAAAYJDQCAAAwSGgEAABgkNAIAADAIPdpBBbdfO8fCQDA8iE0wio2HtYAAGC+lrx7alU9sqo+XVVXVtWZ0zx/m6p6U//8R6pqfT//YVV1aVXt7P/9+ZF17t/Pv7KqXlFVtXR7BAAAsHotaWisqkOSvCrJo5LcO8mvVdW9xxb7zSS7W2v3SPKyJH/Sz/96kv/YWtuQ5PQkbxhZ59VJnp7kxP7nkYu2EwAAAGvIUrc0nprkytba51tr30vyxiSPHVvmsUm29Y/fkuShVVWttX9trX25n/+JJEf0rZJ3SXKH1tqHW2styeuTPG7xdwUAAGD1W+rQeFySq0emd/Xzpl2mtXZTkm8mWTe2zK8k+Vhr7bv98rtm2WaSpKrOqKodVbXjuuuuO+CdAAAAWCtW3EA4VXWfdF1WHz7fdVtr5yY5N0lOOeWUtsClASMMwgMAsDosdUvjNUnuNjJ9fD9v2mWq6tAkd0xyfT99fJK3JXlqa+1zI8sfP8s2AQAAOABLHRovSXJiVZ1QVbdO8uQkF4wtc0G6gW6S5AlJ3t9aa1V1VJJ3JjmztfbP+xZurX0lybeq6gH9qKlPTfKOxd4RAACAtWBJu6e21m6qqmcleXeSQ5Kc11r7RFW9MMmO1toFSf4iyRuq6sok30gXLJPkWUnukeSsqjqrn/fw1trXkmxO8pdJjkjyrv4H1gTdQAEAWExLfk1ja+3CJBeOzTtr5PHeJE+cZr0/SvJHA9vckeSkha0UAACAFTcQDrA8bdq9Z8r0q48+akKVAACwkIRGYE6EQgCAtWmpB8IBAABgBdHSCCyK0ZZJrZIAACuX0AgckPHuqgAArE66pwIAADBIaAQAAGCQ0AgAAMAg1zQC01rIaxbHt7Vh24Yp0ztP37lgrwUAwMLS0ggAAMAgoREAAIBBuqcCS87tOgAAVg4tjQAAAAwSGgEAABgkNAIAADBIaAQAAGCQgXCAmxmgBgCAcVoaAQAAGCQ0AgAAMEj3VGDiNmzbMGV65+k7J1QJAADjtDQCAAAwSEsjMHEG4AEAWL6ERljDhDUAAGajeyoAAACDhEYAAAAGCY0AAAAMEhoBAAAYZCAcYPnZfs7U6Y1bJlMHAABaGgEAABgmNAIAADBIaAQAAGCQaxphDdm0e8+kSwAAYIXR0ggAAMAgLY3AsrP18q1TpjcbPRUAYGK0NAIAADBIaAQAAGCQ7qnA8rf9nFse66oKALCkhEZg2Ru9xtH1jQAAS0v3VAAAAAYJjQAAAAzSPRVWmA3bNky6BAAA1hChEVaxTbv3TLoEAABWOKERljktiwAATJJrGgEAABgkNAIAADBIaAQAAGCQ0AgAAMAgoREAAIBBQiMAAACDhEYAAAAGCY0AAAAMEhoBAAAYJDQCAAAw6NBJFwAwL9vPmTq9cctk6gAAWCOERlhFNu3eM+kSAABYZXRPBQAAYJDQCAAAwCChEQAAgEGuaQRWNgPjAAAsKi2NAAAADBIaAQAAGCQ0AgAAMMg1jcCKsvXyrVOmN5+8eUKVAACsDVoaAQAAGCQ0AgAAMEhoBAAAYJBrGmGF27R7z6RLAABgFdPSCAAAwCChEQAAgEFCIwAAAIOERgAAAAYJjQAAAAwSGgEAABgkNAIAADDIfRphwjZs2zDpEla0rZdvnTK9eeOWCVUCALA6aWkEAABgkNAIAADAIN1TgdVl+zlTp3VXBQA4KFoaAQAAGCQ0AgAAMEhoBAAAYJDQCAAAwCAD4cAKs2n3nkmXAADAGqKlEQAAgEFCIwAAAIOERgAAAAYJjQAAAAwyEA6wum0/Z+r0xi2TqQMAYIXS0ggAAMAgoREAAIBBQiMAAACDhEYAAAAGCY0AAAAMEhoBAAAYJDQCAAAwSGgEAABg0KGTLgDWmg3bNky6BAAAmDOhEZa5Tbv3TLqE1WX7Obc83rhlcnUAAKwQuqcCAAAwSGgEAABgkNAIAADAIKERAACAQUIjAAAAg4RGAAAABgmNAAAADBIaAQAAGCQ0AgAAMGjJQ2NVPbKqPl1VV1bVmdM8f5uqelP//Eeqan0/f11Vba+qG6vqlWPrXNxv87L+59il2RsAAIDV7dClfLGqOiTJq5I8LMmuJJdU1QWttU+OLPabSXa31u5RVU9O8idJfjXJ3iTPT3JS/zPuKa21HYu6AwAAAGvMUrc0nprkytba51tr30vyxiSPHVvmsUm29Y/fkuShVVWttW+31v4pXXgEAABgCSx1aDwuydUj07v6edMu01q7Kck3k6ybw7Zf13dNfX5V1XQLVNUZVbWjqnZcd911868eAABgjVnS7qmL6CmttWuq6vZJ3prkPyV5/fhCrbVzk5ybJKecckpb2hKBZWf7OVOnN26ZTB0AAMvYUrc0XpPkbiPTx/fzpl2mqg5Ncsck18+00dbaNf2/NyQ5P103WAAAAA7SUofGS5KcWFUnVNWtkzw5yQVjy1yQ5PT+8ROSvL+1NtgqWFWHVtUx/ePDkvxiko8veOUAAABr0JJ2T22t3VRVz0ry7iSHJDmvtfaJqnphkh2ttQuS/EWSN1TVlUm+kS5YJkmq6qokd0hy66p6XJKHJ/liknf3gfGQJO9N8udLuFsAAACr1pJf09hauzDJhWPzzhp5vDfJEwfWXT+w2fsvVH0AAADcYrUMhAOrxqbdeyZdAgAA3Gypr2kEAABgBdHSCKwqWy/fOmV688mbJ1QJAMDqoKURAACAQUIjAAAAg4RGAAAABgmNAAAADDIQDrCqGRgHAODgCI3AmjIaIvcLkNvPmTq9ccsSVAQAsLzpngoAAMAgLY2wBDZs2zDpEgAA4IBoaQQAAGCQ0AgAAMAgoREAAIBBrmmECdu0e8+kSwAAgEFaGgEAABgkNAIAADBIaAQAAGCQ0AgAAMAgoREAAIBBRk8FGLL9nKnTG7dMpg4AgAnS0ggAAMAgoREAAIBBQiMAAACDhEYAAAAGCY0AAAAMEhoBAAAY5JYbwJq19fKtU6Y3n7x5QpUAACxfWhoBAAAYpKURFsGGbRsmXQIAACwILY0AAAAM0tIIS2zT7j2TLgEAAOZMSyMAAACD5tXSWFWHtNZ+sFjFACxr28+ZOr1xy2TqAABYQvNtabymql5cVfdalGoxJ5QzAAAgAElEQVQAAABYVuYbGl+T5AlJPl5VH6mqM6rqDotQFwAAAMvAvEJja+3s1tqPJ3lYkk8neWmSr1TVX1fVLyxGgQAAAEzOAQ2E01p7f2vtqUnunOS/JblnkndX1VVVdXZV3XUhiwQAAGAyDnb01FOSPDjJTybZneSDSf5rkiur6jcOctsAAABM2Lzv01hVP5bkaUmemmR9kvcm+S9J3t5a+15VHZLkT5O8JMlfLVilAIts6+Vbp0xvPnnzhCoBAFg+5nvLje1JfjbJNUlel+R1rbUvji7TWvtBVZ2f5NkLViUAAAATMd+Wxq8leXSS97TW2gzLXZbkhAOuCmAlGL1vo3s2AgCr1HxD46uSfGy6wFhVt0vyU621D7TWvp/ki/utDbCC6K4KADD/gXC2J7n3wHP37J8HAABglZhvaKwZnrtdku8cRC0AAAAsM7N2T62qByd5yMis/1pVjxxb7PAkj0myc+FKAwAAYNLmck3jzyT5b/3jluSJSW4aW+Z7Sa5I8tyFKw0AAIBJmzU0ttZeku6ei6mqLyR5fGvtssUuDAAAgMmb1+iprTW30QAAAFhD5nJN46OT/FNr7Vv94xm11i5ckMoAVpLRezYm7tsIAKwac2lp/PskD0jy0f5xy/Aoqi3JIQtTGgAAAJM2l9B4QpKvjDwGAABgjZjLQDhfnO4xAAAAq99crmk8cj4bbK1958DLAQAAYDmZS/fUG9NdqzhXrmmEMZt275l0CQAAcEDmEhr/S+YXGgEAAFgl5nJN418uQR0AAAAsQ3NpaQQgydbLt06Z3nzy5glVAgCwdOYyEM5HkzyttfbJqroks3RVba2dulDFAQAAMFlzaWn8RJJ/G3ns+kYAAIA1Yi7XNP7nkcdPW9RqAAAAWFYO+JrGqqokxyT5emtN6yNr2oZtGyZdAgAALIpbzXeFqnp0Vf1Lkr1Jvppkb1X9S1U9ZsGrAwAAYKLmFRqr6hlJ/i7JjUmeneSJ/b83Jrmgfx4AAIBVYr7dU/9Hkte21sbHmX9NVb0mye8nee2CVAawgux3O46NWyZUCQDAwppv99R1Sd428Nxbk/zIwZUDAADAcjLflsbtSX4uyXumee7nknzgoCsCWCHGWxcBAFajWUNjVd17ZPIVSf5vVa1L8vYkX0tybJLHJ3lUkv+6GEUCAAAwGXNpafx4ktFbalSSZ/Q/rZ/e5x+SHLJg1QEAADBRcwmNGxe9CgAAAJalWUNja+0fl6IQgFVl+zlTp42mCgCsUPMdCOdmVXWrJIePz2+tfeegKgIAAGDZmFdorKpK8ntJnp7khIHFXNPImrdp955JlwAAAAtivvdp/O0kZyb5i3QD4Pxxkhcm+UySq5KcsZDFAQAAMFnzDY1PT/KHSV7cT7+9tfaCJPdJckWSExewNgAAACZsvqHxhCSXtdZ+kOT7SY5KktbaD5NsTXL6wpYHAADAJM03NF6f5Hb94y8lud/Ic0cnOWIhigIAAGB5mO/oqf+c5KeTXJjk/CRnV9WPJPlekmcmed/ClgcAAMAkzTc0np3kuP7x/0rXPfVp6VoY35Pkvy1UYbCcbdi2YdIlAADAkphXaGytfTrJp/vH303y7P4HAACAVWi+LY03q6rjk9wlyZdba9csXEkAq9D2c6ZOb9wymToAAOZpvgPhpKo2VdXVSb6Y5CNJvlRVu6pq84JXBwAAwETNKzRW1VlJXpnkXUkek+SU/t93JXlF/zwAAACrxHy7pz4zyf9qrT1/bP4/VNW1/fMvXJDKAAAAmLj5dk89IskHBp77xySHH1w5AAAALCfzDY1vT/LLA8/9SpK/P7hyAAAAWE5m7Z5aVY8emXxXkhdX1fp0AfJrSY5N8vgk90nyewtfIgAAAJMyl2sa/z5JS1Ij845L8ohplv2rJH+zAHUBAACwDMwlNJ6w6FUArDXu2wgArBCzhsbW2heXohAAAACWn/neciNVdWi6QW8elORHknwjyQeT/L/W2k0LWx4AAACTNK/QWFXHJrkoyb9PclWSa5Oclu7+jJdX1cNba9ctdJGw3G3avWfSJQAAwKKY7y03XppkXZIHtNZ+vLV2Wmvtx5P8TD//pQtdIAAAAJMz39D46CTPa619dHRma+2SJFuSPGahCgMAAGDy5hsab5PkhoHnbkhy64MrBwAAgOVkvqHxw0meV1W3HZ3ZTz+vfx4AAIBVYr6jp/5Oku1Jrq6qi9INhHNskkckqSQPWdDqAAAAmKh5tTS21i5LcmKSc5P8aJKHpQuNr0lyYmvt8gWvEAAAgImZc0tjVR2W5NQkX2itnbl4JQGsQdvPueXxxi2TqwMAYMx8Whp/kOT9SX5ykWoBAABgmZlzaGyt/TDJZ5PcefHKAQAAYDmZ7+ipv5/krKrasBjFAAAAsLzMd/TUP0iyLsllVXVNutFT2+gCrbVTF6g2AAAAJmy+ofHj/Q8AAABrwJxCY1UdkeTR6QLjV5O8t7V27WIWBgAAwOTNGhqr6seTvDfJ+pHZ36qqJ7XWLlqswgAAAJi8uQyE8+IkP0zys0mOTHKfJP+a5LWLWBcAAADLwFxC42lJ/qC19s+ttb2ttU8leUaSu1fVXeb7glX1yKr6dFVdWVVnTvP8barqTf3zH6mq9f38dVW1vapurKpXjq1z/6ra2a/ziqqq+dYFsGxsP2fqDwDABM0lNN4lyefH5n0uSWWe92ysqkOSvCrJo5LcO8mvVdW9xxb7zSS7W2v3SPKyJH/Sz9+b5PlJfneaTb86ydOTnNj/PHI+dQEAADC9uY6e2mZfZE5OTXJla+3zSVJVb0zy2CSfHFnmsUnO7h+/Jckrq6paa99O8k9VdY/RDfatnXdorX24n359kscledcC1Qwwb1sv3zplevPJmydUCQDAwZlraHx3Vd00zfz3jc9vrR07w3aOS3L1yPSuJD8ztExr7aaq+ma6e0N+fYZt7hrb5nHTLVhVZyQ5I0nufve7z1AmzGzT7j2TLgEAAJbEXELjCxa9iiXSWjs3yblJcsoppyxU6ylrwIZtG6ZMb5pQHQAAsNRmDY2ttYUMjdckudvI9PH9vOmW2VVVhya5Y5LrZ9nm8bNsE2CidFcFAFaquQyEs5AuSXJiVZ1QVbdO8uQkF4wtc0GS0/vHT0jy/tbaYKtga+0r6e4b+YB+1NSnJnnHwpcOAACw9sz1msYF0V+j+Kwk705ySJLzWmufqKoXJtnRWrsgyV8keUNVXZnkG+mCZZKkqq5Kcockt66qxyV5eGvtk0k2J/nLJEekGwDHIDgAAAALYElDY5K01i5McuHYvLNGHu9N8sSBddcPzN+R5KSFqxIAAIBkAqERgP2NXvPoekcAYDlZ6msaAQAAWEGERgAAAAbpngqw3G0/Z+r0xi2TqQMAWJOERoAJGL9vIwDAcqV7KgAAAIOERgAAAAYJjQAAAAwSGgEAABgkNAIAADDI6KkAK41bcAAAS0hLIwAAAIO0NMKADds23Px40+49E6wEAAAmR0sjAAAAg7Q0AiwzWy/fOmV688mbJ1QJAICWRgAAAGYgNAIAADBIaAQAAGCQ0AgAAMAgA+EArHTbz5k6vXHLZOoAAFYlLY0AAAAMEhoBAAAYJDQCAAAwyDWNACvM1su3TpnefPLmCVUCAKwFWhoBAAAYJDQCAAAwSGgEAABgkNAIAADAIAPhACxz4wPfAAAsJaERYIXbbzTVjVsmVAkAsBrpngoAAMAgLY0Aq832c255rNURADhIQiMM2LR7z6RLAACAidM9FQAAgEFCIwAAAIOERgAAAAYJjQAAAAwyEA7AajY6kmpiNFUAYN60NAIAADBIaAQAAGCQ0AgAAMAgoREAAIBBQiMAAACDhEYAAAAGueUGwCq29fKtU6Y3u+UGADBPQiP0NmzbMGV604TqgEXlvo0AwDzpngoAAMAgoREAAIBBQiMAAACDXNMIsMqMD34DAHAwhEbobdq9Z9IlwNIzMA4AMAvdUwEAABgkNAIAADBIaAQAAGCQ0AgAAMAgA+EArCHjI6tuPnnzhCoBAFYKoRGAWxhNFQAYo3sqAAAAg4RGAAAABgmNAAAADBIaAQAAGCQ0AgAAMEhoBAAAYJBbbrBmbdi2Ycr0pgnVAQAAy5mWRgAAAAZpaQRg2PZzpk5v3DKZOgCAiREaWbM27d4z6RIAAGDZExoBGLT18q1TpjdraQSANUdoBGDuRrurCpAAsCYIjQBr2H4tiSdvnlAlAMByJTQCcLPxEDnT87qqAsDa4JYbAAAADNLSCMAB2bBtw5TpnafvnFAlAMBi0tIIAADAIC2NABwQ9zoFgLVBSyMAAACDhEYAAAAGCY0AAAAMEhoBAAAYJDQCAAAwSGgEAABgkNAIAADAIPdpZO3Yfs6kK4DVbfx3bOOWydQBACwoLY0AAAAMEhoBAAAYJDQCAAAwyDWNACwO1zgCwKqgpREAAIBBWhpZM7ZevnXSJQAAwIqjpREAAIBBQiMAAACDdE8FYEGMdwHffPLmqQsYGAcAViShEYAlsV+oFBoBYEUQGgFYFLMNPrVh24abH+88fedilwMAHCDXNAIAADBISyMAEzfa6phoeQSA5URoBGAiNu3ec/PjVx991AQrAQBmonsqAAAAg4RGAAAABumeyuo1fk84YNka7aoKACwvQiMAy46BcQBg+dA9FQAAgEFCIwAAAIN0TwVg2XGNIwAsH1oaAQAAGCQ0AgAAMEhoBAAAYNCSh8aqemRVfbqqrqyqM6d5/jZV9ab++Y9U1fqR57b08z9dVY8YmX9VVe2sqsuqasfS7AkAAMDqt6QD4VTVIUleleRhSXYluaSqLmitfXJksd9Msru1do+qenKSP0nyq1V17yRPTnKfJHdN8t6q+onW2g/69Ta21r6+ZDsDwJJx30YAmJylHj311CRXttY+nyRV9cYkj00yGhofm+Ts/vFbkryyqqqf/8bW2neTfKGqruy396Elqh2ACTGaKgBMzlJ3Tz0uydUj07v6edMu01q7Kck3k6ybZd2W5KKqurSqzhh68ao6o6p2VNWO66677qB2BAAAYC1YLfdpfFBr7ZqqOjbJe6rqitbaB8YXaq2dm+TcJDnllFPaUhfJEth+zs0Pt16+dYKFAItp68vvNmV683OuHlgSADhYS93SeE2S0f/pj+/nTbtMVR2a5I5Jrp9p3dbavn+/luRt6bqtAgAAcJCWOjRekuTEqjqhqm6dbmCbC8aWuSDJ6f3jJyR5f2ut9fOf3I+uekKSE5N8tKpuW1W3T5Kqum2Shyf5+BLsCwAAwKq3pN1TW2s3VdWzkrw7ySFJzmutfaKqXphkR2vtgiR/keQN/UA330gXLNMv9+Z0g+bclOSZrbUfVNWdkrytGysnhyY5v7X2D0u5XwAAAKvVkl/T2Fq7MMmFY/POGnm8N8kTB9b94yR/PDbv80lOXvhKAViJ3J4DABbWahkIB4C1bGQQLABgYS31NY0AAACsIFoaAVjxptxi5+ijpjynuyoAHByhEYBVZdPuPVOmXz0WIgGA+dE9FQAAgEFCIwAAAIN0TwVgVRvvrgoAzI/QyKoyZTAMAADgoOmeCgAAwCAtjaxsbugNzNPWl99tyvTm51w9oUoAYGXQ0ggAAMAgoREAAIBBQiMAAACDXNMIwNo2fm30xi2TqQMAliktjQAAAAzS0gjAmjZ+f9fNWhoBYAqhEQBGjXZXFSABQPdUAAAAhmlpBIARo91VdVUFAKERAIYZWRUAhEYAGGKQHAAQGllpxv/qDwAALCqhEQDmSndVANYgo6cCAAAwSEsjAMzRftc4ji+g5RGAVUhLIwAAAIO0NALAATK6KgBrgdAIAAvFQDkArEK6pwIAADBISyMr2njXMAAAYGFpaQQAAGCQlkaWt/HrgwBWktHPMNc3ArBCaWkEAABgkNAIAADAIN1TWVEMfAMsZ/vdt/HkzROqBAAWjtAIAEvBPRwBWKF0TwUAAGCQlkYAmAQtjwCsEFoaAQAAGCQ0AgAAMEj3VABYJKOjqc46kqruqgAsU1oaAQAAGKSlEQCWIy2PACwTQiPLy/iXJIBVYrSrajKH7qoAsEzongoAAMAgLY0AMAHzbnkc7YmhqyoAS0hoBIBlSHdWAJYLoREAVpgN2zZMmd55+s4JVQLAWiA0AsAyMN6yONPzm8aemy1ECpkAHAyhEQBWuE2790y6BABWMaOnAgAAMEhLI5Pn3owAC8vnKgALSEsjAAAAg7Q0AsAqs9+gOkcfNZlCAFgVhEYAWOUMlAPAwRAaWdZmG4IeAFg53P4FViahEQDWmtGBcjZumVwdAKwIBsIBAABgkJZGAFjLxm7PseFL50+Z1n0QAKERALiZQXMAGCc0svTcdBpgokYHGdt88uaZFx7/zHYNJMCaIzQCALBoxkdMBVYeoREA1rDZbm00/vx+7ZLLpOXRrRyWDyERVh+hEQA4cG7fAbDqCY0sPtcwAqxaU66PHAuNWv8AVgehEQCYs5m6s46HxIMZiVXgXDl0R4XVT2gEABbEYt6uQ4gEmByhEQBYcbRuASwdoREAWBru+QiwIgmNLDwD3wAwFzP8fzFbV9dXH33UjM/rzgqwcIRGAGBJzHZPSFYGXYNh7REaWVZ8oQBgLubbEjkadLQ6AsyP0AgArDrjoXK27qwADBMaAYBVbzFvBwKw2gmNAMDaYhRXgHkRGgGANWX8+vlXf+n8KdOueQSYSmjk4LnFBgAr2H5dVxexJXIxbwVisB9gsQiNAAAzmVB31tlubbGYwdBtNYBRQiMAwIjx7qubT948dYGZetjMM1Aul9bBgwmJs41UaxAiWPmERgCAhTLLJRszBaiDbd0b3fZ8t7WQwU5IhNVHaGTixv+iCwDLyUz/T+3XCnkQ5h22ZgioghuwkIRGAIADNFtX1sX8w6g/ugJLRWgEAFggghywGt1q0gUAAACwfAmNAAAADBIaAQAAGOSaRqY3oRsZAwBriO8bsCIIjRyYWe5DBQAwb0IkLEu6pwIAADBISyNzs4Ati4YjBwCAlUNopKO7KQCw3OiuCsuC7qkAAAAM0tIIAMDKMNryqNURloyWRgAAAAZpaVxNZrsucUJ/kTPwDQCw4FzvCEtGSyMAAACDhEYAAAAG6Z4KAMDKp7sqLBotjQAAAAzS0rhWzTZozkEw8A0AMBfj3xk2n7x5QpUAMxEaAQBYfXRXhQUjNAIAsPoJkXDAhEYOmu6oAACwegmNAAAsC/P5Q/RBX/842vKo1RFmJDSuJYs4+A0A/P/t3Xu4XFV5x/Hvz0QiRZBwixAuiRhUhKcIIUi5FFvu8JAiWMI1XFpEgxW0KpdqkadqQO7VglhSwEIAQfCAQIBguNQGkiAPIUEgQICkIRDAhEsSEvL2j7UO7DOZOTmXObPPmfP7PM95Zvaavde8s2ZlZ797r1nbrJG6O4lOcfs1bumk0vo5J419mZNAMzMzM2DNVyk7k1SuVldnroCe9nKb5e2v2b7N8syxMztcl/Vdzfa9O2k0MzMzs6bXXlJZ1/kZKk7qf+3NP7d9r0u2aLO8pmS2vdjWtO32L13f7utFfT2p6W0qv/e+ThHR2DeU9gcuBQYA/xkR4yteHwRcC+wEvA4cERFz82tnAicB7wP/FBGTOlJnNSNHjozp06fX62OVo5dcafREOGZmZmbWGxQT6cpj1MsHr99meeaWR7VZbu+YtnLbziaFlVegewtJMyJi5JrWa+iVRkkDgJ8D+wDzgGmSWiJidmG1k4A3I+LTksYA5wFHSNoWGAN8HtgMuE/SNnmbNdVpdeZE0czMzMx6m/aOUVe76vtmx49nm+3KYWc1enjqKGBORDwPIOkGYDRQTPBGA+fk5zcDP5OkXH5DRCwHXpA0J9dHB+rss4rjoTtzNqSa9s68dGdohJmZmZmZNa9GJ41DgeK12XnALrXWiYiVkhYDG+byqRXbDs3P11QnAJJOBk7Oi29LeroLn6GnbQQsqvaCOKtbFY9rZ/v2XutHara99Ti3fXnc9uVx25fHbV8et3253P4lGXe6emvbb9WRlfrVRDgRcSVwZdlxtEfS9I6MK7b6c9uXx21fHrd9edz25XHbl8dtXy63f3n6ett/pMHvNx8oThm1eS6ruo6kgcAnSBPi1Nq2I3WamZmZmZlZFzQ6aZwGjJA0XNJapIltWirWaQHG5ueHA/dHmuK1BRgjaZCk4cAI4NEO1mlmZmZmZmZd0NDhqfk3iqcCk0i3x5gQEbMknQtMj4gW4CrgV3mimzdISSB5vZtIE9ysBMZFxPsA1eps5Oeqs149fLbJue3L47Yvj9u+PG778rjty+O2L5fbvzx9uu0bfp9GMzMzMzMz6zsaPTzVzMzMzMzM+hAnjWZmZmZmZlaTk8ZeQtL+kp6WNEfSGWXH08wkbSHp95JmS5ol6Zu5/BxJ8yU9nv8OLDvWZiRprqSZuY2n57INJN0r6dn8OLjsOJuRpM8U+vfjkpZIOs19v2dImiDpVUlPFsqq9nUll+X/A56QtGN5kfd9Ndr+p5L+lNv3Vknr5/JhkpYW+v8V5UXe99Vo+5r7GEln5n7/tKT9yom6OdRo+xsL7T5X0uO53P2+jto5tmyafb5/09gLSBoAPAPsA8wjzQh7ZETMLjWwJiVpU2DTiHhM0rrADODvgL8H3o6IC0oNsMlJmguMjIhFhbLzgTciYnw+aTI4Ir5XVoz9Qd7vzAd2AU7Afb/uJO0JvA1cGxHb5bKqfT0fRH8DOJD0nVwaEbuUFXtfV6Pt9yXNyL5S0nkAue2HAXe0rmfdU6Ptz6HKPkbStsBEYBSwGXAfsE3rRIfWOdXavuL1C4HFEXGu+319tXNseTxNss/3lcbeYRQwJyKej4j3gBuA0SXH1LQiYkFEPJafvwU8BQwtN6p+bzRwTX5+DWlHaz3rb4HnIuLFsgNpVhHxIGkW8KJafX006UAvImIqsH4+CLEuqNb2EXFPRKzMi1NJ93W2OqvR72sZDdwQEcsj4gVgDumYyLqgvbaXJNLJ8YkNDaqfaOfYsmn2+U4ae4ehwMuF5Xk4iWmIfKbtC8AjuejUPExggodI9pgA7pE0Q9LJuWxIRCzIz18BhpQTWr8yhrYHD+77jVGrr/v/gcY6EbirsDxc0h8lPSBpj7KCanLV9jHu942zB7AwIp4tlLnf94CKY8um2ec7abR+S9LHgVuA0yJiCXA5sDWwA7AAuLDE8JrZ7hGxI3AAMC4Pp/lApDHzHjffgyStBRwC/DoXue+XwH29HJLOJt3v+bpctADYMiK+AHwLuF7SemXF16S8jynfkbQ9Ueh+3wOqHFt+oK/v85009g7zgS0Ky5vnMushkj5K+kd9XUT8BiAiFkbE+xGxCvglHiLTIyJifn58FbiV1M4LW4dl5MdXy4uwXzgAeCwiFoL7foPV6uv+f6ABJB0PHAwcnQ/gyEMjX8/PZwDPAduUFmQTamcf437fAJIGAl8Gbmwtc7+vv2rHljTRPt9JY+8wDRghaXi+AjAGaCk5pqaVx/VfBTwVERcVyotjyQ8Fnqzc1rpH0jr5B+JIWgfYl9TOLcDYvNpY4LflRNhvtDnj7L7fULX6egtwXJ5R74ukySoWVKvAukbS/sB3gUMi4t1C+cZ5YigkfQoYATxfTpTNqZ19TAswRtIgScNJbf9oo+PrB/YG/hQR81oL3O/rq9axJU20zx9YdgAGeSa3U4FJwABgQkTMKjmsZrYbcCwws3XqaeAs4EhJO5CGDswFvlpOeE1tCHBr2rcyELg+Iu6WNA24SdJJwIukH+tbD8jJ+j607d/nu+/Xn6SJwF7ARpLmAf8KjKd6X7+TNIveHOBd0oy21kU12v5MYBBwb94HTY2IU4A9gXMlrQBWAadEREcncrEKNdp+r2r7mIiYJekmYDZpyPA4z5zaddXaPiKuYvXfsIP7fb3VOrZsmn2+b7lhZmZmZmZmNXl4qpmZmZmZmdXkpNHMzMzMzMxqctJoZmZmZmZmNTlpNDMzMzMzs5qcNJqZmZmZmVlNThrNzKzTJJ0jKSRNqvLazZKmNDCWvXIs2zXqPTtD0uckPSTpnRznsC7WM0XSzfWNrv4kbZL7x7CyYzEzs/pw0mhmZt2xr6Sdyw6il/spsD5wCLAr0Ktv4FwHm5DuzTes5DjMzKxOnDSamVlXvQHMBM4uO5CeJOlj3azis8C9ETE5IqZGxPJ6xFVPktYuO4ZaenNsZmb9hZNGMzPrqgB+BBwiaftaK+WhiouqlIekUwvLcyVdIOkMSQskLZZ0oZIDJc2S9Jak2yQNrvJWm0m6Iw8DfUnSKVXecw9JD0h6V9Lrkn4pad3C68fnuEbl4aBLge+089l2kDQ51/empOskDcmvDZMUwNbA6bneKe3UNUDSmZKekbRc0jxJV1dZ7yhJcyQtkXSXpM0rXh8vaaakt3Md10n6ZMU6c3Pbfl/SPGBJLt9VUktu/3ckPS7p6CoxbCVpoqRF+bM/keMaRjqRAPD7/JmjsN0Gkq6UtFDSMkl/kLRLRd0h6VuSLpH0Wmt9knbPw3yX5L/HJX2lVnuamVn9DCw7ADMz69N+DZxLuto4pg71jQEeBU4AdgL+jXSCc0/g+8DawM+AnwCVSeFVwK+AfwcOBS6XNC8i7gCQtBtwH3AbcDiwITAeGJyXiyYC/wH8EPhztUAlbQxMAZ4CjgI+nuu7V9JI0jDUXYFbgftzXEva+ey/AI4DzgceADYADqtYZxdgM+DbuS0uBa4EDiysswnwY+D/gI3zuvdL2i4iVhXWOwqYBXydD48HtgL+B7gCWAbsBvyXpFURMTF/7k2A/wXeBf4ZeBnYDtgif+ajgeuAccBjhfYaRGr/9UmJ+KvA14D7JI2IiFcKsX0HeBA4FviIpPWAO4DfkvqbgO1zXWZm1sOcNJqZWZdFxCpJPwGukvSDiHimm1UuA74SEe8Dd0saDXwDGBERL/0YhK8AAAT9SURBVABI+ktgLKsnjXdFxFn5+SRJWwP/Qko2ICV0f4iII1o3kDQfmJwTqicLdV0WEZeuIdZv58f9IqL1St2zwFTgsJxkTZW0HFgQEVNrVSTps8BJwDcj4rLCSzdWrLoecFBEvJm3+yRwsaS1I2IpQEScWKh3ACnBmwfsTkrEig6OiGWtCxFxQ2Fb5fU3B/6RlEgDnA58AtgpIlp/nzm5sN0T+ensis98DCm5/HxEPJvXvQ94mtSWxSu6Cyq+p5H5PU+NiLdy8T2YmVlDeHiqmZl1138DLwFn1qGuKTlhbDUHmNuaMBbKNpa0VsW2t1Ys/wbYKQ/7/AvSVb+bJA1s/QMeBlaQrmoW/a4DsY4C7mlNGAEi4hFgLilB64wv5cer17DetNaEMZudH4e2Fkg6IA/7XAysJCWMANtU1DW5mDDmbQdLukzSi6R2WQGcXLHt3wB3FxLGjtobmAG8UGh/SFdVR1ase2fF8nPA28D1kkZL8hVGM7MGctJoZmbdEhErSUMqj5G0VTerqxwK+l6NMgGVSeOrVZYHAhuRhqAOIA05XVH4Ww58lDS0smhhB2LdtMZ6C0lDSztjQ+CdYgJaQ7W2APgYgNJMti2kRPFYUqL8xeI6FXFWuho4gjTj677AzsCEim03pGszwG6UY1lR8XcCa2j/nCjvQ/qubgJek/Q7SZ/qQhxmZtZJHp5qZmb1MIE0FPR7VV5bRkWCV2Mim+7apMrySmARKekJ4BxWv4oF6fd/RVFlnUoLqrwnwBDSFbXOeB1YR9J6HUgc23Mo8BpwREQEpElraqzb5jMqzRJ7MDAuIq4olFeeYH6dlDB31hvAdNLvGCtVzii7Wvvnoa77K82mujdwEXA9HybFZmbWQ3yl0czMui3fRuIC4ERWTyjmAetKGloo27cHwji0yvKMiHg/It4h/dbwMxExvcpfZdLYEY8A+1XMvroz6f6ED3eyrvvz43FdiKNobWBFa8KYrTb7aQ2DSMcFHyRw+bMdUrHeZNLnHlKjnjZXPyu2+zTwUpX2n0kHRcTSiLiddKJi245uZ2ZmXecrjWZmVi+/AM4C/or0O7VWdwNLgQmSLgSGs/okNvVwgKQf5ff+Mmk44+jC698lTXqzCrgZeAvYEjgIOLsLk/hcRLpqNknSeXw4e+pM4JbOVBQRT0u6Ergwz076IGlm0MMjojOz0t4LnCbpEuB20ndxTAdjWCxpGvADSUuAVcAZwGLSBDytLiYltw/l9n4Z+BywTkScT/p961JgbP5d5YqImA5cS/rep0i6AHieNNR1FPBKRFxcKzZJB5FOSNyW6x8KfJUPk20zM+tBvtJoZmZ1ERHvkhKKyvJFpFtHbE466D+GdLuHevsHYMf8Hq3DLFsKcTxMunXHxqRbc9xOSiRfpmO/YWwjIl4jTWCzjDSz6M+Bh4B9IuK99rat4eukW3wcQxpCewnpthadielO0hDhw0i/bfxrUlt01FGkZO5a0u08bsnPi+/xGulWHH/MMd5Bmiznpfz6MtJsqzuREvhphfIvkRLbH5JmP70UGEG6zUp75pCGrP44b3c+6WTEie1tZGZm9aG2I1jMzMzMzMzMPuQrjWZmZmZmZlaTk0YzMzMzMzOryUmjmZmZmZmZ1eSk0czMzMzMzGpy0mhmZmZmZmY1OWk0MzMzMzOzmpw0mpmZmZmZWU1OGs3MzMzMzKym/wfkVojYZ31/AAAAAABJRU5ErkJggg==\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7fb436da47f0>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "train_qs = pd.Series(df_train['question1'].tolist() + df_train['question2'].tolist()).astype(str)\n",
    "test_qs = pd.Series(df_test['question1'].tolist() + df_test['question2'].tolist()).astype(str)\n",
    "\n",
    "dist_train = train_qs.apply(len)\n",
    "dist_test = test_qs.apply(len)\n",
    "plt.figure(figsize=(15, 10))\n",
    "plt.hist(dist_train, bins=200, range=[0, 200], color=pal[2], normed=True, label='train')\n",
    "plt.hist(dist_test, bins=200, range=[0, 200], color=pal[1], normed=True, alpha=0.5, label='test')\n",
    "plt.title('Normalised histogram of character count in questions', fontsize=15)\n",
    "plt.legend()\n",
    "plt.xlabel('Number of characters', fontsize=15)\n",
    "plt.ylabel('Probability', fontsize=15)\n",
    "\n",
    "print('mean-train {:.2f} std-train {:.2f} mean-test {:.2f} std-test {:.2f} max-train {:.2f} max-test {:.2f}'.format(dist_train.mean(), \n",
    "                          dist_train.std(), dist_test.mean(), dist_test.std(), dist_train.max(), dist_test.max()))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "mean-train 11.06 std-train 5.89 mean-test 11.02 std-test 5.84 max-train 237.00 max-test 238.00\n"
     ]
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAA4cAAAJjCAYAAAC/T2S2AAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADl0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uIDIuMS4xLCBodHRwOi8vbWF0cGxvdGxpYi5vcmcvAOZPmwAAIABJREFUeJzs3Xu4HWV9N/zvT0ACooghqIAabNGKpi9qpPqoKFoVsAW1aNXaQmsLJfrUvrVWeFREesDT4+mVVGxFsZYqxWppiQXUWKxHAgUBOQVFCZ4CEgpqVOR+/5iJLhY72YfsvVay8/lc176y1sw9s35rZu1c67vve+6p1loAAADYtt1j3AUAAAAwfsIhAAAAwiEAAADCIQAAABEOAQAAiHAIAABAhENgzKrqxKpqVXXuBOvOqqrPjKGsGenfx8sHnn+mqs4a0WvfVFUnTtLmLvVtpM1Rfbtdpvi6D+vP4X2nUe68V1WHV9WVVfWTqrp+3PVsTFXt0p/vo8Zdy3RU1TOr6k+n2Pb6qnrrXNc0alV1z/53b/+h5Yv7c/ob46oN2HoJh8CW4plV9bhxFzHLliU5ftxFTNM5SZ6Q5IdTbP+wJK9PIhz2qmq7JB9McmmSpyV57ngrmpeemWRK4TDd8X/XHNYyLvdM97u3/9Dyb6f7Hf6vkVcEbPW2H3cBAEm+n+TGJK9J8pzZ3nlV7dRa+9Fs73cyrbWvjvo1N1drbW2SteOuY1OqqpLs2FpbP+5aNuKBSe6T5IzW2ti/oFfVDknubK39bNy1jENr7b/HXcMotdZ+nOSL464D2DrpOQS2BC3JXyc5rKqWbKphVe1fVZ+qqh9W1S1V9Y9Vdf+B9RuGVP1OVX2wqtYl+bd+3fVV9daqOq6qvl1Vt1bV/63OoVV1RVXdVlUfr6rdBvZ5r6p6d1Vd3b/u16vqlKq6zyS13mVYaVXtXVVnVtX3qupHVXVdVf3l0DZPrqr/7F/n5qr6u6q691CbA6vq0qpaX1UXVdX/msIx3mC7qvqbqlrb13FKVe04sO+7DSutquOranX/et+tqv+oqgdU1VM3HNskX++3u35gu02eq77Ng6vqE/3x+Hr/+ncZTtwPnbupqp5UVRcmWZ/k+VM9L31d/29/rm/u9/Xn/bojq+prVbWuqk6rqgWTHcCqekFVXVZVP66qG6rqr6tq+w3HL8kNfdN/7V/7xI3s5xtV9X8Gnh/Tt/+TgWWvrKobB57vXFXvqqrv9Ofjwqp65tB+P9Mfw6Or6rr+eO3Zr/utqrqmP94XJPmVyd5vv91OVfXmvuYf98f65IH12/Xn6Zv9+iuq6sUT1TW07Kn9e35U/3zD7+8LqurU6n5H11TVG6rqHn2bE5O8MslD+ratqj6widrvMqy0qj5QVauq6hlV9ZWq+kFV/VdVPXIKx+FRVfW5/thfWVWH9fv6wECbSd9nv2xBf0xv6I/ZpVV16NB2h1X3O/6D/nfoS1X1lH71bf2/7x84DotrgmGlUzw/UzouVfXSqvpq/xm6qbr/ryY9dsDWQc8hsKX45yQnpes9fOFEDapqUZLPJLkyyYuT7JLkjUnOr6qlrbWfDDR/a5J/SfL8JIM9Ji9M8uUkv5/ksUn+Kt0fyg5M8rokOyV5d5KTk/xxv83OSbbra1ub5EH9439O8qxpvMcP9vs/Osm6JA/NwJfzqnpikk8m+XiSI5Is7N/fbv3zVNWeST7Rv4cj0n3p/8e+xql4ZZJPJ3lJkl/t3+c3krx5osZV9XtJ/k+SVye5oq/paUnuleTiJH+e7lg/L91wth/32016rqqqkpydbkjqH6QLMa9LsijJdUOl7Jzk9L7Oa5J8K9M7L69MN2T2RUl+I8lbqmqPJI9L8idJHpzk7f2+37ixg9cHsY+kO5ev6o/hX/bH5Y/713heus/enyf5XJI1G9ndZ5M8eeD5gf0xeHJ+MQzywL7dBn+X5LB052R1kj9Kck5VHTTUS/nEJL+U7rz9MMmtVfWYvvaPJXlFkkclOXNj73XgPVeSf003VPEvk1yUZK+h2k9K8hdJ3pDkwiS/leQfq6q11v5psteYwJuTfDTdZ/zpSU5I9/k7M8nfJ9k3dx2yO93e7gcneUu6P0r9KN1n+CNVtaS11ibaoKp2SnJukpvSfaZ3SvKOdJ/ty6f5+klyVpID0g0NvS7JC5Kc3f9+XFJVv9S3eWe6z9qCdP9n3a/f/mnpfpf/Kt3nLul+Bx84wWtN9fxs8rhU1YFJ3pPufHwhXQ/5E5LsOoP3D2yJWmt+/PjxM7afJCcmual/fFS6IPew/vlZST4z0PaN6ULVfQaW/Vq6nscX9c8X988/NsFrXZ/uC/V2A8u+nOSOJPsMLHtzku9uoubt0335bkkePLC8JXn5wPPPJDlr4PntSX5zE/v9bJKVQ8ue1u/3UQO13Zxk54E2v9O3OXGSY92SXDC07ONJvjjw/Ki+3S7983cn+egm9vkbffvFQ8uncq6e3T9/3ECbvZL8dOi8n9i3O3yS97ep87Jy4Pk90n2JvmWovjOTfGmS1/jiBOfoL/rP7d5Dn8HfmGRfxyS5Nck9+uff7I/3d/rn1Z/rl/XPH5HkziRHDr2Xy5OcO/S5+1GS+w+93plJvpqkBpa9pq/1qE3U+ay+zWEbWX+/JD9I8vqh5SuSXL2x34d+2VOHPt8bjt0Hh9pdkuTDA8/fmuT6TR3fgbbXJ3nrwPMPpPud33dg2XP61/2VTexnWf/Z3Htg2YbP2wem+T6f3j9/ylC7C5L8c//4iCQ3b6KeXSY6d8Ofv2mcn0mPS7o/eFw0lePux4+frfPHsFJgS/KhdF+QNzaJywFJzmut/c+GBa21L6X78vekobbnZGKfaXe99mp1ui+ZXx9atqiq7rlhQVX9blX9d1Xdnu4L4oZemodt+i3dxSVJTq5u6OSDB1dU1c7p/gJ/ZlVtv+Gnf52fpusxSLpjcH5rbXDCmI9No4bzhp5/Ncnek9R8aD+s74DqJluZiqmcq8elC0IXDrS5MV3P1LCWrsf0LqZxXj418Bp3Jvl6ui+5/zPQZnW6cDqh/r0/Jl3P5KCPpAtpT9jYthtxQbqel/+nqhanOw9vTrJ7Ve2b5JHpvthv6Dl8XLrA+PPX79/LP+fun/+LWmvfHVp2QJKzW2uDPWP/MoU6n5bk+621szey/lHpenEnOi4P63uRp2u6n9Ppur61du3Q/jPJaxyQ7rj+vCe4tfa5JN+bwev/epLvJPnc0O/7p5Is7dtclmTXqjq9utlZ7zWD10mmd34mOy6XJHl0Vb29uuHt9wwwrwiHwBajtXZHui/HL6mqh0zQ5IFJhr/wpl92vwmWTWTd0POfbGRZpZsNMFX13HTDCL+Qbpjq4/OL4WyTXqM24LeTrEo3fPEbVXVJVT29X7dbuiGSy9OFnA0/P06yQ7ohk0nygAx9Ge2D4u1TrGGi97qp93BauiGML0jypSTfraq/mkJInMq5ekAmHg440bJb2l2HDU/3vEz1vG/qWOye7lwMv68Nz4c/g5O5Kt0QxSf3P5e31r6Z7gv4hmXr8oshiw9McvvQHwY2vP7ONXDt6AQ1JhN8diZ4PpGF6XpaN2bDMMbZOi7J9M/NbOw/k7zGRMcvG1k2md37/f106OfE9L/rrbWrkxyebvj5iiQ3VdUZMwjb0zk/mzwurbVPphuSf2C6HtKbqrvOd6bBFdjCuOYQ2NKcluS16a6VGvbtJHtMsPz+uXtv04TXDc3Q89MNN1y2YcHApBBT1veKHVXdxBoHpPsieHbfi7gu/dDQdF8Eh32r//c7GToGfa/jlO5LOIOa70wXZt9eVQ9KN4T1r9NdR/eeTWw6lXP1nXTXFw5blO7au7uUMkG7WTkv03BTui/ww+9rwyQ735/Ozlprrar+K78IgRf0qzZci7ggyef6c5B0x3SXqtp5KCDeP8kPWzdL5c93P8FL3u2zM8Hzidycia9j22BDcNyjbztYV/KL47I+/R9cBuyWrcd3MvEEPsPHcCrvc8MMzZucnbm1dk66a0p3TTcM+x1J/r9s5LrsjZjq+ZmS1trpSU7vQ+rz0v3/cFuS46azH2DLpOcQ2KL0X3Dfmm6CkuEvpF9K8qwamL2zunsjLs7c3tNrp/QTrQz4nZnurLV2Z2vti+kmh9g5yUNaaz9Idz3bw1trqyb42RAOL0zyjD4QbjCS++i11m5orb0x3fDL/frFG+txmcq5ujDJA6rqgIE2e+UXQ2gnM6vnZTL9cOSL0oXSQS9Idy3gF2aw2wvSBcED84twuGHZk3PXyWguTBf6jtiwoJ8s5ohM7fN/YboZgWtg2fOmsN2nktyvNn5T9cvTTXoz0XG5pnW3R0m6PygMh6tnZmZmuydxKi5M8tiq+vnQ034SqeFwOJX3+al0PYe3T/T7PvzCrbVbW2tnpBtCPtnv3rCpnp9paa2tba2dmu4zut9k7YGtg55DYEt0arqhjP8ryX8OLH9bkmOTnFtVb8ovZsC8LN3MhnPl/CSnVNVr0oWeQ9NNKDFl/V/+z003DPKaJDumm0HzO+lm9Ey6iU0+VVV3ppuM57Z0swc+O8lrWmvXpOs5eFmSf6+qt6WbrfT4dBOQzLqqOjVdz8IX002eclC6mSI39Oxe3f97TFV9OF0P1mWZ2rlake5G8WdW1Yb38Pp0w9029JZtymaflxl4fbr39P4kH06yJN0Mnn83eC3aNHw23bG6f34RDv8r3UyjG9YnSVprV1bVPyV5dx+6r0s3W+mvpDvWk3lTuuN0ZlW9L921aC+dwnbnp/vsnlFVJ6WbpfaBSQ5srR3TWvt+Vb0jyWur6o50Q6efl+58vGhgPx9L8tKqenu6a4IPSnLwFF5/IlcluX91tw65PN2kVtfPcF9T9f50oxrOqe52GjulO/c3DbWbyvvccEzP738/rkh3/en+SRa01o6vqmPSXcf6H+lGDuybLuB9MElaN+Pv15O8oKouT9dj+ZXhoqdxfiZVVW9INwz1M/37fnSSp0SvIcwbeg6BLU4/ZO7tEyxfm+6L1vok/5TklHRfnp8xfD3aLDs1yf9NN/3/vyR5SLqp7Kdjfbpg9Ip0t284Pd1f85/ZWvtRkrTuVgQHphtW+Q/p7iH4F+num/fdvs2N6b7U7Z4uZC1Ld1uK4evQZssX+preny7MPTfJH7XWPt7X8410Mxg+L91tG/6tXz7pueonRjk83Rf996ebsv9v002CMThRzMbMxnmZltbaeemG9C1N917/tK/h5TPc5X+nu1702tbad/rXWJvumKxP90V+0B+l++yckO72Eg9JNyvlpD2HfY/UC9N9of94uiGNvz2F7Vq68/7edO/3E+lunzAYik5Id1uUY5P8e7rPzEtaax8e2M856f7oc0S6APWQdOduJs5MN7vmm9P16J04w/1MWf//0rPSzfz54XR/KHhlulvBDLab9H32x/R56YbR/2m6oHhqujC44Vx+Jd3/BW9LN0HPa9PdymRwyP0fp/u/4JPpjsOeGyl/0vMzRRem6yV8T1/zsemO/TunuR9gC1V3nbQMAMan72H9WpJ3t9ZeP+56YDJVtSrdZEJHjbsWgM1lWCkAY1NVf5xuCOm16XpJ/izdkNvTxlkXAGyLhEMAxml9umFyD0k32cqXk/x6P1wVABghw0oBAAAwIQ0AAADzfFjp7rvv3hYvXjzuMgAAAMbioosuuqm1tmgqbed1OFy8eHFWrbrbvWQBAAC2CVU15ev4DSsFAABAOAQAAEA4BAAAIPP8mkMAAGDb9tOf/jRr1qzJ+vXrx13KnFqwYEH23nvv7LDDDjPeh3AIAADMW2vWrMm9733vLF68OFU17nLmRGstN998c9asWZN99tlnxvsxrBQAAJi31q9fn4ULF87bYJgkVZWFCxdudu+ocAgAAMxr8zkYbjAb71E4BAAAwDWHAADAtmPJ6UtmdX+XHXnZJtevW7cuZ5xxRpYtWzat/R566KE544wzct/73ndzypsWPYcAAABzZN26dVm+fPndlt9xxx2b3G7FihUjDYaJnkMAAIA5c9xxx+W6667L/vvvnx122CELFizIbrvtlquuuirXXHNNnvOc5+SGG27I+vXr84pXvCJHH310kmTx4sVZtWpVbr/99hxyyCF50pOelM9//vPZa6+98q//+q/ZaaedZr1WPYcAAABz5I1vfGN+6Zd+KZdcckne8pa35OKLL8473/nOXHPNNUmS0047LRdddFFWrVqVd73rXbn55pvvto9rr702L3vZy3LFFVfkvve9bz760Y/OSa16DgEAAEbkgAMOuMu9CN/1rnflYx/7WJLkhhtuyLXXXpuFCxfeZZt99tkn+++/f5LksY99bK6//vo5qU04BAAAGJF73eteP3/8mc98Jp/85CfzhS98ITvvvHOe+tSnTnivwh133PHnj7fbbrv86Ec/mpPaDCsFAACYI/e+971z2223Tbju1ltvzW677Zadd945V111Vb74xS+OuLq70nMIAABsMya79cRsW7hwYZ74xCfmUY96VHbaaafc//73//m6gw8+OO95z3vyiEc8Ig9/+MPz+Mc/fqS1DavW2lgLmEtLly5tq1atGncZAADAmFx55ZV5xCMeMe4yRmKi91pVF7XWlk5l+5EPK62qg6vq6qpaXVXHTbD+wKq6uKruqKojBpbvX1VfqKorquorVfXbo60cAABg/hppOKyq7ZKckuSQJPsleVFV7TfU7JtJjkpyxtDyHyb5vdbaI5McnOQdVTXau0ICAADMU6O+5vCAJKtba19Lkqr6cJLDk3x1Q4PW2vX9ujsHN2ytXTPw+FtV9b0ki5Ksm/uyAQAA5rdRDyvdK8kNA8/X9MumpaoOSHLPJNdNsO7oqlpVVavWrl0740IBAAC2JVvdrSyq6oFJ/iHJ77fW7hxe31p7b2ttaWtt6aJFi0ZfIAAAwFZo1OHwxiQPGni+d79sSqrqPknOSfKa1tp4bwICAAAwj4z6msMLk+xbVfukC4UvTPLiqWxYVfdM8rEkH2ytnTV3JQIAAPPWypNnd38HHb/J1evWrcsZZ5yRZcuWTXvX73jHO3L00Udn5513nml10zLScNhau6OqXp7k3CTbJTmttXZFVZ2UZFVr7eyqely6ELhbkt+sqjf0M5S+IMmBSRZW1VH9Lo9qrV0yyvcAU7Xk9CUz3nbUN2cFAGBurFu3LsuXL59xOHzJS14yP8NhkrTWViRZMbTshIHHF6Ybbjq83YeSfGjOCwQAAJglxx13XK677rrsv//+ecYznpE99tgjZ555Zn784x/nuc99bt7whjfkBz/4QV7wghdkzZo1+dnPfpbXve51+e53v5tvfetbOeigg7L77rtn5cqVc17ryMMhAADAtuKNb3xjLr/88lxyySU577zzctZZZ+XLX/5yWms57LDDcsEFF2Tt2rXZc889c8455yRJbr311uy6665529velpUrV2b33XcfSa1b3WylAAAAW6Pzzjsv5513Xh796EfnMY95TK666qpce+21WbJkSc4///y8+tWvzmc/+9nsuuuuY6lPzyEAAMAItNZy/PHH55hjjrnbuosvvjgrVqzIa1/72jz96U/PCSecMMEe5paeQwAAgDly73vfO7fddluS5FnPelZOO+203H777UmSG2+8Md/73vfyrW99KzvvvHNe8pKX5FWvelUuvvjiu207CnoOAQCAbcckt56YbQsXLswTn/jEPOpRj8ohhxySF7/4xXnCE56QJNlll13yoQ99KKtXr86rXvWq3OMe98gOO+yQv/3bv02SHH300Tn44IOz5557jmRCmmqtzfmLjMvSpUvbqlWrxl0G2yi3sgAAGL8rr7wyj3jEI8ZdxkhM9F6r6qLW2tKpbG9YKQAAAMIhAAAAwiEAADDPzedL6TaYjfcoHAIAAPPWggULcvPNN8/rgNhay80335wFCxZs1n7MVgoAAMxbe++9d9asWZO1a9eOu5Q5tWDBguy9996btQ/hEAAAmLd22GGH7LPPPuMuY6tgWCkAAADCIQAAAMIhAAAAEQ4BAACIcAgAAECEQwAAACIcAgAAEOEQAACACIcAAABEOAQAACDCIQAAABEOAQAAiHAIAABAhEMAAAAiHAIAABDhEAAAgAiHAAAARDgEAAAgwiEAAAARDgEAAEiy/bgLAO5uyelLNmv7y468bJYqAQBgW6HnEAAAAOEQAAAA4RAAAIAIhwAAAEQ4BAAAIMIhAAAAEQ4BAACIcAgAAECEQwAAACIcAgAAEOEQAACACIcAAABEOAQAACDJ9uMuAOarY29ZN+Nt/3a3+85iJQAAMDk9hwAAAAiHAAAACIcAAABEOAQAACAmpIFNWnL6khlve+ws1gEAAHNNzyEAAAB6DmFLtDm3wQAAgJnQcwgAAIBwCAAAgHAIAABAhEMAAAAiHAIAABDhEAAAgAiHAAAARDgEAAAgwiEAAAARDgEAAIhwCAAAQIRDAAAAIhwCAAAQ4RAAAIAIhwAAAEQ4BAAAIMIhAAAAEQ4BAACIcAgAAECEQwAAACIcAgAAEOEQAACACIcAAABEOAQAACBjCIdVdXBVXV1Vq6vquAnWH1hVF1fVHVV1xNC6I6vq2v7nyNFVDQAAML+NNBxW1XZJTklySJL9kryoqvYbavbNJEclOWNo2/sleX2SX0tyQJLXV9Vuc10zAADAtmDUPYcHJFndWvtaa+0nST6c5PDBBq2161trX0ly59C2z0pyfmvt+621W5Kcn+TgURQNAAAw3406HO6V5IaB52v6ZbO2bVUdXVWrqmrV2rVrZ1woAADAtmTeTUjTWntva21pa23pokWLxl0OAADAVmHU4fDGJA8aeL53v2yutwUAAGATRh0OL0yyb1XtU1X3TPLCJGdPcdtzkzyzqnbrJ6J5Zr8MAACAzTTScNhauyPJy9OFuiuTnNlau6KqTqqqw5Kkqh5XVWuSPD/JqVV1Rb/t95P8ZbqAeWGSk/plAAAAbKbtR/2CrbUVSVYMLTth4PGF6YaMTrTtaUlOm9MCAQAAtkHzbkIaAAAApk84BAAAQDgEAABAOAQAACBjmJAGmHtLTl8y420vO/KyWawEAICthZ5DAAAAhEMAAACEQwAAACIcAgAAEOEQAACAmK0U5qVjb1k37hIAANjK6DkEAABAzyFsih44AAC2FXoOAQAAEA4BAAAQDgEAAIhwCAAAQIRDAAAAIhwCAAAQ4RAAAIAIhwAAAEQ4BAAAIMIhAAAAEQ4BAACIcAgAAECEQwAAACIcAgAAEOEQAACACIcAAABEOAQAACDCIQAAABEOAQAAiHAIAABAhEMAAAAiHAIAABDhEAAAgAiHAAAARDgEAAAgwiEAAAARDgEAAIhwCAAAQIRDAAAAIhwCAAAQ4RAAAIAIhwAAAEQ4BAAAIMIhAAAAEQ4BAACIcAgAAECEQwAAACIcAgAAEOEQAACACIcAAABEOAQAACDCIQAAABEOAQAAiHAIAABAhEMAAAAiHAIAABDhEAAAgAiHAAAARDgEAAAgwiEAAAARDgEAAIhwCAAAQIRDAAAAIhwCAAAQ4RAAAIAIhwAAAEQ4BAAAIMIhAAAAEQ4BAACIcAgAAECEQwAAACIcAgAAEOEQAACACIcAAABEOAQAACBjCIdVdXBVXV1Vq6vquAnW71hVH+nXf6mqFvfLd6iq06vqsqq6sqqOH3XtAAAA89VIw2FVbZfklCSHJNkvyYuqar+hZi9Ncktr7ZeTvD3Jm/rlz0+yY2ttSZLHJjlmQ3AEAABg84y65/CAJKtba19rrf0kyYeTHD7U5vAkp/ePz0ry9KqqJC3Jvapq+yQ7JflJkv8ZTdkAAADz26jD4V5Jbhh4vqZfNmGb1todSW5NsjBdUPxBkm8n+WaSt7bWvj/8AlV1dFWtqqpVa9eunf13AAAAMA9tTRPSHJDkZ0n2TLJPkldW1UOHG7XW3ttaW9paW7po0aJR1wgAALBVGnU4vDHJgwae790vm7BNP4R01yQ3J3lxkv9orf20tfa9JJ9LsnTOKwYAANgGjDocXphk36rap6rumeSFSc4eanN2kiP7x0ck+XRrraUbSvq0JKmqeyV5fJKrRlI1AADAPDfScNhfQ/jyJOcmuTLJma21K6rqpKo6rG/2viQLq2p1kj9LsuF2F6ck2aWqrkgXMt/fWvvKKOsHAACYr7Yf9Qu21lYkWTG07ISBx+vT3bZieLvbJ1oOAADA5tuaJqQBAABgjgiHAAAACIcAAAAIhwAAAGQME9LAqC05fcmMtz12FusAAIAtmZ5DAAAA9BwCQ1aePPNtDzp+9uoAAGCk9BwCAAAgHAIAACAcAgAAEOEQAACACIcAAABEOAQAACDCIQAAABEOAQAAiHAIAABAhEMAAAAiHAIAABDhEAAAgAiHAAAARDgEAAAgwiEAAAARDgEAAEiy/bgLgLl27C3rxl0CAABs8fQcAgAAIBwCAAAgHAIAABDhEAAAgJiQBhiy/NLlM9522UHHz2IlAACMkp5DAAAAhEMAAACEQwAAACIcAgAAkGmGw6rabq4KAQAAYHym23N4Y1W9uaoeMSfVAAAAMBbTDYfvSXJEksur6ktVdXRV3WcO6gIAAGCEphUOW2snttYemuQZSa5O8rYk366qf6yqX5+LAgEAAJh7M5qQprX26dba7yV5QJL/neThSc6tquur6sSq2nM2iwQAAGBube5spUuTHJjkV5LckuSzSf4wyeqqeslm7hsAAIARmXY4rKqHVNXrq+q6JJ9K8sAkf5Bkz9ba7yZ5SJJTk7xlVisFAABgzmw/ncZVtTLJk5PcmOT9Sd7fWvvGYJvW2s+q6owkr5i1KgEAAJhT0wqHSb6X5NAk57fW2ibaXZJknxlXBQAAwEhNd1jpKUk+P1EwrKpdqurAJGmt/XS4RxEAAIAt13TD4cok+21k3cP79QAAAGxlphsOaxPrdknyw82oBQAAgDGZ9JrDfqjoUwcW/WFVHTzUbEGSZye5bPZKAwAAYFSmMiHNr6W70X2StCTPT3LHUJufJLkqyatmrzQAAABGZdJw2Fp7S/p7FlbV15M8t7V2yVwXBgAAwOhM61YWrTW3pwAAAJiHpnLN4aEnZHg/AAAcyklEQVRJ/qu19j/9401qra2YlcoAAAAYman0HP57kscn+XL/uGXjs5a2JNvNTmkAAACMylTC4T5Jvj3wGAAAgHlmKhPSfGOixwAAAMwfU7nmcOfp7LC19sOZlwMAAMA4TGVY6e3priWcKtccAgAAbGWmEg7/INMLhwAAAGxlpnLN4QdGUAcAAABjdI9xFwAAAMD4TWVCmi8nOaq19tWqujCTDDFtrR0wW8UBAAAwGlO55vCKJD8aeOz6QwAAgHlmKtcc/v7A46PmtBoAAADGYsbXHFZnUVXVbBYEAADA6E07HFbVoVX1+STrk3wnyfqq+nxVPXvWqwMAAGAkphUOq+qYJP+W5PYkr0jy/P7f25Oc3a8HAABgKzOVCWkG/Z8kp7bWlg0tf09VvSfJa5KcOiuVAQAAMDLTHVa6MMnHNrLuo0nut3nlAAAAMA7TDYcrkzxlI+uekuSCzSsHAACAcZh0WGlV7Tfw9F1J/r6qFib5eJLvJdkjyXOTHJLkD+eiSAAAAObWVK45vDx3vfF9JTmm/2n98w3+I8l2s1YdAAAAIzGVcHjQnFcBAADAWE0aDltr/zmKQgAAABif6d7K4ueq6h5JFgwvb639cLMqAgAAYOSmNVtpdV5dVauT/DTJbRP8AAAAsJWZ7q0s/iTJcUnel24imr9OclKSa5Jcn+To2SwOAACA0ZhuOPyjJK9P8ub++cdba29I8sgkVyXZdxZrAwAAYESmGw73SXJJa+1n6YaV3jdJWmt3Jlme5MjZLQ8AAIBRmG44vDnJLv3jbyZ59MC63ZLsNBtFAQAAMFrTna30c0kel2RFkjOSnFhV90vykyQvS/Kp2S0PAACAUZhuODwxyV79479JN6z0qHQ9hucn+d+T7aCqDk7yziTbJfn71tobh9bvmOSDSR6brqfyt1tr1/frfjXJqUnuk+TOJI9rra2f5nsA5srKk2e+7UHHz14dAABM27TCYWvt6iRX949/nOQV/c+UVNV2SU5J8owka5JcWFVnt9a+OtDspUluaa39clW9MMmbkvx2VW2f5ENJfre1dmlVLUx33SMAAACbabrXHP5cVe1dVY+rqr0mb/1zByRZ3Vr7WmvtJ0k+nOTwoTaHJzm9f3xWkqdXVSV5ZpKvtNYuTZLW2s39xDgAAABspmmHw6o6tqpuSPKNJF9K8s2qWlNVy6aw+V5Jbhh4via/GKZ6tzattTuS3JpkYZKHJWlVdW5VXVxVf7GR+o6uqlVVtWrt2rXTem8AAADbqmmFw6o6Icm7k3wiybOTLO3//USSd/Xr58r2SZ6U5Hf6f59bVU8fbtRae29rbWlrbemiRYvmsBwAAID5Y7oT0rwsyd+01l43tPw/quq7/fqTNrH9jUkeNPB8737ZRG3W9NcZ7ppuYpo1SS5ord2UJFW1IsljYoZUAACAzTbdYaU7JblgI+v+M8mCSba/MMm+VbVPVd0zyQuTnD3U5uwkR/aPj0jy6dZaS3JukiVVtXMfGp+S5KsBAABgs003HH48yfM2su63kvz7pjburyF8ebqgd2WSM1trV1TVSVV1WN/sfUkWVtXqJH+W5Lh+21uSvC1dwLwkycWttXOmWT8AAAATmHRYaVUdOvD0E0neXFWL0wXF7yXZI8lzkzwyyYSTxAxqra1IsmJo2QkDj9cnef5Gtv1QuttZAAAAMIumcs3hvydpSWpg2V5JnjVB2w8l+adZqAsAAIARmko43GfOqwAAAGCsJg2HrbVvjKIQYOu3/NLlM9522UHHz2IlAABM13RvZZF+ptDfSnevwfsl+X6Szyb5l37CGQAAALYy0wqHVbVHkvOS/GqS65N8N8kT0t3f8NKqemZrbe1sFwkAAMDcmu6tLN6WZGGSx7fWHtpae0Jr7aFJfq1f/rbZLhAAAIC5N91weGiSV7fWvjy4sLV2YZLjkzx7tgoDAABgdKYbDndMcttG1t2W5J6bVw4AAADjMN1w+MUkr66qew0u7J+/ul8PAADAVma6s5W+MsnKJDdU1XnpJqTZI8mzklSSp85qdQAAAIzEtHoOW2uXJNk3yXuTLEryjHTh8D1J9m2tXTrrFQIAADDnptxzWFU7JDkgyddba8fNXUkAAACM2nR6Dn+W5NNJfmWOagEAAGBMphwOW2t3Jrk2yQPmrhwAAADGYbqzlb4myQlVtWQuigEAAGA8pjtb6WuTLExySVXdmG620jbYoLV2wCzVBgAAwIhMNxxe3v8AAAAwj0wpHFbVTkkOTRcMv5Pkk621785lYQAAAIzOpOGwqh6a5JNJFg8s/p+qekFr7by5KgwAAIDRmcqENG9OcmeSJyfZOckjk/x3klPnsC4AAABGaCrh8AlJXtta+1xrbX1r7cokxyR5cFU9cG7LAwAAYBSmEg4fmORrQ8uuS1Jxz0MAAIB5Yar3OWyTNwEAAGBrNdVbWZxbVXdMsPxTw8tba3tsflkAAACM0lTC4RvmvAoAAADGatJw2FoTDgEAAOa5qV5zCAAAwDwmHAIAACAcAgAAIBwCAAAQ4RAAAIAIhwAAAEQ4BAAAIMIhAAAAEQ4BAACIcAgAAECEQwAAACIcAgAAEOEQAACACIcAAABEOAQAACDJ9uMuAKZk5cnjrgAAAOY1PYcAAAAIhwAAAAiHAAAAxDWHwJZic68rPej42akDAGAbpecQAAAA4RAAAADhEAAAgAiHAAAAxIQ0bCWWX7p83CUAAMC8pucQAAAA4RAAAADhEAAAgAiHAAAARDgEAAAgwiEAAAARDgEAAIj7HAJbiM29l+Wyg46fpUoAALZNeg4BAAAQDgEAABAOAQAAiHAIAABAhEMAAAAiHAIAABDhEAAAgAiHAAAARDgEAAAgwiEAAAARDgEAAIhwCAAAQIRDAAAAIhwCAAAQ4RAAAIAIhwAAAEQ4BAAAIMIhAAAAEQ4BAADIGMJhVR1cVVdX1eqqOm6C9TtW1Uf69V+qqsVD6x9cVbdX1Z+PqmYAAID5bqThsKq2S3JKkkOS7JfkRVW131Czlya5pbX2y0nenuRNQ+vfluQTc10rAADAtmTUPYcHJFndWvtaa+0nST6c5PChNocnOb1/fFaSp1dVJUlVPSfJ15NcMaJ6AQAAtgmjDod7Jblh4PmaftmEbVprdyS5NcnCqtolyauTvGFTL1BVR1fVqqpatXbt2lkrHAAAYD7bmiakOTHJ21trt2+qUWvtva21pa21pYsWLRpNZQAAAFu57Uf8ejcmedDA8737ZRO1WVNV2yfZNcnNSX4tyRFV9eYk901yZ1Wtb629e+7LBrZ0S05fMuNtLzvyslmsBABg6zTqcHhhkn2rap90IfCFSV481ObsJEcm+UKSI5J8urXWkjx5Q4OqOjHJ7YIhAADA7BhpOGyt3VFVL09ybpLtkpzWWruiqk5Ksqq1dnaS9yX5h6paneT76QIkwCYde8u6cZcAALBVG3XPYVprK5KsGFp2wsDj9UmeP8k+TpyT4gAAALZRW9OENAAAAMwR4RAAAADhEAAAAOEQAACACIcAAABEOAQAACDCIQAAABEOAQAAiHAIAABAhEMAAAAiHAIAABDhEAAAgAiHAAAARDgEAAAgwiEAAAARDgEAAIhwCAAAQIRDAAAAIhwCAAAQ4RAAAIAIhwAAAEQ4BAAAIMIhAAAAEQ4BAACIcAgAAECS7cddAMC4LTl9yYy3vezIy2axEgCA8dFzCAAAgHAIAACAYaUAOfaWdeMuAQBg7PQcAgAAIBwCAAAgHAIAABDXHDIqK08edwUAAMAm6DkEAABAOAQAAEA4BAAAIMIhAAAAEQ4BAACIcAgAAECEQwAAACIcAgAAEOEQAACACIcAAABEOAQAACDCIQAAABEOAQAAiHAIAABAhEMAAAAiHAIAABDhEAAAgAiHAAAAJNl+3AUAbNVWnjzzbQ86fvbqAADYTMIhwGZYfunyGW+7TDgEALYghpUCAAAgHAIAACAcAgAAEOEQAACACIcAAABEOAQAACDCIQAAAHGfQ0Zkc+4FBwAAzD09hwAAAAiHAAAACIcAAABEOAQAACDCIQAAABEOAQAAiHAIAABAhEMAAAAiHAIAABDhEAAAgAiHAAAARDgEAAAgwiEAAAARDgEAAIhwCAAAQJLtx10AwDZr5ckz3/ag42evDgCA6DkEAAAgYwiHVXVwVV1dVaur6rgJ1u9YVR/p13+pqhb3y59RVRdV1WX9v08bde0AAADz1UjDYVVtl+SUJIck2S/Ji6pqv6FmL01yS2vtl5O8Pcmb+uU3JfnN1tqSJEcm+YfRVA0AADD/jbrn8IAkq1trX2ut/STJh5McPtTm8CSn94/PSvL0qqrW2n+31r7VL78iyU5VteNIqgYAAJjnRh0O90pyw8DzNf2yCdu01u5IcmuShUNtfivJxa21Hw+/QFUdXVWrqmrV2rVrZ61wAACA+Wyrm5Cmqh6ZbqjpMROtb629t7W2tLW2dNGiRaMtDgAAYCs16ltZ3JjkQQPP9+6XTdRmTVVtn2TXJDcnSVXtneRjSX6vtXbd3JcLMHeWX7p8xtsucysLAGCWjbrn8MIk+1bVPlV1zyQvTHL2UJuz0004kyRHJPl0a61V1X2TnJPkuNba50ZWMQAAwDZgpOGwv4bw5UnOTXJlkjNba1dU1UlVdVjf7H1JFlbV6iR/lmTD7S5enuSXk5xQVZf0P3uMsn4AAID5atTDStNaW5FkxdCyEwYer0/y/Am2+6skfzXnBQIAAGyDtroJaQAAAJh9wiEAAADCIQAAAMIhAAAAEQ4BAACIcAgAAEDGcCsLAGbBypM3b/uDjp+dOgCAeUPPIQAAAHoOAbZGyy9dvlnbL9NzCAAM0XMIAACAcAgAAIBwCAAAQIRDAAAAIhwCAAAQ4RAAAIAIhwAAAMR9DpmOlSePuwIAAGCO6DkEAABAzyHANmlzRgIcdPzs1QEAbDGEQ4Bt0PJLl89422XCIQDMS4aVAgAAIBwCAAAgHAIAABDhEAAAgAiHAAAARDgEAAAgwiEAAABxn0MApmnJ6UtmvO1lR142i5UAALNJzyEAAAB6DgGYnmNvWTfuEgCAOaDnEAAAAOEQAAAA4RAAAIAIhwAAAEQ4BAAAIMIhAAAAcSsLAEZoyelLZrztZUdeNouVAADD9BwCAACg5xCA0Tn2lnXjLgEA2Ag9hwAAAAiHAAAACIcAAADENYcAbCU2Z6bTxGynADAZ4RCArYLJbABgbhlWCgAAgJ5Dpm75pcvHXQIAADBHhEMAtg0rT575tgcdP3t1AMAWyrBSAAAA9BwCsG3YnKHxy/QcArAN0HMIAACAcAgAAIBhpQAwqeXveNCMt132pzfMYiUAMHf0HAIAAKDnEADmlFtoALCVEA4BYA5tziyp2ZwZVg1nBWCaDCsFAABAOAQAAMCw0m3P5lz7AgAAzFvCIQDMR5v7x0CT4QBsc4RDAJiHNmsinCTLhEOAbY5rDgEAABAOAQAAEA4BAACIaw4BgAksf8eDZrztsj+9YRYrAWBU9BwCAAAgHAIAACAcAgAAENccAgCzbeXJM9/W/RUBxkbPIQAAAHoOAYDZtfzS5TPedpmeQ4Cx0XMIAACAnkP+//buPdrysq7j+PvjgAJJciduAimKCAIRKHEJWHGRiElBQWQtLIoUKESKQCvJIjAVIUNNYQRJQUTAEZIBB4hB4qrScA3SUUFg4iYIAgLf/vg9Z7E9noGZOefsPfvM+7XWXvv3e36X57v3POv85ruf5/f8JElacvh8RUkaHHsOJUmSJEn2HEqSpCnCWVIlaVxMDiVJ0pTgRDiSND4mh5IkSePpdQR7HiVNCSaHkiRpqTeeXkew51HS1GByKEmSNE7jmmV180MXv2KTUkkTqO/JYZI9gFOAacBpVXXiqO2vAL4IbAU8BOxXVfPatmOBg4HngL+oqll9DF2SJGnCjavXcpw9noPymZVXWuxj5x40dwIjkdSrr8lhkmnAqcCuwD3ADUlmVtVtPbsdDDxSVa9Nsj/wUWC/JJsA+wNvBNYGvpXkdVX1XD8/gyRJksbnfY88utjHjqeXdjx8jqaWBv3uOdwGuLuqvg+Q5BxgOtCbHE4HjmvL5wH/miSt/Jyqehr4QZK72/n+q0+xLzE2O3OzxT52PH+MJUmSllaDSko1PKbCDwj9Tg7XAXq/tXuANy9on6p6NslPgVVb+bWjjl1ndAVJDgEOaas/S3LnxIQ+oVYDHhxExYcNolINwsDamJYKti9NJtuXJpPtS5PmsCOzpLav9Rd2xyk3IU1VfQ743KDjeDFJbqyq3x50HJq6bGOaTLYvTSbblyaT7UuTaSq0r5f1ub57gd4++XVb2Zj7JFkGeBXdxDQLc6wkSZIkaTH0Ozm8AdgoyYZJXk43wczMUfvMBA5qy/sCl1dVtfL9k7wiyYbARsD1fYpbkiRJkqa0vg4rbfcQHg7MonuUxYyqujXJR4Abq2omcDpwVptw5mG6BJK237l0k9c8Cxw2xDOVLtHDXjUl2MY0mWxfmky2L00m25cm09C3r3SdcpIkSZKkpVm/h5VKkiRJkpZAJoeSJEmSJJPDfkuyR5I7k9yd5JhBx6PhlmRGkvlJbukpWyXJZUnuau8rDzJGDa8k6yW5IsltSW5NckQrt41pQiRZLsn1SW5ubezvW/mGSa5r18qvtEnspEWWZFqS7ya5qK3btjRhksxLMjfJ95Lc2MqG+hppcthHSaYBpwJvBTYB3pVkk8FGpSF3BrDHqLJjgNlVtREwu61Li+NZ4Kiq2gR4C3BY+5tlG9NEeRrYpao2B7YA9kjyFuCjwCer6rXAI8DBA4xRw+0I4PaedduWJtrOVbVFz/MNh/oaaXLYX9sAd1fV96vqGeAcYPqAY9IQq6qr6Gb17TUdOLMtnwn8YV+D0pRRVfdV1Xfa8uN0/8FaB9uYJkh1ftZWl22vAnYBzmvltjEtliTrAr8PnNbWg21Lk2+or5Emh/21DvDjnvV7Wpk0kdasqvva8v3AmoMMRlNDkg2ALYHrsI1pArVhf98D5gOXAf8LPFpVz7ZdvFZqcZ0MHA0839ZXxbaliVXApUluSnJIKxvqa2Rfn3Moqb+qqpL4vBqNS5JXAl8D3l9Vj3U/vndsYxqv9sziLZKsBFwAbDzgkDQFJNkLmF9VNyXZadDxaMravqruTbIGcFmSO3o3DuM10p7D/roXWK9nfd1WJk2kB5KsBdDe5w84Hg2xJMvSJYZfqqrzW7FtTBOuqh4FrgC2BVZKMvIDttdKLY7tgL2TzKO7jWcX4BRsW5pAVXVve59P9+PWNgz5NdLksL9uADZqM2W9HNgfmDngmDT1zAQOassHAV8fYCwaYu3+nNOB26vqpJ5NtjFNiCSrtx5DkiwP7Ep3b+sVwL5tN9uYFllVHVtV61bVBnT/37q8qt6NbUsTJMmvJVlxZBnYDbiFIb9GpmqoejqHXpI96cbATwNmVNXxAw5JQyzJ2cBOwGrAA8CHgQuBc4FXAz8E3llVoyetkV5Sku2BOcBcXrhn54N09x3axjRuSd5EN2HDNLofrM+tqo8k+U263p5VgO8CB1bV04OLVMOsDSv9y6ray7alidLa0gVtdRngy1V1fJJVGeJrpMmhJEmSJMlhpZIkSZIkk0NJkiRJEiaHkiRJkiRMDiVJkiRJmBxKkiRJkjA5lCRNsiTHJakks8bYdl6SK/sYy04tlk37VeeiSPKGJHOSPNHi3GDQMS1IkhuTnDHoOCRJE8fkUJLUL7sl2XrQQSzhPgasBOwNbAvcN9hwJElLE5NDSVI/PAzMBT406EAmU5LlxnmKjYHLqmp2VV07yIdzpzPezyNJGiImh5KkfijgeGDvJJstaKc2BPXBMcoryeE96/OSfDzJMUnuS/LTJJ9oCc2eSW5N8niSC5OsPEZVaye5qA3f/FGS945R5w5J/jPJk0keSvL5JCv2bH9Pi2ubJFcm+TnwVy/y2bZIMrud75EkX0qyZtu2QZICXgMc2c575QLOc2aSS3vWX9/2P7+nbKtWtlFP2eFJ7krydJK7kxw56rzHJXkwyfZJbgCeAt7Rtm2a5NtJnkpye5K9x4jrjUkuSfJw+15vT3LYgr4PSdKSx+RQktQvXwXuYuJ6D/cHtgH+CPhn4APAScA/AH8LvBf4XeCEMY49Hfhv4O3AfwCfSbLXyMYk2wHfAu4H9gXeD+wJfGGMc50NfKNtv2isQJOsDlwJrAAcAPx5i+2yJC+nGz66bavvy2350AV87jnAtkmmtfUd6RK57Xv22RF4oKruavX/KfApYCbwB3T/Fp9Icsyoc68AnAmcBuwBXJ9keWAW8MoW+z8CJwOvHnXsN4DngAPphsV+ClgRSdLQWGbQAUiSlg5V9XySE4DTk/xdVf3POE/5FPCOqnoOuCTJdLqka6Oq+gFAks2Bg+gSxV7frKoPtuVZSV4D/A0vJHcnAtdU1X4jByS5F5idZNOquqXnXP9SVae8RKxHtffdq+qxdr67gGuBfarqbODaJE8D91XVtS9yrjl0idqWwI3ADnQJ3cFJNq6qO1rZnFbPy4DjgDOqaiSOS5O8Cjg2yclV9VQrXx74QFV9vedzHwqsAby5qu5pZfOAq3v2WQ3YEJheVXNb8eyX+E4kSUsYew4lSf3078CPgGMn4FxXtsRwxN3AvJHEsKds9dY71+uCUevnA1slmZZkBbqeu3OTLDPyokuGfgFsNerYixci1m2AS0cSQ4Cqug6Yxy/3+L2kqroTmE+XAELXS/hN4Ds9ZdvTkkNgXWBtut7CXl8Bfh3oHeZb7VyjY79pJDFsMXy7xTDiYeDHwGeT7JdkjUX5TJKkJYPJoSSpb6rqWbohoAcmWX+cp3t01PozCygLMDo5nD/G+jLAasDKwDTg03TJ4MjraWBZYL1Rxz6wELGutYD9HgBWWYjjR5sD7JBkPbrhnVf3lL0BWJ0XksO1FhDnyHpv/Y9U1TOj9vsNfvX7oresqp4HdqMbFjsDuL89kmPLRfpUkqSBMjmUJPXbDLrE4q/H2PYUoxK5BUwoM16je7bWAJ4FHqRLMAv4MLD1GK8Zo46thajvvjHqBFiTrtdtUc2h6x3cEbitqh5qZTu0sseAm3vqZoz612zvvfWP9VnuH+PYXzlfVd1RVfvQPYrj94DlgIvbsFZJ0hDwD7Ykqa/a4xk+DvwxL/RqjbgHWDHJOj1lu01CGG8bY/2mqnquqp6guxfw9VV14xivnyxGfdcBu4+a7XRrYAN67t1bBFfR9Q4e0pahSw7Xp5s05prWmwfdd/oT2syjPd5Jl0TO5cXdQDfkdt2e2Ldj7ISRqvpFVV1ONznQWnTJoiRpCJgcSpIG4d+Ax4HfGVV+CfBzYEaS3ZL8GTB6Rs2J8NYkx7c6PgvsCvxTz/ajgX2TnJVkepJd2qMrvprkdYtR30ntfVY737vp7nOcC3xtMc53M11ityMtOayqh4HbWtnIkNKRIZ/HAe9J8rH2mY8H3gec0DMZzYJ8ga5H9eIkb0tyAPDFVgZAkjcluTTJwUl2TvJ2up7hm1tckqQhYHIoSeq7qnoS+OQY5Q8C+9BNonIh3WMRDpiEEP4E+K1Wx17AYVU1syeOq+mSrNWBs+ge03A03aQrC3OP4S+pqv8DdqYbNns2cCpdArfrGPf4Lcz5ngeuaatX9WwaSQqvHrX/54Ej6HpILwLeBRxVVScuRF1PArsDTwDn0A23PQr4Yc9u99N9Lx+im9Dm08DtdI+0kCQNiVQtzK0SkiRJkqSpzJ5DSZIkSZLJoSRJkiTJ5FCSJEmShMmhJEmSJAmTQ0mSJEkSJoeSJEmSJEwOJUmSJEmYHEqSJEmSgP8HUgTI+T7ud5AAAAAASUVORK5CYII=\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7fb436dac2b0>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "dist_train = train_qs.apply(lambda x: len(x.split(' ')))\n",
    "dist_test = test_qs.apply(lambda x: len(x.split(' ')))\n",
    "\n",
    "plt.figure(figsize=(15, 10))\n",
    "plt.hist(dist_train, bins=50, range=[0, 50], color=pal[2], normed=True, label='train')\n",
    "plt.hist(dist_test, bins=50, range=[0, 50], color=pal[1], normed=True, alpha=0.5, label='test')\n",
    "plt.title('Normalised histogram of word count in questions', fontsize=15)\n",
    "plt.legend()\n",
    "plt.xlabel('Number of words', fontsize=15)\n",
    "plt.ylabel('Probability', fontsize=15)\n",
    "\n",
    "print('mean-train {:.2f} std-train {:.2f} mean-test {:.2f} std-test {:.2f} max-train {:.2f} max-test {:.2f}'.format(dist_train.mean(), \n",
    "                          dist_train.std(), dist_test.mean(), dist_test.std(), dist_train.max(), dist_test.max()))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3",
   "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": "3.6.3"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}