The Titanic Survival problem.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [],
   "source": [
    "import pandas as pd\n",
    "import numpy as np\n",
    "import seaborn as sns\n",
    "import matplotlib\n",
    "import pydot\n",
    "import re\n",
    "\n",
    "%matplotlib inline\n",
    "import matplotlib\n",
    "import matplotlib.pyplot as plt\n",
    "\n",
    "\n",
    "import sklearn\n",
    "from sklearn.preprocessing import StandardScaler\n",
    "from sklearn.model_selection import cross_val_score, train_test_split\n",
    "from sklearn.linear_model import LogisticRegression\n",
    "from sklearn import metrics\n",
    "\n",
    "from sklearn.tree import export_graphviz, DecisionTreeClassifier"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "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>PassengerId</th>\n",
       "      <th>Survived</th>\n",
       "      <th>Pclass</th>\n",
       "      <th>Name</th>\n",
       "      <th>Sex</th>\n",
       "      <th>Age</th>\n",
       "      <th>SibSp</th>\n",
       "      <th>Parch</th>\n",
       "      <th>Ticket</th>\n",
       "      <th>Fare</th>\n",
       "      <th>Cabin</th>\n",
       "      <th>Embarked</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>3</td>\n",
       "      <td>Braund, Mr. Owen Harris</td>\n",
       "      <td>male</td>\n",
       "      <td>22.0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>A/5 21171</td>\n",
       "      <td>7.2500</td>\n",
       "      <td>NaN</td>\n",
       "      <td>S</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>2</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>Cumings, Mrs. John Bradley (Florence Briggs Th...</td>\n",
       "      <td>female</td>\n",
       "      <td>38.0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>PC 17599</td>\n",
       "      <td>71.2833</td>\n",
       "      <td>C85</td>\n",
       "      <td>C</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>3</td>\n",
       "      <td>1</td>\n",
       "      <td>3</td>\n",
       "      <td>Heikkinen, Miss. Laina</td>\n",
       "      <td>female</td>\n",
       "      <td>26.0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>STON/O2. 3101282</td>\n",
       "      <td>7.9250</td>\n",
       "      <td>NaN</td>\n",
       "      <td>S</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>4</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>Futrelle, Mrs. Jacques Heath (Lily May Peel)</td>\n",
       "      <td>female</td>\n",
       "      <td>35.0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>113803</td>\n",
       "      <td>53.1000</td>\n",
       "      <td>C123</td>\n",
       "      <td>S</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>5</td>\n",
       "      <td>0</td>\n",
       "      <td>3</td>\n",
       "      <td>Allen, Mr. William Henry</td>\n",
       "      <td>male</td>\n",
       "      <td>35.0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>373450</td>\n",
       "      <td>8.0500</td>\n",
       "      <td>NaN</td>\n",
       "      <td>S</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   PassengerId  Survived  Pclass  \\\n",
       "0            1         0       3   \n",
       "1            2         1       1   \n",
       "2            3         1       3   \n",
       "3            4         1       1   \n",
       "4            5         0       3   \n",
       "\n",
       "                                                Name     Sex   Age  SibSp  \\\n",
       "0                            Braund, Mr. Owen Harris    male  22.0      1   \n",
       "1  Cumings, Mrs. John Bradley (Florence Briggs Th...  female  38.0      1   \n",
       "2                             Heikkinen, Miss. Laina  female  26.0      0   \n",
       "3       Futrelle, Mrs. Jacques Heath (Lily May Peel)  female  35.0      1   \n",
       "4                           Allen, Mr. William Henry    male  35.0      0   \n",
       "\n",
       "   Parch            Ticket     Fare Cabin Embarked  \n",
       "0      0         A/5 21171   7.2500   NaN        S  \n",
       "1      0          PC 17599  71.2833   C85        C  \n",
       "2      0  STON/O2. 3101282   7.9250   NaN        S  \n",
       "3      0            113803  53.1000  C123        S  \n",
       "4      0            373450   8.0500   NaN        S  "
      ]
     },
     "execution_count": 2,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "train_df_raw = pd.read_csv('train.csv')\n",
    "test_df_raw = pd.read_csv('test.csv')\n",
    "train_df_raw.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "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>PassengerId</th>\n",
       "      <th>Survived</th>\n",
       "      <th>Pclass</th>\n",
       "      <th>Age</th>\n",
       "      <th>SibSp</th>\n",
       "      <th>Parch</th>\n",
       "      <th>Fare</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>count</th>\n",
       "      <td>891.000000</td>\n",
       "      <td>891.000000</td>\n",
       "      <td>891.000000</td>\n",
       "      <td>714.000000</td>\n",
       "      <td>891.000000</td>\n",
       "      <td>891.000000</td>\n",
       "      <td>891.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>mean</th>\n",
       "      <td>446.000000</td>\n",
       "      <td>0.383838</td>\n",
       "      <td>2.308642</td>\n",
       "      <td>29.699118</td>\n",
       "      <td>0.523008</td>\n",
       "      <td>0.381594</td>\n",
       "      <td>32.204208</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>std</th>\n",
       "      <td>257.353842</td>\n",
       "      <td>0.486592</td>\n",
       "      <td>0.836071</td>\n",
       "      <td>14.526497</td>\n",
       "      <td>1.102743</td>\n",
       "      <td>0.806057</td>\n",
       "      <td>49.693429</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>min</th>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.420000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>25%</th>\n",
       "      <td>223.500000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>2.000000</td>\n",
       "      <td>20.125000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>7.910400</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>50%</th>\n",
       "      <td>446.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>3.000000</td>\n",
       "      <td>28.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>14.454200</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>75%</th>\n",
       "      <td>668.500000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>3.000000</td>\n",
       "      <td>38.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>31.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>max</th>\n",
       "      <td>891.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>3.000000</td>\n",
       "      <td>80.000000</td>\n",
       "      <td>8.000000</td>\n",
       "      <td>6.000000</td>\n",
       "      <td>512.329200</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "       PassengerId    Survived      Pclass         Age       SibSp  \\\n",
       "count   891.000000  891.000000  891.000000  714.000000  891.000000   \n",
       "mean    446.000000    0.383838    2.308642   29.699118    0.523008   \n",
       "std     257.353842    0.486592    0.836071   14.526497    1.102743   \n",
       "min       1.000000    0.000000    1.000000    0.420000    0.000000   \n",
       "25%     223.500000    0.000000    2.000000   20.125000    0.000000   \n",
       "50%     446.000000    0.000000    3.000000   28.000000    0.000000   \n",
       "75%     668.500000    1.000000    3.000000   38.000000    1.000000   \n",
       "max     891.000000    1.000000    3.000000   80.000000    8.000000   \n",
       "\n",
       "            Parch        Fare  \n",
       "count  891.000000  891.000000  \n",
       "mean     0.381594   32.204208  \n",
       "std      0.806057   49.693429  \n",
       "min      0.000000    0.000000  \n",
       "25%      0.000000    7.910400  \n",
       "50%      0.000000   14.454200  \n",
       "75%      0.000000   31.000000  \n",
       "max      6.000000  512.329200  "
      ]
     },
     "execution_count": 3,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#train_df_raw.head()\n",
    "train_df_raw.describe()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "<class 'pandas.core.frame.DataFrame'>\n",
      "RangeIndex: 891 entries, 0 to 890\n",
      "Data columns (total 12 columns):\n",
      "PassengerId    891 non-null int64\n",
      "Survived       891 non-null int64\n",
      "Pclass         891 non-null int64\n",
      "Name           891 non-null object\n",
      "Sex            891 non-null object\n",
      "Age            714 non-null float64\n",
      "SibSp          891 non-null int64\n",
      "Parch          891 non-null int64\n",
      "Ticket         891 non-null object\n",
      "Fare           891 non-null float64\n",
      "Cabin          204 non-null object\n",
      "Embarked       889 non-null object\n",
      "dtypes: float64(2), int64(5), object(5)\n",
      "memory usage: 83.6+ KB\n"
     ]
    }
   ],
   "source": [
    "train_df_raw.info()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [],
   "source": [
    "# draw_missing_data_table(train_df_raw)\n",
    "#def draw_missing_data_table(df):\n",
    "#    total=df.isnull().sum().sort_values(ascending=False)\n",
    "#    percent=(df.isnull().sum()/df.isnull().count()).sort_values(ascending=False)\n",
    "#    missing_data=pd.concat([total,percent],axis=1,keys=['Total','Percent'])\n",
    "#    return missing_data\n",
    "#There is 77% of missing data in the cabin column, it's usually way too much for this column to be exploitable, but as we have a small numer of data, we will still try to use it in feature engineering.\n",
    "#For the age, we will either interpolate missing values or we will fill it with the median or mean. There is only two missing values for the embarked column, let's replace them by the most recurrent value of Embarqued variable\n",
    "#The most current values seems to be \"S\", which mean that majority of passengers embarked on the boat at Southampton !"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "S    644\n",
       "C    168\n",
       "Q     77\n",
       "Name: Embarked, dtype: int64"
      ]
     },
     "execution_count": 6,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "train_df_raw['Embarked'].value_counts()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAABIcAAANeCAYAAACI527yAAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADl0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uIDIuMi4yLCBodHRwOi8vbWF0cGxvdGxpYi5vcmcvhp/UCwAAIABJREFUeJzs3X+UZWV95/v3R0BF/NEiWmlprq3Xjj9iRzQVwoxJbgnRIGRsspYaDVcaQ27n3sEZc9Mzsc29a0wm8Q6uGyRqjCsdUdoEfzBEAleMVwY9Os4NKCgRFB2QtNJ2S/sD0MJE0+R7/zi7tNJUd53qOvucOme/X2vVOmc/5zl7f7/PqVP11Lf2s0+qCkmSJEmSJHXTQ8YdgCRJkiRJksbH4pAkSZIkSVKHWRySJEmSJEnqMItDkiRJkiRJHWZxSJIkSZIkqcMsDkmSJEmSJHWYxSFJkiRJkqQOszgkaSBJeknuSfKwccciSZKklUmyO8nfJ5lf9PXEccclaW2wOCRpWUk2Aj8HFPDisQYjSZKkI/WvquqRi772ruTJSY5qKzBJ42VxSNIgzgWuBy4Fti40Jnlckv8nyXeSfDrJHyT55KLHn57k2iTfTvKlJC8bfeiSJElaSpKHJLkiydeT3NucKf6MRY//RZK3JflwkvuBn0vy8CRvSnJXkruT/EmSh48xDUlDYHFI0iDOBS5rvn4xyUzT/jbgfuDH6BeNFheOjgOuBd4DPAF4BfAnSX5ihHFLkiTp8D4IbKI/n7sV+PODHv9V4PeARwF/A/wh8GTgJ5vnbQT+jxHFKqklqapxxyBpDUvys8DHgPVV9c0kXwT+FHgL8A/As6rqS03fPwDmqupnk/wK8Oqq+rlF+/pTYG9V/d7IE5EkSeqwJLuBE4ADTVOvqs4+qM8JwDeAR1bV/Un+AvhBVf1a8/hDgO8BT6uqrzRtPwe8s6o2jSYTSW04etwBSFrztgIfqapvNtvvadreS/9nyF2L+i6+/yTgZ5Lcu6jtaB783yhJkiSNxtlV9V8WNpprCP0n4CX0C0f/1Dx0Av2zw+Gfz+9+DHgY8LdJfribNgOWNBoWhyQdUpJjgZcBRyX5etP8MGAdMEP/P08bgP/ePHbSoqffBXy8ql4wonAlSZK0MucCZwKnAV8BHkf/zKHFBZ/FS03uBn5A/8yhu0cVpKT2ec0hSYdzNvAA8Ezg5ObrGcB/pT+Z+ADwu0kekeTpTduCDwI/nuSVSY5pvn568UUOJUmSNFaPAr4PfAt4BPCGw3WuqgeAdwB/lOTx6duQ5IXthyqpTRaHJB3OVuBdVfXVqvr6whfwx8A5wKuBxwBfp79c7L30JxhU1XeBFwIvB/Y2fd5I/8wjSZIkjd+76M/T9gKfB/6/AZ6znf5ZRp8C7gM+Qv/C1JImmBekljQ0Sd4I/FhVbV22syRJkiRpTfDMIUlHLMnTk/xkc0rxKcD5wJXjjkuSJEmSNDgvSC1pNR5FfynZE4H9wEXAVWONSJIkSZK0Ii4rkyRJkiRJ6jCXlUmSJEmSJHXYmlhWdsIJJ9TGjRuHtr/777+f4447bmj7W0umOTeY7vymOTeY7vzMbXJNc35t5XbTTTd9s6oeP/Qda81I8nDgE/Q/PfJo4Iqqen2SS4H/if6nDwGcV1U3JwnwZuBM4HtN+2cOd4xhz+0WTPN7epwc13Y4ru1wXNvhuLZjLYzroHO7NVEc2rhxIzfeeOPQ9tfr9Zibmxva/taSac4Npju/ac4Npjs/c5tc05xfW7kl+crQd6q15vvAaVU1n+QY4JNJ/rp57N9X1RUH9X8R/Y+p3gT8DPD25vaQhj23WzDN7+lxclzb4bi2w3Fth+PajrUwroPO7VxWJkmS1CHVN99sHtN8He4ilFuAdzfPux5Yl2R923FKkqTRWRNnDkmSJGl0khwF3AQ8FXhbVd2Q5H8D3pDkPwDXATuq6vvAicBdi56+p2nbd9A+twHbAGZmZuj1ekOPe35+vpX9dp3j2g7HtR2Oazsc13ZM0rhaHJIkSeqYqnoAODnJOuDKJM8CXgd8HXgosBN4LfAfgSy1iyX2ubN5HrOzs9XGafRr4fT8aeS4tsNxbYfj2g7HtR2TNK4uK5MkSeqoqroX6AFnVNW+ZunY94F3Aac03fYAJy162gZg70gDlSRJrbI4JEmS1CFJHt+cMUSSY4FfAL64cB2h5tPJzgZubZ5yNXBu+k4F7quqfUvsWpIkTSiXlUmSJHXLemBXc92hhwCXV9UHk3w0yePpLyO7Gfhfm/4fov8x9nfQ/yj7V40hZkmS1CKLQ5IkSR1SVZ8DnrNE+2mH6F/ABW3HJUmSxsdlZZIkSZIkSR1mcUiSJEmSJKnDXFamVm3ccc2yfXZfeNYIIpEkSZPulq/dx3nLzC2cV0iStHKrKg4l2Q18F3gAOFBVs0mOB94PbAR2Ay+rqntWF6YkSZIkSZLaMIxlZc+vqpOrarbZ3gFcV1WbgOuabUmSJEmSJK1BbVxzaAuwq7m/Czi7hWNIkiRJkiRpCFZ7zaECPpKkgD+tqp3ATFXtA6iqfUmesNQTk2wDtgHMzMzQ6/VWGcqPzM/PD3V/a8mk5bZ984Fl+yzOZ9LyW4lpzg2mOz9zm1zTnN805yZJkqTRWm1x6HlVtbcpAF2b5IuDPrEpJO0EmJ2drbm5uVWG8iO9Xo9h7m8tmbTclrtoJMDuc+Z+eH/S8luJac4Npjs/c5tc05zfNOcmSZKk0VrVsrKq2tvc7geuBE4B7k6yHqC53b/aICVJkiRJktSOIy4OJTkuyaMW7gMvBG4Frga2Nt22AletNkhJkiRJkiS1YzXLymaAK5Ms7Oc9VfXhJJ8GLk9yPvBV4KWrD1OSJEmSJEltOOLiUFXdCTx7ifZvAaevJihJkiRJkiSNRhsfZS9JkiRJkqQJYXFIkiRJkiSpwywOSZIkSZIkdZjFIUmSJEmSpA6zOCRJkiRJktRhFockSZIkSZI6zOKQJEmSJElShx097gCkSbNxxzXL9tl94VkjiESSJEmSpNXzzCFJkiRJkqQOszgkSZIkSZLUYRaHJEmSJEmSOsxrDmlqLHctIK8DJEkSJHk48AngYfTngldU1euTPBl4H3A88BnglVX1gyQPA94N/BTwLeBXqmr3WIKXJEmt8MwhSZKkbvk+cFpVPRs4GTgjyanAG4GLq2oTcA9wftP/fOCeqnoqcHHTT5IkTRGLQ5IkSR1SffPN5jHNVwGnAVc07buAs5v7W5ptmsdPT5IRhStJkkbAZWWSJEkdk+Qo4CbgqcDbgC8D91bVgabLHuDE5v6JwF0AVXUgyX3A44BvHrTPbcA2gJmZGXq93tDjnjkWtm8+cNg+bRx32s3PzztuLXBc2+G4tsNxbcckjavFIUmSpI6pqgeAk5OsA64EnrFUt+Z2qbOE6kENVTuBnQCzs7M1Nzc3nGAXeetlV3HRLYefvu4+Z/jHnXa9Xo82Xq+uc1zb4bi2w3FtxySNq8vKJEmSOqqq7gV6wKnAuiQLlZcNwN7m/h7gJIDm8ccA3x5tpJIkqU0WhyRJkjokyeObM4ZIcizwC8BtwMeAlzTdtgJXNfevbrZpHv9oVT3ozCFJkjS5XFYmSZLULeuBXc11hx4CXF5VH0zyBeB9Sf4A+CxwSdP/EuDPk9xB/4yhl48jaEmS1B6LQ5IkSR1SVZ8DnrNE+53AKUu0/wPw0hGEJkmSxsRlZZIkSZIkSR1mcUiSJEmSJKnDLA5JkiRJkiR1mMUhSZIkSZKkDrM4JEmSJEmS1GEWhyRJkiRJkjrM4pAkSZIkSVKHWRySJEmSJEnqMItDkiRJkiRJHWZxSJIkSZIkqcMsDkmSJEmSJHWYxSFJkiRJkqQOszgkSZIkSZLUYRaHJEmSJEmSOmzVxaEkRyX5bJIPNttPTnJDktuTvD/JQ1cfpiRJkiRJktowjDOHXgPctmj7jcDFVbUJuAc4fwjHkCRJkiRJUgtWVRxKsgE4C3hHsx3gNOCKpssu4OzVHEOSJEmSJEntOXqVz/8j4LeBRzXbjwPuraoDzfYe4MSlnphkG7ANYGZmhl6vt8pQfmR+fn6o+1tLJi237ZsPLNvnrZdd9cP7M8f+8+0Fm098zKqPNaxxGySnpY41aa/dSk1zfuY2uaY5v2nOTZIkSaN1xMWhJL8E7K+qm5LMLTQv0bWWen5V7QR2AszOztbc3NxS3Y5Ir9djmPtbSyYtt/N2XLOi/ts3H+CiWx78bbn7nLlVH2uQfQxikJyWOtakvXYrNc35mdvkmub8pjk3SZIkjdZqzhx6HvDiJGcCDwceTf9MonVJjm7OHtoA7F19mJIkSZIkSWrDEV9zqKpeV1Ubqmoj8HLgo1V1DvAx4CVNt63Ag9cISZIkSZIkaU0YxqeVHey1wG8luYP+NYguaeEYkiRJkiRJGoLVXpAagKrqAb3m/p3AKcPYryRJkiRJktrVxplDkiRJWqOSnJTkY0luS/L5JK9p2n83ydeS3Nx8nbnoOa9LckeSLyX5xfFFL0mS2jCUM4ckSZI0MQ4A26vqM0keBdyU5NrmsYur6g8Xd07yTPrXl/wJ4InAf0ny41X1wEijliRJrfHMIUmSpA6pqn1V9Znm/neB24ATD/OULcD7qur7VfV3wB14CQFJkqaKZw5JkiR1VJKNwHOAG4DnAa9Oci5wI/2zi+6hXzi6ftHT9rBEMSnJNmAbwMzMDL1eb+jxzhwL2zcfOGyfNo477ebn5x23Fjiu7XBc2+G4tmOSxtXikCRJUgcleSTwl8BvVtV3krwd+H2gmtuLgF8DssTT60ENVTuBnQCzs7M1Nzc39JjfetlVXHTL4aevu88Z/nGnXa/Xo43Xq+sc13Y4ru1wXNsxSePqsjJJkqSOSXIM/cLQZVX1AYCquruqHqiqfwL+jB8tHdsDnLTo6RuAvaOMV5IktcvikCRJUockCXAJcFtVvWlR+/pF3X4ZuLW5fzXw8iQPS/JkYBPwqVHFK0mS2ueyMkmSpG55HvBK4JYkNzdtvwO8IsnJ9JeM7QZ+A6CqPp/kcuAL9D/p7AI/qUySpOlicUiSJKlDquqTLH0doQ8d5jlvAN7QWlCSJGmsXFYmSZIkSZLUYRaHJEmSJEmSOsxlZdIiG3dcM+4QJEmSJEkaKc8ckiRJkiRJ6jCLQ5IkSZIkSR1mcUiSJEmSJKnDvOaQNCaDXN9o94VnjSASSZIkSVKXeeaQJEmSJElSh1kckiRJkiRJ6jCLQ5IkSZIkSR3mNYekCeZ1iyRJkiRJq+WZQ5IkSZIkSR1mcUiSJEmSJKnDLA5JkiRJkiR1mMUhSZIkSZKkDrM4JEmSJEmS1GEWhyRJkiRJkjrM4pAkSZIkSVKHWRySJEmSJEnqMItDkiRJkiRJHWZxSJIkSZIkqcMsDkmSJEmSJHWYxSFJkiRJkqQOszgkSZLUIUlOSvKxJLcl+XyS1zTtxye5Nsntze1jm/YkeUuSO5J8Lslzx5uBJEkatiMuDiV5eJJPJfnbZmLxe037k5Pc0Ews3p/kocMLV5IkSat0ANheVc8ATgUuSPJMYAdwXVVtAq5rtgFeBGxqvrYBbx99yJIkqU2rOXPo+8BpVfVs4GTgjCSnAm8ELm4mFvcA568+TEmSJA1DVe2rqs80978L3AacCGwBdjXddgFnN/e3AO+uvuuBdUnWjzhsSZLUoqOP9IlVVcB8s3lM81XAacCvNu27gN/F/zBJkiStOUk2As8BbgBmqmof9AtISZ7QdDsRuGvR0/Y0bfsO2tc2+mcWMTMzQ6/XG3q8M8fC9s0HDtunjeNOu/n5ecetBY5rOxzXdjiu7ZikcU2/xnOET06OAm4Cngq8Dfi/geur6qnN4ycBf11Vz1riuYsnED/1vve974jjONj8/DyPfOQjh7a/tWTScrvla/etqP/MsXD33z+4ffOJj1n1sYaxj0EtdayDX7tBjrVczMPYx7BM2vfmSpjb5Jrm/NrK7fnPf/5NVTU79B1rzUnySODjwBuq6gNJ7q2qdYsev6eqHpvkGuA/VdUnm/brgN+uqpsOte/Z2dm68cYbhx7zWy+7iotuOfz/NndfeNbQjzvter0ec3Nz4w5j6jiu7XBc2+G4tmMtjGuSgeZ2R3zmEEBVPQCcnGQdcCXwjKW6HeK5O4Gd0J9ADHPA1sIL0JZJy+28HdesqP/2zQeWnPTtPmdu1ccaxj4GtdSxDn7tBjnWcjEPYx/DMmnfmythbpNrmvOb5tzUviTHAH8JXFZVH2ia706yvjlraD2wv2nfA5y06OkbgL2ji1aSJLVtKJ9WVlX3Aj36FzVcl2Thr3snD5IkSWtIkgCXALdV1ZsWPXQ1sLW5vxW4alH7uc2nlp0K3Lew/EySJE2H1Xxa2eObM4ZIcizwC/QvaPgx4CVNt8UTC0mSJI3f84BXAqclubn5OhO4EHhBktuBFzTbAB8C7gTuAP4M+NdjiFmSJLVoNcvK1gO7musOPQS4vKo+mOQLwPuS/AHwWfr/mZI0JhsHWXrm9RkkqTOaawflEA+fvkT/Ai5oNShJkjRWq/m0ss/R/3SLg9vvBE5ZTVCSJEmSJEkajaFcc0iSJEmSJEmTyeKQJEmSJElSh1kckiRJkiRJ6jCLQ5IkSZIkSR22mk8rkyRJktYUP6VTkqSV88whSZIkSZKkDrM4JEmSJEmS1GEuK9NEGOQU8bVkqXi3bz7AeROWx2LLvQbbNx9gbjShSJIkSZKGyDOHJEmSJEmSOszikCRJkiRJUodZHJIkSZIkSeowi0OSJEmSJEkdZnFIkiRJkiSpwywOSZIkSZIkdZjFIUmSJEmSpA6zOCRJkiRJktRhFockSZIkSZI6zOKQJEmSJElSh1kckiRJkiRJ6rCjxx1AmzbuuGbZPrsvPGsEkWgtGOT7QZIkSZKkrvHMIUmSJEmSpA6zOCRJkiRJktRhFockSZI6JMk7k+xPcuuitt9N8rUkNzdfZy567HVJ7kjypSS/OJ6oJUlSmywOSZIkdculwBlLtF9cVSc3Xx8CSPJM4OXATzTP+ZMkR40sUkmSNBIWhyRJkjqkqj4BfHvA7luA91XV96vq74A7gFNaC06SJI3FVH9amSRJkgb26iTnAjcC26vqHuBE4PpFffY0bQ+SZBuwDWBmZoZerzf0AGeOhe2bD6x6P23ENsnm5+cdkxY4ru1wXNvhuLZjksbV4pAkSZLeDvw+UM3tRcCvAVmiby21g6raCewEmJ2drbm5uaEH+dbLruKiW1Y/fd19ztzqg5kivV6PNl6vrnNc2+G4tsNxbcckjavFIUmSpI6rqrsX7if5M+CDzeYe4KRFXTcAe0cYWis27rjmsI/vvvCsEUUiSdLaYHFIh+TESZKkbkiyvqr2NZu/DCx8ktnVwHuSvAl4IrAJ+NQYQpQkSS2yOCRJktQhSd4LzAEnJNkDvB6YS3Iy/SVju4HfAKiqzye5HPgCcAC4oKoeGEfckiSpPRaHJEmSOqSqXrFE8yWH6f8G4A3tRSRJksbNj7KXJEmSJEnqMItDkiRJkiRJHXbExaEkJyX5WJLbknw+yWua9uOTXJvk9ub2scMLV5IkSZIkScO0mjOHDgDbq+oZwKnABUmeCewArquqTcB1zbYkSZIkSZLWoCMuDlXVvqr6THP/u8BtwInAFmBX020XcPZqg5QkSZIkSVI7hvJpZUk2As8BbgBmqmof9AtISZ5wiOdsA7YBzMzM0Ov1hhEKAPPz8/R6PbZvPrBs32EedxQWchuF5cZvkDgGeQ0Wmzl25c+ZFEeS23JjPKyxGsZrOXPs5L2fBjXK992oTXNuMN35TXNukiRJGq1VF4eSPBL4S+A3q+o7SQZ6XlXtBHYCzM7O1tzc3GpD+aFer8fc3Bzn7bhm2b67zxnecUdhIbdRWG78Bhm7QV6DxbZvPsBFtwylZrnmHEluy43xSsf3SI8zyLG2bz7Ay0b0vTlqo3zfjdo05wbTnd805yZJkqTRWtWnlSU5hn5h6LKq+kDTfHeS9c3j64H9qwtRkiRJkiRJbVnNp5UFuAS4raretOihq4Gtzf2twFVHHp4kSZIkSZLatJr1O88DXgnckuTmpu13gAuBy5OcD3wVeOnqQpQkSZIkSVJbjrg4VFWfBA51gaHTj3S/kiRJkiRJGp1VXXNIkiRJkiRJk83ikCRJkiRJUodZHJIkSZIkSeqw1VyQWpJWbOOOaw77+O4LzxrJcYZ5LEmSJEmaZJ45JEmSJEmS1GEWhyRJkiRJkjrM4pAkSZIkSVKHec0haQ0b5Lo5XTTKcdm44xq2bz7AeYc4ptctkiRJkjTpPHNIkiRJkiSpwywOSZIkSZIkdZjFIUmSJEmSpA6zOCRJkiRJktRhFockSZI6JMk7k+xPcuuituOTXJvk9ub2sU17krwlyR1JPpfkueOLXJIktcXikCRJUrdcCpxxUNsO4Lqq2gRc12wDvAjY1HxtA94+ohglSdIIWRySJEnqkKr6BPDtg5q3ALua+7uAsxe1v7v6rgfWJVk/mkglSdKoHD3uACRJkjR2M1W1D6Cq9iV5QtN+InDXon57mrZ9B+8gyTb6ZxcxMzNDr9cbfpDHwvbNB4a+34O1EftaNj8/37mcR8FxbYfj2g7HtR2TNK4WhyRJknQoWaKtlupYVTuBnQCzs7M1Nzc39GDeetlVXHRL+9PX3efMtX6MtaTX69HG69V1jms7HNd2OK7tmKRxdVmZJEmS7l5YLtbc7m/a9wAnLeq3Adg74tgkSVLLLA5JkiTpamBrc38rcNWi9nObTy07FbhvYfmZJEmaHi4rkyRJ6pAk7wXmgBOS7AFeD1wIXJ7kfOCrwEub7h8CzgTuAL4HvGrkAUuSpNZZHBrAxh3XLNtn94VnjSASSZKk1amqVxziodOX6FvABe1GJEmSxs1lZZIkSZIkSR1mcUiSJEmSJKnDLA5JkiRJkiR1mNcc0hEb5FpMkiRJkiRpbfPMIUmSJEmSpA6zOCRJkiRJktRhFockSZIkSZI6zOKQJEmSJElSh1kckiRJkiRJ6jCLQ5IkSZIkSR1mcUiSJEmSJKnDjh53AJLGb+OOa8YdgiRJkiRpTFZ15lCSdybZn+TWRW3HJ7k2ye3N7WNXH6YkSZIkSZLasNplZZcCZxzUtgO4rqo2Adc125IkSZIkSVqDVlUcqqpPAN8+qHkLsKu5vws4ezXHkCRJkiRJUnvauCD1TFXtA2hun9DCMSRJkiRJkjQEY7sgdZJtwDaAmZkZer3e0PY9Pz9Pr9dj++YDy/Yd5LjD2s8wLOQ2CoPkPWwzx47nuKMwzblBP79hvJ+G9Z4cxKDHOtxrN6r3Y1tG+TNlHKY5v2nOTZIkSaPVRnHo7iTrq2pfkvXA/qU6VdVOYCfA7Oxszc3NDS2AXq/H3Nwc5w3wCUy7z1n+uMPazzAs5DYKg+Q9bNs3H+CiW6bzQ/SmOTfo5/eyAb43l/u+GtZ7chCDHutwr92o3vttGeXPlHGY5vymOTdJkiSNVhvLyq4Gtjb3twJXtXAMSZIkSZIkDcFqP8r+vcDfAE9LsifJ+cCFwAuS3A68oNmWJEmSJEnSGrSqNS5V9YpDPHT6avYrqbs2jmE5Y9sGyWn3hWeNIBJJkiRJerA2lpVJkiRJkiRpQlgckiRJkiRJ6rDp/egkSZIkrUiS3cB3gQeAA1U1m+R44P3ARmA38LKqumdcMUqSpOGzOLSGDHJdkkvPOG4EkUhHZhqvFyRJHfT8qvrmou0dwHVVdWGSHc32a8cTmiRJaoPLyiRJknQ4W4Bdzf1dwNljjEWSJLXAM4ckSZK0oICPJCngT6tqJzBTVfsAqmpfkics9cQk24BtADMzM/R6vaEHN3MsbN98YOj7PdhbL7tq2T6bT3xM63GMyvz8fCuvV9c5ru1wXNvhuLZjksbV4pAkSZIWPK+q9jYFoGuTfHHQJzaFpJ0As7OzNTc3N/Tg3nrZVVx0y9qYvu4+Z27cIQxNr9ejjder6xzXdjiu7XBc2zFJ47o2frt2wFq7Fstai0fqukHek7svPGsk+xlWLJImT1XtbW73J7kSOAW4O8n65qyh9cD+sQYpSZKGzmsOSZIkiSTHJXnUwn3ghcCtwNXA1qbbVmD5NVeSJGmieOaQJEmSAGaAK5NAf474nqr6cJJPA5cnOR/4KvDSMcYoSZJaYHFIkiRJVNWdwLOXaP8WcProI5IkSaPisjJJkiRJkqQO6/yZQ5N2YeZbvnYf5y0TsxeKlQYzjPf/pP0MkSSNjhf4lyRNCs8ckiRJkiRJ6jCLQ5IkSZIkSR1mcUiSJEmSJKnDLA5JkiRJkiR1mMUhSZIkSZKkDuv8p5VJkiRJa9lyn3rmJ55JklbLM4ckSZIkSZI6zDOHJKljFv8HevvmA5y3zH+kl9vHofifbEmSJGkyeOaQJEmSJElSh1kckiRJkiRJ6jCLQ5IkSZIkSR3mNYeGZJDrb4zKWopFUnetpesSraVYJEmSpLXG4pAkSZI0JsP4p54FcEnSarmsTJIkSZIkqcM8c0iSJElaIZfxS5KmicUhSZoQ/iGytMONy/bNBzjPcZMkSZIOy2VlkiRJkiRJHWZxSJIkSZIkqcNcViZJkiRpTfET2CRptCwOSZIkSRrate0GKdp4HT1JWlssDkmSxmbS/jM8qngHOc6lZxy36uNIknQok/Y7WtLqtFIcSnIG8GbgKOAdVXVhG8eRJElS+5zbSas3rGKLRRtJbRh6cSjJUcDbgBcAe4BPJ7m6qr4w7GNJkiSpXc7tJElanUko6rZx5tApwB1VdSdAkvcBWwAnEJIkSZPHud0UONwfJts3H+C8Kb0G0CT8QbbYKK/F5HWfJC2WqhruDpOXAGdU1a83268EfqaqXn1Qv23AtmbzacCXhhjGCcA3h7i/tWSac4Ppzm+ac4Ppzs/cJtc059dWbk+qqse3sF9NqDUyt1swze/pcXJc2+G4tsNxbYfj2o61MK4Dze3aOHMoS7Q9qAJVVTuBnS0cnyQ3VtVsG/set2nODaY7v2nODaY7P3ObXNOc3zTnpjVn7HO7Hwbi930rHNd2OK7tcFzb4bi2Y5LG9SEt7HMPcNKi7Q3A3haOI0kVSZG2AAAgAElEQVSSpPY5t5Mkacq1URz6NLApyZOTPBR4OXB1C8eRJElS+5zbSZI05Ya+rKyqDiR5NfD/0v+403dW1eeHfZxltHpK85hNc24w3flNc24w3fmZ2+Sa5vymOTetIWtkbrfA7/t2OK7tcFzb4bi2w3Ftx8SM69AvSC1JkiRJkqTJ0cayMkmSJEmSJE0Ii0OSJEmSJEkdNlXFoSRnJPlSkjuS7Bh3PKuV5J1J9ie5dVHb8UmuTXJ7c/vYccZ4pJKclORjSW5L8vkkr2napyW/hyf5VJK/bfL7vab9yUluaPJ7f3Nhz4mU5Kgkn03ywWZ7KnJLsjvJLUluTnJj0zYV35cASdYluSLJF5v337+YhvySPK15zRa+vpPkN6chtwVJ/vfm58mtSd7b/JyZivedtJxpm+ON2krmlOl7SzPWn0vy3PFFvnatdC7ruA5upfPoJA9rtu9oHt84zvjXskHn747pyqzk74e1/LNgaopDSY4C3ga8CHgm8IokzxxvVKt2KXDGQW07gOuqahNwXbM9iQ4A26vqGcCpwAXN6zUt+X0fOK2qng2cDJyR5FTgjcDFTX73AOePMcbVeg1w26Ltacrt+VV1clXNNtvT8n0J8Gbgw1X1dODZ9F/Dic+vqr7UvGYnAz8FfA+4kinIDSDJicC/BWar6ln0Lwr8cqbrfSctaUrneKN2KYPPKV8EbGq+tgFvH1GMk2alc1nHdXArnUefD9xTVU8FLm76aWmDzt8d05Ub9O+HNfuzYGqKQ8ApwB1VdWdV/QB4H7BlzDGtSlV9Avj2Qc1bgF3N/V3A2SMNakiqal9Vfaa5/136P6ROZHryq6qabzaPab4KOA24ommf2PySbADOAt7RbIcpye0QpuL7MsmjgZ8HLgGoqh9U1b1MSX6LnA58uaq+wnTldjRwbJKjgUcA+5ju9520YOrmeKO2wjnlFuDdzVzmemBdkvWjiXRyHMFc1nEd0BHMoxeP+RXA6c3cVIuscP7umK7exP0smKbi0InAXYu29zRt02amqvZB/5cS8IQxx7NqzWmKzwFuYIrya07bvBnYD1wLfBm4t6oONF0m+Xv0j4DfBv6p2X4c05NbAR9JclOSbU3btHxfPgX4BvCu5pTidyQ5junJb8HLgfc296cit6r6GvCHwFfpF4XuA25iet530uF0ZY43aof6+eh4r9CAc1nHdQVWOI/+4dg2j99Hf26qf24l83fHdGVW8vfDmv1ZME3FoaUqmTXyKLQiSR4J/CXwm1X1nXHHM0xV9UCzxGUD/f96PmOpbqONavWS/BKwv6puWty8RNeJy63xvKp6Lv1TPi9I8vPjDmiIjgaeC7y9qp4D3M+ELrM6lGat/IuB/zzuWIapWae+BXgy8ETgOPrfoweb1PeddDjT9DtmEjjeK7CCuazjugIrnEc7tss4gvm7Y7oyK/n7Yc2O7TQVh/YAJy3a3gDsHVMsbbp74bSz5nb/mOM5YkmOof/L9LKq+kDTPDX5LWiW7fTor0df1ywJgcn9Hn0e8OIku+mf2n8a/f9ETENuVNXe5nY//WvWnML0fF/uAfZU1Q3N9hX0i0XTkh/0fyl/pqrubranJbdfAP6uqr5RVf8IfAD4l0zJ+05aRlfmeKN2qJ+PjveAVjiXdVyPwIDz6B+ObfP4Y3jwMsquW+n83TFdgRX+/bBmfxZMU3Ho08Cm5orrD6W/rODqMcfUhquBrc39rcBVY4zliDVrVi8BbquqNy16aFrye3ySdc39Y+n/YXcb8DHgJU23icyvql5XVRuqaiP999lHq+ocpiC3JMcledTCfeCFwK1MyfdlVX0duCvJ05qm04EvMCX5NV7Bj5aUwfTk9lXg1CSPaH5+Lrx2E/++kwbQlTneqB3q5+PVwLnNJ+qcCty3sDRCP3IEc1nHdUBHMI9ePOYvoT83XRNnYqwVRzB/d0wHdAR/P6zZnwWZptc4yZn0K6BHAe+sqjeMOaRVSfJeYA44AbgbeD3wV8DlwP9A/4+Fl1bVxFVxk/ws8F+BW/jRutffob9Wexry+0n6Fx47in4R9vKq+o9JnkK/Wn888Fngf66q748v0tVJMgf8u6r6pWnIrcnhymbzaOA9VfWGJI9jCr4vAZKcTP9ChA8F7gReRfM9yoTnl+QR9NdwP6Wq7mvapum1+z3gV+h/Qs5ngV+nv0Z9ot930iCmbY43aiuZUzZFjz+m/+lm3wNeVVU3jiPutWylc1nHdXArnUcneTjw5/Sv+/Rt4OVVded4ol/7Bpm/O6aDW+nfD2v5Z8FUFYckSZIkSZK0MtO0rEySJEmSJEkrZHFIkiRJkiSpwywOSZIkSZIkdZjFIUmSJEmSpA6zOCRJkiRJktRhFockSZIkSZI6zOKQJEmSJElSh1kckiRJkiRJ6jCLQ5IkSZIkSR1mcUiSJEmSJKnDLA5JkiRJkiR1mMUhSZIkSZKkDrM4JEmSJEmS1GEWhyRJkiRJkjrM4pAkSZIkSVKHWRySJEmSJEnqMItDkiRJkiRJHWZxSJIkSZIkqcMsDkmSJEmSJHWYxSFJkiRJkqQOszgkSZIkSZLUYRaHJEmSJEmSOszikCRJkiRJUodZHJIkSZIkSeowi0OSJEmSJEkdZnFIkiRJkiSpwywOSZIkSZIkdZjFIUmSJEmSpA6zOCRJkiRJktRhFockSZIkSZI6zOKQJEmSJElSh1kckiRJkiRJ6jCLQ5IkSZIkSR1mcUiSJEmSJKnDLA5JkiRJkiR1mMUhSZIkSZKkDrM4JGmkkvSS/Pq445AkSVJfkrkke8Ydh6TxsTgk6YeS7E7y90nmk9yd5F1JHjnuuCRJkjQ453SSVsrikKSD/auqeiTwXOCngf9zJU9OcnQrUUmSJGklVjWnk9QtFockLamqvgb8NfCsJK9KcluS7ya5M8lvLPRbOA05yWuTfB14V9O+JcnNSb6T5MtJzli0+ycl+W/N/j6S5ITRZidJktQNB83pjm/OItqb5J4kf7XUc5LsaOZv303yhSS/vOixpyb5eJL7knwzyfub9iS5OMn+5rHPJXnWaLKUtFr+h1/SkpKcBJwJfADYD/wScCfw88BfJ/l0VX2m6f5jwPHAk4CHJDkFeDfwEuA6YD3wqEW7/1XgRcBd9Ccr/w7Y0XZOkiRJXXPQnO7PgXngJ5rbf3mIp30Z+Dng68BLgb9I8tSq2gf8PvAR4PnAQ4HZ5jkvpD9P/HHgPuDpwL0tpCSpBRaHJB3sr5IcoP9L/Rrg/6qqv1/0+MeTfIT+hGGhOPRPwOur6vsASc4H3llV1zaPf+2gY7yrqv570/dy4MXtpCJJktRZB8/p/oT+nOxxVXVP0+fjSz2xqv7zos33J3kdcApwFfCP9P8h+MSq2gN8sun3j/T/Gfh04FNVdduQ85HUIpeVSTrY2VW1rqqeVFX/uqr+PsmLklyf5NtJ7qX/36fFS8G+UVX/sGj7JPr/cTqUry+6/z3ACyRKkiQN1z+b09Gfn317UWHokJKc21we4N5m7vcsfjT3+20gwKeSfD7JrwFU1UeBPwbeBtydZGeSR7eRmKThszgk6bCSPAz4S+APgZmqWgd8iP6kYEEd9LS7gP9xNBFKkiRpAHcBxydZd7hOSZ4E/BnwavpnGa0DbqWZ+1XV16vqf6mqJwK/AfxJkqc2j72lqn6K/rK1Hwf+fWvZSBoqi0OSlvNQ4GHAN4ADSV5Ef0354VwCvCrJ6UkekuTEJE9vO1BJkiQtrble0F/TL+Y8NskxSX5+ia7H0f/H3zcAkryK/plDNNsvTbKh2byn6ftAkp9O8jNJjgHuB/4BeKC9jCQNk8UhSYdVVd8F/i1wOf0JwK8CVy/znE8BrwIupr/O/eP016ZLkiRpfF5J/9pAX6T/gSO/eXCHqvoCcBHwN8DdwGbgvy3q8tPADUnm6c8JX1NVfwc8mv4ZR/cAXwG+Rf/Mc0kTIFUHrwaRJEmSJElSV3jmkCRJkiRJUodZHJIkSZIkSeowi0OSJEmSJEkdZnFIkiRJkiSpw45erkOSpwHvX9T0FOA/AO9u2jcCu4GXVdU9SQK8GTgT+B5wXlV95nDHOOGEE2rjxo1HEP7h3X///Rx33HFD3++0cZyW5xgNxnFanmM0GMdpeW2N0U033fTNqnr80HesTnFut3LmNpnMbXJNc37mNpnGPbdbtjhUVV8CTgZIchTwNeBKYAdwXVVdmGRHs/1a4EXApubrZ4C3N7eHtHHjRm688cblQlmxXq/H3Nzc0Pc7bRyn5TlGg3GclucYDcZxWl5bY5TkK0PfqTrHud3KmdtkMrfJNc35mdtkGvfcbqXLyk4HvlxVXwG2ALua9l3A2c39LcC7q+96YF2S9Ss8jiRJkiRJkkZgpcWhlwPvbe7PVNU+gOb2CU37icBdi56zp2mTJEnSGpBkd5Jbktyc5Mam7fgk1ya5vbl9bNOeJG9JckeSzyV57nijlyRJw7bssrIFSR4KvBh43XJdl2irJfa3DdgGMDMzQ6/XGzSUgc3Pz7ey32njOC3PMRqM47Q8x2gwjtPyHCMNwfOr6puLtod2yQBJkjRZBi4O0Z8YfKaq7m62706yvqr2NcvG9jfte4CTFj1vA7D34J1V1U5gJ8Ds7Gy1sbZumtcjDpPjtDzHaDCO0/Ico8E4TstzjNSCLcBcc38X0KNfHPrhJQOA65OsW5gDjiVKSZI0dCtZVvYKfrSkDOBqYGtzfytw1aL2c5tTkE8F7nPyIEmStKYU8JEkNzVnc4OXDJAkqbMGOnMoySOAFwC/saj5QuDyJOcDXwVe2rR/iP7H2N9B/6PsXzW0aCVJkjQMz6uqvUmeAFyb5IuH6eslA0bA3CaTuU2uac7P3CbTuHMbqDhUVd8DHndQ27fof3rZwX0LuGAo0UmSJGnoqmpvc7s/yZXAKXjJgLEyt8lkbpNrmvMzt8k07txW+mllkiRJmmBJjkvyqIX7wAuBW/GSAZIkddZKLkgtSZKkyTcDXJkE+nPB91TVh5N8Gi8ZIElSJ011ceiWr93HeTuuOWyf3ReeNaJoJEmSxq+q7gSevUS7lwyQJKkFG5epSwBcesZxI4jk0FxWJkmSJEmS1GEWhyRJkiRJkjrM4pAkSZIkSVKHWRySJEmSJEnqMItDkiRJkiRJHWZxSJIkSZIkqcMsDkmSJEmSJHWYxSFJkiRJkqQOszgkSZIkSZLUYRaHJEmSJEmSOszikCRJkiRJUodZHJIkSZIkSeowi0OSJEmSJEkdZnFIkiRJkiSpwywOSZIkSZIkdZjFIUmSJEmSpA6zOCRJkiRJktRhFockSZIkSZI6zOKQJEmSJElSh1kckiRJkiRJ6rCBikNJ1iW5IskXk9yW5F8kOT7JtUlub24f2/RNkrckuSPJ55I8t90UJEmSJEmSdKQGPXPozcCHq+rpwLOB24AdwHVVtQm4rtkGeBGwqfnaBrx9qBFLkiRJkiRpaJYtDiV5NPDzwCUAVfWDqroX2ALsarrtAs5u7m8B3l191wPrkqwfeuSSJEmSJElataMH6PMU4BvAu5I8G7gJeA0wU1X7AKpqX5InNP1PBO5a9Pw9Tdu+xTtNso3+mUXMzMzQ6/VWkcbSZo6F7ZsPHLZPG8edNPPz847DMhyjwThOy3OMBuM4Lc8xkiRJ0rAMUhw6Gngu8G+q6oYkb+ZHS8iWkiXa6kENVTuBnQCzs7M1Nzc3QCgr89bLruKiWw6f4u5zhn/cSdPr9Whj/KeJYzQYx2l5jtFgHKflOUaSJEkalkGuObQH2FNVNzTbV9AvFt29sFysud2/qP9Ji56/Adg7nHAlSZIkSZI0TMsWh6rq68BdSZ7WNJ0OfAG4GtjatG0FrmruXw2c23xq2anAfQvLzyRJkiRJkrS2DLKsDODfAJcleShwJ/Aq+oWly5OcD3wVeGnT90PAmcAdwPeavpIkSZIkSVqDBioOVdXNwOwSD52+RN8CLlhlXJIkSZIkSRqBQa45JEmSJEmSpCllcUiSJEmSJKnDLA5JkiRJkiR1mMUhSZIkSZKkDrM4JEmSJEmS1GEWhyRJkiRJkjrM4pAkSVIHJTkqyWeTfLDZfnKSG5LcnuT9SR7atD+s2b6jeXzjOOOWJEnDZ3FIkiSpm14D3LZo+43AxVW1CbgHOL9pPx+4p6qeClzc9JMkSVPE4pAkSVLHJNkAnAW8o9kOcBpwRdNlF3B2c39Ls03z+OlNf0mSNCUsDkmSJHXPHwG/DfxTs/044N6qOtBs7wFObO6fCNwF0Dx+X9NfkiRNiaPHHYAkSZJGJ8kvAfur6qYkcwvNS3StAR5bvN9twDaAmZkZer3e6oM9yPz8fCv7XQvMbTKZ2+Sa5vzMbe3ZvvnAsn3GnZvFIUmSpG55HvDiJGcCDwceTf9MonVJjm7ODtoA7G367wFOAvYkORp4DPDtg3daVTuBnQCzs7M1Nzc39MB7vR5t7HctMLfJZG6Ta5rzM7e157wd1yzb59Izjhtrbi4rkyRJ6pCqel1VbaiqjcDLgY9W1TnAx4CXNN22Alc1969utmke/2hVPejMIUmSNLksDkmSJAngtcBvJbmD/jWFLmnaLwEe17T/FrBjTPFJkqSWuKxMkiSpo6qqB/Sa+3cCpyzR5x+Al440MEmSNFKeOSRJkiRJktRhFockSZIkSZI6zOKQJEmSJElSh1kckiRJkiRJ6jCLQ5IkSZIkSR1mcUiSJEmSJKnDLA5JkiRJkiR1mMUhSZIkSZKkDhuoOJRkd5Jbktyc5Mam7fgk1ya5vbl9bNOeJG9JckeSzyV5bpsJSJIkSZIk6cit5Myh51fVyVU122zvAK6rqk3Adc02/z979x9k2VneB/77BCFQZIzEr45WI3vIMuvARsuPnRAleL0d5MJCOB6lFmVxFBgRpaY2UVx40cYW1CZeb5xaUSkZG62XZNZyGDnCoMUmowXioBX0upysZH4YJEA4GitjNNYEGeuHGbBxDX72j3vGNKPW9B3Nvd19+3w+VbfuOe957znP+945Paefft9zkrw2ya7htS/Ju2cVLAAAAACzdSbTyvYkOTAsH0hyxaryW3ririTnVdUFZ3AcAAAAAObkrCnrdZKPVlUn+RfdvT/JUncfTZLuPlpVLxjqXpjkwVWfPTKUHV29w6ral8nIoiwtLWVlZeUpN+LJLJ2TXHfx8VPWmcdxF82xY8f0wzr00XT00/r00XT00/r0EQAAszJtcuhV3f3QkAC6o6q+eIq6tUZZP6FgkmDanyS7d+/u5eXlKUOZ3k23HsyN9566iYevmv1xF83Kykrm0f/biT6ajn5anz6ajn5anz4CAGBWpppW1t0PDe8PJ/lgklcm+fKJ6WLD+8ND9SNJLlr18R1JHppVwAAAAADMzrrJoao6t6qedWI5yWuSfC7J7Un2DtX2Jjk4LN+e5E3DU8suSfL4ielnAAAAAGwt00wrW0rywao6Uf+93f2rVfWJJLdV1TVJvpTkyqH+R5JcnuRQkq8nefPMowYAAABgJtZNDnX3A0leukb57ye5dI3yTnLtTKIDAAAAYK7O5FH2AAAAACw4ySEAAACAEZMcAgAAABgxySEAAACAEZMcAgAAABgxySEAAACAEZMcAgAAABgxySEAAACAEZMcAgAAABgxySEAAACAEZMcAgAAABgxySEAAACAEZMcAgAAABgxySEAAACAEZMcAgAAABgxySEAAACAEZMcAgAAABgxySEAAACAEZMcAgAYkap6ZlX9RlV9tqo+X1U/OZS/sKrurqr7q+r9VXX2UP6MYf3QsH3nZsYPAMye5BAAwLh8I8mru/ulSV6W5LKquiTJO5K8s7t3JXk0yTVD/WuSPNrdL0ryzqEeALCNSA4BAIxITxwbVp8+vDrJq5N8YCg/kOSKYXnPsJ5h+6VVVRsULgCwASSHAABGpqqeVlWfSfJwkjuS/HaSx7r7+FDlSJILh+ULkzyYJMP2x5M8d2MjBgDm6azNDgAAgI3V3d9M8rKqOi/JB5O8eK1qw/tao4T65IKq2pdkX5IsLS1lZWVlNsGucuzYsbnsdyvQtsWkbYtrO7dP27ae6y4+vm6dzW7b1Mmhqnpakk8m+d3u/sGqemGS9yV5TpJPJ3ljd/9xVT0jyS1J/uskv5/kv+/uwzOPHACAM9Ldj1XVSpJLkpxXVWcNo4N2JHloqHYkyUVJjlTVWUmeneSRNfa1P8n+JNm9e3cvLy/PPN6VlZXMY79bgbYtJm1bXNu5fdq29Vx9/YfXrfOey87d1LadzrSytyS5b9W6mxYCACyYqnr+MGIoVXVOku/P5Brv40leP1Tbm+TgsHz7sJ5h+8e6+wkjhwCAxTVVcqiqdiR5XZKfH9YrbloIALCILkjy8aq6J8knktzR3R9K8uNJ3lpVhzK5p9DNQ/2bkzx3KH9rkus3IWYAYI6mnVb2M0l+LMmzhvXnZsqbFlbViZsWfmX1DjdiXvrSOevP7VvE+YqzttlzGxeBPpqOflqfPpqOflqfPuKp6u57krx8jfIHkrxyjfI/SnLlBoQGAGySdZNDVfWDSR7u7k9V1fKJ4jWqntZNCzdiXvpNtx7MjfeeuomHr5r9cRfNos7b3Ej6aDr6aX36aDr6aX36CACAWZlm5NCrkvxQVV2e5JlJvjOTkURndNNCAAAAADbfuvcc6u63dfeO7t6Z5A2Z3ITwqrhpIQAAAMDCO52nlZ3MTQsBAAAAFty0N6ROknT3SpKVYdlNCwEAAAAW3JmMHAIAAABgwUkOAQAAAIyY5BAAAADAiEkOAQAAAIyY5BAAAADAiEkOAQAAAIyY5BAAAADAiEkOAQAAAIyY5BAAAADAiEkOAQAAAIyY5BAAAADAiEkOAQAAAIyY5BAAAADAiEkOAQAAAIyY5BAAAADAiEkOAQAAAIyY5BAAAADAiEkOAQAAAIyY5BAAAADAiEkOAQAAAIyY5BAAAADAiEkOAQAAAIyY5BAAAADAiK2bHKqqZ1bVb1TVZ6vq81X1k0P5C6vq7qq6v6reX1VnD+XPGNYPDdt3zrcJAAAAADxV04wc+kaSV3f3S5O8LMllVXVJknckeWd370ryaJJrhvrXJHm0u1+U5J1DPQAAAAC2oHWTQz1xbFh9+vDqJK9O8oGh/ECSK4blPcN6hu2XVlXNLGIAAAAAZuasaSpV1dOSfCrJi5L8XJLfTvJYdx8fqhxJcuGwfGGSB5Oku49X1eNJnpvkKyftc1+SfUmytLSUlZWVM2rIWpbOSa67+Pgp68zjuIvm2LFj+mEd+mg6+ml9+mg6+ml9+ogxuvd3H8/V13/4lHUO3/C6DYoGALaPqZJD3f3NJC+rqvOSfDDJi9eqNryvNUqon1DQvT/J/iTZvXt3Ly8vTxPKabnp1oO58d5TN/HwVbM/7qJZWVnJPPp/O9FH09FP69NH09FP69NHAADMymk9ray7H0uykuSSJOdV1YnMy44kDw3LR5JclCTD9mcneWQWwQIAAAAwW9M8rez5w4ihVNU5Sb4/yX1JPp7k9UO1vUkODsu3D+sZtn+su58wcggAAACAzTfNyKELkny8qu5J8okkd3T3h5L8eJK3VtWhTO4pdPNQ/+Ykzx3K35rk+tmHDQDAU1FVF1XVx6vqvqr6fFW9ZSh/TlXdUVX3D+/nD+VVVe+qqkNVdU9VvWJzWwAAzNq69xzq7nuSvHyN8geSvHKN8j9KcuVMogMAYNaOJ7muuz9dVc9K8qmquiPJ1Unu7O4bqur6TP7A9+NJXptk1/D6y0nePbwDANvEad1zCACAxdbdR7v708PyVzO5XcCFSfYkOTBUO5DkimF5T5JbeuKuTO47ecEGhw0AzNFUTysDAGD7qaqdmYwQvzvJUncfTSYJpKp6wVDtwiQPrvrYkaHs6En72pdkX5IsLS1lZWVl5vEunZNcd/HxU9aZx3E3wrFjxxY29vVo22Lazm1Ltnf7tG3rWe//rmTz2yY5BAAwQlX1HUl+OcmPdvcfVNWTVl2j7AkPG+nu/Un2J8nu3bt7eXl5RpF+y023HsyN95768vXwVbM/7kZYWVnJPPpsK9C2xbSd25Zs7/Zp29Zz9fUfXrfOey47d1PbZloZAMDIVNXTM0kM3drdvzIUf/nEdLHh/eGh/EiSi1Z9fEeShzYqVgBg/iSHAABGpCZDhG5Ocl93//SqTbcn2Tss701ycFX5m4anll2S5PET088AgO3BtDIAgHF5VZI3Jrm3qj4zlL09yQ1Jbquqa5J8Kd96+uxHklye5FCSryd588aGCwDMm+QQAMCIdPevZ+37CCXJpWvU7yTXzjUoAGBTmVYGAAAAMGKSQwAAAAAjZloZAAAA3+be33183cdvH77hdRsUDTBvRg4BAAAAjJjkEAAAAMCISQ4BAAAAjJjkEAAAAMCISQ4BAAAAjJjkEAAAAMCISQ4BAAAAjJjkEAAAAMCISQ4BAAAAjJjkEAAAAMCISQ4BAAAAjJjkEAAAAMCISQ4BAAAAjNi6yaGquqiqPl5V91XV56vqLUP5c6rqjqq6f3g/fyivqnpXVR2qqnuq6hXzbgQAAAAAT800I4eOJ7muu1+c5JIk11bVS5Jcn+TO7t6V5M5hPUlem2TX8NqX5N0zjxoAAACAmVg3OdTdR7v708PyV5Pcl+TCJHuSHBiqHUhyxbC8J8ktPXFXkvOq6oKZRw4AAADAGTvrdCpX1c4kL09yd5Kl7j6aTBJIVfWCodqFSR5c9bEjQ9nRk/a1L5ORRVlaWsrKysrpR7+OpXOS6y4+fso68zjuojl27Jh+WIc+mo5+Wp8+mo5+Wp8+AgBgVqZODlXVdyT55SQ/2t1/UFVPWnWNsn5CQff+JPuTZPfu3b28vDxtKFO76daDufHeUzfx8FWzP+6iWVlZyTz6fzvRR9PRT+vTR9PRT+vTRwAAzMpUTyurqqdnkhi6tbt/ZSj+8onpYsP7w5Zd+6EAACAASURBVEP5kSQXrfr4jiQPzSZcAAAAAGZpmqeVVZKbk9zX3T+9atPtSfYOy3uTHFxV/qbhqWWXJHn8xPQzAAAAALaWaaaVvSrJG5PcW1WfGcrenuSGJLdV1TVJvpTkymHbR5JcnuRQkq8nefNMIwYAAABgZtZNDnX3r2ft+wglyaVr1O8k155hXAAAAABsgKnuOQQAAADA9iQ5BAAAADBikkMAAAAAIyY5BAAAADBikkMAAAAAIyY5BAAAADBikkMAAAAAIyY5BAAAADBikkMAAAAAIyY5BAAAADBikkMAACNSVb9QVQ9X1edWlT2nqu6oqvuH9/OH8qqqd1XVoaq6p6pesXmRAwDzIjkEADAu70ly2Ull1ye5s7t3JblzWE+S1ybZNbz2JXn3BsUIAGwgySEAgBHp7l9L8shJxXuSHBiWDyS5YlX5LT1xV5LzquqCjYkUANgoZ212AAAAbLql7j6aJN19tKpeMJRfmOTBVfWODGVHT95BVe3LZHRRlpaWsrKyMvsgz0muu/j4KevM47gb4dixYwsb+3q0bTFt5/Mt2d7fnbZtPeudS8nmt01yCACAJ1NrlPVaFbt7f5L9SbJ79+5eXl6eeTA33XowN9576svXw1fN/rgbYWVlJfPos61A2xbTdj7fku393Wnb1nP19R9et857Ljt3U9tmWhkAAF8+MV1seH94KD+S5KJV9XYkeWiDYwMA5kxyCACA25PsHZb3Jjm4qvxNw1PLLkny+InpZwDA9mFaGQDAiFTVLyVZTvK8qjqS5CeS3JDktqq6JsmXklw5VP9IksuTHEry9SRv3vCAAYC5kxwCABiR7v7hJ9l06Rp1O8m1840IANhsppUBAAAAjJiRQyO1c9Xd0q+7+Piad08/fMPrNjIkAAAAYBMYOQQAAAAwYusmh6rqF6rq4ar63Kqy51TVHVV1//B+/lBeVfWuqjpUVfdU1SvmGTwAAAAAZ2aakUPvSXLZSWXXJ7mzu3cluXNYT5LXJtk1vPYlefdswgQAAABgHtZNDnX3ryV55KTiPUkODMsHklyxqvyWnrgryXlVdcGsggUAAABgtp7qPYeWuvtokgzvLxjKL0zy4Kp6R4YyAAAAALagWT+trNYo6zUrVu3LZOpZlpaWsrKyMuNQkqVzJk/iOpV5HHcRrO6XJ+unsfbNWo4dO6Y/pqCf1qePpqOf1qePAACYlaeaHPpyVV3Q3UeHaWMPD+VHkly0qt6OJA+ttYPu3p9kf5Ls3r27l5eXn2IoT+6mWw/mxntP3cTDV83+uIvg6pMeZb9WP421b9aysrKSefwb3W700/r00XT00/r0EQAAs/JUp5XdnmTvsLw3ycFV5W8anlp2SZLHT0w/AwAAAGDrWXfkUFX9UpLlJM+rqiNJfiLJDUluq6prknwpyZVD9Y8kuTzJoSRfT/LmOcQMAAAAwIysmxzq7h9+kk2XrlG3k1x7pkEBAAAAsDGe6rQyAAAAALaBWT+tDBbazlU36l7tuouP/+lNvA/f8LqNDAkAAADmysghAAAAgBGTHAIAAAAYMckhAAAAgBGTHAIAAAAYMckhAAAAgBGTHAIAAAAYMckhAAAAgBGTHAIAAAAYMckhAAAAgBGTHAIAAAAYMckhAAAAgBGTHAIAAAAYMckhAAAAgBGTHAIAAAAYMckhAAAAgBGTHAIAAAAYMckhAAAAgBGTHAIAAAAYMckhAAAAgBGTHAIAAAAYMckhAAAAgBGbS3Koqi6rqt+qqkNVdf08jgEAwMZwbQcA29vMk0NV9bQkP5fktUlekuSHq+olsz4OAADz59oOALa/s+awz1cmOdTdDyRJVb0vyZ4kX5jDsYAt5N7ffTxXX//hU9Y5fMPrNiiarWPnqj657uLja/bRGPsl+fa+WW11P421b2ALcW0HANtcdfdsd1j1+iSXdfffHdbfmOQvd/c/OKneviT7htXvSfJbMw1k4nlJvjKH/W43+ml9+mg6+ml9+mg6+ml98+qj7+7u589hvywo13YbRtsWk7Ytru3cPm1bTJt6bTePkUO1RtkTMlDdvT/J/jkc/1uBVH2yu3fP8xjbgX5anz6ajn5anz6ajn5anz5iA7m22wDatpi0bXFt5/Zp22La7LbN44bUR5JctGp9R5KH5nAcAADmz7UdAGxz80gOfSLJrqp6YVWdneQNSW6fw3EAAJg/13YAsM3NfFpZdx+vqn+Q5N8meVqSX+juz8/6OFOa69DmbUQ/rU8fTUc/rU8fTUc/rU8fsSFc220YbVtM2ra4tnP7tG0xbWrbZn5DagAAAAAWxzymlQEAAACwICSHAAAAAEZs2yaHquqyqvqtqjpUVddvdjxbUVX9QlU9XFWf2+xYtqqquqiqPl5V91XV56vqLZsd01ZTVc+sqt+oqs8OffSTmx3TVlZVT6uq36yqD212LFtRVR2uqnur6jNV9cnNjmerqqrzquoDVfXF4efTX9nsmOBMrHdNUhPvGq7r7qmqV6zatreq7h9eezcu6ulM0barhjbdU1X/vqpeumrblv6ZOEXblqvq8SH+z1TVP161bUtfq0/Rtn+4ql2fq6pvVtVzhm1b/Xtb9/p2Uc+5Kdu2kOfclG1byHNuyrYt8jm37u9LVfWMqnr/8P3cXVU7V21721D+W1X1A3MLtLu33SuTmyX+dpI/n+TsJJ9N8pLNjmurvZJ8X5JXJPncZseyVV9JLkjyimH5WUn+g39LT+ijSvIdw/LTk9yd5JLNjmurvpK8Ncl7k3xos2PZiq8kh5M8b7Pj2OqvJAeS/N1h+ewk5212TF5eZ/Ja75okyeVJ/s3wf84lSe4eyp+T5IHh/fxh+fzNbs9ptu2vnog5yWtPtG1Y39I/E6do2/Ja/98twrX66VwnJ/nrST62QN/bute3i3rOTdm2hTznpmzbQp5z07TtpPqLds6t+/tSkr+f5J8Py29I8v5h+SXD9/WMJC8cvsenzSPO7Tpy6JVJDnX3A939x0nel2TPJse05XT3ryV5ZLPj2Mq6+2h3f3pY/mqS+5JcuLlRbS09cWxYffrwcqf7NVTVjiSvS/Lzmx0Li6uqvjOTX1puTpLu/uPufmxzo4IzM8U1yZ4ktwz/59yV5LyquiDJDyS5o7sf6e5Hk9yR5LL5Rzy99drW3f9+iD1J7kqyY0MCm4EzuJbc8tfqp9m2H07yS3MMZ6amvL5dyHNumrYt6jl3hr+XbOlz7im0bdHOuWl+X9qTyR//kuQDSS6tqhrK39fd3+ju/5jkUCbf58xt1+TQhUkeXLV+JH6h5wwNQ/tenkmml1VqMlXqM0kezuSCQR+t7WeS/FiSP9nsQLawTvLRqvpUVe3b7GC2qD+f5PeS/MuaTFH8+ao6d7ODgjl7smu77XbNd00mozVO2A4/E//KMJXi31TVfzmUbZvvrar+bCbJkV9eVbww39sprm8X/pyb8tp9Ic+5ddq20Ofcet/bop5zU/y+9KffUXcfT/J4kudmA7+7s+ax0y2g1igzkoGnrKq+I5MfQD/a3X+w2fFsNd39zSQvq6rzknywqv5id7uX1SpV9YNJHu7uT1XV8mbHs4W9qrsfqqoXJLmjqr44/PWWbzkrk6kOP9Ldd1fVzya5Psk/2tywYK6e7Npu21zzVdVfy+QX1e9dVbzoPxM/neS7u/tYVV2e5F8n2ZVt9L1lMr3l33X36lFGC/G9rXN9u9Dn3DTX7ot6zq3TtoU+56b8nWshz7kpfl/a9HNuu44cOpLkolXrO5I8tEmxsOCq6umZ/JC6tbt/ZbPj2cqGqS0r2ULDi7eQVyX5oao6nMlQ3ldX1b/a3JC2nu5+aHh/OMkHM6dhswvuSJIjq/7i9IFMkkWwnT3Ztd22uOarqv8qkynHe7r790+UL/rPxO7+gxNTKbr7I0meXlXPyzb53gZvyEnTWxbhe5vi+nZhz7lprt0X9Zxbr22LfM6dxu9cC3nOnXCK35f+9DuqqrOSPDuTqa0b9t1t1+TQJ5LsqqoXVtXZmfwDun2TY2IBDfM8b05yX3f/9GbHsxVV1fOHDHiq6pwk35/ki5sb1dbT3W/r7h3dvTOTn0kf6+6/vclhbSlVdW5VPevEcpLXJDEC7STd/Z+SPFhV3zMUXZrkC5sYEmyE25O8qSYuSfJ4dx9N8m+TvKaqzq+q8zP5ufFvNzPQ01VV35XkV5K8sbv/w6ryhf+ZWFV/briWSlW9MpPfPX4/2+RavaqeneS/TXJwVdmW/96mvL5dyHNumrYt6jk3ZdsW8pyb9neuBT7npvl96fYkJ57+9/pMflfoofwNNXma2QszGQn2G/OIc1tOK+vu41X1DzL5QfW0JL/Q3Z/f5LC2nKr6pUzuaP+8qjqS5Ce6++bNjWrLeVWSNya5d5gjmiRvHzLxTFyQ5EBVPS2T/4Bu626PaeepWMpkmG0y+f/pvd39q5sb0pb1I0luHS7wHkjy5k2OB87IWtckmdywM939z5N8JJOnJx1K8vUM/+a7+5Gq+ieZ/OKTJP/rSVMNNt0UbfvHmdxX4v8Yfv4d7+7dWYCfiVO07fVJ/l5VHU/yh0neMPyys+Wv1adoW5L8jSQf7e6vrfrolv/e8iTXt0m+K1n4c26ati3qOTdN2xb1nJumbcninnNr/r5UVf9rkk929+2ZJMd+saoOZTJi6A1J0t2fr6rbMvlD4PEk1w5T1GauJv9WAAAAABij7TqtDAAAAIApSA4BAAAAjJjkEAAAAMCISQ4BAAAAjJjkEAAAAMCISQ4BAAAAjJjkEAAAAMCISQ4BAAAAjJjkEAAAAMCISQ4BAAAAjJjkEAAAAMCISQ4BAAAAjJjkEAAAAMCISQ4BAAAAjJjkEAAAAMCISQ4BAAAAjJjkEAAAAMCISQ4BAAAAjJjkEAAAAMCISQ4BAAAAjJjkEAAAAMCISQ4BAAAAjJjkEAAAAMCISQ4BAAAAjJjkEAAAAMCISQ4BAAAAjJjkEAAAAMCISQ4BAAAAjJjkEAAAAMCISQ4BAAAAjJjkEAAAAMCISQ4BAAAAjJjkEAAAAMCISQ4BAAAAjJjkEAAAAMCISQ4BAAAAjJjkEDCVqrqqqj66ar2r6kWbGRMAAPNVVf+8qv7RHPb7v1TVv5r1foGnRnII+DZV9b1V9e+r6vGqeqSq/l1V/aXuvrW7XzPlPs6uqhur6khVHauq/1hV75x37AAAY/Fk12yzPk53/w/d/U9mvV9gazlrswMAto6q+s4kH0ry95LcluTsJP9Nkm+c5q7elmR3klcmOZrku5N83+wiBQAYr1lds1VVJanu/pOZBwksFCOHgNX+iyTp7l/q7m929x9290e7+56qurqqfv2k+pdX1QNV9ZWq+mdVdeJnyl9K8sHufqgnDnf3LSc+VFWHq+ptVfWFqnq0qv5lVT1zg9oIALDoTnXN9m3Ttapq53A7gLOG9ZWq+qdV9e+SfD3J26vqk6t3XlX/Y1XdPiy/p6p+ali+r6p+cFW9s4brwFcM65cMo5keq6rPVtXyqrovrKr/t6q+WlV3JHnevDoHOH2SQ8Bq/yHJN6vqQFW9tqrOX6f+38hkhNArkuxJ8neG8ruSvLWq/n5VXTz8VepkVyX5gST/eSYXOP/zTFoAALD9ne4128nemGRfkmcluSnJ91TVrlXb/1aS967xuV9K8sOr1n8gyVe6+9NVdWGSDyf5qSTPSfI/Jfnlqnr+UPe9ST6VSVLonyTZe5oxA3MkOQT8qe7+gyTfm6ST/J9Jfq+qbq+qpSf5yDu6+5Hu/lKSn8m3Lhb+tyTvyCQB9Mkkv1tVJ18A/O/d/WB3P5Lkn+bbLzQAAHgST+Ga7WTv6e7Pd/fx7n48ycEM12JDkugvJLl9jc+9N8kPVdWfHdZXJ5H+dpKPdPdHuvtPuvuOTK4DL6+q78pkZPk/6u5vdPevJfm/T7fdwPxIDgHfprvv6+6ru3tHkr+Y5D/LJPGzlgdXLf/OUDfD8Oaf6+5XJTkvk+TPL1TVi9f7LAAA6zvNa7aTPXjS+nvzrT/U/a0k/7q7v77GMQ8luS/JXx8SRD+UbyWHvjvJlcOUsseq6rFMElgXDLE92t1fW7W735kyVmADSA4BT6q7v5jkPZlccKzlolXL35XkoTX28Yfd/XNJHk3yktP5LAAA6zvpmu1rSf7sqs1/bq2PnLT+0STPq6qXZZIkWmtK2QknppbtSfKFIWGUTBJOv9jd5616ndvdN2TygJLzq+rcVfv5rulaB2wEySHgT1XVX6iq66pqx7B+USb/+d/1JB/5h1V1/lDvLUneP3zuR6tquarOGW5UuDeTOe2/ueqz11bVjqp6TpK3n/gsAACnts4122eSfF9VfVdVPTuTp8ieUncfT/KBJP8sk/sF3XGK6u9L8ppMnpS2Oon0rzIZUfQDVfW0qnrmcD24o7t/J5MpZj9ZVWdX1fcm+eun225gfiSHgNW+muQvJ7m7qr6WyQXG55Jc9yT1D2ZyY8HPZHIDwpuH8j9McmOS/5TkK0muTfLfdfcDqz773kz+SvXA8PqpmbYEAGD7etJrtuFeP+9Pck8m12kfmnKf703y/Un+ryFZtKbuPprk/0vyV7Pqj3vd/WAmo4nenuT3MhlJ9A/zrd85/9YQ8yNJfiLJLQG2jOo+eUQhwHxV1eEkf7e7/5/NjgUAAGDsjBwCAAAAGDHJIQAAAIARM60MAAAAYMSMHAIAAAAYsbM2O4Aked7zntc7d+6c+X6/9rWv5dxzz535frcCbVtc27l92raYtG0xzattn/rUp77S3c+f+Y4ZFdd2i0n/zpf+nT99PF/6d742+9puSySHdu7cmU9+8pMz3+/KykqWl5dnvt+tQNsW13Zun7YtJm1bTPNqW1X9zsx3yui4tltM+ne+9O/86eP50r/ztdnXdqaVAQAAAIyY5BAAAADAiEkOAQAAAIyY5BAAAADAiEkOAQAAAIyY5BAAAADAiEkOAQAAAIyY5BAAAADAiEkOAQAAAIzYWdNUqqrzkvx8kr+YpJP8nSS/leT9SXYmOZzkb3b3o1VVSX42yeVJvp7k6u7+9Mwjn8K9v/t4rr7+w6esc/iG121QNAAAnAnXdgAwH9OOHPrZJL/a3X8hyUuT3Jfk+iR3dveuJHcO60ny2iS7hte+JO+eacQAAAAAzMy6yaGq+s4k35fk5iTp7j/u7seS7ElyYKh2IMkVw/KeJLf0xF1JzquqC2YeOQAAAABnbJppZX8+ye8l+ZdV9dIkn0ryliRL3X00Sbr7aFW9YKh/YZIHV33+yFB2dPVOq2pfJiOLsrS0lJWVlTNoxtqWzkmuu/j4KevM47gb4dixYwsb+3q2c9uS7d0+bVtM2raYtnPbAADYWNMkh85K8ookP9Ldd1fVz+ZbU8jWUmuU9RMKuvcn2Z8ku3fv7uXl5SlCOT033XowN9576iYevmr2x90IKysrmUefbQXbuW3J9m6fti0mbVtM27ltAABsrGnuOXQkyZHuvntY/0AmyaIvn5guNrw/vKr+Ras+vyPJQ7MJFwAAAIBZWjc51N3/KcmDVfU9Q9GlSb6Q5PYke4eyvUkODsu3J3lTTVyS5PET088AAAAA2FqmepR9kh9JcmtVnZ3kgSRvziSxdFtVXZPkS0muHOp+JJPH2B/K5FH2b55pxAAAAADMzFTJoe7+TJLda2y6dI26neTaM4wLAAAAgA0wzT2HAAAAANimJIcAAAAARkxyCAAAAGDEJIcAAAAARkxyCAAAAGDEJIcAAAAARkxyCAAAAGDEJIcAAEamqg5X1b1V9Zmq+uRQ9pyquqOq7h/ezx/Kq6reVVWHquqeqnrF5kYPAMya5BAAwDj9te5+WXfvHtavT3Jnd+9KcuewniSvTbJreO1L8u4NjxQAmCvJIQAAkmRPkgPD8oEkV6wqv6Un7kpyXlVdsBkBAgDzcdZmBwAAwIbrJB+tqk7yL7p7f5Kl7j6aJN19tKpeMNS9MMmDqz57ZCg7unqHVbUvk5FFWVpaysrKysyDXjonue7i46esM4/jjsWxY8f03xzp3/nTx/Olf+drs/tXcggAYHxe1d0PDQmgO6rqi6eoW2uU9RMKJgmm/Umye/fuXl5enkmgq91068HceO+pL18PXzX7447FyspK5vG9MaF/508fz5f+na/N7l/TygAARqa7HxreH07ywSSvTPLlE9PFhveHh+pHkly06uM7kjy0cdECAPMmOQQAMCJVdW5VPevEcpLXJPlcktuT7B2q7U1ycFi+PcmbhqeWXZLk8RPTzwCA7cG0MgCAcVlK8sGqSibXgu/t7l+tqk8kua2qrknypSRXDvU/kuTyJIeSfD3Jmzc+ZABgniSHAABGpLsfSPLSNcp/P8mla5R3kms3IDQAYJOYVgYAAAAwYpJDAAAAACMmOQQAAAAwYpJDAAAAACMmOQQAAAAwYpJDAAAAACMmOQQAAAAwYpJDAAAAACMmOQQAAAAwYpJDAAAAACM2VXKoqg5X1b1V9Zmq+uRQ9pyquqOq7h/ezx/Kq6reVVWHquqeqnrFPBsAAAAAwFN3OiOH/lp3v6y7dw/r1ye5s7t3JblzWE+S1ybZNbz2JXn3rIIFAAAAYLbOZFrZniQHhuUDSa5YVX5LT9yV5LyquuAMjgMAAADAnJw1Zb1O8tGq6iT/orv3J1nq7qNJ0t1Hq+oFQ90Lkzy46rNHhrKjq3dYVfsyGVmUpaWlrKysPOVGPJmlc5LrLj5+yjrzOO5GOHbs2MLGvp7t3LZke7dP2xaTti2m7dw2AAA21rTJoVd190NDAuiOqvriKerWGmX9hIJJgml/kuzevbuXl5enDGV6N916MDfee+omHr5q9sfdCCsrK5lHn20F27ltyfZun7YtJm1bTNu5bQAAbKypppV190PD+8NJPpjklUm+fGK62PD+8FD9SJKLVn18R5KHZhUwAAAAALOzbnKoqs6tqmedWE7ymiSfS3J7kr1Dtb1JDg7Ltyd50/DUskuSPH5i+hkAAAAAW8s008qWknywqk7Uf293/2pVfSLJbVV1TZIvJblyqP+RJJcnOZTk60nePPOoAQAAAJiJdZND3f1AkpeuUf77SS5do7yTXDuT6AAAAACYqzN5lD0AAAAAC05yCAAAAGDEJIcAAAAARkxyCAAAAGDEJIcAAAAARkxyCAAAAGDEJIcAAAAARkxyCAAAAGDEJIcAAAAARkxyCAAAAGDEJIcAAAAARkxyCABghKrqaVX1m1X1oWH9hVV1d1XdX1Xvr6qzh/JnDOuHhu07NzNuAGD2JIcAAMbpLUnuW7X+jiTv7O5dSR5Ncs1Qfk2SR7v7RUneOdQDALYRySEAgJGpqh1JXpfk54f1SvLqJB8YqhxIcsWwvGdYz7D90qE+ALBNSA4BAIzPzyT5sSR/Mqw/N8lj3X18WD+S5MJh+cIkDybJsP3xoT4AsE2ctdkBAACwcarqB5M83N2fqqrlE8VrVO0ptq3e774k+5JkaWkpKysrZx7sSZbOSa67+Pgp68zjuGNx7Ngx/TdH+nf+9PF86d/52uz+lRwCABiXVyX5oaq6PMkzk3xnJiOJzquqs4bRQTuSPDTUP5LkoiRHquqsJM9O8sjJO+3u/Un2J8nu3bt7eXl55oHfdOvB3HjvqS9fD181++OOxcrKSubxvTGhf+dPH8+X/p2vze5f08oAAEaku9/W3Tu6e2eSNyT5WHdfleTjSV4/VNub5OCwfPuwnmH7x7r7CSOHAIDFJTkEAECS/HiSt1bVoUzuKXTzUH5zkucO5W9Ncv0mxQcAzIlpZQAAI9XdK0lWhuUHkrxyjTp/lOTKDQ0MANhQRg4BAAAAjJjkEAAAAMCISQ4BAAAAjJjkEAAAAMCISQ4BAAAAjJjkEAAAAMCISQ4BAAAAjNjUyaGqelpV/WZVfWhYf2FV3V1V91fV+6vq7KH8GcP6oWH7zvmEDgAAAMCZOp2RQ29Jct+q9XckeWd370ryaJJrhvJrkjza3S9K8s6hHgAAAABb0FTJoarakeR1SX5+WK8kr07ygaHKgSRXDMt7hvUM2y8d6gMAAACwxZw1Zb2fSfJjSZ41rD83yWPdfXxYP5LkwmH5wiQPJkl3H6+qx4f6X1m9w6ral2RfkiwtLWVlZeUpNuHJLZ2TXHfx8VPWmcdxN8KxY8cWNvb1bOe2Jdu7fdq2mLRtMW3ntgEAsLHWTQ5V1Q8mebi7P1VVyyeK16jaU2z7VkH3/iT7k2T37t29vLx8cpUzdtOtB3Pjvadu4uGrZn/cjbCyspJ59NlWsJ3blmzv9mnbYtK2xbSd2wYAwMaaZuTQq5L8UFVdnuSZSb4zk5FE51XVWcPooR1JHhrqH0lyUZIjVXVWkmcneWTmkQMAAABwxta951B3v627d3T3ziRvSPKx7r4qyceTvH6otjfJwWH59mE9w/aPdfcTRg4BAAAAsPlO52llJ/vxJG+tqkOZ3FPo5qH85iTPHcrfmuT6MwsRAAAAgHmZ9obUSZLuXkmyMiw/kOSVa9T5oyRXziA2AAAAAObsTEYOAQAAALDgTmvkEAAAAADT23n9h9et857Lzt2ASJ6ckUMAAAAAIyY5BAAAADBikkMAAAAAIyY5BAAAADBikkMAAAAAIyY5BAAAADBikkMAAAAAIyY5BAAAADBikkMAAAAAIyY5BAAAADBikkMAAAAAIyY5BAAAADBikkMAAAAAIyY5BAAwIlX1zKr6jar6bFV9vqp+cih/YVXdXVX3V9X7q+rsofwZw/qhYfvOzYwfAJg9ySEAgHH5RpJXd/dLk7wsyWVVdUmSdyR5Z3fvSvJokmuG+tckebS7X5TknUM9AGAbkRwCABiRnjg2rD59eHWSVyf5wFB+IMkVw/KeYT3D9kurqjYoXABgA5y12QEAALCxquppST6V5EVJfi7Jbyd5rLuPD1WOJLlwWL4wyYNJ0t3Hq+rxJM9N8pWTsfOW9gAAFTZJREFU9rkvyb4kWVpaysrKyszjXjonue7i46esM4/jjsWxY8f03xzp3/nTx/Olf5+69f7vSja/fyWHAABGpru/meRlVXVekg8mefFa1Yb3tUYJ9RMKuvcn2Z8ku3fv7uXl5dkEu8pNtx7Mjfee+vL18FWzP+5YrKysZB7fGxP6d/708Xzp36fu6us/vG6d91x27qb2r2llAAAj1d2PJVlJckmS86rqROZlR5KHhuUjSS5KkmH7s5M8srGRAgDzJDkEADAiVfX8YcRQquqcJN+f5L4kH0/y+qHa3iQHh+Xbh/UM2z/W3U8YOQQALC7TygAAxuWCJAeG+w79mSS3dfeHquoLSd5XVT+V5DeT3DzUvznJL1bVoUxGDL1hM4IGAOZHcggAYES6+54kL1+j/IEkr1yj/I+SXLkBoQEAm8S0MgAAAIARkxwCAAAAGLF1k0NV9cyq+o2q+mxVfb6qfnIof2FV3V1V91fV+6vq7KH8GcP6oWH7zvk2AQAAAICnapqRQ99I8urufmmSlyW5rKouSfKOJO/s7l1JHk1yzVD/miSPdveLkrxzqAcAAADAFrRucqgnjg2rTx9eneTVST4wlB9IcsWwvGdYz7D90qqqmUUMAAAAwMxM9bSy4VGnn0ryoiQ/l+S3kzzW3ceHKkeSXDgsX5jkwSTp7uNV9XiS5yb5ykn73JdkX5IsLS1lZWXljBqylqVzkusuPn7KOvM47kY4duzYwsa+nu3ctmR7t0/bFpO2Labt3DYAADbWVMmh7v5mkpdV1XlJPpjkxWtVG97XGiXUTyjo3p9kf5Ls3r27l5eXpwnltNx068HceO+pm3j4qtkfdyOsrKxkHn22FWzntiXbu33atpi0bTFt57YBALCxTutpZd39WJKVJJckOa+qTmRediR5aFg+kuSiJBm2PzvJI7MIFgAAAIDZmuZpZc8fRgylqs5J8v1J7kvy8SSvH6rtTXJwWL59WM+w/WPd/YSRQwAAAABsvmmmlV2Q5MBw36E/k+S27v5QVX0hyfuq6qeS/GaSm4f6Nyf5xao6lMmIoTfMIW4AAAAAZmDd5FB335Pk5WuUP5DklWuU/1GSK2cSHQAAAABzdVr3HAIAAABge5EcAgAAABgxySEAAACAEZMcAgAAABgxySEAAACAEZMcAgAAABgxySEAAACAEZMcAgAAABgxySEAAACAEZMcAgAAABgxySEAAACAEZMcAgAAABgxySEAAACAEZMcAgAAABgxySEAAACAEZMcAgAAABgxySEAAACAEZMcAgAAABgxySEAAACAEZMcAgAYkaq6qKo+XlX3VdXnq+otQ/lzquqOqrp/eD9/KK+qeldVHaqqe6rqFZvbAgBg1iSHAADG5XiS67r7xUkuSXJtVb0kyfVJ7uzuXUnuHNaT5LVJdg2vfUnevfEhAwDzJDkEADAi3X20uz89LH81yX1JLkyyJ8mBodqBJFcMy3uS3NITdyU5r6ou2OCwAYA5OmuzAwAAYHNU1c4kL09yd5Kl7j6aTBJIVfWCodqFSR5c9bEjQ9nRk/a1L5ORRVlaWsrKysrM4106J7nu4uOnrDOP447FsWPH9N8c6d/508fzpX+fuvX+70o2v38lhwAARqiqviPJLyf50e7+g6p60qprlPUTCrr3J9mfJLt37+7l5eUZRfotN916MDfee+rL18NXzf64Y7GyspJ5fG9M6N/508fzpX+fuquv//C6dd5z2bmb2r+mlQEAjExVPT2TxNCt3f0rQ/GXT0wXG94fHsqPJLlo1cd3JHloo2IFAOZPcggAYERqMkTo5iT3dfdPr9p0e5K9w/LeJAdXlb9peGrZJUkePzH9DADYHkwrAwAYl1cleWOSe6vqM0PZ25PckOS2qromyZeSXDls+0iSy5McSvL1JG/e2HABgHlbNzlUVRcluSXJn0vyJ0n2d/fPVtVzkrw/yc4kh5P8ze5+dPhr1M9mchHx9SRXn3giBmdu5zBX8bqLjz/pvMXDN7xuI0MCABZId/961r6PUJJcukb9TnLtXIMCADbVNNPKjie5rrtfnOSSJNdW1UuSXJ/kzu7eleTOYT1JXptk1/Dal+TdM48aAAAAgJlYNznU3UdPjPzp7q8muS+Tx5fuSXJgqHYgyRXD8p4kt/TEXUnOO3FzQwAAAAC2ltO651BV7Uzy8iR3J1k6cTPC7j5aVS8Yql2Y5MFVHzsylH3bjQural8mI4uytLSUlZWV049+HUvnTKZfnco8jjtPJ9pzqrYtWptOduzYsYVvw6ls5/Zp22LStsW0ndsGAMDGmjo59P+3d/+xltTlHcffH1ms/BDRooSyVGiCpgabSjZCS2JWV82CBGyCFqIIhJb+AOMP0pbaP+yPf0ittmgN7RYoYBFE1LKptGisG2tTKApUfmm7xS0uUBYrhVLa4tqnf5y59rose4dl5syemfcrubnnzJ2983yfe/ac73nOPN9Jsj+zS56+u6oemy0ttPNdd7KtnrKhagOwAWDNmjW1du3atqG09pGrrueDd+x6iFve1v1x+3TmsjWHnm5sizamHW3atIk+Hg97ijGPz7EtJse2mMY8NkmSJM1Xq0vZJ9mbWWHoqqr6dLP5oaV2seb7tmb7VuCwZf98NfBAN+FKkiRJkiSpSysWh5qrj10K3FNVH1r2o43AGc3tM4Drl21/R2aOBR5daj+TJEmSJEnSnqVNW9lxwOnAHUlub7a9D7gQuDbJ2cB9wFuan93A7DL2m5ldyv6sTiOWJEmSJElSZ1YsDlXVl9n5OkIA63ayfwHnPsu4JEmSJEmSNAet1hySJEmSJEnSOFkckiRJkiRJmjCLQ5IkSZIkSRNmcUiSJEmSJGnCLA5JkiRJkiRNmMUhSZIkSZKkCbM4JEmSJEmSNGEWhyRJkiRJkibM4pAkSZIkSdKEWRySJEmSJEmaMItDkiRJkiRJE2ZxSJIkSZIkacIsDkmSJEmSJE2YxSFJkiRJkqQJszgkSZIkSZI0YRaHJEmSJEmSJszikCRJkiRJ0oRZHJIkSZIkSZowi0OSJEmSJEkTZnFIkiRJkiRpwiwOSZIkSZIkTZjFIUmSJEmSpAmzOCRJkjQhSS5Lsi3Jncu2vSjJ55P8U/P9hc32JPlwks1Jvpbk6OEilyRJfbE4JEmSNC2XA+t32HYB8IWqOhL4QnMf4HjgyObrHODiOcUoSZLmyOKQJEnShFTVl4Dv7LD5ZOCK5vYVwJuXbb+yZm4CDkxyyHwilSRJ82JxSJIkSQdX1YMAzfeXNNsPBb61bL+tzTZJkjQiq1baIcllwInAtqo6qtn2IuATwOHAFuCtVfVIkgAXAScATwBnVtWt/YQuSZKknmUn22qnOybnMGs94+CDD2bTpk2dB3PwPnD+K7fvcp8+jjsVjz/+uPnrkfntnznul/ndfSu9dsHw+V2xOMSsL/0PgSuXbVvqS78wyQXN/V/jB/vSj2HWl35MlwFLkiSpcw8lOaSqHmzaxrY127cChy3bbzXwwM5+QVVtADYArFmzptauXdt5kB+56no+eMeup69b3tb9cadi06ZN9PF304z57Z857pf53X1nXvDZFfe5fP1+g+Z3xbYy+9IlSZJGbyNwRnP7DOD6Zdvf0Vy17Fjg0aX2M0mSNB5tzhzamR/oS0+yUl/6UyYRnnq8e5bGs6uxLdqYdjT06XR9G/P4HNticmyLacxjU7+SXA2sBQ5KshV4P3AhcG2Ss4H7gLc0u9/AbLmAzcyWDDhr7gFLkqTe7W5x6Om07kv31OPds3Q62vmv3P60Y1u0Me1o7Kcrjnl8jm0xObbFNOaxqV9VddrT/GjdTvYt4Nx+I5IkSUPb3auVPbTULra7femSJEmSJEka3u4Wh+xLlyRJkiRJGoE2l7K3L12SJEmSJGmkViwO2ZcuSZIkSZI0XrvbViZJkiRJkqQRsDgkSZIkSZI0YRaHJEmSJEmSJszikCRJkiRJ0oRZHJIkSZIkSZowi0OSJEmSJEkTZnFIkiRJkiRpwiwOSZIkSZIkTZjFIUmSJEmSpAmzOCRJkiRJkjRhFockSZIkSZImzOKQJEmSJEnShFkckiRJkiRJmjCLQ5IkSZIkSRNmcUiSJEmSJGnCLA5JkiRJkiRNmMUhSZIkSZKkCbM4JEmSJEmSNGEWhyRJkiRJkibM4pAkSZIkSdKErRo6AOnwCz77A/fPf+V2ztxh25YL3zTPkCRJkiRJmgzPHJIkSZIkSZowi0OSJEmSJEkTZluZpM7ccf+jT2kJ3JEtgpIkSZK0Z7E4JPVgx3WUlltaU8kiiSRJkiRpT2BbmSRJkiRJ0oRZHJIkSZIkSZqwXtrKkqwHLgL2Ai6pqgv7OI4kSWO0q9bUJZev328OkUgzzu0kSRq3zotDSfYCPgq8AdgK3JJkY1Xd3fWxJKlvS2/Sl9aK2pFrR0kaO+d2kiSNXx9tZa8GNlfVvVX1JHANcHIPx5EkSVL/nNtJkjRyqapuf2FyCrC+qn6uuX86cExVnbfDfucA5zR3Xw58o9NAZg4Cvt3D790TOLbFNebxObbF5NgWU19je2lVvbiH36sF5dxuUsxvv8xv/8xxv8xvvwad2/Wx5lB2su0pFaiq2gBs6OH4/x9I8pWqWtPnMYbi2BbXmMfn2BaTY1tMYx6b9jjO7SbC/PbL/PbPHPfL/PZr6Pz20Va2FThs2f3VwAM9HEeSJEn9c24nSdLI9VEcugU4MskRSZ4LnAps7OE4kiRJ6p9zO0mSRq7ztrKq2p7kPOBGZpc7vayq7ur6OC31emrzwBzb4hrz+BzbYnJsi2nMY9MexLndpJjffpnf/pnjfpnffg2a384XpJYkSZIkSdLi6KOtTJIkSZIkSQvC4pAkSZIkSdKEjbY4lGR9km8k2ZzkgqHj6UqSy5JsS3Ln0LF0LclhSb6Y5J4kdyV519AxdSXJ85L8fZJ/aMb2W0PH1LUkeyW5LclfDB1Ll5JsSXJHktuTfGXoeLqW5MAk1yX5evN/76eGjqkLSV7e/M2Wvh5L8u6h4+pKkvc0zyV3Jrk6yfOGjknq0krzuCQ/lOQTzc9vTnL4/KNcXC3y+94kdyf5WpIvJHnpEHEuqrbvQ5KckqSSeGnwZ6BNfpO8tXkM35Xk4/OOcdG1eI740eZ9223N88QJQ8S5iFZ6P5+ZDze5/1qSo+cW2xjXHEqyF/CPwBuYXX71FuC0qrp70MA6kOQ1wOPAlVV11NDxdCnJIcAhVXVrkucDXwXePJK/W4D9qurxJHsDXwbeVVU3DRxaZ5K8F1gDHFBVJw4dT1eSbAHWVNW3h46lD0muAP6mqi5prkK0b1X9+9Bxdal5TbgfOKaq/mXoeJ6tJIcyew55RVX9V5JrgRuq6vJhI5O60WYel+SXgZ+oql9McirwM1X1s4MEvGBa5ve1wM1V9USSXwLWmt922r4Paea6nwWeC5xXVaP7AKoPLR+/RwLXAq+rqkeSvKSqtg0S8AJqmeMNwG1VdXGSVzCbhxw+RLyLZqX3802h7Z3ACcAxwEVVdcw8YhvrmUOvBjZX1b1V9SRwDXDywDF1oqq+BHxn6Dj6UFUPVtWtze3/AO4BDh02qm7UzOPN3b2br9FUZpOsBt4EXDJ0LGovyQHAa4BLAarqybEVhhrrgH8eQ2FomVXAPklWAfsCDwwcj9SlNvO4k4ErmtvXAeuaD2K0shXzW1VfrKonmrs3AavnHOMia/s+5HeA3wX+e57BjUCb/P488NGqegTAwtAz1ibHBRzQ3H4BzkNaa/F+/mRmhaNqTiQ4sDmJondjLQ4dCnxr2f2tjKTIMBXN6eGvAm4eNpLuNG1XtwPbgM9X1WjGBvwB8KvA/w4dSA8K+FySryY5Z+hgOvZjwMPAnzanBV+SZL+hg+rBqcDVQwfRlaq6H/g94D7gQeDRqvrcsFFJnWozj/v+PlW1HXgU+OG5RLf4nuk8+WzgL3uNaFxWzG+SVwGHVdWoWvHnpM3j92XAy5L8bZKbkqyfW3Tj0CbHvwm8PclW4AZmZ7qoG4PVMsZaHNrZJ0ejOUtj7JLsD3wKeHdVPTZ0PF2pqu9V1U8y+/Tt1UlG0RaY5ERgW1V9dehYenJcVR0NHA+c25wKOhargKOBi6vqVcB/AqNZow2gaZU7Cfjk0LF0JckLmX2qdATwI8B+Sd4+bFRSp9rM45zr7b7WuWueW9YAH+g1onHZZX6TPAf4feD8uUU0Lm0ev6uAI4G1wGnAJUkO7DmuMWmT49OAy6tqNbP2p481j209e4O9vo31D7gVOGzZ/dV4qttCaNbj+RRwVVV9euh4+tC07WwCxvIpxnHASc3aPNcAr0vyZ8OG1J2qeqD5vg34DLNTbcdiK7B12Vls1zErFo3J8cCtVfXQ0IF06PXAN6vq4ar6LvBp4KcHjknqUpt53Pf3adorX8BI2+570GqenOT1wG8AJ1XV/8wptjFYKb/PB44CNjVzp2OBjS5K3Vrb54frq+q7VfVN4BvMikVqp02Oz2a2rhNV9XfA84CD5hLd+A1WyxhrcegW4MgkRzSfGp8KbBw4Jq2gWSvgUuCeqvrQ0PF0KcmLlz6xSLIPszd3Xx82qm5U1a9X1epmEbpTgb+uqlGcxZBkv2bBSJp2qzcCo7lSYFX9K/CtJC9vNq0DFn4B+B2cxohayhr3Accm2bd53lzHbI02aSzazOM2Amc0t09h9trjmUPtrJjfpu3pj5kVhlyv5ZnZZX6r6tGqOqiqDm/mTjcxy7MLUrfT5vnhz4HXAiQ5iFmb2b1zjXKxtcnxfczmHyT5cWbFoYfnGuV4bQTe0Vy17Fhmywc8OI8Dr5rHQeatqrYnOQ+4EdgLuKyq7ho4rE4kuZrZKZIHNT2e76+qS4eNqjPHAacDdzRr8wC8r6puGDCmrhwCXNGs/v8c4Fr7zBfCwcBnmjVOVwEfr6q/Gjakzr0TuKp58b8XOGvgeDqTZF9mV9r4haFj6VJV3ZzkOuBWYDtwG7Bh2Kik7jzdPC7JbwNfqaqNzD5M+liSzczOGDp1uIgXS8v8fgDYH/hk8xp4X1WdNFjQC6RlfrWbWub3RuCNSe4Gvgf8SlX923BRL5aWOT4f+JMk72HW8nSmBfp2dvZ+ntnFiqiqP2K2htMJwGbgCeY4Nx/lpewlSZIkSZLUzljbyiRJkiRJktSCxSFJkiRJkqQJszgkSZIkSZI0YRaHJEmSJEmSJszikCRJkiRJ0oRZHJIkSZIkSZowi0OSJEmSJEkT9n/ykf84bMuo6wAAAABJRU5ErkJggg==\n",
      "text/plain": [
       "<Figure size 1440x1080 with 6 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Let's plot some histograms to have a previzualisation of some of the data ...\n",
    "train_df_raw.drop(['PassengerId'], 1).hist(bins=50, figsize=(20,15))\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [],
   "source": [
    "def preprocess_data(df):\n",
    "    \n",
    "    processed_df=df\n",
    "    #drop useless columns\n",
    "    processed_df= processed_df.drop(['PassengerId'], 1)\n",
    "    \n",
    "    # Deal with missing values\n",
    "    processed_df['Embarked'].fillna('S', inplace=True)\n",
    "    processed_df['Age'].fillna(processed_df['Age'].mean(), inplace=True)\n",
    "    processed_df['Age'] = processed_df['Age'].astype(int)\n",
    "    processed_df['Fare'] = processed_df['Fare'].interpolate()\n",
    "    processed_df['Cabin'].fillna('U', inplace=True)\n",
    "    \n",
    "    # feature engineering on columns\n",
    "    processed_df['Title'] = pd.Series((name.split('.')[0].split(',')[1].strip() for name in train_df_raw['Name']), index=train_df_raw.index)\n",
    "    processed_df['Title'] = processed_df['Title'].replace(['Lady', 'the Countess','Countess','Capt', 'Col','Don', 'Dr', 'Major', 'Rev', 'Sir', 'Jonkheer', 'Dona'], 'Rare')\n",
    "    processed_df['Title'] = processed_df['Title'].replace(['Mlle', 'Ms'], 'Miss')\n",
    "    processed_df['Title'] = processed_df['Title'].replace('Mme', 'Mrs')\n",
    "    processed_df['Title'] = processed_df['Title'].map({\"Mr\": 1, \"Miss\": 2, \"Mrs\": 3, \"Master\": 4, \"Rare\": 5})  #map will replace\n",
    "    \n",
    "    processed_df['Sex'] = processed_df['Sex'].map({'male': 0, 'female': 1})\n",
    "    processed_df['Embarked'] = processed_df['Embarked'].map({'S': 0, 'C': 1, 'Q': 2})\n",
    "    \n",
    "    processed_df['Familly_size'] = processed_df['SibSp'] + processed_df['Parch'] + 1\n",
    "    processed_df['IsAlone'] = np.where(processed_df['Familly_size']!=1, 0, 1)\n",
    "    \n",
    "    #Creation of a deck column corresponding to the letter contained in the cabin value\n",
    "    processed_df['Cabin'] = processed_df['Cabin'].str[:1]\n",
    "    processed_df['Cabin'] = processed_df['Cabin'].map({cabin: p for p, cabin in enumerate(set(cab for cab in processed_df['Cabin']))})\n",
    "    \n",
    "    processed_df = processed_df.drop(['Name', 'Ticket'], 1)    \n",
    "    \n",
    "    return processed_df"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Let try sevel model on our data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {
    "scrolled": true
   },
   "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>Pclass</th>\n",
       "      <th>Sex</th>\n",
       "      <th>Age</th>\n",
       "      <th>SibSp</th>\n",
       "      <th>Parch</th>\n",
       "      <th>Fare</th>\n",
       "      <th>Cabin</th>\n",
       "      <th>Embarked</th>\n",
       "      <th>Title</th>\n",
       "      <th>Familly_size</th>\n",
       "      <th>IsAlone</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>331</th>\n",
       "      <td>-1.566107</td>\n",
       "      <td>-0.737695</td>\n",
       "      <td>1.188306</td>\n",
       "      <td>-0.474545</td>\n",
       "      <td>-0.473674</td>\n",
       "      <td>-0.074583</td>\n",
       "      <td>-2.275539</td>\n",
       "      <td>-0.568837</td>\n",
       "      <td>-0.707550</td>\n",
       "      <td>-0.560975</td>\n",
       "      <td>0.811922</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>733</th>\n",
       "      <td>-0.369365</td>\n",
       "      <td>-0.737695</td>\n",
       "      <td>-0.503160</td>\n",
       "      <td>-0.474545</td>\n",
       "      <td>-0.473674</td>\n",
       "      <td>-0.386671</td>\n",
       "      <td>0.198034</td>\n",
       "      <td>-0.568837</td>\n",
       "      <td>-0.707550</td>\n",
       "      <td>-0.560975</td>\n",
       "      <td>0.811922</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>382</th>\n",
       "      <td>0.827377</td>\n",
       "      <td>-0.737695</td>\n",
       "      <td>0.188804</td>\n",
       "      <td>-0.474545</td>\n",
       "      <td>-0.473674</td>\n",
       "      <td>-0.488854</td>\n",
       "      <td>0.198034</td>\n",
       "      <td>-0.568837</td>\n",
       "      <td>-0.707550</td>\n",
       "      <td>-0.560975</td>\n",
       "      <td>0.811922</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>704</th>\n",
       "      <td>0.827377</td>\n",
       "      <td>-0.737695</td>\n",
       "      <td>-0.272505</td>\n",
       "      <td>0.432793</td>\n",
       "      <td>-0.473674</td>\n",
       "      <td>-0.490280</td>\n",
       "      <td>0.198034</td>\n",
       "      <td>-0.568837</td>\n",
       "      <td>-0.707550</td>\n",
       "      <td>0.059160</td>\n",
       "      <td>-1.231645</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>813</th>\n",
       "      <td>0.827377</td>\n",
       "      <td>1.355574</td>\n",
       "      <td>-1.810201</td>\n",
       "      <td>3.154809</td>\n",
       "      <td>2.008933</td>\n",
       "      <td>-0.018709</td>\n",
       "      <td>0.198034</td>\n",
       "      <td>-0.568837</td>\n",
       "      <td>0.263832</td>\n",
       "      <td>3.159833</td>\n",
       "      <td>-1.231645</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "       Pclass       Sex       Age     SibSp     Parch      Fare     Cabin  \\\n",
       "331 -1.566107 -0.737695  1.188306 -0.474545 -0.473674 -0.074583 -2.275539   \n",
       "733 -0.369365 -0.737695 -0.503160 -0.474545 -0.473674 -0.386671  0.198034   \n",
       "382  0.827377 -0.737695  0.188804 -0.474545 -0.473674 -0.488854  0.198034   \n",
       "704  0.827377 -0.737695 -0.272505  0.432793 -0.473674 -0.490280  0.198034   \n",
       "813  0.827377  1.355574 -1.810201  3.154809  2.008933 -0.018709  0.198034   \n",
       "\n",
       "     Embarked     Title  Familly_size   IsAlone  \n",
       "331 -0.568837 -0.707550     -0.560975  0.811922  \n",
       "733 -0.568837 -0.707550     -0.560975  0.811922  \n",
       "382 -0.568837 -0.707550     -0.560975  0.811922  \n",
       "704 -0.568837 -0.707550      0.059160 -1.231645  \n",
       "813 -0.568837  0.263832      3.159833 -1.231645  "
      ]
     },
     "execution_count": 9,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Let's divide the train dataset in two datasets to evaluate perfomance of machine learning models used\n",
    "\n",
    "train_df = train_df_raw.copy()\n",
    "X = train_df.drop(['Survived'], 1)  # remove survived column from training sert\n",
    "Y = train_df['Survived']\n",
    "\n",
    "X = preprocess_data(X)\n",
    "sc = StandardScaler()\n",
    "X = pd.DataFrame(sc.fit_transform(X.values), index=X.index, columns=X.columns)\n",
    "    \n",
    "# Split dataset for prediction\n",
    "X_train, X_test, Y_train, Y_test = train_test_split(X, Y, test_size=0.2, random_state=42)  # this is function need to import from sklearn\n",
    "\n",
    "X_train.head()\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## 1.1 Logistic regression"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Logistic regression is the \"hello world\" of machine learning algorithms. It is very simple to understand how it works, here is a good article which cover theory of this algorithm."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Create and train model on train data sample\n",
    "logisticRegr = LogisticRegression(random_state=42)\n",
    "logisticRegr.fit(X_train, Y_train)\n",
    "\n",
    "# Predict for test data sample\n",
    "logistic_prediction = logisticRegr.predict(X_test)\n",
    "\n",
    "# Compute error between predicted data and true response and display it in confusion matrix\n",
    "score = metrics.accuracy_score(Y_test, logistic_prediction)     # ---> Output 0.7988826815642458\n",
    "#display_confusion_matrix(Y_test, logistic_prediction, score=score)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## 1.2 Decision Tree"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Decision tree is a quite intuitive model, easy to vizualize and interpret. Here, we are even going to display our tree to improve our understanding on how the algorithm manage to classify our samples\n",
    "\n",
    "https://medium.com/@chiragsehra42/decision-trees-explained-easily-28f23241248\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.8044692737430168"
      ]
     },
     "execution_count": 11,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Create and train model on train data sample\n",
    "dt = DecisionTreeClassifier(min_samples_split=15, min_samples_leaf=20, random_state=42)\n",
    "dt.fit(X_train, Y_train)\n",
    "\n",
    "# Predict for test data sample\n",
    "dt_prediction = dt.predict(X_test)\n",
    "\n",
    "# Compute error between predicted data and true response and display it in confusion matrix\n",
    "metrics.accuracy_score(Y_test, dt_prediction)   # score=0.7932960893854749\n",
    "#display_confusion_matrix(Y_test, dt_prediction, score=score)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {},
   "outputs": [
    {
     "ename": "NameError",
     "evalue": "name 'visualize_tree' is not defined",
     "output_type": "error",
     "traceback": [
      "\u001b[1;31m---------------------------------------------------------------------------\u001b[0m",
      "\u001b[1;31mNameError\u001b[0m                                 Traceback (most recent call last)",
      "\u001b[1;32m<ipython-input-14-baf707cd5497>\u001b[0m in \u001b[0;36m<module>\u001b[1;34m()\u001b[0m\n\u001b[1;32m----> 1\u001b[1;33m \u001b[0mvisualize_tree\u001b[0m\u001b[1;33m(\u001b[0m\u001b[0mdt\u001b[0m\u001b[1;33m,\u001b[0m \u001b[0mX_test\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0mcolumns\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0m\u001b[0;32m      2\u001b[0m \u001b[0mget_ipython\u001b[0m\u001b[1;33m(\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0msystem\u001b[0m\u001b[1;33m(\u001b[0m\u001b[1;34m' dot -Tpng dt.dot > dt.png'\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n",
      "\u001b[1;31mNameError\u001b[0m: name 'visualize_tree' is not defined"
     ]
    }
   ],
   "source": [
    "visualize_tree(dt, X_test.columns)\n",
    "! dot -Tpng dt.dot > dt.png"
   ]
  },
  {
   "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": "3.6.5"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}