Titanic Kaggle Challenge.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Titanic: Machine Learning from Disaster\n",
    "\n",
    "This is my first Kaggle competition applying the knowledge I have acquired due to the last months / year.\n",
    "This version is on purpose not shortened to display the learning process made during the write-up.\n",
    "\n",
    "## Table of Content\n",
    "* [Loading own functions](#first-bullet)\n",
    "* [Data Overview](#second-bullet)\n",
    "* [Data Pre-Processing: Feature Engineering](#third-bullet)\n",
    "* [Data Pre-Processing: Handling Missing Values](#thirdand1-bullet)\n",
    "* [Data Pre Processing: Transforming the variables into numericals](#fourth-bullet)\n",
    "* [Data Pre Processing: Dropping uncessesary features](#fourthand1-bullet)\n",
    "* [Exploratory Data Analysis](#fifth-bullet)\n",
    "* [Machine Learning Models ](#sixth-bullet)\n",
    "* [Cross Validation](#seventh-bullet)\n",
    "* [Grid Search](#eight-bullet)\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Thanks for Feature Engineering:\n",
    "* https://www.kaggle.com/vbmokin/titanic-0-83253-comparison-20-popular-models\n",
    "* https://www.kaggle.com/startupsci/titanic-data-science-solutions\n",
    "* https://www.kaggle.com/gunesevitan/titanic-advanced-feature-engineering-tutorial#2.-Feature-Engineering\n",
    "\n",
    "Thanks for Grid Search:\n",
    "* https://www.kaggle.com/umarsajjad/top-3-titanic-survival-determination#6---Feature-Engineering"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 56,
   "metadata": {},
   "outputs": [],
   "source": [
    "import pandas as pd\n",
    "import numpy as np\n",
    "import sklearn as skl\n",
    "import kaggle\n",
    "import matplotlib.pyplot as plt\n",
    "import seaborn as sns\n",
    "%matplotlib inline\n",

    "\n",
    "import warnings\n",
    "warnings.filterwarnings('ignore')\n",
    "\n",
    "SEED = 42\n",
    "#Pandas - Displaying more rorws and columns\n",
    "pd.set_option(\"display.max_rows\", 500)\n",
    "pd.set_option(\"display.max_columns\", 500)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Loading own functions <a class=\"anchor\" id=\"first-bullet\"></a>"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 57,
   "metadata": {},
   "outputs": [],
   "source": [
    "def my_df_summary(data):\n",
    "    '''Own summary function'''\n",
    "    try:\n",
    "        dat = data.copy()\n",
    "        Q1 = dat.quantile(0.25)\n",
    "        Q3 = dat.quantile(0.75)\n",
    "        IQR = Q3 - Q1        \n",
    "    \n",
    "        df = pd.DataFrame([dat.sum(), dat.min(), dat.max(), ((dat < (Q1 - 1.5 * IQR)) | (dat > (Q3 + 1.5 * IQR))).sum(), dat.mean(), dat.std(), dat.median(), dat.count(), dat.isna().sum(), dat.nunique(), dat.dtypes],\n",
    "                     index=['Summe','Minimum', 'Maximum', 'Ausreißer', 'Mittelwert', 'Stand. Abw.', 'Median', 'Anzahl', '#NA', '#Uniques', 'dtypes'])        \n",
    "        print('In total there are {} rows with {} variables in the dataset.'.format(len(data),(len(data.columns))))\n",
    "\n",
    "        return df\n",
    "    except:\n",
    "        print('A summary of the dataset was not possible.')\n",
    "        return data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 58,
   "metadata": {},
   "outputs": [],
   "source": [
    "def plot_num(df):\n",
    "    try:\n",
    "\n",
    "        '''Function for Plotting all numeric columns (histograms and boxplots) for df'''\n",
    "        fig = plt.figure(figsize=(20,50))\n",
    "\n",
    "        df_plot = df.select_dtypes('number')\n",
    "\n",
    "        for i in range(len(df_plot.columns)):\n",
    "            ax = int(i*2)+1\n",
    "            ax1=fig.add_subplot(9,2,ax)\n",
    "            df_plot.iloc[:,i].plot(kind='hist', bins=50,ax=ax1)\n",
    "            plt.title(df_plot.columns[i])\n",
    "            ax = ax+1\n",
    "            ax2=fig.add_subplot(9,2,ax)\n",
    "            df_plot.iloc[:,i].plot(kind='box', ax=ax2)\n",
    "            plt.title(df_plot.columns[i])\n",
    "        return plt.show()  \n",
    "    except ValueError:\n",
    "        pass  # do nothing!    "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 59,
   "metadata": {},
   "outputs": [],
   "source": [
    "def df_norm(df,target_col,force_num=None,norm_target=True):\n",
    "    '''Normalize dataframe df. Target column will be put at the end. force_num can be an additional column which is forced to be numeric. '''\n",
    "    if force_num != None:\n",
    "        df[force_num] = pd.to_numeric(df[force_num], errors='ignore')\n",
    "\n",
    "    cats = df.select_dtypes('object')\n",
    "    nums = df.select_dtypes('number')\n",
    "\n",
    "    from sklearn.preprocessing import MinMaxScaler\n",
    "    minmax = MinMaxScaler()\n",
    "    num_tf = pd.DataFrame(minmax.fit_transform(nums))\n",
    "    num_tf.columns=nums.columns\n",
    "    if norm_target == False:\n",
    "        num_tf.drop(target_col,axis=1,inplace=True)\n",
    "        cats = cats.apply(LabelEncoder().fit_transform)\n",
    "        cats = cats.reset_index(drop=True)\n",
    "        df = pd.concat([cats, num_tf, df[target_col]],axis=1)\n",
    "    else:\n",
    "        cats = cats.apply(LabelEncoder().fit_transform)\n",
    "        cats = cats.reset_index(drop=True)\n",
    "        df = pd.concat([cats,num_tf],axis=1)\n",
    "    return df"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 60,
   "metadata": {},
   "outputs": [],
   "source": [
    "def fill_na_all(df):\n",
    "    '''Fill all na for dataframe df. For numerics take mean, for categories take most frequent.'''\n",
    "    numerics = ['int16', 'int32', 'int64', 'float16', 'float32', 'float64']\n",
    "    \n",
    "    cats = df.select_dtypes(exclude=numerics)\n",
    "    nums = df.select_dtypes(include=numerics)\n",
    "    \n",
    "    for i in nums.columns:\n",
    "        nums[i].fillna(nums[i].mean(),inplace=True)\n",
    "    for n in cats.columns:\n",
    "        cats[n].fillna(cats[n].mode()[0])\n",
    "    cats = cats.reset_index(drop=True)\n",
    "    nums = nums.reset_index(drop=True)\n",
    "    df = pd.concat([cats,nums],axis=1)\n",
    "    return df"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 61,
   "metadata": {},
   "outputs": [],
   "source": [
    "# ignoring chained assignments error\n",
    "pd.options.mode.chained_assignment = None  # default='warn'"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 62,
   "metadata": {
    "scrolled": false
   },
   "outputs": [],
   "source": [
    "#!kaggle competitions download -c titanic"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 63,
   "metadata": {},
   "outputs": [],
   "source": [
    "train_data = pd.read_csv('train.csv')\n",
    "test_data = pd.read_csv('test.csv')\n",
    "df = pd.concat([train_data, test_data],axis=0,sort=False)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 64,
   "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>0.0</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>1.0</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>1.0</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>1.0</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>0.0</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",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>413</th>\n",
       "      <td>1305</td>\n",
       "      <td>NaN</td>\n",
       "      <td>3</td>\n",
       "      <td>Spector, Mr. Woolf</td>\n",
       "      <td>male</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>A.5. 3236</td>\n",
       "      <td>8.0500</td>\n",
       "      <td>NaN</td>\n",
       "      <td>S</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>414</th>\n",
       "      <td>1306</td>\n",
       "      <td>NaN</td>\n",
       "      <td>1</td>\n",
       "      <td>Oliva y Ocana, Dona. Fermina</td>\n",
       "      <td>female</td>\n",
       "      <td>39.0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>PC 17758</td>\n",
       "      <td>108.9000</td>\n",
       "      <td>C105</td>\n",
       "      <td>C</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>415</th>\n",
       "      <td>1307</td>\n",
       "      <td>NaN</td>\n",
       "      <td>3</td>\n",
       "      <td>Saether, Mr. Simon Sivertsen</td>\n",
       "      <td>male</td>\n",
       "      <td>38.5</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>SOTON/O.Q. 3101262</td>\n",
       "      <td>7.2500</td>\n",
       "      <td>NaN</td>\n",
       "      <td>S</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>416</th>\n",
       "      <td>1308</td>\n",
       "      <td>NaN</td>\n",
       "      <td>3</td>\n",
       "      <td>Ware, Mr. Frederick</td>\n",
       "      <td>male</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>359309</td>\n",
       "      <td>8.0500</td>\n",
       "      <td>NaN</td>\n",
       "      <td>S</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>417</th>\n",
       "      <td>1309</td>\n",
       "      <td>NaN</td>\n",
       "      <td>3</td>\n",
       "      <td>Peter, Master. Michael J</td>\n",
       "      <td>male</td>\n",
       "      <td>NaN</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>2668</td>\n",
       "      <td>22.3583</td>\n",
       "      <td>NaN</td>\n",
       "      <td>C</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>1309 rows × 12 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "     PassengerId  Survived  Pclass  \\\n",
       "0              1       0.0       3   \n",
       "1              2       1.0       1   \n",
       "2              3       1.0       3   \n",
       "3              4       1.0       1   \n",
       "4              5       0.0       3   \n",
       "..           ...       ...     ...   \n",
       "413         1305       NaN       3   \n",
       "414         1306       NaN       1   \n",
       "415         1307       NaN       3   \n",
       "416         1308       NaN       3   \n",
       "417         1309       NaN       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",
       "413                                 Spector, Mr. Woolf    male   NaN      0   \n",
       "414                       Oliva y Ocana, Dona. Fermina  female  39.0      0   \n",
       "415                       Saether, Mr. Simon Sivertsen    male  38.5      0   \n",
       "416                                Ware, Mr. Frederick    male   NaN      0   \n",
       "417                           Peter, Master. Michael J    male   NaN      1   \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  \n",
       "..     ...                 ...       ...   ...      ...  \n",
       "413      0           A.5. 3236    8.0500   NaN        S  \n",
       "414      0            PC 17758  108.9000  C105        C  \n",
       "415      0  SOTON/O.Q. 3101262    7.2500   NaN        S  \n",
       "416      0              359309    8.0500   NaN        S  \n",
       "417      1                2668   22.3583   NaN        C  \n",
       "\n",
       "[1309 rows x 12 columns]"
      ]
     },
     "execution_count": 64,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Data Overview  <a class=\"anchor\" id=\"second-bullet\"></a>"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 65,
   "metadata": {
    "scrolled": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "In total there are 1309 rows with 12 variables in the dataset.\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>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>Summe</th>\n",
       "      <td>857395</td>\n",
       "      <td>342</td>\n",
       "      <td>3004</td>\n",
       "      <td>Braund, Mr. Owen HarrisCumings, Mrs. John Brad...</td>\n",
       "      <td>malefemalefemalefemalemalemalemalemalefemalefe...</td>\n",
       "      <td>31255.7</td>\n",
       "      <td>653</td>\n",
       "      <td>504</td>\n",
       "      <td>A/5 21171PC 17599STON/O2. 31012821138033734503...</td>\n",
       "      <td>43550.5</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Minimum</th>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>Abbing, Mr. Anthony</td>\n",
       "      <td>female</td>\n",
       "      <td>0.17</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>110152</td>\n",
       "      <td>0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Maximum</th>\n",
       "      <td>1309</td>\n",
       "      <td>1</td>\n",
       "      <td>3</td>\n",
       "      <td>van Melkebeke, Mr. Philemon</td>\n",
       "      <td>male</td>\n",
       "      <td>80</td>\n",
       "      <td>8</td>\n",
       "      <td>9</td>\n",
       "      <td>WE/P 5735</td>\n",
       "      <td>512.329</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Ausreißer</th>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>9</td>\n",
       "      <td>57</td>\n",
       "      <td>307</td>\n",
       "      <td>0</td>\n",
       "      <td>171</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Mittelwert</th>\n",
       "      <td>655</td>\n",
       "      <td>0.383838</td>\n",
       "      <td>2.29488</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>29.8811</td>\n",
       "      <td>0.498854</td>\n",
       "      <td>0.385027</td>\n",
       "      <td>NaN</td>\n",
       "      <td>33.2955</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Stand. Abw.</th>\n",
       "      <td>378.02</td>\n",
       "      <td>0.486592</td>\n",
       "      <td>0.837836</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>14.4135</td>\n",
       "      <td>1.04166</td>\n",
       "      <td>0.86556</td>\n",
       "      <td>NaN</td>\n",
       "      <td>51.7587</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Median</th>\n",
       "      <td>655</td>\n",
       "      <td>0</td>\n",
       "      <td>3</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>28</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>14.4542</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Anzahl</th>\n",
       "      <td>1309</td>\n",
       "      <td>891</td>\n",
       "      <td>1309</td>\n",
       "      <td>1309</td>\n",
       "      <td>1309</td>\n",
       "      <td>1046</td>\n",
       "      <td>1309</td>\n",
       "      <td>1309</td>\n",
       "      <td>1309</td>\n",
       "      <td>1308</td>\n",
       "      <td>295</td>\n",
       "      <td>1307</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>#NA</th>\n",
       "      <td>0</td>\n",
       "      <td>418</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>263</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>1014</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>#Uniques</th>\n",
       "      <td>1309</td>\n",
       "      <td>2</td>\n",
       "      <td>3</td>\n",
       "      <td>1307</td>\n",
       "      <td>2</td>\n",
       "      <td>98</td>\n",
       "      <td>7</td>\n",
       "      <td>8</td>\n",
       "      <td>929</td>\n",
       "      <td>281</td>\n",
       "      <td>186</td>\n",
       "      <td>3</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>dtypes</th>\n",
       "      <td>int64</td>\n",
       "      <td>float64</td>\n",
       "      <td>int64</td>\n",
       "      <td>object</td>\n",
       "      <td>object</td>\n",
       "      <td>float64</td>\n",
       "      <td>int64</td>\n",
       "      <td>int64</td>\n",
       "      <td>object</td>\n",
       "      <td>float64</td>\n",
       "      <td>object</td>\n",
       "      <td>object</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "            PassengerId  Survived    Pclass  \\\n",
       "Summe            857395       342      3004   \n",
       "Minimum               1         0         1   \n",
       "Maximum            1309         1         3   \n",
       "Ausreißer             0         0         0   \n",
       "Mittelwert          655  0.383838   2.29488   \n",
       "Stand. Abw.      378.02  0.486592  0.837836   \n",
       "Median              655         0         3   \n",
       "Anzahl             1309       891      1309   \n",
       "#NA                   0       418         0   \n",
       "#Uniques           1309         2         3   \n",
       "dtypes            int64   float64     int64   \n",
       "\n",
       "                                                          Name  \\\n",
       "Summe        Braund, Mr. Owen HarrisCumings, Mrs. John Brad...   \n",
       "Minimum                                    Abbing, Mr. Anthony   \n",
       "Maximum                            van Melkebeke, Mr. Philemon   \n",
       "Ausreißer                                                    0   \n",
       "Mittelwert                                                 NaN   \n",
       "Stand. Abw.                                                NaN   \n",
       "Median                                                     NaN   \n",
       "Anzahl                                                    1309   \n",
       "#NA                                                          0   \n",
       "#Uniques                                                  1307   \n",
       "dtypes                                                  object   \n",
       "\n",
       "                                                           Sex      Age  \\\n",
       "Summe        malefemalefemalefemalemalemalemalemalefemalefe...  31255.7   \n",
       "Minimum                                                 female     0.17   \n",
       "Maximum                                                   male       80   \n",
       "Ausreißer                                                    0        9   \n",
       "Mittelwert                                                 NaN  29.8811   \n",
       "Stand. Abw.                                                NaN  14.4135   \n",
       "Median                                                     NaN       28   \n",
       "Anzahl                                                    1309     1046   \n",
       "#NA                                                          0      263   \n",
       "#Uniques                                                     2       98   \n",
       "dtypes                                                  object  float64   \n",
       "\n",
       "                SibSp     Parch  \\\n",
       "Summe             653       504   \n",
       "Minimum             0         0   \n",
       "Maximum             8         9   \n",
       "Ausreißer          57       307   \n",
       "Mittelwert   0.498854  0.385027   \n",
       "Stand. Abw.   1.04166   0.86556   \n",
       "Median              0         0   \n",
       "Anzahl           1309      1309   \n",
       "#NA                 0         0   \n",
       "#Uniques            7         8   \n",
       "dtypes          int64     int64   \n",
       "\n",
       "                                                        Ticket     Fare  \\\n",
       "Summe        A/5 21171PC 17599STON/O2. 31012821138033734503...  43550.5   \n",
       "Minimum                                                 110152        0   \n",
       "Maximum                                              WE/P 5735  512.329   \n",
       "Ausreißer                                                    0      171   \n",
       "Mittelwert                                                 NaN  33.2955   \n",
       "Stand. Abw.                                                NaN  51.7587   \n",
       "Median                                                     NaN  14.4542   \n",
       "Anzahl                                                    1309     1308   \n",
       "#NA                                                          0        1   \n",
       "#Uniques                                                   929      281   \n",
       "dtypes                                                  object  float64   \n",
       "\n",
       "              Cabin Embarked  \n",
       "Summe           NaN      NaN  \n",
       "Minimum         NaN      NaN  \n",
       "Maximum         NaN      NaN  \n",
       "Ausreißer         0        0  \n",
       "Mittelwert      NaN      NaN  \n",
       "Stand. Abw.     NaN      NaN  \n",
       "Median          NaN      NaN  \n",
       "Anzahl          295     1307  \n",
       "#NA            1014        2  \n",
       "#Uniques        186        3  \n",
       "dtypes       object   object  "
      ]
     },
     "execution_count": 65,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "my_df_summary(df)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "There are 891 people in the dataset with 342 of them have survived the incident.\n",
    "\n",
    "In total the 12 different variables are:\n",
    "\n",
    "|Variable|Definition|Key|\n",
    "|:---|:---|:---|\n",
    "|Passengerid| Unique ID for each passenger | |\n",
    "|survival| \tSurvival of Person| \t0 = No, 1 = Yes|\n",
    "|pclass| \tTicket class | 1 = 1st, 2 = 2nd, 3 = 3rd|\n",
    "|Name| Passenger Name including title| |\n",
    "|sex| \tSex| |\n",
    "|Age| \tAge in years | |\n",
    "|sibsp| # of siblings / spouses aboard the Titanic | |\n",
    "|parch|  of parents / children aboard the Titanic |\t|\n",
    "|ticket| \tTicket number | \t|\n",
    "|fare| \tPassenger fare \t|\n",
    "|cabin| \tCabin number |\t\n",
    "|embarked| \tPort of Embarkation| \tC = Cherbourg, Q = Queenstown, S = Southampton|"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 66,
   "metadata": {
    "scrolled": false
   },
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAABJUAAAhmCAYAAABBku4dAAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADh0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uMy4xLjEsIGh0dHA6Ly9tYXRwbG90bGliLm9yZy8QZhcZAAAgAElEQVR4nOzdebRlZXkv6t8rJbYIKCVBoCi9clHjPSIpjQkeY8QONGISzcF0xHBSmqMxRu+NpWls0gy8aTRGjwbFiL0GRYkQhdjGXLsCOxSNFUQpIQLSqWCDee8fa+64LXZV7VnstdfeVc8zxh5rzm9+c6531xqr6qvf+ua3qrsDAAAAAGPcYtYFAAAAALD6CJUAAAAAGE2oBAAAAMBoQiUAAAAARhMqAQAAADCaUAkAAACA0YRKADdTVX2gqv7nrOsAANhTGH/ByiBUAkapqour6oaq+lZVfb2q/r6qbj/ruqapqp5XVa+fdR0AwJ7J+AtYqYRKwK74ue6+fZKjktwvyR/OuJ6pqao1s64BACDGX8AKJFQCdll3fy3JPyW5d1U9saourKpvVtVFVfWkuX5VdUBVvauqrqmqq6rqX6rqFsOxZ1XV14bzvlhVxwztt6iqTVX171X1jap6a1XdcTi2vqq6qk6sqq9W1ZVV9Qfznu82VXVaVV091PT7VbV13vG7VNXbquqKqvpyVT1t3rHnVdXpVfX6qrouyW9s+3tX1cOq6gtVdW1VvTRJLfkfLgDAAoy/jL9gJREqAbusqg5NclySTya5PMmjk9whyROTvKiqjhq6PjPJ1iRrkxyY5DlJuqqOSPLUJPfr7n2SPCLJxcM5T0vy2CQ/k+QuSa5O8rJtSnhgkiOSHJPkj6vqnkP7c5OsT3K3JA9L8qvzar5Fkn9M8ukkBw/nPr2qHjHvuscnOT3JfknesM3vfECSt2Xy6eABSf49ydE7/9MCALj5jL+Mv2AlESoBu+IdVXVNkg8n+WCSP+/us7r733vig0nOSfLfh/7fT3JQksO6+/vd/S/d3Ul+kORWSe5VVbfs7ou7+9+Hc56U5A+6e2t3fzfJ85I8bpvp0M/v7hu6+9OZDFLuM7T/0lDT1d29NclL5p1zvyRru/sF3f297r4oySuTnDCvz0e6+x3d/Z/dfcM2v/txST7f3ad39/eTvDjJf+zCnyEAwBjGX8ZfsOIIlYBd8dju3q+7D+vu/9XdN1TVsVX10WF69TWZ/ON/wND/L5JsSXLOMDV7U5J095YkT89kwHJ5Vb25qu4ynHNYkjOGKdvXJLkwk0HQgfPqmD+YuD7J3IKVd0lyybxj87cPS3KXuesO137ONted339bP3LtYXC2o/4AAEvB+Gtg/AUrh1AJuNmq6laZTEn+yyQHdvd+Sc7OcK97d3+zu5/Z3XdL8nNJnjF37353v7G7H5jJYKOTvHC47CVJjh0GT3M/tx7WEdiZy5IcMm//0HnblyT58jbX3ae7j5vXp3dy7f+6XlXVNtcHAJg64y/jL1gJhErAUtg7k2nUVyS5saqOTfLwuYNV9eiquvswALguk0+8flBVR1TVQ4ZB0XeS3DAcS5JXJPmzqjpsuMbaqjp+kfW8Ncmzq2r/qjo4k3UD5nw8yXXDApW3qaq9qureVXW/RV77rCQ/XlW/MEwFf1qSH1vkuQAAS8X4C5g5oRJws3X3NzP5x/2tmSzo+MtJzpzX5fAk/5zkW0k+kuR/d/cHMhkInZzkykymUt85k6nQSfI3wzXOqapvJvlokp9cZEkvyGRhyi8Pz3t6ku8Otf4gk0/rjhyOX5nkVUn2XeTvemWSxw91f2P43f51kXUBACwJ4y/jL1gJanI7KsDuq6p+O8kJ3f0zs64FAGBPYPwFewYzlYDdTlUdVFVHV9Uthq/NfWaSM2ZdFwDA7sr4C/ZMa3beBWDV2TvJ3yW5a5Jrkrw5yf+eaUUAALs34y/YA7n9DQAAAIDR3P4GAAAAwGhCJQAAAABG263WVDrggAN6/fr1sy4DAJiS884778ruXjvrOvgh4y8A2P1tbwy2W4VK69evz+bNm2ddBgAwJVX1lVnXwI8y/gKA3d/2xmBufwMAAABgNKESAAAAAKMJlQAAAAAYTagEAAAAwGhCJQAAAABGEyoBAAAAMJpQCQAAAIDRhEoAAAAAjCZUAgAAAGA0oRIAAAAAo62ZdQGrxfpNZ826hJu4+ORH7bSPupeOupeXupfX7lx3svJqX611L9Zifz8Alk9VzbqE7eruWZcA7CIzlQAAAHZz3b1kP4c9611Lej1g9RIqAQAAADDa1G5/q6pXJ3l0ksu7+95D21uSHDF02S/JNd195ALnXpzkm0l+kOTG7t4wrToBAAAAGG+aayq9JslLk7x2rqG7/8fcdlX9VZJrd3D+z3b3lVOrDgAAAIBdNrVQqbs/VFXrFzpWk1XifinJQ6b1/AAAAABMz6zWVPrvSb7e3V/azvFOck5VnVdVG5exLgAAAAAWYZq3v+3IE5K8aQfHj+7uS6vqzknOraovdPeHFuo4hE4bk2TdunVLXykAAAAAN7HsM5Wqak2SX0jylu316e5Lh8fLk5yR5P476HtKd2/o7g1r165d6nIBAAAAWMAsbn97aJIvdPfWhQ5W1e2qap+57SQPT3LBMtYHAAAAwE5MLVSqqjcl+UiSI6pqa1WdNBw6Idvc+lZVd6mqs4fdA5N8uKo+neTjSc7q7ndPq04AAAAAxpvmt789YTvtv7FA26VJjhu2L0pyn2nVBQAAAMDNN6tvfwMAAABgFRMqAQAAADBadfesa1gyGzZs6M2bN0/l2us3nTWV6wLA7ubikx81tWtX1XndvWFqT8Bo0xx/ASvT+k1nTfXvemDl2d4YzEwlAAAAAEYTKgEAAAAwmlAJAAAAgNGESgAAAACMJlQCAAAAYDShEgDAbqyqfq+qPldVF1TVm6rq1lV116r6WFV9qareUlV7D31vNexvGY6vn231AMBKJlQCANhNVdXBSZ6WZEN33zvJXklOSPLCJC/q7sOTXJ3kpOGUk5Jc3d13T/KioR8AwIKESgAAu7c1SW5TVWuS3DbJZUkekuT04fhpSR47bB8/7Gc4fkxV1TLWCgCsIkIlAIDdVHd/LclfJvlqJmHStUnOS3JNd984dNua5OBh++Aklwzn3jj0v9Ny1gwArB5CJQCA3VRV7Z/J7KO7JrlLktslOXaBrj13yg6Ozb/uxqraXFWbr7jiiqUqFwBYZYRKAAC7r4cm+XJ3X9Hd30/y9iQ/nWS/4Xa4JDkkyaXD9tYkhybJcHzfJFdte9HuPqW7N3T3hrVr1077dwAAViihEgDA7uurSR5QVbcd1kY6Jsnnk7w/yeOGPicmeeewfeawn+H4+7r7JjOVAAASoRIAwG6ruz+WyYLb5yf5bCZjv1OSPCvJM6pqSyZrJp06nHJqkjsN7c9IsmnZiwYAVo01O+8CAMBq1d3PTfLcbZovSnL/Bfp+J8njl6MuAGD1M1MJAAAAgNGESgAAAACMJlQCAAAAYDShEgAAAACjCZUAAAAAGE2oBAAAAMBoQiUAAAAARhMqAQAAADDa1EKlqnp1VV1eVRfMa3teVX2tqj41/By3nXMfWVVfrKotVbVpWjUCAAAAsGumOVPpNUkeuUD7i7r7yOHn7G0PVtVeSV6W5Ngk90ryhKq61xTrBAAAAGCkqYVK3f2hJFftwqn3T7Kluy/q7u8leXOS45e0OAAAAABullmsqfTUqvrMcHvc/gscPzjJJfP2tw5tAAAAAKwQyx0qvTzJ/5HkyCSXJfmrBfrUAm29vQtW1caq2lxVm6+44oqlqRIAAACAHVrWUKm7v97dP+ju/0zyykxuddvW1iSHzts/JMmlO7jmKd29obs3rF27dmkLBgAAAGBByxoqVdVB83Z/PskFC3T7RJLDq+quVbV3khOSnLkc9QEAAACwOGumdeGqelOSByc5oKq2JnlukgdX1ZGZ3M52cZInDX3vkuRV3X1cd99YVU9N8p4keyV5dXd/blp1AgAAADDe1EKl7n7CAs2nbqfvpUmOm7d/dpKzp1QaAAAAADfTLL79DQAAAIBVTqgEAAAAwGhCJQAAAABGEyoBAAAAMJpQCQAAAIDRhEoAAAAAjCZUAgAAAGA0oRIAAAAAowmVAAAAABhNqAQAAADAaEIlAAAAAEYTKgEAAAAw2ppZFwAAAMBN3ef55+TaG74/6zIWtH7TWbMu4Sb2vc0t8+nnPnzWZcAeRagEAACwAl17w/dz8cmPmnUZq8ZKDLpgd+f2NwAAAABGEyoBAAAAMJpQCQAAAIDRhEoAAAAAjCZUAgAAAGA0oRIAAAAAowmVAAAAABhNqAQAAADAaEIlAAAAAEYTKgEAAAAwmlAJAAAAgNGmFipV1aur6vKqumBe219U1Req6jNVdUZV7bedcy+uqs9W1aeqavO0agQAAABg10xzptJrkjxym7Zzk9y7u/9bkn9L8uwdnP+z3X1kd2+YUn0AALu9qtqvqk4fPti7sKp+qqruWFXnVtWXhsf9h75VVS+pqi3Dh4BHzbp+AGDlmlqo1N0fSnLVNm3ndPeNw+5HkxwyrecHACBJ8jdJ3t3d90hynyQXJtmU5L3dfXiS9w77SXJsksOHn41JXr785QIAq8Us11T6zST/tJ1jneScqjqvqjYuY00AALuNqrpDkgclOTVJuvt73X1NkuOTnDZ0Oy3JY4ft45O8tic+mmS/qjpomcsGAFaJmYRKVfUHSW5M8obtdDm6u4/K5NOyp1TVg3ZwrY1VtbmqNl9xxRVTqBYAYNW6W5Irkvx9VX2yql5VVbdLcmB3X5Ykw+Odh/4HJ7lk3vlbhzYAgJtY9lCpqk5M8ugkv9LdvVCf7r50eLw8yRlJ7r+963X3Kd29obs3rF27dholAwCsVmuSHJXk5d193yTfzg9vdVtILdB2k/GaD/UAgGSZQ6WqemSSZyV5THdfv50+t6uqfea2kzw8yQUL9QUAYIe2Jtna3R8b9k/PJGT6+txtbcPj5fP6Hzrv/EOSXLrtRX2oBwAkUwyVqupNST6S5Iiq2lpVJyV5aZJ9kpxbVZ+qqlcMfe9SVWcPpx6Y5MNV9ekkH09yVne/e1p1AgDsrrr7P5JcUlVHDE3HJPl8kjOTnDi0nZjkncP2mUl+ffgWuAckuXbuNjkAgG2tmdaFu/sJCzSfup2+lyY5bti+KJNvJgEA4Ob7nSRvqKq9k1yU5ImZfLD41uFDv68mefzQ9+xMxmRbklw/9AUAWNDUQiUAAGavuz+VZMMCh45ZoG8necrUiwIAdgsz+fY3AAAAAFY3oRIAAAAAowmVAAAAABhNqAQAAADAaEIlAAAAAEYTKgEAAAAwmlAJAAAAgNGESgAAAACMJlQCAAAAYDShEgAAAACjCZUAAAAAGG1RoVJV3XvahQAAAACweix2ptIrqurjVfW/qmq/qVYEAAAAwIq3qFCpux+Y5FeSHJpkc1W9saoeNtXKAAAAAFixFr2mUnd/KckfJnlWkp9J8pKq+kJV/cK0igMAAABgZVrsmkr/rapelOTCJA9J8nPdfc9h+0VTrA8AAACAFWjNIvu9NMkrkzynu2+Ya+zuS6vqD6dSGQAAAAAr1mJDpeOS3NDdP0iSqrpFklt39/Xd/bqpVQcAAADAirTYNZX+Oclt5u3fdmgDAAAAYA+02FDp1t39rbmdYfu20ykJAAAAgJVusaHSt6vqqLmdqvqJJDfsoD8AAAAAu7HFrqn09CT/UFWXDvsHJfkf0ykJAAAAgJVuUaFSd3+iqu6R5IgkleQL3f39qVYGAAAAwIq12JlKSXK/JOuHc+5bVenu106lKgAAAABWtEWtqVRVr0vyl0kemEm4dL8kGxZx3qur6vKqumBe2x2r6tyq+tLwuP92zj1x6POlqjpxUb8NAAAAAMtisTOVNiS5V3f3yOu/JslLk8yf0bQpyXu7++Sq2jTsP2v+SVV1xyTPHZ63k5xXVWd299Ujnx8AAACAKVjst79dkOTHxl68uz+U5Kptmo9PctqwfVqSxy5w6iOSnNvdVw1B0rlJHjn2+QEAAACYjsXOVDogyeer6uNJvjvX2N2P2YXnPLC7LxvOv6yq7rxAn4OTXDJvf+vQBgAAAMAKsNhQ6XnTLGIBtUDbgrfeVdXGJBuTZN26ddOsCQAAAIDBom5/6+4PJrk4yS2H7U8kOX8Xn/PrVXVQkgyPly/QZ2uSQ+ftH5Lk0u3Udkp3b+juDWvXrt3FkgAAAAAYY7Hf/vZbSU5P8ndD08FJ3rGLz3lmkrlvczsxyTsX6POeJA+vqv2Hb4d7+NAGAAAAwAqw2IW6n5Lk6CTXJUl3fynJQmsh/YiqelOSjyQ5oqq2VtVJSU5O8rCq+lKShw37qaoNVfWq4fpXJfmTTGZEfSLJC4Y2AAAAAFaAxa6p9N3u/l7VZKmjqlqT7axxNF93P2E7h45ZoO/mJP9z3v6rk7x6kfUBAAAAsIwWO1Ppg1X1nCS3qaqHJfmHJP84vbIAAAAAWMkWGyptSnJFks8meVKSs5P84bSKAgAAAGBlW9Ttb939n0leOfwAAAAwZfvcc1P+r9M2zbqMVWOfeybJo2ZdBuxRFhUqVdWXs8AaSt19tyWvCAAAgHzzwpNz8clCksVav+msWZcAe5zFLtS9Yd72rZM8Pskdl74cAAAAAFaDRa2p1N3fmPfzte5+cZKHTLk2AAAAAFaoxd7+dtS83VtkMnNpn6lUBAAAAMCKt9jb3/5q3vaNSS5O8ktLXg0AAAAAq8Jiv/3tZ6ddCAAAAACrx2Jvf3vGjo53918vTTkAAAAArAaLWqg7kzWUfjvJwcPPk5PcK5N1laytBACwglXVXlX1yap617B/16r6WFV9qareUlV7D+23Gva3DMfXz7JuAGBlW2yodECSo7r7md39zCQ/keSQ7n5+dz9/euUBALAEfjfJhfP2X5jkRd19eJKrk5w0tJ+U5OruvnuSFw39AAAWtNhQaV2S783b/16S9UteDQAAS6qqDknyqCSvGvYryUOSnD50OS3JY4ft44f9DMePGfoDANzEYr/97XVJPl5VZyTpJD+f5LVTqwoAgKXy4iS/nx8uWXCnJNd0943D/tZMljfI8HhJknT3jVV17dD/yuUrFwBYLRY1U6m7/yzJEzOZHn1Nkid2959PszAAAG6eqnp0ksu7+7z5zQt07UUcm3/djVW1uao2X3HFFUtQKQCwGi329rckuW2S67r7b5Jsraq7TqkmAACWxtFJHlNVFyd5cya3vb04yX5VNTdj/ZAklw7bW5McmiTD8X2TXLXtRbv7lO7e0N0b1q5dO93fAABYsRYVKlXVc5M8K8mzh6ZbJnn9tIoCAODm6+5nd/ch3b0+yQlJ3tfdv5Lk/UkeN3Q7Mck7h+0zh/0Mx9/X3TeZqQQAkCx+ptLPJ3lMkm8nSXdfmh/elw8AwOryrCTPqKotmayZdOrQfmqSOw3tz0iyaUb1AQCrwGIX6v5ed3dVdZJU1e2mWBMAAEusuz+Q5APD9kVJ7r9An+8kefyyFgYArFqLnan01qr6u0zuv/+tJP+c5JXTKwsAAACAlWxRM5W6+y+r6mFJrktyRJI/7u5zp1oZAAAAACvWTkOlqtoryXu6+6FJBEkAAAAA7Pz2t+7+QZLrq2rfZagHAAAAgFVgsQt1fyfJZ6vq3AzfAJck3f20qVQFAAAAwIq22FDprOEHAAAAAHYcKlXVuu7+aneftlRPWFVHJHnLvKa7ZbLw94vn9Xlwkncm+fLQ9PbufsFS1QAAAADAzbOzmUrvSHJUklTV27r7F2/uE3b3F5McOVxzryRfS3LGAl3/pbsffXOfDwAAAIClt7OFumve9t2m8PzHJPn37v7KFK4NAAAAwJTsLFTq7WwvlROSvGk7x36qqj5dVf9UVT8+hecGAAAAYBft7Pa3+1TVdZnMWLrNsJ1hv7v7Drv6xFW1d5LHJHn2AofPT3JYd3+rqo7L5Da8w7dznY1JNibJunXrdrUcAAAAAEbY4Uyl7t6ru+/Q3ft095phe25/lwOlwbFJzu/ury/wvNd197eG7bOT3LKqDthOjad094bu3rB27dqbWRIAAAAAi7Gz29+m6QnZzq1vVfVjVVXD9v0zqfMby1gbAAAAADuws9vfpqKqbpvkYUmeNK/tyUnS3a9I8rgkv11VNya5IckJ3T2NNZ0AAAAA2AUzCZW6+/okd9qm7RXztl+a5KXLXRcAAAAAizPL298AAAAAWKWESgAAAACMJlQCAAAAYDShEgAAAACjCZUAAAAAGE2oBAAAAMBoQiUAAAAARhMqAQAAADCaUAkAAACA0YRKAAAAAIwmVAIAAABgNKESAAAAAKMJlQAAAAAYTagEAAAAwGhCJQAAAABGEyoBAAAAMJpQCQAAAIDRhEoAAAAAjCZUAgAAAGA0oRIAAAAAowmVAAAAABhNqAQAAADAaEIlAAAAAEYTKgEAAAAwmlAJAAAAgNFmFipV1cVV9dmq+lRVbV7geFXVS6pqS1V9pqqOmkWdAAAAANzUmhk//89295XbOXZsksOHn59M8vLhEQAAAIAZm3WotCPHJ3ltd3eSj1bVflV1UHdfNuvCAAAAlsP6TWfNuoRVY9/b3HLWJcAeZ5ahUic5p6o6yd919ynbHD84ySXz9rcObUIlAABgt3fxyY+adQkLWr/prBVbG7C8ZrlQ99HdfVQmt7k9paoetM3xWuCc3rahqjZW1eaq2nzFFVdMo04AgFWpqg6tqvdX1YVV9bmq+t2h/Y5VdW5VfWl43H9ot6YlALBoMwuVuvvS4fHyJGckuf82XbYmOXTe/iFJLl3gOqd094bu3rB27dpplQsAsBrdmOSZ3X3PJA/I5IO8eyXZlOS93X14kvcO+8mPrmm5MZM1LQEAFjSTUKmqbldV+8xtJ3l4kgu26XZmkl8fPjF7QJJrracEALB43X1Zd58/bH8zyYWZLCdwfJLThm6nJXnssP1fa1p290eT7FdVBy1z2QDAKjGrNZUOTHJGVc3V8MbufndVPTlJuvsVSc5OclySLUmuT/LEGdUKALDqVdX6JPdN8rEkB859WNfdl1XVnYdu1rQEABZtJqFSd1+U5D4LtL9i3nYnecpy1gUAsDuqqtsneVuSp3f3dcMHewt2XaBtwTUtM7k9LuvWrVuqMgGAVWaWC3UDADBlVXXLTAKlN3T324fmr8/d1jY8Xj60W9MSAFg0oRIAwG6qJlOSTk1yYXf/9bxDZyY5cdg+Mck757Vb0xIAWJRZrakEAMD0HZ3k15J8tqo+NbQ9J8nJSd5aVScl+WqSxw/HrGkJACyaUAkAYDfV3R/OwuskJckxC/S3piUAsGhufwMAAABgNKESAAAAAKMJlQAAAAAYTagEAAAAwGhCJQAAAABGEyoBAAAAMJpQCQAAAIDRhEoAAAAAjCZUAgAAAGA0oRIAAAAAowmVAAAAABhNqAQAAADAaEIlAAAAAEYTKgEAAAAwmlAJAAAAgNGESgAAAACMJlQCAAAAYDShEgAAAACjCZUAAAAAGE2oBAAAAMBoQiUAAAAARlv2UKmqDq2q91fVhVX1uar63QX6PLiqrq2qTw0/f7zcdQIAAACwfWtm8Jw3Jnlmd59fVfskOa+qzu3uz2/T71+6+9EzqA8AAACAnVj2mUrdfVl3nz9sfzPJhUkOXu46AAAAANh1M11TqarWJ7lvko8tcPinqurTVfVPVfXjy1oYAAAAADs0i9vfkiRVdfskb0vy9O6+bpvD5yc5rLu/VVXHJXlHksO3c52NSTYmybp166ZYMQAAAABzZjJTqapumUmg9Ibufvu2x7v7uu7+1rB9dpJbVtUBC12ru0/p7g3dvWHt2rVTrRsAAACAiVl8+1slOTXJhd3919vp82NDv1TV/TOp8xvLVyUAAAAAOzKL29+OTvJrST5bVZ8a2p6TZF2SdPcrkjwuyW9X1Y1JbkhyQnf3DGoFAAAAYAHLHip194eT1E76vDTJS5enIgAAAADGmum3vwEAAACwOgmVAAAAABhNqAQAAADAaEIlAAAAAEYTKgEAAAAwmlAJAAAAgNGESgAAAACMJlQCAAAAYDShEgAAAACjCZUAAAAAGE2oBAAAAMBoQiUAAAAARhMqAQAAADCaUAkAgB9RVY+sqi9W1Zaq2jTregCAlUmoBADAf6mqvZK8LMmxSe6V5AlVda/ZVgUArERCJQAA5rt/ki3dfVF3fy/Jm5McP+OaAIAVSKgEAMB8Bye5ZN7+1qENAOBHrJl1AQAArCi1QFv/SIeqjUk2Jsm6deuWoybgZqpa6K19M673wqW7VnfvvBOwIgmVAACYb2uSQ+ftH5Lk0vkduvuUJKckyYYNG/xvEFYBwQ0wDW5/AwBgvk8kObyq7lpVeyc5IcmZM64JAFiBzFQCAOC/dPeNVfXUJO9JsleSV3f352ZcFgCwAgmVAAD4Ed19dpKzZ10HALCyuf0NAAAAgNGESgAAAACMJlQCAAAAYLSZhEpV9ciq+mJVbamqTQscv1VVvWU4/rGqWr/8VQIAAACwPcseKlXVXkleluTYJPdK8oSqutc23U5KcnV33z3Ji5K8cHmrBAAAAGBHZjFT6f5JtnT3Rd39vSRvTnL8Nn2OT3LasH16kmOqqpaxRgAAAAB2YBah0sFJLpm3v3VoW7BPd9+Y5Nokd1qW6gAAAADYqTUzeM6FZhz1LvSZdKzamGTjsPutqvrizahtew5IcuUUrsvS8PqsXF6blc3rs7Kt2tenpnvT+mFTvTqjnXfeeVdW1VdmXQewrFbtv1HALltwDDaLUGlrkkPn7R+S5NLt9NlaVWuS7JvkqoUu1t2nJDllCnX+l6ra3N0bpvkc7Dqvz8rltVnZvD4rm9eH1aK71866BmB5+TcKmDOL298+keTwqrprVe2d5IQkZ27T58wkJw7bj0vyvu5ecKYSAAAAAMtv2WcqdfeNVfXUJO9JsleSV3f356rqBUk2d/eZSU5N8rqq2pLJDKUTlrtOAAAAALZvFre/pbvPTnL2Nm1/PG/7O0kev9x17cBUb6/jZvP6rFxem5XN67OyeX0AWKn8GwUkScpdZQ7eyGEAACAASURBVAAAAACMNYs1lQAAAABY5YRKO1FVj6yqL1bVlqraNOt69jRVdWhVvb+qLqyqz1XV7w7td6yqc6vqS8Pj/kN7VdVLhtfrM1V11Gx/g91fVe1VVZ+sqncN+3etqo8Nr81bhgX5U1W3Gva3DMfXz7LuPUFV7VdVp1fVF4b30E9576wcVfV7w99rF1TVm6rq1t4/AIxRVT+oqk8N/5b8Q1XddtY17aqqevDceHKBYxdX1QHLXROwc0KlHaiqvZK8LMmxSe6V5AlVda/ZVrXHuTHJM7v7nkkekOQpw2uwKcl7u/vwJO8d9pPJa3X48LMxycuXv+Q9zu8muXDe/guTvGh4ba5OctLQflKSq7v77kleNPRjuv4mybu7+x5J7pPJ6+S9swJU1cFJnpZkQ3ffO5Mvrjgh3j8AjHNDdx85/FvyvSRPnnVBu6KqZrLWL3DzCZV27P5JtnT3Rd39vSRvTnL8jGvao3T3Zd19/rD9zUz+U3xwJq/DaUO305I8dtg+Pslre+KjSfarqoOWuew9RlUdkuRRSV417FeShyQ5feiy7Wsz95qdnuSYoT9TUFV3SPKgTL5NM939ve6+Jt47K8maJLcZBtK3TXJZvH8A2HX/kuTuSVJV76iq84YZsRuHtr2q6jXDrKbPVtXvDe1Pq6rPDzOV3zy03a6qXl1VnxhmpB8/tP9GVb29qt49zKr9f+eevKpOqqp/q6oPVNUrq+qlQ/vaqnrbcK1PVNXRQ/vzquqUqjonyWvn/yJVdaeqOmd47r9L4t88WKEkwjt2cJJL5u1vTfKTM6pljzfc7nHfJB9LcmB3X5ZMgqequvPQbaHX7OBM/rPG0ntxkt9Pss+wf6ck13T3jcP+3J9/Mu+16e4bq+raof+Vy1fuHuVuSa5I8vdVdZ8k52Uyq8x7ZwXo7q9V1V8m+WqSG5Kck8lr5P0DwGjDBxTHJnn30PSb3X1VVd0mySeq6m1J1ic5eJjVlKrab+i7Kcldu/u789r+IMn7uvs3h7aPV9U/D8eOzGRM/t0kX6yqv03ygyR/lOSoJN9M8r4knx76/00ms3A/XFXrkrwnyT2HYz+R5IHdfUNVPXjer/TcJB/u7hdU1aMymUUNrEBmKu3YQom4r8ubgaq6fZK3JXl6d1+3o64LtHnNpqCqHp3k8u4+b37zAl17EcdYemsyGdi9vLvvm+Tb+eGtbgvx+iyjYS2r45PcNcldktwuk/8MbMv7B4AduU1VfSrJ5kw+qDh1aH9aVX06yUeTHJrJ7e0XJblbVf1tVT0yydyY+jNJ3lBVv5rJ0hNJ8vAkm4ZrfyDJrZOsG469t7uv7e7vJPl8ksMyucPjg919VXd/P8k/zKvxoUleOlzrzCR3qKq5DyTP7O4bFvi9HpTk9UnS3Wdlcks4sAKZqbRjWzP5S3jOIUkunVEte6yqumUmgdIbuvvtQ/PXq+qgYabFQUkuH9q9Zsvn6CSPqarjMhlo3CGTmUv7VdWaYbbF/D//uddm6/Bp2r5Jrlr+svcYW5Ns7e6PDfunZxIqee+sDA9N8uXuviJJqurtSX463j8AjHNDdx85v2GY8fPQJD/V3ddX1QeS3Lq7rx5mLz8iyVOS/FKS38xkKYMHJXlMkj+qqh/P5MOMX+zuL25z7Z/MZIbSnB9k8n/KHd2edouhlh8Jj4a7uL+9g/N8eAKrgJlKO/aJJIcP38azdyaLqJ4545r2KMOaIacmubC7/3reoTOTnDhsn5jknfPaf70mHpDk2rlbfVha3f3s7j6ku9dn8t54X3f/SpL3J3nc0G3b12buNXvc0N9gYUq6+z+SXFJVRwxNx2TyaaL3zsrw1SQPqKrbDn/Pzb0+3j8A3Fz7ZvLlDtdX1T0y+bKbDN+edovufluGW9Wq6hZJDu3u92eypMF+SW6fyS1qvzO3fl9V3Xcnz/nxJD9TVfsPH3784rxj5yR56txOVR257ckL+FCSXxn6H5tk/0WcA8yAmUo7MKxb8dRM/lLdK8mru/tzMy5rT3N0kl9L8tlhymySPCfJyUneWlUnZfKfs8cPx85OclySLUmuT/LE5S2XJM9K8uaq+tMkn8wPp2GfmuR1VbUlkxkWJ8yovj3J72QynX3vTKa8PzGTDxO8d2asuz9WVacnOT+TWw0+meSUJGfF+weAm+fdSZ5cVZ9J8sVMboFLJuvz/f0QJCXJszP5P87rq2rfTGYbvai7r6mqP8lkBvpnhmDp4iSP3t4TDmsF/nkma59emskHJdcOh5+W5GVDPWsyCYx29i11z0/ypqo6P8kHMxmzACtQ+aATAACAm6Oqbt/d3xpmKp2RyQfyZ8y6LmC63P4GAADAzfW84c6CC5J8Ock7ZlwPsAzMVAIAAABgNDOVAAAAABhNqAQAAADAaEIlAAAAAEYTKgEAAAAwmlAJAAAAgNGESgAAAACMJlQCAAAAYDShEgAAAACjCZUAAAAAGE2oBAAAAMBoQiUAAAAARhMqAQAAADCaUAkAAACA0YRKAAAAAIwmVAIAAABgNKESAAAAAKMJlQAAAAAYTagErHpV9Yqq+qMpXPd5VfX6pb4uAMBqZ/wFJEIlYIqq6oFV9f9V1bVVdVVV/WtV3W+pn6e7n9zdf7LU1wUAWG2Mv4DltGbWBQC7p6q6Q5J3JfntJG9NsneS/57kuyOvU0mqu/9zyYsEANiNGH8By81MJWBa/s8k6e43dfcPuvuG7j6nuz+z7bTmqlpfVV1Va4b9D1TVn1XVvya5Pslzqmrz/ItX1e9V1ZnD9muq6k+H7Qur6tHz+q2pqiur6qhh/wHDp3fXVNWnq+rB8/retao+WFXfrKpzkxwwrT8cAIApMP4ClpVQCZiWf0vyg6o6raqOrar9R57/a0k2Jtknyd8mOaKqDp93/JeTvHGB896U5Anz9h+R5MruPr+qDk5yVpI/TXLHJP93krdV1dqh7xuTnJfJYOZPkpw4smYAgFky/gKWlVAJmIruvi7JA5N0klcmuaKqzqyqAxd5idd09+e6+8buvjbJOzMMVobBzT2SnLnAeW9M8piquu2wP3/w86tJzu7us7v7P7v73CSbkxxXVeuS3C/JH3X3d7v7Q0n+cezvDQAwK8ZfwHITKgFT090XdvdvdPchSe6d5C5JXrzI0y/ZZv+N+eEnYL+c5B3dff0Cz7klyYVJfm4Y2DwmPxzUHJbk8cPU62uq6ppMBl4HDbVd3d3fnne5ryyyVgCAFcH4C1hOFuoGlkV3f6GqXpPkSUnOT3LbeYd/bKFTttk/J8kBVXVkJoOb39vB081Nwb5Fks8PA51kMlB6XXf/1rYnVNVhSfavqtvNG9isW6AOAIBVwfgLmDYzlYCpqKp7VNUzq+qQYf/QTAYaH03yqSQPqqp1VbVvkmfv7HrdfWOS05P8RSb345+7g+5vTvLwTL75ZP59/6/P5BO0R1TVXlV166p6cFUd0t1fyWQq9vOrau+qemCSnxv7ewMAzIrxF7DchErAtHwzyU8m+VhVfTuTwcwFSZ453Ev/liSfyWRhxnct8ppvTPLQJP8wDHIW1N2XJflIkp8enmeu/ZIkxyd5TpIrMvnk7P/JD/8u/OWh5quSPDfJaxdZFwDASmD8BSyr6jazEAAAAIBxzFQCAAAAYDShEgAAAACjCZUAAAAAGE2oBAAAAMBoQiUAAAAARlsz6wKW0gEHHNDr16+fdRkAwJScd955V3b32lnXwQ8ZfwHA7m97Y7DdKlRav359Nm/ePOsyAIApqaqvzLoGfpTxFwDs/rY3BnP7GwAAAACjCZUAAAAAGE2oBAAAAMBoQiUAAAAARhMqAQAAADCaUAkAAACA0YRKAAB7iKp6dVVdXlUXbOd4VdVLqmpLVX2mqo5a7hoBgNVDqAQAsOd4TZJH7uD4sUkOH342Jnn5MtQEAKxSQiUAgD1Ed38oyVU76HJ8ktf2xEeT7FdVBy1PdQDAaiNUAgBgzsFJLpm3v3VoAwC4iTWzLmC1WL/prJ32ufjkRy1DJQAAU1MLtPVNOlVtzOT2uKxbt27aNcEe6z7PPyfX3vD9JbnWV1746CW5zjQc9qx3Lcl19r3NLfPp5z58Sa4FLI5QCQCAOVuTHDpv/5Akl27bqbtPSXJKkmzYsOEmoROwNK694ftL98H1ybv/W3UxEwGApeX2NwAA5pyZ5NeHb4F7QJJru/uyWRcFAKxMZioBAOwhqupNSR6c5ICq2prkuUlumSTd/YokZyc5LsmWJNcneeJsKgUAVgOhEgDAHqK7n7CT453kKctUDgCwyrn9DQAAAIDRhEoAAAAAjCZUAgAAAGA0oRIAAAAAowmVAAAAABhNqAQAAADAaEIlAAAAAEYTKgEAAAAwmlAJAAAAgNGESgAAAACMJlQCAAAAYDShEgAAAACjCZUAAAAAGG1ZQ6WquriqPltVn6qqzUPbHavq3Kr60vC4/9BeVfWSqtpSVZ+pqqOWs1YAAAAAtm8WM5V+truP7O4Nw/6mJO/t7sOTvHfYT5Jjkxw+/GxM8vJlrxQAAACABa2E29+OT3LasH1aksfOa39tT3w0yX5VddAsCgQAAADgRy13qNRJzqmq86pq49B2YHdfliTD452H9oOTXDLv3K1DGwAAAAAztmaZn+/o7r60qu6c5Nyq+sIO+tYCbX2TTpNwamOSrFu3bmmqBAAAAGCHlnWmUndfOjxenuSMJPdP8vW529qGx8uH7luTHDrv9EOSXLrANU/p7g3dvWHt2rXTLB8AAACAwbKFSlV1u6raZ247ycOTXJDkzCQnDt1OTPLOYfvMJL8+fAvcA5JcO3ebHAAAAACztZy3vx2Y5IyqmnveN3b3u6vqE0neWlUnJflqkscP/c9OclySLUmuT/LEZawVAAAAgB1YtlCpuy9Kcp8F2r+R5JgF2jvJU5ahNAAAAABGWu5vfwMAAABgNyBUAgAAAGA0oRIAAAAAowmVAAAAABhNqAQAAADAaEIlAAAAAEYTKgEAAAAwmlAJAAAAgNGESgAAAACMJlQCAAAAYDShEgAAAACjCZUAAAAAGE2oBAAAAMBoQiUAAAAA/n/27j7Ksrq8E/33kQZfCUQojaG7bSa2icSoaIvOkIkvaC6SCGauRLzGkFxiJxPN6MVJgibLOOZmXU1mJHH52hGvyMQgEl/6GowSBY2ZgDRKQEBiB4l02kirgOIbNnnuH2d3LIvq7jpF1amX/nzWqnX2/u1n73rOtldx/J7f3ntsQiUAAAAAxiZUAgA4QFTViVV1Q1Vtr6qzZtm+vqouqapPV9XVVXXSUvQJAKwMQiUAgANAVR2U5A1JnpHkmCTPrapjZpT9bpILuvvYJKcleeNkuwQAVhKhEgDAgeG4JNu7+8buvjPJ+UlOmVHTSX5gWD4syc4J9gcArDBrlroBAAAm4qgkN09b35HkCTNqXpnkw1X1G0nun+Rpk2kNAFiJzFQCADgw1CxjPWP9uUne3t1rk5yU5LyqutvnxaraXFXbqmrbrl27FqFVAGAlECoBABwYdiRZN219be5+edsZSS5Iku7+uyT3SXLkzAN195bu3tTdm6amphapXQBguRMqAQAcGK5IsrGqjq6qQzK6EffWGTVfSHJCklTVIzIKlUxFAgBmJVQCADgAdPfuJC9K8qEk12f0lLdrq+pVVXXyUPbSJC+oqr9P8udJfqm7Z14iBwCQxI26AQAOGN19UZKLZoy9YtrydUmOn3RfAMDKZKYSAAAAAGMTKgEAAAAwNqESAAAAAGMTKgEAAAAwNqESAAAAAGMTKgEAAAAwNqESAAAAAGMTKgEAAAAwNqESAAAAAGMTKgEAAAAwNqESAAAAAGObaKhUVQdV1aer6gPD+tFVdXlVfa6q3lVVhwzj9x7Wtw/bN0yyTwAAAAD2bdIzlV6c5Ppp669JcnZ3b0xya5IzhvEzktza3Q9LcvZQBwAAAMAyMbFQqarWJvmZJG8d1ivJU5NcOJScm+RZw/Ipw3qG7ScM9QAAAAAsA5OcqfTHSX4ryb8O60ckua27dw/rO5IcNSwfleTmJBm23z7UAwAAALAMTCRUqqqfTXJLd185fXiW0p7DtpnH3lxV26pq265du+5hpwAAAADMxaRmKh2f5OSquinJ+Rld9vbHSQ6vqjVDzdokO4flHUnWJcmw/bAkX53twN29pbs3dfemqampxXsHAAAAAPybiYRK3f2y7l7b3RuSnJbko939vCSXJHn2UHZ6kvcPy1uH9QzbP9rds85UAgAAAGDyJv30t5l+O8mZVbU9o3smnTOMn5PkiGH8zCRnLVF/AAAAAMxizf5LFlZ3X5rk0mH5xiTHzVLz7SSnTrQxAAAAAOZsqWcqAQAAALACCZUAAAAAGJtQCQAAAICxCZUAAAAAGJtQCQAAAICxCZUAAAAAGJtQCQAAAICxCZUAAAAAGJtQCQAAAICxCZUAAAAAGJtQCQAAAICxCZUAAAAAGJtQCQAAAICxCZUAAAAAGJtQCQAAAICxCZUAAAAAGJtQCQAAAICxCZUAAAAAGJtQCQDgAFFVJ1bVDVW1varO2kvNz1fVdVV1bVW9c9I9AgArx5qlbgAAgMVXVQcleUOSpyfZkeSKqtra3ddNq9mY5GVJju/uW6vqQUvTLQCwEsxrplJVPXKhGwEAYFEdl2R7d9/Y3XcmOT/JKTNqXpDkDd19a5J09y0T7hEAWEHmO1PpzVV1SJK3J3lnd9+2cC0BAMvVhrP+cr81N736ZybQCfNwVJKbp63vSPKEGTUPT5Kq+tskByV5ZXf/1WTaAwBWmnnNVOrun0zyvCTrkmyrqndW1dMXtDMAABZSzTLWM9bXJNmY5MlJnpvkrVV1+N0OVLW5qrZV1bZdu3YteKMAwMow7xt1d/fnkvxukt9O8qQkr6uqz1bVf1qo5gAAWDA7MvpCcI+1SXbOUvP+7v5ud38+yQ0ZhUzfp7u3dPem7t40NTW1aA0DAMvbfO+p9KiqOjvJ9UmemuSZ3f2IYfnsBewPAICFcUWSjVV19HAbg9OSbJ1R874kT0mSqjoyo8vhbpxolwDAijHfeyq9PsmfJnl5d39rz2B376yq312QzgAAWDDdvbuqXpTkQxndL+lt3X1tVb0qybbu3jps++mqui7JXUl+s7u/snRdAwDL2XxDpZOSfKu770qSqrpXkvt09ze7+7wF6w4AgAXT3RcluWjG2CumLXeSM4cfAIB9mu89lf46yX2nrd9vGAMAAADgADDfUOk+3X3HnpVh+X4L0xIAAAAAy918Q6VvVNVj96xU1eOSfGsf9QAAAACsIvO9p9JLkry7qvY8hvYhSZ6zMC0BAAAAsNzNK1Tq7iuq6seS/GiSSvLZ7v7ugnYGAAAAwLI135lKSfL4JBuGYxxbVenudyxIVwAAAAAsa/MKlarqvCQ/kuSqJHcNw51EqAQAAABwAJjvTKVNSY7p7l7IZgAAAABYGeb79LfPJPmhhWwEAAAAgJVjvjOVjkxyXVV9Msl39gx298l726Gq7pPk40nuPfzeC7v796rq6CTnJ3lgkk8leX5331lV987ocrrHJflKkud0903z7BcAAACABTTfUOmV89jnO0me2t13VNXBST5RVR9McmaSs7v7/Kp6c5IzkrxpeL21ux9WVacleU2S58yzXwAAAAAW0Lwuf+vujyW5KcnBw/IVGc0y2tc+3d13DKsHDz+d5KlJLhzGz03yrGH5lGE9w/YTqqrm0y8AAAAAC2teoVJVvSCjoOctw9BRSd43h/0OqqqrktyS5OIk/5jktu7ePZTsGI6155g3J8mw/fYkR8ynXwAAAAAW1nxv1P3CJMcn+VqSdPfnkjxofzt1913d/Zgka5Mcl+QRs5UNr7PNSrrb0+aqanNVbauqbbt27Zpj+wAAAADcE/MNlb7T3XfuWamqNZkl8Nmb7r4tyaVJnpjk8GH/ZBQ27RyWdyRZN+34hyX56izH2tLdm7p709TU1DzeCgAAAADjmm+o9LGqenmS+1bV05O8O8n/t68dqmqqqg4flu+b5GlJrk9ySZJnD2WnJ3n/sLx1WM+w/aPdPefgCgAAAIDFM9+nv52V0dPZrknyq0kuSvLW/ezzkCTnVtVBGYVZF3T3B6rquiTnV9X/neTTSc4Z6s9Jcl5Vbc9ohtJp8+wVAAAAgAU2r1Cpu/81yZ8OP3Pd5+okx84yfmNG91eaOf7tJKfOpz8AAAAAFte8QqWq+nxmuYdSd/+7e9wRAAAAAMvefC9/2zRt+T4ZzSh64D1vBwAAAICVYF436u7ur0z7+efu/uMkT13g3gAAAABYpuZ7+dtjp63eK6OZS4cuSEcAAAAALHvzvfztf0xb3p3kpiQ/f4+7AQAAAGBFmO/T356y0I0AAAAAsHLM9/K3M/e1vbtfO792AAAAAFgJ7snT3x6fZOuw/swkH09y80I0BQAAAMDyNt9Q6cgkj+3urydJVb0yybu7+1cWqjEAAAAAlq97zXO/9UnunLZ+Z5IN97gbAAAAAFaE+c5UOi/JJ6vqvUk6yc8leceCdQUAAADAsjbfp7/9QVV9MMl/HIZ+ubs/vXBtAQAAALCczffytyS5X5KvdfefJNlRVUcvUE8AAAAALHPzCpWq6veS/HaSlw1DByf5nwvVFAAAAADL23xnKv1ckpOTfCNJuntnkkMXqikAAAAAlrf5hkp3dndndJPuVNX9F64lAAAAAJa7+YZKF1TVW5IcXlUvSPLXSf504doCAAAAYDmb79Pf/ntVPT3J15L8aJJXdPfFC9oZAAAAAMvW2KFSVR2U5EPd/bQkgiQAAACAA9DYl791911JvllVhy1CPwAALJKqOrGqbqiq7VV11j7qnl1VXVWbJtkfALCyzOvytyTfTnJNVV2c4QlwSdLd/2VBugIAYEENs83fkOTpSXYkuaKqtnb3dTPqDk3yX5JcPvkuAYCVZL6h0l8OPwAArAzHJdne3TcmSVWdn+SUJNfNqPv9JH+Y5L9Otj0AYKUZK1SqqvXd/YXuPnexGgIAYFEcleTmaes7kjxhekFVHZtkXXd/oKqESgDAPo17T6X37Vmoqr9Y4F4AAFg8NctY/9vGqnslOTvJS/d7oKrNVbWtqrbt2rVrAVsEAFaScUOl6R9G/t1CNgIAwKLakWTdtPW1SXZOWz80ySOTXFpVNyV5YpKts92su7u3dPem7t40NTW1iC0DAMvZuKFS72UZAIDl7YokG6vq6Ko6JMlpSbbu2djdt3f3kd29obs3JLksycndvW1p2gUAlrtxb9T96Kr6WkYzlu47LGdY7+7+gQXtDgCABdHdu6vqRUk+lOSgJG/r7mur6lVJtnX31n0fAQDg+40VKnX3QYvVCAAAi6u7L0py0YyxV+yl9smT6AkAWLnGvfwNAAAAAIRKAAAAAIxPqAQAAADA2IRKAAAAAIxNqAQAAADA2IRKAAAAAIxNqAQAAADA2CYSKlXVuqq6pKqur6prq+rFw/gDq+riqvrc8PqDw3hV1euqantVXV1Vj51EnwAAAADMzaRmKu1O8tLufkSSJyZ5YVUdk+SsJB/p7o1JPjKsJ8kzkmwcfjYnedOE+gQAAABgDiYSKnX3F7v7U8Py15Ncn+SoJKckOXcoOzfJs4blU5K8o0cuS3J4VT1kEr0CAAAAsH8Tv6dSVW1IcmySy5M8uLu/mIyCpyQPGsqOSnLztN12DGMAAAAALAMTDZWq6gFJ/iLJS7r7a/sqnWWs93LMzVW1raq27dq1ayHaBAAAAGA/JhYqVdXBGQVKf9bd7xmGv7Tnsrbh9ZZhfEeSddN2X5tk52zH7e4t3b2puzdNTU0tTvMAAAAAfJ9JPf2tkpyT5Prufu20TVuTnD4sn57k/dPGf3F4CtwTk9y+5zI5AAAAAJbemgn9nuOTPD/JNVV11TD28iSvTnJBVZ2R5AtJTh22XZTkpCTbk3wzyS9PqE8AAAAA5mAioVJ3fyKz3ycpSU6Ypb6TvHBRmwIAAABg3ib+9DcAAAAAVj6hEgAAAABjEyoBAAAAMDahEgAAAABjEyoBAAAAMDahEgAAAABjEyoBAAAAMDahEgAAAABjEyoBAAAAMDahEgAAAABjEyoBAAAAMDahEgAAAABjEyoBAAAAMDahEgAAAABjEyoBAAAAMDahEgAAAABjEyoBAAAAMDahEgAAAABjEyoBAAAAMDahEgAAAABjEyoBAAAAMDahEgAAAABjEyoBABwgqurEqrqhqrZX1VmzbD+zqq6rqqur6iNV9dCl6BMAWBmESgAAB4CqOijJG5I8I8kxSZ5bVcfMKPt0kk3d/agkFyb5w8l2CQCsJEIlAIADw3FJtnf3jd19Z5Lzk5wyvaC7L+nubw6rlyVZO+EeAYAVRKgEAHBgOCrJzdPWdwxje3NGkg8uakcAwIq2ZqkbAABgImqWsZ61sOoXkmxK8qS9bN+cZHOSrF+/fqH6AwBWGDOVAAAODDuSrJu2vjbJzplFVfW0JL+T5OTu/s5sB+ruLd29qbs3TU1NLUqzAMDyJ1QCADgwXJFkY1UdXVWHJDktydbpBVV1bJK3ZBQo3bIEPQIAK4hQCQDgANDdu5O8KMmHklyf5ILuvraqXlVVJw9lf5TkAUneXVVXVdXWvRwOAMA9lQAADhTdfVGSi2aMvWLa8tMm3hQAsGKZqQQAAADA2IRKAAAAAIxNqAQAAADA2IRKAAAAAIxNqAQAAADA2CYWKlXV26rqlqr6zLSxB1bVxVX1ueH1B4fxqqrXVdX2qrq6qh47qT4BAAAA2L9JzlR6e5ITZ4ydleQj3b0xyUeG9SR5RpKNw8/mJG+aUI8AAAAAzMHEQqXu/niSr84YPiXJucPyuUmeNW38HT1yWZLDq+ohk+kUAAAAgP1Z6nsqPbi7v5gkw+uDhvGjktw8rW7HMAYAAADAMrDUodLe1CxjPWth1eaq2lZV23bt2rXIbQEAAACQY21yPQAAIABJREFULH2o9KU9l7UNr7cM4zuSrJtWtzbJztkO0N1buntTd2+amppa1GYBAAAAGFnqUGlrktOH5dOTvH/a+C8OT4F7YpLb91wmBwAAAMDSWzOpX1RVf57kyUmOrKodSX4vyauTXFBVZyT5QpJTh/KLkpyUZHuSbyb55Un1CQAAAMD+TSxU6u7n7mXTCbPUdpIXLm5HAAAAAMzXUl/+BgAAAMAKJFQCAAAAYGxCJQAAAADGJlQCAAAAYGxCJQAAAADGJlQCAAAAYGxCJQAAAADGJlQCAAAAYGxCJQAAAADGJlQCAAAAYGxCJQAAAADGJlQCAAAAYGxCJQAAAADGJlQCAAAAYGxCJQAAAADGJlQCAAAAYGxCJQAAAADGJlQCAAAAYGxCJQAAAADGJlQCAAAAYGxCJQAAAADGJlQCAAAAYGxCJQAAAADGJlQCAAAAYGxCJQCAA0RVnVhVN1TV9qo6a5bt966qdw3bL6+qDZPvEgBYKYRKAAAHgKo6KMkbkjwjyTFJnltVx8woOyPJrd39sCRnJ3nNZLsEAFYSoRIAwIHhuCTbu/vG7r4zyflJTplRc0qSc4flC5OcUFU1wR4BgBVEqAQAcGA4KsnN09Z3DGOz1nT37iS3JzliIt0BACvOmqVuAACAiZhtxlHPoyZVtTnJ5iRZv379Pe8MmNWhjzgrP3Hu3W5/xl4c+ogk+ZmlbgMOKEIlAIADw44k66atr02ycy81O6pqTZLDknx15oG6e0uSLUmyadOmu4VOwMK45vRrlroFgH1y+RsAwIHhiiQbq+roqjokyWlJts6o2Zrk9GH52Uk+2t1CIwBgVmYqAQAcALp7d1W9KMmHkhyU5G3dfW1VvSrJtu7emuScJOdV1faMZiidtnQdAwDLnVAJAOAA0d0XJbloxtgrpi1/O8mpk+4LAFiZXP4GAAAAwNiESgAAAACMbVmHSlV1YlXdUFXbq8qzNAEAAACWiWUbKlXVQUnekOQZSY5J8tyqOmZpuwIAAAAgWcahUpLjkmzv7hu7+84k5yc5ZYl7AgAAACDLO1Q6KsnN09Z3DGMAAAAALLE1S93APtQsY323oqrNSTYPq3dU1Q2L1M+RSb68r4J6zSL95gPTfs83C8r5nizne/Kc8wmq1yzq+X7oIh2Xebryyiu/XFX/tNR9ABPlv6tw4Jn1M9hyDpV2JFk3bX1tkp0zi7p7S5Iti91MVW3r7k2L/XsYcb4ny/meLOd78pzzyXK+DyzdPbXUPQCT5e88sMdyvvztiiQbq+roqjokyWlJti5xTwAAAABkGc9U6u7dVfWiJB9KclCSt3X3tUvcFgAAAABZxqFSknT3RUkuWuo+Bot+iR3fx/meLOd7spzvyXPOJ8v5Bljd/J0HkiTVfbd7XwMAAADAPi3neyoBAAAAsEwJlaapqhOr6oaq2l5VZ82y/d5V9a5h++VVtWHyXa4uczjnZ1bVdVV1dVV9pKo8Svoe2N/5nlb37KrqqvJUj3tgLue7qn5++Dd+bVW9c9I9riZz+HuyvqouqapPD39TTlqKPleLqnpbVd1SVZ/Zy/aqqtcN/3tcXVWPnXSPAIxU1e8MnzWurqqrquoJC3DMk/f1eXLMY92xEMcBJs/lb4OqOijJPyR5epIdGT197rndfd20ml9P8qju/rWqOi3Jz3X3c5ak4VVgjuf8KUku7+5vVtV/TvJk53x+5nK+h7pDk/xlkkOSvKi7t02619Vgjv++Nya5IMlTu/vWqnpQd9+yJA2vcHM831uSfLq731RVxyS5qLs3LEW/q0FV/VSSO5K8o7sfOcv2k5L8RpKTkjwhyZ909z3+PzEAjKeq/n2S12b0Ofo7VXVkkkO6e+cc9l3T3bsn0OMd3f2Axf49wMIzU+l7jkuyvbtv7O47k5yf5JQZNackOXdYvjDJCVVVE+xxtdnvOe/uS7r7m8PqZUnWTrjH1WQu/8aT5PeT/GGSb0+yuVVoLuf7BUne0N23JolA6R6Zy/nuJD8wLB+WZL8fptm77v54kq/uo+SUjAKn7u7LkhxeVQ+ZTHcATPOQJF/u7u8kSXd/ubt3VtVNQ8CUqtpUVZcOy6+sqi1V9eEk7xiu0PjxPQerqkur6nFV9UtV9fqqOmw41r2G7ferqpur6uCq+pGq+ququrKq/qaqfmyoObqq/q6qrqiq35/w+QAWkFDpe45KcvO09R3D2Kw1Q2J/e5IjJtLd6jSXcz7dGUk+uKgdrW77Pd9VdWySdd39gUk2tkrN5d/3w5M8vKr+tqouq6oTJ9bd6jOX8/3KJL9QVTsyerLob0ymtQPWuH/jAVgcH06yrqr+oareWFVPmsM+j0tySnf/Hxl9UfPzSTJ8OfDD3X3lnsLuvj3J3yfZc9xnJvlQd383o6fE/UZ3Py7Jf03yxqHmT5K8qbsfn+Rf7vE7BJaMUOl7ZptxNPPawLnUMHdzPp9V9QtJNiX5o0XtaHXb5/kevl06O8lLJ9bR6jaXf99rkmxM8uQkz03y1qo6fJH7Wq3mcr6fm+Tt3b02o0uyztvzrSqLwn8zAZaB7r4jo5Boc5JdSd5VVb+0n922dve3huULkpw6LP98knfPUv+uJHtuUXHa8DsekOQ/JHl3VV2V5C0ZzZpKkuOT/PmwfN5YbwhYVtYsdQPLyI4k66atr83dL43YU7OjqtZkdPnEvqb+s29zOeepqqcl+Z0kT9ozbZd52d/5PjTJI5NcOlzV+UNJtlbVye6rNC9z/Zty2fBN3uer6oaMQqYrJtPiqjKX831GkhOTpLv/rqruk+TIJC47XBxz+hsPwOLr7ruSXJrR57xrkpyeZHe+N8ngPjN2+ca0ff+5qr5SVY/KKDj61Vl+xdYk/09VPTCjAOujSe6f5Lbufsze2prn2wGWEd/Qfs8VSTYO1/ceklHCvnVGzdaM/gAnybOTfLTd6fye2O85Hy7HekuSk91v5h7b5/nu7tu7+8ju3jDcvPiyjM67QGl+5vI35X1JnpIkwz0NHp7kxol2uXrM5Xx/IckJSVJVj8joA/SuiXZ5YNma5BeHp8A9Mcnt3f3FpW4K4EBTVT86PBxkj8ck+ackN2UUACXJ/76fw5yf5LeSHNbd18zcOMyG+mRGl7V9oLvv6u6vZfSl2alDH1VVjx52+duM/ludJM8b/10By4VQaTDcI+lFST6U5PokF3T3tVX1qqo6eSg7J8kRVbU9yZlJFuQRmgeqOZ7zP0rygAzTZqtq5v9JZI7meL5ZIHM83x9K8pWqui7JJUl+s7u/sjQdr2xzPN8vTfKCqvr7jKbc/5IvBuavqv48yd8l+dGq2lFVZ1TVr1XVrw0lF2UUkm5P8qdJfn2JWgU40D0gyblVdV1VXZ3kmIzuM/jfkvxJVf1Nkrv2c4wLMwqBLthHzbuS/MLwusfzkpwx/Lf32nzvIRovTvLCqroio6s/gBWqfJ4GAAAAYFxmKgEAAAAwNqESAAAAAGMTKgEAAAAwNqESAAAAAGMTKgEAAAAwNqESAAAAAGMTKgEAAAAwNqESAAAAAGMTKgEAAAAwNqESAAAAAGMTKgEAAAAwNqESAAAAAGMTKgEAAAAwNqESAAAAAGMTKgEAAAAwNqESAAAAAGMTKgEAAAAwNqESAAAAAGMTKgEAAAAwNqESAAAAAGMTKgErQlU9uap2LHUfAAAHCp+/gP0RKgFLoqpuqqpvVdUdVfWlqvp/q+oBS90XAMBq5fMXsNCESsBSemZ3PyDJY5M8PsnvLnE/AACrnc9fwIIRKgFLrrv/OckHkzyyqh44fGu2s6purar3zbZPVZ1VVf9YVV+vquuq6uembXtYVX2sqm6vqi9X1buG8aqqs6vqlmHb1VX1yMm8SwCA5cPnL2AhrFnqBgCqal2Sk5K8J8l5Se5I8uPD63/Yy27/mOQ/JvmXJKcm+Z9V9bDu/mKS30/y4SRPSXJIkk3DPj+d5KeSPDzJ7Ul+LMlti/CWAACWNZ+/gIUgVAKW0vuqandGHzD+Mskbk/xzkiO6+9ah5mOz7djd7562+q6qelmS45K8P8l3kzw0yQ93944knxjqvpvk0Iw+zHyyu69f4PcDALDc+fwFLBiXvwFL6VndfXh3P7S7fz3JuiRfnfaBZq+q6her6qqquq2qbkvyyCRHDpt/K0kl+WRVXVtV/2eSdPdHk7w+yRuSfKmqtlTVDyzGGwMAWKZ8/gIWjFAJWE5uTvLAqjp8X0VV9dAkf5rkRRl9q3Z4ks9k9EEm3f0v3f2C7v7hJL+a5I1V9bBh2+u6+3EZTe9+eJLfXLR3AwCw/Pn8BcybUAlYNobr8T+Y0YeQH6yqg6vqp2YpvX+STrIrSarqlzP6pizD+qlVtXZYvXWovauqHl9VT6iqg5N8I8m3k9y1eO8IAGB58/kLuCeESsBy8/yMrr3/bJJbkrxkZkF3X5fkfyT5uyRfSvITSf52Wsnjk1xeVXck2Zrkxd39+SQ/kNE3bLcm+ackX0ny3xftnQAArAw+fwHzUt291D0AAAAAsMKYqQQAAADA2IRKAAAAAIxNqAQAAADA2IRKAAAAAIxNqAQAAADA2NYsdQML6cgjj+wNGzYsdRsAwCK58sorv9zdU0vdB9/j8xcArH57+wy2qkKlDRs2ZNu2bUvdBgCwSKrqn5a6B76fz18AsPrt7TOYy98AAAAAGNtEQ6Wquqmqrqmqq6pq2zD2wKq6uKo+N7z+4DBeVfW6qtpeVVdX1WMn2SsAAAAAe7cUM5We0t2P6e5Nw/pZST7S3RuTfGRYT5JnJNk4/GxO8qaJdwoAAADArJbD5W+nJDl3WD43ybOmjb+jRy5LcnhVPWQpGgQAAADg+006VOokH66qK6tq8zD24O7+YpIMrw8axo9KcvO0fXcMYwAAAAAssUmHSsd392MzurTthVX1U/uorVnG+m5FVZuraltVbdu1a9dC9QkAsCJV1X2q6pNV9fdVdW1V/bdZau5dVe8a7l15eVVtmLbtZcP4DVX1v02ydwBgZZloqNTdO4fXW5K8N8lxSb6057K24fWWoXxHknXTdl+bZOcsx9zS3Zu6e9PU1NRitg8AsBJ8J8lTu/vRSR6T5MSqeuKMmjOS3NrdD0tydpLXJElVHZPktCQ/nuTEJG+sqoMm1jkAsKJMLFSqqvtX1aF7lpP8dJLPJNma5PSh7PQk7x+Wtyb5xeEpcE9Mcvuey+QAAJjdcD/KO4bVg4efmbO9p9/T8sIkJ1RVDePnd/d3uvvzSbZn9CUgAMDdrJng73pwkveOPq9kTZJ3dvdfVdUVSS6oqjOSfCHJqUP9RUlOyujDzDeT/PIEewUAWLGG2UVXJnlYkjd09+UzSv7t3pXdvbuqbk9yxDB+2bQ697QEAPZqYqFSd9+Y5NGzjH8lyQmzjHeSF06gNQCAVaW770rymKo6PKMv9R7Z3Z+ZVrK3e1fO+Z6WSTYnyfr16xegY2A2P3HuTyx1CyvONadfs9QtwAFlkjOVAIAVbsNZf7nfmpte/TMT6IS56O7bqurSjO6PND1U2nPvyh1VtSbJYUm+mjHuaZlkS5Js2rTpbqETsDC+fv2r/U0dw1z+GwUsrEk//Q0AgEVUVVPDDKVU1X2TPC3JZ2eUTb+n5bOTfHSYJb41yWnD0+GOTrIxyScn0zkAsNKYqQQAsLo8JMm5w32V7pXkgu7+QFW9Ksm27t6a5Jwk51XV9oxmKJ2WJN19bVVdkOS6JLuTvHC4lA4A4G6ESgAAq0h3X53k2FnGXzFt+dv53sNRZtb9QZI/WLQGAYBVw+VvAAAAAIxNqAQAAADA2IRKAAAAAIxNqAQAAADA2IRKAAAAAIxNqAQAAADA2IRKAAAAAIxNqAQAAADA2IRKAAAAAIxNqAQAAADA2IRKAAAAAIxNqAQAAADA2IRKAAAAAIxNqAQAAADA2IRKAAAAAIxNqAQAAADA2IRKAAAAAIxNqAQAAADA2IRKAAAAAIxNqAQAAADA2NYsdQMAACycqlqX5B1JfijJvybZ0t1/MqPmN5M8b1hdk+QRSaa6+6tVdVOSrye5K8nu7t40qd4BgJVFqAQAsLrsTvLS7v5UVR2a5Mqquri7r9tT0N1/lOSPkqSqnpnk/+rur047xlO6+8sT7RoAWHFc/gYAsIp09xe7+1PD8teTXJ/kqH3s8twkfz6J3gCA1UWoBACwSlXVhiTHJrl8L9vvl+TEJH8xbbiTfLiqrqyqzYvdIwCwcrn8DQBgFaqqB2QUFr2ku7+2l7JnJvnbGZe+Hd/dO6vqQUkurqrPdvfHZxx7c5LNSbJ+/fpF6B4AWAnMVAIAWGWq6uCMAqU/6+737KP0tMy49K27dw6vtyR5b5LjZu7U3Vu6e1N3b5qamlq4xgGAFUWoBACwilRVJTknyfXd/dp91B2W5ElJ3j9t7P7Dzb1TVfdP8tNJPrO4HQMAK5XL3wAAVpfjkzw/yTVVddUw9vIk65Oku988jP1ckg939zem7fvgJO8d5VJZk+Sd3f1XE+kaAFhxhEoAAKtId38iSc2h7u1J3j5j7MYkj16UxgCAVcflbwAAAACMTagEAAAAwNiESgAAAACMTagEAAAAwNiESgAAAACMTagEAAAAwNiESgAAAACMTagEAAAAwNgmGipV1UFV9emq+sCwfnRVXV5Vn6uqd1XVIcP4vYf17cP2DZPsEwAAAIB9m/RMpRcnuX7a+muSnN3dG5PcmuSMYfyMJLd298OSnD3UAQAAALBMTCxUqqq1SX4myVuH9Ury1CQXDiXnJnnWsHzKsJ5h+wlDPQAAAADLwCRnKv1xkt9K8q/D+hFJbuvu3cP6jiRHDctHJbk5SYbttw/1AAAAACwDEwmVqupnk9zS3VdOH56ltOewbeaxN1fVtqratmvXrnvYKQAAAABzMamZSscnObmqbkpyfkaXvf1xksOras1QszbJzmF5R5J1STJsPyzJV2c7cHdv6e5N3b1pampq8d4BAAAAAP9mIqFSd7+su9d294YkpyX5aHc/L8klSZ49lJ2e5P3D8tZhPcP2j3b3rDOVAAAAAJi8ST/9babfTnJmVW3P6J5J5wzj5yQ5Yhg/M8lZS9QfAAAAALNYs/+ShdXdlya5dFi+Mclxs9R8O8mpE20MAAAAgDlb6plKAAAAAKxAQiUAAAAAxiZUAgAAAGBsQiUAAAAAxiZUAgBYRapqXVVdUlXXV9W1VfXiWWqeXFW3V9VVw88rpm07sapuqKrtVeUJvADAXk386W8AACyq3Ule2t2fqqpDk1xZVRd393Uz6v6mu392+kBVHZTkDUmenmRHkiuqauss+wIAmKkEALCadPcXu/tTw/LXk1yf5Kg57n5cku3dfWN335nk/CSnLE6nAMBKJ1QCAFilqmpDkmOTXD7L5n9fVX9fVR+sqh8fxo5KcvO0mh2ZeyAFABxgXP4GALAKVdUDkvxFkpd099dmbP5Ukod29x1VdVKS9yXZmKRmOVTPcuzNSTYnyfr16xe0bwBg5TBTCQBglamqgzMKlP6su98zc3t3f6277xiWL0pycFUdmdHMpHXTStcm2TnL/lu6e1N3b5qamlqU9wAALH9CJQCAVaSqKsk5Sa7v7tfupeaHhrpU1XEZfSb8SpIrkmysqqOr6pAkpyXZOpnOAYCVxuVvAACry/FJnp/kmqq6ahh7eZL1SdLdb07y7CT/uap2J/lWktO6u5PsrqoXJflQkoOSvK27r530GwAAVgahEgDAKtLdn8js90aaXvP6JK/fy7aLkly0CK0BAKuMy98AAAAAGJtQCQAAAICxCZUAAAAAGJtQCQAAAICxCZUAAAAAGJtQCQAAAICxCZUAAAAAGJtQCQAAAICxCZUAAAAAGJtQCQAAAICxCZUAAAAAGJtQCQAAAICxCZUAAAAAGJtQCQAAAICxCZUAAAAAGJtQCQAAAICxCZUAAAAAGJtQCQAAAICxCZUAAAAAGJtQCQAAAICxCZUAAFaRqlpXVZdU1fVVdW1VvXiWmudV1dXDz/+qqkdP23ZTVV1TVVdV1bbJdg8ArCRrlroBAAAW1O4kL+3uT1XVoUmurKqLu/u6aTWfT/Kk7r61qp6RZEuSJ0zb/pTu/vIEewYAViChEgDAKtLdX0zyxWH561V1fZKjklw3reZ/TdvlsiRrJ9okALAquPwNAGCVqqoNSY5Ncvk+ys5I8sFp653kw1V1ZVVtXrzuAICVzkwlAIBVqKoekOQvkryku7+2l5qnZBQq/eS04eO7e2dVPSjJxVX12e7++Iz9NifZnCTr169flP4BgOXPTCUAgFWmqg7OKFD6s+5+z15qHpXkrUlO6e6v7Bnv7p3D6y1J3pvkuJn7dveW7t7U3ZumpqYW4y0AACuAUAkAYBWpqkpyTpLru/u1e6lZn+Q9SZ7f3f8wbfz+w829U1X3T/LTST6z+F0DACuRy98AAFaX45M8P8k1VXXVMPbyJOuTpLvfnOQVSY5I8sZRBpXd3b0pyYOTvHcYW5Pknd39V5NtHwBYKeYVKlXVI7t7rG+tquo+ST6e5N7D772wu3+vqo5Ocn6SByb5VEbfmN1ZVfdO8o4kj0vylSTP6e6b5tMvAMCBors/kaT2U/MrSX5llvEbkzx6kVoDAFaZ+V7+9uaq+mRV/XpVHT7Hfb6T5Knd/egkj0lyYlU9Mclrkpzd3RuT3JrRzSIzvN7a3Q9LcvZQBwAAAMAyMK9Qqbt/MsnzkqxLsq2q3llVT9/PPt3ddwyrBw8/neSpSS4cxs9N8qxh+ZRhPcP2E4Z7BAAAAACwxOZ9o+7u/lyS303y20melOR1VfXZqvpPe9unqg4aru2/JcnFSf4xyW3dvXso2ZHkqGH5qCQ3D79rd5LbM7r2HwAAAIAlNq9QqaoeVVVnJ7k+o5lGz+zuRwzLZ+9tv+6+q7sfk2RtRo+nfcRsZXt+zT62Te9lc1Vtq6ptu3btGvOdAAAAADAf852p9PqMbqr96O5+YXd/Kkm6e2dGs5f2qbtvS3JpkicmObyq9twwfG2SncPyjowur8uw/bAkX53lWFu6e1N3b5qamprn2wEAAABgHPMNlU7K6BGz30qSqrpXVd0vSbr7vNl2qKqpPTf1rqr7JnlaRjOdLkny7KHs9CTvH5a3DusZtn+0u+82UwkAAACAyZtvqPTXSe47bf1+w9i+PCTJJVV1dZIrklzc3R/I6J5MZ1bV9ozumXTOUH9OkiOG8TOTnDXPXgEAAABYYGv2XzKr+0x7klu6+449M5X2pruvTnLsLOM3ZnR/pZnj305y6jz7AwAAAGARzXem0jeq6rF7VqrqcUm+tTAtAQAAALDczXem0kuSvLuq9txU+yFJnrMwLQEAAACw3M0rVOruK6rqx5L8aJJK8tnu/u6CdgYAAADAsjXfmUpJ8vgkG4ZjHFtV6e53LEhXAAAAACxr8wqVquq8JD+S5Kokdw3DnUSoBAAAAHAAmO9MpU1JjunuXshmAAAAAFgZ5vv0t88k+aGFbAQAAACAlWO+M5WOTHJdVX0yyXf2DHb3yQvSFQAAAADL2nxDpVcuZBMAAAAArCzzCpW6+2NV9dAkG7v7r6vqfkkOWtjWAAAAAFiu5nVPpap6QZILk7xlGDoqyfsWqikAAAAAlrf53qj7hUmOT/K1JOnuzyV50EI1BQAAAMDyNt9Q6Tvdfeeelapak6QXpiUAAOarqtZV1SVVdX1VXVtVL56lpqrqdVW1vaqurqrHTtt2elV9bvg5fbLdAwAryXxv1P2xqnp5kvtW/f/s3X+8XXV95/vX2ySUnxWQQBkgBmd4tLEZQe4ppUM6JSoUpYreK0rGq4yPtCmDTfW2M0NKeqVa4yM8psVW2oGGhhodDeIPChVQIoUysVfkpwIeFUpTSEMNym9UMPi5f+wVewwnydmHfc7a++zX8/FYj73Wd33XXp8vm3XOzud8f+Qk4Gzgb3oXliRJkiZpG/C7VXV7kv2A25JsqKqvj6nzWuCoZvtF4CLgF5McCJwHjND5g+FtSa6qqkentwmSJGkQTLan0grgYeAu4DeBa4Df71VQkiRJmpyqeqiqbm/2nwRG6cx/OdZpwEer48vA/kkOBX4V2FBVjzSJpA3AKdMYviRJGiCTXf3tR8AlzSZJkqQ+lGQ+8Erg5h1OHQY8OOZ4c1O2s3JJkqTnmVRSKck/Ms4cSlX1shcckSRJkl6wJPsCnwHeU1VP7Hh6nEtqF+U7vvcyYBnAvHnzXmCkknZl/oqr2w5hYLx4rzlthyANncnOqTQyZn9P4HTgwBcejiRJkl6oJHPoJJQ+XlWfHafKZuCIMceHA1ua8hN3KL9xx4urag2wBmBkZMTFWqQpsmn1qW2HMK75K67u29gkTa9JzalUVd8ds/1zVf0J8KoexyZJkqQuJQmwFhitqgt2Uu0q4B3NKnDHA49X1UPAF4CTkxyQ5ADg5KZMkiTpeSY7/O3YMYcvotNzab+eRCRJkqQX4gTg7cBdSe5sys4F5gFU1cV0Fll5HXAf8D3gnc25R5L8IXBLc937q+qRaYxdkiQNkMkOf/vjMfvbgE3AW15wNJIkSXpBqmoj48+NNLZOAe/ayblLgUunIDRJkjTDTHb1t8W9DkSSJEmSJEmDY7LD335nV+d3MX5fkiRJkiRJM8ALWf3tF+hM8gjweuAm4MFeBCVJkiRJkqT+Ntmk0kHAsVX1JECSPwA+VVW/3qvAJEmSJEmS1L9eNMnr5gHPjjl+Fpj/gqORJEmSJEnSQJhsT6WPAV9JcgVQwJuAj/YsKkmSJEmSJPW1ya7+tirJtcAvN0XvrKo7eheWJEmSJEmS+tlkh78B7A08UVV/CmxOcmSPYpIkSZIkSVKfm1RSKcl5wDnA7zVFc4D/1augJEmSJEmS1N8m21PpTcAbgKcBqmoLsF+vgpIkSZIkSVJ/m2xS6dmqKjqTdJNkn96FJEmSJEmSpH432aSSlOTHAAAgAElEQVTS5Un+Atg/yW8AXwQu6V1YkiRJkiRJ6meTXf3tj5KcBDwB/Czw3qra0NPIJEmSJEmS1Le6TiolmQV8oapeA5hIkiRJkiRJGkJdD3+rqueA7yV58RTEI0mSJEmSpAEwqeFvwA+Au5JsoFkBDqCqfrsnUUmSJEmSJKmvTTapdHWzDY35K3bf3E2rT52GSCRJkiRJktrXVVIpybyqeqCq1k1VQJIkSZIkSep/3c6p9Nfbd5J8psexSJIkSZIkaUB0m1TKmP2X9TIQSZIkSZIkDY5uk0q1k/1dSnJEkhuSjCa5J8m7m/IDk2xIcm/zekBTniQfTnJfkq8lObbLOCVJkoZSkkuTbE1y907O/7ckdzbb3UmeS3Jgc25Tkruac7dOb+SSJGnQdJtUOjrJE0meBF7R7D+R5MkkT+zium3A71bVAuB44F1JXg6sAK6vqqOA65tjgNcCRzXbMuCiLuOUJEkaVh8BTtnZyar6H1V1TFUdA/we8HdV9ciYKoub8yNTHKckSRpwXU3UXVWzJnOTqnoIeKjZfzLJKHAYcBpwYlNtHXAjcE5T/tGqKuDLSfZPcmjzPpIkSdqJqropyfwJVl8CrJ+6aCRJ0kzWbU+lF6z5kvNK4GbgkO2Joub14KbaYcCDYy7b3JRJkiSpB5LsTadH09jFVwq4LsltSZa1E5kkSRoUXfVUeqGS7Evni8t7quqJJDutOk7ZuHM4NV94lgHMmzevF2FKkiQNg9cDX9ph6NsJVbUlycHAhiTfqKqbdrzQ71+SJAmmsadSkjl0Ekofr6rPNsXfTnJoc/5QYGtTvhk4YszlhwNbxnvfqlpTVSNVNTJ37typCV6SJGnmOYMdhr5V1ZbmdStwBXDceBf6/UuSJME0JZXS6ZK0FhitqgvGnLoKOLPZPxO4ckz5O5pV4I4HHnc+JUmSpN5I8mLgV/jX714k2SfJftv3gZOBcVeQkyRJgukb/nYC8HbgriR3NmXnAquBy5MsBR4ATm/OXQO8DrgP+B7wzmmKU5IkaaAlWU9nIZSDkmwGzgPmAFTVxU21NwHXVdXTYy49BLiimZ5gNvCJqvr8dMUtSZIGz7QklapqI+PPkwTw6nHqF/CuKQ1KkiRpBqqqJROo8xHgIzuU3Q8cPTVRSZKkmWjaV3+TJEmSJEnS4DOpJEmSJEmSpK5N15xKktRz81dcvds6m1afOg2RSJIkSdLwsaeSJEmSJEmSumZSSZIkSZIkSV0zqSRJkiRJkqSumVSSJEmSJElS10wqSZIkSZIkqWsmlSRJkiRJktQ1k0qSJEmSJEnqmkklSZIkSZIkdc2kkiRJkiRJkrpmUkmSJEmSJEldM6kkSZIkSZKkrplUkiRJkiRJUtdMKkmSJEmSJKlrJpUkSZIkSZLUNZNKkiRJkiRJ6ppJJUmSJEmSJHXNpJIkSZIkSZK6ZlJJkiRpBklyaZKtSe7eyfkTkzye5M5me++Yc6ck+WaS+5KsmL6oJUnSIDKpJEmSNLN8BDhlN3X+d1Ud02zvB0gyC/hz4LXAy4ElSV4+pZFKkqSBZlJJkiRpBqmqm4BHJnHpccB9VXV/VT0LXAac1tPgJEnSjGJSSZIkafj8UpKvJrk2yc83ZYcBD46ps7kpkyRJGtfstgOQJEnStLodeGlVPZXkdcBfA0cBGadujfcGSZYBywDmzZs3VXFKkqQ+Z08lSZKkIVJVT1TVU83+NcCcJAfR6Zl0xJiqhwNbdvIea6pqpKpG5s6dO+UxS5Kk/mRSSZIkaYgk+ZkkafaPo/N98LvALcBRSY5MsgdwBnBVe5FKkqR+5/A3SZKkGSTJeuBE4KAkm4HzgDkAVXUx8GbgvyTZBnwfOKOqCtiW5LeALwCzgEur6p4WmiBJkgaESSVJkqQZpKqW7Ob8nwF/tpNz1wDXTEVckiRp5nH4myRJkiRJkrpmUkmSJEmSJEldM6kkSZIkSZKkrplUkiRJkiRJUtdMKkmSJEmSJKlrJpUkSZIkSZLUNZNKkiRJkiRJ6ppJJUmSJEmSJHXNpJIkSZIkSZK6ZlJJkiRJkiRJXTOpJEmSJEmSpK5NW1IpyaVJtia5e0zZgUk2JLm3eT2gKU+SDye5L8nXkhw7XXFKkiRJkiRp96azp9JHgFN2KFsBXF9VRwHXN8cArwWOarZlwEXTFKMkSZIkSZImYNqSSlV1E/DIDsWnAeua/XXAG8eUf7Q6vgzsn+TQ6YlUkiRJkiRJu9P2nEqHVNVDAM3rwU35YcCDY+ptbsokSZIkSZLUB9pOKu1MximrcSsmy5LcmuTWhx9+eIrDkiRJkiRJErSfVPr29mFtzevWpnwzcMSYeocDW8Z7g6paU1UjVTUyd+7cKQ1WkiRJkiRJHW0nla4Czmz2zwSuHFP+jmYVuOOBx7cPk5MkSZIkSVL7Zk/XjZKsB04EDkqyGTgPWA1cnmQp8ABwelP9GuB1wH3A94B3TleckiRJkiRJ2r1pSypV1ZKdnHr1OHULeNfURiRJkiRJkqTJanv4myRJknooyaVJtia5eyfn35bka83290mOHnNuU5K7ktyZ5Nbpi1qSJA0ik0qSJEkzy0eAU3Zx/h+BX6mqVwB/CKzZ4fziqjqmqkamKD5JkjRDTNvwN0mSJE29qropyfxdnP/7MYdfprPKriRJUtfsqSRJkjS8lgLXjjku4LoktyVZ1lJMkiRpQNhTSZIkaQglWUwnqbRoTPEJVbUlycHAhiTfqKqbxrl2GbAMYN68edMSryRJ6j/2VJIkSRoySV4B/CVwWlV9d3t5VW1pXrcCVwDHjXd9Va2pqpGqGpk7d+50hCxJkvqQSSVJkqQhkmQe8Fng7VX1rTHl+yTZb/s+cDIw7gpykiRJ4PA3SZKkGSXJeuBE4KAkm4HzgDkAVXUx8F7gJcD/TAKwrVnp7RDgiqZsNvCJqvr8tDdAkiQNDJNKkiRJM0hVLdnN+V8Hfn2c8vuBo6cqLkmSNPM4/E2SJEmSJEldM6kkSZIkSZKkrplUkiRJkiRJUtdMKkmSJEmSJKlrJpUkSZIkSZLUNZNKkiRJkiRJ6ppJJUmSJEmSJHXNpJIkSZIkSZK6ZlJJkiRJkiRJXTOpJEmSJEmSpK6ZVJIkSZIkSVLXTCpJkiRJkiSpayaVJEmSJEmS1DWTSpIkSZIkSeqaSSVJkiRJkiR1zaSSJEmSJEmSumZSSZIkSZIkSV0zqSRJkiRJkqSumVSSJEmSJElS10wqSZIkSZIkqWsmlSRJkmaQJJcm2Zrk7p2cT5IPJ7kvydeSHDvm3JlJ7m22M6cvakmSNIhMKkmSJM0sHwFO2cX51wJHNdsy4CKAJAcC5wG/CBwHnJfkgCmNVJIkDTSTSpIkSTNIVd0EPLKLKqcBH62OLwP7JzkU+FVgQ1U9UlWPAhvYdXJKkiQNOZNKkiRJw+Uw4MExx5ubsp2VS5IkjWt22wFIkiRpWmWcstpF+fPfIFlGZ+gc8+bN611kkqZMMt4j/gLe7/zevVfVuD9qJA0AeypJkiQNl83AEWOODwe27KL8eapqTVWNVNXI3LlzpyxQSb1TVX27SRpcJpUkSZKGy1XAO5pV4I4HHq+qh4AvACcnOaCZoPvkpkySJGlcDn+TJEmaQZKsB04EDkqymc6KbnMAqupi4BrgdcB9wPeAdzbnHknyh8AtzVu9v6p2NeG3JEkaciaVJEmSZpCqWrKb8wW8ayfnLgUunYq4JEnSzOPwN0mSJEmSJHXNpJIkSZIkSZK6ZlJJkiRJkiRJXevrpFKSU5J8M8l9SVa0HY8kSZIkSZI6+japlGQW8OfAa4GXA0uSvLzdqCRJkiRJkgR9nFQCjgPuq6r7q+pZ4DLgtJZjkiRJkiRJEv2dVDoMeHDM8eamTJIkSZIkSS2b3XYAu5Bxyup5lZJlwLLm8Kkk35yieA4CvrOrCjl/iu48fXbbxhnCds4cw/BcwnB8ljAc7RyGNpLzp7SdL52i99Uk3Xbbbd9J8k9txyFpWg3F7zNJP2Hc72D9nFTaDBwx5vhwYMuOlapqDbBmqoNJcmtVjUz1fdo0DG0E2zmTDEMbwXbOJMPQRhiedqqjqua2HYOk6eXPeUnb9fPwt1uAo5IcmWQP4AzgqpZjkiRJkiRJEn3cU6mqtiX5LeALwCzg0qq6p+WwJEmSJEmSRB8nlQCq6hrgmrbjaEz5ELs+MAxtBNs5kwxDG8F2ziTD0EYYnnZK0rDy57wkAFL1vLmvJUmSJEmSpF3q5zmVJEmSJEmS1KeGPqmU5NIkW5PcvZPzSfLhJPcl+VqSY8ecOzPJvc125vRF3Z0JtPFtTdu+luTvkxw95tymJHcluTPJrdMXdfcm0M4TkzzetOXOJO8dc+6UJN9sPucV0xd19ybQzv82po13J3kuyYHNuYH4PJMckeSGJKNJ7kny7nHqDPSzOcE2DvyzOcF2DvyzOcF2zoRnc88kX0ny1aad7xunzk8l+WTzmd2cZP6Yc7/XlH8zya9OZ+ySpJ/U/B7a/jvpU0n23kXdP0jyX6czPkkDoqqGegP+I3AscPdOzr8OuBYIcDxwc1N+IHB/83pAs39A2+2ZZBv/w/bYgddub2NzvAk4qO029KidJwKfG6d8FvAPwMuAPYCvAi9vuz2TbecOdV8P/O2gfZ7AocCxzf5+wLd2/EwG/dmcYBsH/tmcYDsH/tmcSDt3qD+oz2aAfZv9OcDNwPE71DkbuLjZPwP4ZLP/8uYz/CngyOazndV2m9zc3NyGdQOeGrP/ceB3dlH3D4D/2nbMbm5u/bcNfU+lqroJeGQXVU4DPlodXwb2T3Io8KvAhqp6pKoeBTYAp0x9xN3bXRur6u+bNgB8GTh8WgLrsQl8ljtzHHBfVd1fVc8Cl9H53PtSl+1cAqyfwnCmRFU9VFW3N/tPAqPAYTtUG+hncyJtnAnP5gQ/y50ZmGdzEu0c1Gezquqp5nBOs+04OeNpwLpm/9PAq5OkKb+sqp6pqn8E7qPzGUuS2ve/gX8HkOQdTS/pryb52I4Vk/xGklua85/Z3sMpyelNr6evJrmpKfv5pofrnc17HjWtrZI05YY+qTQBhwEPjjne3JTtrHzQLaXT+2O7Aq5LcluSZS3F1Eu/1PyiuzbJzzdlM/KzbH7BnwJ8ZkzxwH2ezdCZV9LpETHWjHk2d9HGsQb+2dxNO2fMs7m7z3PQn80ks5LcCWylk8Dd6bNZVduAx4GXMKCfpyTNdElm0+kRfVfzO3gl8KqqOhp43nBu4LNV9QvN+VE631EA3gv8alP+hqbsLOBPq+oYYITOz35JM8jstgMYABmnrHZRPrCSLKbzS2HRmOITqmpLkoOBDUm+0fSUGUS3Ay+tqqeSvA74a+AoZuBn2Xg98KWqGturaaA+zyT70vmH93uq6okdT49zycA9m7tp4/Y6A/9s7qadM+bZnMjnyYA/m1X1HHBMkv2BK5IsrKqxc7zNiGdTkobAXs0fCaDTU2kt8JvAp6vqOwA7/K7abmGSDwD7A/sCX2jKvwR8JMnlwGebsv8PWJnkcDrJqHunpimS2mJPpd3bDBwx5vhwYMsuygdSklcAfwmcVlXf3V5eVVua163AFQzwUIWqemL7sI2qugaYk+QgZthnOcYZ7DC8ZpA+zyRz6Pzj/ONV9dlxqgz8szmBNs6IZ3N37Zwpz+ZEPs/GQD+b21XVY8CNPH946Y8/t+av3y+mM2R3oD5PSRoC36+qY5pteTPUPOw+4f8R4Leq6t8D7wP2BKiqs4Dfp/Oz/s4kL6mqT9DptfR94AtJXjVFbZHUEpNKu3cV8I50HA88XlUP0cnIn5zkgCQHACfzr1n6gZJkHp2/Jry9qr41pnyfJPtt36fTxnFXHBsESX6mmdeDJMfR+f//u8AtwFFJjkyyB51/8F3VXqQvXJIXA78CXDmmbGA+z+ZzWguMVtUFO6k20M/mRNo4E57NCbZz4J/NCf4/OxOezblNDyWS7AW8BvjGDtWuAravuvhmOhOSV1N+Rjqrwx1JpzfaV6YncknSBF0PvCXJSwDSrFK6g/2Ah5o/prxte2GSf1tVN1fVe4HvAEckeRlwf1V9mM7vgVdMeQskTauhH/6WZD2dlYcOSrIZOI/OxKNU1cXANXRWmboP+B7wzubcI0n+kM4/egDev5Puoa2bQBvfS2e+i//Z/LtuW1WNAIfQGdoAnf9XPlFVn5/2BkzQBNr5ZuC/JNlG568lZzT/0NmW5LfoJB5mAZdW1T0tNGFCJtBOgDcB11XV02MuHaTP8wTg7XTG9m/vln0uMA9mzLM5kTbOhGdzIu2cCc/mRNoJg/9sHgqsSzKLTvLv8qr6XJL3A7dW1VV0kmsfS3IfnR5KZwBU1T3NkIivA9uAdzVD6SRJfaL5Wb0K+LskzwF3AP95h2r/L515A/8JuItOkgngf6QzEXfoJKe+CqwA/u8kPwT+BXj/lDdC0rRK53u7JEmSJEmSNHEOf5MkSZIkSVLXTCpJkiRJkiSpayaVJEmSJEmS1DWTSpIkSZIkSeqaSSVJkiRJkiR1zaSSJEmSJEmSumZSSZIkSZIkSV0zqSRJkiRJkqSumVSSJEmSJElS10wqSZIkSZIkqWsmlSRJkiRJktQ1k0qSJEmSJEnqmkklSZIkSZIkdc2kkiRJkiRJkrpmUkmSJEmSJEldM6kkSZIkSZKkrplUkiRJkiRJUtdMKkmSJEmSJKlrJpUkSZIkSZLUNZNKkiRJkiRJ6ppJJUl9K8mNSR5N8lNtxyJJkjQM/P4lqRsmlST1pSTzgV8GCnhDq8FIkiQNAb9/SeqWSSVJ/eodwJeBjwBnbi9M8pIkf5PkiSS3JPlAko1jzv9ckg1JHknyzSRvmf7QJUmSBpLfvyR1ZXbbAUjSTrwDuAC4GfhykkOq6tvAnwNPAz8DzAe+APwTQJJ9gA3Ae4HXAq8ArktyT1XdM+0tkCRJGix+/5LUFXsqSeo7SRYBLwUur6rbgH8A/lOSWcD/BZxXVd+rqq8D68Zc+mvApqr6q6raVlW3A58B3jzNTZAkSRoofv+SNBkmlST1ozOB66rqO83xJ5qyuXR6WD44pu7Y/ZcCv5jkse0b8DY6f1WTJEnSzvn9S1LXHP4mqa8k2Qt4CzAryb80xT8F7A8cAmwDDge+1Zw7YszlDwJ/V1UnTVO4kiRJA8/vX5ImK1XVdgyS9GNJltAZt38M8OyYU5cDt9D5QvMc8OvAPOA64IGqWpRkP+Bu4PeBy5rrjgGeqqrR6WmBJEnSYPH7l6TJcvibpH5zJvBXVfVAVf3L9g34MzpdqX8LeDHwL8DHgPXAMwBV9SRwMnAGsKWpcz6dv7RJkiRpfH7/kjQp9lSSNNCSnA/8TFWdudvKkiRJesH8/iVpO3sqSRooSX4uySvScRywFLii7bgkSZJmKr9/SdoZJ+qWNGj2o9Pl+t8AW4E/Bq5sNSJJkqSZze9fksbl8DdJkiRJkiR1zeFvkiRJkiRJ6ppJJUmSJEmSJHVtRs2pdNBBB9X8+fPbDkOSJE2R22677TtVNbftOPSv/P4lSdLMt7PvYDMqqTR//nxuvfXWtsOQJElTJMk/tR2DfpLfvyRJmvl29h3M4W+SJEmSJEnqmkklSZIkSZIkdc2kkiRJkiRJkrpmUkmSJEmSJEldM6kkSZIkSZKkrplUkiRJkiRJUtdMKkmSJA2hJP9PknuS3J1kfZI9kxyZ5OYk9yb5ZJI92o5TkiT1L5NKkiRJQybJYcBvAyNVtRCYBZwBnA98qKqOAh4FlrYXpaR+s379ehYuXMisWbNYuHAh69evbzskSS0zqSRJkjScZgN7JZkN7A08BLwK+HRzfh3wxpZik9Rn1q9fz8qVK7nwwgv5wQ9+wIUXXsjKlStNLElDzqSSJEnSkKmqfwb+CHiATjLpceA24LGq2tZU2wwc1k6EkvrNqlWrWLt2LYsXL2bOnDksXryYtWvXsmrVqrZDk9Qik0qSJElDJskBwGnAkcC/AfYBXjtO1drJ9cuS3Jrk1ocffnjqApXUN0ZHR1m0aNFPlC1atIjR0dGWIpLUD2a3HQBAkk3Ak8BzwLaqGklyIPBJYD6wCXhLVT3aVozqP/NXXL3bOptWnzoNkUiSNHBeA/xjVT0MkOSzwH8A9k8yu+mtdDiwZbyLq2oNsAZgZGRk3MSTpJllwYIFbNy4kcWLF/+4bOPGjSxYsKDFqCS1rZ96Ki2uqmOqaqQ5XgFc30wUeX1zLEmSpBfuAeD4JHsnCfBq4OvADcCbmzpnAle2FJ+kPrNy5UqWLl3KDTfcwA9/+ENuuOEGli5dysqVK9sOTVKL+qKn0k6cBpzY7K8DbgTOaSsYSZKkmaKqbk7yaeB2YBtwB52eR1cDlyX5QFO2tr0oJfWTJUuWALB8+XJGR0dZsGABq1at+nG5pOHUL0mlAq5LUsBfNF2qD6mqhwCq6qEkB7caoSRJ0gxSVecB5+1QfD9wXAvhSBoAS5YsMYkk6Sf0S1LphKra0iSONiT5xkQvTLIMWAYwb968qYpPkiRJkiRJY/TFnEpVtaV53QpcQecvZN9OcihA87p1J9euqaqRqhqZO3fudIUsSZIkSZI01FpPKiXZJ8l+2/eBk4G7gavoTBAJThQpSZIkSZLUV/ph+NshwBWdhUeYDXyiqj6f5Bbg8iRL6axQcnqLMUqSJEmSJGmM1pNKVXU/cPQ45d+ls7ytJEmSJEmS+kzrw98kSZIkSZI0eEwqSZIkSZIkqWsmlSRJkiRJktQ1k0qSJEmSJEnqmkklSZIkSZIkdc2kkiRJkiRpt9avX8/ChQuZNWsWCxcuZP369W2HJKlls9sOQJIkSZLU39avX8/KlStZu3YtixYtYuPGjSxduhSAJUuWtBydpLbYU0mSJEmStEurVq1i7dq1LF68mDlz5rB48WLWrl3LqlWr2g5NUotMKkmSJEmSdml0dJRFixb9RNmiRYsYHR1tKSJJ/cCkkiRJkiRplxYsWMDGjRt/omzjxo0sWLCgpYgk9QPnVJL60PwVV0+o3qbVp05xJJIkSRKsXLmSpUuXPm9OJYe/ScPNpJIkSZIkaZe2T8a9fPlyRkdHWbBgAatWrXKSbmnImVSSJEmSJO3WkiVLTCJJ+gnOqSRJkiRJ2q3ly5ez5557koQ999yT5cuXtx2SpJbZU0maoInMc+QcR5IkSZqJli9fzsUXX8z555/PWWedxcUXX8w555wDwIUXXthydJLaYk8lSZKkIZPkZ5PcOWZ7Isl7khyYZEOSe5vXA9qOVVJ/uOSSS3jrW9/KpZdeyn777cell17KW9/6Vi655JK2Q5PUInsqSZIkDZmq+iZwDECSWcA/A1cAK4Drq2p1khXN8TmtBSqpbzzzzDNcddVV/OAHP+BHP/oR3/rWt3jggQd45pln2g5NUovsqSRJkjTcXg38Q1X9E3AasK4pXwe8sbWoJPWdp556itWrV/P000+zevVqnnrqqbZDktQyk0qSJEnD7QxgfbN/SFU9BNC8HtxaVJIkqe+ZVJIkSRpSSfYA3gB8qsvrliW5NcmtDz/88NQEJ6nvnHrqqZx77rnss88+nHvuuZx6qovUSMPOpJIkSdLwei1we1V9uzn+dpJDAZrXreNdVFVrqmqkqkbmzp07TaFKatPs2bP50pe+xLXXXsuzzz7Ltddey5e+9CVmz3aaXmmYmVSSJEkaXkv416FvAFcBZzb7ZwJXTntEkvrSWWedxeOPP86SJUvYY489WLJkCY8//jhnnXVW26FJapFpZUmSpCGUZG/gJOA3xxSvBi5PshR4ADi9jdgk9Z8LL7wQgEsuuQSAxx57jLPPPvvH5ZKGk0klSZKkIVRV3wNeskPZd+msBidJz3PhhReaRJL0Exz+JkmSJEmSpK6ZVJIkSZIkSVLXTCpJkiRJkiSpayaVJEmSJEmS1DWTSpIkSZKk3Vq/fj0LFy5k1qxZLFy4kPXr17cdkqSWufqbJEmSJGmX1q9fz8qVK1m7di2LFi1i48aNLF26FIAlS5a0HJ2ktthTSZIkSZK0S6tWrWLt2rUsXryYOXPmsHjxYtauXcuqVavaDk1Si0wqSZIkSZJ2aXR0lA9+8IO86EUvIgkvetGL+OAHP8jo6GjboUlqkUklSZIkSdIu7bXXXnzxi1/krLPO4rHHHuOss87ii1/8InvttVfboUlqkUklSZIkSdIuPf300+y7776cfvrp7L333px++unsu+++PP30022HJqlFJpUkSZIkSbt1wQUXsHz5cvbcc0+WL1/OBRdc0HZIklpmUkmSJEmStEtJuOOOO7j77rt57rnnuPvuu7njjjtI0nZoklpkUkmSJEmStEsnnXQSF110EWeffTaPP/44Z599NhdddBEnnXRS26FJatHstgOQJEmSJE2tXvUouuiii7jooot+fHzddde94PeuqhcalqSW2FNJkiRJkma4qurZ9tJzPtfT95M0uPomqZRkVpI7knyuOT4yyc1J7k3yySR7tB2jJEmSJEmSOvomqQS8Gxgdc3w+8KGqOgp4FFjaSlSSJEmSJEl6nr5IKiU5HDgV+MvmOMCrgE83VdYBb2wnOkmSJEmSJO2oL5JKwJ8A/x34UXP8EuCxqtrWHG8GDmsjMEmSJEmSJD1f66u/Jfk1YGtV3ZbkxO3F41Qddwa3JMuAZQDz5s2bkhg1uOavuHpC9TatPnWKI5EkSZIkaWbph55KJwBvSLIJuIzOsLc/AfZPsj3pdTiwZbyLq2pNVY1U1cjcuXOnI15JkiRJkqSh13pSqap+r6oOr6r5wBnA31bV24AbgDc31c4ErmwpREmSpBknyf5JPp3kG0lGk/xSkgOTbGhW392Q5IC245QkSf2r9aTSLpwD/E6S++jMsbS25XgkSZJmkj8FPl9VPwccTWcV3hXA9c3qu9c3x5IkSeNqfU6lsarqRuDGZv9+4Lg245EkSTQti8gAACAASURBVJqJkvw08B+B/wxQVc8CzyY5DTixqbaOzveyc6Y/QkmSNAj6uaeSJEmSpsbLgIeBv0pyR5K/TLIPcEhVPQTQvB7cZpCSJKm/mVSSJEkaPrOBY4GLquqVwNN0MdQtybIktya59eGHH56qGCVJUp8zqSRJkjR8NgObq+rm5vjTdJJM305yKEDzunW8i119V5IkgUklSZKkoVNV/wI8mORnm6JXA18HrqKz6i64+q4kSdqNvpqoW5IkSdNmOfDxJHsA9wPvpPMHx8uTLAUeAE5vMT5JktTnTCpJkiQNoaq6ExgZ59SrpzsWSZI0mBz+JkmSJEmSpK6ZVJIkSZIkSVLXTCpJkiRJkiSpayaVJEmSJEmS1DWTSpIkSZIkSeqaSSVJkiRJkiR1zaSSJEmSJEmSumZSSZIkSZIkSV2b3XYA0rCZv+LqtkOQJEmSJOkFs6eSJEmSJEmSumZSSZIkSZIkSV0zqSRJkiRJkqSumVSSJEmSJElS15yoW5rhJjIx+KbVp05DJJIkSZKkmcSeSpIkSZIkSeqaSSVJkiRJkiR1zaSSJEmSJEmSumZSSZIkSZIkSV1zom5JEzKRCb/BSb8lSZIkaVjYU0mSJEmSJElds6eSJEnSEEqyCXgSeA7YVlUjSQ4EPgnMBzYBb6mqR9uKUZIk9Td7KkmSJA2vxVV1TFWNNMcrgOur6ijg+uZYkiRpXCaVJEmStN1pwLpmfx3wxhZjkSRJfc6kkiRJ0nAq4LoktyVZ1pQdUlUPATSvB7cWnSRJ6nvOqSRJkjScTqiqLUkOBjYk+cZEL2ySUMsA5s2bN1XxSZKkPmdPJUmSpCFUVVua163AFcBxwLeTHArQvG7dybVrqmqkqkbmzp07XSFLkqQ+Y1JJkiRpyCTZJ8l+2/eBk4G7gauAM5tqZwJXthOhJEkaBA5/kyRJGj6HAFckgc73wU9U1eeT3AJcnmQp8ABweosxSpKkPtfTpFKShVV1dy/fU5IkSb1VVfcDR49T/l3g1dMfkSRJGkS9Hv52cZKvJDk7yf49fm9JkiRJkiT1iZ4mlapqEfA24Ajg1iSfSHJSL+8hSZIkSZKk9vV8ou6quhf4feAc4FeADyf5RpL/s9f3kiRJkiRJUjt6mlRK8ookHwJGgVcBr6+qBc3+h3p5L0mSJEmSJLWn1z2V/gy4HTi6qt5VVbcDVNUWOr2XnifJns08TF9Nck+S9zXlRya5Ocm9ST6ZZI8exypJkiRJkqRJ6nVS6XV0lqT9PkCSFyXZG6CqPraTa54BXlVVRwPHAKckOR44H/hQVR0FPAos7XGskiRJkiRJmqReJ5W+COw15njvpmynquOp5nBOsxWdIXOfbsrXAW/sbaiSJEmSJEmarF4nlfYckyCi2d97dxclmZXkTmArsAH4B+CxqtrWVNkMHNbjWCVJkiRJkjRJvU4qPZ3k2O0HSf4P4Pu7u6iqnquqY4DDgeOABeNVG+/aJMuS3Jrk1ocffniSYUuSJEmSJKkbs3v8fu8BPpVkS3N8KPDWiV5cVY8luRE4Htg/yeymt9LhwJadXLMGWAMwMjIybuJJkiRJkiRJvdXTpFJV3ZLk54CfBQJ8o6p+uKtrkswFftgklPYCXkNnku4bgDcDlwFnAlf2MlZJkiRJkiRNXq97KgH8AjC/ee9XJqGqPrqL+ocC65LMojMc7/Kq+lySrwOXJfkAcAewdgpilSRJkiRJ0iT0NKmU5GPAvwXuBJ5rigvYaVKpqr4GvHKc8vvpzK8kSZIkSZKkPtPrnkojwMuryrmNpCE1f8XVu62zafWp0xCJJEmSJGkq9Xr1t7uBn+nxe0qSJEmSJKnP9Lqn0kHA15N8BXhme2FVvaHH95EkSZIkSVKLep1U+oMev58kSZIkSZL6UE+TSlX1d0leChxVVV9Msjcwq5f3kCRJkiRJUvt6OqdSkt8APg38RVN0GPDXvbyHJEmSJEmS2tfribrfBZwAPAFQVfcCB/f4HpIkSZIkSWpZr5NKz1TVs9sPkswGqsf3kCRJUg8kmZXkjiSfa46PTHJzknuTfDLJHm3HKEmS+levk0p/l+RcYK8kJwGfAv6mx/eQJElSb7wbGB1zfD7woao6CngUWNpKVJIkaSD0evW3FXS+fNwF/CZwDfCXPb6H1HPzV1zddgiTMqhxT8RE2rZp9anTEIkkzUxJDgdOBVYBv5MkwKuA/9RUWUdnZd+LWglQkiT1vV6v/vYj4JJmkyRJUv/6E+C/A/s1xy8BHquqbc3xZjqLrkiSJI2rp0mlJP/IOHMoVdXLenkfSZIkTV6SXwO2VtVtSU7cXjxO1XHnxkyyDFgGMG/evCmJUZIk9b9eD38bGbO/J3A6cGCP7yFJkqQX5gTgDUleR+c720/T6bm0f5LZTW+lw4Et411cVWuANQAjIyMuyiJJ0pDq6UTdVfXdMds/V9Wf0BmbL0mSpD5RVb9XVYdX1XzgDOBvq+ptwA3Am5tqZwJXthSiJEkaAL0e/nbsmMMX0em5tN9OqkuSJKm/nANcluQDwB3A2pbjkSRJfazXw9/+eMz+NmAT8JYe30OSJEk9UlU3Ajc2+/cDx7UZjyRJGhy9Xv1tcS/fT5IkSZIkSf2p18PffmdX56vqgl7eT5IkSZJmqqPfdx2Pf/+HbYcxrvkrrm47hOd58V5z+Op5J7cdhjRUpmL1t18ArmqOXw/cBDzY4/tIkiRJ0oz2+Pd/yKbVp7YdxsDox0SXNNP1Oql0EHBsVT0JkOQPgE9V1a/3+D6SJEmSJElq0Yt6/H7zgGfHHD8LzO/xPSRJkiRJktSyXvdU+hjwlSRXAAW8Cfhoj+8hSZIkSZKklvV69bdVSa4FfrkpemdV3dHLe0iSJEmSJKl9vR7+BrA38ERV/SmwOcmRU3APSZIkSZIktainPZWSnEdnBbifBf4KmAP8L+CEXt6nDRNZScCVGeSKE5IkSZKkYdHrnkpvAt4APA1QVVuA/Xp8D0mSJEmSJLWs10mlZ6uq6EzSTZJ9evz+kiRJkiRJ6gO9TipdnuQvgP2T/AbwReCSHt9DkiRJkiRJLev16m9/lOQk4Ak68yq9t6o29PIekiRJkiRJal/PkkpJZgFfqKrXACaSJEmSJEmSZrCeDX+rqueA7yV5ca/eU5IkSZIkSf2pp8PfgB8AdyXZQLMCHEBV/XaP7yNJkiRJkqQW9TqpdHWzSZIkSZIkaQbrSVIpybyqeqCq1vXi/TSzzV+x+7zjptWnTkMkkiRJkiRpsno1p9Jfb99J8pkevackSZIkSZL6VK+SShmz/7IevackSZIkSZL6VK+SSrWTfUmSJPWhJHsm+UqSrya5J8n7mvIjk9yc5N4kn0yyR9uxSpKk/tSrpNLRSZ5I8iTwimb/iSRPJnmiR/eQJElS7zwDvKqqjgaOAU5JcjxwPvChqjoKeBRY2mKMkiSpj/UkqVRVs6rqp6tqv6qa3exvP/7pXtxDkiRJvVMdTzWHc5qtgFcBn27K1wFvbCE8SZI0AHrVU2nSkhyR5IYko03X63c35Qcm2dB0vd6Q5IC2Y5UkSZpJksxKciewFdgA/APwWFVta6psBg5rKz5JktTfWk8qAduA362qBcDxwLuSvBxYAVzfdL2+vjmWJElSj1TVc1V1DHA4cBywYLxqOxYkWZbk1iS3Pvzww1MdpiRJ6lOtJ5Wq6qGqur3ZfxIYpfMXsdPodLkGu15LkiRNmap6DLiRzh/49k8yuzl1OLBlnPprqmqkqkbmzp07fYFKkqS+0npSaawk84FXAjcDh1TVQ9BJPAEHtxeZJEnSzJJkbpL9m/29gNfQ+ePeDcCbm2pnAle2E6EkSep3s3dfZXok2Rf4DPCeqnoiyUSvWwYsA5g3b97UBShJkjSzHAqsSzKLzh8aL6+qzyX5OnBZkg8AdwBr2wxSkiT1r75IKiWZQyeh9PGq+mxT/O0kh1bVQ0kOpTOB5PNU1RpgDcDIyMjzxvxLkiTp+arqa3R6iO9Yfj+d+ZUkSZJ2qfXhb+l0SVoLjFbVBWNOXUWnyzXY9VqSJEmSJKmv9ENPpROAtwN3NUvaApwLrAYuT7IUeAA4vaX4JEmSJGna7bdgBf9+nYtgT9R+CwBObTsMaai0nlSqqo3AziZQevV0xiJJkiRJ/eLJ0dVsWm2SZKLmr7i67RCkodP68DdJkiRJkiQNHpNKkiRJkiRJ6ppJJUmSJEmSJHXNpJIkSZIkSZK6ZlJJkiRJkiRJXWt99TdJmkoTXQWkVyurTOR+ruIiSZIkaSawp5IkSZIkSZK6ZlJJkiRJkiRJXTOpJEmSJEmSpK45p5KkCc87NOym+7+T8zNJkiRJ6mf2VJIkSZIkSVLXTCpJkiRJkiSpayaVJEmSJEmS1DWTSpIkSZIkSeqaSSVJkiRJkiR1zaSSJEmSJEmSumZSSZIkSZIkSV0zqSRJkiRJkqSumVSSJEkaMkmOSHJDktEk9yR5d1N+YJINSe5tXg9oO1ZJktS/TCpJkiQNn/+fvbuPsuys6wX//ZlOhCRgQBpu7HTTwWQiGVQIZQSiXCWgBBiC9wZNRjEyaN+5ooIwIw3juurMdU1co7w4crm3BSQo8hZAMiQqIYJeQAKdEHkLSAhN0nRLGnkJASQGfvPH2d1UmupOTlF19umqz2ets87ez9nn1O/pqjq963ue59m3J3l2dz8wycOSPL2qTk+yPcmV3X1qkiuHfQCAJQmVAADWme7e293XDNtfSnJdkk1Jzk1y8XDYxUmeNE6FAMCRYMPYBQAAMJ6q2prkIUmuSnK/7t6bTIKnqrrviKUBSbZuv2zsEo4Y33X3o8cuAdYdoRIAwDpVVccneUOSZ3b3LVV1V5+3Lcm2JNmyZcvqFQjr3K6LHj92CUvauv2yua0NmC3T3wAA1qGqOjqTQOlV3f3GofkzVXXi8PiJSW5e6rndvaO7F7p7YePGjbMpGACYO0YqzdhdGb4q9QcAVlNNhiS9LMl13f38RQ9dmuTCJBcN928eoTwA4AghVAIAWH/OSvKUJB+sqmuHtudlEia9rqqeluTGJE8eqT4A4AggVAIAWGe6+51JDrWA0tmzrAUAOHJZUwkAAACAqRmpxFxy6VQAAACYb0YqAQAAADA1oRIAAAAAUxMqAQAAADA1oRIAAAAAUxMqAQAAADA1oRIAAAAAUxMqAQAAADA1oRIAAAAAUxMqAQAAADC1DWMXAKw/W7dfNnYJAAAAfJtGH6lUVS+vqpur6kOL2u5dVVdU1ceH+3uNWSMAAAAAdzR6qJTkFUkee1Db9iRXdvepSa4c9gEAAACYE6OHSt39d0k+d1DzuUkuHrYvTvKkmRYFAAAAwGGNHiodwv26e2+SDPf3HbkeAAAAABaZ11DpLquqbVW1s6p27tu3b+xyAAAAANaFeQ2VPlNVJybJcH/zoQ7s7h3dvdDdCxs3bpxZgQAAAADr2byGSpcmuXDYvjDJm0esBQAAAICDjB4qVdWrk/x9ktOqandVPS3JRUkeU1UfT/KYYR8AAACAObFh7AK6+4JDPHT2TAsBAFhHqurlSZ6Q5ObuftDQdu8kr02yNcmuJD/d3Z8fq0YAYL6NHioBzIOt2y8bu4RVc1f6tuuix8+gEmDOvCLJHyV55aK27Umu7O6Lqmr7sP+cEWoDAI4Ao09/AwBg9rr775J87qDmc5NcPGxfnORJMy0KADiiCJUAANjvft29N0mG+/uOXA8AMMeESgAATKWqtlXVzqrauW/fvrHLAQBGYk2lNc5aKqxla3kdJICRfKaqTuzuvVV1YpKblzqou3ck2ZEkCwsLPcsCAYD5YaQSAAD7XZrkwmH7wiRvHrEWAGDOCZUAANahqnp1kr9PclpV7a6qpyW5KMljqurjSR4z7AMALMn0NwCAdai7LzjEQ2fPtBAA4IhlpBIAAAAAUzNS6Qg2b4sUz1s9wF23kov6r9RrudAAAADMNyOVAAAAAJiaUAkAAACAqQmVAAAAAJiaUAkAAACAqVmoewUdqQtVWwwXZmsl3yuO1PcdAADgyGekEgAAAABTEyoBAAAAMDWhEgAAAABTEyoBAAAAMDWhEgAAAABTEyoBAAAAMDWhEgAAAABTEyoBAAAAMDWhEgAAAABT2zB2AQCsP1u3XzbT19l10eNX5OsBAADfZKQSAAAAAFMTKgEAAAAwNaESAAAAAFOzptIcWqm1RlbSPNYEcFfdlfewWa67ZC0o5l1VPTbJi5IcleSl3X3RyCUBAHPISCUAAA6oqqOSvDjJOUlOT3JBVZ0+blUAwDwSKgEAsNiZSa7v7hu6+7Ykr0ly7sg1AQBzyPQ3AAAW25TkpkX7u5P88Ei1ACukqlb29X5v5V6ru1fuxYCZEioBALDYUn953uEvvqralmRbkmzZsmUWNQHfJsENsBqESgDcJRbsv2v8O7EG7E6yedH+SUn2LD6gu3ck2ZEkCwsL/lIFgHXKmkoAACz2viSnVtXJVXVMkvOTXDpyTQDAHDJSCQCAA7r79qr6lSR/neSoJC/v7g+PXBYAMIeESgAA3EF3X57k8rHrAADmm+lvAAAAAEzNSCUAyF1bYHvXRY+fQSXTmXXdR+q/EwAAK2+uRypV1WOr6mNVdX1VbR+7HgAAAAAm5jZUqqqjkrw4yTlJTk9yQVWdPm5VAAAAACRzHColOTPJ9d19Q3ffluQ1Sc4duSYAAAAAMt+h0qYkNy3a3z20AQAAADCy6u6xa1hSVT05yU929y8O+09JcmZ3/+pBx21Lsm3YPS3Jx1awjPsk+ewKvt680s+1RT/XFv1cW/Tz23f/7t64Sq/NMlTVviSfGrsOYKbWy/9nwDcteQ42z1d/251k86L9k5LsOfig7t6RZMdqFFBVO7t7YTVee57o59qin2uLfq4t+slaJOSD9cf7PLDfPE9/e1+SU6vq5Ko6Jsn5SS4duSYAAAAAMscjlbr79qr6lSR/neSoJC/v7g+PXBYAAAAAmeNQKUm6+/Ikl49YwqpMq5tD+rm26Ofaop9ri34CsBZ4nweSzPFC3QAAAADMr3leUwkAAACAOSVUOoSqemxVfayqrq+q7WPXs1Kq6uVVdXNVfWhR272r6oqq+vhwf68xa1wJVbW5qt5eVddV1Yer6hlD+5rqa1XdrareW1X/MPTzd4b2k6vqqqGfrx0Wuz+iVdVRVfX+qnrLsL8W+7irqj5YVddW1c6hbU39zCZJVZ1QVZdU1UeH39GHr7V+VtVpw/dx/+2WqnrmWutnklTVrw/vPx+qqlcP70tr7vcTYD2pqp+qqq6q7xu7FmC+CZWWUFVHJXlxknOSnJ7kgqo6fdyqVswrkjz2oLbtSa7s7lOTXDnsH+luT/Ls7n5gkoclefrwPVxrff1akkd19w8meXCSx1bVw5L8XpIXDP38fJKnjVjjSnlGkusW7a/FPibJj3f3gxddpnet/cwmyYuS/FV3f1+SH8zk+7qm+tndHxu+jw9O8tAkX0nypqyxflbVpiS/lmShux+UyYU1zs/a/f0EWC8uSPLOTN7TAQ5JqLS0M5Nc3903dPdtSV6T5NyRa1oR3f13ST53UPO5SS4eti9O8qSZFrUKuntvd18zbH8pkz9aN2WN9bUnbh12jx5uneRRSS4Z2o/4flbVSUken+Slw35ljfXxMNbUz2xV3TPJI5O8LEm6+7bu/kLWWD8PcnaST3T3p7I2+7khyd2rakOSY5Pszfr5/QRYc6rq+CRnZfKBwPlD23dU1X8ZRqa+paour6rzhsceWlV/W1VXV9VfV9WJI5YPzJhQaWmbkty0aH/30LZW3a+79yaTMCbJfUeuZ0VV1dYkD0lyVdZgX4dpYdcmuTnJFUk+keQL3X37cMha+Pl9YZLfSPKNYf+7s/b6mEwCwbcOJ2Xbhra19jP7gCT7kvzJMJ3xpVV1XNZePxc7P8mrh+011c/u/nSS309yYyZh0heTXJ21+fsJsF48KZMRxf+Y5HNVdUaSf5dka5LvT/KLSR6eJFV1dJL/N8l53f3QJC9P8rtjFA2MY8PYBcypWqLNZfKOQMMnLW9I8szuvmUywGVt6e6vJ3lwVZ2QyfSaBy512GyrWjlV9YQkN3f31VX1Y/ublzj0iO3jImd1956qum+SK6rqo2MXtAo2JDkjya9291VV9aIc4VPADmdYS+iJSZ47di2rYVgT6twkJyf5QpLXZzJ1/GBr4fcTYL24IJMP9JLJjI0LMhkN//ru/kaSf6qqtw+Pn5bkQZmctySTadB7Z1suMCah0tJ2J9m8aP+kJHtGqmUWPlNVJ3b33mG46s1jF7QShk9O3pDkVd39xqF5TfY1Sbr7C1X1jkzWkDqhqjYMIwWO9J/fs5I8saoel+RuSe6ZyYnOWupjkqS79wz3N1fVmzKZirvWfmZ3J9nd3VcN+5dkEiqttX7ud06Sa7r7M8P+Wuvno5N8srv3JUlVvTHJI7IGfz8B1oOq+u5MpjA/qKo6k5CoM/ngcsmnJPlwdz98RiUCc8b0t6W9L8mpw9Vrjslk6sKlI9e0mi5NcuGwfWGSN49Yy4oY1tx5WZLruvv5ix5aU32tqo3DCKVU1d0z+QPvuiRvT3LecNgR3c/ufm53n9TdWzP5Xfyb7v7ZrKE+JklVHVdV99i/neQnknwoa+xntrv/KclNVXXa0HR2ko9kjfVzkQvyzalvydrr541JHlZVxw7vu/u/n2vq9xNgHTkvySu7+/7dvbW7Nyf5ZJLPJvn3w9pK90vyY8PxH0uysaoOTIerqv9xjMKBcVS3EelLGUZFvDCTdP7l3b0m5gZX1asz+U/gPkk+k+S3kvxFktcl2ZLJHwhP7u6DF/M+olTVjyT570k+mG+uw/O8TNZVWjN9raofyGQR3KMyCYlf193/Z1U9IJPhyvdO8v4kP9fdXxuv0pUxTH/737r7CWutj0N/9n8KuCHJn3f37w6fGK6Zn9kkqaoHZ7Lo+jFJbkjy1Aw/v1lb/Tw2k/X5HtDdXxza1uL383eS/EwmV918fyZrbWzKGvr9BFgvhlHvF3X3Xy1q+7VMlleoTC628Y9JvjPJ87v7iuH/9T9M8l2ZnMO8sLv/eNa1A+MQKgEAAHBYVXV8d986fEDy3kzWgvynsesCxmVNJQAAAO7MW4ZlF45J8n8JlIDESCUAAAAAlsFC3QAAAABMTagEAAAAwNSESgAAAABMTagEAAAAwNSESgAAAABMTagEAAAAwNSESgAAAABMTagEAAAAwNSESgAAAABMTagEAAAAwNSESgAAAABMTagEAAAAwNSESgAAAABMTagEAAAAwNSESgAAAABMTagEAAAAwNSESgAAAABMTagEzKWq+tmqeuui/a6qU8asCQBgrXMOBkxDqASMqqp+pKreXVVfrKrPVdW7quqHuvtV3f0Td/E1jqmqP6iq3VV1a1V9sqpesNq1AwAcqZyDASthw9gFAOtXVd0zyVuS/Mckr0tyTJIfTfK1KV/quUkWkpyZZG+S+yd55MpVCgCwdjgHA1aKkUrAmP6HJOnuV3f317v7q9391u7+QFX9QlW986DjH1dVN1TVZ6vq/6mq/e9hP5TkTd29pyd2dfcr9z+pqnZV1XOr6iNV9fmq+pOqutuM+ggAMG+cgwErQqgEjOkfk3y9qi6uqnOq6l53cvxPZfJp2BlJzk3yvwzt70nyrKr65ar6/qqqJZ77s0l+Msn3ZnIi9Zsr0gMAgCOPczBgRQiVgNF09y1JfiRJJ/njJPuq6tKqut8hnvJ73f257r4xyQuTXDC0/99Jfi+Tk5adST5dVRce9Nw/6u6buvtzSX530XMBANYV52DAShEqAaPq7uu6+xe6+6QkD0ryPZmcrCzlpkXbnxqOzTBs+8XdfVaSEzI5YXl5VT3wzp4LALAeOQcDVoJQCZgb3f3RJK/I5MRmKZsXbW9JsmeJ1/hqd784yeeTnD7NcwEA1iPnYMByCZWA0VTV91XVs6vqpGF/cyZDot9ziKf871V1r+G4ZyR57fC8Z1bVj1XV3atqwzDs+h5J3r/ouU+vqpOq6t5Jnrf/uQAA641zMGClCJWAMX0pyQ8nuaqqvpzJicyHkjz7EMe/OcnVSa5NclmSlw3tX03yB0n+Kclnkzw9yb/v7hsWPffPk7w1yQ3D7T+vaE8AAI4czsGAFVHdPXYNAKuqqnYl+cXuftvYtQAArBfOwWDtM1IJAAAAgKkJlQAAAACYmulvAAAAAEzNSCUAAAAApiZUAgAAAGBqG8YuYCXd5z736a1bt45dBgCwSq6++urPdvfGsevgm5x/AcDad6hzsDUVKm3dujU7d+4cuwwAYJVU1afGroE7cv4FAGvfoc7BTH8DAAAAYGpCJQAAAACmJlQCAAAAYGpCJQAAAACmJlQCAAAAYGpCJQAAAACmJlQCAOAOqurXq+rDVfWhqnp1Vd1t7JoAgPkjVAIA4ICq2pTk15IsdPeDkhyV5PxxqwLmwfHHH5+qOnA7/vjjxy4JGJlQCQCAg21Icveq2pDk2CR7Rq4HGNnxxx+fL3/5y9m6dWuuv/76bN26NV/+8pcFS7DOzSxUqqoTquqSqvpoVV1XVQ+vqntX1RVV9fHh/l7DsVVVf1hV11fVB6rqjFnVCQCwnnX3p5P8fpIbk+xN8sXufuu4VQFj2x8offKTn8z3fu/35pOf/OSBYAlYvzbM8Gu9KMlfdfd5VXVMJp96PS/Jld19UVVtT7I9yXOSnJPk1OH2w0leMtyPZuv2y+70mF0XPX4GlQAArJ7hQ75zk5yc5AtJXl9VP9fdf7bomG1JtiXJli1bRqkTmL23ve1t37J/yimnjFQNMA9mMlKpqu6Z5JFJXpYk3X1bd38hkxOWi4fDLk7ypGH73CSv7In3JDmhqk6cRa0AAOvco5N8srv3dfe/JnljkkcsPqC7d3T3QncvbNy4cZQigdl79KMffdh9YP2Z1fS34cRcvwAAIABJREFUByTZl+RPqur9VfXSqjouyf26e2+SDPf3HY7flOSmRc/fPbQBALC6bkzysKo6tqoqydlJrhu5JmBkxx13XHbt2pWTTz45n/jEJ3LyySdn165dOe6448YuDRjRrEKlDUnOSPKS7n5Iki9nMtXtUGqJtl7ywKptVbWzqnbu27fv268UAGAd6+6rklyS5JokH8zkfHHHqEUBo7v11lsPBEunnHLKgUDp1ltvHbs0YESzCpV2J9k9nKQkkxOVM5J8Zv+0tuH+5kXHb170/JNyiKuOGH4NALCyuvu3uvv7uvtB3f2U7v7a2DUB47v11lvT3QduAiVgJqFSd/9Tkpuq6rSh6ewkH0lyaZILh7YLk7x52L40yc8PV4F7WCZXHdk7i1oBAAAAuHOzvPrbryZ51XDltxuSPDWTUOt1VfW0TObvP3k49vIkj0tyfZKvDMcCAAAAMCdmFip197VJFpZ46Owlju0kT1/1ogAAAABYllmtqQQAAADAGiJUAgAAAGBqQiUAAAAApiZUAgAAAGBqQiUAAAAApiZUAgAAAGBqQiUAAAAApiZUAgAAAGBqQiUAAAAApiZUAgAAAGBqQiUAAAAApiZUAgAAAGBqQiUAAAAApiZUAgAAAGBqQiUAAAAApiZUAgAAAGBqQiUAAAAApiZUAgDggKo6raquXXS7paqeOXZdAMD82TB2AQAAzI/u/liSBydJVR2V5NNJ3jRqUQDAXDJSCQCAQzk7ySe6+1NjFwIAzB+hEgAAh3J+klePXQQAMJ+ESgAAfIuqOibJE5O8fonHtlXVzqrauW/fvtkXBwDMBaESAABLOSfJNd39mYMf6O4d3b3Q3QsbN24coTQAYB4IlQAAWMoFMfUNADgMoRIAAHdQVccmeUySN45dCwAwvzaMXQAAAPOlu7+S5LvHrgMAmG9GKgEAAAAwNaESAAAAAFMTKgEAAAAwNaESAAAAAFMTKgEAAAAwNaESAAAAAFObaahUVbuq6oNVdW1V7Rza7l1VV1TVx4f7ew3tVVV/WFXXV9UHquqMWdYKAAAAwKGNMVLpx7v7wd29MOxvT3Jld5+a5MphP0nOSXLqcNuW5CUzrxQAAACAJc3D9Ldzk1w8bF+c5EmL2l/ZE+9JckJVnThGgQAAAADc0axDpU7y1qq6uqq2DW336+69STLc33do35TkpkXP3T20AQAAADCyDTP+emd1956qum+SK6rqo4c5tpZo6285aBJObUuSLVu2rEyVAAAAABzWTEcqdfee4f7mJG9KcmaSz+yf1jbc3zwcvjvJ5kVPPynJniVec0d3L3T3wsaNG1ezfAAAAAAGMwuVquq4qrrH/u0kP5HkQ0kuTXLhcNiFSd48bF+a5OeHq8A9LMkX90+TAwAAAGBcs5z+dr8kb6qq/V/3z7v7r6rqfUleV1VPS3JjkicPx1+e5HFJrk/ylSRPnWGtAAAAABzGzEKl7r4hyQ8u0f7PSc5eor2TPH0GpQEAAAAwpVlf/Q0AAACANUCoBAAAAMDUhEoAAAAATE2oBAAAAMDUhEoAAAAATE2oBADAHVTVCVV1SVV9tKquq6qHj10TADB/NoxdAAAAc+dFSf6qu8+rqmOSHDt2QQDA/BEqAQBwQFXdM8kjk/xCknT3bUluG7MmAGA+mf4GAMBiD0iyL8mfVNX7q+qlVXXc2EUBAPNHqAQAwGIbkpyR5CXd/ZAkX06yffEBVbWtqnZW1c59+/aNUSMAMAeESgAALLY7ye7uvmrYvySTkOmA7t7R3QvdvbBx48aZFwgAzAehEgAAB3T3PyW5qapOG5rOTvKREUsCAOaUhboBADjYryZ51XDltxuSPHXkegCAOSRUAgDgDrr72iQLY9cBAMw3098AAAAAmJpQCQAAAICpCZUAAAAAmJpQCQAAAICpCZUAAAAAmJpQCQAAAICpCZUAAAAAmJpQCQAAAICpCZUAAAAAmJpQCQAAAICpCZUAAAAAmJpQCQAAAICpCZUAAAAAmJpQCQAAAICpCZUAAAAAmJpQCQAAAICpCZUAAAAAmNpMQ6WqOqqq3l9Vbxn2T66qq6rq41X12qo6Zmj/zmH/+uHxrbOsEwAAAIDDm/VIpWckuW7R/u8leUF3n5rk80meNrQ/Lcnnu/uUJC8YjgMAAABgTswsVKqqk5I8PslLh/1K8qgklwyHXJzkScP2ucN+hsfPHo4HAAAAYA5smOHXemGS30hyj2H/u5N8obtvH/Z3J9k0bG9KclOSdPftVfXF4fjPzq5cAID1qap2JflSkq8nub27F8atCACYRzMZqVRVT0hyc3dfvbh5iUP7Ljx28Gtvq6qdVbVz375932alAAAMfry7HyxQAvbbsmVLqurAbcuWLWOXBIxsVtPfzkryxOFTr9dkMu3thUlOqKr9o6VOSrJn2N6dZHOSDI9/V5LPLfXC3b2juxe6e2Hjxo2r1wMAAIB1asuWLbnpppvyiEc8Inv27MkjHvGI3HTTTYIlWOdmEip193O7+6Tu3prk/CR/090/m+TtSc4bDrswyZuH7UuH/QyP/013LzlSCQCAFddJ3lpVV1fVtrGLAca3P1B617velRNPPDHvete7DgRLwPo166u/Hew5SZ5VVddnsmbSy4b2lyX57qH9WUm2j1QfAMB6dFZ3n5HknCRPr6pHLn7Q8gOwPl1yySWH3QfWn5mHSt39ju5+wrB9Q3ef2d2ndPeTu/trQ/u/DPunDI/fMOs6AQDWq+7eM9zfnORNSc486HHLD8A6dN555x12H1h/xh6pBADAHKmq46rqHvu3k/xEkg+NWxUwts2bN+fd7353zjrrrOzduzdnnXVW3v3ud2fz5s1jlwaMaMOdHwIAwDpyvyRvqqpkcq745939V+OWBIztxhtvzJYtW/Lud7873/M935NkEjTdeOONI1cGjEmoBADAAcOyAz84dh3A/BEgAQdb1vS3qnrQShcCAAAAwJFjuWsq/deqem9V/XJVnbCiFQEAAAAw95YVKnX3jyT52SSbk+ysqj+vqsesaGUAAAAAzK1lX/2tuz+e5DeTPCfJv03yh1X10ar6dytVHAAAAADzablrKv1AVb0gyXVJHpXkf+ruBw7bL1jB+gAAAACYQ8u9+tsfJfnjJM/r7q/ub+zuPVX1mytSGQAAAABza7mh0uOSfLW7v54kVfUdSe7W3V/p7j9dseoAAAAAmEvLXVPpbUnuvmj/2KENAAAAgHVguaHS3br71v07w/axK1MSAAAAAPNuuaHSl6vqjP07VfXQJF89zPEAAAAArCHLXVPpmUleX1V7hv0Tk/zMypQEAAAAwLxbVqjU3e+rqu9LclqSSvLR7v7XFa0MAAAAgLm13JFKSfJDSbYOr/GQqkp3v3JFqgIAAABgri0rVKqqP03yvUmuTfL1obmTCJUAAAAA1oHljlRaSHJ6d/dKFgMAAADAkWG5V3/7UJJ/s5KFAAAAAHDkWO5Ipfsk+UhVvTfJ1/Y3dvcTV6QqAAAAAObackOl317JIgAAAAA4siwrVOruv62q+yc5tbvfVlXHJjlqZUsDAAAAYF4ta02lqvqlJJck+W9D06Ykf7FSRQEAAAAw35a7UPfTk5yV5JYk6e6PJ7nvShUFAMB4quqoqnp/Vb1l7FoAgPm13FDpa9192/6dqtqQpFemJAAARvaMJNeNXQQwX6rqW27A+rbcUOlvq+p5Se5eVY9J8vok/9/KlQUAwBiq6qQkj0/y0rFrAebH4gDpkksuWbIdWH+We/W37UmeluSDSf5DksvjxAMAYC14YZLfSHKPsQsB5k93H7gXKAHLGqnU3d/o7j/u7id393nDtulvAABHsKp6QpKbu/vqOzluW1XtrKqd+/btm1F1wNgWj1Baah9Yf5Y1UqmqPpkl1lDq7gd82xUBADCWs5I8saoel+RuSe5ZVX/W3T+3+KDu3pFkR5IsLCz4YBHWifPOOy+LxxKcd955I1YDzIPlrqm0kOSHhtuPJvnDJH+2UkUBADB73f3c7j6pu7cmOT/J3xwcKAHrW1XlDW94g6lvQJLlT3/750W3T3f3C5M8aoVrAwAAYA4caoSSVVBgfVvu9LczFu1+RyYjlw67mGNV3S3J3yX5zuHrXtLdv1VVJyd5TZJ7J7kmyVO6+7aq+s4kr0zy0CT/nORnunvXcuoFAGA63f2OJO8YuQxgjgiQgIMt9+pvf7Bo+/Yku5L89J0852tJHtXdt1bV0UneWVV/meRZSV7Q3a+pqv+ayVXlXjLcf767T6mq85P8XpKfWWa9AAAAAKygZYVK3f3jy3hOJ7l12D16uHUm0+b+56H94iS/nUmodO6wnSSXJPmjqipXmQMAAAAY33Knvz3rcI939/MP8byjklyd5JQkL07yiSRf6O7bh0N2J9k0bG9KctPwerdX1ReTfHeSzy6nZgAAAABWzrdz9bf/mEnwsynJ/5rk9EzWVTrk2krd/fXufnCSk5KcmeSBSx023C91OYFvGaVUVduqamdV7dy3b99UnQAAAABgeZa7ptJ9kpzR3V9Kkqr67SSv7+5fvCtP7u4vVNU7kjwsyQlVtWEYrXRSkj3DYbuTbE6yu6o2JPmuJJ9b4rV2JNmRJAsLC6bGAQAAAMzAckcqbUly26L925JsPdwTqmpjVZ0wbN89yaOTXJfk7Un2X5PywiRvHrYvHfYzPP431lMCAAAAmA/LHan0p0neW1VvymRK2k8leeWdPOfEJBcP6yp9R5LXdfdbquojSV5TVf85yfuTvGw4/mVJ/rSqrs9khNL5y6wVAAAAgBW23Ku//W5V/WWSHx2antrd77+T53wgyUOWaL8hk/WVDm7/lyRPXk59AAAAAKyu5U5/S5Jjk9zS3S/KZN2jk1eoJgAAAADm3LJCpar6rSTPSfLcoenoJH+2UkUBAAAAMN+WO1Lpp5I8McmXk6S79yS5x0oVBQAAAMB8W26odNtwJbZOkqo6buVKAgAAAGDeLTdUel1V/bckJ1TVLyV5W5I/XrmyAAAAAJhny7362+9X1WOS3JLktCT/qbuvWNHKAAAAAJhbU4dKVXVUkr/u7kcnESQBAAAArENTT3/r7q8n+UpVfdcq1AMAAADAEWBZ09+S/EuSD1bVFRmuAJck3f1rK1IVAAAAAHNtuaHSZcMNAAAAgHVoqlCpqrZ0943dffFqFQQAAADA/Jt2TaW/2L9RVW9Y4VoAAAAAOEJMGyrVou0HrGQhAACMr6ruVlXvrap/qKoPV9XvjF0TADCfpl1TqQ+xDQDA2vC1JI/q7lur6ugk76yqv+zu94xdGDCuqvqWtm5/FsJ6Nu1IpR+sqluq6ktJfmDYvqWqvlRVt6xGgQAAzE5P3DrsHj3c/NUI69ziQOktb3nLku3A+jPVSKXuPmq1CgEAYD5U1VFJrk5ySpIXd/dVI5cEzIn9I5O6W6AETD1SCQCANa67v97dD05yUpIzq+pBix+vqm1VtbOqdu7bt2+cIoGZWzxCaal9YP0RKgEAsKTu/kKSdyR57EHtO7p7obsXNm7cOEptwOw94QlPOOw+sP4IlQAAOKCqNlbVCcP23ZM8OslHx60KmBdVlcsuu8zUNyDJ9Fd/AwBgbTsxycXDukrfkeR13W2OC6xzi9dQWjxCydXfYH0TKgEAcEB3fyDJQ8auA5g/AiTgYKa/AQAAADA1oRIAAAAAUxMqAQAAADA1oRIAAAAAUxMqAQAAADA1oRIAAAAAUxMqAQAAADA1oRIAAAAAUxMqAQAAADA1oRIAAAAAU9swiy9SVZuTvDLJv0nyjSQ7uvtFVXXvJK9NsjXJriQ/3d2fr6pK8qIkj0vylSS/0N3XzKLW9Wjr9svu9JhdFz1+BpUAAAAAR4pZjVS6Pcmzu/uBSR6W5OlVdXqS7Umu7O5Tk1w57CfJOUlOHW7bkrxkRnUCAAAAcBfMJFTq7r37Rxp195eSXJdkU5Jzk1w8HHZxkicN2+cmeWVPvCfJCVV14ixqBQAAAODOzXxNparamuQhSa5Kcr/u3ptMgqck9x0O25TkpkVP2z20AQAAADAHZhoqVdXxSd6Q5JndfcvhDl2irQ/xmtuqamdV7dy3b99KlAkAAADAnZhZqFRVR2cSKL2qu984NH9m/7S24f7moX13ks2Lnn5Skj1LvW537+juhe5e2Lhx4+oUDwAAAMAdzCRUGq7m9rIk13X38xc9dGmSC4ftC5O8eVH7z9fEw5J8cf80OQAAAADGt2FGX+esJE9J8sGqunZoe16Si5K8rqqeluTGJE8eHrs8yeOSXJ/kK0meOqM6AQAAALgLZhIqdfc7s/Q6SUly9hLHd5Knr2pRAAAAACzbzK/+BgAAAMCRT6gEAMABVbW5qt5eVddV1Yer6hlj1wQAzKdZrakEAMCR4fYkz+7ua6rqHkmurqoruvsjYxcGAMwXI5UAADigu/d29zXD9peSXJdk07hVAQDzSKgEAMCSqmprkockuWrcSgCAeWT6GwAA36Kqjk/yhiTP7O5bDnpsW5JtSbJly5YRqgOmVXWoi3GPb3Lxb+BIZKQSAAB3UFVHZxIovaq733jw4929o7sXunth48aNsy8QmFp3r9jt/s95y4q+HnDkEioBAHBATYYzvCzJdd39/LHrAQDml1AJAIDFzkrylCSPqqprh9vjxi4KAJg/1lQCAOCA7n5nkvldfAUAmBtGKgEAAAAwNaESAAAAAFMTKgEAAAAwNaESAAAAAFMTKgEAAAAwNaESAAAAAFMTKgEAAAAwNaESAAAAAFMTKgEAAAAwNaESAAAAAFMTKgEAAAAwNaESAAAAAFMTKgEAAAAwNaESAAAAAFMTKgEAAAAwNaESAAAAAFMTKgEAAAAwNaESAAAAAFMTKgEAAAAwNaESAAAAAFMTKgEAAAAwtZmFSlX18qq6uao+tKjt3lV1RVV9fLi/19BeVfWHVXV9VX2gqs6YVZ0AAOvZUudsAABLmeVIpVckeexBbduTXNndpya5cthPknOSnDrctiV5yYxqBABY716Rbz1nAwD4FjMLlbr775J87qDmc5NcPGxfnORJi9pf2RPvSXJCVZ04m0oBANavQ5yzAQB8i7HXVLpfd+9NkuH+vkP7piQ3LTpu99AGAAAAwBzYMHYBh1BLtPWSB1Zty2SKXLZs2bKaNQEAEOdfMCs/+DtvzRe/+q9jl7GkrdsvG7uEb/Fddz86//BbPzF2GbCujB0qfaaqTuzuvcP0tpuH9t1JNi867qQke5Z6ge7ekWRHkiwsLCwZPAEAsHKcf8FsfPGr/5pdFz1+7DKOGPMYdMFaN/b0t0uTXDhsX5jkzYvaf364CtzDknxx/zQ5AAAAAMY3s1Cpql6d5O+TnFZVu6vqaUkuSvKYqvp4kscM+0lyeZIbklyf5I+T/PKs6gQAWM8Occ4GAPAtZjb9rbsvOMRDZy9xbCd5+upWBADAwQ5zzgYAcAdjT38DAAAA4AgkVAIAAABgakIlAAAAAKYmVAIAAABgakIlAAAAAKYmVAIAAABgakIlAAAAAKYmVAIAAABgakIlAAAAAKYmVAIAAABgakIlAAAAAKYmVAIAAABgakIlAAAAAKYmVAIAAABgakIlAAAAAKYmVAIAAABgakIlAAAAAKYmVAIAAABgakIlAAAAAKYmVAIAAABgakIlAADuoKoeW1Ufq6rrq2r72PUAAPNJqAQAwAFVdVSSFyc5J8npSS6oqtPHrQoAmEdCJQAAFjszyfXdfUN335bkNUnOHbkmAGAOCZUAAFhsU5KbFu3vHtoAAO5gw9gFwGrauv2yu3Tcrosev8qVAMARo5Zo6zscULUtybYk2bJlyyxqgnXpHg/cnu+/2LJmd9U9HpgkzuthloRKAAAstjvJ5kX7JyXZs/iA7t6RZEeSLCws3CFwAlbOBy/84NglAByW6W8AACz2viSnVtXJVXVMkvOTXDpyTQDAHDJSCQCAA7r79qr6lSR/neSoJC/v7g+PXBYAMIeESsAR666smWW9LIDpdfflSS4fuw4AYL4JleAIJlQBAABgLNZUAgAAAGBqcx0qVdVjq+pjVXV9VbmWJgAAAMCcmNtQqaqOSvLiJOckOT3JBVV1+rhVAQAAAJDM95pKZya5vrtvSJKqek2Sc5N8ZNSqAGbEmlnMIz+XAADsN7cjlZJsSnLTov3dQxsAAAAAI6vuHruGJVXVk5P8ZHf/4rD/lCRndvevHnTctiTbht3TknxslUq6T5LPrtJrz4v10MdEP9eS9dDHRD/XkvXQx2R1+3n/7t64Sq/NMlTVviSfGrsOYKbWy/9nwDcteQ42z9PfdifZvGj/pCR7Dj6ou3ck2bHaxVTVzu5eWO2vM6b10MdEP9eS9dDHRD/XkvXQx2T99JMJIR+sP97ngf3mefrb+5KcWlUnV9UxSc5PcunINQEAAACQOR6p1N23V9WvJPnrJEcleXl3f3jksgAAAADIHIdKSdLdlye5fOw6Bqs+xW4OrIc+Jvq5lqyHPib6uZashz4m66efAOuV93kgyRwv1A0AAADA/JrnNZUAAAAAmFNCpTtRVY+tqo9V1fVVtX3selZDVb28qm6uqg+NXctqqarNVfX2qrquqj5cVc8Yu6bVUFV3q6r3VtU/DP38nbFrWk1VdVRVvb+q3jJ2LauhqnZV1Qer6tqq2jl2Paulqk6oqkuq6qPD7+jDx65ppVXVacP3cf/tlqp65th1rbSq+vXhvedDVfXqqrrb2DUBcOeq6v8Y3r8/MPw/9cNV9dKqOn14/NZDPO9hVXXV8Jzrquq3Z1o4MDrT3w6jqo5K8o9JHpNkdyZXpLuguz8yamErrKoemeTWJK/s7geNXc9qqKoTk5zY3ddU1T2SXJ3kSWvwe1lJjuvuW6vq6CTvTPKM7n7PyKWtiqp6VpKFJPfs7ieMXc9Kq6pdSRa6+7Nj17KaquriJP+9u186XO3z2O7+wth1rZbh/5ZPJ/nh7v7U2PWslKralMl7zund/dWqel2Sy7v7FeNWBsDhDB/mPD/Jj3X316rqPkmO6e49i465tbuPX+K5H0vy0939D8P/b6ettfNr4PCMVDq8M5Nc3903dPdtSV6T5NyRa1px3f13ST43dh2rqbv3dvc1w/aXklyXZNO4Va28ntj/SdLRw21NJsdVdVKSxyd56di1sHxVdc8kj0zysiTp7tvWcqA0ODvJJ9ZSoLTIhiR3r6oNSY5NsudOjgdgfCcm+Wx3fy1Juvuz3b2nqt5RVQv7D6qqP6iqa6rqyqraODTfN8ne4Xlf3x8oVdVvV9WfVtXfVNXHq+qXZtwnYEaESoe3KclNi/Z3Zw0GEetNVW1N8pAkV41byeoYpoRdm+TmJFd095rsZ5IXJvmNJN8Yu5BV1EneWlVXV9W2sYtZJQ9Isi/JnwxTGV9aVceNXdQqOz/Jq8cuYqV196eT/H6SGzP5A+OL3f3WcasC4C54a5LNVfWPVfVfqurfLnHMcUmu6e4zkvxtkt8a2l+Q5GNV9aaq+g8HTXv+gUw+AHx4kv9UVd+zin0ARiJUOrxaom1NjvpYL6rq+CRvSPLM7r5l7HpWw/Ap0YOTnJTkzKpac1Maq+oJSW7u7qvHrmWVnTWcvJ2T5OnDVNW1ZkOSM5K8pLsfkuTLSdbk+nVJMkzve2KS149dy0qrqntlMpr35CTfk+S4qvq5casC4M4Mo9wfmmRbJh/0vLaqfuGgw76R5LX/P3v3H25XXd+J/v2B4G8tVaJFkjR2zFW88kMmpZTY3gKKgC049+pU2qmUYSZjpY59pndGrPOM7dzxau/TW1s7LQ5jtFgVS7GUXGUUqtBeYRSDIqix15ShJA1KWgV/oFLo5/5xVtJDPCTs48leO8nr9Tz72Wt913ft/Tn7gZN13vv7/a5h+91Jnj+c+x8ztxTBNUl+JsmH5p1zVXd/a5jGf13mZoEABxih0p5tS7Jy3v6KGMq/3xrWGHp/kvd09x+PXc++Nkwhuj7JGSOXsi+sS3L2sObQ+5KcWlXvHrekpbdzLYPuvjvJlTkwL8a2Jdk2b0TdFZkLmQ5UZ2bum94vj13IPvCCJP+ju3d0998l+eMkJ49cEwCPwPCl5PXd/YYkv5jkf9vbKfPO/cvuvjhz07uPq6qn7N7nYfaBA4BQac8+mWRNVT1j+Hb55Uk2jlwTizAsYL0hyebu/s2x69lXqmp5VR0+bD82c3/kfWHcqpZed7+uu1d09+rM/X/50e4+oEZEVNXjh0XlM0wHOz3JAXeHxu7+UpKtVfWsoem0JAfyAp/n5gCc+ja4M8lJVfW44XfuaZlbvw6AGTbcoXTNvKbjk+y+7t8hSV46bP9M5m7MkKp68fA7P0nWJHkwyc61Ec8Z7kz8lCQ/kbm/rYADzLKxC5hl3f1AVf1ikg8nOTTJO7r7cyOXteSq6rLM/aI/oqq2JXlDd28Yt6olty7JzyW5bVhvKEl+pbuvHrGmfeHIJJcOd984JMnl3f2BkWticZ6W5MrhOm1Zkvd294f2fMp+69VJ3jOE97cnOX/kevaJqnpc5u4m+q/GrmVf6O5PVNUVST6V5IEkn05yybhVAfAIPCHJ7wxfTD6QZEvmpsJdMa/PN5P8z1V1c5J7k/z00P5zSd5SVfcN5/5sdz84XL/clOSDSVYl+T/m300OOHBUt1GIAAAALI2q+tUk3+ju3xi7FmDfMv0NAAAAgIkZqQQAAADAxIxUAgAAAGBiQiUAAAAAJiZUAgAAAGBiQiUAAAAAJiZUAgAAAGBiQiUAAAAAJiZUAgAAAGBiQiUAAAAAJiZUAgAAAGBiQiUAAAAAJiZUAgAAAGBiQiUAAAAAJiZUAgAAAGBiQiUAAAAAJiZUAgAAAGBiQiUAAAAAJiZUAgAAAGBiQiUAAAAAJiZUAg5oVXV9Vf2LsesAADiYuAaDg4NQCRhNVd1RVd+qqm9U1Zer6p1V9YSx6wIAOJDVsog4AAAgAElEQVS5BgOWilAJGNtPdfcTkpyQ5IeT/PtJTq6qZfukKgCAA5trMOB7JlQCZkJ3/3WS/5bkuVV1flVtrqqvV9XtVfWvdvarqp+oqm1V9dqq+lKSdw7t51TVLVX1tar6y6o6Y97L/2BV3TC83jVVdcR0fzoAgNnkGgz4XgiVgJlQVSuTnJXk00nuTvKTSZ6U5Pwkb6mqE+Z1/4EkT07yg0nWV9WJSd6V5N8mOTzJjye5Y17/nxle56lJHpXkf9+XPwsAwP7CNRjwvTBkERjbn1TVA0nuTfLBJP9nd39r3vE/q6prkvxYkk8NbX+f5A3d/Z0kqaoLkryju68djv/1bu/xzu7+/4a+lyc5e9/8KAAA+w3XYMD3TKgEjO0l3f2n8xuq6swkb0jyP2VuROXjktw2r8uO7v72vP2VSa7ew3t8ad72fUksRAkAHOxcgwHfM9PfgJlSVY9O8v4kv5Hkad19eOYuVmpet97ttK1J/tF0KgQAOPC4BgMWQ6gEzJpHJXl0kh1JHhi+MTt9L+dsSHJ+VZ1WVYdU1VFV9ex9XSgAwAHENRgwMaESMFO6++tJ/nWSy5N8NXMLPG7cyzk3ZVhMMnPrAvxZ5haQBADgEXANBixGde8+ghEAAAAA9sxIJQAAAAAmJlQCAAAAYGJCJQAAAAAmJlQCAAAAYGJCJQAAAAAmtmxab1RVz0ryh/OafijJf0jyrqF9dZI7kvzT7v5qVVWS305yVpL7kvx8d39qT+9xxBFH9OrVq5e8dgBgNtx8881/093Lx66Df+D6CwAOfA93DTa1UKm7/yLJ8UlSVYcm+eskVya5KMlHuvvNVXXRsP/aJGcmWTM8fiTJxcPzw1q9enU2bdq0z34GAGBcVfVXY9fAQ7n+AoAD38Ndg401/e20JH/Z3X+V5Jwklw7tlyZ5ybB9TpJ39ZyPJzm8qo6cfqkAAAAA7G6sUOnlSS4btp/W3XclyfD81KH9qCRb552zbWgDAAAAYGRTD5Wq6lFJzk7yR3vrukBbL/B666tqU1Vt2rFjx1KUCAAAAMBejDFS6cwkn+ruLw/7X945rW14vnto35Zk5bzzViTZvvuLdfcl3b22u9cuX27dTgAAAIBpGCNUOjf/MPUtSTYmOW/YPi/JVfPaX1FzTkpy785pcgAAAACMa6qhUlU9LskLk/zxvOY3J3lhVX1xOPbmof3qJLcn2ZLkvyZ51RRLBQA4aFXV4VV1RVV9oao2V9WPjl0TADB7phoqdfd93f2U7r53Xtvfdvdp3b1meP7K0N7dfWF3/6PuPqa73asWAGA6fjvJh7r72UmOS7J55HqAGXDZZZfluc99bg499NA897nPzWWXXbb3k4AD2rKxCwAAYHZU1ZOS/HiSn0+S7r4/yf1j1gSM77LLLsvrX//6bNiwIc9//vPzsY99LBdccEGS5Nxzzx25OmAsY6ypBADA7PqhJDuSvLOqPl1Vb6+qx49dFDCuN77xjdmwYUNOOeWUHHbYYTnllFOyYcOGvPGNbxy7NGBEQiUAAOZbluSEJBd39/OSfDPJRfM7VNX6qtpUVZt27NgxRo3AlG3evDnPf/7zH9L2/Oc/P5s3mx0LBzPT3x6h1Rd9cK997njzi6dQCQDAPrUtybbu/sSwf0V2C5W6+5IklyTJ2rVre7rlAWM4+uij87GPfSynnHLKrraPfexjOfroo0esChibkUoAAOzS3V9KsrWqnjU0nZbk8yOWBMyA17/+9bngggty3XXX5e/+7u9y3XXX5YILLsjrX//6sUsDRmSkEgAAu3t1kvdU1aOS3J7k/JHrAUa2czHuV7/61dm8eXOOPvrovPGNb7RINxzkhEoAADxEd9+SZO3YdQCz5dxzzxUiAQ9h+hsAAAAAExMqAQAAADAxoRIAAAAAExMqAQAAADAxoRIAAAAAExMqAQAAADAxoRIAAAAAExMqAQAAADAxoRIAAAAAExMqAQAAADAxoRIAAAAAExMqAQAAADAxoRIAAAAAExMqAQAAADAxoRIAAAAAExMqAQAAADAxoRIAAAAAExMqAQAAADAxoRIAAAAAExMqAQAAADCxqYVKVXV4VV1RVV+oqs1V9aNV9eSquraqvjg8f//Qt6rqrVW1papuraoTplUnAAAAAHs3zZFKv53kQ9397CTHJdmc5KIkH+nuNUk+MuwnyZlJ1gyP9UkunmKdAAAAAOzFVEKlqnpSkh9PsiFJuvv+7r4nyTlJLh26XZrkJcP2OUne1XM+nuTwqjpyGrUCAAAAsHfTGqn0Q0l2JHlnVX26qt5eVY9P8rTuvitJhuenDv2PSrJ13vnbhjYAAAAAZsC0QqVlSU5IcnF3Py/JN/MPU90WUgu09YIdq9ZX1aaq2rRjx47vvVIAAAAA9mpaodK2JNu6+xPD/hWZC5m+vHNa2/B897z+K+edvyLJ9oVeuLsv6e613b12+fLl+6R4AAAAAB5qKqFSd38pydaqetbQdFqSzyfZmOS8oe28JFcN2xuTvGK4C9xJSe7dOU0OAAAAgPEtm+J7vTrJe6rqUUluT3J+5kKty6vqgiR3JnnZ0PfqJGcl2ZLkvqEvAAAAADNiaqFSd9+SZO0Ch05boG8nuXCfFwUAAADAokxrTSUAAAAADiDTnP4GAMB+oKruSPL1JA8meaC7FxptDgAc5IxUAgBgIad09/ECJWCnVatWpap2PVatWjV2ScDIhEoAAADs0apVq7J169acfPLJ2b59e04++eRs3bpVsAQHOaESAAC76yTXVNXNVbV+7GKA8e0MlG644YYceeSRueGGG3YFS8DBS6gEAMDu1nX3CUnOTHJhVf34/INVtb6qNlXVph07doxTITB1V1xxxR73gYOPUAkAgIfo7u3D891Jrkxy4m7HL+nutd29dvny5WOUCIzgpS996R73gYOPUAkAgF2q6vFV9cSd20lOT/LZcasCxrZy5crceOONWbduXe66666sW7cuN954Y1auXDl2acCIlo1dAAAAM+VpSa6sqmTuWvG93f2hcUsCxnbnnXdm1apVufHGG/P0pz89yVzQdOedd45cGTAmoRIAALt09+1Jjhu7DmD2CJCA3Zn+BgAAAMDEhEoAAAAATEyoBAAAAMDEhEoAAAAATEyoBAAAAMDEhEoAAAAATEyoBAAAAMDEhEoAAAAATEyoBAAAAMDEhEoAAAAATEyoBAAAAMDEhEoAAAAATEyoBAAAAMDEhEoAAAAATEyoBAAAAMDEhEoAAAAATEyoBAAAAMDEphoqVdUdVXVbVd1SVZuGtidX1bVV9cXh+fuH9qqqt1bVlqq6tapOmGatAAAAADy8MUYqndLdx3f32mH/oiQf6e41ST4y7CfJmUnWDI/1SS6eeqUAAAAALGgWpr+dk+TSYfvSJC+Z1/6unvPxJIdX1ZFjFAgAAADAQ007VOok11TVzVW1fmh7WnfflSTD81OH9qOSbJ137rahDQAAAICRLZvy+63r7u1V9dQk11bVF/bQtxZo6+/qNBdOrU+SVatWLU2VAAAAAOzRVEcqdff24fnuJFcmOTHJl3dOaxue7x66b0uyct7pK5JsX+A1L+nutd29dvny5fuyfAAAAAAGUwuVqurxVfXEndtJTk/y2SQbk5w3dDsvyVXD9sYkrxjuAndSknt3TpMDAAAAYFzTnP72tCRXVtXO931vd3+oqj6Z5PKquiDJnUleNvS/OslZSbYkuS/J+VOsFQAAAIA9mFqo1N23Jzlugfa/TXLaAu2d5MIplAYAAADAhKZ99zcAAAAADgBCJQAAAAAmJlQCAAAAYGJCJQAAAAAmJlQCAOAhqurQqvp0VX1g7FoAgNklVAIAYHevSbJ57CKA2XLsscemqnY9jj322LFLAkYmVAIAYJeqWpHkxUnePnYtwOw49thjc9ttt+Xss8/Ojh07cvbZZ+e2224TLMFBTqgEAMB8v5Xk3yX5+7ELAWbHzkDpqquuyhFHHJGrrrpqV7AEHLyESgAAJEmq6ieT3N3dN++l3/qq2lRVm3bs2DGl6oCxbdiwYY/7wMFHqAQAwE7rkpxdVXckeV+SU6vq3bt36u5Lunttd69dvnz5tGsERnLBBRfscR84+AiVAABIknT367p7RXevTvLyJB/t7n82clnADDjmmGOycePGnHPOOfmbv/mbnHPOOdm4cWOOOeaYsUsDRrRs7AIAAACYbbfeemuOPfbYbNy4MTtHKB5zzDG59dZbR64MGJNQCQCA79Ld1ye5fuQygBkiQAJ2Z/obAAAAABMTKgEAAAAwMaESAAAAABMTKgEAAAAwMaESAAAAABMTKgEAAAAwMaESAAAAABMTKgEAAAAwMaESAAAAABNbVKhUVc9d6kIAAAAA2H8sdqTS26rqpqp6VVUdvqQVAQAAADDzFhUqdffzk/xskpVJNlXVe6vqhUtaGQAAAAAza9FrKnX3F5P8+ySvTfK/JHlrVX2hqv7XpSoOAAAAgNm02DWVjq2qtyTZnOTUJD/V3UcP229ZwvoAAAAAmEGLHan0n5N8Kslx3X1hd38qSbp7e+ZGLy2oqg6tqk9X1QeG/WdU1Seq6otV9YdV9aih/dHD/pbh+OpF1gkAAADAPrDYUOmsJO/t7m8lSVUdUlWPS5Lu/oM9nPeazI1u2unXk7ylu9ck+WqSC4b2C5J8tbufmbmRT7++yDoBAAAA2AcWGyr9aZLHztt/3ND2sKpqRZIXJ3n7sF+Zmy53xdDl0iQvGbbPGfYzHD9t6A8AAADADFhsqPSY7v7Gzp1h+3F7Oee3kvy7JH8/7D8lyT3d/cCwvy3JUcP2UUm2Dq/9QJJ7h/4AAAAAzIDFhkrfrKoTdu5U1T9O8q2H61xVP5nk7u6+eX7zAl37ERzb/bXXV9Wmqtq0Y8eOvVcOAAAAwPds2SLP+6Ukf1RV24f9I5P89B76r0tydlWdleQxSZ6UuZFLh1fVsmE00ookO19vW5KVSbZV1bIk35fkKwu9cHdfkuSSJFm7du2CwRMAAAAAS2tRI5W6+5NJnp3kF5K8KsnRu41C2r3/67p7RXevTvLyJB/t7p9Ncl2Slw7dzkty1bC9cdjPcPyj3S0wAgAAAJgRix2plCQ/nGT18BrPq6p097smfI3XJnlfVf2nJJ9OsmFo35DkD6pqS+ZGKL38e6gTAAAAgCW2qFCpqv4gyT9KckuSB4fmTrLXUKm7r09y/bB9e5ITF+jz7SQvW0xtAAAAAOx7ix2ptDbJc0xJAwAAADg4Lfbub59N8gNLWQgAAAAA+4/FjlQ6Isnnq+qmJN/Z2djdZy9JVQAAAADMtMWGSr+6lEUAAAAAsH9ZVKjU3X9WVT+YZE13/2lVPS7JoUtbGgAAAACzalFrKlXVv0xyRZL/MjQdleRPlqooAAAAAGbbYhfqvjDJuiRfS5Lu/mKSpy5VUQAAAADMtsWGSt/p7vt37lTVsiS9NCUBADCWqnpMVd1UVZ+pqs9V1a+NXRMAMJsWGyr9WVX9SpLHVtULk/xRkv9n6coCAGAk30lyancfl+T4JGdU1Ukj1wTMgFWrVqWqdj1WrVo1dknAyBYbKl2UZEeS25L8qyRXJ/n3S1UUAADj6DnfGHYPGx5GpMNBbtWqVdm6dWtOPvnkbN++PSeffHK2bt0qWIKD3GLv/vb3Sf7r8AAA4ABSVYcmuTnJM5P8bnd/YuSSgJHtDJRuuOGGJMkNN9yQdevW5cYbbxy5MmBMi7372/+oqtt3fyx1cQAATF93P9jdxydZkeTEqnru/ONVtb6qNlXVph07doxTJDB1V1xxxR73gYPPYqe/rU3yw8Pjx5K8Ncm7l6ooAADG1933JLk+yRm7tV/S3Wu7e+3y5ctHqQ2Yvpe+9KV73AcOPosKlbr7b+c9/rq7fyvJqUtcGwAAU1ZVy6vq8GH7sUlekOQL41YFjG3lypW58cYbs27dutx11127pr6tXLly7NKAES1qTaWqOmHe7iGZG7n0xCWpCACAMR2Z5NJhXaVDklze3R8YuSZgZHfeeWdWrVqVG2+8MU9/+tOTzAVNd95558iVAWNaVKiU5P+et/1AkjuS/NPvuRoAAEbV3bcmed7YdQCzR4AE7G6xd387ZakLAQAAAGD/sdjpb/9mT8e7+zcXVw4AAAAA+4PFTn/befe3jcP+TyX58yRbl6IoAAAAAGbbYkOlI5Kc0N1fT5Kq+tUkf9Td/2KpCgMAAABgdh2yyPNWJbl/3v79SVZ/z9UAAAAAsF9Y7EilP0hyU1VdmaST/JMk71qyqgAAAACYaYu9+9sbq+q/Jfmxoen87v700pUFAAAAwCxb7PS3JHlckq91928n2VZVz1iimgAAAACYcYsKlarqDUlem+R1Q9NhSd69VEUBAAAAMNsWO1LpnyQ5O8k3k6S7tyd54lIVBQAAAMBsW2yodH93d+YW6U5VPX7pSgIAAABg1i02VLq8qv5LksOr6l8m+dMk/3VPJ1TVY6rqpqr6TFV9rqp+bWh/RlV9oqq+WFV/WFWPGtofPexvGY6vXmStAAAAACyxRYVK3f0bSa5I8v4kz0ryH7r7d/Zy2neSnNrdxyU5PskZVXVSkl9P8pbuXpPkq0kuGPpfkOSr3f3MJG8Z+gEAAAAwA5ZNekJVHZrkw939giTXPtLzhuly3xh2DxseneTUJD8ztF+a5FeTXJzknGE7mQuw/nNV1fA6AAAAAIxo4pFK3f1gkvuq6vsmPbeqDq2qW5LcnblA6i+T3NPdDwxdtiU5atg+KsnW4T0fSHJvkqdM+p4AAAAALL2JRyoNvp3ktqq6NsMd4JKku//1nk4aAqnjq+rwJFcmOXqhbsNz7eHYLlW1Psn6JFm1atUjKh4AAACA781iQ6UPDo9F6e57qur6JCdlbrHvZcNopBVJtg/dtiVZmWRbVS1L8n1JvrLAa12S5JIkWbt2ralxAAAAAFMwUahUVau6+87uvnTSN6qq5Un+bgiUHpvkBZlbfPu6JC9N8r4k5yW5ajhl47D/34fjH7WeEgAAAMBsmHRNpT/ZuVFV75/w3COTXFdVtyb5ZJJru/sDSV6b5N9U1ZbMrZm0Yei/IclThvZ/k+SiCd8PAAAAgH1k0ulv89c5+qFJTuzuW5M8b4H225OcuED7t5O8bML6AAAAAJiCSUcq9cNsAwAAAHAQmXSk0nFV9bXMjVh67LCdYb+7+0lLWh0AAAAAM2miUKm7D91XhQAAAACw/5h0+hsAAAAACJUAAAAAmJxQCQAAAICJCZUAAAAAmJhQCQAAAICJCZUAAAAAmJhQCQAAAICJCZUAANilqlZW1XVVtbmqPldVrxm7JgBgNi0buwAAAGbKA0l+ubs/VVVPTHJzVV3b3Z8fuzBgXFX1XW3dPUIlwKwwUgkAgF26+67u/tSw/fUkm5McNW5VwNjmB0rPfOYzF2wHDj5GKgEAsKCqWp3keUk+MW4lwKyYPzJJoAQYqQQAwHepqickeX+SX+rur+12bH1VbaqqTTt27BinQGDq5o9QWmgfOPgIlQAAeIiqOixzgdJ7uvuPdz/e3Zd099ruXrt8+fLpFwiMYsuWLXvcBw4+QiUAAHapufksG5Js7u7fHLseYLZUVdasWWPqG5BEqAQAwEOtS/JzSU6tqluGx1ljFwWMa/5aSvNHKLn7GxzcLNQNAMAu3f2xJIYgAN9FgATszkglAAAAACYmVAIAAABgYkIlAAAAACYmVAIAAABgYkIlAAAAACYmVAIAAABgYkIlAAAAACYmVAIAAABgYlMJlapqZVVdV1Wbq+pzVfWaof3JVXVtVX1xeP7+ob2q6q1VtaWqbq2qE6ZRJwAAAACPzLRGKj2Q5Je7++gkJyW5sKqek+SiJB/p7jVJPjLsJ8mZSdYMj/VJLp5SnQAAAAA8AlMJlbr7ru7+1LD99SSbkxyV5Jwklw7dLk3ykmH7nCTv6jkfT3J4VR05jVoBAAAA2Lupr6lUVauTPC/JJ5I8rbvvSuaCpyRPHbodlWTrvNO2DW0AAAAAzICphkpV9YQk70/yS939tT11XaCtH+Y111fVpqratGPHjqUoEwAAAIC9mFqoVFWHZS5Qek93//HQ/OWd09qG57uH9m1JVs47fUWS7Qu9bndf0t1ru3vt8uXL903xAAAAADzEtO7+Vkk2JNnc3b8579DGJOcN2+cluWpe+yuGu8CdlOTendPkAAAAABjfsim9z7okP5fktqq6ZWj7lSRvTnJ5VV2Q5M4kLxuOXZ3krCRbktyX5Pwp1QkAAADAIzCVUKm7P5aF10lKktMW6N9JLtynRQEAAACwaFO/+xsAAAAA+z+hEgAAAAATEyoBAAAAMDGhEgAAAAATEyoBAAAAMDGhEgAAAAATEyoBAAAAMDGhEgAAAAATEyoBAAAAMDGhEgAAAAATEyoBAAAAMDGhEgAAAAATEyoBAAAAMDGhEgAAu1TVO6rq7qr67Ni1AACzTagEAMB8v5/kjLGLAGZPVX3XAzi4CZUAANilu/88yVfGrgOYLQ8XIAmW4OC2bOwCAAAA2D90965tgRJgpBIAABOpqvVVtamqNu3YsWPscgCAkQiVAACYSHdf0t1ru3vt8uXLxy4HABiJ6W8AAAA8Iqa8AfMZqQQAwC5VdVmS/57kWVW1raouGLsmYHzz11J6JO3AwcFIJQAAdunuc8euAZhNAiRgd0YqAQAAADAxoRIAAAAAExMqAQAAADAxayqxZFZf9MFH1O+ON794H1cCAAAA7GtGKgEAAAAwsamFSlX1jqq6u6o+O6/tyVV1bVV9cXj+/qG9quqtVbWlqm6tqhOmVScAAAAAezfNkUq/n+SM3douSvKR7l6T5CPDfpKcmWTN8Fif5OIp1QgAAADAIzC1UKm7/zzJV3ZrPifJpcP2pUleMq/9XT3n40kOr6ojp1MpAAAAAHsz9ppKT+vuu5JkeH7q0H5Ukq3z+m0b2gAAAACYAWOHSg+nFmjrBTtWra+qTVW1aceOHfu4LAAAAACS8UOlL++c1jY83z20b0uycl6/FUm2L/QC3X1Jd6/t7rXLly/fp8UCAAAAMGfsUGljkvOG7fOSXDWv/RXDXeBOSnLvzmlyAAAAAIxv2bTeqKouS/ITSY6oqm1J3pDkzUkur6oLktyZ5GVD96uTnJVkS5L7kpw/rToBAAAA2LuphUrdfe7DHDptgb6d5MJ9WxEAAAAAizX29DcAAAAA9kNCJQAAAAAmNrXpb8Cc1Rd9cK997njzi6dQCQAAACyekUoAAAAATEyoBAAAAMDEhEoAAAAATEyoBAAAAMDEhEoAAAAATEyoBAAAAMDEhEoAAAAATEyoBAAAAMDEhEoAAAAATEyoBAAAAMDEhEoAAAAATEyoBAAAAMDEhEoAADxEVZ1RVX9RVVuq6qKx6wEAZpNQCQCAXarq0CS/m+TMJM9Jcm5VPWfcqgCAWSRUAgBgvhOTbOnu27v7/iTvS3LOyDUBADNIqAQAwHxHJdk6b3/b0AYA8BDLxi4AAICZUgu09UM6VK1Psj5JVq1aNY2a4KB0zKXHjF3Cfue2824buwQ4qAiVAACYb1uSlfP2VyTZPr9Dd1+S5JIkWbt27UMCJ2DpCEiAWWf6GwAA830yyZqqekZVPSrJy5NsHLkmAGAGGakEAMAu3f1AVf1ikg8nOTTJO7r7cyOXBQDMIKESAAAP0d1XJ7l67DoAgNlm+hsAAAAAExMqAQAAADAxoRIAAAAAE5vpUKmqzqiqv6iqLVV10dj1AAAAADBnZkOlqjo0ye8mOTPJc5KcW1XPGbcqAAAAAJLZvvvbiUm2dPftSVJV70tyTpLPj1oVMDNWX/TBvfa5480vnkIlHAj89wQAAJOZ5VDpqCRb5+1vS/IjI9UCsN8SlgAAAPtCdffYNSyoql6W5EXd/S+G/Z9LcmJ3v3q3fuuTrB92n5XkL/ZRSUck+Zt99NoHIp/X5Hxmk/OZTc5nNjmf2eT25Wf2g929fB+9NotQVTuS/NXYdQBT5d9GOPgseA02yyOVtiVZOW9/RZLtu3fq7kuSXLKvi6mqTd29dl+/z4HC5zU5n9nkfGaT85lNzmc2OZ/ZwUXIBwcfv+eBnWZ2oe4kn0yypqqeUVWPSvLyJBtHrgkAAACAzPBIpe5+oKp+McmHkxya5B3d/bmRywIAAAAgMxwqJUl3X53k6rHrGOzzKXYHGJ/X5Hxmk/OZTc5nNjmf2eR8ZgAHNr/ngSQzvFA3AAAAALNrltdUAgAAAGBGCZX2oqrOqKq/qKotVXXR2PXMuqp6R1XdXVWfHbuW/UVVrayq66pqc1V9rqpeM3ZNs66qHlNVN1XVZ4bP7NfGrml/UFWHVtWnq+oDY9eyv6iqO6rqtqq6pao2jV3PrKuqw6vqiqr6wvA77UfHrgmAPauqB4d/5z5bVX9UVY9bgtf8+ar6z0tRHzDbhEp7UFWHJvndJGcmeU6Sc6vqOeNWNfN+P8kZYxexn3kgyS9399FJTkpyof/O9uo7SU7t7uOSHJ/kjKo6aeSa9gevSbJ57CL2Q6d09/FunfyI/HaSD3X3s5McF/+9AewPvjX8O/fcJPcneeUjPXH4ewk4iAmV9uzEJFu6+/buvj/J+5KcM3JNM627/zzJV8auY3/S3Xd196eG7a9n7o+wo8atarb1nG8Mu4cNDwvE7UFVrUjy4iRvH7sWDkxV9aQkP55kQ5J09/3dfc+4VQEwof83yTOTpKr+pKpuHkaFr9/Zoaq+UVX/sao+keRHq+qHq+rGYQT5TVX1xKHr06vqQ1X1xar6v0b4WYApECrt2VFJts7b3xZ/7LMPVdXqJM9L8olxK5l9w1SuW5LcneTa7vaZ7dlvJfl3Sf5+7EL2M53kmuGiev1eex/cfijJjiTvHKZZvr2qHj92UQA8MlW1LHMzNG4bmv55d//jJGuT/OuqesrQ/vgkn+3uH0lyU5I/TPKaYQT5C8wUAmwAACAASURBVJJ8a+h3fJKfTnJMkp+uqpXT+UmAaRIq7Vkt0GY0BPtEVT0hyfuT/FJ3f23semZddz/Y3ccnWZHkxKp67tg1zaqq+skkd3f3zWPXsh9a190nZO4i+8Kq+vGxC5phy5KckOTi7n5ekm8msRYhwOx77PBF3aYkd2YYcZq5IOkzST6eZGWSNUP7g5m7Zk2SZyW5q7s/mSTd/bXufmA49pHuvre7v53k80l+cN//KMC0LRu7gBm3LXO/QHdakWT7SLVwAKuqwzL3j/N7uvuPx65nf9Ld91TV9Zlby8sC8Qtbl+TsqjoryWOSPKmq3t3d/2zkumZed28fnu+uqiszNy36z8etamZtS7Jt3qjBKyJUAtgffGv4om6XqvqJzI06+tHuvm+41nrMcPjb3f3gzq55+C/dvzNv+8H42xMOSEYq7dknk6ypqmdU1aOSvDzJxpFr4gBTVZW5b4Q2d/dvjl3P/qCqllfV4cP2YzN30fOFcauaXd39uu5e0d2rM/d77KMCpb2rqsfvXBdimMZ1egSXD6u7v5Rka1U9a2g6LXPfTAOw//m+JF8dAqVnZ+5mMgv5QubWTvrhJKmqJw7T6ICDhP/h96C7H6iqX0zy4SSHJnlHd39u5LJmWlVdluQnkhxRVduSvKG7N+z5rIPeuiQ/l+S2YehxkvxKd189Yk2z7sgklw53HDkkyeXd/YGRa+LA87QkV87lvlmW5L3d/aFxS5p5r07ynuGLmNuTnD9yPQAszoeSvLKqbk3yF5mbAvdduvv+qvrpJL8zfNH3rcx92QccJKrbEkEAAAAATMb0NwAAAAAmJlQCAAAAYGJCJQAAAAAmJlQCAAAAYGJCJQAAAAAmJlQCAAAAYGJCJQAAAAAmJlQCAAAAYGJCJQAAAAAmJlQCAAAAYGJCJQAAAAAmJlQCAAAAYGJCJQAAAAAmJlQCAAAAYGJCJQAAAAAmJlQCAAAAYGJCJQAAAAAmJlQCAAAAYGJCJQAAAAAmJlQCAAAAYGJCJWBmVNUdVfWtqvrGvMfTx64LAOBA5hoMWCyhEjBrfqq7nzDvsX2Sk6vq0H1VGADAAcw1GDAxoRIw06rqkKq6oqq+VFX3VNX1VXX0vOPvrqrfraoPVdU3k/xYVT2mqn6zqrZW1Zer6veq6jEj/hgAAPsV12DAIyFUAvYHH0iyJskPJPlskj/Y7fjPJPm1JE9M8t+T/EaSZyQ5djhvdZLXT6lWAIADhWswYI+qu8euASDJ3Hz+JEckeWBour67X7JbnyOS7EjyhO7+ZlW9O8n93f3Ph+OHJLkvybO6+6+Gth9L8o7uXjOdnwQAYP/hGgxYrGVjFwCwm5d095/u3Bnm578pyUszd7Hz98OhI5J8c9jeOu/8H0jy6CSfqapdL7MvCwYAOAC4BgMmJlQCZt0rkpyV5NQkf5XkKZn7lmz+Rcr8IZdfTnJ/5r4l+/K0igQAOMC4BgP2yppKwKx7YpLvJPnbJI9L8sY9de7uB5O8PclvVdXymrOiqk7f96UCABwwXIMBeyVUAmbdO5NsHx6fS3LjIzjnlzP3jdpNSe5Nck3mFosEAOCRcQ0G7JWFugEAAACYmJFKAAAAAExMqAQAAADAxIRKAAAAAExMqAQAAADAxKYSKlXVY6rqpqr6TFV9rqp+bWj//ar6H1V1y/A4fmivqnprVW2pqlur6oRp1AkAAADAI7NsSu/znSSndvc3quqwJB+rqv82HPu33X3Fbv3PzNytJ9ck+ZEkFw/Pe3TEEUf06tWrl65qAGCm3HzzzX/T3cvHroN/4PoLAA58D3cNNpVQqbs7yTeG3cOGR+/hlHOSvGs47+NVdXhVHdndd+3pfVavXp1NmzYtSc0AwOypqr8auwYeyvUXABz4Hu4abGprKlXVoVV1S5K7k1zb3Z8YDr1xmOL2lqp69NB2VJKt807fNrQBAAAAMAOmFip194PdfXySFUlOrKrnJnldkmcn+eEkT07y2qF7LfQSC71uVa2vqk1VtWnHjh37oHIAAAAAdjf1u7919z1Jrk9yRnff1XO+k+SdSU4cum1LsnLeaSuSbH+Y17uku9d299rlyy2xAAAAADAN07r72/KqOnzYfmySFyT5QlUdObRVkpck+exwysYkrxjuAndSknv3tp4SAAAAANMzrbu/HZnk0qo6NHNB1uXd/YGq+mhVLc/cdLdbkrxy6H91krOSbElyX5Lzp1QnAAAAAI/AtO7+dmuS5y3QfurD9O8kF+7rugAAAABYnKmvqQQAAADA/k+oBAAAwF696EUvyiGHHJKqyiGHHJIXvehFY5cEjEyoBAAAwB696EUvyjXXXJNXvvKVueeee/LKV74y11xzjWAJDnLTWqgbAACA/dS1116bX/iFX8jv/d7vJcmu57e97W1jlgWMzEglAAAA9qi786Y3vekhbW9605syd48l4GBlpNIjtPqiD+61zx1vfvEUKgEAAJiuqsrrXve6XSOUkuR1r3tdqmrEqoCxGakEAHAAq6o7quq2qrqlqjYNbU+uqmur6ovD8/cP7VVVb62qLVV1a1WdMG71wKx44QtfmIsvvjivetWrcu+99+ZVr3pVLr744rzwhS8cuzRgREIlAIAD3yndfXx3rx32L0ryke5ek+Qjw36SnJlkzfBYn+TiqVcKzKQPf/jDOf300/O2t70thx9+eN72trfl9NNPz4c//OGxSwNGZPobAMDB55wkPzFsX5rk+iSvHdrf1XOLpHy8qg6vqiO7+65RqgRmigAJ2J2RSgAAB7ZOck1V3VxV64e2p+0Miobnpw7tRyXZOu/cbUMbAMB3MVIJAODAtq67t1fVU5NcW1Vf2EPfhVbc/a5bOw3h1PokWbVq1dJUCQDsd4xUAgA4gHX39uH57iRXJjkxyZer6sgkGZ7vHrpvS7Jy3ukrkmxf4DUv6e613b12+fLl+7J8AGCGCZUAAA5QVfX4qnrizu0kpyf5bJKNSc4bup2X5Kphe2OSVwx3gTspyb3WUwIAHo7pbwAAB66nJbmyqpK56773dveHquqTSS6vqguS3JnkZUP/q5OclWRLkvuSnD/9kgGA/YVQCQDgANXdtyc5boH2v01y2gLtneTCKZQGABwATH8DAAAAYGJCJQAAAAAmJlQCAAAAYGJCJQAAAAAmJlQCAAAAYGJCJQAAAAAmJlQCAAAAYGJCJQAAAAAmJlQCAAAAYGJCJQAAAAAmJlQCAAAAYGJCJQAAAAAmNrVQqaoeU1U3VdVnqupzVfVrQ/szquoTVfXFqvrDqnrU0P7oYX/LcHz1tGoFAAAAYM+mOVLpO0lO7e7jkhyf5IyqOinJryd5S3evSfLVJBcM/S9I8tXufmaStwz9AAAAAJgBUwuVes43ht3DhkcnOTXJFUP7pUleMmyfM+xnOH5aVdWUygUAAABgD6a6plJVHVpVtyS5O8m1Sf4yyT3d/cDQZVuSo4bto5JsTZLh+L1JnjLNegEAAABY2FRDpe5+sLuPT7IiyYlJjl6o2/C80Kik3r2hqtZX1aaq2rRjx46lKxYAAACAhzXK3d+6+54k1yc5KcnhVbVsOLQiyfZhe1uSlUkyHP++JF9Z4LUu6e613b12+fLl+7p0AAAAADLdu78tr6rDh+3HJnlBks1Jrkvy0qHbeUmuGrY3DvsZjn+0u79rpBIAAAAA07ds712WzJFJLq2qQzMXZl3e3R+oqs8neV9V/ackn06yYei/IckfVNWWzI1QevkUawUAAABgD6YWKnX3rUmet0D77ZlbX2n39m8nedkUSgMA/v/27j/K7rq+8/jznR9OoFADMgSaBEPb1AXJ8KNT5Cx7zoqUFrCdsbvSxeMqddOkkSSrG9gSoWeh59QIxyLWoZtp0nAIropUtJnVnEpELe1xBSLSDD8UBqRmmkgG5EcMJob43j/ud+JNGCa5NzP3e2fu83HOPff7/Xw/9847nwN3ktf9fD5fSZIkqUal7KkkSZIkSZKkic1QSZIkSZIkSTUzVJIkSZIkSVLNDJUkSZIkSZJUM0MlSZIkSZIk1cxQSZIkSZIkSTUzVJIkSZIkSVLNDJUkSZIkSZJUM0MlSZIkSZIk1cxQSZIkSZIkSTUzVJIkSZIkSVLNDJUkSZIkSZJUM0MlSZIkSZIk1cxQSZIkSZIkSTUzVJIkSZIkSVLNDJUkSZImuYiYGhHfjYgvF+enRsT9EfFkRHw+It5QtLcV5wPF9Xll1i1JkpqboZIkSdLk9yHg8arzm4BbMnM+8AKwsGhfCLyQmb8O3FL0kyRJGpGhkiRJ0iQWEXOAdwJ/W5wH8A7gC0WX9cC7iuPu4pzi+oVFf0mSpNcwVJIkSZrcPgn8KfDz4vxNwIuZ+WpxPgjMLo5nA1sBiusvFf0lSZJew1BJkiRpkoqI3wN2ZOZ3qptH6JqHca36fRdHxOaI2Dw0NDQGlUqSpInIUEmSJGnyOh/oiohngDupLHv7JDAzIqYVfeYA24rjQWAuQHH9jcCPD37TzFyTmZ2Z2dne3j6+fwJJktS0DJUkSZImqcz8SGbOycx5wOXA1zPzvcA3gHcX3a4ANhTHfcU5xfWvZ+ZrZipJkiSBoZIkSVIrugZYEREDVPZMWle0rwPeVLSvAFaWVJ8kSZoAph26iyRJkia6zPwm8M3i+Gng3BH67AYua2hhkiRpwnKmkiRJkiRJkmpmqCRJkiRJkqSaGSpJkiRJkiSpZoZKkiRJkiRJqllDQqWImBsR34iIxyPi0Yj4UNF+Q0T8W0Q8XDwurXrNRyJiICK+HxG/24g6JUmSJEmSdHgadfe3V4GrMvOhiDgW+E5EbCqu3ZKZf1ndOSJOBy4H3gr8CvC1iPiNzNzXoHolSZIkSZI0iobMVMrM7Zn5UHG8E3gcmD3KS7qBOzNzT2b+ABhghNveSpIkSZIkqRwN31MpIuYBZwP3F03LImJLRNwWEccVbbOBrVUvG2T0EEqSJEmSJEkN1NBQKSKOAe4GPpyZLwOrgV8DzgK2AzcPdx3h5fk677k4IjZHxOahoaFxqFqSJEmSJEkHa1ioFBHTqQRKn8nMLwJk5rOZuS8zfw6s5RdL3AaBuVUvnwNsG+l9M3NNZnZmZmd7e/v4/QEkSZIkSZK0X6Pu/hbAOuDxzPxEVfvJVd3+AHikOO4DLo+Itog4FZgPPNCIWiVJkiRJknRojbr72/nA+4D+iHi4aLsWeE9EnEVladszwJ8AZOajEXEX8BiVO8ct9c5vkiRJkiRJzaMhoVJm/jMj75O0cZTXfBT46LgVJUmSJEmSpLo1/O5vkiRJkiRJmvgMlSRJkiRJklQzQyVJkiRJkiTVzFBJkiRJkiRJNTNUkiRJkiRJUs0MlSRJkiRJklQzQyVJkiRJkiTVzFBJkiRJkiRJNTNUkiRJkiRJUs0MlSRJkiRJklQzQyVJkiRJkiTVzFBJkiRJkiRJNTNUkiRJkiRJUs0MlSRJkiRJklQzQyVJkiRJkiTVzFBJkiRJkiRJNTNUkiRJkiRJUs0MlSRJkiRJklQzQyVJkiRJkiTVzFBJkiRJkiRJNTNUkiRJkiRJUs0MlSRJkiRJklQzQyVJkiRJkiTVrK5QKSLOGOtCJEmSJEmSNHHUO1OpNyIeiIgrI2LmmFYkSZKkMRERM4q/s/1LRDwaEX9etJ8aEfdHxJMR8fmIeEPR3lacDxTX55VZvyRJam51hUqZ+R+A9wJzgc0R8dmIuGhMK5MkSdKR2gO8IzPPBM4CLo6I84CbgFsycz7wArCw6L8QeCEzfx24pegnSZI0orr3VMrMJ4E/A64B/iPwqYj4XkT8p7EqTpIkSfXLip8Up9OLRwLvAL5QtK8H3lUcdxfnFNcvjIhoULmSJGmCqXdPpY6IuAV4nMpfSn4/M08rjm8Zof/ciPhGRDxeTL3+UNF+fERsKqZeb4qI44r2iIhPFVOvt0TEOXX/CSVJklpYREyNiIeBHcAm4Cngxcx8tegyCMwujmcDWwGK6y8Bb2psxZIkaaKod6bSrcBDwJmZuTQzHwLIzG1UZi8d7FXgqiJ4Og9YGhGnAyuBe4up1/cW5wCXAPOLx2JgdZ11SpIktbTM3JeZZwFzgHOB00bqVjyPNCspD26IiMURsTkiNg8NDY1dsZIkaUKpN1S6FPhsZv4UICKmRMTRAJn56YM7Z+b2quBpJ5UZTrM5cIr1wVOv7yimbH8bmBkRJ9dZqyRJUsvLzBeBb1L5gm9mREwrLs0BthXHg1T2zKS4/kbgxyO815rM7MzMzvb29vEuXZIkNal6Q6WvAUdVnR9dtB1ScReRs4H7gVmZuR0qwRNwYtFt/9TrQvW0bEmSJB2GiGgfvlNvRBwF/DaVL/e+Aby76HYFsKE47ivOKa5/PTNfM1NJkiQJYNqhu4xoRtWmj2TmT4ZnKo0mIo4B7gY+nJkvj7Lv42FNvS7eczGVJXKccsophypBkiSplZwMrI+IqVS+TLwrM78cEY8Bd0bEXwDfBdYV/dcBn46IASozlC4vo2hJkjQx1Bsq7YqIc4aXtEXEbwI/He0FETGdSqD0mcz8YtH8bEScnJnbi+VtO4r2/VOvC9XTsg+QmWuANQCdnZ1+kyZJklTIzC1UZogf3P40lf2VDm7fDVzWgNIkSdIkUO/ytw8DfxcR/xQR/wR8Hlj2ep2LW9GuAx7PzE9UXaqeYn3w1Ov3F3eBOw94aXiZnCRJkiRJkspX10ylzHwwIv4d8BYqS9W+l5l7R3nJ+cD7gP7ilrYA1wI3AndFxELgh/zim7GNVDYDHwBeAT5QT52SJEmSJEkaH/UufwP4LWBe8R5nRwSZecdIHTPznxl5nySAC0fon8DSI6hNkiRJkiRJ46iuUCkiPg38GvAwsK9oTmDEUEmSJEmSJEmTS70zlTqB073FrCRJkiRJUmuqd6PuR4CTxrIQSZIkSZIkTRz1zlQ6AXgsIh4A9gw3ZmbXmFQlSZIkSZKkplZvqHTDWBYhSZIkSZKkiaWu5W+Z+Y/AM8D04vhB4KExrEuSJEmS1EQ6OjqIiP2Pjo6OskuSVLK6QqWIWAR8Afibomk28PdjVZQkSZIkqXl0dHTQ39/PlCmVf0JOmTKF/v5+gyWpxdW7UfdS4HzgZYDMfBI4cayKkiRJkiQ1j/7+fiKCj3/84+zatYuPf/zjRAT9/f1llyapRPXuqbQnM38WEQBExDQgx6yqCWreyq8css8zN76zAZVIkiRJ0ti64YYbWLFiBQArVqzgJz/5Cddff33JVUkqU70zlf4xIq4FjoqIi4C/A/7v2JUlSZIkSWom999//6jnklpPvaHSSmAI6Af+BNgI/NlYFSVJkiRJah4RwcaNG+nu7ua5556ju7ubjRs3Mrx6RVJrqmv5W2b+HFhbPCRJkiRJk9jSpUu59dZb6evro729/YB2Sa2rrlApIn7ACHsoZeavHnFFkiRJkqSm0tPTA8DatWvZs2cPbW1tLFq0aH+7pNZU70bdnVXHM4DLgOOPvBxJkiRJkiRNBHXtqZSZz1c9/i0zPwm8Y4xrkyRJkiQ1geXLl9Pb28uqVavYtWsXq1atore3l+XLl5ddmqQS1bv87Zyq0ylUZi4dOyYVSZIkSZKaytq1a7nppptYsWIFwP7na6+91iVwUgurd/nbzVXHrwLPAH94xNVIkiRJkprOnj17WLJkyQFtS5Ys4aqrriqpIknNoN67v10w1oVIkiRJkppTW1sbvb29+2coAfT29tLW1lZiVZLKVu/ytxWjXc/MT9RXjiRJkiSp2SxatIhrrrkGqMxQ6u3t5ZprrnnN7CVJreVI7v72W0Bfcf77wH3A1rEoSpIkSZLUPIb3Tbr22mu56qqraGtrY8mSJe6nJLW4ekOlE4BzMnMnQETcAPxdZv7xWBUmSZIkSWoePT09hkiSDjClztedAvys6vxnwLwjrkaSJEmSJEkTQr2h0qeBByLihoi4HrgfuGPsypIkSZIkNZOOjg4iYv+jo6Oj7JIklayuUCkzPwp8AHgBeBH4QGauGsvCJEmSJEnNoaOjg/7+frq6uhgaGqKrq4v+/n6DJanF1TtTCeBo4OXM/CtgMCJOHaOaJEmSJElNpL+/n7PPPpunnnqKWbNm8dRTT3H22WfT399fdmmSSlRXqFQsebsG+EjRNB34P2NVlCRJkiSpuTz//PP09PSwe/duenp6eP7558suSVLJ6p2p9AdAF7ALIDO3AceOVVGSJEmSpOYyZ84cLrjgAqZPn84FF1zAnDlzyi5JUsnqDZV+lpkJJEBE/NKhXhARt0XEjoh4pKrthoj4t4h4uHhcWnXtIxExEBHfj4jfrbNOSZIkSdIY+Na3vkV3dzfPPfcc3d3dfOtb3yq7JEklm1bn6+6KiL8BZkbEIuC/AWsP8ZrbgVt57V3ibsnMv6xuiIjTgcuBtwK/AnwtIn4jM/fVWa8kSZIkqU5vfetbeeWVV+jr66O9vR2AU089laOPPrrkyiSVqd67v/0l8AXgbuAtwP/KzJ5DvOY+4MeH+SO6gTszc09m/gAYAM6tp1ZJkiRJ0pG57rrrGBwcPKBtcHCQ6667rqSKJDWDmkOliJgaEV/LzE2Z+T8z8+rM3HQENSyLiC3F8rjjirbZwNaqPoNFmyRJkiSpwT72sY+xd+9ejjnmGACOOeYY9u7dy8c+9rGSK5NUpppDpWIJ2isR8cYx+PmrgV8DzgK2AzcX7THSjx7pDSJicURsjojNQ0NDY1CSJEmSJKlaf38/XV1d7Ny5k8xk586ddHV10d/fX3ZpkkpU755Ku4H+iNhEcQc4gMz877W8SWY+O3wcEWuBLxeng8Dcqq5zgG2v8x5rgDUAnZ2dIwZPkiRJkqQjs27dutecD++vJKk11RsqfaV4HJGIODkztxenfwAM3xmuD/hsRHyCykbd84EHjvTnNYN5Kw89bM/c+M4GVCJJkia7iJhL5SYpJwE/B9Zk5l9FxPHA54F5wDPAH2bmCxERwF8BlwKvAH+UmQ+VUbuk5rNw4UI2bNhwwLmk1lZTqBQRp2TmDzNzfa0/KCI+B7wdOCEiBoHrgbdHxFlUlrY9A/wJQGY+GhF3AY8BrwJLvfObJElSzV4FrsrMhyLiWOA7xUzzPwLuzcwbI2IlsBK4BriEypd584G3Udmq4G2lVC6pqSxYsIC+vj66u7tZt24dCxcupK+vjwULFpRdmqQS1TpT6e+BcwAi4u7M/M+H+8LMfM8IzetGaBvu/1HgozXWJ0mSpEIxI3x7cbwzIh6ncvOTbipf9gGsB75JJVTqBu7IzAS+HREzD5pZLqlFbdmyhY6ODvr6+vYveVuwYAFbtmwpuTJJZao1VKreQPtXx7IQSZIkjZ+ImAecDdwPzBoOijJze0ScWHR7vTvwGipJMkCS9Bq13v0tX+dYkiRJTSoijgHuBj6cmS+P1nWEttf8nc+770qSJKg9VDozIl6OiJ1AR3H8ckTsjIjR/oIiSZKkEkTEdCqB0mcy84tF87MRcXJx/WRgR9F+WHfgzcw1mdmZmZ3e+UmSpNZVU6iUmVMz85cz89jMnFYcD5//8ngVKUmSpNoVd3NbBzyemZ+outQHXFEcXwFsqGp/f1ScB7zkfkqSJOn11LqnkiRJkiaO84H3Af0R8XDRdi1wI3BXRCwEfghcVlzbCFwKDACvAB9obLmSJGkiMVSSJEmapDLznxl5nySAC0fon8DScS1KkiRNGrXuqSRJkiRJkiQZKkmSJEmSJKl2hkqSJEmSJEmqmaGSJEmSJEmSamaoJEmSJEmSpJoZKkmSJEmSJKlmhkqSJEmSJEmqmaGSJEmSJEmSamaoJEmSJEk6pOXLlzNjxgwighkzZrB8+fKyS5JUMkMlSZIkSdKoli9fTm9vL6tWrWLXrl2sWrWK3t5egyWpxRkqSZIkSZJGtXbtWm666SZWrFjB0UcfzYoVK7jppptYu3Zt2aVJKpGhkiRJkiRpVHv27GHJkiUHtC1ZsoQ9e/aUVJGkZmCoJEmSJEkaVVtbG729vQe09fb20tbWVlJFkpqBoZIkSZIkaVSLFi3i6quvJiL2P66++moWLVpUdmmSSmSoJEmSJEka1RNPPEFmHtCWmTzxxBMlVSSpGRgqSZIkSZJGdc899wDQ1dXF0NAQXV1dB7RLak3Tyi5AkiRJktT85s6dy1e/+lXa29tpa2tj7ty5bN26teyyJJXImUqSJEmSpEPaunUrq1atYteuXaxatcpASZKhkiRJkiTp8AwMDLB3714GBgbKLkVSE3D5myRJkiTpsKxevZrVq1eXXYakJuFMJUmSJEnSqNra2jjppJMOaDvppJNoa2srqSJJzaBhoVJE3BYROyLikaq24yNiU0Q8WTwfV7RHRHwqIgYiYktEnNOoOiVJkiRJB1q0aBE/+tGPDmj70Y9+xKJFi0qqSFIzaORMpduBiw9qWwncm5nzgXuLc4BLgPnFYzHg/EpJkiRJKskTTzxRU7uk1tCwUCkz7wN+fFBzN7C+OF4PvKuq/Y6s+DYwMyJObkylkiRJkqRq99xzDwBdXV0MDQ3R1dV1QLuk1lT2nkqzMnM7QPF8YtE+G6i+P+Vg0SZJkiRJKsEll1zChg0bOOGEE9iwYQOXXHJJ2SVJKlnZodLriRHacsSOEYsjYnNEbB4aGhrnsiRJkiSpNc2aNWvUc0mtp+xQ6dnhZW3F846ifRCYW9VvDrBtpDfIzDWZ2ZmZne3t7eNarCRJkiS1qttvv50rr7ySl156iSuvvJLbb7+97JIklazsUKkPuKI4vgLYUNX+/uIucOcBLw0vk5MkSZIkNdaCBQsAWL16NTNnzmT16tUHtEtqTQ0LlSLic8D/A94SEYMRsRC4EbgoIp4ELirOATYCTwMDwFrgykbVKUmSJEk60JYtW14TIC1YsIAtW7aUVJGkZjCtUT8oM9/zOpcuHKFvAkvHtyJJkiRJkiTVq+zlCZ6fGgAADRZJREFUb5IkSZKkJtfR0UF/fz9dXV0MDQ3R1dVFf38/HR0dZZcmqUSGSpIkSZKkUQ0HShs2bOCEE05gw4YN+4MlSa3LUEmSJEmSdEgvvvgiU6ZMISKYMmUKL774YtklSSpZw/ZUkiRJkiRNXPfdd9/+48w84FxSa3KmkiRJkiRJkmpmqCRJkiRJkqSaGSpJkiRJkg7LzTffzK5du7j55pvLLkVSEzBUkiRJmsQi4raI2BERj1S1HR8RmyLiyeL5uKI9IuJTETEQEVsi4pzyKpfUjAYGBti7dy8DAwNllyKpCRgqSZIkTW63Axcf1LYSuDcz5wP3FucAlwDzi8diYHWDapQ0QaxevZqZM2eyerUfD5IMlSRJkia1zLwP+PFBzd3A+uJ4PfCuqvY7suLbwMyIOLkxlUpqZlOnTq2pXVJrMFSSJElqPbMycztA8Xxi0T4b2FrVb7Bok9TiPvjBDxIR+0OkqVOnEhF88IMfLLkySWUyVJIkSdKwGKEtX9MpYnFEbI6IzUNDQw0oS1LZenp6OOOMM9i3bx8A+/bt44wzzqCnp6fkyiSVyVBJkiSp9Tw7vKyteN5RtA8Cc6v6zQG2HfzizFyTmZ2Z2dne3j7uxUoq3/Lly+nv7z+grb+/n+XLl5dUkaRmYKgkSZLUevqAK4rjK4ANVe3vL+4Cdx7w0vAyOUmt7dZbbwUgIg54Hm6X1JqmlV2AJEmSxk9EfA54O3BCRAwC1wM3AndFxELgh8BlRfeNwKXAAPAK8IGGFyypqWXmAc+SWpuhkiRJ0iSWme95nUsXjtA3gaXjW5EkSZosXP4mSZIkSZKkmhkqSZIkSZIkqWaGSpIkSZIkSaqZoZIkSZIkSZJqZqgkSZIkSZKkmhkqSZIkSZIkqWaGSpIkSZIkSaqZoZIkSZIkSZJqZqgkSZIkSZKkmhkqSZIkSZIOy4wZMw54ltTaDJUkSZIkSYdl9+7dBzxLam3Tyi5AkiRJkjS+IqJp3zszx6gSSY3WFDOVIuKZiOiPiIcjYnPRdnxEbIqIJ4vn48quU5IkSZImosw8oseyZctGfN9ly5Yd8XtLmriaIlQqXJCZZ2VmZ3G+Erg3M+cD9xbnkiRJkqQG6+npYdmyZbS1tQHQ1tbGsmXL6OnpKbkySWVqplDpYN3A+uJ4PfCuEmuRJEmSpJbW09PD7t27efM1X2b37t0GSpKaJlRK4J6I+E5ELC7aZmXmdoDi+cTSqpMkSZIkSdIBmmWj7vMzc1tEnAhsiojvHe4LixBqMcApp5wyXvU11LyVXzmsfs/c+M5xrkSSJEmSJGlkTTFTKTO3Fc87gC8B5wLPRsTJAMXzjtd57ZrM7MzMzvb29kaVLEmSJEmS1NJKD5Ui4pci4tjhY+B3gEeAPuCKotsVwIZyKpQkSZIkSdLBmmH52yzgSxEBlXo+m5n/EBEPAndFxELgh8BlJdYoSZIkSZKkKqWHSpn5NHDmCO3PAxc2viJJkiRJkiQdSunL3yRJkiRJkjTxGCpJkiRJkiSpZoZKkiRJkiRJqlnpeyppfM1b+ZVD9nnmxnc2oBJJkiRJkjSZOFNJkiRJkiRJNXOmkiRJkiQ1oTP//B5e+unesssY0eGsiGi0Nx41nX+5/nfKLkNqKYZKkiRJktSEXvrpXreqqEEzBl3SZOfyN0mSJEmSJNXMUEmSJEmSJEk1M1SSJEmSJElSzdxTSYflcNYnu95bkiRJkqTW4UwlSZIkSZIk1cxQSZIkSZIkSTVz+ZskSZIkNaFjT1vJgvUryy5jwjj2NAC35JAayVBJkiRJkprQzsdvdN/SGhzOPrCSxpbL3yRJknSAiLg4Ir4fEQMR4TQJSZI0ImcqSZIkab+ImAr8NXARMAg8GBF9mflYuZVJrcnZN4fvjUdNL7sEqeUYKkmSJKnaucBAZj4NEBF3At2AoZLUYM269G3eyq80bW2SGstQSWP27cfhvo+/gCRJamqzga1V54PA20qqRZIkNTFDpQnMqbCSJGkcxAhteUCHiMXAYoBTTjmlETVJOkIRI/2vfQTvd9PYvVdmHrqTpKZkqCRJkqRqg8DcqvM5wLbqDpm5BlgD0NnZ6b8GpQnA4EbSePDub5IkSar2IDA/Ik6NiDcAlwN9JdckSZKakDOVJEmStF9mvhoRy4CvAlOB2zLz0ZLLkiRJTchQSZIkSQfIzI3AxrLrkCRJzc1QSTpMh7Mxune2kyRJkiS1CkMlNSUDHEmSJEmSmpuhkiassQyeDue91HwMHyVJkiSpPE1997eIuDgivh8RAxGxsux6JEmSJEmSVNG0M5UiYirw18BFwCDwYET0ZeZj5VYmTT6NnvXl7CFJkiRJmviaNlQCzgUGMvNpgIi4E+gGDJUmuFZfambocnha/b8TqVn5GSZJkqRhzRwqzQa2Vp0PAm8rqRap5RnySJIkSZKqRWaWXcOIIuIy4Hcz84+L8/cB52bm8oP6LQYWF6dvAb4/TiWdADw3Tu+tX3CcG8NxbgzHuTEc58ZolnF+c2a2l12EfiEihoB/LbsOSQ3VLL8TJDXOiH8Ha+aZSoPA3KrzOcC2gztl5hpgzXgXExGbM7NzvH9Oq3OcG8NxbgzHuTEc58ZwnPV6DPmk1uPvBEnDmvnubw8C8yPi1Ih4A3A50FdyTZIkSZIkSaKJZypl5qsRsQz4KjAVuC0zHy25LEmSJEmSJNHEoRJAZm4ENpZdR2Hcl9gJcJwbxXFuDMe5MRznxnCcJUnD/J0gCWjijbolSZIkSZLUvJp5TyVJkiRJkiQ1KUOlQ4iIiyPi+xExEBEry65nIouI2yJiR0Q8UtV2fERsiogni+fjivaIiE8V474lIs4pr/KJJSLmRsQ3IuLxiHg0Ij5UtDvWYygiZkTEAxHxL8U4/3nRfmpE3F+M8+eLGw0QEW3F+UBxfV6Z9U80ETE1Ir4bEV8uzh3nMRYRz0REf0Q8HBGbizY/NySphUTEvuL3wPBjXtk1SWpuhkqjiIipwF8DlwCnA++JiNPLrWpCux24+KC2lcC9mTkfuLc4h8qYzy8ei4HVDapxMngVuCozTwPOA5YW/9061mNrD/COzDwTOAu4OCLOA24CbinG+QVgYdF/IfBCZv46cEvRT4fvQ8DjVeeO8/i4IDPPqrpNtJ8bktRaflr8Hhh+PHM4Lyr+3SSpBRkqje5cYCAzn87MnwF3At0l1zRhZeZ9wI8Pau4G1hfH64F3VbXfkRXfBmZGxMmNqXRiy8ztmflQcbyTyj/EZ+NYj6livH5SnE4vHgm8A/hC0X7wOA+P/xeACyMiGlTuhBYRc4B3An9bnAeOc6P4uSFJLS4i5kXEP0XEQ8Xj3xftby9mx38W6C/a/msxk/vhiPgbwyZp8jNUGt1sYGvV+WDRprEzKzO3QyUMAU4s2h37MVBMWT4buB/HeswVS7IeBnYAm4CngBcz89WiS/VY7h/n4vpLwJsaW/GE9UngT4GfF+dvwnEeDwncExHfiYjFRZufG5LUWo6qWvr2paJtB3BRZp4D/BfgU1X9zwWuy8zTI+K04vr5mXkWsA94byOLl9R408ouoMmN9O22t8trDMf+CEXEMcDdwIcz8+VRJms41nXKzH3AWRExE/gScNpI3Ypnx7kOEfF7wI7M/E5EvH24eYSujvOROz8zt0XEicCmiPjeKH0dZ0manH5aBELVpgO3RsRwUPQbVdceyMwfFMcXAr8JPFj8vfMoKoGUpEnMUGl0g8DcqvM5wLaSapmsno2IkzNze7F0YvgXj2N/BCJiOpVA6TOZ+cWi2bEeJ5n5YkR8k8oeVjMjYloxS6Z6LIfHeTAipgFv5LXLQfVa5wNdEXEpMAP4ZSozlxznMZaZ24rnHcW30+fi54YkCf4H8CxwJpWVLrurru2qOg5gfWZ+pIG1SSqZy99G9yAwv7jL0BuAy4G+kmuabPqAK4rjK4ANVe3vL+4wdB7w0vASDI2u2D9mHfB4Zn6i6pJjPYYior2YoUREHAX8NpX9q74BvLvodvA4D4//u4GvZ6YzOw4hMz+SmXMycx6Vz+CvZ+Z7cZzHVET8UkQcO3wM/A7wCH5uSJIqX9Bsz8yfA+8DXm+fpHuBdxczXofvIPrmBtUoqSTOVBpFZr4aEcuAr1L58LwtMx8tuawJKyI+B7wdOCEiBoHrgRuBuyJiIfBD4LKi+0bgUmAAeAX4QMMLnrjOp/ILv7/Y7wfgWhzrsXYysL7YgHIKcFdmfjkiHgPujIi/AL5LJeCjeP50RAxQmTlzeRlFTyLX4DiPpVnAl4rlCtOAz2bmP0TEg/i5IUmt7n8Dd0fEZVS+1Nk1UqfMfCwi/ozK/nxTgL3AUuBfG1appIYLv8CVJEmSJElSrVz+JkmSJEmSpJoZKkmSJEmSJKlmhkqSJEmSJEmqmaGSJEmSJEmSamaoJEmSJEmSpJoZKkmSJEmSJKlmhkqSJEmSJEmqmaGSJEmSJEmSavb/AVSNuXsEODuVAAAAAElFTkSuQmCC\n",
      "text/plain": [
       "<Figure size 1440x3600 with 14 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "plot_num(train_data)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 67,
   "metadata": {
    "scrolled": false
   },
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1440x3600 with 12 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "plot_num(test_data)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Between the two datasets there does not seem to be any obvious differences in distribution of the features.\n",
    "\n",
    "We see that we have some outliers in Fare, SibSp and Parch. We need to handle them later."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Data Pre-Processing: Feature Engineering <a class=\"anchor\" id=\"third-bullet\"></a>"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Title"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "At a first glance, it seems like the Name variable is unnecessary but let's see if we can extract additional information from it e.g. through extracting the title. "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 68,
   "metadata": {
    "scrolled": false
   },
   "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>LastName</th>\n",
       "      <th>Title</th>\n",
       "      <th>FirstName</th>\n",
       "      <th>Additional</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>Braund</td>\n",
       "      <td>Mr</td>\n",
       "      <td>Owen Harris</td>\n",
       "      <td>None</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>Cumings</td>\n",
       "      <td>Mrs</td>\n",
       "      <td>John Bradley (Florence Briggs Thayer)</td>\n",
       "      <td>None</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>Heikkinen</td>\n",
       "      <td>Miss</td>\n",
       "      <td>Laina</td>\n",
       "      <td>None</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>Futrelle</td>\n",
       "      <td>Mrs</td>\n",
       "      <td>Jacques Heath (Lily May Peel)</td>\n",
       "      <td>None</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>Allen</td>\n",
       "      <td>Mr</td>\n",
       "      <td>William Henry</td>\n",
       "      <td>None</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>413</th>\n",
       "      <td>Spector</td>\n",
       "      <td>Mr</td>\n",
       "      <td>Woolf</td>\n",
       "      <td>None</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>414</th>\n",
       "      <td>Oliva y Ocana</td>\n",
       "      <td>Dona</td>\n",
       "      <td>Fermina</td>\n",
       "      <td>None</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>415</th>\n",
       "      <td>Saether</td>\n",
       "      <td>Mr</td>\n",
       "      <td>Simon Sivertsen</td>\n",
       "      <td>None</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>416</th>\n",
       "      <td>Ware</td>\n",
       "      <td>Mr</td>\n",
       "      <td>Frederick</td>\n",
       "      <td>None</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>417</th>\n",
       "      <td>Peter</td>\n",
       "      <td>Master</td>\n",
       "      <td>Michael J</td>\n",
       "      <td>None</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>1309 rows × 4 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "          LastName    Title                               FirstName Additional\n",
       "0           Braund       Mr                             Owen Harris       None\n",
       "1          Cumings      Mrs   John Bradley (Florence Briggs Thayer)       None\n",
       "2        Heikkinen     Miss                                   Laina       None\n",
       "3         Futrelle      Mrs           Jacques Heath (Lily May Peel)       None\n",
       "4            Allen       Mr                           William Henry       None\n",
       "..             ...      ...                                     ...        ...\n",
       "413        Spector       Mr                                   Woolf       None\n",
       "414  Oliva y Ocana     Dona                                 Fermina       None\n",
       "415        Saether       Mr                         Simon Sivertsen       None\n",
       "416           Ware       Mr                               Frederick       None\n",
       "417          Peter   Master                               Michael J       None\n",
       "\n",
       "[1309 rows x 4 columns]"
      ]
     },
     "execution_count": 68,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "name = df[\"Name\"].copy()\n",
    "name = name.str.split(\"[,.]\", expand=True)\n",
    "name.columns = [\"LastName\",\"Title\",\"FirstName\",\"Additional\"]\n",
    "name"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "We seem to have one name with more than one , or . in it (i.e. column 3 is not NA)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 69,
   "metadata": {
    "scrolled": false
   },
   "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>LastName</th>\n",
       "      <th>Title</th>\n",
       "      <th>FirstName</th>\n",
       "      <th>Additional</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>513</th>\n",
       "      <td>Rothschild</td>\n",
       "      <td>Mrs</td>\n",
       "      <td>Martin (Elizabeth L</td>\n",
       "      <td>Barrett)</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "       LastName Title             FirstName Additional\n",
       "513  Rothschild   Mrs   Martin (Elizabeth L   Barrett)"
      ]
     },
     "execution_count": 69,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "name[name.iloc[:,3].isna()==False]"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Since this is only ony case, we can ignore the additional column"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 70,
   "metadata": {
    "scrolled": true
   },
   "outputs": [],
   "source": [
    "name = name.iloc[:,0:3]"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Let's see what and how many different titles we currently have."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 71,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([' Mr', ' Mrs', ' Miss', ' Master', ' Don', ' Rev', ' Dr', ' Mme',\n",
       "       ' Ms', ' Major', ' Lady', ' Sir', ' Mlle', ' Col', ' Capt',\n",
       "       ' the Countess', ' Jonkheer', ' Dona'], dtype=object)"
      ]
     },
     "execution_count": 71,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "name.iloc[:,1].unique()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 72,
   "metadata": {
    "scrolled": false
   },
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1152x576 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.figure(figsize=(16,8))\n",
    "sns.barplot(x=name.iloc[:,1].value_counts().index,y=name.iloc[:,1].value_counts()).set_title('Title Feature Value Counts Before Grouping', size=20, y=1.05)\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 73,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       " Mr              757\n",
       " Miss            260\n",
       " Mrs             197\n",
       " Master           61\n",
       " Rev               8\n",
       " Dr                8\n",
       " Col               4\n",
       " Mlle              2\n",
       " Ms                2\n",
       " Major             2\n",
       " Dona              1\n",
       " Jonkheer          1\n",
       " Mme               1\n",
       " Capt              1\n",
       " the Countess      1\n",
       " Sir               1\n",
       " Lady              1\n",
       " Don               1\n",
       "Name: Title, dtype: int64"
      ]
     },
     "execution_count": 73,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "name.iloc[:,1].value_counts()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Besides the most common titles Mr, Miss and Mrs, there seem to be more rarely used titles (e.g. Dr, Don or Lady)."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 74,
   "metadata": {
    "scrolled": true
   },
   "outputs": [],
   "source": [
    "name[\"Title\"] = name[\"Title\"].str.strip()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 75,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Mr        757\n",
       "Miss      264\n",
       "Mrs       198\n",
       "Master     61\n",
       "Rare       29\n",
       "Name: Title, dtype: int64"
      ]
     },
     "execution_count": 75,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "name[\"Title\"] = name[\"Title\"].replace(['Lady', 'the Countess','Capt', 'Col','Don', 'Dr', 'Major', 'Rev', 'Sir', 'Jonkheer', 'Dona'], 'Rare')\n",
    "name[\"Title\"].replace(['Mlle','Ms'], 'Miss',inplace=True)\n",
    "name[\"Title\"].replace('Mme', 'Mrs',inplace=True)\n",
    "name[\"Title\"].value_counts()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 76,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 720x432 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.figure(figsize=(10,6))\n",
    "sns.barplot(x=name.iloc[:,1].value_counts().index,y=name.iloc[:,1].value_counts()).set_title('Title Feature Value Counts After Grouping', size=20, y=1.05)\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 77,
   "metadata": {
    "scrolled": true
   },
   "outputs": [],
   "source": [
    "df = pd.concat([df,name[\"Title\"]],axis=1, ignore_index=False)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 78,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Mr        757\n",
       "Miss      264\n",
       "Mrs       198\n",
       "Master     61\n",
       "Rare       29\n",
       "Name: Title, dtype: int64"
      ]
     },
     "execution_count": 78,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df['Title'].value_counts()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 79,
   "metadata": {
    "scrolled": false
   },
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 720x432 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.figure(figsize=(10,6))\n",
    "sns.countplot(x=df['Title'],\n",
    "             hue=df[\"Survived\"],\n",
    "             order = df['Title'].value_counts().index).set_title(\n",
    "                    'Title Feature Value Counts After Grouping', \n",
    "                    size=20, \n",
    "                    y=1.05)\n",
    "plt.legend(loc='upper right', labels=['Not Survived', 'Survived'])\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### FamilySize"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "From the two columns SibSp (Number of Siblings or Spouses on board) and Parch (number of Parents or Children on board) we can create a new variable representing the group size."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 80,
   "metadata": {},
   "outputs": [],
   "source": [
    "df[\"FamilySize\"] = df[\"SibSp\"]+df[\"Parch\"]+1"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 81,
   "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>FamilySize</th>\n",
       "      <th>Survived</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>4</td>\n",
       "      <td>0.724138</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>3</td>\n",
       "      <td>0.578431</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>2</td>\n",
       "      <td>0.552795</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>7</td>\n",
       "      <td>0.333333</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>1</td>\n",
       "      <td>0.303538</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>5</td>\n",
       "      <td>0.200000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>6</td>\n",
       "      <td>0.136364</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>8</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8</th>\n",
       "      <td>11</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   FamilySize  Survived\n",
       "3           4  0.724138\n",
       "2           3  0.578431\n",
       "1           2  0.552795\n",
       "6           7  0.333333\n",
       "0           1  0.303538\n",
       "4           5  0.200000\n",
       "5           6  0.136364\n",
       "7           8  0.000000\n",
       "8          11  0.000000"
      ]
     },
     "execution_count": 81,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df[['FamilySize', 'Survived']].groupby(['FamilySize'], as_index=False).mean().sort_values(by='Survived', ascending=False)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Furthermore, Group Sizes are further grouped:\n",
    "\n",
    "* Group Size with 1 are labeled as Alone\n",
    "* Group Size with 2, 3 and 4 are labeled as Small\n",
    "* Group Size with 5 and 6 are labeled as Medium\n",
    "* Group Size with 7, 8 and 11 are labeled as Large\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 82,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1440x1440 with 4 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "fig, axs = plt.subplots(figsize=(20, 20), ncols=2, nrows=2)\n",
    "plt.subplots_adjust(right=1.5)\n",
    "\n",
    "sns.barplot(x=df['FamilySize'].value_counts().index, y=df['FamilySize'].value_counts().values, ax=axs[0][0])\n",
    "sns.countplot(x='FamilySize', hue='Survived', data=df, ax=axs[0][1])\n",
    "\n",
    "axs[0][0].set_title('Family Size Feature Value Counts', size=20, y=1.05)\n",
    "axs[0][1].set_title('Survival Counts in Family Size ', size=20, y=1.05)\n",
    "\n",
    "family_map = {1: 'Alone', 2: 'Small', 3: 'Small', 4: 'Small', 5: 'Medium', 6: 'Medium', 7: 'Large', 8: 'Large', 11: 'Large'}\n",
    "df['FamilySizeGrouped'] = df['FamilySize'].map(family_map)\n",
    "\n",
    "sns.barplot(x=df['FamilySizeGrouped'].value_counts().index, y=df['FamilySizeGrouped'].value_counts().values, ax=axs[1][0])\n",
    "sns.countplot(x='FamilySizeGrouped', hue='Survived', data=df, ax=axs[1][1])\n",
    "\n",
    "axs[1][0].set_title('Family Size Feature Value Counts After Grouping', size=20, y=1.05)\n",
    "axs[1][1].set_title('Survival Counts in Family Size After Grouping', size=20, y=1.05)\n",
    "\n",
    "for i in range(2):\n",
    "    axs[i][1].legend(['Not Survived', 'Survived'], loc='upper right', prop={'size': 20})\n",
    "    for j in range(2):\n",
    "        axs[i][j].tick_params(axis='x', labelsize=20)\n",
    "        axs[i][j].tick_params(axis='y', labelsize=20)\n",
    "        axs[i][j].set_xlabel('')\n",
    "        axs[i][j].set_ylabel('')\n",
    "\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### TicketFrequency (i.e. GroupSize)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Grouping tickets by their frequencies, reveals that many passengers travelled along with groups consisting of friends, nannies, maids, etc. They were not count as family, but used the same ticket."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 83,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0      1\n",
       "1      2\n",
       "2      1\n",
       "3      2\n",
       "4      1\n",
       "      ..\n",
       "413    1\n",
       "414    3\n",
       "415    1\n",
       "416    1\n",
       "417    3\n",
       "Name: Ticket, Length: 1309, dtype: int64"
      ]
     },
     "execution_count": 83,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.groupby('Ticket')['Ticket'].transform('count')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 84,
   "metadata": {},
   "outputs": [],
   "source": [
    "df['TicketFrequency'] = df.groupby('Ticket')['Ticket'].transform('count')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 85,
   "metadata": {
    "scrolled": false
   },
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 720x432 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.figure(figsize=(10,6))\n",
    "sns.countplot(x='TicketFrequency', hue='Survived', data=df)\n",
    "\n",
    "plt.xlabel('Ticket Frequency', size=15, labelpad=20)\n",
    "plt.ylabel('Passenger Count', size=15, labelpad=20)\n",
    "plt.tick_params(axis='x', labelsize=15)\n",
    "plt.tick_params(axis='y', labelsize=15)\n",
    "\n",
    "plt.legend(['Not Survived', 'Survived'], loc='upper right', prop={'size': 15})\n",
    "plt.title('Count of Survival in {} Feature'.format('Ticket Frequency'), size=15, y=1.05)\n",
    "\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "According to the graph , groups with 2,3 and 4 members had a higher survival rate. Passengers who travel alone has the lowest survival rate. After 4 group members, survival rate decreases drastically. \n",
    "\n",
    "This pattern is very similar to FamilySize feature but there are minor differences. TicketFrequency values are not grouped like FamilySize because that would basically create the same feature with perfect correlation. This kind of feature wouldn't provide any additional information gain."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### HasCabin"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "From the Cabin number we can find out whether the person travelling had a cabin or not. "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 86,
   "metadata": {},
   "outputs": [],
   "source": [
    "df[\"HasCabin\"]= \"\"\n",
    "df[\"HasCabin\"][df[\"Cabin\"].isnull()==True]=0\n",
    "df[\"HasCabin\"][df[\"Cabin\"].isnull()==False]=1"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 87,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 720x432 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.figure(figsize=(10,6))\n",
    "sns.countplot(x='HasCabin', hue='Survived', data=df)\n",
    "\n",
    "plt.xlabel('Has Cabin', size=15, labelpad=20)\n",
    "plt.ylabel('Passenger Count', size=15, labelpad=20)\n",
    "plt.tick_params(axis='x', labelsize=15)\n",
    "plt.tick_params(axis='y', labelsize=15)\n",
    "\n",
    "plt.legend(['Not Survived', 'Survived'], loc='upper right', prop={'size': 15})\n",
    "plt.title('Count of Survival in {} Feature'.format('Has Cabin'), size=15, y=1.05)\n",
    "\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### IsWomanOrBoy\n",
    "\n",
    "We have already taken a look at the different titles. However, we can take a closer look at the title \"Master\".   \n",
    "\n",
    "Taking a look at the origin of the title Master [(source)](https://en.wikipedia.org/wiki/Master_(form_of_address)):_Master was retained as a form of address only for boys who had not yet entered society. By the late 19th century, etiquette dictated that men be addressed as Mister, and boys as Master._ \n",
    "\n",
    "As commonly known, in a ship tragedy, woman or children are rescued first. So, we can use this to create a new variable calld IsWomanOrBoy"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 88,
   "metadata": {},
   "outputs": [],
   "source": [
    "df['IsWomanOrBoy'] = ((df.Title == 'Master') | (df.Sex == 'female'))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 89,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 720x432 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.figure(figsize=(10,6))\n",
    "sns.countplot(x='IsWomanOrBoy', hue='Survived', data=df)\n",
    "\n",
    "plt.xlabel('Is either a Woman or a Boy', size=15, labelpad=20)\n",
    "plt.ylabel('Passenger Count', size=15, labelpad=20)\n",
    "plt.tick_params(axis='x', labelsize=15)\n",
    "plt.tick_params(axis='y', labelsize=15)\n",
    "\n",
    "plt.legend(['Not Survived', 'Survived'], loc='upper right', prop={'size': 15})\n",
    "plt.title('Count of Survival in {} Feature'.format('IsWomanOrBoy'), size=15, y=1.05)\n",
    "\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Data Pre-Processing: Handling Missing Values <a class=\"anchor\" id=\"thirdand1-bullet\"></a>"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 90,
   "metadata": {
    "scrolled": false
   },
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 720x432 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.figure(figsize=(10,6))\n",
    "y = np.round(df.isna().sum().sort_values()/len(df)*100,1)\n",
    "df.isna().sum().sort_values().plot(kind=\"barh\")\n",
    "\n",
    "for index, value in enumerate(y):\n",
    "    plt.text(value*14, index - 0.2, '{:.1f}%'.format(float(value)))\n",
    "\n",
    "plt.title('Percentage of missing values for all variables')\n",
    "plt.xticks([])\n",
    "plt.xlim(0,1200)\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Another way of displaying missing values is the following:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 91,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Text(0.5, 1, 'Missing values heatmap')"
      ]
     },
     "execution_count": 91,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1008x432 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.figure(figsize=(14,6))\n",
    "sns.heatmap(df.isnull(), cbar = False).set_title(\"Missing values heatmap\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "We have missing values in four features:  \n",
    "1. age: 177 - adding up to ~20% so we could try filling the NAs  \n",
    "2. class: 687 - since almost 78% are missing data, we might want to drop this column. Also, we have already used it in the Feature Engineering Part where we created the variable HasCabin.  \n",
    "3. embarked: 2 - adding up to less than 1% so we should fill the NAs\n",
    "4. Fare: 1 - adding up to less than 1% so we should fill the NAs"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Embarked"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Let's start with the embarked column."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 92,
   "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",
       "      <th>Title</th>\n",
       "      <th>FamilySize</th>\n",
       "      <th>FamilySizeGrouped</th>\n",
       "      <th>TicketFrequency</th>\n",
       "      <th>HasCabin</th>\n",
       "      <th>IsWomanOrBoy</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>61</th>\n",
       "      <td>62</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1</td>\n",
       "      <td>Icard, Miss. Amelie</td>\n",
       "      <td>female</td>\n",
       "      <td>38.0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>113572</td>\n",
       "      <td>80.0</td>\n",
       "      <td>B28</td>\n",
       "      <td>NaN</td>\n",
       "      <td>Miss</td>\n",
       "      <td>1</td>\n",
       "      <td>Alone</td>\n",
       "      <td>2</td>\n",
       "      <td>1</td>\n",
       "      <td>True</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>829</th>\n",
       "      <td>830</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1</td>\n",
       "      <td>Stone, Mrs. George Nelson (Martha Evelyn)</td>\n",
       "      <td>female</td>\n",
       "      <td>62.0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>113572</td>\n",
       "      <td>80.0</td>\n",
       "      <td>B28</td>\n",
       "      <td>NaN</td>\n",
       "      <td>Mrs</td>\n",
       "      <td>1</td>\n",
       "      <td>Alone</td>\n",
       "      <td>2</td>\n",
       "      <td>1</td>\n",
       "      <td>True</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "     PassengerId  Survived  Pclass                                       Name  \\\n",
       "61            62       1.0       1                        Icard, Miss. Amelie   \n",
       "829          830       1.0       1  Stone, Mrs. George Nelson (Martha Evelyn)   \n",
       "\n",
       "        Sex   Age  SibSp  Parch  Ticket  Fare Cabin Embarked Title  \\\n",
       "61   female  38.0      0      0  113572  80.0   B28      NaN  Miss   \n",
       "829  female  62.0      0      0  113572  80.0   B28      NaN   Mrs   \n",
       "\n",
       "     FamilySize FamilySizeGrouped  TicketFrequency HasCabin  IsWomanOrBoy  \n",
       "61            1             Alone                2        1          True  \n",
       "829           1             Alone                2        1          True  "
      ]
     },
     "execution_count": 92,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df[df[\"Embarked\"].isnull()==True]"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "There are several way how to handle this problem. Either we could do some research, to see if there is any information about how to fill these two rows. Or we could use backward or forward fill. Finally, we could also use the most frequent place, where people embarked (mode)."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 93,
   "metadata": {},
   "outputs": [],
   "source": [
    "df[\"Embarked\"].fillna(df[\"Embarked\"].mode()[0], inplace = True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 94,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0"
      ]
     },
     "execution_count": 94,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.Embarked.isnull().sum()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Age"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "We saw, that we have 177 missing values in the age column. As a starting point, we could take a look at the mean age looking at different splits like Sex or Pclass (or even both)."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 95,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Sex\n",
       "female    28.687088\n",
       "male      30.585228\n",
       "Name: Age, dtype: float64"
      ]
     },
     "execution_count": 95,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.groupby([\"Sex\"])[\"Age\"].mean()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 96,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Pclass\n",
       "1    39.159930\n",
       "2    29.506705\n",
       "3    24.816367\n",
       "Name: Age, dtype: float64"
      ]
     },
     "execution_count": 96,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.groupby([\"Pclass\"])[\"Age\"].mean()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 97,
   "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></th>\n",
       "      <th>Age</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Sex</th>\n",
       "      <th>Pclass</th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th rowspan=\"3\" valign=\"top\">female</th>\n",
       "      <th>1</th>\n",
       "      <td>36.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>28.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>22.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th rowspan=\"3\" valign=\"top\">male</th>\n",
       "      <th>1</th>\n",
       "      <td>42.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>29.5</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>25.0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                Age\n",
       "Sex    Pclass      \n",
       "female 1       36.0\n",
       "       2       28.0\n",
       "       3       22.0\n",
       "male   1       42.0\n",
       "       2       29.5\n",
       "       3       25.0"
      ]
     },
     "execution_count": 97,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.groupby([\"Sex\",\"Pclass\"])[[\"Age\"]].median()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 98,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.legend.Legend at 0x2275d644548>"
      ]
     },
     "execution_count": 98,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 720x432 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "df.groupby([\"Sex\",\"Pclass\"])[[\"Age\"]].mean().unstack(0).plot(kind=\"barh\",figsize=(10,6))\n",
    "plt.legend(('Female', 'Male'))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 99,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 514.88x475.2 with 6 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "grid = sns.FacetGrid(df, row='Pclass', col='Sex', size=2.2, aspect=1.6)\n",
    "grid.map(plt.hist, 'Age', alpha=.5, bins=20)\n",
    "grid.add_legend()\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "We could use the latter to fill ages per sex and group."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 100,
   "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></th>\n",
       "      <th>Age</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Sex</th>\n",
       "      <th>Pclass</th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th rowspan=\"3\" valign=\"top\">female</th>\n",
       "      <th>1</th>\n",
       "      <td>37.037594</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>27.499223</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>22.185329</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th rowspan=\"3\" valign=\"top\">male</th>\n",
       "      <th>1</th>\n",
       "      <td>41.029272</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>30.815380</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>25.962264</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                     Age\n",
       "Sex    Pclass           \n",
       "female 1       37.037594\n",
       "       2       27.499223\n",
       "       3       22.185329\n",
       "male   1       41.029272\n",
       "       2       30.815380\n",
       "       3       25.962264"
      ]
     },
     "execution_count": 100,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "ages = df.groupby([\"Sex\",\"Pclass\"])[[\"Age\"]].mean()\n",
    "ages"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Let's use these group means for filling the NAs in the Age group."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 101,
   "metadata": {},
   "outputs": [],
   "source": [
    "df[\"Age\"] = df.groupby([\"Sex\",\"Pclass\"])[[\"Age\"]].transform(\n",
    "    lambda grp: grp.fillna(np.mean(grp))\n",
    ")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "We can furthermore, create age bands and determine their correlations with Survived"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 102,
   "metadata": {},
   "outputs": [],
   "source": [
    "def age_to_group(age):\n",
    "    if 0 < age < 12:\n",
    "        # children\n",
    "        return 0\n",
    "    elif 12 <= age < 18:\n",
    "        # teenie\n",
    "        return 1\n",
    "    elif 18 <= age < 50:\n",
    "        # adult\n",
    "        return 2\n",
    "    elif age >= 50:\n",
    "        # elderly people\n",
    "        return 3"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Let us now replace Age with ordinals based on these bands."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 103,
   "metadata": {},
   "outputs": [],
   "source": [
    "df['AgeGroup'] = df['Age'].apply(age_to_group)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Cabin"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "As mentioned before, due to the high number of missing values, we could simply delete the cabin variable. \n",
    "However, we can also fill them with \"NA\", for now."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 104,
   "metadata": {},
   "outputs": [],
   "source": [
    "df[\"Cabin\"].fillna(\"NA\", inplace=True)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Fare\n",
    "\n",
    "Next up is Fare. Let's take a look at the mean here."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 105,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "33.2954792813456"
      ]
     },
     "execution_count": 105,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df[\"Fare\"].mean()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 106,
   "metadata": {},
   "outputs": [],
   "source": [
    "df[\"Fare\"].fillna(np.mean(df['Fare']), inplace=True)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Taking the Fare price, we can create a Fare band, dividing fare into four different groups:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 107,
   "metadata": {},
   "outputs": [],
   "source": [
    "df['FareBand'] = pd.qcut(df['Fare'], 4)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 108,
   "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>FareBand</th>\n",
       "      <th>Survived</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>(-0.001, 7.896]</td>\n",
       "      <td>0.197309</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>(7.896, 14.454]</td>\n",
       "      <td>0.303571</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>(14.454, 31.275]</td>\n",
       "      <td>0.441048</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>(31.275, 512.329]</td>\n",
       "      <td>0.600000</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "            FareBand  Survived\n",
       "0    (-0.001, 7.896]  0.197309\n",
       "1    (7.896, 14.454]  0.303571\n",
       "2   (14.454, 31.275]  0.441048\n",
       "3  (31.275, 512.329]  0.600000"
      ]
     },
     "execution_count": 108,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df[['FareBand', 'Survived']].groupby(['FareBand'], as_index=False).mean().sort_values(by='FareBand', ascending=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 109,
   "metadata": {},
   "outputs": [],
   "source": [
    "df['Fare'][df['Fare'] <= 7.896] = 0\n",
    "df['Fare'][(df['Fare'] > 7.896) & (df['Fare'] <= 14.454)] = 1\n",
    "df['Fare'][(df['Fare'] > 14.454) & (df['Fare'] <=  31.275)]  = 2\n",
    "df['Fare'][df['Fare'] >  31.275] = 3\n",
    "df['Fare'] = df['Fare'].astype(int)\n",
    "\n",
    "df = df.drop(['FareBand'], axis=1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 110,
   "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>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",
       "      <th>Title</th>\n",
       "      <th>FamilySize</th>\n",
       "      <th>FamilySizeGrouped</th>\n",
       "      <th>TicketFrequency</th>\n",
       "      <th>HasCabin</th>\n",
       "      <th>IsWomanOrBoy</th>\n",
       "      <th>AgeGroup</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>1</td>\n",
       "      <td>0.0</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>0</td>\n",
       "      <td>NA</td>\n",
       "      <td>S</td>\n",
       "      <td>Mr</td>\n",
       "      <td>2</td>\n",
       "      <td>Small</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>False</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>2</td>\n",
       "      <td>1.0</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>3</td>\n",
       "      <td>C85</td>\n",
       "      <td>C</td>\n",
       "      <td>Mrs</td>\n",
       "      <td>2</td>\n",
       "      <td>Small</td>\n",
       "      <td>2</td>\n",
       "      <td>1</td>\n",
       "      <td>True</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>3</td>\n",
       "      <td>1.0</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>1</td>\n",
       "      <td>NA</td>\n",
       "      <td>S</td>\n",
       "      <td>Miss</td>\n",
       "      <td>1</td>\n",
       "      <td>Alone</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>True</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>4</td>\n",
       "      <td>1.0</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>3</td>\n",
       "      <td>C123</td>\n",
       "      <td>S</td>\n",
       "      <td>Mrs</td>\n",
       "      <td>2</td>\n",
       "      <td>Small</td>\n",
       "      <td>2</td>\n",
       "      <td>1</td>\n",
       "      <td>True</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>5</td>\n",
       "      <td>0.0</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>1</td>\n",
       "      <td>NA</td>\n",
       "      <td>S</td>\n",
       "      <td>Mr</td>\n",
       "      <td>1</td>\n",
       "      <td>Alone</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>False</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   PassengerId  Survived  Pclass  \\\n",
       "0            1       0.0       3   \n",
       "1            2       1.0       1   \n",
       "2            3       1.0       3   \n",
       "3            4       1.0       1   \n",
       "4            5       0.0       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 Title  FamilySize  \\\n",
       "0      0         A/5 21171     0    NA        S    Mr           2   \n",
       "1      0          PC 17599     3   C85        C   Mrs           2   \n",
       "2      0  STON/O2. 3101282     1    NA        S  Miss           1   \n",
       "3      0            113803     3  C123        S   Mrs           2   \n",
       "4      0            373450     1    NA        S    Mr           1   \n",
       "\n",
       "  FamilySizeGrouped  TicketFrequency HasCabin  IsWomanOrBoy  AgeGroup  \n",
       "0             Small                1        0         False         2  \n",
       "1             Small                2        1          True         2  \n",
       "2             Alone                1        0          True         2  \n",
       "3             Small                2        1          True         2  \n",
       "4             Alone                1        0         False         2  "
      ]
     },
     "execution_count": 110,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Data Pre Processing: Transforming the variables into numericals <a class=\"anchor\" id=\"fourth-bullet\"></a>"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Next, we need to change the type of the data into numericals for modelling."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 111,
   "metadata": {},
   "outputs": [],
   "source": [
    "df['HasCabin'] = df['HasCabin'].astype('int')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 112,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Mapping the sex\n",
    "df['Sex'] = df['Sex'].map( {'female': 0, 'male': 1} ).astype(int)\n",
    "    \n",
    "# Mapping the title\n",
    "title_mapping = {\"Mr\": 1, \"Mrs\": 2, \"Miss\": 3, \"Master\": 4, \"Rare\": 5}\n",
    "df['Title'] = df['Title'].map(title_mapping)\n",
    "\n",
    "df['IsWomanOrBoy'] = df['IsWomanOrBoy'].map( {True: 1, False: 0} )"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Otherwise, we can also use the label encoder. \n",
    "\n",
    "LabelEncoder basically labels the classes from 0 to n. This process is necessary for models to learn from those features."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 113,
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn.preprocessing import LabelEncoder, OneHotEncoder\n",
    "le = LabelEncoder()\n",
    "df['Embarked'] = le.fit_transform(df.Embarked)\n",
    "#df['FamilySizeGrouped'] =le.fit_transform(df.FamilySizeGrouped)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "For those categegories, which are ordinal variables **label encoding** works quite well as you can rank / order the outcomes of a feature (e.g. group of Age or Fare). For tree-based models (like decision trees and random forests), you can expect label encoding to work well with ordinal variables.\n",
    "\n",
    "For most of our categorical variables (i.e. nominal variables), we cannot set a logical order (e.g. woman not better than men) therefore we use **One-Hot-Encoding**. One-hot encoding generally does not perform well if the categorical variable takes on a large number of values (i.e., generally do not use it for variables taking more than 15 different values)."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Therefore, we should convert the categorical features Pclass, Sex, Embarked, \tTitle, GroupSizeGrouped, HasCabin, TravelsAlone and IsWomanOrBoy"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 114,
   "metadata": {},
   "outputs": [],
   "source": [
    "cat_features = ['Pclass', 'Sex', 'Embarked', 'Title', 'FamilySizeGrouped', 'HasCabin', 'IsWomanOrBoy']\n",
    "encoded_features = []"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 115,
   "metadata": {},
   "outputs": [],
   "source": [
    "for feature in cat_features:\n",
    "        encoded_feat = OneHotEncoder().fit_transform(df[feature].values.reshape(-1, 1)).toarray()\n",
    "        n = df[feature].nunique()\n",
    "        cols = ['{}_{}'.format(feature, n) for n in range(1, n + 1)]\n",
    "        encoded_df = pd.DataFrame(encoded_feat, columns=cols)\n",
    "        encoded_df.index = df.index\n",
    "        encoded_features.append(encoded_df)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 119,
   "metadata": {},
   "outputs": [],
   "source": [
    "df_final = df.copy()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 120,
   "metadata": {},
   "outputs": [],
   "source": [
    "df_final = pd.concat([df_final, *encoded_features], axis=1)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Data Pre Processing: Dropping unnecessary features <a class=\"anchor\" id=\"fourthand1-bullet\"></a>"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Before we start with the modeling, we will set PassengerId as index and drop all unnecessary object variables.  \n",
    "Let us also drop Parch and SibSp features in favor of TravelsAlone and GroupSize, as well as the OneHotEncoded features"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 121,
   "metadata": {},
   "outputs": [],
   "source": [
    "df_final.drop(cat_features,axis=1, inplace=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 122,
   "metadata": {},
   "outputs": [],
   "source": [
    "df_final.set_index(\"PassengerId\", inplace=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 123,
   "metadata": {
    "scrolled": true
   },
   "outputs": [],
   "source": [
    "df_final.drop(['Name','Ticket','Cabin','SibSp','Parch', 'Age', 'FamilySize'],axis=1, inplace=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 124,
   "metadata": {},
   "outputs": [],
   "source": [
    "df_final.rename(columns={'IsWomanOrBoy_2':'IsWomanOrBoy_True',\n",
    "                         'IsWomanOrBoy_1':'IsWomanOrBoy_False',\n",
    "                         'HasCabin_1': 'HasCabin_False',\n",
    "                         'HasCabin_2':'HasCabin_True',\n",
    "                         'FamilySizeGrouped_1':'FamilySizeGrouped_Alone',\n",
    "                         'FamilySizeGrouped_4':'FamilySizeGrouped_Small',\n",
    "                         'FamilySizeGrouped_3':'FamilySizeGrouped_Medium',\n",
    "                         'FamilySizeGrouped_2':'FamilySizeGrouped_Large'\n",
    "                        }, inplace=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 131,
   "metadata": {
    "scrolled": false
   },
   "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>Survived</th>\n",
       "      <th>Fare</th>\n",
       "      <th>TicketFrequency</th>\n",
       "      <th>AgeGroup</th>\n",
       "      <th>Pclass_1</th>\n",
       "      <th>Pclass_2</th>\n",
       "      <th>Pclass_3</th>\n",
       "      <th>Sex_1</th>\n",
       "      <th>Sex_2</th>\n",
       "      <th>Embarked_1</th>\n",
       "      <th>Embarked_2</th>\n",
       "      <th>Embarked_3</th>\n",
       "      <th>Title_1</th>\n",
       "      <th>Title_2</th>\n",
       "      <th>Title_3</th>\n",
       "      <th>Title_4</th>\n",
       "      <th>Title_5</th>\n",
       "      <th>FamilySizeGrouped_Alone</th>\n",
       "      <th>FamilySizeGrouped_Large</th>\n",
       "      <th>FamilySizeGrouped_Medium</th>\n",
       "      <th>FamilySizeGrouped_Small</th>\n",
       "      <th>HasCabin_False</th>\n",
       "      <th>HasCabin_True</th>\n",
       "      <th>IsWomanOrBoy_False</th>\n",
       "      <th>IsWomanOrBoy_True</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>PassengerId</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>0.0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>2</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>1.0</td>\n",
       "      <td>3</td>\n",
       "      <td>2</td>\n",
       "      <td>2</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>1.0</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>2</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>1.0</td>\n",
       "      <td>3</td>\n",
       "      <td>2</td>\n",
       "      <td>2</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>0.0</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>2</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "             Survived  Fare  TicketFrequency  AgeGroup  Pclass_1  Pclass_2  \\\n",
       "PassengerId                                                                  \n",
       "1                 0.0     0                1         2       0.0       0.0   \n",
       "2                 1.0     3                2         2       1.0       0.0   \n",
       "3                 1.0     1                1         2       0.0       0.0   \n",
       "4                 1.0     3                2         2       1.0       0.0   \n",
       "5                 0.0     1                1         2       0.0       0.0   \n",
       "\n",
       "             Pclass_3  Sex_1  Sex_2  Embarked_1  Embarked_2  Embarked_3  \\\n",
       "PassengerId                                                               \n",
       "1                 1.0    0.0    1.0         0.0         0.0         1.0   \n",
       "2                 0.0    1.0    0.0         1.0         0.0         0.0   \n",
       "3                 1.0    1.0    0.0         0.0         0.0         1.0   \n",
       "4                 0.0    1.0    0.0         0.0         0.0         1.0   \n",
       "5                 1.0    0.0    1.0         0.0         0.0         1.0   \n",
       "\n",
       "             Title_1  Title_2  Title_3  Title_4  Title_5  \\\n",
       "PassengerId                                                \n",
       "1                1.0      0.0      0.0      0.0      0.0   \n",
       "2                0.0      1.0      0.0      0.0      0.0   \n",
       "3                0.0      0.0      1.0      0.0      0.0   \n",
       "4                0.0      1.0      0.0      0.0      0.0   \n",
       "5                1.0      0.0      0.0      0.0      0.0   \n",
       "\n",
       "             FamilySizeGrouped_Alone  FamilySizeGrouped_Large  \\\n",
       "PassengerId                                                     \n",
       "1                                0.0                      0.0   \n",
       "2                                0.0                      0.0   \n",
       "3                                1.0                      0.0   \n",
       "4                                0.0                      0.0   \n",
       "5                                1.0                      0.0   \n",
       "\n",
       "             FamilySizeGrouped_Medium  FamilySizeGrouped_Small  \\\n",
       "PassengerId                                                      \n",
       "1                                 0.0                      1.0   \n",
       "2                                 0.0                      1.0   \n",
       "3                                 0.0                      0.0   \n",
       "4                                 0.0                      1.0   \n",
       "5                                 0.0                      0.0   \n",
       "\n",
       "             HasCabin_False  HasCabin_True  IsWomanOrBoy_False  \\\n",
       "PassengerId                                                      \n",
       "1                       1.0            0.0                 1.0   \n",
       "2                       0.0            1.0                 0.0   \n",
       "3                       1.0            0.0                 0.0   \n",
       "4                       0.0            1.0                 0.0   \n",
       "5                       1.0            0.0                 1.0   \n",
       "\n",
       "             IsWomanOrBoy_True  \n",
       "PassengerId                     \n",
       "1                          0.0  \n",
       "2                          1.0  \n",
       "3                          1.0  \n",
       "4                          1.0  \n",
       "5                          0.0  "
      ]
     },
     "execution_count": 131,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_final.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Exploratory Data Analysis <a class=\"anchor\" id=\"fifth-bullet\"></a>"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "So far, so good. Now that we have done some feature engineering and filled NAs let's take another look at the data."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 132,
   "metadata": {
    "scrolled": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "In total there are 1309 rows with 25 variables in the dataset.\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>Survived</th>\n",
       "      <th>Fare</th>\n",
       "      <th>TicketFrequency</th>\n",
       "      <th>AgeGroup</th>\n",
       "      <th>Pclass_1</th>\n",
       "      <th>Pclass_2</th>\n",
       "      <th>Pclass_3</th>\n",
       "      <th>Sex_1</th>\n",
       "      <th>Sex_2</th>\n",
       "      <th>Embarked_1</th>\n",
       "      <th>Embarked_2</th>\n",
       "      <th>Embarked_3</th>\n",
       "      <th>Title_1</th>\n",
       "      <th>Title_2</th>\n",
       "      <th>Title_3</th>\n",
       "      <th>Title_4</th>\n",
       "      <th>Title_5</th>\n",
       "      <th>FamilySizeGrouped_Alone</th>\n",
       "      <th>FamilySizeGrouped_Large</th>\n",
       "      <th>FamilySizeGrouped_Medium</th>\n",
       "      <th>FamilySizeGrouped_Small</th>\n",
       "      <th>HasCabin_False</th>\n",
       "      <th>HasCabin_True</th>\n",
       "      <th>IsWomanOrBoy_False</th>\n",
       "      <th>IsWomanOrBoy_True</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>Summe</th>\n",
       "      <td>342</td>\n",
       "      <td>1958</td>\n",
       "      <td>2751</td>\n",
       "      <td>2483</td>\n",
       "      <td>323</td>\n",
       "      <td>277</td>\n",
       "      <td>709</td>\n",
       "      <td>466</td>\n",
       "      <td>843</td>\n",
       "      <td>270</td>\n",
       "      <td>123</td>\n",
       "      <td>916</td>\n",
       "      <td>757</td>\n",
       "      <td>198</td>\n",
       "      <td>264</td>\n",
       "      <td>61</td>\n",
       "      <td>29</td>\n",
       "      <td>790</td>\n",
       "      <td>35</td>\n",
       "      <td>47</td>\n",
       "      <td>437</td>\n",
       "      <td>1014</td>\n",
       "      <td>295</td>\n",
       "      <td>782</td>\n",
       "      <td>527</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Minimum</th>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Maximum</th>\n",
       "      <td>1</td>\n",
       "      <td>3</td>\n",
       "      <td>11</td>\n",
       "      <td>3</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Ausreißer</th>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>62</td>\n",
       "      <td>264</td>\n",
       "      <td>323</td>\n",
       "      <td>277</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>270</td>\n",
       "      <td>123</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>198</td>\n",
       "      <td>264</td>\n",
       "      <td>61</td>\n",
       "      <td>29</td>\n",
       "      <td>0</td>\n",
       "      <td>35</td>\n",
       "      <td>47</td>\n",
       "      <td>0</td>\n",
       "      <td>295</td>\n",
       "      <td>295</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Mittelwert</th>\n",
       "      <td>0.383838</td>\n",
       "      <td>1.4958</td>\n",
       "      <td>2.1016</td>\n",
       "      <td>1.89687</td>\n",
       "      <td>0.246753</td>\n",
       "      <td>0.211612</td>\n",
       "      <td>0.541635</td>\n",
       "      <td>0.355997</td>\n",
       "      <td>0.644003</td>\n",
       "      <td>0.206264</td>\n",
       "      <td>0.0939649</td>\n",
       "      <td>0.699771</td>\n",
       "      <td>0.578304</td>\n",
       "      <td>0.151261</td>\n",
       "      <td>0.201681</td>\n",
       "      <td>0.0466005</td>\n",
       "      <td>0.0221543</td>\n",
       "      <td>0.603514</td>\n",
       "      <td>0.026738</td>\n",
       "      <td>0.0359053</td>\n",
       "      <td>0.333843</td>\n",
       "      <td>0.774637</td>\n",
       "      <td>0.225363</td>\n",
       "      <td>0.597403</td>\n",
       "      <td>0.402597</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Stand. Abw.</th>\n",
       "      <td>0.486592</td>\n",
       "      <td>1.12289</td>\n",
       "      <td>1.77983</td>\n",
       "      <td>0.632381</td>\n",
       "      <td>0.431287</td>\n",
       "      <td>0.408607</td>\n",
       "      <td>0.498454</td>\n",
       "      <td>0.478997</td>\n",
       "      <td>0.478997</td>\n",
       "      <td>0.404777</td>\n",
       "      <td>0.291891</td>\n",
       "      <td>0.458533</td>\n",
       "      <td>0.494019</td>\n",
       "      <td>0.35844</td>\n",
       "      <td>0.401408</td>\n",
       "      <td>0.210862</td>\n",
       "      <td>0.147242</td>\n",
       "      <td>0.489354</td>\n",
       "      <td>0.161378</td>\n",
       "      <td>0.186125</td>\n",
       "      <td>0.471765</td>\n",
       "      <td>0.417981</td>\n",
       "      <td>0.417981</td>\n",
       "      <td>0.490608</td>\n",
       "      <td>0.490608</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Median</th>\n",
       "      <td>0</td>\n",
       "      <td>2</td>\n",
       "      <td>1</td>\n",
       "      <td>2</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Anzahl</th>\n",
       "      <td>891</td>\n",
       "      <td>1309</td>\n",
       "      <td>1309</td>\n",
       "      <td>1309</td>\n",
       "      <td>1309</td>\n",
       "      <td>1309</td>\n",
       "      <td>1309</td>\n",
       "      <td>1309</td>\n",
       "      <td>1309</td>\n",
       "      <td>1309</td>\n",
       "      <td>1309</td>\n",
       "      <td>1309</td>\n",
       "      <td>1309</td>\n",
       "      <td>1309</td>\n",
       "      <td>1309</td>\n",
       "      <td>1309</td>\n",
       "      <td>1309</td>\n",
       "      <td>1309</td>\n",
       "      <td>1309</td>\n",
       "      <td>1309</td>\n",
       "      <td>1309</td>\n",
       "      <td>1309</td>\n",
       "      <td>1309</td>\n",
       "      <td>1309</td>\n",
       "      <td>1309</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>#NA</th>\n",
       "      <td>418</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>#Uniques</th>\n",
       "      <td>2</td>\n",
       "      <td>4</td>\n",
       "      <td>9</td>\n",
       "      <td>4</td>\n",
       "      <td>2</td>\n",
       "      <td>2</td>\n",
       "      <td>2</td>\n",
       "      <td>2</td>\n",
       "      <td>2</td>\n",
       "      <td>2</td>\n",
       "      <td>2</td>\n",
       "      <td>2</td>\n",
       "      <td>2</td>\n",
       "      <td>2</td>\n",
       "      <td>2</td>\n",
       "      <td>2</td>\n",
       "      <td>2</td>\n",
       "      <td>2</td>\n",
       "      <td>2</td>\n",
       "      <td>2</td>\n",
       "      <td>2</td>\n",
       "      <td>2</td>\n",
       "      <td>2</td>\n",
       "      <td>2</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>dtypes</th>\n",
       "      <td>float64</td>\n",
       "      <td>int32</td>\n",
       "      <td>int64</td>\n",
       "      <td>int64</td>\n",
       "      <td>float64</td>\n",
       "      <td>float64</td>\n",
       "      <td>float64</td>\n",
       "      <td>float64</td>\n",
       "      <td>float64</td>\n",
       "      <td>float64</td>\n",
       "      <td>float64</td>\n",
       "      <td>float64</td>\n",
       "      <td>float64</td>\n",
       "      <td>float64</td>\n",
       "      <td>float64</td>\n",
       "      <td>float64</td>\n",
       "      <td>float64</td>\n",
       "      <td>float64</td>\n",
       "      <td>float64</td>\n",
       "      <td>float64</td>\n",
       "      <td>float64</td>\n",
       "      <td>float64</td>\n",
       "      <td>float64</td>\n",
       "      <td>float64</td>\n",
       "      <td>float64</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "             Survived     Fare TicketFrequency  AgeGroup  Pclass_1  Pclass_2  \\\n",
       "Summe             342     1958            2751      2483       323       277   \n",
       "Minimum             0        0               1         0         0         0   \n",
       "Maximum             1        3              11         3         1         1   \n",
       "Ausreißer           0        0              62       264       323       277   \n",
       "Mittelwert   0.383838   1.4958          2.1016   1.89687  0.246753  0.211612   \n",
       "Stand. Abw.  0.486592  1.12289         1.77983  0.632381  0.431287  0.408607   \n",
       "Median              0        2               1         2         0         0   \n",
       "Anzahl            891     1309            1309      1309      1309      1309   \n",
       "#NA               418        0               0         0         0         0   \n",
       "#Uniques            2        4               9         4         2         2   \n",
       "dtypes        float64    int32           int64     int64   float64   float64   \n",
       "\n",
       "             Pclass_3     Sex_1     Sex_2 Embarked_1 Embarked_2 Embarked_3  \\\n",
       "Summe             709       466       843        270        123        916   \n",
       "Minimum             0         0         0          0          0          0   \n",
       "Maximum             1         1         1          1          1          1   \n",
       "Ausreißer           0         0         0        270        123          0   \n",
       "Mittelwert   0.541635  0.355997  0.644003   0.206264  0.0939649   0.699771   \n",
       "Stand. Abw.  0.498454  0.478997  0.478997   0.404777   0.291891   0.458533   \n",
       "Median              1         0         1          0          0          1   \n",
       "Anzahl           1309      1309      1309       1309       1309       1309   \n",
       "#NA                 0         0         0          0          0          0   \n",
       "#Uniques            2         2         2          2          2          2   \n",
       "dtypes        float64   float64   float64    float64    float64    float64   \n",
       "\n",
       "              Title_1   Title_2   Title_3    Title_4    Title_5  \\\n",
       "Summe             757       198       264         61         29   \n",
       "Minimum             0         0         0          0          0   \n",
       "Maximum             1         1         1          1          1   \n",
       "Ausreißer           0       198       264         61         29   \n",
       "Mittelwert   0.578304  0.151261  0.201681  0.0466005  0.0221543   \n",
       "Stand. Abw.  0.494019   0.35844  0.401408   0.210862   0.147242   \n",
       "Median              1         0         0          0          0   \n",
       "Anzahl           1309      1309      1309       1309       1309   \n",
       "#NA                 0         0         0          0          0   \n",
       "#Uniques            2         2         2          2          2   \n",
       "dtypes        float64   float64   float64    float64    float64   \n",
       "\n",
       "            FamilySizeGrouped_Alone FamilySizeGrouped_Large  \\\n",
       "Summe                           790                      35   \n",
       "Minimum                           0                       0   \n",
       "Maximum                           1                       1   \n",
       "Ausreißer                         0                      35   \n",
       "Mittelwert                 0.603514                0.026738   \n",
       "Stand. Abw.                0.489354                0.161378   \n",
       "Median                            1                       0   \n",
       "Anzahl                         1309                    1309   \n",
       "#NA                               0                       0   \n",
       "#Uniques                          2                       2   \n",
       "dtypes                      float64                 float64   \n",
       "\n",
       "            FamilySizeGrouped_Medium FamilySizeGrouped_Small HasCabin_False  \\\n",
       "Summe                             47                     437           1014   \n",
       "Minimum                            0                       0              0   \n",
       "Maximum                            1                       1              1   \n",
       "Ausreißer                         47                       0            295   \n",
       "Mittelwert                 0.0359053                0.333843       0.774637   \n",
       "Stand. Abw.                 0.186125                0.471765       0.417981   \n",
       "Median                             0                       0              1   \n",
       "Anzahl                          1309                    1309           1309   \n",
       "#NA                                0                       0              0   \n",
       "#Uniques                           2                       2              2   \n",
       "dtypes                       float64                 float64        float64   \n",
       "\n",
       "            HasCabin_True IsWomanOrBoy_False IsWomanOrBoy_True  \n",
       "Summe                 295                782               527  \n",
       "Minimum                 0                  0                 0  \n",
       "Maximum                 1                  1                 1  \n",
       "Ausreißer             295                  0                 0  \n",
       "Mittelwert       0.225363           0.597403          0.402597  \n",
       "Stand. Abw.      0.417981           0.490608          0.490608  \n",
       "Median                  0                  1                 0  \n",
       "Anzahl               1309               1309              1309  \n",
       "#NA                     0                  0                 0  \n",
       "#Uniques                2                  2                 2  \n",
       "dtypes            float64            float64           float64  "
      ]
     },
     "execution_count": 132,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "my_df_summary(df_final)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "One way to look at relations in the data is to take a look at correlations."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 133,
   "metadata": {
    "scrolled": false
   },
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1296x720 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.figure(figsize=(18,10))\n",
    "corr = df.iloc[:,1:].corr()\n",
    "mask = np.zeros_like(corr, dtype=np.bool)\n",
    "mask[np.triu_indices_from(mask)] = True\n",
    "\n",
    "sns.heatmap(corr,mask=mask, vmin=-1,cmap='PiYG', annot=False);\n",
    "plt.yticks(rotation=0)\n",
    "plt.xticks(rotation=90)\n",
    "#Nur notwendig, weil Grafik bei einer anderen Version von matplotlib (>3.3) sonst verzerrt. Bekannter Bug.\n",
    "b, t = plt.ylim() # discover the values for bottom and top\n",
    "b += 0.5 # Add 0.5 to the bottom\n",
    "t -= 0.5 # Subtract 0.5 from the top\n",
    "plt.ylim(b, t) # update the ylim(bottom, top) values\n",
    "\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "From here we can see some variables which are highly negative correlated:\n",
    "1. HasCabin & Fare: People without a cabin pay a lower Fare than with a cabin"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Besides, we have some positive correlations:\n",
    "1. HasCabin & Pclass_1: People with a Cabin are usually travelling in Class 1\n",
    "2. FamilySizeGrouped and Fare: The higher the Grouped Family Size (i.e. the larger the group) the higher the Fare\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "However, we have to ignore correlations between two object variables, e.g.:\n",
    "* Title and Sex: There is no meaning of the correlation and obviously a woman with a title \"Mrs\" is in the womanorboy group"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "More interestingly, when looking at the correlation between survived and other features:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 134,
   "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>Fare</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>Fare</th>\n",
       "      <td>1.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Pclass_1</th>\n",
       "      <td>0.607560</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>TicketFrequency</th>\n",
       "      <td>0.591047</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>HasCabin_True</th>\n",
       "      <td>0.506170</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>FamilySizeGrouped_Small</th>\n",
       "      <td>0.449325</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>IsWomanOrBoy_True</th>\n",
       "      <td>0.282711</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Title_2</th>\n",
       "      <td>0.273209</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Sex_1</th>\n",
       "      <td>0.231632</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Embarked_1</th>\n",
       "      <td>0.217211</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>FamilySizeGrouped_Large</th>\n",
       "      <td>0.171490</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>FamilySizeGrouped_Medium</th>\n",
       "      <td>0.167164</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Title_4</th>\n",
       "      <td>0.131599</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Title_5</th>\n",
       "      <td>0.081485</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Pclass_2</th>\n",
       "      <td>0.064425</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Title_3</th>\n",
       "      <td>0.023932</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Embarked_3</th>\n",
       "      <td>-0.012103</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>AgeGroup</th>\n",
       "      <td>-0.048522</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Sex_2</th>\n",
       "      <td>-0.231632</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Embarked_2</th>\n",
       "      <td>-0.282202</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>IsWomanOrBoy_False</th>\n",
       "      <td>-0.282711</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Title_1</th>\n",
       "      <td>-0.298131</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>HasCabin_False</th>\n",
       "      <td>-0.506170</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>FamilySizeGrouped_Alone</th>\n",
       "      <td>-0.553308</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Pclass_3</th>\n",
       "      <td>-0.578503</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                              Fare\n",
       "Fare                      1.000000\n",
       "Pclass_1                  0.607560\n",
       "TicketFrequency           0.591047\n",
       "HasCabin_True             0.506170\n",
       "FamilySizeGrouped_Small   0.449325\n",
       "IsWomanOrBoy_True         0.282711\n",
       "Title_2                   0.273209\n",
       "Sex_1                     0.231632\n",
       "Embarked_1                0.217211\n",
       "FamilySizeGrouped_Large   0.171490\n",
       "FamilySizeGrouped_Medium  0.167164\n",
       "Title_4                   0.131599\n",
       "Title_5                   0.081485\n",
       "Pclass_2                  0.064425\n",
       "Title_3                   0.023932\n",
       "Embarked_3               -0.012103\n",
       "AgeGroup                 -0.048522\n",
       "Sex_2                    -0.231632\n",
       "Embarked_2               -0.282202\n",
       "IsWomanOrBoy_False       -0.282711\n",
       "Title_1                  -0.298131\n",
       "HasCabin_False           -0.506170\n",
       "FamilySizeGrouped_Alone  -0.553308\n",
       "Pclass_3                 -0.578503"
      ]
     },
     "execution_count": 134,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "pd.DataFrame(df_final.iloc[:,1:].corr().iloc[0].sort_values(ascending=False))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "1. Survived & Fare: The higher the Fare, the higher the chance of survival\n",
    "2. Survived & Pclass: The lower the class (1st class is lower than 2nd) the betther the chance of survival\n",
    "3. Survived & HasCabin: People having a Cabin seem to have a higher probability of surviving\n",
    "4. Survived & FamilySizeGrouped_Small: A small family group has a higher chance of survival than when traveling alone"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "By looking at different group means, we can further analyze our data:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 135,
   "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 tr th {\n",
       "        text-align: left;\n",
       "    }\n",
       "\n",
       "    .dataframe thead tr:last-of-type th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr>\n",
       "      <th></th>\n",
       "      <th colspan=\"3\" halign=\"left\">Survived</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th></th>\n",
       "      <th>mean</th>\n",
       "      <th>count</th>\n",
       "      <th>sum</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Pclass</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>0.629630</td>\n",
       "      <td>216</td>\n",
       "      <td>136.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>0.472826</td>\n",
       "      <td>184</td>\n",
       "      <td>87.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>0.242363</td>\n",
       "      <td>491</td>\n",
       "      <td>119.0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "        Survived             \n",
       "            mean count    sum\n",
       "Pclass                       \n",
       "1       0.629630   216  136.0\n",
       "2       0.472826   184   87.0\n",
       "3       0.242363   491  119.0"
      ]
     },
     "execution_count": 135,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df[[\"Pclass\", 'Survived']].groupby([\"Pclass\"], as_index=False).agg(['mean', 'count', 'sum'])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 136,
   "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 tr th {\n",
       "        text-align: left;\n",
       "    }\n",
       "\n",
       "    .dataframe thead tr:last-of-type th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr>\n",
       "      <th></th>\n",
       "      <th colspan=\"3\" halign=\"left\">Survived</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th></th>\n",
       "      <th>mean</th>\n",
       "      <th>count</th>\n",
       "      <th>sum</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>FamilySizeGrouped</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>Alone</th>\n",
       "      <td>0.303538</td>\n",
       "      <td>537</td>\n",
       "      <td>163.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Large</th>\n",
       "      <td>0.160000</td>\n",
       "      <td>25</td>\n",
       "      <td>4.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Medium</th>\n",
       "      <td>0.162162</td>\n",
       "      <td>37</td>\n",
       "      <td>6.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Small</th>\n",
       "      <td>0.578767</td>\n",
       "      <td>292</td>\n",
       "      <td>169.0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                   Survived             \n",
       "                       mean count    sum\n",
       "FamilySizeGrouped                       \n",
       "Alone              0.303538   537  163.0\n",
       "Large              0.160000    25    4.0\n",
       "Medium             0.162162    37    6.0\n",
       "Small              0.578767   292  169.0"
      ]
     },
     "execution_count": 136,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df[[\"FamilySizeGrouped\", 'Survived']].groupby([\"FamilySizeGrouped\"], as_index=False).agg(['mean', 'count', 'sum'])"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Machine Learning Models <a class=\"anchor\" id=\"sixth-bullet\"></a>"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Starting with modeling, we need to divide our data into a training and a test dataset.  \n",
    "We will then use the training data for our model to learn and the test dataset to see which model performs best.\n",
    "  \n",
    "We will use a split of 80% (training) to 20% (test dataset)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 137,
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn.model_selection import train_test_split"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 138,
   "metadata": {},
   "outputs": [],
   "source": [
    "data_test = df_final[df_final[\"Survived\"].isna()==True]\n",
    "data_train = df_final[df_final[\"Survived\"].isna()==False]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 139,
   "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>Survived</th>\n",
       "      <th>Fare</th>\n",
       "      <th>TicketFrequency</th>\n",
       "      <th>AgeGroup</th>\n",
       "      <th>Pclass_1</th>\n",
       "      <th>Pclass_2</th>\n",
       "      <th>Pclass_3</th>\n",
       "      <th>Sex_1</th>\n",
       "      <th>Sex_2</th>\n",
       "      <th>Embarked_1</th>\n",
       "      <th>Embarked_2</th>\n",
       "      <th>Embarked_3</th>\n",
       "      <th>Title_1</th>\n",
       "      <th>Title_2</th>\n",
       "      <th>Title_3</th>\n",
       "      <th>Title_4</th>\n",
       "      <th>Title_5</th>\n",
       "      <th>FamilySizeGrouped_Alone</th>\n",
       "      <th>FamilySizeGrouped_Large</th>\n",
       "      <th>FamilySizeGrouped_Medium</th>\n",
       "      <th>FamilySizeGrouped_Small</th>\n",
       "      <th>HasCabin_False</th>\n",
       "      <th>HasCabin_True</th>\n",
       "      <th>IsWomanOrBoy_False</th>\n",
       "      <th>IsWomanOrBoy_True</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>PassengerId</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>0.0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>2</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>1.0</td>\n",
       "      <td>3</td>\n",
       "      <td>2</td>\n",
       "      <td>2</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>1.0</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>2</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>1.0</td>\n",
       "      <td>3</td>\n",
       "      <td>2</td>\n",
       "      <td>2</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>0.0</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>2</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "             Survived  Fare  TicketFrequency  AgeGroup  Pclass_1  Pclass_2  \\\n",
       "PassengerId                                                                  \n",
       "1                 0.0     0                1         2       0.0       0.0   \n",
       "2                 1.0     3                2         2       1.0       0.0   \n",
       "3                 1.0     1                1         2       0.0       0.0   \n",
       "4                 1.0     3                2         2       1.0       0.0   \n",
       "5                 0.0     1                1         2       0.0       0.0   \n",
       "\n",
       "             Pclass_3  Sex_1  Sex_2  Embarked_1  Embarked_2  Embarked_3  \\\n",
       "PassengerId                                                               \n",
       "1                 1.0    0.0    1.0         0.0         0.0         1.0   \n",
       "2                 0.0    1.0    0.0         1.0         0.0         0.0   \n",
       "3                 1.0    1.0    0.0         0.0         0.0         1.0   \n",
       "4                 0.0    1.0    0.0         0.0         0.0         1.0   \n",
       "5                 1.0    0.0    1.0         0.0         0.0         1.0   \n",
       "\n",
       "             Title_1  Title_2  Title_3  Title_4  Title_5  \\\n",
       "PassengerId                                                \n",
       "1                1.0      0.0      0.0      0.0      0.0   \n",
       "2                0.0      1.0      0.0      0.0      0.0   \n",
       "3                0.0      0.0      1.0      0.0      0.0   \n",
       "4                0.0      1.0      0.0      0.0      0.0   \n",
       "5                1.0      0.0      0.0      0.0      0.0   \n",
       "\n",
       "             FamilySizeGrouped_Alone  FamilySizeGrouped_Large  \\\n",
       "PassengerId                                                     \n",
       "1                                0.0                      0.0   \n",
       "2                                0.0                      0.0   \n",
       "3                                1.0                      0.0   \n",
       "4                                0.0                      0.0   \n",
       "5                                1.0                      0.0   \n",
       "\n",
       "             FamilySizeGrouped_Medium  FamilySizeGrouped_Small  \\\n",
       "PassengerId                                                      \n",
       "1                                 0.0                      1.0   \n",
       "2                                 0.0                      1.0   \n",
       "3                                 0.0                      0.0   \n",
       "4                                 0.0                      1.0   \n",
       "5                                 0.0                      0.0   \n",
       "\n",
       "             HasCabin_False  HasCabin_True  IsWomanOrBoy_False  \\\n",
       "PassengerId                                                      \n",
       "1                       1.0            0.0                 1.0   \n",
       "2                       0.0            1.0                 0.0   \n",
       "3                       1.0            0.0                 0.0   \n",
       "4                       0.0            1.0                 0.0   \n",
       "5                       1.0            0.0                 1.0   \n",
       "\n",
       "             IsWomanOrBoy_True  \n",
       "PassengerId                     \n",
       "1                          0.0  \n",
       "2                          1.0  \n",
       "3                          1.0  \n",
       "4                          1.0  \n",
       "5                          0.0  "
      ]
     },
     "execution_count": 139,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data_train.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 140,
   "metadata": {},
   "outputs": [],
   "source": [
    "i_split = 0.20\n",
    "X_train, X_test, y_train, y_test = train_test_split(data_train.iloc[:,1:], data_train.iloc[:,0], test_size=i_split,\n",
    "                                                    random_state=1234)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Random Forest"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "We will now start with a random forest model:\n",
    "\n",
    "Furthermore we also import some metrics to see e.g. the accuracy of our model."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 141,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "[Parallel(n_jobs=-1)]: Using backend ThreadingBackend with 8 concurrent workers.\n",
      "[Parallel(n_jobs=-1)]: Done  34 tasks      | elapsed:    0.0s\n",
      "[Parallel(n_jobs=-1)]: Done 184 tasks      | elapsed:    0.5s\n",
      "[Parallel(n_jobs=-1)]: Done 434 tasks      | elapsed:    1.3s\n",
      "[Parallel(n_jobs=-1)]: Done 784 tasks      | elapsed:    2.4s\n",
      "[Parallel(n_jobs=-1)]: Done 1100 out of 1100 | elapsed:    3.3s finished\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "RandomForestClassifier(bootstrap=True, class_weight=None, criterion='gini',\n",
       "                       max_depth=5, max_features='auto', max_leaf_nodes=None,\n",
       "                       min_impurity_decrease=0.0, min_impurity_split=None,\n",
       "                       min_samples_leaf=5, min_samples_split=4,\n",
       "                       min_weight_fraction_leaf=0.0, n_estimators=1100,\n",
       "                       n_jobs=-1, oob_score=True, random_state=42, verbose=1,\n",
       "                       warm_start=False)"
      ]
     },
     "execution_count": 141,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "from sklearn.ensemble import RandomForestClassifier\n",
    "from sklearn.metrics import accuracy_score, f1_score, confusion_matrix, precision_score, recall_score, classification_report\n",
    "\n",
    "model = RandomForestClassifier(criterion='gini',\n",
    "                                           n_estimators=1100,\n",
    "                                           max_depth=5,\n",
    "                                           min_samples_split=4,\n",
    "                                           min_samples_leaf=5,\n",
    "                                           max_features='auto',\n",
    "                                           oob_score=True,\n",
    "                                           random_state=SEED,\n",
    "                                           n_jobs=-1,\n",
    "                                           verbose=1) \n",
    "model.fit(X_train,y_train)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 142,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "[Parallel(n_jobs=8)]: Using backend ThreadingBackend with 8 concurrent workers.\n",
      "[Parallel(n_jobs=8)]: Done  34 tasks      | elapsed:    0.0s\n",
      "[Parallel(n_jobs=8)]: Done 184 tasks      | elapsed:    0.0s\n",
      "[Parallel(n_jobs=8)]: Done 434 tasks      | elapsed:    0.1s\n",
      "[Parallel(n_jobs=8)]: Done 784 tasks      | elapsed:    0.3s\n",
      "[Parallel(n_jobs=8)]: Done 1100 out of 1100 | elapsed:    0.4s finished\n"
     ]
    }
   ],
   "source": [
    "predictions = model.predict(X_test)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 143,
   "metadata": {},
   "outputs": [],
   "source": [
    "df_conf_matrix = confusion_matrix(y_test,predictions)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 144,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "              precision    recall  f1-score   support\n",
      "\n",
      "         0.0       0.85      0.93      0.89       109\n",
      "         1.0       0.87      0.74      0.80        70\n",
      "\n",
      "    accuracy                           0.85       179\n",
      "   macro avg       0.86      0.83      0.84       179\n",
      "weighted avg       0.86      0.85      0.85       179\n",
      "\n"
     ]
    }
   ],
   "source": [
    "print(classification_report(y_test,predictions))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Looking at the classification report, we predict with an accuracy of 85%, which is quite good.\n",
    "We seem to have a higher precision with predicting the survivals (i.e. survived = 1) than the non-survivals (i.e. survived = 0). "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Further models & Model comparison"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "How do other models perform in comparison to Random Forest?"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 145,
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn.naive_bayes import MultinomialNB, BernoulliNB, GaussianNB\n",
    "from sklearn.neighbors import KNeighborsClassifier\n",
    "from sklearn.linear_model import LogisticRegression, SGDClassifier\n",
    "from sklearn.svm import LinearSVC, SVC\n",
    "from sklearn.neural_network import MLPClassifier\n",
    "from sklearn.tree import DecisionTreeClassifier\n",
    "from sklearn.ensemble import AdaBoostClassifier\n",
    "import time"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 146,
   "metadata": {},
   "outputs": [],
   "source": [
    "models = [\n",
    "    BernoulliNB(),\n",
    "    GaussianNB(),\n",
    "    KNeighborsClassifier(),\n",
    "    LogisticRegression(solver='lbfgs',max_iter=10000),\n",
    "    SGDClassifier(),\n",
    "    LinearSVC(max_iter=10000),\n",
    "    SVC(gamma='auto',max_iter=-1),\n",
    "    DecisionTreeClassifier(criterion='gini',max_depth=None),\n",
    "    RandomForestClassifier(criterion='gini',n_estimators=1100,max_depth=5,min_samples_split=4,min_samples_leaf=5,max_features='auto',oob_score=True,random_state=SEED,n_jobs=-1,verbose=1),\n",
    "    AdaBoostClassifier(),\n",
    "    MLPClassifier(max_iter=2000)\n",
    "]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 147,
   "metadata": {
    "scrolled": false
   },
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "[Parallel(n_jobs=-1)]: Using backend ThreadingBackend with 8 concurrent workers.\n",
      "[Parallel(n_jobs=-1)]: Done  34 tasks      | elapsed:    0.0s\n",
      "[Parallel(n_jobs=-1)]: Done 184 tasks      | elapsed:    0.5s\n",
      "[Parallel(n_jobs=-1)]: Done 434 tasks      | elapsed:    1.2s\n",
      "[Parallel(n_jobs=-1)]: Done 784 tasks      | elapsed:    2.3s\n",
      "[Parallel(n_jobs=-1)]: Done 1100 out of 1100 | elapsed:    3.3s finished\n",
      "[Parallel(n_jobs=8)]: Using backend ThreadingBackend with 8 concurrent workers.\n",
      "[Parallel(n_jobs=8)]: Done  34 tasks      | elapsed:    0.0s\n",
      "[Parallel(n_jobs=8)]: Done 184 tasks      | elapsed:    0.0s\n",
      "[Parallel(n_jobs=8)]: Done 434 tasks      | elapsed:    0.1s\n",
      "[Parallel(n_jobs=8)]: Done 784 tasks      | elapsed:    0.2s\n",
      "[Parallel(n_jobs=8)]: Done 1100 out of 1100 | elapsed:    0.3s finished\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>Model</th>\n",
       "      <th>Accuracy</th>\n",
       "      <th>Modelling Time</th>\n",
       "      <th>Prediction Time</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>SVC</td>\n",
       "      <td>85.47%</td>\n",
       "      <td>0.0469s</td>\n",
       "      <td>0.0156s</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8</th>\n",
       "      <td>RandomForestClassifier</td>\n",
       "      <td>85.47%</td>\n",
       "      <td>5.7505s</td>\n",
       "      <td>0.4533s</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>LogisticRegression</td>\n",
       "      <td>84.36%</td>\n",
       "      <td>0.0625s</td>\n",
       "      <td>0.0156s</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>LinearSVC</td>\n",
       "      <td>84.36%</td>\n",
       "      <td>0.4375s</td>\n",
       "      <td>0.0158s</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>BernoulliNB</td>\n",
       "      <td>83.8%</td>\n",
       "      <td>0.0156s</td>\n",
       "      <td>0.0156s</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>DecisionTreeClassifier</td>\n",
       "      <td>83.8%</td>\n",
       "      <td>0.0s</td>\n",
       "      <td>0.0156s</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>GaussianNB</td>\n",
       "      <td>83.24%</td>\n",
       "      <td>0.0156s</td>\n",
       "      <td>0.0s</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>KNeighborsClassifier</td>\n",
       "      <td>83.24%</td>\n",
       "      <td>0.0156s</td>\n",
       "      <td>0.0312s</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>10</th>\n",
       "      <td>MLPClassifier</td>\n",
       "      <td>83.24%</td>\n",
       "      <td>6.0943s</td>\n",
       "      <td>0.0156s</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9</th>\n",
       "      <td>AdaBoostClassifier</td>\n",
       "      <td>82.12%</td>\n",
       "      <td>0.2187s</td>\n",
       "      <td>0.0312s</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>SGDClassifier</td>\n",
       "      <td>81.56%</td>\n",
       "      <td>0.0156s</td>\n",
       "      <td>0.0s</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                     Model Accuracy Modelling Time Prediction Time\n",
       "6                      SVC   85.47%        0.0469s         0.0156s\n",
       "8   RandomForestClassifier   85.47%        5.7505s         0.4533s\n",
       "3       LogisticRegression   84.36%        0.0625s         0.0156s\n",
       "5                LinearSVC   84.36%        0.4375s         0.0158s\n",
       "0              BernoulliNB    83.8%        0.0156s         0.0156s\n",
       "7   DecisionTreeClassifier    83.8%           0.0s         0.0156s\n",
       "1               GaussianNB   83.24%        0.0156s            0.0s\n",
       "2     KNeighborsClassifier   83.24%        0.0156s         0.0312s\n",
       "10           MLPClassifier   83.24%        6.0943s         0.0156s\n",
       "9       AdaBoostClassifier   82.12%        0.2187s         0.0312s\n",
       "4            SGDClassifier   81.56%        0.0156s            0.0s"
      ]
     },
     "execution_count": 147,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "comparison = []\n",
    "\n",
    "for model in models:\n",
    "    before_model = time.time()\n",
    "    classifier_name = str(type(model).__name__)\n",
    "    list_of_labels = sorted(list(set(y_train)))\n",
    "    fit = model.fit(X_train, y_train)\n",
    "    before_pred = time.time()\n",
    "    predictions = fit.predict(X_test)\n",
    "\n",
    "    precision = precision_score(y_test, predictions, average=None, pos_label=None, labels=list_of_labels)\n",
    "    recall = recall_score(y_test, predictions, average=None, pos_label=None, labels=list_of_labels)\n",
    "    accuracy = accuracy_score(y_test, predictions)\n",
    "    cms = confusion_matrix(y_test,predictions)\n",
    "    \n",
    "    comparison.append([classifier_name, str(round(accuracy*100,2))+'%',str(round(before_pred-before_model,4))+'s',str(round(time.time()-before_pred,4))+'s'])\n",
    "    \n",
    "comparison = pd.DataFrame(comparison)\n",
    "comparison.columns = ['Model','Accuracy','Modelling Time', 'Prediction Time']\n",
    "comparison.set_index('Model')\n",
    "comparison.sort_values(\"Accuracy\", ascending=False)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "SVC and Random Forest seem to be our best two models, performing better than 85%"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Let's see how we perform when we submit the data, using our random forest  model:"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Random Forest - submission"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 174,
   "metadata": {
    "scrolled": true
   },
   "outputs": [],
   "source": [
    "#data_test.set_index('PassengerId',inplace=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 175,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "[Parallel(n_jobs=-1)]: Using backend ThreadingBackend with 8 concurrent workers.\n",
      "[Parallel(n_jobs=-1)]: Done  34 tasks      | elapsed:    0.0s\n",
      "[Parallel(n_jobs=-1)]: Done 184 tasks      | elapsed:    0.5s\n",
      "[Parallel(n_jobs=-1)]: Done 434 tasks      | elapsed:    1.3s\n",
      "[Parallel(n_jobs=-1)]: Done 784 tasks      | elapsed:    2.3s\n",
      "[Parallel(n_jobs=-1)]: Done 1100 out of 1100 | elapsed:    3.2s finished\n",
      "[Parallel(n_jobs=8)]: Using backend ThreadingBackend with 8 concurrent workers.\n",
      "[Parallel(n_jobs=8)]: Done  34 tasks      | elapsed:    0.0s\n",
      "[Parallel(n_jobs=8)]: Done 184 tasks      | elapsed:    0.0s\n",
      "[Parallel(n_jobs=8)]: Done 434 tasks      | elapsed:    0.1s\n",
      "[Parallel(n_jobs=8)]: Done 784 tasks      | elapsed:    0.3s\n",
      "[Parallel(n_jobs=8)]: Done 1100 out of 1100 | elapsed:    0.4s finished\n"
     ]
    }
   ],
   "source": [
    "model = RandomForestClassifier(criterion='gini',\n",
    "                                           n_estimators=1100,\n",
    "                                           max_depth=5,\n",
    "                                           min_samples_split=4,\n",
    "                                           min_samples_leaf=5,\n",
    "                                           max_features='auto',\n",
    "                                           oob_score=True,\n",
    "                                           random_state=SEED,\n",
    "                                           n_jobs=-1,\n",
    "                                           verbose=1) \n",
    "fit = model.fit(data_train.iloc[:,1:],data_train.iloc[:,0])\n",
    "predictions = fit.predict(data_test.iloc[:,1:])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 176,
   "metadata": {},
   "outputs": [],
   "source": [
    "data_test.reset_index(inplace=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 177,
   "metadata": {},
   "outputs": [],
   "source": [
    "submission = pd.DataFrame({\n",
    "        \"PassengerId\": data_test[\"PassengerId\"],\n",
    "        \"Survived\": predictions\n",
    "    })"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 152,
   "metadata": {},
   "outputs": [],
   "source": [
    "submission = submission.astype('int64')\n",
    "submission.to_csv('submissionRF.csv',index=False)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "The submission reveals an accuracy of 79.425%. A little lower than expected. However, it might have two reasons:\n",
    "1. we need to do cross validation to ensure we are picking a random model\n",
    "2. we might be overfitting, using too many variables\n",
    "\n",
    "Let's start with the first idea in short, but first, let's take a look at the importance of our features."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 191,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1080x1440 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "parameters = pd.Series(fit.feature_importances_, index=data_test.iloc[:,2:].columns).sort_values(ascending=False)\n",
    "plt.figure(figsize=(15, 20))\n",
    "sns.barplot(x=parameters, y=parameters.index)\n",
    "\n",
    "plt.xlabel('')\n",
    "plt.tick_params(axis='x', labelsize=15)\n",
    "plt.tick_params(axis='y', labelsize=15)\n",
    "plt.title('Random Forest Classifier Mean Feature Importance', size=15)\n",
    "\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Cross Validation <a class=\"anchor\" id=\"seventh-bullet\"></a>"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 153,
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn.pipeline import make_pipeline\n",
    "from sklearn.impute import SimpleImputer\n",
    "from sklearn.model_selection import cross_val_score\n",
    "\n",
    "my_pipeline = make_pipeline(SimpleImputer(), RandomForestClassifier(n_estimators=100))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "For a full list of scoring metrics see this [link](https://scikit-learn.org/stable/modules/model_evaluation.html)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 154,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[0.76666667 0.82222222 0.74157303 0.84269663 0.87640449 0.87640449\n",
      " 0.87640449 0.78651685 0.87640449 0.85227273]\n"
     ]
    }
   ],
   "source": [
    "from sklearn.metrics import accuracy_score\n",
    "scores = cross_val_score(my_pipeline, data_train.iloc[:,1:], data_train.iloc[:,0], \n",
    "                         scoring='accuracy', \n",
    "                         cv=10)\n",
    "print(scores)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 155,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Accuracy 83.18%\n"
     ]
    }
   ],
   "source": [
    "print('Accuracy %.02f' %(100*scores.mean()) +'%')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "In comparison to our previous accuracy of 85%, we now get a lower, more realistic accuracy of 83%."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Let's do the same thing again for all of our other models."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 156,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "[Parallel(n_jobs=-1)]: Using backend LokyBackend with 8 concurrent workers.\n",
      "[Parallel(n_jobs=-1)]: Done  34 tasks      | elapsed:    6.4s\n",
      "[Parallel(n_jobs=-1)]: Done 1100 out of 1100 | elapsed:    7.8s finished\n",
      "[Parallel(n_jobs=8)]: Using backend ThreadingBackend with 8 concurrent workers.\n",
      "[Parallel(n_jobs=8)]: Done  34 tasks      | elapsed:    0.0s\n",
      "[Parallel(n_jobs=8)]: Done 184 tasks      | elapsed:    0.0s\n",
      "[Parallel(n_jobs=8)]: Done 434 tasks      | elapsed:    0.0s\n",
      "[Parallel(n_jobs=8)]: Done 784 tasks      | elapsed:    0.1s\n",
      "[Parallel(n_jobs=8)]: Done 1100 out of 1100 | elapsed:    0.2s finished\n",
      "[Parallel(n_jobs=-1)]: Using backend LokyBackend with 8 concurrent workers.\n",
      "[Parallel(n_jobs=-1)]: Done  56 tasks      | elapsed:    0.1s\n",
      "[Parallel(n_jobs=-1)]: Done 1080 tasks      | elapsed:    1.4s\n",
      "[Parallel(n_jobs=-1)]: Done 1100 out of 1100 | elapsed:    1.4s finished\n",
      "[Parallel(n_jobs=8)]: Using backend ThreadingBackend with 8 concurrent workers.\n",
      "[Parallel(n_jobs=8)]: Done  34 tasks      | elapsed:    0.0s\n",
      "[Parallel(n_jobs=8)]: Done 184 tasks      | elapsed:    0.0s\n",
      "[Parallel(n_jobs=8)]: Done 434 tasks      | elapsed:    0.0s\n",
      "[Parallel(n_jobs=8)]: Done 784 tasks      | elapsed:    0.2s\n",
      "[Parallel(n_jobs=8)]: Done 1100 out of 1100 | elapsed:    0.3s finished\n",
      "[Parallel(n_jobs=-1)]: Using backend LokyBackend with 8 concurrent workers.\n",
      "[Parallel(n_jobs=-1)]: Done  56 tasks      | elapsed:    0.1s\n",
      "[Parallel(n_jobs=-1)]: Done 1100 out of 1100 | elapsed:    1.4s finished\n",
      "[Parallel(n_jobs=8)]: Using backend ThreadingBackend with 8 concurrent workers.\n",
      "[Parallel(n_jobs=8)]: Done  34 tasks      | elapsed:    0.0s\n",
      "[Parallel(n_jobs=8)]: Done 184 tasks      | elapsed:    0.0s\n",
      "[Parallel(n_jobs=8)]: Done 434 tasks      | elapsed:    0.0s\n",
      "[Parallel(n_jobs=8)]: Done 784 tasks      | elapsed:    0.1s\n",
      "[Parallel(n_jobs=8)]: Done 1100 out of 1100 | elapsed:    0.2s finished\n",
      "[Parallel(n_jobs=-1)]: Using backend LokyBackend with 8 concurrent workers.\n",
      "[Parallel(n_jobs=-1)]: Done  56 tasks      | elapsed:    0.1s\n",
      "[Parallel(n_jobs=-1)]: Done 1080 tasks      | elapsed:    1.5s\n",
      "[Parallel(n_jobs=-1)]: Done 1100 out of 1100 | elapsed:    1.6s finished\n",
      "[Parallel(n_jobs=8)]: Using backend ThreadingBackend with 8 concurrent workers.\n",
      "[Parallel(n_jobs=8)]: Done  34 tasks      | elapsed:    0.0s\n",
      "[Parallel(n_jobs=8)]: Done 184 tasks      | elapsed:    0.0s\n",
      "[Parallel(n_jobs=8)]: Done 434 tasks      | elapsed:    0.0s\n",
      "[Parallel(n_jobs=8)]: Done 784 tasks      | elapsed:    0.1s\n",
      "[Parallel(n_jobs=8)]: Done 1100 out of 1100 | elapsed:    0.2s finished\n",
      "[Parallel(n_jobs=-1)]: Using backend LokyBackend with 8 concurrent workers.\n",
      "[Parallel(n_jobs=-1)]: Done  56 tasks      | elapsed:    0.1s\n",
      "[Parallel(n_jobs=-1)]: Done 1080 tasks      | elapsed:    1.5s\n",
      "[Parallel(n_jobs=-1)]: Done 1100 out of 1100 | elapsed:    1.5s finished\n",
      "[Parallel(n_jobs=8)]: Using backend ThreadingBackend with 8 concurrent workers.\n",
      "[Parallel(n_jobs=8)]: Done  34 tasks      | elapsed:    0.0s\n",
      "[Parallel(n_jobs=8)]: Done 184 tasks      | elapsed:    0.0s\n",
      "[Parallel(n_jobs=8)]: Done 434 tasks      | elapsed:    0.1s\n",
      "[Parallel(n_jobs=8)]: Done 784 tasks      | elapsed:    0.2s\n",
      "[Parallel(n_jobs=8)]: Done 1100 out of 1100 | elapsed:    0.3s finished\n",
      "[Parallel(n_jobs=-1)]: Using backend LokyBackend with 8 concurrent workers.\n",
      "[Parallel(n_jobs=-1)]: Done  56 tasks      | elapsed:    0.1s\n",
      "[Parallel(n_jobs=-1)]: Done 1080 tasks      | elapsed:    1.4s\n",
      "[Parallel(n_jobs=-1)]: Done 1100 out of 1100 | elapsed:    1.4s finished\n",
      "[Parallel(n_jobs=8)]: Using backend ThreadingBackend with 8 concurrent workers.\n",
      "[Parallel(n_jobs=8)]: Done  34 tasks      | elapsed:    0.0s\n",
      "[Parallel(n_jobs=8)]: Done 184 tasks      | elapsed:    0.0s\n",
      "[Parallel(n_jobs=8)]: Done 434 tasks      | elapsed:    0.1s\n",
      "[Parallel(n_jobs=8)]: Done 784 tasks      | elapsed:    0.2s\n",
      "[Parallel(n_jobs=8)]: Done 1100 out of 1100 | elapsed:    0.3s finished\n",
      "[Parallel(n_jobs=-1)]: Using backend LokyBackend with 8 concurrent workers.\n",
      "[Parallel(n_jobs=-1)]: Done  56 tasks      | elapsed:    0.1s\n",
      "[Parallel(n_jobs=-1)]: Done 1080 tasks      | elapsed:    1.4s\n",
      "[Parallel(n_jobs=-1)]: Done 1100 out of 1100 | elapsed:    1.5s finished\n",
      "[Parallel(n_jobs=8)]: Using backend ThreadingBackend with 8 concurrent workers.\n",
      "[Parallel(n_jobs=8)]: Done  34 tasks      | elapsed:    0.0s\n",
      "[Parallel(n_jobs=8)]: Done 184 tasks      | elapsed:    0.0s\n",
      "[Parallel(n_jobs=8)]: Done 434 tasks      | elapsed:    0.0s\n",
      "[Parallel(n_jobs=8)]: Done 784 tasks      | elapsed:    0.1s\n",
      "[Parallel(n_jobs=8)]: Done 1100 out of 1100 | elapsed:    0.2s finished\n",
      "[Parallel(n_jobs=-1)]: Using backend LokyBackend with 8 concurrent workers.\n",
      "[Parallel(n_jobs=-1)]: Done  56 tasks      | elapsed:    0.1s\n",
      "[Parallel(n_jobs=-1)]: Done 1100 out of 1100 | elapsed:    1.4s finished\n",
      "[Parallel(n_jobs=8)]: Using backend ThreadingBackend with 8 concurrent workers.\n",
      "[Parallel(n_jobs=8)]: Done  34 tasks      | elapsed:    0.0s\n",
      "[Parallel(n_jobs=8)]: Done 184 tasks      | elapsed:    0.0s\n",
      "[Parallel(n_jobs=8)]: Done 434 tasks      | elapsed:    0.0s\n",
      "[Parallel(n_jobs=8)]: Done 784 tasks      | elapsed:    0.1s\n",
      "[Parallel(n_jobs=8)]: Done 1100 out of 1100 | elapsed:    0.3s finished\n",
      "[Parallel(n_jobs=-1)]: Using backend LokyBackend with 8 concurrent workers.\n",
      "[Parallel(n_jobs=-1)]: Done  56 tasks      | elapsed:    0.1s\n",
      "[Parallel(n_jobs=-1)]: Done 1080 tasks      | elapsed:    1.4s\n",
      "[Parallel(n_jobs=-1)]: Done 1100 out of 1100 | elapsed:    1.4s finished\n",
      "[Parallel(n_jobs=8)]: Using backend ThreadingBackend with 8 concurrent workers.\n",
      "[Parallel(n_jobs=8)]: Done  34 tasks      | elapsed:    0.0s\n",
      "[Parallel(n_jobs=8)]: Done 184 tasks      | elapsed:    0.0s\n",
      "[Parallel(n_jobs=8)]: Done 434 tasks      | elapsed:    0.0s\n",
      "[Parallel(n_jobs=8)]: Done 784 tasks      | elapsed:    0.1s\n",
      "[Parallel(n_jobs=8)]: Done 1100 out of 1100 | elapsed:    0.3s finished\n",
      "[Parallel(n_jobs=-1)]: Using backend LokyBackend with 8 concurrent workers.\n",
      "[Parallel(n_jobs=-1)]: Done  56 tasks      | elapsed:    0.1s\n",
      "[Parallel(n_jobs=-1)]: Done 1080 tasks      | elapsed:    1.4s\n",
      "[Parallel(n_jobs=-1)]: Done 1100 out of 1100 | elapsed:    1.5s finished\n",
      "[Parallel(n_jobs=8)]: Using backend ThreadingBackend with 8 concurrent workers.\n",
      "[Parallel(n_jobs=8)]: Done  34 tasks      | elapsed:    0.0s\n",
      "[Parallel(n_jobs=8)]: Done 184 tasks      | elapsed:    0.0s\n",
      "[Parallel(n_jobs=8)]: Done 434 tasks      | elapsed:    0.1s\n",
      "[Parallel(n_jobs=8)]: Done 784 tasks      | elapsed:    0.2s\n",
      "[Parallel(n_jobs=8)]: Done 1100 out of 1100 | elapsed:    0.2s finished\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>Model</th>\n",
       "      <th>Accuracy</th>\n",
       "      <th>Modelling Time</th>\n",
       "      <th>Prediction Time</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>8</th>\n",
       "      <td>RandomForestClassifier</td>\n",
       "      <td>83.5%</td>\n",
       "      <td>0.0s</td>\n",
       "      <td>48.7745s</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>LogisticRegression</td>\n",
       "      <td>83.28%</td>\n",
       "      <td>0.0s</td>\n",
       "      <td>0.7398s</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>SVC</td>\n",
       "      <td>83.27%</td>\n",
       "      <td>0.0s</td>\n",
       "      <td>0.625s</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>LinearSVC</td>\n",
       "      <td>82.83%</td>\n",
       "      <td>0.0s</td>\n",
       "      <td>4.4489s</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>DecisionTreeClassifier</td>\n",
       "      <td>82.39%</td>\n",
       "      <td>0.0s</td>\n",
       "      <td>0.1719s</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>KNeighborsClassifier</td>\n",
       "      <td>82.16%</td>\n",
       "      <td>0.0s</td>\n",
       "      <td>0.2656s</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>10</th>\n",
       "      <td>MLPClassifier</td>\n",
       "      <td>82.16%</td>\n",
       "      <td>0.0s</td>\n",
       "      <td>14.4099s</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>GaussianNB</td>\n",
       "      <td>81.71%</td>\n",
       "      <td>0.0s</td>\n",
       "      <td>0.1562s</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9</th>\n",
       "      <td>AdaBoostClassifier</td>\n",
       "      <td>81.48%</td>\n",
       "      <td>0.0s</td>\n",
       "      <td>2.3283s</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>SGDClassifier</td>\n",
       "      <td>81.36%</td>\n",
       "      <td>0.0s</td>\n",
       "      <td>0.2813s</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>BernoulliNB</td>\n",
       "      <td>80.13%</td>\n",
       "      <td>0.0s</td>\n",
       "      <td>0.2033s</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                     Model Accuracy Modelling Time Prediction Time\n",
       "8   RandomForestClassifier    83.5%           0.0s        48.7745s\n",
       "3       LogisticRegression   83.28%           0.0s         0.7398s\n",
       "6                      SVC   83.27%           0.0s          0.625s\n",
       "5                LinearSVC   82.83%           0.0s         4.4489s\n",
       "7   DecisionTreeClassifier   82.39%           0.0s         0.1719s\n",
       "2     KNeighborsClassifier   82.16%           0.0s         0.2656s\n",
       "10           MLPClassifier   82.16%           0.0s        14.4099s\n",
       "1               GaussianNB   81.71%           0.0s         0.1562s\n",
       "9       AdaBoostClassifier   81.48%           0.0s         2.3283s\n",
       "4            SGDClassifier   81.36%           0.0s         0.2813s\n",
       "0              BernoulliNB   80.13%           0.0s         0.2033s"
      ]
     },
     "execution_count": 156,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "comparison = []\n",
    "\n",
    "for model in models:\n",
    "    before_model = time.time()\n",
    "    classifier_name = str(type(model).__name__)\n",
    "    list_of_labels = sorted(list(set(y_train)))\n",
    "    before_pred = time.time()\n",
    "    \n",
    "    my_pipeline = make_pipeline(SimpleImputer(), model)\n",
    "    scores = cross_val_score(my_pipeline, data_train.iloc[:,1:], data_train.iloc[:,0], scoring='accuracy', cv=10)\n",
    "\n",
    "    \n",
    "    comparison.append([classifier_name, str(round(scores.mean()*100,2))+'%',str(round(before_pred-before_model,4))+'s',str(round(time.time()-before_pred,4))+'s'])\n",
    "    \n",
    "comparison = pd.DataFrame(comparison)\n",
    "comparison.columns = ['Model','Accuracy','Modelling Time', 'Prediction Time']\n",
    "comparison.set_index('Model')\n",
    "comparison.sort_values(\"Accuracy\", ascending=False)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Now with 10-times cross validation, we see more realistic accuracy values for all models.\n",
    "Still, the SVC and Random Forest are among the best models together with Logistic Regression and MLP achieving an accuracy of ~83%."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Logistic Regression - submission\n",
    "As Logistic Regression was performing second-best in our CV model comparison, let's hand in a submission with LogisticRegression"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 157,
   "metadata": {
    "scrolled": true
   },
   "outputs": [],
   "source": [
    "data_test.set_index('PassengerId',inplace=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 158,
   "metadata": {},
   "outputs": [],
   "source": [
    "model = LogisticRegression(solver='lbfgs',max_iter=10000)\n",
    "fit = model.fit(data_train.iloc[:,1:],data_train.iloc[:,0])\n",
    "predictions = fit.predict(data_test.iloc[:,1:])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 159,
   "metadata": {},
   "outputs": [],
   "source": [
    "data_test.reset_index(inplace=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 160,
   "metadata": {},
   "outputs": [],
   "source": [
    "submission = pd.DataFrame({\n",
    "        \"PassengerId\": data_test[\"PassengerId\"],\n",
    "        \"Survived\": predictions\n",
    "    })"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 161,
   "metadata": {},
   "outputs": [],
   "source": [
    "submission = submission.astype('int64')\n",
    "submission.to_csv('submissionLR.csv',index=False)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Logistic regression, however, only achieves 77.99% of accuracy."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 162,
   "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>Feature</th>\n",
       "      <th>Correlation</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>16</th>\n",
       "      <td>FamilySizeGrouped_Alone</td>\n",
       "      <td>1.349651</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>23</th>\n",
       "      <td>IsWomanOrBoy_True</td>\n",
       "      <td>0.974955</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>19</th>\n",
       "      <td>FamilySizeGrouped_Small</td>\n",
       "      <td>0.970146</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>14</th>\n",
       "      <td>Title_4</td>\n",
       "      <td>0.601029</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>Pclass_1</td>\n",
       "      <td>0.581144</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>21</th>\n",
       "      <td>HasCabin_True</td>\n",
       "      <td>0.387437</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>Sex_1</td>\n",
       "      <td>0.373926</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>12</th>\n",
       "      <td>Title_2</td>\n",
       "      <td>0.327537</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8</th>\n",
       "      <td>Embarked_1</td>\n",
       "      <td>0.161565</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>Pclass_2</td>\n",
       "      <td>0.156553</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9</th>\n",
       "      <td>Embarked_2</td>\n",
       "      <td>0.131049</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>Fare</td>\n",
       "      <td>0.122833</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>TicketFrequency</td>\n",
       "      <td>-0.007092</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>13</th>\n",
       "      <td>Title_3</td>\n",
       "      <td>-0.171259</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>10</th>\n",
       "      <td>Embarked_3</td>\n",
       "      <td>-0.292211</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>15</th>\n",
       "      <td>Title_5</td>\n",
       "      <td>-0.361794</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>Sex_2</td>\n",
       "      <td>-0.373524</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>AgeGroup</td>\n",
       "      <td>-0.384131</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>20</th>\n",
       "      <td>HasCabin_False</td>\n",
       "      <td>-0.387035</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>11</th>\n",
       "      <td>Title_1</td>\n",
       "      <td>-0.395110</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>Pclass_3</td>\n",
       "      <td>-0.737295</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>17</th>\n",
       "      <td>FamilySizeGrouped_Large</td>\n",
       "      <td>-0.920857</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>22</th>\n",
       "      <td>IsWomanOrBoy_False</td>\n",
       "      <td>-0.974553</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>18</th>\n",
       "      <td>FamilySizeGrouped_Medium</td>\n",
       "      <td>-1.398537</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                     Feature  Correlation\n",
       "16   FamilySizeGrouped_Alone     1.349651\n",
       "23         IsWomanOrBoy_True     0.974955\n",
       "19   FamilySizeGrouped_Small     0.970146\n",
       "14                   Title_4     0.601029\n",
       "3                   Pclass_1     0.581144\n",
       "21             HasCabin_True     0.387437\n",
       "6                      Sex_1     0.373926\n",
       "12                   Title_2     0.327537\n",
       "8                 Embarked_1     0.161565\n",
       "4                   Pclass_2     0.156553\n",
       "9                 Embarked_2     0.131049\n",
       "0                       Fare     0.122833\n",
       "1            TicketFrequency    -0.007092\n",
       "13                   Title_3    -0.171259\n",
       "10                Embarked_3    -0.292211\n",
       "15                   Title_5    -0.361794\n",
       "7                      Sex_2    -0.373524\n",
       "2                   AgeGroup    -0.384131\n",
       "20            HasCabin_False    -0.387035\n",
       "11                   Title_1    -0.395110\n",
       "5                   Pclass_3    -0.737295\n",
       "17   FamilySizeGrouped_Large    -0.920857\n",
       "22        IsWomanOrBoy_False    -0.974553\n",
       "18  FamilySizeGrouped_Medium    -1.398537"
      ]
     },
     "execution_count": 162,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "coeff_df = pd.DataFrame(data_train.columns.delete(0))\n",
    "coeff_df.columns = ['Feature']\n",
    "coeff_df[\"Correlation\"] = pd.Series(model.coef_[0])\n",
    "\n",
    "coeff_df.sort_values(by='Correlation', ascending=False)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "We can use a Logistic Regression to validate our assumptions and decisions for feature creation. This can be done by calculating the coefficient of the features in the decision function.\n",
    "\n",
    "Positive coefficients increase the log-odds of the response (and thus increase the probability), and negative coefficients decrease the log-odds of the response (and thus decrease the probability).\n",
    "\n",
    "* FamilySizeGrouped_Alone is the highest positive coefficient, meaning that when belonging to this group (i.e. traveling alone) the probability of survival increases the most.\n",
    "* IsWomanOrBoy is the second highest positive coefficient, meaning that when IsWomanOrBoy increases from 0 to 1 (i.e. is True), the probability of Survived = 1 increases the second most.\n",
    "* Likely, HasCabin or Fare both have a positive influence once they increase (i.e. the person traveling has a cabin or has paid a higher Fare)\n",
    "* On the other extreme, the higher the Class (i.e. the lower the Fare paid - see correlation above) the lower the chance of survival\n",
    "* The Same counts for AgeGroup or when the person is not a woman or boy"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### SVC - submission"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Finally, let's also hand in our support vector machine model."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 164,
   "metadata": {},
   "outputs": [],
   "source": [
    "model = SVC(gamma='auto')\n",
    "fit = model.fit(data_train.iloc[:,1:],data_train.iloc[:,0])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 167,
   "metadata": {},
   "outputs": [],
   "source": [
    "predictions = fit.predict(data_test.iloc[:,2:])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 168,
   "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>Fare</th>\n",
       "      <th>TicketFrequency</th>\n",
       "      <th>AgeGroup</th>\n",
       "      <th>Pclass_1</th>\n",
       "      <th>Pclass_2</th>\n",
       "      <th>Pclass_3</th>\n",
       "      <th>Sex_1</th>\n",
       "      <th>Sex_2</th>\n",
       "      <th>Embarked_1</th>\n",
       "      <th>Embarked_2</th>\n",
       "      <th>Embarked_3</th>\n",
       "      <th>Title_1</th>\n",
       "      <th>Title_2</th>\n",
       "      <th>Title_3</th>\n",
       "      <th>Title_4</th>\n",
       "      <th>Title_5</th>\n",
       "      <th>FamilySizeGrouped_Alone</th>\n",
       "      <th>FamilySizeGrouped_Large</th>\n",
       "      <th>FamilySizeGrouped_Medium</th>\n",
       "      <th>FamilySizeGrouped_Small</th>\n",
       "      <th>HasCabin_False</th>\n",
       "      <th>HasCabin_True</th>\n",
       "      <th>IsWomanOrBoy_False</th>\n",
       "      <th>IsWomanOrBoy_True</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>892</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>2</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>893</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>2</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>894</td>\n",
       "      <td>NaN</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>3</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>895</td>\n",
       "      <td>NaN</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>2</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>896</td>\n",
       "      <td>NaN</td>\n",
       "      <td>1</td>\n",
       "      <td>2</td>\n",
       "      <td>2</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   PassengerId  Survived  Fare  TicketFrequency  AgeGroup  Pclass_1  Pclass_2  \\\n",
       "0          892       NaN     0                1         2       0.0       0.0   \n",
       "1          893       NaN     0                1         2       0.0       0.0   \n",
       "2          894       NaN     1                1         3       0.0       1.0   \n",
       "3          895       NaN     1                1         2       0.0       0.0   \n",
       "4          896       NaN     1                2         2       0.0       0.0   \n",
       "\n",
       "   Pclass_3  Sex_1  Sex_2  Embarked_1  Embarked_2  Embarked_3  Title_1  \\\n",
       "0       1.0    0.0    1.0         0.0         1.0         0.0      1.0   \n",
       "1       1.0    1.0    0.0         0.0         0.0         1.0      0.0   \n",
       "2       0.0    0.0    1.0         0.0         1.0         0.0      1.0   \n",
       "3       1.0    0.0    1.0         0.0         0.0         1.0      1.0   \n",
       "4       1.0    1.0    0.0         0.0         0.0         1.0      0.0   \n",
       "\n",
       "   Title_2  Title_3  Title_4  Title_5  FamilySizeGrouped_Alone  \\\n",
       "0      0.0      0.0      0.0      0.0                      1.0   \n",
       "1      1.0      0.0      0.0      0.0                      0.0   \n",
       "2      0.0      0.0      0.0      0.0                      1.0   \n",
       "3      0.0      0.0      0.0      0.0                      1.0   \n",
       "4      1.0      0.0      0.0      0.0                      0.0   \n",
       "\n",
       "   FamilySizeGrouped_Large  FamilySizeGrouped_Medium  FamilySizeGrouped_Small  \\\n",
       "0                      0.0                       0.0                      0.0   \n",
       "1                      0.0                       0.0                      1.0   \n",
       "2                      0.0                       0.0                      0.0   \n",
       "3                      0.0                       0.0                      0.0   \n",
       "4                      0.0                       0.0                      1.0   \n",
       "\n",
       "   HasCabin_False  HasCabin_True  IsWomanOrBoy_False  IsWomanOrBoy_True  \n",
       "0             1.0            0.0                 1.0                0.0  \n",
       "1             1.0            0.0                 0.0                1.0  \n",
       "2             1.0            0.0                 1.0                0.0  \n",
       "3             1.0            0.0                 1.0                0.0  \n",
       "4             1.0            0.0                 0.0                1.0  "
      ]
     },
     "execution_count": 168,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data_test.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 169,
   "metadata": {},
   "outputs": [],
   "source": [
    "submission = pd.DataFrame({\n",
    "        \"PassengerId\": data_test[\"PassengerId\"],\n",
    "        \"Survived\": predictions\n",
    "    })"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 170,
   "metadata": {},
   "outputs": [],
   "source": [
    "submission = submission.astype('int64')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 171,
   "metadata": {},
   "outputs": [],
   "source": [
    "submission.to_csv('submissionSVC.csv',index=False)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Finally, the SVC model reaches an accuracy of 79.425%."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Grid Search <a class=\"anchor\" id=\"eight-bullet\"></a>\n",
    "\n",
    "We can finally improve our best model (RF) through grid search."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 220,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Fitting 5 folds for each of 1386 candidates, totalling 6930 fits\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "[Parallel(n_jobs=-1)]: Using backend LokyBackend with 8 concurrent workers.\n",
      "[Parallel(n_jobs=-1)]: Done  34 tasks      | elapsed:   19.0s\n",
      "[Parallel(n_jobs=-1)]: Done 184 tasks      | elapsed:  2.9min\n",
      "[Parallel(n_jobs=-1)]: Done 434 tasks      | elapsed:  7.3min\n",
      "[Parallel(n_jobs=-1)]: Done 784 tasks      | elapsed: 12.6min\n",
      "[Parallel(n_jobs=-1)]: Done 1234 tasks      | elapsed: 19.3min\n",
      "[Parallel(n_jobs=-1)]: Done 1784 tasks      | elapsed: 27.7min\n",
      "[Parallel(n_jobs=-1)]: Done 2434 tasks      | elapsed: 38.2min\n",
      "[Parallel(n_jobs=-1)]: Done 3184 tasks      | elapsed: 51.1min\n",
      "[Parallel(n_jobs=-1)]: Done 4034 tasks      | elapsed: 65.5min\n",
      "[Parallel(n_jobs=-1)]: Done 4984 tasks      | elapsed: 82.0min\n",
      "[Parallel(n_jobs=-1)]: Done 6034 tasks      | elapsed: 99.9min\n",
      "[Parallel(n_jobs=-1)]: Done 6930 out of 6930 | elapsed: 112.8min finished\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "0.8356741573033708"
      ]
     },
     "execution_count": 220,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "from sklearn.model_selection import GridSearchCV\n",
    "from sklearn.model_selection import StratifiedKFold\n",
    "\n",
    "k_fold = StratifiedKFold(n_splits=5)\n",
    "RF_classifier = RandomForestClassifier()\n",
    "\n",
    "\n",
    "## Search grid for optimal parameters\n",
    "RF_paramgrid = {\"max_depth\": [None],\n",
    "                  \"max_features\": [1, 2 ,3, 4, 5, 10, 'auto'],\n",
    "                  \"min_samples_split\": [2 , 5,  10],\n",
    "                  \"min_samples_leaf\": [2, 5, 10],\n",
    "                  \"bootstrap\": [True],\n",
    "                  \"n_estimators\" :[100,200,300,400,500,1000,1100,1200,1300,1400,1500],\n",
    "                  \"criterion\": [\"gini\"],\n",
    "                  \"oob_score\": [True, False]\n",
    "                   }\n",
    "\n",
    "\n",
    "RF_classifiergrid = GridSearchCV(RF_classifier, param_grid = RF_paramgrid, cv=k_fold, scoring=\"accuracy\", n_jobs= -1, verbose=1)\n",
    "\n",
    "RF_classifiergrid.fit(X_train,y_train)\n",
    "\n",
    "RFC_optimum = RF_classifiergrid.best_estimator_\n",
    "\n",
    "# Best Accuracy Score\n",
    "RF_classifiergrid.best_score_"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 221,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "RandomForestClassifier(bootstrap=True, class_weight=None, criterion='gini',\n",
       "                       max_depth=None, max_features=4, max_leaf_nodes=None,\n",
       "                       min_impurity_decrease=0.0, min_impurity_split=None,\n",
       "                       min_samples_leaf=5, min_samples_split=2,\n",
       "                       min_weight_fraction_leaf=0.0, n_estimators=500,\n",
       "                       n_jobs=None, oob_score=True, random_state=None,\n",
       "                       verbose=0, warm_start=False)"
      ]
     },
     "execution_count": 221,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "RFC_optimum"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 224,
   "metadata": {},
   "outputs": [],
   "source": [
    "model_rffinal=RandomForestClassifier(\n",
    "                                    n_estimators=500,\n",
    "                                    criterion='gini',\n",
    "                                    max_depth=None,\n",
    "                                    min_samples_split=2,\n",
    "                                    min_samples_leaf=5,\n",
    "                                    min_weight_fraction_leaf=0.0,\n",
    "                                    max_features=4,\n",
    "                                    max_leaf_nodes=None,\n",
    "                                    min_impurity_decrease=0.0,\n",
    "                                    min_impurity_split=None,\n",
    "                                    bootstrap=True,\n",
    "                                    oob_score=True,\n",
    "                                    n_jobs=None,\n",
    "                                    random_state=SEED,\n",
    "                                    verbose=0,\n",
    "                                    warm_start=False,\n",
    "                                    class_weight=None\n",
    "                                )"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 202,
   "metadata": {},
   "outputs": [],
   "source": [
    "model_rffinal = RandomForestClassifier(bootstrap=True, class_weight=None, criterion='entropy',\n",
    "                       max_depth=None, max_features=10, max_leaf_nodes=None,\n",
    "                       min_impurity_decrease=0.0, min_impurity_split=None,\n",
    "                       min_samples_leaf=1, min_samples_split=4,\n",
    "                       min_weight_fraction_leaf=0.0, n_estimators=200,\n",
    "                       n_jobs=-1, oob_score=False, random_state=None, verbose=1,\n",
    "                       warm_start=False)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 225,
   "metadata": {},
   "outputs": [],
   "source": [
    "fit = model_rffinal.fit(data_train.iloc[:,1:],data_train.iloc[:,0])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 226,
   "metadata": {},
   "outputs": [],
   "source": [
    "predictions = fit.predict(data_test.iloc[:,2:])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 227,
   "metadata": {},
   "outputs": [],
   "source": [
    "submission = pd.DataFrame({\n",
    "        \"PassengerId\": data_test[\"PassengerId\"],\n",
    "        \"Survived\": predictions\n",
    "    })\n",
    "submission = submission.astype('int64')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 228,
   "metadata": {},
   "outputs": [],
   "source": [
    "submission.to_csv('submissionrfgrid.csv',index=False)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Unfortunately, after the grid search, the handed in model  reaches 80.382% of accuracy. This is so far our best model and a slight improvement in comparison to our 79.425% we achieved earlier. Currently this is ranked among the top 8% in kaggle titanic competition."
   ]
  }
 ],
 "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": "3.7.5"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}