4j_feature_engineering.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Feature Engineering (Titanic Dataset)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "One of the best ways to improve model performance is **feature engineering**. \n",
    "\n",
    "When feature engineering, we create additional columns using \n",
    "- domain knowledge\n",
    "- combinations and interactions of features in the dataset\n",
    "- additional lookups (e.g. the travel time for a patient from the clinic they are visiting rather than just learning from the postcode combinations)\n",
    "\n",
    "Let's see if we can improve performance on the titanic dataset with a little feature engineering.\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 634,
   "metadata": {},
   "outputs": [],
   "source": [
    "import pandas as pd\n",
    "import numpy as np\n",
    "from sklearn.model_selection import train_test_split\n",
    "from xgboost import XGBClassifier\n",
    "from sklearn.impute import SimpleImputer\n",
    "from sklearn.metrics import ConfusionMatrixDisplay, confusion_matrix, f1_score, precision_score, \\\n",
    "                            recall_score, roc_auc_score\n",
    "from sklearn.inspection import permutation_importance\n",
    "import matplotlib.pyplot as plt\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "We're going to import the raw data rather than the preprocessed data we've used in some cases as we want to make use of some of the additional features available in this dataset."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 635,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>PassengerId</th>\n",
       "      <th>Survived</th>\n",
       "      <th>Pclass</th>\n",
       "      <th>Name</th>\n",
       "      <th>Sex</th>\n",
       "      <th>Age</th>\n",
       "      <th>SibSp</th>\n",
       "      <th>Parch</th>\n",
       "      <th>Ticket</th>\n",
       "      <th>Fare</th>\n",
       "      <th>Cabin</th>\n",
       "      <th>Embarked</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>3</td>\n",
       "      <td>Braund, Mr. Owen Harris</td>\n",
       "      <td>male</td>\n",
       "      <td>22.0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>A/5 21171</td>\n",
       "      <td>7.2500</td>\n",
       "      <td>NaN</td>\n",
       "      <td>S</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>2</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>Cumings, Mrs. John Bradley (Florence Briggs Th...</td>\n",
       "      <td>female</td>\n",
       "      <td>38.0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>PC 17599</td>\n",
       "      <td>71.2833</td>\n",
       "      <td>C85</td>\n",
       "      <td>C</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>3</td>\n",
       "      <td>1</td>\n",
       "      <td>3</td>\n",
       "      <td>Heikkinen, Miss. Laina</td>\n",
       "      <td>female</td>\n",
       "      <td>26.0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>STON/O2. 3101282</td>\n",
       "      <td>7.9250</td>\n",
       "      <td>NaN</td>\n",
       "      <td>S</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>4</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>Futrelle, Mrs. Jacques Heath (Lily May Peel)</td>\n",
       "      <td>female</td>\n",
       "      <td>35.0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>113803</td>\n",
       "      <td>53.1000</td>\n",
       "      <td>C123</td>\n",
       "      <td>S</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>5</td>\n",
       "      <td>0</td>\n",
       "      <td>3</td>\n",
       "      <td>Allen, Mr. William Henry</td>\n",
       "      <td>male</td>\n",
       "      <td>35.0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>373450</td>\n",
       "      <td>8.0500</td>\n",
       "      <td>NaN</td>\n",
       "      <td>S</td>\n",
       "    </tr>\n",
       "    <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>886</th>\n",
       "      <td>887</td>\n",
       "      <td>0</td>\n",
       "      <td>2</td>\n",
       "      <td>Montvila, Rev. Juozas</td>\n",
       "      <td>male</td>\n",
       "      <td>27.0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>211536</td>\n",
       "      <td>13.0000</td>\n",
       "      <td>NaN</td>\n",
       "      <td>S</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>887</th>\n",
       "      <td>888</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>Graham, Miss. Margaret Edith</td>\n",
       "      <td>female</td>\n",
       "      <td>19.0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>112053</td>\n",
       "      <td>30.0000</td>\n",
       "      <td>B42</td>\n",
       "      <td>S</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>888</th>\n",
       "      <td>889</td>\n",
       "      <td>0</td>\n",
       "      <td>3</td>\n",
       "      <td>Johnston, Miss. Catherine Helen \"Carrie\"</td>\n",
       "      <td>female</td>\n",
       "      <td>NaN</td>\n",
       "      <td>1</td>\n",
       "      <td>2</td>\n",
       "      <td>W./C. 6607</td>\n",
       "      <td>23.4500</td>\n",
       "      <td>NaN</td>\n",
       "      <td>S</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>889</th>\n",
       "      <td>890</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>Behr, Mr. Karl Howell</td>\n",
       "      <td>male</td>\n",
       "      <td>26.0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>111369</td>\n",
       "      <td>30.0000</td>\n",
       "      <td>C148</td>\n",
       "      <td>C</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>890</th>\n",
       "      <td>891</td>\n",
       "      <td>0</td>\n",
       "      <td>3</td>\n",
       "      <td>Dooley, Mr. Patrick</td>\n",
       "      <td>male</td>\n",
       "      <td>32.0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>370376</td>\n",
       "      <td>7.7500</td>\n",
       "      <td>NaN</td>\n",
       "      <td>Q</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>891 rows × 12 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "     PassengerId  Survived  Pclass  \\\n",
       "0              1         0       3   \n",
       "1              2         1       1   \n",
       "2              3         1       3   \n",
       "3              4         1       1   \n",
       "4              5         0       3   \n",
       "..           ...       ...     ...   \n",
       "886          887         0       2   \n",
       "887          888         1       1   \n",
       "888          889         0       3   \n",
       "889          890         1       1   \n",
       "890          891         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",
       "886                              Montvila, Rev. Juozas    male  27.0      0   \n",
       "887                       Graham, Miss. Margaret Edith  female  19.0      0   \n",
       "888           Johnston, Miss. Catherine Helen \"Carrie\"  female   NaN      1   \n",
       "889                              Behr, Mr. Karl Howell    male  26.0      0   \n",
       "890                                Dooley, Mr. Patrick    male  32.0      0   \n",
       "\n",
       "     Parch            Ticket     Fare Cabin Embarked  \n",
       "0        0         A/5 21171   7.2500   NaN        S  \n",
       "1        0          PC 17599  71.2833   C85        C  \n",
       "2        0  STON/O2. 3101282   7.9250   NaN        S  \n",
       "3        0            113803  53.1000  C123        S  \n",
       "4        0            373450   8.0500   NaN        S  \n",
       "..     ...               ...      ...   ...      ...  \n",
       "886      0            211536  13.0000   NaN        S  \n",
       "887      0            112053  30.0000   B42        S  \n",
       "888      2        W./C. 6607  23.4500   NaN        S  \n",
       "889      0            111369  30.0000  C148        C  \n",
       "890      0            370376   7.7500   NaN        Q  \n",
       "\n",
       "[891 rows x 12 columns]"
      ]
     },
     "execution_count": 635,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "try:\n",
    "    data = pd.read_csv(\"data/raw_data.csv\")\n",
    "\n",
    "except FileNotFoundError:\n",
    "\n",
    "    # Download raw data:\n",
    "    address = 'https://raw.githubusercontent.com/MichaelAllen1966/' + \\\n",
    "                '1804_python_healthcare/master/titanic/data/train.csv'\n",
    "\n",
    "    data = pd.read_csv(address)\n",
    "\n",
    "    # Create a data subfolder if one does not already exist\n",
    "    import os\n",
    "    data_directory ='./data/'\n",
    "    if not os.path.exists(data_directory):\n",
    "        os.makedirs(data_directory)\n",
    "\n",
    "    # Save data\n",
    "    data.to_csv(data_directory + 'raw_data.csv', index=False)\n",
    "\n",
    "data"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Binary features - Simple Conditions"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "First, let's make a feature that reflects whether the passenger is under 16. "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 636,
   "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>Under18</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>3</td>\n",
       "      <td>Braund, Mr. Owen Harris</td>\n",
       "      <td>male</td>\n",
       "      <td>22.0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>A/5 21171</td>\n",
       "      <td>7.2500</td>\n",
       "      <td>NaN</td>\n",
       "      <td>S</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>2</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>Cumings, Mrs. John Bradley (Florence Briggs Th...</td>\n",
       "      <td>female</td>\n",
       "      <td>38.0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>PC 17599</td>\n",
       "      <td>71.2833</td>\n",
       "      <td>C85</td>\n",
       "      <td>C</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>3</td>\n",
       "      <td>1</td>\n",
       "      <td>3</td>\n",
       "      <td>Heikkinen, Miss. Laina</td>\n",
       "      <td>female</td>\n",
       "      <td>26.0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>STON/O2. 3101282</td>\n",
       "      <td>7.9250</td>\n",
       "      <td>NaN</td>\n",
       "      <td>S</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>4</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>Futrelle, Mrs. Jacques Heath (Lily May Peel)</td>\n",
       "      <td>female</td>\n",
       "      <td>35.0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>113803</td>\n",
       "      <td>53.1000</td>\n",
       "      <td>C123</td>\n",
       "      <td>S</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>5</td>\n",
       "      <td>0</td>\n",
       "      <td>3</td>\n",
       "      <td>Allen, Mr. William Henry</td>\n",
       "      <td>male</td>\n",
       "      <td>35.0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>373450</td>\n",
       "      <td>8.0500</td>\n",
       "      <td>NaN</td>\n",
       "      <td>S</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>6</td>\n",
       "      <td>0</td>\n",
       "      <td>3</td>\n",
       "      <td>Moran, Mr. James</td>\n",
       "      <td>male</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>330877</td>\n",
       "      <td>8.4583</td>\n",
       "      <td>NaN</td>\n",
       "      <td>Q</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>7</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>McCarthy, Mr. Timothy J</td>\n",
       "      <td>male</td>\n",
       "      <td>54.0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>17463</td>\n",
       "      <td>51.8625</td>\n",
       "      <td>E46</td>\n",
       "      <td>S</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>8</td>\n",
       "      <td>0</td>\n",
       "      <td>3</td>\n",
       "      <td>Palsson, Master. Gosta Leonard</td>\n",
       "      <td>male</td>\n",
       "      <td>2.0</td>\n",
       "      <td>3</td>\n",
       "      <td>1</td>\n",
       "      <td>349909</td>\n",
       "      <td>21.0750</td>\n",
       "      <td>NaN</td>\n",
       "      <td>S</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8</th>\n",
       "      <td>9</td>\n",
       "      <td>1</td>\n",
       "      <td>3</td>\n",
       "      <td>Johnson, Mrs. Oscar W (Elisabeth Vilhelmina Berg)</td>\n",
       "      <td>female</td>\n",
       "      <td>27.0</td>\n",
       "      <td>0</td>\n",
       "      <td>2</td>\n",
       "      <td>347742</td>\n",
       "      <td>11.1333</td>\n",
       "      <td>NaN</td>\n",
       "      <td>S</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9</th>\n",
       "      <td>10</td>\n",
       "      <td>1</td>\n",
       "      <td>2</td>\n",
       "      <td>Nasser, Mrs. Nicholas (Adele Achem)</td>\n",
       "      <td>female</td>\n",
       "      <td>14.0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>237736</td>\n",
       "      <td>30.0708</td>\n",
       "      <td>NaN</td>\n",
       "      <td>C</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   PassengerId  Survived  Pclass  \\\n",
       "0            1         0       3   \n",
       "1            2         1       1   \n",
       "2            3         1       3   \n",
       "3            4         1       1   \n",
       "4            5         0       3   \n",
       "5            6         0       3   \n",
       "6            7         0       1   \n",
       "7            8         0       3   \n",
       "8            9         1       3   \n",
       "9           10         1       2   \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",
       "5                                   Moran, Mr. James    male   NaN      0   \n",
       "6                            McCarthy, Mr. Timothy J    male  54.0      0   \n",
       "7                     Palsson, Master. Gosta Leonard    male   2.0      3   \n",
       "8  Johnson, Mrs. Oscar W (Elisabeth Vilhelmina Berg)  female  27.0      0   \n",
       "9                Nasser, Mrs. Nicholas (Adele Achem)  female  14.0      1   \n",
       "\n",
       "   Parch            Ticket     Fare Cabin Embarked  Under18  \n",
       "0      0         A/5 21171   7.2500   NaN        S        0  \n",
       "1      0          PC 17599  71.2833   C85        C        0  \n",
       "2      0  STON/O2. 3101282   7.9250   NaN        S        0  \n",
       "3      0            113803  53.1000  C123        S        0  \n",
       "4      0            373450   8.0500   NaN        S        0  \n",
       "5      0            330877   8.4583   NaN        Q        0  \n",
       "6      0             17463  51.8625   E46        S        0  \n",
       "7      1            349909  21.0750   NaN        S        1  \n",
       "8      2            347742  11.1333   NaN        S        0  \n",
       "9      0            237736  30.0708   NaN        C        1  "
      ]
     },
     "execution_count": 636,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data['Under18'] = np.where(data['Age'] < 18 , 1, 0)\n",
    "data.head(10)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Binary Features - Using a combination of other columns"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Next, let's make a column that reflects whether the passenger is travelling alone or with any family."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 637,
   "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>Under18</th>\n",
       "      <th>TravellingWithFamily</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>3</td>\n",
       "      <td>Braund, Mr. Owen Harris</td>\n",
       "      <td>male</td>\n",
       "      <td>22.0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>A/5 21171</td>\n",
       "      <td>7.2500</td>\n",
       "      <td>NaN</td>\n",
       "      <td>S</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>2</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>Cumings, Mrs. John Bradley (Florence Briggs Th...</td>\n",
       "      <td>female</td>\n",
       "      <td>38.0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>PC 17599</td>\n",
       "      <td>71.2833</td>\n",
       "      <td>C85</td>\n",
       "      <td>C</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>3</td>\n",
       "      <td>1</td>\n",
       "      <td>3</td>\n",
       "      <td>Heikkinen, Miss. Laina</td>\n",
       "      <td>female</td>\n",
       "      <td>26.0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>STON/O2. 3101282</td>\n",
       "      <td>7.9250</td>\n",
       "      <td>NaN</td>\n",
       "      <td>S</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>4</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>Futrelle, Mrs. Jacques Heath (Lily May Peel)</td>\n",
       "      <td>female</td>\n",
       "      <td>35.0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>113803</td>\n",
       "      <td>53.1000</td>\n",
       "      <td>C123</td>\n",
       "      <td>S</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>5</td>\n",
       "      <td>0</td>\n",
       "      <td>3</td>\n",
       "      <td>Allen, Mr. William Henry</td>\n",
       "      <td>male</td>\n",
       "      <td>35.0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>373450</td>\n",
       "      <td>8.0500</td>\n",
       "      <td>NaN</td>\n",
       "      <td>S</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>6</td>\n",
       "      <td>0</td>\n",
       "      <td>3</td>\n",
       "      <td>Moran, Mr. James</td>\n",
       "      <td>male</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>330877</td>\n",
       "      <td>8.4583</td>\n",
       "      <td>NaN</td>\n",
       "      <td>Q</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>7</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>McCarthy, Mr. Timothy J</td>\n",
       "      <td>male</td>\n",
       "      <td>54.0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>17463</td>\n",
       "      <td>51.8625</td>\n",
       "      <td>E46</td>\n",
       "      <td>S</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>8</td>\n",
       "      <td>0</td>\n",
       "      <td>3</td>\n",
       "      <td>Palsson, Master. Gosta Leonard</td>\n",
       "      <td>male</td>\n",
       "      <td>2.0</td>\n",
       "      <td>3</td>\n",
       "      <td>1</td>\n",
       "      <td>349909</td>\n",
       "      <td>21.0750</td>\n",
       "      <td>NaN</td>\n",
       "      <td>S</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8</th>\n",
       "      <td>9</td>\n",
       "      <td>1</td>\n",
       "      <td>3</td>\n",
       "      <td>Johnson, Mrs. Oscar W (Elisabeth Vilhelmina Berg)</td>\n",
       "      <td>female</td>\n",
       "      <td>27.0</td>\n",
       "      <td>0</td>\n",
       "      <td>2</td>\n",
       "      <td>347742</td>\n",
       "      <td>11.1333</td>\n",
       "      <td>NaN</td>\n",
       "      <td>S</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9</th>\n",
       "      <td>10</td>\n",
       "      <td>1</td>\n",
       "      <td>2</td>\n",
       "      <td>Nasser, Mrs. Nicholas (Adele Achem)</td>\n",
       "      <td>female</td>\n",
       "      <td>14.0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>237736</td>\n",
       "      <td>30.0708</td>\n",
       "      <td>NaN</td>\n",
       "      <td>C</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   PassengerId  Survived  Pclass  \\\n",
       "0            1         0       3   \n",
       "1            2         1       1   \n",
       "2            3         1       3   \n",
       "3            4         1       1   \n",
       "4            5         0       3   \n",
       "5            6         0       3   \n",
       "6            7         0       1   \n",
       "7            8         0       3   \n",
       "8            9         1       3   \n",
       "9           10         1       2   \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",
       "5                                   Moran, Mr. James    male   NaN      0   \n",
       "6                            McCarthy, Mr. Timothy J    male  54.0      0   \n",
       "7                     Palsson, Master. Gosta Leonard    male   2.0      3   \n",
       "8  Johnson, Mrs. Oscar W (Elisabeth Vilhelmina Berg)  female  27.0      0   \n",
       "9                Nasser, Mrs. Nicholas (Adele Achem)  female  14.0      1   \n",
       "\n",
       "   Parch            Ticket     Fare Cabin Embarked  Under18  \\\n",
       "0      0         A/5 21171   7.2500   NaN        S        0   \n",
       "1      0          PC 17599  71.2833   C85        C        0   \n",
       "2      0  STON/O2. 3101282   7.9250   NaN        S        0   \n",
       "3      0            113803  53.1000  C123        S        0   \n",
       "4      0            373450   8.0500   NaN        S        0   \n",
       "5      0            330877   8.4583   NaN        Q        0   \n",
       "6      0             17463  51.8625   E46        S        0   \n",
       "7      1            349909  21.0750   NaN        S        1   \n",
       "8      2            347742  11.1333   NaN        S        0   \n",
       "9      0            237736  30.0708   NaN        C        1   \n",
       "\n",
       "   TravellingWithFamily  \n",
       "0                     1  \n",
       "1                     1  \n",
       "2                     0  \n",
       "3                     1  \n",
       "4                     0  \n",
       "5                     0  \n",
       "6                     0  \n",
       "7                     1  \n",
       "8                     1  \n",
       "9                     1  "
      ]
     },
     "execution_count": 637,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data['TravellingWithFamily'] = np.where(\n",
    "    (data['SibSp'] + data['Parch']) >= 1, # Condition\n",
    "    1, # Value if True\n",
    "    0 # Value if False\n",
    "    )\n",
    "\n",
    "data.head(10)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Features from Complex String Patterns - Regular Expressions"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Now let's extract groups of titles from the name column. \n",
    "\n",
    "We could extract each individual title, but if there are a lot, this could lead to a very high number of columns when we go to one-hot encode our data. What we want to mainly explore from titles is whether people were noble, general populace, or young. \n",
    "\n",
    "The easiest way to do this here is to use something called **Regular Expressions**.\n",
    "\n",
    "Regular expressions allow us to match varying strings, like telephone numbers that may have different spacings or numbers of digits. We won't be using them to their full potential here! But they have the benefit of being usable with the `contains` method to allow us to pass a series of strings we would like to match with.\n",
    "\n",
    "First, let's import the `re` library."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 638,
   "metadata": {},
   "outputs": [],
   "source": [
    "import re"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Before turning this into a column, let's explore how we would match strings for a couple of categories of passengers."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Let's start with looking at upper-class titles."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Here, we're going to filter to rows where the 'name' column contains any one of\n",
    "\n",
    "- Col.\n",
    "- Capt.\n",
    "- Don.\n",
    "- Countess.\n",
    "- Dr.\n",
    "- Lady.\n",
    "- Sir.\n",
    "- Major.\n",
    "- Rev.\n",
    "- Jonkheer.\n",
    "- Dona.\n",
    "\n",
    "We want to ensure we look out for a `.` after these as in this dataset, a `.` is consistently used after a title. This is handy for us! \n",
    "\n",
    "However, in regex, a `.` has a special meaning - so we need to pass a `\\` before it to tell the regular expression we're looking for an actual full stop. \n",
    "\n",
    "We then use the pipe `|` to denote `or`."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 639,
   "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>Under18</th>\n",
       "      <th>TravellingWithFamily</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>30</th>\n",
       "      <td>31</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>Uruchurtu, Don. Manuel E</td>\n",
       "      <td>male</td>\n",
       "      <td>40.0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>PC 17601</td>\n",
       "      <td>27.7208</td>\n",
       "      <td>NaN</td>\n",
       "      <td>C</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>149</th>\n",
       "      <td>150</td>\n",
       "      <td>0</td>\n",
       "      <td>2</td>\n",
       "      <td>Byles, Rev. Thomas Roussel Davids</td>\n",
       "      <td>male</td>\n",
       "      <td>42.0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>244310</td>\n",
       "      <td>13.0000</td>\n",
       "      <td>NaN</td>\n",
       "      <td>S</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>150</th>\n",
       "      <td>151</td>\n",
       "      <td>0</td>\n",
       "      <td>2</td>\n",
       "      <td>Bateman, Rev. Robert James</td>\n",
       "      <td>male</td>\n",
       "      <td>51.0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>S.O.P. 1166</td>\n",
       "      <td>12.5250</td>\n",
       "      <td>NaN</td>\n",
       "      <td>S</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>245</th>\n",
       "      <td>246</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>Minahan, Dr. William Edward</td>\n",
       "      <td>male</td>\n",
       "      <td>44.0</td>\n",
       "      <td>2</td>\n",
       "      <td>0</td>\n",
       "      <td>19928</td>\n",
       "      <td>90.0000</td>\n",
       "      <td>C78</td>\n",
       "      <td>Q</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>249</th>\n",
       "      <td>250</td>\n",
       "      <td>0</td>\n",
       "      <td>2</td>\n",
       "      <td>Carter, Rev. Ernest Courtenay</td>\n",
       "      <td>male</td>\n",
       "      <td>54.0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>244252</td>\n",
       "      <td>26.0000</td>\n",
       "      <td>NaN</td>\n",
       "      <td>S</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>317</th>\n",
       "      <td>318</td>\n",
       "      <td>0</td>\n",
       "      <td>2</td>\n",
       "      <td>Moraweck, Dr. Ernest</td>\n",
       "      <td>male</td>\n",
       "      <td>54.0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>29011</td>\n",
       "      <td>14.0000</td>\n",
       "      <td>NaN</td>\n",
       "      <td>S</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>398</th>\n",
       "      <td>399</td>\n",
       "      <td>0</td>\n",
       "      <td>2</td>\n",
       "      <td>Pain, Dr. Alfred</td>\n",
       "      <td>male</td>\n",
       "      <td>23.0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>244278</td>\n",
       "      <td>10.5000</td>\n",
       "      <td>NaN</td>\n",
       "      <td>S</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>449</th>\n",
       "      <td>450</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>Peuchen, Major. Arthur Godfrey</td>\n",
       "      <td>male</td>\n",
       "      <td>52.0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>113786</td>\n",
       "      <td>30.5000</td>\n",
       "      <td>C104</td>\n",
       "      <td>S</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>536</th>\n",
       "      <td>537</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>Butt, Major. Archibald Willingham</td>\n",
       "      <td>male</td>\n",
       "      <td>45.0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>113050</td>\n",
       "      <td>26.5500</td>\n",
       "      <td>B38</td>\n",
       "      <td>S</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>556</th>\n",
       "      <td>557</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>Duff Gordon, Lady. (Lucille Christiana Sutherl...</td>\n",
       "      <td>female</td>\n",
       "      <td>48.0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>11755</td>\n",
       "      <td>39.6000</td>\n",
       "      <td>A16</td>\n",
       "      <td>C</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>599</th>\n",
       "      <td>600</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>Duff Gordon, Sir. Cosmo Edmund (\"Mr Morgan\")</td>\n",
       "      <td>male</td>\n",
       "      <td>49.0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>PC 17485</td>\n",
       "      <td>56.9292</td>\n",
       "      <td>A20</td>\n",
       "      <td>C</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>626</th>\n",
       "      <td>627</td>\n",
       "      <td>0</td>\n",
       "      <td>2</td>\n",
       "      <td>Kirkland, Rev. Charles Leonard</td>\n",
       "      <td>male</td>\n",
       "      <td>57.0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>219533</td>\n",
       "      <td>12.3500</td>\n",
       "      <td>NaN</td>\n",
       "      <td>Q</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>632</th>\n",
       "      <td>633</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>Stahelin-Maeglin, Dr. Max</td>\n",
       "      <td>male</td>\n",
       "      <td>32.0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>13214</td>\n",
       "      <td>30.5000</td>\n",
       "      <td>B50</td>\n",
       "      <td>C</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>647</th>\n",
       "      <td>648</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>Simonius-Blumer, Col. Oberst Alfons</td>\n",
       "      <td>male</td>\n",
       "      <td>56.0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>13213</td>\n",
       "      <td>35.5000</td>\n",
       "      <td>A26</td>\n",
       "      <td>C</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>660</th>\n",
       "      <td>661</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>Frauenthal, Dr. Henry William</td>\n",
       "      <td>male</td>\n",
       "      <td>50.0</td>\n",
       "      <td>2</td>\n",
       "      <td>0</td>\n",
       "      <td>PC 17611</td>\n",
       "      <td>133.6500</td>\n",
       "      <td>NaN</td>\n",
       "      <td>S</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>694</th>\n",
       "      <td>695</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>Weir, Col. John</td>\n",
       "      <td>male</td>\n",
       "      <td>60.0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>113800</td>\n",
       "      <td>26.5500</td>\n",
       "      <td>NaN</td>\n",
       "      <td>S</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>745</th>\n",
       "      <td>746</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>Crosby, Capt. Edward Gifford</td>\n",
       "      <td>male</td>\n",
       "      <td>70.0</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>WE/P 5735</td>\n",
       "      <td>71.0000</td>\n",
       "      <td>B22</td>\n",
       "      <td>S</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>759</th>\n",
       "      <td>760</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>Rothes, the Countess. of (Lucy Noel Martha Dye...</td>\n",
       "      <td>female</td>\n",
       "      <td>33.0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>110152</td>\n",
       "      <td>86.5000</td>\n",
       "      <td>B77</td>\n",
       "      <td>S</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>766</th>\n",
       "      <td>767</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>Brewe, Dr. Arthur Jackson</td>\n",
       "      <td>male</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>112379</td>\n",
       "      <td>39.6000</td>\n",
       "      <td>NaN</td>\n",
       "      <td>C</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>796</th>\n",
       "      <td>797</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>Leader, Dr. Alice (Farnham)</td>\n",
       "      <td>female</td>\n",
       "      <td>49.0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>17465</td>\n",
       "      <td>25.9292</td>\n",
       "      <td>D17</td>\n",
       "      <td>S</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>822</th>\n",
       "      <td>823</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>Reuchlin, Jonkheer. John George</td>\n",
       "      <td>male</td>\n",
       "      <td>38.0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>19972</td>\n",
       "      <td>0.0000</td>\n",
       "      <td>NaN</td>\n",
       "      <td>S</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>848</th>\n",
       "      <td>849</td>\n",
       "      <td>0</td>\n",
       "      <td>2</td>\n",
       "      <td>Harper, Rev. John</td>\n",
       "      <td>male</td>\n",
       "      <td>28.0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>248727</td>\n",
       "      <td>33.0000</td>\n",
       "      <td>NaN</td>\n",
       "      <td>S</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>886</th>\n",
       "      <td>887</td>\n",
       "      <td>0</td>\n",
       "      <td>2</td>\n",
       "      <td>Montvila, Rev. Juozas</td>\n",
       "      <td>male</td>\n",
       "      <td>27.0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>211536</td>\n",
       "      <td>13.0000</td>\n",
       "      <td>NaN</td>\n",
       "      <td>S</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "     PassengerId  Survived  Pclass  \\\n",
       "30            31         0       1   \n",
       "149          150         0       2   \n",
       "150          151         0       2   \n",
       "245          246         0       1   \n",
       "249          250         0       2   \n",
       "317          318         0       2   \n",
       "398          399         0       2   \n",
       "449          450         1       1   \n",
       "536          537         0       1   \n",
       "556          557         1       1   \n",
       "599          600         1       1   \n",
       "626          627         0       2   \n",
       "632          633         1       1   \n",
       "647          648         1       1   \n",
       "660          661         1       1   \n",
       "694          695         0       1   \n",
       "745          746         0       1   \n",
       "759          760         1       1   \n",
       "766          767         0       1   \n",
       "796          797         1       1   \n",
       "822          823         0       1   \n",
       "848          849         0       2   \n",
       "886          887         0       2   \n",
       "\n",
       "                                                  Name     Sex   Age  SibSp  \\\n",
       "30                            Uruchurtu, Don. Manuel E    male  40.0      0   \n",
       "149                  Byles, Rev. Thomas Roussel Davids    male  42.0      0   \n",
       "150                         Bateman, Rev. Robert James    male  51.0      0   \n",
       "245                        Minahan, Dr. William Edward    male  44.0      2   \n",
       "249                      Carter, Rev. Ernest Courtenay    male  54.0      1   \n",
       "317                               Moraweck, Dr. Ernest    male  54.0      0   \n",
       "398                                   Pain, Dr. Alfred    male  23.0      0   \n",
       "449                     Peuchen, Major. Arthur Godfrey    male  52.0      0   \n",
       "536                  Butt, Major. Archibald Willingham    male  45.0      0   \n",
       "556  Duff Gordon, Lady. (Lucille Christiana Sutherl...  female  48.0      1   \n",
       "599       Duff Gordon, Sir. Cosmo Edmund (\"Mr Morgan\")    male  49.0      1   \n",
       "626                     Kirkland, Rev. Charles Leonard    male  57.0      0   \n",
       "632                          Stahelin-Maeglin, Dr. Max    male  32.0      0   \n",
       "647                Simonius-Blumer, Col. Oberst Alfons    male  56.0      0   \n",
       "660                      Frauenthal, Dr. Henry William    male  50.0      2   \n",
       "694                                    Weir, Col. John    male  60.0      0   \n",
       "745                       Crosby, Capt. Edward Gifford    male  70.0      1   \n",
       "759  Rothes, the Countess. of (Lucy Noel Martha Dye...  female  33.0      0   \n",
       "766                          Brewe, Dr. Arthur Jackson    male   NaN      0   \n",
       "796                        Leader, Dr. Alice (Farnham)  female  49.0      0   \n",
       "822                    Reuchlin, Jonkheer. John George    male  38.0      0   \n",
       "848                                  Harper, Rev. John    male  28.0      0   \n",
       "886                              Montvila, Rev. Juozas    male  27.0      0   \n",
       "\n",
       "     Parch       Ticket      Fare Cabin Embarked  Under18  \\\n",
       "30       0     PC 17601   27.7208   NaN        C        0   \n",
       "149      0       244310   13.0000   NaN        S        0   \n",
       "150      0  S.O.P. 1166   12.5250   NaN        S        0   \n",
       "245      0        19928   90.0000   C78        Q        0   \n",
       "249      0       244252   26.0000   NaN        S        0   \n",
       "317      0        29011   14.0000   NaN        S        0   \n",
       "398      0       244278   10.5000   NaN        S        0   \n",
       "449      0       113786   30.5000  C104        S        0   \n",
       "536      0       113050   26.5500   B38        S        0   \n",
       "556      0        11755   39.6000   A16        C        0   \n",
       "599      0     PC 17485   56.9292   A20        C        0   \n",
       "626      0       219533   12.3500   NaN        Q        0   \n",
       "632      0        13214   30.5000   B50        C        0   \n",
       "647      0        13213   35.5000   A26        C        0   \n",
       "660      0     PC 17611  133.6500   NaN        S        0   \n",
       "694      0       113800   26.5500   NaN        S        0   \n",
       "745      1    WE/P 5735   71.0000   B22        S        0   \n",
       "759      0       110152   86.5000   B77        S        0   \n",
       "766      0       112379   39.6000   NaN        C        0   \n",
       "796      0        17465   25.9292   D17        S        0   \n",
       "822      0        19972    0.0000   NaN        S        0   \n",
       "848      1       248727   33.0000   NaN        S        0   \n",
       "886      0       211536   13.0000   NaN        S        0   \n",
       "\n",
       "     TravellingWithFamily  \n",
       "30                      0  \n",
       "149                     0  \n",
       "150                     0  \n",
       "245                     1  \n",
       "249                     1  \n",
       "317                     0  \n",
       "398                     0  \n",
       "449                     0  \n",
       "536                     0  \n",
       "556                     1  \n",
       "599                     1  \n",
       "626                     0  \n",
       "632                     0  \n",
       "647                     0  \n",
       "660                     1  \n",
       "694                     0  \n",
       "745                     1  \n",
       "759                     0  \n",
       "766                     0  \n",
       "796                     0  \n",
       "822                     0  \n",
       "848                     1  \n",
       "886                     0  "
      ]
     },
     "execution_count": 639,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data[data['Name'].str.contains(r\"Col\\.|Capt\\.|Don\\.|Countess\\.|Dr\\.|Lady\\.|Sir\\.|Major\\.|Rev\\.|Jonkheer\\.|Dona\\.\")]"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Let's repeat this to look at rows containing other groupings of titles."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "What about 'regular' class people?"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 640,
   "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>Under18</th>\n",
       "      <th>TravellingWithFamily</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>3</td>\n",
       "      <td>Braund, Mr. Owen Harris</td>\n",
       "      <td>male</td>\n",
       "      <td>22.0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>A/5 21171</td>\n",
       "      <td>7.2500</td>\n",
       "      <td>NaN</td>\n",
       "      <td>S</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>2</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>Cumings, Mrs. John Bradley (Florence Briggs Th...</td>\n",
       "      <td>female</td>\n",
       "      <td>38.0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>PC 17599</td>\n",
       "      <td>71.2833</td>\n",
       "      <td>C85</td>\n",
       "      <td>C</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>4</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>Futrelle, Mrs. Jacques Heath (Lily May Peel)</td>\n",
       "      <td>female</td>\n",
       "      <td>35.0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>113803</td>\n",
       "      <td>53.1000</td>\n",
       "      <td>C123</td>\n",
       "      <td>S</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>5</td>\n",
       "      <td>0</td>\n",
       "      <td>3</td>\n",
       "      <td>Allen, Mr. William Henry</td>\n",
       "      <td>male</td>\n",
       "      <td>35.0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>373450</td>\n",
       "      <td>8.0500</td>\n",
       "      <td>NaN</td>\n",
       "      <td>S</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>6</td>\n",
       "      <td>0</td>\n",
       "      <td>3</td>\n",
       "      <td>Moran, Mr. James</td>\n",
       "      <td>male</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>330877</td>\n",
       "      <td>8.4583</td>\n",
       "      <td>NaN</td>\n",
       "      <td>Q</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>883</th>\n",
       "      <td>884</td>\n",
       "      <td>0</td>\n",
       "      <td>2</td>\n",
       "      <td>Banfield, Mr. Frederick James</td>\n",
       "      <td>male</td>\n",
       "      <td>28.0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>C.A./SOTON 34068</td>\n",
       "      <td>10.5000</td>\n",
       "      <td>NaN</td>\n",
       "      <td>S</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>884</th>\n",
       "      <td>885</td>\n",
       "      <td>0</td>\n",
       "      <td>3</td>\n",
       "      <td>Sutehall, Mr. Henry Jr</td>\n",
       "      <td>male</td>\n",
       "      <td>25.0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>SOTON/OQ 392076</td>\n",
       "      <td>7.0500</td>\n",
       "      <td>NaN</td>\n",
       "      <td>S</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>885</th>\n",
       "      <td>886</td>\n",
       "      <td>0</td>\n",
       "      <td>3</td>\n",
       "      <td>Rice, Mrs. William (Margaret Norton)</td>\n",
       "      <td>female</td>\n",
       "      <td>39.0</td>\n",
       "      <td>0</td>\n",
       "      <td>5</td>\n",
       "      <td>382652</td>\n",
       "      <td>29.1250</td>\n",
       "      <td>NaN</td>\n",
       "      <td>Q</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>889</th>\n",
       "      <td>890</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>Behr, Mr. Karl Howell</td>\n",
       "      <td>male</td>\n",
       "      <td>26.0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>111369</td>\n",
       "      <td>30.0000</td>\n",
       "      <td>C148</td>\n",
       "      <td>C</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>890</th>\n",
       "      <td>891</td>\n",
       "      <td>0</td>\n",
       "      <td>3</td>\n",
       "      <td>Dooley, Mr. Patrick</td>\n",
       "      <td>male</td>\n",
       "      <td>32.0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>370376</td>\n",
       "      <td>7.7500</td>\n",
       "      <td>NaN</td>\n",
       "      <td>Q</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>644 rows × 14 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "     PassengerId  Survived  Pclass  \\\n",
       "0              1         0       3   \n",
       "1              2         1       1   \n",
       "3              4         1       1   \n",
       "4              5         0       3   \n",
       "5              6         0       3   \n",
       "..           ...       ...     ...   \n",
       "883          884         0       2   \n",
       "884          885         0       3   \n",
       "885          886         0       3   \n",
       "889          890         1       1   \n",
       "890          891         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",
       "3         Futrelle, Mrs. Jacques Heath (Lily May Peel)  female  35.0      1   \n",
       "4                             Allen, Mr. William Henry    male  35.0      0   \n",
       "5                                     Moran, Mr. James    male   NaN      0   \n",
       "..                                                 ...     ...   ...    ...   \n",
       "883                      Banfield, Mr. Frederick James    male  28.0      0   \n",
       "884                             Sutehall, Mr. Henry Jr    male  25.0      0   \n",
       "885               Rice, Mrs. William (Margaret Norton)  female  39.0      0   \n",
       "889                              Behr, Mr. Karl Howell    male  26.0      0   \n",
       "890                                Dooley, Mr. Patrick    male  32.0      0   \n",
       "\n",
       "     Parch            Ticket     Fare Cabin Embarked  Under18  \\\n",
       "0        0         A/5 21171   7.2500   NaN        S        0   \n",
       "1        0          PC 17599  71.2833   C85        C        0   \n",
       "3        0            113803  53.1000  C123        S        0   \n",
       "4        0            373450   8.0500   NaN        S        0   \n",
       "5        0            330877   8.4583   NaN        Q        0   \n",
       "..     ...               ...      ...   ...      ...      ...   \n",
       "883      0  C.A./SOTON 34068  10.5000   NaN        S        0   \n",
       "884      0   SOTON/OQ 392076   7.0500   NaN        S        0   \n",
       "885      5            382652  29.1250   NaN        Q        0   \n",
       "889      0            111369  30.0000  C148        C        0   \n",
       "890      0            370376   7.7500   NaN        Q        0   \n",
       "\n",
       "     TravellingWithFamily  \n",
       "0                       1  \n",
       "1                       1  \n",
       "3                       1  \n",
       "4                       0  \n",
       "5                       0  \n",
       "..                    ...  \n",
       "883                     0  \n",
       "884                     0  \n",
       "885                     1  \n",
       "889                     0  \n",
       "890                     0  \n",
       "\n",
       "[644 rows x 14 columns]"
      ]
     },
     "execution_count": 640,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data[data['Name'].str.contains(r\"Mrs\\.|Mlle\\.|Mr\\.|Mons\\.\")]"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Now let's look at instances relating to children."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 641,
   "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>Under18</th>\n",
       "      <th>TravellingWithFamily</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>3</td>\n",
       "      <td>1</td>\n",
       "      <td>3</td>\n",
       "      <td>Heikkinen, Miss. Laina</td>\n",
       "      <td>female</td>\n",
       "      <td>26.0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>STON/O2. 3101282</td>\n",
       "      <td>7.9250</td>\n",
       "      <td>NaN</td>\n",
       "      <td>S</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>8</td>\n",
       "      <td>0</td>\n",
       "      <td>3</td>\n",
       "      <td>Palsson, Master. Gosta Leonard</td>\n",
       "      <td>male</td>\n",
       "      <td>2.0</td>\n",
       "      <td>3</td>\n",
       "      <td>1</td>\n",
       "      <td>349909</td>\n",
       "      <td>21.0750</td>\n",
       "      <td>NaN</td>\n",
       "      <td>S</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>10</th>\n",
       "      <td>11</td>\n",
       "      <td>1</td>\n",
       "      <td>3</td>\n",
       "      <td>Sandstrom, Miss. Marguerite Rut</td>\n",
       "      <td>female</td>\n",
       "      <td>4.0</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>PP 9549</td>\n",
       "      <td>16.7000</td>\n",
       "      <td>G6</td>\n",
       "      <td>S</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>11</th>\n",
       "      <td>12</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>Bonnell, Miss. Elizabeth</td>\n",
       "      <td>female</td>\n",
       "      <td>58.0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>113783</td>\n",
       "      <td>26.5500</td>\n",
       "      <td>C103</td>\n",
       "      <td>S</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>14</th>\n",
       "      <td>15</td>\n",
       "      <td>0</td>\n",
       "      <td>3</td>\n",
       "      <td>Vestrom, Miss. Hulda Amanda Adolfina</td>\n",
       "      <td>female</td>\n",
       "      <td>14.0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>350406</td>\n",
       "      <td>7.8542</td>\n",
       "      <td>NaN</td>\n",
       "      <td>S</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>869</th>\n",
       "      <td>870</td>\n",
       "      <td>1</td>\n",
       "      <td>3</td>\n",
       "      <td>Johnson, Master. Harold Theodor</td>\n",
       "      <td>male</td>\n",
       "      <td>4.0</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>347742</td>\n",
       "      <td>11.1333</td>\n",
       "      <td>NaN</td>\n",
       "      <td>S</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>875</th>\n",
       "      <td>876</td>\n",
       "      <td>1</td>\n",
       "      <td>3</td>\n",
       "      <td>Najib, Miss. Adele Kiamie \"Jane\"</td>\n",
       "      <td>female</td>\n",
       "      <td>15.0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>2667</td>\n",
       "      <td>7.2250</td>\n",
       "      <td>NaN</td>\n",
       "      <td>C</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>882</th>\n",
       "      <td>883</td>\n",
       "      <td>0</td>\n",
       "      <td>3</td>\n",
       "      <td>Dahlberg, Miss. Gerda Ulrika</td>\n",
       "      <td>female</td>\n",
       "      <td>22.0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>7552</td>\n",
       "      <td>10.5167</td>\n",
       "      <td>NaN</td>\n",
       "      <td>S</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>887</th>\n",
       "      <td>888</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>Graham, Miss. Margaret Edith</td>\n",
       "      <td>female</td>\n",
       "      <td>19.0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>112053</td>\n",
       "      <td>30.0000</td>\n",
       "      <td>B42</td>\n",
       "      <td>S</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>888</th>\n",
       "      <td>889</td>\n",
       "      <td>0</td>\n",
       "      <td>3</td>\n",
       "      <td>Johnston, Miss. Catherine Helen \"Carrie\"</td>\n",
       "      <td>female</td>\n",
       "      <td>NaN</td>\n",
       "      <td>1</td>\n",
       "      <td>2</td>\n",
       "      <td>W./C. 6607</td>\n",
       "      <td>23.4500</td>\n",
       "      <td>NaN</td>\n",
       "      <td>S</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>222 rows × 14 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "     PassengerId  Survived  Pclass                                      Name  \\\n",
       "2              3         1       3                    Heikkinen, Miss. Laina   \n",
       "7              8         0       3            Palsson, Master. Gosta Leonard   \n",
       "10            11         1       3           Sandstrom, Miss. Marguerite Rut   \n",
       "11            12         1       1                  Bonnell, Miss. Elizabeth   \n",
       "14            15         0       3      Vestrom, Miss. Hulda Amanda Adolfina   \n",
       "..           ...       ...     ...                                       ...   \n",
       "869          870         1       3           Johnson, Master. Harold Theodor   \n",
       "875          876         1       3          Najib, Miss. Adele Kiamie \"Jane\"   \n",
       "882          883         0       3              Dahlberg, Miss. Gerda Ulrika   \n",
       "887          888         1       1              Graham, Miss. Margaret Edith   \n",
       "888          889         0       3  Johnston, Miss. Catherine Helen \"Carrie\"   \n",
       "\n",
       "        Sex   Age  SibSp  Parch            Ticket     Fare Cabin Embarked  \\\n",
       "2    female  26.0      0      0  STON/O2. 3101282   7.9250   NaN        S   \n",
       "7      male   2.0      3      1            349909  21.0750   NaN        S   \n",
       "10   female   4.0      1      1           PP 9549  16.7000    G6        S   \n",
       "11   female  58.0      0      0            113783  26.5500  C103        S   \n",
       "14   female  14.0      0      0            350406   7.8542   NaN        S   \n",
       "..      ...   ...    ...    ...               ...      ...   ...      ...   \n",
       "869    male   4.0      1      1            347742  11.1333   NaN        S   \n",
       "875  female  15.0      0      0              2667   7.2250   NaN        C   \n",
       "882  female  22.0      0      0              7552  10.5167   NaN        S   \n",
       "887  female  19.0      0      0            112053  30.0000   B42        S   \n",
       "888  female   NaN      1      2        W./C. 6607  23.4500   NaN        S   \n",
       "\n",
       "     Under18  TravellingWithFamily  \n",
       "2          0                     0  \n",
       "7          1                     1  \n",
       "10         1                     1  \n",
       "11         0                     0  \n",
       "14         1                     0  \n",
       "..       ...                   ...  \n",
       "869        1                     1  \n",
       "875        1                     0  \n",
       "882        0                     0  \n",
       "887        0                     0  \n",
       "888        0                     1  \n",
       "\n",
       "[222 rows x 14 columns]"
      ]
     },
     "execution_count": 641,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data[data['Name'].str.contains(r\"Master\\.|Miss\\.\")]"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Let's put in all of the ones we've come up with so far and check we haven't missed anyone.\n",
    "\n",
    "We've used the `~` operator to filter our dataframe to instances where a match has NOT been found instead."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 642,
   "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>Under18</th>\n",
       "      <th>TravellingWithFamily</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>369</th>\n",
       "      <td>370</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>Aubart, Mme. Leontine Pauline</td>\n",
       "      <td>female</td>\n",
       "      <td>24.0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>PC 17477</td>\n",
       "      <td>69.3</td>\n",
       "      <td>B35</td>\n",
       "      <td>C</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>443</th>\n",
       "      <td>444</td>\n",
       "      <td>1</td>\n",
       "      <td>2</td>\n",
       "      <td>Reynaldo, Ms. Encarnacion</td>\n",
       "      <td>female</td>\n",
       "      <td>28.0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>230434</td>\n",
       "      <td>13.0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>S</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "     PassengerId  Survived  Pclass                           Name     Sex  \\\n",
       "369          370         1       1  Aubart, Mme. Leontine Pauline  female   \n",
       "443          444         1       2      Reynaldo, Ms. Encarnacion  female   \n",
       "\n",
       "      Age  SibSp  Parch    Ticket  Fare Cabin Embarked  Under18  \\\n",
       "369  24.0      0      0  PC 17477  69.3   B35        C        0   \n",
       "443  28.0      0      0    230434  13.0   NaN        S        0   \n",
       "\n",
       "     TravellingWithFamily  \n",
       "369                     0  \n",
       "443                     0  "
      ]
     },
     "execution_count": 642,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data[~data['Name'].str.contains(r\"Master\\.|Miss\\.|Mrs\\.|Mlle\\.|Mr\\.|Mons\\.|Col\\.|Capt\\.|Don\\.|Countess\\.|Dr\\.|Lady\\.|Sir\\.|Major\\.|Rev\\.|Jonkheer\\.|Dona\\.\")]"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "It looks like we've missed `Mme.` and `Ms.` so we'll make sure to add those into the 'Regular' status for our final query.\n",
    "\n",
    "### Creating the column\n",
    "\n",
    "Now we've explored and built our query, we can turn this into our final column.\n",
    "\n",
    "We're going to use `np.where` again - this does get a bit complex with longer queries!\n",
    "\n",
    "It effectively is structured like a big if-else clause.\n",
    "\n",
    "`np.where(condition, value_if_true, value_if_false)`\n",
    "\n",
    "We nest these by passing additional `np.where` statements in the `value if false` position.\n",
    "\n",
    "This may feel familiar if you've had to do a big if clause in Excel!\n",
    "\n",
    "(note that this would be a lot easier in a newer version of pandas - they released a 'case_when' function in v2.2 - those of you used to SQL may be familiar with that sort of thing)\n",
    "\n",
    "First, let's just run the statement and see the output.\n",
    "\n",
    "I've added in some extra complexity to look at the 'age' column as well for the first two instances.\n",
    "\n",
    "You may wish to indent your code differently for readability - the indentation won't have any impact on the functioning here."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 643,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array(['Regular', 'Regular', 'Unmarried Woman', 'Regular', 'Regular',\n",
       "       'Regular', 'Regular', 'Young', 'Regular', 'Regular', 'Young',\n",
       "       'Unmarried Woman', 'Regular', 'Regular', 'Young', 'Regular',\n",
       "       'Young', 'Regular', 'Regular', 'Regular', 'Regular', 'Regular',\n",
       "       'Young', 'Regular', 'Young', 'Regular', 'Regular', 'Regular',\n",
       "       'Unknown', 'Regular', 'Upper Class', 'Regular', 'Unknown',\n",
       "       'Regular', 'Regular', 'Regular', 'Regular', 'Regular',\n",
       "       'Unmarried Woman', 'Young', 'Regular', 'Regular', 'Regular',\n",
       "       'Young', 'Unmarried Woman', 'Regular', 'Regular', 'Unknown',\n",
       "       'Regular', 'Regular', 'Young', 'Regular', 'Regular', 'Regular',\n",
       "       'Regular', 'Regular', 'Unmarried Woman', 'Regular', 'Young',\n",
       "       'Young', 'Regular', 'Unmarried Woman', 'Regular', 'Young',\n",
       "       'Regular', 'Unknown', 'Regular', 'Regular', 'Young', 'Regular',\n",
       "       'Regular', 'Young', 'Regular', 'Regular', 'Regular', 'Regular',\n",
       "       'Regular', 'Regular', 'Young', 'Unmarried Woman', 'Regular',\n",
       "       'Regular', 'Unknown', 'Regular', 'Young', 'Regular', 'Regular',\n",
       "       'Regular', 'Unmarried Woman', 'Regular', 'Regular', 'Regular',\n",
       "       'Regular', 'Regular', 'Regular', 'Regular', 'Regular', 'Regular',\n",
       "       'Regular', 'Regular', 'Unmarried Woman', 'Regular', 'Regular',\n",
       "       'Regular', 'Regular', 'Regular', 'Unmarried Woman', 'Regular',\n",
       "       'Regular', 'Unknown', 'Regular', 'Young', 'Regular',\n",
       "       'Unmarried Woman', 'Young', 'Regular', 'Regular', 'Regular',\n",
       "       'Regular', 'Young', 'Regular', 'Regular', 'Regular',\n",
       "       'Unmarried Woman', 'Regular', 'Young', 'Regular', 'Regular',\n",
       "       'Unknown', 'Regular', 'Regular', 'Regular', 'Regular', 'Regular',\n",
       "       'Regular', 'Regular', 'Unmarried Woman', 'Regular', 'Regular',\n",
       "       'Regular', 'Regular', 'Unmarried Woman', 'Regular', 'Regular',\n",
       "       'Regular', 'Regular', 'Regular', 'Young', 'Regular', 'Upper Class',\n",
       "       'Upper Class', 'Regular', 'Regular', 'Regular', 'Regular',\n",
       "       'Regular', 'Young', 'Regular', 'Regular', 'Unknown', 'Regular',\n",
       "       'Regular', 'Regular', 'Regular', 'Young', 'Young', 'Regular',\n",
       "       'Regular', 'Regular', 'Regular', 'Regular', 'Young', 'Young',\n",
       "       'Regular', 'Regular', 'Regular', 'Unknown', 'Unmarried Woman',\n",
       "       'Regular', 'Regular', 'Unknown', 'Regular', 'Young', 'Young',\n",
       "       'Young', 'Regular', 'Regular', 'Regular', 'Regular', 'Regular',\n",
       "       'Regular', 'Regular', 'Unmarried Woman', 'Young', 'Regular',\n",
       "       'Unmarried Woman', 'Regular', 'Regular', 'Unknown',\n",
       "       'Unmarried Woman', 'Regular', 'Regular', 'Regular', 'Regular',\n",
       "       'Regular', 'Young', 'Regular', 'Regular', 'Young', 'Regular',\n",
       "       'Regular', 'Unmarried Woman', 'Regular', 'Regular', 'Regular',\n",
       "       'Unmarried Woman', 'Unmarried Woman', 'Regular', 'Unmarried Woman',\n",
       "       'Regular', 'Regular', 'Regular', 'Regular', 'Regular', 'Regular',\n",
       "       'Regular', 'Regular', 'Regular', 'Regular', 'Unknown', 'Regular',\n",
       "       'Regular', 'Regular', 'Young', 'Regular', 'Unknown', 'Regular',\n",
       "       'Young', 'Regular', 'Regular', 'Unknown', 'Unknown', 'Regular',\n",
       "       'Regular', 'Regular', 'Upper Class', 'Unmarried Woman', 'Regular',\n",
       "       'Regular', 'Upper Class', 'Regular', 'Regular', 'Regular',\n",
       "       'Regular', 'Regular', 'Regular', 'Regular', 'Unmarried Woman',\n",
       "       'Unmarried Woman', 'Regular', 'Regular', 'Young', 'Regular',\n",
       "       'Regular', 'Unknown', 'Regular', 'Regular', 'Regular', 'Regular',\n",
       "       'Unmarried Woman', 'Regular', 'Regular', 'Regular', 'Regular',\n",
       "       'Unknown', 'Unmarried Woman', 'Unmarried Woman', 'Regular',\n",
       "       'Young', 'Regular', 'Regular', 'Regular', 'Regular', 'Regular',\n",
       "       'Regular', 'Regular', 'Regular', 'Regular', 'Regular',\n",
       "       'Unmarried Woman', 'Unmarried Woman', 'Regular', 'Regular',\n",
       "       'Unmarried Woman', 'Regular', 'Regular', 'Regular', 'Young',\n",
       "       'Regular', 'Regular', 'Unknown', 'Regular', 'Regular', 'Unknown',\n",
       "       'Regular', 'Young', 'Unknown', 'Regular', 'Regular',\n",
       "       'Unmarried Woman', 'Unmarried Woman', 'Unmarried Woman', 'Regular',\n",
       "       'Regular', 'Regular', 'Unmarried Woman', 'Regular', 'Upper Class',\n",
       "       'Unmarried Woman', 'Regular', 'Regular', 'Regular',\n",
       "       'Unmarried Woman', 'Regular', 'Regular', 'Unmarried Woman',\n",
       "       'Regular', 'Regular', 'Regular', 'Young', 'Unknown', 'Regular',\n",
       "       'Regular', 'Regular', 'Regular', 'Regular', 'Regular',\n",
       "       'Unmarried Woman', 'Regular', 'Regular', 'Young',\n",
       "       'Unmarried Woman', 'Regular', 'Regular', 'Regular',\n",
       "       'Unmarried Woman', 'Unmarried Woman', 'Regular', 'Young',\n",
       "       'Regular', 'Regular', 'Regular', 'Regular', 'Regular', 'Regular',\n",
       "       'Regular', 'Unmarried Woman', 'Unmarried Woman', 'Unknown',\n",
       "       'Unknown', 'Regular', 'Regular', 'Regular', 'Regular', 'Regular',\n",
       "       'Regular', 'Regular', 'Regular', 'Unknown', 'Unknown', 'Regular',\n",
       "       'Regular', 'Regular', 'Regular', 'Young', 'Regular',\n",
       "       'Unmarried Woman', 'Regular', 'Regular', 'Regular',\n",
       "       'Unmarried Woman', 'Young', 'Regular', 'Regular', 'Regular',\n",
       "       'Regular', 'Young', 'Unmarried Woman', 'Regular', 'Young',\n",
       "       'Regular', 'Regular', 'Regular', 'Unmarried Woman', 'Regular',\n",
       "       'Regular', 'Unmarried Woman', 'Regular', 'Upper Class', 'Regular',\n",
       "       'Regular', 'Regular', 'Unmarried Woman', 'Regular',\n",
       "       'Unmarried Woman', 'Regular', 'Regular', 'Young', 'Regular',\n",
       "       'Unknown', 'Regular', 'Regular', 'Unmarried Woman', 'Regular',\n",
       "       'Regular', 'Regular', 'Regular', 'Unmarried Woman', 'Regular',\n",
       "       'Young', 'Regular', 'Regular', 'Regular', 'Regular', 'Regular',\n",
       "       'Regular', 'Regular', 'Unmarried Woman', 'Regular', 'Regular',\n",
       "       'Regular', 'Regular', 'Regular', 'Regular', 'Regular', 'Young',\n",
       "       'Unmarried Woman', 'Regular', 'Regular', 'Regular', 'Regular',\n",
       "       'Regular', 'Regular', 'Unknown', 'Regular', 'Young', 'Young',\n",
       "       'Regular', 'Young', 'Upper Class', 'Regular', 'Regular', 'Regular',\n",
       "       'Regular', 'Regular', 'Regular', 'Regular', 'Regular',\n",
       "       'Unmarried Woman', 'Regular', 'Regular', 'Regular', 'Regular',\n",
       "       'Regular', 'Regular', 'Regular', 'Regular', 'Regular', 'Regular',\n",
       "       'Young', 'Regular', 'Regular', 'Regular', 'Regular',\n",
       "       'Unmarried Woman', 'Regular', 'Regular', 'Regular', 'Regular',\n",
       "       'Young', 'Young', 'Regular', 'Regular', 'Regular', 'Regular',\n",
       "       'Unknown', 'Regular', 'Regular', 'Regular', 'Young', 'Regular',\n",
       "       'Regular', 'Regular', 'Regular', 'Regular', 'Regular',\n",
       "       'Unmarried Woman', 'Regular', 'Regular', 'Regular', 'Regular',\n",
       "       'Unmarried Woman', 'Unknown', 'Unmarried Woman', 'Young',\n",
       "       'Regular', 'Regular', 'Regular', 'Regular', 'Regular', 'Regular',\n",
       "       'Regular', 'Regular', 'Regular', 'Regular', 'Regular', 'Regular',\n",
       "       'Regular', 'Regular', 'Regular', 'Unmarried Woman', 'Regular',\n",
       "       'Regular', 'Regular', 'Regular', 'Regular', 'Unmarried Woman',\n",
       "       'Regular', 'Regular', 'Regular', 'Young', 'Regular', 'Regular',\n",
       "       'Regular', 'Unmarried Woman', 'Young', 'Upper Class',\n",
       "       'Unmarried Woman', 'Regular', 'Unmarried Woman', 'Unmarried Woman',\n",
       "       'Young', 'Young', 'Regular', 'Regular', 'Regular', 'Regular',\n",
       "       'Regular', 'Regular', 'Young', 'Regular', 'Regular', 'Regular',\n",
       "       'Regular', 'Unmarried Woman', 'Regular', 'Upper Class', 'Regular',\n",
       "       'Regular', 'Regular', 'Regular', 'Regular', 'Regular', 'Regular',\n",
       "       'Unknown', 'Regular', 'Regular', 'Regular', 'Regular', 'Regular',\n",
       "       'Regular', 'Regular', 'Regular', 'Unknown', 'Regular', 'Regular',\n",
       "       'Unmarried Woman', 'Regular', 'Regular', 'Regular',\n",
       "       'Unmarried Woman', 'Regular', 'Regular', 'Regular', 'Regular',\n",
       "       'Unmarried Woman', 'Regular', 'Regular', 'Regular', 'Regular',\n",
       "       'Regular', 'Regular', 'Regular', 'Unknown', 'Regular', 'Regular',\n",
       "       'Unknown', 'Regular', 'Regular', 'Upper Class', 'Regular',\n",
       "       'Regular', 'Regular', 'Regular', 'Regular', 'Regular', 'Regular',\n",
       "       'Regular', 'Regular', 'Unmarried Woman', 'Regular', 'Regular',\n",
       "       'Unknown', 'Regular', 'Regular', 'Unmarried Woman', 'Regular',\n",
       "       'Regular', 'Young', 'Regular', 'Regular', 'Regular', 'Regular',\n",
       "       'Regular', 'Regular', 'Regular', 'Upper Class', 'Unmarried Woman',\n",
       "       'Regular', 'Regular', 'Regular', 'Regular', 'Upper Class',\n",
       "       'Regular', 'Young', 'Unmarried Woman', 'Regular', 'Regular',\n",
       "       'Regular', 'Regular', 'Regular', 'Regular', 'Young', 'Regular',\n",
       "       'Young', 'Regular', 'Regular', 'Upper Class', 'Regular',\n",
       "       'Unmarried Woman', 'Regular', 'Unmarried Woman', 'Regular',\n",
       "       'Unknown', 'Unmarried Woman', 'Regular', 'Regular', 'Regular',\n",
       "       'Regular', 'Regular', 'Upper Class', 'Regular', 'Regular',\n",
       "       'Regular', 'Regular', 'Regular', 'Regular', 'Regular', 'Regular',\n",
       "       'Regular', 'Regular', 'Regular', 'Regular', 'Regular', 'Regular',\n",
       "       'Regular', 'Regular', 'Unmarried Woman', 'Regular', 'Regular',\n",
       "       'Unknown', 'Regular', 'Regular', 'Regular', 'Regular', 'Regular',\n",
       "       'Regular', 'Regular', 'Regular', 'Young', 'Regular', 'Young',\n",
       "       'Regular', 'Regular', 'Upper Class', 'Regular', 'Regular',\n",
       "       'Unknown', 'Regular', 'Regular', 'Regular', 'Regular',\n",
       "       'Unmarried Woman', 'Regular', 'Regular', 'Regular', 'Regular',\n",
       "       'Regular', 'Unmarried Woman', 'Unknown', 'Regular', 'Regular',\n",
       "       'Regular', 'Regular', 'Regular', 'Regular', 'Unmarried Woman',\n",
       "       'Unmarried Woman', 'Regular', 'Regular', 'Young', 'Regular',\n",
       "       'Regular', 'Regular', 'Regular', 'Regular', 'Regular', 'Unknown',\n",
       "       'Regular', 'Unmarried Woman', 'Unmarried Woman', 'Regular',\n",
       "       'Regular', 'Regular', 'Regular', 'Regular', 'Regular', 'Regular',\n",
       "       'Regular', 'Regular', 'Regular', 'Regular', 'Unmarried Woman',\n",
       "       'Regular', 'Regular', 'Upper Class', 'Regular', 'Unmarried Woman',\n",
       "       'Regular', 'Regular', 'Young', 'Young', 'Regular', 'Regular',\n",
       "       'Regular', 'Young', 'Regular', 'Regular', 'Regular', 'Upper Class',\n",
       "       'Regular', 'Regular', 'Regular', 'Regular', 'Regular', 'Regular',\n",
       "       'Upper Class', 'Unmarried Woman', 'Regular', 'Regular', 'Regular',\n",
       "       'Regular', 'Regular', 'Regular', 'Regular', 'Regular', 'Regular',\n",
       "       'Young', 'Regular', 'Regular', 'Young', 'Regular', 'Regular',\n",
       "       'Regular', 'Regular', 'Regular', 'Unmarried Woman', 'Young',\n",
       "       'Young', 'Regular', 'Regular', 'Regular', 'Unknown', 'Regular',\n",
       "       'Regular', 'Regular', 'Upper Class', 'Regular', 'Regular',\n",
       "       'Regular', 'Regular', 'Regular', 'Young', 'Young', 'Regular',\n",
       "       'Regular', 'Regular', 'Unmarried Woman', 'Regular', 'Regular',\n",
       "       'Regular', 'Regular', 'Regular', 'Young', 'Regular', 'Regular',\n",
       "       'Unmarried Woman', 'Regular', 'Regular', 'Young', 'Regular',\n",
       "       'Regular', 'Upper Class', 'Regular', 'Young', 'Regular', 'Regular',\n",
       "       'Young', 'Regular', 'Regular', 'Regular', 'Young', 'Regular',\n",
       "       'Regular', 'Regular', 'Unmarried Woman', 'Regular', 'Regular',\n",
       "       'Regular', 'Regular', 'Regular', 'Regular', 'Unmarried Woman',\n",
       "       'Regular', 'Regular', 'Regular', 'Regular', 'Regular',\n",
       "       'Upper Class', 'Regular', 'Young', 'Regular', 'Young', 'Young',\n",
       "       'Regular', 'Regular', 'Regular', 'Regular', 'Regular', 'Regular',\n",
       "       'Regular', 'Regular', 'Regular', 'Unknown', 'Regular', 'Regular',\n",
       "       'Unmarried Woman', 'Regular', 'Regular', 'Young', 'Regular',\n",
       "       'Regular', 'Regular', 'Regular', 'Regular', 'Young', 'Regular',\n",
       "       'Regular', 'Regular', 'Regular', 'Regular', 'Regular',\n",
       "       'Unmarried Woman', 'Regular', 'Regular', 'Regular', 'Upper Class',\n",
       "       'Unmarried Woman', 'Unknown', 'Regular', 'Regular'], dtype='<U15')"
      ]
     },
     "execution_count": 643,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "np.where(\n",
    "    (data['Name'].str.contains(r\"Master\\.|Miss\\.\")) & (data['Age'] <18), \"Young\",\n",
    "    np.where((data['Name'].str.contains(r\"Miss\\.\")) & (data['Age'] >=18), \"Unmarried Woman\",\n",
    "    np.where(data['Name'].str.contains(r\"Mrs\\.|Mlle\\.|Mr\\.|Mons\\.\"), \"Regular\",\n",
    "    np.where(data['Name'].str.contains(r\"Col\\.|Capt\\.|Don\\.|Countess\\.|Dr\\.|Lady\\.|Sir\\.|Major\\.|Rev\\.|Jonkheer\\.|Dona\\.\"), \"Upper Class\",\n",
    "    \"Unknown\"\n",
    "    )))\n",
    "    )"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "We can see it outputs a numpy array - effectively a big list - with the matched statement for each row."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Now let's assign that to a column in our dataframe and take a look at it in context."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 644,
   "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>Name</th>\n",
       "      <th>Age</th>\n",
       "      <th>Title</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>691</th>\n",
       "      <td>Karun, Miss. Manca</td>\n",
       "      <td>4.0</td>\n",
       "      <td>Young</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>194</th>\n",
       "      <td>Brown, Mrs. James Joseph (Margaret Tobin)</td>\n",
       "      <td>44.0</td>\n",
       "      <td>Regular</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>719</th>\n",
       "      <td>Johnson, Mr. Malkolm Joackim</td>\n",
       "      <td>33.0</td>\n",
       "      <td>Regular</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>805</th>\n",
       "      <td>Johansson, Mr. Karl Johan</td>\n",
       "      <td>31.0</td>\n",
       "      <td>Regular</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>125</th>\n",
       "      <td>Nicola-Yarred, Master. Elias</td>\n",
       "      <td>12.0</td>\n",
       "      <td>Young</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>441</th>\n",
       "      <td>Hampe, Mr. Leon</td>\n",
       "      <td>20.0</td>\n",
       "      <td>Regular</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>225</th>\n",
       "      <td>Berglund, Mr. Karl Ivar Sven</td>\n",
       "      <td>22.0</td>\n",
       "      <td>Regular</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>354</th>\n",
       "      <td>Yousif, Mr. Wazli</td>\n",
       "      <td>NaN</td>\n",
       "      <td>Regular</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>260</th>\n",
       "      <td>Smith, Mr. Thomas</td>\n",
       "      <td>NaN</td>\n",
       "      <td>Regular</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>409</th>\n",
       "      <td>Lefebre, Miss. Ida</td>\n",
       "      <td>NaN</td>\n",
       "      <td>Unknown</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>444</th>\n",
       "      <td>Johannesen-Bratthammer, Mr. Bernt</td>\n",
       "      <td>NaN</td>\n",
       "      <td>Regular</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>560</th>\n",
       "      <td>Morrow, Mr. Thomas Rowan</td>\n",
       "      <td>NaN</td>\n",
       "      <td>Regular</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>739</th>\n",
       "      <td>Nankoff, Mr. Minko</td>\n",
       "      <td>NaN</td>\n",
       "      <td>Regular</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>175</th>\n",
       "      <td>Klasen, Mr. Klas Albin</td>\n",
       "      <td>18.0</td>\n",
       "      <td>Regular</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>289</th>\n",
       "      <td>Connolly, Miss. Kate</td>\n",
       "      <td>22.0</td>\n",
       "      <td>Unmarried Woman</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                                          Name   Age            Title\n",
       "691                         Karun, Miss. Manca   4.0            Young\n",
       "194  Brown, Mrs. James Joseph (Margaret Tobin)  44.0          Regular\n",
       "719               Johnson, Mr. Malkolm Joackim  33.0          Regular\n",
       "805                  Johansson, Mr. Karl Johan  31.0          Regular\n",
       "125               Nicola-Yarred, Master. Elias  12.0            Young\n",
       "441                            Hampe, Mr. Leon  20.0          Regular\n",
       "225               Berglund, Mr. Karl Ivar Sven  22.0          Regular\n",
       "354                          Yousif, Mr. Wazli   NaN          Regular\n",
       "260                          Smith, Mr. Thomas   NaN          Regular\n",
       "409                         Lefebre, Miss. Ida   NaN          Unknown\n",
       "444          Johannesen-Bratthammer, Mr. Bernt   NaN          Regular\n",
       "560                   Morrow, Mr. Thomas Rowan   NaN          Regular\n",
       "739                         Nankoff, Mr. Minko   NaN          Regular\n",
       "175                     Klasen, Mr. Klas Albin  18.0          Regular\n",
       "289                       Connolly, Miss. Kate  22.0  Unmarried Woman"
      ]
     },
     "execution_count": 644,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data['Title'] = np.where(\n",
    "    (data['Name'].str.contains(r\"Master\\.|Miss\\.\")) & (data['Age'] <18), \"Young\",\n",
    "    np.where((data['Name'].str.contains(r\"Miss\\.\")) & (data['Age'] >=18), \"Unmarried Woman\",\n",
    "    np.where(data['Name'].str.contains(r\"Mrs\\.|Mlle\\.|Mr\\.|Mons\\.\"), \"Regular\",\n",
    "    np.where(data['Name'].str.contains(r\"Col\\.|Capt\\.|Don\\.|Countess\\.|Dr\\.|Lady\\.|Sir\\.|Major\\.|Rev\\.|Jonkheer\\.|Dona\\.\"), \"Upper Class\",\n",
    "    \"Unknown\"\n",
    "    ))))\n",
    "\n",
    "# Select just the name, age and title columns\n",
    "# Return a random sample of 15 rows\n",
    "data[['Name', 'Age', 'Title']].sample(15)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Finally, let's see how many people fall into each category."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 645,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Regular            644\n",
       "Unmarried Woman     95\n",
       "Young               87\n",
       "Unknown             42\n",
       "Upper Class         23\n",
       "Name: Title, dtype: int64"
      ]
     },
     "execution_count": 645,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data['Title'].value_counts()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Feature from Simple String Patterns - Getting the Deck from the Cabin Number"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Let's also work out which deck they were staying on based on the cabin number. \n",
    "\n",
    "Let's first look at a single cabin number from our dataframe."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 646,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "'C85'"
      ]
     },
     "execution_count": 646,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data['Cabin'][1]"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "On further inspection, it turns out that the first letter of the cabin string is **always** the deck number. \n",
    "\n",
    "We can just pull back the `.str` attribute in each case and pull out the 1st letter (in position 0 - remember Python counts from 0)."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 647,
   "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>Under18</th>\n",
       "      <th>TravellingWithFamily</th>\n",
       "      <th>Title</th>\n",
       "      <th>Deck</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>3</td>\n",
       "      <td>Braund, Mr. Owen Harris</td>\n",
       "      <td>male</td>\n",
       "      <td>22.0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>A/5 21171</td>\n",
       "      <td>7.2500</td>\n",
       "      <td>NaN</td>\n",
       "      <td>S</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>Regular</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>2</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>Cumings, Mrs. John Bradley (Florence Briggs Th...</td>\n",
       "      <td>female</td>\n",
       "      <td>38.0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>PC 17599</td>\n",
       "      <td>71.2833</td>\n",
       "      <td>C85</td>\n",
       "      <td>C</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>Regular</td>\n",
       "      <td>C</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>3</td>\n",
       "      <td>1</td>\n",
       "      <td>3</td>\n",
       "      <td>Heikkinen, Miss. Laina</td>\n",
       "      <td>female</td>\n",
       "      <td>26.0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>STON/O2. 3101282</td>\n",
       "      <td>7.9250</td>\n",
       "      <td>NaN</td>\n",
       "      <td>S</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>Unmarried Woman</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>4</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>Futrelle, Mrs. Jacques Heath (Lily May Peel)</td>\n",
       "      <td>female</td>\n",
       "      <td>35.0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>113803</td>\n",
       "      <td>53.1000</td>\n",
       "      <td>C123</td>\n",
       "      <td>S</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>Regular</td>\n",
       "      <td>C</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>5</td>\n",
       "      <td>0</td>\n",
       "      <td>3</td>\n",
       "      <td>Allen, Mr. William Henry</td>\n",
       "      <td>male</td>\n",
       "      <td>35.0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>373450</td>\n",
       "      <td>8.0500</td>\n",
       "      <td>NaN</td>\n",
       "      <td>S</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>Regular</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>6</td>\n",
       "      <td>0</td>\n",
       "      <td>3</td>\n",
       "      <td>Moran, Mr. James</td>\n",
       "      <td>male</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>330877</td>\n",
       "      <td>8.4583</td>\n",
       "      <td>NaN</td>\n",
       "      <td>Q</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>Regular</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>7</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>McCarthy, Mr. Timothy J</td>\n",
       "      <td>male</td>\n",
       "      <td>54.0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>17463</td>\n",
       "      <td>51.8625</td>\n",
       "      <td>E46</td>\n",
       "      <td>S</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>Regular</td>\n",
       "      <td>E</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>8</td>\n",
       "      <td>0</td>\n",
       "      <td>3</td>\n",
       "      <td>Palsson, Master. Gosta Leonard</td>\n",
       "      <td>male</td>\n",
       "      <td>2.0</td>\n",
       "      <td>3</td>\n",
       "      <td>1</td>\n",
       "      <td>349909</td>\n",
       "      <td>21.0750</td>\n",
       "      <td>NaN</td>\n",
       "      <td>S</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>Young</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8</th>\n",
       "      <td>9</td>\n",
       "      <td>1</td>\n",
       "      <td>3</td>\n",
       "      <td>Johnson, Mrs. Oscar W (Elisabeth Vilhelmina Berg)</td>\n",
       "      <td>female</td>\n",
       "      <td>27.0</td>\n",
       "      <td>0</td>\n",
       "      <td>2</td>\n",
       "      <td>347742</td>\n",
       "      <td>11.1333</td>\n",
       "      <td>NaN</td>\n",
       "      <td>S</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>Regular</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9</th>\n",
       "      <td>10</td>\n",
       "      <td>1</td>\n",
       "      <td>2</td>\n",
       "      <td>Nasser, Mrs. Nicholas (Adele Achem)</td>\n",
       "      <td>female</td>\n",
       "      <td>14.0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>237736</td>\n",
       "      <td>30.0708</td>\n",
       "      <td>NaN</td>\n",
       "      <td>C</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>Regular</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   PassengerId  Survived  Pclass  \\\n",
       "0            1         0       3   \n",
       "1            2         1       1   \n",
       "2            3         1       3   \n",
       "3            4         1       1   \n",
       "4            5         0       3   \n",
       "5            6         0       3   \n",
       "6            7         0       1   \n",
       "7            8         0       3   \n",
       "8            9         1       3   \n",
       "9           10         1       2   \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",
       "5                                   Moran, Mr. James    male   NaN      0   \n",
       "6                            McCarthy, Mr. Timothy J    male  54.0      0   \n",
       "7                     Palsson, Master. Gosta Leonard    male   2.0      3   \n",
       "8  Johnson, Mrs. Oscar W (Elisabeth Vilhelmina Berg)  female  27.0      0   \n",
       "9                Nasser, Mrs. Nicholas (Adele Achem)  female  14.0      1   \n",
       "\n",
       "   Parch            Ticket     Fare Cabin Embarked  Under18  \\\n",
       "0      0         A/5 21171   7.2500   NaN        S        0   \n",
       "1      0          PC 17599  71.2833   C85        C        0   \n",
       "2      0  STON/O2. 3101282   7.9250   NaN        S        0   \n",
       "3      0            113803  53.1000  C123        S        0   \n",
       "4      0            373450   8.0500   NaN        S        0   \n",
       "5      0            330877   8.4583   NaN        Q        0   \n",
       "6      0             17463  51.8625   E46        S        0   \n",
       "7      1            349909  21.0750   NaN        S        1   \n",
       "8      2            347742  11.1333   NaN        S        0   \n",
       "9      0            237736  30.0708   NaN        C        1   \n",
       "\n",
       "   TravellingWithFamily            Title Deck  \n",
       "0                     1          Regular  NaN  \n",
       "1                     1          Regular    C  \n",
       "2                     0  Unmarried Woman  NaN  \n",
       "3                     1          Regular    C  \n",
       "4                     0          Regular  NaN  \n",
       "5                     0          Regular  NaN  \n",
       "6                     0          Regular    E  \n",
       "7                     1            Young  NaN  \n",
       "8                     1          Regular  NaN  \n",
       "9                     1          Regular  NaN  "
      ]
     },
     "execution_count": 647,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data['Deck'] = data['Cabin'].str[0]\n",
    "data.head(10)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Let's see how many people were on each deck, and how many didn't have a cabin number."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 648,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "NaN    687\n",
       "C       59\n",
       "B       47\n",
       "D       33\n",
       "E       32\n",
       "A       15\n",
       "F       13\n",
       "G        4\n",
       "T        1\n",
       "Name: Deck, dtype: int64"
      ]
     },
     "execution_count": 648,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data['Deck'].value_counts(dropna=False)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Numeric Columns - Adding Values - Family Size"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Let's calculate the family size by adding two numeric columns together."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 649,
   "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>Under18</th>\n",
       "      <th>TravellingWithFamily</th>\n",
       "      <th>Title</th>\n",
       "      <th>Deck</th>\n",
       "      <th>FamilySize</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>3</td>\n",
       "      <td>Braund, Mr. Owen Harris</td>\n",
       "      <td>male</td>\n",
       "      <td>22.0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>A/5 21171</td>\n",
       "      <td>7.2500</td>\n",
       "      <td>NaN</td>\n",
       "      <td>S</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>Regular</td>\n",
       "      <td>NaN</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>2</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>Cumings, Mrs. John Bradley (Florence Briggs Th...</td>\n",
       "      <td>female</td>\n",
       "      <td>38.0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>PC 17599</td>\n",
       "      <td>71.2833</td>\n",
       "      <td>C85</td>\n",
       "      <td>C</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>Regular</td>\n",
       "      <td>C</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>3</td>\n",
       "      <td>1</td>\n",
       "      <td>3</td>\n",
       "      <td>Heikkinen, Miss. Laina</td>\n",
       "      <td>female</td>\n",
       "      <td>26.0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>STON/O2. 3101282</td>\n",
       "      <td>7.9250</td>\n",
       "      <td>NaN</td>\n",
       "      <td>S</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>Unmarried Woman</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>4</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>Futrelle, Mrs. Jacques Heath (Lily May Peel)</td>\n",
       "      <td>female</td>\n",
       "      <td>35.0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>113803</td>\n",
       "      <td>53.1000</td>\n",
       "      <td>C123</td>\n",
       "      <td>S</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>Regular</td>\n",
       "      <td>C</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>5</td>\n",
       "      <td>0</td>\n",
       "      <td>3</td>\n",
       "      <td>Allen, Mr. William Henry</td>\n",
       "      <td>male</td>\n",
       "      <td>35.0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>373450</td>\n",
       "      <td>8.0500</td>\n",
       "      <td>NaN</td>\n",
       "      <td>S</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>Regular</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>6</td>\n",
       "      <td>0</td>\n",
       "      <td>3</td>\n",
       "      <td>Moran, Mr. James</td>\n",
       "      <td>male</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>330877</td>\n",
       "      <td>8.4583</td>\n",
       "      <td>NaN</td>\n",
       "      <td>Q</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>Regular</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>7</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>McCarthy, Mr. Timothy J</td>\n",
       "      <td>male</td>\n",
       "      <td>54.0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>17463</td>\n",
       "      <td>51.8625</td>\n",
       "      <td>E46</td>\n",
       "      <td>S</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>Regular</td>\n",
       "      <td>E</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>8</td>\n",
       "      <td>0</td>\n",
       "      <td>3</td>\n",
       "      <td>Palsson, Master. Gosta Leonard</td>\n",
       "      <td>male</td>\n",
       "      <td>2.0</td>\n",
       "      <td>3</td>\n",
       "      <td>1</td>\n",
       "      <td>349909</td>\n",
       "      <td>21.0750</td>\n",
       "      <td>NaN</td>\n",
       "      <td>S</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>Young</td>\n",
       "      <td>NaN</td>\n",
       "      <td>4</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8</th>\n",
       "      <td>9</td>\n",
       "      <td>1</td>\n",
       "      <td>3</td>\n",
       "      <td>Johnson, Mrs. Oscar W (Elisabeth Vilhelmina Berg)</td>\n",
       "      <td>female</td>\n",
       "      <td>27.0</td>\n",
       "      <td>0</td>\n",
       "      <td>2</td>\n",
       "      <td>347742</td>\n",
       "      <td>11.1333</td>\n",
       "      <td>NaN</td>\n",
       "      <td>S</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>Regular</td>\n",
       "      <td>NaN</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9</th>\n",
       "      <td>10</td>\n",
       "      <td>1</td>\n",
       "      <td>2</td>\n",
       "      <td>Nasser, Mrs. Nicholas (Adele Achem)</td>\n",
       "      <td>female</td>\n",
       "      <td>14.0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>237736</td>\n",
       "      <td>30.0708</td>\n",
       "      <td>NaN</td>\n",
       "      <td>C</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>Regular</td>\n",
       "      <td>NaN</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   PassengerId  Survived  Pclass  \\\n",
       "0            1         0       3   \n",
       "1            2         1       1   \n",
       "2            3         1       3   \n",
       "3            4         1       1   \n",
       "4            5         0       3   \n",
       "5            6         0       3   \n",
       "6            7         0       1   \n",
       "7            8         0       3   \n",
       "8            9         1       3   \n",
       "9           10         1       2   \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",
       "5                                   Moran, Mr. James    male   NaN      0   \n",
       "6                            McCarthy, Mr. Timothy J    male  54.0      0   \n",
       "7                     Palsson, Master. Gosta Leonard    male   2.0      3   \n",
       "8  Johnson, Mrs. Oscar W (Elisabeth Vilhelmina Berg)  female  27.0      0   \n",
       "9                Nasser, Mrs. Nicholas (Adele Achem)  female  14.0      1   \n",
       "\n",
       "   Parch            Ticket     Fare Cabin Embarked  Under18  \\\n",
       "0      0         A/5 21171   7.2500   NaN        S        0   \n",
       "1      0          PC 17599  71.2833   C85        C        0   \n",
       "2      0  STON/O2. 3101282   7.9250   NaN        S        0   \n",
       "3      0            113803  53.1000  C123        S        0   \n",
       "4      0            373450   8.0500   NaN        S        0   \n",
       "5      0            330877   8.4583   NaN        Q        0   \n",
       "6      0             17463  51.8625   E46        S        0   \n",
       "7      1            349909  21.0750   NaN        S        1   \n",
       "8      2            347742  11.1333   NaN        S        0   \n",
       "9      0            237736  30.0708   NaN        C        1   \n",
       "\n",
       "   TravellingWithFamily            Title Deck  FamilySize  \n",
       "0                     1          Regular  NaN           1  \n",
       "1                     1          Regular    C           1  \n",
       "2                     0  Unmarried Woman  NaN           0  \n",
       "3                     1          Regular    C           1  \n",
       "4                     0          Regular  NaN           0  \n",
       "5                     0          Regular  NaN           0  \n",
       "6                     0          Regular    E           0  \n",
       "7                     1            Young  NaN           4  \n",
       "8                     1          Regular  NaN           2  \n",
       "9                     1          Regular  NaN           1  "
      ]
     },
     "execution_count": 649,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data['FamilySize'] = data['SibSp'] + data['Parch']\n",
    "\n",
    "data.head(10)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Joining to Counts - Shared Tickets and the Impact on Fare"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Apparently tickets were shared between families. Let's explore this.\n",
    "\n",
    "By using the 'value_counts' method twice in a row \n",
    "- we first get a count of how many times each ticket number appeared, \n",
    "- and next we get a count of how many times a ticket was shared between 1 person, how many times a ticket was shared between 2 people, and so on. "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 650,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "1    547\n",
       "2     94\n",
       "3     21\n",
       "4     11\n",
       "7      3\n",
       "6      3\n",
       "5      2\n",
       "Name: Ticket, dtype: int64"
      ]
     },
     "execution_count": 650,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data['Ticket'].value_counts().value_counts()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "It looks like this might be interesting and might have some impact on our fare value as well - so let's create a column for fare per person.\n",
    "\n",
    "First, though, we'll need to add a column for the counts of how many times a ticket does appear in the dataset.\n",
    "\n",
    "We start by creating a dataframe from the counts."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 651,
   "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>index</th>\n",
       "      <th>NumPeopleTicketShared</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>347082</td>\n",
       "      <td>7</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>CA. 2343</td>\n",
       "      <td>7</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>1601</td>\n",
       "      <td>7</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>3101295</td>\n",
       "      <td>6</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>CA 2144</td>\n",
       "      <td>6</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>676</th>\n",
       "      <td>9234</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>677</th>\n",
       "      <td>19988</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>678</th>\n",
       "      <td>2693</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>679</th>\n",
       "      <td>PC 17612</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>680</th>\n",
       "      <td>370376</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>681 rows × 2 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "        index  NumPeopleTicketShared\n",
       "0      347082                      7\n",
       "1    CA. 2343                      7\n",
       "2        1601                      7\n",
       "3     3101295                      6\n",
       "4     CA 2144                      6\n",
       "..        ...                    ...\n",
       "676      9234                      1\n",
       "677     19988                      1\n",
       "678      2693                      1\n",
       "679  PC 17612                      1\n",
       "680    370376                      1\n",
       "\n",
       "[681 rows x 2 columns]"
      ]
     },
     "execution_count": 651,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "ticket_counts_df = (pd.DataFrame(data['Ticket'].value_counts())\n",
    "                    .reset_index()\n",
    "                    .rename(columns={\"Ticket\": \"NumPeopleTicketShared\"})\n",
    "                    )\n",
    "\n",
    "\n",
    "ticket_counts_df"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Now we can join this back to our main dataframe. "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 652,
   "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>Under18</th>\n",
       "      <th>TravellingWithFamily</th>\n",
       "      <th>Title</th>\n",
       "      <th>Deck</th>\n",
       "      <th>FamilySize</th>\n",
       "      <th>NumPeopleTicketShared</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>3</td>\n",
       "      <td>Braund, Mr. Owen Harris</td>\n",
       "      <td>male</td>\n",
       "      <td>22.0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>A/5 21171</td>\n",
       "      <td>7.2500</td>\n",
       "      <td>NaN</td>\n",
       "      <td>S</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>Regular</td>\n",
       "      <td>NaN</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>2</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>Cumings, Mrs. John Bradley (Florence Briggs Th...</td>\n",
       "      <td>female</td>\n",
       "      <td>38.0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>PC 17599</td>\n",
       "      <td>71.2833</td>\n",
       "      <td>C85</td>\n",
       "      <td>C</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>Regular</td>\n",
       "      <td>C</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>3</td>\n",
       "      <td>1</td>\n",
       "      <td>3</td>\n",
       "      <td>Heikkinen, Miss. Laina</td>\n",
       "      <td>female</td>\n",
       "      <td>26.0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>STON/O2. 3101282</td>\n",
       "      <td>7.9250</td>\n",
       "      <td>NaN</td>\n",
       "      <td>S</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>Unmarried Woman</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>4</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>Futrelle, Mrs. Jacques Heath (Lily May Peel)</td>\n",
       "      <td>female</td>\n",
       "      <td>35.0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>113803</td>\n",
       "      <td>53.1000</td>\n",
       "      <td>C123</td>\n",
       "      <td>S</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>Regular</td>\n",
       "      <td>C</td>\n",
       "      <td>1</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>5</td>\n",
       "      <td>0</td>\n",
       "      <td>3</td>\n",
       "      <td>Allen, Mr. William Henry</td>\n",
       "      <td>male</td>\n",
       "      <td>35.0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>373450</td>\n",
       "      <td>8.0500</td>\n",
       "      <td>NaN</td>\n",
       "      <td>S</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>Regular</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   PassengerId  Survived  Pclass  \\\n",
       "0            1         0       3   \n",
       "1            2         1       1   \n",
       "2            3         1       3   \n",
       "3            4         1       1   \n",
       "4            5         0       3   \n",
       "\n",
       "                                                Name     Sex   Age  SibSp  \\\n",
       "0                            Braund, Mr. Owen Harris    male  22.0      1   \n",
       "1  Cumings, Mrs. John Bradley (Florence Briggs Th...  female  38.0      1   \n",
       "2                             Heikkinen, Miss. Laina  female  26.0      0   \n",
       "3       Futrelle, Mrs. Jacques Heath (Lily May Peel)  female  35.0      1   \n",
       "4                           Allen, Mr. William Henry    male  35.0      0   \n",
       "\n",
       "   Parch            Ticket     Fare Cabin Embarked  Under18  \\\n",
       "0      0         A/5 21171   7.2500   NaN        S        0   \n",
       "1      0          PC 17599  71.2833   C85        C        0   \n",
       "2      0  STON/O2. 3101282   7.9250   NaN        S        0   \n",
       "3      0            113803  53.1000  C123        S        0   \n",
       "4      0            373450   8.0500   NaN        S        0   \n",
       "\n",
       "   TravellingWithFamily            Title Deck  FamilySize  \\\n",
       "0                     1          Regular  NaN           1   \n",
       "1                     1          Regular    C           1   \n",
       "2                     0  Unmarried Woman  NaN           0   \n",
       "3                     1          Regular    C           1   \n",
       "4                     0          Regular  NaN           0   \n",
       "\n",
       "   NumPeopleTicketShared  \n",
       "0                      1  \n",
       "1                      1  \n",
       "2                      1  \n",
       "3                      2  \n",
       "4                      1  "
      ]
     },
     "execution_count": 652,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data = pd.merge(\n",
    "    left=data, # Original Dataframe\n",
    "    right=ticket_counts_df, # Count Dataframe\n",
    "    left_on=\"Ticket\", # Common column from original dataframe\n",
    "    right_on=\"index\", # Common column from count dataframe\n",
    "    how=\"left\" # Keep all rows from the original dataframe\n",
    ").drop(columns='index') # Get rid of the extra column\n",
    "\n",
    "data.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Now we can create a 'fare per person' column by dividing the fare by the number of people it's shared between. "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 653,
   "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>Under18</th>\n",
       "      <th>TravellingWithFamily</th>\n",
       "      <th>Title</th>\n",
       "      <th>Deck</th>\n",
       "      <th>FamilySize</th>\n",
       "      <th>NumPeopleTicketShared</th>\n",
       "      <th>FarePerPerson</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>3</td>\n",
       "      <td>Braund, Mr. Owen Harris</td>\n",
       "      <td>male</td>\n",
       "      <td>22.0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>A/5 21171</td>\n",
       "      <td>7.2500</td>\n",
       "      <td>NaN</td>\n",
       "      <td>S</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>Regular</td>\n",
       "      <td>NaN</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>7.2500</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>2</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>Cumings, Mrs. John Bradley (Florence Briggs Th...</td>\n",
       "      <td>female</td>\n",
       "      <td>38.0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>PC 17599</td>\n",
       "      <td>71.2833</td>\n",
       "      <td>C85</td>\n",
       "      <td>C</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>Regular</td>\n",
       "      <td>C</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>71.2833</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>3</td>\n",
       "      <td>1</td>\n",
       "      <td>3</td>\n",
       "      <td>Heikkinen, Miss. Laina</td>\n",
       "      <td>female</td>\n",
       "      <td>26.0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>STON/O2. 3101282</td>\n",
       "      <td>7.9250</td>\n",
       "      <td>NaN</td>\n",
       "      <td>S</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>Unmarried Woman</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>7.9250</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>4</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>Futrelle, Mrs. Jacques Heath (Lily May Peel)</td>\n",
       "      <td>female</td>\n",
       "      <td>35.0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>113803</td>\n",
       "      <td>53.1000</td>\n",
       "      <td>C123</td>\n",
       "      <td>S</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>Regular</td>\n",
       "      <td>C</td>\n",
       "      <td>1</td>\n",
       "      <td>2</td>\n",
       "      <td>26.5500</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>5</td>\n",
       "      <td>0</td>\n",
       "      <td>3</td>\n",
       "      <td>Allen, Mr. William Henry</td>\n",
       "      <td>male</td>\n",
       "      <td>35.0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>373450</td>\n",
       "      <td>8.0500</td>\n",
       "      <td>NaN</td>\n",
       "      <td>S</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>Regular</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>8.0500</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>886</th>\n",
       "      <td>887</td>\n",
       "      <td>0</td>\n",
       "      <td>2</td>\n",
       "      <td>Montvila, Rev. Juozas</td>\n",
       "      <td>male</td>\n",
       "      <td>27.0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>211536</td>\n",
       "      <td>13.0000</td>\n",
       "      <td>NaN</td>\n",
       "      <td>S</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>Upper Class</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>13.0000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>887</th>\n",
       "      <td>888</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>Graham, Miss. Margaret Edith</td>\n",
       "      <td>female</td>\n",
       "      <td>19.0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>112053</td>\n",
       "      <td>30.0000</td>\n",
       "      <td>B42</td>\n",
       "      <td>S</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>Unmarried Woman</td>\n",
       "      <td>B</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>30.0000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>888</th>\n",
       "      <td>889</td>\n",
       "      <td>0</td>\n",
       "      <td>3</td>\n",
       "      <td>Johnston, Miss. Catherine Helen \"Carrie\"</td>\n",
       "      <td>female</td>\n",
       "      <td>NaN</td>\n",
       "      <td>1</td>\n",
       "      <td>2</td>\n",
       "      <td>W./C. 6607</td>\n",
       "      <td>23.4500</td>\n",
       "      <td>NaN</td>\n",
       "      <td>S</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>Unknown</td>\n",
       "      <td>NaN</td>\n",
       "      <td>3</td>\n",
       "      <td>2</td>\n",
       "      <td>11.7250</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>889</th>\n",
       "      <td>890</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>Behr, Mr. Karl Howell</td>\n",
       "      <td>male</td>\n",
       "      <td>26.0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>111369</td>\n",
       "      <td>30.0000</td>\n",
       "      <td>C148</td>\n",
       "      <td>C</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>Regular</td>\n",
       "      <td>C</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>30.0000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>890</th>\n",
       "      <td>891</td>\n",
       "      <td>0</td>\n",
       "      <td>3</td>\n",
       "      <td>Dooley, Mr. Patrick</td>\n",
       "      <td>male</td>\n",
       "      <td>32.0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>370376</td>\n",
       "      <td>7.7500</td>\n",
       "      <td>NaN</td>\n",
       "      <td>Q</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>Regular</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>7.7500</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>891 rows × 19 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "     PassengerId  Survived  Pclass  \\\n",
       "0              1         0       3   \n",
       "1              2         1       1   \n",
       "2              3         1       3   \n",
       "3              4         1       1   \n",
       "4              5         0       3   \n",
       "..           ...       ...     ...   \n",
       "886          887         0       2   \n",
       "887          888         1       1   \n",
       "888          889         0       3   \n",
       "889          890         1       1   \n",
       "890          891         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",
       "886                              Montvila, Rev. Juozas    male  27.0      0   \n",
       "887                       Graham, Miss. Margaret Edith  female  19.0      0   \n",
       "888           Johnston, Miss. Catherine Helen \"Carrie\"  female   NaN      1   \n",
       "889                              Behr, Mr. Karl Howell    male  26.0      0   \n",
       "890                                Dooley, Mr. Patrick    male  32.0      0   \n",
       "\n",
       "     Parch            Ticket     Fare Cabin Embarked  Under18  \\\n",
       "0        0         A/5 21171   7.2500   NaN        S        0   \n",
       "1        0          PC 17599  71.2833   C85        C        0   \n",
       "2        0  STON/O2. 3101282   7.9250   NaN        S        0   \n",
       "3        0            113803  53.1000  C123        S        0   \n",
       "4        0            373450   8.0500   NaN        S        0   \n",
       "..     ...               ...      ...   ...      ...      ...   \n",
       "886      0            211536  13.0000   NaN        S        0   \n",
       "887      0            112053  30.0000   B42        S        0   \n",
       "888      2        W./C. 6607  23.4500   NaN        S        0   \n",
       "889      0            111369  30.0000  C148        C        0   \n",
       "890      0            370376   7.7500   NaN        Q        0   \n",
       "\n",
       "     TravellingWithFamily            Title Deck  FamilySize  \\\n",
       "0                       1          Regular  NaN           1   \n",
       "1                       1          Regular    C           1   \n",
       "2                       0  Unmarried Woman  NaN           0   \n",
       "3                       1          Regular    C           1   \n",
       "4                       0          Regular  NaN           0   \n",
       "..                    ...              ...  ...         ...   \n",
       "886                     0      Upper Class  NaN           0   \n",
       "887                     0  Unmarried Woman    B           0   \n",
       "888                     1          Unknown  NaN           3   \n",
       "889                     0          Regular    C           0   \n",
       "890                     0          Regular  NaN           0   \n",
       "\n",
       "     NumPeopleTicketShared  FarePerPerson  \n",
       "0                        1         7.2500  \n",
       "1                        1        71.2833  \n",
       "2                        1         7.9250  \n",
       "3                        2        26.5500  \n",
       "4                        1         8.0500  \n",
       "..                     ...            ...  \n",
       "886                      1        13.0000  \n",
       "887                      1        30.0000  \n",
       "888                      2        11.7250  \n",
       "889                      1        30.0000  \n",
       "890                      1         7.7500  \n",
       "\n",
       "[891 rows x 19 columns]"
      ]
     },
     "execution_count": 653,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data['FarePerPerson'] = data['Fare'] / data['NumPeopleTicketShared']\n",
    "data"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Indicating Missing Values\n",
    "\n",
    "We will be one-hot encoding the 'embarked' column. It can be helpful to provide an extra column to indcate when the relevant column contains a missing value."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 654,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0    889\n",
       "1      2\n",
       "Name: Embarked_Missing, dtype: int64"
      ]
     },
     "execution_count": 654,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data['Embarked_Missing'] = data['Embarked'].isna().astype('int')\n",
    "data['Embarked_Missing'].value_counts()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "In a later step, we are going to *impute* values where existing values are missing.\n",
    "\n",
    "It can help to have features indicating that a value was missing and replaced."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 655,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0    714\n",
       "1    177\n",
       "Name: Age_Imputed, dtype: int64"
      ]
     },
     "execution_count": 655,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data['Age_Imputed'] = data['Age'].isna().astype('int')\n",
    "data['Age_Imputed'].value_counts()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Let's remind ourselves what the dataframe looks like at this stage."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 656,
   "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>...</th>\n",
       "      <th>Embarked</th>\n",
       "      <th>Under18</th>\n",
       "      <th>TravellingWithFamily</th>\n",
       "      <th>Title</th>\n",
       "      <th>Deck</th>\n",
       "      <th>FamilySize</th>\n",
       "      <th>NumPeopleTicketShared</th>\n",
       "      <th>FarePerPerson</th>\n",
       "      <th>Embarked_Missing</th>\n",
       "      <th>Age_Imputed</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>3</td>\n",
       "      <td>Braund, Mr. Owen Harris</td>\n",
       "      <td>male</td>\n",
       "      <td>22.0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>A/5 21171</td>\n",
       "      <td>7.2500</td>\n",
       "      <td>...</td>\n",
       "      <td>S</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>Regular</td>\n",
       "      <td>NaN</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>7.2500</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>2</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>Cumings, Mrs. John Bradley (Florence Briggs Th...</td>\n",
       "      <td>female</td>\n",
       "      <td>38.0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>PC 17599</td>\n",
       "      <td>71.2833</td>\n",
       "      <td>...</td>\n",
       "      <td>C</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>Regular</td>\n",
       "      <td>C</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>71.2833</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>3</td>\n",
       "      <td>1</td>\n",
       "      <td>3</td>\n",
       "      <td>Heikkinen, Miss. Laina</td>\n",
       "      <td>female</td>\n",
       "      <td>26.0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>STON/O2. 3101282</td>\n",
       "      <td>7.9250</td>\n",
       "      <td>...</td>\n",
       "      <td>S</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>Unmarried Woman</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>7.9250</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>4</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>Futrelle, Mrs. Jacques Heath (Lily May Peel)</td>\n",
       "      <td>female</td>\n",
       "      <td>35.0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>113803</td>\n",
       "      <td>53.1000</td>\n",
       "      <td>...</td>\n",
       "      <td>S</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>Regular</td>\n",
       "      <td>C</td>\n",
       "      <td>1</td>\n",
       "      <td>2</td>\n",
       "      <td>26.5500</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>5</td>\n",
       "      <td>0</td>\n",
       "      <td>3</td>\n",
       "      <td>Allen, Mr. William Henry</td>\n",
       "      <td>male</td>\n",
       "      <td>35.0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>373450</td>\n",
       "      <td>8.0500</td>\n",
       "      <td>...</td>\n",
       "      <td>S</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>Regular</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>8.0500</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>886</th>\n",
       "      <td>887</td>\n",
       "      <td>0</td>\n",
       "      <td>2</td>\n",
       "      <td>Montvila, Rev. Juozas</td>\n",
       "      <td>male</td>\n",
       "      <td>27.0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>211536</td>\n",
       "      <td>13.0000</td>\n",
       "      <td>...</td>\n",
       "      <td>S</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>Upper Class</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>13.0000</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>887</th>\n",
       "      <td>888</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>Graham, Miss. Margaret Edith</td>\n",
       "      <td>female</td>\n",
       "      <td>19.0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>112053</td>\n",
       "      <td>30.0000</td>\n",
       "      <td>...</td>\n",
       "      <td>S</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>Unmarried Woman</td>\n",
       "      <td>B</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>30.0000</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>888</th>\n",
       "      <td>889</td>\n",
       "      <td>0</td>\n",
       "      <td>3</td>\n",
       "      <td>Johnston, Miss. Catherine Helen \"Carrie\"</td>\n",
       "      <td>female</td>\n",
       "      <td>NaN</td>\n",
       "      <td>1</td>\n",
       "      <td>2</td>\n",
       "      <td>W./C. 6607</td>\n",
       "      <td>23.4500</td>\n",
       "      <td>...</td>\n",
       "      <td>S</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>Unknown</td>\n",
       "      <td>NaN</td>\n",
       "      <td>3</td>\n",
       "      <td>2</td>\n",
       "      <td>11.7250</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>889</th>\n",
       "      <td>890</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>Behr, Mr. Karl Howell</td>\n",
       "      <td>male</td>\n",
       "      <td>26.0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>111369</td>\n",
       "      <td>30.0000</td>\n",
       "      <td>...</td>\n",
       "      <td>C</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>Regular</td>\n",
       "      <td>C</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>30.0000</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>890</th>\n",
       "      <td>891</td>\n",
       "      <td>0</td>\n",
       "      <td>3</td>\n",
       "      <td>Dooley, Mr. Patrick</td>\n",
       "      <td>male</td>\n",
       "      <td>32.0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>370376</td>\n",
       "      <td>7.7500</td>\n",
       "      <td>...</td>\n",
       "      <td>Q</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>Regular</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>7.7500</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>891 rows × 21 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "     PassengerId  Survived  Pclass  \\\n",
       "0              1         0       3   \n",
       "1              2         1       1   \n",
       "2              3         1       3   \n",
       "3              4         1       1   \n",
       "4              5         0       3   \n",
       "..           ...       ...     ...   \n",
       "886          887         0       2   \n",
       "887          888         1       1   \n",
       "888          889         0       3   \n",
       "889          890         1       1   \n",
       "890          891         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",
       "886                              Montvila, Rev. Juozas    male  27.0      0   \n",
       "887                       Graham, Miss. Margaret Edith  female  19.0      0   \n",
       "888           Johnston, Miss. Catherine Helen \"Carrie\"  female   NaN      1   \n",
       "889                              Behr, Mr. Karl Howell    male  26.0      0   \n",
       "890                                Dooley, Mr. Patrick    male  32.0      0   \n",
       "\n",
       "     Parch            Ticket     Fare  ... Embarked Under18  \\\n",
       "0        0         A/5 21171   7.2500  ...        S       0   \n",
       "1        0          PC 17599  71.2833  ...        C       0   \n",
       "2        0  STON/O2. 3101282   7.9250  ...        S       0   \n",
       "3        0            113803  53.1000  ...        S       0   \n",
       "4        0            373450   8.0500  ...        S       0   \n",
       "..     ...               ...      ...  ...      ...     ...   \n",
       "886      0            211536  13.0000  ...        S       0   \n",
       "887      0            112053  30.0000  ...        S       0   \n",
       "888      2        W./C. 6607  23.4500  ...        S       0   \n",
       "889      0            111369  30.0000  ...        C       0   \n",
       "890      0            370376   7.7500  ...        Q       0   \n",
       "\n",
       "     TravellingWithFamily            Title Deck FamilySize  \\\n",
       "0                       1          Regular  NaN          1   \n",
       "1                       1          Regular    C          1   \n",
       "2                       0  Unmarried Woman  NaN          0   \n",
       "3                       1          Regular    C          1   \n",
       "4                       0          Regular  NaN          0   \n",
       "..                    ...              ...  ...        ...   \n",
       "886                     0      Upper Class  NaN          0   \n",
       "887                     0  Unmarried Woman    B          0   \n",
       "888                     1          Unknown  NaN          3   \n",
       "889                     0          Regular    C          0   \n",
       "890                     0          Regular  NaN          0   \n",
       "\n",
       "     NumPeopleTicketShared  FarePerPerson  Embarked_Missing  Age_Imputed  \n",
       "0                        1         7.2500                 0            0  \n",
       "1                        1        71.2833                 0            0  \n",
       "2                        1         7.9250                 0            0  \n",
       "3                        2        26.5500                 0            0  \n",
       "4                        1         8.0500                 0            0  \n",
       "..                     ...            ...               ...          ...  \n",
       "886                      1        13.0000                 0            0  \n",
       "887                      1        30.0000                 0            0  \n",
       "888                      2        11.7250                 0            1  \n",
       "889                      1        30.0000                 0            0  \n",
       "890                      1         7.7500                 0            0  \n",
       "\n",
       "[891 rows x 21 columns]"
      ]
     },
     "execution_count": 656,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Comparing performance"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Finally, we need to do the remaining data preparation steps to make this data usable by machine learning models. "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 657,
   "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>IsMale</th>\n",
       "      <th>Age</th>\n",
       "      <th>SibSp</th>\n",
       "      <th>Parch</th>\n",
       "      <th>Fare</th>\n",
       "      <th>Under18</th>\n",
       "      <th>TravellingWithFamily</th>\n",
       "      <th>...</th>\n",
       "      <th>Title_Upper Class</th>\n",
       "      <th>Title_Young</th>\n",
       "      <th>Deck_A</th>\n",
       "      <th>Deck_B</th>\n",
       "      <th>Deck_C</th>\n",
       "      <th>Deck_D</th>\n",
       "      <th>Deck_E</th>\n",
       "      <th>Deck_F</th>\n",
       "      <th>Deck_G</th>\n",
       "      <th>Deck_T</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>3</td>\n",
       "      <td>1</td>\n",
       "      <td>22.0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>7.2500</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>...</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>1</th>\n",
       "      <td>2</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>38.0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>71.2833</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>...</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>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>3</td>\n",
       "      <td>1</td>\n",
       "      <td>3</td>\n",
       "      <td>0</td>\n",
       "      <td>26.0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>7.9250</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>...</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>3</th>\n",
       "      <td>4</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>35.0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>53.1000</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>...</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>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>5</td>\n",
       "      <td>0</td>\n",
       "      <td>3</td>\n",
       "      <td>1</td>\n",
       "      <td>35.0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>8.0500</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>...</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",
       "  </tbody>\n",
       "</table>\n",
       "<p>5 rows × 31 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "   PassengerId  Survived  Pclass  IsMale   Age  SibSp  Parch     Fare  \\\n",
       "0            1         0       3       1  22.0      1      0   7.2500   \n",
       "1            2         1       1       0  38.0      1      0  71.2833   \n",
       "2            3         1       3       0  26.0      0      0   7.9250   \n",
       "3            4         1       1       0  35.0      1      0  53.1000   \n",
       "4            5         0       3       1  35.0      0      0   8.0500   \n",
       "\n",
       "   Under18  TravellingWithFamily  ...  Title_Upper Class  Title_Young  Deck_A  \\\n",
       "0        0                     1  ...                  0            0       0   \n",
       "1        0                     1  ...                  0            0       0   \n",
       "2        0                     0  ...                  0            0       0   \n",
       "3        0                     1  ...                  0            0       0   \n",
       "4        0                     0  ...                  0            0       0   \n",
       "\n",
       "   Deck_B  Deck_C  Deck_D  Deck_E  Deck_F  Deck_G  Deck_T  \n",
       "0       0       0       0       0       0       0       0  \n",
       "1       0       1       0       0       0       0       0  \n",
       "2       0       0       0       0       0       0       0  \n",
       "3       0       1       0       0       0       0       0  \n",
       "4       0       0       0       0       0       0       0  \n",
       "\n",
       "[5 rows x 31 columns]"
      ]
     },
     "execution_count": 657,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "###############################\n",
    "# Replace unclear data values #\n",
    "###############################\n",
    "embarked_lookup = {\n",
    "    'S': 'Southampton',\n",
    "    'C':  'Cherbourg',\n",
    "    'Q': 'Queenstown'\n",
    "}\n",
    "\n",
    "data['Embarked'] = data['Embarked'].apply(lambda row_value: embarked_lookup.get(row_value))\n",
    "\n",
    "#######################\n",
    "# One hot encoding    #\n",
    "#######################\n",
    "one_hot_embarked = pd.get_dummies(data['Embarked'], prefix='Embarked').astype('int')\n",
    "# Drop the column as it is now encoded\n",
    "data = data.drop('Embarked', axis = 1)\n",
    "# Join the encoded df\n",
    "data = data.join(one_hot_embarked)\n",
    "\n",
    "one_hot_title = pd.get_dummies(data['Title'], prefix='Title').astype('int')\n",
    "# Drop the column as it is now encoded\n",
    "data = data.drop('Title', axis = 1)\n",
    "# Join the encoded df\n",
    "data = data.join(one_hot_title)\n",
    "\n",
    "one_hot_deck = pd.get_dummies(data['Deck'], prefix='Deck').astype('int')\n",
    "# Drop the column as it is now encoded\n",
    "data = data.drop('Deck', axis = 1)\n",
    "# Join the encoded df\n",
    "data = data.join(one_hot_deck)\n",
    "\n",
    "#######################\n",
    "# Dichotomous columns #\n",
    "#######################\n",
    "data['Sex'].replace('male', 1, inplace=True)\n",
    "data['Sex'].replace('female', 0, inplace=True)\n",
    "data = data.rename(columns={'Sex': 'IsMale'})\n",
    "\n",
    "#####################################\n",
    "# Tidying up remaining column names #\n",
    "#####################################\n",
    "data = data.drop(columns=['Name', 'Ticket', 'Cabin'])\n",
    "\n",
    "data.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Original Dataset Performance"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 658,
   "metadata": {},
   "outputs": [],
   "source": [
    "data.drop('PassengerId', inplace=True, axis=1)\n",
    "\n",
    "X = data.drop('Survived',axis=1) # X = all 'data' except the 'survived' column\n",
    "y = data['Survived'] # y = 'survived' column from 'data'\n",
    "\n",
    "X_train, X_validate, y_train, y_validate = train_test_split(X, y, test_size=0.2, random_state=42)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 659,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Training Dataset Samples: 712\n",
      "Validation Dataset Samples: 179\n"
     ]
    }
   ],
   "source": [
    "print(f\"Training Dataset Samples: {len(X_train)}\")\n",
    "print(f\"Validation Dataset Samples: {len(X_validate)}\")\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 660,
   "metadata": {},
   "outputs": [],
   "source": [
    "def impute_missing_df(df, impute_type=\"median\"):\n",
    "    imputed_df = df.copy()\n",
    "    imputed_df.values[:] = SimpleImputer(missing_values=np.nan, strategy=impute_type).fit_transform(df)\n",
    "    return imputed_df\n",
    "\n",
    "X_train = impute_missing_df(X_train, impute_type=\"median\")\n",
    "X_validate = impute_missing_df(X_validate, impute_type=\"median\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 661,
   "metadata": {},
   "outputs": [],
   "source": [
    "def fit_assess(name=\"XGBoost\",\n",
    "              X_train=X_train, X_validate=X_validate,\n",
    "              y_train=y_train, y_validate=y_validate,\n",
    "              model=XGBClassifier(random_state=42)\n",
    "              ):\n",
    "\n",
    "     model.fit(X_train, y_train)\n",
    "\n",
    "     y_pred_train = model.predict(X_train)\n",
    "     y_pred_val = model.predict(X_validate)\n",
    "\n",
    "     tn, fp, fn, tp = confusion_matrix(y_validate, y_pred_val, labels=[0, 1]).ravel()\n",
    "\n",
    "     return (pd.DataFrame({\n",
    "            'Accuracy (training)': np.mean(y_pred_train == y_train),\n",
    "            'Accuracy (validation)': np.mean(y_pred_val == y_validate),\n",
    "            'Precision (validation)': precision_score(y_validate, y_pred_val, average='macro'),\n",
    "            'Recall (validation)': recall_score(y_validate, y_pred_val, average='macro'),\n",
    "            \"AUC\": roc_auc_score(y_validate, y_pred_val),\n",
    "            \"f1\": f1_score(y_validate, y_pred_val, average='macro'),\n",
    "            \"FP\": fp,\n",
    "            \"FN\": fn\n",
    "          }, index=[name]\n",
    ").round(3), model)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 662,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Index(['Pclass', 'IsMale', 'Age', 'SibSp', 'Parch', 'Fare', 'Under18',\n",
       "       'TravellingWithFamily', 'FamilySize', 'NumPeopleTicketShared',\n",
       "       'FarePerPerson', 'Embarked_Missing', 'Age_Imputed',\n",
       "       'Embarked_Cherbourg', 'Embarked_Queenstown', 'Embarked_Southampton',\n",
       "       'Title_Regular', 'Title_Unknown', 'Title_Unmarried Woman',\n",
       "       'Title_Upper Class', 'Title_Young', 'Deck_A', 'Deck_B', 'Deck_C',\n",
       "       'Deck_D', 'Deck_E', 'Deck_F', 'Deck_G', 'Deck_T'],\n",
       "      dtype='object')"
      ]
     },
     "execution_count": 662,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "X_train.columns"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 663,
   "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>Accuracy (training)</th>\n",
       "      <th>Accuracy (validation)</th>\n",
       "      <th>Precision (validation)</th>\n",
       "      <th>Recall (validation)</th>\n",
       "      <th>AUC</th>\n",
       "      <th>f1</th>\n",
       "      <th>FP</th>\n",
       "      <th>FN</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>Without Feature Engineering</th>\n",
       "      <td>0.969</td>\n",
       "      <td>0.804</td>\n",
       "      <td>0.799</td>\n",
       "      <td>0.795</td>\n",
       "      <td>0.795</td>\n",
       "      <td>0.797</td>\n",
       "      <td>16</td>\n",
       "      <td>19</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                             Accuracy (training)  Accuracy (validation)  \\\n",
       "Without Feature Engineering                0.969                  0.804   \n",
       "\n",
       "                             Precision (validation)  Recall (validation)  \\\n",
       "Without Feature Engineering                   0.799                0.795   \n",
       "\n",
       "                               AUC     f1  FP  FN  \n",
       "Without Feature Engineering  0.795  0.797  16  19  "
      ]
     },
     "execution_count": 663,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "original_columns = [\n",
    "       'Pclass', 'IsMale', 'Age', 'SibSp', 'Parch', 'Fare',\n",
    "       'Embarked_Missing', 'Age_Imputed',\n",
    "       'Embarked_Cherbourg', 'Embarked_Queenstown', 'Embarked_Southampton'\n",
    "    ]\n",
    "\n",
    "before_engineering = fit_assess(\n",
    "    X_train=X_train[original_columns],\n",
    "    X_validate=X_validate[original_columns],\n",
    "    name=\"Without Feature Engineering\"\n",
    ")\n",
    "\n",
    "results_df = before_engineering[0]\n",
    "results_df"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 664,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAABdEAAAPdCAYAAABlRyFLAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjguMCwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy81sbWrAAAACXBIWXMAAA9hAAAPYQGoP6dpAACbuElEQVR4nOzde/zW8/0/8Men06f6VJ8yEpaioqKINsupMEQxhy1mKKevmMWcD0MOSVMzY8hYDothbN85H0Y2mTmGyRBihjl1kHTQ5/r94df1dam3PhE53O+32/vW53ofXq/n+3BdfzyuV6+rqlQqlQIAAAAAACyiwfIuAAAAAAAAvqiE6AAAAAAAUECIDgAAAAAABYToAAAAAABQQIgOAAAAAAAFhOgAAAAAAFBAiA4AAAAAAAWE6AAAAAAAUECIDgAAAAAABYToAAB8YV166aWpqqrK1KlTl3cpLAcTJkxIVVVVJkyYsLxLoR6mTp2aqqqqXHrppcu7lM/UWWedlTXXXDMNGzbM+uuvv1THLu4zrV+/funXr98yrfGzMGTIkHTs2PETHftlOUcAKCJEBwD4AlkYsCxuOfbYYz+TPu+7774MHz4806dP/0za/zqbPXt2hg8fLgTmC2Py5MkZPnz4p/pi6sorr8wvf/nLZVbTl8ntt9+eo48+OptssknGjRuXM844Y3mXBAB8Dhot7wIAAFjUqaeemjXWWKNi3brrrvuZ9HXffffllFNOyZAhQ9K6devPpI9Paq+99sruu++e6urq5V3KJzJ79uyccsopSWIU5iew+eab57333kuTJk2WdylfGZMnT84pp5ySfv36feJRxVdeeWX++c9/5rDDDqtY36FDh7z33ntp3Ljxpy/0C+quu+5KgwYNcskll3guAeBrRIgOAPAFtN1226V3797Lu4xP5d13301NTc2naqNhw4Zp2LDhMqro81NXV5d58+Yt7zK+9Bo0aJCmTZsu7zI+U7Nnz07z5s2XdxnLRFVV1Vf+fr3++utp1qyZAB0AvmZM5wIA8CV0yy23ZLPNNktNTU1atmyZAQMG5Mknn6zY5/HHH8+QIUOy5pprpmnTpmnXrl323XffvPXWW+V9hg8fnqOOOipJssYaa5Snjpk6derHzm9cVVWV4cOHV7RTVVWVyZMnZ4899kibNm2y6aablrf/7ne/y4YbbphmzZplhRVWyO67755///vfSzzPxc0f3LFjxwwcODATJkxI796906xZs/To0aM8Zcr111+fHj16pGnTptlwww3z6KOPVrQ5ZMiQtGjRIs8//3y23Xbb1NTUZNVVV82pp56aUqlUse+7776bI444Iu3bt091dXXWXnvtjB49epH9qqqqcsghh2T8+PFZZ511Ul1dnQsvvDArrbRSkuSUU04pX9uF160+9+fD13bKlCnl/y1QW1ubffbZJ7Nnz17kmv3ud7/Lt7/97TRv3jxt2rTJ5ptvnttvv71in/o8P6+99lr22WeffPOb30x1dXVWWWWVfO9731viNCBFcx8vbj7l3//+99lwww3TsmXLtGrVKj169Mg555xT3r64OdH79euXddddN5MnT84WW2yR5s2bZ7XVVsvPf/7zRfp88cUXs+OOO6ampiZt27bNT3/609x22231mmd94XX/17/+lUGDBqVVq1b5xje+kUMPPTRz5sxZZP/6POMLa3/44Yez+eabp3nz5jn++OPL77XRo0fn17/+ddZcc800b94822yzTf7973+nVCrltNNOyze/+c00a9Ys3/ve9/L2229XtP3R9+RCHTt2zJAhQ5J88H76wQ9+kCTZYostys/kwmvxv//7vxkwYEBWXXXVVFdXp1OnTjnttNOyYMGCinO46aab8uKLL5aPX3hfiz4z7rrrrvLz1rp163zve9/LU089tdjrXd/nfHGuvfba8j1YccUVs+eee+Y///lPxT4L3///+c9/stNOO6VFixZZaaWVcuSRR1ac5+JUVVVl3Lhxeffdd8vnfumlly7VZ+WnsfBz5tprr0337t3TrFmz9OnTJ0888USSZOzYsencuXOaNm2afv36Lfa9Wp9rlCR/+tOfsu6666Zp06ZZd91188c//nGxNdXV1eWXv/xl1llnnTRt2jQrr7xyDjzwwEybNm2ZnDMAfFEYiQ4A8AU0Y8aMvPnmmxXrVlxxxSTJFVdckcGDB2fbbbfNqFGjMnv27FxwwQXZdNNN8+ijj5YDrTvuuCPPP/989tlnn7Rr1y5PPvlkLrroojz55JO5//77U1VVlV122SXPPPNMrrrqqpx99tnlPlZaaaW88cYbS133D37wg3Tp0iVnnHFGOWgeMWJETjzxxAwaNCj7779/3njjjZx77rnZfPPN8+ijj36iKWSmTJmSPfbYIwceeGD23HPPjB49OjvssEMuvPDCHH/88Tn44IOTJCNHjsygQYPy9NNPp0GD/xs/smDBgvTv3z/f+c538vOf/zy33nprTj755Lz//vs59dRTkySlUik77rhj7r777uy3335Zf/31c9ttt+Woo47Kf/7zn5x99tkVNd1111255pprcsghh2TFFVfMeuutlwsuuCAHHXRQdt555+yyyy5Jkp49eyap3/35sEGDBmWNNdbIyJEj88gjj+Tiiy9O27ZtM2rUqPI+p5xySoYPH56NN944p556apo0aZJ//OMfueuuu7LNNtskqf/zs+uuu+bJJ5/MT37yk3Ts2DGvv/567rjjjrz00kufeBqQD7vjjjvywx/+MFtttVX5HJ566qlMnDgxhx566MceO23atPTv3z+77LJLBg0alD/84Q855phj0qNHj2y33XZJPvgCZMstt8yrr76aQw89NO3atcuVV16Zu+++e6nqHDRoUDp27JiRI0fm/vvvz69+9atMmzYtl19+eXmfpXnG33rrrWy33XbZfffds+eee2bllVcubxs/fnzmzZuXn/zkJ3n77bfz85//PIMGDcqWW26ZCRMm5JhjjsmUKVNy7rnn5sgjj8xvf/vbpTqXzTffPMOGDcuvfvWrHH/88enWrVuSlP+99NJL06JFixx++OFp0aJF7rrrrpx00kmZOXNmzjrrrCTJCSeckBkzZuTll18uvwdatGhR2Oedd96Z7bbbLmuuuWaGDx+e9957L+eee2422WSTPPLII4s8S/V5zhfn0ksvzT777JNvfetbGTlyZP773//mnHPOycSJExe5BwsWLMi2226bjTbaKKNHj86dd96ZMWPGpFOnTjnooIMK+7jiiity0UUX5YEHHsjFF1+cJNl4440/tq5l7W9/+1v+/Oc/58c//nGSDz7jBg4cmKOPPjrnn39+Dj744EybNi0///nPs+++++auu+4qH1vfa3T77bdn1113Tffu3TNy5Mi89dZb5S/UPurAAw8stzts2LC88MILOe+88/Loo49m4sSJX+mpfQD4mikBAPCFMW7cuFKSxS6lUqn0zjvvlFq3bl064IADKo577bXXSrW1tRXrZ8+evUj7V111VSlJ6a9//Wt53VlnnVVKUnrhhRcq9n3hhRdKSUrjxo1bpJ0kpZNPPrn8+uSTTy4lKf3whz+s2G/q1Kmlhg0blkaMGFGx/oknnig1atRokfVF1+PDtXXo0KGUpHTfffeV1912222lJKVmzZqVXnzxxfL6sWPHlpKU7r777vK6wYMHl5KUfvKTn5TX1dXVlQYMGFBq0qRJ6Y033iiVSqXSn/70p1KS0umnn15R0/e///1SVVVVacqUKRXXo0GDBqUnn3yyYt833nhjkWu1UH3vz8Jru++++1bsu/POO5e+8Y1vlF8/++yzpQYNGpR23nnn0oIFCyr2raurK5VK9X9+pk2bVkpSOuussxapcUn69u1b6tu37yLrBw8eXOrQoUP59aGHHlpq1apV6f333y9s6+67717k/vXt27eUpHT55ZeX182dO7fUrl270q677lpeN2bMmFKS0p/+9Kfyuvfee6/UtWvXRdpcnIXXfccdd6xYf/DBB5eSlB577LFSqbR0z/jC2i+88MKKfRe+11ZaaaXS9OnTy+uPO+64UpLSeuutV5o/f355/Q9/+MNSkyZNSnPmzCmvK3rOOnToUBo8eHD59bXXXlt4/ot7Jg888MBS8+bNK/oaMGBAxb386Hl8+DNj/fXXL7Vt27b01ltvldc99thjpQYNGpT23nvv8rr6PueLM2/evFLbtm1L6667bum9994rr7/xxhtLSUonnXRSed3C9/+pp55a0UavXr1KG2644cf2s/D4mpqaJZ73Qh+9L4v7TCt6zyyurerq6opjF37GtWvXrjRz5szy+oXPzsJ9l+Yarb/++qVVVlml4lm8/fbbS0kq7vvf/va3UpLS+PHjK+q89dZbF1lf33MEgC8q07kAAHwB/frXv84dd9xRsSQfjN6dPn16fvjDH+bNN98sLw0bNsxGG21UMcq2WbNm5b/nzJmTN998M9/5zneSJI888shnUvfQoUMrXl9//fWpq6vLoEGDKupt165dunTpstSjghfq3r17+vTpU3690UYbJUm23HLLrL766ousf/755xdp45BDDin/vXCahHnz5uXOO+9Mktx8881p2LBhhg0bVnHcEUcckVKplFtuuaVifd++fdO9e/d6n8PS3p+PXtvNNtssb731VmbOnJnkg+kX6urqctJJJ1WMul94fkn9n5+Fcz5PmDDhM5uWoXXr1nn33XfLz/bSaNGiRfbcc8/y6yZNmuTb3/52xX2+9dZbs9pqq2XHHXcsr2vatGkOOOCApepr4YjfhX7yk58k+eD5SJb+Ga+urs4+++yz2L5+8IMfpLa2tvx64fO75557plGjRhXr582bt9hpOD6NDz+T77zzTt58881sttlmmT17dv71r38tdXuvvvpqJk2alCFDhmSFFVYor+/Zs2e23nrr8jX8sCU954vz0EMP5fXXX8/BBx9cMSf7gAED0rVr19x000316mdxnxNfNFtttVXF6P2Fz8iuu+6ali1bLrJ+4TnV9xotvGeDBw+ueBa33nrrRT7frr322tTW1mbrrbeuePY33HDDtGjR4hN/vgPAF5HpXAAAvoC+/e1vL/aHRZ999tkkH4TFi9OqVavy32+//XZOOeWU/P73v8/rr79esd+MGTOWYbX/Z4011qh4/eyzz6ZUKqVLly6L3f+T/lf/DwflScphT/v27Re7/qNBcIMGDbLmmmtWrFtrrbWSpDyP8IsvvphVV121IphK/m/qixdffLFi/UfPfUmW9v589JzbtGmT5INza9WqVZ577rk0aNDgY4P8+j4/1dXVGTVqVI444oisvPLK+c53vpOBAwdm7733Trt27ep/kh/j4IMPzjXXXJPtttsuq622WrbZZpsMGjQo/fv3X+Kx3/zmNxeZ7qZNmzZ5/PHHy69ffPHFdOrUaZH9OnfuvFR1fvTZ7dSpUxo0aFB+Tpb2GV9ttdUKf5Ty0z7Xn9aTTz6Zn/3sZ7nrrrsWCa0/yWfGwvfI2muvvci2bt265bbbblvkB4iX9JwvbT9du3bNvffeW7GuadOm5d8r+HA/X4Z5vD/pM1Lfa7Rwv8U9z2uvvXbFF3zPPvtsZsyYkbZt2y621o9+rgHAl5kQHQDgS6Suri7JB3PzLi7M/PBo1UGDBuW+++7LUUcdlfXXXz8tWrRIXV1d+vfvX27n43w0fFzo435878MjWRfWW1VVlVtuuSUNGzZcZP+Pm0v54yyurY9bX/rID4F+Fj567kuytPdnWZzb0jw/hx12WHbYYYf86U9/ym233ZYTTzwxI0eOzF133ZVevXoV9lFVVbXYmj763LRt2zaTJk3KbbfdlltuuSW33HJLxo0bl7333juXXXbZx57H8rzPH31fLO0z/nHPyWfxXC/pxzIXmj59evr27ZtWrVrl1FNPTadOndK0adM88sgjOeaYY+r1mbEsfB73tqiPT+qTfFZ+Ul+kz766urq0bds248ePX+z2j35RAQBfZkJ0AIAvkU6dOiX5IID87ne/W7jftGnT8pe//CWnnHJKTjrppPL6hSORP6woAFo4AnT69OkV6z86AntJ9ZZKpayxxhrlkd5fBHV1dXn++ecranrmmWeSpDxVQocOHXLnnXfmnXfeqRiNvnBaiw4dOiyxn6JruzT3p746deqUurq6TJ48Oeuvv37hPsmSn58P73/EEUfkiCOOyLPPPpv1118/Y8aMye9+97vCY9q0abPYaTEW99w0adIkO+ywQ3bYYYfU1dXl4IMPztixY3PiiScu9Yjxj+rQoUMmT56cUqlUcR+mTJmyVO08++yzFf/LYMqUKamrqys/J1+UZ7xNmzaLvFfnzZuXV199tWJd0TM5YcKEvPXWW7n++uuz+eabl9e/8MILi+xb1MZHLXyPPP3004ts+9e//pUVV1yxYhT6J/Xhfj76vyyefvrper1XP41l8Vn5WavvNVr47+I+iz56Hzt16pQ777wzm2yyyVJ/iQgAXzbmRAcA+BLZdttt06pVq5xxxhmZP3/+ItvfeOONJP83KvGjoxB/+ctfLnLMwhDrowFQq1atsuKKK+avf/1rxfrzzz+/3vXusssuadiwYU455ZRFaimVSnnrrbfq3daydt5551XUct5556Vx48bZaqutkiTbb799FixYULFfkpx99tmpqqrKdtttt8Q+mjdvnmTRa7s096e+dtpppzRo0CCnnnrqIqOGF/ZT3+dn9uzZmTNnTsW2Tp06pWXLlpk7d+7H1tGpU6f861//KreVJI899lgmTpxYsd9H732DBg3Ss2fPJFliH/Wx7bbb5j//+U/+/Oc/l9fNmTMnv/nNb5aqnV//+tcVr88999wkKd//L8oz3qlTp0XeqxdddNEio6GL3u+LeybnzZu32Pd7TU1NvaZ3WWWVVbL++uvnsssuq+jvn//8Z26//fZsv/32S2yjPnr37p22bdvmwgsvrHh2brnlljz11FMZMGDAMumnyLL4rPys1fcaffieffge33HHHZk8eXJFm4MGDcqCBQty2mmnLdLf+++/v8gzBgBfZkaiAwB8ibRq1SoXXHBB9tprr2ywwQbZfffds9JKK+Wll17KTTfdlE022STnnXdeWrVqlc033zw///nPM3/+/Ky22mq5/fbbFzuqdMMNN0ySnHDCCdl9993TuHHj7LDDDqmpqcn++++fM888M/vvv3969+6dv/71r+UR2/XRqVOnnH766TnuuOMyderU7LTTTmnZsmVeeOGF/PGPf8z//M//5Mgjj1xm16e+mjZtmltvvTWDBw/ORhttlFtuuSU33XRTjj/++PIUBDvssEO22GKLnHDCCZk6dWrWW2+93H777fnf//3fHHbYYeVR3R+nWbNm6d69e66++uqstdZaWWGFFbLuuutm3XXXrff9qa/OnTvnhBNOyGmnnZbNNtssu+yyS6qrq/Pggw9m1VVXzciRI+v9/DzzzDPZaqutMmjQoHTv3j2NGjXKH//4x/z3v//N7rvv/rF17LvvvvnFL36RbbfdNvvtt19ef/31XHjhhVlnnXUq5tnef//98/bbb2fLLbfMN7/5zbz44os599xzs/7665fnnf80DjzwwJx33nn54Q9/mEMPPTSrrLJKxo8fX/5RxfqOpn7hhRey4447pn///vn73/+e3/3ud9ljjz2y3nrrJfniPOP7779/hg4dml133TVbb711Hnvssdx2221ZccUVK/Zbf/3107Bhw4waNSozZsxIdXV1ttxyy2y88cZp06ZNBg8enGHDhqWqqipXXHHFYqcD2XDDDXP11Vfn8MMPz7e+9a20aNEiO+yww2LrOuuss7LddtulT58+2W+//fLee+/l3HPPTW1tbYYPH75Mzr1x48YZNWpU9tlnn/Tt2zc//OEP89///jfnnHNOOnbsmJ/+9KfLpJ+P82k/Kz9rS3ONRo4cmQEDBmTTTTfNvvvum7fffjvnnntu1llnncyaNau8X9++fXPggQdm5MiRmTRpUrbZZps0btw4zz77bK699tqcc845+f73v788ThcAlr0SAABfGOPGjSslKT344IMfu9/dd99d2nbbbUu1tbWlpk2bljp16lQaMmRI6aGHHirv8/LLL5d23nnnUuvWrUu1tbWlH/zgB6VXXnmllKR08sknV7R32mmnlVZbbbVSgwYNSklKL7zwQqlUKpVmz55d2m+//Uq1tbWlli1blgYNGlR6/fXXF2nj5JNPLiUpvfHGG4ut97rrrittuummpZqamlJNTU2pa9eupR//+Melp59+ul7XY2E9pVKp1KFDh9KAAQMW2TdJ6cc//nHFuhdeeKGUpHTWWWeV1w0ePLhUU1NTeu6550rbbLNNqXnz5qWVV165dPLJJ5cWLFhQcfw777xT+ulPf1paddVVS40bNy516dKldNZZZ5Xq6uqW2PdC9913X2nDDTcsNWnSpOK61ff+FF3bxV2bUqlU+u1vf1vq1atXqbq6utSmTZtS3759S3fccUfFPkt6ft58883Sj3/841LXrl1LNTU1pdra2tJGG21UuuaaaxZ7jh/1u9/9rrTmmmuWmjRpUlp//fVLt912W2nw4MGlDh06lPf5wx/+UNpmm21Kbdu2LTVp0qS0+uqrlw488MDSq6++WlFnktLdd99dXte3b9/SOuuss0ifH22/VCqVnn/++dKAAQNKzZo1K6200kqlI444onTdddeVkpTuv//+jz2Hhdd98uTJpe9///ulli1bltq0aVM65JBDSu+9994i+9fnGS+qfXHP6YfP/9prr61Yv7jPiQULFpSOOeaY0oorrlhq3rx5adttty1NmTKl1KFDh9LgwYMrjv/Nb35TWnPNNUsNGzasuL4TJ04sfec73yk1a9astOqqq5aOPvro0m233bbIPZg1a1Zpjz32KLVu3bqUpHzdF57HuHHjKvq78847S5tsskmpWbNmpVatWpV22GGH0uTJkxd7vev7nC/O1VdfXX72V1hhhdKPfvSj0ssvv1yxz8L3/0ct7H9Jio6v72fl4s6nb9++pb59+y6x7/p+xpVKxc9Ofa5RqfTB89ytW7dSdXV1qXv37qXrr79+se+xUqlUuuiii0obbrhhqVmzZqWWLVuWevToUTr66KNLr7zyylKfIwB8UVWVSp/Dr+8AAMAXxJAhQ/KHP/yhYkQlXx+//OUv89Of/jQvv/xyVltttcL9hg8fnlNOOSVvvPHGIqO5AQD4ejEnOgAA8JX03nvvVbyeM2dOxo4dmy5dunxsgA4AAB9mTnQAAOAraZdddsnqq6+e9ddfPzNmzMjvfve7/Otf/8r48eOXd2kAAHyJCNEBAICvpG233TYXX3xxxo8fnwULFqR79+75/e9/n9122215lwYAwJeIOdEBAAAAAKCAOdEBAAAAAKCA6Vz4Wqqrq8srr7ySli1bpqqqanmXAwAAAAB8zkqlUt55552suuqqadCgeLy5EJ2vpVdeeSXt27df3mUAAAAAAMvZv//973zzm98s3C5E52upZcuWST54g7Rq1Wo5VwMAAAAAfN5mzpyZ9u3bl7PCIkJ0vpYWTuHSqlUrIToAAAAAfI0tabpnPywKAAAAAAAFhOgAAAAAAFBAiA4AAAAAAAWE6AAAAAAAUECIDgAAAAAABYToAAAAAABQQIgOAAAAAAAFhOgAAAAAAFBAiA4AAAAAAAWE6AAAAAAAUECIDgAAAAAABYToAAAAAABQQIgOAAAAAAAFhOgAAAAAAFBAiA4AAAAAAAWE6AAAAAAAUECIDgAAAAAABYToAAAAAABQoNHyLgBgoY7H3rS8S/jCmnrmgOVdAgAAAMDXkpHoAAAAAABQQIgOAAAAAAAFhOgAAAAAAFBAiA4AAAAAAAWE6AAAAAAAUECIDgAAAAAABYToAAAAAABQQIgOAAAAAAAFhOgAAAAAAFBAiA4AAAAAAAWE6AAAAAAAUECIDgAAAAAABYToAAAAAABQQIgOAAAAAAAFhOgAAAAAAFBguYfow4cPz/rrr/+ZtD1hwoRUVVVl+vTpy6zNqVOnpqqqKpMmTVpmbX5W+vXrl8MOO2x5lwEAAAAA8KW1VCH6kCFDUlVVtcjSv3//z6q+L70//vGP+c53vpPa2tq0bNky66yzzjIPtj+LLws+D1VVVfnTn/60vMsAAAAAACjUaGkP6N+/f8aNG1exrrq6epkVtKzMnz9/eZeQv/zlL9ltt90yYsSI7LjjjqmqqsrkyZNzxx13LO/SAAAAAACoh6WezqW6ujrt2rWrWNq0aZPkg5HFY8eOzcCBA9O8efN069Ytf//73zNlypT069cvNTU12XjjjfPcc88t0u7YsWPTvn37NG/ePIMGDcqMGTPK2x588MFsvfXWWXHFFVNbW5u+ffvmkUceqTi+qqoqF1xwQXbcccfU1NRkxIgRi/Qxe/bsbLfddtlkk03Ko7YvvvjidOvWLU2bNk3Xrl1z/vnnVxzzwAMPpFevXmnatGl69+6dRx99tN7X6oYbbsgmm2ySo446KmuvvXbWWmut7LTTTvn1r39dsd8FF1yQTp06pUmTJll77bVzxRVXlLctbvqY6dOnp6qqKhMmTMjUqVOzxRZbJEnatGmTqqqqDBkypLxvXV1djj766Kywwgpp165dhg8fXtH3L37xi/To0SM1NTVp3759Dj744MyaNau8/dJLL03r1q1z4403Zu21107z5s3z/e9/P7Nnz85ll12Wjh07pk2bNhk2bFgWLFhQPq5jx4457bTT8sMf/jA1NTVZbbXVKs67Y8eOSZKdd945VVVV5ddLuh7JB/f64osvzs4775zmzZunS5cu+fOf//yx92Lu3LmZOXNmxQIAAAAAsCTLfE700047LXvvvXcmTZqUrl27Zo899siBBx6Y4447Lg899FBKpVIOOeSQimOmTJmSa665JjfccENuvfXWPProozn44IPL2995550MHjw49957b+6///506dIl22+/fd55552KdoYPH56dd945TzzxRPbdd9+KbdOnT8/WW2+durq63HHHHWndunXGjx+fk046KSNGjMhTTz2VM844IyeeeGIuu+yyJMmsWbMycODAdO/ePQ8//HCGDx+eI488st7Xol27dnnyySfzz3/+s3CfP/7xjzn00ENzxBFH5J///GcOPPDA7LPPPrn77rvr1Uf79u1z3XXXJUmefvrpvPrqqznnnHPK2y+77LLU1NTkH//4R37+85/n1FNPrRgJ36BBg/zqV7/Kk08+mcsuuyx33XVXjj766Io+Zs+enV/96lf5/e9/n1tvvTUTJkzIzjvvnJtvvjk333xzrrjiiowdOzZ/+MMfKo4766yzst566+XRRx/Nsccem0MPPbTc94MPPpgkGTduXF599dXy6/pej1NOOSWDBg3K448/nu233z4/+tGP8vbbbxdep5EjR6a2tra8tG/fvl7XF5a3unlz8uKogamqqsq77767vMsBAAAA+NqpKpVKpfruPGTIkPzud79L06ZNK9Yff/zxOf7441NVVZWf/exnOe2005Ik999/f/r06ZNLLrmkHGr//ve/zz777JP33nsvyQfB9+mnn54XX3wxq622WpLk1ltvzYABA/Kf//wn7dq1W6SOurq6tG7dOldeeWUGDhz4wYlUVeWwww7L2WefXd5vwoQJ2WKLLfLUU09lt912S5cuXXLllVemSZMmSZLOnTuXR0svdPrpp+fmm2/Offfdl4suuijHH398Xn755fI5X3jhhTnooIPy6KOPLvEHUd99990MGjQoN998czp06JDvfOc72WabbfKjH/2oPAXOJptsknXWWScXXXRR+bhBgwbl3XffzU033ZSpU6dmjTXWqOhv+vTpadOmTe6+++7069evfJ7Tpk1L69aty+3069cvCxYsyN/+9rfyum9/+9vZcsstc+aZZy625j/84Q8ZOnRo3nzzzSQfjETfZ599MmXKlHTq1ClJMnTo0FxxxRX573//mxYtWiT5YJqfjh075sILL0zywUjzbt265ZZbbim3vfvuu2fmzJm5+eaby/fsj3/8Y3baaafyPku6HguP+/Bz9u6776ZFixa55ZZbCufnnzt3bubOnVt+PXPmzLRv3z4zZsxIq1atFnsMn7+Ox960vEv4wpp65oDlXQIAAADAV8rMmTNTW1u7xIxwqUeib7HFFpk0aVLFMnTo0PL2nj17lv9eeeWVkyQ9evSoWDdnzpyK6TRWX331coCeJH369EldXV2efvrpJMl///vfHHDAAenSpUtqa2vTqlWrzJo1Ky+99FJFbb17915szVtvvXU6d+6cq6++uhygv/vuu3nuueey3377pUWLFuXl9NNPL08389RTT6Vnz54VXxr06dOn3teqpqYmN910U6ZMmZKf/exnadGiRY444oh8+9vfzuzZs8t9bLLJJhXHbbLJJnnqqafq3c/H+fD9SJJVVlklr7/+evn1nXfema222iqrrbZaWrZsmb322itvvfVWub4kad68eTlATz64hx07diwH6AvXfbjdZNFr1adPnyWeV32vx4fPq6amJq1atVqk/w+rrq5Oq1atKhYAAAAAgCVZ6h8WrampSefOnQu3N27cuPx3VVVV4bq6urp69zl48OC89dZbOeecc9KhQ4dUV1enT58+mTdv3iK1Lc6AAQNy3XXXZfLkyeVAf+G837/5zW+y0UYbVezfsGHDetdWH506dUqnTp2y//7754QTTshaa62Vq6++Ovvss88Sj23Q4IPvOT78HwaW5kdTP3ztkw+u/8JrP3Xq1AwcODAHHXRQRowYkRVWWCH33ntv9ttvv8ybNy/NmzcvbOPj2v08LO/+AQAAAICvh2U+J/on8dJLL+WVV14pv77//vvToEGDrL322kmSiRMnZtiwYdl+++2zzjrrpLq6ujzdSH2ceeaZGTx4cLbaaqtMnjw5yQcjp1ddddU8//zz6dy5c8WyxhprJEm6deuWxx9/PHPmzKmo7dPo2LFjmjdvXp7buFu3bpk4cWLFPhMnTkz37t2TJCuttFKS5NVXXy1v//CPjCYpj67/8A971sfDDz+curq6jBkzJt/5zney1lprVdyHT+uj1+r+++9Pt27dyq8bN268SM1Luh4AAAAAAJ+npR6JPnfu3Lz22muVjTRqlBVXXPETF9G0adMMHjw4o0ePzsyZMzNs2LAMGjSoPB96ly5dcsUVV6R3796ZOXNmjjrqqDRr1myp+hg9enQWLFiQLbfcMhMmTEjXrl1zyimnZNiwYamtrU3//v0zd+7cPPTQQ5k2bVoOP/zw7LHHHjnhhBNywAEH5LjjjsvUqVMzevToevc5fPjwzJ49O9tvv306dOiQ6dOn51e/+lXmz5+frbfeOkly1FFHZdCgQenVq1e++93v5oYbbsj111+fO++8M0nSrFmzfOc738mZZ56ZNdZYI6+//np+9rOfVfTToUOHVFVV5cYbb8z222+fZs2aVUy1UqRz586ZP39+zj333Oywww6ZOHFieU7zZWHixIn5+c9/np122il33HFHrr322vK85skHXyj85S9/ySabbJLq6uq0adNmidcDAAAAAODztNQj0W+99dasssoqFcumm276qYro3Llzdtlll2y//fbZZptt0rNnz5x//vnl7ZdcckmmTZuWDTbYIHvttVeGDRuWtm3bLnU/Z599dgYNGpQtt9wyzzzzTPbff/9cfPHFGTduXHr06JG+ffvm0ksvLY9Eb9GiRW644YY88cQT6dWrV0444YSMGjWq3v317ds3zz//fPbee+907do12223XV577bXcfvvt5VH2O+20U84555yMHj0666yzTsaOHZtx48alX79+5XZ++9vf5v3338+GG26Yww47LKeffnpFP6uttlpOOeWUHHvssVl55ZVzyCGH1Ku+9dZbL7/4xS8yatSorLvuuhk/fnxGjhxZ7/NbkiOOOCIPPfRQevXqldNPPz2/+MUvsu2225a3jxkzJnfccUfat2+fXr16Janf9QAAAAAA+LxUlT482TYsIx07dsxhhx2Www47bHmXslj1/eVdPl8dj71pyTt9TU09c8DyLgEAAADgK6W+GeEXYk50AAAAAAD4IhKifwpDhw5NixYtFrsMHTp0eZcHAAAAAMCntNQ/LMr/OfXUU3PkkUcudtvXfYqQqVOnLu8SAAAAAAA+NSH6p9C2bdtP9AOnAAAAAAB8OZjOBQAAAAAACgjRAQAAAACggBAdAAAAAAAKCNEBAAAAAKCAEB0AAAAAAAoI0QEAAAAAoIAQHQAAAAAACgjRAQAAAACggBAdAAAAAAAKNFreBQAsNPXMAcu7BAAAAACoYCQ6AAAAAAAUEKIDAAAAAEABIToAAAAAABQQogMAAAAAQAEhOgAAAAAAFBCiAwAAAABAASE6AAAAAAAUEKIDAAAAAEABIToAAAAAABQQogMAAAAAQAEhOgAAAAAAFBCiAwAAAABAASE6AAAAAAAUEKIDAAAAAEABIToAAAAAABQQogMAAAAAQAEhOgAAAAAAFBCiAwAAAABAASE6AAAAAAAUEKIDAAAAAEABIToAAAAAABQQogMAAAAAQAEhOgAAAAAAFBCiAwAAAABAASE6AAAAAAAUEKIDAAAAAEABIToAAAAAABQQogMAAAAAQAEhOgAAAAAAFBCiAwAAAABAASE6AAAAAAAUEKIDAAAAAEABIToAAAAAABQQogMAAAAAQAEhOgAAAAAAFBCiAwAAAABAASE6AAAAAAAUEKIDAAAAAEABIToAAAAAABQQogMAAAAAQIFGy7sA4LPX8diblncJ9TL1zAHLuwQAAAAAqGAkOgAAAAAAFBCiAwAAAABAASE6AAAAAAAUEKIDAAAAAEABIToAAAAAABQQogMAAAAAQAEhOgAAAAAAFBCiAwAAAABAASE6AAAAAAAUEKIDAAAAAEABIToAAAAAABQQogMAAAAAQAEhOgAAAAAAFBCiAwAAAABAASE6AAAAAAAU+MKH6MOHD8/666//mbQ9YcKEVFVVZfr06cuszalTp6aqqiqTJk1aZm0CAAAAALB8LNMQfciQIamqqlpk6d+//7Ls5ivnsssuy7e+9a00b948LVu2TN++fXPjjTcu77I+kX79+uWwww5b3mUAAAAAACwTy3wkev/+/fPqq69WLFddddWy7uZTmz9//vIuIUly5JFH5sADD8xuu+2Wxx9/PA888EA23XTTfO9738t55523vMsDAAAAAPhaW+YhenV1ddq1a1extGnTJklSVVWVsWPHZuDAgWnevHm6deuWv//975kyZUr69euXmpqabLzxxnnuuecWaXfs2LFp3759mjdvnkGDBmXGjBnlbQ8++GC23nrrrLjiiqmtrU3fvn3zyCOPVBxfVVWVCy64IDvuuGNqamoyYsSIRfqYPXt2tttuu2yyySblKV4uvvjidOvWLU2bNk3Xrl1z/vnnVxzzwAMPpFevXmnatGl69+6dRx99tN7X6v7778+YMWNy1lln5cgjj0znzp3TrVu3jBgxIocddlgOP/zw/Pvf/06y+GltfvnLX6Zjx44V65ZU77///e8MGjQorVu3zgorrJDvfe97mTp1ann7kCFDstNOO2X06NFZZZVV8o1vfCM//vGPK750OP/889OlS5c0bdo0K6+8cr7//e+Xj73nnntyzjnnlP8XwsK277nnnnz7299OdXV1VllllRx77LF5//33kyQ33nhjWrdunQULFiRJJk2alKqqqhx77LHlPvfff//sueeeSZJLL700rVu3zm233ZZu3bqlRYsW5S9vAAAAAACWpc99TvTTTjste++9dyZNmpSuXbtmjz32yIEHHpjjjjsuDz30UEqlUg455JCKY6ZMmZJrrrkmN9xwQ2699dY8+uijOfjgg8vb33nnnQwePDj33ntv7r///nTp0iXbb7993nnnnYp2hg8fnp133jlPPPFE9t1334pt06dPz9Zbb526urrccccdad26dcaPH5+TTjopI0aMyFNPPZUzzjgjJ554Yi677LIkyaxZszJw4MB07949Dz/8cIYPH54jjzyy3tfiqquuSosWLXLggQcusu2II47I/Pnzc91119W7vSXVO3/+/Gy77bZp2bJl/va3v2XixInlAHrevHnldu6+++4899xzufvuu3PZZZfl0ksvzaWXXpokeeihhzJs2LCceuqpefrpp3Prrbdm8803T5Kcc8456dOnTw444IDy/0Jo3759/vOf/2T77bfPt771rTz22GO54IILcskll+T0009Pkmy22WZ55513yl9A3HPPPVlxxRUzYcKEck333HNP+vXrV349e/bsjB49OldccUX++te/5qWXXvrYaz937tzMnDmzYvk6ePfdd1NVVZUXRw1M3bw5y7scAAAAAPjSabSsG7zxxhvTokWLinXHH398jj/++CTJPvvsk0GDBiVJjjnmmPTp0ycnnnhitt122yTJoYcemn322afi+Dlz5uTyyy/PaqutliQ599xzM2DAgIwZMybt2rXLlltuWbH/RRddlNatW+eee+7JwIEDy+v32GOPiraff/75JMlrr72W3XbbLV26dMmVV16ZJk2aJElOPvnkjBkzJrvsskuSZI011sjkyZMzduzYDB48OFdeeWXq6upyySWXpGnTpllnnXXy8ssv56CDDqrXtXrmmWfSqVOncn8ftuqqq6ZVq1Z55pln6tVWfeq9+uqrU1dXl4svvjhVVVVJknHjxqV169aZMGFCttlmmyRJmzZtct5556Vhw4bp2rVrBgwYkL/85S854IAD8tJLL6WmpiYDBw5My5Yt06FDh/Tq1StJUltbmyZNmqR58+Zp165dua7zzz8/7du3z3nnnZeqqqp07do1r7zySo455picdNJJqa2tzfrrr58JEyakd+/emTBhQn7605/mlFNOyaxZszJjxoxMmTIlffv2Lbc5f/78XHjhhenUqVOS5JBDDsmpp55aeG1GjhyZU045pd7X8quipqYmpVIpHY+9aXmXAgAAAABfSst8JPoWW2yRSZMmVSxDhw4tb+/Zs2f575VXXjlJ0qNHj4p1c+bMqRgpvPrqq5cD9CTp06dP6urq8vTTTydJ/vvf/+aAAw5Ily5dUltbm1atWmXWrFl56aWXKmrr3bv3Ymveeuut07lz51x99dXlQPvdd9/Nc889l/322y8tWrQoL6effnp5upmnnnoqPXv2TNOmTStqWxqlUuljty8uYF+c+tT72GOPZcqUKWnZsmV5+worrJA5c+ZUTKGzzjrrpGHDhuXXq6yySl5//fUkH1yrDh06ZM0118xee+2V8ePHZ/bs2R9b21NPPZU+ffqUg/sk2WSTTTJr1qy8/PLLSZK+fftmwoQJKZVK+dvf/pZddtkl3bp1y7333pt77rknq666arp06VI+vnnz5uUA/aM1Ls5xxx2XGTNmlJeF0+QAAAAAAHycZT4SvaamJp07dy7c3rhx4/LfC0PVxa2rq6urd5+DBw/OW2+9lXPOOScdOnRIdXV1+vTpUzFFycLaFmfAgAG57rrrMnny5HKgP2vWrCTJb37zm2y00UYV+384YP40unTpknvvvTfz5s1bJCx/5ZVXMnPmzKy11lpJkgYNGiwSuH94nvL61Dtr1qxsuOGGGT9+/CK1rLTSSuW/P3w/kg/uycL70bJlyzzyyCOZMGFCbr/99px00kkZPnx4HnzwwbRu3XppTr9Cv3798tvf/jaPPfZYGjdunK5du6Zfv36ZMGFCpk2bVjEKvajGj/tCorq6OtXV1Z+4PgAAAADg6+lznxP9k3jppZfyyiuvlF/ff//9adCgQdZee+0kycSJEzNs2LBsv/32WWeddVJdXZ0333yz3u2feeaZGTx4cLbaaqtMnjw5yQcj4lddddU8//zz6dy5c8WyxhprJEm6deuWxx9/PHPm/N9c0/fff3+9+/3hD3+YWbNmZezYsYtsGz16dJo2bZrddtstyQch92uvvVYRFE+aNKn8d33q3WCDDfLss8+mbdu2i+xTW1tb77obNWqU7373u/n5z3+exx9/PFOnTs1dd92V5IOR8wt/IHShhT8g++HaJ06cmJYtW+ab3/xmkv+bF/3ss88uB+YLQ/QJEyZUzIcOAAAAAPB5WeYj0efOnZvXXnutspNGjbLiiit+4jabNm2awYMHZ/To0Zk5c2aGDRuWQYMGlefd7tKlS6644or07t07M2fOzFFHHZVmzZotVR+jR4/OggULsuWWW2bChAnp2rVrTjnllAwbNiy1tbXp379/5s6dm4ceeijTpk3L4Ycfnj322CMnnHBCDjjggBx33HGZOnVqRo8eXe8++/Tpk0MPPTRHHXVU5s2bl5122inz58/P7373u/zqV7/KpZdemm984xtJPgiU33jjjfz85z/P97///dx666255ZZb0qpVq3J7S6r3Rz/6Uc4666x873vfy6mnnppvfvObefHFF3P99dfn6KOPLgfaH+fGG2/M888/n8033zxt2rTJzTffnLq6uvIXGh07dsw//vGPTJ06tTxdzMEHH5xf/vKX+clPfpJDDjkkTz/9dE4++eQcfvjhadDgg+9x2rRpk549e2b8+PE577zzkiSbb755Bg0alPnz5y8yEh0AAAAA4POwzEei33rrrVlllVUqlk033fRTtdm5c+fssssu2X777bPNNtukZ8+eOf/888vbL7nkkkybNi0bbLBB9tprrwwbNixt27Zd6n7OPvvsDBo0KFtuuWWeeeaZ7L///rn44oszbty49OjRI3379s2ll15aHtndokWL3HDDDXniiSfSq1evnHDCCRk1atRS9fnLX/4y559/fq666qqsu+666datW84666zcdddd2XPPPcv7devWLeeff35+/etfZ7311ssDDzyQI488sqKtJdXbvHnz/PWvf83qq69ennN8v/32y5w5cyrC+I/TunXrXH/99dlyyy3TrVu3XHjhhbnqqquyzjrrJEmOPPLINGzYMN27d89KK62Ul156KauttlpuvvnmPPDAA1lvvfUydOjQ7LfffvnZz35W0Xbfvn2zYMGC8qjzFVZYId27d0+7du3KIT0AAAAAwOepqrSkX7bkczV16tT07ds3ffr0yfjx45fZ/OtUmjlzZmprazNjxox6f4HwZdbx2JuWdwn1MvXMAcu7BAAAAAC+JuqbEX4p5kT/OunYsWN5OpkPz3kOAAAAAMDnT4j+GRo6dGhatGix2GXo0KGFx62xxhoZPnx4Ntxww8+xWgAAAAAAPmqZ/7Ao/+fUU09dZN7yhb4OU4gAAAAAAHzZCdE/Q23btv1EP3AKAAAAAMAXg+lcAAAAAACggBAdAAAAAAAKCNEBAAAAAKCAEB0AAAAAAAoI0QEAAAAAoIAQHQAAAAAACgjRAQAAAACggBAdAAAAAAAKCNEBAAAAAKBAo+VdAPDZm3rmgOVdAgAAAAB8KRmJDgAAAAAABYToAAAAAABQQIgOAAAAAAAFhOgAAAAAAFBAiA4AAAAAAAWE6AAAAAAAUECIDgAAAAAABYToAAAAAABQQIgOAAAAAAAFhOgAAAAAAFBAiA4AAAAAAAWE6AAAAAAAUECIDgAAAAAABYToAAAAAABQQIgOAAAAAAAFhOgAAAAAAFBAiA4AAAAAAAWE6AAAAAAAUECIDgAAAAAABYToAAAAAABQQIgOAAAAAAAFhOgAAAAAAFBAiA4AAAAAAAWE6AAAAAAAUECIDgAAAAAABYToAAAAAABQQIgOAAAAAAAFhOgAAAAAAFBAiA4AAAAAAAWE6AAAAAAAUECIDgAAAAAABYToAAAAAABQQIgOAAAAAAAFhOgAAAAAAFBAiA4AAAAAAAWE6AAAAAAAUECIDgAAAAAABYToAAAAAABQQIgOAAAAAAAFhOgAAAAAAFCg0fIuAPhy6njsTcu8zalnDljmbQIAAADAp2EkOgAAAAAAFBCiAwAAAABAASE6AAAAAAAUEKIDAAAAAEABIToAAAAAABQQogMAAAAAQAEhOgAAAAAAFBCiAwAAAABAASE6AAAAAAAUEKIDAAAAAEABIToAAAAAABQQogMAAAAAQAEhOgAAAAAAFBCiAwAAAABAASE6AAAAAAAUEKIDAAAAAECBr2SIPnz48Ky//vqfSdsTJkxIVVVVpk+fvszanDp1aqqqqjJp0qRl1uZneQ0+y7YBAAAAAL5IlnuIPmTIkFRVVS2y9O/ff3mX9oV23XXXpV+/fqmtrU2LFi3Ss2fPnHrqqXn77beXd2kAAAAAAF8Zyz1ET5L+/fvn1VdfrViuuuqq5V3WIubPn7+8S0iSnHDCCdltt93yrW99K7fcckv++c9/ZsyYMXnsscdyxRVXfGb9lkqlvP/++59Z+0uyYMGC1NXVLbf+AQAAAICvny9EiF5dXZ127dpVLG3atEmSVFVVZezYsRk4cGCaN2+ebt265e9//3umTJmSfv36paamJhtvvHGee+65RdodO3Zs2rdvn+bNm2fQoEGZMWNGeduDDz6YrbfeOiuuuGJqa2vTt2/fPPLIIxXHV1VV5YILLsiOO+6YmpqajBgxYpE+Zs+ene222y6bbLJJeYqXiy++ON26dUvTpk3TtWvXnH/++RXHPPDAA+nVq1eaNm2a3r1759FHH633tXrggQdyxhlnZMyYMTnrrLOy8cYbp2PHjtl6661z3XXXZfDgwRX7X3HFFenYsWNqa2uz++6755133ilvq6ury8iRI7PGGmukWbNmWW+99fKHP/yhvH3h1DW33HJLNtxww1RXV+fee++t1/Wtq6vLqaeemm9+85uprq7O+uuvn1tvvXWRtj88Lc6kSZNSVVWVqVOnJkkuvfTStG7dOn/+85/TvXv3VFdX56WXXsqrr76aAQMGpFmzZlljjTVy5ZVXpmPHjvnlL39ZeN3mzp2bmTNnVix8Mu+++26qqqry4qiBqZs3Z3mXAwAAAACfqS9EiL4kp512Wvbee+9MmjQpXbt2zR577JEDDzwwxx13XB566KGUSqUccsghFcdMmTIl11xzTW644YbceuutefTRR3PwwQeXt7/zzjsZPHhw7r333tx///3p0qVLtt9++4qQOflg/u+dd945TzzxRPbdd9+KbdOnT8/WW2+durq63HHHHWndunXGjx+fk046KSNGjMhTTz2VM844IyeeeGIuu+yyJMmsWbMycODAdO/ePQ8//HCGDx+eI488st7XYvz48WnRokXFuXxY69aty38/99xz+dOf/pQbb7wxN954Y+65556ceeaZ5e0jR47M5ZdfngsvvDBPPvlkfvrTn2bPPffMPffcU9HmsccemzPPPDNPPfVUevbsWa/re84552TMmDEZPXp0Hn/88Wy77bbZcccd8+yzz9b7XJMPvqQYNWpULr744jz55JNp27Zt9t5777zyyiuZMGFCrrvuulx00UV5/fXXP7adkSNHpra2try0b99+qeoAAAAAAL6eGi3vApLkxhtvTIsWLSrWHX/88Tn++OOTJPvss08GDRqUJDnmmGPSp0+fnHjiidl2222TJIceemj22WefiuPnzJmTyy+/PKuttlqS5Nxzz82AAQMyZsyYtGvXLltuuWXF/hdddFFat26de+65JwMHDiyv32OPPSrafv7555Mkr732Wnbbbbd06dIlV155ZZo0aZIkOfnkkzNmzJjssssuSZI11lgjkydPztixYzN48OBceeWVqauryyWXXJKmTZtmnXXWycsvv5yDDjqoXtfq2WefzZprrpnGjRsvcd+6urpceumladmyZZJkr732yl/+8peMGDEic+fOzRlnnJE777wzffr0SZKsueaauffeezN27Nj07du33M6pp56arbfeeqmu7+jRo3PMMcdk9913T5KMGjUqd999d375y1/m17/+db3ONflgCp3zzz8/6623XpLkX//6V+688848+OCD6d27d5IPRv536dLlY9s57rjjcvjhh5dfz5w5U5D+CdXU1KRUKqXjsTct71IAAAAA4DP3hQjRt9hii1xwwQUV61ZYYYXy3wtHPyfJyiuvnCTp0aNHxbo5c+Zk5syZadWqVZJk9dVXLwe8SdKnT5/U1dXl6aefTrt27fLf//43P/vZzzJhwoS8/vrrWbBgQWbPnp2XXnqpoo6FQe1Hbb311vn2t7+dq6++Og0bNkzywTQXzz33XPbbb78ccMAB5X3ff//91NbWJkl5NHfTpk0raquvUqlU7307duxYDtCTZJVVVimP2J4yZUpmz569SDg+b9689OrVq2Ld4q7Bx13f5s2b55VXXskmm2xSccwmm2ySxx57rN71J0mTJk0q7v/TTz+dRo0aZYMNNiiv69y5c3n6nyLV1dWprq5eqr4BAAAAAL4QIXpNTU06d+5cuP3Do66rqqoK1y3Nj04OHjw4b731Vs4555x06NAh1dXV6dOnT+bNm7dIbYszYMCAXHfddZk8eXI50J81a1aS5De/+U022mijiv0XBu2f1lprrZV777038+fPX+Jo9I9ur6qqKl+jhbXedNNNFWF4kkXC5qJr8Gk0aPDBTEIf/lJgcT/c2qxZs/L9BQAAAAD4vH0p5kT/JF566aW88sor5df3339/GjRokLXXXjtJMnHixAwbNizbb7991llnnVRXV+fNN9+sd/tnnnlmBg8enK222iqTJ09O8sGI+FVXXTXPP/98OnfuXLGsscYaSZJu3brl8ccfz5w5//eDjPfff3+9+91jjz0ya9asRX6sdKEP/1Dnx/nwD3V+tNb6THPycde3VatWWXXVVTNx4sSKYyZOnJju3bsnSVZaaaUkyauvvlrePmnSpCX2u/baa+f999+v+DHWKVOmZNq0aUs8FgAAAABgaX0hRqLPnTs3r732WsW6Ro0aZcUVV/zEbTZt2jSDBw/O6NGjM3PmzAwbNiyDBg1Ku3btkiRdunTJFVdckd69e2fmzJk56qij0qxZs6XqY/To0VmwYEG23HLLTJgwIV27ds0pp5ySYcOGpba2Nv3798/cuXPz0EMPZdq0aTn88MOzxx575IQTTsgBBxyQ4447LlOnTs3o0aPr3edGG22Uo48+OkcccUT+85//ZOedd86qq66aKVOm5MILL8ymm26aQw89dInttGzZMkceeWR++tOfpq6uLptuumlmzJiRiRMnplWrVhk8ePDHHr+k63vUUUfl5JNPTqdOnbL++utn3LhxmTRpUsaPH58k5bB++PDhGTFiRJ555pmMGTNmiXV37do13/3ud/M///M/ueCCC9K4ceMcccQRRqwDAAAAAJ+JL0SIfuutt2aVVVapWLf22mvnX//61ydus3Pnztlll12y/fbb5+23387AgQMrRm9fcskl+Z//+Z9ssMEGad++fc4444wceeSRS93P2WefXRGk77///mnevHnOOuusHHXUUampqUmPHj1y2GGHJUlatGiRG264IUOHDk2vXr3SvXv3jBo1Krvuumu9+xw1alQ23HDD/PrXv86FF16Yurq6dOrUKd///veXGH5/2GmnnZaVVlopI0eOzPPPP5/WrVtngw02KP+g68dZ0vUdNmxYZsyYkSOOOCKvv/56unfvnj//+c/lHwBt3Lhxrrrqqhx00EHp2bNnvvWtb+X000/PD37wgyX2ffnll2e//fbL5ptvnnbt2mXkyJF58sknK+aZBwAAAABYFqpKS/NLlfAF9PLLL6d9+/a58847s9VWW9XrmJkzZ6a2tjYzZswo/xgtS6fjsTct8zannjlgmbcJAAAAAItT34zwCzESHZbGXXfdlVmzZqVHjx559dVXc/TRR6djx47ZfPPNl3dpAAAAAMBXzFf2h0W/rIYOHZoWLVosdhk6dOjyLu8LYf78+Tn++OOzzjrrZOedd85KK62UCRMmpHHjxsu7NAAAAADgK8Z0Ll8wr7/+embOnLnYba1atUrbtm0/54q+mkzn8umZzgUAAACALzPTuXxJtW3bVlAOAAAAAPAFYToXAAAAAAAoIEQHAAAAAIACQnQAAAAAACggRAcAAAAAgAJCdAAAAAAAKCBEBwAAAACAAkJ0AAAAAAAoIEQHAAAAAIACjZZ3AcCX09QzByzvEgAAAADgM2ckOgAAAAAAFBCiAwAAAABAASE6AAAAAAAUEKIDAAAAAEABIToAAAAAABQQogMAAAAAQAEhOgAAAAAAFBCiAwAAAABAASE6AAAAAAAUEKIDAAAAAEABIToAAAAAABQQogMAAAAAQAEhOgAAAAAAFBCiAwAAAABAASE6AAAAAAAUEKIDAAAAAEABIToAAAAAABQQogMAAAAAQAEhOgAAAAAAFBCiAwAAAABAASE6AAAAAAAUEKIDAAAAAEABIToAAAAAABQQogMAAAAAQAEhOgAAAAAAFBCiAwAAAABAASE6AAAAAAAUEKIDAAAAAEABIToAAAAAABQQogMAAAAAQAEhOgAAAAAAFBCiAwAAAABAASE6AAAAAAAUEKIDAAAAAEABIToAAAAAABQQogMAAAAAQAEhOgAAAAAAFBCiAwAAAABAASE6AAAAAAAUEKIDAAAAAEABIToAAAAAABQQogMAAAAAQAEhOgAAAAAAFBCiAwAAAABAASE6AAAAAAAUEKIDAAAAAEABIToAAAAAABQQogMAAAAAQAEhOgAAAAAAFBCiAwAAAABAASE6AAAAAAAUEKIDAAAAAEABIToAAAAAABQQogMAAAAAQAEhOgAAAAAAFBCiAwAAAABAASE6AAAAAAAUEKLzldCvX78cdthhy7sMAAAAAOArRoi+HPz9739Pw4YNM2DAgM+97y9S2PxFqgUAAAAAYHGE6MvBJZdckp/85Cf561//mldeeWV5lwMAAAAAQAEh+uds1qxZufrqq3PQQQdlwIABufTSSyu2//nPf06XLl3StGnTbLHFFrnssstSVVWV6dOnl/e59957s9lmm6VZs2Zp3759hg0blnffffcT1dOxY8ecfvrp2XvvvdOiRYt06NAhf/7zn/PGG2/ke9/7Xlq0aJGePXvmoYceKh9z6aWXpnXr1vnTn/5UrnXbbbfNv//97/I+Q4YMyU477VTR12GHHZZ+/fqVt99zzz0555xzUlVVlaqqqkydOjVJ8s9//jPbbbddWrRokZVXXjl77bVX3nzzzXI77777brneVVZZJWPGjFniec6dOzczZ86sWAAAAAAAlkSI/jm75ppr0rVr16y99trZc88989vf/jalUilJ8sILL+T73/9+dtpppzz22GM58MADc8IJJ1Qc/9xzz6V///7Zdddd8/jjj+fqq6/Ovffem0MOOeQT13T22Wdnk002yaOPPpoBAwZkr732yt57750999wzjzzySDp16pS99967XGeSzJ49OyNGjMjll1+eiRMnZvr06dl9993r3ec555yTPn365IADDsirr76aV199Ne3bt8/06dOz5ZZbplevXnnooYdy66235r///W8GDRpUPvaoo47KPffck//93//N7bffngkTJuSRRx752P5GjhyZ2tra8tK+ffulv1AAAAAAwNeOEP1zdskll2TPPfdMkvTv3z8zZszIPffckyQZO3Zs1l577Zx11llZe+21s/vuu2fIkCEVx48cOTI/+tGPcthhh6VLly7ZeOON86tf/SqXX3555syZ84lq2n777XPggQemS5cuOemkkzJz5sx861vfyg9+8IOstdZaOeaYY/LUU0/lv//9b/mY+fPn57zzzkufPn2y4YYb5rLLLst9992XBx54oF591tbWpkmTJmnevHnatWuXdu3apWHDhjnvvPPSq1evnHHGGenatWt69eqV3/72t7n77rvzzDPPZNasWbnkkksyevTobLXVVunRo0cuu+yyvP/++x/b33HHHZcZM2aUlw+PmgcAAAAAKNJoeRfwdfL000/ngQceyB//+MckSaNGjbLbbrvlkksuSb9+/fL000/nW9/6VsUx3/72tyteP/bYY3n88cczfvz48rpSqZS6urq88MIL6dat21LX1bNnz/LfK6+8cpKkR48ei6x7/fXX065du3LtH661a9euad26dZ566qlFal4ajz32WO6+++60aNFikW3PPfdc3nvvvcybNy8bbbRRef0KK6yQtdde+2Pbra6uTnV19SeuCwAAAAD4ehKif44uueSSvP/++1l11VXL60qlUqqrq3PeeefVq41Zs2blwAMPzLBhwxbZtvrqq3+iuho3blz+u6qqqnBdXV1dvdts0KBBxfQvyQej15dk1qxZ2WGHHTJq1KhFtq2yyiqZMmVKvWsAAAAAAPi0hOifk/fffz+XX355xowZk2222aZi20477ZSrrroqa6+9dm6++eaKbQ8++GDF6w022CCTJ09O586dP/OaP87777+fhx56qDzq/Omnn8706dPLI+FXWmml/POf/6w4ZtKkSRXhfJMmTbJgwYKKfTbYYINcd9116dixYxo1WvTx7NSpUxo3bpx//OMf5S8Npk2blmeeeSZ9+/ZdpucIAAAAAGBO9M/JjTfemGnTpmW//fbLuuuuW7HsuuuuueSSS3LggQfmX//6V4455pg888wzueaaa3LppZcm+b/R4Mccc0zuu+++HHLIIZk0aVKeffbZ/O///u+n+mHRT6Jx48b5yU9+kn/84x95+OGHM2TIkHznO98ph+pbbrllHnrooVx++eV59tlnc/LJJy8Sqnfs2DH/+Mc/MnXq1Lz55pupq6vLj3/847z99tv54Q9/mAcffDDPPfdcbrvttuyzzz5ZsGBBWrRokf322y9HHXVU7rrrrvzzn//MkCFD0qCBRxkAAAAAWPYkj5+TSy65JN/97ndTW1u7yLZdd901Dz30UN5555384Q9/yPXXX5+ePXvmggsuyAknnJAk5fm8e/bsmXvuuSfPPPNMNttss/Tq1SsnnXRSxRQxn4fmzZvnmGOOyR577JFNNtkkLVq0yNVXX13evu222+bEE0/M0UcfnW9961t55513svfee1e0ceSRR6Zhw4bp3r17Vlpppbz00ktZddVVM3HixCxYsCDbbLNNevTokcMOOyytW7cuB+VnnXVWNttss+ywww757ne/m0033TQbbrjh53r+AAAAAMDXQ1XpoxNX84UyYsSIXHjhhfn3v/+9vEspu/TSS3PYYYdl+vTpy7uUT2zmzJmpra3NjBkz0qpVq+VdDgAAAADwOatvRmhO9C+Y888/P9/61rfyjW98IxMnTsxZZ531uU/VAgAAAADAB4ToXzDPPvtsTj/99Lz99ttZffXVc8QRR+S4446r17F/+9vfst122xVunzVr1rIqEwAAAADga8F0Ll8h7733Xv7zn/8Ubu/cufPnWM0Xm+lcAAAAAODrzXQuX0PNmjUTlAMAAAAALEMNlncBAAAAAADwRSVEBwAAAACAAkJ0AAAAAAAoIEQHAAAAAIACQnQAAAAAACggRAcAAAAAgAJCdAAAAAAAKCBEBwAAAACAAkJ0AAAAAAAoIEQHAAAAAIACQnQAAAAAACggRAcAAAAAgAJCdAAAAAAAKCBEBwAAAACAAkJ0AAAAAAAoIEQHAAAAAIACQnQAAAAAACggRAcAAAAAgAJCdAAAAAAAKCBEBwAAAACAAkJ0AAAAAAAoIEQHAAAAAIACQnQAAAAAACggRAcAAAAAgAJCdAAAAAAAKCBEBwAAAACAAkJ0AAAAAAAoIEQHAAAAAIACQnQAAAAAACggRAcAAAAAgAJCdAAAAAAAKCBEBwAAAACAAkJ0AAAAAAAoIEQHAAAAAIACQnQAAAAAACggRAcAAAAAgAJCdAAAAAAAKCBEBwAAAACAAkJ0AAAAAAAoIEQHAAAAAIACQnQAAAAAACggRAcAAAAAgAJCdAAAAAAAKCBEBwAAAACAAkJ0AAAAAAAoIEQHAAAAAIACQnQAAAAAACggRAcAAAAAgAJCdAAAAAAAKCBEBwAAAACAAkJ0AAAAAAAoIEQHAAAAAIACQnQAAAAAACggRAcAAAAAgAJCdAAAAAAAKCBEBwAAAACAAkJ0AAAAAAAoIEQHAAAAAIACQnQAAAAAACggRAcAAAAAgAJCdAAAAAAAKCBEBwAAAACAAkJ0AAAAAAAoIEQHAAAAAIACQnQAAAAAACggRAcAAAAAgAJC9KUwfPjwrL/++p9J2xMmTEhVVVWmT5++zNqcOnVqqqqqMmnSpGXW5kLLqt7PskYAAAAAgE/rKxuiDxkyJFVVVYss/fv3X96lfSEtDLMbNmyY//znPxXbXn311TRq1ChVVVWZOnVqkmTjjTfOq6++mtra2k/Vb/v27fPqq69m3XXX/VTtAAAAAAB8Fr6yIXqS9O/fP6+++mrFctVVVy3vshYxf/785V1C2WqrrZbLL7+8Yt1ll12W1VZbrWJdkyZN0q5du1RVVX2q/ho2bJh27dqlUaNGn6odAAAAAIDPwlc6RK+urk67du0qljZt2iRJqqqqMnbs2AwcODDNmzdPt27d8ve//z1TpkxJv379UlNTk4033jjPPffcIu2OHTs27du3T/PmzTNo0KDMmDGjvO3BBx/M1ltvnRVXXDG1tbXp27dvHnnkkYrjq6qqcsEFF2THHXdMTU1NRowYsUgfs2fPznbbbZdNNtmkPGXKxRdfnG7duqVp06bp2rVrzj///IpjHnjggfTq1StNmzZN79698+ijjy71NRs8eHDGjRtXsW7cuHEZPHhwxbqPTufy4osvZocddkibNm1SU1OTddZZJzfffHOSZNq0afnRj36UlVZaKc2aNUuXLl3KfXx0OpeF7f7lL39J796907x582y88cZ5+umnK/o//fTT07Zt27Rs2TL7779/jj322I+damfu3LmZOXNmxQIAAAAAsCRf6RB9SU477bTsvffemTRpUrp27Zo99tgjBx54YI477rg89NBDKZVKOeSQQyqOmTJlSq655prccMMNufXWW/Poo4/m4IMPLm9/5513Mnjw4Nx77725//7706VLl2y//fZ55513KtoZPnx4dt555zzxxBPZd999K7ZNnz49W2+9derq6nLHHXekdevWGT9+fE466aSMGDEiTz31VM4444yceOKJueyyy5Iks2bNysCBA9O9e/c8/PDDGT58eI488silviY77rhjpk2blnvvvTdJcu+992batGnZYYcdPva4H//4x5k7d27++te/5oknnsioUaPSokWLJMmJJ56YyZMn55ZbbslTTz2VCy64ICuuuOLHtnfCCSdkzJgxeeihh9KoUaOKazR+/PiMGDEio0aNysMPP5zVV189F1xwwce2N3LkyNTW1paX9u3b1+dyAAAAAABfc1/pOTRuvPHGcpC70PHHH5/jjz8+SbLPPvtk0KBBSZJjjjkmffr0yYknnphtt902SXLooYdmn332qTh+zpw5ufzyy8vTm5x77rkZMGBAxowZk3bt2mXLLbes2P+iiy5K69atc88992TgwIHl9XvssUdF288//3yS5LXXXstuu+2WLl265Morr0yTJk2SJCeffHLGjBmTXXbZJUmyxhprZPLkyRk7dmwGDx6cK6+8MnV1dbnkkkvStGnTrLPOOnn55Zdz0EEHLdU1a9y4cfbcc8/89re/zaabbprf/va32XPPPdO4ceOPPe6ll17Krrvumh49eiRJ1lxzzYptvXr1Su/evZMkHTt2XGIdI0aMSN++fZMkxx57bAYMGJA5c+akadOmOffcc7PffvuVr99JJ52U22+/PbNmzSps77jjjsvhhx9efj1z5kxBOgAAAACwRF/pkehbbLFFJk2aVLEMHTq0vL1nz57lv1deeeUkKYfAC9fNmTOnYuqP1VdfvWJ+8D59+qSurq483ch///vfHHDAAenSpUtqa2vTqlWrzJo1Ky+99FJFbQsD5Y/aeuut07lz51x99dXlAP3dd9/Nc889l/322y8tWrQoL6effnp5upmnnnoqPXv2TNOmTStq+yT23XffXHvttXnttddy7bXXLjJSfnGGDRuW008/PZtssklOPvnkPP744+VtBx10UH7/+99n/fXXz9FHH5377rtvie19+N6sssoqSZLXX389SfL000/n29/+dsX+H339UdXV1WnVqlXFAgAAAACwJF/pEL2mpiadO3euWFZYYYXy9g+Prl74A5mLW1dXV1fvPgcPHpxJkyblnHPOyX333ZdJkyblG9/4RubNm7dIbYszYMCA/PWvf83kyZPL6xaOsP7Nb35T8YXAP//5z9x///31rq2+evToka5du+aHP/xhunXrlnXXXXeJx+y///55/vnns9dee+WJJ55I7969c+655yZJtttuu7z44ov56U9/mldeeSVbbbXVEqea+bT3AQAAAABgWfhKh+ifhZdeeimvvPJK+fX999+fBg0aZO21106STJw4McOGDcv222+fddZZJ9XV1XnzzTfr3f6ZZ56ZwYMHZ6uttioH6SuvvHJWXXXVPP/884t8KbDGGmskSbp165bHH388c+bMqajtk9p3330zYcKEeo1CX6h9+/YZOnRorr/++hxxxBH5zW9+U9620korZfDgwfnd736XX/7yl7nooos+cW1rr712HnzwwYp1H30NAAAAALAsfKXnRJ87d25ee+21inWNGjVa4o9afpymTZtm8ODBGT16dGbOnJlhw4Zl0KBBadeuXZKkS5cuueKKK9K7d+/MnDkzRx11VJo1a7ZUfYwePToLFizIlltumQkTJqRr16455ZRTMmzYsNTW1qZ///6ZO3duHnrooUybNi2HH3549thjj5xwwgk54IADctxxx2Xq1KkZPXr0Jz7PAw44ID/4wQ/SunXreu1/2GGHZbvttstaa62VadOm5e677063bt2SfDBn+YYbbph11lknc+fOzY033lje9kn85Cc/yQEHHJDevXtn4403ztVXX53HH3+8Yh52AAAAAIBl4Ssdot96663l+bQXWnvttfOvf/3rE7fZuXPn7LLLLtl+++3z9ttvZ+DAgTn//PPL2y+55JL8z//8TzbYYIO0b98+Z5xxxhKnLlmcs88+uyJI33///dO8efOcddZZOeqoo1JTU5MePXrksMMOS5K0aNEiN9xwQ4YOHZpevXqle/fuGTVqVHbddddPdJ5L+2XDggUL8uMf/zgvv/xyWrVqlf79++fss89OkjRp0qQc7Ddr1iybbbZZfv/733+iupLkRz/6UZ5//vkceeSRmTNnTgYNGpQhQ4bkgQce+MRtAgAAAAAsTlWpVCot7yLg09p6663Trl27XHHFFfXaf+bMmamtrc2MGTP8yCgAAAAAfA3VNyP8So9E56tp9uzZufDCC7PtttumYcOGueqqq3LnnXfmjjvuWN6lAQAAAABfMX5Y9Gti6NChadGixWKXoUOHLu/ylkpVVVVuvvnmbL755tlwww1zww035Lrrrst3v/vd5V0aAAAAAPAVYzqXr4nXX389M2fOXOy2Vq1apW3btp9zRcuX6VwAAAAA4OvNdC5UaNu27dcuKAcAAAAA+LRM5wIAAAAAAAWE6AAAAAAAUECIDgAAAAAABYToAAAAAABQQIgOAAAAAAAFhOgAAAAAAFBAiA4AAAAAAAWE6AAAAAAAUECIDgAAAAAABYToAAAAAABQQIgOAAAAAAAFhOgAAAAAAFBAiA4AAAAAAAWE6AAAAAAAUECIDgAAAAAABYToAAAAAABQQIgOAAAAAAAFhOgAAAAAAFBAiA4AAAAAAAWE6AAAAAAAUECIDgAAAAAABYToAAAAAABQQIgOAAAAAAAFhOgAAAAAAFBAiA4AAAAAAAWE6AAAAAAAUECIDgAAAAAABYToAAAAAABQQIgOAAAAAAAFhOgAAAAAAFBAiA4AAAAAAAWE6AAAAAAAUECIDgAAAAAABYToAAAAAABQQIgOAAAAAAAFhOgAAAAAAFBAiA4AAAAAAAWE6AAAAAAAUECIDgAAAAAABYToAAAAAABQQIgOAAAAAAAFhOgAAAAAAFBAiA4AAAAAAAWE6AAAAAAAUECIDgAAAAAABYToAAAAAABQQIgOAAAAAAAFhOgAAAAAAFBAiA4AAAAAAAWE6AAAAAAAUKDR8i4AYKGOx960vEsAFmPqmQOWdwkAAACw3BiJDgAAAAAABYToAAAAAABQQIgOAAAAAAAFhOgAAAAAAFBAiA4AAAAAAAWE6AAAAAAAUECIDgAAAAAABYToAAAAAABQQIgOAAAAAAAFhOgAAAAAAFBAiA4AAAAAAAWE6AAAAAAAUECIDgAAAAAABYToAAAAAABQQIgOAAAAAAAFhOgAAAAAAFBAiM7nYsiQIamqqlpkmTJlyvIuDQAAAACgkBCdz03//v3z6quvVixrrLHGUrWxYMGC1NXVfUYVAvBhdfPm5MVRA1NVVZV33313eZcDAAAAy4UQnc9NdXV12rVrV7Gcc8456dGjR2pqatK+ffscfPDBmTVrVvmYSy+9NK1bt86f//zndO/ePdXV1XnppZcyd+7cHHnkkVlttdVSU1OTjTbaKBMmTFh+JwcAAAAAfCUJ0VmuGjRokF/96ld58sknc9lll+Wuu+7K0UcfXbHP7NmzM2rUqFx88cV58skn07Zt2xxyyCH5+9//nt///vd5/PHH84Mf/CD9+/fPs88+u9h+5s6dm5kzZ1YsAAAAAABL0mh5F8DXx4033pgWLVqUX2+33Xa59tpry687duyY008/PUOHDs35559fXj9//vycf/75WW+99ZIkL730UsaNG5eXXnopq666apLkyCOPzK233ppx48bljDPOWKTvkSNH5pRTTvmsTg0AAAAA+IoSovO52WKLLXLBBReUX9fU1OTOO+/MyJEj869//SszZ87M+++/nzlz5mT27Nlp3rx5kqRJkybp2bNn+bgnnngiCxYsyFprrVXR/ty5c/ONb3xjsX0fd9xxOfzww8uvZ86cmfbt2y/L0wMAAAAAvoKE6Hxuampq0rlz5/LrqVOnZuDAgTnooIMyYsSIrLDCCrn33nuz3377Zd68eeUQvVmzZqmqqiofN2vWrDRs2DAPP/xwGjZsWNHHh0e6f1h1dXWqq6s/g7MC+Opq0KRpOhxzY6aeOWB5lwIAAADLjRCd5ebhhx9OXV1dxowZkwYNPpie/5prrlnicb169cqCBQvy+uuvZ7PNNvusywQAAAAAvsb8sCjLTefOnTN//vyce+65ef7553PFFVfkwgsvXOJxa621Vn70ox9l7733zvXXX58XXnghDzzwQEaOHJmbbrrpc6gcAAAAAPi6EKKz3Ky33nr5xS9+kVGjRmXdddfN+PHjM3LkyHodO27cuOy999454ogjsvbaa2ennXbKgw8+mNVXX/0zrhoAAAAA+DqpKpVKpeVdBHzeZs6cmdra2syYMSOtWrVa3uXw/3U81v8kgC8ic6IDAADwVVTfjNBIdAAAAAAAKCBEBwAAAACAAkJ0AAAAAAAoIEQHAAAAAIACQnQAAAAAACggRAcAAAAAgAJCdAAAAAAAKCBEBwAAAACAAkJ0AAAAAAAoIEQHAAAAAIACQnQAAAAAACggRAcAAAAAgAJCdAAAAAAAKNBoeRcAsNDUMwcs7xIAAAAAoIKR6AAAAAAAUECIDgAAAAAABYToAAAAAABQQIgOAAAAAAAFhOgAAAAAAFBAiA4AAAAAAAWE6AAAAAAAUECIDgAAAAAABYToAAAAAABQQIgOAAAAAAAFhOgAAAAAAFBAiA4AAAAAAAWE6AAAAAAAUECIDgAAAAAABYToAAAAAABQQIgOAAAAAAAFhOgAAAAAAFBAiA4AAAAAAAWE6AAAAAAAUECIDgAAAAAABYToAAAAAABQQIgOAAAAAAAFhOgAAAAAAFBAiA4AAAAAAAWE6AAAAAAAUECIDgAAAAAABYToAAAAAABQQIgOAAAAAAAFhOgAAAAAAFBAiA4AAAAAAAWE6AAAAAAAUECIDgAAAAAABYToAAAAAABQQIgOAAAAAAAFhOgAAAAAAFBAiA4AAAAAAAWE6AAAAAAAUECIDgAAAAAABYToAAAAAABQQIgOAAAAAAAFGi3vAgAW6njsTcu7hC+sqWcOWN4lAAAAAHwtGYkOAAAAAAAFhOgAAAAAAFBAiA4AAAAAAAWE6AAAAAAAUECIDgAAAAAABYToAAAAAABQQIgOAAAAAAAFhOgAAAAAAFBAiA4AAAAAAAWE6AAAAAAAUECIDgAAAAAABYToAAAAAABQQIgOAAAAAAAFhOgAAAAAAFBAiA4AAAAAAAWE6HzpTJgwIVVVVZk+ffryLgUAAAAA+IoTovOpDRkyJFVVVamqqkqTJk3SuXPnnHrqqXn//feXd2kAAAAAAJ9Ko+VdAF8N/fv3z7hx4zJ37tzcfPPN+fGPf5zGjRvnuOOOW6p2FixYkKqqqjRo4PsdAAAAAGD5k1SyTFRXV6ddu3bp0KFDDjrooHz3u9/Nn//85/ziF79Ijx49UlNTk/bt2+fggw/OrFmzysddeumlad26df785z+ne/fuqa6uzksvvZS5c+fmmGOOSfv27VNdXZ3OnTvnkksuqejz4YcfTu/evdO8efNsvPHGefrppz/v0wYAAAAAvuKE6HwmmjVrlnnz5qVBgwb51a9+lSeffDKXXXZZ7rrrrhx99NEV+86ePTujRo3KxRdfnCeffDJt27bN3nvvnauuuiq/+tWv8tRTT2Xs2LFp0aJFxXEnnHBCxowZk4ceeiiNGjXKvvvuW1jP3LlzM3PmzIoFvgzq5s3Ji6MGpqqqKu++++7yLgcAAADga8d0LixTpVIpf/nLX3LbbbflJz/5SQ477LDyto4dO+b000/P0KFDc/7555fXz58/P+eff37WW2+9JMkzzzyTa665JnfccUe++93vJknWXHPNRfoaMWJE+vbtmyQ59thjM2DAgMyZMydNmzZdZN+RI0fmlFNOWZanCgAAAAB8DRiJzjJx4403pkWLFmnatGm222677Lbbbhk+fHjuvPPObLXVVllttdXSsmXL7LXXXnnrrbcye/bs8rFNmjRJz549y68nTZqUhg0blgPyIh8+ZpVVVkmSvP7664vd97jjjsuMGTPKy7///e9Pc7rwuWnQpGk6HHNjSqVSampqlnc5AAAAAF87QnSWiS222CKTJk3Ks88+m/feey+XXXZZ3njjjQwcODA9e/bMddddl4cffji//vWvkyTz5s0rH9usWbNUVVVVvK6Pxo0bl/9eeHxdXd1i962urk6rVq0qFgAAAACAJRGis0zU1NSkc+fOWX311dOo0QezBD388MOpq6vLmDFj8p3vfCdrrbVWXnnllSW21aNHj9TV1eWee+75rMsGAAAAAPhYQnQ+M507d878+fNz7rnn5vnnn88VV1yRCy+8cInHdezYMYMHD86+++6bP/3pT3nhhRcyYcKEXHPNNZ9D1QAAAAAA/0eIzmdmvfXWyy9+8YuMGjUq6667bsaPH5+RI0fW69gLLrgg3//+93PwwQena9euOeCAA/Luu+9+xhUDAAAAAFSqKpVKpeVdBHzeZs6cmdra2syYMcP86F8gHY+9aXmX8IU19cwBy7sEAAAAgK+U+maERqIDAAAAAEABIToAAAAAABQQogMAAAAAQAEhOgAAAAAAFBCiAwAAAABAASE6AAAAAAAUEKIDAAAAAEABIToAAAAAABQQogMAAAAAQAEhOgAAAAAAFBCiAwAAAABAASE6AAAAAAAUEKIDAAAAAECBRsu7AICFpp45YHmXAAAAAAAVjEQHAAAAAIACQnQAAAAAACggRAcAAAAAgAJCdAAAAAAAKCBEBwAAAACAAkJ0AAAAAAAoIEQHAAAAAIACQnQAAAAAACggRAcA+H/t3Xucl3Wd///nADIgh8FEA5KDxkEkEVFQNEMThcQDqYu1mvBFygOEh3QRFQ+5LVQeUldTWwQ7orZqJG5JBCh4yBRUhFBZFVoPlCkIKAfn8/ujH7M7yYWoMwzI/X67fW4xn+v6XJ/XNfNupEeX1wcAAAAKiOgAAAAAAFBARAcAAAAAgAIiOgAAAAAAFBDRAQAAAACggIgOAAAAAAAFRHQAAAAAACggogMAAAAAQAERHQAAAAAACojoAAAAAABQQEQHAAAAAIACIjoAAAAAABQQ0QEAAAAAoICIDgAAAAAABUR0AAAAAAAoIKIDAAAAAEABER0AAAAAAAqI6AAAAAAAUEBEBwAAAACAAiI6AAAAAAAUENEBAAAAAKCAiA4AAAAAAAVEdAAAAAAAKCCiAwAAAABAAREdAAAAAAAKiOgAAAAAAFBARAcAAAAAgAIiOgAAAAAAFBDRAQAAAACggIgOAAAAAAAFRHQAAAAAACggogMAAAAAQIEGdT0AwAYdLpxa1yNsMS+NH1jXIwAAAACwGVyJDgAAAAAABUR0AAAAAAAoIKIDAAAAAEABER0AAAAAAAqI6AAAAAAAUEBEBwAAAACAAiI6AAAAAAAUENEBAAAAAKCAiA4AAAAAAAVEdAAAAAAAKCCiAwAAAABAAREdAAAAAAAKiOgAAAAAAFBARAcAAAAAgAIiOgAAAAAAFBDRqVVlZWW59957kyQvvfRSysrKMm/evDqdCQAAAABgc4nofCx/+ctfcuaZZ6Zdu3YpLy9Pq1at0r9//8yZMydJ8uqrr+ZLX/rShzrmPffckwMPPDAVFRVp1qxZunXrlnPOOacWpgcAAAAA2LQGdT0A27YTTjgha9euze2335499tgjr7/+eqZPn5433ngjSdKqVasPdbzp06fnpJNOyne+850ce+yxKSsry4IFCzJt2rTaGB8AAAAAYJNcic5H9tZbb+Whhx7Kd7/73Rx22GFp3759evfunTFjxuTYY49NUv12Lhv86U9/ykEHHZRGjRrlc5/7XGbNmlW17de//nUOPvjgXHDBBenSpUs6d+6cQYMG5cYbb6za5/LLL0+PHj1yyy23pG3bttlxxx0zePDgLF++fIucNwAAAACw/RDR+ciaNm2apk2b5t57782aNWs2+3UXXHBBvvWtb2Xu3Lnp06dPjjnmmGpXrj/77LOZP3/+Jo/xwgsv5M4778yvf/3r/OY3v8ncuXNz1llnFe6/Zs2arFixotoDAAAAAOCDiOh8ZA0aNMikSZNy++23p0WLFjn44INz0UUX5emnn97k60aOHJkTTjghXbt2zQ9/+MNUVFRkwoQJSZJvfvOb6dWrV/bee+906NAhX/nKV3Lbbbe9L9K/++67+fGPf5wePXrkC1/4Qm644YZMnjw5r7322kbfc9y4camoqKh6tG3btma+CQAAAADAJ5qIzsdywgkn5JVXXsmUKVMyYMCAzJw5Mz179sykSZMKX9OnT5+qPzdo0CD7779/Fi5cmCRp0qRJpk6dmhdeeCGXXHJJmjZtmm9961vp3bt3Vq9eXfW6du3a5TOf+Uy1Y1ZWVmbRokUbfc8xY8Zk+fLlVY+lS5d+zDMHAAAAALYHIjofW6NGjXLEEUdk7NixefjhhzN06NBcdtllH+uYn/3sZzN8+PD8x3/8R5588sksWLAgd9xxx0c+Xnl5eZo3b17tAQAAAADwQUR0atxee+2VVatWFW5/9NFHq/68fv36PPHEE+natWvh/h06dMiOO+5Y7ZhLlizJK6+8Uu2Y9erVS5cuXT7m9AAAAAAA/6tBXQ/AtuuNN97IP/3TP2XYsGHp3r17mjVrlj/+8Y/53ve+l+OOO67wdTfeeGM6deqUrl275tprr82bb76ZYcOGJUkuv/zyrF69OkcddVTat2+ft956K9dff33WrVuXI444ouoYjRo1ypAhQ3LVVVdlxYoVGTVqVAYPHpxWrVrV+nkDAAAAANsPEZ2PrGnTpjnggANy7bXXZvHixVm3bl3atm2br3/967nooosKXzd+/PiMHz8+8+bNS8eOHTNlypS0bNkySdK3b9/ceOONOfXUU/P6669np512yr777psHHnig2lXmHTt2zPHHH5+jjjoqf/vb33L00UfnpptuqvVzBgAAAAC2L2WlUqlU10PAh3H55Zfn3nvvzbx58z7yMVasWJGKioosX77c/dG3Ih0unFrXI2wxL40fWNcjAAAAAGzXNrcRuic6AAAAAAAUENEBAAAAAKCAiM425/LLL/9Yt3IBAAAAANhcIjoAAAAAABQQ0QEAAAAAoICIDgAAAAAABUR0AAAAAAAoIKIDAAAAAEABER0AAAAAAAqI6AAAAAAAUEBEBwAAAACAAiI6AAAAAAAUENEBAAAAAKCAiA4AAAAAAAUa1PUAABu8NH5gXY8AAAAAANW4Eh0AAAAAAAqI6AAAAAAAUEBEBwAAAACAAiI6AAAAAAAUENEBAAAAAKCAiA4AAAAAAAVEdAAAAAAAKCCiAwAAAABAAREdAAAAAAAKiOgAAAAAAFBARAcAAAAAgAIiOgAAAAAAFBDRAQAAAACggIgOAAAAAAAFRHQAAAAAACggogMAAAAAQAERHQAAAAAACojoAAAAAABQQEQHAAAAAIACIjoAAAAAABQQ0QEAAAAAoICIDgAAAAAABUR0AAAAAAAoIKIDAAAAAEABER0AAAAAAAqI6AAAAAAAUEBEBwAAAACAAiI6AAAAAAAUENEBAAAAAKCAiA4AAAAAAAVEdAAAAAAAKCCiAwAAAABAAREdAAAAAAAKiOgAAAAAAFBARAcAAAAAgAIiOgAAAAAAFBDRAQAAAACggIgOAAAAAAAFRHQAAAAAACggogMAAAAAQIEGdT0AwAYdLpxa1yMAsB16afzAuh4BAADYirkSHQAAAAAACojoAAAAAABQQEQHAAAAAIACIjoAAAAAABQQ0QEAAAAAoICIDgAAAAAABUR0AAAAAAAoIKIDAAAAAEABER0AAAAAAAqI6AAAAAAAUEBEBwAAAACAAiI6AAAAAAAUENEBAAAAAKCAiA4AAAAAAAVEdAAAAAAAKCCiAwAAAABAARGdOvHII4+kfv36GThwYF2PAgAAAABQSESnTkyYMCHf/OY38+CDD+aVV16p63EAAAAAADZKRGeLW7lyZe64446ceeaZGThwYCZNmlRt+5QpU9KpU6c0atQohx12WG6//faUlZXlrbfeqtpn9uzZOeSQQ9K4ceO0bds2o0aNyqpVqwrfc82aNVmxYkW1BwCwfatc+25e/u7RKSsr2+TfIwAAgO2biM4Wd+edd2bPPfdMly5dcsopp+S2225LqVRKkrz44os58cQTM2jQoDz11FM5/fTTc/HFF1d7/eLFizNgwICccMIJefrpp3PHHXdk9uzZGTlyZOF7jhs3LhUVFVWPtm3b1uo5AgAAAACfDGWlDfUStpCDDz44gwcPztlnn53169endevWueuuu3LooYfmwgsvzNSpU/PMM89U7X/JJZfkO9/5Tt588820aNEiw4cPT/369XPLLbdU7TN79uz07ds3q1atSqNGjd73nmvWrMmaNWuqvl6xYkXatm2b5cuXp3nz5rV7wmy2DhdOresRANiOVK59N0uvPTHJ3/9NuSZNmtTxRAAAwJa0YsWKVFRUfGAjdCU6W9SiRYvyhz/8IV/96leTJA0aNMhJJ52UCRMmVG3v1atXtdf07t272tdPPfVUJk2alKZNm1Y9+vfvn8rKyrz44osbfd/y8vI0b9682gMA2L7Va9go7Uffl1KpJKADAACFGtT1AGxfJkyYkPXr16dNmzZVz5VKpZSXl+ff//3fN+sYK1euzOmnn55Ro0a9b1u7du1qbFYAAAAAABGdLWb9+vX58Y9/nKuvvjpHHnlktW2DBg3KL37xi3Tp0iX3339/tW2PP/54ta979uyZBQsWpGPHjrU+MwAAAACwfRPR2WLuu+++vPnmmznttNNSUVFRbdsJJ5yQCRMm5M4778w111yT0aNH57TTTsu8efMyadKkJElZWVmSZPTo0TnwwAMzcuTIDB8+PE2aNMmCBQsybdq0zb6aHQAAAABgc7gnOlvMhAkT0q9fv/cF9OTvEf2Pf/xj3n777fzyl7/M3Xffne7du+eHP/xhLr744iR/v695knTv3j2zZs3Kc889l0MOOST77rtvLr300mq3iAEAAAAAqAllpVKpVNdDwKZ85zvfyc0335ylS5fW2DE395N32bI6XDi1rkcAYDv00viBdT0CAABQBza3EbqdC1udm266Kb169crOO++cOXPm5Pvf/35GjhxZ12MBAAAAANshEZ2tzvPPP59//dd/zd/+9re0a9cu3/rWtzJmzJi6HgsAAAAA2A6J6Gx1rr322lx77bV1PQYAAAAAgA8WBQAAAACAIiI6AAAAAAAUENEBAAAAAKCAiA4AAAAAAAVEdAAAAAAAKCCiAwAAAABAAREdAAAAAAAKiOgAAAAAAFBARAcAAAAAgAIiOgAAAAAAFGhQ1wMAbPDS+IF1PQIAAAAAVONKdAAAAAAAKCCiAwAAAABAAREdAAAAAAAKiOgAAAAAAFBARAcAAAAAgAIiOgAAAAAAFBDRAQAAAACggIgOAAAAAAAFRHQAAAAAACggogMAAAAAQAERHQAAAAAACojoAAAAAABQQEQHAAAAAIACIjoAAAAAABQQ0QEAAAAAoICIDgAAAAAABUR0AAAAAAAoIKIDAAAAAEABER0AAAAAAAqI6AAAAAAAUEBEBwAAAACAAiI6AAAAAAAUENEBAAAAAKCAiA4AAAAAAAVEdAAAAAAAKCCiAwAAAABAAREdAAAAAAAKiOgAAAAAAFBARAcAAAAAgAIiOgAAAAAAFBDRAQAAAACggIgOAAAAAAAFRHQAAAAAACggogMAAAAAQAERHQAAAAAACojoAAAAAABQQEQHAAAAAIACIjoAAAAAABQQ0QEAAAAAoICIDgAAAAAABRrU9QAAG3S4cGpdjwAAAACw3Xtp/MC6HmGr4kp0AAAAAAAoIKIDAAAAAEABER0AAAAAAAqI6AAAAAAAUEBEBwAAAACAAiI6AAAAAAAUENEBAAAAAKCAiA4AAAAAAAVEdAAAAAAAKCCiAwAAAABAAREdAAAAAAAKiOgAAAAAAFBARAcAAAAAgAIiOgAAAAAAFBDRAQAAAACggIjONqGsrCz33ntvXY8BAAAAAGxnRHQ229ChQzNo0KAP3O/yyy9PWVlZBgwY8L5t3//+91NWVpZDDz205gcEAAAAAKhhIjq1onXr1pkxY0b+/Oc/V3v+tttuS7t27epoKgAAAACgSOXad/Pyd49OWVlZVq1aVdfjbDVEdD6SX/7yl9l7773TuHHj7LzzzunXr1+1/2LtuuuuOfLII3P77bdXPffwww/nr3/9awYOHFjtWI8//niOOOKItGzZMhUVFenbt2+efPLJTb7/0qVLM3jw4LRo0SKf+tSnctxxx+Wll14q3H/NmjVZsWJFtQcAAAAAwAcR0fnQXn311Xz1q1/NsGHDsnDhwsycOTPHH398SqVStf2GDRuWSZMmVX1922235eSTT07Dhg2r7ff2229nyJAhmT17dh599NF06tQpRx11VN5+++2Nvv+6devSv3//NGvWLA899FDmzJmTpk2bZsCAAVm7du1GXzNu3LhUVFRUPdq2bfvxvgkAAAAAwHZBROdDe/XVV7N+/focf/zx6dChQ/bee++cddZZadq0abX9jj766KxYsSIPPvhgVq1alTvvvDPDhg173/G++MUv5pRTTsmee+6Zrl275tZbb83q1asza9asjb7/HXfckcrKyvzHf/xH9t5773Tt2jUTJ07MkiVLMnPmzI2+ZsyYMVm+fHnVY+nSpR/7+wAAAAAAfPI1qOsB2Pbss88+Ofzww7P33nunf//+OfLII3PiiSdmp512qrbfDjvskFNOOSUTJ07Mf//3f6dz587p3r37+473+uuv55JLLsnMmTOzbNmyvPfee1m9enWWLFmy0fd/6qmn8sILL6RZs2bVnn/33XezePHijb6mvLw85eXlH/GMAQAAAIDtlYjOh1a/fv1MmzYtDz/8cB544IHccMMNufjii/PYY49l9913r7bvsGHDcsABB2T+/PkbvQo9SYYMGZI33ngj1113Xdq3b5/y8vL06dOn8NYsK1euzH777Zef/exn79u2yy67fPwTBAAAAIDtUL2GjdJ+9H15afzAD955O+J2LnwkZWVlOfjgg3PFFVdk7ty5adiwYe6555737detW7d069Yt8+fPzz//8z9v9Fhz5szJqFGjctRRR6Vbt24pLy/PX//618L37tmzZ55//vnsuuuu6dixY7VHRUVFjZ0jAAAAAICIzof22GOP5d/+7d/yxz/+MUuWLMndd9+dv/zlL+natetG9//973+fV199NS1atNjo9k6dOuUnP/lJFi5cmMceeywnn3xyGjduXPj+J598clq2bJnjjjsuDz30UF588cXMnDkzo0aNyp///OeaOEUAAAAAgCQiOh9B8+bN8+CDD+aoo45K586dc8kll+Tqq6/Ol770pY3u36RJk8KAniQTJkzIm2++mZ49e+ZrX/taRo0alV133bVw/x133DEPPvhg2rVrl+OPPz5du3bNaaedlnfffTfNmzf/uKcHAAAAAFClrFQqlep6CNjSVqxYkYqKiixfvlx434p0uHBqXY8AAAAAsN3bXu6JvrmN0JXoAAAAAABQQEQHAAAAAIACIjoAAAAAABQQ0QEAAAAAoICIDgAAAAAABUR0AAAAAAAoIKIDAAAAAEABER0AAAAAAAqI6AAAAAAAUEBEBwAAAACAAiI6AAAAAAAUENEBAAAAAKCAiA4AAAAAAAUa1PUAABu8NH5gXY8AAAAAANW4Eh0AAAAAAAqI6AAAAAAAUEBEBwAAAACAAiI6AAAAAAAUENEBAAAAAKCAiA4AAAAAAAVEdAAAAAAAKCCiAwAAAABAAREdAAAAAAAKiOgAAAAAAFBARAcAAAAAgAIiOgAAAAAAFBDRAQAAAACggIgOAAAAAAAFRHQAAAAAACggogMAAAAAQAERHQAAAAAACojoAAAAAABQQEQHAAAAAIACIjoAAAAAABQQ0QEAAAAAoICIDgAAAAAABUR0AAAAAAAoIKIDAAAAAEABER0AAAAAAAqI6AAAAAAAUEBEBwAAAACAAiI6AAAAAAAUENEBAAAAAKCAiA4AAAAAAAVEdAAAAAAAKCCiAwAAAABAAREdAAAAAAAKiOgAAAAAAFBARAcAAAAAgAIiOgAAAAAAFBDRAQAAAACggIgOAAAAAAAFRHQAAAAAACggogMAAAAAQIEGdT0AwAYdLpxa1yMAAABsN14aP7CuRwDYJrgSHQAAAAAACojoAAAAAABQQEQHAAAAAIACIjoAAAAAABQQ0QEAAAAAoICIDgAAAAAABUR0AAAAAAAoIKIDAAAAAEABER0AAAAAAAqI6AAAAAAAUEBEBwAAAACAAiI6AAAAAAAUENEBAAAAAKCAiA4AAAAAAAVEdAAAAAAAKCCis8UceuihOeecc+p6DAAAAACAzSai86EMHTo0ZWVlKSsrS8OGDdOxY8d8+9vfzvr16+t6NAAAAACAGtegrgdg2zNgwIBMnDgxa9asyf33358RI0Zkhx12yJgxY+p6NAAAAACAGuVKdD608vLytGrVKu3bt8+ZZ56Zfv36ZcqUKUmSOXPm5NBDD82OO+6YnXbaKf3798+bb7650eP85Cc/yf77759mzZqlVatW+ed//ucsW7asavubb76Zk08+ObvssksaN26cTp06ZeLEiUmStWvXZuTIkWndunUaNWqU9u3bZ9y4cbV/8gAAALCNq1z7bl7+7tEpKyvLqlWr6nocgK2eK9H52Bo3bpw33ngj8+bNy+GHH55hw4bluuuuS4MGDTJjxoy89957G33dunXrcuWVV6ZLly5ZtmxZzjvvvAwdOjT3339/kmTs2LFZsGBB/uu//istW7bMCy+8kHfeeSdJcv3112fKlCm58847065duyxdujRLly4tnHHNmjVZs2ZN1dcrVqyowe8AAAAAAPBJJaLzkZVKpUyfPj2//e1v881vfjPf+973sv/+++emm26q2qdbt26Frx82bFjVn/fYY49cf/316dWrV1auXJmmTZtmyZIl2XfffbP//vsnSTp06FC1/5IlS9KpU6d8/vOfT1lZWdq3b7/JWceNG5crrrjiI54pAAAAALC9cjsXPrT77rsvTZs2TaNGjfKlL30pJ510Ui6//PKqK9E31xNPPJFjjjkm7dq1S7NmzdK3b98kfw/kSXLmmWdm8uTJ6dGjR/7lX/4lDz/8cNVrhw4dmnnz5qVLly4ZNWpUHnjggU2+15gxY7J8+fKqx6auWgcAAIBPsnoNG6X96PtSKpXSpEmTuh4HYKsnovOhHXbYYZk3b16ef/75vPPOO7n99tvTpEmTNG7ceLOPsWrVqvTv3z/NmzfPz372szz++OO55557kvz9fudJ8qUvfSkvv/xyzj333Lzyyis5/PDDc/755ydJevbsmRdffDFXXnll3nnnnQwePDgnnnhi4fuVl5enefPm1R4AAAAAAB9EROdDa9KkSTp27Jh27dqlQYP/vSNQ9+7dM3369M06xp/+9Ke88cYbGT9+fA455JDsueee1T5UdINddtklQ4YMyU9/+tP84Ac/yK233lq1rXnz5jnppJPyox/9KHfccUf+8z//M3/7298+/gkCAAAAAPz/3BOdGjNmzJjsvffeOeuss3LGGWekYcOGmTFjRv7pn/4pLVu2rLZvu3bt0rBhw9xwww0544wzMn/+/Fx55ZXV9rn00kuz3377pVu3blmzZk3uu+++dO3aNUlyzTXXpHXr1tl3331Tr1693HXXXWnVqlVatGixpU4XAAAAANgOuBKdGtO5c+c88MADeeqpp9K7d+/06dMnv/rVr6pdrb7BLrvskkmTJuWuu+7KXnvtlfHjx+eqq66qtk/Dhg0zZsyYdO/ePV/4whdSv379TJ48OUnSrFmzqg8y7dWrV1566aXcf//9qVfPkgYAAAAAak5ZqVQq1fUQsKWtWLEiFRUVWb58ufujb0U6XDi1rkcAAADYbrw0fmBdjwBQpza3EbpsFwAAAAAACojoAAAAAABQQEQHAAAAAIACIjoAAAAAABQQ0QEAAAAAoICIDgAAAAAABUR0AAAAAAAoIKIDAAAAAEABER0AAAAAAAqI6AAAAAAAUEBEBwAAAACAAiI6AAAAAAAUENEBAAAAAKBAg7oeAGCDl8YPrOsRAAAAAKAaV6IDAAAAAEABER0AAAAAAAqI6AAAAAAAUEBEBwAAAACAAiI6AAAAAAAUENEBAAAAAKCAiA4AAAAAAAVEdAAAAAAAKCCiAwAAAABAAREdAAAAAAAKiOgAAAAAAFBARAcAAAAAgAIiOgAAAAAAFBDRAQAAAACggIgOAAAAAAAFRHQAAAAAACggogMAAAAAQAERHQAAAAAACojoAAAAAABQoEFdDwB1oVQqJUlWrFhRx5MAAAAAAHVhQxvc0AqLiOhsl95+++0kSdu2bet4EgAAAACgLr399tupqKgo3F5W+qDMDp9AlZWVeeWVV9KsWbOUlZXV9Tjk7//PX9u2bbN06dI0b968rsfhE8K6ojZYV9QWa4vaYF1RG6wraou1RW2wrtiUUqmUt99+O23atEm9esV3PnclOtulevXqZbfddqvrMdiI5s2b+4caNc66ojZYV9QWa4vaYF1RG6wraou1RW2wriiyqSvQN/DBogAAAAAAUEBEBwAAAACAAiI6sFUoLy/PZZddlvLy8roehU8Q64raYF1RW6wtaoN1RW2wrqgt1ha1wbqiJvhgUQAAAAAAKOBKdAAAAAAAKCCiAwAAAABAAREdAAAAAAAKiOgAAAAAAFBARAdqxY033pgOHTqkUaNGOeCAA/KHP/xhk/vfdddd2XPPPdOoUaPsvffeuf/++6ttL5VKufTSS9O6des0btw4/fr1y/PPP1+bp8BWqibX1rp16zJ69OjsvffeadKkSdq0aZNTTz01r7zySm2fBluZmv6d9X+dccYZKSsryw9+8IManpqtXW2sq4ULF+bYY49NRUVFmjRpkl69emXJkiW1dQpspWp6ba1cuTIjR47MbrvtlsaNG2evvfbKzTffXJunwFbow6yrZ599NieccEI6dOiwyX/Gfdi1yidPTa+rcePGpVevXmnWrFl23XXXDBo0KIsWLarFM2BrVRu/szYYP358ysrKcs4559Ts0GzTRHSgxt1xxx0577zzctlll+XJJ5/MPvvsk/79+2fZsmUb3f/hhx/OV7/61Zx22mmZO3duBg0alEGDBmX+/PlV+3zve9/L9ddfn5tvvjmPPfZYmjRpkv79++fdd9/dUqfFVqCm19bq1avz5JNPZuzYsXnyySdz9913Z9GiRTn22GO35GlRx2rjd9YG99xzTx599NG0adOmtk+DrUxtrKvFixfn85//fPbcc8/MnDkzTz/9dMaOHZtGjRptqdNiK1Aba+u8887Lb37zm/z0pz/NwoULc84552TkyJGZMmXKljot6tiHXVerV6/OHnvskfHjx6dVq1Y1ckw+eWpjXc2aNSsjRozIo48+mmnTpmXdunU58sgjs2rVqto8FbYytbG2Nnj88cdzyy23pHv37rUxOtuyEkAN6927d2nEiBFVX7/33nulNm3alMaNG7fR/QcPHlwaOHBgtecOOOCA0umnn14qlUqlysrKUqtWrUrf//73q7a/9dZbpfLy8tIvfvGLWjgDtlY1vbY25g9/+EMpSenll1+umaHZ6tXWuvrzn/9c+sxnPlOaP39+qX379qVrr722xmdn61Ub6+qkk04qnXLKKbUzMNuM2lhb3bp1K33729+utk/Pnj1LF198cQ1Oztbsw66r/6von3Ef55h8MtTGuvpHy5YtKyUpzZo16+OMyjamttbW22+/XerUqVNp2rRppb59+5bOPvvsGpqYTwJXogM1au3atXniiSfSr1+/qufq1auXfv365ZFHHtnoax555JFq+ydJ//79q/Z/8cUX89prr1Xbp6KiIgcccEDhMfnkqY21tTHLly9PWVlZWrRoUSNzs3WrrXVVWVmZr33ta7ngggvSrVu32hmerVZtrKvKyspMnTo1nTt3Tv/+/bPrrrvmgAMOyL333ltr58HWp7Z+Zx100EGZMmVK/ud//ielUikzZszIc889lyOPPLJ2ToStykdZV3VxTLYtW2oNLF++PEnyqU99qsaOydatNtfWiBEjMnDgwPf9cxMSt3MBathf//rXvPfee/n0pz9d7flPf/rTee211zb6mtdee22T+2/4zw9zTD55amNt/aN33303o0ePzle/+tU0b968ZgZnq1Zb6+q73/1uGjRokFGjRtX80Gz1amNdLVu2LCtXrsz48eMzYMCAPPDAA/nyl7+c448/PrNmzaqdE2GrU1u/s2644Ybstdde2W233dKwYcMMGDAgN954Y77whS/U/Emw1fko66oujsm2ZUusgcrKypxzzjk5+OCD87nPfa5GjsnWr7bW1uTJk/Pkk09m3LhxH3dEPqEa1PUAALA1WLduXQYPHpxSqZQf/vCHdT0O27Annngi1113XZ588smUlZXV9Th8QlRWViZJjjvuuJx77rlJkh49euThhx/OzTffnL59+9bleGzjbrjhhjz66KOZMmVK2rdvnwcffDAjRoxImzZtXI0HbLVGjBiR+fPnZ/bs2XU9Ctu4pUuX5uyzz860adN81gyFXIkO1KiWLVumfv36ef3116s9//rrrxd+gEerVq02uf+G//wwx+STpzbW1gYbAvrLL7+cadOmuQp9O1Ib6+qhhx7KsmXL0q5duzRo0CANGjTIyy+/nG9961vp0KFDrZwHW5faWFctW7ZMgwYNstdee1Xbp2vXrlmyZEkNTs/WrDbW1jvvvJOLLroo11xzTY455ph07949I0eOzEknnZSrrrqqdk6ErcpHWVd1cUy2LbW9BkaOHJn77rsvM2bMyG677faxj8e2ozbW1hNPPJFly5alZ8+eVX9/nzVrVq6//vo0aNAg7733Xk2MzjZORAdqVMOGDbPffvtl+vTpVc9VVlZm+vTp6dOnz0Zf06dPn2r7J8m0adOq9t99993TqlWravusWLEijz32WOEx+eSpjbWV/G9Af/755/O73/0uO++8c+2cAFul2lhXX/va1/L0009n3rx5VY82bdrkggsuyG9/+9vaOxm2GrWxrho2bJhevXpl0aJF1fZ57rnn0r59+xo+A7ZWtbG21q1bl3Xr1qVever/07B+/fpV/wYEn2wfZV3VxTHZttTWGiiVShk5cmTuueee/P73v8/uu+9eE+OyDamNtXX44YfnmWeeqfb39/333z8nn3xy5s2bl/r169fU+GzL6viDTYFPoMmTJ5fKy8tLkyZNKi1YsKD0jW98o9SiRYvSa6+9ViqVSqWvfe1rpQsvvLBq/zlz5pQaNGhQuuqqq0oLFy4sXXbZZaUddtih9Mwzz1TtM378+FKLFi1Kv/rVr0pPP/106bjjjivtvvvupXfeeWeLnx91p6bX1tq1a0vHHntsabfddivNmzev9Oqrr1Y91qxZUyfnyJZXG7+z/lH79u1L1157bW2fCluR2lhXd999d2mHHXYo3XrrraXnn3++dMMNN5Tq169feuihh7b4+VF3amNt9e3bt9StW7fSjBkzSv/93/9dmjhxYqlRo0alm266aYufH3Xjw66rNWvWlObOnVuaO3duqXXr1qXzzz+/NHfu3NLzzz+/2cfkk6821tWZZ55ZqqioKM2cObPa391Xr169xc+PulMba+sf9e3bt3T22WfX9qmwDRHRgVpxww03lNq1a1dq2LBhqXfv3qVHH320alvfvn1LQ4YMqbb/nXfeWercuXOpYcOGpW7dupWmTp1abXtlZWVp7NixpU9/+tOl8vLy0uGHH15atGjRljgVtjI1ubZefPHFUpKNPmbMmLGFzoitQU3/zvpHIvr2qTbW1YQJE0odO3YsNWrUqLTPPvuU7r333to+DbZCNb22Xn311dLQoUNLbdq0KTVq1KjUpUuX0tVXX12qrKzcEqfDVuLDrKuiv0P17dt3s4/J9qGm11XR390nTpy45U6KrUJt/M76v0R0/lFZqVQqbaGL3gEAAAAAYJvinugAAAAAAFBARAcAAAAAgAIiOgAAAAAAFBDRAQAAAACggIgOAAAAAAAFRHQAAAAAACggogMAAAAAQAERHQAAAAAACojoAAAAW8jMmTNTVlaWt956q65HAQBgM4noAACwDRg6dGjKyspyxhlnvG/biBEjUlZWlqFDh275wQAA4BNORAcAgG1E27ZtM3ny5LzzzjtVz7377rv5+c9/nnbt2tXhZFu3tWvX1vUIW0ypVMr69evregwAgE8UER0AALYRPXv2TNu2bXP33XdXPXf33XenXbt22XfffavtW1lZmXHjxmX33XdP48aNs88+++SXv/xl1fb33nsvp512WtX2Ll265Lrrrqt2jKFDh2bQoEG56qqr0rp16+y8884ZMWJE1q1bVzjj4sWLc9xxx+XTn/50mjZtml69euV3v/tdtX3WrFmT0aNHp23btikvL0/Hjh0zYcKEqu3PPvtsjj766DRv3jzNmjXLIYccksWLFydJDj300JxzzjnVjjdo0KBqV+F36NAhV155ZU499dQ0b9483/jGN5Iko0ePTufOnbPjjjtmjz32yNixY993Lr/+9a/Tq1evNGrUKC1btsyXv/zlJMm3v/3tfO5zn3vf+fbo0SNjx44t/H7cf//96dy5cxo3bpzDDjssL7300vv2mT17dg455JA0btw4bdu2zahRo7Jq1arN+n5tuD3Mf/3Xf2W//fZLeXl5Zs+eXSM//5kzZ6Z3795p0qRJWrRokYMPPjgvv/xy1fZf/epX6dmzZxo1apQ99tgjV1xxhYAPAHwiiegAALANGTZsWCZOnFj19W233Zb/9//+3/v2GzduXH784x/n5ptvzrPPPptzzz03p5xySmbNmpXk75F9t912y1133ZUFCxbk0ksvzUUXXZQ777yz2nFmzJiRxYsXZ8aMGbn99tszadKkTJo0qXC+lStX5qijjsr06dMzd+7cDBgwIMccc0yWLFlStc+pp56aX/ziF7n++uuzcOHC3HLLLWnatGmS5H/+53/yhS98IeXl5fn973+fJ554IsOGDfvQcfaqq67KPvvsk7lz51ZF7mbNmmXSpElZsGBBrrvuuvzoRz/KtddeW/WaqVOn5stf/nKOOuqozJ07N9OnT0/v3r2rvu8LFy7M448/XrX/3Llz8/TTT2/0+58kS5cuzfHHH59jjjkm8+bNy/Dhw3PhhRdW22fx4sUZMGBATjjhhDz99NO54447Mnv27IwcOXKzvl8bXHjhhRk/fnwWLlyY7t27f+yf//r16zNo0KD07ds3Tz/9dB555JF84xvfSFlZWZLkoYceyqmnnpqzzz47CxYsyC233JJJkyblO9/5zof6OQEAbAvKSqVSqa6HAAAANm3o0KF566238qMf/Sht27bNokWLkiR77rlnli5dmuHDh6dFixaZNGlS1qxZk0996lP53e9+lz59+lQdY/jw4Vm9enV+/vOfb/Q9Ro4cmddee63qiuWhQ4dm5syZWbx4cerXr58kGTx4cOrVq5fJkydv9uyf+9zncsYZZ2TkyJF57rnn0qVLl0ybNi39+vV7374XXXRRJk+enEWLFmWHHXZ43/ZDDz00PXr0yA9+8IOq5wYNGlR17snfr0Tfd999c88992xyrquuuiqTJ0/OH//4xyTJQQcdlD322CM//elPN7r/UUcdlQ4dOuSmm25KkowaNSrPPPNMZsyYsdH9L7roovzqV7/Ks88+W/XchRdemO9+97t5880306JFiwwfPjz169fPLbfcUrXP7Nmz07dv36xatSpLlizZ5Pdr5syZOeyww3LvvffmuOOOS5Ia+fn/7W9/y84775yZM2emb9++79u3X79+OfzwwzNmzJiq537605/mX/7lX/LKK69s9PgAANuqBnU9AAAAsPl22WWXDBw4MJMmTUqpVMrAgQPTsmXLavu88MILWb16dY444ohqz69du7babV9uvPHG3HbbbVmyZEneeeedrF27Nj169Kj2mm7dulUF9CRp3bp1nnnmmcL5Vq5cmcsvvzxTp07Nq6++mvXr1+edd96puhJ93rx5qV+//kbD7IbthxxyyEYD+oex//77v++5O+64I9dff30WL16clStXZv369WnevHm19/76179eeMyvf/3rGTZsWK655prUq1cvP//5z6tdyf6PFi5cmAMOOKDac/83aifJU089laeffjo/+9nPqp4rlUqprKzMiy++mGeeeWaT36+NnW9N/Pw/9alPZejQoenfv3+OOOKI9OvXL4MHD07r1q2r5p4zZ061K8/fe++9vPvuu1m9enV23HHHTc4LALAtEdEBAGAbM2zYsKrbfdx4443v275y5cokf789yWc+85lq28rLy5MkkydPzvnnn5+rr746ffr0SbNmzfL9738/jz32WLX9/zFml5WVpbKysnC2888/P9OmTctVV12Vjh07pnHjxjnxxBOrPtyzcePGmzy3D9per169/OO/TLuxe7Q3adKk2tePPPJITj755FxxxRXp379/KioqMnny5Fx99dWb/d7HHHNMysvLc88996Rhw4ZZt25dTjzxxE2+5oOsXLkyp59+ekaNGvW+be3atcsLL7ywWcf5v+dbUz//iRMnZtSoUfnNb36TO+64I5dcckmmTZuWAw88MCtXrswVV1yR448//n2zNGrUaLNmBgDYVojoAACwjRkwYEDWrl2bsrKy9O/f/33b99prr5SXl2fJkiWFVzDPmTMnBx10UM4666yq5zZ8eOfHMWfOnAwdOrTqAzlXrlxZ7cM0995771RWVmbWrFkbvT1J9+7dc/vtt2fdunUbvRp9l112yauvvlr19XvvvZf58+fnsMMO2+RcDz/8cNq3b5+LL7646rn/+yGZG957+vTphfc4b9CgQYYMGZKJEyemYcOG+cpXvrLJ8N61a9dMmTKl2nOPPvpota979uyZBQsWpGPHjhs9xgd9vzamJn/+++67b/bdd9+MGTMmffr0yc9//vMceOCB6dmzZxYtWlQ4NwDAJ4mIDgAA25j69etn4cKFVX/+R82aNcv555+fc889N5WVlfn85z+f5cuXZ86cOWnevHmGDBmSTp065cc//nF++9vfZvfdd89PfvKTPP7449l9990/1mydOnXK3XffnWOOOSZlZWUZO3ZstSvXO3TokCFDhmTYsGG5/vrrs88+++Tll1/OsmXLMnjw4IwcOTI33HBDvvKVr2TMmDGpqKjIo48+mt69e6dLly754he/mPPOOy9Tp07NZz/72VxzzTV56623NmuuJUuWZPLkyenVq1emTp36vnumX3bZZTn88MPz2c9+Nl/5yleyfv363H///Rk9enTVPsOHD0/Xrl2T/D1Eb8oZZ5yRq6++OhdccEGGDx+eJ5544n0fyjp69OgceOCBGTlyZIYPH54mTZpkwYIFmTZtWv793//9A79fG1MTP/8XX3wxt956a4499ti0adMmixYtyvPPP59TTz01SXLppZfm6KOPTrt27XLiiSemXr16eeqppzJ//vz867/+6wf+PAAAtiX16noAAADgw2vevHm1+3n/oyuvvDJjx47NuHHj0rVr1wwYMCBTp06tiqSnn356jj/++Jx00kk54IAD8sYbb1S7Kvmjuuaaa7LTTjvloIMOyjHHHJP+/funZ8+e1fb54Q9/mBNPPDFnnXVW9txzz3z961/PqlWrkiQ777xzfv/732flypXp27dv9ttvv/zoRz+quip92LBhGTJkSE499dT07ds3e+yxxwdehZ4kxx57bM4999yMHDkyPXr0yMMPP5yxY8dW2+fQQw/NXXfdlSlTpqRHjx754he/mD/84Q/V9unUqVMOOuig7Lnnnu+73/k/ateuXf7zP/8z9957b/bZZ5/cfPPN+bd/+7dq+3Tv3j2zZs3Kc889l0MOOST77rtvLr300rRp02azvl9FPu7Pf8cdd8yf/vSnnHDCCencuXO+8Y1vZMSIETn99NOTJP379899992XBx54IL169cqBBx6Ya6+9Nu3bt9/kXAAA26Ky0j/eUBAAAICNKpVK6dSpU84666ycd955dT0OAABbgNu5AAAAbIa//OUvmTx5cl577bXC+6YDAPDJI6IDAABshl133TUtW7bMrbfemp122qmuxwEAYAsR0QEAADaDO2ECAGyffLAoAAAAAAAUENEBAAAAAKCAiA4AAAAAAAVEdAAAAAAAKCCiAwAAAABAAREdAAAAAAAKiOgAAAAAAFBARAcAAAAAgAL/H3gh/sgLPFm2AAAAAElFTkSuQmCC",
      "text/plain": [
       "<Figure size 1500x1000 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "result_pfi = permutation_importance(\n",
    "    before_engineering[1], X_validate[original_columns], y_validate,\n",
    "    n_repeats=10, random_state=42, n_jobs=2\n",
    ")\n",
    "\n",
    "feature_importances_pfi = pd.Series(result_pfi.importances_mean, index=X_train[original_columns].columns.tolist())\n",
    "\n",
    "fig, ax = plt.subplots(figsize=(15,10))\n",
    "feature_importances_pfi.plot.barh(yerr=result_pfi.importances_std, ax=ax)\n",
    "ax.set_title(\"Feature importances using permutation on full model\")\n",
    "ax.set_xlabel(\"Mean accuracy decrease\")\n",
    "fig.tight_layout()\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 665,
   "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>Accuracy (training)</th>\n",
       "      <th>Accuracy (validation)</th>\n",
       "      <th>Precision (validation)</th>\n",
       "      <th>Recall (validation)</th>\n",
       "      <th>AUC</th>\n",
       "      <th>f1</th>\n",
       "      <th>FP</th>\n",
       "      <th>FN</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>Without Feature Engineering</th>\n",
       "      <td>0.969</td>\n",
       "      <td>0.804</td>\n",
       "      <td>0.799</td>\n",
       "      <td>0.795</td>\n",
       "      <td>0.795</td>\n",
       "      <td>0.797</td>\n",
       "      <td>16</td>\n",
       "      <td>19</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Before Feature Engineering - subset of columns</th>\n",
       "      <td>0.963</td>\n",
       "      <td>0.810</td>\n",
       "      <td>0.806</td>\n",
       "      <td>0.800</td>\n",
       "      <td>0.800</td>\n",
       "      <td>0.803</td>\n",
       "      <td>15</td>\n",
       "      <td>19</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                                                Accuracy (training)  \\\n",
       "Without Feature Engineering                                   0.969   \n",
       "Before Feature Engineering - subset of columns                0.963   \n",
       "\n",
       "                                                Accuracy (validation)  \\\n",
       "Without Feature Engineering                                     0.804   \n",
       "Before Feature Engineering - subset of columns                  0.810   \n",
       "\n",
       "                                                Precision (validation)  \\\n",
       "Without Feature Engineering                                      0.799   \n",
       "Before Feature Engineering - subset of columns                   0.806   \n",
       "\n",
       "                                                Recall (validation)    AUC  \\\n",
       "Without Feature Engineering                                   0.795  0.795   \n",
       "Before Feature Engineering - subset of columns                0.800  0.800   \n",
       "\n",
       "                                                   f1  FP  FN  \n",
       "Without Feature Engineering                     0.797  16  19  \n",
       "Before Feature Engineering - subset of columns  0.803  15  19  "
      ]
     },
     "execution_count": 665,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "selected_columns = ['Pclass', 'IsMale', 'Age', 'SibSp', 'Parch', 'Fare']\n",
    "\n",
    "before_engineering_2 = fit_assess(\n",
    "    X_train=X_train[selected_columns],\n",
    "    X_validate=X_validate[selected_columns],\n",
    "    name=\"Before Feature Engineering - subset of columns\"\n",
    ")\n",
    "\n",
    "results_df = pd.concat(\n",
    "    [results_df,\n",
    "     before_engineering_2[0]\n",
    "     ]\n",
    ")\n",
    "\n",
    "results_df"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Performance with engineered features "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 666,
   "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>Accuracy (training)</th>\n",
       "      <th>Accuracy (validation)</th>\n",
       "      <th>Precision (validation)</th>\n",
       "      <th>Recall (validation)</th>\n",
       "      <th>AUC</th>\n",
       "      <th>f1</th>\n",
       "      <th>FP</th>\n",
       "      <th>FN</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>Without Feature Engineering</th>\n",
       "      <td>0.969</td>\n",
       "      <td>0.804</td>\n",
       "      <td>0.799</td>\n",
       "      <td>0.795</td>\n",
       "      <td>0.795</td>\n",
       "      <td>0.797</td>\n",
       "      <td>16</td>\n",
       "      <td>19</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Before Feature Engineering - subset of columns</th>\n",
       "      <td>0.963</td>\n",
       "      <td>0.810</td>\n",
       "      <td>0.806</td>\n",
       "      <td>0.800</td>\n",
       "      <td>0.800</td>\n",
       "      <td>0.803</td>\n",
       "      <td>15</td>\n",
       "      <td>19</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>After Feature Engineering</th>\n",
       "      <td>0.975</td>\n",
       "      <td>0.782</td>\n",
       "      <td>0.776</td>\n",
       "      <td>0.774</td>\n",
       "      <td>0.774</td>\n",
       "      <td>0.775</td>\n",
       "      <td>19</td>\n",
       "      <td>20</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                                                Accuracy (training)  \\\n",
       "Without Feature Engineering                                   0.969   \n",
       "Before Feature Engineering - subset of columns                0.963   \n",
       "After Feature Engineering                                     0.975   \n",
       "\n",
       "                                                Accuracy (validation)  \\\n",
       "Without Feature Engineering                                     0.804   \n",
       "Before Feature Engineering - subset of columns                  0.810   \n",
       "After Feature Engineering                                       0.782   \n",
       "\n",
       "                                                Precision (validation)  \\\n",
       "Without Feature Engineering                                      0.799   \n",
       "Before Feature Engineering - subset of columns                   0.806   \n",
       "After Feature Engineering                                        0.776   \n",
       "\n",
       "                                                Recall (validation)    AUC  \\\n",
       "Without Feature Engineering                                   0.795  0.795   \n",
       "Before Feature Engineering - subset of columns                0.800  0.800   \n",
       "After Feature Engineering                                     0.774  0.774   \n",
       "\n",
       "                                                   f1  FP  FN  \n",
       "Without Feature Engineering                     0.797  16  19  \n",
       "Before Feature Engineering - subset of columns  0.803  15  19  \n",
       "After Feature Engineering                       0.775  19  20  "
      ]
     },
     "execution_count": 666,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "after_engineering = fit_assess(\n",
    "    X_train=X_train,\n",
    "    X_validate=X_validate,\n",
    "    name=\"After Feature Engineering\"\n",
    ")\n",
    "\n",
    "results_df = pd.concat([\n",
    "    results_df,\n",
    "    after_engineering[0]\n",
    "])\n",
    "\n",
    "results_df"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 667,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 1500x1000 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "result_pfi = permutation_importance(\n",
    "    after_engineering[1], X_validate, y_validate,\n",
    "    n_repeats=10, random_state=42, n_jobs=2\n",
    ")\n",
    "\n",
    "feature_importances_pfi = pd.Series(result_pfi.importances_mean, index=X_train.columns.tolist())\n",
    "\n",
    "fig, ax = plt.subplots(figsize=(15,10))\n",
    "feature_importances_pfi.plot.barh(yerr=result_pfi.importances_std, ax=ax)\n",
    "ax.set_title(\"Feature importances using permutation on full model\")\n",
    "ax.set_xlabel(\"Mean accuracy decrease\")\n",
    "fig.tight_layout()\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Let's rerun with just some of the features that are the most impactful. "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 668,
   "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>Accuracy (training)</th>\n",
       "      <th>Accuracy (validation)</th>\n",
       "      <th>Precision (validation)</th>\n",
       "      <th>Recall (validation)</th>\n",
       "      <th>AUC</th>\n",
       "      <th>f1</th>\n",
       "      <th>FP</th>\n",
       "      <th>FN</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>Without Feature Engineering</th>\n",
       "      <td>0.969</td>\n",
       "      <td>0.804</td>\n",
       "      <td>0.799</td>\n",
       "      <td>0.795</td>\n",
       "      <td>0.795</td>\n",
       "      <td>0.797</td>\n",
       "      <td>16</td>\n",
       "      <td>19</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Before Feature Engineering - subset of columns</th>\n",
       "      <td>0.963</td>\n",
       "      <td>0.810</td>\n",
       "      <td>0.806</td>\n",
       "      <td>0.800</td>\n",
       "      <td>0.800</td>\n",
       "      <td>0.803</td>\n",
       "      <td>15</td>\n",
       "      <td>19</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>After Feature Engineering</th>\n",
       "      <td>0.975</td>\n",
       "      <td>0.782</td>\n",
       "      <td>0.776</td>\n",
       "      <td>0.774</td>\n",
       "      <td>0.774</td>\n",
       "      <td>0.775</td>\n",
       "      <td>19</td>\n",
       "      <td>20</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>After Feature Engineering - subset of columns</th>\n",
       "      <td>0.969</td>\n",
       "      <td>0.821</td>\n",
       "      <td>0.816</td>\n",
       "      <td>0.816</td>\n",
       "      <td>0.816</td>\n",
       "      <td>0.816</td>\n",
       "      <td>16</td>\n",
       "      <td>16</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                                                Accuracy (training)  \\\n",
       "Without Feature Engineering                                   0.969   \n",
       "Before Feature Engineering - subset of columns                0.963   \n",
       "After Feature Engineering                                     0.975   \n",
       "After Feature Engineering - subset of columns                 0.969   \n",
       "\n",
       "                                                Accuracy (validation)  \\\n",
       "Without Feature Engineering                                     0.804   \n",
       "Before Feature Engineering - subset of columns                  0.810   \n",
       "After Feature Engineering                                       0.782   \n",
       "After Feature Engineering - subset of columns                   0.821   \n",
       "\n",
       "                                                Precision (validation)  \\\n",
       "Without Feature Engineering                                      0.799   \n",
       "Before Feature Engineering - subset of columns                   0.806   \n",
       "After Feature Engineering                                        0.776   \n",
       "After Feature Engineering - subset of columns                    0.816   \n",
       "\n",
       "                                                Recall (validation)    AUC  \\\n",
       "Without Feature Engineering                                   0.795  0.795   \n",
       "Before Feature Engineering - subset of columns                0.800  0.800   \n",
       "After Feature Engineering                                     0.774  0.774   \n",
       "After Feature Engineering - subset of columns                 0.816  0.816   \n",
       "\n",
       "                                                   f1  FP  FN  \n",
       "Without Feature Engineering                     0.797  16  19  \n",
       "Before Feature Engineering - subset of columns  0.803  15  19  \n",
       "After Feature Engineering                       0.775  19  20  \n",
       "After Feature Engineering - subset of columns   0.816  16  16  "
      ]
     },
     "execution_count": 668,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "selected_columns = ['Pclass', 'IsMale', 'Age', 'TravellingWithFamily', 'NumPeopleTicketShared', 'FarePerPerson']\n",
    "\n",
    "after_engineering_2 = fit_assess(\n",
    "    X_train=X_train[selected_columns],\n",
    "    X_validate=X_validate[selected_columns],\n",
    "    name=\"After Feature Engineering - subset of columns\"\n",
    ")\n",
    "\n",
    "results_df = pd.concat(\n",
    "    [results_df,\n",
    "     after_engineering_2[0]\n",
    "     ]\n",
    ")\n",
    "\n",
    "results_df"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "While minor in this case - and potentially further influenced by the number of features we have chosen and the method by which we have selected them - feature engineering can have a positive impact on your models, and this notebook has hopefully given you an insight into different strategies for creating additional columns in pandas."
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "ml",
   "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": "3.11.4"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}