01-predicting-survival-titanic-assignement.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Predicting Survival on the Titanic\n",
    "\n",
    "### History\n",
    "Perhaps one of the most infamous shipwrecks in history, the Titanic sank after colliding with an iceberg, killing 1502 out of 2224 people on board. Interestingly, by analysing the probability of survival based on few attributes like gender, age, and social status, we can make very accurate predictions on which passengers would survive. Some groups of people were more likely to survive than others, such as women, children, and the upper-class. Therefore, we can learn about the society priorities and privileges at the time.\n",
    "\n",
    "### Assignment:\n",
    "\n",
    "Build a Machine Learning Pipeline, to engineer the features in the data set and predict who is more likely to Survive the catastrophe.\n",
    "\n",
    "Follow the Jupyter notebook below, and complete the missing bits of code, to achieve each one of the pipeline steps."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "import re\n",
    "\n",
    "# to handle datasets\n",
    "import pandas as pd\n",
    "import numpy as np\n",
    "\n",
    "# for visualization\n",
    "import matplotlib.pyplot as plt\n",
    "\n",
    "# to divide train and test set\n",
    "from sklearn.model_selection import train_test_split\n",
    "\n",
    "# feature scaling\n",
    "from sklearn.preprocessing import StandardScaler\n",
    "\n",
    "# to build the models\n",
    "from sklearn.linear_model import LogisticRegression\n",
    "\n",
    "# to evaluate the models\n",
    "from sklearn.metrics import accuracy_score, roc_auc_score\n",
    "\n",
    "# to persist the model and the scaler\n",
    "import joblib\n",
    "\n",
    "# to visualise al the columns in the dataframe\n",
    "pd.pandas.set_option('display.max_columns', None)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Prepare the data set"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>pclass</th>\n",
       "      <th>survived</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>boat</th>\n",
       "      <th>body</th>\n",
       "      <th>home.dest</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>Allen, Miss. Elisabeth Walton</td>\n",
       "      <td>female</td>\n",
       "      <td>29</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>24160</td>\n",
       "      <td>211.3375</td>\n",
       "      <td>B5</td>\n",
       "      <td>S</td>\n",
       "      <td>2</td>\n",
       "      <td>?</td>\n",
       "      <td>St Louis, MO</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>Allison, Master. Hudson Trevor</td>\n",
       "      <td>male</td>\n",
       "      <td>0.9167</td>\n",
       "      <td>1</td>\n",
       "      <td>2</td>\n",
       "      <td>113781</td>\n",
       "      <td>151.55</td>\n",
       "      <td>C22 C26</td>\n",
       "      <td>S</td>\n",
       "      <td>11</td>\n",
       "      <td>?</td>\n",
       "      <td>Montreal, PQ / Chesterville, ON</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>Allison, Miss. Helen Loraine</td>\n",
       "      <td>female</td>\n",
       "      <td>2</td>\n",
       "      <td>1</td>\n",
       "      <td>2</td>\n",
       "      <td>113781</td>\n",
       "      <td>151.55</td>\n",
       "      <td>C22 C26</td>\n",
       "      <td>S</td>\n",
       "      <td>?</td>\n",
       "      <td>?</td>\n",
       "      <td>Montreal, PQ / Chesterville, ON</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>Allison, Mr. Hudson Joshua Creighton</td>\n",
       "      <td>male</td>\n",
       "      <td>30</td>\n",
       "      <td>1</td>\n",
       "      <td>2</td>\n",
       "      <td>113781</td>\n",
       "      <td>151.55</td>\n",
       "      <td>C22 C26</td>\n",
       "      <td>S</td>\n",
       "      <td>?</td>\n",
       "      <td>135</td>\n",
       "      <td>Montreal, PQ / Chesterville, ON</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>Allison, Mrs. Hudson J C (Bessie Waldo Daniels)</td>\n",
       "      <td>female</td>\n",
       "      <td>25</td>\n",
       "      <td>1</td>\n",
       "      <td>2</td>\n",
       "      <td>113781</td>\n",
       "      <td>151.55</td>\n",
       "      <td>C22 C26</td>\n",
       "      <td>S</td>\n",
       "      <td>?</td>\n",
       "      <td>?</td>\n",
       "      <td>Montreal, PQ / Chesterville, ON</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   pclass  survived                                             name     sex  \\\n",
       "0       1         1                    Allen, Miss. Elisabeth Walton  female   \n",
       "1       1         1                   Allison, Master. Hudson Trevor    male   \n",
       "2       1         0                     Allison, Miss. Helen Loraine  female   \n",
       "3       1         0             Allison, Mr. Hudson Joshua Creighton    male   \n",
       "4       1         0  Allison, Mrs. Hudson J C (Bessie Waldo Daniels)  female   \n",
       "\n",
       "      age  sibsp  parch  ticket      fare    cabin embarked boat body  \\\n",
       "0      29      0      0   24160  211.3375       B5        S    2    ?   \n",
       "1  0.9167      1      2  113781    151.55  C22 C26        S   11    ?   \n",
       "2       2      1      2  113781    151.55  C22 C26        S    ?    ?   \n",
       "3      30      1      2  113781    151.55  C22 C26        S    ?  135   \n",
       "4      25      1      2  113781    151.55  C22 C26        S    ?    ?   \n",
       "\n",
       "                         home.dest  \n",
       "0                     St Louis, MO  \n",
       "1  Montreal, PQ / Chesterville, ON  \n",
       "2  Montreal, PQ / Chesterville, ON  \n",
       "3  Montreal, PQ / Chesterville, ON  \n",
       "4  Montreal, PQ / Chesterville, ON  "
      ]
     },
     "execution_count": 2,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# load the data - it is available open source and online\n",
    "\n",
    "data = pd.read_csv('https://www.openml.org/data/get_csv/16826755/phpMYEkMl')\n",
    "\n",
    "# display data\n",
    "data.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [],
   "source": [
    "# replace interrogation marks by NaN values\n",
    "\n",
    "data = data.replace('?', np.nan)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [],
   "source": [
    "# retain only the first cabin if more than\n",
    "# 1 are available per passenger\n",
    "\n",
    "def get_first_cabin(row):\n",
    "    try:\n",
    "        return row.split()[0]\n",
    "    except:\n",
    "        return np.nan\n",
    "    \n",
    "data['cabin'] = data['cabin'].apply(get_first_cabin)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [],
   "source": [
    "# extracts the title (Mr, Ms, etc) from the name variable\n",
    "\n",
    "def get_title(passenger):\n",
    "    line = passenger\n",
    "    if re.search('Mrs', line):\n",
    "        return 'Mrs'\n",
    "    elif re.search('Mr', line):\n",
    "        return 'Mr'\n",
    "    elif re.search('Miss', line):\n",
    "        return 'Miss'\n",
    "    elif re.search('Master', line):\n",
    "        return 'Master'\n",
    "    else:\n",
    "        return 'Other'\n",
    "    \n",
    "data['title'] = data['name'].apply(get_title)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [],
   "source": [
    "# cast numerical variables as floats\n",
    "\n",
    "data['fare'] = data['fare'].astype('float')\n",
    "data['age'] = data['age'].astype('float')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>pclass</th>\n",
       "      <th>survived</th>\n",
       "      <th>sex</th>\n",
       "      <th>age</th>\n",
       "      <th>sibsp</th>\n",
       "      <th>parch</th>\n",
       "      <th>fare</th>\n",
       "      <th>cabin</th>\n",
       "      <th>embarked</th>\n",
       "      <th>title</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>female</td>\n",
       "      <td>29.0000</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>211.3375</td>\n",
       "      <td>B5</td>\n",
       "      <td>S</td>\n",
       "      <td>Miss</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>male</td>\n",
       "      <td>0.9167</td>\n",
       "      <td>1</td>\n",
       "      <td>2</td>\n",
       "      <td>151.5500</td>\n",
       "      <td>C22</td>\n",
       "      <td>S</td>\n",
       "      <td>Master</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>female</td>\n",
       "      <td>2.0000</td>\n",
       "      <td>1</td>\n",
       "      <td>2</td>\n",
       "      <td>151.5500</td>\n",
       "      <td>C22</td>\n",
       "      <td>S</td>\n",
       "      <td>Miss</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>male</td>\n",
       "      <td>30.0000</td>\n",
       "      <td>1</td>\n",
       "      <td>2</td>\n",
       "      <td>151.5500</td>\n",
       "      <td>C22</td>\n",
       "      <td>S</td>\n",
       "      <td>Mr</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>female</td>\n",
       "      <td>25.0000</td>\n",
       "      <td>1</td>\n",
       "      <td>2</td>\n",
       "      <td>151.5500</td>\n",
       "      <td>C22</td>\n",
       "      <td>S</td>\n",
       "      <td>Mrs</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   pclass  survived     sex      age  sibsp  parch      fare cabin embarked  \\\n",
       "0       1         1  female  29.0000      0      0  211.3375    B5        S   \n",
       "1       1         1    male   0.9167      1      2  151.5500   C22        S   \n",
       "2       1         0  female   2.0000      1      2  151.5500   C22        S   \n",
       "3       1         0    male  30.0000      1      2  151.5500   C22        S   \n",
       "4       1         0  female  25.0000      1      2  151.5500   C22        S   \n",
       "\n",
       "    title  \n",
       "0    Miss  \n",
       "1  Master  \n",
       "2    Miss  \n",
       "3      Mr  \n",
       "4     Mrs  "
      ]
     },
     "execution_count": 7,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# drop unnecessary variables\n",
    "\n",
    "data.drop(labels=['name','ticket', 'boat', 'body','home.dest'], axis=1, inplace=True)\n",
    "\n",
    "# display data\n",
    "data.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [],
   "source": [
    "# save the data set\n",
    "\n",
    "data.to_csv('titanic.csv', index=False)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Data Exploration\n",
    "\n",
    "### Find numerical and categorical variables"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [],
   "source": [
    "target = 'survived'"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "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>pclass</th>\n",
       "      <th>survived</th>\n",
       "      <th>sex</th>\n",
       "      <th>age</th>\n",
       "      <th>sibsp</th>\n",
       "      <th>parch</th>\n",
       "      <th>fare</th>\n",
       "      <th>cabin</th>\n",
       "      <th>embarked</th>\n",
       "      <th>title</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>female</td>\n",
       "      <td>29.0000</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>211.3375</td>\n",
       "      <td>B5</td>\n",
       "      <td>S</td>\n",
       "      <td>Miss</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>male</td>\n",
       "      <td>0.9167</td>\n",
       "      <td>1</td>\n",
       "      <td>2</td>\n",
       "      <td>151.5500</td>\n",
       "      <td>C22</td>\n",
       "      <td>S</td>\n",
       "      <td>Master</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>female</td>\n",
       "      <td>2.0000</td>\n",
       "      <td>1</td>\n",
       "      <td>2</td>\n",
       "      <td>151.5500</td>\n",
       "      <td>C22</td>\n",
       "      <td>S</td>\n",
       "      <td>Miss</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>male</td>\n",
       "      <td>30.0000</td>\n",
       "      <td>1</td>\n",
       "      <td>2</td>\n",
       "      <td>151.5500</td>\n",
       "      <td>C22</td>\n",
       "      <td>S</td>\n",
       "      <td>Mr</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>female</td>\n",
       "      <td>25.0000</td>\n",
       "      <td>1</td>\n",
       "      <td>2</td>\n",
       "      <td>151.5500</td>\n",
       "      <td>C22</td>\n",
       "      <td>S</td>\n",
       "      <td>Mrs</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   pclass  survived     sex      age  sibsp  parch      fare cabin embarked  \\\n",
       "0       1         1  female  29.0000      0      0  211.3375    B5        S   \n",
       "1       1         1    male   0.9167      1      2  151.5500   C22        S   \n",
       "2       1         0  female   2.0000      1      2  151.5500   C22        S   \n",
       "3       1         0    male  30.0000      1      2  151.5500   C22        S   \n",
       "4       1         0  female  25.0000      1      2  151.5500   C22        S   \n",
       "\n",
       "    title  \n",
       "0    Miss  \n",
       "1  Master  \n",
       "2    Miss  \n",
       "3      Mr  \n",
       "4     Mrs  "
      ]
     },
     "execution_count": 10,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data.head()\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Number of numerical variables: 6\n",
      "Number of categorical variables: 4\n"
     ]
    }
   ],
   "source": [
    "#vars_num = # fill your code here \n",
    "\n",
    "vars_cat = data.select_dtypes(object).columns\n",
    "vars_num = data.select_dtypes(np.number).columns\n",
    "\n",
    "print('Number of numerical variables: {}'.format(len(vars_num)))\n",
    "print('Number of categorical variables: {}'.format(len(vars_cat)))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Find missing values in variables"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "pclass        0\n",
       "survived      0\n",
       "age         263\n",
       "sibsp         0\n",
       "parch         0\n",
       "fare          1\n",
       "dtype: int64"
      ]
     },
     "execution_count": 12,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# first in numerical variables\n",
    "\n",
    "data[vars_num].isnull().sum()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "sex            0\n",
       "cabin       1014\n",
       "embarked       2\n",
       "title          0\n",
       "dtype: int64"
      ]
     },
     "execution_count": 13,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# now in categorical variables\n",
    "\n",
    "data[vars_cat].isnull().sum()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Determine cardinality of categorical variables"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "sex           2\n",
       "cabin       181\n",
       "embarked      3\n",
       "title         5\n",
       "dtype: int64"
      ]
     },
     "execution_count": 18,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data[vars_cat].nunique()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Determine the distribution of numerical variables"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 400x200 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 400x200 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 400x200 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 400x200 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 400x200 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 400x200 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "def plot_distribution(var):\n",
    "    data.hist(column=var, figsize=(4,2))\n",
    "\n",
    "for i in vars_num:\n",
    "    plot_distribution(i)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Separate data into train and test\n",
    "\n",
    "Use the code below for reproducibility. Don't change it."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "((1047, 9), (262, 9))"
      ]
     },
     "execution_count": 21,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "X_train, X_test, y_train, y_test = train_test_split(\n",
    "    data.drop('survived', axis=1),  # predictors\n",
    "    data['survived'],  # target\n",
    "    test_size=0.2,  # percentage of obs in test set\n",
    "    random_state=0)  # seed to ensure reproducibility\n",
    "\n",
    "X_train.shape, X_test.shape"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Feature Engineering\n",
    "\n",
    "### Extract only the letter (and drop the number) from the variable Cabin"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "C:\\Users\\Hany\\AppData\\Roaming\\Python\\Python37\\site-packages\\ipykernel_launcher.py:4: DeprecationWarning: `np.str` is a deprecated alias for the builtin `str`. To silence this warning, use `str` by itself. Doing this will not modify any behavior and is safe. If you specifically wanted the numpy scalar type, use `np.str_` here.\n",
      "Deprecated in NumPy 1.20; for more details and guidance: https://numpy.org/devdocs/release/1.20.0-notes.html#deprecations\n",
      "  after removing the cwd from sys.path.\n",
      "C:\\Users\\Hany\\AppData\\Roaming\\Python\\Python37\\site-packages\\ipykernel_launcher.py:7: DeprecationWarning: `np.str` is a deprecated alias for the builtin `str`. To silence this warning, use `str` by itself. Doing this will not modify any behavior and is safe. If you specifically wanted the numpy scalar type, use `np.str_` here.\n",
      "Deprecated in NumPy 1.20; for more details and guidance: https://numpy.org/devdocs/release/1.20.0-notes.html#deprecations\n",
      "  import sys\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "C    16\n",
       "B    13\n",
       "D    10\n",
       "E     5\n",
       "A     3\n",
       "F     3\n",
       "G     1\n",
       "Name: cabin, dtype: int64"
      ]
     },
     "execution_count": 22,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "\n",
    "# data['cabin'] = data['cabin'].astype(str).str[0]\n",
    "# data['cabin'].unique()\n",
    "X_train['cabin'] = np.where(X_train['cabin'].notnull(),X_train['cabin'].astype(np.str).str[0],np.nan)\n",
    "X_train['cabin'].value_counts()\n",
    "\n",
    "X_test['cabin'] = np.where(X_test['cabin'].notnull(),X_test['cabin'].astype(np.str).str[0],np.nan)\n",
    "X_test['cabin'].value_counts()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([nan, 'E', 'F', 'A', 'C', 'D', 'B', 'T', 'G'], dtype=object)"
      ]
     },
     "execution_count": 23,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "X_train['cabin'].unique()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Fill in Missing data in numerical variables:\n",
    "\n",
    "- Add a binary missing indicator\n",
    "- Fill NA in original variable with the median"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "pclass        0\n",
       "sex           0\n",
       "age         206\n",
       "sibsp         0\n",
       "parch         0\n",
       "fare          1\n",
       "cabin       803\n",
       "embarked      2\n",
       "title         0\n",
       "dtype: int64"
      ]
     },
     "execution_count": 25,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "X_train.isnull().sum()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "28.0\n",
      "14.4542\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "pclass           0\n",
       "sex              0\n",
       "age              0\n",
       "sibsp            0\n",
       "parch            0\n",
       "fare             0\n",
       "cabin          803\n",
       "embarked         2\n",
       "title            0\n",
       "age_isnull       0\n",
       "fare_isnull      0\n",
       "dtype: int64"
      ]
     },
     "execution_count": 32,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "for i in ['age','fare']:\n",
    "\n",
    "    # calculate the mode using the train set\n",
    "    med_val = X_train[i].median()\n",
    "    print(med_val)\n",
    "    # add binary missing indicator _isnull(in train and test)\n",
    "    X_train[i+'_isnull'] = np.where(X_train[i].isnull(), 1, 0)\n",
    "    X_test[i+'_isnull'] = np.where(X_test[i].isnull(), 1, 0)\n",
    "# X_train.head()\n",
    "    # replace missing values by the median value (in train and test)\n",
    "    X_train[i] = X_train[i].fillna(med_val)\n",
    "    X_test[i] = X_test[i].fillna(med_val)\n",
    "\n",
    "X_train.isnull().sum() "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Replace Missing data in categorical variables with the string **Missing**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 33,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "pclass           0\n",
       "sex              0\n",
       "age              0\n",
       "sibsp            0\n",
       "parch            0\n",
       "fare             0\n",
       "cabin          803\n",
       "embarked         2\n",
       "title            0\n",
       "age_isnull       0\n",
       "fare_isnull      0\n",
       "dtype: int64"
      ]
     },
     "execution_count": 33,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "X_train.isnull().sum()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 34,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "pclass           0\n",
       "sex              0\n",
       "age              0\n",
       "sibsp            0\n",
       "parch            0\n",
       "fare             0\n",
       "cabin          211\n",
       "embarked         0\n",
       "title            0\n",
       "age_isnull       0\n",
       "fare_isnull      0\n",
       "dtype: int64"
      ]
     },
     "execution_count": 34,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "X_test.isnull().sum()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 35,
   "metadata": {},
   "outputs": [],
   "source": [
    "X_train[vars_cat] = X_train[vars_cat].fillna('Missing')\n",
    "X_test[vars_cat] = X_test[vars_cat].fillna('Missing')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 39,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "pclass         0\n",
       "sex            0\n",
       "age            0\n",
       "sibsp          0\n",
       "parch          0\n",
       "fare           0\n",
       "cabin          0\n",
       "embarked       0\n",
       "title          0\n",
       "age_isnull     0\n",
       "fare_isnull    0\n",
       "dtype: int64"
      ]
     },
     "execution_count": 39,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "X_test.isnull().sum()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 38,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "pclass         0\n",
       "sex            0\n",
       "age            0\n",
       "sibsp          0\n",
       "parch          0\n",
       "fare           0\n",
       "cabin          0\n",
       "embarked       0\n",
       "title          0\n",
       "age_isnull     0\n",
       "fare_isnull    0\n",
       "dtype: int64"
      ]
     },
     "execution_count": 38,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "X_train.isnull().sum()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Remove rare labels in categorical variables\n",
    "\n",
    "- remove labels present in less than 5 % of the passengers"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 40,
   "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>sex</th>\n",
       "      <th>cabin</th>\n",
       "      <th>embarked</th>\n",
       "      <th>title</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>count</th>\n",
       "      <td>1047</td>\n",
       "      <td>1047</td>\n",
       "      <td>1047</td>\n",
       "      <td>1047</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>unique</th>\n",
       "      <td>2</td>\n",
       "      <td>9</td>\n",
       "      <td>4</td>\n",
       "      <td>5</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>top</th>\n",
       "      <td>male</td>\n",
       "      <td>Missing</td>\n",
       "      <td>S</td>\n",
       "      <td>Mr</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>freq</th>\n",
       "      <td>663</td>\n",
       "      <td>803</td>\n",
       "      <td>743</td>\n",
       "      <td>596</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "         sex    cabin embarked title\n",
       "count   1047     1047     1047  1047\n",
       "unique     2        9        4     5\n",
       "top     male  Missing        S    Mr\n",
       "freq     663      803      743   596"
      ]
     },
     "execution_count": 40,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "X_train[vars_cat].describe()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "#Checking the frequency of each label\n",
    "def label_freq(var):\n",
    "    label_freq = data[var].value_counts() / len(data)\n",
    "    print(label_freq)\n",
    "    fig = label_freq.sort_values(ascending=False).plot.bar()\n",
    "    fig.axhline(y=0.05, color='red')\n",
    "    fig.set_ylabel(f'percentage of {var} within each category')\n",
    "    fig.set_xlabel(f'Variable: {var}')\n",
    "    fig.set_title('Identifying Rare Categories')\n",
    "    plt.show()\n",
    "\n",
    "   \n",
    "for i in vars_cat:\n",
    "    label_freq(i)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "def find_frequent_labels(df, var, rare_perc, target):\n",
    "    df = df.copy()\n",
    "    rare_qty = int(df.shape[0] * rare_perc)\n",
    "    tmp = df[var].value_counts()\n",
    "    tmp = tmp[tmp>rare_qty]\n",
    "    tmp = tmp.index.tolist()\n",
    "    \n",
    "    return df[df[var].isin(tmp)][var]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "for var in vars_cat:\n",
    "    \n",
    "    # find the least frequent labels\n",
    "    frequent_labels = find_frequent_labels(X_train, var, 0.05, target)\n",
    "    \n",
    "    # replace rare categories by the string \"Other\"\n",
    "    X_train[var] = np.where(X_train[var].isin(\n",
    "        frequent_labels), X_train[var], 'Rare')\n",
    "    \n",
    "    X_test[var] = np.where(X_test[var].isin(\n",
    "        frequent_labels), X_test[var], 'Rare')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "X_train['title'].value_counts()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "X_test['title'].value_counts()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Perform one hot encoding of categorical variables into k-1 binary variables\n",
    "\n",
    "- k-1, means that if the variable contains 9 different categories, we create 8 different binary variables\n",
    "- Remember to drop the original categorical variable (the one with the strings) after the encoding"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 41,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "pclass         0\n",
       "sex            0\n",
       "age            0\n",
       "sibsp          0\n",
       "parch          0\n",
       "fare           0\n",
       "cabin          0\n",
       "embarked       0\n",
       "title          0\n",
       "age_isnull     0\n",
       "fare_isnull    0\n",
       "dtype: int64"
      ]
     },
     "execution_count": 41,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "X_train.isnull().sum()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 44,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "ColumnTransformer(remainder='passthrough',\n",
       "                  transformers=[('OHEncoder', OneHotEncoder(drop='first'),\n",
       "                                 Index(['sex', 'cabin', 'embarked', 'title'], dtype='object'))])"
      ]
     },
     "execution_count": 44,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "from sklearn.compose import ColumnTransformer\n",
    "from sklearn.preprocessing import LabelEncoder, OneHotEncoder\n",
    "\n",
    "transformed = ColumnTransformer(transformers=[('OHEncoder',OneHotEncoder(drop='first'),vars_cat)],remainder='passthrough')\n",
    "transformed"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 45,
   "metadata": {},
   "outputs": [],
   "source": [
    "X_train = np.array(transformed.fit_transform(X_train))\n",
    "X_test = np.array(transformed.transform(X_test))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 48,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([[  1.    ,   0.    ,   0.    , ...,   7.925 ,   0.    ,   0.    ],\n",
       "       [  0.    ,   0.    ,   0.    , ..., 134.5   ,   0.    ,   0.    ],\n",
       "       [  1.    ,   0.    ,   0.    , ...,   7.7333,   0.    ,   0.    ],\n",
       "       ...,\n",
       "       [  0.    ,   0.    ,   0.    , ...,   7.7333,   0.    ,   0.    ],\n",
       "       [  0.    ,   0.    ,   0.    , ...,  36.75  ,   0.    ,   0.    ],\n",
       "       [  0.    ,   0.    ,   0.    , ...,  15.5   ,   0.    ,   0.    ]])"
      ]
     },
     "execution_count": 48,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "X_train"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Scale the variables\n",
    "\n",
    "- Use the standard scaler from Scikit-learn"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 50,
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn.preprocessing import StandardScaler\n",
    "scaler = StandardScaler()\n",
    "\n",
    "scaler.fit(X_train)\n",
    "\n",
    "#Scaling both test and train dataframes to the same range\n",
    "X_train = scaler.fit_transform(X_train)\n",
    "X_test = scaler.transform(X_test)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Train the Logistic Regression model\n",
    "\n",
    "- Set the regularization parameter to 0.0005\n",
    "- Set the seed to 0"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 51,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "LogisticRegression(random_state=0, tol=0.0005)"
      ]
     },
     "execution_count": 51,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "log_reg = LogisticRegression(penalty='l2', tol=0.0005, random_state=0)\n",
    "log_reg.fit(X_train, y_train)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Make predictions and evaluate model performance\n",
    "\n",
    "Determine:\n",
    "- roc-auc\n",
    "- accuracy\n",
    "\n",
    "**Important, remember that to determine the accuracy, you need the outcome 0, 1, referring to survived or not. But to determine the roc-auc you need the probability of survival.**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 54,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Accuracy: 0.8015267175572519\n",
      "ROC-AUC : 0.7782716049382716\n"
     ]
    }
   ],
   "source": [
    "y_pred = log_reg.predict(X_test)\n",
    "\n",
    "print(f'Accuracy: {accuracy_score(y_test.values, y_pred)}')\n",
    "print(f'ROC-AUC : {roc_auc_score(y_test.values, y_pred)}')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "That's it! Well done\n",
    "\n",
    "**Keep this code safe, as we will use this notebook later on, to build production code, in our next assignement!!**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.7.1"
  },
  "toc": {
   "base_numbering": 1,
   "nav_menu": {},
   "number_sections": true,
   "sideBar": true,
   "skip_h1_title": false,
   "title_cell": "Table of Contents",
   "title_sidebar": "Contents",
   "toc_cell": false,
   "toc_position": {},
   "toc_section_display": true,
   "toc_window_display": true
  },
  "vscode": {
   "interpreter": {
    "hash": "155e95bcfe30785d6af86bf5b2859a0cb72daaf336d93b2b63095b57782248e5"
   }
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}