Day-1/Regressions/Lab_on_Logistic_Regression.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "<h1 style=\"text-align: center;\" markdown=\"1\">AI Winter School 2019</h1> \n",
    "<h2 style=\"text-align: center;\" markdown=\"2\">Dec 10 - Dec 20, 2019</h2>\n",
    "<h3 style=\"text-align: center;\" markdown=\"3\">Lab on Logistic Regression</h3>\n",
    "\n",
    "\n",
    "> *This notebook is part of the AI Winter School 2019 organized by NAAMII. The objectives of this notebook is to use Logistic Regression to predict whether an individual survived or not during the sinking of the Titanic. The data used in this exercise is adopted from https://www.kaggle.com/c/titanic/data.*"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Table of Contents\n",
    "\n",
    "[Introduction](#introduction)    \n",
    "&nbsp;&nbsp;&nbsp;[Explanation and Theory](#introduction-theory)    \n",
    "&nbsp;&nbsp;&nbsp;[Considerations](#introduction-considerations)    \n",
    "[Data Setup](#setup)  \n",
    "&nbsp;&nbsp;&nbsp;[Meta data](#variables)  \n",
    "&nbsp;&nbsp;&nbsp;[Correlation](#correlation)  \n",
    "&nbsp;&nbsp;&nbsp;[Dummy variables for Categorical Data](#dummy)      \n",
    "[Scikit-Learn Logistic Regression](#sklearn-logreg)  \n",
    "[Summary](#summary)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Logistic Regression Classifier Introduction <a class=\"anchor\" id=\"introduction\"></a>\n",
    "\n",
    "Logistic regression is one of the most simple and well-known machine learning algorithms for classification. Despite its name, it is used for classification rather than regression. \n",
    "\n",
    "In basic terms, it predicts the probability of occurrence of an event by fitting the data to a logistic function. This probability is then translated into a class label based on the set threshold of the function."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Explanation and Theory <a class=\"anchor\" id=\"introduction-theory\"></a>\n",
    "\n",
    "**Assumptions and properties.** \n",
    "Suppose we have a data set that consists of n samples and m features\n",
    "\n",
    "\\begin{equation}\n",
    "\\label{eqn:samples}\n",
    "\\mathbf{X} = \\{\\mathbf{x_1}, \\mathbf{x_2}, \\ldots \\mathbf{x_n}\\}, \\quad   \\mathbf{x_i} \\in \\mathbb{R}^{m}\n",
    "\\end{equation}\n",
    "\n",
    "where\n",
    "\n",
    "\\begin{equation}\n",
    "\\label{eqn:features}\n",
    "\\mathbf{x_i} =  (x_i^{(1)}, x_i^{(2)}, \\ldots x_i^{(m)} ) ^T\n",
    "\\end{equation}\n",
    "\n",
    "The target variable is the probability of a sample belonging to a certain class and is represented by\n",
    "\\begin{equation}\n",
    "\\label{eqn:target}\n",
    "\\mathbf{Y}= \\{y_1, y_2 \\ldots y_n \\} \\quad where \\,\\,\\, y_i \\in (0,1)\n",
    "\\end{equation}\n",
    "\n",
    "Let us assume our problem is a binary classification problem, meaning the response/dependent variables has two classes or labels 0 and 1. If we used linear regression, it would give us a straight line that best separates 0 and 1 responses. However, we could not use this line to give us a probability, since it would give us a negative value for the responses near zero on the x-axis. Instead, when we use logistic regression, we fit this data and estimate the target variable using the following **logistic/sigmoid function**:  \n",
    "\n",
    "\\begin{equation}\n",
    "\\label{eqn:sigmoid}\n",
    "y_i =\\frac {e^{(\\beta_0 + \\mathbf{\\beta} \\cdot \\mathbf{x_i})}} {(1 + e^{(\\beta_0 + \\mathbf{\\beta} \\cdot \\mathbf{x_i})})}\n",
    "\\end{equation}\n",
    "where $\\beta_0$ is called the bias term or the intercept, and $\\mathbf{\\beta}$ are the coefficients associated with the feature vector $\\mathbf{x_i}$. \n",
    "<img alt=\"logistic function diagram\" src=\"logit.png\"/>\n",
    "\n",
    "The function transforms all input variables to the range [0,1], which brings the smallest or most negative numbers close to zero and the largest positive numbers close to one. This allows us to take real-valued inputs and output a probability of the input belonging to either class zero or one. We can then choose a threshold value, such as 0.5, and provide the class output.\n",
    "\n",
    "**Algorithm and Training.** Logistic regression takes the form of a linear model:\n",
    "\n",
    "$$f(i)=\\beta_0+\\beta_1x_{1,i}+...+\\beta_mx_{m,i} $$\n",
    "\n",
    "where $\\beta_0,...,\\beta_m$ are the regression coefficients or weights assigned to each feature $x$. For each data point $i$, a pseudo-variable $x_{0,i}=1$ is added to correspond to the intercept coefficient $\\beta_0$. This allows us to write the model in vector form as:\n",
    "$$f(i)=\\boldsymbol{\\beta}\\cdot\\boldsymbol{X_i}$$\n",
    "\n",
    "When we train the logistic regression classifier, we are trying to find the best values of $\\beta$ to match the data. This is done using an estimation method that attempts to minimize the error of the model. There are several techniques to do this, such as [gradient descent](https://en.wikipedia.org/wiki/Gradient_descent)."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Considerations <a class=\"anchor\" id=\"introduction-considerations\"></a>\n",
    "\n",
    "There are a few things to remember when using logistic regression as a classifier. First, it assumes that there is a linear relationship between the independent variables and the dependent variables. In high-dimensional datasets, this may not be the case, so logistic regression may not be the best choice of classifier.\n",
    "\n",
    "Logisitic regression is also sensitive to highly correlated inputs. Having highly correlated inputs can cause the model to be overfit or will cause the model to fail to converge. We will take a closer look at the correlations between different variables in this notebook. "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Data setup <a class=\"anchor\" id=\"setup\"></a>"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "colab_type": "text",
    "id": "vSzTxrDg46Vv"
   },
   "source": [
    "\n",
    "![alt text](https://upload.wikimedia.org/wikipedia/commons/thumb/f/fd/RMS_Titanic_3.jpg/1280px-RMS_Titanic_3.jpg)\n",
    "\n",
    "The sinking of the RMS Titanic is one of the most infamous shipwrecks in history.  On April 15, 1912, during her maiden voyage, the Titanic sank after colliding with an iceberg, killing 1502 out of 2224 passengers and crew. This sensational tragedy shocked the international community and led to better safety regulations for ships.\n",
    "\n",
    "One of the reasons that the shipwreck led to such loss of life was that there were not enough lifeboats for the passengers and crew. Although there was some element of luck involved in surviving the sinking, some groups of people were more likely to survive than others, such as women, children, and the upper-class.\n",
    "\n",
    "Our goal is to use predict if an individual survived or not in the titanic ship wreck. \n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {
    "colab": {},
    "colab_type": "code",
    "id": "ban1LsHV47fI"
   },
   "outputs": [],
   "source": [
    "# All our imports\n",
    "import pandas as pd\n",
    "from sklearn.model_selection import train_test_split\n",
    "from sklearn.linear_model import LogisticRegression\n",
    "from sklearn.metrics import accuracy_score\n",
    "from sklearn.metrics import confusion_matrix\n",
    "from matplotlib import pyplot as plt\n",
    "%matplotlib inline\n",
    "import seaborn as sns"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {
    "colab": {},
    "colab_type": "code",
    "id": "wNw6JofC4_Nf"
   },
   "outputs": [],
   "source": [
    "#Run this only if you are using Google Colab\n",
    "# from google.colab import drive\n",
    "# drive.mount('/content/drive')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {
    "colab": {},
    "colab_type": "code",
    "id": "MfzC3gW64_3e"
   },
   "outputs": [],
   "source": [
    "#Enter the path of your file inside the quotes\n",
    "path = \"titanic_clean.csv\""
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {
    "colab": {},
    "colab_type": "code",
    "id": "wWaSW9Tv5Q5W"
   },
   "outputs": [],
   "source": [
    "#Write the code to read the csv file to a dataframe df\n",
    "df = pd.read_csv(path)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {
    "colab": {},
    "colab_type": "code",
    "id": "Qyou3IbK5TQu"
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Index(['PassengerId', 'Pclass', 'Name', 'Sex', 'Age', 'SibSp', 'Parch',\n",
       "       'Ticket', 'Fare', 'Embarked', 'Survived', 'Initial'],\n",
       "      dtype='object')"
      ]
     },
     "execution_count": 5,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.columns"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "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>Pclass</th>\n",
       "      <th>Age</th>\n",
       "      <th>SibSp</th>\n",
       "      <th>Parch</th>\n",
       "      <th>Fare</th>\n",
       "      <th>Survived</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>count</th>\n",
       "      <td>891.000000</td>\n",
       "      <td>891.000000</td>\n",
       "      <td>891.000000</td>\n",
       "      <td>891.000000</td>\n",
       "      <td>891.000000</td>\n",
       "      <td>891.000000</td>\n",
       "      <td>891.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>mean</th>\n",
       "      <td>446.000000</td>\n",
       "      <td>2.308642</td>\n",
       "      <td>29.066409</td>\n",
       "      <td>0.523008</td>\n",
       "      <td>0.381594</td>\n",
       "      <td>32.204208</td>\n",
       "      <td>0.383838</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>std</th>\n",
       "      <td>257.353842</td>\n",
       "      <td>0.836071</td>\n",
       "      <td>13.244532</td>\n",
       "      <td>1.102743</td>\n",
       "      <td>0.806057</td>\n",
       "      <td>49.693429</td>\n",
       "      <td>0.486592</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>min</th>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.420000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>25%</th>\n",
       "      <td>223.500000</td>\n",
       "      <td>2.000000</td>\n",
       "      <td>22.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>7.910400</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>50%</th>\n",
       "      <td>446.000000</td>\n",
       "      <td>3.000000</td>\n",
       "      <td>26.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>14.454200</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>75%</th>\n",
       "      <td>668.500000</td>\n",
       "      <td>3.000000</td>\n",
       "      <td>37.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>31.000000</td>\n",
       "      <td>1.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>max</th>\n",
       "      <td>891.000000</td>\n",
       "      <td>3.000000</td>\n",
       "      <td>80.000000</td>\n",
       "      <td>8.000000</td>\n",
       "      <td>6.000000</td>\n",
       "      <td>512.329200</td>\n",
       "      <td>1.000000</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "       PassengerId      Pclass         Age       SibSp       Parch  \\\n",
       "count   891.000000  891.000000  891.000000  891.000000  891.000000   \n",
       "mean    446.000000    2.308642   29.066409    0.523008    0.381594   \n",
       "std     257.353842    0.836071   13.244532    1.102743    0.806057   \n",
       "min       1.000000    1.000000    0.420000    0.000000    0.000000   \n",
       "25%     223.500000    2.000000   22.000000    0.000000    0.000000   \n",
       "50%     446.000000    3.000000   26.000000    0.000000    0.000000   \n",
       "75%     668.500000    3.000000   37.000000    1.000000    0.000000   \n",
       "max     891.000000    3.000000   80.000000    8.000000    6.000000   \n",
       "\n",
       "             Fare    Survived  \n",
       "count  891.000000  891.000000  \n",
       "mean    32.204208    0.383838  \n",
       "std     49.693429    0.486592  \n",
       "min      0.000000    0.000000  \n",
       "25%      7.910400    0.000000  \n",
       "50%     14.454200    0.000000  \n",
       "75%     31.000000    1.000000  \n",
       "max    512.329200    1.000000  "
      ]
     },
     "execution_count": 6,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.describe()"
   ]
  },
  {
   "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>PassengerId</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>Embarked</th>\n",
       "      <th>Survived</th>\n",
       "      <th>Initial</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>1</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>S</td>\n",
       "      <td>0</td>\n",
       "      <td>Mr</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>2</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>C</td>\n",
       "      <td>1</td>\n",
       "      <td>Mrs</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>3</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>S</td>\n",
       "      <td>1</td>\n",
       "      <td>Miss</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>4</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>S</td>\n",
       "      <td>1</td>\n",
       "      <td>Mrs</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>5</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>S</td>\n",
       "      <td>0</td>\n",
       "      <td>Mr</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   PassengerId  Pclass                                               Name  \\\n",
       "0            1       3                            Braund, Mr. Owen Harris   \n",
       "1            2       1  Cumings, Mrs. John Bradley (Florence Briggs Th...   \n",
       "2            3       3                             Heikkinen, Miss. Laina   \n",
       "3            4       1       Futrelle, Mrs. Jacques Heath (Lily May Peel)   \n",
       "4            5       3                           Allen, Mr. William Henry   \n",
       "\n",
       "      Sex   Age  SibSp  Parch            Ticket     Fare Embarked  Survived  \\\n",
       "0    male  22.0      1      0         A/5 21171   7.2500        S         0   \n",
       "1  female  38.0      1      0          PC 17599  71.2833        C         1   \n",
       "2  female  26.0      0      0  STON/O2. 3101282   7.9250        S         1   \n",
       "3  female  35.0      1      0            113803  53.1000        S         1   \n",
       "4    male  35.0      0      0            373450   8.0500        S         0   \n",
       "\n",
       "  Initial  \n",
       "0      Mr  \n",
       "1     Mrs  \n",
       "2    Miss  \n",
       "3     Mrs  \n",
       "4      Mr  "
      ]
     },
     "execution_count": 7,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "colab_type": "text",
    "id": "8XLD5-L35jyb"
   },
   "source": [
    "# Variable Metadata <a class=\"anchor\" id=\"variables\"></a>\n",
    "**Pclass:** A proxy for socio-economic status (SES)\n",
    "\n",
    "1 = Upper\n",
    "\n",
    "2 = Middle\n",
    "\n",
    "3 = Lower\n",
    "\n",
    "**Age:** Age is fractional if less than 1. If the age is estimated, is it in the form of xx.5\n",
    "\n",
    "**SibSp:** The dataset defines family relations in this way:\n",
    "\n",
    "Sibling = brother, sister, stepbrother, stepsister\n",
    "\n",
    "Spouse = husband, wife (mistresses and fiancés were ignored)\n",
    "\n",
    "\n",
    "**Parch:** The dataset defines family relations in this way:\n",
    "Parent = mother, father\n",
    "Child = daughter, son, stepdaughter, stepson\n",
    "Some children travelled only with a nanny, therefore parch=0 for them.\n",
    "\n",
    "**Embarked:** The port from where the particular passenger was embarked/boarded. \n",
    "\n",
    "**Survived:** \n",
    "0: if the person did not survive\n",
    "1: if the person survived"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 202
    },
    "colab_type": "code",
    "id": "iyGlMv-O5L6e",
    "outputId": "deac0d7c-e574-4d9d-ebf0-74b3986c4ef0"
   },
   "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>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>Embarked</th>\n",
       "      <th>Survived</th>\n",
       "      <th>Initial</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>1</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>S</td>\n",
       "      <td>0</td>\n",
       "      <td>Mr</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>2</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>C</td>\n",
       "      <td>1</td>\n",
       "      <td>Mrs</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>3</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>S</td>\n",
       "      <td>1</td>\n",
       "      <td>Miss</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>4</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>S</td>\n",
       "      <td>1</td>\n",
       "      <td>Mrs</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>5</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>S</td>\n",
       "      <td>0</td>\n",
       "      <td>Mr</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   PassengerId  Pclass                                               Name  \\\n",
       "0            1       3                            Braund, Mr. Owen Harris   \n",
       "1            2       1  Cumings, Mrs. John Bradley (Florence Briggs Th...   \n",
       "2            3       3                             Heikkinen, Miss. Laina   \n",
       "3            4       1       Futrelle, Mrs. Jacques Heath (Lily May Peel)   \n",
       "4            5       3                           Allen, Mr. William Henry   \n",
       "\n",
       "      Sex   Age  SibSp  Parch            Ticket     Fare Embarked  Survived  \\\n",
       "0    male  22.0      1      0         A/5 21171   7.2500        S         0   \n",
       "1  female  38.0      1      0          PC 17599  71.2833        C         1   \n",
       "2  female  26.0      0      0  STON/O2. 3101282   7.9250        S         1   \n",
       "3  female  35.0      1      0            113803  53.1000        S         1   \n",
       "4    male  35.0      0      0            373450   8.0500        S         0   \n",
       "\n",
       "  Initial  \n",
       "0      Mr  \n",
       "1     Mrs  \n",
       "2    Miss  \n",
       "3     Mrs  \n",
       "4      Mr  "
      ]
     },
     "execution_count": 8,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {
    "colab": {},
    "colab_type": "code",
    "id": "nHy_Po1H-YNT"
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "PassengerId    False\n",
       "Pclass         False\n",
       "Name           False\n",
       "Sex            False\n",
       "Age            False\n",
       "SibSp          False\n",
       "Parch          False\n",
       "Ticket         False\n",
       "Fare           False\n",
       "Embarked       False\n",
       "Survived       False\n",
       "Initial        False\n",
       "dtype: bool"
      ]
     },
     "execution_count": 9,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.isnull().any()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Correlation between variables<a class=\"anchor\" id=\"correlation\"></a>"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 864x864 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "variables = ['PassengerId', 'Pclass', 'Name', 'Sex', 'Age', 'SibSp', 'Parch',\n",
    "       'Ticket', 'Fare', 'Embarked', 'Survived', 'Initial']\n",
    "# Calculate the correlations\n",
    "corr_mat = df[variables].corr().round(2)\n",
    "\n",
    "# Draw a correlation heatmap\n",
    "plt.rcParams['font.size'] = 18\n",
    "plt.figure(figsize = (12, 12))\n",
    "sns.heatmap(corr_mat, vmin = -0.5, vmax = 0.8, center = 0, \n",
    "            cmap = plt.cm.RdYlGn_r, annot = True);"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Let us look at the dataset. What columns do you think are important variables to detect Survival? Does Name matter? Or, does Ticket number matter? What about Pclass?"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {
    "colab": {},
    "colab_type": "code",
    "id": "peJBQQPvL2YY"
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.axes._subplots.AxesSubplot at 0x7f5821d425c0>"
      ]
     },
     "execution_count": 11,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "df[df['Survived']==1].Age.hist()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {
    "colab": {},
    "colab_type": "code",
    "id": "-c3eYmyGL6bV"
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Sex     Survived\n",
       "female  0            81\n",
       "        1           233\n",
       "male    0           468\n",
       "        1           109\n",
       "Name: Survived, dtype: int64"
      ]
     },
     "execution_count": 12,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.groupby(['Sex', 'Survived'])['Survived'].count()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {
    "colab": {},
    "colab_type": "code",
    "id": "cyIgZPOvL-Mm"
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Embarked  Survived\n",
       "C         0            75\n",
       "          1            93\n",
       "Q         0            47\n",
       "          1            30\n",
       "S         0           427\n",
       "          1           219\n",
       "Name: Survived, dtype: int64"
      ]
     },
     "execution_count": 13,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.groupby(['Embarked', 'Survived'])['Survived'].count()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {
    "colab": {},
    "colab_type": "code",
    "id": "FIxzbAkxMLJl"
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.axes._subplots.AxesSubplot at 0x7f5820603f60>"
      ]
     },
     "execution_count": 14,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "sns.countplot('Embarked',hue='Survived',data=df)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {
    "colab": {},
    "colab_type": "code",
    "id": "iU8EF-eTMM_O"
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Pclass  Survived\n",
       "1       0            80\n",
       "        1           136\n",
       "2       0            97\n",
       "        1            87\n",
       "3       0           372\n",
       "        1           119\n",
       "Name: Survived, dtype: int64"
      ]
     },
     "execution_count": 15,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.groupby(['Pclass', 'Survived'])['Survived'].count()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {
    "colab": {},
    "colab_type": "code",
    "id": "GaYoxpSN11sV"
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<seaborn.axisgrid.FacetGrid at 0x7f5820573a90>"
      ]
     },
     "execution_count": 16,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 360x360 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "sns.catplot(x=\"SibSp\", y=\"Survived\", data=df,  kind=\"bar\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "colab_type": "text",
    "id": "zZnO6iLFrq_P"
   },
   "source": [
    "# Dummy variables for categorical data <a class=\"anchor\" id=\"dummy\"></a>\n",
    "Sex and Embarked are categorical data. We will eventually be using Scikit-learn to create our ML model which expects numeric data. We will have to convert categorical variable into dummy/indicator variables."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 1000
    },
    "colab_type": "code",
    "id": "BRX_39P3tbJl",
    "outputId": "c466d0c5-1ac7-4b17-fc99-8d741eac6aa2"
   },
   "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>female</th>\n",
       "      <th>male</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>886</th>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>887</th>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>888</th>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>889</th>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>890</th>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>891 rows × 2 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "     female  male\n",
       "0         0     1\n",
       "1         1     0\n",
       "2         1     0\n",
       "3         1     0\n",
       "4         0     1\n",
       "..      ...   ...\n",
       "886       0     1\n",
       "887       1     0\n",
       "888       1     0\n",
       "889       0     1\n",
       "890       0     1\n",
       "\n",
       "[891 rows x 2 columns]"
      ]
     },
     "execution_count": 17,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Try this, what do you see\n",
    "pd.get_dummies(df['Sex'])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {
    "colab": {},
    "colab_type": "code",
    "id": "BQbW63S16Qu2"
   },
   "outputs": [],
   "source": [
    "# Let us create dummy variables for \"Sex\" column\n",
    "# As you might have noticed, instead of storing male as 0 and 1,\n",
    "# and females as 0 and 1, storing one of them will suffice as they\n",
    "# are mutually exclusive in our dataset. Therefore, we use the drop_first argument\n",
    "\n",
    "sex = pd.get_dummies(df['Sex'], drop_first = True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "S    646\n",
       "C    168\n",
       "Q     77\n",
       "Name: Embarked, dtype: int64"
      ]
     },
     "execution_count": 19,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.Embarked.value_counts()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {
    "colab": {},
    "colab_type": "code",
    "id": "g5EOQ2h16ZZ-"
   },
   "outputs": [],
   "source": [
    "# Lets do the same with our Embarked variable, what do you think will happen\n",
    "# when three variables are converted to dummy variables with drop_first argument?\n",
    "#We do this to avoid something called the \"dummy variable trap\". \n",
    "\n",
    "#Write the code to create dummy variables for \"Embarked\"\n",
    "embarked = pd.get_dummies(df['Embarked'], drop_first=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {
    "colab": {},
    "colab_type": "code",
    "id": "F5CKONVA6w8t"
   },
   "outputs": [],
   "source": [
    "# Removing Categorical Columns and Columns we decided not to use \n",
    "df.drop(['Name', 'Sex', 'Ticket', 'Embarked'], axis = 1, inplace = True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 202
    },
    "colab_type": "code",
    "id": "L-_Hjvl6-BFE",
    "outputId": "8c351b5b-597c-4e19-9670-6a36b40ec40a"
   },
   "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>Pclass</th>\n",
       "      <th>Age</th>\n",
       "      <th>SibSp</th>\n",
       "      <th>Parch</th>\n",
       "      <th>Fare</th>\n",
       "      <th>Survived</th>\n",
       "      <th>Initial</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>1</td>\n",
       "      <td>3</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>Mr</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>2</td>\n",
       "      <td>1</td>\n",
       "      <td>38.0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>71.2833</td>\n",
       "      <td>1</td>\n",
       "      <td>Mrs</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>3</td>\n",
       "      <td>3</td>\n",
       "      <td>26.0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>7.9250</td>\n",
       "      <td>1</td>\n",
       "      <td>Miss</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>4</td>\n",
       "      <td>1</td>\n",
       "      <td>35.0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>53.1000</td>\n",
       "      <td>1</td>\n",
       "      <td>Mrs</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>5</td>\n",
       "      <td>3</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>Mr</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   PassengerId  Pclass   Age  SibSp  Parch     Fare  Survived Initial\n",
       "0            1       3  22.0      1      0   7.2500         0      Mr\n",
       "1            2       1  38.0      1      0  71.2833         1     Mrs\n",
       "2            3       3  26.0      0      0   7.9250         1    Miss\n",
       "3            4       1  35.0      1      0  53.1000         1     Mrs\n",
       "4            5       3  35.0      0      0   8.0500         0      Mr"
      ]
     },
     "execution_count": 22,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# what does the dataset look like now\n",
    "df.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "metadata": {
    "colab": {},
    "colab_type": "code",
    "id": "WkJOo61n-G8M"
   },
   "outputs": [],
   "source": [
    "# Lets add our dummy/indicator columns back to the dataset\n",
    "df = pd.concat([df, sex, embarked], axis = 1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 202
    },
    "colab_type": "code",
    "id": "xFE1CPFw-Stb",
    "outputId": "07e3fc8d-34ba-4150-b0b0-69d3350d8ae2"
   },
   "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>Pclass</th>\n",
       "      <th>Age</th>\n",
       "      <th>SibSp</th>\n",
       "      <th>Parch</th>\n",
       "      <th>Fare</th>\n",
       "      <th>Survived</th>\n",
       "      <th>Initial</th>\n",
       "      <th>male</th>\n",
       "      <th>Q</th>\n",
       "      <th>S</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>1</td>\n",
       "      <td>3</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>Mr</td>\n",
       "      <td>1</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>38.0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>71.2833</td>\n",
       "      <td>1</td>\n",
       "      <td>Mrs</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>3</td>\n",
       "      <td>26.0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>7.9250</td>\n",
       "      <td>1</td>\n",
       "      <td>Miss</td>\n",
       "      <td>0</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>35.0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>53.1000</td>\n",
       "      <td>1</td>\n",
       "      <td>Mrs</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>5</td>\n",
       "      <td>3</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>Mr</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   PassengerId  Pclass   Age  SibSp  Parch     Fare  Survived Initial  male  \\\n",
       "0            1       3  22.0      1      0   7.2500         0      Mr     1   \n",
       "1            2       1  38.0      1      0  71.2833         1     Mrs     0   \n",
       "2            3       3  26.0      0      0   7.9250         1    Miss     0   \n",
       "3            4       1  35.0      1      0  53.1000         1     Mrs     0   \n",
       "4            5       3  35.0      0      0   8.0500         0      Mr     1   \n",
       "\n",
       "   Q  S  \n",
       "0  0  1  \n",
       "1  0  0  \n",
       "2  0  1  \n",
       "3  0  1  \n",
       "4  0  1  "
      ]
     },
     "execution_count": 24,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "colab_type": "text",
    "id": "jWT9ui6l9XmK"
   },
   "source": [
    "Also, PassengerID is simply used for identification purposes, and Survived holds the \"label\" data for our result. We can get rid of these columns. The Survived column is the data we feed the machine learning algorithms to learn from.  "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "metadata": {
    "colab": {},
    "colab_type": "code",
    "id": "46Ppu76i9_sZ"
   },
   "outputs": [],
   "source": [
    "traning_data_set = df.drop(['PassengerId', 'Initial','Survived'], axis=1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "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>Age</th>\n",
       "      <th>SibSp</th>\n",
       "      <th>Parch</th>\n",
       "      <th>Fare</th>\n",
       "      <th>male</th>\n",
       "      <th>Q</th>\n",
       "      <th>S</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>3</td>\n",
       "      <td>22.0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>7.2500</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>1</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>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>3</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>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>1</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>0</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>3</td>\n",
       "      <td>35.0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>8.0500</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   Pclass   Age  SibSp  Parch     Fare  male  Q  S\n",
       "0       3  22.0      1      0   7.2500     1  0  1\n",
       "1       1  38.0      1      0  71.2833     0  0  0\n",
       "2       3  26.0      0      0   7.9250     0  0  1\n",
       "3       1  35.0      1      0  53.1000     0  0  1\n",
       "4       3  35.0      0      0   8.0500     1  0  1"
      ]
     },
     "execution_count": 26,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "traning_data_set.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(891, 8)"
      ]
     },
     "execution_count": 27,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "traning_data_set.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "metadata": {
    "colab": {},
    "colab_type": "code",
    "id": "ZUs4c3Rw-U60"
   },
   "outputs": [],
   "source": [
    "# Let us split the data in training and testing set. The test_size parameter allows you to control the ratio of \n",
    "# testing set you want to randomly capture from your data set. \n",
    "# With the random_state parameter you can guarantee that the output of Run 1 will be equal to the output of Run 2, \n",
    "# i.e. your split will be always the same.\n",
    "\n",
    "X_train, X_test, y_train, y_test = train_test_split(traning_data_set, df['Survived'], test_size=0.20, random_state=42)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 34
    },
    "colab_type": "code",
    "id": "velHrfoUAKhL",
    "outputId": "4416a7c9-a4b2-4d7c-8537-524eae80e938"
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "((712, 8), (179, 8), (712,), (179,))"
      ]
     },
     "execution_count": 29,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Let us look at the shape our training and testing splits\n",
    "X_train.shape, X_test.shape, y_train.shape, y_test.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "metadata": {
    "colab": {},
    "colab_type": "code",
    "id": "8_6lV7EQAZHz"
   },
   "outputs": [],
   "source": [
    "# We will be using Logistic Regression for our purposes.\n",
    "# If you look at the documentation, https://scikit-learn.org/stable/modules/generated/sklearn.linear_model.LogisticRegression.html\n",
    "# There are several additional parameters you can pass. For our purposes we will be using the defaults.\n",
    "\n",
    "model = LogisticRegression()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 103
    },
    "colab_type": "code",
    "id": "ZpWb4XdmBwIt",
    "outputId": "3b97233d-0640-4154-c6cc-e5b4d447ed63"
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "LogisticRegression(C=1.0, class_weight=None, dual=False, fit_intercept=True,\n",
       "                   intercept_scaling=1, l1_ratio=None, max_iter=100,\n",
       "                   multi_class='auto', n_jobs=None, penalty='l2',\n",
       "                   random_state=None, solver='lbfgs', tol=0.0001, verbose=0,\n",
       "                   warm_start=False)"
      ]
     },
     "execution_count": 31,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "model"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 158
    },
    "colab_type": "code",
    "id": "anixVhJ0Bx4m",
    "outputId": "3c393fe6-054e-476d-84c8-6e8a8862a254"
   },
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/home/rosebay/Documents/adarsh/tensorflow-practice/.venv/lib/python3.6/site-packages/sklearn/linear_model/_logistic.py:939: ConvergenceWarning: lbfgs failed to converge (status=1):\n",
      "STOP: TOTAL NO. of ITERATIONS REACHED LIMIT.\n",
      "\n",
      "Increase the number of iterations (max_iter) or scale the data as shown in:\n",
      "    https://scikit-learn.org/stable/modules/preprocessing.html.\n",
      "Please also refer to the documentation for alternative solver options:\n",
      "    https://scikit-learn.org/stable/modules/linear_model.html#logistic-regression\n",
      "  extra_warning_msg=_LOGISTIC_SOLVER_CONVERGENCE_MSG)\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "LogisticRegression(C=1.0, class_weight=None, dual=False, fit_intercept=True,\n",
       "                   intercept_scaling=1, l1_ratio=None, max_iter=100,\n",
       "                   multi_class='auto', n_jobs=None, penalty='l2',\n",
       "                   random_state=None, solver='lbfgs', tol=0.0001, verbose=0,\n",
       "                   warm_start=False)"
      ]
     },
     "execution_count": 32,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Let us fit our training data\n",
    "# This is where the learning is happening\n",
    "model.fit(X_train, y_train)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 33,
   "metadata": {
    "colab": {},
    "colab_type": "code",
    "id": "NWpwb5QMBxdx"
   },
   "outputs": [],
   "source": [
    "predictions = model.predict(X_test)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 34,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 173
    },
    "colab_type": "code",
    "id": "DIGyYEVsCDjS",
    "outputId": "bdf9da3b-7163-4f73-8e40-af544130fde5"
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([0, 0, 0, 1, 1, 1, 1, 0, 1, 1, 0, 0, 0, 0, 0, 1, 0, 1, 0, 0, 0, 0,\n",
       "       1, 0, 0, 0, 0, 0, 0, 1, 0, 1, 0, 1, 0, 0, 0, 1, 1, 0, 0, 0, 0, 0,\n",
       "       1, 0, 0, 0, 0, 0, 1, 1, 0, 1, 0, 1, 0, 1, 1, 1, 0, 1, 1, 0, 0, 1,\n",
       "       0, 0, 0, 1, 1, 1, 1, 1, 0, 0, 1, 1, 1, 0, 0, 1, 1, 0, 0, 0, 1, 1,\n",
       "       0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 0, 0, 0, 1, 0, 0, 0, 1, 0, 0, 0,\n",
       "       1, 0, 1, 0, 1, 0, 0, 0, 1, 0, 0, 1, 1, 0, 0, 0, 1, 1, 1, 0, 1, 0,\n",
       "       0, 1, 0, 1, 1, 0, 0, 1, 0, 1, 0, 0, 0, 1, 0, 0, 1, 0, 0, 0, 0, 1,\n",
       "       0, 0, 0, 1, 1, 1, 0, 0, 0, 1, 0, 0, 0, 1, 0, 0, 1, 1, 0, 1, 0, 0,\n",
       "       0, 1, 1])"
      ]
     },
     "execution_count": 34,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# What did we predict?\n",
    "predictions"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 35,
   "metadata": {
    "colab": {},
    "colab_type": "code",
    "id": "NGpkTSobAj2b"
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "709    1\n",
       "439    0\n",
       "840    0\n",
       "720    1\n",
       "39     1\n",
       "      ..\n",
       "433    0\n",
       "773    0\n",
       "25     1\n",
       "84     1\n",
       "10     1\n",
       "Name: Survived, Length: 179, dtype: int64"
      ]
     },
     "execution_count": 35,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# What did our test dataset look like?\n",
    "y_test"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 36,
   "metadata": {
    "colab": {},
    "colab_type": "code",
    "id": "LIlp_KLNASj3"
   },
   "outputs": [],
   "source": [
    "df_compare = X_test.reset_index(drop=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 37,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 202
    },
    "colab_type": "code",
    "id": "xITI44wyOEzj",
    "outputId": "1079056d-1442-4694-c4b0-55ded438aaf3"
   },
   "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>Age</th>\n",
       "      <th>SibSp</th>\n",
       "      <th>Parch</th>\n",
       "      <th>Fare</th>\n",
       "      <th>male</th>\n",
       "      <th>Q</th>\n",
       "      <th>S</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>3</td>\n",
       "      <td>24.0</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>15.2458</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>2</td>\n",
       "      <td>31.0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>10.5000</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>3</td>\n",
       "      <td>20.0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>7.9250</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>2</td>\n",
       "      <td>6.0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>33.0000</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>3</td>\n",
       "      <td>14.0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>11.2417</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   Pclass   Age  SibSp  Parch     Fare  male  Q  S\n",
       "0       3  24.0      1      1  15.2458     1  0  0\n",
       "1       2  31.0      0      0  10.5000     1  0  1\n",
       "2       3  20.0      0      0   7.9250     1  0  1\n",
       "3       2   6.0      0      1  33.0000     0  0  1\n",
       "4       3  14.0      1      0  11.2417     0  0  0"
      ]
     },
     "execution_count": 37,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_compare.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 38,
   "metadata": {
    "colab": {},
    "colab_type": "code",
    "id": "4AvyL3elOIXS"
   },
   "outputs": [],
   "source": [
    "#Adding columns 'Survived' and 'Predicted' to df_compare.\n",
    "df_compare['Survived'] = y_test.values"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 39,
   "metadata": {
    "colab": {},
    "colab_type": "code",
    "id": "wLjh6NuyORIu"
   },
   "outputs": [],
   "source": [
    "df_compare['Predicted'] = predictions"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 40,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 202
    },
    "colab_type": "code",
    "id": "kYcz9HmYOX3u",
    "outputId": "de032910-6a79-48f3-accf-fcae72336594"
   },
   "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>Age</th>\n",
       "      <th>SibSp</th>\n",
       "      <th>Parch</th>\n",
       "      <th>Fare</th>\n",
       "      <th>male</th>\n",
       "      <th>Q</th>\n",
       "      <th>S</th>\n",
       "      <th>Survived</th>\n",
       "      <th>Predicted</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>3</td>\n",
       "      <td>24.0</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>15.2458</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>2</td>\n",
       "      <td>31.0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>10.5000</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>3</td>\n",
       "      <td>20.0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>7.9250</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>2</td>\n",
       "      <td>6.0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>33.0000</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>3</td>\n",
       "      <td>14.0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>11.2417</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   Pclass   Age  SibSp  Parch     Fare  male  Q  S  Survived  Predicted\n",
       "0       3  24.0      1      1  15.2458     1  0  0         1          0\n",
       "1       2  31.0      0      0  10.5000     1  0  1         0          0\n",
       "2       3  20.0      0      0   7.9250     1  0  1         0          0\n",
       "3       2   6.0      0      1  33.0000     0  0  1         1          1\n",
       "4       3  14.0      1      0  11.2417     0  0  0         1          1"
      ]
     },
     "execution_count": 40,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_compare.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 41,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 34
    },
    "colab_type": "code",
    "id": "GuphspQJFzQL",
    "outputId": "322e8f46-4529-4ff5-a883-d7cd027b4acc"
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(145, 10)"
      ]
     },
     "execution_count": 41,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# We can now check the accuracy of the model.\n",
    "df_compare.loc[df_compare.Survived==df_compare.Predicted].shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 42,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 34
    },
    "colab_type": "code",
    "id": "lu1T84spGLGR",
    "outputId": "36de079f-b139-4daf-befb-8e3c43dbff1b"
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.8100558659217877"
      ]
     },
     "execution_count": 42,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_compare.loc[df_compare.Survived==df_compare.Predicted].shape[0]/df_compare.shape[0]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 43,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.8100558659217877"
      ]
     },
     "execution_count": 43,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#Alternatively, you can also call the score function to calculate the accuracy of the algorithm\n",
    "model.score(X_test,y_test)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 44,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 51
    },
    "colab_type": "code",
    "id": "EoJbhnYKQmes",
    "outputId": "5decba89-71e9-4398-d258-7614fff17479"
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([[91, 14],\n",
       "       [20, 54]])"
      ]
     },
     "execution_count": 44,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "confusion_matrix(y_test, predictions)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "colab_type": "text",
    "id": "VY_oy4qmSmDH"
   },
   "source": [
    "![alt text](https://upload.wikimedia.org/wikipedia/en/a/a6/Binary_confusion_matrix.png)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "colab_type": "text",
    "id": "iShFpzbEVQ7V"
   },
   "source": [
    "![alt text](https://github.com/ayushsubedi/naivebayes_spirathon2019/raw/e43807530ec059ec26c05dc1f646cb875fab514c/types.jpg)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "colab_type": "text",
    "id": "6TibcRw-WATB"
   },
   "source": [
    "In our context,\n",
    "\n",
    "\n",
    "*   True Positive: (Good) People who really survived were predicted to survive.\n",
    "*   True Negative: (Good) People who really did not survive were predicted to not survive. \n",
    "*   False Positive: (Bad) People who really did not survive were predicted to survive.\n",
    "*   False Negative: (Bad) People who really survived were predicted to not survive. \n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "colab_type": "text",
    "id": "X54LAxE6UmSF"
   },
   "source": [
    "You can also calculate accuracy by (TP+TN)/(TP+FN+FP+TN)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "colab_type": "text",
    "id": "hbOvOiu7Ta-V"
   },
   "source": [
    "But wait, is accuracy a good indication of how good/bad a model is? "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "colab_type": "text",
    "id": "Xu85Qv6qXfo2"
   },
   "source": [
    "There are 284335 good credit card transactions and 472 fradulent transactions. Can you come up with a model that has over 99% accuracy?"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 45,
   "metadata": {
    "colab": {},
    "colab_type": "code",
    "id": "htcAZr3MYFS8"
   },
   "outputs": [],
   "source": [
    "total_transactions = 284335 + 472"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 46,
   "metadata": {
    "colab": {},
    "colab_type": "code",
    "id": "qYC9TYKYYXOt"
   },
   "outputs": [],
   "source": [
    "# Let us say we wrote a bad model which predicts everything to be not fradulent\n",
    "not_fraud = total_transactions\n",
    "correct_prediction = not_fraud - 472"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 47,
   "metadata": {
    "colab": {},
    "colab_type": "code",
    "id": "5nle2eiZYPvD"
   },
   "outputs": [],
   "source": [
    "accuracy = correct_prediction / total_transactions * 100"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 48,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 34
    },
    "colab_type": "code",
    "id": "0WdU6HzxY7VZ",
    "outputId": "0d09f33f-1e2d-4e97-c065-6ef209c66fc9"
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "99.83427373624946"
      ]
     },
     "execution_count": 48,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "accuracy"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "colab_type": "text",
    "id": "81uc4SUcY5IZ"
   },
   "source": [
    "The accuracy of the model is 99%. But what is the problem?"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "colab_type": "text",
    "id": "1RE9yn10Xj-X"
   },
   "source": [
    "Since evaluation of a model based on the accuracy metric does not always provide how good or bad of a model we have built, there are other metrics that can be used:\n",
    "  Learn more here: https://en.wikipedia.org/wiki/Precision_and_recall"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Logistic Regression Summary <a class=\"anchor\" id=\"summary\"></a>\n",
    "In this notebook, we have covered the basic concepts of a logistic regression classifier, and applied it to the titanic dataset from Kaggle to predict if an individual survived the wreck or not. "
   ]
  }
 ],
 "metadata": {
  "colab": {
   "collapsed_sections": [],
   "name": "Copy of 5. Introduction to Machine Learning Logistic Regression",
   "provenance": []
  },
  "kernelspec": {
   "display_name": "Python 3",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.6.9"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}