logistic-regression-heart-disease.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "view-in-github"
   },
   "source": [
    "<a href=\"https://colab.research.google.com/github/dphi-official/Data_Science_Bootcamp/blob/master/Week3/Logistic_Regression/Logistic_Regression_Heart_Disease.ipynb\" target=\"_parent\"><img src=\"https://colab.research.google.com/assets/colab-badge.svg\" alt=\"Open In Colab\"/></a>"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "_cell_guid": "a55f85f4-3e45-4a64-8fd4-d6c4c23c4c77",
    "_uuid": "b34438b648d12caafb20d1ff0b8242ee6093f55e",
    "id": "GYGV8A08rwxU"
   },
   "source": [
    "# Predicting Heart Disease using Logistic Regression"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "_cell_guid": "d8d1ffac-5027-420e-bbfb-cb4aef123890",
    "_uuid": "75521b34607283fa0c48a4aaf4ec9616fcd6c47f",
    "id": "vMwBJ5iRrwxX"
   },
   "source": [
    "### Introduction\n",
    "\n",
    "World Health Organization has  estimated 12 million deaths occur worldwide, every year due to Heart diseases. Half the deaths in the United States and other developed countries are due to cardio vascular diseases. The early prognosis of cardiovascular diseases can aid in making decisions on lifestyle changes in high risk patients and in turn reduce the complications. This research intends to pinpoint the most relevant/risk factors of heart disease as well as predict the overall risk using logistic regression.\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {
    "_cell_guid": "0505f675-ca00-47b6-b43c-b83cd68a8185",
    "_uuid": "a920b2bbb8694dd8e3d715fd9a333c47c5923c49",
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "NJkSF643rwxY",
    "outputId": "01b0e3c6-6f61-4ab7-8b41-5596f5c5d506"
   },
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/usr/local/lib/python3.7/dist-packages/statsmodels/tools/_testing.py:19: FutureWarning: pandas.util.testing is deprecated. Use the functions in the public API at pandas.testing instead.\n",
      "  import pandas.util.testing as tm\n"
     ]
    }
   ],
   "source": [
    "import pandas as pd\n",
    "import numpy as np\n",
    "import statsmodels.api as sm\n",
    "import scipy.stats as st\n",
    "import matplotlib.pyplot as plt\n",
    "import seaborn as sn\n",
    "from sklearn.metrics import confusion_matrix\n",
    "import matplotlib.mlab as mlab\n",
    "%matplotlib inline"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "_cell_guid": "7e94f5d9-568d-4d04-8920-7b8b3d0bcb96",
    "_uuid": "04563e20d085b8b5c7ee9ec44bc8e3135489cbdd",
    "id": "k3NmHrtxrwxe"
   },
   "source": [
    "## Data Preparation\n",
    "\n",
    "### Source:\n",
    "\n",
    "The dataset is publically available on the Kaggle website, and it is from an ongoing ongoing cardiovascular study on residents of the town of Framingham, Massachusetts.  The classification goal is to predict whether the patient has 10-year risk of future coronary heart disease (CHD).The dataset provides the patients’ information. It includes over 4,000 records and 15 attributes."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 206
    },
    "id": "PpjzzEB3v0jY",
    "outputId": "f31e2ff7-6678-44a0-ecce-780d2e3010b6"
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "\n",
       "  <div id=\"df-4f3687b6-281e-4998-9bcc-c03f5f2de5ed\">\n",
       "    <div class=\"colab-df-container\">\n",
       "      <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>male</th>\n",
       "      <th>age</th>\n",
       "      <th>currentSmoker</th>\n",
       "      <th>cigsPerDay</th>\n",
       "      <th>BPMeds</th>\n",
       "      <th>prevalentStroke</th>\n",
       "      <th>prevalentHyp</th>\n",
       "      <th>diabetes</th>\n",
       "      <th>totChol</th>\n",
       "      <th>sysBP</th>\n",
       "      <th>diaBP</th>\n",
       "      <th>BMI</th>\n",
       "      <th>heartRate</th>\n",
       "      <th>glucose</th>\n",
       "      <th>TenYearCHD</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>1</td>\n",
       "      <td>39</td>\n",
       "      <td>0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>195.0</td>\n",
       "      <td>106.0</td>\n",
       "      <td>70.0</td>\n",
       "      <td>26.97</td>\n",
       "      <td>80.0</td>\n",
       "      <td>77.0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>0</td>\n",
       "      <td>46</td>\n",
       "      <td>0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>250.0</td>\n",
       "      <td>121.0</td>\n",
       "      <td>81.0</td>\n",
       "      <td>28.73</td>\n",
       "      <td>95.0</td>\n",
       "      <td>76.0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>1</td>\n",
       "      <td>48</td>\n",
       "      <td>1</td>\n",
       "      <td>20.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>245.0</td>\n",
       "      <td>127.5</td>\n",
       "      <td>80.0</td>\n",
       "      <td>25.34</td>\n",
       "      <td>75.0</td>\n",
       "      <td>70.0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>0</td>\n",
       "      <td>61</td>\n",
       "      <td>1</td>\n",
       "      <td>30.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>225.0</td>\n",
       "      <td>150.0</td>\n",
       "      <td>95.0</td>\n",
       "      <td>28.58</td>\n",
       "      <td>65.0</td>\n",
       "      <td>103.0</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>0</td>\n",
       "      <td>46</td>\n",
       "      <td>1</td>\n",
       "      <td>23.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>285.0</td>\n",
       "      <td>130.0</td>\n",
       "      <td>84.0</td>\n",
       "      <td>23.10</td>\n",
       "      <td>85.0</td>\n",
       "      <td>85.0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>\n",
       "      <button class=\"colab-df-convert\" onclick=\"convertToInteractive('df-4f3687b6-281e-4998-9bcc-c03f5f2de5ed')\"\n",
       "              title=\"Convert this dataframe to an interactive table.\"\n",
       "              style=\"display:none;\">\n",
       "        \n",
       "  <svg xmlns=\"http://www.w3.org/2000/svg\" height=\"24px\"viewBox=\"0 0 24 24\"\n",
       "       width=\"24px\">\n",
       "    <path d=\"M0 0h24v24H0V0z\" fill=\"none\"/>\n",
       "    <path d=\"M18.56 5.44l.94 2.06.94-2.06 2.06-.94-2.06-.94-.94-2.06-.94 2.06-2.06.94zm-11 1L8.5 8.5l.94-2.06 2.06-.94-2.06-.94L8.5 2.5l-.94 2.06-2.06.94zm10 10l.94 2.06.94-2.06 2.06-.94-2.06-.94-.94-2.06-.94 2.06-2.06.94z\"/><path d=\"M17.41 7.96l-1.37-1.37c-.4-.4-.92-.59-1.43-.59-.52 0-1.04.2-1.43.59L10.3 9.45l-7.72 7.72c-.78.78-.78 2.05 0 2.83L4 21.41c.39.39.9.59 1.41.59.51 0 1.02-.2 1.41-.59l7.78-7.78 2.81-2.81c.8-.78.8-2.07 0-2.86zM5.41 20L4 18.59l7.72-7.72 1.47 1.35L5.41 20z\"/>\n",
       "  </svg>\n",
       "      </button>\n",
       "      \n",
       "  <style>\n",
       "    .colab-df-container {\n",
       "      display:flex;\n",
       "      flex-wrap:wrap;\n",
       "      gap: 12px;\n",
       "    }\n",
       "\n",
       "    .colab-df-convert {\n",
       "      background-color: #E8F0FE;\n",
       "      border: none;\n",
       "      border-radius: 50%;\n",
       "      cursor: pointer;\n",
       "      display: none;\n",
       "      fill: #1967D2;\n",
       "      height: 32px;\n",
       "      padding: 0 0 0 0;\n",
       "      width: 32px;\n",
       "    }\n",
       "\n",
       "    .colab-df-convert:hover {\n",
       "      background-color: #E2EBFA;\n",
       "      box-shadow: 0px 1px 2px rgba(60, 64, 67, 0.3), 0px 1px 3px 1px rgba(60, 64, 67, 0.15);\n",
       "      fill: #174EA6;\n",
       "    }\n",
       "\n",
       "    [theme=dark] .colab-df-convert {\n",
       "      background-color: #3B4455;\n",
       "      fill: #D2E3FC;\n",
       "    }\n",
       "\n",
       "    [theme=dark] .colab-df-convert:hover {\n",
       "      background-color: #434B5C;\n",
       "      box-shadow: 0px 1px 3px 1px rgba(0, 0, 0, 0.15);\n",
       "      filter: drop-shadow(0px 1px 2px rgba(0, 0, 0, 0.3));\n",
       "      fill: #FFFFFF;\n",
       "    }\n",
       "  </style>\n",
       "\n",
       "      <script>\n",
       "        const buttonEl =\n",
       "          document.querySelector('#df-4f3687b6-281e-4998-9bcc-c03f5f2de5ed button.colab-df-convert');\n",
       "        buttonEl.style.display =\n",
       "          google.colab.kernel.accessAllowed ? 'block' : 'none';\n",
       "\n",
       "        async function convertToInteractive(key) {\n",
       "          const element = document.querySelector('#df-4f3687b6-281e-4998-9bcc-c03f5f2de5ed');\n",
       "          const dataTable =\n",
       "            await google.colab.kernel.invokeFunction('convertToInteractive',\n",
       "                                                     [key], {});\n",
       "          if (!dataTable) return;\n",
       "\n",
       "          const docLinkHtml = 'Like what you see? Visit the ' +\n",
       "            '<a target=\"_blank\" href=https://colab.research.google.com/notebooks/data_table.ipynb>data table notebook</a>'\n",
       "            + ' to learn more about interactive tables.';\n",
       "          element.innerHTML = '';\n",
       "          dataTable['output_type'] = 'display_data';\n",
       "          await google.colab.output.renderOutput(dataTable, element);\n",
       "          const docLink = document.createElement('div');\n",
       "          docLink.innerHTML = docLinkHtml;\n",
       "          element.appendChild(docLink);\n",
       "        }\n",
       "      </script>\n",
       "    </div>\n",
       "  </div>\n",
       "  "
      ],
      "text/plain": [
       "   male  age  currentSmoker  cigsPerDay  BPMeds  prevalentStroke  \\\n",
       "0     1   39              0         0.0     0.0                0   \n",
       "1     0   46              0         0.0     0.0                0   \n",
       "2     1   48              1        20.0     0.0                0   \n",
       "3     0   61              1        30.0     0.0                0   \n",
       "4     0   46              1        23.0     0.0                0   \n",
       "\n",
       "   prevalentHyp  diabetes  totChol  sysBP  diaBP    BMI  heartRate  glucose  \\\n",
       "0             0         0    195.0  106.0   70.0  26.97       80.0     77.0   \n",
       "1             0         0    250.0  121.0   81.0  28.73       95.0     76.0   \n",
       "2             0         0    245.0  127.5   80.0  25.34       75.0     70.0   \n",
       "3             1         0    225.0  150.0   95.0  28.58       65.0    103.0   \n",
       "4             0         0    285.0  130.0   84.0  23.10       85.0     85.0   \n",
       "\n",
       "   TenYearCHD  \n",
       "0           0  \n",
       "1           0  \n",
       "2           0  \n",
       "3           1  \n",
       "4           0  "
      ]
     },
     "execution_count": 2,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "url = 'https://raw.githubusercontent.com/dphi-official/Datasets/master/framingham.csv'\n",
    "heart_df= pd.read_csv(url)\n",
    "heart_df.drop(['education'],axis=1,inplace=True)\n",
    "heart_df.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "_cell_guid": "76329cfe-8273-432f-9cd0-f25fd24e561a",
    "_uuid": "c2293e6a93a0c89d0ab0faaaa5f37b55fcaa0f17",
    "id": "8NSS_D3vrwxm"
   },
   "source": [
    "### Variables :"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "_cell_guid": "190ccf73-7d24-44d9-98b5-3198fbd13510",
    "_uuid": "920a27387872578520ad578aa8c3b1cf510a9452",
    "id": "z7rSOAkerwxn"
   },
   "source": [
    "Each attribute is a potential risk factor. There are both demographic, behavioural and medical risk factors.\n",
    "\n",
    " - Demographic:\n",
    "sex: male or female;(Nominal)\n",
    "\n",
    "    -  age: age of the patient;(Continuous - Although the recorded ages have been truncated to whole numbers, the concept of age is continuous)\n",
    "\n",
    "\n",
    " - Behavioural:\n",
    "\n",
    "    -  currentSmoker: whether or not the patient is a current smoker (Nominal)\n",
    "\n",
    "    -  cigsPerDay: the number of cigarettes that the person smoked on average in one day.(can be considered continuous as one can have any number of cigarretts, even half a cigarette.)\n",
    "\n",
    " - Medical( history):\n",
    "\n",
    "    -  BPMeds: whether or not the patient was on blood pressure medication (Nominal)\n",
    "\n",
    "    -  prevalentStroke: whether or not the patient had previously had a stroke (Nominal)\n",
    "\n",
    "    -  prevalentHyp: whether or not the patient was hypertensive (Nominal)\n",
    "\n",
    "    -  diabetes: whether or not the patient had diabetes (Nominal)\n",
    "\n",
    " - Medical(current):\n",
    "\n",
    "    -  totChol: total cholesterol level (Continuous)\n",
    "\n",
    "    -  sysBP: systolic blood pressure (Continuous)\n",
    "\n",
    "    -  diaBP: diastolic blood pressure (Continuous)\n",
    "\n",
    "    -  BMI: Body Mass Index (Continuous)\n",
    "\n",
    "    -  heartRate: heart rate (Continuous - In medical research, variables such as heart rate though in fact discrete, yet are considered continuous because of large number of possible values.)\n",
    "\n",
    "    -  glucose: glucose level (Continuous)\n",
    "\n",
    "\n",
    " - Predict variable (desired target):\n",
    "\n",
    "    -  10 year risk of coronary heart disease CHD (binary: “1”, means “Yes”, “0” means “No”)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {
    "_cell_guid": "dd2265c1-70ee-4bbc-82dd-68871fdaa042",
    "_uuid": "d894b86d4cfc1c8e0c4403ed9c844b31ac13446a",
    "id": "Ol8Bv5xFrwxo"
   },
   "outputs": [],
   "source": [
    "heart_df.rename(columns={'male':'Gender_male'},inplace=True)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "_cell_guid": "a4ba245a-9280-46ac-8f97-aca5d1fdedb6",
    "_uuid": "9d599adc546d312c08958a050fbdfbb814e3cf0d",
    "id": "iTF-U6MKrwxu"
   },
   "source": [
    "### Missing values"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {
    "_cell_guid": "6c32b561-0667-4c3d-b8b2-bb7aefc29a38",
    "_uuid": "d34f24db4b4eb3d49a3dfdaecfea442682822d37",
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "wCoR1G0Mrwxv",
    "outputId": "16baa78e-b663-440f-af4d-46fa51824fa9"
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Gender_male          0\n",
       "age                  0\n",
       "currentSmoker        0\n",
       "cigsPerDay          29\n",
       "BPMeds              53\n",
       "prevalentStroke      0\n",
       "prevalentHyp         0\n",
       "diabetes             0\n",
       "totChol             50\n",
       "sysBP                0\n",
       "diaBP                0\n",
       "BMI                 19\n",
       "heartRate            1\n",
       "glucose            388\n",
       "TenYearCHD           0\n",
       "dtype: int64"
      ]
     },
     "execution_count": 4,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "heart_df.isnull().sum()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {
    "_cell_guid": "2e1c244f-856d-482c-b4ff-ae4ec60df36b",
    "_uuid": "05fe0f931aa94963eb044e8b4ab8f3984c536670",
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "kM6UO0Rjrwxz",
    "outputId": "d68c5eab-debd-4dcb-c88b-9a8dbe071e4a"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Total number of rows with missing values is  489\n",
      "since it is only 12 percent of the entire dataset the rows with missing values are excluded.\n"
     ]
    }
   ],
   "source": [
    "count=0\n",
    "for i in heart_df.isnull().sum(axis=1):\n",
    "    if i>0:\n",
    "        count=count+1\n",
    "print('Total number of rows with missing values is ', count)\n",
    "print('since it is only',round((count/len(heart_df.index))*100), 'percent of the entire dataset the rows with missing values are excluded.')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {
    "_cell_guid": "dfb2215d-ade3-4370-9e51-ac53a78e1ab4",
    "_uuid": "6995b6b17ee8452633cd9cb256071fb57ec5a9a7",
    "id": "qpmFfWv0rwx4"
   },
   "outputs": [],
   "source": [
    "heart_df.dropna(axis=0,inplace=True)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "_cell_guid": "2d0e05b9-08fe-43df-85c8-072218145b6e",
    "_uuid": "8f59e41e841e8970a2ea90765aa273de834049b8",
    "id": "KEGL76wurwx-"
   },
   "source": [
    "## Exploratory Analysis"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {
    "_cell_guid": "c4422c44-97d7-44d9-a4ae-793e6e2da4c4",
    "_uuid": "354f82983a107141a40f8952a451123bb30e4a7a",
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 1000
    },
    "id": "7s4ySpfFrwx_",
    "outputId": "29116e6a-61d7-452d-8cf9-0d1636b7940e"
   },
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1440x1440 with 15 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "def draw_histograms(dataframe, features, rows, cols):\n",
    "    fig=plt.figure(figsize=(20,20))\n",
    "    for i, feature in enumerate(features):\n",
    "        ax=fig.add_subplot(rows,cols,i+1)\n",
    "        dataframe[feature].hist(bins=20,ax=ax,facecolor='midnightblue')\n",
    "        ax.set_title(feature+\" Distribution\",color='DarkRed')\n",
    "        \n",
    "    fig.tight_layout()  \n",
    "    plt.show()\n",
    "draw_histograms(heart_df,heart_df.columns,6,3)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {
    "_cell_guid": "721db095-34fd-46b0-a269-9b1ed00df438",
    "_uuid": "aeba17b21c54351fc541d4ab3cc85e18ba8a6d20",
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "_JSDlGBgrwyC",
    "outputId": "5e1bb312-49c0-4a84-d559-b12e5e7b2622"
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0    3179\n",
       "1     572\n",
       "Name: TenYearCHD, dtype: int64"
      ]
     },
     "execution_count": 8,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "heart_df.TenYearCHD.value_counts()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {
    "_cell_guid": "01fcff0d-31f9-4831-89de-ead65ad41158",
    "_uuid": "03a569f5ff9a039256f6584654222a5edbf457fe",
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 297
    },
    "id": "f1PlFmSArwyG",
    "outputId": "2fb34dcb-9d13-40bf-e15e-571d04766b29"
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.axes._subplots.AxesSubplot at 0x7f7b7a8fe910>"
      ]
     },
     "execution_count": 9,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAYsAAAEGCAYAAACUzrmNAAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADh0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uMy4yLjIsIGh0dHA6Ly9tYXRwbG90bGliLm9yZy+WH4yJAAASgElEQVR4nO3df6zd9X3f8eeLX8m60ALhhhLbnVnqbnK2xSSWQ9pNyo8VDFLnNCIZrGmcDM3ZBFUjdZtINQ2alinVCFGTAJo73JiqDSFNUrwKjXksXRYpCVy3LmBTxh2QYsvgG0x+NSuVyXt/nM8dJ/a993Mu+Nx7zX0+pK/O9/v+fr7f8z7I9ovvj/M9qSokSZrPKUvdgCRp+TMsJEldhoUkqcuwkCR1GRaSpK7TlrqBcTj33HNr7dq1S92GJJ1U9uzZ882qmpht3csyLNauXcvk5ORStyFJJ5Uk35hrnaehJEldhoUkqcuwkCR1GRaSpC7DQpLUZVhIkroMC0lSl2EhSeoyLCRJXS/Lb3CfCG/6N7cvdQtahvb8x/ctdQvSkvDIQpLUZVhIkroMC0lSl2EhSeoyLCRJXYaFJKnLsJAkdY0tLJK8Msl9Sf4syb4kv9bqFyT5epKpJJ9Nckarv6ItT7X1a4f29eFWfyTJJePqWZI0u3EeWTwHvL2q3gBsADYnuQj4TeDjVfWTwLPAVW38VcCzrf7xNo4k64ErgNcDm4Fbkpw6xr4lSccYW1jUwPfa4ultKuDtwB+0+k7gnW1+S1umrX9HkrT6HVX1XFU9DkwBm8bVtyTpeGO9ZpHk1CR7gcPAbuD/AN+qqqNtyAFgVZtfBTwJ0NZ/G3j1cH2WbYbfa1uSySST09PT4/g4krRijTUsqur5qtoArGZwNPB3x/he26tqY1VtnJiYGNfbSNKKtCh3Q1XVt4AvAW8Bzkoy8wDD1cDBNn8QWAPQ1v8Y8MxwfZZtJEmLYJx3Q00kOavN/w3gZ4GHGYTG5W3YVuCuNr+rLdPW/4+qqla/ot0tdQGwDrhvXH1Lko43zkeUnw/sbHcunQLcWVV/lGQ/cEeS3wD+FLitjb8N+N0kU8ARBndAUVX7ktwJ7AeOAldX1fNj7FuSdIyxhUVVPQBcOEv9MWa5m6mq/gp49xz7ugG44UT3KEkajd/gliR1GRaSpC7DQpLUZVhIkroMC0lSl2EhSeoyLCRJXYaFJKnLsJAkdRkWkqQuw0KS1GVYSJK6DAtJUpdhIUnqMiwkSV2GhSSpy7CQJHUZFpKkLsNCktRlWEiSugwLSVKXYSFJ6jIsJEldhoUkqWtsYZFkTZIvJdmfZF+SX27165McTLK3TZcNbfPhJFNJHklyyVB9c6tNJbl2XD1LkmZ32hj3fRT4lar6kyRnAnuS7G7rPl5VNw4PTrIeuAJ4PfBa4L8n+am2+mbgZ4EDwP1JdlXV/jH2LkkaMrawqKpDwKE2/90kDwOr5tlkC3BHVT0HPJ5kCtjU1k1V1WMASe5oYw0LSVoki3LNIsla4ELg6610TZIHkuxIcnarrQKeHNrsQKvNVT/2PbYlmUwyOT09fYI/gSStbGMPiySvAj4PfKiqvgPcCrwO2MDgyONjJ+J9qmp7VW2sqo0TExMnYpeSpGac1yxIcjqDoPi9qvoCQFU9PbT+t4E/aosHgTVDm69uNeapS5IWwTjvhgpwG/BwVd00VD9/aNjPAw+1+V3AFUlekeQCYB1wH3A/sC7JBUnOYHARfNe4+pYkHW+cRxY/A/wi8GCSva32q8CVSTYABTwBfBCgqvYluZPBheujwNVV9TxAkmuAe4BTgR1VtW+MfUuSjjHOu6G+AmSWVXfPs80NwA2z1O+ebztJ0nj5DW5JUpdhIUnqMiwkSV2GhSSpy7CQJHUZFpKkLsNCktRlWEiSugwLSVKXYSFJ6jIsJEldhoUkqcuwkCR1GRaSpC7DQpLUZVhIkroMC0lSl2EhSeoyLCRJXYaFJKnLsJAkdRkWkqQuw0KS1DW2sEiyJsmXkuxPsi/JL7f6OUl2J3m0vZ7d6knyiSRTSR5I8sahfW1t4x9NsnVcPUuSZjfOI4ujwK9U1XrgIuDqJOuBa4F7q2odcG9bBrgUWNembcCtMAgX4DrgzcAm4LqZgJEkLY6xhUVVHaqqP2nz3wUeBlYBW4CdbdhO4J1tfgtwew18DTgryfnAJcDuqjpSVc8Cu4HN4+pbknS8RblmkWQtcCHwdeC8qjrUVj0FnNfmVwFPDm12oNXmqkuSFsnYwyLJq4DPAx+qqu8Mr6uqAuoEvc+2JJNJJqenp0/ELiVJzVjDIsnpDILi96rqC638dDu9RHs93OoHgTVDm69utbnqP6SqtlfVxqraODExcWI/iCStcOO8GyrAbcDDVXXT0KpdwMwdTVuBu4bq72t3RV0EfLudrroHuDjJ2e3C9sWtJklaJKeNcd8/A/wi8GCSva32q8BHgTuTXAV8A3hPW3c3cBkwBXwf+ABAVR1J8uvA/W3cR6rqyBj7liQdY2xhUVVfATLH6nfMMr6Aq+fY1w5gx4nrTpK0EH6DW5LUZVhIkroMC0lSl2EhSeoyLCRJXSOFRZJ7R6lJkl6e5r11NskrgR8Bzm1fiJu5FfZH8flMkrRi9L5n8UHgQ8BrgT28EBbfAT41xr4kScvIvGFRVb8F/FaSX6qqTy5ST5KkZWakb3BX1SeT/DSwdnibqrp9TH1JkpaRkcIiye8CrwP2As+3cgGGhSStAKM+G2ojsL49v0mStMKM+j2Lh4AfH2cjkqTla9Qji3OB/UnuA56bKVbVPxlLV5KkZWXUsLh+nE1Ikpa3Ue+G+p/jbkSStHyNejfUdxnc/QRwBnA68JdV9aPjakyStHyMemRx5sx8+23tLcBF42pKkrS8LPipszXwh8AlY+hHkrQMjXoa6l1Di6cw+N7FX42lI0nSsjPq3VA/NzR/FHiCwakoSdIKMOo1iw+MuxFJ0vI16o8frU7yxSSH2/T5JKvH3ZwkaXkY9QL37wC7GPyuxWuB/9JqkqQVYNSwmKiq36mqo236NDAx3wZJdrSjkIeGatcnOZhkb5suG1r34SRTSR5JcslQfXOrTSW5doGfT5J0AowaFs8keW+SU9v0XuCZzjafBjbPUv94VW1o090ASdYDVwCvb9vcMvNewM3ApcB64Mo2VpK0iEYNi38OvAd4CjgEXA68f74NqurLwJER978FuKOqnquqx4EpYFObpqrqsar6a+AOvAtLkhbdqGHxEWBrVU1U1WsYhMevvcj3vCbJA+001dmttgp4cmjMgVabq36cJNuSTCaZnJ6efpGtSZJmM2pY/IOqenZmoaqOABe+iPe7lcEv7m1gcITysRexj1lV1faq2lhVGycm5r2cIklaoFHD4pShowCSnMPoX+j7/6rq6ap6vqp+APw2g9NMAAeBNUNDV7faXHVJ0iIa9R/8jwFfTfK5tvxu4IaFvlmS86vqUFv8eQa/wAeD23J/P8lNDG7NXQfcBwRYl+QCBiFxBfDPFvq+kqSXZtRvcN+eZBJ4eyu9q6r2z7dNks8AbwXOTXIAuA54a5INDB53/gTwwbb/fUnuBPYzeJzI1VX1fNvPNcA9wKnAjqrat6BPKEl6yUY+ldTCYd6AOGb8lbOUb5tn/A3McrTSbq+9e9T3lSSdeAt+RLkkaeUxLCRJXYaFJKnLsJAkdRkWkqQuw0KS1GVYSJK6DAtJUpdhIUnqMiwkSV2GhSSpy7CQJHUZFpKkLsNCktRlWEiSugwLSVKXYSFJ6jIsJEldhoUkqcuwkCR1GRaSpC7DQpLUZVhIkrrGFhZJdiQ5nOShodo5SXYnebS9nt3qSfKJJFNJHkjyxqFttrbxjybZOq5+JUlzG+eRxaeBzcfUrgXurap1wL1tGeBSYF2btgG3wiBcgOuANwObgOtmAkaStHjGFhZV9WXgyDHlLcDONr8TeOdQ/fYa+BpwVpLzgUuA3VV1pKqeBXZzfABJksZssa9ZnFdVh9r8U8B5bX4V8OTQuAOtNlddkrSIluwCd1UVUCdqf0m2JZlMMjk9PX2iditJYvHD4ul2eon2erjVDwJrhsatbrW56sepqu1VtbGqNk5MTJzwxiVpJVvssNgFzNzRtBW4a6j+vnZX1EXAt9vpqnuAi5Oc3S5sX9xqkqRFdNq4dpzkM8BbgXOTHGBwV9NHgTuTXAV8A3hPG343cBkwBXwf+ABAVR1J8uvA/W3cR6rq2IvmkqQxG1tYVNWVc6x6xyxjC7h6jv3sAHacwNYkSQvkN7glSV2GhSSpy7CQJHUZFpKkLsNCktRlWEiSugwLSVKXYSFJ6jIsJEldhoUkqcuwkCR1GRaSpC7DQpLUZVhIkroMC0lSl2EhSeoyLCRJXYaFJKnLsJAkdRkWkqQuw0KS1GVYSJK6DAtJUtdpS92ApIX7i4/8/aVuQcvQT/z7B8e27yU5skjyRJIHk+xNMtlq5yTZneTR9np2qyfJJ5JMJXkgyRuXomdJWsmW8jTU26pqQ1VtbMvXAvdW1Trg3rYMcCmwrk3bgFsXvVNJWuGW0zWLLcDONr8TeOdQ/fYa+BpwVpLzl6JBSVqpliosCvhvSfYk2dZq51XVoTb/FHBem18FPDm07YFW+yFJtiWZTDI5PT09rr4laUVaqgvc/7CqDiZ5DbA7yZ8Pr6yqSlIL2WFVbQe2A2zcuHFB20qS5rckRxZVdbC9Hga+CGwCnp45vdReD7fhB4E1Q5uvbjVJ0iJZ9LBI8jeTnDkzD1wMPATsAra2YVuBu9r8LuB97a6oi4BvD52ukiQtgqU4DXUe8MUkM+//+1X1X5PcD9yZ5CrgG8B72vi7gcuAKeD7wAcWv2VJWtkWPSyq6jHgDbPUnwHeMUu9gKsXoTVJ0hyW062zkqRlyrCQJHUZFpKkLsNCktRlWEiSugwLSVKXYSFJ6jIsJEldhoUkqcuwkCR1GRaSpC7DQpLUZVhIkroMC0lSl2EhSeoyLCRJXYaFJKnLsJAkdRkWkqQuw0KS1GVYSJK6DAtJUpdhIUnqMiwkSV0nTVgk2ZzkkSRTSa5d6n4kaSU5KcIiyanAzcClwHrgyiTrl7YrSVo5ToqwADYBU1X1WFX9NXAHsGWJe5KkFeO0pW5gRKuAJ4eWDwBvHh6QZBuwrS1+L8kji9TbSnAu8M2lbmI5yI1bl7oFHc8/nzOuy0vdw9+aa8XJEhZdVbUd2L7UfbwcJZmsqo1L3Yc0G/98Lo6T5TTUQWDN0PLqVpMkLYKTJSzuB9YluSDJGcAVwK4l7kmSVoyT4jRUVR1Ncg1wD3AqsKOq9i1xWyuJp/e0nPnncxGkqpa6B0nSMneynIaSJC0hw0KS1GVYaF4+ZkXLUZIdSQ4neWipe1kpDAvNycesaBn7NLB5qZtYSQwLzcfHrGhZqqovA0eWuo+VxLDQfGZ7zMqqJepF0hIyLCRJXYaF5uNjViQBhoXm52NWJAGGheZRVUeBmcesPAzc6WNWtBwk+QzwVeDvJDmQ5Kql7unlzsd9SJK6PLKQJHUZFpKkLsNCktRlWEiSugwLSVKXYaEVK8mrk+xt01NJDg4tnzHC9q9J8kSSHx+q3Zzkwyegt01Jvtye+PunSf5zkh9J8v4knzpm7B8n2djmn0jyYJv2J/mNJK98qf1IJ8XPqkrjUFXPABsAklwPfK+qblzA9oeTfBS4EXhvkjcC/wh404vtKclpwKuBzwFXVNVXW/1y4MwRd/O2qvpmklcx+MnR/wRsfbE9SWBYSD8kyZuAm4BXAd8E3l9Vh5L8MfB14G3AWcBVVfW/GPxjvDXJ24D/wOBLjD+R5GZgAvg+8C+q6s+T/Bzw74AzgGeAX6iqp1tQvQ7428BfAI8CO2eCAqCq/qD1N/JnqarvJfmXwJNJzqkqn9KqF83TUNILAnwSuLyq3gTsAG4YWn9aVW0CPgRcB1BVPwD+FfB54JH26OztwC+1ffxr4Ja2/VeAi6rqQgaPe/+3Q/teD/zjqroS+HvAnnn6/KdDp8v2AhvnGlhV3wEeB9aN8h9AmotHFtILXsHgH+rd7f/gTwUODa3/QnvdA6ydKVbV3vaLbbe0Uz8/DXxu6CjgFe11NfDZJOczOLp4fGjfu6rq/47Y52er6pqZhXbUM5/RD0ekORgW0gsC7Kuqt8yx/rn2+jzH/935QZtOAb5VVRtm2f6TwE1VtSvJW4Hrh9b95dD8PgbXPe5aUPezSHImg2D73y91X1rZPA0lveA5YCLJWwCSnJ7k9QvZwcxpnyTvbvtIkje01T/GC494n++C86cYXAd580whybuSnLeQXtpRzi3AH1bVswvZVjqWYSG94AfA5cBvJvkzYC+DU0oL9QvAVW0f+3jhp2ivZ3B6ag+Di+ezqqqnGTwO/sZ26+zDwCXAd0d8/y+102L3Mbhg/sEX8RmkH+JTZyVJXR5ZSJK6DAtJUpdhIUnqMiwkSV2GhSSpy7CQJHUZFpKkrv8H6sHy05C/EdQAAAAASUVORK5CYII=\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "sn.countplot(x='TenYearCHD',data=heart_df)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "_cell_guid": "955ded98-f1bd-4756-badb-70ada8067f2f",
    "_uuid": "bbc3663c10718e4b1e85d6be97c4c2bf2fa098fd",
    "id": "dRb1BWtWrwyN"
   },
   "source": [
    "There are 3179 patents with no heart disease and 572 patients with risk of heart disease."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {
    "_cell_guid": "1706a968-b8e0-4393-b9bd-72a8761ee3c0",
    "_uuid": "f3913969f492436b291f154646e0584be9c64ec4",
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 364
    },
    "id": "U8NFhHrxrwyS",
    "outputId": "9417511a-825d-4ee2-bcff-0f6d781fb0f8"
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "\n",
       "  <div id=\"df-e250be52-e44d-478c-8ac3-991b7d6db2eb\">\n",
       "    <div class=\"colab-df-container\">\n",
       "      <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>Gender_male</th>\n",
       "      <th>age</th>\n",
       "      <th>currentSmoker</th>\n",
       "      <th>cigsPerDay</th>\n",
       "      <th>BPMeds</th>\n",
       "      <th>prevalentStroke</th>\n",
       "      <th>prevalentHyp</th>\n",
       "      <th>diabetes</th>\n",
       "      <th>totChol</th>\n",
       "      <th>sysBP</th>\n",
       "      <th>diaBP</th>\n",
       "      <th>BMI</th>\n",
       "      <th>heartRate</th>\n",
       "      <th>glucose</th>\n",
       "      <th>TenYearCHD</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>count</th>\n",
       "      <td>3751.000000</td>\n",
       "      <td>3751.000000</td>\n",
       "      <td>3751.000000</td>\n",
       "      <td>3751.000000</td>\n",
       "      <td>3751.000000</td>\n",
       "      <td>3751.000000</td>\n",
       "      <td>3751.000000</td>\n",
       "      <td>3751.000000</td>\n",
       "      <td>3751.000000</td>\n",
       "      <td>3751.000000</td>\n",
       "      <td>3751.000000</td>\n",
       "      <td>3751.000000</td>\n",
       "      <td>3751.000000</td>\n",
       "      <td>3751.000000</td>\n",
       "      <td>3751.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>mean</th>\n",
       "      <td>0.445215</td>\n",
       "      <td>49.573447</td>\n",
       "      <td>0.488403</td>\n",
       "      <td>9.008531</td>\n",
       "      <td>0.030392</td>\n",
       "      <td>0.005599</td>\n",
       "      <td>0.311917</td>\n",
       "      <td>0.027193</td>\n",
       "      <td>236.928019</td>\n",
       "      <td>132.368435</td>\n",
       "      <td>82.938550</td>\n",
       "      <td>25.808288</td>\n",
       "      <td>75.704079</td>\n",
       "      <td>81.880032</td>\n",
       "      <td>0.152493</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>std</th>\n",
       "      <td>0.497056</td>\n",
       "      <td>8.570204</td>\n",
       "      <td>0.499932</td>\n",
       "      <td>11.925097</td>\n",
       "      <td>0.171686</td>\n",
       "      <td>0.074623</td>\n",
       "      <td>0.463338</td>\n",
       "      <td>0.162666</td>\n",
       "      <td>44.611594</td>\n",
       "      <td>22.046522</td>\n",
       "      <td>11.932779</td>\n",
       "      <td>4.065599</td>\n",
       "      <td>11.956382</td>\n",
       "      <td>23.882233</td>\n",
       "      <td>0.359546</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>min</th>\n",
       "      <td>0.000000</td>\n",
       "      <td>32.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>113.000000</td>\n",
       "      <td>83.500000</td>\n",
       "      <td>48.000000</td>\n",
       "      <td>15.540000</td>\n",
       "      <td>44.000000</td>\n",
       "      <td>40.000000</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>25%</th>\n",
       "      <td>0.000000</td>\n",
       "      <td>42.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>206.000000</td>\n",
       "      <td>117.000000</td>\n",
       "      <td>75.000000</td>\n",
       "      <td>23.085000</td>\n",
       "      <td>68.000000</td>\n",
       "      <td>71.000000</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>50%</th>\n",
       "      <td>0.000000</td>\n",
       "      <td>49.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>234.000000</td>\n",
       "      <td>128.000000</td>\n",
       "      <td>82.000000</td>\n",
       "      <td>25.410000</td>\n",
       "      <td>75.000000</td>\n",
       "      <td>78.000000</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>75%</th>\n",
       "      <td>1.000000</td>\n",
       "      <td>56.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>20.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>264.000000</td>\n",
       "      <td>144.000000</td>\n",
       "      <td>90.000000</td>\n",
       "      <td>28.060000</td>\n",
       "      <td>82.000000</td>\n",
       "      <td>87.000000</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>max</th>\n",
       "      <td>1.000000</td>\n",
       "      <td>70.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>70.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>696.000000</td>\n",
       "      <td>295.000000</td>\n",
       "      <td>142.500000</td>\n",
       "      <td>56.800000</td>\n",
       "      <td>143.000000</td>\n",
       "      <td>394.000000</td>\n",
       "      <td>1.000000</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>\n",
       "      <button class=\"colab-df-convert\" onclick=\"convertToInteractive('df-e250be52-e44d-478c-8ac3-991b7d6db2eb')\"\n",
       "              title=\"Convert this dataframe to an interactive table.\"\n",
       "              style=\"display:none;\">\n",
       "        \n",
       "  <svg xmlns=\"http://www.w3.org/2000/svg\" height=\"24px\"viewBox=\"0 0 24 24\"\n",
       "       width=\"24px\">\n",
       "    <path d=\"M0 0h24v24H0V0z\" fill=\"none\"/>\n",
       "    <path d=\"M18.56 5.44l.94 2.06.94-2.06 2.06-.94-2.06-.94-.94-2.06-.94 2.06-2.06.94zm-11 1L8.5 8.5l.94-2.06 2.06-.94-2.06-.94L8.5 2.5l-.94 2.06-2.06.94zm10 10l.94 2.06.94-2.06 2.06-.94-2.06-.94-.94-2.06-.94 2.06-2.06.94z\"/><path d=\"M17.41 7.96l-1.37-1.37c-.4-.4-.92-.59-1.43-.59-.52 0-1.04.2-1.43.59L10.3 9.45l-7.72 7.72c-.78.78-.78 2.05 0 2.83L4 21.41c.39.39.9.59 1.41.59.51 0 1.02-.2 1.41-.59l7.78-7.78 2.81-2.81c.8-.78.8-2.07 0-2.86zM5.41 20L4 18.59l7.72-7.72 1.47 1.35L5.41 20z\"/>\n",
       "  </svg>\n",
       "      </button>\n",
       "      \n",
       "  <style>\n",
       "    .colab-df-container {\n",
       "      display:flex;\n",
       "      flex-wrap:wrap;\n",
       "      gap: 12px;\n",
       "    }\n",
       "\n",
       "    .colab-df-convert {\n",
       "      background-color: #E8F0FE;\n",
       "      border: none;\n",
       "      border-radius: 50%;\n",
       "      cursor: pointer;\n",
       "      display: none;\n",
       "      fill: #1967D2;\n",
       "      height: 32px;\n",
       "      padding: 0 0 0 0;\n",
       "      width: 32px;\n",
       "    }\n",
       "\n",
       "    .colab-df-convert:hover {\n",
       "      background-color: #E2EBFA;\n",
       "      box-shadow: 0px 1px 2px rgba(60, 64, 67, 0.3), 0px 1px 3px 1px rgba(60, 64, 67, 0.15);\n",
       "      fill: #174EA6;\n",
       "    }\n",
       "\n",
       "    [theme=dark] .colab-df-convert {\n",
       "      background-color: #3B4455;\n",
       "      fill: #D2E3FC;\n",
       "    }\n",
       "\n",
       "    [theme=dark] .colab-df-convert:hover {\n",
       "      background-color: #434B5C;\n",
       "      box-shadow: 0px 1px 3px 1px rgba(0, 0, 0, 0.15);\n",
       "      filter: drop-shadow(0px 1px 2px rgba(0, 0, 0, 0.3));\n",
       "      fill: #FFFFFF;\n",
       "    }\n",
       "  </style>\n",
       "\n",
       "      <script>\n",
       "        const buttonEl =\n",
       "          document.querySelector('#df-e250be52-e44d-478c-8ac3-991b7d6db2eb button.colab-df-convert');\n",
       "        buttonEl.style.display =\n",
       "          google.colab.kernel.accessAllowed ? 'block' : 'none';\n",
       "\n",
       "        async function convertToInteractive(key) {\n",
       "          const element = document.querySelector('#df-e250be52-e44d-478c-8ac3-991b7d6db2eb');\n",
       "          const dataTable =\n",
       "            await google.colab.kernel.invokeFunction('convertToInteractive',\n",
       "                                                     [key], {});\n",
       "          if (!dataTable) return;\n",
       "\n",
       "          const docLinkHtml = 'Like what you see? Visit the ' +\n",
       "            '<a target=\"_blank\" href=https://colab.research.google.com/notebooks/data_table.ipynb>data table notebook</a>'\n",
       "            + ' to learn more about interactive tables.';\n",
       "          element.innerHTML = '';\n",
       "          dataTable['output_type'] = 'display_data';\n",
       "          await google.colab.output.renderOutput(dataTable, element);\n",
       "          const docLink = document.createElement('div');\n",
       "          docLink.innerHTML = docLinkHtml;\n",
       "          element.appendChild(docLink);\n",
       "        }\n",
       "      </script>\n",
       "    </div>\n",
       "  </div>\n",
       "  "
      ],
      "text/plain": [
       "       Gender_male          age  currentSmoker   cigsPerDay       BPMeds  \\\n",
       "count  3751.000000  3751.000000    3751.000000  3751.000000  3751.000000   \n",
       "mean      0.445215    49.573447       0.488403     9.008531     0.030392   \n",
       "std       0.497056     8.570204       0.499932    11.925097     0.171686   \n",
       "min       0.000000    32.000000       0.000000     0.000000     0.000000   \n",
       "25%       0.000000    42.000000       0.000000     0.000000     0.000000   \n",
       "50%       0.000000    49.000000       0.000000     0.000000     0.000000   \n",
       "75%       1.000000    56.000000       1.000000    20.000000     0.000000   \n",
       "max       1.000000    70.000000       1.000000    70.000000     1.000000   \n",
       "\n",
       "       prevalentStroke  prevalentHyp     diabetes      totChol        sysBP  \\\n",
       "count      3751.000000   3751.000000  3751.000000  3751.000000  3751.000000   \n",
       "mean          0.005599      0.311917     0.027193   236.928019   132.368435   \n",
       "std           0.074623      0.463338     0.162666    44.611594    22.046522   \n",
       "min           0.000000      0.000000     0.000000   113.000000    83.500000   \n",
       "25%           0.000000      0.000000     0.000000   206.000000   117.000000   \n",
       "50%           0.000000      0.000000     0.000000   234.000000   128.000000   \n",
       "75%           0.000000      1.000000     0.000000   264.000000   144.000000   \n",
       "max           1.000000      1.000000     1.000000   696.000000   295.000000   \n",
       "\n",
       "             diaBP          BMI    heartRate      glucose   TenYearCHD  \n",
       "count  3751.000000  3751.000000  3751.000000  3751.000000  3751.000000  \n",
       "mean     82.938550    25.808288    75.704079    81.880032     0.152493  \n",
       "std      11.932779     4.065599    11.956382    23.882233     0.359546  \n",
       "min      48.000000    15.540000    44.000000    40.000000     0.000000  \n",
       "25%      75.000000    23.085000    68.000000    71.000000     0.000000  \n",
       "50%      82.000000    25.410000    75.000000    78.000000     0.000000  \n",
       "75%      90.000000    28.060000    82.000000    87.000000     0.000000  \n",
       "max     142.500000    56.800000   143.000000   394.000000     1.000000  "
      ]
     },
     "execution_count": 10,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "heart_df.describe()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "_cell_guid": "d791f658-cfbb-496b-93d1-b2d0ef639b7e",
    "_uuid": "a14fae661d7902e8c58a9b57b6f6873b6aeb0a79",
    "id": "AnamVECbrwyV"
   },
   "source": [
    "## Logistic Regression"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "_cell_guid": "17d7af59-bd19-4448-b26c-f2c52b6c5381",
    "_uuid": "2244062f695fd7f76c0b86af5b49d6a597679c4d",
    "id": "4iP2WCHhrwyW"
   },
   "source": [
    "Logistic regression is a type of regression analysis in statistics used for prediction of outcome of a categorical dependent variable from a set of predictor or independent variables. In logistic regression the dependent variable is always binary. Logistic regression is mainly used to for prediction and also calculating the probability of success. \n",
    "\n",
    "## Logistic Regression is one of the basic and popular algorithms to solve a binary classification problems\n",
    "\n",
    "For each input, logistic regression outputs a probability that this input belongs to the 2 classes\n",
    "\n",
    "Set a probability threshold boundary and that determines which class the input belongs to\n",
    "\n",
    "Binary classification problems (2 classes)\n",
    "\n",
    "Emails (Spam / Not Spam)\n",
    "\n",
    "Transactions (Fraudulent / Not Fraudulent)\n",
    "\n",
    "Loan Default (Yes / No)\n",
    "\n",
    "# Why Not Use Linear Regression for Classification?\n",
    "\n",
    "Linear regression is used to solve regression problems with continuous values\n",
    "\n",
    "Logistic regression is used to solve classification problems with discrete categories\n",
    "\n",
    "Binary classification (Classes 0 and 1)\n",
    "\n",
    " The difference between Linear Regression and Logistic Regression can be represented diagrammatically as follows-\n",
    "   \n",
    "   \n",
    "   ![Linear Regression Vs Logistic Regression](https://external-content.duckduckgo.com/iu/?u=http%3A%2F%2Fres.cloudinary.com%2Fdyd911kmh%2Fimage%2Fupload%2Ff_auto%2Cq_auto%3Abest%2Fv1534281070%2Flinear_vs_logistic_regression_edxw03.png&f=1&nofb=1)\n",
    "   \n",
    "\n",
    "## Implement linear equation\n",
    "\n",
    "\n",
    "Logistic Regression algorithm works by implementing a linear equation with independent or explanatory variables to predict a response value. For example, we consider the example of number of hours studied and probability of passing the exam. Here, number of hours studied is the explanatory variable and it is denoted by x1. Probability of passing the exam is the response or target variable and it is denoted by z. \n",
    "\n",
    "\n",
    "If we have one explanatory variable (x1) and one response variable (z), then the linear equation would be given mathematically with the following equation- \n",
    "\n",
    "\n",
    " \tz = β0 + β1x1\n",
    "\n",
    "\n",
    "Here, the coefficients β0 and β1 are the parameters of the model.\n",
    "\n",
    "\n",
    "If there are multiple explanatory variables, then the above equation can be extended to \n",
    "\n",
    "\n",
    "\tz = β0 + β1x1+ β2x2+……..+ βnxn\n",
    "\n",
    "\n",
    "Here, the coefficients β0, β1, β2 and βn are the parameters of the model.\n",
    "\n",
    "\n",
    "So, the predicted response value is given by the above equations and is denoted by z.\n",
    "\n",
    " \n",
    "# Sigmoid Function\n",
    "\n",
    "\n",
    "This predicted response value, denoted by z is then converted into a probability value that lie between 0 and 1. We use the **sigmoid function** in order to map predicted values to probability values. This sigmoid function then maps any real value into a probability value between 0 and 1. \n",
    "\n",
    "\n",
    "In machine learning, sigmoid function is used to map predictions to probabilities. The sigmoid function has an `S` shaped curve. It is also called **sigmoid curve**.  \n",
    "\n",
    "\n",
    "A `Sigmoid function` is a special case of the `Logistic function`. It is given by the following mathematical formula. \n",
    "\n",
    "The logisitc also called as logit is denoted by σ is a sigmoid function and it outputs a number between 0 and 1.\n",
    "\n",
    "![](https://imgur.com/DNpFtRM.png)\n",
    "\n",
    "*source* : https://www.oreilly.com/library/view/hands-on-machine-learning/9781491962282/\n",
    "\n",
    "\n",
    "* Positive values are predictive of class 1\n",
    "* Negative values are predictive of class 0\n",
    "\n",
    "# Types of Logistic Regression\n",
    "\n",
    "\n",
    "Logistic Regression model can be classified into three groups based on the target variable categories. These three groups are described below:-\n",
    "\n",
    "\n",
    "1.\t**Binary Logistic Regression**\n",
    "\n",
    "\n",
    "In **Binary Logistic Regression**, the target variable has two possible categories. The common examples of categories are `yes or no`, `good or bad`, `true or false`, `spam or no spam` and `pass or fail`.\n",
    "\n",
    "\n",
    "2.\t**Multinomial Logistic Regression**\n",
    "\n",
    "\n",
    "In **Multinomial Logistic Regression**, the target variable has three or more categories which are not in any particular order. So, there are three or more nominal categories. The examples include the type of categories of fruits - `apple`, `mango`, `orange` and `banana`.\n",
    "\n",
    "\n",
    "3.\t**Ordinal Logistic Regression**\n",
    "\n",
    "\n",
    "In **Ordinal Logistic Regression**, the target variable has three or more ordinal categories. So, there is intrinsic order involved with the categories. For example, the student performance can be categorized as `poor`, `average`, `good` and `excellent`.\n",
    "\n",
    "\n",
    "# Decision boundary\n",
    "\n",
    "\n",
    "The sigmoid function returns a probability value between 0 and 1. This probability value is then mapped to a discrete class which is either “0” or “1”. In order to map this probability value to a discrete class (pass/fail, yes/no, true/false), we select a threshold value. This threshold value is called **Decision boundary**. Above this threshold value, we will map the probability values into class 1 and below which we will map values into class 0.\n",
    "\n",
    "\n",
    "Mathematically, it can be expressed as follows:-\n",
    "\n",
    "\n",
    "\tp ≥ 0.5 => class = 1\n",
    "\t\n",
    "\t\n",
    "\tp < 0.5 => class = 0 \n",
    "\t\n",
    "\n",
    "Generally, the decision boundary is set to 0.5. So, if the probability value is 0.8 (> 0.5), we will map this observation to class 1.  Similarly, if the probability value is 0.2 (< 0.5), we will map this observation to class 0.\n",
    "\n",
    "\n",
    "![Decision boundary](https://external-content.duckduckgo.com/iu/?u=http%3A%2F%2Fwww.holehouse.org%2Fmlclass%2F06_Logistic_Regression_files%2FImage%2520%5B5%5D.png&f=1&nofb=1)\n",
    "\n",
    "\n",
    "# Making predictions\n",
    "\n",
    "\n",
    "Now, we know about `sigmoid function` and `decision boundary` in logistic regression. We can use our knowledge of `sigmoid function` \n",
    "and `decision boundary` to write a prediction function. A prediction function in logistic regression returns the probability of the observation being positive, `Yes` or `True`. We call this as `class 1` and it is denoted by `P(class = 1)`. If the probability inches closer to one, then we will be more confident about our model that the observation is in class 1.\n",
    "\n",
    "\n",
    "In the previous example, suppose the sigmoid function returns the probability value of 0.4. It means that there is only 40% chance of passing the exam. If the decision boundary is 0.5, then we predict this observation as `fail`.\n",
    "\n",
    "\n",
    "\n",
    "\n",
    "\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "uhvfhOqG8vGg"
   },
   "source": [
    "Logistic Regression an extension of Linear regression where the dependent variable is categorical and not continuous. It predicts the probability of the outcome variable.\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "9HGsIMgV80Eb"
   },
   "source": [
    "# Representation of Logistic regression\n",
    "\n",
    "Just like a linear regression model, a logistic regression model also computes a weighted sum of input features(and adds a bias term to it). However, unlike linear regression, it calculates the logistic of the results so that the output is always between o and 1.\n",
    "\n",
    "\n",
    "\n",
    "The output of a Logistic regression model is a probability. We can select a threshold value. If the probability is greater than this threshold value, the event is predicted to happen otherwise it is predicted not to happen.\n",
    "\n",
    "Once the mpdel estimates the probabilities(p hat), it can then easily make the predictions as follows:\n",
    "\n",
    "![](https://imgur.com/cm53tL8.png)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "_cell_guid": "a6a3a722-08a7-48c0-96fa-0144c9013a48",
    "_uuid": "1a9aa9a7484a8a1e402041f3dda067e5e5dbebce",
    "collapsed": true,
    "id": "tniNy9nYrwyv"
   },
   "source": [
    "### Splitting data to train and test split"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {
    "_cell_guid": "1f21df9a-2ede-495c-a485-a992727663af",
    "_uuid": "8ff55a6b66ba18a19f84efd73104b8d3836247d4",
    "id": "KxCqRfG8rwyw"
   },
   "outputs": [],
   "source": [
    "import sklearn\n",
    "new_features=heart_df[['age','Gender_male','cigsPerDay','totChol','sysBP','glucose','TenYearCHD']]\n",
    "x=new_features.iloc[:,:-1]\n",
    "y=new_features.iloc[:,-1]\n",
    "from sklearn.model_selection import train_test_split\n",
    "x_train,x_test,y_train,y_test=train_test_split(x,y,test_size=.20,random_state=5)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {
    "_cell_guid": "6c5e51fd-3e41-4ca6-ba5e-71f36434dcf0",
    "_uuid": "2b9b52e6c72728b4bfff82e6c483f3430d88f6c9",
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "cA-PX9qJrwyz",
    "outputId": "2f0626a3-5de6-4314-c5a2-8170a6237ece"
   },
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/usr/local/lib/python3.7/dist-packages/sklearn/linear_model/_logistic.py:818: 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"
     ]
    }
   ],
   "source": [
    "from sklearn.linear_model import LogisticRegression\n",
    "logreg=LogisticRegression()\n",
    "logreg.fit(x_train,y_train)\n",
    "y_pred=logreg.predict(x_test)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "_cell_guid": "ffc0c6b7-4be0-46bd-8984-eb982eba7e79",
    "_uuid": "c8f5c46cf02412b0f6b6ce0d5fdab68f75f8c4ba",
    "id": "cTA8Jnbirwy3"
   },
   "source": [
    "## Model Evaluation\n",
    "\n",
    "### Model accuracy"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {
    "_cell_guid": "ef135c4f-33c4-41a9-809d-6b352dc0397c",
    "_uuid": "b5deedf9531410f34bb01e717ab90d883233c6e3",
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "EryuwI5Crwy3",
    "outputId": "37d0235c-5e78-432f-cdcb-5a653ef98d68"
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.8748335552596538"
      ]
     },
     "execution_count": 13,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "sklearn.metrics.accuracy_score(y_test,y_pred)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "_cell_guid": "170ecc2a-080f-4d54-8a91-357f2167f83b",
    "_uuid": "7db798cfbb6b481419ad1086bcbcdb20c8465640",
    "id": "NbKAbxIlrwy7"
   },
   "source": [
    "####  Accuracy of the model is 0.88"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "4dzoB2lh9BNz"
   },
   "source": [
    "# Describing the Performance of a Logistic model\n",
    "\n",
    "![](https://media.licdn.com/dms/image/C5112AQHRd51OS1-Oaw/article-cover_image-shrink_720_1280/0?e=1575504000&v=beta&t=kzm0mOJoLPQeShqeFN6jgAQwblDum8S1Nc5I2FiPTBk)\n",
    "\n",
    "[source](https://medium.com/google-design/human-centered-machine-learning-a770d10562cd)\n",
    "\n",
    "A **confusion matrix** is a table that is often used to describe the performance of a classification model (or “classifier”) on a set of test data for which the true values are known.Let us look at some of the important terms of confusion matrix.\n",
    "\n",
    "   ![](https://miro.medium.com/max/386/1*GMlSubndVt3g7FmeQjpeMA.png)\n",
    "\n",
    "  confusion matrix whether employees will leave a company or not\n",
    "\n",
    "### The Confusion Matrix tells us the following:\n",
    "\n",
    "* There are two possible predicted classes: “yes” and “no”. If we were predicting that employees would leave an organisation, for example, “yes” would mean they will, and “no” would mean they won’t leave the organisation.\n",
    "* The classifier made a total of 165 predictions (e.g., 165 employees were being studied).\n",
    "* Out of those 165 cases, the classifier predicted “yes” 110 times, and “no” 55 times.\n",
    "* In reality, 105 employees in the sample leave the organisation, and 60 do not.\n",
    "\n",
    "\n",
    "### Basic terms related to Confusion matrix:\n",
    "\n",
    "* **True positives (TP)**: These are cases in which we predicted yes (employees will leave the organisation), and employees actually leave i.e 100\n",
    "* **True negatives (TN)**: We predicted no(employees will not leave the organisation) and they don’t leave i.e 50\n",
    "* **False positives (FP)**: We predicted yes they will leave, but they don’t leave. (Also known as a “Type I error.”) i.e 10\n",
    "* **False negatives (FN)**: We predicted no they will not leave, but they actually leave (Also known as a “Type II error.”) i.e 5\n",
    "\n",
    "### Evaluating a Classification Model\n",
    "\n",
    "* **Accuracy** : (TP+TN)/Total . Describes overall, how often the classifier correct. i.e 100+50/165\n",
    "Measures of Accuracy\n",
    "\n",
    "Sensitivity and specificity are statistical measures of the performance of a binary classification test:\n",
    "\n",
    "* **Sensitivity/Recall** = TP/(TP + FN). When it’s actually yes, how often does it predict yes? i.e 100/(100+5)\n",
    "\n",
    "* **Specificity** = TN/(TN + FP).When it’s actually no, how often does it predict no?? i.e 50/(50+10)\n",
    "* **Precision** = TP/predicted yes. When it predicts yes, how often is it correct?100/(10+100)\n",
    "\n",
    "### Evaluation metrics for a Classification model’s performance.\n",
    "\n",
    "**ROC curve**\n",
    "\n",
    "A **ROC(Receiver Operator Characteristic Curve)** can help in deciding the best threshold value. It is generated by plotting the True Positive Rate (y-axis) against the False Positive Rate (x-axis) as you vary the threshold for assigning observations to a given class.ROC curve will always end at (1,1). The threshold at this point will be 0. This means that we will always classify these observations falling into the class 1(Specificity will be 0. False positive rate is 1).\n",
    "\n",
    "One should select the best threshold for the trade off you want to make. According to the criticality of the business, we need to compare the cost of failing to detect positives vs cost of raising false alarms.\n",
    "\n",
    "![](https://miro.medium.com/max/469/1*Y65IEOXvxLRKKqWxlQovsg.png)\n",
    "\n",
    "An animation to demonstrate how an ROC curve relates to sensitivity and specificity for all possible cutoffs.\n",
    "\n",
    "![Alt Text](https://github.com/dariyasydykova/open_projects/blob/master/ROC_animation/animations/ROC.gif?raw=true)\n",
    "\n",
    "[Source](https://github.com/dariyasydykova/open_projects/blob/master/ROC_animation/animations/ROC.gif)\n",
    "\n",
    "**High Threshold:**\n",
    "* High specificity\n",
    "* Low sensitivity\n",
    "\n",
    "\n",
    "**Low Threshold**\n",
    "* Low specificity\n",
    "* High sensitivity\n",
    "\n",
    "The area under ROC is called *Area Under the Curve(AUC)*. AUC gives the rate of successful classification by the logistic model. To get a more in-depth idea of what a ROC-AUC curve is and how is it calculated, here is a link to the [article](https://towardsdatascience.com/understanding-the-roc-and-auc-curves-a05b68550b69) I wrote on the same topic."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "_cell_guid": "3c44a90d-cbfb-4787-acee-56da3f180f86",
    "_uuid": "7b6090270c785e1b60309025e966cc44c2724dcc",
    "id": "D-pCykvKrwy8"
   },
   "source": [
    "### Confusion matrix"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {
    "_cell_guid": "3b57df93-e9b9-4d32-8fe8-8a1afe1d28d9",
    "_uuid": "6dadfc64d1a62b14d472972cd3b5b3595711a5e8",
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 338
    },
    "id": "v1_nAdADrwy9",
    "outputId": "a80ec3dc-19e5-48c2-d3d4-d76ebb674ca4"
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.axes._subplots.AxesSubplot at 0x7f7b7f1b5350>"
      ]
     },
     "execution_count": 14,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 576x360 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "from sklearn.metrics import confusion_matrix\n",
    "cm=confusion_matrix(y_test,y_pred)\n",
    "conf_matrix=pd.DataFrame(data=cm,columns=['Predicted:0','Predicted:1'],index=['Actual:0','Actual:1'])\n",
    "plt.figure(figsize = (8,5))\n",
    "sn.heatmap(conf_matrix, annot=True,fmt='d',cmap=\"YlGnBu\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "_cell_guid": "abfa561a-b375-4edd-b493-d69c3ff3ee61",
    "_uuid": "6819f4ddb5cc364ae41bce422812137afbee4186",
    "id": "_mXVP5-XrwzA"
   },
   "source": [
    "The confusion matrix shows 652+5 = 657 correct predictions and 87+7= 94 incorrect ones.\n",
    "\n",
    "- True Positives:  5\n",
    "- True Negatives:  652\n",
    "- False Positives: 7 (*Type I error*)\n",
    "- False Negatives: 87 ( *Type II error*)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {
    "_cell_guid": "838dd1cc-78df-452d-8306-e06e8baa0e6d",
    "_uuid": "7679a3bd16a92619001cf7e349920e39938c4dc0",
    "id": "iTO0hqWurwzB"
   },
   "outputs": [],
   "source": [
    "TN=cm[0,0]\n",
    "TP=cm[1,1]\n",
    "FN=cm[1,0]\n",
    "FP=cm[0,1]\n",
    "sensitivity=TP/float(TP+FN)\n",
    "specificity=TN/float(TN+FP)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "_cell_guid": "32fd2aba-f6ed-41cc-b6e2-676a626411d8",
    "_uuid": "6ad4604c589dd46bb03dba2157dc97f5abf80bc0",
    "id": "m2lUba9wrwzG"
   },
   "source": [
    "### Model Evaluation - Statistics"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {
    "_cell_guid": "4b314554-2e62-4b78-8fd2-8a6bbcdc5fd9",
    "_uuid": "7448c4155697b69d089237d133be1c4df4376723",
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "jdYsG6HTrwzH",
    "outputId": "a302a3e0-48e2-4327-a743-43f5c892874b"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "The acuuracy of the model = TP+TN/(TP+TN+FP+FN) =  0.8748335552596538 \n",
      " The Missclassification = 1-Accuracy =  0.12516644474034622 \n",
      " Sensitivity or True Positive Rate = TP/(TP+FN) =  0.05434782608695652 \n",
      " Specificity or True Negative Rate = TN/(TN+FP) =  0.9893778452200304 \n",
      " Positive Predictive value = TP/(TP+FP) =  0.4166666666666667 \n",
      " Negative predictive Value = TN/(TN+FN) =  0.8822733423545331 \n",
      " Positive Likelihood Ratio = Sensitivity/(1-Specificity) =  5.116459627329198 \n",
      " Negative likelihood Ratio = (1-Sensitivity)/Specificity =  0.9558048813016804\n"
     ]
    }
   ],
   "source": [
    "print('The acuuracy of the model = TP+TN/(TP+TN+FP+FN) = ',(TP+TN)/float(TP+TN+FP+FN),'\\n',\n",
    "\n",
    "'The Missclassification = 1-Accuracy = ',1-((TP+TN)/float(TP+TN+FP+FN)),'\\n',\n",
    "\n",
    "'Sensitivity or True Positive Rate = TP/(TP+FN) = ',TP/float(TP+FN),'\\n',\n",
    "\n",
    "'Specificity or True Negative Rate = TN/(TN+FP) = ',TN/float(TN+FP),'\\n',\n",
    "\n",
    "'Positive Predictive value = TP/(TP+FP) = ',TP/float(TP+FP),'\\n',\n",
    "\n",
    "'Negative predictive Value = TN/(TN+FN) = ',TN/float(TN+FN),'\\n',\n",
    "\n",
    "'Positive Likelihood Ratio = Sensitivity/(1-Specificity) = ',sensitivity/(1-specificity),'\\n',\n",
    "\n",
    "'Negative likelihood Ratio = (1-Sensitivity)/Specificity = ',(1-sensitivity)/specificity)\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "_cell_guid": "161e6ad5-13bb-46ca-9a59-ce88a2572d72",
    "_uuid": "1a21c02ad27ae1f35d8b2c1a7470f1fe3ad99a7c",
    "id": "cdgTJ47WrwzL"
   },
   "source": [
    "**From the above statistics it is clear that the model is highly specific than sensitive. The negative values are predicted more accurately than the positives.**"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "_cell_guid": "a14bc45c-f000-4803-983e-934f331ae0f6",
    "_uuid": "2f936ed959075964b25fcb753548ea00f3fe0af2",
    "id": "Dzvsk3IlrwzZ"
   },
   "source": [
    "A common way to visualize the trade-offs of different thresholds is by using an ROC curve, a plot of the true positive rate (# true positives/ total # positives) versus the false positive rate (# false positives /\n",
    "total # negatives) for all possible choices of thresholds. A model with good classification accuracy should have significantly more true positives than false positives at all thresholds. \n",
    "\n",
    "The optimum position for roc curve is towards the top left corner where the specificity and sensitivity are at optimum levels"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "WVbM6JfkHpLW"
   },
   "source": [
    "### Predicted probabilities of  0 (No Coronary Heart Disease) and 1 ( Coronary Heart Disease: Yes)  for the test data with a default classification threshold of 0.5"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 206
    },
    "id": "Q0Momj1dHros",
    "outputId": "1b249a2c-9aae-4965-c3f1-33009b213f29"
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "\n",
       "  <div id=\"df-2657f0f1-7c0f-41a2-952f-bbf58e919d19\">\n",
       "    <div class=\"colab-df-container\">\n",
       "      <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>Prob of no heart disease (0)</th>\n",
       "      <th>Prob of Heart Disease (1)</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>0.875026</td>\n",
       "      <td>0.124974</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>0.956183</td>\n",
       "      <td>0.043817</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>0.783473</td>\n",
       "      <td>0.216527</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>0.806581</td>\n",
       "      <td>0.193419</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>0.892879</td>\n",
       "      <td>0.107121</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>\n",
       "      <button class=\"colab-df-convert\" onclick=\"convertToInteractive('df-2657f0f1-7c0f-41a2-952f-bbf58e919d19')\"\n",
       "              title=\"Convert this dataframe to an interactive table.\"\n",
       "              style=\"display:none;\">\n",
       "        \n",
       "  <svg xmlns=\"http://www.w3.org/2000/svg\" height=\"24px\"viewBox=\"0 0 24 24\"\n",
       "       width=\"24px\">\n",
       "    <path d=\"M0 0h24v24H0V0z\" fill=\"none\"/>\n",
       "    <path d=\"M18.56 5.44l.94 2.06.94-2.06 2.06-.94-2.06-.94-.94-2.06-.94 2.06-2.06.94zm-11 1L8.5 8.5l.94-2.06 2.06-.94-2.06-.94L8.5 2.5l-.94 2.06-2.06.94zm10 10l.94 2.06.94-2.06 2.06-.94-2.06-.94-.94-2.06-.94 2.06-2.06.94z\"/><path d=\"M17.41 7.96l-1.37-1.37c-.4-.4-.92-.59-1.43-.59-.52 0-1.04.2-1.43.59L10.3 9.45l-7.72 7.72c-.78.78-.78 2.05 0 2.83L4 21.41c.39.39.9.59 1.41.59.51 0 1.02-.2 1.41-.59l7.78-7.78 2.81-2.81c.8-.78.8-2.07 0-2.86zM5.41 20L4 18.59l7.72-7.72 1.47 1.35L5.41 20z\"/>\n",
       "  </svg>\n",
       "      </button>\n",
       "      \n",
       "  <style>\n",
       "    .colab-df-container {\n",
       "      display:flex;\n",
       "      flex-wrap:wrap;\n",
       "      gap: 12px;\n",
       "    }\n",
       "\n",
       "    .colab-df-convert {\n",
       "      background-color: #E8F0FE;\n",
       "      border: none;\n",
       "      border-radius: 50%;\n",
       "      cursor: pointer;\n",
       "      display: none;\n",
       "      fill: #1967D2;\n",
       "      height: 32px;\n",
       "      padding: 0 0 0 0;\n",
       "      width: 32px;\n",
       "    }\n",
       "\n",
       "    .colab-df-convert:hover {\n",
       "      background-color: #E2EBFA;\n",
       "      box-shadow: 0px 1px 2px rgba(60, 64, 67, 0.3), 0px 1px 3px 1px rgba(60, 64, 67, 0.15);\n",
       "      fill: #174EA6;\n",
       "    }\n",
       "\n",
       "    [theme=dark] .colab-df-convert {\n",
       "      background-color: #3B4455;\n",
       "      fill: #D2E3FC;\n",
       "    }\n",
       "\n",
       "    [theme=dark] .colab-df-convert:hover {\n",
       "      background-color: #434B5C;\n",
       "      box-shadow: 0px 1px 3px 1px rgba(0, 0, 0, 0.15);\n",
       "      filter: drop-shadow(0px 1px 2px rgba(0, 0, 0, 0.3));\n",
       "      fill: #FFFFFF;\n",
       "    }\n",
       "  </style>\n",
       "\n",
       "      <script>\n",
       "        const buttonEl =\n",
       "          document.querySelector('#df-2657f0f1-7c0f-41a2-952f-bbf58e919d19 button.colab-df-convert');\n",
       "        buttonEl.style.display =\n",
       "          google.colab.kernel.accessAllowed ? 'block' : 'none';\n",
       "\n",
       "        async function convertToInteractive(key) {\n",
       "          const element = document.querySelector('#df-2657f0f1-7c0f-41a2-952f-bbf58e919d19');\n",
       "          const dataTable =\n",
       "            await google.colab.kernel.invokeFunction('convertToInteractive',\n",
       "                                                     [key], {});\n",
       "          if (!dataTable) return;\n",
       "\n",
       "          const docLinkHtml = 'Like what you see? Visit the ' +\n",
       "            '<a target=\"_blank\" href=https://colab.research.google.com/notebooks/data_table.ipynb>data table notebook</a>'\n",
       "            + ' to learn more about interactive tables.';\n",
       "          element.innerHTML = '';\n",
       "          dataTable['output_type'] = 'display_data';\n",
       "          await google.colab.output.renderOutput(dataTable, element);\n",
       "          const docLink = document.createElement('div');\n",
       "          docLink.innerHTML = docLinkHtml;\n",
       "          element.appendChild(docLink);\n",
       "        }\n",
       "      </script>\n",
       "    </div>\n",
       "  </div>\n",
       "  "
      ],
      "text/plain": [
       "   Prob of no heart disease (0)  Prob of Heart Disease (1)\n",
       "0                      0.875026                   0.124974\n",
       "1                      0.956183                   0.043817\n",
       "2                      0.783473                   0.216527\n",
       "3                      0.806581                   0.193419\n",
       "4                      0.892879                   0.107121"
      ]
     },
     "execution_count": 17,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "y_pred_prob=logreg.predict_proba(x_test)[:,:]\n",
    "y_pred_prob_df=pd.DataFrame(data=y_pred_prob, columns=['Prob of no heart disease (0)','Prob of Heart Disease (1)'])\n",
    "y_pred_prob_df.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "bP9GSW6VHtn8"
   },
   "source": [
    "### Lower the threshold"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "EdoHWobuHwUC"
   },
   "source": [
    "Since the model is predicting Heart disease too many type II errors is not advisable. A False Negative ( ignoring the probability of disease when there actualy is one) is more dangerous than a False Positive in this case. Hence inorder to increase the sensitivity,  threshold can be lowered."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "fGgXqCLoHyUJ",
    "outputId": "03959ade-9135-412e-be4f-c49280c56dbc"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "With 0.1 threshold the Confusion Matrix is  \n",
      " [[311 348]\n",
      " [ 12  80]] \n",
      " with 391 correct predictions and 12 Type II errors( False Negatives) \n",
      "\n",
      " Sensitivity:  0.8695652173913043 Specificity:  0.47192716236722304 \n",
      "\n",
      "\n",
      "\n",
      "With 0.2 threshold the Confusion Matrix is  \n",
      " [[518 141]\n",
      " [ 43  49]] \n",
      " with 567 correct predictions and 43 Type II errors( False Negatives) \n",
      "\n",
      " Sensitivity:  0.532608695652174 Specificity:  0.7860394537177542 \n",
      "\n",
      "\n",
      "\n",
      "With 0.3 threshold the Confusion Matrix is  \n",
      " [[600  59]\n",
      " [ 64  28]] \n",
      " with 628 correct predictions and 64 Type II errors( False Negatives) \n",
      "\n",
      " Sensitivity:  0.30434782608695654 Specificity:  0.9104704097116844 \n",
      "\n",
      "\n",
      "\n",
      "With 0.4 threshold the Confusion Matrix is  \n",
      " [[640  19]\n",
      " [ 80  12]] \n",
      " with 652 correct predictions and 80 Type II errors( False Negatives) \n",
      "\n",
      " Sensitivity:  0.13043478260869565 Specificity:  0.9711684370257967 \n",
      "\n",
      "\n",
      "\n"
     ]
    }
   ],
   "source": [
    "from sklearn.preprocessing import binarize\n",
    "for i in range(1,5):\n",
    "    cm2=0\n",
    "    y_pred_prob_yes=logreg.predict_proba(x_test)\n",
    "    y_pred2=binarize(y_pred_prob_yes,threshold=i/10)[:,1]\n",
    "    cm2=confusion_matrix(y_test,y_pred2)\n",
    "    print ('With',i/10,'threshold the Confusion Matrix is ','\\n',cm2,'\\n',\n",
    "            'with',cm2[0,0]+cm2[1,1],'correct predictions and',cm2[1,0],'Type II errors( False Negatives)','\\n\\n',\n",
    "          'Sensitivity: ',cm2[1,1]/(float(cm2[1,1]+cm2[1,0])),'Specificity: ',cm2[0,0]/(float(cm2[0,0]+cm2[0,1])),'\\n\\n\\n')\n",
    "    "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "hTURCvuXH2MF"
   },
   "source": [
    "### ROC curve<font>"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 295
    },
    "id": "7SES9eXXH38L",
    "outputId": "2229e89b-8f8f-4cda-8f62-18203b4016d1"
   },
   "outputs": [
    {
     "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": [
    "from sklearn.metrics import roc_curve\n",
    "fpr, tpr, thresholds = roc_curve(y_test, y_pred_prob_yes[:,1])\n",
    "plt.plot(fpr,tpr)\n",
    "plt.xlim([0.0, 1.0])\n",
    "plt.ylim([0.0, 1.0])\n",
    "plt.title('ROC curve for Heart disease classifier')\n",
    "plt.xlabel('False positive rate (1-Specificity)')\n",
    "plt.ylabel('True positive rate (Sensitivity)')\n",
    "plt.grid(True)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "yPilqnDnH7v3"
   },
   "source": [
    "A common way to visualize the trade-offs of different thresholds is by using an ROC curve, a plot of the true positive rate (# true positives/ total # positives) versus the false positive rate (# false positives /\n",
    "total # negatives) for all possible choices of thresholds. A model with good classification accuracy should have significantly more true positives than false positives at all thresholds. \n",
    "\n",
    "The optimum position for roc curve is towards the top left corner where the specificity and sensitivity are at optimum levels"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "_cell_guid": "8b48402f-20b5-4408-ae09-79576c61bc85",
    "_uuid": "25342d42365c10d4b8c397324c09bfd658776462",
    "id": "WN8FK_EIrwza"
   },
   "source": [
    "### Area Under The Curve (AUC)\n",
    "\n",
    "The area under the ROC curve quantifies model classification accuracy; the higher the area, the greater the disparity between true and false positives, and the stronger the model in classifying members of the training dataset. An area of 0.5 corresponds to a model that performs no better than random classification and a good classifier stays as far away from that  as possible. An area of 1 is ideal. The closer the AUC to 1 the better."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {
    "_cell_guid": "719c8a37-8cf1-44ed-b312-d4d7ef6414ff",
    "_uuid": "62755c5a95c67810df506bf3972ecf319a86bf98",
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "xWIA9LrYrwza",
    "outputId": "9f0d8d82-3cc9-41e5-b2f7-93212937121b"
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.7386191198786038"
      ]
     },
     "execution_count": 20,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "sklearn.metrics.roc_auc_score(y_test,y_pred_prob_yes[:,1])"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "_cell_guid": "bda76cab-79ca-4f3f-a864-19443787a3e4",
    "_uuid": "c2e31f1f363af6d137ebf7df90d62b9cfae6f136",
    "id": "daPma_JCrwzd"
   },
   "source": [
    "\n",
    "## Conclusions:\n",
    "\n",
    "\n",
    "\n",
    "\n",
    " - All attributes selected after the elimination process show Pvalues lower than 5% and thereby suggesting significant role in the Heart disease prediction. \n",
    "\n",
    " - Men seem to be more susceptible to heart disease than women.Increase in Age,number of cigarettes smoked per day and systolic Blood Pressure also show increasing odds of having heart disease.\n",
    "\n",
    " - Total cholesterol shows no significant change in the odds of CHD. This could be due to the presence of 'good cholesterol(HDL) in the total cholesterol reading.Glucose too causes a very negligible change in odds (0.2%)\n",
    "\n",
    " - The model predicted with 0.88 accuracy. The model is more specific than sensitive.\n",
    "\n",
    " - The Area under the ROC curve is 73.5 which is somewhat satisfactory.\n",
    "\n",
    " - Overall model could be improved with more data."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "_cell_guid": "a3e3d46a-61a6-4b14-a518-cefa27b44b5a",
    "_uuid": "5e0b2ae2206ad6758fc87492ea41999afb4887d0",
    "id": "2q8maGOWrwze"
   },
   "source": [
    "## Resource Credits\n",
    "\n",
    "http://www.who.int/mediacentre/factsheets/fs317/en/\n",
    "\n",
    "https://www.kaggle.com/parulpandey/deep-dive-into-logistic-regression-for-beginners\n",
    "\n",
    "\n",
    "https://www.kaggle.com/amanajmera1/framingham-heart-study-dataset/data\n",
    "\n",
    "https://www.kaggle.com/neisha/heart-disease-prediction-using-logistic-regression"
   ]
  }
 ],
 "metadata": {
  "colab": {
   "collapsed_sections": [],
   "name": "logistic-regression-heart-disease-97e2e78290044a9393d3f26cb3d818d9 (1).ipynb",
   "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": "3.8.8"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 1
}