DAY 16 - Liver Disease Prediction - Logistic Regression.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "id": "protected-exercise",
   "metadata": {},
   "source": [
    "# Logistic Regression\n",
    "\n",
    "- It is a supervised learning algorithm used to predict the probability of target variable.\n",
    "- The nature of target variable is dichotomous.\n",
    "- It means that there would be only two possible classes.\n",
    "- The dependent variable is binary in nature having data coded as 1(stands for success) and 0(stands for failure)\n",
    "- A logistic regression model predicts P(Y=1) as a function of X.\n",
    "- It is one of the simplest algorithms that can be used for various classification problems.\n",
    "- The classification problems can be spam detection, diabetes prediction and cancer prediction.\n",
    "\n",
    "# Types of Logistic Regression\n",
    "\n",
    "- Logistic Regression means binary logistic regression having two target variables.\n",
    "- There can be two more categories of target variables that can be predicted by it.\n",
    "- Based on the number of categories, logistic regression can be classified into the following types:\n",
    "    1. BInary or Binomial - In this, a dependent variable will have only two possible types, 1 and 0. These variables may represent success or failure, yes or no, win or loss.\n",
    "    2. Multinomial - In this kind of classification, dependent variables can have three or more possible unordered types. These types have no quantitative significance. These variables may represent \"Type A\" or \"Type B\" or \"Type C\".\n",
    "    3. Ordinal - In this kind of classification, dependent variable can have three or more possible ordered types. All these types have a quantitative significance. These variables may represent \"poor\" or \"good\", \"very good\" and \"excellent\". Each category can have the scores like 0,1,2,3."
   ]
  },
  {
   "cell_type": "markdown",
   "id": "infinite-thompson",
   "metadata": {},
   "source": [
    "# Liver Disease Prediction"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "successful-counter",
   "metadata": {},
   "source": [
    "### Objective\n",
    "- To develop and implement a Machine Learning approach that characterizes the features of the Liver Disease revolution, and proposes a Liver Disease processing model, from the data mining perspective using Naive Bayes Classification.\n",
    "\n",
    "### Modules\n",
    "- Dataset\n",
    "- Data preprocessing\n",
    "- Classification Techniques\n",
    "- Logistic Regression"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "id": "editorial-universe",
   "metadata": {},
   "outputs": [],
   "source": [
    "# for numerical computing\n",
    "import numpy as np\n",
    "\n",
    "# for dataframes\n",
    "import pandas as pd\n",
    "\n",
    "# for easier visualization\n",
    "import seaborn as sns\n",
    "\n",
    "# for visualization and to display plots\n",
    "from matplotlib import pyplot as plt\n",
    "%matplotlib inline\n",
    "\n",
    "# import color maps\n",
    "from matplotlib.colors import ListedColormap\n",
    "\n",
    "# ignore warnings\n",
    "import warnings\n",
    "warnings.filterwarnings(\"ignore\")\n",
    "\n",
    "# to perform hyperparameter tuning\n",
    "from sklearn.model_selection import GridSearchCV\n",
    "\n",
    "from math import sqrt\n",
    "\n",
    "# to split train and test set\n",
    "from sklearn.model_selection import train_test_split\n",
    "from sklearn.linear_model import LogisticRegression\n",
    "from sklearn.svm import SVC\n",
    "from sklearn.metrics import roc_curve, auc, roc_auc_score, confusion_matrix\n",
    "\n",
    "from sklearn.preprocessing import StandardScaler\n",
    "\n",
    "from sklearn.metrics import accuracy_score"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "id": "coordinated-puppy",
   "metadata": {},
   "outputs": [],
   "source": [
    "df=pd.read_csv('indian_liver_patient.csv')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "id": "capable-mainland",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(583, 11)"
      ]
     },
     "execution_count": 5,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "id": "hybrid-rebel",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Index(['Age', 'Gender', 'Total_Bilirubin', 'Direct_Bilirubin',\n",
       "       'Alkaline_Phosphotase', 'Alamine_Aminotransferase',\n",
       "       'Aspartate_Aminotransferase', 'Total_Protiens', 'Albumin',\n",
       "       'Albumin_and_Globulin_Ratio', 'Dataset'],\n",
       "      dtype='object')"
      ]
     },
     "execution_count": 6,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.columns"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "id": "missing-possible",
   "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>Age</th>\n",
       "      <th>Gender</th>\n",
       "      <th>Total_Bilirubin</th>\n",
       "      <th>Direct_Bilirubin</th>\n",
       "      <th>Alkaline_Phosphotase</th>\n",
       "      <th>Alamine_Aminotransferase</th>\n",
       "      <th>Aspartate_Aminotransferase</th>\n",
       "      <th>Total_Protiens</th>\n",
       "      <th>Albumin</th>\n",
       "      <th>Albumin_and_Globulin_Ratio</th>\n",
       "      <th>Dataset</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>65</td>\n",
       "      <td>Female</td>\n",
       "      <td>0.7</td>\n",
       "      <td>0.1</td>\n",
       "      <td>187</td>\n",
       "      <td>16</td>\n",
       "      <td>18</td>\n",
       "      <td>6.8</td>\n",
       "      <td>3.3</td>\n",
       "      <td>0.90</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>62</td>\n",
       "      <td>Male</td>\n",
       "      <td>10.9</td>\n",
       "      <td>5.5</td>\n",
       "      <td>699</td>\n",
       "      <td>64</td>\n",
       "      <td>100</td>\n",
       "      <td>7.5</td>\n",
       "      <td>3.2</td>\n",
       "      <td>0.74</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>62</td>\n",
       "      <td>Male</td>\n",
       "      <td>7.3</td>\n",
       "      <td>4.1</td>\n",
       "      <td>490</td>\n",
       "      <td>60</td>\n",
       "      <td>68</td>\n",
       "      <td>7.0</td>\n",
       "      <td>3.3</td>\n",
       "      <td>0.89</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>58</td>\n",
       "      <td>Male</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.4</td>\n",
       "      <td>182</td>\n",
       "      <td>14</td>\n",
       "      <td>20</td>\n",
       "      <td>6.8</td>\n",
       "      <td>3.4</td>\n",
       "      <td>1.00</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>72</td>\n",
       "      <td>Male</td>\n",
       "      <td>3.9</td>\n",
       "      <td>2.0</td>\n",
       "      <td>195</td>\n",
       "      <td>27</td>\n",
       "      <td>59</td>\n",
       "      <td>7.3</td>\n",
       "      <td>2.4</td>\n",
       "      <td>0.40</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   Age  Gender  Total_Bilirubin  Direct_Bilirubin  Alkaline_Phosphotase  \\\n",
       "0   65  Female              0.7               0.1                   187   \n",
       "1   62    Male             10.9               5.5                   699   \n",
       "2   62    Male              7.3               4.1                   490   \n",
       "3   58    Male              1.0               0.4                   182   \n",
       "4   72    Male              3.9               2.0                   195   \n",
       "\n",
       "   Alamine_Aminotransferase  Aspartate_Aminotransferase  Total_Protiens  \\\n",
       "0                        16                          18             6.8   \n",
       "1                        64                         100             7.5   \n",
       "2                        60                          68             7.0   \n",
       "3                        14                          20             6.8   \n",
       "4                        27                          59             7.3   \n",
       "\n",
       "   Albumin  Albumin_and_Globulin_Ratio  Dataset  \n",
       "0      3.3                        0.90        1  \n",
       "1      3.2                        0.74        1  \n",
       "2      3.3                        0.89        1  \n",
       "3      3.4                        1.00        1  \n",
       "4      2.4                        0.40        1  "
      ]
     },
     "execution_count": 7,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "greek-liverpool",
   "metadata": {},
   "source": [
    "## Exploratory Data Analysis\n",
    "\n",
    "Filtering categorical data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "id": "serial-heating",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Gender    object\n",
       "dtype: object"
      ]
     },
     "execution_count": 8,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.dtypes[df.dtypes=='object']"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "fleet-management",
   "metadata": {},
   "source": [
    "## Distribution of Numerical Features "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "id": "equal-notebook",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1080x1080 with 12 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Plot histogram grid\n",
    "df.hist(figsize=(15,15), xrot=-45, bins=10) ## Display the labels rotated by 45 degress\n",
    "\n",
    "# Clear the text \"residue\"\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "id": "exceptional-inspiration",
   "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>Age</th>\n",
       "      <th>Total_Bilirubin</th>\n",
       "      <th>Direct_Bilirubin</th>\n",
       "      <th>Alkaline_Phosphotase</th>\n",
       "      <th>Alamine_Aminotransferase</th>\n",
       "      <th>Aspartate_Aminotransferase</th>\n",
       "      <th>Total_Protiens</th>\n",
       "      <th>Albumin</th>\n",
       "      <th>Albumin_and_Globulin_Ratio</th>\n",
       "      <th>Dataset</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>count</th>\n",
       "      <td>583.000000</td>\n",
       "      <td>583.000000</td>\n",
       "      <td>583.000000</td>\n",
       "      <td>583.000000</td>\n",
       "      <td>583.000000</td>\n",
       "      <td>583.000000</td>\n",
       "      <td>583.000000</td>\n",
       "      <td>583.000000</td>\n",
       "      <td>579.000000</td>\n",
       "      <td>583.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>mean</th>\n",
       "      <td>44.746141</td>\n",
       "      <td>3.298799</td>\n",
       "      <td>1.486106</td>\n",
       "      <td>290.576329</td>\n",
       "      <td>80.713551</td>\n",
       "      <td>109.910806</td>\n",
       "      <td>6.483190</td>\n",
       "      <td>3.141852</td>\n",
       "      <td>0.947064</td>\n",
       "      <td>1.286449</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>std</th>\n",
       "      <td>16.189833</td>\n",
       "      <td>6.209522</td>\n",
       "      <td>2.808498</td>\n",
       "      <td>242.937989</td>\n",
       "      <td>182.620356</td>\n",
       "      <td>288.918529</td>\n",
       "      <td>1.085451</td>\n",
       "      <td>0.795519</td>\n",
       "      <td>0.319592</td>\n",
       "      <td>0.452490</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>min</th>\n",
       "      <td>4.000000</td>\n",
       "      <td>0.400000</td>\n",
       "      <td>0.100000</td>\n",
       "      <td>63.000000</td>\n",
       "      <td>10.000000</td>\n",
       "      <td>10.000000</td>\n",
       "      <td>2.700000</td>\n",
       "      <td>0.900000</td>\n",
       "      <td>0.300000</td>\n",
       "      <td>1.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>25%</th>\n",
       "      <td>33.000000</td>\n",
       "      <td>0.800000</td>\n",
       "      <td>0.200000</td>\n",
       "      <td>175.500000</td>\n",
       "      <td>23.000000</td>\n",
       "      <td>25.000000</td>\n",
       "      <td>5.800000</td>\n",
       "      <td>2.600000</td>\n",
       "      <td>0.700000</td>\n",
       "      <td>1.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>50%</th>\n",
       "      <td>45.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.300000</td>\n",
       "      <td>208.000000</td>\n",
       "      <td>35.000000</td>\n",
       "      <td>42.000000</td>\n",
       "      <td>6.600000</td>\n",
       "      <td>3.100000</td>\n",
       "      <td>0.930000</td>\n",
       "      <td>1.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>75%</th>\n",
       "      <td>58.000000</td>\n",
       "      <td>2.600000</td>\n",
       "      <td>1.300000</td>\n",
       "      <td>298.000000</td>\n",
       "      <td>60.500000</td>\n",
       "      <td>87.000000</td>\n",
       "      <td>7.200000</td>\n",
       "      <td>3.800000</td>\n",
       "      <td>1.100000</td>\n",
       "      <td>2.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>max</th>\n",
       "      <td>90.000000</td>\n",
       "      <td>75.000000</td>\n",
       "      <td>19.700000</td>\n",
       "      <td>2110.000000</td>\n",
       "      <td>2000.000000</td>\n",
       "      <td>4929.000000</td>\n",
       "      <td>9.600000</td>\n",
       "      <td>5.500000</td>\n",
       "      <td>2.800000</td>\n",
       "      <td>2.000000</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "              Age  Total_Bilirubin  Direct_Bilirubin  Alkaline_Phosphotase  \\\n",
       "count  583.000000       583.000000        583.000000            583.000000   \n",
       "mean    44.746141         3.298799          1.486106            290.576329   \n",
       "std     16.189833         6.209522          2.808498            242.937989   \n",
       "min      4.000000         0.400000          0.100000             63.000000   \n",
       "25%     33.000000         0.800000          0.200000            175.500000   \n",
       "50%     45.000000         1.000000          0.300000            208.000000   \n",
       "75%     58.000000         2.600000          1.300000            298.000000   \n",
       "max     90.000000        75.000000         19.700000           2110.000000   \n",
       "\n",
       "       Alamine_Aminotransferase  Aspartate_Aminotransferase  Total_Protiens  \\\n",
       "count                583.000000                  583.000000      583.000000   \n",
       "mean                  80.713551                  109.910806        6.483190   \n",
       "std                  182.620356                  288.918529        1.085451   \n",
       "min                   10.000000                   10.000000        2.700000   \n",
       "25%                   23.000000                   25.000000        5.800000   \n",
       "50%                   35.000000                   42.000000        6.600000   \n",
       "75%                   60.500000                   87.000000        7.200000   \n",
       "max                 2000.000000                 4929.000000        9.600000   \n",
       "\n",
       "          Albumin  Albumin_and_Globulin_Ratio     Dataset  \n",
       "count  583.000000                  579.000000  583.000000  \n",
       "mean     3.141852                    0.947064    1.286449  \n",
       "std      0.795519                    0.319592    0.452490  \n",
       "min      0.900000                    0.300000    1.000000  \n",
       "25%      2.600000                    0.700000    1.000000  \n",
       "50%      3.100000                    0.930000    1.000000  \n",
       "75%      3.800000                    1.100000    2.000000  \n",
       "max      5.500000                    2.800000    2.000000  "
      ]
     },
     "execution_count": 10,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.describe()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "revolutionary-device",
   "metadata": {},
   "source": [
    "It seems there is outlier in Aspartate_Aminotransferase as the max value is very high than mean value\n",
    "\n",
    "Dataset i.e output value has '1' for liver disease and '2' for no liver disease so let's make it 0 for no disease to make it convinient"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "id": "senior-edmonton",
   "metadata": {},
   "outputs": [],
   "source": [
    "## if score==negative, mark 0 ;else 1 \n",
    "def partition(x):\n",
    "    if x == 2:\n",
    "        return 0\n",
    "    return 1\n",
    "\n",
    "df['Dataset'] = df['Dataset'].map(partition)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "ahead-choir",
   "metadata": {},
   "source": [
    "## Distribution of categorical data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "id": "pharmaceutical-catch",
   "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>Gender</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>count</th>\n",
       "      <td>583</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>unique</th>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>top</th>\n",
       "      <td>Male</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>freq</th>\n",
       "      <td>441</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "       Gender\n",
       "count     583\n",
       "unique      2\n",
       "top      Male\n",
       "freq      441"
      ]
     },
     "execution_count": 12,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.describe(include=['object'])"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "crazy-concern",
   "metadata": {},
   "source": [
    "## Bar plots for categorical Features"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "id": "corporate-compatibility",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<AxesSubplot:xlabel='count', ylabel='Gender'>"
      ]
     },
     "execution_count": 13,
     "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": [
    "plt.figure(figsize=(5,5))\n",
    "sns.countplot(y='Gender', data=df)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "id": "sublime-temperature",
   "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>Dataset</th>\n",
       "      <th>Gender</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>1</td>\n",
       "      <td>Male</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>1</td>\n",
       "      <td>Male</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>1</td>\n",
       "      <td>Male</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>1</td>\n",
       "      <td>Male</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>1</td>\n",
       "      <td>Male</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   Dataset Gender\n",
       "1        1   Male\n",
       "2        1   Male\n",
       "3        1   Male\n",
       "4        1   Male\n",
       "5        1   Male"
      ]
     },
     "execution_count": 14,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df[df['Gender'] == 'Male'][['Dataset', 'Gender']].head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "id": "broad-africa",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 407.125x360 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "sns.factorplot (x=\"Age\", y=\"Gender\", hue=\"Dataset\", data=df);"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "assured-gross",
   "metadata": {},
   "source": [
    "Age seems to be a factor for liver disease for both male and female genders"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "id": "atmospheric-intention",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Number of patients that are male:  441\n",
      "Number of patients that are female:  142\n"
     ]
    },
    {
     "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(data=df, x = 'Gender', label='Count')\n",
    "\n",
    "M, F = df['Gender'].value_counts()\n",
    "print('Number of patients that are male: ',M)\n",
    "print('Number of patients that are female: ',F)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "chemical-brazil",
   "metadata": {},
   "source": [
    "There are more male patients than female patients\n",
    "\n",
    "Label Male as 0 and Female as 1"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "id": "cloudy-formation",
   "metadata": {},
   "outputs": [],
   "source": [
    "## if score==negative, mark 0 ;else 1 \n",
    "def partition(x):\n",
    "    if x =='Male':\n",
    "        return 0\n",
    "    return 1\n",
    "\n",
    "df['Gender'] = df['Gender'].map(partition)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "heard-title",
   "metadata": {},
   "source": [
    "#### 2-D Scatter Plot"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "id": "foreign-greene",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<seaborn.axisgrid.FacetGrid at 0x1bdf6d8d308>"
      ]
     },
     "execution_count": 18,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 402.75x360 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "sns.set_style('whitegrid')   ## Background Grid\n",
    "sns.FacetGrid(df, hue = 'Dataset', size = 5).map(plt.scatter, 'Total_Bilirubin', 'Direct_Bilirubin').add_legend()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "id": "endless-listing",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<seaborn.axisgrid.FacetGrid at 0x1bdf6df6d48>"
      ]
     },
     "execution_count": 19,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 402.75x360 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "sns.set_style('whitegrid')   ## Background Grid\n",
    "sns.FacetGrid(df, hue = 'Dataset', size = 5).map(plt.scatter, 'Total_Bilirubin', 'Albumin').add_legend()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "id": "large-madrid",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<seaborn.axisgrid.FacetGrid at 0x1bdf6e2ec48>"
      ]
     },
     "execution_count": 20,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 402.75x360 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "sns.set_style('whitegrid')   ## Background Grid\n",
    "sns.FacetGrid(df, hue = 'Dataset', size = 5).map(plt.scatter, 'Total_Protiens', 'Albumin_and_Globulin_Ratio').add_legend()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "supported-unemployment",
   "metadata": {},
   "source": [
    "## Correlations\n",
    "* Finally, let's take a look at the relationships between numeric features and other numeric features.\n",
    "* ***Correlation*** is a value between -1 and 1 that represents how closely values for two separate features move in unison.\n",
    "* Positive correlation means that as one feature increases, the other increases; eg. a child's age and her height.\n",
    "* Negative correlation means that as one feature increases, the other decreases; eg. hours spent studying and number of parties attended.\n",
    "* Correlations near -1 or 1 indicate a strong relationship.\n",
    "* Those closer to 0 indicate a weak relationship.\n",
    "* 0 indicates no relationship."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "id": "instrumental-vulnerability",
   "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>Age</th>\n",
       "      <th>Gender</th>\n",
       "      <th>Total_Bilirubin</th>\n",
       "      <th>Direct_Bilirubin</th>\n",
       "      <th>Alkaline_Phosphotase</th>\n",
       "      <th>Alamine_Aminotransferase</th>\n",
       "      <th>Aspartate_Aminotransferase</th>\n",
       "      <th>Total_Protiens</th>\n",
       "      <th>Albumin</th>\n",
       "      <th>Albumin_and_Globulin_Ratio</th>\n",
       "      <th>Dataset</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>Age</th>\n",
       "      <td>1.000000</td>\n",
       "      <td>-0.056560</td>\n",
       "      <td>0.011763</td>\n",
       "      <td>0.007529</td>\n",
       "      <td>0.080425</td>\n",
       "      <td>-0.086883</td>\n",
       "      <td>-0.019910</td>\n",
       "      <td>-0.187461</td>\n",
       "      <td>-0.265924</td>\n",
       "      <td>-0.216408</td>\n",
       "      <td>0.137351</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Gender</th>\n",
       "      <td>-0.056560</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>-0.089291</td>\n",
       "      <td>-0.100436</td>\n",
       "      <td>0.027496</td>\n",
       "      <td>-0.082332</td>\n",
       "      <td>-0.080336</td>\n",
       "      <td>0.089121</td>\n",
       "      <td>0.093799</td>\n",
       "      <td>0.003424</td>\n",
       "      <td>-0.082416</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Total_Bilirubin</th>\n",
       "      <td>0.011763</td>\n",
       "      <td>-0.089291</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.874618</td>\n",
       "      <td>0.206669</td>\n",
       "      <td>0.214065</td>\n",
       "      <td>0.237831</td>\n",
       "      <td>-0.008099</td>\n",
       "      <td>-0.222250</td>\n",
       "      <td>-0.206267</td>\n",
       "      <td>0.220208</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Direct_Bilirubin</th>\n",
       "      <td>0.007529</td>\n",
       "      <td>-0.100436</td>\n",
       "      <td>0.874618</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.234939</td>\n",
       "      <td>0.233894</td>\n",
       "      <td>0.257544</td>\n",
       "      <td>-0.000139</td>\n",
       "      <td>-0.228531</td>\n",
       "      <td>-0.200125</td>\n",
       "      <td>0.246046</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Alkaline_Phosphotase</th>\n",
       "      <td>0.080425</td>\n",
       "      <td>0.027496</td>\n",
       "      <td>0.206669</td>\n",
       "      <td>0.234939</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.125680</td>\n",
       "      <td>0.167196</td>\n",
       "      <td>-0.028514</td>\n",
       "      <td>-0.165453</td>\n",
       "      <td>-0.234166</td>\n",
       "      <td>0.184866</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Alamine_Aminotransferase</th>\n",
       "      <td>-0.086883</td>\n",
       "      <td>-0.082332</td>\n",
       "      <td>0.214065</td>\n",
       "      <td>0.233894</td>\n",
       "      <td>0.125680</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.791966</td>\n",
       "      <td>-0.042518</td>\n",
       "      <td>-0.029742</td>\n",
       "      <td>-0.002375</td>\n",
       "      <td>0.163416</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Aspartate_Aminotransferase</th>\n",
       "      <td>-0.019910</td>\n",
       "      <td>-0.080336</td>\n",
       "      <td>0.237831</td>\n",
       "      <td>0.257544</td>\n",
       "      <td>0.167196</td>\n",
       "      <td>0.791966</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>-0.025645</td>\n",
       "      <td>-0.085290</td>\n",
       "      <td>-0.070040</td>\n",
       "      <td>0.151934</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Total_Protiens</th>\n",
       "      <td>-0.187461</td>\n",
       "      <td>0.089121</td>\n",
       "      <td>-0.008099</td>\n",
       "      <td>-0.000139</td>\n",
       "      <td>-0.028514</td>\n",
       "      <td>-0.042518</td>\n",
       "      <td>-0.025645</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.784053</td>\n",
       "      <td>0.234887</td>\n",
       "      <td>-0.035008</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Albumin</th>\n",
       "      <td>-0.265924</td>\n",
       "      <td>0.093799</td>\n",
       "      <td>-0.222250</td>\n",
       "      <td>-0.228531</td>\n",
       "      <td>-0.165453</td>\n",
       "      <td>-0.029742</td>\n",
       "      <td>-0.085290</td>\n",
       "      <td>0.784053</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.689632</td>\n",
       "      <td>-0.161388</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Albumin_and_Globulin_Ratio</th>\n",
       "      <td>-0.216408</td>\n",
       "      <td>0.003424</td>\n",
       "      <td>-0.206267</td>\n",
       "      <td>-0.200125</td>\n",
       "      <td>-0.234166</td>\n",
       "      <td>-0.002375</td>\n",
       "      <td>-0.070040</td>\n",
       "      <td>0.234887</td>\n",
       "      <td>0.689632</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>-0.163131</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Dataset</th>\n",
       "      <td>0.137351</td>\n",
       "      <td>-0.082416</td>\n",
       "      <td>0.220208</td>\n",
       "      <td>0.246046</td>\n",
       "      <td>0.184866</td>\n",
       "      <td>0.163416</td>\n",
       "      <td>0.151934</td>\n",
       "      <td>-0.035008</td>\n",
       "      <td>-0.161388</td>\n",
       "      <td>-0.163131</td>\n",
       "      <td>1.000000</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                                 Age    Gender  Total_Bilirubin  \\\n",
       "Age                         1.000000 -0.056560         0.011763   \n",
       "Gender                     -0.056560  1.000000        -0.089291   \n",
       "Total_Bilirubin             0.011763 -0.089291         1.000000   \n",
       "Direct_Bilirubin            0.007529 -0.100436         0.874618   \n",
       "Alkaline_Phosphotase        0.080425  0.027496         0.206669   \n",
       "Alamine_Aminotransferase   -0.086883 -0.082332         0.214065   \n",
       "Aspartate_Aminotransferase -0.019910 -0.080336         0.237831   \n",
       "Total_Protiens             -0.187461  0.089121        -0.008099   \n",
       "Albumin                    -0.265924  0.093799        -0.222250   \n",
       "Albumin_and_Globulin_Ratio -0.216408  0.003424        -0.206267   \n",
       "Dataset                     0.137351 -0.082416         0.220208   \n",
       "\n",
       "                            Direct_Bilirubin  Alkaline_Phosphotase  \\\n",
       "Age                                 0.007529              0.080425   \n",
       "Gender                             -0.100436              0.027496   \n",
       "Total_Bilirubin                     0.874618              0.206669   \n",
       "Direct_Bilirubin                    1.000000              0.234939   \n",
       "Alkaline_Phosphotase                0.234939              1.000000   \n",
       "Alamine_Aminotransferase            0.233894              0.125680   \n",
       "Aspartate_Aminotransferase          0.257544              0.167196   \n",
       "Total_Protiens                     -0.000139             -0.028514   \n",
       "Albumin                            -0.228531             -0.165453   \n",
       "Albumin_and_Globulin_Ratio         -0.200125             -0.234166   \n",
       "Dataset                             0.246046              0.184866   \n",
       "\n",
       "                            Alamine_Aminotransferase  \\\n",
       "Age                                        -0.086883   \n",
       "Gender                                     -0.082332   \n",
       "Total_Bilirubin                             0.214065   \n",
       "Direct_Bilirubin                            0.233894   \n",
       "Alkaline_Phosphotase                        0.125680   \n",
       "Alamine_Aminotransferase                    1.000000   \n",
       "Aspartate_Aminotransferase                  0.791966   \n",
       "Total_Protiens                             -0.042518   \n",
       "Albumin                                    -0.029742   \n",
       "Albumin_and_Globulin_Ratio                 -0.002375   \n",
       "Dataset                                     0.163416   \n",
       "\n",
       "                            Aspartate_Aminotransferase  Total_Protiens  \\\n",
       "Age                                          -0.019910       -0.187461   \n",
       "Gender                                       -0.080336        0.089121   \n",
       "Total_Bilirubin                               0.237831       -0.008099   \n",
       "Direct_Bilirubin                              0.257544       -0.000139   \n",
       "Alkaline_Phosphotase                          0.167196       -0.028514   \n",
       "Alamine_Aminotransferase                      0.791966       -0.042518   \n",
       "Aspartate_Aminotransferase                    1.000000       -0.025645   \n",
       "Total_Protiens                               -0.025645        1.000000   \n",
       "Albumin                                      -0.085290        0.784053   \n",
       "Albumin_and_Globulin_Ratio                   -0.070040        0.234887   \n",
       "Dataset                                       0.151934       -0.035008   \n",
       "\n",
       "                             Albumin  Albumin_and_Globulin_Ratio   Dataset  \n",
       "Age                        -0.265924                   -0.216408  0.137351  \n",
       "Gender                      0.093799                    0.003424 -0.082416  \n",
       "Total_Bilirubin            -0.222250                   -0.206267  0.220208  \n",
       "Direct_Bilirubin           -0.228531                   -0.200125  0.246046  \n",
       "Alkaline_Phosphotase       -0.165453                   -0.234166  0.184866  \n",
       "Alamine_Aminotransferase   -0.029742                   -0.002375  0.163416  \n",
       "Aspartate_Aminotransferase -0.085290                   -0.070040  0.151934  \n",
       "Total_Protiens              0.784053                    0.234887 -0.035008  \n",
       "Albumin                     1.000000                    0.689632 -0.161388  \n",
       "Albumin_and_Globulin_Ratio  0.689632                    1.000000 -0.163131  \n",
       "Dataset                    -0.161388                   -0.163131  1.000000  "
      ]
     },
     "execution_count": 21,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.corr()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "id": "sweet-clause",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<AxesSubplot:>"
      ]
     },
     "execution_count": 22,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 720x720 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.figure(figsize=(10,10))\n",
    "sns.heatmap(df.corr())"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "id": "representative-hydrogen",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 720x720 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "mask=np.zeros_like(df.corr())\n",
    "mask[np.triu_indices_from(mask)] = True\n",
    "plt.figure(figsize=(10,10))\n",
    "with sns.axes_style(\"white\"):\n",
    "    ax = sns.heatmap(df.corr()*100, mask=mask, fmt='.0f', annot=True, lw=1, cmap=ListedColormap(['green', 'yellow', 'red','blue']))"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "covered-thunder",
   "metadata": {},
   "source": [
    "## Data Cleaning"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "id": "modular-stake",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "(570, 11)\n"
     ]
    }
   ],
   "source": [
    "df = df.drop_duplicates()\n",
    "print( df.shape )"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "devoted-solomon",
   "metadata": {},
   "source": [
    "There were 13 duplicates"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "exterior-bullet",
   "metadata": {},
   "source": [
    "## Removing Outliers"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "id": "waiting-virgin",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<AxesSubplot:xlabel='Aspartate_Aminotransferase'>"
      ]
     },
     "execution_count": 25,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "sns.boxplot(df.Aspartate_Aminotransferase)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "id": "solid-flooring",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "135    4929\n",
       "117    2946\n",
       "118    1600\n",
       "207    1500\n",
       "199    1050\n",
       "Name: Aspartate_Aminotransferase, dtype: int64"
      ]
     },
     "execution_count": 26,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.Aspartate_Aminotransferase.sort_values(ascending=False).head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "id": "relative-mouse",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(569, 11)"
      ]
     },
     "execution_count": 27,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df = df[df.Aspartate_Aminotransferase <=3000 ]\n",
    "df.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "id": "conscious-implementation",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<AxesSubplot:xlabel='Aspartate_Aminotransferase'>"
      ]
     },
     "execution_count": 28,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "sns.boxplot(df.Aspartate_Aminotransferase)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "id": "artistic-midnight",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "117    2946\n",
       "118    1600\n",
       "207    1500\n",
       "119    1050\n",
       "199    1050\n",
       "Name: Aspartate_Aminotransferase, dtype: int64"
      ]
     },
     "execution_count": 29,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.Aspartate_Aminotransferase.sort_values(ascending=False).head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "id": "interpreted-transformation",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(568, 11)"
      ]
     },
     "execution_count": 30,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df = df[df.Aspartate_Aminotransferase <=2500 ]\n",
    "df.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "id": "necessary-forwarding",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "True"
      ]
     },
     "execution_count": 31,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.isnull().values.any()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "id": "laden-eclipse",
   "metadata": {},
   "outputs": [],
   "source": [
    "df=df.dropna(how='any')  "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 33,
   "id": "confused-drinking",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(564, 11)"
      ]
     },
     "execution_count": 33,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 34,
   "id": "prerequisite-species",
   "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>Age</th>\n",
       "      <th>Gender</th>\n",
       "      <th>Total_Bilirubin</th>\n",
       "      <th>Direct_Bilirubin</th>\n",
       "      <th>Alkaline_Phosphotase</th>\n",
       "      <th>Alamine_Aminotransferase</th>\n",
       "      <th>Aspartate_Aminotransferase</th>\n",
       "      <th>Total_Protiens</th>\n",
       "      <th>Albumin</th>\n",
       "      <th>Albumin_and_Globulin_Ratio</th>\n",
       "      <th>Dataset</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>65</td>\n",
       "      <td>1</td>\n",
       "      <td>0.7</td>\n",
       "      <td>0.1</td>\n",
       "      <td>187</td>\n",
       "      <td>16</td>\n",
       "      <td>18</td>\n",
       "      <td>6.8</td>\n",
       "      <td>3.3</td>\n",
       "      <td>0.90</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>62</td>\n",
       "      <td>0</td>\n",
       "      <td>10.9</td>\n",
       "      <td>5.5</td>\n",
       "      <td>699</td>\n",
       "      <td>64</td>\n",
       "      <td>100</td>\n",
       "      <td>7.5</td>\n",
       "      <td>3.2</td>\n",
       "      <td>0.74</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>62</td>\n",
       "      <td>0</td>\n",
       "      <td>7.3</td>\n",
       "      <td>4.1</td>\n",
       "      <td>490</td>\n",
       "      <td>60</td>\n",
       "      <td>68</td>\n",
       "      <td>7.0</td>\n",
       "      <td>3.3</td>\n",
       "      <td>0.89</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>58</td>\n",
       "      <td>0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.4</td>\n",
       "      <td>182</td>\n",
       "      <td>14</td>\n",
       "      <td>20</td>\n",
       "      <td>6.8</td>\n",
       "      <td>3.4</td>\n",
       "      <td>1.00</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>72</td>\n",
       "      <td>0</td>\n",
       "      <td>3.9</td>\n",
       "      <td>2.0</td>\n",
       "      <td>195</td>\n",
       "      <td>27</td>\n",
       "      <td>59</td>\n",
       "      <td>7.3</td>\n",
       "      <td>2.4</td>\n",
       "      <td>0.40</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   Age  Gender  Total_Bilirubin  Direct_Bilirubin  Alkaline_Phosphotase  \\\n",
       "0   65       1              0.7               0.1                   187   \n",
       "1   62       0             10.9               5.5                   699   \n",
       "2   62       0              7.3               4.1                   490   \n",
       "3   58       0              1.0               0.4                   182   \n",
       "4   72       0              3.9               2.0                   195   \n",
       "\n",
       "   Alamine_Aminotransferase  Aspartate_Aminotransferase  Total_Protiens  \\\n",
       "0                        16                          18             6.8   \n",
       "1                        64                         100             7.5   \n",
       "2                        60                          68             7.0   \n",
       "3                        14                          20             6.8   \n",
       "4                        27                          59             7.3   \n",
       "\n",
       "   Albumin  Albumin_and_Globulin_Ratio  Dataset  \n",
       "0      3.3                        0.90        1  \n",
       "1      3.2                        0.74        1  \n",
       "2      3.3                        0.89        1  \n",
       "3      3.4                        1.00        1  \n",
       "4      2.4                        0.40        1  "
      ]
     },
     "execution_count": 34,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "atomic-anderson",
   "metadata": {},
   "source": [
    "## Machine Learning Models"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "incorporated-impact",
   "metadata": {},
   "source": [
    "### Data Preparation"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 35,
   "id": "complete-wealth",
   "metadata": {},
   "outputs": [],
   "source": [
    "# Create separate object for target variable\n",
    "y = df.Dataset\n",
    "\n",
    "# Create separate object for input features\n",
    "X = df.drop('Dataset', axis=1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 36,
   "id": "continuing-panama",
   "metadata": {},
   "outputs": [],
   "source": [
    "# Split X and y into train and test sets\n",
    "X_train, X_test, y_train, y_test = train_test_split(X, y, \n",
    "                                                    test_size=0.2, \n",
    "                                                    random_state=1234,\n",
    "                                                    stratify=df.Dataset)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 37,
   "id": "fantastic-cookie",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "(451, 10) (113, 10) (451,) (113,)\n"
     ]
    }
   ],
   "source": [
    "# Print number of observations in X_train, X_test, y_train, and y_test\n",
    "print(X_train.shape, X_test.shape, y_train.shape, y_test.shape)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "spectacular-tower",
   "metadata": {},
   "source": [
    "### Data standardization\n",
    "* In Data Standardization we perform zero mean centring and unit scaling; i.e. we make the mean of all the features as zero and the standard deviation as 1.\n",
    "* Thus we use **mean** and **standard deviation** of each feature.\n",
    "* It is very important to save the **mean** and **standard deviation** for each of the feature from the **training set**, because we use the same mean and standard deviation in the test set."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 38,
   "id": "confirmed-berlin",
   "metadata": {},
   "outputs": [],
   "source": [
    "train_mean = X_train.mean()\n",
    "train_std = X_train.std()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 39,
   "id": "intimate-syntax",
   "metadata": {},
   "outputs": [],
   "source": [
    "## Standardize the train data set\n",
    "X_train = (X_train - train_mean) / train_std"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 40,
   "id": "above-lighter",
   "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>Age</th>\n",
       "      <th>Gender</th>\n",
       "      <th>Total_Bilirubin</th>\n",
       "      <th>Direct_Bilirubin</th>\n",
       "      <th>Alkaline_Phosphotase</th>\n",
       "      <th>Alamine_Aminotransferase</th>\n",
       "      <th>Aspartate_Aminotransferase</th>\n",
       "      <th>Total_Protiens</th>\n",
       "      <th>Albumin</th>\n",
       "      <th>Albumin_and_Globulin_Ratio</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>count</th>\n",
       "      <td>4.510000e+02</td>\n",
       "      <td>4.510000e+02</td>\n",
       "      <td>4.510000e+02</td>\n",
       "      <td>4.510000e+02</td>\n",
       "      <td>4.510000e+02</td>\n",
       "      <td>4.510000e+02</td>\n",
       "      <td>4.510000e+02</td>\n",
       "      <td>4.510000e+02</td>\n",
       "      <td>4.510000e+02</td>\n",
       "      <td>4.510000e+02</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>mean</th>\n",
       "      <td>1.378547e-16</td>\n",
       "      <td>-8.271285e-17</td>\n",
       "      <td>-3.938707e-18</td>\n",
       "      <td>1.772418e-17</td>\n",
       "      <td>6.301931e-17</td>\n",
       "      <td>-7.877414e-18</td>\n",
       "      <td>-2.363224e-17</td>\n",
       "      <td>1.486862e-15</td>\n",
       "      <td>1.417935e-16</td>\n",
       "      <td>-3.150966e-17</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>std</th>\n",
       "      <td>1.000000e+00</td>\n",
       "      <td>1.000000e+00</td>\n",
       "      <td>1.000000e+00</td>\n",
       "      <td>1.000000e+00</td>\n",
       "      <td>1.000000e+00</td>\n",
       "      <td>1.000000e+00</td>\n",
       "      <td>1.000000e+00</td>\n",
       "      <td>1.000000e+00</td>\n",
       "      <td>1.000000e+00</td>\n",
       "      <td>1.000000e+00</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>min</th>\n",
       "      <td>-2.459710e+00</td>\n",
       "      <td>-5.980292e-01</td>\n",
       "      <td>-4.613674e-01</td>\n",
       "      <td>-5.013974e-01</td>\n",
       "      <td>-9.009611e-01</td>\n",
       "      <td>-3.943772e-01</td>\n",
       "      <td>-4.865614e-01</td>\n",
       "      <td>-3.476087e+00</td>\n",
       "      <td>-2.818030e+00</td>\n",
       "      <td>-2.023543e+00</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>25%</th>\n",
       "      <td>-7.595886e-01</td>\n",
       "      <td>-5.980292e-01</td>\n",
       "      <td>-3.996190e-01</td>\n",
       "      <td>-4.658144e-01</td>\n",
       "      <td>-4.636908e-01</td>\n",
       "      <td>-3.157860e-01</td>\n",
       "      <td>-4.017376e-01</td>\n",
       "      <td>-6.503894e-01</td>\n",
       "      <td>-7.121980e-01</td>\n",
       "      <td>-6.338007e-01</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>50%</th>\n",
       "      <td>2.975347e-02</td>\n",
       "      <td>-5.980292e-01</td>\n",
       "      <td>-3.687448e-01</td>\n",
       "      <td>-4.302314e-01</td>\n",
       "      <td>-3.270438e-01</td>\n",
       "      <td>-2.432402e-01</td>\n",
       "      <td>-3.056039e-01</td>\n",
       "      <td>7.882282e-02</td>\n",
       "      <td>3.103678e-02</td>\n",
       "      <td>1.145223e-01</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>75%</th>\n",
       "      <td>7.583769e-01</td>\n",
       "      <td>1.668451e+00</td>\n",
       "      <td>-1.063141e-01</td>\n",
       "      <td>-7.440091e-02</td>\n",
       "      <td>2.042996e-02</td>\n",
       "      <td>-9.814860e-02</td>\n",
       "      <td>-7.092474e-02</td>\n",
       "      <td>6.257320e-01</td>\n",
       "      <td>8.362078e-01</td>\n",
       "      <td>4.199603e-01</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>max</th>\n",
       "      <td>2.762091e+00</td>\n",
       "      <td>1.668451e+00</td>\n",
       "      <td>1.105471e+01</td>\n",
       "      <td>6.472879e+00</td>\n",
       "      <td>7.090935e+00</td>\n",
       "      <td>9.701578e+00</td>\n",
       "      <td>8.504762e+00</td>\n",
       "      <td>2.813368e+00</td>\n",
       "      <td>2.880103e+00</td>\n",
       "      <td>5.612406e+00</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                Age        Gender  Total_Bilirubin  Direct_Bilirubin  \\\n",
       "count  4.510000e+02  4.510000e+02     4.510000e+02      4.510000e+02   \n",
       "mean   1.378547e-16 -8.271285e-17    -3.938707e-18      1.772418e-17   \n",
       "std    1.000000e+00  1.000000e+00     1.000000e+00      1.000000e+00   \n",
       "min   -2.459710e+00 -5.980292e-01    -4.613674e-01     -5.013974e-01   \n",
       "25%   -7.595886e-01 -5.980292e-01    -3.996190e-01     -4.658144e-01   \n",
       "50%    2.975347e-02 -5.980292e-01    -3.687448e-01     -4.302314e-01   \n",
       "75%    7.583769e-01  1.668451e+00    -1.063141e-01     -7.440091e-02   \n",
       "max    2.762091e+00  1.668451e+00     1.105471e+01      6.472879e+00   \n",
       "\n",
       "       Alkaline_Phosphotase  Alamine_Aminotransferase  \\\n",
       "count          4.510000e+02              4.510000e+02   \n",
       "mean           6.301931e-17             -7.877414e-18   \n",
       "std            1.000000e+00              1.000000e+00   \n",
       "min           -9.009611e-01             -3.943772e-01   \n",
       "25%           -4.636908e-01             -3.157860e-01   \n",
       "50%           -3.270438e-01             -2.432402e-01   \n",
       "75%            2.042996e-02             -9.814860e-02   \n",
       "max            7.090935e+00              9.701578e+00   \n",
       "\n",
       "       Aspartate_Aminotransferase  Total_Protiens       Albumin  \\\n",
       "count                4.510000e+02    4.510000e+02  4.510000e+02   \n",
       "mean                -2.363224e-17    1.486862e-15  1.417935e-16   \n",
       "std                  1.000000e+00    1.000000e+00  1.000000e+00   \n",
       "min                 -4.865614e-01   -3.476087e+00 -2.818030e+00   \n",
       "25%                 -4.017376e-01   -6.503894e-01 -7.121980e-01   \n",
       "50%                 -3.056039e-01    7.882282e-02  3.103678e-02   \n",
       "75%                 -7.092474e-02    6.257320e-01  8.362078e-01   \n",
       "max                  8.504762e+00    2.813368e+00  2.880103e+00   \n",
       "\n",
       "       Albumin_and_Globulin_Ratio  \n",
       "count                4.510000e+02  \n",
       "mean                -3.150966e-17  \n",
       "std                  1.000000e+00  \n",
       "min                 -2.023543e+00  \n",
       "25%                 -6.338007e-01  \n",
       "50%                  1.145223e-01  \n",
       "75%                  4.199603e-01  \n",
       "max                  5.612406e+00  "
      ]
     },
     "execution_count": 40,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "## Check for mean and std dev.\n",
    "X_train.describe()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 41,
   "id": "facial-covering",
   "metadata": {},
   "outputs": [],
   "source": [
    "## Note: We use train_mean and train_std_dev to standardize test data set\n",
    "X_test = (X_test - train_mean) / train_std"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 42,
   "id": "arctic-democrat",
   "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>Age</th>\n",
       "      <th>Gender</th>\n",
       "      <th>Total_Bilirubin</th>\n",
       "      <th>Direct_Bilirubin</th>\n",
       "      <th>Alkaline_Phosphotase</th>\n",
       "      <th>Alamine_Aminotransferase</th>\n",
       "      <th>Aspartate_Aminotransferase</th>\n",
       "      <th>Total_Protiens</th>\n",
       "      <th>Albumin</th>\n",
       "      <th>Albumin_and_Globulin_Ratio</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>count</th>\n",
       "      <td>113.000000</td>\n",
       "      <td>113.000000</td>\n",
       "      <td>113.000000</td>\n",
       "      <td>113.000000</td>\n",
       "      <td>113.000000</td>\n",
       "      <td>113.000000</td>\n",
       "      <td>113.000000</td>\n",
       "      <td>113.000000</td>\n",
       "      <td>113.000000</td>\n",
       "      <td>113.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>mean</th>\n",
       "      <td>0.109816</td>\n",
       "      <td>-0.216940</td>\n",
       "      <td>-0.062052</td>\n",
       "      <td>-0.033465</td>\n",
       "      <td>-0.048222</td>\n",
       "      <td>-0.017203</td>\n",
       "      <td>0.007819</td>\n",
       "      <td>-0.082507</td>\n",
       "      <td>-0.175052</td>\n",
       "      <td>-0.218486</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>std</th>\n",
       "      <td>0.937258</td>\n",
       "      <td>0.851421</td>\n",
       "      <td>0.835512</td>\n",
       "      <td>1.046110</td>\n",
       "      <td>0.734043</td>\n",
       "      <td>0.661163</td>\n",
       "      <td>0.874507</td>\n",
       "      <td>0.962887</td>\n",
       "      <td>0.922474</td>\n",
       "      <td>0.851370</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>min</th>\n",
       "      <td>-2.338273</td>\n",
       "      <td>-0.598029</td>\n",
       "      <td>-0.445930</td>\n",
       "      <td>-0.501397</td>\n",
       "      <td>-0.768218</td>\n",
       "      <td>-0.394377</td>\n",
       "      <td>-0.469597</td>\n",
       "      <td>-2.655723</td>\n",
       "      <td>-2.198667</td>\n",
       "      <td>-1.718105</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>25%</th>\n",
       "      <td>-0.577433</td>\n",
       "      <td>-0.598029</td>\n",
       "      <td>-0.399619</td>\n",
       "      <td>-0.465814</td>\n",
       "      <td>-0.444170</td>\n",
       "      <td>-0.321831</td>\n",
       "      <td>-0.401738</td>\n",
       "      <td>-0.741541</td>\n",
       "      <td>-0.836070</td>\n",
       "      <td>-0.801792</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>50%</th>\n",
       "      <td>0.090472</td>\n",
       "      <td>-0.598029</td>\n",
       "      <td>-0.384182</td>\n",
       "      <td>-0.430231</td>\n",
       "      <td>-0.342661</td>\n",
       "      <td>-0.261377</td>\n",
       "      <td>-0.316914</td>\n",
       "      <td>-0.012329</td>\n",
       "      <td>-0.216708</td>\n",
       "      <td>-0.190916</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>75%</th>\n",
       "      <td>0.940533</td>\n",
       "      <td>-0.598029</td>\n",
       "      <td>-0.198937</td>\n",
       "      <td>-0.181150</td>\n",
       "      <td>-0.018612</td>\n",
       "      <td>-0.007466</td>\n",
       "      <td>-0.034168</td>\n",
       "      <td>0.625732</td>\n",
       "      <td>0.526527</td>\n",
       "      <td>0.114522</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>max</th>\n",
       "      <td>1.851312</td>\n",
       "      <td>1.668451</td>\n",
       "      <td>4.231512</td>\n",
       "      <td>5.974717</td>\n",
       "      <td>3.147694</td>\n",
       "      <td>4.834965</td>\n",
       "      <td>4.263572</td>\n",
       "      <td>2.175308</td>\n",
       "      <td>2.136869</td>\n",
       "      <td>2.558026</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "              Age      Gender  Total_Bilirubin  Direct_Bilirubin  \\\n",
       "count  113.000000  113.000000       113.000000        113.000000   \n",
       "mean     0.109816   -0.216940        -0.062052         -0.033465   \n",
       "std      0.937258    0.851421         0.835512          1.046110   \n",
       "min     -2.338273   -0.598029        -0.445930         -0.501397   \n",
       "25%     -0.577433   -0.598029        -0.399619         -0.465814   \n",
       "50%      0.090472   -0.598029        -0.384182         -0.430231   \n",
       "75%      0.940533   -0.598029        -0.198937         -0.181150   \n",
       "max      1.851312    1.668451         4.231512          5.974717   \n",
       "\n",
       "       Alkaline_Phosphotase  Alamine_Aminotransferase  \\\n",
       "count            113.000000                113.000000   \n",
       "mean              -0.048222                 -0.017203   \n",
       "std                0.734043                  0.661163   \n",
       "min               -0.768218                 -0.394377   \n",
       "25%               -0.444170                 -0.321831   \n",
       "50%               -0.342661                 -0.261377   \n",
       "75%               -0.018612                 -0.007466   \n",
       "max                3.147694                  4.834965   \n",
       "\n",
       "       Aspartate_Aminotransferase  Total_Protiens     Albumin  \\\n",
       "count                  113.000000      113.000000  113.000000   \n",
       "mean                     0.007819       -0.082507   -0.175052   \n",
       "std                      0.874507        0.962887    0.922474   \n",
       "min                     -0.469597       -2.655723   -2.198667   \n",
       "25%                     -0.401738       -0.741541   -0.836070   \n",
       "50%                     -0.316914       -0.012329   -0.216708   \n",
       "75%                     -0.034168        0.625732    0.526527   \n",
       "max                      4.263572        2.175308    2.136869   \n",
       "\n",
       "       Albumin_and_Globulin_Ratio  \n",
       "count                  113.000000  \n",
       "mean                    -0.218486  \n",
       "std                      0.851370  \n",
       "min                     -1.718105  \n",
       "25%                     -0.801792  \n",
       "50%                     -0.190916  \n",
       "75%                      0.114522  \n",
       "max                      2.558026  "
      ]
     },
     "execution_count": 42,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "## Check for mean and std dev. - not exactly 0 and 1\n",
    "X_test.describe()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "veterinary-penguin",
   "metadata": {},
   "source": [
    "##  Model-1 Logistic Regression"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 43,
   "id": "broken-mileage",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "GridSearchCV(estimator=LogisticRegression(), n_jobs=-1,\n",
       "             param_grid={'C': [0.0001, 0.001, 0.01, 0.1, 1, 10, 100, 1000,\n",
       "                               10000],\n",
       "                         'penalty': ['l1', 'l2']},\n",
       "             scoring='roc_auc')"
      ]
     },
     "execution_count": 43,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "tuned_params = {'C': [0.0001, 0.001, 0.01, 0.1, 1, 10, 100, 1000, 10000], 'penalty': ['l1', 'l2']}\n",
    "model = GridSearchCV(LogisticRegression(), tuned_params, scoring = 'roc_auc', n_jobs=-1)\n",
    "model.fit(X_train, y_train)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 44,
   "id": "democratic-metadata",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "LogisticRegression(C=0.1)"
      ]
     },
     "execution_count": 44,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "model.best_estimator_"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 45,
   "id": "cooperative-digest",
   "metadata": {},
   "outputs": [],
   "source": [
    "## Predict Train set results\n",
    "y_train_pred = model.predict(X_train)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 46,
   "id": "coordinate-indiana",
   "metadata": {},
   "outputs": [],
   "source": [
    "## Predict Test set results\n",
    "y_pred = model.predict(X_test)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 47,
   "id": "laughing-startup",
   "metadata": {},
   "outputs": [],
   "source": [
    "# Get just the prediction for the positive class (1)\n",
    "y_pred_proba = model.predict_proba(X_test)[:,1]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 48,
   "id": "purple-generic",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([0.62886101, 0.60635075, 0.65628802, 0.69577814, 0.45236153,\n",
       "       0.97879646, 0.56694329, 0.67530199, 0.79401715, 0.50727069])"
      ]
     },
     "execution_count": 48,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Display first 10 predictions\n",
    "y_pred_proba[:10]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 49,
   "id": "romantic-virus",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "For test point 28, actual class = 1, precited class = 1, predicted probability = 0.8175521111565488\n"
     ]
    }
   ],
   "source": [
    "i=28  ## Change the value of i to get the details of any point (56, 213, etc.)\n",
    "print('For test point {}, actual class = {}, precited class = {}, predicted probability = {}'.\n",
    "      format(i, y_test.iloc[i], y_pred[i], y_pred_proba[i]))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 50,
   "id": "damaged-metabolism",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([[ 1,  4],\n",
       "       [31, 77]], dtype=int64)"
      ]
     },
     "execution_count": 50,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "confusion_matrix(y_test, y_pred).T"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 51,
   "id": "false-arctic",
   "metadata": {},
   "outputs": [],
   "source": [
    "# Calculate ROC curve from y_test and pred\n",
    "fpr, tpr, thresholds = roc_curve(y_test, y_pred_proba)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 52,
   "id": "experimental-default",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 576x576 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Plot the ROC curve\n",
    "fig = plt.figure(figsize=(8,8))\n",
    "plt.title('Receiver Operating Characteristic')\n",
    "\n",
    "# Plot ROC curve\n",
    "plt.plot(fpr, tpr, label='l1')\n",
    "plt.legend(loc='lower right')\n",
    "\n",
    "# Diagonal 45 degree line\n",
    "plt.plot([0,1],[0,1],'k--')\n",
    "\n",
    "# Axes limits and labels\n",
    "plt.xlim([-0.1,1.1])\n",
    "plt.ylim([-0.1,1.1])\n",
    "plt.ylabel('True Positive Rate')\n",
    "plt.xlabel('False Positive Rate')\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 53,
   "id": "designing-liverpool",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "0.5380781212556913\n"
     ]
    }
   ],
   "source": [
    "# Calculate AUC for Train set\n",
    "print(roc_auc_score(y_train, y_train_pred))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 54,
   "id": "younger-elevation",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "0.6940586419753086\n"
     ]
    }
   ],
   "source": [
    "# Calculate AUC for Test set\n",
    "print(auc(fpr, tpr))"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "chemical-array",
   "metadata": {},
   "source": [
    "#### Feature Importance"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 55,
   "id": "elder-comparison",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "LogisticRegression(C=1)"
      ]
     },
     "execution_count": 55,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "## Building the model again with the best hyperparameters\n",
    "model = LogisticRegression(C=1, penalty = 'l2')\n",
    "model.fit(X_train, y_train)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 56,
   "id": "severe-albany",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "The features in order of importance are:\n",
      "--------------------------------------------------\n",
      "Alamine_Aminotransferase\n",
      "Direct_Bilirubin\n",
      "Aspartate_Aminotransferase\n",
      "Albumin\n",
      "Total_Protiens\n",
      "Total_Bilirubin\n",
      "Alkaline_Phosphotase\n",
      "Age\n",
      "Gender\n",
      "Albumin_and_Globulin_Ratio\n"
     ]
    }
   ],
   "source": [
    "indices = np.argsort(-abs(model.coef_[0,:]))\n",
    "print(\"The features in order of importance are:\")\n",
    "print(50*'-')\n",
    "for feature in X.columns[indices]:\n",
    "    print(feature)"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.7.6"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}