Healthcare Analytics.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Healthcare Analytics\n",
    "### The goal of this project is to accurately predict the Length of Stay for each patient so that the hospitals can optimize resources and function better."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "import numpy as np\n",
    "import pandas as pd\n",
    "import matplotlib.pyplot as plt\n",
    "np.set_printoptions(suppress=True)\n",
    "import warnings\n",
    "warnings.filterwarnings('ignore')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Importing datasets\n",
    "train = pd.read_csv('train.csv')\n",
    "test = pd.read_csv('test.csv')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Data Exploration"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Overview of Data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "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>case_id</th>\n",
       "      <th>Hospital_code</th>\n",
       "      <th>Hospital_type_code</th>\n",
       "      <th>City_Code_Hospital</th>\n",
       "      <th>Hospital_region_code</th>\n",
       "      <th>Available Extra Rooms in Hospital</th>\n",
       "      <th>Department</th>\n",
       "      <th>Ward_Type</th>\n",
       "      <th>Ward_Facility_Code</th>\n",
       "      <th>Bed Grade</th>\n",
       "      <th>patientid</th>\n",
       "      <th>City_Code_Patient</th>\n",
       "      <th>Type of Admission</th>\n",
       "      <th>Severity of Illness</th>\n",
       "      <th>Visitors with Patient</th>\n",
       "      <th>Age</th>\n",
       "      <th>Admission_Deposit</th>\n",
       "      <th>Stay</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>1</td>\n",
       "      <td>8</td>\n",
       "      <td>c</td>\n",
       "      <td>3</td>\n",
       "      <td>Z</td>\n",
       "      <td>3</td>\n",
       "      <td>radiotherapy</td>\n",
       "      <td>R</td>\n",
       "      <td>F</td>\n",
       "      <td>2.0</td>\n",
       "      <td>31397</td>\n",
       "      <td>7.0</td>\n",
       "      <td>Emergency</td>\n",
       "      <td>Extreme</td>\n",
       "      <td>2</td>\n",
       "      <td>51-60</td>\n",
       "      <td>4911.0</td>\n",
       "      <td>0-10</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>2</td>\n",
       "      <td>2</td>\n",
       "      <td>c</td>\n",
       "      <td>5</td>\n",
       "      <td>Z</td>\n",
       "      <td>2</td>\n",
       "      <td>radiotherapy</td>\n",
       "      <td>S</td>\n",
       "      <td>F</td>\n",
       "      <td>2.0</td>\n",
       "      <td>31397</td>\n",
       "      <td>7.0</td>\n",
       "      <td>Trauma</td>\n",
       "      <td>Extreme</td>\n",
       "      <td>2</td>\n",
       "      <td>51-60</td>\n",
       "      <td>5954.0</td>\n",
       "      <td>41-50</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>3</td>\n",
       "      <td>10</td>\n",
       "      <td>e</td>\n",
       "      <td>1</td>\n",
       "      <td>X</td>\n",
       "      <td>2</td>\n",
       "      <td>anesthesia</td>\n",
       "      <td>S</td>\n",
       "      <td>E</td>\n",
       "      <td>2.0</td>\n",
       "      <td>31397</td>\n",
       "      <td>7.0</td>\n",
       "      <td>Trauma</td>\n",
       "      <td>Extreme</td>\n",
       "      <td>2</td>\n",
       "      <td>51-60</td>\n",
       "      <td>4745.0</td>\n",
       "      <td>31-40</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>4</td>\n",
       "      <td>26</td>\n",
       "      <td>b</td>\n",
       "      <td>2</td>\n",
       "      <td>Y</td>\n",
       "      <td>2</td>\n",
       "      <td>radiotherapy</td>\n",
       "      <td>R</td>\n",
       "      <td>D</td>\n",
       "      <td>2.0</td>\n",
       "      <td>31397</td>\n",
       "      <td>7.0</td>\n",
       "      <td>Trauma</td>\n",
       "      <td>Extreme</td>\n",
       "      <td>2</td>\n",
       "      <td>51-60</td>\n",
       "      <td>7272.0</td>\n",
       "      <td>41-50</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>5</td>\n",
       "      <td>26</td>\n",
       "      <td>b</td>\n",
       "      <td>2</td>\n",
       "      <td>Y</td>\n",
       "      <td>2</td>\n",
       "      <td>radiotherapy</td>\n",
       "      <td>S</td>\n",
       "      <td>D</td>\n",
       "      <td>2.0</td>\n",
       "      <td>31397</td>\n",
       "      <td>7.0</td>\n",
       "      <td>Trauma</td>\n",
       "      <td>Extreme</td>\n",
       "      <td>2</td>\n",
       "      <td>51-60</td>\n",
       "      <td>5558.0</td>\n",
       "      <td>41-50</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   case_id  Hospital_code Hospital_type_code  City_Code_Hospital  \\\n",
       "0        1              8                  c                   3   \n",
       "1        2              2                  c                   5   \n",
       "2        3             10                  e                   1   \n",
       "3        4             26                  b                   2   \n",
       "4        5             26                  b                   2   \n",
       "\n",
       "  Hospital_region_code  Available Extra Rooms in Hospital    Department  \\\n",
       "0                    Z                                  3  radiotherapy   \n",
       "1                    Z                                  2  radiotherapy   \n",
       "2                    X                                  2    anesthesia   \n",
       "3                    Y                                  2  radiotherapy   \n",
       "4                    Y                                  2  radiotherapy   \n",
       "\n",
       "  Ward_Type Ward_Facility_Code  Bed Grade  patientid  City_Code_Patient  \\\n",
       "0         R                  F        2.0      31397                7.0   \n",
       "1         S                  F        2.0      31397                7.0   \n",
       "2         S                  E        2.0      31397                7.0   \n",
       "3         R                  D        2.0      31397                7.0   \n",
       "4         S                  D        2.0      31397                7.0   \n",
       "\n",
       "  Type of Admission Severity of Illness  Visitors with Patient    Age  \\\n",
       "0         Emergency             Extreme                      2  51-60   \n",
       "1            Trauma             Extreme                      2  51-60   \n",
       "2            Trauma             Extreme                      2  51-60   \n",
       "3            Trauma             Extreme                      2  51-60   \n",
       "4            Trauma             Extreme                      2  51-60   \n",
       "\n",
       "   Admission_Deposit   Stay  \n",
       "0             4911.0   0-10  \n",
       "1             5954.0  41-50  \n",
       "2             4745.0  31-40  \n",
       "3             7272.0  41-50  \n",
       "4             5558.0  41-50  "
      ]
     },
     "execution_count": 3,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "train.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "<class 'pandas.core.frame.DataFrame'>\n",
      "RangeIndex: 318438 entries, 0 to 318437\n",
      "Data columns (total 18 columns):\n",
      " #   Column                             Non-Null Count   Dtype  \n",
      "---  ------                             --------------   -----  \n",
      " 0   case_id                            318438 non-null  int64  \n",
      " 1   Hospital_code                      318438 non-null  int64  \n",
      " 2   Hospital_type_code                 318438 non-null  object \n",
      " 3   City_Code_Hospital                 318438 non-null  int64  \n",
      " 4   Hospital_region_code               318438 non-null  object \n",
      " 5   Available Extra Rooms in Hospital  318438 non-null  int64  \n",
      " 6   Department                         318438 non-null  object \n",
      " 7   Ward_Type                          318438 non-null  object \n",
      " 8   Ward_Facility_Code                 318438 non-null  object \n",
      " 9   Bed Grade                          318325 non-null  float64\n",
      " 10  patientid                          318438 non-null  int64  \n",
      " 11  City_Code_Patient                  313906 non-null  float64\n",
      " 12  Type of Admission                  318438 non-null  object \n",
      " 13  Severity of Illness                318438 non-null  object \n",
      " 14  Visitors with Patient              318438 non-null  int64  \n",
      " 15  Age                                318438 non-null  object \n",
      " 16  Admission_Deposit                  318438 non-null  float64\n",
      " 17  Stay                               318438 non-null  object \n",
      "dtypes: float64(3), int64(6), object(9)\n",
      "memory usage: 43.7+ MB\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "array(['0-10', '41-50', '31-40', '11-20', '51-60', '21-30', '71-80',\n",
       "       'More than 100 Days', '81-90', '61-70', '91-100'], dtype=object)"
      ]
     },
     "execution_count": 4,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "train.info()\n",
    "train.Stay.unique()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "City_Code_Patient                    4532\n",
       "Bed Grade                             113\n",
       "Stay                                    0\n",
       "Ward_Type                               0\n",
       "Hospital_code                           0\n",
       "Hospital_type_code                      0\n",
       "City_Code_Hospital                      0\n",
       "Hospital_region_code                    0\n",
       "Available Extra Rooms in Hospital       0\n",
       "Department                              0\n",
       "Ward_Facility_Code                      0\n",
       "Admission_Deposit                       0\n",
       "patientid                               0\n",
       "Type of Admission                       0\n",
       "Severity of Illness                     0\n",
       "Visitors with Patient                   0\n",
       "Age                                     0\n",
       "case_id                                 0\n",
       "dtype: int64"
      ]
     },
     "execution_count": 5,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# NA values in train dataset\n",
    "train.isnull().sum().sort_values(ascending = False)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "City_Code_Patient                    2157\n",
       "Bed Grade                              35\n",
       "Admission_Deposit                       0\n",
       "Department                              0\n",
       "Hospital_code                           0\n",
       "Hospital_type_code                      0\n",
       "City_Code_Hospital                      0\n",
       "Hospital_region_code                    0\n",
       "Available Extra Rooms in Hospital       0\n",
       "Ward_Facility_Code                      0\n",
       "Ward_Type                               0\n",
       "Age                                     0\n",
       "patientid                               0\n",
       "Type of Admission                       0\n",
       "Severity of Illness                     0\n",
       "Visitors with Patient                   0\n",
       "case_id                                 0\n",
       "dtype: int64"
      ]
     },
     "execution_count": 6,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# NA values in test dataset\n",
    "test.isnull().sum().sort_values(ascending = False)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(318438, 18)"
      ]
     },
     "execution_count": 7,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Dimension of train dataset\n",
    "train.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(137057, 17)"
      ]
     },
     "execution_count": 8,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Dimension of test dataset\n",
    "test.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "case_id : 318438\n",
      "Hospital_code : 32\n",
      "Hospital_type_code : 7\n",
      "City_Code_Hospital : 11\n",
      "Hospital_region_code : 3\n",
      "Available Extra Rooms in Hospital : 18\n",
      "Department : 5\n",
      "Ward_Type : 6\n",
      "Ward_Facility_Code : 6\n",
      "Bed Grade : 4\n",
      "patientid : 92017\n",
      "City_Code_Patient : 37\n",
      "Type of Admission : 3\n",
      "Severity of Illness : 3\n",
      "Visitors with Patient : 28\n",
      "Age : 10\n",
      "Admission_Deposit : 7300\n",
      "Stay : 11\n"
     ]
    }
   ],
   "source": [
    "# Number of distinct observations in train dataset \n",
    "for i in train.columns:\n",
    "    print(i, ':', train[i].nunique())"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "case_id : 137057\n",
      "Hospital_code : 32\n",
      "Hospital_type_code : 7\n",
      "City_Code_Hospital : 11\n",
      "Hospital_region_code : 3\n",
      "Available Extra Rooms in Hospital : 15\n",
      "Department : 5\n",
      "Ward_Type : 6\n",
      "Ward_Facility_Code : 6\n",
      "Bed Grade : 4\n",
      "patientid : 39607\n",
      "City_Code_Patient : 37\n",
      "Type of Admission : 3\n",
      "Severity of Illness : 3\n",
      "Visitors with Patient : 27\n",
      "Age : 10\n",
      "Admission_Deposit : 6609\n"
     ]
    }
   ],
   "source": [
    "# Number of distinct observations in test dataset\n",
    "for i in test.columns:\n",
    "    print(i, ':', test[i].nunique())"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Data Preparation"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [],
   "source": [
    "#Replacing NA values in Bed Grade Column for both Train and Test datssets\n",
    "train['Bed Grade'].fillna(train['Bed Grade'].mode()[0], inplace = True)\n",
    "test['Bed Grade'].fillna(test['Bed Grade'].mode()[0], inplace = True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [],
   "source": [
    "#Replacing NA values in  Column for both Train and Test datssets\n",
    "train['City_Code_Patient'].fillna(train['City_Code_Patient'].mode()[0], inplace = True)\n",
    "test['City_Code_Patient'].fillna(test['City_Code_Patient'].mode()[0], inplace = True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Label Encoding Stay column in train dataset\n",
    "from sklearn.preprocessing import LabelEncoder\n",
    "le = LabelEncoder()\n",
    "train['Stay'] = le.fit_transform(train['Stay'].astype('str'))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "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>case_id</th>\n",
       "      <th>Hospital_code</th>\n",
       "      <th>Hospital_type_code</th>\n",
       "      <th>City_Code_Hospital</th>\n",
       "      <th>Hospital_region_code</th>\n",
       "      <th>Available Extra Rooms in Hospital</th>\n",
       "      <th>Department</th>\n",
       "      <th>Ward_Type</th>\n",
       "      <th>Ward_Facility_Code</th>\n",
       "      <th>Bed Grade</th>\n",
       "      <th>patientid</th>\n",
       "      <th>City_Code_Patient</th>\n",
       "      <th>Type of Admission</th>\n",
       "      <th>Severity of Illness</th>\n",
       "      <th>Visitors with Patient</th>\n",
       "      <th>Age</th>\n",
       "      <th>Admission_Deposit</th>\n",
       "      <th>Stay</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>1</td>\n",
       "      <td>8</td>\n",
       "      <td>c</td>\n",
       "      <td>3</td>\n",
       "      <td>Z</td>\n",
       "      <td>3</td>\n",
       "      <td>radiotherapy</td>\n",
       "      <td>R</td>\n",
       "      <td>F</td>\n",
       "      <td>2.0</td>\n",
       "      <td>31397</td>\n",
       "      <td>7.0</td>\n",
       "      <td>Emergency</td>\n",
       "      <td>Extreme</td>\n",
       "      <td>2</td>\n",
       "      <td>51-60</td>\n",
       "      <td>4911.0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>2</td>\n",
       "      <td>2</td>\n",
       "      <td>c</td>\n",
       "      <td>5</td>\n",
       "      <td>Z</td>\n",
       "      <td>2</td>\n",
       "      <td>radiotherapy</td>\n",
       "      <td>S</td>\n",
       "      <td>F</td>\n",
       "      <td>2.0</td>\n",
       "      <td>31397</td>\n",
       "      <td>7.0</td>\n",
       "      <td>Trauma</td>\n",
       "      <td>Extreme</td>\n",
       "      <td>2</td>\n",
       "      <td>51-60</td>\n",
       "      <td>5954.0</td>\n",
       "      <td>4</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>3</td>\n",
       "      <td>10</td>\n",
       "      <td>e</td>\n",
       "      <td>1</td>\n",
       "      <td>X</td>\n",
       "      <td>2</td>\n",
       "      <td>anesthesia</td>\n",
       "      <td>S</td>\n",
       "      <td>E</td>\n",
       "      <td>2.0</td>\n",
       "      <td>31397</td>\n",
       "      <td>7.0</td>\n",
       "      <td>Trauma</td>\n",
       "      <td>Extreme</td>\n",
       "      <td>2</td>\n",
       "      <td>51-60</td>\n",
       "      <td>4745.0</td>\n",
       "      <td>3</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>4</td>\n",
       "      <td>26</td>\n",
       "      <td>b</td>\n",
       "      <td>2</td>\n",
       "      <td>Y</td>\n",
       "      <td>2</td>\n",
       "      <td>radiotherapy</td>\n",
       "      <td>R</td>\n",
       "      <td>D</td>\n",
       "      <td>2.0</td>\n",
       "      <td>31397</td>\n",
       "      <td>7.0</td>\n",
       "      <td>Trauma</td>\n",
       "      <td>Extreme</td>\n",
       "      <td>2</td>\n",
       "      <td>51-60</td>\n",
       "      <td>7272.0</td>\n",
       "      <td>4</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>5</td>\n",
       "      <td>26</td>\n",
       "      <td>b</td>\n",
       "      <td>2</td>\n",
       "      <td>Y</td>\n",
       "      <td>2</td>\n",
       "      <td>radiotherapy</td>\n",
       "      <td>S</td>\n",
       "      <td>D</td>\n",
       "      <td>2.0</td>\n",
       "      <td>31397</td>\n",
       "      <td>7.0</td>\n",
       "      <td>Trauma</td>\n",
       "      <td>Extreme</td>\n",
       "      <td>2</td>\n",
       "      <td>51-60</td>\n",
       "      <td>5558.0</td>\n",
       "      <td>4</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   case_id  Hospital_code Hospital_type_code  City_Code_Hospital  \\\n",
       "0        1              8                  c                   3   \n",
       "1        2              2                  c                   5   \n",
       "2        3             10                  e                   1   \n",
       "3        4             26                  b                   2   \n",
       "4        5             26                  b                   2   \n",
       "\n",
       "  Hospital_region_code  Available Extra Rooms in Hospital    Department  \\\n",
       "0                    Z                                  3  radiotherapy   \n",
       "1                    Z                                  2  radiotherapy   \n",
       "2                    X                                  2    anesthesia   \n",
       "3                    Y                                  2  radiotherapy   \n",
       "4                    Y                                  2  radiotherapy   \n",
       "\n",
       "  Ward_Type Ward_Facility_Code  Bed Grade  patientid  City_Code_Patient  \\\n",
       "0         R                  F        2.0      31397                7.0   \n",
       "1         S                  F        2.0      31397                7.0   \n",
       "2         S                  E        2.0      31397                7.0   \n",
       "3         R                  D        2.0      31397                7.0   \n",
       "4         S                  D        2.0      31397                7.0   \n",
       "\n",
       "  Type of Admission Severity of Illness  Visitors with Patient    Age  \\\n",
       "0         Emergency             Extreme                      2  51-60   \n",
       "1            Trauma             Extreme                      2  51-60   \n",
       "2            Trauma             Extreme                      2  51-60   \n",
       "3            Trauma             Extreme                      2  51-60   \n",
       "4            Trauma             Extreme                      2  51-60   \n",
       "\n",
       "   Admission_Deposit  Stay  \n",
       "0             4911.0     0  \n",
       "1             5954.0     4  \n",
       "2             4745.0     3  \n",
       "3             7272.0     4  \n",
       "4             5558.0     4  "
      ]
     },
     "execution_count": 14,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "train.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(455495, 18)"
      ]
     },
     "execution_count": 15,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#Imputing dummy Stay column in test datset to concatenate with train dataset\n",
    "test['Stay'] = -1\n",
    "df = pd.concat([train, test])\n",
    "df.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {},
   "outputs": [],
   "source": [
    "#Label Encoding all the columns in Train and test datasets\n",
    "for i in ['Hospital_type_code', 'Hospital_region_code', 'Department',\n",
    "          'Ward_Type', 'Ward_Facility_Code', 'Type of Admission', 'Severity of Illness', 'Age']:\n",
    "    le = LabelEncoder()\n",
    "    df[i] = le.fit_transform(df[i].astype(str))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {},
   "outputs": [],
   "source": [
    "#Spearating Train and Test Datasets\n",
    "train = df[df['Stay']!=-1]\n",
    "test = df[df['Stay']==-1]"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Feature Engineering"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {},
   "outputs": [],
   "source": [
    "def get_countid_enocde(train, test, cols, name):\n",
    "  temp = train.groupby(cols)['case_id'].count().reset_index().rename(columns = {'case_id': name})\n",
    "  temp2 = test.groupby(cols)['case_id'].count().reset_index().rename(columns = {'case_id': name})\n",
    "  train = pd.merge(train, temp, how='left', on= cols)\n",
    "  test = pd.merge(test,temp2, how='left', on= cols)\n",
    "  train[name] = train[name].astype('float')\n",
    "  test[name] = test[name].astype('float')\n",
    "  train[name].fillna(np.median(temp[name]), inplace = True)\n",
    "  test[name].fillna(np.median(temp2[name]), inplace = True)\n",
    "  return train, test"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {},
   "outputs": [],
   "source": [
    "train, test = get_countid_enocde(train, test, ['patientid'], name = 'count_id_patient')\n",
    "train, test = get_countid_enocde(train, test, \n",
    "                                 ['patientid', 'Hospital_region_code'], name = 'count_id_patient_hospitalCode')\n",
    "train, test = get_countid_enocde(train, test, \n",
    "                                 ['patientid', 'Ward_Facility_Code'], name = 'count_id_patient_wardfacilityCode')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Droping duplicate columns\n",
    "test1 = test.drop(['Stay', 'patientid', 'Hospital_region_code', 'Ward_Facility_Code'], axis =1)\n",
    "train1 = train.drop(['case_id', 'patientid', 'Hospital_region_code', 'Ward_Facility_Code'], axis =1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Splitting train data for Naive Bayes and XGBoost\n",
    "X1 = train1.drop('Stay', axis =1)\n",
    "y1 = train1['Stay']\n",
    "from sklearn.model_selection import train_test_split\n",
    "X_train, X_test, y_train, y_test = train_test_split(X1, y1, test_size =0.20, random_state =100)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Models"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Naive Bayes Model"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn.naive_bayes import GaussianNB\n",
    "target = y_train.values\n",
    "features = X_train.values\n",
    "classifier_nb = GaussianNB()\n",
    "model_nb = classifier_nb.fit(features, target)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Acurracy: 34.55439015199096\n"
     ]
    }
   ],
   "source": [
    "prediction_nb = model_nb.predict(X_test)\n",
    "from sklearn.metrics import accuracy_score\n",
    "acc_score_nb = accuracy_score(prediction_nb,y_test)\n",
    "print(\"Acurracy:\", acc_score_nb*100)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### XGBoost Model"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "metadata": {},
   "outputs": [],
   "source": [
    "import xgboost\n",
    "classifier_xgb = xgboost.XGBClassifier(max_depth=4, learning_rate=0.1, n_estimators=800,\n",
    "                                  objective='multi:softmax', reg_alpha=0.5, reg_lambda=1.5,\n",
    "                                  booster='gbtree', n_jobs=4, min_child_weight=2, base_score= 0.75)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "metadata": {},
   "outputs": [],
   "source": [
    "model_xgb = classifier_xgb.fit(X_train, y_train)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Accuracy: 43.047355859816605\n"
     ]
    }
   ],
   "source": [
    "prediction_xgb = model_xgb.predict(X_test)\n",
    "acc_score_xgb = accuracy_score(prediction_xgb,y_test)\n",
    "print(\"Accuracy:\", acc_score_xgb*100)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Neural Network"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Index(['case_id', 'Hospital_code', 'Hospital_type_code', 'City_Code_Hospital',\n",
      "       'Hospital_region_code', 'Available Extra Rooms in Hospital',\n",
      "       'Department', 'Ward_Type', 'Ward_Facility_Code', 'Bed Grade',\n",
      "       'patientid', 'City_Code_Patient', 'Type of Admission',\n",
      "       'Severity of Illness', 'Visitors with Patient', 'Age',\n",
      "       'Admission_Deposit', 'count_id_patient',\n",
      "       'count_id_patient_hospitalCode', 'count_id_patient_wardfacilityCode'],\n",
      "      dtype='object')\n",
      "Index(['case_id', 'Hospital_code', 'Hospital_type_code', 'City_Code_Hospital',\n",
      "       'Hospital_region_code', 'Available Extra Rooms in Hospital',\n",
      "       'Department', 'Ward_Type', 'Ward_Facility_Code', 'Bed Grade',\n",
      "       'patientid', 'City_Code_Patient', 'Type of Admission',\n",
      "       'Severity of Illness', 'Visitors with Patient', 'Age',\n",
      "       'Admission_Deposit', 'count_id_patient',\n",
      "       'count_id_patient_hospitalCode', 'count_id_patient_wardfacilityCode'],\n",
      "      dtype='object')\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "(318438, 20)"
      ]
     },
     "execution_count": 27,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Segregation of features and target variable\n",
    "X = train.drop('Stay', axis =1)\n",
    "y = train['Stay']\n",
    "print(X.columns)\n",
    "z = test.drop('Stay', axis = 1)\n",
    "print(z.columns)\n",
    "\n",
    "# Data Scaling\n",
    "from sklearn import preprocessing\n",
    "X_scale = preprocessing.scale(X)\n",
    "X_scale.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "metadata": {},
   "outputs": [],
   "source": [
    "X_train, X_test, y_train, y_test = train_test_split(X_scale, y, test_size =0.20, random_state =100)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "metadata": {},
   "outputs": [],
   "source": [
    "import keras\n",
    "from keras.models import Sequential\n",
    "from keras.layers import Dense\n",
    "import tensorflow as tf"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "metadata": {},
   "outputs": [],
   "source": [
    "from keras.utils import to_categorical\n",
    "#Sparse Matrix\n",
    "a = to_categorical(y_train)\n",
    "b = to_categorical(y_test)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "metadata": {},
   "outputs": [],
   "source": [
    "model = Sequential()\n",
    "model.add(Dense(64, activation='relu', input_shape = (254750, 20))) \n",
    "model.add(Dense(128, activation='relu'))\n",
    "model.add(Dense(256, activation='relu'))\n",
    "model.add(Dense(512, activation='relu'))\n",
    "model.add(Dense(512, activation='relu'))\n",
    "model.add(Dense(11, activation='softmax'))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Model: \"sequential\"\n",
      "_________________________________________________________________\n",
      "Layer (type)                 Output Shape              Param #   \n",
      "=================================================================\n",
      "dense (Dense)                (None, 254750, 64)        1344      \n",
      "_________________________________________________________________\n",
      "dense_1 (Dense)              (None, 254750, 128)       8320      \n",
      "_________________________________________________________________\n",
      "dense_2 (Dense)              (None, 254750, 256)       33024     \n",
      "_________________________________________________________________\n",
      "dense_3 (Dense)              (None, 254750, 512)       131584    \n",
      "_________________________________________________________________\n",
      "dense_4 (Dense)              (None, 254750, 512)       262656    \n",
      "_________________________________________________________________\n",
      "dense_5 (Dense)              (None, 254750, 11)        5643      \n",
      "=================================================================\n",
      "Total params: 442,571\n",
      "Trainable params: 442,571\n",
      "Non-trainable params: 0\n",
      "_________________________________________________________________\n"
     ]
    }
   ],
   "source": [
    "model.summary()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 33,
   "metadata": {},
   "outputs": [],
   "source": [
    "model.compile(optimizer= 'SGD', \n",
    "              loss='categorical_crossentropy', \n",
    "              metrics=['accuracy'])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 34,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 1/20\n",
      "WARNING:tensorflow:Model was constructed with shape (None, 254750, 20) for input Tensor(\"dense_input:0\", shape=(None, 254750, 20), dtype=float32), but it was called on an input with incompatible shape (None, 20).\n",
      "WARNING:tensorflow:Model was constructed with shape (None, 254750, 20) for input Tensor(\"dense_input:0\", shape=(None, 254750, 20), dtype=float32), but it was called on an input with incompatible shape (None, 20).\n",
      "   1/6369 [..............................] - ETA: 0s - loss: 2.3814 - accuracy: 0.2188WARNING:tensorflow:From /Users/arjunmac/opt/anaconda3/lib/python3.8/site-packages/tensorflow/python/ops/summary_ops_v2.py:1277: stop (from tensorflow.python.eager.profiler) is deprecated and will be removed after 2020-07-01.\n",
      "Instructions for updating:\n",
      "use `tf.profiler.experimental.stop` instead.\n",
      "WARNING:tensorflow:Callbacks method `on_train_batch_end` is slow compared to the batch time (batch time: 0.0027s vs `on_train_batch_end` time: 0.0102s). Check your callbacks.\n",
      "6348/6369 [============================>.] - ETA: 0s - loss: 1.6578 - accuracy: 0.3684WARNING:tensorflow:Model was constructed with shape (None, 254750, 20) for input Tensor(\"dense_input:0\", shape=(None, 254750, 20), dtype=float32), but it was called on an input with incompatible shape (None, 20).\n",
      "6369/6369 [==============================] - 15s 2ms/step - loss: 1.6575 - accuracy: 0.3686 - val_loss: 1.5884 - val_accuracy: 0.3953\n",
      "Epoch 2/20\n",
      "6369/6369 [==============================] - 17s 3ms/step - loss: 1.5681 - accuracy: 0.3995 - val_loss: 1.5608 - val_accuracy: 0.4016\n",
      "Epoch 3/20\n",
      "6369/6369 [==============================] - 19s 3ms/step - loss: 1.5468 - accuracy: 0.4073 - val_loss: 1.5455 - val_accuracy: 0.4088\n",
      "Epoch 4/20\n",
      "6369/6369 [==============================] - 18s 3ms/step - loss: 1.5339 - accuracy: 0.4105 - val_loss: 1.5370 - val_accuracy: 0.4148\n",
      "Epoch 5/20\n",
      "6369/6369 [==============================] - 17s 3ms/step - loss: 1.5244 - accuracy: 0.4146 - val_loss: 1.5325 - val_accuracy: 0.4147\n",
      "Epoch 6/20\n",
      "6369/6369 [==============================] - 17s 3ms/step - loss: 1.5175 - accuracy: 0.4168 - val_loss: 1.5269 - val_accuracy: 0.4153\n",
      "Epoch 7/20\n",
      "6369/6369 [==============================] - 16s 3ms/step - loss: 1.5117 - accuracy: 0.4188 - val_loss: 1.5245 - val_accuracy: 0.4177\n",
      "Epoch 8/20\n",
      "6369/6369 [==============================] - 16s 2ms/step - loss: 1.5069 - accuracy: 0.4198 - val_loss: 1.5201 - val_accuracy: 0.4173\n",
      "Epoch 9/20\n",
      "6369/6369 [==============================] - 15s 2ms/step - loss: 1.5026 - accuracy: 0.4219 - val_loss: 1.5183 - val_accuracy: 0.4181\n",
      "Epoch 10/20\n",
      "6369/6369 [==============================] - 15s 2ms/step - loss: 1.4989 - accuracy: 0.4230 - val_loss: 1.5179 - val_accuracy: 0.4148\n",
      "Epoch 11/20\n",
      "6369/6369 [==============================] - 15s 2ms/step - loss: 1.4955 - accuracy: 0.4252 - val_loss: 1.5131 - val_accuracy: 0.4205\n",
      "Epoch 12/20\n",
      "6369/6369 [==============================] - 16s 2ms/step - loss: 1.4924 - accuracy: 0.4257 - val_loss: 1.5162 - val_accuracy: 0.4186\n",
      "Epoch 13/20\n",
      "6369/6369 [==============================] - 18s 3ms/step - loss: 1.4895 - accuracy: 0.4273 - val_loss: 1.5191 - val_accuracy: 0.4189\n",
      "Epoch 14/20\n",
      "6369/6369 [==============================] - 17s 3ms/step - loss: 1.4868 - accuracy: 0.4277 - val_loss: 1.5153 - val_accuracy: 0.4167\n",
      "Epoch 15/20\n",
      "6369/6369 [==============================] - 17s 3ms/step - loss: 1.4840 - accuracy: 0.4289 - val_loss: 1.5116 - val_accuracy: 0.4181\n",
      "Epoch 16/20\n",
      "6369/6369 [==============================] - 16s 3ms/step - loss: 1.4815 - accuracy: 0.4298 - val_loss: 1.5066 - val_accuracy: 0.4215\n",
      "Epoch 17/20\n",
      "6369/6369 [==============================] - 16s 2ms/step - loss: 1.4790 - accuracy: 0.4303 - val_loss: 1.5139 - val_accuracy: 0.4196\n",
      "Epoch 18/20\n",
      "6369/6369 [==============================] - 15s 2ms/step - loss: 1.4764 - accuracy: 0.4315 - val_loss: 1.5125 - val_accuracy: 0.4168\n",
      "Epoch 19/20\n",
      "6369/6369 [==============================] - 16s 2ms/step - loss: 1.4740 - accuracy: 0.4322 - val_loss: 1.5080 - val_accuracy: 0.4235\n",
      "Epoch 20/20\n",
      "6369/6369 [==============================] - 15s 2ms/step - loss: 1.4718 - accuracy: 0.4334 - val_loss: 1.5090 - val_accuracy: 0.4209\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "<tensorflow.python.keras.callbacks.History at 0x7ff88fd7b9a0>"
      ]
     },
     "execution_count": 34,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "callbacks = [tf.keras.callbacks.TensorBoard(\"logs_keras\")]\n",
    "model.fit(X_train, a, epochs=20, callbacks=callbacks, validation_split = 0.2)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Genrating tensorboard\n",
    "!tensorboard --logdir logs_keras"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 36,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 1/4\n",
      "6369/6369 [==============================] - 14s 2ms/step - loss: 1.4689 - accuracy: 0.4347 - val_loss: 1.5081 - val_accuracy: 0.4209\n",
      "Epoch 2/4\n",
      "6369/6369 [==============================] - 15s 2ms/step - loss: 1.4669 - accuracy: 0.4353 - val_loss: 1.5071 - val_accuracy: 0.4222\n",
      "Epoch 3/4\n",
      "6369/6369 [==============================] - 14s 2ms/step - loss: 1.4647 - accuracy: 0.4361 - val_loss: 1.5067 - val_accuracy: 0.4188\n",
      "Epoch 4/4\n",
      "6369/6369 [==============================] - 14s 2ms/step - loss: 1.4619 - accuracy: 0.4362 - val_loss: 1.5066 - val_accuracy: 0.4222\n",
      "\n",
      " Model Evaluation\n",
      "1991/1991 [==============================] - 2s 844us/step - loss: 1.5048 - accuracy: 0.4220\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "[1.5047825574874878, 0.42199471592903137]"
      ]
     },
     "execution_count": 36,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Retraining the model with 4 epochs\n",
    "model.fit(X_train, a, epochs=4, validation_split = 0.2)\n",
    "print(\"\\n Model Evaluation\")\n",
    "model.evaluate(X_test,b)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Predictions"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 37,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Naive Bayes\n",
    "pred_nb = classifier_nb.predict(test1.iloc[:,1:])\n",
    "result_nb = pd.DataFrame(pred_nb, columns=['Stay'])\n",
    "result_nb['case_id'] = test1['case_id']\n",
    "result_nb = result_nb[['case_id', 'Stay']]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 38,
   "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>case_id</th>\n",
       "      <th>Stay</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>318439</td>\n",
       "      <td>21-30</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>318440</td>\n",
       "      <td>51-60</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>318441</td>\n",
       "      <td>21-30</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>318442</td>\n",
       "      <td>21-30</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>318443</td>\n",
       "      <td>31-40</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   case_id   Stay\n",
       "0   318439  21-30\n",
       "1   318440  51-60\n",
       "2   318441  21-30\n",
       "3   318442  21-30\n",
       "4   318443  31-40"
      ]
     },
     "execution_count": 38,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "result_nb['Stay'] = result_nb['Stay'].replace({0:'0-10', 1: '11-20', 2: '21-30', 3:'31-40', 4: '41-50', 5: '51-60', 6: '61-70', 7: '71-80', 8: '81-90', 9: '91-100', 10: 'More than 100 Days'})\n",
    "result_nb.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 39,
   "metadata": {},
   "outputs": [],
   "source": [
    "# XGBoost\n",
    "pred_xgb = classifier_xgb.predict(test1.iloc[:,1:])\n",
    "result_xgb = pd.DataFrame(pred_xgb, columns=['Stay'])\n",
    "result_xgb['case_id'] = test1['case_id']\n",
    "result_xgb = result_xgb[['case_id', 'Stay']]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 40,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>case_id</th>\n",
       "      <th>Stay</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>318439</td>\n",
       "      <td>0-10</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>318440</td>\n",
       "      <td>51-60</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>318441</td>\n",
       "      <td>21-30</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>318442</td>\n",
       "      <td>21-30</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>318443</td>\n",
       "      <td>51-60</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   case_id   Stay\n",
       "0   318439   0-10\n",
       "1   318440  51-60\n",
       "2   318441  21-30\n",
       "3   318442  21-30\n",
       "4   318443  51-60"
      ]
     },
     "execution_count": 40,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "result_xgb['Stay'] = result_xgb['Stay'].replace({0:'0-10', 1: '11-20', 2: '21-30', 3:'31-40', 4: '41-50', 5: '51-60', 6: '61-70', 7: '71-80', 8: '81-90', 9: '91-100', 10: 'More than 100 Days'})\n",
    "result_xgb.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 41,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(137057, 20)"
      ]
     },
     "execution_count": 41,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Neural Network\n",
    "test_scale = preprocessing.scale(z)\n",
    "test_scale.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 42,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "WARNING:tensorflow:From <ipython-input-42-b90ff66bbd9d>:1: Sequential.predict_classes (from tensorflow.python.keras.engine.sequential) is deprecated and will be removed after 2021-01-01.\n",
      "Instructions for updating:\n",
      "Please use instead:* `np.argmax(model.predict(x), axis=-1)`,   if your model does multi-class classification   (e.g. if it uses a `softmax` last-layer activation).* `(model.predict(x) > 0.5).astype(\"int32\")`,   if your model does binary classification   (e.g. if it uses a `sigmoid` last-layer activation).\n",
      "WARNING:tensorflow:Model was constructed with shape (None, 254750, 20) for input Tensor(\"dense_input:0\", shape=(None, 254750, 20), dtype=float32), but it was called on an input with incompatible shape (None, 20).\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "array([0, 5, 2, ..., 1, 1, 5])"
      ]
     },
     "execution_count": 42,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "pred = model.predict_classes(test_scale)\n",
    "pred"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 43,
   "metadata": {},
   "outputs": [],
   "source": [
    "result_nn = pd.DataFrame(pred, columns=['Stay'])\n",
    "result_nn['case_id'] = test['case_id']\n",
    "result_nn = result_nn[['case_id', 'Stay']]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 44,
   "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>case_id</th>\n",
       "      <th>Stay</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>318439</td>\n",
       "      <td>0-10</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>318440</td>\n",
       "      <td>51-60</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>318441</td>\n",
       "      <td>21-30</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>318442</td>\n",
       "      <td>21-30</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>318443</td>\n",
       "      <td>51-60</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   case_id   Stay\n",
       "0   318439   0-10\n",
       "1   318440  51-60\n",
       "2   318441  21-30\n",
       "3   318442  21-30\n",
       "4   318443  51-60"
      ]
     },
     "execution_count": 44,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "result_nn['Stay'] = result_nn['Stay'].replace({0:'0-10', 1: '11-20', 2: '21-30', 3:'31-40', 4: '41-50', 5: '51-60', 6: '61-70', 7: '71-80', 8: '81-90', 9: '91-100', 10: 'More than 100 Days'})\n",
    "result_nn.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Results"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 45,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Stay\n",
      "0-10                   2598\n",
      "11-20                 26827\n",
      "21-30                 72206\n",
      "31-40                 15639\n",
      "41-50                   469\n",
      "51-60                 13651\n",
      "61-70                    92\n",
      "71-80                   955\n",
      "81-90                   296\n",
      "91-100                    2\n",
      "More than 100 Days     4322\n",
      "Name: case_id, dtype: int64\n"
     ]
    }
   ],
   "source": [
    "# Naive Bayes\n",
    "print(result_nb.groupby('Stay')['case_id'].nunique())"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 46,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Stay\n",
      "0-10                   4373\n",
      "11-20                 39337\n",
      "21-30                 58261\n",
      "31-40                 12100\n",
      "41-50                    61\n",
      "51-60                 19217\n",
      "61-70                    16\n",
      "71-80                   302\n",
      "81-90                  1099\n",
      "91-100                   78\n",
      "More than 100 Days     2213\n",
      "Name: case_id, dtype: int64\n"
     ]
    }
   ],
   "source": [
    "# XGBoost\n",
    "print(result_xgb.groupby('Stay')['case_id'].nunique())"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 47,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Stay\n",
      "0-10                   5379\n",
      "11-20                 41215\n",
      "21-30                 55240\n",
      "31-40                 10926\n",
      "41-50                     9\n",
      "51-60                 20016\n",
      "71-80                    29\n",
      "81-90                  1126\n",
      "More than 100 Days     3117\n",
      "Name: case_id, dtype: int64\n"
     ]
    }
   ],
   "source": [
    "# Neural Networks\n",
    "print(result_nn.groupby('Stay')['case_id'].nunique())"
   ]
  }
 ],
 "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": "3.8.3"
  },
  "varInspector": {
   "cols": {
    "lenName": 16,
    "lenType": 16,
    "lenVar": 40
   },
   "kernels_config": {
    "python": {
     "delete_cmd_postfix": "",
     "delete_cmd_prefix": "del ",
     "library": "var_list.py",
     "varRefreshCmd": "print(var_dic_list())"
    },
    "r": {
     "delete_cmd_postfix": ") ",
     "delete_cmd_prefix": "rm(",
     "library": "var_list.r",
     "varRefreshCmd": "cat(var_dic_list()) "
    }
   },
   "position": {
    "height": "234px",
    "left": "1112px",
    "right": "20px",
    "top": "157px",
    "width": "330px"
   },
   "types_to_exclude": [
    "module",
    "function",
    "builtin_function_or_method",
    "instance",
    "_Feature"
   ],
   "window_display": false
  }
 },
 "nbformat": 4,
 "nbformat_minor": 4
}