{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "id": "f2b656df",
   "metadata": {},
   "outputs": [],
   "source": [
    "import pandas as pd\n",
    "import matplotlib.pyplot as plt\n",
    "import seaborn as sns\n",
    "import numpy as np"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "baa3758b",
   "metadata": {},
   "source": [
    "# Objectives"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "bdb464ad",
   "metadata": {},
   "source": [
    "The objectives of this analysis are to find an accurate machine learning model that is not opaque and to determine what features contribute the most significantly to covid death from the CSV file provided."
   ]
  },
  {
   "cell_type": "markdown",
   "id": "932008d3",
   "metadata": {},
   "source": [
    "## Importing Data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "id": "73e187df",
   "metadata": {},
   "outputs": [],
   "source": [
    "# Import CSV file to a dataframe and format the columns\n",
    "df = pd.read_csv(\"data/COVID-19_Reported_Patient_Impact_and_Hospital_Capacity_by_State_Timeseries__RAW_.csv\")\n",
    "df['date'] = pd.to_datetime(df['date'])"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "a71b9337",
   "metadata": {},
   "source": [
    "## Describing Data"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "dad4fe6b",
   "metadata": {},
   "source": [
    "The dataset is extremely wide with 133 columns, each with different levels of data completeness. All data is numeric except for state and date. \n",
    "\n",
    "There are 64703 entries from the years 2020 to 2023. "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "id": "002907f0",
   "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>state</th>\n",
       "      <th>date</th>\n",
       "      <th>critical_staffing_shortage_today_yes</th>\n",
       "      <th>critical_staffing_shortage_today_no</th>\n",
       "      <th>critical_staffing_shortage_today_not_reported</th>\n",
       "      <th>critical_staffing_shortage_anticipated_within_week_yes</th>\n",
       "      <th>critical_staffing_shortage_anticipated_within_week_no</th>\n",
       "      <th>critical_staffing_shortage_anticipated_within_week_not_reported</th>\n",
       "      <th>hospital_onset_covid</th>\n",
       "      <th>hospital_onset_covid_coverage</th>\n",
       "      <th>...</th>\n",
       "      <th>previous_day_admission_pediatric_covid_confirmed_5_11</th>\n",
       "      <th>previous_day_admission_pediatric_covid_confirmed_5_11_coverage</th>\n",
       "      <th>previous_day_admission_pediatric_covid_confirmed_unknown</th>\n",
       "      <th>previous_day_admission_pediatric_covid_confirmed_unknown_coverage</th>\n",
       "      <th>staffed_icu_pediatric_patients_confirmed_covid</th>\n",
       "      <th>staffed_icu_pediatric_patients_confirmed_covid_coverage</th>\n",
       "      <th>staffed_pediatric_icu_bed_occupancy</th>\n",
       "      <th>staffed_pediatric_icu_bed_occupancy_coverage</th>\n",
       "      <th>total_staffed_pediatric_icu_beds</th>\n",
       "      <th>total_staffed_pediatric_icu_beds_coverage</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>RI</td>\n",
       "      <td>2021-02-26</td>\n",
       "      <td>4</td>\n",
       "      <td>10</td>\n",
       "      <td>1</td>\n",
       "      <td>4</td>\n",
       "      <td>10</td>\n",
       "      <td>1</td>\n",
       "      <td>6.0</td>\n",
       "      <td>14</td>\n",
       "      <td>...</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0</td>\n",
       "      <td>68.0</td>\n",
       "      <td>14</td>\n",
       "      <td>80.0</td>\n",
       "      <td>14</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>MA</td>\n",
       "      <td>2021-02-24</td>\n",
       "      <td>10</td>\n",
       "      <td>90</td>\n",
       "      <td>1</td>\n",
       "      <td>9</td>\n",
       "      <td>91</td>\n",
       "      <td>1</td>\n",
       "      <td>40.0</td>\n",
       "      <td>100</td>\n",
       "      <td>...</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>NE</td>\n",
       "      <td>2021-02-17</td>\n",
       "      <td>10</td>\n",
       "      <td>90</td>\n",
       "      <td>1</td>\n",
       "      <td>17</td>\n",
       "      <td>83</td>\n",
       "      <td>1</td>\n",
       "      <td>3.0</td>\n",
       "      <td>100</td>\n",
       "      <td>...</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>12</td>\n",
       "      <td>0.0</td>\n",
       "      <td>12</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>ME</td>\n",
       "      <td>2021-01-30</td>\n",
       "      <td>2</td>\n",
       "      <td>29</td>\n",
       "      <td>8</td>\n",
       "      <td>4</td>\n",
       "      <td>27</td>\n",
       "      <td>8</td>\n",
       "      <td>2.0</td>\n",
       "      <td>38</td>\n",
       "      <td>...</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0</td>\n",
       "      <td>47.0</td>\n",
       "      <td>38</td>\n",
       "      <td>54.0</td>\n",
       "      <td>38</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>NH</td>\n",
       "      <td>2021-01-30</td>\n",
       "      <td>6</td>\n",
       "      <td>23</td>\n",
       "      <td>1</td>\n",
       "      <td>8</td>\n",
       "      <td>21</td>\n",
       "      <td>1</td>\n",
       "      <td>8.0</td>\n",
       "      <td>30</td>\n",
       "      <td>...</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0</td>\n",
       "      <td>28.0</td>\n",
       "      <td>17</td>\n",
       "      <td>39.0</td>\n",
       "      <td>17</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>5 rows × 135 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "  state       date  critical_staffing_shortage_today_yes  \\\n",
       "0    RI 2021-02-26                                     4   \n",
       "1    MA 2021-02-24                                    10   \n",
       "2    NE 2021-02-17                                    10   \n",
       "3    ME 2021-01-30                                     2   \n",
       "4    NH 2021-01-30                                     6   \n",
       "\n",
       "   critical_staffing_shortage_today_no  \\\n",
       "0                                   10   \n",
       "1                                   90   \n",
       "2                                   90   \n",
       "3                                   29   \n",
       "4                                   23   \n",
       "\n",
       "   critical_staffing_shortage_today_not_reported  \\\n",
       "0                                              1   \n",
       "1                                              1   \n",
       "2                                              1   \n",
       "3                                              8   \n",
       "4                                              1   \n",
       "\n",
       "   critical_staffing_shortage_anticipated_within_week_yes  \\\n",
       "0                                                  4        \n",
       "1                                                  9        \n",
       "2                                                 17        \n",
       "3                                                  4        \n",
       "4                                                  8        \n",
       "\n",
       "   critical_staffing_shortage_anticipated_within_week_no  \\\n",
       "0                                                 10       \n",
       "1                                                 91       \n",
       "2                                                 83       \n",
       "3                                                 27       \n",
       "4                                                 21       \n",
       "\n",
       "   critical_staffing_shortage_anticipated_within_week_not_reported  \\\n",
       "0                                                  1                 \n",
       "1                                                  1                 \n",
       "2                                                  1                 \n",
       "3                                                  8                 \n",
       "4                                                  1                 \n",
       "\n",
       "   hospital_onset_covid  hospital_onset_covid_coverage  ...  \\\n",
       "0                   6.0                             14  ...   \n",
       "1                  40.0                            100  ...   \n",
       "2                   3.0                            100  ...   \n",
       "3                   2.0                             38  ...   \n",
       "4                   8.0                             30  ...   \n",
       "\n",
       "   previous_day_admission_pediatric_covid_confirmed_5_11  \\\n",
       "0                                                NaN       \n",
       "1                                                NaN       \n",
       "2                                                NaN       \n",
       "3                                                NaN       \n",
       "4                                                NaN       \n",
       "\n",
       "   previous_day_admission_pediatric_covid_confirmed_5_11_coverage  \\\n",
       "0                                                  0                \n",
       "1                                                  0                \n",
       "2                                                  0                \n",
       "3                                                  0                \n",
       "4                                                  0                \n",
       "\n",
       "   previous_day_admission_pediatric_covid_confirmed_unknown  \\\n",
       "0                                                NaN          \n",
       "1                                                NaN          \n",
       "2                                                NaN          \n",
       "3                                                NaN          \n",
       "4                                                NaN          \n",
       "\n",
       "   previous_day_admission_pediatric_covid_confirmed_unknown_coverage  \\\n",
       "0                                                  0                   \n",
       "1                                                  0                   \n",
       "2                                                  0                   \n",
       "3                                                  0                   \n",
       "4                                                  0                   \n",
       "\n",
       "   staffed_icu_pediatric_patients_confirmed_covid  \\\n",
       "0                                             NaN   \n",
       "1                                             NaN   \n",
       "2                                             NaN   \n",
       "3                                             NaN   \n",
       "4                                             NaN   \n",
       "\n",
       "   staffed_icu_pediatric_patients_confirmed_covid_coverage  \\\n",
       "0                                                  0         \n",
       "1                                                  0         \n",
       "2                                                  0         \n",
       "3                                                  0         \n",
       "4                                                  0         \n",
       "\n",
       "   staffed_pediatric_icu_bed_occupancy  \\\n",
       "0                                 68.0   \n",
       "1                                  NaN   \n",
       "2                                  0.0   \n",
       "3                                 47.0   \n",
       "4                                 28.0   \n",
       "\n",
       "   staffed_pediatric_icu_bed_occupancy_coverage  \\\n",
       "0                                            14   \n",
       "1                                             0   \n",
       "2                                            12   \n",
       "3                                            38   \n",
       "4                                            17   \n",
       "\n",
       "   total_staffed_pediatric_icu_beds  total_staffed_pediatric_icu_beds_coverage  \n",
       "0                              80.0                                         14  \n",
       "1                               NaN                                          0  \n",
       "2                               0.0                                         12  \n",
       "3                              54.0                                         38  \n",
       "4                              39.0                                         17  \n",
       "\n",
       "[5 rows x 135 columns]"
      ]
     },
     "execution_count": 3,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "id": "e3571ddf",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "<class 'pandas.core.frame.DataFrame'>\n",
      "RangeIndex: 64703 entries, 0 to 64702\n",
      "Columns: 135 entries, state to total_staffed_pediatric_icu_beds_coverage\n",
      "dtypes: datetime64[ns](1), float64(77), int64(56), object(1)\n",
      "memory usage: 66.6+ MB\n"
     ]
    }
   ],
   "source": [
    "# Print initial dataframe info\n",
    "df.info()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "id": "e4781cfd",
   "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>date</th>\n",
       "      <th>critical_staffing_shortage_today_yes</th>\n",
       "      <th>critical_staffing_shortage_today_no</th>\n",
       "      <th>critical_staffing_shortage_today_not_reported</th>\n",
       "      <th>critical_staffing_shortage_anticipated_within_week_yes</th>\n",
       "      <th>critical_staffing_shortage_anticipated_within_week_no</th>\n",
       "      <th>critical_staffing_shortage_anticipated_within_week_not_reported</th>\n",
       "      <th>hospital_onset_covid</th>\n",
       "      <th>hospital_onset_covid_coverage</th>\n",
       "      <th>inpatient_beds</th>\n",
       "      <th>...</th>\n",
       "      <th>previous_day_admission_pediatric_covid_confirmed_5_11</th>\n",
       "      <th>previous_day_admission_pediatric_covid_confirmed_5_11_coverage</th>\n",
       "      <th>previous_day_admission_pediatric_covid_confirmed_unknown</th>\n",
       "      <th>previous_day_admission_pediatric_covid_confirmed_unknown_coverage</th>\n",
       "      <th>staffed_icu_pediatric_patients_confirmed_covid</th>\n",
       "      <th>staffed_icu_pediatric_patients_confirmed_covid_coverage</th>\n",
       "      <th>staffed_pediatric_icu_bed_occupancy</th>\n",
       "      <th>staffed_pediatric_icu_bed_occupancy_coverage</th>\n",
       "      <th>total_staffed_pediatric_icu_beds</th>\n",
       "      <th>total_staffed_pediatric_icu_beds_coverage</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>count</th>\n",
       "      <td>64703</td>\n",
       "      <td>64703.000000</td>\n",
       "      <td>64703.000000</td>\n",
       "      <td>64703.000000</td>\n",
       "      <td>64703.000000</td>\n",
       "      <td>64703.000000</td>\n",
       "      <td>64703.000000</td>\n",
       "      <td>64446.000000</td>\n",
       "      <td>64703.000000</td>\n",
       "      <td>64698.000000</td>\n",
       "      <td>...</td>\n",
       "      <td>27677.000000</td>\n",
       "      <td>64703.000000</td>\n",
       "      <td>28407.000000</td>\n",
       "      <td>64703.000000</td>\n",
       "      <td>34554.000000</td>\n",
       "      <td>64703.000000</td>\n",
       "      <td>56400.000000</td>\n",
       "      <td>64703.000000</td>\n",
       "      <td>56387.000000</td>\n",
       "      <td>64703.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>mean</th>\n",
       "      <td>2021-10-23 20:33:20.105095424</td>\n",
       "      <td>9.334343</td>\n",
       "      <td>55.179404</td>\n",
       "      <td>39.791323</td>\n",
       "      <td>14.525092</td>\n",
       "      <td>63.044789</td>\n",
       "      <td>26.735190</td>\n",
       "      <td>23.640924</td>\n",
       "      <td>98.615644</td>\n",
       "      <td>13338.298804</td>\n",
       "      <td>...</td>\n",
       "      <td>0.639412</td>\n",
       "      <td>40.596958</td>\n",
       "      <td>0.815890</td>\n",
       "      <td>42.540114</td>\n",
       "      <td>2.922990</td>\n",
       "      <td>48.024466</td>\n",
       "      <td>169.880177</td>\n",
       "      <td>74.827844</td>\n",
       "      <td>257.038963</td>\n",
       "      <td>74.662025</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>min</th>\n",
       "      <td>2020-01-01 00:00:00</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>...</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>-1.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>25%</th>\n",
       "      <td>2020-12-26 00:00:00</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>6.000000</td>\n",
       "      <td>3.000000</td>\n",
       "      <td>2.000000</td>\n",
       "      <td>16.000000</td>\n",
       "      <td>2.000000</td>\n",
       "      <td>2.000000</td>\n",
       "      <td>38.000000</td>\n",
       "      <td>3158.000000</td>\n",
       "      <td>...</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>17.000000</td>\n",
       "      <td>13.000000</td>\n",
       "      <td>29.000000</td>\n",
       "      <td>13.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>50%</th>\n",
       "      <td>2021-10-26 00:00:00</td>\n",
       "      <td>3.000000</td>\n",
       "      <td>37.000000</td>\n",
       "      <td>14.000000</td>\n",
       "      <td>8.000000</td>\n",
       "      <td>52.000000</td>\n",
       "      <td>8.000000</td>\n",
       "      <td>7.000000</td>\n",
       "      <td>86.000000</td>\n",
       "      <td>8891.500000</td>\n",
       "      <td>...</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>4.000000</td>\n",
       "      <td>72.000000</td>\n",
       "      <td>52.000000</td>\n",
       "      <td>110.000000</td>\n",
       "      <td>52.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>75%</th>\n",
       "      <td>2022-08-22 00:00:00</td>\n",
       "      <td>12.000000</td>\n",
       "      <td>87.000000</td>\n",
       "      <td>47.000000</td>\n",
       "      <td>21.000000</td>\n",
       "      <td>90.000000</td>\n",
       "      <td>26.000000</td>\n",
       "      <td>23.000000</td>\n",
       "      <td>129.000000</td>\n",
       "      <td>17106.750000</td>\n",
       "      <td>...</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>58.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>62.000000</td>\n",
       "      <td>3.000000</td>\n",
       "      <td>84.000000</td>\n",
       "      <td>191.000000</td>\n",
       "      <td>108.000000</td>\n",
       "      <td>319.000000</td>\n",
       "      <td>108.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>max</th>\n",
       "      <td>2023-06-17 00:00:00</td>\n",
       "      <td>191.000000</td>\n",
       "      <td>494.000000</td>\n",
       "      <td>523.000000</td>\n",
       "      <td>204.000000</td>\n",
       "      <td>469.000000</td>\n",
       "      <td>523.000000</td>\n",
       "      <td>1334.000000</td>\n",
       "      <td>597.000000</td>\n",
       "      <td>108966.000000</td>\n",
       "      <td>...</td>\n",
       "      <td>101.000000</td>\n",
       "      <td>591.000000</td>\n",
       "      <td>155.000000</td>\n",
       "      <td>595.000000</td>\n",
       "      <td>346.000000</td>\n",
       "      <td>595.000000</td>\n",
       "      <td>2580.000000</td>\n",
       "      <td>597.000000</td>\n",
       "      <td>3917.000000</td>\n",
       "      <td>597.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>std</th>\n",
       "      <td>NaN</td>\n",
       "      <td>16.287815</td>\n",
       "      <td>62.544193</td>\n",
       "      <td>66.802128</td>\n",
       "      <td>20.534351</td>\n",
       "      <td>61.075534</td>\n",
       "      <td>58.251265</td>\n",
       "      <td>53.715488</td>\n",
       "      <td>91.323635</td>\n",
       "      <td>14742.114302</td>\n",
       "      <td>...</td>\n",
       "      <td>1.605329</td>\n",
       "      <td>73.441597</td>\n",
       "      <td>5.183662</td>\n",
       "      <td>75.923520</td>\n",
       "      <td>6.180913</td>\n",
       "      <td>77.132074</td>\n",
       "      <td>287.636544</td>\n",
       "      <td>86.462506</td>\n",
       "      <td>438.611268</td>\n",
       "      <td>86.485956</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>8 rows × 134 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "                                date  critical_staffing_shortage_today_yes  \\\n",
       "count                          64703                          64703.000000   \n",
       "mean   2021-10-23 20:33:20.105095424                              9.334343   \n",
       "min              2020-01-01 00:00:00                              0.000000   \n",
       "25%              2020-12-26 00:00:00                              0.000000   \n",
       "50%              2021-10-26 00:00:00                              3.000000   \n",
       "75%              2022-08-22 00:00:00                             12.000000   \n",
       "max              2023-06-17 00:00:00                            191.000000   \n",
       "std                              NaN                             16.287815   \n",
       "\n",
       "       critical_staffing_shortage_today_no  \\\n",
       "count                         64703.000000   \n",
       "mean                             55.179404   \n",
       "min                               0.000000   \n",
       "25%                               6.000000   \n",
       "50%                              37.000000   \n",
       "75%                              87.000000   \n",
       "max                             494.000000   \n",
       "std                              62.544193   \n",
       "\n",
       "       critical_staffing_shortage_today_not_reported  \\\n",
       "count                                   64703.000000   \n",
       "mean                                       39.791323   \n",
       "min                                         0.000000   \n",
       "25%                                         3.000000   \n",
       "50%                                        14.000000   \n",
       "75%                                        47.000000   \n",
       "max                                       523.000000   \n",
       "std                                        66.802128   \n",
       "\n",
       "       critical_staffing_shortage_anticipated_within_week_yes  \\\n",
       "count                                       64703.000000        \n",
       "mean                                           14.525092        \n",
       "min                                             0.000000        \n",
       "25%                                             2.000000        \n",
       "50%                                             8.000000        \n",
       "75%                                            21.000000        \n",
       "max                                           204.000000        \n",
       "std                                            20.534351        \n",
       "\n",
       "       critical_staffing_shortage_anticipated_within_week_no  \\\n",
       "count                                       64703.000000       \n",
       "mean                                           63.044789       \n",
       "min                                             0.000000       \n",
       "25%                                            16.000000       \n",
       "50%                                            52.000000       \n",
       "75%                                            90.000000       \n",
       "max                                           469.000000       \n",
       "std                                            61.075534       \n",
       "\n",
       "       critical_staffing_shortage_anticipated_within_week_not_reported  \\\n",
       "count                                       64703.000000                 \n",
       "mean                                           26.735190                 \n",
       "min                                             0.000000                 \n",
       "25%                                             2.000000                 \n",
       "50%                                             8.000000                 \n",
       "75%                                            26.000000                 \n",
       "max                                           523.000000                 \n",
       "std                                            58.251265                 \n",
       "\n",
       "       hospital_onset_covid  hospital_onset_covid_coverage  inpatient_beds  \\\n",
       "count          64446.000000                   64703.000000    64698.000000   \n",
       "mean              23.640924                      98.615644    13338.298804   \n",
       "min                0.000000                       0.000000        0.000000   \n",
       "25%                2.000000                      38.000000     3158.000000   \n",
       "50%                7.000000                      86.000000     8891.500000   \n",
       "75%               23.000000                     129.000000    17106.750000   \n",
       "max             1334.000000                     597.000000   108966.000000   \n",
       "std               53.715488                      91.323635    14742.114302   \n",
       "\n",
       "       ...  previous_day_admission_pediatric_covid_confirmed_5_11  \\\n",
       "count  ...                                       27677.000000       \n",
       "mean   ...                                           0.639412       \n",
       "min    ...                                           0.000000       \n",
       "25%    ...                                           0.000000       \n",
       "50%    ...                                           0.000000       \n",
       "75%    ...                                           1.000000       \n",
       "max    ...                                         101.000000       \n",
       "std    ...                                           1.605329       \n",
       "\n",
       "       previous_day_admission_pediatric_covid_confirmed_5_11_coverage  \\\n",
       "count                                       64703.000000                \n",
       "mean                                           40.596958                \n",
       "min                                             0.000000                \n",
       "25%                                             0.000000                \n",
       "50%                                             0.000000                \n",
       "75%                                            58.000000                \n",
       "max                                           591.000000                \n",
       "std                                            73.441597                \n",
       "\n",
       "       previous_day_admission_pediatric_covid_confirmed_unknown  \\\n",
       "count                                       28407.000000          \n",
       "mean                                            0.815890          \n",
       "min                                            -1.000000          \n",
       "25%                                             0.000000          \n",
       "50%                                             0.000000          \n",
       "75%                                             0.000000          \n",
       "max                                           155.000000          \n",
       "std                                             5.183662          \n",
       "\n",
       "       previous_day_admission_pediatric_covid_confirmed_unknown_coverage  \\\n",
       "count                                       64703.000000                   \n",
       "mean                                           42.540114                   \n",
       "min                                             0.000000                   \n",
       "25%                                             0.000000                   \n",
       "50%                                             0.000000                   \n",
       "75%                                            62.000000                   \n",
       "max                                           595.000000                   \n",
       "std                                            75.923520                   \n",
       "\n",
       "       staffed_icu_pediatric_patients_confirmed_covid  \\\n",
       "count                                    34554.000000   \n",
       "mean                                         2.922990   \n",
       "min                                          0.000000   \n",
       "25%                                          0.000000   \n",
       "50%                                          1.000000   \n",
       "75%                                          3.000000   \n",
       "max                                        346.000000   \n",
       "std                                          6.180913   \n",
       "\n",
       "       staffed_icu_pediatric_patients_confirmed_covid_coverage  \\\n",
       "count                                       64703.000000         \n",
       "mean                                           48.024466         \n",
       "min                                             0.000000         \n",
       "25%                                             0.000000         \n",
       "50%                                             4.000000         \n",
       "75%                                            84.000000         \n",
       "max                                           595.000000         \n",
       "std                                            77.132074         \n",
       "\n",
       "       staffed_pediatric_icu_bed_occupancy  \\\n",
       "count                         56400.000000   \n",
       "mean                            169.880177   \n",
       "min                               0.000000   \n",
       "25%                              17.000000   \n",
       "50%                              72.000000   \n",
       "75%                             191.000000   \n",
       "max                            2580.000000   \n",
       "std                             287.636544   \n",
       "\n",
       "       staffed_pediatric_icu_bed_occupancy_coverage  \\\n",
       "count                                  64703.000000   \n",
       "mean                                      74.827844   \n",
       "min                                        0.000000   \n",
       "25%                                       13.000000   \n",
       "50%                                       52.000000   \n",
       "75%                                      108.000000   \n",
       "max                                      597.000000   \n",
       "std                                       86.462506   \n",
       "\n",
       "       total_staffed_pediatric_icu_beds  \\\n",
       "count                      56387.000000   \n",
       "mean                         257.038963   \n",
       "min                            0.000000   \n",
       "25%                           29.000000   \n",
       "50%                          110.000000   \n",
       "75%                          319.000000   \n",
       "max                         3917.000000   \n",
       "std                          438.611268   \n",
       "\n",
       "       total_staffed_pediatric_icu_beds_coverage  \n",
       "count                               64703.000000  \n",
       "mean                                   74.662025  \n",
       "min                                     0.000000  \n",
       "25%                                    13.000000  \n",
       "50%                                    52.000000  \n",
       "75%                                   108.000000  \n",
       "max                                   597.000000  \n",
       "std                                    86.485956  \n",
       "\n",
       "[8 rows x 134 columns]"
      ]
     },
     "execution_count": 5,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Print initial dataframe description\n",
    "df.describe()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "id": "735d71be",
   "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>null_values</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>column</th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>geocoded_state</th>\n",
       "      <td>64703</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>previous_day_admission_pediatric_covid_confirmed_12_17</th>\n",
       "      <td>37040</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>previous_day_admission_pediatric_covid_confirmed_5_11</th>\n",
       "      <td>37026</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>previous_day_admission_pediatric_covid_confirmed_0_4</th>\n",
       "      <td>36410</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>previous_day_admission_pediatric_covid_confirmed_unknown</th>\n",
       "      <td>36296</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>staffed_icu_pediatric_patients_confirmed_covid</th>\n",
       "      <td>30149</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>on_hand_supply_therapeutic_c_bamlanivimab_etesevimab_courses</th>\n",
       "      <td>21210</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>previous_week_therapeutic_c_bamlanivimab_etesevimab_courses_used</th>\n",
       "      <td>21193</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>on_hand_supply_therapeutic_b_bamlanivimab_courses</th>\n",
       "      <td>17718</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>previous_week_therapeutic_b_bamlanivimab_courses_used</th>\n",
       "      <td>17686</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>on_hand_supply_therapeutic_a_casirivimab_imdevimab_courses</th>\n",
       "      <td>16528</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>previous_week_therapeutic_a_casirivimab_imdevimab_courses_used</th>\n",
       "      <td>16527</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>previous_day_deaths_covid_and_influenza</th>\n",
       "      <td>12888</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>total_patients_hospitalized_confirmed_influenza_and_covid</th>\n",
       "      <td>12884</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>previous_day_deaths_influenza</th>\n",
       "      <td>12746</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>total_patients_hospitalized_confirmed_influenza</th>\n",
       "      <td>11766</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>icu_patients_confirmed_influenza</th>\n",
       "      <td>11710</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>previous_day_admission_influenza_confirmed</th>\n",
       "      <td>11709</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>total_staffed_pediatric_icu_beds</th>\n",
       "      <td>8316</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>all_pediatric_inpatient_beds</th>\n",
       "      <td>8314</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>all_pediatric_inpatient_bed_occupied</th>\n",
       "      <td>8303</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>staffed_pediatric_icu_bed_occupancy</th>\n",
       "      <td>8303</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>previous_day_admission_adult_covid_suspected_80+</th>\n",
       "      <td>8158</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>previous_day_admission_adult_covid_suspected_50-59</th>\n",
       "      <td>8156</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>previous_day_admission_adult_covid_suspected_40-49</th>\n",
       "      <td>8154</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>previous_day_admission_adult_covid_suspected_70-79</th>\n",
       "      <td>8151</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>previous_day_admission_adult_covid_suspected_60-69</th>\n",
       "      <td>8150</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>previous_day_admission_adult_covid_suspected_20-29</th>\n",
       "      <td>8146</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>previous_day_admission_adult_covid_suspected_30-39</th>\n",
       "      <td>8145</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>previous_day_admission_adult_covid_confirmed_40-49</th>\n",
       "      <td>8127</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                                                    null_values\n",
       "column                                                         \n",
       "geocoded_state                                            64703\n",
       "previous_day_admission_pediatric_covid_confirme...        37040\n",
       "previous_day_admission_pediatric_covid_confirme...        37026\n",
       "previous_day_admission_pediatric_covid_confirme...        36410\n",
       "previous_day_admission_pediatric_covid_confirme...        36296\n",
       "staffed_icu_pediatric_patients_confirmed_covid            30149\n",
       "on_hand_supply_therapeutic_c_bamlanivimab_etese...        21210\n",
       "previous_week_therapeutic_c_bamlanivimab_etesev...        21193\n",
       "on_hand_supply_therapeutic_b_bamlanivimab_courses         17718\n",
       "previous_week_therapeutic_b_bamlanivimab_course...        17686\n",
       "on_hand_supply_therapeutic_a_casirivimab_imdevi...        16528\n",
       "previous_week_therapeutic_a_casirivimab_imdevim...        16527\n",
       "previous_day_deaths_covid_and_influenza                   12888\n",
       "total_patients_hospitalized_confirmed_influenza...        12884\n",
       "previous_day_deaths_influenza                             12746\n",
       "total_patients_hospitalized_confirmed_influenza           11766\n",
       "icu_patients_confirmed_influenza                          11710\n",
       "previous_day_admission_influenza_confirmed                11709\n",
       "total_staffed_pediatric_icu_beds                           8316\n",
       "all_pediatric_inpatient_beds                               8314\n",
       "all_pediatric_inpatient_bed_occupied                       8303\n",
       "staffed_pediatric_icu_bed_occupancy                        8303\n",
       "previous_day_admission_adult_covid_suspected_80+           8158\n",
       "previous_day_admission_adult_covid_suspected_50-59         8156\n",
       "previous_day_admission_adult_covid_suspected_40-49         8154\n",
       "previous_day_admission_adult_covid_suspected_70-79         8151\n",
       "previous_day_admission_adult_covid_suspected_60-69         8150\n",
       "previous_day_admission_adult_covid_suspected_20-29         8146\n",
       "previous_day_admission_adult_covid_suspected_30-39         8145\n",
       "previous_day_admission_adult_covid_confirmed_40-49         8127"
      ]
     },
     "execution_count": 6,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Create null_df showing null value counts\n",
    "null_df = (df\n",
    "           .isnull()\n",
    "           .sum()\n",
    "           .to_frame()\n",
    "           .reset_index()\n",
    "           .rename(columns={'index':'column', 0:'null_values'})\n",
    "           .sort_values(by='null_values', ascending=False)\n",
    "           .reset_index(drop=True)\n",
    "           .set_index('column')\n",
    "          )\n",
    "\n",
    "# Filter null_df to only columns that have null values\n",
    "null_df = null_df[null_df['null_values'] != 0]\n",
    "\n",
    "# Print top 30 null value counts\n",
    "null_df.head(30)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "6f984a76",
   "metadata": {},
   "source": [
    "# Visualizing Data"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "32c5e231",
   "metadata": {},
   "source": [
    "The data clearly has holes in it, as demonstrated in the two following line charts. This is going to be fixed later through fixing NaN values in the dataframe."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "id": "c2a3a63f",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<Axes: xlabel='date'>"
      ]
     },
     "execution_count": 7,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Plots with similar trends showing missing values\n",
    "df[['date', 'percent_of_inpatients_with_covid', 'inpatient_bed_covid_utilization', 'adult_icu_bed_covid_utilization']].groupby('date').mean().resample('W').mean().plot()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "id": "f698ca81",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<Axes: xlabel='date'>"
      ]
     },
     "execution_count": 8,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Adult ICU Bed Utilization and Inpatient Beds Utilization line charts\n",
    "df[['date', 'adult_icu_bed_utilization', 'inpatient_beds_utilization']].groupby('date').mean().resample('W').mean().plot()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "152b4541",
   "metadata": {},
   "source": [
    "# Preprocessing Data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "id": "e6622e75",
   "metadata": {},
   "outputs": [],
   "source": [
    "# Drop geocoded_state as it is empty\n",
    "df.drop(columns='geocoded_state', inplace=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "id": "eff4b2ad",
   "metadata": {},
   "outputs": [],
   "source": [
    "# Create a mask and remove the beginning days of pandemic with little information\n",
    "start_date = '2020-01-01'\n",
    "end_date = '2020-08-01'\n",
    "mask = (df['date'] >= start_date) & (df['date'] <= end_date)\n",
    "\n",
    "# Apply mask to dataframe to filter by date\n",
    "df = df.loc[~mask].reset_index(drop=True)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "5730c055",
   "metadata": {},
   "source": [
    "## Fix NaN Values"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "id": "868f4490",
   "metadata": {},
   "outputs": [],
   "source": [
    "# Forward fill all null values and remove the rest\n",
    "df = df.fillna(method='ffill').dropna().reset_index(drop=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "id": "f9db538d",
   "metadata": {},
   "outputs": [],
   "source": [
    "# Create null_df showing null value counts\n",
    "null_df = (df\n",
    "           .isnull()\n",
    "           .sum()\n",
    "           .to_frame()\n",
    "           .reset_index()\n",
    "           .rename(columns={'index':'column', 0:'null_values'})\n",
    "           .sort_values(by='null_values', ascending=False)\n",
    "           .reset_index(drop=True)\n",
    "           .set_index('column')\n",
    "          )\n",
    "\n",
    "# Filter null_df to only columns that have null values\n",
    "null_df = null_df[null_df['null_values'] != 0]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "id": "df9fdde7",
   "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>null_values</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>column</th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "Empty DataFrame\n",
       "Columns: [null_values]\n",
       "Index: []"
      ]
     },
     "execution_count": 13,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Print null values\n",
    "# There are no more null values\n",
    "null_df"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "237df7d5",
   "metadata": {},
   "source": [
    "# Model Training"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "c24179d4",
   "metadata": {},
   "source": [
    "I chose two different models here including:\n",
    "* Random Forest\n",
    "* Decision Tree\n",
    "* Linear Regression\n",
    "\n",
    "Each of these models are first analyzed using a .20 test and .80 train split.\n",
    "\n",
    "The results are shown under the Accuracy Results section. Each model showed improvement, especially decision trees which shot up by 8% accuracy."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "id": "d8100292",
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn.model_selection import train_test_split\n",
    "from sklearn.preprocessing import StandardScaler\n",
    "from sklearn.metrics import mean_squared_error"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "id": "d80875c2",
   "metadata": {},
   "outputs": [],
   "source": [
    "# Remove unnecessary columns and the result variable\n",
    "X = df.drop(columns={'state', 'date', 'deaths_covid'})\n",
    "\n",
    "# Extract result variable\n",
    "y = df['deaths_covid']\n",
    "\n",
    "# Set random_state constant\n",
    "random_state = 42"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "id": "ffa1b405",
   "metadata": {},
   "outputs": [],
   "source": [
    "# Split data into train/test splits\n",
    "X_train, X_test, y_train, y_test = train_test_split(X, y, test_size = 0.2, random_state = random_state)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "id": "be61a67e",
   "metadata": {},
   "outputs": [],
   "source": [
    "# Scale data using StandardScaler()\n",
    "sc = StandardScaler()\n",
    "X_train = sc.fit_transform(X_train)\n",
    "X_test = sc.transform(X_test)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "0818ebc6",
   "metadata": {},
   "source": [
    "## Random Forest"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "id": "d909773e",
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn.ensemble import RandomForestClassifier"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "id": "20cf3543",
   "metadata": {},
   "outputs": [],
   "source": [
    "# Define classifier\n",
    "rfc = RandomForestClassifier(random_state=random_state)\n",
    "\n",
    "# Run predictions using random forest classifier\n",
    "rfc.fit(X_train, y_train)\n",
    "pred_rfc = rfc.predict(X_test)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "id": "2dd749f9",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Random Forest Root Mean Squared Error: 10.288897409494494\n"
     ]
    }
   ],
   "source": [
    "# Calculate mean squared error\n",
    "mse_rf = mean_squared_error(y_test, pred_rfc)\n",
    "\n",
    "# Calculate root mean squared error\n",
    "rmse_rf = np.sqrt(mse_rf)  \n",
    "\n",
    "# Print RMSE\n",
    "print(\"Random Forest Root Mean Squared Error:\", rmse_rf)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "id": "fc764cdc",
   "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>Feature</th>\n",
       "      <th>Importance</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>staffed_icu_adult_patients_confirmed_covid</td>\n",
       "      <td>0.026500</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>staffed_icu_adult_patients_confirmed_and_suspe...</td>\n",
       "      <td>0.025898</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>adult_icu_bed_covid_utilization_numerator</td>\n",
       "      <td>0.023353</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>percent_of_inpatients_with_covid_numerator</td>\n",
       "      <td>0.020792</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>total_adult_patients_hospitalized_confirmed_covid</td>\n",
       "      <td>0.019461</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>total_adult_patients_hospitalized_confirmed_an...</td>\n",
       "      <td>0.019099</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>deaths_covid_coverage</td>\n",
       "      <td>0.018980</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>inpatient_bed_covid_utilization_numerator</td>\n",
       "      <td>0.018109</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8</th>\n",
       "      <td>previous_day_admission_adult_covid_confirmed_5...</td>\n",
       "      <td>0.016019</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9</th>\n",
       "      <td>inpatient_beds_used_covid</td>\n",
       "      <td>0.015906</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>10</th>\n",
       "      <td>adult_icu_bed_covid_utilization</td>\n",
       "      <td>0.015842</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>11</th>\n",
       "      <td>previous_day_admission_adult_covid_confirmed</td>\n",
       "      <td>0.014609</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>12</th>\n",
       "      <td>inpatient_beds_utilization</td>\n",
       "      <td>0.013619</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>13</th>\n",
       "      <td>inpatient_bed_covid_utilization</td>\n",
       "      <td>0.013484</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>14</th>\n",
       "      <td>adult_icu_bed_utilization</td>\n",
       "      <td>0.013188</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>15</th>\n",
       "      <td>percent_of_inpatients_with_covid</td>\n",
       "      <td>0.013176</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>16</th>\n",
       "      <td>previous_day_admission_adult_covid_confirmed_4...</td>\n",
       "      <td>0.012333</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>17</th>\n",
       "      <td>adult_icu_bed_covid_utilization_denominator</td>\n",
       "      <td>0.012239</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>18</th>\n",
       "      <td>critical_staffing_shortage_today_no</td>\n",
       "      <td>0.012147</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>19</th>\n",
       "      <td>previous_day_admission_adult_covid_suspected</td>\n",
       "      <td>0.012105</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                                              Feature  Importance\n",
       "0          staffed_icu_adult_patients_confirmed_covid    0.026500\n",
       "1   staffed_icu_adult_patients_confirmed_and_suspe...    0.025898\n",
       "2           adult_icu_bed_covid_utilization_numerator    0.023353\n",
       "3          percent_of_inpatients_with_covid_numerator    0.020792\n",
       "4   total_adult_patients_hospitalized_confirmed_covid    0.019461\n",
       "5   total_adult_patients_hospitalized_confirmed_an...    0.019099\n",
       "6                               deaths_covid_coverage    0.018980\n",
       "7           inpatient_bed_covid_utilization_numerator    0.018109\n",
       "8   previous_day_admission_adult_covid_confirmed_5...    0.016019\n",
       "9                           inpatient_beds_used_covid    0.015906\n",
       "10                    adult_icu_bed_covid_utilization    0.015842\n",
       "11       previous_day_admission_adult_covid_confirmed    0.014609\n",
       "12                         inpatient_beds_utilization    0.013619\n",
       "13                    inpatient_bed_covid_utilization    0.013484\n",
       "14                          adult_icu_bed_utilization    0.013188\n",
       "15                   percent_of_inpatients_with_covid    0.013176\n",
       "16  previous_day_admission_adult_covid_confirmed_4...    0.012333\n",
       "17        adult_icu_bed_covid_utilization_denominator    0.012239\n",
       "18                critical_staffing_shortage_today_no    0.012147\n",
       "19       previous_day_admission_adult_covid_suspected    0.012105"
      ]
     },
     "execution_count": 21,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Add random forest importances to dataframe\n",
    "Random_Forest_Importances = pd.DataFrame({\n",
    "    \"Feature\": X.columns, \n",
    "    \"Importance\": rfc.feature_importances_\n",
    "}).sort_values(\"Importance\", ascending=False).reset_index(drop=True)\n",
    "\n",
    "# Print top 20 values of the dataframe\n",
    "Random_Forest_Importances.head(20)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "c3fa8ae5",
   "metadata": {},
   "source": [
    "# Decision Trees"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "id": "e00fb574",
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn.tree import DecisionTreeRegressor"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "id": "8763eacb",
   "metadata": {},
   "outputs": [],
   "source": [
    "# Define classification\n",
    "dt = DecisionTreeRegressor(random_state=random_state)\n",
    "\n",
    "# Run prediction using decision trees\n",
    "dt.fit(X_train, y_train)\n",
    "pred_dt = dt.predict(X_test)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "id": "f870375a",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Random Forest Root Mean Squared Error: 11.982188903538527\n"
     ]
    }
   ],
   "source": [
    "# Calculate mean squared error\n",
    "mse_dt = mean_squared_error(y_test, pred_dt)\n",
    "\n",
    "# Calculate root mean squared error\n",
    "rmse_dt = np.sqrt(mse_dt)  \n",
    "\n",
    "# Print RMSE\n",
    "print(\"Random Forest Root Mean Squared Error:\", rmse_dt)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "id": "08205ed2",
   "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>Feature</th>\n",
       "      <th>Importance</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>staffed_icu_adult_patients_confirmed_covid</td>\n",
       "      <td>0.619773</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>inpatient_beds_coverage</td>\n",
       "      <td>0.067250</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>adult_icu_bed_covid_utilization_numerator</td>\n",
       "      <td>0.046706</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>total_adult_patients_hospitalized_confirmed_an...</td>\n",
       "      <td>0.036031</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>previous_day_admission_adult_covid_confirmed_u...</td>\n",
       "      <td>0.029505</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>previous_day_admission_adult_covid_suspected_6...</td>\n",
       "      <td>0.024762</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>previous_day_admission_adult_covid_suspected_7...</td>\n",
       "      <td>0.023170</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>deaths_covid_coverage</td>\n",
       "      <td>0.012864</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8</th>\n",
       "      <td>staffed_pediatric_icu_bed_occupancy</td>\n",
       "      <td>0.010547</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9</th>\n",
       "      <td>total_adult_patients_hospitalized_confirmed_covid</td>\n",
       "      <td>0.007443</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>10</th>\n",
       "      <td>previous_day_admission_adult_covid_suspected_5...</td>\n",
       "      <td>0.007377</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>11</th>\n",
       "      <td>critical_staffing_shortage_today_not_reported</td>\n",
       "      <td>0.006928</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>12</th>\n",
       "      <td>adult_icu_bed_covid_utilization</td>\n",
       "      <td>0.006112</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>13</th>\n",
       "      <td>critical_staffing_shortage_today_yes</td>\n",
       "      <td>0.005054</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>14</th>\n",
       "      <td>adult_icu_bed_utilization_denominator</td>\n",
       "      <td>0.004770</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>15</th>\n",
       "      <td>staffed_adult_icu_bed_occupancy</td>\n",
       "      <td>0.004643</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>16</th>\n",
       "      <td>staffed_icu_pediatric_patients_confirmed_covid...</td>\n",
       "      <td>0.004381</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>17</th>\n",
       "      <td>total_staffed_pediatric_icu_beds</td>\n",
       "      <td>0.004099</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>18</th>\n",
       "      <td>staffed_icu_adult_patients_confirmed_and_suspe...</td>\n",
       "      <td>0.003889</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>19</th>\n",
       "      <td>inpatient_beds_used_covid</td>\n",
       "      <td>0.003455</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                                              Feature  Importance\n",
       "0          staffed_icu_adult_patients_confirmed_covid    0.619773\n",
       "1                             inpatient_beds_coverage    0.067250\n",
       "2           adult_icu_bed_covid_utilization_numerator    0.046706\n",
       "3   total_adult_patients_hospitalized_confirmed_an...    0.036031\n",
       "4   previous_day_admission_adult_covid_confirmed_u...    0.029505\n",
       "5   previous_day_admission_adult_covid_suspected_6...    0.024762\n",
       "6   previous_day_admission_adult_covid_suspected_7...    0.023170\n",
       "7                               deaths_covid_coverage    0.012864\n",
       "8                 staffed_pediatric_icu_bed_occupancy    0.010547\n",
       "9   total_adult_patients_hospitalized_confirmed_covid    0.007443\n",
       "10  previous_day_admission_adult_covid_suspected_5...    0.007377\n",
       "11      critical_staffing_shortage_today_not_reported    0.006928\n",
       "12                    adult_icu_bed_covid_utilization    0.006112\n",
       "13               critical_staffing_shortage_today_yes    0.005054\n",
       "14              adult_icu_bed_utilization_denominator    0.004770\n",
       "15                    staffed_adult_icu_bed_occupancy    0.004643\n",
       "16  staffed_icu_pediatric_patients_confirmed_covid...    0.004381\n",
       "17                   total_staffed_pediatric_icu_beds    0.004099\n",
       "18  staffed_icu_adult_patients_confirmed_and_suspe...    0.003889\n",
       "19                          inpatient_beds_used_covid    0.003455"
      ]
     },
     "execution_count": 25,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Add decision tree importances to dataframe\n",
    "Decision_Tree_Importances = pd.DataFrame({\n",
    "    \"Feature\": X.columns, \n",
    "    \"Importance\": dt.feature_importances_\n",
    "}).sort_values(\"Importance\", ascending=False).reset_index(drop=True)\n",
    "\n",
    "# Print top 20 values of the dataframe\n",
    "Decision_Tree_Importances.head(20)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "30bb6cf8",
   "metadata": {},
   "source": [
    "## Linear Regression"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "id": "bb38e278",
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn.linear_model import LinearRegression"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "id": "cae203a5",
   "metadata": {},
   "outputs": [],
   "source": [
    "# Define classification\n",
    "lr = LinearRegression()\n",
    "\n",
    "# Run prediction using linear regression\n",
    "lr.fit(X_train, y_train)\n",
    "pred_lr = lr.predict(X_test)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "id": "ac3c270f",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Linear Regression Root Mean Squared Error: 11.579237777122788\n"
     ]
    }
   ],
   "source": [
    "# Calculate mean squared error\n",
    "mse_lr = mean_squared_error(y_test, pred_lr)\n",
    "\n",
    "# Calculate root mean squared error\n",
    "rmse_lr = np.sqrt(mse_lr)\n",
    "\n",
    "# Print RMSE\n",
    "print(\"Linear Regression Root Mean Squared Error:\", rmse_lr)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "id": "416424f9",
   "metadata": {
    "scrolled": false
   },
   "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>Feature</th>\n",
       "      <th>Coefficient</th>\n",
       "      <th>Abs_Coefficient</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>critical_staffing_shortage_today_no</td>\n",
       "      <td>-1.407022e+12</td>\n",
       "      <td>1.407022e+12</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>critical_staffing_shortage_today_not_reported</td>\n",
       "      <td>-1.399735e+12</td>\n",
       "      <td>1.399735e+12</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>critical_staffing_shortage_anticipated_within_...</td>\n",
       "      <td>1.342986e+12</td>\n",
       "      <td>1.342986e+12</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>critical_staffing_shortage_anticipated_within_...</td>\n",
       "      <td>1.129891e+12</td>\n",
       "      <td>1.129891e+12</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>critical_staffing_shortage_anticipated_within_...</td>\n",
       "      <td>4.675849e+11</td>\n",
       "      <td>4.675849e+11</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>critical_staffing_shortage_today_yes</td>\n",
       "      <td>-3.752257e+11</td>\n",
       "      <td>3.752257e+11</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>inpatient_beds_coverage</td>\n",
       "      <td>1.577878e+02</td>\n",
       "      <td>1.577878e+02</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>inpatient_bed_covid_utilization_coverage</td>\n",
       "      <td>-1.557128e+02</td>\n",
       "      <td>1.557128e+02</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8</th>\n",
       "      <td>adult_icu_bed_utilization_numerator</td>\n",
       "      <td>1.069342e+02</td>\n",
       "      <td>1.069342e+02</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9</th>\n",
       "      <td>staffed_adult_icu_bed_occupancy</td>\n",
       "      <td>-1.059861e+02</td>\n",
       "      <td>1.059861e+02</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>10</th>\n",
       "      <td>percent_of_inpatients_with_covid_coverage</td>\n",
       "      <td>1.012391e+02</td>\n",
       "      <td>1.012391e+02</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>11</th>\n",
       "      <td>previous_day_admission_adult_covid_confirmed_5...</td>\n",
       "      <td>-1.000766e+02</td>\n",
       "      <td>1.000766e+02</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>12</th>\n",
       "      <td>all_pediatric_inpatient_beds_coverage</td>\n",
       "      <td>-8.345058e+01</td>\n",
       "      <td>8.345058e+01</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>13</th>\n",
       "      <td>inpatient_beds_utilization_coverage</td>\n",
       "      <td>-8.208582e+01</td>\n",
       "      <td>8.208582e+01</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>14</th>\n",
       "      <td>icu_patients_confirmed_influenza_coverage</td>\n",
       "      <td>-7.412975e+01</td>\n",
       "      <td>7.412975e+01</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>15</th>\n",
       "      <td>staffed_icu_adult_patients_confirmed_and_suspe...</td>\n",
       "      <td>-7.337561e+01</td>\n",
       "      <td>7.337561e+01</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>16</th>\n",
       "      <td>staffed_icu_adult_patients_confirmed_covid_cov...</td>\n",
       "      <td>7.295163e+01</td>\n",
       "      <td>7.295163e+01</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>17</th>\n",
       "      <td>previous_day_admission_adult_covid_suspected_7...</td>\n",
       "      <td>-7.136121e+01</td>\n",
       "      <td>7.136121e+01</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>18</th>\n",
       "      <td>staffed_pediatric_icu_bed_occupancy_coverage</td>\n",
       "      <td>7.107138e+01</td>\n",
       "      <td>7.107138e+01</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>19</th>\n",
       "      <td>adult_icu_bed_covid_utilization_numerator</td>\n",
       "      <td>-7.093283e+01</td>\n",
       "      <td>7.093283e+01</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                                              Feature   Coefficient  \\\n",
       "0                 critical_staffing_shortage_today_no -1.407022e+12   \n",
       "1       critical_staffing_shortage_today_not_reported -1.399735e+12   \n",
       "2   critical_staffing_shortage_anticipated_within_...  1.342986e+12   \n",
       "3   critical_staffing_shortage_anticipated_within_...  1.129891e+12   \n",
       "4   critical_staffing_shortage_anticipated_within_...  4.675849e+11   \n",
       "5                critical_staffing_shortage_today_yes -3.752257e+11   \n",
       "6                             inpatient_beds_coverage  1.577878e+02   \n",
       "7            inpatient_bed_covid_utilization_coverage -1.557128e+02   \n",
       "8                 adult_icu_bed_utilization_numerator  1.069342e+02   \n",
       "9                     staffed_adult_icu_bed_occupancy -1.059861e+02   \n",
       "10          percent_of_inpatients_with_covid_coverage  1.012391e+02   \n",
       "11  previous_day_admission_adult_covid_confirmed_5... -1.000766e+02   \n",
       "12              all_pediatric_inpatient_beds_coverage -8.345058e+01   \n",
       "13                inpatient_beds_utilization_coverage -8.208582e+01   \n",
       "14          icu_patients_confirmed_influenza_coverage -7.412975e+01   \n",
       "15  staffed_icu_adult_patients_confirmed_and_suspe... -7.337561e+01   \n",
       "16  staffed_icu_adult_patients_confirmed_covid_cov...  7.295163e+01   \n",
       "17  previous_day_admission_adult_covid_suspected_7... -7.136121e+01   \n",
       "18       staffed_pediatric_icu_bed_occupancy_coverage  7.107138e+01   \n",
       "19          adult_icu_bed_covid_utilization_numerator -7.093283e+01   \n",
       "\n",
       "    Abs_Coefficient  \n",
       "0      1.407022e+12  \n",
       "1      1.399735e+12  \n",
       "2      1.342986e+12  \n",
       "3      1.129891e+12  \n",
       "4      4.675849e+11  \n",
       "5      3.752257e+11  \n",
       "6      1.577878e+02  \n",
       "7      1.557128e+02  \n",
       "8      1.069342e+02  \n",
       "9      1.059861e+02  \n",
       "10     1.012391e+02  \n",
       "11     1.000766e+02  \n",
       "12     8.345058e+01  \n",
       "13     8.208582e+01  \n",
       "14     7.412975e+01  \n",
       "15     7.337561e+01  \n",
       "16     7.295163e+01  \n",
       "17     7.136121e+01  \n",
       "18     7.107138e+01  \n",
       "19     7.093283e+01  "
      ]
     },
     "execution_count": 29,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Add linear regression coefficients to dataframe\n",
    "Linear_Regression_Coefficients = pd.DataFrame({\n",
    "    'Feature': X.columns,\n",
    "    'Coefficient': lr.coef_\n",
    "}).sort_values(by='Coefficient', ascending=False).reset_index(drop=True)\n",
    "\n",
    "# Create an absolute value column and sort by that column\n",
    "Linear_Regression_Coefficients['Abs_Coefficient'] = Linear_Regression_Coefficients['Coefficient'].apply(lambda x: abs(float(x)))\n",
    "Linear_Regression_Coefficients = Linear_Regression_Coefficients.sort_values(by='Abs_Coefficient', ascending=False).head(20).reset_index(drop=True)\n",
    "\n",
    "# Print top 20 values of the dataframe\n",
    "Linear_Regression_Coefficients.head(20)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "bcd0249e",
   "metadata": {},
   "source": [
    "# Results"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "ba850582",
   "metadata": {},
   "source": [
    "## Key Findings and Optimal Model\n",
    "The results show the reliability of the models as follow: \n",
    "*   The Decision Tree model is the most reliable, with a root mean squared error of 11.2 \n",
    "*   The Random Forest model is next with a RMSE of 12.3 \n",
    "*   The Linear Regression model is last with a RMSE of 15.7. \n",
    "\n",
    "The most common features among the algorithms are the following:\n",
    "*   staffed_icu_adult_patients_confirmed_covid\n",
    "*   adult_icu_bed_covid_utilization_numerator\n",
    "*   total_adult_patients_hospitalized_confirmed_covid\n",
    "*   total_adult_patients_hospitalized_confirmed_and_suspected_covid\n",
    "*   deaths_covid_coverage\n",
    "*   inpatient_beds_coverage\n",
    "\n",
    "One potential issue is that the features extracted from Linear Regression are assumed to be more important the higher the coefficient is. This is not necessarily the case and can skew the results. More research may be needed."
   ]
  },
  {
   "cell_type": "markdown",
   "id": "0fa7b66e",
   "metadata": {},
   "source": [
    "## Model Accuracy"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "id": "1260e6c5",
   "metadata": {},
   "outputs": [],
   "source": [
    "# Create dataframe using model results\n",
    "result_models = ['Random Forest', 'Decision Tree', 'Linear Regression']\n",
    "result_stats = [rmse_rf, rmse_dt, rmse_lr]\n",
    "results = pd.DataFrame([result_models, result_stats])\n",
    "results = results.T.rename(columns={0:'Models', 1:'RMSE'}).set_index('Models').sort_values(by='RMSE')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "id": "428ff430",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Plot the results dataframe with appropriate labels\n",
    "ax = results.plot(kind='bar', legend=False)\n",
    "\n",
    "ax.set_title('Root Mean Squared Error for Models')\n",
    "ax.set_xlabel('Models')\n",
    "ax.set_ylabel('Root Mean Squared Error')\n",
    "\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "58c5805d",
   "metadata": {},
   "source": [
    "## Top Features"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "id": "84f75fbe",
   "metadata": {},
   "outputs": [],
   "source": [
    "from collections import Counter"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 33,
   "id": "4dbe690e",
   "metadata": {},
   "outputs": [],
   "source": [
    "# Create temporary dataframes with top 10 features from each\n",
    "RF_top10 = Random_Forest_Importances[['Feature']].rename(columns={'Feature':'Random Forest'}).head(10)\n",
    "DT_top10 = Decision_Tree_Importances[['Feature']].rename(columns={'Feature':'Decision Tree'}).head(10)\n",
    "LR_top10 = Linear_Regression_Coefficients[['Feature']].rename(columns={'Feature':'Linear Regression'}).head(10)\n",
    "\n",
    "# Combine dataframes into a top 10 list for each\n",
    "top10 = RF_top10.merge(DT_top10, left_index=True, right_index=True, how='outer')\n",
    "top10 = top10.merge(LR_top10, left_index=True, right_index=True, how='outer')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 34,
   "id": "2bef7ad6",
   "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>Random Forest</th>\n",
       "      <th>Decision Tree</th>\n",
       "      <th>Linear Regression</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>staffed_icu_adult_patients_confirmed_covid</td>\n",
       "      <td>staffed_icu_adult_patients_confirmed_covid</td>\n",
       "      <td>critical_staffing_shortage_today_no</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>staffed_icu_adult_patients_confirmed_and_suspe...</td>\n",
       "      <td>inpatient_beds_coverage</td>\n",
       "      <td>critical_staffing_shortage_today_not_reported</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>adult_icu_bed_covid_utilization_numerator</td>\n",
       "      <td>adult_icu_bed_covid_utilization_numerator</td>\n",
       "      <td>critical_staffing_shortage_anticipated_within_...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>percent_of_inpatients_with_covid_numerator</td>\n",
       "      <td>total_adult_patients_hospitalized_confirmed_an...</td>\n",
       "      <td>critical_staffing_shortage_anticipated_within_...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>total_adult_patients_hospitalized_confirmed_covid</td>\n",
       "      <td>previous_day_admission_adult_covid_confirmed_u...</td>\n",
       "      <td>critical_staffing_shortage_anticipated_within_...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>total_adult_patients_hospitalized_confirmed_an...</td>\n",
       "      <td>previous_day_admission_adult_covid_suspected_6...</td>\n",
       "      <td>critical_staffing_shortage_today_yes</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>deaths_covid_coverage</td>\n",
       "      <td>previous_day_admission_adult_covid_suspected_7...</td>\n",
       "      <td>inpatient_beds_coverage</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>inpatient_bed_covid_utilization_numerator</td>\n",
       "      <td>deaths_covid_coverage</td>\n",
       "      <td>inpatient_bed_covid_utilization_coverage</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8</th>\n",
       "      <td>previous_day_admission_adult_covid_confirmed_5...</td>\n",
       "      <td>staffed_pediatric_icu_bed_occupancy</td>\n",
       "      <td>adult_icu_bed_utilization_numerator</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9</th>\n",
       "      <td>inpatient_beds_used_covid</td>\n",
       "      <td>total_adult_patients_hospitalized_confirmed_covid</td>\n",
       "      <td>staffed_adult_icu_bed_occupancy</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                                       Random Forest  \\\n",
       "0         staffed_icu_adult_patients_confirmed_covid   \n",
       "1  staffed_icu_adult_patients_confirmed_and_suspe...   \n",
       "2          adult_icu_bed_covid_utilization_numerator   \n",
       "3         percent_of_inpatients_with_covid_numerator   \n",
       "4  total_adult_patients_hospitalized_confirmed_covid   \n",
       "5  total_adult_patients_hospitalized_confirmed_an...   \n",
       "6                              deaths_covid_coverage   \n",
       "7          inpatient_bed_covid_utilization_numerator   \n",
       "8  previous_day_admission_adult_covid_confirmed_5...   \n",
       "9                          inpatient_beds_used_covid   \n",
       "\n",
       "                                       Decision Tree  \\\n",
       "0         staffed_icu_adult_patients_confirmed_covid   \n",
       "1                            inpatient_beds_coverage   \n",
       "2          adult_icu_bed_covid_utilization_numerator   \n",
       "3  total_adult_patients_hospitalized_confirmed_an...   \n",
       "4  previous_day_admission_adult_covid_confirmed_u...   \n",
       "5  previous_day_admission_adult_covid_suspected_6...   \n",
       "6  previous_day_admission_adult_covid_suspected_7...   \n",
       "7                              deaths_covid_coverage   \n",
       "8                staffed_pediatric_icu_bed_occupancy   \n",
       "9  total_adult_patients_hospitalized_confirmed_covid   \n",
       "\n",
       "                                   Linear Regression  \n",
       "0                critical_staffing_shortage_today_no  \n",
       "1      critical_staffing_shortage_today_not_reported  \n",
       "2  critical_staffing_shortage_anticipated_within_...  \n",
       "3  critical_staffing_shortage_anticipated_within_...  \n",
       "4  critical_staffing_shortage_anticipated_within_...  \n",
       "5               critical_staffing_shortage_today_yes  \n",
       "6                            inpatient_beds_coverage  \n",
       "7           inpatient_bed_covid_utilization_coverage  \n",
       "8                adult_icu_bed_utilization_numerator  \n",
       "9                    staffed_adult_icu_bed_occupancy  "
      ]
     },
     "execution_count": 34,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Print significant features on models \n",
    "top10"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 35,
   "id": "b9a796f1",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "staffed_icu_adult_patients_confirmed_covid\n",
      "adult_icu_bed_covid_utilization_numerator\n",
      "total_adult_patients_hospitalized_confirmed_covid\n",
      "total_adult_patients_hospitalized_confirmed_and_suspected_covid\n",
      "deaths_covid_coverage\n",
      "inpatient_beds_coverage\n"
     ]
    }
   ],
   "source": [
    "# Combine top10 dataframes into a single dataframe\n",
    "all_entries = pd.concat([top10['Random Forest'], top10['Decision Tree'], top10['Linear Regression']])\n",
    "\n",
    "# Count the frequency of each entry\n",
    "counter = Counter(all_entries)\n",
    "\n",
    "# Sort counter by value in descending order and get the most common entries\n",
    "most_common_entries = counter.most_common()\n",
    "\n",
    "# Get the highest count (the count of the first entry in the sorted list)\n",
    "highest_count = most_common_entries[0][1]\n",
    "\n",
    "# Print only the most common entries (those with a count equal to highest_count)\n",
    "for entry, count in most_common_entries:\n",
    "    if count == highest_count:\n",
    "        print(entry)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "e4e2235e",
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3 (ipykernel)",
   "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": "3.10.11"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}