Naive_Bayes.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Naive Bayes Classification"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [],
   "source": [
    "import pre_ml\n",
    "import mlpipeline as pipeline\n",
    "import pandas as pd\n",
    "import numpy as np\n",
    "import sklearn"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Requesting data to datos abiertos Mexico\n",
      "Getting zip raw data into directory, will delete soon\n",
      "Raw data deleted. If you specified filename, clean data will be saved in data directory\n"
     ]
    }
   ],
   "source": [
    "df = pre_ml.data()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "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>entidad_res</th>\n",
       "      <th>municipio_res</th>\n",
       "      <th>CVE_MUN</th>\n",
       "      <th>embarazo</th>\n",
       "      <th>edad</th>\n",
       "      <th>diabetes</th>\n",
       "      <th>epoc</th>\n",
       "      <th>asma</th>\n",
       "      <th>inmusupr</th>\n",
       "      <th>hipertension</th>\n",
       "      <th>...</th>\n",
       "      <th>obesidad</th>\n",
       "      <th>tabaquismo</th>\n",
       "      <th>pobreza</th>\n",
       "      <th>Densidad_pob</th>\n",
       "      <th>medicos</th>\n",
       "      <th>camas_hosp</th>\n",
       "      <th>enfermeras</th>\n",
       "      <th>hospitalizado</th>\n",
       "      <th>muertos</th>\n",
       "      <th>covid_grave</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>25</td>\n",
       "      <td>6</td>\n",
       "      <td>25006</td>\n",
       "      <td>0.0</td>\n",
       "      <td>27</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>...</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>31.2</td>\n",
       "      <td>152.715706</td>\n",
       "      <td>2.637944</td>\n",
       "      <td>10.489463</td>\n",
       "      <td>30.502528</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>27</td>\n",
       "      <td>4</td>\n",
       "      <td>27004</td>\n",
       "      <td>0.0</td>\n",
       "      <td>44</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>...</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>36.3</td>\n",
       "      <td>430.537057</td>\n",
       "      <td>5.448810</td>\n",
       "      <td>13.926240</td>\n",
       "      <td>41.575909</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>27</td>\n",
       "      <td>5</td>\n",
       "      <td>27005</td>\n",
       "      <td>0.0</td>\n",
       "      <td>52</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>...</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>63.6</td>\n",
       "      <td>283.191451</td>\n",
       "      <td>3.401376</td>\n",
       "      <td>2.436121</td>\n",
       "      <td>8.319582</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>15</td>\n",
       "      <td>58</td>\n",
       "      <td>15058</td>\n",
       "      <td>0.0</td>\n",
       "      <td>55</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>...</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>37.4</td>\n",
       "      <td>17954.252292</td>\n",
       "      <td>0.810012</td>\n",
       "      <td>2.870259</td>\n",
       "      <td>10.248409</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>2</td>\n",
       "      <td>4</td>\n",
       "      <td>2004</td>\n",
       "      <td>0.0</td>\n",
       "      <td>47</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>...</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>31.1</td>\n",
       "      <td>1449.318075</td>\n",
       "      <td>0.726447</td>\n",
       "      <td>4.889549</td>\n",
       "      <td>11.701390</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>5 rows × 21 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "   entidad_res  municipio_res  CVE_MUN  embarazo  edad  diabetes  epoc  asma  \\\n",
       "0           25              6    25006       0.0    27       0.0   0.0   0.0   \n",
       "1           27              4    27004       0.0    44       0.0   0.0   0.0   \n",
       "2           27              5    27005       0.0    52       1.0   0.0   0.0   \n",
       "3           15             58    15058       0.0    55       1.0   0.0   0.0   \n",
       "4            2              4     2004       0.0    47       0.0   0.0   0.0   \n",
       "\n",
       "   inmusupr  hipertension  ...  obesidad  tabaquismo  pobreza  Densidad_pob  \\\n",
       "0       0.0           0.0  ...       0.0         0.0     31.2    152.715706   \n",
       "1       0.0           0.0  ...       0.0         0.0     36.3    430.537057   \n",
       "2       0.0           1.0  ...       0.0         0.0     63.6    283.191451   \n",
       "3       0.0           1.0  ...       1.0         0.0     37.4  17954.252292   \n",
       "4       0.0           0.0  ...       0.0         0.0     31.1   1449.318075   \n",
       "\n",
       "    medicos  camas_hosp  enfermeras  hospitalizado  muertos  covid_grave  \n",
       "0  2.637944   10.489463   30.502528              0        0            0  \n",
       "1  5.448810   13.926240   41.575909              0        0            0  \n",
       "2  3.401376    2.436121    8.319582              1        0            1  \n",
       "3  0.810012    2.870259   10.248409              0        0            0  \n",
       "4  0.726447    4.889549   11.701390              1        1            1  \n",
       "\n",
       "[5 rows x 21 columns]"
      ]
     },
     "execution_count": 4,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.head(5)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Check data:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "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>entidad_res</th>\n",
       "      <th>municipio_res</th>\n",
       "      <th>CVE_MUN</th>\n",
       "      <th>embarazo</th>\n",
       "      <th>edad</th>\n",
       "      <th>diabetes</th>\n",
       "      <th>epoc</th>\n",
       "      <th>asma</th>\n",
       "      <th>inmusupr</th>\n",
       "      <th>hipertension</th>\n",
       "      <th>...</th>\n",
       "      <th>obesidad</th>\n",
       "      <th>tabaquismo</th>\n",
       "      <th>pobreza</th>\n",
       "      <th>Densidad_pob</th>\n",
       "      <th>medicos</th>\n",
       "      <th>camas_hosp</th>\n",
       "      <th>enfermeras</th>\n",
       "      <th>hospitalizado</th>\n",
       "      <th>muertos</th>\n",
       "      <th>covid_grave</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>count</th>\n",
       "      <td>93435.000000</td>\n",
       "      <td>93435.000000</td>\n",
       "      <td>93435.000000</td>\n",
       "      <td>93435.000000</td>\n",
       "      <td>93435.000000</td>\n",
       "      <td>93435.000000</td>\n",
       "      <td>93435.000000</td>\n",
       "      <td>93435.000000</td>\n",
       "      <td>93435.000000</td>\n",
       "      <td>93435.000000</td>\n",
       "      <td>...</td>\n",
       "      <td>93435.000000</td>\n",
       "      <td>93435.000000</td>\n",
       "      <td>93435.000000</td>\n",
       "      <td>93435.000000</td>\n",
       "      <td>93435.000000</td>\n",
       "      <td>93435.000000</td>\n",
       "      <td>93435.000000</td>\n",
       "      <td>93435.000000</td>\n",
       "      <td>93435.000000</td>\n",
       "      <td>93435.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>mean</th>\n",
       "      <td>15.057537</td>\n",
       "      <td>36.730593</td>\n",
       "      <td>15094.267919</td>\n",
       "      <td>0.006336</td>\n",
       "      <td>46.241034</td>\n",
       "      <td>0.174282</td>\n",
       "      <td>0.019693</td>\n",
       "      <td>0.029026</td>\n",
       "      <td>0.015701</td>\n",
       "      <td>0.208070</td>\n",
       "      <td>...</td>\n",
       "      <td>0.203885</td>\n",
       "      <td>0.082603</td>\n",
       "      <td>36.873778</td>\n",
       "      <td>4743.009499</td>\n",
       "      <td>2.770005</td>\n",
       "      <td>11.096877</td>\n",
       "      <td>29.604093</td>\n",
       "      <td>0.348981</td>\n",
       "      <td>0.108814</td>\n",
       "      <td>0.360143</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>std</th>\n",
       "      <td>8.105571</td>\n",
       "      <td>52.420438</td>\n",
       "      <td>8121.567737</td>\n",
       "      <td>0.079347</td>\n",
       "      <td>16.014916</td>\n",
       "      <td>0.379353</td>\n",
       "      <td>0.138943</td>\n",
       "      <td>0.167879</td>\n",
       "      <td>0.124316</td>\n",
       "      <td>0.405929</td>\n",
       "      <td>...</td>\n",
       "      <td>0.402887</td>\n",
       "      <td>0.275282</td>\n",
       "      <td>14.105819</td>\n",
       "      <td>5873.026693</td>\n",
       "      <td>2.008568</td>\n",
       "      <td>10.075898</td>\n",
       "      <td>24.243500</td>\n",
       "      <td>0.476650</td>\n",
       "      <td>0.311407</td>\n",
       "      <td>0.480044</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>min</th>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1001.000000</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>3.200000</td>\n",
       "      <td>0.446811</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>9.000000</td>\n",
       "      <td>6.000000</td>\n",
       "      <td>9007.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>34.000000</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>28.700000</td>\n",
       "      <td>254.086894</td>\n",
       "      <td>1.506770</td>\n",
       "      <td>5.474526</td>\n",
       "      <td>15.946373</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>50%</th>\n",
       "      <td>15.000000</td>\n",
       "      <td>15.000000</td>\n",
       "      <td>15024.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>45.000000</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>34.500000</td>\n",
       "      <td>1814.604740</td>\n",
       "      <td>2.248080</td>\n",
       "      <td>8.238176</td>\n",
       "      <td>21.720906</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>75%</th>\n",
       "      <td>21.000000</td>\n",
       "      <td>50.000000</td>\n",
       "      <td>21119.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>57.000000</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>41.200000</td>\n",
       "      <td>8183.315370</td>\n",
       "      <td>3.759853</td>\n",
       "      <td>13.572164</td>\n",
       "      <td>36.168421</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>1.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>max</th>\n",
       "      <td>32.000000</td>\n",
       "      <td>999.000000</td>\n",
       "      <td>32057.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>120.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>...</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>99.100000</td>\n",
       "      <td>17954.252292</td>\n",
       "      <td>50.068722</td>\n",
       "      <td>297.467112</td>\n",
       "      <td>559.591596</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>8 rows × 21 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "        entidad_res  municipio_res       CVE_MUN      embarazo          edad  \\\n",
       "count  93435.000000   93435.000000  93435.000000  93435.000000  93435.000000   \n",
       "mean      15.057537      36.730593  15094.267919      0.006336     46.241034   \n",
       "std        8.105571      52.420438   8121.567737      0.079347     16.014916   \n",
       "min        1.000000       1.000000   1001.000000      0.000000      0.000000   \n",
       "25%        9.000000       6.000000   9007.000000      0.000000     34.000000   \n",
       "50%       15.000000      15.000000  15024.000000      0.000000     45.000000   \n",
       "75%       21.000000      50.000000  21119.000000      0.000000     57.000000   \n",
       "max       32.000000     999.000000  32057.000000      1.000000    120.000000   \n",
       "\n",
       "           diabetes          epoc          asma      inmusupr  hipertension  \\\n",
       "count  93435.000000  93435.000000  93435.000000  93435.000000  93435.000000   \n",
       "mean       0.174282      0.019693      0.029026      0.015701      0.208070   \n",
       "std        0.379353      0.138943      0.167879      0.124316      0.405929   \n",
       "min        0.000000      0.000000      0.000000      0.000000      0.000000   \n",
       "25%        0.000000      0.000000      0.000000      0.000000      0.000000   \n",
       "50%        0.000000      0.000000      0.000000      0.000000      0.000000   \n",
       "75%        0.000000      0.000000      0.000000      0.000000      0.000000   \n",
       "max        1.000000      1.000000      1.000000      1.000000      1.000000   \n",
       "\n",
       "       ...      obesidad    tabaquismo       pobreza  Densidad_pob  \\\n",
       "count  ...  93435.000000  93435.000000  93435.000000  93435.000000   \n",
       "mean   ...      0.203885      0.082603     36.873778   4743.009499   \n",
       "std    ...      0.402887      0.275282     14.105819   5873.026693   \n",
       "min    ...      0.000000      0.000000      3.200000      0.446811   \n",
       "25%    ...      0.000000      0.000000     28.700000    254.086894   \n",
       "50%    ...      0.000000      0.000000     34.500000   1814.604740   \n",
       "75%    ...      0.000000      0.000000     41.200000   8183.315370   \n",
       "max    ...      1.000000      1.000000     99.100000  17954.252292   \n",
       "\n",
       "            medicos    camas_hosp    enfermeras  hospitalizado       muertos  \\\n",
       "count  93435.000000  93435.000000  93435.000000   93435.000000  93435.000000   \n",
       "mean       2.770005     11.096877     29.604093       0.348981      0.108814   \n",
       "std        2.008568     10.075898     24.243500       0.476650      0.311407   \n",
       "min        0.000000      0.000000      0.000000       0.000000      0.000000   \n",
       "25%        1.506770      5.474526     15.946373       0.000000      0.000000   \n",
       "50%        2.248080      8.238176     21.720906       0.000000      0.000000   \n",
       "75%        3.759853     13.572164     36.168421       1.000000      0.000000   \n",
       "max       50.068722    297.467112    559.591596       1.000000      1.000000   \n",
       "\n",
       "        covid_grave  \n",
       "count  93435.000000  \n",
       "mean       0.360143  \n",
       "std        0.480044  \n",
       "min        0.000000  \n",
       "25%        0.000000  \n",
       "50%        0.000000  \n",
       "75%        1.000000  \n",
       "max        1.000000  \n",
       "\n",
       "[8 rows x 21 columns]"
      ]
     },
     "execution_count": 5,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.describe()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Drop outlier observations for age"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [],
   "source": [
    "df = df.drop(df[df.edad > (df.edad.mean()+3*df.edad.std())].index)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Re-scale poverty measurement"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [],
   "source": [
    "df['pobreza'] = df['pobreza'] / 100"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Probability of dying:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "P(dying) =     0.1086\n",
      "P(not dying) =     0.8914\n"
     ]
    }
   ],
   "source": [
    "deaths = df[df['muertos']==1]['muertos'].count()\n",
    "no_deaths = df[df['muertos']==0]['muertos'].count()\n",
    "cases = df['muertos'].count()\n",
    "\n",
    "# Probability of dying for COVID-19\n",
    "p_rip = [deaths / cases,\n",
    "        no_deaths / cases]\n",
    "\n",
    "print(\"P(dying) = {:10.4f}\\nP(not dying) = {:10.4f}\".format(p_rip[0],p_rip[1]))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Split data into train/test sets:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [],
   "source": [
    "train, test = pipeline.split_data(df.loc[:,'embarazo':], .20, seed=0)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Check for balance in train/test:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0    66566\n",
       "1     8094\n",
       "Name: muertos, dtype: int64"
      ]
     },
     "execution_count": 11,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "train['muertos'].value_counts()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0    48608\n",
       "1    26052\n",
       "Name: hospitalizado, dtype: int64"
      ]
     },
     "execution_count": 12,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "train['hospitalizado'].value_counts()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0    16624\n",
       "1     2041\n",
       "Name: muertos, dtype: int64"
      ]
     },
     "execution_count": 13,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "test['muertos'].value_counts()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0    12166\n",
       "1     6499\n",
       "Name: hospitalizado, dtype: int64"
      ]
     },
     "execution_count": 14,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "test['hospitalizado'].value_counts()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "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>embarazo</th>\n",
       "      <th>edad</th>\n",
       "      <th>diabetes</th>\n",
       "      <th>epoc</th>\n",
       "      <th>asma</th>\n",
       "      <th>inmusupr</th>\n",
       "      <th>hipertension</th>\n",
       "      <th>cardiovascular</th>\n",
       "      <th>obesidad</th>\n",
       "      <th>tabaquismo</th>\n",
       "      <th>pobreza</th>\n",
       "      <th>Densidad_pob</th>\n",
       "      <th>medicos</th>\n",
       "      <th>camas_hosp</th>\n",
       "      <th>enfermeras</th>\n",
       "      <th>hospitalizado</th>\n",
       "      <th>muertos</th>\n",
       "      <th>covid_grave</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>count</th>\n",
       "      <td>74660.000000</td>\n",
       "      <td>74660.000000</td>\n",
       "      <td>74660.000000</td>\n",
       "      <td>74660.000000</td>\n",
       "      <td>74660.000000</td>\n",
       "      <td>74660.000000</td>\n",
       "      <td>74660.000000</td>\n",
       "      <td>74660.000000</td>\n",
       "      <td>74660.000000</td>\n",
       "      <td>74660.000000</td>\n",
       "      <td>74660.000000</td>\n",
       "      <td>74660.000000</td>\n",
       "      <td>74660.000000</td>\n",
       "      <td>74660.000000</td>\n",
       "      <td>74660.000000</td>\n",
       "      <td>74660.000000</td>\n",
       "      <td>74660.000000</td>\n",
       "      <td>74660.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>mean</th>\n",
       "      <td>0.006389</td>\n",
       "      <td>46.201326</td>\n",
       "      <td>0.173841</td>\n",
       "      <td>0.019756</td>\n",
       "      <td>0.028891</td>\n",
       "      <td>0.015617</td>\n",
       "      <td>0.207822</td>\n",
       "      <td>0.025837</td>\n",
       "      <td>0.203523</td>\n",
       "      <td>0.082681</td>\n",
       "      <td>0.368844</td>\n",
       "      <td>4750.381673</td>\n",
       "      <td>2.771698</td>\n",
       "      <td>11.100244</td>\n",
       "      <td>29.591547</td>\n",
       "      <td>0.348942</td>\n",
       "      <td>0.108411</td>\n",
       "      <td>0.360139</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>std</th>\n",
       "      <td>0.079676</td>\n",
       "      <td>15.967045</td>\n",
       "      <td>0.378976</td>\n",
       "      <td>0.139162</td>\n",
       "      <td>0.167501</td>\n",
       "      <td>0.123991</td>\n",
       "      <td>0.405752</td>\n",
       "      <td>0.158650</td>\n",
       "      <td>0.402621</td>\n",
       "      <td>0.275402</td>\n",
       "      <td>0.141022</td>\n",
       "      <td>5873.716516</td>\n",
       "      <td>2.012931</td>\n",
       "      <td>10.098444</td>\n",
       "      <td>24.226843</td>\n",
       "      <td>0.476639</td>\n",
       "      <td>0.310901</td>\n",
       "      <td>0.480044</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>min</th>\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>0.000000</td>\n",
       "      <td>0.032000</td>\n",
       "      <td>0.446811</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>0.000000</td>\n",
       "      <td>34.000000</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>0.287000</td>\n",
       "      <td>254.086894</td>\n",
       "      <td>1.506770</td>\n",
       "      <td>5.474526</td>\n",
       "      <td>15.903346</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>50%</th>\n",
       "      <td>0.000000</td>\n",
       "      <td>45.000000</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>0.345000</td>\n",
       "      <td>1814.604740</td>\n",
       "      <td>2.248080</td>\n",
       "      <td>8.238176</td>\n",
       "      <td>21.720906</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>75%</th>\n",
       "      <td>0.000000</td>\n",
       "      <td>57.000000</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>0.412000</td>\n",
       "      <td>8183.315370</td>\n",
       "      <td>3.759853</td>\n",
       "      <td>13.672404</td>\n",
       "      <td>36.168421</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>1.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>max</th>\n",
       "      <td>1.000000</td>\n",
       "      <td>94.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.991000</td>\n",
       "      <td>17954.252292</td>\n",
       "      <td>50.068722</td>\n",
       "      <td>297.467112</td>\n",
       "      <td>559.591596</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "           embarazo          edad      diabetes          epoc          asma  \\\n",
       "count  74660.000000  74660.000000  74660.000000  74660.000000  74660.000000   \n",
       "mean       0.006389     46.201326      0.173841      0.019756      0.028891   \n",
       "std        0.079676     15.967045      0.378976      0.139162      0.167501   \n",
       "min        0.000000      0.000000      0.000000      0.000000      0.000000   \n",
       "25%        0.000000     34.000000      0.000000      0.000000      0.000000   \n",
       "50%        0.000000     45.000000      0.000000      0.000000      0.000000   \n",
       "75%        0.000000     57.000000      0.000000      0.000000      0.000000   \n",
       "max        1.000000     94.000000      1.000000      1.000000      1.000000   \n",
       "\n",
       "           inmusupr  hipertension  cardiovascular      obesidad    tabaquismo  \\\n",
       "count  74660.000000  74660.000000    74660.000000  74660.000000  74660.000000   \n",
       "mean       0.015617      0.207822        0.025837      0.203523      0.082681   \n",
       "std        0.123991      0.405752        0.158650      0.402621      0.275402   \n",
       "min        0.000000      0.000000        0.000000      0.000000      0.000000   \n",
       "25%        0.000000      0.000000        0.000000      0.000000      0.000000   \n",
       "50%        0.000000      0.000000        0.000000      0.000000      0.000000   \n",
       "75%        0.000000      0.000000        0.000000      0.000000      0.000000   \n",
       "max        1.000000      1.000000        1.000000      1.000000      1.000000   \n",
       "\n",
       "            pobreza  Densidad_pob       medicos    camas_hosp    enfermeras  \\\n",
       "count  74660.000000  74660.000000  74660.000000  74660.000000  74660.000000   \n",
       "mean       0.368844   4750.381673      2.771698     11.100244     29.591547   \n",
       "std        0.141022   5873.716516      2.012931     10.098444     24.226843   \n",
       "min        0.032000      0.446811      0.000000      0.000000      0.000000   \n",
       "25%        0.287000    254.086894      1.506770      5.474526     15.903346   \n",
       "50%        0.345000   1814.604740      2.248080      8.238176     21.720906   \n",
       "75%        0.412000   8183.315370      3.759853     13.672404     36.168421   \n",
       "max        0.991000  17954.252292     50.068722    297.467112    559.591596   \n",
       "\n",
       "       hospitalizado       muertos   covid_grave  \n",
       "count   74660.000000  74660.000000  74660.000000  \n",
       "mean        0.348942      0.108411      0.360139  \n",
       "std         0.476639      0.310901      0.480044  \n",
       "min         0.000000      0.000000      0.000000  \n",
       "25%         0.000000      0.000000      0.000000  \n",
       "50%         0.000000      0.000000      0.000000  \n",
       "75%         1.000000      0.000000      1.000000  \n",
       "max         1.000000      1.000000      1.000000  "
      ]
     },
     "execution_count": 15,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "train.describe()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "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>embarazo</th>\n",
       "      <th>edad</th>\n",
       "      <th>diabetes</th>\n",
       "      <th>epoc</th>\n",
       "      <th>asma</th>\n",
       "      <th>inmusupr</th>\n",
       "      <th>hipertension</th>\n",
       "      <th>cardiovascular</th>\n",
       "      <th>obesidad</th>\n",
       "      <th>tabaquismo</th>\n",
       "      <th>pobreza</th>\n",
       "      <th>Densidad_pob</th>\n",
       "      <th>medicos</th>\n",
       "      <th>camas_hosp</th>\n",
       "      <th>enfermeras</th>\n",
       "      <th>hospitalizado</th>\n",
       "      <th>muertos</th>\n",
       "      <th>covid_grave</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>count</th>\n",
       "      <td>18665.000000</td>\n",
       "      <td>18665.000000</td>\n",
       "      <td>18665.000000</td>\n",
       "      <td>18665.000000</td>\n",
       "      <td>18665.000000</td>\n",
       "      <td>18665.000000</td>\n",
       "      <td>18665.000000</td>\n",
       "      <td>18665.000000</td>\n",
       "      <td>18665.000000</td>\n",
       "      <td>18665.000000</td>\n",
       "      <td>18665.000000</td>\n",
       "      <td>18665.000000</td>\n",
       "      <td>18665.000000</td>\n",
       "      <td>18665.000000</td>\n",
       "      <td>18665.000000</td>\n",
       "      <td>18665.000000</td>\n",
       "      <td>18665.000000</td>\n",
       "      <td>18665.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>mean</th>\n",
       "      <td>0.006161</td>\n",
       "      <td>46.098312</td>\n",
       "      <td>0.176266</td>\n",
       "      <td>0.018805</td>\n",
       "      <td>0.029574</td>\n",
       "      <td>0.015751</td>\n",
       "      <td>0.208090</td>\n",
       "      <td>0.025181</td>\n",
       "      <td>0.206322</td>\n",
       "      <td>0.082400</td>\n",
       "      <td>0.368470</td>\n",
       "      <td>4710.340932</td>\n",
       "      <td>2.761847</td>\n",
       "      <td>11.070872</td>\n",
       "      <td>29.613717</td>\n",
       "      <td>0.348192</td>\n",
       "      <td>0.109349</td>\n",
       "      <td>0.358961</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>std</th>\n",
       "      <td>0.078254</td>\n",
       "      <td>15.767218</td>\n",
       "      <td>0.381056</td>\n",
       "      <td>0.135840</td>\n",
       "      <td>0.169414</td>\n",
       "      <td>0.124516</td>\n",
       "      <td>0.405952</td>\n",
       "      <td>0.156678</td>\n",
       "      <td>0.404675</td>\n",
       "      <td>0.274981</td>\n",
       "      <td>0.141023</td>\n",
       "      <td>5869.689261</td>\n",
       "      <td>1.991319</td>\n",
       "      <td>9.976615</td>\n",
       "      <td>24.270544</td>\n",
       "      <td>0.476410</td>\n",
       "      <td>0.312085</td>\n",
       "      <td>0.479708</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>min</th>\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>0.000000</td>\n",
       "      <td>0.032000</td>\n",
       "      <td>0.446811</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>0.000000</td>\n",
       "      <td>34.000000</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>0.287000</td>\n",
       "      <td>253.234919</td>\n",
       "      <td>1.506770</td>\n",
       "      <td>5.474526</td>\n",
       "      <td>16.131462</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>50%</th>\n",
       "      <td>0.000000</td>\n",
       "      <td>45.000000</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>0.346000</td>\n",
       "      <td>1813.733810</td>\n",
       "      <td>2.248080</td>\n",
       "      <td>8.238176</td>\n",
       "      <td>21.720906</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>75%</th>\n",
       "      <td>0.000000</td>\n",
       "      <td>57.000000</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>0.412000</td>\n",
       "      <td>8183.315370</td>\n",
       "      <td>3.772233</td>\n",
       "      <td>13.572164</td>\n",
       "      <td>35.940020</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>1.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>max</th>\n",
       "      <td>1.000000</td>\n",
       "      <td>94.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.976000</td>\n",
       "      <td>17954.252292</td>\n",
       "      <td>50.068722</td>\n",
       "      <td>297.467112</td>\n",
       "      <td>559.591596</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "           embarazo          edad      diabetes          epoc          asma  \\\n",
       "count  18665.000000  18665.000000  18665.000000  18665.000000  18665.000000   \n",
       "mean       0.006161     46.098312      0.176266      0.018805      0.029574   \n",
       "std        0.078254     15.767218      0.381056      0.135840      0.169414   \n",
       "min        0.000000      0.000000      0.000000      0.000000      0.000000   \n",
       "25%        0.000000     34.000000      0.000000      0.000000      0.000000   \n",
       "50%        0.000000     45.000000      0.000000      0.000000      0.000000   \n",
       "75%        0.000000     57.000000      0.000000      0.000000      0.000000   \n",
       "max        1.000000     94.000000      1.000000      1.000000      1.000000   \n",
       "\n",
       "           inmusupr  hipertension  cardiovascular      obesidad    tabaquismo  \\\n",
       "count  18665.000000  18665.000000    18665.000000  18665.000000  18665.000000   \n",
       "mean       0.015751      0.208090        0.025181      0.206322      0.082400   \n",
       "std        0.124516      0.405952        0.156678      0.404675      0.274981   \n",
       "min        0.000000      0.000000        0.000000      0.000000      0.000000   \n",
       "25%        0.000000      0.000000        0.000000      0.000000      0.000000   \n",
       "50%        0.000000      0.000000        0.000000      0.000000      0.000000   \n",
       "75%        0.000000      0.000000        0.000000      0.000000      0.000000   \n",
       "max        1.000000      1.000000        1.000000      1.000000      1.000000   \n",
       "\n",
       "            pobreza  Densidad_pob       medicos    camas_hosp    enfermeras  \\\n",
       "count  18665.000000  18665.000000  18665.000000  18665.000000  18665.000000   \n",
       "mean       0.368470   4710.340932      2.761847     11.070872     29.613717   \n",
       "std        0.141023   5869.689261      1.991319      9.976615     24.270544   \n",
       "min        0.032000      0.446811      0.000000      0.000000      0.000000   \n",
       "25%        0.287000    253.234919      1.506770      5.474526     16.131462   \n",
       "50%        0.346000   1813.733810      2.248080      8.238176     21.720906   \n",
       "75%        0.412000   8183.315370      3.772233     13.572164     35.940020   \n",
       "max        0.976000  17954.252292     50.068722    297.467112    559.591596   \n",
       "\n",
       "       hospitalizado       muertos   covid_grave  \n",
       "count   18665.000000  18665.000000  18665.000000  \n",
       "mean        0.348192      0.109349      0.358961  \n",
       "std         0.476410      0.312085      0.479708  \n",
       "min         0.000000      0.000000      0.000000  \n",
       "25%         0.000000      0.000000      0.000000  \n",
       "50%         0.000000      0.000000      0.000000  \n",
       "75%         1.000000      0.000000      1.000000  \n",
       "max         1.000000      1.000000      1.000000  "
      ]
     },
     "execution_count": 16,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "test.describe()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 65,
   "metadata": {},
   "outputs": [],
   "source": [
    "feats = ['embarazo', 'edad', 'diabetes', 'epoc', 'asma', 'inmusupr', 'hipertension', 'cardiovascular',\n",
    "            'obesidad', 'tabaquismo', 'pobreza', 'Densidad_pob', 'medicos', 'camas_hosp', 'enfermeras']\n",
    "\n",
    "feats_train = train.loc[:, feats]\n",
    "target1_train = train.loc[:, 'muertos']\n",
    "target2_train = train.loc[:, 'hospitalizado']\n",
    "target3_train = train.loc[:, 'covid_grave']\n",
    "\n",
    "feats_test = test.loc[:, feats]\n",
    "target1_test = test.loc[:, 'muertos']\n",
    "target2_test = test.loc[:, 'hospitalizado']\n",
    "target3_test = test.loc[:, 'covid_grave']"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### 1.1 Gaussian Naive Bayes (GNB) T1 (Deaths)\n",
    "\n",
    "Train for  Gaussian distribution:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "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>params</th>\n",
       "      <th>mean_test_accuracy</th>\n",
       "      <th>mean_test_precision</th>\n",
       "      <th>mean_test_recall</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>{}</td>\n",
       "      <td>0.855304</td>\n",
       "      <td>0.295961</td>\n",
       "      <td>0.242649</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "  params  mean_test_accuracy  mean_test_precision  mean_test_recall\n",
       "0     {}            0.855304             0.295961          0.242649"
      ]
     },
     "execution_count": 28,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "from sklearn.model_selection import GridSearchCV\n",
    "from sklearn.naive_bayes import GaussianNB\n",
    "\n",
    "params = {}\n",
    "\n",
    "nbg = GridSearchCV(estimator=GaussianNB(), param_grid=params, cv=10 ,\n",
    "                   return_train_score=True,\n",
    "                   scoring=['accuracy', 'precision', 'recall'], \n",
    "                   refit='accuracy')\n",
    "\n",
    "nbg.fit(feats_train, target1_train)\n",
    "cv_results_nbg = pd.DataFrame(nbg.cv_results_)\n",
    "cols = ['params', 'mean_test_accuracy', 'mean_test_precision', 'mean_test_recall']\n",
    "\n",
    "cv_results_nbg[cols]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Confusion matrix\n",
      "[[15518  1106]\n",
      " [ 1500   541]]\n",
      "Accuracy score\n",
      "0.8603803911063488\n",
      "Precision score\n",
      "0.3284760170006072\n",
      "Recall score\n",
      "0.26506614404703577\n",
      "F1 score\n",
      "0.2933839479392624\n"
     ]
    }
   ],
   "source": [
    "predicted_target1 = nbg.predict(feats_test)\n",
    "\n",
    "print('Confusion matrix')\n",
    "print(sklearn.metrics.confusion_matrix(target1_test, predicted_target1))\n",
    "\n",
    "print('Accuracy score')\n",
    "print(sklearn.metrics.accuracy_score(target1_test, predicted_target1))\n",
    "\n",
    "print('Precision score')\n",
    "print(sklearn.metrics.precision_score(target1_test, predicted_target1))\n",
    "      \n",
    "print('Recall score')\n",
    "print(sklearn.metrics.recall_score(target1_test, predicted_target1))\n",
    "      \n",
    "print('F1 score')\n",
    "print(sklearn.metrics.f1_score(target1_test, predicted_target1))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Upsampling"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 56,
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn.utils import resample\n",
    "negatives = train[train.muertos == 0]\n",
    "positives = train[train.muertos == 1]\n",
    "\n",
    "upsample_positives = resample(positives,\n",
    "                              replace=True, \n",
    "                              n_samples=len(negatives) // 2, \n",
    "                              random_state=0)\n",
    "upsampled_train = pd.concat([negatives, upsample_positives])\n",
    "\n",
    "feats_train_up = upsampled_train.loc[:, feats]\n",
    "target1_train_up = upsampled_train.loc[:, 'muertos']"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 57,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0    66566\n",
       "1    33283\n",
       "Name: muertos, dtype: int64"
      ]
     },
     "execution_count": 57,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "upsampled_train.muertos.value_counts()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 58,
   "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>params</th>\n",
       "      <th>mean_test_accuracy</th>\n",
       "      <th>mean_test_precision</th>\n",
       "      <th>mean_test_recall</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>{}</td>\n",
       "      <td>0.716342</td>\n",
       "      <td>0.598583</td>\n",
       "      <td>0.452724</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "  params  mean_test_accuracy  mean_test_precision  mean_test_recall\n",
       "0     {}            0.716342             0.598583          0.452724"
      ]
     },
     "execution_count": 58,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "nbg = GridSearchCV(estimator=GaussianNB(), param_grid=params, cv=10 ,\n",
    "                   return_train_score=True,\n",
    "                   scoring=['accuracy', 'precision', 'recall'], \n",
    "                   refit='accuracy')\n",
    "\n",
    "nbg.fit(feats_train_up, target1_train_up)\n",
    "cv_results_nbg = pd.DataFrame(nbg.cv_results_)\n",
    "cols = ['params', 'mean_test_accuracy', 'mean_test_precision', 'mean_test_recall']\n",
    "\n",
    "cv_results_nbg[cols]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 59,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Confusion matrix\n",
      "[[14171  2453]\n",
      " [ 1055   986]]\n",
      "Accuracy score\n",
      "0.812054647736405\n",
      "Precision score\n",
      "0.2867112532712998\n",
      "Recall score\n",
      "0.48309652131308184\n",
      "F1 score\n",
      "0.3598540145985402\n"
     ]
    }
   ],
   "source": [
    "predicted_target1 = nbg.predict(feats_test)\n",
    "\n",
    "print('Confusion matrix')\n",
    "print(sklearn.metrics.confusion_matrix(target1_test, predicted_target1))\n",
    "\n",
    "print('Accuracy score')\n",
    "print(sklearn.metrics.accuracy_score(target1_test, predicted_target1))\n",
    "\n",
    "print('Precision score')\n",
    "print(sklearn.metrics.precision_score(target1_test, predicted_target1))\n",
    "      \n",
    "print('Recall score')\n",
    "print(sklearn.metrics.recall_score(target1_test, predicted_target1))\n",
    "      \n",
    "print('F1 score')\n",
    "print(sklearn.metrics.f1_score(target1_test, predicted_target1))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Undersampling"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 41,
   "metadata": {},
   "outputs": [],
   "source": [
    "undersample_negatives = resample(negatives,\n",
    "                                 replace=True, # sample with replacement\n",
    "                                 n_samples=len(positives), # match number in majority class\n",
    "                                 random_state=0)\n",
    "undersample_train = pd.concat([positives, undersample_negatives])\n",
    "\n",
    "feats_train_und = undersample_train.loc[:, feats]\n",
    "target1_train_und = undersample_train.loc[:, 'muertos']"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 43,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "1    8094\n",
       "0    8094\n",
       "Name: muertos, dtype: int64"
      ]
     },
     "execution_count": 43,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "undersample_train.muertos.value_counts()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 60,
   "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>params</th>\n",
       "      <th>mean_test_accuracy</th>\n",
       "      <th>mean_test_precision</th>\n",
       "      <th>mean_test_recall</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>{}</td>\n",
       "      <td>0.674451</td>\n",
       "      <td>0.720415</td>\n",
       "      <td>0.570299</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "  params  mean_test_accuracy  mean_test_precision  mean_test_recall\n",
       "0     {}            0.674451             0.720415          0.570299"
      ]
     },
     "execution_count": 60,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "nbg = GridSearchCV(estimator=GaussianNB(), param_grid=params, cv=10 ,\n",
    "                   return_train_score=True,\n",
    "                   scoring=['accuracy', 'precision', 'recall'], \n",
    "                   refit='accuracy')\n",
    "\n",
    "nbg.fit(feats_train_und, target1_train_und)\n",
    "cv_results_nbg = pd.DataFrame(nbg.cv_results_)\n",
    "cols = ['params', 'mean_test_accuracy', 'mean_test_precision', 'mean_test_recall']\n",
    "\n",
    "cv_results_nbg[cols]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 61,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Confusion matrix\n",
      "[[13178  3446]\n",
      " [  830  1211]]\n",
      "Accuracy score\n",
      "0.7709081167961425\n",
      "Precision score\n",
      "0.2600386514923771\n",
      "Recall score\n",
      "0.5933365997060265\n",
      "F1 score\n",
      "0.3616004777545536\n"
     ]
    }
   ],
   "source": [
    "predicted_target1 = nbg.predict(feats_test)\n",
    "\n",
    "print('Confusion matrix')\n",
    "print(sklearn.metrics.confusion_matrix(target1_test, predicted_target1))\n",
    "\n",
    "print('Accuracy score')\n",
    "print(sklearn.metrics.accuracy_score(target1_test, predicted_target1))\n",
    "\n",
    "print('Precision score')\n",
    "print(sklearn.metrics.precision_score(target1_test, predicted_target1))\n",
    "      \n",
    "print('Recall score')\n",
    "print(sklearn.metrics.recall_score(target1_test, predicted_target1))\n",
    "      \n",
    "print('F1 score')\n",
    "print(sklearn.metrics.f1_score(target1_test, predicted_target1))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### SMOTE"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 69,
   "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>params</th>\n",
       "      <th>mean_test_accuracy</th>\n",
       "      <th>mean_test_precision</th>\n",
       "      <th>mean_test_recall</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>{}</td>\n",
       "      <td>0.723725</td>\n",
       "      <td>0.233318</td>\n",
       "      <td>0.677251</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "  params  mean_test_accuracy  mean_test_precision  mean_test_recall\n",
       "0     {}            0.723725             0.233318          0.677251"
      ]
     },
     "execution_count": 69,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "from imblearn.pipeline import Pipeline, make_pipeline\n",
    "from imblearn.over_sampling import SMOTE\n",
    "from sklearn.model_selection import RepeatedStratifiedKFold\n",
    "from sklearn.model_selection import KFold\n",
    "\n",
    "pipeline = make_pipeline((SMOTE(random_state=0)), GaussianNB())\n",
    "\n",
    "kf = RepeatedStratifiedKFold(n_splits=10, n_repeats=3, random_state=1)\n",
    "\n",
    "nbg = GridSearchCV(estimator=pipeline, param_grid=params, cv=kf ,\n",
    "                   return_train_score=True,\n",
    "                   scoring=['accuracy', 'precision', 'recall'], \n",
    "                   refit='accuracy')\n",
    "\n",
    "nbg.fit(feats_train, target1_train)\n",
    "cv_results_nbg = pd.DataFrame(nbg.cv_results_)\n",
    "cols = ['params', 'mean_test_accuracy', 'mean_test_precision', 'mean_test_recall']\n",
    "\n",
    "cv_results_nbg[cols]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 70,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Confusion matrix\n",
      "[[12273  4351]\n",
      " [  630  1411]]\n",
      "Accuracy score\n",
      "0.7331368872220734\n",
      "Precision score\n",
      "0.24488024991322457\n",
      "Recall score\n",
      "0.6913277804997551\n",
      "F1 score\n",
      "0.36165577342047933\n"
     ]
    }
   ],
   "source": [
    "predicted_target1 = nbg.predict(feats_test)\n",
    "\n",
    "print('Confusion matrix')\n",
    "print(sklearn.metrics.confusion_matrix(target1_test, predicted_target1))\n",
    "\n",
    "print('Accuracy score')\n",
    "print(sklearn.metrics.accuracy_score(target1_test, predicted_target1))\n",
    "\n",
    "print('Precision score')\n",
    "print(sklearn.metrics.precision_score(target1_test, predicted_target1))\n",
    "      \n",
    "print('Recall score')\n",
    "print(sklearn.metrics.recall_score(target1_test, predicted_target1))\n",
    "      \n",
    "print('F1 score')\n",
    "print(sklearn.metrics.f1_score(target1_test, predicted_target1))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### 1.2 GNB T2 (Hospitalization)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 62,
   "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>params</th>\n",
       "      <th>mean_test_accuracy</th>\n",
       "      <th>mean_test_precision</th>\n",
       "      <th>mean_test_recall</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>{}</td>\n",
       "      <td>0.705585</td>\n",
       "      <td>0.606167</td>\n",
       "      <td>0.446607</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "  params  mean_test_accuracy  mean_test_precision  mean_test_recall\n",
       "0     {}            0.705585             0.606167          0.446607"
      ]
     },
     "execution_count": 62,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "nbg = GridSearchCV(estimator=GaussianNB(), param_grid=params, cv=10 ,\n",
    "                   return_train_score=True,\n",
    "                   scoring=['accuracy', 'precision', 'recall'], \n",
    "                   refit='accuracy')\n",
    "\n",
    "nbg.fit(feats_train, target2_train)\n",
    "cv_results_nbg = pd.DataFrame(nbg.cv_results_)\n",
    "cols = ['params', 'mean_test_accuracy', 'mean_test_precision', 'mean_test_recall']\n",
    "\n",
    "cv_results_nbg[cols]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 63,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Confusion matrix\n",
      "[[10291  1875]\n",
      " [ 3579  2920]]\n",
      "Accuracy score\n",
      "0.7077953388695419\n",
      "Precision score\n",
      "0.6089676746611054\n",
      "Recall score\n",
      "0.4492998922911217\n",
      "F1 score\n",
      "0.5170887196741634\n"
     ]
    }
   ],
   "source": [
    "predicted_target2 = nbg.predict(feats_test)\n",
    "\n",
    "print('Confusion matrix')\n",
    "print(sklearn.metrics.confusion_matrix(target2_test, predicted_target2))\n",
    "\n",
    "print('Accuracy score')\n",
    "print(sklearn.metrics.accuracy_score(target2_test, predicted_target2))\n",
    "\n",
    "print('Precision score')\n",
    "print(sklearn.metrics.precision_score(target2_test, predicted_target2))\n",
    "      \n",
    "print('Recall score')\n",
    "print(sklearn.metrics.recall_score(target2_test, predicted_target2))\n",
    "      \n",
    "print('F1 score')\n",
    "print(sklearn.metrics.f1_score(target2_test, predicted_target2))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### SMOTE"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 71,
   "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>params</th>\n",
       "      <th>mean_test_accuracy</th>\n",
       "      <th>mean_test_precision</th>\n",
       "      <th>mean_test_recall</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>{}</td>\n",
       "      <td>0.704049</td>\n",
       "      <td>0.577317</td>\n",
       "      <td>0.56734</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "  params  mean_test_accuracy  mean_test_precision  mean_test_recall\n",
       "0     {}            0.704049             0.577317           0.56734"
      ]
     },
     "execution_count": 71,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "pipeline = make_pipeline((SMOTE(random_state=0)), GaussianNB())\n",
    "\n",
    "kf = RepeatedStratifiedKFold(n_splits=10, n_repeats=3, random_state=1)\n",
    "\n",
    "nbg = GridSearchCV(estimator=pipeline, param_grid=params, cv=kf ,\n",
    "                   return_train_score=True,\n",
    "                   scoring=['accuracy', 'precision', 'recall'], \n",
    "                   refit='accuracy')\n",
    "\n",
    "nbg.fit(feats_train, target2_train)\n",
    "cv_results_nbg = pd.DataFrame(nbg.cv_results_)\n",
    "cols = ['params', 'mean_test_accuracy', 'mean_test_precision', 'mean_test_recall']\n",
    "\n",
    "cv_results_nbg[cols]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 72,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Confusion matrix\n",
      "[[9521 2645]\n",
      " [2810 3689]]\n",
      "Accuracy score\n",
      "0.7077417626573801\n",
      "Precision score\n",
      "0.5824123776444585\n",
      "Recall score\n",
      "0.567625788582859\n",
      "F1 score\n",
      "0.5749240240006234\n"
     ]
    }
   ],
   "source": [
    "predicted_target2 = nbg.predict(feats_test)\n",
    "\n",
    "print('Confusion matrix')\n",
    "print(sklearn.metrics.confusion_matrix(target2_test, predicted_target2))\n",
    "\n",
    "print('Accuracy score')\n",
    "print(sklearn.metrics.accuracy_score(target2_test, predicted_target2))\n",
    "\n",
    "print('Precision score')\n",
    "print(sklearn.metrics.precision_score(target2_test, predicted_target2))\n",
    "      \n",
    "print('Recall score')\n",
    "print(sklearn.metrics.recall_score(target2_test, predicted_target2))\n",
    "      \n",
    "print('F1 score')\n",
    "print(sklearn.metrics.f1_score(target2_test, predicted_target2))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### 1.3 GNB T3 (Hospitalization/Death)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 66,
   "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>params</th>\n",
       "      <th>mean_test_accuracy</th>\n",
       "      <th>mean_test_precision</th>\n",
       "      <th>mean_test_recall</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>{}</td>\n",
       "      <td>0.707929</td>\n",
       "      <td>0.629515</td>\n",
       "      <td>0.459723</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "  params  mean_test_accuracy  mean_test_precision  mean_test_recall\n",
       "0     {}            0.707929             0.629515          0.459723"
      ]
     },
     "execution_count": 66,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "nbg = GridSearchCV(estimator=GaussianNB(), param_grid=params, cv=10 ,\n",
    "                   return_train_score=True,\n",
    "                   scoring=['accuracy', 'precision', 'recall'], \n",
    "                   refit='accuracy')\n",
    "\n",
    "nbg.fit(feats_train, target3_train)\n",
    "cv_results_nbg = pd.DataFrame(nbg.cv_results_)\n",
    "cols = ['params', 'mean_test_accuracy', 'mean_test_precision', 'mean_test_recall']\n",
    "\n",
    "cv_results_nbg[cols]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 67,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Confusion matrix\n",
      "[[10159  1806]\n",
      " [ 3588  3112]]\n",
      "Accuracy score\n",
      "0.71100991159925\n",
      "Precision score\n",
      "0.6327775518503457\n",
      "Recall score\n",
      "0.4644776119402985\n",
      "F1 score\n",
      "0.5357204338096058\n"
     ]
    }
   ],
   "source": [
    "predicted_target3 = nbg.predict(feats_test)\n",
    "\n",
    "print('Confusion matrix')\n",
    "print(sklearn.metrics.confusion_matrix(target3_test, predicted_target3))\n",
    "\n",
    "print('Accuracy score')\n",
    "print(sklearn.metrics.accuracy_score(target3_test, predicted_target3))\n",
    "\n",
    "print('Precision score')\n",
    "print(sklearn.metrics.precision_score(target3_test, predicted_target3))\n",
    "      \n",
    "print('Recall score')\n",
    "print(sklearn.metrics.recall_score(target3_test, predicted_target3))\n",
    "      \n",
    "print('F1 score')\n",
    "print(sklearn.metrics.f1_score(target3_test, predicted_target3))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### SMOTE"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 73,
   "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>params</th>\n",
       "      <th>mean_test_accuracy</th>\n",
       "      <th>mean_test_precision</th>\n",
       "      <th>mean_test_recall</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>{}</td>\n",
       "      <td>0.708416</td>\n",
       "      <td>0.602032</td>\n",
       "      <td>0.5618</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "  params  mean_test_accuracy  mean_test_precision  mean_test_recall\n",
       "0     {}            0.708416             0.602032            0.5618"
      ]
     },
     "execution_count": 73,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "pipeline = make_pipeline((SMOTE(random_state=0)), GaussianNB())\n",
    "\n",
    "kf = RepeatedStratifiedKFold(n_splits=10, n_repeats=3, random_state=1)\n",
    "\n",
    "nbg = GridSearchCV(estimator=pipeline, param_grid=params, cv=kf ,\n",
    "                   return_train_score=True,\n",
    "                   scoring=['accuracy', 'precision', 'recall'], \n",
    "                   refit='accuracy')\n",
    "\n",
    "nbg.fit(feats_train, target3_train)\n",
    "cv_results_nbg = pd.DataFrame(nbg.cv_results_)\n",
    "cols = ['params', 'mean_test_accuracy', 'mean_test_precision', 'mean_test_recall']\n",
    "\n",
    "cv_results_nbg[cols]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 74,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Confusion matrix\n",
      "[[9506 2459]\n",
      " [2941 3759]]\n",
      "Accuracy score\n",
      "0.7106884543262791\n",
      "Precision score\n",
      "0.6045352203280797\n",
      "Recall score\n",
      "0.561044776119403\n",
      "F1 score\n",
      "0.5819786344635391\n"
     ]
    }
   ],
   "source": [
    "predicted_target3 = nbg.predict(feats_test)\n",
    "\n",
    "print('Confusion matrix')\n",
    "print(sklearn.metrics.confusion_matrix(target3_test, predicted_target3))\n",
    "\n",
    "print('Accuracy score')\n",
    "print(sklearn.metrics.accuracy_score(target3_test, predicted_target3))\n",
    "\n",
    "print('Precision score')\n",
    "print(sklearn.metrics.precision_score(target3_test, predicted_target3))\n",
    "      \n",
    "print('Recall score')\n",
    "print(sklearn.metrics.recall_score(target3_test, predicted_target3))\n",
    "      \n",
    "print('F1 score')\n",
    "print(sklearn.metrics.f1_score(target3_test, predicted_target3))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### 2.1 Complement Naive Bayes (CNB) T1 (Deaths) <---Model with highest Recall!\n",
    "\n",
    "CNB is an adaptation of the standard multinomial naive Bayes (MNB) algorithm that is particularly suited for imbalanced data sets, where the algorithm uses statistics from the complement of each class to compute the model’s weight. \n",
    "\n",
    "https://scikit-learn.org/stable/modules/naive_bayes.html"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Categorize continuous features:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 76,
   "metadata": {},
   "outputs": [],
   "source": [
    "import mlpipeline as pipeline\n",
    "\n",
    "df_c = df.loc[:,:]\n",
    "def label_age (row):\n",
    "    if row['edad'] >= 0  and row['edad'] <= 17 :\n",
    "        return '0-17'\n",
    "    if row['edad'] >= 18 and row['edad'] <= 44 :\n",
    "        return '18-44'\n",
    "    if row['edad'] >= 45 and row['edad'] <= 64 :\n",
    "        return '45-64'\n",
    "    if row['edad'] >= 65 and row['edad'] <= 74 :\n",
    "        return '65-74'\n",
    "    return '75 +'\n",
    "\n",
    "df_c['age_group'] = df_c.apply (lambda row: label_age(row), axis=1)\n",
    "df_c['pobreza'] = np.where(df_c['pobreza']>=df_c['pobreza'].mean(), 1, 0)\n",
    "df_c['Densidad_pob'] = np.where(df_c['Densidad_pob']>=df_c['Densidad_pob'].mean(), 1, 0)\n",
    "df_c['medicos'] = np.where(df_c['medicos']>=df_c['medicos'].mean(), 1, 0)\n",
    "df_c['camas_hosp'] = np.where(df_c['camas_hosp']>=df_c['camas_hosp'].mean(), 1, 0)\n",
    "df_c['camas_hosp'] = np.where(df_c['enfermeras']>=df_c['enfermeras'].mean(), 1, 0)\n",
    "\n",
    "train_c, test_c = pipeline.split_data(df_c.loc[:,'embarazo':], .20, seed=0)\n",
    "\n",
    "feats_train_c = train_c.loc[:, feats]\n",
    "target1_train_c = train_c.loc[:, 'muertos']\n",
    "target2_train_c = train_c.loc[:, 'hospitalizado']\n",
    "target3_train_c = train_c.loc[:, 'covid_grave']\n",
    "\n",
    "feats_test_c = test_c.loc[:, feats]\n",
    "target1_test_c = test_c.loc[:, 'muertos']\n",
    "target2_test_c = test_c.loc[:, 'hospitalizado']\n",
    "target3_test_c = test_c.loc[:, 'covid_grave']"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 77,
   "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>params</th>\n",
       "      <th>mean_test_accuracy</th>\n",
       "      <th>mean_test_precision</th>\n",
       "      <th>mean_test_recall</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>{}</td>\n",
       "      <td>0.536539</td>\n",
       "      <td>0.15327</td>\n",
       "      <td>0.723745</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "  params  mean_test_accuracy  mean_test_precision  mean_test_recall\n",
       "0     {}            0.536539              0.15327          0.723745"
      ]
     },
     "execution_count": 77,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "from sklearn.naive_bayes import ComplementNB\n",
    "\n",
    "params = {}\n",
    "\n",
    "nbg = GridSearchCV(estimator=ComplementNB(), param_grid=params, cv=10 ,\n",
    "                   return_train_score=True,\n",
    "                   scoring=['accuracy', 'precision', 'recall'], \n",
    "                   refit='accuracy')\n",
    "\n",
    "nbg.fit(feats_train_c, target1_train_c)\n",
    "cv_results_nbg = pd.DataFrame(nbg.cv_results_)\n",
    "cols = ['params', 'mean_test_accuracy', 'mean_test_precision', 'mean_test_recall']\n",
    "\n",
    "cv_results_nbg[cols]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 79,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Confusion matrix\n",
      "[[8544 8080]\n",
      " [ 528 1513]]\n",
      "Accuracy score\n",
      "0.5388159657112243\n",
      "Precision score\n",
      "0.15771917022829146\n",
      "Recall score\n",
      "0.7413032827045566\n",
      "F1 score\n",
      "0.26009970775313734\n"
     ]
    }
   ],
   "source": [
    "predicted_target1_c = nbg.predict(feats_test_c)\n",
    "\n",
    "print('Confusion matrix')\n",
    "print(sklearn.metrics.confusion_matrix(target1_test_c, predicted_target1_c))\n",
    "\n",
    "print('Accuracy score')\n",
    "print(sklearn.metrics.accuracy_score(target1_test_c, predicted_target1_c))\n",
    "\n",
    "print('Precision score')\n",
    "print(sklearn.metrics.precision_score(target1_test_c, predicted_target1_c))\n",
    "      \n",
    "print('Recall score')\n",
    "print(sklearn.metrics.recall_score(target1_test_c, predicted_target1_c))\n",
    "      \n",
    "print('F1 score')\n",
    "print(sklearn.metrics.f1_score(target1_test_c, predicted_target1_c))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### SMOTE"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 80,
   "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>params</th>\n",
       "      <th>mean_test_accuracy</th>\n",
       "      <th>mean_test_precision</th>\n",
       "      <th>mean_test_recall</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>{}</td>\n",
       "      <td>0.53791</td>\n",
       "      <td>0.153538</td>\n",
       "      <td>0.7228</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "  params  mean_test_accuracy  mean_test_precision  mean_test_recall\n",
       "0     {}             0.53791             0.153538            0.7228"
      ]
     },
     "execution_count": 80,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "pipeline = make_pipeline((SMOTE(random_state=0)), ComplementNB())\n",
    "\n",
    "kf = RepeatedStratifiedKFold(n_splits=10, n_repeats=3, random_state=1)\n",
    "\n",
    "nbg = GridSearchCV(estimator=pipeline, param_grid=params, cv=kf ,\n",
    "                   return_train_score=True,\n",
    "                   scoring=['accuracy', 'precision', 'recall'], \n",
    "                   refit='accuracy')\n",
    "\n",
    "nbg.fit(feats_train_c, target1_train_c)\n",
    "cv_results_nbg = pd.DataFrame(nbg.cv_results_)\n",
    "cols = ['params', 'mean_test_accuracy', 'mean_test_precision', 'mean_test_recall']\n",
    "\n",
    "cv_results_nbg[cols]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 81,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Confusion matrix\n",
      "[[8593 8031]\n",
      " [ 533 1508]]\n",
      "Accuracy score\n",
      "0.5411733190463435\n",
      "Precision score\n",
      "0.1580878498794423\n",
      "Recall score\n",
      "0.7388535031847133\n",
      "F1 score\n",
      "0.26044905008635577\n"
     ]
    }
   ],
   "source": [
    "predicted_target1_c = nbg.predict(feats_test_c)\n",
    "\n",
    "print('Confusion matrix')\n",
    "print(sklearn.metrics.confusion_matrix(target1_test_c, predicted_target1_c))\n",
    "\n",
    "print('Accuracy score')\n",
    "print(sklearn.metrics.accuracy_score(target1_test_c, predicted_target1_c))\n",
    "\n",
    "print('Precision score')\n",
    "print(sklearn.metrics.precision_score(target1_test_c, predicted_target1_c))\n",
    "      \n",
    "print('Recall score')\n",
    "print(sklearn.metrics.recall_score(target1_test_c, predicted_target1_c))\n",
    "      \n",
    "print('F1 score')\n",
    "print(sklearn.metrics.f1_score(target1_test_c, predicted_target1_c))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### 2.2 CNB T2 (Hospitalization)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 82,
   "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>params</th>\n",
       "      <th>mean_test_accuracy</th>\n",
       "      <th>mean_test_precision</th>\n",
       "      <th>mean_test_recall</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>{}</td>\n",
       "      <td>0.587008</td>\n",
       "      <td>0.443112</td>\n",
       "      <td>0.714609</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "  params  mean_test_accuracy  mean_test_precision  mean_test_recall\n",
       "0     {}            0.587008             0.443112          0.714609"
      ]
     },
     "execution_count": 82,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "nbg = GridSearchCV(estimator=ComplementNB(), param_grid=params, cv=10 ,\n",
    "                   return_train_score=True,\n",
    "                   scoring=['accuracy', 'precision', 'recall'], \n",
    "                   refit='accuracy')\n",
    "\n",
    "nbg.fit(feats_train_c, target2_train_c)\n",
    "cv_results_nbg = pd.DataFrame(nbg.cv_results_)\n",
    "cols = ['params', 'mean_test_accuracy', 'mean_test_precision', 'mean_test_recall']\n",
    "\n",
    "cv_results_nbg[cols]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 83,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Confusion matrix\n",
      "[[6312 5854]\n",
      " [1847 4652]]\n",
      "Accuracy score\n",
      "0.5874095901419769\n",
      "Precision score\n",
      "0.4427945935655816\n",
      "Recall score\n",
      "0.7158024311432528\n",
      "F1 score\n",
      "0.5471331961187886\n"
     ]
    }
   ],
   "source": [
    "predicted_target2_c = nbg.predict(feats_test_c)\n",
    "\n",
    "print('Confusion matrix')\n",
    "print(sklearn.metrics.confusion_matrix(target2_test_c, predicted_target2_c))\n",
    "\n",
    "print('Accuracy score')\n",
    "print(sklearn.metrics.accuracy_score(target2_test_c, predicted_target2_c))\n",
    "\n",
    "print('Precision score')\n",
    "print(sklearn.metrics.precision_score(target2_test_c, predicted_target2_c))\n",
    "      \n",
    "print('Recall score')\n",
    "print(sklearn.metrics.recall_score(target2_test_c, predicted_target2_c))\n",
    "      \n",
    "print('F1 score')\n",
    "print(sklearn.metrics.f1_score(target2_test_c, predicted_target2_c))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### SMOTE"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 84,
   "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>params</th>\n",
       "      <th>mean_test_accuracy</th>\n",
       "      <th>mean_test_precision</th>\n",
       "      <th>mean_test_recall</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>{}</td>\n",
       "      <td>0.587021</td>\n",
       "      <td>0.443114</td>\n",
       "      <td>0.714519</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "  params  mean_test_accuracy  mean_test_precision  mean_test_recall\n",
       "0     {}            0.587021             0.443114          0.714519"
      ]
     },
     "execution_count": 84,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "pipeline = make_pipeline((SMOTE(random_state=0)), ComplementNB())\n",
    "\n",
    "kf = RepeatedStratifiedKFold(n_splits=10, n_repeats=3, random_state=1)\n",
    "\n",
    "nbg = GridSearchCV(estimator=pipeline, param_grid=params, cv=kf ,\n",
    "                   return_train_score=True,\n",
    "                   scoring=['accuracy', 'precision', 'recall'], \n",
    "                   refit='accuracy')\n",
    "\n",
    "nbg.fit(feats_train_c, target2_train_c)\n",
    "cv_results_nbg = pd.DataFrame(nbg.cv_results_)\n",
    "cols = ['params', 'mean_test_accuracy', 'mean_test_precision', 'mean_test_recall']\n",
    "\n",
    "cv_results_nbg[cols]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 85,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Confusion matrix\n",
      "[[6326 5840]\n",
      " [1854 4645]]\n",
      "Accuracy score\n",
      "0.5877846236271096\n",
      "Precision score\n",
      "0.4430138292799237\n",
      "Recall score\n",
      "0.7147253423603631\n",
      "F1 score\n",
      "0.5469853980216675\n"
     ]
    }
   ],
   "source": [
    "predicted_target2_c = nbg.predict(feats_test_c)\n",
    "\n",
    "print('Confusion matrix')\n",
    "print(sklearn.metrics.confusion_matrix(target2_test_c, predicted_target2_c))\n",
    "\n",
    "print('Accuracy score')\n",
    "print(sklearn.metrics.accuracy_score(target2_test_c, predicted_target2_c))\n",
    "\n",
    "print('Precision score')\n",
    "print(sklearn.metrics.precision_score(target2_test_c, predicted_target2_c))\n",
    "      \n",
    "print('Recall score')\n",
    "print(sklearn.metrics.recall_score(target2_test_c, predicted_target2_c))\n",
    "      \n",
    "print('F1 score')\n",
    "print(sklearn.metrics.f1_score(target2_test_c, predicted_target2_c))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### 2.3 CNB T3 (Hospitalization/Death)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 87,
   "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>params</th>\n",
       "      <th>mean_test_accuracy</th>\n",
       "      <th>mean_test_precision</th>\n",
       "      <th>mean_test_recall</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>{}</td>\n",
       "      <td>0.591776</td>\n",
       "      <td>0.457447</td>\n",
       "      <td>0.717346</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "  params  mean_test_accuracy  mean_test_precision  mean_test_recall\n",
       "0     {}            0.591776             0.457447          0.717346"
      ]
     },
     "execution_count": 87,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "nbg = GridSearchCV(estimator=ComplementNB(), param_grid=params, cv=10 ,\n",
    "                   return_train_score=True,\n",
    "                   scoring=['accuracy', 'precision', 'recall'], \n",
    "                   refit='accuracy')\n",
    "\n",
    "nbg.fit(feats_train_c, target3_train_c)\n",
    "cv_results_nbg = pd.DataFrame(nbg.cv_results_)\n",
    "cols = ['params', 'mean_test_accuracy', 'mean_test_precision', 'mean_test_recall']\n",
    "\n",
    "cv_results_nbg[cols]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 88,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Confusion matrix\n",
      "[[6230 5735]\n",
      " [1897 4803]]\n",
      "Accuracy score\n",
      "0.5911063487811412\n",
      "Precision score\n",
      "0.4557790852154109\n",
      "Recall score\n",
      "0.716865671641791\n",
      "F1 score\n",
      "0.5572572224155935\n"
     ]
    }
   ],
   "source": [
    "predicted_target3_c = nbg.predict(feats_test_c)\n",
    "\n",
    "print('Confusion matrix')\n",
    "print(sklearn.metrics.confusion_matrix(target3_test_c, predicted_target3_c))\n",
    "\n",
    "print('Accuracy score')\n",
    "print(sklearn.metrics.accuracy_score(target3_test_c, predicted_target3_c))\n",
    "\n",
    "print('Precision score')\n",
    "print(sklearn.metrics.precision_score(target3_test_c, predicted_target3_c))\n",
    "      \n",
    "print('Recall score')\n",
    "print(sklearn.metrics.recall_score(target3_test_c, predicted_target3_c))\n",
    "      \n",
    "print('F1 score')\n",
    "print(sklearn.metrics.f1_score(target3_test_c, predicted_target3_c))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### SMOTE"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 89,
   "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>params</th>\n",
       "      <th>mean_test_accuracy</th>\n",
       "      <th>mean_test_precision</th>\n",
       "      <th>mean_test_recall</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>{}</td>\n",
       "      <td>0.591388</td>\n",
       "      <td>0.457128</td>\n",
       "      <td>0.717284</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "  params  mean_test_accuracy  mean_test_precision  mean_test_recall\n",
       "0     {}            0.591388             0.457128          0.717284"
      ]
     },
     "execution_count": 89,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "pipeline = make_pipeline((SMOTE(random_state=0)), ComplementNB())\n",
    "\n",
    "kf = RepeatedStratifiedKFold(n_splits=10, n_repeats=3, random_state=1)\n",
    "\n",
    "nbg = GridSearchCV(estimator=pipeline, param_grid=params, cv=kf ,\n",
    "                   return_train_score=True,\n",
    "                   scoring=['accuracy', 'precision', 'recall'], \n",
    "                   refit='accuracy')\n",
    "\n",
    "nbg.fit(feats_train_c, target3_train_c)\n",
    "cv_results_nbg = pd.DataFrame(nbg.cv_results_)\n",
    "cols = ['params', 'mean_test_accuracy', 'mean_test_precision', 'mean_test_recall']\n",
    "\n",
    "cv_results_nbg[cols]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 90,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Confusion matrix\n",
      "[[6231 5734]\n",
      " [1891 4809]]\n",
      "Accuracy score\n",
      "0.5914813822662738\n",
      "Precision score\n",
      "0.4561320307312909\n",
      "Recall score\n",
      "0.7177611940298507\n",
      "F1 score\n",
      "0.5577915675926464\n"
     ]
    }
   ],
   "source": [
    "predicted_target3_c = nbg.predict(feats_test_c)\n",
    "\n",
    "print('Confusion matrix')\n",
    "print(sklearn.metrics.confusion_matrix(target3_test_c, predicted_target3_c))\n",
    "\n",
    "print('Accuracy score')\n",
    "print(sklearn.metrics.accuracy_score(target3_test_c, predicted_target3_c))\n",
    "\n",
    "print('Precision score')\n",
    "print(sklearn.metrics.precision_score(target3_test_c, predicted_target3_c))\n",
    "      \n",
    "print('Recall score')\n",
    "print(sklearn.metrics.recall_score(target3_test_c, predicted_target3_c))\n",
    "      \n",
    "print('F1 score')\n",
    "print(sklearn.metrics.f1_score(target3_test_c, predicted_target3_c))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.7.6"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 4
}