lec/29 Regression 1/lec1.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "id": "1a2549e7",
   "metadata": {},
   "outputs": [],
   "source": [
    "import geopandas as gpd"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "id": "a1c5c16f",
   "metadata": {},
   "outputs": [],
   "source": [
    "# DO ONCE:\n",
    "#df = gpd.read_file(\"https://opendata.arcgis.com/datasets/559d482bad0643f69cbe1538243e0baa_9.geojson\")\n",
    "#df.to_file(\"lec1download.geojson\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "id": "b40ed21e",
   "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>RptDt</th>\n",
       "      <th>GEOID</th>\n",
       "      <th>GEOName</th>\n",
       "      <th>POS_CUM_CP</th>\n",
       "      <th>POS_NEW_CP</th>\n",
       "      <th>POS_7DAYAVG_CP</th>\n",
       "      <th>DTH_CUM_CP</th>\n",
       "      <th>DTH_NEW_CP</th>\n",
       "      <th>DTH_7DAYAVG_CP</th>\n",
       "      <th>DTH_OVER_30DAYS_CP</th>\n",
       "      <th>...</th>\n",
       "      <th>HOSP_NO_CP</th>\n",
       "      <th>HOSP_UNK_CP</th>\n",
       "      <th>Date</th>\n",
       "      <th>POP</th>\n",
       "      <th>POP_MOE</th>\n",
       "      <th>OBJECTID</th>\n",
       "      <th>GEO</th>\n",
       "      <th>ShapeSTArea</th>\n",
       "      <th>ShapeSTLength</th>\n",
       "      <th>geometry</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>2022-04-07T19:00:00+00:00</td>\n",
       "      <td>55079000101</td>\n",
       "      <td>None</td>\n",
       "      <td>1158</td>\n",
       "      <td>0</td>\n",
       "      <td>0.14</td>\n",
       "      <td>24</td>\n",
       "      <td>0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>511</td>\n",
       "      <td>568</td>\n",
       "      <td>2022-04-06T06:00:00+00:00</td>\n",
       "      <td>4476</td>\n",
       "      <td>437.0</td>\n",
       "      <td>260</td>\n",
       "      <td>Census tract</td>\n",
       "      <td>7.500609e+06</td>\n",
       "      <td>11159.571995</td>\n",
       "      <td>POLYGON ((-87.99418 43.19256, -87.99429 43.192...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>2022-04-07T19:00:00+00:00</td>\n",
       "      <td>55079000102</td>\n",
       "      <td>None</td>\n",
       "      <td>1159</td>\n",
       "      <td>1</td>\n",
       "      <td>0.43</td>\n",
       "      <td>19</td>\n",
       "      <td>0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>561</td>\n",
       "      <td>508</td>\n",
       "      <td>2022-04-06T06:00:00+00:00</td>\n",
       "      <td>3468</td>\n",
       "      <td>404.0</td>\n",
       "      <td>261</td>\n",
       "      <td>Census tract</td>\n",
       "      <td>1.224474e+07</td>\n",
       "      <td>15500.211793</td>\n",
       "      <td>POLYGON ((-87.99449 43.17769, -87.99512 43.177...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>2022-04-07T19:00:00+00:00</td>\n",
       "      <td>55079000201</td>\n",
       "      <td>None</td>\n",
       "      <td>1410</td>\n",
       "      <td>0</td>\n",
       "      <td>0.29</td>\n",
       "      <td>10</td>\n",
       "      <td>0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>722</td>\n",
       "      <td>609</td>\n",
       "      <td>2022-04-06T06:00:00+00:00</td>\n",
       "      <td>5561</td>\n",
       "      <td>699.0</td>\n",
       "      <td>262</td>\n",
       "      <td>Census tract</td>\n",
       "      <td>1.488298e+07</td>\n",
       "      <td>17849.912522</td>\n",
       "      <td>POLYGON ((-88.02435 43.17761, -88.02434 43.177...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>2022-04-07T19:00:00+00:00</td>\n",
       "      <td>55079000202</td>\n",
       "      <td>None</td>\n",
       "      <td>1728</td>\n",
       "      <td>0</td>\n",
       "      <td>0.71</td>\n",
       "      <td>11</td>\n",
       "      <td>0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>913</td>\n",
       "      <td>738</td>\n",
       "      <td>2022-04-06T06:00:00+00:00</td>\n",
       "      <td>6439</td>\n",
       "      <td>621.0</td>\n",
       "      <td>263</td>\n",
       "      <td>Census tract</td>\n",
       "      <td>1.480524e+07</td>\n",
       "      <td>17742.064041</td>\n",
       "      <td>POLYGON ((-88.02435 43.17761, -88.02518 43.177...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>2022-04-07T19:00:00+00:00</td>\n",
       "      <td>55079000301</td>\n",
       "      <td>None</td>\n",
       "      <td>411</td>\n",
       "      <td>1</td>\n",
       "      <td>0.14</td>\n",
       "      <td>4</td>\n",
       "      <td>0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>218</td>\n",
       "      <td>172</td>\n",
       "      <td>2022-04-06T06:00:00+00:00</td>\n",
       "      <td>1485</td>\n",
       "      <td>92.0</td>\n",
       "      <td>264</td>\n",
       "      <td>Census tract</td>\n",
       "      <td>3.733793e+06</td>\n",
       "      <td>9142.114479</td>\n",
       "      <td>POLYGON ((-88.05488 43.14699, -88.05511 43.146...</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>5 rows × 90 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "                       RptDt        GEOID GEOName  POS_CUM_CP  POS_NEW_CP  \\\n",
       "0  2022-04-07T19:00:00+00:00  55079000101    None        1158           0   \n",
       "1  2022-04-07T19:00:00+00:00  55079000102    None        1159           1   \n",
       "2  2022-04-07T19:00:00+00:00  55079000201    None        1410           0   \n",
       "3  2022-04-07T19:00:00+00:00  55079000202    None        1728           0   \n",
       "4  2022-04-07T19:00:00+00:00  55079000301    None         411           1   \n",
       "\n",
       "   POS_7DAYAVG_CP  DTH_CUM_CP  DTH_NEW_CP  DTH_7DAYAVG_CP  DTH_OVER_30DAYS_CP  \\\n",
       "0            0.14          24           0             0.0                   0   \n",
       "1            0.43          19           0             0.0                   0   \n",
       "2            0.29          10           0             0.0                   0   \n",
       "3            0.71          11           0             0.0                   0   \n",
       "4            0.14           4           0             0.0                   0   \n",
       "\n",
       "   ...  HOSP_NO_CP HOSP_UNK_CP                       Date   POP POP_MOE  \\\n",
       "0  ...         511         568  2022-04-06T06:00:00+00:00  4476   437.0   \n",
       "1  ...         561         508  2022-04-06T06:00:00+00:00  3468   404.0   \n",
       "2  ...         722         609  2022-04-06T06:00:00+00:00  5561   699.0   \n",
       "3  ...         913         738  2022-04-06T06:00:00+00:00  6439   621.0   \n",
       "4  ...         218         172  2022-04-06T06:00:00+00:00  1485    92.0   \n",
       "\n",
       "   OBJECTID           GEO   ShapeSTArea  ShapeSTLength  \\\n",
       "0       260  Census tract  7.500609e+06   11159.571995   \n",
       "1       261  Census tract  1.224474e+07   15500.211793   \n",
       "2       262  Census tract  1.488298e+07   17849.912522   \n",
       "3       263  Census tract  1.480524e+07   17742.064041   \n",
       "4       264  Census tract  3.733793e+06    9142.114479   \n",
       "\n",
       "                                            geometry  \n",
       "0  POLYGON ((-87.99418 43.19256, -87.99429 43.192...  \n",
       "1  POLYGON ((-87.99449 43.17769, -87.99512 43.177...  \n",
       "2  POLYGON ((-88.02435 43.17761, -88.02434 43.177...  \n",
       "3  POLYGON ((-88.02435 43.17761, -88.02518 43.177...  \n",
       "4  POLYGON ((-88.05488 43.14699, -88.05511 43.146...  \n",
       "\n",
       "[5 rows x 90 columns]"
      ]
     },
     "execution_count": 3,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df = gpd.read_file(\"lec1download.geojson\")\n",
    "df.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "id": "7ade3f4f",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<AxesSubplot:>"
      ]
     },
     "execution_count": 4,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "df.plot()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "id": "09cec218",
   "metadata": {},
   "outputs": [],
   "source": [
    "# How can we get a clean dataset of COVID deaths in WI?"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "id": "08bb6bd8",
   "metadata": {},
   "outputs": [],
   "source": [
    "# cleanup based on documentation\n",
    "# TODO: when sharing results, mention this, and % of times we swapped in value\n",
    "df = df.replace(-999, 2) # -999 means between 0 and 4"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "id": "b75a23d8",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Index(['RptDt', 'GEOID', 'GEOName', 'POS_CUM_CP', 'POS_NEW_CP',\n",
       "       'POS_7DAYAVG_CP', 'DTH_CUM_CP', 'DTH_NEW_CP', 'DTH_7DAYAVG_CP',\n",
       "       'DTH_OVER_30DAYS_CP', 'DTH_CP_Daily', 'TESTS_POS_CONF',\n",
       "       'TESTS_POS_CONF_NEW', 'TESTS_POS_CONF_7DAYAVG', 'PercentNAATPos',\n",
       "       'TESTS_CUM', 'TESTS_NEW', 'TESTS_7DAYAVG', 'POS_CUM_CONF',\n",
       "       'POS_NEW_CONF', 'POS_7DAYAVG_CONF', 'POS_CUM_PROB', 'POS_NEW_PROB',\n",
       "       'POS_7DAYAVG_PROB', 'DTH_CUM_CONF', 'DTH_NEW_CONF', 'DTH_7DAYAVG_CONF',\n",
       "       'DTH_CONF_Daily', 'DTH_CUM_PROB', 'DTH_NEW_PROB', 'DTH_7DAYAVG_PROB',\n",
       "       'DTH_PROB_Daily', 'POS_MALE_CP', 'POS_FEM_CP', 'POS_OTH_CP',\n",
       "       'POS_0_9_CP', 'POS_10_19_CP', 'POS_20_29_CP', 'POS_30_39_CP',\n",
       "       'POS_40_49_CP', 'POS_50_59_CP', 'POS_60_69_CP', 'POS_70_79_CP',\n",
       "       'POS_80_89_CP', 'POS_90_CP', 'DTHS_FEM_CP', 'DTHS_MALE_CP',\n",
       "       'DTHS_OTH_CP', 'DTHS_0_9_CP', 'DTHS_10_19_CP', 'DTHS_20_29_CP',\n",
       "       'DTHS_30_39_CP', 'DTHS_40_49_CP', 'DTHS_50_59_CP', 'DTHS_60_69_CP',\n",
       "       'DTHS_70_79_CP', 'DTHS_80_89_CP', 'DTHS_90_CP', 'POS_AIAN_CP',\n",
       "       'POS_ASN_CP', 'POS_BLK_CP', 'POS_WHT_CP', 'POS_MLTOTH_CP', 'POS_UNK_CP',\n",
       "       'POS_E_HSP_CP', 'POS_E_NHSP_CP', 'POS_E_UNK_CP', 'DTH_AIAN_CP',\n",
       "       'DTH_ASN_CP', 'DTH_BLK_CP', 'DTH_WHT_CP', 'DTH_MLTOTH_CP', 'DTH_UNK_CP',\n",
       "       'DTH_E_HSP_CP', 'DTH_E_NHSP_CP', 'DTH_E_UNK_CP', 'POS_HC_Y_CP',\n",
       "       'POS_HC_N_CP', 'POS_HC_UNK_CP', 'HOSP_YES_CP', 'HOSP_NO_CP',\n",
       "       'HOSP_UNK_CP', 'Date', 'POP', 'POP_MOE', 'OBJECTID', 'GEO',\n",
       "       'ShapeSTArea', 'ShapeSTLength', 'geometry'],\n",
       "      dtype='object')"
      ]
     },
     "execution_count": 7,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# POP, POS_CUM_CP, POS_????_CP, DTH_CUM_CP\n",
    "df.columns"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "id": "b1829df6",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<AxesSubplot:xlabel='POP', ylabel='POS_CUM_CP'>"
      ]
     },
     "execution_count": 8,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "df.plot.scatter(x=\"POP\", y=\"POS_CUM_CP\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "id": "ec9447a1",
   "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>RptDt</th>\n",
       "      <th>GEOID</th>\n",
       "      <th>GEOName</th>\n",
       "      <th>POS_CUM_CP</th>\n",
       "      <th>POS_NEW_CP</th>\n",
       "      <th>POS_7DAYAVG_CP</th>\n",
       "      <th>DTH_CUM_CP</th>\n",
       "      <th>DTH_NEW_CP</th>\n",
       "      <th>DTH_7DAYAVG_CP</th>\n",
       "      <th>DTH_OVER_30DAYS_CP</th>\n",
       "      <th>...</th>\n",
       "      <th>HOSP_NO_CP</th>\n",
       "      <th>HOSP_UNK_CP</th>\n",
       "      <th>Date</th>\n",
       "      <th>POP</th>\n",
       "      <th>POP_MOE</th>\n",
       "      <th>OBJECTID</th>\n",
       "      <th>GEO</th>\n",
       "      <th>ShapeSTArea</th>\n",
       "      <th>ShapeSTLength</th>\n",
       "      <th>geometry</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>916</th>\n",
       "      <td>2022-04-07T19:00:00+00:00</td>\n",
       "      <td>TRACT N/A</td>\n",
       "      <td>None</td>\n",
       "      <td>39838</td>\n",
       "      <td>21</td>\n",
       "      <td>12.43</td>\n",
       "      <td>406</td>\n",
       "      <td>0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>17979</td>\n",
       "      <td>20213</td>\n",
       "      <td>2022-04-06T06:00:00+00:00</td>\n",
       "      <td>0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>2668</td>\n",
       "      <td>Census tract</td>\n",
       "      <td>4.661914e+02</td>\n",
       "      <td>80.071546</td>\n",
       "      <td>POLYGON ((-87.04159 44.13490, -87.04161 44.134...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>241</th>\n",
       "      <td>2022-04-07T19:00:00+00:00</td>\n",
       "      <td>55009940001</td>\n",
       "      <td>None</td>\n",
       "      <td>6549</td>\n",
       "      <td>3</td>\n",
       "      <td>1.86</td>\n",
       "      <td>29</td>\n",
       "      <td>0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>4691</td>\n",
       "      <td>1688</td>\n",
       "      <td>2022-04-06T06:00:00+00:00</td>\n",
       "      <td>18610</td>\n",
       "      <td>629.0</td>\n",
       "      <td>1201</td>\n",
       "      <td>Census tract</td>\n",
       "      <td>2.763480e+08</td>\n",
       "      <td>106951.691005</td>\n",
       "      <td>POLYGON ((-88.10048 44.54828, -88.10064 44.548...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1012</th>\n",
       "      <td>2022-04-07T19:00:00+00:00</td>\n",
       "      <td>55027960300</td>\n",
       "      <td>None</td>\n",
       "      <td>3879</td>\n",
       "      <td>0</td>\n",
       "      <td>0.57</td>\n",
       "      <td>29</td>\n",
       "      <td>0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>1363</td>\n",
       "      <td>2386</td>\n",
       "      <td>2022-04-06T06:00:00+00:00</td>\n",
       "      <td>7928</td>\n",
       "      <td>40.0</td>\n",
       "      <td>2766</td>\n",
       "      <td>Census tract</td>\n",
       "      <td>2.580188e+07</td>\n",
       "      <td>27162.474970</td>\n",
       "      <td>POLYGON ((-88.77561 43.59656, -88.77558 43.596...</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>3 rows × 90 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "                          RptDt        GEOID GEOName  POS_CUM_CP  POS_NEW_CP  \\\n",
       "916   2022-04-07T19:00:00+00:00    TRACT N/A    None       39838          21   \n",
       "241   2022-04-07T19:00:00+00:00  55009940001    None        6549           3   \n",
       "1012  2022-04-07T19:00:00+00:00  55027960300    None        3879           0   \n",
       "\n",
       "      POS_7DAYAVG_CP  DTH_CUM_CP  DTH_NEW_CP  DTH_7DAYAVG_CP  \\\n",
       "916            12.43         406           0             0.0   \n",
       "241             1.86          29           0             0.0   \n",
       "1012            0.57          29           0             0.0   \n",
       "\n",
       "      DTH_OVER_30DAYS_CP  ...  HOSP_NO_CP HOSP_UNK_CP  \\\n",
       "916                    0  ...       17979       20213   \n",
       "241                    0  ...        4691        1688   \n",
       "1012                   0  ...        1363        2386   \n",
       "\n",
       "                           Date    POP POP_MOE  OBJECTID           GEO  \\\n",
       "916   2022-04-06T06:00:00+00:00      0     NaN      2668  Census tract   \n",
       "241   2022-04-06T06:00:00+00:00  18610   629.0      1201  Census tract   \n",
       "1012  2022-04-06T06:00:00+00:00   7928    40.0      2766  Census tract   \n",
       "\n",
       "       ShapeSTArea  ShapeSTLength  \\\n",
       "916   4.661914e+02      80.071546   \n",
       "241   2.763480e+08  106951.691005   \n",
       "1012  2.580188e+07   27162.474970   \n",
       "\n",
       "                                               geometry  \n",
       "916   POLYGON ((-87.04159 44.13490, -87.04161 44.134...  \n",
       "241   POLYGON ((-88.10048 44.54828, -88.10064 44.548...  \n",
       "1012  POLYGON ((-88.77561 43.59656, -88.77558 43.596...  \n",
       "\n",
       "[3 rows x 90 columns]"
      ]
     },
     "execution_count": 9,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.sort_values(by=\"POS_CUM_CP\", ascending=False).head(3)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "id": "07af703d",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "True     1392\n",
       "False       1\n",
       "Name: GEOID, dtype: int64"
      ]
     },
     "execution_count": 10,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df[\"GEOID\"].str.fullmatch(r\"\\d+\").value_counts()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "id": "12ccc2a3",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<AxesSubplot:xlabel='POP', ylabel='POS_CUM_CP'>"
      ]
     },
     "execution_count": 11,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "# cleanup based on browsing the data\n",
    "# TODO: in presentation, state what % of deaths are excluded because we don't know geo\n",
    "df = df[df[\"GEOID\"].str.fullmatch(r\"\\d+\")]\n",
    "df.plot.scatter(x=\"POP\", y=\"POS_CUM_CP\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "id": "7e16357d",
   "metadata": {},
   "outputs": [],
   "source": [
    "# How can we train/fit models to known data to predict unknowns?\n",
    "# Population => Deaths, Cases => Deaths, Cases by Age => Deaths\n",
    "\n",
    "# GENERAL PATTERN\n",
    "# model = ????\n",
    "# model.fit(X, y)\n",
    "# y2 = model.predict(X)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "id": "ac8bd7be",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0       24\n",
       "1       19\n",
       "2       10\n",
       "3       11\n",
       "4        4\n",
       "        ..\n",
       "1388    15\n",
       "1389    23\n",
       "1390     7\n",
       "1391    22\n",
       "1392    21\n",
       "Name: DTH_CUM_CP, Length: 1392, dtype: int64"
      ]
     },
     "execution_count": 13,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# for X values, sklearn wants DataFrames\n",
    "# for y values, sklearn wants Series\n",
    "\n",
    "df[[\"POP\"]] # X\n",
    "df[\"DTH_CUM_CP\"] # y"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "id": "e49d1025",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([10.66799181,  8.62647161, 12.86546147, ...,  9.7606495 ,\n",
       "       15.96419749,  8.59001589])"
      ]
     },
     "execution_count": 14,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "from sklearn.linear_model import LinearRegression\n",
    "\n",
    "model = LinearRegression()\n",
    "model.fit(df[[\"POP\"]], df[\"DTH_CUM_CP\"])  # fit means training (learn the relation between X and y)\n",
    "y2 = model.predict(df[[\"POP\"]])\n",
    "y2"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "id": "e0cc0dea",
   "metadata": {},
   "outputs": [],
   "source": [
    "import pandas as pd\n",
    "#pd.Series(y2)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "id": "7d7d6948",
   "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>POP</th>\n",
       "      <th>predicted_deaths</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>1</td>\n",
       "      <td>1.604695</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>100</td>\n",
       "      <td>1.805202</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>1000</td>\n",
       "      <td>3.627988</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "    POP  predicted_deaths\n",
       "0     1          1.604695\n",
       "1   100          1.805202\n",
       "2  1000          3.627988"
      ]
     },
     "execution_count": 16,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "prediction_df = pd.DataFrame({\"POP\": [1, 100, 1000]})\n",
    "prediction_df[\"predicted_deaths\"] = model.predict(prediction_df)\n",
    "prediction_df"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "id": "33d12863",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<AxesSubplot:xlabel='POP', ylabel='DTH_CUM_CP'>"
      ]
     },
     "execution_count": 17,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "# How can we visualize model predictions?\n",
    "prediction_df = pd.DataFrame({\"POP\": range(1, 20000)})\n",
    "prediction_df[\"predicted_deaths\"] = model.predict(prediction_df)\n",
    "\n",
    "ax = prediction_df.plot.line(x=\"POP\", y=\"predicted_deaths\", color=\"red\")\n",
    "df.plot.scatter(x=\"POP\", y=\"DTH_CUM_CP\", ax=ax, color=\"black\", alpha=0.05)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "id": "0e7532e4",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "1.6026699609658888"
      ]
     },
     "execution_count": 18,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# How can we get a formula for the relationship?\n",
    "model.intercept_"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "id": "ae9dd84d",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([0.00202532])"
      ]
     },
     "execution_count": 19,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "model.coef_"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "id": "603d0636",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.00202531765994764"
      ]
     },
     "execution_count": 20,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "model.coef_[0]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "id": "dc233aa0",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "'deaths ~= 0.002025*population + 1.6'"
      ]
     },
     "execution_count": 21,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "f\"deaths ~= {round(model.coef_[0], 6)}*population + {round(model.intercept_, 2)}\""
   ]
  },
  {
   "cell_type": "markdown",
   "id": "4570327c",
   "metadata": {},
   "source": [
    "y = 3*x1 + 2*x2 + 10*x3 + 100"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "id": "d76f7258",
   "metadata": {},
   "outputs": [],
   "source": [
    "# How well does our model fit the data?\n",
    "# r^2 score (explained variance)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "id": "54893182",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "64.45856160704695"
      ]
     },
     "execution_count": 23,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "variance_before = df[\"DTH_CUM_CP\"].var()\n",
    "variance_before"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "id": "b6ca1e46",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "51.94852682334216"
      ]
     },
     "execution_count": 24,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "variance_after = (df[\"DTH_CUM_CP\"] - model.predict(df[[\"POP\"]])).var()\n",
    "variance_after"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "id": "07fe6e5f",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "12.510034783704782"
      ]
     },
     "execution_count": 25,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "improvement = variance_before - variance_after\n",
    "improvement"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "id": "f2346628",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.19407871463171342"
      ]
     },
     "execution_count": 26,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "improvement / variance_before # r^2 score"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "id": "206646b9",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.19407871463171344"
      ]
     },
     "execution_count": 27,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "model.score(df[[\"POP\"]], df[\"DTH_CUM_CP\"]) # calls predict for us, computes the r^2"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "id": "0aaeacec",
   "metadata": {},
   "outputs": [],
   "source": [
    "# Did our model learn, or just memorize (that is, \"overfit\")?\n",
    "from sklearn.model_selection import train_test_split, cross_val_score"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "id": "3a4a59f5",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(1044, 348)"
      ]
     },
     "execution_count": 29,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "train, test = train_test_split(df)\n",
    "len(train), len(test)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "id": "01421126",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.23358299890443257"
      ]
     },
     "execution_count": 30,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "model = LinearRegression()\n",
    "model.fit(train[[\"POP\"]], train[\"DTH_CUM_CP\"])\n",
    "model.score(test[[\"POP\"]], test[\"DTH_CUM_CP\"])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "id": "9d9b5159",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([0.08339123, 0.234877  , 0.2264941 , 0.16216738, 0.21302395])"
      ]
     },
     "execution_count": 31,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# How can we minimize noise due to random train/test splits?\n",
    "scores = cross_val_score(model, train[[\"POP\"]], train[\"DTH_CUM_CP\"])\n",
    "scores"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "id": "e2fe6d44",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.18399073025506374"
      ]
     },
     "execution_count": 32,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "scores.mean()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 33,
   "id": "767617f1",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.056276806592814664"
      ]
     },
     "execution_count": 33,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "scores.std()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 34,
   "id": "ad4dfea9",
   "metadata": {},
   "outputs": [],
   "source": [
    "# How can we compare models?\n",
    "model1 = LinearRegression()   # pop => deaths\n",
    "model2 = LinearRegression()   # cases => deaths\n",
    "\n",
    "scores1 = cross_val_score(model1, train[[\"POP\"]], train[\"DTH_CUM_CP\"])\n",
    "scores2 = cross_val_score(model2, train[[\"POS_CUM_CP\"]], train[\"DTH_CUM_CP\"])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 35,
   "id": "5ae467a3",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Text(0, 0.5, 'R^2 Score')"
      ]
     },
     "execution_count": 35,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "avg = pd.Series({\"Pop Model\": scores1.mean(), \"Cases Model\": scores2.mean()})\n",
    "std = pd.Series({\"Pop Model\": scores1.std(), \"Cases Model\": scores2.std()})\n",
    "ax = avg.plot.bar(yerr=std)\n",
    "ax.set_ylabel(\"R^2 Score\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 36,
   "id": "47c6a827",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.6867737981016822"
      ]
     },
     "execution_count": 36,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# How can we use multiple x variables (multiple regression)?\n",
    "xcols = ['POS_0_9_CP', 'POS_10_19_CP', 'POS_20_29_CP', 'POS_30_39_CP',\n",
    "       'POS_40_49_CP', 'POS_50_59_CP', 'POS_60_69_CP', 'POS_70_79_CP',\n",
    "       'POS_80_89_CP', 'POS_90_CP']\n",
    "\n",
    "model = LinearRegression()\n",
    "model.fit(train[xcols], train[\"DTH_CUM_CP\"])\n",
    "model.score(test[xcols], test[\"DTH_CUM_CP\"])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 37,
   "id": "9950af84",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Index(['RptDt', 'GEOID', 'GEOName', 'POS_CUM_CP', 'POS_NEW_CP',\n",
       "       'POS_7DAYAVG_CP', 'DTH_CUM_CP', 'DTH_NEW_CP', 'DTH_7DAYAVG_CP',\n",
       "       'DTH_OVER_30DAYS_CP', 'DTH_CP_Daily', 'TESTS_POS_CONF',\n",
       "       'TESTS_POS_CONF_NEW', 'TESTS_POS_CONF_7DAYAVG', 'PercentNAATPos',\n",
       "       'TESTS_CUM', 'TESTS_NEW', 'TESTS_7DAYAVG', 'POS_CUM_CONF',\n",
       "       'POS_NEW_CONF', 'POS_7DAYAVG_CONF', 'POS_CUM_PROB', 'POS_NEW_PROB',\n",
       "       'POS_7DAYAVG_PROB', 'DTH_CUM_CONF', 'DTH_NEW_CONF', 'DTH_7DAYAVG_CONF',\n",
       "       'DTH_CONF_Daily', 'DTH_CUM_PROB', 'DTH_NEW_PROB', 'DTH_7DAYAVG_PROB',\n",
       "       'DTH_PROB_Daily', 'POS_MALE_CP', 'POS_FEM_CP', 'POS_OTH_CP',\n",
       "       'POS_0_9_CP', 'POS_10_19_CP', 'POS_20_29_CP', 'POS_30_39_CP',\n",
       "       'POS_40_49_CP', 'POS_50_59_CP', 'POS_60_69_CP', 'POS_70_79_CP',\n",
       "       'POS_80_89_CP', 'POS_90_CP', 'DTHS_FEM_CP', 'DTHS_MALE_CP',\n",
       "       'DTHS_OTH_CP', 'DTHS_0_9_CP', 'DTHS_10_19_CP', 'DTHS_20_29_CP',\n",
       "       'DTHS_30_39_CP', 'DTHS_40_49_CP', 'DTHS_50_59_CP', 'DTHS_60_69_CP',\n",
       "       'DTHS_70_79_CP', 'DTHS_80_89_CP', 'DTHS_90_CP', 'POS_AIAN_CP',\n",
       "       'POS_ASN_CP', 'POS_BLK_CP', 'POS_WHT_CP', 'POS_MLTOTH_CP', 'POS_UNK_CP',\n",
       "       'POS_E_HSP_CP', 'POS_E_NHSP_CP', 'POS_E_UNK_CP', 'DTH_AIAN_CP',\n",
       "       'DTH_ASN_CP', 'DTH_BLK_CP', 'DTH_WHT_CP', 'DTH_MLTOTH_CP', 'DTH_UNK_CP',\n",
       "       'DTH_E_HSP_CP', 'DTH_E_NHSP_CP', 'DTH_E_UNK_CP', 'POS_HC_Y_CP',\n",
       "       'POS_HC_N_CP', 'POS_HC_UNK_CP', 'HOSP_YES_CP', 'HOSP_NO_CP',\n",
       "       'HOSP_UNK_CP', 'Date', 'POP', 'POP_MOE', 'OBJECTID', 'GEO',\n",
       "       'ShapeSTArea', 'ShapeSTLength', 'geometry'],\n",
       "      dtype='object')"
      ]
     },
     "execution_count": 37,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.columns"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 39,
   "id": "17962c9d",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([-3.06228488e-02,  1.86787556e-04, -3.71295962e-04,  1.64123894e-02,\n",
       "        1.22816578e-02, -1.21099247e-02,  1.37278694e-02,  2.15745405e-02,\n",
       "        1.70873273e-01,  3.05043707e-01])"
      ]
     },
     "execution_count": 39,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# How can we interpret what features the model is relying on?\n",
    "model.coef_"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 40,
   "id": "6c493f28",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "['POS_0_9_CP',\n",
       " 'POS_10_19_CP',\n",
       " 'POS_20_29_CP',\n",
       " 'POS_30_39_CP',\n",
       " 'POS_40_49_CP',\n",
       " 'POS_50_59_CP',\n",
       " 'POS_60_69_CP',\n",
       " 'POS_70_79_CP',\n",
       " 'POS_80_89_CP',\n",
       " 'POS_90_CP']"
      ]
     },
     "execution_count": 40,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "xcols"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 41,
   "id": "c30ece77",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<AxesSubplot:>"
      ]
     },
     "execution_count": 41,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "pd.Series(model.coef_, index=xcols).plot.bar()"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3 (ipykernel)",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.8.10"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}