aggregate_counts_over_intersections.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "id": "8a18b75f",
   "metadata": {
    "tags": []
   },
   "outputs": [],
   "source": [
    "import pandas as pd"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "cd0e0dfa",
   "metadata": {},
   "source": [
    "# Aggregate Counts over Intersections"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "0da399c8",
   "metadata": {
    "tags": []
   },
   "source": [
    "## Load dataset"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "id": "5ea0f919",
   "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>PanoramaId</th>\n",
       "      <th>n_cameras_median</th>\n",
       "      <th>attached_street_median</th>\n",
       "      <th>attached_building_median</th>\n",
       "      <th>attached_unknown_median</th>\n",
       "      <th>type_dome_median</th>\n",
       "      <th>type_bullet_median</th>\n",
       "      <th>type_unknown_median</th>\n",
       "      <th>n_cameras_agreement</th>\n",
       "      <th>attached_street_agreement</th>\n",
       "      <th>...</th>\n",
       "      <th>attached_unknown_agreement</th>\n",
       "      <th>type_dome_agreement</th>\n",
       "      <th>type_bullet_agreement</th>\n",
       "      <th>type_unknown_agreement</th>\n",
       "      <th>Lat</th>\n",
       "      <th>Long</th>\n",
       "      <th>geometry_pano</th>\n",
       "      <th>BoroName</th>\n",
       "      <th>URL</th>\n",
       "      <th>ImageDate</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>--48RnFczF55WcDFpy5y_Q</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>2 vs 1</td>\n",
       "      <td>Unanimous</td>\n",
       "      <td>...</td>\n",
       "      <td>Unanimous</td>\n",
       "      <td>Unanimous</td>\n",
       "      <td>Unanimous</td>\n",
       "      <td>Unanimous</td>\n",
       "      <td>40.870564</td>\n",
       "      <td>-73.869327</td>\n",
       "      <td>POINT (-73.86932652339482 40.87056425110688)</td>\n",
       "      <td>Bronx</td>\n",
       "      <td>https://www.google.com/maps/@?api=1&amp;map_action...</td>\n",
       "      <td>2019-08</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>--T2iuvfxXJrzbgD7R3W-g</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>2 vs 1</td>\n",
       "      <td>Unanimous</td>\n",
       "      <td>...</td>\n",
       "      <td>2 vs 1</td>\n",
       "      <td>Unanimous</td>\n",
       "      <td>Unanimous</td>\n",
       "      <td>Unanimous</td>\n",
       "      <td>40.669099</td>\n",
       "      <td>-73.762447</td>\n",
       "      <td>POINT (-73.76244689609075 40.6690985670363)</td>\n",
       "      <td>Queens</td>\n",
       "      <td>https://www.google.com/maps/@?api=1&amp;map_action...</td>\n",
       "      <td>2020-10</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>--VGAsqPgTXxgZFXuhuhbg</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>Unanimous</td>\n",
       "      <td>Unanimous</td>\n",
       "      <td>...</td>\n",
       "      <td>Unanimous</td>\n",
       "      <td>Unanimous</td>\n",
       "      <td>Unanimous</td>\n",
       "      <td>Unanimous</td>\n",
       "      <td>40.766790</td>\n",
       "      <td>-73.771211</td>\n",
       "      <td>POINT (-73.7712107859631 40.76679030231237)</td>\n",
       "      <td>Queens</td>\n",
       "      <td>https://www.google.com/maps/@?api=1&amp;map_action...</td>\n",
       "      <td>2020-11</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>--YJY0GGvay_3W_O5ObeNQ</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>2 vs 1</td>\n",
       "      <td>Unanimous</td>\n",
       "      <td>...</td>\n",
       "      <td>Unanimous</td>\n",
       "      <td>Unanimous</td>\n",
       "      <td>Unanimous</td>\n",
       "      <td>Unanimous</td>\n",
       "      <td>40.554253</td>\n",
       "      <td>-74.201603</td>\n",
       "      <td>POINT (-74.20160298846051 40.55425348634071)</td>\n",
       "      <td>Staten Island</td>\n",
       "      <td>https://www.google.com/maps/@?api=1&amp;map_action...</td>\n",
       "      <td>2019-09</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>--fUkaf5nn1-bm4KiJ9pVA</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>2 vs 1</td>\n",
       "      <td>Unanimous</td>\n",
       "      <td>...</td>\n",
       "      <td>Unanimous</td>\n",
       "      <td>Unanimous</td>\n",
       "      <td>Unanimous</td>\n",
       "      <td>Unanimous</td>\n",
       "      <td>40.611084</td>\n",
       "      <td>-74.162803</td>\n",
       "      <td>POINT (-74.16280280196275 40.61108412672856)</td>\n",
       "      <td>Staten Island</td>\n",
       "      <td>https://www.google.com/maps/@?api=1&amp;map_action...</td>\n",
       "      <td>2019-09</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>43401</th>\n",
       "      <td>zzj23G_f-_hZa1LrXJ7tig</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>2 vs 1</td>\n",
       "      <td>2 vs 1</td>\n",
       "      <td>...</td>\n",
       "      <td>Unanimous</td>\n",
       "      <td>Unanimous</td>\n",
       "      <td>Unanimous</td>\n",
       "      <td>2 vs 1</td>\n",
       "      <td>40.889133</td>\n",
       "      <td>-73.821303</td>\n",
       "      <td>POINT (-73.82130317259254 40.88913324730224)</td>\n",
       "      <td>Bronx</td>\n",
       "      <td>https://www.google.com/maps/@?api=1&amp;map_action...</td>\n",
       "      <td>2019-11</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>43402</th>\n",
       "      <td>zzpZoM4JHrGNywYitkYT8w</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>2 vs 1</td>\n",
       "      <td>2 vs 1</td>\n",
       "      <td>...</td>\n",
       "      <td>Unanimous</td>\n",
       "      <td>2 vs 1</td>\n",
       "      <td>Unanimous</td>\n",
       "      <td>Unanimous</td>\n",
       "      <td>40.750994</td>\n",
       "      <td>-73.706451</td>\n",
       "      <td>POINT (-73.7064507165946 40.75099428476829)</td>\n",
       "      <td>Queens</td>\n",
       "      <td>https://www.google.com/maps/@?api=1&amp;map_action...</td>\n",
       "      <td>2019-07</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>43403</th>\n",
       "      <td>zztPjLBFQjwEISsA75a-zg</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>All disagree</td>\n",
       "      <td>All disagree</td>\n",
       "      <td>...</td>\n",
       "      <td>Unanimous</td>\n",
       "      <td>All disagree</td>\n",
       "      <td>2 vs 1</td>\n",
       "      <td>Unanimous</td>\n",
       "      <td>40.730852</td>\n",
       "      <td>-73.947876</td>\n",
       "      <td>POINT (-73.9478762022405 40.73085200510598)</td>\n",
       "      <td>Brooklyn</td>\n",
       "      <td>https://www.google.com/maps/@?api=1&amp;map_action...</td>\n",
       "      <td>2020-10</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>43404</th>\n",
       "      <td>zzv84J0v5tQobN5uCXO1xw</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>2 vs 1</td>\n",
       "      <td>Unanimous</td>\n",
       "      <td>...</td>\n",
       "      <td>Unanimous</td>\n",
       "      <td>Unanimous</td>\n",
       "      <td>Unanimous</td>\n",
       "      <td>Unanimous</td>\n",
       "      <td>40.675203</td>\n",
       "      <td>-73.778650</td>\n",
       "      <td>POINT (-73.77864970575715 40.67520299190087)</td>\n",
       "      <td>Queens</td>\n",
       "      <td>https://www.google.com/maps/@?api=1&amp;map_action...</td>\n",
       "      <td>2018-07</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>43405</th>\n",
       "      <td>zzyhPWndbm7wxqTN6LuYLA</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>2 vs 1</td>\n",
       "      <td>Unanimous</td>\n",
       "      <td>...</td>\n",
       "      <td>Unanimous</td>\n",
       "      <td>Unanimous</td>\n",
       "      <td>Unanimous</td>\n",
       "      <td>Unanimous</td>\n",
       "      <td>40.713562</td>\n",
       "      <td>-73.779712</td>\n",
       "      <td>POINT (-73.77971167180961 40.71356164178761)</td>\n",
       "      <td>Queens</td>\n",
       "      <td>https://www.google.com/maps/@?api=1&amp;map_action...</td>\n",
       "      <td>2016-11</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>43406 rows × 21 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "                   PanoramaId  n_cameras_median  attached_street_median  \\\n",
       "0      --48RnFczF55WcDFpy5y_Q                 0                       0   \n",
       "1      --T2iuvfxXJrzbgD7R3W-g                 0                       0   \n",
       "2      --VGAsqPgTXxgZFXuhuhbg                 0                       0   \n",
       "3      --YJY0GGvay_3W_O5ObeNQ                 0                       0   \n",
       "4      --fUkaf5nn1-bm4KiJ9pVA                 1                       0   \n",
       "...                       ...               ...                     ...   \n",
       "43401  zzj23G_f-_hZa1LrXJ7tig                 0                       0   \n",
       "43402  zzpZoM4JHrGNywYitkYT8w                 1                       0   \n",
       "43403  zztPjLBFQjwEISsA75a-zg                 1                       1   \n",
       "43404  zzv84J0v5tQobN5uCXO1xw                 0                       0   \n",
       "43405  zzyhPWndbm7wxqTN6LuYLA                 0                       0   \n",
       "\n",
       "       attached_building_median  attached_unknown_median  type_dome_median  \\\n",
       "0                             0                        0                 0   \n",
       "1                             0                        0                 0   \n",
       "2                             0                        0                 0   \n",
       "3                             0                        0                 0   \n",
       "4                             1                        0                 0   \n",
       "...                         ...                      ...               ...   \n",
       "43401                         0                        0                 0   \n",
       "43402                         0                        0                 0   \n",
       "43403                         0                        0                 1   \n",
       "43404                         0                        0                 0   \n",
       "43405                         0                        0                 0   \n",
       "\n",
       "       type_bullet_median  type_unknown_median n_cameras_agreement  \\\n",
       "0                       0                    0              2 vs 1   \n",
       "1                       0                    0              2 vs 1   \n",
       "2                       0                    0           Unanimous   \n",
       "3                       0                    0              2 vs 1   \n",
       "4                       0                    0              2 vs 1   \n",
       "...                   ...                  ...                 ...   \n",
       "43401                   0                    0              2 vs 1   \n",
       "43402                   0                    0              2 vs 1   \n",
       "43403                   0                    0        All disagree   \n",
       "43404                   0                    0              2 vs 1   \n",
       "43405                   0                    0              2 vs 1   \n",
       "\n",
       "      attached_street_agreement  ... attached_unknown_agreement  \\\n",
       "0                     Unanimous  ...                  Unanimous   \n",
       "1                     Unanimous  ...                     2 vs 1   \n",
       "2                     Unanimous  ...                  Unanimous   \n",
       "3                     Unanimous  ...                  Unanimous   \n",
       "4                     Unanimous  ...                  Unanimous   \n",
       "...                         ...  ...                        ...   \n",
       "43401                    2 vs 1  ...                  Unanimous   \n",
       "43402                    2 vs 1  ...                  Unanimous   \n",
       "43403              All disagree  ...                  Unanimous   \n",
       "43404                 Unanimous  ...                  Unanimous   \n",
       "43405                 Unanimous  ...                  Unanimous   \n",
       "\n",
       "      type_dome_agreement type_bullet_agreement type_unknown_agreement  \\\n",
       "0               Unanimous             Unanimous              Unanimous   \n",
       "1               Unanimous             Unanimous              Unanimous   \n",
       "2               Unanimous             Unanimous              Unanimous   \n",
       "3               Unanimous             Unanimous              Unanimous   \n",
       "4               Unanimous             Unanimous              Unanimous   \n",
       "...                   ...                   ...                    ...   \n",
       "43401           Unanimous             Unanimous                 2 vs 1   \n",
       "43402              2 vs 1             Unanimous              Unanimous   \n",
       "43403        All disagree                2 vs 1              Unanimous   \n",
       "43404           Unanimous             Unanimous              Unanimous   \n",
       "43405           Unanimous             Unanimous              Unanimous   \n",
       "\n",
       "             Lat       Long                                 geometry_pano  \\\n",
       "0      40.870564 -73.869327  POINT (-73.86932652339482 40.87056425110688)   \n",
       "1      40.669099 -73.762447   POINT (-73.76244689609075 40.6690985670363)   \n",
       "2      40.766790 -73.771211   POINT (-73.7712107859631 40.76679030231237)   \n",
       "3      40.554253 -74.201603  POINT (-74.20160298846051 40.55425348634071)   \n",
       "4      40.611084 -74.162803  POINT (-74.16280280196275 40.61108412672856)   \n",
       "...          ...        ...                                           ...   \n",
       "43401  40.889133 -73.821303  POINT (-73.82130317259254 40.88913324730224)   \n",
       "43402  40.750994 -73.706451   POINT (-73.7064507165946 40.75099428476829)   \n",
       "43403  40.730852 -73.947876   POINT (-73.9478762022405 40.73085200510598)   \n",
       "43404  40.675203 -73.778650  POINT (-73.77864970575715 40.67520299190087)   \n",
       "43405  40.713562 -73.779712  POINT (-73.77971167180961 40.71356164178761)   \n",
       "\n",
       "            BoroName                                                URL  \\\n",
       "0              Bronx  https://www.google.com/maps/@?api=1&map_action...   \n",
       "1             Queens  https://www.google.com/maps/@?api=1&map_action...   \n",
       "2             Queens  https://www.google.com/maps/@?api=1&map_action...   \n",
       "3      Staten Island  https://www.google.com/maps/@?api=1&map_action...   \n",
       "4      Staten Island  https://www.google.com/maps/@?api=1&map_action...   \n",
       "...              ...                                                ...   \n",
       "43401          Bronx  https://www.google.com/maps/@?api=1&map_action...   \n",
       "43402         Queens  https://www.google.com/maps/@?api=1&map_action...   \n",
       "43403       Brooklyn  https://www.google.com/maps/@?api=1&map_action...   \n",
       "43404         Queens  https://www.google.com/maps/@?api=1&map_action...   \n",
       "43405         Queens  https://www.google.com/maps/@?api=1&map_action...   \n",
       "\n",
       "      ImageDate  \n",
       "0       2019-08  \n",
       "1       2020-10  \n",
       "2       2020-11  \n",
       "3       2019-09  \n",
       "4       2019-09  \n",
       "...         ...  \n",
       "43401   2019-11  \n",
       "43402   2019-07  \n",
       "43403   2020-10  \n",
       "43404   2018-07  \n",
       "43405   2016-11  \n",
       "\n",
       "[43406 rows x 21 columns]"
      ]
     },
     "execution_count": 2,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "full_aggregated_data = pd.read_csv(\"data/counts_per_intersections.csv\")\n",
    "full_aggregated_data"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "74a3ab28",
   "metadata": {},
   "source": [
    "## Plot strata frequencies for importance sampling"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "id": "ba2c46de",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 1600x1200 with 9 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/plain": [
       "<ggplot: (402564559)>"
      ]
     },
     "execution_count": 3,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "from plotnine import ggplot, aes, geom_histogram, facet_wrap, theme, ylab, element_text\n",
    "\n",
    "to_plot = full_aggregated_data[[\"attached_street_agreement\", \"attached_street_median\"]].copy()\n",
    "to_plot.columns=[\"Agreement\", \"Median\"]\n",
    "agreement_type = pd.CategoricalDtype(\n",
    "    categories=[\"Unanimous\", \"2 vs 1\", \"All disagree\"],\n",
    "    ordered=True\n",
    ")\n",
    "to_plot[\"Agreement\"] = to_plot[\"Agreement\"].astype(agreement_type)\n",
    "\n",
    "\n",
    "(ggplot(to_plot, aes(\"Agreement\")) \n",
    "+ geom_histogram(aes(fill=\"Median\"), color=\"white\", binwidth=1)\n",
    "+ facet_wrap(\"Median\", scales='free_y', labeller=\"label_both\")\n",
    "+ ylab(\"Intersections\")\n",
    "+ theme(\n",
    "    subplots_adjust={'wspace': 0.25}, \n",
    "    axis_text_x=element_text(angle=45, ha=\"right\"),\n",
    "    figure_size=(8, 6),\n",
    "    dpi = 200,\n",
    "    legend_position=\"none\")\n",
    ") "
   ]
  },
  {
   "cell_type": "markdown",
   "id": "1c1f4d6c",
   "metadata": {},
   "source": [
    "## Go from wide to long"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "id": "afaa92e5",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "['n_cameras_median',\n",
       " 'attached_street_median',\n",
       " 'attached_building_median',\n",
       " 'attached_unknown_median',\n",
       " 'type_dome_median',\n",
       " 'type_bullet_median',\n",
       " 'type_unknown_median']"
      ]
     },
     "execution_count": 4,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "median_columns = [x for x in full_aggregated_data.columns if x.endswith(\"_median\")]\n",
    "median_columns"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "id": "1ba5397c",
   "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>PanoramaId</th>\n",
       "      <th>BoroName</th>\n",
       "      <th>subset</th>\n",
       "      <th>count</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>--48RnFczF55WcDFpy5y_Q</td>\n",
       "      <td>Bronx</td>\n",
       "      <td>n_cameras</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>--T2iuvfxXJrzbgD7R3W-g</td>\n",
       "      <td>Queens</td>\n",
       "      <td>n_cameras</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>--VGAsqPgTXxgZFXuhuhbg</td>\n",
       "      <td>Queens</td>\n",
       "      <td>n_cameras</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>--YJY0GGvay_3W_O5ObeNQ</td>\n",
       "      <td>Staten Island</td>\n",
       "      <td>n_cameras</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>--fUkaf5nn1-bm4KiJ9pVA</td>\n",
       "      <td>Staten Island</td>\n",
       "      <td>n_cameras</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>303837</th>\n",
       "      <td>zzj23G_f-_hZa1LrXJ7tig</td>\n",
       "      <td>Bronx</td>\n",
       "      <td>type_unknown</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>303838</th>\n",
       "      <td>zzpZoM4JHrGNywYitkYT8w</td>\n",
       "      <td>Queens</td>\n",
       "      <td>type_unknown</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>303839</th>\n",
       "      <td>zztPjLBFQjwEISsA75a-zg</td>\n",
       "      <td>Brooklyn</td>\n",
       "      <td>type_unknown</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>303840</th>\n",
       "      <td>zzv84J0v5tQobN5uCXO1xw</td>\n",
       "      <td>Queens</td>\n",
       "      <td>type_unknown</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>303841</th>\n",
       "      <td>zzyhPWndbm7wxqTN6LuYLA</td>\n",
       "      <td>Queens</td>\n",
       "      <td>type_unknown</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>303842 rows × 4 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "                    PanoramaId       BoroName        subset  count\n",
       "0       --48RnFczF55WcDFpy5y_Q          Bronx     n_cameras      0\n",
       "1       --T2iuvfxXJrzbgD7R3W-g         Queens     n_cameras      0\n",
       "2       --VGAsqPgTXxgZFXuhuhbg         Queens     n_cameras      0\n",
       "3       --YJY0GGvay_3W_O5ObeNQ  Staten Island     n_cameras      0\n",
       "4       --fUkaf5nn1-bm4KiJ9pVA  Staten Island     n_cameras      1\n",
       "...                        ...            ...           ...    ...\n",
       "303837  zzj23G_f-_hZa1LrXJ7tig          Bronx  type_unknown      0\n",
       "303838  zzpZoM4JHrGNywYitkYT8w         Queens  type_unknown      0\n",
       "303839  zztPjLBFQjwEISsA75a-zg       Brooklyn  type_unknown      0\n",
       "303840  zzv84J0v5tQobN5uCXO1xw         Queens  type_unknown      0\n",
       "303841  zzyhPWndbm7wxqTN6LuYLA         Queens  type_unknown      0\n",
       "\n",
       "[303842 rows x 4 columns]"
      ]
     },
     "execution_count": 5,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "final_long = full_aggregated_data.reset_index().melt(id_vars = ['PanoramaId', 'BoroName'],\n",
    "                                 value_vars = median_columns,\n",
    "                                 var_name = \"subset\",\n",
    "                                 value_name = \"count\"\n",
    "                                )\n",
    "final_long[\"subset\"] = final_long[\"subset\"].str.replace(\"_median\", \"\")\n",
    "final_long"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "50a44f02",
   "metadata": {},
   "source": [
    "## Summary of camera counts"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "id": "603f4062",
   "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>count</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>subset</th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>attached_building</th>\n",
       "      <td>22133</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>attached_street</th>\n",
       "      <td>3317</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>attached_unknown</th>\n",
       "      <td>86</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>n_cameras</th>\n",
       "      <td>28954</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>type_bullet</th>\n",
       "      <td>444</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>type_dome</th>\n",
       "      <td>2266</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>type_unknown</th>\n",
       "      <td>63</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                   count\n",
       "subset                  \n",
       "attached_building  22133\n",
       "attached_street     3317\n",
       "attached_unknown      86\n",
       "n_cameras          28954\n",
       "type_bullet          444\n",
       "type_dome           2266\n",
       "type_unknown          63"
      ]
     },
     "execution_count": 6,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "counts_per_subset = final_long.groupby(\"subset\").sum()\n",
    "counts_per_subset"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "id": "368382f6",
   "metadata": {},
   "outputs": [],
   "source": [
    "total_cameras = counts_per_subset.loc[[\"attached_building\", \"attached_street\", \"attached_unknown\"]].sum()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "id": "546a211d",
   "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>count</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>subset</th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>attached_building</th>\n",
       "      <td>86.673716</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>attached_street</th>\n",
       "      <td>12.989505</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>attached_unknown</th>\n",
       "      <td>0.336779</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>n_cameras</th>\n",
       "      <td>113.385025</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                        count\n",
       "subset                       \n",
       "attached_building   86.673716\n",
       "attached_street     12.989505\n",
       "attached_unknown     0.336779\n",
       "n_cameras          113.385025"
      ]
     },
     "execution_count": 8,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "counts_per_subset.loc[[\"attached_building\", \"attached_street\", \"attached_unknown\", \"n_cameras\"]] * 100 / total_cameras"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "id": "d227611a",
   "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>count</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>subset</th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>type_bullet</th>\n",
       "      <td>444</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>type_dome</th>\n",
       "      <td>2266</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>type_unknown</th>\n",
       "      <td>63</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "              count\n",
       "subset             \n",
       "type_bullet     444\n",
       "type_dome      2266\n",
       "type_unknown     63"
      ]
     },
     "execution_count": 9,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "counts_per_subset.loc[[\"type_bullet\", \"type_dome\", \"type_unknown\"]]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "id": "bfce4480",
   "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>count</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>subset</th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>type_bullet</th>\n",
       "      <td>1.738722</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>type_dome</th>\n",
       "      <td>8.873747</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>type_unknown</th>\n",
       "      <td>0.246711</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                 count\n",
       "subset                \n",
       "type_bullet   1.738722\n",
       "type_dome     8.873747\n",
       "type_unknown  0.246711"
      ]
     },
     "execution_count": 10,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "counts_per_subset.loc[[\"type_bullet\", \"type_dome\", \"type_unknown\"]] * 100 / total_cameras"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "725cbd4a",
   "metadata": {},
   "source": [
    "## Focus on Borough"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "id": "8508377a",
   "metadata": {
    "tags": []
   },
   "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></th>\n",
       "      <th>count</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>BoroName</th>\n",
       "      <th>subset</th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th rowspan=\"3\" valign=\"top\">Bronx</th>\n",
       "      <th>attached_building</th>\n",
       "      <td>3275</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>attached_street</th>\n",
       "      <td>456</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>attached_unknown</th>\n",
       "      <td>6</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th rowspan=\"3\" valign=\"top\">Brooklyn</th>\n",
       "      <th>attached_building</th>\n",
       "      <td>8330</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>attached_street</th>\n",
       "      <td>871</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>attached_unknown</th>\n",
       "      <td>31</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th rowspan=\"3\" valign=\"top\">Manhattan</th>\n",
       "      <th>attached_building</th>\n",
       "      <td>3213</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>attached_street</th>\n",
       "      <td>724</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>attached_unknown</th>\n",
       "      <td>11</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th rowspan=\"3\" valign=\"top\">Queens</th>\n",
       "      <th>attached_building</th>\n",
       "      <td>6459</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>attached_street</th>\n",
       "      <td>1088</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>attached_unknown</th>\n",
       "      <td>33</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th rowspan=\"3\" valign=\"top\">Staten Island</th>\n",
       "      <th>attached_building</th>\n",
       "      <td>852</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>attached_street</th>\n",
       "      <td>178</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>attached_unknown</th>\n",
       "      <td>5</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                                 count\n",
       "BoroName      subset                  \n",
       "Bronx         attached_building   3275\n",
       "              attached_street      456\n",
       "              attached_unknown       6\n",
       "Brooklyn      attached_building   8330\n",
       "              attached_street      871\n",
       "              attached_unknown      31\n",
       "Manhattan     attached_building   3213\n",
       "              attached_street      724\n",
       "              attached_unknown      11\n",
       "Queens        attached_building   6459\n",
       "              attached_street     1088\n",
       "              attached_unknown      33\n",
       "Staten Island attached_building    852\n",
       "              attached_street      178\n",
       "              attached_unknown       5"
      ]
     },
     "execution_count": 11,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "by_borough = final_long[final_long[\"subset\"].str.startswith(\"attached_\")].groupby([\"BoroName\", \"subset\"]).sum()\n",
    "by_borough"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "id": "2311ba42",
   "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></th>\n",
       "      <th>count</th>\n",
       "      <th>Percentage</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>BoroName</th>\n",
       "      <th>subset</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th rowspan=\"3\" valign=\"top\">Bronx</th>\n",
       "      <th>attached_building</th>\n",
       "      <td>3275</td>\n",
       "      <td>87.637142</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>attached_street</th>\n",
       "      <td>456</td>\n",
       "      <td>12.202301</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>attached_unknown</th>\n",
       "      <td>6</td>\n",
       "      <td>0.160557</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th rowspan=\"3\" valign=\"top\">Brooklyn</th>\n",
       "      <th>attached_building</th>\n",
       "      <td>8330</td>\n",
       "      <td>90.229636</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>attached_street</th>\n",
       "      <td>871</td>\n",
       "      <td>9.434575</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>attached_unknown</th>\n",
       "      <td>31</td>\n",
       "      <td>0.335789</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th rowspan=\"3\" valign=\"top\">Manhattan</th>\n",
       "      <th>attached_building</th>\n",
       "      <td>3213</td>\n",
       "      <td>81.382979</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>attached_street</th>\n",
       "      <td>724</td>\n",
       "      <td>18.338399</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>attached_unknown</th>\n",
       "      <td>11</td>\n",
       "      <td>0.278622</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th rowspan=\"3\" valign=\"top\">Queens</th>\n",
       "      <th>attached_building</th>\n",
       "      <td>6459</td>\n",
       "      <td>85.211082</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>attached_street</th>\n",
       "      <td>1088</td>\n",
       "      <td>14.353562</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>attached_unknown</th>\n",
       "      <td>33</td>\n",
       "      <td>0.435356</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th rowspan=\"3\" valign=\"top\">Staten Island</th>\n",
       "      <th>attached_building</th>\n",
       "      <td>852</td>\n",
       "      <td>82.318841</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>attached_street</th>\n",
       "      <td>178</td>\n",
       "      <td>17.198068</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>attached_unknown</th>\n",
       "      <td>5</td>\n",
       "      <td>0.483092</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                                 count  Percentage\n",
       "BoroName      subset                              \n",
       "Bronx         attached_building   3275   87.637142\n",
       "              attached_street      456   12.202301\n",
       "              attached_unknown       6    0.160557\n",
       "Brooklyn      attached_building   8330   90.229636\n",
       "              attached_street      871    9.434575\n",
       "              attached_unknown      31    0.335789\n",
       "Manhattan     attached_building   3213   81.382979\n",
       "              attached_street      724   18.338399\n",
       "              attached_unknown      11    0.278622\n",
       "Queens        attached_building   6459   85.211082\n",
       "              attached_street     1088   14.353562\n",
       "              attached_unknown      33    0.435356\n",
       "Staten Island attached_building    852   82.318841\n",
       "              attached_street      178   17.198068\n",
       "              attached_unknown       5    0.483092"
      ]
     },
     "execution_count": 12,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "by_borough[\"Percentage\"] = by_borough.groupby(level=0).apply(lambda x: 100*x / x.sum())\n",
    "by_borough"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "id": "30811966",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "BoroName\n",
       "Bronx             456\n",
       "Brooklyn          871\n",
       "Manhattan         724\n",
       "Queens           1088\n",
       "Staten Island     178\n",
       "Name: count, dtype: int64"
      ]
     },
     "execution_count": 13,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "public_per_borough = by_borough.xs(\"attached_street\", level=\"subset\")[\"count\"]\n",
    "public_per_borough"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "id": "b3952d37",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "3317"
      ]
     },
     "execution_count": 14,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "public_per_borough.sum()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "id": "82a81006",
   "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>count</th>\n",
       "      <th>Percentage</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>BoroName</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>Bronx</th>\n",
       "      <td>3737</td>\n",
       "      <td>100.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Brooklyn</th>\n",
       "      <td>9232</td>\n",
       "      <td>100.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Manhattan</th>\n",
       "      <td>3948</td>\n",
       "      <td>100.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Queens</th>\n",
       "      <td>7580</td>\n",
       "      <td>100.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Staten Island</th>\n",
       "      <td>1035</td>\n",
       "      <td>100.0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "               count  Percentage\n",
       "BoroName                        \n",
       "Bronx           3737       100.0\n",
       "Brooklyn        9232       100.0\n",
       "Manhattan       3948       100.0\n",
       "Queens          7580       100.0\n",
       "Staten Island   1035       100.0"
      ]
     },
     "execution_count": 15,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "total_per_borough = by_borough.groupby(level=0).sum()\n",
    "total_per_borough"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "id": "9f59146b",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "count         25532.0\n",
       "Percentage      500.0\n",
       "dtype: float64"
      ]
     },
     "execution_count": 16,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "total_per_borough.sum()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "69e9aaa5",
   "metadata": {},
   "source": [
    "Number of intersections per borough:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "id": "a321d8d8",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "BoroName\n",
       "Bronx             6257\n",
       "Brooklyn         10384\n",
       "Manhattan         3934\n",
       "Queens           16344\n",
       "Staten Island     6474\n",
       "Name: PanoramaId, dtype: int64"
      ]
     },
     "execution_count": 17,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "intersections_per_borough = final_long.groupby(\"BoroName\")[\"PanoramaId\"].nunique()\n",
    "intersections_per_borough"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "id": "e7aa8a96",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "BoroName\n",
       "Bronx            0.597251\n",
       "Brooklyn         0.889060\n",
       "Manhattan        1.003559\n",
       "Queens           0.463779\n",
       "Staten Island    0.159870\n",
       "dtype: float64"
      ]
     },
     "execution_count": 18,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "cameras_per_intersection = total_per_borough[\"count\"] / intersections_per_borough\n",
    "cameras_per_intersection"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "id": "0bef465e",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "3.1135188970648806"
      ]
     },
     "execution_count": 19,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "cameras_per_intersection.sum()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "id": "5ae3b655",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "BoroName\n",
       "Bronx            0.072878\n",
       "Brooklyn         0.083879\n",
       "Manhattan        0.184037\n",
       "Queens           0.066569\n",
       "Staten Island    0.027495\n",
       "dtype: float64"
      ]
     },
     "execution_count": 20,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "public_per_intersection = public_per_borough / intersections_per_borough\n",
    "public_per_intersection"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "id": "b43a7ab5",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.43485739004243457"
      ]
     },
     "execution_count": 21,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "public_per_intersection.sum()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "a573c40a",
   "metadata": {},
   "source": [
    "## Focus on Intersection"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "id": "8e99c4a8",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "43406"
      ]
     },
     "execution_count": 22,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "final_long.PanoramaId.nunique()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "id": "4f32015b",
   "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>PanoramaId</th>\n",
       "      <th>BoroName</th>\n",
       "      <th>subset</th>\n",
       "      <th>count</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>--48RnFczF55WcDFpy5y_Q</td>\n",
       "      <td>Bronx</td>\n",
       "      <td>n_cameras</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>--T2iuvfxXJrzbgD7R3W-g</td>\n",
       "      <td>Queens</td>\n",
       "      <td>n_cameras</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>--VGAsqPgTXxgZFXuhuhbg</td>\n",
       "      <td>Queens</td>\n",
       "      <td>n_cameras</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>--YJY0GGvay_3W_O5ObeNQ</td>\n",
       "      <td>Staten Island</td>\n",
       "      <td>n_cameras</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>--fUkaf5nn1-bm4KiJ9pVA</td>\n",
       "      <td>Staten Island</td>\n",
       "      <td>n_cameras</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>303837</th>\n",
       "      <td>zzj23G_f-_hZa1LrXJ7tig</td>\n",
       "      <td>Bronx</td>\n",
       "      <td>type_unknown</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>303838</th>\n",
       "      <td>zzpZoM4JHrGNywYitkYT8w</td>\n",
       "      <td>Queens</td>\n",
       "      <td>type_unknown</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>303839</th>\n",
       "      <td>zztPjLBFQjwEISsA75a-zg</td>\n",
       "      <td>Brooklyn</td>\n",
       "      <td>type_unknown</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>303840</th>\n",
       "      <td>zzv84J0v5tQobN5uCXO1xw</td>\n",
       "      <td>Queens</td>\n",
       "      <td>type_unknown</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>303841</th>\n",
       "      <td>zzyhPWndbm7wxqTN6LuYLA</td>\n",
       "      <td>Queens</td>\n",
       "      <td>type_unknown</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>303842 rows × 4 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "                    PanoramaId       BoroName        subset  count\n",
       "0       --48RnFczF55WcDFpy5y_Q          Bronx     n_cameras      0\n",
       "1       --T2iuvfxXJrzbgD7R3W-g         Queens     n_cameras      0\n",
       "2       --VGAsqPgTXxgZFXuhuhbg         Queens     n_cameras      0\n",
       "3       --YJY0GGvay_3W_O5ObeNQ  Staten Island     n_cameras      0\n",
       "4       --fUkaf5nn1-bm4KiJ9pVA  Staten Island     n_cameras      1\n",
       "...                        ...            ...           ...    ...\n",
       "303837  zzj23G_f-_hZa1LrXJ7tig          Bronx  type_unknown      0\n",
       "303838  zzpZoM4JHrGNywYitkYT8w         Queens  type_unknown      0\n",
       "303839  zztPjLBFQjwEISsA75a-zg       Brooklyn  type_unknown      0\n",
       "303840  zzv84J0v5tQobN5uCXO1xw         Queens  type_unknown      0\n",
       "303841  zzyhPWndbm7wxqTN6LuYLA         Queens  type_unknown      0\n",
       "\n",
       "[303842 rows x 4 columns]"
      ]
     },
     "execution_count": 23,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "final_long"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "id": "c160d978",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "14100"
      ]
     },
     "execution_count": 24,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "sum(final_long[final_long[\"subset\"] == \"n_cameras\"][\"count\"] > 0)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "id": "836f8505",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "2442"
      ]
     },
     "execution_count": 25,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "sum(final_long[final_long[\"subset\"] == \"attached_street\"][\"count\"] > 0)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "id": "947944a3",
   "metadata": {
    "tags": []
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "1614"
      ]
     },
     "execution_count": 26,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "sum(final_long[final_long[\"subset\"] == \"type_dome\"][\"count\"] > 0)"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3 (ipykernel)",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.10.1"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}