{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Evolution of urban patterns: urban morphology as an open reproducible data science\n",
    "\n",
    "## Summary statistics\n",
    "\n",
    "This is the second notebook in a sequence of three. The notebook summarise morphometric data obtained in the previous notebook on the basis of historical periods.\n",
    "\n",
    "It requires `data/case_studies.csv` input with origins of case studies and data generated by the first notebook.\n",
    "\n",
    "Date: May 17, 2021\n",
    "\n",
    "---\n",
    "\n",
    "We start with the import of libraries used in this notebook."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Matplotlib created a temporary config/cache directory at /tmp/matplotlib-x7v3hb4j because the default path (/home/jovyan/.config/matplotlib) is not a writable directory; it is highly recommended to set the MPLCONFIGDIR environment variable to a writable directory, in particular to speed up the import of Matplotlib and to better support multiprocessing.\n"
     ]
    }
   ],
   "source": [
    "import pandas as pd\n",
    "import seaborn as sns\n",
    "from shapely.geometry import Point\n",
    "import geopandas as gpd\n",
    "import matplotlib.pyplot as plt\n",
    "import glob\n",
    "import pathlib\n",
    "from palettable.wesanderson import Moonrise5_6"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Using seaborn we can specify global settings for all matplotlib-based plots."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [],
   "source": [
    "sns.set()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "We will need the original input of our case studies as we needed in the first notebook."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>case</th>\n",
       "      <th>period</th>\n",
       "      <th>origin</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>Athens</td>\n",
       "      <td>pre-industrial</td>\n",
       "      <td>(23.729297894645065, 37.977742321097296)</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>Brugge</td>\n",
       "      <td>pre-industrial</td>\n",
       "      <td>(3.222135599564842, 51.20663413385126)</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>Havana</td>\n",
       "      <td>pre-industrial</td>\n",
       "      <td>(-82.35387869733259, 23.137058811383366)</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>Kyoto</td>\n",
       "      <td>pre-industrial</td>\n",
       "      <td>(135.77513789847612, 35.00441991880096)</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>Nuremberg</td>\n",
       "      <td>pre-industrial</td>\n",
       "      <td>(11.080741895925831, 49.4558122318182)</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>Pavia</td>\n",
       "      <td>pre-industrial</td>\n",
       "      <td>(9.155202505665285, 45.18542787308676)</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>Recife</td>\n",
       "      <td>pre-industrial</td>\n",
       "      <td>(-34.87950134898907, -8.066819977715335)</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>Barcelona</td>\n",
       "      <td>industrial</td>\n",
       "      <td>(2.1599658311051333, 41.39207228345335)</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8</th>\n",
       "      <td>Brisbane</td>\n",
       "      <td>industrial</td>\n",
       "      <td>(153.00890689998246, -27.483072513877765)</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9</th>\n",
       "      <td>Buenos Aires</td>\n",
       "      <td>industrial</td>\n",
       "      <td>(-58.37436109734774, -34.618526518125655)</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>10</th>\n",
       "      <td>Chicago</td>\n",
       "      <td>industrial</td>\n",
       "      <td>(-87.6496541987247, 41.91601382423912)</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>11</th>\n",
       "      <td>De Pijp</td>\n",
       "      <td>industrial</td>\n",
       "      <td>(4.890787, 52.353532)</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>12</th>\n",
       "      <td>Paris</td>\n",
       "      <td>industrial</td>\n",
       "      <td>(2.3036982060349636, 48.86864235146261)</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>13</th>\n",
       "      <td>Philadelphia</td>\n",
       "      <td>industrial</td>\n",
       "      <td>(-75.17713599972483, 39.92784202217783)</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>14</th>\n",
       "      <td>Akademgorodok</td>\n",
       "      <td>modernist</td>\n",
       "      <td>(83.10742859587872, 54.83929583879922)</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>15</th>\n",
       "      <td>Brasilia</td>\n",
       "      <td>modernist</td>\n",
       "      <td>(-47.91657209425542, -15.823763507567802)</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>16</th>\n",
       "      <td>Drumul Taberei</td>\n",
       "      <td>modernist</td>\n",
       "      <td>(26.02122171749766, 44.41721907853648)</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>17</th>\n",
       "      <td>Kilamba</td>\n",
       "      <td>modernist</td>\n",
       "      <td>(13.273496897999385, -8.998426816984628)</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>18</th>\n",
       "      <td>Marzhan</td>\n",
       "      <td>modernist</td>\n",
       "      <td>(13.558283202822224, 52.55398573559633)</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>19</th>\n",
       "      <td>Tblisi</td>\n",
       "      <td>modernist</td>\n",
       "      <td>(44.72275760065603, 41.72302532410621)</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>20</th>\n",
       "      <td>Karvina</td>\n",
       "      <td>modernist</td>\n",
       "      <td>(18.557418, 49.853861)</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>21</th>\n",
       "      <td>Ciudad Guayana</td>\n",
       "      <td>garden city</td>\n",
       "      <td>(-62.67209868783639, 8.34826427219966)</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>22</th>\n",
       "      <td>Frohnau</td>\n",
       "      <td>garden city</td>\n",
       "      <td>(13.277062904022877, 52.631536235815815)</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>23</th>\n",
       "      <td>Hilversum</td>\n",
       "      <td>garden city</td>\n",
       "      <td>(5.170650196333054, 52.23187803549469)</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>24</th>\n",
       "      <td>Montreal</td>\n",
       "      <td>garden city</td>\n",
       "      <td>(-73.64192800272671, 45.522340227287295)</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>25</th>\n",
       "      <td>Greenhills</td>\n",
       "      <td>garden city</td>\n",
       "      <td>(-87.996019, 42.941374)</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>26</th>\n",
       "      <td>Stratton</td>\n",
       "      <td>garden city</td>\n",
       "      <td>(116.04114530164892, -31.864639316563675)</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>27</th>\n",
       "      <td>Tel Aviv</td>\n",
       "      <td>garden city</td>\n",
       "      <td>(34.977140396112986, 32.0925043168361)</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>28</th>\n",
       "      <td>Auckland</td>\n",
       "      <td>neo-traditional</td>\n",
       "      <td>(174.9160552559242, -36.95408615897924)</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>29</th>\n",
       "      <td>Brandevoort</td>\n",
       "      <td>neo-traditional</td>\n",
       "      <td>(5.62261399486988, 51.46170483425585)</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>30</th>\n",
       "      <td>Cornell Village</td>\n",
       "      <td>neo-traditional</td>\n",
       "      <td>(-79.23037000333382, 43.88964002638408)</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>31</th>\n",
       "      <td>Kentlands</td>\n",
       "      <td>neo-traditional</td>\n",
       "      <td>(-77.24424129756238, 39.120324272641994)</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>32</th>\n",
       "      <td>Mueller</td>\n",
       "      <td>neo-traditional</td>\n",
       "      <td>(-97.70314980255172, 30.292836815637383)</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>33</th>\n",
       "      <td>Miami Lakes</td>\n",
       "      <td>neo-traditional</td>\n",
       "      <td>(-80.31195, 25.905058)</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>34</th>\n",
       "      <td>Rieselfield</td>\n",
       "      <td>neo-traditional</td>\n",
       "      <td>(7.792109457726385, 47.99888843253804)</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>35</th>\n",
       "      <td>Abdurahmangazi</td>\n",
       "      <td>informal</td>\n",
       "      <td>(29.25333679557981, 40.96419132348267)</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>36</th>\n",
       "      <td>Bicentennaire</td>\n",
       "      <td>informal</td>\n",
       "      <td>(-72.35389200521153, 18.535389208950473)</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>37</th>\n",
       "      <td>Jardim Portal I e II</td>\n",
       "      <td>informal</td>\n",
       "      <td>(-46.57587250888991, -23.440949292392443)</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>38</th>\n",
       "      <td>Kricak</td>\n",
       "      <td>informal</td>\n",
       "      <td>(110.35677760126997, -7.773518503607216)</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>39</th>\n",
       "      <td>Mafalala</td>\n",
       "      <td>informal</td>\n",
       "      <td>(32.572846900841455, -25.952561712972226)</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>40</th>\n",
       "      <td>Tandale</td>\n",
       "      <td>informal</td>\n",
       "      <td>(39.24312098621779, -6.792734470840269)</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>41</th>\n",
       "      <td>Tondo</td>\n",
       "      <td>informal</td>\n",
       "      <td>(120.9641471044341, 14.616058307179102)</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                    case           period  \\\n",
       "0                 Athens   pre-industrial   \n",
       "1                 Brugge   pre-industrial   \n",
       "2                 Havana   pre-industrial   \n",
       "3                  Kyoto   pre-industrial   \n",
       "4              Nuremberg   pre-industrial   \n",
       "5                  Pavia   pre-industrial   \n",
       "6                 Recife   pre-industrial   \n",
       "7              Barcelona       industrial   \n",
       "8               Brisbane       industrial   \n",
       "9           Buenos Aires       industrial   \n",
       "10               Chicago       industrial   \n",
       "11               De Pijp       industrial   \n",
       "12                 Paris       industrial   \n",
       "13          Philadelphia       industrial   \n",
       "14         Akademgorodok        modernist   \n",
       "15              Brasilia        modernist   \n",
       "16        Drumul Taberei        modernist   \n",
       "17               Kilamba        modernist   \n",
       "18               Marzhan        modernist   \n",
       "19                Tblisi        modernist   \n",
       "20               Karvina        modernist   \n",
       "21        Ciudad Guayana      garden city   \n",
       "22               Frohnau      garden city   \n",
       "23             Hilversum      garden city   \n",
       "24              Montreal      garden city   \n",
       "25            Greenhills      garden city   \n",
       "26              Stratton      garden city   \n",
       "27              Tel Aviv      garden city   \n",
       "28              Auckland  neo-traditional   \n",
       "29           Brandevoort  neo-traditional   \n",
       "30       Cornell Village  neo-traditional   \n",
       "31             Kentlands  neo-traditional   \n",
       "32               Mueller  neo-traditional   \n",
       "33           Miami Lakes  neo-traditional   \n",
       "34           Rieselfield  neo-traditional   \n",
       "35        Abdurahmangazi         informal   \n",
       "36         Bicentennaire         informal   \n",
       "37  Jardim Portal I e II         informal   \n",
       "38                Kricak         informal   \n",
       "39              Mafalala         informal   \n",
       "40               Tandale         informal   \n",
       "41                 Tondo         informal   \n",
       "\n",
       "                                       origin  \n",
       "0    (23.729297894645065, 37.977742321097296)  \n",
       "1      (3.222135599564842, 51.20663413385126)  \n",
       "2    (-82.35387869733259, 23.137058811383366)  \n",
       "3     (135.77513789847612, 35.00441991880096)  \n",
       "4      (11.080741895925831, 49.4558122318182)  \n",
       "5      (9.155202505665285, 45.18542787308676)  \n",
       "6    (-34.87950134898907, -8.066819977715335)  \n",
       "7     (2.1599658311051333, 41.39207228345335)  \n",
       "8   (153.00890689998246, -27.483072513877765)  \n",
       "9   (-58.37436109734774, -34.618526518125655)  \n",
       "10     (-87.6496541987247, 41.91601382423912)  \n",
       "11                      (4.890787, 52.353532)  \n",
       "12    (2.3036982060349636, 48.86864235146261)  \n",
       "13    (-75.17713599972483, 39.92784202217783)  \n",
       "14     (83.10742859587872, 54.83929583879922)  \n",
       "15  (-47.91657209425542, -15.823763507567802)  \n",
       "16     (26.02122171749766, 44.41721907853648)  \n",
       "17   (13.273496897999385, -8.998426816984628)  \n",
       "18    (13.558283202822224, 52.55398573559633)  \n",
       "19     (44.72275760065603, 41.72302532410621)  \n",
       "20                     (18.557418, 49.853861)  \n",
       "21     (-62.67209868783639, 8.34826427219966)  \n",
       "22   (13.277062904022877, 52.631536235815815)  \n",
       "23     (5.170650196333054, 52.23187803549469)  \n",
       "24   (-73.64192800272671, 45.522340227287295)  \n",
       "25                    (-87.996019, 42.941374)  \n",
       "26  (116.04114530164892, -31.864639316563675)  \n",
       "27     (34.977140396112986, 32.0925043168361)  \n",
       "28    (174.9160552559242, -36.95408615897924)  \n",
       "29      (5.62261399486988, 51.46170483425585)  \n",
       "30    (-79.23037000333382, 43.88964002638408)  \n",
       "31   (-77.24424129756238, 39.120324272641994)  \n",
       "32   (-97.70314980255172, 30.292836815637383)  \n",
       "33                     (-80.31195, 25.905058)  \n",
       "34     (7.792109457726385, 47.99888843253804)  \n",
       "35     (29.25333679557981, 40.96419132348267)  \n",
       "36   (-72.35389200521153, 18.535389208950473)  \n",
       "37  (-46.57587250888991, -23.440949292392443)  \n",
       "38   (110.35677760126997, -7.773518503607216)  \n",
       "39  (32.572846900841455, -25.952561712972226)  \n",
       "40    (39.24312098621779, -6.792734470840269)  \n",
       "41    (120.9641471044341, 14.616058307179102)  "
      ]
     },
     "execution_count": 3,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "cases = pd.read_csv(\"data/case_studies.csv\")\n",
    "cases"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "It will be easier to use `case` column as an index."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [],
   "source": [
    "cases = cases.set_index(\"case\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Using `glob` library, we get a list of all GeoPackages created in the first notebook."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [],
   "source": [
    "files = glob.glob(\"data/*gpkg\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "We need to generate a single DataFrame with all the results of all case studies combined. Let's prepare an empty DataFrame for that."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [],
   "source": [
    "data = pd.DataFrame()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Now we loop through the list of GeoPackages and:\n",
    "1. Read the file.\n",
    "2. Select only those geometries within 400 m buffer around the origin. That way we minimise potential edge effect affecting the results.\n",
    "3. Store period and case name within the DataFrame and append it together with morphometric data from GPKG to our `data` DataFrame."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/opt/conda/lib/python3.8/site-packages/geopandas/geodataframe.py:422: RuntimeWarning: Sequential read of iterator was interrupted. Resetting iterator. This can negatively impact the performance.\n",
      "  for feature in features_lst:\n"
     ]
    }
   ],
   "source": [
    "for f in files:\n",
    "    tessellation = gpd.read_file(f, layer=\"tessellation\")\n",
    "    case = pathlib.Path(f).stem\n",
    "    coords = cases.origin.loc[case]\n",
    "    buffer = gpd.GeoSeries([Point(tuple(map(float, coords[1:-1].split(', '))))], crs=4326).to_crs(tessellation.crs).buffer(400)\n",
    "    tessellation['period'] = cases.period.loc[case]\n",
    "    tessellation['case'] = case\n",
    "    casegdf = tessellation[tessellation.centroid.within(buffer.iloc[0])]\n",
    "    data = data.append(casegdf.drop(columns=[\"uID\", \"nID\", \"mm_len\", \"node_start\", \"node_end\", \"nodeID\", \"geometry\"]))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Since `period` is an ordered categorical column, let's encode it as such."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [],
   "source": [
    "data['period'] = pd.Categorical(data['period'], categories=['pre-industrial', 'industrial', 'garden city', 'modernist', 'neo-traditional', 'informal'], ordered=True)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Now we can sort the `data` based on the `period` in a way which follows the time."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [],
   "source": [
    "data = data.sort_values('period').reset_index(drop=True)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "For our figures, we want `paper` context, `whitegrid` style and `palettable`'s `Moonrise5_6` color palette."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [],
   "source": [
    "sns.set_context(\"paper\")\n",
    "sns.set_style(\"whitegrid\")\n",
    "sns.set_palette(Moonrise5_6.hex_colors)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "It is better to rename final labels to nicers ones, so let's prepare a dictionary for that."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [],
   "source": [
    "labels = {\n",
    "    \"cell_area\": \"Area of a tessellation cell [m]\",\n",
    "    \"car\": \"Covered area ratio\",\n",
    "    \"blg_area\": \"Area of a building footprint [m]\",\n",
    "    \"wall\": \"Length of a perimeter wall [m]\",\n",
    "    \"adjacency\": \"Building adjacency\",\n",
    "    \"neighbour_distance\": \"Mean neighbor distance between buildings [m]\",\n",
    "    \"length\": \"Length of a street segment [m]\",\n",
    "    \"linearity\": \"Linearity of a street segment\",\n",
    "    \"width\": \"Width of a street profile [m]\",\n",
    "    \"width_deviation\": \"Width deviation of a street profile [m]\",\n",
    "    \"openness\": \"Openness of a street profile\",\n",
    "    \"meshedness\": \"Meshedness of a street network\",\n",
    "}"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Now we loop through all morphometric characters and create box plots for each city, once with normal y axis and once with log y axis."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [],
   "source": [
    "for ch in data.columns.drop(['case', 'period']):\n",
    "    plt.figure(figsize=(12, 8))\n",
    "    boxen = sns.boxplot(x=\"case\", y=ch, hue='period', dodge=False, data=data, showfliers=False)\n",
    "    boxen.set_xticklabels(boxen.get_xticklabels(), rotation=90)\n",
    "    boxen.set_ylabel(labels[ch])\n",
    "    boxen.set_xlabel(\"Case study\")\n",
    "    sns.despine()\n",
    "    plt.legend(loc=\"lower left\", borderaxespad=0., ncol=6, frameon=False)\n",
    "    plt.savefig(f\"figures/{ch}_all_normal.png\", bbox_inches=\"tight\")\n",
    "    plt.close(\"all\")\n",
    "    \n",
    "    plt.figure(figsize=(12, 8))\n",
    "    boxen = sns.boxplot(x=\"case\", y=ch, hue='period', dodge=False, data=data, showfliers=False)\n",
    "    boxen.set_xticklabels(boxen.get_xticklabels(), rotation=90)\n",
    "    boxen.set_ylabel(labels[ch])\n",
    "    boxen.set_xlabel(\"Case study\")\n",
    "    boxen.set_yscale(\"log\")\n",
    "    sns.despine()\n",
    "    plt.legend(loc=\"lower left\", borderaxespad=0., ncol=6, frameon=False)\n",
    "    plt.savefig(f\"figures/{ch}_all_log.png\", bbox_inches=\"tight\")\n",
    "    plt.close(\"all\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "We can also create box plots for each character grouped by period."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [],
   "source": [
    "for ch in data.columns.drop(['case', 'period']):\n",
    "    plt.figure(figsize=(12, 8))\n",
    "    box = sns.boxplot(x=\"period\", y=ch,dodge=False, data=data, showfliers=False)\n",
    "    box.set_ylabel(labels[ch])\n",
    "    box.set_xlabel(\"Historical period\")\n",
    "    sns.despine()\n",
    "    plt.savefig(f\"figures/{ch}_grouped_normal.png\", bbox_inches=\"tight\")\n",
    "    plt.close(\"all\")\n",
    "    \n",
    "    plt.figure(figsize=(12, 8))\n",
    "    box = sns.boxplot(x=\"period\", y=ch,dodge=False, data=data, showfliers=False)\n",
    "    box.set_ylabel(labels[ch])\n",
    "    box.set_xlabel(\"Historical period\")\n",
    "    box.set_yscale(\"log\")\n",
    "    sns.despine()\n",
    "    plt.savefig(f\"figures/{ch}_grouped_log.png\", bbox_inches=\"tight\")\n",
    "    plt.close(\"all\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "To report median values per character and period, we can create a grouped DataFrame."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th>period</th>\n",
       "      <th>pre-industrial</th>\n",
       "      <th>industrial</th>\n",
       "      <th>garden city</th>\n",
       "      <th>modernist</th>\n",
       "      <th>neo-traditional</th>\n",
       "      <th>informal</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>Area of a tessellation cell [m]</th>\n",
       "      <td>177.03</td>\n",
       "      <td>171.40</td>\n",
       "      <td>589.16</td>\n",
       "      <td>3377.37</td>\n",
       "      <td>504.60</td>\n",
       "      <td>109.79</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Covered area ratio</th>\n",
       "      <td>0.62</td>\n",
       "      <td>0.53</td>\n",
       "      <td>0.22</td>\n",
       "      <td>0.16</td>\n",
       "      <td>0.27</td>\n",
       "      <td>0.51</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Area of a building footprint [m]</th>\n",
       "      <td>98.74</td>\n",
       "      <td>80.32</td>\n",
       "      <td>141.49</td>\n",
       "      <td>545.36</td>\n",
       "      <td>143.27</td>\n",
       "      <td>52.87</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Length of a perimeter wall [m]</th>\n",
       "      <td>282.85</td>\n",
       "      <td>335.94</td>\n",
       "      <td>60.26</td>\n",
       "      <td>169.43</td>\n",
       "      <td>68.33</td>\n",
       "      <td>32.48</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Building adjacency</th>\n",
       "      <td>0.20</td>\n",
       "      <td>0.20</td>\n",
       "      <td>0.84</td>\n",
       "      <td>0.92</td>\n",
       "      <td>1.00</td>\n",
       "      <td>1.00</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Mean neighbor distance between buildings [m]</th>\n",
       "      <td>3.44</td>\n",
       "      <td>6.56</td>\n",
       "      <td>14.67</td>\n",
       "      <td>32.56</td>\n",
       "      <td>12.94</td>\n",
       "      <td>3.00</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Length of a street segment [m]</th>\n",
       "      <td>91.81</td>\n",
       "      <td>136.51</td>\n",
       "      <td>154.63</td>\n",
       "      <td>196.23</td>\n",
       "      <td>131.60</td>\n",
       "      <td>133.96</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Linearity of a street segment</th>\n",
       "      <td>1.00</td>\n",
       "      <td>1.00</td>\n",
       "      <td>1.00</td>\n",
       "      <td>1.00</td>\n",
       "      <td>1.00</td>\n",
       "      <td>0.99</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Width of a street profile [m]</th>\n",
       "      <td>10.16</td>\n",
       "      <td>15.16</td>\n",
       "      <td>29.12</td>\n",
       "      <td>33.25</td>\n",
       "      <td>24.70</td>\n",
       "      <td>12.44</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Width deviation of a street profile [m]</th>\n",
       "      <td>3.16</td>\n",
       "      <td>2.31</td>\n",
       "      <td>3.59</td>\n",
       "      <td>3.30</td>\n",
       "      <td>2.87</td>\n",
       "      <td>4.75</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Openness of a street profile</th>\n",
       "      <td>0.15</td>\n",
       "      <td>0.18</td>\n",
       "      <td>0.38</td>\n",
       "      <td>0.61</td>\n",
       "      <td>0.35</td>\n",
       "      <td>0.13</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Meshedness of a street network</th>\n",
       "      <td>0.17</td>\n",
       "      <td>0.26</td>\n",
       "      <td>0.15</td>\n",
       "      <td>0.14</td>\n",
       "      <td>0.20</td>\n",
       "      <td>0.12</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "period                                        pre-industrial  industrial  \\\n",
       "Area of a tessellation cell [m]                       177.03      171.40   \n",
       "Covered area ratio                                      0.62        0.53   \n",
       "Area of a building footprint [m]                       98.74       80.32   \n",
       "Length of a perimeter wall [m]                        282.85      335.94   \n",
       "Building adjacency                                      0.20        0.20   \n",
       "Mean neighbor distance between buildings [m]            3.44        6.56   \n",
       "Length of a street segment [m]                         91.81      136.51   \n",
       "Linearity of a street segment                           1.00        1.00   \n",
       "Width of a street profile [m]                          10.16       15.16   \n",
       "Width deviation of a street profile [m]                 3.16        2.31   \n",
       "Openness of a street profile                            0.15        0.18   \n",
       "Meshedness of a street network                          0.17        0.26   \n",
       "\n",
       "period                                        garden city  modernist  \\\n",
       "Area of a tessellation cell [m]                    589.16    3377.37   \n",
       "Covered area ratio                                   0.22       0.16   \n",
       "Area of a building footprint [m]                   141.49     545.36   \n",
       "Length of a perimeter wall [m]                      60.26     169.43   \n",
       "Building adjacency                                   0.84       0.92   \n",
       "Mean neighbor distance between buildings [m]        14.67      32.56   \n",
       "Length of a street segment [m]                     154.63     196.23   \n",
       "Linearity of a street segment                        1.00       1.00   \n",
       "Width of a street profile [m]                       29.12      33.25   \n",
       "Width deviation of a street profile [m]              3.59       3.30   \n",
       "Openness of a street profile                         0.38       0.61   \n",
       "Meshedness of a street network                       0.15       0.14   \n",
       "\n",
       "period                                        neo-traditional  informal  \n",
       "Area of a tessellation cell [m]                        504.60    109.79  \n",
       "Covered area ratio                                       0.27      0.51  \n",
       "Area of a building footprint [m]                       143.27     52.87  \n",
       "Length of a perimeter wall [m]                          68.33     32.48  \n",
       "Building adjacency                                       1.00      1.00  \n",
       "Mean neighbor distance between buildings [m]            12.94      3.00  \n",
       "Length of a street segment [m]                         131.60    133.96  \n",
       "Linearity of a street segment                            1.00      0.99  \n",
       "Width of a street profile [m]                           24.70     12.44  \n",
       "Width deviation of a street profile [m]                  2.87      4.75  \n",
       "Openness of a street profile                             0.35      0.13  \n",
       "Meshedness of a street network                           0.20      0.12  "
      ]
     },
     "execution_count": 16,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "grouper = data.drop(columns='case').groupby('period')\n",
    "medians = grouper.median()\n",
    "medians = medians.rename(columns=labels)\n",
    "med = medians.T.round(2)\n",
    "med"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "It may also be useful to report interquartile range to illustrate the spread of each distribution."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "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>period</th>\n",
       "      <th>pre-industrial</th>\n",
       "      <th>industrial</th>\n",
       "      <th>garden city</th>\n",
       "      <th>modernist</th>\n",
       "      <th>neo-traditional</th>\n",
       "      <th>informal</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>Area of a tessellation cell [m]</th>\n",
       "      <td>262.39</td>\n",
       "      <td>276.77</td>\n",
       "      <td>642.93</td>\n",
       "      <td>3742.74</td>\n",
       "      <td>639.15</td>\n",
       "      <td>122.36</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Covered area ratio</th>\n",
       "      <td>0.28</td>\n",
       "      <td>0.20</td>\n",
       "      <td>0.15</td>\n",
       "      <td>0.10</td>\n",
       "      <td>0.17</td>\n",
       "      <td>0.27</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Area of a building footprint [m]</th>\n",
       "      <td>136.78</td>\n",
       "      <td>113.72</td>\n",
       "      <td>136.75</td>\n",
       "      <td>592.86</td>\n",
       "      <td>137.69</td>\n",
       "      <td>56.23</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Length of a perimeter wall [m]</th>\n",
       "      <td>351.53</td>\n",
       "      <td>399.06</td>\n",
       "      <td>42.09</td>\n",
       "      <td>101.07</td>\n",
       "      <td>39.54</td>\n",
       "      <td>22.24</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Building adjacency</th>\n",
       "      <td>0.29</td>\n",
       "      <td>0.20</td>\n",
       "      <td>0.34</td>\n",
       "      <td>0.42</td>\n",
       "      <td>0.44</td>\n",
       "      <td>0.05</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Mean neighbor distance between buildings [m]</th>\n",
       "      <td>3.57</td>\n",
       "      <td>4.40</td>\n",
       "      <td>10.45</td>\n",
       "      <td>19.40</td>\n",
       "      <td>9.54</td>\n",
       "      <td>3.32</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Length of a street segment [m]</th>\n",
       "      <td>71.42</td>\n",
       "      <td>51.10</td>\n",
       "      <td>121.79</td>\n",
       "      <td>169.47</td>\n",
       "      <td>100.27</td>\n",
       "      <td>124.83</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Linearity of a street segment</th>\n",
       "      <td>0.01</td>\n",
       "      <td>0.00</td>\n",
       "      <td>0.04</td>\n",
       "      <td>0.17</td>\n",
       "      <td>0.03</td>\n",
       "      <td>0.06</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Width of a street profile [m]</th>\n",
       "      <td>5.81</td>\n",
       "      <td>10.96</td>\n",
       "      <td>6.93</td>\n",
       "      <td>13.07</td>\n",
       "      <td>9.86</td>\n",
       "      <td>6.86</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Width deviation of a street profile [m]</th>\n",
       "      <td>2.44</td>\n",
       "      <td>1.88</td>\n",
       "      <td>1.88</td>\n",
       "      <td>2.82</td>\n",
       "      <td>1.99</td>\n",
       "      <td>1.74</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Openness of a street profile</th>\n",
       "      <td>0.13</td>\n",
       "      <td>0.13</td>\n",
       "      <td>0.23</td>\n",
       "      <td>0.22</td>\n",
       "      <td>0.23</td>\n",
       "      <td>0.14</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Meshedness of a street network</th>\n",
       "      <td>0.07</td>\n",
       "      <td>0.05</td>\n",
       "      <td>0.09</td>\n",
       "      <td>0.05</td>\n",
       "      <td>0.06</td>\n",
       "      <td>0.08</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "period                                        pre-industrial  industrial  \\\n",
       "Area of a tessellation cell [m]                       262.39      276.77   \n",
       "Covered area ratio                                      0.28        0.20   \n",
       "Area of a building footprint [m]                      136.78      113.72   \n",
       "Length of a perimeter wall [m]                        351.53      399.06   \n",
       "Building adjacency                                      0.29        0.20   \n",
       "Mean neighbor distance between buildings [m]            3.57        4.40   \n",
       "Length of a street segment [m]                         71.42       51.10   \n",
       "Linearity of a street segment                           0.01        0.00   \n",
       "Width of a street profile [m]                           5.81       10.96   \n",
       "Width deviation of a street profile [m]                 2.44        1.88   \n",
       "Openness of a street profile                            0.13        0.13   \n",
       "Meshedness of a street network                          0.07        0.05   \n",
       "\n",
       "period                                        garden city  modernist  \\\n",
       "Area of a tessellation cell [m]                    642.93    3742.74   \n",
       "Covered area ratio                                   0.15       0.10   \n",
       "Area of a building footprint [m]                   136.75     592.86   \n",
       "Length of a perimeter wall [m]                      42.09     101.07   \n",
       "Building adjacency                                   0.34       0.42   \n",
       "Mean neighbor distance between buildings [m]        10.45      19.40   \n",
       "Length of a street segment [m]                     121.79     169.47   \n",
       "Linearity of a street segment                        0.04       0.17   \n",
       "Width of a street profile [m]                        6.93      13.07   \n",
       "Width deviation of a street profile [m]              1.88       2.82   \n",
       "Openness of a street profile                         0.23       0.22   \n",
       "Meshedness of a street network                       0.09       0.05   \n",
       "\n",
       "period                                        neo-traditional  informal  \n",
       "Area of a tessellation cell [m]                        639.15    122.36  \n",
       "Covered area ratio                                       0.17      0.27  \n",
       "Area of a building footprint [m]                       137.69     56.23  \n",
       "Length of a perimeter wall [m]                          39.54     22.24  \n",
       "Building adjacency                                       0.44      0.05  \n",
       "Mean neighbor distance between buildings [m]             9.54      3.32  \n",
       "Length of a street segment [m]                         100.27    124.83  \n",
       "Linearity of a street segment                            0.03      0.06  \n",
       "Width of a street profile [m]                            9.86      6.86  \n",
       "Width deviation of a street profile [m]                  1.99      1.74  \n",
       "Openness of a street profile                             0.23      0.14  \n",
       "Meshedness of a street network                           0.06      0.08  "
      ]
     },
     "execution_count": 17,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "iqr = (grouper.quantile(.75).values - grouper.quantile(.25))\n",
    "iqr = iqr.rename(columns=labels)\n",
    "iq = iqr.T.round(2)\n",
    "iq"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "We want to report both in the same table."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "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>period</th>\n",
       "      <th>pre-industrial</th>\n",
       "      <th>industrial</th>\n",
       "      <th>garden city</th>\n",
       "      <th>modernist</th>\n",
       "      <th>neo-traditional</th>\n",
       "      <th>informal</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>Area of a tessellation cell [m]</th>\n",
       "      <td>177.03 (262.39)</td>\n",
       "      <td>171.4 (276.77)</td>\n",
       "      <td>589.16 (642.93)</td>\n",
       "      <td>3377.37 (3742.74)</td>\n",
       "      <td>504.6 (639.15)</td>\n",
       "      <td>109.79 (122.36)</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Covered area ratio</th>\n",
       "      <td>0.62 (0.28)</td>\n",
       "      <td>0.53 (0.2)</td>\n",
       "      <td>0.22 (0.15)</td>\n",
       "      <td>0.16 (0.1)</td>\n",
       "      <td>0.27 (0.17)</td>\n",
       "      <td>0.51 (0.27)</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Area of a building footprint [m]</th>\n",
       "      <td>98.74 (136.78)</td>\n",
       "      <td>80.32 (113.72)</td>\n",
       "      <td>141.49 (136.75)</td>\n",
       "      <td>545.36 (592.86)</td>\n",
       "      <td>143.27 (137.69)</td>\n",
       "      <td>52.87 (56.23)</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Length of a perimeter wall [m]</th>\n",
       "      <td>282.85 (351.53)</td>\n",
       "      <td>335.94 (399.06)</td>\n",
       "      <td>60.26 (42.09)</td>\n",
       "      <td>169.43 (101.07)</td>\n",
       "      <td>68.33 (39.54)</td>\n",
       "      <td>32.48 (22.24)</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Building adjacency</th>\n",
       "      <td>0.2 (0.29)</td>\n",
       "      <td>0.2 (0.2)</td>\n",
       "      <td>0.84 (0.34)</td>\n",
       "      <td>0.92 (0.42)</td>\n",
       "      <td>1.0 (0.44)</td>\n",
       "      <td>1.0 (0.05)</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Mean neighbor distance between buildings [m]</th>\n",
       "      <td>3.44 (3.57)</td>\n",
       "      <td>6.56 (4.4)</td>\n",
       "      <td>14.67 (10.45)</td>\n",
       "      <td>32.56 (19.4)</td>\n",
       "      <td>12.94 (9.54)</td>\n",
       "      <td>3.0 (3.32)</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Length of a street segment [m]</th>\n",
       "      <td>91.81 (71.42)</td>\n",
       "      <td>136.51 (51.1)</td>\n",
       "      <td>154.63 (121.79)</td>\n",
       "      <td>196.23 (169.47)</td>\n",
       "      <td>131.6 (100.27)</td>\n",
       "      <td>133.96 (124.83)</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Linearity of a street segment</th>\n",
       "      <td>1.0 (0.01)</td>\n",
       "      <td>1.0 (0.0)</td>\n",
       "      <td>1.0 (0.04)</td>\n",
       "      <td>1.0 (0.17)</td>\n",
       "      <td>1.0 (0.03)</td>\n",
       "      <td>0.99 (0.06)</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Width of a street profile [m]</th>\n",
       "      <td>10.16 (5.81)</td>\n",
       "      <td>15.16 (10.96)</td>\n",
       "      <td>29.12 (6.93)</td>\n",
       "      <td>33.25 (13.07)</td>\n",
       "      <td>24.7 (9.86)</td>\n",
       "      <td>12.44 (6.86)</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Width deviation of a street profile [m]</th>\n",
       "      <td>3.16 (2.44)</td>\n",
       "      <td>2.31 (1.88)</td>\n",
       "      <td>3.59 (1.88)</td>\n",
       "      <td>3.3 (2.82)</td>\n",
       "      <td>2.87 (1.99)</td>\n",
       "      <td>4.75 (1.74)</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Openness of a street profile</th>\n",
       "      <td>0.15 (0.13)</td>\n",
       "      <td>0.18 (0.13)</td>\n",
       "      <td>0.38 (0.23)</td>\n",
       "      <td>0.61 (0.22)</td>\n",
       "      <td>0.35 (0.23)</td>\n",
       "      <td>0.13 (0.14)</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Meshedness of a street network</th>\n",
       "      <td>0.17 (0.07)</td>\n",
       "      <td>0.26 (0.05)</td>\n",
       "      <td>0.15 (0.09)</td>\n",
       "      <td>0.14 (0.05)</td>\n",
       "      <td>0.2 (0.06)</td>\n",
       "      <td>0.12 (0.08)</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "period                                         pre-industrial  \\\n",
       "Area of a tessellation cell [m]               177.03 (262.39)   \n",
       "Covered area ratio                                0.62 (0.28)   \n",
       "Area of a building footprint [m]               98.74 (136.78)   \n",
       "Length of a perimeter wall [m]                282.85 (351.53)   \n",
       "Building adjacency                                 0.2 (0.29)   \n",
       "Mean neighbor distance between buildings [m]      3.44 (3.57)   \n",
       "Length of a street segment [m]                  91.81 (71.42)   \n",
       "Linearity of a street segment                      1.0 (0.01)   \n",
       "Width of a street profile [m]                    10.16 (5.81)   \n",
       "Width deviation of a street profile [m]           3.16 (2.44)   \n",
       "Openness of a street profile                      0.15 (0.13)   \n",
       "Meshedness of a street network                    0.17 (0.07)   \n",
       "\n",
       "period                                             industrial  \\\n",
       "Area of a tessellation cell [m]                171.4 (276.77)   \n",
       "Covered area ratio                                 0.53 (0.2)   \n",
       "Area of a building footprint [m]               80.32 (113.72)   \n",
       "Length of a perimeter wall [m]                335.94 (399.06)   \n",
       "Building adjacency                                  0.2 (0.2)   \n",
       "Mean neighbor distance between buildings [m]       6.56 (4.4)   \n",
       "Length of a street segment [m]                  136.51 (51.1)   \n",
       "Linearity of a street segment                       1.0 (0.0)   \n",
       "Width of a street profile [m]                   15.16 (10.96)   \n",
       "Width deviation of a street profile [m]           2.31 (1.88)   \n",
       "Openness of a street profile                      0.18 (0.13)   \n",
       "Meshedness of a street network                    0.26 (0.05)   \n",
       "\n",
       "period                                            garden city  \\\n",
       "Area of a tessellation cell [m]               589.16 (642.93)   \n",
       "Covered area ratio                                0.22 (0.15)   \n",
       "Area of a building footprint [m]              141.49 (136.75)   \n",
       "Length of a perimeter wall [m]                  60.26 (42.09)   \n",
       "Building adjacency                                0.84 (0.34)   \n",
       "Mean neighbor distance between buildings [m]    14.67 (10.45)   \n",
       "Length of a street segment [m]                154.63 (121.79)   \n",
       "Linearity of a street segment                      1.0 (0.04)   \n",
       "Width of a street profile [m]                    29.12 (6.93)   \n",
       "Width deviation of a street profile [m]           3.59 (1.88)   \n",
       "Openness of a street profile                      0.38 (0.23)   \n",
       "Meshedness of a street network                    0.15 (0.09)   \n",
       "\n",
       "period                                                modernist  \\\n",
       "Area of a tessellation cell [m]               3377.37 (3742.74)   \n",
       "Covered area ratio                                   0.16 (0.1)   \n",
       "Area of a building footprint [m]                545.36 (592.86)   \n",
       "Length of a perimeter wall [m]                  169.43 (101.07)   \n",
       "Building adjacency                                  0.92 (0.42)   \n",
       "Mean neighbor distance between buildings [m]       32.56 (19.4)   \n",
       "Length of a street segment [m]                  196.23 (169.47)   \n",
       "Linearity of a street segment                        1.0 (0.17)   \n",
       "Width of a street profile [m]                     33.25 (13.07)   \n",
       "Width deviation of a street profile [m]              3.3 (2.82)   \n",
       "Openness of a street profile                        0.61 (0.22)   \n",
       "Meshedness of a street network                      0.14 (0.05)   \n",
       "\n",
       "period                                        neo-traditional         informal  \n",
       "Area of a tessellation cell [m]                504.6 (639.15)  109.79 (122.36)  \n",
       "Covered area ratio                                0.27 (0.17)      0.51 (0.27)  \n",
       "Area of a building footprint [m]              143.27 (137.69)    52.87 (56.23)  \n",
       "Length of a perimeter wall [m]                  68.33 (39.54)    32.48 (22.24)  \n",
       "Building adjacency                                 1.0 (0.44)       1.0 (0.05)  \n",
       "Mean neighbor distance between buildings [m]     12.94 (9.54)       3.0 (3.32)  \n",
       "Length of a street segment [m]                 131.6 (100.27)  133.96 (124.83)  \n",
       "Linearity of a street segment                      1.0 (0.03)      0.99 (0.06)  \n",
       "Width of a street profile [m]                     24.7 (9.86)     12.44 (6.86)  \n",
       "Width deviation of a street profile [m]           2.87 (1.99)      4.75 (1.74)  \n",
       "Openness of a street profile                      0.35 (0.23)      0.13 (0.14)  \n",
       "Meshedness of a street network                     0.2 (0.06)      0.12 (0.08)  "
      ]
     },
     "execution_count": 20,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "med.astype('str') + ' (' + iq.astype('str') + ')'"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Following cells pick individual boxplots (same we created above) and generate combined figures used in the final paper."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 67,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 576x504 with 3 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "fig = plt.figure(figsize=(8, 7), constrained_layout=True)\n",
    "gs = fig.add_gridspec(6, 2)\n",
    "top = fig.add_subplot(gs[0:4, :])\n",
    "left = fig.add_subplot(gs[4:, 0])\n",
    "right = fig.add_subplot(gs[4:, 1])\n",
    "\n",
    "sns.boxplot(x=\"case\", y='neighbour_distance', hue='period', dodge=False, data=data, showfliers=False, ax=top, linewidth=.75)\n",
    "top.set_xticklabels(top.get_xticklabels(), rotation=90)\n",
    "top.set_ylabel(labels['neighbour_distance'])\n",
    "top.set_xlabel(\"\")\n",
    "\n",
    "sns.boxplot(x=\"period\", y='cell_area',dodge=False, data=data, showfliers=False, ax=left, linewidth=.75)\n",
    "left.set_ylabel(labels['cell_area'])\n",
    "left.set_xlabel(\"Historical period\")\n",
    "left.set_xticklabels(left.get_xticklabels(), rotation=90)\n",
    "left.set_yscale(\"log\")\n",
    "\n",
    "sns.boxplot(x=\"period\", y='blg_area',dodge=False, data=data, showfliers=False, ax=right, linewidth=.75)\n",
    "right.set_ylabel(labels['blg_area'])\n",
    "right.set_xlabel(\"Historical period\")\n",
    "right.set_xticklabels(right.get_xticklabels(), rotation=90)\n",
    "right.set_yscale(\"log\")\n",
    "sns.despine()\n",
    "\n",
    "top.legend(loc=\"upper left\", borderaxespad=0., ncol=3, frameon=True)\n",
    "plt.savefig(f\"figures/results_1.png\", bbox_inches=\"tight\", dpi=300)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 66,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 576x504 with 3 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "fig = plt.figure(figsize=(8, 7), constrained_layout=True)\n",
    "gs = fig.add_gridspec(6, 2)\n",
    "top = fig.add_subplot(gs[0:4, :])\n",
    "left = fig.add_subplot(gs[4:, 0])\n",
    "right = fig.add_subplot(gs[4:, 1])\n",
    "\n",
    "sns.boxplot(x=\"case\", y='openness', hue='period', dodge=False, data=data, showfliers=False, ax=top, linewidth=.75)\n",
    "top.set_xticklabels(top.get_xticklabels(), rotation=90)\n",
    "top.set_ylabel(labels['openness'])\n",
    "top.set_xlabel(\"\")\n",
    "\n",
    "sns.boxplot(x=\"period\", y='car',dodge=False, data=data, showfliers=False, ax=left, linewidth=.75)\n",
    "left.set_ylabel(labels['car'])\n",
    "left.set_xlabel(\"Historical period\")\n",
    "left.set_xticklabels(left.get_xticklabels(), rotation=90)\n",
    "left.set_yscale(\"log\")\n",
    "\n",
    "sns.boxplot(x=\"period\", y='width',dodge=False, data=data, showfliers=False, ax=right, linewidth=.75)\n",
    "right.set_ylabel(labels['width'])\n",
    "right.set_xlabel(\"Historical period\")\n",
    "right.set_xticklabels(right.get_xticklabels(), rotation=90)\n",
    "right.set_yscale(\"log\")\n",
    "sns.despine()\n",
    "\n",
    "top.legend(loc=\"upper left\", borderaxespad=0., ncol=3, frameon=True)\n",
    "plt.savefig(f\"figures/results_2.png\", bbox_inches=\"tight\", dpi=300)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 72,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 576x216 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "fig = plt.figure(figsize=(8, 3), constrained_layout=True)\n",
    "gs = fig.add_gridspec(1, 2)\n",
    "left = fig.add_subplot(gs[0, 0])\n",
    "right = fig.add_subplot(gs[0, 1])\n",
    "\n",
    "sns.boxplot(x=\"period\", y='wall',dodge=False, data=data, showfliers=False, ax=left, linewidth=.75)\n",
    "left.set_ylabel(labels['wall'])\n",
    "left.set_xlabel(\"Historical period\")\n",
    "left.set_xticklabels(left.get_xticklabels(), rotation=90)\n",
    "\n",
    "sns.boxplot(x=\"period\", y='adjacency',dodge=False, data=data, showfliers=False, ax=right, linewidth=.75)\n",
    "right.set_ylabel(labels['adjacency'])\n",
    "right.set_xlabel(\"Historical period\")\n",
    "right.set_xticklabels(right.get_xticklabels(), rotation=90)\n",
    "sns.despine()\n",
    "\n",
    "plt.savefig(f\"figures/results_3.png\", bbox_inches=\"tight\", dpi=300)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 75,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 576x504 with 3 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "fig = plt.figure(figsize=(8, 7), constrained_layout=True)\n",
    "gs = fig.add_gridspec(6, 2)\n",
    "top = fig.add_subplot(gs[0:4, :])\n",
    "left = fig.add_subplot(gs[4:, 0])\n",
    "right = fig.add_subplot(gs[4:, 1])\n",
    "\n",
    "sns.boxplot(x=\"case\", y='meshedness', hue='period', dodge=False, data=data, showfliers=False, ax=top, linewidth=.75)\n",
    "top.set_xticklabels(top.get_xticklabels(), rotation=90)\n",
    "top.set_ylabel(labels['meshedness'])\n",
    "top.set_xlabel(\"\")\n",
    "\n",
    "sns.boxplot(x=\"period\", y='linearity',dodge=False, data=data, showfliers=False, ax=left, linewidth=.75)\n",
    "left.set_ylabel(labels['linearity'])\n",
    "left.set_xlabel(\"Historical period\")\n",
    "left.set_xticklabels(left.get_xticklabels(), rotation=90)\n",
    "\n",
    "sns.boxplot(x=\"period\", y='width_deviation',dodge=False, data=data, showfliers=False, ax=right, linewidth=.75)\n",
    "right.set_ylabel(labels['width_deviation'])\n",
    "right.set_xlabel(\"Historical period\")\n",
    "right.set_xticklabels(right.get_xticklabels(), rotation=90)\n",
    "sns.despine()\n",
    "\n",
    "top.legend(loc=\"upper right\", borderaxespad=0., ncol=3, frameon=True)\n",
    "plt.savefig(f\"figures/results_4.png\", bbox_inches=\"tight\", dpi=300)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Statistical comparison\n",
    "\n",
    "We also want to compare similarity of distributions statistically, not only visually. We use Kruskal–Wallis one-way analysis of variance and Pairwise Mann–Whitney *U* test for that."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [],
   "source": [
    "from scipy import stats\n",
    "\n",
    "data = data.set_index('case')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [],
   "source": [
    "data = data.rename(columns=labels)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Kruskal–Wallis one-way analysis of variance\n",
    "\n",
    "Kruskal–Wallis one-way analysis of variance is chosen as a non-parametric equivalent of ANOVA due to skewed distributions of underlying data."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {},
   "outputs": [],
   "source": [
    "kruskal = pd.DataFrame()\n",
    "for col in data.columns.drop(\"period\"):\n",
    "    H, p = stats.kruskal(*[data[data.period == per][col] for per in data.period.cat.categories])\n",
    "    kruskal.loc[col, 'H'] = H\n",
    "    kruskal.loc[col, 'p'] = p"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "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>H</th>\n",
       "      <th>p</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>Area of a tessellation cell [m]</th>\n",
       "      <td>16091.729256</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Covered area ratio</th>\n",
       "      <td>14840.457954</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Area of a building footprint [m]</th>\n",
       "      <td>10044.313353</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Length of a perimeter wall [m]</th>\n",
       "      <td>20996.802901</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Building adjacency</th>\n",
       "      <td>25942.003837</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Mean neighbor distance between buildings [m]</th>\n",
       "      <td>21449.270408</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Length of a street segment [m]</th>\n",
       "      <td>2870.986793</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Linearity of a street segment</th>\n",
       "      <td>10832.608655</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Width of a street profile [m]</th>\n",
       "      <td>20907.856962</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Width deviation of a street profile [m]</th>\n",
       "      <td>17022.101966</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Openness of a street profile</th>\n",
       "      <td>15489.762454</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Meshedness of a street network</th>\n",
       "      <td>22363.576671</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                                                         H    p\n",
       "Area of a tessellation cell [m]               16091.729256  0.0\n",
       "Covered area ratio                            14840.457954  0.0\n",
       "Area of a building footprint [m]              10044.313353  0.0\n",
       "Length of a perimeter wall [m]                20996.802901  0.0\n",
       "Building adjacency                            25942.003837  0.0\n",
       "Mean neighbor distance between buildings [m]  21449.270408  0.0\n",
       "Length of a street segment [m]                 2870.986793  0.0\n",
       "Linearity of a street segment                 10832.608655  0.0\n",
       "Width of a street profile [m]                 20907.856962  0.0\n",
       "Width deviation of a street profile [m]       17022.101966  0.0\n",
       "Openness of a street profile                  15489.762454  0.0\n",
       "Meshedness of a street network                22363.576671  0.0"
      ]
     },
     "execution_count": 15,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "kruskal"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "The results indicate that the distributions of morphometric values obtained from samples from different historical periods cannot be considered the same."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Pairwise Mann–Whitney _U_ test"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "metadata": {},
   "outputs": [],
   "source": [
    "from itertools import product\n",
    "\n",
    "mann_whitney_U = {}\n",
    "mann_whitney_p = {}\n",
    "for col in data.columns.drop(\"period\"):\n",
    "    df_u = pd.DataFrame(columns=data.period.cat.categories, index=data.period.cat.categories)\n",
    "    df_p = pd.DataFrame(columns=data.period.cat.categories, index=data.period.cat.categories)\n",
    "    for a, b in product(data.period.cat.categories, repeat=2):\n",
    "        if pd.isna(df_u.loc[b, a]):\n",
    "            U, p = stats.mannwhitneyu(data[data.period == a][col], data[data.period == b][col])\n",
    "            df_u.loc[a, b] = U\n",
    "            df_p.loc[a, b] = p\n",
    "        else:\n",
    "            df_u.loc[a, b] = df_u.loc[b, a]\n",
    "            df_p.loc[a, b] = df_p.loc[b, a]\n",
    "            \n",
    "    mann_whitney_U[col] = df_u\n",
    "    mann_whitney_p[col] = df_p"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Building adjacency\n",
      "Linearity of a street segment\n",
      "Width deviation of a street profile [m]\n"
     ]
    }
   ],
   "source": [
    "for col in data.columns.drop(\"period\"):\n",
    "    if not (mann_whitney_p[col] > .05).sum().sum() == 6:\n",
    "        print(f\"{col}\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "In three cases, `Building adjacency`, `Linearity of a street segment`, and `Width deviation of a street profile`, Mann–Whitney _U_ is not significant (`p < 0.05`) in all pairwise cases."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>pre-industrial</th>\n",
       "      <th>industrial</th>\n",
       "      <th>garden city</th>\n",
       "      <th>modernist</th>\n",
       "      <th>neo-traditional</th>\n",
       "      <th>informal</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>pre-industrial</th>\n",
       "      <td>True</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>industrial</th>\n",
       "      <td>False</td>\n",
       "      <td>True</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>garden city</th>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>True</td>\n",
       "      <td>True</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>modernist</th>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>True</td>\n",
       "      <td>True</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>neo-traditional</th>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>True</td>\n",
       "      <td>False</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>informal</th>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>True</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                 pre-industrial  industrial  garden city  modernist  \\\n",
       "pre-industrial             True       False        False      False   \n",
       "industrial                False        True        False      False   \n",
       "garden city               False       False         True       True   \n",
       "modernist                 False       False         True       True   \n",
       "neo-traditional           False       False        False      False   \n",
       "informal                  False       False        False      False   \n",
       "\n",
       "                 neo-traditional  informal  \n",
       "pre-industrial             False     False  \n",
       "industrial                 False     False  \n",
       "garden city                False     False  \n",
       "modernist                  False     False  \n",
       "neo-traditional             True     False  \n",
       "informal                   False      True  "
      ]
     },
     "execution_count": 28,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "mann_whitney_p['Building adjacency'] > .05"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "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>pre-industrial</th>\n",
       "      <th>industrial</th>\n",
       "      <th>garden city</th>\n",
       "      <th>modernist</th>\n",
       "      <th>neo-traditional</th>\n",
       "      <th>informal</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>pre-industrial</th>\n",
       "      <td>0.5</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>industrial</th>\n",
       "      <td>0.0</td>\n",
       "      <td>0.5</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>garden city</th>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.499998</td>\n",
       "      <td>0.404944</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>modernist</th>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.404944</td>\n",
       "      <td>0.499981</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>neo-traditional</th>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.499998</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>informal</th>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.5</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                pre-industrial industrial garden city modernist  \\\n",
       "pre-industrial             0.5        0.0         0.0       0.0   \n",
       "industrial                 0.0        0.5         0.0       0.0   \n",
       "garden city                0.0        0.0    0.499998  0.404944   \n",
       "modernist                  0.0        0.0    0.404944  0.499981   \n",
       "neo-traditional            0.0        0.0         0.0       0.0   \n",
       "informal                   0.0        0.0         0.0       0.0   \n",
       "\n",
       "                neo-traditional informal  \n",
       "pre-industrial              0.0      0.0  \n",
       "industrial                  0.0      0.0  \n",
       "garden city                 0.0      0.0  \n",
       "modernist                   0.0      0.0  \n",
       "neo-traditional        0.499998      0.0  \n",
       "informal                    0.0      0.5  "
      ]
     },
     "execution_count": 31,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "mann_whitney_p['Building adjacency'].round(3)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "In the case of `Building adjacency`, the distributions of values in Garden city and Modernist periods cannot be considered significantly different. "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "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>pre-industrial</th>\n",
       "      <th>industrial</th>\n",
       "      <th>garden city</th>\n",
       "      <th>modernist</th>\n",
       "      <th>neo-traditional</th>\n",
       "      <th>informal</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>pre-industrial</th>\n",
       "      <td>True</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>industrial</th>\n",
       "      <td>False</td>\n",
       "      <td>True</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>garden city</th>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>True</td>\n",
       "      <td>True</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>modernist</th>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>True</td>\n",
       "      <td>True</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>neo-traditional</th>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>True</td>\n",
       "      <td>False</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>informal</th>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>True</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                 pre-industrial  industrial  garden city  modernist  \\\n",
       "pre-industrial             True       False        False      False   \n",
       "industrial                False        True        False      False   \n",
       "garden city               False       False         True       True   \n",
       "modernist                 False       False         True       True   \n",
       "neo-traditional           False       False        False      False   \n",
       "informal                  False       False        False      False   \n",
       "\n",
       "                 neo-traditional  informal  \n",
       "pre-industrial             False     False  \n",
       "industrial                 False     False  \n",
       "garden city                False     False  \n",
       "modernist                  False     False  \n",
       "neo-traditional             True     False  \n",
       "informal                   False      True  "
      ]
     },
     "execution_count": 21,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "mann_whitney_p['Linearity of a street segment'] > .05"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "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>pre-industrial</th>\n",
       "      <th>industrial</th>\n",
       "      <th>garden city</th>\n",
       "      <th>modernist</th>\n",
       "      <th>neo-traditional</th>\n",
       "      <th>informal</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>pre-industrial</th>\n",
       "      <td>0.5</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>industrial</th>\n",
       "      <td>0.0</td>\n",
       "      <td>0.5</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>garden city</th>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.499998</td>\n",
       "      <td>0.153011</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>modernist</th>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.153011</td>\n",
       "      <td>0.499981</td>\n",
       "      <td>0.000139</td>\n",
       "      <td>0.000064</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>neo-traditional</th>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.000139</td>\n",
       "      <td>0.499998</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>informal</th>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.000064</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.5</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                pre-industrial industrial garden city modernist  \\\n",
       "pre-industrial             0.5        0.0         0.0       0.0   \n",
       "industrial                 0.0        0.5         0.0       0.0   \n",
       "garden city                0.0        0.0    0.499998  0.153011   \n",
       "modernist                  0.0        0.0    0.153011  0.499981   \n",
       "neo-traditional            0.0        0.0         0.0  0.000139   \n",
       "informal                   0.0        0.0         0.0  0.000064   \n",
       "\n",
       "                neo-traditional  informal  \n",
       "pre-industrial              0.0       0.0  \n",
       "industrial                  0.0       0.0  \n",
       "garden city                 0.0       0.0  \n",
       "modernist              0.000139  0.000064  \n",
       "neo-traditional        0.499998       0.0  \n",
       "informal                    0.0       0.5  "
      ]
     },
     "execution_count": 32,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "mann_whitney_p['Linearity of a street segment'].round(3)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "In the case of `Linearity of a street segment`, the distributions of values in Garden city and Modernist periods cannot be considered significantly different."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "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>pre-industrial</th>\n",
       "      <th>industrial</th>\n",
       "      <th>garden city</th>\n",
       "      <th>modernist</th>\n",
       "      <th>neo-traditional</th>\n",
       "      <th>informal</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>pre-industrial</th>\n",
       "      <td>True</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>True</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>industrial</th>\n",
       "      <td>False</td>\n",
       "      <td>True</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>garden city</th>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>True</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>modernist</th>\n",
       "      <td>True</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>True</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>neo-traditional</th>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>True</td>\n",
       "      <td>False</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>informal</th>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>True</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                 pre-industrial  industrial  garden city  modernist  \\\n",
       "pre-industrial             True       False        False       True   \n",
       "industrial                False        True        False      False   \n",
       "garden city               False       False         True      False   \n",
       "modernist                  True       False        False       True   \n",
       "neo-traditional           False       False        False      False   \n",
       "informal                  False       False        False      False   \n",
       "\n",
       "                 neo-traditional  informal  \n",
       "pre-industrial             False     False  \n",
       "industrial                 False     False  \n",
       "garden city                False     False  \n",
       "modernist                  False     False  \n",
       "neo-traditional             True     False  \n",
       "informal                   False      True  "
      ]
     },
     "execution_count": 22,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "mann_whitney_p['Width deviation of a street profile [m]'] > .05"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 33,
   "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>pre-industrial</th>\n",
       "      <th>industrial</th>\n",
       "      <th>garden city</th>\n",
       "      <th>modernist</th>\n",
       "      <th>neo-traditional</th>\n",
       "      <th>informal</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>pre-industrial</th>\n",
       "      <td>0.5</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.173361</td>\n",
       "      <td>0.002794</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>industrial</th>\n",
       "      <td>0.0</td>\n",
       "      <td>0.5</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>garden city</th>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.499998</td>\n",
       "      <td>0.000003</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>modernist</th>\n",
       "      <td>0.173361</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.000003</td>\n",
       "      <td>0.499981</td>\n",
       "      <td>0.01815</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>neo-traditional</th>\n",
       "      <td>0.002794</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.01815</td>\n",
       "      <td>0.499998</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>informal</th>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.5</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                pre-industrial industrial garden city modernist  \\\n",
       "pre-industrial             0.5        0.0         0.0  0.173361   \n",
       "industrial                 0.0        0.5         0.0       0.0   \n",
       "garden city                0.0        0.0    0.499998  0.000003   \n",
       "modernist             0.173361        0.0    0.000003  0.499981   \n",
       "neo-traditional       0.002794        0.0         0.0   0.01815   \n",
       "informal                   0.0        0.0         0.0       0.0   \n",
       "\n",
       "                neo-traditional informal  \n",
       "pre-industrial         0.002794      0.0  \n",
       "industrial                  0.0      0.0  \n",
       "garden city                 0.0      0.0  \n",
       "modernist               0.01815      0.0  \n",
       "neo-traditional        0.499998      0.0  \n",
       "informal                    0.0      0.5  "
      ]
     },
     "execution_count": 33,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "mann_whitney_p['Width deviation of a street profile [m]'].round(3)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "In the case of `Width deviation of a street profile`, the distributions of values in Pre-industrial and Modernist periods cannot be considered significantly different."
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.9.2"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 4
}