notebooks/Demo-analysing-scan-characteristics.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "id": "82f38ca9",
   "metadata": {},
   "source": [
    "# Analysing Characteristics of an entire PageXML Archive\n",
    "\n",
    "**Note**: This tutorial follows and depends on the tutorial on [Reading PageXML files from archives](./Demo-reading-pagexml-files-from-archive.ipynb). It assumes you have downloaded the PageXML archives and derived line format files from them. \n",
    "\n",
    "This tutorials shows how you can analyse the layout and text of sets of PageXML files.\n",
    "\n",
    "As an example of a zipped archive, this tutorial uses two datasets provided by the [National Archives of the Netherlands](https://www.nationaalarchief.nl/en) (NA) via their HTR repository on [Zenodo](https://zenodo.org/): https://zenodo.org/record/6414086#.Y8Elk-zMIUo. The repository contains many other HTR PageXML datasets that NA made available.\n",
    "\n",
    "The datasets contains HTR output in [PageXML](https://www.primaresearch.org/tools/PAGELibraries) format of scans from the following archive: \n",
    "- (small) _Notaris mr. D.A.M.de Fremery te Assen, 1899-1915, 114.11, 1_ ([EAD](https://www.drentsarchief.nl/onderzoeken/archiefstukken?mivast=34&mizig=210&miadt=34&micode=0114.11&miview=inv2)). This is an archive maintained by the [Drents Archief](https://www.drentsarchief.nl).\n",
    "- (medium) _Verspreide West-Indische stukken, 1614-1875, 1.05.06, 1-1413_ ([EAD](https://www.nationaalarchief.nl/onderzoeken/archief/1.05.06/invnr/%401?query=1.05.06&search-type=inventory)). This is an archive maintained by the [Nationaal Archief](https://www.nationaalarchief.nl/en). \n",
    "\n",
    "You can download the datasets via the following URLs:\n",
    "- https://zenodo.org/record/6414086/files/HTR%20results%20DA%200114.11%20PAGE.zip?download=1\n",
    "- https://zenodo.org/record/6414086/files/HTR%20results%201.05.06%20PAGE.zip?download=1\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "id": "28955c8e",
   "metadata": {},
   "outputs": [],
   "source": [
    "%reload_ext autoreload\n",
    "%autoreload 2\n"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "84d5e74d",
   "metadata": {},
   "source": [
    "## Extracting PageXML files from Zip/Tar files\n",
    "\n",
    "The example zip files contains of smaller zip files, that each have a number of PageXML files. "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "id": "ba14c6ba",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<generator object parse_pagexml_files_from_archive at 0x291485000>"
      ]
     },
     "execution_count": 18,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "import json\n",
    "\n",
    "import matplotlib.pyplot as plt\n",
    "import numpy as np\n",
    "import seaborn as sns\n",
    "\n",
    "from pagexml.analysis.stats import get_doc_stats\n",
    "from pagexml.helper.file_helper import read_page_archive_files\n",
    "from pagexml.parser import parse_pagexml_files_from_archive\n",
    "\n",
    "sns.set_theme()\n",
    "\n",
    "da_archive_file = '../data/HTR results DA 0114.11 PAGE.zip'\n",
    "na_archive_file = '../data/HTR results 1.05.06 PAGE.zip'\n",
    "\n",
    "parse_pagexml_files_from_archive(na_archive_file)\n"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "38158d3c",
   "metadata": {},
   "source": [
    "# Analysing documents based on layout\n",
    "\n",
    "## Categorising lines by their widths\n",
    "\n",
    "Lines can vary strongly in how much text they contain, from one or a few characters to a dozen or more words. How much text they contain and how large the characters are, are related to a line's width. \n",
    "\n",
    "The distribution of the width of text lines can tell us whether most lines are roughly of similar length, of whether there are different clusters of line widths. Based on these _line width_ clusters, we can cluster scans based on their distribution of line widths, which can help us identify different types of scans. \n",
    "\n",
    "We start with an analysis of the distribution of line widths. Scans often have a width of several thousand pixels, with individual characters taking up somewhere between 15 and 150 pixels. "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "id": "593d4e25",
   "metadata": {},
   "outputs": [],
   "source": [
    "sns.set_theme()\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "id": "9ed4a4f0",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Number of lines: 1007175\n",
      "CPU times: user 17.4 s, sys: 107 ms, total: 17.6 s\n",
      "Wall time: 17.6 s\n"
     ]
    }
   ],
   "source": [
    "%%time\n",
    "import seaborn as sns\n",
    "\n",
    "import pagexml.analysis.layout_stats as layout_stats\n",
    "from pagexml.parser import parse_pagexml_files_from_archive\n",
    "from pagexml.helper.text_helper import read_pagexml_docs_from_line_file\n",
    "\n",
    "sns.set_theme()\n",
    "\n",
    "da_line_file = '../data/line_format-NL-AsnDA_0114.11.tsv.gz'\n",
    "na_line_file = '../data/line_format-NL-HaNA_1.05.06.tsv.gz'\n",
    "\n",
    "# reading from line file is much faster \n",
    "# than reading from archived PageXML files\n",
    "scans = read_pagexml_docs_from_line_file(na_line_file, add_bounding_box=True)\n",
    "\n",
    "\n",
    "line_bin_size = 100\n",
    "line_widths = layout_stats.get_line_widths(scans, line_bin_size)\n",
    "print(f'Number of lines: {len(line_widths)}')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "id": "3c6e43be",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "sns.kdeplot(data=line_widths);"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "08dbda8a",
   "metadata": {},
   "source": [
    "We can use the local maxima as centroids of clusters of lines with similar width, and the local minima as boundary points between these clusters. That way, we can categorise the lines of a scan according to these clusters and get summary statistics per scan on the categories of line widths."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "id": "7a95fed7",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1152x432 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "from demo_scripts import plot_line_widths_boundary_points\n",
    "\n",
    "boundary_points = layout_stats.find_line_width_boundary_points(line_widths, line_bin_size)\n",
    "\n",
    "plot_line_widths_boundary_points(line_widths, boundary_points)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "829c5004",
   "metadata": {},
   "source": [
    "The distribution has several peaks and valleys. The valleys can be used as boundary points between categories of lines."
   ]
  },
  {
   "cell_type": "markdown",
   "id": "a21090d3",
   "metadata": {},
   "source": [
    "# Collecting document statistics\n",
    "\n",
    "It's simple to get some statistics on the text on a scan using the `.stats` property. But this is very minimal. "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "id": "bdae7519",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "NL-HaNA_1.05.06_1_0001.jpg {'lines': 33, 'words': 165, 'text_regions': 1, 'columns': 0, 'extra': 0, 'pages': 0}\n",
      "NL-HaNA_1.05.06_1_0002.jpg {'lines': 59, 'words': 286, 'text_regions': 2, 'columns': 0, 'extra': 0, 'pages': 0}\n",
      "NL-HaNA_1.05.06_1_0003.jpg {'lines': 54, 'words': 263, 'text_regions': 2, 'columns': 0, 'extra': 0, 'pages': 0}\n",
      "NL-HaNA_1.05.06_1_0004.jpg {'lines': 0, 'words': 0, 'text_regions': 0, 'columns': 0, 'extra': 0, 'pages': 0}\n",
      "NL-HaNA_1.05.06_1_0005.jpg {'lines': 0, 'words': 0, 'text_regions': 0, 'columns': 0, 'extra': 0, 'pages': 0}\n",
      "NL-HaNA_1.05.06_10_0001.jpg {'lines': 34, 'words': 254, 'text_regions': 2, 'columns': 0, 'extra': 0, 'pages': 0}\n"
     ]
    }
   ],
   "source": [
    "scans = parse_pagexml_files_from_archive(na_archive_file)\n",
    "\n",
    "for si, scan in enumerate(scans):\n",
    "    print(scan.id, scan.stats)\n",
    "    if (si+1) > 5:\n",
    "        break"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "52292b53",
   "metadata": {},
   "source": [
    "## Gathering descriptive statistics per scan\n",
    "\n",
    "The `pagexml.analysis.stats` module contains a function called `get_doc_stats` that gives a richer set of statistics. You can pass it a set of scans (as an iterable, so it can also be an iterable that reads scans from file) to calculate basic statistics per scan, such as:\n",
    "\n",
    "- the number of lines and words,\n",
    "- the nubmer of lines of different line width categories,\n",
    "- the number of words that start with an uppercase character,\n",
    "- the number of words that are numbers,\n",
    "- the number of words binned by their lengths.\n",
    "\n",
    "These allow simple analysis of sequences of scans and cluster of scans based on these characteristics. This is a low-resource approach to clustering and classifying documents. \n",
    "\n",
    "**Note**: don't use line-format files for gathering statistics per scan, because line-format files only capture scans that contain at least one text line. **In other words, empty scans are not in the line-format files, and therefore will not be represented in scan statistics!**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "id": "0f7d3d20",
   "metadata": {
    "scrolled": true
   },
   "outputs": [],
   "source": [
    "import pagexml.analysis.text_stats as text_stats\n",
    "\n",
    "scans = parse_pagexml_files_from_archive(na_archive_file)\n",
    "\n",
    "scans_stats = []\n",
    "\n",
    "scans_stats = get_doc_stats(scans, line_width_boundary_points=boundary_points)\n",
    "  "
   ]
  },
  {
   "cell_type": "markdown",
   "id": "fa888262",
   "metadata": {},
   "source": [
    "The statistics are best viewed and analysed using Pandas."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "id": "5f490256",
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "(15857, 39)\n"
     ]
    },
    {
     "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>doc_id</th>\n",
       "      <th>doc_num</th>\n",
       "      <th>doc_width</th>\n",
       "      <th>doc_height</th>\n",
       "      <th>lines</th>\n",
       "      <th>words</th>\n",
       "      <th>text_regions</th>\n",
       "      <th>columns</th>\n",
       "      <th>extra</th>\n",
       "      <th>pages</th>\n",
       "      <th>...</th>\n",
       "      <th>num_words_length_10</th>\n",
       "      <th>num_words_length_15</th>\n",
       "      <th>num_words_length_20</th>\n",
       "      <th>num_words_length_25</th>\n",
       "      <th>num_words_length_30</th>\n",
       "      <th>line_width_range_0-600</th>\n",
       "      <th>line_width_range_600-800</th>\n",
       "      <th>line_width_range_800-1300</th>\n",
       "      <th>line_width_range_1300-1600</th>\n",
       "      <th>line_width_range_1600-</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>NL-HaNA_1.05.06_1_0001.jpg</td>\n",
       "      <td>1</td>\n",
       "      <td>2522</td>\n",
       "      <td>3957</td>\n",
       "      <td>33</td>\n",
       "      <td>165</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>59</td>\n",
       "      <td>17</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>4</td>\n",
       "      <td>0</td>\n",
       "      <td>2</td>\n",
       "      <td>0</td>\n",
       "      <td>27</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>NL-HaNA_1.05.06_1_0002.jpg</td>\n",
       "      <td>2</td>\n",
       "      <td>4861</td>\n",
       "      <td>3963</td>\n",
       "      <td>59</td>\n",
       "      <td>286</td>\n",
       "      <td>2</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>134</td>\n",
       "      <td>17</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>3</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>2</td>\n",
       "      <td>54</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>NL-HaNA_1.05.06_1_0003.jpg</td>\n",
       "      <td>3</td>\n",
       "      <td>4886</td>\n",
       "      <td>3960</td>\n",
       "      <td>54</td>\n",
       "      <td>263</td>\n",
       "      <td>2</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>103</td>\n",
       "      <td>26</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>2</td>\n",
       "      <td>0</td>\n",
       "      <td>4</td>\n",
       "      <td>1</td>\n",
       "      <td>47</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>NL-HaNA_1.05.06_1_0004.jpg</td>\n",
       "      <td>4</td>\n",
       "      <td>4882</td>\n",
       "      <td>3955</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>...</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>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>NL-HaNA_1.05.06_1_0005.jpg</td>\n",
       "      <td>5</td>\n",
       "      <td>2493</td>\n",
       "      <td>3964</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>...</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>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>5 rows × 39 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "                       doc_id  doc_num  doc_width  doc_height  lines  words  \\\n",
       "0  NL-HaNA_1.05.06_1_0001.jpg        1       2522        3957     33    165   \n",
       "1  NL-HaNA_1.05.06_1_0002.jpg        2       4861        3963     59    286   \n",
       "2  NL-HaNA_1.05.06_1_0003.jpg        3       4886        3960     54    263   \n",
       "3  NL-HaNA_1.05.06_1_0004.jpg        4       4882        3955      0      0   \n",
       "4  NL-HaNA_1.05.06_1_0005.jpg        5       2493        3964      0      0   \n",
       "\n",
       "   text_regions  columns  extra  pages  ...  num_words_length_10  \\\n",
       "0             1        0      0      0  ...                   59   \n",
       "1             2        0      0      0  ...                  134   \n",
       "2             2        0      0      0  ...                  103   \n",
       "3             0        0      0      0  ...                    0   \n",
       "4             0        0      0      0  ...                    0   \n",
       "\n",
       "   num_words_length_15  num_words_length_20  num_words_length_25  \\\n",
       "0                   17                    0                    0   \n",
       "1                   17                    0                    0   \n",
       "2                   26                    0                    0   \n",
       "3                    0                    0                    0   \n",
       "4                    0                    0                    0   \n",
       "\n",
       "  num_words_length_30  line_width_range_0-600  line_width_range_600-800  \\\n",
       "0                   0                       4                         0   \n",
       "1                   0                       3                         0   \n",
       "2                   0                       2                         0   \n",
       "3                   0                       0                         0   \n",
       "4                   0                       0                         0   \n",
       "\n",
       "   line_width_range_800-1300  line_width_range_1300-1600  \\\n",
       "0                          2                           0   \n",
       "1                          0                           2   \n",
       "2                          4                           1   \n",
       "3                          0                           0   \n",
       "4                          0                           0   \n",
       "\n",
       "   line_width_range_1600-  \n",
       "0                      27  \n",
       "1                      54  \n",
       "2                      47  \n",
       "3                       0  \n",
       "4                       0  \n",
       "\n",
       "[5 rows x 39 columns]"
      ]
     },
     "execution_count": 8,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "import pandas as pd\n",
    "\n",
    "df = pd.DataFrame(scans_stats)\n",
    "\n",
    "print(df.shape)\n",
    "\n",
    "df.head(5)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "fcb47383",
   "metadata": {},
   "source": [
    "If you already have written the statistics to file (or downloaded the file from elsewhere), you can skip the steps above and load them from file."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "b65a7c6d",
   "metadata": {},
   "outputs": [],
   "source": [
    "stats_file = '../data/na_scan_stats.tsv.gz'\n",
    "\n",
    "df = pd.read_csv(stats_file, sep='\\t', compression='gzip')"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "0c932930",
   "metadata": {},
   "source": [
    "## Grouping scans by inventory\n",
    "\n",
    "The archive contains 15,857 scans belonging to 533 inventories. We use the document identifier to add information per scan on the inventory number and the sequence number of the scan in the inventory's list of scans.\n",
    "\n",
    "Once we have added this inventory and sequence information, we can:\n",
    "\n",
    "- group scans by inventory and compare inventories on aggregate statistics,\n",
    "- visualise characteristics of scans per inventory in sequence order."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "id": "4ebf43c6",
   "metadata": {
    "scrolled": false
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>doc_id</th>\n",
       "      <th>doc_num</th>\n",
       "      <th>doc_width</th>\n",
       "      <th>doc_height</th>\n",
       "      <th>lines</th>\n",
       "      <th>words</th>\n",
       "      <th>text_regions</th>\n",
       "      <th>columns</th>\n",
       "      <th>extra</th>\n",
       "      <th>pages</th>\n",
       "      <th>...</th>\n",
       "      <th>num_words_length_30</th>\n",
       "      <th>line_width_range_0-600</th>\n",
       "      <th>line_width_range_600-800</th>\n",
       "      <th>line_width_range_800-1300</th>\n",
       "      <th>line_width_range_1300-1600</th>\n",
       "      <th>line_width_range_1600-</th>\n",
       "      <th>archive_id</th>\n",
       "      <th>inv_num</th>\n",
       "      <th>scan_num</th>\n",
       "      <th>num_scans</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>NL-HaNA_1.05.06_1_0001.jpg</td>\n",
       "      <td>1</td>\n",
       "      <td>2522</td>\n",
       "      <td>3957</td>\n",
       "      <td>33</td>\n",
       "      <td>165</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>4</td>\n",
       "      <td>0</td>\n",
       "      <td>2</td>\n",
       "      <td>0</td>\n",
       "      <td>27</td>\n",
       "      <td>1.05.06</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>5</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>NL-HaNA_1.05.06_1_0002.jpg</td>\n",
       "      <td>2</td>\n",
       "      <td>4861</td>\n",
       "      <td>3963</td>\n",
       "      <td>59</td>\n",
       "      <td>286</td>\n",
       "      <td>2</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>3</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>2</td>\n",
       "      <td>54</td>\n",
       "      <td>1.05.06</td>\n",
       "      <td>1</td>\n",
       "      <td>2</td>\n",
       "      <td>5</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>NL-HaNA_1.05.06_1_0003.jpg</td>\n",
       "      <td>3</td>\n",
       "      <td>4886</td>\n",
       "      <td>3960</td>\n",
       "      <td>54</td>\n",
       "      <td>263</td>\n",
       "      <td>2</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>2</td>\n",
       "      <td>0</td>\n",
       "      <td>4</td>\n",
       "      <td>1</td>\n",
       "      <td>47</td>\n",
       "      <td>1.05.06</td>\n",
       "      <td>1</td>\n",
       "      <td>3</td>\n",
       "      <td>5</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>NL-HaNA_1.05.06_1_0004.jpg</td>\n",
       "      <td>4</td>\n",
       "      <td>4882</td>\n",
       "      <td>3955</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>...</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>1.05.06</td>\n",
       "      <td>1</td>\n",
       "      <td>4</td>\n",
       "      <td>5</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>NL-HaNA_1.05.06_1_0005.jpg</td>\n",
       "      <td>5</td>\n",
       "      <td>2493</td>\n",
       "      <td>3964</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>...</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>1.05.06</td>\n",
       "      <td>1</td>\n",
       "      <td>5</td>\n",
       "      <td>5</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>15852</th>\n",
       "      <td>NL-HaNA_1.05.06_99_0036.jpg</td>\n",
       "      <td>15853</td>\n",
       "      <td>5103</td>\n",
       "      <td>4067</td>\n",
       "      <td>46</td>\n",
       "      <td>265</td>\n",
       "      <td>3</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>6</td>\n",
       "      <td>0</td>\n",
       "      <td>3</td>\n",
       "      <td>1</td>\n",
       "      <td>35</td>\n",
       "      <td>1.05.06</td>\n",
       "      <td>99</td>\n",
       "      <td>36</td>\n",
       "      <td>40</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>15853</th>\n",
       "      <td>NL-HaNA_1.05.06_99_0037.jpg</td>\n",
       "      <td>15854</td>\n",
       "      <td>5125</td>\n",
       "      <td>4065</td>\n",
       "      <td>44</td>\n",
       "      <td>297</td>\n",
       "      <td>3</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>4</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>39</td>\n",
       "      <td>1.05.06</td>\n",
       "      <td>99</td>\n",
       "      <td>37</td>\n",
       "      <td>40</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>15854</th>\n",
       "      <td>NL-HaNA_1.05.06_99_0038.jpg</td>\n",
       "      <td>15855</td>\n",
       "      <td>5116</td>\n",
       "      <td>4104</td>\n",
       "      <td>44</td>\n",
       "      <td>290</td>\n",
       "      <td>3</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>4</td>\n",
       "      <td>1</td>\n",
       "      <td>4</td>\n",
       "      <td>2</td>\n",
       "      <td>33</td>\n",
       "      <td>1.05.06</td>\n",
       "      <td>99</td>\n",
       "      <td>38</td>\n",
       "      <td>40</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>15855</th>\n",
       "      <td>NL-HaNA_1.05.06_99_0039.jpg</td>\n",
       "      <td>15856</td>\n",
       "      <td>5111</td>\n",
       "      <td>4054</td>\n",
       "      <td>44</td>\n",
       "      <td>252</td>\n",
       "      <td>3</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>7</td>\n",
       "      <td>3</td>\n",
       "      <td>3</td>\n",
       "      <td>1</td>\n",
       "      <td>30</td>\n",
       "      <td>1.05.06</td>\n",
       "      <td>99</td>\n",
       "      <td>39</td>\n",
       "      <td>40</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>15856</th>\n",
       "      <td>NL-HaNA_1.05.06_99_0040.jpg</td>\n",
       "      <td>15857</td>\n",
       "      <td>2672</td>\n",
       "      <td>4049</td>\n",
       "      <td>27</td>\n",
       "      <td>129</td>\n",
       "      <td>2</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>4</td>\n",
       "      <td>3</td>\n",
       "      <td>5</td>\n",
       "      <td>0</td>\n",
       "      <td>14</td>\n",
       "      <td>1.05.06</td>\n",
       "      <td>99</td>\n",
       "      <td>40</td>\n",
       "      <td>40</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>15857 rows × 43 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "                            doc_id  doc_num  doc_width  doc_height  lines  \\\n",
       "0       NL-HaNA_1.05.06_1_0001.jpg        1       2522        3957     33   \n",
       "1       NL-HaNA_1.05.06_1_0002.jpg        2       4861        3963     59   \n",
       "2       NL-HaNA_1.05.06_1_0003.jpg        3       4886        3960     54   \n",
       "3       NL-HaNA_1.05.06_1_0004.jpg        4       4882        3955      0   \n",
       "4       NL-HaNA_1.05.06_1_0005.jpg        5       2493        3964      0   \n",
       "...                            ...      ...        ...         ...    ...   \n",
       "15852  NL-HaNA_1.05.06_99_0036.jpg    15853       5103        4067     46   \n",
       "15853  NL-HaNA_1.05.06_99_0037.jpg    15854       5125        4065     44   \n",
       "15854  NL-HaNA_1.05.06_99_0038.jpg    15855       5116        4104     44   \n",
       "15855  NL-HaNA_1.05.06_99_0039.jpg    15856       5111        4054     44   \n",
       "15856  NL-HaNA_1.05.06_99_0040.jpg    15857       2672        4049     27   \n",
       "\n",
       "       words  text_regions  columns  extra  pages  ...  num_words_length_30  \\\n",
       "0        165             1        0      0      0  ...                    0   \n",
       "1        286             2        0      0      0  ...                    0   \n",
       "2        263             2        0      0      0  ...                    0   \n",
       "3          0             0        0      0      0  ...                    0   \n",
       "4          0             0        0      0      0  ...                    0   \n",
       "...      ...           ...      ...    ...    ...  ...                  ...   \n",
       "15852    265             3        0      0      0  ...                    0   \n",
       "15853    297             3        0      0      0  ...                    0   \n",
       "15854    290             3        0      0      0  ...                    0   \n",
       "15855    252             3        0      0      0  ...                    0   \n",
       "15856    129             2        0      0      0  ...                    0   \n",
       "\n",
       "       line_width_range_0-600  line_width_range_600-800  \\\n",
       "0                           4                         0   \n",
       "1                           3                         0   \n",
       "2                           2                         0   \n",
       "3                           0                         0   \n",
       "4                           0                         0   \n",
       "...                       ...                       ...   \n",
       "15852                       6                         0   \n",
       "15853                       4                         0   \n",
       "15854                       4                         1   \n",
       "15855                       7                         3   \n",
       "15856                       4                         3   \n",
       "\n",
       "       line_width_range_800-1300 line_width_range_1300-1600  \\\n",
       "0                              2                          0   \n",
       "1                              0                          2   \n",
       "2                              4                          1   \n",
       "3                              0                          0   \n",
       "4                              0                          0   \n",
       "...                          ...                        ...   \n",
       "15852                          3                          1   \n",
       "15853                          1                          0   \n",
       "15854                          4                          2   \n",
       "15855                          3                          1   \n",
       "15856                          5                          0   \n",
       "\n",
       "       line_width_range_1600-  archive_id  inv_num  scan_num  num_scans  \n",
       "0                          27     1.05.06        1         1          5  \n",
       "1                          54     1.05.06        1         2          5  \n",
       "2                          47     1.05.06        1         3          5  \n",
       "3                           0     1.05.06        1         4          5  \n",
       "4                           0     1.05.06        1         5          5  \n",
       "...                       ...         ...      ...       ...        ...  \n",
       "15852                      35     1.05.06       99        36         40  \n",
       "15853                      39     1.05.06       99        37         40  \n",
       "15854                      33     1.05.06       99        38         40  \n",
       "15855                      30     1.05.06       99        39         40  \n",
       "15856                      14     1.05.06       99        40         40  \n",
       "\n",
       "[15857 rows x 43 columns]"
      ]
     },
     "execution_count": 9,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df['archive_id'] = df.doc_id.apply(lambda x: x.split('_')[1])\n",
    "df['inv_num'] = df.doc_id.apply(lambda x: x.split('_')[2])\n",
    "df['scan_num'] = df.doc_id.apply(lambda x: int(x.split('_')[3].replace('.jpg', '')))\n",
    "s_inv_nums = df.inv_num.value_counts()\n",
    "df['num_scans'] = df.inv_num.apply(lambda x: s_inv_nums.loc[x])\n",
    "df = df.sort_values(['inv_num', 'scan_num'])\n",
    "df"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "id": "a98785f2",
   "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>doc_id</th>\n",
       "      <th>doc_num</th>\n",
       "      <th>doc_width</th>\n",
       "      <th>doc_height</th>\n",
       "      <th>lines</th>\n",
       "      <th>words</th>\n",
       "      <th>text_regions</th>\n",
       "      <th>columns</th>\n",
       "      <th>extra</th>\n",
       "      <th>pages</th>\n",
       "      <th>...</th>\n",
       "      <th>num_words_length_30</th>\n",
       "      <th>line_width_range_0-600</th>\n",
       "      <th>line_width_range_600-800</th>\n",
       "      <th>line_width_range_800-1300</th>\n",
       "      <th>line_width_range_1300-1600</th>\n",
       "      <th>line_width_range_1600-</th>\n",
       "      <th>archive_id</th>\n",
       "      <th>inv_num</th>\n",
       "      <th>scan_num</th>\n",
       "      <th>num_scans</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>NL-HaNA_1.05.06_1_0001.jpg</td>\n",
       "      <td>1</td>\n",
       "      <td>2522</td>\n",
       "      <td>3957</td>\n",
       "      <td>33</td>\n",
       "      <td>165</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>4</td>\n",
       "      <td>0</td>\n",
       "      <td>2</td>\n",
       "      <td>0</td>\n",
       "      <td>27</td>\n",
       "      <td>1.05.06</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>5</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>NL-HaNA_1.05.06_1_0002.jpg</td>\n",
       "      <td>2</td>\n",
       "      <td>4861</td>\n",
       "      <td>3963</td>\n",
       "      <td>59</td>\n",
       "      <td>286</td>\n",
       "      <td>2</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>3</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>2</td>\n",
       "      <td>54</td>\n",
       "      <td>1.05.06</td>\n",
       "      <td>1</td>\n",
       "      <td>2</td>\n",
       "      <td>5</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>NL-HaNA_1.05.06_1_0003.jpg</td>\n",
       "      <td>3</td>\n",
       "      <td>4886</td>\n",
       "      <td>3960</td>\n",
       "      <td>54</td>\n",
       "      <td>263</td>\n",
       "      <td>2</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>2</td>\n",
       "      <td>0</td>\n",
       "      <td>4</td>\n",
       "      <td>1</td>\n",
       "      <td>47</td>\n",
       "      <td>1.05.06</td>\n",
       "      <td>1</td>\n",
       "      <td>3</td>\n",
       "      <td>5</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>NL-HaNA_1.05.06_1_0004.jpg</td>\n",
       "      <td>4</td>\n",
       "      <td>4882</td>\n",
       "      <td>3955</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>...</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>1.05.06</td>\n",
       "      <td>1</td>\n",
       "      <td>4</td>\n",
       "      <td>5</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>NL-HaNA_1.05.06_1_0005.jpg</td>\n",
       "      <td>5</td>\n",
       "      <td>2493</td>\n",
       "      <td>3964</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>...</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>1.05.06</td>\n",
       "      <td>1</td>\n",
       "      <td>5</td>\n",
       "      <td>5</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>NL-HaNA_1.05.06_10_0001.jpg</td>\n",
       "      <td>6</td>\n",
       "      <td>2622</td>\n",
       "      <td>4011</td>\n",
       "      <td>34</td>\n",
       "      <td>254</td>\n",
       "      <td>2</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>4</td>\n",
       "      <td>1</td>\n",
       "      <td>26</td>\n",
       "      <td>1.05.06</td>\n",
       "      <td>10</td>\n",
       "      <td>1</td>\n",
       "      <td>53</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>NL-HaNA_1.05.06_10_0002.jpg</td>\n",
       "      <td>7</td>\n",
       "      <td>5135</td>\n",
       "      <td>4010</td>\n",
       "      <td>67</td>\n",
       "      <td>519</td>\n",
       "      <td>3</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>4</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>3</td>\n",
       "      <td>58</td>\n",
       "      <td>1.05.06</td>\n",
       "      <td>10</td>\n",
       "      <td>2</td>\n",
       "      <td>53</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>NL-HaNA_1.05.06_10_0003.jpg</td>\n",
       "      <td>8</td>\n",
       "      <td>5141</td>\n",
       "      <td>4008</td>\n",
       "      <td>70</td>\n",
       "      <td>546</td>\n",
       "      <td>4</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>6</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>62</td>\n",
       "      <td>1.05.06</td>\n",
       "      <td>10</td>\n",
       "      <td>3</td>\n",
       "      <td>53</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8</th>\n",
       "      <td>NL-HaNA_1.05.06_10_0004.jpg</td>\n",
       "      <td>9</td>\n",
       "      <td>5140</td>\n",
       "      <td>4010</td>\n",
       "      <td>70</td>\n",
       "      <td>559</td>\n",
       "      <td>2</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>5</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>63</td>\n",
       "      <td>1.05.06</td>\n",
       "      <td>10</td>\n",
       "      <td>4</td>\n",
       "      <td>53</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9</th>\n",
       "      <td>NL-HaNA_1.05.06_10_0005.jpg</td>\n",
       "      <td>10</td>\n",
       "      <td>5137</td>\n",
       "      <td>4007</td>\n",
       "      <td>71</td>\n",
       "      <td>548</td>\n",
       "      <td>3</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>5</td>\n",
       "      <td>0</td>\n",
       "      <td>2</td>\n",
       "      <td>0</td>\n",
       "      <td>64</td>\n",
       "      <td>1.05.06</td>\n",
       "      <td>10</td>\n",
       "      <td>5</td>\n",
       "      <td>53</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>10 rows × 43 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "                        doc_id  doc_num  doc_width  doc_height  lines  words  \\\n",
       "0   NL-HaNA_1.05.06_1_0001.jpg        1       2522        3957     33    165   \n",
       "1   NL-HaNA_1.05.06_1_0002.jpg        2       4861        3963     59    286   \n",
       "2   NL-HaNA_1.05.06_1_0003.jpg        3       4886        3960     54    263   \n",
       "3   NL-HaNA_1.05.06_1_0004.jpg        4       4882        3955      0      0   \n",
       "4   NL-HaNA_1.05.06_1_0005.jpg        5       2493        3964      0      0   \n",
       "5  NL-HaNA_1.05.06_10_0001.jpg        6       2622        4011     34    254   \n",
       "6  NL-HaNA_1.05.06_10_0002.jpg        7       5135        4010     67    519   \n",
       "7  NL-HaNA_1.05.06_10_0003.jpg        8       5141        4008     70    546   \n",
       "8  NL-HaNA_1.05.06_10_0004.jpg        9       5140        4010     70    559   \n",
       "9  NL-HaNA_1.05.06_10_0005.jpg       10       5137        4007     71    548   \n",
       "\n",
       "   text_regions  columns  extra  pages  ...  num_words_length_30  \\\n",
       "0             1        0      0      0  ...                    0   \n",
       "1             2        0      0      0  ...                    0   \n",
       "2             2        0      0      0  ...                    0   \n",
       "3             0        0      0      0  ...                    0   \n",
       "4             0        0      0      0  ...                    0   \n",
       "5             2        0      0      0  ...                    0   \n",
       "6             3        0      0      0  ...                    0   \n",
       "7             4        0      0      0  ...                    0   \n",
       "8             2        0      0      0  ...                    0   \n",
       "9             3        0      0      0  ...                    0   \n",
       "\n",
       "   line_width_range_0-600  line_width_range_600-800  \\\n",
       "0                       4                         0   \n",
       "1                       3                         0   \n",
       "2                       2                         0   \n",
       "3                       0                         0   \n",
       "4                       0                         0   \n",
       "5                       1                         1   \n",
       "6                       4                         0   \n",
       "7                       6                         0   \n",
       "8                       5                         0   \n",
       "9                       5                         0   \n",
       "\n",
       "   line_width_range_800-1300 line_width_range_1300-1600  \\\n",
       "0                          2                          0   \n",
       "1                          0                          2   \n",
       "2                          4                          1   \n",
       "3                          0                          0   \n",
       "4                          0                          0   \n",
       "5                          4                          1   \n",
       "6                          1                          3   \n",
       "7                          0                          0   \n",
       "8                          0                          1   \n",
       "9                          2                          0   \n",
       "\n",
       "   line_width_range_1600-  archive_id  inv_num  scan_num  num_scans  \n",
       "0                      27     1.05.06        1         1          5  \n",
       "1                      54     1.05.06        1         2          5  \n",
       "2                      47     1.05.06        1         3          5  \n",
       "3                       0     1.05.06        1         4          5  \n",
       "4                       0     1.05.06        1         5          5  \n",
       "5                      26     1.05.06       10         1         53  \n",
       "6                      58     1.05.06       10         2         53  \n",
       "7                      62     1.05.06       10         3         53  \n",
       "8                      63     1.05.06       10         4         53  \n",
       "9                      64     1.05.06       10         5         53  \n",
       "\n",
       "[10 rows x 43 columns]"
      ]
     },
     "execution_count": 10,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.head(10)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "38b5a691",
   "metadata": {},
   "source": [
    "The number of scans per inventory is a heavily skewed distribution. Most inventories have only a handful of scans."
   ]
  },
  {
   "cell_type": "markdown",
   "id": "8a9a5a42",
   "metadata": {},
   "source": [
    "### Viewing scans\n",
    "\n",
    "Because the scans of this archive are also viewable online, we can make visual checks of scans. The scan identifier allows us to construct a viewer URL:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "id": "747ff900",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<style type=\"text/css\">\n",
       "</style>\n",
       "<table id=\"T_b59dd\">\n",
       "  <thead>\n",
       "    <tr>\n",
       "      <th class=\"blank level0\" >&nbsp;</th>\n",
       "      <th id=\"T_b59dd_level0_col0\" class=\"col_heading level0 col0\" >doc_id</th>\n",
       "      <th id=\"T_b59dd_level0_col1\" class=\"col_heading level0 col1\" >iiif_url</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th id=\"T_b59dd_level0_row0\" class=\"row_heading level0 row0\" >0</th>\n",
       "      <td id=\"T_b59dd_row0_col0\" class=\"data row0 col0\" >NL-HaNA_1.05.06_1_0001.jpg</td>\n",
       "      <td id=\"T_b59dd_row0_col1\" class=\"data row0 col1\" ><a target=\"_blank\" href=\"https://www.nationaalarchief.nl/onderzoeken/archief/1.05.06/invnr/1/file/NL-HaNA_1.05.06_1_0001?eadId=1.05.06&unitID=1\">https://www.nationaalarchief.nl/onderzoeken/archief/1.05.06/invnr/1/file/NL-HaNA_1.05.06_1_0001?eadId=1.05.06&unitID=1</a></td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_b59dd_level0_row1\" class=\"row_heading level0 row1\" >1</th>\n",
       "      <td id=\"T_b59dd_row1_col0\" class=\"data row1 col0\" >NL-HaNA_1.05.06_1_0002.jpg</td>\n",
       "      <td id=\"T_b59dd_row1_col1\" class=\"data row1 col1\" ><a target=\"_blank\" href=\"https://www.nationaalarchief.nl/onderzoeken/archief/1.05.06/invnr/1/file/NL-HaNA_1.05.06_1_0002?eadId=1.05.06&unitID=1\">https://www.nationaalarchief.nl/onderzoeken/archief/1.05.06/invnr/1/file/NL-HaNA_1.05.06_1_0002?eadId=1.05.06&unitID=1</a></td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_b59dd_level0_row2\" class=\"row_heading level0 row2\" >2</th>\n",
       "      <td id=\"T_b59dd_row2_col0\" class=\"data row2 col0\" >NL-HaNA_1.05.06_1_0003.jpg</td>\n",
       "      <td id=\"T_b59dd_row2_col1\" class=\"data row2 col1\" ><a target=\"_blank\" href=\"https://www.nationaalarchief.nl/onderzoeken/archief/1.05.06/invnr/1/file/NL-HaNA_1.05.06_1_0003?eadId=1.05.06&unitID=1\">https://www.nationaalarchief.nl/onderzoeken/archief/1.05.06/invnr/1/file/NL-HaNA_1.05.06_1_0003?eadId=1.05.06&unitID=1</a></td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_b59dd_level0_row3\" class=\"row_heading level0 row3\" >3</th>\n",
       "      <td id=\"T_b59dd_row3_col0\" class=\"data row3 col0\" >NL-HaNA_1.05.06_1_0004.jpg</td>\n",
       "      <td id=\"T_b59dd_row3_col1\" class=\"data row3 col1\" ><a target=\"_blank\" href=\"https://www.nationaalarchief.nl/onderzoeken/archief/1.05.06/invnr/1/file/NL-HaNA_1.05.06_1_0004?eadId=1.05.06&unitID=1\">https://www.nationaalarchief.nl/onderzoeken/archief/1.05.06/invnr/1/file/NL-HaNA_1.05.06_1_0004?eadId=1.05.06&unitID=1</a></td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_b59dd_level0_row4\" class=\"row_heading level0 row4\" >4</th>\n",
       "      <td id=\"T_b59dd_row4_col0\" class=\"data row4 col0\" >NL-HaNA_1.05.06_1_0005.jpg</td>\n",
       "      <td id=\"T_b59dd_row4_col1\" class=\"data row4 col1\" ><a target=\"_blank\" href=\"https://www.nationaalarchief.nl/onderzoeken/archief/1.05.06/invnr/1/file/NL-HaNA_1.05.06_1_0005?eadId=1.05.06&unitID=1\">https://www.nationaalarchief.nl/onderzoeken/archief/1.05.06/invnr/1/file/NL-HaNA_1.05.06_1_0005?eadId=1.05.06&unitID=1</a></td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_b59dd_level0_row5\" class=\"row_heading level0 row5\" >5</th>\n",
       "      <td id=\"T_b59dd_row5_col0\" class=\"data row5 col0\" >NL-HaNA_1.05.06_10_0001.jpg</td>\n",
       "      <td id=\"T_b59dd_row5_col1\" class=\"data row5 col1\" ><a target=\"_blank\" href=\"https://www.nationaalarchief.nl/onderzoeken/archief/1.05.06/invnr/10/file/NL-HaNA_1.05.06_10_0001?eadId=1.05.06&unitID=10\">https://www.nationaalarchief.nl/onderzoeken/archief/1.05.06/invnr/10/file/NL-HaNA_1.05.06_10_0001?eadId=1.05.06&unitID=10</a></td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_b59dd_level0_row6\" class=\"row_heading level0 row6\" >6</th>\n",
       "      <td id=\"T_b59dd_row6_col0\" class=\"data row6 col0\" >NL-HaNA_1.05.06_10_0002.jpg</td>\n",
       "      <td id=\"T_b59dd_row6_col1\" class=\"data row6 col1\" ><a target=\"_blank\" href=\"https://www.nationaalarchief.nl/onderzoeken/archief/1.05.06/invnr/10/file/NL-HaNA_1.05.06_10_0002?eadId=1.05.06&unitID=10\">https://www.nationaalarchief.nl/onderzoeken/archief/1.05.06/invnr/10/file/NL-HaNA_1.05.06_10_0002?eadId=1.05.06&unitID=10</a></td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_b59dd_level0_row7\" class=\"row_heading level0 row7\" >7</th>\n",
       "      <td id=\"T_b59dd_row7_col0\" class=\"data row7 col0\" >NL-HaNA_1.05.06_10_0003.jpg</td>\n",
       "      <td id=\"T_b59dd_row7_col1\" class=\"data row7 col1\" ><a target=\"_blank\" href=\"https://www.nationaalarchief.nl/onderzoeken/archief/1.05.06/invnr/10/file/NL-HaNA_1.05.06_10_0003?eadId=1.05.06&unitID=10\">https://www.nationaalarchief.nl/onderzoeken/archief/1.05.06/invnr/10/file/NL-HaNA_1.05.06_10_0003?eadId=1.05.06&unitID=10</a></td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_b59dd_level0_row8\" class=\"row_heading level0 row8\" >8</th>\n",
       "      <td id=\"T_b59dd_row8_col0\" class=\"data row8 col0\" >NL-HaNA_1.05.06_10_0004.jpg</td>\n",
       "      <td id=\"T_b59dd_row8_col1\" class=\"data row8 col1\" ><a target=\"_blank\" href=\"https://www.nationaalarchief.nl/onderzoeken/archief/1.05.06/invnr/10/file/NL-HaNA_1.05.06_10_0004?eadId=1.05.06&unitID=10\">https://www.nationaalarchief.nl/onderzoeken/archief/1.05.06/invnr/10/file/NL-HaNA_1.05.06_10_0004?eadId=1.05.06&unitID=10</a></td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_b59dd_level0_row9\" class=\"row_heading level0 row9\" >9</th>\n",
       "      <td id=\"T_b59dd_row9_col0\" class=\"data row9 col0\" >NL-HaNA_1.05.06_10_0005.jpg</td>\n",
       "      <td id=\"T_b59dd_row9_col1\" class=\"data row9 col1\" ><a target=\"_blank\" href=\"https://www.nationaalarchief.nl/onderzoeken/archief/1.05.06/invnr/10/file/NL-HaNA_1.05.06_10_0005?eadId=1.05.06&unitID=10\">https://www.nationaalarchief.nl/onderzoeken/archief/1.05.06/invnr/10/file/NL-HaNA_1.05.06_10_0005?eadId=1.05.06&unitID=10</a></td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n"
      ],
      "text/plain": [
       "<pandas.io.formats.style.Styler at 0x29150afe0>"
      ]
     },
     "execution_count": 11,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# https://www.nationaalarchief.nl/onderzoeken/archief/1.05.06/invnr/1/file/NL-HaNA_1.05.06_1_0001?eadID=1.05.06&unitID=1&query=1.05.06\n",
    "def make_iiif_url(row):\n",
    "    base_url = 'https://www.nationaalarchief.nl/onderzoeken/archief'\n",
    "    return f\"{base_url}/{row['archive_id']}/invnr/{row['inv_num']}/file/{row['doc_id'][:-4]}?eadId={row['archive_id']}&unitID={row['inv_num']}\"\n",
    "\n",
    "\n",
    "def make_clickable(val):\n",
    "    return f'<a target=\"_blank\" href=\"{val}\">{val}</a>'\n",
    "\n",
    "\n",
    "\n",
    "df['iiif_url'] = df.apply(lambda row: make_iiif_url(row), axis=1)\n",
    "df[['doc_id', 'iiif_url']].head(10).style.format({'iiif_url': make_clickable})\n"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "01ea52fe",
   "metadata": {},
   "source": [
    "### Writing stats to file\n",
    "\n",
    "Now that statistics of more than 14,000 scans have been calculated and various columns have been added, it is worthwhile writing the statistics to file, so they can be re-used in other analyses, e.g.:\n",
    "\n",
    "- [book section detection](./Demo-analysing-scan-characteristics-book-sections.ipynb)\n",
    "- [comparing subsets](./Demo-analysing-scan-characteristics-comparing-subsets.ipynb)\n",
    "- [checking quality](./Demo-analysing-scan-characteristics-checking-quality.ipynb)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "id": "91727ee6",
   "metadata": {},
   "outputs": [],
   "source": [
    "stats_file = '../data/na_scan_stats.tsv.gz'\n",
    "\n",
    "df.to_csv(stats_file, sep='\\t', compression='gzip', index=False)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "ae3f4247",
   "metadata": {},
   "source": [
    "## Comparing size and shape of scans\n",
    "\n",
    "### Width and height of scans\n",
    "\n",
    "To get an idea of how homogeneous or heterogeneous the scans are in terms of shape, we can plot the distribution of scan widths and heights."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "id": "d24ce8d3",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1152x432 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "import matplotlib.pyplot as plt\n",
    "\n",
    "fig, ax = plt.subplots(1, 2)\n",
    "\n",
    "sns.kdeplot(data=df.doc_width, ax=ax[0]);\n",
    "sns.kdeplot(data=df.doc_height, ax=ax[1]);\n",
    "fig.set_size_inches(16, 6)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "id": "dc46d6f5",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(5030.511794106549, 4044.1712745763934)"
      ]
     },
     "execution_count": 20,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "import numpy as np\n",
    "\n",
    "df.doc_width.value_counts()\n",
    "data = ax[0].lines[0].get_xydata()\n",
    "peak_width = data[np.where(data[:, 1] == max(data[:, 1]))][0][0]\n",
    "data = ax[1].lines[0].get_xydata()\n",
    "peak_height = data[np.where(data[:, 1] == max(data[:, 1]))][0][0]\n",
    "peak_width, peak_height"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "f015b58c",
   "metadata": {},
   "source": [
    "The vast majority of scans are roughly the same shape, with a width around 5000 pixels and a height around 4000 pixels. There are a few bumps at other widths and heights."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "id": "0dfb98ce",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 576x432 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "ax = sns.displot(data=df, x='doc_width', y='doc_height', binwidth=(200, 200));\n",
    "ax.figure.set_size_inches(8, 6)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "5a30be5e",
   "metadata": {},
   "source": [
    "There seem to be some small clusters of pages with different sizes:\n",
    "\n",
    "- smaller documents with both a lower width and height than 5000 by 4000\n",
    "- bigger documents with both a higher width and height than 5000 by 4000\n",
    "- wider and narrower documents with a height around 4000 but lower or higher width than 5000\n",
    "\n",
    "If we plot the ratio of width to height, we can see the distribution of scans that are roughly shaped as _landscape_, _portrait_ or _square_."
   ]
  },
  {
   "cell_type": "markdown",
   "id": "cf7ad676",
   "metadata": {},
   "source": [
    "### Shape of scans"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "id": "7f0c0096",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<Axes: ylabel='Density'>"
      ]
     },
     "execution_count": 22,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "sns.kdeplot(data=df.doc_width.div(df.doc_height, axis=0), clip=(0,1.5))"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "547b9b9a",
   "metadata": {},
   "source": [
    "We use this information to classify scan shape and scan sizes."
   ]
  },
  {
   "cell_type": "markdown",
   "id": "1d80c322",
   "metadata": {},
   "source": [
    "### Classifying scans by shape, width and height"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "id": "1a80f294",
   "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>doc_id</th>\n",
       "      <th>doc_num</th>\n",
       "      <th>doc_width</th>\n",
       "      <th>doc_height</th>\n",
       "      <th>lines</th>\n",
       "      <th>words</th>\n",
       "      <th>text_regions</th>\n",
       "      <th>columns</th>\n",
       "      <th>extra</th>\n",
       "      <th>pages</th>\n",
       "      <th>...</th>\n",
       "      <th>line_width_range_1300-1600</th>\n",
       "      <th>line_width_range_1600-</th>\n",
       "      <th>archive_id</th>\n",
       "      <th>inv_num</th>\n",
       "      <th>scan_num</th>\n",
       "      <th>num_scans</th>\n",
       "      <th>iiif_url</th>\n",
       "      <th>shape_class</th>\n",
       "      <th>width_class</th>\n",
       "      <th>height_class</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>NL-HaNA_1.05.06_1_0001.jpg</td>\n",
       "      <td>1</td>\n",
       "      <td>2522</td>\n",
       "      <td>3957</td>\n",
       "      <td>33</td>\n",
       "      <td>165</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>27</td>\n",
       "      <td>1.05.06</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>5</td>\n",
       "      <td>https://www.nationaalarchief.nl/onderzoeken/ar...</td>\n",
       "      <td>portrait</td>\n",
       "      <td>narrow</td>\n",
       "      <td>medium</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>NL-HaNA_1.05.06_1_0002.jpg</td>\n",
       "      <td>2</td>\n",
       "      <td>4861</td>\n",
       "      <td>3963</td>\n",
       "      <td>59</td>\n",
       "      <td>286</td>\n",
       "      <td>2</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>2</td>\n",
       "      <td>54</td>\n",
       "      <td>1.05.06</td>\n",
       "      <td>1</td>\n",
       "      <td>2</td>\n",
       "      <td>5</td>\n",
       "      <td>https://www.nationaalarchief.nl/onderzoeken/ar...</td>\n",
       "      <td>landscape</td>\n",
       "      <td>medium</td>\n",
       "      <td>medium</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>NL-HaNA_1.05.06_1_0003.jpg</td>\n",
       "      <td>3</td>\n",
       "      <td>4886</td>\n",
       "      <td>3960</td>\n",
       "      <td>54</td>\n",
       "      <td>263</td>\n",
       "      <td>2</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>1</td>\n",
       "      <td>47</td>\n",
       "      <td>1.05.06</td>\n",
       "      <td>1</td>\n",
       "      <td>3</td>\n",
       "      <td>5</td>\n",
       "      <td>https://www.nationaalarchief.nl/onderzoeken/ar...</td>\n",
       "      <td>landscape</td>\n",
       "      <td>medium</td>\n",
       "      <td>medium</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>NL-HaNA_1.05.06_1_0004.jpg</td>\n",
       "      <td>4</td>\n",
       "      <td>4882</td>\n",
       "      <td>3955</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>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1.05.06</td>\n",
       "      <td>1</td>\n",
       "      <td>4</td>\n",
       "      <td>5</td>\n",
       "      <td>https://www.nationaalarchief.nl/onderzoeken/ar...</td>\n",
       "      <td>landscape</td>\n",
       "      <td>medium</td>\n",
       "      <td>medium</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>NL-HaNA_1.05.06_1_0005.jpg</td>\n",
       "      <td>5</td>\n",
       "      <td>2493</td>\n",
       "      <td>3964</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>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1.05.06</td>\n",
       "      <td>1</td>\n",
       "      <td>5</td>\n",
       "      <td>5</td>\n",
       "      <td>https://www.nationaalarchief.nl/onderzoeken/ar...</td>\n",
       "      <td>portrait</td>\n",
       "      <td>narrow</td>\n",
       "      <td>medium</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>15852</th>\n",
       "      <td>NL-HaNA_1.05.06_99_0036.jpg</td>\n",
       "      <td>15853</td>\n",
       "      <td>5103</td>\n",
       "      <td>4067</td>\n",
       "      <td>46</td>\n",
       "      <td>265</td>\n",
       "      <td>3</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>1</td>\n",
       "      <td>35</td>\n",
       "      <td>1.05.06</td>\n",
       "      <td>99</td>\n",
       "      <td>36</td>\n",
       "      <td>40</td>\n",
       "      <td>https://www.nationaalarchief.nl/onderzoeken/ar...</td>\n",
       "      <td>landscape</td>\n",
       "      <td>medium</td>\n",
       "      <td>medium</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>15853</th>\n",
       "      <td>NL-HaNA_1.05.06_99_0037.jpg</td>\n",
       "      <td>15854</td>\n",
       "      <td>5125</td>\n",
       "      <td>4065</td>\n",
       "      <td>44</td>\n",
       "      <td>297</td>\n",
       "      <td>3</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>39</td>\n",
       "      <td>1.05.06</td>\n",
       "      <td>99</td>\n",
       "      <td>37</td>\n",
       "      <td>40</td>\n",
       "      <td>https://www.nationaalarchief.nl/onderzoeken/ar...</td>\n",
       "      <td>landscape</td>\n",
       "      <td>medium</td>\n",
       "      <td>medium</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>15854</th>\n",
       "      <td>NL-HaNA_1.05.06_99_0038.jpg</td>\n",
       "      <td>15855</td>\n",
       "      <td>5116</td>\n",
       "      <td>4104</td>\n",
       "      <td>44</td>\n",
       "      <td>290</td>\n",
       "      <td>3</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>2</td>\n",
       "      <td>33</td>\n",
       "      <td>1.05.06</td>\n",
       "      <td>99</td>\n",
       "      <td>38</td>\n",
       "      <td>40</td>\n",
       "      <td>https://www.nationaalarchief.nl/onderzoeken/ar...</td>\n",
       "      <td>landscape</td>\n",
       "      <td>medium</td>\n",
       "      <td>medium</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>15855</th>\n",
       "      <td>NL-HaNA_1.05.06_99_0039.jpg</td>\n",
       "      <td>15856</td>\n",
       "      <td>5111</td>\n",
       "      <td>4054</td>\n",
       "      <td>44</td>\n",
       "      <td>252</td>\n",
       "      <td>3</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>1</td>\n",
       "      <td>30</td>\n",
       "      <td>1.05.06</td>\n",
       "      <td>99</td>\n",
       "      <td>39</td>\n",
       "      <td>40</td>\n",
       "      <td>https://www.nationaalarchief.nl/onderzoeken/ar...</td>\n",
       "      <td>landscape</td>\n",
       "      <td>medium</td>\n",
       "      <td>medium</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>15856</th>\n",
       "      <td>NL-HaNA_1.05.06_99_0040.jpg</td>\n",
       "      <td>15857</td>\n",
       "      <td>2672</td>\n",
       "      <td>4049</td>\n",
       "      <td>27</td>\n",
       "      <td>129</td>\n",
       "      <td>2</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>14</td>\n",
       "      <td>1.05.06</td>\n",
       "      <td>99</td>\n",
       "      <td>40</td>\n",
       "      <td>40</td>\n",
       "      <td>https://www.nationaalarchief.nl/onderzoeken/ar...</td>\n",
       "      <td>portrait</td>\n",
       "      <td>narrow</td>\n",
       "      <td>medium</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>15857 rows × 47 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "                            doc_id  doc_num  doc_width  doc_height  lines  \\\n",
       "0       NL-HaNA_1.05.06_1_0001.jpg        1       2522        3957     33   \n",
       "1       NL-HaNA_1.05.06_1_0002.jpg        2       4861        3963     59   \n",
       "2       NL-HaNA_1.05.06_1_0003.jpg        3       4886        3960     54   \n",
       "3       NL-HaNA_1.05.06_1_0004.jpg        4       4882        3955      0   \n",
       "4       NL-HaNA_1.05.06_1_0005.jpg        5       2493        3964      0   \n",
       "...                            ...      ...        ...         ...    ...   \n",
       "15852  NL-HaNA_1.05.06_99_0036.jpg    15853       5103        4067     46   \n",
       "15853  NL-HaNA_1.05.06_99_0037.jpg    15854       5125        4065     44   \n",
       "15854  NL-HaNA_1.05.06_99_0038.jpg    15855       5116        4104     44   \n",
       "15855  NL-HaNA_1.05.06_99_0039.jpg    15856       5111        4054     44   \n",
       "15856  NL-HaNA_1.05.06_99_0040.jpg    15857       2672        4049     27   \n",
       "\n",
       "       words  text_regions  columns  extra  pages  ...  \\\n",
       "0        165             1        0      0      0  ...   \n",
       "1        286             2        0      0      0  ...   \n",
       "2        263             2        0      0      0  ...   \n",
       "3          0             0        0      0      0  ...   \n",
       "4          0             0        0      0      0  ...   \n",
       "...      ...           ...      ...    ...    ...  ...   \n",
       "15852    265             3        0      0      0  ...   \n",
       "15853    297             3        0      0      0  ...   \n",
       "15854    290             3        0      0      0  ...   \n",
       "15855    252             3        0      0      0  ...   \n",
       "15856    129             2        0      0      0  ...   \n",
       "\n",
       "       line_width_range_1300-1600  line_width_range_1600-  archive_id  \\\n",
       "0                               0                      27     1.05.06   \n",
       "1                               2                      54     1.05.06   \n",
       "2                               1                      47     1.05.06   \n",
       "3                               0                       0     1.05.06   \n",
       "4                               0                       0     1.05.06   \n",
       "...                           ...                     ...         ...   \n",
       "15852                           1                      35     1.05.06   \n",
       "15853                           0                      39     1.05.06   \n",
       "15854                           2                      33     1.05.06   \n",
       "15855                           1                      30     1.05.06   \n",
       "15856                           0                      14     1.05.06   \n",
       "\n",
       "       inv_num scan_num  num_scans  \\\n",
       "0            1        1          5   \n",
       "1            1        2          5   \n",
       "2            1        3          5   \n",
       "3            1        4          5   \n",
       "4            1        5          5   \n",
       "...        ...      ...        ...   \n",
       "15852       99       36         40   \n",
       "15853       99       37         40   \n",
       "15854       99       38         40   \n",
       "15855       99       39         40   \n",
       "15856       99       40         40   \n",
       "\n",
       "                                                iiif_url  shape_class  \\\n",
       "0      https://www.nationaalarchief.nl/onderzoeken/ar...     portrait   \n",
       "1      https://www.nationaalarchief.nl/onderzoeken/ar...    landscape   \n",
       "2      https://www.nationaalarchief.nl/onderzoeken/ar...    landscape   \n",
       "3      https://www.nationaalarchief.nl/onderzoeken/ar...    landscape   \n",
       "4      https://www.nationaalarchief.nl/onderzoeken/ar...     portrait   \n",
       "...                                                  ...          ...   \n",
       "15852  https://www.nationaalarchief.nl/onderzoeken/ar...    landscape   \n",
       "15853  https://www.nationaalarchief.nl/onderzoeken/ar...    landscape   \n",
       "15854  https://www.nationaalarchief.nl/onderzoeken/ar...    landscape   \n",
       "15855  https://www.nationaalarchief.nl/onderzoeken/ar...    landscape   \n",
       "15856  https://www.nationaalarchief.nl/onderzoeken/ar...     portrait   \n",
       "\n",
       "       width_class  height_class  \n",
       "0           narrow        medium  \n",
       "1           medium        medium  \n",
       "2           medium        medium  \n",
       "3           medium        medium  \n",
       "4           narrow        medium  \n",
       "...            ...           ...  \n",
       "15852       medium        medium  \n",
       "15853       medium        medium  \n",
       "15854       medium        medium  \n",
       "15855       medium        medium  \n",
       "15856       narrow        medium  \n",
       "\n",
       "[15857 rows x 47 columns]"
      ]
     },
     "execution_count": 23,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "def classify_scan_width(width, peak_width):\n",
    "    if width / peak_width < 0.9:\n",
    "        return 'narrow'\n",
    "    elif width / peak_width > 1.1:\n",
    "        return 'wide'\n",
    "    else:\n",
    "        return 'medium'\n",
    "\n",
    "\n",
    "def classify_scan_height(height, peak_height):\n",
    "    if height / peak_height < 0.9:\n",
    "        return 'high'\n",
    "    elif height / peak_height > 1.1:\n",
    "        return 'low'\n",
    "    else:\n",
    "        return 'medium'\n",
    "\n",
    "\n",
    "def classify_scan_shape(row):\n",
    "    if row['doc_width'] / row['doc_height'] > 1.1:\n",
    "        return 'landscape'\n",
    "    if row['doc_width'] / row['doc_height'] < 0.9:\n",
    "        return 'portrait'\n",
    "    else:\n",
    "        return 'square'\n",
    "    \n",
    "df['shape_class'] = df.apply(classify_scan_shape, axis=1)\n",
    "df['width_class'] = df.doc_width.apply(lambda width: classify_scan_width(width, peak_width))\n",
    "df['height_class'] = df.doc_height.apply(lambda height: classify_scan_width(height, peak_height))\n",
    "\n",
    "df"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "id": "c6a8f004",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAA8oAAAF2CAYAAAC2+GETAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjcuMSwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy/bCgiHAAAACXBIWXMAAAsTAAALEwEAmpwYAACA2ElEQVR4nO3de3xU5Z0/8M+Z+0wm5DoTIHhBvGBBSmtURA0/WiFckoqoqwttai+4dXdbpTUWBaSolNbSQt1K3dKLWwSXFDUpLQasLt6gmtBaUZGKCGICM7knk8zl3H5/TGaSIbdJMpOZOfm8X6/dZs6Zy/OciTz5nu/zfB9BVVUVRERERERERAQA0CW6AURERERERETJhIEyERERERERUQ8MlImIiIiIiIh6YKBMRERERERE1AMDZSIiIiIiIqIeGCgTERERERER9cBAmYiIiIiIiKgHQ6IbMJqamzugKKOzbXROjh2NjZ5R+Swt4XUbOl6z4eF1Gzpes246nYCsrLREN0MTRnNs7osWf6/Zp9ShxX5psU+ANvultT7FemyOa6C8Z88e/PKXv4QoirjzzjuxfPnyiPNHjx7FmjVr4PF4UFBQgPXr18Ng6G7Sz3/+c+h0Onz7298GALS1teG+++7D6dOnkZ2djS1btsDhcETdHkVRR3UwTuTAn8p43YaO12x4eN2GjteMYm20x+b+2qA17FPq0GK/tNgnQJv90mKfYiVuU69dLhc2b96MnTt3orKyErt27cLx48cjnlNWVoa1a9di3759UFUV5eXlAID29nY8+OCD+O1vfxvx/C1btqCgoAAvvPACbrvtNmzYsCFezSciIiIiIqIxKm6B8sGDBzFr1ixkZmbCZrOhqKgIVVVV4fO1tbXw+XyYOXMmAGDp0qXh8y+99BIuvPBCfO1rX4t4zwMHDqCkpAQAUFxcjFdffRWiKMarC0RERERERDQGxS1QdrvdEdOinU4nXC5Xv+cdDkf4/JIlS3DXXXdBr9f3+54GgwF2ux1NTU3x6gIRERERERGNQXFbo6yqvee7C4IQ9flo6XTRx/o5OfYhv/9IOBzpo/p5WsHrNnS8ZsPD6zZ0vGapY7h1Qurq6lBWVobGxkZMnjwZmzZtQlpad3GU3bt3o6amBj/60Y8AAIFAAI899hhqamogiiIeeOABXH/99aPaVyIioliLW6Ccl5eHmpqa8GO32w2n0xlxvqGhIfy4vr4+4nxfnE4nGhoaMH78eEiSBI/Hg8zMzKjb1NjoGbUF6w5HOurr20fls7SE123oeM2Gh9dt6HjNuul0wqjffB2KUJ2Q5557DiaTCXfccQeuueYaXHzxxeHnlJWV4dFHH8XMmTPx4IMPory8HMuWLcP69euxbNkyLF68GE888QS2bt2KsrIy+P1+/Nd//Rd27NiBoqKi8Pv8+te/RnNzM55//nkcP34cX//61/Hqq68O6+Y3ERFRsojb1OvZs2fj0KFDaGpqgtfrxf79+1FYWBg+n5+fD7PZjMOHDwMAKioqIs73Zc6cOaioqAAA7N27FwUFBTAajfHqAhERUUoabp0QURRRXV0dDoR71g+prq6GoigoKyuL+KwXXngBK1asgCAIuOSSS/C73/2uz1ljREREqSRugXJeXh5WrlyJ0tJSLFmyBMXFxZgxYwZWrFiBI0eOAAA2bdqEjRs3YuHChfB6vSgtLR3wPe+55x68/fbbWLx4MXbu3ImHHnooXs0nIiJKWcOtE9Lc3Ay73R7eqrFn/ZDrr78e999/PywWS8RnnTp1CtXV1Vi6dCluv/12NDQ0DGlZFBERUTKK6z7KJSUl4SrVIdu2bQv/PHXqVOzevbvf14f2Tw7JzMzEk08+GdtGEhERacxw64QMp36ILMs4e/Ysnn32WRw7dgzf/OY38cILLyA9Pbr17MkwhV2La+/Zp9ShxX5psU+ANvulxT7FSlwDZSIiIhp9w60Tkp2dDY/HA1mWodfro6ofkpubi8WLF0MQBEydOhXjx4/Hxx9/jBkzZkTV1tGsH9IXLa69Z59Shxb7pcU+Adrsl9b6FOv6IZwbRUREpDHDrRNiNBpRUFCAvXv3RhwfyNy5c8PPP336NM6cOYPJkyfHqWdERESjg4EyERGRxoykTsi6detQXl6ORYsWoaamBvfee++An3XffffB7XZj8eLF+Na3voVHH3006mnXREREyUpQx1BpSi1tDyXJCgx67d3n0NoUkNHAazY8vG5Dx2vWLdm3h0oliZ56nZtrR319u6a2s9Lif6ta7BOgzX5psU+ANvultT5x6jXhzfdduOfx1/DP0y2JbgoREVHKUlUV39zwIl55uy7RTSEioiTDQDkFnTrbDq9fxpY//AMn6toS3RwiIqKU5PVLcDd78f6p5kQ3hYiIkgwD5RTU2OZDpt0Eu9WIX/3xvUQ3h4iIKCW1e0UAwGm3J8EtISKiZMNAOQU1tvkwMTcNs6ePR32LF7KiJLpJREREKcfTGQyU3U2d8ItygltDRETJhIFyCmps9SFnnAXj0kxQ0T3QExERUfRCGWUVQG19R2IbQ0RESYWBcooRJQWtHYFgoGwzAQBaOwIJbhUREVHq6Xmj+dN6Tr8mIqJuDJRTTFO7DwCQk2FBhj0YKLcxUCYiIhoyT1dG2aAXuE6ZiIgiMFBOMU2twUA5u2vqNcCMMhER0XC0ewMw6AVcMD6dgTIREUVgoJxiGtq6M8qhqddtnQyUiYiIhsrTKWJcmgnnOdPxqdsDVVUT3SQiIkoSDJRTTFObHwKA7HQzLCY9TEYdWj0MlImIiIbK4xUxLs2M8xxp6PRLaGrzJ7pJRESUJBgop5jGVh8y7CYY9DoIgoBxNhMzykRERMPQ7u3OKAPAaRb0IiKiLgyUU0xjW3BrqJCMNBMzykRERMPg6RSRnmZC9jgzAKDFw4wyEREFMVBOMY1tPuRkdAfK49KYUSYiIhoOT1dG2Wo2AAC8finBLSIiomTBQDmFKKqKpj4yytweioiIaGgURUVHV6BsMekhCAyUiYioGwPlFNLeEYAkq8geF5lR9nSKkBUlgS0jIiJKLZ1+CSqC46ggCLCaDPD65EQ3i4iIkgQD5RTSc2uokIw0E1QA7Z1iglpFRESUetq7li2NSwuuT7aaDehkRpmIiLowUE4hLe3BIiNZdnP42Li0rr2UOf2aiIgoah5v8AZzaBy1mg2cek1ERGEMlFNIKGscGtQBIKPrTngrA2UiIqKoec4ZU21mPQNlIiIKY6CcQkLTxOxWY/jYuLTgz8woExERRa89lFG2MaNMRES9MVBOIW2dIqxmA4yG7q8tdCecGWUiIqLo9Zp6beEaZSIi6sZAOYW0dwaQbjNGHLOYDDAb9cwoExERDYGnU4TRoIPZpAfAjDIREUVioJxC2jvFXoEyEJx+zUCZiIgoeu3eAOxWIwRBAADYzAZ4/TJUVU1wy4iIKBkwUE4h7Z0i0q2mXsfHpZk49ZqIiGgIPJ0i0nvU/LCZDVBUFQFRGdH7tnUEcPJs20ibR0RECcZAOYW0e3tPvQYAu8WIDh/3USYiIoqWxyvC3mNMtZoNADDidcpPVr6LR/6nBn997+yI3oeIiBKLgXKKUFUVnk4xYmuoEKuF66qIiIiGwuOTkGbpHSiPZDz9qLYVH3zSArvViG1/eh81H7hH3E4iIkoMBsopotMvQVbUiGliIVZTcF0VERERRScgyjAb9eHHsQiU9/71FNIsBjz8jWtwvjMd5f93nGueiYhSFAPlFNHeGZxanW7rI6PcVamTgzEREVF0REmJ2G7RNsKp17UNHfj7hw344pWTkJFmwtzP56Oh1YdPXJ6YtJeIiEYXA+UU0d4ZLNbV1xplq1kPWVEhSiMrQEJERDRWnBsoW83B7PJwM8rVR10QBOCLV04CAMy8JBeCAPztn/UjbywREY06BsopYqCMsi0G08WIiIjGClVVEZBkmIw9A+WRZZRPuz3Iy7KFx+lxNhMuOy+TgTIRUYpioJwi2gbIKFtiVKmTiIhoLJAVFaoKGPW9A+Xh3nSure/AJEdaxLHPXepAbUMHzjZ1Dr+xRESUEAyUU8Rga5QBsKAXERFRFEJLlYyG7mJeFpMegjC8QNkfkFHf4sUkhz3i+JWXOgBw+jURUSpioJwi2jsDsJj0EeupQsJTrwPMKBMRUdCePXuwaNEizJs3Dzt27Oh1/ujRo7jllltQVFSE1atXQ5KCY0hdXR2WL1+OBQsW4O6770ZHR0fE63bv3o1Vq1b1ej+Px4Mbb7wRb775Znw6FEOBcKDcPaYKghDcRcI39JvOtQ0dUAHknxMoZ4+z4HynHe+eaBxRe4mIaPQxUE4Rnk6xz2nXQPAuOAB4fQyUiYgIcLlc2Lx5M3bu3InKykrs2rULx48fj3hOWVkZ1q5di3379kFVVZSXlwMA1q9fj2XLlqGqqgrTp0/H1q1bAQB+vx+bNm3Chg0b+vzMRx55BG1tbfHtWIyIUjAYNp1z89lqNgxrGdOn9cHK1uc503qdm3pBFo7XtoU/k4iIUgMD5RTR1hnoc9o1wGJeREQU6eDBg5g1axYyMzNhs9lQVFSEqqqq8Pna2lr4fD7MnDkTALB06VJUVVVBFEVUV1ejqKgo4jgAVFdXQ1EUlJWV9fq8vXv3Ii0tDZdddln8OxcD4anXxsg/g2wWw7DG0k/dHpiNeuRmWnudm3p+FiRZwYm61LiJQEREQQyUU0R7p4hx/QTKVkto6jXvVhMREeB2u+FwOMKPnU4nXC5Xv+cdDgdcLheam5tht9thMBgijgPA9ddfj/vvvx8WiyXis+rq6vA///M/uP/+++PZpZgKB8p6fcRxq3mYgXK9BxNz06AThF7nLj0vA4IAHD3VPLzGEhFRQhgS3QCKTntnABeMT+/znNXEjDIREXVTVbXXMaFHENff+cFedy5FUbB69WqsXbu2VwAdrZwc++BPirHGjmCBTEducKq0wxEcXzPTLWho8YYfR0NVVdQ1duKaaeP7fd2U/Ax8dKZ9SO87UqP5WaNFi30CtNkvLfYJ0Ga/tNinWGGgnAJUVUX7AGuUdToBZqOegTIREQEA8vLyUFNTE37sdrvhdDojzjc0NIQf19fXw+l0Ijs7Gx6PB7IsQ6/Xh4/358SJEzhx4gRWr14NAPjkk0+wZs0aPPLII5g1a1ZUbW1s9EBRegfo8eSubwcAdHr8AID6rsd6QUVbhz/8OBqtHj/aOgLITTf3+7opE8fhpcOforauBSajvs/nxJLDkT6kPqQCLfYJ0Ga/tNgnQJv90lqfdDohpjdfOfU6BXj9MmRFRbq176nXAGA167mPMhERAQBmz56NQ4cOoampCV6vF/v370dhYWH4fH5+PsxmMw4fPgwAqKioQGFhIYxGIwoKCrB3796I4/25+OKL8corr6CyshKVlZWYPn06Hn300aiD5EQR5b7XKA9n6vWnDcGq4OfuodxTcJ2yio9qW4fYUiIiShQGyimgrTMAABiX1ndGGQgO7j4GykREhGDGeOXKlSgtLcWSJUtQXFyMGTNmYMWKFThy5AgAYNOmTdi4cSMWLlwIr9eL0tJSAMC6detQXl6ORYsWoaamBvfee28CexIfATEYKJsMfa1Rlvucgt6fhhYvAMCZZev3OZeelwmdIOB9rlMmIkoZnHqdAhrbfACAnHH9r/8abgESIiLSppKSEpSUlEQc27ZtW/jnqVOnYvfu3b1el5+fj+3bt/f7vkuXLsXSpUv7PDfQ65JJOKN8zvZQNrMBiqoiICowm6KbIt3Y5ocgAJnpA836MuCi/HF49+Mm3DJnyvAbTkREo4YZ5RTQ2BpdoNzpZ9VrIiKiwYSqXve1jzKAIS1lam7zIdNuhl438J9U0ydn45Oz7eFZYkRElNwYKKeAhlYfdIKArHHmfp/DjDIREVF0AmLwxrKhn0B5KONpU7sf2QOMzyHTJ+dABfD+x03RN5SIiBKGgXIKaGz1ISvdNODdaptZD2+AgTIREdFgQlOv+8soDyVQbmzzDTjjK+TC8elIsxjwLgNlIqKUwEA5BTS2epGTYR3wORYTM8pERETREMX+1ygDiPrGs6qqaG73Izt98EBZpxMwbXI23vu4aUjFwoiIKDEYKKeAhijuVtvMBgREBVLXXXIiIiLqW0BSoNcJvWZqWczBAl6+KGt+tHtFiJIy4NKonqZPzkFrRwCn3Z6hNZiIiEYdA+UkJ8kKmtv9yM0YOFAOTRfzBVjQi4iIaCCipPTKJgOA1TS0qddNXbtSRJNRBoArpuRAJwj46/uuKFtKRESJwkA5yTW3+6GqQM4ggXLoLjinXxMREQ1MlORe65MBwDrEsbSpzQ8AyMmILqOckWbCzEty8fo7Z8KVt4mIKDkxUE5yoa2hBsso24ZRgISIiGgs6i+jbAlllKOcnTXUjDIAzJk5ER6viL9/WB/1a4iIaPQxUE5yDaE9lKOces1AmYiIaGABSYHRoO91XKcTYDbph5RRNuh1SLcZo/7saZOzkTPOglferov6NURENPoYKCeZv753Fgferg0/bmzzQcDgd6u7A2WuUSYiIhpIfxllALCa9PBFWfW6qd2H7HFmCIIQ9WfrBAGFMyfi6KlmHK9tjfp1REQ0uhgoJ5FWjx9PVX2A//3Lh+G72Q2tXmTYTf0O6CHMKBMREUWnvzXKQHA8jfamc1ObH9np0a1P7ukLn8+HI9OCJ54/ghaPf8ivJyKi+ItroLxnzx4sWrQI8+bNw44dO3qdP3r0KG655RYUFRVh9erVkKRgkFdXV4fly5djwYIFuPvuu9HR0QEAaG1txYoVK/ClL30Jt956K44ePRrP5o+6Px48iYCoICApqP7ADSC4Rjl3kD2Uge5AuZOBMhER0YACA2SULSZD1DedG9t8yB5k+8a+pFmM+M+lM+D1S/j5H97B4WNutHj8cLd40dDqRadP4l7LREQJFrdA2eVyYfPmzdi5cycqKyuxa9cuHD9+POI5ZWVlWLt2Lfbt2wdVVVFeXg4AWL9+PZYtW4aqqipMnz4dW7duBQD87ne/w6WXXoo//vGP+Pd//3c8/PDD8Wr+qHM1d+LVt+vw/z6Xjwk5Nrx+5AyA4BrlwdYnA4AttPdjlNPFiIiIxiqxnzXKQLDytTeKsVRWFLR4/MMKlAHgPKcdK4qnobHNhyeefxff/cUbWPXkIdz/y0P4zy2v4uGnalDzgRsKA2YiooSIW6B88OBBzJo1C5mZmbDZbCgqKkJVVVX4fG1tLXw+H2bOnAkAWLp0KaqqqiCKIqqrq1FUVBRxHAAURQlnl71eLyyW4Q1OyeiVt+sgCMBN112I666YgOOftuKPr3+MxjYfnJmDZ5QNeh30OoEZZSIiokGIkjLg1GtfFFOvWz0BqCqQPW7oU69DrrzMgc3fvg7fu2Mmvjz/Unxj8eX42qKpuGXORfAFJGyteBdPvfABs8tERAlgiNcbu91uOByO8GOn04l33nmn3/MOhwMulwvNzc2w2+0wGAwRxwHg61//Om6//XZcf/316OjowG9/+9t4NX/UNbX5kDPOggy7GddOG49nX/kIFa9/jKnnZ+LGgkmDvl4QhCGtqyIiIhqrApI8QDEvQ1QZ5eb24NriLPvwA2UA0Ot0mHZhNqZdmB1xfOE1F+D5107gz4dOIdNuwtLCKSP6HCIiGpq4Bcp93f3sWRWyv/MDve6RRx7B8uXLUVpair///e9YuXIl/vznPyMtLS2qNuXk2KNtfkw4HOlRP9cbUJCTaYXDkQ6HIx13Lv4MdDodSm64CHpddNU07TYjVAhD+txklOrtTwRes+HhdRs6XjPSgoGqXlvM+qhuOrd2BAAAmSMMlPuj0wlYWngR2jtF/OngKVw0MQMzL86Ny2cREVFvcQuU8/LyUFNTE37sdrvhdDojzjc0NIQf19fXw+l0Ijs7Gx6PB7IsQ6/Xh48DwEsvvRRel/y5z30OOTk5+OijjzBjxoyo2tTY6IGijM70JYcjHfX17VE/v6GlExNz08KvuWH6eABAU6Mn6vcw6XVobfcN6XOTzVCvG/GaDRev29DxmnXT6YRRv/lKsROcet3PGmWTAT6/BEVVoRtg26fWrmrVGXZTXNoIBBMFX55/KY6easKeNz7GZ6fkDGkrKiIiGr64rVGePXs2Dh06hKamJni9Xuzfvx+FhYXh8/n5+TCbzTh8+DAAoKKiAoWFhTAajSgoKMDevXsjjgPA1KlT8Ze//AUAcPLkSbjdbkyePDleXRhVbR0BjEsb2WBrMem5PRQREdEgBtxH2WyACsAfGDir3OIJQACQbjPGvoE9GPQ6LJp1AT4+0473TzbH9bOIiKhb3ALlvLw8rFy5EqWlpViyZAmKi4sxY8YMrFixAkeOHAEAbNq0CRs3bsTChQvh9XpRWloKAFi3bh3Ky8uxaNEi1NTU4N577wUA/OhHP8Kzzz6L4uJifPe738WPf/xjpKen/jRASVbQ4ZOQMdJA2WyAb5CBnYiIaCxTVXXg7aHCu0gMPJ62dgSQnmaCXhfXnTYBALOnT0BWuhl7Dp6M+2cREVFQ3KZeA0BJSQlKSkoijm3bti3889SpU7F79+5er8vPz8f27dt7Hb/wwgvx+9//PvYNTbC2rnVOscgon+X2UERERP2SZAUA+g2Ubebgn0Zev4Ss9P7XH7d6/Mgc4bgdLaNBhwVXn49nXvoQp90enOfktH8ioniL/21QGlRbZzBQzrCNNFBmRpmIiGggohQMlPvbHspi6gqUB7nx3NIRwLg4rk8+11WXB+u1vPNRwyDPJCKiWGCgnARaPbHJKFvNeviYUSYiIupXoCtQNhr7KeYVmno9SOXrto4AMtPiU/G6L5l2M87Ps+PIR42j9plERGMZA+UkELup1wYERAWyosSiWURERJoTyigb9f3vowxgwOKYiqqirSMQ14rXfZkxJQfHa9vQ4RNH9XOJiMYiBspJIDT1esQZZVPwLvhglTqJiIjGqlBG2WQcuJjXQIGyp1OErKgjLsI5VDMuyoWiqnjv46ZR/VwiorGIgXISaO0IwGLSw9zPNLBoWcIFSBgoExER9UWUgmPkQNtDAYB3gJvOrV0zwTLtozf1GgAumjgOaRYDp18TEY0CBspx0tLux0uHP8XbHw5edCMWeygDwarXALhOmYiIqB/hqdf9BcpdU699A2SUWz1+ABj1qdc6nYDpF+XgyMdNUFV1VD+biGisiev2UGPVX987i9/8+ShkRcXE3DTMvCR3wOfHLlDuGtw59ZqIiKhP4anXhr5ncel0AsxG/YBVr1u6inCO9tRrALjsvEy8+b4Lja0+5GZaR/3ziYjGCmaU4+DjM+3Q63WYdmFWVNnd1o7AiLeGArozyoNtaUFERDRWDZZRBoLrlAdao9zaEcooj+7UawA4Py8dAHDK1T7qn01ENJYwUI4DUVZgMxswPjstqsJabTHaizG0rmqwLS2IiIjGqmgCZavJMGC9j1ZPAFbzyGuLDMckRxp0goBTLs+ofzYR0VjCQDkOREmG0aiD2aQfdBq0JCvo8EnMKBMREY2CgBgcl00DBcrmQaZedwQwbhT3UO7JZNRjQo4NnzCjTEQUVwyU40CUFJgMOlhMesiKGr573Zf2zuBeiLFYoxzOKHONMhHRmLdnzx4sWrQI8+bNw44dO3qdP3r0KG655RYUFRVh9erVkKRgYFhXV4fly5djwYIFuPvuu9HR0RHxut27d2PVqlXhx263G9/4xjdw00034eabb8ahQ4fi27EREuVQRrn/bLDFZBhwdlabx4/MBKxPDjk/L52BMhFRnDFQjgNRUmA06GEO7WssDrTFRHCdU2yrXjNQJiIay1wuFzZv3oydO3eisrISu3btwvHjxyOeU1ZWhrVr12Lfvn1QVRXl5eUAgPXr12PZsmWoqqrC9OnTsXXrVgCA3+/Hpk2bsGHDhoj3eeyxxzB37lxUVlbipz/9Ke677z7IcvKOQ6I4+NRrm9kwaEZ5tCte93RBnh0tnkB4myoiIoo9BspxIMoKTEZdVNs1tXUNcrEIlA16HQx6YcAtLYiISPsOHjyIWbNmITMzEzabDUVFRaiqqgqfr62thc/nw8yZMwEAS5cuRVVVFURRRHV1NYqKiiKOA0B1dTUURUFZWVnEZ82fPx8lJSUAgAsuuAB+vx+dnZ2j0Mvh6c4oD1zMa+DtoQLISNDUa6C7oBezykRE8cNAOQ6kroxyNNs1tcYwUAa6posxo0xENKa53W44HI7wY6fTCZfL1e95h8MBl8uF5uZm2O12GAyGiOMAcP311+P++++HxWKJ+Kz58+cjIyMDAPCb3/wGl19+OdLT0+PWt5EKrVEerJhXZz9Tr71+CX5RRmYCM8rn59kBMFAmIoon7qMcB6KkwGoxhqthDlT5usMbvGOdbjXG5LMtJn1UW1IREZF2qara65ggCIOeH+x1A3nqqaewa9cuPP3000NoKZCTYx/S80fKaDLAaNAhzzkufMzhiAzsc7Js8AUk5Obae/X/dFdwev7EjF6vG0152TacbfH124ZEti1etNgnQJv90mKfAG32S4t9ihUGynEgSgpMRn331OsB1ih7/RIEILyeeaSs5oG3tCAiIu3Ly8tDTU1N+LHb7YbT6Yw439DQEH5cX18Pp9OJ7OxseDweyLIMvV4fPj6Yxx57DK+88gp27NiB8ePHD6mtjY0eKErvAD1eWtt8MOh1qK8PBrwOR3r45xBFlqGqwOnalnChzJATnzQBAPSq2ut1o2lSbhqOf9LcZxv66lOq02KfAG32S4t9ArTZL631SacTYnrzlVOv40CUFRgNPdYoDxC4ev0SLGY9dFHesR8MM8pERDR79mwcOnQITU1N8Hq92L9/PwoLC8Pn8/PzYTabcfjwYQBARUUFCgsLYTQaUVBQgL1790YcH8hTTz2FN998E88888yQg+RECHTtTDEQuyU4y6vDK/Y619weLMKZlZ64NcoAMD7HhvoWHyS5/501iIho+Bgox0Eoo9xd9br/wNXrl3rdrR4Jq9kAL9coExGNaXl5eVi5ciVKS0uxZMkSFBcXY8aMGVixYgWOHDkCANi0aRM2btyIhQsXwuv1orS0FACwbt06lJeXY9GiRaipqcG9997b7+eoqoonnngCTU1N+MpXvoKbbroJN910U8R66GQjSvKA65OB7rohrZ29q0q3eIKBcmaCA+W8LBsUVUVjqy+h7SAi0ipOvY6D7oxy8PIOtEa5M8aBssWkh6uZgTIR0VhXUlISrkYdsm3btvDPU6dOxe7du3u9Lj8/H9u3b+/3fZcuXYqlS5cCCK5frq6ujlGLR0dwC8foAuX2jt4Z5Zb2AGxmQ7gOSaLkZVsBAK7mTuRl2xLaFiIiLWJGOQ6k0Bpl4+D7Gsc6o8yp10RERP2LKlC2BQPltj4yys0ef8KzyUAwowwAriZvgltCRKRNDJTjQOxa/2Qy6iBgkEA5IMMW00DZMOCaaCIiorEsNOtrIOPSgmuU2zr6CJTb/chK4NZQIek2I6xmPVzNybtnNRFRKmOgHGOqqnbdrdZDEASYTXr4B6l6HeuMsl+UR7WCKBERUaqQJAVG/cB//hgNeljN+j4D5ZYkySgLgoC8LBtczcwoExHFAwPlGJMVFSoAkzF4ac2DTIWORzEvYOAsNhER0VglKSr0gwTKAJBuM/Waeq0oKlo9gYRXvA7Jy7bB1cSMMhFRPDBQjjFRCm7TYDQE1ydbjPoo1ijHriBIeEsqrlMmIiLqJZqMMhAs6HVuRrm1IwBFVZFlT5JAOcuKxjZf+G8PIiKKHQbKMdYdKAcvrcVk6LfqtSjJkGQVVlNs1ygDzCgTERH1RZQVGPTCoM8bZzOhvTOy6nWybA0Vkpdlg6oC9S2cfk1EFGsMlGMsFCibDD2nXvcdtHZ2Fd2K7dTrYEbZy4wyERFRL5KswDBIMS8gmFFuPSej3NIeDJSTZeq1s8cWUUREFFsMlGNMlLsyyl1bQ1lMevj6Kebl9QeD2VhXvQaYUSYiIuqLJKsw6KIIlG1GdHhFyEr3tObmroxy8ky95hZRRETxwkA5xs7NKFtM+n6nXocC5VhXvQbALaKIiIj6MJSMsgrA02P6dXO7HzpBQLot8dtDAYDdaoTdaoSbGWUiophjoBxj4UC5K6NsNvZf9bo7UI5hMa9w1WtOvSYiIjqXNIQ1ygDQ1iNQbmn3I8Nugk43+OtHizPLyi2iiIjigIFyjIlSMJNr7LFGub99lOORUbaGq14zo0xERHQuUVKjrnoNIKLydbPHnzTrk0NyMyxobPUluhlERJrDQDnGQmuUTaHtoUwG+AIyVFXt9dzOOK5RDgXhREREFKSqaldGeQiBco+9lJvb/UmzPjkkNyO4RZSi9P47g4iIho+BcoyFt4cydq9RVlUg0Mceh95Q1WtL7AJlo0EHg17HQJmIiOgcclcwGd3UayOA7oyyJCtwN3vDlaaTRW6mBbKihreuIiKi2GCgHGO9tofqWqvcV0Gv8NTrGO6jDAA2s56BMhER0Tmkrllf0RTzspoNMOiFcEb5TGMnZEXF+c70uLZxqHIzLACABk6/JiKKKQbKMXZuMa9wFeo+1il7/RLMRn3Mi4JYzYbwtG4iIiIKkuRQRnnwP3+ErurWoYzyJ652AMB5Tnv8GjgMuRnBDHdDKwt6ERHFEgPlGAvvo9xjeygA8PURuHb6pZhWvA6xmg3had1EREQUFF4eFUWgDATXKbd3Vb0+7fbAaNAhL8mmXueMY0aZiCgeGCjHWK/toboC5b4qX3v9UkwrXofYLAZ0+sXBn0hERDSGhKZe66NYowwEt4hq7coon3Z7MMmRBr0uuf50Mhp0yLSb0NDCQJmIKJaS6197DZCkczPKwUC4rzXKPr8U04rXIcwoExER9RYKlKPPKBvR4vFDVVWcdnuSbtp1SG6GlVOviYhijIFyjIWndYW2hzL2v69xp1+OS0Y5GChzjTIREVFPQ1mjDACXTMpEqyeAg++ehccr4rwkK+QVkpth4dRrIqIYY6AcY6KsQK8ToO8q0BWaet1XoBy3qdcs5kVERNTLUKpeA8C10/KQkWbC/770IYDkK+QVkptpQVObH7LSeytKIiIaHgbKMSZKSnjaNdBdzGs01yhbzQb4AzIHTCIioh6GWszLaNBj/lXnocMXvPmctIFyhhWKqqK5nXspExHFCgPlGBPlvgNlX6B3htcbxzXKwffnOmUiIqIQOZRRjrKYFwDMmZkPq1kPR6YlLje3YyGnay/lRk6/JiKKmeT8Fz+FiZISsfbJoNdBrxN6Tb2WZAUBSYElDttD2cKBsgS71Rjz9yciIkpF4hDXKAPBnSS+tvByyIoar2aNWG5G9xZRlyW4LUREWsFAOcakc6ZeC4IAs1Hfq+p1qNhWvKZe9/wMIiIi6rFGeQiBMgAUTHXGozkxkzPOAgFAfQsrXxMRxQqnXsfYuWuUgWBBr3MzyqEgNh5Tr21dWWoGykRERN2GWswrVRj0OmSmmzn1mogohrQ1UiQBUVZ6FQmxmPS91iiH1g/HJaNsCb4nK18TERF16y7mFf0a5VTBLaKIiGKLgXKM9ZVRtpkN8HLqNRERUUKF1hkPdep1KsjNsKKhlVOviYhiRXsjRYL1FShbLQZ0+iKD1tBWE2mW+AXK534mERHRWBbKKGszULagqd0fnl5OREQjo72RIsFEqffUa5vZ0GsadIdPBACkWWJfldrGjDIREVEvwy3mlQpyMyxQVXAvZSKiGNHeSJFg5+6jDHRNve4KjENC2d40a+wzyga9DiaDjvsoExER9RAKlI0Gba5RBoAGVr4mIooJBsoxJkpy31Ov+8go63XBraPiwdpHFpuIiGgsC0291msxo5xpBQAW9CIiihHtjRQJ1l8xL0lWIUrdGd4Or4g0iwGCEJ+72lazgVOviYjGsD179mDRokWYN28eduzY0ev80aNHccstt6CoqAirV6+GJAXHjLq6OixfvhwLFizA3XffjY6OjojX7d69G6tWrQo/DgQCKCsrw8KFC3HzzTfjo48+im/HRkBWVOh1AnRxGnsTKSvdDEFgoExEFCsMlGMsuEY5Mkts61qH3LO4lscnIc0a+/XJIQyUiYjGLpfLhc2bN2Pnzp2orKzErl27cPz48YjnlJWVYe3atdi3bx9UVUV5eTkAYP369Vi2bBmqqqowffp0bN26FQDg9/uxadMmbNiwIeJ9tm/fDqvVihdeeAEPPvhgRBCdbERJ0eT6ZCC47Co73cxAmYgoRrQ5WiRQf2uUgch9jYMZ5fgFyjaznlOviYjGqIMHD2LWrFnIzMyEzWZDUVERqqqqwudra2vh8/kwc+ZMAMDSpUtRVVUFURRRXV2NoqKiiOMAUF1dDUVRUFZWFvFZBw4cwJe+9CUAwFVXXYXm5mbU1dWNQi+HTpIVGDS4h3JIDreIIiKKGQbKMaSqavBu9bmBsqX3dk2dPikuW0OFWC1GZpSJiMYot9sNh8MRfux0OuFyufo973A44HK50NzcDLvdDoPBEHEcAK6//nrcf//9sFgsA36Ww+HA2bNn49KvkZLk3mO0ljgyLMwoExHFSPwitTFIVlSoKnoX8+oro+wTMTE3LW5tYUaZiGjsUlW117GeNTH6Oz/Y66Kl00UfjObk2If8/sNlMBpgNurhcKRHHD/3cao6f2IGDr53FqKkaKZPPWmxT4A2+6XFPgHa7JcW+xQrcQ2U9+zZg1/+8pcQRRF33nknli9fHnH+6NGjWLNmDTweDwoKCrB+/XoYDAbU1dWhrKwMjY2NmDx5MjZt2oS0tDR4PB6sW7cuXChkw4YNmDZtWjy7MCShapp97aMMRGaUO3xiXLaGCuEaZSKisSsvLw81NTXhx263G06nM+J8Q0ND+HF9fT2cTieys7Ph8XggyzL0en34+ECcTifq6+txwQUXRLxXtBobPVCU3gF6PHg6/BAEAfX17eFjDkd6xONUZjXooKpAfUsnjH3c9EhlWvqeetJiv7TYJ0Cb/dJan3Q6IaY3X+M2/ygehUQ2btyICRMmoKKiAt/97nfxgx/8IF7NHxYxvD9j31OvQ4GrJCvw+mXY47hG2Wo2ICAq4T0jiYho7Jg9ezYOHTqEpqYmeL1e7N+/H4WFheHz+fn5MJvNOHz4MACgoqIChYWFMBqNKCgowN69eyOOD2TOnDmorKwEANTU1MBsNmPixIlx6tnIBAtuaneNsiMzOC3e3dSZ4JYQEaW+uAXKsS4koqoq9u/fj7vuugsAUFhYiB/+8Ifxav6wSFI/gfI5U69D/xvvqtcAmFUmIhqD8vLysHLlSpSWlmLJkiUoLi7GjBkzsGLFChw5cgQAsGnTJmzcuBELFy6E1+tFaWkpAGDdunUoLy/HokWLUFNTg3vvvXfAz/rKV76CQCCAxYsXY8OGDXjsscfi3b1hk2RVs1WvASAnIxgou5pY0IuIaKTiNve3r0Ii77zzTr/nBysk0tjYCJPJhKeffhr79+/HuHHj8OCDD8ar+cPS39Rro0EHg14IT70O/W88i3nZegTK6TZT3D6HiIiSU0lJCUpKSiKObdu2Lfzz1KlTsXv37l6vy8/Px/bt2/t936VLl2Lp0qXhx2azGT/+8Y9j0OL403oxr6x0M3SCAHdzJ4DsRDeHiCilxS1Si3UhEVmW0dDQgIyMDFRUVOCNN97Af/zHf+Cll16Kuk3xLhjSIQXbnpMdLNLVc3F8mtUIVRDgcKSjsVMEAEzIGxe3BfTjnR4AgNlmTrlF+qnW3mTAazY8vG5Dx2tGqUySFRh02p16rdfpkD3ODFcjp14TEY1U3ALlWBcSycrKgsFgQHFxMQDguuuuQ2dnJxobG5GTkxNVm+JdMMTdtRi+s9MPABGL4y1GPRpbOlFf347aM60AACkgxm0BvegLBuN1Z9uQYdbH5TPiQWtFBUYDr9nw8LoNHa9Zt1gXDKHRIckKzKb4LXtKBrkZlq6MMhERjUTc5h/FupCIyWTC7Nmz8ec//xkA8Pbbb8NqtSIrKyteXRgysZ81ykCwoJfXLwMAOrzBqdfxLuYFRFbaJiIiGstESe21PEprcjOscLGYFxHRiMVttIhHIZENGzbg1VdfRXFxMX7wgx9g8+bNQ9qrMd76W6MMBNcMd/qDWV5PV7Y3nsW8QltPdXR9FhER0VgnyYqmi3kBwYxyU5sPoiQnuilERCktrvsox7qQiNPpxJNPPhn7hsZIf9tDAYDVYkRTe3BKdijLGyq4FQ+hAl7tnYEBn9fi8cNi0sNiiuuvAhERUcIFA2XtrlEGuitfN7b5MT7bluDWEBGlLm3fVh1l/W0PBYQyysEAucMrwmo2QBfHgiJmox4mgw4e78AZ5R89/TeUv3x8wOcQERFpwVjIKDsyrQCAhlZuEUVENBJMI8bQgGuUzQZ4uzLJHT4xrltDhaTbjGjv7D9QFiUZ7hYvlI/jV+CMiIgoWUiyquntoYDg1GsAaGj1JbglRESpTdujxSgLT73u42611WJAQFIgyQo6fFJc1yeH2G2mATPKjW3BqeANrT40tXFAJSIibRNlRfPFvDLtZuh1AhpaOK4TEY2EtkeLUSZ1Bcp9TesKrUfu9Evo8Iqwj0ZG2WoccI1yY4+7zf883RL39hARESWSJGl/6rVOJ8CRZeXUayKiEdL2aDHKJDk4hVnfR6EQW1dg7PVJo5ZRHmzqdWgQ1QkCA2UiItI0VVUhK6rmi3kBQF62LeJmOBERDR0D5RiSQxnlPrasisgo+0TY4riHcojdakL7gFOvfdAJAi6/MAvHGCgTEZGGhW5maz2jDADOLBvqGSgTEY1IVKPFt7/9bRw8eDDebUl5ktJ/Rtlq7t7XuMMrjVoxL39A7ncvxcZWH7LSzfjMBVk409iJto6Bt5IiIqLRxzE4NgZaHqU1edk2tHUEEBC5lzIR0XBFNVrMnz8fW7duRVFREX7zm9+gpaUlzs1KTaGMsr6PbZ9CU6+b2/xQVBVpo5FRtgU/o7/p142tPuRkWHDpeZkAuE6ZiCgZcQyOjXDBTY1XvQaCgTIQnDlGRETDE9VoUVJSgqeffhpbt25FY2MjbrvtNpSVleGdd96Jd/tSiiSr0OsECEIfgXJXRvmv77sAAJl2U9zbk24Nfka/gXKbDznjLDg/zw4AONPUGfc2ERHR0HAMjg2pawvHvmZ9aY2zK1DmFlFERMMX9W1VRVFw6tQpnDx5EpIkIScnBz/4wQ/wk5/8JJ7tSymy0n81zVBG+eipZkw9PxNXXuaMe3vSuzLKfW0RJckKmtr9yMmwwGjQQ68T4AtIcW8TERENHcfgkQstj9L69lBAd0a5oYWVr4mIhiuqhbKbN2/Gc889h/POOw/Lli3Dz3/+cxiNRnR2dmLu3LkoKyuLdztTQiij3BezUQ+TQQdHlhX/ufSKUZn6lR6eet177XFLux+qCuRmWAAE11D7AlzLRESUbDgGx0YoozwW1ihnpVtg0AvMKBMRjUBUgXJTUxO2bduGqVOnRhy32Wz46U9/GpeGpSJZVvrddkIQBJT96+eQl20blYrXAJBu63/qdWjdUk5XoGwx6eHzM6NMRJRsOAbHxlgq5qXTCcgZZ2GgTEQ0AlGNFrIs9xqgv/3tbwMArr/++ti3KkVJigr9AAPwlPwM2Edh/+QQm8UAQUCfW0SFBs/ccT0CZWaUiYiSDsfg2Ogu5qX9NcpAcMZYQyunXhMRDdeAGeV169bB5XLh8OHDaGpqCh+XJAknTpyIe+NSjSwr/U69TgSdIMBuNcLTx9Trxq5AOXucGQBg4dRrIqKkwjE4tmQ5tIWj9jPKAJCTYcUnH9YnuhlERClrwED51ltvxYcffohjx46hqKgofFyv1+Nzn/tc3BuXaiRZTbopXXarse+McpsPGWkmGA16AMGMsqef6thERDT6OAbHVjijnGTjdLzkZljQ3inCH5BhNukT3RwiopQzYKB8xRVX4IorrsB1112HvLy80WpTypIVtd81yomSbjP1vUa5aw/lEKvJgIYWrmUiIkoWHINjaywV8wKA3MzgGN/Q5kN+blqCW0NElHoGDJTvuece/PznP8c3v/nNPs/v2bMnLo1KVZKsJN2UrnSbEXUNHb2ON7b5cEFeevhxcI0yi3kRESULjsGx1V3MK7luaMdLboYVQHCLKAbKRERDN2CgvGLFCgDA2rVrR6UxqU6WFRiSaI0yAKRbjb32UVZUFU1tPlx5qSN8zGIywMs1ykRESYNjcGx1F/NKrhva8RLa/pGVr4mIhmfA0WL69OkAgKuvvhoTJkzA1VdfjZaWFrz11lu4/PLLR6WBqUQepOp1IthtJni8IhRVDR9r9QQgyWrk1GuzHv6AHPE8IiJKHI7BsTXWinkF65DowsU7iYhoaKIaLR566CFs27YNH330ER5++GHU1tZi9erV8W5bygkW80q+jLKqAp2+7mnV4T2Ux3UHyhZTcHKBn1llIqKkwjE4NsZaMS9BCO6lXM8tooiIhiWq0eLdd9/FD37wA7z44ou4+eabsXHjRtTW1sa7bSlHkhXodck1AKfbgvs2t/fYIip0d7lnRtnSVRGTW0QRESUXjsGx0V3MK7luaMdTcC9lZpSJiIYjqqhOVVXodDq88cYbmDVrFgDA6+UdynMlY0Y5uytrXN+jonWfGWVzKFBmQS8iomTCMTg2pK6p12Ol6jUQDJQ59ZqIaHiiGi3OP/98rFixAp9++imuvvpqfO9738Nll10W77alHFlJvqrX47NtAICzTZ3hY42tPqRZDLCau2u5haZeM6NMRJRcOAbHhiiPre2hACA30wqPV4TXz5vgRERDNWDV65CNGzfixRdfxJVXXgmj0YiCggIsWbIkzk1LPbKsJl/Va5sRNrMhIlBuaPVFZJMBwNo19ZqDKRFRcuEYHBvyGNseCgAcmcEtoupbvDi/x5aQREQ0uKhuq9psNhQUFKCtrQ3vvfceZsyYgRMnTsS7bSlHUhTok2wAFgQB43NsONvYvZdyY5svYn0ywIwyEVGy4hgcG6KswKAXIAjJNU7HU15WMFB2NXOqPhHRUEWVUf7JT36Cp59+Gjk5OeFjgiDgpZdeilvDUpEsq0k5pWt8tg3vn2wCEFzr1tjqw2cuzIp4DtcoExElp+GOwXv27MEvf/lLiKKIO++8E8uXL484f/ToUaxZswYejwcFBQVYv349DAYD6urqUFZWhsbGRkyePBmbNm1CWloa2tracN999+H06dPIzs7Gli1b4HA4EAgE8MADD+Cf//wndDodvv/972P27NlxuRYjIUnJOUbHk7MrUO45q4yIiKITVaD8wgsvYP/+/cjLy4t3e1JasOp18t2pHp9tw8F3z8LrlyArKvyijNxeU6+DvwpePzPKRETJZDhjsMvlwubNm/Hcc8/BZDLhjjvuwDXXXIOLL744/JyysjI8+uijmDlzJh588EGUl5dj2bJlWL9+PZYtW4bFixfjiSeewNatW1FWVoYtW7agoKAAv/rVr1BRUYENGzZgy5YtqKyshKIo2LNnD44dO4YVK1bg1VdfjcelGBFJVsZcoGwxGZCVboaLgTIR0ZBFNWJMmDCBQXIUJCU571aHCnq5mjvR0LWfYu+p18woExElo+GMwQcPHsSsWbOQmZkJm82GoqIiVFVVhc/X1tbC5/Nh5syZAIClS5eiqqoKoiiiuroaRUVFEccB4MCBAygpKQEAFBcX49VXX4UoilAUBV6vF7Isw+v1wmKJHF+ShSgrMBqSb4yOt7wsKwNlIqJhiCqjfO211+Kxxx7DF7/4xYgBcNq0aXFrWCqSZTXp1igDwPic7srXxq5A/txA2WjQQa8TuEaZiCjJDGcMdrvdcDgc4cdOpxPvvPNOv+cdDgdcLheam5tht9thMBgijp/7GoPBALvdjqamJtx88814/vnnccMNN6CtrQ0/+9nPYtPxGJMkJTwGjiXjs22o/sCd6GYQEaWcqALl5557DgAi7kZzjXIkVVUhywr0uuQbhPOyrBAAnG3shK1rS6hzq14LggCLSQ8fp14TESWV4YzBqqr2OtaziFV/5wd73bl0Oh1+8YtfYObMmXjmmWdw8uRJ3HnnnZg2bRry8/P7fV1POTn2qJ43UjqDDmazAQ5H7+rPfR1LdaE+TTk/CwferoPZZsa4NFOCWzUyWvyeAG32S4t9ArTZLy32KVaiCpRffvnleLcj5SmqChXJue2E0aBHbqYFZ5s6YTLoYTbqYbcaez3PYtJz6jURUZIZzhicl5eHmpqa8GO32w2n0xlxvqGhIfy4vr4eTqcT2dnZ8Hg8kGUZer0+fBwIZqUbGhowfvx4SJIEj8eDzMxMvPTSS9i8eTMEQcDkyZPx2c9+Fu+8807UgXJjoweK0jtAjzVPRwA6APX17RHHHY70XsdSXc8+pXUtrXr/Qzem5GckslkjosXvCdBmv7TYJ0Cb/dJan3Q6IaY3X6NKf3Z0dODhhx/GV7/6VbS0tOChhx5CR0fH4C8cQ2Q5OMgn4xplABifnYYjJxrx+pEzuOpyZ58ZAovZAC+nXhMRJZXhjMGzZ8/GoUOH0NTUBK/Xi/3796OwsDB8Pj8/H2azGYcPHwYAVFRUoLCwMLxP8969eyOOA8CcOXNQUVEBANi7dy8KCgpgNBoxdepU/OUvfwEANDU14d1338Xll18e68swYpKswGBIvpvZ8RaqU8LK10REQxNVVPfoo48iPT0djY2NMJvN8Hg8eOihh+LdtpQidQXKyVj1GggOlF6/jEsmZeAr8y/r8znMKBMRJZ/hjMF5eXlYuXIlSktLsWTJEhQXF2PGjBlYsWIFjhw5AgDYtGkTNm7ciIULF8Lr9aK0tBQAsG7dOpSXl2PRokWoqanBvffeCwC455578Pbbb2Px4sXYuXNnuA0PPPAAjhw5gsWLF+OrX/0qvvvd7+LCCy+M2/UYrrG6Rjk3wwKdIMDVzECZiGgoopp6ffToUWzcuBGvvPIKrFYrNm3ahOLi4ni3LaVIigIgeTPKn7skF67mTnyz+DP9Vv20mAzw+hkoExElk+GOwSUlJeEq1SHbtm0L/zx16lTs3r271+vy8/Oxffv2XsczMzPx5JNP9jqem5uLX/7yl9F0JaFEWUGaqfeyI60z6HVwZFpwtsmb6KYQEaWUqAJl3TkFqmRZ7nVsrAtNvU7GqtcAMPWCLEy9IGvA51hNejS1+UapRUREFA2OwbEhSuqYzCgDQF62jVtEERENUVSB8lVXXYWf/OQn8Pl8eO211/D000/jmmuuiXfbUoosd2WUU/iPF4vJwO2hiIiSDMfg2JBkJWlnfcVbXpYNH3zSDFVVB6xiTkRE3aIaMe677z7YbDakp6djy5YtmDp1Ku6///54ty2lSEpyZ5SjYTHrGSgTESUZjsGxIUpKv0uPtG58thUBUUGLJ5DophARpYxBM8ovvvgifvOb3+DYsWOwWCy47LLL8PnPfx5ms3k02pcywhnlFL5bHcwoS7zjTESUJDgGx86Yzij3qHydlc7fHSKiaAwYKFdUVGDr1q34zne+g6lTp0IQBBw5cgQbNmyA3+/H/PnzR6udSS9U9dqQpFWvo2E16aGqQEBUYO7ad5GIiBKDY3BsSfLYrHoNdG8R5WrqxOWD1CshIqKgAQPl7du346mnnsLEiRPDx6ZMmYLPfvazePDBBzlI9xCqep3aU6+Dvw6+gMRAmYgowTgGx9ZYnnqdmW6GyaDjXspEREMw4IghimLEAB0yefJk+P3+uDUqFXVXvU7dQdjSFRx7uU6ZiCjhOAbHligrMBhS92b2SOgEAc4sVr4mIhqKAaM6vb7/rKKqqjFvTCrrrnqduoNwKFD2BbiXMhFRonEMjh1ZUaCqqV1HZKTGZ1txtpl7KRMRRWvsjhgx1l31OnUvqdXUNfXaz4wyERFphygFb2aP1anXQLCgV0OLF3LXUjEiIhrYgGuUjx07hs9//vO9jquqikCAWwz0JIWrXqdwRtkcmnrNjDIRUaJxDI6dcMHNFL6ZPVLjs22QFRUNrT7kZdkS3RwioqQ3YKD84osvjlY7Up4crnqduoOw2RgMlAMi7zYTESUax+DYCWeUx3CgHAqOXU2dDJSJiKIwYKCcn58/Wu1IeVqoeh0KlP0ip14TESUax+DYEWVOvc7LtgIAzjZ5MWNKghtDRJQCxu6IEWNaqHptCmeUGSgTEZF2SFJoeVTqjtEjZbcakWYxsPI1EVGUxu6IEWOyEpp6nboZZVPXnXZmlImISEu664iM3T97BEFAXraNeykTEUVp7I4YMRYahFM5o2w06CCAa5SJiEhbWPU6KC+LgTIRUbTG9ogRQ90VNVM3oywIAkxGPQISM8pERKQdoZvZxhQeo2NhYq4Nze1+eP3c3YKIaDAMlGMktC9hKle9BgCTUQc/M8pERKQhoWJehjGeUZ6QkwYAONPIrDIR0WDG9ogRQ1K4mFdq3602G/Us5kVERJrCqddBE3KC20KdaexIcEuIiJLf2B4xYkgOrVFO4WJeQLDyNQNlIiLSku7lUWP7zx5nlhV6nYA6BspERIMa2yNGDMmKCr1OgCCkdqBs5tRrIiLSmND2UMYxHijrdTrkZdtwpoFTr4mIBjO2R4wYkmRFE3eqTQZmlImISFtCa5TH+tRrIDj9mlOviYgGF9cRY8+ePVi0aBHmzZuHHTt29Dp/9OhR3HLLLSgqKsLq1ashScEqjHV1dVi+fDkWLFiAu+++Gx0dkf+gnz17FldffTU+/fTTeDZ/SCRZTflp10Bw6jX3USYiIi0JrVHWwg3tkZqQkwZ3izd8TYiIqG9xGzFcLhc2b96MnTt3orKyErt27cLx48cjnlNWVoa1a9di3759UFUV5eXlAID169dj2bJlqKqqwvTp07F169bwaxRFwerVqyGKYryaPiyyoqb01lAhZqMOAQ6eRESkIaHtoRgoAxNzbFBVwNXM6ddERAOJ24hx8OBBzJo1C5mZmbDZbCgqKkJVVVX4fG1tLXw+H2bOnAkAWLp0KaqqqiCKIqqrq1FUVBRxPOTXv/41Zs+ejaysrHg1fVgkWYFeAwMwi3kREZHWsOp1N24RRUQUnbiNGG63Gw6HI/zY6XTC5XL1e97hcMDlcqG5uRl2ux0GgyHiOAC8++67ePPNN/G1r30tXs0eNllWOPWaiIgoCXVnlFN/nB6p8Tk2CADqGrhOmYhoIIZ4vbGqqr2O9awI3d/5/o57vV48/PDD2LJlC3S64cX3OTn2Yb0uGgajAWaTAQ5HevhYz59TReY4C0RJSWjbU/G6JRqv2fDwug0drxmlIlFWYNCn/s4UsWA26pGTYWFBLyKiQcQtUM7Ly0NNTU34sdvthtPpjDjf0NAQflxfXw+n04ns7Gx4PB7Isgy9Xh8+XlNTg4aGBtx9993h97vrrrvwi1/8AhdddFFUbWps9EBRegfisdDRGYAAFfX17QCCf0yGfk4lsijDH5Dhdrcl5A+KVL1uicRrNjy8bkPHa9ZNpxPievOVYkuUFE677mGSw45P6xkoExENJG6jxuzZs3Ho0CE0NTXB6/Vi//79KCwsDJ/Pz8+H2WzG4cOHAQAVFRUoLCyE0WhEQUEB9u7dG3H8hhtuwMsvv4zKykpUVlbC6XTiV7/6VdRBcrxJsgL9MDPdycRs0kMFWA2TiIg0Q5JVFvLqYZIzDWcbOyFKXGpFRNSfuI0aeXl5WLlyJUpLS7FkyRIUFxdjxowZWLFiBY4cOQIA2LRpEzZu3IiFCxfC6/WitLQUALBu3TqUl5dj0aJFqKmpwb333huvZsaM3DWtK9WZuu64s/I1ERFphSQpDJR7OM+ZDkVVUdfAgl5ERP2J29RrACgpKUFJSUnEsW3btoV/njp1Knbv3t3rdfn5+di+ffuA7/3yyy/HppExIiuqZqpeA4A/IMNuNSa4NURERCMnypx63dMkR7Dy9Wm3BxeMZ90BIqK+cNSIEUlWNVH12twVKAc4HYuIiDRCkhQYNXAzO1bysmwwGXQ47fYkuilEREmLo0aMSLI2pnWZjME+cIsoIqLUtmfPHixatAjz5s3Djh07ep0/evQobrnlFhQVFWH16tWQJAkAUFdXh+XLl2PBggW4++670dERLPrU1taGu+66CwsXLsTy5ctRX18PAAgEAnj00UexZMkSLF68GK+//vrodTJKokbG6FjR6QTkO9LwaT0DZSKi/nDUiBFZUbWxRjmUURa5RpmIKFW5XC5s3rwZO3fuRGVlJXbt2oXjx49HPKesrAxr167Fvn37oKoqysvLAQDr16/HsmXLUFVVhenTp2Pr1q0AgC1btqCgoAAvvPACbrvtNmzYsAEA8Otf/xrNzc14/vnnsWXLFjzwwAN9bvWYSKx63dt5TjtOuz1J910RESULjhoxEqx6nfqBcnjqNTPKREQp6+DBg5g1axYyMzNhs9lQVFSEqqqq8Pna2lr4fD7MnDkTALB06VJUVVVBFEVUV1ejqKgo4jgAHDhwIFx3pLi4GK+++ipEUcQLL7yAFStWQBAEXHLJJfjd736XdMGXpJGCm7E0yWGHxyuixRNIdFOIiJISA+UYkTWy9USo6jWnXhMRpS632w2HwxF+7HQ64XK5+j3vcDjgcrnQ3NwMu90Og8EQcfzc1xgMBtjtdjQ1NeHUqVOorq7G0qVLcfvtt6OhoQG6JNsuUZIVGJhRjnCeM7gPOKdfExH1La5Vr8cSWVGg18DdarOJU6+JiFJdXxldQRAGPT/Y686l0+kgyzLOnj2LZ599FseOHcM3v/lNvPDCC0hPj66ack6OParnjYQKAXabCQ5H323q73gqG6xP1jQzAKCpQ0yZ/qdKO4dKi/3SYp8AbfZLi32KFQbKMRKsep36d6tNhq7toZhRJiJKWXl5eaipqQk/drvdcDqdEecbGhrCj+vr6+F0OpGdnQ2PxwNZlqHX68PHgWBWuqGhAePHj4ckSfB4PMjMzERubi4WL14MQRAwdepUjB8/Hh9//DFmzJgRVVsbGz1QlPhO1fb6JSiygvr69l7nHI70Po+nsmj75My04t0P6zHnivGj0KqR0eL3BGizX1rsE6DNfmmtTzqdENObr6kf2SUJrax/MndVveYaZSKi1DV79mwcOnQITU1N8Hq92L9/PwoLC8Pn8/PzYTabcfjwYQBARUUFCgsLYTQaUVBQgL1790YcB4A5c+agoqICALB3714UFBTAaDRi7ty54eefPn0aZ86cweTJk0ext4OTJFa97suU/Ax8WNuadGvKiYiSAUeNGAlWvU79yxmqeu2XOPWaiChV5eXlYeXKlSgtLcWSJUtQXFyMGTNmYMWKFThy5AgAYNOmTdi4cSMWLlwIr9eL0tJSAMC6detQXl6ORYsWoaamBvfeey8A4J577sHbb7+NxYsXY+fOnXjooYcAAPfddx/cbjcWL16Mb33rW3j00UejnnY9WkSZVa/7cvGkDLR1BFDf6kt0U4iIkg6nXsdIcOp16meUDXod9DqBGWUiohRXUlISrlIdsm3btvDPU6dOxe7du3u9Lj8/H9u3b+91PDMzE08++WSv43a7HY899lgMWhw/zCj37ZL8DADA8U9b4My0Jrg1RETJhaNGjMiyAr1GBmGTUcc1ykREpBmSrMCokTE6libmpsFq1uN4bVuim0JElHQ4asSAoqhQAU2sUQaC06+ZUSYiIi1QVRWixO2h+qLTCbhoYgaOf9qS6KYQESUdjhoxIMnB9bxamdZlNui5PRQREWmC3HUz26iRm9mxdkl+BmrrO9DpkxLdFCKipKKNyC7BJDlYLVILa5QBTr0mIiLtELuKUxq7tj+kSFMmZUAFcKKuNdFNISJKKgyUY0BWNJZR5tRrIiLSiO5ZX9q4mR1rUyaOg0Gvw5ETTYluChFRUtFGZJdg4YyyRgZhk1HP7aGIiEgTQmM01yj3zWIy4DMXZuFv/6znfspERD1w1IgBuetutVamXpuNegQCzCgTEVHqE6XgeMaq1/278lIHGtt8+MTlSXRTiIiSBkeNGJCVrrvVGhmETUYdM8pERKQJYldG2ciMcr8+e0kuBAE4/M/6RDeFiChpcNSIAa1VvTYZuEaZiIi0QZK0NUbHwzibCZdOysTfGSgTEYVx1IgBrVW9ZjEvIiLSClEOVb3mnzwD+fylDtQ2dOBMY0eim0JElBQ4asSApGiromZweyhOvSYiotTHjHJ0rrrcCb1OwMuHaxPdFCKipMBRIwbkcNVrbVxOk1EPSVagKKx+SUREqS20PIrFvAaWaTdj1rQ8vHakDh6vmOjmEBElHEeNGAivUdbQ1GsA8HP6NRERpThR4tTraBVdfT4CooL/+zuzykREHDViIHy32qBPcEtiw2QM/loEWPmaiIhSnChra3lUPE1y2DH9omy8dPhT1iohojGPgXIMiJK2BmFmlImISCvCs76YUY7K4lkXoK0jgP3VpxPdFCKihOKoEQNam9Zl6gqUAwEGykRElNrCYzTXKEflsvOz8PlLHfjzoVNo8fgT3RwiooThqBEDWtt6wmoKBso+BspERJTiwrO+NDJGj4bb5k6BJCt47tUTiW4KEVHCcNSIAUljd6stJgMAwBeQEtwSIiKikQkFyqFlRTS4vCwb5hWchzfeOYNTZ9sT3RwiooTQRmSXYFqbem3pyih7mVEmIqIUF6q3oZUxerQUz74QdpsR//vSh1DV+G8X+cxfPsTGpw/j1X/UsZAYESUFjhoxoLWp16FA2ednRpmIiFJbQFRgMuigE7RRcHO02CwGLLnhIhw73YK//bM+rp/1ab0Hf6k5jbqGDjz1wgd45qUP4/p5RETR0EZkl2DdVa+1cTkt5tDUa97RJSKi1OaX5HCRShqaws9OQL4jDeX/dzz8t0487HnjJEwmPTb+27W4+nInDh+rh6LEP4tNRDQQbUR2CSbKCgx6AYJG7laHM8pco0xERCkuIMowGfnnznDodTrc8YVLUN/iw0uHP43LZ9TWe1DzgRs3XjkJdqsRn7/UAY9XxPHa1rh8HhFRtDhyxIAkqZqZdg0EM+MGvY5rlImIKOUFp14zozxc0yZn47NTcrDn4Mdo6wjE/P33VZ+GyahH0dXnAwCmT86BXifgH8cbYv5ZRERDoZ3oLoFEWdFMxesQi0nPqddERJTymFEeuX/5wsUIiAqefy2220VJsoK//7Men780F3arEUBwbfSl52XibQbKRJRgHDliQJRkze3PaDXrOfWaiIhSXkBSuEZ5hCbkpOGLV07Cq2/X4eMzbTF732OftKDDJ+HKy5wRx2dekoszjZ1wNXfG7LOIiIZKW9FdgkiyqsGMsgE+PzPKRESU2gKiDLPGbmYnwk3XT8a4NBOe3n8MSoy2i6o55obZqMf0ydkRx2denAsAeOejxph8DhHRcHDkiAFRUjS1RhkITb1mRpmIiFKbX2RGORasZgP+Ze7F+PhMOw78vXbE76coKv7+z3rMmJLT6/txZFoxLs2ET1ztI/4cIqLh0lZ0lyCipGhma6gQq9nAYl5ERJTyApIMMwPlmJg1LQ/TLsxC+cvHcaaxY0Tv9eGnLWjrFFEw1dnn+fzcNNTWj+wziIhGQlvRXYJIslYzygyUiYgotbGYV+wIgoCvL/4MTEY9/vuP741ob+WaD+phNOhwxUXZfZ6f5LCjrqEjZtO8iYiGiiNHDGh26rWfU6+JiCi1cXuo2MpKN+Priy7HJy4Pnnrhg2EFsoqq4vA/3bjiohxYTIY+n5PvSENAUlDf4h1pk4mIhkVb0V2CaHHqtcVkYEaZiCiF7dmzB4sWLcK8efOwY8eOXuePHj2KW265BUVFRVi9ejUkKXhztK6uDsuXL8eCBQtw9913o6MjOP21ra0Nd911FxYuXIjly5ejvr4+4v08Hg9uvPFGvPnmm/Hv3BAEJJlrlGNs5iW5uPmGyTj03lns/r+Phvz6E7VtaPEEcOVljn6fk+9IAwBOvyaihNFWdJcgWp167RdlTnkiIkpBLpcLmzdvxs6dO1FZWYldu3bh+PHjEc8pKyvD2rVrsW/fPqiqivLycgDA+vXrsWzZMlRVVWH69OnYunUrAGDLli0oKCjACy+8gNtuuw0bNmyIeL9HHnkEbW2x2zooFmRFgSSrnHodB8WzL8QXPp+Pqrc+wbOvfAR1CH8v1Bxzw6AX8Nkpuf0+Jz83FCh7RtxWIqLh4MgRA9qceh2cCuVnVpmIKOUcPHgQs2bNQmZmJmw2G4qKilBVVRU+X1tbC5/Ph5kzZwIAli5diqqqKoiiiOrqahQVFUUcB4ADBw6gpKQEAFBcXIxXX30VoigCAPbu3Yu0tDRcdtllo9jLwQXE4BpaTr2OPUEQsOzGSzFn5kT8+dAp7Hjxn1HdXFdVFYeP1WPahdmwWfqedg0E/w7JzbDgU2aUiShBtBXdJYgoa3DqtTn4R4WX65SJiFKO2+2Gw9E9rdXpdMLlcvV73uFwwOVyobm5GXa7HQaDIeL4ua8xGAyw2+1oampCXV0d/ud//gf333//aHRtSAJdxabMzCjHhU4noLToMiy4+ny8/Lda/PbPRyErAxf4OlHXhsY2H668rO9q1z1NcthR28BAmYgSo/9beRQ1LWaUrV0ZZa5TJiJKPX1NgxUEYdDzg72uL6tXr8batWthsViG0VIgJ8c+rNdFQ9YFg6yc7DQ4HOn9Pm+gc6lqNPv07/8yE7nZNjxd9QFUQUDZl6+EsZ8s/pN/fB9pViPmz56MNKtxwPe95IIsHDnRiMwsGwBtfk+ANvulxT4B2uyXFvsUKwyUY0CUFRi1llE2BQc4BspERKknLy8PNTU14cdutxtOpzPifENDQ/hxfX09nE4nsrOz4fF4IMsy9Hp9+DgQzEo3NDRg/PjxkCQJHo8Hzc3NOHHiBFavXg0A+OSTT7BmzRo88sgjmDVrVlRtbWz0QFHiUw/jbNf6Vr9PRH19e5/PcTjS+z2XqhLRpy/MnAhZlPHMSx/ioScP4j+WXtFr/+qP6lrx1vtnsbTwInR6fOj0+AZ8z6w0I2RFxZFjbnx+2gTNfU8Af/9SiRb7pbU+6XRCTG++aiu6SxBJgxnl7kCZU6+JiFLN7NmzcejQITQ1NcHr9WL//v0oLCwMn8/Pz4fZbMbhw4cBABUVFSgsLITRaERBQQH27t0bcRwA5syZg4qKCgDBNckFBQWYOnUqXnnlFVRWVqKyshLTp0/Ho48+GnWQHG+hqdcmjY3RyWreVefhawun4r2TTdj0v39Hi8cfcb7itY9htxpxY8GkqN4vPzf4B29tAwt6EdHo48gxQoqiQlZUDWaUg5MNvH5mlImIUk1eXh5WrlyJ0tJSLFmyBMXFxZgxYwZWrFiBI0eOAAA2bdqEjRs3YuHChfB6vSgtLQUArFu3DuXl5Vi0aBFqampw7733AgDuuecevP3221i8eDF27tyJhx56KFHdi1pADI5h3B5q9Nzw2Yn49yXTcdrtwcNPVeOtoy40tvqwbc/7eO/jJiyadUG/eyefy5llBQC4m7mXMhGNPk69HiFRDt6t1lpG2WpmRpmIKJWVlJSEq1SHbNu2Lfzz1KlTsXv37l6vy8/Px/bt23sdz8zMxJNPPjngZ/b1ukTyh6pes5jXqLryMiecWTb84rl38GTlewAAnSDgS9ddiHlXRZdNBgCzUY+sdDNcTQyUiWj0MVAeIbFrWpdBY4GyhcW8iIgoxYUyymZuDzXqznPasWHFLJyoa8M/T7dg2uRsTJ4wbsjvk5dlhbulMw4tJCIaGAPlEZJCGWXNTb1mRpmIiFJbQApNvdbWGJ0qDHodLj0vE5eelzns93Bm2fC3f9bHrlFERFHiyDFCoYyy1qZeGw066ASBGWUiIkpZgfDUa2aUU1VelhUerwiPV0x0U4hojNFWdJcAWg2UBUGA1ayHj8W8iIgoRflDxbw49TplObv2UD7DytdENMq0Fd0lQGjqtUFjU6+B4PRrTr0mIqJU1V31Wntj9FiR11X5+kxDR4JbQkRjDUeOEdJqRhkIFvTycuo1ERGlqICkQK8TNHkze6xwdAXKdQyUiWiUxXXk2LNnDxYtWoR58+Zhx44dvc4fPXoUt9xyC4qKirB69WpIUjB7WVdXh+XLl2PBggW4++670dER/Mfxo48+wrJly3DTTTfh9ttvx9GjR+PZ/KiEA2UNDsIWMzPKRESUuvyizPXJKS60RRQzykQ02uIW3blcLmzevBk7d+5EZWUldu3ahePHj0c8p6ysDGvXrsW+ffugqirKy8sBAOvXr8eyZctQVVWF6dOnY+vWrQCANWvWYMWKFaisrMS9996L73//+/FqftRC+yhrbXsoIJhRZjEvIiJKVQFR4bRrDcjLsqKunmuUiWh0xW30OHjwIGbNmoXMzEzYbDYUFRWhqqoqfL62thY+nw8zZ84EACxduhRVVVUQRRHV1dUoKiqKOA4At912GwoLCwEAl112Gc6cOROv5kdN0nJG2aRnoExERCkrIMncQ1kDnFlWnGlkRpmIRlfcoju32w2HwxF+7HQ64XK5+j3vcDjgcrnQ3NwMu90Og8EQcRwIBs16fXDAe/zxx3HjjTfGq/lRC2WUtblGWQ+vn1OviYgoNTGjrA15WTa0egLo9PFvEiIaPYZ4vbGqqr2OCYIw6PloXvfYY4/hH//4B37/+98PqU05OfYhPT8alpPNAIA8ZzocOWkR5xyO9Jh/3mjKzrQhIDaMej9S/bolAq/Z8PC6DR2vGaWSANcoa4Kzq6CXu6UTF44fl+DWENFYEbdAOS8vDzU1NeHHbrcbTqcz4nxDQ0P4cX19PZxOJ7Kzs+HxeCDLMvR6ffg4AEiShO9///twuVz4/e9/j/T0of3B1tjogaL0DsRHoqmlEwDQ1uqFXlHCxx2OdNTXt8f0s0abKsvo9Etwu9siblbEkxau22jjNRseXreh4zXrptMJcbn5SrEVEGWYNDjja6zJ69pL2dXkZaBMRKMmbqPH7NmzcejQITQ1NcHr9WL//v3h9cUAkJ+fD7PZjMOHDwMAKioqUFhYCKPRiIKCAuzduzfiOAD8+Mc/hsfjwW9/+9shB8nxouXtoWxmI1QVnH5NREQpyS8pzChrQGiLKHdzZ4JbQkRjSVwzyitXrkRpaSlEUcStt96KGTNmYMWKFfjOd76DK664Aps2bcKaNWvQ0dGBz3zmMygtLQUArFu3DqtWrcIvf/lLTJgwAT/72c/Q1NSEHTt2YNKkSbjtttvCn1NZWRmvLkRFy8W8MuwmAEBrRwA2izHBrSEiIhoaTr3WBrNRj5wMC1zN3kQ3hYjGkLgFygBQUlKCkpKSiGPbtm0L/zx16lTs3r271+vy8/Oxffv2Xsfff//92DdyhLq3hxqdqcmjKSMtGCi3dQQw4Zz110RERMkuICowa3DG11g0ITcNbgbKRDSKOHqMkCgp0OsE6HXau5Tj0rozykRERKkmIDGjrBUTc+1wceo1EY0i7UV3o0yUFBg0OO0a6M4oM1AmIqJUxO2htGNibhraO0VuEUVEo4ajxwhJsqLJQl4AkGY1QicIaGOgTEREKUZV1a6q18woa8GE3OASMHcLs8pENDq0GeGNIlHSbqCsEwSMSzMyo0xERClHkhWoADPKGjHREdyOjeuUiWi0cPQYIVFWNFnxOiQjzcyMMhERpRy/GCy2yTXK2jA+J7SXMjPKRDQ6tBvhjRJJUmDQaEYZCBb0YkaZiIhSTUCUAQS3FqLUZzEZkJVuZkaZiEaNdiO8USJKWs8om5hRJiKilBOQujLKGr6ZPdY4M63cS5mIRg1HjxESNVzMCwhmlNs6AlBUNdFNISIiilooo8yp19qRl23lFlFENGq0G+GNEklSYNALiW5G3GSkmSArKrdjICKilBLoWqPMqdfakZdl69oiSkx0U4hoDGCgPELBjLJ2B+EMO/dSJiKi1OMTgzd4WfVaO0JbRNU1MqtMRPHH0WOEtLw9FACMswUD5TaPP8EtISIiil5oJpTNbEhwSyhWJoYC5YaOBLeEiMYC7UZ4o0SUVW1PvQ5llDuZUSYiotQRDpQtxgS3hGIld5wFJoOOgTIRjQoGyiMkSbK2M8ppoYwyA2UiIkodHV3rWNMszChrhU4nYHyOjYEyEY0K7UZ4oyQ49Vq7a5RtZgMMeoFrlImIKKV0+iQY9DpWvdaY/Nw01DUyUCai+GOgPEJan3otCAL3UiYiopTT4ZOYTdagiblpaGrzw+vnbhxEFF8MlEdI68W8gOD0a2aUiYgolXT6RNgYKGvOxJxQ5WtmlYkovrQd4cWZqqqQZAVGvbYvY0aamRllIqIUs2fPHixatAjz5s3Djh07ep0/evQobrnlFhQVFWH16tWQpGCGrq6uDsuXL8eCBQtw9913o6MjGJC0tbXhrrvuwsKFC7F8+XLU19cDANxuN77xjW/gpptuws0334xDhw6NXicHEMwos5CX1rDyNRGNFm1HeHEmySoAaD6jnD3ODHeLF4qqJropREQUBZfLhc2bN2Pnzp2orKzErl27cPz48YjnlJWVYe3atdi3bx9UVUV5eTkAYP369Vi2bBmqqqowffp0bN26FQCwZcsWFBQU4IUXXsBtt92GDRs2AAAee+wxzJ07F5WVlfjpT3+K++67D7Isj26H+9Dpk5hR1iBHphUGvQ5nGriXMhHFl7YjvDgTJQUANJ9RviAvHb6ADHezN9FNISKiKBw8eBCzZs1CZmYmbDYbioqKUFVVFT5fW1sLn8+HmTNnAgCWLl2KqqoqiKKI6upqFBUVRRwHgAMHDqCkpAQAUFxcjFdffRWiKGL+/Pnh4xdccAH8fj86OxMfxHT4RK5R1iCdTsCEHBunXhNR3Gk7woszUQ4GygaNZ5QvGJ8OADh5ti3BLSEiomi43W44HI7wY6fTCZfL1e95h8MBl8uF5uZm2O12GAyGiOPnvsZgMMBut6OpqQnz589HRkYGAOA3v/kNLr/8cqSnp8e9j4MJZpQ59VqLJuamobbek+hmEJHG8VbrCPgCwfVcFpO2t56YmJsGg16HU2fbMesz4xPdHCIiGoTax1IZQRAGPT/Y686l03XfKH7qqaewa9cuPP3000Nqa06OfUjPj4aiqPAGJDiy0+BwDB60R/OcVKPlPn3mohy8+b4LJqsJGXZzgls1clr+rrRGi/3SYp9ihYHyCHT6goGy1u9YG/Q6nOe049TZ9kQ3hYiIopCXl4eamprwY7fbDafTGXG+oaEh/Li+vh5OpxPZ2dnweDyQZRl6vT58HAhmpRsaGjB+/HhIkgSPx4PMzEwAwXXKr7zyCnbs2IHx44d2Q7Wx0QNFiW0NjA6fCFUFoCiorx947HI40gd9TqrRep8c6cHguObdOsyYkpvIZo2Y1r8rLdFiv7TWJ51OiOnNV23PGY6zcKBs1v79hgvHp+OUq50FvYiIUsDs2bNx6NAhNDU1wev1Yv/+/SgsLAyfz8/Ph9lsxuHDhwEAFRUVKCwshNFoREFBAfbu3RtxHADmzJmDiooKAMDevXtRUFAAo9GIp556Cm+++SaeeeaZIQfJ8dLRNT5zjbI2XTA+HQKAk2e08wc+ESUfBsoj0OETAYyNgfiC8enw+mXUs6AXEVHSy8vLw8qVK1FaWoolS5aguLgYM2bMwIoVK3DkyBEAwKZNm7Bx40YsXLgQXq8XpaWlAIB169ahvLwcixYtQk1NDe69914AwD333IO3334bixcvxs6dO/HQQw9BVVU88cQTaGpqwle+8hXcdNNNuOmmmyLWQydCZ9f4zKrX2mQ1GzA+x4aPz7B2ChHFD0eQEej0j42p10AwowwAJ8+2Iy/bluDWEBHRYEpKSsLVqEO2bdsW/nnq1KnYvXt3r9fl5+dj+/btvY5nZmbiySef7HW8uro6Bq2Nre6MsvbH57Fq8oRxePfjJqiqOuA6eiKi4WJGeQS61yhr/35DsKCXwHXKRESU9MbS0qixavKEcWjrCKC53Z/ophCRRjFQHoEOnwiDXoBJ49tDAcGCXufnpeP9U019VkUlIiJKFh2ceq15F04IznTj9GsiihftR3hx5PVJsJkNY2bKz3XTx+MTlwcnOCgREVES6+TUa80732mHXifgJGe6EVGcMFAegQ6fNCbWJ4fMmjYeFpMeLx/+NNFNISIi6leHT4ReJ8Bk5J85WmU06DHJacdHta2JbgoRaRRHkBHo9EtjalqX1WzAdVdMQPUHbrR2BBLdHCIioj51+iSkWcbOjK+xaur5mThe2wa/KCe6KUSkQQyUR6DTJ46pQBkAvvD5fEiyiv/7G7PKRESUnMbajK+xatqF2ZBkBR+ebkl0U4hIgxgoj0BH1xrlsWRCThquvMyBqrc+QVObL9HNISIi6qXTJyJtjN3IHosuOS8TBr2A9042JbopRKRBDJRHIDi1a+zdsb79CxdDVYFdLx9PdFOIiIh6YUZ5bDAb9bhkUibe+7g50U0hIg1ioDxMqqqi0ze21iiH5GZYsfjaC1D9gRtHeReXiIiSDDPKY8e0ydn4tN6DVg/3Uyai2GKgPEy+gAxFVcdkoAwAC685H7kZFuz4y4eQZCXRzSEiIgobqzeyx6JpF2YDAN4/yawyEcUWA+Vh8vrH9h6NRoMe/3rjJahr6MBL3C6KiIiShBKe8TU2x+ex5rw8O9JtRvzjo4ZEN4WINIaB8jB1+IKB8lgr5tXTzItzccVFOah8/WNOeSIioqTg88tQAU69HiN0goCCqU78/cOGcBKDiCgWGCgPU6dPBIAxPbVLEAT8642XICAq+NPBU4luDhEREdq9AQCA3cqM8lgxe9p4iJKCmmPuRDeFiDSEgfIwdYYyymM4UAaA8dk2XD9jPF75Ry0aW7ldFBERJVZDS3Asys2wJLglNFoumjgOziwr/vqeK9FNISINYaA8TOGp11wDheLZF0JVgT8dOpnophAR0RjnbvECAByZ1gS3hEaLIAi4dtp4fHCqGU1tvGlPRLHBQHmYOsPFvMZ2RhkIbhc1Z+ZEvP7OGbiaOxPdHCIiGsPqm70w6HXITDcnuik0imZNy4MK4LV3ziS6KUSkEQyUhym0RtlqYqAMBLPKBr0Ozx74KNFNISKiMay+xQtHpgU6QUh0U2gU5WXZMPPiXPyl5jSLehFRTDBQHqYOnwSr2QCdjgMxAGTazVhwzfmoOVaP45+2Jro5REQ0RrlbvJx2PUaVXHchOnwSXv4bt60kopFjoDxMnT5pTG8N1ZcFV5+PDLsJ//vyh1AUNdHNISKiMUZVVbhbvHAyUB6TJk8YhysuysG+t07DF2BWmYhGhoHyMHX6RK5PPofZpMe//L+LcaKuDZWvf5zo5hAR0RjT7hXhD8jMKI9hX7ruQni8Iipe498hRDQyDJSHqdMvjfmtofpy7fTxuP6KCdhz8CTe+agh0c0hIqIxpL65q+J1FgPlsWpKfgb+3+fy8WL1afzzdEuim0NEKYyB8jB1+iRuDdWPL8+/FOc57Xji+Xfx2jt1iW4OERGNEaGtoTj1emz7l7lTkJNhwW//fDRcfJWIaKgYKA9Th09kRrkfJqMe37t9Ji7Oz8Dv9n6Arc8fQV1DR6KbRUREGlffFSjnZlgS3BJKJIvJgG8WfwaNbT78fPc78ItyoptERCmIgfIwyIoCj1eC3cqMcn/GpZnwvdtn4uYbJuPIiSas/fWb+NWe93C2ifssExFRfNQ3e5GVbobJqE90UyjBLj0vE3d9aRqO17biieeOsLgXEQ0ZA+VhqK3vgCQrON9pT3RTkppOJ6Dkusl47O5rsWDW+fjbP+uxettfsW3P+3A1M2AmIqLY4tZQ1NNVU524c8FUvHeyCT/c/jc0tHoT3SQiSiEMlIfh4zNtAIDJE8cluCWpId1mwm3/72I89q3ZKLrqfBw+5sbqX72J8v87zulQREQUM/UtXjgyOe2aut3w2YlYedtn0djmw/rfVeOto65EN4mIUgQD5WE4UdeGNIuBxUKGaFyaCf/yhYvx429di+uuGI+qNz/But+8haOnmhPdNCIiSnGu5k60eAI4z5me6KZQkpl+UQ4e+moBnFk2PFn5Hn71x/fQwSJfRDQIBsrD8PGZNkyeOA6CICS6KSkpw27G1xZdjrI7ZgIAfvLM3/GbP70PN6djExHRMNV84AYAXHmpI8EtoWSUl23Dg1/5PJbcMBlvHXXjod+8hXdPNCa6WUSUxFi2eYh8AQm1DR34PAfiEbv8wmw8/I2rUfn6x3ix5jQOvncWV10+Hpfkj8NFE8chL8vKLbiIiCgq1R+4MWXiOOSw4jX1Q6/T4UvXTcYVF+Xg1396Hz8r/weu+Uwe/mXuxchKNye6eUSUZBgoD9Gps+1QVeAirk+OCZNRj9vmXox5V52HF6tP4/CHDXjr/bPh83arEc4sK5yZ1uD/ZlnhzLLBmWlFus0YkdVXVRWyokKWVciKApNRD4OekyaIiLTO3dyJT1we3P6FixPdFEoBkyeMww++djVe+Osp/OnQSfztn/X44pWTcOOVk5A9jjdaiCgoroHynj178Mtf/hKiKOLOO+/E8uXLI84fPXoUa9asgcfjQUFBAdavXw+DwYC6ujqUlZWhsbERkydPxqZNm5CWloa2tjbcd999OH36NLKzs7FlyxY4HKOb2T0RKuQ1gYFyLGXazbht7sW4+7aZeP9DN067O1Df4oW7uROuZi8+/LQVb77vgtrjNXqdAINBBwGAJKuQZSXiPABYTHrkZlgxPseGyRPScdGEcbhw/DiYTdw6hIi0bbTG4EAggNWrV+Pdd9+FxWLBpk2bMGXKlFHta3XXtOuCy5yj+rmUuowGHb50/WRcO308Kl47gX1vfoJ9b32CaZOzMX1yDi7Oz4Azy4o0i4FL7YjGqLgFyi6XC5s3b8Zzzz0Hk8mEO+64A9dccw0uvrj7bm9ZWRkeffRRzJw5Ew8++CDKy8uxbNkyrF+/HsuWLcPixYvxxBNPYOvWrSgrK8OWLVtQUFCAX/3qV6ioqMCGDRuwZcuWeHWhTx/XtcGRaUG6zTSqnztWCIIQzBhn2XqdEyUFDa1euJu9cLd40eoJQJIVKKoKo14Hg14Hg16AQa+DTifAL8rwdIqob/Hi5Jm28Po1QQDyc+3B6d3ZVtgtRtitRqRZjTAadMEAXK+DIAQ/M/R/qqpCr9chzWpEdroZVnP0//n4AzJOuz1o9vjR4RORZgm+xySnHeZh7vepqCr8ARltHQGIkgyjgcE/EQWN5hi8fft2WK1WvPDCC6iursaqVavwhz/8YVT7+8EnLZx2TcPiyLRiRck03HTDRXjtH3V466gL755oCp83GnSwWQxIsxhhsxhgMxuQZjHAZjZ2HTfAGjpvNoSfa7caYTLqGGQTpbC4BcoHDx7ErFmzkJmZCQAoKipCVVUV/vM//xMAUFtbC5/Ph5kzZwIAli5discffxy33XYbqqur8cQTT4SPf/nLX0ZZWRkOHDiAHTt2AACKi4vx8MMPQxRFGI2jt441w27GNTlpo/Z51M1o0GFCThomDPP6t3UG8HFdG07UteFEV+Dc6ZeG3R6rWY/sdAtyMyxwZFrhyLIiy26GwaCDJCloavejtt6Dj8+0o7bBA/XcdDcAnSBgkjMNF00Yh0lOOzLtZpiNeqiqCqVrKnmnT4LHK8LjFdHaEUBjqw+NbT40tfkgyd1vmmYxINNuRqbdhKx0C7LHmZE9zoJMuyl4I8EQvJmgEwSExm1VBfyijE6fhA6fiE6fhHavCF/XdTEadcgdZ0FOhgU54yzI6rpBwIGfKLmN5hh84MAB3HPPPQCAq666Cs3Nzairq8PEiRNHrb+3zpkCq5k3C2n4nJlW3DJnCm6ZMwVNbT6cPNuOhhYvWjwBdPpFdPgkdPoktHoCqGvoQKdPgtcv9ZrJ1pNBLyDNaoTdErwZbzXpYTbpYTEZYDHpYTHpkZlhhcfjh4rgEjJFUYOz5BQFsqzCoNfBaNDBZNTBZNTD3hWwp1mNSOsKyiOXmakQZRWiJHff7JcVSF0/C4IAiznYBqtJD5vFALNRz3GdqA9xC5TdbnfEtGin04l33nmn3/MOhwMulwvNzc2w2+0wGAwRx899jcFggN1uR1NTE/Ly8qJqk0438n8EvlJ0WdTPjcXnjUXxum6ZdjM+d6kDn+sqxKaqKvyiAq9PRIc/OABKsgpFUSDJKlSoMOr1MBh0MOqDwaUsq/D4RLR6AmjtCKDF40djmx8fnG7BO31Uz7SZDbhwYjoKZ07AJEc6sseZYTHp4fXLaGrz4XS9B6dd7Thxpg3vD7JNll4nwGY1IssezERn2c1IsxqRnm5Gc4sX7Z0BtHWIaOsMwNXixYeftgw4gPdHAGA26SFAQECSISuR76LXC0i3mpBmMUKnCwbbKgCogBr8f8EBH+GDwfcVgpl6vV6AQSdA3zUDQK8L/W/ovA49x+vQDQYV4R96nIw8p/Y63nf/IAAWswH+rhsCAoQeJxHxc+hczzYJiDwHocdLe7yu198dfb1Hz2NdD3q+l3DOK/p6Hfp4XUT70Pf1675eva+U2sc1tFqN8HoDva5zzx/VKN+rdyP6en7fbVbV8BkoqtrjwwEIwX9DdELwdwpC8L8dnRC8MaXT6fDZKTnIHGHhnmT/9300x+C+3uvs2bNRB8qxuJaTR1g3JNm/z+Fgn4YvN9OK3Ci2AA3N7PL5JXT6ZXj9InwBGZ1+CR0+Cd7QTWi/hE6/BH9ARrtXREObHwFRRkCU+3xffdcYqRMEyEowwI0nvU6AtSsbbjUH/89mDgXQXWOJgOC/rz3Gm9Dx0LgKBM9brEb4fGLwOT2Oo+u9eh4PJQUUFVAVQFFUKAjeMFCU7qRBMIGAiONK+HVK8D0AKLIKSem+0SArCmQFkOXgNQz9HWLsmn1oNOhg0AkwGvUwdt2UMOh1MOp1PQd76AQBVkuoX13ju4CIn8PXRQiO56HrErpiPfuNiPPBg+eOeUMbowf+O+jcsTn00Go1odMb6HGiv8/pr23nNOyc34fIaxXsf8/fBV3XQZ0g4LMX5yDTnlxjc9wC5b7+WDq38FJf5wd73bl0uuiLNWVljW4mOCfHPqqfpxW8bkREI5PoMTiZx+a+aHHcYZ+IiEYmbiWB8/Ly0NDQEH7sdrvhdDr7PV9fXw+n04ns7Gx4PB7IshxxHAjeEQ+9RpIkeDye8LQyIiIiChrNMdjpdKK+vr7XexEREaWyuAXKs2fPxqFDh9DU1ASv14v9+/ejsLAwfD4/Px9msxmHDx8GAFRUVKCwsBBGoxEFBQXYu3dvxHEAmDNnDioqKgAAe/fuRUFBwaiuTyYiIkoFozkGz5kzB5WVlQCAmpoamM3mUV2fTEREFA+C2tc8qxjZs2cP/vu//xuiKOLWW2/FihUrsGLFCnznO9/BFVdcgQ8++ABr1qxBR0cHPvOZz2Djxo0wmUyora3FqlWr0NjYiAkTJuBnP/sZMjIy0NLSglWrVuH06dNIT0/Hpk2bMGnSpHg1n4iIKGWN1hjs9/vx0EMP4d1334XJZMKjjz6KadOmJbr7REREIxLXQJmIiIiIiIgo1cRt6jURERERERFRKmKgTERERERERNQDA2UiIiIiIiKiHhgoExEREREREfXAQJmIiIiIiIioBwbKcbBnzx4sWrQI8+bNw44dOxLdnKTyi1/8AosXL8bixYvx2GOPAQAOHjyIkpISzJ8/H5s3bw4/9+jRo7jllltQVFSE1atXQ5KkRDU7Kfz4xz/GqlWrAPR/berq6rB8+XIsWLAAd999Nzo6OhLZ5IR6+eWXsXTpUixYsACPPvooAP6uRaOysjL83+iPf/xjAPx9I21K5bG6tLQUixcvxk033YSbbroJ//jHP/rtT3//7iULj8eD4uJifPrppwCG/u90sv47dG6/HnjgAcyfPz/8nb344osAUmdcisXfb8n4XfXVr1T/rn7+859j0aJFWLx4MX73u98B0MZ31Ve/4v5dqRRTZ8+eVefOnas2NzerHR0daklJifrhhx8mullJ4Y033lBvv/121e/3q4FAQC0tLVX37NmjzpkzR/3kk09UURTVr3/96+qBAwdUVVXVxYsXq3//+99VVVXVBx54QN2xY0cCW59YBw8eVK+55hr1+9//vqqq/V+bu+66S/3Tn/6kqqqq/uIXv1Afe+yxhLQ30T755BP1+uuvV8+cOaMGAgH1X//1X9UDBw7wd20QnZ2d6lVXXaU2Njaqoiiqt956q/rGG2/w9400J5XHakVR1Ouuu04VRTF8rL/+eL3efv/dSwZvv/22WlxcrE6bNk09ffr0gO1NpX+Hzu2XqqpqcXGx6nK5Ip43nP4mQqz+fku276qvfu3fvz+lv6s333xTveOOO1RRFFWv16vOnTtXPXr0aMp/V33166OPPor7d8WMcowdPHgQs2bNQmZmJmw2G4qKilBVVZXoZiUFh8OBVatWwWQywWg0YsqUKTh58iQuuOACnHfeeTAYDCgpKUFVVRVqa2vh8/kwc+ZMAMDSpUvH7HVsaWnB5s2b8a1vfQsA+r02oiiiuroaRUVFEcfHohdffBGLFi3C+PHjYTQasXnzZlitVv6uDUKWZSiKAq/XC0mSIEkSDAYDf99Ic1J5rD5x4gQEQcCKFSvwpS99CU8//XS//XnnnXf6/HcvWZSXl2PdunVwOp0A0G97U23cO7dfnZ2dqKurw9q1a1FSUoLHH38ciqIMub+JEou/35Lxu+qrX3V1dSn9XV199dX4/e9/D4PBgMbGRsiyjLa2tpT/rvrql9lsjvt3ZRiFvo0pbrcbDocj/NjpdOKdd95JYIuSxyWXXBL++eTJk9i7dy++8pWv9LpeLper13V0OBxwuVyj2t5k8dBDD2HlypU4c+YMgN6/Y6Fr09zcDLvdDoPBEHF8LDp16hSMRiO+8Y1voL6+HnPnzsUll1zC37VB2O123HPPPVi4cCEsFguuvvpqGI1G/r6R5qTyWN3W1oZrr70WP/jBD+Dz+VBaWoqFCxf22Z+++plM/51u2LAh4nF/7U21ce/cfjU2NmLWrFl4+OGHYbPZ8G//9m/YvXs3bDZbSoxLsfj7LRm/q776tXPnTrz11lsp+10BgNFoxOOPP47f/va3WLBggWb+uzq3X7Isx/2/K2aUY0xV1V7HBEFIQEuS14cffoivf/3r+P73v4/zzz+/13lBEHgdu/zhD3/AhAkTcO2114aP9XdteM26ybKMQ4cO4Sc/+QnKy8tx5MiR8DqxnnjdIn3wwQd49tln8X//9394/fXXodPp8MYbb/R6Hq8bpbpU/v393Oc+h8ceeww2mw3Z2dm49dZb8fjjj/d6Xir+dzrU8S1V+nfeeefhiSeeQE5ODqxWK77yla/glVdeSbl+jeTvt2TtExDZr4suukgT39V3vvMdHDp0CGfOnMHJkyd7nU/V76pnvw4dOhT374qBcozl5eWhoaEh/Njtdoen3hBw+PBh3Hnnnfje976Hm2++ud/rde7x+vr6MXkd9+7dizfeeAM33XQTHn/8cbz88sv4wx/+0Oe1yc7OhsfjgSzLEcfHotzcXFx77bXIzs6GxWLBF7/4Rbzxxhv8XRvE66+/jmuvvRY5OTkwmUxYunQp3nzzTf6+keak8lhdU1ODQ4cOhR+rqor8/Pyo/n1L9n4O9W+CVPl36NixY9i3b1/4saqqMBgMKfU30Ej/fkvW7+rcfqX6d/XRRx/h6NGjAACr1Yr58+f3GsdT8bvqq1979+6N+3fFQDnGZs+ejUOHDqGpqQlerxf79+9HYWFhopuVFM6cOYP/+I//wKZNm7B48WIAwGc/+1l8/PHHOHXqFGRZxp/+9CcUFhYiPz8fZrMZhw8fBgBUVFSMyev4u9/9Dn/6059QWVmJ73znO/jCF76AjRs39nltjEYjCgoKsHfv3ojjY9HcuXPx+uuvo62tDbIs47XXXsOCBQv4uzaIqVOn4uDBg+js7ISqqnj55Zdx9dVX8/eNNCeVx+r29nY89thj8Pv98Hg8eP755/GTn/ykz/70N8Ymq6H+TZAq/w6pqoof/vCHaG1thSiK2LVrF+bNm5cyfwPF4u+3ZPyu+upXqn9Xn376KdasWYNAIIBAIICXXnoJd9xxR8p/V33166qrror7dyWofeWhaUT27NmD//7v/4Yoirj11luxYsWKRDcpKTz66KN49tlnI6br3HHHHbjwwguxceNG+P1+zJkzBw888AAEQcAHH3yANWvWoKOjA5/5zGewceNGmEymBPYgsZ577jm89dZb+NGPftTvtamtrcWqVavQ2NiICRMm4Gc/+xkyMjIS3fSE2L17N5566imIoojrrrsOa9aswZtvvsnftUH86le/wnPPPQej0YgrrrgC69atw8cff8zfN9KcVB6rt2zZgn379kFRFCxbtgxf/epX++3PoUOH+vx3L5l84QtfwO9//3tMmjSp3/am4rjXs187duzAjh07IEkS5s+fj/vuuw9A/99PMo1Lsfr7Ldm+q/76pShKyn5XAPD444+jqqoKer0e8+fPx7e//W1N/HfVV7/i/d8VA2UiIiIiIiKiHjj1moiIiIiIiKgHBspEREREREREPTBQJiIiIiIiIuqBgTIRERERERFRDwyUiYiIiIiIiHpgoEw0Brz55psoLi7Gz3/+c1RUVCS6OURERBQDTU1NuOyyyxLdDCJNMiS6AUQ0eu65555EN4GIiIiIKOkxUCYaQ1atWoVLLrkE3/jGN3DFFVfgrrvuwhtvvAG3243S0lLceeedAIA//OEPeOaZZ6AoCjIzM7F27VpMmTIFNTU1+NGPfgRFUQAA//Zv/4aioqIE9oiIiCj5LVmyBPfffz9mz56NP//5z1i1ahWqq6thsViwZs0aXHLJJThy5Ag++OADCIKAG264Ad/97ndhMBgwffp0fPGLX8QHH3yATZs24cyZM9i8eTOsViumT58e/oz6+np8//vfR3NzMwBgzpw5uPfeexPUY6LUx6nXRGNUIBBAVlYW/vd//xePP/44fvrTn8Lv9+Ott95CRUUFduzYgYqKCnzzm9/Et7/9bQDAf/3Xf+FrX/sannvuOfzwhz/EX//61wT3goiIKPndeOONeO211wAAr732GjIyMlBTUwNFUXDgwAG8+uqryMzMxJ49e/Dss8/i2LFj+O1vfwsAEEURc+fOxb59+zBhwgQ8+OCD+K//+i8899xzyM/PD39GeXk5Jk2ahOeffx47duzAqVOn0N7enpD+EmkBA2WiMeyLX/wiAGDatGkIBALo7OzEgQMHcOrUKdxxxx246aab8JOf/AStra1oaWnBwoUL8fDDD+N73/se3nvvPXz3u99NcA+IiIiS37x58/Dqq68CAGpqanDnnXfijTfewD/+8Q+cf/75eP/99/HlL38ZgiDAZDLhjjvuCD8fAAoKCgAAhw8fxqWXXoqLL74YAHD77beHn3PDDTdg//79WLFiBXbt2oXvfe97SE9PH8VeEmkLA2WiMcxsNgMABEEAAKiqCkVRcNNNN6GyshKVlZV4/vnn8eyzzyIjIwN33HEH/vjHP+K6667D66+/ji996Uu8W01ERDSIyy67DKIo4qWXXsIFF1yAuXPn4o033sDLL7+M+fPnh5c0hSiKAkmSwo9tNhuA4Hitqmr4uMHQvYpyxowZeOmll3D77bejtrYWt912G/72t7/FuWdE2sVAmYgiXHfddfjzn/8Mt9sNAHjmmWfw1a9+FQBwxx134OjRo1i6dCkeeeQRtLW1obW1NZHNJSIiSgk33ngjNm3ahOuuuw5TpkyBx+PBnj17UFRUhOuvvx47duyAqqoIBAIoLy/H7Nmze71HQUEBjh8/jg8++AAA8Nxzz4XPbdq0CVu3bsWNN96I1atX4+KLL8bJkydHq3tEmsNiXkQU4YYbbsCKFSvw9a9/HYIgwG634xe/+AUEQcB9992HH/7wh9iyZQt0Oh3+8z//E5MmTUp0k4mIiJLevHnz8Jvf/CYcAM+ePRvHjh3DhAkTsGbNGjz66KMoKSmBKIq44YYb8K1vfavXe2RnZ2PTpk247777YDQacdVVV4XPffWrX8WqVatQXFwMk8mEyy67DMXFxaPWPyKtEdSe8zeIiIiIiIiIxjhOvSYiIiIiIiLqgYEyERERERERUQ8MlImIiIiIiIh6YKBMRERERERE1AMDZSIiIiIiIqIeGCgTERERERER9cBAmYiIiIiIiKgHBspEREREREREPfx/jrqdJuApkxoAAAAASUVORK5CYII=\n",
      "text/plain": [
       "<Figure size 1152x432 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "fig, ax = plt.subplots(1, 2)\n",
    "\n",
    "sns.kdeplot(data=df.lines, ax=ax[0]);\n",
    "sns.kdeplot(data=df.words, ax=ax[1]);\n",
    "fig.set_size_inches(16, 6)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "46ce5a48",
   "metadata": {},
   "source": [
    "### Comparing scans of different shapes"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "id": "3ebda9b1",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1152x360 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "portrait = df[df.shape_class == 'portrait']\n",
    "landscape = df[df.shape_class == 'landscape']\n",
    "square = df[df.shape_class == 'square']\n",
    "\n",
    "fig, ax = plt.subplots(1, 2)\n",
    "\n",
    "clip_words = (0, 4000)\n",
    "clip_lines = (0, 1000)\n",
    "\n",
    "sns.kdeplot(data=df, x='lines', hue='shape_class', common_norm=False, clip=clip_lines, ax=ax[0]);\n",
    "sns.kdeplot(data=df, x='words', hue='shape_class', common_norm=False, clip=clip_words, ax=ax[1]);\n",
    "\n",
    "ax[0].set_title('Distribution of lines in scans')\n",
    "ax[1].set_title('Distribution of words in scans')\n",
    "\n",
    "\n",
    "fig.set_size_inches(16, 5)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "f1f6bc11",
   "metadata": {},
   "source": [
    "## Analysing scans with many lines and many words\n",
    "\n",
    "### Scans with high numbers of lines"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "id": "a9facf63",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<style type=\"text/css\">\n",
       "</style>\n",
       "<table id=\"T_8de24\">\n",
       "  <thead>\n",
       "    <tr>\n",
       "      <th class=\"blank level0\" >&nbsp;</th>\n",
       "      <th id=\"T_8de24_level0_col0\" class=\"col_heading level0 col0\" >doc_id</th>\n",
       "      <th id=\"T_8de24_level0_col1\" class=\"col_heading level0 col1\" >lines</th>\n",
       "      <th id=\"T_8de24_level0_col2\" class=\"col_heading level0 col2\" >doc_width</th>\n",
       "      <th id=\"T_8de24_level0_col3\" class=\"col_heading level0 col3\" >doc_height</th>\n",
       "      <th id=\"T_8de24_level0_col4\" class=\"col_heading level0 col4\" >width_class</th>\n",
       "      <th id=\"T_8de24_level0_col5\" class=\"col_heading level0 col5\" >height_class</th>\n",
       "      <th id=\"T_8de24_level0_col6\" class=\"col_heading level0 col6\" >iiif_url</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th id=\"T_8de24_level0_row0\" class=\"row_heading level0 row0\" >12420</th>\n",
       "      <td id=\"T_8de24_row0_col0\" class=\"data row0 col0\" >NL-HaNA_1.05.06_709_0006.jpg</td>\n",
       "      <td id=\"T_8de24_row0_col1\" class=\"data row0 col1\" >904</td>\n",
       "      <td id=\"T_8de24_row0_col2\" class=\"data row0 col2\" >5165</td>\n",
       "      <td id=\"T_8de24_row0_col3\" class=\"data row0 col3\" >6145</td>\n",
       "      <td id=\"T_8de24_row0_col4\" class=\"data row0 col4\" >medium</td>\n",
       "      <td id=\"T_8de24_row0_col5\" class=\"data row0 col5\" >wide</td>\n",
       "      <td id=\"T_8de24_row0_col6\" class=\"data row0 col6\" ><a target=\"_blank\" href=\"https://www.nationaalarchief.nl/onderzoeken/archief/1.05.06/invnr/709/file/NL-HaNA_1.05.06_709_0006?eadId=1.05.06&unitID=709\">https://www.nationaalarchief.nl/onderzoeken/archief/1.05.06/invnr/709/file/NL-HaNA_1.05.06_709_0006?eadId=1.05.06&unitID=709</a></td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_8de24_level0_row1\" class=\"row_heading level0 row1\" >7000</th>\n",
       "      <td id=\"T_8de24_row1_col0\" class=\"data row1 col0\" >NL-HaNA_1.05.06_176_0002.jpg</td>\n",
       "      <td id=\"T_8de24_row1_col1\" class=\"data row1 col1\" >890</td>\n",
       "      <td id=\"T_8de24_row1_col2\" class=\"data row1 col2\" >9934</td>\n",
       "      <td id=\"T_8de24_row1_col3\" class=\"data row1 col3\" >3844</td>\n",
       "      <td id=\"T_8de24_row1_col4\" class=\"data row1 col4\" >wide</td>\n",
       "      <td id=\"T_8de24_row1_col5\" class=\"data row1 col5\" >medium</td>\n",
       "      <td id=\"T_8de24_row1_col6\" class=\"data row1 col6\" ><a target=\"_blank\" href=\"https://www.nationaalarchief.nl/onderzoeken/archief/1.05.06/invnr/176/file/NL-HaNA_1.05.06_176_0002?eadId=1.05.06&unitID=176\">https://www.nationaalarchief.nl/onderzoeken/archief/1.05.06/invnr/176/file/NL-HaNA_1.05.06_176_0002?eadId=1.05.06&unitID=176</a></td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_8de24_level0_row2\" class=\"row_heading level0 row2\" >6999</th>\n",
       "      <td id=\"T_8de24_row2_col0\" class=\"data row2 col0\" >NL-HaNA_1.05.06_176_0001.jpg</td>\n",
       "      <td id=\"T_8de24_row2_col1\" class=\"data row2 col1\" >788</td>\n",
       "      <td id=\"T_8de24_row2_col2\" class=\"data row2 col2\" >9705</td>\n",
       "      <td id=\"T_8de24_row2_col3\" class=\"data row2 col3\" >3864</td>\n",
       "      <td id=\"T_8de24_row2_col4\" class=\"data row2 col4\" >wide</td>\n",
       "      <td id=\"T_8de24_row2_col5\" class=\"data row2 col5\" >medium</td>\n",
       "      <td id=\"T_8de24_row2_col6\" class=\"data row2 col6\" ><a target=\"_blank\" href=\"https://www.nationaalarchief.nl/onderzoeken/archief/1.05.06/invnr/176/file/NL-HaNA_1.05.06_176_0001?eadId=1.05.06&unitID=176\">https://www.nationaalarchief.nl/onderzoeken/archief/1.05.06/invnr/176/file/NL-HaNA_1.05.06_176_0001?eadId=1.05.06&unitID=176</a></td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_8de24_level0_row3\" class=\"row_heading level0 row3\" >7091</th>\n",
       "      <td id=\"T_8de24_row3_col0\" class=\"data row3 col0\" >NL-HaNA_1.05.06_186_0015.jpg</td>\n",
       "      <td id=\"T_8de24_row3_col1\" class=\"data row3 col1\" >704</td>\n",
       "      <td id=\"T_8de24_row3_col2\" class=\"data row3 col2\" >7857</td>\n",
       "      <td id=\"T_8de24_row3_col3\" class=\"data row3 col3\" >4948</td>\n",
       "      <td id=\"T_8de24_row3_col4\" class=\"data row3 col4\" >wide</td>\n",
       "      <td id=\"T_8de24_row3_col5\" class=\"data row3 col5\" >wide</td>\n",
       "      <td id=\"T_8de24_row3_col6\" class=\"data row3 col6\" ><a target=\"_blank\" href=\"https://www.nationaalarchief.nl/onderzoeken/archief/1.05.06/invnr/186/file/NL-HaNA_1.05.06_186_0015?eadId=1.05.06&unitID=186\">https://www.nationaalarchief.nl/onderzoeken/archief/1.05.06/invnr/186/file/NL-HaNA_1.05.06_186_0015?eadId=1.05.06&unitID=186</a></td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_8de24_level0_row4\" class=\"row_heading level0 row4\" >2168</th>\n",
       "      <td id=\"T_8de24_row4_col0\" class=\"data row4 col0\" >NL-HaNA_1.05.06_117_0043.jpg</td>\n",
       "      <td id=\"T_8de24_row4_col1\" class=\"data row4 col1\" >690</td>\n",
       "      <td id=\"T_8de24_row4_col2\" class=\"data row4 col2\" >5026</td>\n",
       "      <td id=\"T_8de24_row4_col3\" class=\"data row4 col3\" >4010</td>\n",
       "      <td id=\"T_8de24_row4_col4\" class=\"data row4 col4\" >medium</td>\n",
       "      <td id=\"T_8de24_row4_col5\" class=\"data row4 col5\" >medium</td>\n",
       "      <td id=\"T_8de24_row4_col6\" class=\"data row4 col6\" ><a target=\"_blank\" href=\"https://www.nationaalarchief.nl/onderzoeken/archief/1.05.06/invnr/117/file/NL-HaNA_1.05.06_117_0043?eadId=1.05.06&unitID=117\">https://www.nationaalarchief.nl/onderzoeken/archief/1.05.06/invnr/117/file/NL-HaNA_1.05.06_117_0043?eadId=1.05.06&unitID=117</a></td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_8de24_level0_row5\" class=\"row_heading level0 row5\" >7253</th>\n",
       "      <td id=\"T_8de24_row5_col0\" class=\"data row5 col0\" >NL-HaNA_1.05.06_220_0005.jpg</td>\n",
       "      <td id=\"T_8de24_row5_col1\" class=\"data row5 col1\" >660</td>\n",
       "      <td id=\"T_8de24_row5_col2\" class=\"data row5 col2\" >5102</td>\n",
       "      <td id=\"T_8de24_row5_col3\" class=\"data row5 col3\" >4085</td>\n",
       "      <td id=\"T_8de24_row5_col4\" class=\"data row5 col4\" >medium</td>\n",
       "      <td id=\"T_8de24_row5_col5\" class=\"data row5 col5\" >medium</td>\n",
       "      <td id=\"T_8de24_row5_col6\" class=\"data row5 col6\" ><a target=\"_blank\" href=\"https://www.nationaalarchief.nl/onderzoeken/archief/1.05.06/invnr/220/file/NL-HaNA_1.05.06_220_0005?eadId=1.05.06&unitID=220\">https://www.nationaalarchief.nl/onderzoeken/archief/1.05.06/invnr/220/file/NL-HaNA_1.05.06_220_0005?eadId=1.05.06&unitID=220</a></td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_8de24_level0_row6\" class=\"row_heading level0 row6\" >1310</th>\n",
       "      <td id=\"T_8de24_row6_col0\" class=\"data row6 col0\" >NL-HaNA_1.05.06_1166_0293.jpg</td>\n",
       "      <td id=\"T_8de24_row6_col1\" class=\"data row6 col1\" >636</td>\n",
       "      <td id=\"T_8de24_row6_col2\" class=\"data row6 col2\" >5058</td>\n",
       "      <td id=\"T_8de24_row6_col3\" class=\"data row6 col3\" >4036</td>\n",
       "      <td id=\"T_8de24_row6_col4\" class=\"data row6 col4\" >medium</td>\n",
       "      <td id=\"T_8de24_row6_col5\" class=\"data row6 col5\" >medium</td>\n",
       "      <td id=\"T_8de24_row6_col6\" class=\"data row6 col6\" ><a target=\"_blank\" href=\"https://www.nationaalarchief.nl/onderzoeken/archief/1.05.06/invnr/1166/file/NL-HaNA_1.05.06_1166_0293?eadId=1.05.06&unitID=1166\">https://www.nationaalarchief.nl/onderzoeken/archief/1.05.06/invnr/1166/file/NL-HaNA_1.05.06_1166_0293?eadId=1.05.06&unitID=1166</a></td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_8de24_level0_row7\" class=\"row_heading level0 row7\" >7251</th>\n",
       "      <td id=\"T_8de24_row7_col0\" class=\"data row7 col0\" >NL-HaNA_1.05.06_220_0003.jpg</td>\n",
       "      <td id=\"T_8de24_row7_col1\" class=\"data row7 col1\" >620</td>\n",
       "      <td id=\"T_8de24_row7_col2\" class=\"data row7 col2\" >5126</td>\n",
       "      <td id=\"T_8de24_row7_col3\" class=\"data row7 col3\" >4107</td>\n",
       "      <td id=\"T_8de24_row7_col4\" class=\"data row7 col4\" >medium</td>\n",
       "      <td id=\"T_8de24_row7_col5\" class=\"data row7 col5\" >medium</td>\n",
       "      <td id=\"T_8de24_row7_col6\" class=\"data row7 col6\" ><a target=\"_blank\" href=\"https://www.nationaalarchief.nl/onderzoeken/archief/1.05.06/invnr/220/file/NL-HaNA_1.05.06_220_0003?eadId=1.05.06&unitID=220\">https://www.nationaalarchief.nl/onderzoeken/archief/1.05.06/invnr/220/file/NL-HaNA_1.05.06_220_0003?eadId=1.05.06&unitID=220</a></td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_8de24_level0_row8\" class=\"row_heading level0 row8\" >7252</th>\n",
       "      <td id=\"T_8de24_row8_col0\" class=\"data row8 col0\" >NL-HaNA_1.05.06_220_0004.jpg</td>\n",
       "      <td id=\"T_8de24_row8_col1\" class=\"data row8 col1\" >604</td>\n",
       "      <td id=\"T_8de24_row8_col2\" class=\"data row8 col2\" >5128</td>\n",
       "      <td id=\"T_8de24_row8_col3\" class=\"data row8 col3\" >4104</td>\n",
       "      <td id=\"T_8de24_row8_col4\" class=\"data row8 col4\" >medium</td>\n",
       "      <td id=\"T_8de24_row8_col5\" class=\"data row8 col5\" >medium</td>\n",
       "      <td id=\"T_8de24_row8_col6\" class=\"data row8 col6\" ><a target=\"_blank\" href=\"https://www.nationaalarchief.nl/onderzoeken/archief/1.05.06/invnr/220/file/NL-HaNA_1.05.06_220_0004?eadId=1.05.06&unitID=220\">https://www.nationaalarchief.nl/onderzoeken/archief/1.05.06/invnr/220/file/NL-HaNA_1.05.06_220_0004?eadId=1.05.06&unitID=220</a></td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_8de24_level0_row9\" class=\"row_heading level0 row9\" >7128</th>\n",
       "      <td id=\"T_8de24_row9_col0\" class=\"data row9 col0\" >NL-HaNA_1.05.06_213_0002.jpg</td>\n",
       "      <td id=\"T_8de24_row9_col1\" class=\"data row9 col1\" >585</td>\n",
       "      <td id=\"T_8de24_row9_col2\" class=\"data row9 col2\" >5064</td>\n",
       "      <td id=\"T_8de24_row9_col3\" class=\"data row9 col3\" >4012</td>\n",
       "      <td id=\"T_8de24_row9_col4\" class=\"data row9 col4\" >medium</td>\n",
       "      <td id=\"T_8de24_row9_col5\" class=\"data row9 col5\" >medium</td>\n",
       "      <td id=\"T_8de24_row9_col6\" class=\"data row9 col6\" ><a target=\"_blank\" href=\"https://www.nationaalarchief.nl/onderzoeken/archief/1.05.06/invnr/213/file/NL-HaNA_1.05.06_213_0002?eadId=1.05.06&unitID=213\">https://www.nationaalarchief.nl/onderzoeken/archief/1.05.06/invnr/213/file/NL-HaNA_1.05.06_213_0002?eadId=1.05.06&unitID=213</a></td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n"
      ],
      "text/plain": [
       "<pandas.io.formats.style.Styler at 0x291cf65f0>"
      ]
     },
     "execution_count": 26,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# sort scans by descending number of lines and show shape information and a clickable viewer URL\n",
    "(df.sort_values('lines', ascending=False)\n",
    " .head(10)[['doc_id', 'lines', 'doc_width', 'doc_height', 'width_class', 'height_class', 'iiif_url']]\n",
    " .style.format({'iiif_url': make_clickable}))"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "90dac037",
   "metadata": {},
   "source": [
    "Many of the scans with a high number of lines contain lists and tables. Each cell in the table is represented by a separate line, which is mostly short. **In this archive, a high number of lines per scan seems to signal that the scan represents a table**.\n"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "cdad2703",
   "metadata": {},
   "source": [
    "### Scans with high numbers of words"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "id": "75eeab6c",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<style type=\"text/css\">\n",
       "</style>\n",
       "<table id=\"T_a2d99\">\n",
       "  <thead>\n",
       "    <tr>\n",
       "      <th class=\"blank level0\" >&nbsp;</th>\n",
       "      <th id=\"T_a2d99_level0_col0\" class=\"col_heading level0 col0\" >doc_id</th>\n",
       "      <th id=\"T_a2d99_level0_col1\" class=\"col_heading level0 col1\" >lines</th>\n",
       "      <th id=\"T_a2d99_level0_col2\" class=\"col_heading level0 col2\" >words</th>\n",
       "      <th id=\"T_a2d99_level0_col3\" class=\"col_heading level0 col3\" >doc_width</th>\n",
       "      <th id=\"T_a2d99_level0_col4\" class=\"col_heading level0 col4\" >doc_height</th>\n",
       "      <th id=\"T_a2d99_level0_col5\" class=\"col_heading level0 col5\" >width_class</th>\n",
       "      <th id=\"T_a2d99_level0_col6\" class=\"col_heading level0 col6\" >height_class</th>\n",
       "      <th id=\"T_a2d99_level0_col7\" class=\"col_heading level0 col7\" >iiif_url</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th id=\"T_a2d99_level0_row0\" class=\"row_heading level0 row0\" >4542</th>\n",
       "      <td id=\"T_a2d99_row0_col0\" class=\"data row0 col0\" >NL-HaNA_1.05.06_1200_0003.jpg</td>\n",
       "      <td id=\"T_a2d99_row0_col1\" class=\"data row0 col1\" >502</td>\n",
       "      <td id=\"T_a2d99_row0_col2\" class=\"data row0 col2\" >3198</td>\n",
       "      <td id=\"T_a2d99_row0_col3\" class=\"data row0 col3\" >5987</td>\n",
       "      <td id=\"T_a2d99_row0_col4\" class=\"data row0 col4\" >6356</td>\n",
       "      <td id=\"T_a2d99_row0_col5\" class=\"data row0 col5\" >wide</td>\n",
       "      <td id=\"T_a2d99_row0_col6\" class=\"data row0 col6\" >wide</td>\n",
       "      <td id=\"T_a2d99_row0_col7\" class=\"data row0 col7\" ><a target=\"_blank\" href=\"https://www.nationaalarchief.nl/onderzoeken/archief/1.05.06/invnr/1200/file/NL-HaNA_1.05.06_1200_0003?eadId=1.05.06&unitID=1200\">https://www.nationaalarchief.nl/onderzoeken/archief/1.05.06/invnr/1200/file/NL-HaNA_1.05.06_1200_0003?eadId=1.05.06&unitID=1200</a></td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_a2d99_level0_row1\" class=\"row_heading level0 row1\" >3196</th>\n",
       "      <td id=\"T_a2d99_row1_col0\" class=\"data row1 col0\" >NL-HaNA_1.05.06_1174_0207.jpg</td>\n",
       "      <td id=\"T_a2d99_row1_col1\" class=\"data row1 col1\" >228</td>\n",
       "      <td id=\"T_a2d99_row1_col2\" class=\"data row1 col2\" >2538</td>\n",
       "      <td id=\"T_a2d99_row1_col3\" class=\"data row1 col3\" >3392</td>\n",
       "      <td id=\"T_a2d99_row1_col4\" class=\"data row1 col4\" >5424</td>\n",
       "      <td id=\"T_a2d99_row1_col5\" class=\"data row1 col5\" >narrow</td>\n",
       "      <td id=\"T_a2d99_row1_col6\" class=\"data row1 col6\" >wide</td>\n",
       "      <td id=\"T_a2d99_row1_col7\" class=\"data row1 col7\" ><a target=\"_blank\" href=\"https://www.nationaalarchief.nl/onderzoeken/archief/1.05.06/invnr/1174/file/NL-HaNA_1.05.06_1174_0207?eadId=1.05.06&unitID=1174\">https://www.nationaalarchief.nl/onderzoeken/archief/1.05.06/invnr/1174/file/NL-HaNA_1.05.06_1174_0207?eadId=1.05.06&unitID=1174</a></td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_a2d99_level0_row2\" class=\"row_heading level0 row2\" >4541</th>\n",
       "      <td id=\"T_a2d99_row2_col0\" class=\"data row2 col0\" >NL-HaNA_1.05.06_1200_0002.jpg</td>\n",
       "      <td id=\"T_a2d99_row2_col1\" class=\"data row2 col1\" >487</td>\n",
       "      <td id=\"T_a2d99_row2_col2\" class=\"data row2 col2\" >2534</td>\n",
       "      <td id=\"T_a2d99_row2_col3\" class=\"data row2 col3\" >5987</td>\n",
       "      <td id=\"T_a2d99_row2_col4\" class=\"data row2 col4\" >6343</td>\n",
       "      <td id=\"T_a2d99_row2_col5\" class=\"data row2 col5\" >wide</td>\n",
       "      <td id=\"T_a2d99_row2_col6\" class=\"data row2 col6\" >wide</td>\n",
       "      <td id=\"T_a2d99_row2_col7\" class=\"data row2 col7\" ><a target=\"_blank\" href=\"https://www.nationaalarchief.nl/onderzoeken/archief/1.05.06/invnr/1200/file/NL-HaNA_1.05.06_1200_0002?eadId=1.05.06&unitID=1200\">https://www.nationaalarchief.nl/onderzoeken/archief/1.05.06/invnr/1200/file/NL-HaNA_1.05.06_1200_0002?eadId=1.05.06&unitID=1200</a></td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_a2d99_level0_row3\" class=\"row_heading level0 row3\" >4540</th>\n",
       "      <td id=\"T_a2d99_row3_col0\" class=\"data row3 col0\" >NL-HaNA_1.05.06_1200_0001.jpg</td>\n",
       "      <td id=\"T_a2d99_row3_col1\" class=\"data row3 col1\" >377</td>\n",
       "      <td id=\"T_a2d99_row3_col2\" class=\"data row3 col2\" >2402</td>\n",
       "      <td id=\"T_a2d99_row3_col3\" class=\"data row3 col3\" >5847</td>\n",
       "      <td id=\"T_a2d99_row3_col4\" class=\"data row3 col4\" >6339</td>\n",
       "      <td id=\"T_a2d99_row3_col5\" class=\"data row3 col5\" >wide</td>\n",
       "      <td id=\"T_a2d99_row3_col6\" class=\"data row3 col6\" >wide</td>\n",
       "      <td id=\"T_a2d99_row3_col7\" class=\"data row3 col7\" ><a target=\"_blank\" href=\"https://www.nationaalarchief.nl/onderzoeken/archief/1.05.06/invnr/1200/file/NL-HaNA_1.05.06_1200_0001?eadId=1.05.06&unitID=1200\">https://www.nationaalarchief.nl/onderzoeken/archief/1.05.06/invnr/1200/file/NL-HaNA_1.05.06_1200_0001?eadId=1.05.06&unitID=1200</a></td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_a2d99_level0_row4\" class=\"row_heading level0 row4\" >11545</th>\n",
       "      <td id=\"T_a2d99_row4_col0\" class=\"data row4 col0\" >NL-HaNA_1.05.06_56_0016.jpg</td>\n",
       "      <td id=\"T_a2d99_row4_col1\" class=\"data row4 col1\" >253</td>\n",
       "      <td id=\"T_a2d99_row4_col2\" class=\"data row4 col2\" >2377</td>\n",
       "      <td id=\"T_a2d99_row4_col3\" class=\"data row4 col3\" >5101</td>\n",
       "      <td id=\"T_a2d99_row4_col4\" class=\"data row4 col4\" >4053</td>\n",
       "      <td id=\"T_a2d99_row4_col5\" class=\"data row4 col5\" >medium</td>\n",
       "      <td id=\"T_a2d99_row4_col6\" class=\"data row4 col6\" >medium</td>\n",
       "      <td id=\"T_a2d99_row4_col7\" class=\"data row4 col7\" ><a target=\"_blank\" href=\"https://www.nationaalarchief.nl/onderzoeken/archief/1.05.06/invnr/56/file/NL-HaNA_1.05.06_56_0016?eadId=1.05.06&unitID=56\">https://www.nationaalarchief.nl/onderzoeken/archief/1.05.06/invnr/56/file/NL-HaNA_1.05.06_56_0016?eadId=1.05.06&unitID=56</a></td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_a2d99_level0_row5\" class=\"row_heading level0 row5\" >4543</th>\n",
       "      <td id=\"T_a2d99_row5_col0\" class=\"data row5 col0\" >NL-HaNA_1.05.06_1200_0004.jpg</td>\n",
       "      <td id=\"T_a2d99_row5_col1\" class=\"data row5 col1\" >344</td>\n",
       "      <td id=\"T_a2d99_row5_col2\" class=\"data row5 col2\" >2357</td>\n",
       "      <td id=\"T_a2d99_row5_col3\" class=\"data row5 col3\" >4662</td>\n",
       "      <td id=\"T_a2d99_row5_col4\" class=\"data row5 col4\" >6375</td>\n",
       "      <td id=\"T_a2d99_row5_col5\" class=\"data row5 col5\" >medium</td>\n",
       "      <td id=\"T_a2d99_row5_col6\" class=\"data row5 col6\" >wide</td>\n",
       "      <td id=\"T_a2d99_row5_col7\" class=\"data row5 col7\" ><a target=\"_blank\" href=\"https://www.nationaalarchief.nl/onderzoeken/archief/1.05.06/invnr/1200/file/NL-HaNA_1.05.06_1200_0004?eadId=1.05.06&unitID=1200\">https://www.nationaalarchief.nl/onderzoeken/archief/1.05.06/invnr/1200/file/NL-HaNA_1.05.06_1200_0004?eadId=1.05.06&unitID=1200</a></td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_a2d99_level0_row6\" class=\"row_heading level0 row6\" >8111</th>\n",
       "      <td id=\"T_a2d99_row6_col0\" class=\"data row6 col0\" >NL-HaNA_1.05.06_327_0004.jpg</td>\n",
       "      <td id=\"T_a2d99_row6_col1\" class=\"data row6 col1\" >189</td>\n",
       "      <td id=\"T_a2d99_row6_col2\" class=\"data row6 col2\" >2346</td>\n",
       "      <td id=\"T_a2d99_row6_col3\" class=\"data row6 col3\" >3187</td>\n",
       "      <td id=\"T_a2d99_row6_col4\" class=\"data row6 col4\" >5048</td>\n",
       "      <td id=\"T_a2d99_row6_col5\" class=\"data row6 col5\" >narrow</td>\n",
       "      <td id=\"T_a2d99_row6_col6\" class=\"data row6 col6\" >wide</td>\n",
       "      <td id=\"T_a2d99_row6_col7\" class=\"data row6 col7\" ><a target=\"_blank\" href=\"https://www.nationaalarchief.nl/onderzoeken/archief/1.05.06/invnr/327/file/NL-HaNA_1.05.06_327_0004?eadId=1.05.06&unitID=327\">https://www.nationaalarchief.nl/onderzoeken/archief/1.05.06/invnr/327/file/NL-HaNA_1.05.06_327_0004?eadId=1.05.06&unitID=327</a></td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_a2d99_level0_row7\" class=\"row_heading level0 row7\" >3195</th>\n",
       "      <td id=\"T_a2d99_row7_col0\" class=\"data row7 col0\" >NL-HaNA_1.05.06_1174_0206.jpg</td>\n",
       "      <td id=\"T_a2d99_row7_col1\" class=\"data row7 col1\" >223</td>\n",
       "      <td id=\"T_a2d99_row7_col2\" class=\"data row7 col2\" >2304</td>\n",
       "      <td id=\"T_a2d99_row7_col3\" class=\"data row7 col3\" >3377</td>\n",
       "      <td id=\"T_a2d99_row7_col4\" class=\"data row7 col4\" >5421</td>\n",
       "      <td id=\"T_a2d99_row7_col5\" class=\"data row7 col5\" >narrow</td>\n",
       "      <td id=\"T_a2d99_row7_col6\" class=\"data row7 col6\" >wide</td>\n",
       "      <td id=\"T_a2d99_row7_col7\" class=\"data row7 col7\" ><a target=\"_blank\" href=\"https://www.nationaalarchief.nl/onderzoeken/archief/1.05.06/invnr/1174/file/NL-HaNA_1.05.06_1174_0206?eadId=1.05.06&unitID=1174\">https://www.nationaalarchief.nl/onderzoeken/archief/1.05.06/invnr/1174/file/NL-HaNA_1.05.06_1174_0206?eadId=1.05.06&unitID=1174</a></td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_a2d99_level0_row8\" class=\"row_heading level0 row8\" >11540</th>\n",
       "      <td id=\"T_a2d99_row8_col0\" class=\"data row8 col0\" >NL-HaNA_1.05.06_56_0011.jpg</td>\n",
       "      <td id=\"T_a2d99_row8_col1\" class=\"data row8 col1\" >185</td>\n",
       "      <td id=\"T_a2d99_row8_col2\" class=\"data row8 col2\" >2270</td>\n",
       "      <td id=\"T_a2d99_row8_col3\" class=\"data row8 col3\" >5082</td>\n",
       "      <td id=\"T_a2d99_row8_col4\" class=\"data row8 col4\" >4047</td>\n",
       "      <td id=\"T_a2d99_row8_col5\" class=\"data row8 col5\" >medium</td>\n",
       "      <td id=\"T_a2d99_row8_col6\" class=\"data row8 col6\" >medium</td>\n",
       "      <td id=\"T_a2d99_row8_col7\" class=\"data row8 col7\" ><a target=\"_blank\" href=\"https://www.nationaalarchief.nl/onderzoeken/archief/1.05.06/invnr/56/file/NL-HaNA_1.05.06_56_0011?eadId=1.05.06&unitID=56\">https://www.nationaalarchief.nl/onderzoeken/archief/1.05.06/invnr/56/file/NL-HaNA_1.05.06_56_0011?eadId=1.05.06&unitID=56</a></td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_a2d99_level0_row9\" class=\"row_heading level0 row9\" >8110</th>\n",
       "      <td id=\"T_a2d99_row9_col0\" class=\"data row9 col0\" >NL-HaNA_1.05.06_327_0003.jpg</td>\n",
       "      <td id=\"T_a2d99_row9_col1\" class=\"data row9 col1\" >286</td>\n",
       "      <td id=\"T_a2d99_row9_col2\" class=\"data row9 col2\" >2180</td>\n",
       "      <td id=\"T_a2d99_row9_col3\" class=\"data row9 col3\" >3193</td>\n",
       "      <td id=\"T_a2d99_row9_col4\" class=\"data row9 col4\" >5053</td>\n",
       "      <td id=\"T_a2d99_row9_col5\" class=\"data row9 col5\" >narrow</td>\n",
       "      <td id=\"T_a2d99_row9_col6\" class=\"data row9 col6\" >wide</td>\n",
       "      <td id=\"T_a2d99_row9_col7\" class=\"data row9 col7\" ><a target=\"_blank\" href=\"https://www.nationaalarchief.nl/onderzoeken/archief/1.05.06/invnr/327/file/NL-HaNA_1.05.06_327_0003?eadId=1.05.06&unitID=327\">https://www.nationaalarchief.nl/onderzoeken/archief/1.05.06/invnr/327/file/NL-HaNA_1.05.06_327_0003?eadId=1.05.06&unitID=327</a></td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n"
      ],
      "text/plain": [
       "<pandas.io.formats.style.Styler at 0x29427bc10>"
      ]
     },
     "execution_count": 27,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# sort scans by descending number of words and show shape information and a clickable viewer URL\n",
    "(df.sort_values('words', ascending=False)\n",
    " .head(10)[['doc_id', 'lines', 'words', 'doc_width', 'doc_height', 'width_class', 'height_class', 'iiif_url']]\n",
    " .style.format({'iiif_url': make_clickable}))"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "7b02a7ee",
   "metadata": {},
   "source": [
    "**Inventories with a high median number of words per page are mostly printed material.** They seem to be a mixture of patents (NL: 'Octrooi'), newspapers and lists/tables."
   ]
  },
  {
   "cell_type": "markdown",
   "id": "1c08a361",
   "metadata": {},
   "source": [
    "## Categorising scans by number of words per line\n",
    "\n",
    "**Assumptions**: \n",
    "- many lines with few words (3 or fewer) signals the table contains a table or list\n",
    "- many lines with several words (4 or more) signals the table contains a running text.\n",
    "\n",
    "**TO DO**:\n",
    "- It probably works better after some further pre-processing, e.g. \n",
    "    - calculate these statistics per text region and classify text regions,\n",
    "    - cut scans into pages and classify individual pages. A two-page scan (the majority of scans?) can have a full-text page and a table/list or empty page."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "id": "cbb2cdcf",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "['words_per_line_0-0',\n",
       " 'words_per_line_1-1',\n",
       " 'words_per_line_2-2',\n",
       " 'words_per_line_3-3',\n",
       " 'words_per_line_4-5',\n",
       " 'words_per_line_6-9',\n",
       " 'words_per_line_10-15',\n",
       " 'words_per_line_16-25',\n",
       " 'words_per_line_26-42',\n",
       " 'words_per_line_43-70',\n",
       " 'words_per_line_71-100']"
      ]
     },
     "execution_count": 28,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "wpl_cols = [col for col in df.columns if 'words_per_line_' in col]\n",
    "wpl_cols"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "id": "c8ffb403",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1152x360 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "ax = sns.barplot(data=df[wpl_cols].rename(columns={col: col.replace('words_per_line_', '') for col in wpl_cols}))\n",
    "ax.figure.set_size_inches(16,5)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "id": "254c9151",
   "metadata": {},
   "outputs": [],
   "source": [
    "wpl_few_cols = [col for col in wpl_cols if int(col.split('-')[-1]) <= 3]\n",
    "wpl_many_cols = [col for col in wpl_cols if int(col.split('-')[-1]) > 3]\n",
    "wpl_few_cols, wpl_many_cols\n",
    "\n",
    "df['wpl_few'] = df[wpl_few_cols].sum(axis=1)\n",
    "df['wpl_many'] = df[wpl_many_cols].sum(axis=1)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "id": "08ae688d",
   "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>wpl_few</th>\n",
       "      <th>wpl_many</th>\n",
       "      <th>lines</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>0.151515</td>\n",
       "      <td>0.848485</td>\n",
       "      <td>33</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>0.084746</td>\n",
       "      <td>0.915254</td>\n",
       "      <td>59</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>0.055556</td>\n",
       "      <td>0.944444</td>\n",
       "      <td>54</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>15852</th>\n",
       "      <td>0.195652</td>\n",
       "      <td>0.782609</td>\n",
       "      <td>46</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>15853</th>\n",
       "      <td>0.113636</td>\n",
       "      <td>0.886364</td>\n",
       "      <td>44</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>15854</th>\n",
       "      <td>0.113636</td>\n",
       "      <td>0.886364</td>\n",
       "      <td>44</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>15855</th>\n",
       "      <td>0.227273</td>\n",
       "      <td>0.772727</td>\n",
       "      <td>44</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>15856</th>\n",
       "      <td>0.333333</td>\n",
       "      <td>0.629630</td>\n",
       "      <td>27</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>15857 rows × 3 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "        wpl_few  wpl_many  lines\n",
       "0      0.151515  0.848485     33\n",
       "1      0.084746  0.915254     59\n",
       "2      0.055556  0.944444     54\n",
       "3           NaN       NaN      0\n",
       "4           NaN       NaN      0\n",
       "...         ...       ...    ...\n",
       "15852  0.195652  0.782609     46\n",
       "15853  0.113636  0.886364     44\n",
       "15854  0.113636  0.886364     44\n",
       "15855  0.227273  0.772727     44\n",
       "15856  0.333333  0.629630     27\n",
       "\n",
       "[15857 rows x 3 columns]"
      ]
     },
     "execution_count": 31,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "wpl_rel_df = df[['wpl_few', 'wpl_many']].div(df.lines, axis=0)\n",
    "wpl_rel_df['lines'] = df.lines\n",
    "wpl_rel_df"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "id": "37fb1d8e",
   "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>wpl_few</th>\n",
       "      <th>wpl_many</th>\n",
       "      <th>lines</th>\n",
       "      <th>wpl_class</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>0.151515</td>\n",
       "      <td>0.848485</td>\n",
       "      <td>33</td>\n",
       "      <td>running_text</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>0.084746</td>\n",
       "      <td>0.915254</td>\n",
       "      <td>59</td>\n",
       "      <td>running_text</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>0.055556</td>\n",
       "      <td>0.944444</td>\n",
       "      <td>54</td>\n",
       "      <td>running_text</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0</td>\n",
       "      <td>partially_empty</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0</td>\n",
       "      <td>partially_empty</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>15852</th>\n",
       "      <td>0.195652</td>\n",
       "      <td>0.782609</td>\n",
       "      <td>46</td>\n",
       "      <td>running_text</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>15853</th>\n",
       "      <td>0.113636</td>\n",
       "      <td>0.886364</td>\n",
       "      <td>44</td>\n",
       "      <td>running_text</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>15854</th>\n",
       "      <td>0.113636</td>\n",
       "      <td>0.886364</td>\n",
       "      <td>44</td>\n",
       "      <td>running_text</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>15855</th>\n",
       "      <td>0.227273</td>\n",
       "      <td>0.772727</td>\n",
       "      <td>44</td>\n",
       "      <td>running_text</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>15856</th>\n",
       "      <td>0.333333</td>\n",
       "      <td>0.629630</td>\n",
       "      <td>27</td>\n",
       "      <td>mixed</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>15857 rows × 4 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "        wpl_few  wpl_many  lines        wpl_class\n",
       "0      0.151515  0.848485     33     running_text\n",
       "1      0.084746  0.915254     59     running_text\n",
       "2      0.055556  0.944444     54     running_text\n",
       "3           NaN       NaN      0  partially_empty\n",
       "4           NaN       NaN      0  partially_empty\n",
       "...         ...       ...    ...              ...\n",
       "15852  0.195652  0.782609     46     running_text\n",
       "15853  0.113636  0.886364     44     running_text\n",
       "15854  0.113636  0.886364     44     running_text\n",
       "15855  0.227273  0.772727     44     running_text\n",
       "15856  0.333333  0.629630     27            mixed\n",
       "\n",
       "[15857 rows x 4 columns]"
      ]
     },
     "execution_count": 32,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "def classify_wpl_scan_type(row):\n",
    "    if row['lines'] < 10:\n",
    "        return 'partially_empty'\n",
    "    elif row['wpl_few'] > 0.7:\n",
    "        if row['lines'] > 100:\n",
    "            return 'table'\n",
    "        else:\n",
    "            return 'list'\n",
    "    elif row['wpl_few'] < 0.3:\n",
    "        return 'running_text'\n",
    "    else:\n",
    "        return 'mixed'\n",
    "    \n",
    "    \n",
    "wpl_rel_df['wpl_class'] = wpl_rel_df.apply(classify_wpl_scan_type, axis=1)\n",
    "wpl_rel_df"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 33,
   "id": "ff859cc5",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "running_text       9739\n",
       "mixed              2697\n",
       "partially_empty    2390\n",
       "list                523\n",
       "table               508\n",
       "Name: wpl_class, dtype: int64"
      ]
     },
     "execution_count": 33,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "wpl_rel_df.wpl_class.value_counts()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "2f9d7cb6",
   "metadata": {},
   "source": [
    "## Grouping inventories by line categories"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "9d5f36a9",
   "metadata": {},
   "source": [
    "There are five categories of line widths. We can analyse scans by their composition of lines of different width categories."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 34,
   "id": "53df4090",
   "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>line_width_range_0-600</th>\n",
       "      <th>line_width_range_600-800</th>\n",
       "      <th>line_width_range_800-1300</th>\n",
       "      <th>line_width_range_1300-1600</th>\n",
       "      <th>line_width_range_1600-</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>4</td>\n",
       "      <td>0</td>\n",
       "      <td>2</td>\n",
       "      <td>0</td>\n",
       "      <td>27</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>3</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>2</td>\n",
       "      <td>54</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>2</td>\n",
       "      <td>0</td>\n",
       "      <td>4</td>\n",
       "      <td>1</td>\n",
       "      <td>47</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>15852</th>\n",
       "      <td>6</td>\n",
       "      <td>0</td>\n",
       "      <td>3</td>\n",
       "      <td>1</td>\n",
       "      <td>35</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>15853</th>\n",
       "      <td>4</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>39</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>15854</th>\n",
       "      <td>4</td>\n",
       "      <td>1</td>\n",
       "      <td>4</td>\n",
       "      <td>2</td>\n",
       "      <td>33</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>15855</th>\n",
       "      <td>7</td>\n",
       "      <td>3</td>\n",
       "      <td>3</td>\n",
       "      <td>1</td>\n",
       "      <td>30</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>15856</th>\n",
       "      <td>4</td>\n",
       "      <td>3</td>\n",
       "      <td>5</td>\n",
       "      <td>0</td>\n",
       "      <td>14</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>15857 rows × 5 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "       line_width_range_0-600  line_width_range_600-800  \\\n",
       "0                           4                         0   \n",
       "1                           3                         0   \n",
       "2                           2                         0   \n",
       "3                           0                         0   \n",
       "4                           0                         0   \n",
       "...                       ...                       ...   \n",
       "15852                       6                         0   \n",
       "15853                       4                         0   \n",
       "15854                       4                         1   \n",
       "15855                       7                         3   \n",
       "15856                       4                         3   \n",
       "\n",
       "       line_width_range_800-1300  line_width_range_1300-1600  \\\n",
       "0                              2                           0   \n",
       "1                              0                           2   \n",
       "2                              4                           1   \n",
       "3                              0                           0   \n",
       "4                              0                           0   \n",
       "...                          ...                         ...   \n",
       "15852                          3                           1   \n",
       "15853                          1                           0   \n",
       "15854                          4                           2   \n",
       "15855                          3                           1   \n",
       "15856                          5                           0   \n",
       "\n",
       "       line_width_range_1600-  \n",
       "0                          27  \n",
       "1                          54  \n",
       "2                          47  \n",
       "3                           0  \n",
       "4                           0  \n",
       "...                       ...  \n",
       "15852                      35  \n",
       "15853                      39  \n",
       "15854                      33  \n",
       "15855                      30  \n",
       "15856                      14  \n",
       "\n",
       "[15857 rows x 5 columns]"
      ]
     },
     "execution_count": 34,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "lwr_cols = [col for col in df.columns if 'line_width_range_' in col]\n",
    "df[lwr_cols]"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "6887bb7c",
   "metadata": {},
   "source": [
    "Many scans seem to have mostly lines wider than 1600 pixels, but some may have mainly short lines (e.g. scans with tables)."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 35,
   "id": "74fb010c",
   "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>mean</th>\n",
       "      <th>std</th>\n",
       "      <th>min</th>\n",
       "      <th>25%</th>\n",
       "      <th>50%</th>\n",
       "      <th>75%</th>\n",
       "      <th>max</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>inv_num</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>108A</th>\n",
       "      <td>3.0</td>\n",
       "      <td>1.333333</td>\n",
       "      <td>2.309401</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.00</td>\n",
       "      <td>0.0</td>\n",
       "      <td>2.00</td>\n",
       "      <td>4.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>946</th>\n",
       "      <td>4.0</td>\n",
       "      <td>2.250000</td>\n",
       "      <td>4.500000</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.00</td>\n",
       "      <td>0.0</td>\n",
       "      <td>2.25</td>\n",
       "      <td>9.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>37</th>\n",
       "      <td>3.0</td>\n",
       "      <td>0.666667</td>\n",
       "      <td>1.154701</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.00</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.00</td>\n",
       "      <td>2.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>47B</th>\n",
       "      <td>3.0</td>\n",
       "      <td>0.333333</td>\n",
       "      <td>0.577350</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.00</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.50</td>\n",
       "      <td>1.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>933</th>\n",
       "      <td>3.0</td>\n",
       "      <td>16.666667</td>\n",
       "      <td>28.867513</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.00</td>\n",
       "      <td>0.0</td>\n",
       "      <td>25.00</td>\n",
       "      <td>50.0</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",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>156</th>\n",
       "      <td>7.0</td>\n",
       "      <td>203.714286</td>\n",
       "      <td>138.431348</td>\n",
       "      <td>1.0</td>\n",
       "      <td>132.50</td>\n",
       "      <td>276.0</td>\n",
       "      <td>289.50</td>\n",
       "      <td>305.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>176</th>\n",
       "      <td>4.0</td>\n",
       "      <td>306.750000</td>\n",
       "      <td>354.915365</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.00</td>\n",
       "      <td>293.0</td>\n",
       "      <td>599.75</td>\n",
       "      <td>641.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>150</th>\n",
       "      <td>28.0</td>\n",
       "      <td>327.321429</td>\n",
       "      <td>141.682235</td>\n",
       "      <td>0.0</td>\n",
       "      <td>325.25</td>\n",
       "      <td>382.0</td>\n",
       "      <td>400.25</td>\n",
       "      <td>448.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>152</th>\n",
       "      <td>13.0</td>\n",
       "      <td>289.461538</td>\n",
       "      <td>211.368405</td>\n",
       "      <td>0.0</td>\n",
       "      <td>11.00</td>\n",
       "      <td>441.0</td>\n",
       "      <td>455.00</td>\n",
       "      <td>463.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>220</th>\n",
       "      <td>9.0</td>\n",
       "      <td>332.000000</td>\n",
       "      <td>275.267597</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.00</td>\n",
       "      <td>465.0</td>\n",
       "      <td>574.00</td>\n",
       "      <td>620.0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>533 rows × 8 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "         count        mean         std  min     25%    50%     75%    max\n",
       "inv_num                                                                  \n",
       "108A       3.0    1.333333    2.309401  0.0    0.00    0.0    2.00    4.0\n",
       "946        4.0    2.250000    4.500000  0.0    0.00    0.0    2.25    9.0\n",
       "37         3.0    0.666667    1.154701  0.0    0.00    0.0    1.00    2.0\n",
       "47B        3.0    0.333333    0.577350  0.0    0.00    0.0    0.50    1.0\n",
       "933        3.0   16.666667   28.867513  0.0    0.00    0.0   25.00   50.0\n",
       "...        ...         ...         ...  ...     ...    ...     ...    ...\n",
       "156        7.0  203.714286  138.431348  1.0  132.50  276.0  289.50  305.0\n",
       "176        4.0  306.750000  354.915365  0.0    0.00  293.0  599.75  641.0\n",
       "150       28.0  327.321429  141.682235  0.0  325.25  382.0  400.25  448.0\n",
       "152       13.0  289.461538  211.368405  0.0   11.00  441.0  455.00  463.0\n",
       "220        9.0  332.000000  275.267597  0.0    0.00  465.0  574.00  620.0\n",
       "\n",
       "[533 rows x 8 columns]"
      ]
     },
     "execution_count": 35,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.groupby('inv_num')['line_width_range_0-600'].describe().sort_values('50%')"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "7b5a47c4",
   "metadata": {},
   "source": [
    "We can also aggregate the number of lines in each category per inventory and check if their median number of lines per scan differs."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 37,
   "id": "05c21f86",
   "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>line_width_range_0-600</th>\n",
       "      <th>line_width_range_600-800</th>\n",
       "      <th>line_width_range_800-1300</th>\n",
       "      <th>line_width_range_1300-1600</th>\n",
       "      <th>line_width_range_1600-</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>inv_num</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>2.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>27.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>10</th>\n",
       "      <td>4.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>51.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>100</th>\n",
       "      <td>3.0</td>\n",
       "      <td>2.0</td>\n",
       "      <td>2.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>2.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1005</th>\n",
       "      <td>13.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>3.0</td>\n",
       "      <td>7.5</td>\n",
       "      <td>42.5</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1007</th>\n",
       "      <td>3.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>37.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>981</th>\n",
       "      <td>3.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>44.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>983</th>\n",
       "      <td>4.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>4.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>2.5</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>983A</th>\n",
       "      <td>7.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>2.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>3.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>984</th>\n",
       "      <td>3.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>5.0</td>\n",
       "      <td>2.0</td>\n",
       "      <td>4.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>99</th>\n",
       "      <td>5.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>3.5</td>\n",
       "      <td>1.5</td>\n",
       "      <td>30.0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>533 rows × 5 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "         line_width_range_0-600  line_width_range_600-800  \\\n",
       "inv_num                                                     \n",
       "1                           2.0                       0.0   \n",
       "10                          4.0                       0.0   \n",
       "100                         3.0                       2.0   \n",
       "1005                       13.0                       1.0   \n",
       "1007                        3.0                       0.0   \n",
       "...                         ...                       ...   \n",
       "981                         3.0                       0.0   \n",
       "983                         4.0                       1.0   \n",
       "983A                        7.0                       1.0   \n",
       "984                         3.0                       0.0   \n",
       "99                          5.0                       1.0   \n",
       "\n",
       "         line_width_range_800-1300  line_width_range_1300-1600  \\\n",
       "inv_num                                                          \n",
       "1                              0.0                         0.0   \n",
       "10                             1.0                         1.0   \n",
       "100                            2.0                         0.0   \n",
       "1005                           3.0                         7.5   \n",
       "1007                           1.0                         1.0   \n",
       "...                            ...                         ...   \n",
       "981                            1.0                         1.0   \n",
       "983                            4.0                         0.0   \n",
       "983A                           2.0                         1.0   \n",
       "984                            5.0                         2.0   \n",
       "99                             3.5                         1.5   \n",
       "\n",
       "         line_width_range_1600-  \n",
       "inv_num                          \n",
       "1                          27.0  \n",
       "10                         51.0  \n",
       "100                         2.0  \n",
       "1005                       42.5  \n",
       "1007                       37.0  \n",
       "...                         ...  \n",
       "981                        44.0  \n",
       "983                         2.5  \n",
       "983A                        3.0  \n",
       "984                         4.0  \n",
       "99                         30.0  \n",
       "\n",
       "[533 rows x 5 columns]"
      ]
     },
     "execution_count": 37,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "g = df.groupby('inv_num')[lwr_cols].describe()#.sort_values('50%')\n",
    "g.swaplevel(0, 1, axis=1)['50%']"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "523277e6",
   "metadata": {},
   "source": [
    "Another steps is to turn the absolute frequencies into relative frequencies, so inventories are easier to compare."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 38,
   "id": "1a17964c",
   "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>line_width_range_0-600</th>\n",
       "      <th>line_width_range_600-800</th>\n",
       "      <th>line_width_range_800-1300</th>\n",
       "      <th>line_width_range_1300-1600</th>\n",
       "      <th>line_width_range_1600-</th>\n",
       "      <th>num_scans</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>inv_num</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>0.061644</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.041096</td>\n",
       "      <td>0.020548</td>\n",
       "      <td>0.876712</td>\n",
       "      <td>5</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>10</th>\n",
       "      <td>0.080519</td>\n",
       "      <td>0.007506</td>\n",
       "      <td>0.025247</td>\n",
       "      <td>0.016035</td>\n",
       "      <td>0.845786</td>\n",
       "      <td>53</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>100</th>\n",
       "      <td>0.510638</td>\n",
       "      <td>0.106383</td>\n",
       "      <td>0.234043</td>\n",
       "      <td>0.042553</td>\n",
       "      <td>0.106383</td>\n",
       "      <td>3</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1005</th>\n",
       "      <td>0.224906</td>\n",
       "      <td>0.018868</td>\n",
       "      <td>0.089811</td>\n",
       "      <td>0.110189</td>\n",
       "      <td>0.550189</td>\n",
       "      <td>18</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1007</th>\n",
       "      <td>0.041558</td>\n",
       "      <td>0.005195</td>\n",
       "      <td>0.041558</td>\n",
       "      <td>0.023377</td>\n",
       "      <td>0.875325</td>\n",
       "      <td>7</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",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>981</th>\n",
       "      <td>0.080645</td>\n",
       "      <td>0.016129</td>\n",
       "      <td>0.043011</td>\n",
       "      <td>0.032258</td>\n",
       "      <td>0.822581</td>\n",
       "      <td>5</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>983</th>\n",
       "      <td>0.268902</td>\n",
       "      <td>0.051762</td>\n",
       "      <td>0.356067</td>\n",
       "      <td>0.035485</td>\n",
       "      <td>0.276552</td>\n",
       "      <td>312</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>983A</th>\n",
       "      <td>0.561142</td>\n",
       "      <td>0.064548</td>\n",
       "      <td>0.098605</td>\n",
       "      <td>0.062601</td>\n",
       "      <td>0.178073</td>\n",
       "      <td>102</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>984</th>\n",
       "      <td>0.427632</td>\n",
       "      <td>0.078947</td>\n",
       "      <td>0.197368</td>\n",
       "      <td>0.078947</td>\n",
       "      <td>0.217105</td>\n",
       "      <td>7</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>99</th>\n",
       "      <td>0.223842</td>\n",
       "      <td>0.028110</td>\n",
       "      <td>0.143675</td>\n",
       "      <td>0.033837</td>\n",
       "      <td>0.563769</td>\n",
       "      <td>40</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>533 rows × 6 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "         line_width_range_0-600  line_width_range_600-800  \\\n",
       "inv_num                                                     \n",
       "1                      0.061644                  0.000000   \n",
       "10                     0.080519                  0.007506   \n",
       "100                    0.510638                  0.106383   \n",
       "1005                   0.224906                  0.018868   \n",
       "1007                   0.041558                  0.005195   \n",
       "...                         ...                       ...   \n",
       "981                    0.080645                  0.016129   \n",
       "983                    0.268902                  0.051762   \n",
       "983A                   0.561142                  0.064548   \n",
       "984                    0.427632                  0.078947   \n",
       "99                     0.223842                  0.028110   \n",
       "\n",
       "         line_width_range_800-1300  line_width_range_1300-1600  \\\n",
       "inv_num                                                          \n",
       "1                         0.041096                    0.020548   \n",
       "10                        0.025247                    0.016035   \n",
       "100                       0.234043                    0.042553   \n",
       "1005                      0.089811                    0.110189   \n",
       "1007                      0.041558                    0.023377   \n",
       "...                            ...                         ...   \n",
       "981                       0.043011                    0.032258   \n",
       "983                       0.356067                    0.035485   \n",
       "983A                      0.098605                    0.062601   \n",
       "984                       0.197368                    0.078947   \n",
       "99                        0.143675                    0.033837   \n",
       "\n",
       "         line_width_range_1600-  num_scans  \n",
       "inv_num                                     \n",
       "1                      0.876712          5  \n",
       "10                     0.845786         53  \n",
       "100                    0.106383          3  \n",
       "1005                   0.550189         18  \n",
       "1007                   0.875325          7  \n",
       "...                         ...        ...  \n",
       "981                    0.822581          5  \n",
       "983                    0.276552        312  \n",
       "983A                   0.178073        102  \n",
       "984                    0.217105          7  \n",
       "99                     0.563769         40  \n",
       "\n",
       "[533 rows x 6 columns]"
      ]
     },
     "execution_count": 38,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "g = df.groupby('inv_num')[lwr_cols + ['lines']].sum()\n",
    "lwr_rel_df = g[lwr_cols].div(g.lines, axis=0)\n",
    "lwr_rel_df['num_scans'] = s_inv_nums\n",
    "lwr_rel_df"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 39,
   "id": "561dd19e",
   "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>line_width_range_0-600</th>\n",
       "      <th>line_width_range_600-800</th>\n",
       "      <th>line_width_range_800-1300</th>\n",
       "      <th>line_width_range_1300-1600</th>\n",
       "      <th>line_width_range_1600-</th>\n",
       "      <th>num_scans</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>inv_num</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>1084</th>\n",
       "      <td>0.036145</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.963855</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>3</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1242</th>\n",
       "      <td>0.028944</td>\n",
       "      <td>0.013025</td>\n",
       "      <td>0.955137</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>11</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>887</th>\n",
       "      <td>0.053763</td>\n",
       "      <td>0.010753</td>\n",
       "      <td>0.935484</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>3</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>353</th>\n",
       "      <td>0.095556</td>\n",
       "      <td>0.004444</td>\n",
       "      <td>0.895556</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>9</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1153</th>\n",
       "      <td>0.058696</td>\n",
       "      <td>0.013043</td>\n",
       "      <td>0.892391</td>\n",
       "      <td>0.001087</td>\n",
       "      <td>0.030435</td>\n",
       "      <td>19</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>108A</th>\n",
       "      <td>0.074074</td>\n",
       "      <td>0.037037</td>\n",
       "      <td>0.888889</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>3</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>886</th>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.884615</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.076923</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1243</th>\n",
       "      <td>0.099089</td>\n",
       "      <td>0.017084</td>\n",
       "      <td>0.882688</td>\n",
       "      <td>0.001139</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>15</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>169</th>\n",
       "      <td>0.092527</td>\n",
       "      <td>0.021352</td>\n",
       "      <td>0.882562</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.001779</td>\n",
       "      <td>12</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>888</th>\n",
       "      <td>0.077586</td>\n",
       "      <td>0.008621</td>\n",
       "      <td>0.879310</td>\n",
       "      <td>0.008621</td>\n",
       "      <td>0.017241</td>\n",
       "      <td>3</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "         line_width_range_0-600  line_width_range_600-800  \\\n",
       "inv_num                                                     \n",
       "1084                   0.036145                  0.000000   \n",
       "1242                   0.028944                  0.013025   \n",
       "887                    0.053763                  0.010753   \n",
       "353                    0.095556                  0.004444   \n",
       "1153                   0.058696                  0.013043   \n",
       "108A                   0.074074                  0.037037   \n",
       "886                    0.000000                  0.000000   \n",
       "1243                   0.099089                  0.017084   \n",
       "169                    0.092527                  0.021352   \n",
       "888                    0.077586                  0.008621   \n",
       "\n",
       "         line_width_range_800-1300  line_width_range_1300-1600  \\\n",
       "inv_num                                                          \n",
       "1084                      0.963855                    0.000000   \n",
       "1242                      0.955137                    0.000000   \n",
       "887                       0.935484                    0.000000   \n",
       "353                       0.895556                    0.000000   \n",
       "1153                      0.892391                    0.001087   \n",
       "108A                      0.888889                    0.000000   \n",
       "886                       0.884615                    0.000000   \n",
       "1243                      0.882688                    0.001139   \n",
       "169                       0.882562                    0.000000   \n",
       "888                       0.879310                    0.008621   \n",
       "\n",
       "         line_width_range_1600-  num_scans  \n",
       "inv_num                                     \n",
       "1084                   0.000000          3  \n",
       "1242                   0.000000         11  \n",
       "887                    0.000000          3  \n",
       "353                    0.000000          9  \n",
       "1153                   0.030435         19  \n",
       "108A                   0.000000          3  \n",
       "886                    0.076923          2  \n",
       "1243                   0.000000         15  \n",
       "169                    0.001779         12  \n",
       "888                    0.017241          3  "
      ]
     },
     "execution_count": 39,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "lwr_rel_df.sort_values('line_width_range_800-1300', ascending=False).head(10)\n",
    "\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 40,
   "id": "69f9a930",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<style type=\"text/css\">\n",
       "</style>\n",
       "<table id=\"T_b16bb\">\n",
       "  <thead>\n",
       "    <tr>\n",
       "      <th class=\"blank level0\" >&nbsp;</th>\n",
       "      <th id=\"T_b16bb_level0_col0\" class=\"col_heading level0 col0\" >doc_id</th>\n",
       "      <th id=\"T_b16bb_level0_col1\" class=\"col_heading level0 col1\" >iiif_url</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th class=\"index_name level0\" >inv_num</th>\n",
       "      <th class=\"blank col0\" >&nbsp;</th>\n",
       "      <th class=\"blank col1\" >&nbsp;</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th id=\"T_b16bb_level0_row0\" class=\"row_heading level0 row0\" >1084</th>\n",
       "      <td id=\"T_b16bb_row0_col0\" class=\"data row0 col0\" >NL-HaNA_1.05.06_1084_0001.jpg</td>\n",
       "      <td id=\"T_b16bb_row0_col1\" class=\"data row0 col1\" ><a target=\"_blank\" href=\"https://www.nationaalarchief.nl/onderzoeken/archief/1.05.06/invnr/1084/file/NL-HaNA_1.05.06_1084_0001?eadId=1.05.06&unitID=1084\">https://www.nationaalarchief.nl/onderzoeken/archief/1.05.06/invnr/1084/file/NL-HaNA_1.05.06_1084_0001?eadId=1.05.06&unitID=1084</a></td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_b16bb_level0_row1\" class=\"row_heading level0 row1\" >108A</th>\n",
       "      <td id=\"T_b16bb_row1_col0\" class=\"data row1 col0\" >NL-HaNA_1.05.06_108A_0001.jpg</td>\n",
       "      <td id=\"T_b16bb_row1_col1\" class=\"data row1 col1\" ><a target=\"_blank\" href=\"https://www.nationaalarchief.nl/onderzoeken/archief/1.05.06/invnr/108A/file/NL-HaNA_1.05.06_108A_0001?eadId=1.05.06&unitID=108A\">https://www.nationaalarchief.nl/onderzoeken/archief/1.05.06/invnr/108A/file/NL-HaNA_1.05.06_108A_0001?eadId=1.05.06&unitID=108A</a></td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_b16bb_level0_row2\" class=\"row_heading level0 row2\" >1153</th>\n",
       "      <td id=\"T_b16bb_row2_col0\" class=\"data row2 col0\" >NL-HaNA_1.05.06_1153_0001.jpg</td>\n",
       "      <td id=\"T_b16bb_row2_col1\" class=\"data row2 col1\" ><a target=\"_blank\" href=\"https://www.nationaalarchief.nl/onderzoeken/archief/1.05.06/invnr/1153/file/NL-HaNA_1.05.06_1153_0001?eadId=1.05.06&unitID=1153\">https://www.nationaalarchief.nl/onderzoeken/archief/1.05.06/invnr/1153/file/NL-HaNA_1.05.06_1153_0001?eadId=1.05.06&unitID=1153</a></td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_b16bb_level0_row3\" class=\"row_heading level0 row3\" >1242</th>\n",
       "      <td id=\"T_b16bb_row3_col0\" class=\"data row3 col0\" >NL-HaNA_1.05.06_1242_0001.jpg</td>\n",
       "      <td id=\"T_b16bb_row3_col1\" class=\"data row3 col1\" ><a target=\"_blank\" href=\"https://www.nationaalarchief.nl/onderzoeken/archief/1.05.06/invnr/1242/file/NL-HaNA_1.05.06_1242_0001?eadId=1.05.06&unitID=1242\">https://www.nationaalarchief.nl/onderzoeken/archief/1.05.06/invnr/1242/file/NL-HaNA_1.05.06_1242_0001?eadId=1.05.06&unitID=1242</a></td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_b16bb_level0_row4\" class=\"row_heading level0 row4\" >1243</th>\n",
       "      <td id=\"T_b16bb_row4_col0\" class=\"data row4 col0\" >NL-HaNA_1.05.06_1243_0001.jpg</td>\n",
       "      <td id=\"T_b16bb_row4_col1\" class=\"data row4 col1\" ><a target=\"_blank\" href=\"https://www.nationaalarchief.nl/onderzoeken/archief/1.05.06/invnr/1243/file/NL-HaNA_1.05.06_1243_0001?eadId=1.05.06&unitID=1243\">https://www.nationaalarchief.nl/onderzoeken/archief/1.05.06/invnr/1243/file/NL-HaNA_1.05.06_1243_0001?eadId=1.05.06&unitID=1243</a></td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_b16bb_level0_row5\" class=\"row_heading level0 row5\" >169</th>\n",
       "      <td id=\"T_b16bb_row5_col0\" class=\"data row5 col0\" >NL-HaNA_1.05.06_169_0001.jpg</td>\n",
       "      <td id=\"T_b16bb_row5_col1\" class=\"data row5 col1\" ><a target=\"_blank\" href=\"https://www.nationaalarchief.nl/onderzoeken/archief/1.05.06/invnr/169/file/NL-HaNA_1.05.06_169_0001?eadId=1.05.06&unitID=169\">https://www.nationaalarchief.nl/onderzoeken/archief/1.05.06/invnr/169/file/NL-HaNA_1.05.06_169_0001?eadId=1.05.06&unitID=169</a></td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_b16bb_level0_row6\" class=\"row_heading level0 row6\" >353</th>\n",
       "      <td id=\"T_b16bb_row6_col0\" class=\"data row6 col0\" >NL-HaNA_1.05.06_353_0001.jpg</td>\n",
       "      <td id=\"T_b16bb_row6_col1\" class=\"data row6 col1\" ><a target=\"_blank\" href=\"https://www.nationaalarchief.nl/onderzoeken/archief/1.05.06/invnr/353/file/NL-HaNA_1.05.06_353_0001?eadId=1.05.06&unitID=353\">https://www.nationaalarchief.nl/onderzoeken/archief/1.05.06/invnr/353/file/NL-HaNA_1.05.06_353_0001?eadId=1.05.06&unitID=353</a></td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_b16bb_level0_row7\" class=\"row_heading level0 row7\" >886</th>\n",
       "      <td id=\"T_b16bb_row7_col0\" class=\"data row7 col0\" >NL-HaNA_1.05.06_886_0001.jpg</td>\n",
       "      <td id=\"T_b16bb_row7_col1\" class=\"data row7 col1\" ><a target=\"_blank\" href=\"https://www.nationaalarchief.nl/onderzoeken/archief/1.05.06/invnr/886/file/NL-HaNA_1.05.06_886_0001?eadId=1.05.06&unitID=886\">https://www.nationaalarchief.nl/onderzoeken/archief/1.05.06/invnr/886/file/NL-HaNA_1.05.06_886_0001?eadId=1.05.06&unitID=886</a></td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_b16bb_level0_row8\" class=\"row_heading level0 row8\" >887</th>\n",
       "      <td id=\"T_b16bb_row8_col0\" class=\"data row8 col0\" >NL-HaNA_1.05.06_887_0001.jpg</td>\n",
       "      <td id=\"T_b16bb_row8_col1\" class=\"data row8 col1\" ><a target=\"_blank\" href=\"https://www.nationaalarchief.nl/onderzoeken/archief/1.05.06/invnr/887/file/NL-HaNA_1.05.06_887_0001?eadId=1.05.06&unitID=887\">https://www.nationaalarchief.nl/onderzoeken/archief/1.05.06/invnr/887/file/NL-HaNA_1.05.06_887_0001?eadId=1.05.06&unitID=887</a></td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_b16bb_level0_row9\" class=\"row_heading level0 row9\" >888</th>\n",
       "      <td id=\"T_b16bb_row9_col0\" class=\"data row9 col0\" >NL-HaNA_1.05.06_888_0001.jpg</td>\n",
       "      <td id=\"T_b16bb_row9_col1\" class=\"data row9 col1\" ><a target=\"_blank\" href=\"https://www.nationaalarchief.nl/onderzoeken/archief/1.05.06/invnr/888/file/NL-HaNA_1.05.06_888_0001?eadId=1.05.06&unitID=888\">https://www.nationaalarchief.nl/onderzoeken/archief/1.05.06/invnr/888/file/NL-HaNA_1.05.06_888_0001?eadId=1.05.06&unitID=888</a></td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n"
      ],
      "text/plain": [
       "<pandas.io.formats.style.Styler at 0x291c87280>"
      ]
     },
     "execution_count": 40,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "line_cat_invs = lwr_rel_df.sort_values('line_width_range_800-1300', ascending=False).head(10).index\n",
    "\n",
    "(df[df.inv_num.isin(line_cat_invs)]\n",
    " .groupby('inv_num')\n",
    " .first()[['doc_id', 'iiif_url']]\n",
    " .style.format({'iiif_url': make_clickable}))"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "0804b8d0",
   "metadata": {},
   "source": [
    "All these books consist mainly of scans with a single text column on the right half of each page. Moreover, these books are all homogeneous in their line widths."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 41,
   "id": "8c02f22d",
   "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>line_width_range_0-600</th>\n",
       "      <th>line_width_range_600-800</th>\n",
       "      <th>line_width_range_800-1300</th>\n",
       "      <th>line_width_range_1300-1600</th>\n",
       "      <th>line_width_range_1600-</th>\n",
       "      <th>num_scans</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>inv_num</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>152</th>\n",
       "      <td>0.985079</td>\n",
       "      <td>0.004188</td>\n",
       "      <td>0.002356</td>\n",
       "      <td>0.001309</td>\n",
       "      <td>0.000785</td>\n",
       "      <td>13</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>114</th>\n",
       "      <td>0.979724</td>\n",
       "      <td>0.000922</td>\n",
       "      <td>0.002765</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.002765</td>\n",
       "      <td>5</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>156</th>\n",
       "      <td>0.974710</td>\n",
       "      <td>0.012303</td>\n",
       "      <td>0.002051</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>7</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>954</th>\n",
       "      <td>0.974106</td>\n",
       "      <td>0.006165</td>\n",
       "      <td>0.007398</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.001233</td>\n",
       "      <td>5</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>113</th>\n",
       "      <td>0.969639</td>\n",
       "      <td>0.011385</td>\n",
       "      <td>0.003795</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.007590</td>\n",
       "      <td>5</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1401</th>\n",
       "      <td>0.956772</td>\n",
       "      <td>0.020173</td>\n",
       "      <td>0.005764</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.008646</td>\n",
       "      <td>3</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>150</th>\n",
       "      <td>0.954787</td>\n",
       "      <td>0.002917</td>\n",
       "      <td>0.002917</td>\n",
       "      <td>0.001667</td>\n",
       "      <td>0.000625</td>\n",
       "      <td>28</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>928</th>\n",
       "      <td>0.942973</td>\n",
       "      <td>0.022003</td>\n",
       "      <td>0.014818</td>\n",
       "      <td>0.000898</td>\n",
       "      <td>0.000449</td>\n",
       "      <td>13</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>220</th>\n",
       "      <td>0.937853</td>\n",
       "      <td>0.050847</td>\n",
       "      <td>0.000628</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.001255</td>\n",
       "      <td>9</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>151</th>\n",
       "      <td>0.936875</td>\n",
       "      <td>0.018476</td>\n",
       "      <td>0.025404</td>\n",
       "      <td>0.000770</td>\n",
       "      <td>0.000770</td>\n",
       "      <td>7</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "         line_width_range_0-600  line_width_range_600-800  \\\n",
       "inv_num                                                     \n",
       "152                    0.985079                  0.004188   \n",
       "114                    0.979724                  0.000922   \n",
       "156                    0.974710                  0.012303   \n",
       "954                    0.974106                  0.006165   \n",
       "113                    0.969639                  0.011385   \n",
       "1401                   0.956772                  0.020173   \n",
       "150                    0.954787                  0.002917   \n",
       "928                    0.942973                  0.022003   \n",
       "220                    0.937853                  0.050847   \n",
       "151                    0.936875                  0.018476   \n",
       "\n",
       "         line_width_range_800-1300  line_width_range_1300-1600  \\\n",
       "inv_num                                                          \n",
       "152                       0.002356                    0.001309   \n",
       "114                       0.002765                    0.000000   \n",
       "156                       0.002051                    0.000000   \n",
       "954                       0.007398                    0.000000   \n",
       "113                       0.003795                    0.000000   \n",
       "1401                      0.005764                    0.000000   \n",
       "150                       0.002917                    0.001667   \n",
       "928                       0.014818                    0.000898   \n",
       "220                       0.000628                    0.000000   \n",
       "151                       0.025404                    0.000770   \n",
       "\n",
       "         line_width_range_1600-  num_scans  \n",
       "inv_num                                     \n",
       "152                    0.000785         13  \n",
       "114                    0.002765          5  \n",
       "156                    0.000000          7  \n",
       "954                    0.001233          5  \n",
       "113                    0.007590          5  \n",
       "1401                   0.008646          3  \n",
       "150                    0.000625         28  \n",
       "928                    0.000449         13  \n",
       "220                    0.001255          9  \n",
       "151                    0.000770          7  "
      ]
     },
     "execution_count": 41,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "lwr_rel_df.sort_values('line_width_range_0-600', ascending=False).head(10)\n",
    "\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 42,
   "id": "68f23393",
   "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>line_width_range_0-600</th>\n",
       "      <th>line_width_range_600-800</th>\n",
       "      <th>line_width_range_800-1300</th>\n",
       "      <th>line_width_range_1300-1600</th>\n",
       "      <th>line_width_range_1600-</th>\n",
       "      <th>num_scans</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>inv_num</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>1220</th>\n",
       "      <td>0.074828</td>\n",
       "      <td>0.009166</td>\n",
       "      <td>0.028821</td>\n",
       "      <td>0.885775</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>109</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>260</th>\n",
       "      <td>0.060261</td>\n",
       "      <td>0.004886</td>\n",
       "      <td>0.047231</td>\n",
       "      <td>0.884365</td>\n",
       "      <td>0.001629</td>\n",
       "      <td>8</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1226</th>\n",
       "      <td>0.069069</td>\n",
       "      <td>0.015015</td>\n",
       "      <td>0.045045</td>\n",
       "      <td>0.867868</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>5</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1195</th>\n",
       "      <td>0.093750</td>\n",
       "      <td>0.025568</td>\n",
       "      <td>0.022727</td>\n",
       "      <td>0.852273</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>5</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>24</th>\n",
       "      <td>0.135225</td>\n",
       "      <td>0.035058</td>\n",
       "      <td>0.013356</td>\n",
       "      <td>0.816361</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>7</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1193</th>\n",
       "      <td>0.137085</td>\n",
       "      <td>0.041126</td>\n",
       "      <td>0.019481</td>\n",
       "      <td>0.787157</td>\n",
       "      <td>0.012987</td>\n",
       "      <td>15</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1197</th>\n",
       "      <td>0.181818</td>\n",
       "      <td>0.017408</td>\n",
       "      <td>0.017408</td>\n",
       "      <td>0.783366</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>9</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1148</th>\n",
       "      <td>0.086957</td>\n",
       "      <td>0.086957</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.782609</td>\n",
       "      <td>0.043478</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>399</th>\n",
       "      <td>0.135965</td>\n",
       "      <td>0.028509</td>\n",
       "      <td>0.059211</td>\n",
       "      <td>0.774123</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>7</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>120</th>\n",
       "      <td>0.150259</td>\n",
       "      <td>0.020725</td>\n",
       "      <td>0.056995</td>\n",
       "      <td>0.766839</td>\n",
       "      <td>0.005181</td>\n",
       "      <td>5</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "         line_width_range_0-600  line_width_range_600-800  \\\n",
       "inv_num                                                     \n",
       "1220                   0.074828                  0.009166   \n",
       "260                    0.060261                  0.004886   \n",
       "1226                   0.069069                  0.015015   \n",
       "1195                   0.093750                  0.025568   \n",
       "24                     0.135225                  0.035058   \n",
       "1193                   0.137085                  0.041126   \n",
       "1197                   0.181818                  0.017408   \n",
       "1148                   0.086957                  0.086957   \n",
       "399                    0.135965                  0.028509   \n",
       "120                    0.150259                  0.020725   \n",
       "\n",
       "         line_width_range_800-1300  line_width_range_1300-1600  \\\n",
       "inv_num                                                          \n",
       "1220                      0.028821                    0.885775   \n",
       "260                       0.047231                    0.884365   \n",
       "1226                      0.045045                    0.867868   \n",
       "1195                      0.022727                    0.852273   \n",
       "24                        0.013356                    0.816361   \n",
       "1193                      0.019481                    0.787157   \n",
       "1197                      0.017408                    0.783366   \n",
       "1148                      0.000000                    0.782609   \n",
       "399                       0.059211                    0.774123   \n",
       "120                       0.056995                    0.766839   \n",
       "\n",
       "         line_width_range_1600-  num_scans  \n",
       "inv_num                                     \n",
       "1220                   0.000000        109  \n",
       "260                    0.001629          8  \n",
       "1226                   0.000000          5  \n",
       "1195                   0.000000          5  \n",
       "24                     0.000000          7  \n",
       "1193                   0.012987         15  \n",
       "1197                   0.000000          9  \n",
       "1148                   0.043478          2  \n",
       "399                    0.000000          7  \n",
       "120                    0.005181          5  "
      ]
     },
     "execution_count": 42,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "lwr_rel_df.sort_values('line_width_range_1300-1600', ascending=False).head(10)\n",
    "\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 43,
   "id": "8112569d",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<style type=\"text/css\">\n",
       "</style>\n",
       "<table id=\"T_b68ee\">\n",
       "  <thead>\n",
       "    <tr>\n",
       "      <th class=\"blank level0\" >&nbsp;</th>\n",
       "      <th id=\"T_b68ee_level0_col0\" class=\"col_heading level0 col0\" >doc_id</th>\n",
       "      <th id=\"T_b68ee_level0_col1\" class=\"col_heading level0 col1\" >iiif_url</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th class=\"index_name level0\" >inv_num</th>\n",
       "      <th class=\"blank col0\" >&nbsp;</th>\n",
       "      <th class=\"blank col1\" >&nbsp;</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th id=\"T_b68ee_level0_row0\" class=\"row_heading level0 row0\" >1148</th>\n",
       "      <td id=\"T_b68ee_row0_col0\" class=\"data row0 col0\" >NL-HaNA_1.05.06_1148_0001.jpg</td>\n",
       "      <td id=\"T_b68ee_row0_col1\" class=\"data row0 col1\" ><a target=\"_blank\" href=\"https://www.nationaalarchief.nl/onderzoeken/archief/1.05.06/invnr/1148/file/NL-HaNA_1.05.06_1148_0001?eadId=1.05.06&unitID=1148\">https://www.nationaalarchief.nl/onderzoeken/archief/1.05.06/invnr/1148/file/NL-HaNA_1.05.06_1148_0001?eadId=1.05.06&unitID=1148</a></td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_b68ee_level0_row1\" class=\"row_heading level0 row1\" >1193</th>\n",
       "      <td id=\"T_b68ee_row1_col0\" class=\"data row1 col0\" >NL-HaNA_1.05.06_1193_0001.jpg</td>\n",
       "      <td id=\"T_b68ee_row1_col1\" class=\"data row1 col1\" ><a target=\"_blank\" href=\"https://www.nationaalarchief.nl/onderzoeken/archief/1.05.06/invnr/1193/file/NL-HaNA_1.05.06_1193_0001?eadId=1.05.06&unitID=1193\">https://www.nationaalarchief.nl/onderzoeken/archief/1.05.06/invnr/1193/file/NL-HaNA_1.05.06_1193_0001?eadId=1.05.06&unitID=1193</a></td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_b68ee_level0_row2\" class=\"row_heading level0 row2\" >1195</th>\n",
       "      <td id=\"T_b68ee_row2_col0\" class=\"data row2 col0\" >NL-HaNA_1.05.06_1195_0001.jpg</td>\n",
       "      <td id=\"T_b68ee_row2_col1\" class=\"data row2 col1\" ><a target=\"_blank\" href=\"https://www.nationaalarchief.nl/onderzoeken/archief/1.05.06/invnr/1195/file/NL-HaNA_1.05.06_1195_0001?eadId=1.05.06&unitID=1195\">https://www.nationaalarchief.nl/onderzoeken/archief/1.05.06/invnr/1195/file/NL-HaNA_1.05.06_1195_0001?eadId=1.05.06&unitID=1195</a></td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_b68ee_level0_row3\" class=\"row_heading level0 row3\" >1197</th>\n",
       "      <td id=\"T_b68ee_row3_col0\" class=\"data row3 col0\" >NL-HaNA_1.05.06_1197_0001.jpg</td>\n",
       "      <td id=\"T_b68ee_row3_col1\" class=\"data row3 col1\" ><a target=\"_blank\" href=\"https://www.nationaalarchief.nl/onderzoeken/archief/1.05.06/invnr/1197/file/NL-HaNA_1.05.06_1197_0001?eadId=1.05.06&unitID=1197\">https://www.nationaalarchief.nl/onderzoeken/archief/1.05.06/invnr/1197/file/NL-HaNA_1.05.06_1197_0001?eadId=1.05.06&unitID=1197</a></td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_b68ee_level0_row4\" class=\"row_heading level0 row4\" >120</th>\n",
       "      <td id=\"T_b68ee_row4_col0\" class=\"data row4 col0\" >NL-HaNA_1.05.06_120_0001.jpg</td>\n",
       "      <td id=\"T_b68ee_row4_col1\" class=\"data row4 col1\" ><a target=\"_blank\" href=\"https://www.nationaalarchief.nl/onderzoeken/archief/1.05.06/invnr/120/file/NL-HaNA_1.05.06_120_0001?eadId=1.05.06&unitID=120\">https://www.nationaalarchief.nl/onderzoeken/archief/1.05.06/invnr/120/file/NL-HaNA_1.05.06_120_0001?eadId=1.05.06&unitID=120</a></td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_b68ee_level0_row5\" class=\"row_heading level0 row5\" >1220</th>\n",
       "      <td id=\"T_b68ee_row5_col0\" class=\"data row5 col0\" >NL-HaNA_1.05.06_1220_0001.jpg</td>\n",
       "      <td id=\"T_b68ee_row5_col1\" class=\"data row5 col1\" ><a target=\"_blank\" href=\"https://www.nationaalarchief.nl/onderzoeken/archief/1.05.06/invnr/1220/file/NL-HaNA_1.05.06_1220_0001?eadId=1.05.06&unitID=1220\">https://www.nationaalarchief.nl/onderzoeken/archief/1.05.06/invnr/1220/file/NL-HaNA_1.05.06_1220_0001?eadId=1.05.06&unitID=1220</a></td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_b68ee_level0_row6\" class=\"row_heading level0 row6\" >1226</th>\n",
       "      <td id=\"T_b68ee_row6_col0\" class=\"data row6 col0\" >NL-HaNA_1.05.06_1226_0001.jpg</td>\n",
       "      <td id=\"T_b68ee_row6_col1\" class=\"data row6 col1\" ><a target=\"_blank\" href=\"https://www.nationaalarchief.nl/onderzoeken/archief/1.05.06/invnr/1226/file/NL-HaNA_1.05.06_1226_0001?eadId=1.05.06&unitID=1226\">https://www.nationaalarchief.nl/onderzoeken/archief/1.05.06/invnr/1226/file/NL-HaNA_1.05.06_1226_0001?eadId=1.05.06&unitID=1226</a></td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_b68ee_level0_row7\" class=\"row_heading level0 row7\" >24</th>\n",
       "      <td id=\"T_b68ee_row7_col0\" class=\"data row7 col0\" >NL-HaNA_1.05.06_24_0001.jpg</td>\n",
       "      <td id=\"T_b68ee_row7_col1\" class=\"data row7 col1\" ><a target=\"_blank\" href=\"https://www.nationaalarchief.nl/onderzoeken/archief/1.05.06/invnr/24/file/NL-HaNA_1.05.06_24_0001?eadId=1.05.06&unitID=24\">https://www.nationaalarchief.nl/onderzoeken/archief/1.05.06/invnr/24/file/NL-HaNA_1.05.06_24_0001?eadId=1.05.06&unitID=24</a></td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_b68ee_level0_row8\" class=\"row_heading level0 row8\" >260</th>\n",
       "      <td id=\"T_b68ee_row8_col0\" class=\"data row8 col0\" >NL-HaNA_1.05.06_260_0001.jpg</td>\n",
       "      <td id=\"T_b68ee_row8_col1\" class=\"data row8 col1\" ><a target=\"_blank\" href=\"https://www.nationaalarchief.nl/onderzoeken/archief/1.05.06/invnr/260/file/NL-HaNA_1.05.06_260_0001?eadId=1.05.06&unitID=260\">https://www.nationaalarchief.nl/onderzoeken/archief/1.05.06/invnr/260/file/NL-HaNA_1.05.06_260_0001?eadId=1.05.06&unitID=260</a></td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_b68ee_level0_row9\" class=\"row_heading level0 row9\" >399</th>\n",
       "      <td id=\"T_b68ee_row9_col0\" class=\"data row9 col0\" >NL-HaNA_1.05.06_399_0001.jpg</td>\n",
       "      <td id=\"T_b68ee_row9_col1\" class=\"data row9 col1\" ><a target=\"_blank\" href=\"https://www.nationaalarchief.nl/onderzoeken/archief/1.05.06/invnr/399/file/NL-HaNA_1.05.06_399_0001?eadId=1.05.06&unitID=399\">https://www.nationaalarchief.nl/onderzoeken/archief/1.05.06/invnr/399/file/NL-HaNA_1.05.06_399_0001?eadId=1.05.06&unitID=399</a></td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n"
      ],
      "text/plain": [
       "<pandas.io.formats.style.Styler at 0x291c85ae0>"
      ]
     },
     "execution_count": 43,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "line_cat_invs = lwr_rel_df.sort_values('line_width_range_1300-1600', ascending=False).head(10).index\n",
    "\n",
    "(df[df.inv_num.isin(line_cat_invs)]\n",
    " .groupby('inv_num')\n",
    " .first()[['doc_id', 'iiif_url']]\n",
    " .style.format({'iiif_url': make_clickable}))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 44,
   "id": "22c6f23a",
   "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>line_width_range_0-600</th>\n",
       "      <th>line_width_range_600-800</th>\n",
       "      <th>line_width_range_800-1300</th>\n",
       "      <th>line_width_range_1300-1600</th>\n",
       "      <th>line_width_range_1600-</th>\n",
       "      <th>num_scans</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>inv_num</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>315</th>\n",
       "      <td>0.090909</td>\n",
       "      <td>0.272727</td>\n",
       "      <td>0.363636</td>\n",
       "      <td>0.181818</td>\n",
       "      <td>0.090909</td>\n",
       "      <td>3</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>109</th>\n",
       "      <td>0.200000</td>\n",
       "      <td>0.233333</td>\n",
       "      <td>0.066667</td>\n",
       "      <td>0.066667</td>\n",
       "      <td>0.433333</td>\n",
       "      <td>3</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1367</th>\n",
       "      <td>0.111111</td>\n",
       "      <td>0.222222</td>\n",
       "      <td>0.111111</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.555556</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>314</th>\n",
       "      <td>0.419890</td>\n",
       "      <td>0.220994</td>\n",
       "      <td>0.243094</td>\n",
       "      <td>0.044199</td>\n",
       "      <td>0.071823</td>\n",
       "      <td>5</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1151</th>\n",
       "      <td>0.237557</td>\n",
       "      <td>0.205882</td>\n",
       "      <td>0.190045</td>\n",
       "      <td>0.031674</td>\n",
       "      <td>0.332579</td>\n",
       "      <td>9</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>387A</th>\n",
       "      <td>0.162162</td>\n",
       "      <td>0.189189</td>\n",
       "      <td>0.567568</td>\n",
       "      <td>0.081081</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>3</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1402</th>\n",
       "      <td>0.753012</td>\n",
       "      <td>0.186747</td>\n",
       "      <td>0.054217</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.004016</td>\n",
       "      <td>5</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1366</th>\n",
       "      <td>0.571429</td>\n",
       "      <td>0.171429</td>\n",
       "      <td>0.257143</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1365</th>\n",
       "      <td>0.083333</td>\n",
       "      <td>0.166667</td>\n",
       "      <td>0.083333</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.666667</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>305</th>\n",
       "      <td>0.617391</td>\n",
       "      <td>0.165217</td>\n",
       "      <td>0.182609</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.026087</td>\n",
       "      <td>3</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "         line_width_range_0-600  line_width_range_600-800  \\\n",
       "inv_num                                                     \n",
       "315                    0.090909                  0.272727   \n",
       "109                    0.200000                  0.233333   \n",
       "1367                   0.111111                  0.222222   \n",
       "314                    0.419890                  0.220994   \n",
       "1151                   0.237557                  0.205882   \n",
       "387A                   0.162162                  0.189189   \n",
       "1402                   0.753012                  0.186747   \n",
       "1366                   0.571429                  0.171429   \n",
       "1365                   0.083333                  0.166667   \n",
       "305                    0.617391                  0.165217   \n",
       "\n",
       "         line_width_range_800-1300  line_width_range_1300-1600  \\\n",
       "inv_num                                                          \n",
       "315                       0.363636                    0.181818   \n",
       "109                       0.066667                    0.066667   \n",
       "1367                      0.111111                    0.000000   \n",
       "314                       0.243094                    0.044199   \n",
       "1151                      0.190045                    0.031674   \n",
       "387A                      0.567568                    0.081081   \n",
       "1402                      0.054217                    0.000000   \n",
       "1366                      0.257143                    0.000000   \n",
       "1365                      0.083333                    0.000000   \n",
       "305                       0.182609                    0.000000   \n",
       "\n",
       "         line_width_range_1600-  num_scans  \n",
       "inv_num                                     \n",
       "315                    0.090909          3  \n",
       "109                    0.433333          3  \n",
       "1367                   0.555556          2  \n",
       "314                    0.071823          5  \n",
       "1151                   0.332579          9  \n",
       "387A                   0.000000          3  \n",
       "1402                   0.004016          5  \n",
       "1366                   0.000000          2  \n",
       "1365                   0.666667          2  \n",
       "305                    0.026087          3  "
      ]
     },
     "execution_count": 44,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "lwr_rel_df.sort_values('line_width_range_600-800', ascending=False).head(10)\n",
    "\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 45,
   "id": "f3d0f9be",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<style type=\"text/css\">\n",
       "</style>\n",
       "<table id=\"T_0034a\">\n",
       "  <thead>\n",
       "    <tr>\n",
       "      <th class=\"blank level0\" >&nbsp;</th>\n",
       "      <th id=\"T_0034a_level0_col0\" class=\"col_heading level0 col0\" >doc_id</th>\n",
       "      <th id=\"T_0034a_level0_col1\" class=\"col_heading level0 col1\" >iiif_url</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th class=\"index_name level0\" >inv_num</th>\n",
       "      <th class=\"blank col0\" >&nbsp;</th>\n",
       "      <th class=\"blank col1\" >&nbsp;</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th id=\"T_0034a_level0_row0\" class=\"row_heading level0 row0\" >109</th>\n",
       "      <td id=\"T_0034a_row0_col0\" class=\"data row0 col0\" >NL-HaNA_1.05.06_109_0001.jpg</td>\n",
       "      <td id=\"T_0034a_row0_col1\" class=\"data row0 col1\" ><a target=\"_blank\" href=\"https://www.nationaalarchief.nl/onderzoeken/archief/1.05.06/invnr/109/file/NL-HaNA_1.05.06_109_0001?eadId=1.05.06&unitID=109\">https://www.nationaalarchief.nl/onderzoeken/archief/1.05.06/invnr/109/file/NL-HaNA_1.05.06_109_0001?eadId=1.05.06&unitID=109</a></td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_0034a_level0_row1\" class=\"row_heading level0 row1\" >1151</th>\n",
       "      <td id=\"T_0034a_row1_col0\" class=\"data row1 col0\" >NL-HaNA_1.05.06_1151_0001.jpg</td>\n",
       "      <td id=\"T_0034a_row1_col1\" class=\"data row1 col1\" ><a target=\"_blank\" href=\"https://www.nationaalarchief.nl/onderzoeken/archief/1.05.06/invnr/1151/file/NL-HaNA_1.05.06_1151_0001?eadId=1.05.06&unitID=1151\">https://www.nationaalarchief.nl/onderzoeken/archief/1.05.06/invnr/1151/file/NL-HaNA_1.05.06_1151_0001?eadId=1.05.06&unitID=1151</a></td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_0034a_level0_row2\" class=\"row_heading level0 row2\" >1365</th>\n",
       "      <td id=\"T_0034a_row2_col0\" class=\"data row2 col0\" >NL-HaNA_1.05.06_1365_0001.jpg</td>\n",
       "      <td id=\"T_0034a_row2_col1\" class=\"data row2 col1\" ><a target=\"_blank\" href=\"https://www.nationaalarchief.nl/onderzoeken/archief/1.05.06/invnr/1365/file/NL-HaNA_1.05.06_1365_0001?eadId=1.05.06&unitID=1365\">https://www.nationaalarchief.nl/onderzoeken/archief/1.05.06/invnr/1365/file/NL-HaNA_1.05.06_1365_0001?eadId=1.05.06&unitID=1365</a></td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_0034a_level0_row3\" class=\"row_heading level0 row3\" >1366</th>\n",
       "      <td id=\"T_0034a_row3_col0\" class=\"data row3 col0\" >NL-HaNA_1.05.06_1366_0001.jpg</td>\n",
       "      <td id=\"T_0034a_row3_col1\" class=\"data row3 col1\" ><a target=\"_blank\" href=\"https://www.nationaalarchief.nl/onderzoeken/archief/1.05.06/invnr/1366/file/NL-HaNA_1.05.06_1366_0001?eadId=1.05.06&unitID=1366\">https://www.nationaalarchief.nl/onderzoeken/archief/1.05.06/invnr/1366/file/NL-HaNA_1.05.06_1366_0001?eadId=1.05.06&unitID=1366</a></td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_0034a_level0_row4\" class=\"row_heading level0 row4\" >1367</th>\n",
       "      <td id=\"T_0034a_row4_col0\" class=\"data row4 col0\" >NL-HaNA_1.05.06_1367_0001.jpg</td>\n",
       "      <td id=\"T_0034a_row4_col1\" class=\"data row4 col1\" ><a target=\"_blank\" href=\"https://www.nationaalarchief.nl/onderzoeken/archief/1.05.06/invnr/1367/file/NL-HaNA_1.05.06_1367_0001?eadId=1.05.06&unitID=1367\">https://www.nationaalarchief.nl/onderzoeken/archief/1.05.06/invnr/1367/file/NL-HaNA_1.05.06_1367_0001?eadId=1.05.06&unitID=1367</a></td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_0034a_level0_row5\" class=\"row_heading level0 row5\" >1402</th>\n",
       "      <td id=\"T_0034a_row5_col0\" class=\"data row5 col0\" >NL-HaNA_1.05.06_1402_0001.jpg</td>\n",
       "      <td id=\"T_0034a_row5_col1\" class=\"data row5 col1\" ><a target=\"_blank\" href=\"https://www.nationaalarchief.nl/onderzoeken/archief/1.05.06/invnr/1402/file/NL-HaNA_1.05.06_1402_0001?eadId=1.05.06&unitID=1402\">https://www.nationaalarchief.nl/onderzoeken/archief/1.05.06/invnr/1402/file/NL-HaNA_1.05.06_1402_0001?eadId=1.05.06&unitID=1402</a></td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_0034a_level0_row6\" class=\"row_heading level0 row6\" >305</th>\n",
       "      <td id=\"T_0034a_row6_col0\" class=\"data row6 col0\" >NL-HaNA_1.05.06_305_0001.jpg</td>\n",
       "      <td id=\"T_0034a_row6_col1\" class=\"data row6 col1\" ><a target=\"_blank\" href=\"https://www.nationaalarchief.nl/onderzoeken/archief/1.05.06/invnr/305/file/NL-HaNA_1.05.06_305_0001?eadId=1.05.06&unitID=305\">https://www.nationaalarchief.nl/onderzoeken/archief/1.05.06/invnr/305/file/NL-HaNA_1.05.06_305_0001?eadId=1.05.06&unitID=305</a></td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_0034a_level0_row7\" class=\"row_heading level0 row7\" >314</th>\n",
       "      <td id=\"T_0034a_row7_col0\" class=\"data row7 col0\" >NL-HaNA_1.05.06_314_0001.jpg</td>\n",
       "      <td id=\"T_0034a_row7_col1\" class=\"data row7 col1\" ><a target=\"_blank\" href=\"https://www.nationaalarchief.nl/onderzoeken/archief/1.05.06/invnr/314/file/NL-HaNA_1.05.06_314_0001?eadId=1.05.06&unitID=314\">https://www.nationaalarchief.nl/onderzoeken/archief/1.05.06/invnr/314/file/NL-HaNA_1.05.06_314_0001?eadId=1.05.06&unitID=314</a></td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_0034a_level0_row8\" class=\"row_heading level0 row8\" >315</th>\n",
       "      <td id=\"T_0034a_row8_col0\" class=\"data row8 col0\" >NL-HaNA_1.05.06_315_0001.jpg</td>\n",
       "      <td id=\"T_0034a_row8_col1\" class=\"data row8 col1\" ><a target=\"_blank\" href=\"https://www.nationaalarchief.nl/onderzoeken/archief/1.05.06/invnr/315/file/NL-HaNA_1.05.06_315_0001?eadId=1.05.06&unitID=315\">https://www.nationaalarchief.nl/onderzoeken/archief/1.05.06/invnr/315/file/NL-HaNA_1.05.06_315_0001?eadId=1.05.06&unitID=315</a></td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_0034a_level0_row9\" class=\"row_heading level0 row9\" >387A</th>\n",
       "      <td id=\"T_0034a_row9_col0\" class=\"data row9 col0\" >NL-HaNA_1.05.06_387A_0001.jpg</td>\n",
       "      <td id=\"T_0034a_row9_col1\" class=\"data row9 col1\" ><a target=\"_blank\" href=\"https://www.nationaalarchief.nl/onderzoeken/archief/1.05.06/invnr/387A/file/NL-HaNA_1.05.06_387A_0001?eadId=1.05.06&unitID=387A\">https://www.nationaalarchief.nl/onderzoeken/archief/1.05.06/invnr/387A/file/NL-HaNA_1.05.06_387A_0001?eadId=1.05.06&unitID=387A</a></td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n"
      ],
      "text/plain": [
       "<pandas.io.formats.style.Styler at 0x291c850f0>"
      ]
     },
     "execution_count": 45,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "line_cat_invs = lwr_rel_df.sort_values('line_width_range_600-800', ascending=False).head(10).index\n",
    "\n",
    "(df[df.inv_num.isin(line_cat_invs)]\n",
    " .groupby('inv_num')\n",
    " .first()[['doc_id', 'iiif_url']]\n",
    " .style.format({'iiif_url': make_clickable}))"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "9ed8bf18",
   "metadata": {},
   "source": [
    "#### Heterogeneous line widths\n",
    "\n",
    "If we represent inventories by the line width category with the maximum fraction, and then sort inventories by the lowest maximum fraction, we get a list of inventories that are most diverse in terms of line width."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 46,
   "id": "0d8e6c97",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "inv_num\n",
       "262     0.283019\n",
       "690     0.306094\n",
       "1342    0.314286\n",
       "40      0.319568\n",
       "708     0.325000\n",
       "43      0.325301\n",
       "1151    0.332579\n",
       "1378    0.333333\n",
       "1397    0.338583\n",
       "481     0.344519\n",
       "dtype: float64"
      ]
     },
     "execution_count": 46,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "lwr_rel_df[lwr_cols].max(axis=1).sort_values().head(10)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 47,
   "id": "acb06f65",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<style type=\"text/css\">\n",
       "</style>\n",
       "<table id=\"T_4ed60\">\n",
       "  <thead>\n",
       "    <tr>\n",
       "      <th class=\"blank level0\" >&nbsp;</th>\n",
       "      <th id=\"T_4ed60_level0_col0\" class=\"col_heading level0 col0\" >doc_id</th>\n",
       "      <th id=\"T_4ed60_level0_col1\" class=\"col_heading level0 col1\" >num_scans</th>\n",
       "      <th id=\"T_4ed60_level0_col2\" class=\"col_heading level0 col2\" >iiif_url</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th class=\"index_name level0\" >inv_num</th>\n",
       "      <th class=\"blank col0\" >&nbsp;</th>\n",
       "      <th class=\"blank col1\" >&nbsp;</th>\n",
       "      <th class=\"blank col2\" >&nbsp;</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th id=\"T_4ed60_level0_row0\" class=\"row_heading level0 row0\" >1151</th>\n",
       "      <td id=\"T_4ed60_row0_col0\" class=\"data row0 col0\" >NL-HaNA_1.05.06_1151_0001.jpg</td>\n",
       "      <td id=\"T_4ed60_row0_col1\" class=\"data row0 col1\" >9</td>\n",
       "      <td id=\"T_4ed60_row0_col2\" class=\"data row0 col2\" ><a target=\"_blank\" href=\"https://www.nationaalarchief.nl/onderzoeken/archief/1.05.06/invnr/1151/file/NL-HaNA_1.05.06_1151_0001?eadId=1.05.06&unitID=1151\">https://www.nationaalarchief.nl/onderzoeken/archief/1.05.06/invnr/1151/file/NL-HaNA_1.05.06_1151_0001?eadId=1.05.06&unitID=1151</a></td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_4ed60_level0_row1\" class=\"row_heading level0 row1\" >1342</th>\n",
       "      <td id=\"T_4ed60_row1_col0\" class=\"data row1 col0\" >NL-HaNA_1.05.06_1342_0001.jpg</td>\n",
       "      <td id=\"T_4ed60_row1_col1\" class=\"data row1 col1\" >3</td>\n",
       "      <td id=\"T_4ed60_row1_col2\" class=\"data row1 col2\" ><a target=\"_blank\" href=\"https://www.nationaalarchief.nl/onderzoeken/archief/1.05.06/invnr/1342/file/NL-HaNA_1.05.06_1342_0001?eadId=1.05.06&unitID=1342\">https://www.nationaalarchief.nl/onderzoeken/archief/1.05.06/invnr/1342/file/NL-HaNA_1.05.06_1342_0001?eadId=1.05.06&unitID=1342</a></td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_4ed60_level0_row2\" class=\"row_heading level0 row2\" >1378</th>\n",
       "      <td id=\"T_4ed60_row2_col0\" class=\"data row2 col0\" >NL-HaNA_1.05.06_1378_0001.jpg</td>\n",
       "      <td id=\"T_4ed60_row2_col1\" class=\"data row2 col1\" >3</td>\n",
       "      <td id=\"T_4ed60_row2_col2\" class=\"data row2 col2\" ><a target=\"_blank\" href=\"https://www.nationaalarchief.nl/onderzoeken/archief/1.05.06/invnr/1378/file/NL-HaNA_1.05.06_1378_0001?eadId=1.05.06&unitID=1378\">https://www.nationaalarchief.nl/onderzoeken/archief/1.05.06/invnr/1378/file/NL-HaNA_1.05.06_1378_0001?eadId=1.05.06&unitID=1378</a></td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_4ed60_level0_row3\" class=\"row_heading level0 row3\" >1397</th>\n",
       "      <td id=\"T_4ed60_row3_col0\" class=\"data row3 col0\" >NL-HaNA_1.05.06_1397_0001.jpg</td>\n",
       "      <td id=\"T_4ed60_row3_col1\" class=\"data row3 col1\" >3</td>\n",
       "      <td id=\"T_4ed60_row3_col2\" class=\"data row3 col2\" ><a target=\"_blank\" href=\"https://www.nationaalarchief.nl/onderzoeken/archief/1.05.06/invnr/1397/file/NL-HaNA_1.05.06_1397_0001?eadId=1.05.06&unitID=1397\">https://www.nationaalarchief.nl/onderzoeken/archief/1.05.06/invnr/1397/file/NL-HaNA_1.05.06_1397_0001?eadId=1.05.06&unitID=1397</a></td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_4ed60_level0_row4\" class=\"row_heading level0 row4\" >262</th>\n",
       "      <td id=\"T_4ed60_row4_col0\" class=\"data row4 col0\" >NL-HaNA_1.05.06_262_0001.jpg</td>\n",
       "      <td id=\"T_4ed60_row4_col1\" class=\"data row4 col1\" >3</td>\n",
       "      <td id=\"T_4ed60_row4_col2\" class=\"data row4 col2\" ><a target=\"_blank\" href=\"https://www.nationaalarchief.nl/onderzoeken/archief/1.05.06/invnr/262/file/NL-HaNA_1.05.06_262_0001?eadId=1.05.06&unitID=262\">https://www.nationaalarchief.nl/onderzoeken/archief/1.05.06/invnr/262/file/NL-HaNA_1.05.06_262_0001?eadId=1.05.06&unitID=262</a></td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_4ed60_level0_row5\" class=\"row_heading level0 row5\" >40</th>\n",
       "      <td id=\"T_4ed60_row5_col0\" class=\"data row5 col0\" >NL-HaNA_1.05.06_40_0001.jpg</td>\n",
       "      <td id=\"T_4ed60_row5_col1\" class=\"data row5 col1\" >146</td>\n",
       "      <td id=\"T_4ed60_row5_col2\" class=\"data row5 col2\" ><a target=\"_blank\" href=\"https://www.nationaalarchief.nl/onderzoeken/archief/1.05.06/invnr/40/file/NL-HaNA_1.05.06_40_0001?eadId=1.05.06&unitID=40\">https://www.nationaalarchief.nl/onderzoeken/archief/1.05.06/invnr/40/file/NL-HaNA_1.05.06_40_0001?eadId=1.05.06&unitID=40</a></td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_4ed60_level0_row6\" class=\"row_heading level0 row6\" >43</th>\n",
       "      <td id=\"T_4ed60_row6_col0\" class=\"data row6 col0\" >NL-HaNA_1.05.06_43_0001.jpg</td>\n",
       "      <td id=\"T_4ed60_row6_col1\" class=\"data row6 col1\" >3</td>\n",
       "      <td id=\"T_4ed60_row6_col2\" class=\"data row6 col2\" ><a target=\"_blank\" href=\"https://www.nationaalarchief.nl/onderzoeken/archief/1.05.06/invnr/43/file/NL-HaNA_1.05.06_43_0001?eadId=1.05.06&unitID=43\">https://www.nationaalarchief.nl/onderzoeken/archief/1.05.06/invnr/43/file/NL-HaNA_1.05.06_43_0001?eadId=1.05.06&unitID=43</a></td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_4ed60_level0_row7\" class=\"row_heading level0 row7\" >481</th>\n",
       "      <td id=\"T_4ed60_row7_col0\" class=\"data row7 col0\" >NL-HaNA_1.05.06_481_0001.jpg</td>\n",
       "      <td id=\"T_4ed60_row7_col1\" class=\"data row7 col1\" >21</td>\n",
       "      <td id=\"T_4ed60_row7_col2\" class=\"data row7 col2\" ><a target=\"_blank\" href=\"https://www.nationaalarchief.nl/onderzoeken/archief/1.05.06/invnr/481/file/NL-HaNA_1.05.06_481_0001?eadId=1.05.06&unitID=481\">https://www.nationaalarchief.nl/onderzoeken/archief/1.05.06/invnr/481/file/NL-HaNA_1.05.06_481_0001?eadId=1.05.06&unitID=481</a></td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_4ed60_level0_row8\" class=\"row_heading level0 row8\" >690</th>\n",
       "      <td id=\"T_4ed60_row8_col0\" class=\"data row8 col0\" >NL-HaNA_1.05.06_690_0001.jpg</td>\n",
       "      <td id=\"T_4ed60_row8_col1\" class=\"data row8 col1\" >23</td>\n",
       "      <td id=\"T_4ed60_row8_col2\" class=\"data row8 col2\" ><a target=\"_blank\" href=\"https://www.nationaalarchief.nl/onderzoeken/archief/1.05.06/invnr/690/file/NL-HaNA_1.05.06_690_0001?eadId=1.05.06&unitID=690\">https://www.nationaalarchief.nl/onderzoeken/archief/1.05.06/invnr/690/file/NL-HaNA_1.05.06_690_0001?eadId=1.05.06&unitID=690</a></td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th id=\"T_4ed60_level0_row9\" class=\"row_heading level0 row9\" >708</th>\n",
       "      <td id=\"T_4ed60_row9_col0\" class=\"data row9 col0\" >NL-HaNA_1.05.06_708_0001.jpg</td>\n",
       "      <td id=\"T_4ed60_row9_col1\" class=\"data row9 col1\" >3</td>\n",
       "      <td id=\"T_4ed60_row9_col2\" class=\"data row9 col2\" ><a target=\"_blank\" href=\"https://www.nationaalarchief.nl/onderzoeken/archief/1.05.06/invnr/708/file/NL-HaNA_1.05.06_708_0001?eadId=1.05.06&unitID=708\">https://www.nationaalarchief.nl/onderzoeken/archief/1.05.06/invnr/708/file/NL-HaNA_1.05.06_708_0001?eadId=1.05.06&unitID=708</a></td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n"
      ],
      "text/plain": [
       "<pandas.io.formats.style.Styler at 0x28a863730>"
      ]
     },
     "execution_count": 47,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "line_cat_invs = lwr_rel_df[lwr_cols].max(axis=1).sort_values().head(10).index\n",
    "\n",
    "(df[df.inv_num.isin(line_cat_invs)]\n",
    " .groupby('inv_num')\n",
    " .first()[['doc_id', 'num_scans', 'iiif_url']]\n",
    " .style.format({'iiif_url': make_clickable}))"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "8494a268",
   "metadata": {},
   "source": [
    "We can also analyse the distribution of line widths across scans and check what the maximum fraction is. This shows how homogeneous the line widths of a scan are."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 62,
   "id": "23608645",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0        0.818182\n",
       "1        0.915254\n",
       "2        0.870370\n",
       "3             NaN\n",
       "4             NaN\n",
       "           ...   \n",
       "15852    0.760870\n",
       "15853    0.886364\n",
       "15854    0.750000\n",
       "15855    0.681818\n",
       "15856    0.518519\n",
       "Length: 15857, dtype: float64"
      ]
     },
     "execution_count": 62,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df[lwr_cols].div(df.lines, axis=0).max(axis=1)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "8d30879f",
   "metadata": {},
   "source": [
    "The above shows that in the first scan, the line width category with the highest fraction represents 82% of the lines. **In other words, this scans is fairly homogeneous in terms of line widths.**\n",
    "\n",
    "If we plot the distribution of these maximum fractions, we can see how homogeneous or heterogeneous scans in this archive are."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 64,
   "id": "0c350557",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "sns.kdeplot(data=df[lwr_cols].div(df.lines, axis=0).max(axis=1));"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "e0461ecc",
   "metadata": {},
   "source": [
    "This shows that many scans are homogeneous (over 80% of lines are in the same line width category), but also a large chunk of scans are more heterogeneous in line widths."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "4d7babe0",
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3 (ipykernel)",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.10.9"
  },
  "toc": {
   "base_numbering": 1,
   "nav_menu": {},
   "number_sections": true,
   "sideBar": true,
   "skip_h1_title": false,
   "title_cell": "Table of Contents",
   "title_sidebar": "Contents",
   "toc_cell": false,
   "toc_position": {},
   "toc_section_display": true,
   "toc_window_display": false
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}