12-DA_Pandas_realworld.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# 12. A complete example"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "import pandas as pd\n",
    "import numpy as np\n",
    "import matplotlib.pyplot as plt\n",
    "\n",
    "import seaborn as sns"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "We have seen now most of the basic features of Pandas including importing data, combining dataframes, aggregating information and plotting it. In this chapter, we are going to re-use these concepts with the real data seen in the [introduction chapter](06-DA_Pandas_introduction.ipynb). We are also going to explore some more advanced plotting libraries that exploit to the maximum dataframe structures."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## 12.1 Importing data"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "We are importing here two tables provided openly by the Swiss National Science Foundation. One contains a list of all *projects* to which funds have been allocated since 1975. The other table contains a list of all *people* to which funds have been awarded during the same period:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [],
   "source": [
    "# local import\n",
    "projects = pd.read_csv('Data/P3_GrantExport.csv',sep = ';')\n",
    "persons = pd.read_csv('Data/P3_PersonExport.csv',sep = ';')\n",
    "\n",
    "# import from url\n",
    "#projects = pd.read_csv('http://p3.snf.ch/P3Export/P3_GrantExport.csv',sep = ';')\n",
    "#persons = pd.read_csv('http://p3.snf.ch/P3Export/P3_PersonExport.csv',sep = ';')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "We can have a brief look at both tables:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "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>Project Number</th>\n",
       "      <th>Project Number String</th>\n",
       "      <th>Project Title</th>\n",
       "      <th>Project Title English</th>\n",
       "      <th>Responsible Applicant</th>\n",
       "      <th>Funding Instrument</th>\n",
       "      <th>Funding Instrument Hierarchy</th>\n",
       "      <th>Institution</th>\n",
       "      <th>Institution Country</th>\n",
       "      <th>University</th>\n",
       "      <th>Discipline Number</th>\n",
       "      <th>Discipline Name</th>\n",
       "      <th>Discipline Name Hierarchy</th>\n",
       "      <th>All disciplines</th>\n",
       "      <th>Start Date</th>\n",
       "      <th>End Date</th>\n",
       "      <th>Approved Amount</th>\n",
       "      <th>Keywords</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>1</td>\n",
       "      <td>1000-000001</td>\n",
       "      <td>Schlussband (Bd. VI) der Jacob Burckhardt-Biog...</td>\n",
       "      <td>NaN</td>\n",
       "      <td>Kaegi Werner</td>\n",
       "      <td>Project funding (Div. I-III)</td>\n",
       "      <td>Project funding</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>Unassignable - NA</td>\n",
       "      <td>10302</td>\n",
       "      <td>Swiss history</td>\n",
       "      <td>Humanities and Social Sciences;Theology &amp; reli...</td>\n",
       "      <td>10302</td>\n",
       "      <td>01.10.1975</td>\n",
       "      <td>30.09.1976</td>\n",
       "      <td>11619.00</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>4</td>\n",
       "      <td>1000-000004</td>\n",
       "      <td>Batterie de tests à l'usage des enseignants po...</td>\n",
       "      <td>NaN</td>\n",
       "      <td>Massarenti Léonard</td>\n",
       "      <td>Project funding (Div. I-III)</td>\n",
       "      <td>Project funding</td>\n",
       "      <td>Faculté de Psychologie et des Sciences de l'Ed...</td>\n",
       "      <td>Switzerland</td>\n",
       "      <td>University of Geneva - GE</td>\n",
       "      <td>10104</td>\n",
       "      <td>Educational science and Pedagogy</td>\n",
       "      <td>Humanities and Social Sciences;Psychology, edu...</td>\n",
       "      <td>10104</td>\n",
       "      <td>01.10.1975</td>\n",
       "      <td>30.09.1976</td>\n",
       "      <td>41022.00</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>5</td>\n",
       "      <td>1000-000005</td>\n",
       "      <td>Kritische Erstausgabe der 'Evidentiae contra D...</td>\n",
       "      <td>NaN</td>\n",
       "      <td>Kommission für das Corpus philosophorum medii ...</td>\n",
       "      <td>Project funding (Div. I-III)</td>\n",
       "      <td>Project funding</td>\n",
       "      <td>Kommission für das Corpus philosophorum medii ...</td>\n",
       "      <td>Switzerland</td>\n",
       "      <td>Non-profit organisations (libraries, museums, ...</td>\n",
       "      <td>10101</td>\n",
       "      <td>Philosophy</td>\n",
       "      <td>Humanities and Social Sciences;Linguistics and...</td>\n",
       "      <td>10101</td>\n",
       "      <td>01.03.1976</td>\n",
       "      <td>28.02.1985</td>\n",
       "      <td>79732.00</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>6</td>\n",
       "      <td>1000-000006</td>\n",
       "      <td>Katalog der datierten Handschriften in der Sch...</td>\n",
       "      <td>NaN</td>\n",
       "      <td>Burckhardt Max</td>\n",
       "      <td>Project funding (Div. I-III)</td>\n",
       "      <td>Project funding</td>\n",
       "      <td>Abteilung Handschriften und Alte Drucke Univer...</td>\n",
       "      <td>Switzerland</td>\n",
       "      <td>University of Basel - BS</td>\n",
       "      <td>10302</td>\n",
       "      <td>Swiss history</td>\n",
       "      <td>Humanities and Social Sciences;Theology &amp; reli...</td>\n",
       "      <td>10302</td>\n",
       "      <td>01.10.1975</td>\n",
       "      <td>30.09.1976</td>\n",
       "      <td>52627.00</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>7</td>\n",
       "      <td>1000-000007</td>\n",
       "      <td>Wissenschaftliche Mitarbeit am Thesaurus Lingu...</td>\n",
       "      <td>NaN</td>\n",
       "      <td>Schweiz. Thesauruskommission</td>\n",
       "      <td>Project funding (Div. I-III)</td>\n",
       "      <td>Project funding</td>\n",
       "      <td>Schweiz. Thesauruskommission</td>\n",
       "      <td>Switzerland</td>\n",
       "      <td>Non-profit organisations (libraries, museums, ...</td>\n",
       "      <td>10303</td>\n",
       "      <td>Ancient history and Classical studies</td>\n",
       "      <td>Humanities and Social Sciences;Theology &amp; reli...</td>\n",
       "      <td>10303</td>\n",
       "      <td>01.01.1976</td>\n",
       "      <td>30.04.1978</td>\n",
       "      <td>120042.00</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   Project Number Project Number String  \\\n",
       "0               1           1000-000001   \n",
       "1               4           1000-000004   \n",
       "2               5           1000-000005   \n",
       "3               6           1000-000006   \n",
       "4               7           1000-000007   \n",
       "\n",
       "                                       Project Title Project Title English  \\\n",
       "0  Schlussband (Bd. VI) der Jacob Burckhardt-Biog...                   NaN   \n",
       "1  Batterie de tests à l'usage des enseignants po...                   NaN   \n",
       "2  Kritische Erstausgabe der 'Evidentiae contra D...                   NaN   \n",
       "3  Katalog der datierten Handschriften in der Sch...                   NaN   \n",
       "4  Wissenschaftliche Mitarbeit am Thesaurus Lingu...                   NaN   \n",
       "\n",
       "                               Responsible Applicant  \\\n",
       "0                                       Kaegi Werner   \n",
       "1                                 Massarenti Léonard   \n",
       "2  Kommission für das Corpus philosophorum medii ...   \n",
       "3                                     Burckhardt Max   \n",
       "4                       Schweiz. Thesauruskommission   \n",
       "\n",
       "             Funding Instrument Funding Instrument Hierarchy  \\\n",
       "0  Project funding (Div. I-III)              Project funding   \n",
       "1  Project funding (Div. I-III)              Project funding   \n",
       "2  Project funding (Div. I-III)              Project funding   \n",
       "3  Project funding (Div. I-III)              Project funding   \n",
       "4  Project funding (Div. I-III)              Project funding   \n",
       "\n",
       "                                         Institution Institution Country  \\\n",
       "0                                                NaN                 NaN   \n",
       "1  Faculté de Psychologie et des Sciences de l'Ed...         Switzerland   \n",
       "2  Kommission für das Corpus philosophorum medii ...         Switzerland   \n",
       "3  Abteilung Handschriften und Alte Drucke Univer...         Switzerland   \n",
       "4                       Schweiz. Thesauruskommission         Switzerland   \n",
       "\n",
       "                                          University  Discipline Number  \\\n",
       "0                                  Unassignable - NA              10302   \n",
       "1                          University of Geneva - GE              10104   \n",
       "2  Non-profit organisations (libraries, museums, ...              10101   \n",
       "3                           University of Basel - BS              10302   \n",
       "4  Non-profit organisations (libraries, museums, ...              10303   \n",
       "\n",
       "                         Discipline Name  \\\n",
       "0                          Swiss history   \n",
       "1       Educational science and Pedagogy   \n",
       "2                             Philosophy   \n",
       "3                          Swiss history   \n",
       "4  Ancient history and Classical studies   \n",
       "\n",
       "                           Discipline Name Hierarchy All disciplines  \\\n",
       "0  Humanities and Social Sciences;Theology & reli...           10302   \n",
       "1  Humanities and Social Sciences;Psychology, edu...           10104   \n",
       "2  Humanities and Social Sciences;Linguistics and...           10101   \n",
       "3  Humanities and Social Sciences;Theology & reli...           10302   \n",
       "4  Humanities and Social Sciences;Theology & reli...           10303   \n",
       "\n",
       "   Start Date    End Date Approved Amount Keywords  \n",
       "0  01.10.1975  30.09.1976        11619.00      NaN  \n",
       "1  01.10.1975  30.09.1976        41022.00      NaN  \n",
       "2  01.03.1976  28.02.1985        79732.00      NaN  \n",
       "3  01.10.1975  30.09.1976        52627.00      NaN  \n",
       "4  01.01.1976  30.04.1978       120042.00      NaN  "
      ]
     },
     "execution_count": 8,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "projects.head(5)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "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>Last Name</th>\n",
       "      <th>First Name</th>\n",
       "      <th>Gender</th>\n",
       "      <th>Institute Name</th>\n",
       "      <th>Institute Place</th>\n",
       "      <th>Person ID SNSF</th>\n",
       "      <th>OCRID</th>\n",
       "      <th>Projects as responsible Applicant</th>\n",
       "      <th>Projects as Applicant</th>\n",
       "      <th>Projects as Partner</th>\n",
       "      <th>Projects as Practice Partner</th>\n",
       "      <th>Projects as Employee</th>\n",
       "      <th>Projects as Contact Person</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>a Marca</td>\n",
       "      <td>Davide</td>\n",
       "      <td>male</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>53856</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>36549</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>a Marca</td>\n",
       "      <td>Andrea</td>\n",
       "      <td>male</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>132628</td>\n",
       "      <td>NaN</td>\n",
       "      <td>67368</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>A. Jafari</td>\n",
       "      <td>Golnaz</td>\n",
       "      <td>female</td>\n",
       "      <td>Universität Luzern</td>\n",
       "      <td>Luzern</td>\n",
       "      <td>747886</td>\n",
       "      <td>NaN</td>\n",
       "      <td>191432</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>Aaberg</td>\n",
       "      <td>Johan</td>\n",
       "      <td>male</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>575257</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>119868</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>Aahman</td>\n",
       "      <td>Josefin</td>\n",
       "      <td>female</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>629557</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>141014</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   Last Name First Name  Gender      Institute Name Institute Place  \\\n",
       "0    a Marca     Davide    male                 NaN             NaN   \n",
       "1    a Marca     Andrea    male                 NaN             NaN   \n",
       "2  A. Jafari     Golnaz  female  Universität Luzern          Luzern   \n",
       "3     Aaberg      Johan    male                 NaN             NaN   \n",
       "4     Aahman    Josefin  female                 NaN             NaN   \n",
       "\n",
       "   Person ID SNSF OCRID Projects as responsible Applicant  \\\n",
       "0           53856   NaN                               NaN   \n",
       "1          132628   NaN                             67368   \n",
       "2          747886   NaN                            191432   \n",
       "3          575257   NaN                               NaN   \n",
       "4          629557   NaN                               NaN   \n",
       "\n",
       "  Projects as Applicant Projects as Partner Projects as Practice Partner  \\\n",
       "0                   NaN                 NaN                          NaN   \n",
       "1                   NaN                 NaN                          NaN   \n",
       "2                   NaN                 NaN                          NaN   \n",
       "3                   NaN                 NaN                          NaN   \n",
       "4                   NaN                 NaN                          NaN   \n",
       "\n",
       "  Projects as Employee Projects as Contact Person  \n",
       "0                36549                        NaN  \n",
       "1                  NaN                        NaN  \n",
       "2                  NaN                        NaN  \n",
       "3               119868                        NaN  \n",
       "4               141014                        NaN  "
      ]
     },
     "execution_count": 9,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "persons.head(5)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "We see that the ```persons``` table gives information such as the role of a person in various projects  (applicant, employee etc.), her/his gender etc. The *project* table on the other side gives information such as the period of a grant, how much money was awarded etc.\n",
    "\n",
    "What if we now wish to know for example:\n",
    "- How much money is awarded on average depending on gender?\n",
    "- How long does it typically take for a researcher to go from employee to applicant status on a grant?\n",
    "\n",
    "We need a way to *link* the two tables, i.e. create a large table where *each row* corresponds to a single *observation* containing information from the two tables such as: applicant, gender, awarded funds, dates etc. We will now go through all necessary steps to achieve that goal."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## 12.2 Merging tables\n",
    "\n",
    "If each row of the persons table contained a single observation with a single person and a single project (the same person would appear of course multiple times), we could just *join* the two tables based e.g. on the project ID. Unfortunately, in the persons table, each line corresponds to a *single researcher* with all projects IDs lumped together in a list. For example:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Last Name                                                                    Bodenmann\n",
       "First Name                                                                         Guy\n",
       "Gender                                                                            male\n",
       "Institute Name                       Lehrstuhl für Klinische Psychologie Kinder/Jug...\n",
       "Institute Place                                                                 Zürich\n",
       "Person ID SNSF                                                                   47670\n",
       "OCRID                                                              0000-0003-0964-6409\n",
       "Projects as responsible Applicant    46820;56660;62901;109547;115948;128960;129627;...\n",
       "Projects as Applicant                                             112141;122090;166348\n",
       "Projects as Partner                                                                NaN\n",
       "Projects as Practice Partner                                                       NaN\n",
       "Projects as Employee                                                             62901\n",
       "Projects as Contact Person                                                         NaN\n",
       "Name: 10041, dtype: object"
      ]
     },
     "execution_count": 12,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "persons.iloc[10041]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "'46820;56660;62901;109547;115948;128960;129627;129699;133004;146775;147634;173270'"
      ]
     },
     "execution_count": 13,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "persons.iloc[10041]['Projects as responsible Applicant']"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Therefore the first thing we need to do is to split those strings into actual lists. We can do that by using classic Python string splitting. We simply ```apply``` that function to the relevant columns. We need to take care of rows containing NaNs on which we cannot use ```split()```. We create two series, one for applicants, one for employees:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {},
   "outputs": [],
   "source": [
    "projID_a = persons['Projects as responsible Applicant'].apply(lambda x: x.split(';') if not pd.isna(x) else np.nan)\n",
    "projID_e = persons['Projects as Employee'].apply(lambda x: x.split(';') if not pd.isna(x) else np.nan)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0                                             NaN\n",
       "1                                         [67368]\n",
       "2                                        [191432]\n",
       "3                                             NaN\n",
       "4                                             NaN\n",
       "                           ...                   \n",
       "110811    [52821, 143769, 147153, 165510, 183584]\n",
       "110812                                        NaN\n",
       "110813                                        NaN\n",
       "110814                                        NaN\n",
       "110815                                        NaN\n",
       "Name: Projects as responsible Applicant, Length: 110816, dtype: object"
      ]
     },
     "execution_count": 15,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "projID_a"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "['46820',\n",
       " '56660',\n",
       " '62901',\n",
       " '109547',\n",
       " '115948',\n",
       " '128960',\n",
       " '129627',\n",
       " '129699',\n",
       " '133004',\n",
       " '146775',\n",
       " '147634',\n",
       " '173270']"
      ]
     },
     "execution_count": 17,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "projID_a[10041]"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Now, to avoid problems later we'll only keep rows that are not NaNs. We first add the two series to the dataframe and then remove NaNs:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0          True\n",
       "1         False\n",
       "2         False\n",
       "3          True\n",
       "4          True\n",
       "          ...  \n",
       "110811    False\n",
       "110812     True\n",
       "110813     True\n",
       "110814     True\n",
       "110815     True\n",
       "Name: Projects as responsible Applicant, Length: 110816, dtype: bool"
      ]
     },
     "execution_count": 18,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "pd.isna(projID_a)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {},
   "outputs": [],
   "source": [
    "applicants = persons.copy()\n",
    "applicants['projID'] = projID_a\n",
    "applicants = applicants[~pd.isna(projID_a)]\n",
    "\n",
    "employees = persons.copy()\n",
    "employees['projID'] = projID_e\n",
    "employees = employees[~pd.isna(projID_e)]"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Now we want each of these projects to become a single line in the dataframe. Here we use a function that we haven't used before called ```explode``` which turns every element in a list into a row (a good illustration of the variety of available functions in Pandas):"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {},
   "outputs": [],
   "source": [
    "applicants = applicants.explode('projID')\n",
    "employees = employees.explode('projID')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>Last Name</th>\n",
       "      <th>First Name</th>\n",
       "      <th>Gender</th>\n",
       "      <th>Institute Name</th>\n",
       "      <th>Institute Place</th>\n",
       "      <th>Person ID SNSF</th>\n",
       "      <th>OCRID</th>\n",
       "      <th>Projects as responsible Applicant</th>\n",
       "      <th>Projects as Applicant</th>\n",
       "      <th>Projects as Partner</th>\n",
       "      <th>Projects as Practice Partner</th>\n",
       "      <th>Projects as Employee</th>\n",
       "      <th>Projects as Contact Person</th>\n",
       "      <th>projID</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>a Marca</td>\n",
       "      <td>Andrea</td>\n",
       "      <td>male</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>132628</td>\n",
       "      <td>NaN</td>\n",
       "      <td>67368</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>67368</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>A. Jafari</td>\n",
       "      <td>Golnaz</td>\n",
       "      <td>female</td>\n",
       "      <td>Universität Luzern</td>\n",
       "      <td>Luzern</td>\n",
       "      <td>747886</td>\n",
       "      <td>NaN</td>\n",
       "      <td>191432</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>191432</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>Aapro</td>\n",
       "      <td>Matti S.</td>\n",
       "      <td>male</td>\n",
       "      <td>Clinique de Genolier F.M.H. Oncologie-Hématolo...</td>\n",
       "      <td>Genolier</td>\n",
       "      <td>3268</td>\n",
       "      <td>NaN</td>\n",
       "      <td>8532;9513</td>\n",
       "      <td>8155</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>8532</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>Aapro</td>\n",
       "      <td>Matti S.</td>\n",
       "      <td>male</td>\n",
       "      <td>Clinique de Genolier F.M.H. Oncologie-Hématolo...</td>\n",
       "      <td>Genolier</td>\n",
       "      <td>3268</td>\n",
       "      <td>NaN</td>\n",
       "      <td>8532;9513</td>\n",
       "      <td>8155</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>9513</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>11</th>\n",
       "      <td>Aas</td>\n",
       "      <td>Gregor</td>\n",
       "      <td>male</td>\n",
       "      <td>Lehrstuhl für Pflanzenphysiologie Universität ...</td>\n",
       "      <td>Bayreuth</td>\n",
       "      <td>36412</td>\n",
       "      <td>NaN</td>\n",
       "      <td>52037</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>52037</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "    Last Name First Name  Gender  \\\n",
       "1     a Marca     Andrea    male   \n",
       "2   A. Jafari     Golnaz  female   \n",
       "7       Aapro   Matti S.    male   \n",
       "7       Aapro   Matti S.    male   \n",
       "11        Aas     Gregor    male   \n",
       "\n",
       "                                       Institute Name Institute Place  \\\n",
       "1                                                 NaN             NaN   \n",
       "2                                  Universität Luzern          Luzern   \n",
       "7   Clinique de Genolier F.M.H. Oncologie-Hématolo...        Genolier   \n",
       "7   Clinique de Genolier F.M.H. Oncologie-Hématolo...        Genolier   \n",
       "11  Lehrstuhl für Pflanzenphysiologie Universität ...        Bayreuth   \n",
       "\n",
       "    Person ID SNSF OCRID Projects as responsible Applicant  \\\n",
       "1           132628   NaN                             67368   \n",
       "2           747886   NaN                            191432   \n",
       "7             3268   NaN                         8532;9513   \n",
       "7             3268   NaN                         8532;9513   \n",
       "11           36412   NaN                             52037   \n",
       "\n",
       "   Projects as Applicant Projects as Partner Projects as Practice Partner  \\\n",
       "1                    NaN                 NaN                          NaN   \n",
       "2                    NaN                 NaN                          NaN   \n",
       "7                   8155                 NaN                          NaN   \n",
       "7                   8155                 NaN                          NaN   \n",
       "11                   NaN                 NaN                          NaN   \n",
       "\n",
       "   Projects as Employee Projects as Contact Person  projID  \n",
       "1                   NaN                        NaN   67368  \n",
       "2                   NaN                        NaN  191432  \n",
       "7                   NaN                        NaN    8532  \n",
       "7                   NaN                        NaN    9513  \n",
       "11                  NaN                        NaN   52037  "
      ]
     },
     "execution_count": 21,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "applicants.head(5)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "So now we have one large table, where each row corresponds to a *single* applicant and a *single* project. We can finally do our merging operation where we combined information on persons and projects. We will do two such operations: one for applicants using the ```projID_a``` column for merging and one using the ```projID_e``` column. We have one last problem to fix:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "'67368'"
      ]
     },
     "execution_count": 22,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "applicants.loc[1].projID"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "4"
      ]
     },
     "execution_count": 23,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "projects.loc[1]['Project Number']"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "We need the project ID in the persons table to be a *number* and not a *string*. We can try to convert but get an error:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "metadata": {},
   "outputs": [
    {
     "ename": "ValueError",
     "evalue": "invalid literal for int() with base 10: ''",
     "output_type": "error",
     "traceback": [
      "\u001b[0;31m---------------------------------------------------------------------------\u001b[0m",
      "\u001b[0;31mValueError\u001b[0m                                Traceback (most recent call last)",
      "\u001b[0;32m<ipython-input-24-fca9460da04e>\u001b[0m in \u001b[0;36m<module>\u001b[0;34m\u001b[0m\n\u001b[0;32m----> 1\u001b[0;31m \u001b[0mapplicants\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mprojID\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mapplicants\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mprojID\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mastype\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mint\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m      2\u001b[0m \u001b[0memployees\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mprojID\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0memployees\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mprojID\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mastype\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mint\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
      "\u001b[0;32m~/miniconda3/envs/danalytics/lib/python3.8/site-packages/pandas/core/generic.py\u001b[0m in \u001b[0;36mastype\u001b[0;34m(self, dtype, copy, errors)\u001b[0m\n\u001b[1;32m   5696\u001b[0m         \u001b[0;32melse\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m   5697\u001b[0m             \u001b[0;31m# else, only a single dtype is given\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m-> 5698\u001b[0;31m             \u001b[0mnew_data\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0m_data\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mastype\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mdtype\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0mdtype\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mcopy\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0mcopy\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0merrors\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0merrors\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m   5699\u001b[0m             \u001b[0;32mreturn\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0m_constructor\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mnew_data\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0m__finalize__\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mself\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m   5700\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n",
      "\u001b[0;32m~/miniconda3/envs/danalytics/lib/python3.8/site-packages/pandas/core/internals/managers.py\u001b[0m in \u001b[0;36mastype\u001b[0;34m(self, dtype, copy, errors)\u001b[0m\n\u001b[1;32m    580\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    581\u001b[0m     \u001b[0;32mdef\u001b[0m \u001b[0mastype\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mself\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mdtype\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mcopy\u001b[0m\u001b[0;34m:\u001b[0m \u001b[0mbool\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0;32mFalse\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0merrors\u001b[0m\u001b[0;34m:\u001b[0m \u001b[0mstr\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0;34m\"raise\"\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 582\u001b[0;31m         \u001b[0;32mreturn\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mapply\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m\"astype\"\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mdtype\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0mdtype\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mcopy\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0mcopy\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0merrors\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0merrors\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m    583\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    584\u001b[0m     \u001b[0;32mdef\u001b[0m \u001b[0mconvert\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mself\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0;34m**\u001b[0m\u001b[0mkwargs\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
      "\u001b[0;32m~/miniconda3/envs/danalytics/lib/python3.8/site-packages/pandas/core/internals/managers.py\u001b[0m in \u001b[0;36mapply\u001b[0;34m(self, f, filter, **kwargs)\u001b[0m\n\u001b[1;32m    440\u001b[0m                 \u001b[0mapplied\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mb\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mapply\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mf\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0;34m**\u001b[0m\u001b[0mkwargs\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    441\u001b[0m             \u001b[0;32melse\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 442\u001b[0;31m                 \u001b[0mapplied\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mgetattr\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mb\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mf\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m**\u001b[0m\u001b[0mkwargs\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m    443\u001b[0m             \u001b[0mresult_blocks\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0m_extend_blocks\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mapplied\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mresult_blocks\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    444\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n",
      "\u001b[0;32m~/miniconda3/envs/danalytics/lib/python3.8/site-packages/pandas/core/internals/blocks.py\u001b[0m in \u001b[0;36mastype\u001b[0;34m(self, dtype, copy, errors)\u001b[0m\n\u001b[1;32m    623\u001b[0m             \u001b[0mvals1d\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mvalues\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mravel\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    624\u001b[0m             \u001b[0;32mtry\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 625\u001b[0;31m                 \u001b[0mvalues\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mastype_nansafe\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mvals1d\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mdtype\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mcopy\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0;32mTrue\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m    626\u001b[0m             \u001b[0;32mexcept\u001b[0m \u001b[0;34m(\u001b[0m\u001b[0mValueError\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mTypeError\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    627\u001b[0m                 \u001b[0;31m# e.g. astype_nansafe can fail on object-dtype of strings\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
      "\u001b[0;32m~/miniconda3/envs/danalytics/lib/python3.8/site-packages/pandas/core/dtypes/cast.py\u001b[0m in \u001b[0;36mastype_nansafe\u001b[0;34m(arr, dtype, copy, skipna)\u001b[0m\n\u001b[1;32m    872\u001b[0m         \u001b[0;31m# work around NumPy brokenness, #1987\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    873\u001b[0m         \u001b[0;32mif\u001b[0m \u001b[0mnp\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0missubdtype\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mdtype\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mtype\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mnp\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0minteger\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 874\u001b[0;31m             \u001b[0;32mreturn\u001b[0m \u001b[0mlib\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mastype_intsafe\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0marr\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mravel\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mdtype\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mreshape\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0marr\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mshape\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m    875\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    876\u001b[0m         \u001b[0;31m# if we have a datetime/timedelta array of objects\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
      "\u001b[0;32mpandas/_libs/lib.pyx\u001b[0m in \u001b[0;36mpandas._libs.lib.astype_intsafe\u001b[0;34m()\u001b[0m\n",
      "\u001b[0;31mValueError\u001b[0m: invalid literal for int() with base 10: ''"
     ]
    }
   ],
   "source": [
    "applicants.projID = applicants.projID.astype(int)\n",
    "employees.projID = employees.projID.astype(int)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "It looks like we have a row that doesn't conform to expectation and only contains ''. Let's try to figure out what happened. First we find the location with the issue:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "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>Last Name</th>\n",
       "      <th>First Name</th>\n",
       "      <th>Gender</th>\n",
       "      <th>Institute Name</th>\n",
       "      <th>Institute Place</th>\n",
       "      <th>Person ID SNSF</th>\n",
       "      <th>OCRID</th>\n",
       "      <th>Projects as responsible Applicant</th>\n",
       "      <th>Projects as Applicant</th>\n",
       "      <th>Projects as Partner</th>\n",
       "      <th>Projects as Practice Partner</th>\n",
       "      <th>Projects as Employee</th>\n",
       "      <th>Projects as Contact Person</th>\n",
       "      <th>projID</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>50947</th>\n",
       "      <td>Kleinewefers</td>\n",
       "      <td>Henner</td>\n",
       "      <td>male</td>\n",
       "      <td>Séminaire de politique économique, d'économie ...</td>\n",
       "      <td>Fribourg</td>\n",
       "      <td>10661</td>\n",
       "      <td>NaN</td>\n",
       "      <td>8;</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td></td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>62384</th>\n",
       "      <td>Massarenti</td>\n",
       "      <td>Léonard</td>\n",
       "      <td>male</td>\n",
       "      <td>Faculté de Psychologie et des Sciences de l'Ed...</td>\n",
       "      <td>Genève 4</td>\n",
       "      <td>11138</td>\n",
       "      <td>NaN</td>\n",
       "      <td>4;</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td></td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "          Last Name First Name Gender  \\\n",
       "50947  Kleinewefers     Henner   male   \n",
       "62384    Massarenti    Léonard   male   \n",
       "\n",
       "                                          Institute Name Institute Place  \\\n",
       "50947  Séminaire de politique économique, d'économie ...        Fribourg   \n",
       "62384  Faculté de Psychologie et des Sciences de l'Ed...        Genève 4   \n",
       "\n",
       "       Person ID SNSF OCRID Projects as responsible Applicant  \\\n",
       "50947           10661   NaN                                8;   \n",
       "62384           11138   NaN                                4;   \n",
       "\n",
       "      Projects as Applicant Projects as Partner Projects as Practice Partner  \\\n",
       "50947                   NaN                 NaN                          NaN   \n",
       "62384                   NaN                 NaN                          NaN   \n",
       "\n",
       "      Projects as Employee Projects as Contact Person projID  \n",
       "50947                  NaN                        NaN         \n",
       "62384                  NaN                        NaN         "
      ]
     },
     "execution_count": 25,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "applicants[applicants.projID=='']"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Then we look in the original table:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Last Name                                                                 Kleinewefers\n",
       "First Name                                                                      Henner\n",
       "Gender                                                                            male\n",
       "Institute Name                       Séminaire de politique économique, d'économie ...\n",
       "Institute Place                                                               Fribourg\n",
       "Person ID SNSF                                                                   10661\n",
       "OCRID                                                                              NaN\n",
       "Projects as responsible Applicant                                                   8;\n",
       "Projects as Applicant                                                              NaN\n",
       "Projects as Partner                                                                NaN\n",
       "Projects as Practice Partner                                                       NaN\n",
       "Projects as Employee                                                               NaN\n",
       "Projects as Contact Person                                                         NaN\n",
       "Name: 50947, dtype: object"
      ]
     },
     "execution_count": 26,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "persons.loc[50947]"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Unfortunately, as is often the case, we have a misformatting in the original table. The project as applicant entry has a single number but still contains the ```;``` sign. Therefore when we split the text, we end up with ```['8','']```. Can we fix this? We can for example filter the table and remove rows where ```projID``` has length 0:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "metadata": {},
   "outputs": [],
   "source": [
    "applicants = applicants[applicants.projID.apply(lambda x: len(x) > 0)]\n",
    "employees = employees[employees.projID.apply(lambda x: len(x) > 0)]"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Now we can convert the ```projID``` column to integer:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "metadata": {},
   "outputs": [],
   "source": [
    "applicants.projID = applicants.projID.astype(int)\n",
    "employees.projID = employees.projID.astype(int)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Finally we can use ```merge``` to combine both tables. We will combine the projects (on 'Project Number') and persons table (on 'projID_a' and 'projID_e'): "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "metadata": {},
   "outputs": [],
   "source": [
    "merged_appl = pd.merge(applicants, projects, left_on='projID', right_on='Project Number')\n",
    "merged_empl = pd.merge(employees, projects, left_on='projID', right_on='Project Number')\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 33,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>Last Name</th>\n",
       "      <th>First Name</th>\n",
       "      <th>Gender</th>\n",
       "      <th>Institute Name</th>\n",
       "      <th>Institute Place</th>\n",
       "      <th>Person ID SNSF</th>\n",
       "      <th>OCRID</th>\n",
       "      <th>Projects as responsible Applicant</th>\n",
       "      <th>Projects as Applicant</th>\n",
       "      <th>Projects as Partner</th>\n",
       "      <th>Projects as Practice Partner</th>\n",
       "      <th>Projects as Employee</th>\n",
       "      <th>Projects as Contact Person</th>\n",
       "      <th>projID</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>a Marca</td>\n",
       "      <td>Andrea</td>\n",
       "      <td>male</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>132628</td>\n",
       "      <td>NaN</td>\n",
       "      <td>67368</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>67368</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>A. Jafari</td>\n",
       "      <td>Golnaz</td>\n",
       "      <td>female</td>\n",
       "      <td>Universität Luzern</td>\n",
       "      <td>Luzern</td>\n",
       "      <td>747886</td>\n",
       "      <td>NaN</td>\n",
       "      <td>191432</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>191432</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>Aapro</td>\n",
       "      <td>Matti S.</td>\n",
       "      <td>male</td>\n",
       "      <td>Clinique de Genolier F.M.H. Oncologie-Hématolo...</td>\n",
       "      <td>Genolier</td>\n",
       "      <td>3268</td>\n",
       "      <td>NaN</td>\n",
       "      <td>8532;9513</td>\n",
       "      <td>8155</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>8532</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>Aapro</td>\n",
       "      <td>Matti S.</td>\n",
       "      <td>male</td>\n",
       "      <td>Clinique de Genolier F.M.H. Oncologie-Hématolo...</td>\n",
       "      <td>Genolier</td>\n",
       "      <td>3268</td>\n",
       "      <td>NaN</td>\n",
       "      <td>8532;9513</td>\n",
       "      <td>8155</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>9513</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>11</th>\n",
       "      <td>Aas</td>\n",
       "      <td>Gregor</td>\n",
       "      <td>male</td>\n",
       "      <td>Lehrstuhl für Pflanzenphysiologie Universität ...</td>\n",
       "      <td>Bayreuth</td>\n",
       "      <td>36412</td>\n",
       "      <td>NaN</td>\n",
       "      <td>52037</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>52037</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "    Last Name First Name  Gender  \\\n",
       "1     a Marca     Andrea    male   \n",
       "2   A. Jafari     Golnaz  female   \n",
       "7       Aapro   Matti S.    male   \n",
       "7       Aapro   Matti S.    male   \n",
       "11        Aas     Gregor    male   \n",
       "\n",
       "                                       Institute Name Institute Place  \\\n",
       "1                                                 NaN             NaN   \n",
       "2                                  Universität Luzern          Luzern   \n",
       "7   Clinique de Genolier F.M.H. Oncologie-Hématolo...        Genolier   \n",
       "7   Clinique de Genolier F.M.H. Oncologie-Hématolo...        Genolier   \n",
       "11  Lehrstuhl für Pflanzenphysiologie Universität ...        Bayreuth   \n",
       "\n",
       "    Person ID SNSF OCRID Projects as responsible Applicant  \\\n",
       "1           132628   NaN                             67368   \n",
       "2           747886   NaN                            191432   \n",
       "7             3268   NaN                         8532;9513   \n",
       "7             3268   NaN                         8532;9513   \n",
       "11           36412   NaN                             52037   \n",
       "\n",
       "   Projects as Applicant Projects as Partner Projects as Practice Partner  \\\n",
       "1                    NaN                 NaN                          NaN   \n",
       "2                    NaN                 NaN                          NaN   \n",
       "7                   8155                 NaN                          NaN   \n",
       "7                   8155                 NaN                          NaN   \n",
       "11                   NaN                 NaN                          NaN   \n",
       "\n",
       "   Projects as Employee Projects as Contact Person  projID  \n",
       "1                   NaN                        NaN   67368  \n",
       "2                   NaN                        NaN  191432  \n",
       "7                   NaN                        NaN    8532  \n",
       "7                   NaN                        NaN    9513  \n",
       "11                  NaN                        NaN   52037  "
      ]
     },
     "execution_count": 33,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "applicants.head(5)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## 12.3 Reformatting columns: time\n",
    "\n",
    "We now have in those tables information on both scientists and projects. Among other things we now when each project of each scientist has started via the ```Start Date``` column:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 34,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0         01.04.1993\n",
       "1         01.04.1993\n",
       "2         01.04.1993\n",
       "3         01.04.1993\n",
       "4         01.04.1993\n",
       "             ...    \n",
       "127126    01.04.1990\n",
       "127127    01.04.1991\n",
       "127128    01.11.1998\n",
       "127129    01.11.1992\n",
       "127130    01.10.2008\n",
       "Name: Start Date, Length: 127131, dtype: object"
      ]
     },
     "execution_count": 34,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "merged_empl['Start Date']"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "If we want to do computations with dates (e.g. measuring time spans) we have to change the type of the column. Currently it is indeed just a string. We could parse that string, but Pandas already offers tools to handle dates. For example we can use ```pd.to_datetime``` to transform the string into a Python ```datetime``` format. Let's create a new ```date``` column:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 35,
   "metadata": {},
   "outputs": [],
   "source": [
    "merged_empl['date'] = pd.to_datetime(merged_empl['Start Date'])\n",
    "merged_appl['date'] = pd.to_datetime(merged_appl['Start Date'])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 36,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Timestamp('1993-01-04 00:00:00')"
      ]
     },
     "execution_count": 36,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "merged_empl.iloc[0]['date']"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 37,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "1993"
      ]
     },
     "execution_count": 37,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "merged_empl.iloc[0]['date'].year"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Let's add a year column to our dataframe:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 38,
   "metadata": {},
   "outputs": [],
   "source": [
    "merged_empl['year'] = merged_empl.date.apply(lambda x: x.year)\n",
    "merged_appl['year'] = merged_appl.date.apply(lambda x: x.year)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## 12.4 Completing information\n",
    "\n",
    "As we did in the introduction, we want to be able to broadly classify projects into three categories. We therefore search for a specific string ('Humanities', 'Mathematics','Biology') within the 'Discipline Name Hierarchy' column to create a new column called 'Field'^:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 39,
   "metadata": {},
   "outputs": [],
   "source": [
    "science_types = ['Humanities', 'Mathematics','Biology']\n",
    "merged_appl['Field'] = merged_appl['Discipline Name Hierarchy'].apply(\n",
    "    lambda el: next((y for y in [x for x in science_types if x in el] if y is not None),None) if not pd.isna(el) else el)\n",
    "                                            "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "We will use the amounts awarded in our analysis. Let's look at that column:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 40,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0                       20120.00\n",
       "1        data not included in P3\n",
       "2                      211427.00\n",
       "3                      174021.00\n",
       "4                        8865.00\n",
       "                  ...           \n",
       "74650                  150524.00\n",
       "74651                  346000.00\n",
       "74652                  262960.00\n",
       "74653                  449517.00\n",
       "74654                 1433628.00\n",
       "Name: Approved Amount, Length: 74655, dtype: object"
      ]
     },
     "execution_count": 40,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "merged_appl['Approved Amount']"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Problem: we have rows that are not numerical. Let's coerce that column to numerical:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 41,
   "metadata": {},
   "outputs": [],
   "source": [
    "merged_appl['Approved Amount'] = pd.to_numeric(merged_appl['Approved Amount'], errors='coerce')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 42,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0          20120.0\n",
       "1              NaN\n",
       "2         211427.0\n",
       "3         174021.0\n",
       "4           8865.0\n",
       "           ...    \n",
       "74650     150524.0\n",
       "74651     346000.0\n",
       "74652     262960.0\n",
       "74653     449517.0\n",
       "74654    1433628.0\n",
       "Name: Approved Amount, Length: 74655, dtype: float64"
      ]
     },
     "execution_count": 42,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "merged_appl['Approved Amount']"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## 12.5 Data anaylsis\n",
    "\n",
    "We are finally done tidying up our tables so that we can do proper data analysis. We can *aggregate* data to answer some questions."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### 12.5.1 Amounts by gender\n",
    " Let's see for example what is the average amount awarded every year, split by gender. We keep only the 'Project funding' category to avoid obscuring the results with large funds awarded for specific projects (PNR etc):"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 44,
   "metadata": {},
   "outputs": [],
   "source": [
    "merged_projects = merged_appl[merged_appl['Funding Instrument Hierarchy'] == 'Project funding']"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 45,
   "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>Gender</th>\n",
       "      <th>year</th>\n",
       "      <th>Approved Amount</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>female</td>\n",
       "      <td>1975.0</td>\n",
       "      <td>101433.200000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>female</td>\n",
       "      <td>1976.0</td>\n",
       "      <td>145017.750000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>female</td>\n",
       "      <td>1977.0</td>\n",
       "      <td>177826.157895</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>female</td>\n",
       "      <td>1978.0</td>\n",
       "      <td>141489.857143</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>female</td>\n",
       "      <td>1979.0</td>\n",
       "      <td>218496.904762</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>87</th>\n",
       "      <td>male</td>\n",
       "      <td>2016.0</td>\n",
       "      <td>429717.055907</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>88</th>\n",
       "      <td>male</td>\n",
       "      <td>2017.0</td>\n",
       "      <td>507521.397098</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>89</th>\n",
       "      <td>male</td>\n",
       "      <td>2018.0</td>\n",
       "      <td>582461.020513</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>90</th>\n",
       "      <td>male</td>\n",
       "      <td>2019.0</td>\n",
       "      <td>624826.387985</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>91</th>\n",
       "      <td>male</td>\n",
       "      <td>2020.0</td>\n",
       "      <td>617256.523404</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>92 rows × 3 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "    Gender    year  Approved Amount\n",
       "0   female  1975.0    101433.200000\n",
       "1   female  1976.0    145017.750000\n",
       "2   female  1977.0    177826.157895\n",
       "3   female  1978.0    141489.857143\n",
       "4   female  1979.0    218496.904762\n",
       "..     ...     ...              ...\n",
       "87    male  2016.0    429717.055907\n",
       "88    male  2017.0    507521.397098\n",
       "89    male  2018.0    582461.020513\n",
       "90    male  2019.0    624826.387985\n",
       "91    male  2020.0    617256.523404\n",
       "\n",
       "[92 rows x 3 columns]"
      ]
     },
     "execution_count": 45,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "grouped_gender = merged_projects.groupby(['Gender','year'])['Approved Amount'].mean().reset_index()\n",
    "grouped_gender"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "To generate a plot, we use here Seaborn which uses some elements of a grammar of graphics. For example we can assign variables to each \"aspect\" of our plot. Here x and y axis are year and amount while color ('hue') is the gender. In one line, we can generate a plot that compiles all the information:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 46,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.axes._subplots.AxesSubplot at 0x122c5d0d0>"
      ]
     },
     "execution_count": 46,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAZgAAAEGCAYAAABYV4NmAAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADh0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uMy4yLjEsIGh0dHA6Ly9tYXRwbG90bGliLm9yZy+j8jraAAAgAElEQVR4nOzdd3gVVfrA8e+bTgmB0EkgBAiE3kJoohQpFhQVBCuuKPa2q66u+lPXxbK66FpAUVCwoBRRdAWlBURKSOhIh1RqCCmE9JzfHzOBQNol5KbA+3me+9zcM3NmzgS9b04XYwxKKaVUeXOp7AIopZS6NGmAUUop5RQaYJRSSjmFBhillFJOoQFGKaWUU7hVdgGqigYNGpiWLVtWdjGUUqpaiYyMTDDGNCzqmAYYW8uWLYmIiKjsYiilVLUiItHFHdMmMqWUUk6hAUYppZRTaIBRSinlFNoHU4Ls7Gzi4uLIyMio7KJUSV5eXvj7++Pu7l7ZRVFKVUEaYEoQFxeHt7c3LVu2REQquzhVijGGEydOEBcXR2BgYGUXRylVBWkTWQkyMjKoX7++BpciiAj169fX2p1SqlgaYEqhwaV4+rtRSpVEA4xSSl3qkuNhw2eQllCht9UAUw0cPXqU22+/nVatWtGzZ0/69u3LggULLvq6YWFhXH/99eVQQqVUlfbbi/C/v8Hk9rDgQYiPrJDbaoCp4owxjBo1iiuvvJIDBw4QGRnJt99+S1xcXIWXJScnp8LvqZS6SCmHYedC6Hwr9BgPO3+CTwfDp0Ngy3eQk+m0W2uAqeKWL1+Oh4cHDz744Jm0gIAAHnvsMXJzc3nmmWfo1asXXbp04ZNPPgGsmsnAgQMZPXo0wcHB3HHHHeTvXLp48WKCg4O54oor+P77789cMy0tjXvvvZdevXrRvXt3fvzxRwC++OILxowZw8iRIxk2bFgFPrlSqlxEfgF5uTDwObjuHfjrTrjmbchIhgUTYXIHWPE6OGF3Yx2mXMXt2LGDHj16FHls+vTp+Pj4sGHDBjIzM+nfv/+ZILBp0yZ27NhBs2bN6N+/P3/88QchISHcf//9LF++nDZt2jB27Ngz15o0aRKDBw9mxowZJCUlERoaytVXXw3A2rVr2bp1K76+vs5/YKVU+cnJgsjPIWgo1G9tpXnVgd4Todd9cDAMwj+F47vACYN2NMBUM4888girV6/Gw8ODgIAAtm7dyrx58wBITk5m7969eHh4EBoair+/PwDdunUjKiqK2rVrExgYSFBQEAB33nkn06ZNA+C3335j4cKFvPPOO4A1RDsmJgaAoUOHanBRqjrauRBOHYXQBwofc3GB1oOtV16uU26vAaaK69ixI/Pnzz/z+aOPPiIhIYGQkBBatGjBBx98wPDhw8/JExYWhqen55nPrq6uZ/pPihtabIxh/vz5tGvX7pz09evXU6tWrfJ6HKVURQqfBr6trCBSEhdXp9zeqX0wIlJXROaJyC4R2SkifUXEV0SWiMhe+71egfOfF5F9IrJbRIYXSO8pItvsY++L/S0pIp4i8p2dvl5EWhbIM96+x14RGe/M53SmwYMHk5GRwdSpU8+knT59GoDhw4czdepUsrOzAdizZw9paWnFXis4OJiDBw+yf/9+AGbPnn3m2PDhw/nggw/O9NVs2rSp3J9FKVWBDm2G2PXQ636rtlIJnH3X/wKLjTHBQFdgJ/AcsMwYEwQssz8jIh2AcUBHYAQwRUTyw+pUYCIQZL9G2OkTgJPGmDbAu8Bb9rV8gZeB3kAo8HLBQFadiAg//PADK1euJDAwkNDQUMaPH89bb73FfffdR4cOHejRowedOnXigQceKHGkl5eXF9OmTeO6667jiiuuICAg4Myxl156iezsbLp06UKnTp146aWXKuLxlFLOEv4puNeEbrdXWhHEOGHkAICI1AG2AK1MgZuIyG5goDHmsIg0BcKMMe1E5HkAY8wb9nm/Aq8AUcAKO0ghIrfZ+R/IP8cYs1ZE3IAjQEOsQDXQGPOAnecT+z5n/2Q/T0hIiDl/w7GdO3fSvn37cvhtXLr0d6RUFXQ6Ef4TDN3vgOvfdeqtRCTSGBNS1DFn1mBaAceBz0Vkk4h8JiK1gMbGmMMA9nsj+3w/ILZA/jg7zc/++fz0c/IYY3KAZKB+CddSSqlL38ZZkJtpNY9VImcGGDegBzDVGNMdSMNuDitGUb3PpoT0suY5e0ORiSISISIRx48fL6FoSilVTeTlwobp0HIANO5QqUVxZoCJA+KMMevtz/OwAs5Ru2kM+/1YgfObF8jvDxyy0/2LSD8nj91E5gMklnCtcxhjphljQowxIQ0bNizjYyqlVBWyZzEkx0Bo5dZewIkBxhhzBIgVkfxxr0OAP4GFQP6orvHAj/bPC4Fx9siwQKzO/HC7GS1VRPrYo8fuPi9P/rVGA8vt/p5fgWEiUs/u3B9mpyml1KUtfBrU8YN211V2SZw+D+Yx4GsR8QAOAH/BCmpzRGQCEAOMATDG7BCROVhBKAd4xBiTP/vnIeALoAawyH4BTAe+FJF9WDWXcfa1EkXkNWCDfd4/jTGJznxQpZSqdMd3w4EwGPwSuFb+NEenlsAYsxkoanTBkGLOnwRMKiI9AuhURHoGdoAq4tgMYMaFlFcppaq1DZ+Bq4e1qGUBxhjSs3Op6VGxQUcXu6wG3n//fdq3b88dd9zhlOu/8sorZ5aIUUpVUxkpsPkb6Hgz1D7bp5yakc39syLo9s8lvPPrbk5nVdyq6JVfh1KlmjJlCosWLSIwMLCyi6KUqqr2L4OsU9DzbO3lYEIa98+KICohjb6t6/Phin3Mi4zj+WuDuaFrM6fvSqs1mCruwQcf5MCBA9xwww1MmjSp2CX1R40axciRIwkMDOTDDz9k8uTJdO/enT59+pCYaHU/ffrpp/Tq1YuuXbtyyy23nFlypqD9+/czYsQIevbsyYABA9i1a1eFPq9Sqoyi11oz9/17AbBqz3Fu/HA1J05l8uWE3nw5oTfzH+pLQ29Pnvh2M2M+Xsv2+GSnFklrMA569acd/HkopVyv2aFZHV4e2bHEcz7++GMWL17MihUrmDx5crFL6m/fvp1NmzaRkZFBmzZteOutt9i0aRNPPfUUs2bN4sknn+Tmm2/m/vutoYsvvvgi06dP57HHHjvnfhMnTuTjjz8mKCiI9evX8/DDD7N8+fJyfW6llBNErwH/XhgXN6b/foDXf9lJ28befHp3CM19awLQM8CXHx7pz7zIWP69eDcjP1zN2JDmPD28HQ1qe5ZygwunAaYaKWlJ/UGDBuHt7Y23tzc+Pj6MHDkSgM6dO7N161bACkIvvvgiSUlJnDp1qtAqzKdOnWLNmjWMGXN23ERmpvN2u1NKlZP0JDi6newBz/Lc3K3M3xjHiI5N+M+tXanlee7XvKuLMLZXC67p3JT3l+7lizVRRESfZMlTV5Z7k5kGGAeVVtOoCCUtqV9weX4XF5czn11cXM4sgHnPPffwww8/0LVrV7744gvCwsLOuU5eXh5169Zl8+bNzn0QpVT5ig0HDK9t82H+kTievDqIxwcH4eJSfMCo4+XOi9d3YFxoC46mZDilP0b7YKqRi11SPzU1laZNm5Kdnc3XX39d6HidOnUIDAxk7ty5gBXQtmzZcvEFV0o5V8wa8sSNOUea8vboLjx5ddsSg0tBbRrVpn+bBk4plgaYauRil9R/7bXX6N27N0OHDiU4OLjIc77++mumT59O165d6dix45mBBEqpKix6LUdrBZMpnlzfpVlll+YMpy3XX93ocv1lo78jpSpZdga82Zzfao/ijdw7WPH0wAq9fWUt16+UUsrZ4iMhN4uwzCCCm3hXdmnOoQFGKaWqs+g1APySHEA7DTBKKaXKTcwa0uu1I8nU1hqMUkqpcpKbA7HhxHt3AyC4SZ1KLtC5NMAopVR1dXQbZJ1ii2t7ari70sKesV9VaIBRSqnqKnotACvS29C2cW2H575UFA0wl7CwsDCuv/76yi6GUspZYtZg6rZgzXGvKtc8BhpglFKqejIGoteS0awPiWlZVW4EGWiAqfKioqIIDg7mvvvuo1OnTtxxxx0sXbqU/v37ExQURHh4OOHh4fTr14/u3bvTr18/du/eXeg6aWlpRS71r5SqphL2wukEYr27AhDctOoFGF3s0lGLnoMj28r3mk06wzVvlnravn37mDt3LtOmTaNXr1588803rF69moULF/L6668za9YsVq1ahZubG0uXLuUf//gH8+fPP+cakyZNKnKp/1q1apXvMymlKkaMNf9lMx2A9CrZRKYBphoIDAykc+fOAHTs2JEhQ4YgInTu3JmoqCiSk5MZP348e/fuRUTIzs4udI3ilvrXZV6Uqqai10KthqxP8aWhdwK+tTwqu0SFaIBxlAM1DWcpbSn+l156iUGDBrFgwQKioqIYOHBgoWsUt9S/UqqailkDLfqw62hqlZtgmU/7YC4BycnJ+Pn5Adb2yUW52KX+lVJVSHI8JMWQ27wve4+d0gCjnOfZZ5/l+eefp3///uTm5hZ5zsUu9a+UqkJirPkvh3x6kJWTVyX7X0CX6z9Dl+svG/0dKVUJfn4Kts7l5+vW8ujsrfz82BV08vOplKLocv1KKXUpiV4LzUPZffQ0ri5Cm0a1K7tERdIAo5RS1cnpRDi+EwL6sutIKoENauHl7lrZpSqSBphSaBNi8fR3o1QliFlnvbfox64jKVVyBn8+DTAl8PLy4sSJE/pFWgRjDCdOnMDLy6uyi6LU5SVmDbh6cKphV2IT0wluXHUDjM6DKYG/vz9xcXEcP368sotSJXl5eeHv71/ZxVDq8hK9Fvx6sjvBmlAd3LRqjiADJwcYEYkCUoFcIMcYEyIivsB3QEsgCrjVGHPSPv95YIJ9/uPGmF/t9J7AF0AN4BfgCWOMERFPYBbQEzgBjDXGRNl5xgMv2kX5lzFm5oWW393dncDAwAt+bqWUcoqs03B4M/R7jN1HUgGq7BwYqJgmskHGmG4FhrE9BywzxgQBy+zPiEgHYBzQERgBTBGR/J6rqcBEIMh+jbDTJwAnjTFtgHeBt+xr+QIvA72BUOBlEann1KdUSilni4+EvBxo0ZfdR1Ko7emGX90alV2qYlVGH8yNQH5tYiYwqkD6t8aYTGPMQWAfECoiTYE6xpi1xuoMmXVenvxrzQOGiIgAw4ElxphEu3a0hLNBSSmlqqdYu4Pfvxc7j6RWyU3GCnJ2gDHAbyISKSIT7bTGxpjDAPZ7IzvdD4gtkDfOTvOzfz4//Zw8xpgcIBmoX8K1ziEiE0UkQkQitJ9FKVXlxayHhsGYGvXYfSSVdlV0Bn8+Z3fy9zfGHBKRRsASEdlVwrlFhWFTQnpZ85xNMGYaMA2smfwllE0ppSpXXh7EhUOHURxNySQ5PZv2VXAPmIKcWoMxxhyy348BC7D6Q47azV7Y78fs0+OA5gWy+wOH7HT/ItLPySMiboAPkFjCtZRSqno6vgsykqFFH3YeSQGgXRUeogxODDAiUktEvPN/BoYB24GFwHj7tPFA/taKC4FxIuIpIoFYnfnhdjNaqoj0sftX7j4vT/61RgPL7X6aX4FhIlLP7twfZqcppVT1lN//0rx3gRFkl28TWWNggRUTcAO+McYsFpENwBwRmQDEAGMAjDE7RGQO8CeQAzxijMlfGvghzg5TXmS/AKYDX4rIPqyayzj7Woki8hqwwT7vn8aYRCc+q1JKOVfMeqjVEHxbsevwZpr6eOFT072yS1UipwUYY8wBoGsR6SeAIcXkmQRMKiI9AuhURHoGdoAq4tgMYMaFlVoppaqo2HXQvDeIsOtIapVeIiafLhWjlFJVXepROBkFLfqQnZvH/uOnqnzzGGiAUUqpqu9M/0sfDhxPIzvXVOkZ/Pk0wCilVFUXsx5cPaFpV3bljyC7FAKMiLzlSJpSSikniV0Hfj3AzYPdR1JxcxFaN6yam4wV5EgNZmgRadeUd0GUUkoVIes0HN5idfADu46k0rphbTzcqn4DVLGjyETkIeBhoJWIbC1wyBv4w9kFU0opBRzaaC9w2QeA3UdSCWlZPdbuLWmY8jdY803ewF7x2Jaqc0qUUqqCxJydYHniVCbxSenc1TSgcsvkoGLrWMaYZGNMlDHmNqylV7Kx1vOqLSItKqqASil1WYtdDw3aQk1fNkRZf9v3aulbyYVyTKkTLUXkUeAV4CiQZycboIvziqWUUoq8PIgNh/YjAVh/MBEvdxc6+/lUcsEc48hM/ieBdvYMfKWUUhUlYQ9kJJ3pfwk/mEj35vWqRQc/ODaKLBZrnxWllFIVqcAEy5SMbHYeTiE0sHo0j4FjNZgDQJiI/A/IzE80xkx2WqmUUkpZEyxrNoD6rYncc5w8A70vsQATY7887JdSSqmKUGCBy/CDibi5CN1bVI8hyuBAgDHGvFoRBVFKKVXAqWOQeAB63gNY/S9d/H2o4eFaueW6AI6MIltB0dsND3ZKiZRSSlnDkwGa9yE9K5etcUnce0Vg5ZbpAjnSRPZ0gZ+9gFuwNgRTSinlLDHrrAUum3VjU/RJsnNNtep/AceayCLPS/pDRFY6qTxKKaXAqsE06w5unmw4GIMI9Ay4xAKMiBR8IhegJ9DEaSVSSqnLXXY6HNoMfR8GIDzqBO2b1MGnRtXeIvl8jjSRRWL1wQhW09hBYIIzC6WUUpe1Q5sgLxua9yErJ4/I6JOM61X9VuhypImsevUqKaVUdVdggcvth5LJyM6rdv0v4FgTmTvwEHClnRQGfGKMyXZiuZRS6vIVvQbqB0Gt+oRH7AcgpJoscFmQI0vFTMXqd5liv3raaUoppcrb0T9h39IzC1yGH0ykVcNaNPT2rOSCXThH+mB6GWO6Fvi8XES2OKtASil1WQt7AzxqQ7/HyM0zbIhK5PouTSu7VGXiSA0mV0Ra538QkVZArvOKpJRSl6nDW2HnQujzENT0ZfeRVFIzcqrVApcFOVKDeQZYISIHsEaSBQB/cWqplFLqchT2Bnj5QN9HAAg/aO2SEhpYvzJLVWaOjCJbJiJBQDusALPLGJNZSjallFIXIj4Sdv8Cg16EGnUBCI9KxK9uDfzq1qjkwpWNI6PIXIHhQEv7/CEiosv1K6VUeVrxOtSoB30eBMAYQ/jBRAYENazkgpWdI01kPwEZwDbObpmslFKqvMSst0aOXf0KeHoDcCAhjYRTWdW2/wUcCzD+xpguTi+JUkpdrlZMgloNIXTimaQNBxMBqnWAcWQU2SIRGVbWG4iIq4hsEpGf7c++IrJERPba7/UKnPu8iOwTkd0iMrxAek8R2WYfe19ExE73FJHv7PT1ItKyQJ7x9j32isj4spZfKaWcKmo1HFwJVzwFHrXOJIcfTKRBbQ9aNahVQuaqzZEAsw5YICLpIpIiIqkiknIB93gC2Fng83PAMmNMELDM/oyIdADGAR2BEcAUu/8HrImdE4Eg+zXCTp8AnDTGtAHeBd6yr+ULvAz0BkKBlwsGMqWUqhKMgeWToHYTCLn3nEPrDyYSGuiL/fd0teRIgPkP0BeoaYypY4zxNsbUceTiIuIPXAd8ViD5RmCm/fNMYFSB9G+NMZnGmIPAPiBURJoCdYwxa40xBph1Xp78a83DGoAgWIMSlhhjEo0xJ4ElnA1KSilVNRxYATFrYMDfwP3sSLG4k6eJT0qnVzVcHqYgRwLMXmC7/eV+od4DnuXcwQGNjTGHAez3Rna6HxBb4Lw4O83P/vn89HPyGGNygGSgfgnXOoeITBSRCBGJOH78eBkeTymlysgYa+RYHX/oeW4r/oao6t//Ao518h8GwkRkEXBm/ktpw5RF5HrgmDEmUkQGOnCfouqBpoT0suY5m2DMNGAaQEhISFkCqFJKlc3eJRC3Aa5/D9zOrjN2LCWDRduO4O3lRnAThxqLqixHAsxB++Vhv6CIL+si9AduEJFrsbZariMiXwFHRaSpMeaw3fx1zD4/DmheIL8/cMhO9y8ivWCeOBFxA3yARDt94Hl5whwos1JKOV9eLix9BeoGEB94C+s3xrH+QCLhUYkcTEgDYGTXZri6VN/+F3BsJv+rBT+LiBcw0oF8zwPP23kGAk8bY+4UkbeB8cCb9vuPdpaFwDciMhlohtWZH26MybUHFvQB1gN3Ax8UyDMeWAuMBpYbY4yI/Aq8XqBjf1h+WZRSqtJtmQ3HdvB/7n9j1tu/A1DHy43QQF9uC21O78D6dGxWvWsv4FgNJn82/zDgNvt9NTC3jPd8E5gjIhOAGGAMgDFmh4jMAf7E2jnzEWNM/qKaDwFfADWARfYLYDrwpYjsw6q5jLOvlSgirwEb7PP+aYxJLGN5lVKq/GSdxiz/FztdgljpfgUvjwwkNNCX4CZ1qn2N5XxSUt+9iFwJ3I41Eiwcq9mrlTHmdMUUr+KEhISYiIiIyi6GUupSt+ptWP4vxmT+H3+5/Xau7Vw9l+LPJyKRxpiQoo4VW4MRkTisGsZU4BljTKqIHLwUg4tSSlWIU8cxq//LOvc+JNTuyfCOTSq7RE5V0jDl+VhDe8cCI0WkFo517iullCrKyjch+zQvnBrNA1e2uuSaxM5XbIAxxjyBtYLyZGAQsAdoKCK3ikjtiimeUkpdIhL2QsTnLK15LadqB3JTj0JT8y45JU60NJblxpj7sYLN7Viz6KOcXzSllLqELH2FXLcaPHfiWu4bEIinm2vpeao5h0aRARhjsrGW7v9JRKrn7jdKKVUZotfArp/52fdesrPrc1toi8ouUYVwZKmYQowx6eVdEKWUuiQZA7+9RE6tJjx3eAB3922Jt5d7ZZeqQpQpwCillHLQjgUQH8GCuveQ51qDe/q3rOwSVRgNMEop5SxZabDsVbIbtOfFqM6M7dWcBrU9S893iShpHsxPlDAs2Rhzg1NKpJRS1d3JaNjwKWz8EjKSmNv2PXIOuXD/gFaVXbIKVVIn/zv2+81AE+Ar+/Nt6CgypZQ6lzFwIAzCp8HuRSAu0P56Urvdx6SvTjOyS2Oa+9as7FJWqGIDjDFmJYCIvGaMubLAoZ9EZJXTS6aUUtXF5m9g9XuQsBtq1ocBf7V2qPTx54tle0nL2sODA1tXdikrnCPDlBuKSCtjzAEAEQkEGjq3WEopVU0cXAU/PARNusCoqdDxZnD3AiA9K5fP10QxqF3Dar+3S1k4EmCewtpw7ID9uSXwgNNKpJRS1YUxsPRVqOMHE5acCSz5ZofHkJiWxUMD21RSASuXI/vBLBaRICDYTtpljMksKY9SSl0Wdi+C+AgY+X6h4JKRncvUlfvp08q32m99XFalDlMWkZrAM8CjxpgtQAt7O2SllLp85eXC8tfAtzV0u6PQ4dnhMRxPzeSJIW0roXBVgyPzYD4HsoC+9uc44F9OK5FSSlUH2+bBsT9h8Avgem5jUEZ2LlPD9tM70Je+retXUgErnyMBprUx5t9ANpxZJubSXmNaKaVKkpMFKyZBk87Q4aZCh78Nj+FYaiZPXB1UCYWrOhzp5M+yF7c0ACLSGtA+GKXU5WvTLEiKhjvmgcu5f6fn972EBvrSt9XlW3sBx2owrwCLgeYi8jWwDHjWmYVSSqkqK+s0rHwbWvSFNlcXOvzdhliOpmTy5NVBiFzejT2OjCL7TUQigT5YTWNPGGMSnF4ypZSqisI/gVNHYMwXcF4AycjOZUrYPkJbau0FHAgwIrIQmA0sNMakOb9ISilVRaUnWTP2g4ZBQN9Ch+dEWLWXd2/tdtnXXsCxJrL/AAOAP0VkroiMFhGv0jIppdQlZ80HkJEEg18qdCgjO5cpK/bTq2W9y3rkWEGlBhhjzEpjzMNAK2AacCtwzNkFU0qpi3J8D5zYb434Kg+njsE6eymYpl0KHZ4TEcuRlAyevLqt1l5sDm2ZbI8iGwmMBXoAM51ZKKWUuiiHt8InV2INfhWo0wzqtrBfAdCkE7S/oVAfSolWvQ05GTD4xUKHMnOs2ktIQD36ae3lDEf6YL4DemONJPsICDPG5Dm7YEopVWZrPwKPWjDiDUiOh6QYa1hx9BrYNhdMHlz7DoTe79j1otdA+KfWCsn1C6+KPGeDVXt5Z0xXrb0U4EgN5nPgdmNMrrMLo5RSFy3lEGyfB73uhx53Fz6emw2zb4NfX4CA/tC4Q8nXy0iG7x+Aei1h6KuFDh9LzeCjFfvpGVCP/m209lKQI538y4BHRGSe/XpMRNydXTCllCqT8GlWDaV3MYu+u7rDqCngVQfmT4Ds9JKv98uzkBIPN38Knt7nHDpw/BS3TF1Dcno2L1zXXmsv53EkwEwFegJT7FcPO00ppaqWzFMQMQOCrwffwOLPq90IRn1srSW25P+KP2/7fNj6LVz5DDTvdc6hTTEnGf3xWk5n5vLtxD70aFGvnB7i0uFIE1kvY0zXAp+Xi8gWZxVIKaXKbMtsq0mr76Olnxt0NfR5GNZNgdZDoN0I8vIMWbl5eLm7QnIc/PwU+IVYAaaA5buO8sjXm2jo7cmse0Np2aCWkx6oenOkBpNrrz8GgIi0AkrtjxERLxEJF5EtIrJDRF61031FZImI7LXf6xXI87yI7BOR3SIyvEB6TxHZZh97X+x6qIh4ish3dvp6EWlZIM94+x57RWS8I78MpVQ1lpdrde7794IWvR3Lc/Ur0Lgz/PgwpB7hHwu20eXV33ho1gZOfHUvJjcHbp52zmrJ322I4f5ZkbRpVJv5D/XT4FICRwLM08AKEQkTkZXAcuBvDuTLBAbbtZ9uwAgR6QM8BywzxgRh9e88ByAiHYBxQEdgBDBFRFzta00FJgJB9muEnT4BOGmMaQO8C7xlX8sXeBlr9Fso8HLBQKaUugTtXgQnD0LfRxzP4+YJo6dD1mmSZ9/Hdxui6ervQ7uoWdQ/vp7/y76LZ1ec4o99CeTmGd5ftpe/z99G/zYN+HZiHxp6ezrveS4BJTaR2V/wXbG+1NthrUXm0I6WxhgDnLI/utsvA9wIDLTTZwJhwN/t9G/tax8UkX1AqIhEAXWMMWvtMs0CRgGL7Dyv2NeaB3xo126GA0uMMYl2niVYQWl2aeVWSlVTaz8CnxYQPPLC8jVsR97wN/D535M8Was1E6+7B6/PZ8qpeZsAACAASURBVJPQfBina9/Gr9uOMCciDm8vN1Izcri5hx9v3dIFd1dH/j6/vJUYYIwxuSJygzHmXWDrhV7cDlCRQBvgI2PMehFpbIw5bF//sIg0sk/3A9YVyB5np2XbP5+fnp8n1r5WjogkA/ULpheRp2D5JmLVjGjRosWFPp5SqqqIj4SYNTD89UKbfzni65xBNMztxWPyNS7fh0HN+jQY9zH/qVWfSdm5LNt5jEXbD9OxmQ8PXtVKR4s5yJF/iTUi8iHwHXBmsUtjzMbSMtpzZ7qJSF1ggYh0KuH0ov7FTAnpZc1TsHzTsJa/ISQkpNBxpVQ1sfYj8KwD3e+64KyJaVm889seevs9y/C0J+BkFNz5PdSy5rR4ubtyXZemXNelaTkX+tLnSIDpZ7//s0CaAQY7ehNjTJKIhGE1Ux0VkaZ27aUpZ9c1iwOaF8jmDxyy0/2LSC+YJ05E3AAfINFOH3henjBHy6uUqkaSYmHHD9DnIWtuywV6+9ddpGXm8MxNAxBZACf2QpshTijo5ceRxS4HFfEqNbiISEO75pK/ltnVwC5gIZA/qms88KP980JgnD0yLBCr3yfcbk5LFZE+dv/K3eflyb/WaGC53ffzKzBMROrZnfvD7DSl1KUm/BPrvbiJlSXYEpvEtxtiuadfS4Iae0OjYGh/gX04qliOrEVWH2tE1hVYNZfVwD+NMSdKydoUmGn3w7gAc4wxP4vIWmCOiEwAYoAxAMaYHSIyB/gTyAEeKbA8zUPAF0ANrM79RXb6dOBLe0BAItYoNIwxiSLyGrDBPu+f+R3+SqlLSEYKRM6EDjdaC1legLw8w/8t3EGD2p48cXWQkwp4eXOkiexbYBVwi/35Dqz+mMJ7hRZgjNkKdC8i/QRQZP3TGDMJmFREegRQqP/GGJOBHaCKODYDmFFSGZVS1dvJP2ZQLzPFsYmV55kXGceW2CTeHdsVby9d/coZHBln52uMec0Yc9B+/Quo6+yCKaVUSdKTjpH9+3tE5LXjoTBYd+AEVgt56ZJPZ/Pm4l30almPUd0KDTBV5cSRGswKERkHzLE/jwb+57wiKaUqXfpJOLQZsk5B2xHWApFVSV4uSV/fg69JZUWrN1mz/wSLth8huIk34/u1ZFQ3P2p4uBabffKS3SSdzuKVG0J1yLETSWkRX0RSgVpA/h4wLpwdrmyMMRc+bKMKCgkJMREREZVdDKUqXkYKHN4ChzadfZ08ePZ43RYw4GnodnvVCTRhb0LYG7zr9TBP/v11MrLzWLglni/WRLPzcAo+NdwZ3dOfht6enM7M4XRWLmlZuZzOyiEtM5flu45yZ58A/nljSTMnlCNEJNIYE1LkMUerlJc6DTCqysjLhYQ90Ki9c++THAcr/w2bv4a8HCvNpwU06wbNuluvnAxrJ8f4yLOBputt4Obh3LKVZO9SzNej+T73CtKu+YC7+51dNdkYw4aok8xcE8XiHUfIzTOIQE13V2p6ulHLw5UaHm4E+NbkrVu64FOzigTMauyiA4yI3MzZUWS/G2N+KN8iVj4NMKpKyM6ABRPhzx/h1lnW6Kjyduo4rJ4MG6Zb+6b0HA9tr7ECS60Ghc83BvYthbA3Kj/QJMXCJwM4lFeP69NfYeU/ri22g/50Vg6C4OXuos1gTlRSgHFkmPIUrKVe8tfxelBEhhpjLmBFOaVUqdJPwrd3QPQfULM+LJ9k7WviUnxfwoVdPwnWfghrp0BOOnS9Ha56FuoFlJxPBIKGQpurzwaanx639l255+dCm3CV6GS0FaDK8oWfkwlzx2Nyc7jn9GNc07NViaO/anpc+JIxqnw58i9wFdDJnsCIiMwEtjm1VEpdbpLj4atbIHE/jJ4B4gpzx8O2edB17MVdOy8P1n0Eq96BjCToeBMM/Ac0bHth1ykYaHZ8D/Pvhznj4fbvHOubWf0eLH0ZQibAdf+58CDz6z8gPpJlnd9hz4bG/LdPKYFRVTpHAsxuoAUQbX9uThkWvlRKFePYTiu4ZKbCnfMh8EorKDTpbNUWOt1c9s719JPw/UTY+xu0GQpDXoKmXUvPVxIR6HQLZKXBwsfgpyfhxg9LDhhrPrSCi29riJhu1dAGv+D4PbfOhQ2fYfo+xr+2tqFXS0/aN70kxhdd0hwJMPWBnSISbn/uBawTkYUAxpgbnFU4pS5IXh4c3Qb7llmvjCRoHgot+kKLPhc807tCRK+B2ePArQb85RcrqAC4uMCgF2H2WNj8jdVPcqGO7rCa3JLj4LrJEHJv2ZqmitPjbqvmtfJN8PGHQc8Xfd66qfDbC9BhFNzyGfzvr7Dq31DT11o/rNTn+NNqkmvRj98DHiFqxUaeGnqBtS9VKRwJMCVsWK1UJUtLgP3LrYCyfxmkHbfSG3e29l3fNs/qKwCo428FmhZ9rC9EAMT+0rXf3bwgoF/59XsUJzcH/vwBfnjYCnx3fV84ALYdbm3Xu/Lf0HWctTmWo7bNs2oXnnWswNU8tHzLn2/gc/ZotDfBx88KOgWFfwqLn7PW97rlM6smdv17Vs1q8XPgVRe63Vb0tfPyrNrO0lfAozaM+ZwvF8TRoLYHIzo1cc7zqHJVaoAxxqws+FlE+gO3aye/qnTrp8Hiv1sjoWr4QuvBVv9A68Hg3dg6Jy/X+ks+Zh3ErLU60LfPK/m6XW+HUVPK96/9rNMQHwHRa619S2I3QHYa+IdafRg1fQvnEYHBL8KXo6z1tnpPLP0+uTlWU9TaD62a25iZZ38XziACI9+D1ENWU5l3U6ufBqxRar88De2ug1tmnG3mc3GFW6bD12Pgx0fAyweCrz33ugl7rQAZsxZaD4GR7xGf68OynZE8NLA1nm5O/gNAlQtHhyl3A24HbgUOAvONMR86uWwVSocpVzPb58O8CRA0DAb+HZp2c6zWYQwkxcDp/LVajb1TkLGO7fwR1nxgdUL3uu/iypgUazVv7VtqTV7MywYEGne0akkt+kK7a8Hdq+TyzhxpzYt5fDN41Cz+3LQEmHsPRP0OoRNh2KSKG0acmQqfXwMnDlg1psNbrGatoOEw9suia1+ZqTDzBusPgLu+h5ZXQG42rHkfwt4C9xow4g1rOLQI7/y6mylh+1j17CD865Xwe1AVqkzzYESkLdbqxLcBJ7AWuHzaGHNJDt3QAFONHFhpdYr794K7FpT8BX2h8vKsfo/9K+Ce/0GL3heWPzcb9vwKG2fC3iVWmn8IBPS3gkrz3lDjApfyi14Ln4+Aoa9B/8eLPidqtTWqKz3RaoIqrtnJmVKPwGdXQ2aKtTpAmyEw9uuS/33STpAzfQSkxLO+wwv0OTIb12PbrPk/17x9pvaVlZNHvzeX0a15PT4bX+R3maokZQ0wecDvwARjzD477YAxppXTSlqJNMBUE4e3wufXQt3m8JdFF/5l7Yj0kzBtEGSnwwOrHGtiOhkFG2fBpq/g1FGrqaj7ndYOi6XNM3HEV7dA/EZ4cuu5807ycq2Z9ivfgnqBMObzix8ldjGO7bKCYbPuMO4bqxZSgDGGqBOn2RCVSERUIhFRJzmdEMM8z1fxlwSOU5fITi9y1Q33nrOW2MIth3h89iZm3hvKVW0bVvRTqRKUdaLlLVg1mBUishhr2X6dDqsqz8ko+Hq01WZ/xzznBBeAGvVg7Fcwfag1F2X8T8UPE049as3P2D4PxMVqsut5jzUkuAx7wxdr0Avw6SBrRNZVz1ppKYesWkv0augyDq5758ImPTpDo2B4age41yzUh3U4OZ1x09YRfeI0AHVruhMSUI9evfpz0nc+rlE/86+jfflfRAYNd6/g8cFtGNurBR5uLny1NpqA+jUZ0KaIlQZUleXIYpe1gFFYTWWDgZnAAmPMb84vXsXRGkwVl5YA04dZfScTfoOG7Zx/z23zYP4E6P0gXPPWucfy8qxmsKUvWzWdvo9afTY+Tlz6/ds74OAqeGILxG2ABQ9as9uv+0/lNIldoP/7cTvfrI/h5Rs60ifQl9YNa+PiUvhv1g1Riby9eDfhUYn416vBuF7Neee3PbxwbXvuv/KSbECp1sptsUsR8cXa4GusI9smVycaYKqwzFNWR/exP+HuhRfeL3IxFv/DmgV/07SzM+qP7bRGTMWug5YD4Pp3oUEF7Ih4dAdM7W8NEji63ZozM/oLaNDG+fe+SMdSMrji3yu4qZsfb43uUur5xhhW7U3g7V93sT0+BU83F9Y9P4R6tSpxkU1VpItai6wge9vhT+yXUs6XnW41Ux3ebHUYV2RwARj6qj0i6gnwbQV7FsEf/7Xml4yaemaEU4Vo3NGa1b99PoQ+AEP/Wb4DHJxo2qoD5OYZHh7U2qHzRYSr2jbkyqAG/PbnUVxENLhUQ7oanKq6Eg/CnLvgyDYY+X7huRJOkJyeTWpGNsYetWwwMOQj/L4bjtt0e5fwrrfDsH9BrfpOL08hN3wA/Z8ot4788IOJvPrTDv46tC1D2js+X+ZoSgbZuXkODRc+cSqTr9fHcGPXZgTUr3VB5RMRhnfUSZXVlQYYVTXt+RW+v9/6+fY51qx2JzuUlM6gd8LIzMkrdKyLPMJTbvOY63kTXlmD6Lcrnf5t0mnqU6OIKzmRR61yCy6R0Sf5y+fhpGfnMvHLSN68uTNjQpqXmm/lnuM8+s1G3FyExU9eSeM6Jdeipq8+SEZOLg8PqvpNeap8aYBRVUterrVb4ap/W30Mt34JvoGl5ysH8yLjyMzJ49UbOlLDwxUXEVzsFWRcpBtxGTficuAEYXuO8/2meABaNaxF/9YNGNXdj54B9SqknOVhS2wS98wIp6G3J9Pv6cUrC3fwzLytJJzK4sGrWhW5f4oxhumrD/L6LzsJauRNTOJpnvx2M1/d1xvXIjrrAZJOZzFrbTTXdm5Km0a1nf1YqorRAKOqjtOJ1qit/cuh253WsFv3iqkh5OUZ5kTE0q91fcb3a1nseXf1CSAvz7DrSCpr9ifwx74E5m+M47sNsax8dmDF12jKYHt8MndNX0/dWu58c38fmtWtwfTxvXh67hbeWryL46mZvHhd+3NGeGVk5/LCgu3M3xjHNZ2a8M6Yrvxv62Genb+Vj1fu55FiaidfrIniVGYOj2rt5bKkAUZVDfEbYc7d1iTFkf+FHuMrrvMcWHfgBHEn03lmeOnDn11chA7N6tChWR3uG9CK2MTTDHonjI/D9vPqRe7xfiozh5lroli15zh/6d+SEZ2aXtT1zrfzcAp3Tl+Pt5c739xnBRcADzcX3hvbDd9aHsz44yAn0jJ5e3RXPNxcOJaSwQNfRbIpJoknrw7i8cFBuLgIY0L8WbX3OJOX7KFv6/r0aHFuDS41I5sZqw8ytENjXVr/MqUBRlWubHvP9z/es2a/3/sr+PWo8GJ8FxGLt5dbmTqUm/vW5JYe/szeEMvDg9qU2idRlNSMbGatjebT3w+QdDqbht6ePPjVRkZ1a8YrN3Skbs2SR1Bl5eSxcMshkk5n0SOgHp2a+eDh5nLOOXuPpnLnZ+vxcnPlm/t709z33A56Fxfh5ZEdaFTHk38v3k1iWhYPD2zDU99tJjk9m6l39OCazmcDnogw6abObI5N4vHZm/jliQHUKbDD5JfroknJyOGxwVp7uVxpgKlGMrJzGfKflTw9vC03dfcvPUNVF73WWjH3xF5rZNbwSUWvKlyC7Nw8jiRnEJ+UTvzJ9HPeAxvU4p83dix1P/bk09ks2n6EsSHN8XIv2yq9jwxqw7yNcUwN288rN3R0OF9qRjYz10Tx2eqDJJ3OZlC7hjxxdVs6NqvDRyv28eHyfazZf4I3b+nM4ODCo7zSs3L5dkMM01Yd4HByxpl0TzcXujavS0hAPUJa1qN+LU8mzIzAxUX45v7exY7mEhEeHtiGBrU8ee77rfy+NwG/ujWY/1A/OjQrXAvxqeHOf8d159ZP1vKP77fxwW3dERFOZ+Xw2e8HuaptQ7r4O2nFBVXlaYCpRrbFJxOflM6KXcerd4DJSIFlr8KGz6w9UO783loY8QIt+fMoj3yzkazzRn019Pakjpcbq/clMLh9Iwa1a1TidRZuiScrJ49bHRhBVZwW9Wtyc3c/ZofH8PDA1jQqpRZjjOGTVQeYGraf5PRshgQ34vEhQXRtfvbL+Mmr23J1+8b8bc4W7v0igjE9/XlpZAfqeLmTmpHNl+uimf77QU6kZRHa0pe3bulCcBNvIqNPEmG/pq06wJQwazJ1/VoefDuxD60alt7Zfmuv5jT09mTR9sP8fUQw9WsXvxdNz4B6/HVoW97+dTdXBjXk1l7N+WZ9DIlpWTw+RGsvlzMNMNXIhqhEALbGJVVySUqQkQIHwqxl5Wv4Wut61fS1JiaKwO7F1o6GKYegz8PWGlueFz66KDMnl1d/2kGAb03uGxCIX92a+NWrQVMfL7zcXcnKyWPouyt585ddXBnUsNhRTgBzIuJo37QOnfwurp/g0cFt+H5TPB+vPMD/jexQ4rkz/ojizUW7GNSuIX8d2o7O/j5FntfJz4eFj/Xnv0v38vHK/fyxL4FrOjdlbkQsKRk5XNm2IY8OakNo4Nma3zWdm55pykrPymVLXBI7DqUwOLgRgQ0cn4cyKLgRg4JLDs75HryqNav3JvDywh109KvDtFUH6NuqPj0DLqxGqi4tGmCqkciokwBEnThN8ulsfGqWcZ92Z0k5BF/eDMd3Fj4mrtYilemJ0KgD3DrLWsa+jL5cG03cyXS+nBDKgKDCq+t6uLnw9xHBPPz1RuZFxjK2V9HbJf95KIVt8cm8PLJDqU1ppQmoX4tR3fz4en00Dw5sRSPvomsxEVGJvPHLToZ1aMwnd/Us9b6ebq48OyKYoR0a87e5W5i++iDDOjTm0cFtSm1+quHhSp9W9enTyrmTQl1dhPfGdWPEe6u49eO1pGXl8t64bk69p6r6NMBUE3l5hojokzT3rUFsYjrb4pO5IqgKrSx7fA98dTOkJ1lzV2o3spa9Tz9pDT/O/9k30Frm5CI2wkpOz+bDFfsYENSgyOCS75pOTejRoi7/+W0PI7s2o6ZH4f/c50TE4uHqwqhu5bNI5aOD27BgUxzTVh7gxesL12KOp2byyDcb8a9Xg3du7XpBQa17i3osfuJKUjKyaVBCk1VlaVzHi3fGdGXCzAhCAurR18lBTVV9LqWfUjYi0lxEVojIThHZISJP2Om+IrJERPba7/UK5HleRPaJyG4RGV4gvaeIbLOPvS/2/5Ui4iki39np60WkZYE84+177BWR8c56zoqy//gpktOzGd+3JQBb46tQM1lcJMwYDjkZ8Jf/QYcbrH3v210D3W6Hfo/CkJfg+snQ77GL3mUxv9/iuWuCSzxPRHjhuvYcS83ks98PFjqemZPLD5vjGdaxcbmtcxXYwKrFfLU+muOpmeccy8nN4/HZm6wRWXf2PGfElaM83FyqZHDJN6R9Y6aPD+G9cd0uukaoqj+nBRggB/ibMaY90Ad4REQ6AM8By4wxQcAy+zP2sXFAR2AEMEVE8of0TAUmAkH2a4SdPgE4aYxpA7wLvGVfyxd4GegNhAIvFwxkVV76SchIPidpg908NqR9YwLq12RrbHJROSve3qUw83rwqmMNMXbyZlfxSenM+OMgN3Xzo2OzovstCuoZ4MuIjk34eOV+jqVmnHNsyZ9HSTqdfVGd+0V5dHAbsnLy+PT3A+ekT16yh7UHTvCvUZ0v6XkhQ9o31i2NFeDEJjJjzGHgsP1zqojsBPyAG4GB9mkzgTDg73b6t8aYTOCgiOwDQkUkCqhjjFkLICKzsPanWWTnecW+1jzgQ7t2MxxYYq/+jIgswQpKs531vOUm6g9rgcfsDGtHxD4PgW8gEdGJNKjtQcv6NeniX5eN0VbAwRiIDYfwabBzIeRmFX3dWo3gyqeh51/Kb5/2rXPgh4egUXu4Y75jOz9epMm/7QHgr8PaOpzn79cEs3TnUd5bupfXb+p8Jv27DbH41a1B/3LexKpVw9rc0LUZX66NZuKVrWhQ25Mlfx5lSth+bgttweie1XgEoFIXoEL6YOymq+7AeqCxHXwwxhwWkfxhKn7AugLZ4uy0bPvn89Pz88Ta18oRkWSgfsH0IvIULNdErJoRLVoU3QlcoSJnwv/+Zm2x2yYEImbAhk+h/UjS9venZ4DVIdzFz4fftkSRuvZzvLfMgCNbwdPH2p63VjF9EtF/wKJnrR0Rh/wfdLyp7DPl83Jh3RT47UVrP5RxX1sd+E6283AK32+KY+KAVhf0F3Jgg1rc2SeAL9dFc2//lrRp5E18Ujqr9yXw2OCgEkeYldWjg4P4ccshPv39ALeHtuCvczbT2c+Hl0sZXabUpcTpAUZEagPzgSeNMSkltMsWdcCUkF7WPGcTjJkGTANrw7HiCuZ0uTnw2wuw/mNoPQRGz7C2A776FQj/hLwNM/gk80eOJHSDbY9x3dH1jPb8Bu9fT1kjsq5/FzrfWvJwX2Ng7xJrB8Z5f4E1H1j7iQQOcKyMOZlwYCXs+gl2L4K049B+JNz8WYXtSfLmol3U8XLn4YEXPrfi8SFBzI+M481Fu/hsfC/mRcRhDIxxUm2iTaPajOxi1WJW7j6OiwhT7uhR5omcSlVHTg0wIuKOFVy+NsZ8bycfFZGmdu2lKXDMTo8DCjaG+wOH7HT/ItIL5okTETfAB0i00weelyesnB6rfKWfhLl/gQMroM8j1pd+/l7udZrC1a+wpP6drJ33X57PXQ7zJ9BUXFmc15O0rvcy+pZxjtVERKDtMGtC45ZvYcUkq+8kaBj0uNtaBt69Jrh5We/uXuDiDjFrYOfPVnDKSgUPbwgaagWXDjeCS8V8Yf6xL4GVe47zwrXtyzQ827eWBw8Nas2/F+9mzf4E5kbG0r9N/ULLpZSnx4e04aeth9h1JJXP7+nl1HspVRU5LcDYfSHTgZ3GmMkFDi0ExgNv2u8/Fkj/RkQmA82wOvPDjTG5IpIqIn2wmtjuBj4471prgdHAcmOMEZFfgdcLdOwPA5530qOW3fE9MHssJMXCDR9Cj7uKPG1dXCbfulzLPx7/D8StReq3ZvL0/bQ4VZPRF9rM5eIK3e+wdkZc/wn8Phn2/lZynpoNoNNNEDwSWl0FbhU7iikvz/DGop341a3BXX0Dynyde/sH8tXaaB6fvYmEU1kOLWx5Mdo08ua5EcF4e7k7PGFRqUuJM2sw/YG7gG0istlO+wdWYJkjIhOAGGAMgDFmh4jMAf7EGoH2iDEm1873EPAFUAOrc3+RnT4d+NIeEJCINQoNY0yiiLwGbLDP+2d+h3+VkJMJW7+DX18AVw8Y/xME9C329Mjok3T1r4uHh4f1BQ909k9g1Z4EjDFlGw7qXgOueBJC7oXE/dagguzT1lDj7HTrlZNhbdPbvHeF1VSK8tPWQ2yPT+HdsV0vqonJy92Vvw1rx9/mbqFOGRe2vFAPXOXYFsFKXYqcOYpsNUX3hQAUufCUMWYSMKmI9Aig0DroxpgM7ABVxLEZwAxHy1shMpIh4nOrryX1MDTrAbfOtNbjKkZaZg47DqXw0HlfVF396/L9xniOpGRc3B4kXnWgWfey53eyjOxc3v51Nx2a1uHGrhc/GfKm7n78uOUQPVvU0/4QpZxMZ/JXhOR4a9RV5EyrHyPwKrjxQ6tDv5Tax5bYJHLzDD1bnjuNJ3/tqi2xyU7Z5CovzzD+83Bu6eHPqO7lM8s9PimdjOxcWjuw2GK+13/ZSdzJdL65r8s5G2CVlYuLMOve0Iu+jlKqdM6caKnSTsCCB+G/XazhwW2HwcSVMH4ha6U77y3bW+olNkSdRIRCmzl1aFoHNxdhmwMz+qMS0pizIbbU8wraEpfE73sTmLbqQOknO2Dpn0cZ/u4qbvhgNTsPpziUZ/H2I8xaG839AwLpV85zVZRSzqcBxlkyUqy1ubZ/D73ug8c3WcOPm3UjMyeXZ+Zt4b2le1l34ESJl4mITqRdY298apw7csrL3ZW2jb3ZGlf6jP43F+3i2flbiUpIc7j4S/48CsCfh1PYezTV4Xzny8szvL9sL/fNiqBlg5p4e7lz38yIQrPqzxeflM7f52+li78PzwwveUkYpVTVpAHGGbLTYfZtcHQ7jP0SrnnLmjxpy18JuKaHKx8sL74Wk5tn2BSTREjLole56drch61xyRhT/BSehFOZLN1pBYtF2484/AhL/jxKcBNvXAR+3Hyo9AxFOJWZw0NfRzJ5yR5u6u7HvAf78dn4EBLTspg4K5KM7Nwi8+Xk5vHkt5vIyc3j/XHdC+3MqJSqHvT/3PKWm23Na4n+A0Z9DG2Hn3M4+XQ2Hyzfx5VtG/LXoW35Y98JIqKKHuC260gKpzJz6NWy6D01OvvVJTk9m5jE08UWZ8HGeHLyDE19vFi0/bBDjxCVkMbeY6cYE9Kc/m0a8OOW+BKDWHHXuHnKHyzdeYyXru/A5FutEWCd/Hx4b1w3tsQl8fTcLUVe9/3l+9gQdZJJN3Wm5QXsX6KUqlo0wJSnvDz48VHYswiuewe6FB7gNmXlPlIysnluRDB39A6gQW0P/ltMX0yEvcBlz4CiazBd7I7+4prJjDF8FxFL9xZ1uadfS7bGJRNbQjDKl1/jGdahMTd28yM2MZ2NMY6v3rxyz3Fu+HA1x1IzmXVvKBOuCDxnKPXwjk34+4hgft56mPeWnvvs6w6c4MPle8t1cIFSqnJogCkvxsDi52DrtzD4Ravf5TzxSel8/kcUN3f3p0OzOtTwcOX+Aa34fW8CG2NOFjp/Q1QiTX288Ktb9Cixdk288XBzKXaHy40xSew7doqxIc25ppO1w+FiB5rJfrObx5r71mR4x8Z4urnw4+b4UvP9f3t3Hl1lmR9w/PvLTvaAIWRBAsoWw5qIoI64sM8UcBg7llJR7NGZ6lHb0VbbnjOt1hZ7ejoz1lOXo1AdpyKKUx1XEB1RZAsgCGELkrCaBBIgJMAlN0//eJ+b3ITcw6nz7AAAD3hJREFUy83yZrn8Pufcc988933fvO+PkF+e5X0egE92fs89SzeSldqHPzx4Y8CJJO+/aQh3FOTwm9X7Gs9dVevhkWXfkNsvgSfnhL6uvVKqZ9IE01n+uBg2vgiTHoQfPNrqLr6ZgH/hNxPwgomDSIuP5r9a1GKMMRSVVlMwKC3gg5TRkRHkZSazLUANZvmmQ8THRPKjMVlc2S+ea7KSL9lMVl3roai0iql5zszISXHRTBmZwfvbj3HB2xD02AveBv71w10My0jinb+6PujUKCLC07ePYsLgvjz21naKSqv427e3UVXr4dk/G0dCrI6gV6q30wTTGdY/D18sdqbXn/YvrT7bUnzUmQn4nhtyyfKrkSTERvGXPxjC53sqm9VEjpw8y/enzwXsf/EZnZPCziOn8DY078uoPV/P+9uP8sNRmSTaX9azRmWy5eBJjp06G/B8n+2uoMHAlJFNU+/PGZtFVa2Hr0qOB72WFZsPU3aijkenDW919ciWYqIieHFBAZmpccx/eQOf7qrg8ZkjyM92f2ZmpZT7NMF0VOVe+OTvnckff/SbgA9OLv54Nyl9Wp8J+K5Jg0jpE82zq0sayzaXBe9/8Rmdk0qtx8t3lWealX+w/Ri1Hi8/vbZp/tCZ+c7UKMGayVYVl5ORHMsov1/yNw/vT0qfaN7dGriZ7Hy9l2dX72PswFRuGxn6vFtpCTEsufta4mMimTIyg3tuyA35WKVUz6YJpqPSh8H85TDvlaZZkFv4at9x1uyt5MFbrr7oeRZwmqHuvXEwn+4qZ8cRp7lrU2kVibFRjBiQFPTbB+rof7PoEEPSE5olqCHpiYwYkMRH37aeYM5d8LJmXyVTRmY0e2o+JiqCWaMGsLK4nDpPfavHvrHhIEdPnePRacPbPDfaVemJfPV3t/LiXxToMrtKhRFNMJ1h6NSAMwz7ZgLOSQs+E/DC63NJiotqfC6mqLSacVemEhUZ/J/oqvRE4mMimzWvlVTUsLmsmp8WDrzoF/bM/Ew2lVVRcfriBx3X7T9Bncfb2P/ib87YbOo83sYHMP2d9Xh57vP9XDe4Lzdc3S/o9QaSGBvlysJfSqnuownGZe9tO8rOo6d5bPpwYqMCT66Y0ieaRTcM5pOd5Ww8UMWe8hoKBwXvfwGIjBDys1LYfqSpBrO86DBREcKPx1+8mNasUQMwxhnt1dLK4nISYiKZdNXFSWJCbl8yU+JafejytXWlHD9znkent732opQKX5pgXOSbCTg/O5k/GZ11yf0X3TCYxNgoHlm2FWMI+AR/S6NzUig+epoL3gY89Q2s2HyYW0f0Jz3p4lrV0Iwkru6fyIctmskaGgyrd5UzeXh6q4kwIkKYPSaLNXsrqar1NJbXnLvA81/sZ/Kw9EsOSFBKXV40wbjonS1HOHLyLI/PGBnSTMAp8dHcfX0uR0+dIzJCGDswNaTvM3pgKufrG9hbXsNnu8s5Uetp1rnf0qz8AWw4cILjZ843lm0/coqKmvOtNo/5zBmbTX2D4YNvm4Y6L/mqlJN1F3h0mruLdymleh9NMC4xxvDb9WWMzExuU7/EvTcOJj4mkrzM5JCfBRmd3dTR/+amQ/RPimXysPSA+88clUmDgZU7m/pTVhV/T2SEcMvwwCPARmYmMbR/YuNospN1Hl7+8jumX5PRuHyAUkr5aIJxyZaDJ9l17DQLJl7Zpn6JtIQYXlhQwD/Nzgv5mEH94kmOi2JVcTlf7K3kJwU5QQcHjBiQxOArEpo9dPlpcQXX5qaRGh8T8DgRYe64bIrKqjlUVcdLa77jjKeev5mqtRel1MU0wbjkd+vLSIyNYu7Yts+nddOwdApC6OD3ERFG56Q2PiT5p4WBm8d8+8/IH8DX+09QXevh4Ik69pTXMDXv0ksIzx7j9CUtWXuApWtLmT0mi+GXGEqtlLo8aYJxQVWth/e/PcaPx2d32ZQnvudhJg7pG9IMxLPyM/E2GFYVl7Oy2Onwnxak/8VnYN94CgalsXRtKR5vAw/fNrRjF66UCls64ZML3io6hKe+gQUTAz/30tnG2AEBwTr3/eVnJ5OT1ocPdxzjrMfbOLllKOaOzWJzWTXzxmczpA3LHyulLi+aYDpZQ4PhfzceZEJuX4ZldF3T0ZSRGbywYHxIzVzgNJPNGpXJ0rUHaDDw88lXhfy95o7LZvf3NTyktRelVBDaRNbJviw5TtmJOv584pVd+n0jI4QZ+Zltehp+Zv4ALngN3gYTdHhyS0lx0Tx9+ygykuPac6lKqcuE1mA62evry+iXEMOM/NBqEt1p7MBUslLi8BrTbHJLpZTqDJpgOtGRk2dZvauc+ydfFXRamJ5CRFg8bzReY0J6EFQppdpCE0wnWrbxIAaYP6Frm8c64qYgD2QqpVRHaB9MJ7ngbWDZpkPcMrx/yKOxlFIqnGmC6SQrd5ZTWXOeBV3cua+UUj2VJphO8vr6MrJT+zB5WOirOSqlVDhzLcGIyBIRqRCRHX5lfUVklYjss+9pfp89ISIlIrJHRKb7lReIyLf2s2fFTuwlIrEi8qYt3yAiuX7HLLTfY5+ILHTrHn1KKmpY990J5l93pS6apZRSlps1mP8BZrQoexxYbYwZCqy2XyMiecCdwDX2mP8WEd8wrOeB+4Ch9uU7571AtTHmauBXwDP2XH2BXwLXAROAX/onMje8vv4g0ZES8lP0Sil1OXAtwRhj1gBVLYrnAK/a7VeBuX7ly4wx540xB4ASYIKIZALJxph1xhgDvNbiGN+53gZus7Wb6cAqY0yVMaYaWMXFia7T1HnqWbHlMDPyM7kisfVlk5VS6nLU1X0wGcaYYwD23ddhkQ0c8tvvsC3Lttsty5sdY4ypB04B/YKc6yIicp+IFIlIUWVlZbtuqOZcPZOHpXPXpK6bd0wppXqDnvIcTGsdFyZIeXuPaV5ozEvASwCFhYWt7nMpGclxPDd/fHsOVUqpsNbVNZhy2+yFfa+w5YcB/w6MHOCoLc9ppbzZMSISBaTgNMkFOpdSSqku1NUJ5j3AN6prIfCuX/mddmTYYJzO/I22Ga1GRCba/pW7WhzjO9dPgM9sP80nwDQRSbOd+9NsmVJKqS7kWhOZiLwB3AxcISKHcUZ2LQaWi8i9wEHgDgBjzE4RWQ4UA/XAA8YYrz3Vz3FGpPUBPrIvgFeA34pICU7N5U57rioReQrYZPd70hjTcrCBUkopl4nzR78qLCw0RUVF3X0ZSinVq4jIZmNMYWuf6ZP8SimlXKEJRimllCs0wSillHKFJhillFKu0E5+S0QqgbLuvo4OugI43t0X0YNoPJrTeDTRWDTXkXgMMsa0unKhJpgwIiJFgUZzXI40Hs1pPJpoLJpzKx7aRKaUUsoVmmCUUkq5QhNMeHmpuy+gh9F4NKfxaKKxaM6VeGgfjFJKKVdoDUYppZQrNMEopZRyhSaYHk5ElohIhYjs8CsbIyLrRORbEfmDiCTb8mgRedWW7xKRJ/yOKbDlJSLyrF3+oFdpYyxiRGSpLd8mIjf7HdPrYwEgIgNF5HP7b71TRB625X1FZJWI7LPvaX7HPGHve4+ITPcr79UxaWssRKSf3f+MiDzX4ly9OhbQrnhMFZHN9r43i8itfudqfzyMMfrqwS/gJmA8sMOvbBMw2W4vAp6y2/OBZXY7HigFcu3XG4FJOCt+fgTM7O57czkWDwBL7XZ/YDMQES6xsPeRCYy320nAXiAP+HfgcVv+OPCM3c4DtgGxwGBgPxAZDjFpRywSgBuBnwHPtThXr45FO+MxDsiy2/nAkc6Ih9ZgejhjzBqc9W78DQfW2O1VwDzf7kCCXeGzD+ABTtvVQ5ONMeuM8xPzGjDX9YvvZG2MRR6w2h5XAZwECsMlFgDGmGPGmC12uwbYBWQDc4BX7W6v0nR/c3D+ADlvjDkAlAATwiEmbY2FMabWGPMVcM7/POEQC2hXPLYaY3wr/+4E4uwCkB2KhyaY3mkHMNtu30HTEtFvA7XAMZwF3f7DOIutZeMsJe1z2JaFg0Cx2AbMEZEou0pqgf0sLGMhIrk4f4VuADKMsxos9r2/3S0bOOR3mO/ewyomIcYikLCKBbQrHvOArcaY83QwHppgeqdFwAMishmn+uux5RMAL5CF0wTyCxEZglO1bSlcxqcHisUSnP8MRcCvga9xVksNu1iISCKwAnjEGHM62K6tlJkg5b1OG2IR8BStlPXKWEDb4yEi1wDPAPf7ilrZLeR4uLZksnKPMWY3MA1ARIYBP7QfzQc+NsZcACpEZC1QCHwJ5PidIgc4ShgIFAtjTD3w1779RORrYB9QTRjFQkSicX6B/M4Y844tLheRTGPMMdvEUWHLD9NUw4Omez9MGMSkjbEIJCxiAW2Ph4jkAL8H7jLG7LfFHYqH1mB6IRHpb98jgH8EXrAfHQRuFUcCMBHYbavCNSIy0Y4AuQt4txsuvdMFioWIxNsYICJTgXpjTHE4xcJe/yvALmPMf/p99B6w0G4vpOn+3gPutG3rg4GhwMZwiEk7YtGqcIgFtD0eIpIKfAA8YYxZ69u5w/Ho7tEO+rrkaJA3cPpULuD8NXEv8DDOqJC9wGKaZmRIBN7C6aQrBh7zO08hTn/FfuA53zG96dXGWOQCe3A6Nz/FmVI8bGJh7+NGnOaK7cA39jUL6IczwGGffe/rd8w/2Pveg99ooN4ek3bGohRn0MgZ+/OUFw6xaE88cP44q/Xb9xugf0fjoVPFKKWUcoU2kSmllHKFJhillFKu0ASjlFLKFZpglFJKuUITjFJKKVdoglFKKeUKTTBKhRERiezua1DKRxOMUt1ERJ7yrdNhv35aRB4SkcdEZJOIbBeRf/b7/P/sWh07ReQ+v/IzIvKkiGzAmVZdqR5BE4xS3ecV7LQddqqbO4FynClcJgBjgQIRucnuv8gYU4DzZPVDItLPlifgrJFznXGmoFeqR9DJLpXqJsaYUhE5ISLjgAxgK3AtzuSdW+1uiTgJZw1OUrndlg+05SdwZtBe0ZXXrlQoNMEo1b1eBu4GBuAsMXAb8G/GmBf9dxJnyecpwCRjTJ2I/BGIsx+fM8Z4u+qClQqVNpEp1b1+D8zAqbl8Yl+L7DoeiEi2nTE6Bai2yWUEzkzZSvVoWoNRqhsZYzwi8jlw0tZCVorISGCdMzs6Z4AFwMfAz0RkO85MyOu765qVCpXOpqxUN7Kd+1uAO4wx+7r7epTqTNpEplQ3EZE8oARYrclFhSOtwSillHKF1mCUUkq5QhOMUkopV2iCUUop5QpNMEoppVyhCUYppZQr/h+BwXOIkRvK1gAAAABJRU5ErkJggg==\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "sns.lineplot(data = grouped_gender, x='year', y='Approved Amount', hue='Gender')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "There seems to be a small but systematic difference in the average amount awarded.\n",
    "\n",
    "We can now use a plotting library that is essentially a Python port of ggplot to add even more complexity to this plot. For example, let's split the data also by Field:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 47,
   "metadata": {},
   "outputs": [],
   "source": [
    "import plotnine as p9"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 48,
   "metadata": {},
   "outputs": [],
   "source": [
    "grouped_gender_field = merged_projects.groupby(['Gender','year','Field'])['Approved Amount'].mean().reset_index()\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 49,
   "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>Gender</th>\n",
       "      <th>year</th>\n",
       "      <th>Field</th>\n",
       "      <th>Approved Amount</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>female</td>\n",
       "      <td>1975.0</td>\n",
       "      <td>Biology</td>\n",
       "      <td>95049.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>female</td>\n",
       "      <td>1975.0</td>\n",
       "      <td>Humanities</td>\n",
       "      <td>95451.666667</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>female</td>\n",
       "      <td>1975.0</td>\n",
       "      <td>Mathematics</td>\n",
       "      <td>125762.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>female</td>\n",
       "      <td>1976.0</td>\n",
       "      <td>Biology</td>\n",
       "      <td>183154.200000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>female</td>\n",
       "      <td>1976.0</td>\n",
       "      <td>Humanities</td>\n",
       "      <td>68590.750000</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>271</th>\n",
       "      <td>male</td>\n",
       "      <td>2019.0</td>\n",
       "      <td>Humanities</td>\n",
       "      <td>523397.013072</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>272</th>\n",
       "      <td>male</td>\n",
       "      <td>2019.0</td>\n",
       "      <td>Mathematics</td>\n",
       "      <td>632188.796040</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>273</th>\n",
       "      <td>male</td>\n",
       "      <td>2020.0</td>\n",
       "      <td>Biology</td>\n",
       "      <td>694705.243590</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>274</th>\n",
       "      <td>male</td>\n",
       "      <td>2020.0</td>\n",
       "      <td>Humanities</td>\n",
       "      <td>520925.507246</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>275</th>\n",
       "      <td>male</td>\n",
       "      <td>2020.0</td>\n",
       "      <td>Mathematics</td>\n",
       "      <td>624141.068182</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>276 rows × 4 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "     Gender    year        Field  Approved Amount\n",
       "0    female  1975.0      Biology     95049.000000\n",
       "1    female  1975.0   Humanities     95451.666667\n",
       "2    female  1975.0  Mathematics    125762.000000\n",
       "3    female  1976.0      Biology    183154.200000\n",
       "4    female  1976.0   Humanities     68590.750000\n",
       "..      ...     ...          ...              ...\n",
       "271    male  2019.0   Humanities    523397.013072\n",
       "272    male  2019.0  Mathematics    632188.796040\n",
       "273    male  2020.0      Biology    694705.243590\n",
       "274    male  2020.0   Humanities    520925.507246\n",
       "275    male  2020.0  Mathematics    624141.068182\n",
       "\n",
       "[276 rows x 4 columns]"
      ]
     },
     "execution_count": 49,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "grouped_gender_field"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 50,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 640x480 with 3 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/plain": [
       "<ggplot: (305412337)>"
      ]
     },
     "execution_count": 50,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "(p9.ggplot(grouped_gender_field, p9.aes('year', 'Approved Amount', color='Gender'))\n",
    " + p9.geom_point()\n",
    " + p9.geom_line()\n",
    " + p9.facet_wrap('~Field'))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### 12.5.2 From employee to applicant\n",
    "\n",
    "One of the questions we wanted to answer above was how much time goes by between the first time a scientist is mentioned as \"employee\" on an application and the first time he applies as main applicant. We have therefore to:\n",
    "1. Find all rows corresponding to a specific scientist\n",
    "2. Find the earliest date of project\n",
    "\n",
    "For (1) we can use ```groupby``` and use the ```Person ID SNSF``` ID which is a unique ID assigned to each researcher. Once this *aggregation* is done, we can summarize each group by looking for the \"minimal\" date:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 51,
   "metadata": {},
   "outputs": [],
   "source": [
    "first_empl = merged_empl.groupby('Person ID SNSF').date.min().reset_index()\n",
    "first_appl = merged_appl.groupby('Person ID SNSF').date.min().reset_index()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "We have now two dataframes indexed by the ```Person ID```:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 52,
   "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>Person ID SNSF</th>\n",
       "      <th>date</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>1611</td>\n",
       "      <td>1990-01-10</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>1659</td>\n",
       "      <td>1988-01-11</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>1661</td>\n",
       "      <td>1978-01-07</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>1694</td>\n",
       "      <td>1978-01-06</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>1712</td>\n",
       "      <td>1982-01-04</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   Person ID SNSF       date\n",
       "0            1611 1990-01-10\n",
       "1            1659 1988-01-11\n",
       "2            1661 1978-01-07\n",
       "3            1694 1978-01-06\n",
       "4            1712 1982-01-04"
      ]
     },
     "execution_count": 52,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "first_empl.head(5)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Now we can again merge the two series to be able to compare applicant/employee start dates for single people:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 53,
   "metadata": {},
   "outputs": [],
   "source": [
    "merge_first = pd.merge(first_appl, first_empl, on = 'Person ID SNSF', suffixes=('_appl', '_empl'))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 54,
   "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>Person ID SNSF</th>\n",
       "      <th>date_appl</th>\n",
       "      <th>date_empl</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>1659</td>\n",
       "      <td>1975-01-10</td>\n",
       "      <td>1988-01-11</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>1661</td>\n",
       "      <td>1978-01-07</td>\n",
       "      <td>1978-01-07</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>1694</td>\n",
       "      <td>1985-01-01</td>\n",
       "      <td>1978-01-06</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>1712</td>\n",
       "      <td>1982-01-04</td>\n",
       "      <td>1982-01-04</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>1726</td>\n",
       "      <td>1985-01-03</td>\n",
       "      <td>1985-01-03</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>10336</th>\n",
       "      <td>748652</td>\n",
       "      <td>2019-01-12</td>\n",
       "      <td>2019-01-12</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>10337</th>\n",
       "      <td>748760</td>\n",
       "      <td>2020-01-03</td>\n",
       "      <td>2020-01-03</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>10338</th>\n",
       "      <td>749430</td>\n",
       "      <td>2020-01-04</td>\n",
       "      <td>2020-01-04</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>10339</th>\n",
       "      <td>749991</td>\n",
       "      <td>2020-01-03</td>\n",
       "      <td>2020-01-03</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>10340</th>\n",
       "      <td>750593</td>\n",
       "      <td>2020-01-01</td>\n",
       "      <td>2020-01-01</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>10341 rows × 3 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "       Person ID SNSF  date_appl  date_empl\n",
       "0                1659 1975-01-10 1988-01-11\n",
       "1                1661 1978-01-07 1978-01-07\n",
       "2                1694 1985-01-01 1978-01-06\n",
       "3                1712 1982-01-04 1982-01-04\n",
       "4                1726 1985-01-03 1985-01-03\n",
       "...               ...        ...        ...\n",
       "10336          748652 2019-01-12 2019-01-12\n",
       "10337          748760 2020-01-03 2020-01-03\n",
       "10338          749430 2020-01-04 2020-01-04\n",
       "10339          749991 2020-01-03 2020-01-03\n",
       "10340          750593 2020-01-01 2020-01-01\n",
       "\n",
       "[10341 rows x 3 columns]"
      ]
     },
     "execution_count": 54,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "merge_first"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Finally we merge with the full table, based on the index to recover the other paramters:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 55,
   "metadata": {},
   "outputs": [],
   "source": [
    "full_table = pd.merge(merge_first, merged_appl,on = 'Person ID SNSF')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Finally we can add a column to that dataframe as a \"difference in dates\":"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 56,
   "metadata": {},
   "outputs": [],
   "source": [
    "full_table['time_diff'] = full_table.date_appl-full_table.date_empl"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 57,
   "metadata": {},
   "outputs": [],
   "source": [
    "full_table.time_diff = full_table.time_diff.apply(lambda x: x.days/365)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 58,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([[<matplotlib.axes._subplots.AxesSubplot object at 0x12ba24970>]],\n",
       "      dtype=object)"
      ]
     },
     "execution_count": 58,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "full_table.hist(column='time_diff',bins = 50)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "We see that we have one strong peak at $\\Delta T==0$ which corresponds to people who were paid for the first time through an SNSF grant when they applied themselves. The remaining cases have a peak around $\\Delta T==5$ which typically corresponds to the case where a PhD student was payed on a grant and then applied for a postdoc grant ~4-5 years later.\n",
    "\n",
    "We can go further and ask how dependent this waiting time is on the Field of research. Obviously Humanities are structured very differently "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 60,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "sns.boxplot(data=full_table, y='time_diff', x='Field');"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 61,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAYgAAAEGCAYAAAB/+QKOAAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADh0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uMy4yLjEsIGh0dHA6Ly9tYXRwbG90bGliLm9yZy+j8jraAAAgAElEQVR4nOzdd5hU9dXA8e+Zme27tC30IuyCIk2a5dWELmhUNKhgw1hQX1s0vrbEWLDHxBqTmIgaY0ONkVCkiogiTcrSBJbetsHC9p1y3j9mBlZc2DYzd2bn93meffbOnXZ2Z+4999dFVTEMwzCM49msDsAwDMMITyZBGIZhGDUyCcIwDMOokUkQhmEYRo1MgjAMwzBq5LA6gEBJS0vTLl26WB2GYRhGRFm5cmWBqqbXdF+TSRBdunRhxYoVVodhGIYRUURk54nuM1VMhmEYRo1MgjAMwzBqZBKEYRiGUaOwSBAiYheRVSIy3Xe7lYjMFZEtvt8trY7RMAwj2oRFggDuBjZWu/0gMF9Vs4D5vtuGYRhGCFmeIESkA3Ah8I9quy8B3vFtvwOMDXVchmEY0c7yBAG8BNwPeKrta62q+wF8vzNqeqKITBKRFSKyIj8/P/iRGoZhRBFLE4SI/ALIU9WVDXm+qr6hqgNVdWB6eo3jPAwjLJhp9Y1IZHUJ4n+Ai0VkB/AhMExE/gXkikhbAN/vPOtCNIzGmTJlCnfeeafVYRhGvVmaIFT1IVXtoKpdgPHAAlW9BpgGTPQ9bCLwuUUhGkajvf3226xdu9bqMAyj3qwuQZzIs8BIEdkCjPTdNgzDMEIobOZiUtWFwELfdiEw3Mp4DMMwol24liAMwzAMi5kEYRiGcQKqyiuvvMKmTZusDsUSJkEYhmGcQGVlJZ988glPPfWU1aFYwiQIwzCMWuzfv9/qECxhEoRhGEYtRMTqECxhEoRhGEYtonUkvEkQEaKiosLqEAwj6ng83iniTAnCCFubNm1i1KhRZs1twwixaC05+JkEEQH80zQsWbLE4kgMI7qYBGFEjGj/shqGVaL12DMJIoJEaz2oYRjWMAkigkTrVYxhWCXajzmTICKIKUEYhjWi9dgzCSKCRPvVjGGEWrQfcyZBRJBovYoxDKu43W6rQ7CUSRARJNqvZgwj1KqqqqwOwVImQUQQU4IwjNDyJ4hovTgzCSKCROuX1DCsUl5eDkTvxZmlCUJE4kVkmYisEZH1IvK4b38rEZkrIlt8v1taGadhGNGprKzM6hAsZXUJohIYpqp9gX7AaBE5C3gQmK+qWcB83+2oF61XMYZhFVOCsJB6lfhuxvh+FLgEeMe3/x1grAXhhR1TxWQYoeUvQfhndY02VpcgEBG7iKwG8oC5qroUaK2q+wF8vzOsjNEwjOjkL0EQnQUI6xOEqrpVtR/QARgsIr3q+lwRmSQiK0RkRX5+fvCCNAwjKvkTRLSW3i1PEH6qWgQsBEYDuSLSFsD3O+8Ez3lDVQeq6sD09PSQxWqVaK0HNQyrVFZWWh2CpazuxZQuIi182wnACGATMA2Y6HvYROBzayIML9F6FWMYVon2cRAOi9+/LfCOiNjxJqupqjpdRJYAU0XkRmAXcLmVQYYLU4IwjNByOp0AuF3ROeWGpQlCVdcCZ9SwvxAYHvqIwlu0XsU0FapqknyE8fdeitZjL2zaIIzamZNLZHO5XFaHYDRCNCYJkyAiSDR+QZuSaG/wjHTReIFmEkQE8E85HI1f0KakoqLC6hCMevIfc9F67JkEEQGivSdFU3F00JURMUyCMMKevydFtA73byqifeK3SBatF2cmQUQAf4Lw/zYiR/UTi0kQkad6L6ZoTBImQUQAf2IwvWAiT/UlK6N9dbJIVP2iLBqPP5MgIoD/ixmNX9BIV/0EY0qAkad6z7No7GRgEkQE8BdtTRtE5KleLRGNVRSRrrS09Oh2NFYRmgQRQaK1J0Ukq57UTYKIPIcPHz66XVRUZGEk1jAJwjCCqHq1oKkijDy5+blogjexFxYWWhxN6JkEEQFMySFyRXsddiTzeDzk5eahad4EsX//fosjCj2TICKAqZqIXEeOHKlx2wh/+/fvp6qyCjJAYoVt27ZZHVLImQQRAfz12KaROvIcOnSoxm0j/G3atAkAbaF4WnjYuGmjxRGFnkkQEcAkiMhVUFBwdDsa67Aj2ffff4/ECrQATVNytuZEXSnQJIgI4G/cNP3oI48/KXROcZsEEUFUlSXfLcGT5gEbaBvvSOply5ZZHVpImQQRAfwjcM100ZHn4MGDJMQI6fFuCvJrXFrdCEMbN26kIL8Abe9r/2sFkiB89dVX1gYWYiZBRAB/sba4uNjiSIz6OnjwIC3ilOZxHg4ePGh1OEYdLVq0yFtyaOdLEALudm6WfLckqi7UTIKIAIW+E0tBoTnBRJr8vDxaxDhpGasUl5RG1cklkn216CtIB2KP7dP2SlVlFcuXL7csrlCzNEGISEcR+VJENorIehG527e/lYjMFZEtvt8trYzTSh6Ph7179gJQWJBvTjARZt++PWQkeMhI9E7aF4196SNNXl4ee/fsxdPmuE4h6SAOYeXKldYEZgGrSxAu4DeqehpwFnC7iPQEHgTmq2oWMN93Oyrt2bMHp7MKV/MOAOTk5FgckVFXRUVFHDx0mLaJbtomek8227dvtzgqozbZ2dkARwfIHWUDTysPa7PXWhCVNSxNEKq6X1W/920XAxuB9sAlwDu+h70DjLUmQuutW7cOAGfb3j+6bYS/jRu9/eZzDjtYtC8Whw02bNhgcVRGbfbt2+fdaP7T+7SZsnfv3tAGZCGrSxBHiUgX4AxgKdBaVfeDN4kAGSd4ziQRWSEiK/Lz80MVakitWbMGiYnH06wdxDdjzZo1Vodk1NGKFSuIscFhp7C7xE5mcxcrlkdXN8lIVFhY6B3/YK/hzngoKy2LmmlTwiJBiEgy8Cnwa1Wt80gUVX1DVQeq6sD09PTgBWgRVWX5ipU4k9uACM6UNqz8/vsfLUJjhCdV5dtvFnNaS+fRg6xfahU527Zz4MABS2MzTq6qqgpxnGD+M1/SiJYxSZYnCBGJwZsc3lPVf/t254pIW9/9bYGo7EB+4MABCvLzcDdrC4AnpS1lpaWmHjsC/PDDD+zdt59BGcdWkRuU4T2pLFiwwKqwjDpwOp1Hz4yyWpDV1ZKFrdpjooDVvZgEeBPYqKp/qnbXNGCib3si8HmoYwsH/sYyT0obANzNvL9NNVP4mzdvHg4bDM44diJpneghs7mbuXPmWBiZUZuysjLU4W2gliJBiqoliJhjj4kGVpcg/ge4FhgmIqt9PxcAzwIjRWQLMNJ3O+qsW7cOccTiSfT28tXYZCQuifXr11scmXEyqsrCLxfQu1UVSTE/7glzZkYlOdu2sXv3bouiM2qTn5+PJ6bmec80zvt5Vp9jqymzuhfTYlUVVe2jqv18PzNVtVBVh6tqlu93VI4QW5udjSspjdhdS4nducTbDpGYztps05MpnO3atYu8/AIGpP+0GmKgr0QRTYOtIonL5SInJwdteYIp9lt4f23evDl0QVnI6hKEcQLl5eXs2L4dd1IGttJCbKXeid7cKRnk5R4wU0eHsa1btwJwSrOfdiZIi/eQHCtmPEuYWrNmDU6nE009QYKIB0kWli5dGtrALGISRJjaunUrHo8HT/KPe/h6kry3/XPVG+HHP2trWvxPqylEIC3eHTVVFJFm7ty5SIxA2xM/xt3BzcqVK6PiIs0kiDC1Y8cOgKPtD37+2/77jfDj74Zsk5qvQm2iZm2PMFRcXMy8+fNwt3fXPAbCRzt7P7/p06eHLjiLmAQRpvbs2QM2Oxqb/OM7HHFITHxUjeaMNP7p2WNOcHQ5RI8+xggfM2fOpKqyCs2qZYnfZkBr+PTfnx5dq6WpMgkiTB0+fBiJSfDWSRxHYxI4fPiwBVEZdVFcXIzdBo4THF3xdqU4ylYmC3cul4upH0/1zuDaovbHuzPdHCw82OTXhzAJIkwdOXIEdcTWeJ/bHmsSRBhbtvQ7spqf+Moyq7mLnG3bTDtEGFmyZAn5efm4M+s4S0FbkBTh448/Dm5gFjMJIkwdPnwYty2uxvvUEU9RkUkQ4WjZsmXs2LmLwRknrkIanFGFqjb5k0skmfbfaUiiQLs6PkHA3dXNhg0bmvTMBiZBhKn8gkI0Jr7G+9QRb9Y3DkPFxcU8+8zTtEtWhrQ78bod7ZM9DGlXyUcffnh0tLxhnYKCApYtXYa7s7teZ0TtrIhNmDVrVvCCs5hJEGHI6XSSl3sAT3wN8w0DGt+c4uIjZgnSMOJyuXj00d9z6OBBbu1ZTOxJesEAXNW9jNQE5feP/I68vKicaixsfPXVV6gq2qmWxunjxYGntYd58+ehWs/nRgiTIMLQ0TEQiTUvpOffb8ZChI+XXnqJFStWcsNppXStYYDc8RIdcG+fw5QdKeLBB+6nvLw8BFEaNfnyyy+R5uLtnVRP2kEpyC84uvZHU2MSRBhasWIFAO5mNVeIulPagM129HGGtaZPn860adO4sHMFP29X9+6rHZM93NHrCDk52/jDH/7QZK9Cw9nhw4fJzs7G3a5hU+hrOwWBb775JsCRhQeTIMLQ/AVfosnpEJNQ8wPsMbhT2rLgy4XmpGKxrVu38uKLf6JXKxdXZta/FNA3zcW4buXMmzePzz+PykmLLXW0eql9A4+jWCAd5s6b2ySPRZMgwkxOTg7bcrZSlZp50se5UjPJPbCftWujZ33ccFNVVcXkJx4nyebif3uVYDvBGjO1uahLBX1Tnfz5tVfNLK8hpKrMnDXTW71Uh7EPJ+Lp5OHA/gNNssOBSRBhZvr06YjNjiu120kf52rVBXHEMmPGjBBFZhzvr3/9K9t37OSm04ppFtvwq0ebwE09S3Hg5PHHH6Oy8sQ9oIzAWbp0KRvWb8B9ihsamNzB2w4h8cLf//H3JleKMAkijFRWVvLF7Nk4W3aGE3RxPcoeQ1WrrsxfsICSkpLQBGgcNW/ePD755BPO71hBv7TGT7fQMk6ZdGoJmzdv4aWXXgpAhMbJlJSU8PIrLyMpgnZr5Ek9BtynuVmzeg1zmthiUCZBhJGlS5dSWlKCKy2rTo93pXXHWVXF119/HeTIjOq+/vprnnnmabq3cDMhK3C9jwZkOLm4SzkzZszgr3/9a5O7Gg0X3i7Jj7J3315c/V0BOQtqV4V0eO7555pUVZNJEGFk3rx5SGwC7ubt6/R4T3I6xDdjzty5QY7M8Js1axaPPPIInRIruadP8QnnW2qocd0qGN6+kvfff58XXnihyU8GF2qVlZU89vhjLF++HE9/D2TU/pw6sYH7bDfueDf3P3A/q1evDtALW8skiDBRXFzM4m++oarlKSB1/FhEqGrVle9XrjTz+gRZRUUFzz33HM888wyntqjiwTOOkNKIdocTsQlcf2oZF3Up57///S933H47+/fvD/j7RKNDhw5x1113seirRXj6etBTAvz5xYHrPBdltjLuufce5jaBCzeTIMLE7NmzcTmdda5e8nOlZaGqUTE3vVU2b97MpJtuZOaMGVzUpZz7+xWT4Aje+4nAlZkV3NG7hO1bN3HDr65n/vz5psqpEZYtW8bE6yeyafMm3Ge70e5B+l8mgWuoC3dLN5MnT+YPf/hDRA+CtDxBiMgUEckTkXXV9rUSkbkissX3u+YhxU1EWVkZb7/zTzzN2uJJSqvXczWhOa6Wnfjggw/NDK8B5nK5eOutt7jllkkU5e7m/jOKuTKzIuDVSidyVmsnTw0uom1MCY8//ji///3vKSoqCs2bNxEVFRW8/PLL3HfffRz2HMY1zAUdgvymseD+mRtPDw///e9/+dUNv2LDhg1BftPgOOlXXUROCUEMbwOjj9v3IDBfVbOA+b7bTdbrr7/OkcNFVHYYVOP6D7Wp6jCI8opyXnzxRXOVGSA7d+7k1lsm8dZbb3FmejnPnlVE79TQtwdkJHh4ZMARrsgs45uvv+K6a69h8eLFIY8jEq1cuZLrJl7Hp59+iifLg2u4q1HjHerFBtpHcf/czf6i/dx222289tprEVeaqO1a6BMAEZkfrABUdRFw8LjdlwDv+LbfAcYG6/2t9p///Idp06ZR1bYPnpSGtZhpYkuq2g9gwYIFvPfeewGOMLqoKp9//jk33XgD+3du5e4+JfxvrzKSY6xLvHYbXNylksmDD9PcU8TDDz/MCy+8QEVFhWUxhbPS0lKee+457rnnHnJLcnEPcaP99KTLiAZNBrhGunCf4mbq1KlcN/E6Vq1aZUEgDVNbTapNRB4FuovIvcffqap/Ck5YtFbV/b732C8iNZ45RWQSMAmgU6dOQQoleD7++GNefe013C064uw4sFGv5WzXF1v5Qd544w3cbjfXXXcd0oDSSDQrLS3l2Wee4atFi+id6mJSzxJaxoVPiaxjsofHBh3mk5wE/jttGmtWr+LJp56mc+fOVocWNrKzs3li8hPk5ubi6eFBT7coMVQXAzpAcXdyk7cyj1//+teMHz+em266iZiYGIuDO7naShDjgQq8iSSlhh9LqeobqjpQVQemp6dbHU6duVwuXnnlFV599VVcLTpRkTm87j2XTkSEyq4/x5mWyZtvvsnzzz9vRuTWw7Zt27j5phv5+utFXJVVxv/1Kw5Icnj3hwR2FtvZWWznyRXJvPvDCebXqqMYG0zIKueBM4o5lLubSTffxIIFCxodZ6RTVf71r39xxx13kFeS5y019GlccpDVAkVAEdgW2ry3GyMdXCO8pYkPPviAW269hQMHDjTuNYOsthLEaFV9TkTiVPWJkETklSsibX2lh7ZAk5kwf8eOHUye/CRbtmzG2aYXVZ0GNz45+NnsVHX9ORqXwowZM1i3fj2/f+QRsrLq1zMqmqgqs2bN4sU//ZEEm5OH+xdzasvAtTXsLLZT7vZ+vpuKAte63SvVxZODinh1XQqPPfYYa9as4fbbbyc2tuZlapuyyspKnn/+eebOnYunowcdoBCAC3MpEsTpSwr53l9KIy8aHL7SRBs3OStyuHnSzTz7zLOcfvrpjXvdIKntG/sr3+9QtwFMAyb6ticCET/Npcvl4sMPP+SGG29k645dVGQNp6rzWYFLDn4iODsMoKLH+ezan8/NkybxzjvvUFVV92moo0VpaSlPP/00zz77LJnJ5Tw5qCigySHYWsUrv+1/hAs6VfDZZ5/xv7fdGnWT/eXm5nLnXXd6k0MvD3pmYJJD0LX3doc94jrCnXfdGbbd1Gs7O20UkR1ADxFZW+0nW0QCMo2oiHwALPG9xx4RuRF4FhgpIluAkb7bEWvt2rXccOONvP7661Qmt6W012W4WwW3g5i7RUdKel1KVYvOvPnmm0y8/nqzfkQ1S5Ys4bprr2HO7Nlc1tVbZdMijNob6sphg6u6l3NP3xL27szhV9dfz/vvvx8VI7CXLl3Kr274FT9s/cE7tuE0bdSkeyHXDFzDXLhauXj++ed5+umnw67jgdTWLVJE2gCzgYuPv09VdwYprnobOHCghtsJsKCggL/97W/Mnj0b4lOo6Hgm7pad692VNX6D9+qioucvGhSHvWg38buWQPkRhgwZwm233Ubbtm0b9FqRLi8vj7/85S/Mnz+fDsnKTacVk9m8YYvF1MWTK5LZVHTskvbUFk5+NzA4kyseqhTe3pTIyvxYunfP4p577g3bqovG2LZtG1OmTGHRokVIC8F1lisoLaK2hTYk/9ixqumKZ4gn8G+kIBsE2wYbaelpXD/xesaMGROyBmwRWamqNfaSqTVBRIpwShDl5eV89NFH/Ou993A6XVS26YWzXT+wN+wDb2yCAMDjImZ/NnH712C3CVdecQXXXHMNSUlJDX/NCFJeXs4HH3zAB++/h9vl5KLO5VxySvAHvYUyQQCowvK8GP65OZmiShg+fDi33HILbdq0Cdp7hkpOTg7/fPefLPxyITjAneVGe2jtLakNFLIE4ZcP9mw7FEJG6wyun3g9I0eOJC4uLnjvSSMShIhMVdUrRCQbftQ6I4Cqap/Ahtpw4ZAg3G43c+bM4W9vvMHBwkJcrU6hquMgNL4Bi91WE5AE4SOVpcTsWU5MwVaaNW/BzTfdyIUXXojDEcS5IyzkcrmYOXMmb015k8KDhzirdRVXZJaTkRDEA72aUCcIv3IXTN8Zz8xdiYjNwS/HjePqq6+mWbPGfRdDLT8/n3nz5jF7zmy25WxDHII705cYgtweH/IEAd6z7AGwb7DDQYhPiGfY0GGMHDmSfv36YbcHvs9uYxKEvydRjR2tTRXTMcuWLePPr7/O9m3b0OR0KjqdiSclMFdtgUwQfraSfOJ2L8V25ADtO3Tkf2+7lXPPPbfJjJ3weDwsWLCAN//xd/bu209WCzcTMkvp3iJ41Uk1sSpB+BVWCB9vTeCbA3EkJCYwYcJVXH755SQmJoYshvo6cuQIixcvZs6cOaxatco7O0Aq3h5KnRSCe0F9lCUJ4uibAfkgOwX7XjvqVFqltmLUyFEMGzaMHj16BOxYNVVMQbRv3z7+9OKLLFu6FOKbUdFhAO5WXRs0ZUZNYncuwZG/GQBPYiqepFSqOp8dkNdGFfuhXcTvWQ7lRfTt14/7fvObiB94tXbtWl555WU2b95CpxQPl3cto1+aM1AfSb1YnSD8dpfY+CQnkZX5MbRs0ZxJt9zKmDFjsNksn44NgKKiIhYvXsyXX37Jyu9X4nF7kBTB3dHtTQoWjLqyNEFU5wbZJ8hOQXIFPN4qqGFDhzFkyBBOO+20RiWLxpQgiuHEHX9VNWzKq6FOEC6Xi6lTp/LmlCm4PEpFu/64WvcEW2CLgPEbpmMvPjaYxp3SJqAlCQDUgyNvE/F7V2JTN9decw3XXHNNxPWpz83N5S9/eZ0FC76kVQJc0bWUc9pUNXit6EAIlwTht/Wwnfe2JLGlyE737lncffev6d27tyWxuN1uvv76az7//HNWrVqFx+NBkgV3ezfaQaEllvZKCpsEUV0VyF5B9gi2PBvqUdIy0hg1YhTjxo0jLa1+k31CAEoQIvIEcAB4F+9HdjWQoqrP1zuaIAllgigrK+M3993H+nXrcLXsTFXnc9C44DT2hiRB+DnLiNu5FEdhDt0yM3n5pZcips56x44d3H3XnZQWH+HCTmVc2KWCeKunWCD8EgR4G7KX5MbwYU4yRZU2HnnkEYYPHx6y96+qqmLOnDn8671/sW/vPm9S6OBLCi0Im66qYZkgqqvylSz2CHJAsNvtjD5/NFdddRUdO3as88ucLEHUtWXyfFU9s9rtv4jIUiBsEkSolJeX83//dz/r16+notsQ3GmZVocUODGJVGYOxZXalW1bF3DPPffy0ksvkpJi+awqJ7V7927uvutOtOIwkwcfpn1SGB3EYUgEzmnj5Iy0Q/xxTTMmP/EENpuNoUOHBv29d+/ezR133cGhwkNIK8F9thvaEzZJIaLEgnZRtItCCXg2e5j5xUxmzpzJtddey0033dTot6hrBaRbRK4WEbuI2ETkaiC0rX1h4s033yR7XTYV3YY2reRQjbtlZ8qzRrA1Zyt//vOfrQ6nVl999RWHig5zx+nFYZccyl1CQkIC48aNIyEhgXJX+JwJExxwX98jxDuUadNCM1nBokWLOFR4CPd57mNrM4TPv+THnPzos8NpdUAnkQzaX3Fd4MKT7uHTzz4NyNT/dU0QVwFXALm+n8t9+6JOdvY6PCltcKd2tTqUoHK36IizeUeys9fV/mCLnXvuuYC3/78rvPIDZS7hwgsv5K677uLCCy+kLIwSBED2wRjKnHDeeT8Lyfvt3bsXBKTc29ga1pz86LML6wThVwl4oLS4lOLi4ka/XJ2qmFR1B941GmokIg+p6jONjiYC5OXn4ZEQ9bOzmtjIzw/v2SYBunTpwtAhQ5i9cCGrD8ZxRddSBmdY02vpeIkOZcaMGQDMmDGDDEd49BrcfsTOR1sTWXfQQft2bRkxYkRI3nfMmDFs2LiBbSu2Iet9Yxq6Bn9MQ4PE8KPPLlTda+vN1yXW9oMNOSDExcdx2YTLSE5ObvRLB2p01OVAVCSIsZdcwptvvomraA/uFsFeu9A6tuJcHAe3M3b8eKtDqZPHHn+c0d99x19e/zOvZu+ic4qHn7Wt4Kw2VTSPte6knOBQykvK+eSTT7y3W1gXS5Ubvi+IYfH+OFYXxNAsJZk77riesWPHhqzHWu/evXlrylusWLGC995/j+9Xfg/rgFbgSfegrb1jHixfwwEgBsqLjn12NP58GxgKlILkCeSCo8CBp8JDsxbNuOKmKxg7dmzAOpcEKkGEwbVaaEyYMIEvZs9m39b5VHQ+G1daVsDGPIQLe+E2EnYsplVaGtdff73V4dSJiHD22WczePBgZs+ezaeffMy7m3N4b0sifVKdnNumkv7pTmLD4cQTQh6FzUUOFu+PZWl+POVOJbVVSyZOvJgrr7wyIFeZ9SUiDBo0iEGDBrF582YWLVrEihUr2LRpE55NHsQueFI9aIaiGb7uruExXMM65b6EkAf2fDta6r3QaNmqJYN/PpiBAwcyZMiQgE/LEagEER7l5hCIjY3l1Vde4fEnnmDN6kXYD++jsss54AjHMnI9uZ3E7lpKTN4mepx6Go899mhYj7itid1u54ILLuCCCy5g+/btzJ49m7lzZvPaukISYoSBaRWc06aKni1d2JvwSWdXsZ1vD8TyXV48BeUQHxfHz4cN4fzzz+eMM84IypQNDdG9e3e6d+/OTTfdRGlpKWvWrGHlypUsX7GcHet2ACAOX8JIUzRdoRXhUcIIFn8JIV+gAOwFdrTEe4pNSk5iwIAB9O/fnwEDBtCpU6egzn4QkJHUIrJKVc8IQDwNFuqBcm63m3/9619MmTIFYhMp7/w/uFsGftnTUI2DsB3eR8KOxVBxhAkTJnDzzTc3mfmZ3G43q1evZu7cuSz88kvKystpHgdnZlRwXtsqTmnWNGZzLawQvtkfx7d58ewpFmw2G4MHD2LEiJGcd9553p44EeTQoUOsWbOGNWvWsGrVKrZv346qIjZBWymeNA/azpcwgnCODOk4iErvADjywFHowFPmfZ/klGT69e1Hvz3us3gAACAASURBVH796Nu3L5mZmQFP7kGfakNEHlbVpxv9Qo1g1VQbGzdu5Omnn2Hnzh0407tTdcp5Aa1yCkWCiN35HTEH1tG2XTseevBB+vXrF9DXDyeVlZV89913zJs3l2+//Ran08WpLV1c1LmcPqmugNcWhiJB7C6xMWNnPEsOxOFW6N3rdEaMHMXQoUNp0aJFQN/LSsXFxWRnZ7NmzRq+X/U9WzZv8Y6+Tqo20C6Ao6+DniAqfQPddou3+ki9VUZn9DvjaELo3Llz0KdDCcRI6u7AX4DWqtpLRPoAF6vqk4ENteGsnKyvqqqKP/3pT8ycOZOyfleicYEbWBaf/W+StZwLL7yQGTNmUCIJVPS+LGCvj6uSpJXvMnToUB566CHi4+MD99phrri4mFmzZvHRhx+QX1BIR9+8Tf3TA9efMZgJYm+JjQ+2JrK6IIb4uFh+cdHFjBs3jnbt2gXk9cNdcXEx33zzDQsWLGD58uW43W7vqOxMN5oVgDEAwUoQxWBbbfMmBQ+0aduGEcNHMHToUDIzM0M+YWYgRlL/Hfg/4G8AqrpWRN4HwiZBWOnQoUPHBqUEeAlRcVVx4cXevtgAU6d9EdDX98erqhQWFtK+ffvAvn4YS0lJ4YorruDSSy9l/vz5vPevd3lp7W7uP6OYXq3Ce0W2okrhudXNcTqSuPHG8YwdO5bmzZtbHVZIpaSkMHr0aEaPHk1xcTGLFy9m5syZrFm9Bo960O5h2DRaBo6vHSTZkrh4/MUMHTqU7t27h+0synVNEImquuy4PyK8j6Ag27dvHwsXLuTLhQv5YdMmADSlNeoIbD2vOmJ/1Bc70K+PzYGneXsWLlzIwoUL6ZaZybChQ/n5z39Op06Bb1MJRzExMYwePZqf/exn3HrLJF5bt4tnziyiZZguQaoKr65LoUxjef2VV8nMbJoj+usjJSWFMWPGMGrUKB577DG++uor3M3cEE7rJCk4vnUQ54njpZdfonv37lZHVKu6JogCEemGr7eSiIwD9gctqjBUUFBAdnY22dnZfP/9KrZtywFAk9NxdhyEq1UXND4IV3D2WMqLDx7ri50S4PcQofzUMUhlMfaDO9iSu4Ocv/+dv//973Tu3IX+/c+gd+/e9O7dm9atWwf2vcNMYmIi/c7oz+c7d1HmkvBNEEBBhZ02bdvSpUsXq8MJKx6Ph+KSYm87RCP7WGgLhSLfjRa+242kDsXlclFWVtbo1wqFuv4LbwfeAE4Vkb3AduCaoEUFiMho4GW8Hdr+oarPBvP9qnO73Wzfvv1oQlizdi35eXneuOwO3EnpODudibtVl4C2N1hJ41Jwte2Nq21vpLIU+6EdbCvaxa5p0/nss88ASE1No2/fPvTu3ZtevXrRrVu3JtPTCeCHH35g2uefM6xDRdjN6VSdTeDqzBJeyd7Jv//9b6644gqrQwoLbrebJ596ku9Xfo9nkAfqP/P1j2g/RYq8tSYBaXsQcJ/tRr4SHnjwAV595dWwL0XUdaqNbcAIEUkCbKra+Ek+TkJE7MCfgZHAHmC5iExT1Q3BfF+Px8NHH33E22+/Q3m5N8NLXBLOpAzcnc7Ck9IaT2IqhMkiK8GicUm42pyOq83poB5sZQexFeeSW5LLl98uY8GCBQDExcVz1VUTuO6668KmX31DlZeX88Tjj9E8Trm8W4XV4dRqUIaTM9Kc/O1vf2XAgAF069bN6pAs5fF4eP755/lywZd4+ni8M5yGozhwneuiYmEF99x7D39+7c9hXQqsU4IQkRbAdUAXwOFvi1DVu4IU12Bgqy8xISIf4p0LKmgJ4vDhwzz19NN8t2SJd6K6doPxpLRGY5Ob3EjpehEbnqQ0PElpuDidSkAqS7CV5OIq3M5bb73FqlWrefTR35Oammp1tA32n//8h9179nJ3nxKSYwJ3cumc4mZnsf3odueUwIy5EIGbTivl7m9i+MtfXueFF/4YkNeNVEuXLmXWrFl4unq861WHs0RwneeieHYxL/zxBV579TWrIzqhul4Kz8SbHLKBldV+gqU9sLva7T2+fT8iIpNEZIWIrMjPz2/UGz7w4IN8t2QJnvjmVLU/A3dqN2/1UTQnhxPQuGTcrbribN8Pd2Iqq1ev4u67f211WI0yePBgYhx2Fu2LwxPA88u1PcqPJobfDSzh2h7lAXvtbw/E4vKEbibWcNa7d2/atG2D/YAdwr0AqN61plEYM3qM1dGcVF0rkONV9d6gRvJjNZ2Vf3LYquobeNtGGDhwYKMO6/FXXsl//vMfVq9ejW39NCQuiarmnXC16ownpW3AlxKNSB43tpJcHAd3EnN4F1QUIyL07tuXi34RpFXuQqRbt27ccuttvPbaazy8tAWXnVLKwAynpcuVnsj2I3Y+3ZbA6oIYzjnnbC6++GKrQ7JccnIyk5+YzG233QazwN3ZNxYinJoI3SC7BPsWO3pYOf/887ngggusjuqk6pog3hWRm4HpeGccB0BVDwYlKm+JofqaeR2AfUF6LwCGDBnCkCFDKC4uZsmSJSxatIjvli6lKm8j2Ox4ElNxJ2fgTs7Ak9w6aEuMhhOpKsNWkoutJA9HSR720kLU4yImJpbBgwdx3nnncc455zSZ0bqXX345qampTHnzH7ySvZfOKR4u71ZK3yCMsG6IvaU2PtmawPL8WFKSk5g06WrGjRsXtn3oQ61Hjx5MmTKFDz74gDlz5+DOcaPtFE93X4O1Vf+mSpBtgj3HjpYrnU/pzIT/ncDIkSPD/rOr60jq24Gn8Hb68j9BVTUoq+aIiAPYDAwH9gLLgatUdf2JnhOMkdSVlZWsXLmSNWvWkL1uHT9s+gGns8obo7/xOjkDT0obPElpQamOCtma1Kq+xuj92EvyiCnNRyu8fRHsDgc9unenV69e9OnTh4EDB0bcJH714XK5mDdvHm9PeZN9B3I5vZWLCZlldGngnE1PrvDOmNrQEdSHK4VPtyWwcF8c8fHxXHHleK644gpLZmKNFIWFhXz22Wf8+7N/U1JcAqng7l7/5U1tC7218A3qxVQCskWw77CjLmXQoEGMHz+egQMHhlViCMRUGznAmapaEOjgTvKeFwAv4e3mOkVVnzrZ40Mx1YbT6WTr1q2sX7+e9evXszY7+2j3VxKaU5WahSstE40L3IEb7AQhVWXYC7cSV7AFyg4B3u6svXv3olevXpx++ulkZWWFbL2AcOJ0Opk2bRpvTXmT4uISftaukok9yuo9ZXhjEsTc3XF8lJOEU4WxYy9l4sSJTabEFgoVFRV88cUXvP/B+xzYf8A7FcfpbrSj1ilRNChBlIGsFWx7bNhtdkaOHMn48ePp2jU8V6EMRIKYBoxX1bAd3WHVXEwFBQUsW7aMmbNmsXbNGgDczdvjTO+Bu9UpjS5VBCtB2A/tJCZvE/bDe0CVnj17MmbMGM4++2wyMjIa/fpNSXFxMe+++y4ffvghp7dycU+fYuLrMfyjIQlCFabmxPPfHQmceeaZ3HXXXXTs2LH2Jxo1crvdLF68mH+++0+2bN6Ctlc8Azy1rhJXrwShvjaG1XZiJIbLx13OL3/5S9LSGjkgI8gCkSA+A04HvuTHbRDB6uZab1ZO1ue3b98+vvjiC2bOmkVebi6Vnc/C1aZXo14zGAnCkb+ZuG2LSE1LY4xvLptomVajMb744gueffZZMps5ebj/ERx17APYkATx0VZvcrjooou49957I36cSbhwu91MnTqVN/7+BupQnOc4vSvYnUCdE4QHbMtsyG6hV+9e/Pbh30bMvGYnSxB17eb6H7xtEN8Smm6uEaldu3bccMMNfPThh5x77rnE7VqKvWh37U88CU9SKmqPQe0xuFPa4Elq3FgDW/EB4nYspn///nw8dSqTJk0yyaGORo8eze9+9zs2F9n5dFvwZr3NLnTw3x0JXHjhhdx3330mOQSQ3W5nwoQJvPmPN2ndqjWOrx3Q2IpzN9iWeJPDzTffzKuvvBoxyaE2dUoQqvpOTT/BDi5S2e12HnnkETp16kT87mXe+oIGqup8Np7EVDyJqVT0/AVVnc9uVGxxu5eTkZ7O5MmTm9Q0GaEyYsQIfvGLXzB9RwLrCgP//yuqFP66IYUunTvx61//OqwaM5uSrl278ufX/ky71u28SeJQA19IwfadDdkn3HvvvVx77bVNKqGfNEGIyFTf72wRWXvcz5rQhBiZEhISuHzcOCg7hK2s0OpwAJCKw9iKc7ns0ktJSQmnDuKR5c4776Rzp468viGFQ5WBO4F7FF5fl0yFxvDY408EfH1h48fS0tJ49ZVXSWuVhuNbBzRgDKNkC7JPuPPOOxk7dmzgg7RYbSWIu32/NwIXVfu5GPghiHE1CcOGDSMhMYm4nd81qhQREKrE7fyOmNhYRo0aZW0sES4hIYEnnnyKKmJ5eW0KVQFasfTDrQlsOOTgnnt/E7Y9XpqatLQ0/vD8H4jTOOxL7FCf3qy7wfaDjbFjxzJu3LigxWilkyYIVfVP6Z2pqjur/ewATg16dBEuJSWFX999F7biA8Tss7bA5cjbiL1oN7fdemvY96qIBF26dOHh3/6OrYftvLkxsdH5/6t9sczcGc+ll14a9qNrm5quXbvywP0PQCHIpjqWCMvBscpBj1N7cNdddzXZqsDaqphuE5FsoMdx1UvbgbWhCTGyjR49miFDhxK7ZwWO/essicGR9wNxO5YwaPBgLrssgMuVRrkhQ4Zw44038s2BOKbtaHij9aZDDqZsSmLAgP7ceeedAYzQqKvhw4czfPhwbBttUIfOZrJGcKiD3z/y+ybdllfbX/Y+MAt4Bniw2v7iIE6z0aSICL9/5BHU4+Grr75CPE6c7fqFZhJAVRy564nb+R2DBg/m6aeeCvoC6NHmuuuuY9eunXw8dx5tE90Mbl2/9axzy2y8lN2M9u3b88QTpuOAlW6//Xa+Xvw1FdkV6NknKRIeBNtuG1dNvKrJj02prYrpsKruUNUJx1UxmeRQDw6Hg0cffZTzzz+f2D0rictZCJ4gr9jq8RC74xvidn7HeeedxzNPP20aPYNARLj//gfoedpp/G1DCruK696DpdwFL65thsQm8uxzz5uOAxZLS0tj/JXjse2xwZETP862wUZySjITJkwIXXAWMZeTIeJwOHj44Ye5+eabcRTmkLhpFjiDNC+xq4r4zbOJydvENddcw+TJk6NyqoxQiYuL48mnniKlRUtezG5GibP20qEqvLE+iX1ldp6Y/CQdOnQIQaRGbcaNG0dMbAzywwk+wyMg+4XLx13epOcj8zMJIoREhGuvvZYnnniCmIqDJG2cjlQGdnE+qSolcdMMYkoO8NBDDzFp0iRTrRQCaWlpPPnU0xyqtPO39Um1rikxa1ccy/NjufXWWxkwYEBogjRq1aJFCy4YcwH23Xao+un9kiM4YhxceumloQ/OAubMYYEhQ4bw4p/+RKLN6U0SVQGa4spVSeKmmcS7Snn+uecYMya8FyNpanr27Mntd9zBqoIY5u0+cXXezmI7H21N5LzzzuPKK68MYYRGXVxyySWoW72L+lTnBvsuO0OHDI2aCRNNgrBI3759eeXll3Gok/icL8HTyEXRVYnPWYi9qoQ//vEFBg8eHJhAjXq57LLLGDx4EFO3JVFY8dNqCo/CmxuTadasGQ888ECT7R4ZyTIzM+mW2Q3bruNOj/tBqzSquiGbBGGh7t2788D992M7sh9H3sZGvZajYCv2ot3ccccd9OnTJ0ARGvUlIvzmN/ehNgef5iT85P4lB2LZdsTGnXf/mmbNmlkQoVEXI4aPgINAtUGQslto1qIZffv2tSyuUDMJwmKjRo0iK6s7sQVbGvU6joLNtG3XzoxzCANt27Zl9JgLWJIbx5GqH5cQ5uxJoGOH9gwbNsyi6Iy6OPfcc70b/nYID9jz7Jx7zrlR1RXZJIgwcP75o5DSAqTiJH3rTsZZgf3Ifs4fNcpUWYSJiy++GKcHVuXHHN2XX24j57CNiy8ZazoOhLlOnTqRlp52LEEUeauXBg0aZGlcoWa+pWHAX2S1leY36Pl23/P69esXsJiMxsnMzKRZSjKbio5dbf7g2za9lsKfiNC7V29EBW2hyCHvhVfPnj0tjiy0TIIIA127diUmNhZ7cW6Dnm8rPoCI0L179wBHZjSUiNDj1NPYU3qsBLGnxE6Mw84pp5xiYWRGXXXr1g2coL0UDkNCYgJt2rSxOqyQsixBiMjlIrJeRDwiMvC4+x4Ska0i8oOInG9VjKESExPD4EGDiS3a1aBZX2OLdtGnb1+ziH2YadeuHfkVx0ZW55XbaNO6dZNaL6ApS09P925UgFQI6enpUVeFa2UJYh1wGbCo+k4R6QmMx7vE6WjgdRFp8kfUsGFD0coS7xrR9WArPgBlhxg2dGiQIjMaKj09nZIqxZ/yi6pspLeOrivQSNa8eXPvRpX3p2WLlpbGYwXLEoSqblTVmtaUuAT4UFUrVXU7sBVo8p36hw4dSnpGa2L3rapXKSJ232qaNWtuBsWFIf9gKrfHe9V5xOmImgFWTcHR0oJ6f6KxY0E4/sXtgeoLOe/x7fsJEZkkIitEZEV+fsMaeMOFw+Fg4nXXYivOw35oZ52eYzu8D3vRHq6++iri44O3RrLRMP7J99y+fF/uEjMhXwRxOn0z89oAO5RXNGDJuQgX1AQhIvNEZF0NP5ec7Gk17KvxklpV31DVgao68Gh9YQS74IIL6NSpM/F7loGnlmXK1EP87qWkZ7Q2Yx/ClH8yN/+8TGUujYoJ3pqKoqIi70YcaKweux1FgpogVHWEqvaq4efzkzxtD1B9kvUOwL5gxhkuHA4Hd9xxO5QfwZG36eSPLchBSgu57dZbzDTeYcpfqvNPouJ0Y0p6ESQ3N9d7uRoPJEBBQQGexk6JE2HCsYppGjBeROJE5BQgC1hmcUwhc+aZZ9K3Xz/i968G9wkWn/F4iNv3PZlZWWZEbhjzJ4PUeA/tk9w/2meEv/3792NLsnnPksngcrqI9Krs+rKym+ulIrIHOBuYISKzAVR1PTAV2AB8AdyuqgFaFj78iQg33XgjWlWOo2BrjY+xH9wOFcXcdOONUdlwFin87Q2DM5xcfErFj/YZ4S8vPw93gvfUowneesLCwkIrQwo5K3sxfaaqHVQ1TlVbq+r51e57SlW7qWoPVZ1lVYxW6dOnD1lZ3YnLW19jj6a4vPW0b9+Bs846y4LojLryT8ZX4hSKq7yHmkkQkePgoYPgX2fLV4sbbe0Q5vIzDIkIl112KZQV/WT6DSk/hBTnMXbsJab0EOaSkpKIjYmhqNJGkW/SvtTUVIujMuoqKSHp2Gyuvt8JCT+dobcpM2eYMDVkyBBiYmNx5P94lldHwVZsNhsjRoywKDKjrkSEtLRWHKy0cajSe6g1hd520SI1NRVbufdzk1Jvgm/VqpWVIYWcSRBhKikpif855xxii3Yc26lK7MHt9O8/wFyJRoiMjDYcrLRxsMKGiJjPLYL07t0bPaxQAeRDSrMUOnbsWOvzmhKTIMLYsGHD0KpycHnnHLaVFULFEYYPNz2XIkVaejpFTgeHKm20bNEsqtYSiHT+WXclV7Dn2xnQf0DUVetG118bYQYPHozdbkfclQDYi7wDzM8++2wrwzLqIS0tjUPlcKjSRmpqmtXhGPXQvXt3mrVohmwUtEyj8rgzCSKMJSYm0rtPH8TjxpOUiv3IPrplZkZdPWgkS01NxemBfWV2UtNM+0MksdlsDBowCCn2tj/079/f4ohCzySIMNe7Vy/EXUVVh0E4ygro07u31SEZ9ZCW5i015JfbTPtDBMrKygJAbEJGRobF0YSeSRBhLisrC9SDvWgn6nKaRYEijD9BAFF5gol0rVu3BiDGERN1a0GASRBhr3PnzgA4DnnbHzp16mRlOEY9Ve/WWj1ZGJEh2idXNAkizLVr1w4A+2FvgujQoYOV4Rj1VL1aySSIyON2e0fIRWPpAUyCCHtxcXG0bNUKcVUSFxdvFpyJMNVn2j26QpkRMSoqvHNoaQOWAm4KTIKIAG3btAUgPSMjaq9kmoJor66IRKWlpd6NKD3sTIKIAK1bexs327ZpbXEkRmPY7U1+afUmp6yszLsRnQUIkyAigX/cgxn/ENmitZoikrlcLu+GKUEY4crf7pCcnGxxJEZjVFVVWR2C0VBRmttNgogA/lXITPtDZKusrLQ6BKOeor3dyCSICGKqKCKb03mCJWSNsOUvtUfrsWcSRATwlxyibSbJpiZaTzKRLNq7JpszTgQxJ5jIZqb6jjz+ZWOjtXrXsgQhIn8QkU0islZEPhORFtXue0hEtorIDyJy/sleJ5pE65e0qag+aM6IDNHeMcTKEsRcoJeq9gE2Aw8BiEhPYDxwOjAaeF1EoroDub/k4PF4LI7EaIxoW8+4KfAn9Wi9OLMsQajqHFX1dTLmO8A/ydAlwIeqWqmq24GtwGArYgwX/r7Y0folbSpMgog8/mrBaK3eDZc2iBuAWb7t9sDuavft8e37CRGZJCIrRGRFfn5+kEO0Tnl5OWBKEJHOVDFFrmi9OAtqq5mIzAPa1HDXb1X1c99jfgu4gPf8T6vh8TWmb1V9A3gDYODAgU02xfsnDDP96CNbbGys1SEY9RStJQe/oCYIVR1xsvtFZCLwC2C4Hvsk9gAdqz2sA7AvOBFGBn//eTMSN7LFxMRYHYJRT0en2ohSVvZiGg08AFysqmXV7poGjBeROBE5BcgCllkRY7jwj38w4yAim/n8Io//4ixaSxJWdsx+DYgD5vrq975T1VtVdb2ITAU24K16ul1V3RbGaTl/Q5npR28YoeWv3o1Wlp1xVDXzJPc9BTwVwnDCmr/3i39OJsMwQsPfQSRamTJvBDDdI5sG0wst8pgEYYQ9U3fdNER7g2ck8ncMidZurubMYxiGcQLR3khtEoRhGMYJ+EvvpgRhhC0z3XfTYHqhRZ5o/8zMGSeCRGsxt6kwCT7yRHsHEfONNQzDOAH/kqPRenFmEoRhGMYJRPt68CZBRJBo/ZJGOjPAMXJF+wSL0d0CE2GitZgb6f74xz+ye/fu2h9ohB1/I3W0XpyZBBFBovVLGul69+5N7969rQ7DaAC73buYZbSOgjdVTBHElCAMI7T8CSJamQQRQUwJwjBCywyUMyKGKUEYRmhF+9iV6P7rI0y0XsUYhlVMCcKIGKYEYRih5U8Q0XrsmQQRQaL1KsYwrBatx55JEBEkWq9iDMMqMTExAFx99dUWR2INMw7CMAzjBBwOB1988UXUTtpnWQlCRCaLyFoRWS0ic0SkXbX7HhKRrSLyg4icb1WM4SZai7mGYaXExMSoPfasrGL6g6r2UdV+wHTg9wAi0hMYD5wOjAZeF5HoHq3iY6qYDMMIJcsShKoeqXYzCfCf/S4BPlTVSlXdDmwFBoc6vnAUrVcxhmFYw9I2CBF5CrgOOAwM9e1uD3xX7WF7fPtqev4kYBJAp06dghdomDAlCMMwQimoJQgRmSci62r4uQRAVX+rqh2B94A7/E+r4aVqPDOq6huqOlBVB6anpwfnjwgjpgRhGEYoBbUEoaoj6vjQ94EZwKN4Swwdq93XAdgX4NAikilBGIYRSlb2YsqqdvNiYJNvexowXkTiROQUIAtYFur4wpEpQRiGEUpWtkE8KyI9AA+wE7gVQFXXi8hUYAPgAm5XVbd1YYYPU4IwDCOULEsQqvrLk9z3FPBUCMOJCKYEYRhGKJmpNiJAz549ARg4cKDFkRiGEU3MVBsRoFevXkyfPp1mzZpZHYphGFHElCAihEkOhmGEmkkQhmEYRo1MgjAMwzBqZBKEYRiGUSOTIAzDMIwamQRhGIZh1MgkCMMwDKNGJkEYhmEYNZKmMr+PiOTjndOpqUoDCqwOwmgw8/lFrqb+2XVW1RrXS2gyCaKpE5EVqmrm2ohQ5vOLXNH82ZkqJsMwDKNGJkEYhmEYNTIJInK8YXUARqOYzy9yRe1nZ9ogDMMwjBqZEoRhGIZRI5MgDMMwjBqZBNEIIlJy3O3rReQ1C+P51ve7i4hcVW3/QBF5xaq4IpmIuEVktYisEZHvReQc3/52IvJJLc8dIiLTQxNp0yYiKiLvVrvtEJH82v6/ItJPRC6odvsxEbkvmLEe9/4RfSyaBNGEqOo5vs0uwFXV9q9Q1bssCSrylatqP1XtCzwEPAOgqvtUdZy1oUWVUqCXiCT4bo8E9tbhef2AC2p9VPB0IYKPRZMggkRE3haRcdVul/h+DxGRr0RkqohsFpFnReRqEVkmItki0s33uItEZKmIrBKReSLS2rf/MRGZIiILRWSbiNx1/HsAzwLn+a5876l+JSsiSb7nL/e99iW+/af7YlgtImtFJCs0/6mI0gw4BEevDNf5tuNF5C3f57dKRIYe/0QRaSUi//H9b78TkT6+/ekiMtdXOvmbiOwUkTQRmSwid1d7/lPVP+soNQu40Lc9AfjAf4eIDBaRb33//29FpIeIxAJPAFf6vtdX+h7e8wTHzzXVjoG/iYjdt79ERJ4TkZW+Y3Fwtedf7HtMFxH52vc5Hi1pcvJjMbna92atiPxSROy+c8c63/57gvkPrZWqmp8G/gBuYHW1n13Aa7773gbGVXtsie/3EKAIaAvE4b0Ketx3393AS77tlhzrZXYT8Eff9mPAt77npgGFQEwN7zG92nsfvQ08DVzj224BbAaSgFeBq337Y4EEq/+/4fBT7TPeBBwGBvj2dwHW+bZ/A7zl2z7V9z2IP+7//irwqG97GLDat/0a8JBvezSgvs+1C/C9b78NyAFSrf5/WPg5lAB9gE98/9vVx/1/mwEO3/YI4FPf9vX+Y9J3u8bjBzgN+G+1Y+l14DrftgJjfNufAXN8z+lb7XNMBOJ921nACt/2yY7F5/Ad777bLYEBwNxq+1pY+X93YDRGuar2898QkeuBugzJX66q+33PycH70k2yawAABJ5JREFUhQPIBvxXnx2Aj0SkLd4T9vZqz5+hqpVApYjkAa2BPXWMeRRwcbV62HigE7AE+K2IdAD+rapb6vh6Td3Rz1hEzgb+KSK9jnvMuXgTAKq6SUR2At1reMwvfY9ZICKpItLct/9S3/4vROSQb3uHiBSKyBl4P99VqloYnD8xMqjqWhHpgrf0MPO4u5sD7/hKvor3BH4iNR0/w/GenJeLCEACkOd7fBXwhW87G6hUVaeIZONN5Pje7zUR6Yf3ouL4z78mI4Dx1f6+QyKyDegqIq8CMzh2brCESRDB48JXhSfeb1xstfsqq217qt32cOwzeRX4k6pOE5EheK98anq+m/p9jgL8UlV/OG7/RhFZircIP1tEblLVBfV43SZPVZeISBpw/MRmUoen1/QYreW5/8B7BdwGmFKXGKPANOAFvFfiqdX2Twa+VNVLfUlk4Uleo6bjR4B3VPWhGh7vVN/lPNWOV1X1iIj/2LsHyMVbqrABFXX4WwTvd+AoX5LoC5wP3A5cAdxQh9cKCtMGETw78F6RAFzCya9oatKcY41wE+v53GIg5QT3zQbu9CUtfFeoiEhXYJuqvoL3IOxTz/ds8kTkVMCOt1qiukXA1b7HdMdbIjs+AVd/zBCgQFWPAIvxngQQkVF4qxn8PsNb7TQI7+dmeBPlE6qafdz+6sfL9dX2n+xYqG4+ME5EMuBom1HnesTVHNivqh7gWrzfk9refw5wh/+GiLT0XYDYVPVT4BGgfz1iCDiTIILn78DPRWQZcCbeXhj18RjwsYh8Tf2nGl4LuMTbNfP4Rq7JeJPVWl8j62Tf/iuBdSKyGm89+j/r+Z5NVYKvgXE18BEwUVXdxz3mdcDuq3L4CLjeV4VR3WPAQBFZi7fh0p/0HwdGicj3wBhgP96TCqpaBXwJTK3hPaOSqu5R1ZdruOt54BkR+YZjJ2fw/v96HtdIXdPrbgB+B8zxfUZz8bYT1tXrwEQR+Q5v9ZL/eD/Zsfgk0NLXIL0Gb/Vye2Ch7/v2Nt6ec5YxU20YhoVEJA5wq6rL18bxl2ptHjbge+By0yZkWMG0QRiGtToBU33JoAq4GUBEegLTgc9McjCsYkoQhmEYRo1MG4RhGIZRI5MgDMMwjBqZBGEYhmHUyCQIw2ggOTbTq/+ni/hm1K3leSUn2P+j+bsMw2qmF5NhNNyPplrxOafGRxpGBDIlCMMIoOqlAxH5P/HOmrtWRB6v4bEiIq+JyAYRmQFkhDRYw6iFSRCG0XAJ1aqXPqt+h2/ajCxgMN41CQaIyM+Oe/6lQA/g/9u7Q5sIoigKoPf5bYBQAYIWEAhqQKBohhqogBrAsQaJpwAqwH/E/M2ueIGEZMiIc5IRk3wx7mZe8u+7zHL/wd8Hm2LEBH/XjZgObubzPt93WQJjf3LmKsnTrNH4rCrliGyKgIB1VJKHMcbjL+fcVGWzjJhgHc9J7qtqlyRVdX5oCj2xT3I7t4id5bgLBDbBHwSsYIzxUlUXSd5ms/pXkrscl9AkS533dZYlNB9JXv/7O+EnupgAaBkxAdASEAC0BAQALQEBQEtAANASEAC0BAQArW8/qNQ9kEQh3AAAAABJRU5ErkJggg==\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "sns.violinplot(data=full_table, y='time_diff', x='Field', );"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.8.2"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 4
}