code/simulated_data/simulate_data.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "id": "a592e39f",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "/Users/jruffle/Library/CloudStorage/OneDrive-UniversityCollegeLondon/PhD/scripts_JKR/WellcomeHackathon/WIdeathonRetention23/code/simulated_data\n"
     ]
    }
   ],
   "source": [
    "# Author: Dr James K Ruffle\n",
    "# Email: j.ruffle@ucl.ac.uk\n",
    "\n",
    "import pandas as pd\n",
    "import numpy as np\n",
    "from random_username.generate import generate_username\n",
    "import random\n",
    "import glob\n",
    "import datetime\n",
    "import names\n",
    "from faker import Faker\n",
    "from tqdm import tqdm\n",
    "import os\n",
    "import seaborn as sns\n",
    "print(os.getcwd())\n",
    "\n",
    "import warnings #suppress buggy pandas warning\n",
    "warnings.filterwarnings(\"ignore\")"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "91e1a466",
   "metadata": {},
   "source": [
    "# simulate patient data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "id": "3f268b5f",
   "metadata": {},
   "outputs": [],
   "source": [
    "sample_size = 150\n",
    "timepoints = 500\n",
    "\n",
    "max_points_per_epoch = 1150"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "id": "cd6819d9",
   "metadata": {},
   "outputs": [],
   "source": [
    "def generate_random_date(start_year, end_year):\n",
    "    start_date = datetime.date(start_year, 1, 1)\n",
    "    end_date = datetime.date(end_year, 12, 31)\n",
    "\n",
    "    days_between = (end_date - start_date).days\n",
    "    random_days = random.randint(0, days_between)\n",
    "    \n",
    "    random_date = start_date + datetime.timedelta(days=random_days)\n",
    "    return random_date"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "id": "7d5cd50a",
   "metadata": {},
   "outputs": [],
   "source": [
    "ethnic_groups = ['Asian or Asian British','Black, Black British, Caribbean or African',\n",
    "                 'Mixed or multiple ethnic groups','White','Not disclosed']\n",
    "# categories taken from https://www.ethnicity-facts-figures.service.gov.uk/style-guide/ethnic-groups"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "id": "414a9527",
   "metadata": {},
   "outputs": [],
   "source": [
    "#HADS questionnaire data is based on\n",
    "# https://www.svri.org/sites/default/files/attachments/2016-01-13/HADS.pdf"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "id": "8fd0d871",
   "metadata": {},
   "outputs": [],
   "source": [
    "# prototypic treatment data is based on\n",
    "# https://www.nice.org.uk"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "id": "1fa6183d",
   "metadata": {
    "scrolled": false
   },
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "150it [01:07,  2.21it/s]"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "(150, 8054)\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>username</th>\n",
       "      <th>icon</th>\n",
       "      <th>Trial name</th>\n",
       "      <th>First name</th>\n",
       "      <th>Last name</th>\n",
       "      <th>DOB</th>\n",
       "      <th>Address</th>\n",
       "      <th>email</th>\n",
       "      <th>Latitude</th>\n",
       "      <th>Longitude</th>\n",
       "      <th>...</th>\n",
       "      <th>Epoch points: timepoint 499</th>\n",
       "      <th>HADS-Anxiety: timepoint 499</th>\n",
       "      <th>HADS-Depression: timepoint 499</th>\n",
       "      <th>Personal health rating: timepoint 499</th>\n",
       "      <th>Total points</th>\n",
       "      <th>Total attendance points</th>\n",
       "      <th>Total log points</th>\n",
       "      <th>Total check in points</th>\n",
       "      <th>Total answered question points</th>\n",
       "      <th>Total spoke to bot points</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>enragedBass3</td>\n",
       "      <td>../../assets/icons/sharks_235052.png</td>\n",
       "      <td>MindfulAI</td>\n",
       "      <td>Shannon</td>\n",
       "      <td>Clarke</td>\n",
       "      <td>1993-01-10</td>\n",
       "      <td>792 Jones shoal\\nWest Bryanmouth\\nCW6M 0RR</td>\n",
       "      <td>Shannon_Clarke@mindfulai.com</td>\n",
       "      <td>52.66277</td>\n",
       "      <td>-2.01111</td>\n",
       "      <td>...</td>\n",
       "      <td>1070.0</td>\n",
       "      <td>20.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>91.0</td>\n",
       "      <td>142360.0</td>\n",
       "      <td>110000.0</td>\n",
       "      <td>12500.0</td>\n",
       "      <td>2280.0</td>\n",
       "      <td>4140.0</td>\n",
       "      <td>5040.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>thrilledCaviar1</td>\n",
       "      <td>../../assets/icons/bunny_1469202.png</td>\n",
       "      <td>MindfulAI</td>\n",
       "      <td>Richard</td>\n",
       "      <td>Hayden</td>\n",
       "      <td>1973-11-20</td>\n",
       "      <td>613 Wood viaduct\\nStanleyborough\\nG6U 3QH</td>\n",
       "      <td>Richard_Hayden@mindfulai.com</td>\n",
       "      <td>51.40148</td>\n",
       "      <td>-1.32471</td>\n",
       "      <td>...</td>\n",
       "      <td>1080.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>100.0</td>\n",
       "      <td>171470.0</td>\n",
       "      <td>140000.0</td>\n",
       "      <td>13450.0</td>\n",
       "      <td>2480.0</td>\n",
       "      <td>3260.0</td>\n",
       "      <td>4980.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>yearningBaboon5</td>\n",
       "      <td>../../assets/icons/lion_2298426.png</td>\n",
       "      <td>MindfulAI</td>\n",
       "      <td>Jeremy</td>\n",
       "      <td>Swarey</td>\n",
       "      <td>1996-02-19</td>\n",
       "      <td>9 Miller fall\\nLloydmouth\\nDD8B 4WF</td>\n",
       "      <td>Jeremy_Swarey@mindfulai.com</td>\n",
       "      <td>53.69786</td>\n",
       "      <td>-2.68758</td>\n",
       "      <td>...</td>\n",
       "      <td>20.0</td>\n",
       "      <td>18.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>77.0</td>\n",
       "      <td>160600.0</td>\n",
       "      <td>129000.0</td>\n",
       "      <td>13350.0</td>\n",
       "      <td>2650.0</td>\n",
       "      <td>3160.0</td>\n",
       "      <td>4840.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>pacifiedTuna1</td>\n",
       "      <td>../../assets/icons/cow_1998610.png</td>\n",
       "      <td>MindfulAI</td>\n",
       "      <td>Norman</td>\n",
       "      <td>Kearney</td>\n",
       "      <td>1988-09-17</td>\n",
       "      <td>7 Megan lock\\nKarlside\\nL3 8HY</td>\n",
       "      <td>Norman_Kearney@mindfulai.com</td>\n",
       "      <td>55.82737</td>\n",
       "      <td>-4.0573</td>\n",
       "      <td>...</td>\n",
       "      <td>1030.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>85.0</td>\n",
       "      <td>159180.0</td>\n",
       "      <td>127000.0</td>\n",
       "      <td>12200.0</td>\n",
       "      <td>2680.0</td>\n",
       "      <td>3500.0</td>\n",
       "      <td>5000.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>emptyChough6</td>\n",
       "      <td>../../assets/icons/elephant_10197317.png</td>\n",
       "      <td>MindfulAI</td>\n",
       "      <td>Donald</td>\n",
       "      <td>Hernandez</td>\n",
       "      <td>1985-04-05</td>\n",
       "      <td>32 Norris corners\\nNew Alexandraberg\\nHU01 2FE</td>\n",
       "      <td>Donald_Hernandez@mindfulai.com</td>\n",
       "      <td>52.68333</td>\n",
       "      <td>0.93333</td>\n",
       "      <td>...</td>\n",
       "      <td>1150.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>100.0</td>\n",
       "      <td>195920.0</td>\n",
       "      <td>162000.0</td>\n",
       "      <td>12150.0</td>\n",
       "      <td>2620.0</td>\n",
       "      <td>3480.0</td>\n",
       "      <td>5120.0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>5 rows × 8054 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "          username                                      icon Trial name  \\\n",
       "0     enragedBass3      ../../assets/icons/sharks_235052.png  MindfulAI   \n",
       "1  thrilledCaviar1      ../../assets/icons/bunny_1469202.png  MindfulAI   \n",
       "2  yearningBaboon5       ../../assets/icons/lion_2298426.png  MindfulAI   \n",
       "3    pacifiedTuna1        ../../assets/icons/cow_1998610.png  MindfulAI   \n",
       "4     emptyChough6  ../../assets/icons/elephant_10197317.png  MindfulAI   \n",
       "\n",
       "  First name  Last name         DOB  \\\n",
       "0    Shannon     Clarke  1993-01-10   \n",
       "1    Richard     Hayden  1973-11-20   \n",
       "2     Jeremy     Swarey  1996-02-19   \n",
       "3     Norman    Kearney  1988-09-17   \n",
       "4     Donald  Hernandez  1985-04-05   \n",
       "\n",
       "                                          Address  \\\n",
       "0      792 Jones shoal\\nWest Bryanmouth\\nCW6M 0RR   \n",
       "1       613 Wood viaduct\\nStanleyborough\\nG6U 3QH   \n",
       "2             9 Miller fall\\nLloydmouth\\nDD8B 4WF   \n",
       "3                  7 Megan lock\\nKarlside\\nL3 8HY   \n",
       "4  32 Norris corners\\nNew Alexandraberg\\nHU01 2FE   \n",
       "\n",
       "                            email  Latitude Longitude  ...  \\\n",
       "0    Shannon_Clarke@mindfulai.com  52.66277  -2.01111  ...   \n",
       "1    Richard_Hayden@mindfulai.com  51.40148  -1.32471  ...   \n",
       "2     Jeremy_Swarey@mindfulai.com  53.69786  -2.68758  ...   \n",
       "3    Norman_Kearney@mindfulai.com  55.82737   -4.0573  ...   \n",
       "4  Donald_Hernandez@mindfulai.com  52.68333   0.93333  ...   \n",
       "\n",
       "  Epoch points: timepoint 499 HADS-Anxiety: timepoint 499  \\\n",
       "0                      1070.0                        20.0   \n",
       "1                      1080.0                         0.0   \n",
       "2                        20.0                        18.0   \n",
       "3                      1030.0                         0.0   \n",
       "4                      1150.0                         0.0   \n",
       "\n",
       "  HADS-Depression: timepoint 499 Personal health rating: timepoint 499  \\\n",
       "0                            0.0                                  91.0   \n",
       "1                            0.0                                 100.0   \n",
       "2                            0.0                                  77.0   \n",
       "3                            0.0                                  85.0   \n",
       "4                            0.0                                 100.0   \n",
       "\n",
       "  Total points Total attendance points Total log points Total check in points  \\\n",
       "0     142360.0                110000.0          12500.0                2280.0   \n",
       "1     171470.0                140000.0          13450.0                2480.0   \n",
       "2     160600.0                129000.0          13350.0                2650.0   \n",
       "3     159180.0                127000.0          12200.0                2680.0   \n",
       "4     195920.0                162000.0          12150.0                2620.0   \n",
       "\n",
       "  Total answered question points Total spoke to bot points  \n",
       "0                         4140.0                    5040.0  \n",
       "1                         3260.0                    4980.0  \n",
       "2                         3160.0                    4840.0  \n",
       "3                         3500.0                    5000.0  \n",
       "4                         3480.0                    5120.0  \n",
       "\n",
       "[5 rows x 8054 columns]"
      ]
     },
     "execution_count": 7,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df = pd.DataFrame(generate_username(sample_size),columns=['username'])\n",
    "\n",
    "df['icon']=''\n",
    "icons = glob.glob('../../assets/icons/*png')\n",
    "\n",
    "df['Trial name']='MindfulAI'\n",
    "\n",
    "#constitutional data\n",
    "df['First name']=''\n",
    "df['Last name']=''\n",
    "df['DOB']=''\n",
    "df['Address']=''\n",
    "df['email']=''\n",
    "df['Latitude']=0\n",
    "df['Longitude']=0\n",
    "df['Postcode']=''\n",
    "df['Telephone number']=''\n",
    "df['Sex']=''\n",
    "df['Gender']=''\n",
    "df['Ethnicity']=''\n",
    "df['Weight']=''\n",
    "df['Height']=''\n",
    "df['Employment status']=''\n",
    "df['Smoking status']=''\n",
    "df['ETOH units per week']=''\n",
    "df['Recreational drug use']=''\n",
    "\n",
    "#Healthcare acess data\n",
    "df['GP']=''\n",
    "df['GP address']=''\n",
    "df['GP postcode']=''\n",
    "df['GP latitude']=''\n",
    "df['GP longitude']=''\n",
    "df['GP telephone number']=''\n",
    "df['Hospital Consultant']=''\n",
    "df['Hospital address']=''\n",
    "df['Hospital postcode']=''\n",
    "df['Hospital latitude']=0\n",
    "df['Hospital longitude']=0\n",
    "df['Hospital telephone number']=''\n",
    "\n",
    "#Medical data\n",
    "df['Diagnosis']=''\n",
    "df['Duration of diagnosis']=0\n",
    "df['Family History']=''\n",
    "df['Under Community Mental Health Team']=''\n",
    "df['Has had inpatient treatment']=''\n",
    "df['Has had therapy']=''\n",
    "df['On anxiolytic']=''\n",
    "df['On anti-depressant']=''\n",
    "df['On lithium']=''\n",
    "df['On valproate']=''\n",
    "df['Polypharmacy']=''\n",
    "df['Physical disability']=''\n",
    "df['Carers']=''\n",
    "\n",
    "#start of gamification\n",
    "df['Enrollment points']=1000\n",
    "\n",
    "for i, row in tqdm(df.iterrows()):\n",
    "    total_points=0\n",
    "    total_attendance_points=0\n",
    "    total_log_points=0\n",
    "    total_check_in_points=0\n",
    "    total_asked_question_points=0\n",
    "    total_answered_question_points=0\n",
    "    total_spoke_to_bot_points=0\n",
    "    \n",
    "    df.loc[i,'icon']=icons[random.randint(0, len(icons)-1)]\n",
    "    \n",
    "    df.loc[i,'DOB']=np.random.choice([generate_random_date(1970, 2005),'Not disclosed'], p=[0.9,0.1])\n",
    "    df.loc[i,'Sex']=np.random.choice(['Male','Female','Not disclosed'],p=[0.45,0.45,0.1])\n",
    "    df.loc[i,'Gender']=np.random.choice(['Male','Female','Non-binary','Not disclosed'],p=[0.35,0.35,0.2,0.1])\n",
    "    \n",
    "    if df.loc[i,'Gender']=='Male':\n",
    "        df.loc[i,'First name']=names.get_first_name(gender='male')\n",
    "    elif df.loc[i,'Gender']=='Female':\n",
    "        df.loc[i,'First name']=names.get_first_name(gender='female')\n",
    "    else:\n",
    "        df.loc[i,'First name']=names.get_first_name()\n",
    "    df.loc[i,'Last name']=names.get_last_name()\n",
    "    \n",
    "    df.loc[i,'email']=df.loc[i,'First name']+'_'+df.loc[i,'Last name']+'@mindfulai.com'\n",
    "    \n",
    "    df.loc[i,'Ethnicity']=np.random.choice(ethnic_groups)\n",
    "    df.loc[i,'Weight']=np.random.choice([np.round(80*random.uniform(0.5, 2),1),'Not disclosed'], p=[0.9,0.1])\n",
    "    df.loc[i,'Height']=np.random.choice([np.round(random.uniform(100, 200),1),'Not disclosed'], p=[0.9,0.1])\n",
    "    df.loc[i,'Employment status']=np.random.choice(['Employed','Unemployed','Not disclosed'], p=[0.45,0.45,0.1])\n",
    "    \n",
    "    df.loc[i,'Smoking status']=np.random.choice(['Current smoker','Ex-smoker','Never smoked'])\n",
    "    df.loc[i,'ETOH units per week']=int(random.uniform(0, 35))\n",
    "    df.loc[i,'Recreational drug use']=np.random.choice(['Yes','No'])\n",
    "    \n",
    "    fake = Faker('en-UK')\n",
    "    df.loc[i,'Address']=fake.address()\n",
    "    df.loc[i,'Postcode']=fake.postcode()\n",
    "    df.loc[i,'Telephone number']=fake.phone_number()\n",
    "    \n",
    "    country='US'\n",
    "    while country != 'GB':\n",
    "        result = fake.location_on_land()\n",
    "        country = result[3]\n",
    "    df.loc[i,'Latitude']=result[0]\n",
    "    df.loc[i,'Longitude']=result[1]\n",
    "    \n",
    "    fake = Faker('en-UK')\n",
    "    df.loc[i,'GP']='Dr '+names.get_full_name()\n",
    "    df.loc[i,'GP address']=fake.address()\n",
    "    df.loc[i,'GP postcode']=fake.postcode()\n",
    "    \n",
    "    country='US'\n",
    "    while country != 'GB':\n",
    "        result = fake.location_on_land()\n",
    "        country = result[3]\n",
    "    df.loc[i,'GP Latitude']=result[0]\n",
    "    df.loc[i,'GP longitude']=result[1]\n",
    "    df.loc[i,'GP telephone number']=fake.phone_number()\n",
    "    \n",
    "    fake = Faker('en-UK')\n",
    "    df.loc[i,'Hospital Consultant']='Dr '+names.get_full_name()\n",
    "    df.loc[i,'Hospital address']=fake.address()\n",
    "    df.loc[i,'Hospital postcode']=fake.postcode()\n",
    "    \n",
    "    country='US'\n",
    "    while country != 'GB':\n",
    "        result = fake.location_on_land()\n",
    "        country = result[3]\n",
    "        \n",
    "    df.loc[i,'Hospital latitude']=result[0]\n",
    "    df.loc[i,'Hospital longitude']=result[0]\n",
    "    df.loc[i,'Hospital telephone number']=fake.phone_number()\n",
    "    \n",
    "    df.loc[i,'Diagnosis']=np.random.choice(['Major depressive disorder','Bipolar depression','Generalized Anxiety Disorder'])\n",
    "    if df.loc[i,'DOB']!='Not disclosed':\n",
    "        df.loc[i,'Duration of diagnosis']=(datetime.date.today().year-df.loc[i,'DOB'].year)*random.uniform(0, 0.5)\n",
    "    else:\n",
    "        df.loc[i,'Duration of diagnosis']=random.uniform(0, 10)\n",
    "    df.loc[i,'Family History']=np.random.choice(['Positive','Negative'])\n",
    "    df.loc[i,'Under Community Mental Health Team']=np.random.choice(['Yes','No'])\n",
    "    df.loc[i,'Has had inpatient treatment']=np.random.choice(['Yes','No'])\n",
    "    df.loc[i,'Has had therapy']=np.random.choice(['Yes','No'])\n",
    "\n",
    "    df.loc[i,'On anxiolytic']=np.random.choice(['Yes','No'])\n",
    "    df.loc[i,'On anti-depressant']=np.random.choice(['Yes','No'])\n",
    "    df.loc[i,'On lithium']=np.random.choice(['Yes','No'])\n",
    "    df.loc[i,'On valproate']=np.random.choice(['Yes','No'])\n",
    "    df.loc[i,'Polypharmacy']=np.random.choice(['Yes','No'])\n",
    "    df.loc[i,'Physical disability']=np.random.choice(['Yes','No'])\n",
    "    df.loc[i,'Carers']=np.random.choice(['Yes','No'])\n",
    "    \n",
    "    for timepoint in range(timepoints):\n",
    "        #prime an initial value with probability as shown\n",
    "        if timepoint==0:\n",
    "            df.loc[i,'Attended last appointment points: timepoint '+str(timepoint)]=np.random.choice([0,1000], p=[0.5,0.5])\n",
    "            df.loc[i,'Logged symptoms today points: timepoint '+str(timepoint)]=np.random.choice([0,50], p=[0.5,0.5])\n",
    "            df.loc[i,'Checked in today points: timepoint '+str(timepoint)]=np.random.choice([0,10], p=[0.5,0.5])\n",
    "            df.loc[i,'Asked a question points: timepoint '+str(timepoint)]=np.random.choice([0,20], p=[0.8,0.2])\n",
    "            df.loc[i,'Answered a question points: timepoint '+str(timepoint)]=np.random.choice([0,50], p=[0.8,0.2])\n",
    "            df.loc[i,'Spoke to trialbot points: timepoint '+str(timepoint)]=np.random.choice([0,20], p=[0.5,0.5])\n",
    "            df.loc[i,'Epoch points: timepoint '+str(timepoint)]=df.loc[i,'Spoke to trialbot points: timepoint '+str(timepoint)]+df.loc[i,'Answered a question points: timepoint '+str(timepoint)]+df.loc[i,'Asked a question points: timepoint '+str(timepoint)]+df.loc[i,'Checked in today points: timepoint '+str(timepoint)]+df.loc[i,'Logged symptoms today points: timepoint '+str(timepoint)]+df.loc[i,'Attended last appointment points: timepoint '+str(timepoint)]\n",
    "            \n",
    "            total_attendance_points+=df.loc[i,'Attended last appointment points: timepoint '+str(timepoint)]\n",
    "            total_log_points+=df.loc[i,'Logged symptoms today points: timepoint '+str(timepoint)]\n",
    "            total_check_in_points+=df.loc[i,'Checked in today points: timepoint '+str(timepoint)]\n",
    "            total_asked_question_points+=df.loc[i,'Asked a question points: timepoint '+str(timepoint)]\n",
    "            total_answered_question_points+=df.loc[i,'Answered a question points: timepoint '+str(timepoint)]\n",
    "            total_spoke_to_bot_points+=df.loc[i,'Spoke to trialbot points: timepoint '+str(timepoint)]\n",
    "            total_points+=df.loc[i,'Epoch points: timepoint '+str(timepoint)]\n",
    "            \n",
    "            df.loc[i,'Cumulative points: timepoint '+str(timepoint)]=total_points\n",
    "            df.loc[i,'Cumulative attendance points: timepoint '+str(timepoint)]=total_attendance_points\n",
    "            df.loc[i,'Cumulative log points: timepoint '+str(timepoint)]=total_log_points\n",
    "            df.loc[i,'Cumulative check in points: timepoint '+str(timepoint)]=total_check_in_points\n",
    "            df.loc[i,'Cumulative answered question points: timepoint '+str(timepoint)]=total_asked_question_points\n",
    "            df.loc[i,'Cumulative spoke to bot points: timepoint '+str(timepoint)]=total_spoke_to_bot_points\n",
    "\n",
    "            #simulate randomly anxiety as a function of depression, and vice versa\n",
    "            order = np.random.choice(['Anxiety','Depression'])\n",
    "            if order == 'Anxiety':\n",
    "                df.loc[i,'HADS-Anxiety: timepoint '+str(timepoint)]=random.randint(12, 21)\n",
    "                df.loc[i,'HADS-Depression: timepoint '+str(timepoint)]=int(df.loc[i,'HADS-Anxiety: timepoint '+str(timepoint)]*random.uniform(0.2, 1))\n",
    "            if order == 'Depression':\n",
    "                df.loc[i,'HADS-Depression: timepoint '+str(timepoint)]=random.randint(12, 21)\n",
    "                df.loc[i,'HADS-Anxiety: timepoint '+str(timepoint)]=int(df.loc[i,'HADS-Depression: timepoint '+str(timepoint)]*random.uniform(0.2, 1))\n",
    "            \n",
    "        #work to the hypothesis that prior values have SOME temporal relationship    \n",
    "        if timepoint>0:      \n",
    "            if df.loc[i,'Logged symptoms today points: timepoint '+str(timepoint-1)]==0:\n",
    "#                 p0,p1 = 0.7,0.3\n",
    "                p0=random.uniform(0.6, 1)\n",
    "                p1=1-p0\n",
    "            else:\n",
    "#                 p0,p1 = 0.4,0.6    \n",
    "                p1=random.uniform(0.6, 1)\n",
    "                p0=1-p1\n",
    "            df.loc[i,'Logged symptoms today points: timepoint '+str(timepoint)]=np.random.choice([0,50], p=[p0,p1])\n",
    "            \n",
    "            if df.loc[i,'Checked in today points: timepoint '+str(timepoint-1)]==0:\n",
    "#                 p0,p1 = 0.6,0.4\n",
    "                p0=random.uniform(0.6, 1)\n",
    "                p1=1-p0\n",
    "            else:\n",
    "                p0,p1 = 0.4,0.6\n",
    "                p1=random.uniform(0.6, 1)\n",
    "                p0=1-p1\n",
    "            df.loc[i,'Checked in today points: timepoint '+str(timepoint)]=np.random.choice([0,10], p=[p0,p1])\n",
    "            \n",
    "            if df.loc[i,'Asked a question points: timepoint '+str(timepoint-1)]==0:\n",
    "#                 p0,p1 = 0.8,0.2\n",
    "                p0=random.uniform(0.8, 1)\n",
    "                p1=1-p0\n",
    "            else:\n",
    "#                 p0,p1 = 0.4,0.6\n",
    "                p1=random.uniform(0.6, 1)\n",
    "                p0=1-p1\n",
    "            df.loc[i,'Asked a question points: timepoint '+str(timepoint)]=np.random.choice([0,20], p=[p0,p1])\n",
    "            \n",
    "            if df.loc[i,'Answered a question points: timepoint '+str(timepoint-1)]==0:\n",
    "#                 p0,p1 = 0.8,0.2\n",
    "                p0=random.uniform(0.8, 1)\n",
    "                p1=1-p0\n",
    "            else:\n",
    "#                 p0,p1 = 0.4,0.6\n",
    "                p1=random.uniform(0.6, 1)\n",
    "                p0=1-p1\n",
    "            df.loc[i,'Answered a question points: timepoint '+str(timepoint)]=np.random.choice([0,50], p=[p0,p1])\n",
    "            \n",
    "            if df.loc[i,'Spoke to trialbot points: timepoint '+str(timepoint-1)]==0:\n",
    "#                 p0,p1 = 0.6,0.4\n",
    "                p0=random.uniform(0.6, 1)\n",
    "                p1=1-p0\n",
    "            else:\n",
    "#                 p0,p1 = 0.4,0.6\n",
    "                p1=random.uniform(0.6, 1)\n",
    "                p0=1-p1\n",
    "            df.loc[i,'Spoke to trialbot points: timepoint '+str(timepoint)]=np.random.choice([0,20], p=[p0,p1])\n",
    "\n",
    "            \n",
    "            engagement = df.loc[i,'Epoch points: timepoint '+str(timepoint-1)]/max_points_per_epoch\n",
    "            \n",
    "            #multi-level engagement probabalistic attendence from app engagement\n",
    "            if df.loc[i,'Attended last appointment points: timepoint '+str(timepoint-1)]>0 and engagement>0.9:\n",
    "                p1=random.uniform(0.7, 1)\n",
    "                p0=1-p1\n",
    "            if df.loc[i,'Attended last appointment points: timepoint '+str(timepoint-1)]>0 and engagement>0.8:\n",
    "                p1=random.uniform(0.6, 1)\n",
    "                p0=1-p1\n",
    "            if df.loc[i,'Attended last appointment points: timepoint '+str(timepoint-1)]>0 and engagement>0.7:\n",
    "                p1=random.uniform(0.5, 1)\n",
    "                p0=1-p1\n",
    "            if df.loc[i,'Attended last appointment points: timepoint '+str(timepoint-1)]>0 and engagement>0.6:\n",
    "                p1=random.uniform(0.4, 1)\n",
    "                p0=1-p1\n",
    "            if df.loc[i,'Attended last appointment points: timepoint '+str(timepoint-1)]>0 and engagement>0.5:\n",
    "                p1=random.uniform(0.3, 1)\n",
    "                p0=1-p1\n",
    "            else:\n",
    "                p0=random.uniform(0.75, 1)\n",
    "                p1=1-p0\n",
    "\n",
    "            df.loc[i,'Attended last appointment points: timepoint '+str(timepoint)]=np.random.choice([0,1000], p=[p0,p1])\n",
    "        \n",
    "            total_attendance_points+=df.loc[i,'Attended last appointment points: timepoint '+str(timepoint)]\n",
    "            total_log_points+=df.loc[i,'Logged symptoms today points: timepoint '+str(timepoint)]\n",
    "            total_check_in_points+=df.loc[i,'Checked in today points: timepoint '+str(timepoint)]\n",
    "            total_asked_question_points+=df.loc[i,'Asked a question points: timepoint '+str(timepoint)]\n",
    "            total_answered_question_points+=df.loc[i,'Answered a question points: timepoint '+str(timepoint)]\n",
    "            total_spoke_to_bot_points+=df.loc[i,'Spoke to trialbot points: timepoint '+str(timepoint)]\n",
    "            \n",
    "            df.loc[i,'Cumulative points: timepoint '+str(timepoint)]=total_points\n",
    "            df.loc[i,'Cumulative attendance points: timepoint '+str(timepoint)]=total_attendance_points\n",
    "            df.loc[i,'Cumulative log points: timepoint '+str(timepoint)]=total_log_points\n",
    "            df.loc[i,'Cumulative check in points: timepoint '+str(timepoint)]=total_check_in_points\n",
    "            df.loc[i,'Cumulative answered question points: timepoint '+str(timepoint)]=total_asked_question_points\n",
    "            df.loc[i,'Cumulative spoke to bot points: timepoint '+str(timepoint)]=total_spoke_to_bot_points\n",
    "            df.loc[i,'Epoch points: timepoint '+str(timepoint)]=df.loc[i,'Spoke to trialbot points: timepoint '+str(timepoint)]+df.loc[i,'Answered a question points: timepoint '+str(timepoint)]+df.loc[i,'Asked a question points: timepoint '+str(timepoint)]+df.loc[i,'Checked in today points: timepoint '+str(timepoint)]+df.loc[i,'Logged symptoms today points: timepoint '+str(timepoint)]+df.loc[i,'Attended last appointment points: timepoint '+str(timepoint)]\n",
    "            total_points+=df.loc[i,'Epoch points: timepoint '+str(timepoint)]\n",
    "            \n",
    "            df.loc[i,'HADS-Anxiety: timepoint '+str(timepoint)]=int((df.loc[i,'HADS-Anxiety: timepoint '+str(timepoint-1)]-engagement)*random.uniform(0.95, 1.2))\n",
    "            df.loc[i,'HADS-Depression: timepoint '+str(timepoint)]=int((df.loc[i,'HADS-Depression: timepoint '+str(timepoint-1)]-engagement)*random.uniform(0.95, 1.2))\n",
    "    \n",
    "            if df.loc[i,'HADS-Anxiety: timepoint '+str(timepoint)]>21:\n",
    "                df.loc[i,'HADS-Anxiety: timepoint '+str(timepoint)]=21\n",
    "            if df.loc[i,'HADS-Depression: timepoint '+str(timepoint)]>21:\n",
    "                df.loc[i,'HADS-Depression: timepoint '+str(timepoint)]=21\n",
    "            if df.loc[i,'HADS-Anxiety: timepoint '+str(timepoint)]<0:\n",
    "                df.loc[i,'HADS-Anxiety: timepoint '+str(timepoint)]=0\n",
    "            if df.loc[i,'HADS-Depression: timepoint '+str(timepoint)]<0:\n",
    "                df.loc[i,'HADS-Depression: timepoint '+str(timepoint)]=0\n",
    "                \n",
    "        #simulate personal health rating as a function of perfect health MINUS anxiety and depression, plus a metric of app engagement, multipled a random variable of life event\n",
    "        engagement = df.loc[i,'Epoch points: timepoint '+str(timepoint)]/max_points_per_epoch\n",
    "        enagement_weight=15*engagement\n",
    "        df.loc[i,'Personal health rating: timepoint '+str(timepoint)]=int((100-(df.loc[i,'HADS-Depression: timepoint '+str(timepoint)]+df.loc[i,'HADS-Anxiety: timepoint '+str(timepoint)])+enagement_weight)*random.uniform(0.7, 1.1))\n",
    "        if df.loc[i,'Personal health rating: timepoint '+str(timepoint)]>100:\n",
    "            df.loc[i,'Personal health rating: timepoint '+str(timepoint)]=100\n",
    "    \n",
    "    df.loc[i,'Total points']=total_points\n",
    "    df.loc[i,'Total attendance points']=total_attendance_points\n",
    "    df.loc[i,'Total log points']=total_log_points\n",
    "    df.loc[i,'Total check in points']=total_check_in_points\n",
    "    df.loc[i,'Total answered question points']=total_asked_question_points\n",
    "    df.loc[i,'Total spoke to bot points']=total_spoke_to_bot_points\n",
    "        \n",
    "print(df.shape)\n",
    "df.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "id": "811385e3",
   "metadata": {},
   "outputs": [],
   "source": [
    "#simulate award badges for users\n",
    "#similar to git achievements\n",
    "badges = glob.glob('../../assets/badges/*png')\n",
    "\n",
    "for i, row in df.iterrows():\n",
    "    for b in range(len(badges)):\n",
    "        png = badges[b]\n",
    "        award = badges[b].split('/')[-1].split('.')[0].replace(\"_\", \" \").title()\n",
    "        \n",
    "        #random allocation of if award or not\n",
    "        achievement_unlocked = np.random.choice([0,1], p=[0.7,0.3])\n",
    "        if achievement_unlocked ==1:\n",
    "            df.loc[i,award]=png\n",
    "        else:\n",
    "            df.loc[i,award]=''"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "id": "4bc839ee",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Example temporal health rating over time of users\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>username</th>\n",
       "      <th>variable</th>\n",
       "      <th>value</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>8100</th>\n",
       "      <td>enragedBass3</td>\n",
       "      <td>Cumulative points: timepoint 0</td>\n",
       "      <td>1000.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8101</th>\n",
       "      <td>thrilledCaviar1</td>\n",
       "      <td>Cumulative points: timepoint 0</td>\n",
       "      <td>50.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8102</th>\n",
       "      <td>yearningBaboon5</td>\n",
       "      <td>Cumulative points: timepoint 0</td>\n",
       "      <td>60.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8103</th>\n",
       "      <td>pacifiedTuna1</td>\n",
       "      <td>Cumulative points: timepoint 0</td>\n",
       "      <td>1020.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8104</th>\n",
       "      <td>emptyChough6</td>\n",
       "      <td>Cumulative points: timepoint 0</td>\n",
       "      <td>20.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1205695</th>\n",
       "      <td>amazedWildfowl9</td>\n",
       "      <td>Cumulative points: timepoint 499</td>\n",
       "      <td>161420.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1205696</th>\n",
       "      <td>scornfulMuesli2</td>\n",
       "      <td>Cumulative points: timepoint 499</td>\n",
       "      <td>141910.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1205697</th>\n",
       "      <td>soreTacos1</td>\n",
       "      <td>Cumulative points: timepoint 499</td>\n",
       "      <td>127240.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1205698</th>\n",
       "      <td>jealousTortoise1</td>\n",
       "      <td>Cumulative points: timepoint 499</td>\n",
       "      <td>179910.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1205699</th>\n",
       "      <td>dejectedIguana0</td>\n",
       "      <td>Cumulative points: timepoint 499</td>\n",
       "      <td>136370.0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>75000 rows × 3 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "                 username                          variable     value\n",
       "8100         enragedBass3    Cumulative points: timepoint 0    1000.0\n",
       "8101      thrilledCaviar1    Cumulative points: timepoint 0      50.0\n",
       "8102      yearningBaboon5    Cumulative points: timepoint 0      60.0\n",
       "8103        pacifiedTuna1    Cumulative points: timepoint 0    1020.0\n",
       "8104         emptyChough6    Cumulative points: timepoint 0      20.0\n",
       "...                   ...                               ...       ...\n",
       "1205695   amazedWildfowl9  Cumulative points: timepoint 499  161420.0\n",
       "1205696   scornfulMuesli2  Cumulative points: timepoint 499  141910.0\n",
       "1205697        soreTacos1  Cumulative points: timepoint 499  127240.0\n",
       "1205698  jealousTortoise1  Cumulative points: timepoint 499  179910.0\n",
       "1205699   dejectedIguana0  Cumulative points: timepoint 499  136370.0\n",
       "\n",
       "[75000 rows x 3 columns]"
      ]
     },
     "execution_count": 9,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "print(\"Example temporal health rating over time of users\")\n",
    "df_melt = pd.melt(df,id_vars='username')\n",
    "df_melt = df_melt[df_melt['variable'].str.contains('Cumulative points')]\n",
    "df_melt"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "id": "e0c94fe6",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<Axes: xlabel='variable', ylabel='value'>"
      ]
     },
     "execution_count": 10,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "sns.lineplot(data=df_melt[df_melt['username']==np.unique(df_melt['username'])[0]],x='variable',y='value')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "id": "f7f15a46",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<Axes: xlabel='variable', ylabel='value'>"
      ]
     },
     "execution_count": 11,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "sns.lineplot(data=df_melt[df_melt['username']==np.unique(df_melt['username'])[1]],x='variable',y='value')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "id": "b792f6db",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<Axes: xlabel='variable', ylabel='value'>"
      ]
     },
     "execution_count": 12,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "sns.lineplot(data=df_melt[df_melt['username']==np.unique(df_melt['username'])[2]],x='variable',y='value')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "id": "a3f43534",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<Axes: xlabel='variable', ylabel='value'>"
      ]
     },
     "execution_count": 13,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "sns.lineplot(data=df_melt[df_melt['username']==np.unique(df_melt['username'])[3]],x='variable',y='value')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "id": "a7b95786",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<Axes: xlabel='variable', ylabel='value'>"
      ]
     },
     "execution_count": 14,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAArcAAAGwCAYAAACzcE1SAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjcuMSwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy/bCgiHAAAACXBIWXMAAA9hAAAPYQGoP6dpAABU/UlEQVR4nO3deXhU5f3+8fdkmcmeECAbhCTssssWgoBaUoIiSt1AUBFQ1KIVcUGqULH9icW6Viu1FrGtFuVbpS0oFlFBICIgOxJZEggkEyAhM9m3Ob8/kKkpYU3IyUzu13XNReaczznzmbmAuXl4znMshmEYiIiIiIh4AR+zGxARERERaSgKtyIiIiLiNRRuRURERMRrKNyKiIiIiNdQuBURERERr6FwKyIiIiJeQ+FWRERERLyGn9kNNFcul4ucnBxCQ0OxWCxmtyMiIiLnwTAMioqKiIuLw8dHY4RNkcKtSXJycoiPjze7DREREbkI2dnZtG3b1uw2pA4KtyYJDQ0FTv7hCAsLM7kbEREROR9Op5P4+Hj397g0PQq3Jjk1FSEsLEzhVkRExMNoSmHTpckiIiIiIuI1FG5FRERExGso3IqIiIiI11C4FRERERGvoXArIiIiIl5D4VZEREREvIbCrYiIiIh4DYVbEREREfEaCrciIiIi4jUUbkVERETEayjcioiIiIjXULgVEREREa/hZ3YDIiIiIg2pxmVgd5bjY4HY8ECz25FGpnArIiIiTVZVjYvdOU4OFZRSUFJJfkklReVVFJdXU1xx8nGsqILC0ioAXIZBQUkl1S6D8cntePZnPU1+B9LYFG5FRETEVPuOFnOksIwal4vqGoOyqhr2HythU1YBWw4VUlZVc8Hn9Pe1UFntugTdSlOncCsiIiKm2XnEwfWvrcVlnLkmPNCfLtGhtAq1EhlsJSzAn5AAP0JtfgTb/GgRbKV1iM1d3zLESlRoAL4+lkZ4B9LUKNyKiIiIaRas3o/LgOgwGzFhJwOp1c+HNhFB9E2IYGBiJB1ah+CjoCrnqcmslrBmzRpGjx5NXFwcFouFpUuX1tpvsVjqfDz//PPumsTExNP2P/fcc7XOs337doYOHUpAQADx8fHMnz//tF6WLFlC165dCQgIoGfPnnz88ce19huGwZw5c4iNjSUwMJDU1FT27t3bcB+GiIhIM5BdUMrHO3IBePuugfzzgSF8+PMrWDw1hRdu7c2E5AQ6RYcq2MoFaTLhtqSkhN69e/P666/XuT83N7fWY+HChVgsFm666aZadc8880ytugcffNC9z+l0MmLECBISEti8eTPPP/88Tz/9NG+++aa7Zv369dx2221MmTKFLVu2MGbMGMaMGcPOnTvdNfPnz+fVV19lwYIFbNiwgeDgYNLS0igvL2/gT0VERMR7LVyXicuAoZ1a0S0uzOx2xEtYDMM4yywXc1gsFj766CPGjBlzxpoxY8ZQVFTEqlWr3NsSExOZPn0606dPr/OYN954gyeffBK73Y7VagXgiSeeYOnSpezZsweAsWPHUlJSwrJly9zHDRo0iD59+rBgwQIMwyAuLo5HHnmERx99FACHw0F0dDSLFi1i3Lhx5/UenU4n4eHhOBwOwsL0B1pERJoXR2kVKc+torSyhr9MHsiwzq3Nbum86Pu76WsyI7cXIi8vj+XLlzNlypTT9j333HO0bNmSyy+/nOeff57q6mr3vvT0dIYNG+YOtgBpaWlkZGRw4sQJd01qamqtc6alpZGeng5AZmYmdru9Vk14eDjJycnumrpUVFTgdDprPURERLydYRgUlVdxtKic7IJS9h8rZucRB69/uY/Syhq6xoQytFMrs9sUL+KRF5S98847hIaGcuONN9ba/otf/IK+ffsSGRnJ+vXrmTVrFrm5ubz44osA2O12kpKSah0THR3t3teiRQvsdrt7249r7Ha7u+7Hx9VVU5d58+Yxd+7ci3i3IiIiTd83mQUsXJtJWVUNhWVVFJVXUVZZw4nSSsqrzrwk191D22OxaE6tNByPDLcLFy5kwoQJBAQE1No+Y8YM98+9evXCarVy7733Mm/ePGw22/+eplHNmjWrVn9Op5P4+HgTOxIREWkYNS6Dx/5vGwfzS89YY7GA1dcHm58PQVY/Aq2+dIsN4/recY3YqTQHHhduv/rqKzIyMnj//ffPWZucnEx1dTVZWVl06dKFmJgY8vLyatWceh4TE+P+ta6aH+8/tS02NrZWTZ8+fc7Yi81mMz1gi4iIXAqf7Mx1B9tnbuhObHggYQEnA2yLICutQmwE+PtohFYahcfNuf3zn/9Mv3796N279zlrt27dio+PD1FRUQCkpKSwZs0aqqqq3DUrV66kS5cutGjRwl3z44vUTtWkpKQAkJSURExMTK0ap9PJhg0b3DUiIiLNhWEYLFi9H4CHhnfizpREftotmuT2LenVNoL4yCACrb4KttJomszIbXFxMfv27XM/z8zMZOvWrURGRtKuXTvgZIhcsmQJL7zwwmnHp6ens2HDBq6++mpCQ0NJT0/n4Ycf5vbbb3cH1/HjxzN37lymTJnCzJkz2blzJ6+88govvfSS+zwPPfQQV155JS+88AKjRo1i8eLFbNq0yb1cmMViYfr06fzmN7+hU6dOJCUlMXv2bOLi4s66uoOIiIi3KKusYfvhQnYccfDtoRPsPOIk0N+XiYMTzW5NpOmE202bNnH11Ve7n5+anzpx4kQWLVoEwOLFizEMg9tuu+204202G4sXL+bpp5+moqKCpKQkHn744VrzXMPDw/nPf/7DtGnT6NevH61atWLOnDlMnTrVXTN48GDee+89nnrqKX75y1/SqVMnli5dSo8ePdw1jz/+OCUlJUydOpXCwkKGDBnCihUrTpsDLCIi4m0cZVX89MXVHC2qqLX99kHtiAy2nuEokcbTJNe5bQ60Tp6IiHiiP67ez7xP9hAa4Meg9i3p2SacXm3DGdKxFX6+Hjfb8YLp+7vpazIjtyIiItK0vbnmZLAFmD2qG7cO0Ko/0vR4/z+xREREpN4O5Ze6g23vtuFc30dLeEnTpJFbEREROaMal8HOIw6eWbYbw4A+8RH84/7B+Ppo9QNpmhRuRUREpE7fHjrB1L9s5njxfy8eeyi1k4KtNGkKtyIiIlKn/7f8O44XVxBq82Nwx5bc2LctV3eJMrstkbNSuBUREZFayipreOLD7Ww+eAJ/XwurHr2SqFAtdymeQReUiYiISC0L12Xyz605AIzuFadgKx5F4VZERETcKqpreGd9FgA3Xt6GX4/pcfYDRJoYTUsQERERt39tzeFoUQXRYTaeu6kXVj+Ng4ln0e9YERERAcAwDN76KhOAuwYnKdiKR9LvWhEREQFgzd7jZOQVEWz1ZXxyO7PbEbkoCrciIiICwFtfHQBg7IB2hAf6m9yNyMVRuBURERF25zj5au9xfCww6YpEs9sRuWgKtyIiIsJba0+O2l7TM5b4yCCTuxG5eAq3IiIizZzdUc6/fljXdurQ9iZ3I1I/CrciIiLN3KL1WVS7DAYmRdI7PsLsdkTqReFWRESkGSuuqObdDQcBjdqKd9BNHERERJqh/OIKFm/MZmt2IUXl1bRvHcxPukaZ3ZZIvSncioiINENP/3s3/96W435+95D2+PhYTOxIpGEo3IqIiDQzh/JLWb79ZLCdfEUSSa2DGTsg3uSuRBqGwq2IiEgz86evDuAy4MrOrZkzupvZ7Yg0KF1QJiIi0owcL67gg03ZANx7pS4gE++jcCsiItKM/GV9FhXVLnq3DSelfUuz2xFpcAq3IiIizURJRTXvpJ9c9uu+KztgsegCMvE+CrciIiLNxOKN2TjKqkhqFcyI7jFmtyNySeiCMhERES9TWe3im8wCTpRWUlxRTXF5Nc7yKn7/+T4A7hnaHl8t+yVeSuFWRETEyzz78XcsWp9V576oUBs39m3TuA2JNCKFWxERES9yvLiCv39zCID+CS2ICLISYvMlJMCP2PBARvaIIcDf1+QuRS4dhVsREREv8tf0g+7VEJbcl6KLxqTZ0QVlIiIiXqKy2sV7P4zaThnaXsFWmiWFWxERES/xyc5cjhVVEBVqY6RWQ5BmSuFWRETES/zlhzVsxye3w+qnr3hpnvQ7X0RExAvsPOJg88ET+PtaGJ/czux2REyjcCsiIuIF/pKeBcA1PWKJCg0wtxkRE2m1BBEREQ9UWe3i8IlSduc62ZhZwAebDgMwcXCCyZ2JmEvhVkREpAlzlFXxXa6TzOMlZB4v4fu8IvYdLSansAyXUbv2io4t6duuhTmNijQRTWZawpo1axg9ejRxcXFYLBaWLl1aa/9dd92FxWKp9Rg5cmStmoKCAiZMmEBYWBgRERFMmTKF4uLiWjXbt29n6NChBAQEEB8fz/z580/rZcmSJXTt2pWAgAB69uzJxx9/XGu/YRjMmTOH2NhYAgMDSU1NZe/evQ3zQYiIiPwgz1nOkN9+zrg3v2bWhzt4c80Bvsw4xuETJ4NtkNWXnm3CuTMlgdfH9+XPEwdo+S9p9prMyG1JSQm9e/dm8uTJ3HjjjXXWjBw5krffftv93Gaz1do/YcIEcnNzWblyJVVVVUyaNImpU6fy3nvvAeB0OhkxYgSpqaksWLCAHTt2MHnyZCIiIpg6dSoA69ev57bbbmPevHlcd911vPfee4wZM4Zvv/2WHj16ADB//nxeffVV3nnnHZKSkpg9ezZpaWns3r2bgADNcxIRkYaxcF0mReXVRAT507ttBEmtgukYFUKnqBCSWgfTOsSmMCvyPyyGYRjnLmtcFouFjz76iDFjxri33XXXXRQWFp42onvKd999R7du3di4cSP9+/cHYMWKFVx77bUcPnyYuLg43njjDZ588knsdjtWqxWAJ554gqVLl7Jnzx4Axo4dS0lJCcuWLXOfe9CgQfTp04cFCxZgGAZxcXE88sgjPProowA4HA6io6NZtGgR48aNO6/36HQ6CQ8Px+FwEBYWdqEfkYiIeJnvcp28tPJ7CsuqKKmo5mhRBceKKgD40539+Wm3aJM7FND3tydoMtMSzseXX35JVFQUXbp04f777yc/P9+9Lz09nYiICHewBUhNTcXHx4cNGza4a4YNG+YOtgBpaWlkZGRw4sQJd01qamqt101LSyM9PR2AzMxM7HZ7rZrw8HCSk5PdNXWpqKjA6XTWeoiIiJwy5587+c/uPL7JLGBXjtMdbId1bs3wrlEmdyfiOZrMtIRzGTlyJDfeeCNJSUns37+fX/7yl1xzzTWkp6fj6+uL3W4nKqr2H34/Pz8iIyOx2+0A2O12kpKSatVER0e797Vo0QK73e7e9uOaH5/jx8fVVVOXefPmMXfu3It45yIi4k0MwyC/pJKD+SUcKiglu6CMA8eK2Zh1AquvD/Nv7kV4kD+RQVY6RIUQYvOYr2qRJsFj/sT8+L/7e/bsSa9evejQoQNffvklw4cPN7Gz8zNr1ixmzJjhfu50OomPjzexIxERuVQqq13kFJaRfeJkeM0+UcrB/BIO5pdyML+U4orqOo+7qV9bxlzeppG7FfEuHhNu/1f79u1p1aoV+/btY/jw4cTExHD06NFaNdXV1RQUFBATc/L+2jExMeTl5dWqOfX8XDU/3n9qW2xsbK2aPn36nLFfm8122gVwIiLiXX73aQYffnsYu7P8tGW6fsxigdiwANq1DKJdZBDxLYJIaBXMTy/TvFqR+vLYcHv48GHy8/PdATMlJYXCwkI2b95Mv379APj8889xuVwkJye7a5588kmqqqrw9/cHYOXKlXTp0oUWLVq4a1atWsX06dPdr7Vy5UpSUlIASEpKIiYmhlWrVrnDrNPpZMOGDdx///2N8dZFRKQJ+j6viNe+2Od+bvPzIT4yiPgWgcRHBpHQMpjElkEktAyibYsgAvx9TexWxHs1mXBbXFzMvn3//UshMzOTrVu3EhkZSWRkJHPnzuWmm24iJiaG/fv38/jjj9OxY0fS0tIAuOyyyxg5ciT33HMPCxYsoKqqigceeIBx48YRFxcHwPjx45k7dy5Tpkxh5syZ7Ny5k1deeYWXXnrJ/boPPfQQV155JS+88AKjRo1i8eLFbNq0iTfffBM4uZLD9OnT+c1vfkOnTp3cS4HFxcXVWt1BRESal4VrMwH4Sdconrupp5bpEjGL0UR88cUXBnDaY+LEiUZpaakxYsQIo3Xr1oa/v7+RkJBg3HPPPYbdbq91jvz8fOO2224zQkJCjLCwMGPSpElGUVFRrZpt27YZQ4YMMWw2m9GmTRvjueeeO62XDz74wOjcubNhtVqN7t27G8uXL6+13+VyGbNnzzaio6MNm81mDB8+3MjIyLig9+twOAzAcDgcF3SciIg0PceKyo1OT35sJMxcZmzMzDe7HbmE9P3d9DXJdW6bA62TJyLiPV75bC8vffY9vduGs3TaFRqx9WL6/m76PGqdWxERkaamvKqGv36dBcCUoe0VbEVMpnArIiJSD//alsPx4kpiwwO4pkeM2e2INHsKtyIiIhfJMAz3hWR3DU7E31dfqyJm059CERGRi7RuXz577EUEWX0ZN7Cd2e2ICAq3IiIiF+3Paw8AcGv/eMID/U3uRkSgCa1zKyIi4gnW7z/OgWMl5DrK+CLjGBYLTLoi0ey2ROQHCrciIiLnaVt2IeP/tKHWtp9eFk1Cy2CTOhKR/6VwKyIicp4WrN7v/vm2gfHERwZxa/94EzsSkf+lcCsiInIeMo+XsGKXHYBPpw+jS0yoyR2JSF10QZmIiMg52B3lPPj3bzEM+EnXKAVbkSZMI7ciIiJnUeMyGPdmOln5pQDcO6y9yR2JyNlo5FZEROQs/rPL7g62T157GQOTIk3uSETORiO3IiIiZ7Fw3ck7kD1wdUfu0aitSJOnkVsREZEz2H64kI1ZJ/DzsXBHSoLZ7YjIeVC4FREROYOFa0+O2l7XK5bosACTuxGR86FwKyIiUoc8ZznLtucCMGWIpiOIeArNuRURkWavuKKaDHsRe+xO9uQW8V2uk505DqpdBgMSW9CzbbjZLYrIeVK4FRERr2UYBs7yauyOcnIdZRSUVFJYWkVhaSWFZVXkOsrJsBdxqKC0zuPDA/15+KedG7lrEakPhVsREfFKhmFw19sbWf39sfOqjw6z0TUmjK6xoXSNCaVnmwjatwrGx8dyiTsVkYakcCsiIl7pq73H3cG2RZA/0WEBtA61ERFkJSLQn4ggf1oGW+kcE0rXmDAig60mdywiDUHhVkREvNIf1+wHYNIVifxqdHeTuxGRxqLVEkRExOvsPOJg3b58fH0sTBmSZHY7ItKIFG5FRMTr/HHNAQBG94qlbYsgk7sRkcakcCsiIl4lu6CU5dtzAJg6rIPJ3YhIY1O4FRERr/LWVwdwGTC0Uyu6xYWZ3Y6INDJdUCYiIl6hqsZFbmE572/KBuC+KzVqK9IcKdyKiIjH+fe2HP65NYdjxRUcL6qgsLSSksoa9/7ucWEM7tDSxA5FxCwKtyIi4lGOFVXwyJJtVFa76tzfKsTKE9d0xWLRzRdEmiOFWxERaZLKq2o4XlzB8eJKCkoqKCqvZt/RYt5Zn0VltYtusWFMT+1Eq1AbkUFWIoL8CQ3wx1d3FBNp1hRuRUTEVKWV1fxraw77jxVzML+UQwWl5BSW4SyvPutxD6V2YkT3mEbqUkQ8hcKtiIiY6neffs/CdZl17vP3tdA6xEbLEBtBVl+SWgXTMSqEDq1DuLprVCN3KiKeQOFWRERMU1xRzZIfVje4pV9buseFkdAymLYtAokKDSAs0E9zZ0XkgijcioiIaT7acoSiimratwrmtzf1wkfzZUWknnQTBxERMYVhGPw1PQuA2wclKNiKSIPQyK2IiDQoR2kVhwpKcZRV4SirorKmhqoag6oaF4WlVWQeLyHzeAlZx0vIL6kk0N+Xm/q1NbttEfESCrciItJgDuWXMvKVNZT+6IYK5zJ1WHvCA/0vYVci0pwo3IqIyEUrq6zh8IlSKqpdVFS7WLQ+i9LKGkID/GgTEUhYgD82fx/8fCz4+foQavMjqVUwia2C3b+G2PRVJCINp8nMuV2zZg2jR48mLi4Oi8XC0qVL3fuqqqqYOXMmPXv2JDg4mLi4OO68805ycnJqnSMxMRGLxVLr8dxzz9Wq2b59O0OHDiUgIID4+Hjmz59/Wi9Lliyha9euBAQE0LNnTz7++ONa+w3DYM6cOcTGxhIYGEhqaip79+5tuA9DRKQJc5RV8fS/dnHNK1/R/Vcr+OlLa7ju92u56Y31/Hvbyb+XF941gBXTh/HBfSn8dUoyb08ayJ/u7M+LY/vw4PBOjO4dR4824Qq2ItLgmky4LSkpoXfv3rz++uun7SstLeXbb79l9uzZfPvtt3z44YdkZGRw/fXXn1b7zDPPkJub6348+OCD7n1Op5MRI0aQkJDA5s2bef7553n66ad588033TXr16/ntttuY8qUKWzZsoUxY8YwZswYdu7c6a6ZP38+r776KgsWLGDDhg0EBweTlpZGeXl5A38qIiJNzx++3Mei9Vl8l+vEZUBogB/RYTbiIwPp0DqYyVckMSAx0uw2RaSZshiGYZjdxP+yWCx89NFHjBkz5ow1GzduZODAgRw8eJB27doBJ0dup0+fzvTp0+s85o033uDJJ5/EbrdjtVoBeOKJJ1i6dCl79uwBYOzYsZSUlLBs2TL3cYMGDaJPnz4sWLAAwzCIi4vjkUce4dFHHwXA4XAQHR3NokWLGDduXJ2vXVFRQUVFhfu50+kkPj4eh8NBWFjYeX82IiJmKq+qYdC8VRSWVjHrmq7c0KcN0WE2rUUrzYbT6SQ8PFzf301Ykxm5vVAOhwOLxUJERESt7c899xwtW7bk8ssv5/nnn6e6+r+3b0xPT2fYsGHuYAuQlpZGRkYGJ06ccNekpqbWOmdaWhrp6ekAZGZmYrfba9WEh4eTnJzsrqnLvHnzCA8Pdz/i4+Mv+r2LiJjlX9tyKCytok1EIHcPbU9MeICCrYg0KR4ZbsvLy5k5cya33XZbrX81/eIXv2Dx4sV88cUX3HvvvTz77LM8/vjj7v12u53o6Oha5zr13G63n7Xmx/t/fFxdNXWZNWsWDofD/cjOzr7Qty0iYirDMHhnfRYAd6Qk4Kt1aUWkCfK4mfxVVVXceuutGIbBG2+8UWvfjBkz3D/36tULq9XKvffey7x587DZbI3dai02m830HkRE6uPbQ4XsynFi8/NhbH/975OINE0eNXJ7KtgePHiQlStXnnOuS3JyMtXV1WRlZQEQExNDXl5erZpTz2NiYs5a8+P9Pz6urhoREW/0lx/uJnZ97zhaBFvPXiwiYhKPCbengu3evXv57LPPaNmy5TmP2bp1Kz4+PkRFRQGQkpLCmjVrqKqqctesXLmSLl260KJFC3fNqlWrap1n5cqVpKSkAJCUlERMTEytGqfTyYYNG9w1IiLe5qiznI935AIwcXCiuc2IiJxFk5mWUFxczL59+9zPMzMz2bp1K5GRkcTGxnLzzTfz7bffsmzZMmpqatzzWyMjI7FaraSnp7NhwwauvvpqQkNDSU9P5+GHH+b22293B9fx48czd+5cpkyZwsyZM9m5cyevvPIKL730kvt1H3roIa688kpeeOEFRo0axeLFi9m0aZN7uTCLxcL06dP5zW9+Q6dOnUhKSmL27NnExcWddXUHERFP9vb6LKpqDPoltKBHm3Cz2xEROTOjifjiiy8M4LTHxIkTjczMzDr3AcYXX3xhGIZhbN682UhOTjbCw8ONgIAA47LLLjOeffZZo7y8vNbrbNu2zRgyZIhhs9mMNm3aGM8999xpvXzwwQdG586dDavVanTv3t1Yvnx5rf0ul8uYPXu2ER0dbdhsNmP48OFGRkbGBb1fh8NhAIbD4biwD0pEpJEVlVcZPX61wkiYucxYsTPX7HZETKXv76avSa5z2xxonTwR8RR/XpvJr5ftpn2rYD6bcSU+WiVBmjF9fzd9HjPnVkREGl9VjYuFazMBuHtoewVbEWnyFG5FROSMPt6Ry5HCMloGW7mxbxuz2xEROSeFWxERqZNhGPxx9QHg5AoJAf6+JnckInJuCrciIlKn9fvz2Z3rJNDflzsGJZjdjojIeVG4FRGROv1xzclR21v7t9VNG0TEYyjciojIab7LdbLm+2P4WGDKkPZmtyMict4UbkVE5DR/+urkqO01PWJp1zLI5G5ERM6fwq2IiNSS6yjjX1tzAJg6TKO2IuJZFG5FRKSWt9dlUe0ySE6KpHd8hNntiIhcEIVbERFxc5ZX8d6GQwDce6VGbUXE8yjcioiI2+JvDlFcUU3HqBCu6hxldjsiIhdM4VZERADYd7SY367IAGCqbrUrIh7Kz+wGRETEfFuzCxnz+joAokJt3HB5nMkdiYhcHI3ciogIC77c7/7512N6YPPTrXZFxDNp5FZEpJk7mF/Cp7vtAPzn4WF0jg41uSMRkYunkVsRkWbMMAze+ioTw4CrurRWsBURj6eRWxGRZuj7vCJe+3wfG7MKyHWUA3DPUC39JSKeT+FWRKSZqax2cdfCb8j5IdT6+Vi4dUA8gzu0NLkzEZH6U7gVEWlmlm49Qo6jnNahNl4Z24c+7SIIsurrQES8g/42ExFpRmpcBgtWn1wZ4e4hSQzu2MrkjkREGpYuKBMRaUb+s8vOgWMlhAX4MWFQgtntiIg0OIVbEZFmwjAM3vhh1Hbi4ERCbPrPOxHxPgq3IiLNwLGiCm5ekM72ww4C/H24a3Ci2S2JiFwSCrciIs3A7z/fy+aDJwC4Y1ACLUNsJnckInJp6P+kRES82BcZR0nfn8+STYcBeGh4Jx74SUeTuxIRuXQUbkVEvFSGvYjJizZiGCef92gTxvTUTlgsFnMbExG5hBRuRUQ8XFF5FceLKympqKa0soaSympKK2r48NvDGAb0jo9gTJ84rukRq2ArIl5P4VZExINtzS7klgXrqaoxzlgz57rL6JcQ2YhdiYiYR+FWRMSDvfb5PqpqDAL8fQgP9CfY6kewzY8gqy/BNj/6JbSgb7sWZrcpItJoFG5FRDzUgWPFrNqTB8DyXwylQ+sQkzsSETGflgITEfFQf/rqAIYBqZdFKdiKiPxAI7ciIh7AMAz2Hi3m8z1H2ZhZwI4jDo4WVQAwZUh7k7sTEWk6FG5FRJqoo0Xl7DjsYPX3x/h8z1EOnyg7reaqLq0Z1F4Xi4mInKJwKyLSBO3KcXDDa+uodv13FQSrnw8p7VsyrHNrercNp1NUKGGBflreS0TkRxRuRUSaoDe+3O8OtuOT2/GTLlEM7tiSIKv+2hYRORv9LSkiYqLqGhdHiyrIyi9hd46TnUcc7DjiYP+xEgCWPTiEHm3CTe5SRMRzNJnVEtasWcPo0aOJi4vDYrGwdOnSWvsNw2DOnDnExsYSGBhIamoqe/furVVTUFDAhAkTCAsLIyIigilTplBcXFyrZvv27QwdOpSAgADi4+OZP3/+ab0sWbKErl27EhAQQM+ePfn4448vuBcRkXNZ9V0ePZ7+lMHPfc74P23gN8u/Y+nWHHewva5XrIKtiMgFajLhtqSkhN69e/P666/XuX/+/Pm8+uqrLFiwgA0bNhAcHExaWhrl5eXumgkTJrBr1y5WrlzJsmXLWLNmDVOnTnXvdzqdjBgxgoSEBDZv3szzzz/P008/zZtvvumuWb9+PbfddhtTpkxhy5YtjBkzhjFjxrBz584L6kVE5GxcLoPfrthDeZULPx8L8ZGBpHWP5pGfdubtuwbwzZPDeW18X7PbFBHxPEYTBBgfffSR+7nL5TJiYmKM559/3r2tsLDQsNlsxt///nfDMAxj9+7dBmBs3LjRXfPJJ58YFovFOHLkiGEYhvGHP/zBaNGihVFRUeGumTlzptGlSxf381tvvdUYNWpUrX6Sk5ONe++997x7OR8Oh8MADIfDcd7HiIj3+M8uu5Ewc5nRY84Ko7C00ux2ROQ86fu76WsyI7dnk5mZid1uJzU11b0tPDyc5ORk0tPTAUhPTyciIoL+/fu7a1JTU/Hx8WHDhg3ummHDhmG1Wt01aWlpZGRkcOLECXfNj1/nVM2p1zmfXupSUVGB0+ms9RCR5skwDP7w5T4Abk9JIDzQ3+SORES8h0eEW7vdDkB0dHSt7dHR0e59drudqKioWvv9/PyIjIysVVPXOX78Gmeq+fH+c/VSl3nz5hEeHu5+xMfHn+Ndi4i3+vpAAVsOFWL182HyFUlmtyMi4lU8Itx6g1mzZuFwONyP7Oxss1sSkUZWUFLJjsMOXl118gLUW/u3pXWozeSuRES8i0csBRYTEwNAXl4esbGx7u15eXn06dPHXXP06NFax1VXV1NQUOA+PiYmhry8vFo1p56fq+bH+8/VS11sNhs2m77ERJqTovIqNmYVsG5fPuv2HWePvci9z9fHwr3DOpjYnYiId2qwkdt9+/bx6aefUlZ28vaQhmGc44jzl5SURExMDKtWrXJvczqdbNiwgZSUFABSUlIoLCxk8+bN7prPP/8cl8tFcnKyu2bNmjVUVVW5a1auXEmXLl1o0aKFu+bHr3Oq5tTrnE8vItK8HSksY+wf0+nzzEomL9rEn9dmuoNtdJiN9q2CeWREZ+Ijg0zuVETE+9R75DY/P5+xY8fy+eefY7FY2Lt3L+3bt2fKlCm0aNGCF1544bzOU1xczL59+9zPMzMz2bp1K5GRkbRr147p06fzm9/8hk6dOpGUlMTs2bOJi4tjzJgxAFx22WWMHDmSe+65hwULFlBVVcUDDzzAuHHjiIuLA2D8+PHMnTuXKVOmMHPmTHbu3Mkrr7zCSy+95H7dhx56iCuvvJIXXniBUaNGsXjxYjZt2uReLsxisZyzFxFp3uZ9/B0bMgsAaBcZxBUdWzK4QytSOrSkVYj+B0dE5JKq73ILd9xxh5GWlmZkZ2cbISEhxv79+w3DMIwVK1YY3bp1O+/zfPHFFwZw2mPixImGYZxcgmv27NlGdHS0YbPZjOHDhxsZGRm1zpGfn2/cdtttRkhIiBEWFmZMmjTJKCoqqlWzbds2Y8iQIYbNZjPatGljPPfcc6f18sEHHxidO3c2rFar0b17d2P58uW19p9PL+eipUREvM/qjKPG3H/tMpKeWGYkzFxmbMrKN7slEWlg+v5u+iyGUb/5AzExMXz66af07t2b0NBQtm3bRvv27Tlw4AC9evU67Q5hcpLT6SQ8PByHw0FYWJjZ7YhIPe084uC63691P7+6S2venjTQxI5E5FLQ93fTV+9pCSUlJQQFnT5vrKCgQBdQiUizUF5Vw/9b/h0AfdtFcPugBH7aLfocR4mIyKVQ7wvKhg4dyl/+8hf3c4vFgsvlYv78+Vx99dX1Pb2ISJNW4zK44bV1pB/IB2DO6O7c2LctoQG6MYOIiBnqPXI7f/58hg8fzqZNm6isrOTxxx9n165dFBQUsG7duoboUUSkyfrsuzwy8k6uhHDvle3pEx9hbkMiIs1cvUdue/Towffff8+QIUO44YYbKCkp4cYbb2TLli106KA1HEXEu72zPguA+67swKxrLjO3GRERaZibOISHh/Pkk082xKlERDzG93lFrN+fj48Fbh/Uzux2RESEBgi3a9asOev+YcOG1fclRERMZRgGJZU1FJdXU1xRjaOskl05Tj7YdPI22j/tFk3bFrohg4hIU1DvcHvVVVedts1isbh/rqmpqe9LiIiY4oON2bz3zSG+zyuitLLuv8usfj7ce6WmYImINBX1DrcnTpyo9byqqootW7Ywe/Zs/t//+3/1Pb2IiCnynOU8uXQHVTX/XQrc18dCsNWX0AB/kloFc0XHVqR1j6Z96xATOxURkR+rd7gNDw8/bdtPf/pTrFYrM2bMYPPmzfV9CRGRRrdofRZVNQZ94iP43S29aBMRRIC/T63/mRIRkaanQS4oq0t0dDQZGRmX6vQiIpdMSUU17359EICfX9WBjlGhJnckIiLnq97hdvv27bWeG4ZBbm4uzz33HH369Knv6UVEGt0Hm7JxlleT1CqY1Mt0pzEREU9S73Dbp08fLBYLhmHU2j5o0CAWLlxY39OLiDSq6hoXC9dlAjBlSBI+PpqGICLiSeodbjMzM2s99/HxoXXr1gQEBNT31CIije7TXXlkF5TRIsifm/q2NbsdERG5QPUOtwkJCQ3Rh4iI6QzD4E9fHQDgjkEJBFp9Te5IREQu1EWF21dfffW8a3/xi19czEuIiDS6zQdPsDW7EKufD3ekJJrdjoiIXISLCrcvvfTSedVZLBaFWxHxGKdGbW+8vA2tQ20mdyMiIhfjosLt/86zFRHxdJnHS/jP7jwA7h6aZHI3IiJysXzMbkBEpClYuDYTw4CfdI3SurYiIh6sQW7icPjwYf71r39x6NAhKisra+178cUXG+IlREQumRMllSzZnA1o1FZExNPVO9yuWrWK66+/nvbt27Nnzx569OhBVlYWhmHQt2/fhuhRROSS+tvXBymvctE9LoyU9i3NbkdEROqh3tMSZs2axaOPPsqOHTsICAjgH//4B9nZ2Vx55ZXccsstDdGjiMglU15VwzvpJ2+1O3VYeywW3bRBRMST1Tvcfvfdd9x5550A+Pn5UVZWRkhICM888wy//e1v692giMil9K+tORwvriA2PIBre8aa3Y6IiNRTvcNtcHCwe55tbGws+/fvd+87fvx4fU8vInLJ/PimDZOuSMTfV9fYioh4unrPuR00aBBr167lsssu49prr+WRRx5hx44dfPjhhwwaNKghehQRuSS+/P4Ye48WE2LzY9zAdma3IyIiDaDe4fbFF1+kuLgYgLlz51JcXMz7779Pp06dtFKCiDRZ6/cf5+d/+xaAcQPiCQvwN7kjERFpCPUOt88++yy33347cHKKwoIFC+rdlIjIpVRQUsmURZsoq6rBz8fCpCFa/ktExFvUO9weO3aMkSNH0rp1a8aNG8ftt99O7969G6I3EZF6K6us4cDxYrKOl3L4RCkllTXsOFxIWVUNNj8fFk0aSJuIQLPbFBGRBmIxDMOo70lOnDjBkiVLeO+99/jqq6/o2rUrEyZMYPz48SQmJjZAm97H6XQSHh6Ow+EgLCzM7HZEvNKBY8Xc8No6iiqq69z/2vjLua5XXCN3JSKeTN/fTV+DhNsfO3z4MH//+99ZuHAhe/fupbq67i+V5k5/OEQuvZn/t533N2UTavOjU3QI7SKDCAnwI8DPl4RWwdye3E7r2orIBdH3d9PXILffPaWqqopNmzaxYcMGsrKyiI6ObsjTi4ict6NF5Xy05QgAiyYPoF9CpMkdiYhIY2iQRR2/+OIL7rnnHqKjo7nrrrsICwtj2bJlHD58uCFOLyJywf6afpDKGheXt4tQsBURaUbqPXLbpk0bCgoKGDlyJG+++SajR4/GZrM1RG8iIhelrLKGv379wy11h7Y3uRsREWlM9Q63Tz/9NLfccgsREREN0I6ISP0YhsGSzdkUllbRLjKIEd1jzG5JREQaUb3D7T333NMQfYiInLejReVk2Is46qzgSGEZ2QWlHCoo5fCJMvKc5VS7Tl4nO2VIEr4+umBMRKQ5adALykRELrWjReUM/93qMy7vdUrHqBBu6d+2kboSEZGmQuFWRDxGRXUNc/+1m6KKaiKDrVwWG0qbiEDiWwQRH3nyERseQGiAHyE2Py3zJSLSDDXIagmNJTExEYvFctpj2rRpAFx11VWn7bvvvvtqnePQoUOMGjWKoKAgoqKieOyxx05bi/fLL7+kb9++2Gw2OnbsyKJFi07r5fXXXycxMZGAgACSk5P55ptvLtn7FhFwuQxuWZDO8h25ADx/cy/evXsQ82/uzYPDOzHm8jb0S2hBXEQgoQH+CrYiIs2UR4XbjRs3kpub636sXLkSgFtuucVdc88999SqmT9/vntfTU0No0aNorKykvXr1/POO++waNEi5syZ467JzMxk1KhRXH311WzdupXp06dz99138+mnn7pr3n//fWbMmMGvfvUrvv32W3r37k1aWhpHjx5thE9BpPmprHbx4ZYjbD/sAODW/m25ukuUyV2JiEhT1OB3KGtM06dPZ9myZezduxeLxcJVV11Fnz59ePnll+us/+STT7juuuvIyclx32BiwYIFzJw5k2PHjmG1Wpk5cybLly9n586d7uPGjRtHYWEhK1asACA5OZkBAwbw2muvAeByuYiPj+fBBx/kiSeeOK/edYcTkf/KdZSx6rujHDhWwqGCUnIdZTjLqyirrKGkooayqhp37d1Dknjqum4mdisizZm+v5s+jxq5/bHKykr+9re/MXny5Fr//fjuu+/SqlUrevTowaxZsygtLXXvS09Pp2fPnrXunJaWlobT6WTXrl3umtTU1FqvlZaWRnp6uvt1N2/eXKvGx8eH1NRUd01dKioqcDqdtR4iAtU1Lsb/aQNPLd3JwnWZfPZdHrtynGQXlHG8uLJWsG0XGcTdWrdWRETOwmMvKFu6dCmFhYXcdddd7m3jx48nISGBuLg4tm/fzsyZM8nIyODDDz8EwG63n3ZL4FPP7Xb7WWucTidlZWWcOHGCmpqaOmv27Nlzxn7nzZvH3LlzL/r9inirFbvsZB4vISzAj7ED4mnXMpg2EQGEB1oJsvoSZPUlItBKaIAfPlrWS0REzsFjw+2f//xnrrnmGuLi4tzbpk6d6v65Z8+exMbGMnz4cPbv30+HDh3MaNNt1qxZzJgxw/3c6XQSHx9vYkci5jMMgwWr9wMweUgS01M7m9yRiIh4Oo8MtwcPHuSzzz5zj8ieSXJyMgD79u2jQ4cOxMTEnLaqQV5eHgAxMTHuX09t+3FNWFgYgYGB+Pr64uvrW2fNqXPUxWaz6bbEIj/Yll1ITmEZ+44Ws/OIkwB/H+5MSTS7LRER8QIeOef27bffJioqilGjRp21buvWrQDExsYCkJKSwo4dO2qtarBy5UrCwsLo1q2bu2bVqlW1zrNy5UpSUlIAsFqt9OvXr1aNy+Vi1apV7hoRObPl23O54fV13P/ut7yw8nsAxg1oR2Sw1eTORETEG3jcyK3L5eLtt99m4sSJ+Pn9t/39+/fz3nvvce2119KyZUu2b9/Oww8/zLBhw+jVqxcAI0aMoFu3btxxxx3Mnz8fu93OU089xbRp09yjqvfddx+vvfYajz/+OJMnT+bzzz/ngw8+YPny5e7XmjFjBhMnTqR///4MHDiQl19+mZKSEiZNmtS4H4aIh8kuKOWJf2wHTt5BrF1kENFhAfxieCeTOxMREW/hceH2s88+49ChQ0yePLnWdqvVymeffeYOmvHx8dx000089dRT7hpfX1+WLVvG/fffT0pKCsHBwUycOJFnnnnGXZOUlMTy5ct5+OGHeeWVV2jbti1vvfUWaWlp7pqxY8dy7Ngx5syZg91up0+fPqxYseK0i8xE5L/2Hytm+AurAQjw92HJvSm00GitiIg0MI9e59aTaZ08aW5m/t923t+UDcCvx/TgjkEJJnckInLh9P3d9HncyK2IeJ6jznI+2nIEgH/cn0K/hEiTOxIREW/lkReUiYhnWbQ+i8oaF/0SWijYiojIJaVwKyKXVHFFNX/7+iAAU4fp7mIiInJpKdyKyCX1/sZsnOXVJLUK5qeX6aJLERG5tBRuReSSqapxsXBtJgB3D03S7XNFROSSU7gVkUvm4x25HCkso2WwlZv6tjW7HRERaQYUbkXkkjAMgzfXHABg4uBEAvx9Te5IRESaAy0FJiINrsZ1MtjuynES6O+rNW1FRKTRKNyKSIOb/+ke/rj65Kjtrf3b6k5kIiLSaDQtQUQalKOsir+ln1z6q1fbcKb9pKPJHYmISHOikVsRaVDvbzxESWUNXaJD+ee0K7BYtEKCiIg0Ho3cikiDqapx8fa6LACmDElSsBURkUancCsiDebjHbnkOsppFWLjhsvjzG5HRESaIYVbEWkQhmHw5x9u2HBnSgI2Py39JSIijU/hVkQaxMasE2w/7MDm58OE5HZmtyMiIs2Uwq2INIi3vjq59NeNfdvSMsRmcjciItJcKdyKSL1lHS9h5Xd5AEwZkmhuMyIi0qwp3IpIvS1cl4lhwNVdWtMxKtTsdkREpBnTOrcickGqalxsP1zIsaJK9h0tYu2+42zMOgHA3UPbm9ydiIg0dwq3InJB5v57F3/7+tBp29O6RzO4Q0sTOhIREfkvhVsROW+5jjLe35gNQJ/4CNpEBDKoQ0uGdGxFYssg3bRBRERMp3ArIuftT2syqaoxGJgUyQf3ppjdjoiIyGl0QZmInJeCkkr+/s3J6QjTru5ocjciIiJ1U7gVkfOyaH0WZVU1dI8LY1inVma3IyIiUieFWxE5p+KKat5ZnwXAz6/qqLm1IiLSZCncisg5vbfhII6yKtq3CmZkjxiz2xERETkjhVsROauK6hre+ioTgHuvbI+vj0ZtRUSk6VK4FZGz+ufWHI4WVRATFsDPLm9rdjsiIiJnpXArImdkGAYL154ctb3rikSsfvorQ0REmjZ9U4nIGaUfyGePvYhAf1/GDYg3ux0REZFzUrgVkTNauDYLgJv6tSEiyGpuMyIiIudB4VZE6pR1vIRVe/IAuGtwksndiIiInB+FWxGp06L1WRgGXNWlNR2jQsxuR0RE5Lwo3IrIaZzlVSzZlA3A5Cs0aisiIp5D4VZETvPBxmxKKmvoGBXCUN1qV0REPIjCrYjUUuMyWPTDrXYnX5GkW+2KiIhHUbgVkVre++YQh0+UERHkz88ub2N2OyIiIhfEo8Lt008/jcViqfXo2rWre395eTnTpk2jZcuWhISEcNNNN5GXl1frHIcOHWLUqFEEBQURFRXFY489RnV1da2aL7/8kr59+2Kz2ejYsSOLFi06rZfXX3+dxMREAgICSE5O5ptvvrkk71mkMb382ffMXroTgPED2xFo9TW5IxERkQvjUeEWoHv37uTm5rofa9eude97+OGH+fe//82SJUtYvXo1OTk53Hjjje79NTU1jBo1isrKStavX88777zDokWLmDNnjrsmMzOTUaNGcfXVV7N161amT5/O3Xffzaeffuquef/995kxYwa/+tWv+Pbbb+nduzdpaWkcPXq0cT4EkUugqLyKt746eTey3vERTB3W3uSORERELpzFMAzD7CbO19NPP83SpUvZunXrafscDgetW7fmvffe4+abbwZgz549XHbZZaSnpzNo0CA++eQTrrvuOnJycoiOjgZgwYIFzJw5k2PHjmG1Wpk5cybLly9n586d7nOPGzeOwsJCVqxYAUBycjIDBgzgtddeA8DlchEfH8+DDz7IE088UWfvFRUVVFRUuJ87nU7i4+NxOByEhYU1yOcjUh8L12byzLLddIwKYeXDwzTXVkSkDk6nk/DwcH1/N2EeN3K7d+9e4uLiaN++PRMmTODQoUMAbN68maqqKlJTU921Xbt2pV27dqSnpwOQnp5Oz5493cEWIC0tDafTya5du9w1Pz7HqZpT56isrGTz5s21anx8fEhNTXXX1GXevHmEh4e7H/HxupWpNB01LoN30rMAmHRFooKtiIh4LI8Kt8nJySxatIgVK1bwxhtvkJmZydChQykqKsJut2O1WomIiKh1THR0NHa7HQC73V4r2J7af2rf2WqcTidlZWUcP36cmpqaOmtOnaMus2bNwuFwuB/Z2dkX9RmIXAqf7znKwfxSwgP9ufHytma3IyIictH8zG7gQlxzzTXun3v16kVycjIJCQl88MEHBAYGmtjZudlsNmw2m9ltiNTp7XUn59qOGxivi8hERMSjedTI7f+KiIigc+fO7Nu3j5iYGCorKyksLKxVk5eXR0xMDAAxMTGnrZ5w6vm5asLCwggMDKRVq1b4+vrWWXPqHCKeZI/dyfr9+fj6WLgzJdHsdkREROrFo8NtcXEx+/fvJzY2ln79+uHv78+qVavc+zMyMjh06BApKSkApKSksGPHjlqrGqxcuZKwsDC6devmrvnxOU7VnDqH1WqlX79+tWpcLherVq1y14h4kkXrsgBI6x5Nm4im/T8gIiIi5+JR4fbRRx9l9erVZGVlsX79en72s5/h6+vLbbfdRnh4OFOmTGHGjBl88cUXbN68mUmTJpGSksKgQYMAGDFiBN26deOOO+5g27ZtfPrppzz11FNMmzbNPWXgvvvu48CBAzz++OPs2bOHP/zhD3zwwQc8/PDD7j5mzJjBn/70J9555x2+++477r//fkpKSpg0aZIpn4vIxSooqeSjLUcAmHRFksndiIiI1J9Hzbk9fPgwt912G/n5+bRu3ZohQ4bw9ddf07p1awBeeuklfHx8uOmmm6ioqCAtLY0//OEP7uN9fX1ZtmwZ999/PykpKQQHBzNx4kSeeeYZd01SUhLLly/n4Ycf5pVXXqFt27a89dZbpKWluWvGjh3LsWPHmDNnDna7nT59+rBixYrTLjITaer+/s0hKqpd9GgTRv+EFma3IyIiUm8etc6tN9E6eWK2qhoXQ3/7BXZnOS/e2psb+2qVBBGRc9H3d9PnUdMSRKThfLLTjt1ZTqsQG6N6xZrdjoiISINQuBVppk4t/3X7oHbY/LT8l4iIeAeFW5FmaGt2IVsOFWL19WFCcoLZ7YiIiDQYhVuRZujUqO11vWNpHaqbi4iIiPdQuBVpZvKc5SzfngvAZC3/JSIiXsajlgITkYtnGAY5jnJe/2If1S6DgYmR9GgTbnZbIiIiDUrhVsTL7TtazK+X7Wb74UJOlFa5t0+6ItG8pkRERC4RhVsRLzfv4+9Y/f0xAPx8LHSKDmVop1aM6B5jcmciIiINT+FWxAsdL65gU9YJduc4WLXnKAB/nTKQAYmRBPhr2S8REfFeCrciXuSDjdksWL2fA8dLam1PvSyKoZ1am9SViIhI41G4FfES2QWlPPHhdlwGWCzQJTqUnm3C6RITys8ub2N2eyIiIo1C4VbES/z164O4DBjUPpI/3t6f8CB/s1sSERFpdFrnVsQLlFZWs/ibQwBMHdZewVZERJothVsRL/DRliM4y6tJaBnEVZ2jzG5HRETENAq3Ih7OMAwWrcsCYGJKIj4+FnMbEhERMZHm3Ip4KEdpFfuPF/OPzYfZe7SYYKsvN/dva3ZbIiIiplK4FfEg5VU1PPz+Vr7JLCC/pLLWvpv7tSUsQHNtRUSkeVO4FfEg/9x6hE922t3PY8ICSGgZRJsWgTzwk04mdiYiItI0KNyKeAjDMFi4NguAX/ykI/de2YFgm/4Ii4iI/JguKBPxEOv355ORV0SQ1ZcpQ9or2IqIiNRB4VbEQyxcmwmcnFurdWxFRETqpnAr4gEyj5ewas9RAO4anGhuMyIiIk2Ywq1IE5dhL2L8n74GYHjXKNq3DjG5IxERkaZL4VakCatxGUz96yZyHeUATB6SZHJHIiIiTZuuSBFpYgzD4GhRBZnHS1j8zSEO5pcCMP/mXlzRsZXJ3YmIiDRtCrciJtpy6AT/3JrD8eIKCkoqyS+uJPtEKaWVNbXqpl3dgVv7x5vUpYiIiOdQuBUxSXlVDVP/upljRRWn7fP1sdC2RSCJLYPpHB3C/Vd1NKFDERERz6NwK2KSpVuOcKyogugwG1OHdaBlsJXIYCttWgQS3yIIq5+mxIuIiFwohVsRE7hcBm+uOQDAPUPbM0UXiomIiDQIhVuRRpJfXMGuHCcH80tYs/c4B46XEBrgx7iB7cxuTURExGso3Io0gn1Hi7j21bVUVrtqbb9jUAIhuo2uiIhIg9G3qkgjeHPNASqrXUSF2ugdH0HLYCsdo0KYkJxgdmsiIiJeReFWpIFU17g4WlRBrqOMnMJyso6XsCvHyc4cB4dPlAHwxu396JfQwuRORUREvJfCrchFMgyDF1d+z/r9+eQUlnG0qIIal3HG+lG9YhVsRURELjGFW5GLtHbfcX7/+b5a2/x9LUSHBRAXHkjbFoF0iwujW1wY3WPDCQ/yN6lTERGR5kPhVuQsqmtcZJ8oY29eEQeOl3CipJLC0iqKK6rZdrgQgBv6xHHX4ETaRATSKsSGj4/F3KZFRESaMY9aJX7evHkMGDCA0NBQoqKiGDNmDBkZGbVqrrrqKiwWS63HfffdV6vm0KFDjBo1iqCgIKKionjssceorq6uVfPll1/St29fbDYbHTt2ZNGiRaf18/rrr5OYmEhAQADJycl88803Df6epXFUVrvILihl9ffH+MOX+5j27rf89MXVXDZnBVf/7kum/nUzz32yhz+uOcD7m7JZviOXwyfKsPr58FhaFy5v14KosAAFWxEREZN51Mjt6tWrmTZtGgMGDKC6uppf/vKXjBgxgt27dxMcHOyuu+eee3jmmWfcz4OCgtw/19TUMGrUKGJiYli/fj25ubnceeed+Pv78+yzzwKQmZnJqFGjuO+++3j33XdZtWoVd999N7GxsaSlpQHw/vvvM2PGDBYsWEBycjIvv/wyaWlpZGRkEBUV1UifiJyvDHsRH245zDFnBceKK3CWVVFUUU1xeTXFFdWUVtac8Vibnw8dWofQKTqE1iE2IoL8CQ3wJ8TmR7e4MNq2CDrjsSIiItK4LIZhnPkKmCbu2LFjREVFsXr1aoYNGwacHLnt06cPL7/8cp3HfPLJJ1x33XXk5OQQHR0NwIIFC5g5cybHjh3DarUyc+ZMli9fzs6dO93HjRs3jsLCQlasWAFAcnIyAwYM4LXXXgPA5XIRHx/Pgw8+yBNPPHHO3p1OJ+Hh4TgcDsLCwurzMcg5uFwGw19cTebxkrPWWf18aNsikO5x4fSIC6NrbBgdWgcTFx6oEVkREQH0/e0JPGrk9n85HA4AIiMja21/9913+dvf/kZMTAyjR49m9uzZ7tHb9PR0evbs6Q62AGlpadx///3s2rWLyy+/nPT0dFJTU2udMy0tjenTpwNQWVnJ5s2bmTVrlnu/j48PqamppKen19lrRUUFFRUV7udOp/Pi37hckFV7jpJ5vISwAD/uv6ojrUKstAiyEhrgR0iAH6E2f0ID/IgI8sdiUYgVERHxZB4bbl0uF9OnT+eKK66gR48e7u3jx48nISGBuLg4tm/fzsyZM8nIyODDDz8EwG631wq2gPu53W4/a43T6aSsrIwTJ05QU1NTZ82ePXvq7HfevHnMnTu3fm9aLspbXx0A4Lbkdtx/VQeTuxEREZFLyWPD7bRp09i5cydr166ttX3q1Knun3v27ElsbCzDhw9n//79dOhgXrCZNWsWM2bMcD93Op3Ex8eb1k9zsfOIgw2ZBfj5WLhrcKLZ7YiIiMgl5pHh9oEHHmDZsmWsWbOGtm3bnrU2OTkZgH379tGhQwdiYmJOW9UgLy8PgJiYGPevp7b9uCYsLIzAwEB8fX3x9fWts+bUOf6XzWbDZrOd/5uUBrFwbSYA1/aMJTY80ORuRERE5FLzqKXADMPggQce4KOPPuLzzz8nKSnpnMds3boVgNjYWABSUlLYsWMHR48eddesXLmSsLAwunXr5q5ZtWpVrfOsXLmSlJQUAKxWK/369atV43K5WLVqlbtGzJfnLOff23MAmDLk3L9XRERExPN51MjttGnTeO+99/jnP/9JaGioe45seHg4gYGB7N+/n/fee49rr72Wli1bsn37dh5++GGGDRtGr169ABgxYgTdunXjjjvuYP78+djtdp566immTZvmHlm97777eO2113j88ceZPHkyn3/+OR988AHLly939zJjxgwmTpxI//79GThwIC+//DIlJSVMmjSp8T8YqdNf0rOoqjEYkNiC3vERZrcjIiIijcHwIECdj7ffftswDMM4dOiQMWzYMCMyMtKw2WxGx44djccee8xwOBy1zpOVlWVcc801RmBgoNGqVSvjkUceMaqqqmrVfPHFF0afPn0Mq9VqtG/f3v0aP/b73//eaNeunWG1Wo2BAwcaX3/99Xm/F4fDYQCn9SYNo7Si2ug991MjYeYy45MdOWa3IyIiXkLf302fR69z68m0Tt6l9e6Ggzz50U7iIwP58tGr8dU6tSIi0gD0/d30edS0BJFzMQwDu7PcfSHZpMFJCrYiIiLNiMKteLRjRRX8a1sO39uL+P5oEfvyiimqqAYg1ObHrQO03JqIiEhzonArHu3Bv3/L1wcKam3z9bGQ1CqY+67sQIhNv8VFRESaE33zS5NV4zLIKSzjaFE5FVUuKqpdVFTXnPy1ykWes5yvDxTg72vhvis70Dk6lM7RoSS1Csbq51Gr3ImIiEgDUbiVJmnxN4f41b92UVHtOmftmD5teGREl0boSkRERJo6hVtpcsoqa5j/aQYV1S6svj5Eh9sI9PfF5ueLzc+HAP+Tv9r8fQgPtPLwTzuZ3bKIiIg0EQq30uT83+ZsCkoqiY8MZNWMqzTFQERERM6bUoM0KTUugz//sIzX3UPaK9iKiIjIBVFykCZl1Xd5ZOWXEh7ozy3925rdjoiIiHgYhVtpUt766uSo7YTkdgRZNWtGRERELozCrTQZ27IL+Sbr5NJeEwcnmt2OiIiIeCCFW2ky3vphru3o3nFEhwWY3I2IiIh4Iv2/r5jq0112Pt1l52B+KVsOnQBOXkgmIiIicjEUbsU0uY4ypr37LdUuw70t9bIousWFmdiViIiIeDKFWzFFWWUNj//fdqpdBt3jwrj/qg4ktgyma0yo2a2JiIiIB1O4FVPc85dNrN13HIBfDO9EWvcYkzsSERERb6ALyqTRbTl0wh1sn7z2MkZ0iza5IxEREfEWGrmVRrdwXRYAN/Vtyz3DdPGYiIiINByN3EqjynWU8fGOXAAmD0k0txkRERHxOgq30qj+kn6QGpfBoPaRdI8LN7sdERER8TIKt9JoSiureW/DIQAmX5FkcjciIiLijRRupdF8+O0RHGVVJLQMYvhluohMREREGp7CrTQKl8tg4bqTt9e9a3Aivj4WkzsSERERb6RwK41i9d5jHDhWQqjNj1v6x5vdjoiIiHgphVtpFAvXnhy1HTsgnhCbVqATERGRS0MpQxqEs7yKQ/mlHD5RxuETJ3/NdZSR56zgWFEFRwrL8LHAxMGJZrcqIiIiXkzhVi6IYRjkOMrZk+tk5xEnu3Ic7MpxcqSw7JzH3tyvLfGRQY3QpYiIiDRXCrdyRtU1LvYdK+b7vGK+PXiCrdmF7DtaTHFFdZ31rUKstGkRRNsWgcS3CCIuIoDosACiQm1EhwUQGx7QyO9AREREmhuFW6GgpJKPd+SSfaKUo84KCksrKSitYm9eEaWVNafV+/lY6NA6hO5xYXSLC6NHm3C6xYURFuBvQvciIiIi/6VwKzy0eAtf7T1e574Qmx+dokPo3TaCvgktuCwmlISWwVj9dC2iiIiIND0Kt83crhwHX+09jq+PhTsGJRAbHkCLICthgf4ktQqmY1SI1qQVERERj6Fw28y99dXJJbqu7RnL09d3N7kbERERkfrR/y03Y7mOMv69LQeAe4YmmdyNiIiISP0p3DZji9ZlUe0ySE6KpFfbCLPbEREREak3hdtmKsNexJ9/uGvY1GHtTe5GREREpGFozm0zUVXjYlPWCb7IOMrne46y72gxAO1bB3N1lyiTuxMRERFpGAq39fT666/z/PPPY7fb6d27N7///e8ZOHCgaf2UVFRzorSSymoX3+cVsSvHyc4jDjYdPEFR+X9vvuDrY6F323DmXt8DH62GICIiIl5C4bYe3n//fWbMmMGCBQtITk7m5ZdfJi0tjYyMDKKizBkNXbk7j+nvb61zX4sgf67uEsXVXaMY1qk14UG66YKIiIh4F4XbenjxxRe55557mDRpEgALFixg+fLlLFy4kCeeeMKUnnx8LNj8fPCxWGjfOpgeceF0bxNGr7YR9GwTrjVrRURExKsp3F6kyspKNm/ezKxZs9zbfHx8SE1NJT09/bT6iooKKioq3M+dTucl6ev63nFc3zvukpxbREREpKnTagkX6fjx49TU1BAdHV1re3R0NHa7/bT6efPmER4e7n7Ex8c3VqsiIiIizYbCbSOZNWsWDofD/cjOzja7JRERERGvo2kJF6lVq1b4+vqSl5dXa3teXh4xMTGn1dtsNmw2W2O1JyIiItIsaeT2IlmtVvr168eqVavc21wuF6tWrSIlJcXEzkRERESaL43c1sOMGTOYOHEi/fv3Z+DAgbz88suUlJS4V08QERERkcalcFsPY8eO5dixY8yZMwe73U6fPn1YsWLFaReZiYiIiEjjsBiGYZjdRHPkdDoJDw/H4XAQFhZmdjsiIiJyHvT93fRpzq2IiIiIeA2FWxERERHxGgq3IiIiIuI1FG5FRERExGso3IqIiIiI11C4FRERERGvoXArIiIiIl5DN3EwyanlhZ1Op8mdiIiIyPk69b2t2wQ0XQq3JikqKgIgPj7e5E5ERETkQhUVFREeHm52G1IH3aHMJC6Xi5ycHEJDQ7FYLA16bqfTqdAsIiICZGdnN+idxAzDoKioiLi4OHx8NLuzKVK49UKnbg0oIiLS3Ok2uc2P/skhIiIiIl5D4VZEREREvIYuKPNCNpuNmTNnsnbtWlJSUvD19T1jbXV1NV9//fVZ686npiHP5e2v58m967Nquq/nyb3rs2q6r+fpvfv5+WGz2c64X7yT5tyKiIiIiNfQtAQRERER8RoKtyIiIiLiNRRuRURERMRreN0FZQ19QwQREREx36nv91OXCp3p+SkXU3exNeHh4VRWVuJyuWjduvUZ30N+fv45a8637mJr+vbty9KlS8/6+p7ugi8os9vt/PKXv2TJkiUUFxdfqr5EREREpBnz8fHh97//PT//+c8v6LgLCrdZWVkkJye7/yUA//3XjIiIiIjIhYqIiKCwsND9fMqUKbz77rv4+/tTVlbGhx9+yOjRo8/7fBc05/bnP/85DocDPz8/rFYrVqv1Qg4XERERESEgIMD9c2FhIYGBgcDJtfqvu+46KioqeOSRRwgICOC3v/3tBZ37vEduCwoKaNmy5QWdXERERETkQiQlJdGmTRt8fHxwOp3s2rWLkpIS/P39z+v48x653bdv30U3KSIiIiJyNqfCa2ZmJmvXrsUwDHJzc6mqquL48ePnfZ7zDreaWysiIiIil8LNN99MVVVVrW3p6elERUUBJy8uO1+aliAiIiIipvrxMmun/O1vf+P2228nJCQEh8Nx3gH3vGNwZGQkI0eOxGaz1XqIiIiIiJxJXXNlT11ABidHZa+88kpCQ0MB6Ny5Mz4+PrRp0waA1NTUSzNyC3DgwAGSk5M5ceKElgITERERkXoLCgoiMDCQ/Px8AFq2bEnXrl3ZvXs3TqeTffv2kZiYeN7nu6ClwNq3b8/27du54447CAoKUrAVERERkXopLS11B1s4eWe19evXExwczOeff35BwRYu4g5l0nB0q2ARERE5Xx999BFjxowxu40m74JGbkVEREREmjKFWxERERHxGn5mN9CcaUaIiIiISMPSyK2IiIiIeA2FWxERERHxGgq3IiIiIuI1FG5FRERExGso3IqIiIiI11C4FRGph6ysLCwWC1u3bj3vY+66665zLsR+1VVXMX369Hr1JiLSHGkpMBGReoiPjyc3N5dWrVqZ3YqIiKBwKyJy0SorK7FarcTExJjdioiI/EDTEkSkWXjzzTeJi4vD5XLV2n7DDTcwefJk9u/fzw033EB0dDQhISEMGDCAzz77rFZtYmIiv/71r7nzzjsJCwtj6tSpp01LqKmpYcqUKSQlJREYGEiXLl145ZVX6uxp7ty5tG7dmrCwMO677z4qKyvP2H9FRQWPPvoobdq0ITg4mOTkZL788st6fSYiIt5I4VZEmoVbbrmF/Px8vvjiC/e2goICVqxYwYQJEyguLubaa69l1apVbNmyhZEjRzJ69GgOHTpU6zy/+93v6N27N1u2bGH27NmnvY7L5aJt27YsWbKE3bt3M2fOHH75y1/ywQcf1KpbtWoV3333HV9++SV///vf+fDDD5k7d+4Z+3/ggQdIT09n8eLFbN++nVtuuYWRI0eyd+/een4yIiLexWLoHrAi0kyMGTOGli1b8uc//xk4OZo7d+5csrOz8fE5/d/6PXr04L777uOBBx4ATo7cXn755Xz00UfumqysLJKSktiyZQt9+vSp83UfeOAB7HY7//d//wecvKDs3//+N9nZ2QQFBQGwYMECHnvsMRwOBz4+Plx11VX06dOHl19+mUOHDtG+fXsOHTpEXFyc+7ypqakMHDiQZ599tkE+HxERb6CRWxFpNiZMmMA//vEPKioqAHj33XcZN24cPj4+FBcX8+ijj3LZZZcRERFBSEgI33333Wkjt/379z/n67z++uv069eP1q1bExISwptvvnnaeXr37u0OtgApKSkUFxeTnZ192vl27NhBTU0NnTt3JiQkxP1YvXo1+/fvv5iPQkTEa+mCMhFpNkaPHo1hGCxfvpwBAwbw1Vdf8dJLLwHw6KOPsnLlSn73u9/RsWNHAgMDufnmm0+bBxscHHzW11i8eDGPPvooL7zwAikpKYSGhvL888+zYcOGi+67uLgYX19fNm/ejK+vb619ISEhF31eERFvpHArIs1GQEAAN954I++++y779u2jS5cu9O3bF4B169Zx11138bOf/Qw4GSizsrIu+DXWrVvH4MGD+fnPf+7eVtfo6rZt2ygrKyMwMBCAr7/+mpCQEOLj40+rvfzyy6mpqeHo0aMMHTr0gnsSEWlONC1BRJqVCRMmsHz5chYuXMiECRPc2zt16sSHH37I1q1b2bZtG+PHjz9tZYXz0alTJzZt2sSnn37K999/z+zZs9m4ceNpdZWVlUyZMoXdu3fz8ccf86tf/YoHHnigzrm/nTt3ZsKECdx55518+OGHZGZm8s033zBv3jyWL19+wT2KiHgzhVsRaVZ+8pOfEBkZSUZGBuPHj3dvf/HFF2nRogWDBw9m9OjRpKWluUd1L8S9997LjTfeyNixY0lOTiY/P7/WKO4pw4cPp1OnTgwbNoyxY8dy/fXX8/TTT5/xvG+//TZ33nknjzzyCF26dGHMmDFs3LiRdu3aXXCPIiLeTKsliIiIiIjX0MitiIiIiHgNhVsRERER8RoKtyIiIiLiNRRuRURERMRrKNyKiIiIiNdQuBURERERr6FwKyIiIiJeQ+FWRERERLyGwq2IiIiIeA2FWxERERHxGgq3IiIiIuI1/j+dmLXBnmy5DQAAAABJRU5ErkJggg==",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "sns.lineplot(data=df_melt[df_melt['username']==np.unique(df_melt['username'])[4]],x='variable',y='value')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "id": "1a7b8b8a",
   "metadata": {},
   "outputs": [],
   "source": [
    "df.to_csv('simulated_data.csv')"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "0decc78c",
   "metadata": {},
   "source": [
    "# now simulate researcher data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "id": "4c7d90f7",
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "15it [00:00, 35.82it/s]"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "(15, 12)\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>username</th>\n",
       "      <th>icon</th>\n",
       "      <th>Trial name</th>\n",
       "      <th>First name</th>\n",
       "      <th>Last name</th>\n",
       "      <th>email</th>\n",
       "      <th>Address</th>\n",
       "      <th>Latitude</th>\n",
       "      <th>Longitude</th>\n",
       "      <th>Postcode</th>\n",
       "      <th>Telephone number</th>\n",
       "      <th>Department</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>mereOwl9</td>\n",
       "      <td>../../assets/staff_icons/administrator_2.png</td>\n",
       "      <td>MindfullAI</td>\n",
       "      <td>Charlotte</td>\n",
       "      <td>Treadwell</td>\n",
       "      <td>charlotte_treadwell@mindfulai.com</td>\n",
       "      <td>Studio 0\\nNatasha street\\nPort Derektown\\nTW0R...</td>\n",
       "      <td>53.69786</td>\n",
       "      <td>-2.68758</td>\n",
       "      <td>N7 0BF</td>\n",
       "      <td>(0113)4960410</td>\n",
       "      <td>Administrator</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>excitedHyena4</td>\n",
       "      <td>../../assets/staff_icons/administrator_2.png</td>\n",
       "      <td>MindfullAI</td>\n",
       "      <td>Catherine</td>\n",
       "      <td>Cross</td>\n",
       "      <td>catherine_cross@mindfulai.com</td>\n",
       "      <td>Flat 44Y\\nGraham flat\\nSouth Naomi\\nW0A 1FR</td>\n",
       "      <td>51.35084</td>\n",
       "      <td>-1.99421</td>\n",
       "      <td>NE6B 6GA</td>\n",
       "      <td>(028) 9018 0419</td>\n",
       "      <td>Administrator</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>zestyRaisins7</td>\n",
       "      <td>../../assets/staff_icons/developer_1.png</td>\n",
       "      <td>MindfullAI</td>\n",
       "      <td>Matthew</td>\n",
       "      <td>Souders</td>\n",
       "      <td>matthew_souders@mindfulai.com</td>\n",
       "      <td>Flat 88\\nManning walks\\nWest Iainberg\\nS92 3RL</td>\n",
       "      <td>52.68333</td>\n",
       "      <td>0.93333</td>\n",
       "      <td>DY4 4XW</td>\n",
       "      <td>+44909 8790043</td>\n",
       "      <td>Developer</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>similarSeagull2</td>\n",
       "      <td>../../assets/staff_icons/healthcare_2.png</td>\n",
       "      <td>MindfullAI</td>\n",
       "      <td>Edward</td>\n",
       "      <td>Park</td>\n",
       "      <td>edward_park@mindfulai.com</td>\n",
       "      <td>1 Barnett extension\\nWest Mohammedville\\nYO8E 8GY</td>\n",
       "      <td>50.75767</td>\n",
       "      <td>-1.5443</td>\n",
       "      <td>LN83 6BT</td>\n",
       "      <td>0113 496 0415</td>\n",
       "      <td>Healthcare</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>sadOwl1</td>\n",
       "      <td>../../assets/staff_icons/administrator_1.png</td>\n",
       "      <td>MindfullAI</td>\n",
       "      <td>Kelsey</td>\n",
       "      <td>Simmons</td>\n",
       "      <td>kelsey_simmons@mindfulai.com</td>\n",
       "      <td>4 Rachel unions\\nNorth Joelhaven\\nBB97 2HF</td>\n",
       "      <td>53.78333</td>\n",
       "      <td>-1.06667</td>\n",
       "      <td>W19 4TJ</td>\n",
       "      <td>0113 496 0055</td>\n",
       "      <td>Administrator</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "          username                                          icon  Trial name  \\\n",
       "0         mereOwl9  ../../assets/staff_icons/administrator_2.png  MindfullAI   \n",
       "1    excitedHyena4  ../../assets/staff_icons/administrator_2.png  MindfullAI   \n",
       "2    zestyRaisins7      ../../assets/staff_icons/developer_1.png  MindfullAI   \n",
       "3  similarSeagull2     ../../assets/staff_icons/healthcare_2.png  MindfullAI   \n",
       "4          sadOwl1  ../../assets/staff_icons/administrator_1.png  MindfullAI   \n",
       "\n",
       "  First name  Last name                              email  \\\n",
       "0  Charlotte  Treadwell  charlotte_treadwell@mindfulai.com   \n",
       "1  Catherine      Cross      catherine_cross@mindfulai.com   \n",
       "2    Matthew    Souders      matthew_souders@mindfulai.com   \n",
       "3     Edward       Park          edward_park@mindfulai.com   \n",
       "4     Kelsey    Simmons       kelsey_simmons@mindfulai.com   \n",
       "\n",
       "                                             Address  Latitude Longitude  \\\n",
       "0  Studio 0\\nNatasha street\\nPort Derektown\\nTW0R...  53.69786  -2.68758   \n",
       "1        Flat 44Y\\nGraham flat\\nSouth Naomi\\nW0A 1FR  51.35084  -1.99421   \n",
       "2     Flat 88\\nManning walks\\nWest Iainberg\\nS92 3RL  52.68333   0.93333   \n",
       "3  1 Barnett extension\\nWest Mohammedville\\nYO8E 8GY  50.75767   -1.5443   \n",
       "4         4 Rachel unions\\nNorth Joelhaven\\nBB97 2HF  53.78333  -1.06667   \n",
       "\n",
       "   Postcode Telephone number     Department  \n",
       "0    N7 0BF    (0113)4960410  Administrator  \n",
       "1  NE6B 6GA  (028) 9018 0419  Administrator  \n",
       "2   DY4 4XW   +44909 8790043      Developer  \n",
       "3  LN83 6BT    0113 496 0415     Healthcare  \n",
       "4   W19 4TJ    0113 496 0055  Administrator  "
      ]
     },
     "execution_count": 16,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "researcher_count = 15\n",
    "# Name, Department (rightnow values: Clinician, IT, Developer), Email, Avatar (icon), Address, Phone\n",
    "\n",
    "df_researcher = pd.DataFrame(generate_username(researcher_count),columns=['username'])\n",
    "df_researcher['icon']=''\n",
    "\n",
    "df_researcher['Trial name']='MindfullAI'\n",
    "\n",
    "#constitutional data\n",
    "df_researcher['First name']=''\n",
    "df_researcher['Last name']=''\n",
    "df_researcher['email']=''\n",
    "df_researcher['Address']=''\n",
    "df_researcher['Latitude']=0\n",
    "df_researcher['Longitude']=0\n",
    "df_researcher['Postcode']=''\n",
    "df_researcher['Telephone number']=''\n",
    "df_researcher['Department']=''\n",
    "\n",
    "for i, row in tqdm(df_researcher.iterrows()):\n",
    "\n",
    "    df_researcher.loc[i,'First name']=names.get_first_name()\n",
    "    df_researcher.loc[i,'Last name']=names.get_last_name()\n",
    "    df_researcher.loc[i,'email']=df_researcher.loc[i,'First name'].lower()+'_'+df_researcher.loc[i,'Last name'].lower()+'@mindfulai.com'\n",
    "    \n",
    "    fake = Faker('en-UK')\n",
    "    df_researcher.loc[i,'Address']=fake.address()\n",
    "    \n",
    "    country='US'\n",
    "    while country != 'GB':\n",
    "        result = fake.location_on_land()\n",
    "        country = result[3]\n",
    "    \n",
    "    df_researcher.loc[i,'Latitude']=result[0]\n",
    "    df_researcher.loc[i,'Longitude']=result[1]\n",
    "    df_researcher.loc[i,'Postcode']=fake.postcode()\n",
    "    df_researcher.loc[i,'Telephone number']=fake.phone_number()\n",
    "    \n",
    "    df_researcher.loc[i,'Department']=np.random.choice(['Healthcare','Developer','Administrator'])\n",
    "    staff_icons = glob.glob('../../assets/staff_icons/*'+df_researcher.loc[i,'Department'].lower()+'*')\n",
    "    df_researcher.loc[i,'icon']=staff_icons[random.randint(0, len(staff_icons)-1)]\n",
    "        \n",
    "print(df_researcher.shape)\n",
    "df_researcher.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "id": "c6f9b3a4",
   "metadata": {},
   "outputs": [],
   "source": [
    "df_researcher.to_csv('simulated_staff_data.csv')"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3 (ipykernel)",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.11.4"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}