diff --git a/.ipynb_checkpoints/Among_Us_Prediction-checkpoint.ipynb b/.ipynb_checkpoints/Among_Us_Prediction-checkpoint.ipynb
new file mode 100644
index 0000000..4d2dd90
--- /dev/null
+++ b/.ipynb_checkpoints/Among_Us_Prediction-checkpoint.ipynb
@@ -0,0 +1,1349 @@
+{
+ "cells": [
+  {
+   "cell_type": "code",
+   "execution_count": 1,
+   "id": "02949a29",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "import numpy as np\n",
+    "import pandas as pd\n",
+    "import matplotlib.pyplot as plt\n",
+    "import seaborn as sns"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 2,
+   "id": "bcdba050",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "files = []\n",
+    "for i in range(29):\n",
+    "    tempdf = pd.read_csv(f'data/User{i+1}.csv', delimiter = ',')\n",
+    "    tempdf['User'] = i\n",
+    "    files.append(tempdf)\n",
+    "df = pd.concat(files, ignore_index = True)"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 3,
+   "id": "25f68e6f",
+   "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>Game Completed Date</th>\n",
+       "      <th>Team</th>\n",
+       "      <th>Outcome</th>\n",
+       "      <th>Task Completed</th>\n",
+       "      <th>All Tasks Completed</th>\n",
+       "      <th>Murdered</th>\n",
+       "      <th>Imposter Kills</th>\n",
+       "      <th>Game Length</th>\n",
+       "      <th>Ejected</th>\n",
+       "      <th>Sabotages Fixed</th>\n",
+       "      <th>Time to complete all tasks</th>\n",
+       "      <th>Rank Change</th>\n",
+       "      <th>Region/Game Code</th>\n",
+       "      <th>User</th>\n",
+       "    </tr>\n",
+       "  </thead>\n",
+       "  <tbody>\n",
+       "    <tr>\n",
+       "      <th>0</th>\n",
+       "      <td>12/13/2020 at 1:26:56 am EST</td>\n",
+       "      <td>Crewmate</td>\n",
+       "      <td>Win</td>\n",
+       "      <td>3</td>\n",
+       "      <td>No</td>\n",
+       "      <td>Yes</td>\n",
+       "      <td>-</td>\n",
+       "      <td>07m 04s</td>\n",
+       "      <td>No</td>\n",
+       "      <td>2.0</td>\n",
+       "      <td>-</td>\n",
+       "      <td>++</td>\n",
+       "      <td>NA / WYMSBF</td>\n",
+       "      <td>0</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>1</th>\n",
+       "      <td>12/13/2020 at 1:17:42 am EST</td>\n",
+       "      <td>Crewmate</td>\n",
+       "      <td>Loss</td>\n",
+       "      <td>7</td>\n",
+       "      <td>Yes</td>\n",
+       "      <td>No</td>\n",
+       "      <td>-</td>\n",
+       "      <td>16m 21s</td>\n",
+       "      <td>No</td>\n",
+       "      <td>1.0</td>\n",
+       "      <td>09m 48s</td>\n",
+       "      <td>--</td>\n",
+       "      <td>NA / WYMSBF</td>\n",
+       "      <td>0</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>2</th>\n",
+       "      <td>12/13/2020 at 12:57:47 am EST</td>\n",
+       "      <td>Crewmate</td>\n",
+       "      <td>Win</td>\n",
+       "      <td>3</td>\n",
+       "      <td>No</td>\n",
+       "      <td>No</td>\n",
+       "      <td>-</td>\n",
+       "      <td>11m 33s</td>\n",
+       "      <td>No</td>\n",
+       "      <td>0.0</td>\n",
+       "      <td>-</td>\n",
+       "      <td>++</td>\n",
+       "      <td>NA / WYMSBF</td>\n",
+       "      <td>0</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>3</th>\n",
+       "      <td>12/13/2020 at 12:41:55 am EST</td>\n",
+       "      <td>Imposter</td>\n",
+       "      <td>Win</td>\n",
+       "      <td>-</td>\n",
+       "      <td>-</td>\n",
+       "      <td>-</td>\n",
+       "      <td>2</td>\n",
+       "      <td>08m 05s</td>\n",
+       "      <td>No</td>\n",
+       "      <td>NaN</td>\n",
+       "      <td>-</td>\n",
+       "      <td>+++</td>\n",
+       "      <td>Europe / QIRTNF</td>\n",
+       "      <td>0</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>4</th>\n",
+       "      <td>12/13/2020 at 12:30:37 am EST</td>\n",
+       "      <td>Crewmate</td>\n",
+       "      <td>Loss</td>\n",
+       "      <td>4</td>\n",
+       "      <td>No</td>\n",
+       "      <td>No</td>\n",
+       "      <td>-</td>\n",
+       "      <td>05m 10s</td>\n",
+       "      <td>No</td>\n",
+       "      <td>0.0</td>\n",
+       "      <td>-</td>\n",
+       "      <td>---</td>\n",
+       "      <td>Europe / QIRTNF</td>\n",
+       "      <td>0</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>...</th>\n",
+       "      <td>...</td>\n",
+       "      <td>...</td>\n",
+       "      <td>...</td>\n",
+       "      <td>...</td>\n",
+       "      <td>...</td>\n",
+       "      <td>...</td>\n",
+       "      <td>...</td>\n",
+       "      <td>...</td>\n",
+       "      <td>...</td>\n",
+       "      <td>...</td>\n",
+       "      <td>...</td>\n",
+       "      <td>...</td>\n",
+       "      <td>...</td>\n",
+       "      <td>...</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>2222</th>\n",
+       "      <td>12/01/2020 at 11:07:41 am EST</td>\n",
+       "      <td>Imposter</td>\n",
+       "      <td>Loss</td>\n",
+       "      <td>-</td>\n",
+       "      <td>-</td>\n",
+       "      <td>-</td>\n",
+       "      <td>2</td>\n",
+       "      <td>14m 10s</td>\n",
+       "      <td>No</td>\n",
+       "      <td>NaN</td>\n",
+       "      <td>-</td>\n",
+       "      <td>-</td>\n",
+       "      <td>NA / SNNGZF</td>\n",
+       "      <td>28</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>2223</th>\n",
+       "      <td>12/01/2020 at 10:52:25 am EST</td>\n",
+       "      <td>Crewmate</td>\n",
+       "      <td>Win</td>\n",
+       "      <td>7</td>\n",
+       "      <td>Yes</td>\n",
+       "      <td>No</td>\n",
+       "      <td>-</td>\n",
+       "      <td>14m 11s</td>\n",
+       "      <td>No</td>\n",
+       "      <td>1.0</td>\n",
+       "      <td>14m 10s</td>\n",
+       "      <td>++</td>\n",
+       "      <td>NA / SNNGZF</td>\n",
+       "      <td>28</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>2224</th>\n",
+       "      <td>11/26/2020 at 11:23:14 am EST</td>\n",
+       "      <td>Crewmate</td>\n",
+       "      <td>Win</td>\n",
+       "      <td>7</td>\n",
+       "      <td>Yes</td>\n",
+       "      <td>No</td>\n",
+       "      <td>-</td>\n",
+       "      <td>19m 45s</td>\n",
+       "      <td>No</td>\n",
+       "      <td>1.0</td>\n",
+       "      <td>15m 16s</td>\n",
+       "      <td>++</td>\n",
+       "      <td>Europe / NZWLXQ</td>\n",
+       "      <td>28</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>2225</th>\n",
+       "      <td>11/26/2020 at 11:00:36 am EST</td>\n",
+       "      <td>Crewmate</td>\n",
+       "      <td>Win</td>\n",
+       "      <td>7</td>\n",
+       "      <td>Yes</td>\n",
+       "      <td>No</td>\n",
+       "      <td>-</td>\n",
+       "      <td>10m 18s</td>\n",
+       "      <td>No</td>\n",
+       "      <td>0.0</td>\n",
+       "      <td>06m 13s</td>\n",
+       "      <td>++</td>\n",
+       "      <td>Europe / NZWLXQ</td>\n",
+       "      <td>28</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>2226</th>\n",
+       "      <td>11/26/2020 at 10:48:50 am EST</td>\n",
+       "      <td>Crewmate</td>\n",
+       "      <td>Win</td>\n",
+       "      <td>7</td>\n",
+       "      <td>Yes</td>\n",
+       "      <td>No</td>\n",
+       "      <td>-</td>\n",
+       "      <td>19m 37s</td>\n",
+       "      <td>No</td>\n",
+       "      <td>1.0</td>\n",
+       "      <td>16m 02s</td>\n",
+       "      <td>++</td>\n",
+       "      <td>Europe / NZWLXQ</td>\n",
+       "      <td>28</td>\n",
+       "    </tr>\n",
+       "  </tbody>\n",
+       "</table>\n",
+       "<p>2227 rows × 14 columns</p>\n",
+       "</div>"
+      ],
+      "text/plain": [
+       "                Game Completed Date      Team Outcome Task Completed  \\\n",
+       "0      12/13/2020 at 1:26:56 am EST  Crewmate     Win              3   \n",
+       "1      12/13/2020 at 1:17:42 am EST  Crewmate    Loss              7   \n",
+       "2     12/13/2020 at 12:57:47 am EST  Crewmate     Win              3   \n",
+       "3     12/13/2020 at 12:41:55 am EST  Imposter     Win              -   \n",
+       "4     12/13/2020 at 12:30:37 am EST  Crewmate    Loss              4   \n",
+       "...                             ...       ...     ...            ...   \n",
+       "2222  12/01/2020 at 11:07:41 am EST  Imposter    Loss              -   \n",
+       "2223  12/01/2020 at 10:52:25 am EST  Crewmate     Win              7   \n",
+       "2224  11/26/2020 at 11:23:14 am EST  Crewmate     Win              7   \n",
+       "2225  11/26/2020 at 11:00:36 am EST  Crewmate     Win              7   \n",
+       "2226  11/26/2020 at 10:48:50 am EST  Crewmate     Win              7   \n",
+       "\n",
+       "     All Tasks Completed Murdered Imposter Kills Game Length Ejected  \\\n",
+       "0                     No      Yes              -     07m 04s      No   \n",
+       "1                    Yes       No              -     16m 21s      No   \n",
+       "2                     No       No              -     11m 33s      No   \n",
+       "3                      -        -              2     08m 05s      No   \n",
+       "4                     No       No              -     05m 10s      No   \n",
+       "...                  ...      ...            ...         ...     ...   \n",
+       "2222                   -        -              2     14m 10s      No   \n",
+       "2223                 Yes       No              -     14m 11s      No   \n",
+       "2224                 Yes       No              -     19m 45s      No   \n",
+       "2225                 Yes       No              -     10m 18s      No   \n",
+       "2226                 Yes       No              -     19m 37s      No   \n",
+       "\n",
+       "      Sabotages Fixed Time to complete all tasks Rank Change Region/Game Code  \\\n",
+       "0                 2.0                          -          ++      NA / WYMSBF   \n",
+       "1                 1.0                    09m 48s          --      NA / WYMSBF   \n",
+       "2                 0.0                          -          ++      NA / WYMSBF   \n",
+       "3                 NaN                          -         +++  Europe / QIRTNF   \n",
+       "4                 0.0                          -         ---  Europe / QIRTNF   \n",
+       "...               ...                        ...         ...              ...   \n",
+       "2222              NaN                          -           -      NA / SNNGZF   \n",
+       "2223              1.0                    14m 10s          ++      NA / SNNGZF   \n",
+       "2224              1.0                    15m 16s          ++  Europe / NZWLXQ   \n",
+       "2225              0.0                    06m 13s          ++  Europe / NZWLXQ   \n",
+       "2226              1.0                    16m 02s          ++  Europe / NZWLXQ   \n",
+       "\n",
+       "      User  \n",
+       "0        0  \n",
+       "1        0  \n",
+       "2        0  \n",
+       "3        0  \n",
+       "4        0  \n",
+       "...    ...  \n",
+       "2222    28  \n",
+       "2223    28  \n",
+       "2224    28  \n",
+       "2225    28  \n",
+       "2226    28  \n",
+       "\n",
+       "[2227 rows x 14 columns]"
+      ]
+     },
+     "execution_count": 3,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "df"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 4,
+   "id": "e6f11cc9",
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "text/plain": [
+       "Game Completed Date            object\n",
+       "Team                           object\n",
+       "Outcome                        object\n",
+       "Task Completed                 object\n",
+       "All Tasks Completed            object\n",
+       "Murdered                       object\n",
+       "Imposter Kills                 object\n",
+       "Game Length                    object\n",
+       "Ejected                        object\n",
+       "Sabotages Fixed               float64\n",
+       "Time to complete all tasks     object\n",
+       "Rank Change                    object\n",
+       "Region/Game Code               object\n",
+       "User                            int64\n",
+       "dtype: object"
+      ]
+     },
+     "execution_count": 4,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "df.dtypes"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 5,
+   "id": "888d9bdd",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "df[['Task Completed', 'Imposter Kills']] = df[['Task Completed', 'Imposter Kills']].apply(pd.to_numeric, errors = 'coerce')"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 6,
+   "id": "127740d3",
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "text/plain": [
+       "NA         1436\n",
+       "Europe      791\n",
+       "Name: Region, dtype: int64"
+      ]
+     },
+     "execution_count": 6,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "df['Region'] = df['Region/Game Code'].str.extract(r'^(.*?)/')\n",
+    "df['Region'].value_counts()\n"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 7,
+   "id": "8cb080c6",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "df['Game Length'] = df['Game Length'].apply(pd.to_timedelta)\n",
+    "df['Game Length'] = df['Game Length'] / np.timedelta64(1, 's')"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 8,
+   "id": "44941ec1",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "df['Time to complete all tasks'] = df['Time to complete all tasks'].apply(pd.to_timedelta, errors = 'coerce')\n",
+    "df['Time to complete all tasks'] = df['Time to complete all tasks'] / np.timedelta64(1, 's')"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 9,
+   "id": "e84a07c1",
+   "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>Team</th>\n",
+       "      <th>Outcome</th>\n",
+       "      <th>Task Completed</th>\n",
+       "      <th>All Tasks Completed</th>\n",
+       "      <th>Murdered</th>\n",
+       "      <th>Game Length</th>\n",
+       "      <th>Ejected</th>\n",
+       "      <th>Sabotages Fixed</th>\n",
+       "      <th>Region</th>\n",
+       "      <th>User</th>\n",
+       "    </tr>\n",
+       "  </thead>\n",
+       "  <tbody>\n",
+       "    <tr>\n",
+       "      <th>0</th>\n",
+       "      <td>Crewmate</td>\n",
+       "      <td>Win</td>\n",
+       "      <td>3.0</td>\n",
+       "      <td>No</td>\n",
+       "      <td>Yes</td>\n",
+       "      <td>424.0</td>\n",
+       "      <td>No</td>\n",
+       "      <td>2.0</td>\n",
+       "      <td>NA</td>\n",
+       "      <td>0</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>1</th>\n",
+       "      <td>Crewmate</td>\n",
+       "      <td>Loss</td>\n",
+       "      <td>7.0</td>\n",
+       "      <td>Yes</td>\n",
+       "      <td>No</td>\n",
+       "      <td>981.0</td>\n",
+       "      <td>No</td>\n",
+       "      <td>1.0</td>\n",
+       "      <td>NA</td>\n",
+       "      <td>0</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>2</th>\n",
+       "      <td>Crewmate</td>\n",
+       "      <td>Win</td>\n",
+       "      <td>3.0</td>\n",
+       "      <td>No</td>\n",
+       "      <td>No</td>\n",
+       "      <td>693.0</td>\n",
+       "      <td>No</td>\n",
+       "      <td>0.0</td>\n",
+       "      <td>NA</td>\n",
+       "      <td>0</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>4</th>\n",
+       "      <td>Crewmate</td>\n",
+       "      <td>Loss</td>\n",
+       "      <td>4.0</td>\n",
+       "      <td>No</td>\n",
+       "      <td>No</td>\n",
+       "      <td>310.0</td>\n",
+       "      <td>No</td>\n",
+       "      <td>0.0</td>\n",
+       "      <td>Europe</td>\n",
+       "      <td>0</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>5</th>\n",
+       "      <td>Crewmate</td>\n",
+       "      <td>Loss</td>\n",
+       "      <td>7.0</td>\n",
+       "      <td>Yes</td>\n",
+       "      <td>Yes</td>\n",
+       "      <td>982.0</td>\n",
+       "      <td>No</td>\n",
+       "      <td>0.0</td>\n",
+       "      <td>Europe</td>\n",
+       "      <td>0</td>\n",
+       "    </tr>\n",
+       "  </tbody>\n",
+       "</table>\n",
+       "</div>"
+      ],
+      "text/plain": [
+       "       Team Outcome  Task Completed All Tasks Completed Murdered  Game Length  \\\n",
+       "0  Crewmate     Win             3.0                  No      Yes        424.0   \n",
+       "1  Crewmate    Loss             7.0                 Yes       No        981.0   \n",
+       "2  Crewmate     Win             3.0                  No       No        693.0   \n",
+       "4  Crewmate    Loss             4.0                  No       No        310.0   \n",
+       "5  Crewmate    Loss             7.0                 Yes      Yes        982.0   \n",
+       "\n",
+       "  Ejected  Sabotages Fixed   Region  User  \n",
+       "0      No              2.0      NA      0  \n",
+       "1      No              1.0      NA      0  \n",
+       "2      No              0.0      NA      0  \n",
+       "4      No              0.0  Europe      0  \n",
+       "5      No              0.0  Europe      0  "
+      ]
+     },
+     "execution_count": 9,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "crewmate = df[['Team',\n",
+    "               'Outcome',\n",
+    "               'Task Completed',\n",
+    "               'All Tasks Completed',\n",
+    "               'Murdered',\n",
+    "               'Game Length',\n",
+    "               'Ejected',\n",
+    "              'Sabotages Fixed',\n",
+    "              'Region',\n",
+    "              'User']][df['Team'] == 'Crewmate']\n",
+    "crewmate.head()"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 10,
+   "id": "13385be6",
+   "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>Team</th>\n",
+       "      <th>Outcome</th>\n",
+       "      <th>Task Completed</th>\n",
+       "      <th>All Tasks Completed</th>\n",
+       "      <th>Murdered</th>\n",
+       "      <th>Game Length</th>\n",
+       "      <th>Ejected</th>\n",
+       "      <th>Sabotages Fixed</th>\n",
+       "      <th>Region</th>\n",
+       "      <th>User</th>\n",
+       "    </tr>\n",
+       "  </thead>\n",
+       "  <tbody>\n",
+       "    <tr>\n",
+       "      <th>0</th>\n",
+       "      <td>Crewmate</td>\n",
+       "      <td>1</td>\n",
+       "      <td>3.0</td>\n",
+       "      <td>0</td>\n",
+       "      <td>1</td>\n",
+       "      <td>424.0</td>\n",
+       "      <td>0</td>\n",
+       "      <td>2.0</td>\n",
+       "      <td>0</td>\n",
+       "      <td>0</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>1</th>\n",
+       "      <td>Crewmate</td>\n",
+       "      <td>0</td>\n",
+       "      <td>7.0</td>\n",
+       "      <td>1</td>\n",
+       "      <td>0</td>\n",
+       "      <td>981.0</td>\n",
+       "      <td>0</td>\n",
+       "      <td>1.0</td>\n",
+       "      <td>0</td>\n",
+       "      <td>0</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>2</th>\n",
+       "      <td>Crewmate</td>\n",
+       "      <td>1</td>\n",
+       "      <td>3.0</td>\n",
+       "      <td>0</td>\n",
+       "      <td>0</td>\n",
+       "      <td>693.0</td>\n",
+       "      <td>0</td>\n",
+       "      <td>0.0</td>\n",
+       "      <td>0</td>\n",
+       "      <td>0</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>4</th>\n",
+       "      <td>Crewmate</td>\n",
+       "      <td>0</td>\n",
+       "      <td>4.0</td>\n",
+       "      <td>0</td>\n",
+       "      <td>0</td>\n",
+       "      <td>310.0</td>\n",
+       "      <td>0</td>\n",
+       "      <td>0.0</td>\n",
+       "      <td>1</td>\n",
+       "      <td>0</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>5</th>\n",
+       "      <td>Crewmate</td>\n",
+       "      <td>0</td>\n",
+       "      <td>7.0</td>\n",
+       "      <td>1</td>\n",
+       "      <td>1</td>\n",
+       "      <td>982.0</td>\n",
+       "      <td>0</td>\n",
+       "      <td>0.0</td>\n",
+       "      <td>1</td>\n",
+       "      <td>0</td>\n",
+       "    </tr>\n",
+       "  </tbody>\n",
+       "</table>\n",
+       "</div>"
+      ],
+      "text/plain": [
+       "       Team  Outcome  Task Completed  All Tasks Completed  Murdered  \\\n",
+       "0  Crewmate        1             3.0                    0         1   \n",
+       "1  Crewmate        0             7.0                    1         0   \n",
+       "2  Crewmate        1             3.0                    0         0   \n",
+       "4  Crewmate        0             4.0                    0         0   \n",
+       "5  Crewmate        0             7.0                    1         1   \n",
+       "\n",
+       "   Game Length  Ejected  Sabotages Fixed  Region  User  \n",
+       "0        424.0        0              2.0       0     0  \n",
+       "1        981.0        0              1.0       0     0  \n",
+       "2        693.0        0              0.0       0     0  \n",
+       "4        310.0        0              0.0       1     0  \n",
+       "5        982.0        0              0.0       1     0  "
+      ]
+     },
+     "execution_count": 10,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "crewmate['Outcome'] = crewmate['Outcome'].replace(['Loss', 'Win'],[0, 1])\n",
+    "crewmate['All Tasks Completed'] = crewmate['All Tasks Completed'].replace(['No', 'Yes'],[0, 1])\n",
+    "crewmate['Murdered'] = crewmate['Murdered'].replace(['No', 'Yes'],[0, 1])\n",
+    "crewmate['Ejected'] = crewmate['Ejected'].replace(['No', 'Yes'],[0, 1])\n",
+    "crewmate['Region'] = crewmate['Region'].replace(['NA ', 'Europe '],[0, 1])\n",
+    "crewmate.head()"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 11,
+   "id": "342da2cd",
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "image/png": "iVBORw0KGgoAAAANSUhEUgAABBwAAANSCAYAAADGbt2XAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjQuMywgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy/MnkTPAAAACXBIWXMAAAsTAAALEwEAmpwYAADbx0lEQVR4nOzdd3hUZd6H8ftJ6EhPIBS72EVRQQEVFQHFuljA3nXtotjQRdeua1t776tiexUVRYqiFAXFVbC7rkhLQkJvAsnz/pEIREgyGzJp3p/rmis55zyT+Q6HM5n85ikhxogkSZIkSVJ5SqnsAJIkSZIkqeax4CBJkiRJksqdBQdJkiRJklTuLDhIkiRJkqRyZ8FBkiRJkiSVOwsOkiRJkiSp3FlwkCRJkiSpBgshPBlCyA4hTC3meAgh3BtC+CmE8FUIYdfyeFwLDpIkSZIk1WxPAweWcPwgoH3h7SzgofJ4UAsOkiRJkiTVYDHGj4C5JTQ5HHg2FvgEaBpCaL2hj1trQ39AAmIFPIYkSZIkqXKEyg6QTCtzfq7yf9PWSd/ybAp6Jvzu0Rjjo//Dj2gLTF9re0bhvtkbkqsiCg6szPm5Ih5GpaidtgWHb3JIZcdQoTd/fZsTN+1b2TEEPDftdWrVaVvZMQSsWjGTE7wuqoTnp73OlmnlMnxTG+g/OZPp0a5XZcdQoVEz3vf9VBXx5q9vc8DGvSs7hoCR04dXdoQ/vcLiwv9SYKgQDqmQJEmSJOnPbSaw8Vrb7Qr3bRALDpIkSZIk/bkNBU4qXK1iT2BBjHGDhlNABQ2pkCRJkiSpWsrPq+wEGyyE8CKwL5AWQpgBXAvUBogxPgwMA/oAPwFLgVPL43EtOEiSJEmSVIPFGI8t5XgEzivvx3VIhSRJkiRJKnf2cJAkSZIkqTgxv7ITVFv2cJAkSZIkSeXOgoMkSZIkSSp3DqmQJEmSJKk4+Q6pKCt7OEiSJEmSpHJnwUGSJEmSJJU7h1RIkiRJklSM6CoVZWYPB0mSJEmSVO4sOEiSJEmSpHLnkApJkiRJkorjKhVlZg8HSZIkSZJU7iw4SJIkSZKkcmfBQZIkSZIklTvncJAkSZIkqTgui1lm9nCQJEmSJEnlzoKDJEmSJEkqdw6pkCRJkiSpOPl5lZ2g2rKHgyRJkiRJKncWHCRJkiRJUrlLaEhFCKEBcCmwSYzxzBBCe2CbGOPbSU0nSZIkSVJlcpWKMku0h8NTwG9Al8LtmcCNSUkkSZIkSZKqvUQLDlvGGG8HVgLEGJcCIWmpJEmSJElStZboKhUrQgj1gQgQQtiSgh4PkiRJkiTVXPkOqSirRAsO1wLvARuHEP4FdANOSVYoSZIkSZJUvSVUcIgxjgghTAb2pGAoxUUxxpykJpMkSZIkSdVWoj0cANoCqYX32SeEQIzx9eTEkiRJkiSp8kVXqSizRJfFfBLoAHwN/P6vHQELDpIkSZIkaR2J9nDYM8a4fVKTSJIkSZKkGiPRgsOEEML2McZvkppGkiRJkqSqxFUqyizRgsOzFBQdMilYDjMAMcbYIWnJJEmSJElStZVoweEJ4ERgCmvmcJAkSZIkSVqvRAsOc2KMQ5OaRJIkSZIk1RiJFhy+CCG8ALxFwZAKAJfFlCRJkiTVbC6LWWaJFhzqU1Bo6LXWPpfFlCRJkiRJ65WSSKMY46nruZ2W7HAV4Zqb72Kfg/tzxAl/Xe/xGCM33/0QBx1zGn856Ry++f6n1cfeHDaCPv1Op0+/03lz2IiKilyjdey+Kw9+8DAPf/QoR5571DrHa9WpxWUPXM7DHz3KP968k5btWhY5ntYmnZe+fYUjzvpLRUWusXbq3pHbR9/HHWMe4JBz1v33rFWnFufdfyl3jHmA6964lbR26QCk1krlrDsv4Obhd3PrqHs59Ny+FR29Rrr7ruv57puxTP58BB132XGd4/Xr12PoG88ydcoYvvz3aG6+6arVx/beaw8mfvoey5dOo2/fgysydo3ToXtH/jH6Pu4c8wCHFnNdnH//pdy5nuvi7Dsv4Jbhd3Ob10W5GXzzZYye+CbvjBnCDh22XW+bSwedx9gvh/HVL2OL7K9Tpzb3Pn4roye+yWvDn6Htxq0rInKN1Wnf3Xl6zBM8O/Yp+p/Xb53jtevU5poHB/Hs2Ke4/617adWu1epjx57Xn2fHPsXTY55g9+67VWTsGsn3UlVHp31356kPH+eZj5+i/7nHrHP89+vimY+f4r6h/1x9XTRu2og7htzOW9+9wfk3nFfRsaVylVDBIYTQLoTwfyGE7MLbayGEdskOVxGO6NOTh++6sdjjH0+YxK8zZjFsyBNcd/mF3HDH/QAsWLiIh556gRcfu4cXH7uHh556gQULF1VU7BopJSWFs288h7+ffC3n9ziXvQ/rzsbtNy7Spme/XixesIS/7nMWQx9/k5OvOqXI8dMHn8HkDz+vwNQ1U0hJ4eQbzuQfJ9/IFQdcRJfD9qZN+6KXfPd+B7BkwWIGdj+P9554i35XngRA54O7UrtObQb1HsDggwey33G9Vv/RpbI56MD9ab/V5my7/V6cc84VPHD/Lettd9fdD7PjTt3ZvVNvunbpxIG99wPg1+kzOf2MAbz40hsVmLrm+f26uP3kG7n8gIvYcz3Xxb6F18WlhddF/7Wui1p1anNV7wH87eCB7O91scH2PaAbm22xCft3PpyrL7mR6/9x1XrbjRr+EX/pddI6+48+/ggWzF/I/p0P56mH/8UV116U7Mg1VkpKChfeeD5XnXg1p+13Jvsfvi+btt+kSJuD+h/I4gWLOWmvU3ntsdc5c9DpAGzafhP2O7w7p+9/FleecDUX3XQBKSkJvT3VevhequpISUnhghvPY9BJ13D6/mey3+H7sck610VvFs1fzMl7n8prj6+5Llb8toKn73iGR258rDKia33y86r+rYpK9BX9KWAo0Kbw9lbhvmpv9112oknjRsUe/2DsJxx2YA9CCOy843YsWrSYOTlzGffp53Tp1JEmjRvRpHEjunTqyLhPfXHeEO132ZrMX2aT9WsWq1au4uO3PqJzrz2LtNmj156MfnUUAOOGjaVDt52LHMv6NZNff/i1QnPXRFvushVZv8xmzvQs8lau4pO3xrJbz85F2uzasxNjX/sAgInDJrBDt52Agl5BdRvUJSU1hTr16rBq5SqWLVpW4c+hJjn00N48969XAfh04mSaNG1CRkbRT6SWLVvOh2PGA7By5UomfzGFtm0LPrGdNm0GU6Z8S75rSG+QRK+Lj9dzXeB1Ue4OOGhf/u/ltwH49+dTaNykEemt0tZp9+/PpzAnK2e993/9pYL7vzt0FF327pTcwDXYtrtsw8xfZjH710xWrVzFB2+OoWuvrkXadO3VhfdfKegNOuadj9h1r46F+7vywZtjWLliJZnTM5n5yyy23WWbCn8ONYXvpaqObXbZhllrXRcfDv2Qbr26FGnTtVcX3n+14Lr46J2P6dhtFwCWL/uNqZO+ZsVvKyo6tlTuEi04pMcYn4oxriq8PQ38KT6ayZqTS0bLNW9gWrVMI2tODllzcshoueafoFV6wX6VXYuMFuTMmrN6O3d2Di1atSjSpvlabfLz8lmyaCmNmjWmXoN69D3nKF6658UKzVxTNctowdzZuau3587OpVlG8yJtmme0IHdWQZv8vHyWLlrKRs0aMWnYBH5b+hv3TXqCeyY8yruPvsmSBYsrNH9N07ZNBjOmz1q9PXPGbNq2ySi2fZMmjTnk4J6M/mBssW30v0vkumiW0YK567kuJhZeF/cXXhfDvC42WKvWLZk1M2v1duasbDJaJ/7WJKN1OrNnZgKQl5fHooWLada8aXnH/FNIa53GnNlrfn/PyZxDWuuiv7/TMtLInr3W7++FS2jcrDFprVsUuW9OZg5prdctHCkxvpeqOtIyWpC91rmYMzuHFhlF/2+3yEhjTpFzUXBdSDVJopNG5oYQTgB+fwU6Fsgtob1UofoPOI6hT7zB8qXLKzvKn94Wu7QnPz+fCzufQcMmG3HNKzcydexXzJmeVfqdtcFSU1P513MPcP8DT/Lf//oJVVXx+3VxQeF18TevC0lVjO+lpBK4SkWZJVpwOA24D7ibgtUpxgOnFtc4hHAWcBbAI488wql9D9jAmJWnVXoLMrPX9FzIys6hVXoardLTmPTFV2v2z8mhU8cOlRGxxsjNzCWtzZpPp1q0TiM3q2hda25hm9zMXFJSU2jYqAGL5i1k647b0LVPN06+6lQaNm5IjJEVv61k2DNvV/TTqBHmZebSfK1Pp5q3bsG8zLlF2szNzKVFmxbMKzwXDRo1YPG8RXQ9fG+++vAL8lblsTB3AT98/h2bd9jSP6z+R+f89WROP/14AD777N+027jN6mNt27Vm5qzM9d7v4Ydu58ef/su99z1eITn/TBK5LuZl5tK8TQvmJnBdbOF18T874bRj6HdiwUR2U/79NW3atuL3wYwZbVqSudYn5aXJnD2H1m0zyJydTWpqKo0ab8S8ufPLP/SfQM7sHNLX6l2SnpFOzuyiv79zMnNo2TqdnNk5Bb+/Gzdk4byF5MzOLXLftIw0cmbbY7SsfC9VdeRk5tJyrXOR3jqN3Myi/7dzM3NIb5NOTmbhddGo4LqQapJEV6mYFmM8LMaYHmNsGWM8IsZY7EdnMcZHY4y7xxh3P+uss8ovbSXYd689GfreKGKMfDn1WzbaqCHpac3ptsdujJ84mQULF7Fg4SLGT5xMtz2cWXlD/PjlD7TevA0tN25Frdq12PvQfZg44tMibSaO+JT9j+oBQLc+e/HV+IKiz6CjruCsbqdzVrfTeevJobx6/8v+gtwAP3/5ExmbtyZ945ak1q7FnofuxeQRk4q0+WLkJPY6smBSws59uvDN+CkA5MzMYfuuBePW69avy1Ydt2b2f2ZW7BOoAR56+Bl279SL3Tv1YujQ4Zx4fMFM43t03pWFCxaSmZm9zn2u//vlNGnSiEsuvbai4/4pJHJdTB45ib3Xc13kzsxhhz9cF7O8Lv5nzz/5MofudyyH7ncs7w/7kL8ccwgAu+y2E4sWLl7vXA3FGfXeGPr2L7j/QYf1YMLHk0q5h4rz3Zff03bztmRsnEGt2rXY7/DujB8xoUibCSMm0OvongB0P3gfvhj3bwDGj5jAfod3p3ad2mRsnEHbzdvy3b+/r+inUGP4Xqrq+P7L72m7WVsyCs/Fvofty/gRnxRpM37EJ/Q6quC62Ofgvfn3uC8rI6qUVAn1cAghPANcFGOcX7jdDLizJiyNedm1tzLpi6+YP38hPY44gXNPP5FVq1YB0O8vB7NPl058PGESBx1zGvXr1eOGQQMAaNK4EWefciz9zyiY1fqvpx5X4uSTKl1+Xj6P/u1hrnvuelJSUxg1ZATTf/iV4y45np+m/MjEERMZMeR9BtxzKQ9/9CiL5i/mjvNvq+zYNVJ+Xj7PDn6cy54dTEpqCh+9PIqZP06n7yX9+e9X/+GLkZMYM2QUf737Iu4Y8wCL5y/mgfPvAmDks+9y1h3nc8uIewgh8NEro5n+3bRKfkbV27B3R3Hggfvz/bfjWLpsGWecccnqY59Nep/dO/WibdvWDLrqIr797kcmTRwOwIMPPsWTT73I7rvtzKuvPEGzZk045OCeXDv4UnbeZf/KejrVVn5ePs8MfpzLC6+LMYXXxZGF18Xkta6LOwuvi/sLr4sRhdfFrV4X5ebDEWPZ94C9GD3pTZYvW84VF163+thbH7zIofsdC8AV117EoUceSP0G9Rj71bu8/Pwb3Hv7I7z8rze488EbGD3xTebPX8BFZ65/lQuVLj8vn/v+dj+3/etmUlJSeHfIcKb9MI1TBp7E91/+wIQRnzDspfe46p9X8OzYp1g0fxE3nnszANN+mMaHb33Ek6MfIy8vj/uuud8JbjeA76WqjoLr4gFuff5mUlJTeG/I+0z7YRonX3oSP3xVcF28+9J7XHnP5TzzccF1cdN5N6++//Pjn6FBo4bUrl2Lbr27cMXxg/j1R4dKVhpfl8osxBhLbxTCFzHGjqXtK0ZcmfNzWfOpHNVO24LDNzmksmOo0Ju/vs2Jm/at7BgCnpv2OrXqtK3sGAJWrZjJCV4XVcLz015ny7RdKzuGgP/kTKZHu16VHUOFRs143/dTVcSbv77NARv3ruwYAkZOHx4qO0My/fb1qNL/aK5kdXfoUSXPQaKrVKQU9moAIITQnMTnf5AkSZIkSX8yiRYN7gQmhBBeKdw+Gri5hPaSJEmSJFV/rlJRZgkVHGKMz4YQPgN+H/TbN8b4TfJiSZIkSZKk6izRSSOfizGeCHyznn2SJEmSJElFJDqHww5rb4QQUgHXgJQkSZIkSetVYg+HEMJVwCCgfghhIfD7zJcrgEeTnE2SJEmSpMrlsphlVmIPhxjjLTHGRsA/YoyNY4yNCm8tYowuWC1JkiRJktYr0VUq3g0h7PPHnTHGj8o5jyRJkiRJqgESLThcttb39YDOwOesWbVCkiRJkqQaJ8a8yo5QbSW6LOaha2+HEDYG7klGIEmSJEmSVP0lukrFH80AtivPIJIkSZIkqeZIqIdDCOE+IBZupgAdgcnJCiVJkiRJUpUQXaWirBKdw+EbILXw+/nAizHGcUlJJEmSJEmSqr0SCw4hhFrAzcBpwK+FuzcBngwhTIwxrkxyPkmSJEmSVA2V1sPhH0AjYPMY4yKAEEJj4I7C20XJjSdJkiRJUiXKd0hFWZU2aeQhwJm/FxsAYowLgXOAPskMJkmSJEmSqq/SCg4xxhjXszOPNZNISpIkSZIkFVHakIpvQggnxRifXXtnCOEE4LvkxZIkSZIkqQpwlYoyK63gcB7wegjhNODzwn27A/WBvyQzmCRJkiRJqr5KLDjEGGcCe4QQ9gd2KNw9LMY4KunJJEmSJElStVVaDwcAYoyjgdFJziJJkiRJUtWSn1fZCaqt0iaNlCRJkiRJ+p9ZcJAkSZIkSeXOgoMkSZIkSSp3Cc3hIEmSJEnSn5LLYpaZPRwkSZIkSVK5s+AgSZIkSZLKnUMqJEmSJEkqTr5DKsrKHg6SJEmSJKncWXCQJEmSJEnlziEVkiRJkiQVx1UqysweDpIkSZIkqdxZcJAkSZIkSeXOIRWSJEmSJBXHVSrKzB4OkiRJkiSp3FlwkCRJkiRJ5c4hFZIkSZIkFcchFWVmDwdJkiRJklTuLDhIkiRJkqRyZ8FBkiRJkiSVO+dwkCRJkiSpGDHmVXaEaivEGJP9GEl/AEmSJElSpQmVHSCZln30dJX/m7b+PqdUyXNQIT0cDt/kkIp4GJXizV/fZmXOz5UdQ4Vqp23BsicGVnYMAfVPv4MLN+tX2TEE3PvLEBZfclhlxxCw0V1D2a5l58qOIeDb7Ikcu+kRlR1DhV6c9gZnbnZ0ZccQ8Ngvr7BJ850qO4aAX+dOqewIqqIcUiFJkiRJUnFcFrPMnDRSkiRJkiSVOwsOkiRJkiSp3DmkQpIkSZKk4kSHVJSVPRwkSZIkSVK5s+AgSZIkSZLKXbFDKkIIU4Bi1xuNMXZISiJJkiRJkqoKV6kos5LmcDik8Ot5hV+fK/x6fPLiSJIkSZKkmqDYgkOMcRpACKFnjLHjWoeuDCFMBq5MdjhJkiRJklQ9JbJKRQghdIsxjivc6IpzP0iSJEmS/gxcpaLMEik4nA48GUJoUrg9HzgtaYkkSZIkSVK1V2rBIcb4ObDz7wWHGOOCpKeSJEmSJEnVWqlDI0IIrUIITwAvxRgXhBC2DyGcXgHZJEmSJElSNZXIXAxPA8OBNoXbPwAXJymPJEmSJElVR35+1b9VUYkUHNJijC8D+QAxxlVAXlJTSZIkSZKkai2RgsOSEEILIAKEEPYEnMdBkiRJkiQVK5FVKi4BhgJbhhDGAenA0UlNJUmSJElSVeCymGWWSMHha6A7sA0QgO9JrGeEJEmSJEn6k0qkcDAhxrgqxvh1jHFqjHElMCHZwSRJkiRJUvVVbA+HEEIG0BaoH0LoSEHvBoDGQIMKyCZJkiRJUuWqwqtAVHUlDanoDZwCtAPuWmv/QmBQEjNJkiRJkqRqrtiCQ4zxGeCZEMKRMcbXKjCTJEmSJEmq5hKZNHJcCOEJoE2M8aAQwvZAlxjjE0nOJkmSJElS5XJIRZklMmnkU8BwoE3h9g/AxckKJEmSJEmSqr9ECg5pMcaXgXyAGOMqIC+pqSRJkiRJUrWWyJCKJSGEFkAECCHsCSxIaipJkiRJkqqC6JCKskqk4HAJMBTYMoQwDkgHjkpqKkmSJEmSVK2VWnCIMU4OIXQHtgEC8H2McWXSk0mSJEmSpGqr2IJDCKFvMYe2DiEQY3w9SZkkSZIkSaoaXKWizErq4XBoCcciYMFBkiRJkiStV7EFhxjjqRUZRJIkSZIk1RylLosZQmgRQrg3hDA5hPB5COGfhatWSJIkSZIkrVciq1S8BHwEHFm4fTwwBDggWaEkSZIkSaoSXBazzBIpOLSOMd6w1vaNIYR+yQokSZIkSZKqv1KHVADvhxD6hxBSCm/HAMOTHUySJEmSJFVfifRwOBO4GHi+cDsFWBJCOBuIMcbGScomSZIkSVLlclnMMiu14BBjbFQRQSRJkiRJUs2RSA8HQggdgM3Wbh9jfD1JmSRJkiRJUjVXasEhhPAk0AH4Gvi9L0kELDhIkiRJkmo2V6kos0R6OOwZY9w+6UkqScfuu3LmdWeRkprCiJfe57UHXy1yvFadWgy4+xK23GkrFs1bxD/Ou43sGdmrj6e1Sef+UQ/y0t0v8Maj/1fR8WuUa26+i4/GTaR5s6a88fzD6xyPMXLLPQ/z8YRJ1KtXl5uuvpTtt9kKgDeHjeCRZ14C4OyT+3N4n54Vmr0mGvdzNrePmkp+jPylwyactmf7Isf/MWoqk6bnArB8ZR5zl/7G2IsOAmD2wqX8/b0vyVq4nBDgvqP2oG2TBhX+HGqK7brvTN/Bp5CSmsKEIaMZ+dCbRY5v2Xk7+g4+mTbbbsIzF/yTf7/7KQDN2qZxxiMDCSmB1FqpfPTMe4z718jKeAo1Ruq2u1L3iDMgJZWVn7zPytGvrdOm1s7dqNP7WCKQP+u//Pb8nWsO1q1PgyseYNXUT1nx+iMVF7yGGnTTpexzQFeWL1vOoAuu55sp36/TZvsO23LLvYOpW78uH40cz81XF5yPbXdsz3W3X0mdenXJW5XH9VfcxpQvvqnop1Bj7Ny9IyddewYpqSl88NIIhj5U9HOpWnVqce5dF7P5TluyeN4i/nn+HeTMyKbbEftwyFl/Wd1uk+02ZdDBlzLtm/9W9FOoMXbovgv9B59KSmoKHw8ZxXsPvVHkePvO29Fv8Cm023ZTHr3gHia/+8nqYxc9czVbdGzPT5O+477Tb63g5DXT32+5kv167s2yZcu59LxrmPrVt+u0uezqCziy/2E0adKY7TbZY/X+M849iWNP7MuqVXnMzZnLwAsGM3PG7IqMrxokhHAg8E8gFXg8xnjrH45vAjwDNC1sc2WMcdiGPm4iq1RMCCHUyIJDSkoKZ994Dn8/+VrO73Euex/WnY3bb1ykTc9+vVi8YAl/3ecshj7+JidfdUqR46cPPoPJH35egalrriP69OThu24s9vjHEybx64xZDBvyBNddfiE33HE/AAsWLuKhp17gxcfu4cXH7uGhp15gwcJFFRW7RsrLj9wycgoPHL0Hr5++H+99O4v/5BT9N72sx468fEp3Xj6lO8fuujk9tm69+tg17/ybkztvxf+dsR/Pn7g3zRvUqeinUGOElMDR15/Gw6fcws09L2G3w7qRsVXbIm3mzcrhXwMf5PM3xxXZvzB7Hnf3vYbb+1zBnUdczQHnHE7jls0qMn7NElKo2/dslj36d5bedh61dt2H0Kro74yQ1praPY5m6X1XsOz281nxxuNFjtc56Hjyfv66IlPXWPv06MqmW2zMgXscybWX3sLg269Yb7trb7+CwZfezIF7HMmmW2zM3vt3AWDg4At44I7H6bv/Cdx32yMMHHxBRcavUUJKCqfecDa3nXw9Aw+4gK6H7U3b9u2KtNmvX0+WLFjMgO7nMOyJoRx35UkAjHvjI67qM4Cr+gzgwQH3MGd6tsWGDRBSUjju+tP55yk3MbjnADof1o3WWxU9F3Nn5fDUwAeY+ObYde4//JE3eWLAfRUVt8bb74C92WzLTdln94O5csDfuenOa9bbbuTwMRx2wLHr7P/6q285eP/+9N77SN4ZOoJBf78k2ZFVQ4UQUoEHgIOA7YFj1/M3/jXAyzHGjkB/4MHyeOxECg7PUlB0+D6E8FUIYUoI4avyePDK1n6Xrcn8ZTZZv2axauUqPn7rIzr32rNImz167cnoV0cBMG7YWDp027nIsaxfM/n1h18rNHdNtfsuO9GkcfFzlH4w9hMOO7AHIQR23nE7Fi1azJycuYz79HO6dOpIk8aNaNK4EV06dWTcpxaBNsTU2fPYuGlD2jVtSO3UFHpv14YPf8ostv27387kwO0K/gj+T84i8vLz6bJZOgAN6tSifu2EpovRemy6y1bMmZZF7vRs8lbmMfmt8ezUq1ORNnNnzGHWd78S/9DdL29lHqtWrAKgVp3ahJDIS76Kk7JJe/JzZhPnZkHeKlZ98TG1dtyjSJvae/Zm5bh3YNkSAOLiBWvu325LQqOm5H3/RYXmrqn2P2gf3ny54IOXLz+fSuMmjUhv2aJIm/SWLdioUUO+/HwqAG++PIwefboDECNs1KghABs13ojszJwKTF+zbLVLezJ/mU329CzyVq5iwltj2b1n0Wtjt56d+ei1DwD4dNh4duzWYZ2f0/WwvRn/1scVkrmm2nyXrZgzLZOc6dnkrVzFpLfGsUuv3Yu0yZ0xh5nf/UqMcZ37fzd+KsuXLKuouDVerz778dpLQwH44rOvaNy4ES1bpa3T7ovPviI7a93XoAljJ7F82fLVbVq3aZXcwCpefn7Vv5WsM/BTjPHnGOMK4CXg8D+0icDvK1A2AWaVxz9dIu8+nwBOBA4EDgUOKfxa7bXIaEHOrDmrt3Nn59CiVdE3K83XapOfl8+SRUtp1Kwx9RrUo+85R/HSPS9WaOY/s6w5uWS0XPMi3aplGllzcsiak0NGy/Q1+9ML9qvsshcvJ6NR/dXbrRrVI3vR8vW2nbVgKbMWLKXzJgXnZtq8xTSqW5tL/m8S/Z4ew10ffENe/rpvapSYpq2aM39W7urt+bNzadIq8V4KTVu34Ip3b+f6CQ8y6uE3WZg9Lxkx/xRCkxbE+WteW+L8HEKTor8zUtLbkJLelvoX3Eb9i/5B6ra7Ft45UPew01gx9KmKjFyjtcpoSeasrNXbmbOyadm6ZZE2LVu3JGv2mmGQWbOyaZVR0OaWa+5i4LUXMvqLt7j8ugu5+6YHKiZ4DdQsozm5s9dcG7mzc2mW0bxIm+YZzcmdVdAmPy+fpYuW0qhZ0Q8Zuhy6F+PftOCwIZq2as7ctX5nzJs9l6Z/eG+ripPRuiWzZ675wCZzVhYZf3idSlS/E/rywch1e6VIvwshnBVC+Gyt21lrHW4LTF9re0bhvrVdB5wQQpgBDAPKpetfIgWHOTHGoTHG/8YYp/1+K+kOaz/ZRx99tDxyVjn9BxzH0CfeYPnS9f8RJv1ZDP9uFgds05rUlAAUDMf4YsZcLtlve/510t7MXLCEoVOnl/JTlCzzZ+dy20GXc333i+h8ZHcapTWp7Eg1W0oqKemtWfbAIJY/dwd1jz4P6jWkdrc+rPr2c+KC3NJ/hipE/1OO5NbBd7N/x0O59W/3cOM96+/qrIqx5S7t+W3Zb8yw16i0jr8cfQgdOm7PI/dZtFbxYoyPxhh3X+v2v/4hfizwdIyxHdAHeC6UQ/fYRPo5fxFCeAF4C/jt950lLYtZ+OR+f4LxnRuHblDIZMnNzCWtzZpPxlu0TiM3q+ibwbmFbXIzc0lJTaFhowYsmreQrTtuQ9c+3Tj5qlNp2LghMUZW/LaSYc+8XdFP40+jVXoLMrPXfIKSlZ1Dq/Q0WqWnMemLNaN8subk0Knjul01lbiWG9Ujc9GaLpVZi5bTslG99bZ979uZXNVzp9XbrRrVZ5uWjWnXtKCr8n7tM/hq1nz+st57qzTzs+bStM2aT6eatm7Bgqz/vZfCwux5zP5hOlt22nb1pJL638QFuYSma3pZhaZp6xQQ4oIc8qb9APl5xLlZ5M+ZRUp6a1I23YbULXagdreDCHXqQ61a8NsyVrzzbEU/jWrtuNOO4qgTjgBg6hffkLFW9+KMNi3JXqs3A0D27GxarfVpYqs2LcnKLGhzRL+DV08g+d7Qkdxw96Akp6+55mXOpUXrNddGi9YtmJc5t0ibuZlzadEmjbmF76caNGrAonlr5gbqeujejB9q74YNNT9rLs3X+p3RrHVz5mdZ6KxIJ53en2NPOhKAr76YSuu2GauPZbRpReYfXqdKs1f3PTn/0jM55pBTWbFiZblm1f+g9CELVd1MYO2Jp9oV7lvb6RSMaiDGOCGEUA9IA/63/7R/kEjFoj4FhYZeFAyl+H1YRbX345c/0HrzNrTcuBW1atdi70P3YeKIom/EJ474lP2P6gFAtz578dX4gj9sBx11BWd1O52zup3OW08O5dX7X7bYkGT77rUnQ98bRYyRL6d+y0YbNSQ9rTnd9tiN8RMns2DhIhYsXMT4iZPptsdulR23WtuhdVN+nbeEmfOXsjIvn+HfzqL7VhnrtPtv7iIWLl/Jzm3WdPHfIaMpi35bxdylBfXJidNy2aLFRhWWvab59cv/kL5ZBs3bpZNaO5VdD+3KlBGfJXTfphnNqV23NgD1Gzdki923IevnchmO96eUP/1HUtLbEJq3gtRa1Oq4N3lTi/7OWDX1U1K3KizANWxESnob8nOz+O1fd7H0htNZeuOZ/PbWk6z87AOLDWXwwpOv0nf/E+i7/wmMencMhx/TB4Cdd9uRRQsXMye76B9Wc7JzWbxoCTvvtiMAhx/Th9HvfgRAduYcOnUtGPKy596dmPazPbHK6j9f/kjG5q1J37glqbVr0eXQvfh8xMQibT4fOZF9jtwPgD36dOXr8VNWHwshsOch3ZhgwWGD/fLlT7TcrDVp7QrORadDu/Flgr8zVD6efeIlDup+NAd1P5rh74zmyP6HAdBx9w4sWrh4vXM1FGeHnbbllrsGc/pxF5CbM7f0O0jFmwS0DyFsHkKoQ8GkkH/sFfAr0AMghLAdUA+YwwYqtYdDjPHUDX2Qqio/L59H//Yw1z13PSmpKYwaMoLpP/zKcZccz09TfmTiiImMGPI+A+65lIc/epRF8xdzx/m3VXbsGuuya29l0hdfMX/+QnoccQLnnn4iq1YVTHjX7y8Hs0+XTnw8YRIHHXMa9evV44ZBAwBo0rgRZ59yLP3PuAiAv556XImTT6p0tVJSuPKAHTnnlU/Ij5HDd9qYrdIa8eDH37F9RlP2bV9QfHjv21kcuF1bQgir75uaEhiw3/acPWQCMcJ2GU04cudNK+upVHv5efm8OvhJzn12ECmpKXzy8odk/jiDPgOO5tcpPzN15Ods0mFLznjkUuo3aciOPXbjoAFHc0uvgbTaqi1HXH3i6p81+rG3mf29f1SVWX4+v73+CPXPug5SUlg5cST5WdOpc+Bx5E3/ibyvJ5L33WRSt96FBpffT4z5rHjraVjqqjnJMGbkOPY5oCvDJ77O8qXLGXTRDauPvT76efrufwIA119x++plMT8eNZ6PRo0HYPClNzPoxktIrVWL35b/xuBLb6mU51ET5Ofl8/Tgx7jq2WtJSU3lw5dHMuPH6Rx1ybH896uf+HzkJD4cMpJz776Yu8c8xOL5i7jv/DXLxW67xw7kzsohe3pWCY+iROTn5fPC4Ce4+NmrCakpjHv5A2b9OIPDBvRj2pT/8OXIz9isw5ac+8hlNGjSkA49duPwAcdwba+C1Q8uf/l6MrZsS92G9bh9wsM8c8VDfP3Rl5X8rKqv0SM+Zr+e+/Dx58NYtmw5A89fM3Tr3TGvcFD3owEYdN0ADj/qYOo3qMenU0fy0nOvcfdtD3H13y+lQcMGPPRUwfUya8ZsTj/+wkp5LqreYoyrQgjnA8MpWPLyyRjj1yGE64HPYoxDgUuBx0IIAyiYQPKUuL7ZZf9HobSfEUJoB9wHdCvc9TFwUYxxRoKPEQ/fpEZ0iKj23vz1bVbm/FzZMVSodtoWLHtiYGXHEFD/9Du4cLN+lR1DwL2/DGHxJYdVdgwBG901lO1adq7sGAK+zZ7IsZseUdkxVOjFaW9w5mZHV3YMAY/98gqbNN+p9IZKul/nTgmlt6q+lr18fZWfAb3+MYOr5DlIZEjFUxR0t2hTeHurcJ8kSZIkSTVbjFX/VkUlUnBIjzE+FWNcVXh7Gkgv7U6SJEmSJOnPK5GCQ24I4YQQQmrh7QTA6W4lSZIkSVKxElkW8zQK5nC4m4LJI8YDNXYiSUmSJEmSVqv+y2JWmkRWqZgGOIOXJEmSJElKWLFDKkII/wghnL2e/WeHEG5NbixJkiRJklSdldTDYX/g8vXsfwz4CrgyKYkkSZIkSaoqHFJRZiVNGlk3xnXX14gx5gNVco1PSZIkSZJUNZRUcFgWQmj/x52F+5YlL5IkSZIkSaruShpSMRh4N4RwI/B54b7dgauAi5OcS5IkSZKkyhcdUlFWxRYcYozvhhCOAC4DLijcPRU4MsY4pQKySZIkSZKkaqrEZTFjjFOBkysoiyRJkiRJqiFKLDhIkiRJkvSn5ioVZVbSpJGSJEmSJEllUmrBIYTQfD37Nk9OHEmSJEmSVBMk0sPhrRBC4983QgjbA28lL5IkSZIkSVVEjFX/VkUlUnC4mYKiw0YhhN2AV4ATkhtLkiRJkiRVZ6VOGhljfCeEUBt4H2gE/CXG+EPSk0mSJEmSpGqr2IJDCOE+YO2+GU2A/wDnhxCIMV6Y7HCSJEmSJKl6KqmHw2d/2P48mUEkSZIkSapyXBazzIotOMQYn/njvhBCM2DjGONXSU0lSZIkSZKqtUSWxfwwhNC4cHnMycBjIYS7kh9NkiRJkiRVV6VOGgk0iTEuDCGcATwbY7w2hGAPB0mSJElSzeeQijJLZFnMWiGE1sAxwNtJziNJkiRJkmqARAoO1wPDgZ9ijJNCCFsAPyY3liRJkiRJqs5KHVIRY3wFeGWt7Z+BI5MZSpIkSZKkKiE6pKKsSi04hBDqAacDOwD1ft8fYzwtibkkSZIkSVI1lsiQiueADKA3MAZoByxKZihJkiRJklS9FdvDIYRQK8a4Ctgqxnh0COHwGOMzIYQXgI8rLqIkSZIkSZUj5sfKjlBtldTDYWLh15WFX+eHEHYEmgAtk5pKkiRJkiRVa6XO4QA8GkJoBlwDDAU2Av6W1FSSJEmSJKlaK6ng0DKEcEnh96cWfn2g8GvD5EWSJEmSJKmKyHeVirIqqeCQSkFvhrCeYw5ikSRJkiRJxSqp4DA7xnh9hSWRJEmSJEk1RkmTRq6vZ4MkSZIkSVKpSurh0KPCUkiSJEmSVBVF53Aoq2J7OMQY51ZkEEmSJEmSVHOUNKRCkiRJkiSpTEoaUiFJkiRJ0p9bvos0llWxBYcQwhRKWP4yxtghKYkkSZIkSVK1V1IPh0MKv55X+PW5wq/HJy+OJEmSJEmqCUKMJXcPCSF8EWPs+Id9k2OMuyb4GPY/kSRJkqSaK1R2gGRaet+5Vf5v2gYXPFglz0EicziEEEK3GOO4wo2u/I+TTZ64ad+yZFM5e27a6yx7YmBlx1Ch+qffwcqcnys7hoDaaVtw/mb9KjuGgPt/GUJO7+6VHUNA2vAxdMzoVtkxBHyROY6+mx5W2TFU6PVpQzljs6MqO4aAx395le1adq7sGAK+zZ5Y2RFURSVScDgdeDKE0KRwez5wWtISSZIkSZKkaq/UgkOM8XNg598LDjHGBUlPJUmSJElSVZCfX9kJqq1Sh0aEEFqFEJ4AXooxLgghbB9COL0CskmSJEmSpGoqkbkYngaGA20Kt38ALk5SHkmSJEmSVAMkUnBIizG+DOQDxBhXAXlJTSVJkiRJUlUQY9W/VVGJFByWhBBaULi8ZQhhT8B5HCRJkiRJUrESWaXiEmAosGUIYRyQDhyd1FSSJEmSJKlaS6Tg8DXQHdgGCMD3JNYzQpIkSZIk/UklUnCYEGPclYLCAwAhhMnArklLJUmSJElSVeCymGVWbMEhhJABtAXqhxA6UtC7AaAx0KACskmSJEmSpGqqpB4OvYFTgHbAXWvtXwgMSmImSZIkSZJUzRVbcIgxPgM8E0I4Msb4WgVmkiRJkiSpasivustOVnWJTP44LoTwRAjhXYAQwvYhhNOTnEuSJEmSJFVjiRQcngKGA20Kt38ALk5WIEmSJEmSVP0lUnBIizG+DOQDxBhXAXlJTSVJkiRJUlUQ86v+rYpKpOCwJITQAogAIYQ9gQVJTSVJkiRJkqq1klap+N0lwFBgyxDCOCAdOCqpqSRJkiRJUrVWasEhxjg5hNAd2AYIwPcxxpVJTyZJkiRJUmVzlYoyK7bgEELoW8yhrUMIxBhfT1ImSZIkSZJUzZXUw+HQEo5FwIKDJEmSJElar2ILDjHGUysyiCRJkiRJVU3Mr7qrQFR1pa5SEUJoEUK4N4QwOYTweQjhn4WrVkiSJEmSJK1XIstivgTMAY6kYHWKOcCQZIaSJEmSJEnVWyLLYraOMd6w1vaNIYR+yQokSZIkSVKV4SoVZZZID4f3Qwj9QwgphbdjgOHJDiZJkiRJkqqvRAoOZwIvACsKby8BZ4cQFoUQFiYznCRJkiRJqp5KHVIRY2xUEUEkSZIkSVLNkcgcDoQQOgCbrd0+xvh6kjJJkiRJklQ1RJfFLKtSCw4hhCeBDsDXwO//0hGw4CBJkiRJktYrkR4Oe8YYt096EkmSJEmSVGMkUnCYEELYPsb4TdLTSJIkSZJUlbgsZpklUnB4loKiQybwGxCAGGPskNRkkiRJkiSp2kqk4PAEcCIwhTVzOEiSJEmSJBUrkYLDnBjj0KQnkSRJkiSpqsn3c/eySqTg8EUI4QXgLQqGVAAuiylJkiRJkoqXSMGhPgWFhl5r7XNZTEmSJEmSVKxSCw4xxlMrIogkSZIkSVWOq1SUWUppDUII7UII/xdCyC68vRZCaFcR4SRJkiRJUvVUasEBeAoYCrQpvL1VuE+SJEmSJGm9EpnDIT3GuHaB4ekQwsVJylPhdurekROvPY2U1BQ+fGkkbz/0f0WO16pTi7PvuojNd9qCxfMWcf/5d5IzYw6ptVI5/bZz2WzHLUiplcq41z7krQed1mJDjfs5m9tHTSU/Rv7SYRNO27N9keP/GDWVSdNzAVi+Mo+5S39j7EUHATB74VL+/t6XZC1cTghw31F70LZJgwp/DjXBNTffxUfjJtK8WVPeeP7hdY7HGLnlnof5eMIk6tWry01XX8r222wFwJvDRvDIMy8BcPbJ/Tm8T88KzV4Tbdd9Z44afAopqSmMHzKaEQ+9WeT4lp2346jBJ9Nm20146oJ/8u93PwWgWds0znpkICElkForlTHPvMfYf42sjKdQY9TevTMN/3oBITWF5e++w7KXXyhyvG7PA2l4xjnk584BYNnQ/+O3995ZfTw0aEDTR59hxYSxLHngnxWavSa6/MaL6dajC8uXLefai27iuyk/rNNmuw7b8Pd/Xk3denUZN2oCt19zDwBnDzyNvscfxrzc+QDcf8sjjB01oQLT1ywdu+/KadeeQUpqKiNfep//e+i1Isdr1anFRXcNYIudtmLRvIXcef4/mDMjm/R2Lbl31APM+s9MAH744nseufqhyngKNcYO3Xfh2MGnkpKawsdDRvHuQ28UOd6+83b0H3wq7bbdlEcvuJvP3/1k9bGuR3bn4POPAuCd+19l/GtjKjJ6jTTopkvZ54CuLF+2nEEXXM83U75fp832HbbllnsHU7d+XT4aOZ6br74TgG13bM91t19JnXp1yVuVx/VX3MaUL76p6KcggOgqFWWVSMEhN4RwAvBi4faxQG7yIlWckJLCyTecyW3H/525mblcP/R2Jo+cxKwfZ6xu073fASxZsJiB3c9jz0O70e/Kk3jg/DvpfHBXatepzaDeA6hTrw63jryXCUM/JmfGnEp8RtVbXn7klpFTePiYPWnVqD7HP/sx3bfKYMu0RqvbXNZjx9Xfv/j5f/kue8Hq7Wve+TdndGlPl83SWbpiFSFUaPwa5Yg+PTnuyMMYdMMd6z3+8YRJ/DpjFsOGPMFXX3/HDXfcz4uP3cOChYt46KkXGPLEvQD0O/1C9t1rT5o0brTen6PShZTAMdefxv0n3MT8zFwuG3oLU0Z8RuZPM1e3mTcrh+cGPkiPMw8tct+F2fO4s+81rFqxijoN6nL1+3cwZcTnLMieV9FPo2ZISWGj8y5mwVWXkp8zh6b3PcKKT8aR9+u0Is1++2h0scWEBiedzsqpX1VE2hpvrx5d2GSLdhzepR877boDg24byEl9zlqn3aDbBnLDpbcxZfLX3P/CHXTbf0/GjS74A+v5R4fw3EMvrnMf/W9SUlI484az+fvxg8nNzOX2oXcyaeREZvw4fXWbA/r1ZPGCxZzX/Wy6Hbo3J115Mnee/w8AsqZlcmmfiyspfc0SUlI4/vozuOuE65mXOZdrht7Kv0d8xuyf1ry3nTsrh6cGPkCvMw8rct+GTTbi0IuO4cZDryDGyN/evp1/j/iMpQuXVPTTqDH26dGVTbfYmAP3OJKdd9uRwbdfQf+DTlun3bW3X8HgS2/my8+n8siL97D3/l34ePQEBg6+gAfueJyPR09gnx5dGTj4Ak7+yzmV8EyksktkSMVpwDFAJjAbOAqoERNJbrnLVmT9Mps507PIW7mKT94ay249Oxdps2vPTox97QMAJg6bwA7ddgIKPuGt26AuKakp1KlXh1UrV7Fs0bIKfw41ydTZ89i4aUPaNW1I7dQUem/Xhg9/yiy2/bvfzuTA7doC8J+cReTl59Nls3QAGtSpRf3aidTTtD6777JTiUWCD8Z+wmEH9iCEwM47bseiRYuZkzOXcZ9+TpdOHWnSuBFNGjeiS6eOjPv08wpMXvNststW5EzLInd6Nnkr85j81ng69OpUpM3cGXOY9d2vxD9U3/NW5rFqxSoAatepTQiJvOSrOLW22Y68WTPJz5wNq1bx24ejqdNlr4Tvn7rV1qQ0a8bKzyclMeWfR/fee/H2y+8BMGXy1zRq3Ii0li2KtElr2YKGGzVkyuSvAXj75ffY98C9KzxrTbfVLu2Z/ctssqZnsWrlKsa+9TGde+5RpE2nnnvwwWujAZgwbBw7ddu5MqLWeJvvshXZ0zLJmZ5N3spVTHxrHLv84XdG7ow5zPhu2jq/M3bovjPfjP2SJQsWs3ThEr4Z+yU77rtLBaavefY/aB/efHkYAF9+PpXGTRqR/ofXqfSWLdioUUO+/HwqAG++PIwefboDECNs1KghABs13ojszJwKTC+Vj0RWqZgGHFZau+qoWUYL5s5e01lj7uxctuxYtAt/84wW5M4qaJOfl8/SRUvZqFkjJg2bwG49O3PfpCeoW78u/7r+KZYsWFyh+Wua7MXLyWhUf/V2q0b1mDJr/nrbzlqwlFkLltJ5kzQAps1bTKO6tbnk/yYxc8FS9tg0nYu6b0dqit0ckiFrTi4ZLdNWb7dqmUbWnByy5uSQ0TJ9zf70gv0quyatmjNv1prXqXmzc9lsl60Svn/T1i0458krSN8sgzduft7eDRsgpUUa+XOyV2/n58yh1rbbrdOubrfu1N5xZ/JmTmfJI/eTP2cOhMBGZ53LottvonbH3Soydo3VsnU6mbPWnI+s2dm0bJ1OTnZukTbZs9duM4eWrde8RvU/7UgOOfpAvvnyO+667n4WLVhUMeFrmBYZLcidvea1Pnd2Du07brNum1kFbQreTy2hUbOCwnbLjVtxx7B7WLZoKS/c8TzfTrLLeFk1a9WcebPWnIt5s3PZYpf2JdxjjaatWjC3yO+buTRt1aKEe6g0rTJakjkra/V25qxsWrZuyZwir1MtyVr7dWpWNq0yWgJwyzV38diQe7nsuotISQkcd/AZFRdeKifFftwVQvhHCOHs9ew/O4Rwa0k/NIRwVgjhsxDCZ48++mh55KxyttilPfn5+VzY+Qwu2escDjrzMNI3blXZsf40hn83iwO2ab26oJCXH/lixlwu2W97/nXS3sxcsIShU6eX8lOkmm/+7FxuOehy/t79Ijof2Z1GaU0qO1KNtuKT8cw9uR/zzzmNlZM/Y6OBgwCod+gRrJj0Kfk5DrurKl55+v84dI9j6N/jFHKycrnkuvMrO9Kf0rzsuZzV5XQG9rmYp254ggH3Xkr9jeqXfkfpT6D/KUdy6+C72b/jodz6t3u48Z5rKjvSn1d+rPq3Kqqk/rX7A+urFjwGHFLSD40xPhpj3D3GuPtZZ607nrKqmJeZS/PWayq3zVu3YF7m3CJt5mbm0qJNQZuU1BQaNGrA4nmL6Hr43nz14RfkrcpjYe4Cfvj8OzbvsGWF5q9pWm5Uj8y1hqVkLVpOy0b11tv2vbWGUwC0alSfbVo2pl3ThtRKSWG/9hl8m7VgvffVhmuV3oLM7DWfoGRl59AqPY1W6WlkZq/5gyprTsF+ld2CrLk0a7PmdapZ6xYsyPrfeyksyJ7H7B+ms2Wnbcsz3p9Kfm4OKektV2+npKWTn1O0B09ctBBWrgRg+XvvUKv91gDU2m4H6h32F5o98xINzzyHuj160+C0qvv7sao65tS+vDTyaV4a+TQ5WblktFlzPlq1bkn27KIFnezZc2jZeu026avbzM2ZR35+PjFGXv/XUHbsuH3FPIkaKDczlxat17zWt2idxtzM3HXbtCloU/B+qiGL5i1i1YpVLJ5f0LPk56n/IXNaJm02b4vKZl7WXJq1WXMumrVuwbysuSXcY435Wbk0L/L7pjnzs2rEtG0V6rjTjuL10c/z+ujnmZOVQ0abNR9IZrRpWaTXFUD27Gxarf061aYlWZkFbY7odzAj3i4Y2v3e0JHs5OuUqqGSCg51Y4zrlEpiwYCvGtFP/ecvfyJj89akb9yS1Nq12PPQvZg8oujY2i9GTmKvI/cDoHOfLnwzfgoAOTNz2L5rwXwOdevXZauOWzP7PzNR2e3Quim/zlvCzPlLWZmXz/BvZ9F9q4x12v03dxELl69k5zbN1tw3oymLflvF3KW/ATBxWi5btNiowrL/2ey7154MfW8UMUa+nPotG23UkPS05nTbYzfGT5zMgoWLWLBwEeMnTqbbHnYf3xDTvvwP6Ztl0KJdOqm1U9n10K58NeKzhO7bNKM5tevWBqB+44Zsufs2ZP88K5lxa7RV339Hatt2pLTKgFq1qLvv/qz4ZFyRNqF589Xf19mz2+oJJRffdiPzTjyGeSf3Z8ljD/HbqOEsfbJm9gBMppefep3+B5xC/wNO4YP3PuKQYw4EYKddd2DxosVFhlMA5GTnsmTxEnbadQcADjnmQMYMHwtQZL6H/Q/qzn+++7mCnkXN89OXP9J68za03LgVtWrXYq9D92bSiE+LtJk0ciL7Hbk/AF36dGPK+ILJUxs3b0xKSsHb0VYbt6L15m3I+rX4+ZtUsl++/IlWm7UmrV3Be9vOh3bjyxGJzRvz9Zgv2X7vnWnQuCENGjdk+7135usxXyY5cc3zwpOv0nf/E+i7/wmMencMhx/TB4Cdd9uRRQsXFxlOATAnO5fFi5aw824FE6MffkwfRr/7EQDZmXPo1HVXAPbcuxPTfrb3rqqfkuZwWBZCaB9j/HHtnSGE9kCNmB0xPy+fZwc/zmXPDiYlNYWPXh7FzB+n0/eS/vz3q//wxchJjBkyir/efRF3jHmAxfMX88D5dwEw8tl3OeuO87llxD2EEPjoldFM/25aKY+oktRKSeHKA3bknFc+IT9GDt9pY7ZKa8SDH3/H9hlN2bd9QfHhvW9nceB2bQlrLUORmhIYsN/2nD1kAjHCdhlNOHLnTSvrqVR7l117K5O++Ir58xfS44gTOPf0E1m1qmDywX5/OZh9unTi4wmTOOiY06hfrx43DBoAQJPGjTj7lGPpf8ZFAPz11ONcoWID5efl8/LgJznv2UGE1BQ+eflDMn+cwcEDjubXKT8zZeTnbNJhS8585FIaNGnITj124+ABR3NTr4FkbNWWv1x9IpGCKvGox95m1ve+WSmz/DwWP3APTW6+A1JSWP7+MPKm/UKDk05j1Q/fseKT8dQ//EjqdOkGeXnkL1rE4jtLHIGoDTB25AT26tGFoZ+8zPJly7nu4ptXH3tp5NP0P+AUAG658s41y2KO/mT10pcX/e1cttmxPTFGZk/P5MbLbq+Mp1Ej5Ofl8/jgRxj87HWkpKYw6uWRTP9xOv0vOY7/fPUTk0ZOZNSQEVx09yU8MOYRFs9fxF2FK1Rsv8cO9L/kePJWriLGyCODHmSxc2KVWX5ePi8MfpyLn72GlNQUxr08mlk/zuDwAf34Zcp/+HLkZ2zWYUvOfeRyGjZpyM49duewAf24ttcAlixYzNv3vsY1Qwtet96+91XnJ9tAY0aOY58DujJ84ussX7qcQRfdsPrY66Ofp+/+JwBw/RW3r14W8+NR4/lo1HgABl96M4NuvITUWrX4bflvDL70lkp5HoKY77KYZRXW04mh4EAIBwH3ATcCv08zvztwFXBxjHFYgo8RT9y074bmVDl4btrrLHtiYGXHUKH6p9/Byhw/UasKaqdtwfmb9avsGALu/2UIOb27V3YMAWnDx9Axo1tlxxDwReY4+m5aI+fvrpZenzaUMzY7qrJjCHj8l1fZrmXn0hsq6b7NnlgjesAXZ/FVR1bdSRIKbXTLa1XyHBTbwyHG+G4I4QjgMuCCwt1TgSNjjFMqIJskSZIkSaqmSlwWM8Y4FTi5grJIkiRJklS1VOFVIKq6kiaNlCRJkiRJKhMLDpIkSZIkqdyVOKRCkiRJkqQ/NYdUlFmpPRxCCLeHEBqHEGqHEEaFEOaEEE6oiHCSJEmSJKl6SmRIRa8Y40LgEOAXYCsKVq6QJEmSJElar0SGVNQu/How8EqMcUEIVXKJT0mSJEmSylfMr+wE1VYiBYe3QgjfAcuAc0II6cDy5MaSJEmSJEnVWSJDKq4CugK7xxhXAkuBw5KaSpIkSZIkVWuJFByeiDHOjTHmFW4H4IUkZpIkSZIkqWrIj1X/VkUlUnCYGUJ4ECCE0Ax4H3g+qakkSZIkSVK1VmrBIcb4N2BxCOFhCooNd8YYn0p6MkmSJEmSVG0VO2lkCKHvWpufAn8DJgIxhNA3xvh6ssNJkiRJkqTqqaRVKg79w/YXFCyReSgQAQsOkiRJkqQaLVbhORKqumILDjHGUysyiCRJkiRJqjlKncMhhHB7CKFxCKF2CGFUCGFOCOGEiggnSZIkSZKqp0RWqegVY1wIHAL8AmwFXJbMUJIkSZIkVQmVveRlDV8W8/dhFwcDr8QYFyQxjyRJkiRJqgFKmjTyd2+HEL4DlgHnhBDSgeXJjSVJkiRJkqqzUgsOMcYrQwi3AwtijHkhhCXA4cmPJkmSJElSJcvPr+wE1VYiPRwA2gAHhBDqrbXv2STkkSRJkiRJNUCpBYcQwrXAvsD2wDDgIGAsFhwkSZIkSVIxEunhcBSwM/BFjPHUEEIr4PnkxpIkSZIkqQqowqtAVHWJrFKxLMaYD6wKITQGsoGNkxtLkiRJkiRVZ8UWHEIITxd++1kIoSnwGPA5MBmYkPRkkiRJkiSp2ippSEUHgBjjuYXbD4cQ3gMaxxi/SnoySZIkSZIqm0MqyqykgkODEEJHIPzxQAhh1xjj5OTFkiRJkiRJ1VlJBYe2wJ2sp+AARGD/pCSSJEmSJEnVXkkFh59ijBYVJEmSJEnS/yyRZTElSZIkSfpTitE5HMqqpGUxr6iwFJIkSZIkqUYptuAQY3y/IoNIkiRJkqSawyEVkiRJkiQVx2Uxy6ykIRWSJEmSJEllUmwPhxDCWxQsf7leMcbDkpJIkiRJkiSVmxDCgcA/gVTg8RjjretpcwxwHQV1gC9jjMdt6OOWNKTijg394ZIkSZIkVWvVfEhFCCEVeADoCcwAJoUQhsYYv1mrTXvgKqBbjHFeCKFleTx2sQWHGOOY8ngASZIkSZJUaToDP8UYfwYIIbwEHA58s1abM4EHYozzAGKM2eXxwCUNqZjC+odUhILHjx3KI4AkSZIkSSq7EMJZwFlr7Xo0xvho4fdtgelrHZsB7PGHH7F14c8ZR8Gwi+tijO9taK6ShlQcsqE/XJIkSZKk6ixWgyEVhcWFR0ttWLxaQHtgX6Ad8FEIYacY4/wNyRVi/N/+8UIIewHHxhjPS/AuVf/sSJIkSZLKKlR2gGRacOoBVf5v2iZPjSz2HIQQulDQY6F34fZVADHGW9Zq8zDwaYzxqcLtUcCVMcZJG5KrpB4OawfsCBwHHA38F3j9f3qQOm3/92Qqd6tWzOTCzfpVdgwVuveXIZzv+agS7v9lCCtzfq7sGAJqp23hdVFF3P/LEBo22KyyYwhYsvQXMppuV9kxVChz/rd0a7t/ZccQMG7maPZr17OyYwj4YMaIyo6gkk0C2ocQNgdmAv0p+Pt+bW8AxwJPhRDSKBhiscFvkEuaw2Hrwgc8FsgBhlDQI2K/DX1QSZIkSZKqhWowpKIkMcZVIYTzgeEUzM/wZIzx6xDC9cBnMcahhcd6hRC+AfKAy2KMuRv62CX1cPgO+Bg4JMb4E0AIYcCGPqAkSZIkSao4McZhwLA/7Bu81vcRuKTwVm5SSjjWF5gNfBBCeCyE0IMaPjZHkiRJkiSVj2ILDjHGN2KM/YFtgQ+Ai4GWIYSHQgi9KiifJEmSJEmqhkrq4QBAjHFJjPGFGOOhFCyP8QVwRdKTSZIkSZJU2fKrwa2KKrXgsLYY47wY46Mxxh7JCiRJkiRJkqq//6ngIEmSJEmSlIiSVqmQJEmSJOlPLVbzZTErkz0cJEmSJElSubPgIEmSJEmSyp1DKiRJkiRJKo5DKsrMHg6SJEmSJKncWXCQJEmSJEnlziEVkiRJkiQVJ7+yA1Rf9nCQJEmSJEnlzoKDJEmSJEkqdw6pkCRJkiSpGNFVKsrMHg6SJEmSJKncWXCQJEmSJEnlziEVkiRJkiQVx1UqysweDpIkSZIkqdxZcJAkSZIkSeXOgoMkSZIkSSp3Jc7hEEJYBBS7BkiMsXG5J5IkSZIkqYpwWcyyK7HgEGNsBBBCuAGYDTwHBOB4oHXS00mSJEmSpGop0SEVh8UYH4wxLooxLowxPgQcnsxgkiRJkiSp+kp0WcwlIYTjgZcoGGJxLLAkaakkSZIkSaoKXBazzBLt4XAccAyQVXg7unCfJEmSJEnSOhLq4RBj/AWHUEiSJEmSpAQl1MMhhLB1CGFUCGFq4XaHEMI1yY0mSZIkSVLlivlV/1ZVJTqk4jHgKmAlQIzxK6B/skJJkiRJkqTqLdGCQ4MY48Q/7FtV3mEkSZIkSVLNkOgqFTkhhC0pWKGCEMJRwOykpZIkSZIkqSqowkMWqrpECw7nAY8C24YQZgL/BY5PWipJkiRJklStlVpwCCGkAufGGA8IITQEUmKMi5IfTZIkSZIkVVelFhxijHkhhL0Kv1+S/EiSJEmSJFUNVXkViKou0SEVX4QQhgKvAKuLDjHG15OSSpIkSZIkVWuJFhzqAbnA/mvti4AFB0mSJEmStI6ECg4xxlOTHUSSJEmSJNUcKYk0CiFsHUIYFUKYWrjdIYRwTXKjVZy777qe774Zy+TPR9Bxlx3XOV6/fj2GvvEsU6eM4ct/j+bmm65afWzvvfZg4qfvsXzpNPr2PbgiY9dI23XfmatH3c3fPvwnB5xz+DrHt+y8HZe9fSt3//QCuxy0x+r9zdqmcdnbt3L5sNu46v076Hb8ARUZu0barvvO/G3U3Vz74T/pWcy5uOLtW/nnes7FFW/fypXDbuPq9+9gL8/FBrnm5rvY5+D+HHHCX9d7PMbIzXc/xEHHnMZfTjqHb77/afWxN4eNoE+/0+nT73TeHDaioiLXaF4XVcs/7riWr6Z8yKefvssuu+yw3jbXXjeQ738YT1b210X2t2vXhmHvvsj4Ce/w6afv0rv3vskPXMPdeNsgJkx+j9Hj3mCnnbdfb5srr7mIz6eO5j8zPiuy/6RT+/HBuDcZ+fHrvPnu82y9zZYVEbnGuvj68xky9jmeGfEYW+/Yfr1tttmpPc+OfJwhY5/j4uvPX73/vGvO5oUxT/PMiMe4+fHr2ahxw4qKXSNdcP25PD/2aR4f8Qjtd9xqvW223qk9T4x8lOfHPs0F15+7en+jpo34xwu38tzHT/OPF25loyYbVVRs/VF+NbhVUQkVHIDHgKuAlQAxxq+A/skKVZEOOnB/2m+1OdtuvxfnnHMFD9x/y3rb3XX3w+y4U3d279Sbrl06cWDv/QD4dfpMTj9jAC++9EYFpq6ZQkrg6OtP4+FTbuHmnpew22HdyNiqbZE282bl8K+BD/L5m+OK7F+YPY+7+17D7X2u4M4jruaAcw6ncctmFRm/RgkpgWOuP40HT7mFG0s4F88NfJDP1nMu7ux7Dbf2uYJ/HHE1Pc85nCaeizI7ok9PHr7rxmKPfzxhEr/OmMWwIU9w3eUXcsMd9wOwYOEiHnrqBV587B5efOweHnrqBRYsdIGhDeF1UbX07r0vW221OR122pfzzx/EPf+8ab3thr0ziu77rFscuuLK83n99Xfo2uVgTj75Au6+p/jrTKXr0XMftthiU7rseiADL7qW2+4cvN5277/3IQf16LfO/tdffZv9uh3OAXv35YF7n+C6m65IduQaq8v+e9Bu87b02+tEbr/iLgbecvF62w28ZQC3XX4n/fY6kXabt2XP/ToDMOmjzzlx/9M4ueeZTP95Oieef1wFpq9Z9ti/M203b8sJe53CnVfcw4BbLlxvu4tvuZA7Lr+bE/Y6hbabt6Xzfp0AOO68fkwe9wUn7n0Kk8d9wXHn1Yg/v/Qnk2jBoUGMceIf9q0q7zCV4dBDe/Pcv14F4NOJk2nStAkZGS2LtFm2bDkfjhkPwMqVK5n8xRTatm0NwLRpM5gy5Vvy86twWama2HSXrZgzLYvc6dnkrcxj8lvj2alXpyJt5s6Yw6zvfiX+YarYvJV5rFpR8F+yVp3ahJDof22tz2a7bEXOH85FhzKci9qeiw22+y470aRxo2KPfzD2Ew47sAchBHbecTsWLVrMnJy5jPv0c7p06kiTxo1o0rgRXTp1ZNynn1dg8prH66JqOfiQXrzwr4KppCZN+oImTRqRkZG+TrtJk74gM3POOvtjhMaNCj4tbNy4MbNnZyU3cA3Xu8/+vPzSmwBM/uxLGjdpTMtW656PyZ99SXbWuudj8aI1C6E1aFC/4ASpTPbq3ZX3Xi3o1fb15G9p1GQjWrRsXqRNi5bNadioAV9P/haA914dwd4HdgNg4kefkZeXv/r+LVuvex6VmG69uvD+qyMB+HbytzRsvBHN/3AumrdsTsONGvBt4bl4/9WR7NW7KwBde3Vl+CsF53L4KyPoVrhfqk4SnTQyJ4SwJQUTRRJCOAqYnbRUFahtmwxmTJ+1envmjNm0bZNBZmb2ets3adKYQw7uyX33P1FREf80mrZqzvxZuau358/OZdNd1t/1bL33b92Cs5+8gvTNMnjz5udZmD0vGTH/FJq0as68tc7FvNm5bPY/notzCs/FGzc/zwLPRdJkzcklo2Xa6u1WLdPImpND1pwcMlqueZPYKr1gv8rO66JqadOmFTNmrPn9PWtmJq3bZKy3uLA+N990N0OHPsdfzzmZBg0acMghxycr6p9C69atmDUzc/X27FmZtG7dcr3FheKcesZxnH3eydSuXZujDnP6sLJKz0gje9aa97HZs+eQnpFGbvbcom1mrzk3cwrb/NHB/Q9i1NAPkhu4Bkv7w7nImZ1DWkYac9c6F2kZacyZveb385zZc0grPBfN05qtbjs3ey7N0+wZV1lcFrPsEv2I5TzgEWDbEMJM4GLgnOIahxDOCiF8FkL47NFHH93wlFVEamoq/3ruAe5/4En++99fKzuO/mD+7FxuO+hyru9+EZ2P7E6jtCaVHelPa/7sXG456HL+7rmQVvO6qFqOPvownn/+VbZu34W+fzmVxx+/mxBCZcf6U3vq8RfYs2NvbrzuTgZctv55a1RxTrrwePJW5fH+6yMrO4oKRXv+qBpKqOAQY/w5xngAkA5sG2PcK8b4SwntH40x7h5j3P2ss84qp6jl55y/nsxnk97ns0nvMzszi3Ybt1l9rG271syclbne+z380O38+NN/ufe+xysq6p/K/Ky5NG3TYvV209YtWJD1v38CuDB7HrN/mM6WnbYtz3h/Kguy5tJsrXPRrIznYoHnIulapbcgM3vNJyNZ2Tm0Sk+jVXoamdlrPr3KmlOwX2XndVH5zjr7RCZ8MowJnwwjMzObdu3W/P5u0zaD2cX8/l6fk07ux2uvvQPAxImTqVevLmlpzUu5l9Z26hnHMfLj1xn58etkZc2hTduM1cdat8lg9uz19xYtzRuvDePAPj3KK+afQt+TD+fp9x/l6fcfJTdrLi3brBke3LJ1OnMyi/Zwm5OZU2SoRPof2vQ5pjfdDtiTv5+//rlRVLwjTj6Mx4Y/zGPDHyY3u+i5SGudRs4fzkVOZg7prdf8fk5vnb66zdyceauHYDRv2Zx5ufOT/wSkclZiwSGEcMnaN+Bs4My1tqulhx5+ht079WL3Tr0YOnQ4Jx5/FAB7dN6VhQsWrnc4xfV/v5wmTRpxyaXXVnTcP41fv/wP6Ztl0LxdOqm1U9n10K5MGfFZ6XcEmmY0p3bd2gDUb9yQLXbfhqyfZ5VyLxVnWuG5aLHWufiqjOdiy923IdtzkTT77rUnQ98bRYyRL6d+y0YbNSQ9rTnd9tiN8RMns2DhIhYsXMT4iZPptsdulR23WvO6qHyPPvIcXfbsQ5c9+/DWW+9z3PF9AejUqSMLFy5KeDgFwIwZs9hvv4Ix69tssyX16tVlzpzcUu6ltT31+AscsHdfDti7L++9M4pj+hdMzrnr7juzaOGi/2k4xeZbbLr6+wN6d+e/P08r97w12evPvMkpvc7ilF5n8dHwsRx4VE8Adth1OxYvXFJkOAVAbvZclixayg67bgfAgUf1ZOzwgvnK9ti3E8ed048rTrmG35b/VrFPpAZ445mhnNn7r5zZ+6+Me28cvY4qWJVou123Y8miJUWGU0DBUIkli5eyXeG56HXUAYx7fwIA40dMoPfRBeey99E9Gf/++Ap8JlpbzK/6t6qqtDkcfp+pbBugEzC0cPtQ4I+TSFZLw94dxYEH7s/3345j6bJlnHHGmjrKZ5PeZ/dOvWjbtjWDrrqIb7/7kUkThwPw4INP8eRTL7L7bjvz6itP0KxZEw45uCfXDr6UnXfZv7KeTrWWn5fPq4Of5NxnB5GSmsInL39I5o8z6DPgaH6d8jNTR37OJh225IxHLqV+k4bs2GM3DhpwNLf0GkirrdpyxNUnrv5Zox97m9nfT6/EZ1O95efl8/LgJznv2UGEtc7FwYXnYkrhuTjzkUtp0KQhO/XYjYMHHM1NvQaSsVVb/nL1iUQgAKMee5tZnosyu+zaW5n0xVfMn7+QHkecwLmnn8iqVQWTD/b7y8Hs06UTH0+YxEHHnEb9evW4YdAAAJo0bsTZpxxL/zMuAuCvpx5X4uSTKp3XRdUy/L0P6N17P6ZMHcOypcs4+6+XrT424ZNhdNmzDwA33nglx/Q7nAYN6vPDjxN4+ukh3HzTPVx15Y3c/8CtnH/+6UQiZ581sLKeSo0w8v0x9Oi5D598MZxlS5dz8XmD1hz7+HUO2LugOPS3vw/kL0cdTP0G9Zn89Qe88Nyr3HHrA5x21nHs070rK1etZMH8hVx4zlXFPZRKMWHUp3TZfw9eHvc8y5ct5+ZLbl997On3H+WUXgW9j+8cdA9X330FdevV5ZMPJjJh9KcAXHLjhdSuW5t7XvoHAF9P/oZ/XHlPhT+PmuCT0RPZY/89eH7sM/y2/Dduu+SO1cceG/4wZ/YuGDp0z6D7uPKugdSpV5eJH07i09EFf2a9eP9LXPvw3+jT/yCyZmTx93NcTUfVT0hkLFAI4SPg4BjjosLtRsA7McZ9EniMWKtO29JbKelWrZjJhZutuxSVKse9vwzhfM9HlXD/L0NYmfNzZccQUDttC6+LKuL+X4bQsMFmlR1DwJKlv5DRdLvKjqFCmfO/pVtbP1yqCsbNHM1+7XpWdgwBH8wYUaMnwcnu0b3KT6DRctSYKnkOEl2lohWwYq3tFYX7JEmSJEmqsarykIWqLtGCw7PAxBDC/xVuHwE8nYxAkiRJkiSp+iu14BAK1oh6FngX2Ltw96kxxi+SGUySJEmSJFVfpRYcYowxhDAsxrgTMLkCMkmSJEmSVDXEKjk9QrVQ4rKYa5kcQuiU1CSSJEmSJKnGSHQOhz2A40MI04AlFKzqFWOMHZKWTJIkSZIkVVuJFhx6JzWFJEmSJElVkKtUlF2iBYcqv+6oJEmSJEmqOhItOLxDQdEhAPWAzYHvgR2SlEuSJEmSJFVjCRUcCleoWC2EsCtwblISSZIkSZKkai/RHg5FxBgnhxD2KO8wkiRJkiRVJTHfZTHLKqGCQwjhkrU2U4BdgVlJSSRJkiRJkqq9RHs4NFrr+1UUzOnwWvnHkSRJkiRJNUGiczj8PdlBJEmSJEmqalwWs+xKLDiEEIaWdDzGeFj5xpEkSZIkSTVBaT0cugDTgReBTylYFlOSJEmSJKlEpRUcMoCewLHAcRTM3fBijPHrZAeTJEmSJKmyxejn7mWVUtLBGGNejPG9GOPJwJ7AT8CHIYTzKySdJEmSJEmqlkqdNDKEUBc4mIJeDpsB9wL/l9xYkiRJkiSpOitt0shngR2BYcDfY4xTKySVJEmSJElVgKtUlF1pPRxOAJYAFwEXhrB67EoAYoyxcRKzSZIkSZKkaqrEgkOMscQ5HiRJkiRJktan1DkcJEmSJEn6s4r5rlJRVvZgkCRJkiRJ5c6CgyRJkiRJKncWHCRJkiRJUrlzDgdJkiRJkooRY2UnqL7s4SBJkiRJksqdBQdJkiRJklTuHFIhSZIkSVIxXBaz7OzhIEmSJEmSyp0FB0mSJEmSVO4cUiFJkiRJUjEcUlF29nCQJEmSJEnlzoKDJEmSJEkqdw6pkCRJkiSpGDFWdoLqyx4OkiRJkiSp3FlwkCRJkiRJ5c4hFZIkSZIkFcNVKsrOHg6SJEmSJKncWXCQJEmSJEnlLsTkT7npnJ6SJEmSVHPV6DEHP+/Uq8r/TbvFlPer5DmokDkcTti0b0U8jErx/LTXWXzJYZUdQ4U2umsoOb27V3YMAWnDx3D+Zv0qO4aA+38Zwsqcnys7hoDaaVvQuY2vUVXBxFm+RlUl9/8yhKs3O66yYwi46ZcX2LzFzpUdQ8B/c7+s7AhJFWOV/Fu+WnBIhSRJkiRJKncWHCRJkiRJUrlzWUxJkiRJkooR8ys7QfVlDwdJkiRJklTuLDhIkiRJkqRy55AKSZIkSZKKke8qFWVmDwdJkiRJklTuLDhIkiRJkqRy55AKSZIkSZKKER1SUWb2cJAkSZIkSeXOgoMkSZIkSSp3DqmQJEmSJKkYMd8hFWWVUMEhhNAV2Gzt9jHGZ5OUSZIkSZIkVXOlFhxCCM8BWwL/BvIKd0fAgoMkSZIkSVqvRHo47A5sH2OMyQ4jSZIkSVJV4l/CZZfIpJFTgYxkB5EkSZIkSTVHsT0cQghvUTB0ohHwTQhhIvDb78djjIclP54kSZIkSaqOShpScUeFpZAkSZIkSTVKsQWHGOMYgBDCbTHGK9Y+FkK4DRiT5GySJEmSJFUql8Usu0TmcOi5nn0HlXcQSZIkSZJUc5Q0h8M5wLnAFiGEr9Y61AgYl+xgkiRJkiSp+ippDocXgHeBW4Ar19q/KMY4N6mpJEmSJEmqAvKjQyrKqqQ5HBYAC0II5/3xWAihdoxxZVKTSZIkSZKkaiuRORwmA3OAH4AfC7//JYQwOYSwWzLDSZIkSZKk6qmkIRW/GwG8GmMcDhBC6AUcCTwFPAjskbx4kiRJkiRVnuiQijJLpIfDnr8XGwBijO8DXWKMnwB1k5ZMkiRJkiRVW4n0cJgdQrgCeKlwux+QFUJIBfKTlkySJEmSJFVbiRQcjgOuBd4o3B5XuC8VOCY5sSRJkiRJqnwxVnaC6qvUgkOMMQe4oJjDP5VvHEmSJEmSVBOUWnAIIWwNDAQ2W7t9jHH/5MWSJEmSJEnVWSJDKl4BHgYeB/KSG0eSJEmSpKoj31UqyiyRgsOqGONDSU8iSZIkSZJqjESWxXwrhHBuCKF1CKH577ekJ5MkSZIkSdVWIj0cTi78etla+yKwRfnHkSRJkiRJNUEiq1RsXhFBJEmSJEmqaqJzOJRZqUMqQggNQgjXhBAeLdxuH0I4JPnRJEmSJElSdZXIHA5PASuAroXbM4Ebk5ZIkiRJkiRVe4kUHLaMMd4OrASIMS4F7FMiSZIkSarxYqz6t9KEEA4MIXwfQvgphHBlCe2ODCHEEMLu5fFvl0jBYUUIoT4FE0USQtgS+K08HlySJEmSJCVPCCEVeAA4CNgeODaEsP162jUCLgI+La/HTmSVimuB94CNQwj/AroBp5RXgMrWoXtHTrz2NFJSU/jwpZG89dD/FTleq04t/nrXRWy+0xYsmreI+8+/k5wZc0itlcoZt53LZjtuQUqtVMa+9iFvPfh6JT2LmiN1212pe8QZkJLKyk/eZ+Xo19ZpU2vnbtTpfSwRyJ/1X357/s41B+vWp8EVD7Bq6qeseP2RigteA9XevTMN/3oBITWF5e++w7KXXyhyvG7PA2l4xjnk584BYNnQ/+O3995ZfTw0aEDTR59hxYSxLHngnxWavabZrvvOHDX4FFJSUxg/ZDQjHnqzyPEtO2/HUYNPps22m/DUBf/k3+8W/I5o1jaNsx4ZSEgJpNZKZcwz7zH2XyMr4ynUGNfcfBcfjZtI82ZNeeP5h9c5HmPklnse5uMJk6hXry43XX0p22+zFQBvDhvBI8+8BMDZJ/fn8D49KzR7TXTpDRfSdf89WL7sN64fcAvfT/lxnTbb7rQ1g++5irr16jB+9Kfc+bd7ATj7stPYp/dexJjP3Jz5XH/xLeRk5Vb0U6gxyvo61Xb7Tel/4xnU26g++Xn5DH/g/5j89oTKeAo1RvvuHTh48EmkpKbw2ZAP+Oiht4oc36zzthw8+ERabbsJQy64j6/fnQjA5l225+C/nbC6XdqWbRhywf18+/5nFZq/prn2livY94C9WL5sOQPP/xtff/XdOm0GXn0+f+l3KE2aNGbHTbus3t+5y6787abL2XaH9lx4xhW8+5a/w1VmnYGfYow/A4QQXgIOB775Q7sbgNsoukLlBim1h0OMcQTQl4Iiw4vA7sCy8gpQmUJKCiffcCa3n3wjlx9wEXsetjdt2rcr0mbffgewZMFiLu1+Hu898Rb9rzwJgM4Hd6VWndpc1XsAfzt4IPsf14u0dumV8TRqjpBC3b5ns+zRv7P0tvOotes+hFYbF22S1praPY5m6X1XsOz281nxxuNFjtc56Hjyfv66IlPXTCkpbHTexSy85nLmnXkydffrQeomm67T7LePRjP/3DOYf+4ZRYoNAA1OOp2VU7+qqMQ1VkgJHHP9aTx4yi3c2PMSdjusGxlbtS3SZt6sHJ4b+CCfvTmuyP6F2fO4s+813NrnCv5xxNX0POdwmrRsVpHxa5wj+vTk4buKn8bo4wmT+HXGLIYNeYLrLr+QG+64H4AFCxfx0FMv8OJj9/DiY/fw0FMvsGDhooqKXSN13X8PNt68HUd2O55bLr+DK265ZL3trrj1Em6+7B8c2e14Nt68HV322wOA5x96ieMPOI0Tep7B2JETOGPAyeu9v0q3Ia9TK5et4NlLHuCmXgN58ORbOHLwydRv3KAi49coISVw6PWn8swpt/PPnpfR4bCupP/hXMyflcOrAx/mqzfHF9n/3wnfcH+fQdzfZxBPHHsTK5et4KeP/D2+IfY9YC8222IT9ut0KFddcj033nHNetuNHD6GI3oev87+mTMyuez8vzH0tXeTHVWlyI+hyt9CCGeFED5b63bWWk+hLTB9re0ZhftWCyHsCmwcYyz6pn4DJTKkghhjbozxnRjj2zHGHOCV8gxRWbbcZSuyfpnNnOlZ5K1cxSdvjWW3np2LtNm1Zyc+fu0DACYOm8AO3XYqOBAjdRvUJSU1hTr16rBq5SqWLaoRdZhKk7JJe/JzZhPnZkHeKlZ98TG1dtyjSJvae/Zm5bh3YNkSAOLiBWvu325LQqOm5H3/RYXmrolqbbMdebNmkp85G1at4rcPR1Ony14J3z91q61JadaMlZ9PSmLKP4fNdtmKnGlZ5E7PJm9lHpPfGk+HXp2KtJk7Yw6zvvuVGPOL7M9bmceqFasAqF2nNiEk9JKvEuy+y040adyo2OMfjP2Eww7sQQiBnXfcjkWLFjMnZy7jPv2cLp060qRxI5o0bkSXTh0Z9+nnFZi85tmn914Me3U4AFMnf0OjJhvRomXzIm1atGxOw0YNmDq54AOcYa8Op/uBBa9lSxYvXd2ufv16CY1/1fptyOtU9n9nM+eXTAAWZM9jUe5CNmreuMKy1zTtdtmKudOymFd4Lr56awLb9dqtSJv5M3LI+m76OudibTv22YMfPvySlctXJDtyjdbzoP14fUhBD5N/fzaFxk0akd4qbZ12//5sCnOyctbZP3P6LL775kfy84s/V9LvYoyPxhh3X+v2aKL3DQVvEu8CLi3vXGV991kjJo1sltGCubPXdJ+cOzuXZhnN120zq6BNfl4+SxctZaNmjZg4bAK/Lf2N+yc9wT0THmXYo2+yZMHiCs1f04QmLYjz17zYxvk5hCYtirRJSW9DSnpb6l9wG/Uv+gep2+5aeOdA3cNOY8XQpyoyco2V0iKN/DnZq7fzc+aQkrbuL8i63brT9KEnaXTN30lJL+zhEwIbnXUuSx57qKLi1mhNWjVn3qw1r1PzZufSpFXivRSatm7BVe/ezg0THmTkw2+yIHteMmKqUNacXDJarrlWWrVMI2tODllzcshouaYXXKv0gv0qu5YZaWTNWvM6lT1rDi0z0v/QJp3s2XP+0GbN+TnnijN467NXOLDvATzyjyeSH7qG2tDXqd9tuvOW1Kpdi5xpWeUZ70+lcatmLFjrXCycPZcmrZqXcI/12+nQLnw1dHzpDVWiVq1bMnvmmv/Ps2dlkdG6ZSUm0p/YTGDtruPtCvf9rhGwI/BhCOEXYE9gaHlMHFnWgkOJnwOs3Z3j0UcTLqxUK1vs0p78/Hwu6HwGl+x1Dn3OPIz0jVtVdqyaLyWVlPTWLHtgEMufu4O6R58H9RpSu1sfVn37OXGB428ryopPxjP35H7MP+c0Vk7+jI0GDgKg3qFHsGLSp+TnzCnlJ6gizJ+dyy0HXc7fu19E5yO70yitSWVHkqqMh257nEN3P5r3Xh/J0af1rew4f2qN05ty0l3n8/xlDxHtblKpGqU3JWObjfnR4RTSajGGKn8rxSSgfQhh8xBCHaA/MHTN84sLYoxpMcbNYoybAZ8Ah8UYN3gSl2InjQwhvMX6CwsBaLGe/asVdt/4vdIQP7rpvTIHTKZ5mbk0b73mqTRv3YJ5mXPXbdOmBXMzc0lJTaFBowYsnreIrofvzVcffkHeqjwW5i7gh8+/Y4sOWzJnulX5sooLcglN13zyFJqmrVNAiAtyyJv2A+TnEedmkT9nFinprUnZdBtSt9iB2t0OItSpD7VqwW/LWPHOsxX9NGqE/NwcUtLXVOBT0tLJzyn6aWxctHD198vfe4cGZ/wVgFrb7UDtHTtQ75DDCfXrQ63axGXLWPpkzSw+JtuCrLk0a7PmdapZ6xYsyPrfeyksyJ7H7B+ms2WnbVdP1qby1yq9BZnZa66VrOwcWqWn0So9jUlfrHnznjUnh04dO1RGxGrtqFOO4IjjDwHgm39/T6s2a16nWrZJJzuzaKEzO3MOLVun/6HNuj1L3vu/Edzz3G08doe95MpiQ1+n6m1Un3OeupK37niJX75Yd+JPJW5h1jyarHUuGrduzoKsuSXcY107HrIn3wz/jPxVeeUd70/hxNP70f/EggLmV198Teu2az6QbN2mFZmzs4u7q5Q0McZVIYTzgeFAKvBkjPHrEML1wGcxxqEl/4SyK6mHwx3Aneu53QH0SVagivTzlz+RsXlr0jduSWrtWux56F5MHlF0zPnkkZPY+8j9AOjcpwvfjJ8CQO7MHHboWjCfQ936ddmq49bM+s9MVHb5038kJb0NoXkrSK1FrY57kze16B9Gq6Z+SupWhfNoNGxESnob8nOz+O1fd7H0htNZeuOZ/PbWk6z87AOLDRtg1fffkdq2HSmtMqBWLeruuz8rPik60VdovqaLZp09u5H36zQAFt92I/NOPIZ5J/dnyWMP8duo4RYbNsC0L/9D+mYZtGiXTmrtVHY9tCtfjUis2Nw0ozm169YGoH7jhmy5+zZk/zwrmXH/9Pbda0+GvjeKGCNfTv2WjTZqSHpac7rtsRvjJ05mwcJFLFi4iPETJ9Ntj91K/4Eq4tWn3+CEnmdwQs8zGPPex/Q5qjcAO+66PYsXLiE3u+gfVrnZc1myaCk77lqw8lefo3rz0fCxAGy8+Zq5srr33otffvq1gp5FzbMhr1OptVM585FL+fT1jyyGloOZX/6HFptl0KzwXHQ4tAvfjfjf5ovpcFgXvnzL4RRl9dwTQzh4334cvG8/3h/2AX37HQrALrvvxKKFi9c7V4NUEWKMw2KMW8cYt4wx3lS4b/D6ig0xxn3Lo3cDlNDDIcY4pjweoCrLz8vnmcGPc/mzg0lJTWHMy6OY+eN0jrykP//96j9MHjmJMUNG8de7L+LOMQ+weP5i7j//LgBGPPsuZ91xPreOuIcQAh+9Mprp302r5GdUzeXn89vrj1D/rOsgJYWVE0eSnzWdOgceR970n8j7eiJ5300mdetdaHD5/cSYz4q3noalzvRe7vLzWPzAPTS5+Q5ISWH5+8PIm/YLDU46jVU/fMeKT8ZT//AjqdOlG+Tlkb9oEYvvvLWyU9dI+Xn5vDz4Sc57dhAhNYVPXv6QzB9ncPCAo/l1ys9MGfk5m3TYkjMfuZQGTRqyU4/dOHjA0dzUayAZW7XlL1efSKSga9qox95m1vfTS3tIleCya29l0hdfMX/+QnoccQLnnn4iq1YVTMzZ7y8Hs0+XTnw8YRIHHXMa9evV44ZBAwBo0rgRZ59yLP3PuAiAv556XImTT6p040Z9Qtcee/L6+BdYvuw3bhiw5jXo+RGPc0LPMwC4/aq7GXzPldStV5fxH3zK+NEFf9SeN+hsNt1yY/LzI5kzs7j1ijvX+zgq3Ya8Tu16cBe26rwdDZs1Ys+jugPw3MAHmfmN76nKIj8vn7cGP80pz15JSE1h8ssfkv3jTHoMOIqZU37mu5GTadthC45/ZAD1mzRk2x670mPAUdzb63IAmrZLo2nrFvzyybeV/Exqhg9GfMx+Pffiw8/eZtmy5Vx+weDVx975sKAwAXDltRdz2FF9qN+gHuOnvM+Q517nn7c/TIeOO/Dws3fTpEljevTuzsVXnkvvbg7/qgz5pQ9ZUDFCBYyTiyds6oVRFTw/7XUWX3JYZcdQoY3uGkpO7+6VHUNA2vAxnL9Zv8qOIeD+X4awMufnyo4hoHbaFnRu42tUVTBxlq9RVcn9vwzh6s2Oq+wYAm765QU2b7FzZccQ8N/cL2v0X+Sftulb5SeX2WPW61XyHLhGmiRJkiRJKnfFDqn4oxBCgxjj0tJbSpIkSZJUM1T57g1VWKk9HEIIXUMI3wDfFW7vHEJ4MOnJJEmSJElStZXIkIq7gd5ALkCM8Utgn2SGkiRJkiRJ1VtCczjEGP84rbkL80qSJEmSpGIlMofD9BBCVyCGEGoDFwGulSNJkiRJqvFcFrPsEunh8FfgPKAtMBPYpXBbkiRJkiRpvUrt4RBjzAGOr4AskiRJkiSphii14BBC2By4ANhs7fYxxsOSF0uSJEmSpMoXHVJRZonM4fAG8ATwFpCf1DSSJEmSJKlGSKTgsDzGeG/Sk0iSJEmSpBojkYLDP0MI1wLvA7/9vjPGODlpqSRJkiRJqgLs5l92iRQcdgJOBPZnzb91LNyWJEmSJElaRyIFh6OBLWKMK5IdRpIkSZIk1QyJFBymAk2B7ORGkSRJkiSpaom4SkVZJVJwaAp8F0KYRNE5HFwWU5IkSZIkrVciBYdrk55CkiRJkiTVKKUWHGKMYyoiiCRJkiRJVU1+rOwE1VdKaQ1CCHuGECaFEBaHEFaEEPJCCAsrIpwkSZIkSaqeSi04APcDxwI/AvWBM4AHkhlKkiRJkiRVb4kUHIgx/gSkxhjzYoxPAQcmN5YkSZIkSarOEpk0cmkIoQ7w7xDC7cBsEixUSJIkSZJUneW7LGaZJVI4OLGw3fnAEmBj4MhkhpIkSZIkSdVbIqtUTCv8djnw9+TGkSRJkiRJNUGxPRxCCIeHEM5ba/vTEMLPhbejKiaeJEmSJEmVJxKq/K2qKmlIxeXA0LW26wKdgH2Bc5KYSZIkSZIkVXMlDamoE2Ocvtb22BhjLpAbQmiY5FySJEmSJKkaK6ng0GztjRjj+WttpicnjiRJkiRJVUd+ZQeoxkoaUvFpCOHMP+4MIZwNTExeJEmSJEmSVN2V1MNhAPBGCOE4YHLhvt0omMvhiCTnkiRJkiRJ1VixBYcYYzbQNYSwP7BD4e53YoyjKySZJEmSJEmVrCqvAlHVldTDAYDCAoNFBkmSJEmSlLCS5nCQJEmSJEkqk1J7OEiSJEmS9GflKhVlZw8HSZIkSZJU7iw4SJIkSZKkcmfBQZIkSZIklTvncJAkSZIkqRjO4VB29nCQJEmSJEnlzoKDJEmSJEkqdyHGmOzHSPoDSJIkSZIqTajsAMn0Tqtjq/zftAdnvVglz0GFzOGwZdquFfEwKsV/ciazXcvOlR1Dhb7NnkjHjG6VHUPAF5njaNhgs8qOIWDJ0l/o3KZ7ZccQMHHWGFbm/FzZMQTUTtuCQzY5uLJjqNDbv77Drq33quwYAibPHsvhmxxS2TEEvPnr25UdQVWUQyokSZIkSVK5c5UKSZIkSZKKkV8lBytUD/ZwkCRJkiRJ5c6CgyRJkiRJKncOqZAkSZIkqRj5NXsRjqSyh4MkSZIkSSp3FhwkSZIkSVK5c0iFJEmSJEnFiJUdoBqzh4MkSZIkSSp3FhwkSZIkSVK5c0iFJEmSJEnFyK/sANWYPRwkSZIkSVK5s+AgSZIkSZLKnQUHSZIkSZJU7pzDQZIkSZKkYuSHUNkRqi17OEiSJEmSpHJnwUGSJEmSJJU7h1RIkiRJklSMWNkBqjF7OEiSJEmSpHJnwUGSJEmSJJU7h1RIkiRJklSM/MoOUI3Zw0GSJEmSJJU7Cw6SJEmSJKncOaRCkiRJkqRi5IfKTlB92cNBkiRJkiSVOwsOkiRJkiSp3DmkQpIkSZKkYuTjmIqysoeDJEmSJEkqdyX2cAgh7FrS8Rjj5PKNI0mSJEmSaoLShlTcWfi1HrA78CUQgA7AZ0CX5EWTJEmSJEnVVYkFhxjjfgAhhNeBXWOMUwq3dwSuS3o6SZIkSZIqUazsANVYonM4bPN7sQEgxjgV2C45kSRJkiRJUnWX6CoVX4UQHgeeL9w+HvgqOZEkSZIkSVJ1l2jB4VTgHOCiwu2PgIeSkkiSJEmSpCoi31UxyyyhgkOMcXkI4WFgWIzx+yRnkiRJkiRJ1VxCcziEEA4D/g28V7i9SwhhaBJzSZIkSZKkaizRIRXXAp2BDwFijP8OIWyerFCSJEmSJFUF+ZUdoBpLdJWKlTHGBX/Y5+ogkiRJkiRpvRLt4fB1COE4IDWE0B64EBifvFiSJEmSJKk6S7SHwwXADsBvwAvAAtasWFHtDb75MkZPfJN3xgxhhw7brrfNpYPOY+yXw/jql7FF9tepU5t7H7+V0RPf5LXhz9B249YVEblGG3TTpbz36Wu88eG/2H6nbdbbZvsO2/Lmhy/w3qevMeimS1fv33bH9rw07AleH/08r7z/DDt13L6iYtdIl994MW9OGMKQ0c+w7U5br7fNdh224eUPnuXNCUO4/MaLV+8/e+BpDP/iDV4a+TQvjXyavXp0qaDUNdM/7riWr6Z8yKefvssuu+yw3jbXXjeQ738YT1b210X2t2vXhmHvvsj4Ce/w6afv0rv3vskPXINdesOFvDbuX/xr5JNss1P79bbZdqeteWHUU7w27l9cesOFq/effdlp/Gvkkzw/4nHuffEO0lq1qKjYNc41N9/FPgf354gT/rre4zFGbr77IQ465jT+ctI5fPP9T6uPvTlsBH36nU6ffqfz5rARFRW5Rtu1+248/MEjPPrRYxx17tHrHK9VpxaXP3AFj370GP/f3n3HR1Gtfxz/PAEEKUEhlaao2Aui9Ca9iViogoING4qAFRUbil3sgHrVq1cBsQBeBEJRehEUEFH0+hNEUkhoAQWBnN8fMySbkJAQNtkkfN+89sXuzJndZ2cyZ2afOefMC5NfJKpGFACnX3A6r3z1Kq989SqvTn+Vxh10rAiGe54YzORF45kw+73DHr8nzHmfyYvGc88TGaf1Nw+7nukrP+fjuHf5OO5dmrZuVFhhlzgXtqzHG3PHMGbeOK66rfsh80sfV5p7Xr+XMfPG8dzkF9L3i4MiqkUyft0nXD7wisIKWXLgisGjqMprwqGLc+5B51x9//EQcFlBBlZYLmnblJNPqUXrBt14cOhIHn/ugWzLzZ4xjyvaX3vI9B59L2fH9p20btCNd8f8h/seKTF5mJBo0aYJJ51Sk44Nr+KRYaMY8ex92ZZ75Nn7GDHsKTo2vIqTTqlJ89beCcrdI+7g9eff5srW/Xj1mbHcPeKOwgy/RGnWpjG1TqlBt8a9GHn3swx/5u5syw1/5m6eGPYM3Rr3otYpNTKdmHw4bgK92w6gd9sBLJi9uLBCL3E6dLiE006rzfnnXcKgQcMZ/fKT2Zab9t/ZtGzR7ZDp990/iM8++y9NGnehf/87eGn0yIIOucRq0rohNWvX4KqmfRl17/PcN2potuXue3ooT93zHFc17UvN2jVo3KohAB++OZ6+ba+nX7sbWTBrMTcO6V+Y4Zcol3dux5gXc/5bnr94ORs3bWbahHd49N47eeL51wDYsTOVN9/9iI/fGs3Hb43mzXc/YsfO1MIKu0QKCwvj1pG38kj/R7itza20vKwFNevUzFSmfa8O7N6xi4EtbmLy218w4IHrANjw8wbuunQwd3a6gxHXjuD2UYMIK5XX01PJTtPWjah1Sk26NenNyHue44Gnsz9+P/D0MEbe/SzdmvSm1ik1aRJw/P7PuIn0aXcdfdpdx8I5Swor9BIlLCyMm0feymP9H2FQm9toflnLQ/aLdr3as2vHbm5pMZApb0+m/wMDMs2/YcSNrPx6RSFGLRJ8ea3Rs/sVnv0v82KmbadL+HzilwB8v2IN4ZUrERkdcUi571esYUticrbLfzbeW/6rKbNp3Lx+wQZcwrXu1ILJE6cBsGrFD972iMp8BTAyqioVK1Vg1YofAJg8cRptOrcEwDmoWKkCABXDK5KUcOg2k7xp2aEZX06cDsCalWupFF6JiCzbIiKqKhUqVmDNSu+K+pcTp3NJx+aFHmtJ1+XS9nz0n88AWL78OypXrkRMTOQh5ZYv/46EhC2HTHcOwitVBCA8PJz4+MSCDbgEa9GhGdMmzQDgh5U/UqlyRapGVclUpmpUFSpUKs8PK38EYNqkGbTs2AyA3bv+Si93/PHlcEX5kkQRd3Hd86gcXinH+XMXLOGyjm0wMy449yxSU3exJXkrC5euoHH9C6kcXonK4ZVoXP9CFi7VCf3ROL3u6cT/vpnEjQns37efeVPn0ah95qvijdo3ZPak2QAsmLaAC5peAMDePXtJO+ANx3Zc2eNw2imO2iUdm/PlJ4HH74rZH78rBRy/P5lOKx2/g6pO3dNJ+D2exI2J7N+3n/lT59Egy37RsH0j5vj7xcJpCzjf3y8OzkvcmMDG9RsLNW6RYDvsGA5m1gnoDFQ3s1cCZoUD+wsysMISHRvF5j8zTr4TNicRExuZbXIhOzGxkcT/mQDAgQMHSN25ixOrnMC2rdsLItwSLzomioTNmbdHVGwUW5JS0qdFxUaRGJ+U/jpxcxLRMV4TtFEPvchbE17hnkcHExZmXN3lxsILvoSJio0kYXPAeo5PIio2kuRM2yKSpMBtEb+FqNiMH8K9r7+KS3t05MdVP/Hio6+RukNXEfOjWrVoNm3anP56858JxFaLyTa5kJ2nnnyJKVM+4JZb+1O+fHkuvbRvQYVa4kXFRJAYsF8kbd5CVEwkKUlbA8pEkhS/JUuZjET2rffdSOceHdi1cxe3dr+rUOI+FiVuSSEmKmO9R0dFkLglmcQtycREZdRT0ZHedMm/qjFV2bI5Yx0mxydzRt0zsinj7RdpB9L4K/Uvwk8MZ+e2nZxe9wwGPz+YqOpRvHjXC+kJCMmfQ+qp+CQiYyMyHb8jYyNI2rwlU5nAeqrX9VdyaY8O/LjqZ158TMfv/KgaU5XkgHWcEp/M6Vn2iyoBZdIOpLE79S8qnRjOvr3/cOWt3Xmk70NcfvOVhRq3ZC/NQh1B8ZVbC4fNwLfAHmBFwGMK0KFgQxM5cr0HXMXTI16i9YVdefrh0Ywc/VCoQzpmffLe53Rt2JPebQaQnJjC0EcHhTqkY1aPHpfx4YeTOL1OY6684jrefvslzHTkDJU3n3mbrhf3YPpns+hxvU4kRdZ//zO3t72NIV2H0OP2HpQpWybUIR3TPnn/cy5r1Iveba8jOSmFoY/o+F3Yeg+5minvfMGev/aEOhSRo3bYFg7OuVXAKjP7HNjtnDsAYGalgLI5LWdmA4GBAGPHjg1etEHS7/qe9LrGG3xlzfdrqVY9moONKWOqRZEQn7erhgAJ8VuIrR5DQnwSpUqVolJ4RbVuOEJXX9+d7v0uB+CH734kplp0+ryYalGZrqCDl4WPjs0YVCe6WhSJCV6Zy3t14akHXwBg+pRZPPHS8AKOvmTped2VXNnXG55l7ffriKkWsJ5jozJdtQVIit9CVOC2iM24srs1eVv69M/+M4VXPniuIEMvcQbefA3XXdcHgBUrVlGjRrX0edWqxxC/OSHP73Vt/15c3s0bK2DZspWUK1eWiIgqbNmSksuSAtB9wOVc3vdSAH78/meiA/aLqGqRJGVpaZKUkLmlj1fm0Cvo0z+PY/QHz/DW8+8WUOTHtujIqiQkZaz3xKRkoiMjiI6MYPl3qzOmb0mm/oXnhyLEEiMlIYXIahlXxyNiI0hJTMmmTCQpCSmElQqjfKXy7Ny2M1OZTb/+wd+793DSGSfx6+pfkbzrOeBKrujbFYC1q9Zlrqdio9gSn7kO2hKfTFS1yExlDtZTmY7fH07h5Q+eLcjQS6yUhBQiAtZx1Wz2i61+mYP7RYVK5UndtpPTLzyDJp2b0v+B66gQXgHnHP/s3ce0978s7K8hctTyOobDTOD4gNfHA7NyKuycG+ecu9g5d/HAgQOPJr4C8eG/JtK1VR+6turDzGlfc0VP70Sy7kXnkbpzV567UwDMnv4NV/b2lu90WRsWz19eIDGXZB/9axJXtu7Hla37Mfurb+jWszMAF1x0rrc9kjJXzluSUtiVupsLLjoXgG49OzPnq3mAd6Jfv0k9ABo1r8+G3/4oxG9S/E1897P0QR7nTp/HpT07AnBevXPYlborU3NMgOSkFHbv2s159by7JlzasyPfzPDu5BLYX7R1p5b876ffCulblAzjxn5A40adadyoM1OnzuTqvt6V8Pr1L2TnztQ8d6cA2LRpM61aNQXgjDNOpVy5sko2HIFJ731Bv3Y30q/djXwzfT6du3sN/M6tdza7du7O1J0CICVpK7tT/+Lcet5dcjp378A8f7+oWbt6ermWHZrx+6/qm1tQLmnWiCnTZ+OcY9UP66hYsQKREVVo2vAiFi1byY6dqezYmcqiZStp2vCiUIdbrK1ftZ5qtasTXTOa0mVK06JrC5bGLc1UZmncUtp0bwNAs87NWL3IS/pE14xOHyQysnokNU6rQdIfmS80SO4mvvdZ+iCPX381n0t75OH4nRpw/O7Rka+nzweyHL87t9DxO59+WbWe2NrViPL3i+ZdW7Asy36xLG4prf39omnAfjG8+30MbHoDA5vewNR/TWHSaxOVbAixtGLwKKoO28IhQDnn3K6DL5xzu8ysfAHFVKi+jlvAJW2bMWf5ZPb8vYf77nw0fd7UuR/TtZV3hfG+RwbT9aqOHF++HAtWf8XED7/glWfHMvE/X/DCG08wZ9lktm/fweCbSsRYmiHzzayFtGjbhBnLPmPPX3sYPviJ9HmfzfmQK1v3A+Dx+55l1CsjKHt8WebPXsS82YsAGDHsKYaPHEqp0qXZu2cvI4aNCsn3KAkWzFpMszaNmbJkInv+3sOjdz2VPm/8rPfo3XYAAKPuf4HHXn6QsuXKsnDOkvS7UQx++DbOOLcOzjni/0hg5D26QpJfM6bPpUOHVqz54Rv+/utvbr7lnvR5i5dMo3EjL0k3cuT99OzVjfLlj2f9L4t5770JPPXkaB64fySvvf40gwbdgMNx88DsRyyX3C2cvYQmbRrx2aKP2PP3Xp4Y8nT6vA/j3qZfO2/cmGcfeIkRo++nbLmyLJq7lEVzvJPM24ffzEmn1iQtzZHwZyJP3/dCSL5HSXDPI0+z/LvVbN++kzaX9+O2G65h/35veKleV3ShReP6zF+8nE49r+f4cuV4YvgQACqHV+LmAX3ofaN3V6lbrrv6sINPSu7SDqQx5uE3efyDJwgrFUbchDg2rt9I36H9+GXNLyyLW8rMCTMZNvpuxs17i13bU3lmkHdMOLv+2XS/rQcH9h0gLS2NNx9845CWD3JkFsz2jt+TF0/wjt9DMo7fH8e9S5923h1CRj3wAo+N9o7fi+YsSb8bxeCHb+X0c+qAc2z+I4En71ULxfxIO5DGuIfH8OgHjxNWKozZE+L4Y/1Grh7al1/X/MKyuGXETZjJkNHDGDNvHKnbd/H8oGdCHbZI0FleRgM2s4XAHc65lf7ri4DXnHN5uVmyOzWi3tFFKUHxv+SVnBXVINRhiG9d0jIujGka6jAE+C5hIRXKnxzqMATY/dfvNKjWMtRhCLBs8zfsS9aVzaKgTMQpXFqrS6jDEN+XG/9LvdhmoQ5DgJXxC+hW69JQhyHA5I1flujBod6q0a/I30Lnpk0fFsltkNcWDncBn5jZZsCAGKBXQQUlIiIiIiIiIsVbnhIOzrnlZnYmcPBeLj875/YVXFgiIiIiIiIioVeUx0go6vI0aKQ/XsN9wGDn3A/AyWam9ksiIiIiIiIikq283qXiXeAf4OCYDX8CIwskIhEREREREREp9vI6hsOpzrleZtYHwDn3l5kVyUEpRERERERERILF6ZdvvuW1hcM/ZnY84ADM7FRgb4FFJSIiIiIiIiLFWl5bODwCTAdqmtl/gKbAgIIKSkRERERERESKt7zepSLOzFYCjfBuiznYOZdcoJGJiIiIiIiIhJjuUpF/h004mNmZzrmfzKyePyne/7+WmdUEtjrnNhRohCIiIiIiIiJS7OTWwmEYcBPwQg7zq5rZKufcNcENS0RERERERESKs8MmHJxzN/n/t8qpjJnNDHZQIiIiIiIiIkWBulTk32HvUmFm9wY875Fl3lMAzrn2BROaiIiIiIiIiBRXud0Ws3fA8weyzOsY5FhEREREREREpITIbQwHy+F5dq9FREREREREShQX6gCKsdxaOLgcnmf3WkREREREREQEyL2FwwVmthOvNcPx/nP81+UKNDIRERERERERKbZyu0tFqcIKRERERERERERKjtxaOIiIiIiIiIgcs9I0emG+5TaGg4iIiIiIiIjIEVPCQURERERERESCTl0qRERERERERHKQFuoAijG1cBARERERERGRoFPCQURERERERESCTl0qRERERERERHKgLhX5pxYOIiIiIiIiIhJ0SjiIiIiIiIiISNCpS4WIiIiIiIhIDlyoAyjG1MJBRERERERERIJOCQcRERERERERCTp1qRARERERERHJQZqFOoLiSy0cRERERERERCTolHAQERERERERkaBTwkFEREREREREgk5jOIiIiIiIiIjkIC3UARRjauEgIiIiIiIiUoKZWUcz+9nMfjWz+7OZP9TMfjSz1WY228xOCsbnKuEgIiIiIiIiUkKZWSngdaATcDbQx8zOzlLsO+Bi59z5wCTg2WB8thIOIiIiIiIiIjlwxeCRiwbAr86535xz/wDjgW6ZvqNzc51zf/kvlwA18rh6DksJBxEREREREZFizMwGmtm3AY+BAbOrA38EvN7kT8vJDcBXwYhLg0aKiIiIiIiIFGPOuXHAuKN9HzPrB1wMtDzqoFDCQURERERERCRHaXnptFC0/QnUDHhdw5+WiZm1BR4EWjrn9gbjg9WlQkRERERERKTkWg7UMbPaZnYc0BuYEljAzC4ExgKXOeeSgvXB5lyBZ2uKfTpIREREREREcmShDqAgPXlS3yL/m/bBDf857DYws87AaKAU8C/n3JNm9jjwrXNuipnNAs4D4v1FNjrnLjvauAqlS0WbGu0L42MkF7M3zaTPSZeHOgzxfbzhC6486aj3YQmCzzZMIeaEs0IdhgAJ29cx6OReoQ5DgNd+n8CltbqEOgwBvtz4X/Yl/xbqMMRXJuIUmldvE+owBJj/52yuqNU11GEI8PnGqaEOoUClhTqAIHDOTQOmZZk2IuB524L4XHWpEBEREREREZGgU8JBRERERERERIJOd6kQERERERERyUGRH8ChCFMLBxEREREREREJOiUcRERERERERCTocuxSYWavcpjWI865OwskIhEREREREZEioiTcpSJUDtfC4VtgBVAOqAf84j/qAscVeGQiIiIiIiIiUmzl2MLBOfc+gJndCjRzzu33X48B5hdOeCIiIiIiIiJSHOVlDIcTgfCA1xX9aSIiIiIiIiIi2crLbTGfBr4zs7mAAS2ARwsyKBEREREREZGiIM1CHUHxlWvCwTn3rpl9BTT0J93nnEso2LBEREREREREpDjLtUuFmRnQFrjAOTcZOM7MGhR4ZCIiIiIiIiJSbOWlS8UbeHcCaQ08DqQCnwL1CzAuERERERERkZBLw4U6hGIrLwmHhs65emb2HYBzbpuZ6baYIiIiIiIiIpKjvNylYp+ZlQIvrWNmkXgtHkREREREREREspWXFg6vAJ8DUWb2JNAdeLhAoxIREREREREpAtShIv/ycpeK/5jZCqAN3m0xL3fOrSvwyERERERERESk2MrLXSqeAH51zr3unHsN+NPM3i340ERERERERESkuMrLGA6lgaVmdr6ZtQOWAysKNiwRERERERGR0EsrBo+iKi9dKh4ws1nAUmAb0MI592uBRyYiIiIiIiIixVZeulS0wBs48nHga+BVM6tWwHGJiIiIiIiISDGWl7tUPA/0cM79CGBmVwJzgDMLMjARERERERGRUEvTfSryLS8Jh8bOuQMHXzjnPjOzbwowJhEREREREREp5nJMOJhZP+fch8BgM8uuyIsFFpWIiIiIiIiIFGuHa+FQ3v+/UmEEIiIiIiIiIiIlx+ESDicDOOceM7N2zrm4wglJREREREREpGjQCA75d7i7VHQMeP5MQQciIiIiIiIiIiVHrrfFFBERERERERE5UofrUhFlZkMBC3iezjmnQSNFRERERESkREsLdQDF2OESDm+RMWBk4HMRERERERERkcPKMeHgnHusMAMRERERERERkZLjcC0cRERERERERI5pabpPRb5p0EgRERERERERCTolHEREREREREQk6HJNOJjZYDMLN887ZrbSzNoXRnAiIiIiIiIioeSKwaOoyksLh+udczuB9sCJwDXA0wUalYiIiIiIiIgUa3lJOJj/f2fgA+fc2oBpIiIiIiIiIiKHyMtdKlaY2UygNvCAmVUC0go2LBEREREREZHQ04/f/MtLwuEGoC7wm3PuLzOrClxXoFGJiIiIiIiISLGWl4SDA84GLgUeByoA5QoyqMJU/5KLuf2xWwkrFca0j6cz/vUJmeaXOa4M942+h9PPr8PObak8ceuTJG5KBKDP7b3p1KcDaQfSeG3EG3z7zYpQfIUS5YKWF3LtIzcSViqMuePjmPLmZ5nmlz6uNLe9eBe1zzuVXdtSeXnQ8yRvSqLp5S24dOAV6eVqnXUSw7sMY8OP/1fYX6HEuLBlPa5/5EbCSpVi1viZfP7mp5nmlz6uNINfHMIp551G6radvDDoObZsSiKyRhSvzH6dzf/7E4D13/3M2AffDMVXKFFGPjOcNu1a8Pffexh823DWrPrxkDL3PzSYHr27ccIJ4Zxa4+L06dde14vrbryaA2kH2L3rL+656xHW//y/wgy/xDir5QV0HzGAsFJhLJowh7g3J2eaf2qDs+g+oj/VzqzFu3e8zPdfLQWg+tkn0XvkjZSreDxpB9KY8frnrPxycSi+QolRr+VFDHx0IGGlwpg5fiaT3vgk0/zSx5Vm6EvDOO2800jdlsoztz9N0qYkTr/gdAY9fQcAZvDRSx+xeIa2xdF46KkXmbdwGVVOPIEvPhxzyHznHKNGj2H+4uWUK1eWJx8cxtlnnAbA5GlxjH1/PAA39+9Nt87tCjX2kmrw47fTqHVD9v69l6eGPMv6H345pMzp59Vh+Ev3UrZcWZbMWcrLI14H4LaHBtKkXWP2/7OfPzdsZtTQZ9m1c3dhf4US4cKW9bjh0ZsIKxXGrPFxfPbGpEzzSx9XmsEvDeXU804ldVsqz9/+bPq51Ktz3sh0LjVm+Buh+AoiRy0vYzi8ATQG+vivU4HXCyyiQhQWFsadIwfxwDUPcn2rm2jd7RJOqlMrU5lOvTuya8curm12HZ++9Rk3Db8BgJPq1KJVt5bc0Hog9/d7kMFP3kFYmO4yejQsLIzrnriZZ/o/zt1t76DJZc2pXqdGpjKterVj945dDGl5K9PemcLV918LwMIv5vFA5yE80HkIbwwZzZY/kpRsOAphYWHc9MTNjOz/GIPb3k7zy1pQo07NTGXa9mrHrh27uL3lzUx9ZwrX3t8/fV7ihgSGdb6LYZ3vUrIhCNq0a8Epp5xE43oduXvwIzzzwohsy82c/jWd2vQ6ZPpnk76kVdNutG1+Ja+/8g6PPnlfQYdcIlmY0fPx63ljwChGthvKRZc1Jea06pnKbNuczAd3v8G3kxdmmr7v73/499DXebL93bzRfxRXjejP8eHlCzP8EiUsLIxbR97KI/0f4bY2t9LyshbUzFJHte/Vgd07djGwxU1MfvsLBjzgNc7c8PMG7rp0MHd2uoMR147g9lGDCCul4/fRuLxzO8a8ODLH+fMXL2fjps1Mm/AOj957J088/xoAO3am8ua7H/HxW6P5+K3RvPnuR+zYmVpYYZdYjVo3oEbtGvRpdi3P3vciw0YNzrbcsFF38ey9L9Kn2bXUqF2Dhq0aALB83gr6t76BAe1u4o/fNtFv0NWFGX6JERYWxsCRt/BE/0e5s83tNMv2XKo9u3fs4rYWNzP17clc+8CA9HmJGxIY2mkwQzsNVrKhCHDF4F9RlZcjbEPn3O3AHgDn3DbguAKNqpCcWfcM/vx9M/EbE9i/bz9zJ39Dk/ZNMpVp0r4xMz+JA+Cb/86jXrML/elNmDv5G/b9s4+EPxL48/fNnFn3jEL/DiXJaXXrkPB7PEl/JHJg334WT13Axe0aZipzUbsGzPt0LgBLpy3i3KbnH/I+TS5rzqKp8wsl5pLqtLp1iP89nsQ/Etm/bz8Lps6nQZZtUb9dQ+Z+OgeAxdMWcl7TC0IR6jGhQ+fWTBzvXUlf+e0qwiuHExUdeUi5ld+uIilxyyHTd6VmXJkqX/54cEX3oFSUnVz3NJI3JJLyRxIH9h1g5dRFnN++fqYyWzdtYfNPG3Euc2/PpP+LZ8vvCQDsSNpGaspOKlYJL7TYS5rT655O/O+bSfSP3/OmzqNR+0aZyjRq35DZk2YDsGDaAi7w66i9e/aSdsDbPseVPQ6n/eGoXVz3PCqHV8px/twFS7isYxvMjAvOPYvU1F1sSd7KwqUraFz/QiqHV6JyeCUa17+QhUvVWvRoNevQlOmTZgLw48p1VKxckapRVTKVqRpVhQqVyvPjynUATJ80k+YdmwJewuGAv4+sXfkjkbERhRh9yVHn4LnUxoPnUvNo0D7zuVSD9g2Z69dTi6Yt5HydS0kJlJeEwz4zK4V/e08zi6SEjJsRERvBlviMk/MtCVuIiK2auUxMBEl+mbQDaezeuZvwE8OJiK2aadnkhGQiVCEflRNjqpASn5z+OiU+hRNjMh8gq8RUIWWzVybtQBp/pf5FpRMzn+Q07tqMRZOVcDgaVWOqZtkWyVSJqXpomUzbYnf6toiqGc3z00bzxISnOKv+2YUXeAkVGxvN5j8T0l/Hb04gNjbqiN7juhuvZsl3M3j4sbt58L6ngh3iMaFydBW2bU5Jf70tPoXK0Sce8fucdMGplC5TmuQNicEM75hSNaYqWzZn1FHJ8clUjT60jtqyOeP4/VfqX4Sf6CV5Tq97Bq/PeoPXZr7OG8NfT09ASMFI3JJCTFTGOVJ0VASJW5JJ3JJMTFRG8jQ60psuRycyJoKkzQHnt/FbiIjJfI4aEZPlHDg+mciYQ89ju/TuxNK5ywsu2BKsSkxVkjdnPq/Nrp5KPuRcyqunompG88K00YycOIqzGuhcSoqvvCQcXgE+B6LM7ElgAXDYs1UzG2hm35rZt+PGjQtCmCJ5c2rdOuz9ey+b1m8MdSjHrG1JWxnY+Abu7nwX7z7xDkNeGcbxFY8PdVjHvHff/ohGF3Zg5KMvMOSeW0IdzjErPPIErn1xEB/e86aurIfQ+u9/5va2tzGk6xB63N6DMmXLhDokkSLnmjuv5sD+A8z8bFaoQznmbEvaysBG1zOs813864m3GfrK3TqXkmIr14SDc+4/wL3AKCAeuNw590kuy4xzzl3snLt44MCBwYm0ACTHJxMZm5FZj4yJJDk+JXOZhGSi/DJhpcKoEF6Bndt2khyfkmnZiJgIkuOVlT8a2xK2UjWglUjV2KpsS9iaqczWhK1UreaVCSsVRvlK5UndltHfs0nX5iyaotYNRyslISXLtohga0LKoWUybYsKpG5LZf8/+9m13dsmv/3wPxI2JFCtduZ+7pK76268mlnzP2PW/M9ITNxCteox6fNiq8UQH5+Ur/f94tNpdOzcJlhhHlN2JG7lxGoZV6dOjK3KjsRteV6+XMXjufXd+5n6/Hh+/+7QAdwk71ISUoisllFHRcRGkJJ4aB0VWS3j+F2+Unl2btuZqcymX//g7917OOmMkwo+6GNYdGRVEpIyzpESk5KJjowgOjKChKSMq+yJW7zpcuSu6N+Nf80cy79mjiUlMYWoagHnt7GRJCdkPkdNTshyDhwbwZaAMp16dqBJ28Y8Pkgt4vJra0IKEdUyn9dmV09FHHIutZP9/+wn9eC51Br/XOoUnUuFUloxeBRVuSYczKwKkAR8DHwEJJpZibgU8NOqn6leuzoxNWMoXaY0rbq1ZFFc5pGqF8ctpn0Pb8Tkll1a8N3C7wFYFLeYVt1aUua4MsTUjKF67er89P3Phf0VSpT/rfqFmNqxRNaMolSZ0jTu2owVccsylVkxaxktrmoFQMPOTVi7aE36PDOj0aVNWayEw1H7ddUvxNauRlTNaEqXKU2zrs1ZHrc0U5nls5bR6qrWADTu3JQ1i1YDEF4lPH0A1eia0cTWrkbixgTkyLz79ke0bX4lbZtfyfT/zqZn724A1Lv4AlJ3pmY7VkNOap+S8WOqbYeW/N9vG4Ie77Fgw6r/EXlyDFVrRFKqTCnqdW3C6rhv87RsqTKluGnsMJZ+Ni/9zhWSf+tXrada7epE+3VUi64tWJqljloat5Q23b3kWrPOzVjt11HRNaPTB4mMrB5JjdNqkPRH/hJ4kjeXNGvElOmzcc6x6od1VKxYgciIKjRteBGLlq1kx85UduxMZdGylTRteFGowy2WPn9/Mte3v5nr29/M/BkL6di9PQBn1zuLXTt3k5KU+QJOStJWdqf+xdn1zgKgY/f2LJjhDXbb4JL6XH1rLx4Y8BB79+wt3C9SgvxyyLlUC5ZnOa9dHreUVn491SSnc6la/rnUBp1LSfGUl9tirgRqAtsAA04AEswsEbjJOVdsR/dJO5DGqw+/xjP/eYqwsDC+mjCDDes3MODua/l51XoWxy1h2vjpPPDyffx7wbukbk9l5G1epnfD+g18PXUe/5rzFgcOHODVh14jLa0o55aKvrQDabw34i0e+PcjhJUqxdcTZ7Hplz/oPrQP/7f6V1bMWs7XE2Zx20t38dI3b7JreyqvDnohffkzG55DyuZkkv5Qv+ijlXYgjbdHjGXEvx8lrFQYsyfO4o9f/qD30Kv53+pfWT5rGbMnxDH4paG8/s1Ydm1P5cVBzwFwdsNz6D20Lwf27cc5x9jhb7Brx67QfqFibtbMb2jTrgVLvpvB33/t4a7bh2fMm/8ZbZtfCcDDj93NFd27cHz541m5di4ffTCJ559+nesHXk2Llk3Yt38fO7bv5M5bHwjVVynW0g6kMXHEv7j938OxUmEsmfg1Cb9sosuQHmxc8xtrZq2g1vmnctPYYZSvXIHz2lxElyE9eLL93dTr0pjTGpxFhRMr0ah7SwA+uPsN/vxRyZ/8SDuQxpiH3+TxD54grFQYcRPi2Lh+I32H9uOXNb+wLG4pMyfMZNjouxk37y12bU/lmUHPAnB2/bPpflsPDuw7QFpaGm8++MYhLR/kyNzzyNMs/24127fvpM3l/bjthmvYv38/AL2u6EKLxvWZv3g5nXpez/HlyvHE8CEAVA6vxM0D+tD7Ru8uCrdcd/VhB5+UvFk8eymNWjdk/MIP2PP3HkYNfS593r9mjuX69jcD8OLwlzNuizl3GUvmeD+Gh4y8gzJly/DieG+fWbtyHS/cP7rQv0dxl3YgjbceHsMjHzzmnUtNmMUf6zfSZ2hffl3zC8vjljFrQhx3jR7KG/PGsmv7Ll44WE81PJc+w7xzqbQ0x5jhr+tcSooty60PqZm9BUxyzs3wX7cHrgLeBV52zjU83PKAa1OjfTBilaM0e9NM+px0eajDEN/HG77gypMuC3UYAny2YQoxJ5wV6jAESNi+jkEnH3prTyl8r/0+gUtrdQl1GAJ8ufG/7Ev+LdRhiK9MxCk0r66uaUXB/D9nc0WtrqEOQ4DPN061UMdQkG47uWeRH3jpjd8nFsltkJdBIxsdTDYAOOdmAo2dc0uAsgUWmYiIiIiIiIgUW3npUhFvZvcB4/3XvfDGcShF0R6fQkRERERERERCJC8tHK4GagBf+I9a/rRSQM+CCkxEREREREQk1FwxeBRVubZwcM4lA3fkMPvX4IYjIiIiIiIiIiVBrgkHM4sE7gXOAcodnO6ca12AcYmIiIiIiIhIMZaXMRz+A0wALgVuAfoDeb8BvIiIiIiIiEgxlVakOy0UbXkZw6Gqc+4dYJ9z7hvn3PWAWjeIiIiIiIiISI7y0sJhn/9/vJl1ATYDVQouJBEREREREREp7vKScBhpZpWBYcCrQDhwV0EGJSIiIiIiIlIUpIU6gGIsLwmHbc65HcAOoBWAmTUt0KhEREREREREpFjLyxgOr+ZxmoiIiIiIiIgIcJgWDmbWGGgCRJrZ0IBZ4UCpgg5MRERERERERIqvw3WpOA6o6JepFDB9J9C9IIMSERERERERKQqcbouZbzkmHJxz3wDfmNl7zrkNZlbRn76r0KITERERERERkWIpL4NGVjKz7/BvhWlmyUB/59wPBRqZiIiIiIiIiBRbeUk4jAOGOufmApjZJf60JgUXloiIiIiIiEjo6baY+ZeXu1RUOJhsAHDOfQ1UKLCIRERERERERKTYy0sLh9/M7GHgA/91P+C3ggtJRERERERERIq7vCQcrgceAz7zX8/3p4mIiIiIiIiUaLpLRf7lmnBwzm0D7jSzykCacy614MMSERERERERkeIs1zEczKy+ma0BVgFrzGyVmV1U8KGJiIiIiIiISHGVly4V7wC3OefmA5hZM+Bd4PyCDExEREREREQk1HSXivzLy10qDhxMNgA45xYA+wsuJBEREREREREp7nJs4WBm9fyn35jZWOBjwAG9gK8LPjQRERERERERKa4O16XihSyvHwl4rmE6RUREREREpMRLc/r5m185Jhycc60KMxARERERERERKTnyMmgkZtYFOAcod3Cac+7xggpKRERERERERIq3vNwWcwzeuA13AAb0AE4q4LhEREREREREpBjLy10qmjjnrgW2OeceAxoDpxdsWCIiIiIiIiKh54rBo6jKS8Lhb///v8ysGrAPiC24kERERERERESkuMvLGA5fmtkJwHPASrwEylsFGZSIiIiIiIiIFG+5Jhycc0/4Tz81sy+Bcs65HQUbloiIiIiIiEjopRXpTgtFW45dKsysvpnFBLy+FpgIPGFmVQojOBEREREREREpng43hsNY4B8AM2sBPA38G9gBjCv40ERERERERESkuDpcl4pSzrmt/vNewDjn3Kd4XSu+L/DIRERERERERELMqUtFvh2uhUMpMzuYkGgDzAmYl5fBJkVERERERETkGHW4xMHHwDdmlox3a8z5AGZ2Gl63ChERERERERGRbOWYcHDOPWlms4FYYKZz7mA7kjDgjsIITkRERERERCSU0kIdQDF22K4Rzrkl2UxbX3DhiIiIiIiIiEhJcLgxHERERERERERE8kWDP4qIiIiIiIjkIE13qcg3yxiaocBo64iIiIiIiJRcFuoAClKPk7oV+d+0n2yYXCS3QaG0cOhW69LC+BjJxeSNX3LTyT1CHYb43vr9E248uXuowxDg7d8n0bR661CHIcDCP+fw4MlXhzoMAZ78/SPqxTYLdRgCrIxfQPPqbUIdhvjm/zmbfcm/hToMAcpEnELj6q1CHYYAi/+cG+oQpIhSlwoRERERERGRHDg12s83DRopIiIiIiIiIkGnhIOIiIiIiIiIBJ0SDiIiIiIiIiISdBrDQURERERERCQHaaEOoBhTCwcRERERERERCTolHEREREREREQk6NSlQkRERERERCQHzum2mPmlFg4iIiIiIiIiEnRKOIiIiIiIiIhI0KlLhYiIiIiIiEgO0lCXivxSCwcRERERERERCTolHEREREREREQk6NSlQkRERERERCQHaaEOoBhTCwcRERERERERCTolHEREREREREQk6NSlQkRERERERCQHTnepyDe1cBARERERERGRoFPCQURERERERESCTgkHEREREREREQk6jeEgIiIiIiIikoM0jeGQb2rhICIiIiIiIiJBp4SDiIiIiIiIiASdulSIiIiIiIiI5MA5danIL7VwEBEREREREZGgU8JBRERERERERIJOXSpEREREREREcpAW6gCKMbVwEBEREREREZGgU8JBRERERERERIIuz10qzKw6cFLgMs65eQURlIiIiIiIiEhR4NBdKvIrTwkHM3sG6AX8CBzwJztACQcREREREREROUReWzhcDpzhnNtbgLGIiIiIiIiISAmR14TDb0AZQAkHEREREREROWakqUtFvuU14fAX8L2ZzSYg6eCcu7NAohIRERERERGRYi2vCYcp/kNEREREREREJFd5Sjg45943s+OA0/1JPzvn9hVcWCIiIiIiIiKh51zx71JhZh2Bl4FSwNvOuaezzC8L/Bu4CEgBejnnfj/azw3LY3CXAL8ArwNvAOvNrMXRfriIiIiIiIiIFBwzK4X3W74TcDbQx8zOzlLsBmCbc+404CXgmWB8dp4SDsALQHvnXEvnXAuggx+EiIiIiIiIiBRdDYBfnXO/Oef+AcYD3bKU6Qa87z+fBLQxMzvaD85rwqGMc+7ngy+cc+vx7lohIiIiIiIiIiFkZgPN7NuAx8CA2dWBPwJeb/KnkV0Z59x+YAdQ9Wjjyuugkd+a2dvAh/7rvsC3R/vhRcGFLetx06MDCSsVRtz4mXz6xqRM80sfV5ohLw3l1PNOI3VbKs/d/gxJm5LS50dUi+S12W8w/qWP+GLc54UdfolzTsu69B5xHWGlwpg/YTbT3/wi0/w6Dc6i14gB1DjzJMbdMZqVXy1Jnzf4/Qc55cI6/Lr8J1694Wnk6JzTsi59ArbFV9lsi94jrvO3xUusCNgWTa5qSZdB3QH472uTWPTpN4UZeol01+ODaNy6IXv+3sOTQ55l/Q+/HFLmjPPq8OBL91G2XFkWz1nK6BGvAXD7QzfTtF1j9v2zjz83xPPU0GfYtXN3YX+FEqFOy/PpMuJawkqF8e2Eucx7c2qm+Sc3OJMuI64h+sxaTLjjVdZ+tQyA2o3PpsvD/dLLRZxajQl3vMa6mSXiUBoy9zwxmGZtGrPn7z08ctdT/LRm/SFlzjr/DB4dPZxy5cqyYPZinnv4ZQBuHnY9V/TtyraU7QC8NmosC+csOWR5ybvBj99Oo9YN2fv3Xp7KoZ46/bw6DH/pXsqWK8uSOUt5ecTrANz20ECatGvM/n/28+eGzYwa+qzqqXx46KkXmbdwGVVOPIEvPhxzyHznHKNGj2H+4uWUK1eWJx8cxtlnnAbA5GlxjH1/PAA39+9Nt87tCjX2kmrI43fQxD9+PzHkmRyO36fzsH/8XjRnKS+NeBWAQQ/dTLN2Tfzj92ZG6vgdMsXhtpjOuXHAuFDHkVVeWzjcCvwI3Ok/fvSnFWthYWHcPPJWHuv/CIPa3Ebzy1pSs07NTGXa9WrPrh27uaXFQKa8PZn+DwzINP+GETey8usVhRh1yWVhYVz9+A28POBJRrQbQoPLmhJ7Wo1MZbZuTubdu19n2eQFhyw/Y+xk3hnyamGFW6JZWBh9H7+R0QOe5OF2Q2hwWbMct8XSLNuiQuWKdB3ck6cuf4Anu91P18E9KR9eoTDDL3Eat25IjdrV6dXsGp6970XuHnVXtuXuHjWEZ+59gV7NrqFG7eo0atUAgOXzVnBN6+vp3+4m/vjtD64ZdHUhRl9yWJjR9fHreH/As7zc7h7Ov6wJkadlvjiwfXMyk+4ew+rJizJN/7/FP/Ja5+G81nk47/R5kn1//8Ov81YXZvglTtPWjah1Sk26NenNyHue44Gn78623ANPD2Pk3c/SrUlvap1SkyatG6XP+8+4ifRpdx192l2nZMNRatS6ATVq16BPs2t59r4XGTZqcLblho26i2fvfZE+za6lRu0aNAyop/q3voEB7W7ij9820U/1VL5c3rkdY14cmeP8+YuXs3HTZqZNeIdH772TJ573EtM7dqby5rsf8fFbo/n4rdG8+e5H7NiZWlhhl1iNWzekZu3q9GjWj6fve4F7Rw3Jtty9o+5i1L3P06NZP2oGHL+XzVtB39bXcU27G9n42yauHdS3MMOXkuVPIPCHbg1/WrZlzKw0UBlv8MijkqeEg3Nur3PuRefclf7jJefc3qP98FCrU/d0En6PJ3FjIvv37Wf+1Hk0aN8oU5mG7RsxZ9JsABZOW8D5TS/INC9xYwIb128s1LhLqtp1T2PLhgSS/0jiwL79LJ+6kLrtL85UJmXTFv78aWO2I8X+tOgH9uz+u7DCLdFq1z2NpIBtsWzqQuq2r5+pTMqmLWz6aQPOpWWafk7LC/hxwSp279jFXzt38+OCVZx7Sd1CjL7kadahCdMnxQGwduU6KlWuSNWoKpnKVI2qQoVK5Vm7ch0A0yfF0bxjUwCWzfuWAwfS0pePio0sxOhLjhp1T2PrhkS2/ZHEgX0HWD11MWe1vyhTme2bkkn86Y9D9otA53ZuyPqvV7Fvzz8FHXKJdknH5nz5yXQA1qxcS6XwikREZW75GRFVlQqVKrBm5VoAvvxkOq06Ni/0WI8FzTo0ZfqkmQD8uHIdFQ9TT/2YXk/NTK+nls9bEVBP/UhkbEQhRl9yXFz3PCqHV8px/twFS7isYxvMjAvOPYvU1F1sSd7KwqUraFz/QiqHV6JyeCUa17+QhUt1Qe1otejQlK/8/WLtynVUrFwhh/2iQvrx+6tJM2nZsRmQ9fj9o47fcjSWA3XMrLZ/98newJQsZaYA/f3n3YE5Lgi35zhswsHMJvr/rzGz1VkfR/vhoVY1pirJm7ekv06JT6ZqdOaTlSoBZdIOpLE79S8qnRhOufLluPLW7owf/XGhxlySnRBdha2bM5Jo2+K3ckL0UXcbknw4MboK2zYnp7/eFp/CidFVDrNEhhOiq2o7BllkTARJmzO6ciXFbyEyJuLQMvEZ9dmWbMoAdOndicVzlxVcsCVYePSJ7Aj4294Zv5XKedwvAp3XtTGrpyzKvaAcVlRMBImZ9oukQ36kRsZGkBRwnE+KTyIqYL/odf2VTJj9Ho+8+ACVKuf8I01y59VTmeugiCx1UERMBFsy1VPJOdZTS+cuL7hgj2GJW1KIicpY59FRESRuSSZxSzIxURk/ZqMjvelydCKz1FPZ/c1nPX5nd4wHuLR3JxbPXVpwwcphuWLw77Dxe2MyDAJmAOuAic65tWb2uJld5hd7B6hqZr8CQ4H7g7HuchvD4WB7uEuP5E39ASoGAowdOzYfYRV9vYdczZR3vmDPX3tCHYqISJ5ce2dfDuw/wMzPZoU6lGNWpcgTiDmjJr+oO0XIffL+57z10ns457jtvpsY+sggHhs6KtRhHfOuufNq1VMiWfT3j98ztF/IUXDOTQOmZZk2IuD5HqBHsD/3sAkH51y8//+GI3nTLANWuP+OzNpao2hISUgholpGNrdqbAQpiZm7qWz1y6QkpBBWKowKlcqTum0np194Bk06N6X/A9dRIbwCzjn+2buPae9/Wdhfo8TYnriVKtUyroSfGFuF7YlH3W1I8mFb4lZOrJaRXT8xtirbErfmadntiSmc0eicgGWr8POStUGPsaS7sn83LuvbBYB13/9MVLWo9HlRsZFsSch85WlLQnKmppaRWcp07tmBpm0bcWfP7Pu5S+52Jm6jckAdFR5bhR153C8OOvfSRvw441vS9h8IdnjHhJ4DruSKvl0BWLtqHdGZ9osotsRn2S/ik4kKOM5HxUaR5O8XW5O3pU//7MMpvPzBswUZeol0Rf9udO3bGYCfvv8507qOjI0kOUs9lZyQTGSmeioiUz3VqWcHmrRtzF2qpwpMdGRVEpIy1nliUjLRkRFER0aw/LuMRGjilmTqX3h+KEIs9q7qf3nA8funTPVU1r95OPT4nfUY7x2/G3NHz2EFHLlIwcjTGA5mlmpmO7M8/jCzz83slIIOsqD8smo9sbWrEVUzmtJlStO8awuWxWVuqrQsbimtu7cBoGnnZqxe5FXGw7vfx8CmNzCw6Q1M/dcUJr02UcmGo/T7ql+JOjmWiBpRlCpTmvpdm7IqTiO4h8Lvq34lOmBbNOjalFVxeWveuvabVZzd/ALKh1egfHgFzm5+AWu/WVXAEZc8n70/mQHtBzKg/UDmzVhAx+7eaOHn1DuLXTt3k5KU+YduStJWdqf+xTn1zgKgY/d2LJjhNdtveEl9rr61F/cNeIi9e4r98Dsh8+eq/1H15BhOrBFJqTKlOL9rY36KO7I+zudf1phVU9WdIr8mvvdZ+iCPX381n0t7dATgvHrnsCt1F8lJmZPUyUkp7E7dzXn1vCTopT068vX0+QCZxnto3bkF//vpt0L6FiXH5+9P5vr2N3N9+5uZP2MhHbu3B+DsXOqps9PrqfYsmLEQgAZ+PfWA6qkCdUmzRkyZPhvnHKt+WEfFihWIjKhC04YXsWjZSnbsTGXHzlQWLVtJ04YX5f6GcohP3/+C/u1von/7m5g3YyGd/P3inHpnsTvH/WJ3+vG7U/f2zPP3i0aX1Kffrb25d8CD2i9CLM25Iv8oqvJ6W8zRePfq/AgwvEEmTgVWAv8CLimA2Apc2oE0xj08hkc/eJywUmHMnhDHH+s3cvXQvvy65heWxS0jbsJMhowexph540jdvovnBz0T6rBLrLQDaXw04h3u+veDWKkwFk6cy+ZfNnHZkF5sWPM/Vs36lpPPP5Xbxt5D+coVOL/NRXQb0pNH2g8F4N6JjxNzanXKVijHs4vH8P59b7J2nn7o5oe3Ld7mrn8/RFipMBZOnMPmXzbRbUgvfs+0Le6lQuUKXNDmYi4b0otH2g9h945dfPnKpzw0xbs16ZevTGL3jl0h/kbF2+LZS2ncuiETF37Inr/38NTQjCux780cx4D23m2WXxg+Ov22mEvmLmPxHC+BOnTknZQpW4bR458DvIGnnrt/dKF/j+Iu7UAaU0e8x4B/34+VCmPlxK9J+uVP2gzpzp9rfuOnWSupfv4p9B07hOMrV+DMNvVoM6Q7r7S/F4ATakRwQmxVfl+yLsTfpGRYMHsxzdo0ZvLiCez5ew+PDnkqfd7Hce/Sp911AIx64AUeG/2gf7u5Jel3oxj88K2cfk4dcI7NfyTw5L3PheR7lBSLZy+lUeuGjF/4AXv+3sOooRnr818zx3J9+5sBeHH4yxm3xZy7jCVzvDFlhoy8gzJly/DieK9+W7tyHS+onjpi9zzyNMu/W8327Ttpc3k/brvhGvbv3w9Aryu60KJxfeYvXk6nntdzfLlyPDHcu2tC5fBK3DygD71v9HpT33Ld1YcdfFLyZtHsJTRp3ZBPFn7I3r/3MnJoxu+I92e+Rf/2NwHw3PDRPPTS/ZQtd1ym4/ewkYMpU7YML49/HvCO38/e/1LhfxGRo2B5GXjSzFY55y7IMu1751zd7OZl4brVOqIhIKSATN74JTedHPRuOZJPb/3+CTee3D3UYQjw9u+TaFq9dajDEGDhn3N48GTdDq8oePL3j6gX2yzUYQiwMn4Bzau3CXUY4pv/52z2JatFTFFQJuIUGldvFeowBFj851wLdQwFqUX1NkW3CYFv3p+zi+Q2yFOXCuAvM+tpZmH+oydwcLTEIr/yRURERERERPLDFYNHUZXXhENf4BogCUj0n/czs+Pxbq8hIiIiIiIiIpIuT2M4OOd+A7rmMHtB8MIRERERERERkZIgr3epON3MZpvZD/7r883soYINTURERERERCS00nBF/lFU5bVLxVvAA8A+AOfcarw7VYiIiIiIiIiIHCKvCYfyzrllWabtD3YwIiIiIiIiIlIy5DXhkGxmp+IPgGlm3YH4AotKRERERERERIq1PA0aCdwOjAPONLM/gf/Du3OFiIiIiIiISIlVlMdIKOqO5C4Vbc2sAl6riL/wxnDYUICxiYiIiIiIiEgxddguFWYWbmYPmNlrZtYOL9HQH/gV6FkYAYqIiIiIiIhI8ZNbC4cPgG3AYuAm4EHAgCucc98XbGgiIiIiIiIioeWculTkV24Jh1Occ+cBmNnbeANF1nLO7SnwyERERERERESk2MrtLhX7Dj5xzh0ANinZICIiIiIiIiK5ya2FwwVmttN/bsDx/msDnHMuvECjExEREREREQkh3aUi/w6bcHDOlSqsQERERERERESk5MitS4WIiIiIiIiIyBHLrUuFiIiIiIiIyDHLqUtFvqmFg4iIiIiIiIgEnRIOIiIiIiIiIhJ06lIhIiIiIiIikgPn1KUiv9TCQURERERERESCTgkHEREREREREQk6JRxEREREREREJOg0hoOIiIiIiIhIDtJ0W8x8UwsHEREREREREQk6JRxEREREREREJOjUpUJEREREREQkB7otZv6phYOIiIiIiIiIBJ0SDiIiIiIiIiISdOpSISIiIiIiIpID3aUi/9TCQURERERERESCTgkHEREREREREQk6dakQERERERERyYFTl4p8UwsHEREREREREQk6JRxEREREREREJOjUpUJEREREREQkB2lOXSrySy0cRERERERERCTolHAQERERERERkaBTlwoRERERERGRHOguFfmnFg4iIiIiIiIiEnTmCn4ADKWDRERERERESi4LdQAF6ZzohkX+N+3axKVFchsUSpeKtjU7FMbHSC5m/TGDWlXOC3UY4tu4dQ1nRTUIdRgCrEtaRqsa7UIdhgBzN8VRu+oFoQ5DgP9LWUW3WpeGOgwBJm/8kitqdQ11GOL7fONUGldvFeowBFj851z2Jf8W6jAEKBNxSqhDkCJKYziIiIiIiIiI5EC3xcw/jeEgIiIiIiIiIkGnhIOIiIiIiIiIBJ26VIiIiIiIiIjkQLfFzD+1cBARERERERGRoFPCQURERERERESCTl0qRERERERERHKgu1Tkn1o4iIiIiIiIiEjQKeEgIiIiIiIiIkGnLhUiIiIiIiIiOdBdKvJPLRxEREREREREJOiUcBARERERERGRoFOXChEREREREZEc6C4V+acWDiIiIiIiIiISdEo4iIiIiIiIiEjQKeEgIiIiIiIiIkGnMRxEREREREREcqDbYuafWjiIiIiIiIiISNAp4SAiIiIiIiIiQacuFSIiIiIiIiI5cC4t1CEUW2rhICIiIiIiIiJBp4SDiIiIiIiIiASdulSIiIiIiIiI5CBNd6nIN7VwEBEREREREZGgU8JBRERERERERIJOXSpEREREREREcuCculTkl1o4iIiIiIiIiEjQKeEgIiIiIiIiIkGnLhUiIiIiIiIiOdBdKvJPLRxEREREREREJOiUcBARERERERGRoFOXChEREREREZEc6C4V+acWDiIiIiIiIiISdEo4iIiIiIiIiEjQKeEgIiIiIiIiIkGX6xgOZlYKmOWca1UI8YiIiIiIiIgUGWkawyHfcm3h4Jw7AKSZWeVCiEdERERERERESoC83qViF7DGzOKA3QcnOufuLJCoRERERERERKRYy2vC4TP/ISIiIiIiInLMcKhLRX7ladBI59z7wERgiXPu/YOPgg2tcNS/5GLe/fpt3p//Lr1v63nI/DLHleGhN4bz/vx3eXXKy0TXiAYg/IRKPD/hWab+9AWDnri9sMMu0R4bdT/zvv0vM+Z/yrnnn5VtmXsevIMla+JYt3Fppuk33nYtsxd/wYz5n/Lx529RvUZsYYRcYg1/chjTl37KF1//h7PPOyPbMmeffyaTv/6I6Us/ZfiTw9Knn3luHcZPe4fP5nzIJzPf57wLzy6ssEukOx6/jQ8XvMfbcWOpc+5p2ZY5/bw6vDNrHB8ueI87Hr8tfXqlEyrx3EdP88H893juo6epWLliYYVdIj0y6j7mLp/KV/M+4Zzzz8y2zN0PDmLh6hn8sGFxpukNGtdj6pzx/JK4gk5d2xZGuCXWhS3r8cbcMYyZN46rbut+yPzSx5XmntfvZcy8cTw3+QWiakRlmh9RLZLx6z7h8oFXFFbIJdqFLevx2tw3eWPeWK7MYXsMe/1e3pg3lmcmP0+kvz0ia0Qxfv0kXvzqZV786mVueeq2Q5aVIzfk8Tv4ZMGHfBD3NqefWyfbMmecdzofznqHTxZ8yJDH70ifPuihmxn/zft8EPc2T7/9OBXDKxRW2CXKQ0+9SIsuvbm83y3ZznfO8dRLb9Kp5/Vcce2t/Pjzr+nzJk+Lo3OvG+jc6wYmT4srrJBFCkSeEg5m1hX4Hpjuv65rZlMKMK5CERYWxh0jb2f4tQ9xQ+ubaNWtFbXq1MpUplPvDqRu30X/5tfx6dufcdPwGwD4Z+8/vPf8+4wd+VYoQi+xWrVtzsmnnkSLi7tw/5DHePKFh7ItN2vGN1zWts8h09euXkeX1r3p0Pwq/jsljuGPDS3okEusFm2acNIpNenY8CoeGTaKEc/el225R569jxHDnqJjw6s46ZSaNG/dGIC7R9zB68+/zZWt+/HqM2O5e8Qd2S4vuWvYugHVa1enX7MBvHDfaIaMyr43212j7uT5e1+iX7MBVK9dnQat6gNw9e29WLnwO65pPoCVC7/j6tt7F2b4JcolbZtx8im1aFW/Kw8MfZyRz+dcR13eru8h0//clMA9gx5myqdfFXSoJVpYWBg3j7yVx/o/wqA2t9H8spbUrFMzU5l2vdqza8dubmkxkClvT6b/AwMyzb9hxI2s/HpFIUZdcoWFhTFw5C080f9R7mxzO80ua0GNLNujba/27N6xi9ta3MzUtydzbcD2SNyQwNBOgxnaaTBjhr9RyNGXPI1bN6Rm7er0aNaPp+97gXtHDcm23L2j7mLUvc/To1k/atauTqNWDQBYNm8FfVtfxzXtbmTjb5u4dtChdZnk7vLO7Rjz4sgc589fvJyNmzYzbcI7PHrvnTzx/GsA7NiZypvvfsTHb43m47dG8+a7H7FjZ2phhS0SdHm9LeajQANgO4Bz7nvglAKJqBCdUfcMNv++mfiNCezft5+vp3xN0/aNM5Vp0r4xMyd5mcV5/53PhU3rArDn7738sHwt/+z9p7DDLtHad27Fp+O9XNZ3364mPLwSUdERh5T77tvVJCUmHzJ98YLl7Pl7T3qZ2GrRBRtwCda6UwsmT5wGwKoVPxBeuRKRUVUzlYmMqkrFShVYteIHACZPnEabzi0BcA4qVvKuilQMr0hSwqHbS/KmafvGzJw0C4B1K9dRIbwiVaKqZCpTJaoKFSqWZ93KdQDMnDSLZh2aANCkfRNmfOLVYzM+iaOpP12OXLtOrfhswlQAvv92jbdfZFNHff/tGrZkU0f9+cdmfvrxF9LS0go81pKsTt3TSfg9nsSNiezft5/5U+fRoH2jTGUatm/EnEmzAVg4bQHnN70g07zEjQlsXL+xUOMuqerUrUN8wPZYMHUeDdo3zFSmQfuGzPW3x6JpCzNtDwmuFh2a8tWkmQCsXbmOipUrUDXLMaNqVBUqVKrAWv+Y8dWkmbTs2AyAZfO+5cCBNH/5H4mKjSzE6EuOi+ueR+XwSjnOn7tgCZd1bIOZccG5Z5GauostyVtZuHQFjetfSOXwSlQOr0Tj+heycKmSo6HmnCvyj6IqrwmHfc65HVmmFfuzpYiYqiRt3pL+ekt8MlVjMp84Vo2JYItfJu1AGrtTdxN+YnihxnksiYmNIv7PhPTXCZsTiYmNOswSOevV70rmzloQrNCOOdExUSRsTkx/nbA5iags2yIqNorE+KT014mbk4iO8cqMeuhF7n7kTuZ8N5V7H72Tl558vXACL4EiYiJI2pyxnpPjk4nIUldFxESwJT7jB+6W+C3pZapEnMjWpK0AbE3aSpWIEwsh6pIpOjaK+D8z9ov4o6ijJP+qxlQlOeD4nRKfTNXozAnRKgFlvOP3X1Q6MZxy5ctx5a3dGT/640KNuSTz1nVG/ZMSn3LI9qgaUCbtQBp/pe6mkn8+FVUzmhemjWbkxFGc1UDd745WZEwEiQHHjC3xyURmOWZExkSQFJ+xDyXFbzmkDMClvTuxeO7SQ6bL0UvckkJMVMY6j46KIHFLMolbkomJykjyREd600WKq7wOGrnWzK4GSplZHeBOYFFOhc1sIDAQYOzYsUcdpMiRuqLHpZx/4dn0vPS6UIdyzOo94CqeHvEScV/OpeNlbRk5+iGu7z4o1GEJFOksuEhB6z3kaqa88wV7/toT6lAE2Ja0lYGNrid1eyqnnHcqD7z1IHe2vZ2/d/0d6tCOef3v7MuB/QeY8dmsUIciIsVYXhMOdwAPAnuBj4EZwBM5FXbOjQPGHXw58YlPjybGApOckEJUtYwMYmRsBClZmn2nJCQTWS2S5IRkwkqFUaFSBXZu21nYoZZo197Qmz7XXgXA6u9+ILZ6TPq8mGrRJARcQc+LZi0bMWjYTfS89Dr++WdfUGMt6a6+vjvd+10OwA/f/UhMQJeUmGpRJGXZFknxSUQHXN2NrhZFYoJX5vJeXXjqwRcAmD5lFk+8NLyAoy9ZLu9/GV2u7gzAT6t+JqpaFLAWgIjYCJKz1FXJCclExmZcKYmMjUwvszV5G1WiqnitG6KqsC1le6F8h5Limht60fuaKwFY/d1aYqtn7Bex+aij5OilJKQQEXD8rhobQUpiSqYyW/0yKQkp/vG7PKnbdnL6hWfQpHNT+j9wHRXCK+Cc45+9+5j2/peF/TVKDG9dZ9Q/VWOrHrI9UvwyB7dH+UoVSPXPp1L/8fqn/7bmfyRsSKDaKdX53+pfkby7qv/lXNa3CwDrvv+J6GoZx+bI2Ai2ZDlmbElIztRVIio2MlOZzj070LRtY+7oOQwpGNGRVUlIyljniUnJREdGEB0ZwfLvVmdM35JM/QvPD0WIEiBNd6nIt7zepeIv59yDzrn6QEPgGedcsb808POqn6l+cnViakZTukxpLrnsEhbFLclUZlHcEtp3bwdAiy7N+X7hqlCEWqL9+53xdGrZg04tezDjv3O4qvdlAFx48fmk7tyV7VgNOTnnvDMZ9eIIbrj6DlKStxZUyCXWR/+axJWt+3Fl637M/uobuvX0fvBecNG5pO7cxZakzCeQW5JS2JW6mwsuOheAbj07M+ereQAkJWyhfpN6ADRqXp8Nv/1RiN+k+Pvi/Snc1OEWbupwCwunL6R9d++OBmfVO4vdqbvTu0gctDVpK7t3/cVZ9bw7u7Tv3paFM707JCyKW0yHHl491qFHOxbNzLGBmmTjg3cm0OWSXnS5pBczp83lyl5dAah78XnefnEEdZQExy+r1hNbuxpR/vG7edcWLIvL3Ox7WdxSWndvA0DTzs1Yvcg7gR/e/T4GNr2BgU1vYOq/pjDptYlKNhylX1b9kml7NOvaguVxyzKVWR63lFb+9mjSuSlr/O0RXiWcsDDvdDS6VjSxtauRuCEBOTKfvv8F/dvfRP/2NzFvxkI6dW8PwDn1zmL3zt2kZDlmpCRtZXfqbs7xjxmdurdn3oyFADS6pD79bu3NvQMeZO+evYX7RY4hlzRrxJTps3HOseqHdVSsWIHIiCo0bXgRi5atZMfOVHbsTGXRspU0bXhRqMMVybc8tXAws4+AW4ADwHIg3Mxeds49V5DBFbS0A2m8+vDrPP3hU4SVCmP6hJlsWL+B/sOuZf3q9SyOW8JX46dz/+h7eX/+u6RuT+XJ259KX/7DRe9TvlIFypQpTdMOjbmv73A2/qIBqI7GnLj5tGrXgvkrpvH333u4e1DGCPBfffMJnVr2AGD4o0Po1r0Lx5cvx9IfZjH+g0956Zk3efCxYZSvUJ433/WurG/eFM8NfbMf0V8O75tZC2nRtgkzln3Gnr/2MHxwRqOmz+Z8yJWt+wHw+H3PMuqVEZQ9vizzZy9i3mzvx+yIYU8xfORQSpUuzd49exkxbFRIvkdJsGTOMhq2bsiHC95n7569PDP0+fR5b80Yw00dvFtujR7+Kve/eDfHlSvLsq+Xs3SOd8L/8WvjeWTMw3Tu3YnETYk8dmvOo2bL4c2Nm0+rds34+tsv+fvvPdx7x4j0ef/92ktMANz/yF1c1r0zx5cvx6I1M5nwwWe8/OwYzr/wHMb8+yUqVw6nTYeW3HX/bXRoemWovk6xlXYgjXEPj+HRDx4nrFQYsyfE8cf6jVw9tC+/rvmFZXHLiJswkyGjhzFm3jhSt+/i+UHPhDrsEivtQBpvPTyGRz54zN8es/hj/Ub6+NtjedwyZk2I467RQ3lj3lh2bd/FC4OeBeDshufSZ1hfDuzbT1qaY8zw19m1Y1eIv1Hxtmj2Epq0bsgnCz9k7997GTk042///Zlv0b/9TQA8N3w0D710P2XLHceSuctYPMdL2g0bOZgyZcvw8njvWLN25Y88e/9Lhf9Firl7Hnma5d+tZvv2nbS5vB+33XAN+/fvB6DXFV1o0bg+8xcvp1PP6zm+XDmeGO7dTaRyeCVuHtCH3jcOBuCW664+7OCTIkWd5aUvr5l975yra2Z9gXrA/cAK51xe2ve4tjU7HGWYEgyz/phBrSrnhToM8W3cuoazohqEOgwB1iUto1WNdqEOQ4C5m+KoXVWj1xcF/5eyim61Lg11GAJM3vglV9TqGuowxPf5xqk0rt4q1GEIsPjPuexL/i3UYQhQJuIUC3UMBSki/PQi36cieef6IrkN8nqXijJmVga4HJjinNsH6sgiIiIiIiIiItnLa8JhDPB/QAVgnpmdBGjkRBERERERERHJ1mHHcDCzoQEvX8Jr1dAPWACoLZmIiIiIiIiIZCu3QSOzG6HkJLxbZD4KjA92QCIiIiIiIiJFRVoexj2U7B024eCceyy76WZWBZiFEg4iIiIiIiIiko28juGQiXNuK1AkR8EUERERERERkdDLrUtFtsysFbAtyLGIiIiIiIiIFClOXSryLbdBI9dw6O0vqwCbgWsLKigRERERERERKd5ya+FwaZbXDkhxzu0uoHhEREREREREpATIbdDIDYUViIiIiIiIiEhRk3ZIo3/Jq3wNGikiIiIiIiIicjhKOIiIiIiIiIhI0OXrLhUiIiIiIiIixwLdpSL/1MJBRERERERERIJOCQcRERERERERCTp1qRARERERERHJQZq6VOSbWjiIiIiIiIiISNAp4SAiIiIiIiIiQaeEg4iIiIiIiIgEncZwEBEREREREcmBQ2M45JdaOIiIiIiIiIhI0CnhICIiIiIiIiJBpy4VIiIiIiIiIjnQbTHzTy0cRERERERERCTolHAQERERERERkaBTlwoRERERERGRHDh1qcg3tXAQERERERERkaBTwkFEREREREREgk5dKkRERERERERy4FCXivxSCwcRERERERERCTolHEREREREREQk6NSlQkRERERERCQHuktF/qmFg4iIiIiIiMgxysyqmFmcmf3i/39iNmXqmtliM1trZqvNrFde3lsJBxEREREREZFj1/3AbOdcHWC2/zqrv4BrnXPnAB2B0WZ2Qm5vrC4VIiIiIiIiIjk4BrpUdAMu8Z+/D3wN3BdYwDm3PuD5ZjNLAiKB7Yd7Y7VwEBERERERESnGzGygmX0b8Bh4BItHO+fi/ecJQHQun9UAOA74X25vrBYOIiIiIiIiIsWYc24cMC6n+WY2C4jJZtaDWd7HmVmOTTrMLBb4AOjvnEvLLS4lHERERERERERKMOdc25zmmVmimcU65+L9hEJSDuXCgf8CDzrnluTlc9WlQkRERERERCQHrhg8jtIUoL//vD8wOWsBMzsO+Bz4t3NuUl7fWAkHERERERERkWPX00A7M/sFaOu/xswuNrO3/TI9gRbAADP73n/Uze2N1aVCRERERERE5BjlnEsB2mQz/VvgRv/5h8CHR/redgzc4iMozGygPxCHhJi2RdGhbVF0aFsULdoeRYe2RdGhbVF0aFsUHdoWUtKpS0XeHcltRaRgaVsUHdoWRYe2RdGi7VF0aFsUHdoWRYe2RdGhbSElmhIOIiIiIiIiIhJ0SjiIiIiIiIiISNAp4ZB36ltVdGhbFB3aFkWHtkXRou1RdGhbFB3aFkWHtkXRoW0hJZoGjRQRERERERGRoFMLBxEREREREREJOiUcRERERERERCToSmTCwcxqmNlkM/vFzP5nZi+b2XG5LDO8sOIrDsysqpl97z8SzOzPgNeHXZf+8gPM7LU8lOtkZt+a2Y9m9p2ZvRCcb5Dj571nZt1zKTPAzKod4fuebGY/5COey83MmdmZ2b2XmV1iZl9mWaZDwLbYZWY/+8//fQSfm+t6OMyyx/Q2Oxr+tv4w4HVpM9uSdRvn430fNbO7jz7CbN97V0G8b0Eys2gz+8jMfjOzFWa22MyuKMTPP2S/DfL7n2BmtxXW54WKmR0IqOu+N7P7/elvm9nZ+Xi/umbWOR/LfW1mFx/pcqFgZg+a2VozW+2vs4a5lD+i7+b/rTU5+kiPnF/PBZ6LPG1mt5jZtUF47zydsxQlAfvHD2Y21cxOyOf7VDOzSUEOT8j+PKMgj9ciRVWJSziYmQGfAV845+oApwMVgSdzWVQJhwDOuRTnXF3nXF1gDPDSwdfOuX+C8Rlmdi7wGtDPOXc2cDHwazDe+ygNAI7ox+tR6AMs8P/PE+fcjIBt8y3Q13991CddudE2O2q7gXPN7Hj/dTvgzyN5A/McVd1tZqWPZvmizD8GfAHMc86d4py7COgN1AhpYMF1AnBbboVKgL8Djjt1nXNPAzjnbnTO/ZiP96sLHHHCobgws8bApUA959z5QFvgjyB/zCVASBIOvsBzkfudc2Occ3lOtpcwB/ePc4GtwO35eRPn3GbnXL4uQEjhK8nHbym5SlzCAWgN7HHOvQvgnDsADAGuN7PbAjPYZvaln61/GjjezxT/x593rX+FYJWZfeBPO9nM5vjTZ5tZLX/6e2b2ppkt8a+oXWJm/zKzdWb2XsDntfevtK00s0/MrGKhrZUgMLObzGy5v04+NbPy/vQefoZ9lZnNy2a5Lv73jsgy617gSefcT+BtK+fcm/4yR7uud5nZS/6VntlmFplNXBeZ2TfmXQGdYWax5l1Jvxj4j//3cHx25QKWX2Vmq8jHgd7f/s2AG/B+EB0Vf71863/nxwKmP21ea4TVZvZ8Nss94a/XUrmV5RjfZkEyDejiP+8DfHxwhmW58uHvVyf7j5/Na8XyA1DTvCuZ681sAXBGwDKnmtl0/7vPN7/1jL8dxpjZUuDZw5Sr7e+va8xsZIGvjeBrDfzjnBtzcIJzboNz7lVI/zudb149vNL8q7X+3+U35rWO+83fF/qa2TJ/XZzql4s0r/5b7j+a5jUwy+EYYGa/m9lj/vQ1Adsi0szi/H3ibTPbYF49+jRwqv/3/pz/9hXNbJKZ/WRm/zEzC8bKLIos4Kr8YdZpfTNb5O/vy8ysMvA40Mtfb73MrIJfFy0zr7VWN3/Z481svF9HfQ4cn2MwRUsskOyc2wvgnEt2zm0GMLMR/t/rD2Y2LsvfxzWWcaW8gV++ipl9YV59vsTMzjezk4FbgCF++eZm1tXMlvrrb5aZRfvL5/S3i5n189f592Y21rxjTym/jvrB3weG5OULm19nmtdabLmZXeJPH2VmT+b0ef7068yrQ5cBed6Pi6jFQHU47DHgVH9brjGzkea3XrPMrSrLmdm7fpnvzKyVP32AmX3mv+8vZvZsiL5niWFmd1rG+dZ4f1pOddIAM5tiZnOA2SENXCQ/nHMl6gHciZcBzzr9O3/eawHTvgQu8Z/vCph+DrAeiPBfV/H/nwr0959fj9eKAuA9YDxgQDdgJ3AeXkJnBd5VlQhgHlDBX+Y+YESo11ce1+mjwN1A1YBpI4E7/OdrgOr+8xP8/wfgXQm/ApgPnJjN+64ELsjhM/O9rv1yDu/KP8CIg9vdX747UAZYBET603sB//Kffw1c7D8/XLnVQAv/+XPAD0e4XvsC7/jPFwEX+c9PPvheeFeTvjzMewTGevDvtJQ//XygKvAzGXekOSHLengOrwWL5VRW2yyo+9Iuf7tMAsoB3wduY/x9LaD8D/7fw8lAGtDIn34R3n5XHgjHa2Vytz9vNlDHf94QmBOwHr8ESuVSbgpwrf/8dgLqxuLwIIdjQMD88kA5/3kd4NuAfW073o+2sngtTx7z5w0GRvvPPwKa+c9rAeuy+Yz0bRowLcdjAPA7GfXpbcDb/vPXgAf85x3x9pEIAuqIgM/bgdeKIwzvx0ezUG+LIGzLA/4+cvDRy5/+NV6SMdt1ChwH/AbU96eHA6Xxj0sB7/8UXmst8FqNrAcqAEPJqDPOB/bj1y9F+YHXmvN7/3u8AbQMmFcl4PkHQNeAdfmW/7wFGceeV4FH/Oetge/954+SuY46kYxjxo3AC7n87Z6Fd6wo4897A7gWr06LC3jfE7L5fo/i7ZcH/x46BMaDd+62Dq9lx3f+30FOnxcLbAQi/XILA/82isMDv27GO+Z/AnT0X+dUt38J9PGf3xKw/MkB231YwN/+mf46KufvO78Blf3XG4CaoV4HRf1Blro64O/4bmAzUNafdoL/f0510gBgEwH7sR56FKeHmuVkrzXwiXMuGcA5t9Wf3hi40n/+ARCY4Z3qnHNmtgZIdM6tATCztXgVTg3gbGChf2HhOLyTwuLkXPOueJ6Ad2Izw5++EHjPzCbidWc5qDXeSWF759zOI/yso1nX3+P9OJvgl/8wS1zgXRE+F4jzt0cpID6bOLItZ15fyROccwdbdHwAdDrC79gHeNl/Pt5/veII3yNQTzMbiHdiHYv39/YjsAd4x7w+3oH9vB8GljrnBgKY2Y7DlM2LY2GbHTXn3Gr/SmEfvNYOebXBObfEf94c+Nw59xeAmU3x/6+I19z5k4ALmGUD3uMT59yBXMo1Ba7yn38APHMEMRY5ZvY6Xkuif5xz9fESUq+ZWV28H7SnBxRf7pyL95f7HzDTn74GaOU/bwucHbDews2sonMut7EuGnH4Y8DBv/cVZOxHzfCStjjnppvZtsO8/zLn3CY/9u/x9qsFucRU1P3tvK5jOclpnZ4BxDvnlgMcPP7YoY0+2gOXWUaronJ4SaQWwCv+sqvNbHUwvkxBc87tMrOL8OqHVsAEM7vfOfce0MrM7sVLuFUB1uL9EAe/lZVzbp6Zhft1ZTP8esA5N8e8cZ3Cs/nYGv7nxOKt///zp+f0t9sGL7mw3N8exwNJfiynmNmrwH/J2Peyesk5l976zrxuJAe//1rzWqR+CTR2zv1jZjl9XkPga+fcFv99JpC5LigOjvf39ep4iZa4XOr2xsDl/vOPgOxaMTbDSzbhnPvJzDaQsV5mO+d2AJjZj8BJBL/LTknjDjN9NV7LzC/wugFCznUSeAm5rYgUQyUx4fAj3tXQdP5Bshbe1avAbiTlgvi5e/3/0wKeH3xdGu/ENs45l+e++kXQe8DlzrlVZjYA76oazrlbzBuYqguwwj/hAfgfcArewerbbN5vLd6JwKojjCO3dZ2drJW+AWudc42zK5xbOcvn4EwBy1fBS8icZ2YO70exM7N78vl+tfEy5vWdc9vM66pQzjm337wmsm3w9otB/ucCLAcuMrMqzrmtuZQ96JjdZkE2Be9k7xK8liUH7SfnOmp3Ht43DNh+mB9pu/NYLqeTpOJgLRkJE5xzt5vXlPtgHTQESAQuwFsPewKWzfq3Gfh3e/DvNAyvpUngcnlhHP4YcPCzDpC/Y3Ng7Pl9j+Im23VqZucdwfJXOed+zrJ8kMIrfM7rRvo18LWf4O3vN9d+A6+Vxh9m9iiZ65as+/uR7P+vAi8656aY153h0VzKG/C+c+6BQ2aYXYDXauEWoCdeS7kjdR7euV7U4T7PzC7Px3sXNX875+qa1711Bl6LtPc4fN1+NI7FOuZopeC1AgpUBS8x1wUvudkVeNCvt3KqkxqSt3MAkSKpJI7hMBsob/6oxeb11XsBrxL+DahrZmFmVhNoELDcPjMr4z+fA/Qws6r+e1Txpy8io699X7yuAnm1BGhqZqf571nBzIpbNr0S3pXiMnjfH/D6BTrnljrnRgBbgJr+rA14J/7/NrNzsnm/54DhB9eDv11u8ecdzboG72/7YOLpag690vczEHnw6oiZlQmIMdX/rjmWc85tB7abWbOAGI9Ed+AD59xJzrmTnXM18Q5AzY/wfQ4KxzsY7TCvD20nP96KQGXn3DS8H1oXBCwzHa8v+H/NrFIuZQ86lrdZMP0Lr7n+mizTfwfqAZhZPaB2DsvPAy43r695JbwTloNXcv/PzHr472H+SXwmuZRbSObtWNzMAcqZ2a0B08oHPK+Md/U7DbgGL9l3JGYCdxx84beUyIv8HAMW4v3wwszak3HiGvj3fizLaZ3+DMSaWX1/eiXzBlrLut5mAHeYn2Ewswv96fPw6iDMGyj3/ML4MkfLzM4wszoBk+riHYcPJheS/Xo+6wCBvfzlmwE7/KvY8/H3fz+RkOzXG1nXYWUyBr7tHzA9p7/d2UB3M4vy51Uxs5P8pGCYc+5T4CH8evAIv/+VeD/mWgCv+knmbD8PWAq0NK/lRhmgx5F+XlHht3S7E687xF/kXLcvISMZm9O4UYHb/XS8i3U/51BWcuG3fIs3s9aQ/nuiI975TU3n3Fy8rmCVyWg5nF2dJFKslbiEg3PO4TXj62Fmv+D1f9qDdxeKhXg/6n7Eay65MmDRccBqM/uPc24t3l0tvjFvcLkX/TJ3ANeZ17zyGrx+vXmNawteH6yP/eUX4/WPK04exjtILwR+Cpj+nHkDDP2A96Mz/eq38wYX7IvXvO/UwDdzzq0G7sJbJ+vw+quf4s/O97r27QYa+DG1xhssLPCz/8E76XrG38bfkzHy9nvAGPOaKpY6TLnrgNf9ckd6SawP8HmWaZ9yBHerCOScW4XXZ/UnvKaSC/1ZlYAv/fW4AK9vcuBynwBv4V1xP2xZv/yxvM2Cxjm3yTn3SjazPgWqmNfVZBBe/ZXd8ivxup+sAr7Ca61yUF/gBv+7r8UbNyM7OZUbDNxu3tXR6kf0xYoA/xhwOd6Pif8zb0C49/FO6sC70tvf/95ncuRXje4ELjZvoK8f8a7GZqeNmW06+ABO48iPAY8B7f19ogeQAKQ651LwuhH8YBmDRpZEBwdzPvh4OmCey+m46tcVvfB+dK4C4vB+dM/F6w7zvZn1Ap7A62Kz2t/nnvDf+028QTjX4dVDR9PVrTBVBN43fyA6vO4mj/rJ1rfw6usZZK4vAPaY2Xd44/nc4E97FK8F3Gq8xPTBZMJU4Ap/HTb3y31iZiuA5ID3zOlv90e8hMJM/73j8LoAVsdrlfE9Xpe6Q1pAHI5lDKZ6o3NuPd4YEi/n9Hl+16lH8f5mFuJ1SSi2nHPf4TXR70POdftdwFB/PZyGN+5LVm8AYX79PwEY4PxBSCXfrgUe9v+25+DtGxuBD/31/B3wir+f5lQniRRrBwf6ESlRzGyXc65Y3QXkWKdtJpKZmZUFDvjdnRoDbxZQU+lixT9Jv8w593+5FpaQ0N9u0WNe14u//fGUeuMNIJlTQlpEJGjU/0pERKRoqgVMNLMw4B/gphDHE3JmFgesUbKhyNPfbtFzEd6AuYY3zkV+xsgQETliauEgIiIiIiIiIkFX4sZwEBEREREREZHQU8JBRERERERERIJOCQcRERERERERCTolHEREREREREQk6JRwEBEREREREZGg+39yNNsnBCTskwAAAABJRU5ErkJggg==\n",
+      "text/plain": [
+       "<Figure size 1440x1080 with 2 Axes>"
+      ]
+     },
+     "metadata": {
+      "needs_background": "light"
+     },
+     "output_type": "display_data"
+    }
+   ],
+   "source": [
+    "%matplotlib inline\n",
+    "corrlation_matrix = crewmate.corr()\n",
+    "fig, ax = plt.subplots(figsize = (20, 15))\n",
+    "ax = sns.heatmap(corrlation_matrix,\n",
+    "                annot = True,\n",
+    "                linewidths = 0.5,\n",
+    "                fmt = \".2f\")"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 12,
+   "id": "bdf14fd8",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "crewmate = crewmate.drop(['All Tasks Completed'], axis =1)"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 13,
+   "id": "0ca1ad6d",
+   "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>Team</th>\n",
+       "      <th>Outcome</th>\n",
+       "      <th>Task Completed</th>\n",
+       "      <th>Murdered</th>\n",
+       "      <th>Game Length</th>\n",
+       "      <th>Ejected</th>\n",
+       "      <th>Sabotages Fixed</th>\n",
+       "      <th>Region</th>\n",
+       "      <th>User</th>\n",
+       "    </tr>\n",
+       "  </thead>\n",
+       "  <tbody>\n",
+       "    <tr>\n",
+       "      <th>0</th>\n",
+       "      <td>Crewmate</td>\n",
+       "      <td>1</td>\n",
+       "      <td>3.0</td>\n",
+       "      <td>1</td>\n",
+       "      <td>424.0</td>\n",
+       "      <td>0</td>\n",
+       "      <td>2.0</td>\n",
+       "      <td>0</td>\n",
+       "      <td>0</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>1</th>\n",
+       "      <td>Crewmate</td>\n",
+       "      <td>0</td>\n",
+       "      <td>7.0</td>\n",
+       "      <td>0</td>\n",
+       "      <td>981.0</td>\n",
+       "      <td>0</td>\n",
+       "      <td>1.0</td>\n",
+       "      <td>0</td>\n",
+       "      <td>0</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>2</th>\n",
+       "      <td>Crewmate</td>\n",
+       "      <td>1</td>\n",
+       "      <td>3.0</td>\n",
+       "      <td>0</td>\n",
+       "      <td>693.0</td>\n",
+       "      <td>0</td>\n",
+       "      <td>0.0</td>\n",
+       "      <td>0</td>\n",
+       "      <td>0</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>4</th>\n",
+       "      <td>Crewmate</td>\n",
+       "      <td>0</td>\n",
+       "      <td>4.0</td>\n",
+       "      <td>0</td>\n",
+       "      <td>310.0</td>\n",
+       "      <td>0</td>\n",
+       "      <td>0.0</td>\n",
+       "      <td>1</td>\n",
+       "      <td>0</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>5</th>\n",
+       "      <td>Crewmate</td>\n",
+       "      <td>0</td>\n",
+       "      <td>7.0</td>\n",
+       "      <td>1</td>\n",
+       "      <td>982.0</td>\n",
+       "      <td>0</td>\n",
+       "      <td>0.0</td>\n",
+       "      <td>1</td>\n",
+       "      <td>0</td>\n",
+       "    </tr>\n",
+       "  </tbody>\n",
+       "</table>\n",
+       "</div>"
+      ],
+      "text/plain": [
+       "       Team  Outcome  Task Completed  Murdered  Game Length  Ejected  \\\n",
+       "0  Crewmate        1             3.0         1        424.0        0   \n",
+       "1  Crewmate        0             7.0         0        981.0        0   \n",
+       "2  Crewmate        1             3.0         0        693.0        0   \n",
+       "4  Crewmate        0             4.0         0        310.0        0   \n",
+       "5  Crewmate        0             7.0         1        982.0        0   \n",
+       "\n",
+       "   Sabotages Fixed  Region  User  \n",
+       "0              2.0       0     0  \n",
+       "1              1.0       0     0  \n",
+       "2              0.0       0     0  \n",
+       "4              0.0       1     0  \n",
+       "5              0.0       1     0  "
+      ]
+     },
+     "execution_count": 13,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "crewmate.head()"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 14,
+   "id": "858d7136",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "def user_train_valid_split(users_i,valid_ratio,possible=False):\n",
+    "    n = sum(users_i)\n",
+    "    target = int(np.floor(valid_ratio*n))\n",
+    "    while not possible:\n",
+    "        possible, validation_indices = user_split(users_i,target)\n",
+    "        target -= 1\n",
+    "    training_indices = [i for i in range(len(users_i)) if i not in validation_indices]\n",
+    "    return training_indices, validation_indices\n",
+    "\n",
+    "def user_split(users_i,target):  \n",
+    "    # Reference: https://levelup.gitconnected.com/dynamic-programming-subset-sum-c386126621cd\n",
+    "    n = len(users_i)\n",
+    "    solution = [[False for j in range(int(target+1))] for i in range(n+1)]\n",
+    "    # base cases\n",
+    "    for i in range(n):\n",
+    "        solution[i][0] = True\n",
+    "    # other cases\n",
+    "    for i in range(1,n+1):\n",
+    "        for j in range(1,target+1):\n",
+    "            solution[i][j] = solution[i-1][j]\n",
+    "            if(solution[i][j] == False and j >= users_i[i-1]):\n",
+    "                solution[i][j] = solution[i][j] or solution[i-1][j-users_i[i-1]]\n",
+    "    # check if the subset sum is possible\n",
+    "    possible = solution[len(users_i)][target]\n",
+    "    subset = []\n",
+    "    if not possible: return possible, subset\n",
+    "    # return the subset solution if one exists\n",
+    "    y = len(users_i)\n",
+    "    x = target\n",
+    "    while x != 0:\n",
+    "        if solution[y-1][x] == False:\n",
+    "            subset.append(y-1)\n",
+    "            x -= users_i[y-1]\n",
+    "        else: \n",
+    "            y -= 1\n",
+    "    return possible, subset"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 15,
+   "id": "b5a1d348",
+   "metadata": {},
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "expected valid ratio: 0.25, actual: 0.2498580352072686\n"
+     ]
+    }
+   ],
+   "source": [
+    "crewmates_i = list(crewmate[\"User\"].value_counts(sort = False))\n",
+    "test_ratio = .25\n",
+    "training_indices, test_indices = user_train_valid_split(crewmates_i,test_ratio)\n",
+    "\n",
+    "#check\n",
+    "train_sum = sum([crewmates_i[i] for i in training_indices])\n",
+    "test_sum = sum([crewmates_i[i] for i in test_indices])\n",
+    "print(\"expected valid ratio: {}, actual: {}\".format(test_ratio, test_sum/(train_sum+test_sum)))\n",
+    "\n",
+    "train_data = crewmate.loc[crewmate['User'].isin(training_indices)]\n",
+    "test_data = crewmate.loc[crewmate['User'].isin(test_indices)]\n",
+    "train_data.reset_index(drop=True, inplace=True)\n",
+    "test_data.reset_index(drop=True, inplace=True)"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 16,
+   "id": "fbbf9b9e",
+   "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>Team</th>\n",
+       "      <th>Outcome</th>\n",
+       "      <th>Task Completed</th>\n",
+       "      <th>Murdered</th>\n",
+       "      <th>Game Length</th>\n",
+       "      <th>Ejected</th>\n",
+       "      <th>Sabotages Fixed</th>\n",
+       "      <th>Region</th>\n",
+       "      <th>User</th>\n",
+       "    </tr>\n",
+       "  </thead>\n",
+       "  <tbody>\n",
+       "    <tr>\n",
+       "      <th>0</th>\n",
+       "      <td>Crewmate</td>\n",
+       "      <td>1</td>\n",
+       "      <td>3.0</td>\n",
+       "      <td>1</td>\n",
+       "      <td>424.0</td>\n",
+       "      <td>0</td>\n",
+       "      <td>2.0</td>\n",
+       "      <td>0</td>\n",
+       "      <td>0</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>1</th>\n",
+       "      <td>Crewmate</td>\n",
+       "      <td>0</td>\n",
+       "      <td>7.0</td>\n",
+       "      <td>0</td>\n",
+       "      <td>981.0</td>\n",
+       "      <td>0</td>\n",
+       "      <td>1.0</td>\n",
+       "      <td>0</td>\n",
+       "      <td>0</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>2</th>\n",
+       "      <td>Crewmate</td>\n",
+       "      <td>1</td>\n",
+       "      <td>3.0</td>\n",
+       "      <td>0</td>\n",
+       "      <td>693.0</td>\n",
+       "      <td>0</td>\n",
+       "      <td>0.0</td>\n",
+       "      <td>0</td>\n",
+       "      <td>0</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>3</th>\n",
+       "      <td>Crewmate</td>\n",
+       "      <td>0</td>\n",
+       "      <td>4.0</td>\n",
+       "      <td>0</td>\n",
+       "      <td>310.0</td>\n",
+       "      <td>0</td>\n",
+       "      <td>0.0</td>\n",
+       "      <td>1</td>\n",
+       "      <td>0</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>4</th>\n",
+       "      <td>Crewmate</td>\n",
+       "      <td>0</td>\n",
+       "      <td>7.0</td>\n",
+       "      <td>1</td>\n",
+       "      <td>982.0</td>\n",
+       "      <td>0</td>\n",
+       "      <td>0.0</td>\n",
+       "      <td>1</td>\n",
+       "      <td>0</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>...</th>\n",
+       "      <td>...</td>\n",
+       "      <td>...</td>\n",
+       "      <td>...</td>\n",
+       "      <td>...</td>\n",
+       "      <td>...</td>\n",
+       "      <td>...</td>\n",
+       "      <td>...</td>\n",
+       "      <td>...</td>\n",
+       "      <td>...</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>1316</th>\n",
+       "      <td>Crewmate</td>\n",
+       "      <td>0</td>\n",
+       "      <td>5.0</td>\n",
+       "      <td>0</td>\n",
+       "      <td>660.0</td>\n",
+       "      <td>1</td>\n",
+       "      <td>2.0</td>\n",
+       "      <td>1</td>\n",
+       "      <td>28</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>1317</th>\n",
+       "      <td>Crewmate</td>\n",
+       "      <td>1</td>\n",
+       "      <td>7.0</td>\n",
+       "      <td>0</td>\n",
+       "      <td>851.0</td>\n",
+       "      <td>0</td>\n",
+       "      <td>1.0</td>\n",
+       "      <td>0</td>\n",
+       "      <td>28</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>1318</th>\n",
+       "      <td>Crewmate</td>\n",
+       "      <td>1</td>\n",
+       "      <td>7.0</td>\n",
+       "      <td>0</td>\n",
+       "      <td>1185.0</td>\n",
+       "      <td>0</td>\n",
+       "      <td>1.0</td>\n",
+       "      <td>1</td>\n",
+       "      <td>28</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>1319</th>\n",
+       "      <td>Crewmate</td>\n",
+       "      <td>1</td>\n",
+       "      <td>7.0</td>\n",
+       "      <td>0</td>\n",
+       "      <td>618.0</td>\n",
+       "      <td>0</td>\n",
+       "      <td>0.0</td>\n",
+       "      <td>1</td>\n",
+       "      <td>28</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>1320</th>\n",
+       "      <td>Crewmate</td>\n",
+       "      <td>1</td>\n",
+       "      <td>7.0</td>\n",
+       "      <td>0</td>\n",
+       "      <td>1177.0</td>\n",
+       "      <td>0</td>\n",
+       "      <td>1.0</td>\n",
+       "      <td>1</td>\n",
+       "      <td>28</td>\n",
+       "    </tr>\n",
+       "  </tbody>\n",
+       "</table>\n",
+       "<p>1321 rows × 9 columns</p>\n",
+       "</div>"
+      ],
+      "text/plain": [
+       "          Team  Outcome  Task Completed  Murdered  Game Length  Ejected  \\\n",
+       "0     Crewmate        1             3.0         1        424.0        0   \n",
+       "1     Crewmate        0             7.0         0        981.0        0   \n",
+       "2     Crewmate        1             3.0         0        693.0        0   \n",
+       "3     Crewmate        0             4.0         0        310.0        0   \n",
+       "4     Crewmate        0             7.0         1        982.0        0   \n",
+       "...        ...      ...             ...       ...          ...      ...   \n",
+       "1316  Crewmate        0             5.0         0        660.0        1   \n",
+       "1317  Crewmate        1             7.0         0        851.0        0   \n",
+       "1318  Crewmate        1             7.0         0       1185.0        0   \n",
+       "1319  Crewmate        1             7.0         0        618.0        0   \n",
+       "1320  Crewmate        1             7.0         0       1177.0        0   \n",
+       "\n",
+       "      Sabotages Fixed  Region  User  \n",
+       "0                 2.0       0     0  \n",
+       "1                 1.0       0     0  \n",
+       "2                 0.0       0     0  \n",
+       "3                 0.0       1     0  \n",
+       "4                 0.0       1     0  \n",
+       "...               ...     ...   ...  \n",
+       "1316              2.0       1    28  \n",
+       "1317              1.0       0    28  \n",
+       "1318              1.0       1    28  \n",
+       "1319              0.0       1    28  \n",
+       "1320              1.0       1    28  \n",
+       "\n",
+       "[1321 rows x 9 columns]"
+      ]
+     },
+     "execution_count": 16,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "train_data"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "id": "5a917985",
+   "metadata": {},
+   "source": [
+    "### Hypothesis Testing"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 95,
+   "id": "58a04df9",
+   "metadata": {},
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "0.05237999155491819\n",
+      "-1.6786449499637606\n"
+     ]
+    }
+   ],
+   "source": [
+    "from scipy import stats\n",
+    "\n",
+    "avg_task_completed = crewmate.groupby(by = 'User')['Task Completed'].mean()\n",
+    "avg_wins = crewmate.groupby(by = 'User')['Outcome'].mean()\n",
+    "\n",
+    "thresh = avg_task_completed.mean()\n",
+    "less_tasks = avg_wins[avg_task_completed < thresh]\n",
+    "greater_tasks = avg_wins[avg_task_completed >= thresh]\n",
+    "avg_task_completed = crewmate.groupby(by = 'User')['Task Completed'].mean()\n",
+    "avg_wins = crewmate.groupby(by = 'User')['Outcome'].mean()\n",
+    "thresh = avg_task_completed.mean()\n",
+    "less_tasks = avg_wins[avg_task_completed < thresh]\n",
+    "greater_tasks = avg_wins[avg_task_completed >= thresh]\n",
+    "\n",
+    "t1, p1 = stats.ttest_ind(less_tasks, greater_tasks, alternative = 'less')\n",
+    "print(p1)\n",
+    "print(t1)"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 94,
+   "id": "88550759",
+   "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": [
+    "from scipy.stats import t\n",
+    "df = 27\n",
+    "fig = plt.figure()\n",
+    "x = np.linspace(t.ppf(0.00001, df),\n",
+    "                t.ppf(0.99999, df), 100)\n",
+    "plt.plot(x, t.pdf(x, df),'r-', alpha=0.6, label='t pdf')\n",
+    "plt.axvline(t1, color = 'black', ls = '--')\n",
+    "plt.xlabel('t')\n",
+    "plt.ylabel('pdf')\n",
+    "plt.title(\"pdf of Student's t-test\")\n",
+    "plt.xlim(-5,5)\n",
+    "plt.ylim(0,.4)\n",
+    "\n",
+    "def f(x):\n",
+    "    return t.pdf(x, 27)\n",
+    "\n",
+    "section = np.linspace(-5,t1,100)\n",
+    "plt.fill_between(section,f(section))\n",
+    "plt.annotate('t=-1.678', (-3.2,.3))\n",
+    "plt.annotate('p=0.052', (-3.6,.05))\n",
+    "\n",
+    "plt.show()"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "id": "fd22b33a",
+   "metadata": {},
+   "source": [
+    "### Logistic Regression"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 18,
+   "id": "8f7c0308",
+   "metadata": {
+    "scrolled": false
+   },
+   "outputs": [
+    {
+     "data": {
+      "text/plain": [
+       "<BarContainer object of 6 artists>"
+      ]
+     },
+     "execution_count": 18,
+     "metadata": {},
+     "output_type": "execute_result"
+    },
+    {
+     "data": {
+      "image/png": "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\n",
+      "text/plain": [
+       "<Figure size 1080x576 with 1 Axes>"
+      ]
+     },
+     "metadata": {
+      "needs_background": "light"
+     },
+     "output_type": "display_data"
+    }
+   ],
+   "source": [
+    "from sklearn.linear_model import LogisticRegression\n",
+    "from sklearn.metrics import roc_auc_score\n",
+    "\n",
+    "#train data\n",
+    "X_train = train_data.drop([\"Outcome\", \"Team\", \"User\"], axis=1)\n",
+    "y_train = train_data[\"Outcome\"]\n",
+    "clf = LogisticRegression(random_state=0).fit(X_train, y_train)\n",
+    "coefs = clf.coef_[0]\n",
+    "\n",
+    "plt.figure(figsize=(15,8))\n",
+    "plt.bar(X_train.columns, coefs)"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 93,
+   "id": "6f153480",
+   "metadata": {},
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "AUC: 0.6136017272034544\n"
+     ]
+    }
+   ],
+   "source": [
+    "# test data\n",
+    "X_test = test_data.drop([\"Outcome\", \"Team\", \"User\"], axis=1)\n",
+    "y_true = test_data[\"Outcome\"]\n",
+    "y_predict = clf.predict(X_test)\n",
+    "\n",
+    "auc = roc_auc_score(y_true, y_predict)\n",
+    "print(\"AUC:\", auc)"
+   ]
+  }
+ ],
+ "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": "3.9.7"
+  }
+ },
+ "nbformat": 4,
+ "nbformat_minor": 5
+}
diff --git a/.ipynb_checkpoints/Hypothesis_Testing-checkpoint.ipynb b/.ipynb_checkpoints/Hypothesis_Testing-checkpoint.ipynb
new file mode 100644
index 0000000..363fcab
--- /dev/null
+++ b/.ipynb_checkpoints/Hypothesis_Testing-checkpoint.ipynb
@@ -0,0 +1,6 @@
+{
+ "cells": [],
+ "metadata": {},
+ "nbformat": 4,
+ "nbformat_minor": 5
+}
diff --git a/Among_Us_Prediction.ipynb b/Among_Us_Prediction.ipynb
new file mode 100644
index 0000000..4d2dd90
--- /dev/null
+++ b/Among_Us_Prediction.ipynb
@@ -0,0 +1,1349 @@
+{
+ "cells": [
+  {
+   "cell_type": "code",
+   "execution_count": 1,
+   "id": "02949a29",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "import numpy as np\n",
+    "import pandas as pd\n",
+    "import matplotlib.pyplot as plt\n",
+    "import seaborn as sns"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 2,
+   "id": "bcdba050",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "files = []\n",
+    "for i in range(29):\n",
+    "    tempdf = pd.read_csv(f'data/User{i+1}.csv', delimiter = ',')\n",
+    "    tempdf['User'] = i\n",
+    "    files.append(tempdf)\n",
+    "df = pd.concat(files, ignore_index = True)"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 3,
+   "id": "25f68e6f",
+   "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>Game Completed Date</th>\n",
+       "      <th>Team</th>\n",
+       "      <th>Outcome</th>\n",
+       "      <th>Task Completed</th>\n",
+       "      <th>All Tasks Completed</th>\n",
+       "      <th>Murdered</th>\n",
+       "      <th>Imposter Kills</th>\n",
+       "      <th>Game Length</th>\n",
+       "      <th>Ejected</th>\n",
+       "      <th>Sabotages Fixed</th>\n",
+       "      <th>Time to complete all tasks</th>\n",
+       "      <th>Rank Change</th>\n",
+       "      <th>Region/Game Code</th>\n",
+       "      <th>User</th>\n",
+       "    </tr>\n",
+       "  </thead>\n",
+       "  <tbody>\n",
+       "    <tr>\n",
+       "      <th>0</th>\n",
+       "      <td>12/13/2020 at 1:26:56 am EST</td>\n",
+       "      <td>Crewmate</td>\n",
+       "      <td>Win</td>\n",
+       "      <td>3</td>\n",
+       "      <td>No</td>\n",
+       "      <td>Yes</td>\n",
+       "      <td>-</td>\n",
+       "      <td>07m 04s</td>\n",
+       "      <td>No</td>\n",
+       "      <td>2.0</td>\n",
+       "      <td>-</td>\n",
+       "      <td>++</td>\n",
+       "      <td>NA / WYMSBF</td>\n",
+       "      <td>0</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>1</th>\n",
+       "      <td>12/13/2020 at 1:17:42 am EST</td>\n",
+       "      <td>Crewmate</td>\n",
+       "      <td>Loss</td>\n",
+       "      <td>7</td>\n",
+       "      <td>Yes</td>\n",
+       "      <td>No</td>\n",
+       "      <td>-</td>\n",
+       "      <td>16m 21s</td>\n",
+       "      <td>No</td>\n",
+       "      <td>1.0</td>\n",
+       "      <td>09m 48s</td>\n",
+       "      <td>--</td>\n",
+       "      <td>NA / WYMSBF</td>\n",
+       "      <td>0</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>2</th>\n",
+       "      <td>12/13/2020 at 12:57:47 am EST</td>\n",
+       "      <td>Crewmate</td>\n",
+       "      <td>Win</td>\n",
+       "      <td>3</td>\n",
+       "      <td>No</td>\n",
+       "      <td>No</td>\n",
+       "      <td>-</td>\n",
+       "      <td>11m 33s</td>\n",
+       "      <td>No</td>\n",
+       "      <td>0.0</td>\n",
+       "      <td>-</td>\n",
+       "      <td>++</td>\n",
+       "      <td>NA / WYMSBF</td>\n",
+       "      <td>0</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>3</th>\n",
+       "      <td>12/13/2020 at 12:41:55 am EST</td>\n",
+       "      <td>Imposter</td>\n",
+       "      <td>Win</td>\n",
+       "      <td>-</td>\n",
+       "      <td>-</td>\n",
+       "      <td>-</td>\n",
+       "      <td>2</td>\n",
+       "      <td>08m 05s</td>\n",
+       "      <td>No</td>\n",
+       "      <td>NaN</td>\n",
+       "      <td>-</td>\n",
+       "      <td>+++</td>\n",
+       "      <td>Europe / QIRTNF</td>\n",
+       "      <td>0</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>4</th>\n",
+       "      <td>12/13/2020 at 12:30:37 am EST</td>\n",
+       "      <td>Crewmate</td>\n",
+       "      <td>Loss</td>\n",
+       "      <td>4</td>\n",
+       "      <td>No</td>\n",
+       "      <td>No</td>\n",
+       "      <td>-</td>\n",
+       "      <td>05m 10s</td>\n",
+       "      <td>No</td>\n",
+       "      <td>0.0</td>\n",
+       "      <td>-</td>\n",
+       "      <td>---</td>\n",
+       "      <td>Europe / QIRTNF</td>\n",
+       "      <td>0</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>...</th>\n",
+       "      <td>...</td>\n",
+       "      <td>...</td>\n",
+       "      <td>...</td>\n",
+       "      <td>...</td>\n",
+       "      <td>...</td>\n",
+       "      <td>...</td>\n",
+       "      <td>...</td>\n",
+       "      <td>...</td>\n",
+       "      <td>...</td>\n",
+       "      <td>...</td>\n",
+       "      <td>...</td>\n",
+       "      <td>...</td>\n",
+       "      <td>...</td>\n",
+       "      <td>...</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>2222</th>\n",
+       "      <td>12/01/2020 at 11:07:41 am EST</td>\n",
+       "      <td>Imposter</td>\n",
+       "      <td>Loss</td>\n",
+       "      <td>-</td>\n",
+       "      <td>-</td>\n",
+       "      <td>-</td>\n",
+       "      <td>2</td>\n",
+       "      <td>14m 10s</td>\n",
+       "      <td>No</td>\n",
+       "      <td>NaN</td>\n",
+       "      <td>-</td>\n",
+       "      <td>-</td>\n",
+       "      <td>NA / SNNGZF</td>\n",
+       "      <td>28</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>2223</th>\n",
+       "      <td>12/01/2020 at 10:52:25 am EST</td>\n",
+       "      <td>Crewmate</td>\n",
+       "      <td>Win</td>\n",
+       "      <td>7</td>\n",
+       "      <td>Yes</td>\n",
+       "      <td>No</td>\n",
+       "      <td>-</td>\n",
+       "      <td>14m 11s</td>\n",
+       "      <td>No</td>\n",
+       "      <td>1.0</td>\n",
+       "      <td>14m 10s</td>\n",
+       "      <td>++</td>\n",
+       "      <td>NA / SNNGZF</td>\n",
+       "      <td>28</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>2224</th>\n",
+       "      <td>11/26/2020 at 11:23:14 am EST</td>\n",
+       "      <td>Crewmate</td>\n",
+       "      <td>Win</td>\n",
+       "      <td>7</td>\n",
+       "      <td>Yes</td>\n",
+       "      <td>No</td>\n",
+       "      <td>-</td>\n",
+       "      <td>19m 45s</td>\n",
+       "      <td>No</td>\n",
+       "      <td>1.0</td>\n",
+       "      <td>15m 16s</td>\n",
+       "      <td>++</td>\n",
+       "      <td>Europe / NZWLXQ</td>\n",
+       "      <td>28</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>2225</th>\n",
+       "      <td>11/26/2020 at 11:00:36 am EST</td>\n",
+       "      <td>Crewmate</td>\n",
+       "      <td>Win</td>\n",
+       "      <td>7</td>\n",
+       "      <td>Yes</td>\n",
+       "      <td>No</td>\n",
+       "      <td>-</td>\n",
+       "      <td>10m 18s</td>\n",
+       "      <td>No</td>\n",
+       "      <td>0.0</td>\n",
+       "      <td>06m 13s</td>\n",
+       "      <td>++</td>\n",
+       "      <td>Europe / NZWLXQ</td>\n",
+       "      <td>28</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>2226</th>\n",
+       "      <td>11/26/2020 at 10:48:50 am EST</td>\n",
+       "      <td>Crewmate</td>\n",
+       "      <td>Win</td>\n",
+       "      <td>7</td>\n",
+       "      <td>Yes</td>\n",
+       "      <td>No</td>\n",
+       "      <td>-</td>\n",
+       "      <td>19m 37s</td>\n",
+       "      <td>No</td>\n",
+       "      <td>1.0</td>\n",
+       "      <td>16m 02s</td>\n",
+       "      <td>++</td>\n",
+       "      <td>Europe / NZWLXQ</td>\n",
+       "      <td>28</td>\n",
+       "    </tr>\n",
+       "  </tbody>\n",
+       "</table>\n",
+       "<p>2227 rows × 14 columns</p>\n",
+       "</div>"
+      ],
+      "text/plain": [
+       "                Game Completed Date      Team Outcome Task Completed  \\\n",
+       "0      12/13/2020 at 1:26:56 am EST  Crewmate     Win              3   \n",
+       "1      12/13/2020 at 1:17:42 am EST  Crewmate    Loss              7   \n",
+       "2     12/13/2020 at 12:57:47 am EST  Crewmate     Win              3   \n",
+       "3     12/13/2020 at 12:41:55 am EST  Imposter     Win              -   \n",
+       "4     12/13/2020 at 12:30:37 am EST  Crewmate    Loss              4   \n",
+       "...                             ...       ...     ...            ...   \n",
+       "2222  12/01/2020 at 11:07:41 am EST  Imposter    Loss              -   \n",
+       "2223  12/01/2020 at 10:52:25 am EST  Crewmate     Win              7   \n",
+       "2224  11/26/2020 at 11:23:14 am EST  Crewmate     Win              7   \n",
+       "2225  11/26/2020 at 11:00:36 am EST  Crewmate     Win              7   \n",
+       "2226  11/26/2020 at 10:48:50 am EST  Crewmate     Win              7   \n",
+       "\n",
+       "     All Tasks Completed Murdered Imposter Kills Game Length Ejected  \\\n",
+       "0                     No      Yes              -     07m 04s      No   \n",
+       "1                    Yes       No              -     16m 21s      No   \n",
+       "2                     No       No              -     11m 33s      No   \n",
+       "3                      -        -              2     08m 05s      No   \n",
+       "4                     No       No              -     05m 10s      No   \n",
+       "...                  ...      ...            ...         ...     ...   \n",
+       "2222                   -        -              2     14m 10s      No   \n",
+       "2223                 Yes       No              -     14m 11s      No   \n",
+       "2224                 Yes       No              -     19m 45s      No   \n",
+       "2225                 Yes       No              -     10m 18s      No   \n",
+       "2226                 Yes       No              -     19m 37s      No   \n",
+       "\n",
+       "      Sabotages Fixed Time to complete all tasks Rank Change Region/Game Code  \\\n",
+       "0                 2.0                          -          ++      NA / WYMSBF   \n",
+       "1                 1.0                    09m 48s          --      NA / WYMSBF   \n",
+       "2                 0.0                          -          ++      NA / WYMSBF   \n",
+       "3                 NaN                          -         +++  Europe / QIRTNF   \n",
+       "4                 0.0                          -         ---  Europe / QIRTNF   \n",
+       "...               ...                        ...         ...              ...   \n",
+       "2222              NaN                          -           -      NA / SNNGZF   \n",
+       "2223              1.0                    14m 10s          ++      NA / SNNGZF   \n",
+       "2224              1.0                    15m 16s          ++  Europe / NZWLXQ   \n",
+       "2225              0.0                    06m 13s          ++  Europe / NZWLXQ   \n",
+       "2226              1.0                    16m 02s          ++  Europe / NZWLXQ   \n",
+       "\n",
+       "      User  \n",
+       "0        0  \n",
+       "1        0  \n",
+       "2        0  \n",
+       "3        0  \n",
+       "4        0  \n",
+       "...    ...  \n",
+       "2222    28  \n",
+       "2223    28  \n",
+       "2224    28  \n",
+       "2225    28  \n",
+       "2226    28  \n",
+       "\n",
+       "[2227 rows x 14 columns]"
+      ]
+     },
+     "execution_count": 3,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "df"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 4,
+   "id": "e6f11cc9",
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "text/plain": [
+       "Game Completed Date            object\n",
+       "Team                           object\n",
+       "Outcome                        object\n",
+       "Task Completed                 object\n",
+       "All Tasks Completed            object\n",
+       "Murdered                       object\n",
+       "Imposter Kills                 object\n",
+       "Game Length                    object\n",
+       "Ejected                        object\n",
+       "Sabotages Fixed               float64\n",
+       "Time to complete all tasks     object\n",
+       "Rank Change                    object\n",
+       "Region/Game Code               object\n",
+       "User                            int64\n",
+       "dtype: object"
+      ]
+     },
+     "execution_count": 4,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "df.dtypes"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 5,
+   "id": "888d9bdd",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "df[['Task Completed', 'Imposter Kills']] = df[['Task Completed', 'Imposter Kills']].apply(pd.to_numeric, errors = 'coerce')"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 6,
+   "id": "127740d3",
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "text/plain": [
+       "NA         1436\n",
+       "Europe      791\n",
+       "Name: Region, dtype: int64"
+      ]
+     },
+     "execution_count": 6,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "df['Region'] = df['Region/Game Code'].str.extract(r'^(.*?)/')\n",
+    "df['Region'].value_counts()\n"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 7,
+   "id": "8cb080c6",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "df['Game Length'] = df['Game Length'].apply(pd.to_timedelta)\n",
+    "df['Game Length'] = df['Game Length'] / np.timedelta64(1, 's')"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 8,
+   "id": "44941ec1",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "df['Time to complete all tasks'] = df['Time to complete all tasks'].apply(pd.to_timedelta, errors = 'coerce')\n",
+    "df['Time to complete all tasks'] = df['Time to complete all tasks'] / np.timedelta64(1, 's')"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 9,
+   "id": "e84a07c1",
+   "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>Team</th>\n",
+       "      <th>Outcome</th>\n",
+       "      <th>Task Completed</th>\n",
+       "      <th>All Tasks Completed</th>\n",
+       "      <th>Murdered</th>\n",
+       "      <th>Game Length</th>\n",
+       "      <th>Ejected</th>\n",
+       "      <th>Sabotages Fixed</th>\n",
+       "      <th>Region</th>\n",
+       "      <th>User</th>\n",
+       "    </tr>\n",
+       "  </thead>\n",
+       "  <tbody>\n",
+       "    <tr>\n",
+       "      <th>0</th>\n",
+       "      <td>Crewmate</td>\n",
+       "      <td>Win</td>\n",
+       "      <td>3.0</td>\n",
+       "      <td>No</td>\n",
+       "      <td>Yes</td>\n",
+       "      <td>424.0</td>\n",
+       "      <td>No</td>\n",
+       "      <td>2.0</td>\n",
+       "      <td>NA</td>\n",
+       "      <td>0</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>1</th>\n",
+       "      <td>Crewmate</td>\n",
+       "      <td>Loss</td>\n",
+       "      <td>7.0</td>\n",
+       "      <td>Yes</td>\n",
+       "      <td>No</td>\n",
+       "      <td>981.0</td>\n",
+       "      <td>No</td>\n",
+       "      <td>1.0</td>\n",
+       "      <td>NA</td>\n",
+       "      <td>0</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>2</th>\n",
+       "      <td>Crewmate</td>\n",
+       "      <td>Win</td>\n",
+       "      <td>3.0</td>\n",
+       "      <td>No</td>\n",
+       "      <td>No</td>\n",
+       "      <td>693.0</td>\n",
+       "      <td>No</td>\n",
+       "      <td>0.0</td>\n",
+       "      <td>NA</td>\n",
+       "      <td>0</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>4</th>\n",
+       "      <td>Crewmate</td>\n",
+       "      <td>Loss</td>\n",
+       "      <td>4.0</td>\n",
+       "      <td>No</td>\n",
+       "      <td>No</td>\n",
+       "      <td>310.0</td>\n",
+       "      <td>No</td>\n",
+       "      <td>0.0</td>\n",
+       "      <td>Europe</td>\n",
+       "      <td>0</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>5</th>\n",
+       "      <td>Crewmate</td>\n",
+       "      <td>Loss</td>\n",
+       "      <td>7.0</td>\n",
+       "      <td>Yes</td>\n",
+       "      <td>Yes</td>\n",
+       "      <td>982.0</td>\n",
+       "      <td>No</td>\n",
+       "      <td>0.0</td>\n",
+       "      <td>Europe</td>\n",
+       "      <td>0</td>\n",
+       "    </tr>\n",
+       "  </tbody>\n",
+       "</table>\n",
+       "</div>"
+      ],
+      "text/plain": [
+       "       Team Outcome  Task Completed All Tasks Completed Murdered  Game Length  \\\n",
+       "0  Crewmate     Win             3.0                  No      Yes        424.0   \n",
+       "1  Crewmate    Loss             7.0                 Yes       No        981.0   \n",
+       "2  Crewmate     Win             3.0                  No       No        693.0   \n",
+       "4  Crewmate    Loss             4.0                  No       No        310.0   \n",
+       "5  Crewmate    Loss             7.0                 Yes      Yes        982.0   \n",
+       "\n",
+       "  Ejected  Sabotages Fixed   Region  User  \n",
+       "0      No              2.0      NA      0  \n",
+       "1      No              1.0      NA      0  \n",
+       "2      No              0.0      NA      0  \n",
+       "4      No              0.0  Europe      0  \n",
+       "5      No              0.0  Europe      0  "
+      ]
+     },
+     "execution_count": 9,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "crewmate = df[['Team',\n",
+    "               'Outcome',\n",
+    "               'Task Completed',\n",
+    "               'All Tasks Completed',\n",
+    "               'Murdered',\n",
+    "               'Game Length',\n",
+    "               'Ejected',\n",
+    "              'Sabotages Fixed',\n",
+    "              'Region',\n",
+    "              'User']][df['Team'] == 'Crewmate']\n",
+    "crewmate.head()"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 10,
+   "id": "13385be6",
+   "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>Team</th>\n",
+       "      <th>Outcome</th>\n",
+       "      <th>Task Completed</th>\n",
+       "      <th>All Tasks Completed</th>\n",
+       "      <th>Murdered</th>\n",
+       "      <th>Game Length</th>\n",
+       "      <th>Ejected</th>\n",
+       "      <th>Sabotages Fixed</th>\n",
+       "      <th>Region</th>\n",
+       "      <th>User</th>\n",
+       "    </tr>\n",
+       "  </thead>\n",
+       "  <tbody>\n",
+       "    <tr>\n",
+       "      <th>0</th>\n",
+       "      <td>Crewmate</td>\n",
+       "      <td>1</td>\n",
+       "      <td>3.0</td>\n",
+       "      <td>0</td>\n",
+       "      <td>1</td>\n",
+       "      <td>424.0</td>\n",
+       "      <td>0</td>\n",
+       "      <td>2.0</td>\n",
+       "      <td>0</td>\n",
+       "      <td>0</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>1</th>\n",
+       "      <td>Crewmate</td>\n",
+       "      <td>0</td>\n",
+       "      <td>7.0</td>\n",
+       "      <td>1</td>\n",
+       "      <td>0</td>\n",
+       "      <td>981.0</td>\n",
+       "      <td>0</td>\n",
+       "      <td>1.0</td>\n",
+       "      <td>0</td>\n",
+       "      <td>0</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>2</th>\n",
+       "      <td>Crewmate</td>\n",
+       "      <td>1</td>\n",
+       "      <td>3.0</td>\n",
+       "      <td>0</td>\n",
+       "      <td>0</td>\n",
+       "      <td>693.0</td>\n",
+       "      <td>0</td>\n",
+       "      <td>0.0</td>\n",
+       "      <td>0</td>\n",
+       "      <td>0</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>4</th>\n",
+       "      <td>Crewmate</td>\n",
+       "      <td>0</td>\n",
+       "      <td>4.0</td>\n",
+       "      <td>0</td>\n",
+       "      <td>0</td>\n",
+       "      <td>310.0</td>\n",
+       "      <td>0</td>\n",
+       "      <td>0.0</td>\n",
+       "      <td>1</td>\n",
+       "      <td>0</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>5</th>\n",
+       "      <td>Crewmate</td>\n",
+       "      <td>0</td>\n",
+       "      <td>7.0</td>\n",
+       "      <td>1</td>\n",
+       "      <td>1</td>\n",
+       "      <td>982.0</td>\n",
+       "      <td>0</td>\n",
+       "      <td>0.0</td>\n",
+       "      <td>1</td>\n",
+       "      <td>0</td>\n",
+       "    </tr>\n",
+       "  </tbody>\n",
+       "</table>\n",
+       "</div>"
+      ],
+      "text/plain": [
+       "       Team  Outcome  Task Completed  All Tasks Completed  Murdered  \\\n",
+       "0  Crewmate        1             3.0                    0         1   \n",
+       "1  Crewmate        0             7.0                    1         0   \n",
+       "2  Crewmate        1             3.0                    0         0   \n",
+       "4  Crewmate        0             4.0                    0         0   \n",
+       "5  Crewmate        0             7.0                    1         1   \n",
+       "\n",
+       "   Game Length  Ejected  Sabotages Fixed  Region  User  \n",
+       "0        424.0        0              2.0       0     0  \n",
+       "1        981.0        0              1.0       0     0  \n",
+       "2        693.0        0              0.0       0     0  \n",
+       "4        310.0        0              0.0       1     0  \n",
+       "5        982.0        0              0.0       1     0  "
+      ]
+     },
+     "execution_count": 10,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "crewmate['Outcome'] = crewmate['Outcome'].replace(['Loss', 'Win'],[0, 1])\n",
+    "crewmate['All Tasks Completed'] = crewmate['All Tasks Completed'].replace(['No', 'Yes'],[0, 1])\n",
+    "crewmate['Murdered'] = crewmate['Murdered'].replace(['No', 'Yes'],[0, 1])\n",
+    "crewmate['Ejected'] = crewmate['Ejected'].replace(['No', 'Yes'],[0, 1])\n",
+    "crewmate['Region'] = crewmate['Region'].replace(['NA ', 'Europe '],[0, 1])\n",
+    "crewmate.head()"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 11,
+   "id": "342da2cd",
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "image/png": "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\n",
+      "text/plain": [
+       "<Figure size 1440x1080 with 2 Axes>"
+      ]
+     },
+     "metadata": {
+      "needs_background": "light"
+     },
+     "output_type": "display_data"
+    }
+   ],
+   "source": [
+    "%matplotlib inline\n",
+    "corrlation_matrix = crewmate.corr()\n",
+    "fig, ax = plt.subplots(figsize = (20, 15))\n",
+    "ax = sns.heatmap(corrlation_matrix,\n",
+    "                annot = True,\n",
+    "                linewidths = 0.5,\n",
+    "                fmt = \".2f\")"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 12,
+   "id": "bdf14fd8",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "crewmate = crewmate.drop(['All Tasks Completed'], axis =1)"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 13,
+   "id": "0ca1ad6d",
+   "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>Team</th>\n",
+       "      <th>Outcome</th>\n",
+       "      <th>Task Completed</th>\n",
+       "      <th>Murdered</th>\n",
+       "      <th>Game Length</th>\n",
+       "      <th>Ejected</th>\n",
+       "      <th>Sabotages Fixed</th>\n",
+       "      <th>Region</th>\n",
+       "      <th>User</th>\n",
+       "    </tr>\n",
+       "  </thead>\n",
+       "  <tbody>\n",
+       "    <tr>\n",
+       "      <th>0</th>\n",
+       "      <td>Crewmate</td>\n",
+       "      <td>1</td>\n",
+       "      <td>3.0</td>\n",
+       "      <td>1</td>\n",
+       "      <td>424.0</td>\n",
+       "      <td>0</td>\n",
+       "      <td>2.0</td>\n",
+       "      <td>0</td>\n",
+       "      <td>0</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>1</th>\n",
+       "      <td>Crewmate</td>\n",
+       "      <td>0</td>\n",
+       "      <td>7.0</td>\n",
+       "      <td>0</td>\n",
+       "      <td>981.0</td>\n",
+       "      <td>0</td>\n",
+       "      <td>1.0</td>\n",
+       "      <td>0</td>\n",
+       "      <td>0</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>2</th>\n",
+       "      <td>Crewmate</td>\n",
+       "      <td>1</td>\n",
+       "      <td>3.0</td>\n",
+       "      <td>0</td>\n",
+       "      <td>693.0</td>\n",
+       "      <td>0</td>\n",
+       "      <td>0.0</td>\n",
+       "      <td>0</td>\n",
+       "      <td>0</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>4</th>\n",
+       "      <td>Crewmate</td>\n",
+       "      <td>0</td>\n",
+       "      <td>4.0</td>\n",
+       "      <td>0</td>\n",
+       "      <td>310.0</td>\n",
+       "      <td>0</td>\n",
+       "      <td>0.0</td>\n",
+       "      <td>1</td>\n",
+       "      <td>0</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>5</th>\n",
+       "      <td>Crewmate</td>\n",
+       "      <td>0</td>\n",
+       "      <td>7.0</td>\n",
+       "      <td>1</td>\n",
+       "      <td>982.0</td>\n",
+       "      <td>0</td>\n",
+       "      <td>0.0</td>\n",
+       "      <td>1</td>\n",
+       "      <td>0</td>\n",
+       "    </tr>\n",
+       "  </tbody>\n",
+       "</table>\n",
+       "</div>"
+      ],
+      "text/plain": [
+       "       Team  Outcome  Task Completed  Murdered  Game Length  Ejected  \\\n",
+       "0  Crewmate        1             3.0         1        424.0        0   \n",
+       "1  Crewmate        0             7.0         0        981.0        0   \n",
+       "2  Crewmate        1             3.0         0        693.0        0   \n",
+       "4  Crewmate        0             4.0         0        310.0        0   \n",
+       "5  Crewmate        0             7.0         1        982.0        0   \n",
+       "\n",
+       "   Sabotages Fixed  Region  User  \n",
+       "0              2.0       0     0  \n",
+       "1              1.0       0     0  \n",
+       "2              0.0       0     0  \n",
+       "4              0.0       1     0  \n",
+       "5              0.0       1     0  "
+      ]
+     },
+     "execution_count": 13,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "crewmate.head()"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 14,
+   "id": "858d7136",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "def user_train_valid_split(users_i,valid_ratio,possible=False):\n",
+    "    n = sum(users_i)\n",
+    "    target = int(np.floor(valid_ratio*n))\n",
+    "    while not possible:\n",
+    "        possible, validation_indices = user_split(users_i,target)\n",
+    "        target -= 1\n",
+    "    training_indices = [i for i in range(len(users_i)) if i not in validation_indices]\n",
+    "    return training_indices, validation_indices\n",
+    "\n",
+    "def user_split(users_i,target):  \n",
+    "    # Reference: https://levelup.gitconnected.com/dynamic-programming-subset-sum-c386126621cd\n",
+    "    n = len(users_i)\n",
+    "    solution = [[False for j in range(int(target+1))] for i in range(n+1)]\n",
+    "    # base cases\n",
+    "    for i in range(n):\n",
+    "        solution[i][0] = True\n",
+    "    # other cases\n",
+    "    for i in range(1,n+1):\n",
+    "        for j in range(1,target+1):\n",
+    "            solution[i][j] = solution[i-1][j]\n",
+    "            if(solution[i][j] == False and j >= users_i[i-1]):\n",
+    "                solution[i][j] = solution[i][j] or solution[i-1][j-users_i[i-1]]\n",
+    "    # check if the subset sum is possible\n",
+    "    possible = solution[len(users_i)][target]\n",
+    "    subset = []\n",
+    "    if not possible: return possible, subset\n",
+    "    # return the subset solution if one exists\n",
+    "    y = len(users_i)\n",
+    "    x = target\n",
+    "    while x != 0:\n",
+    "        if solution[y-1][x] == False:\n",
+    "            subset.append(y-1)\n",
+    "            x -= users_i[y-1]\n",
+    "        else: \n",
+    "            y -= 1\n",
+    "    return possible, subset"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 15,
+   "id": "b5a1d348",
+   "metadata": {},
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "expected valid ratio: 0.25, actual: 0.2498580352072686\n"
+     ]
+    }
+   ],
+   "source": [
+    "crewmates_i = list(crewmate[\"User\"].value_counts(sort = False))\n",
+    "test_ratio = .25\n",
+    "training_indices, test_indices = user_train_valid_split(crewmates_i,test_ratio)\n",
+    "\n",
+    "#check\n",
+    "train_sum = sum([crewmates_i[i] for i in training_indices])\n",
+    "test_sum = sum([crewmates_i[i] for i in test_indices])\n",
+    "print(\"expected valid ratio: {}, actual: {}\".format(test_ratio, test_sum/(train_sum+test_sum)))\n",
+    "\n",
+    "train_data = crewmate.loc[crewmate['User'].isin(training_indices)]\n",
+    "test_data = crewmate.loc[crewmate['User'].isin(test_indices)]\n",
+    "train_data.reset_index(drop=True, inplace=True)\n",
+    "test_data.reset_index(drop=True, inplace=True)"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 16,
+   "id": "fbbf9b9e",
+   "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>Team</th>\n",
+       "      <th>Outcome</th>\n",
+       "      <th>Task Completed</th>\n",
+       "      <th>Murdered</th>\n",
+       "      <th>Game Length</th>\n",
+       "      <th>Ejected</th>\n",
+       "      <th>Sabotages Fixed</th>\n",
+       "      <th>Region</th>\n",
+       "      <th>User</th>\n",
+       "    </tr>\n",
+       "  </thead>\n",
+       "  <tbody>\n",
+       "    <tr>\n",
+       "      <th>0</th>\n",
+       "      <td>Crewmate</td>\n",
+       "      <td>1</td>\n",
+       "      <td>3.0</td>\n",
+       "      <td>1</td>\n",
+       "      <td>424.0</td>\n",
+       "      <td>0</td>\n",
+       "      <td>2.0</td>\n",
+       "      <td>0</td>\n",
+       "      <td>0</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>1</th>\n",
+       "      <td>Crewmate</td>\n",
+       "      <td>0</td>\n",
+       "      <td>7.0</td>\n",
+       "      <td>0</td>\n",
+       "      <td>981.0</td>\n",
+       "      <td>0</td>\n",
+       "      <td>1.0</td>\n",
+       "      <td>0</td>\n",
+       "      <td>0</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>2</th>\n",
+       "      <td>Crewmate</td>\n",
+       "      <td>1</td>\n",
+       "      <td>3.0</td>\n",
+       "      <td>0</td>\n",
+       "      <td>693.0</td>\n",
+       "      <td>0</td>\n",
+       "      <td>0.0</td>\n",
+       "      <td>0</td>\n",
+       "      <td>0</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>3</th>\n",
+       "      <td>Crewmate</td>\n",
+       "      <td>0</td>\n",
+       "      <td>4.0</td>\n",
+       "      <td>0</td>\n",
+       "      <td>310.0</td>\n",
+       "      <td>0</td>\n",
+       "      <td>0.0</td>\n",
+       "      <td>1</td>\n",
+       "      <td>0</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>4</th>\n",
+       "      <td>Crewmate</td>\n",
+       "      <td>0</td>\n",
+       "      <td>7.0</td>\n",
+       "      <td>1</td>\n",
+       "      <td>982.0</td>\n",
+       "      <td>0</td>\n",
+       "      <td>0.0</td>\n",
+       "      <td>1</td>\n",
+       "      <td>0</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>...</th>\n",
+       "      <td>...</td>\n",
+       "      <td>...</td>\n",
+       "      <td>...</td>\n",
+       "      <td>...</td>\n",
+       "      <td>...</td>\n",
+       "      <td>...</td>\n",
+       "      <td>...</td>\n",
+       "      <td>...</td>\n",
+       "      <td>...</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>1316</th>\n",
+       "      <td>Crewmate</td>\n",
+       "      <td>0</td>\n",
+       "      <td>5.0</td>\n",
+       "      <td>0</td>\n",
+       "      <td>660.0</td>\n",
+       "      <td>1</td>\n",
+       "      <td>2.0</td>\n",
+       "      <td>1</td>\n",
+       "      <td>28</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>1317</th>\n",
+       "      <td>Crewmate</td>\n",
+       "      <td>1</td>\n",
+       "      <td>7.0</td>\n",
+       "      <td>0</td>\n",
+       "      <td>851.0</td>\n",
+       "      <td>0</td>\n",
+       "      <td>1.0</td>\n",
+       "      <td>0</td>\n",
+       "      <td>28</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>1318</th>\n",
+       "      <td>Crewmate</td>\n",
+       "      <td>1</td>\n",
+       "      <td>7.0</td>\n",
+       "      <td>0</td>\n",
+       "      <td>1185.0</td>\n",
+       "      <td>0</td>\n",
+       "      <td>1.0</td>\n",
+       "      <td>1</td>\n",
+       "      <td>28</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>1319</th>\n",
+       "      <td>Crewmate</td>\n",
+       "      <td>1</td>\n",
+       "      <td>7.0</td>\n",
+       "      <td>0</td>\n",
+       "      <td>618.0</td>\n",
+       "      <td>0</td>\n",
+       "      <td>0.0</td>\n",
+       "      <td>1</td>\n",
+       "      <td>28</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>1320</th>\n",
+       "      <td>Crewmate</td>\n",
+       "      <td>1</td>\n",
+       "      <td>7.0</td>\n",
+       "      <td>0</td>\n",
+       "      <td>1177.0</td>\n",
+       "      <td>0</td>\n",
+       "      <td>1.0</td>\n",
+       "      <td>1</td>\n",
+       "      <td>28</td>\n",
+       "    </tr>\n",
+       "  </tbody>\n",
+       "</table>\n",
+       "<p>1321 rows × 9 columns</p>\n",
+       "</div>"
+      ],
+      "text/plain": [
+       "          Team  Outcome  Task Completed  Murdered  Game Length  Ejected  \\\n",
+       "0     Crewmate        1             3.0         1        424.0        0   \n",
+       "1     Crewmate        0             7.0         0        981.0        0   \n",
+       "2     Crewmate        1             3.0         0        693.0        0   \n",
+       "3     Crewmate        0             4.0         0        310.0        0   \n",
+       "4     Crewmate        0             7.0         1        982.0        0   \n",
+       "...        ...      ...             ...       ...          ...      ...   \n",
+       "1316  Crewmate        0             5.0         0        660.0        1   \n",
+       "1317  Crewmate        1             7.0         0        851.0        0   \n",
+       "1318  Crewmate        1             7.0         0       1185.0        0   \n",
+       "1319  Crewmate        1             7.0         0        618.0        0   \n",
+       "1320  Crewmate        1             7.0         0       1177.0        0   \n",
+       "\n",
+       "      Sabotages Fixed  Region  User  \n",
+       "0                 2.0       0     0  \n",
+       "1                 1.0       0     0  \n",
+       "2                 0.0       0     0  \n",
+       "3                 0.0       1     0  \n",
+       "4                 0.0       1     0  \n",
+       "...               ...     ...   ...  \n",
+       "1316              2.0       1    28  \n",
+       "1317              1.0       0    28  \n",
+       "1318              1.0       1    28  \n",
+       "1319              0.0       1    28  \n",
+       "1320              1.0       1    28  \n",
+       "\n",
+       "[1321 rows x 9 columns]"
+      ]
+     },
+     "execution_count": 16,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "train_data"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "id": "5a917985",
+   "metadata": {},
+   "source": [
+    "### Hypothesis Testing"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 95,
+   "id": "58a04df9",
+   "metadata": {},
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "0.05237999155491819\n",
+      "-1.6786449499637606\n"
+     ]
+    }
+   ],
+   "source": [
+    "from scipy import stats\n",
+    "\n",
+    "avg_task_completed = crewmate.groupby(by = 'User')['Task Completed'].mean()\n",
+    "avg_wins = crewmate.groupby(by = 'User')['Outcome'].mean()\n",
+    "\n",
+    "thresh = avg_task_completed.mean()\n",
+    "less_tasks = avg_wins[avg_task_completed < thresh]\n",
+    "greater_tasks = avg_wins[avg_task_completed >= thresh]\n",
+    "avg_task_completed = crewmate.groupby(by = 'User')['Task Completed'].mean()\n",
+    "avg_wins = crewmate.groupby(by = 'User')['Outcome'].mean()\n",
+    "thresh = avg_task_completed.mean()\n",
+    "less_tasks = avg_wins[avg_task_completed < thresh]\n",
+    "greater_tasks = avg_wins[avg_task_completed >= thresh]\n",
+    "\n",
+    "t1, p1 = stats.ttest_ind(less_tasks, greater_tasks, alternative = 'less')\n",
+    "print(p1)\n",
+    "print(t1)"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 94,
+   "id": "88550759",
+   "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": [
+    "from scipy.stats import t\n",
+    "df = 27\n",
+    "fig = plt.figure()\n",
+    "x = np.linspace(t.ppf(0.00001, df),\n",
+    "                t.ppf(0.99999, df), 100)\n",
+    "plt.plot(x, t.pdf(x, df),'r-', alpha=0.6, label='t pdf')\n",
+    "plt.axvline(t1, color = 'black', ls = '--')\n",
+    "plt.xlabel('t')\n",
+    "plt.ylabel('pdf')\n",
+    "plt.title(\"pdf of Student's t-test\")\n",
+    "plt.xlim(-5,5)\n",
+    "plt.ylim(0,.4)\n",
+    "\n",
+    "def f(x):\n",
+    "    return t.pdf(x, 27)\n",
+    "\n",
+    "section = np.linspace(-5,t1,100)\n",
+    "plt.fill_between(section,f(section))\n",
+    "plt.annotate('t=-1.678', (-3.2,.3))\n",
+    "plt.annotate('p=0.052', (-3.6,.05))\n",
+    "\n",
+    "plt.show()"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "id": "fd22b33a",
+   "metadata": {},
+   "source": [
+    "### Logistic Regression"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 18,
+   "id": "8f7c0308",
+   "metadata": {
+    "scrolled": false
+   },
+   "outputs": [
+    {
+     "data": {
+      "text/plain": [
+       "<BarContainer object of 6 artists>"
+      ]
+     },
+     "execution_count": 18,
+     "metadata": {},
+     "output_type": "execute_result"
+    },
+    {
+     "data": {
+      "image/png": "iVBORw0KGgoAAAANSUhEUgAAA3IAAAHSCAYAAACpVPkmAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjQuMywgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy/MnkTPAAAACXBIWXMAAAsTAAALEwEAmpwYAAAiWElEQVR4nO3de7xudV0n8M8XDohKiAghosdjihWZUW5NvKKio1KCJqGZQpOdcTSzi6+GhhmHbjOUUzap6ZzIAbXSvCCoFALmYOQFUO6IeCNRvE5aaOUL+M0fa23Pw+bZe59zns3Z/M7zfr9ez2uvy2+t32/vtZ61ns/6rfXsaq0FAACAfuy23g0AAABg+whyAAAAnRHkAAAAOiPIAQAAdEaQAwAA6IwgBwAA0JkN692Aley///5t06ZN690MAACAdXHJJZd8rbV2wNLpd+ogt2nTplx88cXr3QwAAIB1UVXXT5vu1koAAIDOCHIAAACdEeQAAAA6I8gBAAB0RpADAADojCAHAADQGUEOAACgM4IcAABAZwQ5AACAzghyAAAAnRHkAAAAOiPIAQAAdEaQAwAA6IwgBwAA0Jk1CXJV9dSquraqPlVVJ06Zf5eqeus4/yNVtWkt6gUAAJhHMwe5qto9yWuTPC3JoUmeW1WHLin280n+sbX2oCSvSvJ7s9YLAAAwr9aiR+4RST7VWvtMa+07Sd6S5OglZY5Ocvo4/PYkT6qqWoO6AQAA5s5aBLmDk3x+YvyGcdrUMq21m5N8M8m91qBuAACAubNhvRuwVFVtTrI5STZu3LjOrYGVbTrxvevdBFbxuVOOWu8mMGccF+78dtZxwb5w57czzxH2hzu/3j4zrEWP3BeS3G9i/L7jtKllqmpDknsk+fq0lbXWtrTWFlprCwcccMAaNA8AAGDXshZB7qIkh1TVA6pqzyTPSXLWkjJnJTl+HH52kve31toa1A0AADB3Zr61srV2c1X9YpJzkuye5A2ttauq6reSXNxaOyvJnyV5U1V9Ksn/yxD2AAAA2AFr8oxca+3sJGcvmfaKieF/TXLsWtQFAAAw79bkH4IDAACw89zpvrUSAAB2Nb19IyJ3fnrkAAAAOiPIAQAAdEaQAwAA6IwgBwAA0BlBDgAAoDOCHAAAQGcEOQAAgM4IcgAAAJ0R5AAAADojyAEAAHRGkAMAAOiMIAcAANAZQQ4AAKAzghwAAEBnBDkAAIDOCHIAAACdEeQAAAA6I8gBAAB0RpADAADojCAHAADQGUEOAACgM4IcAABAZwQ5AACAzghyAAAAnRHkAAAAOiPIAQAAdEaQAwAA6IwgBwAA0BlBDgAAoDOCHAAAQGcEOQAAgM4IcgAAAJ0R5AAAADojyAEAAHRGkAMAAOiMIAcAANCZDevdAACAXdHnTjlqvZsA7ML0yAEAAHRGkAMAAOiMIAcAANAZQQ4AAKAzghwAAEBnBDkAAIDOCHIAAACdEeQAAAA6I8gBAAB0RpADAADozExBrqr2q6pzq+q68ec9p5Q5rKo+VFVXVdXlVXXcLHUCAADMu1l75E5Mcn5r7ZAk54/jS307yQtaaz+U5KlJ/qiq9p2xXgAAgLk1a5A7Osnp4/DpSY5ZWqC19snW2nXj8BeTfCXJATPWCwAAMLdmDXIHttZuHIe/lOTAlQpX1SOS7Jnk0zPWCwAAMLc2rFagqs5Lcu8ps06aHGmttapqK6znoCRvSnJ8a+3WFcptTrI5STZu3Lha8wAAAObOqkGutXbkcvOq6stVdVBr7cYxqH1lmXL7JHlvkpNaax9epb4tSbYkycLCwrLBEAAAYF7NemvlWUmOH4ePT3Lm0gJVtWeSM5K8sbX29hnrAwAAmHuzBrlTkjy5qq5LcuQ4nqpaqKpTxzI/neRxSU6oqkvH12Ez1gsAADC3Vr21ciWtta8nedKU6RcneeE4/OYkb56lHgAAALaatUcOAACAnUyQAwAA6IwgBwAA0BlBDgAAoDOCHAAAQGcEOQAAgM4IcgAAAJ0R5AAAADojyAEAAHRGkAMAAOiMIAcAANAZQQ4AAKAzghwAAEBnBDkAAIDOCHIAAACdEeQAAAA6I8gBAAB0RpADAADojCAHAADQGUEOAACgM4IcAABAZwQ5AACAzghyAAAAnRHkAAAAOiPIAQAAdEaQAwAA6IwgBwAA0BlBDgAAoDOCHAAAQGcEOQAAgM4IcgAAAJ0R5AAAADojyAEAAHRGkAMAAOiMIAcAANAZQQ4AAKAzghwAAEBnBDkAAIDOCHIAAACdEeQAAAA6I8gBAAB0RpADAADojCAHAADQGUEOAACgM4IcAABAZwQ5AACAzghyAAAAnRHkAAAAOjNzkKuq/arq3Kq6bvx5zxXK7lNVN1TVa2atFwAAYF6tRY/ciUnOb60dkuT8cXw5v53kgjWoEwAAYG6tRZA7Osnp4/DpSY6ZVqiqHpbkwCTvW4M6AQAA5tZaBLkDW2s3jsNfyhDWbqOqdkvyB0levgb1AQAAzLUN21Koqs5Lcu8ps06aHGmttapqU8q9OMnZrbUbqmq1ujYn2ZwkGzdu3JbmAQAAzJVtCnKttSOXm1dVX66qg1prN1bVQUm+MqXY4UkeW1UvTrJ3kj2r6qbW2u2ep2utbUmyJUkWFhamhUIAAIC5tk1BbhVnJTk+ySnjzzOXFmitPW9xuKpOSLIwLcQBAACwurV4Ru6UJE+uquuSHDmOp6oWqurUNVg/AAAAE2bukWutfT3Jk6ZMvzjJC6dMPy3JabPWCwAAMK/WokcOAACAnUiQAwAA6IwgBwAA0BlBDgAAoDOCHAAAQGcEOQAAgM4IcgAAAJ0R5AAAADojyAEAAHRGkAMAAOiMIAcAANAZQQ4AAKAzghwAAEBnBDkAAIDOCHIAAACdEeQAAAA6I8gBAAB0RpADAADojCAHAADQGUEOAACgM4IcAABAZwQ5AACAzghyAAAAnRHkAAAAOiPIAQAAdEaQAwAA6IwgBwAA0BlBDgAAoDOCHAAAQGcEOQAAgM4IcgAAAJ0R5AAAADojyAEAAHRGkAMAAOiMIAcAANAZQQ4AAKAzghwAAEBnBDkAAIDOCHIAAACdEeQAAAA6I8gBAAB0RpADAADojCAHAADQGUEOAACgM4IcAABAZwQ5AACAzghyAAAAnRHkAAAAOiPIAQAAdGamIFdV+1XVuVV13fjznsuU21hV76uqa6rq6qraNEu9AAAA82zWHrkTk5zfWjskyfnj+DRvTPLK1toPJnlEkq/MWC8AAMDcmjXIHZ3k9HH49CTHLC1QVYcm2dBaOzdJWms3tda+PWO9AAAAc2vWIHdga+3GcfhLSQ6cUubBSb5RVe+sqo9X1SuravcZ6wUAAJhbG1YrUFXnJbn3lFknTY601lpVtWXqeGySH03yD0nemuSEJH+2TH2bk2xOko0bN67WPAAAgLmzapBrrR253Lyq+nJVHdRau7GqDsr0Z99uSHJpa+0z4zLvSvLILBPkWmtbkmxJkoWFhWnBEAAAYK7NemvlWUmOH4ePT3LmlDIXJdm3qg4Yx5+Y5OoZ6wUAAJhbswa5U5I8uaquS3LkOJ6qWqiqU5OktXZLkpcnOb+qrkhSSf50xnoBAADm1qq3Vq6ktfb1JE+aMv3iJC+cGD83yUNnqQsAAIDBrD1yAAAA7GSCHAAAQGcEOQAAgM4IcgAAAJ0R5AAAADojyAEAAHRGkAMAAOiMIAcAANAZQQ4AAKAzghwAAEBnBDkAAIDOCHIAAACdEeQAAAA6I8gBAAB0RpADAADojCAHAADQGUEOAACgM4IcAABAZwQ5AACAzghyAAAAnRHkAAAAOiPIAQAAdEaQAwAA6IwgBwAA0BlBDgAAoDOCHAAAQGcEOQAAgM4IcgAAAJ0R5AAAADojyAEAAHRGkAMAAOiMIAcAANAZQQ4AAKAzghwAAEBnBDkAAIDOCHIAAACdEeQAAAA6I8gBAAB0RpADAADojCAHAADQGUEOAACgM4IcAABAZwQ5AACAzghyAAAAnRHkAAAAOiPIAQAAdEaQAwAA6IwgBwAA0BlBDgAAoDMzB7mq2q+qzq2q68af91ym3O9X1VVVdU1V/XFV1ax1AwAAzKO16JE7Mcn5rbVDkpw/jt9GVT0qyaOTPDTJQ5I8PMnj16BuAACAubMWQe7oJKePw6cnOWZKmZZkryR7JrlLkj2SfHkN6gYAAJg7axHkDmyt3TgOfynJgUsLtNY+lORvk9w4vs5prV0zbWVVtbmqLq6qi7/61a+uQfMAAAB2LRu2pVBVnZfk3lNmnTQ50lprVdWmLP+gJD+Y5L7jpHOr6rGttQ8uLdta25JkS5IsLCzcbl0AAADzbpuCXGvtyOXmVdWXq+qg1tqNVXVQkq9MKfbMJB9urd00LvPXSQ5PcrsgBwAAwMrW4tbKs5IcPw4fn+TMKWX+Icnjq2pDVe2R4YtOpt5aCQAAwMrWIsidkuTJVXVdkiPH8VTVQlWdOpZ5e5JPJ7kiyWVJLmutvXsN6gYAAJg723Rr5Upaa19P8qQp0y9O8sJx+JYk/2HWugAAAFibHjkAAAB2IkEOAACgM4IcAABAZwQ5AACAzghyAAAAnRHkAAAAOiPIAQAAdEaQAwAA6IwgBwAA0BlBDgAAoDOCHAAAQGcEOQAAgM4IcgAAAJ0R5AAAADojyAEAAHRGkAMAAOiMIAcAANAZQQ4AAKAzghwAAEBnBDkAAIDOCHIAAACdEeQAAAA6I8gBAAB0RpADAADojCAHAADQGUEOAACgM4IcAABAZwQ5AACAzghyAAAAnRHkAAAAOrNhvRsAPfvcKUetdxMAAJhDeuQAAAA6I8gBAAB0RpADAADojCAHAADQGUEOAACgM4IcAABAZwQ5AACAzghyAAAAnRHkAAAAOiPIAQAAdEaQAwAA6IwgBwAA0BlBDgAAoDOCHAAAQGcEOQAAgM4IcgAAAJ0R5AAAADozU5CrqmOr6qqqurWqFlYo99SquraqPlVVJ85SJwAAwLybtUfuyiTPSnLBcgWqavckr03ytCSHJnluVR06Y70AAABza8MsC7fWrkmSqlqp2COSfKq19pmx7FuSHJ3k6lnqBgAAmFc74xm5g5N8fmL8hnEaAAAAO2DVHrmqOi/JvafMOqm1duZaN6iqNifZnCQbN25c69UDAAB0b9Ug11o7csY6vpDkfhPj9x2nLVffliRbkmRhYaHNWDcAAMAuZ2fcWnlRkkOq6gFVtWeS5yQ5ayfUCwAAsEua9d8PPLOqbkhyeJL3VtU54/T7VNXZSdJauznJLyY5J8k1Sf6qtXbVbM0GAACYX7N+a+UZSc6YMv2LSZ4+MX52krNnqQsAAIDBzri1EgAAgDUkyAEAAHRGkAMAAOiMIAcAANAZQQ4AAKAzghwAAEBnBDkAAIDOCHIAAACdEeQAAAA6I8gBAAB0RpADAADojCAHAADQGUEOAACgM4IcAABAZwQ5AACAzghyAAAAnRHkAAAAOiPIAQAAdEaQAwAA6IwgBwAA0BlBDgAAoDOCHAAAQGcEOQAAgM4IcgAAAJ0R5AAAADojyAEAAHRGkAMAAOiMIAcAANAZQQ4AAKAzghwAAEBnBDkAAIDOCHIAAACdEeQAAAA6I8gBAAB0RpADAADojCAHAADQGUEOAACgM4IcAABAZwQ5AACAzghyAAAAnRHkAAAAOiPIAQAAdEaQAwAA6IwgBwAA0BlBDgAAoDOCHAAAQGcEOQAAgM4IcgAAAJ0R5AAAADozU5CrqmOr6qqqurWqFpYpc7+q+tuqunos+7JZ6gQAAJh3s/bIXZnkWUkuWKHMzUl+rbV2aJJHJnlJVR06Y70AAABza8MsC7fWrkmSqlqpzI1JbhyH/7mqrklycJKrZ6kbAABgXu3UZ+SqalOSH03ykZ1ZLwAAwK5k1R65qjovyb2nzDqptXbmtlZUVXsneUeSX26t/dMK5TYn2ZwkGzdu3NbVAwAAzI1Vg1xr7chZK6mqPTKEuD9vrb1zlfq2JNmSJAsLC23WugEAAHY1d/itlTU8QPdnSa5prf3hHV0fAADArm7Wfz/wzKq6IcnhSd5bVeeM0+9TVWePxR6d5PlJnlhVl46vp8/UagAAgDk267dWnpHkjCnTv5jk6ePw3yVZ/mstAQAA2C479VsrAQAAmJ0gBwAA0BlBDgAAoDOCHAAAQGcEOQAAgM4IcgAAAJ0R5AAAADojyAEAAHRGkAMAAOiMIAcAANAZQQ4AAKAzghwAAEBnBDkAAIDOCHIAAACdEeQAAAA6I8gBAAB0RpADAADojCAHAADQGUEOAACgM4IcAABAZwQ5AACAzghyAAAAnRHkAAAAOiPIAQAAdEaQAwAA6IwgBwAA0BlBDgAAoDOCHAAAQGcEOQAAgM4IcgAAAJ0R5AAAADojyAEAAHRGkAMAAOiMIAcAANAZQQ4AAKAzghwAAEBnBDkAAIDOCHIAAACdEeQAAAA6I8gBAAB0RpADAADojCAHAADQGUEOAACgM4IcAABAZzasdwMAYFfyuVOOWu8mADAH9MgBAAB0RpADAADojCAHAADQmZmCXFUdW1VXVdWtVbWwStndq+rjVfWeWeoEAACYd7P2yF2Z5FlJLtiGsi9Lcs2M9QEAAMy9mYJca+2a1tq1q5WrqvsmOSrJqbPUBwAAwM57Ru6Pkvx6klt3Un0AAAC7rFX/j1xVnZfk3lNmndRaO3Mblv+JJF9prV1SVUdsQ/nNSTYnycaNG1crDgAAMHdWDXKttSNnrOPRSZ5RVU9PsleSfarqza21n12mvi1JtiTJwsJCm7FuAACAXc4dfmtla+03Wmv3ba1tSvKcJO9fLsQBAACwuln//cAzq+qGJIcneW9VnTNOv09Vnb0WDQQAAOC2Vr21ciWttTOSnDFl+heTPH3K9A8k+cAsdQIAAMy7nfWtlQAAAKwRQQ4AAKAzghwAAEBnBDkAAIDOCHIAAACdEeQAAAA6I8gBAAB0plpr692GZVXVV5Ncv97tmAP7J/naejeCOwX7AovsC0yyP7DIvsAi+8LOc//W2gFLJ96pgxw7R1Vd3FpbWO92sP7sCyyyLzDJ/sAi+wKL7Avrz62VAAAAnRHkAAAAOiPIkSRb1rsB3GnYF1hkX2CS/YFF9gUW2RfWmWfkAAAAOqNHDgAAoDOC3E5SVfeqqkvH15eq6gsT43tuw/InVNVrtqHc06rq4qq6uqo+XlV/sDa/wbL1nVZVz16lzAlVdZ/tXO+mqrpyttbtmqqqVdWbJ8Y3VNVXq+o9M6735Kp6+ewtnLrum+6I9e7KqurAqvqLqvpMVV1SVR+qqmfuxPqPmHWfWmX9+1bVi3dWffOqqm6ZONdcWlUnjtNPrapDd2B9h1XV03dguQ9UlW+32wZVdVJVXVVVl4/b7MdXKb9df9vxvfao2Vu6/cbzzOTnn1Oq6kVV9YI1WPc2fU5iq4njw5VV9e6q2ncH13Ofqnr7GjePbbBhvRswL1prX09yWDIcyJLc1Fr7n2tZR1U9JMlrkhzVWvtEVe2eZPNa1rGDTkhyZZIvrnM7dhXfSvKQqrpra+1fkjw5yRe2ZwVVVRlurb51RxtRVRtaazfv6PIsb9w+70pyemvtZ8Zp90/yjPVs1xrbN8mLk/zJOrdjV/cvrbXDlk5srb1wB9d3WJKFJGfP0CaWUVWHJ/mJJD/WWvu3qto/yaoXe7fTEUluSvL3a7zebfWqtf78ww777vGhqk5P8pIkv7u9K2mtfTHJihf1uWPokVtHVfULVXVRVV1WVe+oqruN048dr45cVlUXTFnuqPHq/P5LZv16kt9trX0iSVprt7TWXjcus6mq3j9e4Tu/qjaO00+rqtdV1YfHK/9HVNUbquqaqjptos6bqupV41XC86vq9v+UsOphVfV/x96Dc6rqoLG3biHJn49Xfe46rdzE8pdV1WUZDiYs7+wkR43Dz03yl4szlvasjfvSpvF1bVW9MUOwvt945feTVfV3Sb5/YpkHVtXfjNvog1X1A+P006rq9VX1kSS/v0K5B4z76BVV9Tt3+F9j1/PEJN9prb1+cUJr7frW2quT776fP1hVHxtfjxqnHzG+t84c38+nVNXzquqj47Z44FjugPGYc9H4evS2NqyqnjJu249V1duqau9x+ueq6jfH6VdM7AsHVNW547Hj1Kq6fjx2nZLkgeNx4ZXj6veuqrdX1Seq6s+rqtbij8nt1UQvzgrb9OFV9ffjcfmjVXWPJL+V5Lhxux1XVXcfzxkfreEukKPHZe9aVW8ZzyVnJLnruv2yfTkoyddaa/+WJK21r40fklNVrxjfr1dW1ZYl74/n19aelUeM5ferqnfVcN7/cFU9tKo2JXlRkl8Zyz+2qn6yqj4ybr/zqurAcfnl3rupqp8dt/mlVfW/q2r38XXa2IYrqupXtuUXrvGcVcPdJRdV1RHj9P9RVb+7XH3j9J+r4Rz20STbfBxjqg8lOThZ8TPAA8d96Yqq+p0a77apibuoqmqvqvo/Y5mPV9UTxuknVNU7x/VeV1W/v06/566ltea1k19JTk7y8iT3mpj2O0leOg5fkeTgcXjf8ecJGXrbnpnkg0nuOWW9H0vyI8vU+e4kx4/D/z7Ju8bh05K8JUklOTrJPyX54Qwh/5Ikh43lWpLnjcOvSPKaieWfnWSPDFf3DhinH5fkDePwB5IsjMMrlbs8yePG4VcmuXK9t9Wd8ZXhSupDk7w9yV5JLs1whfU9k/vXRPkrk2waX7cmeeQ4/WHjvna3JPsk+dTicknOT3LIOPzjSd4/sb3fk2T3VcqdleQF4/BLMvRAr/vfrpdXkl/KcNV6ufl3S7LXOHxIkovH4SOSfCPDh8G7ZOip/c1x3suS/NE4/BdJHjMOb0xyzZQ6vrtPTUzbP8kFSe4+jv+nJK8Yhz+XrcewFyc5dRx+TZLfGIefmuFYsv+4P165pL5vJrlvhuPPhxbb6DXTvnTLeIxYfB03Tv9AhotsU7dphl6gzyR5+Dh9nwx38ZyQ8fg/Tv/vSX52HN43ySeT3D3Jr2brsf2hSW7OeB7wWnF77T1up09m6K1+/MS8/SaG35TkJye25Z+Ow49bfF8leXWS/zYOPzHJpePwybntOeKe2frldy9M8gfj8HLv3R/M8Jlij3HenyR5QYZzyrkT6913yu93cobj0uL++O8m25Pkh5Jck+TIJB8f98Pl6jsoyT8kOWAsd+Hkvum1TfvbTePP3ZO8LclTx/Hlzu3vSfLccfhFE8tvmtjvfm3ivf8D4zbaazx2fCbJPcbx65Pcb73/Br2/3Fq5vh5SQ2/FvhkO3ueM0y9MclpV/VWSd06Uf2KGE+9TWmv/tJ11HZ7kWePwm5JMXgl5d2utVdUVSb7cWrsiSarqqgxvzkszBIC3juXfvKRdydCb85Ak544XCXdPcuOUdkwtV8N92fu21hZ7IN+U5Gnb+TvOjdba5eOV1edm+25xur619uFx+LFJzmitfTtJquqs8efeSR6V5G0TF3zvMrGOt7XWblml3KOT/NQ4/KYkv7cdbWSJqnptksdk6KV7eIYLIq+pqsMyfFB/8ETxi1prN47LfTrJ+8bpVyR5wjh8ZJJDJ7bbPlW1d2tttWcZH5nk0CQXjsvumSFwLVo8LlySrcebx2S4AJXW2t9U1T+usP6PttZuGNt+aYbjz9+t0iZWNvXWygnLbdPvT3Jja+2iJFk850zpJH1KkmfU1rsA9spwceBxSf54XPbyqrp8LX6ZXV1r7aaqeliG4/MTkry1qk5srZ2W5AlV9esZLuTsl+SqDAEnGe/KaK1dUFX7jOfUx2Q8DrfW3l/Ds/r7TKn2vmM9B2XY/p8dpy/33n1ShtB20bg/3DXJV8a2fF9VvTrJe7P12LPUq9rErZU13E66+PtfVVVvyhAYDm+tfaeqlqvvx5N8oLX21XE9b81tj4Ws7q7jsfbgDAH63FXO7YcnOWYc/osk026RfUyGiwhpw2M+12frdjm/tfbNJKmqq5PcP8nn1/D3mTuC3Po6LckxrbXLquqEDFek01p7UQ0PNx+V5JLxoJ4kn07yfRneEBdPWd9VGQ52l21nO/5t/HnrxPDi+HL7yNL/W1FJrmqtHT6t8GrlagcfsJ1zZ2U4iB6R5F4T02/ObW+b3mti+FvbsN7dknxjhQ9/39rGcv63yY67KluDcFprLxlvaVp83/9Kki8n+ZEM2+FfJ5Zd+h6efH8vvp93y9AzO7nctqgMV9yfu8z8xbpuyY6dXybbvqPrYPtM3aZV9cPbsfxPtdauXbL8GjVv/rTWbsnQy/aB8QLr8VX1lgw9UQuttc/X8Kz95LF96fF2e46/r07yh621s8bbGk9epXxleH73N243o+pHMvSyvSjJT2e4A2h7/XCGOwu+d6X6quqYHVg3t/UvrbXDani055wMd9CclpXP7bNwjF9jnpFbX9+ToTdqjyTPW5xYVQ9srX2ktfaKJF9Ncr9x1vUZPty9sap+aMr6XpnkP1fVg8f17FZVLxrn/X2S54zDz8twe+b22C1bH2T9mdz+Kvm1SQ5YvLJWVXtMtPGfx9912XKttW8k+UZVPWaijazsDRlum7tiyfTPJfmxJKmqH0vygGWWvyDJMTU8y/I9SX4y+e6V989W1bHjOmo8Od/GKuUuzG33N7bP+5PsVVX/cWLa3SaG75Ght+TWJM/P0LO9Pd6X5KWLI2PP3rb4cJJHV9WDxuXuvni8WcGFGT7QpaqekuE2ruS2xwXWz3Lb9NokB1XVw8fp31NVG3L77XZOkpfWmNyq6kfH6RdkOFcsfhHXQ3fGL9O7qvr+qjpkYtJhGc79i6Hta2OPydIvljhuXP4xSb459np8MOPxdwxoXxuP20u34T2y9Quzjp+Yvtx79/wkz66q7x3n7VdV9x8vNu3WWntHkv+S8Ty0nb//szL0Nj4uyavHi7xT60vykSSPH3sa90hy7PbWx2C8M+eXMtwW+e0sf27/cLZeZHzO7VY0mNzvHpyhh/7aZcoyI0Fuff3XDAeiC5N8YmL6K2t4SPTKDAHsuz1sbfgik+dl6PJ+4OTKWmuXJ/nlJH9ZVddkeDbq+8bZL03yc+PtLc/P8LzM9vhWkkeMbXpihgfeJ+v+ToYTy+/V8GUll2bomk+GqzuvH7vvd1+h3M8lee1YzuXcVbTWbmit/fGUWe9Isl8Nt8b+YoZnLaYt/7EMt8teluSvk1w0Mft5SX5+3EZXZXh+cprlyr0syUvGq8kHb9cvRlprLcPtK4+vqs/W8CD/6RmeX0qGK/PHj3/3H8i29bRO+qUkCzV8CcLVGa6eT/Okqrph8ZXkQRmec/jL8VjyobH+lfxmkqeMx45jk3wpyT+34Zt8L6zhixFeueIamMVd67b/fuCUiXltvC3thCzZpuMx/bgMH6YvS3JuhjDxtxluy720qo5L8tsZbvW9fDzm/Pa47tdl+PKaazKcLy6543/VXcLeSU6v4V8IXZ7htteTx4udf5rhvH5Obnu8TpJ/raqPJ3l9kp8fp52c5GHjek7J1pD27iTPHLfhY8dyb6uqS5J8bWKdy713r84Q1N43rvvcDM+rHZyhF/HSDI9g3K7HbiW19UuQXtha+2SGZ/T+13L1jbeQn5xhn70ww62B7KDW2sczfFfBc7P8uf2Xk/zquB0elOG55qX+JMlu4/n/rUlOaOOX97D2Fh9uhRVV1U2ttb3Xux1AX6rqLkluaa3dPPbEv+4OumWH7TB+yHpGa+2zqxZmLnnvslQNt2D+y/i9Cs/J8MUny13oZSdwbyoAd6SNSf6qqnZL8p0kv7DO7Zl7VXVukiuEOFbhvctSD8vwRVuV4TnGHXkGkjWkRw4AAKAznpEDAADojCAHAADQGUEOAACgM4IcAABAZwQ5AACAzghyAAAAnfn/2bl7YtIPz9wAAAAASUVORK5CYII=\n",
+      "text/plain": [
+       "<Figure size 1080x576 with 1 Axes>"
+      ]
+     },
+     "metadata": {
+      "needs_background": "light"
+     },
+     "output_type": "display_data"
+    }
+   ],
+   "source": [
+    "from sklearn.linear_model import LogisticRegression\n",
+    "from sklearn.metrics import roc_auc_score\n",
+    "\n",
+    "#train data\n",
+    "X_train = train_data.drop([\"Outcome\", \"Team\", \"User\"], axis=1)\n",
+    "y_train = train_data[\"Outcome\"]\n",
+    "clf = LogisticRegression(random_state=0).fit(X_train, y_train)\n",
+    "coefs = clf.coef_[0]\n",
+    "\n",
+    "plt.figure(figsize=(15,8))\n",
+    "plt.bar(X_train.columns, coefs)"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 93,
+   "id": "6f153480",
+   "metadata": {},
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "AUC: 0.6136017272034544\n"
+     ]
+    }
+   ],
+   "source": [
+    "# test data\n",
+    "X_test = test_data.drop([\"Outcome\", \"Team\", \"User\"], axis=1)\n",
+    "y_true = test_data[\"Outcome\"]\n",
+    "y_predict = clf.predict(X_test)\n",
+    "\n",
+    "auc = roc_auc_score(y_true, y_predict)\n",
+    "print(\"AUC:\", auc)"
+   ]
+  }
+ ],
+ "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": "3.9.7"
+  }
+ },
+ "nbformat": 4,
+ "nbformat_minor": 5
+}
diff --git a/Among_Us_Prediction_Jason.ipynb b/Among_Us_Prediction_Jason.ipynb
deleted file mode 100644
index 6356ce8..0000000
--- a/Among_Us_Prediction_Jason.ipynb
+++ /dev/null
@@ -1,668 +0,0 @@
-{
- "cells": [
-  {
-   "cell_type": "code",
-   "execution_count": 1,
-   "id": "9c9469a8",
-   "metadata": {},
-   "outputs": [],
-   "source": [
-    "import numpy as np\n",
-    "import pandas as pd"
-   ]
-  },
-  {
-   "cell_type": "code",
-   "execution_count": 2,
-   "id": "c03c47a8",
-   "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>Game Completed Date</th>\n",
-       "      <th>Team</th>\n",
-       "      <th>Outcome</th>\n",
-       "      <th>Task Completed</th>\n",
-       "      <th>All Tasks Completed</th>\n",
-       "      <th>Murdered</th>\n",
-       "      <th>Imposter Kills</th>\n",
-       "      <th>Game Length</th>\n",
-       "      <th>Ejected</th>\n",
-       "      <th>Sabotages Fixed</th>\n",
-       "      <th>Time to complete all tasks</th>\n",
-       "      <th>Rank Change</th>\n",
-       "      <th>Region/Game Code</th>\n",
-       "      <th>User</th>\n",
-       "    </tr>\n",
-       "  </thead>\n",
-       "  <tbody>\n",
-       "    <tr>\n",
-       "      <th>0</th>\n",
-       "      <td>12/13/2020 at 1:26:56 am EST</td>\n",
-       "      <td>Crewmate</td>\n",
-       "      <td>Win</td>\n",
-       "      <td>3</td>\n",
-       "      <td>No</td>\n",
-       "      <td>Yes</td>\n",
-       "      <td>-</td>\n",
-       "      <td>07m 04s</td>\n",
-       "      <td>No</td>\n",
-       "      <td>2.0</td>\n",
-       "      <td>-</td>\n",
-       "      <td>++</td>\n",
-       "      <td>NA / WYMSBF</td>\n",
-       "      <td>1</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>1</th>\n",
-       "      <td>12/13/2020 at 1:17:42 am EST</td>\n",
-       "      <td>Crewmate</td>\n",
-       "      <td>Loss</td>\n",
-       "      <td>7</td>\n",
-       "      <td>Yes</td>\n",
-       "      <td>No</td>\n",
-       "      <td>-</td>\n",
-       "      <td>16m 21s</td>\n",
-       "      <td>No</td>\n",
-       "      <td>1.0</td>\n",
-       "      <td>09m 48s</td>\n",
-       "      <td>--</td>\n",
-       "      <td>NA / WYMSBF</td>\n",
-       "      <td>1</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>2</th>\n",
-       "      <td>12/13/2020 at 12:57:47 am EST</td>\n",
-       "      <td>Crewmate</td>\n",
-       "      <td>Win</td>\n",
-       "      <td>3</td>\n",
-       "      <td>No</td>\n",
-       "      <td>No</td>\n",
-       "      <td>-</td>\n",
-       "      <td>11m 33s</td>\n",
-       "      <td>No</td>\n",
-       "      <td>0.0</td>\n",
-       "      <td>-</td>\n",
-       "      <td>++</td>\n",
-       "      <td>NA / WYMSBF</td>\n",
-       "      <td>1</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>3</th>\n",
-       "      <td>12/13/2020 at 12:41:55 am EST</td>\n",
-       "      <td>Imposter</td>\n",
-       "      <td>Win</td>\n",
-       "      <td>-</td>\n",
-       "      <td>-</td>\n",
-       "      <td>-</td>\n",
-       "      <td>2</td>\n",
-       "      <td>08m 05s</td>\n",
-       "      <td>No</td>\n",
-       "      <td>NaN</td>\n",
-       "      <td>-</td>\n",
-       "      <td>+++</td>\n",
-       "      <td>Europe / QIRTNF</td>\n",
-       "      <td>1</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>4</th>\n",
-       "      <td>12/13/2020 at 12:30:37 am EST</td>\n",
-       "      <td>Crewmate</td>\n",
-       "      <td>Loss</td>\n",
-       "      <td>4</td>\n",
-       "      <td>No</td>\n",
-       "      <td>No</td>\n",
-       "      <td>-</td>\n",
-       "      <td>05m 10s</td>\n",
-       "      <td>No</td>\n",
-       "      <td>0.0</td>\n",
-       "      <td>-</td>\n",
-       "      <td>---</td>\n",
-       "      <td>Europe / QIRTNF</td>\n",
-       "      <td>1</td>\n",
-       "    </tr>\n",
-       "  </tbody>\n",
-       "</table>\n",
-       "</div>"
-      ],
-      "text/plain": [
-       "             Game Completed Date      Team Outcome Task Completed  \\\n",
-       "0   12/13/2020 at 1:26:56 am EST  Crewmate     Win              3   \n",
-       "1   12/13/2020 at 1:17:42 am EST  Crewmate    Loss              7   \n",
-       "2  12/13/2020 at 12:57:47 am EST  Crewmate     Win              3   \n",
-       "3  12/13/2020 at 12:41:55 am EST  Imposter     Win              -   \n",
-       "4  12/13/2020 at 12:30:37 am EST  Crewmate    Loss              4   \n",
-       "\n",
-       "  All Tasks Completed Murdered Imposter Kills Game Length Ejected  \\\n",
-       "0                  No      Yes              -     07m 04s      No   \n",
-       "1                 Yes       No              -     16m 21s      No   \n",
-       "2                  No       No              -     11m 33s      No   \n",
-       "3                   -        -              2     08m 05s      No   \n",
-       "4                  No       No              -     05m 10s      No   \n",
-       "\n",
-       "   Sabotages Fixed Time to complete all tasks Rank Change Region/Game Code  \\\n",
-       "0              2.0                          -          ++      NA / WYMSBF   \n",
-       "1              1.0                    09m 48s          --      NA / WYMSBF   \n",
-       "2              0.0                          -          ++      NA / WYMSBF   \n",
-       "3              NaN                          -         +++  Europe / QIRTNF   \n",
-       "4              0.0                          -         ---  Europe / QIRTNF   \n",
-       "\n",
-       "   User  \n",
-       "0     1  \n",
-       "1     1  \n",
-       "2     1  \n",
-       "3     1  \n",
-       "4     1  "
-      ]
-     },
-     "execution_count": 2,
-     "metadata": {},
-     "output_type": "execute_result"
-    }
-   ],
-   "source": [
-    "files = []\n",
-    "for i in range(29):\n",
-    "    tempdf = pd.read_csv(f'data/User{i+1}.csv', delimiter = ',')\n",
-    "    tempdf['User'] = i+1\n",
-    "    files.append(tempdf)\n",
-    "df = pd.concat(files, ignore_index = True)\n",
-    "df.head()"
-   ]
-  },
-  {
-   "cell_type": "code",
-   "execution_count": 3,
-   "id": "f1220e69",
-   "metadata": {},
-   "outputs": [
-    {
-     "data": {
-      "text/plain": [
-       "Game Completed Date            object\n",
-       "Team                           object\n",
-       "Outcome                        object\n",
-       "Task Completed                 object\n",
-       "All Tasks Completed            object\n",
-       "Murdered                       object\n",
-       "Imposter Kills                 object\n",
-       "Game Length                    object\n",
-       "Ejected                        object\n",
-       "Sabotages Fixed               float64\n",
-       "Time to complete all tasks     object\n",
-       "Rank Change                    object\n",
-       "Region/Game Code               object\n",
-       "User                            int64\n",
-       "dtype: object"
-      ]
-     },
-     "execution_count": 3,
-     "metadata": {},
-     "output_type": "execute_result"
-    }
-   ],
-   "source": [
-    "df.dtypes"
-   ]
-  },
-  {
-   "cell_type": "code",
-   "execution_count": 4,
-   "id": "ac647218",
-   "metadata": {},
-   "outputs": [],
-   "source": [
-    "df[['Task Completed', 'Imposter Kills']] = df[['Task Completed', 'Imposter Kills']].apply(pd.to_numeric, errors = 'coerce')\n"
-   ]
-  },
-  {
-   "cell_type": "code",
-   "execution_count": 5,
-   "id": "c8322c13",
-   "metadata": {},
-   "outputs": [
-    {
-     "data": {
-      "text/plain": [
-       "NA         1436\n",
-       "Europe      791\n",
-       "Name: Region, dtype: int64"
-      ]
-     },
-     "execution_count": 5,
-     "metadata": {},
-     "output_type": "execute_result"
-    }
-   ],
-   "source": [
-    "df['Region'] = df['Region/Game Code'].str.extract(r'^(.*?)/')\n",
-    "df['Region'].value_counts()\n"
-   ]
-  },
-  {
-   "cell_type": "code",
-   "execution_count": 6,
-   "id": "b4b42965",
-   "metadata": {},
-   "outputs": [],
-   "source": [
-    "df['Game Length'] = df['Game Length'].apply(pd.to_timedelta)\n",
-    "df['Game Length'] = df['Game Length'] / np.timedelta64(1, 's')"
-   ]
-  },
-  {
-   "cell_type": "code",
-   "execution_count": 7,
-   "id": "8cbfe4e2",
-   "metadata": {},
-   "outputs": [],
-   "source": [
-    "df['Time to complete all tasks'] = df['Time to complete all tasks'].apply(pd.to_timedelta, errors = 'coerce')\n",
-    "df['Time to complete all tasks'] = df['Time to complete all tasks'] / np.timedelta64(1, 's')"
-   ]
-  },
-  {
-   "cell_type": "code",
-   "execution_count": 8,
-   "id": "fe3b13f2",
-   "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>Team</th>\n",
-       "      <th>Outcome</th>\n",
-       "      <th>Task Completed</th>\n",
-       "      <th>All Tasks Completed</th>\n",
-       "      <th>Murdered</th>\n",
-       "      <th>Game Length</th>\n",
-       "      <th>Ejected</th>\n",
-       "      <th>Sabotages Fixed</th>\n",
-       "      <th>Region</th>\n",
-       "    </tr>\n",
-       "  </thead>\n",
-       "  <tbody>\n",
-       "    <tr>\n",
-       "      <th>0</th>\n",
-       "      <td>Crewmate</td>\n",
-       "      <td>Win</td>\n",
-       "      <td>3.0</td>\n",
-       "      <td>No</td>\n",
-       "      <td>Yes</td>\n",
-       "      <td>424.0</td>\n",
-       "      <td>No</td>\n",
-       "      <td>2.0</td>\n",
-       "      <td>NA</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>1</th>\n",
-       "      <td>Crewmate</td>\n",
-       "      <td>Loss</td>\n",
-       "      <td>7.0</td>\n",
-       "      <td>Yes</td>\n",
-       "      <td>No</td>\n",
-       "      <td>981.0</td>\n",
-       "      <td>No</td>\n",
-       "      <td>1.0</td>\n",
-       "      <td>NA</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>2</th>\n",
-       "      <td>Crewmate</td>\n",
-       "      <td>Win</td>\n",
-       "      <td>3.0</td>\n",
-       "      <td>No</td>\n",
-       "      <td>No</td>\n",
-       "      <td>693.0</td>\n",
-       "      <td>No</td>\n",
-       "      <td>0.0</td>\n",
-       "      <td>NA</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>4</th>\n",
-       "      <td>Crewmate</td>\n",
-       "      <td>Loss</td>\n",
-       "      <td>4.0</td>\n",
-       "      <td>No</td>\n",
-       "      <td>No</td>\n",
-       "      <td>310.0</td>\n",
-       "      <td>No</td>\n",
-       "      <td>0.0</td>\n",
-       "      <td>Europe</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>5</th>\n",
-       "      <td>Crewmate</td>\n",
-       "      <td>Loss</td>\n",
-       "      <td>7.0</td>\n",
-       "      <td>Yes</td>\n",
-       "      <td>Yes</td>\n",
-       "      <td>982.0</td>\n",
-       "      <td>No</td>\n",
-       "      <td>0.0</td>\n",
-       "      <td>Europe</td>\n",
-       "    </tr>\n",
-       "  </tbody>\n",
-       "</table>\n",
-       "</div>"
-      ],
-      "text/plain": [
-       "       Team Outcome  Task Completed All Tasks Completed Murdered  Game Length  \\\n",
-       "0  Crewmate     Win             3.0                  No      Yes        424.0   \n",
-       "1  Crewmate    Loss             7.0                 Yes       No        981.0   \n",
-       "2  Crewmate     Win             3.0                  No       No        693.0   \n",
-       "4  Crewmate    Loss             4.0                  No       No        310.0   \n",
-       "5  Crewmate    Loss             7.0                 Yes      Yes        982.0   \n",
-       "\n",
-       "  Ejected  Sabotages Fixed   Region  \n",
-       "0      No              2.0      NA   \n",
-       "1      No              1.0      NA   \n",
-       "2      No              0.0      NA   \n",
-       "4      No              0.0  Europe   \n",
-       "5      No              0.0  Europe   "
-      ]
-     },
-     "execution_count": 8,
-     "metadata": {},
-     "output_type": "execute_result"
-    }
-   ],
-   "source": [
-    "crewmate = df[['Team',\n",
-    "               'Outcome',\n",
-    "               'Task Completed',\n",
-    "               'All Tasks Completed',\n",
-    "               'Murdered',\n",
-    "               'Game Length',\n",
-    "               'Ejected',\n",
-    "              'Sabotages Fixed',\n",
-    "              'Region']][df['Team'] == 'Crewmate']\n",
-    "crewmate.head()"
-   ]
-  },
-  {
-   "cell_type": "code",
-   "execution_count": 9,
-   "id": "30fb4f2c",
-   "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>Team</th>\n",
-       "      <th>Outcome</th>\n",
-       "      <th>Task Completed</th>\n",
-       "      <th>All Tasks Completed</th>\n",
-       "      <th>Murdered</th>\n",
-       "      <th>Game Length</th>\n",
-       "      <th>Ejected</th>\n",
-       "      <th>Sabotages Fixed</th>\n",
-       "      <th>Region</th>\n",
-       "    </tr>\n",
-       "  </thead>\n",
-       "  <tbody>\n",
-       "    <tr>\n",
-       "      <th>0</th>\n",
-       "      <td>Crewmate</td>\n",
-       "      <td>1</td>\n",
-       "      <td>3.0</td>\n",
-       "      <td>0</td>\n",
-       "      <td>1</td>\n",
-       "      <td>424.0</td>\n",
-       "      <td>0</td>\n",
-       "      <td>2.0</td>\n",
-       "      <td>0</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>1</th>\n",
-       "      <td>Crewmate</td>\n",
-       "      <td>0</td>\n",
-       "      <td>7.0</td>\n",
-       "      <td>1</td>\n",
-       "      <td>0</td>\n",
-       "      <td>981.0</td>\n",
-       "      <td>0</td>\n",
-       "      <td>1.0</td>\n",
-       "      <td>0</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>2</th>\n",
-       "      <td>Crewmate</td>\n",
-       "      <td>1</td>\n",
-       "      <td>3.0</td>\n",
-       "      <td>0</td>\n",
-       "      <td>0</td>\n",
-       "      <td>693.0</td>\n",
-       "      <td>0</td>\n",
-       "      <td>0.0</td>\n",
-       "      <td>0</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>4</th>\n",
-       "      <td>Crewmate</td>\n",
-       "      <td>0</td>\n",
-       "      <td>4.0</td>\n",
-       "      <td>0</td>\n",
-       "      <td>0</td>\n",
-       "      <td>310.0</td>\n",
-       "      <td>0</td>\n",
-       "      <td>0.0</td>\n",
-       "      <td>1</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>5</th>\n",
-       "      <td>Crewmate</td>\n",
-       "      <td>0</td>\n",
-       "      <td>7.0</td>\n",
-       "      <td>1</td>\n",
-       "      <td>1</td>\n",
-       "      <td>982.0</td>\n",
-       "      <td>0</td>\n",
-       "      <td>0.0</td>\n",
-       "      <td>1</td>\n",
-       "    </tr>\n",
-       "  </tbody>\n",
-       "</table>\n",
-       "</div>"
-      ],
-      "text/plain": [
-       "       Team  Outcome  Task Completed  All Tasks Completed  Murdered  \\\n",
-       "0  Crewmate        1             3.0                    0         1   \n",
-       "1  Crewmate        0             7.0                    1         0   \n",
-       "2  Crewmate        1             3.0                    0         0   \n",
-       "4  Crewmate        0             4.0                    0         0   \n",
-       "5  Crewmate        0             7.0                    1         1   \n",
-       "\n",
-       "   Game Length  Ejected  Sabotages Fixed  Region  \n",
-       "0        424.0        0              2.0       0  \n",
-       "1        981.0        0              1.0       0  \n",
-       "2        693.0        0              0.0       0  \n",
-       "4        310.0        0              0.0       1  \n",
-       "5        982.0        0              0.0       1  "
-      ]
-     },
-     "execution_count": 9,
-     "metadata": {},
-     "output_type": "execute_result"
-    }
-   ],
-   "source": [
-    "crewmate['Outcome'] = crewmate['Outcome'].replace(['Loss', 'Win'],[0, 1])\n",
-    "crewmate['All Tasks Completed'] = crewmate['All Tasks Completed'].replace(['No', 'Yes'],[0, 1])\n",
-    "crewmate['Murdered'] = crewmate['Murdered'].replace(['No', 'Yes'],[0, 1])\n",
-    "crewmate['Ejected'] = crewmate['Ejected'].replace(['No', 'Yes'],[0, 1])\n",
-    "crewmate['Region'] = crewmate['Region'].replace(['NA ', 'Europe '],[0, 1])\n",
-    "crewmate.head()"
-   ]
-  },
-  {
-   "cell_type": "code",
-   "execution_count": 10,
-   "id": "d46ac201",
-   "metadata": {},
-   "outputs": [],
-   "source": [
-    "target = crewmate['Outcome']\n",
-    "crewmate = crewmate.drop('Outcome', axis =1)"
-   ]
-  },
-  {
-   "cell_type": "code",
-   "execution_count": 11,
-   "id": "cce623f6",
-   "metadata": {},
-   "outputs": [],
-   "source": [
-    "def user_train_valid_split(users_i,valid_ratio,possible=False):\n",
-    "    n = sum(users_i)\n",
-    "    target = int(np.floor(valid_ratio*n))\n",
-    "    while not possible:\n",
-    "        possible, validation_indices = user_split(users_i,target)\n",
-    "        target -= 1\n",
-    "    training_indices = [i for i in range(len(users_i)) if i not in validation_indices]\n",
-    "    return training_indices, validation_indices\n",
-    "\n",
-    "def user_split(users_i,target):  \n",
-    "    # Reference: https://levelup.gitconnected.com/dynamic-programming-subset-sum-c386126621cd\n",
-    "    n = len(users_i)\n",
-    "    solution = [[False for j in range(int(target+1))] for i in range(n+1)]\n",
-    "    # base cases\n",
-    "    for i in range(n):\n",
-    "        solution[i][0] = True\n",
-    "    # other cases\n",
-    "    for i in range(1,n+1):\n",
-    "        for j in range(1,target+1):\n",
-    "            solution[i][j] = solution[i-1][j]\n",
-    "            if(solution[i][j] == False and j >= users_i[i-1]):\n",
-    "                solution[i][j] = solution[i][j] or solution[i-1][j-users_i[i-1]]\n",
-    "    # check if the subset sum is possible\n",
-    "    possible = solution[len(users_i)][target]\n",
-    "    subset = []\n",
-    "    if not possible: return possible, subset\n",
-    "    # return the subset solution if one exists\n",
-    "    y = len(users_i)\n",
-    "    x = target\n",
-    "    while x != 0:\n",
-    "        if solution[y-1][x] == False:\n",
-    "            subset.append(y-1)\n",
-    "            x -= users_i[y-1]\n",
-    "        else: \n",
-    "            y -= 1\n",
-    "    return possible, subset"
-   ]
-  },
-  {
-   "cell_type": "code",
-   "execution_count": 12,
-   "id": "b4765783",
-   "metadata": {},
-   "outputs": [
-    {
-     "ename": "NameError",
-     "evalue": "name 'users_i' is not defined",
-     "output_type": "error",
-     "traceback": [
-      "\u001b[1;31m---------------------------------------------------------------------------\u001b[0m",
-      "\u001b[1;31mNameError\u001b[0m                                 Traceback (most recent call last)",
-      "\u001b[1;32m~\\AppData\\Local\\Temp/ipykernel_11152/1415368763.py\u001b[0m in \u001b[0;36m<module>\u001b[1;34m\u001b[0m\n\u001b[0;32m      1\u001b[0m \u001b[0mvalid_ratio\u001b[0m \u001b[1;33m=\u001b[0m \u001b[1;36m.25\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[1;32m----> 2\u001b[1;33m \u001b[0mtraining_indices\u001b[0m\u001b[1;33m,\u001b[0m \u001b[0mvalidation_indices\u001b[0m \u001b[1;33m=\u001b[0m \u001b[0muser_train_valid_split\u001b[0m\u001b[1;33m(\u001b[0m\u001b[0musers_i\u001b[0m\u001b[1;33m,\u001b[0m\u001b[0mvalid_ratio\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0m\u001b[0;32m      3\u001b[0m \u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0;32m      4\u001b[0m \u001b[0mprint\u001b[0m\u001b[1;33m(\u001b[0m\u001b[0musers_i\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0;32m      5\u001b[0m \u001b[0mprint\u001b[0m\u001b[1;33m(\u001b[0m\u001b[0mtraining_indices\u001b[0m\u001b[1;33m,\u001b[0m \u001b[0mvalidation_indices\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n",
-      "\u001b[1;31mNameError\u001b[0m: name 'users_i' is not defined"
-     ]
-    }
-   ],
-   "source": [
-    "valid_ratio = .25\n",
-    "training_indices, validation_indices = user_train_valid_split(users_i,valid_ratio)\n",
-    "\n",
-    "print(users_i)\n",
-    "print(training_indices, validation_indices)\n",
-    "train_sum = sum([users_i[i] for i in training_indices])\n",
-    "valid_sum = sum([users_i[i] for i in validation_indices])\n",
-    "print(valid_sum/(train_sum+valid_sum))"
-   ]
-  },
-  {
-   "cell_type": "code",
-   "execution_count": null,
-   "id": "14b9197c",
-   "metadata": {},
-   "outputs": [],
-   "source": []
-  }
- ],
- "metadata": {
-  "kernelspec": {
-   "display_name": "Python 3 (ipykernel)",
-   "language": "python",
-   "name": "python3"
-  },
-  "language_info": {
-   "codemirror_mode": {
-    "name": "ipython",
-    "version": 3
-   },
-   "file_extension": ".py",
-   "mimetype": "text/x-python",
-   "name": "python",
-   "nbconvert_exporter": "python",
-   "pygments_lexer": "ipython3",
-   "version": "3.9.7"
-  }
- },
- "nbformat": 4,
- "nbformat_minor": 5
-}