From decd34a927881d70cb7a81af5a1e29a800447b68 Mon Sep 17 00:00:00 2001
From: Eljas Roellin <eljas.roellin@ikmail.com>
Date: Mon, 12 Feb 2024 18:36:08 +0100
Subject: [PATCH 01/46] nb with stuff that at least does not just fail

---
 try_flowchart.ipynb | 1592 +++++++++++++++++++++++++++++++++++++++++++
 1 file changed, 1592 insertions(+)
 create mode 100644 try_flowchart.ipynb

diff --git a/try_flowchart.ipynb b/try_flowchart.ipynb
new file mode 100644
index 00000000..61047371
--- /dev/null
+++ b/try_flowchart.ipynb
@@ -0,0 +1,1592 @@
+{
+ "cells": [
+  {
+   "cell_type": "code",
+   "execution_count": 261,
+   "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>ethnicity</th>\n",
+       "      <th>gender</th>\n",
+       "      <th>proficiency</th>\n",
+       "    </tr>\n",
+       "  </thead>\n",
+       "  <tbody>\n",
+       "    <tr>\n",
+       "      <th>0</th>\n",
+       "      <td>asian</td>\n",
+       "      <td>female</td>\n",
+       "      <td>high prof.</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>1</th>\n",
+       "      <td>white</td>\n",
+       "      <td>male</td>\n",
+       "      <td>limited prof.</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>2</th>\n",
+       "      <td>black</td>\n",
+       "      <td>female</td>\n",
+       "      <td>high prof.</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>3</th>\n",
+       "      <td>black</td>\n",
+       "      <td>male</td>\n",
+       "      <td>limited prof.</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>4</th>\n",
+       "      <td>black</td>\n",
+       "      <td>male</td>\n",
+       "      <td>limited prof.</td>\n",
+       "    </tr>\n",
+       "  </tbody>\n",
+       "</table>\n",
+       "</div>"
+      ],
+      "text/plain": [
+       "  ethnicity  gender    proficiency\n",
+       "0     asian  female     high prof.\n",
+       "1     white    male  limited prof.\n",
+       "2     black  female     high prof.\n",
+       "3     black    male  limited prof.\n",
+       "4     black    male  limited prof."
+      ]
+     },
+     "execution_count": 261,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "import pandas as pd\n",
+    "import matplotlib.pyplot as plt\n",
+    "import numpy as np\n",
+    "import ehrapy as ep\n",
+    "from tableone import TableOne\n",
+    "import seaborn as sns\n",
+    "\n",
+    "# generate dataset\n",
+    "rng = np.random.default_rng(42)\n",
+    "\n",
+    "random_dataset = pd.DataFrame({\n",
+    "    \"ethnicity\": rng.choice([\"asian\", \"black\", \"white\"], size=150),\n",
+    "    \"gender\": rng.choice([\"male\", \"female\"], size=150),\n",
+    "    \"proficiency\": rng.choice([\"high prof.\", \"limited prof.\"], size=150)\n",
+    "    })\n",
+    "random_dataset.head()"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 122,
+   "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></th>\n",
+       "      <th>Missing</th>\n",
+       "      <th>Overall</th>\n",
+       "    </tr>\n",
+       "  </thead>\n",
+       "  <tbody>\n",
+       "    <tr>\n",
+       "      <th>n</th>\n",
+       "      <th></th>\n",
+       "      <td></td>\n",
+       "      <td>150</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th rowspan=\"3\" valign=\"top\">ethnicity, n (%)</th>\n",
+       "      <th>asian</th>\n",
+       "      <td>0</td>\n",
+       "      <td>43 (28.7)</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>black</th>\n",
+       "      <td></td>\n",
+       "      <td>52 (34.7)</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>white</th>\n",
+       "      <td></td>\n",
+       "      <td>55 (36.7)</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th rowspan=\"2\" valign=\"top\">gender, n (%)</th>\n",
+       "      <th>female</th>\n",
+       "      <td>0</td>\n",
+       "      <td>66 (44.0)</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>male</th>\n",
+       "      <td></td>\n",
+       "      <td>84 (56.0)</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th rowspan=\"2\" valign=\"top\">proficiency, n (%)</th>\n",
+       "      <th>high prof.</th>\n",
+       "      <td>0</td>\n",
+       "      <td>78 (52.0)</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>limited prof.</th>\n",
+       "      <td></td>\n",
+       "      <td>72 (48.0)</td>\n",
+       "    </tr>\n",
+       "  </tbody>\n",
+       "</table>\n",
+       "</div><br />"
+      ],
+      "text/plain": [
+       "                                 Missing    Overall\n",
+       "n                                               150\n",
+       "ethnicity, n (%)   asian               0  43 (28.7)\n",
+       "                   black                  52 (34.7)\n",
+       "                   white                  55 (36.7)\n",
+       "gender, n (%)      female              0  66 (44.0)\n",
+       "                   male                   84 (56.0)\n",
+       "proficiency, n (%) high prof.          0  78 (52.0)\n",
+       "                   limited prof.          72 (48.0)"
+      ]
+     },
+     "execution_count": 122,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "t1 = TableOne(random_dataset, columns=[\"ethnicity\", \"gender\", \"proficiency\"])\n",
+    "t1"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 141,
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "text/plain": [
+       "{'asian': [43.0],\n",
+       " 'black': [52.0],\n",
+       " 'white': [55.0],\n",
+       " 'male': [84.0],\n",
+       " 'female': [66.0],\n",
+       " 'high prof.': [78.0],\n",
+       " 'limited prof.': [72.0]}"
+      ]
+     },
+     "execution_count": 141,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "def get_dicts(table_one):\n",
+    "\n",
+    "    cols = [\"ethnicity\", \"gender\", \"proficiency\"]\n",
+    "    col_cats = {\"ethnicity\": [\"asian\", \"black\", \"white\"],\n",
+    "                \"gender\": [\"male\", \"female\"],\n",
+    "                \"proficiency\": [\"high prof.\", \"limited prof.\"]}\n",
+    "\n",
+    "    pd.DataFrame(columns=cols)\n",
+    "    dvals = {}\n",
+    "    dpcts = {}\n",
+    "    for col in cols:\n",
+    "        for cat in col_cats[col]:\n",
+    "\n",
+    "            val = float(table_one.cat_table[\"Overall\"].loc[(col, cat)].split(\" \")[0])\n",
+    "            pct = float(table_one.cat_table[\"Overall\"].loc[(col, cat)].split(\"(\")[1].split(\")\")[0])\n",
+    "            if cat not in dvals.keys():\n",
+    "                dvals[cat] = [val]\n",
+    "            else:\n",
+    "                dvals[cat].append(val)\n",
+    "\n",
+    "            if cat not in dpcts.keys():\n",
+    "                dpcts[cat] = [pct]\n",
+    "            else:\n",
+    "                dpcts[cat].append(pct)\n",
+    "\n",
+    "    return dvals, dpcts\n",
+    "\n",
+    "dvals, dcpts = get_dicts(t1)\n",
+    "dvals"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 142,
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "text/plain": [
+       "{'asian': [28.7],\n",
+       " 'black': [34.7],\n",
+       " 'white': [36.7],\n",
+       " 'male': [56.0],\n",
+       " 'female': [44.0],\n",
+       " 'high prof.': [52.0],\n",
+       " 'limited prof.': [48.0]}"
+      ]
+     },
+     "execution_count": 142,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "dpcts"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 143,
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "image/png": "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",
+      "text/latex": [
+       "$\\displaystyle \\left( 50, \\  3\\right)$"
+      ],
+      "text/plain": [
+       "(50, 3)"
+      ]
+     },
+     "execution_count": 143,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "random_dataset_2 = random_dataset[:50]\n",
+    "random_dataset_2.shape"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 144,
+   "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></th>\n",
+       "      <th>Missing</th>\n",
+       "      <th>Overall</th>\n",
+       "    </tr>\n",
+       "  </thead>\n",
+       "  <tbody>\n",
+       "    <tr>\n",
+       "      <th>n</th>\n",
+       "      <th></th>\n",
+       "      <td></td>\n",
+       "      <td>50</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th rowspan=\"3\" valign=\"top\">ethnicity, n (%)</th>\n",
+       "      <th>asian</th>\n",
+       "      <td>0</td>\n",
+       "      <td>12 (24.0)</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>black</th>\n",
+       "      <td></td>\n",
+       "      <td>17 (34.0)</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>white</th>\n",
+       "      <td></td>\n",
+       "      <td>21 (42.0)</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th rowspan=\"2\" valign=\"top\">gender, n (%)</th>\n",
+       "      <th>female</th>\n",
+       "      <td>0</td>\n",
+       "      <td>22 (44.0)</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>male</th>\n",
+       "      <td></td>\n",
+       "      <td>28 (56.0)</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th rowspan=\"2\" valign=\"top\">proficiency, n (%)</th>\n",
+       "      <th>high prof.</th>\n",
+       "      <td>0</td>\n",
+       "      <td>24 (48.0)</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>limited prof.</th>\n",
+       "      <td></td>\n",
+       "      <td>26 (52.0)</td>\n",
+       "    </tr>\n",
+       "  </tbody>\n",
+       "</table>\n",
+       "</div><br />"
+      ],
+      "text/plain": [
+       "                                 Missing    Overall\n",
+       "n                                                50\n",
+       "ethnicity, n (%)   asian               0  12 (24.0)\n",
+       "                   black                  17 (34.0)\n",
+       "                   white                  21 (42.0)\n",
+       "gender, n (%)      female              0  22 (44.0)\n",
+       "                   male                   28 (56.0)\n",
+       "proficiency, n (%) high prof.          0  24 (48.0)\n",
+       "                   limited prof.          26 (52.0)"
+      ]
+     },
+     "execution_count": 144,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "t2 = TableOne(random_dataset_2)\n",
+    "t2"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 146,
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "text/plain": [
+       "{'asian': [12.0],\n",
+       " 'black': [17.0],\n",
+       " 'white': [21.0],\n",
+       " 'male': [28.0],\n",
+       " 'female': [22.0],\n",
+       " 'high prof.': [24.0],\n",
+       " 'limited prof.': [26.0]}"
+      ]
+     },
+     "execution_count": 146,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "dvals2, dcpts2 = get_dicts(t2)\n",
+    "dvals2"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 147,
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "text/plain": [
+       "{'asian': [24.0],\n",
+       " 'black': [34.0],\n",
+       " 'white': [42.0],\n",
+       " 'male': [56.0],\n",
+       " 'female': [44.0],\n",
+       " 'high prof.': [48.0],\n",
+       " 'limited prof.': [52.0]}"
+      ]
+     },
+     "execution_count": 147,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "dcpts2"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 151,
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "text/plain": [
+       "{'asian': [28.7, 24.0],\n",
+       " 'black': [34.7, 34.0],\n",
+       " 'white': [36.7, 42.0],\n",
+       " 'male': [56.0, 56.0],\n",
+       " 'female': [44.0, 44.0],\n",
+       " 'high prof.': [52.0, 48.0],\n",
+       " 'limited prof.': [48.0, 52.0]}"
+      ]
+     },
+     "execution_count": 151,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "collated_data = {key: values1 + values2 for key, values1, values2 in zip(dpcts.keys(), dpcts.values(), dcpts2.values())}\n",
+    "collated_data"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 155,
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "text/plain": [
+       "{'asian': [28.7, 24.0], 'black': [34.7, 34.0], 'white': [36.7, 42.0]}"
+      ]
+     },
+     "execution_count": 155,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "data_ethnicity = {key: value for key, value in collated_data.items() if key in [\"asian\", \"black\", \"white\"]}\n",
+    "data_ethnicity"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 191,
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "text/plain": [
+       "{'male': [56.0, 56.0], 'female': [44.0, 44.0]}"
+      ]
+     },
+     "execution_count": 191,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "data_gender = {key: value for key, value in collated_data.items() if key in [\"male\", \"female\"]}\n",
+    "data_gender"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 248,
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "text/plain": [
+       "{'high prof.': [52.0, 48.0], 'limited prof.': [48.0, 52.0]}"
+      ]
+     },
+     "execution_count": 248,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "data_prof = {key: value for key, value in collated_data.items() if key in [\"high prof.\", \"limited prof.\"]}\n",
+    "data_prof"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 192,
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "text/plain": [
+       "{'male': [56.0], 'female': [44.0]}"
+      ]
+     },
+     "execution_count": 192,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "{'male': [56.0], 'female': [44.0]}"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 171,
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "image/png": "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",
+      "text/plain": [
+       "<Figure size 1000x500 with 2 Axes>"
+      ]
+     },
+     "metadata": {
+      "image/png": {
+       "height": 488,
+       "width": 989
+      }
+     },
+     "output_type": "display_data"
+    }
+   ],
+   "source": [
+    "# Convert dictionary to DataFrame\n",
+    "df = pd.DataFrame(data_ethnicity)\n",
+    "\n",
+    "# Transpose DataFrame\n",
+    "df= df.T\n",
+    "\n",
+    "# Plotting\n",
+    "fig, axes = plt.subplots(1, 2, figsize=(10, 5))\n",
+    "\n",
+    "# First plot for the first numbers\n",
+    "df.iloc[:,0].plot(kind='barh', stacked=True, ax=axes[0], color=['skyblue', 'lightcoral', 'lightgreen'])\n",
+    "axes[0].set_title('First Numbers')\n",
+    "axes[0].set_xlabel('Percentage')\n",
+    "axes[0].set_ylabel('Ethnicity')\n",
+    "\n",
+    "# Second plot for the second numbers\n",
+    "df.iloc[:,0].plot(kind='barh', stacked=True, ax=axes[1], color=['skyblue', 'lightcoral', 'lightgreen'])\n",
+    "axes[1].set_title('Second Numbers')\n",
+    "axes[1].set_xlabel('Percentage')\n",
+    "axes[1].set_ylabel('Ethnicity')\n",
+    "\n",
+    "plt.tight_layout()\n",
+    "plt.show()"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 180,
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "text/plain": [
+       "Index(['asian', 'black', 'white'], dtype='object')"
+      ]
+     },
+     "execution_count": 180,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "df.index"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 189,
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "image/png": "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",
+      "text/plain": [
+       "<Figure size 1000x500 with 1 Axes>"
+      ]
+     },
+     "metadata": {
+      "image/png": {
+       "height": 488,
+       "width": 990
+      }
+     },
+     "output_type": "display_data"
+    }
+   ],
+   "source": [
+    "import pandas as pd\n",
+    "import matplotlib.pyplot as plt\n",
+    "\n",
+    "# Given dictionary\n",
+    "data = {'asian': [28.7],\n",
+    "        'black': [34.7],\n",
+    "        'white': [36.7]}\n",
+    "\n",
+    "# Convert dictionary to DataFrame\n",
+    "df = pd.DataFrame(data)\n",
+    "\n",
+    "# Plotting\n",
+    "fig, axes = plt.subplots(figsize=(10, 5))\n",
+    "\n",
+    "# Transpose DataFrame\n",
+    "#df = df.T\n",
+    "\n",
+    "# Plotting for the first subplot\n",
+    "df.plot(kind='barh', stacked=True, ax=axes, color=['skyblue', 'lightcoral', 'lightgreen'])\n",
+    "axes.set_title('First Numbers')\n",
+    "axes.set_xlabel('Percentage')\n",
+    "axes.set_ylabel('Ethnicity')\n",
+    "\n",
+    "# # Plotting for the second subplot\n",
+    "# df.plot(kind='barh', stacked=True, ax=axes[1], color=['skyblue', 'lightcoral', 'lightgreen'])\n",
+    "# axes[1].set_title('Second Numbers')\n",
+    "# axes[1].set_xlabel('Percentage')\n",
+    "# axes[1].set_ylabel('Ethnicity')\n",
+    "\n",
+    "plt.tight_layout()\n",
+    "plt.show()\n"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 251,
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "image/png": "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",
+      "text/plain": [
+       "<Figure size 1000x500 with 1 Axes>"
+      ]
+     },
+     "metadata": {
+      "image/png": {
+       "height": 488,
+       "width": 990
+      }
+     },
+     "output_type": "display_data"
+    }
+   ],
+   "source": [
+    "import pandas as pd\n",
+    "import matplotlib.pyplot as plt\n",
+    "\n",
+    "# Given data dictionaries\n",
+    "data_ethnicity = {'asian': [28.7],\n",
+    "                  'black': [34.7],\n",
+    "                  'white': [36.7]}\n",
+    "\n",
+    "data_gender = {'male': [56.0],\n",
+    "               'female': [44.0]}\n",
+    "\n",
+    "data_prof = {'high prof.': [52.0], 'limited prof.': [48.0]}\n",
+    "\n",
+    "# Convert dictionaries to DataFrames\n",
+    "df_ethnicity = pd.DataFrame(data_ethnicity)\n",
+    "df_gender = pd.DataFrame(data_gender)\n",
+    "\n",
+    "# Plotting\n",
+    "fig, ax = plt.subplots(figsize=(10, 5))\n",
+    "\n",
+    "# Plotting for the ethnicity\n",
+    "df_ethnicity.plot(kind='barh', stacked=True, ax=ax, color=['skyblue', 'lightcoral', 'lightgreen'], position=0.1, width=0.2)\n",
+    "\n",
+    "# Plotting for the gender with a small gap\n",
+    "df_gender.plot(kind='barh', stacked=True, ax=ax, color=['skyblue', 'lightcoral'], position=1.5, width=0.2)\n",
+    "\n",
+    "# Set labels and title\n",
+    "ax.set_title('Ethnicity and Gender')\n",
+    "ax.set_xlabel('Percentage')\n",
+    "ax.set_ylabel('Category')\n",
+    "\n",
+    "# Add a legend\n",
+    "ax.legend(['Asian', 'Black', 'White', 'Male', 'Female'])\n",
+    "\n",
+    "plt.tight_layout()\n",
+    "plt.show()\n"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 211,
+   "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>asian</th>\n",
+       "      <th>black</th>\n",
+       "      <th>white</th>\n",
+       "    </tr>\n",
+       "  </thead>\n",
+       "  <tbody>\n",
+       "    <tr>\n",
+       "      <th>0</th>\n",
+       "      <td>28.7</td>\n",
+       "      <td>34.7</td>\n",
+       "      <td>36.7</td>\n",
+       "    </tr>\n",
+       "  </tbody>\n",
+       "</table>\n",
+       "</div>"
+      ],
+      "text/plain": [
+       "   asian  black  white\n",
+       "0   28.7   34.7   36.7"
+      ]
+     },
+     "execution_count": 211,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "df_ethnicity"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 215,
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "image/png": "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",
+      "text/plain": [
+       "<Figure size 1000x500 with 1 Axes>"
+      ]
+     },
+     "metadata": {
+      "image/png": {
+       "height": 488,
+       "width": 990
+      }
+     },
+     "output_type": "display_data"
+    }
+   ],
+   "source": [
+    "import pandas as pd\n",
+    "import matplotlib.pyplot as plt\n",
+    "\n",
+    "# Given data dictionaries\n",
+    "data_ethnicity = {'asian': [28.7],\n",
+    "                  'black': [34.7],\n",
+    "                  'white': [36.7]}\n",
+    "\n",
+    "data_gender = {'male': [56.0],\n",
+    "               'female': [44.0]}\n",
+    "\n",
+    "# Convert dictionaries to DataFrames\n",
+    "df_ethnicity = pd.DataFrame(data_ethnicity)\n",
+    "df_gender = pd.DataFrame(data_gender)\n",
+    "\n",
+    "# Plotting\n",
+    "fig, ax = plt.subplots(figsize=(10, 5))\n",
+    "\n",
+    "# Plotting for the ethnicity with a small gap\n",
+    "df_ethnicity.plot(kind='barh', stacked=True, ax=ax, color=['skyblue', 'lightcoral', 'lightgreen'], position=0.67, width=0.3)\n",
+    "\n",
+    "# Plotting for the gender with a small gap\n",
+    "df_gender.plot(kind='barh', stacked=True, ax=ax, color=['skyblue', 'lightcoral'], position=0.33, width=0.3)\n",
+    "\n",
+    "\n",
+    "# Set labels and title\n",
+    "ax.set_title('Ethnicity and Gender')\n",
+    "ax.set_xlabel('Percentage')\n",
+    "ax.set_ylabel('Category')\n",
+    "\n",
+    "# Add a legend\n",
+    "ax.legend(['Asian', 'Black', 'White', 'Male', 'Female'])\n",
+    "\n",
+    "plt.tight_layout()\n",
+    "plt.show()\n"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 235,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "l = df_ethnicity.T"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 242,
+   "metadata": {},
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "28.7\n",
+      "34.7\n",
+      "36.7\n"
+     ]
+    }
+   ],
+   "source": [
+    "for v in l.iterrows():\n",
+    "    print(v[1].values[0])"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 232,
+   "metadata": {},
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "[28.7 34.7 36.7]\n"
+     ]
+    }
+   ],
+   "source": [
+    "for u in df_ethnicity.values:\n",
+    "    print(u)"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 233,
+   "metadata": {},
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "0\n"
+     ]
+    }
+   ],
+   "source": [
+    "for i, value in enumerate(df_ethnicity.T):\n",
+    "    print(value)"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 271,
+   "metadata": {},
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "28.7\n",
+      "34.7\n",
+      "36.7\n",
+      "56.0\n",
+      "44.0\n",
+      "52.0\n",
+      "48.0\n"
+     ]
+    },
+    {
+     "data": {
+      "image/png": "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",
+      "text/plain": [
+       "<Figure size 1000x500 with 1 Axes>"
+      ]
+     },
+     "metadata": {
+      "image/png": {
+       "height": 495,
+       "width": 970
+      }
+     },
+     "output_type": "display_data"
+    }
+   ],
+   "source": [
+    "import pandas as pd\n",
+    "import matplotlib.pyplot as plt\n",
+    "\n",
+    "# Given data dictionaries\n",
+    "data_ethnicity = {'asian': [28.7],\n",
+    "                  'black': [34.7],\n",
+    "                  'white': [36.7]}\n",
+    "\n",
+    "data_gender = {'male': [56.0],\n",
+    "               'female': [44.0]}\n",
+    "\n",
+    "data_prof = {'high prof.': [52.0], 'limited prof.': [48.0]}\n",
+    "\n",
+    "\n",
+    "# Convert dictionaries to DataFrames\n",
+    "df_ethnicity = pd.DataFrame(data_ethnicity)\n",
+    "df_gender = pd.DataFrame(data_gender)\n",
+    "df_prof = pd.DataFrame(data_prof)\n",
+    "\n",
+    "# Plotting\n",
+    "fig, ax = plt.subplots(figsize=(10, 5))\n",
+    "\n",
+    "# Plotting for the ethnicity with a small gap\n",
+    "#df_ethnicity.plot(kind='barh', stacked=True, ax=ax, color=['skyblue', 'lightcoral', 'lightgreen'], position=0.67, width=0.3)\n",
+    "categories = [df_ethnicity, df_gender, df_prof]\n",
+    "category_labels = ['Ethnicity', 'Gender', 'Proficiency']\n",
+    "\n",
+    "colors=['skyblue', 'lightcoral', 'lightgreen'] # TODO: make this dynamic, more colors, pallette per category\n",
+    "legend_labels = []\n",
+    "\n",
+    "for pos, data in enumerate([df_ethnicity.T.iterrows(), df_gender.T.iterrows(), df_prof.T.iterrows()]):\n",
+    "    cumwidth = 0\n",
+    "    category = categories[pos]\n",
+    "\n",
+    "    hue_shift = pos / len(categories)  # Adjust the hue shift based on the category position\n",
+    "    colors = sns.color_palette(\"husl\", len(category.columns))\n",
+    "    adjusted_colors = [((color[0] + hue_shift) % 1, color[1], color[2]) for color in colors]\n",
+    "\n",
+    "    for i, value in enumerate(data):\n",
+    "        value = value[1].values[0]\n",
+    "        print(value)\n",
+    "        ax.barh(pos, value, left=cumwidth, color=adjusted_colors[i], height=0.8)\n",
+    "        \n",
+    "\n",
+    "        if value > 5:\n",
+    "            # Add data labels\n",
+    "            width = value\n",
+    "            ax.text(cumwidth + width/2, pos, '{:.1f}'.format(value), ha='center', va='center', color='white', fontweight='bold')\n",
+    "            \n",
+    "        ax.set_yticks([])\n",
+    "        ax.set_xticks([])\n",
+    "        cumwidth += value\n",
+    "        \n",
+    "        #if pos == 0:\n",
+    "        legend_labels.append(category.columns[i])\n",
+    "        \n",
+    "# Plotting for the gender with a small gap\n",
+    "#df_gender.plot(kind='barh', stacked=True, ax=ax, color=['skyblue', 'lightcoral'], position=0.33, width=0.3)\n",
+    "\n",
+    "# ax.barh(1, df_gender, color=['skyblue', 'lightcoral', 'lightgreen'])\n",
+    "\n",
+    "# # Set labels and title\n",
+    "# ax.set_title('Ethnicity and Gender')\n",
+    "# ax.set_xlabel('Percentage')\n",
+    "# ax.set_ylabel('Category')\n",
+    "\n",
+    "# # Add a legend\n",
+    "# ax.legend(['Asian', 'Black', 'White', 'Male', 'Female'])\n",
+    "            \n",
+    "ax.legend(legend_labels, loc='lower center', ncol=len(legend_labels), bbox_to_anchor=(0.5, -0.1))\n",
+    "\n",
+    "plt.tight_layout()\n",
+    "plt.show()\n"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 9,
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "image/png": "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",
+      "text/plain": [
+       "<Figure size 400x900 with 6 Axes>"
+      ]
+     },
+     "metadata": {
+      "image/png": {
+       "height": 988,
+       "width": 390
+      }
+     },
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "text/plain": [
+       "<Figure size 640x480 with 0 Axes>"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "text/plain": [
+       "<Figure size 640x480 with 0 Axes>"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "text/plain": [
+       "<Figure size 640x480 with 0 Axes>"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "text/plain": [
+       "<Figure size 640x480 with 0 Axes>"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "text/plain": [
+       "<Figure size 640x480 with 0 Axes>"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "text/plain": [
+       "<Figure size 640x480 with 0 Axes>"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "text/plain": [
+       "<Figure size 640x480 with 0 Axes>"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "text/plain": [
+       "<Figure size 640x480 with 0 Axes>"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "text/plain": [
+       "<Figure size 640x480 with 0 Axes>"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "text/plain": [
+       "<Figure size 640x480 with 0 Axes>"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "text/plain": [
+       "<Figure size 640x480 with 0 Axes>"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "text/plain": [
+       "<Figure size 640x480 with 0 Axes>"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "text/plain": [
+       "<Figure size 640x480 with 0 Axes>"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "text/plain": [
+       "<Figure size 640x480 with 0 Axes>"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "text/plain": [
+       "<Figure size 640x480 with 0 Axes>"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "text/plain": [
+       "<Figure size 640x480 with 0 Axes>"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "text/plain": [
+       "<Figure size 640x480 with 0 Axes>"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "text/plain": [
+       "<Figure size 640x480 with 0 Axes>"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "text/plain": [
+       "<Figure size 640x480 with 0 Axes>"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "text/plain": [
+       "<Figure size 640x480 with 0 Axes>"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "text/plain": [
+       "<Figure size 640x480 with 0 Axes>"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "text/plain": [
+       "<Figure size 640x480 with 0 Axes>"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "text/plain": [
+       "<Figure size 640x480 with 0 Axes>"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "text/plain": [
+       "<Figure size 640x480 with 0 Axes>"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "text/plain": [
+       "<Figure size 640x480 with 0 Axes>"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "text/plain": [
+       "<Figure size 640x480 with 0 Axes>"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "text/plain": [
+       "<Figure size 640x480 with 0 Axes>"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "text/plain": [
+       "<Figure size 640x480 with 0 Axes>"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "text/plain": [
+       "<Figure size 640x480 with 0 Axes>"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "text/plain": [
+       "<Figure size 640x480 with 0 Axes>"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "text/plain": [
+       "<Figure size 640x480 with 0 Axes>"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "text/plain": [
+       "<Figure size 640x480 with 0 Axes>"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "text/plain": [
+       "<Figure size 640x480 with 0 Axes>"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "text/plain": [
+       "<Figure size 640x480 with 0 Axes>"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "text/plain": [
+       "<Figure size 640x480 with 0 Axes>"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "text/plain": [
+       "<Figure size 640x480 with 0 Axes>"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "text/plain": [
+       "<Figure size 640x480 with 0 Axes>"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "text/plain": [
+       "<Figure size 640x480 with 0 Axes>"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "text/plain": [
+       "<Figure size 640x480 with 0 Axes>"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "text/plain": [
+       "<Figure size 640x480 with 0 Axes>"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    }
+   ],
+   "source": [
+    "\n",
+    "colors_racial = ['midnightblue', 'gold', 'saddlebrown', 'coral', 'lightgrey']\n",
+    "colors_gender = ['tab:blue', 'orchid']\n",
+    "colors_proficiency = ['yellowgreen', 'firebrick']\n",
+    "fig, axes = plt.subplots(6, 1, figsize=(4, 9))\n",
+    "labels = ['Race and Ethnicity', 'Sex', 'English Proficiency']\n",
+    "for i, ax in enumerate(axes):\n",
+    "    # Extract rows\n",
+    "    data_racial = df_racial.iloc[i]\n",
+    "    data_gender = df_gender.iloc[i]\n",
+    "    data_proficiency = df_proficiency.iloc[i]\n",
+    "    \n",
+    "    # For vertical positioning within each lane\n",
+    "    positions = [2, 1, 0]\n",
+    "    for j, (data, colors) in enumerate(zip( [data_racial, data_gender, data_proficiency], [colors_racial, colors_gender, colors_proficiency] )):\n",
+    "        position = positions[j]\n",
+    "        cum_width = 0\n",
+    "        for value, color in zip(data, colors):\n",
+    "            ax.barh(position, value, left=cum_width, color=color, height=.8)\n",
+    "            if value > 5:\n",
+    "                # Add data labels\n",
+    "                width = value\n",
+    "                ax.text(cum_width + width/2, position, '{:.1f}'.format(value), ha='center', va='center', color='white', fontweight='bold')\n",
+    "                cum_width += value\n",
+    "                ax.set_xlim(0, 100)\n",
+    "                ax.set_ylim(-1, 3)\n",
+    "                ax.set_yticks([])\n",
+    "                ax.set_xticks([])\n",
+    "                \n",
+    "                for spine in ax.spines.values():\n",
+    "                    spine.set_visible(False)\n",
+    "                    ax.set_xlabel('Group Proportion (%)')\n",
+    "                    # Add legend to the bottom\n",
+    "                    racial_patches = [plt.Rectangle((0,0),1,1, color=color) for color in colors_racial]\n",
+    "                    gender_patches = [plt.Rectangle((0,0),1,1, color=color) for color in colors_gender]\n",
+    "                    proficiency_patches = [plt.Rectangle((0,0),1,1, color=color) for color in colors_proficiency]\n",
+    "                    racial_labels = ['Black', 'Asian', 'Hispanic', 'Other', 'White']\n",
+    "                    gender_labels = ['Male', 'Female']\n",
+    "                    proficiency_labels = ['Proficient', 'Limited Prof.']\n",
+    "                    legend_handles = racial_patches + gender_patches + proficiency_patches\n",
+    "                    legend_labels = racial_labels + gender_labels + proficiency_labels\n",
+    "                    \n",
+    "                    fig.legend(legend_handles, legend_labels, loc='lower center', ncol=3, bbox_to_anchor=(0.5, -0.1))\n",
+    "                \n",
+    "                plt.tight_layout()\n",
+    "                plt.show() "
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 4,
+   "metadata": {},
+   "outputs": [
+    {
+     "ename": "IndexError",
+     "evalue": "list index out of range",
+     "output_type": "error",
+     "traceback": [
+      "\u001b[0;31m---------------------------------------------------------------------------\u001b[0m",
+      "\u001b[0;31mIndexError\u001b[0m                                Traceback (most recent call last)",
+      "Cell \u001b[0;32mIn[4], line 38\u001b[0m\n\u001b[1;32m     36\u001b[0m positions \u001b[38;5;241m=\u001b[39m [\u001b[38;5;241m2\u001b[39m, \u001b[38;5;241m1\u001b[39m, \u001b[38;5;241m0\u001b[39m]\n\u001b[1;32m     37\u001b[0m \u001b[38;5;28;01mfor\u001b[39;00m j, (index, row) \u001b[38;5;129;01min\u001b[39;00m \u001b[38;5;28menumerate\u001b[39m(data\u001b[38;5;241m.\u001b[39miterrows()):\n\u001b[0;32m---> 38\u001b[0m     position \u001b[38;5;241m=\u001b[39m \u001b[43mpositions\u001b[49m\u001b[43m[\u001b[49m\u001b[43mj\u001b[49m\u001b[43m]\u001b[49m\n\u001b[1;32m     39\u001b[0m     cum_width \u001b[38;5;241m=\u001b[39m \u001b[38;5;241m0\u001b[39m\n\u001b[1;32m     40\u001b[0m     \u001b[38;5;28;01mfor\u001b[39;00m value, color \u001b[38;5;129;01min\u001b[39;00m \u001b[38;5;28mzip\u001b[39m(row, colors):\n",
+      "\u001b[0;31mIndexError\u001b[0m: list index out of range"
+     ]
+    },
+    {
+     "data": {
+      "image/png": "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",
+      "text/plain": [
+       "<Figure size 400x900 with 3 Axes>"
+      ]
+     },
+     "metadata": {
+      "image/png": {
+       "height": 734,
+       "width": 330
+      }
+     },
+     "output_type": "display_data"
+    }
+   ],
+   "source": [
+    "import pandas as pd\n",
+    "import matplotlib.pyplot as plt\n",
+    "%config InlineBackend.figure_format = 'retina'\n",
+    "\n",
+    "# Data dictionaries\n",
+    "racial_data = {'Black': [10.8, 10.8, 10.8, 10, 10, 11.8], 'Asian': [2.9, 2.9, 2.9, 2.8, 2.8, 3.3],\n",
+    "               'Hispanic': [2.9, 2.9, 2.9, 2.8, 2.8, 3.3], 'Other': [14.6, 14.6, 14.6, 15.8, 15.8, 0],\n",
+    "               'White': [68.8, 68.8, 68.8, 68.6, 68.6, 81.6]}\n",
+    "\n",
+    "gender_data = {'Male': [55.8, 55.8, 55.9, 58.2, 58.2, 57.6], 'Female': [44.2, 44.2, 44.1, 41.8, 41.8, 42.4]}\n",
+    "\n",
+    "proficiency_data = {'Proficient': [89.8, 89.8, 89.2, 89.4, 89.4, 90.2], 'Limited': [10.2, 10.2, 10.8, 10.6, 10.6, 9.8]}\n",
+    "\n",
+    "# Creating DataFrames\n",
+    "df_racial = pd.DataFrame(racial_data)\n",
+    "df_gender = pd.DataFrame(gender_data)\n",
+    "df_proficiency = pd.DataFrame(proficiency_data)\n",
+    "\n",
+    "colors_racial = ['midnightblue', 'gold', 'saddlebrown', 'coral', 'lightgrey']\n",
+    "colors_gender = ['tab:blue', 'orchid']\n",
+    "colors_proficiency = ['yellowgreen', 'firebrick']\n",
+    "\n",
+    "fig, axes = plt.subplots(3, 1, figsize=(4, 9))\n",
+    "labels = ['Race and Ethnicity', 'Sex', 'English Proficiency']\n",
+    "\n",
+    "for i, ax in enumerate(axes):\n",
+    "    ax.set_title(labels[i])\n",
+    "    ax.set_xlim(0, 100)\n",
+    "    ax.set_ylim(-1, 3)\n",
+    "    ax.set_yticks([])\n",
+    "    ax.set_xticks([])\n",
+    "    for spine in ax.spines.values():\n",
+    "        spine.set_visible(False)\n",
+    "\n",
+    "for ax, data, colors in zip(axes, [df_racial, df_gender, df_proficiency], [colors_racial, colors_gender, colors_proficiency]):\n",
+    "    positions = [2, 1, 0]\n",
+    "    for j, (index, row) in enumerate(data.iterrows()):\n",
+    "        position = positions[j]\n",
+    "        cum_width = 0\n",
+    "        for value, color in zip(row, colors):\n",
+    "            ax.barh(position, value, left=cum_width, color=color, height=0.8)\n",
+    "            if value > 5:\n",
+    "                ax.text(cum_width + value / 2, position, '{:.1f}'.format(value), ha='center', va='center', color='white',\n",
+    "                        fontweight='bold')\n",
+    "            cum_width += value\n",
+    "\n",
+    "# Add legend\n",
+    "racial_patches = [plt.Rectangle((0, 0), 1, 1, color=color) for color in colors_racial]\n",
+    "gender_patches = [plt.Rectangle((0, 0), 1, 1, color=color) for color in colors_gender]\n",
+    "proficiency_patches = [plt.Rectangle((0, 0), 1, 1, color=color) for color in colors_proficiency]\n",
+    "racial_labels = ['Black', 'Asian', 'Hispanic', 'Other', 'White']\n",
+    "gender_labels = ['Male', 'Female']\n",
+    "proficiency_labels = ['Proficient', 'Limited Prof.']\n",
+    "legend_handles = racial_patches + gender_patches + proficiency_patches\n",
+    "legend_labels = racial_labels + gender_labels + proficiency_labels\n",
+    "fig.legend(legend_handles, legend_labels, loc='lower center', ncol=3, bbox_to_anchor=(0.5, -0.1))\n",
+    "\n",
+    "plt.tight_layout()\n",
+    "plt.show()\n"
+   ]
+  }
+ ],
+ "metadata": {
+  "kernelspec": {
+   "display_name": "ehrapy_venv_feb",
+   "language": "python",
+   "name": "python3"
+  },
+  "language_info": {
+   "codemirror_mode": {
+    "name": "ipython",
+    "version": 3
+   },
+   "file_extension": ".py",
+   "mimetype": "text/x-python",
+   "name": "python",
+   "nbconvert_exporter": "python",
+   "pygments_lexer": "ipython3",
+   "version": "3.11.7"
+  }
+ },
+ "nbformat": 4,
+ "nbformat_minor": 2
+}

From 64076233a9120de8a8832401de3ff99f091fee67 Mon Sep 17 00:00:00 2001
From: Eljas Roellin <eljas.roellin@ikmail.com>
Date: Tue, 13 Feb 2024 21:26:30 +0100
Subject: [PATCH 02/46] population logging and tracking first support

---
 ehrapy/tools/__init__.py                      |   1 +
 ehrapy/tools/population_logging/__init__.py   |   0
 .../tools/population_logging/_pop_logger.py   | 193 ++++++++++++++++++
 3 files changed, 194 insertions(+)
 create mode 100644 ehrapy/tools/population_logging/__init__.py
 create mode 100644 ehrapy/tools/population_logging/_pop_logger.py

diff --git a/ehrapy/tools/__init__.py b/ehrapy/tools/__init__.py
index aefb73bf..d89d00fc 100644
--- a/ehrapy/tools/__init__.py
+++ b/ehrapy/tools/__init__.py
@@ -2,6 +2,7 @@
 from ehrapy.tools._scanpy_tl_api import *  # noqa: F403
 from ehrapy.tools.causal._dowhy import causal_inference
 from ehrapy.tools.feature_ranking._rank_features_groups import filter_rank_features_groups, rank_features_groups
+from ehrapy.tools.population_logging._pop_logger import PopulationLogger
 
 try:  # pragma: no cover
     from ehrapy.tools.nlp._medcat import (
diff --git a/ehrapy/tools/population_logging/__init__.py b/ehrapy/tools/population_logging/__init__.py
new file mode 100644
index 00000000..e69de29b
diff --git a/ehrapy/tools/population_logging/_pop_logger.py b/ehrapy/tools/population_logging/_pop_logger.py
new file mode 100644
index 00000000..bdf0c342
--- /dev/null
+++ b/ehrapy/tools/population_logging/_pop_logger.py
@@ -0,0 +1,193 @@
+import copy
+from typing import Any, Union
+
+import graphviz
+import matplotlib.pyplot as plt
+import pandas as pd
+import seaborn as sns
+from scanpy import AnnData
+from tableone import TableOne
+
+
+def _check_columns_exist(df, columns):
+    if not all(col in df.columns for col in columns):
+        missing_columns = [col for col in columns if col not in df.columns]
+        raise ValueError(f"Columns {missing_columns} not found in dataframe.")
+
+
+def get_column_structure(df, columns):
+    columns = df.columns if columns is None else columns
+    column_structure = {}
+
+    for column in columns:
+        if isinstance(df[column], pd.CategoricalDtype):
+            column_structure[column] = {category: [] for category in df[column].cat.categories}
+        elif pd.api.types.is_numeric_dtype(df[column]):
+            column_structure[column] = []
+        else:
+            # Coerce to categorical
+            df[column] = df[column].astype("category")
+            column_structure[column] = {category: [] for category in df[column].cat.categories}
+    return column_structure
+
+
+def log_from_tableone():
+    pass  # roughly get_column_dicts I think
+
+
+class PopulationLogger:
+    def __init__(self, adata: AnnData, columns: list = None, *args: Any):
+        """
+        TODO: write docsring
+        """
+        if columns is not None:
+            _check_columns_exist(adata.obs, columns)
+
+        self.columns = columns if columns is not None else adata.obs.columns
+
+        self.log = get_column_structure(adata.obs, columns)
+
+        self._logged_steps: int = 0
+
+        self._logged_text: list = []
+
+        self._logged_operations: list = []
+
+        self._log_backup = copy.deepcopy(self.log)
+
+        self.columns = args
+
+    def __call__(
+        self, adata: AnnData, label: str = None, operations_done: str = None, *args: Any, **tableone_kwargs: Any
+    ) -> Any:
+        _check_columns_exist(adata, self.columns)
+
+        # log a small text with each logging step, for the flowchart
+        log_text = label if label is not None else f"Step {self.logged_steps}"
+        log_text += "\n (n=" + str(adata.n_obs) + ")"
+        self._logged_text.append(log_text)
+
+        # log a small text with the operations done
+        self._logged_operations.append(operations_done)
+
+        self._logged_steps += 1
+
+        t1 = TableOne(adata.obs, **tableone_kwargs)
+        # log new stuff
+        self._get_column_dicts(t1)
+
+    def _get_column_dicts(self, table_one):
+        for key, value in self.log.items():
+            if isinstance(value, dict):
+                self._get_cat_dicts(table_one, key)
+            else:
+                # self.log[key] = self.get_num_dicts(table_one, key)
+                pass
+
+    def _get_cat_dicts(self, table_one, col):
+        for cat in self.log[col].keys():
+            pct = float(table_one.cat_table["Overall"].loc[(col, cat)].split("(")[1].split(")")[0])
+            self.log[col][cat].append(pct)
+
+    def _get_num_dicts(self, table_one, col):
+        return 0  # TODO
+
+    def reset(self):
+        self.log = self._log_backup
+        self._logged_steps = 0
+        self._logged_text = []
+
+    @property
+    def logged_steps(self):
+        return self._logged_steps
+
+    def plot_population_change(self, save: str = None, return_plot: bool = False):
+        """
+        Plot the population change over the logged steps.
+        TODO: write docstring
+        """
+        # Plotting
+        fig, axes = plt.subplots(self.logged_steps, 1, figsize=(7, 7))
+
+        legend_labels = []
+
+        # if only one step is logged, axes object is not iterable
+        if self.logged_steps == 1:
+            axes = [axes]
+
+        # each logged step is a subplot
+        for idx, ax in enumerate(axes):
+            # TODO: continue here
+            for pos, (_cols, data) in enumerate(self.log.items()):
+                data = pd.DataFrame(data).loc[idx]
+
+                cumwidth = 0
+
+                # Adjust the hue shift based on the category position such that the colors are more distinguishable
+                hue_shift = (pos + 1) / len(data)
+                colors = sns.color_palette("husl", len(data))
+                adjusted_colors = [((color[0] + hue_shift) % 1, color[1], color[2]) for color in colors]
+
+                for i, value in enumerate(data):
+                    # value = value[1].values[0]  # Take the value based on idx
+                    ax.barh(pos, value, left=cumwidth, color=adjusted_colors[i], height=0.8)
+
+                    if value > 5:
+                        # Add proportion numbers to the bars
+                        width = value
+                        ax.text(
+                            cumwidth + width / 2,
+                            pos,
+                            f"{value:.1f}",
+                            ha="center",
+                            va="center",
+                            color="white",
+                            fontweight="bold",
+                        )
+
+                    ax.set_yticks([])
+                    ax.set_xticks([])
+                    cumwidth += value
+
+                    legend_labels.append(data.index[i])
+
+        # makes the frames invisible
+        # for ax in axes:
+        #     ax.axis('off')
+
+        # Add legend for the first subplot
+        plt.legend(legend_labels, loc="best", bbox_to_anchor=(-0.5, 1))
+
+        if save is not None:
+            if not isinstance(save, str):
+                raise ValueError("'save' must be a string.")
+            plt.savefig(
+                save,
+            )
+
+        if return_plot:
+            return fig, axes
+
+        else:
+            plt.tight_layout()
+            plt.show()
+
+    def plot_flowchart(self, save: str = None, return_plot: bool = False):
+        """
+        Plot the flowchart of the logged steps.
+        """
+
+        # Create Digraph object
+        dot = graphviz.Digraph()
+
+        # Define nodes (edgy nodes)
+        for i, text in enumerate(self._logged_text):
+            dot.node(name=str(i), label=text, style="filled", shape="box")
+
+        for i, op in enumerate(self._logged_operations[1:]):
+            dot.edge(str(i), str(i + 1), label=op, labeldistance="10.5")
+
+        # Render the graph
+        dot.render("flow_diagram_edgy_nodes", format="png", cleanup=True)
+
+        return dot

From cd7e11391fd1fe85663f9f78979d01ad86822159 Mon Sep 17 00:00:00 2001
From: Eljas Roellin <eljas.roellin@ikmail.com>
Date: Wed, 14 Feb 2024 13:58:24 +0100
Subject: [PATCH 03/46] refined and first-line debugged population tracking

---
 ehrapy/tools/__init__.py                      |    2 +-
 .../tools/population_logging/_pop_logger.py   |  193 --
 .../__init__.py                               |    0
 .../tools/population_tracking/_pop_tracker.py |  315 +++
 try_flowchart.ipynb                           | 2273 ++++++++++-------
 5 files changed, 1614 insertions(+), 1169 deletions(-)
 delete mode 100644 ehrapy/tools/population_logging/_pop_logger.py
 rename ehrapy/tools/{population_logging => population_tracking}/__init__.py (100%)
 create mode 100644 ehrapy/tools/population_tracking/_pop_tracker.py

diff --git a/ehrapy/tools/__init__.py b/ehrapy/tools/__init__.py
index d89d00fc..d594c8d2 100644
--- a/ehrapy/tools/__init__.py
+++ b/ehrapy/tools/__init__.py
@@ -2,7 +2,7 @@
 from ehrapy.tools._scanpy_tl_api import *  # noqa: F403
 from ehrapy.tools.causal._dowhy import causal_inference
 from ehrapy.tools.feature_ranking._rank_features_groups import filter_rank_features_groups, rank_features_groups
-from ehrapy.tools.population_logging._pop_logger import PopulationLogger
+from ehrapy.tools.population_tracking._pop_tracker import PopulationTracker
 
 try:  # pragma: no cover
     from ehrapy.tools.nlp._medcat import (
diff --git a/ehrapy/tools/population_logging/_pop_logger.py b/ehrapy/tools/population_logging/_pop_logger.py
deleted file mode 100644
index bdf0c342..00000000
--- a/ehrapy/tools/population_logging/_pop_logger.py
+++ /dev/null
@@ -1,193 +0,0 @@
-import copy
-from typing import Any, Union
-
-import graphviz
-import matplotlib.pyplot as plt
-import pandas as pd
-import seaborn as sns
-from scanpy import AnnData
-from tableone import TableOne
-
-
-def _check_columns_exist(df, columns):
-    if not all(col in df.columns for col in columns):
-        missing_columns = [col for col in columns if col not in df.columns]
-        raise ValueError(f"Columns {missing_columns} not found in dataframe.")
-
-
-def get_column_structure(df, columns):
-    columns = df.columns if columns is None else columns
-    column_structure = {}
-
-    for column in columns:
-        if isinstance(df[column], pd.CategoricalDtype):
-            column_structure[column] = {category: [] for category in df[column].cat.categories}
-        elif pd.api.types.is_numeric_dtype(df[column]):
-            column_structure[column] = []
-        else:
-            # Coerce to categorical
-            df[column] = df[column].astype("category")
-            column_structure[column] = {category: [] for category in df[column].cat.categories}
-    return column_structure
-
-
-def log_from_tableone():
-    pass  # roughly get_column_dicts I think
-
-
-class PopulationLogger:
-    def __init__(self, adata: AnnData, columns: list = None, *args: Any):
-        """
-        TODO: write docsring
-        """
-        if columns is not None:
-            _check_columns_exist(adata.obs, columns)
-
-        self.columns = columns if columns is not None else adata.obs.columns
-
-        self.log = get_column_structure(adata.obs, columns)
-
-        self._logged_steps: int = 0
-
-        self._logged_text: list = []
-
-        self._logged_operations: list = []
-
-        self._log_backup = copy.deepcopy(self.log)
-
-        self.columns = args
-
-    def __call__(
-        self, adata: AnnData, label: str = None, operations_done: str = None, *args: Any, **tableone_kwargs: Any
-    ) -> Any:
-        _check_columns_exist(adata, self.columns)
-
-        # log a small text with each logging step, for the flowchart
-        log_text = label if label is not None else f"Step {self.logged_steps}"
-        log_text += "\n (n=" + str(adata.n_obs) + ")"
-        self._logged_text.append(log_text)
-
-        # log a small text with the operations done
-        self._logged_operations.append(operations_done)
-
-        self._logged_steps += 1
-
-        t1 = TableOne(adata.obs, **tableone_kwargs)
-        # log new stuff
-        self._get_column_dicts(t1)
-
-    def _get_column_dicts(self, table_one):
-        for key, value in self.log.items():
-            if isinstance(value, dict):
-                self._get_cat_dicts(table_one, key)
-            else:
-                # self.log[key] = self.get_num_dicts(table_one, key)
-                pass
-
-    def _get_cat_dicts(self, table_one, col):
-        for cat in self.log[col].keys():
-            pct = float(table_one.cat_table["Overall"].loc[(col, cat)].split("(")[1].split(")")[0])
-            self.log[col][cat].append(pct)
-
-    def _get_num_dicts(self, table_one, col):
-        return 0  # TODO
-
-    def reset(self):
-        self.log = self._log_backup
-        self._logged_steps = 0
-        self._logged_text = []
-
-    @property
-    def logged_steps(self):
-        return self._logged_steps
-
-    def plot_population_change(self, save: str = None, return_plot: bool = False):
-        """
-        Plot the population change over the logged steps.
-        TODO: write docstring
-        """
-        # Plotting
-        fig, axes = plt.subplots(self.logged_steps, 1, figsize=(7, 7))
-
-        legend_labels = []
-
-        # if only one step is logged, axes object is not iterable
-        if self.logged_steps == 1:
-            axes = [axes]
-
-        # each logged step is a subplot
-        for idx, ax in enumerate(axes):
-            # TODO: continue here
-            for pos, (_cols, data) in enumerate(self.log.items()):
-                data = pd.DataFrame(data).loc[idx]
-
-                cumwidth = 0
-
-                # Adjust the hue shift based on the category position such that the colors are more distinguishable
-                hue_shift = (pos + 1) / len(data)
-                colors = sns.color_palette("husl", len(data))
-                adjusted_colors = [((color[0] + hue_shift) % 1, color[1], color[2]) for color in colors]
-
-                for i, value in enumerate(data):
-                    # value = value[1].values[0]  # Take the value based on idx
-                    ax.barh(pos, value, left=cumwidth, color=adjusted_colors[i], height=0.8)
-
-                    if value > 5:
-                        # Add proportion numbers to the bars
-                        width = value
-                        ax.text(
-                            cumwidth + width / 2,
-                            pos,
-                            f"{value:.1f}",
-                            ha="center",
-                            va="center",
-                            color="white",
-                            fontweight="bold",
-                        )
-
-                    ax.set_yticks([])
-                    ax.set_xticks([])
-                    cumwidth += value
-
-                    legend_labels.append(data.index[i])
-
-        # makes the frames invisible
-        # for ax in axes:
-        #     ax.axis('off')
-
-        # Add legend for the first subplot
-        plt.legend(legend_labels, loc="best", bbox_to_anchor=(-0.5, 1))
-
-        if save is not None:
-            if not isinstance(save, str):
-                raise ValueError("'save' must be a string.")
-            plt.savefig(
-                save,
-            )
-
-        if return_plot:
-            return fig, axes
-
-        else:
-            plt.tight_layout()
-            plt.show()
-
-    def plot_flowchart(self, save: str = None, return_plot: bool = False):
-        """
-        Plot the flowchart of the logged steps.
-        """
-
-        # Create Digraph object
-        dot = graphviz.Digraph()
-
-        # Define nodes (edgy nodes)
-        for i, text in enumerate(self._logged_text):
-            dot.node(name=str(i), label=text, style="filled", shape="box")
-
-        for i, op in enumerate(self._logged_operations[1:]):
-            dot.edge(str(i), str(i + 1), label=op, labeldistance="10.5")
-
-        # Render the graph
-        dot.render("flow_diagram_edgy_nodes", format="png", cleanup=True)
-
-        return dot
diff --git a/ehrapy/tools/population_logging/__init__.py b/ehrapy/tools/population_tracking/__init__.py
similarity index 100%
rename from ehrapy/tools/population_logging/__init__.py
rename to ehrapy/tools/population_tracking/__init__.py
diff --git a/ehrapy/tools/population_tracking/_pop_tracker.py b/ehrapy/tools/population_tracking/_pop_tracker.py
new file mode 100644
index 00000000..ca1be393
--- /dev/null
+++ b/ehrapy/tools/population_tracking/_pop_tracker.py
@@ -0,0 +1,315 @@
+import copy
+from typing import Any, Union
+
+import graphviz
+import matplotlib.pyplot as plt
+import numpy as np
+import pandas as pd
+import seaborn as sns
+from scanpy import AnnData
+from tableone import TableOne
+
+
+def _check_columns_exist(df, columns):
+    if not all(col in df.columns for col in columns):
+        missing_columns = [col for col in columns if col not in df.columns]
+        raise ValueError(f"Columns {missing_columns} not found in dataframe.")
+
+
+# from tableone: https://github.com/tompollard/tableone/blob/bfd6fbaa4ed3e9f59e1a75191c6296a2a80ccc64/tableone/tableone.py#L555
+def _detect_categorical_columns(data) -> list:
+    """
+    Detect categorical columns if they are not specified.
+
+    Parameters
+    ----------
+        data : pandas DataFrame
+            The input dataset.
+
+    Returns
+    ----------
+        likely_cat : list
+            List of variables that appear to be categorical.
+    """
+    # assume all non-numerical and date columns are categorical
+    numeric_cols = set(data._get_numeric_data().columns.values)
+    date_cols = set(data.select_dtypes(include=[np.datetime64]).columns)
+    likely_cat = set(data.columns) - numeric_cols
+    # mypy absolutely looses it if likely_cat is overwritten to be a list
+    likely_cat_no_dates = list(likely_cat - date_cols)
+
+    # check proportion of unique values if numerical
+    for var in data._get_numeric_data().columns:
+        likely_flag = 1.0 * data[var].nunique() / data[var].count() < 0.005
+        if likely_flag:
+            likely_cat_no_dates.append(var)
+    return likely_cat_no_dates
+
+
+class PopulationTracker:
+    def __init__(self, adata: AnnData, columns: list = None, categorical: list = None, *args: Any):
+        """
+
+        categorical : list, optional
+        List of columns that contain categorical variables.
+        """
+        if columns is not None:
+            _check_columns_exist(adata.obs, columns)
+            _check_columns_exist(adata.obs, categorical)
+
+        self._tracked_steps: int = 0
+        self._tracked_text: list = []
+        self._tracked_operations: list = []
+
+        self.columns = columns if columns is not None else adata.obs.columns
+
+        # if categorical columns specified, use them
+        # else, follow tableone's logic
+        self.categorical = categorical if categorical is not None else _detect_categorical_columns(adata.obs[columns])
+        self.track = self._get_column_structure(adata.obs)
+
+        self._track_backup = copy.deepcopy(self.track)
+
+    def __call__(
+        self, adata: AnnData, label: str = None, operations_done: str = None, *args: Any, **tableone_kwargs: Any
+    ) -> Any:
+        _check_columns_exist(adata.obs, self.columns)
+
+        # track a small text with each tracking step, for the flowchart
+        track_text = label if label is not None else f"Cohort {self.tracked_steps}"
+        track_text += "\n (n=" + str(adata.n_obs) + ")"
+        self._tracked_text.append(track_text)
+
+        # track a small text with the operations done
+        self._tracked_operations.append(operations_done)
+
+        self._tracked_steps += 1
+
+        t1 = TableOne(adata.obs, categorical=self.categorical, **tableone_kwargs)
+        # track new stuff
+        self._get_column_dicts(t1)
+
+    def _get_column_structure(self, df):
+        column_structure = {}
+        for column in self.columns:
+            if column in self.categorical:
+                # if e.g. a column containing integers is deemed categorical, coerce it to categorical
+                df[column] = df[column].astype("category")
+                column_structure[column] = {category: [] for category in df[column].cat.categories}
+            else:
+                column_structure[column] = []
+
+        return column_structure
+
+    def _get_column_dicts(self, table_one):
+        for col, value in self.track.items():
+            if isinstance(value, dict):
+                self._get_cat_dicts(table_one, col)
+            else:
+                self._get_num_dicts(table_one, col)
+
+    def _get_cat_dicts(self, table_one, col):
+        for cat in self.track[col].keys():
+            # if tableone does not have the category of this column anymore, set the percentage to 0
+            # for categorized columns (e.g. gender 1.0/0.0), str(cat) helps to avoid considering the category as a float
+            if (col, str(cat)) in table_one.cat_table["Overall"].index:
+                pct = float(table_one.cat_table["Overall"].loc[(col, str(cat))].split("(")[1].split(")")[0])
+            else:
+                pct = 0
+            self.track[col][cat].append(pct)
+
+    def _get_num_dicts(self, table_one, col):
+        summary = table_one.cont_table["Overall"].loc[(col, "")]
+        self.track[col].append(summary)
+
+    def reset(self):
+        self.track = self._track_backup
+        self._tracked_steps = 0
+        self._tracked_text = []
+
+    @property
+    def tracked_steps(self):
+        return self._tracked_steps
+
+    def plot_population_change(
+        self,
+        set_axis_labels=True,
+        subfigure_title: bool = False,
+        sns_color_palette: str = "husl",
+        save: str = None,
+        return_plot: bool = False,
+        subplots_kwargs: dict = None,
+        legend_kwargs: dict = None,
+    ):
+        """Plot the population change over the tracked steps.
+
+        Create stacked bar plots to monitor population changes over the steps tracked with `PopulationTracker`.
+
+        Args:
+            set_axis_labels: If `True`, the y-axis labels will be set to the column names.
+            subfigure_title: If `True`, each subplot will have a title with the `label` provided during tracking.
+            sns_color_palette: The color palette to use for the plot. Default is "husl".
+            save: If a string is provided, the plot will be saved to the path specified.
+            return_plot: If `True`, the plot will be returned as a tuple of (fig, ax).
+            subplot_kwargs: Additional keyword arguments for the subplots.
+            legend_kwargs: Additional keyword arguments for the legend.
+
+        Returns:
+            If `return_plot` a :class:`~matplotlib.figure.Figure` and a :class:`~matplotlib.axes.Axes` or a list of it.
+
+        Example:
+            .. code-block:: python
+
+                import ehrapy as ep
+
+                adata = ep.dt.diabetes_130(columns_obs_only=["gender", "race", "weight", "age"])
+                pop_track = ep.tl.PopulationTracker(adata)
+                pop_track(adata, label="original")
+                adata = adata[:1000]
+                pop_track(adata, label="filtered cohort", operations_done="filtered to first 1000 entries")
+                pop_track.plot_flowchart()
+        Preview:
+            .. image:: /_static/docstring_previews/flowchart.png
+        """
+        # Plotting
+        subplots_kwargs = {} if subplots_kwargs is None else subplots_kwargs
+
+        fig, axes = plt.subplots(self.tracked_steps, 1, **subplots_kwargs)
+
+        legend_labels = []
+
+        # if only one step is tracked, axes object is not iterable
+        if self.tracked_steps == 1:
+            axes = [axes]
+
+        # each tracked step is a subplot
+        for idx, ax in enumerate(axes):
+            if subfigure_title:
+                ax.set_title(self._tracked_text[idx])
+
+            # iterate over the tracked columns in the dataframe
+            for pos, (_cols, data) in enumerate(self.track.items()):
+                data = pd.DataFrame(data).loc[idx]
+
+                cumwidth = 0
+
+                # Adjust the hue shift based on the category position such that the colors are more distinguishable
+                hue_shift = (pos + 1) / len(data)
+                colors = sns.color_palette(sns_color_palette, len(data))
+                adjusted_colors = [((color[0] + hue_shift) % 1, color[1], color[2]) for color in colors]
+
+                # for categoricals, plot multiple bars
+                if _cols in self.categorical:
+                    for i, value in enumerate(data):
+                        ax.barh(pos, value, left=cumwidth, color=adjusted_colors[i], height=0.7)
+
+                        if value > 5:
+                            # Add proportion numbers to the bars
+                            width = value
+                            ax.text(
+                                cumwidth + width / 2,
+                                pos,
+                                f"{value:.1f}",
+                                ha="center",
+                                va="center",
+                                color="white",
+                                fontweight="bold",
+                            )
+
+                        ax.set_yticks([])
+                        ax.set_xticks([])
+                        cumwidth += value
+                        legend_labels.append(data.index[i])
+
+                # for numericals, plot a single bar
+                else:
+                    ax.barh(pos, 100, left=cumwidth, color=adjusted_colors[0], height=0.8)
+                    ax.text(
+                        100 / 2,
+                        pos,
+                        data[0],
+                        ha="center",
+                        va="center",
+                        color="white",
+                        fontweight="bold",
+                    )
+                    legend_labels.append(_cols)
+
+            # Set y-axis labels
+            if set_axis_labels:
+                ax.set_yticks(
+                    range(len(self.track.keys()))
+                )  # Set ticks at positions corresponding to the number of columns
+                ax.set_yticklabels(self.track.keys())  # Set y-axis labels to the column names
+
+            # makes the frames invisible
+            # for ax in axes:
+            #     ax.axis('off')
+
+        # Add legend
+        tot_legend_kwargs = {"loc": "best", "bbox_to_anchor": (1, 1)}
+        tot_legend_kwargs.update(legend_kwargs)
+
+        plt.legend(legend_labels, **tot_legend_kwargs)
+
+        if save is not None:
+            if not isinstance(save, str):
+                raise ValueError("'save' must be a string.")
+            plt.savefig(
+                save,
+            )
+
+        if return_plot:
+            return fig, axes
+
+        else:
+            plt.tight_layout()
+            plt.show()
+
+    def plot_flowchart(self, save: str = None, return_plot: bool = True):
+        """Flowchart over the tracked steps.
+
+        Create a simple flowchart of data preparation steps tracked with `PopulationTracker`.
+
+        Args:
+            save: If a string is provided, the plot will be saved to the path specified.
+            return_plot: If `True`, the plot will be returned as a :class:`~graphviz.Digraph`.
+
+        Returns:
+            If `return_plot` a :class:`~graphviz.Digraph`.
+
+        Example:
+            .. code-block:: python
+
+                import ehrapy as ep
+
+                adata = ep.dt.diabetes_130(columns_obs_only=["gender", "race", "weight", "age"])
+                pop_track = ep.tl.PopulationTracker(adata)
+                pop_track(adata, label="original")
+                adata = adata[:1000]
+                pop_track(adata, label="filtered cohort", operations_done="filtered to first 1000 entries")
+                pop_track.plot_flowchart()
+        Preview:
+            .. image:: /_static/docstring_previews/flowchart.png
+
+        """
+
+        # Create Digraph object
+        dot = graphviz.Digraph()
+
+        # Define nodes (edgy nodes)
+        for i, text in enumerate(self._tracked_text):
+            dot.node(name=str(i), label=text, style="filled", shape="box")
+
+        for i, op in enumerate(self._tracked_operations[1:]):
+            dot.edge(str(i), str(i + 1), label=op, labeldistance="2.5")
+
+        # Render the graph
+        if save is not None:
+            if not isinstance(save, str):
+                raise ValueError("'save' must be a string.")
+            dot.render(save, format="png", cleanup=True)
+
+        # Think that to be shown, the plot can a) be rendered (as above) or be "printed" by the notebook
+        if return_plot:
+            return dot
diff --git a/try_flowchart.ipynb b/try_flowchart.ipynb
index 61047371..a1bafc39 100644
--- a/try_flowchart.ipynb
+++ b/try_flowchart.ipynb
@@ -2,7 +2,7 @@
  "cells": [
   {
    "cell_type": "code",
-   "execution_count": 261,
+   "execution_count": 55,
    "metadata": {},
    "outputs": [
     {
@@ -26,56 +26,50 @@
        "  <thead>\n",
        "    <tr style=\"text-align: right;\">\n",
        "      <th></th>\n",
-       "      <th>ethnicity</th>\n",
-       "      <th>gender</th>\n",
-       "      <th>proficiency</th>\n",
+       "      <th>haircolor</th>\n",
+       "      <th>height</th>\n",
        "    </tr>\n",
        "  </thead>\n",
        "  <tbody>\n",
        "    <tr>\n",
        "      <th>0</th>\n",
-       "      <td>asian</td>\n",
-       "      <td>female</td>\n",
-       "      <td>high prof.</td>\n",
+       "      <td>colored</td>\n",
+       "      <td>179</td>\n",
        "    </tr>\n",
        "    <tr>\n",
        "      <th>1</th>\n",
-       "      <td>white</td>\n",
-       "      <td>male</td>\n",
-       "      <td>limited prof.</td>\n",
+       "      <td>black</td>\n",
+       "      <td>165</td>\n",
        "    </tr>\n",
        "    <tr>\n",
        "      <th>2</th>\n",
-       "      <td>black</td>\n",
-       "      <td>female</td>\n",
-       "      <td>high prof.</td>\n",
+       "      <td>colored</td>\n",
+       "      <td>174</td>\n",
        "    </tr>\n",
        "    <tr>\n",
        "      <th>3</th>\n",
        "      <td>black</td>\n",
-       "      <td>male</td>\n",
-       "      <td>limited prof.</td>\n",
+       "      <td>173</td>\n",
        "    </tr>\n",
        "    <tr>\n",
        "      <th>4</th>\n",
-       "      <td>black</td>\n",
-       "      <td>male</td>\n",
-       "      <td>limited prof.</td>\n",
+       "      <td>colored</td>\n",
+       "      <td>181</td>\n",
        "    </tr>\n",
        "  </tbody>\n",
        "</table>\n",
        "</div>"
       ],
       "text/plain": [
-       "  ethnicity  gender    proficiency\n",
-       "0     asian  female     high prof.\n",
-       "1     white    male  limited prof.\n",
-       "2     black  female     high prof.\n",
-       "3     black    male  limited prof.\n",
-       "4     black    male  limited prof."
+       "  haircolor  height\n",
+       "0   colored     179\n",
+       "1     black     165\n",
+       "2   colored     174\n",
+       "3     black     173\n",
+       "4   colored     181"
       ]
      },
-     "execution_count": 261,
+     "execution_count": 55,
      "metadata": {},
      "output_type": "execute_result"
     }
@@ -87,21 +81,24 @@
     "import ehrapy as ep\n",
     "from tableone import TableOne\n",
     "import seaborn as sns\n",
+    "import scanpy as sc\n",
     "\n",
     "# generate dataset\n",
-    "rng = np.random.default_rng(42)\n",
-    "\n",
-    "random_dataset = pd.DataFrame({\n",
-    "    \"ethnicity\": rng.choice([\"asian\", \"black\", \"white\"], size=150),\n",
-    "    \"gender\": rng.choice([\"male\", \"female\"], size=150),\n",
-    "    \"proficiency\": rng.choice([\"high prof.\", \"limited prof.\"], size=150)\n",
-    "    })\n",
+    "rng = np.random.default_rng(151)\n",
+    "\n",
+    "random_dataset = pd.DataFrame(\n",
+    "    {\n",
+    "        \"haircolor\": rng.choice([\"blond\", \"black\", \"colored\"], size=150),\n",
+    "        \"height\": rng.integers(low=160, high=190, size=150),\n",
+    "        # \"proficiency\": rng.choice([\"high prof.\", \"limited prof.\"], size=150),\n",
+    "    }\n",
+    ")\n",
     "random_dataset.head()"
    ]
   },
   {
    "cell_type": "code",
-   "execution_count": 122,
+   "execution_count": 72,
    "metadata": {},
    "outputs": [
     {
@@ -125,190 +122,177 @@
        "  <thead>\n",
        "    <tr style=\"text-align: right;\">\n",
        "      <th></th>\n",
-       "      <th></th>\n",
-       "      <th>Missing</th>\n",
-       "      <th>Overall</th>\n",
+       "      <th>haircolor</th>\n",
+       "      <th>height</th>\n",
        "    </tr>\n",
        "  </thead>\n",
        "  <tbody>\n",
        "    <tr>\n",
-       "      <th>n</th>\n",
-       "      <th></th>\n",
-       "      <td></td>\n",
-       "      <td>150</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th rowspan=\"3\" valign=\"top\">ethnicity, n (%)</th>\n",
-       "      <th>asian</th>\n",
-       "      <td>0</td>\n",
-       "      <td>43 (28.7)</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>black</th>\n",
-       "      <td></td>\n",
-       "      <td>52 (34.7)</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>white</th>\n",
-       "      <td></td>\n",
-       "      <td>55 (36.7)</td>\n",
+       "      <th>0</th>\n",
+       "      <td>colored</td>\n",
+       "      <td>176</td>\n",
        "    </tr>\n",
        "    <tr>\n",
-       "      <th rowspan=\"2\" valign=\"top\">gender, n (%)</th>\n",
-       "      <th>female</th>\n",
-       "      <td>0</td>\n",
-       "      <td>66 (44.0)</td>\n",
+       "      <th>1</th>\n",
+       "      <td>black</td>\n",
+       "      <td>173</td>\n",
        "    </tr>\n",
        "    <tr>\n",
-       "      <th>male</th>\n",
-       "      <td></td>\n",
-       "      <td>84 (56.0)</td>\n",
+       "      <th>2</th>\n",
+       "      <td>colored</td>\n",
+       "      <td>175</td>\n",
        "    </tr>\n",
        "    <tr>\n",
-       "      <th rowspan=\"2\" valign=\"top\">proficiency, n (%)</th>\n",
-       "      <th>high prof.</th>\n",
-       "      <td>0</td>\n",
-       "      <td>78 (52.0)</td>\n",
+       "      <th>3</th>\n",
+       "      <td>black</td>\n",
+       "      <td>175</td>\n",
        "    </tr>\n",
        "    <tr>\n",
-       "      <th>limited prof.</th>\n",
-       "      <td></td>\n",
-       "      <td>72 (48.0)</td>\n",
+       "      <th>4</th>\n",
+       "      <td>colored</td>\n",
+       "      <td>176</td>\n",
        "    </tr>\n",
        "  </tbody>\n",
        "</table>\n",
-       "</div><br />"
+       "</div>"
       ],
       "text/plain": [
-       "                                 Missing    Overall\n",
-       "n                                               150\n",
-       "ethnicity, n (%)   asian               0  43 (28.7)\n",
-       "                   black                  52 (34.7)\n",
-       "                   white                  55 (36.7)\n",
-       "gender, n (%)      female              0  66 (44.0)\n",
-       "                   male                   84 (56.0)\n",
-       "proficiency, n (%) high prof.          0  78 (52.0)\n",
-       "                   limited prof.          72 (48.0)"
-      ]
-     },
-     "execution_count": 122,
-     "metadata": {},
-     "output_type": "execute_result"
-    }
-   ],
-   "source": [
-    "t1 = TableOne(random_dataset, columns=[\"ethnicity\", \"gender\", \"proficiency\"])\n",
-    "t1"
-   ]
-  },
-  {
-   "cell_type": "code",
-   "execution_count": 141,
-   "metadata": {},
-   "outputs": [
-    {
-     "data": {
-      "text/plain": [
-       "{'asian': [43.0],\n",
-       " 'black': [52.0],\n",
-       " 'white': [55.0],\n",
-       " 'male': [84.0],\n",
-       " 'female': [66.0],\n",
-       " 'high prof.': [78.0],\n",
-       " 'limited prof.': [72.0]}"
+       "  haircolor  height\n",
+       "0   colored     176\n",
+       "1     black     173\n",
+       "2   colored     175\n",
+       "3     black     175\n",
+       "4   colored     176"
       ]
      },
-     "execution_count": 141,
+     "execution_count": 72,
      "metadata": {},
      "output_type": "execute_result"
     }
    ],
    "source": [
-    "def get_dicts(table_one):\n",
-    "\n",
-    "    cols = [\"ethnicity\", \"gender\", \"proficiency\"]\n",
-    "    col_cats = {\"ethnicity\": [\"asian\", \"black\", \"white\"],\n",
-    "                \"gender\": [\"male\", \"female\"],\n",
-    "                \"proficiency\": [\"high prof.\", \"limited prof.\"]}\n",
-    "\n",
-    "    pd.DataFrame(columns=cols)\n",
-    "    dvals = {}\n",
-    "    dpcts = {}\n",
-    "    for col in cols:\n",
-    "        for cat in col_cats[col]:\n",
-    "\n",
-    "            val = float(table_one.cat_table[\"Overall\"].loc[(col, cat)].split(\" \")[0])\n",
-    "            pct = float(table_one.cat_table[\"Overall\"].loc[(col, cat)].split(\"(\")[1].split(\")\")[0])\n",
-    "            if cat not in dvals.keys():\n",
-    "                dvals[cat] = [val]\n",
-    "            else:\n",
-    "                dvals[cat].append(val)\n",
-    "\n",
-    "            if cat not in dpcts.keys():\n",
-    "                dpcts[cat] = [pct]\n",
-    "            else:\n",
-    "                dpcts[cat].append(pct)\n",
-    "\n",
-    "    return dvals, dpcts\n",
-    "\n",
-    "dvals, dcpts = get_dicts(t1)\n",
-    "dvals"
+    "# generate dataset\n",
+    "rng = np.random.default_rng(151)\n",
+    "\n",
+    "random_dataset = pd.DataFrame(\n",
+    "    {\n",
+    "        \"haircolor\": rng.choice([\"blond\", \"black\", \"colored\"], size=150),\n",
+    "        \"height\": rng.integers(low=173, high=178, size=150),\n",
+    "        # \"proficiency\": rng.choice([\"high prof.\", \"limited prof.\"], size=150),\n",
+    "    }\n",
+    ")\n",
+    "random_dataset.head()"
    ]
   },
   {
    "cell_type": "code",
-   "execution_count": 142,
+   "execution_count": 57,
    "metadata": {},
    "outputs": [
     {
      "data": {
+      "image/png": "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",
+      "text/latex": [
+       "$\\displaystyle 0.0333333333333333$"
+      ],
       "text/plain": [
-       "{'asian': [28.7],\n",
-       " 'black': [34.7],\n",
-       " 'white': [36.7],\n",
-       " 'male': [56.0],\n",
-       " 'female': [44.0],\n",
-       " 'high prof.': [52.0],\n",
-       " 'limited prof.': [48.0]}"
+       "0.03333333333333333"
       ]
      },
-     "execution_count": 142,
+     "execution_count": 57,
      "metadata": {},
      "output_type": "execute_result"
     }
    ],
    "source": [
-    "dpcts"
+    "1.0 * random_dataset[\"height\"].nunique() / random_dataset[\"height\"].count()"
    ]
   },
   {
    "cell_type": "code",
-   "execution_count": 143,
+   "execution_count": 76,
    "metadata": {},
    "outputs": [
     {
      "data": {
-      "image/png": "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",
-      "text/latex": [
-       "$\\displaystyle \\left( 50, \\  3\\right)$"
+      "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></th>\n",
+       "      <th>Missing</th>\n",
+       "      <th>Overall</th>\n",
+       "    </tr>\n",
+       "  </thead>\n",
+       "  <tbody>\n",
+       "    <tr>\n",
+       "      <th>n</th>\n",
+       "      <th></th>\n",
+       "      <td></td>\n",
+       "      <td>150</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th rowspan=\"3\" valign=\"top\">haircolor, n (%)</th>\n",
+       "      <th>black</th>\n",
+       "      <td>0</td>\n",
+       "      <td>45 (30.0)</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>blond</th>\n",
+       "      <td></td>\n",
+       "      <td>55 (36.7)</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>colored</th>\n",
+       "      <td></td>\n",
+       "      <td>50 (33.3)</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>height, mean (SD)</th>\n",
+       "      <th></th>\n",
+       "      <td>0</td>\n",
+       "      <td>175.0 (1.5)</td>\n",
+       "    </tr>\n",
+       "  </tbody>\n",
+       "</table>\n",
+       "</div><br />"
       ],
       "text/plain": [
-       "(50, 3)"
+       "                          Missing      Overall\n",
+       "n                                          150\n",
+       "haircolor, n (%)  black         0    45 (30.0)\n",
+       "                  blond              55 (36.7)\n",
+       "                  colored            50 (33.3)\n",
+       "height, mean (SD)               0  175.0 (1.5)"
       ]
      },
-     "execution_count": 143,
+     "execution_count": 76,
      "metadata": {},
      "output_type": "execute_result"
     }
    ],
    "source": [
-    "random_dataset_2 = random_dataset[:50]\n",
-    "random_dataset_2.shape"
+    "TableOne(random_dataset, categorical=[\"haircolor\"])"
    ]
   },
   {
    "cell_type": "code",
-   "execution_count": 144,
+   "execution_count": 59,
    "metadata": {},
    "outputs": [
     {
@@ -332,432 +316,465 @@
        "  <thead>\n",
        "    <tr style=\"text-align: right;\">\n",
        "      <th></th>\n",
-       "      <th></th>\n",
-       "      <th>Missing</th>\n",
-       "      <th>Overall</th>\n",
+       "      <th>haircolor</th>\n",
+       "      <th>height</th>\n",
        "    </tr>\n",
        "  </thead>\n",
        "  <tbody>\n",
        "    <tr>\n",
-       "      <th>n</th>\n",
-       "      <th></th>\n",
-       "      <td></td>\n",
-       "      <td>50</td>\n",
+       "      <th>0</th>\n",
+       "      <td>colored</td>\n",
+       "      <td>176</td>\n",
        "    </tr>\n",
        "    <tr>\n",
-       "      <th rowspan=\"3\" valign=\"top\">ethnicity, n (%)</th>\n",
-       "      <th>asian</th>\n",
-       "      <td>0</td>\n",
-       "      <td>12 (24.0)</td>\n",
+       "      <th>1</th>\n",
+       "      <td>black</td>\n",
+       "      <td>173</td>\n",
        "    </tr>\n",
        "    <tr>\n",
-       "      <th>black</th>\n",
-       "      <td></td>\n",
-       "      <td>17 (34.0)</td>\n",
+       "      <th>2</th>\n",
+       "      <td>colored</td>\n",
+       "      <td>175</td>\n",
        "    </tr>\n",
        "    <tr>\n",
-       "      <th>white</th>\n",
-       "      <td></td>\n",
-       "      <td>21 (42.0)</td>\n",
+       "      <th>3</th>\n",
+       "      <td>black</td>\n",
+       "      <td>175</td>\n",
        "    </tr>\n",
        "    <tr>\n",
-       "      <th rowspan=\"2\" valign=\"top\">gender, n (%)</th>\n",
-       "      <th>female</th>\n",
-       "      <td>0</td>\n",
-       "      <td>22 (44.0)</td>\n",
+       "      <th>4</th>\n",
+       "      <td>colored</td>\n",
+       "      <td>176</td>\n",
        "    </tr>\n",
        "    <tr>\n",
-       "      <th>male</th>\n",
-       "      <td></td>\n",
-       "      <td>28 (56.0)</td>\n",
+       "      <th>...</th>\n",
+       "      <td>...</td>\n",
+       "      <td>...</td>\n",
        "    </tr>\n",
        "    <tr>\n",
-       "      <th rowspan=\"2\" valign=\"top\">proficiency, n (%)</th>\n",
-       "      <th>high prof.</th>\n",
-       "      <td>0</td>\n",
-       "      <td>24 (48.0)</td>\n",
+       "      <th>145</th>\n",
+       "      <td>blond</td>\n",
+       "      <td>173</td>\n",
        "    </tr>\n",
        "    <tr>\n",
-       "      <th>limited prof.</th>\n",
-       "      <td></td>\n",
-       "      <td>26 (52.0)</td>\n",
+       "      <th>146</th>\n",
+       "      <td>black</td>\n",
+       "      <td>177</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>147</th>\n",
+       "      <td>black</td>\n",
+       "      <td>173</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>148</th>\n",
+       "      <td>blond</td>\n",
+       "      <td>174</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>149</th>\n",
+       "      <td>blond</td>\n",
+       "      <td>173</td>\n",
        "    </tr>\n",
        "  </tbody>\n",
        "</table>\n",
-       "</div><br />"
+       "<p>150 rows × 2 columns</p>\n",
+       "</div>"
       ],
       "text/plain": [
-       "                                 Missing    Overall\n",
-       "n                                                50\n",
-       "ethnicity, n (%)   asian               0  12 (24.0)\n",
-       "                   black                  17 (34.0)\n",
-       "                   white                  21 (42.0)\n",
-       "gender, n (%)      female              0  22 (44.0)\n",
-       "                   male                   28 (56.0)\n",
-       "proficiency, n (%) high prof.          0  24 (48.0)\n",
-       "                   limited prof.          26 (52.0)"
+       "    haircolor  height\n",
+       "0     colored     176\n",
+       "1       black     173\n",
+       "2     colored     175\n",
+       "3       black     175\n",
+       "4     colored     176\n",
+       "..        ...     ...\n",
+       "145     blond     173\n",
+       "146     black     177\n",
+       "147     black     173\n",
+       "148     blond     174\n",
+       "149     blond     173\n",
+       "\n",
+       "[150 rows x 2 columns]"
       ]
      },
-     "execution_count": 144,
+     "execution_count": 59,
      "metadata": {},
      "output_type": "execute_result"
     }
    ],
    "source": [
-    "t2 = TableOne(random_dataset_2)\n",
-    "t2"
+    "adata = sc.AnnData(X=rng.normal(size=(150, 100)), obs=random_dataset)\n",
+    "adata.obs"
    ]
   },
   {
    "cell_type": "code",
-   "execution_count": 146,
+   "execution_count": 60,
    "metadata": {},
    "outputs": [
     {
-     "data": {
-      "text/plain": [
-       "{'asian': [12.0],\n",
-       " 'black': [17.0],\n",
-       " 'white': [21.0],\n",
-       " 'male': [28.0],\n",
-       " 'female': [22.0],\n",
-       " 'high prof.': [24.0],\n",
-       " 'limited prof.': [26.0]}"
-      ]
-     },
-     "execution_count": 146,
-     "metadata": {},
-     "output_type": "execute_result"
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "The autoreload extension is already loaded. To reload it, use:\n",
+      "  %reload_ext autoreload\n"
+     ]
     }
    ],
    "source": [
-    "dvals2, dcpts2 = get_dicts(t2)\n",
-    "dvals2"
+    "%load_ext autoreload\n",
+    "%autoreload 2"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 61,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "pl = ep.tl.PopulationTracker(adata)\n",
+    "pl(adata)\n",
+    "adata = adata[:75]\n",
+    "pl(adata, label=\"filtered\", operations_done=\"filtered to first 75 entries\")"
    ]
   },
   {
    "cell_type": "code",
-   "execution_count": 147,
+   "execution_count": 62,
    "metadata": {},
    "outputs": [
     {
      "data": {
       "text/plain": [
-       "{'asian': [24.0],\n",
-       " 'black': [34.0],\n",
-       " 'white': [42.0],\n",
-       " 'male': [56.0],\n",
-       " 'female': [44.0],\n",
-       " 'high prof.': [48.0],\n",
-       " 'limited prof.': [52.0]}"
+       "[None, 'filtered to first 75 entries']"
       ]
      },
-     "execution_count": 147,
+     "execution_count": 62,
      "metadata": {},
      "output_type": "execute_result"
     }
    ],
    "source": [
-    "dcpts2"
+    "pl._tracked_operations"
    ]
   },
   {
    "cell_type": "code",
-   "execution_count": 151,
+   "execution_count": 63,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "# pl.reset()"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 64,
    "metadata": {},
    "outputs": [
     {
      "data": {
+      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAA0AAAAPCAYAAAA/I0V3AAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjguMiwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy8g+/7EAAAACXBIWXMAABJ0AAASdAHeZh94AAABBUlEQVR4nJXSPUscQBDG8d+d9w0srey1thZLQVCxVL+ACQcWAUGGKQJ2KtopeLVgqZjSMoIgKBJSWUoIQlrfzuL25Lzc+TLN7OzOf+aZ3a00m02ftVp7kZmDmMYkRjGEO1xgD3sR8QTVjgJz2MEYfmIDBxjBLvYzs/KqE35jCoftikXBCk4xixkcVD4yUwG/YzsivlTfA4rdF//QPVO/LjUslPD4QxDWtC7jKCJ+vAtl5lcs4xfm2/t9ocxcwiauMB4Rt29CmVnHFi4LcNN5/h+Umd+wjvMC/OnOqXYBq1qDn2EiIv72UvLyuJm5iAYei7R/PfKvI6LR+Y2Gix9AvVcHnKDxDEnuUnOCo1FOAAAAAElFTkSuQmCC",
+      "text/latex": [
+       "$\\displaystyle 2$"
+      ],
       "text/plain": [
-       "{'asian': [28.7, 24.0],\n",
-       " 'black': [34.7, 34.0],\n",
-       " 'white': [36.7, 42.0],\n",
-       " 'male': [56.0, 56.0],\n",
-       " 'female': [44.0, 44.0],\n",
-       " 'high prof.': [52.0, 48.0],\n",
-       " 'limited prof.': [48.0, 52.0]}"
+       "2"
       ]
      },
-     "execution_count": 151,
+     "execution_count": 64,
      "metadata": {},
      "output_type": "execute_result"
     }
    ],
    "source": [
-    "collated_data = {key: values1 + values2 for key, values1, values2 in zip(dpcts.keys(), dpcts.values(), dcpts2.values())}\n",
-    "collated_data"
+    "pl.tracked_steps"
    ]
   },
   {
    "cell_type": "code",
-   "execution_count": 155,
+   "execution_count": 65,
    "metadata": {},
    "outputs": [
     {
      "data": {
       "text/plain": [
-       "{'asian': [28.7, 24.0], 'black': [34.7, 34.0], 'white': [36.7, 42.0]}"
+       "{'haircolor': {'black': [30.0, 25.3],\n",
+       "  'blond': [36.7, 40.0],\n",
+       "  'colored': [33.3, 34.7]},\n",
+       " 'height': []}"
       ]
      },
-     "execution_count": 155,
+     "execution_count": 65,
      "metadata": {},
      "output_type": "execute_result"
     }
    ],
    "source": [
-    "data_ethnicity = {key: value for key, value in collated_data.items() if key in [\"asian\", \"black\", \"white\"]}\n",
-    "data_ethnicity"
+    "pl.track"
    ]
   },
   {
    "cell_type": "code",
-   "execution_count": 191,
+   "execution_count": 66,
    "metadata": {},
    "outputs": [
+    {
+     "ename": "KeyError",
+     "evalue": "0",
+     "output_type": "error",
+     "traceback": [
+      "\u001b[0;31m---------------------------------------------------------------------------\u001b[0m",
+      "\u001b[0;31mValueError\u001b[0m                                Traceback (most recent call last)",
+      "File \u001b[0;32m~/Documents/ehrapy_clean/ehrapy_venv_feb/lib/python3.11/site-packages/pandas/core/indexes/range.py:413\u001b[0m, in \u001b[0;36mRangeIndex.get_loc\u001b[0;34m(self, key)\u001b[0m\n\u001b[1;32m    412\u001b[0m \u001b[38;5;28;01mtry\u001b[39;00m:\n\u001b[0;32m--> 413\u001b[0m     \u001b[38;5;28;01mreturn\u001b[39;00m \u001b[38;5;28;43mself\u001b[39;49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43m_range\u001b[49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43mindex\u001b[49m\u001b[43m(\u001b[49m\u001b[43mnew_key\u001b[49m\u001b[43m)\u001b[49m\n\u001b[1;32m    414\u001b[0m \u001b[38;5;28;01mexcept\u001b[39;00m \u001b[38;5;167;01mValueError\u001b[39;00m \u001b[38;5;28;01mas\u001b[39;00m err:\n",
+      "\u001b[0;31mValueError\u001b[0m: 0 is not in range",
+      "\nThe above exception was the direct cause of the following exception:\n",
+      "\u001b[0;31mKeyError\u001b[0m                                  Traceback (most recent call last)",
+      "Cell \u001b[0;32mIn[66], line 1\u001b[0m\n\u001b[0;32m----> 1\u001b[0m fig, axes \u001b[38;5;241m=\u001b[39m \u001b[43mpl\u001b[49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43mplot_population_change\u001b[49m\u001b[43m(\u001b[49m\u001b[43mreturn_plot\u001b[49m\u001b[38;5;241;43m=\u001b[39;49m\u001b[38;5;28;43;01mTrue\u001b[39;49;00m\u001b[43m)\u001b[49m\n",
+      "File \u001b[0;32m~/Documents/ehrapy_clean/ehrapy/ehrapy/tools/population_tracking/_pop_tracker.py:142\u001b[0m, in \u001b[0;36mPopulationTracker.plot_population_change\u001b[0;34m(self, set_axis_labels, save, return_plot)\u001b[0m\n\u001b[1;32m    140\u001b[0m \u001b[38;5;28;01mfor\u001b[39;00m idx, ax \u001b[38;5;129;01min\u001b[39;00m \u001b[38;5;28menumerate\u001b[39m(axes):\n\u001b[1;32m    141\u001b[0m     \u001b[38;5;28;01mfor\u001b[39;00m pos, (_cols, data) \u001b[38;5;129;01min\u001b[39;00m \u001b[38;5;28menumerate\u001b[39m(\u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39mtrack\u001b[38;5;241m.\u001b[39mitems()):\n\u001b[0;32m--> 142\u001b[0m         data \u001b[38;5;241m=\u001b[39m \u001b[43mpd\u001b[49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43mDataFrame\u001b[49m\u001b[43m(\u001b[49m\u001b[43mdata\u001b[49m\u001b[43m)\u001b[49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43mloc\u001b[49m\u001b[43m[\u001b[49m\u001b[43midx\u001b[49m\u001b[43m]\u001b[49m\n\u001b[1;32m    144\u001b[0m         cumwidth \u001b[38;5;241m=\u001b[39m \u001b[38;5;241m0\u001b[39m\n\u001b[1;32m    146\u001b[0m         \u001b[38;5;66;03m# Adjust the hue shift based on the category position such that the colors are more distinguishable\u001b[39;00m\n",
+      "File \u001b[0;32m~/Documents/ehrapy_clean/ehrapy_venv_feb/lib/python3.11/site-packages/pandas/core/indexing.py:1192\u001b[0m, in \u001b[0;36m_LocationIndexer.__getitem__\u001b[0;34m(self, key)\u001b[0m\n\u001b[1;32m   1190\u001b[0m maybe_callable \u001b[38;5;241m=\u001b[39m com\u001b[38;5;241m.\u001b[39mapply_if_callable(key, \u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39mobj)\n\u001b[1;32m   1191\u001b[0m maybe_callable \u001b[38;5;241m=\u001b[39m \u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39m_check_deprecated_callable_usage(key, maybe_callable)\n\u001b[0;32m-> 1192\u001b[0m \u001b[38;5;28;01mreturn\u001b[39;00m \u001b[38;5;28;43mself\u001b[39;49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43m_getitem_axis\u001b[49m\u001b[43m(\u001b[49m\u001b[43mmaybe_callable\u001b[49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[43maxis\u001b[49m\u001b[38;5;241;43m=\u001b[39;49m\u001b[43maxis\u001b[49m\u001b[43m)\u001b[49m\n",
+      "File \u001b[0;32m~/Documents/ehrapy_clean/ehrapy_venv_feb/lib/python3.11/site-packages/pandas/core/indexing.py:1432\u001b[0m, in \u001b[0;36m_LocIndexer._getitem_axis\u001b[0;34m(self, key, axis)\u001b[0m\n\u001b[1;32m   1430\u001b[0m \u001b[38;5;66;03m# fall thru to straight lookup\u001b[39;00m\n\u001b[1;32m   1431\u001b[0m \u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39m_validate_key(key, axis)\n\u001b[0;32m-> 1432\u001b[0m \u001b[38;5;28;01mreturn\u001b[39;00m \u001b[38;5;28;43mself\u001b[39;49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43m_get_label\u001b[49m\u001b[43m(\u001b[49m\u001b[43mkey\u001b[49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[43maxis\u001b[49m\u001b[38;5;241;43m=\u001b[39;49m\u001b[43maxis\u001b[49m\u001b[43m)\u001b[49m\n",
+      "File \u001b[0;32m~/Documents/ehrapy_clean/ehrapy_venv_feb/lib/python3.11/site-packages/pandas/core/indexing.py:1382\u001b[0m, in \u001b[0;36m_LocIndexer._get_label\u001b[0;34m(self, label, axis)\u001b[0m\n\u001b[1;32m   1380\u001b[0m \u001b[38;5;28;01mdef\u001b[39;00m \u001b[38;5;21m_get_label\u001b[39m(\u001b[38;5;28mself\u001b[39m, label, axis: AxisInt):\n\u001b[1;32m   1381\u001b[0m     \u001b[38;5;66;03m# GH#5567 this will fail if the label is not present in the axis.\u001b[39;00m\n\u001b[0;32m-> 1382\u001b[0m     \u001b[38;5;28;01mreturn\u001b[39;00m \u001b[38;5;28;43mself\u001b[39;49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43mobj\u001b[49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43mxs\u001b[49m\u001b[43m(\u001b[49m\u001b[43mlabel\u001b[49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[43maxis\u001b[49m\u001b[38;5;241;43m=\u001b[39;49m\u001b[43maxis\u001b[49m\u001b[43m)\u001b[49m\n",
+      "File \u001b[0;32m~/Documents/ehrapy_clean/ehrapy_venv_feb/lib/python3.11/site-packages/pandas/core/generic.py:4295\u001b[0m, in \u001b[0;36mNDFrame.xs\u001b[0;34m(self, key, axis, level, drop_level)\u001b[0m\n\u001b[1;32m   4293\u001b[0m             new_index \u001b[38;5;241m=\u001b[39m index[loc]\n\u001b[1;32m   4294\u001b[0m \u001b[38;5;28;01melse\u001b[39;00m:\n\u001b[0;32m-> 4295\u001b[0m     loc \u001b[38;5;241m=\u001b[39m \u001b[43mindex\u001b[49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43mget_loc\u001b[49m\u001b[43m(\u001b[49m\u001b[43mkey\u001b[49m\u001b[43m)\u001b[49m\n\u001b[1;32m   4297\u001b[0m     \u001b[38;5;28;01mif\u001b[39;00m \u001b[38;5;28misinstance\u001b[39m(loc, np\u001b[38;5;241m.\u001b[39mndarray):\n\u001b[1;32m   4298\u001b[0m         \u001b[38;5;28;01mif\u001b[39;00m loc\u001b[38;5;241m.\u001b[39mdtype \u001b[38;5;241m==\u001b[39m np\u001b[38;5;241m.\u001b[39mbool_:\n",
+      "File \u001b[0;32m~/Documents/ehrapy_clean/ehrapy_venv_feb/lib/python3.11/site-packages/pandas/core/indexes/range.py:415\u001b[0m, in \u001b[0;36mRangeIndex.get_loc\u001b[0;34m(self, key)\u001b[0m\n\u001b[1;32m    413\u001b[0m         \u001b[38;5;28;01mreturn\u001b[39;00m \u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39m_range\u001b[38;5;241m.\u001b[39mindex(new_key)\n\u001b[1;32m    414\u001b[0m     \u001b[38;5;28;01mexcept\u001b[39;00m \u001b[38;5;167;01mValueError\u001b[39;00m \u001b[38;5;28;01mas\u001b[39;00m err:\n\u001b[0;32m--> 415\u001b[0m         \u001b[38;5;28;01mraise\u001b[39;00m \u001b[38;5;167;01mKeyError\u001b[39;00m(key) \u001b[38;5;28;01mfrom\u001b[39;00m \u001b[38;5;21;01merr\u001b[39;00m\n\u001b[1;32m    416\u001b[0m \u001b[38;5;28;01mif\u001b[39;00m \u001b[38;5;28misinstance\u001b[39m(key, Hashable):\n\u001b[1;32m    417\u001b[0m     \u001b[38;5;28;01mraise\u001b[39;00m \u001b[38;5;167;01mKeyError\u001b[39;00m(key)\n",
+      "\u001b[0;31mKeyError\u001b[0m: 0"
+     ]
+    },
     {
      "data": {
+      "image/png": "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",
       "text/plain": [
-       "{'male': [56.0, 56.0], 'female': [44.0, 44.0]}"
+       "<Figure size 700x700 with 2 Axes>"
       ]
      },
-     "execution_count": 191,
      "metadata": {},
-     "output_type": "execute_result"
+     "output_type": "display_data"
     }
    ],
    "source": [
-    "data_gender = {key: value for key, value in collated_data.items() if key in [\"male\", \"female\"]}\n",
-    "data_gender"
+    "fig, axes = pl.plot_population_change(return_plot=True)"
    ]
   },
   {
    "cell_type": "code",
-   "execution_count": 248,
+   "execution_count": 67,
    "metadata": {},
    "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "\u001b[1;35m2024-02-14 11:07:38,156\u001b[0m - \u001b[1;34mroot\u001b[0m \u001b[1;37mINFO - Transformed passed DataFrame into an AnnData object with n_obs x n_vars = `101766` x `47`.\u001b[0m\n"
+     ]
+    },
     {
      "data": {
+      "image/svg+xml": [
+       "<?xml version=\"1.0\" encoding=\"UTF-8\" standalone=\"no\"?>\n",
+       "<!DOCTYPE svg PUBLIC \"-//W3C//DTD SVG 1.1//EN\"\n",
+       " \"http://www.w3.org/Graphics/SVG/1.1/DTD/svg11.dtd\">\n",
+       "<!-- Generated by graphviz version 9.0.0 (20230911.1827)\n",
+       " -->\n",
+       "<!-- Pages: 1 -->\n",
+       "<svg width=\"208pt\" height=\"143pt\"\n",
+       " viewBox=\"0.00 0.00 207.62 142.50\" xmlns=\"http://www.w3.org/2000/svg\" xmlns:xlink=\"http://www.w3.org/1999/xlink\">\n",
+       "<g id=\"graph0\" class=\"graph\" transform=\"scale(1 1) rotate(0) translate(4 138.5)\">\n",
+       "<polygon fill=\"white\" stroke=\"none\" points=\"-4,4 -4,-138.5 203.62,-138.5 203.62,4 -4,4\"/>\n",
+       "<!-- 0 -->\n",
+       "<g id=\"node1\" class=\"node\">\n",
+       "<title>0</title>\n",
+       "<polygon fill=\"lightgrey\" stroke=\"black\" points=\"100.75,-134.5 0,-134.5 0,-93.5 100.75,-93.5 100.75,-134.5\"/>\n",
+       "<text text-anchor=\"middle\" x=\"50.38\" y=\"-117.2\" font-family=\"Times,serif\" font-size=\"14.00\">Baseline cohort</text>\n",
+       "<text text-anchor=\"middle\" x=\"50.38\" y=\"-100.7\" font-family=\"Times,serif\" font-size=\"14.00\"> (n=101766)</text>\n",
+       "</g>\n",
+       "<!-- 1 -->\n",
+       "<g id=\"node2\" class=\"node\">\n",
+       "<title>1</title>\n",
+       "<polygon fill=\"lightgrey\" stroke=\"black\" points=\"98.5,-41 2.25,-41 2.25,0 98.5,0 98.5,-41\"/>\n",
+       "<text text-anchor=\"middle\" x=\"50.38\" y=\"-23.7\" font-family=\"Times,serif\" font-size=\"14.00\">Filtered cohort</text>\n",
+       "<text text-anchor=\"middle\" x=\"50.38\" y=\"-7.2\" font-family=\"Times,serif\" font-size=\"14.00\"> (n=1000)</text>\n",
+       "</g>\n",
+       "<!-- 0&#45;&gt;1 -->\n",
+       "<g id=\"edge1\" class=\"edge\">\n",
+       "<title>0&#45;&gt;1</title>\n",
+       "<path fill=\"none\" stroke=\"black\" d=\"M50.38,-93.15C50.38,-81.28 50.38,-65.94 50.38,-52.5\"/>\n",
+       "<polygon fill=\"black\" stroke=\"black\" points=\"53.88,-52.83 50.38,-42.83 46.88,-52.83 53.88,-52.83\"/>\n",
+       "<text text-anchor=\"middle\" x=\"125\" y=\"-62.2\" font-family=\"Times,serif\" font-size=\"14.00\">filtered to first 1000 entries</text>\n",
+       "</g>\n",
+       "</g>\n",
+       "</svg>\n"
+      ],
       "text/plain": [
-       "{'high prof.': [52.0, 48.0], 'limited prof.': [48.0, 52.0]}"
+       "<graphviz.graphs.Digraph at 0x2ebecee50>"
       ]
      },
-     "execution_count": 248,
+     "execution_count": 67,
      "metadata": {},
      "output_type": "execute_result"
     }
    ],
    "source": [
-    "data_prof = {key: value for key, value in collated_data.items() if key in [\"high prof.\", \"limited prof.\"]}\n",
-    "data_prof"
+    "adata = ep.dt.diabetes_130(columns_obs_only=[\"gender\", \"race\", \"weight\", \"age\"])\n",
+    "pop_track = ep.tl.PopulationTracker(adata)\n",
+    "pop_track(adata, label=\"Baseline cohort\")\n",
+    "adata = adata[:1000]\n",
+    "pop_track(adata, label=\"Filtered cohort\", operations_done=\"filtered to first 1000 entries\")\n",
+    "pop_track.plot_flowchart(save=\"flowchart.png\")"
    ]
   },
   {
    "cell_type": "code",
-   "execution_count": 192,
+   "execution_count": 68,
    "metadata": {},
    "outputs": [
     {
-     "data": {
-      "text/plain": [
-       "{'male': [56.0], 'female': [44.0]}"
-      ]
-     },
-     "execution_count": 192,
-     "metadata": {},
-     "output_type": "execute_result"
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "\u001b[1;35m2024-02-14 11:07:40,029\u001b[0m - \u001b[1;34mroot\u001b[0m \u001b[1;37mINFO - Transformed passed DataFrame into an AnnData object with n_obs x n_vars = `101766` x `47`.\u001b[0m\n"
+     ]
     }
    ],
    "source": [
-    "{'male': [56.0], 'female': [44.0]}"
+    "adata = ep.dt.diabetes_130(columns_obs_only=[\"gender\", \"race\", \"weight\", \"age\"])\n",
+    "adata.obs.gender = adata.obs.gender.astype(\"category\")"
    ]
   },
   {
    "cell_type": "code",
-   "execution_count": 171,
+   "execution_count": 69,
    "metadata": {},
    "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "\u001b[1;35m2024-02-14 11:07:40,619\u001b[0m - \u001b[1;34mroot\u001b[0m \u001b[1;37mINFO - Transformed passed DataFrame into an AnnData object with n_obs x n_vars = `101766` x `47`.\u001b[0m\n"
+     ]
+    },
     {
      "data": {
-      "image/png": "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",
       "text/plain": [
-       "<Figure size 1000x500 with 2 Axes>"
+       "weight\n",
+       "[75-100)     1336\n",
+       "[50-75)       897\n",
+       "[100-125)     625\n",
+       "[125-150)     145\n",
+       "[25-50)        97\n",
+       "[0-25)         48\n",
+       "[150-175)      35\n",
+       "[175-200)      11\n",
+       ">200            3\n",
+       "Name: count, dtype: int64"
       ]
      },
-     "metadata": {
-      "image/png": {
-       "height": 488,
-       "width": 989
-      }
-     },
-     "output_type": "display_data"
+     "execution_count": 69,
+     "metadata": {},
+     "output_type": "execute_result"
     }
    ],
    "source": [
-    "# Convert dictionary to DataFrame\n",
-    "df = pd.DataFrame(data_ethnicity)\n",
-    "\n",
-    "# Transpose DataFrame\n",
-    "df= df.T\n",
-    "\n",
-    "# Plotting\n",
-    "fig, axes = plt.subplots(1, 2, figsize=(10, 5))\n",
-    "\n",
-    "# First plot for the first numbers\n",
-    "df.iloc[:,0].plot(kind='barh', stacked=True, ax=axes[0], color=['skyblue', 'lightcoral', 'lightgreen'])\n",
-    "axes[0].set_title('First Numbers')\n",
-    "axes[0].set_xlabel('Percentage')\n",
-    "axes[0].set_ylabel('Ethnicity')\n",
-    "\n",
-    "# Second plot for the second numbers\n",
-    "df.iloc[:,0].plot(kind='barh', stacked=True, ax=axes[1], color=['skyblue', 'lightcoral', 'lightgreen'])\n",
-    "axes[1].set_title('Second Numbers')\n",
-    "axes[1].set_xlabel('Percentage')\n",
-    "axes[1].set_ylabel('Ethnicity')\n",
-    "\n",
-    "plt.tight_layout()\n",
-    "plt.show()"
+    "adata = ep.dt.diabetes_130(columns_obs_only=[\"gender\", \"race\", \"weight\", \"age\"])\n",
+    "adata.obs.weight.value_counts()"
    ]
   },
   {
    "cell_type": "code",
-   "execution_count": 180,
+   "execution_count": 70,
    "metadata": {},
    "outputs": [
     {
      "data": {
       "text/plain": [
-       "Index(['asian', 'black', 'white'], dtype='object')"
+       "array([55629189, 'Emergency', 'Discharged to home', ' Emergency Room', 3,\n",
+       "       nan, nan, 59, 0, 18, 0, 0, 0,\n",
+       "       'endocrine/nutritional/metabolic diseases and immunity disorders',\n",
+       "       'diabetes',\n",
+       "       'endocrine/nutritional/metabolic diseases and immunity disorders',\n",
+       "       9, nan, nan, 'No', 'No', 'No', 'No', 'No', 'No', 'No', 'No', 'No',\n",
+       "       'No', 'No', 'No', 'No', 'No', 'No', 'No', 'No', 'Up', 'No', 'No',\n",
+       "       'No', 'No', 'No', True, True, '>30', 15.0, nan], dtype=object)"
       ]
      },
-     "execution_count": 180,
+     "execution_count": 70,
      "metadata": {},
      "output_type": "execute_result"
     }
    ],
    "source": [
-    "df.index"
+    "adata.X[1, :]"
    ]
   },
   {
    "cell_type": "code",
-   "execution_count": 189,
+   "execution_count": 71,
    "metadata": {},
    "outputs": [
     {
-     "data": {
-      "image/png": "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",
-      "text/plain": [
-       "<Figure size 1000x500 with 1 Axes>"
-      ]
-     },
-     "metadata": {
-      "image/png": {
-       "height": 488,
-       "width": 990
-      }
-     },
-     "output_type": "display_data"
+     "ename": "SyntaxError",
+     "evalue": "unmatched ']' (2079747778.py, line 2)",
+     "output_type": "error",
+     "traceback": [
+      "\u001b[0;36m  Cell \u001b[0;32mIn[71], line 2\u001b[0;36m\u001b[0m\n\u001b[0;31m    adata[]]\u001b[0m\n\u001b[0m           ^\u001b[0m\n\u001b[0;31mSyntaxError\u001b[0m\u001b[0;31m:\u001b[0m unmatched ']'\n"
+     ]
     }
    ],
    "source": [
-    "import pandas as pd\n",
-    "import matplotlib.pyplot as plt\n",
-    "\n",
-    "# Given dictionary\n",
-    "data = {'asian': [28.7],\n",
-    "        'black': [34.7],\n",
-    "        'white': [36.7]}\n",
-    "\n",
-    "# Convert dictionary to DataFrame\n",
-    "df = pd.DataFrame(data)\n",
-    "\n",
-    "# Plotting\n",
-    "fig, axes = plt.subplots(figsize=(10, 5))\n",
-    "\n",
-    "# Transpose DataFrame\n",
-    "#df = df.T\n",
-    "\n",
-    "# Plotting for the first subplot\n",
-    "df.plot(kind='barh', stacked=True, ax=axes, color=['skyblue', 'lightcoral', 'lightgreen'])\n",
-    "axes.set_title('First Numbers')\n",
-    "axes.set_xlabel('Percentage')\n",
-    "axes.set_ylabel('Ethnicity')\n",
-    "\n",
-    "# # Plotting for the second subplot\n",
-    "# df.plot(kind='barh', stacked=True, ax=axes[1], color=['skyblue', 'lightcoral', 'lightgreen'])\n",
-    "# axes[1].set_title('Second Numbers')\n",
-    "# axes[1].set_xlabel('Percentage')\n",
-    "# axes[1].set_ylabel('Ethnicity')\n",
-    "\n",
-    "plt.tight_layout()\n",
-    "plt.show()\n"
+    "adata = ep.dt.diabetes_130()\n",
+    "adata[]]"
    ]
   },
   {
    "cell_type": "code",
-   "execution_count": 251,
+   "execution_count": 253,
    "metadata": {},
    "outputs": [
     {
      "data": {
-      "image/png": "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",
       "text/plain": [
-       "<Figure size 1000x500 with 1 Axes>"
+       "array([[1],\n",
+       "       [3],\n",
+       "       [2],\n",
+       "       ...,\n",
+       "       [1],\n",
+       "       [10],\n",
+       "       [6]], dtype=object)"
       ]
      },
-     "metadata": {
-      "image/png": {
-       "height": 488,
-       "width": 990
-      }
-     },
-     "output_type": "display_data"
+     "execution_count": 253,
+     "metadata": {},
+     "output_type": "execute_result"
     }
    ],
    "source": [
-    "import pandas as pd\n",
-    "import matplotlib.pyplot as plt\n",
-    "\n",
-    "# Given data dictionaries\n",
-    "data_ethnicity = {'asian': [28.7],\n",
-    "                  'black': [34.7],\n",
-    "                  'white': [36.7]}\n",
-    "\n",
-    "data_gender = {'male': [56.0],\n",
-    "               'female': [44.0]}\n",
-    "\n",
-    "data_prof = {'high prof.': [52.0], 'limited prof.': [48.0]}\n",
-    "\n",
-    "# Convert dictionaries to DataFrames\n",
-    "df_ethnicity = pd.DataFrame(data_ethnicity)\n",
-    "df_gender = pd.DataFrame(data_gender)\n",
-    "\n",
-    "# Plotting\n",
-    "fig, ax = plt.subplots(figsize=(10, 5))\n",
-    "\n",
-    "# Plotting for the ethnicity\n",
-    "df_ethnicity.plot(kind='barh', stacked=True, ax=ax, color=['skyblue', 'lightcoral', 'lightgreen'], position=0.1, width=0.2)\n",
-    "\n",
-    "# Plotting for the gender with a small gap\n",
-    "df_gender.plot(kind='barh', stacked=True, ax=ax, color=['skyblue', 'lightcoral'], position=1.5, width=0.2)\n",
-    "\n",
-    "# Set labels and title\n",
-    "ax.set_title('Ethnicity and Gender')\n",
-    "ax.set_xlabel('Percentage')\n",
-    "ax.set_ylabel('Category')\n",
-    "\n",
-    "# Add a legend\n",
-    "ax.legend(['Asian', 'Black', 'White', 'Male', 'Female'])\n",
-    "\n",
-    "plt.tight_layout()\n",
-    "plt.show()\n"
+    "adata.X[:, adata.var_names == \"time_in_hospital_days\"]"
    ]
   },
   {
    "cell_type": "code",
-   "execution_count": 211,
+   "execution_count": null,
+   "metadata": {},
+   "outputs": [],
+   "source": []
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 78,
    "metadata": {},
    "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "\u001b[1;35m2024-02-14 11:15:44,787\u001b[0m - \u001b[1;34mroot\u001b[0m \u001b[1;37mINFO - Transformed passed DataFrame into an AnnData object with n_obs x n_vars = `101766` x `47`.\u001b[0m\n",
+      "gender                    float64\n",
+      "race                     category\n",
+      "weight                   category\n",
+      "time_in_hospital_days     float64\n",
+      "dtype: object\n"
+     ]
+    },
     {
      "data": {
       "text/html": [
@@ -779,645 +796,801 @@
        "  <thead>\n",
        "    <tr style=\"text-align: right;\">\n",
        "      <th></th>\n",
-       "      <th>asian</th>\n",
-       "      <th>black</th>\n",
-       "      <th>white</th>\n",
+       "      <th></th>\n",
+       "      <th>Missing</th>\n",
+       "      <th>Overall</th>\n",
        "    </tr>\n",
        "  </thead>\n",
        "  <tbody>\n",
        "    <tr>\n",
-       "      <th>0</th>\n",
-       "      <td>28.7</td>\n",
-       "      <td>34.7</td>\n",
-       "      <td>36.7</td>\n",
+       "      <th>n</th>\n",
+       "      <th></th>\n",
+       "      <td></td>\n",
+       "      <td>101766</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th rowspan=\"2\" valign=\"top\">gender, n (%)</th>\n",
+       "      <th>0.0</th>\n",
+       "      <td>3</td>\n",
+       "      <td>54708 (53.8)</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>1.0</th>\n",
+       "      <td></td>\n",
+       "      <td>47055 (46.2)</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th rowspan=\"5\" valign=\"top\">race, n (%)</th>\n",
+       "      <th>AfricanAmerican</th>\n",
+       "      <td>2273</td>\n",
+       "      <td>19210 (19.3)</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>Asian</th>\n",
+       "      <td></td>\n",
+       "      <td>641 (0.6)</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>Caucasian</th>\n",
+       "      <td></td>\n",
+       "      <td>76099 (76.5)</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>Hispanic</th>\n",
+       "      <td></td>\n",
+       "      <td>2037 (2.0)</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>Other</th>\n",
+       "      <td></td>\n",
+       "      <td>1506 (1.5)</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th rowspan=\"9\" valign=\"top\">weight, n (%)</th>\n",
+       "      <th>&gt;200</th>\n",
+       "      <td>98569</td>\n",
+       "      <td>3 (0.1)</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>[0-25)</th>\n",
+       "      <td></td>\n",
+       "      <td>48 (1.5)</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>[100-125)</th>\n",
+       "      <td></td>\n",
+       "      <td>625 (19.5)</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>[125-150)</th>\n",
+       "      <td></td>\n",
+       "      <td>145 (4.5)</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>[150-175)</th>\n",
+       "      <td></td>\n",
+       "      <td>35 (1.1)</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>[175-200)</th>\n",
+       "      <td></td>\n",
+       "      <td>11 (0.3)</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>[25-50)</th>\n",
+       "      <td></td>\n",
+       "      <td>97 (3.0)</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>[50-75)</th>\n",
+       "      <td></td>\n",
+       "      <td>897 (28.1)</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>[75-100)</th>\n",
+       "      <td></td>\n",
+       "      <td>1336 (41.8)</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>time_in_hospital_days, mean (SD)</th>\n",
+       "      <th></th>\n",
+       "      <td>0</td>\n",
+       "      <td>4.9 (3.0)</td>\n",
        "    </tr>\n",
        "  </tbody>\n",
        "</table>\n",
-       "</div>"
+       "</div><br />"
       ],
       "text/plain": [
-       "   asian  black  white\n",
-       "0   28.7   34.7   36.7"
+       "                                                 Missing       Overall\n",
+       "n                                                               101766\n",
+       "gender, n (%)                    0.0                   3  54708 (53.8)\n",
+       "                                 1.0                      47055 (46.2)\n",
+       "race, n (%)                      AfricanAmerican    2273  19210 (19.3)\n",
+       "                                 Asian                       641 (0.6)\n",
+       "                                 Caucasian                76099 (76.5)\n",
+       "                                 Hispanic                   2037 (2.0)\n",
+       "                                 Other                      1506 (1.5)\n",
+       "weight, n (%)                    >200              98569       3 (0.1)\n",
+       "                                 [0-25)                       48 (1.5)\n",
+       "                                 [100-125)                  625 (19.5)\n",
+       "                                 [125-150)                   145 (4.5)\n",
+       "                                 [150-175)                    35 (1.1)\n",
+       "                                 [175-200)                    11 (0.3)\n",
+       "                                 [25-50)                      97 (3.0)\n",
+       "                                 [50-75)                    897 (28.1)\n",
+       "                                 [75-100)                  1336 (41.8)\n",
+       "time_in_hospital_days, mean (SD)                       0     4.9 (3.0)"
       ]
      },
-     "execution_count": 211,
+     "execution_count": 78,
      "metadata": {},
      "output_type": "execute_result"
     }
    ],
    "source": [
-    "df_ethnicity"
+    "adata = ep.dt.diabetes_130(columns_obs_only=[\"gender\", \"race\", \"weight\", \"time_in_hospital_days\"])\n",
+    "adata.obs.time_in_hospital_days = adata.obs.time_in_hospital_days.astype(\"float\") + np.random.random(adata.n_obs)\n",
+    "print(adata.obs.dtypes)\n",
+    "TableOne(adata.obs)"
    ]
   },
   {
    "cell_type": "code",
-   "execution_count": 215,
+   "execution_count": 156,
    "metadata": {},
    "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "\u001b[1;35m2024-02-14 13:01:31,453\u001b[0m - \u001b[1;34mroot\u001b[0m \u001b[1;37mINFO - Transformed passed DataFrame into an AnnData object with n_obs x n_vars = `101766` x `48`.\u001b[0m\n"
+     ]
+    },
     {
      "data": {
-      "image/png": "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",
+      "image/png": "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",
       "text/plain": [
-       "<Figure size 1000x500 with 1 Axes>"
+       "<Figure size 700x700 with 2 Axes>"
       ]
      },
-     "metadata": {
-      "image/png": {
-       "height": 488,
-       "width": 990
-      }
-     },
+     "metadata": {},
      "output_type": "display_data"
     }
    ],
    "source": [
-    "import pandas as pd\n",
-    "import matplotlib.pyplot as plt\n",
-    "\n",
-    "# Given data dictionaries\n",
-    "data_ethnicity = {'asian': [28.7],\n",
-    "                  'black': [34.7],\n",
-    "                  'white': [36.7]}\n",
-    "\n",
-    "data_gender = {'male': [56.0],\n",
-    "               'female': [44.0]}\n",
-    "\n",
-    "# Convert dictionaries to DataFrames\n",
-    "df_ethnicity = pd.DataFrame(data_ethnicity)\n",
-    "df_gender = pd.DataFrame(data_gender)\n",
-    "\n",
-    "# Plotting\n",
-    "fig, ax = plt.subplots(figsize=(10, 5))\n",
-    "\n",
-    "# Plotting for the ethnicity with a small gap\n",
-    "df_ethnicity.plot(kind='barh', stacked=True, ax=ax, color=['skyblue', 'lightcoral', 'lightgreen'], position=0.67, width=0.3)\n",
-    "\n",
-    "# Plotting for the gender with a small gap\n",
-    "df_gender.plot(kind='barh', stacked=True, ax=ax, color=['skyblue', 'lightcoral'], position=0.33, width=0.3)\n",
-    "\n",
-    "\n",
-    "# Set labels and title\n",
-    "ax.set_title('Ethnicity and Gender')\n",
-    "ax.set_xlabel('Percentage')\n",
-    "ax.set_ylabel('Category')\n",
-    "\n",
-    "# Add a legend\n",
-    "ax.legend(['Asian', 'Black', 'White', 'Male', 'Female'])\n",
-    "\n",
-    "plt.tight_layout()\n",
-    "plt.show()\n"
+    "adata = ep.dt.diabetes_130(columns_obs_only=[\"gender\", \"race\", \"time_in_hospital_days\"])\n",
+    "pop_track = ep.tl.PopulationTracker(adata, categorical=[\"gender\", \"race\"])\n",
+    "pop_track(adata, label=\"Initial cohort\")\n",
+    "adata = adata[:1000]\n",
+    "pop_track(adata, label=\"Cohort 1\", operations_done=\"filtered to first 1000 entries\")\n",
+    "pop_track.plot_population_change(\n",
+    "    subfigure_title=True, subplots_kwargs={\"figsize\": (7, 7)}, legend_kwargs={\"bbox_to_anchor\": (1, 1)}\n",
+    ")"
    ]
   },
   {
    "cell_type": "code",
-   "execution_count": 235,
+   "execution_count": 127,
    "metadata": {},
-   "outputs": [],
+   "outputs": [
+    {
+     "data": {
+      "image/svg+xml": [
+       "<?xml version=\"1.0\" encoding=\"UTF-8\" standalone=\"no\"?>\n",
+       "<!DOCTYPE svg PUBLIC \"-//W3C//DTD SVG 1.1//EN\"\n",
+       " \"http://www.w3.org/Graphics/SVG/1.1/DTD/svg11.dtd\">\n",
+       "<!-- Generated by graphviz version 9.0.0 (20230911.1827)\n",
+       " -->\n",
+       "<!-- Pages: 1 -->\n",
+       "<svg width=\"201pt\" height=\"143pt\"\n",
+       " viewBox=\"0.00 0.00 200.50 142.50\" xmlns=\"http://www.w3.org/2000/svg\" xmlns:xlink=\"http://www.w3.org/1999/xlink\">\n",
+       "<g id=\"graph0\" class=\"graph\" transform=\"scale(1 1) rotate(0) translate(4 138.5)\">\n",
+       "<polygon fill=\"white\" stroke=\"none\" points=\"-4,4 -4,-138.5 196.5,-138.5 196.5,4 -4,4\"/>\n",
+       "<!-- 0 -->\n",
+       "<g id=\"node1\" class=\"node\">\n",
+       "<title>0</title>\n",
+       "<polygon fill=\"lightgrey\" stroke=\"black\" points=\"86.5,-134.5 0,-134.5 0,-93.5 86.5,-93.5 86.5,-134.5\"/>\n",
+       "<text text-anchor=\"middle\" x=\"43.25\" y=\"-117.2\" font-family=\"Times,serif\" font-size=\"14.00\">Initial cohort</text>\n",
+       "<text text-anchor=\"middle\" x=\"43.25\" y=\"-100.7\" font-family=\"Times,serif\" font-size=\"14.00\"> (n=101766)</text>\n",
+       "</g>\n",
+       "<!-- 1 -->\n",
+       "<g id=\"node2\" class=\"node\">\n",
+       "<title>1</title>\n",
+       "<polygon fill=\"lightgrey\" stroke=\"black\" points=\"78.62,-41 7.88,-41 7.88,0 78.62,0 78.62,-41\"/>\n",
+       "<text text-anchor=\"middle\" x=\"43.25\" y=\"-23.7\" font-family=\"Times,serif\" font-size=\"14.00\">Cohort 1</text>\n",
+       "<text text-anchor=\"middle\" x=\"43.25\" y=\"-7.2\" font-family=\"Times,serif\" font-size=\"14.00\"> (n=1000)</text>\n",
+       "</g>\n",
+       "<!-- 0&#45;&gt;1 -->\n",
+       "<g id=\"edge1\" class=\"edge\">\n",
+       "<title>0&#45;&gt;1</title>\n",
+       "<path fill=\"none\" stroke=\"black\" d=\"M43.25,-93.15C43.25,-81.28 43.25,-65.94 43.25,-52.5\"/>\n",
+       "<polygon fill=\"black\" stroke=\"black\" points=\"46.75,-52.83 43.25,-42.83 39.75,-52.83 46.75,-52.83\"/>\n",
+       "<text text-anchor=\"middle\" x=\"117.88\" y=\"-62.2\" font-family=\"Times,serif\" font-size=\"14.00\">filtered to first 1000 entries</text>\n",
+       "</g>\n",
+       "</g>\n",
+       "</svg>\n"
+      ],
+      "text/plain": [
+       "<graphviz.graphs.Digraph at 0x2c950aa50>"
+      ]
+     },
+     "execution_count": 127,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
    "source": [
-    "l = df_ethnicity.T"
+    "pop_track.plot_flowchart()"
    ]
   },
   {
    "cell_type": "code",
-   "execution_count": 242,
+   "execution_count": 107,
    "metadata": {},
    "outputs": [
     {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      "28.7\n",
-      "34.7\n",
-      "36.7\n"
-     ]
+     "data": {
+      "text/plain": [
+       "{'gender': {0.0: [53.8, 52.4], 1.0: [46.2, 47.6]},\n",
+       " 'race': {'AfricanAmerican': [19.3, 25.9],\n",
+       "  'Asian': [0.6, 0.7],\n",
+       "  'Caucasian': [76.5, 70.4],\n",
+       "  'Hispanic': [2.0, 0.8],\n",
+       "  'Other': [1.5, 2.2]},\n",
+       " 'weight': {'>200': [0.1, 0],\n",
+       "  '[0-25)': [1.5, 0],\n",
+       "  '[100-125)': [19.5, 0],\n",
+       "  '[125-150)': [4.5, 0],\n",
+       "  '[150-175)': [1.1, 0],\n",
+       "  '[175-200)': [0.3, 0],\n",
+       "  '[25-50)': [3.0, 0],\n",
+       "  '[50-75)': [28.1, 0],\n",
+       "  '[75-100)': [41.8, 0]},\n",
+       " 'time_in_hospital_days': ['4.4 (3.0)', '4.6 (3.2)']}"
+      ]
+     },
+     "execution_count": 107,
+     "metadata": {},
+     "output_type": "execute_result"
     }
    ],
    "source": [
-    "for v in l.iterrows():\n",
-    "    print(v[1].values[0])"
+    "pop_track.track"
    ]
   },
   {
    "cell_type": "code",
-   "execution_count": 232,
+   "execution_count": 116,
    "metadata": {},
    "outputs": [
     {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      "[28.7 34.7 36.7]\n"
-     ]
+     "data": {
+      "image/png": "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",
+      "text/plain": [
+       "<Figure size 700x700 with 2 Axes>"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
     }
    ],
    "source": [
-    "for u in df_ethnicity.values:\n",
-    "    print(u)"
+    "pop_track.plot_population_change()"
    ]
   },
   {
    "cell_type": "code",
-   "execution_count": 233,
+   "execution_count": 99,
    "metadata": {},
    "outputs": [
     {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      "0\n"
-     ]
+     "data": {
+      "text/plain": [
+       "{'gender': {0.0: [53.8, 52.4], 1.0: [46.2, 47.6]},\n",
+       " 'race': {'AfricanAmerican': [19.3, 25.9],\n",
+       "  'Asian': [0.6, 0.7],\n",
+       "  'Caucasian': [76.5, 70.4],\n",
+       "  'Hispanic': [2.0, 0.8],\n",
+       "  'Other': [1.5, 2.2]},\n",
+       " 'weight': {'>200': [0.1, 0],\n",
+       "  '[0-25)': [1.5, 0],\n",
+       "  '[100-125)': [19.5, 0],\n",
+       "  '[125-150)': [4.5, 0],\n",
+       "  '[150-175)': [1.1, 0],\n",
+       "  '[175-200)': [0.3, 0],\n",
+       "  '[25-50)': [3.0, 0],\n",
+       "  '[50-75)': [28.1, 0],\n",
+       "  '[75-100)': [41.8, 0]},\n",
+       " 'time_in_hospital_days': ['4.4 (3.0)', '4.6 (3.2)']}"
+      ]
+     },
+     "execution_count": 99,
+     "metadata": {},
+     "output_type": "execute_result"
     }
    ],
    "source": [
-    "for i, value in enumerate(df_ethnicity.T):\n",
-    "    print(value)"
+    "pop_track.track"
    ]
   },
   {
    "cell_type": "code",
-   "execution_count": 271,
+   "execution_count": 97,
    "metadata": {},
    "outputs": [
     {
      "name": "stdout",
      "output_type": "stream",
      "text": [
-      "28.7\n",
-      "34.7\n",
-      "36.7\n",
-      "56.0\n",
-      "44.0\n",
-      "52.0\n",
-      "48.0\n"
+      "\u001b[1;35m2024-02-14 11:56:20,020\u001b[0m - \u001b[1;34mroot\u001b[0m \u001b[1;37mINFO - Transformed passed DataFrame into an AnnData object with n_obs x n_vars = `101766` x `47`.\u001b[0m\n"
+     ]
+    },
+    {
+     "ename": "KeyError",
+     "evalue": "0",
+     "output_type": "error",
+     "traceback": [
+      "\u001b[0;31m---------------------------------------------------------------------------\u001b[0m",
+      "\u001b[0;31mValueError\u001b[0m                                Traceback (most recent call last)",
+      "File \u001b[0;32m~/Documents/ehrapy_clean/ehrapy_venv_feb/lib/python3.11/site-packages/pandas/core/indexes/range.py:413\u001b[0m, in \u001b[0;36mRangeIndex.get_loc\u001b[0;34m(self, key)\u001b[0m\n\u001b[1;32m    412\u001b[0m \u001b[38;5;28;01mtry\u001b[39;00m:\n\u001b[0;32m--> 413\u001b[0m     \u001b[38;5;28;01mreturn\u001b[39;00m \u001b[38;5;28;43mself\u001b[39;49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43m_range\u001b[49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43mindex\u001b[49m\u001b[43m(\u001b[49m\u001b[43mnew_key\u001b[49m\u001b[43m)\u001b[49m\n\u001b[1;32m    414\u001b[0m \u001b[38;5;28;01mexcept\u001b[39;00m \u001b[38;5;167;01mValueError\u001b[39;00m \u001b[38;5;28;01mas\u001b[39;00m err:\n",
+      "\u001b[0;31mValueError\u001b[0m: 0 is not in range",
+      "\nThe above exception was the direct cause of the following exception:\n",
+      "\u001b[0;31mKeyError\u001b[0m                                  Traceback (most recent call last)",
+      "Cell \u001b[0;32mIn[97], line 7\u001b[0m\n\u001b[1;32m      5\u001b[0m adata \u001b[38;5;241m=\u001b[39m adata[:\u001b[38;5;241m1000\u001b[39m]\n\u001b[1;32m      6\u001b[0m pop_track(adata, label\u001b[38;5;241m=\u001b[39m\u001b[38;5;124m\"\u001b[39m\u001b[38;5;124mFiltered cohort\u001b[39m\u001b[38;5;124m\"\u001b[39m, operations_done\u001b[38;5;241m=\u001b[39m\u001b[38;5;124m\"\u001b[39m\u001b[38;5;124mfiltered to first 1000 entries\u001b[39m\u001b[38;5;124m\"\u001b[39m)\n\u001b[0;32m----> 7\u001b[0m \u001b[43mpop_track\u001b[49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43mplot_population_change\u001b[49m\u001b[43m(\u001b[49m\u001b[43m)\u001b[49m\n",
+      "File \u001b[0;32m~/Documents/ehrapy_clean/ehrapy/ehrapy/tools/population_tracking/_pop_tracker.py:173\u001b[0m, in \u001b[0;36mPopulationTracker.plot_population_change\u001b[0;34m(self, set_axis_labels, save, return_plot)\u001b[0m\n\u001b[1;32m    171\u001b[0m \u001b[38;5;28;01mfor\u001b[39;00m idx, ax \u001b[38;5;129;01min\u001b[39;00m \u001b[38;5;28menumerate\u001b[39m(axes):\n\u001b[1;32m    172\u001b[0m     \u001b[38;5;28;01mfor\u001b[39;00m pos, (_cols, data) \u001b[38;5;129;01min\u001b[39;00m \u001b[38;5;28menumerate\u001b[39m(\u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39mtrack\u001b[38;5;241m.\u001b[39mitems()):\n\u001b[0;32m--> 173\u001b[0m         data \u001b[38;5;241m=\u001b[39m \u001b[43mpd\u001b[49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43mDataFrame\u001b[49m\u001b[43m(\u001b[49m\u001b[43mdata\u001b[49m\u001b[43m)\u001b[49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43mloc\u001b[49m\u001b[43m[\u001b[49m\u001b[43midx\u001b[49m\u001b[43m]\u001b[49m\n\u001b[1;32m    175\u001b[0m         cumwidth \u001b[38;5;241m=\u001b[39m \u001b[38;5;241m0\u001b[39m\n\u001b[1;32m    177\u001b[0m         \u001b[38;5;66;03m# Adjust the hue shift based on the category position such that the colors are more distinguishable\u001b[39;00m\n",
+      "File \u001b[0;32m~/Documents/ehrapy_clean/ehrapy_venv_feb/lib/python3.11/site-packages/pandas/core/indexing.py:1192\u001b[0m, in \u001b[0;36m_LocationIndexer.__getitem__\u001b[0;34m(self, key)\u001b[0m\n\u001b[1;32m   1190\u001b[0m maybe_callable \u001b[38;5;241m=\u001b[39m com\u001b[38;5;241m.\u001b[39mapply_if_callable(key, \u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39mobj)\n\u001b[1;32m   1191\u001b[0m maybe_callable \u001b[38;5;241m=\u001b[39m \u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39m_check_deprecated_callable_usage(key, maybe_callable)\n\u001b[0;32m-> 1192\u001b[0m \u001b[38;5;28;01mreturn\u001b[39;00m \u001b[38;5;28;43mself\u001b[39;49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43m_getitem_axis\u001b[49m\u001b[43m(\u001b[49m\u001b[43mmaybe_callable\u001b[49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[43maxis\u001b[49m\u001b[38;5;241;43m=\u001b[39;49m\u001b[43maxis\u001b[49m\u001b[43m)\u001b[49m\n",
+      "File \u001b[0;32m~/Documents/ehrapy_clean/ehrapy_venv_feb/lib/python3.11/site-packages/pandas/core/indexing.py:1432\u001b[0m, in \u001b[0;36m_LocIndexer._getitem_axis\u001b[0;34m(self, key, axis)\u001b[0m\n\u001b[1;32m   1430\u001b[0m \u001b[38;5;66;03m# fall thru to straight lookup\u001b[39;00m\n\u001b[1;32m   1431\u001b[0m \u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39m_validate_key(key, axis)\n\u001b[0;32m-> 1432\u001b[0m \u001b[38;5;28;01mreturn\u001b[39;00m \u001b[38;5;28;43mself\u001b[39;49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43m_get_label\u001b[49m\u001b[43m(\u001b[49m\u001b[43mkey\u001b[49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[43maxis\u001b[49m\u001b[38;5;241;43m=\u001b[39;49m\u001b[43maxis\u001b[49m\u001b[43m)\u001b[49m\n",
+      "File \u001b[0;32m~/Documents/ehrapy_clean/ehrapy_venv_feb/lib/python3.11/site-packages/pandas/core/indexing.py:1382\u001b[0m, in \u001b[0;36m_LocIndexer._get_label\u001b[0;34m(self, label, axis)\u001b[0m\n\u001b[1;32m   1380\u001b[0m \u001b[38;5;28;01mdef\u001b[39;00m \u001b[38;5;21m_get_label\u001b[39m(\u001b[38;5;28mself\u001b[39m, label, axis: AxisInt):\n\u001b[1;32m   1381\u001b[0m     \u001b[38;5;66;03m# GH#5567 this will fail if the label is not present in the axis.\u001b[39;00m\n\u001b[0;32m-> 1382\u001b[0m     \u001b[38;5;28;01mreturn\u001b[39;00m \u001b[38;5;28;43mself\u001b[39;49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43mobj\u001b[49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43mxs\u001b[49m\u001b[43m(\u001b[49m\u001b[43mlabel\u001b[49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[43maxis\u001b[49m\u001b[38;5;241;43m=\u001b[39;49m\u001b[43maxis\u001b[49m\u001b[43m)\u001b[49m\n",
+      "File \u001b[0;32m~/Documents/ehrapy_clean/ehrapy_venv_feb/lib/python3.11/site-packages/pandas/core/generic.py:4295\u001b[0m, in \u001b[0;36mNDFrame.xs\u001b[0;34m(self, key, axis, level, drop_level)\u001b[0m\n\u001b[1;32m   4293\u001b[0m             new_index \u001b[38;5;241m=\u001b[39m index[loc]\n\u001b[1;32m   4294\u001b[0m \u001b[38;5;28;01melse\u001b[39;00m:\n\u001b[0;32m-> 4295\u001b[0m     loc \u001b[38;5;241m=\u001b[39m \u001b[43mindex\u001b[49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43mget_loc\u001b[49m\u001b[43m(\u001b[49m\u001b[43mkey\u001b[49m\u001b[43m)\u001b[49m\n\u001b[1;32m   4297\u001b[0m     \u001b[38;5;28;01mif\u001b[39;00m \u001b[38;5;28misinstance\u001b[39m(loc, np\u001b[38;5;241m.\u001b[39mndarray):\n\u001b[1;32m   4298\u001b[0m         \u001b[38;5;28;01mif\u001b[39;00m loc\u001b[38;5;241m.\u001b[39mdtype \u001b[38;5;241m==\u001b[39m np\u001b[38;5;241m.\u001b[39mbool_:\n",
+      "File \u001b[0;32m~/Documents/ehrapy_clean/ehrapy_venv_feb/lib/python3.11/site-packages/pandas/core/indexes/range.py:415\u001b[0m, in \u001b[0;36mRangeIndex.get_loc\u001b[0;34m(self, key)\u001b[0m\n\u001b[1;32m    413\u001b[0m         \u001b[38;5;28;01mreturn\u001b[39;00m \u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39m_range\u001b[38;5;241m.\u001b[39mindex(new_key)\n\u001b[1;32m    414\u001b[0m     \u001b[38;5;28;01mexcept\u001b[39;00m \u001b[38;5;167;01mValueError\u001b[39;00m \u001b[38;5;28;01mas\u001b[39;00m err:\n\u001b[0;32m--> 415\u001b[0m         \u001b[38;5;28;01mraise\u001b[39;00m \u001b[38;5;167;01mKeyError\u001b[39;00m(key) \u001b[38;5;28;01mfrom\u001b[39;00m \u001b[38;5;21;01merr\u001b[39;00m\n\u001b[1;32m    416\u001b[0m \u001b[38;5;28;01mif\u001b[39;00m \u001b[38;5;28misinstance\u001b[39m(key, Hashable):\n\u001b[1;32m    417\u001b[0m     \u001b[38;5;28;01mraise\u001b[39;00m \u001b[38;5;167;01mKeyError\u001b[39;00m(key)\n",
+      "\u001b[0;31mKeyError\u001b[0m: 0"
      ]
     },
     {
      "data": {
-      "image/png": "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",
+      "image/png": "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",
       "text/plain": [
-       "<Figure size 1000x500 with 1 Axes>"
+       "<Figure size 700x700 with 2 Axes>"
       ]
      },
-     "metadata": {
-      "image/png": {
-       "height": 495,
-       "width": 970
-      }
-     },
+     "metadata": {},
      "output_type": "display_data"
     }
    ],
    "source": [
-    "import pandas as pd\n",
-    "import matplotlib.pyplot as plt\n",
-    "\n",
-    "# Given data dictionaries\n",
-    "data_ethnicity = {'asian': [28.7],\n",
-    "                  'black': [34.7],\n",
-    "                  'white': [36.7]}\n",
-    "\n",
-    "data_gender = {'male': [56.0],\n",
-    "               'female': [44.0]}\n",
-    "\n",
-    "data_prof = {'high prof.': [52.0], 'limited prof.': [48.0]}\n",
-    "\n",
-    "\n",
-    "# Convert dictionaries to DataFrames\n",
-    "df_ethnicity = pd.DataFrame(data_ethnicity)\n",
-    "df_gender = pd.DataFrame(data_gender)\n",
-    "df_prof = pd.DataFrame(data_prof)\n",
-    "\n",
-    "# Plotting\n",
-    "fig, ax = plt.subplots(figsize=(10, 5))\n",
-    "\n",
-    "# Plotting for the ethnicity with a small gap\n",
-    "#df_ethnicity.plot(kind='barh', stacked=True, ax=ax, color=['skyblue', 'lightcoral', 'lightgreen'], position=0.67, width=0.3)\n",
-    "categories = [df_ethnicity, df_gender, df_prof]\n",
-    "category_labels = ['Ethnicity', 'Gender', 'Proficiency']\n",
-    "\n",
-    "colors=['skyblue', 'lightcoral', 'lightgreen'] # TODO: make this dynamic, more colors, pallette per category\n",
-    "legend_labels = []\n",
-    "\n",
-    "for pos, data in enumerate([df_ethnicity.T.iterrows(), df_gender.T.iterrows(), df_prof.T.iterrows()]):\n",
-    "    cumwidth = 0\n",
-    "    category = categories[pos]\n",
-    "\n",
-    "    hue_shift = pos / len(categories)  # Adjust the hue shift based on the category position\n",
-    "    colors = sns.color_palette(\"husl\", len(category.columns))\n",
-    "    adjusted_colors = [((color[0] + hue_shift) % 1, color[1], color[2]) for color in colors]\n",
-    "\n",
-    "    for i, value in enumerate(data):\n",
-    "        value = value[1].values[0]\n",
-    "        print(value)\n",
-    "        ax.barh(pos, value, left=cumwidth, color=adjusted_colors[i], height=0.8)\n",
-    "        \n",
-    "\n",
-    "        if value > 5:\n",
-    "            # Add data labels\n",
-    "            width = value\n",
-    "            ax.text(cumwidth + width/2, pos, '{:.1f}'.format(value), ha='center', va='center', color='white', fontweight='bold')\n",
-    "            \n",
-    "        ax.set_yticks([])\n",
-    "        ax.set_xticks([])\n",
-    "        cumwidth += value\n",
-    "        \n",
-    "        #if pos == 0:\n",
-    "        legend_labels.append(category.columns[i])\n",
-    "        \n",
-    "# Plotting for the gender with a small gap\n",
-    "#df_gender.plot(kind='barh', stacked=True, ax=ax, color=['skyblue', 'lightcoral'], position=0.33, width=0.3)\n",
-    "\n",
-    "# ax.barh(1, df_gender, color=['skyblue', 'lightcoral', 'lightgreen'])\n",
-    "\n",
-    "# # Set labels and title\n",
-    "# ax.set_title('Ethnicity and Gender')\n",
-    "# ax.set_xlabel('Percentage')\n",
-    "# ax.set_ylabel('Category')\n",
-    "\n",
-    "# # Add a legend\n",
-    "# ax.legend(['Asian', 'Black', 'White', 'Male', 'Female'])\n",
-    "            \n",
-    "ax.legend(legend_labels, loc='lower center', ncol=len(legend_labels), bbox_to_anchor=(0.5, -0.1))\n",
-    "\n",
-    "plt.tight_layout()\n",
-    "plt.show()\n"
+    "adata = ep.dt.diabetes_130(columns_obs_only=[\"gender\", \"race\", \"weight\", \"time_in_hospital_days\"])\n",
+    "adata.obs.gender = adata.obs.gender.astype(\"category\")\n",
+    "pop_track = ep.tl.PopulationTracker(adata, categorical=[\"gender\", \"race\", \"weight\"])\n",
+    "pop_track(adata, label=\"Baseline cohort\")\n",
+    "adata = adata[:1000]\n",
+    "pop_track(adata, label=\"Filtered cohort\", operations_done=\"filtered to first 1000 entries\")\n",
+    "pop_track.plot_population_change()"
    ]
   },
   {
    "cell_type": "code",
-   "execution_count": 9,
+   "execution_count": 187,
    "metadata": {},
    "outputs": [
     {
-     "data": {
-      "image/png": "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",
-      "text/plain": [
-       "<Figure size 400x900 with 6 Axes>"
-      ]
-     },
-     "metadata": {
-      "image/png": {
-       "height": 988,
-       "width": 390
-      }
-     },
-     "output_type": "display_data"
-    },
-    {
-     "data": {
-      "text/plain": [
-       "<Figure size 640x480 with 0 Axes>"
-      ]
-     },
-     "metadata": {},
-     "output_type": "display_data"
-    },
-    {
-     "data": {
-      "text/plain": [
-       "<Figure size 640x480 with 0 Axes>"
-      ]
-     },
-     "metadata": {},
-     "output_type": "display_data"
-    },
-    {
-     "data": {
-      "text/plain": [
-       "<Figure size 640x480 with 0 Axes>"
-      ]
-     },
-     "metadata": {},
-     "output_type": "display_data"
-    },
-    {
-     "data": {
-      "text/plain": [
-       "<Figure size 640x480 with 0 Axes>"
-      ]
-     },
-     "metadata": {},
-     "output_type": "display_data"
-    },
-    {
-     "data": {
-      "text/plain": [
-       "<Figure size 640x480 with 0 Axes>"
-      ]
-     },
-     "metadata": {},
-     "output_type": "display_data"
-    },
-    {
-     "data": {
-      "text/plain": [
-       "<Figure size 640x480 with 0 Axes>"
-      ]
-     },
-     "metadata": {},
-     "output_type": "display_data"
-    },
-    {
-     "data": {
-      "text/plain": [
-       "<Figure size 640x480 with 0 Axes>"
-      ]
-     },
-     "metadata": {},
-     "output_type": "display_data"
-    },
-    {
-     "data": {
-      "text/plain": [
-       "<Figure size 640x480 with 0 Axes>"
-      ]
-     },
-     "metadata": {},
-     "output_type": "display_data"
-    },
-    {
-     "data": {
-      "text/plain": [
-       "<Figure size 640x480 with 0 Axes>"
-      ]
-     },
-     "metadata": {},
-     "output_type": "display_data"
-    },
-    {
-     "data": {
-      "text/plain": [
-       "<Figure size 640x480 with 0 Axes>"
-      ]
-     },
-     "metadata": {},
-     "output_type": "display_data"
-    },
-    {
-     "data": {
-      "text/plain": [
-       "<Figure size 640x480 with 0 Axes>"
-      ]
-     },
-     "metadata": {},
-     "output_type": "display_data"
-    },
-    {
-     "data": {
-      "text/plain": [
-       "<Figure size 640x480 with 0 Axes>"
-      ]
-     },
-     "metadata": {},
-     "output_type": "display_data"
-    },
-    {
-     "data": {
-      "text/plain": [
-       "<Figure size 640x480 with 0 Axes>"
-      ]
-     },
-     "metadata": {},
-     "output_type": "display_data"
-    },
-    {
-     "data": {
-      "text/plain": [
-       "<Figure size 640x480 with 0 Axes>"
-      ]
-     },
-     "metadata": {},
-     "output_type": "display_data"
-    },
-    {
-     "data": {
-      "text/plain": [
-       "<Figure size 640x480 with 0 Axes>"
-      ]
-     },
-     "metadata": {},
-     "output_type": "display_data"
-    },
-    {
-     "data": {
-      "text/plain": [
-       "<Figure size 640x480 with 0 Axes>"
-      ]
-     },
-     "metadata": {},
-     "output_type": "display_data"
-    },
-    {
-     "data": {
-      "text/plain": [
-       "<Figure size 640x480 with 0 Axes>"
-      ]
-     },
-     "metadata": {},
-     "output_type": "display_data"
-    },
-    {
-     "data": {
-      "text/plain": [
-       "<Figure size 640x480 with 0 Axes>"
-      ]
-     },
-     "metadata": {},
-     "output_type": "display_data"
-    },
-    {
-     "data": {
-      "text/plain": [
-       "<Figure size 640x480 with 0 Axes>"
-      ]
-     },
-     "metadata": {},
-     "output_type": "display_data"
-    },
-    {
-     "data": {
-      "text/plain": [
-       "<Figure size 640x480 with 0 Axes>"
-      ]
-     },
-     "metadata": {},
-     "output_type": "display_data"
-    },
-    {
-     "data": {
-      "text/plain": [
-       "<Figure size 640x480 with 0 Axes>"
-      ]
-     },
-     "metadata": {},
-     "output_type": "display_data"
-    },
-    {
-     "data": {
-      "text/plain": [
-       "<Figure size 640x480 with 0 Axes>"
-      ]
-     },
-     "metadata": {},
-     "output_type": "display_data"
-    },
-    {
-     "data": {
-      "text/plain": [
-       "<Figure size 640x480 with 0 Axes>"
-      ]
-     },
-     "metadata": {},
-     "output_type": "display_data"
-    },
-    {
-     "data": {
-      "text/plain": [
-       "<Figure size 640x480 with 0 Axes>"
-      ]
-     },
-     "metadata": {},
-     "output_type": "display_data"
-    },
-    {
-     "data": {
-      "text/plain": [
-       "<Figure size 640x480 with 0 Axes>"
-      ]
-     },
-     "metadata": {},
-     "output_type": "display_data"
-    },
-    {
-     "data": {
-      "text/plain": [
-       "<Figure size 640x480 with 0 Axes>"
-      ]
-     },
-     "metadata": {},
-     "output_type": "display_data"
-    },
-    {
-     "data": {
-      "text/plain": [
-       "<Figure size 640x480 with 0 Axes>"
-      ]
-     },
-     "metadata": {},
-     "output_type": "display_data"
-    },
-    {
-     "data": {
-      "text/plain": [
-       "<Figure size 640x480 with 0 Axes>"
-      ]
-     },
-     "metadata": {},
-     "output_type": "display_data"
-    },
-    {
-     "data": {
-      "text/plain": [
-       "<Figure size 640x480 with 0 Axes>"
-      ]
-     },
-     "metadata": {},
-     "output_type": "display_data"
-    },
+     "ename": "AttributeError",
+     "evalue": "'Digraph' object has no attribute 'savefig'",
+     "output_type": "error",
+     "traceback": [
+      "\u001b[0;31m---------------------------------------------------------------------------\u001b[0m",
+      "\u001b[0;31mAttributeError\u001b[0m                            Traceback (most recent call last)",
+      "Cell \u001b[0;32mIn[187], line 1\u001b[0m\n\u001b[0;32m----> 1\u001b[0m \u001b[43mdot\u001b[49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43msavefig\u001b[49m(\u001b[38;5;124m\"\u001b[39m\u001b[38;5;124mflowchart.png\u001b[39m\u001b[38;5;124m\"\u001b[39m)\n",
+      "\u001b[0;31mAttributeError\u001b[0m: 'Digraph' object has no attribute 'savefig'"
+     ]
+    }
+   ],
+   "source": [
+    "dot.savefig(\"flowchart.png\")"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 117,
+   "metadata": {},
+   "outputs": [
     {
      "data": {
+      "image/png": "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",
+      "text/latex": [
+       "$\\displaystyle 2$"
+      ],
       "text/plain": [
-       "<Figure size 640x480 with 0 Axes>"
+       "2"
       ]
      },
+     "execution_count": 117,
      "metadata": {},
-     "output_type": "display_data"
-    },
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "pl.tracked_steps"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 120,
+   "metadata": {},
+   "outputs": [
     {
      "data": {
       "text/plain": [
-       "<Figure size 640x480 with 0 Axes>"
+       "[None, 'filtered nothing lol']"
       ]
      },
+     "execution_count": 120,
      "metadata": {},
-     "output_type": "display_data"
-    },
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "pl._tracked_operations"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 136,
+   "metadata": {},
+   "outputs": [
     {
      "data": {
+      "image/svg+xml": [
+       "<?xml version=\"1.0\" encoding=\"UTF-8\" standalone=\"no\"?>\n",
+       "<!DOCTYPE svg PUBLIC \"-//W3C//DTD SVG 1.1//EN\"\n",
+       " \"http://www.w3.org/Graphics/SVG/1.1/DTD/svg11.dtd\">\n",
+       "<!-- Generated by graphviz version 9.0.0 (20230911.1827)\n",
+       " -->\n",
+       "<!-- Pages: 1 -->\n",
+       "<svg width=\"176pt\" height=\"143pt\"\n",
+       " viewBox=\"0.00 0.00 175.75 142.50\" xmlns=\"http://www.w3.org/2000/svg\" xmlns:xlink=\"http://www.w3.org/1999/xlink\">\n",
+       "<g id=\"graph0\" class=\"graph\" transform=\"scale(1 1) rotate(0) translate(4 138.5)\">\n",
+       "<polygon fill=\"white\" stroke=\"none\" points=\"-4,4 -4,-138.5 171.75,-138.5 171.75,4 -4,4\"/>\n",
+       "<!-- 0 -->\n",
+       "<g id=\"node1\" class=\"node\">\n",
+       "<title>0</title>\n",
+       "<polygon fill=\"lightgrey\" stroke=\"black\" points=\"64,-134.5 0,-134.5 0,-93.5 64,-93.5 64,-134.5\"/>\n",
+       "<text text-anchor=\"middle\" x=\"32\" y=\"-117.2\" font-family=\"Times,serif\" font-size=\"14.00\">Step 0</text>\n",
+       "<text text-anchor=\"middle\" x=\"32\" y=\"-100.7\" font-family=\"Times,serif\" font-size=\"14.00\"> (n=150)</text>\n",
+       "</g>\n",
+       "<!-- 1 -->\n",
+       "<g id=\"node2\" class=\"node\">\n",
+       "<title>1</title>\n",
+       "<polygon fill=\"lightgrey\" stroke=\"black\" points=\"60.62,-41 3.38,-41 3.38,0 60.62,0 60.62,-41\"/>\n",
+       "<text text-anchor=\"middle\" x=\"32\" y=\"-23.7\" font-family=\"Times,serif\" font-size=\"14.00\">filtered</text>\n",
+       "<text text-anchor=\"middle\" x=\"32\" y=\"-7.2\" font-family=\"Times,serif\" font-size=\"14.00\"> (n=75)</text>\n",
+       "</g>\n",
+       "<!-- 0&#45;&gt;1 -->\n",
+       "<g id=\"edge1\" class=\"edge\">\n",
+       "<title>0&#45;&gt;1</title>\n",
+       "<path fill=\"none\" stroke=\"black\" d=\"M32,-93.15C32,-81.28 32,-65.94 32,-52.5\"/>\n",
+       "<polygon fill=\"black\" stroke=\"black\" points=\"35.5,-52.83 32,-42.83 28.5,-52.83 35.5,-52.83\"/>\n",
+       "<text text-anchor=\"middle\" x=\"99.88\" y=\"-62.2\" font-family=\"Times,serif\" font-size=\"14.00\">filtered to first 75 entries</text>\n",
+       "</g>\n",
+       "</g>\n",
+       "</svg>\n"
+      ],
       "text/plain": [
-       "<Figure size 640x480 with 0 Axes>"
+       "<graphviz.graphs.Digraph at 0x2b37649d0>"
       ]
      },
+     "execution_count": 136,
      "metadata": {},
-     "output_type": "display_data"
-    },
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "pl.plot_flowchart()"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": []
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 20,
+   "metadata": {},
+   "outputs": [
     {
      "data": {
+      "text/html": [
+       "<div>\n",
+       "<style scoped>\n",
+       "    .dataframe tbody tr th:only-of-type {\n",
+       "        vertical-align: middle;\n",
+       "    }\n",
+       "\n",
+       "    .dataframe tbody tr th {\n",
+       "        vertical-align: top;\n",
+       "    }\n",
+       "\n",
+       "    .dataframe thead th {\n",
+       "        text-align: right;\n",
+       "    }\n",
+       "</style>\n",
+       "<table border=\"1\" class=\"dataframe\">\n",
+       "  <thead>\n",
+       "    <tr style=\"text-align: right;\">\n",
+       "      <th></th>\n",
+       "      <th></th>\n",
+       "      <th>Missing</th>\n",
+       "      <th>Overall</th>\n",
+       "    </tr>\n",
+       "  </thead>\n",
+       "  <tbody>\n",
+       "    <tr>\n",
+       "      <th>n</th>\n",
+       "      <th></th>\n",
+       "      <td></td>\n",
+       "      <td>150</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th rowspan=\"3\" valign=\"top\">ethnicity, n (%)</th>\n",
+       "      <th>asian</th>\n",
+       "      <td>0</td>\n",
+       "      <td>43 (28.7)</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>black</th>\n",
+       "      <td></td>\n",
+       "      <td>52 (34.7)</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>white</th>\n",
+       "      <td></td>\n",
+       "      <td>55 (36.7)</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th rowspan=\"2\" valign=\"top\">gender, n (%)</th>\n",
+       "      <th>female</th>\n",
+       "      <td>0</td>\n",
+       "      <td>66 (44.0)</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>male</th>\n",
+       "      <td></td>\n",
+       "      <td>84 (56.0)</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th rowspan=\"2\" valign=\"top\">proficiency, n (%)</th>\n",
+       "      <th>high prof.</th>\n",
+       "      <td>0</td>\n",
+       "      <td>78 (52.0)</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>limited prof.</th>\n",
+       "      <td></td>\n",
+       "      <td>72 (48.0)</td>\n",
+       "    </tr>\n",
+       "  </tbody>\n",
+       "</table>\n",
+       "</div><br />"
+      ],
       "text/plain": [
-       "<Figure size 640x480 with 0 Axes>"
+       "                                 Missing    Overall\n",
+       "n                                               150\n",
+       "ethnicity, n (%)   asian               0  43 (28.7)\n",
+       "                   black                  52 (34.7)\n",
+       "                   white                  55 (36.7)\n",
+       "gender, n (%)      female              0  66 (44.0)\n",
+       "                   male                   84 (56.0)\n",
+       "proficiency, n (%) high prof.          0  78 (52.0)\n",
+       "                   limited prof.          72 (48.0)"
       ]
      },
+     "execution_count": 20,
      "metadata": {},
-     "output_type": "display_data"
-    },
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "t1 = TableOne(random_dataset, columns=[\"ethnicity\", \"gender\", \"proficiency\"])\n",
+    "t1"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 151,
+   "metadata": {},
+   "outputs": [
     {
      "data": {
       "text/plain": [
-       "<Figure size 640x480 with 0 Axes>"
+       "{'asian': [28.7, 24.0],\n",
+       " 'black': [34.7, 34.0],\n",
+       " 'white': [36.7, 42.0],\n",
+       " 'male': [56.0, 56.0],\n",
+       " 'female': [44.0, 44.0],\n",
+       " 'high prof.': [52.0, 48.0],\n",
+       " 'limited prof.': [48.0, 52.0]}"
       ]
      },
+     "execution_count": 151,
      "metadata": {},
-     "output_type": "display_data"
-    },
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "collated_data = {key: values1 + values2 for key, values1, values2 in zip(dpcts.keys(), dpcts.values(), dcpts2.values())}\n",
+    "collated_data"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 155,
+   "metadata": {},
+   "outputs": [
     {
      "data": {
       "text/plain": [
-       "<Figure size 640x480 with 0 Axes>"
+       "{'asian': [28.7, 24.0], 'black': [34.7, 34.0], 'white': [36.7, 42.0]}"
       ]
      },
+     "execution_count": 155,
      "metadata": {},
-     "output_type": "display_data"
-    },
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "data_ethnicity = {key: value for key, value in collated_data.items() if key in [\"asian\", \"black\", \"white\"]}\n",
+    "data_ethnicity"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 191,
+   "metadata": {},
+   "outputs": [
     {
      "data": {
       "text/plain": [
-       "<Figure size 640x480 with 0 Axes>"
+       "{'male': [56.0, 56.0], 'female': [44.0, 44.0]}"
       ]
      },
+     "execution_count": 191,
      "metadata": {},
-     "output_type": "display_data"
-    },
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "data_gender = {key: value for key, value in collated_data.items() if key in [\"male\", \"female\"]}\n",
+    "data_gender"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 248,
+   "metadata": {},
+   "outputs": [
     {
      "data": {
       "text/plain": [
-       "<Figure size 640x480 with 0 Axes>"
+       "{'high prof.': [52.0, 48.0], 'limited prof.': [48.0, 52.0]}"
       ]
      },
+     "execution_count": 248,
      "metadata": {},
-     "output_type": "display_data"
-    },
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "data_prof = {key: value for key, value in collated_data.items() if key in [\"high prof.\", \"limited prof.\"]}\n",
+    "data_prof"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 211,
+   "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>asian</th>\n",
+       "      <th>black</th>\n",
+       "      <th>white</th>\n",
+       "    </tr>\n",
+       "  </thead>\n",
+       "  <tbody>\n",
+       "    <tr>\n",
+       "      <th>0</th>\n",
+       "      <td>28.7</td>\n",
+       "      <td>34.7</td>\n",
+       "      <td>36.7</td>\n",
+       "    </tr>\n",
+       "  </tbody>\n",
+       "</table>\n",
+       "</div>"
+      ],
       "text/plain": [
-       "<Figure size 640x480 with 0 Axes>"
+       "   asian  black  white\n",
+       "0   28.7   34.7   36.7"
       ]
      },
+     "execution_count": 211,
      "metadata": {},
-     "output_type": "display_data"
-    },
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "df_ethnicity"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 242,
+   "metadata": {},
+   "outputs": [
     {
-     "data": {
-      "text/plain": [
-       "<Figure size 640x480 with 0 Axes>"
-      ]
-     },
-     "metadata": {},
-     "output_type": "display_data"
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "28.7\n",
+      "34.7\n",
+      "36.7\n"
+     ]
+    }
+   ],
+   "source": [
+    "for v in l.iterrows():\n",
+    "    print(v[1].values[0])"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 80,
+   "metadata": {},
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "28.7\n",
+      "34.7\n",
+      "36.7\n",
+      "56.0\n",
+      "44.0\n",
+      "52.0\n",
+      "48.0\n",
+      "24.0\n",
+      "34.0\n",
+      "42.0\n",
+      "56.0\n",
+      "44.0\n",
+      "48.0\n",
+      "52.0\n"
+     ]
     },
     {
      "data": {
+      "image/png": "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",
       "text/plain": [
-       "<Figure size 640x480 with 0 Axes>"
+       "<Figure size 700x700 with 2 Axes>"
       ]
      },
      "metadata": {},
@@ -1425,81 +1598,101 @@
     }
    ],
    "source": [
+    "import pandas as pd\n",
+    "import matplotlib.pyplot as plt\n",
+    "import seaborn as sns\n",
+    "\n",
+    "# Given data dictionaries\n",
+    "data_ethnicity = {\"asian\": [28.7, 24.0], \"black\": [34.7, 34.0], \"white\": [36.7, 42.0]}\n",
+    "\n",
+    "data_gender = {\"male\": [56.0, 56.0], \"female\": [44.0, 44.0]}\n",
+    "\n",
+    "data_prof = {\"high prof.\": [52.0, 48.0], \"limited prof.\": [48.0, 52.0]}\n",
+    "\n",
+    "# Convert dictionaries to DataFrames\n",
+    "df_ethnicity = pd.DataFrame(data_ethnicity)\n",
+    "df_gender = pd.DataFrame(data_gender)\n",
+    "df_prof = pd.DataFrame(data_prof)\n",
+    "\n",
+    "# Plotting\n",
+    "fig, axes = plt.subplots(2, 1, figsize=(7, 7))\n",
+    "\n",
+    "# Plotting for the ethnicity with a small gap\n",
+    "categories = [df_ethnicity, df_gender, df_prof]\n",
+    "legend_labels = []\n",
+    "\n",
+    "for idx, ax in enumerate(axes):\n",
+    "    for pos, data in enumerate([df_ethnicity.loc[idx], df_gender.loc[idx], df_prof.loc[idx]]):\n",
+    "        cumwidth = 0\n",
+    "        category = categories[pos]\n",
+    "\n",
+    "        hue_shift = (pos + 1) / len(categories[pos])  # Adjust the hue shift based on the category position\n",
+    "        colors = sns.color_palette(\"husl\", len(category.columns))\n",
+    "        adjusted_colors = [((color[0] + hue_shift) % 1, color[1], color[2]) for color in colors]\n",
+    "\n",
+    "        for i, value in enumerate(data):\n",
+    "            print(value)\n",
+    "            # value = value[1].values[0]  # Take the value based on idx\n",
+    "            ax.barh(pos, value, left=cumwidth, color=adjusted_colors[i], height=0.8)\n",
     "\n",
-    "colors_racial = ['midnightblue', 'gold', 'saddlebrown', 'coral', 'lightgrey']\n",
-    "colors_gender = ['tab:blue', 'orchid']\n",
-    "colors_proficiency = ['yellowgreen', 'firebrick']\n",
-    "fig, axes = plt.subplots(6, 1, figsize=(4, 9))\n",
-    "labels = ['Race and Ethnicity', 'Sex', 'English Proficiency']\n",
-    "for i, ax in enumerate(axes):\n",
-    "    # Extract rows\n",
-    "    data_racial = df_racial.iloc[i]\n",
-    "    data_gender = df_gender.iloc[i]\n",
-    "    data_proficiency = df_proficiency.iloc[i]\n",
-    "    \n",
-    "    # For vertical positioning within each lane\n",
-    "    positions = [2, 1, 0]\n",
-    "    for j, (data, colors) in enumerate(zip( [data_racial, data_gender, data_proficiency], [colors_racial, colors_gender, colors_proficiency] )):\n",
-    "        position = positions[j]\n",
-    "        cum_width = 0\n",
-    "        for value, color in zip(data, colors):\n",
-    "            ax.barh(position, value, left=cum_width, color=color, height=.8)\n",
     "            if value > 5:\n",
     "                # Add data labels\n",
     "                width = value\n",
-    "                ax.text(cum_width + width/2, position, '{:.1f}'.format(value), ha='center', va='center', color='white', fontweight='bold')\n",
-    "                cum_width += value\n",
-    "                ax.set_xlim(0, 100)\n",
-    "                ax.set_ylim(-1, 3)\n",
-    "                ax.set_yticks([])\n",
-    "                ax.set_xticks([])\n",
-    "                \n",
-    "                for spine in ax.spines.values():\n",
-    "                    spine.set_visible(False)\n",
-    "                    ax.set_xlabel('Group Proportion (%)')\n",
-    "                    # Add legend to the bottom\n",
-    "                    racial_patches = [plt.Rectangle((0,0),1,1, color=color) for color in colors_racial]\n",
-    "                    gender_patches = [plt.Rectangle((0,0),1,1, color=color) for color in colors_gender]\n",
-    "                    proficiency_patches = [plt.Rectangle((0,0),1,1, color=color) for color in colors_proficiency]\n",
-    "                    racial_labels = ['Black', 'Asian', 'Hispanic', 'Other', 'White']\n",
-    "                    gender_labels = ['Male', 'Female']\n",
-    "                    proficiency_labels = ['Proficient', 'Limited Prof.']\n",
-    "                    legend_handles = racial_patches + gender_patches + proficiency_patches\n",
-    "                    legend_labels = racial_labels + gender_labels + proficiency_labels\n",
-    "                    \n",
-    "                    fig.legend(legend_handles, legend_labels, loc='lower center', ncol=3, bbox_to_anchor=(0.5, -0.1))\n",
-    "                \n",
-    "                plt.tight_layout()\n",
-    "                plt.show() "
+    "                ax.text(\n",
+    "                    cumwidth + width / 2,\n",
+    "                    pos,\n",
+    "                    \"{:.1f}\".format(value),\n",
+    "                    ha=\"center\",\n",
+    "                    va=\"center\",\n",
+    "                    color=\"white\",\n",
+    "                    fontweight=\"bold\",\n",
+    "                )\n",
+    "\n",
+    "            ax.set_yticks([])\n",
+    "            ax.set_xticks([])\n",
+    "            cumwidth += value\n",
+    "\n",
+    "            legend_labels.append(category.columns[i])\n",
+    "\n",
+    "# makes the frames invisible\n",
+    "# for ax in axes:\n",
+    "#     ax.axis('off')\n",
+    "\n",
+    "\n",
+    "# Add legend for the first subplot\n",
+    "plt.legend(legend_labels, loc=\"best\", bbox_to_anchor=(-0.5, 1))  # ncol=len(legend_labels), bbox_to_anchor=(0.5, -0.3)\n",
+    "\n",
+    "plt.tight_layout()\n",
+    "plt.show()"
    ]
   },
   {
    "cell_type": "code",
-   "execution_count": 4,
+   "execution_count": 369,
    "metadata": {},
    "outputs": [
     {
-     "ename": "IndexError",
-     "evalue": "list index out of range",
+     "ename": "AttributeError",
+     "evalue": "'str' object has no attribute 'values'",
      "output_type": "error",
      "traceback": [
       "\u001b[0;31m---------------------------------------------------------------------------\u001b[0m",
-      "\u001b[0;31mIndexError\u001b[0m                                Traceback (most recent call last)",
-      "Cell \u001b[0;32mIn[4], line 38\u001b[0m\n\u001b[1;32m     36\u001b[0m positions \u001b[38;5;241m=\u001b[39m [\u001b[38;5;241m2\u001b[39m, \u001b[38;5;241m1\u001b[39m, \u001b[38;5;241m0\u001b[39m]\n\u001b[1;32m     37\u001b[0m \u001b[38;5;28;01mfor\u001b[39;00m j, (index, row) \u001b[38;5;129;01min\u001b[39;00m \u001b[38;5;28menumerate\u001b[39m(data\u001b[38;5;241m.\u001b[39miterrows()):\n\u001b[0;32m---> 38\u001b[0m     position \u001b[38;5;241m=\u001b[39m \u001b[43mpositions\u001b[49m\u001b[43m[\u001b[49m\u001b[43mj\u001b[49m\u001b[43m]\u001b[49m\n\u001b[1;32m     39\u001b[0m     cum_width \u001b[38;5;241m=\u001b[39m \u001b[38;5;241m0\u001b[39m\n\u001b[1;32m     40\u001b[0m     \u001b[38;5;28;01mfor\u001b[39;00m value, color \u001b[38;5;129;01min\u001b[39;00m \u001b[38;5;28mzip\u001b[39m(row, colors):\n",
-      "\u001b[0;31mIndexError\u001b[0m: list index out of range"
+      "\u001b[0;31mAttributeError\u001b[0m                            Traceback (most recent call last)",
+      "Cell \u001b[0;32mIn[369], line 37\u001b[0m\n\u001b[1;32m     34\u001b[0m adjusted_colors \u001b[38;5;241m=\u001b[39m [((color[\u001b[38;5;241m0\u001b[39m] \u001b[38;5;241m+\u001b[39m hue_shift) \u001b[38;5;241m%\u001b[39m \u001b[38;5;241m1\u001b[39m, color[\u001b[38;5;241m1\u001b[39m], color[\u001b[38;5;241m2\u001b[39m]) \u001b[38;5;28;01mfor\u001b[39;00m color \u001b[38;5;129;01min\u001b[39;00m colors]\n\u001b[1;32m     36\u001b[0m \u001b[38;5;28;01mfor\u001b[39;00m i, value \u001b[38;5;129;01min\u001b[39;00m \u001b[38;5;28menumerate\u001b[39m(data):\n\u001b[0;32m---> 37\u001b[0m     value \u001b[38;5;241m=\u001b[39m \u001b[43mvalue\u001b[49m\u001b[43m[\u001b[49m\u001b[38;5;241;43m1\u001b[39;49m\u001b[43m]\u001b[49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43mvalues\u001b[49m[idx]  \u001b[38;5;66;03m# Take the value based on idx\u001b[39;00m\n\u001b[1;32m     38\u001b[0m     ax\u001b[38;5;241m.\u001b[39mbarh(pos, value, left\u001b[38;5;241m=\u001b[39mcumwidth, color\u001b[38;5;241m=\u001b[39madjusted_colors[i], height\u001b[38;5;241m=\u001b[39m\u001b[38;5;241m0.8\u001b[39m)\n\u001b[1;32m     40\u001b[0m     \u001b[38;5;28;01mif\u001b[39;00m value \u001b[38;5;241m>\u001b[39m \u001b[38;5;241m5\u001b[39m:\n\u001b[1;32m     41\u001b[0m         \u001b[38;5;66;03m# Add data labels\u001b[39;00m\n",
+      "\u001b[0;31mAttributeError\u001b[0m: 'str' object has no attribute 'values'"
      ]
     },
     {
      "data": {
-      "image/png": "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",
+      "image/png": "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",
       "text/plain": [
-       "<Figure size 400x900 with 3 Axes>"
+       "<Figure size 1000x1000 with 2 Axes>"
       ]
      },
      "metadata": {
       "image/png": {
-       "height": 734,
-       "width": 330
+       "height": 819,
+       "width": 837
       }
      },
      "output_type": "display_data"
@@ -1508,63 +1701,193 @@
    "source": [
     "import pandas as pd\n",
     "import matplotlib.pyplot as plt\n",
-    "%config InlineBackend.figure_format = 'retina'\n",
+    "import seaborn as sns\n",
+    "\n",
+    "# Given data dictionaries\n",
+    "data_ethnicity = {\"asian\": [28.7, 24.0], \"black\": [34.7, 34.0], \"white\": [36.7, 42.0]}\n",
+    "\n",
+    "data_gender = {\"male\": [56.0, 56.0], \"female\": [44.0, 44.0]}\n",
     "\n",
-    "# Data dictionaries\n",
-    "racial_data = {'Black': [10.8, 10.8, 10.8, 10, 10, 11.8], 'Asian': [2.9, 2.9, 2.9, 2.8, 2.8, 3.3],\n",
-    "               'Hispanic': [2.9, 2.9, 2.9, 2.8, 2.8, 3.3], 'Other': [14.6, 14.6, 14.6, 15.8, 15.8, 0],\n",
-    "               'White': [68.8, 68.8, 68.8, 68.6, 68.6, 81.6]}\n",
+    "data_prof = {\"high prof.\": [52.0, 48.0], \"limited prof.\": [48.0, 52.0]}\n",
     "\n",
-    "gender_data = {'Male': [55.8, 55.8, 55.9, 58.2, 58.2, 57.6], 'Female': [44.2, 44.2, 44.1, 41.8, 41.8, 42.4]}\n",
+    "# Convert dictionaries to DataFrames\n",
+    "df_ethnicity = pd.DataFrame(data_ethnicity)\n",
+    "df_gender = pd.DataFrame(data_gender)\n",
+    "df_prof = pd.DataFrame(data_prof)\n",
     "\n",
-    "proficiency_data = {'Proficient': [89.8, 89.8, 89.2, 89.4, 89.4, 90.2], 'Limited': [10.2, 10.2, 10.8, 10.6, 10.6, 9.8]}\n",
+    "# Plotting\n",
+    "fig, axes = plt.subplots(2, 1, figsize=(10, 10))\n",
     "\n",
-    "# Creating DataFrames\n",
-    "df_racial = pd.DataFrame(racial_data)\n",
-    "df_gender = pd.DataFrame(gender_data)\n",
-    "df_proficiency = pd.DataFrame(proficiency_data)\n",
+    "# Plotting for the ethnicity with a small gap\n",
+    "categories = [df_ethnicity, df_gender, df_prof]\n",
+    "legend_labels = []\n",
     "\n",
-    "colors_racial = ['midnightblue', 'gold', 'saddlebrown', 'coral', 'lightgrey']\n",
-    "colors_gender = ['tab:blue', 'orchid']\n",
-    "colors_proficiency = ['yellowgreen', 'firebrick']\n",
+    "for idx, ax in enumerate(axes):\n",
+    "    for pos, data in enumerate(categories[idx].T.iterrows()):\n",
+    "        cumwidth = 0\n",
+    "        category = categories[idx]\n",
     "\n",
-    "fig, axes = plt.subplots(3, 1, figsize=(4, 9))\n",
-    "labels = ['Race and Ethnicity', 'Sex', 'English Proficiency']\n",
+    "        hue_shift = (pos + 1) / len(categories[idx])  # Adjust the hue shift based on the category position\n",
+    "        colors = sns.color_palette(\"husl\", len(category.columns))\n",
+    "        adjusted_colors = [((color[0] + hue_shift) % 1, color[1], color[2]) for color in colors]\n",
     "\n",
-    "for i, ax in enumerate(axes):\n",
-    "    ax.set_title(labels[i])\n",
-    "    ax.set_xlim(0, 100)\n",
-    "    ax.set_ylim(-1, 3)\n",
-    "    ax.set_yticks([])\n",
-    "    ax.set_xticks([])\n",
-    "    for spine in ax.spines.values():\n",
-    "        spine.set_visible(False)\n",
+    "        for i, value in enumerate(data):\n",
+    "            value = value[1].values[idx]  # Take the value based on idx\n",
+    "            ax.barh(pos, value, left=cumwidth, color=adjusted_colors[i], height=0.8)\n",
     "\n",
-    "for ax, data, colors in zip(axes, [df_racial, df_gender, df_proficiency], [colors_racial, colors_gender, colors_proficiency]):\n",
-    "    positions = [2, 1, 0]\n",
-    "    for j, (index, row) in enumerate(data.iterrows()):\n",
-    "        position = positions[j]\n",
-    "        cum_width = 0\n",
-    "        for value, color in zip(row, colors):\n",
-    "            ax.barh(position, value, left=cum_width, color=color, height=0.8)\n",
     "            if value > 5:\n",
-    "                ax.text(cum_width + value / 2, position, '{:.1f}'.format(value), ha='center', va='center', color='white',\n",
-    "                        fontweight='bold')\n",
-    "            cum_width += value\n",
+    "                # Add data labels\n",
+    "                width = value\n",
+    "                ax.text(\n",
+    "                    cumwidth + width / 2,\n",
+    "                    pos,\n",
+    "                    \"{:.1f}\".format(value),\n",
+    "                    ha=\"center\",\n",
+    "                    va=\"center\",\n",
+    "                    color=\"white\",\n",
+    "                    fontweight=\"bold\",\n",
+    "                )\n",
+    "\n",
+    "            ax.set_yticks([])\n",
+    "            ax.set_xticks([])\n",
+    "            cumwidth += value\n",
+    "\n",
+    "            legend_labels.append(category.columns[i])\n",
+    "\n",
+    "# Set labels and title for the first subplot\n",
+    "axes[0].set_title(\"First Numbers\")\n",
+    "axes[0].set_xlabel(\"Percentage\")\n",
+    "axes[0].set_yticks(range(len(categories[0])))\n",
+    "axes[0].set_yticklabels(category_labels)\n",
+    "\n",
+    "# Set labels and title for the second subplot\n",
+    "axes[1].set_title(\"Second Numbers\")\n",
+    "axes[1].set_xlabel(\"Percentage\")\n",
+    "axes[1].set_yticks(range(len(categories[0])))\n",
+    "axes[1].set_yticklabels(category_labels)\n",
+    "\n",
+    "# Add legend for the first subplot\n",
+    "axes[0].legend(legend_labels, loc=\"lower center\", ncol=len(legend_labels), bbox_to_anchor=(0.5, -0.1))\n",
+    "\n",
+    "plt.tight_layout()\n",
+    "plt.show()"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 343,
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "image/svg+xml": [
+       "<?xml version=\"1.0\" encoding=\"UTF-8\" standalone=\"no\"?>\n",
+       "<!DOCTYPE svg PUBLIC \"-//W3C//DTD SVG 1.1//EN\"\n",
+       " \"http://www.w3.org/Graphics/SVG/1.1/DTD/svg11.dtd\">\n",
+       "<!-- Generated by graphviz version 9.0.0 (20230911.1827)\n",
+       " -->\n",
+       "<!-- Pages: 1 -->\n",
+       "<svg width=\"250pt\" height=\"143pt\"\n",
+       " viewBox=\"0.00 0.00 250.38 142.50\" xmlns=\"http://www.w3.org/2000/svg\" xmlns:xlink=\"http://www.w3.org/1999/xlink\">\n",
+       "<g id=\"graph0\" class=\"graph\" transform=\"scale(1 1) rotate(0) translate(4 138.5)\">\n",
+       "<polygon fill=\"white\" stroke=\"none\" points=\"-4,4 -4,-138.5 246.38,-138.5 246.38,4 -4,4\"/>\n",
+       "<!-- A -->\n",
+       "<g id=\"node1\" class=\"node\">\n",
+       "<title>A</title>\n",
+       "<polygon fill=\"lightgrey\" stroke=\"black\" points=\"80.12,-134.5 16.12,-134.5 16.12,-93.5 80.12,-93.5 80.12,-134.5\"/>\n",
+       "<text text-anchor=\"middle\" x=\"48.12\" y=\"-117.2\" font-family=\"Times,serif\" font-size=\"14.00\">Original</text>\n",
+       "<text text-anchor=\"middle\" x=\"48.12\" y=\"-100.7\" font-family=\"Times,serif\" font-size=\"14.00\"> (n=200)</text>\n",
+       "</g>\n",
+       "<!-- B -->\n",
+       "<g id=\"node2\" class=\"node\">\n",
+       "<title>B</title>\n",
+       "<polygon fill=\"lightgrey\" stroke=\"black\" points=\"96.25,-41 0,-41 0,0 96.25,0 96.25,-41\"/>\n",
+       "<text text-anchor=\"middle\" x=\"48.12\" y=\"-23.7\" font-family=\"Times,serif\" font-size=\"14.00\">Filtered values</text>\n",
+       "<text text-anchor=\"middle\" x=\"48.12\" y=\"-7.2\" font-family=\"Times,serif\" font-size=\"14.00\"> (n=100)</text>\n",
+       "</g>\n",
+       "<!-- A&#45;&gt;B -->\n",
+       "<g id=\"edge1\" class=\"edge\">\n",
+       "<title>A&#45;&gt;B</title>\n",
+       "<path fill=\"none\" stroke=\"black\" d=\"M48.12,-93.15C48.12,-81.28 48.12,-65.94 48.12,-52.5\"/>\n",
+       "<polygon fill=\"black\" stroke=\"black\" points=\"51.63,-52.83 48.13,-42.83 44.63,-52.83 51.63,-52.83\"/>\n",
+       "<text text-anchor=\"middle\" x=\"145.25\" y=\"-62.2\" font-family=\"Times,serif\" font-size=\"14.00\"> &#160;Filtering Process: Removes lalalla</text>\n",
+       "</g>\n",
+       "</g>\n",
+       "</svg>\n"
+      ],
+      "text/plain": [
+       "<graphviz.graphs.Digraph at 0x2b01d6a50>"
+      ]
+     },
+     "execution_count": 343,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "import graphviz\n",
+    "\n",
+    "# Create Digraph object\n",
+    "dot = graphviz.Digraph()\n",
+    "\n",
+    "# Define nodes (edgy nodes)\n",
+    "dot.node(name=\"A\", label=\"Original\\n (n=200)\", style=\"filled\", shape=\"box\")\n",
+    "dot.node(name=\"B\", label=\"Filtered values\\n (n=100)\", style=\"filled\", shape=\"box\")\n",
+    "\n",
+    "# Define edges (arrows) with labels\n",
+    "dot.edge(\"A\", \"B\", label=\"  Filtering Process: Removes lalalla\", labeldistance=\"10.5\")\n",
+    "\n",
+    "# Render the graph\n",
+    "dot.render(\"flow_diagram_edgy_nodes\", format=\"png\", cleanup=True)\n",
+    "\n",
+    "dot"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 317,
+   "metadata": {},
+   "outputs": [
+    {
+     "ename": "UnidentifiedImageError",
+     "evalue": "cannot identify image file 'dot_figure.png'",
+     "output_type": "error",
+     "traceback": [
+      "\u001b[0;31m---------------------------------------------------------------------------\u001b[0m",
+      "\u001b[0;31mUnidentifiedImageError\u001b[0m                    Traceback (most recent call last)",
+      "Cell \u001b[0;32mIn[317], line 8\u001b[0m\n\u001b[1;32m      5\u001b[0m dot\u001b[38;5;241m.\u001b[39mrender(\u001b[38;5;124m'\u001b[39m\u001b[38;5;124mdot_figure.png\u001b[39m\u001b[38;5;124m'\u001b[39m, \u001b[38;5;28mformat\u001b[39m\u001b[38;5;241m=\u001b[39m\u001b[38;5;124m'\u001b[39m\u001b[38;5;124mpng\u001b[39m\u001b[38;5;124m'\u001b[39m)\n\u001b[1;32m      7\u001b[0m \u001b[38;5;66;03m# Convert Dot figure to image\u001b[39;00m\n\u001b[0;32m----> 8\u001b[0m dot_img \u001b[38;5;241m=\u001b[39m \u001b[43mImage\u001b[49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43mopen\u001b[49m\u001b[43m(\u001b[49m\u001b[38;5;124;43m'\u001b[39;49m\u001b[38;5;124;43mdot_figure.png\u001b[39;49m\u001b[38;5;124;43m'\u001b[39;49m\u001b[43m)\u001b[49m\n\u001b[1;32m     10\u001b[0m \u001b[38;5;66;03m# Create subplots for both figures\u001b[39;00m\n\u001b[1;32m     11\u001b[0m fig, axes \u001b[38;5;241m=\u001b[39m plt\u001b[38;5;241m.\u001b[39msubplots(\u001b[38;5;241m1\u001b[39m, \u001b[38;5;241m2\u001b[39m, figsize\u001b[38;5;241m=\u001b[39m(\u001b[38;5;241m15\u001b[39m, \u001b[38;5;241m5\u001b[39m))  \u001b[38;5;66;03m# Adjust figsize as needed\u001b[39;00m\n",
+      "File \u001b[0;32m~/Documents/ehrapy_clean/ehrapy_venv_feb/lib/python3.11/site-packages/PIL/Image.py:3309\u001b[0m, in \u001b[0;36mopen\u001b[0;34m(fp, mode, formats)\u001b[0m\n\u001b[1;32m   3307\u001b[0m     warnings\u001b[38;5;241m.\u001b[39mwarn(message)\n\u001b[1;32m   3308\u001b[0m msg \u001b[38;5;241m=\u001b[39m \u001b[38;5;124m\"\u001b[39m\u001b[38;5;124mcannot identify image file \u001b[39m\u001b[38;5;132;01m%r\u001b[39;00m\u001b[38;5;124m\"\u001b[39m \u001b[38;5;241m%\u001b[39m (filename \u001b[38;5;28;01mif\u001b[39;00m filename \u001b[38;5;28;01melse\u001b[39;00m fp)\n\u001b[0;32m-> 3309\u001b[0m \u001b[38;5;28;01mraise\u001b[39;00m UnidentifiedImageError(msg)\n",
+      "\u001b[0;31mUnidentifiedImageError\u001b[0m: cannot identify image file 'dot_figure.png'"
+     ]
+    }
+   ],
+   "source": [
+    "import matplotlib.pyplot as plt\n",
+    "from PIL import Image\n",
+    "\n",
+    "# Save Dot figure to a file\n",
+    "dot.render(\"dot_figure.png\", format=\"png\")\n",
+    "\n",
+    "# Convert Dot figure to image\n",
+    "dot_img = Image.open(\"dot_figure.png\")\n",
     "\n",
-    "# Add legend\n",
-    "racial_patches = [plt.Rectangle((0, 0), 1, 1, color=color) for color in colors_racial]\n",
-    "gender_patches = [plt.Rectangle((0, 0), 1, 1, color=color) for color in colors_gender]\n",
-    "proficiency_patches = [plt.Rectangle((0, 0), 1, 1, color=color) for color in colors_proficiency]\n",
-    "racial_labels = ['Black', 'Asian', 'Hispanic', 'Other', 'White']\n",
-    "gender_labels = ['Male', 'Female']\n",
-    "proficiency_labels = ['Proficient', 'Limited Prof.']\n",
-    "legend_handles = racial_patches + gender_patches + proficiency_patches\n",
-    "legend_labels = racial_labels + gender_labels + proficiency_labels\n",
-    "fig.legend(legend_handles, legend_labels, loc='lower center', ncol=3, bbox_to_anchor=(0.5, -0.1))\n",
+    "# Create subplots for both figures\n",
+    "fig, axes = plt.subplots(1, 2, figsize=(15, 5))  # Adjust figsize as needed\n",
     "\n",
+    "# Plot Dot figure\n",
+    "axes[0].imshow(dot_img)\n",
+    "axes[0].axis(\"off\")\n",
+    "\n",
+    "# Plot Matplotlib figure with barplots\n",
+    "# Assuming `fig` contains your Matplotlib figure with barplots\n",
+    "axes[1].imshow(fig.canvas.renderer._renderer)\n",
+    "axes[1].axis(\"off\")\n",
+    "\n",
+    "# Adjust layout\n",
     "plt.tight_layout()\n",
-    "plt.show()\n"
+    "\n",
+    "# Show the plots\n",
+    "plt.show()"
    ]
   }
  ],

From 5ee6b5df487844f181bee84d56fc5b9fcbd60194 Mon Sep 17 00:00:00 2001
From: Eljas Roellin <eljas.roellin@ikmail.com>
Date: Wed, 14 Feb 2024 14:13:59 +0100
Subject: [PATCH 04/46] population to cohort

---
 ehrapy/tools/__init__.py                      |  2 +-
 .../__init__.py                               |  0
 .../_cohort_tracker.py}                       | 51 +++++++++----------
 3 files changed, 25 insertions(+), 28 deletions(-)
 rename ehrapy/tools/{population_tracking => cohort_tracking}/__init__.py (100%)
 rename ehrapy/tools/{population_tracking/_pop_tracker.py => cohort_tracking/_cohort_tracker.py} (88%)

diff --git a/ehrapy/tools/__init__.py b/ehrapy/tools/__init__.py
index d594c8d2..d0a0525f 100644
--- a/ehrapy/tools/__init__.py
+++ b/ehrapy/tools/__init__.py
@@ -1,8 +1,8 @@
 from ehrapy.tools._sa import anova_glm, cox_ph, glm, kmf, ols, test_kmf_logrank, test_nested_f_statistic
 from ehrapy.tools._scanpy_tl_api import *  # noqa: F403
 from ehrapy.tools.causal._dowhy import causal_inference
+from ehrapy.tools.cohort_tracking._cohort_tracker import CohortTracker
 from ehrapy.tools.feature_ranking._rank_features_groups import filter_rank_features_groups, rank_features_groups
-from ehrapy.tools.population_tracking._pop_tracker import PopulationTracker
 
 try:  # pragma: no cover
     from ehrapy.tools.nlp._medcat import (
diff --git a/ehrapy/tools/population_tracking/__init__.py b/ehrapy/tools/cohort_tracking/__init__.py
similarity index 100%
rename from ehrapy/tools/population_tracking/__init__.py
rename to ehrapy/tools/cohort_tracking/__init__.py
diff --git a/ehrapy/tools/population_tracking/_pop_tracker.py b/ehrapy/tools/cohort_tracking/_cohort_tracker.py
similarity index 88%
rename from ehrapy/tools/population_tracking/_pop_tracker.py
rename to ehrapy/tools/cohort_tracking/_cohort_tracker.py
index ca1be393..2ef6d74e 100644
--- a/ehrapy/tools/population_tracking/_pop_tracker.py
+++ b/ehrapy/tools/cohort_tracking/_cohort_tracker.py
@@ -18,19 +18,6 @@ def _check_columns_exist(df, columns):
 
 # from tableone: https://github.com/tompollard/tableone/blob/bfd6fbaa4ed3e9f59e1a75191c6296a2a80ccc64/tableone/tableone.py#L555
 def _detect_categorical_columns(data) -> list:
-    """
-    Detect categorical columns if they are not specified.
-
-    Parameters
-    ----------
-        data : pandas DataFrame
-            The input dataset.
-
-    Returns
-    ----------
-        likely_cat : list
-            List of variables that appear to be categorical.
-    """
     # assume all non-numerical and date columns are categorical
     numeric_cols = set(data._get_numeric_data().columns.values)
     date_cols = set(data.select_dtypes(include=[np.datetime64]).columns)
@@ -46,12 +33,22 @@ def _detect_categorical_columns(data) -> list:
     return likely_cat_no_dates
 
 
-class PopulationTracker:
+class CohortTracker:
     def __init__(self, adata: AnnData, columns: list = None, categorical: list = None, *args: Any):
-        """
+        """Track cohort changes over multiple filtering or processing steps.
+
+        This class offers functionality to track and plot cohort changes over multiple filtering or processing steps,
+        enabling the user to monitor the impact of each step on the cohort.
+
+        Tightly interacting with the `tableone` package [1].
 
         categorical : list, optional
         List of columns that contain categorical variables.
+
+        References
+        ----------
+        [1] Tom Pollard, Alistair E.W. Johnson, Jesse D. Raffa, Roger G. Mark; tableone: An open source Python package for producing summary statistics for research papers, Journal of the American Medical Informatics Association, Volume 24, Issue 2, 1 March 2017, Pages 267–271, https://doi.org/10.1093/jamia/ocw117
+
         """
         if columns is not None:
             _check_columns_exist(adata.obs, columns)
@@ -131,7 +128,7 @@ def reset(self):
     def tracked_steps(self):
         return self._tracked_steps
 
-    def plot_population_change(
+    def plot_cohort_change(
         self,
         set_axis_labels=True,
         subfigure_title: bool = False,
@@ -141,9 +138,9 @@ def plot_population_change(
         subplots_kwargs: dict = None,
         legend_kwargs: dict = None,
     ):
-        """Plot the population change over the tracked steps.
+        """Plot the cohort change over the tracked steps.
 
-        Create stacked bar plots to monitor population changes over the steps tracked with `PopulationTracker`.
+        Create stacked bar plots to monitor cohort changes over the steps tracked with `CohortTracker`.
 
         Args:
             set_axis_labels: If `True`, the y-axis labels will be set to the column names.
@@ -163,11 +160,11 @@ def plot_population_change(
                 import ehrapy as ep
 
                 adata = ep.dt.diabetes_130(columns_obs_only=["gender", "race", "weight", "age"])
-                pop_track = ep.tl.PopulationTracker(adata)
-                pop_track(adata, label="original")
+                cohort_tracker = ep.tl.CohortTracker(adata)
+                cohort_tracker(adata, label="original")
                 adata = adata[:1000]
-                pop_track(adata, label="filtered cohort", operations_done="filtered to first 1000 entries")
-                pop_track.plot_flowchart()
+                cohort_tracker(adata, label="filtered cohort", operations_done="filtered to first 1000 entries")
+                cohort_tracker.plot_cohort_change()
         Preview:
             .. image:: /_static/docstring_previews/flowchart.png
         """
@@ -269,7 +266,7 @@ def plot_population_change(
     def plot_flowchart(self, save: str = None, return_plot: bool = True):
         """Flowchart over the tracked steps.
 
-        Create a simple flowchart of data preparation steps tracked with `PopulationTracker`.
+        Create a simple flowchart of data preparation steps tracked with `CohortTracker`.
 
         Args:
             save: If a string is provided, the plot will be saved to the path specified.
@@ -284,11 +281,11 @@ def plot_flowchart(self, save: str = None, return_plot: bool = True):
                 import ehrapy as ep
 
                 adata = ep.dt.diabetes_130(columns_obs_only=["gender", "race", "weight", "age"])
-                pop_track = ep.tl.PopulationTracker(adata)
-                pop_track(adata, label="original")
+                cohort_tracker = ep.tl.CohortTracker(adata)
+                cohort_tracker(adata, label="original")
                 adata = adata[:1000]
-                pop_track(adata, label="filtered cohort", operations_done="filtered to first 1000 entries")
-                pop_track.plot_flowchart()
+                cohort_tracker(adata, label="filtered cohort", operations_done="filtered to first 1000 entries")
+                cohort_tracker.plot_flowchart()
         Preview:
             .. image:: /_static/docstring_previews/flowchart.png
 

From 2e28f2b962ab28cd51cf3d0f2124746b0f08bd9b Mon Sep 17 00:00:00 2001
From: Eljas Roellin <eljas.roellin@ikmail.com>
Date: Wed, 14 Feb 2024 18:57:01 +0100
Subject: [PATCH 05/46] cohort logging with tests

---
 .../tools/cohort_tracking/_cohort_tracker.py  |  46 +++--
 .../cohort_tracking/test_cohort_tracking.py   | 169 ++++++++++++++++++
 tests/tools/ehrapy_data/dataset1.csv          |  13 ++
 3 files changed, 215 insertions(+), 13 deletions(-)
 create mode 100644 tests/tools/cohort_tracking/test_cohort_tracking.py
 create mode 100644 tests/tools/ehrapy_data/dataset1.csv

diff --git a/ehrapy/tools/cohort_tracking/_cohort_tracker.py b/ehrapy/tools/cohort_tracking/_cohort_tracker.py
index 2ef6d74e..17eb2c74 100644
--- a/ehrapy/tools/cohort_tracking/_cohort_tracker.py
+++ b/ehrapy/tools/cohort_tracking/_cohort_tracker.py
@@ -34,43 +34,61 @@ def _detect_categorical_columns(data) -> list:
 
 
 class CohortTracker:
-    def __init__(self, adata: AnnData, columns: list = None, categorical: list = None, *args: Any):
+    def __init__(self, adata: AnnData | pd.DataFrame, columns: list = None, categorical: list = None, *args: Any):
         """Track cohort changes over multiple filtering or processing steps.
 
         This class offers functionality to track and plot cohort changes over multiple filtering or processing steps,
         enabling the user to monitor the impact of each step on the cohort.
 
         Tightly interacting with the `tableone` package [1].
-
-        categorical : list, optional
-        List of columns that contain categorical variables.
+        Args:
+            adata: :class:`~anndata.AnnData` or :class:`~pandas.DataFrame` object to track.
+            columns: List of columns to track. If `None`, all columns will be tracked.
+            categorical: List of columns that contain categorical variables, if not given will be inferred from the data.
 
         References
         ----------
         [1] Tom Pollard, Alistair E.W. Johnson, Jesse D. Raffa, Roger G. Mark; tableone: An open source Python package for producing summary statistics for research papers, Journal of the American Medical Informatics Association, Volume 24, Issue 2, 1 March 2017, Pages 267–271, https://doi.org/10.1093/jamia/ocw117
 
         """
+        if isinstance(adata, AnnData):
+            df = adata.obs
+        elif isinstance(adata, pd.DataFrame):
+            df = adata
+        else:
+            raise ValueError("adata must be an AnnData or a DataFrame.")
+
+        self.columns = columns if columns is not None else list(df.columns)
+
         if columns is not None:
-            _check_columns_exist(adata.obs, columns)
-            _check_columns_exist(adata.obs, categorical)
+            _check_columns_exist(df, columns)
+        if categorical is not None:
+            _check_columns_exist(df, categorical)
+            if set(categorical).difference(set(self.columns)):
+                raise ValueError("categorical columns must be in the (selected) columns.")
 
         self._tracked_steps: int = 0
         self._tracked_text: list = []
         self._tracked_operations: list = []
 
-        self.columns = columns if columns is not None else adata.obs.columns
-
         # if categorical columns specified, use them
         # else, follow tableone's logic
-        self.categorical = categorical if categorical is not None else _detect_categorical_columns(adata.obs[columns])
-        self.track = self._get_column_structure(adata.obs)
+        self.categorical = categorical if categorical is not None else _detect_categorical_columns(df[self.columns])
+        self.track = self._get_column_structure(df)
 
         self._track_backup = copy.deepcopy(self.track)
 
     def __call__(
         self, adata: AnnData, label: str = None, operations_done: str = None, *args: Any, **tableone_kwargs: Any
     ) -> Any:
-        _check_columns_exist(adata.obs, self.columns)
+        if isinstance(adata, AnnData):
+            df = adata.obs
+        elif isinstance(adata, pd.DataFrame):
+            df = adata
+        else:
+            raise ValueError("adata must be an AnnData or a DataFrame.")
+
+        _check_columns_exist(df, self.columns)
 
         # track a small text with each tracking step, for the flowchart
         track_text = label if label is not None else f"Cohort {self.tracked_steps}"
@@ -82,7 +100,7 @@ def __call__(
 
         self._tracked_steps += 1
 
-        t1 = TableOne(adata.obs, categorical=self.categorical, **tableone_kwargs)
+        t1 = TableOne(df, columns=self.columns, categorical=self.categorical, **tableone_kwargs)
         # track new stuff
         self._get_column_dicts(t1)
 
@@ -123,6 +141,7 @@ def reset(self):
         self.track = self._track_backup
         self._tracked_steps = 0
         self._tracked_text = []
+        self._tracked_operations = []
 
     @property
     def tracked_steps(self):
@@ -245,7 +264,8 @@ def plot_cohort_change(
 
         # Add legend
         tot_legend_kwargs = {"loc": "best", "bbox_to_anchor": (1, 1)}
-        tot_legend_kwargs.update(legend_kwargs)
+        if legend_kwargs is not None:
+            tot_legend_kwargs.update(legend_kwargs)
 
         plt.legend(legend_labels, **tot_legend_kwargs)
 
diff --git a/tests/tools/cohort_tracking/test_cohort_tracking.py b/tests/tools/cohort_tracking/test_cohort_tracking.py
new file mode 100644
index 00000000..a9b0cbfa
--- /dev/null
+++ b/tests/tools/cohort_tracking/test_cohort_tracking.py
@@ -0,0 +1,169 @@
+from pathlib import Path
+
+import numpy as np
+import pandas as pd
+import pytest
+
+import ehrapy as ep
+import ehrapy.tools.feature_ranking._rank_features_groups as _utils
+from ehrapy.io._read import read_csv
+
+CURRENT_DIR = Path(__file__).parent
+_TEST_DATA_PATH = f"{CURRENT_DIR.parent}/test_data_features_ranking"
+
+
+def _compare_dict_equal(dict1, dict2, tolerance=1e-9):
+    if isinstance(dict1, dict) and isinstance(dict2, dict):
+        if set(dict1.keys()) != set(dict2.keys()):
+            return False
+        for key in dict1.keys():
+            if not _compare_dict_equal(dict1[key], dict2[key], tolerance):
+                return False
+        return True
+    elif isinstance(dict1, list) and isinstance(dict2, list):
+        if len(dict1) != len(dict2):
+            return False
+        for val1, val2 in zip(dict1, dict2):
+            if not _compare_dict_equal(val1, val2, tolerance):
+                return False
+        return True
+    elif isinstance(dict1, float) and isinstance(dict2, float):
+        return abs(dict1 - dict2) < tolerance
+    elif isinstance(dict1, str) and isinstance(dict2, str):
+        return dict1 == dict2
+    else:
+        return dict1 == dict2
+
+
+class TestCohortTracker:
+    @pytest.mark.parametrize("columns", [None, ["glucose", "weight", "disease", "station"]])
+    def test_CohortTracker_init_vanilla(self, columns):
+        adata = ep.io.read_csv(
+            f"{_TEST_DATA_PATH}/dataset1.csv", columns_obs_only=["glucose", "weight", "disease", "station"]
+        )
+
+        ct = ep.tl.CohortTracker(adata, columns)
+        assert ct._tracked_steps == 0
+        assert ct.tracked_steps == 0
+        assert ct._tracked_text == []
+        assert ct._tracked_operations == []
+
+        target_track = {
+            "glucose": [],
+            "weight": [],
+            "disease": {"A": [], "B": [], "C": []},
+            "station": {"ICU": [], "MICU": []},
+        }
+        assert _compare_dict_equal(ct.track, target_track)
+
+    def test_CohortTracker_init_set_columns(self):
+        adata = ep.io.read_csv(
+            f"{_TEST_DATA_PATH}/dataset1.csv", columns_obs_only=["glucose", "weight", "disease", "station"]
+        )
+        # limit columns
+        ct = ep.tl.CohortTracker(adata, columns=["glucose", "disease"])
+        target_track = {
+            "glucose": [],
+            "disease": {"A": [], "B": [], "C": []},
+        }
+        assert _compare_dict_equal(ct.track, target_track)
+
+        # invalid column
+        with pytest.raises(ValueError):
+            ep.tl.CohortTracker(
+                adata,
+                columns=["glucose", "disease", "non_existing_column"],
+            )
+
+        # force categoricalization
+        ct = ep.tl.CohortTracker(adata, columns=["glucose", "disease"], categorical=["glucose", "disease"])
+        target_track = {
+            "glucose": {70: [], 80: [], 85: [], 90: [], 95: [], 120: [], 125: [], 130: [], 135: []},
+            "disease": {"A": [], "B": [], "C": []},
+        }
+        assert _compare_dict_equal(ct.track, target_track)
+
+        # invalid category
+        with pytest.raises(ValueError):
+            ep.tl.CohortTracker(
+                adata,
+                columns=["glucose", "disease"],
+                categorical=["station"],
+            )
+
+    def test_CohortTracker_call(self):
+        adata = ep.io.read_csv(
+            f"{_TEST_DATA_PATH}/dataset1.csv", columns_obs_only=["glucose", "weight", "disease", "station"]
+        )
+
+        ct = ep.tl.CohortTracker(adata)
+
+        ct(adata)
+        assert ct.tracked_steps == 1
+        assert ct._tracked_text == ["Cohort 0\n (n=12)"]
+        target_track_1 = {
+            "glucose": ["105.0 (23.6)"],
+            "weight": ["76.0 (14.9)"],
+            "disease": {"A": [33.3], "B": [33.3], "C": [33.3]},
+            "station": {"ICU": [50.0], "MICU": [50.0]},
+        }
+        assert _compare_dict_equal(ct.track, target_track_1)
+
+        ct(adata)
+        assert ct.tracked_steps == 2
+        assert ct._tracked_text == ["Cohort 0\n (n=12)", "Cohort 1\n (n=12)"]
+        target_track_2 = {
+            "glucose": ["105.0 (23.6)", "105.0 (23.6)"],
+            "weight": ["76.0 (14.9)", "76.0 (14.9)"],
+            "disease": {"A": [33.3, 33.3], "B": [33.3, 33.3], "C": [33.3, 33.3]},
+            "station": {"ICU": [50.0, 50.0], "MICU": [50.0, 50.0]},
+        }
+        assert _compare_dict_equal(ct.track, target_track_2)
+
+    def test_CohortTracker_reset(self):
+        adata = ep.io.read_csv(
+            f"{_TEST_DATA_PATH}/dataset1.csv", columns_obs_only=["glucose", "weight", "disease", "station"]
+        )
+
+        ct = ep.tl.CohortTracker(adata)
+
+        ct(adata)
+        ct(adata)
+
+        ct.reset()
+        assert ct.tracked_steps == 0
+        assert ct._tracked_text == []
+        assert ct._tracked_operations == []
+
+        target_track = {
+            "glucose": [],
+            "weight": [],
+            "disease": {"A": [], "B": [], "C": []},
+            "station": {"ICU": [], "MICU": []},
+        }
+        assert _compare_dict_equal(ct.track, target_track)
+        assert _compare_dict_equal(ct._track_backup, target_track)
+
+    def test_CohortTracker_flowchart(self):
+        adata = ep.io.read_csv(
+            f"{_TEST_DATA_PATH}/dataset1.csv", columns_obs_only=["glucose", "weight", "disease", "station"]
+        )
+
+        ct = ep.tl.CohortTracker(adata)
+
+        ct(adata, label="First step", operations_done="Some operations")
+        ct(adata, label="Second step", operations_done="Some other operations")
+
+        ct.plot_flowchart()
+
+    def test_CohortTracker_plot_cohort_change(self):
+        adata = ep.io.read_csv(
+            f"{_TEST_DATA_PATH}/dataset1.csv", columns_obs_only=["glucose", "weight", "disease", "station"]
+        )
+
+        ct = ep.tl.CohortTracker(adata)
+
+        ct(adata)
+        ct(adata)
+
+        ct.plot_cohort_change(return_plot=True)
diff --git a/tests/tools/ehrapy_data/dataset1.csv b/tests/tools/ehrapy_data/dataset1.csv
new file mode 100644
index 00000000..1569f780
--- /dev/null
+++ b/tests/tools/ehrapy_data/dataset1.csv
@@ -0,0 +1,13 @@
+idx,sys_bp_entry,dia_bp_entry,glucose,weight,disease,station
+1,138,78,80,77,A,ICU
+2,139,79,90,76,A,ICU
+3,140,80,120,60,A,MICU
+4,141,81,130,90,A,MICU
+5,148,77,80,110,B,ICU
+6,149,78,135,78,B,ICU
+7,150,79,125,56,B,MICU
+8,151,80,95,76,B,MICU
+9,158,55,70,67,C,ICU
+10,159,56,85,82,C,ICU
+11,160,57,125,59,C,MICU
+12,161,58,125,81,C,MICU

From 3673d285ff1581c8fd6a75279c47a4fd9843d562 Mon Sep 17 00:00:00 2001
From: Eljas Roellin <eljas.roellin@ikmail.com>
Date: Thu, 15 Feb 2024 09:14:35 +0100
Subject: [PATCH 06/46] toy notebook cleaned

---
 cohort_tracking.ipynb |  460 ++++++++++
 try_flowchart.ipynb   | 1915 -----------------------------------------
 2 files changed, 460 insertions(+), 1915 deletions(-)
 create mode 100644 cohort_tracking.ipynb
 delete mode 100644 try_flowchart.ipynb

diff --git a/cohort_tracking.ipynb b/cohort_tracking.ipynb
new file mode 100644
index 00000000..6ad27c3d
--- /dev/null
+++ b/cohort_tracking.ipynb
@@ -0,0 +1,460 @@
+{
+ "cells": [
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "## Cohort Tracking with ehrapy\n",
+    "Important for many reasons"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 2,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "import pandas as pd\n",
+    "import matplotlib.pyplot as plt\n",
+    "import numpy as np\n",
+    "import ehrapy as ep\n",
+    "from tableone import TableOne\n",
+    "import seaborn as sns\n",
+    "import scanpy as sc"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "### Tableone\n",
+    "nice package"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 15,
+   "metadata": {},
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "\u001b[1;35m2024-02-15 09:11:49,986\u001b[0m - \u001b[1;34mroot\u001b[0m \u001b[1;37mINFO - Transformed passed DataFrame into an AnnData object with n_obs x n_vars = `101766` x `48`.\u001b[0m\n"
+     ]
+    },
+    {
+     "data": {
+      "text/html": [
+       "<div>\n",
+       "<style scoped>\n",
+       "    .dataframe tbody tr th:only-of-type {\n",
+       "        vertical-align: middle;\n",
+       "    }\n",
+       "\n",
+       "    .dataframe tbody tr th {\n",
+       "        vertical-align: top;\n",
+       "    }\n",
+       "\n",
+       "    .dataframe thead th {\n",
+       "        text-align: right;\n",
+       "    }\n",
+       "</style>\n",
+       "<table border=\"1\" class=\"dataframe\">\n",
+       "  <thead>\n",
+       "    <tr style=\"text-align: right;\">\n",
+       "      <th></th>\n",
+       "      <th></th>\n",
+       "      <th>Missing</th>\n",
+       "      <th>Overall</th>\n",
+       "    </tr>\n",
+       "  </thead>\n",
+       "  <tbody>\n",
+       "    <tr>\n",
+       "      <th>n</th>\n",
+       "      <th></th>\n",
+       "      <td></td>\n",
+       "      <td>101766</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th rowspan=\"2\" valign=\"top\">gender, n (%)</th>\n",
+       "      <th>0.0</th>\n",
+       "      <td>3</td>\n",
+       "      <td>54708 (53.8)</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>1.0</th>\n",
+       "      <td></td>\n",
+       "      <td>47055 (46.2)</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th rowspan=\"5\" valign=\"top\">race, n (%)</th>\n",
+       "      <th>AfricanAmerican</th>\n",
+       "      <td>2273</td>\n",
+       "      <td>19210 (19.3)</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>Asian</th>\n",
+       "      <td></td>\n",
+       "      <td>641 (0.6)</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>Caucasian</th>\n",
+       "      <td></td>\n",
+       "      <td>76099 (76.5)</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>Hispanic</th>\n",
+       "      <td></td>\n",
+       "      <td>2037 (2.0)</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>Other</th>\n",
+       "      <td></td>\n",
+       "      <td>1506 (1.5)</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th rowspan=\"14\" valign=\"top\">time_in_hospital_days, n (%)</th>\n",
+       "      <th>1</th>\n",
+       "      <td>0</td>\n",
+       "      <td>14208 (14.0)</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>10</th>\n",
+       "      <td></td>\n",
+       "      <td>2342 (2.3)</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>11</th>\n",
+       "      <td></td>\n",
+       "      <td>1855 (1.8)</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>12</th>\n",
+       "      <td></td>\n",
+       "      <td>1448 (1.4)</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>13</th>\n",
+       "      <td></td>\n",
+       "      <td>1210 (1.2)</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>14</th>\n",
+       "      <td></td>\n",
+       "      <td>1042 (1.0)</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>2</th>\n",
+       "      <td></td>\n",
+       "      <td>17224 (16.9)</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>3</th>\n",
+       "      <td></td>\n",
+       "      <td>17756 (17.4)</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>4</th>\n",
+       "      <td></td>\n",
+       "      <td>13924 (13.7)</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>5</th>\n",
+       "      <td></td>\n",
+       "      <td>9966 (9.8)</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>6</th>\n",
+       "      <td></td>\n",
+       "      <td>7539 (7.4)</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>7</th>\n",
+       "      <td></td>\n",
+       "      <td>5859 (5.8)</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>8</th>\n",
+       "      <td></td>\n",
+       "      <td>4391 (4.3)</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>9</th>\n",
+       "      <td></td>\n",
+       "      <td>3002 (2.9)</td>\n",
+       "    </tr>\n",
+       "  </tbody>\n",
+       "</table>\n",
+       "</div><br />"
+      ],
+      "text/plain": [
+       "                                             Missing       Overall\n",
+       "n                                                           101766\n",
+       "gender, n (%)                0.0                   3  54708 (53.8)\n",
+       "                             1.0                      47055 (46.2)\n",
+       "race, n (%)                  AfricanAmerican    2273  19210 (19.3)\n",
+       "                             Asian                       641 (0.6)\n",
+       "                             Caucasian                76099 (76.5)\n",
+       "                             Hispanic                   2037 (2.0)\n",
+       "                             Other                      1506 (1.5)\n",
+       "time_in_hospital_days, n (%) 1                     0  14208 (14.0)\n",
+       "                             10                         2342 (2.3)\n",
+       "                             11                         1855 (1.8)\n",
+       "                             12                         1448 (1.4)\n",
+       "                             13                         1210 (1.2)\n",
+       "                             14                         1042 (1.0)\n",
+       "                             2                        17224 (16.9)\n",
+       "                             3                        17756 (17.4)\n",
+       "                             4                        13924 (13.7)\n",
+       "                             5                          9966 (9.8)\n",
+       "                             6                          7539 (7.4)\n",
+       "                             7                          5859 (5.8)\n",
+       "                             8                          4391 (4.3)\n",
+       "                             9                          3002 (2.9)"
+      ]
+     },
+     "execution_count": 15,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "adata = ep.dt.diabetes_130(columns_obs_only=[\"gender\", \"race\", \"time_in_hospital_days\"])\n",
+    "TableOne(adata.obs)"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "### CohortTracker\n",
+    "A visualization aid automated in ehrapy: summarizing tableone information graphically.\n",
+    "Especially useful for cohort processing, as the overview component becomes even more important there"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 16,
+   "metadata": {},
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "{'gender': {0.0: [53.8, 52.4], 1.0: [46.2, 47.6]}, 'race': {'AfricanAmerican': [19.3, 25.9], 'Asian': [0.6, 0.7], 'Caucasian': [76.5, 70.4], 'Hispanic': [2.0, 0.8], 'Other': [1.5, 2.2]}, 'time_in_hospital_days': ['4.4 (3.0)', '4.6 (3.2)']}\n"
+     ]
+    },
+    {
+     "data": {
+      "image/png": "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",
+      "text/plain": [
+       "<Figure size 640x480 with 2 Axes>"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "image/svg+xml": [
+       "<?xml version=\"1.0\" encoding=\"UTF-8\" standalone=\"no\"?>\n",
+       "<!DOCTYPE svg PUBLIC \"-//W3C//DTD SVG 1.1//EN\"\n",
+       " \"http://www.w3.org/Graphics/SVG/1.1/DTD/svg11.dtd\">\n",
+       "<!-- Generated by graphviz version 9.0.0 (20230911.1827)\n",
+       " -->\n",
+       "<!-- Pages: 1 -->\n",
+       "<svg width=\"201pt\" height=\"143pt\"\n",
+       " viewBox=\"0.00 0.00 200.50 142.50\" xmlns=\"http://www.w3.org/2000/svg\" xmlns:xlink=\"http://www.w3.org/1999/xlink\">\n",
+       "<g id=\"graph0\" class=\"graph\" transform=\"scale(1 1) rotate(0) translate(4 138.5)\">\n",
+       "<polygon fill=\"white\" stroke=\"none\" points=\"-4,4 -4,-138.5 196.5,-138.5 196.5,4 -4,4\"/>\n",
+       "<!-- 0 -->\n",
+       "<g id=\"node1\" class=\"node\">\n",
+       "<title>0</title>\n",
+       "<polygon fill=\"lightgrey\" stroke=\"black\" points=\"86.5,-134.5 0,-134.5 0,-93.5 86.5,-93.5 86.5,-134.5\"/>\n",
+       "<text text-anchor=\"middle\" x=\"43.25\" y=\"-117.2\" font-family=\"Times,serif\" font-size=\"14.00\">Initial cohort</text>\n",
+       "<text text-anchor=\"middle\" x=\"43.25\" y=\"-100.7\" font-family=\"Times,serif\" font-size=\"14.00\"> (n=101766)</text>\n",
+       "</g>\n",
+       "<!-- 1 -->\n",
+       "<g id=\"node2\" class=\"node\">\n",
+       "<title>1</title>\n",
+       "<polygon fill=\"lightgrey\" stroke=\"black\" points=\"78.62,-41 7.88,-41 7.88,0 78.62,0 78.62,-41\"/>\n",
+       "<text text-anchor=\"middle\" x=\"43.25\" y=\"-23.7\" font-family=\"Times,serif\" font-size=\"14.00\">Cohort 1</text>\n",
+       "<text text-anchor=\"middle\" x=\"43.25\" y=\"-7.2\" font-family=\"Times,serif\" font-size=\"14.00\"> (n=1000)</text>\n",
+       "</g>\n",
+       "<!-- 0&#45;&gt;1 -->\n",
+       "<g id=\"edge1\" class=\"edge\">\n",
+       "<title>0&#45;&gt;1</title>\n",
+       "<path fill=\"none\" stroke=\"black\" d=\"M43.25,-93.15C43.25,-81.28 43.25,-65.94 43.25,-52.5\"/>\n",
+       "<polygon fill=\"black\" stroke=\"black\" points=\"46.75,-52.83 43.25,-42.83 39.75,-52.83 46.75,-52.83\"/>\n",
+       "<text text-anchor=\"middle\" x=\"117.88\" y=\"-62.2\" font-family=\"Times,serif\" font-size=\"14.00\">filtered to first 1000 entries</text>\n",
+       "</g>\n",
+       "</g>\n",
+       "</svg>\n"
+      ],
+      "text/plain": [
+       "<graphviz.graphs.Digraph at 0x2bedb7f90>"
+      ]
+     },
+     "execution_count": 16,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "# instantiate the cohort tracker\n",
+    "pop_track = ep.tl.CohortTracker(adata, categorical=[\"gender\", \"race\"])\n",
+    "\n",
+    "# track the initial state of the dataset\n",
+    "pop_track(adata, label=\"Initial cohort\")\n",
+    "\n",
+    "# do a filtering step\n",
+    "adata = adata[:1000]\n",
+    "\n",
+    "# track the filtered dataset\n",
+    "pop_track(adata, label=\"Cohort 1\", operations_done=\"filtered to first 1000 entries\")\n",
+    "\n",
+    "print(pop_track.track)\n",
+    "\n",
+    "# plot the change of the cohort\n",
+    "pop_track.plot_cohort_change()\n",
+    "\n",
+    "# plot a flowchart\n",
+    "pop_track.plot_flowchart()"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "The tracking steps can be reset for convenience"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 17,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "pop_track.reset()"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 18,
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAnUAAAHLCAYAAACu327MAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjguMiwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy8g+/7EAAAACXBIWXMAAA9hAAAPYQGoP6dpAAB3XUlEQVR4nO3deVxU9f7H8dcMOww7KogoqKCg4oZ7Kprmfl0yzbT0lm1mpkaWLTfzVnorTb1Wtigut7Jut6yfmWa4pGQuKGbua7jgLouiLMP8/iAnJwRBQXR6Px8PHjHnfM/3fM7YwJvv9ywGi8ViQURERERua8aKLkBEREREbpxCnYiIiIgdUKgTERERsQMKdSIiIiJ2QKFORERExA4o1ImIiIjYAYU6ERERETugUCciIiJiBxwrugCR0sjPz+fYsWN4enpiMBgquhwRuQaLxUJmZiZVq1bFaNQ4gkh5UqiT28qxY8cICQmp6DJEpJQOHz5MtWrVKroMEbumUCe3FU9PT6DgF4SXl1cFVyMi15KRkUFISIj1sysi5UehTm4rl6dcvby8FOpEbiM6XUKk/OkEBxERERE7oFAnIiIiYgcU6kRERETsgEKdiIiIiB1QqBMRERGxAwp1IiIiInZAtzSR29Kl8dNwdnGt6DJE/rJcp46r6BJE5E80UiciIiJiBxTqREREROyAQp2IiIiIHVCoExEREbEDCnUiIiIidkChTkRERMQOKNSJiIiI2AGFOhERERE7oFAnIiIiYgcU6kRERETsgEKdiIiIiB1QqBMRERGxAwp1IiIiInZAoU5ERETEDijUiYiIiNgBhToRERERO6BQJyIiImIHFOpERERE7IBCnYiIiIgdUKgTERERsQMKdSIiIiJ2QKFORG4ax+5tcZ06DscubW7aPg0BPri8FYfzC4+Ag37kiYj9cryRjVetWkWHDh04d+4cPj4+ZVRS6cydO5fRo0eTlpZWJv3FxsbSqFEjpk2bVib9XY8JEyawaNEikpOTK7SfYcOGkZaWxqJFi26oDrl1OT3wNxwa1QXAvGUnuQv+79obebjheFdrHOrVBi8TXMom/9gp8j5fiuVserHbObRtiiUvj7y1m62LHXu0w1g/HIO3CRwcIPMC5r2/kff9T3Auo+j+XJxx7HYHxrBqGHy9wMUJy7lM8pN3kbdyA2TnAGA5nUb+r3txiK6DQ/NozOuSS/LWiIjcdkr1Z2tsbCyjR4+2vm7dujWpqal4e3uXdV0lNnDgQPbs2VNh+y8PcXFxJCQkWF8PGzaMPn36VFxBYpccmtW3BroS83DDefT9OLZtCl4eWE6dxXI+C2NoVfA2Fb+/mHoYXJzJ330ILly0LjfWCcPg7ITl1DksaZkY/LxxbBGN8yP3FNufwd0Vx3YxGAIDsKRnQnYuxsp+ON7VGqcHetm0NW/eWVBDm0alO14RkdvIDY3UOTs7ExgYWFa1XBc3Nzfc3NwqtIayZjKZMJmK/wUpciMM/j449u1E/sGjGHw8C0a6SsCxW1uM/j7kp54iZ9bnkHmhYIWDETAUu61D40gA8rfvt1meM+M/kGe2vna6rwcOMfUwVvEHd1fIunTV/ix5ZnK/WYl53daCUTlHB5xH3IsxNBiHyFrkurnAxeyCfe48gMWcj7FqZQxV/LGcOFOi4xURuZ2UeKRu2LBhrF69munTp2MwGDAYDMydOxeDwWCd+pw7dy4+Pj4sXryYOnXq4O7uTv/+/cnKymLevHmEhobi6+vLqFGjMJv/+CGenZ1NXFwcwcHBeHh40KJFC1atWlWiui7v87IJEybQqFEjFixYQGhoKN7e3tx7771kZmaW9FDJz89n3Lhx+Pn5ERgYyIQJE2zWp6Sk0Lt3b0wmE15eXgwYMIATJ05Y12/dupUOHTrg6emJl5cXTZs2ZdOmTTb1Llq0iPDwcFxdXenSpQuHDx8udAyXv583bx5ff/219X2//N48++yzRERE4O7uTs2aNXnppZfIzc0t8XFeyWw2M3bsWHx8fPD392fcuHFYLBabNkuXLuWOO+6wtunZsyf79//xC7pjx46MHDnSZptTp07h7OxsHXl89913rcddpUoV+vfvf131yg0wGnAa0hMsFnI/Xgx/+ncujkOjOgBY0jJxfmwALpNG4xw3DGN0HbjiM12IsxOG4CoA5B9OtV2XZ8ahdSOcnxqC8/iHcYipV9Du+OkiAx1QME27aqN1mpU8M/mHjxfUl58P+VccV24elhOnCw6/ZrUSH6+IyO2kxKFu+vTptGrViocffpjU1FRSU1MJCQkp1C4rK4sZM2awcOFCli5dyqpVq+jbty9LlixhyZIlLFiwgPfff58vvvjCus3IkSNZt24dCxcu5JdffuGee+6ha9eu7N2797oOav/+/SxatIjFixezePFiVq9ezeTJk0u8/bx58/Dw8GD9+vW88cYbTJw4keXLlwMFga93796cPXuW1atXs3z5cg4cOMDAgQOt2w8ePJhq1aqxceNGkpKSeO6553BycrJ5j1577TXmz59PYmIiaWlp3HvvvVetJS4ujgEDBtC1a1fr+966dWsAPD09mTt3Ljt27GD69Ol8+OGHvP3229fzljFlyhTmzp3LnDlzWLt2LWfPnuWrr76yaXPhwgXGjh3Lpk2bSEhIwGg00rdvX/Lz8wEYPnw4n3zyCdnZ2dZt/vOf/xAcHEzHjh3ZtGkTo0aNYuLEiezevZulS5fSrl27YuvKzs4mIyPD5ktujONdbTDWqEru/5YXfw7cn5ncMbgXjIo7RNbE4OYKF7MxVq2M8/29MEZHFLmpwc8bw+8XKVxtnwZfL4w1qmKs5AtA/pETBSOBpWFyx+H3GvK37Poj7P3Ocrbg/x1DgG/p+hURuU2UePrV29sbZ2dn3N3drVOuu3btKtQuNzeX9957j1q1agHQv39/FixYwIkTJzCZTERFRdGhQwdWrlzJwIEDSUlJIT4+npSUFKpWrQoUBJmlS5cSHx/P66+/XuqDys/PZ+7cuXh6egJw//33k5CQwGuvvVai7aOjo3n55ZcBCA8PZ+bMmSQkJNC5c2cSEhLYtm0bBw8etIba+fPnU69ePTZu3EizZs1ISUnhmWeeoW7dutY+/vwezZw5kxYtWgAFITIyMpINGzbQvHlzm7Ymkwk3Nzeys7MLTXW/+OKL1u9DQ0OJi4tj4cKFjBs3rqRvldW0adMYP348/fr1A2DWrFksW7bMps3dd99t83rOnDlUqlSJHTt2UL9+ffr168fIkSP5+uuvGTBgAFAwMjls2DAMBgMpKSl4eHjQs2dPPD09qVGjBo0bNy62rkmTJvHKK6+U+njk6gzVAnG4syXmTdvJ37yjdBsb//gbMP/4aXKmzAXA+elhGAMDcLyjCTm/FHF+q6vLH9//KWwB5H37I3lL1mAI8MWxf2ccwmvgNKQnue99VqKRRIO/D04P98fg7Un+gSPkfvF94UaX/9i4shYRETtS5tf3u7u7WwMdQJUqVQgNDbU5R6xKlSqcPHkSgG3btmE2m4mIiLCeS2YymVi9erXN1F5phIaGWgMdQFBQkHV/JREdHW3z+srtd+7cSUhIiM0oZVRUFD4+PuzcWXAy9tixYxk+fDidOnVi8uTJhY7D0dGRZs2aWV/XrVvXZvuS+uyzz2jTpg2BgYGYTCZefPFFUlJSStUHQHp6OqmpqdaQebnGmJgYm3Z79+5l0KBB1KxZEy8vL0JDQwGs+3R1deX+++9nzpw5AGzevJlff/2VYcOGAdC5c2dq1KhBzZo1uf/++/n444/Jysoqtrbx48eTnp5u/bpymlpKzxgUgMHBiLFhBC6TRuMyaTT4FJxPZ4wuWIar89U3Pp+FJS8PAMuxU2DOB3N+wfcUjMYV6YrRW5yL6N9iwXLqLOYfkwBwqF0dY3iNax6ToUZVnJ8agrGyH+Zf95Hz/udXDY64/B7mLmUXXiciYgfKPNRdOc0IYDAYrrrs8pTd+fPncXBwICkpieTkZOvXzp07mT59epnVcHl/N2P7CRMmsH37dnr06MGKFSuIiooqNJV5o9atW8fgwYPp3r07ixcvZsuWLbzwwgvk5Fzll1kZ6dWrF2fPnuXDDz9k/fr1rF+/HsBmn8OHD2f58uUcOXKE+Ph4OnbsSI0aBb+YPT092bx5M59++ilBQUH84x//oGHDhsXejsbFxQUvLy+bL7lxBicnDC7OBV/GggscDA4OGFycuXzBg9NjA3F+9iEce/w+RZ6fT/7+IwVtq1YqGLkzGgu+B/JPnStyf5Yz6VjMBZ8hg98f/4aGAF+M9Wr/cY2FAYx1w/7Y0OX3z6K3CednH8L52YcwNvhj5NsYHYHziIEYTO7krUkiN/5LyM27+jH/vl/L6aLrFBG5nZXq6ldnZ2ebCxzKQuPGjTGbzZw8eZK2bduWad/lITIyksOHD3P48GHraN2OHTtIS0sjKirK2i4iIoKIiAjGjBnDoEGDiI+Pp2/fvgDk5eWxadMm61Tr7t27SUtLIzIy8qr7vNr7/tNPP1GjRg1eeOEF67Lffvvtuo7J29uboKAg1q9fbz3HLS8vj6SkJJo0aQLAmTNn2L17Nx9++KH132nt2rWF+mrQoAExMTF8+OGHfPLJJ8ycOdNmvaOjI506daJTp068/PLL+Pj4sGLFCuu0r5Qv88ZfMW/81WaZy4uPYvDzLnSfOmOADwY/byy/HbMuy/tuDcZa1TAGBuDy4qMAGHw8sZjzMSf8XPSOc3KxHD2BoXoQxpBAzJdH97xNOD/UD8ulHCxn0zCY3DF4FYzqW85lkL+n4P9pg9FYcDUsYLg8feplwumB3hiMBix5eRirB+E8aoh1l7lfLMdy9PcLmJwcMVQJACD/wJHSvm0iIreFUoW60NBQ1q9fz6FDhzCZTKUavSpKREQEgwcP5oEHHmDKlCk0btyYU6dOkZCQQHR0ND169LjhfZSlTp060aBBAwYPHsy0adPIy8tjxIgRtG/fnpiYGC5evMgzzzxD//79CQsL48iRI2zcuNHmfDQnJyeefPJJZsyYgaOjIyNHjqRly5aFzqe7LDQ0lGXLlrF79278/f3x9vYmPDyclJQUFi5cSLNmzfj2229vaDTwqaeeYvLkyYSHh1O3bl2mTp1qM4Lm6+uLv78/H3zwAUFBQaSkpPDcc89dta/hw4czcuRIPDw8rEEWYPHixRw4cIB27drh6+vLkiVLyM/Pp06dOtddt9xclpRUct79DMfubTGGBEJuHubdh8j7bg2WlNRitzVv2YmxehDGqNqY128r6O9cBuZtezBWC8RQyQ8MkH/6HPl7fiNv+U9Xn0b9ncHB+Mcoo6MjhhpVbde7OnP5bDxjZE0MDkbyj53U7UxExG6VKtTFxcUxdOhQoqKiuHjxIvHx8WVSRHx8PK+++ipPP/00R48eJSAggJYtW9KzZ88y6b8sGQwGvv76a5588knatWuH0Wika9eu/Pvf/wbAwcGBM2fO8MADD3DixAkCAgLo16+fzcn+7u7uPPvss9x3330cPXqUtm3bMnv27CL3+fDDD7Nq1SpiYmI4f/48K1eu5G9/+xtjxoxh5MiRZGdn06NHD1566aVCt18pqaeffprU1FSGDh2K0WjkwQcfpG/fvqSnF1ypaDQaWbhwIaNGjaJ+/frUqVOHGTNmEBsbW6ivQYMGMXr0aAYNGoSrq6t1uY+PD19++SUTJkzg0qVLhIeH8+mnn1KvXr3rqlnKRvar75dqueXQUXLfXVjq/Zg3bcex6x0F06sebnDhIpaz6eTGL7rmtpZzGVwa+8Y1lxXFoUnBKLo5cUup6xYRuV0YLH++GZmUq7J+rNmt6NChQ9SqVYuNGzdap2/LSkZGBt7e3pwY8QpeLq7X3kBuKY7d2+LYqRV5yxLJW5Z4U/ZpCPDB+bnhWM5lkDPpIyiDGQYB16klu8r+8mc2PT1d58SKlLMbeqKEyJVyc3M5c+YML774Ii1btizzQCe3v7wla8hbsuam7tNyOo3suLdu6j5FRCpCmV/9Wta6detmc6uTK79Kew+7lJSUIvsymUzXdTuQ20Vxx71mTdn8kk1MTCQoKIiNGzcya9asMulTRERESuaWn349evQoFy9evOo6Pz8//Pz8StxXXl4ehw4dKnJ9aGgojo72OXi5b9++ItcFBwffNs/P1fSryK1B068it55bPsEEBweXWV+Ojo7Url27zPq7nfxVj1tEROSv4paffhURERGRa1OoExEREbEDCnUiIiIidkChTkRERMQOKNSJiIiI2AGFOhERERE7oFAnIiIiYgcU6kRERETsgEKdiIiIiB1QqBMRERGxAwp1IiIiInZAoU5ERETEDijUiYiIiNgBhToRERERO6BQJyIiImIHFOpERERE7IBCnYiIiIgdUKgTERERsQOOFV2AyPVwnTQaVy+vii5DRETklqGROhERERE7oFAnIiIiYgcU6kRERETsgEKdiIiIiB1QqBMRERGxAwp1IiIiInZAoU5ERETEDijUiYiIiNgBhToRERERO6BQJyIiImIHFOpERERE7IBCnYiIiIgdUKgTERERsQMKdSIiIiJ2QKFORERExA4o1ImIiIjYAceKLkDkejw6fhrOLq4VXUa5+q7+QgBe9e9WwZXIX5VTdg8A3tnasNC6Ta953OxyROQaNFInIiIiYgcU6kRERETsgEKdiIiIiB1QqBMRERGxAwp1IiIiInZAoU5ERETEDijUiYiIiNgBhToRERERO6BQJyIiImIHFOpERERE7IBCnYiIiIgdUKgTERERsQMKdSIiIiJ2QKFORERExA4o1ImIiIjYAYU6ERERETugUCciIiJiBxTqREREROyAY0UXIHIzdG3fjEb1ahFU2Q8Pd1fSMy6wa/9hFi1L5NTZdACeG3EvkbWrF9p2z4EjvDbzk2L7D/D1ok+XNkTWro6XpwdnzmXw4/pf+G7VBiyWcjmkvwSTmw/33fVMkeuTdiWQtHsFTo7OxNTtRFjV+ri5eHDhYjp7DiezZc8qLJb8Eu3L3zuIPm0fw8Gh4MfiZwlvk37+dJkch4jIzaBQJ38Jndo2wd/Hi+OnzpKTm0dlfx/u8POmfp1Qnp30EZeyc6xtT54+R8aFi9bXR08U/4vd08ONl0ffj5enBxcv5ZB68gzVAisxsFcsPt4mPlm0otyOy96Z882cOJtis8zFyQ0fz0oAZF3KBAx0aXE/VQNqYs7PI/PCObxM/sTUvRMvDz9Wbf7imvtxMDrSsekAa6ATEbkd6SeYlEhOTg7Ozs4VXcZ1W/3zVhI3bedsWiYA9/XuSJf2Mfh4magXUYOkbXutbb9evo61G38tcd/NGtbBy9MDgH9OX8DRE2do3zKaBwd0pVObJixdtdG6Xymdi9mZfL3mfZtlbRr0wsezEpdysth7ZCuhQZFUDagJwPINn5ByYjf1wlrSJroXESGN2bb/J86kHyt2P63qd8fXszL7j26jVnCDcjseEZHypHPq5KpiY2MZOXIko0ePJiAggC5dujB16lQaNGiAh4cHISEhjBgxgvPnz9tsl5iYSGxsLO7u7vj6+tKlSxfOnTsHQH5+PpMmTSIsLAw3NzcaNmzIF19cexSlLPzfDz/bBKvdBw5bv8/NM9u0va93Rz56YyxvvvAIf7+nC14m92L7NhoN1u/zf59rtfz+XwcH41WndOX6uDi5EVG9CQA7D20gz5xDSOUIAPLyckg5sQeAg6nbrduEVA4vts/qVeoSFdaCXw/8xOETu8upchGR8qdQJ0WaN28ezs7OJCYmMmvWLIxGIzNmzGD79u3MmzePFStWMG7cOGv75ORk7rzzTqKioli3bh1r166lV69emM0FoWnSpEnMnz+fWbNmsX37dsaMGcOQIUNYvXr1TT0ug8FAbKuGQMFU6449v1nXZefkci49k8zzF6ns70Nsq4a89NQQnJ2diuxv644DXLxUMH378uj7mfj0UIb2v8u63tfbVE5H8tcTFdYSJ0dn8sy5/HpgHQAmN28ALuVmAQVh+uKlP/7YMLn5FNmfm4uJ9o37cib9OOu3Lyu3ukVEbgZNv0qRwsPDeeONN6yv69SpY/0+NDSUV199lccee4x3330XgDfeeIOYmBjra4B69eoBkJ2dzeuvv84PP/xAq1atAKhZsyZr167l/fffp3379letITs7m+zsbOvrjIyMGzomZ2cnRgzpRXTdmqRlnOft2V+S93vo/OTrFRw7fsb6un/3tvTq1IrK/j7ENAjnp6QdV+3z1Nl03nz/c/p3b0uNalXw9TKxduOvtGsejdFowGwu2Yn6Ujyj0YF6YS0A2HdkKxezzxfd2GAoet0V2jbsg5OjCyuSZmPOzyuLMkVEKoxCnRSpadOmNq9/+OEHJk2axK5du8jIyCAvL49Lly6RlZWFu7s7ycnJ3HPPPVfta9++fWRlZdG5c2eb5Tk5OTRu3LjIGiZNmsQrr7xy4wcDeHt6MGb43YSFBJJ68ixTPviv9cpXgJSjJ23ar9u8k16dCgKon49XsX3v/+0Y/3rvM+vrWjWqEtuyYDQw9eTZMqn/ry4ipDHurp5YLPn8sm+tdfn5iwX/hq7OHoABsODm4nHF+rQi+/T3DsRodKBPu8cAMBj+mLzo1/4Jth/8mQ07NIInIrcHhTopkofHH78YDx06RM+ePXn88cd57bXX8PPzY+3atTz00EPk5OTg7u6Om5tbkX1dPvfu22+/JTg42Gadi4tLkduNHz+esWPHWl9nZGQQEhJS6mMJruLPmIf7U8nPm937DzM9/isuZF2yrvc0udO6aRSrf/7FeiVsi0Z1retPnysIDr7eJsY9NhCAL5b8aL3AIjwsmH2HjmGxWHB3c2HQ32IL6j2fxY69f0zvyvWLrnUHACkn9pB2/pR1+ZGTe4kMbYajgxPVq0SQcmI3YUH1rOsPnyz4NwoNiqJ5ZMG0+OKf5pB1qWDU12gwYnQs/P+gk6MzDkaHcjseEZGyplAnJZKUlER+fj5TpkzBaCwYzfj8889t2kRHR5OQkHDVkbWoqChcXFxISUkpcqr1alxcXIoNfSX15N/7Usmv4NwrVxdnxj7c37rux59/YfueQ9zXuyMDerbn5Ok0XJyd8PctGJ07evw0Sb8UnIDvYDRStYo/AG6uf9Q1rP9d+Hp7cjYtg8r+Pri4OGM25zPvi+/JydW03o2qEVjXehuTrfvW2Kw7lLqD1DOHCPIPpXPz+8i4cBZvUwAAew8nW698dXZ0sfZh/H1E7tPlb9n0FRHSmNgmBf9v6D51InK7UaiTEqlduza5ubn8+9//plevXtaLJ640fvx4GjRowIgRI3jsscdwdnZm5cqV3HPPPQQEBBAXF8eYMWPIz8/njjvuID09ncTERLy8vBg6dGi51u/o+MeIS41qVWzWbdt1kIwLF/lm+U/UrxNGZX8fnJwcOXbiDJt/3cu3K9YXukL2z37dfYiY6AgCK/uRm2tm266D/N8P69h94Ei5HM9fzeVRupPnDnP8zCGbdRYsLP15PjF1O1Gzan28PPw4n5XG3sPJbN6zsgKqFRGpGAaLRfe7l8JiY2Np1KgR06ZNsy57++23efPNN0lLS6Ndu3YMHjyYBx54gHPnzuHj4wPA6tWref7550lKSsLNzY0WLVqwcOFCfHx8sFgszJgxg/fee48DBw7g4+NDkyZNeP7552nXrl2J6srIyMDb25t7R7yCs4trORz5reO7+gsBeNW/WwVXIn9VTtk9AHhna8NC6za95lFo2dVc/symp6fj5VX8uakicmMU6uS2olAncvMo1IncXnSfOhERERE7oFAnIiIiYgcU6kRERETsgEKdiIiIiB1QqBMRERGxAwp1IiIiInZAoU5ERETEDijUiYiIiNgBhToRERERO6BQJyIiImIHFOpERERE7IBCnYiIiIgdUKgTERERsQMKdSIiIiJ2QKFORERExA4o1ImIiIjYAYU6ERERETugUCciIiJiBwwWi8VS0UWIlFRGRgbe3t6kp6fj5eVV0eWIyDXoMyty82ikTkRERMQOKNSJiIiI2AGFOhERERE7oFAnIiIiYgcU6kRERETsgEKdiIiIiB1QqBMRERGxAwp1IiIiInZAoU5ERETEDijUiYiIiNgBhToRERERO6BQJyIiImIHFOpERERE7IBCnYiIiIgdUKgTERERsQMKdSIiIiJ2QKFORERExA4o1ImIiIjYAYU6ERERETugUCciIiJiBxwrugCR6/Hy+Gm4uLhWdBlShtrdEVzRJUgpdO83uKJLEJE/0UidiIiIiB1QqBMRERGxAwp1IiIiInZAoU5ERETEDijUiYiIiNgBhToRERERO6BQJyIiImIHFOpERERE7IBCnYiIiIgdUKgTERERsQMKdSIiIiJ2QKFORERExA4o1ImIiIjYAYU6ERERETugUCciIiJiBxTqREREROyAQp2IiIiIHVCoExEREbEDjhVdgNz+hg0bRlpaGosWLaroUm45nbq0oVOXNldd93zcmzg6OXFXtzsIDauGr68Xzi5OpJ3L5JfkXaxeuYGc7JwS7cfZxZmnnh6Kf4AvAF/993vWr0suq8OQ3zVufgdB1WoAcOzwIZI3JhJcvSYNY1oVuc3PPy7n7OmTxfbr61+Z2nXr4ePrj9HBkexLFzmReoSdvySVaf0iYt8U6kRugvPnszh7Js1mmcUCHu6u3NEuhtzcPE6dPIOXtyeVKvtx512tCQ6pwtwP/1ei/nv362QNdFI+qtWoaQ10V8rJucS5s6dtlrm5uePq5g5A9qVLxfYbGFydRs3aYDQaycm+xPnMdJycnKkcWFWhTkRKRaFOKpzFYsFsNuPoaL//O+7esZ//Lvyu0PLcPDPffrOS9eu2kpOdg6OjAw+PuJcaocHUjayFm5sLFy9mF9t3g4Z1aNqsPlu37KJh47rldQh/ae4eJqKiYzh35hSubu64uXtY1506foxTx4/ZtL/jzu64urlz6kQqF85nFNmvg4MD9Rs1x2g0sn/3dvbs2IrFYilYZ8efBxEpHzqnzo5kZmYyePBgPDw8CAoK4u233yY2NpbRo0cDkJ2dTVxcHMHBwXh4eNCiRQtWrVpl3X7u3Ln4+PiwbNkyIiMjMZlMdO3aldTUVGsbs9nM2LFj8fHxwd/fn3Hjxll/CV2Wn5/PpEmTCAsLw83NjYYNG/LFF19Y169atQqDwcB3331H06ZNcXFxYe3ateX63lS0+tF1+Oe/xvD8hBEMfehuqgZXBuB85gXWrNponWbNyzNz5PBxoOB9NOdbiuwTwNvHk373dOHI4eN8/92a8j2IvyiDwUDDZm2wYCF5Y2Kh/9//LKBKEF7eBaOmB/fuKLatf+UgnF1cAHBxdaVjt7506tGfpq3a4+LiWjYHICJ/GQp1dmTs2LEkJibyzTffsHz5ctasWcPmzZut60eOHMm6detYuHAhv/zyC/fccw9du3Zl79691jZZWVm89dZbLFiwgB9//JGUlBTi4uKs66dMmcLcuXOZM2cOa9eu5ezZs3z11Vc2dUyaNIn58+cza9Ystm/fzpgxYxgyZAirV6+2affcc88xefJkdu7cSXR0dDm9KxXPbM4nM/MC585m4OVlIrJeLR4fNcQa7K7kYXKnfnQEAFu37Cr2nDqDAQbe1wOjg5GF//k/8s3mcjuGv7LakQ3w9Qtge/JGLmZduGb7muFRAGSkneX0yePFtjWZvKzfB1evSU52Ng4ODlQJqkaLdp1xdHS6seJF5C9F4/t2IjMzk3nz5vHJJ59w5513AhAfH0/VqlUBSElJIT4+npSUFOuyuLg4li5dSnx8PK+//joAubm5zJo1i1q1agEFQXDixInW/UybNo3x48fTr18/AGbNmsWyZcus67Ozs3n99df54YcfaNWq4OTxmjVrsnbtWt5//33at29vbTtx4kQ6d+5c7HFlZ2eTnf3H9GNGRtFTWbei5M07SFyTxMWsgvOqwuuE8tCjA3BycqRVmyb87/Ol1rZ+/j78/eH+eHt7cujAEb764vti+27TNoaatavzxWffcfrUOXx9vYptL6Xn7eNHrYh6HE05yLHDh67Z3svbl4DKgQAc2Lvzmu0NRoP1+z07fmH/7l/x9a9Eq/Z34ebmTpWqIRxNOXDd9YvIX4tCnZ04cOAAubm5NG/e3LrM29ubOnXqALBt2zbMZjMRERE222VnZ+Pv72997e7ubg10AEFBQZw8WXDlXnp6OqmpqbRo0cK63tHRkZiYGOuU1L59+8jKyioU1nJycmjcuLHNspiYmGse16RJk3jllVeu2e5WdfrUOZvXe3cf4sL5LDxM7nj7elqXV69RlQce6ofJ5M6OX/fx6YJvyM3NK7bvoKqVAOjV50569bkTwxXrevbpSJOYerz374/L7Fj+ikxePhiNRgKDQ6hSdQAADg4FPzYDg6tzV1AwK5Z8RV5eLgBh4ZEAXMy6QOqR367Z/6WLWdbv08+dsfkvgLuHR6FtRESKolD3F3H+/HkcHBxISkrCwcHBZp3JZLJ+7+RkO91jMBiueQ7Rn/cD8O233xIcHGyzzuX3c4cu8yjBL6zx48czduxY6+uMjAxCQkJKXE9Fa9+xOcmbd5KelglA7YgaeJgKroo8d7Zg1LF+dAQDB/fAycmJxDVJLF6UwJ/fci9vE8MfGwjAsiU/sn3bH1PmLi7Ohfbr5OSIk7M+3mXlcpC7ktFoxGg0cjlNu7q5W6+OPbRvV6HPjbevv/XWJ1s3rSP93BnOnDqBxZKPwWDE29eP0ydT8fb544+sC+czy+mIRMQe6ae+nahZsyZOTk5s3LiR6tWrAwUja3v27KFdu3Y0btwYs9nMyZMnadu27XXtw9vbm6CgINavX0+7du0AyMvLIykpiSZNmgAQFRWFi4sLKSkpNlOt18vFxaVQGLydtGzdmC7d25OelkFOTi6VKhf8ws7OziHxx014epm474HeGI0G8vLyCKkexOOjhli3X/TFco4dPYGD0UjlKgXburoWvB//XfidzRW1vr5ePPvSY4DuU1dWjqYcKDT9GdulN+4eJut96i4Lq10Xo9FIbk4OKYf2FerLwcEBk6e39XsoGKk7tH8PYbXrEhHVkKrVQnHzKPgjKzMjjeNHU8rr0ETEDinU2QlPT0+GDh3KM888g5+fH5UrV+bll1/GaDRiMBiIiIhg8ODBPPDAA0yZMoXGjRtz6tQpEhISiI6OpkePHiXaz1NPPcXkyZMJDw+nbt26TJ06lbS0NJs64uLiGDNmDPn5+dxxxx2kp6eTmJiIl5cXQ4cOLad34Na08oefadCwDlUCA/Dz9yDtXDq/HTpKwvfrOH3qLL6+Xhh/P6/K0dGR6jWq2mzv6lp4FE5uPY6OTlQLrQ1AyqF9mPOKnzq/0s5fksi+dJGQ0Fq4mzzJvpTFkd+OsWfHL+Tn55dXySJihxTq7MjUqVN57LHH6NmzJ15eXowbN47Dhw/j6lpwa4T4+HheffVVnn76aY4ePUpAQAAtW7akZ8+eJd7H008/TWpqKkOHDsVoNPLggw/St29f0tPTrW3++c9/UqlSJSZNmsSBAwfw8fGhSZMmPP/882V+zLe6DT9vZcPPW4tcf+5cBs+NfeOa/ZSkXUn7khuzatnXhZbl5eWy/P8+L3a7s6dPsuTLq5/jeGDPDg7sKf72JyIi12KwlOaEKbmtXLhwgeDgYKZMmcJDDz1U0eWUiYyMDLy9vRk94hXdx8vOtLsj+NqN5JbRvd/gErW7/JlNT0/Hy0tXaIuUJ43U2ZEtW7awa9cumjdvTnp6uvVWJL17967gykRERKS8KdTZmbfeeovdu3fj7OxM06ZNWbNmDQEBARVdloiIiJQzhTo70rhxY5KS9ABwERGRvyI9JkxERETEDijUiYiIiNgBhToRERERO6BQJyIiImIHFOpERERE7IBCnYiIiIgdUKgTERERsQMKdSIiIiJ2QKFORERExA4o1ImIiIjYAYU6ERERETugUCciIiJiBxTqREREROyAQp2IiIiIHVCoExEREbEDBovFYqnoIkRKKiMjA29vb9LT0/Hy8qrockTkGvSZFbl5NFInIiIiYgcU6kRERETsgEKdiIiIiB1QqBMRERGxAwp1IiIiInZAoU5ERETEDijUiYiIiNgBhToRERERO+BY0QWIlMble2VnZGRUcCUiUhKXP6u6z71I+VOok9tKZmYmACEhIRVciYiURmZmJt7e3hVdhohd02PC5LaSn5/PsWPH8PT0xGAwVHQ5InINFouFzMxMqlatitGoM35EypNCnYiIiIgd0J9NIiIiInZAoU5ERETEDijUiYiIiNgBhToRERERO6BQJyIiImIHFOpERERE7IBCnYiIiIgdUKgTERERsQMKdSIiIiJ2QKFORERExA4o1ImIiIjYAceKLkCkNPLz8zl27Bienp4YDIaKLkdErsFisZCZmUnVqlUxGjWOIFKeFOrktnLs2DFCQkIqugwRKaXDhw9TrVq1ii5DxK4p1MltxdPTEyj4BeHl5VXB1YjItWRkZBASEmL97IpI+VGok9vK5SlXLy8vhTqR24hOlxApfzrBQURERMQOKNSJiIiI2AGFOhERERE7oHPqRESkwpnNZnJzcyu6DJFbjpOTEw4ODiVqq1AnIiIVxmKxcPz4cdLS0iq6FJFblo+PD4GBgde84EihTkREKszlQFe5cmXc3d11lazIFSwWC1lZWZw8eRKAoKCgYtsr1ImISIUwm83WQOfv71/R5Yjcktzc3AA4efIklStXLnYqVqFObkuXxk/D2cW1ossQ+ctynTruhvu4fA6du7v7DfclYs8uf0Zyc3OLDXW6+lVERCqUplxFilfSz4hCnYiIiIgdUKgTERERsQM6p05ERG45z41946bta/J1nh/4zjvv8Oabb3L8+HEaNmzIv//9b5o3b15k+//+97+89NJLHDp0iPDwcP71r3/RvXv36y1bpBCN1ImIiJTSZ599xtixY3n55ZfZvHkzDRs2pEuXLtZbT/zZTz/9xKBBg3jooYfYsmULffr0oU+fPvz66683uXKxZwp1IiIipTR16lQefvhh/v73vxMVFcWsWbNwd3dnzpw5V20/ffp0unbtyjPPPENkZCT//Oc/adKkCTNnzrzJlYs9U6gTEREphZycHJKSkujUqZN1mdFopFOnTqxbt+6q26xbt86mPUCXLl2KbC9yPRTqRERESuH06dOYzWaqVKlis7xKlSocP378qtscP368VO1FrodCnYiIiIgdUKgTEREphYCAABwcHDhx4oTN8hMnThAYGHjVbQIDA0vVXuR6KNSJiIiUgrOzM02bNiUhIcG6LD8/n4SEBFq1anXVbVq1amXTHmD58uVFthe5HrpPnYiISCmNHTuWoUOHEhMTQ/PmzZk2bRoXLlzg73//OwAPPPAAwcHBTJo0CYCnnnqK9u3bM2XKFHr06MHChQvZtGkTH3zwQUUehtgZhToREbnlXO8NgW+WgQMHcurUKf7xj39w/PhxGjVqxNKlS60XQ6SkpGA0/jEZ1rp1az755BNefPFFnn/+ecLDw1m0aBH169evqEMQO2SwWCyWii5CpKQyMjLw9vbmxIhX8HJxrehyRP6yXEsYui5/ZtPT0/Hy8rJZd+nSJQ4ePEhYWBiurvo8ixSlpJ8VnVMnIiIiYgcU6kRERETsgEKdiIiIiB1QqBMRERGxAwp1IiIiInZAoU5ERETEDijUichN49i9La5Tx+HYpc1N26chwAeXt+JwfuERcNCPPBGxXzd08+FVq1bRoUMHzp07h4+PTxmVVDpz585l9OjRpKWllUl/sbGxNGrUiGnTppVJf9djwoQJLFq0iOTk5ArtZ9iwYaSlpbFo0aIbqkNuXU4P/A2HRnUBMG/ZSe6C/yuyrcHXC5eXHityfd6yRPKWJRa9Mw83HNo2xZKXR97azdbFjj3aYawfjsHbBA4OkHkB897fyPv+JziXUXR/Ls44drsDY1g1DL5e4OKE5Vwm+cm7yFu5AbJzALCcTiP/1704RNfBoXk05nXJRfcpInIbK9WfrbGxsYwePdr6unXr1qSmpuLt7V3WdZXYwIED2bNnT4XtvzzExcXZPCNw2LBh9OnTp+IKErvk0Ky+NdCVhCXPTP5vx2y/Tpz5Y33G+eL3F1MPg4sz+bsPwYWL1uXGOmEYnJ2wnDqHJS0Tg583ji2icX7knmL7M7i74tguBkNgAJb0TMjOxVjZD8e7WuP0QC+btubNOwtqaNOoxMcrInK7uaGROmdnZwIDA8uqluvi5uaGm5tbhdZQ1kwmEyaTqaLLEDtm8PfBsW8n8g8exeDjWTDSdS2ZF8iZ/h+bRY79OmGs4o8l6yLmpB3Fbu7QOBKA/O37bZbnzPgP5Jmtr53u64FDTD2MVfzB3RWyLl21P0uemdxvVmJet7VgVM7RAecR92IMDcYhsha5bi5wMbtgnzsPYDHnY6xaGUMVfyxXhFEREXtR4pG6YcOGsXr1aqZPn47BYMBgMDB37lwMBoN16nPu3Ln4+PiwePFi6tSpg7u7O/379ycrK4t58+YRGhqKr68vo0aNwmz+44d4dnY2cXFxBAcH4+HhQYsWLVi1alWJ6rq8z8smTJhAo0aNWLBgAaGhoXh7e3PvvfeSmZlZ0kMlPz+fcePG4efnR2BgIBMmTLBZn5KSQu/evTGZTHh5eTFgwABOnDhhXb9161Y6dOiAp6cnXl5eNG3alE2bNtnUu2jRIsLDw3F1daVLly4cPny40DFc/n7evHl8/fXX1vf98nvz7LPPEhERgbu7OzVr1uSll14iNze3xMd5JbPZzNixY/Hx8cHf359x48bx5yfILV26lDvuuMPapmfPnuzf/8cv6I4dOzJy5EibbU6dOoWzs7N15PHdd9+1HneVKlXo37//ddUrN8BowGlIT7BYyP14MVzvkwLdXXFoVvDcSnNiMuQU8/+esxOG4IJnYuYfTrVdl2fGoXUjnJ8agvP4h3GIqVfQ7vjpIgMdUDBNu2qjdZqVPDP5h48DYMnPh/wrjis3D8uJ0wAYa1Yr+TFKhVny5cc37et6/Pjjj/Tq1YuqVatiMBhKdJrKqlWraNKkCS4uLtSuXZu5c+de175FilLiUDd9+nRatWrFww8/TGpqKqmpqYSEhBRql5WVxYwZM1i4cCFLly5l1apV9O3blyVLlrBkyRIWLFjA+++/zxdffGHdZuTIkaxbt46FCxfyyy+/cM8999C1a1f27t17XQe1f/9+Fi1axOLFi1m8eDGrV69m8uTJJd5+3rx5eHh4sH79et544w0mTpzI8uXLgYLA17t3b86ePcvq1atZvnw5Bw4cYODAgdbtBw8eTLVq1di4cSNJSUk899xzODk52bxHr732GvPnzycxMZG0tDTuvffeq9YSFxfHgAED6Nq1q/V9b926NQCenp7MnTuXHTt2MH36dD788EPefvvt63nLmDJlCnPnzmXOnDmsXbuWs2fP8tVXX9m0uXDhAmPHjmXTpk0kJCRgNBrp27cv+fn5AAwfPpxPPvmE7Oxs6zb/+c9/CA4OpmPHjmzatIlRo0YxceJEdu/ezdKlS2nXrt111SvXz/GuNhhrVCX3f8uxnE2/7n4c2jTG4OKMJdf2HLmrMfh5Y/j9IoWr7dPg64WxRlWMlXwByD9ygpxZn5euIJM7DtERBdtv2fVH2Pud5WzB+XmGAN/S9StyFRcuXKBhw4a88847JWp/8OBBevToQYcOHUhOTmb06NEMHz6cZcuWlXOl8ldS4ulXb29vnJ2dcXd3t0657tq1q1C73Nxc3nvvPWrVqgVA//79WbBgASdOnMBkMhEVFUWHDh1YuXIlAwcOJCUlhfj4eFJSUqhatSpQEGSWLl1KfHw8r7/+eqkPKj8/n7lz5+Lp6QnA/fffT0JCAq+99lqJto+Ojubll18GIDw8nJkzZ5KQkEDnzp1JSEhg27ZtHDx40Bpq58+fT7169di4cSPNmjUjJSWFZ555hrp161r7+PN7NHPmTFq0aAEUhMjIyEg2bNhA8+bNbdqaTCbc3NzIzs4uNNX94osvWr8PDQ0lLi6OhQsXMm5cyR60faVp06Yxfvx4+vXrB8CsWbMK/bC5++67bV7PmTOHSpUqsWPHDurXr0+/fv0YOXIkX3/9NQMGDAAKRiaHDRuGwWAgJSUFDw8PevbsiaenJzVq1KBx48bF1pWdnW0TEjMyijlxXq7JUC0QhztbYt60nfzNxU+XFsvBAcc2Bf925qQdkHmh+PauLn98/6ewBZD37Y/kLVmDIcAXx/6dcQivgdOQnuS+91mJRhIN/j44Pdwfg7cn+QeOkPvF94UbXf7/6MpaRK5Tt27d6NatW4nbz5o1i7CwMKZMmQJAZGQka9eu5e2336ZLly7lVab8xZT59f3u7u7WQAdQpUoVQkNDbc4Rq1KlCidPngRg27ZtmM1mIiIirOeSmUwmVq9ebTO1VxqhoaHWQAcQFBRk3V9JREdH27y+cvudO3cSEhJiM0oZFRWFj48PO3cWnIw9duxYhg8fTqdOnZg8eXKh43B0dKRZs2bW13Xr1rXZvqQ+++wz2rRpQ2BgICaTiRdffJGUlJRS9QGQnp5OamqqNWRerjEmJsam3d69exk0aBA1a9bEy8uL0NBQAOs+XV1duf/++5kzZw4Amzdv5tdff2XYsGEAdO7cmRo1alCzZk3uv/9+Pv74Y7KysoqtbdKkSXh7e1u/rjY6LCVnDArA4GDE2DACl0mjcZk0GnwKzqczRhcsw9X5mv04NKuHwcuEJd+CedWGa+/4imCOcxH9WyxYTp3F/GNSwT5qV8cYXuOaXRtqVMX5qSEYK/th/nUfOe9/ftXgiMvvYe5SduF1IuVs3bp1dOrUyWZZly5dWLduXQVVJPaozEPdldOMAAaD4arLLk/ZnT9/HgcHB5KSkkhOTrZ+7dy5k+nTp5dZDZf3dzO2nzBhAtu3b6dHjx6sWLGCqKioQlOZN2rdunUMHjyY7t27s3jxYrZs2cILL7xATs5VfpmVkV69enH27Fk+/PBD1q9fz/r16wFs9jl8+HCWL1/OkSNHiI+Pp2PHjtSoUfCL2dPTk82bN/Ppp58SFBTEP/7xDxo2bFjs7WjGjx9Penq69evKcw/l+hmcnDC4OBd8GQ0FyxwcMLg4AwWvnR4biPOzD+HYo/AUuUP7gj9K8nfux3Ly7DX3ZzmTjsVc8Bky+P1xUYYhwBdjvdqXdwkGMNYN+2NDl98/i94mnJ99COdnH8LY4I+Rb2N0BM4jBmIwuZO3Jonc+C8hN+/qx/z7fi2nz12zXpGydvz4capUqWKzrEqVKmRkZHDx4sUithIpnVJd/ers7GxzgUNZaNy4MWazmZMnT9K2bdsy7bs8REZGcvjwYQ4fPmwdNdqxYwdpaWlERUVZ20VERBAREcGYMWMYNGgQ8fHx9O3bF4C8vDw2bdpknWrdvXs3aWlpREZGXnWfV3vff/rpJ2rUqMELL7xgXfbbb79d1zF5e3sTFBTE+vXrree45eXlkZSURJMmTQA4c+YMu3fv5sMPP7T+O61du7ZQXw0aNCAmJoYPP/yQTz75hJkzZ9qsd3R0pFOnTnTq1ImXX34ZHx8fVqxYYZ32/TMXFxdcXDRdVlbMG3/FvPFXm2UuLz6Kwc+70H3qjAE+GPy8sfx2zKa9sV7tgitToeB+cCWRk4vl6AkM1YMwhgRiPnYKAIO3CeeH+mG5lIPlbBoGkzsGr4JRfcu5DPL3FPw/bTAarfs0XJ4+9TLh9EBvDEYDlrw8jNWDcB41xLrL3C+WYzn6+wVMTo4YqgQAkH/gSMlqFhG5zZQq1IWGhrJ+/XoOHTqEyWQq1ehVUSIiIhg8eDAPPPAAU6ZMoXHjxpw6dYqEhASio6Pp0aPHDe+jLHXq1IkGDRowePBgpk2bRl5eHiNGjKB9+/bExMRw8eJFnnnmGfr3709YWBhHjhxh48aNNuejOTk58eSTTzJjxgwcHR0ZOXIkLVu2LHQ+3WWhoaEsW7aM3bt34+/vj7e3N+Hh4aSkpLBw4UKaNWvGt99+e0OjgU899RSTJ08mPDycunXrMnXqVJsRNF9fX/z9/fnggw8ICgoiJSWF55577qp9DR8+nJEjR+Lh4WENsgCLFy/mwIEDtGvXDl9fX5YsWUJ+fj516tS57rrl5nOM/X2U7rdjWEoRkMxbdmKsHoQxqjbm9duAguBm3rYHY7VADJX8wAD5p8+Rv+c38pb/dPVp1N8ZHIx/jDI6OmKoUdV2vaszl8/GM0bWxOBgJP/YSd3ORCpEYGCgzV0SAE6cOIGXl5fd3ZZLKk6pQl1cXBxDhw4lKiqKixcvEh8fXyZFxMfH8+qrr/L0009z9OhRAgICaNmyJT179iyT/suSwWDg66+/5sknn6Rdu3YYjUa6du3Kv//9bwAcHBw4c+YMDzzwACdOnCAgIIB+/frxyiuvWPtwd3fn2Wef5b777uPo0aO0bduW2bNnF7nPhx9+mFWrVhETE8P58+dZuXIlf/vb3xgzZgwjR44kOzubHj168NJLLxW6/UpJPf3006SmpjJ06FCMRiMPPvggffv2JT294EpFo9HIwoULGTVqFPXr16dOnTrMmDGD2NjYQn0NGjSI0aNHM2jQIFxdXa3LfXx8+PLLL5kwYQKXLl0iPDycTz/9lHr16l1XzVI2sl99v1TLc9759Lr2Y960HceudxRMr3q4wYWLWM6mkxu/6JrbWs5lcGnsG9dcVhSHJgWj6ObELaWuW6QstGrViiVLltgsW758Oa1ataqgisQeGSx/vhmZlKuyfqzZrejQoUPUqlWLjRs3Wqdvy0pGRgbe3t6cGPEKXi6u195AbimO3dvi2KnVtR8pVoYMAT44Pzccy7kMciZ9BGUwwyDgOrVkV9lf/symp6fj5WV7k+tLly5x8OBBwsLCbP4ABK77/nHXo3u/waXe5vz58+zbtw8oOI1o6tSpdOjQAT8/P6pXr8748eM5evQo8+fPBwpuaVK/fn2eeOIJHnzwQVasWMGoUaP49ttvdfWrXFNxn5Ur3dATJUSulJuby5kzZ3jxxRdp2bJlmQc6uf3lLVlD3pI1N3WfltNpZMe9dVP3KTfueoLWzbRp0yY6dOhgfT127FgAhg4dyty5c0lNTbW5G0FYWBjffvstY8aMYfr06VSrVo2PPvpIgU7K1C0f6rp168aaNVf/JfD888/z/PPPl7ivlJQUm4sZ/mzHjh1Ur1691DXeDop77Nh3331XJhepJCYm0qFDByIiImxuLi0iYm9iY2MLPXXnSld7WkRsbCxbtugUACk/t3yo++ijj4q83NvPz69UfVWtWpXk5ORi15e3YcOGWe/bdjMVd9zBwcFlso9r/ZATERGR8nPLh7qyChxQcDuN2rVrl1l/t5O/6nGLiIj8VZT5zYdFRERE5OZTqBMRERGxAwp1IiIiInZAoU5ERETEDijUiYiIiNgBhToRERERO6BQJyIiUsYsFguPPPIIfn5+GAyGYu8VajAYWLRo0U2r7VYVGhrKtGnTKrqM25pCnYiIyHVat24dDg4O9OjRw2b50qVLmTt3LosXLyY1NZX69esX2UdqairdunUr71Kv6tFHH8XBwYH//ve/FbL/K23cuJFHHnmkosu4rd3yNx8WEZG/nqFj37hp+5o3ddx1bzt79myefPJJZs+ezbFjx6xPJtq/fz9BQUG0bt26yG1zcnJwdnYmMDDwuvd/I7Kysli4cCHjxo1jzpw53HPPPRVSx+X3oVKlShWyf3uikToREZHrcP78eT777DMef/xxevToYX3e67Bhw3jyySdJSUnBYDAQGhoKFDxKceTIkYwePZqAgAC6dOkCFJ5+PXLkCIMGDcLPzw8PDw9iYmJYv349UBAWe/fuTZUqVTCZTDRr1owffvjBpq7Q0FBef/11HnzwQTw9PalevToffPBBofr/+9//EhUVxXPPPcePP/7I4cOHbdYPGzaMPn368Prrr1OlShV8fHyYOHEieXl5PPPMM/j5+VGtWjXi4+Nttjt8+DADBgzAx8cHPz8/evfuzaFDhwr1+9prr1G1alXq1KljrfvK6de0tDQeffRRqlSpgqurK/Xr12fx4sUAnDlzhkGDBhEcHIy7uzsNGjTg008/takjNjaWUaNGMW7cOPz8/AgMDGTChAlF/4PaAYU6ERGR6/D5559Tt25d6tSpw5AhQ5gzZw4Wi4Xp06czceJEqlWrRmpqKhs3brRuM2/ePJydnUlMTGTWrFmF+jx//jzt27fn6NGjfPPNN2zdupVx48aRn59vXd+9e3cSEhLYsmULXbt2pVevXqSkpNj0M2XKFGJiYtiyZQsjRozg8ccfZ/fu3TZtZs+ezZAhQ/D29qZbt27WUHqlFStWcOzYMX788UemTp3Kyy+/TM+ePfH19WX9+vU89thjPProoxw5cgSA3NxcunTpgqenJ2vWrCExMRGTyUTXrl3Jycmx9puQkMDu3btZvny5NahdKT8/n27dupGYmMh//vMfduzYweTJk3FwcADg0qVLNG3alG+//ZZff/2VRx55hPvvv58NGzbY9DNv3jw8PDxYv349b7zxBhMnTmT58uXF/bPe1jT9KiIich0uhyKArl27kp6ezurVq4mNjcXT0xMHB4dCU6vh4eG88UbRU8uffPIJp06dYuPGjfj5+QG2z+5u2LAhDRs2tL7+5z//yVdffcU333zDyJEjrcu7d+/OiBEjAHj22Wd5++23WblypXVUbO/evfz88898+eWXAAwZMoSxY8fy4osvYjAYrP34+fkxY8YMjEYjderU4Y033iArK4vnn38egPHjxzN58mTWrl3Lvffey2effUZ+fj4fffSRtZ/4+Hh8fHxYtWoVd911FwAeHh589NFHODs7X/V9+OGHH9iwYQM7d+4kIiICgJo1a1rXBwcHExcXZ3395JNPsmzZMj7//HOaN29uXR4dHc3LL79sfe9nzpxJQkICnTt3LvLf4HamkToREZFS2r17Nxs2bGDQoEEAODo6MnDgQGbPnl3sdk2bNi12fXJyMo0bN7YGuj87f/48cXFxREZG4uPjg8lkYufOnYVG6qKjo63fGwwGAgMDOXnypHXZnDlz6NKlCwEBAUBBCExPT2fFihU2/dSrVw+j8Y+oUKVKFRo0aGB97eDggL+/v7XvrVu3sm/fPjw9PTGZTJhMJvz8/Lh06RL79++3btegQYMiA93l96FatWrWQPdnZrOZf/7znzRo0AA/Pz9MJhPLli0r9n0ACAoKsnkf7I1G6kREREpp9uzZ5OXlWS+MgILbmLi4uDBz5swit/Pw8Ci2Xzc3t2LXx8XFsXz5ct566y1q166Nm5sb/fv3t5naBHBycrJ5bTAYrFO4ZrOZefPmcfz4cRwd/4gBZrOZOXPmcOeddxbbT3F9nz9/nqZNm/Lxxx8Xqv3KCyFu9H148803mT59OtOmTaNBgwZ4eHgwevToUr0P9kihTkREpBTy8vKYP38+U6ZMsU4nXtanT59CJ+yXRnR0NB999BFnz5696mhdYmIiw4YNo2/fvkBBiLryIoSSWLJkCZmZmWzZssV6jhrAr7/+yt///nfS0tLw8fG5rvqbNGnCZ599RuXKlfHy8rquPqDgfThy5Ah79uy56mhdYmIivXv3tk5/5+fns2fPHqKioq57n/ZA068iIiKlsHjxYs6dO8dDDz1E/fr1bb7uvvvua07BFmfQoEEEBgbSp08fEhMTOXDgAP/73/9Yt24dUHBe2JdffklycjJbt27lvvvuK/XI0+zZs+nRowcNGza0qf3yFatXG2UrqcGDBxMQEEDv3r1Zs2YNBw8eZNWqVYwaNcp6MUVJtG/fnnbt2nH33XezfPlyDh48yHfffcfSpUuBgvdh+fLl/PTTT+zcuZNHH32UEydOXHfd9kKhTkREpBRmz55Np06d8Pb2LrTu7rvvZtOmTWRkZFxX387Oznz//fdUrlyZ7t2706BBA5urPqdOnYqvry+tW7emV69edOnShSZNmpS4/xMnTvDtt99y9913F1pnNBrp27fvDYVSd3d3fvzxR6pXr06/fv2IjIzkoYce4tKlS6Ueufvf//5Hs2bNGDRoEFFRUYwbNw6z2QzAiy++SJMmTejSpQuxsbHWIPxXZ7BYLJaKLkKkpDIyMvD29ubEiFfwcnGt6HJE/rJcS3jD3suf2fT09EK/1C9dusTBgwcJCwvD1VWfZ5GilPSzonPq5LbkOmk0rjdwvoaIiIi90fSriIiIiB1QqBMRERGxAwp1IiIiInZAoU5ERETEDijUiYiIiNgBhToRERERO6BQJyIiImIHFOpERERE7IBCnYiIyE106NAhDAYDycnJFV2K2BmFOhERkeu0bt06HBwc6NGjR4m3CQkJITU1lfr165djZfJXpMeEiYjILafynJI/pP5GnXxw83VvO3v2bJ588klmz57NsWPHqFq16jW3cXBwIDAw8Lr3KVIUjdSJiIhch/Pnz/PZZ5/x+OOP06NHD+bOnWtdd+7cOQYPHkylSpVwc3MjPDyc+Ph4oPD0q9ls5qGHHiIsLAw3Nzfq1KnD9OnTbfY1bNgw+vTpw1tvvUVQUBD+/v488cQT5Obm3qzDlduARupERESuw+eff07dunWpU6cOQ4YMYfTo0YwfPx6DwcBLL73Ejh07+O677wgICGDfvn1cvHjxqv3k5+dTrVo1/vvf/+Lv789PP/3EI488QlBQEAMGDLC2W7lyJUFBQaxcuZJ9+/YxcOBAGjVqxMMPP3yzDllucQp1IiIi12H27NkMGTIEgK5du5Kens7q1auJjY0lJSWFxo0bExMTA0BoaGiR/Tg5OfHKK69YX4eFhbFu3To+//xzm1Dn6+vLzJkzcXBwoG7duvTo0YOEhASFOrHS9KuIiEgp7d69mw0bNjBo0CAAHB0dGThwILNnzwbg8ccfZ+HChTRq1Ihx48bx008/FdvfO++8Q9OmTalUqRImk4kPPviAlJQUmzb16tXDwcHB+jooKIiTJ0+W8ZHJ7UyhTkREpJRmz55NXl4eVatWxdHREUdHR9577z3+97//kZ6eTrdu3fjtt98YM2YMx44d48477yQuLu6qfS1cuJC4uDgeeughvv/+e5KTk/n73/9OTk6OTTsnJyeb1waDgfz8/HI7Rrn9aPpVbkuPjp+Gs4trRZdRbr6rv5BX/btVdBnyF+WUXXB7jne2NiyyzabXPG5WObecvLw85s+fz5QpU7jrrrts1vXp04dPP/2Uxx57jEqVKjF06FCGDh1K27ZteeaZZ3jrrbcK9ZeYmEjr1q0ZMWKEddn+/fvL/TjE/ijUiYiIlMLixYs5d+4cDz30EN7e3jbr7r77buvtTZo2bUq9evXIzs5m8eLFREZGXrW/8PBw5s+fz7JlywgLC2PBggVs3LiRsLCwm3E4Ykc0/SoiIlIKs2fPplOnToUCHRSEuk2bNuHo6Mj48eOJjo6mXbt2ODg4sHDhwqv29+ijj9KvXz8GDhxIixYtOHPmjM2onUhJGSwWi6WiixApqYyMDLy9vbl3xCuafhUpJ2U5/Xr5M5ueno6Xl5fNukuXLnHw4EHCwsJwdbXfz7PIjSrpZ0UjdSIiIiJ2QKFORERExA4o1ImIiIjYAYU6ERERETugUCciIiJiBxTqREREROyAQp2IiIiIHVCoExEREbEDCnUiIiIidkChTkRE5C9o1apVGAwG0tLSKroUKSOOFV2AiIjIn33w9Qs3bV+P9H7turY7fvw4r732Gt9++y1Hjx6lcuXKNGrUiNGjR3PnnXeWcZVlr3Xr1qSmpl71GbZye1KoExERKaVDhw7Rpk0bfHx8ePPNN2nQoAG5ubksW7aMJ554gl27dlV0idfk7OxMYGBgRZchZUjTryIiIqU0YsQIDAYDGzZs4O677yYiIoJ69eoxduxYfv75ZwCmTp1KgwYN8PDwICQkhBEjRnD+/HlrHxMmTKBRo0Y2/U6bNo3Q0FCbZXPmzKFevXq4uLgQFBTEyJEjreuutY/ffvuNXr164evri4eHB/Xq1WPJkiVA4enXM2fOMGjQIIKDg3F3d6dBgwZ8+umnNrXExsYyatQoxo0bh5+fH4GBgUyYMOEG300pKwp1IiIipXD27FmWLl3KE088gYeHR6H1Pj4+ABiNRmbMmMH27duZN28eK1asYNy4caXa13vvvccTTzzBI488wrZt2/jmm2+oXbu2df219vHEE0+QnZ3Njz/+yLZt2/jXv/6FyWS66r4uXbpE06ZN+fbbb/n111955JFHuP/++9mwYYNNu3nz5uHh4cH69et54403mDhxIsuXLy/VcUn50PSriIhIKezbtw+LxULdunWLbTd69Gjr96Ghobz66qs89thjvPvuuyXe16uvvsrTTz/NU089ZV3WrFmzEu8jJSWFu+++mwYNGgBQs2bNIvcVHBxMXFyc9fWTTz7JsmXL+Pzzz2nevLl1eXR0NC+//DIA4eHhzJw5k4SEBDp37lzi45LyoVAndq9r+2Y0qleLoMp+eLi7kp5xgV37D7NoWSKnzqYD8NyIe4msXb3QtnsOHOG1mZ+UaD+uLs788+mhVA7wBWDuf79n5brkMjuOvxKTmw/33fVMkeuTdiWQtHsFTo7OxNTtRFjV+ri5eHDhYjp7DiezZc8qLJb8Yvfh5uJB88guVA+sg7OjKxlZZ9lxcD3bD/5c1ocjdsZisZSo3Q8//MCkSZPYtWsXGRkZ5OXlcenSJbKysnB3d7/m9idPnuTYsWPFXnRxrX2MGjWKxx9/nO+//55OnTpx9913Ex0dfdW+zGYzr7/+Op9//jlHjx4lJyeH7OzsQrX+efugoCBOnjxZgndEypumX8XudWrbhDo1Q8i6mM259PME+HlzR7P6vDhqMK4uzjZtT54+x77fjlm/jp44XeL93N+vkzXQyY0x55s5cTbF5ist85R1fdalTMBAlxb306BWG9xcPMi8cA6Tuy8xde+kfeN+xfbv6OBErzYPU6dGU5wcnMm8mIavZ2XaRPeiad1b/6pFqVjh4eEYDIZiL4Y4dOgQPXv2JDo6mv/9738kJSXxzjvvAJCTkwMUTJ3+OSDm5uZav3dzcyu2jpLsY/jw4Rw4cID777+fbdu2ERMTw7///e+r9vfmm28yffp0nn32WVauXElycjJdunSx9nWZk5OTzWuDwUB+fvF/RMnNoZE6KZGcnBycnZ2v3fAWtPrnrSRu2s7ZtEwA7uvdkS7tY/DxMlEvogZJ2/Za2369fB1rN/5a6n00b1iHO5rVZ/2WXbRoXPyUjFzbxexMvl7zvs2yNg164eNZiUs5Wew9spXQoEiqBhRMJS3f8AkpJ3ZTL6wlbaJ7ERHSmG37f+JM+rGr9h8Z2hwfz0pYLPksWjOLsxknaFmvG9G176BReDt2HPyZi9kXyv045fbk5+dHly5deOeddxg1alSh8+rS0tJISkoiPz+fKVOmYDQWjJ98/vnnNu0qVarE8ePHsVgsGAwGAJKTk63rPT09CQ0NJSEhgQ4dOhSqoyT7AAgJCeGxxx7jscceY/z48Xz44Yc8+eSThdolJibSu3dvhgwZAkB+fj579uwhKiqqFO+OVCSN1MlVxcbGMnLkSEaPHk1AQABdunS55lVWUPBDITY2Fnd3d3x9fenSpQvnzp0DCn5ATJo0ibCwMNzc3GjYsCFffPFFuR/L//3wszXQAew+cNj6fW6e2abtfb078tEbY3nzhUf4+z1d8DJde4rEz8eTYfd04eDh4/zvuzVlV7hYuTi5EVG9CQA7D20gz5xDSOUIAPLyckg5sQeAg6nbrduEVA4vsr/L26ZfOMPZjBMF2x4r2NbB6EjVgFplfxBiV9555x3MZjPNmzfnf//7H3v37mXnzp3MmDGDVq1aUbt2bXJzc/n3v//NgQMHWLBgAbNmzbLpIzY2llOnTvHGG2+wf/9+3nnnHb777jubNhMmTGDKlCnMmDGDvXv3snnzZutIW0n2MXr0aJYtW8bBgwfZvHkzK1euJDIy8qrHFB4ezvLly/npp5/YuXMnjz76KCdOnCjDd03Km0KdFGnevHk4OzuTmJjIrFmzrnmVVXJyMnfeeSdRUVGsW7eOtWvX0qtXL8zmguA0adIk5s+fz6xZs9i+fTtjxoxhyJAhrF69+qYdk8FgILZVQ6BgqnXHnt+s67JzcjmXnknm+YtU9vchtlVDXnpqCM7OTkV1h8EAj9zXAwcHI+/95/+sxyplKyqsJU6OzuSZc/n1wDoATG4FN0y9lJsFFExhXbz0xx8ZJjefIvu7vO2Vo3FZ2Vds6170tiJQcMHB5s2b6dChA08//TT169enc+fOJCQk8N5779GwYUOmTp3Kv/71L+rXr8/HH3/MpEmTbPqIjIzk3Xff5Z133qFhw4Zs2LDB5kIFgKFDhzJt2jTeffdd6tWrR8+ePdm7t2B2oST7MJvNPPHEE0RGRtK1a1ciIiKKvFDjxRdfpEmTJnTp0oXY2FgCAwPp06dP2b1pUu4MlpKe8Sl/KbGxsWRkZLB58+Yi23zxxRc89thjnD5dcN7ZfffdR0pKCmvXri3UNjs7Gz8/P3744QdatWplXT58+HCysrL45JOrX4yQnZ1Ndna29XVGRgYhISHcO+IVnF1cS3VMzs5OjBjSi8b1a5OWcZ5/vfcZx06cAaB6cGWOHT9D3u+hrH/3tvTqVFDn+x8v5qekHVfts0u7GO7r05HZn33Hj+u3EeDrxZSXHgNu7EKJ7+ov5FX/bte1rb0xGh24r/MzuLt6suu3TfyY/BUA3VoOJaRKBOcvpvHJ928CYDAYefhv/wRgx8ENrP3l66v2OaDjaHw8K5F65hD/t/ZDALw8/Li309MArN+xjK17fyzvQ7tlOWX3AOCdrQ2LbLPptcK38riajIwMvL29SU9Px8vLy2bdpUuXOHjwIGFhYbi6lu7zLPJXUtLPis6pkyI1bdrU5vW1rrJKTk7mnnvuuWpf+/btIysrq9Al7zk5OTRu3LjIGiZNmsQrr7xyw8fi7enBmOF3ExYSSOrJs0z54L/WK18BUo7aXrm1bvNOa6jz87H9RXSlkKqVABjc504G97kTwxXr7uvTkTYx9Xj13x/fcP1/ZREhjXF39cRiyeeXfX/8wXD+YsG/n6uzB2AALLi5eFyxPq3IPs9fTMfHs5JNezfnP+7ddT6r6G1FRG5VCnVSpCtP/r18ldXjjz/Oa6+9hp+fH2vXruWhhx4iJycHd3f3Yq/Uunzu3bfffktwcLDNOhcXlyK3Gz9+PGPHjrW+vjxSVxrBVfwZ83B/Kvl5s3v/YabHf8WFrEvW9Z4md1o3jWL1z79wKbvgKq8Wjf642OH0uYLw4OttYtxjAwH4YsmPNhdY/PkqWgBnJ0ecnfURu1HRte4AIOXEHtLO/3EF7JGTe4kMbYajgxPVq0SQcmI3YUH1rOsPnyz49wkNiqJ55F0ALP5pDlmXMjh8cg/VKtfG28MfP68qnM04QVjVgm3N+XkcO73/Zh2eiEiZ0W8cKZGSXGUVHR1NQkLCVUfWoqKicHFxISUlhfbt25d4vy4uLsWGvpJ48u99qeRXcA6Vq4szYx/ub13348+/sH3PIe7r3ZEBPdtz8nQaLs5O+PsWjM4dPX6apF8KTsJ3MBqpWsUfADfXgpo+WvgdHy3848Tmspp+lQI1Auvi41kwGrp1n+1FKIdSd5B65hBB/qF0bn4fGRfO4m0KAGDv4WTrla/Oji7WPoyGgv93dx7aWHAFrCmAPm0f4/yldHxMBW1+2bdWV76KyG1JoU5K5MqrrHr16mW9eOJK48ePp0GDBowYMYLHHnsMZ2dnVq5cyT333ENAQABxcXGMGTOG/Px87rjjDtLT00lMTMTLy4uhQ4eWW+2Ojg7W72tUq2Kzbtuug2RcuMg3y3+ifp0wKvv74OTkyLETZ9j8616+XbG+0BWycvNcHqU7ee4wx88csllnwcLSn+cTU7cTNavWx8vDj/NZaew9nMzmPSuL7TfPnMP/rf2I5lF3Ub1KHTzdfTmXeZKdhzZYL8QQEbnd6EIJuarY2FgaNWrEtGnTrMvefvtt3nzzTdLS0mjXrh2DBw/mgQce4Ny5c9ZnHa5evZrnn3+epKQk3NzcaNGiBQsXLsTHxweLxcKMGTN47733OHDgAD4+PjRp0oTnn3+edu3alaiuyyddX8+FErcTXSghFUkXSojcWkr6WVGok9uKQp1I+VOoE7m1lPSzovvUiYiIiNgBhToRERERO6BQJyIiImIHFOpERETKwdy5c60Xkd0qDh06hMFgIDk5uaJLkXKgW5qIiMgtJ/7zn27avv4+oHWptxk2bBhpaWksWrTIZvmqVavo0KED586dY+DAgXTv3r2MqiwbISEhpKamEhAQUNGlSDlQqBMRESkHbm5uxT5ppyI4ODgQGBhY0WVIOdH0q4iISDn48/Tr1q1b6dChA56ennh5edG0aVM2bdpk03bRokWEh4fj6upKly5dOHz4sHX7/fv307t3b6pUqYLJZKJZs2b88MMPNvsMDQ3l9ddf58EHH8TT05Pq1avzwQcfWNdfbfp1+/bt9OzZEy8vLzw9PWnbti379+tRebcjhToREZGbYPDgwVSrVo2NGzeSlJTEc889h5OTk3V9VlYWr732GvPnzycxMZG0tDTuvfde6/rz58/TvXt3EhIS2LJlC127dqVXr16kpKTY7GfKlCnExMSwZcsWRowYweOPP87u3buvWtPRo0dp164dLi4urFixgqSkJB588EHy8vLK502QcqXpVxERkeuwePFiTCaTzTKzuejHCqakpPDMM89Qt25dAMLDw23W5+bmMnPmTFq0aAHAvHnziIyMZMOGDTRv3pyGDRvSsOEfN4T+5z//yVdffcU333zDyJEjrcu7d+/OiBEjAHj22Wd5++23WblyJXXq1ClU0zvvvIO3tzcLFy60BsyIiIjSvA1yC9FInYiIyHXo0KEDycnJNl8fffRRke3Hjh3L8OHD6dSpE5MnTy40xeno6EizZs2sr+vWrYuPjw87d+4ECkbq4uLiiIyMxMfHB5PJxM6dOwuN1EVHR1u/NxgMBAYGcvLkyavWlJycTNu2bW1GDOX2pVAnIiJyHTw8PKhdu7bNV3BwcJHtJ0yYwPbt2+nRowcrVqwgKiqKr776qsT7i4uL46uvvuL1119nzZo1JCcn06BBA3Jycmza/TmgGQwG8vPzr9rnrXYhh9wYhToREZGbJCIigjFjxvD999/Tr18/4uPjrevy8vKsF04A7N69m7S0NCIjIwFITExk2LBh9O3blwYNGhAYGMihQ4duqJ7o6GjWrFlDbm7uDfUjtwaFOhERkXJ28eJFRo4cyapVq/jtt99ITExk48aN1sAGBSNsTz75JOvXrycpKYlhw4bRsmVLmjdvDhScg/fll1+SnJzM1q1bue+++4ocgSupkSNHkpGRwb333sumTZvYu3cvCxYsKPLCCrm1KdSJiIiUMwcHB86cOcMDDzxAREQEAwYMoFu3brzyyivWNu7u7jz77LPcd999tGnTBpPJxGeffWZdP3XqVHx9fWndujW9evWiS5cuNGnS5Ibq8vf3Z8WKFZw/f5727dvTtGlTPvzwQ51jd5syWCwWS0UXIVJSGRkZeHt7c++IV3B2ca3ocsrNd/UX8qp/t4ouQ/6inLJ7APDO1oZFttn0mkeJ+rr8mU1PT8fLy8tm3aVLlzh48CBhYWG4utrv57kk5s6dy+jRo0lLS6voUuQWVNLPikbqREREROyAQp2IiIiIHVCoExERqWDDhg3T1KvcMD1RQm5L708aXej8HPsyrqILEOHvAyq6AhEpDY3UiYiIiNgBhToRERERO6BQJyIiImIHFOpERERE7IBCnYiIiIgdUKgTERG5iebOnYuPj09FlyF2SLc0ERGRW07MCxdu2r5K+sizPzt8+DAvv/wyS5cu5fTp0wQFBdGnTx/+8Y9/4O/vD0BoaCijR49m9OjRZVixyNVppE5ERKSUDhw4QExMDHv37uXTTz9l3759zJo1i4SEBFq1asXZs2dvek25ubk3fZ9ya1GoExERKaUnnngCZ2dnvv/+e9q3b0/16tXp1q0bP/zwA0ePHuWFF14gNjaW3377jTFjxmAwGDAYDDZ9LFu2jMjISEwmE127diU1NdVm/UcffURkZCSurq7UrVuXd99917ru0KFDGAwGPvvsM9q3b4+rqysff/zxTTl2uXVp+lVERKQUzp49y7Jly3jttddwc3OzWRcYGMjgwYP57LPP2Lt3L40aNeKRRx7h4YcftmmXlZXFW2+9xYIFCzAajQwZMoS4uDhrMPv444/5xz/+wcyZM2ncuDFbtmzh4YcfxsPDg6FDh1r7ee6555gyZQqNGzfG1dW1/A9ebmkKdSIiIqWwd+9eLBYLkZGRV10fGRnJuXPnMJvNODg44OnpSWBgoE2b3NxcZs2aRa1atQAYOXIkEydOtK5/+eWXmTJlCv369QMgLCyMHTt28P7779uEutGjR1vbiCjUiYiIXAeLxXLd27q7u1sDHUBQUBAnT54E4MKFC+zfv5+HHnrIZoQvLy8Pb29vm35iYmKuuwaxPwp1IiIipVC7dm0MBgM7d+6kb9++hdbv3LkTX19fKlWqVGQfTk5ONq8NBoM1JJ4/fx6ADz/8kBYtWti0c3BwsHnt4XF9V+6KfdKFEiIiIqXg7+9P586deffdd7l48aLNuuPHj/Pxxx8zcOBADAYDzs7OmM3mUvVfpUoVqlatyoEDB6hdu7bNV1hYWFkeitgZhToREZFSmjlzJtnZ2XTp0oUff/yRw4cPs3TpUjp37kxwcDCvvfYaUHCfuh9//JGjR49y+vTpEvf/yiuvMGnSJGbMmMGePXvYtm0b8fHxTJ06tbwOSeyAQp2IiEgphYeHs2nTJmrWrMmAAQOoVasWjzzyCB06dGDdunX4+fkBMHHiRA4dOkStWrWKnY79s+HDh/PRRx8RHx9PgwYNaN++PXPnztVInRTLYLmRMz1FbrKMjAy8vb1JT0/Hy8urossRkWso7jN76dIlDh48SFhYmG7HIVKMkn5WNFInIiIiYgcU6kRERETsgEKdiIiIiB3QferktvTy+Gm4uOgcHHvV7o7gii5BrqF7v8EVXYKI/IlG6kRERETsgEKdiIiIiB1QqBMRERGxAwp1IiIiInZAoU5ERETEDijUiYiI3KBVq1ZhMBhIS0ursBrmzp2Lj49PmfUXGxvL6NGjy6y/6zFhwgQaNWpU4f0MGzaMPn363HAd5U23NBERkVvOpbFv3LR9uU4dV+ptYmNjadSoEdOmTQOgdevWpKam4u3tXcbVldzAgQPp3r17he2/PMTFxfHkk09aXw8bNoy0tDQWLVpUcUXdwhTqREREbpCzszOBgYEVWoObmxtubm4VWkNZM5lMmEymii7jtqHpVxERkVIYNmwYq1evZvr06RgMBgwGA3PnzrWZfr08Fbp48WLq1KmDu7s7/fv3Jysri3nz5hEaGoqvry+jRo3CbDZb+87OziYuLo7g4GA8PDxo0aIFq1atKlFdf55+vTzluGDBAkJDQ/H29ubee+8lMzOzxMean5/PuHHj8PPzIzAwkAkTJtisT0lJoXfv3phMJry8vBgwYAAnTpywrt+6dSsdOnTA09MTLy8vmjZtyqZNm2zqXbRoEeHh4bi6utKlSxcOHz5c6Bgufz9v3jy+/vpr6/t++b159tlniYiIwN3dnZo1a/LSSy+Rm5tb4uO8ktlsZuzYsfj4+ODv78+4ceOwWCw2bZYuXcodd9xhbdOzZ0/2799vXd+xY0dGjhxps82pU6dwdnYmISEBgHfffdd63FWqVKF///7XVe+VFOpERERKYfr06bRq1YqHH36Y1NRUUlNTCQkJKdQuKyuLGTNmsHDhQpYuXcqqVavo27cvS5YsYcmSJSxYsID333+fL774wrrNyJEjWbduHQsXLuSXX37hnnvuoWvXruzdu/e6at2/fz+LFi1i8eLFLF68mNWrVzN58uQSbz9v3jw8PDxYv349b7zxBhMnTmT58uVAQeDr3bs3Z8+eZfXq1SxfvpwDBw4wcOBA6/aDBw+mWrVqbNy4kaSkJJ577jmcnJxs3qPXXnuN+fPnk5iYSFpaGvfee+9Va4mLi2PAgAF07drV+r63bt0aAE9PT+bOncuOHTuYPn06H374IW+//fb1vGVMmTKFuXPnMmfOHNauXcvZs2f56quvbNpcuHCBsWPHsmnTJhISEjAajfTt25f8/HwAhg8fzieffEJ2drZ1m//85z8EBwfTsWNHNm3axKhRo5g4cSK7d+9m6dKltGvX7rrqvZKmX0VERErB29sbZ2dn3N3drVOuu3btKtQuNzeX9957j1q1agHQv39/FixYwIkTJzCZTERFRdGhQwdWrlzJwIEDSUlJIT4+npSUFKpWrQoUBJmlS5cSHx/P66+/Xupa8/PzmTt3Lp6engDcf//9JCQk8Nprr5Vo++joaF5++WUAwsPDmTlzJgkJCXTu3JmEhAS2bdvGwYMHraF2/vz51KtXj40bN9KsWTNSUlJ45plnqFu3rrWPP79HM2fOpEWLFkBBiIyMjGTDhg00b97cpq3JZMLNzY3s7OxCU90vvvii9fvQ0FDi4uJYuHAh48aV/nzJadOmMX78ePr16wfArFmzWLZsmU2bu+++2+b1nDlzqFSpEjt27KB+/fr069ePkSNH8vXXXzNgwACgYGRy2LBhGAwGUlJS8PDwoGfPnnh6elKjRg0aN25c6lr/TCN1IiIi5cDd3d0a6ACqVKlCaGiozTliVapU4eTJkwBs27YNs9lMRESE9Vwyk8nE6tWrbab2SiM0NNQa6ACCgoKs+yuJ6Ohom9dXbr9z505CQkJsRimjoqLw8fFh586dAIwdO5bhw4fTqVMnJk+eXOg4HB0dadasmfV13bp1bbYvqc8++4w2bdoQGBiIyWTixRdfJCUlpVR9AKSnp5OammoNmZdrjImJsWm3d+9eBg0aRM2aNfHy8iI0NBTAuk9XV1fuv/9+5syZA8DmzZv59ddfGTZsGACdO3emRo0a1KxZk/vvv5+PP/6YrKysUtf7Zwp1IiIi5eDKaUYAg8Fw1WWXp+zOnz+Pg4MDSUlJJCcnW7927tzJ9OnTy6yGy/u7GdtPmDCB7du306NHD1asWEFUVFShqcwbtW7dOgYPHkz37t1ZvHgxW7Zs4YUXXiAnJ6dM93OlXr16cfbsWT788EPWr1/P+vXrAWz2OXz4cJYvX86RI0eIj4+nY8eO1KhRAyiYLt68eTOffvopQUFB/OMf/6Bhw4Y3fEschToREZFScnZ2trnAoSw0btwYs9nMyZMnqV27ts1XRV9ZezWRkZEcPnzY5sKGHTt2kJaWRlRUlHVZREQEY8aM4fvvv6dfv37Ex8db1+Xl5VkvnADYvXs3aWlpREZGXnWfV3vff/rpJ2rUqMELL7xATEwM4eHh/Pbbb9d1TN7e3gQFBVlD2uUak5KSrK/PnDnD7t27efHFF7nzzjuJjIzk3Llzhfpq0KABMTExfPjhh3zyySc8+OCDNusdHR3p1KkTb7zxBr/88guHDh1ixYoV11W3tc8b2lpEROQvKDQ0lPXr13Po0CFMJlOpRq+KEhERweDBg3nggQeYMmUKjRs35tSpUyQkJBAdHU2PHj3KoPKy06lTJxo0aMDgwYOZNm0aeXl5jBgxgvbt2xMTE8PFixd55pln6N+/P2FhYRw5coSNGzfanI/m5OTEk08+yYwZM3B0dGTkyJG0bNmy0Pl0l4WGhrJs2TJ2796Nv78/3t7ehIeHk5KSwsKFC2nWrBnffvvtDY0GPvXUU0yePJnw8HDq1q3L1KlTbUbQfH198ff354MPPiAoKIiUlBSee+65q/Y1fPhwRo4ciYeHB3379rUuX7x4MQcOHKBdu3b4+vqyZMkS8vPzqVOnznXXDRqpExERKbW4uDgcHByIioqiUqVK13X+1tXEx8fzwAMP8PTTT1OnTh369OnDxo0bqV69epn0X5YMBgNff/01vr6+tGvXjk6dOlGzZk0+++wzABwcHDhz5gwPPPAAERERDBgwgG7duvHKK69Y+3B3d+fZZ5/lvvvuo02bNphMJuv2V/Pwww9Tp04dYmJiqFSpEomJifztb39jzJgxjBw5kkaNGvHTTz/x0ksvXfdxPf3009x///0MHTqUVq1a4enpaRPIjEYjCxcuJCkpifr16zNmzBjefPPNq/Y1aNAgHB0dGTRoEK6urtblPj4+fPnll3Ts2JHIyEhmzZrFp59+Sr169a67bgCD5c83XxG5hWVkZODt7c3oEa/g4uJ67Q3kttTujuCKLkGuoXu/wSVqd/kzm56ejpeXl826S5cucfDgQcLCwmx+4clfw9y5cxk9enSFPlqtvB06dIhatWqxceNGmjRpct39lPSzoulXERERkTKUm5vLmTNnePHFF2nZsuUNBbrSUKgTKWedurShU5c2V133fNyb5Odb6D+wK6E1Q/D08sAAZGZeYNfOA/ywLJGLWZeK7b9qcBXuvKs1IdWDcPdw5eLFbI4dOcHKH37m0MEj5XBE0rj5HQRVK7iK7djhQyRvTLSu8/TyoXZkA/wCKuPk5EROdjbnzpxiy4a1Rfbn6uZG7ToN8PWvhKubGwajkYsXLnAk5QCH9u0qdDd7+Wvq1q0ba9asueq6559/nueff77EfaWkpNhczPBnO3bsuCWnfMtCcY8d++6772jbtu0N7yMxMZEOHToQERFhc3Pp8qZQJzdMD1gumfPnszh7Js1m2eXf1ZH1w7l0KZtTJ8/i4eGGf4Avbdo2JaCSL/EfFP0DwdXVhYcfH4ibuyvZl3I4cfwMlSr5UieyJjVrV2fyxPe4cOFiOR7VX0+1GjWtge7PfP0r0bxNRxwcHcnNzSEzIx1HR0cqB1Urtk93D0+q1wwnLzeXCxcycfcw4entQ2SDJrh7mNievLE8DkVuMx999BEXL1798+zn51eqvqpWrUpycnKx68vbsGHDrPdtu5mKO+7g4LI59SM2NrZC/hhTqBO5SXbv2M9/F3531XWTXnmXvLw/LtN/dOQgwmqGEBpafBioEhSAm3vB+RX/+3wpvyTvommz+twzqDtOTo6YPD0U6sqQu4eJqOgYzp05haubO27uHjbrGzRugYOjI0dTDrJt83ry8wv+TR0ci/9Rm5uTwy+bf+ZYykHy8/NxdHLmjo5dcffwpGpIqEKdAGUXOKDgdhq1a9cus/5uJ/Z83Ap1UuEsFgtmsxnHa/ziu93Vj65DdOO6XLyYzdHDJ1i+dA3HjhbcmT0vz0znrncQXicUT08PfP28Aa45fXry+Gmysi7i7u7G3QO60q5DcypX9iMnJ5e1qzdx4vjpcj+uvwqDwUDDZm2wYCF5YyIt2nayWe/p7YPJy9v6uv1dvXB0ciL93Fl2/bqFjLSzRfadmZFGZkaa9XXe76N87h6e5Jtv/FYZIvLXoFua2JHMzEwGDx6Mh4cHQUFBvP3228TGxjJ69GgAsrOziYuLIzg4GA8PD1q0aMGqVaus28+dOxcfHx+WLVtGZGQkJpPJ+uDky8xmM2PHjsXHxwd/f3/GjRtXaIg5Pz+fSZMmERYWhpubGw0bNrQ5p2DVqlUYDAa+++47mjZtiouLC2vXFn2+kT0wm/PJzLzAubMZeHmZiKxXi8dHDaFqcGVrm4BKvlSvUdUa6PbuPsTH878ptt+LF7OZ9e9POHP6HC6uzlQLCcTZxZkL57NIPVbyRwHJtdWObICvXwDbkzdyMetCofUepj+u7AyuHobZnAdAQOVAWrbtVGhUrzgeJk/8K1UB4PChfTdY+a1P5wyKFK+knxGFOjsyduxYEhMT+eabb1i+fDlr1qxh8+bN1vUjR45k3bp1LFy4kF9++YV77rmHrl27snfvXmubrKws3nrrLRYsWMCPP/5ISkoKcXFx1vVTpkxh7ty5zJkzh7Vr13L27NlCN3mcNGkS8+fPZ9asWWzfvp0xY8YwZMgQVq9ebdPuueeeY/LkyezcubPQ8wUvy87OJiMjw+brdpO8eQevvjyTtyZ9yNR/zWb2+58D4OTkSKs2f1wR9emC/+P5uLeY/tZcjqeeIrxOKH36dSqq24I+nJ24Z1B3/AN8+fbrlbz03Nss/noFvn7eDLr/bzahUa6ft48ftSLqcTTlIMcOH7pqG6Phjx+nhw/t48fli1mbsOT36VQnqtWoWbJ9+frRsl1nHB2dOH40hb07fymLQ7glXX4EVVk881LEnl3+jPz5sW1/Zt/zXX8hmZmZzJs3j08++YQ777wTKLiJ5eWTXVNSUoiPjyclJcW6LC4ujqVLlxIfH8/rr78OFFyGPWvWLOtDqEeOHMnEiROt+5k2bRrjx4+nX79+AMyaNYtly5ZZ12dnZ/P666/zww8/0KpVKwBq1qzJ2rVref/992nfvr217cSJE+ncuXOxxzVp0iSbG1Xejk6fsn18zN7dh7hwPgsPkzvevp426/Lz80k9dpINP2/lb3070aRZfVb8sK5QH5c1ahJJSPUgADZt+IXcnFySNmyjZ++OGI0GaoXXsE7xyvUzeflgNBoJDA6hStUBADg4FPz4DAyuzl1BwWz6aZW1fdq5MwBczLpATk42rq5uJRqpqxxUjUbN2uDo6EjKwb38umUjYL+jWA4ODvj4+FgfEO/u7o7BYKjgqkRuHRaLhaysLE6ePImPjw8ODg7FtleosxMHDhwgNzfX5tEq3t7e1keObNu2DbPZTEREhM122dnZ+Pv7W1+7u7tbAx1AUFCQ9Qdueno6qamptGjRwrre0dGRmJgY69Dwvn37yMrKKhTWcnJyaNy4sc2ymJiYax7X+PHjGTt2rPV1RkYGISEh19zuVtK+Y3OSN+8kPS0TgNoRNfAwuQNw7mxGwZSpsxMH9hc8P9HBwUjt8FDr9s7OBX+ZVaseyIBBBY8J+vzTbzmSchxXVxdru+CQQPbt+Y3gkD+eEZmTk1uux/ZXcznIXcloNGI0GrlwPoPc3BycnJzx9vHnMPtwdfPA2bng3+jC+YJ/f29ffxrGFPzBs3XTOtJ/D4ChteoQGd0EMLBr2xYO7N1xcw6qgl1+punlnzMiUpiPj0+Jnv+rUPcXcf78eRwcHEhKSiqU9K+8Z8+fh3YNBkOpznc5f/48AN9++22hK7VcXFxsXnt4XHvkwsXFpdB2t5uWrRvTpXt70tMyyMnJpVLlghCdnZ1D4o+bCKkexD2DupOVdZG0cxl4+3jh4eEGwLEjJ6znxjk7OVG5ir/1e4BdO/bTpXtbHB0dGTb8bk6dPEtApYJbG1y8eIkd2/b+uRy5DkdTDnA05YDNstguvXH3MNncp27vzm1ERTelelht/Pwr4eLqhtFo5NKlixw+WHBunIODAyZPb+v3AD5+AUQ1LPgjJzc3lyrBIVQJ/uOPl80/ryb7UvH3K7xdGQwGgoKCqFy5Mrm5+iNE5M+cnJyuOUJ3mUKdnahZsyZOTk42zwhMT09nz549tGvXjsaNG2M2mzl58uR131jR29uboKAg1q9fT7t27QDIy8sjKSnJerfsqKgoXFxcSElJsZlq/Stb+cPPNGhYhyqBAfj5e5B2Lp3fDh0l4ft1nD51FicnJ3bvPEBQ1cpUrhKAxZLPieOn2bXzAKt+WEdxmfrUybO8/86nxHZsSbXqgVSq7Mf5zKzf+/+JzMzCJ/RL+Tm0bxd5ubmE1a6Lu8mTnOxLnEg9wu7tyeTkZBe5ndH4x/l4Tk5O+PoF/Gl9yX6g384cHBxK/ItLRK5Ooc5OeHp6MnToUJ555hn8/PyoXLkyL7/8MkajEYPBQEREBIMHD+aBBx5gypQpNG7cmFOnTpGQkEB0dDQ9evQo0X6eeuopJk+eTHh4OHXr1mXq1Kk2z+3z9PQkLi6OMWPGkJ+fzx133EF6ejqJiYl4eXkxdOjQcnoHbl0bft7Khp+3Frn+6JHjxH947TuOH9h/mOfGvlFo+eHfUlkQ/9VVtpDytGrZ11ddfuS3/Rz5bX+R2509fZIlX358zWUiIqWlUGdHpk6dymOPPUbPnj3x8vJi3LhxHD582Prw3/j4eF599VWefvppjh49SkBAAC1btqRnz54l3sfTTz9NamoqQ4cOxWg08uCDD9K3b1/S09Otbf75z39SqVIlJk2axIEDB/Dx8aFJkyaleoSNiIiIlI7BohsE2a0LFy4QHBzMlClTeOihhyq6nDKRkZGBt7c3o0e8gouLa0WXI+Wk3R1ld+d8KR/d+w0uUbvLn9n09HS8vLyuvYGIXDeN1NmRLVu2sGvXLpo3b056err1ViS9e/eu4MpERESkvCnU2Zm33nqL3bt34+zsTNOmTVmzZg0BAQHX3lBERERuawp1dqRx48YkJSVVdBkiIiJSAfSYMBERERE7oFAnIiIiYgcU6kRERETsgEKdiIiIiB1QqBMRERGxAwp1IiIiInZAoU5ERETEDijUiYiIiNgBhToRERERO6BQJyIiImIHFOpERERE7IBCnYiIiIgdUKgTERERsQMKdSIiIiJ2wGCxWCwVXYRISWVkZODt7U16ejpeXl4VXY6IXIM+syI3j0bqREREROyAQp2IiIiIHVCoExEREbEDCnUiIiIidkChTkRERMQOKNSJiIiI2AGFOhERERE7oFAnIiIiYgccK7oAkdK4fK/sjIyMCq5EREri8mdV97kXKX8KdXJbyczMBCAkJKSCKxGR0sjMzMTb27uiyxCxa3pMmNxW8vPzOXbsGJ6enhgMhoouR0SuwWKxkJmZSdWqVTEadcaPSHlSqBMRERGxA/qzSURERMQOKNSJiIiI2AGFOhERERE7oFAnIiIiYgcU6kRERETsgEKdiIiIiB1QqBMRERGxA/8P8t0CjQpNdbcAAAAASUVORK5CYII=",
+      "text/plain": [
+       "<Figure size 640x480 with 2 Axes>"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "image/svg+xml": [
+       "<?xml version=\"1.0\" encoding=\"UTF-8\" standalone=\"no\"?>\n",
+       "<!DOCTYPE svg PUBLIC \"-//W3C//DTD SVG 1.1//EN\"\n",
+       " \"http://www.w3.org/Graphics/SVG/1.1/DTD/svg11.dtd\">\n",
+       "<!-- Generated by graphviz version 9.0.0 (20230911.1827)\n",
+       " -->\n",
+       "<!-- Pages: 1 -->\n",
+       "<svg width=\"227pt\" height=\"143pt\"\n",
+       " viewBox=\"0.00 0.00 226.75 142.50\" xmlns=\"http://www.w3.org/2000/svg\" xmlns:xlink=\"http://www.w3.org/1999/xlink\">\n",
+       "<g id=\"graph0\" class=\"graph\" transform=\"scale(1 1) rotate(0) translate(4 138.5)\">\n",
+       "<polygon fill=\"white\" stroke=\"none\" points=\"-4,4 -4,-138.5 222.75,-138.5 222.75,4 -4,4\"/>\n",
+       "<!-- 0 -->\n",
+       "<g id=\"node1\" class=\"node\">\n",
+       "<title>0</title>\n",
+       "<polygon fill=\"lightgrey\" stroke=\"black\" points=\"152.5,-134.5 0,-134.5 0,-93.5 152.5,-93.5 152.5,-134.5\"/>\n",
+       "<text text-anchor=\"middle\" x=\"76.25\" y=\"-117.2\" font-family=\"Times,serif\" font-size=\"14.00\">Restarted cohort analysis</text>\n",
+       "<text text-anchor=\"middle\" x=\"76.25\" y=\"-100.7\" font-family=\"Times,serif\" font-size=\"14.00\"> (n=1000)</text>\n",
+       "</g>\n",
+       "<!-- 1 -->\n",
+       "<g id=\"node2\" class=\"node\">\n",
+       "<title>1</title>\n",
+       "<polygon fill=\"lightgrey\" stroke=\"black\" points=\"108.25,-41 44.25,-41 44.25,0 108.25,0 108.25,-41\"/>\n",
+       "<text text-anchor=\"middle\" x=\"76.25\" y=\"-23.7\" font-family=\"Times,serif\" font-size=\"14.00\">Cohort 2</text>\n",
+       "<text text-anchor=\"middle\" x=\"76.25\" y=\"-7.2\" font-family=\"Times,serif\" font-size=\"14.00\"> (n=500)</text>\n",
+       "</g>\n",
+       "<!-- 0&#45;&gt;1 -->\n",
+       "<g id=\"edge1\" class=\"edge\">\n",
+       "<title>0&#45;&gt;1</title>\n",
+       "<path fill=\"none\" stroke=\"black\" d=\"M76.25,-93.15C76.25,-81.28 76.25,-65.94 76.25,-52.5\"/>\n",
+       "<polygon fill=\"black\" stroke=\"black\" points=\"79.75,-52.83 76.25,-42.83 72.75,-52.83 79.75,-52.83\"/>\n",
+       "<text text-anchor=\"middle\" x=\"147.5\" y=\"-62.2\" font-family=\"Times,serif\" font-size=\"14.00\">filtered to first 500 entries</text>\n",
+       "</g>\n",
+       "</g>\n",
+       "</svg>\n"
+      ],
+      "text/plain": [
+       "<graphviz.graphs.Digraph at 0x2bff4ff90>"
+      ]
+     },
+     "execution_count": 18,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "pop_track(adata, label=\"Restarted cohort analysis\")\n",
+    "adata = adata[:500]\n",
+    "pop_track(adata, label=\"Cohort 2\", operations_done=\"filtered to first 500 entries\")\n",
+    "pop_track.plot_cohort_change()\n",
+    "pop_track.plot_flowchart()"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "Some nice plotting options are available"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 19,
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAnUAAAHVCAYAAACXAw0nAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjguMiwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy8g+/7EAAAACXBIWXMAAA9hAAAPYQGoP6dpAACXKklEQVR4nOzdd1yV1R8H8M+9F+5lXPYGkSFDEMGBuAUVBVfuvUitzNDUyLTs5yizLE3NTC3F0UCztHIrbjTFgbk3gorgYMq+9/n9QTx6YwgKotfP+/W6L7nPOc853/PAlS/nPEMiCIIAIiIiInqpSWs6ACIiIiJ6dkzqiIiIiLQAkzoiIiIiLcCkjoiIiEgLMKkjIiIi0gJM6oiIiIi0AJM6IiIiIi3ApI6IiIhICzCpIyIiItICTOqIqFQrV66ERCJBfHz8c+/z2LFjz63Pl8HevXshkUiwd+/eamlfIpFg+vTp1dI2ET0/TOrolVGcMBS/dHR04ODggLCwMNy6dava+t2yZUu1/cLMzs7G9OnTq+2XvbZbvHgxVq5cWdNhEBFVCZ2aDoDoeZs5cyZcXFyQm5uLv//+GytXrsTBgwdx5swZ6OnpVXl/W7ZswbffflstiV12djZmzJgBAAgKCqry9rXd4sWLYWlpibCwsJoOpUbl5ORAR4e/DohedvwU0yunU6dO8Pf3BwCMGjUKlpaW+OKLL/Dnn3+iX79+NRxdxajVauTn59d0GC+t7OxsGBgY1HQYL4zq+GOGiJ4/Lr/SK69169YAgKtXr2psv3DhAvr06QNzc3Po6enB398ff/75p0adgoICzJgxA+7u7tDT04OFhQVatWqFnTt3AgDCwsLw7bffAoDG0m+xr776Ci1atICFhQX09fXRuHFjrF+/vkSMEokE4eHh+Omnn1CvXj0oFAosWbIEVlZWAIAZM2aIbT8+I1iRMQDA2bNn0a5dO+jr66NWrVr49NNPoVarK3wML1y4gH79+sHKygr6+vrw9PTERx99pFHn5MmT6NSpE4yNjaFUKtG+fXv8/fffpbaXl5eHiRMnwsrKCoaGhujZsyfu3r1bot7ixYvF42Fvb4933nkHaWlpGnWCgoLg4+OD48ePo02bNjAwMMCHH34IZ2dnnD17Fvv27ROP3ZNmOyv7/dq4cSN8fHygUChQr149bNu2TaPejRs3MGbMGHh6ekJfXx8WFhbo27fvE89jnDZtGnR1dUs9Jm+++SZMTU2Rm5sLADh27BhCQkJgaWkJfX19uLi4YMSIESXiffznJjMzE+PHj4ezszMUCgWsra3RoUMHnDhxoty4iKhmcaaOXnnFv0DNzMzEbWfPnkXLli3h4OCAyZMnw9DQEOvWrUOPHj3w22+/oWfPngCA6dOnY/bs2Rg1ahQCAgKQkZGBY8eO4cSJE+jQoQPeeust3L59Gzt37sSaNWtK9L1gwQK89tprGDx4MPLz8xEVFYW+ffti06ZN6NKli0bd3bt3Y926dQgPD4elpSX8/Pzw3Xff4e2330bPnj3Rq1cvAICvr2+lxnDnzh20bdsWhYWFYr1ly5ZBX1+/Qsfvn3/+QevWraGrq4s333wTzs7OuHr1Kv766y/MmjVLjKV169YwNjbGpEmToKuri6VLlyIoKAj79u1D06ZNNdocO3YszMzMMG3aNMTHx2P+/PkIDw/H2rVrxTrTp0/HjBkzEBwcjLfffhsXL17Ed999h9jYWMTExEBXV1ese//+fXTq1AkDBgzAkCFDYGNjg6CgIIwdOxZKpVJMQG1sbModa2W+XwcPHsTvv/+OMWPGwMjICAsXLkTv3r2RkJAACwsLAEBsbCwOHTqEAQMGoFatWoiPj8d3332HoKAgnDt3rszZxKFDh2LmzJlYu3YtwsPDxe35+flYv349evfuDT09PaSkpKBjx46wsrLC5MmTYWpqivj4ePz+++/ljnP06NFYv349wsPD4e3tjfv37+PgwYM4f/48GjVqVO6+RFSDBKJXRGRkpABA2LVrl3D37l0hMTFRWL9+vWBlZSUoFAohMTFRrNu+fXuhfv36Qm5urrhNrVYLLVq0ENzd3cVtfn5+QpcuXcrt95133hHK+qhlZ2drvM/Pzxd8fHyEdu3aaWwHIEilUuHs2bMa2+/evSsAEKZNm1ai7YqOYfz48QIA4ciRI+K2lJQUwcTERAAgXL9+vdzxtWnTRjAyMhJu3LihsV2tVotf9+jRQ5DL5cLVq1fFbbdv3xaMjIyENm3aiNuKv0fBwcEa+0+YMEGQyWRCWlqaGJ9cLhc6duwoqFQqsd6iRYsEAMKKFSvEbYGBgQIAYcmSJSVir1evnhAYGFju+B5Xme+XXC4Xrly5Im47deqUAED45ptvymxPEATh8OHDAgBh9erV4rY9e/YIAIQ9e/aI25o3by40bdpUY9/ff/9do96GDRsEAEJsbGy54/rvz5CJiYnwzjvvlLsPEb14uPxKr5zg4GBYWVnB0dERffr0gaGhIf7880/UqlULAPDgwQPs3r0b/fr1Q2ZmJu7du4d79+7h/v37CAkJweXLl8WrZU1NTXH27Flcvnz5qWJ5fDYsNTUV6enpaN26danLXIGBgfD29q5Qu5UZw5YtW9CsWTMEBASI+1tZWWHw4MFP7Ofu3bvYv38/RowYgdq1a2uUFS8zq1Qq7NixAz169ICrq6tYbmdnh0GDBuHgwYPIyMjQ2PfNN9/UWKZu3bo1VCoVbty4AQDYtWsX8vPzMX78eEilj/4be+ONN2BsbIzNmzdrtKdQKPD6668/cTxPUpnvV3BwMOrUqSO+9/X1hbGxMa5du1ZqewUFBbh//z7c3Nxgamr6xKXOYcOG4ciRIxqnDfz0009wdHREYGAggKKfTwDYtGkTCgoKKjxOU1NTHDlyBLdv367wPkRU85jU0Svn22+/xc6dO7F+/Xp07twZ9+7dg0KhEMuvXLkCQRDw8ccfw8rKSuM1bdo0AEBKSgqAoitp09LS4OHhgfr16+P999/HP//8U+FYNm3ahGbNmkFPTw/m5uawsrLCd999h/T09BJ1XVxcKtxuZcZw48YNuLu7l2jD09Pzif0UJyg+Pj5l1rl79y6ys7NLbc/LywtqtRqJiYka2/+bIBYvjaempooxlxajXC6Hq6urWF7MwcEBcrn8ieN5ksp8v/47huJxFI8BKLrq9H//+x8cHR2hUChgaWkJKysrpKWlldrm4/r37w+FQoGffvoJAJCeno5NmzZh8ODBYkIcGBiI3r17Y8aMGbC0tET37t0RGRmJvLy8ctueM2cOzpw5A0dHRwQEBGD69OkaySgRvZh4Th29cgICAsSrX3v06IFWrVph0KBBuHjxIpRKpXiBQEREBEJCQkptw83NDQDQpk0bXL16FX/88Qd27NiBH374AV9//TWWLFmCUaNGlRvHgQMH8Nprr6FNmzZYvHgx7OzsoKuri8jISPz8888l6lf0HDcAlRrDi0gmk5W6XRCEp2qvMseuLJX9flVkDGPHjkVkZCTGjx+P5s2bw8TEBBKJBAMGDHjihSpmZmbo2rUrfvrpJ/zvf//D+vXrkZeXhyFDhoh1JBIJ1q9fj7///ht//fUXtm/fjhEjRmDu3Ln4+++/oVQqS227X79+aN26NTZs2IAdO3bgyy+/xBdffIHff/8dnTp1qsjhIqIawKSOXmkymQyzZ89G27ZtsWjRIkyePFlcItTV1UVwcPAT2zA3N8frr7+O119/HVlZWWjTpg2mT58uJnWPLyM+7rfffoOenh62b9+uMVMYGRlZ4fjLarsyY3Bycip1+fjixYtP7L+4nzNnzpRZx8rKCgYGBqW2d+HCBUilUjg6Oj6xr8c5OTmJMT6+pJufn4/r169X6PsGlH38SlMV36//Wr9+PYYPH465c+eK23Jzc0tcwVuWYcOGoXv37oiNjcVPP/2Ehg0bol69eiXqNWvWDM2aNcOsWbPw888/Y/DgwYiKiir3Dw87OzuMGTMGY8aMQUpKCho1aoRZs2YxqSN6gXH5lV55QUFBCAgIwPz585Gbmwtra2sEBQVh6dKlSEpKKlH/8dtI3L9/X6NMqVTCzc1NY3nL0NAQAEr8opbJZJBIJFCpVOK2+Ph4bNy4scKxF18d+d+2KzOGzp074++//8bRo0c1youX9cpjZWWFNm3aYMWKFUhISNAoK56Rkslk6NixI/744w+NW3UkJyfj559/RqtWrWBsbPzEvh4XHBwMuVyOhQsXasx8LV++HOnp6SWuRC2LoaFhhROoqvh+ldbmf2cfv/nmG40+ytOpUyfxPov79u3TmKUDipar/9t+gwYNAKDMJViVSlVi6dfa2hr29vZPXLYloprFmToiAO+//z769u2LlStXYvTo0fj222/RqlUr1K9fH2+88QZcXV2RnJyMw4cP4+bNmzh16hQAwNvbG0FBQWjcuDHMzc1x7Ngx8VYQxRo3bgwAGDduHEJCQiCTyTBgwAB06dIF8+bNQ2hoKAYNGoSUlBR8++23cHNzq/B5efr6+vD29sbatWvh4eEBc3Nz+Pj4wMfHp8JjmDRpEtasWYPQ0FC8++674i1NnJycKhTHwoUL0apVKzRq1AhvvvkmXFxcEB8fj82bNyMuLg4A8Omnn2Lnzp1o1aoVxowZAx0dHSxduhR5eXmYM2dOZb5VAIqSySlTpmDGjBkIDQ3Fa6+9hosXL2Lx4sVo0qRJieSmLI0bN8Z3332HTz/9FG5ubrC2tka7du1KrVsV36//6tq1K9asWQMTExN4e3vj8OHD2LVrl3jLkyfR1dXFgAEDsGjRIshkMgwcOFCjfNWqVVi8eDF69uyJOnXqIDMzE99//z2MjY3RuXPnUtvMzMxErVq10KdPH/j5+UGpVGLXrl2IjY3VmFEkohdQzV14S/R8Fd8uo7TbO6hUKqFOnTpCnTp1hMLCQkEQBOHq1avCsGHDBFtbW0FXV1dwcHAQunbtKqxfv17c79NPPxUCAgIEU1NTQV9fX6hbt64wa9YsIT8/X6xTWFgojB07VrCyshIkEonG7U2WL18uuLu7CwqFQqhbt64QGRkpTJs2rcQtUACUeYuJQ4cOCY0bNxbkcnmJW1NUZAyCIAj//POPEBgYKOjp6QkODg7CJ598IixfvrxCtzQRBEE4c+aM0LNnT8HU1FTQ09MTPD09hY8//lijzokTJ4SQkBBBqVQKBgYGQtu2bYVDhw5p1Cnre1TaLT0EoegWJnXr1hV0dXUFGxsb4e233xZSU1M16gQGBgr16tUrNe47d+4IXbp0EYyMjAQAT7y9ybN+v5ycnIThw4eL71NTU4XXX39dsLS0FJRKpRASEiJcuHChRL2yxi8IgnD06FEBgNCxY8cSZSdOnBAGDhwo1K5dW1AoFIK1tbXQtWtX4dixYyXiLf65ycvLE95//33Bz89PMDIyEgwNDQU/Pz9h8eLF5R4bIqp5EkF4yjOPiYioxp06dQoNGjTA6tWrMXTo0JoOh4hqEM+pIyJ6iX3//fdQKpXiE0WI6NXFc+qIiF5Cf/31F86dO4dly5YhPDxcvCCHiF5dXH4lInoJOTs7Izk5GSEhIVizZg2MjIxqOiQiqmFM6oiIiIi0AM+pIyIiItICTOqIiIiItACTOqJX3NGjRyGXy3Hjxo2aDuWFs23bNiiVSo0ncBARvaiY1BG94j766CMMHDhQfJ7q85CVlYVp06YhNDQU5ubmkEgkWLlyZZn1z58/j9DQUCiVSpibm2Po0KGlJlpqtRpz5syBi4sL9PT04Ovri19++eWp2wwNDYWbmxtmz579TOMlInoemNQRvcLi4uKwa9cujB49+rn2e+/ePcycORPnz5+Hn59fuXVv3ryJNm3a4MqVK/jss88QERGBzZs3o0OHDsjPz9eo+9FHH+GDDz5Ahw4d8M0336B27doYNGgQoqKinrrNt956C0uXLkVmZmbVDJ6IqLrU5OMsiKhmjRs3Tqhdu7agVqufa7+5ublCUlKSIAiCEBsbKwAQIiMjS6379ttvC/r6+sKNGzfEbTt37hQACEuXLhW33bx5U9DV1dV4PJdarRZat24t1KpVS3z8W2XaFARBSE5OFmQymbB8+fJnGjMRUXXjTB3RK2zjxo1o164dJBKJxnZnZ2d07doVBw8eREBAAPT09ODq6orVq1dXSb8KhQK2trYVqvvbb7+ha9euqF27trgtODgYHh4eWLdunbjtjz/+QEFBAcaMGSNuk0gkePvtt3Hz5k0cPny40m0CgLW1NXx9ffHHH39UepxERM8TkzqiV9StW7eQkJCARo0alVp+5coV9OnTBx06dMDcuXNhZmaGsLAwnD17VqyjVqtx7969Cr0KCgqeKsaUlBT4+/uXKAsICMDJkyfF9ydPnoShoSG8vLxK1Csur2ybxRo3boxDhw5VOn4ioueJjwkjekVduHABAODi4lJq+cWLF7F//360bt0aANCvXz84OjoiMjISX331FQAgISGhzP3/a8+ePQgKCqpUjElJSQAAOzu7EmV2dnZ48OAB8vLyoFAokJSUBBsbmxKzjsX73r59u9JtFnN1dcW9e/eQkpICa2vrSo2BiOh5YVJH9Iq6f/8+AMDMzKzUcm9vbzGhAwArKyt4enri2rVr4jZbW1vs3LmzQv096YKI0uTk5ACARoJVTE9PT6yjUCjEf8urV9k2ixUfo3v37jGpI6IXFpM6olecUMaTAh8/36yYmZkZUlNTxfd6enoIDg6uttj09fUBAHl5eSXKcnNzNero6+tXuF5F2yxWfIz+OwtIRPQiYVJH9IqysLAAAI0k7XEymazU7Y8ngSqVqsI35jU3N4dcLq9UjMVLpMVLpo9LSkqCubm5OKNmZ2eHPXv2QBAEjeSreF97e/tKt1ms+BhZWlpWKn4ioueJSR3RK6pu3boAgOvXrz91G4mJidV6Tp2DgwOsrKxw7NixEmVHjx5FgwYNxPcNGjTADz/8gPPnz8Pb21vcfuTIEbG8sm0Wu379OiwtLWFlZVWp+ImInicmdUSvKAcHBzg6Opaa3FRUdZ9TBwC9e/fGqlWrkJiYCEdHRwBAdHQ0Ll26hAkTJoj1unfvjgkTJmDx4sVYtGgRgKJZxSVLlsDBwQEtWrSodJvFjh8/jubNmz9V/EREzwuTOqJXWPfu3bFhw4YSS5YV9Szn1C1atAhpaWniVal//fUXbt68CQAYO3YsTExMAAAffvghfv31V7Rt2xbvvvsusrKy8OWXX6J+/fp4/fXXxfZq1aqF8ePH48svv0RBQQGaNGmCjRs34sCBA/jpp580lpMr2iYApKSk4J9//sE777zzVOMkInpuavLOx0RUs06cOCEAEA4cOKCx3cnJSejSpUuJ+oGBgUJgYGCV9O3k5CQAKPV1/fp1jbpnzpwROnbsKBgYGAimpqbC4MGDhTt37pRoU6VSCZ999png5OQkyOVyoV69esKPP/5Yav8VbfO7774TDAwMhIyMjCoZNxFRdZEIQhmXvhHRK6F9+/awt7fHmjVrajqUF1LDhg0RFBSEr7/+uqZDISIqF5M6olfckSNH0Lp1a1y+fBlOTk41Hc4LZdu2bejTpw+uXbvG+9MR0QuPSR0RERGRFuCzX4mIiIi0AJM6IiIiIi3ApI6IiIhICzCpIyIiItICvPkwvVTUajVu374NIyMjPlyd6CUgCAIyMzNhb28PqZTzCETViUkdvVRu374tPtaJiF4eiYmJqFWrVk2HQaTVmNTRS8XIyAhA0S8IY2PjGo6GiJ4kIyMDjo6O4meXiKoPkzp6qRQvuRobGzOpI3qJ8HQJourHExyIiIiItACTOiIiIiItwKSOiIiISAswqSMiIiLSAkzqiIiIiLQAr36ll1LulPmQK/RqOgyiV5bevEk1HQIR/Qdn6oiIiIi0AJM6IiIiIi3ApI6IiIhICzCpIyIiItICTOqIiIiItACTOiIiIiItwKSOiIiISAswqSMiIiLSAkzqiIiIiLQAkzoiIiIiLcCkjoiIiEgLMKkjIiIi0gJM6oiIiIi0AJM6IiIiIi3ApI6Inhudzq2hN28SdEJaPrc+JZamUHwVAflHbwIy/pdHRNpLp6YDIKKaozvsNcga1AUAqE6eR8Gav568k6E+dDq2gKyeG2CsBHLzoL59F4XrtkF4kF7ufrLWjSEUFqLw4Alxs06XNpD6uENiogRkMiDzIVSXb6BwxyEgNaPs9hRy6HRqBalLLUjMjAGFLoTUTKjjLqBwz1EgLx8AINxLg/rMZch8PSEL8IXqcFxFDg0R0Uvnmf5s3bt3LyQSCdLS0qoonMpbuXIlTE1Nq6y9oKAgjB8/vsraexrTp09HgwYNarydsLAw9OjR45njoBeTrImPmNBVmKE+5OOHQqd1Y8DYEMLdBxCysiF1tgdMlOX3518PEoUc6ovxwMMccbvU0wUSuS6Eu6kQ0jIhMTeBTlNfyN/sW257EgM96LTxh8TWEkJ6JpBXAKm1OXQ6toDusG4adVUnzhfF0LJB5cZLRPQSqdRMXVBQEBo0aID58+cDAFq0aIGkpCSYmJhUR2wV0r9/f3Tu3LnG+q8OERERGDt2rPg+LCwMaWlp2LhxY80FRVpFYmEKnZ7BUF+/BYmpUdFMVwXodGoNqYUp1El3kb9kHZD5sKhAJgUgKXdfWUMvAID67FWN7fkLfwQKVeJ73UFdIPOvB6mNBWCgB2TnltqeUKhCwZ97oDp8qmhWTkcG+ZgBkDo7QOZVBwX6CiAnr6jP89cgqNSQ2ltDYmMBIfl+hcZLRPQyeablV7lcDltb26qK5ano6+tDX1+/RmOoakqlEkpl+bMeRE9NKoHukK6AIKDgp02QjxlQ4V1lDTwBAEJaJuSj+0FibgLhfhoKo49AffJ82TvKdSFxsAEAqBOTNMsKVZC1aABZEx/AQB9SK7OienfulZnQAShapt0bq9GOOvEOpM4OENRqQC08KisohJB8DxJ7a0hda0HFpI6ItFCFl1/DwsKwb98+LFiwABKJBBKJBCtXrtRYfi1eCt20aRM8PT1hYGCAPn36IDs7G6tWrYKzszPMzMwwbtw4qFSP/jLPy8tDREQEHBwcYGhoiKZNm2Lv3r0Viuu/y6/FS45r1qyBs7MzTExMMGDAAGRmZlZ0qFCr1Zg0aRLMzc1ha2uL6dOna5QnJCSge/fuUCqVMDY2Rr9+/ZCcnCyWnzp1Cm3btoWRkRGMjY3RuHFjHDt2TCPejRs3wt3dHXp6eggJCUFiYmKJMRR/vWrVKvzxxx/icS8+Nh988AE8PDxgYGAAV1dXfPzxxygoKKjwOB+nUqkwceJEmJqawsLCApMmTYIgCBp1tm3bhlatWol1unbtiqtXH826tGvXDuHh4Rr73L17F3K5HNHR0QCAxYsXi+O2sbFBnz59nipeeno6HVtC6mSPgt92ln8O3H8pDSAxKPoDSublCom+HpCTB6m9NeRDu0Hq61HmrhJzE0j+vUihtD4lZsaQOtk/SuhuJhfNBFaG0gCyf2NQn7wgnlNXTHhQdH6exNKscu0SEb0kKpzULViwAM2bN8cbb7yBpKQkJCUlwdHRsUS97OxsLFy4EFFRUdi2bRv27t2Lnj17YsuWLdiyZQvWrFmDpUuXYv369eI+4eHhOHz4MKKiovDPP/+gb9++CA0NxeXLl59qUFevXsXGjRuxadMmbNq0Cfv27cPnn39e4f1XrVoFQ0NDHDlyBHPmzMHMmTOxc+dOAEUJX/fu3fHgwQPs27cPO3fuxLVr19C/f39x/8GDB6NWrVqIjY3F8ePHMXnyZOjq6moco1mzZmH16tWIiYlBWloaBgwofbYkIiIC/fr1Q2hoqHjcW7RoAQAwMjLCypUrce7cOSxYsADff/89vv7666c5ZJg7dy5WrlyJFStW4ODBg3jw4AE2bNigUefhw4eYOHEijh07hujoaEilUvTs2RNqtRoAMGrUKPz888/Iy8sT9/nxxx/h4OCAdu3a4dixYxg3bhxmzpyJixcvYtu2bWjTpk25ceXl5SEjI0PjRU9PUssWsvbNoDp2FuoT5yq3s/TRfxfqO/eQN2sp8mYtLZpRA6DTqlHZ++opHn39n2QLAAo370fue18ib/YPUF2+AWktm6LZREn5S7rFJBamkIcPgsTECOprN1GwfkfJSsU/l4/HQkSkRSq8/GpiYgK5XA4DAwNxyfXChQsl6hUUFOC7775DnTp1AAB9+vTBmjVrkJycDKVSCW9vb7Rt2xZ79uxB//79kZCQgMjISCQkJMDe3h5AUSKzbds2REZG4rPPPqv0oNRqNVauXAkjIyMAwNChQxEdHY1Zs2ZVaH9fX19MmzYNAODu7o5FixYhOjoaHTp0QHR0NE6fPo3r16+LSe3q1atRr149xMbGokmTJkhISMD777+PunXrim389xgtWrQITZs2BVCURHp5eeHo0aMICAjQqKtUKqGvr4+8vLwSS91Tp04Vv3Z2dkZERASioqIwadKkih4q0fz58zFlyhT06tULALBkyRJs375do07v3r013q9YsQJWVlY4d+4cfHx80KtXL4SHh+OPP/5Av379ABTNTIaFhUEikSAhIQGGhobo2rUrjIyM4OTkhIYNG5Yb1+zZszFjxoxKj4dKJ7WzhEQmhdTPA4r6//5c/vsHh9TXA4rZ45E3YzGQWzLxQlY2hMJCSHR0INy+C6iKknnh9l3A1hIS83LOrX0s0YdcDuTmlawjCBDuPoBq/3HI3J0gc6sNlbsT1Jfiyx2TxMke8pG9IFEaQHXmCgrW/AkUFJasqPg3mSutbyIiLVDlN20yMDAQEzoAsLGxgbOzs8Y5YjY2NkhJSQEAnD59GiqVCh4eHuK5ZEqlEvv27dNY2qsMZ2dnMaEDADs7O7G/ivD19dV4//j+58+fh6Ojo8Yspbe3N0xNTXH+fNE5RRMnTsSoUaMQHByMzz//vMQ4dHR00KRJE/F93bp1NfavqLVr16Jly5awtbWFUqnE1KlTkZCQUKk2ACA9PR1JSUliklkco7+/v0a9y5cvY+DAgXB1dYWxsTGcnZ0BQOxTT08PQ4cOxYoVKwAAJ06cwJkzZxAWFgYA6NChA5ycnODq6oqhQ4fip59+QnZ2drmxTZkyBenp6eLr8WVqenoSXV1IFPKil7RoNkwik0GikKP4ggfd0f0h/2AkdLr8O5uqVkN99WZRXXuropk7qbToawDqu6ll9ifcT4fwbxIoMX90UYbE0gzSem6PrrGQANK6Lo92VPw7w22ihPyDkZB/MBLS+o/+SJL6ekA+pj8kSgMUHjiOgsjfS0/oHutXuFd2nEREL7Mqv0/d48uMACCRSErdVrxkl5WVBZlMhuPHj0Mmk2nUe9qLBcrr73nsP336dAwaNAibN2/G1q1bMW3aNERFRaFnz54VbuNJDh8+jMGDB2PGjBkICQmBiYkJoqKiMHfu3Crr47+6desGJycnfP/997C3t4darYaPjw/y8x/N6owaNQoNGjTAzZs3ERkZiXbt2sHJyQlA0XLxiRMnsHfvXuzYsQP/+9//MH36dMTGxpZ5WxqFQgGFgstlVUUVewaq2DMa2xRT34LE3KTEfeqklqZFF0LcuC1uK9x6ANI6tSC1tYRi6lsAAImpEQSVGqrov8vuOL8Awq1kSGrbQepoC9Xtu0X7mighH9kLQm4+hAdpkCgNIDEu+twLqRlQX7pRVE8qLboaFoCkePnUWAndYd0hkUogFBZCWtsO8nFDxC4L1u+EcOvfc111dSCxsQQAqK/drOxhIyJ6KVRqpk4ul2tc4FAVGjZsCJVKhZSUFLi5uWm8avrK2tJ4eXkhMTFRY8bo3LlzSEtLg7e3t7jNw8MDEyZMwI4dO9CrVy9ERkaKZYWFheKFEwBw8eJFpKWlwcvLq9Q+Szvuhw4dgpOTEz766CP4+/vD3d0dN27ceKoxmZiYwM7ODkeOHNGI8fjx4+L7+/fv4+LFi5g6dSrat28PLy8vpKaWnPGoX78+/P398f333+Pnn3/GiBEjNMp1dHQQHByMOXPm4J9//kF8fDx27979VHHT8yckJCF/8VqoriQA+gpARwbVxXjkf/MT1FfKnyVW/Xt1rNTb7VF7qRlQnb4E5ORCYmUOGOhBfS8VhYfikLfwx1LPvysmkUkfzTLq6BRdaPHYS6InF+tKvVwhkUmhvp3C25kQkdaq1Eyds7Mzjhw5gvj4eCiVykrNXpXFw8MDgwcPxrBhwzB37lw0bNgQd+/eRXR0NHx9fdGlS5dn7qMqBQcHo379+hg8eDDmz5+PwsJCjBkzBoGBgfD390dOTg7ef/999OnTBy4uLrh58yZiY2M1zkfT1dXF2LFjsXDhQujo6CA8PBzNmjUrcT5dMWdnZ2zfvh0XL16EhYUFTExM4O7ujoSEBERFRaFJkybYvHlziQsbKuPdd9/F559/Dnd3d9StWxfz5s3TuKm0mZkZLCwssGzZMtjZ2SEhIQGTJ08uta1Ro0YhPDwchoaGGrOTmzZtwrVr19CmTRuYmZlhy5YtUKvV8PT0fOq46dnlfbq0UtuF+FsoWBxV6X5Ux85CJ7RV0fKqoT7wMAfCg3QURG584r5CagZyJ8554rayyBoV/cGlijlZ6biJiF4WlZqpi4iIgEwmg7e3N6ysrJ7q/K3SREZGYtiwYXjvvffg6emJHj16IDY2FrVr166S9quSRCLBH3/8ATMzM7Rp0wbBwcFwdXXF2rVrAQAymQz379/HsGHD4OHhgX79+qFTp04aJ/sbGBjggw8+wKBBg9CyZUsolUpx/9K88cYb8PT0hL+/P6ysrBATE4PXXnsNEyZMQHh4OBo0aIBDhw7h448/fupxvffeexg6dCiGDx+O5s2bw8jISCMhk0qliIqKwvHjx+Hj44MJEybgyy+/LLWtgQMHQkdHBwMHDoSenp643dTUFL///jvatWsHLy8vLFmyBL/88gvq1av31HHTS+RhDlQHjkOiq1P+lbJVTGJpCqmPG9T306A6cvq59UtE9LxJhP/ejIyq1cqVKzF+/PgafbRadYuPj0edOnUQGxuLRo2q9pd3RkYGTExMkDxmBowVek/egYiqhd68il1lX/yZTU9Ph7FxxZ5cQkRPp8ovlKBXV0FBAe7fv4+pU6eiWbNmVZ7QERERUdmq/JYmVa1Tp04atzp5/FXZe9glJCSU2ZZSqayy5eQXUXnjPnDgQJX0ERMTAzs7O8TGxmLJkiVV0iYRERFVzAu//Hrr1i3k5OSUWmZubg5zc/MKt1VYWIj4+Pgyy52dnaGjo52Tl1euXCmzzMHB4aV5fi6XX4leDFx+JXrxvPAZjIODQ5W1paOjAzc3tydX1EKv6riJiIheFS/88isRERERPRmTOiIiIiItwKSOiIiISAswqSMiIiLSAkzqiIiIiLQAkzoiIiIiLcCkjoiIiEgLMKkjIiIi0gJM6oiIiIi0AJM6IiIiIi3ApI6IiIhIC7zwz34lKo3e7PHQ48PBiYiIRJypIyIiItICTOqIiIiItACTOiIiIiItwKSOiIiISAswqSMiIiLSAkzqiIiIiLQAkzoiIiIiLcCkjoiIiEgLMKkjIiIi0gJM6oiIiIi0AJM6IiIiIi3ApI6IiIhICzCpIyIiItICOjUdANHTeGvKfMgVejUdRrXa6hMFAPjUolMNR0KvKt28LgCAb0/5lSg7NsvweYdDRE/AmToiIiIiLcCkjoiIiEgLMKkjIiIi0gJM6oiIiIi0AJM6IiIiIi3ApI6IiIhICzCpIyIiItICTOqIiIiItACTOiIiIiItwKSOiIiISAswqSMiIiLSAkzqiIiIiLQAkzoiIiIiLaBT0wEQPQ+hgU3QoF4d2Fmbw9BAD+kZD3HhaiI2bo/B3QfpAIDJYwbAy612iX0vXbuJWYt+Lrd9SzNj9AhpCS+32jA2MsT91AzsP/IPtu49CkGoliG9EpT6phjU8f0yy49fiMbxi7uhqyOHf91guNj7QF9hiIc56biUGIeTl/ZCENQV6svCxA49Wo+GTFb03+La6K+RnnWvSsZBRPQ8MKmjV0Jw60awMDXGnbsPkF9QCGsLU7QyN4GPpzM+mP0DcvPyxbop91KR8TBHfH8rufxf7EaG+pg2fiiMjQyRk5uPpJT7qGVrhf7dgmBqosTPG3dX27i0nUqtQvKDBI1tCl19mBpZAQCyczMBSBDSdCjsLV2hUhci82EqjJUW8K/bHsaG5th7Yv0T+5FJddCucT8xoSMiehnxfzB6Jez7+xRijp3Fg7RMAMCg7u0QEugPU2Ml6nk44fjpy2LdP3YexsHYMxVuu4mfJ4yNDAEAnyxYg1vJ9xHYzBcj+oUiuGUjbNsbK/ZLlZOTl4k/DizV2NayfjeYGlkhNz8bl2+egrOdF+wtXQEAO4/+jITki6jn0gwtfbvBw7EhTl89hPvpt8vtp7lPZ5gZWePqrdOo41C/2sZDRFSdeE4dVUh+fv6TK73A/tr1t0ZidfFaovh1QaFKo+6g7u3ww5yJ+PKjN/F63xAYKw3KbVsqlYhfq/9daxX+/Vcmk5a6pEtPR6GrD4/ajQAA5+OPolCVD0drDwBAYWE+EpIvAQCuJ50V93G0di+3zdo2deHt0hRnrh1CYvLFaoqciKj6MamjUgUFBSE8PBzjx4+HpaUlQkJCMG/ePNSvXx+GhoZwdHTEmDFjkJWVpbFfTEwMgoKCYGBgADMzM4SEhCA1NRUAoFarMXv2bLi4uEBfXx9+fn5Yv/7JS2NVTSKRIKi5H4CipdZzl26IZXn5BUhNz0RmVg6sLUwR1NwPH787BHK5bpntnTp3DTm5RUnvtPFDMfO94Rjep6NYbmairKaRvHq8XZpBV0eOQlUBzlw7DABQ6psAAHILsgEUJdM5uY9+LpX6pmW2p69QIrBhT9xPv4MjZ7dXW9xERM8Dkzoq06pVqyCXyxETE4MlS5ZAKpVi4cKFOHv2LFatWoXdu3dj0qRJYv24uDi0b98e3t7eOHz4MA4ePIhu3bpBpSqaCZs9ezZWr16NJUuW4OzZs5gwYQKGDBmCffv2lRlDXl4eMjIyNF7PQi7Xxbuv94RvXVekZWTh6+W/o/Df+H7+YzfGfLQQH30ZiQkzv8Nfu4qSBmsLU/jXL3u25+6DdHy5dB3OXb4BtSDAzFiJg7FnoFYXJRgqVcVO1KfySaUy1HNpCgC4cvMUcvKyyq4skZRd9pjWfj2gq6PA7uNroVIXVkWYREQ1hufUUZnc3d0xZ84c8b2np6f4tbOzMz799FOMHj0aixcvBgDMmTMH/v7+4nsAqFevHoCi5Oyzzz7Drl270Lx5cwCAq6srDh48iKVLlyIwMLDUGGbPno0ZM2ZUyXhMjAwxYVRvuDjaIinlAeYu+1W88hUAEm6laNQ/fOI8ugUXxWpualxu21dv3MYX360V39dxskdQs6LZwKSUB1US/6vOw7EhDPSMIAhq/HPloLg9K6foe6gnNwQgASBAX2H4WHlamW1amNhCKpWhR5vRAACJ5NHfub0C38HZ63/j6DnO4BHRy4FJHZWpcePGGu937dqF2bNn48KFC8jIyEBhYSFyc3ORnZ0NAwMDxMXFoW/fvqW2deXKFWRnZ6NDhw4a2/Pz89GwYcMyY5gyZQomTpwovs/IyICjo2Olx+JgY4EJb/SBlbkJLl5NxILIDXiYnSuWGykN0KKxN/b9/Y94JWzTBnXF8nupRYmDmYkSk0b3BwCs37JfvMDC3cUBV+JvQxAEGOgrMPC1oKJ4s7Jx7vKj5V16er51WgEAEpIvIS3rrrj9ZspleDk3gY5MF7VtPJCQfBEudvXE8sSUou+Rs503AryKlsU3HVqB7NyiWV+pRAqpjqJEf7o6csiksmobDxFRVWNSR2UyNHw02xEfH4+uXbvi7bffxqxZs2Bubo6DBw9i5MiRyM/Ph4GBAfT19ctsq/jcu82bN8PBwUGjTKEo+Qv18bLyyitq7Os9YWVedO6VnkKOiW/0Ecv2//0Pzl6Kx6Du7dCvayBS7qVBIdeFhVnR7NytO/dw/J+iE/BlUinsbSwAAPp6j+IK69MRZiZGeJCWAWsLUygUcqhUaqxavwP5BVzWe1ZOtnXF25icunJAoyw+6RyS7sfDzsIZHQIGIePhA5goLQEAlxPjxCtf5ToKsQ3pvzNyv+z8SqMtD8eGCGpU9LPB+9QR0cuGSR1VyPHjx6FWqzF37lxIpUW/ENetW6dRx9fXF9HR0aUul3p7e0OhUCAhIaHMpdbqpKPzaMbFqZaNRtnpC9eR8TAHf+48BB9PF1hbmEJXVwe3k+/jxJnL2Lz7SIkrZP/rzMV4+Pt6wNbaHAUFKpy+cB1/7TqMi9duVst4XjXFs3QpqYm4cz9eo0yAgG1/r4Z/3WC42vvA2NAcWdlpuJwYhxOX9tRAtERENUMiCLzfPZUUFBSEBg0aYP78+QCAU6dOie+7deuGmJgYTJkyBbdu3UJqaipMTU1x6dIl1K9fHyNHjsTo0aMhl8uxZ88e9O3bF5aWlpg6dSqWLFmCuXPnolWrVkhPT0dMTAyMjY0xfPjwCsWVkZEBExMTDBgzA3KFXjUegZq31ScKAPCpRacajoReVbp5XQAA357yK1F2bJZhiW2lKf7Mpqenw9i4/HNTiejZ8OpXqhA/Pz/MmzcPX3zxBXx8fPDTTz9h9uzZGnU8PDywY8cOnDp1CgEBAWjevDn++OMP6OgUTQh/8skn+PjjjzF79mx4eXkhNDQUmzdvhouLS00MiYiISKtwpo5eKpypI3p+OFNH9HLhTB0RERGRFmBSR0RERKQFmNQRERERaQEmdURERERagEkdERERkRZgUkdERESkBZjUEREREWkBJnVEREREWoBJHREREZEWYFJHREREpAWY1BERERFpASZ1RERERFqASR0RERGRFpAIgiDUdBBEFZWRkQETExOkp6fD2Ni4psMhoifgZ5bo+eFMHREREZEWYFJHREREpAWY1BERERFpASZ1RERERFqASR0RERGRFmBSR0RERKQFmNQRERERaQEmdURERERagEkdERERkRZgUkdERESkBZjUEREREWkBJnVEREREWoBJHREREZEWYFJHREREpAWY1BERERFpAZ2aDoDoaUybMh8KhV5Nh0FVqE0rh5oOgSqhc6/BNR0CEf0HZ+qIiIiItACTOiIiIiItwKSOiIiISAswqSMiIiLSAkzqiIiIiLQAkzoiIiIiLcCkjoiIiEgLMKkjIiIi0gJM6oiIiIi0AJM6IiIiIi3ApI6IiIhICzCpIyIiItICTOqIiIiItIBOTQdAL7+wsDCkpaVh48aNNR3KCyc4pCWCQ1qWWvZhxJfQ0dVFx06t4OxSC2ZmxpArdJGWmol/4i5g356jyM/Lr1A/coUc7743HBaWZgCADb/uwJHDcVU1DPpXw4BWsKvlBAC4nRiPuNgYONR2hZ9/8zL3+Xv/Tjy4l1Juu2YW1nCrWw+mZhaQynSQl5uD5KSbOP/P8SqNn4i0G5M6oucgKysbD+6naWwTBMDQQA+t2vijoKAQd1Puw9jECFbW5mjfsQUcHG2w8vvfKtR+917BYkJH1aOWk6uY0D0uPz8XqQ/uaWzT1zeAnr4BACAvN7fcdm0daqNBk5aQSqXIz8tFVmY6dHXlsLa1Z1JHRJXCpI5qnCAIUKlU0NHR3h/Hi+eu4teorSW2FxSqsPnPPThy+BTy8/KhoyPDG2MGwMnZAXW96kBfX4GcnLxy267v54nGTXxw6uQF+DWsW11DeKUZGCrh7euP1Pt3oadvAH0DQ7Hs7p3buHvntkb9Vu07Q0/fAHeTk/AwK6PMdmUyGXwaBEAqleLqxbO4dO4UBEEoKtPizwMRVQ+eU6dFMjMzMXjwYBgaGsLOzg5ff/01goKCMH78eABAXl4eIiIi4ODgAENDQzRt2hR79+4V91+5ciVMTU2xfft2eHl5QalUIjQ0FElJSWIdlUqFiRMnwtTUFBYWFpg0aZL4S6iYWq3G7Nmz4eLiAn19ffj5+WH9+vVi+d69eyGRSLB161Y0btwYCoUCBw8erNZjU9N8fD3xyRcT8OH0MRg+sjfsHawBAFmZD3Fgb6y4zFpYqMLNxDsAio6jSi2U2SYAmJgaoVffENxMvIMdWw9U7yBeURKJBH5NWkKAgLjYmBI/7/9laWMHY5OiWdPrl8+VW9fC2g5yhQIAoNDTQ7tOPRHcpQ8aNw+EQqFXNQMgolcGkzotMnHiRMTExODPP//Ezp07ceDAAZw4cUIsDw8Px+HDhxEVFYV//vkHffv2RWhoKC5fvizWyc7OxldffYU1a9Zg//79SEhIQEREhFg+d+5crFy5EitWrMDBgwfx4MEDbNiwQSOO2bNnY/Xq1ViyZAnOnj2LCRMmYMiQIdi3b59GvcmTJ+Pzzz/H+fPn4evrW01HpeapVGpkZj5E6oMMGBsr4VWvDt4eN0RM7B5nqDSAj68HAODUyQvlnlMnkQD9B3WBVCZF1I9/Qa1SVdsYXmVuXvVhZm6Js3GxyMl++MT6ru7eAICMtAe4l3Kn3LpKpbH4tUNtV+Tn5UEmk8HGrhaatukAHR3dZwueiF4pnN/XEpmZmVi1ahV+/vlntG/fHgAQGRkJe3t7AEBCQgIiIyORkJAgbouIiMC2bdsQGRmJzz77DABQUFCAJUuWoE6dOgCKEsGZM2eK/cyfPx9TpkxBr169AABLlizB9u3bxfK8vDx89tln2LVrF5o3Lzp53NXVFQcPHsTSpUsRGBgo1p05cyY6dOhQ7rjy8vKQl/do+TEjo+ylrBdR3IlziDlwHDnZRedVuXs6Y+Rb/aCrq4PmLRvht3XbxLrmFqZ4/Y0+MDExQvy1m9iwfke5bbds7Q9Xt9pYv3Yr7t1NhZmZcbn1qfJMTM1Rx6MebiVcx+3E+CfWNzYxg6W1LQDg2uXzT6wvkUrEry+d+wdXL56BmYUVmgd2hL6+AWzsHXEr4dpTx09ErxYmdVri2rVrKCgoQEBAgLjNxMQEnp6eAIDTp09DpVLBw8NDY7+8vDxYWFiI7w0MDMSEDgDs7OyQklJ05V56ejqSkpLQtGlTsVxHRwf+/v7iktSVK1eQnZ1dIlnLz89Hw4YNNbb5+/s/cVyzZ8/GjBkznljvRXXvbqrG+8sX4/EwKxuGSgOYmBmJ22s72WPYyF5QKg1w7swV/LLmTxQUFJbbtp29FQCgW4/26NajPSSPlXXt0Q6N/Ovhu29+qrKxvIqUxqaQSqWwdXCEjX0/AIBMVvTfpq1DbXS0c8DuLRtQWFgAAHBx9wIA5GQ/RNLNG09sPzcnW/w6PfW+xr8AYGBoWGIfIqKyMKl7RWRlZUEmk+H48eOQyWQaZUqlUvxaV1dzuUcikTzxHKL/9gMAmzdvhoODg0aZ4t9zh4oZVuAX1pQpUzBx4kTxfUZGBhwdHSscT00LbBeAuBPnkZ6WCQBw83CCobLoqsjUB0Wzjj6+Hug/uAt0dXURc+A4Nm2Mxn8PubGJEqNG9wcAbN+yH2dPP1oyVyjkJfrV1dWBrpwf76pSnMg9TiqVQiqVojib1tM3EK+Ojb9yocTnxsTMQrz1yaljh5Geeh/37yZDENSQSKQwMTPHvZQkmJg++iPrYVZmNY2IiLQR/9fXEq6urtDV1UVsbCxq164NoGhm7dKlS2jTpg0aNmwIlUqFlJQUtG7d+qn6MDExgZ2dHY4cOYI2bdoAAAoLC3H8+HE0atQIAODt7Q2FQoGEhASNpdanpVAoSiSDL5NmLRoipHMg0tMykJ9fACvrol/YeXn5iNl/DEbGSgwa1h1SqQSFhYVwrG2Ht8cNEfffuH4nbt9KhkwqhbVN0b56ekXH49eorRpX1JqZGeODj0cD4H3qqsqthGsllj+DQrrDwFAp3qeumItbXUilUhTk5yMh/kqJtmQyGZRGJuLXQNFMXfzVS3BxqwsPbz/Y13KGvmHRH1mZGWm4cyuhuoZGRFqISZ2WMDIywvDhw/H+++/D3Nwc1tbWmDZtGqRSKSQSCTw8PDB48GAMGzYMc+fORcOGDXH37l1ER0fD19cXXbp0qVA/7777Lj7//HO4u7ujbt26mDdvHtLS0jTiiIiIwIQJE6BWq9GqVSukp6cjJiYGxsbGGD58eDUdgRfTnl1/o76fJ2xsLWFuYYi01HTciL+F6B2Hce/uA5iZGUP673lVOjo6qO1kr7G/nl7JWTh68ejo6KKWsxsAICH+ClSF5S+dP+78P8eRl5sDR+c6MFAaIS83Gzdv3Malc/9ArVZXV8hEpIWY1GmRefPmYfTo0ejatSuMjY0xadIkJCYmQk+v6NYIkZGR+PTTT/Hee+/h1q1bsLS0RLNmzdC1a9cK9/Hee+8hKSkJw4cPh1QqxYgRI9CzZ0+kp6eLdT755BNYWVlh9uzZuHbtGkxNTdGoUSN8+OGHVT7mF93Rv0/h6N+nyixPTc3A5IlznthORepVtC16Nnu3/1FiW2FhAXb+ta7c/R7cS8GW30s/x/HapXO4dqn8258QET2JRKjMCVP0Unn48CEcHBwwd+5cjBw5sqbDqRIZGRkwMTHB+DEzeB8vLdOmlcOTK9ELo3OvwRWqV/yZTU9Ph7Exr9Amqk6cqdMiJ0+exIULFxAQEID09HTxViTdu3ev4ciIiIioujGp0zJfffUVLl68CLlcjsaNG+PAgQOwtLSs6bCIiIiomjGp0yINGzbE8eN8ADgREdGriI8JIyIiItICTOqIiIiItACTOiIiIiItwKSOiIiISAswqSMiIiLSAkzqiIiIiLQAkzoiIiIiLcCkjoiIiEgLMKkjIiIi0gJM6oiIiIi0gEQQBKGmgyCqqIyMDJiYmCA9PR3GxsY1HQ4RPQE/s0TPD2fqiIiIiLQAkzoiIiIiLcCkjoiIiEgLMKkjIiIi0gJM6oiIiIi0AJM6IiIiIi2gU9MBEFVG8R14MjIyajgSIqqI4s8q755FVP2Y1NFLJTMzEwDg6OhYw5EQUWVkZmbCxMSkpsMg0mq8+TC9VNRqNW7fvg0jIyNIJJKaDoeInkAQBGRmZsLe3h5SKc/4IapOTOqIiIiItAD/bCIiIiLSAkzqiIiIiLQAkzoiIiIiLcCkjoiIiEgLMKkjIiIi0gJM6oiIiIi0AJM6IiIiIi3ApI6IiIhICzCpIyIiItICTOqIiIiItACTOiIiIiItwKSOiIiISAswqSOiauHs7IyuXbvWdBhERK8MJnVEBAC4evUq3nrrLbi6ukJPTw/GxsZo2bIlFixYgJycnJoOr1znzp3D9OnTER8fX6H60dHRGDFiBDw8PGBgYABXV1eMGjUKSUlJ1RsoEVE10qnpAIio5m3evBl9+/aFQqHAsGHD4OPjg/z8fBw8eBDvv/8+zp49i2XLltV0mGU6d+4cZsyYgaCgIDg7Oz+x/gcffIAHDx6gb9++cHd3x7Vr17Bo0SJs2rQJcXFxsLW1rf6giYiqGJM6olfc9evXMWDAADg5OWH37t2ws7MTy9555x1cuXIFmzdvrsEIy5abmwu5XF7p/ebNm4dWrVpBKn20WBEaGorAwEAsWrQIn376aVWGSUT0XHD5legVN2fOHGRlZWH58uUaCV0xNzc3vPvuu+L7wsJCfPLJJ6hTpw4UCgWcnZ3x4YcfIi8vr9T2Dx48iICAAOjp6cHV1RWrV68uUefatWvo27cvzM3NYWBggGbNmpVIJPfu3QuJRIKoqChMnToVDg4OMDAwwMKFC9G3b18AQNu2bSGRSCCRSLB3794yx9ymTRuNhK54m7m5Oc6fP1/mfkRELzKJIAhCTQdBRDWnVq1aUCgUuHr1aoXqh4WFYdWqVejTpw/atm2LI0eOYPXq1ejRowc2bNgg1nN2doaenh7S0tIwcuRI2NvbY8WKFTh58iROnz6NevXqAQCSk5Ph5+eH7OxsjBs3DhYWFli1ahVOnz6N9evXo2fPngCKkrq2bdvC29sbcrkcw4YNQ15eHl577TUsXboUCxcuxIcffggvLy8AQIcOHWBjY1Ph45CVlQULCwuEhYVh6dKlFd6PiOiFIRDRKys9PV0AIHTv3r1C9ePi4gQAwqhRozS2R0RECACE3bt3i9ucnJwEAML+/fvFbSkpKYJCoRDee+89cdv48eMFAMKBAwfEbZmZmYKLi4vg7OwsqFQqQRAEYc+ePQIAwdXVVcjOztbo/9dffxUACHv27Kno0Ev45JNPBABCdHT0U7dBRFSTuPxK9ArLyMgAABgZGVWo/pYtWwAAEydO1Nj+3nvvAUCJJVNvb2+0bt1afG9lZQVPT09cu3ZNo82AgAC0atVK3KZUKvHmm28iPj4e586d02hz+PDh0NfXr1C8FbV//37MmDED/fr1Q7t27aq0bSKi54VJHdErzNjYGACQmZlZofo3btyAVCqFm5ubxnZbW1uYmprixo0bGttr165dog0zMzOkpqZqtOnp6VmiXvEy6n/bdHFxqVCsFXXhwgX07NkTPj4++OGHH6q0bSKi54lJHdErzNjYGPb29jhz5kyl9pNIJBWqJ5PJSt0uPMOpvFU5S5eYmIiOHTvCxMQEW7ZsqfCMJRHRi4hJHdErrmvXrrh69SoOHz78xLpOTk5Qq9W4fPmyxvbk5GSkpaXBycmp0v07OTnh4sWLJbZfuHBBLH+SiiaZj7t//z46duyIvLw8bN++vdQrf4mIXiZM6ohecZMmTYKhoSFGjRqF5OTkEuVXr17FggULAACdO3cGAMyfP1+jzrx58wAAXbp0qXT/nTt3xtGjRzWSyocPH2LZsmVwdnaGt7f3E9swNDQEAKSlpVWoz4cPH6Jz5864desWtmzZAnd390rHTUT0ouHNh4lecXXq1MHPP/+M/v37w8vLS+OJEocOHcKvv/6KsLAwAICfnx+GDx+OZcuWIS0tDYGBgTh69ChWrVqFHj16oG3btpXuf/Lkyfjll1/QqVMnjBs3Dubm5li1ahWuX7+O3377rcT95ErToEEDyGQyfPHFF0hPT4dCoUC7du1gbW1dav3Bgwfj6NGjGDFiBM6fP69xbzqlUokePXpUehxERDWupi+/JaIXw6VLl4Q33nhDcHZ2FuRyuWBkZCS0bNlS+Oabb4Tc3FyxXkFBgTBjxgzBxcVF0NXVFRwdHYUpU6Zo1BGEoluadOnSpUQ/gYGBQmBgoMa2q1evCn369BFMTU0FPT09ISAgQNi0aZNGneJbmvz666+lxv/9998Lrq6ugkwme+LtTYpvt1Lay8nJqfwDRUT0guLNh4mIiIi0AM+pIyIiItICTOqIiIiItACTOiIiIiItwKSOiIiISAswqSMiIiLSAkzqiIiIiLQAkzoiKuHo0aOQy+W4ceNGTYfyXBQUFMDR0RGLFy+u6VCIiJ4akzoiKuGjjz7CwIEDn+pZrk9r7969kEgkpb7+/vvvEvUPHTqEVq1awcDAALa2thg3bhyysrJK1MvLy8MHH3wAe3t76Ovro2nTpti5c6dGHV1dXUycOBGzZs1Cbm5utY2RiKg68TFhRKQhLi4Ou3btwqFDh2qk/3HjxqFJkyYa29zc3DTex8XFoX379vDy8sK8efNw8+ZNfPXVV7h8+TK2bt2qUTcsLAzr16/H+PHj4e7ujpUrV6Jz587Ys2cPWrVqJdZ7/fXXMXnyZPz8888YMWJE9Q2QiKia8IkSRKTh3XffxcaNGxEfHw+JRPLc+t27dy/atm2LX3/9FX369Cm3bufOnREXF4cLFy7A2NgYAPDDDz/gjTfewPbt29GxY0cARcvITZs2xZdffomIiAgAQG5uLnx8fGBtbV0ice3WrRvS09Oxf//+ahghEVH14vIrEWnYuHEj2rVrVyKhc3Z2RteuXXHw4EEEBARAT08Prq6uWL16dZXHkJmZicLCwlLLMjIysHPnTgwZMkRM6ABg2LBhUCqVWLdunbht/fr1kMlkePPNN8Vtenp6GDlyJA4fPozExESNtjt06ICDBw/iwYMHVTwiIqLqx6SOiES3bt1CQkICGjVqVGr5lStX0KdPH3To0AFz586FmZkZwsLCcPbsWbGOWq3GvXv3KvQqKCgo0cfrr78OY2Nj6OnpoW3btjh27JhG+enTp1FYWAh/f3+N7XK5HA0aNMDJkyfFbSdPnoSHh4dG8gcAAQEBAIqWcR/XuHFjCIJQY0vPRETPgufUEZHowoULAAAXF5dSyy9evIj9+/ejdevWAIB+/frB0dERkZGR+OqrrwAACQkJZe7/X3v27EFQUBCAoqSsd+/e6Ny5MywtLXHu3Dl89dVXaN26NQ4dOoSGDRsCAJKSkgAAdnZ2Jdqzs7PDgQMHxPdJSUll1gOA27dva2x3dXUFAJw7dw5du3at0BiIiF4UTOqISHT//n0AgJmZWanl3t7eYkIHAFZWVvD09MS1a9fEbba2tiWuLi2Ln5+f+HWLFi3QokUL8f1rr72GPn36wNfXF1OmTMG2bdsAADk5OQAAhUJRoj09PT2xvLhuWfUeb6tY8bjv3btXofiJiF4kTOqIqISyrp+qXbt2iW1mZmZITU0V3+vp6SE4OLhK4nBzc0P37t3x+++/Q6VSQSaTQV9fH0DRrUr+Kzc3VywHAH19/TLrFZc/rnjcz/MCESKiqsKkjohEFhYWAKCRpD1OJpOVuv3xJFClUuHu3bsV6s/c3BxyubzcOo6OjsjPz8fDhw9hbGwsLp0WL8M+LikpCfb29uJ7Ozs73Lp1q9R6ADTqAo/GbWlpWaH4iYheJLxQgohEdevWBQBcv379qdtITEyEnZ1dhV4VuSDh2rVr0NPTg1KpBAD4+PhAR0enxAUU+fn5iIuLQ4MGDcRtDRo0wKVLl5CRkaFR98iRI2L544rH7eXlVdlhExHVOM7UEZHIwcEBjo6OJRKmynjac+ru3r0LKysrjfJTp07hzz//RKdOnSCVFv0NamJiguDgYPz444/4+OOPYWRkBABYs2YNsrKy0LdvX3H/Pn364KuvvsKyZcvE+9Tl5eUhMjISTZs2haOjo0Z/x48fh0QiQfPmzSs/cCKiGsakjog0dO/eHRs2bIAgCE91btnTnlPXv39/6Ovro0WLFrC2tsa5c+ewbNkyGBgY4PPPP9eoO2vWLLRo0QKBgYF48803cfPmTcydOxcdO3ZEaGioWK9p06bo27cvpkyZgpSUFLi5uWHVqlWIj4/H8uXLS8Swc+dOtGzZUlyGJiJ6mXD5lYg0jBgxArdu3UJMTMxz7bdHjx64d+8e5s2bhzFjxmDt2rXo1asXjh07VmI5tFGjRti1axf09fUxYcIELFu2DCNHjsT69etLtLt69WqMHz8ea9aswbhx41BQUIBNmzahTZs2GvXS09OxY8cOhIWFVecwiYiqDR8TRkQltG/fHvb29lizZk1Nh/LczJ8/H3PmzMHVq1dLXBVLRPQyYFJHRCUcOXIErVu3xuXLl+Hk5FTT4VS7goIC1KlTB5MnT8aYMWNqOhwioqfCpI6IiIhIC/CcOiIiIiItwKSOiIiISAswqSMiIiLSAkzqiIiIiLQAbz5MLxW1Wo3bt2/DyMiID10negkIgoDMzEzY29uLTwUhourBpI5eKrdv3y7xaCcievElJiaiVq1aNR0GkVZjUkcvleLnfCYmJsLY2LiGoyGiJ8nIyICjo6P42SWi6sOkjl4qxUuuxsbGTOqIXiI8XYKo+vEEByIiIiItwKSOiIiISAswqSMiIiLSAjynjoiIapxKpUJBQUFNh0H0wtHV1YVMJqtQXSZ1RERUYwRBwJ07d5CWllbToRC9sExNTWFra/vEC46Y1NFLKXfKfMgVejUdBtErS2/epCpppzihs7a2hoGBAa+SJXqMIAjIzs5GSkoKAMDOzq7c+kzqiIioRqhUKjGhs7CwqOlwiF5I+vr6AICUlBRYW1uXuxTLCyWIiKhGFJ9DZ2BgUMOREL3Yij8jTzrvlEkdERHVKC65EpWvop8RJnVEREREWoBJHREREZEW4IUSRET0wpk8cc5z6+vzp7yS99tvv8WXX36JO3fuwM/PD9988w0CAgLKrP/rr7/i448/Rnx8PNzd3fHFF1+gc+fOTxs2UQmcqSMiIqqktWvXYuLEiZg2bRpOnDgBPz8/hISEiLee+K9Dhw5h4MCBGDlyJE6ePIkePXqgR48eOHPmzHOOnLQZkzoiIqJKmjdvHt544w28/vrr8Pb2xpIlS2BgYIAVK1aUWn/BggUIDQ3F+++/Dy8vL3zyySdo1KgRFi1a9JwjJ23GpI6IiKgS8vPzcfz4cQQHB4vbpFIpgoODcfjw4VL3OXz4sEZ9AAgJCSmzPtHTYFJHRERUCffu3YNKpYKNjY3GdhsbG9y5c6fUfe7cuVOp+kRPg0kdERERkRZgUkdERFQJlpaWkMlkSE5O1tienJwMW1vbUvextbWtVH2ip8GkjoiIqBLkcjkaN26M6OhocZtarUZ0dDSaN29e6j7NmzfXqA8AO3fuLLM+0dPgfeqIiIgqaeLEiRg+fDj8/f0REBCA+fPn4+HDh3j99dcBAMOGDYODgwNmz54NAHj33XcRGBiIuXPnokuXLoiKisKxY8ewbNmymhwGaRkmdUT03Oh0bg2d4OYo3B6Dwu0xz6VPiaUp5JNHQUjNQP7nPwAq9XPpl7Rb//79cffuXfzvf//DnTt30KBBA2zbtk28GCIhIQFS6aPFsBYtWuDnn3/G1KlT8eGHH8Ld3R0bN26Ej49PTQ2BtJBEEAShpoMgqqiMjAyYmJggecwMGCv0ajqcl57usNcga1AXAKA6eR4Fa/4qs67EzBiKj0eXWf7ERM1QH4qpbwEyKfJmfAc8zAEA6HRpA6mPOyQmSkAmAzIfQnX5Bgp3HAJSM8puTyGHTqdWkLrUgsTMGFDoQkjNhDruAgr3HAXy8h+NM6w7ZL6eKPh1B1SH48pukypMr4JPYSj+zKanp8PY2FijLDc3F9evX4eLiwv09Ph5JipLRT8rz3RO3d69eyGRSJCWlvYszTyTlStXwtTUtMraCwoKwvjx46usvacxffp0NGjQoMbbCQsLQ48ePZ45DnoxyZr4iAldRQiFKqhv3NZ8Jd9/VJ6RVX5//vUgUcihvhgvJnQAIPV0gUSuC+FuKoS0TEjMTaDT1BfyN/uW257EQA86bfwhsbWEkJ4J5BVAam0OnY4toDusm0Zd1YnzRTG0bFDh8RIRvWwqtfwaFBSEBg0aYP78+QCKppOTkpJgYmJSHbFVSP/+/bXu2XkREREYO3as+D4sLAxpaWnYuHFjzQVFWkViYQqdnsFQX78FialR0UzXk2Q+RP6CHzU26fQKhtTGAkJ2DlTHz5W7u6yhFwBAffaqxvb8hT8ChSrxve6gLpD514PUxgIw0AOyc0ttTyhUoeDPPVAdPlU0K6cjg3zMAEidHSDzqoMCfQWQk1fU5/lrEFRqSO2tIbGxgPBYMkpEpC2e6Zw6uVxe45dj6+vrQ19fv0ZjqGpKpRJKpbKmwyBtJZVAd0hXQBBQ8NMmyMcMeLp2DPQga1J0PpAqJg7ILyi7rlwXEoeic43UiUmaZYUqyFo0KGrLQB9SK7OienfulZnQAShapt0bq9GOOvEOpM4OENRqQP3YmSUFhRCS70Fibw2pay2omNQRkRaq8PJrWFgY9u3bhwULFkAikUAikWDlypUay6/FS6GbNm2Cp6cnDAwM0KdPH2RnZ2PVqlVwdnaGmZkZxo0bB5Xq0V/meXl5iIiIgIODAwwNDdG0aVPs3bu3QnH9d/m1eMlxzZo1cHZ2homJCQYMGIDMzMyKDhVqtRqTJk2Cubk5bG1tMX36dI3yhIQEdO/eHUqlEsbGxujXr5/G/YdOnTqFtm3bwsjICMbGxmjcuDGOHTumEe/GjRvh7u4OPT09hISEIDExscQYir9etWoV/vjjD/G4Fx+bDz74AB4eHjAwMICrqys+/vhjFBSU84u1HCqVChMnToSpqSksLCwwadIk/Pd0y23btqFVq1Zina5du+Lq1UezLu3atUN4eLjGPnfv3oVcLhcv5V+8eLE4bhsbG/Tp0+ep4qWnp9OxJaRO9ij4bSeEB+lP3Y6sZUNIFHIIBYUoPHii3LoScxNIZEX/3ZTWp8TMGFIn+0cJ3c1k5C9ZV7mAlAaQ+XoU7X/ygsY5dUX9Fp2fJ7E0q1y7REQviQondQsWLEDz5s3xxhtvICkpCUlJSXB0dCxRLzs7GwsXLkRUVBS2bduGvXv3omfPntiyZQu2bNmCNWvWYOnSpVi/fr24T3h4OA4fPoyoqCj8888/6Nu3L0JDQ3H58uWnGtTVq1exceNGbNq0CZs2bcK+ffvw+eefV3j/VatWwdDQEEeOHMGcOXMwc+ZM7Ny5E0BRwte9e3c8ePAA+/btw86dO3Ht2jX0799f3H/w4MGoVasWYmNjcfz4cUyePBm6uroax2jWrFlYvXo1YmJikJaWhgEDSp8tiYiIQL9+/RAaGioe9xYtWgAAjIyMsHLlSpw7dw4LFizA999/j6+//vppDhnmzp2LlStXYsWKFTh48CAePHiADRs2aNR5+PAhJk6ciGPHjiE6OhpSqRQ9e/aEWl10NeGoUaPw888/Iy8vT9znxx9/hIODA9q1a4djx45h3LhxmDlzJi5evIht27ahTZs25caVl5eHjIwMjRc9PUktW8jaN4Pq2FmoT5S/XFoumQw6LRsCQNGya+bD8uvrKR59/Z9kCwAKN+9H7ntfIm/2D1BdvgFpLZui2USJpELhSCxMIQ8fBImJEdTXbqJg/Y6SlYp/Lh+PhYhIi1R4+dXExARyuRwGBgbikuuFCxdK1CsoKMB3332HOnXqAAD69OmDNWvWIDk5GUqlEt7e3mjbti327NmD/v37IyEhAZGRkUhISIC9vT2AokRm27ZtiIyMxGeffVbpQanVaqxcuRJGRkYAgKFDhyI6OhqzZs2q0P6+vr6YNm0aAMDd3R2LFi1CdHQ0OnTogOjoaJw+fRrXr18Xk9rVq1ejXr16iI2NRZMmTZCQkID3338fdevWFdv47zFatGgRmjZtCqAoifTy8sLRo0cREBCgUVepVEJfXx95eXkllrqnTp0qfu3s7IyIiAhERUVh0qSKXZX2uPnz52PKlCno1asXAGDJkiXYvn27Rp3evXtrvF+xYgWsrKxw7tw5+Pj4oFevXggPD8cff/yBfv36ASiamQwLC4NEIkFCQgIMDQ3RtWtXGBkZwcnJCQ0bNiw3rtmzZ2PGjBmVHg+VTmpnCYlMCqmfBxT1//25/PcPDqmvBxSzxyNvxmIgt2Ti9ThZk3qQGCshqAWo9h59csePJfqQy4HcvJJ1BAHC3QdQ7T8OmbsTZG61oXJ3gvpSfLlNS5zsIR/ZCxKlAVRnrqBgzZ9AQWHJiop/k7nS+iYi0gJV/kQJAwMDMaEDih5Y7OzsrHGOmI2NDVJSUgAAp0+fhkqlgoeHh3gumVKpxL59+zSW9irD2dlZTOgAwM7OTuyvInx9fTXeP77/+fPn4ejoqDFL6e3tDVNTU5w/X3SF3cSJEzFq1CgEBwfj888/LzEOHR0dNGnSRHxft25djf0rau3atWjZsiVsbW2hVCoxdepUJCQkVKoNAEhPT0dSUpKYZBbH6O/vr1Hv8uXLGDhwIFxdXWFsbAxnZ2cAEPvU09PD0KFDsWLFCgDAiRMncObMGYSFhQEAOnToACcnJ7i6umLo0KH46aefkJ2dXW5sU6ZMQXp6uvh6fJmanp5EVxcShbzoJS2aDZPIZJAo5ACK3uuO7g/5ByOh06XkbKossOjnV33+KoSUB0/sT7ifDuHf+8NJzB9dlCGxNIO0nltxl4AEkNZ1ebSj4t8ZbhMl5B+MhPyDkZDWf/RHktTXA/Ix/SFRGqDwwHEURP5eekL3WL/CvdQnxktE9DKq8psPP77MCAASiaTUbcVLdllZWZDJZDh+/DhkMplGvae9WKC8/p7H/tOnT8egQYOwefNmbN26FdOmTUNUVBR69uxZ4Tae5PDhwxg8eDBmzJiBkJAQmJiYICoqCnPnzq2yPv6rW7ducHJywvfffw97e3uo1Wr4+PggP//RrM6oUaPQoEED3Lx5E5GRkWjXrh2cnJwAFC0XnzhxAnv37sWOHTvwv//9D9OnT0dsbGyZt6VRKBRQKLhcVlVUsWegij2jsU0x9S1IzE1K3KdOamkKibkJhBu3NepL67kVXZkKFN0PriLyCyDcSoakth2kjrZQ3b4LAJCYKCEf2QtCbj6EB2mQKA0gMS763AupGVBfulFUTyoV+5QUL58aK6E7rDskUgmEwkJIa9tBPm6I2GXB+p0Qbv17rquuDiQ2lgAA9bWbFYuZiOglU6mZOrlcrnGBQ1Vo2LAhVCoVUlJS4ObmpvGq6StrS+Pl5YXExESNGaNz584hLS0N3t7e4jYPDw9MmDABO3bsQK9evRAZGSmWFRYWihdOAMDFixeRlpYGLy+vUvss7bgfOnQITk5O+Oijj+Dv7w93d3fcuHHjqcZkYmICOzs7HDlyRCPG48ePi+/v37+PixcvYurUqWjfvj28vLyQmlpyxqN+/frw9/fH999/j59//hkjRozQKNfR0UFwcDDmzJmDf/75B/Hx8di9e/dTxU01Qyfo31m6G7chVCJBUp0smomWeruJ24TUDKhOXwJyciGxMgcM9KC+l4rCQ3HIW/hjqeffFZPIpI9mGXV0ii60eOwl0ZOLdaVerpDIpFDfTuHtTIhIa1Vqps7Z2RlHjhxBfHw8lEplpWavyuLh4YHBgwdj2LBhmDt3Lho2bIi7d+8iOjoavr6+6NKlyzP3UZWCg4NRv359DB48GPPnz0dhYSHGjBmDwMBA+Pv7IycnB++//z769OkDFxcX3Lx5E7GxsRrno+nq6mLs2LFYuHAhdHR0EB4ejmbNmpU4n66Ys7Mztm/fjosXL8LCwgImJiZwd3dHQkICoqKi0KRJE2zevLnEhQ2V8e677+Lzzz+Hu7s76tati3nz5mncVNrMzAwWFhZYtmwZ7OzskJCQgMmTJ5fa1qhRoxAeHg5DQ0ON2clNmzbh2rVraNOmDczMzLBlyxao1Wp4eno+ddz07PI+XVqp7fnf/vJU/aiOnYVOaKui5VVDfeBhDoQH6SiI3PjEfYXUDOROnPPEbWWRNSr6g0sVc7LScRMRvSwqldRFRERg+PDh8Pb2Rk5Ojsbs07OIjIzEp59+ivfeew+3bt2CpaUlmjVrhq5du1ZJ+1VJIpHgjz/+wNixY9GmTRtIpVKEhobim2++AQDIZDLcv38fw4YNQ3JyMiwtLdGrVy+Nk/0NDAzwwQcfYNCgQbh16xZat26N5cuXl9nnG2+8gb1798Lf3x9ZWVnYs2cPXnvtNUyYMAHh4eHIy8tDly5d8PHHH5e4/UpFvffee0hKSsLw4cMhlUoxYsQI9OzZE+npRbefkEqliIqKwrhx4+Dj4wNPT08sXLgQQUFBJdoaOHAgxo8fj4EDB2o8zsTU1BS///47pk+fjtzcXLi7u+OXX35BvXr1nipmesk8zIHqwHHoBDeHTqtGz/XZr1IfN6jvp0F15PRz6ZOe3Zbff3pufXXuNbjS++zfvx9ffvkljh8/jqSkJGzYsOGJT+DZu3cvJk6ciLNnz8LR0RFTp04Vzzkmqgp89utztnLlSowfP75GH61W3eLj41GnTh3ExsaiUaNGVdo2n/1K9GKo7me/vuhJ3datWxETE4PGjRujV69eT0zqrl+/Dh8fH4wePRqjRo1CdHQ0xo8fj82bNyMkJOQZoqdXQUWf/VrlF0rQq6ugoAD379/H1KlT0axZsypP6IiIXhSdOnVCp06dKlx/yZIlcHFxES9m8/LywsGDB/H1118zqaMqU+W3NKlqnTp10rjVyeOvyt7DLiEhocy2lErlU90O5GVR3rgPHDhQJX3ExMTAzs4OsbGxWLJkSZW0SUSkDQ4fPozg4GCNbSEhITh8+HANRUTa6IWfqfvhhx+Qk5NTapm5uXml2rK3t0dcXFy55dUtLCysRs6hKG/cDg4OVdJHUFBQiUeLERERcOfOHdjY2Ghss7GxQUZGBnJycrTuGeZUM174pK6qEg6g6HYabm5uT66ohV7VcRMREb0qXvjlVyIiopedra0tkpOTNbYlJyfD2NiYs3RUZZjUERERVbPmzZsjOjpaY9vOnTvRvHnzGoqItBGTOiIiokrKyspCXFyceL7y9evXERcXJ15wN2XKFAwbNkysP3r0aFy7dg2TJk3ChQsXsHjxYqxbtw4TJkyoifBJSzGpIyIiqqRjx46hYcOGaNiwIQBg4sSJaNiwIf73v/8BAJKSkjTuqODi4oLNmzdj586d8PPzw9y5c/HDDz/wdiZUpV74CyWIiOjV8zQ3BH6ennS1/8qVK0vd5+RJPqqOqg9n6oiIiIi0AJM6IiIiIi3ApI6IiIhICzCpIyIiItICTOqIiIiItACTOiIiIiItwKSOiIiISAvwPnX0UtKbPR56xsY1HQYREdELgzN1RERERFqASR0REVEVEwQBb775JszNzSGRSMRnxJZGIpFg48aNzy22F5WzszPmz59f02G81Lj8SkREL5zhE+c8t75WzZv01PsePnwYrVq1QmhoKDZv3ixu37ZtG1auXIm9e/fC1dUVlpaWZbaRlJQEMzOzp47hWbz11lv44YcfEBUVhb59+9ZIDMViY2NhaGhYozG87DhTR0RE9JSWL1+OsWPHYv/+/bh9+7a4/erVq7Czs0OLFi1ga2sLHZ2Scyj5+fkAAFtbWygUiucWc7Hs7GxERUVh0qRJWLFixXPvv1jxcbCysoKBgUGNxaENmNQRERE9haysLKxduxZvv/02unTpgpUrVwIAwsLCMHbsWCQkJEAikcDZ2RkAEBQUhPDwcIwfPx6WlpYICQkBUHL59ebNmxg4cCDMzc1haGgIf39/HDlyBEBRsti9e3fY2NhAqVSiSZMm2LVrl0Zczs7O+OyzzzBixAgYGRmhdu3aWLZsWYn4f/31V3h7e2Py5MnYv38/EhMTNcrDwsLQo0cPfPbZZ7CxsYGpqSlmzpyJwsJCvP/++zA3N0etWrUQGRmpsV9iYiL69esHU1NTmJubo3v37oiPjy/R7qxZs2Bvbw9PT08x7seXX9PS0vDWW2/BxsYGenp68PHxwaZNmwAA9+/fx8CBA+Hg4AADAwPUr18fv/zyi0YcQUFBGDduHCZNmgRzc3PY2tpi+vTpZX9DtQCTOiIioqewbt061K1bF56enhgyZAhWrFgBQRCwYMECzJw5E7Vq1UJSUhJiY2PFfVatWgW5XI6YmBgsWbKkRJtZWVkIDAzErVu38Oeff+LUqVOYNGkS1Gq1WN65c2dER0fj5MmTCA0NRbdu3ZCQkKDRzty5c+Hv74+TJ09izJgxePvtt3Hx4kWNOsuXL8eQIUNgYmKCTp06iUnp43bv3o3bt29j//79mDdvHqZNm4auXbvCzMwMR44cwejRo/HWW2/h5s2bAICCggKEhITAyMgIBw4cQExMDJRKJUJDQ8UZOQCIjo7GxYsXsXPnTjFRe5xarUanTp0QExODH3/8EefOncPnn38OmUwGAMjNzUXjxo2xefNmnDlzBm+++SaGDh2Ko0eParSzatUqGBoa4siRI5gzZw5mzpyJnTt3lvdtfanxnDoiIqKnUJwUAUBoaCjS09Oxb98+BAUFwcjICDKZDLa2thr7uLu7Y86css8X/Pnnn3H37l3ExsbC3NwcAODm5iaW+/n5wc/PT3z/ySefYMOGDfjzzz8RHh4ubu/cuTPGjBkDAPjggw/w9ddfY8+ePeKs2OXLl/H333/j999/BwAMGTIEEydOxNSpUyGRSMR2zM3NsXDhQkilUnh6emLOnDnIzs7Ghx9+CACYMmUKPv/8cxw8eBADBgzA2rVroVar8cMPP4jtREZGwtTUFHv37kXHjh0BAIaGhvjhhx8gl8tLPQ67du3C0aNHcf78eXh4eAAAXF1dxXIHBwdERESI78eOHYvt27dj3bp1CAgIELf7+vpi2rRp4rFftGgRoqOj0aFDhzK/By8zztQRERFV0sWLF3H06FEMHDgQAKCjo4P+/ftj+fLl5e7XuHHjcsvj4uLQsGFDMaH7r6ysLERERMDLywumpqZQKpU4f/58iZk6X19f8WuJRAJbW1ukpKSI21asWIGQkBDxAo7OnTsjPT0du3fv1minXr16kEofpQo2NjaoX7+++F4mk8HCwkJs+9SpU7hy5QqMjIygVCqhVCphbm6O3NxcXL16Vdyvfv36ZSZ0xcehVq1aYkL3XyqVCp988gnq168Pc3NzKJVKbN++vdzjAAB2dnYax0HbcKaOiIiokpYvX47CwkLY29uL2wRBgEKhwKJFi8rc70lXd+rr65dbHhERgZ07d+Krr76Cm5sb9PX10adPH42lTQDQ1dXVeC+RSMQlXJVKhVWrVuHOnTsaF3CoVCqsWLEC7du3L7ed8trOyspC48aN8dNPP5WI3crKSvz6WY/Dl19+iQULFmD+/PmoX78+DA0NMX78+EodB23EpI6IiKgSCgsLsXr1asydO1dcTizWo0ePEifsV4avry9++OEHPHjwoNTZupiYGISFhaFnz54AipKoxy9CqIgtW7YgMzMTJ0+eFM9RA4AzZ87g9ddfR1paGkxNTZ8q/kaNGmHt2rWwtraG8TM89cfX1xc3b97EpUuXSp2ti4mJQffu3cXlb7VajUuXLsHb2/up+9QGXH4lIiKqhE2bNiE1NRUjR46Ej4+Pxqt3795PXIItz8CBA2Fra4sePXogJiYG165dw2+//YbDhw8DKDov7Pfff0dcXBxOnTqFQYMGVXrmafny5ejSpQv8/Pw0Yi++YrW0WbaKGjx4MCwtLdG9e3ccOHAA169fx969ezFu3DjxYoqKCAwMRJs2bdC7d2/s3LkT169fx9atW7Ft2zYARcdh586dOHToEM6fP4+33noLycnJTx23tmBSR0REVAnLly9HcHAwTExMSpT17t0bx44dQ0ZGxlO1LZfLsWPHDlhbW6Nz586oX7++xlWf8+bNg5mZGVq0aIFu3bohJCQEjRo1qnD7ycnJ2Lx5M3r37l2iTCqVomfPns+UlBoYGGD//v2oXbs2evXqBS8vL4wcORK5ubmVnrn77bff0KRJEwwcOBDe3t6YNGkSVCoVAGDq1Klo1KgRQkJCEBQUJCbCrzqJIAhCTQdBVFEZGRkwMTFBenr6M03tE9HzUd5nNjc3F9evX4eLiwv09PRqKEKiF19FPys8p45eSm9NmQ+5Qnt/CWz1icKnFp1qOgx6RenmdQEAfHvKr8w6x2bxcU5ELxouvxIRERFpASZ1RERERFqASR0RERGRFmBSR0RERKQFmNQRERERaQEmdURERERagEkdERERkRZgUkdERESkBZjUERERPUfx8fGQSCSIi4ur6VBIy/CJEkRE9MKxXlHx55k+q5QRJ55638OHD6NVq1YIDQ3F5s2bK7SPo6MjkpKSYGlp+dT9EpWGM3VERERPafny5Rg7diz279+P27dvV2gfmUwGW1tb6OhwXoWqFpM6IiKip5CVlYW1a9fi7bffRpcuXbBy5UqxLDU1FYMHD4aVlRX09fXh7u6OyMhIACWXX1UqFUaOHAkXFxfo6+vD09MTCxYs0OgrLCwMPXr0wFdffQU7OztYWFjgnXfeQUFBwfMaLr0E+GcCab3QwCZoUK8O7KzNYWigh/SMh7hwNREbt8fg7oN0AMDkMQPg5Va7xL6Xrt3ErEU/V6gfPYUcn7w3HNaWZgCAlb/uwJ7DcVU2jleJUt8Ugzq+X2b58QvROH5xN3R15PCvGwwXex/oKwzxMCcdlxLjcPLSXgiCutw+9BWGCPAKQW1bT8h19JCR/QDnrh/B2et/V/VwSEutW7cOdevWhaenJ4YMGYLx48djypQpkEgk+Pjjj3Hu3Dls3boVlpaWuHLlCnJyckptR61Wo1atWvj1119hYWGBQ4cO4c0334SdnR369esn1tuzZw/s7OywZ88eXLlyBf3790eDBg3wxhtvPK8h0wuOSR1pveDWjWBhaow7dx8gv6AQ1hamaGVuAh9PZ3ww+wfk5uWLdVPupSLj4aP/eG8l36twP0N7BYsJHT0blVqF5AcJGtsUuvowNbICAGTnZgKQIKTpUNhbukKlLkTmw1QYKy3gX7c9jA3NsffE+jLb15HpolvLN2BqZIXCwnxk5qTBzMgaLX27QU9hiOMXoqtzeKQlli9fjiFDhgAAQkNDkZ6ejn379iEoKAgJCQlo2LAh/P39AQDOzs5ltqOrq4sZM2aI711cXHD48GGsW7dOI6kzMzPDokWLIJPJULduXXTp0gXR0dFM6kjEpI603r6/TyHm2Fk8SMsEAAzq3g4hgf4wNVainocTjp++LNb9Y+dhHIw9U+k+Avw80aqJD46cvICmDetWWeyvqpy8TPxxYKnGtpb1u8HUyAq5+dm4fPMUnO28YG/pCgDYefRnJCRfRD2XZmjp2w0ejg1x+uoh3E8v/RwnL+cAmBpZQRDU2HhgCR5kJKNZvU7wdWuFBu5tcO7638jJe1jt46SX18WLF3H06FFs2LABAKCjo4P+/ftj+fLlCAoKwttvv43evXvjxIkT6NixI3r06IEWLVqU2d63336LFStWICEhATk5OcjPz0eDBg006tSrVw8ymUx8b2dnh9OnT1fL+OjlxHPqqELy8/OfXOkF9deuv8WEDgAuXksUvy4oVGnUHdS9HX6YMxFffvQmXu8bAmOlwRPbNzc1QljfEFxPvIPfth6ousBJpNDVh0ftoqshz8cfRaEqH47WHgCAwsJ8JCRfAgBcTzor7uNo7V5me8X7pj+8jwcZyUX73i7aVybVgb1lnaofBGmV5cuXo7CwEPb29tDR0YGOjg6+++47/Pbbb0hPT0enTp1w48YNTJgwAbdv30b79u0RERFRaltRUVGIiIjAyJEjsWPHDsTFxeH1118v8f+urq6uxnuJRAK1uvzTDOjVwqSOShUUFITw8HCMHz8elpaWCAkJwbx581C/fn0YGhrC0dERY8aMQVZWlsZ+MTExCAoKgoGBAczMzBASEoLU1FQAReeNzJ49WzwZ2M/PD+vXl71EVh0kEgmCmvsBKFpqPXfphliWl1+A1PRMZGblwNrCFEHN/fDxu0Mgl+uW1RwkEuDNQV0gk0nx3Y9/QaVSlVmXnp63SzPo6shRqCrAmWuHAQBKfRMAQG5BNgABAJCT++jnUalvWmZ7xfs+PhuXnffYvgZl70tUWFiI1atXY+7cuYiLixNfp06dgr29PX755RcAgJWVFYYPH44ff/wR8+fPx7Jly0ptLyYmBi1atMCYMWPQsGFDuLm54erVq89zSKQluPxKZVq1ahXefvttxMTEAAC2bt2KhQsXwsXFBdeuXcOYMWMwadIkLF68GAAQFxeH9u3bY8SIEViwYAF0dHSwZ88eMdGZPXs2fvzxRyxZsgTu7u7Yv38/hgwZAisrKwQGBpYaQ15eHvLy8sT3GRkZTz0euVwXY4Z0g29dV6RlZOHr5b+j8N/Yfv5jN27fuS++79O5NboFN4e1hSn867vj0PFzpbbZsbU/vNxqY/narUi+mwpLM+Onjo9KJ5XKUM+lKQDgys1TyMnLKruyRPLU/TzDrvSK2bRpE1JTUzFy5EiYmJholPXu3RvLly/H7du30bhxY9SrVw95eXnYtGkTvLy8Sm3P3d0dq1evxvbt2+Hi4oI1a9YgNjYWLi4uz2M4pEWY1FGZ3N3dMWfOHPG9p6en+LWzszM+/fRTjB49Wkzq5syZA39/f/E9UHQOCFCUnH322WfYtWsXmjdvDgBwdXXFwYMHsXTp0jKTutmzZ2ucQPy0TIwMMWFUb7g42iIp5QHmLvtVvPIVABJupWjUP3ziPLoFF8Vpblp2ouZoX3Ti/uAe7TG4R3s8nhcM6tEOLf3r4dNvfnrm+F9lHo4NYaBnBEFQ458rB8XtWTlF3z89uSEACQAB+grDx8rTymwzKycdpkZWGvX15cpH5dll70u0fPlyBAcHl0jogKKkbs6cOejWrRumTJmC+Ph46Ovro3Xr1oiKiiq1vbfeegsnT55E//79IZFIMHDgQIwZMwZbt26t7qGQlmFSR2Vq3Lixxvtdu3Zh9uzZuHDhAjIyMlBYWIjc3FxkZ2fDwMAAcXFx6Nu3b6ltXblyBdnZ2ejQoYPG9vz8fDRs2LDMGKZMmYKJEyeK7zMyMuDo6FipcTjYWGDCG31gZW6Ci1cTsSByAx5m54rlRkoDtGjsjX1//yNeCdu0waOLHe6lFiUPZiZKTBrdHwCwfst+jQss9BTyEv3KdXUgl/Mj9qx867QCACQkX0Ja1l1x+82Uy/BybgIdmS5q23ggIfkiXOzqieWJKUXfH2c7bwR4dQQAbDq0Atm5GUhMuYRa1m4wMbSAubENHmQkw8W+aF+VuhC373Hpq6Y9y1Meqttff/1VZllAQAAEoeh0gP/973+l1nF2dhbrAIBCoUBkZKR4H7tis2fPFr9+/B54xebPn1+JqOlVwN84VCZDw0ezGPHx8ejatSvefvttzJo1C+bm5jh48CBGjhyJ/Px8GBgYQF9fv8y2is+927x5MxwcHDTKFApFmfspFIpyyyti7Os9YWVe9Be1nkKOiW/0Ecv2//0Pzl6Kx6Du7dCvayBS7qVBIdeFxb/LqLfu3MPxf4pOwpdJpbC3sQAA6OsVxfRD1Fb8EPXor2lLM2PM/Xg0AN6nrio42dYVb2Ny6ormRSjxSeeQdD8edhbO6BAwCBkPH8BEWfTYpcuJceKVr3IdhdiGVFJ0GvH5+NiiK2CVlujRejSyctNhqiyq88+Vg7zylYheSkzqqEKOHz8OtVqNuXPnQiot+sW4bt06jTq+vr6Ijo4udbnU29sbCoUCCQkJZS61VhcdnUe3AHCqZaNRdvrCdWQ8zMGfOw/Bx9MF1ham0NXVwe3k+zhx5jI27z5S4gpZen6KZ+lSUhNx5368RpkAAdv+Xg3/usFwtfeBsaE5srLTcDkxDicu7Sm33UJVPv46+AMCvDuito0njAzMkJqZgvPxR8ULMYiIXjZM6qhC3NzcUFBQgG+++QbdunVDTEwMlixZolFnypQpqF+/PsaMGYPRo0dDLpdjz5496Nu3LywtLREREYEJEyZArVajVatWSE9PR0xMDIyNjTF8+PBqiz3i06VPrPPb1oP4bevBcuvcS83A8IlznrkOVdxfMT+UW15QmIfDZzbj8JmyH6R+KfEkLiWeLLE9Jy8T+07+9swxEhG9KHhLE6oQPz8/zJs3D1988QV8fHzw008/aZzvAQAeHh7YsWMHTp06hYCAADRv3hx//PGH+NDqTz75BB9//DFmz54NLy8vhIaGYvPmzbzCi4iIqApIhMfP1iR6wWVkZMDExAQDxsyAXKFX0+FUm60+UfjUolNNh0GvKN28LgCAb0/5lVnn2CzDMsseV/yZTU9Ph7Gx5pXkubm5uH79OlxcXKCnp72fZ6JnVdHPCmfqiIiIiLQAkzoiIiIiLcCkjoiIiEgLMKkjIiIi0gJM6oiIiF5Be/fuhUQiQVpaWk2HQlWE96kjIqIXzrI/Pnpufb3ZfdZT7Xfnzh3MmjULmzdvxq1bt2BtbY0GDRpg/PjxaN++fRVHWfVatGiBpKSkUp9hSy8nJnVERESVFB8fj5YtW8LU1BRffvkl6tevj4KCAmzfvh3vvPMOLly4UNMhPpFcLoetrW1Nh0FViMuvRERElTRmzBhIJBIcPXoUvXv3hoeHB+rVq4eJEyfi77//BgDMmzcP9evXh6GhIRwdHTFmzBjxOdgAMH36dDRo0ECj3fnz58PZ2Vlj24oVK1CvXj0oFArY2dkhPDxcLHtSHzdu3EC3bt1gZmYGQ0ND1KtXD1u2bAFQcvn1/v37GDhwIBwcHGBgYID69evjl19+0YglKCgI48aNw6RJk2Bubg5bW1tMnz79GY8mVRUmdURERJXw4MEDbNu2De+88w4MDUvehNnU1BQAIJVKsXDhQpw9exarVq3C7t27MWnSpEr19d133+Gdd97Bm2++idOnT+PPP/+Em5ubWP6kPt555x3k5eVh//79OH36NL744gsolcpS+8rNzUXjxo2xefNmnDlzBm+++SaGDh2Ko0ePatRbtWoVDA0NceTIEcyZMwczZ87Ezp07KzUuqh5cfiUiIqqEK1euQBAE1K1bt9x648ePF792dnbGp59+itGjR2Px4sUV7uvTTz/Fe++9h3fffVfc1qRJkwr3kZCQgN69e6N+/foAAFdX1zL7cnBwQEREhPh+7Nix2L59O9atW4eAgABxu6+vL6ZNmwYAcHd3x6JFixAdHY0OHTpUeFxUPZjUERERVUJFn665a9cuzJ49GxcuXEBGRgYKCwuRm5uL7OxsGBgYPHH/lJQU3L59u9yLLp7Ux7hx4/D2229jx44dCA4ORu/eveHr61tqWyqVCp999hnWrVuHW7duIT8/H3l5eSVi/e/+dnZ2SElJqcARoerG5VciIqJKcHd3h0QiKfdiiPj4eHTt2hW+vr747bffcPz4cXz77bcAgPz8fABFS6f/TRALCgrEr/X19cuNoyJ9jBo1CteuXcPQoUNx+vRp+Pv745tvvim1vS+//BILFizABx98gD179iAuLg4hISFiW8V0dXU13kskEqjV6nJjpeeDSR0REVElmJubIyQkBN9++y0ePnxYojwtLQ3Hjx+HWq3G3Llz0axZM3h4eOD27dsa9aysrHDnzh2NxC4uLk782sjICM7OzoiOji41jor0AQCOjo4YPXo0fv/9d7z33nv4/vvvS20vJiYG3bt3x5AhQ+Dn5wdXV1dcunSpIoeEXhBcfqWX0tLZ42FsbFzTYVSjyp1MTVQdXu9X0xG8uL799lu0bNkSAQEBmDlzJnx9fVFYWIidO3fiu+++Q1RUFAoKCvDNN9+gW7duiImJwZIlSzTaCAoKwt27dzFnzhz06dMH27Ztw9atWzX+b5s+fTpGjx4Na2trdOrUCZmZmYiJicHYsWPh5ub2xD7Gjx+PTp06wcPDA6mpqdizZw+8vLxKHZO7uzvWr1+PQ4cOwczMDPPmzUNycjK8vb2r/gBS9RCIXiLp6ekCACE9Pb2mQyGiCijvM5uTkyOcO3dOyMnJqYHInt3t27eFd955R3BychLkcrng4OAgvPbaa8KePXsEQRCEefPmCXZ2doK+vr4QEhIirF69WgAgpKamim189913gqOjo2BoaCgMGzZMmDVrluDk5KTRz5IlSwRPT09BV1dXsLOzE8aOHSuWPamP8PBwoU6dOoJCoRCsrKyEoUOHCvfu3RMEQRD27NmjUff+/ftC9+7dBaVSKVhbWwtTp04Vhg0bJnTv3l3sLzAwUHj33Xc14uvevbswfPjwKjiiVJaKflYkglDBMz6JXgAZGRkwMTFBenq6ls/UEWmH8j6zubm5uH79OlxcXKCnp1dDERK9+Cr6WeE5dURERERagEkdERERkRZgUkdERESkBZjUEREREWkBJnVEREREWoBJHREREZEWYFJHREREpAWY1BERERFpASZ1RERE1WDlypUwNTWt6TA0xMfHQyKRaDxjlrQHkzoiIqJKCgsLQ48ePUps37t3LyQSCdLS0tC/f39cunTp+QdXDkdHRyQlJcHHx6emQ6FqoFPTARAREf1X5LpDz62v1/u1qJZ29fX1oa+vXy1tPy2ZTAZbW9uaDoOqCWfqiIiIqsF/l19PnTqFtm3bwsjICMbGxmjcuDGOHTumUXfjxo1wd3eHnp4eQkJCkJiYKO5/9epVdO/eHTY2NlAqlWjSpAl27dql0aezszM+++wzjBgxAkZGRqhduzaWLVsmlpe2/Hr27Fl07doVxsbGMDIyQuvWrXH16tXqOShUrZjUERERPQeDBw9GrVq1EBsbi+PHj2Py5MnQ1dUVy7OzszFr1iysXr0aMTExSEtLw4ABA8TyrKwsdO7cGdHR0Th58iRCQ0PRrVs3JCQkaPQzd+5c+Pv74+TJkxgzZgzefvttXLx4sdSYbt26hTZt2kChUGD37t04fvw4RowYgcLCwuo5CFStuPxKL6VpU+ZDodCr6TComrRp5VDTIdATdO41uKZDqHGbNm2CUqnU2KZSqcqsn5CQgPfffx9169YFALi7u2uUFxQUYNGiRWjatCkAYNWqVfDy8sLRo0cREBAAPz8/+Pn5ifU/+eQTbNiwAX/++SfCw8PF7Z07d8aYMWMAAB988AG+/vpr7NmzB56eniVi+vbbb2FiYoKoqCgxwfTw8KjMYaAXCGfqiIiInkLbtm0RFxen8frhhx/KrD9x4kSMGjUKwcHB+Pzzz0sscero6KBJkybi+7p168LU1BTnz58HUDRTFxERAS8vL5iamkKpVOL8+fMlZup8fX3FryUSCWxtbZGSklJqTHFxcWjdurXGjCG9vJjUERERPQVDQ0O4ublpvBwcyp5lnj59Os6ePYsuXbpg9+7d8Pb2xoYNGyrcX0REBDZs2IDPPvsMBw4cQFxcHOrXr4/8/HyNev9N0CQSCdRqdaltvmgXctCzYVJHRET0nHh4eGDChAnYsWMHevXqhcjISLGssLBQvHACAC5evIi0tDR4eXkBAGJiYhAWFoaePXuifv36sLW1RXx8/DPF4+vriwMHDqCgoOCZ2qEXA5M6IiKiapaTk4Pw8HDs3bsXN27cQExMDGJjY8WEDSiaYRs7diyOHDmC48ePIywsDM2aNUNAQACAonPwfv/9d8TFxeHUqVMYNGhQmTNwFRUeHo6MjAwMGDAAx44dw+XLl7FmzZoyL6ygFxuTOiIiomomk8lw//59DBs2DB4eHujXrx86deqEGTNmiHUMDAzwwQcfYNCgQWjZsiWUSiXWrl0rls+bNw9mZmZo0aIFunXrhpCQEDRq1OiZ4rKwsMDu3buRlZWFwMBANG7cGN9//z3PsXtJSQRBEGo6CKKKysjIgImJCcaPmcGrX7UYr3598VX06tfiz2x6ejqMjY01ynJzc3H9+nW4uLhAT+/V/jyvXLkS48ePR1paWk2HQi+gin5WOFNHREREpAWY1BERERFpASZ1RERENSwsLIxLr/TMmNQRERERaQEmdURERERagEkdERERkRZgUkdERESkBXRqOgB6+RWf4Ltx48aaDuWFFBzSEsEhLUst+zDiS6jVAvr0D4WzqyOMjA0hAZCZ+RAXzl/Dru0xyMnOLbd9ewcbtO/YAo617WBgqIecnDzcvpmMPbv+Rvz1m9UwImoY0Ap2tZwAALcT4xEXGyOWGRmbws2rPswtraGrq4v8vDyk3r+Lk0cPltmenr4+3Dzrw8zCCnr6+pBIpch5+BA3E64h/soF8HaiRFQRTOqInpOsrGw8uJ+msa34d7WXjztyc/NwN+UBDA31YWFphpatG8PSygyRy9aX2aaengJvvN0f+gZ6yMvNR/Kd+7CyMoOnlytc3Wrj85nf4eHDnGoc1aunlpOrmND9l5mFFQJatoNMRwcFBfnIzEiHjo4OrO1qldumgaERaru6o7CgAA8fZsLAUAkjE1N41W8EA0MlzsbFVsdQiEjLMKmjGicIAlQqFXR0tPvH8eK5q/g1amupZbNnLEZhoUp8/1b4QLi4OsLZufxkwMbOEvoGRXcX/23dNvwTdwGNm/ig78DO0NXVgdLIkEldFTIwVMLb1x+p9+9CT98A+gaGGuX1GzaFTEcHtxKu4/SJI1Cri76nsif8bBfk5+OfE3/jdsJ1qNVq6OjK0apdKAwMjWDv6MykTsvw6RFUXXhOnRbJzMzE4MGDYWhoCDs7O3z99dcICgrC+PHjAQB5eXmIiIiAg4MDDA0N0bRpU+zdu1fcf+XKlTA1NcX27dvh5eUFpVKJ0NBQJCUliXVUKhUmTpwIU1NTWFhYYNKkSSWWhtRqNWbPng0XFxfo6+vDz88P69c/mm3au3cvJBIJtm7disaNG0OhUODgwbKXprSFj68nPvliAj6cPgbDR/aGvYO1WFZYqEKH0FYY8+4QfDD1Lbi4OgLAE5dPU+7cQ3Z2UdLWu18owicMQ/dewcjPL8DunYeRfOde9Q3oFSORSODXpCUECIiLjSnxc29kYgqlsYn4PrBjN3To1hcBrdrDUGn83+Y0ZGak4Wb8VfHh7IX/zvIBgFr1bA9sp+qTmJiIESNGwN7eHnK5HE5OTnj33Xdx//59sY6zszPmz59fc0HSK0W7p0ZeMRMnTkRMTAz+/PNP2NjY4H//+x9OnDiBBg0aAADCw8Nx7tw5REVFwd7eHhs2bEBoaChOnz4Nd3d3AEB2dja++uorrFmzBlKpFEOGDEFERAR++uknAMDcuXOxcuVKrFixAl5eXpg7dy42bNiAdu3aiXHMnj0bP/74I5YsWQJ3d3fs378fQ4YMgZWVFQIDA8V6kydPxldffQVXV1eYmZk9vwNVA1QqNTIzH0KtVsPaxgLG9ZRw83DCdwt/xO1bKQAASysz1HayF/e5fDH+/+3de1SU1d7A8e8MDHIZGG4KiCgXQSC1MEw0xUuSppwMMs3DUViGrbM6ZIpompbm0fR1pan5evRognlMPees0vdFj2Z4ecu8AFJaECGIg4RiolwEh+Hy/kFMjngBRNHx91mLteZ59n72s59hDfxm//azH7Z++j93bLeqSse6jz8j+rVInJwd6OLhCsCVklKKfim+fxf0GOoe0AsHR2e+Sz1CVeW1JuU3Bm7uXb2oKC/FXKXCuZMrIYOG83XK7lsedys2alucOroAUJB/pm0u4BETPLd571VbSFtsc/dKN8nLy6N///74+fmxbds2vLy8+PHHH5k5cyb/+c9/OHbsGI6Ojveht7en1+tRqVQP9Jzi4SIjdSaivLyczZs38+GHH/Lcc8/Rs2dPEhMTqa1tSP9otVoSExP517/+xaBBg/Dx8SEhIYGBAweSmJhoaEev17Nu3TqCg4Pp06cPcXFxpKSkGMpXrlzJnDlziIyMJCAggHXr1qHR/D46odPp+OCDD9i0aRMjRozA29ubmJgY/vSnP7F+/XqjPi9cuJCwsDB8fHxu+8dPp9NRVlZm9POo+e5kJovmr+HDJRtY8V+f8Mn6fwKgUpnT/9k+hnrbtvwv7yR8yKoPk7hQdAnfHp68FDn8jm2rLFS8MmEUTs4O7N51kHdnf0TyrgM4OGqYMPFFo9FA0Xoae0d8/J6gUHuWXwryb1lHqfj9z2lB/hn+b38y36Ts+S2dqqJLN+/mncvBkZDQMMzNVVwo1JKTdaotLkG0sb/85S9YWFjw5ZdfMnjwYLp27coLL7zAV199RWFhIXPnzmXIkCGcO3eO6dOno1AoUCgURm3cKSsCsHHjRgICArC0tMTf35+1a9cayvLz81EoFOzYsYPBgwdjaWlp+PItHl8yUmci8vLy0Ov1PPPMM4Z9Go2GHj16AHD69Glqa2vx8/MzOk6n0+Hk5GTYtra2xsfHx7Dt5uZGcXHDiE9paSlFRUX069fPUG5ubk5wcLAhFXXmzBkqKysJCwszOk91dTVBQUFG+4KDg+96XUuWLOH999+/a72H2a+Xrhht52Tnc62iEhu1NRoHW6Oyuro6in4p5sSx73kxYjh9+vbkwFdHm7TR6Kk+AXh0dQMg7cQp9NV60k+cJnzMMJRKBT6+3QwjgaL11Hb2KJVKXN09cOk8DgAzs4Y/n67uXXnezZ20bw8Z6l+90pB+q6q8RnW1DktLqybz726lk1sXnur7LObm5mjP5vBDRiogd74+bEpKSti3bx+LFy/GysrKqMzV1ZWoqCh27NhBTk4OTz31FK+//jpTpkwxqne3rMjWrVt57733WLNmDUFBQWRkZDBlyhRsbGyIjo42tDN79myWL19OUFAQlpaW9//ixUNNgrrHREVFBWZmZqSnp2NmZmZUplarDa9vHrpXKBQtWk6hoqICgN27d+Pu7m5U1qFDB6NtG5u7/5ObM2cO8fHxhu2ysjI8PDya3Z+HweBhz/DdySxKr5YD0N2vGzZqawCulJTRxcMVCwsVebkFAJiZKenu62k43sKi4XfSpasr4yaMBuCf23ZzXnsBS8vf31N3D1fO/HwO999SsADV1fr7em2Pm8ZA7kZKpRKlUsm1ijL0+mpUKgs09k4UcAZLKxssLBp+R9cqGn7/GgcnngzuD8D3aUcp/S0A9PTpQUDvPoCCn05nkJeT+WAuSrRYTk4O9fX1BAQE3LI8ICCAK1euUFtbi5mZGba2tri6uhrVacyKNH6JjouLY+HChYby+fPns3z5ciIjIwHw8vIiMzOT9evXGwV106ZNM9QRQoI6E+Ht7Y1KpSI1NZWuXbsCDSNrP//8M6GhoQQFBVFbW0txcTGDBg1q1Tk0Gg1ubm4cP36c0NBQAGpqakhPT6dPn4Y0YmBgIB06dECr1RrNn2utDh06NAkGHzUhA4IYMWowpVfLqK7W07FTw8ioTlfNkf9Lw6OrG69MGEVlZRVXr5ShsbfDxqbh2/8v5y8a5sZZqFR0cnEyvAb4KTOXEaMGYW5uTkzsy1wqLsG5Y0Mqu6rqOpmncx705ZqkQm0ehdo8o31DRozB2kZttE5dTtZpAns/TVev7jg6daSDpRVKpZLr16soONswN87MzAy1rcbwGsDe0ZnAJxtGrvV6PS7uHri4//7l5eSxw+iu33m9QvHg3cv6gXfKily7do3c3Fxee+01oxG+mpoao+ku0LyMh3h8SFBnImxtbYmOjmbmzJk4OjrSqVMn5s+fj1KpRKFQ4OfnR1RUFJMmTTIM1V+6dImUlBR69+7N6NGjm3Wet956i6VLl+Lr64u/vz8rVqwwui3f1taWhIQEpk+fTl1dHQMHDqS0tJQjR45gZ2dn9A3zcXHwq2P0erIHLq7OODrZcPVKKefyC0n58ii/XipBpVKRnZWHW+dOdHJxpr6+josXfuWnrDwOfXWUO/3fuFRcwvr/3saQYSF06epKx06OVJRX/tb+t5SXP7jJ5gLyz/xEjV6PV3d/rNW2VOuuc7HoPNk/fkd1te62xymVv8/HU6lUODg631RudvMhoh11794dhUJBVlYWERERTcqzsrJwcHCgY8eOt23jTlmRxozHhg0bjKa7AE0yLc3JeIjHhwR1JmTFihX8+c9/Jjw8HDs7O2bNmkVBQYFhnkViYiKLFi1ixowZFBYW4uzsTEhICOHh4c0+x4wZMygqKiI6OhqlUsnkyZOJiIigtLTUUOevf/0rHTt2ZMmSJeTl5WFvb0+fPn1455132vyaHwUnjn3PiWPf37a88PwFEjfcfoHhRnm5BcyOX9Zkf8G5IrYkfnFPfRQtd2jfrlvuP38ul/Pncm97XMmvxez5fOtd94mHl5OTE2FhYaxdu5bp06cbzau7cOECW7duZdKkSSgUCiwsLAw3rDWXi4sLnTt3Ji8vj6ioqLbuvjBhinp5/ozJunbtGu7u7ixfvpzXXnutvbvTJsrKytBoNEx74306dJBJwaYqdKD73SuJdjUqsnnBRuNntrS0FDs74/X6rl+/ztmzZ/Hy8moyyf9hX9IkJyeHAQMGEBAQwKJFi4yWNNHpdIYlTZ5//nmsrKxYu3YtHTp0wNnZ+ZaLD+/cuZOIiAjDaN3GjRuZOnUqS5cuZeTIkeh0OtLS0rhy5Qrx8fHk5+fj5eVFRkaGYdkqYbru9Fm5kSxpYkIyMjLYtm0bubm5nDx50vANb8yYMe3cMyGEMC2+vr6kpaXh7e3NuHHj8PHx4fXXX2fo0KEcPXrUsEzTwoULyc/Px8fH547p2JvFxsayceNGEhMT6dWrF4MHDyYpKQkvL6/7dUnCBMhInQnJyMggNjaW7OxsLCwsePrpp1mxYgW9evVq7661GRmpezzISN3D736P1Akhftfcz4rMqTMhQUFBpKent3c3hBBCCNEOJP0qhBBCCGECJKgTQgghhDABEtQJIYQQQpgACeqEEEIIIUyABHVCCCGEECZAgjohhBBCCBMgQZ0QQgghhAmQoE4IIYQQwgRIUCeEEELco0OHDqFQKIye5/qgJSUlYW9v32btDRkyhGnTprVZe62xYMGCNnm27b22ExMTw0svvXTP/bjfJKgTQgghWujmgGfAgAEUFRWh0WjarU/jx4/n559/brfz3w8JCQmkpKQYth+V4Kq9yGPCxCPp/SXTmjxHUghhOq7HL3tg57JcMeue27CwsMDV1bUNetN6VlZWWFlZtWsf2pparUatVrd3Nx4ZMlInhBBCtEBMTAyHDx9m1apVKBQKFAoFSUlJRunXxlRocnIyPXr0wNramrFjx1JZWcnmzZvx9PTEwcGBqVOnUltba2hbp9ORkJCAu7s7NjY29OvXj0OHDjWrXzenXxtTjlu2bMHT0xONRsOrr75KeXl5s6+1rq6OWbNm4ejoiKurKwsWLDAq12q1jBkzBrVajZ2dHePGjePixYuG8u+//56hQ4dia2uLnZ0dTz/9NGlpaUb93blzJ76+vlhaWjJixAgKCgqaXEPj682bN7Nr1y7D+9743rz99tv4+flhbW2Nt7c37777Lnq9vtnXeaPa2lri4+Oxt7fHycmJWbNmUV9fb1Rn7969DBw40FAnPDyc3NxcQ/mwYcOIi4szOubSpUtYWFgYRh7Xrl1ruG4XFxfGjh3bqv7eSII6IYQQogVWrVpF//79mTJlCkVFRRQVFeHh4dGkXmVlJatXr2b79u3s3buXQ4cOERERwZ49e9izZw9btmxh/fr1/Pvf/zYcExcXx9GjR9m+fTunTp3ilVdeYeTIkeTk5LSqr7m5uezcuZPk5GSSk5M5fPgwS5cubfbxmzdvxsbGhuPHj7Ns2TIWLlzI/v37gYaAb8yYMZSUlHD48GH2799PXl4e48ePNxwfFRVFly5dSE1NJT09ndmzZ6NSqYzeo8WLF/Ppp59y5MgRrl69yquvvnrLviQkJDBu3DhGjhxpeN8HDBgAgK2tLUlJSWRmZrJq1So2bNjARx991Jq3jOXLl5OUlMSmTZv45ptvKCkp4YsvvjCqc+3aNeLj40lLSyMlJQWlUklERAR1dXUAxMbG8tlnn6HT6QzH/OMf/8Dd3Z1hw4aRlpbG1KlTWbhwIdnZ2ezdu5fQ0NBW9fdGkn4VQgghWkCj0WBhYYG1tbUh5frTTz81qafX6/nb3/6Gj48PAGPHjmXLli1cvHgRtVpNYGAgQ4cO5eDBg4wfPx6tVktiYiJarZbOnTsDDYHM3r17SUxM5IMPPmhxX+vq6khKSsLW1haAiRMnkpKSwuLFi5t1fO/evZk/fz4Avr6+rFmzhpSUFMLCwkhJSeH06dOcPXvWENR++umnPPHEE6SmptK3b1+0Wi0zZ87E39/f0MbN79GaNWvo168f0BBEBgQEcOLECZ555hmjumq1GisrK3Q6XZNU97x58wyvPT09SUhIYPv27cya1fLU+sqVK5kzZw6RkZEArFu3jn379hnVefnll422N23aRMeOHcnMzKRnz55ERkYSFxfHrl27GDduHNAwMhkTE4NCoUCr1WJjY0N4eDi2trZ069aNoKCgFvf1ZjJSJ4QQQtwH1tbWhoAOwMXFBU9PT6M5Yi4uLhQXFwNw+vRpamtr8fPzM8wlU6vVHD582Ci11xKenp6GgA7Azc3NcL7m6N27t9H2jcdnZWXh4eFhNEoZGBiIvb09WVlZAMTHxxMbG8vw4cNZunRpk+swNzenb9++hm1/f3+j45trx44dPPvss7i6uqJWq5k3bx5arbZFbQCUlpZSVFRkCDIb+xgcHGxULycnhwkTJuDt7Y2dnR2enp4AhnNaWloyceJENm3aBMDJkyf54YcfiImJASAsLIxu3brh7e3NxIkT2bp1K5WVlS3u780kqBNCCCHugxvTjAAKheKW+xpTdhUVFZiZmZGens53331n+MnKymLVqlVt1ofG8z2I4xcsWMCPP/7I6NGjOXDgAIGBgU1Smffq6NGjREVFMWrUKJKTk8nIyGDu3LlUV1e36Xlu9Ic//IGSkhI2bNjA8ePHOX78OIDROWNjY9m/fz/nz58nMTGRYcOG0a1bN6AhXXzy5Em2bduGm5sb7733Hk8++eQ9L4kjQZ0QQgjRQhYWFkY3OLSFoKAgamtrKS4upnv37kY/7X1n7a0EBARQUFBgdGNDZmYmV69eJTAw0LDPz8+P6dOn8+WXXxIZGUliYqKhrKamxnDjBEB2djZXr14lICDglue81fv+7bff0q1bN+bOnUtwcDC+vr6cO3euVdek0Whwc3MzBGmNfUxPTzdsX758mezsbObNm8dzzz1HQEAAV65cadJWr169CA4OZsOGDXz22WdMnjzZqNzc3Jzhw4ezbNkyTp06RX5+PgcOHGhVvw1t3tPRQjxgjXcglZWVtXNPhBDN0fhZvfnuwUedp6cnx48fJz8/H7Va3aLRq9vx8/MjKiqKSZMmsXz5coKCgrh06RIpKSn07t2b0aNHt0HP287w4cPp1asXUVFRrFy5kpqaGt544w0GDx5McHAwVVVVzJw5k7Fjx+Ll5cX58+dJTU01mo+mUql48803Wb16Nebm5sTFxRESEtJkPl0jT09P9u3bR3Z2Nk5OTmg0Gnx9fdFqtWzfvp2+ffuye/fuexoNfOutt1i6dCm+vr74+/uzYsUKoxE0BwcHnJyc+Pvf/46bmxtarZbZs2ffsq3Y2Fji4uKwsbEhIiLCsD85OZm8vDxCQ0NxcHBgz5491NXV0aNHj1b3GySoE4+Yxlvxb3WnmRDi4VVeXt6uC/O2tYSEBKKjowkMDKSqqspo9OleJCYmsmjRImbMmEFhYSHOzs6EhIQQHh7eJu23JYVCwa5du3jzzTcJDQ1FqVQycuRIPv74YwDMzMy4fPkykyZN4uLFizg7OxMZGcn7779vaMPa2pq3336bP/7xjxQWFjJo0CA++eST255zypQpHDp0iODgYCoqKjh48CAvvvgi06dPJy4uDp1Ox+jRo3n33XebLL/SXDNmzKCoqIjo6GiUSiWTJ08mIiKC0tJSAJRKJdu3b2fq1Kn07NmTHj16sHr1aoYMGdKkrQkTJjBt2jQmTJiApaWlYb+9vT2ff/45CxYs4Pr16/j6+rJt2zaeeOKJVvW5kaLe1L4+CZNWV1fHL7/8gq2tLQqFor27I4S4i/r6esrLy+ncuTNKpfGMn+vXr3P27Fm8vLyM/uGJx0NSUhLTpk1r10er3W/5+fn4+PiQmppKnz59Wt1Ocz8rMlInHilKpZIuXbq0dzeEEC1gSiN0QjSHXq/n8uXLzJs3j5CQkHsK6FpCbpQQQgghHgEvvPCC0VInN/60dA07rVZ727bUanWrlgN5VNzpur/++us2OceRI0dwc3MjNTWVdevWtUmbzSHpVyGEEO1C0q8tU1hYSFVV1S3LHB0dcXR0bHZbNTU15Ofn37bc09MTc3PTTOadOXPmtmXu7u4P5fNzJf0qhBBCmBB3d/c2a8vc3Jzu3bu3WXuPElO+bkm/CiGEEEKYAAnqhBBCtCuZBSTEnTX3MyJBnRBCiHbR+AiqtnjmpRCmrPEzcvNj224mc+qEEEK0CzMzM+zt7Q0PiLe2tpb1J4W4QX19PZWVlRQXF2Nvb4+Zmdkd68vdr0IIIdpNfX09Fy5cMOkFaIW4V/b29ri6ut71S48EdUIIIdpdbW0ter2+vbshxENHpVLddYSukQR1QgghhBAmQG6UEEIIIYQwARLUCSGEEEKYAAnqhBBCCCFMgAR1QgghhBAmQII6IYQQQggTIEGdEEIIIYQJkKBOCCGEEMIE/D+t7W5M3POyVQAAAABJRU5ErkJggg==",
+      "text/plain": [
+       "<Figure size 640x480 with 2 Axes>"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    }
+   ],
+   "source": [
+    "pop_track.plot_cohort_change(subfigure_title=True)"
+   ]
+  }
+ ],
+ "metadata": {
+  "kernelspec": {
+   "display_name": "ehrapy_venv_feb",
+   "language": "python",
+   "name": "python3"
+  },
+  "language_info": {
+   "codemirror_mode": {
+    "name": "ipython",
+    "version": 3
+   },
+   "file_extension": ".py",
+   "mimetype": "text/x-python",
+   "name": "python",
+   "nbconvert_exporter": "python",
+   "pygments_lexer": "ipython3",
+   "version": "3.11.7"
+  }
+ },
+ "nbformat": 4,
+ "nbformat_minor": 2
+}
diff --git a/try_flowchart.ipynb b/try_flowchart.ipynb
deleted file mode 100644
index a1bafc39..00000000
--- a/try_flowchart.ipynb
+++ /dev/null
@@ -1,1915 +0,0 @@
-{
- "cells": [
-  {
-   "cell_type": "code",
-   "execution_count": 55,
-   "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>haircolor</th>\n",
-       "      <th>height</th>\n",
-       "    </tr>\n",
-       "  </thead>\n",
-       "  <tbody>\n",
-       "    <tr>\n",
-       "      <th>0</th>\n",
-       "      <td>colored</td>\n",
-       "      <td>179</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>1</th>\n",
-       "      <td>black</td>\n",
-       "      <td>165</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>2</th>\n",
-       "      <td>colored</td>\n",
-       "      <td>174</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>3</th>\n",
-       "      <td>black</td>\n",
-       "      <td>173</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>4</th>\n",
-       "      <td>colored</td>\n",
-       "      <td>181</td>\n",
-       "    </tr>\n",
-       "  </tbody>\n",
-       "</table>\n",
-       "</div>"
-      ],
-      "text/plain": [
-       "  haircolor  height\n",
-       "0   colored     179\n",
-       "1     black     165\n",
-       "2   colored     174\n",
-       "3     black     173\n",
-       "4   colored     181"
-      ]
-     },
-     "execution_count": 55,
-     "metadata": {},
-     "output_type": "execute_result"
-    }
-   ],
-   "source": [
-    "import pandas as pd\n",
-    "import matplotlib.pyplot as plt\n",
-    "import numpy as np\n",
-    "import ehrapy as ep\n",
-    "from tableone import TableOne\n",
-    "import seaborn as sns\n",
-    "import scanpy as sc\n",
-    "\n",
-    "# generate dataset\n",
-    "rng = np.random.default_rng(151)\n",
-    "\n",
-    "random_dataset = pd.DataFrame(\n",
-    "    {\n",
-    "        \"haircolor\": rng.choice([\"blond\", \"black\", \"colored\"], size=150),\n",
-    "        \"height\": rng.integers(low=160, high=190, size=150),\n",
-    "        # \"proficiency\": rng.choice([\"high prof.\", \"limited prof.\"], size=150),\n",
-    "    }\n",
-    ")\n",
-    "random_dataset.head()"
-   ]
-  },
-  {
-   "cell_type": "code",
-   "execution_count": 72,
-   "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>haircolor</th>\n",
-       "      <th>height</th>\n",
-       "    </tr>\n",
-       "  </thead>\n",
-       "  <tbody>\n",
-       "    <tr>\n",
-       "      <th>0</th>\n",
-       "      <td>colored</td>\n",
-       "      <td>176</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>1</th>\n",
-       "      <td>black</td>\n",
-       "      <td>173</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>2</th>\n",
-       "      <td>colored</td>\n",
-       "      <td>175</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>3</th>\n",
-       "      <td>black</td>\n",
-       "      <td>175</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>4</th>\n",
-       "      <td>colored</td>\n",
-       "      <td>176</td>\n",
-       "    </tr>\n",
-       "  </tbody>\n",
-       "</table>\n",
-       "</div>"
-      ],
-      "text/plain": [
-       "  haircolor  height\n",
-       "0   colored     176\n",
-       "1     black     173\n",
-       "2   colored     175\n",
-       "3     black     175\n",
-       "4   colored     176"
-      ]
-     },
-     "execution_count": 72,
-     "metadata": {},
-     "output_type": "execute_result"
-    }
-   ],
-   "source": [
-    "# generate dataset\n",
-    "rng = np.random.default_rng(151)\n",
-    "\n",
-    "random_dataset = pd.DataFrame(\n",
-    "    {\n",
-    "        \"haircolor\": rng.choice([\"blond\", \"black\", \"colored\"], size=150),\n",
-    "        \"height\": rng.integers(low=173, high=178, size=150),\n",
-    "        # \"proficiency\": rng.choice([\"high prof.\", \"limited prof.\"], size=150),\n",
-    "    }\n",
-    ")\n",
-    "random_dataset.head()"
-   ]
-  },
-  {
-   "cell_type": "code",
-   "execution_count": 57,
-   "metadata": {},
-   "outputs": [
-    {
-     "data": {
-      "image/png": "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",
-      "text/latex": [
-       "$\\displaystyle 0.0333333333333333$"
-      ],
-      "text/plain": [
-       "0.03333333333333333"
-      ]
-     },
-     "execution_count": 57,
-     "metadata": {},
-     "output_type": "execute_result"
-    }
-   ],
-   "source": [
-    "1.0 * random_dataset[\"height\"].nunique() / random_dataset[\"height\"].count()"
-   ]
-  },
-  {
-   "cell_type": "code",
-   "execution_count": 76,
-   "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></th>\n",
-       "      <th>Missing</th>\n",
-       "      <th>Overall</th>\n",
-       "    </tr>\n",
-       "  </thead>\n",
-       "  <tbody>\n",
-       "    <tr>\n",
-       "      <th>n</th>\n",
-       "      <th></th>\n",
-       "      <td></td>\n",
-       "      <td>150</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th rowspan=\"3\" valign=\"top\">haircolor, n (%)</th>\n",
-       "      <th>black</th>\n",
-       "      <td>0</td>\n",
-       "      <td>45 (30.0)</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>blond</th>\n",
-       "      <td></td>\n",
-       "      <td>55 (36.7)</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>colored</th>\n",
-       "      <td></td>\n",
-       "      <td>50 (33.3)</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>height, mean (SD)</th>\n",
-       "      <th></th>\n",
-       "      <td>0</td>\n",
-       "      <td>175.0 (1.5)</td>\n",
-       "    </tr>\n",
-       "  </tbody>\n",
-       "</table>\n",
-       "</div><br />"
-      ],
-      "text/plain": [
-       "                          Missing      Overall\n",
-       "n                                          150\n",
-       "haircolor, n (%)  black         0    45 (30.0)\n",
-       "                  blond              55 (36.7)\n",
-       "                  colored            50 (33.3)\n",
-       "height, mean (SD)               0  175.0 (1.5)"
-      ]
-     },
-     "execution_count": 76,
-     "metadata": {},
-     "output_type": "execute_result"
-    }
-   ],
-   "source": [
-    "TableOne(random_dataset, categorical=[\"haircolor\"])"
-   ]
-  },
-  {
-   "cell_type": "code",
-   "execution_count": 59,
-   "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>haircolor</th>\n",
-       "      <th>height</th>\n",
-       "    </tr>\n",
-       "  </thead>\n",
-       "  <tbody>\n",
-       "    <tr>\n",
-       "      <th>0</th>\n",
-       "      <td>colored</td>\n",
-       "      <td>176</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>1</th>\n",
-       "      <td>black</td>\n",
-       "      <td>173</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>2</th>\n",
-       "      <td>colored</td>\n",
-       "      <td>175</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>3</th>\n",
-       "      <td>black</td>\n",
-       "      <td>175</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>4</th>\n",
-       "      <td>colored</td>\n",
-       "      <td>176</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>...</th>\n",
-       "      <td>...</td>\n",
-       "      <td>...</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>145</th>\n",
-       "      <td>blond</td>\n",
-       "      <td>173</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>146</th>\n",
-       "      <td>black</td>\n",
-       "      <td>177</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>147</th>\n",
-       "      <td>black</td>\n",
-       "      <td>173</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>148</th>\n",
-       "      <td>blond</td>\n",
-       "      <td>174</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>149</th>\n",
-       "      <td>blond</td>\n",
-       "      <td>173</td>\n",
-       "    </tr>\n",
-       "  </tbody>\n",
-       "</table>\n",
-       "<p>150 rows × 2 columns</p>\n",
-       "</div>"
-      ],
-      "text/plain": [
-       "    haircolor  height\n",
-       "0     colored     176\n",
-       "1       black     173\n",
-       "2     colored     175\n",
-       "3       black     175\n",
-       "4     colored     176\n",
-       "..        ...     ...\n",
-       "145     blond     173\n",
-       "146     black     177\n",
-       "147     black     173\n",
-       "148     blond     174\n",
-       "149     blond     173\n",
-       "\n",
-       "[150 rows x 2 columns]"
-      ]
-     },
-     "execution_count": 59,
-     "metadata": {},
-     "output_type": "execute_result"
-    }
-   ],
-   "source": [
-    "adata = sc.AnnData(X=rng.normal(size=(150, 100)), obs=random_dataset)\n",
-    "adata.obs"
-   ]
-  },
-  {
-   "cell_type": "code",
-   "execution_count": 60,
-   "metadata": {},
-   "outputs": [
-    {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      "The autoreload extension is already loaded. To reload it, use:\n",
-      "  %reload_ext autoreload\n"
-     ]
-    }
-   ],
-   "source": [
-    "%load_ext autoreload\n",
-    "%autoreload 2"
-   ]
-  },
-  {
-   "cell_type": "code",
-   "execution_count": 61,
-   "metadata": {},
-   "outputs": [],
-   "source": [
-    "pl = ep.tl.PopulationTracker(adata)\n",
-    "pl(adata)\n",
-    "adata = adata[:75]\n",
-    "pl(adata, label=\"filtered\", operations_done=\"filtered to first 75 entries\")"
-   ]
-  },
-  {
-   "cell_type": "code",
-   "execution_count": 62,
-   "metadata": {},
-   "outputs": [
-    {
-     "data": {
-      "text/plain": [
-       "[None, 'filtered to first 75 entries']"
-      ]
-     },
-     "execution_count": 62,
-     "metadata": {},
-     "output_type": "execute_result"
-    }
-   ],
-   "source": [
-    "pl._tracked_operations"
-   ]
-  },
-  {
-   "cell_type": "code",
-   "execution_count": 63,
-   "metadata": {},
-   "outputs": [],
-   "source": [
-    "# pl.reset()"
-   ]
-  },
-  {
-   "cell_type": "code",
-   "execution_count": 64,
-   "metadata": {},
-   "outputs": [
-    {
-     "data": {
-      "image/png": "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",
-      "text/latex": [
-       "$\\displaystyle 2$"
-      ],
-      "text/plain": [
-       "2"
-      ]
-     },
-     "execution_count": 64,
-     "metadata": {},
-     "output_type": "execute_result"
-    }
-   ],
-   "source": [
-    "pl.tracked_steps"
-   ]
-  },
-  {
-   "cell_type": "code",
-   "execution_count": 65,
-   "metadata": {},
-   "outputs": [
-    {
-     "data": {
-      "text/plain": [
-       "{'haircolor': {'black': [30.0, 25.3],\n",
-       "  'blond': [36.7, 40.0],\n",
-       "  'colored': [33.3, 34.7]},\n",
-       " 'height': []}"
-      ]
-     },
-     "execution_count": 65,
-     "metadata": {},
-     "output_type": "execute_result"
-    }
-   ],
-   "source": [
-    "pl.track"
-   ]
-  },
-  {
-   "cell_type": "code",
-   "execution_count": 66,
-   "metadata": {},
-   "outputs": [
-    {
-     "ename": "KeyError",
-     "evalue": "0",
-     "output_type": "error",
-     "traceback": [
-      "\u001b[0;31m---------------------------------------------------------------------------\u001b[0m",
-      "\u001b[0;31mValueError\u001b[0m                                Traceback (most recent call last)",
-      "File \u001b[0;32m~/Documents/ehrapy_clean/ehrapy_venv_feb/lib/python3.11/site-packages/pandas/core/indexes/range.py:413\u001b[0m, in \u001b[0;36mRangeIndex.get_loc\u001b[0;34m(self, key)\u001b[0m\n\u001b[1;32m    412\u001b[0m \u001b[38;5;28;01mtry\u001b[39;00m:\n\u001b[0;32m--> 413\u001b[0m     \u001b[38;5;28;01mreturn\u001b[39;00m \u001b[38;5;28;43mself\u001b[39;49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43m_range\u001b[49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43mindex\u001b[49m\u001b[43m(\u001b[49m\u001b[43mnew_key\u001b[49m\u001b[43m)\u001b[49m\n\u001b[1;32m    414\u001b[0m \u001b[38;5;28;01mexcept\u001b[39;00m \u001b[38;5;167;01mValueError\u001b[39;00m \u001b[38;5;28;01mas\u001b[39;00m err:\n",
-      "\u001b[0;31mValueError\u001b[0m: 0 is not in range",
-      "\nThe above exception was the direct cause of the following exception:\n",
-      "\u001b[0;31mKeyError\u001b[0m                                  Traceback (most recent call last)",
-      "Cell \u001b[0;32mIn[66], line 1\u001b[0m\n\u001b[0;32m----> 1\u001b[0m fig, axes \u001b[38;5;241m=\u001b[39m \u001b[43mpl\u001b[49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43mplot_population_change\u001b[49m\u001b[43m(\u001b[49m\u001b[43mreturn_plot\u001b[49m\u001b[38;5;241;43m=\u001b[39;49m\u001b[38;5;28;43;01mTrue\u001b[39;49;00m\u001b[43m)\u001b[49m\n",
-      "File \u001b[0;32m~/Documents/ehrapy_clean/ehrapy/ehrapy/tools/population_tracking/_pop_tracker.py:142\u001b[0m, in \u001b[0;36mPopulationTracker.plot_population_change\u001b[0;34m(self, set_axis_labels, save, return_plot)\u001b[0m\n\u001b[1;32m    140\u001b[0m \u001b[38;5;28;01mfor\u001b[39;00m idx, ax \u001b[38;5;129;01min\u001b[39;00m \u001b[38;5;28menumerate\u001b[39m(axes):\n\u001b[1;32m    141\u001b[0m     \u001b[38;5;28;01mfor\u001b[39;00m pos, (_cols, data) \u001b[38;5;129;01min\u001b[39;00m \u001b[38;5;28menumerate\u001b[39m(\u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39mtrack\u001b[38;5;241m.\u001b[39mitems()):\n\u001b[0;32m--> 142\u001b[0m         data \u001b[38;5;241m=\u001b[39m \u001b[43mpd\u001b[49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43mDataFrame\u001b[49m\u001b[43m(\u001b[49m\u001b[43mdata\u001b[49m\u001b[43m)\u001b[49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43mloc\u001b[49m\u001b[43m[\u001b[49m\u001b[43midx\u001b[49m\u001b[43m]\u001b[49m\n\u001b[1;32m    144\u001b[0m         cumwidth \u001b[38;5;241m=\u001b[39m \u001b[38;5;241m0\u001b[39m\n\u001b[1;32m    146\u001b[0m         \u001b[38;5;66;03m# Adjust the hue shift based on the category position such that the colors are more distinguishable\u001b[39;00m\n",
-      "File \u001b[0;32m~/Documents/ehrapy_clean/ehrapy_venv_feb/lib/python3.11/site-packages/pandas/core/indexing.py:1192\u001b[0m, in \u001b[0;36m_LocationIndexer.__getitem__\u001b[0;34m(self, key)\u001b[0m\n\u001b[1;32m   1190\u001b[0m maybe_callable \u001b[38;5;241m=\u001b[39m com\u001b[38;5;241m.\u001b[39mapply_if_callable(key, \u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39mobj)\n\u001b[1;32m   1191\u001b[0m maybe_callable \u001b[38;5;241m=\u001b[39m \u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39m_check_deprecated_callable_usage(key, maybe_callable)\n\u001b[0;32m-> 1192\u001b[0m \u001b[38;5;28;01mreturn\u001b[39;00m \u001b[38;5;28;43mself\u001b[39;49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43m_getitem_axis\u001b[49m\u001b[43m(\u001b[49m\u001b[43mmaybe_callable\u001b[49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[43maxis\u001b[49m\u001b[38;5;241;43m=\u001b[39;49m\u001b[43maxis\u001b[49m\u001b[43m)\u001b[49m\n",
-      "File \u001b[0;32m~/Documents/ehrapy_clean/ehrapy_venv_feb/lib/python3.11/site-packages/pandas/core/indexing.py:1432\u001b[0m, in \u001b[0;36m_LocIndexer._getitem_axis\u001b[0;34m(self, key, axis)\u001b[0m\n\u001b[1;32m   1430\u001b[0m \u001b[38;5;66;03m# fall thru to straight lookup\u001b[39;00m\n\u001b[1;32m   1431\u001b[0m \u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39m_validate_key(key, axis)\n\u001b[0;32m-> 1432\u001b[0m \u001b[38;5;28;01mreturn\u001b[39;00m \u001b[38;5;28;43mself\u001b[39;49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43m_get_label\u001b[49m\u001b[43m(\u001b[49m\u001b[43mkey\u001b[49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[43maxis\u001b[49m\u001b[38;5;241;43m=\u001b[39;49m\u001b[43maxis\u001b[49m\u001b[43m)\u001b[49m\n",
-      "File \u001b[0;32m~/Documents/ehrapy_clean/ehrapy_venv_feb/lib/python3.11/site-packages/pandas/core/indexing.py:1382\u001b[0m, in \u001b[0;36m_LocIndexer._get_label\u001b[0;34m(self, label, axis)\u001b[0m\n\u001b[1;32m   1380\u001b[0m \u001b[38;5;28;01mdef\u001b[39;00m \u001b[38;5;21m_get_label\u001b[39m(\u001b[38;5;28mself\u001b[39m, label, axis: AxisInt):\n\u001b[1;32m   1381\u001b[0m     \u001b[38;5;66;03m# GH#5567 this will fail if the label is not present in the axis.\u001b[39;00m\n\u001b[0;32m-> 1382\u001b[0m     \u001b[38;5;28;01mreturn\u001b[39;00m \u001b[38;5;28;43mself\u001b[39;49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43mobj\u001b[49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43mxs\u001b[49m\u001b[43m(\u001b[49m\u001b[43mlabel\u001b[49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[43maxis\u001b[49m\u001b[38;5;241;43m=\u001b[39;49m\u001b[43maxis\u001b[49m\u001b[43m)\u001b[49m\n",
-      "File \u001b[0;32m~/Documents/ehrapy_clean/ehrapy_venv_feb/lib/python3.11/site-packages/pandas/core/generic.py:4295\u001b[0m, in \u001b[0;36mNDFrame.xs\u001b[0;34m(self, key, axis, level, drop_level)\u001b[0m\n\u001b[1;32m   4293\u001b[0m             new_index \u001b[38;5;241m=\u001b[39m index[loc]\n\u001b[1;32m   4294\u001b[0m \u001b[38;5;28;01melse\u001b[39;00m:\n\u001b[0;32m-> 4295\u001b[0m     loc \u001b[38;5;241m=\u001b[39m \u001b[43mindex\u001b[49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43mget_loc\u001b[49m\u001b[43m(\u001b[49m\u001b[43mkey\u001b[49m\u001b[43m)\u001b[49m\n\u001b[1;32m   4297\u001b[0m     \u001b[38;5;28;01mif\u001b[39;00m \u001b[38;5;28misinstance\u001b[39m(loc, np\u001b[38;5;241m.\u001b[39mndarray):\n\u001b[1;32m   4298\u001b[0m         \u001b[38;5;28;01mif\u001b[39;00m loc\u001b[38;5;241m.\u001b[39mdtype \u001b[38;5;241m==\u001b[39m np\u001b[38;5;241m.\u001b[39mbool_:\n",
-      "File \u001b[0;32m~/Documents/ehrapy_clean/ehrapy_venv_feb/lib/python3.11/site-packages/pandas/core/indexes/range.py:415\u001b[0m, in \u001b[0;36mRangeIndex.get_loc\u001b[0;34m(self, key)\u001b[0m\n\u001b[1;32m    413\u001b[0m         \u001b[38;5;28;01mreturn\u001b[39;00m \u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39m_range\u001b[38;5;241m.\u001b[39mindex(new_key)\n\u001b[1;32m    414\u001b[0m     \u001b[38;5;28;01mexcept\u001b[39;00m \u001b[38;5;167;01mValueError\u001b[39;00m \u001b[38;5;28;01mas\u001b[39;00m err:\n\u001b[0;32m--> 415\u001b[0m         \u001b[38;5;28;01mraise\u001b[39;00m \u001b[38;5;167;01mKeyError\u001b[39;00m(key) \u001b[38;5;28;01mfrom\u001b[39;00m \u001b[38;5;21;01merr\u001b[39;00m\n\u001b[1;32m    416\u001b[0m \u001b[38;5;28;01mif\u001b[39;00m \u001b[38;5;28misinstance\u001b[39m(key, Hashable):\n\u001b[1;32m    417\u001b[0m     \u001b[38;5;28;01mraise\u001b[39;00m \u001b[38;5;167;01mKeyError\u001b[39;00m(key)\n",
-      "\u001b[0;31mKeyError\u001b[0m: 0"
-     ]
-    },
-    {
-     "data": {
-      "image/png": "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",
-      "text/plain": [
-       "<Figure size 700x700 with 2 Axes>"
-      ]
-     },
-     "metadata": {},
-     "output_type": "display_data"
-    }
-   ],
-   "source": [
-    "fig, axes = pl.plot_population_change(return_plot=True)"
-   ]
-  },
-  {
-   "cell_type": "code",
-   "execution_count": 67,
-   "metadata": {},
-   "outputs": [
-    {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      "\u001b[1;35m2024-02-14 11:07:38,156\u001b[0m - \u001b[1;34mroot\u001b[0m \u001b[1;37mINFO - Transformed passed DataFrame into an AnnData object with n_obs x n_vars = `101766` x `47`.\u001b[0m\n"
-     ]
-    },
-    {
-     "data": {
-      "image/svg+xml": [
-       "<?xml version=\"1.0\" encoding=\"UTF-8\" standalone=\"no\"?>\n",
-       "<!DOCTYPE svg PUBLIC \"-//W3C//DTD SVG 1.1//EN\"\n",
-       " \"http://www.w3.org/Graphics/SVG/1.1/DTD/svg11.dtd\">\n",
-       "<!-- Generated by graphviz version 9.0.0 (20230911.1827)\n",
-       " -->\n",
-       "<!-- Pages: 1 -->\n",
-       "<svg width=\"208pt\" height=\"143pt\"\n",
-       " viewBox=\"0.00 0.00 207.62 142.50\" xmlns=\"http://www.w3.org/2000/svg\" xmlns:xlink=\"http://www.w3.org/1999/xlink\">\n",
-       "<g id=\"graph0\" class=\"graph\" transform=\"scale(1 1) rotate(0) translate(4 138.5)\">\n",
-       "<polygon fill=\"white\" stroke=\"none\" points=\"-4,4 -4,-138.5 203.62,-138.5 203.62,4 -4,4\"/>\n",
-       "<!-- 0 -->\n",
-       "<g id=\"node1\" class=\"node\">\n",
-       "<title>0</title>\n",
-       "<polygon fill=\"lightgrey\" stroke=\"black\" points=\"100.75,-134.5 0,-134.5 0,-93.5 100.75,-93.5 100.75,-134.5\"/>\n",
-       "<text text-anchor=\"middle\" x=\"50.38\" y=\"-117.2\" font-family=\"Times,serif\" font-size=\"14.00\">Baseline cohort</text>\n",
-       "<text text-anchor=\"middle\" x=\"50.38\" y=\"-100.7\" font-family=\"Times,serif\" font-size=\"14.00\"> (n=101766)</text>\n",
-       "</g>\n",
-       "<!-- 1 -->\n",
-       "<g id=\"node2\" class=\"node\">\n",
-       "<title>1</title>\n",
-       "<polygon fill=\"lightgrey\" stroke=\"black\" points=\"98.5,-41 2.25,-41 2.25,0 98.5,0 98.5,-41\"/>\n",
-       "<text text-anchor=\"middle\" x=\"50.38\" y=\"-23.7\" font-family=\"Times,serif\" font-size=\"14.00\">Filtered cohort</text>\n",
-       "<text text-anchor=\"middle\" x=\"50.38\" y=\"-7.2\" font-family=\"Times,serif\" font-size=\"14.00\"> (n=1000)</text>\n",
-       "</g>\n",
-       "<!-- 0&#45;&gt;1 -->\n",
-       "<g id=\"edge1\" class=\"edge\">\n",
-       "<title>0&#45;&gt;1</title>\n",
-       "<path fill=\"none\" stroke=\"black\" d=\"M50.38,-93.15C50.38,-81.28 50.38,-65.94 50.38,-52.5\"/>\n",
-       "<polygon fill=\"black\" stroke=\"black\" points=\"53.88,-52.83 50.38,-42.83 46.88,-52.83 53.88,-52.83\"/>\n",
-       "<text text-anchor=\"middle\" x=\"125\" y=\"-62.2\" font-family=\"Times,serif\" font-size=\"14.00\">filtered to first 1000 entries</text>\n",
-       "</g>\n",
-       "</g>\n",
-       "</svg>\n"
-      ],
-      "text/plain": [
-       "<graphviz.graphs.Digraph at 0x2ebecee50>"
-      ]
-     },
-     "execution_count": 67,
-     "metadata": {},
-     "output_type": "execute_result"
-    }
-   ],
-   "source": [
-    "adata = ep.dt.diabetes_130(columns_obs_only=[\"gender\", \"race\", \"weight\", \"age\"])\n",
-    "pop_track = ep.tl.PopulationTracker(adata)\n",
-    "pop_track(adata, label=\"Baseline cohort\")\n",
-    "adata = adata[:1000]\n",
-    "pop_track(adata, label=\"Filtered cohort\", operations_done=\"filtered to first 1000 entries\")\n",
-    "pop_track.plot_flowchart(save=\"flowchart.png\")"
-   ]
-  },
-  {
-   "cell_type": "code",
-   "execution_count": 68,
-   "metadata": {},
-   "outputs": [
-    {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      "\u001b[1;35m2024-02-14 11:07:40,029\u001b[0m - \u001b[1;34mroot\u001b[0m \u001b[1;37mINFO - Transformed passed DataFrame into an AnnData object with n_obs x n_vars = `101766` x `47`.\u001b[0m\n"
-     ]
-    }
-   ],
-   "source": [
-    "adata = ep.dt.diabetes_130(columns_obs_only=[\"gender\", \"race\", \"weight\", \"age\"])\n",
-    "adata.obs.gender = adata.obs.gender.astype(\"category\")"
-   ]
-  },
-  {
-   "cell_type": "code",
-   "execution_count": 69,
-   "metadata": {},
-   "outputs": [
-    {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      "\u001b[1;35m2024-02-14 11:07:40,619\u001b[0m - \u001b[1;34mroot\u001b[0m \u001b[1;37mINFO - Transformed passed DataFrame into an AnnData object with n_obs x n_vars = `101766` x `47`.\u001b[0m\n"
-     ]
-    },
-    {
-     "data": {
-      "text/plain": [
-       "weight\n",
-       "[75-100)     1336\n",
-       "[50-75)       897\n",
-       "[100-125)     625\n",
-       "[125-150)     145\n",
-       "[25-50)        97\n",
-       "[0-25)         48\n",
-       "[150-175)      35\n",
-       "[175-200)      11\n",
-       ">200            3\n",
-       "Name: count, dtype: int64"
-      ]
-     },
-     "execution_count": 69,
-     "metadata": {},
-     "output_type": "execute_result"
-    }
-   ],
-   "source": [
-    "adata = ep.dt.diabetes_130(columns_obs_only=[\"gender\", \"race\", \"weight\", \"age\"])\n",
-    "adata.obs.weight.value_counts()"
-   ]
-  },
-  {
-   "cell_type": "code",
-   "execution_count": 70,
-   "metadata": {},
-   "outputs": [
-    {
-     "data": {
-      "text/plain": [
-       "array([55629189, 'Emergency', 'Discharged to home', ' Emergency Room', 3,\n",
-       "       nan, nan, 59, 0, 18, 0, 0, 0,\n",
-       "       'endocrine/nutritional/metabolic diseases and immunity disorders',\n",
-       "       'diabetes',\n",
-       "       'endocrine/nutritional/metabolic diseases and immunity disorders',\n",
-       "       9, nan, nan, 'No', 'No', 'No', 'No', 'No', 'No', 'No', 'No', 'No',\n",
-       "       'No', 'No', 'No', 'No', 'No', 'No', 'No', 'No', 'Up', 'No', 'No',\n",
-       "       'No', 'No', 'No', True, True, '>30', 15.0, nan], dtype=object)"
-      ]
-     },
-     "execution_count": 70,
-     "metadata": {},
-     "output_type": "execute_result"
-    }
-   ],
-   "source": [
-    "adata.X[1, :]"
-   ]
-  },
-  {
-   "cell_type": "code",
-   "execution_count": 71,
-   "metadata": {},
-   "outputs": [
-    {
-     "ename": "SyntaxError",
-     "evalue": "unmatched ']' (2079747778.py, line 2)",
-     "output_type": "error",
-     "traceback": [
-      "\u001b[0;36m  Cell \u001b[0;32mIn[71], line 2\u001b[0;36m\u001b[0m\n\u001b[0;31m    adata[]]\u001b[0m\n\u001b[0m           ^\u001b[0m\n\u001b[0;31mSyntaxError\u001b[0m\u001b[0;31m:\u001b[0m unmatched ']'\n"
-     ]
-    }
-   ],
-   "source": [
-    "adata = ep.dt.diabetes_130()\n",
-    "adata[]]"
-   ]
-  },
-  {
-   "cell_type": "code",
-   "execution_count": 253,
-   "metadata": {},
-   "outputs": [
-    {
-     "data": {
-      "text/plain": [
-       "array([[1],\n",
-       "       [3],\n",
-       "       [2],\n",
-       "       ...,\n",
-       "       [1],\n",
-       "       [10],\n",
-       "       [6]], dtype=object)"
-      ]
-     },
-     "execution_count": 253,
-     "metadata": {},
-     "output_type": "execute_result"
-    }
-   ],
-   "source": [
-    "adata.X[:, adata.var_names == \"time_in_hospital_days\"]"
-   ]
-  },
-  {
-   "cell_type": "code",
-   "execution_count": null,
-   "metadata": {},
-   "outputs": [],
-   "source": []
-  },
-  {
-   "cell_type": "code",
-   "execution_count": 78,
-   "metadata": {},
-   "outputs": [
-    {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      "\u001b[1;35m2024-02-14 11:15:44,787\u001b[0m - \u001b[1;34mroot\u001b[0m \u001b[1;37mINFO - Transformed passed DataFrame into an AnnData object with n_obs x n_vars = `101766` x `47`.\u001b[0m\n",
-      "gender                    float64\n",
-      "race                     category\n",
-      "weight                   category\n",
-      "time_in_hospital_days     float64\n",
-      "dtype: object\n"
-     ]
-    },
-    {
-     "data": {
-      "text/html": [
-       "<div>\n",
-       "<style scoped>\n",
-       "    .dataframe tbody tr th:only-of-type {\n",
-       "        vertical-align: middle;\n",
-       "    }\n",
-       "\n",
-       "    .dataframe tbody tr th {\n",
-       "        vertical-align: top;\n",
-       "    }\n",
-       "\n",
-       "    .dataframe thead th {\n",
-       "        text-align: right;\n",
-       "    }\n",
-       "</style>\n",
-       "<table border=\"1\" class=\"dataframe\">\n",
-       "  <thead>\n",
-       "    <tr style=\"text-align: right;\">\n",
-       "      <th></th>\n",
-       "      <th></th>\n",
-       "      <th>Missing</th>\n",
-       "      <th>Overall</th>\n",
-       "    </tr>\n",
-       "  </thead>\n",
-       "  <tbody>\n",
-       "    <tr>\n",
-       "      <th>n</th>\n",
-       "      <th></th>\n",
-       "      <td></td>\n",
-       "      <td>101766</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th rowspan=\"2\" valign=\"top\">gender, n (%)</th>\n",
-       "      <th>0.0</th>\n",
-       "      <td>3</td>\n",
-       "      <td>54708 (53.8)</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>1.0</th>\n",
-       "      <td></td>\n",
-       "      <td>47055 (46.2)</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th rowspan=\"5\" valign=\"top\">race, n (%)</th>\n",
-       "      <th>AfricanAmerican</th>\n",
-       "      <td>2273</td>\n",
-       "      <td>19210 (19.3)</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>Asian</th>\n",
-       "      <td></td>\n",
-       "      <td>641 (0.6)</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>Caucasian</th>\n",
-       "      <td></td>\n",
-       "      <td>76099 (76.5)</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>Hispanic</th>\n",
-       "      <td></td>\n",
-       "      <td>2037 (2.0)</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>Other</th>\n",
-       "      <td></td>\n",
-       "      <td>1506 (1.5)</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th rowspan=\"9\" valign=\"top\">weight, n (%)</th>\n",
-       "      <th>&gt;200</th>\n",
-       "      <td>98569</td>\n",
-       "      <td>3 (0.1)</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>[0-25)</th>\n",
-       "      <td></td>\n",
-       "      <td>48 (1.5)</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>[100-125)</th>\n",
-       "      <td></td>\n",
-       "      <td>625 (19.5)</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>[125-150)</th>\n",
-       "      <td></td>\n",
-       "      <td>145 (4.5)</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>[150-175)</th>\n",
-       "      <td></td>\n",
-       "      <td>35 (1.1)</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>[175-200)</th>\n",
-       "      <td></td>\n",
-       "      <td>11 (0.3)</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>[25-50)</th>\n",
-       "      <td></td>\n",
-       "      <td>97 (3.0)</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>[50-75)</th>\n",
-       "      <td></td>\n",
-       "      <td>897 (28.1)</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>[75-100)</th>\n",
-       "      <td></td>\n",
-       "      <td>1336 (41.8)</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>time_in_hospital_days, mean (SD)</th>\n",
-       "      <th></th>\n",
-       "      <td>0</td>\n",
-       "      <td>4.9 (3.0)</td>\n",
-       "    </tr>\n",
-       "  </tbody>\n",
-       "</table>\n",
-       "</div><br />"
-      ],
-      "text/plain": [
-       "                                                 Missing       Overall\n",
-       "n                                                               101766\n",
-       "gender, n (%)                    0.0                   3  54708 (53.8)\n",
-       "                                 1.0                      47055 (46.2)\n",
-       "race, n (%)                      AfricanAmerican    2273  19210 (19.3)\n",
-       "                                 Asian                       641 (0.6)\n",
-       "                                 Caucasian                76099 (76.5)\n",
-       "                                 Hispanic                   2037 (2.0)\n",
-       "                                 Other                      1506 (1.5)\n",
-       "weight, n (%)                    >200              98569       3 (0.1)\n",
-       "                                 [0-25)                       48 (1.5)\n",
-       "                                 [100-125)                  625 (19.5)\n",
-       "                                 [125-150)                   145 (4.5)\n",
-       "                                 [150-175)                    35 (1.1)\n",
-       "                                 [175-200)                    11 (0.3)\n",
-       "                                 [25-50)                      97 (3.0)\n",
-       "                                 [50-75)                    897 (28.1)\n",
-       "                                 [75-100)                  1336 (41.8)\n",
-       "time_in_hospital_days, mean (SD)                       0     4.9 (3.0)"
-      ]
-     },
-     "execution_count": 78,
-     "metadata": {},
-     "output_type": "execute_result"
-    }
-   ],
-   "source": [
-    "adata = ep.dt.diabetes_130(columns_obs_only=[\"gender\", \"race\", \"weight\", \"time_in_hospital_days\"])\n",
-    "adata.obs.time_in_hospital_days = adata.obs.time_in_hospital_days.astype(\"float\") + np.random.random(adata.n_obs)\n",
-    "print(adata.obs.dtypes)\n",
-    "TableOne(adata.obs)"
-   ]
-  },
-  {
-   "cell_type": "code",
-   "execution_count": 156,
-   "metadata": {},
-   "outputs": [
-    {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      "\u001b[1;35m2024-02-14 13:01:31,453\u001b[0m - \u001b[1;34mroot\u001b[0m \u001b[1;37mINFO - Transformed passed DataFrame into an AnnData object with n_obs x n_vars = `101766` x `48`.\u001b[0m\n"
-     ]
-    },
-    {
-     "data": {
-      "image/png": "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",
-      "text/plain": [
-       "<Figure size 700x700 with 2 Axes>"
-      ]
-     },
-     "metadata": {},
-     "output_type": "display_data"
-    }
-   ],
-   "source": [
-    "adata = ep.dt.diabetes_130(columns_obs_only=[\"gender\", \"race\", \"time_in_hospital_days\"])\n",
-    "pop_track = ep.tl.PopulationTracker(adata, categorical=[\"gender\", \"race\"])\n",
-    "pop_track(adata, label=\"Initial cohort\")\n",
-    "adata = adata[:1000]\n",
-    "pop_track(adata, label=\"Cohort 1\", operations_done=\"filtered to first 1000 entries\")\n",
-    "pop_track.plot_population_change(\n",
-    "    subfigure_title=True, subplots_kwargs={\"figsize\": (7, 7)}, legend_kwargs={\"bbox_to_anchor\": (1, 1)}\n",
-    ")"
-   ]
-  },
-  {
-   "cell_type": "code",
-   "execution_count": 127,
-   "metadata": {},
-   "outputs": [
-    {
-     "data": {
-      "image/svg+xml": [
-       "<?xml version=\"1.0\" encoding=\"UTF-8\" standalone=\"no\"?>\n",
-       "<!DOCTYPE svg PUBLIC \"-//W3C//DTD SVG 1.1//EN\"\n",
-       " \"http://www.w3.org/Graphics/SVG/1.1/DTD/svg11.dtd\">\n",
-       "<!-- Generated by graphviz version 9.0.0 (20230911.1827)\n",
-       " -->\n",
-       "<!-- Pages: 1 -->\n",
-       "<svg width=\"201pt\" height=\"143pt\"\n",
-       " viewBox=\"0.00 0.00 200.50 142.50\" xmlns=\"http://www.w3.org/2000/svg\" xmlns:xlink=\"http://www.w3.org/1999/xlink\">\n",
-       "<g id=\"graph0\" class=\"graph\" transform=\"scale(1 1) rotate(0) translate(4 138.5)\">\n",
-       "<polygon fill=\"white\" stroke=\"none\" points=\"-4,4 -4,-138.5 196.5,-138.5 196.5,4 -4,4\"/>\n",
-       "<!-- 0 -->\n",
-       "<g id=\"node1\" class=\"node\">\n",
-       "<title>0</title>\n",
-       "<polygon fill=\"lightgrey\" stroke=\"black\" points=\"86.5,-134.5 0,-134.5 0,-93.5 86.5,-93.5 86.5,-134.5\"/>\n",
-       "<text text-anchor=\"middle\" x=\"43.25\" y=\"-117.2\" font-family=\"Times,serif\" font-size=\"14.00\">Initial cohort</text>\n",
-       "<text text-anchor=\"middle\" x=\"43.25\" y=\"-100.7\" font-family=\"Times,serif\" font-size=\"14.00\"> (n=101766)</text>\n",
-       "</g>\n",
-       "<!-- 1 -->\n",
-       "<g id=\"node2\" class=\"node\">\n",
-       "<title>1</title>\n",
-       "<polygon fill=\"lightgrey\" stroke=\"black\" points=\"78.62,-41 7.88,-41 7.88,0 78.62,0 78.62,-41\"/>\n",
-       "<text text-anchor=\"middle\" x=\"43.25\" y=\"-23.7\" font-family=\"Times,serif\" font-size=\"14.00\">Cohort 1</text>\n",
-       "<text text-anchor=\"middle\" x=\"43.25\" y=\"-7.2\" font-family=\"Times,serif\" font-size=\"14.00\"> (n=1000)</text>\n",
-       "</g>\n",
-       "<!-- 0&#45;&gt;1 -->\n",
-       "<g id=\"edge1\" class=\"edge\">\n",
-       "<title>0&#45;&gt;1</title>\n",
-       "<path fill=\"none\" stroke=\"black\" d=\"M43.25,-93.15C43.25,-81.28 43.25,-65.94 43.25,-52.5\"/>\n",
-       "<polygon fill=\"black\" stroke=\"black\" points=\"46.75,-52.83 43.25,-42.83 39.75,-52.83 46.75,-52.83\"/>\n",
-       "<text text-anchor=\"middle\" x=\"117.88\" y=\"-62.2\" font-family=\"Times,serif\" font-size=\"14.00\">filtered to first 1000 entries</text>\n",
-       "</g>\n",
-       "</g>\n",
-       "</svg>\n"
-      ],
-      "text/plain": [
-       "<graphviz.graphs.Digraph at 0x2c950aa50>"
-      ]
-     },
-     "execution_count": 127,
-     "metadata": {},
-     "output_type": "execute_result"
-    }
-   ],
-   "source": [
-    "pop_track.plot_flowchart()"
-   ]
-  },
-  {
-   "cell_type": "code",
-   "execution_count": 107,
-   "metadata": {},
-   "outputs": [
-    {
-     "data": {
-      "text/plain": [
-       "{'gender': {0.0: [53.8, 52.4], 1.0: [46.2, 47.6]},\n",
-       " 'race': {'AfricanAmerican': [19.3, 25.9],\n",
-       "  'Asian': [0.6, 0.7],\n",
-       "  'Caucasian': [76.5, 70.4],\n",
-       "  'Hispanic': [2.0, 0.8],\n",
-       "  'Other': [1.5, 2.2]},\n",
-       " 'weight': {'>200': [0.1, 0],\n",
-       "  '[0-25)': [1.5, 0],\n",
-       "  '[100-125)': [19.5, 0],\n",
-       "  '[125-150)': [4.5, 0],\n",
-       "  '[150-175)': [1.1, 0],\n",
-       "  '[175-200)': [0.3, 0],\n",
-       "  '[25-50)': [3.0, 0],\n",
-       "  '[50-75)': [28.1, 0],\n",
-       "  '[75-100)': [41.8, 0]},\n",
-       " 'time_in_hospital_days': ['4.4 (3.0)', '4.6 (3.2)']}"
-      ]
-     },
-     "execution_count": 107,
-     "metadata": {},
-     "output_type": "execute_result"
-    }
-   ],
-   "source": [
-    "pop_track.track"
-   ]
-  },
-  {
-   "cell_type": "code",
-   "execution_count": 116,
-   "metadata": {},
-   "outputs": [
-    {
-     "data": {
-      "image/png": "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",
-      "text/plain": [
-       "<Figure size 700x700 with 2 Axes>"
-      ]
-     },
-     "metadata": {},
-     "output_type": "display_data"
-    }
-   ],
-   "source": [
-    "pop_track.plot_population_change()"
-   ]
-  },
-  {
-   "cell_type": "code",
-   "execution_count": 99,
-   "metadata": {},
-   "outputs": [
-    {
-     "data": {
-      "text/plain": [
-       "{'gender': {0.0: [53.8, 52.4], 1.0: [46.2, 47.6]},\n",
-       " 'race': {'AfricanAmerican': [19.3, 25.9],\n",
-       "  'Asian': [0.6, 0.7],\n",
-       "  'Caucasian': [76.5, 70.4],\n",
-       "  'Hispanic': [2.0, 0.8],\n",
-       "  'Other': [1.5, 2.2]},\n",
-       " 'weight': {'>200': [0.1, 0],\n",
-       "  '[0-25)': [1.5, 0],\n",
-       "  '[100-125)': [19.5, 0],\n",
-       "  '[125-150)': [4.5, 0],\n",
-       "  '[150-175)': [1.1, 0],\n",
-       "  '[175-200)': [0.3, 0],\n",
-       "  '[25-50)': [3.0, 0],\n",
-       "  '[50-75)': [28.1, 0],\n",
-       "  '[75-100)': [41.8, 0]},\n",
-       " 'time_in_hospital_days': ['4.4 (3.0)', '4.6 (3.2)']}"
-      ]
-     },
-     "execution_count": 99,
-     "metadata": {},
-     "output_type": "execute_result"
-    }
-   ],
-   "source": [
-    "pop_track.track"
-   ]
-  },
-  {
-   "cell_type": "code",
-   "execution_count": 97,
-   "metadata": {},
-   "outputs": [
-    {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      "\u001b[1;35m2024-02-14 11:56:20,020\u001b[0m - \u001b[1;34mroot\u001b[0m \u001b[1;37mINFO - Transformed passed DataFrame into an AnnData object with n_obs x n_vars = `101766` x `47`.\u001b[0m\n"
-     ]
-    },
-    {
-     "ename": "KeyError",
-     "evalue": "0",
-     "output_type": "error",
-     "traceback": [
-      "\u001b[0;31m---------------------------------------------------------------------------\u001b[0m",
-      "\u001b[0;31mValueError\u001b[0m                                Traceback (most recent call last)",
-      "File \u001b[0;32m~/Documents/ehrapy_clean/ehrapy_venv_feb/lib/python3.11/site-packages/pandas/core/indexes/range.py:413\u001b[0m, in \u001b[0;36mRangeIndex.get_loc\u001b[0;34m(self, key)\u001b[0m\n\u001b[1;32m    412\u001b[0m \u001b[38;5;28;01mtry\u001b[39;00m:\n\u001b[0;32m--> 413\u001b[0m     \u001b[38;5;28;01mreturn\u001b[39;00m \u001b[38;5;28;43mself\u001b[39;49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43m_range\u001b[49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43mindex\u001b[49m\u001b[43m(\u001b[49m\u001b[43mnew_key\u001b[49m\u001b[43m)\u001b[49m\n\u001b[1;32m    414\u001b[0m \u001b[38;5;28;01mexcept\u001b[39;00m \u001b[38;5;167;01mValueError\u001b[39;00m \u001b[38;5;28;01mas\u001b[39;00m err:\n",
-      "\u001b[0;31mValueError\u001b[0m: 0 is not in range",
-      "\nThe above exception was the direct cause of the following exception:\n",
-      "\u001b[0;31mKeyError\u001b[0m                                  Traceback (most recent call last)",
-      "Cell \u001b[0;32mIn[97], line 7\u001b[0m\n\u001b[1;32m      5\u001b[0m adata \u001b[38;5;241m=\u001b[39m adata[:\u001b[38;5;241m1000\u001b[39m]\n\u001b[1;32m      6\u001b[0m pop_track(adata, label\u001b[38;5;241m=\u001b[39m\u001b[38;5;124m\"\u001b[39m\u001b[38;5;124mFiltered cohort\u001b[39m\u001b[38;5;124m\"\u001b[39m, operations_done\u001b[38;5;241m=\u001b[39m\u001b[38;5;124m\"\u001b[39m\u001b[38;5;124mfiltered to first 1000 entries\u001b[39m\u001b[38;5;124m\"\u001b[39m)\n\u001b[0;32m----> 7\u001b[0m \u001b[43mpop_track\u001b[49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43mplot_population_change\u001b[49m\u001b[43m(\u001b[49m\u001b[43m)\u001b[49m\n",
-      "File \u001b[0;32m~/Documents/ehrapy_clean/ehrapy/ehrapy/tools/population_tracking/_pop_tracker.py:173\u001b[0m, in \u001b[0;36mPopulationTracker.plot_population_change\u001b[0;34m(self, set_axis_labels, save, return_plot)\u001b[0m\n\u001b[1;32m    171\u001b[0m \u001b[38;5;28;01mfor\u001b[39;00m idx, ax \u001b[38;5;129;01min\u001b[39;00m \u001b[38;5;28menumerate\u001b[39m(axes):\n\u001b[1;32m    172\u001b[0m     \u001b[38;5;28;01mfor\u001b[39;00m pos, (_cols, data) \u001b[38;5;129;01min\u001b[39;00m \u001b[38;5;28menumerate\u001b[39m(\u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39mtrack\u001b[38;5;241m.\u001b[39mitems()):\n\u001b[0;32m--> 173\u001b[0m         data \u001b[38;5;241m=\u001b[39m \u001b[43mpd\u001b[49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43mDataFrame\u001b[49m\u001b[43m(\u001b[49m\u001b[43mdata\u001b[49m\u001b[43m)\u001b[49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43mloc\u001b[49m\u001b[43m[\u001b[49m\u001b[43midx\u001b[49m\u001b[43m]\u001b[49m\n\u001b[1;32m    175\u001b[0m         cumwidth \u001b[38;5;241m=\u001b[39m \u001b[38;5;241m0\u001b[39m\n\u001b[1;32m    177\u001b[0m         \u001b[38;5;66;03m# Adjust the hue shift based on the category position such that the colors are more distinguishable\u001b[39;00m\n",
-      "File \u001b[0;32m~/Documents/ehrapy_clean/ehrapy_venv_feb/lib/python3.11/site-packages/pandas/core/indexing.py:1192\u001b[0m, in \u001b[0;36m_LocationIndexer.__getitem__\u001b[0;34m(self, key)\u001b[0m\n\u001b[1;32m   1190\u001b[0m maybe_callable \u001b[38;5;241m=\u001b[39m com\u001b[38;5;241m.\u001b[39mapply_if_callable(key, \u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39mobj)\n\u001b[1;32m   1191\u001b[0m maybe_callable \u001b[38;5;241m=\u001b[39m \u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39m_check_deprecated_callable_usage(key, maybe_callable)\n\u001b[0;32m-> 1192\u001b[0m \u001b[38;5;28;01mreturn\u001b[39;00m \u001b[38;5;28;43mself\u001b[39;49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43m_getitem_axis\u001b[49m\u001b[43m(\u001b[49m\u001b[43mmaybe_callable\u001b[49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[43maxis\u001b[49m\u001b[38;5;241;43m=\u001b[39;49m\u001b[43maxis\u001b[49m\u001b[43m)\u001b[49m\n",
-      "File \u001b[0;32m~/Documents/ehrapy_clean/ehrapy_venv_feb/lib/python3.11/site-packages/pandas/core/indexing.py:1432\u001b[0m, in \u001b[0;36m_LocIndexer._getitem_axis\u001b[0;34m(self, key, axis)\u001b[0m\n\u001b[1;32m   1430\u001b[0m \u001b[38;5;66;03m# fall thru to straight lookup\u001b[39;00m\n\u001b[1;32m   1431\u001b[0m \u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39m_validate_key(key, axis)\n\u001b[0;32m-> 1432\u001b[0m \u001b[38;5;28;01mreturn\u001b[39;00m \u001b[38;5;28;43mself\u001b[39;49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43m_get_label\u001b[49m\u001b[43m(\u001b[49m\u001b[43mkey\u001b[49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[43maxis\u001b[49m\u001b[38;5;241;43m=\u001b[39;49m\u001b[43maxis\u001b[49m\u001b[43m)\u001b[49m\n",
-      "File \u001b[0;32m~/Documents/ehrapy_clean/ehrapy_venv_feb/lib/python3.11/site-packages/pandas/core/indexing.py:1382\u001b[0m, in \u001b[0;36m_LocIndexer._get_label\u001b[0;34m(self, label, axis)\u001b[0m\n\u001b[1;32m   1380\u001b[0m \u001b[38;5;28;01mdef\u001b[39;00m \u001b[38;5;21m_get_label\u001b[39m(\u001b[38;5;28mself\u001b[39m, label, axis: AxisInt):\n\u001b[1;32m   1381\u001b[0m     \u001b[38;5;66;03m# GH#5567 this will fail if the label is not present in the axis.\u001b[39;00m\n\u001b[0;32m-> 1382\u001b[0m     \u001b[38;5;28;01mreturn\u001b[39;00m \u001b[38;5;28;43mself\u001b[39;49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43mobj\u001b[49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43mxs\u001b[49m\u001b[43m(\u001b[49m\u001b[43mlabel\u001b[49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[43maxis\u001b[49m\u001b[38;5;241;43m=\u001b[39;49m\u001b[43maxis\u001b[49m\u001b[43m)\u001b[49m\n",
-      "File \u001b[0;32m~/Documents/ehrapy_clean/ehrapy_venv_feb/lib/python3.11/site-packages/pandas/core/generic.py:4295\u001b[0m, in \u001b[0;36mNDFrame.xs\u001b[0;34m(self, key, axis, level, drop_level)\u001b[0m\n\u001b[1;32m   4293\u001b[0m             new_index \u001b[38;5;241m=\u001b[39m index[loc]\n\u001b[1;32m   4294\u001b[0m \u001b[38;5;28;01melse\u001b[39;00m:\n\u001b[0;32m-> 4295\u001b[0m     loc \u001b[38;5;241m=\u001b[39m \u001b[43mindex\u001b[49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43mget_loc\u001b[49m\u001b[43m(\u001b[49m\u001b[43mkey\u001b[49m\u001b[43m)\u001b[49m\n\u001b[1;32m   4297\u001b[0m     \u001b[38;5;28;01mif\u001b[39;00m \u001b[38;5;28misinstance\u001b[39m(loc, np\u001b[38;5;241m.\u001b[39mndarray):\n\u001b[1;32m   4298\u001b[0m         \u001b[38;5;28;01mif\u001b[39;00m loc\u001b[38;5;241m.\u001b[39mdtype \u001b[38;5;241m==\u001b[39m np\u001b[38;5;241m.\u001b[39mbool_:\n",
-      "File \u001b[0;32m~/Documents/ehrapy_clean/ehrapy_venv_feb/lib/python3.11/site-packages/pandas/core/indexes/range.py:415\u001b[0m, in \u001b[0;36mRangeIndex.get_loc\u001b[0;34m(self, key)\u001b[0m\n\u001b[1;32m    413\u001b[0m         \u001b[38;5;28;01mreturn\u001b[39;00m \u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39m_range\u001b[38;5;241m.\u001b[39mindex(new_key)\n\u001b[1;32m    414\u001b[0m     \u001b[38;5;28;01mexcept\u001b[39;00m \u001b[38;5;167;01mValueError\u001b[39;00m \u001b[38;5;28;01mas\u001b[39;00m err:\n\u001b[0;32m--> 415\u001b[0m         \u001b[38;5;28;01mraise\u001b[39;00m \u001b[38;5;167;01mKeyError\u001b[39;00m(key) \u001b[38;5;28;01mfrom\u001b[39;00m \u001b[38;5;21;01merr\u001b[39;00m\n\u001b[1;32m    416\u001b[0m \u001b[38;5;28;01mif\u001b[39;00m \u001b[38;5;28misinstance\u001b[39m(key, Hashable):\n\u001b[1;32m    417\u001b[0m     \u001b[38;5;28;01mraise\u001b[39;00m \u001b[38;5;167;01mKeyError\u001b[39;00m(key)\n",
-      "\u001b[0;31mKeyError\u001b[0m: 0"
-     ]
-    },
-    {
-     "data": {
-      "image/png": "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",
-      "text/plain": [
-       "<Figure size 700x700 with 2 Axes>"
-      ]
-     },
-     "metadata": {},
-     "output_type": "display_data"
-    }
-   ],
-   "source": [
-    "adata = ep.dt.diabetes_130(columns_obs_only=[\"gender\", \"race\", \"weight\", \"time_in_hospital_days\"])\n",
-    "adata.obs.gender = adata.obs.gender.astype(\"category\")\n",
-    "pop_track = ep.tl.PopulationTracker(adata, categorical=[\"gender\", \"race\", \"weight\"])\n",
-    "pop_track(adata, label=\"Baseline cohort\")\n",
-    "adata = adata[:1000]\n",
-    "pop_track(adata, label=\"Filtered cohort\", operations_done=\"filtered to first 1000 entries\")\n",
-    "pop_track.plot_population_change()"
-   ]
-  },
-  {
-   "cell_type": "code",
-   "execution_count": 187,
-   "metadata": {},
-   "outputs": [
-    {
-     "ename": "AttributeError",
-     "evalue": "'Digraph' object has no attribute 'savefig'",
-     "output_type": "error",
-     "traceback": [
-      "\u001b[0;31m---------------------------------------------------------------------------\u001b[0m",
-      "\u001b[0;31mAttributeError\u001b[0m                            Traceback (most recent call last)",
-      "Cell \u001b[0;32mIn[187], line 1\u001b[0m\n\u001b[0;32m----> 1\u001b[0m \u001b[43mdot\u001b[49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43msavefig\u001b[49m(\u001b[38;5;124m\"\u001b[39m\u001b[38;5;124mflowchart.png\u001b[39m\u001b[38;5;124m\"\u001b[39m)\n",
-      "\u001b[0;31mAttributeError\u001b[0m: 'Digraph' object has no attribute 'savefig'"
-     ]
-    }
-   ],
-   "source": [
-    "dot.savefig(\"flowchart.png\")"
-   ]
-  },
-  {
-   "cell_type": "code",
-   "execution_count": 117,
-   "metadata": {},
-   "outputs": [
-    {
-     "data": {
-      "image/png": "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",
-      "text/latex": [
-       "$\\displaystyle 2$"
-      ],
-      "text/plain": [
-       "2"
-      ]
-     },
-     "execution_count": 117,
-     "metadata": {},
-     "output_type": "execute_result"
-    }
-   ],
-   "source": [
-    "pl.tracked_steps"
-   ]
-  },
-  {
-   "cell_type": "code",
-   "execution_count": 120,
-   "metadata": {},
-   "outputs": [
-    {
-     "data": {
-      "text/plain": [
-       "[None, 'filtered nothing lol']"
-      ]
-     },
-     "execution_count": 120,
-     "metadata": {},
-     "output_type": "execute_result"
-    }
-   ],
-   "source": [
-    "pl._tracked_operations"
-   ]
-  },
-  {
-   "cell_type": "code",
-   "execution_count": 136,
-   "metadata": {},
-   "outputs": [
-    {
-     "data": {
-      "image/svg+xml": [
-       "<?xml version=\"1.0\" encoding=\"UTF-8\" standalone=\"no\"?>\n",
-       "<!DOCTYPE svg PUBLIC \"-//W3C//DTD SVG 1.1//EN\"\n",
-       " \"http://www.w3.org/Graphics/SVG/1.1/DTD/svg11.dtd\">\n",
-       "<!-- Generated by graphviz version 9.0.0 (20230911.1827)\n",
-       " -->\n",
-       "<!-- Pages: 1 -->\n",
-       "<svg width=\"176pt\" height=\"143pt\"\n",
-       " viewBox=\"0.00 0.00 175.75 142.50\" xmlns=\"http://www.w3.org/2000/svg\" xmlns:xlink=\"http://www.w3.org/1999/xlink\">\n",
-       "<g id=\"graph0\" class=\"graph\" transform=\"scale(1 1) rotate(0) translate(4 138.5)\">\n",
-       "<polygon fill=\"white\" stroke=\"none\" points=\"-4,4 -4,-138.5 171.75,-138.5 171.75,4 -4,4\"/>\n",
-       "<!-- 0 -->\n",
-       "<g id=\"node1\" class=\"node\">\n",
-       "<title>0</title>\n",
-       "<polygon fill=\"lightgrey\" stroke=\"black\" points=\"64,-134.5 0,-134.5 0,-93.5 64,-93.5 64,-134.5\"/>\n",
-       "<text text-anchor=\"middle\" x=\"32\" y=\"-117.2\" font-family=\"Times,serif\" font-size=\"14.00\">Step 0</text>\n",
-       "<text text-anchor=\"middle\" x=\"32\" y=\"-100.7\" font-family=\"Times,serif\" font-size=\"14.00\"> (n=150)</text>\n",
-       "</g>\n",
-       "<!-- 1 -->\n",
-       "<g id=\"node2\" class=\"node\">\n",
-       "<title>1</title>\n",
-       "<polygon fill=\"lightgrey\" stroke=\"black\" points=\"60.62,-41 3.38,-41 3.38,0 60.62,0 60.62,-41\"/>\n",
-       "<text text-anchor=\"middle\" x=\"32\" y=\"-23.7\" font-family=\"Times,serif\" font-size=\"14.00\">filtered</text>\n",
-       "<text text-anchor=\"middle\" x=\"32\" y=\"-7.2\" font-family=\"Times,serif\" font-size=\"14.00\"> (n=75)</text>\n",
-       "</g>\n",
-       "<!-- 0&#45;&gt;1 -->\n",
-       "<g id=\"edge1\" class=\"edge\">\n",
-       "<title>0&#45;&gt;1</title>\n",
-       "<path fill=\"none\" stroke=\"black\" d=\"M32,-93.15C32,-81.28 32,-65.94 32,-52.5\"/>\n",
-       "<polygon fill=\"black\" stroke=\"black\" points=\"35.5,-52.83 32,-42.83 28.5,-52.83 35.5,-52.83\"/>\n",
-       "<text text-anchor=\"middle\" x=\"99.88\" y=\"-62.2\" font-family=\"Times,serif\" font-size=\"14.00\">filtered to first 75 entries</text>\n",
-       "</g>\n",
-       "</g>\n",
-       "</svg>\n"
-      ],
-      "text/plain": [
-       "<graphviz.graphs.Digraph at 0x2b37649d0>"
-      ]
-     },
-     "execution_count": 136,
-     "metadata": {},
-     "output_type": "execute_result"
-    }
-   ],
-   "source": [
-    "pl.plot_flowchart()"
-   ]
-  },
-  {
-   "cell_type": "markdown",
-   "metadata": {},
-   "source": []
-  },
-  {
-   "cell_type": "code",
-   "execution_count": 20,
-   "metadata": {},
-   "outputs": [
-    {
-     "data": {
-      "text/html": [
-       "<div>\n",
-       "<style scoped>\n",
-       "    .dataframe tbody tr th:only-of-type {\n",
-       "        vertical-align: middle;\n",
-       "    }\n",
-       "\n",
-       "    .dataframe tbody tr th {\n",
-       "        vertical-align: top;\n",
-       "    }\n",
-       "\n",
-       "    .dataframe thead th {\n",
-       "        text-align: right;\n",
-       "    }\n",
-       "</style>\n",
-       "<table border=\"1\" class=\"dataframe\">\n",
-       "  <thead>\n",
-       "    <tr style=\"text-align: right;\">\n",
-       "      <th></th>\n",
-       "      <th></th>\n",
-       "      <th>Missing</th>\n",
-       "      <th>Overall</th>\n",
-       "    </tr>\n",
-       "  </thead>\n",
-       "  <tbody>\n",
-       "    <tr>\n",
-       "      <th>n</th>\n",
-       "      <th></th>\n",
-       "      <td></td>\n",
-       "      <td>150</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th rowspan=\"3\" valign=\"top\">ethnicity, n (%)</th>\n",
-       "      <th>asian</th>\n",
-       "      <td>0</td>\n",
-       "      <td>43 (28.7)</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>black</th>\n",
-       "      <td></td>\n",
-       "      <td>52 (34.7)</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>white</th>\n",
-       "      <td></td>\n",
-       "      <td>55 (36.7)</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th rowspan=\"2\" valign=\"top\">gender, n (%)</th>\n",
-       "      <th>female</th>\n",
-       "      <td>0</td>\n",
-       "      <td>66 (44.0)</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>male</th>\n",
-       "      <td></td>\n",
-       "      <td>84 (56.0)</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th rowspan=\"2\" valign=\"top\">proficiency, n (%)</th>\n",
-       "      <th>high prof.</th>\n",
-       "      <td>0</td>\n",
-       "      <td>78 (52.0)</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>limited prof.</th>\n",
-       "      <td></td>\n",
-       "      <td>72 (48.0)</td>\n",
-       "    </tr>\n",
-       "  </tbody>\n",
-       "</table>\n",
-       "</div><br />"
-      ],
-      "text/plain": [
-       "                                 Missing    Overall\n",
-       "n                                               150\n",
-       "ethnicity, n (%)   asian               0  43 (28.7)\n",
-       "                   black                  52 (34.7)\n",
-       "                   white                  55 (36.7)\n",
-       "gender, n (%)      female              0  66 (44.0)\n",
-       "                   male                   84 (56.0)\n",
-       "proficiency, n (%) high prof.          0  78 (52.0)\n",
-       "                   limited prof.          72 (48.0)"
-      ]
-     },
-     "execution_count": 20,
-     "metadata": {},
-     "output_type": "execute_result"
-    }
-   ],
-   "source": [
-    "t1 = TableOne(random_dataset, columns=[\"ethnicity\", \"gender\", \"proficiency\"])\n",
-    "t1"
-   ]
-  },
-  {
-   "cell_type": "code",
-   "execution_count": 151,
-   "metadata": {},
-   "outputs": [
-    {
-     "data": {
-      "text/plain": [
-       "{'asian': [28.7, 24.0],\n",
-       " 'black': [34.7, 34.0],\n",
-       " 'white': [36.7, 42.0],\n",
-       " 'male': [56.0, 56.0],\n",
-       " 'female': [44.0, 44.0],\n",
-       " 'high prof.': [52.0, 48.0],\n",
-       " 'limited prof.': [48.0, 52.0]}"
-      ]
-     },
-     "execution_count": 151,
-     "metadata": {},
-     "output_type": "execute_result"
-    }
-   ],
-   "source": [
-    "collated_data = {key: values1 + values2 for key, values1, values2 in zip(dpcts.keys(), dpcts.values(), dcpts2.values())}\n",
-    "collated_data"
-   ]
-  },
-  {
-   "cell_type": "code",
-   "execution_count": 155,
-   "metadata": {},
-   "outputs": [
-    {
-     "data": {
-      "text/plain": [
-       "{'asian': [28.7, 24.0], 'black': [34.7, 34.0], 'white': [36.7, 42.0]}"
-      ]
-     },
-     "execution_count": 155,
-     "metadata": {},
-     "output_type": "execute_result"
-    }
-   ],
-   "source": [
-    "data_ethnicity = {key: value for key, value in collated_data.items() if key in [\"asian\", \"black\", \"white\"]}\n",
-    "data_ethnicity"
-   ]
-  },
-  {
-   "cell_type": "code",
-   "execution_count": 191,
-   "metadata": {},
-   "outputs": [
-    {
-     "data": {
-      "text/plain": [
-       "{'male': [56.0, 56.0], 'female': [44.0, 44.0]}"
-      ]
-     },
-     "execution_count": 191,
-     "metadata": {},
-     "output_type": "execute_result"
-    }
-   ],
-   "source": [
-    "data_gender = {key: value for key, value in collated_data.items() if key in [\"male\", \"female\"]}\n",
-    "data_gender"
-   ]
-  },
-  {
-   "cell_type": "code",
-   "execution_count": 248,
-   "metadata": {},
-   "outputs": [
-    {
-     "data": {
-      "text/plain": [
-       "{'high prof.': [52.0, 48.0], 'limited prof.': [48.0, 52.0]}"
-      ]
-     },
-     "execution_count": 248,
-     "metadata": {},
-     "output_type": "execute_result"
-    }
-   ],
-   "source": [
-    "data_prof = {key: value for key, value in collated_data.items() if key in [\"high prof.\", \"limited prof.\"]}\n",
-    "data_prof"
-   ]
-  },
-  {
-   "cell_type": "code",
-   "execution_count": 211,
-   "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>asian</th>\n",
-       "      <th>black</th>\n",
-       "      <th>white</th>\n",
-       "    </tr>\n",
-       "  </thead>\n",
-       "  <tbody>\n",
-       "    <tr>\n",
-       "      <th>0</th>\n",
-       "      <td>28.7</td>\n",
-       "      <td>34.7</td>\n",
-       "      <td>36.7</td>\n",
-       "    </tr>\n",
-       "  </tbody>\n",
-       "</table>\n",
-       "</div>"
-      ],
-      "text/plain": [
-       "   asian  black  white\n",
-       "0   28.7   34.7   36.7"
-      ]
-     },
-     "execution_count": 211,
-     "metadata": {},
-     "output_type": "execute_result"
-    }
-   ],
-   "source": [
-    "df_ethnicity"
-   ]
-  },
-  {
-   "cell_type": "code",
-   "execution_count": 242,
-   "metadata": {},
-   "outputs": [
-    {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      "28.7\n",
-      "34.7\n",
-      "36.7\n"
-     ]
-    }
-   ],
-   "source": [
-    "for v in l.iterrows():\n",
-    "    print(v[1].values[0])"
-   ]
-  },
-  {
-   "cell_type": "code",
-   "execution_count": 80,
-   "metadata": {},
-   "outputs": [
-    {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      "28.7\n",
-      "34.7\n",
-      "36.7\n",
-      "56.0\n",
-      "44.0\n",
-      "52.0\n",
-      "48.0\n",
-      "24.0\n",
-      "34.0\n",
-      "42.0\n",
-      "56.0\n",
-      "44.0\n",
-      "48.0\n",
-      "52.0\n"
-     ]
-    },
-    {
-     "data": {
-      "image/png": "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",
-      "text/plain": [
-       "<Figure size 700x700 with 2 Axes>"
-      ]
-     },
-     "metadata": {},
-     "output_type": "display_data"
-    }
-   ],
-   "source": [
-    "import pandas as pd\n",
-    "import matplotlib.pyplot as plt\n",
-    "import seaborn as sns\n",
-    "\n",
-    "# Given data dictionaries\n",
-    "data_ethnicity = {\"asian\": [28.7, 24.0], \"black\": [34.7, 34.0], \"white\": [36.7, 42.0]}\n",
-    "\n",
-    "data_gender = {\"male\": [56.0, 56.0], \"female\": [44.0, 44.0]}\n",
-    "\n",
-    "data_prof = {\"high prof.\": [52.0, 48.0], \"limited prof.\": [48.0, 52.0]}\n",
-    "\n",
-    "# Convert dictionaries to DataFrames\n",
-    "df_ethnicity = pd.DataFrame(data_ethnicity)\n",
-    "df_gender = pd.DataFrame(data_gender)\n",
-    "df_prof = pd.DataFrame(data_prof)\n",
-    "\n",
-    "# Plotting\n",
-    "fig, axes = plt.subplots(2, 1, figsize=(7, 7))\n",
-    "\n",
-    "# Plotting for the ethnicity with a small gap\n",
-    "categories = [df_ethnicity, df_gender, df_prof]\n",
-    "legend_labels = []\n",
-    "\n",
-    "for idx, ax in enumerate(axes):\n",
-    "    for pos, data in enumerate([df_ethnicity.loc[idx], df_gender.loc[idx], df_prof.loc[idx]]):\n",
-    "        cumwidth = 0\n",
-    "        category = categories[pos]\n",
-    "\n",
-    "        hue_shift = (pos + 1) / len(categories[pos])  # Adjust the hue shift based on the category position\n",
-    "        colors = sns.color_palette(\"husl\", len(category.columns))\n",
-    "        adjusted_colors = [((color[0] + hue_shift) % 1, color[1], color[2]) for color in colors]\n",
-    "\n",
-    "        for i, value in enumerate(data):\n",
-    "            print(value)\n",
-    "            # value = value[1].values[0]  # Take the value based on idx\n",
-    "            ax.barh(pos, value, left=cumwidth, color=adjusted_colors[i], height=0.8)\n",
-    "\n",
-    "            if value > 5:\n",
-    "                # Add data labels\n",
-    "                width = value\n",
-    "                ax.text(\n",
-    "                    cumwidth + width / 2,\n",
-    "                    pos,\n",
-    "                    \"{:.1f}\".format(value),\n",
-    "                    ha=\"center\",\n",
-    "                    va=\"center\",\n",
-    "                    color=\"white\",\n",
-    "                    fontweight=\"bold\",\n",
-    "                )\n",
-    "\n",
-    "            ax.set_yticks([])\n",
-    "            ax.set_xticks([])\n",
-    "            cumwidth += value\n",
-    "\n",
-    "            legend_labels.append(category.columns[i])\n",
-    "\n",
-    "# makes the frames invisible\n",
-    "# for ax in axes:\n",
-    "#     ax.axis('off')\n",
-    "\n",
-    "\n",
-    "# Add legend for the first subplot\n",
-    "plt.legend(legend_labels, loc=\"best\", bbox_to_anchor=(-0.5, 1))  # ncol=len(legend_labels), bbox_to_anchor=(0.5, -0.3)\n",
-    "\n",
-    "plt.tight_layout()\n",
-    "plt.show()"
-   ]
-  },
-  {
-   "cell_type": "code",
-   "execution_count": 369,
-   "metadata": {},
-   "outputs": [
-    {
-     "ename": "AttributeError",
-     "evalue": "'str' object has no attribute 'values'",
-     "output_type": "error",
-     "traceback": [
-      "\u001b[0;31m---------------------------------------------------------------------------\u001b[0m",
-      "\u001b[0;31mAttributeError\u001b[0m                            Traceback (most recent call last)",
-      "Cell \u001b[0;32mIn[369], line 37\u001b[0m\n\u001b[1;32m     34\u001b[0m adjusted_colors \u001b[38;5;241m=\u001b[39m [((color[\u001b[38;5;241m0\u001b[39m] \u001b[38;5;241m+\u001b[39m hue_shift) \u001b[38;5;241m%\u001b[39m \u001b[38;5;241m1\u001b[39m, color[\u001b[38;5;241m1\u001b[39m], color[\u001b[38;5;241m2\u001b[39m]) \u001b[38;5;28;01mfor\u001b[39;00m color \u001b[38;5;129;01min\u001b[39;00m colors]\n\u001b[1;32m     36\u001b[0m \u001b[38;5;28;01mfor\u001b[39;00m i, value \u001b[38;5;129;01min\u001b[39;00m \u001b[38;5;28menumerate\u001b[39m(data):\n\u001b[0;32m---> 37\u001b[0m     value \u001b[38;5;241m=\u001b[39m \u001b[43mvalue\u001b[49m\u001b[43m[\u001b[49m\u001b[38;5;241;43m1\u001b[39;49m\u001b[43m]\u001b[49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43mvalues\u001b[49m[idx]  \u001b[38;5;66;03m# Take the value based on idx\u001b[39;00m\n\u001b[1;32m     38\u001b[0m     ax\u001b[38;5;241m.\u001b[39mbarh(pos, value, left\u001b[38;5;241m=\u001b[39mcumwidth, color\u001b[38;5;241m=\u001b[39madjusted_colors[i], height\u001b[38;5;241m=\u001b[39m\u001b[38;5;241m0.8\u001b[39m)\n\u001b[1;32m     40\u001b[0m     \u001b[38;5;28;01mif\u001b[39;00m value \u001b[38;5;241m>\u001b[39m \u001b[38;5;241m5\u001b[39m:\n\u001b[1;32m     41\u001b[0m         \u001b[38;5;66;03m# Add data labels\u001b[39;00m\n",
-      "\u001b[0;31mAttributeError\u001b[0m: 'str' object has no attribute 'values'"
-     ]
-    },
-    {
-     "data": {
-      "image/png": "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",
-      "text/plain": [
-       "<Figure size 1000x1000 with 2 Axes>"
-      ]
-     },
-     "metadata": {
-      "image/png": {
-       "height": 819,
-       "width": 837
-      }
-     },
-     "output_type": "display_data"
-    }
-   ],
-   "source": [
-    "import pandas as pd\n",
-    "import matplotlib.pyplot as plt\n",
-    "import seaborn as sns\n",
-    "\n",
-    "# Given data dictionaries\n",
-    "data_ethnicity = {\"asian\": [28.7, 24.0], \"black\": [34.7, 34.0], \"white\": [36.7, 42.0]}\n",
-    "\n",
-    "data_gender = {\"male\": [56.0, 56.0], \"female\": [44.0, 44.0]}\n",
-    "\n",
-    "data_prof = {\"high prof.\": [52.0, 48.0], \"limited prof.\": [48.0, 52.0]}\n",
-    "\n",
-    "# Convert dictionaries to DataFrames\n",
-    "df_ethnicity = pd.DataFrame(data_ethnicity)\n",
-    "df_gender = pd.DataFrame(data_gender)\n",
-    "df_prof = pd.DataFrame(data_prof)\n",
-    "\n",
-    "# Plotting\n",
-    "fig, axes = plt.subplots(2, 1, figsize=(10, 10))\n",
-    "\n",
-    "# Plotting for the ethnicity with a small gap\n",
-    "categories = [df_ethnicity, df_gender, df_prof]\n",
-    "legend_labels = []\n",
-    "\n",
-    "for idx, ax in enumerate(axes):\n",
-    "    for pos, data in enumerate(categories[idx].T.iterrows()):\n",
-    "        cumwidth = 0\n",
-    "        category = categories[idx]\n",
-    "\n",
-    "        hue_shift = (pos + 1) / len(categories[idx])  # Adjust the hue shift based on the category position\n",
-    "        colors = sns.color_palette(\"husl\", len(category.columns))\n",
-    "        adjusted_colors = [((color[0] + hue_shift) % 1, color[1], color[2]) for color in colors]\n",
-    "\n",
-    "        for i, value in enumerate(data):\n",
-    "            value = value[1].values[idx]  # Take the value based on idx\n",
-    "            ax.barh(pos, value, left=cumwidth, color=adjusted_colors[i], height=0.8)\n",
-    "\n",
-    "            if value > 5:\n",
-    "                # Add data labels\n",
-    "                width = value\n",
-    "                ax.text(\n",
-    "                    cumwidth + width / 2,\n",
-    "                    pos,\n",
-    "                    \"{:.1f}\".format(value),\n",
-    "                    ha=\"center\",\n",
-    "                    va=\"center\",\n",
-    "                    color=\"white\",\n",
-    "                    fontweight=\"bold\",\n",
-    "                )\n",
-    "\n",
-    "            ax.set_yticks([])\n",
-    "            ax.set_xticks([])\n",
-    "            cumwidth += value\n",
-    "\n",
-    "            legend_labels.append(category.columns[i])\n",
-    "\n",
-    "# Set labels and title for the first subplot\n",
-    "axes[0].set_title(\"First Numbers\")\n",
-    "axes[0].set_xlabel(\"Percentage\")\n",
-    "axes[0].set_yticks(range(len(categories[0])))\n",
-    "axes[0].set_yticklabels(category_labels)\n",
-    "\n",
-    "# Set labels and title for the second subplot\n",
-    "axes[1].set_title(\"Second Numbers\")\n",
-    "axes[1].set_xlabel(\"Percentage\")\n",
-    "axes[1].set_yticks(range(len(categories[0])))\n",
-    "axes[1].set_yticklabels(category_labels)\n",
-    "\n",
-    "# Add legend for the first subplot\n",
-    "axes[0].legend(legend_labels, loc=\"lower center\", ncol=len(legend_labels), bbox_to_anchor=(0.5, -0.1))\n",
-    "\n",
-    "plt.tight_layout()\n",
-    "plt.show()"
-   ]
-  },
-  {
-   "cell_type": "code",
-   "execution_count": 343,
-   "metadata": {},
-   "outputs": [
-    {
-     "data": {
-      "image/svg+xml": [
-       "<?xml version=\"1.0\" encoding=\"UTF-8\" standalone=\"no\"?>\n",
-       "<!DOCTYPE svg PUBLIC \"-//W3C//DTD SVG 1.1//EN\"\n",
-       " \"http://www.w3.org/Graphics/SVG/1.1/DTD/svg11.dtd\">\n",
-       "<!-- Generated by graphviz version 9.0.0 (20230911.1827)\n",
-       " -->\n",
-       "<!-- Pages: 1 -->\n",
-       "<svg width=\"250pt\" height=\"143pt\"\n",
-       " viewBox=\"0.00 0.00 250.38 142.50\" xmlns=\"http://www.w3.org/2000/svg\" xmlns:xlink=\"http://www.w3.org/1999/xlink\">\n",
-       "<g id=\"graph0\" class=\"graph\" transform=\"scale(1 1) rotate(0) translate(4 138.5)\">\n",
-       "<polygon fill=\"white\" stroke=\"none\" points=\"-4,4 -4,-138.5 246.38,-138.5 246.38,4 -4,4\"/>\n",
-       "<!-- A -->\n",
-       "<g id=\"node1\" class=\"node\">\n",
-       "<title>A</title>\n",
-       "<polygon fill=\"lightgrey\" stroke=\"black\" points=\"80.12,-134.5 16.12,-134.5 16.12,-93.5 80.12,-93.5 80.12,-134.5\"/>\n",
-       "<text text-anchor=\"middle\" x=\"48.12\" y=\"-117.2\" font-family=\"Times,serif\" font-size=\"14.00\">Original</text>\n",
-       "<text text-anchor=\"middle\" x=\"48.12\" y=\"-100.7\" font-family=\"Times,serif\" font-size=\"14.00\"> (n=200)</text>\n",
-       "</g>\n",
-       "<!-- B -->\n",
-       "<g id=\"node2\" class=\"node\">\n",
-       "<title>B</title>\n",
-       "<polygon fill=\"lightgrey\" stroke=\"black\" points=\"96.25,-41 0,-41 0,0 96.25,0 96.25,-41\"/>\n",
-       "<text text-anchor=\"middle\" x=\"48.12\" y=\"-23.7\" font-family=\"Times,serif\" font-size=\"14.00\">Filtered values</text>\n",
-       "<text text-anchor=\"middle\" x=\"48.12\" y=\"-7.2\" font-family=\"Times,serif\" font-size=\"14.00\"> (n=100)</text>\n",
-       "</g>\n",
-       "<!-- A&#45;&gt;B -->\n",
-       "<g id=\"edge1\" class=\"edge\">\n",
-       "<title>A&#45;&gt;B</title>\n",
-       "<path fill=\"none\" stroke=\"black\" d=\"M48.12,-93.15C48.12,-81.28 48.12,-65.94 48.12,-52.5\"/>\n",
-       "<polygon fill=\"black\" stroke=\"black\" points=\"51.63,-52.83 48.13,-42.83 44.63,-52.83 51.63,-52.83\"/>\n",
-       "<text text-anchor=\"middle\" x=\"145.25\" y=\"-62.2\" font-family=\"Times,serif\" font-size=\"14.00\"> &#160;Filtering Process: Removes lalalla</text>\n",
-       "</g>\n",
-       "</g>\n",
-       "</svg>\n"
-      ],
-      "text/plain": [
-       "<graphviz.graphs.Digraph at 0x2b01d6a50>"
-      ]
-     },
-     "execution_count": 343,
-     "metadata": {},
-     "output_type": "execute_result"
-    }
-   ],
-   "source": [
-    "import graphviz\n",
-    "\n",
-    "# Create Digraph object\n",
-    "dot = graphviz.Digraph()\n",
-    "\n",
-    "# Define nodes (edgy nodes)\n",
-    "dot.node(name=\"A\", label=\"Original\\n (n=200)\", style=\"filled\", shape=\"box\")\n",
-    "dot.node(name=\"B\", label=\"Filtered values\\n (n=100)\", style=\"filled\", shape=\"box\")\n",
-    "\n",
-    "# Define edges (arrows) with labels\n",
-    "dot.edge(\"A\", \"B\", label=\"  Filtering Process: Removes lalalla\", labeldistance=\"10.5\")\n",
-    "\n",
-    "# Render the graph\n",
-    "dot.render(\"flow_diagram_edgy_nodes\", format=\"png\", cleanup=True)\n",
-    "\n",
-    "dot"
-   ]
-  },
-  {
-   "cell_type": "code",
-   "execution_count": 317,
-   "metadata": {},
-   "outputs": [
-    {
-     "ename": "UnidentifiedImageError",
-     "evalue": "cannot identify image file 'dot_figure.png'",
-     "output_type": "error",
-     "traceback": [
-      "\u001b[0;31m---------------------------------------------------------------------------\u001b[0m",
-      "\u001b[0;31mUnidentifiedImageError\u001b[0m                    Traceback (most recent call last)",
-      "Cell \u001b[0;32mIn[317], line 8\u001b[0m\n\u001b[1;32m      5\u001b[0m dot\u001b[38;5;241m.\u001b[39mrender(\u001b[38;5;124m'\u001b[39m\u001b[38;5;124mdot_figure.png\u001b[39m\u001b[38;5;124m'\u001b[39m, \u001b[38;5;28mformat\u001b[39m\u001b[38;5;241m=\u001b[39m\u001b[38;5;124m'\u001b[39m\u001b[38;5;124mpng\u001b[39m\u001b[38;5;124m'\u001b[39m)\n\u001b[1;32m      7\u001b[0m \u001b[38;5;66;03m# Convert Dot figure to image\u001b[39;00m\n\u001b[0;32m----> 8\u001b[0m dot_img \u001b[38;5;241m=\u001b[39m \u001b[43mImage\u001b[49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43mopen\u001b[49m\u001b[43m(\u001b[49m\u001b[38;5;124;43m'\u001b[39;49m\u001b[38;5;124;43mdot_figure.png\u001b[39;49m\u001b[38;5;124;43m'\u001b[39;49m\u001b[43m)\u001b[49m\n\u001b[1;32m     10\u001b[0m \u001b[38;5;66;03m# Create subplots for both figures\u001b[39;00m\n\u001b[1;32m     11\u001b[0m fig, axes \u001b[38;5;241m=\u001b[39m plt\u001b[38;5;241m.\u001b[39msubplots(\u001b[38;5;241m1\u001b[39m, \u001b[38;5;241m2\u001b[39m, figsize\u001b[38;5;241m=\u001b[39m(\u001b[38;5;241m15\u001b[39m, \u001b[38;5;241m5\u001b[39m))  \u001b[38;5;66;03m# Adjust figsize as needed\u001b[39;00m\n",
-      "File \u001b[0;32m~/Documents/ehrapy_clean/ehrapy_venv_feb/lib/python3.11/site-packages/PIL/Image.py:3309\u001b[0m, in \u001b[0;36mopen\u001b[0;34m(fp, mode, formats)\u001b[0m\n\u001b[1;32m   3307\u001b[0m     warnings\u001b[38;5;241m.\u001b[39mwarn(message)\n\u001b[1;32m   3308\u001b[0m msg \u001b[38;5;241m=\u001b[39m \u001b[38;5;124m\"\u001b[39m\u001b[38;5;124mcannot identify image file \u001b[39m\u001b[38;5;132;01m%r\u001b[39;00m\u001b[38;5;124m\"\u001b[39m \u001b[38;5;241m%\u001b[39m (filename \u001b[38;5;28;01mif\u001b[39;00m filename \u001b[38;5;28;01melse\u001b[39;00m fp)\n\u001b[0;32m-> 3309\u001b[0m \u001b[38;5;28;01mraise\u001b[39;00m UnidentifiedImageError(msg)\n",
-      "\u001b[0;31mUnidentifiedImageError\u001b[0m: cannot identify image file 'dot_figure.png'"
-     ]
-    }
-   ],
-   "source": [
-    "import matplotlib.pyplot as plt\n",
-    "from PIL import Image\n",
-    "\n",
-    "# Save Dot figure to a file\n",
-    "dot.render(\"dot_figure.png\", format=\"png\")\n",
-    "\n",
-    "# Convert Dot figure to image\n",
-    "dot_img = Image.open(\"dot_figure.png\")\n",
-    "\n",
-    "# Create subplots for both figures\n",
-    "fig, axes = plt.subplots(1, 2, figsize=(15, 5))  # Adjust figsize as needed\n",
-    "\n",
-    "# Plot Dot figure\n",
-    "axes[0].imshow(dot_img)\n",
-    "axes[0].axis(\"off\")\n",
-    "\n",
-    "# Plot Matplotlib figure with barplots\n",
-    "# Assuming `fig` contains your Matplotlib figure with barplots\n",
-    "axes[1].imshow(fig.canvas.renderer._renderer)\n",
-    "axes[1].axis(\"off\")\n",
-    "\n",
-    "# Adjust layout\n",
-    "plt.tight_layout()\n",
-    "\n",
-    "# Show the plots\n",
-    "plt.show()"
-   ]
-  }
- ],
- "metadata": {
-  "kernelspec": {
-   "display_name": "ehrapy_venv_feb",
-   "language": "python",
-   "name": "python3"
-  },
-  "language_info": {
-   "codemirror_mode": {
-    "name": "ipython",
-    "version": 3
-   },
-   "file_extension": ".py",
-   "mimetype": "text/x-python",
-   "name": "python",
-   "nbconvert_exporter": "python",
-   "pygments_lexer": "ipython3",
-   "version": "3.11.7"
-  }
- },
- "nbformat": 4,
- "nbformat_minor": 2
-}

From 9687e4d2c65cd39c16108a517d776c79edd624c6 Mon Sep 17 00:00:00 2001
From: Eljas Roellin <eljas.roellin@ikmail.com>
Date: Thu, 29 Feb 2024 17:28:50 +0100
Subject: [PATCH 07/46] small comments included

---
 .../tools/cohort_tracking/_cohort_tracker.py  | 43 ++++++++++---------
 .../cohort_tracking/test_cohort_tracking.py   |  3 +-
 2 files changed, 24 insertions(+), 22 deletions(-)

diff --git a/ehrapy/tools/cohort_tracking/_cohort_tracker.py b/ehrapy/tools/cohort_tracking/_cohort_tracker.py
index 17eb2c74..bd3faea6 100644
--- a/ehrapy/tools/cohort_tracking/_cohort_tracker.py
+++ b/ehrapy/tools/cohort_tracking/_cohort_tracker.py
@@ -1,4 +1,5 @@
 import copy
+from collections.abc import Iterable
 from typing import Any, Union
 
 import graphviz
@@ -10,10 +11,10 @@
 from tableone import TableOne
 
 
-def _check_columns_exist(df, columns):
-    if not all(col in df.columns for col in columns):
-        missing_columns = [col for col in columns if col not in df.columns]
-        raise ValueError(f"Columns {missing_columns} not found in dataframe.")
+def _check_columns_exist(df, columns) -> None:
+    missing_columns = set(columns) - set(df.columns)
+    if missing_columns:
+        raise ValueError(f"Columns {list(missing_columns)} not found in dataframe.")
 
 
 # from tableone: https://github.com/tompollard/tableone/blob/bfd6fbaa4ed3e9f59e1a75191c6296a2a80ccc64/tableone/tableone.py#L555
@@ -34,7 +35,9 @@ def _detect_categorical_columns(data) -> list:
 
 
 class CohortTracker:
-    def __init__(self, adata: AnnData | pd.DataFrame, columns: list = None, categorical: list = None, *args: Any):
+    def __init__(
+        self, adata: AnnData | pd.DataFrame, columns: Iterable = None, categorical: Iterable = None, *args: Any
+    ):
         """Track cohort changes over multiple filtering or processing steps.
 
         This class offers functionality to track and plot cohort changes over multiple filtering or processing steps,
@@ -43,8 +46,8 @@ def __init__(self, adata: AnnData | pd.DataFrame, columns: list = None, categori
         Tightly interacting with the `tableone` package [1].
         Args:
             adata: :class:`~anndata.AnnData` or :class:`~pandas.DataFrame` object to track.
-            columns: List of columns to track. If `None`, all columns will be tracked.
-            categorical: List of columns that contain categorical variables, if not given will be inferred from the data.
+            columns: Iterable of columns to track. If `None`, all columns will be tracked.
+            categorical: Iterable of columns that contain categorical variables, if not given will be inferred from the data.
 
         References
         ----------
@@ -80,7 +83,7 @@ def __init__(self, adata: AnnData | pd.DataFrame, columns: list = None, categori
 
     def __call__(
         self, adata: AnnData, label: str = None, operations_done: str = None, *args: Any, **tableone_kwargs: Any
-    ) -> Any:
+    ) -> None:
         if isinstance(adata, AnnData):
             df = adata.obs
         elif isinstance(adata, pd.DataFrame):
@@ -151,9 +154,9 @@ def plot_cohort_change(
         self,
         set_axis_labels=True,
         subfigure_title: bool = False,
-        sns_color_palette: str = "husl",
+        color_palette: str = "husl",
         save: str = None,
-        return_plot: bool = False,
+        return_figure: bool = False,
         subplots_kwargs: dict = None,
         legend_kwargs: dict = None,
     ):
@@ -164,16 +167,16 @@ def plot_cohort_change(
         Args:
             set_axis_labels: If `True`, the y-axis labels will be set to the column names.
             subfigure_title: If `True`, each subplot will have a title with the `label` provided during tracking.
-            sns_color_palette: The color palette to use for the plot. Default is "husl".
+            color_palette: The color palette to use for the plot. Default is "husl".
             save: If a string is provided, the plot will be saved to the path specified.
-            return_plot: If `True`, the plot will be returned as a tuple of (fig, ax).
+            return_figure: If `True`, the plot will be returned as a tuple of (fig, ax).
             subplot_kwargs: Additional keyword arguments for the subplots.
             legend_kwargs: Additional keyword arguments for the legend.
 
         Returns:
-            If `return_plot` a :class:`~matplotlib.figure.Figure` and a :class:`~matplotlib.axes.Axes` or a list of it.
+            If `return_figure` a :class:`~matplotlib.figure.Figure` and a :class:`~matplotlib.axes.Axes` or a list of it.
 
-        Example:
+        Examples:
             .. code-block:: python
 
                 import ehrapy as ep
@@ -211,7 +214,7 @@ def plot_cohort_change(
 
                 # Adjust the hue shift based on the category position such that the colors are more distinguishable
                 hue_shift = (pos + 1) / len(data)
-                colors = sns.color_palette(sns_color_palette, len(data))
+                colors = sns.color_palette(color_palette, len(data))
                 adjusted_colors = [((color[0] + hue_shift) % 1, color[1], color[2]) for color in colors]
 
                 # for categoricals, plot multiple bars
@@ -276,24 +279,24 @@ def plot_cohort_change(
                 save,
             )
 
-        if return_plot:
+        if return_figure:
             return fig, axes
 
         else:
             plt.tight_layout()
             plt.show()
 
-    def plot_flowchart(self, save: str = None, return_plot: bool = True):
+    def plot_flowchart(self, save: str = None, return_figure: bool = True):
         """Flowchart over the tracked steps.
 
         Create a simple flowchart of data preparation steps tracked with `CohortTracker`.
 
         Args:
             save: If a string is provided, the plot will be saved to the path specified.
-            return_plot: If `True`, the plot will be returned as a :class:`~graphviz.Digraph`.
+            return_figure: If `True`, the plot will be returned as a :class:`~graphviz.Digraph`.
 
         Returns:
-            If `return_plot` a :class:`~graphviz.Digraph`.
+            If `return_figure` a :class:`~graphviz.Digraph`.
 
         Example:
             .. code-block:: python
@@ -328,5 +331,5 @@ def plot_flowchart(self, save: str = None, return_plot: bool = True):
             dot.render(save, format="png", cleanup=True)
 
         # Think that to be shown, the plot can a) be rendered (as above) or be "printed" by the notebook
-        if return_plot:
+        if return_figure:
             return dot
diff --git a/tests/tools/cohort_tracking/test_cohort_tracking.py b/tests/tools/cohort_tracking/test_cohort_tracking.py
index a9b0cbfa..a24359f5 100644
--- a/tests/tools/cohort_tracking/test_cohort_tracking.py
+++ b/tests/tools/cohort_tracking/test_cohort_tracking.py
@@ -5,7 +5,6 @@
 import pytest
 
 import ehrapy as ep
-import ehrapy.tools.feature_ranking._rank_features_groups as _utils
 from ehrapy.io._read import read_csv
 
 CURRENT_DIR = Path(__file__).parent
@@ -166,4 +165,4 @@ def test_CohortTracker_plot_cohort_change(self):
         ct(adata)
         ct(adata)
 
-        ct.plot_cohort_change(return_plot=True)
+        ct.plot_cohort_change(return_figure=True)

From c6f5955a501351192cb9231f36bc73ccd671bb3d Mon Sep 17 00:00:00 2001
From: Eljas Roellin <eljas.roellin@ikmail.com>
Date: Thu, 29 Feb 2024 18:29:11 +0100
Subject: [PATCH 08/46] documentation working somewhat

---
 docs/usage/usage.md                           | 10 +++
 .../tools/cohort_tracking/_cohort_tracker.py  | 70 +++++++++----------
 2 files changed, 45 insertions(+), 35 deletions(-)

diff --git a/docs/usage/usage.md b/docs/usage/usage.md
index 9c749245..c4c1108f 100644
--- a/docs/usage/usage.md
+++ b/docs/usage/usage.md
@@ -253,6 +253,16 @@ In contrast to a preprocessing function, a tool usually adds an easily interpret
     tools.causal_inference
 ```
 
+### Cohort Tracking
+
+```{eval-rst}
+.. autosummary::
+    :toctree: tools
+    :nosignatures:
+
+    tools.CohortTracker
+```
+
 ## Plotting
 
 The plotting module `ehrapy.pl.\*` largely parallels the `tl.\*` and a few of the `pp.\*` functions.
diff --git a/ehrapy/tools/cohort_tracking/_cohort_tracker.py b/ehrapy/tools/cohort_tracking/_cohort_tracker.py
index bd3faea6..8d214627 100644
--- a/ehrapy/tools/cohort_tracking/_cohort_tracker.py
+++ b/ehrapy/tools/cohort_tracking/_cohort_tracker.py
@@ -35,25 +35,25 @@ def _detect_categorical_columns(data) -> list:
 
 
 class CohortTracker:
-    def __init__(
-        self, adata: AnnData | pd.DataFrame, columns: Iterable = None, categorical: Iterable = None, *args: Any
-    ):
-        """Track cohort changes over multiple filtering or processing steps.
+    """Track cohort changes over multiple filtering or processing steps.
 
-        This class offers functionality to track and plot cohort changes over multiple filtering or processing steps,
-        enabling the user to monitor the impact of each step on the cohort.
+    This class offers functionality to track and plot cohort changes over multiple filtering or processing steps,
+    enabling the user to monitor the impact of each step on the cohort.
 
-        Tightly interacting with the `tableone` package [1].
-        Args:
-            adata: :class:`~anndata.AnnData` or :class:`~pandas.DataFrame` object to track.
-            columns: Iterable of columns to track. If `None`, all columns will be tracked.
-            categorical: Iterable of columns that contain categorical variables, if not given will be inferred from the data.
+    Tightly interacting with the `tableone` package [1].
 
-        References
-        ----------
+    Args:
+        adata: Object to track.
+        columns: Columns to track. If `None`, all columns will be tracked.
+        categorical: Columns that contain categorical variables, if None will be inferred from the data.
+
+    References:
         [1] Tom Pollard, Alistair E.W. Johnson, Jesse D. Raffa, Roger G. Mark; tableone: An open source Python package for producing summary statistics for research papers, Journal of the American Medical Informatics Association, Volume 24, Issue 2, 1 March 2017, Pages 267–271, https://doi.org/10.1093/jamia/ocw117
+    """
 
-        """
+    def __init__(
+        self, adata: AnnData | pd.DataFrame, columns: Iterable = None, categorical: Iterable = None, *args: Any
+    ):
         if isinstance(adata, AnnData):
             df = adata.obs
         elif isinstance(adata, pd.DataFrame):
@@ -140,7 +140,11 @@ def _get_num_dicts(self, table_one, col):
         summary = table_one.cont_table["Overall"].loc[(col, "")]
         self.track[col].append(summary)
 
-    def reset(self):
+    def reset(self) -> None:
+        """Resets the `CohortTracker` object.
+
+        A full reset of the `CohortTracker` object.
+        """
         self.track = self._track_backup
         self._tracked_steps = 0
         self._tracked_text = []
@@ -148,6 +152,7 @@ def reset(self):
 
     @property
     def tracked_steps(self):
+        """list: List of tableone objects of each logging step."""
         return self._tracked_steps
 
     def plot_cohort_change(
@@ -177,17 +182,14 @@ def plot_cohort_change(
             If `return_figure` a :class:`~matplotlib.figure.Figure` and a :class:`~matplotlib.axes.Axes` or a list of it.
 
         Examples:
-            .. code-block:: python
-
-                import ehrapy as ep
+                >>> import ehrapy as ep
+                >>> adata = ep.dt.diabetes_130(columns_obs_only=["gender", "race", "weight", "age"])
+                >>> cohort_tracker = ep.tl.CohortTracker(adata)
+                >>> cohort_tracker(adata, label="original")
+                >>> adata = adata[:1000]
+                >>> cohort_tracker(adata, label="filtered cohort", operations_done="filtered to first 1000 entries")
+                >>> cohort_tracker.plot_cohort_change()
 
-                adata = ep.dt.diabetes_130(columns_obs_only=["gender", "race", "weight", "age"])
-                cohort_tracker = ep.tl.CohortTracker(adata)
-                cohort_tracker(adata, label="original")
-                adata = adata[:1000]
-                cohort_tracker(adata, label="filtered cohort", operations_done="filtered to first 1000 entries")
-                cohort_tracker.plot_cohort_change()
-        Preview:
             .. image:: /_static/docstring_previews/flowchart.png
         """
         # Plotting
@@ -298,18 +300,16 @@ def plot_flowchart(self, save: str = None, return_figure: bool = True):
         Returns:
             If `return_figure` a :class:`~graphviz.Digraph`.
 
-        Example:
-            .. code-block:: python
+        Examples:
 
-                import ehrapy as ep
+                >>> import ehrapy as ep
+                >>> adata = ep.dt.diabetes_130(columns_obs_only=["gender", "race", "weight", "age"])
+                >>> cohort_tracker = ep.tl.CohortTracker(adata)
+                >>> cohort_tracker(adata, label="original")
+                >>> adata = adata[:1000]
+                >>> cohort_tracker(adata, label="filtered cohort", operations_done="filtered to first 1000 entries")
+                >>> cohort_tracker.plot_flowchart()
 
-                adata = ep.dt.diabetes_130(columns_obs_only=["gender", "race", "weight", "age"])
-                cohort_tracker = ep.tl.CohortTracker(adata)
-                cohort_tracker(adata, label="original")
-                adata = adata[:1000]
-                cohort_tracker(adata, label="filtered cohort", operations_done="filtered to first 1000 entries")
-                cohort_tracker.plot_flowchart()
-        Preview:
             .. image:: /_static/docstring_previews/flowchart.png
 
         """

From dcc3841ef7478588adb51bdc90da35b27fe51c3b Mon Sep 17 00:00:00 2001
From: Eljas Roellin <eljas.roellin@ikmail.com>
Date: Fri, 1 Mar 2024 12:06:10 +0100
Subject: [PATCH 09/46] remove class in tests

---
 .../cohort_tracking/test_cohort_tracking.py   | 236 +++++++++---------
 1 file changed, 120 insertions(+), 116 deletions(-)

diff --git a/tests/tools/cohort_tracking/test_cohort_tracking.py b/tests/tools/cohort_tracking/test_cohort_tracking.py
index a24359f5..56d0ad1f 100644
--- a/tests/tools/cohort_tracking/test_cohort_tracking.py
+++ b/tests/tools/cohort_tracking/test_cohort_tracking.py
@@ -34,135 +34,139 @@ def _compare_dict_equal(dict1, dict2, tolerance=1e-9):
         return dict1 == dict2
 
 
-class TestCohortTracker:
-    @pytest.mark.parametrize("columns", [None, ["glucose", "weight", "disease", "station"]])
-    def test_CohortTracker_init_vanilla(self, columns):
-        adata = ep.io.read_csv(
-            f"{_TEST_DATA_PATH}/dataset1.csv", columns_obs_only=["glucose", "weight", "disease", "station"]
+@pytest.mark.parametrize("columns", [None, ["glucose", "weight", "disease", "station"]])
+def test_CohortTracker_init_vanilla(columns):
+    adata = ep.io.read_csv(
+        f"{_TEST_DATA_PATH}/dataset1.csv", columns_obs_only=["glucose", "weight", "disease", "station"]
+    )
+
+    ct = ep.tl.CohortTracker(adata, columns)
+    assert ct._tracked_steps == 0
+    assert ct.tracked_steps == 0
+    assert ct._tracked_text == []
+    assert ct._tracked_operations == []
+
+    target_track = {
+        "glucose": [],
+        "weight": [],
+        "disease": {"A": [], "B": [], "C": []},
+        "station": {"ICU": [], "MICU": []},
+    }
+    assert _compare_dict_equal(ct.track, target_track)
+
+
+def test_CohortTracker_init_set_columns():
+    adata = ep.io.read_csv(
+        f"{_TEST_DATA_PATH}/dataset1.csv", columns_obs_only=["glucose", "weight", "disease", "station"]
+    )
+    # limit columns
+    ct = ep.tl.CohortTracker(adata, columns=["glucose", "disease"])
+    target_track = {
+        "glucose": [],
+        "disease": {"A": [], "B": [], "C": []},
+    }
+    assert _compare_dict_equal(ct.track, target_track)
+
+    # invalid column
+    with pytest.raises(ValueError):
+        ep.tl.CohortTracker(
+            adata,
+            columns=["glucose", "disease", "non_existing_column"],
         )
 
-        ct = ep.tl.CohortTracker(adata, columns)
-        assert ct._tracked_steps == 0
-        assert ct.tracked_steps == 0
-        assert ct._tracked_text == []
-        assert ct._tracked_operations == []
-
-        target_track = {
-            "glucose": [],
-            "weight": [],
-            "disease": {"A": [], "B": [], "C": []},
-            "station": {"ICU": [], "MICU": []},
-        }
-        assert _compare_dict_equal(ct.track, target_track)
-
-    def test_CohortTracker_init_set_columns(self):
-        adata = ep.io.read_csv(
-            f"{_TEST_DATA_PATH}/dataset1.csv", columns_obs_only=["glucose", "weight", "disease", "station"]
-        )
-        # limit columns
-        ct = ep.tl.CohortTracker(adata, columns=["glucose", "disease"])
-        target_track = {
-            "glucose": [],
-            "disease": {"A": [], "B": [], "C": []},
-        }
-        assert _compare_dict_equal(ct.track, target_track)
-
-        # invalid column
-        with pytest.raises(ValueError):
-            ep.tl.CohortTracker(
-                adata,
-                columns=["glucose", "disease", "non_existing_column"],
-            )
-
-        # force categoricalization
-        ct = ep.tl.CohortTracker(adata, columns=["glucose", "disease"], categorical=["glucose", "disease"])
-        target_track = {
-            "glucose": {70: [], 80: [], 85: [], 90: [], 95: [], 120: [], 125: [], 130: [], 135: []},
-            "disease": {"A": [], "B": [], "C": []},
-        }
-        assert _compare_dict_equal(ct.track, target_track)
-
-        # invalid category
-        with pytest.raises(ValueError):
-            ep.tl.CohortTracker(
-                adata,
-                columns=["glucose", "disease"],
-                categorical=["station"],
-            )
-
-    def test_CohortTracker_call(self):
-        adata = ep.io.read_csv(
-            f"{_TEST_DATA_PATH}/dataset1.csv", columns_obs_only=["glucose", "weight", "disease", "station"]
+    # force categoricalization
+    ct = ep.tl.CohortTracker(adata, columns=["glucose", "disease"], categorical=["glucose", "disease"])
+    target_track = {
+        "glucose": {70: [], 80: [], 85: [], 90: [], 95: [], 120: [], 125: [], 130: [], 135: []},
+        "disease": {"A": [], "B": [], "C": []},
+    }
+    assert _compare_dict_equal(ct.track, target_track)
+
+    # invalid category
+    with pytest.raises(ValueError):
+        ep.tl.CohortTracker(
+            adata,
+            columns=["glucose", "disease"],
+            categorical=["station"],
         )
 
-        ct = ep.tl.CohortTracker(adata)
-
-        ct(adata)
-        assert ct.tracked_steps == 1
-        assert ct._tracked_text == ["Cohort 0\n (n=12)"]
-        target_track_1 = {
-            "glucose": ["105.0 (23.6)"],
-            "weight": ["76.0 (14.9)"],
-            "disease": {"A": [33.3], "B": [33.3], "C": [33.3]},
-            "station": {"ICU": [50.0], "MICU": [50.0]},
-        }
-        assert _compare_dict_equal(ct.track, target_track_1)
-
-        ct(adata)
-        assert ct.tracked_steps == 2
-        assert ct._tracked_text == ["Cohort 0\n (n=12)", "Cohort 1\n (n=12)"]
-        target_track_2 = {
-            "glucose": ["105.0 (23.6)", "105.0 (23.6)"],
-            "weight": ["76.0 (14.9)", "76.0 (14.9)"],
-            "disease": {"A": [33.3, 33.3], "B": [33.3, 33.3], "C": [33.3, 33.3]},
-            "station": {"ICU": [50.0, 50.0], "MICU": [50.0, 50.0]},
-        }
-        assert _compare_dict_equal(ct.track, target_track_2)
-
-    def test_CohortTracker_reset(self):
-        adata = ep.io.read_csv(
-            f"{_TEST_DATA_PATH}/dataset1.csv", columns_obs_only=["glucose", "weight", "disease", "station"]
-        )
 
-        ct = ep.tl.CohortTracker(adata)
+def test_CohortTracker_call():
+    adata = ep.io.read_csv(
+        f"{_TEST_DATA_PATH}/dataset1.csv", columns_obs_only=["glucose", "weight", "disease", "station"]
+    )
 
-        ct(adata)
-        ct(adata)
+    ct = ep.tl.CohortTracker(adata)
 
-        ct.reset()
-        assert ct.tracked_steps == 0
-        assert ct._tracked_text == []
-        assert ct._tracked_operations == []
+    ct(adata)
+    assert ct.tracked_steps == 1
+    assert ct._tracked_text == ["Cohort 0\n (n=12)"]
+    target_track_1 = {
+        "glucose": ["105.0 (23.6)"],
+        "weight": ["76.0 (14.9)"],
+        "disease": {"A": [33.3], "B": [33.3], "C": [33.3]},
+        "station": {"ICU": [50.0], "MICU": [50.0]},
+    }
+    assert _compare_dict_equal(ct.track, target_track_1)
 
-        target_track = {
-            "glucose": [],
-            "weight": [],
-            "disease": {"A": [], "B": [], "C": []},
-            "station": {"ICU": [], "MICU": []},
-        }
-        assert _compare_dict_equal(ct.track, target_track)
-        assert _compare_dict_equal(ct._track_backup, target_track)
+    ct(adata)
+    assert ct.tracked_steps == 2
+    assert ct._tracked_text == ["Cohort 0\n (n=12)", "Cohort 1\n (n=12)"]
+    target_track_2 = {
+        "glucose": ["105.0 (23.6)", "105.0 (23.6)"],
+        "weight": ["76.0 (14.9)", "76.0 (14.9)"],
+        "disease": {"A": [33.3, 33.3], "B": [33.3, 33.3], "C": [33.3, 33.3]},
+        "station": {"ICU": [50.0, 50.0], "MICU": [50.0, 50.0]},
+    }
+    assert _compare_dict_equal(ct.track, target_track_2)
 
-    def test_CohortTracker_flowchart(self):
-        adata = ep.io.read_csv(
-            f"{_TEST_DATA_PATH}/dataset1.csv", columns_obs_only=["glucose", "weight", "disease", "station"]
-        )
 
-        ct = ep.tl.CohortTracker(adata)
+def test_CohortTracker_reset():
+    adata = ep.io.read_csv(
+        f"{_TEST_DATA_PATH}/dataset1.csv", columns_obs_only=["glucose", "weight", "disease", "station"]
+    )
 
-        ct(adata, label="First step", operations_done="Some operations")
-        ct(adata, label="Second step", operations_done="Some other operations")
+    ct = ep.tl.CohortTracker(adata)
 
-        ct.plot_flowchart()
+    ct(adata)
+    ct(adata)
 
-    def test_CohortTracker_plot_cohort_change(self):
-        adata = ep.io.read_csv(
-            f"{_TEST_DATA_PATH}/dataset1.csv", columns_obs_only=["glucose", "weight", "disease", "station"]
-        )
+    ct.reset()
+    assert ct.tracked_steps == 0
+    assert ct._tracked_text == []
+    assert ct._tracked_operations == []
+
+    target_track = {
+        "glucose": [],
+        "weight": [],
+        "disease": {"A": [], "B": [], "C": []},
+        "station": {"ICU": [], "MICU": []},
+    }
+    assert _compare_dict_equal(ct.track, target_track)
+    assert _compare_dict_equal(ct._track_backup, target_track)
+
+
+def test_CohortTracker_flowchart():
+    adata = ep.io.read_csv(
+        f"{_TEST_DATA_PATH}/dataset1.csv", columns_obs_only=["glucose", "weight", "disease", "station"]
+    )
+
+    ct = ep.tl.CohortTracker(adata)
+
+    ct(adata, label="First step", operations_done="Some operations")
+    ct(adata, label="Second step", operations_done="Some other operations")
+
+    ct.plot_flowchart()
+
+
+def test_CohortTracker_plot_cohort_change():
+    adata = ep.io.read_csv(
+        f"{_TEST_DATA_PATH}/dataset1.csv", columns_obs_only=["glucose", "weight", "disease", "station"]
+    )
 
-        ct = ep.tl.CohortTracker(adata)
+    ct = ep.tl.CohortTracker(adata)
 
-        ct(adata)
-        ct(adata)
+    ct(adata)
+    ct(adata)
 
-        ct.plot_cohort_change(return_figure=True)
+    ct.plot_cohort_change(return_figure=True)

From b8440965c429c8f998573c28071b325800fd76c8 Mon Sep 17 00:00:00 2001
From: Eljas Roellin <eljas.roellin@ikmail.com>
Date: Fri, 1 Mar 2024 13:48:24 +0100
Subject: [PATCH 10/46] move read_csv to fixture

---
 .../cohort_tracking/test_cohort_tracking.py   | 74 +++++++------------
 1 file changed, 28 insertions(+), 46 deletions(-)

diff --git a/tests/tools/cohort_tracking/test_cohort_tracking.py b/tests/tools/cohort_tracking/test_cohort_tracking.py
index 56d0ad1f..52daff5a 100644
--- a/tests/tools/cohort_tracking/test_cohort_tracking.py
+++ b/tests/tools/cohort_tracking/test_cohort_tracking.py
@@ -34,13 +34,14 @@ def _compare_dict_equal(dict1, dict2, tolerance=1e-9):
         return dict1 == dict2
 
 
-@pytest.mark.parametrize("columns", [None, ["glucose", "weight", "disease", "station"]])
-def test_CohortTracker_init_vanilla(columns):
-    adata = ep.io.read_csv(
-        f"{_TEST_DATA_PATH}/dataset1.csv", columns_obs_only=["glucose", "weight", "disease", "station"]
-    )
+@pytest.fixture
+def mini_adata():
+    return read_csv(f"{_TEST_DATA_PATH}/dataset1.csv", columns_obs_only=["glucose", "weight", "disease", "station"])
+
 
-    ct = ep.tl.CohortTracker(adata, columns)
+@pytest.mark.parametrize("columns", [None, ["glucose", "weight", "disease", "station"]])
+def test_CohortTracker_init_vanilla(columns, mini_adata):
+    ct = ep.tl.CohortTracker(mini_adata, columns)
     assert ct._tracked_steps == 0
     assert ct.tracked_steps == 0
     assert ct._tracked_text == []
@@ -55,12 +56,9 @@ def test_CohortTracker_init_vanilla(columns):
     assert _compare_dict_equal(ct.track, target_track)
 
 
-def test_CohortTracker_init_set_columns():
-    adata = ep.io.read_csv(
-        f"{_TEST_DATA_PATH}/dataset1.csv", columns_obs_only=["glucose", "weight", "disease", "station"]
-    )
+def test_CohortTracker_init_set_columns(mini_adata):
     # limit columns
-    ct = ep.tl.CohortTracker(adata, columns=["glucose", "disease"])
+    ct = ep.tl.CohortTracker(mini_adata, columns=["glucose", "disease"])
     target_track = {
         "glucose": [],
         "disease": {"A": [], "B": [], "C": []},
@@ -70,12 +68,12 @@ def test_CohortTracker_init_set_columns():
     # invalid column
     with pytest.raises(ValueError):
         ep.tl.CohortTracker(
-            adata,
+            mini_adata,
             columns=["glucose", "disease", "non_existing_column"],
         )
 
     # force categoricalization
-    ct = ep.tl.CohortTracker(adata, columns=["glucose", "disease"], categorical=["glucose", "disease"])
+    ct = ep.tl.CohortTracker(mini_adata, columns=["glucose", "disease"], categorical=["glucose", "disease"])
     target_track = {
         "glucose": {70: [], 80: [], 85: [], 90: [], 95: [], 120: [], 125: [], 130: [], 135: []},
         "disease": {"A": [], "B": [], "C": []},
@@ -85,20 +83,16 @@ def test_CohortTracker_init_set_columns():
     # invalid category
     with pytest.raises(ValueError):
         ep.tl.CohortTracker(
-            adata,
+            mini_adata,
             columns=["glucose", "disease"],
             categorical=["station"],
         )
 
 
-def test_CohortTracker_call():
-    adata = ep.io.read_csv(
-        f"{_TEST_DATA_PATH}/dataset1.csv", columns_obs_only=["glucose", "weight", "disease", "station"]
-    )
-
-    ct = ep.tl.CohortTracker(adata)
+def test_CohortTracker_call(mini_adata):
+    ct = ep.tl.CohortTracker(mini_adata)
 
-    ct(adata)
+    ct(mini_adata)
     assert ct.tracked_steps == 1
     assert ct._tracked_text == ["Cohort 0\n (n=12)"]
     target_track_1 = {
@@ -109,7 +103,7 @@ def test_CohortTracker_call():
     }
     assert _compare_dict_equal(ct.track, target_track_1)
 
-    ct(adata)
+    ct(mini_adata)
     assert ct.tracked_steps == 2
     assert ct._tracked_text == ["Cohort 0\n (n=12)", "Cohort 1\n (n=12)"]
     target_track_2 = {
@@ -121,15 +115,11 @@ def test_CohortTracker_call():
     assert _compare_dict_equal(ct.track, target_track_2)
 
 
-def test_CohortTracker_reset():
-    adata = ep.io.read_csv(
-        f"{_TEST_DATA_PATH}/dataset1.csv", columns_obs_only=["glucose", "weight", "disease", "station"]
-    )
+def test_CohortTracker_reset(mini_adata):
+    ct = ep.tl.CohortTracker(mini_adata)
 
-    ct = ep.tl.CohortTracker(adata)
-
-    ct(adata)
-    ct(adata)
+    ct(mini_adata)
+    ct(mini_adata)
 
     ct.reset()
     assert ct.tracked_steps == 0
@@ -146,27 +136,19 @@ def test_CohortTracker_reset():
     assert _compare_dict_equal(ct._track_backup, target_track)
 
 
-def test_CohortTracker_flowchart():
-    adata = ep.io.read_csv(
-        f"{_TEST_DATA_PATH}/dataset1.csv", columns_obs_only=["glucose", "weight", "disease", "station"]
-    )
-
-    ct = ep.tl.CohortTracker(adata)
+def test_CohortTracker_flowchart(mini_adata):
+    ct = ep.tl.CohortTracker(mini_adata)
 
-    ct(adata, label="First step", operations_done="Some operations")
-    ct(adata, label="Second step", operations_done="Some other operations")
+    ct(mini_adata, label="First step", operations_done="Some operations")
+    ct(mini_adata, label="Second step", operations_done="Some other operations")
 
     ct.plot_flowchart()
 
 
-def test_CohortTracker_plot_cohort_change():
-    adata = ep.io.read_csv(
-        f"{_TEST_DATA_PATH}/dataset1.csv", columns_obs_only=["glucose", "weight", "disease", "station"]
-    )
-
-    ct = ep.tl.CohortTracker(adata)
+def test_CohortTracker_plot_cohort_change(mini_adata):
+    ct = ep.tl.CohortTracker(mini_adata)
 
-    ct(adata)
-    ct(adata)
+    ct(mini_adata)
+    ct(mini_adata)
 
     ct.plot_cohort_change(return_figure=True)

From 75451bdbc4b03b83724c80ea7fc934e9fe98a2da Mon Sep 17 00:00:00 2001
From: Eljas Roellin <eljas.roellin@ikmail.com>
Date: Tue, 5 Mar 2024 10:05:05 +0100
Subject: [PATCH 11/46] remove tracking dict, use tableones for tracking
 instead

---
 .../tools/cohort_tracking/_cohort_tracker.py  | 73 +++++++-----------
 .../cohort_tracking/test_cohort_tracking.py   | 77 +++----------------
 2 files changed, 37 insertions(+), 113 deletions(-)

diff --git a/ehrapy/tools/cohort_tracking/_cohort_tracker.py b/ehrapy/tools/cohort_tracking/_cohort_tracker.py
index 8d214627..b386412c 100644
--- a/ehrapy/tools/cohort_tracking/_cohort_tracker.py
+++ b/ehrapy/tools/cohort_tracking/_cohort_tracker.py
@@ -1,6 +1,5 @@
-import copy
-from collections.abc import Iterable
-from typing import Any, Union
+from collections.abc import Sequence
+from typing import Any
 
 import graphviz
 import matplotlib.pyplot as plt
@@ -52,8 +51,8 @@ class CohortTracker:
     """
 
     def __init__(
-        self, adata: AnnData | pd.DataFrame, columns: Iterable = None, categorical: Iterable = None, *args: Any
-    ):
+        self, adata: AnnData | pd.DataFrame, columns: Sequence = None, categorical: Sequence = None, *args: Any
+    ) -> None:
         if isinstance(adata, AnnData):
             df = adata.obs
         elif isinstance(adata, pd.DataFrame):
@@ -77,9 +76,7 @@ def __init__(
         # if categorical columns specified, use them
         # else, follow tableone's logic
         self.categorical = categorical if categorical is not None else _detect_categorical_columns(df[self.columns])
-        self.track = self._get_column_structure(df)
-
-        self._track_backup = copy.deepcopy(self.track)
+        self.track_t1: list = []
 
     def __call__(
         self, adata: AnnData, label: str = None, operations_done: str = None, *args: Any, **tableone_kwargs: Any
@@ -100,52 +97,32 @@ def __call__(
 
         # track a small text with the operations done
         self._tracked_operations.append(operations_done)
-
         self._tracked_steps += 1
 
-        t1 = TableOne(df, columns=self.columns, categorical=self.categorical, **tableone_kwargs)
         # track new stuff
-        self._get_column_dicts(t1)
-
-    def _get_column_structure(self, df):
-        column_structure = {}
-        for column in self.columns:
-            if column in self.categorical:
-                # if e.g. a column containing integers is deemed categorical, coerce it to categorical
-                df[column] = df[column].astype("category")
-                column_structure[column] = {category: [] for category in df[column].cat.categories}
-            else:
-                column_structure[column] = []
-
-        return column_structure
-
-    def _get_column_dicts(self, table_one):
-        for col, value in self.track.items():
-            if isinstance(value, dict):
-                self._get_cat_dicts(table_one, col)
-            else:
-                self._get_num_dicts(table_one, col)
+        t1 = TableOne(df, columns=self.columns, categorical=self.categorical, **tableone_kwargs)
+        self.track_t1.append(t1)
 
     def _get_cat_dicts(self, table_one, col):
-        for cat in self.track[col].keys():
+        cat_pct = {category: [] for category in table_one.cat_table.loc[col].index}
+        for cat in cat_pct.keys():
             # if tableone does not have the category of this column anymore, set the percentage to 0
             # for categorized columns (e.g. gender 1.0/0.0), str(cat) helps to avoid considering the category as a float
-            if (col, str(cat)) in table_one.cat_table["Overall"].index:
-                pct = float(table_one.cat_table["Overall"].loc[(col, str(cat))].split("(")[1].split(")")[0])
-            else:
-                pct = 0
-            self.track[col][cat].append(pct)
+            # if (col, str(cat)) in table_one.cat_table["Overall"].index:
+            pct = float(table_one.cat_table["Overall"].loc[(col, str(cat))].split("(")[1].split(")")[0])
+            # else:
+            #    pct = 0
+            cat_pct[cat] = [pct]
+        return pd.DataFrame(cat_pct).T[0]
 
     def _get_num_dicts(self, table_one, col):
-        summary = table_one.cont_table["Overall"].loc[(col, "")]
-        self.track[col].append(summary)
+        return table_one.cont_table["Overall"].loc[(col, "")]
 
     def reset(self) -> None:
         """Resets the `CohortTracker` object.
 
         A full reset of the `CohortTracker` object.
         """
-        self.track = self._track_backup
         self._tracked_steps = 0
         self._tracked_text = []
         self._tracked_operations = []
@@ -209,8 +186,12 @@ def plot_cohort_change(
                 ax.set_title(self._tracked_text[idx])
 
             # iterate over the tracked columns in the dataframe
-            for pos, (_cols, data) in enumerate(self.track.items()):
-                data = pd.DataFrame(data).loc[idx]
+            # TODO: allow for new/disappearing columns during logging?
+            for pos, col in enumerate(self.columns):
+                if col in self.categorical:
+                    data = self._get_cat_dicts(self.track_t1[idx], col)
+                else:
+                    data = [self._get_num_dicts(self.track_t1[idx], col)]
 
                 cumwidth = 0
 
@@ -220,7 +201,7 @@ def plot_cohort_change(
                 adjusted_colors = [((color[0] + hue_shift) % 1, color[1], color[2]) for color in colors]
 
                 # for categoricals, plot multiple bars
-                if _cols in self.categorical:
+                if col in self.categorical:
                     for i, value in enumerate(data):
                         ax.barh(pos, value, left=cumwidth, color=adjusted_colors[i], height=0.7)
 
@@ -254,14 +235,12 @@ def plot_cohort_change(
                         color="white",
                         fontweight="bold",
                     )
-                    legend_labels.append(_cols)
+                    legend_labels.append(col)
 
             # Set y-axis labels
             if set_axis_labels:
-                ax.set_yticks(
-                    range(len(self.track.keys()))
-                )  # Set ticks at positions corresponding to the number of columns
-                ax.set_yticklabels(self.track.keys())  # Set y-axis labels to the column names
+                ax.set_yticks(range(len(self.columns)))  # Set ticks at positions corresponding to the number of columns
+                ax.set_yticklabels(self.columns)  # Set y-axis labels to the column names
 
             # makes the frames invisible
             # for ax in axes:
diff --git a/tests/tools/cohort_tracking/test_cohort_tracking.py b/tests/tools/cohort_tracking/test_cohort_tracking.py
index 52daff5a..a0a8453d 100644
--- a/tests/tools/cohort_tracking/test_cohort_tracking.py
+++ b/tests/tools/cohort_tracking/test_cohort_tracking.py
@@ -11,29 +11,6 @@
 _TEST_DATA_PATH = f"{CURRENT_DIR.parent}/test_data_features_ranking"
 
 
-def _compare_dict_equal(dict1, dict2, tolerance=1e-9):
-    if isinstance(dict1, dict) and isinstance(dict2, dict):
-        if set(dict1.keys()) != set(dict2.keys()):
-            return False
-        for key in dict1.keys():
-            if not _compare_dict_equal(dict1[key], dict2[key], tolerance):
-                return False
-        return True
-    elif isinstance(dict1, list) and isinstance(dict2, list):
-        if len(dict1) != len(dict2):
-            return False
-        for val1, val2 in zip(dict1, dict2):
-            if not _compare_dict_equal(val1, val2, tolerance):
-                return False
-        return True
-    elif isinstance(dict1, float) and isinstance(dict2, float):
-        return abs(dict1 - dict2) < tolerance
-    elif isinstance(dict1, str) and isinstance(dict2, str):
-        return dict1 == dict2
-    else:
-        return dict1 == dict2
-
-
 @pytest.fixture
 def mini_adata():
     return read_csv(f"{_TEST_DATA_PATH}/dataset1.csv", columns_obs_only=["glucose", "weight", "disease", "station"])
@@ -47,23 +24,12 @@ def test_CohortTracker_init_vanilla(columns, mini_adata):
     assert ct._tracked_text == []
     assert ct._tracked_operations == []
 
-    target_track = {
-        "glucose": [],
-        "weight": [],
-        "disease": {"A": [], "B": [], "C": []},
-        "station": {"ICU": [], "MICU": []},
-    }
-    assert _compare_dict_equal(ct.track, target_track)
-
 
 def test_CohortTracker_init_set_columns(mini_adata):
     # limit columns
-    ct = ep.tl.CohortTracker(mini_adata, columns=["glucose", "disease"])
-    target_track = {
-        "glucose": [],
-        "disease": {"A": [], "B": [], "C": []},
-    }
-    assert _compare_dict_equal(ct.track, target_track)
+    ep.tl.CohortTracker(mini_adata, columns=["glucose", "disease"])
+
+    # TODO: check plot?
 
     # invalid column
     with pytest.raises(ValueError):
@@ -73,12 +39,9 @@ def test_CohortTracker_init_set_columns(mini_adata):
         )
 
     # force categoricalization
-    ct = ep.tl.CohortTracker(mini_adata, columns=["glucose", "disease"], categorical=["glucose", "disease"])
-    target_track = {
-        "glucose": {70: [], 80: [], 85: [], 90: [], 95: [], 120: [], 125: [], 130: [], 135: []},
-        "disease": {"A": [], "B": [], "C": []},
-    }
-    assert _compare_dict_equal(ct.track, target_track)
+    ep.tl.CohortTracker(mini_adata, columns=["glucose", "disease"], categorical=["glucose", "disease"])
+
+    # TODO: check plot?
 
     # invalid category
     with pytest.raises(ValueError):
@@ -95,24 +58,15 @@ def test_CohortTracker_call(mini_adata):
     ct(mini_adata)
     assert ct.tracked_steps == 1
     assert ct._tracked_text == ["Cohort 0\n (n=12)"]
-    target_track_1 = {
-        "glucose": ["105.0 (23.6)"],
-        "weight": ["76.0 (14.9)"],
-        "disease": {"A": [33.3], "B": [33.3], "C": [33.3]},
-        "station": {"ICU": [50.0], "MICU": [50.0]},
-    }
-    assert _compare_dict_equal(ct.track, target_track_1)
+
+    # TODO: check plot?
 
     ct(mini_adata)
     assert ct.tracked_steps == 2
     assert ct._tracked_text == ["Cohort 0\n (n=12)", "Cohort 1\n (n=12)"]
-    target_track_2 = {
-        "glucose": ["105.0 (23.6)", "105.0 (23.6)"],
-        "weight": ["76.0 (14.9)", "76.0 (14.9)"],
-        "disease": {"A": [33.3, 33.3], "B": [33.3, 33.3], "C": [33.3, 33.3]},
-        "station": {"ICU": [50.0, 50.0], "MICU": [50.0, 50.0]},
-    }
-    assert _compare_dict_equal(ct.track, target_track_2)
+
+
+# TODO: check plot?
 
 
 def test_CohortTracker_reset(mini_adata):
@@ -126,15 +80,6 @@ def test_CohortTracker_reset(mini_adata):
     assert ct._tracked_text == []
     assert ct._tracked_operations == []
 
-    target_track = {
-        "glucose": [],
-        "weight": [],
-        "disease": {"A": [], "B": [], "C": []},
-        "station": {"ICU": [], "MICU": []},
-    }
-    assert _compare_dict_equal(ct.track, target_track)
-    assert _compare_dict_equal(ct._track_backup, target_track)
-
 
 def test_CohortTracker_flowchart(mini_adata):
     ct = ep.tl.CohortTracker(mini_adata)

From c42d84728059704d53e6f646d16c6d96cb37e358 Mon Sep 17 00:00:00 2001
From: Eljas Roellin <eljas.roellin@ikmail.com>
Date: Tue, 5 Mar 2024 10:11:21 +0100
Subject: [PATCH 12/46] remove DataFrame as accepted input

---
 .../tools/cohort_tracking/_cohort_tracker.py  | 39 ++++++++-----------
 1 file changed, 16 insertions(+), 23 deletions(-)

diff --git a/ehrapy/tools/cohort_tracking/_cohort_tracker.py b/ehrapy/tools/cohort_tracking/_cohort_tracker.py
index b386412c..4ad97e26 100644
--- a/ehrapy/tools/cohort_tracking/_cohort_tracker.py
+++ b/ehrapy/tools/cohort_tracking/_cohort_tracker.py
@@ -10,6 +10,11 @@
 from tableone import TableOne
 
 
+def _check_adata_type(adata) -> None:
+    if not isinstance(adata, AnnData):
+        raise ValueError("adata must be an AnnData.")
+
+
 def _check_columns_exist(df, columns) -> None:
     missing_columns = set(columns) - set(df.columns)
     if missing_columns:
@@ -50,22 +55,15 @@ class CohortTracker:
         [1] Tom Pollard, Alistair E.W. Johnson, Jesse D. Raffa, Roger G. Mark; tableone: An open source Python package for producing summary statistics for research papers, Journal of the American Medical Informatics Association, Volume 24, Issue 2, 1 March 2017, Pages 267–271, https://doi.org/10.1093/jamia/ocw117
     """
 
-    def __init__(
-        self, adata: AnnData | pd.DataFrame, columns: Sequence = None, categorical: Sequence = None, *args: Any
-    ) -> None:
-        if isinstance(adata, AnnData):
-            df = adata.obs
-        elif isinstance(adata, pd.DataFrame):
-            df = adata
-        else:
-            raise ValueError("adata must be an AnnData or a DataFrame.")
+    def __init__(self, adata: AnnData, columns: Sequence = None, categorical: Sequence = None, *args: Any) -> None:
+        _check_adata_type(adata)
 
-        self.columns = columns if columns is not None else list(df.columns)
+        self.columns = columns if columns is not None else list(adata.obs.columns)
 
         if columns is not None:
-            _check_columns_exist(df, columns)
+            _check_columns_exist(adata.obs, columns)
         if categorical is not None:
-            _check_columns_exist(df, categorical)
+            _check_columns_exist(adata.obs, categorical)
             if set(categorical).difference(set(self.columns)):
                 raise ValueError("categorical columns must be in the (selected) columns.")
 
@@ -75,20 +73,16 @@ def __init__(
 
         # if categorical columns specified, use them
         # else, follow tableone's logic
-        self.categorical = categorical if categorical is not None else _detect_categorical_columns(df[self.columns])
+        self.categorical = (
+            categorical if categorical is not None else _detect_categorical_columns(adata.obs[self.columns])
+        )
         self.track_t1: list = []
 
     def __call__(
         self, adata: AnnData, label: str = None, operations_done: str = None, *args: Any, **tableone_kwargs: Any
     ) -> None:
-        if isinstance(adata, AnnData):
-            df = adata.obs
-        elif isinstance(adata, pd.DataFrame):
-            df = adata
-        else:
-            raise ValueError("adata must be an AnnData or a DataFrame.")
-
-        _check_columns_exist(df, self.columns)
+        _check_adata_type(adata)
+        _check_columns_exist(adata.obs, self.columns)
 
         # track a small text with each tracking step, for the flowchart
         track_text = label if label is not None else f"Cohort {self.tracked_steps}"
@@ -100,7 +94,7 @@ def __call__(
         self._tracked_steps += 1
 
         # track new stuff
-        t1 = TableOne(df, columns=self.columns, categorical=self.categorical, **tableone_kwargs)
+        t1 = TableOne(adata.obs, columns=self.columns, categorical=self.categorical, **tableone_kwargs)
         self.track_t1.append(t1)
 
     def _get_cat_dicts(self, table_one, col):
@@ -186,7 +180,6 @@ def plot_cohort_change(
                 ax.set_title(self._tracked_text[idx])
 
             # iterate over the tracked columns in the dataframe
-            # TODO: allow for new/disappearing columns during logging?
             for pos, col in enumerate(self.columns):
                 if col in self.categorical:
                     data = self._get_cat_dicts(self.track_t1[idx], col)

From 232c2b157f3ad027d49611ff9e1fe5f27836b75e Mon Sep 17 00:00:00 2001
From: Eljas Roellin <eljas.roellin@ikmail.com>
Date: Tue, 5 Mar 2024 11:10:17 +0100
Subject: [PATCH 13/46] legend label order matching bar order

---
 .../tools/cohort_tracking/_cohort_tracker.py  | 33 ++++++++++++-------
 1 file changed, 22 insertions(+), 11 deletions(-)

diff --git a/ehrapy/tools/cohort_tracking/_cohort_tracker.py b/ehrapy/tools/cohort_tracking/_cohort_tracker.py
index 4ad97e26..db521dc6 100644
--- a/ehrapy/tools/cohort_tracking/_cohort_tracker.py
+++ b/ehrapy/tools/cohort_tracking/_cohort_tracker.py
@@ -2,10 +2,12 @@
 from typing import Any
 
 import graphviz
+import matplotlib.colors as mcolors
 import matplotlib.pyplot as plt
 import numpy as np
 import pandas as pd
 import seaborn as sns
+from matplotlib.patches import Patch
 from scanpy import AnnData
 from tableone import TableOne
 
@@ -186,17 +188,17 @@ def plot_cohort_change(
                 else:
                     data = [self._get_num_dicts(self.track_t1[idx], col)]
 
-                cumwidth = 0
-
-                # Adjust the hue shift based on the category position such that the colors are more distinguishable
-                hue_shift = (pos + 1) / len(data)
-                colors = sns.color_palette(color_palette, len(data))
-                adjusted_colors = [((color[0] + hue_shift) % 1, color[1], color[2]) for color in colors]
+                # Assign a unique color to each level (i.e. column)
+                level_color = sns.color_palette(color_palette, len(self.columns))[pos]
 
+                cumwidth = 0
                 # for categoricals, plot multiple bars
                 if col in self.categorical:
+                    col_legend_labels = []
                     for i, value in enumerate(data):
-                        ax.barh(pos, value, left=cumwidth, color=adjusted_colors[i], height=0.7)
+                        # Use different shades of the level color for the stacked bars
+                        stacked_bar_color = mcolors.to_rgb(level_color) + (0.5 + 0.5 * (i / len(data)),)
+                        ax.barh(pos, value, left=cumwidth, color=stacked_bar_color, height=0.7)
 
                         if value > 5:
                             # Add proportion numbers to the bars
@@ -214,11 +216,13 @@ def plot_cohort_change(
                         ax.set_yticks([])
                         ax.set_xticks([])
                         cumwidth += value
-                        legend_labels.append(data.index[i])
+                        if idx == 0:
+                            col_legend_labels.append(Patch(color=stacked_bar_color, label=data.index[i]))
+                    legend_labels.append(col_legend_labels)
 
                 # for numericals, plot a single bar
                 else:
-                    ax.barh(pos, 100, left=cumwidth, color=adjusted_colors[0], height=0.8)
+                    ax.barh(pos, 100, left=cumwidth, color=level_color, height=0.8)
                     ax.text(
                         100 / 2,
                         pos,
@@ -228,7 +232,9 @@ def plot_cohort_change(
                         color="white",
                         fontweight="bold",
                     )
-                    legend_labels.append(col)
+                    # legend_labels.append(col)
+                    if idx == 0:
+                        legend_labels.append([Patch(color=level_color, label=col)])
 
             # Set y-axis labels
             if set_axis_labels:
@@ -240,11 +246,16 @@ def plot_cohort_change(
             #     ax.axis('off')
 
         # Add legend
+        # These list of lists is needed to reverse the order of the legend labels,
+        # making the plot much more readable
+        legend_labels.reverse()
+        legend_labels = [item for sublist in legend_labels for item in sublist]
+
         tot_legend_kwargs = {"loc": "best", "bbox_to_anchor": (1, 1)}
         if legend_kwargs is not None:
             tot_legend_kwargs.update(legend_kwargs)
 
-        plt.legend(legend_labels, **tot_legend_kwargs)
+        plt.legend(handles=legend_labels, **tot_legend_kwargs)
 
         if save is not None:
             if not isinstance(save, str):

From 4090f6fca1265bfc1431a79a66bfd79b89956611 Mon Sep 17 00:00:00 2001
From: Eljas Roellin <eljas.roellin@ikmail.com>
Date: Tue, 5 Mar 2024 15:08:16 +0100
Subject: [PATCH 14/46] prepare type detection for alignment, added test

---
 .../tools/cohort_tracking/_cohort_tracker.py   | 18 +++++-------------
 .../cohort_tracking/test_cohort_tracking.py    |  5 +++++
 2 files changed, 10 insertions(+), 13 deletions(-)

diff --git a/ehrapy/tools/cohort_tracking/_cohort_tracker.py b/ehrapy/tools/cohort_tracking/_cohort_tracker.py
index db521dc6..f33cb0e6 100644
--- a/ehrapy/tools/cohort_tracking/_cohort_tracker.py
+++ b/ehrapy/tools/cohort_tracking/_cohort_tracker.py
@@ -25,19 +25,11 @@ def _check_columns_exist(df, columns) -> None:
 
 # from tableone: https://github.com/tompollard/tableone/blob/bfd6fbaa4ed3e9f59e1a75191c6296a2a80ccc64/tableone/tableone.py#L555
 def _detect_categorical_columns(data) -> list:
-    # assume all non-numerical and date columns are categorical
-    numeric_cols = set(data._get_numeric_data().columns.values)
-    date_cols = set(data.select_dtypes(include=[np.datetime64]).columns)
-    likely_cat = set(data.columns) - numeric_cols
-    # mypy absolutely looses it if likely_cat is overwritten to be a list
-    likely_cat_no_dates = list(likely_cat - date_cols)
-
-    # check proportion of unique values if numerical
-    for var in data._get_numeric_data().columns:
-        likely_flag = 1.0 * data[var].nunique() / data[var].count() < 0.005
-        if likely_flag:
-            likely_cat_no_dates.append(var)
-    return likely_cat_no_dates
+    # TODO grab this from ehrapy once https://github.com/theislab/ehrapy/issues/662 addressed
+    numeric_cols = set(data.select_dtypes("number").columns)
+    categorical_cols = set(data.columns) - numeric_cols
+
+    return list(categorical_cols)
 
 
 class CohortTracker:
diff --git a/tests/tools/cohort_tracking/test_cohort_tracking.py b/tests/tools/cohort_tracking/test_cohort_tracking.py
index a0a8453d..5729081b 100644
--- a/tests/tools/cohort_tracking/test_cohort_tracking.py
+++ b/tests/tools/cohort_tracking/test_cohort_tracking.py
@@ -25,6 +25,11 @@ def test_CohortTracker_init_vanilla(columns, mini_adata):
     assert ct._tracked_operations == []
 
 
+def test_CohortTracker_type_detection(mini_adata):
+    ct = ep.tl.CohortTracker(mini_adata, ["glucose", "weight", "disease", "station"])
+    assert set(ct.categorical) == {"disease", "station"}
+
+
 def test_CohortTracker_init_set_columns(mini_adata):
     # limit columns
     ep.tl.CohortTracker(mini_adata, columns=["glucose", "disease"])

From 460615543a9b45ae53694302e8c54e06db147aeb Mon Sep 17 00:00:00 2001
From: Eljas Roellin <eljas.roellin@ikmail.com>
Date: Wed, 6 Mar 2024 09:34:16 +0100
Subject: [PATCH 15/46] add ax and remove unused args, return not solved yet

---
 .../tools/cohort_tracking/_cohort_tracker.py  | 67 +++++++++++--------
 1 file changed, 38 insertions(+), 29 deletions(-)

diff --git a/ehrapy/tools/cohort_tracking/_cohort_tracker.py b/ehrapy/tools/cohort_tracking/_cohort_tracker.py
index f33cb0e6..612ae3a7 100644
--- a/ehrapy/tools/cohort_tracking/_cohort_tracker.py
+++ b/ehrapy/tools/cohort_tracking/_cohort_tracker.py
@@ -7,6 +7,7 @@
 import numpy as np
 import pandas as pd
 import seaborn as sns
+from matplotlib.axes import Axes
 from matplotlib.patches import Patch
 from scanpy import AnnData
 from tableone import TableOne
@@ -49,7 +50,7 @@ class CohortTracker:
         [1] Tom Pollard, Alistair E.W. Johnson, Jesse D. Raffa, Roger G. Mark; tableone: An open source Python package for producing summary statistics for research papers, Journal of the American Medical Informatics Association, Volume 24, Issue 2, 1 March 2017, Pages 267–271, https://doi.org/10.1093/jamia/ocw117
     """
 
-    def __init__(self, adata: AnnData, columns: Sequence = None, categorical: Sequence = None, *args: Any) -> None:
+    def __init__(self, adata: AnnData, columns: Sequence = None, categorical: Sequence = None) -> None:
         _check_adata_type(adata)
 
         self.columns = columns if columns is not None else list(adata.obs.columns)
@@ -72,9 +73,7 @@ def __init__(self, adata: AnnData, columns: Sequence = None, categorical: Sequen
         )
         self.track_t1: list = []
 
-    def __call__(
-        self, adata: AnnData, label: str = None, operations_done: str = None, *args: Any, **tableone_kwargs: Any
-    ) -> None:
+    def __call__(self, adata: AnnData, label: str = None, operations_done: str = None, **tableone_kwargs: Any) -> None:
         _check_adata_type(adata)
         _check_columns_exist(adata.obs, self.columns)
 
@@ -120,13 +119,19 @@ def tracked_steps(self):
         """list: List of tableone objects of each logging step."""
         return self._tracked_steps
 
+    # IMMEDIATE NEXT TODO:
+    # I ALLOWED FOR THE AX ARGUMENT, BUT NEED TO CHECK
+    # WHAT IS A GOOD RETURN TYPE WITH AND WITHOUT, AND WHETHER
+    # THE AX RETURN THING IS ENOUGH IN GENERAL
+    # THEN ASK WHAT IS BEST FOR KEYWORD ARGUMENTS, THEN THIS IS DONE?
+    # ONLY TEST FOR PLOT SIMILARITY I THINK AFTERWARDS LEFT
     def plot_cohort_change(
         self,
         set_axis_labels=True,
         subfigure_title: bool = False,
-        color_palette: str = "husl",
-        save: str = None,
+        color_palette: str = "colorblind",
         return_figure: bool = False,
+        ax: Axes | np.ndarray = None,
         subplots_kwargs: dict = None,
         legend_kwargs: dict = None,
     ):
@@ -138,7 +143,6 @@ def plot_cohort_change(
             set_axis_labels: If `True`, the y-axis labels will be set to the column names.
             subfigure_title: If `True`, each subplot will have a title with the `label` provided during tracking.
             color_palette: The color palette to use for the plot. Default is "husl".
-            save: If a string is provided, the plot will be saved to the path specified.
             return_figure: If `True`, the plot will be returned as a tuple of (fig, ax).
             subplot_kwargs: Additional keyword arguments for the subplots.
             legend_kwargs: Additional keyword arguments for the legend.
@@ -160,7 +164,10 @@ def plot_cohort_change(
         # Plotting
         subplots_kwargs = {} if subplots_kwargs is None else subplots_kwargs
 
-        fig, axes = plt.subplots(self.tracked_steps, 1, **subplots_kwargs)
+        if ax is None:
+            _, axes = plt.subplots(self.tracked_steps, 1, **subplots_kwargs)
+        else:
+            axes = ax
 
         legend_labels = []
 
@@ -190,7 +197,15 @@ def plot_cohort_change(
                     for i, value in enumerate(data):
                         # Use different shades of the level color for the stacked bars
                         stacked_bar_color = mcolors.to_rgb(level_color) + (0.5 + 0.5 * (i / len(data)),)
-                        ax.barh(pos, value, left=cumwidth, color=stacked_bar_color, height=0.7)
+                        ax.barh(
+                            pos,
+                            value,
+                            left=cumwidth,
+                            color=stacked_bar_color,
+                            height=0.7,
+                            edgecolor="black",
+                            linewidth=0.6,
+                        )
 
                         if value > 5:
                             # Add proportion numbers to the bars
@@ -214,7 +229,15 @@ def plot_cohort_change(
 
                 # for numericals, plot a single bar
                 else:
-                    ax.barh(pos, 100, left=cumwidth, color=level_color, height=0.8)
+                    ax.barh(
+                        pos,
+                        100,
+                        left=cumwidth,
+                        color=level_color,
+                        height=0.8,
+                        edgecolor="black",
+                        linewidth=0.6,
+                    )
                     ax.text(
                         100 / 2,
                         pos,
@@ -249,27 +272,19 @@ def plot_cohort_change(
 
         plt.legend(handles=legend_labels, **tot_legend_kwargs)
 
-        if save is not None:
-            if not isinstance(save, str):
-                raise ValueError("'save' must be a string.")
-            plt.savefig(
-                save,
-            )
-
         if return_figure:
-            return fig, axes
+            return axes
 
-        else:
-            plt.tight_layout()
-            plt.show()
+        # else:
+        #     plt.tight_layout()
+        #     plt.show()
 
-    def plot_flowchart(self, save: str = None, return_figure: bool = True):
+    def plot_flowchart(self, return_figure: bool = True):
         """Flowchart over the tracked steps.
 
         Create a simple flowchart of data preparation steps tracked with `CohortTracker`.
 
         Args:
-            save: If a string is provided, the plot will be saved to the path specified.
             return_figure: If `True`, the plot will be returned as a :class:`~graphviz.Digraph`.
 
         Returns:
@@ -299,12 +314,6 @@ def plot_flowchart(self, save: str = None, return_figure: bool = True):
         for i, op in enumerate(self._tracked_operations[1:]):
             dot.edge(str(i), str(i + 1), label=op, labeldistance="2.5")
 
-        # Render the graph
-        if save is not None:
-            if not isinstance(save, str):
-                raise ValueError("'save' must be a string.")
-            dot.render(save, format="png", cleanup=True)
-
         # Think that to be shown, the plot can a) be rendered (as above) or be "printed" by the notebook
         if return_figure:
             return dot

From 2728ba5148047f71a77211a1837caa3cf99e8a03 Mon Sep 17 00:00:00 2001
From: Eljas Roellin <eljas.roellin@ikmail.com>
Date: Wed, 6 Mar 2024 19:33:22 +0100
Subject: [PATCH 16/46] tests for plots, move to pyplot for flowchart

---
 .../docstring_previews/cohort_tracking.png    | Bin 0 -> 34537 bytes
 docs/_static/docstring_previews/flowchart.png | Bin 0 -> 23797 bytes
 .../tools/cohort_tracking/_cohort_tracker.py  | 208 ++++++++++++------
 tests/conftest.py                             |  31 +++
 ...ttracker_adata_mini_flowchart_expected.png | Bin 0 -> 7766 bytes
 ...ohorttracker_adata_mini_step1_expected.png | Bin 0 -> 15461 bytes
 ...ohorttracker_adata_mini_step2_expected.png | Bin 0 -> 24087 bytes
 .../cohort_tracking/test_cohort_tracking.py   | 119 ++++++----
 8 files changed, 252 insertions(+), 106 deletions(-)
 create mode 100644 docs/_static/docstring_previews/cohort_tracking.png
 create mode 100644 docs/_static/docstring_previews/flowchart.png
 create mode 100644 tests/tools/_images/cohorttracker_adata_mini_flowchart_expected.png
 create mode 100644 tests/tools/_images/cohorttracker_adata_mini_step1_expected.png
 create mode 100644 tests/tools/_images/cohorttracker_adata_mini_step2_expected.png

diff --git a/docs/_static/docstring_previews/cohort_tracking.png b/docs/_static/docstring_previews/cohort_tracking.png
new file mode 100644
index 0000000000000000000000000000000000000000..c811ec94f2f011bd5dc64735567fd4cbd867ec59
GIT binary patch
literal 34537
zcmbrm1yok=)-U=ZAWDd!G=hSZAl)fS2?!D*(jg$y-60}^l!%n1q;yHA(k+5?r_$YV
z=W@rm_uTJ{Z-3`7Hsik$cv<gS&ok$5{^ALEtRRhtO^%I1q3|BcNGPLF7aUM1wEjyM
z;a{|LCw$;P0``*X_9|A!_D=e?MyN;n_SWWB_U5JrHyw>^?M$sKxmfvGIazMLu(!9i
z6J%qv`1cR6TG^Vg-8?>PgcrGNE%VF{g~HcIKGD92WtgH+ergXT#8sW+R>qy2RiB)-
zZ0>hH*KoDMxX-AJVfs_@@%!YWn|I!oUoaTeW)@R=roCb$c~xz!spx`Z|C2FoA`D+l
zf-uX^GWnl<cLduyPfrO*FEib~6+FG|n&1*Gus^0ke|tOLu3KdAIjP?(fB1;&>)ahj
z5x|Fl8}^a}e7Jl22qIroCnYJsmyKAAU#`KIn`r<2w^L>1>8~jYZ%U+X@mh?0EHrFU
zs|XQs+tcHWiRLm<i0H|^e0sE-qFv><t6b(U<G}Rgnm|uy3>Q{-TjNNHnOSP}dhKxt
zLxQu<&dv@wQ@&sbrNH;s_@Xk)1G$^~`wTa4db4U*Zgg_>^qW=Pnf>wkf$(PIHS@9Z
z!rE9GnnnJ)BTH}pIL)oCt&Q#DRmB#kNspFrMbWmQLc_~=cz45`MoTTCyGoA7ZMr`3
zXt~TsJU211+wS7)OOqo(4diKe;ILKKZ_Rfc9xdmamEf8D{R@Y4;oR_MLT(5RIN3?e
zQKrLf$!eUoRu#-2x6E&fyyxmFIH+ZZM;!PDkAn9{rNc5|OQOf#vkB+fU=5204n#ym
z)0@sqRb(_X54cb7&;;EOimb1{cZEv0bz{2WGA{1zJ9pk;QM+F)8`Y*+&oo;~aGtqL
zuMl_Xp0Y9_efVX#Hwhwc-apNjlvPw*Hojq*PS%RNdi4qe3ybS{>s7eXs=1DJdn0C)
z!p`qa`_dz@f1?Zx44hXATVzvDPuIN`zvHkqT9#d-rtV2R+sUxac)@2i(P<Q=?N?VP
z7EI1F(?p`n60!KZHzkN#EFn6Yl=2GJ4MBUowXq7itvD^(gP(G293DsZpZKgo#w{1u
zz0QT;31aGvmD^10=hqGpzH6;`MC(9v!FzPo_(zhs{rW_Fl#XjaQy>YK!;&ieS~vCC
z##_!FG7bZby0ar@YWEE^wz?C}u&^+cgrsCdjMreEc4wkk;&OgnLTc)5<f)tG*GNs)
zdJI>)aBXgGK6$xacYbOWr@%0%?WCQyB?gaO7#<8RDZ97W$uz$CXlbz5$t-owvtqvp
z=BEZF8Z|C`KYx-6JG0YD1?kPThPiLGF$kY*G<>z1Ot|&H57ng73ro?kCFI7K$Nu<j
zI=aox7HTfX74@HeeHE|@W&6e4_RP1ow<~5tgbGGiwd}fho*Nl)xNJRNE1$h`wDSl3
zUcOFE){dyZKT6NP)x~IUW%!~VTtIUyuW=_SI(vfaQYN8`ki*h*`#(R^B*^uI&kklM
z236AKIZ%fCYvXB4vhwoW%|T?d9`T8ZR^{Vsgf5Qj6Ba|OdxQ?_6E&l#py6VZr$bQ%
z1?(vg-d-VE5cmVn$gj(o>*5tk^<+FQlWuO*Y;M`aR{HMh=pZVnY|?WyR_*a)y~gHd
z@<AQ9Ts{l9&Wn20$7@v++)*7yJ1fI|mvC{pIk~vxb{P2ihxlE0UIaMiJ}XXF_8=k<
zxb~>-oj=^+@h3A5iJ~4uC1y{9Bdp3sYe*Ez$1Ovs#T4}PZsux~7{IN2!pg?Q6`quo
zBs;HN?aaX%6?nW}SJ&7?0aqdP?b|o^%_b7MEv$<dZDE7C>5iK7!p=~eIX&3wU<(Ki
zHtPSPG{K$ytG#A_Le9~V_xyPBT=t{OYRO=v&^sm?+tTa(0hb=Tqc#3PlU7$JH8nNm
zQG44mJ3Swzn{Lz*Ikf6XLrWX*?%hlawbyN0+LZQrOf0PMh=?bd)o%N1m`P8v)#!d&
z@KV4=-=-2{x^)XJyiJv6b!XoYZtYj|;i7(`T8ojdOiGIn;UnT@Hk2cz(4km}<#Go%
z$LxAI^uJ@U*6b)xc<gHvvud>_O9qy2e8<UlpnC!3=x}#X*QP>?XJTM$0>6PhOFO;q
zPtINi4UI`M)Q9f&KPk_MV1X)@uM~A0EQ+1C1@ld`*1tk0W7oS^!gq;E(4Nu7#pPsK
z_gv6$qOnmT$$7n6r9z7pR=}!=4pBfrfca!?jp{wPX#<1~YN>bD6a8f*;ij}cyL{;q
z>WiwLaiO|wu)W90W(#cm0u6^rhe7Serlw7=^L?)qC}3$J)FQ4QO%hx#dZVFV)GHo7
z7@a&*OU64tSrlt)ZJo->&dahreX8R^_ZFYpc%s^6;%jm;jgHGa(FJt$Kci*qWF24H
zj~Qr`2`o(EZOzZ!prrEHQL?Pwl*&<Kh6TBMkvXICXerxeDKmqV&yupnb$2Es(Q^i#
zXv(u<<6l`qEG$7iqQ^nZ`PCnwd`6b1G#>3PA5Oi-|2<sH@VL6GJ@sL@Gi*#U*r#>J
z<Lg7E7O_tG<tl9NzT^n4mA{5H91KmQ$Z}lp6R$-(L!$fl_6X(}Q4iNN)|nR*)dHVC
zfA0Lm6PT2A^TLG-zb9&{1||b!BA~5=75{|Ao8Wn5YTWgy$p;f>!&!I*zTcU&*5h@?
zB_JSBeSUV_8Os;i9WS&E1#qL4HdM9hY;C-1d%H)B%c%W&?ZFJrgv;WMPGc@uDAQrm
zRMTr!>w60xGXV_#@7@_KbjAp8)T8^peQVcp!)XInOvF%1UJ=x&j>x;S@Z<VWK%W~M
z|IB*AY}}KeZO}LWPCZv`p*y|<-gfzNUKQh=I|fiSps%L3%Qp|_91%gE4ZA3A_CM~!
zFFWzYPQ1H}nH^}|bO%zT@u^&Ye?GGG^Y@2do+%J2dc>F_6`VETp_<JsBoqsKChps7
ztZP(MA#KkMy5j_l7H$YRve1WT_NTylquW{?F3#w5rw<Q+mo1Q(cuVaW{Z!W@qTwz6
zpc;v!#2rbek5|LU3vFhlh}m>Hq0K(c*iBDO-RM_Q>`i@mP0;>#eUVYeoldA7F0lRO
zQs4OcevEN~J&$x<x}ORYjw`A94Socal%133XHH11%yPOXk?IF~FpQi7O7ACb)BB-R
zB1>N<%FFpbgoh(FV|Ap&9zMUHczSw9)b^MycE|Tb51RAZjXMl(?u}Y)!-g~9jJY%J
zND%fZ_0Lm+OoHvjo<*oQY()J?6VJ}C{bW^l8fiIRIc?4VVjzbMzH$uiH!B;3*U$Ya
zU|AO1+1V*O-__RkZQk}t^x`|_JUyrw2*!|Iy-JYz5T4dt&JAd_9>;sV6?Ti9q3c71
zui<HiFz1!~WU6Ek5)(HA;5dXzRqY~7h0nyq6ybWEm{_eUx1yo~y7o0f!u!t7{9Z?W
ziLKq;tX%5=Q655Jz4G>To_5t16qcx%e7PuARE}0f(a0(xfq>>GPol8mQmCJ(4<A0X
z3)fnV@<y`hDnN}!kqX$}4yNEUhJ{qU^HUDBGFp~DwyjuZ|3?`%+tYc+W~d229^k{Z
zn!wfI5VK4n9XyuT!eF?_=-`m({2_O27cM3Ld$YU>KdE5yLbE~E^hdGH?{7Ug*snY9
zdh%5JN%9zcL%(=<sM^IYV)QnRvdZ5|B;<L*kJ8uIH=?o{EzRXiLWg~k43&3fq@*L3
zn1Z4MYUZ6Mx^J)C$okd2nCMm8xJ+SH9freN?k!F2@ljFq&=<ZY@TPzCsla;rJE4mg
zw1K`-3(d3j^Rs-_$5>4siVq)Vs#Qryy!1g<GKZi3$CwG&KFX%>XH(PCTy%vfuUy0>
zeQajI+=GrUT(5h!MFMMb1#hXHkcz6i+-BB!`V9%c^%Uyz<;!#-6n>urZ%d@={4Ej0
z1O#Z0X=H>Vnhcn@PWx~ESaq~p87h?hMFJb6%Q)MBxnsdzFu>pcCJ)bUkP8K$CGNeK
zFJERRv5JVqx3;#Dn}+|RYlIa$+HOybS2;bc3OYUBA3XY0Sp;Aq?8Arqs;d2D+X->3
zq9W_C<Dg@F1&Ag!0=MwS-<8DkvfsZsn4;Kmb>wA~(x}-IKn^ArmfMVsUnGwD6vcw|
zNpY!!hhiP!!Z^{($|hU`16eEm%gf6NTLyA9vHpUPF;-abNZ)zcSfJl<uu|NURa3fR
zKG5G4ClJ1~V}Ip_AV$5{**5hRN`6$m$o|*^8JSB^_0v@{6oSZk+J%bBt*7tTo@~io
z#HZ?nm#Cf>+4=EcWvn9n?DQA`j(|X&p*Od-wdn(5Nn6tW`hwumC+^Si0>56?`O7?F
zC!5bTCnm`BK`<~ZHMJb|Z*);wGmosG$-<tUMetj|2}6crkG*Do`wRG2=+k$%hoIt`
z!WV!3{DBH8;(mU*Pbp}BsciI7An{#N5m#<}(Sw_6*}vSn0V3A}yc#Mp!T~hk>*x0~
zLy?-wZKY7zZ4>qxAt52Dn5VF!*y+x*eMyPbN?4&X!2$wByWK+T{Z1-NohOIe6@SuW
zhv4cmO`j2jRi+|$taIsC_Mg5C{LwNi1`ZAwBvcq&_f{Ul)^Y{}vs?vOBthjd0}Tz$
z47&aL)BkcZ8p&AWrV4!vJ(1x0_0(hjcmDqUR~y({+uLbGMX8FP8OoN3i;tVaW0<R{
zOM$7YKL?gcRHai<KQ5+2hJXvsXZZRWC8btRE;4H2#l*yLTaK0Y=N_dlFn$s0YGWFl
zn$p8`fUDq<my3NFN?#)RpNwjGo_D3AuB;Zku#+IjG%;-nB_Jl(nCyDLEpKm5*-@vf
z3IG3&^5dr=m_0%Ct&=*FU01`lwU&E(3Z}YoU}UjPAUhe!-yO%uyoe_W8^Qko_4guZ
zMDK~mBjY3f<A2^7xZD5Lw||H_(3t2I9bWrU$+lOn7JWmU^bO{9BT9O^2Ct2|beHXz
z8{d5Q&(V!zuX*o8tj`EmrhG5D`>hw_>)Kl$VYK2dYNtILTc;cvugT_-l@F}~x5}4G
zFiNR?bY*bq#VFRsaIvqP-9Kt~ViYdL@%9yS_tt%3o@rO?r6bYqm5^O<@?M8;cCv`1
z^!Gyai8*S;!{(-6Rl$M~tuj^?x3UT8MW-<vtm*0LXnA^G%Q5v_Sr`a9U@Xga$dl3L
z96q_c?j&^9Rb^O;HPbB`l5vy5%-N8oJ4wZk(0!9<!aI+XMO?{uzXne_r8Jtwv%ENT
zmW1$ub<{L}?EaU-ZvQyLm0%&gJ^at?r&r(nd0T+tckM?ybzw?|evzQ;oI~2m=cy^<
z^%Ta{DY6dy7vm)G>SGmTCBHDk1fQgZRbt{0vzv5NM&8wwDmL!&4+_d*KZl(tBP(mr
z;D=|wGSnL@YS@g6+wev0wd9hr-e^Z5hOg2`mybJElg%pcwcYEVqxD%Go@nlUCytOf
zdzEN&`)ju9a0vYkj>BR0F-gkvBch=1!WTw#6z*lqU4AhhT0L&|O+^)_hERYn)&0yq
zK&kkLTnoX-F)>T`)f|FH);G>p^RO~;GQ#VuR3h5Nwl;5%J!A6K;q?0<YG{@<OtYIV
zD?Sd8Dj+a0<!Qc-$rDo<nbv3VY|2kpf>|-@<735t)0OFM;-@h(M2zXyrA^iMpbp+N
z68zY+#r{whR_11r_F6Ff&&MzCuj$$<xDRuTQmQwQD>=Jc1@B7EX&IK|Y}QQ?-ggc9
z{(QfMe1a6^q(J^A^H^>!n0)iLadvoeC+*x><p)?wdgZpvER*pO%4QZMnfMV2o(RI(
zAG5<h-dk;W&e{9bRE3yDy&h`R{TDA7Vb^JS?EZp^*XO1Hb2lnp(BV7$RKBj5)w2wR
zI%639U{(T@=^q$4iwq&Ub)I8h=cg0GfRYZk7o%RNz{}VzbZ*v8)Om@~N{8I$=Ki>`
zVeN-U(G1V*&dO79aWvT1bW}7bdQER9!CzFPgyFwlWw2A5$*|ng&TpWaBxX_bsnQ9|
zxEa;CoA@;I<m7=SyD*hJu^3HWdDKS7xwlugTun<CUAmKGw_>C>`%3VLL7@4&HuKNy
zt5@GnCVEe|Z!c_KpPTBI-DC~E^v8ylgRfR5VA?)x`(E1XN-{2E1fhk!e}B)`meXt?
z=N1FQ+soJJ3mjK9B3YjK44W@}GRylmyNTLpz!yUmIByyN@H*V-Q|w@I`|f)g@W=<G
zL$hdn^(SIhEATu$y5o>qQ1D*-^+hOrjqiw<)hlf2fSZ7pMH<!dWX+{V>|FTPty>e)
zv(3TTs`&@3;fH0qp~;<wl)C({XK_p(e{{h`nG{=Ml&*%-vHNs7B*xpQPT5Ol9=9Ew
zp<}Dydk_1xEwBi9*w5_C%(+LFOpC3!oN?~exuOfi)_o%<R_xzsi&k)-7CaiBU2!{f
zqC8r(Sn-;7E%G`dM@gsP6f$V3Cq5#{w1X9Zclq)yM#f8lP4~^sHzTBAy08ckwyN1B
zR#H+jpQw(39!x+)gn3Uw!goKe&j=p5)0)<V*SW{h{yKjB%5ZT&XlTB&^g>rGGKqHa
zP3|a|btQ<z{`m12wrDd<cOAl3KT{vdN3(xB7Ubemc@>XpnmogEBA^eXmLz-o%^Lmj
zM;p?*DMpsF&lA<RL@Y666*B#H%3t|B&}`E>&bM8*zA+rK-0HY)=5}8wF5pJsa@#=v
z`__kRohhka`&4KaY%Q>n?~R^4w>q&3aBv`{vMpt$EAZHybgP>p#XhJ=&|p#C?|zwB
zxjI_L^|GJw6OWmcwKe;rSYAv$pi?4P4S{L_F#0oBlg~;6PykHxS<0F~A~nK<b_pAs
zhMoOAuv6a~8rph#*f{9i1*^sttgWpB-d;lawN#~dvE9BXZ_FMOn25d>H+#?KrpaEh
zKK!ppy(S};MQz=zu0YBczdUvW&1sfopf$FIJnXlJrQ=0t0MsbS7-<_5s@z{|0o>1x
zP1w;^^HAGXhWJom&mG5>>dA*c)(7VKhSxdKbD`xPxo$(FxL8j-_({sCGp+ZAU13v6
zUhEbB{@2AL8J53)e7;Rjztv1(b(NYrv}N`)a8g7HPHItY&+j-ind?tmJ(V;vF+r%k
z3Agnc<;4U_j_a9OD_xUUiNpMLt+@$auVK~C;5KhKw+@&3eE%pyx1~uNOXxUrr&47<
zlj~W|0j8vDV&e6mX53J9=1dGP2h~5{Y+$qvtC*fFncxfa>om6`<k{(wz)iTPG;*tR
z54BEmM^XHWz^IxYUNGHOR)7R<W$5>>uhrvjsb6S7YaK04@OfQ#J<6!WF!;rcho|z~
zI#m<Z<i+CIvfb>>vYv}KS7QfTPaP!Vcb=rCs|tNV@HJ4Dv%noBDJF`d+>ckw=(dK;
zn_>kj$JcB-Oksdg%Z}hN8~Bm>aFb12O6nrQuSMw`jV5n?`N9ahw<f>LYEr>+&B`oV
z&}pr++GX26BxK-4Y=3m}*Vn0DLQ&qOQHFSCzSFGV9qbu~W9u^eMb%AaWGDQ1ikKpk
zWP*|}U|%_x?H)Lv?iN-?&m|FXa@i_wo2bTc(=t2xKEn}n%Gl0-|LpA1bN79!wN6@6
zNmfoZFW+KQlJysr%fA(DJ)bF07rhNXxj<W(z-fEMSvPj|tmG$K*of+C5-TtS{{EMA
zyiOfvW@d)Utf)hUw`2l@dwYBN;}xJj-@AYR!j-ov{f_kF;^IMMTz&!97y#ER*xPf5
zik<Nr7gsbMrFO8@#l>)${46$69JAzuk(FI-*@~7bMLN{D{UZ}D1#hRDc*{hM8&b(!
zet&)|BO~+0^69+r)9A{?9jiq(R;zZZ5YIh{v_9tTZ?v|eTi55_Zyj<GEw(+PU5)j6
zucpu1I$0JV(xa<(=XB#3KTZ-yl04yMekNCJm;1r=+Z@fZ5U3rS_2~HQM(rO7T~1C;
z5VEBiAu-+Eg`rn77v})(?}aPOt37%N1Is-aD&b%7uC!PQ`6<Q=M!*oz2Y}YH@((?g
z`-b?*sn;?6{Q7zi(9xzpQ!fLJnFPBSVQ5cI+=IsuwAI5!hPN?r1;`9;m?B{G<n%l_
z5X=<<Uhe(-tB%#jm&DKZxDs#DC<khUdiw2p{ZT=%U+2`Ir_pjkv95t_R(N7Q85tS5
zc_y8S!)J*{!v*^2nq`)0@cI1tbEH|IeH1D}z%LCQkAKpYskA5^16Ywmz~<Ti`3?oY
z+u7<Ff{B?1Zu(Rz#0wU}X#JOX@NX4Z&f%qo8H*syPyD||i{2M)K1)2{J}BHTh$h;8
zCv#dfvV!ZmItVWiXOe6<yHs^q84r7@z0k6+XM|D-5N|gb`8z^*$^S%+_)h?`<u$bs
z$9O@5QpN?MOlGFcOkZEW0VggA4FA9Y{nhp-yW_Z-G?)(aq#}!Sl&{d>%dqX{J`d?1
zrx5}01p!*=1Xo&Z?9E?EpAii4e<GS7msu}RbGQMd!9ztwle3e<`Gr4&`P2Ydo&$gD
zFu<BgHpsJD|3<$MbrKiZS+{p`>7IX=c-;^&Hy$>_j~=N?=0v;}{EhKmkqB0y)|ZR;
zxo)#N#TPc#LL62)WivNL{o~9mO>H;~H5_5$BEm7A=D=m4=i=j8@2r=~^|!@%zt%qM
z)g}~EbBeV!E@_gJ;-O9n-<6O3LAqj$lZjd3?qrXjZ~5$5-#CFle*cTdPh)Oo(|1L4
zpf|8DrLtyf-e{PuK^xCJIn5dK6vw}i<A^;bnL4+4brpB+omB3nB0+U$YqPZWI{LyI
zzNBkmKYwbwP_}om#5m9NtTyhQmM1GQkNLfQEkM?Ih-Ig(jHQ7gy)}h@o}OIS?qL|a
z{o*Z}dZ#g`mdi?4_w64qGc;AN!5m_eDIU@foT)kQDA_hJO#R|!sd;aP<8T#TG@Pw)
z`&eALZ?3RjSo*mTLwyx>vOTyi5SjR3(I3N)x8l8}Qt&eed;FL;heVN+LV^LQNtZko
z<kGd(wQ~0WX_g%Q*3mgJHO<_K{1&Fw`-KRVZ>iI^c<h;OVo%d!U-2P?Ao%x<@WwJ}
z$EauN4KppLBs%Y!%oiH8DGnC%vp3@o4A#E>mWym1&Kmesi7*{|tE01j*#y0MZ#4Nh
z{xBeqwhk8)4LM`jW8_&hOL4tn2;{Pzlez!k!DW1Wlc55Az#J`_6R<GIMBVv;N@W2p
z<m<~PCs?xx(F(lQ{s2NgzK@UBUaAhawY4pukJJQ)|28XY2-E|$3^(Oa`DoJK9U5F*
z(edMJ8TgH#*j*A6g&l3b4w!THaWhtb-!hCB*CHPDu^^Lv7F$#&HOe|8chtx`+wrPw
zyVs~}!#wKN*GnN7NoK0otP+T%9nm&Uduy)8MLVvN52xDeGuVg7+#Fg!y{oJ4s#+ly
zT^0+;5W>IJxyG_DOpqCO2g{+Ytqn+qcbZl;5qC993(SW(rGiMm&4!9~w}#Oo)Gz$N
zZ3c$E?5AwlI5=e7FRntZ&rvV*0VRi!l(co~+iT!WCTycvo_#O17*z-A;T&Ww_$_@<
zn|$BADKzdP7d<;LM7YRtr%7$gQIOAC&(HS7M9)v`YR`_$5O@Y`81stazd*RM*HIG>
z!>yYi+4ik&jYm9PI@s$poWKtq)!3ctRuiendZ@X+Idv_?*=Un1rlRzZ<Xok;d+q+>
zC#!1zP-{2gRN^^Xezw3%(kPMS*e%&mvlj%E55Lm>h-IruI&kVoacVWu&h>WJ7rnzs
zvbpzV|BxtHBW$|8#klYXb2?LBB|dHadM6h*jdJZB#X5Qh1{GD+Mt=fY8F_h(Bw!X%
z(EhNovHgQjE$;SLt=9(r18_Dzze8mdBjUyj#5b&%pjYTv7p`8tS|4jw`zPMf1Aszn
zPtW3N*(CkVn<#JSiJbPoAD<l!))8L2rjo2ZJUo1|op?S6H!02Q{0!9@FBIkL>x+H7
zeX`wykf&<97VkB#F5)hHh`!C6Mm3Z1b&hj8wQ9@RsYQb|)aqdzSM<&7_(v`Hy-djj
zpiMeqnz3Lt>fZAe@pf{`qwUdF*Brkpkl|tXLMU<dG$7~u1r%$Dd8clbPH@*uO9B0=
zHAMnv4_8dPJnJ`uBMVCX?mcQ}t3}t5w@n;1E!N4Due1om8^6CEeVIy4vzP*l2ZYEZ
zxKu>8gDHc<dg@+Y)tcb<@88=y>rDITfff^g{P=Oe4m3r_wXr`Q<Fy{ggv-B6&4=#;
z?O2U=18KCI&cdJ&=Bhp}vs+ZqVkH5lZa@9a+qZ5RwjEriKX22^M^Aag$HY8@DTP^B
z`la+&uB*ju@}oaUM|>sF^>{Q<FRqa{qy`>wtuCJ1+~j**=9IIW5H&@8f3bCD>eSg&
zp<=#;KkyuX;(b3s{V^T-l%Mg9;ZHUn6&1DgnQdzY-^VJ=?2ddqeO|tGsZb!m@nR9v
z?)!c9FXIniinG$+9MTZZb>(J@-AcWFvUgl@ca{{T$ra+<r*IbXPG(ot<>;3|VZn=D
zn%bi!=8;m14iHrCXRGB7)l{v#DTId#0^=LtfFqp|-Dk|c@6m*N#ziZKHV}P*Q<}N8
z+wrngulPW}_`n67EwC-9YX&QQhAA$5I=W$IQTuDW%5Q_G12(9c(+3ACu@fhmkE~`+
zy{eL$pL4o&h?21<<n+D#SsF~k$Gfa<@UBEo!e`$gt)b#sjzDRTTaKbU7s^$F!JzDV
zC_&CcCeevIw?7SgSUz`5q%d`uksI-w#j2+DAn}zTI;Wsqy}LeTik9ZD`luw&lP5-g
zXKg$dkQo=qs<1>4fw?RgHvcPy3jG_c<|)_rr$3?qz2$*|LI~fKCd`-s{5pHfU=lMd
z(Rfzl%Jb;cb+-OTa&ohSx?X8kleNG8r1uv-%6uyubOALsxTk!7wVUv`Uwgd#uDBW&
z)O_FJxFHr#XIy;Ul}Y*D0DRo+VfncLeERGPe)ZTY!$$!{*U%eCM~f(uJeDLB$VU$D
zH|zd*ZFo35SA}Cr`B)ey{HbbdhMU)OD@Ss<D;61E?s8}SrF;#9qXg@^CuOHC{SR`4
zZRWEZW~9UG+1Ks74G#sNe3(da*ZAAqt+IKhg)<Y^w7AwLYjijrpTQ?Bk(Jc`;9B<6
z{Dq3e1VbQ9fyWO8GV>!y)V=WT5&a7gZz~i)nTHQQ!_eKEB7KdFtPR*&^+E&8DyMa0
z;Pb~|)S=|JP8xHS3Z)8enAYoz<~V`P&7xiT(QU2bEr>1qY3Vv>An}m_TY8N_VGe4?
zG;s6AKrlylk$Y_N2o;B@Ofk`Xr}D;neItgcbG$b4Zif7<zszoRotX$l2z9*y=QXGi
zGCIV8_-F0flOEqER=81TN3(p@H6C}Rw~GSy@#z~PS~A=`=X%WQa;!rIz3Z%=wNZ?!
z*)a0G_Vs@JQ_ID@bdh20_>2o%#3+C8T3CDLmW(z`svS{mCIEUsH~SDB-P)h2LdN<0
z5{a%m2O_G?1hUbCe3d2_^%IC0P(ZqYr|5*T2~2CUQi{~h`efbrFd;85FGR2fU0)ie
zT`-O;j+o^ejeIqoPe6pdqT*sQkYAz6oJ^jdQUm<$7K`c{s;*WY8N@2v+$AGcS?^Mq
zZNtOq5e#iTTnIeXJTN$q#NIB)!?$zFVd6RR$bzNy`}~)fY#K%V+bZ6z>AYZ*VPyh=
z__iN+$z)qQJ8$2<?JFI5jw7vinzMaPfLdhS3BzNtMimHF0=;Ds75;L=r@@gfTiZOe
zv=5dC%>owa|06>8f3fDMF7xUyzTR&Z^&d`-ewv!cZ|?XHKzfVjo)(#t<p)1Sacu$5
zi2|iqS|Fuc1s*(5LcsigO5grB?C!t5ZNENft2LH+gSgm0hkzh6p}+t7_xlfkKmJ8{
zgQ8_C!fd&UOj!~E^1n-BFJH0l2ZRkynqTfyDlDgFJx7*OC0ctM3rChds;~W$0TBO>
zO-L^i1;e%@hTn00W*gX0i?;ZtS8hY{d`k8(%;!P&yW*h-*0*<>*ixB{&2A?6*kgxS
zrcV8EA|hUu`5jnnGie)Yl9Tu<<7SFf6|HSvp=2pT(yw6k2DW+we9U)Ve6~)f`(^U<
zIInW0r{yXAo5TneJq2B`qQXzNDRG@V58uRE_&Ox|>b}5X31P=oBuJ>9oufX9%(=Vn
zlTqL(#ISo`JigVXUCi{--tQ1pxm~D1(a<K2|2MWN?2YTwvWIVy_R922FY1-!STCmO
zZH6|B*`SOyZ@ReEvX+^LGKVMUpws>A46*K1JD<UiO2!rqlIXtX`T1@^i+yu^I4*-v
zW#1dElbdG&=<K+BYdNpgFU(#tcey+8gE;lfibn6FOG2AL6V>1ir*0)|t~~=AsVR~%
z-}<RocHq{=*i-fd>3606ULjlgWz-u}q1K5b(dg4ltAYgK!|#Tv-rY^UYix!>#fa6s
zVOS?E{jGq(P5n-!l;ZGhZHu*0<g?P{-xsx{`A#DB_GnYLJ(^lVEG{^SuR1EsQNAy;
z;;5*qt)IC>3nabVjeQ)8S6(wH6er)=>#_Jm!_Z}M=k<(2X@`1($=H(zlkj3n(GAl*
zo>)>!Ha>0g{Ye4`n9cfu)dd4bw|C``tIT4wqb8Q-A1ENka4N?MZwhHMPh+=ZJAA9A
zANayf)-t8Bjy{~s%vwc@NlYz^Ra5ot41rj~&GR4oJfj3*b|0tPvo^|UOJnz4&b|?$
zuHVjYwkFXl8Y*nj(XxE}R7;%TmNxwq$p_yivNug+Xx_~8i^W#e+%kAu_XGZaFn(@s
z4z;lmYxNT<G`Oa6HOm%2T&>!#M!+tj)A{(!D50V{Br_D^5oq3?o1K*84GJ?Me%QA7
zS{@rCf8}LVpcl8X+a<q@OHOM9LCM0{eWiwRq)IrJd<jW`YxQD(?yMggYd%g}wr$)+
zC9$tQ6+Ub+D<Gk%?N|xg65XH6NOU%K9!&Xkn$n~0r)#ZZKAl^Gradvy^xV?DBt7;H
zp2TaMC7^=HcwSxy{g7F+)Lb80MGz^+g$8!#^_u-?VHZwhxPWg=L-(0w^E$+_TkOW1
zvhq5n0CqumtCbcdeDpgdH6w!yC~n(jx3j}V&{H=>$@wg`7*4NeTJ4on4JETC!?XEg
zrbA;mJIOJ|N?AB=O~2Qo@?K$JVV|*cDEQT@VPfCqA$r^t=da!D!Vy_O;kO^Z`rze<
z!y=K&DId}8;{K=RwwEg%liYese3TED4eFJJWnNS1Lywo%efa?SuM-h*7&Kl2uz>ag
zd~hg)1p*P$2*5K6TwH}N+r~O>t2`i5J^)D_j9jdMrs)_N{Xyd^22I@O_3Mt@|AnMI
z5G|(eZa#%6tX#QF@A|yK^?BuwCDfYE%sIEZTG7Xbv05iiLiV}k^e*g0ccUVxZ0s6$
zFC;N8qt<dxc_xSmbm!ZI7J%jSooQ3Vl#gC*N;FLvONn6l^8Wj5-TN{xvR5Qp(t68E
z&#jIhn`wGHhUWnmiF@Z~r-+yD^+h~`srpx1PU8;1s{o2^JJ_79INckqIShRF?h@SQ
z+BNXYAVMer0&axgiqf)u_-1!_)p5nqEq7Ns^%iGLZgeLxiUrMvf$*;SNnnN>uG6L?
zR%B9hdGQt0MfyPW^G0UW#`T5A(S_Okih9Cmcm2@c7aNMMtvr};WqEb^0F%V);ZMeR
znO}3%Iu9OK*`WBzcLji+agvXm>B;eFFheWwE1~(i`~a1j?-KU?tHn^V(6o(_(#KkK
zRCv>?yIaXKi^e$g#a&g^_uZx89kLMPfP^~(!^<rA`W7Od>P+`aQ_qW?AL65|sy8nJ
z2xoKXeh9iL;@B%xhbdP&zEpcoOaJ=3Qgj80nKjMcm%t4ibKf?D7IPd71+CB67~ECZ
z;dd`7iqjKIlhOy`HEg`ETE^M9uKUD<Q?rntwx&mM-hchFx6nHOn@r8+EH|1(4Cxa(
z3~hUgXwPF@9CJ4uG<X#HOKLy8oL^OYOir2G3Vge9e(&D)r2b`=GeR}%73nSfT#WiT
zlYN~o%Pr<B2j*d`mtMNztetQRA7-jwFsON2yhyF_MoE2BlHAtvrM37C@3!a3PU{m0
zmTe1X2nXB(E+x<kOaOfZQHfNJ*=5Mbd<V7u5a3r!lKAUrnBWjWq+uX1j>5LIyP@|>
znbWp@gSE;>2XCJSG3WdcX8FlCJTr|nW{L!F{s^^+75F6Qc$lzEu3Ty$pTK6UHBgk0
z<$QL!iDRxx(=~Gm{X4V{R1>Y0<S>>E{grw`(zh8mU!R~ew;mbs&aWzAJ$81-!Ib8k
zA&CR-o5nAr;9!4$3cIMxLp8AG!_jk2g*;cwn`-$wT_5P=NcpXA0t1_)Q}YSHVGO8S
zb{kXTYc=~iI?mIWU=fgqDcWo>kM8c>U_?IBZ@B2X^ZOfO&jL%X%jxbQVpAP1x1k01
zUAF2es9rE-7MS$j_CcpC8Pf#%?nxJxXcJp8TlTuKYKIZNmGN@-YaWe-ww}SDpdjqQ
zO1WL0@~}$Kyb+MM^QW&3AQcBk7{YOa=M{BwdK$xNgbncZHh6-8S+N7(KzL$e4~R0I
z2wejB4uwd&y6&5p;5=A=R_pE(V5)&obh<yMd@4ckDA4=nV;edN(-JZ<8BraBswbP~
z0hNtsfERu)%rEQPs>)yP^GOCc1^x$CXgtfrO23(RLX%AfYMsn9G`|c>1j<*YyLQKe
zUIv5&XRp)C@GyR1ZW+N7Tjr$)JZ9Vd<MGG;L{I-0;`-mn?f?3=eQbg4*X1Fy;$h{<
zE`Byjxkq%|++S;BZ_&^&G&VIBIQ}a8H?w*AQ+NgK6<pf(cP`q>-j_3DfBD<E0*m(O
zzlbbd_x}W%{>wFvXRtjD7V!5UKQ@9C`WqgO2Vcg-jy!J46Z@M4XuNT1T*H`_d#MXs
zb@43^JD4%*&3#0>XnvDY%NI1rhGhHeUNI@Om&Rr-vuPG2Y>3-2B^FP)iltT@U>+5l
zD_#7oVE2WElVa+(z{^G3-@3HhMMRytJQwoD6_v}y3@drlE$nZ8yTW|TwRJF<{+!r6
z`(#}yuoYz-@*=cL`DEO+6N_RfF~MoT1Dp>}Utd~ERo`!9zfb9zI#<qU>vn0othZ3<
zQWq_HbcF#QrX_yENYl?-q8Oq2??27bbWTe3q)#_u3Po2LixItXAZf<1S(UVj?xMHV
zmZa=pYMeGnH+qi3S;~o`!F&37v|)kv8Qsqd7Q~-?57DiK2PF-n!((ViX1{8*wzAFL
zlcMUPz2N+Nix^|*?#Xh?9M^zhakPv5Jvpm;V8gt2y`Sgi@|D!>+0XKvTY)Ij{yXmy
z9yMGpIu5%&tth2)>&DCjlzPD~qu7evqR$tU<x<?eK5H)xX{@687mhJTns@2*UQ)as
zHlG<CZ#gt}!%nzHkkn<66@C`LZJ$~3X6*EjBHhomn%e>N`e)A<NQw>y4aaP44CB;r
z@@VkOs_S0m_|RCc4*14a-DY2yX+BwuqAedwz+OHyTu;C4cUWvLjeV*iF<=}eF(<5?
zp?gh)3ZJd^fab2o*LuscKClHQc<c=;AA&?P+Y-va#B@2M?;)@l(Hw?S0m3lerGV;y
zfq`+AoLm+-n_*}VC~&FG6m_uDLjQ(ZfJhDy6If0SH8E_;Q?<H>Rlv4dmPuOHk|e3>
zy}!HA(3Em{G<5#4^9*{b<eXKD;vkDK$}1Mr`(7;dwIkle&m9`<St_&_R^P2kwiBA~
z8$MosfjT?wAlM2Nj#fOp@|;vEzzcoV^J_@nK|-&{wT6T7H-+v;hMd{^X}=U2B<HGC
zGZGiV=^xF2A~yx~V&Q`v>n#C+7yy2e?Teevpe#)Ru?bt+2T3v@9#(KKN`+89R8k@U
z_v9j&6c-`TP(N?G38SkXC((Jy4KO)@+hB(J`~V9w1i<5feZ9k61!^oNCg#K-RL_#W
z!#kX4+T?F9y#KUpFu19u5x=D$HfxKnhpJ%ZyQRQ&tGnp<(p56OZPE3s)SZRqB>HD+
z8j@IqB<731BHS1mvz+b4=iIle91ACuDtbKcB&cZ@GDn!Py=z%8S%{An4F9l9!!$PO
z)}2{rmVslZ;j&Rgyds>?U^clvZ5#YEKn5?XIEj^+IS?!a>~;(C0I$&RZ7+0H%zcm>
z0#PfV8e~YQ|38lfnV15=0k{dRo&%_CaGE1t<E89;@cJyE!2H%6>jgqI!hb_UME?Oc
zQQOg7U?`%#WZx>#kk@ij6241=-g<{|=JhU?4aIzZB-ayzy`BEvH)%r3>b1-f^q&<h
zS0{Y+-bV2DzA_i?XTP8k=2!ocm!n#n1r=CIy{(h8(-Fo(b^MaKu8#jzy8fB((CLBe
zD96P_(q87RkLOi_pRMdpa*fZdHXX5<ZS=v!_oV;X<1(WTVxU}qPu6**ZPkIVxq5d%
z6JcShtA({#qxRc}z|GMN_8@5B<umV?N!j&Kh|PnQ^@=xiLF-YAvQ4m>vFp{N^6Ji9
ztu~x0ABNK-F7ox|&Q5ti%I=QwA~>&jarhh*Zi+UF%Z3E*58T-BjD9{x;<}AF#PYmq
zEE<(j;;g|#n_rMagVxG&V4<<EC*>L7g?68rbRKvFJaZz+@;baw{f*7$2XrC16H#gg
zOFLV{VG~Jt$;|4p(g(99yxU9udnk2<Ln+?n%E0LJZT*P@Ppe0c4<x#29ZJ)?|JbVu
zu7jnw6&7MEKusj=f;4MDBn|VnF58`8X@E!A3w%5fY%Q21TJ`&rUikHp?D+<<jqhLx
zZQ`4Be*iilLfEzEbo`TNRMJhpU_LfL8+LFN%T2}ZaB-<-!uyJzv~0+}%jnQel!g?h
zo`?<Rc-w67eaf-%Dki6WOIfJps45k^%{bE4K(R2#EnUXiEJX=VJCC={D7E!RhLfhR
zMkH<OPUHQt3pmovHc7sXXI88zbYV_?NVQ%V8wwNPb7bM5GvkQ0*0{@`OKk30cY4`T
z<8I}PhZ5bg@xlEi=EH;;eQkhi%2!KA?{rq#&da*)E;Ye^leQWGuaeySa)0JE*n$S&
z-GGQhIK10LDiK$d^*2ls<P%)p`0QY&02g-`&!Cn{aeS6OpZ&|Z8yv-lF=XQc-`vbk
zU*Qf=cTQ!@$&a*Pa+N!9Yg4=>T+=<#FXZEi%4NuNFZuLH;^Ih64;tro;DNrbpa@|@
zd9~#2RfE@;g_H(+S1#~UvmR$X3)@!8PuHKYkY9-BmBrdPd1BGZw)H-8ndVkwKv$2v
z&ga<74oBy0jJP7Ack_+w(~s73*_67I)RctXT{gF_B6!6lyW2<bZNoQ&8hd~*;@f4M
z0^pQV$T3LQ^DwRdqdAmX7cde$E4RJ{(TD2oZXqN#1GyU%gb%s^t9P$^?hDP#&JLGa
zP=GCRF;dg=78_eANbj~_v<`z#4!VPSu`wRvpwV_H897mSc0f$B!f3NH$2evpMrSA7
zBKh_}@B696T=}~p_uH{nrPSF73;0Ge^uv?K5{pu|Brw$5+Sv>mzEMW5$z)o0CA1zI
z&jlz|-jyyo;HcQq=zHR(T)`;n+x%I<7H8bOjxD%saK&UWKuM4w?C5dE)4H>HHm`3z
zJ)wZO?SB8b1@-~NQZGAS1@8xuvH8ZV(ZQMptl(yg*jZS9?GZK#h{00J@%FI_I{|Q3
z*x1;d{899pflZtG<%>TwCA*HhW!7`81Q0EOtdhj*D$-<@zNmFUL#j};4o51J;qKY_
zh2ikXmTNd=5wd*V5mhSBDui~;KQYa0ubrPnVTXzpJ0FvEsF$yj;*{MGS{A0Ei7?{C
z#p{i`^AAy;>TpnF1&wH(kB(<HhG4vWCoLaIhSAu5FJ&_1FJJD|)ybNfTQrlBlB&tm
z-8ZU~4*tJ@g-HtQj5e$;>)$BYWA7pe&u84yaP%VwqFfUQ++w9q|0`W5<xz@RY31L@
z2b?z-_ZfWi@bSNm|FM4L2U9W%7}^0lIfSZfY3unFo>;8BeZ4=Po|zfP8>|e2|M-@o
zni?_Kzu@mBKIgp788x-(%y0i2(f6X|(~m4e;Jqob<G#+xVS(kw5nFPU&HFFhh#m2N
zgF65Ftl~dO#;`lkzwlw92-qz+_aV_cd3u#oo~(CSKPQ|fT;4*H0*L|AgP2L+_}f_?
z%-<s)y4T;_Q<B38=%gL+Y_`@woPZs$2cXt22DxNwN<Tp|e&C@WczJ_HnFEyS4E{s$
zX9GwmB-37?M`WK8VhQ;*egm%nIB-Jf49UoW6aE`)J{{&=cgPMP!N!PgG0?X-y+q-s
zSR)2!H#dL)|M2mnZ^QIdav<vr)bB6f1_J%LDLo;>8RqvpGb4e9i78V;qsnm=l*HR?
zZ1~x~G^P)czX-9e8-PIao74vKAV39?L;zkF0i125d%?uUqtO$k>)8eY35~S+Zp*T5
zV6`G;a4uhFidaX+_BSLtQvhBvVC-{;5eXTYz%vVBpl0l4Bx;y-z4hrJh!gCw*=MIm
zkQ)%;n{cLetg0c_prcs?3k8i4=|7l!o0t~LE7CX0($zgZMR{X$HPG%kP>Q%-uNPW%
z{m^ABG+R+Jt|lfXW?6f<fW3``8e!_&$`CsZ06p^p(q?<htaK5D7m^a8O$M0-SH$W@
z&aXWiPwF%%R2}jp5IBEv^duF#T|>o5FuYAdLE#FpS-T2&36%+dKujKUUCwQ6Y}6|o
zt$;k2%W`g+?8ib-r6IdY?Rju3RAi47Qjy;$CljG<qmY>xehJvFZ}2amE*9C&w<}v5
z!p-<IQo;<O4QP_vATT`1N(6(>Jj4jGZr-_r9o{x~<DrnpBW+2GyDThoZ48O2aNGGQ
zufVuh4zLN4u^$UZ0M5vFoh%CvZMUu(pV&`fN%T4sc=P7X$!4h6PI`<yJtQZ^tf%VH
z)N(bR=pP#8dU3_bBauZUFA1h27cWnUbTB%28ciOn3|8f8R}n*C6e^tDye1TQSlFQW
z)ZH64Tf%LaiZMxm-G~NdkhN-s(-W4N>~3x~rx>W@d6i3dU=J3Ku5w#V#DTkj9O9_P
zUJAR@Z>fKRht9{x=hBrc=|HA|qBI+{7$CFzB2^(37u-Xssi_}j0>PZzU$I`jeWh&N
zkz%~k0b-MDs3ZjRQ3}}lLOXSS{TeK3h=kc3O08&YOb>0=E1(F;Xqod;AnX>vT_C{V
zu~YcrlHHGD)pmw(8%Jyx0!R>#7kEy8fNv4ydGv<?B9Jnws@EZGj3Fs0i9{>mk;WC`
zgNF*yzZZ&oL}w914Q5m=`(8<+1Qj*SMA+R4US}@r$6+xsZ7@kZtMe3js%3)-8=V2v
zVuRjfNu(V{xO$=4+Sy&ca)q9c4~btw_@znr{OBpDW;HIn=0o@H-9sVaL7TO<eG_nZ
zY1y{XMdWR#BGDp5vlBbp48_32+$^g*4F(TbB-}NqsA#I^HFXCwfrx95+GB^nusMjy
z27IS3U}4~R@gvE$ClMiHg|~5$P#*FL)y4#v!E&v<JNw(IP@%4Z9m4nWHE?(!vYr0c
zTpN^OPVk)B_GrR-f{a%(h~+1{`CgcM(B=`tE@VIRM^*{Kut{lVcpz7W&bX~l>R0Ht
zZ;8$hfM!)>5mqdp!ub>IOPAOI*Q;cmY|ge&Ky#Fmm6Z+HfwI&K93%(a5WEKVAkf~(
zy?b+>aanzWmuZ>h&?q%G#U<x%g@E0}&e)h1EfZ540*|Xrq1y0KUU^&Sw60?{0-FRc
zut_<q#z<f|A2iJfnRN&%Av<|}d28e)sbBBs2UxR6Cb1wv3?|Wtmpu=`3+TTIR%8m8
zG`K_$rqB&x?OoIZo^6zog=J@<YAGu_2=aXqf?^B|EvO)ADwovnuI)z^7m$!Fl+Ni#
zqhC$K(Ma>Fx8CvP-dXp6Ty(%WGYoK?IcnKqpFS-;VrFDy2fs})5ca=f9k<SxAdiB)
zPcRRX^Tu|S!=I2nnLGg3ggJ_Ci80&CD8iIfDl;)tpR_sNX~Oj2_?Y7G@NnWQhhZ}*
zq&6o8_s5+QA=hROOm&2yuEfi?z@M~lz&0;>(VI*dT5Lpm@f;i(q(D*m>A(N*Va5Y)
zA9(w52l91wXX4;>rtAf4^7$;sT7lR?5-a?kC*}JmadDL3Lq`07U}u0hE+TP4QUT0?
z0sALN3Y`C^9j}5F^Y#1|gvuJK^>7RR_1D+|`EPcZ#gODCi0BQlH9CNxMAE6%hme(T
z*UERaee_&MN2g?onqAM^X^seKFEsr8(cq~?i`#YcYQW=rtghY)Wnt;jF^mh~Q(N?-
zuw>k?hLHEHZAJNFf}5;-Zi48M=97#w`QoA?M5G4G>fFLY5hMz2dvNgZ=olC>q&%G@
zQf+~~`uG{9eTZ)-HO`uY(Jcs4N{9<IIXPKY81MtuiKET;hZ(xB+&KBkxs2sAIVaq=
zAHjGAv8yTQxpVXLy)Zf|ufF(|_3PPw8cRkQ(Yt&a1-LFB99G0ZgG`vzUMGdk%bruO
zu-Id}Sahm~Vj<7P#K_1P!6zm*=~gYMfY+}8o+H}3ck!ljCXXSY1NnV3SgVVVV)?A7
zAfT=VRdPO|4A!u>@~)JmBmpt8W3y|=Hq;|)7zVCFmJd{GFpy6Ks&aC2wu`KXdY#ac
zXj!8@@WX{-`wA9TNoUDh8fCk$=?bYK&cakEq_Zh6rzr?}jF0?z)zZe9E+2Cr37)Xf
zva=U->ZO-P5Z%^x{gZ|Su)skdmtS}G2_~*^=+H5cthIq2`0K?ul$6x;bk{zBzUr0s
zOc0GDt&mkvxOh`kv{u6*!l?<?=&w8ZEC>=t<Q)hiB1H)Vs#rk9zlMI5&ua(S9yQ@Z
z_G)K!Gzu6N$YC-eF@G=)K*Sr#%TJt#8U=)e7{hSLqE-GLVDvlW6O!o&SpMj83yzQw
zf$m<Cl!M;|^&R>IG_qe*he86YBw#+#Cxd&#AK%THAyQ^Bs+L=s%`7Y&XEWQ(VL8SR
zN_+)m7F{4wTP?2rXsl#<?yH`v>PNm@s1VhM3$c*XegmMwwGXl+5R0aGIa5<N?MH!R
zh7Skp&LrjKae*E&8GFjc!9l=C)nze|6T^23>tjNxF<c>JEG;7=YN%*o?#1Lo#6JMT
zuKUrS3@x~|6FZ<kJk6r!jSc?%;4Na4f^Z~qga8sjKj_dsQ#M$CK#)lW$;(VkG&D4j
zzM^D`)mgOMR!fl7_6%4iSRJ|{o!qT%39SlABeT`)5P}<LK2pC>ofK?r0P;+L>zJ5D
z6u0T`Z=S3?+Pxv>S>3n{4ttHSFO&>!bAV4PssH#%B>0z#ii%{@AimZ|>h}#0M~x|T
zmWG!lTL7lUV~t5KLImiq$pHduS>u@>KRK^Mb$bADmKa#7P$FqrS@R?wKZ!)r-X$84
zvwbQXL=G9z8r6w#-77$(uRRi~!0Y+r@#;g$d)BDzMmLRwKCeCF;$a~|ZED&!nnm>B
z{(ycTIAQ~degJ+&%*!w$A_qE<=s3SZd>1utPL>dv1s9Us%a?bM2DLfUM3;`=VZa-U
zM1B9-HFTVB!Fox7RffPf_%%Yv`6Bs4RD%1~L&Vk%@wx-(&oV%ZL2<^0SJ%(YF=Lhk
zP64sLYFFCdf@M--HV98T(Hcss+cub-FulRv5(PdmxWVl=A_UzJ*paavIZ9yE=PoQy
zpeH^+%#zD|h}~+!*B4{0?%WgET>ARApp+SbwHEOe?oYVotMRSn2JD|f5z~V%3a|`7
ziC1%QTuO4XGl&;1fN?7Tn^kPJGb7WauE*|^%vEOubwKw(f`*_yG#V6c3}{+GX5<OY
z)zx3wB#=o(JnC?fBmlyMAkm+hxiD&372Myi%360~3mL^Bpf!<qjHHoDx`O=s-@8_=
zsimR`R-`pRSfgE7XZY?T7*by%i`1|04@*gWV;HS;5YiK+;HKhb|9XiW?*LX&KBQn;
zC4`Uqx&ux!Dm8@l(hO!8#8Ka1Sw;cfs2LoV!E|zwA6sD&o%a0xnHC5sUZiJ|!D_nW
zBox^9#6d2+XygdaR3Qb_#s$3)FQd=&tL6BJ^WLxtUR1|?7kCq&BjHRb*qZtfuQu1z
z(D-Bn51RtU2X@e6@T$fwwIFBP3_tWVxifssvJ5#vfefwy>BOwn8~10yb0}gssknNc
ziJDdPI#S83&ajd*AdT=^?+_eh$2Ej?Jh5tD3$mwjOUH<3<EkSC$nB{FBqS!_g-n6r
zs~Jk6mhlZZe1WUdY26Z_VhRv!&A=(Pv)IkIM}Xr=1Q)Kov89CqUu5TfA!J(7*&|uC
zm%5yeBIy&|w*nCVAg!R_h-M4YyTNoCwG`;zj~lMvKeYtT74WW*X#ACYX3+l|h2{VC
zZRKpwgJ~>tUd>F+$MI%xZzW-g*z;i!m=?fK$>h53^h$&F{&9aI)d3OSVf5vso`hIY
z=E<bT{teJ_FQ1*A!9%qLEk3dmZxrem7+$O4NEe4+cm!d0dM&<7qz*s?Jq)_-Hr&Wx
za;LC{*<eZE4RbsSo&&5PgNVtHC+JImGz&o_Bnl1(NW?>}ZjbGmwq}PTNsv%CWPe|`
zJ<ke<Q%h7Mkwi65u}M!l+{PU^T101Q9fGhhC~CiPIfGLou0fI=nk?Y{i%Ia>W1h#W
z?l3AsR{Te-6P&414PX{AmO(G9RvTrdQGSBt)HS%g&fH<OZz-o}AOjH~pM_DYIwVs7
z+YHutGkm2zYk_#SzI^%8X$Tb@Nt6Mg9j5q!HfGy#7vti^Ef_i@S|?%hfGE--@WW=+
z@o)+SI6(aS-@V4dFS42xg+Pne!Y1q_YoMiUnuwmV!*MfU4JC!W1%E$~4`M;L=;+dv
zL*XBl9fBzMLZHWiPa{g(kqideyEZ3)Xy79J1KDcJB`Sd_gT7-vRPb3fCSJ(N7}nG;
z7oa#^MgW9MRWl#2B=Uw~hcgt??!ogNQAl1R$@yh5bTec`firI|djs-_a69`0t4_Ag
z^VEALX$3~Z`F7^~+nk&qx?=gZ;V_+E_`TZWH6cNVB@85qhXSyj)*cubXxQ|QDD7t~
z+@im=gEzpxpALGfGBX^#a^;@7x_V3t6v$<~-zdOIfP9Ex&fiLJYHmiLrR=XkfY9sP
zIyJ5cb-TVidAjC=m<o{jNB$Fc0HjG#@1vryyvHrso+LC(Z%~OnfumL67!p8fGwbWR
znWgXtF_Ax*joZM;RRhZ+?I+wi89BM_T1S3L1g=Sqz>tK4gc{-2y&GL^HYfzRiq<~n
z#)?3y6IF)elx(<LpOqjOTmrN1g(Qjhj&}h18y{tgCylNy!0L|3ViOjojO|j>PXMe3
z_z`mDx8OJ$U;u2%_f)~=0ed(~R#8r_VREvr(_2v@RTN3=YxDv-Ls&}fRnEQjN!?r?
zhxy?WGZGlWp`)T;qI;W|NDWWc2v+BWJDf>_xMF35OyRg4C==;h@o<6zEppz58j_fS
zza3RM^Yu<vRr(rS6>=02e=V4b&%j-d9Ie#3FoSdc;QoDt=|`L-UWXmJ5Rtnu1;yhQ
z_z;1usZ|w0dLZncU#lPdWpe+%Yh!TkyF>9L^;62_dGSkI#R>2;+@(LzB0yJ1qK05o
zy#Zr5=;?0=U69kFfRUC2tbm+0gdA3YSXyB6gB)eK!~_6nm{r<AXU`g=CkPu!2TyA1
z077FRC%<&>J5WG60MMO$EN?I}dLqY~5YS3NtaC(JB?DQ>(4(P<QaMkdAw2_HT@Q#g
z%2{LJo{5IT7k*n3()_#`6ST$p=o&#L8uS(hQPH!w$<bVJD?!-+|4*nj)T&=_0G5Po
zuMF(LEI|ArFx|s+H*sXy{tPjtg9cQ!R+){Q#idLSPy*P!1Ju9%3(E>V{%t&jV0Z@1
zF_885htpfeM<Lvk>PPA)4um0O947GU(~OxQ_zu|Y1)MYG!S?{<cZSu;x<qgifIh9L
z{MhMJc-iF<$lAhTL%U<-P`oGK|HHyv^G`BzVeX@uwRHjFnG7ROPkU4%9X|5^xrl$s
z<$%nA4kX8gQwN}hd+?=vE6~VQXNi=ZfATlpzT3@?dxct5-qqFBgD>qHuz_^Eyn8_;
zF(BgqJ716XGlZI!<*&-FjD(le&rU}KbGM)0Be`)SxUQ})??W2@kFKt+d8o7~IH>1I
zv@UoviG{g7-Sl!p-XOX6KuV(iw|El79=U|1v~v57pq26^1^WAc>8qp=%;>vZ54K;x
z<9XJ1)c#sImGc$yK5y^u=PyQb%QH+yapm%;{=H^ApLDGbwc&lqe@3r<d_qojGcAC8
zxwBS<gL2gHLTHm%s=`thB&^`wr;6Jy8e6YyXV{pe^*{80=wGtm>3kgC3UX!{4GrFh
zmWV(hX3X%e06NBGHER0UYffsiB~q9OG6Hk*yN2l`IM4wm27IE0^Mlz?)A33k*q1QE
zZo@RHwtx4|KSE&KFN~|ijHMs1imv>s+HwT?1LCr;kP~Tkmk0Zxf?dQTpBvX)vqyR(
z5E=$22b)lmUrjf><#Io;_5@&7>=O9TV+el1+{=yNQP|4TP#R-+%!r|vBFdf7LL%^+
z&?$^(8UtYS`$4_>1ZodrU@Nwiix}u9^ZN!H8Sth1eGLtA3SRRhsQyp#AXZ<97~Q<i
zPVOW!D?dZaJtp{(t{c-5((Q$fix08u3mUq9Je&l@BB+#BNu`1Rr?s<ys%rh({YF3(
z5EM`lDN#zLQ$SDzLAtv^x+DZ74OBpB1Vlh36hTtDyFuyBO-RG0^UkH`U-!G`e%~GA
zo^crGaP(~VTCDxPbI#{^ey<Yn392JYzfI!HnjPqDRt+RwfpBa5Qz`^a&_0t)VW-jN
zDut#Z=RJ71vSQjkV=u6^+ZAmiU(Hr_-lJLq^6!Q@f)3`?4CXD$T;<1SYa%rsel#Pv
zTy~GN-|nn;l~SQoXVXCO>y2r9BeI!3@6YJc8OI{c0y}CbG}cpM+RPwOSou{gty$Q-
z=c^jr&yV7Rpi)|0W&EnFSYTLn0%`kUCr5xc(9@ou8H1TO6!ytaD?rdSq_~$yc|%zZ
z-tq8FO9oPfgJXgvF9NtwLf3N=S`Aq{FOifvFMa5Ddo!R5Bn>-06C$qpnS}jfqAD&f
zfPs$i_`<C<0{<(ALq)zv!GZ6|`n?@u403L?8$z7@r4stn>W;{bJE592mq-C6V3SSm
zu#tE_nLtPatwwyV-;VA@gPhXfoddC{Q@<YF+KPTl<?VA~?^eWML|E9<p(RE}#uV|#
z(CR@2!HRhgW2LB=W0-uHkHlZYb%C4!l6Hb!FHSA_^5xkXH3mn!YXYTl>6;jt<XGw|
z%@ftA#(|uO;ic=@_xKpnJ2YX?Yj%8=ndu9Lz#q8`)PN`(oqZp1zcD-A_gT6APw$L_
z?d7UIvgY+Ng*y2gSO`yO4BrKO!WuzPJ-fi5M=OMYi_x^Ewg<#7t`mpXNw>~LJ>)4q
zdpvWCL+Au#g`>1GSW_6DES)?NQtZa4wJ<Yf^D{hpA)zL;|J*SF+vV-P^Ddo|p)WoF
z)en)Bp)9L`(K2WV2&D)%YKJC<D8vXME{+4>C0@A5jrj102%OtsB!%AvIz)#~FWiH&
zUFaGSObWE92>c5oVO?O2NBSOn8X%roS#SY{0MRxKz#E?codgNID$s|ZS1<vpOQbh<
zE&ssO(AmurNq>rUpXVE;rz=7<QNp`gWDj7F{h{r_bG=7Qmq+X?R|}hla%uV|mZslY
z23k&d={q=ZqcUoSIEU*B9I_<}m@&ekm>{t1uF&At+-<H%zzzOfYXp3T7*S7joEh)1
z-)J)Ni-tE752JLx=eqN+V1~c7Kqz|cR($o-$*H#DNcVUtIsezMp{;U2OEqEng1Ub8
z(j}dlo7vgfYRv}cBZ=nCTBEB}6mCgCEEuW;M7M;o7{qrfk!fy@>1ngv7{C~v?C>`j
z43FAi(=UxiqfJGir{-#j=P){PT_Xbx9z1?9CK^Mnm2OZ`E=>)%9a9LB!E7YkxDH=E
zLl=QD!NO5Lea*)Uv0*@g%0M$IL>P>xaEEVs*!e~f-C{QR+1~AZ(kF3u_5zB?QmCsT
z8cs?1a$u&Gx%D&it(brmTltZiMwJk!96n*(x~S^dj-R)J#I3X~5;*%W<xTJOaYJqz
zvk^8RN_MR7<vV$9BcQi?7x$2Fa<*Mgbgi{AJW9z^_-gUSeR|$Tv)rFh2?L{HQV!!L
zd_ybeIUpt8y34~)PzH?$Qi4*^ZrZ2Lu1uIX!F51;T>2B8*Qc*T-vRR{2^f^aI4cGE
z1*mbb&YnA`WQX%EGO`Y20(UUZ{TizfML{3H2!kVOYGj%Krc9VOb$JAThd~c1C1o=#
z9^eWzjG4W4e1Jj7i)vqG+RyCbdDFYjTD`8GB*xWob`tYArU|bj2AqSb?GEvsxQu$r
z3ku6YE5Fm?vNB@D!B@XbsTE>!9T=j@7(eM2+kbo&#kzGu>~Rp|{%mH`?t`_9n0>!`
zQl}aM2;YZR95V&6N2|BJ+w_>TmiIQYd*y=c6)+)n05ne*NC7Dd3GyKCKwpeK4l4CW
zm^cW-8$=BQk_;JXfpX;7qo%P2B1R5CE};0?%w4Y08c{Nac2hQ9IHo5}BOFGs)E`0T
zriIi^n(9TUsXsxy#%{#>5+m{H(;vYKp=Q+rRS=Yjc%WP|bYd_V8QIebc3>A^$Pfc-
zViI%!YF~9I8MpmYd%|@kT_;%@>uk0gCJC!3BcI>eI)Pf>w#9M1jQf&rilw!|#>Ftu
zO5ILzw!LjtgntNomCO{Iv6YVr6|WK$_&(D{={|UQVJAs$!mUfRXH26;9jixmqL0f>
zk6RZw@g1}Htz<S_I*iwd8C>_JgW7y36Tf_@UJuOi9>Qy#qzw>aI?rMB=p;F92eU^4
z$4%F0+at}V^sujwi*!Z);H@Mf^B+*pJDj&GoHz;8hkA5V_)Qq4KqF;Z^%QqB@G@-L
zF9!~WYAt|p=>`voI|zKp2wWD1XfdKb-pNxb0q@lm`C)1XMr?P89li>NUTf_K@XPQ&
z3#GM~azz#!=VxX8*cXO68DX{pgCSCI8~w^iYw?i~$bvJJ*Ur4~X&FV+?hFW>H+f?<
zGosrW7aKG%zdEx?5crvSisNnNDOvBjV_JDT8RJJd=N>mvpmKb)USD^=o{=atUOgH5
zz#}oAi7mpyzsVSvS%i%q75y#wxc=P@?>gejfq7;6mq!<Kd*o*Z74GZtgm*=WYM!{_
zUuJm?W!W-Hx7}yoG?e!utGobfd~C0ilw5jawKu9DZBp+rc_Me`<yqcE_A^JLneZU-
z!<TDJ17t|muPwpwdltSj9dDT{FMU{si6SHm!42m2OvEdo;)&N^ga$s;@bWr&F`soW
zAuukX>9v3iKMKS$B%$|v{^$vFQ5ZmY7(iMrZg_aiyAdhWp&SyIltfZ{BX<784;ft;
z#;dX1*10$W@UwBw+gDAoIEI#BJcMxv+;rWeZRMW|>c8>HU5>`C4yd^8HcUn(kX=$R
z@N~@IN#W#EiC<0VmE+TrQ(HLtJQU`+8L$;%IV3M8k1L-4Y%AfZxTQeQbYyAd|NRd*
zDcv<eP{t5PsZSWNfc4CdM#0u{!>kpy)sH<VoOLDhO>#7|?veT2i(bM)6#^7ui(Tea
zX_n{&_A=uaj6=Fnos26y=k(isVF~vg?Q(WrzR%q@{$|}nG!IQAGy~%x*fPOh4L$N2
zbQBSdq)-`m{X!l($V-ho2uRzSlK@8sgz0F3hxCADf_s61TGE6sw#<48g^bY5+TY)Y
zRqtWR2r``pp5n?HRK!TE3wqL-vJVK}6PfV}1HTMjRJ9<@F{pXh1vluPZ8naC$>~$I
z;j!`)Ylmo?{RCGlKHFW<(}m;8>7jC2IQPa`S`+CW5Z@enoA6M{b*8w(!JM`M>Vvnw
ztE92S_KfB&{1)~%LYq0Y4me^0KBvff^8tb9t$=rm4uA5&wtGM)(b*q%biG+_!$rT)
zc6#!I>pq{x?%o^&l*eDF2qd5P-MTLb)LcXzY%8%n?F%id&;HPu8(=h%B!<`C28$Dg
zBgEB)f#``)iGUf10fJw)&#^aE&{#@CihNvi;lrsp7}h|q1e(}0=n&n5_xJa6J$D^x
zymyQsV2yNAAY&tXAZ3w|!t{!k<WaoK`59OSLW(jL({dEn(J4+FxlGcu9KvWDy?gDw
z4|?pA8r^3qMejwr=|wzz4#0~?^_b?i*@xlf#zeoa&vYviNVMxQCSr{aBIqEHMLA*O
z0_~*1XBTrbGZ=4H2f5N-ZZxo7636)$a^h?%_cMe2$ZsA*y%sS{q6XYtT+(Oq(X+F;
zph5!BP?~T~D!aK{^zRT998mv1wu^x$sIlnd9UK=FGH8ctA7h6ufNh+6R7V{e81Vog
zRAfNo%+T9_w4PEus(XA!NWa(V!22o`>3%|U$R5zJJ|s^VG+iUEV&54qGymNkp0f-I
z;0n(1spn;~!_wHlfi=NXzLM+2m({hA2Cu{;1)Q<ro2ft$i;k<x+vC_BaF|U}cUN`T
zU7seT7L3^0af6x?#{HsGH(`lP1iJwMYGJ}(1;&uC+b84hT>*r|LBEG)Yk<21PCB0u
z+>bo4WB+axfpWnRolex6AF#sam@n+Bq`K+RH;YDtmmbfqY*QV6m=Eg@AmBLz^5`VQ
zpAo{U2^iXj9}U?;(T8ZqR^Jc!{{m$cllsAXR<&phxlDMGE2(qW943;0w1PW>iy%Z0
zh)n6hdW$)Aw<R+anBqx6>DqY#pc+I-T>{ktEV+UUIp=@3@7<wLZIeQFoSv1D@del+
zKrU?mYV#kxyc9K{GpAwD%+LwoW<(8>|BOj*R{!0FDoX-P5AohfC&8@v6;v(pE(67T
zYVgNW@+5N5Do}STQG;dzJy*^CV;IvQQ%>+%0{ycZFpmOmt8^f$^PymjMGm}U^v9Pp
zlQ43hhGF4EVz3DKVaf#-M1FgQLQghL(mrLNqjHtE%G(M_bM(TR&^);N)wvyN!y>@&
zmS!@^Mx!c`)rH%f2+Bt=cXoIPgJBhH%1`lV+<c)K3JD28()8}VLoujm!WxeDVlvt=
z7%KoO30CjH&oC%RdZD_p`wqeeAQj)2mad*~h!9Bes8ByRdM^zaY<S)bcOh!gcsuQK
zrymLO0RC(0v0;64&}IHlT2uI6xRIm_zg_v$17rEp^KYLPJHC#`kOWow!lXm$*{1W@
zIcJ{-#>kAos;bu`rQ4M>iIRtm0bcpP&Z&RHW0R_y9i~HdR5h7gcK5YTXWv{%${75^
zbEixq)S&6G(Uq0<eiYYD{_7MvI|#ddpi;go6M5@&(-+Q%g(bY{>X3eZ<%2q{yy$h7
z-GLs;^PZbM8kXRBfSw+K<bu0UHToF5&MC0=z+88*z#cIWX^j+o!TJ4obbi>~A2^jW
zXuUbHVxFcxgjh+zx`0?Z5ZeU;7>A-8j6dI7e~-cczSWnjkiJ!zVTs2^7>GMQ``z>?
zZ?=J?!PbnxHGf&6aD>+PczNH?rGZ#g)9i1kD0!WDp-yeAtZZhoEg27JN{p*+2*0T6
zwyQ`q5l>WZ@R63AIqJb&zi~WYZ+r2q4;JgwD-ZXyuqa9nu(xKD5^TF9D3S5uL-^!y
zgKUXHn8J`ru7cXS?n<z*^0i}-eGu3<Dbs}&41yWTg|t3wRY-@@{*`rhLuv%7M}T;^
z`MCp!qTz6R%z$YrlzV}KX64LWGxybxZ}>l?5&DQ;FMyaiphX<C(~|(7V(028k*H2+
z)xBpEdt*`ss3|t`qRNv`l(;%R4!<U8*f#qdpFM1+B!|!^UR=+7tIgKATByqguC)eV
zEbLNC=!f8gon@sq;ELAFvr_XvKtbH>8o!TP_uTLI^cQi4SIedddaEZ!6z1*BXT+l0
zn#P~=Eq2IX>7QYCx@-j`1i^belW_9~{p!@d4*z#Z72ZMeP^X^xL=X41&&lU!++7<4
zFVGCDCp(%S<h=j>R5-)J8H-rYd3S(E3H=^0ssHyb|K!cXdB>uxfp*~nY7!D<yl<iJ
z46AEjrH~sLuKpTLpy>Lv!K0YfQ`Pf^9j$10QYPG0RqzY9(YlD}#i<0stu{t(xo6e8
zoSQ!eT*YefHvDbxRTN$@CBKk$3qz4xkZB}oChafmW6>46KeWlfAo_BX1ur`H5U;bH
z<AiG|T>)P#D}AKmw_kn#CH{$c4F1OymnO?@^8~NAuvcj3S>aZKPpB3KW6R%W!6-#1
zBvjJv&NPiiJAkSVE*~jfT`E|xZ<!b%o`=@f8*3r&`q(pWp2qs<S>!|TYuZylH@u%F
zu&EYkK(B^YRf`vwVj7qo$gv!B(rAVUzA|a>S}xq8@iZn=Kzq(^xk<4$PCMvdI2Q!0
z2L9<tn!X?O0-do;V|r53-b5EOgeMQ56&2stIqG%7A(DAUg?%<v113e9H!i@Mu1tRj
z%X@A8+p=zEKE8ZuXf>Pf=z(wnpIKxz245R1LL-5h^D0zp2*L&liNSAxMRsGpC$PF&
z45oVGii&5itEU@+tO5cLm)oi#*y;<0L7vP4p78}%?OK2f8UVxw(avEGi*)zMfEq+E
zCs2T+hVsFIybd~M@IO2e@CRoDgzyi)1hC4JUI;bytG)mGJ3uOtF%IjGc>F4%Gpwbx
z>&mszlsgkQ1hOK#uX#@?xK6Jv2jwlc3kDfc;Akz`2zoaPY0vhOm7N-+8f?!SYc*<J
zCoXiUdnsHxQ-K0h1fg{&k*|-<VbDb2*kr!_?8B)}S2rK?wXqJj<H<3jb1Y5&vmU+K
zpbVn8EI|H-#3}&nPW%9$4ho>*V18rx)CR2vd2mAw>_Glas={)7VAujhRx97;A~qfo
z4jqKukiJ}KX%Dz}A+QC3lqLseG!*QOra;6)oL{WGv~h89ZIH86K+yHS*@e*P!w&W(
zOB4o%#z1oXohrd!w9@=L&`B=$f&l6nsW!L3M|9m8_?NEB!tn+9Q57p4PlgFcN3(nP
z1*=4#w};k=?oaLB<Q=39b^7@PmZZ$SltreRNXb`i_u;iabn`5R-5n}QbvH+SqQ2wO
zr5MVEzEITuJ~e(m8s;%zft!FmJOw6uvI!z_Fw0r#$toCjBtVX(05%fdIOt@F&>_hJ
zAuTPVH~_-0A&4sgx~K#f+OJgIE^|Mg0He4I?jk&5wIF>KEf+7GA|SZVef-a!o#B`c
zsWt)rva$fZZIq@Y683-y?l?$kP<6a%m@<NTGWlsS^k9>YMF>frpH*0RSSV75req{&
zB;!|6u>)5KkfKfNjX6Cu>IRylC)c=jN@q`MEe&^08h_ZqHn+aFCUW}?w%E;(@E|OK
z8{R;T+L?_9wo0GBuM*nxkx6clr;`Pqq{3qB7z2HkM|S7L@C`9r`IEH;Xkbh<I+5F#
z53*+zZ6DOQYoWvkv7<X}nNjvf#qS~|AN4)pM+;A_m@{;qgHJ4<yNg3wmXsvj5mP=o
zHT{xZ;S-o$BZpHUG6h(Pw6YAJ@XHUtP{RykPnw9J`C-qQ&2FHXLiqQ<VKN0X1au<H
z=|cw7AKL%o%YZ0};F6F-01(zccwhC*jusbfkmMC~E*kTiU;)_PO|6REc?$@Pe&<0y
z{O!E52*r}zR<HRFZevCJ?27~u!=Kx2xX%4zrcujEAVKN>pQ$TK=X+=UcfNjn!)|(j
zdrRPLf?Rn?NIh1aK`kZ#f7f{>xS!Sg;G>+qoqf-<XGJk^Lu&;BD_#Sd<3U}h9twD1
z?{F?8TykqR`&Ljee<vZ@j6vk=@!+OafG~~7&^~Fx(qU>w@imdFwKGVGLwQhU03zP9
zW7NyI((-XtP~dLCmsjM%&2X!4p8rD@4WocSVW$fsC%{ZKBs_fT%nl3;{ReE=oHo9J
z90m<3+%!4(n4v&5{WluT`@4_Ks6WeD5T4sY1rYKJaj9rNbR*w)?3^#Lq(4)=U((yf
zpMTT;$wckVXE(#5>eww`x|`1m5}}i(vUXsta?z!Vxn1gWiZ9TwY``*II49P<gimqq
zp@yb+bld$z-!LJP_k{S*oU_B;`<;!&ye}iS)=P6#EO@+W1!Kc%_*@_^uFd%#BnSek
z39@)p`w<`!4`_@)RO0}=Em&o$L1jcbQAn~O-1~pD($g94mte|_r0tQ<tUpsc;=)5r
zAqQYDX6N0+N35wZ4T*uq9bOP=+~*R5-%_kcyfCSsfyb%CVH$&e;TI`?yf@BQo-tXj
zO+OpXzGXiFkACy5_!Tzetu5MUy{bVq-z)A~8^`ms$(;rVmcK4;SCEDds!e}j13w~&
z?%Z$^4W!EM%gUP0cc(&!^Bj7BXxIrJ!b~t67Cvw*BNie6$;`so&kUAXK(YV)qI3mT
zLW7CMI5_l%fSe7DEG(cDgMZTr9A8{u=XLBrGR4r0AT1q;k_d!87~;r|7|`Ed?cj%h
z)ZvOiBu3~3f^Qs?2i^XLA!X@6Xn?MxGDxN(6<22^rn-u{#OizNAF^WdNf!XR50rWE
zgWMuO6-4>1k6d-@@aJ$6yyV~0vt2DNK(~5w!9xGS#f$^N)a$h~bc628qa+Hy)jS?4
zfb39Yn-ZmFlnSP@56_^xj!ZiKRl`3hT)*!!ZN#s0DjVu=_@|W1wA%k;yXt1Y_$-<g
z+STT%tbu{Y9T7+>xcg}#*YdDx0@Vf41BIeK2U`|_nx)ld*wKJ}PF)DH8{}qcFJj=y
ziA9!ckb?uksS*r(JYw7jGY=Af|Jr544ot-z_nP`^BWXzQ-YyDmNh9c3A@X3|djn<y
z(Sk156g@Nd5lc1n6}#!r1pg4oyF+0h%LxoU5=#V}0D70XUjA<^PN<;c&Z?2IeC%kD
zzyD>(+mx2(me7g+2YOOj(%8i0CKM?C-w}0zit2UOFX6O*cDvA@=9o5xkywI2@X5R*
z9$q@b8&2g2J;^$z*KN1}=QWtY>N?to;Fe(O42ORp!B&vaYr%Ggq^hux6jJm}Cq{4}
zokeIHW1g5Zh}#>I=yh7#e`%mnm-+A}_?Y0wXSH-$Lk<4*GM{8*yA~LFz?%}Yy$PKX
zih|pW5X4F(UkVd5_h2SCqzE~w1WBU-^&6%wZs5V{cq~MU3I{i`R@CHO_+UU#;W99V
z_L2U*J0w8F0H)RdeEt}oT^rfYo7|pNrA&oGy+Ti3AH3L2hyCq^6uC6=Kq7K|<kY16
z(r+H;tDR5(B>9|CcLrj-M#i@lGpX||kX`@OXU8b;!77kJLL5E(<2T7?VZi?KL>ili
zo0Av+ov(NP8`O{<Go;b!&%FIihpnJw7uWq@3)5gD>Y5qwXL4{R3;+h-dpaiCWm3cU
zZnR5P7a#9}9R1sPCG`x=(p(3Ckb#aTO&t~L3taNu(=5&392K|0ZUlySGGrfwN74;k
zkaN%tvF3gE{7dHM-rY6H?^*s<EsT;-t7upRBrZ~lAnr;Ku#qwZ42IcIk$y@{1k=%Z
zuyjlkw5q>9?UntapxTDuWX+orI&6*RU@|Ig7j6AZRn*IXx51+`?;|e^N>3~F48JIu
zD3c}OQfWg_&)iz&;>qp49|a_!w}K3g%UKft4i};`IV1$SP=UBeMab0?L&L9G=jmH}
zQNp+v3zfYudu?)`e`~+)MB+9<@%{~4-QuabIFHw*Wq}Bwha!G({;g^?Y(<6OnY#$*
z@1Wokkb%fJ90@gvM2))g=jJue(T*qZ;7BIwYra3{hTq9CjJtzdJD%W%yVakOydF<D
zcTLY5Gd5u}Oq_cQN8t`7qctjMhXyB~MlBpre&)M=`IR(X7i%Kb>}afO{lxXR_}`z7
zGUn1@Y3>hHQuV!ZB<q^bajtjY@`R4Q4%LqbOeUvaU7oAve6D+7*!BocN}F~BCafSR
z)Z=n61Vv8lLKGDTv!x&4BOzWTusd<vXfSBfe%VY^tVF)nG~ZZ+(+Xwy#v<@mF?L3T
zL6(HRW@-*AR(5|P+X*`R4l#V9%qq%9SWmv$*HU(pS-V-|s}%IQC5X2(xbI&G2ObDe
zm_B8C7PJa~)MGSUcM2(R<@~jve=?YWnZ_^e)s=GotuI8Vq;HSIo}znp_-XG+eJ)v3
zD1G{VDM2}+Jkd<(Y9uT`|JjoE9*)(Wys5kmZULlVhk6}0IpkWPWJMv?bQsq@0}BKU
zLk_?_v>opXu_DKx668vXiYPM4A7E3R1c53w%KKG!<fm3IlxB_g+2IK`F~OtL;RDh}
zGy3ZavK8-as&@EG0zNI?t%GyiOwZU(cr6Q-<N8*AH{g_)Ez!Ok+o#Ge-mdlHddsDx
zNZ?t;Y`2gzlKy;VPReZ~G7(k&ih<)}b&f(%{w0ZBva83a+ecOQGSuXj4u8|G8B&}C
z&mm1Xlfuk^c>>bM2?HYk0HT%1;YFx}C3~MM5Umh;c?g?%SbTgt9M;$JE%ToYBPY4P
zVHlxKXcIP==7i)Ps&EWPvJKe~Y>DvWJITB2m)dt1J^oc8rMFJ7$Liq3_bd3J@tyKp
z#iZxF45}w-RYp^z4--+Vg<M#PJMP}^whQaD3If-r`Tt8dTHovLK{>o(4Q(bQfPiN7
z6pqnC!5oUX5@CxX=r}lU4h8$}Z^IQFB?ZDEQl$WqMh2`_r}6QPfi#C07{|xQ|1gc*
z!ES6_VM3qRNYjU{(=9>vVEKy2kYsfF5ssd~1xXuz4q-QZuB;b44Cb>HYv|9tigadQ
zpKt+=i!pH9a(OwkQIr2npwA7ZXS0U+ds8-oA&YG*svk?WEfQ71rjZDz`XJ&hcv-L^
z3aJT#HKHxW6@0G3{TtFHM593bHw6<6V!s?cfJ=S;{9p^K2U7P4CmM-QArpT#h~ncf
z#PV2@kTS}ymbd>AC3(j<LkEqwY6c2vf!3)pf8q5fUHCg7XUl}Z-j6rwqQt;U)T+H8
z#c|3rnfJ?M&SwF+)41pUD5j1)n23y93!4K!S3#Iv9}msa_~nhh{|=eL=))CEb+;{W
z%!sF0q#7n)S56uneF<XPrf_gn9mb)EUNRb;S}sk@>zsK}%iNrtpSi~S)x9qOGV+KE
zm92co7!Eb7N1!fEO>6pRbfPGA70t6Xkw6^c`7!W0<V76P!Ek8R@ot9?3eHjFhmOo6
z)k%KcO&BswB!P9+LDc5@0#*e>5Ln|KIvpx#%B8~@6X=8Y7Xe9Lc7S4eET4)~%jUuR
z6+AQRH7A|3cWix#Z;zHUkfqoqa(FSdGe%(>23ky&DONjAzOUYh%j@DiM>{D!67ZP=
zFP9f(-H)5KA4n?0Ach4(5Kbj!Z@`ZXVyuwz83{miDLI&)Icm$OHO&Ty*}XgKR#b1i
zFJ&<cl%}-rt7x=(A8Ec1t#gg|QcvTk_Tb??<BUnh1|*3A<%0=0&Z$H^qd|MkF|2y3
zOs>%Fcappt3j&qUjLJdr6azXCn5U8a7R-yl$5#iYv6fq|fHMadY9xZcMM6U$&iBcD
zzoD#wP{X*ljnBx}(zu-Se)5JxYcbmP392y2_S1NywxjFOkVnt}jhNKzzHiumcZ{sr
zjrPXL*$a56<@ct#ETTq%Ho>{S=oUH2O+>O8kkv?U_u2`8N%qTgpN6Dmb5G`%`S}uL
zByzZ1|0P;_duyiP^ztg+L9CuXYYw%UqyS?sMCjm{7B4K{rjF&Z57lT5;qefkRiTv?
zL(PtczyhArRybrnpq5-WXxlmqJ9$yn*#0IQr<M`*tGA1{@Pi$6CdelPg5@OC)NK%P
zxcO+20*dkpSaH!H9y0(?5lSo-MMcEwabHHp1jZPMFAVbE7-Us0akWPb&It4oKHQM8
z)ADkaOX@}2#Jx;ZvR|30c`=a!#4g;IK9!N{n1QH@DW_rg)VfP4g4LV>j?4n(ex+WK
z6#?qCQ8DR9JGgw;Ld_mN$1)n-e>-DyOr~JOhQ%s5PTmjfDZ!$qw8p((32wUQSxr_$
z_LfZQ7EDR^3{StViSvkn?YU`mU4O~t>l2QlyUusi8Dc<2m@hf^c@R!6A{Jgnfnk&h
zs#934mWyU->~Ri`_E!)Alh1xs7G^nca>FeA7jvkF5w5EYThVjK*|+J8LT3i4DkHdy
z3@Mf^;eQ$tj|+120+mYEhh&h8?}9SyfNMVofQrvt25MpliZO=&piCxq<GzNV8XSTb
z3sHY6BCoCZjGPsPXO~i~!8VR$NVj+7_(y8tOLZn2K?SG|LdHek#1=XI)$$AFFdXFr
ze<{2CuA}~*(0FDUiVmnP!-tC`7Zx8q2ejRV<2k|9zoOkB)4*N2a$H2D;H@3kZT{%_
zIqr>c(Ares7^3UTp5%dAxn1=1<&9QO4sZ!TjzmgNT`zttl){~R4{w`io8)vadkvm(
z#fk?c7w$zh{CNZt6~Ur{IO1SUL9Dlgv?8Tjl-gL(W@wg7%qk1%Lk}l4!VRB`Z%29B
zFJDep!s&;OjxaOBMyhnc>wnpOq4wlYBVk1eQDx{zQdi7Sf5fU<rhn$r|Idf>A(a-~
zrXPgxwgkT<3PQALAlu~S_9loW2wVfiY#xZ}rM%N%`Ku1)8oJ9NRK1G+tR^(~RoN?u
zqjf9fS{xh%rj3MiR~BHsL?H_%c=uGo+k~$CQB~-)4ifz@?%jV@zHJQ#a}L79x7w<_
ze_paGXGb#Ox8z=|C<a|sEuKYE2Gx?_qk5Zs@RXti-oBHzu(0@TrfnyE$jO<mw`mLC
zeU$<u>b(jcVvpN~Ui{d;=J%WvduogRPy!TeV!8fc0{&Yr02qs)qqve8GA!mA*F9~Y
zcXsl7?|nTu`wBTLibv|AyPt6Gjz-XflKG_aFSjoRpDFy%{A2_&=diA+2p|>|LFaJ*
z#Ua_iyzb_8@O&dL$x`s64VX6aX}CT*OUu-;3)eEP7yQsm{$Tf)dlWKpKOvX|ZlHE(
zFX0r)OFw_Qcrd})T56_w(Q|#mVZA?};0PT!cn|_Pso$CT*~(UunwwooKiy*{Khlea
ziH#;%`)+K3-nZ^)>l;z$>7$1ySGk?;e!i#MEBrK?PDl&au6!uGId*J?VQ89>tFpdL
zkHmuflDO&zYSCQ(8ZF!#!;d3;4kKO<sjMAR#Hn*n74v<x$D9(2Q-{ml+3QOzEm=mH
z3#f_Dhg7tQg_LdGeOqcJi;)M*Hq3D2z(9n>0iEyfLnj|Eu+q4rL*LmM`%%{Vk$7w^
zR+)0aVT<)(w7#g<!}wzK{)9MAzf=or(uc;ovLSI#zJ2=Dd-eM;MvxbOqL^q6h*XVi
z%+)yG-<&b_+-Ah-T`wR8JtkKp->9|jjV0M6`u1ZH`LviZJ&KmVyX>_#Pw!nO@^&@4
zk$s1V$b4pMi{h%^zLQ2~%J7I#YITlWW$xWU2fdeWa5Z7v82{5b-1*j|G{xx>BhHHA
z?UvxPhq@Pfa+LkM0UI;G+k>F|wSA}yj)ZW!AI`yq8VY`Z*<RUNZ7wKbarQu+a*4R5
zYt32ZOF)m+?<o*?@%kgoeDeqyS<&h5`#(QxiI4E;qp89uW|CAA&!Wt{OJavaM{D&G
z;M{|Sx!ZerudZINcop?_Y5d)QsKYFisAE^jf<QzJ@$KWcq37TF+*cN$R){#>skVK6
z=y1iypy-k>+#$=07Zr8%NYBy!Z+a7)gO`EXV*Zl$sm#(H_5C$II=JkeV+X^1O;)q5
zl$rIB_-?Ka@ZyjdUZ>$3^}I^pn|^{HuQO28Q+uxFZW0*;{k`lIg~_cbD%$)XS&~&}
zbnBK6J4RHKIwcRV@p~vcVU=1=!_SA%_dtw4ZTkXyoM!n#;kH{J{=HENamr7EXXYV$
z2paVNe^k~VjMQ)c;`!YbcFBG>4s72!I-$$**Cl9(C36UXE)(G{|7xnh1$8rHY-S->
zPa-)eP%7KYvS?8-9=q~Z6wlP=f)K|-G12H$C|17mEDw>3o8)$Rzh_eQdg6(|j*c75
zlcqba^LxE>xAzBl48+^9|52*5>~7Ee7LJfoeV*qFQk}AwuAV@06mXIx_X@aFQ!mRf
z6CJ@UV+WKY<g{m`7ePkU$jk%<)h?{$@4*_o4mEv_RnIl#@Z}Z$eenFi87GxstU?xj
z2#TLlw;oVQ{P*Lgt}ov0NHKDLMQmoE&BpGX`%I%^<V{-R-2z6VQI_6jE+auF#!yT|
zgRf+HfAuGW;LuP5Dc!8NgfQ-#(MrPF`p$z=x@li->(lfSeeS7V*K3H$H8b$y(E*AQ
zghr8&q-hh0{X38jz~2DapxpK$74y5|T3T9&BL`vc!{UuX$R2RUST#(Rq|tfW-w^NQ
zXdg5si01-I+*e@9Mf?SDbPr<mg@h0uV$v8r0E40l;#{{Iq4Wjo+1sIvf1^pYI${f@
z4b(nhrRR*1wGj!Ev-=o3EXKdziLs!j7SCNpjccZ2V5uwZz?Ls!o<w6*f&&}vl^u13
zD@;qa^LmUY4NQtpWF6P8<NE5$)wnrV;sQG92-9L@mfr7DpQfR(>~Vf_wFZacc0bDy
z1yvi{mu&A>uU6juoo3^b+Wxc{R0Ux1BUsbxk;gq62GcNzFot3b!rBucs`O`HLBTeN
zuu)*b@;k2uO9lc*1F=)_;N%$CSrGgJ(pE#XX7tiQ9Dqnn3N-BA)md?xaFPdS)Zq!C
zR=HzKyLd84$z_2~ccSW%Xg*b&S<@C7(F1ho!nNwL8Ofn&I&YEp5+NgEVipfC3Mo5Y
zQ>$p)MM!*jT<eY3Hw^iz^lrtHxEbp_dmyBOqWCEiB)VZ0(1eruZLp^F%}eRXE<wQ%
zNjCQKPn&-Bj{q>pC%ULG^);Wf|L}&h0^!ttnWgUIW(gVVt>iv$Jc%S_N4-2BN&orY
zhY<8}SnTIyRETedChr$qGE!MNG3Mm=XerJ*3to@g2ImC23Xc|<UL&w67&Ieecc^&F
z+!4SOczaslyy9qZ;UXjwWDqGCOoQNHfVpy9;Od$!7w1aV;8QQMl`UfvE=E~`-6Pdh
zEPt&<;<|pZoNV}4UBWxJ#RepDE+>6BYkA@fg$kD1yx!yNSSu3#QrlO7a#G=n&G`Hl
zsXTs!nho#BzoeZbagchpRLhzD#&v>`{&{I(Ejv|tg3!J&)9lI>hS^!B+1&(#c~A@H
z=jMJu;@Nq5_@?bp{!)?1UHZt9)BA6vgGY%TZx(k%v}D`t(L0%NxQgm0H?UL;DN6A0
zhleytNA1(jk&CAH>hzr7d&7UVz$mK4sBgzq`}kCFtGn#@1NCFG$ax>d<dUE0jg`o&
zhtu^R9@CpVUdWra@?h^PHudECMw{v7Qg5}Mn6Nb$D=y8}!1YaQBJ^qzRZ)q&V9nUv
z*j>&t>ecz0?TXAQP31An^pU*Y{3Uf!gM8mG48@~w+RNRm{b<+u%TbwXEph7<y9F1m
zt*upbbab$;H7ooS)02~X16dn3gYt9;hIa_%=jN)wfv>98)(aaRaD-MoZ$?Ijyo-wq
zgpRWS0rCO-(pHd>4@(*x)b8o+wSwo|%EqP}B<gsf?)#lZ)~=hQgVJSMYL_!xKFac?
zJ$XsGH+y_<vrf^o{jeDC)4NvT-g^FPdGkXvHOFPNeaYoUb+qSaV;)ro&a^Z}dv6pg
zU}dtzCXF^mA3egv3qGa~bIhGp(cgW<cyM3A&S!8cePf@%V7_l9qgSD?{OgB4Rnx`L
zqf!2JQ5Rw|GzQfe++jQSETM0Gu}|Wf_7`qB6_|&1LJu|VESLu=w==}V3t-C0Jz~&N
zQAva><HBTJAikfNpwrCj;^Ja5{OR)0mo-KZSl5O8x)KM4s=a;Y-u`}1cXyJ%|7mIg
zhucDr78RMvIx8!E;eYrJLu!Uys-_JSy?KuhGG4?pRG6374Q^o$*<VazPKu(qy?Xi6
zJNC@RJ_}CQ8_I%kJWFu&zNnr-`Jz|nZl1`mgFW|V9L3ur1#4n>-;A$WvA=;CedC0~
z95la_R8*>TXMeylYWBw3(()&Wz~4d79kAi!<>ULBm#0QeO&xYMxFIn<o(6_((pRor
z!N)l<SmK(gt*sp)|CyJI>r+Y!<4A>%0fYcl@7}F>?MIG6q0~)UL<ZVO1L};|cRlw;
z3^JQHo7a`2uH4`IGNE^fo(fWq!a}*--^k}*Sk^csAa1Sc-sGix)4Z_R-v3sX@jTJX
z^Z0(I?`Fbi`tUD!c)NYVL7{wc&(6b3-|$t$L!k_@o(sY^1a#e4D3s=HC7pNh9r_B1
zPm%?fT7=22>|M3<fga8lH|gOw_~wxLi+1LR>ruur%YV_C;B!9GSV5m(%+M|Rw6*of
z#nsi?!a{1cBcTsw+5{SC;ZrV7I~aRctAU(@gk$wnY*8v@OKr()*r?U8QxW4v8yzHa
zDLjIqVO@6NznB+!Z^pD+IMXLva7Ai-xkO9b2ZI>UfZ2F?c*wG<{m#wLtR^fRzw4no
zheDAh{K{&Xx^W=hV#*k|EwZ_}`86|>{WkBe$;ZFDM8l7@Dmeq`S6MCuMi!EwQ0XS~
z#eyzzb)8qbZ!ypW%Y-3%@}x8hl@;$M%$Cp*)hE*AoN~KI_?WZ4zTO#v-M{)K95nkf
z4KyA+NQMqRS9O{d|Kl=Cm7${R^01PGgddW!ZoxuTlE9xcxmNhJYW&moTFeXe!ZUB=
zlgd>}LOs7ZTw0I0r@yaT_E-%<5Aoh*V4J<K?BtXyA1_=KG0oce_5lqU3KgNq6{9LZ
zrOr$iX8Ddge=Ry|x1%FGIu#p*dXV<DTJ7zfG^*^q^C(p1iBm4{BfD_>3;a?RN{kwX
zVkSr*@yxG_C;$BoOHE_mx%*d17&4NC<>I1JF+Akz6(>->VL3xf>-&cR;`GRac{fvY
z+H8)*|2ci=s{`+f#Yg+4Q@>Q;0g+~lZJUSj)?*N_Qe{|#yekZPdU^;>>ivH4OEI$d
zQdJ+mJ5zCrutwA2g&qD!BIG@3GQ;EapU)QR&#(NKck7FLK>Nak=4mN~ZxLJw>b``$
Kc!rqa)BgpQbd5>?

literal 0
HcmV?d00001

diff --git a/docs/_static/docstring_previews/flowchart.png b/docs/_static/docstring_previews/flowchart.png
new file mode 100644
index 0000000000000000000000000000000000000000..358f210c35369a3deda5e722c5c904b7ffadc93c
GIT binary patch
literal 23797
zcmeFZbyU{f_xE`L64DaVARr=LBHbWZpfn;aA>G|6A)!)=q=1BUcY}a*Nry;xO3uE%
zznOXFne{wt)~q#u&CI<P-1mI}*Y)|Fv(G;J?AQDK34Ng`kB3c(jX)sq9xKQyBM>Ow
z@Rf&!0k3qjT<F8UgdF8G9aU^h99<0TjS<fc9BnOa94+4(-f=dzcX(@K&BrOo$;*Dn
z%+b--L70on>VN+QP8)kut~=*fZSW>HZ56Z}5C{ST<QK|!iR`xs#B0OHvQnz9Nn0~6
zsw0|I7`vwT`3}WN&{<e%Zpf0LJGSeN&$jCp)z+-ngw2%K9!J)n%;HQGv9YEy_}}**
zB3ZL-l+fH48p1%A?>!d@aY*>`<%<B1I5BUcY!kWBoj}4CD*F;nUqb1wD7j`OEP7Fy
z{~O=0N%{Er4%dd|$BY#c9$Hb}Kt&PvJiW7_-0<|t6Ko7za)gYG4Au?Q2wJgMV@xr;
zL)l7n35kh6Yj7EuW^Y{o{q=>O5DWV<QP3_uzvn!sH<1rBgn*7S-cG<~HX=<X#Bi#@
zYIQVEBgX4jZ*P?kosf`_+1t0x&CMtQ7<kezUJSU5zogwbSr%tvW)680%bNE1<Gm`+
zfA)!7;n6#3cQ#PH%P9Cvnx>~;r~jIr)qOO2k@UAWF(8sgWbtIqhill5OSkTwu+w@|
zf4@@hm7=2J=~9#g0+Ax_C9R-<b-Xnda<nmS)SJMgS?9VZF#1w=0~7I{O5oAXY<)Ai
zjI8Vp^&hW-g7K&p+d_%WX6xJ%xpqv8y)WJW{QcYgpL<02Ir#f)VSSx-gV$R)g6CUy
zwv3im^>o}VGBQkUZS9<Mje^${Vjjnz%)SWN(AK!@L^3^%OLae5Cl_`!(~AG6!_L9M
zWNtlPKtoMKQ|!1l#1J4MCPpKjMrdzOdO~0;Cnu*BFZm>Gar>>rKu<i^V2%p&>guYb
zoE*B-`bcvqF>}>g1kdW)n!)Bo@oQ~sYtFK=vO8Q{QQyAZE%1UXe(~akBuv%#ZA!|p
zzON!O?6=_pDb&@~6GYwY{e(Y!_%Hx-27@l8qC)I-ek^Bho-0e|<Ky$da)R*e?2Ir}
zwy3<keSI_!C&JjoBp3!{0jAN`$tkF$gm<h!2ggs+w?9R~s3(qN`fd0J@q>5a;WF1(
zmj?R!2qq?`k?$|jv@0wZvh*4}DNy|ow{G1Mv|GF(u$`w-7)Ddc-QC?iQDH^p=;+Ap
zact@7=I}cRhb-NCre<Jt6!{6}WBH+%=O<5l^`_6{9zALqA18y0f1$1(Tu{IXGkf^l
z>92#Wt#5KNEdv7s;=)IJbGDw!=lW8BRiglxl$6x-;`A1x$YxG_BwvddK5bxlSf)Q&
zY!6}+b?HfIf1TR_#^u>jfS(_VEW@qM&CQ~>!&#_C9C{7Z&d$y+UcU6JsCf8VN2l0f
zMbY{!TPcOx{m=|zT<R{7l%5_PL`z4#$4QmAUdocs(K;!tklUQE-#KlJrJ0ZB!svZh
z%e!qo*PxRc45QmTHg@~TlP5)$m7Ng%s}2G7ZMBS7fByVIrTehHwnq9};^D*F@V77v
ztdAn8Z|dmiSpF_Sx0)<#zPdagEYJ~w#Ws?sL4bmSGVK{fD<*Ak&kKKUyEVyQX8D^4
zK}b%HWol~rJtxQRQdCSV?fdt4QUMsvy}h^#3k!nIo8%__$s8}AuJnC1u(pn@c34@Q
zt@oHHGC(^#I^waJ6)ot8nKm#r4S8raDJdm|(iuq;P*})i@8G~=Il&#e*V585@IzYw
zmsXS<4HG{wHkP!)YAP574WsGTFFaVYs0h^`ugbf&KE}lbW8hKFw}+FFi+c+8{`yr6
zQvjI+^YP=y<oxC&mX?-9R#OinDEKfsIy$D;sERdT|EX$!siGoN>%3KDxAaKV>x=|u
z0O1RhNd_Yz1-BcWm>4=!>yn|8ttjYyAy|1CuH86QX|r^8w6Qu_&KB|h{oeX0TfS~R
zB_8F2rr~U*wDff4^;g8i#4!m8?_fe6m-QzJ5ySFY>Ub~n=n+PsH2p(JMGao(=@8io
zA}*YUMn-!(jn|8l<rY;w*W$385DJz3Z5<uyieH5Zv2Gx5r_GQD{wD~gWNE20Y8rE6
zW4y4lD~55f*hoHtk{^3*ZO!#?^_IWC|3JNm3!*8SUZ%+B+8dWnoVuc-;$2i!C*-_a
zL`0}$WbuaMJ}?`nhik+b1hlxYnEoBCWI(96{@Z6d(P)7cK9Kp8Oi)lze|x%G;n_1;
zSX(cv?U$8pY#5k2h_9}$a3#|u^mzICjm5>qr#A*O<oOWn+}zO?=kT6<0s^KcCMN8V
zK6yGO{Lyg|>gq(#Pxly&B8rWB;&@b4RQ~?`+i~w%0&Et@-+KG@Z848ghpmyDn;Val
zlhbVq3YM4hkXKG>Q&Q;AnqZ`v2eWl*ou6rFXry!S^OrZ(2v|*i($dr<(Nscg!W5&Q
z<>lr+u?$#b*Qxo0q7U=4GRe!wmsndX0w1?jA%DdS8G&WAis?wF<jqaArbud`mU6H9
z?Zm{y-r3n?e#?m$>Gj`_*es`Js_X(#^tCIkIf_jX`99ZIn~)3GIXNTI(b0KqY-})m
zIbM~$gFDN78c&WeF)*MnuIuf6mQ`5Tzqz%g_39Pnxc9}W9ft4oufnkwA5Bu$<{BmV
zczF%!#66=|hs!NAvT|~Ma@1N)RrL1t6-*orjf}i$4aQ@uh>D6L;N$0C)|Ho)<wn5G
z#xvc$Yr7>MNo|0bC^qV>dWeaMX<}*FEADmHR<6_FnTQfi{_zvd?ALXzYI`O~waj}e
zNgtV&T3}(&($c>E^7%6*tcXC3LfwV6;cNs#Nl6J-?mP@AVn1?IOiXNHZ7sB^NfOZn
zF%|Ra(+&6qMAP}n&MiVhU+>GaG}!-$dPuQupqo?hzIoSX0Y7i+>~zS!iBL$&=)c~`
zf?q;L&4-U4nf?k02*?j-JrCEp{*WZ<Mid+zygK%SYCH=PLJXrk0}~TE>`ea(_QF40
zlfgPksjA=_7vtf%_4DV?-NVCAf=n-yxK{eRyQMunMRV2iFgQ3kaylof>=+^TAy>St
zb+&q4>x^gA8OdS0mJ6fRwh!?cP*o*(dUhuL_;IL(&F>PX{2xE6Rl<ih`676z1#J;O
zv@0=;U=?`Idw+gt753-P8*49EV1zU@I;(5(p08fL!otRGfq==J#L!zG%ET!vFRvc|
zcYXys9}2=|u0hV$mfQVg``6_cR7ptM3-j|Y)|YRs?Q)*ZOivGdf5{k{vM39?M0a<t
zWbOO+cmgzt*RNlvrl+?Ad3x`9xVbIB#3(2!tz<1XhW^>P7}8ZcJn}ibxPB2D8v1&w
z++rO0Cf{>&rJp{<F)SwcxjZ66kWgOCvGMW}!J6gTZT>Spm7Txr^C-Qbps$=p!kZfQ
zVZRR_h+qf1b342kwszDFolj+D51xG$CcSs>9&7=y<pgN5KT=lyFk)0@HFX=-dn>H+
zTO=fb1a#v1gXwbU75(E=;yxmz6@^;b+782+zfETweUO`YzIG+PVX@|7Hrzd8UpyYg
z_a^Z@$>-1UtgWrT>(&cHaT07XUhp_Vz~{<S)cb-QKGN2~p|$MoaFy^A<stDejk<a%
z*lfA)KjF=OUE<;j3k&<MQ~MAV6?J#99jDIY1V1@BnVXwiQeFKvBrN0xVqvjiBVNhD
z#ieH^a=wGsx4UcW>fzBkIY}8##uXeIdK0pyVzOu`lrkh$(*>f1^DH(|bP^c5*S1)=
zxUyd!TFoyn2dU+$H+OYm!RPIsoW!4|mIyLFituf0lz0?Hi|69vVrFhGeTOg=E?!wL
z*UZxL^<wjl7Y$S5jbq7T9xa`n*#qK{k&%e~+0@pS7F$@wN#b6VnE2F95RTkV>#EOE
z#AzgFO`pQ$3p%b6_9csjqvPDxhnh#wbyq(${(D?mKhf{hc`m=492VHOu**V%gYAb8
z+}zy_k-y|~C3<*xsBdM3=O>8}5fQn8fx*DdO$gi9;+eLlCgRlRomU~mH$5XGntT*(
ztl&_M<C=0#2g~LIm6y++Ki^v&WQ02K3H9jh3a@V1)v@tugm1uBv&|I_d!yvz$2U#;
zlKd3zE}os8pOYmtem|OO^pWuN^ei$PWGLVe6VqAD&ENmf_!YUS78%HdzBMo~n09=7
zOi`?f3#k{fgTm3hZ@IZ)oH{kBPvbeedV6KhkGFWBP<r<HL8!>qn>TN6=;`Ufieup9
z#OLJXOn_3;#YceNr(U<t^`-tR1FLVAyW6TtZ_CXFQ7f%yTA*C1sz8=&Gc``!e0(3P
zIA?~3i76;3G;17LL`6lbeD@9y+h9-MJ>JsZre=;lp&RMx!GS52R#pAfX5YB{?*8@a
zdXL-3j~{XH8W&-yJfU%hB-~W%vQuQ*Pg}qtB0>erH3ZV#`goyUu5tWb4i3C}_akX(
zY1Dg4$-P#{tp&;$WZB}SSMRR2D&QFYO2W<UAL^9Xzx`z*qc}3yTUIwVjDCIQ@9OHB
zpHza)HaePcJNGUO4*a#_>HrEv!u><MMc5CfclP%7h=w63Ev>AeY==O$x3|Ym`a3i}
zE}!`%_U`(HdMvA&e|^1Jh4l=pUFbn9n+72@^~KE6$CklcJ~!_vCcT3lEzRxVFS(HY
zU0BIQ#yz)XWo74^{LqNxtGBnek!7@|Cb>(_9fYf!o89=355)zd$nrO@tE+2MdpkOm
zKK|9!Lb~<tLB!08^UVR6nYra~SRup?GMlO4hpj#0(<hRAy+&H7so#Z$#<(Ya=C@e9
z_;&y&4<1&v+#3!K4kHCRwF3?CRxMC`3_@8xUGG5#yY~cy4g4q=Hn}HB0bD=e@T3oM
z0NchsoMghFSXh9f2zj*OhR83lfl|EKb<gmD#TYKUF8}&9wR*wpP^f+upg2(ep_ixr
z16g~)0&IbLRKR+=39?aIUf%hyjXNB0;^_S%K!7|7K!zQJVsG1C5l(?nsReJa{SrUu
zm$;rE%6y7UA{hA8Sa8a@d-v`<l#$34MMOk|__M_a+s${4LRws0TtXrue^NH}R46uH
zXlTeQD-%JHhOCp22Lp3+bL1aEdCajFKX{VBXNo^rVHFNJG!4=n6gzEUq-^>&He7oq
zlHJk01<Ys7`&=t^UgyrmmkkX`A9*@jT3fZCMoo;1yRD+EYzRql^K@@1s1NRu1H%`p
z1Usw2!9iRE9}f?UV{&H{U8(cdq}`IXmR2!~N|x>L0oRJ>^_5rJ2Fa~k&E+&=?w^E&
zg?ZuJ0>j<`M}$tOgH==Y?BbuoE%9=3ab@M__aZ*IzBRS9WZ#&o<Qg6xHj++@77`TX
z_VKw!y?o*!#C%mdt$v4)8xIf9#MHDC2I+pWRbv^HwCeTlPWOmH6Uxi^4^B>2D!xA0
zh)+o9`hi?57L#QxICyv*jw$zXeTld{dK;4*<{G?$6z57!2^^fAO&DU{t*x#W(Tcej
z!3hvPY0{Lu=iG(jp(Q!oodM+8-hqK=HxG|eI6&BW!O^UzuC6Wtj!DF!viUIkmKUdc
z>_S3`JRO=)`BF)Ei{`z|z);c9FwM=)&42Ws(sDFcji`rp-M$YJDM`)Oi;Ii9lg>M{
zzhE0rRJdz+e0;pQI+&3!6N2A6IaxEY^|;}^FVEShAj16vJ9~S7O?J+`4;ISqj*%&Z
z3@Tb$T3vm8ab~ri|9EM|J>Nm`)0OpIkD6wWSo#f|GBnIazTa{`-VBAhhW^0=5{t2X
zzpSi#v=ZJIGU7%HRA;vF@Gg^+li#&U?>)4f@IBd?lTlM6<<PEZleoT+Nz+qYeV0Ng
z?Hb$%?}*j^V)y_5{(n6J_q%Ek`wsdM^!M&zz{=?A?*8B;RG0ewJC>1=5jh2g)E&aE
z8tN8Cl8YNRZ{D1!Fv6Uz3zxya@#2MQ>K#_o2%a~l#xk_Hv3N&=dGZ7hh+Q@8EySoD
z<R6`G$vQ)szqhkbdYTX$?mqk8z|tH;NJtp%6uD0_2=9{?=m=us$q7YJ@=!~sy%;mT
zMVorPAbPU-o3AchN%G*B=4Jb|zw|XGRIoCn4mb~N=5FmQ67Ry#wc=%6{jcH=reSMI
zOUuCIpwZEaZ_$1wTr)5**a|Ni>tHY3#P&{ps-zV1A?!(Fgfa72Q_#oG@U4UKU3&7t
z+AkwU4WVIyUp3L_o1+>k>xg12DoVb6OZPuKbWCLbvP<Aq_0(5dyzA29l+g1mA!1qE
z<*``ra%X>9PA2*tuA(m^j!jqL8U{=AT}O!zOliB9j55387c?v1&|AN*4h%dc7!Y{i
zZ}eQ_aE}JXgE^j8PxozXmDpOV<38HU-H*JC54rCbN7V$IByyeX1$eB^HQ?oHO*Ppp
zC?{xCwyk_3c`fXPij$?-5totO?U^{#{qtuU+qW*I5XB?HoWr+Ms1Enucebpl%<I-;
z*IDKnEMEu_c{PZR4u8u+L!=tX5KM&f4`f^6AkdYvZ|Dk!By)U8Ra7G$DSr@TbE5bz
zEc)uTfD!ITdEB5JIW<cWZ;Yca50HaAAt>15a&eb^?)aK<(^M%X)uvY6mS<32=c&r+
zdz#RG78dl5L%nL;qpjSGf=m{ZD<Sno54@j`LlYiY5>Dr-N&RBQAfL_@V`$a09VXbC
z^b=NlZo7B*xU4_}r!V$s_e4RD`gbWYnbKEZ7<*?!yy=?T_m6~$l*jI$hK@vT{ziu;
z&8E25EuYh0P3u7iwwL?u(&yXa9ybw1drdE2dtXpL6|mtd(|cvk>KTmRnd}f>yw}8o
z_2&<Ft`Y&V08>*j+hwf7!t;yT*_AF-7#bKsMSSzVijev`B_*SR<I$(vvG`uwj&GjV
zM2VVt@Oo@ivd)_n<+_!b+q9h$d2t~wJM&pev^*qHbkm(-R<*@Tf84k`xW8%3hYEz0
zm8n|3uDHP$m){hVSGhYr{m*QyLQItMVG9I@-M>8t!O>if(X(61w5zHFF`qMk(0Fz#
zp&}C0e*}M}8ecGJA|0_BD3a8p(Gf%VW>qK_XZwCGvu4|KA8X8;H?WG_Eqb9^S<F$K
z1<`~+gqz2;AB))8((&>R@whC%`u;sPUD)@h1DU>NKPS0}6UO1;5ekCx)vJVye(A2B
zp4RjlraDS~xfe~wj_aeCF$-L_Yln|5CG7C~tlgT$e12iLiu_(&{JCT{SwVxiBf^HG
z@|P>PydJI6Gv)&ktk-YCG>;VAG8nxX_@8QvgpM1c$=vmHz1A^$6x`TUCFM0ZGJ4DU
zZ|vYiJ1v5c`0`cmzs$?j)8$fb&EG<ja%!;_4h1@x)z#ImM~gz^>c6@v_xJawEsFh6
z79~6ph(HM+4A_MQ#~hclEY?<A^39zBB?y)#NrI#`fB$}}Mn+#EDSshrjvW?8REUo1
zH<JA(j5ptujGW?&{@J22Ar+<H^566yrdBySWN*~VDG~F_;ig4)`FlF$2=wt=s3@8&
z<1ZgSP81(tlHeC!WL}ouYt%+Y1<7w_7JAxST**dV6A~{w^H6eMy`mRl$x-bocD9vO
zW3#Z-Do-B}NQ?UEwxp%`ihlDkx3YxK->eyenw!vtPEgC^94*OJuatFmen>l0&sdCK
z*piZwo9R@1z=|Ez^^?V2n{b~I7Pfs_26OO-65aw`!6*VdyVFCSYDqdg4lA>@RsFe#
zKC%tCoxhsc5+t0IbK)2YvA8XMv({nz`04)m%rHMq0fFdO#2i)ZyoVaG8ta1oG7tYN
zJrT=7;eyPr=pPpU0u22=y`{k#6^-o+@5j&b)MZ-#ax4fq-`Jk5^Bd@FPT+RDbBOnT
zpvbMqaOH=kGJ1cqxKvyCib6zGwn!>bV){G`3H0$kn)G+h{<u2*)%*cxo(m571`EPI
zqHd>%Mvu|vI+l;zeKzVj@yn}=Vf8Cq8DZ7v7aftZ8}8-Ybs8FVSUAmXTGt4~CpFa~
z>mB;L%uMM|KVpy%7m}HKJZ$>ob34vlt2FM1zYZQV?#3X6*sBMqMz|KvTOo_%+3y;L
ziZ$pGH1dM`-mkZ|PEw!P%&;H16Dg~&ZYPOWy6ou6wlQZoWfzoi28~B<P2oA{Ja^ol
z)*8zsq)lb?md0A3!2X|pS2id1!Ze)uFr<U=)5YVVO`r`yjbX?I_5IVwIf3H%kDC6K
zuJYI_RC6p{*1C{ORMEW}aSpC;xc65v`Gr{{pFDnAtjBXfs_?y94VoKCB8)tj(|Z}@
z6FD5~uRh;CBPqPd_8>iW_}fJv+7RNrC0Tp6ZTs+c>u?7u-!(NG&sQFd>)4)w^a{Qt
z){O-LOtyTTI^Qp|1s+_1qVDf%1qyunv*^D)8>n|z$l07Oy8V2rEFkyut7;BLK0Y+z
z8h3J0XVHoREj6>0^GbhiZ(E6oD-X(bfw#fLn8${JN-mp?gk3YEd>lu7NfxykZ@(Px
zQi-HDkq5`VM`oSS3a_b)g+@aP(K;#zt1lXbO>t^Hd!1#8wvKjkkSk~FB0nV9Z0q?f
zo-K!y3wS+bx3Y{$`t<Hko2+%si~F)Y_+N{>E-6PG3(tQ|1xJ@iHY<IL`}i@Sv@=qF
zcKwKG>%GqAV2%cztXy4NVn#?>T4qqQQD=t3PQzSTQn%VOp0|CTh^Dl!L35ox=QX3d
zW)oPK!BMD)h@=PT0tK&I`rdEMSK6{5QlGEX+2AyWhOs;1BF}i0j~i?k1wWTxUQ=n?
zD>KCPe+`YDI!TSxKr5JwcbUWWxg9L_MPSiZpRo73b|<PhN6T-S-)x8)O}Jdn{T1b>
zPq9xYumeBQ36IKrJKp?h1ba1Zt#q<Mt=p0PgpiQYb-SeFmI#8cGfJwBtWv~!X#qn+
zC4<2S!{x1B?kn2qYNlIlh!;BoHlkGy8(+ro5Xqu-Q3IV(0}b*v1zR1<PrH8olC$j9
zE-Tmw3yvHkpPIwuu}QBMkMB?NFV3MTND_O0kDq1$(xHkxfxNPzf7_|EVqe$KzK#qJ
z7s+?*gSRANCZ|@%qvPX43N&Bqf7Zyjw>oBX2ch58#WPXi%w(;DHy`OczB#G4`Tpk<
zj~f^CoVRQAW?nzn$d5X$noxd!kclQz(b6&$=Fx?XsEhOr0=$*4SA#`t+8<C8j1l+i
zYmgt!)Z%<B6`U^L!OtrP3+?`{()a2V-xjJCN;o;W8?iP{brllnV8oL5UGljMnoXnd
zbJ?z_N60x&PR2-6P+hJ6%VBlxvo3RoP*P(Kv8N~cE%U?USUTmU$|p)sTYsXR^ZbdD
zV4Bpj<}4{K;fBc`IsZk4DAM1$H&U4wtgV4IKUj1NxIj!|j9)d|IAr!&s+rzQwt<`&
zG6r4ImDvm5I*+65?99t5QpRtDB#`7Mg=-X^K9@5wxfggnV<VB`@BgRZuJ0eTXOpf~
z3tJoMDY<H;ks@Q;Lie3G+|MxHMTLJf3-_VJ)VZ|rC|me#6cQ5`*X_j_tz@Z>h4^lM
zbbY?HvAQOutr0IW1Lc*kzdtJ3ndI!{wO<``s29Zr6e$B&JH?*@yRgwEVm_DGGa~MY
zvXeZ-kIejBnXhytSY5MVGEaI)%Nt=)Yqw04^m^tAzhXmxNX0^B7Ds9ZGevh_`GMkj
zo%<eja4aht;sX(JG&x<=la70CUeoJ$pwsd478lg4423?YI0#85Vq)w3@9*Nf6qLA*
zDj2`zzuYCxSzBFe9vi%Q;&1&)xbBHZF7x&0^O63P81icrC`425lcZ1j$j^#FU7S+!
ztJrPS-j&}vOeN%aX)oYBp~8$p;E}Q+s$qJ@-jxn6!nS{GbXXv|>NE4cvEcS}Rq$x-
zG1>Q4*i|1`&JsB=D>aqL^`vGML^73THP_WJO^enc5ZgyS0Y5T`vmz*iLrbruID6Xj
zZ~CNK2h82Q!mjc-z<37eRRB}^O;Eu)edbV_sZEk#QRQ^}aB&)@@WaEVk0B>^i%sH=
zmaBegGs5=6UuVpXu0Fq?s2ru+q8e%cZ6sgh*`v5k>5ZszJX4+ZoDo4mbl|OrN8*VU
zx%4~=HKv)Yroo71by(SLd<YedU`<s^r3dqAnH4_t8uDixqCHN-l9H&^w<pk`tUt6K
zrk1I7@|SaBW7k~a;_3Sta6XV`no=?$8t6&pPKeVVdo0Di_V%sWnl5qp*m-M?`ql^2
zI|yI0H2T%?22#c8I!q_LU#P=tQtW@^(;2;$S)X)wC)C<=5D^nS*?uOtV>D;b7gin=
zEKV;$`E`j|lI42UA#X)XG3cUZdgEH%6Z5R~%Y~$4TfYM|^%Rtz$2w8e#V3EphOU5C
zyp*fIf659wJNqur0@??nML2iZFfp|)8MR<)Y7UHw|LIJfUlDrx^eOrJ`us@q<-0J~
z&}ps)N4ahdfGFt<x{L54_sFPUW@cvoQwC8fjUz(npXg6T(zqUQ;tL^2NJw$&InY={
z{&UolqLnn3@j6@|L=zSkF5wK{y@4))5=|xxfVvf+Q}|^t2a5<C{keH~{L<3s>jyv3
zh57$a55*s`xda8t0k4;E21H{A_@}74Lf^Inc!RMH20bYTx&(dTF7E%^zyD+6>EJdS
zxP#>AcL<TaI<tXP6zEF-I^yBuYlYsHvAU4~fLwe!@e7$ar_)^n)jV}PsB6i^+#dig
z#O2`lL04kQu(sh<S_dVd-VgwtVz&bmYQO<JJUmFe3-8L~@VA#?<zGzN`HybiJIEFw
zQkMXNg9IUmd(cbvH|<aM8c*;#J8TBvsB75h>x(F}o}q+ZC98ccUk9?ujXaS9uYXU;
z-*T|h&vXW%k_ye}fq?;EXnCET>_kCZF;pDTsDE}g^QTXr^bHODp_}K^&-&!YgL9R~
z=O$Vi^URIkpob6dl?Gs)`@w@i$du6KOf4wT-R|k@?q)qP_AG=qYHM##1N!}4HHT<q
zbTlnLAJ@z4l40d%Z!df4vd`581;6=7Q+F&|_$D9F26pWVTxi-zy?8;Ws-`wr_ErVt
z4-Q`LhpT?fO3Cxn)ecBT3!nix*w$yzd>j1sm^k6{=f&-rTFoMTl&t4pf}vLx`@6<7
z>cZ+Q<M=r~<2V1QY6oWI*U4mOX9&u?H?-O*j3P!0^-^FI!Zu9J&Aa2+wOTH9CqyH7
zoMCeQZf^Dhx<{f;AYMk$cbaxYAbVN)@aE6t<bwYF`}de1+Gza=Fh<Y}>+R?W{wn4X
zhoT>XPs8dc)~L&cU}s?P2adSh6R!mXAmPRRH_-62Yze$c)N=+A5sw|=ur|c>cW0;n
z0Y87WH~jOTjv2H_dwY8$Zr*?nh!ON;ji5*N4Cohh>bQYV#(ntE0$nt0n##LDKO7(d
z@I83&2In^W{3Hl35}>69tlKbx=lF0<1;cly!RyP0!}<uzJwmJrkc*HTMZl$my|u6?
zjflW4=nuR}oDdtUR?Lr%iCJp$>obp<ni`UHLJ#dH(4zk5g?f$RR6K?^kgdJpu`v@2
z42<bU5hJ-BB{{3%!joUn={WAB^C^Oc8xI`10sP)8=70`-`>S&EcHnRcVPV*%<>kiA
zib?mKzc$D=E3wuU21u~5g#Jq4380{&DoscrzfDHgNzNvIy0o}Rj)3vt;!%>6L<|oM
z=))zOfXZNV^LHt7OrC#v_?`+AF+4tQ3i!Yfe!|YnThjIqdh-d;ia$6$HX)!D)dvh|
z(AU@3j^P8nVFwo%Q(z}=U=gw3yBBP1YFfIwYL*53&eO|F3nV)tKzg9%`VlLg%lOTk
zqM{;h3||^STi^D!wxWoL2<ZSlHMLlNi?Il3d`F3So<`SuoyW1<zh4AGgwl8Ku#N%R
zRM;;+IbIuPMt+S|JD85;Yw;B{aG_P+hdFX}6+~;=o@?Y9++JR82MxwgawAI+3p|6A
zZs-nZ;bVNfDJB8!n+TpU;DKWmR@$NBWL&y$5Sjou(vI%lyT@Cs*?aH#=QjBN36O?p
z!e9TkX}s(>%7OVo4iqv<L4slg6x;yS&leerJnDy6suK@?(e3njIW+mA<ikQ`zjMd8
z-gVEPL$k;?RF>fw+F%f5(EfY_6vqUXJaSAy_G7dOOOnZQ?ONsji3AJQzh;Qgw1NVB
z043y<lu!7hG}$o_pzwLz%jb3G_$*2A76=6p2pk-oUeH4LT>WE)5Uuh?q5^0*SC2zX
zvpF|Fx?zN_EFdf?U?>O#Br_?{I)&?SrM<v{=n19N`0)xIa96%&F)9dq#>U26&yP7k
zN#L8Kk}a*MxF1pQ;{G2+SP82nweO+RQ0%m!Zhbyo>q7AR_iwYg1~FLKC<uIf{MSCl
z#xR?26zXTf_qS(65WaqXezLWo$Pw6rt3p?NzML|2UJ_MZ?dHd`pA8FwO!FAX!2ITB
zgpk8uTPuh!f54K!4KZMv5WX21jFkkcg}O9s>iJEeCmN`-TVgn2PMv>v`ASul<l^%3
z7Aa}_78T4)E2Qp)ot-FnpF6Cq*fhe9^A+_$L2_wEq(S~E(T1X)r-b7LI;{}Y)fG^s
zK#F7J=O<-lWknz~G{%*Gk%5c@3H>A<BC2X@eN^7l)6?6!xP(aFnV*C%cL=0aS$X*k
zSaWj^I1!Njp4?D{L^i**g$D|P2Ue3L&p$tq0)~NuHxRbI&>B+ZxJEQGGJ^dKhkzg+
z>xv8D96cvzLEB#AY?dAlKK>&`Mf}PnPk0$5Qs{^XY9S(M$hZG4H5~+Y4QMP)Qmj2~
zp7|R&3kyqvxR>x`rH!8cXF${^wYj<L0>4g5&)w_)EHB$_PYb<n^rnGrsC|FAhsSOI
z%?)&PQo)id+JEvti;Ii7d3hHnp)I`d=TF<wKVh@d($bjd=%!ybW@hguCNx9E{}qOx
z<=q7>0f7Au5K>`3J1@BnMP2Ay2ngs2F%?XWPn8ud_w@8ELC%LGQG7z&waEkhCIs@v
zA<s0yW%Iafzh-|`h6Q-|zTH2B&ZrEQ{L7cOpbJA}1xynQ2S=vdd=&e_(+fNE!NKo|
ziFw#VfuSw@t%f}}lpi~&1(kmiUha5Lr5|Bu1=AH!QzHzVDO1NQ2U0UqFa;WiPGE)f
zB-nN^LJS+{w;}z(R@esNQB?s|5X1fZ@6LO=W@czWp;YB@!a2UO*dC5N3~o)9SBrz-
zib&(tR}mL{ke|t2#V#%``ok{Q)Z3fi)Tr|~?J~3FnE9=shzKJyGp3!L9eUjL@|Rc$
z%3xdy{{T#a#VCm@q<AJ%7g=RpBMS}q<Xwn%$QY=IEXA*3Fo=d{hik6;e{MqRDDpV5
zzPLP@YmZbL@`Ql#&&*^3cD%d0AOob@aR1Lw<xz4_5Fqs=tsw+(p+O+j(rxs9e~a<a
z9ay;_Rq+GKPDdn-_U4!2N~E$PNyx!=rH@8dPHq9p9@pzDF_2+ZodHq`giPvrx|`PZ
zR2w+j?tC+vsOv5QL3rzyA0#{E%8;D>(gW~^$6}0~TFjjUipbTOS^*%JZC*-E9UU0J
zYFh!9+HH(+AcY}tN%Ale)hcGNf<vI+0J5dMKR=0K9bsNyopT{C>yI!#)k9ps3`l7T
zK*0A<q`)O&0DI(6izcg(c-%TPL?q^ZC<*1R-R`^;63LGJ&_>F%6q5u|5XgG0FG(l>
z5<4=j#B=H_Ug3RXIO=Q)y!h78kPDh8C>0p&9^W%+*BfmmH-gg17zzU;Y;0_zgJf~9
z&!a!I<5}3)_<+?Rbx`o;ozLZOX)Jzvpa;13V0Tv^;2!9RN+5p7Jk*q~L_zLh05ZZk
zPa$R_H;sb%ydIA5nxuoLN-5&Tu;m#W7AHWmHhKH@XB?Ma-}9*Oa3f^82GrTE+ov_b
zl!-_HY0)^dfI$57MBX3@a&qn%g|@o-dQea$v<U*r2FZstq{jDe&CH5Gm{tPs{P@uA
zbeA4^IY0r`3SK8(0)~~dj`u`vBM_fPAf0)prIiTcky2p1rO+|x&}{Y!<>>&P%J$O7
z;9SK37UCXc;ol=UDj>9?P*GKV^XSo|d;s5@@Fx*G9l->2eYxsCx}laL4sXwgz}kda
zWGc`QM={d7@KR7fFS8iuL`Z^;;LYos^rUEC*XaWZlOfgof9*S|GSn*m0RcE=kir>R
zSk^kpWEIe*p<en16%w_mD<K3d+rVvNV#d~PtbhODjDY>KyBix|W|jRih;a!C5mr`K
znp#>vi+H8Iy-6;<Ndxl=h>s_emXV1`PL709EigE^1*8}D_GQ|{%G76c(yZ+A^73XD
z7L4@tC=ezCV`IS?85#LK41coa;iK~}GOl2=@w~c7j6A*>Lv+bSPx={lLK0F^kbx@5
zp9d;D29aM0r+@U$EeeX82yTA<SUaXbRa}w_P>J(7=7u0)>VKUc{}10ENsXaL@5g@m
z5)L^Zw!Z-J&cXyZ9U%`J`udV{DtC<i{jbT|$&m?-&~XL-gGA<?c-}_okHm=2>*n9L
z*0G@y2Ehj~JXUV(E4`7urB1tl9UPh><FYhQ+3x3kf1#=R0>smqS&P6Gl4@MhL4^pi
ziE_9%QgZZbu2oolRo)v9&E3aBRDMjBaE5$B;jp41S9=(o`<b#pkDv+aD-3>qeyw;p
z*89Q+r#S9c7ec%a_6Al;mE5PJno}YSe9UB^dx>{q`G%S?e~s=1r<pJYru!^xJvAiC
z9ne-xyRR($<VjfmQ~LjW<~yEanZm0oCk{ejR-Vrk&kTCld%OQR3^Yj9)0MD`6k~8E
ziC)RX9r43PRU;klNrvrB`w!qDa5#PDF-bf|ng=;+zCpv;cNoQV#xgHom`2e^rBrad
zm}(&}$a-X^F-Mbk`zqjD+YPOF49N-`)<C-Zr*0h?28>ANmAxa{5}4>vY!fb_-kZX>
zQR1E8W7^+T<>#fpv+2)YOE9l6|2!f^oXQ}IZfRn6hW7^tAD>EA5A%gob3dPS43;4p
z8ghJG!qvYojUog%^%k>M{{A(*9Fj9r9!BB3cV;%8%GpOP#?JGoshZF1<rD?HwtC?4
zv@SI4LxS!EOZ*QD_4MNKdfBOFnyG<~Ry@VnqZe}p!Y0p81HK&d2;GGKXGlo7&~z_b
z1~5Y`FKTw4>oXs+)}9Jh^F+0lPcK`eW0GHP)=1R2U~a^5ww<n(@>ox@3)YT4T4v4q
zK9x#;HB(1pRqMQRwtYP3Yf?`AGUEg1t1-Lg!TMV4h<NC<#fZ<mEDb~XTr0G&xb~4~
z*ha55jPp6A?M9_*FkMM$Y4w`dWJMFSv-G$B-285891P6QZp!SNk_*fNwk4l*H!8YB
zlo)|q7-nTa>V4HelVe;ufzYTmSodcVBq9%-4G1wv-;tbda3)0%9^9kMI6IaN$ogeB
zKg932`J3vg$yu!X@7hZN{joNQ0qBNP<5&5SPPrfniGY`-4_%v&YZOd|{@g#dBBy-6
zE1Fjh*+z<V;_uXO{u++xc~5v;VISh+7ESrc6!(Z0`*5kMjt-R1HA!b%n3<y>`cedg
zt3NU~nXY0{->eH&6u(6B(i=K>(W{N_3ld0Xhp*d!U-;-bw?P-8pkX4Rl6<nBar5hI
z+uePqXF5eUFkP(MC)*F0Is{8h$$+XM$)7h)ykY<E9)zz}kNVR7(#o`jbT^SsXFs1t
zUvh9>&MkZU66o5ZK-@mNGNnh{;bTPGxYh~8JCCv*zy6Ytxa2*6w>eXDvwYo6KkE8X
z0=&{s;g%iJo2S-Unypl9_wR%9C=ywn;?hLw26$CJ#9aA|@U0cFxpO+Jb0auV<~|Yh
z0F_eS<1j}pZ5A`GPS<i({Ll_Mp7BVN>7PaTzVvzC{>!(go%Tq+0XL*Lkphxb_$C_S
ziY<QOq(E$7t5M*cxJ=&C;Mb{%I(rIp4_CX(N$0J|xGj-o4xbXbibB1Bc@LL`wHfkS
z0V`jF%<fZ<Sq8{4nyQtu)3xO6$qONAOZV*Q8os<zZ5M1zL^mH87zvzcN^RRx=&DSe
zU!f2n;ChzWNd7h_XQIZ0iG+k?;G2^+LQ=KuDdZtT$<8^ZzmT#6-^<Lppz4Yf1Z0Hg
zwK-<Lvd+w<zq<5js}Gl=DZ~g;B=8`L80lyEZt?csU~)`y@OUA^{<`Zi?pLp~$Il8E
z){&j^@Ye-p8ZC96w84t?=L%&)wWA%!M;3P{^DVD^C?CnGk(#elDxB^nQ)L9_eQB%A
z{5rFK=g0wjw%#iMTFnT=6s4l_sH}okI$X9fDQUthRdht}s|svn`_*qtdk|W--@N;3
zA9^13c?}5gen{ZD?$V}26%yiB2;Piesfhg)8}?P==)m+H9`1tAsjT1If7wP2L0Y5c
zRb-CD&@W-4J$fp$kjDdA-TtYY%3Wj9*qwra=%I9n3(<QZ3ir#nZYrtfi%E_P&Q%{H
z43mB5c0ieRTaC#(f|Uvl!Jey4e|n`8YDYU;&9V1NUV~uIu9!%EA}xx}t}!n+C&Qjp
zWg55YwV6&uB>yhp4k*9S*9XQ%f916f5k+m!#w6?v&P)a2b-w+bMyxE6t$&hr6r|XH
zO^fUTPdAd@<=MLWaYwbhjMe#qOG~w~rB$TzZbkE@wm@2L7JloK*zm6!QvW$ID1$L*
z2*#;8%W6_eSC_W-;g68q&%$S?kxf83ZCy|BkvwZSgrLail9GD19t|;2tt5ejv@LIa
z7#G_es3nTxzr+=)iz(gpJoSCw6VJf9$mDGOI;06_<u1HU$~%rc4SfJ!?{IyZIWQ5>
zF%G6Ium*vNNP$j`ufWTgDjzni>%L^aPJ(R|ggZWUm`rWO(m>yqE9n(A+uvUu;R6GP
z<D<nU9EVgqlsFgnj1~@ATx~_LZ1m(oEMSa8H}cLqHxP$Z;@BWeXtKV>gU|B&m8UU(
z5t^~eXZdXegvyO5t|!NMAZQ4Q?@w7+%pukk^1%mv1lwMZQ|$BHd6RkLx~TVZ_R8ra
zJ$Geakz9yor>6Q^Ut$^zhh-{UUQjSHKW~mM_AzX23?Hn~<-pyZq}AyA>SssF2Oz_U
znEhxj7*=nFS3u1~tsC(prMSA>?A%P!tX{ncoiAhXx;s{b3z?S}lgt6ESZ=#9iHWU7
z6f04S)=GI(7*Nb;m+m*ewl8g+3d0-a&(h^+$Reml$ZRq)QrX5Z5f5y}QDi!<Lgo0f
zQa`$Q)d7msIXmXE^SVwDMmrd{l-->XAs?w0n{*xNsLSzb9_ov%O&YG1V|(^ljNpp~
zhYVH!<V2-0u?dGZdZl}^vW*5>#uLqY2MPb3Ik|HTCmye9v$CC8s({Wbt;M<})81y7
z%Ym++7&KH*8lOpjW70pI+0j#k)|H}Oby%{Dd>51f_!Wh^ktRoCAYWFqH<iYglK)R-
z^k4PSfB*g4>t}~F@0_0A?yiZTW7usGDzkY(07qWH*n~MRK<`6#CP<F5(x2YhM88}S
zak4XOa53+V+-8r@J|~E{NvB=*Q40mCZ}Tr$vUlw0IuY7#N3=P1kjGu~;eghrwOy9(
zQlw1k%_&Z(_~X!<rw$4}Rf!spzUJOb;58=~_u0F_uC&tuvaF`POOI}ExtlDQP)7D{
z{jZ?eTu2;O^uKZD{7q`?Er)A$!VZG%kYh1-z@&L*6Ln0FP8~ZBqO)J8z~~ruo|W%v
zSCP~z>IIgBVPiPG<=V6Td{YKvUe4_P!%c~ZSdrt!zV{Xh$MVfPOrBm`bkg4H@R`q+
z!Nvet-fZAoqO6AznMb~M387Eap&Uo)lU?VMcF-Nn)%|bCxNK{PW}T8U==gVoH9v@G
zq#GA9UW^W5Lpb{VFv8oMuELj`9lWY2&*OI8bVX7vb?0Lwy;+=XpKr;jW7ZFnz-b)~
zU*R!Yvlkr85Blshy*ECMF6gw1fM2ogF&`)sN*1k(0uDD<rx<u~A9;mh9{c1e;`dtn
zer7h6{mP4n|DZVM!brE6LrD<w#zpeVtjayxfV`ZSF%ki_Wn5|;1l!ZBwwMF%jc!ZL
zU$qgwt6dtWgk)?WjFKUa<ScM?6)Eg*ovKVt%RxWq8tbgTvWW&wbd|;nSUjs?M|LNF
z-z$$Qykj4j5qg~HPIZaY#SD;!<2*Ys9|^4Oy!4GRDze&oGF?N7I4~bYL4aOD9$E`A
zMr<r>@wssXcEtmu<MC$0q#KOloX3r8b_j%+U9<SVVez(hX?Z0j!y0JfKPWHXTWGY~
zU!$U=zxQhY?VG@h2oOIcK2cH%cyaeb6@iy{57)LH7a>+^o{jJ6^|l@XBKBGCMKonk
zxWeO`P-Y`Z5D`+JmiZ$iylh1Ejfsi$eUc~7R9-|+&)gvN^|HJ9hKfJ{)@MyuvBj8O
z=cE{BW`Ud%AmeMA>XrZCS+9WWtQDFyA6Gie;<PkbItvd>>j<zm-0c7X=p%bWhQmXq
zl;|TKmRrI#AAD#C_xV7UV%$9`zVldT!60o~rS_ZJShhRhS6jYaY<B{Rj`Eo*`<sdd
zZc4a(5BUR<8&LdF3vJ7@JkQ?ApKRN@oaMjl!K3^5are)H*TkqT2I2;~t`v0FsV%Oq
zAHShkTYKkD`f}arp6qSpU<?foBl+hx`pkVNthi%UB?a5rRvl3LJTkn%47?%XpZ?;V
zx%wEwH#s0UU1{p?M4h=%9i>2fj?DSqX|vZe?J``+*|jVvbdWWVYVmjMCDts>tA6%X
zD$3?sH#US@Wx_GYAH0no8Lqc5b@g?3Y=$lk)ahWZQe9oM7!&Nw2EVv3jW<3d7W{!w
zD;4g&0IDFt8W%EmT=LfiJ!=yK0q2)LTh9gU^yyd=Rhz3lqEP;bF}na})qLo3FRm{c
zUnM+Ov%#1-cWT1^;E0Zg@@-BEkJY9EQvDD=xvg2df{=Z3?IZf9j-QI#V^Q$XW)1@y
z?Pk-tc=vuKa=q*UiaMBUWa|EjA%}3)`=arV)?Ac?7zZB5jD_K#yAM*s_L0*2TidB?
znWIbdtb>NWrMdccPsaJ)#~r!xy^p-v-|{1X^E2?!OO0h-5ZdU<8+O;VOWZxY$$|q%
z$4}uyqFF`hr7rd{r~YL05+cHj*Wh7hWnkm`xPAK{PB%2t)J^U|od?7;r|JHwj}$mn
zB&{DBriVA&OICfxGgwI9^{ek(;NBmqm%G>gVBd^*V#tHq|5Zv^x-P$m-uNyEy8Kr9
zniSX{EqBKn>)_`!t>z}{se-TIq2<23r8L7xxl`J=T;)|OlCJ(Jp5Vj&oYPO_1ZXr?
zxU1u2{{>K0^+_%oi@+l$>T^ZSrPly%A&k|O8bczi8%V2bQIX`)`Y6scHnRHpuf~SM
z|6mXJ))F2W`N+|+EWhW@w-%UD(peBR+a0Z|0U=fhk~&$ZC#BU%c%-m4aO?l&O!faM
znko*BcmnV1OBbzpdO!b^iw}w^;_rWb5m*8}KQw#&02$%aioQ$aGrcX;AB8((b9D4&
z{DfT8m0Q?(^Bp)%jKKN53Z5~b0LMB;PpIqxvjj-qi9Ri^sEAwr2u&4()?g_O4KirO
zK?@iSjGjzq&<agMS`)zL+t4815k(gT4&a?vm~TJ*O<kDtxo!s!Ku>w{kaDyWO5mrE
z|3*<?e~FCjBhYdNA_Ot*(^b--1`Z(T3J(AWRn1imwa`#k=d@3TNW%(h;mvFA>%(8^
zPg&7ryRXnOn80IXyV!PfbE=Zy^XJb<ODQzzqreJ<f`C;!P-6TP$ju_Oi{Z|tpqGU7
z@dBavr1f=<75dB2U57s_;mo}ANCCYPjL8-7wOtN<MYN_%Kn%Pc<atOI2tM2K{*jSS
z99kuT=k-Ny<VKsj3$4Ase<vbal-AaLKwvvcN=mBM5EJ04(%cyPkq8X%;Nak+c>`bp
z9<Uf9{R-fu6Y)4^M{5Fylj^FwcfBq`6U_18AqC}iDbPcnv=?2tczC}7QQnO^0$j>T
zO1=Wvw0CYU<*m87{Us14O=x0Iz%3!0kHlCvpy3Gq`%bWFCIXrm9vn27ot<T8VnWhd
zQekU@{Z7s&&}BomDS@jlO+rhY9O1mwfd>o>8t-ploHn7g{5k~|9<qHK7ZdaH&&11@
zFZ011{7gmVBa)kfd&72$$48{nYAOzi(I_Y>d1+{9{%&maKr58O$nAV^eB&qiQpFsY
zB9XU=q<dgdpLRuJ$_Z#jgVUP==_~;&8qr1~&=2tN_knLL0WKPRG?Y^?-7y;n9I(*@
zojC0>Gcgsz6A8G1nn3rj72<*1s2$S>EI4RQfSs8B-o(Pvr=+B`1eC@uDw>33e=s1$
zK;7+@5c*&avV^cj{sheKae{ildP*xRO<`y@K~aDNeK32~6Z1+>o*05H$r2n&NKlHz
zpPQSOw`pj8sj8|b00_XjdD8&swxPSb`!v<SI4>`+CN#8R2B`&`3=Ivz0rDuVxiB}k
z8*E4&z-BtZqR}<m;MD^s;RygAS}?yJ;9=GVGb_?ajQrY|uI2;aPobgr5@mB`r4#v4
zF){xE?=TalF)=aIhd5-M+z56ao|lFVzy^@m6`<u8UgIZz0#Wf?db|j*=X(Tk3V`<v
zNKO%0R*Ru2Uknb^61WOJHa4~bbvin_thBUdASW$Ih6-+5B$<Y;I1NPH>8UHyc5Bj`
z(EI~B@Xdwazjv*V{pbOMR0lLqf5RffxpAGZRf=@kh<Ken0E}h|F6n1?iIC9>S3?Dk
z83%iN!%CYu6PW8w5Go*TEkOnSm5}mDXUCiS$&H@IvBwCSo93U+n?{gQ*1!Z<0%;%u
zqp+0`j2+Rzds?v?)q}4>4#7yg43;e0VX&#GS?-cm-~-Vg*Yn|ln~D1Z>Ep-i>JK?N
zImJmyNz$E<pFLx@mDSbUWn*QvxA6xF!^yS|hA$Z8atG`h@vGCpuKNM7XFQT#e+5KQ
zU0Yj<97<qRu&$?yjWCI^gV6ux!E*)9fdYUw0*_iSn*Z&Ptl=5lJO#oT*pPHQ($kDO
z#mdMS0DPmA&$Q2WLm#9Q$flY2S79eZof^jlU7z|L?zqZJL=iw9g)*P&=R?E8$P7^H
zeOV)$0{-?=;LtoB<QCvz1Gke++EF~IaHo<257N*B>B<10^lNPhQuw==&!4e@3nP5<
z@+@-xLj3uGr?%~%lZ(rVO)k&fyEh<S!9V|68*VlF!v}u=`>Hw0%U?Rs5nu~{*KWA&
z^graHX+M7G<l2F_pzm^fb@d%s<61{XNIXuq<p3!)R?mJwB3gLPhi^?yjjV^eyD)gd
zs>k8U0e~A<bg6)z3JO-W?FC8G!y`7rldiz^`YkIf-EOIaXbYB}t+O)*0+&XZ1ki62
z;Q9g3s9ao~?+~S^#iE1sI4lOP74{~iCm%fvkSemJk?TF|DN$QE{uEjaNH;G4NxS1s
zEp~Wl1X#Wwfg%Q;c+lu31X!l1tgH=evj_y}NstCkZ~?b5Wg+S6g9m@krfZy@8YW%j
zdGDWlasWsL=|UT*gaD-O<85ycte`_j#=+s~YXTtTp#t+QaRdmUrjk(qFl9jwN4VL9
z!Lk!{+rJCOW=!a_;{s8GHOOxFZz#P-=;ptu^mGPDWp^jR2TgeUb};zBNKa!vvYj<5
zJ^p`+rG9(_Tb~QlZ3OZW8z4rjvRyzOXJ%)|g={>n_Xy@43$Qih!0{DWYe?<Vj~`ZP
zJwQpDL1mCuSg1EYm-#?r&U3>Z2#NGvB6?QV+@}5K&z@Czg9#hd3KI=pB5|BLVMw)*
zyH<M|2*a<8;Atdqr>Wq%M0%{p#yWs(4d$w`K~$mBh&TrVtw)9_=#Gp);vx6nbJQaJ
zW>n;(L14P#aywE}@6@>;VMApwFgWN3=@a>3z)|&2cNc&^2mGHnDpFDaVFDIa#{2iN
zcIFyGVC&GBgW3d!7y};K0oKaf?r%3hb}&%mWPy}CZB13~9vlR~(-&%-HZTwaBO^#6
z%HzDLiIkzq$e<z3(eQ+X^^qJjgxS!yo8V;~1c6C!iUb`L0Ig7qyodV5PQG6c9`S&A
zn_bfvT)KuG5ft1HA3l2X<}R{00$ZHQy1)}~Yu>nV1L>^(FZ20Fb8)Z}M^Fji8+Ak=
z&q-kPjZR96fM1Y;PloYL5O!jPC#~EB8#L%J=Aq=8KIUL^Y*Xs;xy|t^&|#&oxvecw
zWNs36JEX`BY6CgYCcVg3q+!=8xsjEX^&d<PfrhkV4-E~?|NV=6Du{w7?Tb&a(vYn<
zc=82dc>BAsFlmThyZt|pA@*^Qf--nI#z3Qw4-$6IHNI>}PfPPdD&fGf92XaNdcM<G
z41VeDq)VX72FAwd;0;D5i_hnUa-ge#U7@I?L>d+$7+Aw0OFT)k)`F)>Kv5C|)k+f(
zG+21x2)`qs0~Z0o5CU^BJu@>{YN~X&JwwH3(tE$e8xj|C$Uvom6s7<bM_v)YE_l3=
z%=#9$Q3ozEKETo*`(29?l4;u~78P#LElf;Iq-vF#ARlStlKbpFiQ&+<$F`1+fk5&=
zVNncON|__$F6!~}Tu7j*B&fYm5!+@lgvz4eRr&uKHT|#qB~^wK7u4Q3Wt|-z42+C2
z7^Jt9?Vw|Z%>o$%G_CURNSpCE2v10=jO^^sf@5#(AAp9D^aDI?229=sy7k@VI`!_P
z@DvwlnDFLDM5oZ9OXNfYw}yGO13T{<V`F+YHXJaF%7f3Ho+$K<4lTU|Keo5B0kr6(
z?vpS8LIvFc=;z__JaPXWLWOO15d^W0WlHZpA&WYY1z^Af0{kQ;|3gp#h7#_J!ADUF
zSodBkVzDIp@^rk>xj`QoO^3AV{SS?(18QQtZU1FVcW{97FFmx10w6KpK=lLMxE$fH
zhXMHa7CHGlc@R!6Ws_*dg9b)M2QSju8J@-w2<r<Dz{`Q0D;Lf1i^bkc_u8_*HMex!
z2Det%&z_zdcIoIZ9|KN`cXoOIusDu>;wkcxz`oWaLBA$&hJ(h9SvOzjde77Xsq37s
z{*~d*H`J3A=(^=e@azd3_^@siud4F^G57r&w;xuJoc+npEJ(YFL!~Y1d7(L$8Er4!
ze2oTTJb2oT(g5hsomc`%o<a+VPQqK-*|`FF55E4_m~h5NN4vwd<Z#Be(l2$6arEX@
zYo#kO=}my8(*ATm%Oc^~K441xNLP~I!FdeaWtl*0EFKn0Goog3lT5Dp`s3+;5mf98
zXdz!0*lo2@R@v^mQm8$C!nkUV2zK1^t%=L}uGNnCu`$|mk>f-E@b8cOr)YO-1xOh@
zNG>F9MFqdNMcQNNj8sZ(DB1Y<N-e|LP>i0f52FI<SbqI}b8DvLCYo@WAGDR?&TKjd
z^D5^n^;wj~z1#k(^%%dJikcPZ>JRb1W@aD>z4MSe`ZOj!zEiI?L|=Btn#gNKG1*rf
zoD=P!*XQ82XrY5<F6z2gqQQ$5XR}R10tETmFbyE%ZKnp{jA_N;V&>Z9M$=8=r%C&v
zU!WcgKYB`*_V_twRl;fp2$cjK4!H#`$I2yvSQ5Hku?xF-H0PAiBD+;ZivO}r6={SK
zDGz@kyTs7wMM0cbT1jox6o3&a1I$aClV!Swv(OYvZMC6AzUGi|hOW3r;@rL5@^h((
z*g~O9N7w-yjEEtn#pFl%z1&0DA_5#h{D3?xfoG}eGiAEP@SN?&D}BFZunMM|kfKE#
zu0{T73q;-cm+4|N2L?1BWEVT3+>&&9-7OtpPSx0t5Q?W7`m0qpU(bPIDD`~^6?mZF
zX1&sEutA%D>hX;W7_fikM~-9bqxgiY%{Agft8r{V%s|Sp8;cqDgqE-82!77_jB)Dv
zVjRbrE$b2^S?<*M%LD;qqLXo^1%flwajrjI$vUe;l@b*xf|#&t4h3hwM~9w29!&FH
zac$R2R#Y;@<fdF2k3Alin?{(Y`mK#mxy|=g5lv^A;kKUFv`9z_sLF1o!wEX}Sm44O
z$z5rFU+OycAR^@<PDhr>z`zh|Nd-723@#3T7unecDtrrEIzQ2GultQOS%N{9;qK`z
zFvR&nSKNDN5-mkVnLA(Jv}*hs??Am!-f@KOg3nDlW{GUM`}Z*U@-R(6M?~v%>_&8K
zVrbKWY4uv^y+|2X&HJ|t4nigd2+0&_JLfsK>`uG!kS0-i7aODXRRi+NlLxMqW}$gx
zeMZX{)$N=Iy@N7c-MzBc3mG_H3(^uRzapJV5*VjVC;gAjt~@-Z8=Y`U??nV<Rp3uL
zZ8RUv)wv5jdUipDCV%aRbaDMX<@v5Qj`nvnml4<>?)o%7|Ld8el}%%S)&z|_jL1G=
zuU2s|)e?AAzvtu`wO%nolNk-o(d-2QjmL>pXR_CmdY#AZkD9NEtL={;yng<^YCL)N
zees)N^_&%f^6z1G-h4P^K(eJR@3=ADGOGSx3OUnwDAzuY-x{(nagf7Uj?}3Xibi%K
zWJ{xxeaVu2A1V93q);epmLp^*YsN$=$x;oDWMs(}<Je=E^PlHDFV2hS^LbwT823GM
z&*gtzzwh^#S!X9_Li|jJR4C$Jf9QCBIJ)$0@7oTb0&L;My~?f%A~tr8$-C;9YTV+{
z_z4ji+cOL2oK5ZAUmj%Fn!4uRGz<|LVOyZC`J6rWY*cKt9qC#qWxRpWP-rX`sHR?v
z7H%jnGO2r^4u>b1G9gKyzo3wzQ<ibxmlSNaquyC$f!m5qxUahkjW?iYMU<l3ZN~%1
zif%_Lt(5M=`eU<B3K@BEnru)<jT=Oex-#Nl69HV6R&AH3&xl*cTrjXi*<I+gbNx~q
z#IO)Z>>f}(>SZ6R${J1W36+Q-l^gT;En>J3X`AW8Zsx7rjg<#G+RXg3<(oJK1m<Z$
zg^9^kV|2%KGpjUstVuZvsGyx41cBUE-@5hy&}4$w4XuRB%3K&W*;f|<MTu{^9cbG_
zc|GRJ%S*iS!PC4`ccC2)>AjwEA^>qf42d8#iv$km+2*!eyOe<Qf~WMMbDghGy36eV
z!_#Br(md6vjdCV^w6YMKLi;S8KmX_D;kSx2E2}pSOS&(KUAZ#($FP@wn#W8#*N%MC
zTk+9ZqdZYKNjl}ys#EXe;Hai&K8sh#CA2!6jGu?cB!N9dTE`!smC>L3)YkW%lumOK
zkpQ6#{Nw7}i*~%VOU3%ZDO3wsTGW+&yv!>bhgpP|NF|@_YgiP;VEC&bQi&ja95VcE
zof!}3xtbu4+PD}usI|ga-o78<A0Q_3;6(W|otKgf3|jHS2bh}lwyvno=7?Cm^c7jZ
zmvm%@86VX4wEjJ%)_P&K9|@aJ%pee6UoEh{b=v<?8qdQf%l5ld+h*`H*ywe)v(mw}
zQJVC$t{N;Tv^$zLoKOkYf3u4tUs77ghym3s6iY{MQ{2{DE4zE}=f+@cD{t6~T;cKV
z_B3Z!U{{E3;2^(!_wG1xwan}W5Da49Ke$5F{%BjK%*~>#hNZBiiZl3vI0$)HZ77uT
zOi9U#l^^r=H;YPq*OXiCnA-`#+H%+MXV=RVgrPC9xI)?Epw;Pm4nb{gsJapTwzqr0
ze>!i!y1Jg0ZE~+EQ_7O7<ZkUL(x?rKn)-oxNHm&i$fTCzGE3tXS<>ZZ1`8N9R=tx^
z%FO8_2e%?UfU4$jB$n|jR>AOKS<;XFg~F@m&MBpI+e<`y^M@vch0ilFf>Ih6ik>vq
z>hg}UwIyxU)aV#qdgvm2Rn8x(3Bqq<kA7QOM$?|C5>64D?j1S#wmaRp=kKWOyK7Qz
z+c=pQRp$4uf9wAX)@&hAw?HjLka{-~Gko45_SzF#roSJ6pqSd@z-=%8*)KY^mSsOj
z84V_!H>G3zz{XF=$kQ0N>|yz20+|yk1#CKNC6Lv=(#C}mfwUOTH_cPphtNM+?k|5O
z&=%>LAGZ*-{_qQ48<V%Ut1M#PuJDi4*q*B7#XVsRGj_1o&pglnWw_!E07YKAqgUI4
z(b0F*F)~Z{jZN#ERgnDtfkRCKUJT{@A3DodX5nrap_?pLrCmG`P1UPV@QZ=<93R6;
zfSi-5<=3$Dq4%^G4K15LDQT5j6tmDQI>2f`n9c}#Tr^oYQH9Xl{s*f2T?O;l!QywG
z+u5lR=W<#v$&t4Oe7&YtC1ibS)w^SIuSoe$7ac!*X48oFu9*#&M-wc+O!YLMm9Dhg
zyyq%bYA+=M-+3WWhdant%PY6O2tl47=FN*m!T?*l=0S>tH>Uuq8z_oSN8cFch{?pH
zzu@>@VRb)j&1tZ*UC~f@jFY5#On2S8d^Y~IVB_S+dF<SZx$?RziG{GW?DNU#l=(fl
zG`n!p&3eD8_>`Bepbg%UKGWB#UmsI0_=l8L^3vk9!adIos8!Hn6IN{R+E@e`b%?B(
zL}Q8mEKxKg`bU#8zg@;&*CtUTzP^az;d38}z{YaqS2&^JX=3a=rk}oT(|Dxk#0k^J
zEw0ow?!jVR6krOh4a$NxtE#Me**I8u1J@K-KDZFWVjG%)IN%lenzP2()7#tiS>Wuq
z@p@heu6kO8G!oXqYWK)KpDcha8Hjy_7yZBcAk{8SQn+bj(X<|0Xn@o!jlqFIvq@HL
zl!+_^yJOM+w!Fou=LW5*fIU2#YxJE`kPWDz4Lz1w`!U*ENAQobQgV|YhbC)5vjt2&
zwkCPG5a@OrR5?1ZcRy*UcKDFsW6rP9Bgmk79H)qRuJ!5MQhwRy(iT2tC_W1(A$+s&
zB%kjmn8(6?VyO}2ruY$@7}vD`j>BWk%9AAg>`rgi8D>~8X>AiVGjqCdA%tftsn*Y2
zhAS91w>^ty2UA%RBk~@W!xEUS96-<hL^jRY^M~DizzFzu!%{q<N+3P(!G{ct^_N;l
z28GRa_10x&W9U=>qDw>fh5*1`C<G$EBoKrr`vrV2ch{->aFMqkswL705>E2i<Qs3#
z0WJ_<n|${caKn0WlCZP#+4#qkmh?9NemCp7RBTS#*-dLU&Ea;Gv@hkbW?NvNf#<(H
zX`5aS{O$a9OMkYx5x%X6{ZLdrEint;rn#h3qp+M^hDHVc1tVb<wSa4UV~d!H9U!lw
zQfiom?;fJS#%ySl1Ju$ToVy@+r5^9~dx-xi(HZhHsZ~vkNl3>N%iY9al`3dX*lHc=
z;DS4R1)=<7fyeUI!K^#`;|FLMf;;=MBNAPJ)R(BCTc9{A@Gmxm1`s7^h;2nlR)X2$
zohyPs0m2FZRr45nG1Wt`$*Y9ZNYP*7S%Rf?y*UrltqSvrpMs3N83O`sK>0mDw5t7=
zE>Y@Sf@OR|1fd4w<EM{3cP%tR=xM)P@9&qA{v-hurURJW6Z3T9?n+|`%4L-|y3wv%
zLXti$k$MKpbSqFjL-QH7Mfs7*smc73JRCV#IT&@*!Ew^wC$-Bsl(M_U;NhK}oyo*=
zwxv*_G!u4{&9yK3@0i#JnmUJfpGz;2EeT@b0Zzm&61WEDXpNB}uUCM@t$3tt{r(!J
z9a^m2xF3li0dm>-`Oo6A=&t`msX@a-Tzm+7NtN*K_HVNE24+;NvJJ6mCv}2R&!gmA
zVK;P+m(<+F(nvh~3FsJLhGEL2q4uKV#e<ukq=@{ElcspG^l-P+TQ7>79n+8O>~9m3
z3GS}f#ZJlbF12jXBhOr&3ohIr&s5!za&~crsR#}IGqKoHnE2T}=+V&xLq47}oViqo
zc2|o7b(rMkuhGJ&gBf4WK%gWYx?ml5haV;3;9xpBYTp@N_-yIh>y2CEl_>5>RhX>L
z!&i?LI04Xq)h1>8fOLhTXz3PDBY6J`#y!UY=d+7<$Wdw%@fVbpBnt9~sqzGIz31#2
zW~?bHDSldf$3W7I%voaA;9%K@13^(UakMWJ?<j8JGMwUnekMk^Yt^-I5v~n;<=UIE
zN3JQE4`{X3=KAZdY`^YxOQCnBhqX(BLPWumVnS~6uZ7Q4^;G1p_wVjKer#q!E6KT$
zVZ03PyQtc@ie8AaUl$bxJMB(2$&0;NP~EJdQ!{=M9}{k>PoO3iOi&3z=pToEO8(+m
zIB%XeJ-dExN3I|@mM}<YpYAKtDB#3}(o$zPQVq%|7awI++%rJL<QJY;#lp;?e%C@$
zr7alMy1+O0XCY&7^)9?(M)>pRlJty>Q|~^N#2M*m8X6{pFnolZdr4umx~?HQG7|Nl
oG@bkSvHt?xP`mZN0PgRhQUVndgtq9jP(hC9Y8h%)YS={l4e*EBlmGw#

literal 0
HcmV?d00001

diff --git a/ehrapy/tools/cohort_tracking/_cohort_tracker.py b/ehrapy/tools/cohort_tracking/_cohort_tracker.py
index 612ae3a7..7ac56e07 100644
--- a/ehrapy/tools/cohort_tracking/_cohort_tracker.py
+++ b/ehrapy/tools/cohort_tracking/_cohort_tracker.py
@@ -1,13 +1,13 @@
 from collections.abc import Sequence
 from typing import Any
 
-import graphviz
 import matplotlib.colors as mcolors
 import matplotlib.pyplot as plt
 import numpy as np
 import pandas as pd
 import seaborn as sns
 from matplotlib.axes import Axes
+from matplotlib.figure import Figure
 from matplotlib.patches import Patch
 from scanpy import AnnData
 from tableone import TableOne
@@ -71,9 +71,9 @@ def __init__(self, adata: AnnData, columns: Sequence = None, categorical: Sequen
         self.categorical = (
             categorical if categorical is not None else _detect_categorical_columns(adata.obs[self.columns])
         )
-        self.track_t1: list = []
+        self._track_t1: list = []
 
-    def __call__(self, adata: AnnData, label: str = None, operations_done: str = None, **tableone_kwargs: Any) -> None:
+    def __call__(self, adata: AnnData, label: str = None, operations_done: str = None, **tableone_kwargs: dict) -> None:
         _check_adata_type(adata)
         _check_columns_exist(adata.obs, self.columns)
 
@@ -88,7 +88,7 @@ def __call__(self, adata: AnnData, label: str = None, operations_done: str = Non
 
         # track new stuff
         t1 = TableOne(adata.obs, columns=self.columns, categorical=self.categorical, **tableone_kwargs)
-        self.track_t1.append(t1)
+        self._track_t1.append(t1)
 
     def _get_cat_dicts(self, table_one, col):
         cat_pct = {category: [] for category in table_one.cat_table.loc[col].index}
@@ -111,30 +111,30 @@ def reset(self) -> None:
         A full reset of the `CohortTracker` object.
         """
         self._tracked_steps = 0
+        self._track_t1 = []
         self._tracked_text = []
         self._tracked_operations = []
 
     @property
     def tracked_steps(self):
-        """list: List of tableone objects of each logging step."""
+        """int: Number of tracked steps."""
         return self._tracked_steps
 
-    # IMMEDIATE NEXT TODO:
-    # I ALLOWED FOR THE AX ARGUMENT, BUT NEED TO CHECK
-    # WHAT IS A GOOD RETURN TYPE WITH AND WITHOUT, AND WHETHER
-    # THE AX RETURN THING IS ENOUGH IN GENERAL
-    # THEN ASK WHAT IS BEST FOR KEYWORD ARGUMENTS, THEN THIS IS DONE?
-    # ONLY TEST FOR PLOT SIMILARITY I THINK AFTERWARDS LEFT
+    @property
+    def track_t1(self):
+        """list: List of :class:`~tableone.TableOne` objects of each logging step."""
+        return self._track_t1
+
     def plot_cohort_change(
         self,
         set_axis_labels=True,
         subfigure_title: bool = False,
         color_palette: str = "colorblind",
-        return_figure: bool = False,
-        ax: Axes | np.ndarray = None,
+        show: bool = True,
+        ax: Axes | Sequence[Axes] = None,
         subplots_kwargs: dict = None,
         legend_kwargs: dict = None,
-    ):
+    ) -> None | list[Axes] | tuple[Figure, list[Axes]]:
         """Plot the cohort change over the tracked steps.
 
         Create stacked bar plots to monitor cohort changes over the steps tracked with `CohortTracker`.
@@ -142,43 +142,44 @@ def plot_cohort_change(
         Args:
             set_axis_labels: If `True`, the y-axis labels will be set to the column names.
             subfigure_title: If `True`, each subplot will have a title with the `label` provided during tracking.
-            color_palette: The color palette to use for the plot. Default is "husl".
-            return_figure: If `True`, the plot will be returned as a tuple of (fig, ax).
+            color_palette: The color palette to use for the plot. Default is "colorblind".
+            show: If `True`, the plot will be shown. If `False`, returns plotting handels are returned.
+            ax: If `None`, a new figure and axes will be created. If an axes object is provided, the plot will be added to it.
             subplot_kwargs: Additional keyword arguments for the subplots.
             legend_kwargs: Additional keyword arguments for the legend.
 
         Returns:
-            If `return_figure` a :class:`~matplotlib.figure.Figure` and a :class:`~matplotlib.axes.Axes` or a list of it.
+            If `show=True`, returns `None`. Else, if no ax is passed, returns a tuple  (:class:`~matplotlib.figure.Figure`, :class:`~list`(:class:`~matplotlib.axes.Axes`), else a :class:`~list`(:class:`~matplotlib.axes.Axes`).
 
         Examples:
                 >>> import ehrapy as ep
-                >>> adata = ep.dt.diabetes_130(columns_obs_only=["gender", "race", "weight", "age"])
-                >>> cohort_tracker = ep.tl.CohortTracker(adata)
-                >>> cohort_tracker(adata, label="original")
-                >>> adata = adata[:1000]
-                >>> cohort_tracker(adata, label="filtered cohort", operations_done="filtered to first 1000 entries")
-                >>> cohort_tracker.plot_cohort_change()
-
-            .. image:: /_static/docstring_previews/flowchart.png
+                >>> adata = ep.dt.hepatitis(columns_obs_only=["class", "sex", "steroid", "fatigue"])
+                >>> cohort_tracker = ep.tl.CohortTracker(adata, categorical=["class", "sex", "steroid", "fatigue"])
+                >>> cohort_tracker(adata, label="Initial Cohort")
+                >>> adata = adata[:50]
+                >>> cohort_tracker(adata, label="Filtered first 50 individuals", operations_done="Filtered to first 50 entries")
+                >>> cohort_tracker.plot_cohort_change(subfigure_title=True)
+
+            .. image:: /_static/docstring_previews/cohort_tracking.png
         """
         # Plotting
         subplots_kwargs = {} if subplots_kwargs is None else subplots_kwargs
 
         if ax is None:
-            _, axes = plt.subplots(self.tracked_steps, 1, **subplots_kwargs)
+            fig, axes = plt.subplots(self.tracked_steps, 1, **subplots_kwargs)
         else:
             axes = ax
 
         legend_labels = []
 
         # if only one step is tracked, axes object is not iterable
-        if self.tracked_steps == 1:
+        if isinstance(axes, Axes):
             axes = [axes]
 
         # each tracked step is a subplot
-        for idx, ax in enumerate(axes):
+        for idx, single_ax in enumerate(axes):
             if subfigure_title:
-                ax.set_title(self._tracked_text[idx])
+                single_ax.set_title(self._tracked_text[idx])
 
             # iterate over the tracked columns in the dataframe
             for pos, col in enumerate(self.columns):
@@ -197,7 +198,7 @@ def plot_cohort_change(
                     for i, value in enumerate(data):
                         # Use different shades of the level color for the stacked bars
                         stacked_bar_color = mcolors.to_rgb(level_color) + (0.5 + 0.5 * (i / len(data)),)
-                        ax.barh(
+                        single_ax.barh(
                             pos,
                             value,
                             left=cumwidth,
@@ -210,7 +211,7 @@ def plot_cohort_change(
                         if value > 5:
                             # Add proportion numbers to the bars
                             width = value
-                            ax.text(
+                            single_ax.text(
                                 cumwidth + width / 2,
                                 pos,
                                 f"{value:.1f}",
@@ -220,8 +221,8 @@ def plot_cohort_change(
                                 fontweight="bold",
                             )
 
-                        ax.set_yticks([])
-                        ax.set_xticks([])
+                        single_ax.set_yticks([])
+                        single_ax.set_xticks([])
                         cumwidth += value
                         if idx == 0:
                             col_legend_labels.append(Patch(color=stacked_bar_color, label=data.index[i]))
@@ -229,7 +230,7 @@ def plot_cohort_change(
 
                 # for numericals, plot a single bar
                 else:
-                    ax.barh(
+                    single_ax.barh(
                         pos,
                         100,
                         left=cumwidth,
@@ -238,7 +239,7 @@ def plot_cohort_change(
                         edgecolor="black",
                         linewidth=0.6,
                     )
-                    ax.text(
+                    single_ax.text(
                         100 / 2,
                         pos,
                         data[0],
@@ -253,12 +254,10 @@ def plot_cohort_change(
 
             # Set y-axis labels
             if set_axis_labels:
-                ax.set_yticks(range(len(self.columns)))  # Set ticks at positions corresponding to the number of columns
-                ax.set_yticklabels(self.columns)  # Set y-axis labels to the column names
-
-            # makes the frames invisible
-            # for ax in axes:
-            #     ax.axis('off')
+                single_ax.set_yticks(
+                    range(len(self.columns))
+                )  # Set ticks at positions corresponding to the number of columns
+                single_ax.set_yticklabels(self.columns)  # Set y-axis labels to the column names
 
         # Add legend
         # These list of lists is needed to reverse the order of the legend labels,
@@ -272,48 +271,123 @@ def plot_cohort_change(
 
         plt.legend(handles=legend_labels, **tot_legend_kwargs)
 
-        if return_figure:
-            return axes
+        if show:
+            plt.tight_layout()
+            plt.show()
+            return None
 
-        # else:
-        #     plt.tight_layout()
-        #     plt.show()
+        # to be able to e.g. save the figure, the fig object helps a lot
+        # if users have passed an ax, they likely have the one belonging to ax.
+        # else, give the one belonging to the created axes
+        else:
+            if ax is None:
+                return fig, axes
+            else:
+                return axes
 
-    def plot_flowchart(self, return_figure: bool = True):
+    def plot_flowchart(
+        self,
+        title: str = None,
+        arrow_size: float = 0.7,
+        show: bool = True,
+        ax=None,
+        bbox_kwargs: dict = None,
+        arrowprops_kwargs: dict = None,
+    ) -> None | list[Axes] | tuple[Figure, list[Axes]]:
         """Flowchart over the tracked steps.
 
         Create a simple flowchart of data preparation steps tracked with `CohortTracker`.
 
         Args:
-            return_figure: If `True`, the plot will be returned as a :class:`~graphviz.Digraph`.
+            arrow_size: The size of the arrows in the plot. Default is 0.7.
+            show: If `True`, the plot will be displayed. If `False`, plotting handels are returned.
+            ax: If `None`, a new figure and axes will be created. If an axes object is provided, the plot will be added to it.
+            bbox_kwargs: Additional keyword arguments for the node boxes.
+            arrowprops_kwargs: Additional keyword arguments for the arrows.
 
         Returns:
-            If `return_figure` a :class:`~graphviz.Digraph`.
+            If `show=True`, returns `None`. Else, if no ax is passed, returns a tuple  (:class:`~matplotlib.figure.Figure`, :class:`list`(:class:`~matplotlib.axes.Axes`), else a :class:`list`(:class:`~matplotlib.axes.Axes`).
 
         Examples:
-
                 >>> import ehrapy as ep
                 >>> adata = ep.dt.diabetes_130(columns_obs_only=["gender", "race", "weight", "age"])
                 >>> cohort_tracker = ep.tl.CohortTracker(adata)
-                >>> cohort_tracker(adata, label="original")
+                >>> cohort_tracker(adata, label="Initial Cohort")
                 >>> adata = adata[:1000]
-                >>> cohort_tracker(adata, label="filtered cohort", operations_done="filtered to first 1000 entries")
-                >>> cohort_tracker.plot_flowchart()
+                >>> cohort_tracker(adata, label="Reduced Cohort", operations_done="filtered to first 1000 entries")
+                >>> adata = adata[:500]
+                >>> cohort_tracker(adata, label="Further reduced Cohort", operations_done="filtered to first 500 entries")
+                >>> cohort_tracker.plot_flowchart(title="Flowchart of Data Processing", show=True)
 
-            .. image:: /_static/docstring_previews/flowchart.png
 
+            .. image:: /_static/docstring_previews/flowchart.png
         """
 
-        # Create Digraph object
-        dot = graphviz.Digraph()
-
-        # Define nodes (edgy nodes)
-        for i, text in enumerate(self._tracked_text):
-            dot.node(name=str(i), label=text, style="filled", shape="box")
-
-        for i, op in enumerate(self._tracked_operations[1:]):
-            dot.edge(str(i), str(i + 1), label=op, labeldistance="2.5")
-
-        # Think that to be shown, the plot can a) be rendered (as above) or be "printed" by the notebook
-        if return_figure:
-            return dot
+        # Create figure and axes
+        if ax is None:
+            fig, axes = plt.subplots()
+        else:
+            axes = ax
+        axes.set_aspect("equal")
+
+        if title is not None:
+            axes.set_title(title)
+        # Define positions for the nodes
+        # heuristic to avoid oversized gaps
+        max_pos = min(0.3 * self.tracked_steps, 1)
+        y_positions = np.linspace(max_pos, 0, self.tracked_steps)
+
+        # Define node labels
+        node_labels = self._tracked_text
+
+        # Draw nodes
+        tot_bbox_kwargs = {"boxstyle": "round,pad=0.3", "fc": "lightblue", "alpha": 0.5}
+        if bbox_kwargs is not None:
+            tot_bbox_kwargs.update(bbox_kwargs)
+        for _, (y, label) in enumerate(zip(y_positions, node_labels)):
+            axes.annotate(
+                label,
+                xy=(0, y),
+                xytext=(0, y),
+                ha="center",
+                va="center",
+                bbox=tot_bbox_kwargs,
+            )
+
+        # Draw operation text
+        for i in range(len(self._tracked_operations) - 1):
+            axes.annotate(
+                self._tracked_operations[i + 1],
+                xy=(0, (y_positions[i] + y_positions[i + 1]) / 2),
+                xytext=(0.01, (y_positions[i] + y_positions[i + 1]) / 2),
+            )
+
+        # Draw arrows
+        tot_arrowprops_kwargs = {"arrowstyle": "->", "connectionstyle": "arc3", "color": "gray"}
+        if arrowprops_kwargs is not None:
+            tot_arrowprops_kwargs.update(arrowprops_kwargs)
+        for i in range(len(self._tracked_operations) - 1):
+            arrow_length = (
+                y_positions[i] - y_positions[i + 1] - (y_positions[i] - y_positions[i + 1]) * (1 - arrow_size)
+            )
+            axes.annotate(
+                "",
+                xy=(0, (y_positions[i] + y_positions[i + 1]) / 2 - arrow_length / 2),
+                xytext=(0, (y_positions[i] + y_positions[i + 1]) / 2 + arrow_length / 2),
+                arrowprops=tot_arrowprops_kwargs,
+            )
+
+        # Set the limits of the axes to center the plot
+        axes.set_xlim(-0.5, 0.5)
+        axes.set_ylim(0, 1.1)
+
+        axes.set_axis_off()
+        # Show or return plotting handles
+        if show:
+            plt.show()
+            return None
+        else:
+            if ax is None:
+                return fig, axes
+            else:
+                return axes
diff --git a/tests/conftest.py b/tests/conftest.py
index a6cde99b..2c56f293 100644
--- a/tests/conftest.py
+++ b/tests/conftest.py
@@ -1,8 +1,39 @@
+import os
 from pathlib import Path
 
 import pytest
+from matplotlib import pyplot as plt
+from matplotlib.figure import Figure
+from matplotlib.testing.compare import compare_images
 
 
 @pytest.fixture
 def root_dir():
     return Path(__file__).resolve().parent
+
+
+# simplified from https://github.com/scverse/scanpy/blob/main/scanpy/tests/conftest.py
+@pytest.fixture
+def check_same_image(tmp_path):
+    def check_same_image(
+        fig: Figure,
+        base_path: Path | os.PathLike,
+        *,
+        tol: float,
+    ) -> None:
+        expected = Path(base_path).parent / (Path(base_path).name + "_expected.png")
+        if not Path(expected).is_file():
+            raise OSError(f"No expected output found at {expected}.")
+        actual = tmp_path / "actual.png"
+
+        fig.tight_layout()
+        fig.savefig(actual, dpi=80)
+
+        result = compare_images(expected, actual, tol=tol, in_decorator=True)
+
+        if result is None:
+            return None
+
+        raise AssertionError(result)
+
+    return check_same_image
diff --git a/tests/tools/_images/cohorttracker_adata_mini_flowchart_expected.png b/tests/tools/_images/cohorttracker_adata_mini_flowchart_expected.png
new file mode 100644
index 0000000000000000000000000000000000000000..781c157ef9a177f7e8cd892e2b49787b1b3ec578
GIT binary patch
literal 7766
zcmeHMXH-+$w%(wK<p2sIiV7AG=^h278c-1fN(m)O6%;g~6EM^Oa*%@{XcQZv3PMN#
zk!Gki1W|e*3ML>e^w0x@khgN~eeaDo?i=I1aetpNlAS%)-g{-PIluX>IlmomX>KC2
zU3xnNK_c*r=MfMj2u=l}wr&AGeTRDM!H-U$@s&VpUynenYk)g+*)`D5%Qw&q<#y<{
zdjJOI>vKv?TkWLEp<980ei&VKb?^VWLCrVdrurey#wRezHouEkF%a~dIsYpViA_8T
zK?gs=&l}kU6Q=1Q;WkT@#yQ{RYvP9P#3O=l_8$6r@z9P3JB{}ErpPxqZ8j<k@Gti3
zPB?C(B4ATMtw8J$yq982B5&Dz=lkH6Ch3b;G%sJ3_`?w2&)uX)yV!o{@T1vqM&x_K
zDW?9Z8jXzg(Qk<vnAU9&^lXqdDF8u+BftRY#Gl~s?ht>?F<Zgmw<CWK{_T*z!SK&S
zINna%1ljGDlJaa!I@ntqR(bE<y>J8b6!E=#^Y4o6pC}*|Aq)+piW>zk&0<=0T*L$%
z$C6}WMfG)ack4FTM%}*OFGd>~)yWV|6?H=plxpn68(pc%nd%wDty}rUCW30gWC%(*
zbn29K+>V59>_F}I--cn;Z+1&cj>lj<l)$)CRp8u`_B)|uPbkWCwtG|+J&yV)#*^Iz
zOn8#6wT52U<B}}+>>clZl&PiPV*ZCCM@;EG<(D7diHVUEw?C&~v0Le+?uE9x**9w2
zcj&PerkWX9QWx%1KL@$w+f65o$PN^zuF(tYxNE#IZdFBmCo8Xx_@v`^X5hx<Nt&Xb
zCuxIZ(T}K%Gd`a`FxG+5)ry*T6@v2Ra5!8ZU2CIU@kOb=^#YQ6PRJzb7USDKW<*lW
zPsgR|tZLoBne2v)BAT9TQF{hY3)#?+f^l`TvU;fHd!<+ib2Fo&kj&irZq*2aY~Ov6
zPzap*{(Cps$pK#$G=!{LcuJ>MV&3>nwVOv09?a0OKeB};x)k9k;z6pij?umyjHiJU
z!km{eYm-lWh!bh}{P^n^QbLL<oq45$;(e2^aGPo>y9**peD<BVh%bYwTZoCgJy&Zj
z-VL6POkE2la?~SuPIO#^R8T^KnkKFFV4vm~(xA39V&#2*ZJ#}@kcj+zCNH)6c4w1`
zOuO>+j4qg6p^ZIS*npE7*ET}jG<#}J92z)Rzfa%raF%A7FI%nr-cr6vu?8Fwqg+g#
z#4xP=F8-M$tM^K8gG9pX<|<Wq*Wh^Tx9%U{;q|A*Z4ij;JL^aBISO%%yfE2~`@KP9
z`Ptb=L@3;2sxM$jdf1X;Q0sjcTviU%vrW6}nSu;*=v9lRfAeqCZC|@uXt`q(Jy!R;
z_p8I3Vc}PN)mAKp*(_x8?UD%$C#@jz$A~nJXJA@go8dD9v(&4+IbJbYo$nr)sOB4-
zEB6I<$S~4j{-GRd#^KcCk^xre>itTkbsZtgGn>JxhFns!0S=UulsZe34RWL|7!<m`
z{5-Z&G9+y<W#!Fy?Tp*lSo2`X@J~y;hf<#Xgtx&o9qB$(h0DaWZh*Re8G`}~J6sMo
zB!qE2GOWz-$JM^p&{xSwl`cIq<VCz%p#ze}oxc#V-Vxvnt<NJ*!|`f$iwLKV{F3FV
zpTyZew|d*Nb2}L>N8o5`+4EjGx$!PdQmw#)+hO&X&?3l2MG(pt6x_6_d+BbvG;eCF
z$bQWW5*cB3<?0lh(vx<ICr()MqaNFsY=*y^jB~!4dV2hkt@7gVrurU{`l_pyjEH2p
z@xoYHrC|lexqhXdm+HJ#i*ee>ff?cAuflQbytz$+f~~>R@5Ef%iW3v|3p|#k0l0or
zAlc=^OGoiJALB&FrJkO)yml?~O=$&cY2KZ|z3tUk;PFs3sL_Bu-XjoGHq#ZUq@wRQ
zchFj6m8~Ak=Fm>NOr}PJS?INSW5s%UlaZ*I)UMjfgF4|&9Q#N_bNLHR?*h5;clEr&
zTvmPFGHf0nEf*VWlJ9<rJX#kt@k|fvMZ1|1PJWvs>OLbYvny-;s7doe8KiA-MF0vP
zgV7+sTI2J|(i7uz>To=cJ2teI!K*_q^y92W_Mf&o8*q<(hVLj4k=?s@liQ^=7Z=}3
z$-}R^q>&f%GAxUSH?OX)5<eSDp@vuD*BVAdOw&qd)NxrTmTKpYml+)-iO>8sMMoP|
zc@JEJcS>CE)C~Q86y)J?CSfLQ`1HMaPxhl^_bJjN&51da#Zi|H71%0@nov9j%z8rc
zbu(4Cy!w&VooCHY1&#UYMv=_<S*v%3@_QehG2ce_ZFd(bnq4|>+e-^$k3BMuq&@m5
z!(KWLxbW$}R4JL#eKoGNGxdcm3)U5LSyku!)T*4OuFRzLNOUmKyNkD85Gt>bk2TF+
z3pDKvV|`onai|N9zHnykT5OBJ!1|CAWT?5^>6_6p5GO-=J*0{5wRx#v-kFVMPV~u!
zD_WE`J$bRQ;bu%F=N7YPtomCIO7>|vBef`|M0yb|r=-o3xvT=7B7Yba@Y{GT2-#S3
zmZ$7`p0wOEL#xAkt3C6?Q?;mk61nVzpuTh>+o@a)v}^SGSw=bXGW+X#@8GA$Heh;r
zcoq4*WPx=tj@Y(uj6q8FXbB!VAhRo~`O|jD@^Yb5t*KSIR<6%iSilukdt=ntY+l*?
zNBK3n^~aRX@@l#9lCUj=xtGxjEx{ZzBYS1e<ZIw?dlrJpF1wFh>$2ELG!c9!wH3;I
z(M+$jc-m9wFj4WK-7=^)D^KoAmzr}m=1pziu}kTQ!%TYBoHCacb4~-0eC`=Wr8zg$
z9XEd{svsVM3WxwQf!VKLQ@!86Kg?(wV6(+6{ryX@4#s5YhW8KhVeC(GhlGNn+V4ba
z#fPET?S)oDUb0&t*~3s2y1;nyxN0cYnZkzAB%`JsD#teZ)+U~BljuHEzJTh(deO5<
z4kl?`I*^Twn&9RFqSiKeGdDCt^q{__YolYS8%4MAJbHrkNaN;bji97>o6jG_<N*8V
zTmqZWSgh62lqbDLV52Bgx=v)*kxo7-=(}I5<h2K4N-2XtOYRPw*bzSZ?b)m7bFzS{
z_oG^t>5OUDU%%FrSW|WW>c7j|?wizlO7Wg2uimWbf6wKV`_rYQbzBFprin$d_~E^d
zWrRf%cg%&mV_QWH3})nogoQbwyP*+&n6}*!++<OGt@y}hVd2T1Y8G*JD7|n#+{rJ5
zppA2UZ)7AHpaQ{?B%w@k4bRT=I(-#up@~ZB3ZAJiG*N?-ze5u11WYUK93)ls^QR-%
zf8Hf<JmqrWh;7U6Prj5~f_@01`t}VJugHrITaoD|A=_w21*G%=8Onu(gsctA@+_BI
z({w}LG|`4^6pf6c(Vh7Z#RrFM*}}3un+Mhp{gptT|B^=keg|*@J=s&(TNOkqZiEJo
zZicSe+`e7$rnEG|;IEMX-D><_ZI&};^qZBVBSBv#`N+|u-4x8tl*f-(Ldi<r)4k<Y
zBTpQ{4Yv68{1X=W;R$<G@p3^dh7*2qiRPlLt81TuTO&$qqUGRgQ>ATzEt<5ZwzjsL
zn2pDvpb*-!Wi9l#%wABqMae0iK8+L*5U3i)o7Gg7qq;tGK1PE7K=;?xy1BXSmX$63
z_U#fJVQ!v;rWF<LI(i}ha;g@ZQA0U->eN!x(d4F~x#LPoru((LW@RwFWhm2&7cEn@
zydUH7+3NM{Z@z^{pY|a)%0@c+`}?DN$j?9D-;YH1&Hnl*38voQa!8q(hMR;%ZES3S
zY1xaw+E6yV8e6=!JWU!KyVdaN)2&Z`?!xD+uZN4s=wF$eo4Xeu-%-nEkR%k$FXC~W
zzWxIDuTSQK^}(l?g2_udsQ1djhhki4UyktdB3A`X9)sK0g;VrpEQn0E;iiWS3T$Oy
z6H8@**@pc3L{)PCe$uaxaU}CJ-R3V}Rw=w2?<oGzCHuF2Sq|XKW;HZ4sHv*9IEBwO
z6k%BmH3I|s^pDa;9j2qJYnJEt*I8IziZZflZE3O^82L0UZCZ{(!7z21gjD^IH#>Ih
z!0g+%uT_WH+12ILUl);vX$7V=J4JGrW>)5gzy&+<m0$MdXxq{7`_h^L+_fc)yu7@c
zx_X<Y*Y|dJ50B7F5owJk003<!0r>e@HFvqs39zg3@#B6n$&D=poi1Pu8eOpelt&`_
zEpLrh5BLIj>-*(2-EMFR4v*ijaqEGZg#}hv^nhPEx;vmB$0@|LQeK)QS6y#RqUc~o
z-}=z2LXt}(c{nBk1N?$1DJcQQMW}hSyci*1eBR#?Dl$&k1BL=H7`psZ$3ZRc-eCe}
z1|SZLcf5W3cE|=LYI`G4PvH3a`g)JH^g_TzB$H&B=5OC@03oELq~xkPR5bPX_j3Y_
zKoezKkBf+yYmma{w6|L>&Cnh2)CzBh*todKUeDqbxrW9@&%O$8yg7Hv`p8okrNf7f
zWRgw9^k<_)4`^osvr}}sGL2$}nan)hpo!zEs+Y~4sxA5KKXEMz8{{_LnJcXy{Cb_s
zmE5!ERrO4jnb{K+E5Lqsh4#6PGGX7s>H#)P@uMxNdHzG6zYNWdz3?pRXiib)+xbG%
zvmj()v}&PMZz&JPg#*7?f-cSN`gM>a$3*s>Jnvi2t+;dNPBTGAL&FAzLg90Aa^$@G
z>zwdq9xc)Ux5rPOMBu6Qk&$38$e_c-Eu({f{bs26@ZrOIiHXz*Ha$<qAasn5?YE^B
zIa&0vI5eR@{`e!9cVOqv%=Y$nM5MZs5?)S+$)Do>zEh(&Tb8+)yK!+(J?l^Tx1yG(
zexN(D8&G{c<f|`%qdiyWM@%nWqB{82<_-_v@c#MM3gFJZDv$`8(@~L3){br4=F(w@
z4jBSKj|AT^9JvniKgYT_g~tqvfa?a&f|9kgy9{OR6tUE;q^Q_|mp7AxlaR$dr~AW)
zBBrZrLYJqA<6U{4z+QX~9xq<#RK1+Yv3P#c7#}$BPM{@C4+%O{MNIu<QMGyG>Q|Rx
zLRZI26HV`4jiAY%%fEj8GKIrC{@5l_!x<eJf#XBjGleDB8$*YN2j0gtH#Z}2xH<r?
z6as<FCmImpwe^Lz@|$fk{Oa>&Te>t1S3|&n-WyAi4RdqL2K7oFK-5M2D~pVo+EHj$
zNRYoIWLL7Zrm3oZ*@MO*tAKzvYiw4rb8V=iw)Pdks~t@bk4j2Nyo{5@T`BcwS?lQk
z@D&gKa&a+mX}rq>z#dJnD)(q5gnZ@#)RNx5)#f8{tUU{mV#IUK{k^}RH8L`~my~38
z_Us+LPf~sBu67a=M1fyOz@)DLN-0R^hjO(mg9!yzFYf{t2V6?cHIo_{9v%bSXKH4)
z;H(imKMY@D^}9^ciji@m2X_M^@+_iU00dxt>Cz>z!rIU!jga{p{LBEr=ie3_{1(u#
z(}u#%e8U=%ohK&Zd^eHNe2N8NO<PhrXZs{0qxjE1x5D#^jDiQ_70LiwWi5nAZ-IoV
z^yRC1J-M<U=Kv5;aJBT6$G2y{CLdngs|-b1*+vLK?^YW$K^3s~JUCq233S0Iqs`!S
z%dWo%|KX6mLj+}-H>BnTN}6x$s|)3>AGfgjPkVjOTwP#t{Ek3%h7j~w1J!vRq+USl
z7oEJs#NRYgou6pe1@dozhb{#MmaVU^KYsc&uck&HpjuH^xA#9UCIZ$(AP{9(mh&{d
z8mXwLIL~HA3PowIBgOVEX4pbFj_}m>#U&8Q3~*}|mZb)R!N6eRJf?;x3_+NSk^g?p
z_OCi$rlE@%lnLw&k60Ve>rGF4DWPDo7D~W~AK0Dq>aw)5nwt5pUAyXJgx|4ObeM$V
zILNOVY^6(iv}0e@$*-g}eX|Azc7j!|hLSTHbn+w=wwGi2?7*%Q*gp_vL$XtKNfsK=
z+YDL8s$6|@ezKIEAC}=_+b6pY1I&?4DV3xTZx5R8?WPSuE28{PyQybdk2*Zjv|}``
zUXD)+U%DUmRvhi4y@Xa|K*4*~oS>b>C(lYrpqhydrPD;<{u3#L%rdH|^y*lQZoo==
zvWzEUZLa~3I=lTj|4dQ_hWk7%)=#$}TB%a2d|FO2w~L6_y1Kf?eiO`e+X)#;o*vBN
z_1slILrWAg(~KNZcv0(>(u=LKixZR4XED|IJk53e$+Ec;JPN&ae`na@{3L|I6tMIU
zmI7;={#oLb(4U%@o8fCy*ehkGTb{$HMar_y6xW}$TJUIn0D@^>hA!MSu6{gwl$bER
zk(Ns0cT9AhAE5QCa}MW<bl0tjOG-&iHgp^<2Mr!RXGQc&XgGB**jpyQWYs?bQ`~lE
zwMR!XiB54h;GIsVmmkzeMmdMwOb5D3C6xeAD=JLP+hi!@2IRUAT7?`0(z)A4$|V%_
zybr17M5g$fcP+A|JQ?>i3;eB~IUnFAVmVc9J?L*=izjNoMNV|;1}(Ny>rL^DkfGw<
z+0~n&H^;Ywg%W|4p+m`!YS<h!T`F{`5WC1uhIKdbl?FzzZ`P<voBnhhENX(^Czb%U
z2bs)36<j4%$o-{;gofP3hBda*h|BVMgNK14cAH74+Q}M9enO%ZVvlRfosW)=p6K(9
zI<s->IU7g!ohp3o$EXx@Yyw^DcF%nhBqzO!$@XaONcBfChQZDQ;)C4MNW&<$Cbx@a
zw}z97jZLFxLIno_IzC7{b=x@kjuCf>i7h!hu?Ss*>`_mwc?BMUDEHfNlpL0Q7>GvI
zbIa2r=ikE95KA-tjLvuo1yG&l{OHOnh?scJlG-Dab?2yw((vyB`O<H<?#v&&V6h&v
zv3R!1XOeMMAUZlgUjClhMCPZ*=B9N;sTJOgoWA1vBC2r|Yy^}cJu?I^I(399gq!Zg
z+SGII68w1r=k89V*VXQ}AId(dE?C2>sOz2Y-U3GyCv8Iq+FD<J&U>$;kA|50Gd6Fv
zIjp|1;ck2^(B|c<mT&{p*j|c(St{&mu8B<eJAN97ssrNwpX_a@SrPIvlpg8Ozg>@E
zk{~tlN=<QIBOlYt-g=3*K;q{ILouE&r?P5Kbya(>4hr#;MN&&RY9Sdb0Cj&itB<%4
zBO6NokVMNzS>I3Bh9@x6NWRH3s^x`OZ<t+<j4U~V?&$>DOujEO322bT%BhaIR3DmW
z=XYi~CagyYD{#~=i;ulm%hNv!$*QT6PpQuMGAzwpUducGFKM+ZC*P#{O5v3^W8|bC
zZ&4WI3&zSPW*7TK#H|sv{lEKNjZwDAbB>58^)1GRv7I9{JU{Pe&i)D69J2!|0(yGF
z_TGG%Vvxhe!;k0=CDU_pu-p_`wGLLgbkXA?x;#9Yeu}R-HtYA~?DL&`vjzE>^HVmm
z?2?J-L8Ty^MPG0(j62*GEC+>0^CgX+5gc=lVfuf|Au~J}#|8htsUaud0yQMpZZl-#
z4an?rY)^icL~LxX&kj-1Ib2aaNW^;y3AKGnZbFAVzj}yJhCKK#^pA(<R!GXI+uXV@
zeUoznTG=W9gd$aY6I4g_LUS#TQUrPRR~fVWdys9Ho?RJbZ<+dDWsIy$(#^v_`TC9k
z4A$P__{W^3GY55pGe0CzJQ@Ok{z0W86;Q2-sEGp-#KhRVGhMt@g*f6s)2`YR=$uZ3
z-f`9P7&~{BB?i{^Ga{(E1R2{RII#X#Y(PO9*!S-Ji&NKMY!QvPC{pB=quGZItBi|{
zn<Ukga<<e1UCN!@=QvvNrF1l74hIxw8hCkdYxs}&n?Yxwx{pBbxK9sZadXu^Eu_$O
z^nj4~boYS+Q<GNEV+kO)NQ&R?FxIuVw4C9_#KZ(m(%FQ@_-2J;Mf7AJn7w7@>jbre
z92E3e_^*3^WK$}9uH=EH)g>9>3SPLB1%1>luc?YHHW8RLz=39D@-RdQSYPq4J>y=m
z5v1+%^mLu6(c!~&GUEh-iVC%kaIgTWcXktpjr_+;mRUX#xS6Nx6TSN?6crRq6wC>W
zOWI61ZS9Nq_n)>~7;TZ<y?c1l9|{-ys{(S|^#84c+Ia#U63<T54ytfL!34pL&CeH|
Iy>{n60MdpiasU7T

literal 0
HcmV?d00001

diff --git a/tests/tools/_images/cohorttracker_adata_mini_step1_expected.png b/tests/tools/_images/cohorttracker_adata_mini_step1_expected.png
new file mode 100644
index 0000000000000000000000000000000000000000..8fb392b824816c1946b64878106bc822d5de37b5
GIT binary patch
literal 15461
zcmd731yogExGlaZK~WJ=kdRbBI;0y3kw&DuySqgckP_(@Bt*KqyA`Cn^U!gq!&}GS
zz4zUB-~GpXW4!S$V;l~9v(MRkul23(n{$412P?=)-n&C|2ZEq`Qm@67AP5!wiQ<QK
z3%pH}Pc?xz9w%`%r?<8yPOkb6#*nPOlbw~Vla;vvxr?!bqq(gOI}<k(8w0tSlarkz
zFEg|Cza7A2>tM=Eeu-!YUvk^-wYnn&-IYVWP<&lOsUV1NSV~Mp*)171>*}U#j6fWm
zl^^T;Krs-F`g)iAokD;g?dN>|3E}ob=kd~@>im4B&xM89^o6CDcpMvde&lDo7iS<B
zrw|hfN<V{8(3Pa*v#$_XMq%jYQ8N28Mcaw=Jv>AF>yGy+9ETiJdAe3y>Slw$PkMZV
zwjX%^BZL9r;^LwyK@V_oS;_x*Hu9>l_wjXgbu+$yZ_xe}6r@ig<db}O=wKU1K|{z1
ziAhRE#l)<_*u=!{Qc+WPPCxIK-*|*PH_8hL{C@HU_3s9u%YE?vlEkK}?t)^y%ZN>%
zGwtTm9eG6uL~(r`C5BzyLqp>sH{!*UR^?Wc56Q^V9DTx(A9fo+Twk&#4YYT58k9}e
zYY#r}fw@FQMYX+0!!(<%J>e0OM-dhit8ZO(P*hQoNaD0E);Mu+Vx^>_>KPd5SanD_
z+Jv6kgA;vz`wBrFmAmGUgI3kIc>qi3$CruKUGJOAHFOiqQS?cX6crO=lCsJ9U`TdD
zC^@Geab|tbR^oe7PAUgFIQ`eJU;SroOc7V-GD)0at*xT|{{CiT`Gm~%`=}{=uAN3~
zk#TX7YHDf)L)IIQAThDb7Q4D+?!)_rw=F;SSRJ@E*~A?sOQOvATw3dR9f&I_DQ)~p
zM}y{lZv+By$g&;R2TAh=H2?XeIab@r+iaP|lX2Z5p99Q|v4SK5`a=e7tk+^<E2eaG
zblh(H<PDx@r!Fs?C6EKq{=r8uw<P`VT%Qw?k&zh~7^u~`zzC<2lN1&8)==HWg)@wZ
z<LCAve`~QLbm-t%<Nn$g%%nr&Y@I9ad}&Qp--BF=d_~7O!~&jeoMKU``x+s&OuVp`
z771*+3R7>nGYmQZEgvu)VHM+SmSYTv)2b$v{oT5~hXe$YBiXVI>Zd&HR*+pyay`QL
z8iY)@`7QqX_k<r%!QZd%MKB5|7fl}>J3~yFO`dxV1irq$yY)wMpFe*t)+lK2>N;8q
zVMg8Wr7vRfxpI5>@F7dmz)SPt034yKU*Da_pSRrt5o7KqV}d=lZ{qCo)X(wUOy9)B
z<mAkC#vT*ABnx0*;uF|S!*hD`T<<`Xl*4-Q3@Svk{O?pz^D%iJf*s$$js}HsH19je
zVe><dx7gj+f5-4&H_8~m?1|X((b@4mCEXi0DlsU%aI2q}jzLD`%O^IH7?(Fb+9}$Y
z-$s|yQvOB<n9bJoSAigYp?tLxe-yM^`<L@Ksf%ra?BG_0M@CR+>FAbnb5acsmO9XH
z-7?@zBILFsXKsX39<2>fJ%9c&Iy!m_p<1Y=zI;BH0g+Qu-upDY*&ONmL|P9$xlP=2
zFrhD)YLCSxXE|8{MWfQHCqq1>dNo!PdMzin_L;tryFsTaiP)z4J*V|lvH1w~)%A65
ze*QBao&>O9hemZVUt)qACL%4%+G-Ub^+>jjuU=2}j2H0agfK8LGPAN+jk>>#Ofl*<
zCWCl58uCTpf$4#)IEirt;p8GBA_Tr9TYndH_mGzRYK)PIi35CJf6n*DTOyn!xWmCC
z2#&bi+Z-z(78Hz|n?Xe`??{bMQh`L*IrqTHGE;V&nQH8<HY2u_@xzNu<Si<Nuo`ym
zZ{-y&t*=XhwQFHLQlQDjX*D4%E-qfM{r&s*o}QlE5(YlLB#AI0_xib`tbaacGLU!x
zV&!mHnmMIpzz+qjdZ)C{?fAFk3wnCA=0&^xPGVaHWo5CURDsD7L&=lS8!!l8qoN4;
z-R!(~TA7D)z_e#B+Vj|ZRySeW4+)%IEqO%swCAMx^rj(>X?R_Bf|{CynBqdfq#`d)
zLE*v8`H(MscF#vrO6oTF2gR#T`y1aQbQA<q|K7{JPdx@5M05Frg=lt6u7v&l{rS%4
z>_hx-udn5W+Q>*q`tyn!d%;U%`{?3qPu{{}>pWV6`JeU9qV8>W0ynLFs3)KR(sksm
z4x^Lb*v<mKo(ca?DSVg&xe63Wm}umxB!$qi{5p(Uw|a>0YA^ju)~QaGncT++l)l41
z=UV2ckP7pF@qF|H87q;~y7!0TcMveAZjATdqx$`!bJ%xdc(2b9yPD%I+z|<bS8R#u
zP5Z=Ozp}$$S1doFrwg|dGSo$}zFZ_5e_AFFh2wS6P95{a@L+Da;jE0)u2t<B-m_FP
zoGgh*%`h4#@_$%JQ$F-m?ZfxnejypePop<3g@T5<=?#0&s<&m;P(^dO8zW|nZ!-BU
z-)isqCx0A)t=UPT$fAiB$Md_!PHno_S7$;w)82B-UPtRgIhs{AURMWQSIcBKX8W$E
zbq}05G0EB4qsDc8ual>Rj{4b0;}p^1M`<_x_3lSo883Mx!6m+c?|$D3q}qE>*{u@b
z)NniQdva^&P>U6Hc~^m&KcywCt;m>C<SFJ7Lw~l}2m!dP+c0lT>Lg!*y|60{Wg}Au
zeq)Q1?^oUTy)>Ua<T$>qwc4@5loZ@!_56ZsNH5Gx(UR<aO40NB8*k+`67sV=U;TU5
zGRu+7dZ*Zh>kX0YCGq`g1F4>1k%7TB8P1f_avpuxpC;54emY!7Bm=pfPMTVx`{lmP
zaad1r9TPWq2nVN7CYhBwPalXTn+na#e{cgQg&x^zim0!7ym84gU`sk^%J(erhIhTI
zV4gYV$+{d<OsqjE8^CFb4L+$PJKSEP8_Y41C+e71QK+}Z7|gH|t-Xgyev{-%I|J9f
z)0h&`V?~j<En9JF|I>R{^(ND<P4v!)P{$5|n8Lc{wmEYq#dNJRJthw6%9Lf%>YV4^
z^=K;U{%eSqk<s|i4@IYqU+>%7#ctt{_J6<;><)j*S3C{7OU!xKFBp$zxWUtP$AwQ-
z+xan{%MP^Af=Pz{O}EKs0_Fk^?3^eu1Svxm0!h$drrO?V>rb#;3h%*A*^r8Do$n1o
zCYd|Z=lTMdj7)bnPK#etnLfidS?0Jn2Wyk(PLw<p+x*k@qA+TGUF-QdWJ-bpclTUZ
zp9)D1v+YCI5NHfgk@pfg9vCt>#p%GuMe|#3FZo%7A&F9)s)cZ@+&{8p?);jZ+#k<s
zSggjN<$HayGhG#7JzeQu#}IMUP2x8sbd!8_<yC1jdq>#DhKZ1nFqBNFvGe3Bjcj83
z@#Y8$eBlEQ{CdxKXg$q$aJ#5!Fkih4zLsQ*g^kTua0BvaILI$-Y;0htvTWxY%WdaV
z6~9ZM6q^o_s^qJQ7#lx-%Il;Do2oe8ot?(3Gf~UB<XU1FY=Qk~7b@iQ8DPScG-hNR
zI!hg!Yr?G5s=dQ5*cKW0F)t(Zi%r{YE7!-ny@w-HDSHKD#LWehS?_=3=Nk<s^GsNA
zf@u<f-n@AO;;ayiV8wnnmau^V<#eU>vTM`jQ?Q1*Ov*;5Y&M6pz`2vasIxk6{UKUh
zh-1*K487cIA}gDe(bFS0><rZh$Bb@Eh7o-nwdk0iC##zsOcn5y+g@Ic)1{`S&c0Z*
z)0s#~56v*@qU$Z=e0#8b|8rFV)J!z>#MFM&yI=rCpQh;(>H@=Hfk#HqT@(5ooekf)
zxx=GV%Jp=;fx1Vw+H#w_%7=}q_x$9)bJIDNLB*e5-fDI@%%j><7{wdFm%-N@-J?-R
zaQ$>z6S!q6nc1RIZiZ`bZ%@MO6x81@!^X}oB`s|T?p#@A?38+Acr_AwKtRwnR-j49
zW`t!qmY@DMM{Z?xb=1spp&5nL2d=+1n9{d4kaP|v&?5Orh%7s;!&ykEX={%WANzjS
zCRb!Ui$1Ely80+io_HFD0G{+~!`6$?PN7BupMqNDG=hR|9^uV7X9!(qj;v|I2EBaq
z!}Mx~lkVu@0#lor)5u$RhYQ*d8?u>mjVtEua9=A`t(ev5r>~1{o<>P0GV+FZR$YgE
z*`>mIQym_Z83@t#P01{N`UnM_)FqK$h6S9Ge4Jj**CBPD71Ow0l=g?cLZQ8Cv~X{4
zp2wXzp$6ObF$8bR&#{?z68>~<x-8nc?=Cvk4Vu@1(wH+KRCjgQXFgZYa|5Dt+d#FU
zlPU))EiJ{u!7-gE*1t>077!i%2!Ns&wD9?LZtOdE07Q9dJ@u$+dZ0gn-SQZhghU1u
z_ktV<09+)L0M$4<Gc$n4agD`%_?dh(Q-xg|!Lkbvmf17`NH(hOxX<gxmCBuYxaoOH
z4w}#_VbXR>C)^>9JSDrYl0QyXoW64*e_G~?g^1G@QD}WYf+6{0>KJ49^RUMpea)Ym
zH-`XNVZogFOq$4(J8B+zol%_UeYNN7?LqA91)SEVbhku+2#8;P@pcGqKawGvW_UZ8
zy1Fk+tlTT;v*Bg;^_Fh+8~Hy+Iz2<`GSrC}0b`TS5Q8bte3?<tSB*H^iQ<;!6SpKu
z)aGp2B)7vJ8vD9gZj$>F5u~zdg5DbwB~&TA&Xi0{;o>3q_S4o?zX3o114Jn3<rXGL
z>U(8>>mHGqurQ?Sb8Z~wEH8BZGD7fz7C`)4dMn+LGalOoskGV_i*(Ex<gjb@Lgv1!
z)YASvOKsI^ijS@LUU0<5?7$PhKGpRs6=NkgwP)%Q^w<jB4jCqR8{Jqv<~*6!D0Z5|
zJgTNrY;qsNa?Pz=qQ*_7?4xO&P98nV7+Tb{Hm2nH!-c)|KEzqtu?rz(4&nZOAA?VI
z5)z;<mSlB!m0s0_YyrmgDL+4dD=kHihR=036XY{BCymn#t?rUP1AhNAs{IGzI&lF+
zVXi=@Ek{!5csYYOFHRhoIDX-0@ef*7u1%h)aUc!cTHu8S&_ol)ZwAXg`Atlyd3k#u
zoSo&=*C#V+*Y1PV_wdP+jf2E`Ddg7@-3{#PNdJnOagAD%y3Xk3TcYdwuIYZ<Hh`*Y
zLunDSHeIU@_RC>hhl6~3`@de{xaX&S68`&1lDwYKj-3{B3j#NZ^o$JuuU{XwwzUbU
zN~5Jq`B73*3b9-Kp#_yiHt}VP73U3IkzgJY0v@u1%tj+1kg~I5WngIN8*`SXh~&9E
zXnCW(j8!wOAW`qal=$V4Cs`q20I1dlHsxM-H(~?{sm$$2Y{oI<XZQ?#*Y!bdgI``Q
z+i{zX{(M~2gs5HrY)<|UmI8zG5(lfGY#UM=T8zR-NJt=^cX{<McyGW;U+0F;`FXR)
zx4(G(^MKX=2WIO{xeo?D;>?m=bH8H4c<<5E+u&MQ!0tk=D<YxT)P1*k*{fG-j%`%y
zk5EbOWBQ)=>AKxqpOt77s1>Mq9B;JDBTl(mTU+^clDKH0prF?u6gM3X4L5I#S~bSY
zP)D3>yVRYR9HYlDt@B{`0ge?nGov+Ds4ea0#%)uT1R^;I?Dz}KYYo^E0O;b99dIX&
zdpy(N$wlS^r@*{V@}u!yagUJ!2zJuE9E!gXG%^h&6Ey3oatb=S-s!AAFaA*z*mhn(
zWs|KR%2Yh+?naKpUw?47GB5d89%&n=U|Z_~>n+r->+b2HU}GDeUj0MZcFS)+2ul#*
zl?9FiNH^Pgdu+y`k8uWEZ2i$7U%<zYi{KvM%MqX)VT0U1$Ml=Z2}#z7NUEm8Mqq~%
zZ7`R#@l(^GR9R5z1UyeO4ZFgT=!CiH>MM{Pc&C;xOeb)WDA~=*1RD(K)($VU?h!?i
zG|*W#B-jUD+XuWvcaS$>&f>+WGB&dxREVyI<Ma$k{0=JAPDHhF{_>^&SAxjX+pNBL
z9?mEjU*MCvclPivIl1?#PqDM5%D_Hm)uZ0FOc$MTw=aL>fnL@0ITA43zmMo}xxPFz
zy@DUyB@?_VcYg8W1(M|IG<b|E#=QIIPVbwFr_{sI-d&G%bhpZ9=H#TEblg}0lKEO%
z`ba_CG#eRk4#UK@^(|;P{7^F}fUF5JUgxWEkW31148WD8mg7vH?mvBVwmU2J`nBF#
zf5PYxSAxvS`a0T|#Kh0DHaS3W2qoqUZ*Lb<0jzV@<z#F0czZko4tFE<+82Rh8Pr<=
z#f;~)CUU7ushac18$5l7q)lRS0B$T5`CijQ(s4{#u6uJxBmq=`GN(-?#Km}%!zHYW
z-E~)Ms>)U#aHR&1lMFCdPJ43=wP~L}-vLanqi(OUk7Hb`?ek-LH*mD2l_g+mNaEbv
z`-azPqj{>rvIcRnO~`5W1f2X}xodp0(E=U0lr!JDUQE}yGWW$YE_V><8l4}mK6?DP
z#CcoA|HFsLQe*6@Sr=mP8A0z01vRy9Fff+KW!{%A`6~3|T8{T1<gbH+1CFUsB91$G
zqn`oG14n}(`ZFY#nu!dW$h6ns{y>Zvju&ccd7q3{&)VQKs0oAXFHr-^fY_JRRG+!)
zM44<Nq094YTegf2gO)`*EvwCSaViLzvLd6S+sn;IHhzD{gwSr?)$5IZ@$C8Y--X(I
zqN1V}ch|u;CC$tv;t@<&SV{tHn5R;JsyF)c-OgOYWpX9i!-t|GBJTrF*ZTj86FvqD
z&F@LL?ffz`pFyBDwF5;AsU4`~#dL%a936C#MSb}q+!g+`=4{UM1tVj@@4%uWW-_mR
zlz*vPYp|;MZ0$N1u(DE|FTYDhtO?zm<7n49<&H@CQCCgF3d`n*i{s;;0NhajSLVCA
z+AAE{G+G}_DYu@+59vVFx)?`+=YqVm6@gx;Srx9|_NfG5TU1n3{>Fl-@YPAPzY5<I
zlRPpI?pN=$Zrio^V*tR72A1rog)u;E&WB6B>3_4T2&?Axf~sB59MJvPTz1rK<Lo68
z(Lg{O2qTLS<*mQ@_1`v-WnlPQ^gG^MyITIaqKJ;rG}w$Fr7LwwNl9jACRkVkZ=(r-
z9J#W#`rCKnWRB?-Q}eENojESSHjHfrb!R?FYD9$<xm#F9nF@dK`rBaDTb0dcYc*Lb
zvC1CxdqEJfx;#$p#)ExJ{RwEniKk|5a^AD<ZVf~%IV0H@><a2e4U|Oc_W8$6-Qcuw
zH%52jiF$)oPV45+C<0%<nXhl?k_#R0K-dIKD-pg|RDyznNaVG$Vl-BytNUH{260U&
z<ip<<h=Zxu9)vq%*LE-TDPOF=KPnP$3=a<jTH>@l76wF;x5~=x-y;NjUFW?qo4X^a
z64;FIK!Bg{1DU98e7vgmlS+X`iOno8HJ~Pdc-ir$$fWWoL{drW93QY)A!}gn02tUO
zTs9;~JnVgWT7P{yy|S{BtV*`r711?6pB5T=x8H=lD}oGjp=LexJE-Yk&l_p!&FLy;
zp!AvUFSN*{38qj<hA-^xA(ya8SpuL(q;3l+AS;Lhp|59fFxzsx5KyB)0Iqfa_*gvr
z78KMi4`ifB5R8|nFqG=c-8!()c!~t^nRWWXrLrdF>6idX2Su&YitudSR|tG-bF*+~
zC}G=hmh|o0x6@^lIA@&4wfhq}LO_way1IHTD{BlEi<axG^91a4>k@BgqGUPa?CsxW
zM$l|gvE;;?=gUM-&$lNiH%mETQ1T@<R>H!9j$XAeQ?*E^{swUgRxeO<URzlF?Aj9i
zmj(JePn8+4wV(6GLn(Z5pFfZ6$^NQ!wrspU)r3xA(_LGmd3odah&e+yM{<@O-gzf3
z{&|61en3G2VDb?G0lVoS2_V8U`uY@-5u{mw=<;75j~v6M<HCr!hI8b}kl#{j-1qZR
zPF}vKBK05L>JSeIuuBD1a|ZDJMZk=KQYWUM@XNgF@Vbo*(F~H|#!L;nyx{pWkdN|p
zxC}e*_V)E{f_)%G^+eI2?H`ZG8<tHXj{-*+ftrkjqmLdvvO5~!8mVzK1)3?KhATjk
zrRL{P2G}*4K*tTq+L1|>>;~SorDxCjPl>rp7LpJIX{LDJ{nh1Qv*#K0^(%b9)(1`a
z&6_(R1kBV7452{7+TsKiIA61>aBMv=AmB@Kve(q%w{PSoL#f_VhahEu*tLM=flvSl
z>neFmknA5A37=#z5)kO#CP4K7QgOVHub^BG$WjkLu&f_=)XV;bevgEbI<^V66ZqX7
zPKvLKV%7kj`L~|I-v)we^2OcyY1pa5^SH*hUHgAn1ql8o3LSmUq=TcQ<zMNqUfRq&
zNg7y#eFf5eRZ?;?i4~{jNet%D$G_bJS&&+}U3Q)T;6`-28jCY2ES6C-b=(n|huST{
zA^@;5-F~?f1j$D~Y(2Jw!R?fr7sz{9a;A>jtRK|v>-Y;8|5w{6+E^P8S+kv86XBT4
zhLs9CKlD6X6dU==KdhnX&rFn}r3)g!IUV|>;cFW@m1|@L6>0bN!w4tx87Vk*lWn-V
zdOxfAxDh0X*PPT#b{?2s>S<tnTsAFHQc&!y-TBZz*^tUfp*S;N$ffA>zS{oCP>LdL
zXI^aT;JV^;gueE1RhR4YRvIX9T1v$pg=}ZGIhOrOv#-N5Z;y5=kh68(f~u-Mpy>YI
z(q5A4*9UXaLau}|csCE42M`%vb2BBeS%I!&Q$q`DFt~SMp4keSSXN(8YAcD(>><Z?
zDbCp5{ot#BE(s_jBcr>&f8}Dk$P^Ty)GZY`UJVa_<3QI#^D&iP?fV{LRT-8>cbTd}
zWNReh#|tWBB(suKH{>?KvdvvqxfNu9H+<S*KzQ2G6FZce;XdgMnh5ZiBR1?|=rx{8
zz4vVIDwfUk`747d_HXLMBsK(j5xc7eDdr3_O$qq#_4%jL$q=zqwn0jZ4u_+QDfKmm
zjC~P1+16$+oA%t^0X45=hnBlDlrax3eWCRE>EjvP`Iscl=L(xH2b(R{SF3?p;0)no
z6pHnKt|d?l1S?M8?mF!&uHnec!k?4v&5Id`v1gxDm!zIaWKu^&Kn&`o?IYTE_B+y_
z!qu(d>iJGsmZIgl4{XTuGX2_~)9jaN1Rvj9c{6StFT2LBqvW_Ayu9<Gzh&GYt6x!S
zPom_cmQ3P!clQKc!yBvo;um}E;{mj`=6iK=MrBnjflwaK0IFzp&+P4xp8kZwPqh>@
zA+qj)2UT!|l^+YA@Skou3x%3ZWj_v>xWcMDXZzvKx8zTS!bMmUIGK1x)xxQ(fgx^Z
zJ58!4W72O5@6tS3CLJT~d)B?x7u?I%S>Bl8<0_VkPe20UGitPFN<{%N02HcB4|gjo
z27p6BPD=GfxPw)Sgb>JbmiaDE1uXw?sVHjurS{v&ax_si@+N>907i|3mvVCVv$C?D
zQBxoOLkYV9*nh(3!T|L9YAq(4+D&TY7t+zwyI*YPs5g4?JbwHbV2W7vG80@<QlA6K
zocHdTuXIu68r!<N1^PcFGeAqPGODQ>>7mDRm}xsh-)1(s4TZF+467SER-tHaIylQw
zAa9|T&L&Ut12Q3~@!XzS_vs{tFqkRo=XcS+^{e;v7v#OLS57TSLy?_kDYTY{<dmaX
z4g1EOwQ8<he>DVL-hztZ%_{gtCeW`Odj@vTc~s%t@BTF7zHju6&j~MUcsSqp1WQ3H
zF>(9ws*tT}he%x)ksJuAWD1<kwE1ITcXxMx1H=WC)jwuYadBOs5>5h!w7I#NnVmgc
zZEtvZbmYG99y6ZHmUMf($P{oHegu3O=vY`G9j$=sbX}ZmPu9D$r>CdapH5jq{mDF0
z$nFi`VDCX4WYup)hk$+|0i?Nk@3W`ii=B375^_>KwkUC=+g|$G@{;XJuPfN2q_ige
zVgkTbz2h926^6<xQNH}iXY`3>H?4zVNG8->da962hNX3Melvlwd{46blNHBRWYoF#
zZbvXuHcg_$DdaQCPYD#n*T!g97XUh*9%v77hu$q_x+G`5Dh|Ebk*4WhYtufME#A0A
z=WNkhg`EJOrQ}>C)qgGHdBJ6UPgr6X7hyXPSRZ}x7>aPNP1H-;BfM0%cOz)hKBBmp
zsZ9hEMeVc|F`DT<u_Qq75*1mcMMY7c@+#jG0Rm<N|6w;3$5cQ-z^`ms;@#S9CZxy=
zl6!D)us#6zJF|6h+l6(75|>Z;Tz-|CzcuNPZvm~E&uxp}zJ2opP#>}zeW!ZRjwcVA
zKJBBby0S`2@4m`ONu`5-V_@S$3W|y%R#rRa98Ewjs|T-9A-<=B*ptphyb8&x*Vo;B
zJI^FI4K-U7-=xIsF!t-)>fb+TQ7pe5wpRM>b>(A{4w=!Xm0C6gIX0#0YU*<O#yp;P
z$~GU31+2YKe|d_t_ImU7-cCG<6+O4g^|fJ2VYY9STO*&3Uu&-e+hO>ZNY{Y&AQPwH
z&~!sLJg^@M=?R!LYF%_#7CbMC)6@~Qalb9JLYgiAT}>lU>^B=rvUg&;x@n^j&+sjl
zhAtKxi^)ikRln*g_D&fl<O{?~rAfs7E|MY<_^w&9voDuVBag|g#Uyt<#HXRs{=1ZN
zjHJ&9Ob@T*%Pd*5m(wqU-`2NNIYJhE^HwElfJ9PpaKyyWs}2JogzVtW*l>ZNea69Y
zQoh~~Akk`HT$#lv9iS_2C!?y!bqiI^IAD8U93w?yWUQ*{xPN>u9lPLC=MF0T6MnaS
z<}q&l)_^TVG)ULCDP(5*0*{PDd_9PogrOJno1_yx6dsGumg(*Z$^xi+SZ4?8r3<gY
zkb+VNRhJk`?nDl#d+A}%uN&Dr=AnnjJ+{v%4W&l3h+SeTF6MhQy6J!=e!<xo`V^xJ
z!UH7%t+<IcqE6E_=Z|_%L>7(N4uMS0em1u@Q|ad;6t{2MwjekMlK6(zT-<0PoHn}B
zm%K{*LW1voCZ}czWs`g^6pQa+i|N}@X!WS?gps$Hb^QVL!f8~_q{|}`Q>As3M?Aee
zYRU34!cV{`FWJ>hcvBMlz0fXP%~nZqvssSwia$s5Z7`tqyyM8Nq4B!zmqdG@nBvlp
z1wF+%W@U1vDkfS8skwXZ)&v4rJ<{my32>7vNR45fA3uJ4gpWU5q$>pSKo<Zd6V>)q
zfSQ@iH+lm)oqwV2<>A2#ng-;|%n?6+(1MwuWnfrIah*#y?u&(XrYb@JQ3u%hRRpQP
zPsv-KaL5Kj3E4LQf=59?0lnm6+j&8Nd)-*OvP3{K{MF4If0`b~nk4f`CARFMt0AAa
zG?1QHS7Q9f6aV~1Ap4Opz0e!@?Y$CBFg~T4Q1*5AHzQUtTw&LtcwLM$&u8XU&33=z
zt(DftDzPJ!EXliyj3x0hE3t83N`3Ljj1kSUjC`!1Mz?xhy>R87(2Rf8ix?nb%JqJh
zU0s?iGBAy6aFo@q*n?9OsQ%>~Eo+c{i0}(`oT}oR?Ju;qURr<Xuv$_x$F+sMUOs{4
z&Wu!IR<%<<SS5a)Fwzox^}T=Y97ZT4=b2HOtQhA^F*BZ)(d(cTVRCqJs{EPcaYZFT
zE)J5)3VN1}FltqQ5kedQB?Hc4)cpW}cJ(G70ie0ifHpfAONr1YBKG#I)KS3zfU|g>
zT8X}Tg-qUfG;+^qXaa!*0ov<Q00ytFt|FBw5`MQY8yjY|&fA@UdP_-3Eh2@7-I*G2
zf(3^f_(&R}9e{d}jr6wu<az<cd6Wt_iduFFh@Dp*7v9&`@yWIOqAuF?JFB>7tSbdg
zZb&|vCb<8|-q&+w=U!Gz$-qVEo{+AOqIY|H`<Iv)=hIJ0%uL*n*4874N&5~I$AsZ0
z9f#Un6p{Vy^M3%z|LU^&A1pWj_czpm3#m%2A74f0Jg}<o3-*JZ<)QbRj=%__<@X?u
ztWrJgFRNx%46E1ncZ=6Y6BcvhrYB%L0E#yyrB3?W|3nHvs$T!53;ZA7P_ZRgV05qH
zxl1<DGi@9cr1h|ssA4JQOz6aga7pCTJ%}mp10mNgQlh7f3+O^@!|t??{3TcaxoU_c
zJ^$tYlRw8KoJp}CQu8^RDmJCE{!olsy=)p)@9;K7#-1)#P}A)YTki0sww1zEnVv$7
ziqp;Maspgc`7SOxTvK(<k4bZsH`7hYHJ{e5&3F&#OPf4`T4ikwZrg|)FBo|dJ9q35
zi<*zo`COU(G9SV=yCx=@XCeD`QB#ro^9BQT$+mrmSmJ!KVkO@zw0FiD64T|vyixhk
z<54G2cv!D1s7LA0WBGsz0t6VSPi2+c^G!kks8$q8OG!~MG7hxG1S=}9SS?R{t@}wP
z&WAt~K9AV`IsLN6oPy%^jpSrYj1jYMe1`QAOM-R9sp;vxtOHr>l8+VC>)SLxy60rv
z_FkYn*RO|i1n8cV43_eLu585i>sbmZTY1vqsWb>DkydcbxP`!0?U=l|5xW;J^|}=;
zdgnU2p2N4Z$X$pCM-y86d4_^PjCkL{oczlsBBg>Ex8|552m&ghoSogi_(?iMPSEzw
zBY7$1Cf5fpMuVzQjM#(V9a>?!NAxfF$`hB<=4B7Uak}M`-hej?7<2I~Ozo=m#nhqg
z9d;cW(^YgxbO&d2Y7T!Ax5>ntcJfkpX3zm5Ij`eEOlm($yGxbBj_VDZ2EGM_de24m
zv)?zR6wpe6&sRL+8<F#pdlRi1%kx_r5M6j-4w&DTr+X+erZ9OiCTr5$kZt|JEgb%R
z2m)4%4kNY$(CkHJao$qSmP<h@_fL48Xn`MyhJxZf@Kvb@%>$oXS{jZp5Lv*S45sh_
zR}c3?JUsOp2V>{y(LB`=P}7!{m)-YUFj3S>jc$kFGa5{ln*(Yw=dsNwzZ3f;`o{Rx
zL9m%52n)J!YpA)(o(p1lFKc$9z*IoW^lj2QcY>1l&r~2|JqHxt`hlitw@8C@toGEp
zk_1Z_#LBvs>BeKcgW#?H1LjVY!m>(Ne6oTml^y2^N<MvTTgTFD_!j4C;7Bn#XW@Ne
z%EN8YCH?{=*;3<o>+*3-+HZiQWO>DJw;*iObnSvboHdmJJq;)sZlFV@R%N3d<^vpd
zPdF?dv77Yk0ksS1VG0cmC9*!<nM6v~69BjCdL3W@*<%$+eSr?c55Omo=^X)9cQ>eN
z|0Y@&hWSnJx^J{^hKUB)7qX3fPaa@wBR<(i(Fc)w+31fr#k&5<(Fu4euD<>!jU%69
zu^~sH*oTNZgv>;96f3bqh^S20s9>l^orly^ohhu)GcNu&akJB5Q`4re?=2GXz@Ha5
zVSG8fVk`f2!BtPX48V;cwRi$uuSkI4kWwPhzi;2aZvxcv8JB5mFbNmOo3DUs1tz=T
z_YhF2KxawIaexCt{sIoXaWiDU4T$9k1OgllZrD+dqY(gRP?wa2NW#mS3?D8ZnE%1J
ziO<XJo_u4zW;)tOf8>g$5ZREh1lv>G_a4oC|8_5<C7@>sn`RwddF@?A)u(dw=7(A*
zqmjO`^Qu*s$y?>OINFR3C^F`C&FzOcQKzp!Q%G>=!PrsHPYDyvcd|5Xl9TU?xZ_EN
z7I6xnZO^Ct;O1#s{6i}_dMb}-V%dD;*Iks-C5VBXR?{@^P7tWo7h8EioMTWc_M2<)
zG%QgAViLK4z;*p4$Mp^lh*{l}GyD*Yz&3liYngbK_qXw=m2Y<TR4-Q;Su20pylbMu
zm>>F4Qtxy_puD!%$u%YSI!?0qqpCOCx8r69=*r;Yw^KV?Z@5=q<oPx2+yTm>s@(G*
zZBOWq&K^GCdgO;EbX7J6Q^DUsYbGLj+t6=>{j3;P02fn9F)2oW`TWPSfHKMSYw<yq
zaHhtBTI<(KAK+CRL&37PO66G=5A~yx;ugppIfQ&1(6TO@o{{fLSmDC#Vo5rfvxN`^
zw-ggkIE>YGRnHtPF2nsB2;9)cSCke@r2GIJLHbci1w0@Yy=Ev|CP7q5=^=2MP(6DV
zP&MyUS9`rXTL-+b1-~VM9~5aE0i6aYU!&qPSk`GHr>_9Aq@~^eH5Y`d?yvb^;=+Wa
zLy~q1p)(KzTqy*zqLY!ztwmm`&G5xXHs@TEy4RIsjasYcB--30k_a8jqSSDCCscRy
zA;oHUmjamdnQ98#k@eKJ?+IKISHKKEUE3GOPhLzj19O0q93i2VH5t8wsJ>XF<%fhj
zN}T9Q+2!UTze$PLOSZv8@Ea^Mj(9x6GVa<9E{;SvCoC9b;>TUS+}j5aBc+po<aFyk
zNjEV6f*uX3-mh;WNTC{R8dPYg*SoO-f!h>-V(t_<piH^#H6;2Pd3$>!%^z+X>4A1N
zT1b^0G=g#Q@m<DX06HfEsT>#|T3_J^WV%BRbIBhud%PI;L$~Q2s}7d^EkRp9_Jre7
z=ODKv|H_UoP-w$ZZo064Qf6|o0=<0}3!ltRWr4ford)97*=65lH==ZH?`o)s(2kBF
zIaq$#W<qqiTbB#wn}aI0qT09OlC~_><)H*z!q<__KtEu$nqUEI?3J*vpJdhYo~hxs
zu5;>JqVTr_RJ}T<ns>Np#xGCvn9+|1SF3v+h7jni(Orn?*ytioU1c&>R#qe9lxZ>5
zT+e&EZ~U<F8T7!t16dI)U&pon19(W&>sQLtw4&a;h)J^|YfZ5&-4EChXqFlS9Ub&T
zKs!yfKt$`d0%&S&#UTaL|6ttyzmFpSpReFAzKkDGA-%M%M|P5Ho&|7Ntim3)x7{U%
zTB#_d&MnH2`*9EMq@ez9JK+E6bN~Oe(3g7DLAE_T16j-O#sRp+jg%I(ih3=0J){L}
z)}%(DwPaU2F}CcA9Qu{v{n@5^?e}i+awz{%_Wm!5=>LbgKn{Np*Y#qeTjz<H)ONRU
zzX{&>Rg<y`tS48;*D$d2YW3o@JowJaez1d%Mq?7!@F^OLv7Ki2T^rT!P0<5F4o7_3
z7PBhFMKQj2_uBX6WQ)~hyejCE0tM&V)BQ^|rZoqsIfZ_wf#%vpePtrN8#0-jNeIZM
z2)LZ?rlVa>vphYBusucj#mwZdPW8~7a4mwEt1g`2s8zKNPP^Pqc-G|oom{7O(V;Lr
zq}_tEj3y@@_eN)^aK}KWx=<kEEFx#5JZ7htNw8>{j`Cibq$C84BvSJ7XGU?9&b>nx
ziz5Y+JYRG@rfR{t1#2(wpnB(r>Q;YT7GJFwq<sJ0I>Vye^EgU&M)1tjChbzdSGQ(8
zblSt%O|sL)w&ju-=b-9w9ifN2;apwWHsN#xY|g*(x5(9UQUlZ;aE|wQqm5p7Z#5##
zhKjPeGc<|U{#9Tnq|7^8Uqulz@15Kt-RC-zE5))mI~k)lKjr8XrKj8Iv+?FuMey29
zmdRl^O_Ev#4JMi8Xxc|r)x9<=wuAK+Yu|J7&)v_Zs2~Hgw{g4Z-(0QGOK?#*#;$eu
zR3;cqO!BAN@+&SUlis!x;2bY4(lamTX-|&A8fV;-uidkv`Ls-D<iM2t);8>BDqKj9
z;6kypu9G+p32C2W2smz|k@FlxYtT?sS~OZ*BSTS3N{~ELZOqjnBgSop=Pea;Ve21^
zeb0mq8D6(|wO(15hut?C_8U_hpzhC}vh*l+rO!b1cJtYkOVzBzizA-XaV_?^j`+Bm
zKNr^*6p+n-(PMJ>hJMK5C8%x=B2<4Hs)v2dG_0(VK!L`6{8${={lhr_;qqA*>$p@B
z`+#Lb3JMC%-e<FtIdUnlWn_$i#S{Xr09F7^d7Zb!z+R-`HZSew>S)Ljm?R9Vrc0(_
zJM+36ck2N3yLI<T^J7&l5r7Th`)xSPMsZ?tBfNB2-#nc3e06R^*5A=A%cn2eIUY6H
z2_&U$3@udtt~h;6K<GDWu(lvCL3Fwvle)vaQGSYx!dc6S^ZeQl#hB)U_{Sq(l(r!Y
z6TKu_Eher$*Bj<gt!bu2!y%o%t-woEmWzh4N6|bM1-Y6bERfCCpMb9ffl61cCtloZ
zcl^B=N3E6E9S<rrs(A%V#{@yfaGs-rK%2!7+#c|segf$2=yci^Jl)}VN`q*JA1nih
z77Y~@TG?bjuuP!1?bfW*GBblcABKmOE>2<6pf3v?@GK)a^6Q|r3`|dNm6ZG^t*C{B
z(lANg{D7W^R98#O$~LDe=z-Xm?tZkkUS~<!)GanpHB5~2cz?eusk@&(-SS8QEaWap
zS6z{QM`<4WPXx1x*D2~+MvzHau7n=<^FKqqU50fdHRf0m%u}KkCORjz<=5w0ml8yA
zTz8h3$E((hUNM_a7Fr!zx}WVj)u6B>$;=$?InRtVJg<88(pcAq)mN?lo?pKIPgs}a
zt>jZL+?w)~Gb~{Z4I<Dg1_oPFKVTvu22TndZ5e`RDu9C+x2(qE==bk;dx=0#>%2aj
zj{vvE!tw<?V^9pNDL_Iq8_B*0ddUQ3CIggQTm(WwLP~GnP69W{xVGz4KxdE+3zwY<
z`zcPO`Om<>;3pYV&XJd^_FWTK7hI^IHBaYC<I^RYvV4$c`6}(rn*pPqN<;o8D+0c7
z0^AkL=Ypjc0nJ?ooo6fO#|=Esxs#bR0?t*>ZjKNZ%P!+rJk)<8Y)Kox`oxLb-T3<G
zb&h*l<`2Ed)!Q6guCMhV%R60pKtR~-H}=jEKi>qeQN{7(#rs*Bsyesr2!<(#Xji%W
zH&wwG^Fi%Bp?*Eb()QwFOya;1kL-kB^@LcdtHan+e0y?*c=1C%mY2T-cL#XcbKt;P
z0>jY-a~T0%N3e5fYCPaJ9&o;jo(ys73Oz!})YU?_{_Y=iKmSh2D`s0Ukk7w1@ih~h
zV?LQ@4nI)sP~5d+EtJHDObQDVoo{~6sush{DC){Wq|As!13OztnCVQr4t_2adk=H8
zOQ74GOVdx-*9W;1<Jg})W{f{ik|SZH%Q5Y<8{6qKUQYKP*_w)bdop;L^PU0>$sfs!
zhj?p4b=x>f%8Po@4TP5t((qkj!J?Tzk$A@UUDhAr5#TmAb1H)Ith`dgY9^+_Bb;u)
zXwv@D0<u{`g^N(caYJZi@W4cjc7KYtAq2M=KiIOa#44CV%q2gJAa%t(XgnfC*%-|u
z0Osz|y;5Lpkro&4Fo$VW*$e=fjNRPmeW}*q;e1jHye+KW7fxD*KOGZu<WPWfCo(xX
zbQ+djT^)~f#k%e5-NDA@@i?ZvzBpOl%t^}zj&G}(>PXOsK&q}F_*7<HJv=-LPJ;fd
zrS3AiAB~AEeg8REA-^-9$Ym>w{&S8Z!t2Zvm`3V(>+!<;z1%mJ@5ZJiL-F=-bgGAm
z2Su6>pE!zR3(ot5TF$@zo|$oKc77Q>P+x~2cXbKVALg)zTV1k(qT^j)?f8ZcE=Zro
zH1H}<5_jo4YA_EWR;Si)|FgZV&P!M4<(9b;h=lp4-tlABexk@`DsN_xn^A2LXBS(-
zhmLinAvGtndoK!{4u+EW@IV?LNwYRxO9YMo2E(>oyK1!>n!3Nq$Jn9HSFqd7CC%&b
zHs}F%j*rKQe7JQ0_^;#sf(USE{?k7^-JR_N&zR8C(|-p*6IikBk!~>H@{vpCZZV7y
zkN^u9wBZ3KIRYI+rT@h30?gQ<PgwnzmJDR5@qpz}9{}WlPoGNl+ip?IB?F7{{g{{-
z@RZ<JU~C2+5g&Upt41GgvflwB=GD7<lV8<AaGdXY;9s?xo^QWd*3-Q&7}G9EC|PYd
z_sL_v@g+l??-_HY*Y%|M2`(PqD;1U4K)SCX&LSw^ii^K2Ea*YNW{0#5PgYtdb%o?f
z-+_jIrPMx9aIwemHZ9COgQbCu^-cOocP#=`HeL)r-8Ae(w0-c*1gkH?3v@^+C@A`q
zxQ0fYgP)|ft!7_K)7MvC_kdwNG@Wj!^v2EABzOc-%cu6Q^h610hXC~V2XFQ+c(U)m
zoH+Q`fA$|9j{Em2X7?5hwZ6lGb>y_p_rC*P+c<JklL_bnA#O`-k<iiqc|8{?{GYnu
z&Mj<jOFKE0BT+wVD4>FGb#!)t1MJz~=LT`9MqU7S-~XzS<KGj2Z#OuKPt#uXsvgpT
QM}8qGaXGO<Q9b|v296GL4gdfE

literal 0
HcmV?d00001

diff --git a/tests/tools/_images/cohorttracker_adata_mini_step2_expected.png b/tests/tools/_images/cohorttracker_adata_mini_step2_expected.png
new file mode 100644
index 0000000000000000000000000000000000000000..853042a888893fa9ef951030b8c086513b5602b3
GIT binary patch
literal 24087
zcmdSBWn3KJx+UC5kPt!$1lJHexI2LW3GQye-Jx+v&;*O&4#C~sf&~xm?(S~0JO6Xf
z%sF%J{O*}M?}zt8LsL|DRqfjQdDgSmdg`ORtOWW)f`<?Y1pV!s*NPAbJa`M|i-H7x
zO_NPEfnPjMV(LyxwkA%{cMir7nRiZhR<=%7<_2Ug#tx3=wl-{x+>ER*$;_Ob>>PQS
zn5_T41EZ~jDHGZCT^slkRJ%7Cju6NrS=bBC2O3HNf#~?Zef>(=HDPxa>WS@q*LgTz
z`66ETy)X(ndAa`!Ww=*gp1xA%eo*o}%^~R-X2kID@@s(std^g%a<4yQ$@7RID1HwY
z2_C^C`!4*@U)X)Je78vmKJSZvtr5bpn;^+G3g7YUzNZQKjz`mRB7WqoO%;Xz6A=_K
zg}Pb@Ie5n|fC>>oA@_%W3jv$SvERV^g5RmaPa&|~9(;w6f!{Qr5&!J`{sjc=`sCpY
zIB<Uc@?`(?(d{jG?e^K(#Lu6fs6~VQJ}IcH2M?M#IpZ%qYDe+~n;je-b#wy37mBN^
z$K**pX(A8$H;zMl#@;;i_30jg;IC0*6#MXW3AW&ew#!?r{@I;PVuXUXM_S;x32#c^
znoY`%21$`?Y9EpJvs`Qn^pdQt!ggD~bf(tP`C!M&gZOV;*FO%4;!c2X+vNjFS@Gpn
zJzEWp?&ug<;-Bd{(0bm-uM5uGHB}^yuaKm(9&=1Zn|8k6pzug{Ic`%_5MFKEXTc?H
z>h>Tjlz8WJ1z+Hk=ic+M(K}!()Y`RkHc;>Wdhp6$pRp(1s>d@o!pgdwsmU8-Vb~Cx
zKu$<HK;wJ<Jy*w!mzoEzzZqxY6pFBm3n!`9$)H%({q_94thIH)^5y*98CTRx4F?WM
zEZi<dg*TVsMZ_3Vn{S<7?H=s1JdAG-Cg1gt&pX_9#30yDLApLbiRk1rK~WXOjJ)<1
zUfvASYr0n0JdvsAT5^9!*PY319=QCD-x=?`@(?k?6EnW#)7Mi+`J4*>o5g)IglW!+
zAhjwjRJ;?swWa0nRu>qWO;P1U*BB!tg5WDes%6ViAw>;mnhg+U!;VKJt_$z;6ymDi
za2h_#yrIC8+v{|75USKfxXKoUC=<`TwmxIbW&5g-GrBQ-Zc~imWu)WL+PI(7HT72J
zjAVotga7Apyp>SfX}hY5_-P4t?qi#@tp%$5NU@7EB6Q(NIW(4?+wi00=rl!3aJ9KD
zg8h|DlLxksJfaNW=mbo)67JhnaSrODEV|3`Aj&H#wfduCO;p*E@+O8yF=#p5T-ZA}
zIJ|lDhK!#-;TgRuv-6g6rPbtzNvn?W@o2Z>wZR-E>Lk}CR0z^zqG0j#ce_&+H<trG
zROICEyS`B?X=rqn4G8vx6Z1jv_*`BNXUQS!|4w;zvNal#>~j}eU(X+iPU!NsMU-6@
zcP9KZpKgaug?{fqbMZMJyRP&6iS_8Jsdp3{J1_jf_1PF&Iqs6kOzX1Q`A>@-9IVk!
zMH*c;G#!0BNYtkfwaKfO5VL`o_X0Z285Ouo?knFR2e;?*N~)^j?(V!0N=nKf85t~Y
zM@EWDO2oUFGVvX~v21V`M?Fk#do}A9ch~!u8|e`-d{9QA+Y_w2ySw$1`>X9DLw16<
zQc~%enSpQ!NWW_wO*_H}k()~lJ15G`u}sQh3w7#By{=uZPPf-Xb(Hes5*pO=K9y+o
zf?2*iSc|6Ki(am*V=xdsr5mh=KQyLK+t+TA+g`P9R?j=@KWoLGmALivK32!VgS-$K
z+=C~iqysH}$Zkhn)Ec!;w7iLhEiT(*1?hTC-fbNntcD$-?u!eaV?}xc%Mn81ngty_
zJt2q7on<z&yq#C}%N?ns7K7WTFJ8Q`+nFeFyo%wl#CiAbU3gHtMMJ)q%BQVXE#|eo
zPNyqo6=W{BlvB%(6}fnjimdzCWN(Oip5&HpwxXYA&m_jxeAW0(W5bi^6whf><qX>&
zYi6t&LX42+0<jlsU$r6SjSO(V6E4kZ1ssw2Dv%<@)z0^^n4>v|8ei|ZQh#ryg(G8S
z{d#iZ6o^6EUuiuZ92S;QUS6&nzuXydd@?Lo=6Yywwm&}*t>>+=A9J4}f6G|t;p)nL
zmp^D$tp0Mn-lY3G?aJC3FNtQ@T4`kgBEfvr#kmiPm$DxJlrxkFm*W{}7L_A4L{|rC
z5pxi0VB>DWn$x0mE9%;*=^IYEZVN>kwQqWZT$R>f)qa>@8Zu|um#O6(nI^-fxJ<N<
zKUN#=e^#C2)h~-b;Xa{-SU8>=PLvTr?%!i9C)y!<bA`6rl<`I(N5n|h2^F<dzk2}*
zoOSL&2;%=J7ol+RBDBpUKh5DDgK>h5TpM<^TwJ)3lI@Amxh^oRN~U)aA3RXd)!j2w
z($x)@&7xpu$F7_0PvCLfI$c~|Rw}x+MJML{etEoZST&uKn`<zbCia4giZ|V95K>q2
zu&(ceX(!x8Tg2=1^Ngzv7rOJJcJ1#n?Q{F`+x!O~o6vVS>0G=6M&(EJ^(s1*a-`>j
zJE-yYs~6StdT8hGcwCfZu=XRX$0l=oinXLKP8%Xpw5ZchQ8PU|)xO(Y2w6mADVDG4
zjO@)JQVRObnq)AWi>GJBtL|MV4UyuepgpFx+fY*PM*DTYRWxBi7w1~!Z%3Tf3kM4g
zObsVbUFwKK+JA~-2)w?(yZ!j_BRMB$)URJ}RP)ta8qVjWq@^Fddi5%lM4(oel&Ssz
zzBLfNb$Iw&h2{8|*7;Um(a}!Xz)*n}SBlNk1%dOK<<KE91gj3oZly*uCok(W8;de1
zS9O0eHEL_p5BHv2jQcmIi=>GfY4T*2DyR?inua#6x<?|FGYhTkbOv*)Xf;n=8jM$`
zaTFS@zbIK^FKMQgPVc&?;VwRO$ACICx_13ZjxBNn2CKsYuylLM-&FX?mTRv%P4MT>
zpTGM0oE62EB)c+v1ODw4|BtEAp*zF3NX()lkb2F=#Tj2DLA2m}@zPnPkWn<){YI<Y
z-Wm007d9aT9v<Ox?Ya+a2C%^n^}msP|1}=@G2sFC{d*MaM*p<7PGI#T3AXSdPS&5J
z#=$+Gzs1U!)iDEHz$mJ2hZe<jHw1=<TG0wpJe5ClH020RN;|N@kc7f>%kR&7uW!2K
z7ZDVHzNp4=yt)=~(Z~ViUjm+g9F70%+|b~3EIXhZGV~1?HWuhAI9Ng5#~L0-%+~z3
zH5B#q-Zt(#E_{?77T(<hFB)DpwRqQFRT;h64!_u%DVf^E9#Ji2Ve;GQ{z^4Rv3j8;
z@_e@*;o!F!E!Svi*#=@r_}=p=u9!}~Q;LaJG(HzuHNp!{Q&QRh9`v(et{=qbVy9%k
z`>;Z0C2drA^i1{^g(-Z-m%6=J^@$R1HB-j7XjoLs$eH}0=eAUgWZ#>d<$grY;TC^W
z(&(sfoz{qk{%G-jI_<PnQ^++avL7~SvKo^bgYo$0SAQCoOr5C6P2jLN<b$7ISAYLn
z<+RQ6!D4IwmTJNmJ*AebdW!sRnNG8^-_$>d1wQ;Np0wVf=V{Ny+O+B!NDG*YucKJ9
zQ?ex3=`<)S$)btgm;F(e#T{L#dl&jK*m9{sA!?`lX%x|P>kWmFk|hTle?FU^mV{x+
z^h8{87qx_<dQw?3*qMLYnan+`u*F-Nlb-M>oO%Dig)@m)QljZeQb<kC$8&Lw$I-aV
z*UdO{(mE26?QORT0@ZmsL?(4%D%6jU)|j?qe2kEZJ^kG3&WT;kXTj;1RU9~<G4)Q=
zMcF!>VUkY7L9b>8+dL><d&!c*sW-S+l+|H3V^Cu(rw%Tr4i1erJtHTXn|MF7<#3cs
zfAHG^)!}?eT+3sGj)N{Tx5YbL8QY;(KD16;s-RuE2?tZAmQ?a&a!>~x>|m(YUi1y$
zl<TD*F)Q7tU+mb+InL=7Sxo6WC1%PN1ebY+a`ueZlE=R-XR^y<svE4wsb(=K-A(7I
zJe^4{HfHe0GO@l0zVb~h|H?lkCAXaJn3V`6i*VWT6VDKz{iQb-HQYI&`#G#?Eb*Zu
z!!qBA#S}4S8eB9n1Qxx@&4+()Iz1v4oGIad{`|v_A1|Ihed@X##-(0rgu1u42YIR2
znBcNIRpPiV3m}BJnAii=yhxD5VgVvBC{f4bad^?*-aaxi0!!reZpQ~lN6#rJ{7|u|
z<TW%t_L&%tWGmDkb`Zw&ndo|*s&LxO1a}#w8Fq&ECh|r5`ufuQK;vRfYacfs_i}}T
zOE|+`hkvq4*bD9vgjs{ZGyYrPpoMvPGRdoVGh>x#kWm~gJ>z?aNP67;S!Mz~E)0k_
zZ(>)V5<a92xza>)<_W~2z+~-*ZN`*=O?h%l>Y((FD+1(Mrltw$#UY&<V<X`i4cuN5
zB%eR2zjTKka#a<oFWXcHmt7L$YCcZ)$iy)HcIFvoXV0N191Vx+@y`MJLWfn6gn46P
zcp4g-g)b5$?<Px)0kj%S6-E9O60%)E%fMi~)*HJsSr#-hqWtySH}lC-N@HVV={WX}
zlu}W}mg9_L1zMeGO!dtGBMzFed){5H?@U#M++LqAudjCok$R&vgG^a{dp3Q!8m$LQ
zI!c++a0?*0wg9+8%;)ksE>5z_X4Z8zQmyd7+B_-BB=bl`m5w^E`y;xU81+03WJNWy
zWMhM|aX<j2`S%#{;cxb&-N%vZWfvPZoQb@jg2X$E#}eg-e~rs7VwauSnw;G+5Eam*
z`GlajBaP2uGNq>lb199NFdHI6&LvLGobd4VO$Sl$4tw11q9Z?5^@ky+yQi63{SxMG
z|DGJp=G^)4EHc5sRoxURzy(59u!8L~bN0+O%Nnw9IT3$2Si?0nkXireu6&pNF0t_$
zxKEjFk45snx=s9aX0${kqofR|uI8?FDIomdc`_*Bwiau;RdINFO3BL`hfM1ExvlNB
zcsS8pb#**KE}Iu@Y~L^`w&pQ+Ky2|33}p2<vr;Y43}`n<t*=kGyS+xR8<ycZTnb@Q
z1OU8`Yrgk)s`Fc3TH23CM-BkiBl&h2J%9duDIw0ZFRndbql{>v0B3u3r&fx%h|c-O
z&FUI)$xqX<m_L~yFIHNVB{t9?&GW;@GqltO=o93gV}*2)48?b4)5Xbd1ua%>`kY`f
zzs~oe2a9%4albVu?&0`O3_Q_GCcB$UaP{~jbf1U=I$2`EencicFF3(Id|$9hf-NpB
zOz(nCB9X3`s+v<gu`@Anvm1$*N6VsmEcKS;R_pce7Ooh(B1oyx({+`y>)tnn&Aoz9
zRyzK`Q#T13w$b7SviRKxlP9)=N$(4u*O31p-aZab5pd*gHYl?{<6>&6{CYDbS^3Ne
z7A9xmMZx(mrbsVgGcJASfuIxHj+wFl`0?X&LPGCaU;O5D6_cLl;RBbt#F!X7FlPtc
z`j*;5Cadi!v?{GyK(T`%c;OpH$OUhAJ_{Xg^6`O9VK7Nu_Z!YZ0OSB4#1Ogx_o?l8
zjN7;yyE7Zkw{C9v;d$-m>r2F8hT><`bUly_tf9xgHv|Z+(@q^`W%363RLwGS9?4`@
zE`$#%A|W%FB%&hngsLP$&3jAQdltA7@2n}Gt!Zj*dZ7$@7?zGT=yL0M>RTd}TMaB8
z!3(z^6!$|`l%tO9g62ll`w|LsQZj-tO5jq}tvMa?x*;(mIi!9fe9ZfH55d?RNRPiy
z$fryHr7q?VC3~lPH=@~_P{>kkgIr|DRT;ccS*bIr<@xlP-F$4TRZtIF?3gukBn}02
zZa2#D$$H>tyO*4V67-)SF>|-7;uRsQrwx{ubl13dla?<!AGNQH<87>rf1sJy&t;&J
zRgOx^f+sI5?0E>ey*gb6QDX6Ts;JZ6OzZ<R9DUZ<tqK;rM|h8rK-u`K`8@)U-2%kp
zd~Y>1h~zmveaPwdIA3FXR%Yf<g{7LLloaaY$DrQC^=oPV8Gwe0fq_g0LTYw4O-BT&
zT(Y1S5d}iJM1;xN&g3IJW{8}eTy9=oS9kYwDylWfxo8s*`qBb|FhFuoQ_W$H39}MN
z?21zAtS&M3a7PSUEO73$D*LLY(_eY>@>7@!`k9I)$1vXyh9m;6P)@W?x=#}hS9s!D
z%Z^BGt86i0<);r*pmy8@k)NI#a=hiazTmya6%XiSt#)R=vvChqMFXqltHaqhwb_F=
z(vg$Dia8SxMr)L;Szxoer_?94W9lLS&i+zThQsLq-Nep}ouJ~=hd4<=o`jzgL0{c|
z7?SRV2UASUhG^ymAQ)f1<UVpt&l*;Z2@AEHPlP}PB26f|E+;*mXksNpa$3$r*dM?+
zEWPT<nkXi{e!aUR+?kmfJV8(H<Mjd4@j{(P1nj{ej{!u2&t~#SMMWiBnJz6oU87hZ
z&fD7?7Nv8PvJ|zoLnp1MK$Q=64Hc(=&U<lD%O;S$c;TNU-~rZysG{Oi+lEt|5L~8?
zzCQH$z7D<nD>YC!lf1-1V>9Vpn6_<l4t*^h&l!dcSdIGAx@`eh9=A&nRqRSM3`q<x
zsBLR&AKBxp%D4yo@)2-c3{0CG?9*x8+vn0%AO$rlDE4%#Q_Y@&OkC*v6jG@3zr2K4
zAE^8*e(k>kH~)#6<5{)w9e*CC*_yezUk5uO&zTdV<eCW>G+xL%73PA5<@G#?CVyES
z0O&Z`%n8U7yeG|T&+`7zbSR5Y|C`7+G76i`{Y^eNhDfmG2cHm^mp2BH;4j%96A=9}
zjTo;^%s*7&xc_>Sqw|BGlPgdD@A=1n93RCUP~coJL=?xP5SQI0tgd~sa=>*$6kc7E
zTJYsM3?&HY5P(RjkWgyrW%3$o_<7X^olzdF#a1JC8Wu48EZ)cJ_VzA2B$#^i!f6NT
z6DNPbyLo<gL2J~v2@xJ4B-yiGIbE|qH;aN2k9h|k`=qSyO2H8oS-;V_y>DjQdOWgs
zhj(15RDbE#ONk7Jf?tB-RrlIKXg9P^-TjVB8pXcP5{CkUii(Q&C=jiZklpOb+qZAg
z3#B9_pHot{cSWLRFh3*xp?~TZF0s0J52w%HFg%z!n!S3_ar<QoE1q`hf;|#BIz@4t
zU`K9uPohVmXYdl5a?y#)@@$tnW!|aYki@(^(3eoJQ&S5NlBSSs_C57Hv~1p}kqt}s
zIcpS>BT}nOwEH_mj3htXnHRZiY6`Wxwyx#)Z+iHm-`DPW2wx_h54s{Qj2~a3ol*K%
zpK*jh02KKY5i#)WlpIc28&yk&Tt(t;-L*QP4MDzfTL;4b)21W6FKi@fMso2SL&^C<
zs$0=S-ua>j>H2ufiIgA$LWgmKD>@~CZJPw|Yx`3Py>k3P`@&$Y&}i)ZyD9>>;_t*6
zc71-ziKrCoikJ@DSuvl6vUzK7K3r$p3K97Z6)-83oO^{n@c4ZR6`tQyg-Bn5;?8Ip
z>lh`aR^JCqvUnx<=1x+d-XCwjLEgy8Nn|LNxa?{|OnPHlLh;#t{raW9^6SgV-fTQT
zAON>5yEaH>q~+&7@?BXm0_b;Rtbh=UTK0vYU{YT^7vGf4Y;ELw1mvgBp0#7i@xwK5
zjpk{T7(CdoKYk4EohKl!`X?At<*&{t?9p~!%k1GuKk$Iq*-m0X!qn)e1XM)ovU<$D
zY*}lY3(?O~(KaBy?RHDO0|~XOcf<s9CKpd2p>ESmwX#jRO7>`b;)@tqaMKT_wepU(
zE1$8z>&U;X_1$3O5aXG`J68PdTH%C9M&5SXLkp+B@jEqNPY-(IG+XO@vOQijcT6rC
zXgBY5ez;e+UkU)bqN3u(#d5?XAn_W!rKzx%JHiY>(%6}<iU=Zc4+iv<yEULHI{>hs
zXz~#fx;a3YuCiT+6><H1`z!6ZdM`LQB0EVudhzQhvAc`bjUe67K8KfO)A<mydJ)>G
z1MF44`v8wTUH=!l+7rq;3xubW81JNv_ry-KPWOd{n>N4iyCX+-AXF6EAj+g;yxPSa
zS+M-XCD);;B@%*}jCuB(L_YEIRPFX^-Z$gONax=B3$9AglclNE?~o_>8t2JkCMHzi
z($)*A<_y5;E32x0O-c%nh`>O8OtdiPu}=wzR{#JFneo7#E*c+56%G1rQ8DAt!vKN(
zcr7M2YRP6Z!(Fb);NFnr?~e$s*F}M&KaL}$xR|-K)u85piz=1hhM_hcqYhumo|j|N
zlBlQ8SOSju*>2X>@AdEdg?#LGXIkvp^5!#(y$KLUa~+QYTW$zJa&2-{DcX%k#WYi{
zYys3#nScu(`R-d~1W_CP6$S+H<eu-2t6q8hO+jAXru0R&K;wI?<MdzGN8%3mgdv-~
zH-Tm8M9$Mf#~!I)MIj>E`w1>M%>XoLl$oI8<KxeG9`{Y`x&Zhp-sFAz+l(C%3F&KG
zoO{=8uJ<|Vm!P1a<pF{)q>y$8g<I)Z)&N?iOzxy@0fMA@%I|$ern607j%LfY8H&0t
z;T$IwW<;MJ?A|@9^{r(?un4zBY^JaxOEnx*^4Kz!Wgvbty2ylDZI3jk(`m@q)5hlT
z0`jcsrmkf-uBQz0`^ujuD#uCJ)wv$IDy6pd5y5~w2rVtj#kgR+F6sB3%TL36;Jv0_
zDeuHftiG>(>tC>^878=6BJ1P9$H8NH7+Q~@qfebbdg<zqE;4U}$5A7s7=N9fo?JFv
z_1#o=jgQ8^T9w=S=>oWVxAjEm#&8w}#19$6<mzPe5nv5v6L}Z6w|laZy|>JKPBs)I
zB_+2`VN86l2^&DGaQZz_3`7F%Cux=Y+uKrZZZ+S|#A&{CWHIOx@BLJC{?sj+Jg#S6
z_PJ@v#zKw~%Wj#_RV=K-ZlMX=`a;RiYckY$g^4sQ<K?%Mc+(}ErxNDf+jwV8X2o7I
zva{rD=f4}9#?mpUj3%^i-1(7xwe%%K0{hxQfVbihQzpIbI4s5sN6@Ex)2{s8We*85
z!t1$w|N7wbeQTM{^JnIcU3!N>?lLo?w#}M4zBvccXpMTXAO%R1z(@E$0hf)>>qG+q
zl%XLAMvw4W5erm%x+Cdh`Q2E{Ob0qFDxMM%^{yw6-^+~wWsRxn=4)1x`@{G?@qr|P
z&5;}!bB{Z72e`>Lpk=&~lw8WmO8V*sS^BDhPW|idci!OX+E>2>T^CEgZ3{=A7uy0=
z?meMyT`OLb_i2^ZjB|3<b1wW8`;ag~bs2R!hyKkn7BNfGbTjK)pW~(`ljh0d-*{C~
z0)v-gij~KSdKeE-$P2aT>TaK7@N#1MwJdN`+O78|&VU<7#OLw|pwTQFiTzZH+hm6X
zC3aF%CUNERJ2||w0o$WpK1RgyVzkPsuV_ueG~LJZ&M(Yc@94KEUcL;SnF)#Cpet-N
zfl@Xb_e8OP5)W`<ngvl}Avip@t$M+0*{KTtnab5`<k8kzD1#?)I*=IVckozL(sqD%
zL?Lf;F3`|Mm4~U5GE)DD1O7EF_fI(KpVZI3J;i1jv4XLq8>qq`*0K=wj$x_>Mg@LC
zhrqpmrRHf$5015@N%`*zm`h)u8rTpl0yXl<Qrwe4x?q)itsU?e7A+@4J)iUOnH!9t
zByX#D*Xg6J^ZvgjRsPeZV|&o*lSA%2rLeKC{fEhap;c(MGu%h_Mj5t;pF*9iyW*>y
zsdy(s>K{_r+VXdI((*v0<#v9Zw8@@cNlp30)#2<}*=Y&xY0i-n_iRk4Zm`pF8et!f
za9wk2=uYc<6lk{J;UX(2T4&w|X?=WnWYn)S8;o{JDO^NRRUhSJ+p~0=TnfR%!~ZqI
za_E1(U1CU`Ui`_&>es?{xl-7k&pOUry_W~BP52{`JXOd1>Q5~9vtr^3&+hRCV{wth
zFt7VfSU=@5ZhAcvVfT?Gh6@f3Cgb23jn#Q*ui|FLkuN#0?yAwpvz_*V|MF^4*R{qx
zw*CeZ5GgK5fYteVW5Inl>DGnxoMEodD+GGHw`Le-v)wSrBE2yi#dMamq8A^|S}G+a
z=#wImzDs>#Bh~rDcJ7`Q%a_?lphefF{@jqtN5lEEupDujM7x8>coZb2z(sKWThF)o
zABSfh>1~xqi^s_B3(IqM{kE&i{Sk7rzlt}W+%fUQn>NmNf>EwE-@enj(6_{~i2T$<
z>B4^>4`Lrx7mBZZIsSGy<Tumv&%we2?%Q`0MZ~;eag$o%EKrZfvF)Qb%sEVQ^Akzv
z_+sHMmJ!%rl{EbZ4J@7bz|A1p!W37C{Fv1^iQHU}sL9OC8tgyT_+w2a;W+NC*R+Cl
zX{qPEIi%7wdg!_#nP{5=ULe*}=>=+mmqu)ooTMymo$U9;qVYkNwh(haz7&-_`8<Nc
z&5z!@C}$4KVa^h+UbmQ5E*DPgsGV2CNEBLSl6Y9S-DUH8S>+v`@1aU4_6#-on7*q$
z(Zd;1xJ%2+v7FWfEiElZrl#YTY-Yd7zJC2mG|(A}-wMiEI2tCV^^#5!3Lv3eT)0xn
zCZGr#8&kp%7uYf}p|0z(%h9<#@4Ad$>$E8%A_5Ok*2>BVdh6ay%{L$w4(F*c)k63C
z)%3iadFo!Y@$e#gAaIK#J4oiiZQw1b6zN=rRLP%4Oi$b|kv-)LWS>wo`b@1E7Khb@
zZ~lw6fM+=QoM#t7!1=^%S;9h>r^56@K-ZOJ^TF?2{HMgBeI};MgL<vM3KB(VTz6F~
z1qC3!L4xyAZUq@Faqg9;$ok4w^ttl{Z%2h}mh%o94qV2$r(T#wFBFd2IV#Khd4ce#
z6U`4VFE8KR_7Q8c(wZ=l54tuiCzMfLeesj&CJc~(>Eqi#WECw?RZ>?+Y2Ges3Li9k
z!N3p}7e|2f2u~Qq)3?&nMj#&~@H#(i(3Nv`=CG)UD=lTEl8I{r02CD+JtI3im_)$+
zH=t&w%FP{f>Zsb>a<cDL-1oa2lI~1MCo@LD1$Kg^>v?O(^ix+)LX3j6_Vn6P8;piU
zMXs(sch7)m4;PmEFvCbRt|}23Pu)lfBV|^omXK_!6q@Z^eF57&(PY3WzSwt;gWXw+
zGK3V>oa*Z)KDQrGZfIw~R~5zfWHk~*vZ{KUYlT-y>-D-Q_SSVPn7-e?V=pZ&ZOqiL
zK>$g_Y-3{+J<<uwX}r#K7S6M4q9R+390ht!NdTr3RcMAqN8<vS)D+;@i>s@r#Ke7m
z7=piMY8+{4X<J%b{Qxd!Wn;r<F+c=U0RRGQ-_{_E<gz}1lrV8{)?6p@$E0SRQ<)_~
z=-|w7<<A#!Nm7vQq=U8W$&&A1n3i4X!`bE!EZDuHoZHh(`>-XBGQ(~$Ge9Ia@UpPO
zpki=LZ-3+d{FbB5rs$r(cfc<>sqxGET?DrqpC;z(&R+NJkUW`uKE+?iT4fUs`7yu9
zn2?~^@nvA4*p25B2O)+hP40D{4Cixy9{9=DdOz6G6#gcm3ng25J%k>l`azX=gp^T>
zlDjq0%F4>c*#33LGa=$IaZyRhhh1G=R&({Sjb7JB0B-A7O~b@R9-hk&UNV2jbTa~<
zQOJ<Qk%m|%;-3`NKH)2sLgPhrWAW*<7nJ4cUwQvUx{>y2dV5Ekpfy}fS%uBm?KsBy
zl78d+bTO5-fUBX33=9wgALEWa2ofE(A%Fh+8>6l55*Q9C5c3vwtz(aa)Q{-HW}c8V
zSLQ!GOL{4kRhYk2YdhD3{^*@I$qOMiSO84k<$aaW^|RW4%qUX1Y}plmKkQR#+St#*
zCA#Mb0<fjiZn4~KjsPKz2s0Jgue>qWAaJZ&X9n?W;E~zD#Ko8qw#uP!1TzGr4BpDg
zu_Bvm5ZGkB<^!G-Lc1s+Ue{-o3=DmT3?7F|?J?4sj#JJXzuyDj1io*6{>w_6*)4S_
zK!>c3YtmMgwT5w~7;rFj_|KZM0W;pOfdSKn=J)kiTX}#X^KI}r9|Fsyw6s*c$-5za
zAE*^b+n8zO5hssL=L8O4KN$D2Oy<o-INtk+ywyk^acjjjN<~i^8lDkQ#V6<Wt>cOf
z3o?8}hle#DcRxeid7*24c_$B7<((i>k+W!asXgQ^IPLra+1;OIH_B{`h;E^=y^hdI
zOWWE2YJ)#JeZeD+nf9Rpg%1m*<J_VL7vpRt1li;0brTBYA-xR$2z_Xc;gYL~u!l<`
z&3fVeo&)07`5v0wrKABDCBw=?K-*wqZgk~YK4vl1s!W$@3WiAI5XMJ5<CuVzN*Sfo
zK0?1--_3hpGXQkR3RIg%gq)!ff>*&HsDA$ZSr|}RVBr{<n6QFM$imVR%(2w+^6%Gs
zbrCR~MnR$FctA)F%tqjXnQQQ1vz_Coqo;Q}n>3v)0eS!sYm5OB2YeN)NiSY*ZmxWe
z@@ol+GW{0#o12@7GE)qtELjlT@X98ofujMC7_P3a2~bj}JrecgFQ|TLm+m)#c5C5b
zraPu}>Mlu+mqk%gf_n+B#x#0aiA#5s9&w)6X>OaP%R)pK_I=Mz!J-zou_^pU_f&1G
z6yCQ*zqzv$*%uIXy6&5lk4gA9PL>L(P$6kGd>-T}X>?c!YgU#=yU7ht<f~%ic@$}I
z;T3sNA>+wCaR>*CL8PTln~M3W^o@;;FEvWrCY@1JU2)<Q`)zF<CeQSA9fPCCiNL#n
zCu-xy4}4+L7I;lgR=|*C+Q(mgzW=@r?q9;s-^Mx*FWgf`vp1`kh5iO9F$+Fu!^~*F
z8<*1DOy5`*3G{gT0$4o=4*y?s<SDz<&F*bI*KYYweRmLWB_dt_?|8ML+FFR(7Y{fT
z^4G&(J>b}x03s$QCkGP*FrL5_7z#>jwP-SL(Z^ty|J^J|>+%>de8Fq2HdTBiOGAL3
z9odq)7i9WlDeRfB;(Yk<Aq_qKH;jAwl;KSj42;gczTGrEiogE9va*t*n)fj{SU79A
zrZ&Oy>#R*yb#)BLAH$&d5AjYkKoB;6iwZ?2T>+SyyK^ud|8+o>J*W{zmgm#Zlf5qs
zl&;GstwzSiK&A;(&6|jz$U=GeFzv?=#P~j!!5{?e>+oqSRl%!eGC)kW@gNB`1WmTq
z12+K7MzQx}_?n+@rARSni>XSpcsgHpK1s6RRYr}V&{=<DP_zTy>Xj!o7;o)wJE~|8
z=5NrY%#((%$B@c1+jZC&@!)|F1&XxifVaqyiar87i#{{n!P!}+^Y+-yGU<JaR+SAH
z5g-B_o}b$uRP-hOF<$ls&YH9<#szf<VMl5%pLp+bZi_|0dtPm3)t`=Oy`ZEV1j+%#
zWq(daSC<6jQAbVd!soC7dtEk66u6xwm)tKw>a}ulfV?qTmW4*|z*zI-Gni|Kd{o~)
zf-v#BNKZ&D+F|s~9}j54#`!mbc=|u)&bIyPni{YoJI;2e39Yf7J{7UIw{I{O6B8>p
z8(gx8UdR4p6b(kYx!w0M1tumUUZ-CrR{L`e`8sttzFM{oku{F%yklM<oV*qne@Z|w
zwNjz`I2MrxEU!uXGiL~6Rueck>;~Nd1Pp!{76erv5Ig4Fw;3RS;t`;dOOn>sCISEm
zh<=BoYI-~sdwFl+ea+<MmyFnT?@ps4Gdw?Tj*tVRRA=6@@QAx!zDkC2rfsv+mV7HQ
zToN&ITEtG+R3Da<7+mCaW2d&~4c-3xYLbB4O|<g6f-MI6i>qh3Y?X_|P(tt9eTi+X
z4>d0%N*RuG1WPLh8*i=ze1QkbjQ^JTnTbN_-8`P-xz4W66WkO<%il=ljTYl@)iFwF
zWo2c=K8^Qx*EQfpXoZW7-IgX6lAxmdt<D0DjE^sVzv(^+W*$tNx(7qG&g656MnXaY
zx!JY#Ne2;>|DYK`E$_8GT+}C#+{f0Jr*3~YZ8}cU<_<M>velA$d+ntfwKkQg;v^3T
zNx1UQv4}E1Gg3^1J}F}KJJ^N_wF}ip%xVvRucIi$%1pEBuJT@UwfUd+&E-vs?>F_>
z^MdD3j)IyT9p)(1C1n$a|HH${p-~6yR9JTi)W02egnr#NU6+g!#5q9Q_OsbFHD@?P
z3lS#}3#W*DcXq4<uQt*{Z0@Hlixjl9wp|5*z4P(_$`W7lQ0Wdcl=|F|zlKz+0M1L9
z6auqpa(dv7;&)|_Se)#7a7_dWmKHX=n?D~h+WkT0fnkw|OtN?=Q%SmXWspagex?2p
z*!2{g?XbqWpL=?QBc<$Q1>l`|OI!{}DdccmoZ*e?lbTLPTU~Gk3iPo2C=}UU41-o_
z#hWp;U4k<m;e}=it+X}?X4MmNaBxV+Fuw=jB6WDv3@G!R5v0`A)Iop?{Mp(HV{*aR
z1IB>?I21g+ykfa*NfLRT9YUWzBI14$Cf+(#VR?C+d@l_Ok#E#;nWd#)O_RN%0E-A=
zcG*$SluLfxobt+VV>Ay}=>CSy&CLxY-<_`?g3J;atBJ{>^`WyCEiuXO-|a)q>OVk=
zRicwGHIRHe__A9Bz31oL-=B$S2Ya{@?dqzB%^IYED{r?nNkw~v(5$E+e_qB<3fYZa
zZOTHcE(beCBs_(H8RpU5-3{nY<dh#j{L9PPVe7-fA~$^&m^os29H}6B-j|dE0_Q~Y
zKKBB^9*pU@-oFaSvs|dmYMHLjtuxTO%+1XK3S!o8hVVl7daLc1A>1xIc=4RpKdm^2
zK`5!To%hz!t52mJS=j-8=FnLKaneb`h%udJ1Dll9jHG>c>jYBzIBL-;f|nUJ6VesH
ze;cNj6<-`5wq9O0GF)8rWw<iHjxLJ`J~$#`eWAsV|8nKqOP$*CK@Z*{t*UVE+bc{$
z!X7}#Zw2@rYczQaL<rq+TT~D*F);y|bD-9Fn`pZ;j4((wPpoMC+vm?P3u^zS;aI+g
ze?R~a2~OT|1e1<jE{1JOOdaX)W-D+HBl?aNQI6%-F5g(l-LnbZLW)%*M;hcakSe0e
zNKcw<i!Qx9$QfT<Rt{uK&cEp!G?MhuhWKiQyy>N^ftYz|!Kd&X4=cK_J$A7`rwc3G
zo2S-iwEp4s@p#{r`nBU9N*ezy@YaIr!*sgR8u(KcH8nMh`T&RL1|qRWy(`Q0`Tlwj
zQ&T>)_g?Ylj!w#y`Y}9EJYVAU@D#<e8b7S5sR<8KU9OLp*0yDDWH{Wve{z-G=vuzL
z$IWO#CnzhkCJU*ree@HUIg<u&OQx+xj7AyGH@&GD#2Y?Hyq-NR9jH9tbfLdGYZ#O`
zX*S_h^x&`9gi31<JCT1WuRqtjI3w|c&Y-{t2|f_E-tz{=XM_czgLm~<d(HcS_7m2s
zokt6pO6KS56~NPp*j!v^<F@db`f!uwVY2_1iBJ{gz5A929#sC8+^bK7ZEP6f;NW1?
z9&iZ3l+q6$K7jHaT29NvB#||YNkH&v*2WaLJWsC<4iAADqrOEHA{zAA`0nNs#$PL|
zsg;=xJ_q{38qD&z(?Pexi3Y-P%9xuzo2sIh6JO*eVT(nk!9LX0naEe}#u`|1vnMW}
zsfsH2d5x0Di4~lS%{SuC*|(N_UstNxN8X|0K@ip^&~nCi2|;aYJWoA4=+IOV0Va2v
zX;#v@_BVrZ(8FyH(pMh`^cnlo|8Dan3A05JoZTm5zM++VpaJu#SH^v!nxh|lQrGT3
z>%NCuL!U8ZreHpsFt=aP4aAlCMz8vlAt}JX1HBhR;N%G~>rhis1p<Ns_?uv1<9yD8
ziG_s)_+DAuPd41W>JNURfwd1~Ks!~U`rS`Q)nE{sxJnQ3qA$;993$8yanXzM>1pLI
z(*{o~ChKzg=4KaoTa#Wq^7y*NkSBOUF&gCKWb#queZC4Mw&$&6`Zm3inqq8iX^XR1
z-dv{H=rqW6wpy<*D=p87_O>Hu4W6&hVyf-(LO}R%Y<cno6Y~|I3aI6hoQQ_zfL|Ms
z=xa6WNropIfCa8rQB^(arY|aXSd|2~N<MECu+9EebM7I)rS$px_t%PwU(@e_=9SDQ
z(f<sD8u3uPmS9YZevL*4<6FIs^Ld|S@R~ooiB$8xHBJ)rsx>qbf>{)Zea_C#Oj<rX
zK)3NYUh6we^0_@7k5;S&D9@w~INWLbM3<kT;Mxx+g_x8X(Dt?3RA`UTG*6a|?=@cA
zK-LD5<%)D0`U}Qw*3YJGseuOw7Ouy)S>pS8lZ8mr4`)zw`##Q~!SKv+`q+q0b?4g%
zDRis|{-faK5K}id9>5fUDSDIO0_;L78-wXLpdUiKXdD3mNY|y0)Bs>L9-S}J{Y8h~
z^oMcHy#c)wAX<AKw&R8p^Wn%KeCvb)_ykCGd&j!7HJrbY5W&H3QH&D1MG>5r>M%No
zDKBC^xxyuNRR#bMfS(3^!Qww|K_*>TTzuu^<m6}VezwyEECkH4(&&X4sHmyc)puTs
zpFVvu8%zrr-fmF-8(J#2nT>-Debu~o-5@8pPnqD7`LB1%!Y!xOU*ZVfT^NEItY$Hg
zklXX3*s!B%(98&|D&U#sY-db}p22?w8w<lS1E_;fG|X?pv+oZ4*tsfs7u$OGlE9Jd
z-arCe3%|Zl93C7rPfy3gKnE-cgGFgrHVHFK3<zHD7>_PX|2ZXgLb@;<YvJuY7h>O#
z8VY?E7Z<;t`@8cdzPp=g#|tk{&x-{=4A_gna~5R0H?p#g-xcHi{{~Mbx5evl$9_dZ
z;(h<x@=0==igZLqxCBnPVC<l62Il<@cFaB%1Xkp3z;7!=_=Ip4GQQ%D1%O>xpKRW<
zX%a?z`#)hM&w_ghs~dhyV8Tk5zHjX&?H16jx4sZFWxDeI`j+1uvarz)Hfc2|WAz%q
zSW$`QYuF!4@~$6S9t(L0hvfJE_iuU7xX^iXd7P*#<w*7y6n<AR-Id2kJu-|N*6k*d
zQ{Y1AHoB_5dlquFpaq%_1U$}G;%yrjK^KMX8T71<h!yz{f=*Y+8O%*QuO4A$@wxMa
zbyCQ1&9)6?NEHp~)<*rg+$FA|;b9*8B}8yf=|Sne;p~AT{AY$762g2bHy4K>4cOG}
zasn2D+w~wNug3w@s4(3EA2d|}k>haw{w_K=7=_nqL)h8*)S2hupOg0IHf;a&3;<_`
zqm{R%Wo5u+O9jkx-~;bC<8^eBl>X=zgFEWGoG&@o1@c{{#jK%VqQKDPX_WKkPyna(
z)WIxtUMEb)8J`Co2qdq|QWbK{b~mA;g$OAMIsMZnr(2`wjPqrt5yRW|9H18cI;4m+
z>lP35g)EwNC`^fqSHqkm4y<LU*0^h1tZl|9Xf`cJ?RvVWyJn*Vy4&`7+Ni~-G(f+w
zR<&{4<$;;4Tjy9JFa=Lv{dwQ`w&=-CC*~h}^-PUroRyYlg?J6AtjwD=$iJ!)|3i&g
z(*XZ9Q|;xX$C++Vtn#5Px8UIAL2G{XxBG8ZcGSE^nNa_59L{Wcv-9gK3~gq_*RNkM
z?Cpi?7Im7j4>fo|f!{4CIy!pT(t+~NkjJ5i^mTM?If(Gr|JGe%@En*fYADnT>2py?
zhe$;;0!<kXtQ=zKObdu@Zd*B73954cF@@4qat5;#xn*`)BWPlzsetne6$Qmkm!Z!e
z(bux=;&Fp{#(9fOL29?blDwOsr5s6SSe6jfAKE@369FpKHuXNa%h(^$t6URuYWGo(
z%x5w668Ti?5<g5ya;_PF*3lSh^Y}sJBlYG-A{0VGz_6OB>4KA}KRBrK`WorjHOm{(
z!}T2iLKX<~YdaJf%vmkoSApfrjoBoKwQdF_T3rQn>pi^!L}n}dn%XsTNTKOvdT1w_
z#K2w7xng{<ch}eU!2hUf?_>GUHN1-hT~7?um2dkfoP9#hAYPV#{FF*_qk&~QpNI|n
zxaB?Rq=r+PIrNQoIjq%Yd>hQZOw)m+8OH&ELdLGm1e~J(hVk3dD7feUC%$jScCgyt
z%E-V-cd$lbfM)A!LjNWOBy9H0EA3cPS&|OX^j6XIj65HDFs(`uYDQOARMZwyN1v|j
zwTD%nWvf^yhs_);y%8e^=D}*qa$rV0vBVtwah>`)nRqrJdJLVZfRM|-JUuL0Xo4Kr
zae)yvNs~*{<BGMLBt$$Tv&18Ehqm8X&z!_Pj4%=htEy^u3ij#!ooWA~egx7)zSb$H
z=R?#e#otuFO1N*DdphlvA3t6h4EOq-J(5j$*)t7{ebZ>Ahhw!$?+T4QgLGwp|Gw0`
zLKBvSbv+No=Df~z)S{P_|GzMY@5i{se)-Y@H0td9{5F#^7K;&@IUw=?4Zo=IiVo!M
zHY3)-5<`i#{zMNQk|MXvZ3zE>XX_jKE|>!zqR`kfLpq88`8N8qipLzUAWHmt6BE0`
z+PV%RZ_RS3a>k@q<8U~Ko_^?j3|Yt;j<w>BgcAnl(ad&qxRfyq6WJ@(T!yfDM^9iu
z#NwNS2nXauO(J{OcSnZo2MS2~)=@*~BeKd~@}c|QF;K(|@7q8uUQ+2!RXp^YoG=rD
zinSd&+{i;D!$#{5jz8sT*Vsh?f9w#jA*e+!Iz>C9q@<+yN|3nugJSYaRFrtoV`3Uc
z#=`MTK()rQnU<Okkbq9Sx3aRCV2AIKku2szFF>Q_PtZ`OUS&fJ>l*^V?REO#!R2~#
z88D55cCm=$WKvS^3v!6#TJIu&M(co@d<UQ@Fnm1lh0eJlgKEiTXX2HG1sx!LmLt`4
zYkpQ`xM{}4cB^UR0P`y~XUelgPc6u2EtQV6^tubdb25mi7PL)t*tL>RIcZ=iy>8NJ
zi-v@NxK$G>(syb5g|@Ia|296h!3Pa^UpMBTe@JR^hPL%&(=7FkIKS+jfNKWUd;(%|
zqzMhD4?Z`gN%zJXO9hV0E!j!9^$j`ItV-Ux{UgNm6KW2h7T+b}V^6{PmVZ^cv7c6}
zPppLVT{pDtkXWBH6N*1#y%?@6x`1A(dd~94Jmve9Q#Mn?g;R%~*XM>2ayw)K7f#3(
z4(fCRLqi_d12Vv$+^I|{jl;CYzbQ;p6*eeGgA$~xv8G8m<FFphjSVvr&x2P_v9Rz#
zYXY!YTFq2{>Ft#k56$+GC<c`q76CywFo9}RSUkfC2>=H1tMh$~W>`WGVQPE>*e9-e
zPd3q@G0II)?{xR{XjEF^%X1mB{Xg)Jb9*?-_)P}4SgpjQ2CLtzmLDFaXUqjDaXc%W
zWO=}?uDft?d|blCmaT%#_b|ecneZ*WfSBec?Gl34umDYHjIZhr4S%?JhDd?pyOPP0
z<jZM#7F8vK{I%$>{8v@MDSo!hi@I<|T`&6cOwMD!aQeo2?HMV%J_257d6V3C9<MmT
z>ffAC6F|@sa{x^zW9!7PJl5=d+f9~tY<0b@DK9G`t>}N509SwQgTSY|xIbUTwrNDD
z7JwW!uTuBe_w0H9%U}KdE?1TD6relKY{RuoJe?fC5ung<+RjOXnRy9Z_{Gz^z!wgj
z=AQkQI83^|0EmH>2{b^!dmgkPgBybBTesgB4}4p&Hcp^&wtDP0bOKouv=Mq<ECvCP
zn^e^SGc<{#LuO7%)T+m2sJp3SY~B}eD`+4ICx3Q<0Q)uGTqc`%d;qW|!`Zk`T_XaA
zK23H3;zIZV-qR0L#YY!#F1YIsYIWQXE$MTZ$A={MMT6%Ezg6s={=5W3%9dj60s4X<
zbb&$6NMu>1--Gy0GCM5RoMav_2zUmgqqJL;ebG`uYW^X6i-$e+-YrN;JsIi`dR118
zRPe<*=$_XjpdKG(jTx*tg|#^Ixf0ivh#^*0>~_yBearT}?k6(g)7Je}ZsV6zj>n^*
zA|8_^#O3dASTf12sessgvRB2LtMjsGU#CZHvgu$zmx9NHhN`JUn7Rq)^(7K8P}`@K
zp71}T`6gTNt}6w3MbcMhp%h*in893EzDRI8Z9D<GGOXwVQklE;BP0ZXdVx=o)owxf
zF%fq&5OUl$QiFbVcN+jldbsUjPEL-~?o{N^kRr(0pw*&XpSc(mHh=_!1D;4SI=WAw
zlmQ*fq@BN-K?M~n;K7N2^r#v1F>Y;H0Mr`%@ng-`^9+rg${4f7MN!kHsB9I`FGHav
zXIi>C!eV1JnMBg^&ed`jpa%fR%NT@03=CC38L!!Q^4L5is{=i~?`O%U=amV9frDdz
zJmKKe)(!9*#6)0Hr?sieDE13B_TL34ES5a7Ar5PbCrKX_$xMy;k#M`B!o~ko$(WNh
zUxwP#c~4oZ_eSw+XX%b!b_u7TsjzmVq7pjj;KC;&2oo<zOUe8y_c-fdFx7};b0P>N
zITV-kv@wfCVGTjF@SB;9&GKg^6w6ak>Si6ejLA0+-khYwl9@vW7R3!in<E65DK_wQ
zb+a~2c4dwXp#Q1qW|aw$;4GdOcA!)p00A5CFS9NpGBU7&|Cb~&yfATCmX%H5e)Z-J
z3aB(*01gsF7#Pb9)<|f4JRxU%Skv8U5e!^`3J_*j1;ybu-p3f=roaGr-fbH??s4F7
zbksBHyl7C?!QXKpHwSyP0&9NU$-^)A?rT5M#mds->jG!zZR`Km>3Mtl`O~PyDkuY;
zb|x^x#03wE+E(kM|Apy1uByk4$@pO|sA^M%9e_^=R9>G3En2J`?_xE1E0E5Q6P_2a
zAIZW1IVbS_a~0lv9#&f{Ur}|}tDL>-6$^lX9$8R7b>ZrHz6Z2j-sljRP2h@r7Wen}
zPo}j0Cun!5&4kxW^hl3pZz2MnB*PrfVyr>o1_6~54QO1qnr}?>^@RWfn4Zrq2iWje
z(VQoip_ppF6#29u+D>Qzv_KpIF~A=5>~|Wks?a?HI;FFwK;L_5cx+Htc^iUK9>|nf
zL_|AViiLFl@T3ZF@L>`oBclYzZ*|*Yf2-g?<<$%drM~nCK}ih_5cEIVfv^Q+)7naI
z>Oagu+9Zfjpwh|`U_4E$Lc@7U228?)Hr4Mz4^6GhZa3hHQ6D|3m-Lu9!S=xM6)$)<
ze6qcN*DtxSoVk9{+YT##kn{?WYilh3G?%_>VTd#0<E;o(b)WxJcQ<r>md#qO7}tfp
z9Nmfg0zh%BFARxS%ZGK>f~#uPNjIL}9r#md*JFV#anGP(s)E_^zP)hZ&i$v4@8~Us
zu@(6fFDvvpc$y5u2rUYNj;|ZU|0*A?!1jcT7w}rvABV41JrIY^N_O%H;DH<ZVlyil
zaC<zqb+7!8U{4bek09~tuLUz#v%&&!b6!P?k<NdlMIh)|)Y6gO3F|be85%8e0kFe#
zvFJk%^kC3;#GRAYLF<B;QvM@mP}Ja0Z-F)zXb?7B#Ld&z=}<X8>?DPO9vJZfN@*Zb
z1LfA6jD_V3=;DCDT+Oh917(nzAPm-62l#9PNTj5gxj8K`K!U||yz))Xph9zVwvMX~
zy1xc$cvJA?1&9@JZ-eHJ=)WGl&;f)dn12Sw-+`wL7%u({;J-Z?=0BS-R5R9lMhLOu
zL@V3pUi!<McW?tOEKD(nnGAgHj*}tzfAQ8O-nZ@mUuXc(c7L-jRI}fBtq!PP5REfy
zYvX{G!=V+O>vtX0Rze7fYoKsCoHR|g?_sDaHXo)E1c!5<=KW`~z2CZOwgdG0z5PpO
zXw1E)`(rB%W&>;^+^^nVBM`)4QXdS?bazXFAv!!fTm-E|Gaz?@#~;9aKp+QfZf>4l
z@qpX_G&C)nng_n%M?iLVw!1Y*!vDaf$C*$M$nT*1A+bIK<+RkKkARhxwfb@`p12Br
zZ@h?<AS|%CSr~R3YxE)|RrYIwI*YzxauA^7gse(RHN9Ykhdn{%rsMzU+SRh<z6aUl
z`}dLL>WTlOWjE9QuT?&e$7+NEL>g1zZ->B~3<(KDz$yDdHP3jpI})(UTy5>`m{?e?
zuyraUBhw%_Gdp_;d#nXuTQ`AuKiT_=7DzZh=jZP~O#6Twcmb#tiSxZ#Kz_*;b}+*K
z>1cWFm}W1!i|GnT5Z{j754(WcE}pasUk5V}kiMPAK!I$$E?m_3(qxSbT3dm$0=~IF
zk?#^r%Z-T=3P?PcZQ*%sQPbVj;X25*j=*n%w*ULt-{OS4rGVMB`kAx+|L$g8+}xZr
znr?dX7ZvfWVho|sPwGq-XjH(AQC%A<-bDlfu~+1uAGH&UK!&)0>4F7p+sS1!+FxY$
zKGbT|(Q)a)BiDa49id_0gHz8vqM5v69mVe<*clcmXL`>y&?dl5Vrh*T=#?ygb{roz
z{@VojzPaSf7ik||ucaQEOTAh0oY6WF-rbV|tqZa~0y_!gZY~!H47Pd?7hJ7Hx3=z)
zlN6vSF6W&wRf$V|1BQ$|D8j#j#WtoXzh0Z0T6GK^H@yQcC*3kUIPl<<W_@Pu;y?VL
zq@;zNj?%2aTkT}&c4S)h+mvMmw+kN#p4`4uxjBagPourhkFz$DPn?1>t1Xn-Beh7^
zN_E$R4kHducf>SzhhoxV*Qgs@#QQN41O?;^9pfcqQHrs5_z=1<zOgO%+;B>cD}pd6
z;fJ6b<oL~>O+qbx>kEWjAGdl{zS~?Al2{`R9rYWh6ke0aH?DiJlfYe^dft8TLbc}@
zjVn+diA%i8=Flptt^^f*Pwhi}$B~!fkc2}MK(BrI{Ml}FKlx>$TxFa{Zurf>;Y`zh
z9t1Qq;6gw*(V>|mMFn-0;>B%iHoMJ4L5KK%^tN<E2Rc5VVRCeGE$}Qt8d2pG+!lr%
zr?x*B9GK0Lie8z9dovhI&}tu(UAcU`a3Y>>R#o<&xuVTXV}BX=?ZV0Hp#JY|F7Mc)
zmPhR$k)(2^*h1+!H1e2N_kU`C36txk_)aB<ytBS57o3+mW<|Mk>ik_fK5-73k`+Fq
z7POc;%-yhW`9(<_0t=?VF^DJ(L@T<;PvChJ@UW-w0BG?_TpWo)BK!xSyuoT>G3_Vn
z+y*HWbS*<bECq6#FlaZ9;kJJcqRc5{OtJH}DrjANqpXZA$A5_EyL|TXZ~ZT7r`>Ot
zuCi=}o>n!>j;z>#wgAvU93=n;T2<0!l_<+SY7@rJsnXGQru0c`eIG;AA5r(;^yX;o
z&8IF}2V%fA$5h>>ZJ%s&JgZu;!M<b0EY>M*sGw`I(ZZzo{h&P!Em%-#xQL5}Wi2D+
z(Bh-kn;Wrla(}ki{K~weMGSdG>kgxagZ+oiN2@)XLm6n`d|^F#3k&*yEZwfSyv;s|
zVKG#%wj+nEuCBrYJ&^65f&~r8d^o^F7%jDZglz#@q-9aVn(Hm0ZT+igMjhBQfPhCM
zA))r3%Vlq-3n-YIK;xaQgRbWlHP+s6EpjEW!l(S0rv9{@tT)gFCP&<=x=Gb2I<duH
ztuZ$56A$-~aYddAk?;51;qE8}Y}IcMc%x|1IWpyn7x*B<U&&;#U)pA~lVm#!I-hQe
z%6lNb%VQ8+=V*_o9$u=XEXPxT2Deze<Wrg{nn!pUNxgk_9@3^9<E~JFV&6bZL!%E)
z-fFrs48)~lkf#;3J@@uUE~S(B@quSlQ<DH#CVu`Df&44YSK7efIh$$!%1DmV#;H4?
zF%|Mg|3*rv7U{~Enb81;2at{YK$SKO_%8Y>xvWr#Hd>Wj!rvdU<6uuKPZOdQvbWJ{
z)1jOoD|1HB>ENg_E5i>~ZVvAW&uZQFWAubu9CSKm0w^~db+fkSxb8OLW8aRfy{wEK
z_<P-5T&|tMQ=2f!W$}e(pP{oaey(EoaK65GKArc0(OU#LK+Z};C9UY>vNQ0;cGH`C
z^FKN{^Khv5_Ky!PMT8c66izy1$-c&OLXoX=Lc&;!k~P~*h(wOHESc<EvhU1{oscA3
zma&BFgt4y;o_n0<JiqIAo!|9b&-wi`f4Ht0*LV4RKKFgU->)~($4V?QcFf5=I^E%i
z%~MgmSw^92CY``K{T>pM4pKKlk+jl?uR!jceqUD(QrA13#9$-jwixH>)0FiPTZAU9
zw!p=r&}gxzBotuHfrW*l-QC@aply&t2_hp;D=I32yup8jonHqB-8Z~)!+2w0Axb$>
z`glskn36O2Twn&%ylS%Cy)q@7YWMW&!a-Q{auU)Jt=+-+rY(GeJJQ#2+op@UH<A68
zix(e)+k%Rgt2Alowz7-u{6BISd^GCABvy#>+%9+dlOp&ivUaA&PC5`wlBj+AFMTf7
z%2<D&J9!mP-N$9!g?cf;wNCTSujWl~PpcWaFg@RqbY^atA<c43QIN6MyV22&V7p-<
z<FYqAoZWgi#Ab?Xj#9xQzoUiEj%m-mF{>4b!LR4J_7g^GGZfdw$A007R>|$NEI*Jh
zF2Z8BB7eyKV9)G+pst|-=F?b^0DEwbX0<yLbWnnW#tB=3LO!t+043kw9Fl)7FylzZ
zqTKLwntcL-jNYa`8SU2&f~&HxzSgP_bNt3E#nJfe@VByY-IH7QC1N7#yd+~vwEP^B
zVDX;I5Gb@7s7q<sl9F@Fyigf<ARnavU&1y0D-1bqJ0A{%zVRFe+El3uewm4^n_%3G
z!v~J~mRBNfkBhQ;)%oK=lo-Z<FQNsv*c`%YC7$T%=*Pe%zXB#ls#mXHH^U?6S|C=h
z>sWUr%;U4W2LG*D!@&>sM#HiPr=SV|T6=AMon6-H0(m|Ot*|+Gnl_fllEIAHUwF7F
zIep|Kt<clqgh|L&5Ui#=j~`9o?v?;}1q_GafcH8;?;qal993EcDCfO(X*Uo(nw8!Y
z|3m6bfEC`QzkoKNh(qVqe(Ig{h4zcb<!ToDj}iK24%GWSHv?(H(5<Xe%7}WC>g1v7
zCDirNUp;W*xwM*U9dh2%?;b2!oz2fbbUfH5(93eNa7^}>T#~O-|Kwh~zL1LGIhzGq
z5t&kicp?rD4-W_+dx67k%O$IkYg%?0hjTDF<O5|J`k01d2jBx04@v8;3;ib(b#A~F
zlMOL&19B!=8pux2=_F4gkBkgP%<97}ef@x&8N8s@LZ!~8tgH<7tY+9x(BlEDSK{0K
z0|RBA+$3e(8eC#UcNLTGr<d1ve#s@U@hdD>N5PND?K9kb!pJp$7(JPgwkzu?mX@~Z
z^S;REX_VubGfzqYvfxv`VRQM??y`JjnZ@D1oT0?RGst{{%QwQ$rNVDKkaWwOk;3#K
ze16=Gj*f296s>6jN+8iRs#J2r)Kt9`&rt6uwLN_4gmby1WN=sX34-EOi9MZjOrVYm
zF}G1<&$UK#{N1l+8&U`*(uzuq2of6RW*8-cDlS{Yedm!oA3|I0I;0d%Oi1<7_~I2Y
z>!#8nniO<UrKLUmQ>i)<`!KSe4As|Osw(siZm(<;I^>eAWsaC6@vJ^u*i~fFYd;Aa
zV^uSw=*o}`RU<mi&SLV39nRAYA%N3DXWdrzh!F1QM+t}5{hz*_m~)kf`Hbki)KO`S
z4>ihvr=M$3W;b}cYD9$Rn|PL9j^^Hs&mS73Nak<ZBJ(@t_{ZW$|MM^DjKwa9UR&W#
zA69Y_BCto_#?li1WWJf$T$>5yEYLv~=>IL^+ftWGNJM~V<{eqwgKyJ<r`2$?j@;|Y
zMp#n$f+^fQ86p=U^U;LUoz&9DOmivf@AzB)bK+z%hLVNyTPIFctXNl;%MV-)`d`<4
zTJ7Wv77cxF5Mrv)#x`lsYd+>)Spz`FD<P7qexoTIpeDV-y0`n86_3U{j^U;!3#RwG
zi$lZh19Z|YRj!}X^)7uGHC+yDa*qC@k2BN7KI=Un{Oha>x-(z<!tl4uC^zIEw>{e3
z#tH!+MCbb~;@M}eGrIUXZUuMo6S+GrFX(G^6#C{4JRHW`hjAX`GaoO2a@i7`%eFp-
zB>UiBo=QYp+&WvlY{#FT^*wU*4ygIFy{CLZU4o7)aWI4F&g(#f#+(hWa7<r9sfggm
zsKWEa08D&LYa8P&!-rGpV$SqI3<U(@pyRgl*u=<J97<-{L8*TZJIcw`=s#+|Yq_j(
zP(J@NUsc%@KuMBAPr;P$O20a@pDa#6vrag(IQ_v>*?JA5f1GxsSIkpS2<Kr>>pN&W
zyMDDoaaWrI2(X)Wm(K$YbzF&fa+YMEL?rF>hS7&Gj2}jpqGyJ^;Vy^9V~KSEj96<3
zHQ6z?Y+0`hf@UBx6hWgFUreMy?DJ9v(QM@5h%fhtQhTyuYo^%%3V3zbq@6KpZ*3!N
zY4FvrQx9-k>O688Qf$K5>$pjH7^&PAXZ2%I-{ZXzUx%`(sra6&KUmVVHEjkLRWFVU
z-1%+W$;Zq+-34mKezX=V0n*SaqZZjUgn2!``*te2WI>hv;0YI(O7$CQJe5)TiSnHB
zxTHBkxVfmzsG)~HfpEUTI#Aam$th3S{z{2duD_Ubs^X8Or#BycW<|*aFcBi#I4BHj
zz2uOUt~@a0dD!#r)VXt7V2Y@betzsEuYf@7s@LZ?VnQZ(CSMQ{xIGueK&A+K`O*ez
zG6ZnS;?5JQ<iCW4O_OIi)d%1?Nda{kdUdiVUq@%;$|BTl?24!>HBUCvAPvL|97C8=
z&1(tiDn&BBR<0K^%Q$?)c}zElGU`xt`iJ<ieNhJ9Vk`CM-fG9tX?<C8tML&p628Pq
z?4y%=l)7%;cq#q#?ab@jp8SSy^6HM!Gv>$fbT<sxxl42JOND+N67csUINV?8p+cbM
z)@z~(MjUzBHpV`P>&@7-z5EEh{YmLsjxjn%kFa#@$L_R5qmwc>sQ-Wr+<?!Zrl!_P
zkd3e(#&!cy4ug!_5Pu{}dO!9CP*tF-;mGZ^l?n(5KvW;>Ex^F@%a<=%+1VELz7*sj
z6o~YJ1Gg2zH@X%!!Gw}-kSlQcvelQj8hNHQap2bglMGA>j)T!EQ#)Q7Q~Bsx0w#FR
zEMq%KXm;LlD*|bx4HpPYw+ZkTUg&!CB-c`n>*_v`?}kC{cj_0VB*ii>iBy@9Yu;#!
zW{SvCO|FIC9)H4bPM)nSTbXGemS69TKpH0hs(P1dY&2h50Rb3Z)nQfCjVFOF9wtBE
zy{Jz@Pyd2=+_r!{&pd_hSb#SHtI174E{sG(!r^&Qt*kUMPR`7@Yk23GR7Aqnq!|$G
zm413lH17HT6_;e@n*vNeD{{mCF0I>bYkHBaCtS+1PW!r3xSfAGPVofM;*V!OzP79T
zSUQ)HW#z*jJp-%3#8SaU!7K;=dhPTN`=3`gK2O*R8F1R~Z-yWEjG3t{ZtZn8k`#C5
z=Q-I2lwP$QZqt?ee9ptw;)j|l`2xI(q(X#Sf`jPSES&-?<>+9cc;pW`tACq|T`JI0
zZV#r~6;hNSo#!Gg{N5BB>U-0z7SHRCXF-<BA&ekG$x-Rvqeq-c^Il3ehdmH%^&j$u
zL<b1b0=M6Y-w&Rw=oq02LRgyLyTJ9lhITcFt{Aqc6qQ%3>mKK3C!;5|xY=5#;(5i4
z?6ekpSbXJ-7pzBofMH{p;gmJfQ!~jDMH=ssw?HQgq?4@V5(_Wo)%mq8ez+Ya_m(xd
ziG4A0KEGeA`||PRIE7uCpv7|Qn#=c3f(1Wj_J`}4W@dj`lc0hPaDA}gj9#2jsOAH0
z0}UyKaLnpx7Mq~0@HycEUh%FdSUO}HH$fV)vO0j`-*9M4j>#a?n6Q7XeD>W64Rh26
zG?j6KhRR&!da6~t{r-1s@?Ddq-fbl0%K`$}1w}NhfpcRetT5qky(gcE1W|Yg<XsA~
zECU0`a5oTfFG1zlWSy(<4!y-hei>qweUxA*o|7ZUB5b4zeHsWmB$5|7ItW_NPrejX
z163ZPWIy7+qJ#Eyw6*PvxOq6%)R&!RM$YK!IgQ}9vY*ReSx!Qw5INTd03@<6S!<Xx
zG(uIa_sTrXQ<k8(B3$Sv=A6IVIpQ3;+!wRV$n3PxBA^$~5gyZdacQ;`c?p$rgRRFQ
z$htk1l4x9A)qrj8<9DBp>8=S&wmQorI1o}K@yy(-5bL#Su3L*Lx}>`if=DU)(xg)J
z{*3~ExXrOL8$!@y;@+~zjSIUqp=pA|T{{^mYtF;kT+UO^lBz3bm|U>;6^1rKXzK43
z7w;N^>61)P0$jfRJd|~72{#5}hv+1c*imHBxC3dmvf!`+lnHnT7}5L#vLPT7A{~p3
zf%!WS6XZ!OnF(iRW+vCL@HT?|Zw>~+;7LsXt8rTg?5&V5o0^kz0tU&hu97g&ukY-9
z3E~Vn_ZK!d@E<}I8x<W5pDpO^UuKEVHrZpZ{2W)<;z(md96)_|iVe9xD~CC>eK<ea
z5p?Ipn%j^7(X1r16(<^p5m7j3e0H8-@@L$Sd^WFhhR9{aQ6=n!AhY@z7}E(DE$E06
zZDDd4S5?YNDVQTJX5NA2A3GSNqi~C_h~*1-%yS_l3gNV0ktc^B+X3JhX|^q&fJ}tQ
zi<${Y3{$X8@!&jZiyvcuI4u>3Lbn4XZ${(nhBQE+AW{xui7<)C5VAfH_xo3Rbtp9d
zQXC%DuYWJ~8`%%xpT~i$4OciIXK??4vw!TnNOS479fIZref5NbF)dWd1i`-*_<0{U
zyi_3ao*Cx4@UY3~4oxZ>oGWmofSIi0mfaDg60P-r<3<0gAph?!DCO`tXT4hLWZrvr
Q2x&)XsOhK{Tr>CkC%kV#9smFU

literal 0
HcmV?d00001

diff --git a/tests/tools/cohort_tracking/test_cohort_tracking.py b/tests/tools/cohort_tracking/test_cohort_tracking.py
index 5729081b..f09364df 100644
--- a/tests/tools/cohort_tracking/test_cohort_tracking.py
+++ b/tests/tools/cohort_tracking/test_cohort_tracking.py
@@ -1,7 +1,5 @@
 from pathlib import Path
 
-import numpy as np
-import pandas as pd
 import pytest
 
 import ehrapy as ep
@@ -9,76 +7,68 @@
 
 CURRENT_DIR = Path(__file__).parent
 _TEST_DATA_PATH = f"{CURRENT_DIR.parent}/test_data_features_ranking"
+_TEST_IMAGE_PATH = f"{CURRENT_DIR.parent}/_images"
 
 
 @pytest.fixture
-def mini_adata():
+def adata_mini():
     return read_csv(f"{_TEST_DATA_PATH}/dataset1.csv", columns_obs_only=["glucose", "weight", "disease", "station"])
 
 
 @pytest.mark.parametrize("columns", [None, ["glucose", "weight", "disease", "station"]])
-def test_CohortTracker_init_vanilla(columns, mini_adata):
-    ct = ep.tl.CohortTracker(mini_adata, columns)
+def test_CohortTracker_init_vanilla(columns, adata_mini):
+    ct = ep.tl.CohortTracker(adata_mini, columns)
     assert ct._tracked_steps == 0
     assert ct.tracked_steps == 0
     assert ct._tracked_text == []
     assert ct._tracked_operations == []
 
 
-def test_CohortTracker_type_detection(mini_adata):
-    ct = ep.tl.CohortTracker(mini_adata, ["glucose", "weight", "disease", "station"])
+def test_CohortTracker_type_detection(adata_mini):
+    ct = ep.tl.CohortTracker(adata_mini, ["glucose", "weight", "disease", "station"])
     assert set(ct.categorical) == {"disease", "station"}
 
 
-def test_CohortTracker_init_set_columns(mini_adata):
+def test_CohortTracker_init_set_columns(adata_mini):
     # limit columns
-    ep.tl.CohortTracker(mini_adata, columns=["glucose", "disease"])
-
-    # TODO: check plot?
+    ep.tl.CohortTracker(adata_mini, columns=["glucose", "disease"])
 
     # invalid column
     with pytest.raises(ValueError):
         ep.tl.CohortTracker(
-            mini_adata,
+            adata_mini,
             columns=["glucose", "disease", "non_existing_column"],
         )
 
     # force categoricalization
-    ep.tl.CohortTracker(mini_adata, columns=["glucose", "disease"], categorical=["glucose", "disease"])
-
-    # TODO: check plot?
+    ep.tl.CohortTracker(adata_mini, columns=["glucose", "disease"], categorical=["glucose", "disease"])
 
     # invalid category
     with pytest.raises(ValueError):
         ep.tl.CohortTracker(
-            mini_adata,
+            adata_mini,
             columns=["glucose", "disease"],
             categorical=["station"],
         )
 
 
-def test_CohortTracker_call(mini_adata):
-    ct = ep.tl.CohortTracker(mini_adata)
+def test_CohortTracker_call(adata_mini):
+    ct = ep.tl.CohortTracker(adata_mini)
 
-    ct(mini_adata)
+    ct(adata_mini)
     assert ct.tracked_steps == 1
     assert ct._tracked_text == ["Cohort 0\n (n=12)"]
 
-    # TODO: check plot?
-
-    ct(mini_adata)
+    ct(adata_mini)
     assert ct.tracked_steps == 2
     assert ct._tracked_text == ["Cohort 0\n (n=12)", "Cohort 1\n (n=12)"]
 
 
-# TODO: check plot?
-
-
-def test_CohortTracker_reset(mini_adata):
-    ct = ep.tl.CohortTracker(mini_adata)
+def test_CohortTracker_reset(adata_mini):
+    ct = ep.tl.CohortTracker(adata_mini)
 
-    ct(mini_adata)
-    ct(mini_adata)
+    ct(adata_mini)
+    ct(adata_mini)
 
     ct.reset()
     assert ct.tracked_steps == 0
@@ -86,19 +76,70 @@ def test_CohortTracker_reset(mini_adata):
     assert ct._tracked_operations == []
 
 
-def test_CohortTracker_flowchart(mini_adata):
-    ct = ep.tl.CohortTracker(mini_adata)
+def test_CohortTracker_plot_cohort_change_test_sensitivity(adata_mini, check_same_image):
+    ct = ep.tl.CohortTracker(adata_mini)
+
+    # check that e.g. different color triggers error
+    ct(adata_mini, label="First step", operations_done="Some operations")
+    fig1, _ = ct.plot_cohort_change(show=False, color_palette="husl")
+
+    with pytest.raises(AssertionError):
+        check_same_image(
+            fig=fig1,
+            base_path=f"{_TEST_IMAGE_PATH}/cohorttracker_adata_mini_step1",
+            tol=1e-1,
+        )
+
+
+def test_CohortTracker_plot_cohort_change(adata_mini, check_same_image):
+    ct = ep.tl.CohortTracker(adata_mini)
 
-    ct(mini_adata, label="First step", operations_done="Some operations")
-    ct(mini_adata, label="Second step", operations_done="Some other operations")
+    ct(adata_mini, label="First step", operations_done="Some operations")
+    fig1, _ = ct.plot_cohort_change(show=False)
+
+    check_same_image(
+        fig=fig1,
+        base_path=f"{_TEST_IMAGE_PATH}/cohorttracker_adata_mini_step1",
+        tol=1e-1,
+    )
+
+    ct(adata_mini, label="Second step", operations_done="Some other operations")
+    fig2, _ = ct.plot_cohort_change(show=False)
+
+    check_same_image(
+        fig=fig2,
+        base_path=f"{_TEST_IMAGE_PATH}/cohorttracker_adata_mini_step2",
+        tol=1e-1,
+    )
+
+
+def test_CohortTracker_flowchart_sensitivity(adata_mini, check_same_image):
+    ct = ep.tl.CohortTracker(adata_mini)
+
+    ct(adata_mini, label="Base Cohort")
+    ct(adata_mini, operations_done="Some processing")
+
+    # check that e.g. different arrow size triggers error
+    fig, _ = ct.plot_flowchart(show=False, arrow_size=0.5)
+
+    with pytest.raises(AssertionError):
+        check_same_image(
+            fig=fig,
+            base_path=f"{_TEST_IMAGE_PATH}/cohorttracker_adata_mini_flowchart",
+            tol=1e-1,
+        )
 
-    ct.plot_flowchart()
 
+def test_CohortTracker_flowchart(adata_mini, check_same_image):
+    ct = ep.tl.CohortTracker(adata_mini)
 
-def test_CohortTracker_plot_cohort_change(mini_adata):
-    ct = ep.tl.CohortTracker(mini_adata)
+    ct(adata_mini, label="Base Cohort")
+    ct(adata_mini, operations_done="Some processing")
 
-    ct(mini_adata)
-    ct(mini_adata)
+    fig, _ = ct.plot_flowchart(show=False)
 
-    ct.plot_cohort_change(return_figure=True)
+    check_same_image(
+        fig=fig,
+        base_path=f"{_TEST_IMAGE_PATH}/cohorttracker_adata_mini_flowchart",
+        tol=1e-1,
+    )

From c8ff205330e0c35979f817e615e2aad14efb2eaa Mon Sep 17 00:00:00 2001
From: Eljas Roellin <65244425+eroell@users.noreply.github.com>
Date: Sat, 9 Mar 2024 09:46:38 +0100
Subject: [PATCH 17/46] Update ehrapy/tools/cohort_tracking/_cohort_tracker.py

Co-authored-by: Lukas Heumos <lukas.heumos@posteo.net>
---
 ehrapy/tools/cohort_tracking/_cohort_tracker.py | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/ehrapy/tools/cohort_tracking/_cohort_tracker.py b/ehrapy/tools/cohort_tracking/_cohort_tracker.py
index 7ac56e07..df53ee66 100644
--- a/ehrapy/tools/cohort_tracking/_cohort_tracker.py
+++ b/ehrapy/tools/cohort_tracking/_cohort_tracker.py
@@ -117,7 +117,7 @@ def reset(self) -> None:
 
     @property
     def tracked_steps(self):
-        """int: Number of tracked steps."""
+        """Number of tracked steps."""
         return self._tracked_steps
 
     @property

From 3c7cdc99b000deec86c7b2b02c05da4fca7f596f Mon Sep 17 00:00:00 2001
From: Eljas Roellin <65244425+eroell@users.noreply.github.com>
Date: Sat, 9 Mar 2024 09:46:47 +0100
Subject: [PATCH 18/46] Update ehrapy/tools/cohort_tracking/_cohort_tracker.py

Co-authored-by: Lukas Heumos <lukas.heumos@posteo.net>
---
 ehrapy/tools/cohort_tracking/_cohort_tracker.py | 1 -
 1 file changed, 1 deletion(-)

diff --git a/ehrapy/tools/cohort_tracking/_cohort_tracker.py b/ehrapy/tools/cohort_tracking/_cohort_tracker.py
index df53ee66..dd754a65 100644
--- a/ehrapy/tools/cohort_tracking/_cohort_tracker.py
+++ b/ehrapy/tools/cohort_tracking/_cohort_tracker.py
@@ -323,7 +323,6 @@ def plot_flowchart(
             .. image:: /_static/docstring_previews/flowchart.png
         """
 
-        # Create figure and axes
         if ax is None:
             fig, axes = plt.subplots()
         else:

From 088983c7914537e67a6af8dc21c40a6033c10189 Mon Sep 17 00:00:00 2001
From: Eljas Roellin <65244425+eroell@users.noreply.github.com>
Date: Sat, 9 Mar 2024 09:47:43 +0100
Subject: [PATCH 19/46] Update ehrapy/tools/cohort_tracking/_cohort_tracker.py

Co-authored-by: Lukas Heumos <lukas.heumos@posteo.net>
---
 ehrapy/tools/cohort_tracking/_cohort_tracker.py | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/ehrapy/tools/cohort_tracking/_cohort_tracker.py b/ehrapy/tools/cohort_tracking/_cohort_tracker.py
index dd754a65..cd89f586 100644
--- a/ehrapy/tools/cohort_tracking/_cohort_tracker.py
+++ b/ehrapy/tools/cohort_tracking/_cohort_tracker.py
@@ -122,7 +122,7 @@ def tracked_steps(self):
 
     @property
     def track_t1(self):
-        """list: List of :class:`~tableone.TableOne` objects of each logging step."""
+        """List of :class:`~tableone.TableOne` objects of each logging step."""
         return self._track_t1
 
     def plot_cohort_change(

From e9dedbb53b50bc40aae3626fb9149c53f5571edc Mon Sep 17 00:00:00 2001
From: Eljas Roellin <65244425+eroell@users.noreply.github.com>
Date: Sat, 9 Mar 2024 09:48:01 +0100
Subject: [PATCH 20/46] Update ehrapy/tools/cohort_tracking/_cohort_tracker.py

Co-authored-by: Lukas Heumos <lukas.heumos@posteo.net>
---
 ehrapy/tools/cohort_tracking/_cohort_tracker.py | 1 -
 1 file changed, 1 deletion(-)

diff --git a/ehrapy/tools/cohort_tracking/_cohort_tracker.py b/ehrapy/tools/cohort_tracking/_cohort_tracker.py
index cd89f586..d098030a 100644
--- a/ehrapy/tools/cohort_tracking/_cohort_tracker.py
+++ b/ehrapy/tools/cohort_tracking/_cohort_tracker.py
@@ -336,7 +336,6 @@ def plot_flowchart(
         max_pos = min(0.3 * self.tracked_steps, 1)
         y_positions = np.linspace(max_pos, 0, self.tracked_steps)
 
-        # Define node labels
         node_labels = self._tracked_text
 
         # Draw nodes

From 711c7101097adce5b045f642b3413d2dcd4c6774 Mon Sep 17 00:00:00 2001
From: Eljas Roellin <65244425+eroell@users.noreply.github.com>
Date: Sat, 9 Mar 2024 09:48:53 +0100
Subject: [PATCH 21/46] Update ehrapy/tools/cohort_tracking/_cohort_tracker.py

Co-authored-by: Lukas Heumos <lukas.heumos@posteo.net>
---
 ehrapy/tools/cohort_tracking/_cohort_tracker.py | 1 -
 1 file changed, 1 deletion(-)

diff --git a/ehrapy/tools/cohort_tracking/_cohort_tracker.py b/ehrapy/tools/cohort_tracking/_cohort_tracker.py
index d098030a..9789485c 100644
--- a/ehrapy/tools/cohort_tracking/_cohort_tracker.py
+++ b/ehrapy/tools/cohort_tracking/_cohort_tracker.py
@@ -380,7 +380,6 @@ def plot_flowchart(
         axes.set_ylim(0, 1.1)
 
         axes.set_axis_off()
-        # Show or return plotting handles
         if show:
             plt.show()
             return None

From 7f25d6633dc39152f93efc103f4c29d6cd12415d Mon Sep 17 00:00:00 2001
From: Eljas Roellin <65244425+eroell@users.noreply.github.com>
Date: Sat, 9 Mar 2024 09:51:01 +0100
Subject: [PATCH 22/46] Update ehrapy/tools/cohort_tracking/_cohort_tracker.py

Co-authored-by: Lukas Heumos <lukas.heumos@posteo.net>
---
 ehrapy/tools/cohort_tracking/_cohort_tracker.py | 1 -
 1 file changed, 1 deletion(-)

diff --git a/ehrapy/tools/cohort_tracking/_cohort_tracker.py b/ehrapy/tools/cohort_tracking/_cohort_tracker.py
index 9789485c..1c7c0e7b 100644
--- a/ehrapy/tools/cohort_tracking/_cohort_tracker.py
+++ b/ehrapy/tools/cohort_tracking/_cohort_tracker.py
@@ -162,7 +162,6 @@ def plot_cohort_change(
 
             .. image:: /_static/docstring_previews/cohort_tracking.png
         """
-        # Plotting
         subplots_kwargs = {} if subplots_kwargs is None else subplots_kwargs
 
         if ax is None:

From 331c095342f91e5b26c004d5fd71e0c4801b19c8 Mon Sep 17 00:00:00 2001
From: Eljas Roellin <65244425+eroell@users.noreply.github.com>
Date: Sat, 9 Mar 2024 09:52:08 +0100
Subject: [PATCH 23/46] Update ehrapy/tools/cohort_tracking/_cohort_tracker.py

Co-authored-by: Lukas Heumos <lukas.heumos@posteo.net>
---
 ehrapy/tools/cohort_tracking/_cohort_tracker.py | 1 -
 1 file changed, 1 deletion(-)

diff --git a/ehrapy/tools/cohort_tracking/_cohort_tracker.py b/ehrapy/tools/cohort_tracking/_cohort_tracker.py
index 1c7c0e7b..edaf0338 100644
--- a/ehrapy/tools/cohort_tracking/_cohort_tracker.py
+++ b/ehrapy/tools/cohort_tracking/_cohort_tracker.py
@@ -247,7 +247,6 @@ def plot_cohort_change(
                         color="white",
                         fontweight="bold",
                     )
-                    # legend_labels.append(col)
                     if idx == 0:
                         legend_labels.append([Patch(color=level_color, label=col)])
 

From ee0b68fa261e3a3cc4ba344e16f10d0ff69f0915 Mon Sep 17 00:00:00 2001
From: Eljas Roellin <65244425+eroell@users.noreply.github.com>
Date: Sat, 9 Mar 2024 09:52:33 +0100
Subject: [PATCH 24/46] Update ehrapy/tools/cohort_tracking/_cohort_tracker.py

Co-authored-by: Lukas Heumos <lukas.heumos@posteo.net>
---
 ehrapy/tools/cohort_tracking/_cohort_tracker.py | 1 -
 1 file changed, 1 deletion(-)

diff --git a/ehrapy/tools/cohort_tracking/_cohort_tracker.py b/ehrapy/tools/cohort_tracking/_cohort_tracker.py
index edaf0338..2069b5e4 100644
--- a/ehrapy/tools/cohort_tracking/_cohort_tracker.py
+++ b/ehrapy/tools/cohort_tracking/_cohort_tracker.py
@@ -320,7 +320,6 @@ def plot_flowchart(
 
             .. image:: /_static/docstring_previews/flowchart.png
         """
-
         if ax is None:
             fig, axes = plt.subplots()
         else:

From 7472eaa5b5278af715436f2733d7eb71a999b057 Mon Sep 17 00:00:00 2001
From: Eljas Roellin <eljas.roellin@ikmail.com>
Date: Sat, 9 Mar 2024 10:03:13 +0100
Subject: [PATCH 25/46] remove reset, add updated notebook for quick check

---
 cohort_tracking.ipynb                         | 151 ++----------------
 .../tools/cohort_tracking/_cohort_tracker.py  |  10 --
 .../cohort_tracking/test_cohort_tracking.py   |  12 --
 3 files changed, 11 insertions(+), 162 deletions(-)

diff --git a/cohort_tracking.ipynb b/cohort_tracking.ipynb
index 6ad27c3d..e916ec62 100644
--- a/cohort_tracking.ipynb
+++ b/cohort_tracking.ipynb
@@ -10,7 +10,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 2,
+   "execution_count": 6,
    "metadata": {},
    "outputs": [],
    "source": [
@@ -33,14 +33,14 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 15,
+   "execution_count": 7,
    "metadata": {},
    "outputs": [
     {
      "name": "stdout",
      "output_type": "stream",
      "text": [
-      "\u001b[1;35m2024-02-15 09:11:49,986\u001b[0m - \u001b[1;34mroot\u001b[0m \u001b[1;37mINFO - Transformed passed DataFrame into an AnnData object with n_obs x n_vars = `101766` x `48`.\u001b[0m\n"
+      "\u001b[1;35m2024-03-09 10:00:23,246\u001b[0m - \u001b[1;34mroot\u001b[0m \u001b[1;37mINFO - Transformed passed DataFrame into an AnnData object with n_obs x n_vars = `101766` x `48`.\u001b[0m\n"
      ]
     },
     {
@@ -214,7 +214,7 @@
        "                             9                          3002 (2.9)"
       ]
      },
-     "execution_count": 15,
+     "execution_count": 7,
      "metadata": {},
      "output_type": "execute_result"
     }
@@ -235,19 +235,12 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 16,
+   "execution_count": 8,
    "metadata": {},
    "outputs": [
-    {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      "{'gender': {0.0: [53.8, 52.4], 1.0: [46.2, 47.6]}, 'race': {'AfricanAmerican': [19.3, 25.9], 'Asian': [0.6, 0.7], 'Caucasian': [76.5, 70.4], 'Hispanic': [2.0, 0.8], 'Other': [1.5, 2.2]}, 'time_in_hospital_days': ['4.4 (3.0)', '4.6 (3.2)']}\n"
-     ]
-    },
     {
      "data": {
-      "image/png": "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",
+      "image/png": "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",
       "text/plain": [
        "<Figure size 640x480 with 2 Axes>"
       ]
@@ -257,48 +250,13 @@
     },
     {
      "data": {
-      "image/svg+xml": [
-       "<?xml version=\"1.0\" encoding=\"UTF-8\" standalone=\"no\"?>\n",
-       "<!DOCTYPE svg PUBLIC \"-//W3C//DTD SVG 1.1//EN\"\n",
-       " \"http://www.w3.org/Graphics/SVG/1.1/DTD/svg11.dtd\">\n",
-       "<!-- Generated by graphviz version 9.0.0 (20230911.1827)\n",
-       " -->\n",
-       "<!-- Pages: 1 -->\n",
-       "<svg width=\"201pt\" height=\"143pt\"\n",
-       " viewBox=\"0.00 0.00 200.50 142.50\" xmlns=\"http://www.w3.org/2000/svg\" xmlns:xlink=\"http://www.w3.org/1999/xlink\">\n",
-       "<g id=\"graph0\" class=\"graph\" transform=\"scale(1 1) rotate(0) translate(4 138.5)\">\n",
-       "<polygon fill=\"white\" stroke=\"none\" points=\"-4,4 -4,-138.5 196.5,-138.5 196.5,4 -4,4\"/>\n",
-       "<!-- 0 -->\n",
-       "<g id=\"node1\" class=\"node\">\n",
-       "<title>0</title>\n",
-       "<polygon fill=\"lightgrey\" stroke=\"black\" points=\"86.5,-134.5 0,-134.5 0,-93.5 86.5,-93.5 86.5,-134.5\"/>\n",
-       "<text text-anchor=\"middle\" x=\"43.25\" y=\"-117.2\" font-family=\"Times,serif\" font-size=\"14.00\">Initial cohort</text>\n",
-       "<text text-anchor=\"middle\" x=\"43.25\" y=\"-100.7\" font-family=\"Times,serif\" font-size=\"14.00\"> (n=101766)</text>\n",
-       "</g>\n",
-       "<!-- 1 -->\n",
-       "<g id=\"node2\" class=\"node\">\n",
-       "<title>1</title>\n",
-       "<polygon fill=\"lightgrey\" stroke=\"black\" points=\"78.62,-41 7.88,-41 7.88,0 78.62,0 78.62,-41\"/>\n",
-       "<text text-anchor=\"middle\" x=\"43.25\" y=\"-23.7\" font-family=\"Times,serif\" font-size=\"14.00\">Cohort 1</text>\n",
-       "<text text-anchor=\"middle\" x=\"43.25\" y=\"-7.2\" font-family=\"Times,serif\" font-size=\"14.00\"> (n=1000)</text>\n",
-       "</g>\n",
-       "<!-- 0&#45;&gt;1 -->\n",
-       "<g id=\"edge1\" class=\"edge\">\n",
-       "<title>0&#45;&gt;1</title>\n",
-       "<path fill=\"none\" stroke=\"black\" d=\"M43.25,-93.15C43.25,-81.28 43.25,-65.94 43.25,-52.5\"/>\n",
-       "<polygon fill=\"black\" stroke=\"black\" points=\"46.75,-52.83 43.25,-42.83 39.75,-52.83 46.75,-52.83\"/>\n",
-       "<text text-anchor=\"middle\" x=\"117.88\" y=\"-62.2\" font-family=\"Times,serif\" font-size=\"14.00\">filtered to first 1000 entries</text>\n",
-       "</g>\n",
-       "</g>\n",
-       "</svg>\n"
-      ],
+      "image/png": "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",
       "text/plain": [
-       "<graphviz.graphs.Digraph at 0x2bedb7f90>"
+       "<Figure size 640x480 with 1 Axes>"
       ]
      },
-     "execution_count": 16,
      "metadata": {},
-     "output_type": "execute_result"
+     "output_type": "display_data"
     }
    ],
    "source": [
@@ -314,8 +272,6 @@
     "# track the filtered dataset\n",
     "pop_track(adata, label=\"Cohort 1\", operations_done=\"filtered to first 1000 entries\")\n",
     "\n",
-    "print(pop_track.track)\n",
-    "\n",
     "# plot the change of the cohort\n",
     "pop_track.plot_cohort_change()\n",
     "\n",
@@ -323,91 +279,6 @@
     "pop_track.plot_flowchart()"
    ]
   },
-  {
-   "cell_type": "markdown",
-   "metadata": {},
-   "source": [
-    "The tracking steps can be reset for convenience"
-   ]
-  },
-  {
-   "cell_type": "code",
-   "execution_count": 17,
-   "metadata": {},
-   "outputs": [],
-   "source": [
-    "pop_track.reset()"
-   ]
-  },
-  {
-   "cell_type": "code",
-   "execution_count": 18,
-   "metadata": {},
-   "outputs": [
-    {
-     "data": {
-      "image/png": "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",
-      "text/plain": [
-       "<Figure size 640x480 with 2 Axes>"
-      ]
-     },
-     "metadata": {},
-     "output_type": "display_data"
-    },
-    {
-     "data": {
-      "image/svg+xml": [
-       "<?xml version=\"1.0\" encoding=\"UTF-8\" standalone=\"no\"?>\n",
-       "<!DOCTYPE svg PUBLIC \"-//W3C//DTD SVG 1.1//EN\"\n",
-       " \"http://www.w3.org/Graphics/SVG/1.1/DTD/svg11.dtd\">\n",
-       "<!-- Generated by graphviz version 9.0.0 (20230911.1827)\n",
-       " -->\n",
-       "<!-- Pages: 1 -->\n",
-       "<svg width=\"227pt\" height=\"143pt\"\n",
-       " viewBox=\"0.00 0.00 226.75 142.50\" xmlns=\"http://www.w3.org/2000/svg\" xmlns:xlink=\"http://www.w3.org/1999/xlink\">\n",
-       "<g id=\"graph0\" class=\"graph\" transform=\"scale(1 1) rotate(0) translate(4 138.5)\">\n",
-       "<polygon fill=\"white\" stroke=\"none\" points=\"-4,4 -4,-138.5 222.75,-138.5 222.75,4 -4,4\"/>\n",
-       "<!-- 0 -->\n",
-       "<g id=\"node1\" class=\"node\">\n",
-       "<title>0</title>\n",
-       "<polygon fill=\"lightgrey\" stroke=\"black\" points=\"152.5,-134.5 0,-134.5 0,-93.5 152.5,-93.5 152.5,-134.5\"/>\n",
-       "<text text-anchor=\"middle\" x=\"76.25\" y=\"-117.2\" font-family=\"Times,serif\" font-size=\"14.00\">Restarted cohort analysis</text>\n",
-       "<text text-anchor=\"middle\" x=\"76.25\" y=\"-100.7\" font-family=\"Times,serif\" font-size=\"14.00\"> (n=1000)</text>\n",
-       "</g>\n",
-       "<!-- 1 -->\n",
-       "<g id=\"node2\" class=\"node\">\n",
-       "<title>1</title>\n",
-       "<polygon fill=\"lightgrey\" stroke=\"black\" points=\"108.25,-41 44.25,-41 44.25,0 108.25,0 108.25,-41\"/>\n",
-       "<text text-anchor=\"middle\" x=\"76.25\" y=\"-23.7\" font-family=\"Times,serif\" font-size=\"14.00\">Cohort 2</text>\n",
-       "<text text-anchor=\"middle\" x=\"76.25\" y=\"-7.2\" font-family=\"Times,serif\" font-size=\"14.00\"> (n=500)</text>\n",
-       "</g>\n",
-       "<!-- 0&#45;&gt;1 -->\n",
-       "<g id=\"edge1\" class=\"edge\">\n",
-       "<title>0&#45;&gt;1</title>\n",
-       "<path fill=\"none\" stroke=\"black\" d=\"M76.25,-93.15C76.25,-81.28 76.25,-65.94 76.25,-52.5\"/>\n",
-       "<polygon fill=\"black\" stroke=\"black\" points=\"79.75,-52.83 76.25,-42.83 72.75,-52.83 79.75,-52.83\"/>\n",
-       "<text text-anchor=\"middle\" x=\"147.5\" y=\"-62.2\" font-family=\"Times,serif\" font-size=\"14.00\">filtered to first 500 entries</text>\n",
-       "</g>\n",
-       "</g>\n",
-       "</svg>\n"
-      ],
-      "text/plain": [
-       "<graphviz.graphs.Digraph at 0x2bff4ff90>"
-      ]
-     },
-     "execution_count": 18,
-     "metadata": {},
-     "output_type": "execute_result"
-    }
-   ],
-   "source": [
-    "pop_track(adata, label=\"Restarted cohort analysis\")\n",
-    "adata = adata[:500]\n",
-    "pop_track(adata, label=\"Cohort 2\", operations_done=\"filtered to first 500 entries\")\n",
-    "pop_track.plot_cohort_change()\n",
-    "pop_track.plot_flowchart()"
-   ]
-  },
   {
    "cell_type": "markdown",
    "metadata": {},
@@ -417,12 +288,12 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 19,
+   "execution_count": 9,
    "metadata": {},
    "outputs": [
     {
      "data": {
-      "image/png": "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",
+      "image/png": "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",
       "text/plain": [
        "<Figure size 640x480 with 2 Axes>"
       ]
diff --git a/ehrapy/tools/cohort_tracking/_cohort_tracker.py b/ehrapy/tools/cohort_tracking/_cohort_tracker.py
index 2069b5e4..288e2b4e 100644
--- a/ehrapy/tools/cohort_tracking/_cohort_tracker.py
+++ b/ehrapy/tools/cohort_tracking/_cohort_tracker.py
@@ -105,16 +105,6 @@ def _get_cat_dicts(self, table_one, col):
     def _get_num_dicts(self, table_one, col):
         return table_one.cont_table["Overall"].loc[(col, "")]
 
-    def reset(self) -> None:
-        """Resets the `CohortTracker` object.
-
-        A full reset of the `CohortTracker` object.
-        """
-        self._tracked_steps = 0
-        self._track_t1 = []
-        self._tracked_text = []
-        self._tracked_operations = []
-
     @property
     def tracked_steps(self):
         """Number of tracked steps."""
diff --git a/tests/tools/cohort_tracking/test_cohort_tracking.py b/tests/tools/cohort_tracking/test_cohort_tracking.py
index f09364df..0673d643 100644
--- a/tests/tools/cohort_tracking/test_cohort_tracking.py
+++ b/tests/tools/cohort_tracking/test_cohort_tracking.py
@@ -64,18 +64,6 @@ def test_CohortTracker_call(adata_mini):
     assert ct._tracked_text == ["Cohort 0\n (n=12)", "Cohort 1\n (n=12)"]
 
 
-def test_CohortTracker_reset(adata_mini):
-    ct = ep.tl.CohortTracker(adata_mini)
-
-    ct(adata_mini)
-    ct(adata_mini)
-
-    ct.reset()
-    assert ct.tracked_steps == 0
-    assert ct._tracked_text == []
-    assert ct._tracked_operations == []
-
-
 def test_CohortTracker_plot_cohort_change_test_sensitivity(adata_mini, check_same_image):
     ct = ep.tl.CohortTracker(adata_mini)
 

From 89e5d5b584de7cf5772a6d2e2ce515d843cf92ce Mon Sep 17 00:00:00 2001
From: Eljas Roellin <eljas.roellin@ikmail.com>
Date: Sat, 9 Mar 2024 10:20:06 +0100
Subject: [PATCH 26/46] typehints and review comments

---
 .../tools/cohort_tracking/_cohort_tracker.py  | 28 ++++++++-----------
 1 file changed, 11 insertions(+), 17 deletions(-)

diff --git a/ehrapy/tools/cohort_tracking/_cohort_tracker.py b/ehrapy/tools/cohort_tracking/_cohort_tracker.py
index 288e2b4e..4942c6c7 100644
--- a/ehrapy/tools/cohort_tracking/_cohort_tracker.py
+++ b/ehrapy/tools/cohort_tracking/_cohort_tracker.py
@@ -86,23 +86,21 @@ def __call__(self, adata: AnnData, label: str = None, operations_done: str = Non
         self._tracked_operations.append(operations_done)
         self._tracked_steps += 1
 
-        # track new stuff
+        # track new tableone object
         t1 = TableOne(adata.obs, columns=self.columns, categorical=self.categorical, **tableone_kwargs)
         self._track_t1.append(t1)
 
-    def _get_cat_dicts(self, table_one, col):
-        cat_pct = {category: [] for category in table_one.cat_table.loc[col].index}
+    def _get_cat_dicts(self, table_one: TableOne, col: str) -> pd.DataFrame:
+        # mypy error if not specifying dict below
+        cat_pct: dict = {category: [] for category in table_one.cat_table.loc[col].index}
         for cat in cat_pct.keys():
-            # if tableone does not have the category of this column anymore, set the percentage to 0
             # for categorized columns (e.g. gender 1.0/0.0), str(cat) helps to avoid considering the category as a float
-            # if (col, str(cat)) in table_one.cat_table["Overall"].index:
             pct = float(table_one.cat_table["Overall"].loc[(col, str(cat))].split("(")[1].split(")")[0])
-            # else:
-            #    pct = 0
+
             cat_pct[cat] = [pct]
         return pd.DataFrame(cat_pct).T[0]
 
-    def _get_num_dicts(self, table_one, col):
+    def _get_num_dicts(self, table_one: TableOne, col: str):
         return table_one.cont_table["Overall"].loc[(col, "")]
 
     @property
@@ -117,7 +115,7 @@ def track_t1(self):
 
     def plot_cohort_change(
         self,
-        set_axis_labels=True,
+        set_axis_labels: bool = True,
         subfigure_title: bool = False,
         color_palette: str = "colorblind",
         show: bool = True,
@@ -240,14 +238,10 @@ def plot_cohort_change(
                     if idx == 0:
                         legend_labels.append([Patch(color=level_color, label=col)])
 
-            # Set y-axis labels
             if set_axis_labels:
-                single_ax.set_yticks(
-                    range(len(self.columns))
-                )  # Set ticks at positions corresponding to the number of columns
-                single_ax.set_yticklabels(self.columns)  # Set y-axis labels to the column names
+                single_ax.set_yticks(range(len(self.columns)))
+                single_ax.set_yticklabels(self.columns)
 
-        # Add legend
         # These list of lists is needed to reverse the order of the legend labels,
         # making the plot much more readable
         legend_labels.reverse()
@@ -278,7 +272,7 @@ def plot_flowchart(
         title: str = None,
         arrow_size: float = 0.7,
         show: bool = True,
-        ax=None,
+        ax: Axes = None,
         bbox_kwargs: dict = None,
         arrowprops_kwargs: dict = None,
     ) -> None | list[Axes] | tuple[Figure, list[Axes]]:
@@ -362,7 +356,7 @@ def plot_flowchart(
                 arrowprops=tot_arrowprops_kwargs,
             )
 
-        # Set the limits of the axes to center the plot
+        # required to center the plot
         axes.set_xlim(-0.5, 0.5)
         axes.set_ylim(0, 1.1)
 

From ced853eb1e875c70f3280696a43cad013284fc33 Mon Sep 17 00:00:00 2001
From: Eljas Roellin <eljas.roellin@ikmail.com>
Date: Sat, 9 Mar 2024 10:22:14 +0100
Subject: [PATCH 27/46] remove comment in test

---
 tests/tools/cohort_tracking/test_cohort_tracking.py | 1 -
 1 file changed, 1 deletion(-)

diff --git a/tests/tools/cohort_tracking/test_cohort_tracking.py b/tests/tools/cohort_tracking/test_cohort_tracking.py
index 0673d643..55d8b380 100644
--- a/tests/tools/cohort_tracking/test_cohort_tracking.py
+++ b/tests/tools/cohort_tracking/test_cohort_tracking.py
@@ -30,7 +30,6 @@ def test_CohortTracker_type_detection(adata_mini):
 
 
 def test_CohortTracker_init_set_columns(adata_mini):
-    # limit columns
     ep.tl.CohortTracker(adata_mini, columns=["glucose", "disease"])
 
     # invalid column

From 0d44608da50d97d0e2530caa5b3f1f11ee62dde6 Mon Sep 17 00:00:00 2001
From: Eljas Roellin <eljas.roellin@ikmail.com>
Date: Sat, 9 Mar 2024 10:33:20 +0100
Subject: [PATCH 28/46] tableone to requirements?

---
 pyproject.toml | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)

diff --git a/pyproject.toml b/pyproject.toml
index dd0cf06e..4b78a4f3 100644
--- a/pyproject.toml
+++ b/pyproject.toml
@@ -62,7 +62,8 @@ dependencies = [
     "missingno",
     "thefuzz[speedup]",
     "dowhy",
-    "fhiry"
+    "fhiry",
+    "tableone"
 ]
 
 [project.optional-dependencies]

From 482d0cb43cd56be805c8e164139db895296c9886 Mon Sep 17 00:00:00 2001
From: eroell <eljas.roellin@ikmail.com>
Date: Tue, 12 Mar 2024 09:54:13 +0100
Subject: [PATCH 29/46] allow typehint union

---
 ehrapy/tools/cohort_tracking/_cohort_tracker.py | 2 ++
 1 file changed, 2 insertions(+)

diff --git a/ehrapy/tools/cohort_tracking/_cohort_tracker.py b/ehrapy/tools/cohort_tracking/_cohort_tracker.py
index 4942c6c7..e2ae3e9e 100644
--- a/ehrapy/tools/cohort_tracking/_cohort_tracker.py
+++ b/ehrapy/tools/cohort_tracking/_cohort_tracker.py
@@ -1,3 +1,5 @@
+from __future__ import annotations
+
 from collections.abc import Sequence
 from typing import Any
 

From da4f922636106e356df2e5379e111720ecba99f9 Mon Sep 17 00:00:00 2001
From: "pre-commit-ci[bot]"
 <66853113+pre-commit-ci[bot]@users.noreply.github.com>
Date: Tue, 12 Mar 2024 08:54:41 +0000
Subject: [PATCH 30/46] [pre-commit.ci] auto fixes from pre-commit.com hooks

for more information, see https://pre-commit.ci
---
 ehrapy/tools/cohort_tracking/_cohort_tracker.py | 9 ++++++---
 1 file changed, 6 insertions(+), 3 deletions(-)

diff --git a/ehrapy/tools/cohort_tracking/_cohort_tracker.py b/ehrapy/tools/cohort_tracking/_cohort_tracker.py
index e2ae3e9e..5eabbf72 100644
--- a/ehrapy/tools/cohort_tracking/_cohort_tracker.py
+++ b/ehrapy/tools/cohort_tracking/_cohort_tracker.py
@@ -1,7 +1,6 @@
 from __future__ import annotations
 
-from collections.abc import Sequence
-from typing import Any
+from typing import TYPE_CHECKING, Any
 
 import matplotlib.colors as mcolors
 import matplotlib.pyplot as plt
@@ -9,11 +8,15 @@
 import pandas as pd
 import seaborn as sns
 from matplotlib.axes import Axes
-from matplotlib.figure import Figure
 from matplotlib.patches import Patch
 from scanpy import AnnData
 from tableone import TableOne
 
+if TYPE_CHECKING:
+    from collections.abc import Sequence
+
+    from matplotlib.figure import Figure
+
 
 def _check_adata_type(adata) -> None:
     if not isinstance(adata, AnnData):

From 83c0cca87893c2bb92ace216a7f57047e7bd48c4 Mon Sep 17 00:00:00 2001
From: zethson <lukas.heumos@posteo.net>
Date: Tue, 12 Mar 2024 10:38:35 +0100
Subject: [PATCH 31/46] Fix scanpy pre-release compat

Signed-off-by: zethson <lukas.heumos@posteo.net>
---
 ehrapy/plot/_scanpy_pl_api.py | 4 +++-
 1 file changed, 3 insertions(+), 1 deletion(-)

diff --git a/ehrapy/plot/_scanpy_pl_api.py b/ehrapy/plot/_scanpy_pl_api.py
index 87bc36a3..9e0b6fd7 100644
--- a/ehrapy/plot/_scanpy_pl_api.py
+++ b/ehrapy/plot/_scanpy_pl_api.py
@@ -603,7 +603,9 @@ def stacked_violin(
     stripplot: bool = StackedViolin.DEFAULT_STRIPPLOT,
     jitter: float | bool = StackedViolin.DEFAULT_JITTER,
     size: int = StackedViolin.DEFAULT_JITTER_SIZE,
-    scale: Literal["area", "count", "width"] = StackedViolin.DEFAULT_SCALE,
+    scale: Literal[
+        "area", "count", "width"
+    ] = "width",  # TODO This should be StackedViolin.DEFAULT_DENSITY_NORM -> wait for next releases
     yticklabels: bool | None = StackedViolin.DEFAULT_PLOT_YTICKLABELS,
     order: Sequence[str] | None = None,
     swap_axes: bool = False,

From 6327638f4587c46df86c9cbb38e76cb694878b84 Mon Sep 17 00:00:00 2001
From: zethson <lukas.heumos@posteo.net>
Date: Tue, 12 Mar 2024 10:44:32 +0100
Subject: [PATCH 32/46] Remove anndata warning ignore

Signed-off-by: zethson <lukas.heumos@posteo.net>
---
 ehrapy/plot/_scanpy_pl_api.py | 2 +-
 pyproject.toml                | 1 -
 2 files changed, 1 insertion(+), 2 deletions(-)

diff --git a/ehrapy/plot/_scanpy_pl_api.py b/ehrapy/plot/_scanpy_pl_api.py
index 9e0b6fd7..5b44a0bb 100644
--- a/ehrapy/plot/_scanpy_pl_api.py
+++ b/ehrapy/plot/_scanpy_pl_api.py
@@ -605,7 +605,7 @@ def stacked_violin(
     size: int = StackedViolin.DEFAULT_JITTER_SIZE,
     scale: Literal[
         "area", "count", "width"
-    ] = "width",  # TODO This should be StackedViolin.DEFAULT_DENSITY_NORM -> wait for next releases
+    ] = "width",  # TODO This should be StackedViolin.DEFAULT_DENSITY_NORM -> wait for next release
     yticklabels: bool | None = StackedViolin.DEFAULT_PLOT_YTICKLABELS,
     order: Sequence[str] | None = None,
     swap_axes: bool = False,
diff --git a/pyproject.toml b/pyproject.toml
index 4b78a4f3..35614a74 100644
--- a/pyproject.toml
+++ b/pyproject.toml
@@ -122,7 +122,6 @@ markers = [
 ]
 filterwarnings = [
     "ignore::DeprecationWarning",
-    "ignore::anndata._core.anndata.ImplicitModificationWarning",
     "ignore::anndata.OldFormatWarning:",
 ]
 minversion = 6.0

From 3e583158715cc06cb11bcbf02b108bf1a110aefc Mon Sep 17 00:00:00 2001
From: eroell <eljas.roellin@ikmail.com>
Date: Tue, 12 Mar 2024 10:50:36 +0100
Subject: [PATCH 33/46] future import fixed in test conf

---
 tests/conftest.py | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)

diff --git a/tests/conftest.py b/tests/conftest.py
index 2c56f293..31a0af4c 100644
--- a/tests/conftest.py
+++ b/tests/conftest.py
@@ -1,8 +1,9 @@
+from __future__ import annotations
+
 import os
 from pathlib import Path
 
 import pytest
-from matplotlib import pyplot as plt
 from matplotlib.figure import Figure
 from matplotlib.testing.compare import compare_images
 

From 92f91129918f8d41f799733841cdad2d08f94ffd Mon Sep 17 00:00:00 2001
From: "pre-commit-ci[bot]"
 <66853113+pre-commit-ci[bot]@users.noreply.github.com>
Date: Tue, 12 Mar 2024 09:50:56 +0000
Subject: [PATCH 34/46] [pre-commit.ci] auto fixes from pre-commit.com hooks

for more information, see https://pre-commit.ci
---
 tests/conftest.py | 8 ++++++--
 1 file changed, 6 insertions(+), 2 deletions(-)

diff --git a/tests/conftest.py b/tests/conftest.py
index 31a0af4c..c40dc90d 100644
--- a/tests/conftest.py
+++ b/tests/conftest.py
@@ -1,12 +1,16 @@
 from __future__ import annotations
 
-import os
 from pathlib import Path
+from typing import TYPE_CHECKING
 
 import pytest
-from matplotlib.figure import Figure
 from matplotlib.testing.compare import compare_images
 
+if TYPE_CHECKING:
+    import os
+
+    from matplotlib.figure import Figure
+
 
 @pytest.fixture
 def root_dir():

From 057dd8f6efff8eefb395d8f5229223967b161ab1 Mon Sep 17 00:00:00 2001
From: eroell <eljas.roellin@ikmail.com>
Date: Tue, 12 Mar 2024 10:52:38 +0100
Subject: [PATCH 35/46] track_t1 -> tracked_tables

---
 ehrapy/tools/cohort_tracking/_cohort_tracker.py | 12 ++++++------
 1 file changed, 6 insertions(+), 6 deletions(-)

diff --git a/ehrapy/tools/cohort_tracking/_cohort_tracker.py b/ehrapy/tools/cohort_tracking/_cohort_tracker.py
index 5eabbf72..14a78c6e 100644
--- a/ehrapy/tools/cohort_tracking/_cohort_tracker.py
+++ b/ehrapy/tools/cohort_tracking/_cohort_tracker.py
@@ -76,7 +76,7 @@ def __init__(self, adata: AnnData, columns: Sequence = None, categorical: Sequen
         self.categorical = (
             categorical if categorical is not None else _detect_categorical_columns(adata.obs[self.columns])
         )
-        self._track_t1: list = []
+        self._tracked_tables: list = []
 
     def __call__(self, adata: AnnData, label: str = None, operations_done: str = None, **tableone_kwargs: dict) -> None:
         _check_adata_type(adata)
@@ -93,7 +93,7 @@ def __call__(self, adata: AnnData, label: str = None, operations_done: str = Non
 
         # track new tableone object
         t1 = TableOne(adata.obs, columns=self.columns, categorical=self.categorical, **tableone_kwargs)
-        self._track_t1.append(t1)
+        self._tracked_tables.append(t1)
 
     def _get_cat_dicts(self, table_one: TableOne, col: str) -> pd.DataFrame:
         # mypy error if not specifying dict below
@@ -114,9 +114,9 @@ def tracked_steps(self):
         return self._tracked_steps
 
     @property
-    def track_t1(self):
+    def tracked_tables(self):
         """List of :class:`~tableone.TableOne` objects of each logging step."""
-        return self._track_t1
+        return self._tracked_tables
 
     def plot_cohort_change(
         self,
@@ -176,9 +176,9 @@ def plot_cohort_change(
             # iterate over the tracked columns in the dataframe
             for pos, col in enumerate(self.columns):
                 if col in self.categorical:
-                    data = self._get_cat_dicts(self.track_t1[idx], col)
+                    data = self._get_cat_dicts(self.tracked_tables[idx], col)
                 else:
-                    data = [self._get_num_dicts(self.track_t1[idx], col)]
+                    data = [self._get_num_dicts(self.tracked_tables[idx], col)]
 
                 # Assign a unique color to each level (i.e. column)
                 level_color = sns.color_palette(color_palette, len(self.columns))[pos]

From 5589a22d59dc3e6dc751eb2d70b283a3f1792485 Mon Sep 17 00:00:00 2001
From: eroell <eljas.roellin@ikmail.com>
Date: Wed, 13 Mar 2024 11:25:31 +0100
Subject: [PATCH 36/46] updates with better names, label-dicts, better colors,
 more tests

---
 cohort_tracking.ipynb                         |  67 ++++++---
 .../docstring_previews/cohort_tracking.png    | Bin 34537 -> 34136 bytes
 docs/_static/docstring_previews/flowchart.png | Bin 23797 -> 23579 bytes
 .../tools/cohort_tracking/_cohort_tracker.py  | 133 +++++++++++++-----
 tests/conftest.py                             |   8 +-
 ...ttracker_adata_mini_flowchart_expected.png | Bin 7766 -> 7766 bytes
 ...ohorttracker_adata_mini_step1_expected.png | Bin 15461 -> 0 bytes
 ...adata_mini_step1_use_settings_expected.png | Bin 0 -> 15182 bytes
 ...cker_adata_mini_step1_vanilla_expected.png | Bin 0 -> 15313 bytes
 ...ohorttracker_adata_mini_step2_expected.png | Bin 24087 -> 0 bytes
 ...ata_mini_step2_loose_category_expected.png | Bin 0 -> 22658 bytes
 ...cker_adata_mini_step2_vanilla_expected.png | Bin 0 -> 23928 bytes
 .../cohort_tracking/test_cohort_tracking.py   |  64 +++++++--
 13 files changed, 208 insertions(+), 64 deletions(-)
 delete mode 100644 tests/tools/_images/cohorttracker_adata_mini_step1_expected.png
 create mode 100644 tests/tools/_images/cohorttracker_adata_mini_step1_use_settings_expected.png
 create mode 100644 tests/tools/_images/cohorttracker_adata_mini_step1_vanilla_expected.png
 delete mode 100644 tests/tools/_images/cohorttracker_adata_mini_step2_expected.png
 create mode 100644 tests/tools/_images/cohorttracker_adata_mini_step2_loose_category_expected.png
 create mode 100644 tests/tools/_images/cohorttracker_adata_mini_step2_vanilla_expected.png

diff --git a/cohort_tracking.ipynb b/cohort_tracking.ipynb
index e916ec62..84582c20 100644
--- a/cohort_tracking.ipynb
+++ b/cohort_tracking.ipynb
@@ -10,17 +10,21 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 6,
+   "execution_count": 1,
    "metadata": {},
-   "outputs": [],
+   "outputs": [
+    {
+     "name": "stderr",
+     "output_type": "stream",
+     "text": [
+      "/Users/eljasroellin/Documents/ehrapy_clean/ehrapy_venv_march_II/lib/python3.11/site-packages/tqdm/auto.py:21: TqdmWarning: IProgress not found. Please update jupyter and ipywidgets. See https://ipywidgets.readthedocs.io/en/stable/user_install.html\n",
+      "  from .autonotebook import tqdm as notebook_tqdm\n"
+     ]
+    }
+   ],
    "source": [
-    "import pandas as pd\n",
-    "import matplotlib.pyplot as plt\n",
-    "import numpy as np\n",
     "import ehrapy as ep\n",
-    "from tableone import TableOne\n",
-    "import seaborn as sns\n",
-    "import scanpy as sc"
+    "from tableone import TableOne"
    ]
   },
   {
@@ -33,14 +37,14 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 7,
+   "execution_count": 2,
    "metadata": {},
    "outputs": [
     {
      "name": "stdout",
      "output_type": "stream",
      "text": [
-      "\u001b[1;35m2024-03-09 10:00:23,246\u001b[0m - \u001b[1;34mroot\u001b[0m \u001b[1;37mINFO - Transformed passed DataFrame into an AnnData object with n_obs x n_vars = `101766` x `48`.\u001b[0m\n"
+      "\u001b[1;35m2024-03-13 11:17:19,282\u001b[0m - \u001b[1;34mroot\u001b[0m \u001b[1;37mINFO - Transformed passed DataFrame into an AnnData object with n_obs x n_vars = `101766` x `48`.\u001b[0m\n"
      ]
     },
     {
@@ -214,7 +218,7 @@
        "                             9                          3002 (2.9)"
       ]
      },
-     "execution_count": 7,
+     "execution_count": 2,
      "metadata": {},
      "output_type": "execute_result"
     }
@@ -235,12 +239,12 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 8,
+   "execution_count": 3,
    "metadata": {},
    "outputs": [
     {
      "data": {
-      "image/png": "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",
+      "image/png": "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",
       "text/plain": [
        "<Figure size 640x480 with 2 Axes>"
       ]
@@ -250,7 +254,7 @@
     },
     {
      "data": {
-      "image/png": "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",
+      "image/png": "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",
       "text/plain": [
        "<Figure size 640x480 with 1 Axes>"
       ]
@@ -273,7 +277,7 @@
     "pop_track(adata, label=\"Cohort 1\", operations_done=\"filtered to first 1000 entries\")\n",
     "\n",
     "# plot the change of the cohort\n",
-    "pop_track.plot_cohort_change()\n",
+    "pop_track.plot_cohort_barplot()\n",
     "\n",
     "# plot a flowchart\n",
     "pop_track.plot_flowchart()"
@@ -288,22 +292,47 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 9,
+   "execution_count": 11,
    "metadata": {},
    "outputs": [
     {
      "data": {
-      "image/png": "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",
+      "image/png": "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",
       "text/plain": [
        "<Figure size 640x480 with 2 Axes>"
       ]
      },
      "metadata": {},
      "output_type": "display_data"
+    },
+    {
+     "data": {
+      "image/png": "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",
+      "text/plain": [
+       "<Figure size 640x480 with 1 Axes>"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
     }
    ],
    "source": [
-    "pop_track.plot_cohort_change(subfigure_title=True)"
+    "pop_track.plot_cohort_barplot(\n",
+    "    subfigure_title=True,\n",
+    "    yticks_labels={\"time_in_hospital_days\": \"Time in hospital (days)\", \"race\": \"Race\", \"gender\": \"Gender\"},\n",
+    "    color_palette=\"tab20\",\n",
+    "    legend_labels={\n",
+    "        \"time_in_hospital_days\": \"Time in hospital (days)\",\n",
+    "        \"AfricanAmerican\": \"African American\",\n",
+    "        0.0: \"Female\",\n",
+    "        1.0: \"Male\",\n",
+    "    },\n",
+    "    legend_kwargs={\"title\": \"Variables\", \"bbox_to_anchor\": (1, 1)},\n",
+    ")\n",
+    "\n",
+    "pop_track.plot_flowchart(\n",
+    "    title=\"Cohort flowchart\", arrow_size=0.75, bbox_kwargs={\"fc\": \"lightgreen\"}, arrowprops_kwargs={\"color\": \"black\"}\n",
+    ")"
    ]
   }
  ],
@@ -323,7 +352,7 @@
    "name": "python",
    "nbconvert_exporter": "python",
    "pygments_lexer": "ipython3",
-   "version": "3.11.7"
+   "version": "3.11.8"
   }
  },
  "nbformat": 4,
diff --git a/docs/_static/docstring_previews/cohort_tracking.png b/docs/_static/docstring_previews/cohort_tracking.png
index c811ec94f2f011bd5dc64735567fd4cbd867ec59..889e867538236954b58418ffe8d31de08c6f0e63 100644
GIT binary patch
literal 34136
zcmbrm1z1+?wl4f4Dk>!+f`XtRpwitSN+~JbUD6%WsHA{Mw?RmQbfdJCgmiaz=NZ$r
z*8cak&ffpo=Q>`O-}mwH#+=U_&lvZ($9+HkGE$=0*X~?Hp-|Y*#e`&0s0;Qe6k6|<
z%kY2HwZ^>RUtD&=%64*=hIWoRHU=n39Xl&COFOeyy7wFmY;0dyTCmV_(zDRsGq$s{
zvgKxAF#nGi&|BIVF*MC7%fW|SwGvaYMWJwYkiTf(1kzriP;VQa3kk|Q#jlJyI>{@X
zHEr&<=^M&Y2g%>Nzccm}@4DcV0Ka>RLYIU!`sryNFJJqTSxWxU^zd_@x4nU|u%kk<
z*hREZlh094u0Or{Am%}G=jkcl?W>PId}!Qo=Ii8bi+5Rj^z7qd2iJU}DKl2x6CwCR
zZ#*|-;pqo|;sq|*lfxg<CxQvc&loZ?)bQgJ+Ra{n_|XaVfBv=XeX213v?TYV_$_vm
zUcY?3CdJZVx8pq>=9p*>OU-wk+1KEw6!YQYBH0o@F@s0n@XMOp;y7@_TI!E>mot+q
z|70p{@AOEAMMf%8&MGv9QiwY3bcwzi{7zY2U43wPh(|`&Ievb&uWH_3#%Z^3U0UFj
zHIBnFXEoKY`BFz_G4VZe!pM~sv&zb|gP#1?yC$g;F+T_L^{!sK_Bhh9Dfn)5`=8@c
z>(O$1b?4a#eM3W=R=SewWMsIt@cqt{?+BIibuR@HGUW8J$A>)g#bGk|b@L9Ji4eTg
zw1SJ=22BnYg<xuR)vd#M)xm|W_+oc=cj;WLX`V8z*-)W>B%RtVp0g3#P9~SV6%+Gk
z1_U>6PS}*0RNN+?5@bDnO#c2Z&n*uR4;)h7W<NZNjs1OUTH5!bWNzU<fBw{Ume!Kn
zhude-`FRmFH8sT;BM~h@wYNHq5w0%hI8}d@)NSYGP?4!`O^pDK@0<J2d^UQd1r#zy
zEvpZyFmOrr8{gliyub~Q^UB#?NezLx%XAQjaZf6?Lq)>Eg7I3#dNp&Jaf!v~uh}TA
z;M`n>-R1uI20|^`(D{X~ukVRDg5jN&RaJ|>KclZ#ZcF9!ruAK^-su*v-1<es`l|b4
z#b^hyWu;($js{xk>E7_@x|Vr1<;_scIFTixi=ID!KD~^6M`(X-)Oj%_O1a$bQBRt5
z+Ve1~lfzCA`rUYVn%`~brOr;*J+`-68RT*``A~4%FWPGy*EMg^t2giVYe}oClZXbB
zOty23Crc;sn~j!7EG#TsD>@8hsIF~l@?FlUSc~5hINj-yh+%$*hJJZtUf?|DE|0^8
z$N9cTmRd=$LZ+flk?N?koSfWkUMEHh(cl>$Vhimb-j{gK4yFrD`-5TKP{MLu5Q+wH
zts@@ir}YgDo10B!EDkHm@I0LsKKlmz5pY~nx3RG)JzlHGA6`|rZD-Th*Jpgy{a~&7
z*y8p68upXUL_V4wem_5ywqLWb!P-dKWo@{Cjs$kYHX?M!MCZ8}U1uKqCH?xQCPH!a
z_bp-66<X49?Cq%c_4V~P>>C;z@8~tYx31}EX<@pCi5WQiTtb3%#la=N93LMayEdA~
z(W1=lDh>`ix7~t_MWS3VDWA!J8N1#@tyja)#*k_5ZO_fUVatl1qn$-PR3kjyWvSNo
zb~bVfiXhY6s=kE1Uo^Qyo||XK>lHm)TU+l*`Cb#9ot^pgEYkLo`PS)>@VzcnuIS5F
zbChx;z~ds8tO@WN=Gz~UTd`jss~W6uw2X+bEEygpl)8EAmd(k5^~ve!tFAAXh&e2>
ztemCN!aO}aZGTnREy}{DI~`2=25MAV_iH##T5%r00}h<s-`}5)RI{+IDIPSlTkgvq
zINtoDq=k3;cFVVyUq@akW=3F;xqfJClc<JO)-}*{SuHXM)(!BBXMkjL;id-4+OQUn
zghax`#3bU+q)nOA;cV2azO362&#!lv@NV5|>=Zbcc>VhIyH;CH5_mrGa$~2*`{=*F
z2nXsevd0BA1`$_;Bz1ohM3<nF%P(XZ4}o9zWT_O~P&z0!`IDiXkOt2_K~7Cr=<zei
zdnoOE<&3J}`6Q3T&FLoFL4zn{^|m1z^YmM83>ChTOQ#W3D>iHCNZ@H{ZOz;&j@)%8
z@_D()?B?qF*e%DedGO&k{LpQiIb%iVTeQlxu+chd&UP6r$``3F<B&9aqM<*o65Pns
zscY`&m<yBU=Q14|7*PIXyk6NeJ3AX19lgM>iI<<BzcN&mDfFZqf>hT&GL3va$xqA~
zGI6u>=Jqx^XIgEWZhkn?;?co$2$$1_cFI!qe4>kAK)@7(#~~(<;~K5?32f;T*lHNs
zwVr4YHfOL2Z<*F17t`I-bDP`F_}y|1>}xL91IF$7PRBGzRU1p0xl`-wu<_hFR7Hfx
z1n@$)jK%TsxERvbCN^fmq-}a6Ibe6)W;cuGQRW&|P||2%>Ex|jEg6@B-2hKV>tNyy
zuJK5zb;=SB4$fRbC;v1AVX|IRaKqZeeCufuWcT-c6~+Bz92en_Pt6Po_A)MM8-zuB
z@Liq}mxcuQm-S`k5D(6<?6-LwXo=a)KE3=ZLaUtT3441JJ}$mTm_#y`_5Q<$ekuiq
zc-OCcCMA)LRyg(_@!+L1Lg?6NmOHHef-i4{N3C3Fj7J|82zyTUO*!0tuD*f6=A<tf
zv&We$x8qt{l$Lu7zBKR4#iI(x^^V+{vtNa;dXV^aUa#6`jB776`4d2$=;VF8UNfGs
zznGRp1u45<(<Mlh)cNg)57&`=kmzxCsO!?3Kto1GcH_nkrzwAGr}+d&EAE6s<1X(q
z+xc6=c4fQ}?~Y^6zHr+%oqYQ;GBULzWwr#p{$lT9-fyey(jev6-n;LATyOtXt>t6L
zJWlZ9+7+in_7Yb<iqlW#H<4zp8v6Mm!I-fP502TS_uGq6(W39nEG#j+&dhHxaH$?V
zxCn)}(s}zuI}c<&*dhL+t&wzrv%@8pTIq)pRQP02Z}KZc1kN~NS-5?1$#Qb&VRzz>
zzI-Z_Q9sunrvgbnQz<v6>$RM$tkYbKk)OZ+XQ-PG9y|z-8@DK7G3todedXoll`?mi
z&E%$%rn5nQp<Kyb9)p98Nlp@+F{f#4BvA%9J`qX|K(6*N6Xac8UER`|kK!hUW8c0#
z{1_JI|Ng!H>Cvv;>QK?hiKeFJylYZF?7$?p+(EOs^5vZAQO9wEU*V6UB~ooID*VJ0
zx_2`D<o(}a5_acm)!Y!2{3;Sux|ET50GoNSFPnI5Y>d@nB-(M@t>nd0(eSE}h=}4s
zJk+mvsd=v1_Fuoq85nX$c$L(xFCFi%8*R_E6PWnI$I3$WpI@E$@#X;~WfweE*$~K+
zaZr<jX>v-v7kkq1Nl4lNZVY}m2;>QAoh7HFoQ)?vJUooBScdd(%}H|c@@0sRpKkl3
zqK1Zs>CR~qG2T$k;ez1iA{qh-qhcBO`ALyTS18vYC!ZJZ$tAZLE4CcRd7`eap5ElR
zGVmjSfM#K|f&;}dVug-P%&rfmrxL0(3a%=``79=qK}%}9+KpzU16GgJWrYc%{C(@J
zZr2y#%FRaN6F?a7d?scUcBs3|*jcP5p5&Cx-C63-?HsFeiDA;a{LBYy66&q?!tc)w
zzp}Mz_#u;KtCc90N|KW+2t%5TFPUx%LBCEyqQ0~V<)Z?f+cPhZQMJ(c2YkZltI*I;
zT9xlJ@k`6UKR-(u;<8`r)y=hp3>eE{*#SAP#rqOw>0)Xe3Zi8)&a(PZ<bI!;Ws3cB
z-^nb4$0NL6!I$F@Pmll5<Ar8?`XnWk@m_~0CAF!(K3ohF2ZxG{?a%4Z@NgWIe|fdx
zUt7%-AsfAN()mz`9A%VZVNY7roH}SaO<r<8UgfyMYAmx-szX#R_Eg9u!kZh9%kuN*
z&!`6P!X6dcNZB4~+zXhPm|PAk16+RyZ`|;1H_X(g5ucx)er9a^kXkC9({#)SoA`p&
zRK5QVK@C;aJG)DNGC#8W`^o$K{jfClbx8?{EW6tnI3(Zp*z8kUnw#%YQohT}%PW}U
zW7BXo9E08VF)B(lQ!zUu8P+i3Gu%^$8btJ~fxP?6{kgezxBpb7(o<9Wo}9Q|xNspL
zMc{a)fXL?ugxcsU0G~8$RxZb@C9lS-`Ch(!*$`|#kcZ}($mgP13NiBck;wZ`*7<G^
z8YL;bHXl69N{3R~FD5QN4Nm}dvR`vPTQcsBKH-HydXJIuL!B>fPr4knNYI^GQk9dr
zILiY_{6v73;1`Zj`vZC*`9__jNBis2c6O}Cd#lq>03XoM1jeu!MSk;W`A7-L^mn>k
z+KVDBq2$NW@4qXu2l~m^rb@;wjFhp^DCNv_i!<1#aX=GLY(BhIbAFnud?weEW>k^g
zUuN9<4G(q|!>+QfE_u)$HgY<;APiC`PcrudFGw9Pm6gSMQYCNmy=D#JKYFN``NypT
zKETqjJ+=WpOiWUe67oL8*>*|IX?DG)E97>7os1SEoQZCGgWGF?|L812AKRMfFIVl4
zWv-_7MnfvFfqGJ4)R}mr=Q}w$+j9tG`IsWp{%(Ld4y!}LW@cvQ9BCpu3texJ51$HT
z=yQ-0njuevXC)+5kG?RLi{$73tvxD~9=)wjPw0ztm;BznC;Qen@$k~geRMvAh6=*s
z85T5#XT0$)XCdLXYj*Px#e?6fLZCdfUe`~*b8V@^YU$Qf__=n{s@KL8zKHit&tkGp
zoe!nOA8G|r!O`1Nl4aZ^IDeZj1qt!Ej1a0KVdQn55Pa|l7ug@k^+EE4A2(6gmf%PD
zb&fmTvwoaS<+jKn3m-rAncx(%CV|?T;<)>SG0n|leOW4U!|flg$;v4Cuyvji`sVI+
z<An|>SV|B>m(d09QP<aey_hm&;Vti-s`Wqm%742e<0S6Tg8Ue!60F*ri`ZkgnjLjG
zx2?@zya{{fF|*CCw@B#KqPU}xZe@S1v1??j)xk-}VzKt78G77pW=yNqmU=O66}MBo
zvL}U><B5vdsz{-aiXt2w93<g)jfURE79c6gqH0%>1V_~07Rij=DdMek0{r}mrKOy(
zOMM$;3kOdC13aXqrQPQZEzfvLVeg}3wLj{J4?Dd@Ul&q)M26kUK>m9orZ2Md0lg37
zZi^=|wgsL3u<&ZT&}>g|NoTBKol`d$*BZla{y{BftAS4RSGQVS@3spPUW&8{Z<yFN
zGpB<6)SV!MhJ=LR0xsU!v2%5Edo@(}h(hFjZENO1bUhS(PS=CH<Iu!J_r9$-cJmi9
zGS}f&{`|QVdPm`NXt{9UEhft`UZ@TLpO_%W&~ThryB+%l2I|3ffJ^L3;J2NVz^eNu
z|3=v4V_5`xSQ4QROVUq;M<13Sb88=+Zt}>nc(my_sMjRakJRpu{0iP|m@#O4$of#-
zhG+}BEt@1-?V`5D;wMHXOm&5E>QKyvsfNJ7pjb$}K{*-~(AG*?SusMou_XiiwGIuo
z%)4Lb&~J+>D&Eqla7a-ru~6u<2py9B{K*3Fo{+F`7wo9*3|2<Qjmi$f^hc>j{3xQq
zcgnB-plf036HE=ocv)}kTkd-)!nc=xEMVF*_`!AAK85D!yWaQM`vaGowmS+i$O&_x
z_CmvpWDlq{pFZ6~NhR_=S5;NbuY_XsqqQ{<nuNWz(N<{FTRJ-#N=~w}vP?|M#8x@y
z%f0~7uvLqW`yo91ZHxXF1S_f*`iV1Cg~JCuefF#jD*r%%QOI#TL<Nf5ehGP%<#_el
zcc~=z_U2}7*z*XtKtRyY%1|Q#rPCRDnNeo|K$wK|w0)`YF@SaE-Q6Z?(JV?QFME4?
zU%7f!2O!MKa7pfn`|RA@lpS{<pAaf9X0#`L?uunz?EUeFvE-|FydMn_(^AoDc;D|=
zC~VBRpx(koM2{ZU@YdkT#mLq=Cbsiw_qtS;R7YWY5ui8#h(G{%51LW&@g+cSQ>k1V
z35CPZkjmNF`QoKZ06L6vs*jdb&5Dp)|70=O7Q?&U#=>g++jC|{A8Il3sGv9-L2;o~
zE5<4qY-8(Jf77bKYP2&QB5)xOY9s)vL%;yybS@_cn^~$w0Swg#-jw2zc{Z~z*eyns
zKL^ju&$Af+z6X^^xy*(V%GS^E@uVMbFv6mg_Kcf))8(#WVc7}XVK-C${JP;y!rm}!
z>__UZ(LWn}nc{`XsOB2aqSbh6$1(dG<(o4n)EE7E)lEb6;+xS6u9_LXM^DqZ_nGn<
zx!K_Ncl}x(2?!{L@WLY^YOeJ{&sWWU()Illm+c%QAaq#xCSX9&{NBHR@c3~CdmO#&
zbcgd18&ok*PZaGfOaL1wb*Od?Ta`dbynXjB-+ozfsMuUwS(yk0g?`PpBRn+p2hVzC
zoiCZY*!kHh8g$n5U0q!SOU);T+ty|3Z80nuVJ$QC)KFO7zP$*2n(~q}KDCrStj}7%
zRGOSu!p#<Yotw9Ul5cmM$gzAS>Ts37p*c_;X<cd?eHM@yeWq+|v_={b8XsipW)h=7
zBx>`BcdM!Vo}3$*-=lJ|pUt%2y4U>Fn*n_f=INlf$FfndvW7!RgNyCcU0%Yf(*`P{
z&}{Gq%ERM)$RtNBn=_YA66#Vz^7Am|3J1DJj~=bzl_w`ZfLn8sGFTa{h&nkr(G8SG
zM@NUAT=?D47H~oNF8daL@x%Pg%+0tP{9&UOO%^3e-{{N4h_p#M8*Z*?+^jz(&5a=u
zrno3#h(=YAz99ZKD!qP@D)4IEmdoo81mdoYsCUG&_gDARzy9v&Yd=(GETKI*WQ#wX
zke8t*WT+UGGBcxvR>{GmxW2KmJVY&9@#)j2KoM#~<|tLChiwDRne{xtAk;lx-grQ1
zC}9>+(Gb$?ABldw@!Qvy08fd)hd}P2;^w{+)haw(Y~Gr%mPH%ZB0m){*wZf$)Z7i=
z`hXsgnHrI!@F6Fd9h$4ZY0@;4r+ZtOb;xWhU%M3ReZuWh*?{w`xTV27b=941ha-M_
zDhB*iROazvdXm0)h|ELQdrg<lG=;siy#jw;%jv}5WQ7ESpmoTtuxaxL&DI!JVRIo1
z3IPTArO6;lL*(^qZhn6LgQHs--*$#AYpf48bWoL*l~-_Z90UYr188#Yk&}NJAUuX-
zgO&$8BjhGE3O9oE#}PP7!DIp*(91^ixv*Ec>|dUM#*rbi$*=$|ydqR@x-kd>93iTr
zqlr+^i4T_9Qb%bx#PDcAz_zrsto2totWM3e2=tNJuZ_eUZqMVBl4>rAgp)^nR4t6~
zhr}<JM{o^WUDVdf-1e!-p;BK0=Cl<~stnIXCqXZ-WigzbRjHuv7GVYsj;I74#~XNf
zUUieTX6#|G4WKz2fB?LG`}WP7H}jmg4FMfLdGZ7W#BF%^O(focQnl;jhS++s(rC4`
zo8HL~iupX{P5H!xHsn(0<?Irm8OR?!@`s-Ii*n+|hLx+UE7D_Q-(jsC9<Hcokb%zw
zVoT@e+pBNizP(OL+A-4-#uV4SR%DHWv@~7RulW_C0XU^Cpf4&2e_a4(D)v=RDwFGh
z)&8Zs01oFNmS>@f{VMq8a>Qze`ij)dq-unXluA%I@3mW`yz@I-eZ&6hZNNJzf7gFr
z=KQ^&K&8m!k4h_WW!IrueeJ));wXQ;?6<<vSgv6~3_wXgXqA2!o2!vJ&)@-4Riq68
zFmZk|FMtpY{=vb9kk3;h+4LH*UhgdYK&TBMYZm@w5@xFw{Tv=1PA~$Vjn!=61HeOD
zc+2*}*N}4G!Osid_X%heujAv_HHT8X`jbHbkC=j*T9m>O;NsM#gvi~>qGynm0k9yS
z05ofk$g%@>0<?lQCvVI-UC}$-z5T@4GB{GPJNeGl%{iXYGc#1}-0=7JS5Tcx@9YOS
zfn6*gwp}jsB?DoZSBFb_veh^sc%XGQLj58cM4(^j)RK@4hVb|`=w!Uw4JiNxMbIcE
z?v4T6B`23nmbz%4z-_Nv=Yw4|iiNx$Q!#}h{5tYq<p}rnUvU1O+}59kG<v67bp8WI
zW+~FZwaNayNJza!qZiXJUa~h(-tW0=9{do#h!6{*5&vfc19<uWhNu61+)#3r=2;qh
zLS8h%c0lh<BgJdMsB;9-qVPgrW~p!M!IeA~6V$jpR()o0vg!`^78=<2i6dw3jcCp{
z;6Sc09?eqzy-aN3|LGv(zq&5^g?3*^aN^JaBB+2x8gRnmKt4ro<<^U!BJm87k#EIs
z3qw#7_0pnZCksj_Gr=(-Y(I)m){KfC#Llm{lCuQrLnEGR$Eyu0`X=<@+BdcKU#rb2
zi?dqeeMxtEz3hXKBbn6cTAesw@!B7rA67cdYJN-1;lCBD@(wZNKCI5v=j@W4OtPpS
zmKkN&36yhnY4}bNb`cds1)YQ~Ca;bPbq9DT**gl+KUTQYn;`x{A;PdD-d`tvI;~V)
z=IQj;5wy|8%9E_Z%3K<_q<3YPW=MGarQeca`>@t6<!uk2J0w(D;j@;1F5>xwANp~t
zfTdYd;I)WJ*Z8mqx<#61Ya$UnA=fL>JPaSa!B_XHb(mHT?C#9ERa~>^i=cY5)~H|n
zAt^y{&;OazpxafmUQ&3L@lo7dvQbe{zU_uQrq#8zwRenzdRf){aIt;asBJy&6hE<q
zg<8}5vHCOri15_R%tMclOSIryKU?fTX!ezwWHeDls?a3OGciG-b!8ICr<|MHp|a?r
zcwRg8Nn+|p4*quIc|=P^D_5U%2jFrYZI@hZwEuLN=WE4k{M;%`DtT>C*(Ao0fjH=B
zETF~gzVoTlN%5Q<qq=ZY)>uj+1^IlAtmzU^wND`JZftJehq??EixEaJW6tw;`Oi;m
zp`bx}1Ym%|A>pjW<`{Vb(Ea}X`&eZB@xafPwGsnM2|$0MY(9~Kn>+RvohoVrkmT=N
zEdfV9AelH~Vqz#ecqpNR9=52!hWXEzFJGeh-Ci4%Qx~<Ek*_tfYd@Jfp`#Sll}r^#
zJa=D_Ck<<mcrra<6S(h>bMGjiG1;82F`vvmvpY4=>ElGIhF^{2t-Fowp9b@MH?RZI
z?lBfu@klSNVWX0%&)#R=!$pajRWP^O)fYClHj$w+bKe9zN~HB@oF^|T;A-#M%PT1f
z(cdA!SpqD08;Ci_5o-*n!4&uJd%={3*?CJJ!MQ-pO4!&iLs8mU84Lqz0vYc>a)$<A
z|4)X(!SOK*aL<r~M`r2Oiea?nfUx9ez&3<NL^OO7d;{guaJ1as^<=Y&R--%|3SmuK
zv1Bx(*LwAFFu>b*AiTmpe5h><O7YO+rR3lsOzTluxnD$G?m=?K<n_R<I-#;R{zg+z
zNt#Tw8tO@_OGWpRmqmTB@gl#4?LaH-us&PQK_mOk_%@?)>fwy)O>)woRs<hCm=n~E
zuCdqL^Pb}Vy>vvqi5>Xk)<zRy+R2KX2LtxC?}61;!{|?c(f0E^>^|i!RFK}XWoC3%
zxoR=uAKB0E(=aDT2aTt)ywedI7gvn8_XUw)lIMU}P^&`(ZY<jnCQw_On+4ph$34#7
zp#8U<Yx}Wt{IC@WU6?hJ!#aAQV_cY+pKprivK{HkL7)!MI}i+n0s;akm~^lK2x=5K
z+r%X&50H*CFf`oC^*9aOANNQ)J3U6<=-^rZ4J&AK%woSJ4^vFpt`flvFUwOD`Ek9C
zs~j^mJJ};PX-yNHim@@Fom#2`5e#knf#-uCB&z7UqT8sjlm<<mmvBu2suCG0<}6Y-
z^xf_4pUrqXxHjRE3|5B!9HB*#Fqs=OO)*4QWxXv~_nmx9dnzQi*wXMlnr>|#!$VnY
ztNeVDZ1lZizhwO!*;ARY4wCWWjK}5oZC++r$@G7IhJi3dux*G~jjy9);LHFMOTk&X
z^C}lc!2ZC;<6plnL(2!Z&F8XLs#RjzpEJN+4%6Q*c&p<RjNh3`46t;-JwL?jT^%g&
zt*qoNcU)hS!n}IbT6#9p02sITWCCNSLyLCZFfp;|3|*SUpA2cyr7AV&6wIzeF{%af
zU-TaK!8@=cb5A_tQ+Y1s&^|cB-^tJ@s9bV{g8Zgpcv$^XxQAcT;?dC!UFP%6N4+zV
zd$65ErbN{;isi@&qelv-w7$hZS8ML<s9nPMq`UX_{kTJUh>9D@Op*kKgp)Jjsr#&9
zxEq19reVghb6;q^?)im*ySo`*Fv$tCuVL8V3VesOPXGgNF=)1B=2rhgNKP2kink}l
z#5{L(<t2075cJp6(a~9(Ps6ZeYq2QND~aMWpk`Z_zvsvu*6_YTw^5&Yl8{(TQI?xX
zdvov0j}L~Z8Kb$$5xE_&so`3|k=<-Gu@fg6Y7d%ND&`nT<rPYgm3Bo;!`zR%0qSPt
z)4o=^O{!bWf$mWx*lK#akB(#$mwBa*&a{kw+de@3kev3Yk@>=BnEQocw&S!0H9?3c
zIzr%h;X|oDj*94wmIHHQpVoIRW{nKII7Z#KPJR-XEq6SpaeJ#N)FGLY<IWZ%v9t6i
zP_yba6U;z$_4E!`3Oe%!%>c!b#jZo&4l^WbUS1O4=@G)8Fj@<6-kyUjFpF~l%+0$&
zGkW`-Y#0SAXDjyQXoyQlba{{U_R2e~jXZ~8)8ih=be$#+DqB}QoBYD!kllLH6*;q{
z@&P=${L4cv;;lbOmjW9Me>}7cEwthhIZLNmY;*EGE=xjRiPvy-OYu;@PPZ0}Q7E0z
z{AhBOK25OGw565P3-aLl#Ftstcg|c8@^9nueL1VAF`lZaKW#JMz#pq7+pD!gG2^V=
zvzP9ASU%Y5BP|ww%yl!U_wK3lZM?g=pWH+6-m<Ry@Z~MEy6tVgNnX>`>KP^2d`M6&
zqYJYL-t&{~EUlVE=*sv{cjN%`i$Jf8@S6hXCoB-x2n`HrQzV2`vg-9N5x*-Zq|OIS
zOrcOrNFgNLkF!@YaRCblm6UJ*WS)d>*x+z$mUp8TedP6WHf5?N8q8d8b6AkzQ%Ouf
zBQlG;0_dvf_VQOgY1V7}f;v)YV><2=%wprt@d>u<p~bVx<{XkU-cgln;_(>8I?OIQ
zj&DQfc{?^&O$Uy;SZ3lmFFu+2=wWTOIzMAT(nLxmCFpQjl$}9mo{)uW7-!-+4gXYa
zMo5dKfRCfq#Y5@@lEYVzhHptUJf!{RMH=9gb1kHoiv&k8OZf*)Zl!n6;sbW}NSJ-F
z=r><4w_Ai>@PbCU-F+68u-VqA9vCGduLSO)1K=_AgrD3`k9O9_YeqtOw6wIU59f$U
zUoYH;d0hYq5@4)G$70y_a62B_fzX(k7TESzu3VWwrkab3iv+7*dhgd%mzE0+S!%6K
zl`yo9>vOU&1!3byB{-6X2DazZja!?wI?m29slOvpJ?8T5bJsLXeE)_=dB2nA1y5WZ
zarNAWPi|*H3b_qc*k?87o6~mOKxRCc4DH-3>F?kc2o3B)z~i!!W!AsI<B9~YOo~C`
zk*lP%bnb5(S}w=fIAD4I6W980#OJ?LvH$$FO#1L!4Xm|;13cPt;&);}o8O0sPPab(
z3(WotN)Hqogt3dl7ldN^_<Z>XnfpKcg8z;R{<n7nlWpK<I)1r8g(=t1Pd@n3F42!O
z0M74UTn)+S{6)sm@4%AmeX}>>{%$^AwCl>>cA@OarO|Wdq}|q>RYUn-Y~y_nn90nn
zHz@%s|KCW2{vzw+UVrsu*D5oKC8tXiy2G;L93Nbj?jN(rIeT=9q@Ks3o@9A2hVkGe
zTguCkSDp?gMf@7(@z6iY9oNwgSJ_JrMvobJr5)2m>~TT@@5(F}Cnvl)eyi&0-J^^A
z)*72OZ_z&Ru9M0})f|da1rqN1oJM@o)XVc>{Z#9GbG|%;>!?MJs7%Em(){~tx${Ka
zjo{VyO4=7iboei&V_4^$hlevHHZ*r;udO|&0~M4_=<C&wrbIniG^m<XQmWr`lE)v6
zHL?7u(ioEtT{yPu6map@rlm#PafW+By4g8Dde~J37;9lndz^e}`;KdvvRT)4+Rz=#
zwS>2{Dk>$&pI$~d=aWUQ(85#9a%P@$Z+$@1QN}TK!P*kPcUcij>AthCqYkE}#KI@h
zjo`)lGmjnguR9D%RGA5$!AhY~-;+k(sBK)!3E?Xz&`}cbZx9HtTR$n|uAnLs5=>|o
z&QKB}K?S-Gcl@Tb-6(bnwb?E>a8mK##hoiJ#d&*tfc|^iBlkPo>J>-!^^UlzocU_S
zs_6V#9{1K43`Hv*t4E>mw+kk0Q^Hfb9_SNQe8g6B^YXUqLQB;<3hSZeCz6#HD2&<N
zwEU;X@)NEs+0&qW1)ThPQ*x)rHy`Vg9x}LRAKm!veDO9hKTVNYe~Vr5=qm2JL*c82
zn3CuD_-amiu5r5=IrPuzdSbal1AF&HZ=ey4e4-QxatdGV=l2SUiXiaaXP)j~XN;3r
ztIQb}MVC5VHYm1o*6*>+F_q}af!mm7R&P7lt$4ihuK8Ui<MKrjE-kB*VO(p0PUZcp
zs2Bk^e;S74{pmNIvV#0C9owiZwp3%=#eUuzBk~NZs21pq6xUFoep7x4^VaVi4IUsd
zu3%#ufL21^fFS#BG>Z|=Kl=VS&~$Az5?hvbR!j%;7{dgvtW4PS5}@9N-gNwuQ3tgc
z`~M(-=`VX|R(U!)=3;y@w|anm>fKC7I}z2ZFT4Hy@TiuZ+5Nm$#rDq|v(h8Hu?6Ss
z;}hfYTHB1KxBkAv>X&QJam)<Twv`F?h3Z+H-?7>WFj$e&z1AhVX_$=C#`fnjy}fs2
zzWHhd+ti8`^(u3Ge&ZNd(M>N!>YJ^g1^{)Ka!&>GtZ4(1hXF66&5RgU9mp~GwY!s_
z^X`w>nhh5RA<~w##Mc49xng*FoNXr|YKVG#X%r&w_!t@42+AQvLpQlJDN07h->1gF
z7R5LV^S0EP(@#fIB^;iKk>im`rLLy+G&gt7B9U{)gVH~Sn&_n6tXgs`sFl%g4|^s-
zW7+C9J!&iHr)`FjlLT=r#;t6^*NahaO_qY1g=Y$1?i38}o8cA*6@(q^i?>MgiOF=T
zU8y)qH!+LvP3100rT=%>U4*W^HPa%I#Ge2S%7yEMgiSzpBh1Zh7DFr$HC3LU9q%mm
z_yUKg={!pSGw|TL$%h5ih{nYI#A)9&ttY)!iZ?N&;TKB+h8}y&q1||A@{D2V$GC>2
zum0+Hdm_)e+R+Xs2i$X%SzF!4D0T8j2B&VI98!~p{e};U-B8N5c4V2sYnemAeolGb
z_!uLdRV0%)lgTcJT@Cbjh&rVH*pW{1^<Kz|My6Fxg7m_~W-sxcl5Djr8ih0=t#d3l
zhwOg>&825Y%P~L^rl+UFBS7Y@AXWlpU(N9*G*Af$WbRz=`ig-JK|A@6X*~{GwMt>i
z-<sWNQ8I1C8NesRn~)ntkU4naiEGtm)Bv4kAx^->djNVoH7A2cOqadQ_$vo?AESb2
z^}nzDrbm_cnKnn7lIpdxg>)5ZZZZh*{nQ&J)HkL}5Npss``&!Y`^mJc8{-`38Pl+9
zK=8S`-AIMCQmn+NNLc#thK0JFaD~>gyHk^cO_JsJ<H_MK#<$<LOucJa(x7nr5UVgR
zvIKOzcqFY346kT4E2EXP+#;xxTt9+dvj)|c$QQQ;94qJi&8XIf424WkOs<SoRfr7J
z+kECsboscBv;J9HA76ySKTS;;3r*93+%2%WH(u3rU!-&S+xqgv6U94{7nUZCie64U
zo*Da-^x^ba!M<9PUG?(2iln492J_<!d9BXxJNM32pH<<PF<z|KdR8CA;C&bpGL;wF
zd7XRa<@f!p*@?K457l$eF1U<H%q+yb-KoJ<bHYb`cgVV@H~kdV#;&!}MK{`s<{Gz8
zwGx~5$ci-O-tP6jjYehH0QVXeb~i5?@nfOaU*KV4Xfwwk{GeQ_gbs}8Kw||^R~UxA
zF8gaKfL|I&t*x!sY||v;8exKQ2op7gCbgaK$Zh~0T!SL2eY!y_vVh|BVHdyG0IlYU
zm7PF|`#xGhaqea9mLGRHw{2hH5lMNyY@T|<LzK55$v=ADs?utC<NB{Z*u$ueUy%{B
z7K*esOw7ledBUs>xB8kB=4P)n-RioTJ8^^}iRbgt!%X|&!Fv_%@EUWlwd<w?P8*Tx
z9H~8udjI4}U-7Iq(ZF)a4}Ipa0uL5Z0q1~m&6>~UP)F+ZiDpmad%a;k{@rYl873lP
zPEMTAz|Df3PiOMyI}%=}4~TY;P}9i^84Sh6A)~qHl9C26GeI~?ar&~ikXjkd1{lX)
zFVi8)Rd&lU3E)OM)7{g0o@;C0OYL88n$gRh8w0v9?OOS{fuiPo#y`=8UXxcwmF#Kn
zyy@%L<&oS9VcZ<2OIDnSJQoPJYcOL4)m*A{Qpbn;!pFM@M@P)&L#$L1(TEZZkoh#|
zyO4fc&7$lbOb40aDo}_{tNLV192pn`Bz<WXtB4HF!+Og?%ZE}!VVa(P?M^z46ovht
zT&bM&zm#gg%l|78v?q7Tp_K;S$<)&}!-h&AUAd^@5=^UiL5n5-!b$%0?=aoFocY07
zrEnYt+;;P&XU|>&Ci*|qp#L>N`|qOc-{wDmUGGo*x^}AitSzSHA{TwzF?^5H7r$m>
z#jHW~e@*=HDN=R6ANM)1&|kY>kgL=6?C>whI~<*eDH^?yj`jaWl`hrK)q!e+f}TED
zGLF3q9>9==RQB!hG$y5@;7`KQ0{5#Gag1g@=FM|<(vCf!@()8f#YmnMd9YY4cY9v;
z;wNcuGx*Wj!dqm_qvHI|DEq79^mVE|Q<44Xd#+ix;H_H?k#)Q$hmlKbiDC)!vwkU)
zcPoDJCLYxlB%f~3_WdrtsI_%-@0&$*)w4%pqryk2G5Qv6`|-=)-`PJnYK(}kTtS`e
zN%3sSImSDZ3C&C?P<wMfvhm2})b(~9u%-=LIeB`=_lPMqkO4i;w`+^6wWXPELW{#;
zrtjeL&2lV<6#XvF`Y2xpQ;DV{oTE<FH#*@OXS;X598Ao0IH|NH>(A!zic$TjR`e;v
zQj(#*Ky;~W=0Pr9q+v0J4(2PgCpwDEeIFEs#tUy0{0aHFGJP|7weeP8aGsYf<1o6l
z#p~Uy-Dujex7g18cYlM>=+Y&7`{wwf=anUs95~4a=#0}#?7XfjdS6J!b}SV`PSXcR
z?u?^6eFe@V`ju9vGy3pctwlMG*wdT**h5|SmM3Lf_li`U^z>aSX$>{itK+C^#-u)H
zKD@9~qGpo%hTL9#vZd}p=eh<hUEw;-z*AM3Xj|{L)pM$E$K#A@r&a_r8^e`-r8aU`
z*AB4+M-%TYd`xRTaAp_Lkt*ALyA$I>mrNPMLpU*hQogkE(NA5-#7;7yb#@B0<;jo;
z2UZ<G(dV>QHveLVhL%<z<e{Ltodktd;4Qjnu7ZQ01(<p-fH2S9-5r^712MF>K8{-p
zv#5H|lu7~T2zvK}1E;=REiDQ+Ub0Sx;aQvol;RrSntP(!gsxbt)&8=yyVT{0R&&wj
z4u)8VyRJi<EZD}1Au<BQsL3-2oD*|YQ!1A3<f?}MM!&I_*sHd&`v)br(VsMSTI$Rg
zfk#2A_2o`OGow06c|~x5q9R-^ReD5*Wc6I=D7C+NY~Q}9P)&$lij7;A!+J`{ursk8
zINr9DD9slDy+AU_5ZXCw(*;sv{<8x;7?r71y=FtmUl44KSk+;}yp2(}{>obUGCiV!
z4U39u0%7ICnI6yU9b*tB)SsV@pX=l+4^%p{C}tM*vID667#FwfzG}yvnYK-!&;COr
zs{TSAHD#bdg=}fLBtYP<wlD0&mm%7^7@K{41o;gj`^nqJPfv;uE}QYi3Qk`a$g#kG
zK7Zs@==KQBNJ@}xS7herjIl?Ei+o4fp*>VRhWf5W!8Xk?R-%fCpKbbCyF;a&`3L<B
z1-yZl7j={>pO58hiJiKWyVF^G+vE3ZZur{hUH<+(77P|z04X*`GYOBo#<9No3G46+
zMydg`MXK2#&2Kp6hTYF*+P9H(e&S1PQHlmRABL(*8ukP*9Y*^E0>`zJ=g*%@02>Sl
z`@(Nf=`atoAyPs6KbNZ6G0B40+9o0=AMn_gzFrHNAt?XVX8Cx#nXIzSV<z^b9Vbz(
zI8c20sj{^e!<~zVR+7VFV=Cur1cjRQ&y>wB+LdB>X04LRI&ED+m9+?1%><fWT{=@T
z`<%^m{PsiHs@nzC@ynhsqqPh}Kc9}*>SInUP`_ckX4~3EO)h(zjr|mON_HDt+r{yk
zbFNvB)7^gmkPz@{U;%lL)U%HtKcWx~2v9LtcUb*Eu*C%89AMK&=VAPt2x@~c#GLRG
zlOYczXUxknoYv$(V5~I(E!>g&b^Z}~t<=vX7aRLHF$a?NneKkY{8ce36h`-%avJ{G
zad>*mTNl;co%CUryLM;RYvt&#P$$vhb5f=^r_>ZQu5VsJ%eMH@%5bC3DzthsLbvJr
zzMJIM$=HeqML4?cRP$I(y&J87y-|$8(9a<q=3&2TYs&on#CwI~%WHaF9Oq9qQaY8K
z1t9ai)Duhyksg$LV{`=bOvG)2uuWirF$8JU2<RtgW@pQ`+Y)%9;9>qQFp{3@Op=CQ
z9YTdgIUOygfz}lhWP7tfP9X|T*S$fbls<!!VGz4EkQSxbzZEbgEe)r~exsbf&&j8C
zI#v~Nwf4)Z=xArIIoYM!-->t#%5)9Ej^;%5eNFU6%X(2gMLO3}<7+RhwTQZA_Y3K<
z(MM-jxg!<KPj8f6=i(N)@Wf%Ne<tmg7S)u-c}U<R+B57cOJ_Gx&X3fZ1<q(t)sjAP
z=fzszJcb#RPri&EN*ZDs#(8wrwGJm(6d&n)AIQiSAfGxkF*_;(`S`)XK^{PywaV=d
zo9!G>!+@;P29)3BJuisOm!ZW&hNi$oBf>yjfs+paY%Y3&Dy65=*%lckgNU*LSpPCG
z0Rf$?uoA|kF-l>G=D<*0Sx|92P}4`XnM)RSU^Q(yGE#<?i)Gmg{dCuZUR+NCmyiN0
z@vewmh<3WVQvjXr;+>Pb3jVwU6-=F2CO?AietkuvD&HSjcboR$FeFqeG%&$(I&)+=
zI@vN>bIPfHqm${H6vX&LO+iT}7w73;o#Y0Cd~3Xh6?GauEwY8F`z0lDgfvYxyV-BC
zPn5ctR^>pzoZjfoZE0yq>TzO?sDsVeiGiG)0@*U?aZ>8T%?b?Loj@UX2)a;EpG|{=
zQV)hWkg?G|L`OHPSynZGT3-@yE-;+h-~}3S7&g~8GD2KS;LhO&9oGRE0+dTE?gE`>
zE@@xXy`t%X74MKiro1x1!4n`d+2~#vGx|tnWGLmn<kM5m$cR>I0qX4%;o;WNuQ(H-
z48>N>F?FRDOsu%u)7r0bw9I*GlPw*>D-5M)bny?4A8K?=lC^LhxcN_0<JZ(zxLaoZ
zXt$m<_z-cxmTr13<^GF-ZB!l)k0VScp6=#1ZcvDVSe4mj&kW}-w~)O2P3VPjp;><H
zAnHFdq7HS5l>c~z1?GJ4KgmH^1OkHC_BamPnPxoXm%v!Jf*r&^AV9Y*nn@y&7nulw
z*QyTM7*3m+E70%g%`}Hf#Ia*rTCNuqcP-l8AR@Z5@A0-J$KH_bzWyFtR@@h!K7YSn
zf;UE&^C>nL23Ztq|4#OIR5g}nT1;5KRa}4gEz=oMZ!X=sWKT%t_VNE^if*#Pu0t66
z;T=9x9Km+q#e%4WA1tSc4UK%O!moMZKY*ykVAkDS5O>aafByUsIL-fFPf8!o`9_Bb
z#hK8dYJtqr=Q(%FUN6}yW+Z~T6|6U?-@w)W^H+!hnvpTDIBqgj!sqFn?tkcGQ6TUc
zZb!J&EB}(n7VddvQ~gV{`qbe6_Zt3<$oy}3*8ejNntTBoGGCB?j#WCZ6@p{&4qTT@
zP{yZ#!hHz$uLiA=FsJ~TA8KK^gaKx}+0~c(@cwL`tE$F;Vp$imJ^{Tcy!jvYJjcGu
zi0rTAji3&I0Lm!B!F!bC%^lhkfAooV6(LvRZMk6D$qTxHtZJV=Ptb25R<l?rn=69_
zh@>VXB{8@9DB!Pmay#TMPjT;g5>&;==4CKFZ8i}42L{%G;2o}dd)?!_;+h%*xvZnp
z9|v@yr7uQ_&OW83rMj<RvPKD1JlG$z6iv&(7k9F)?l@|XC|mESsZ;^PvI4DJJXaW~
zU&?z388p>Xc3pS>C{+RzL2zwl#eoN=n2#cF5fLeqk0TQ-Ooqx@4aZSo*yD#M+nr^Q
z5tw2mM(lbdXZto0_4QLPSA*4oeDs2{sqtJ(=&KN$6FAt|3SS7(3&BHu!Ss)zfb0?z
z?2J1WlA-bOBxdYtDHl9{alYPp=;<=Fdo!-xq0ct6A{&eyBjCrJodZYHJ+KbWgDVUX
z3!zWIFcfhEB3NNLw}v#ZLDu4~rskh>>hIqLV;XcHm7J2>Y<BHrid_mXPF<6LX?<1}
zEi%5}@10m&jzQd(cO6Frg@i7GbBx`5Xaj6>gfuYiMgs#3xG(F+&-OSF`7Vqgq|)#`
zq3vac$@T?w^v&Jf6!>OPNoEMCRiB@^AcG1R)&Pb<Jc_(&?ugSBydCJ*Z{6}~(YFan
zPAE&wsVtrY2_<-&V%bc6;N_r0R0lCNxFRk-jD6t-aAI_QIK8P1bjfx#WGTq%f~<9Z
z%w^p^Ek=Us3{*+mP*C$hA!yUeV163cqCW-R6(i`p3(bf7PR~n>yT06*Pg-{IIC@<I
z$|61H7{qY&V&)jKHuy$QAWYC-XvrB7UG_LX9)I=w6B@`%)!h$gA<56iA1NxvBzc^1
zLE)wZ1s-Vn!l)#jpmaY1p>&Zz+A&}OKL9-U?xB3^1YyJ*4RTDCPW;ecU1AAhk3e|@
z(CtyAVy<SDEjO}drjbVk^57A0wxAHJs3#aNty(CfO5eNzYe@CkktsrN^(@W+ZlG7}
z18-JfyWw<~C>b*Kk85uPdqV4n`_JG@jezHiZa19kND!}aKScznP<_xRz^0`G&PNb<
zZi0AK4$M=aDuYDCX-b9=eq(O?8b}g9TR%U;1&}rbgQqpFl%icQY#A=~rmyYSE!ur|
zKYa}f*Jc7G^&d%RJJe7P>OWG7J3+bcSUkPN>Adv-`iwAY=^NnV0gK{6lfao5P;xw-
zuw^#3x4!~vZRp_JJ^RaWfDS5tnRgac4S(fo@bD<_-TTX_B6?IX<KW%7b4~uNYP~xJ
zGl-aDPT>3y7lwR~9L&y64liP0fW-C_b7vCks~gB8Dg!Sb@^`S*ngWLD55R%+^OV;B
zk_@7=016WZfrk*cG4OhzKcKCOa2|t;Cj$8fm>TEF&JQq9!2IfgGJ|*L_p=C#5Wn!c
z>v$#LQ@Zmc9k-4VY^Or#jIgfD{Ky(N#?-ukehVPH-ivSnuuo7te*8-ed93c>^1BVD
ztwxAzY<M5c@rXUUO0D%AV|9ppT4$Ldstp15HbXvH+kxe}14LINEb62bF_Z4w$<<Z!
zm2oi36@wDD4t$Enman9CmjPfjbOT4L2Q+QdX?a%h9UmRLh2KAjz=~M5I|8lJd~c#$
zqSMsXY4GB50M4{LIUYY}heE4|5`jp252IAaYEUrVfz_n9c2`u4L?e>s-4rU@4c|Mz
zM#LW1K2YUS!m|o#lo>S*AjLjSVWui1D2Q&afEjwWI+LP1VCb<*gVySfprBwQB!x*S
zp521j=xAdiAH)ya3nT4D@XQ3Bnw{+Cda!xw+K2a*bpK+-`dWXUu0zy_giT$4e}8jJ
zi*;JiZI(vApdkAicd!g&fuk0hyl_q}p$uPeOuX1UTFx!dZ~3SjvTZ<gImvbFDroQO
z7_nq!Wk*6l*a%jD4%ifd2{7>_g89QpE{Ek<8wl}BA<=TVpOhYh%AFQ87M)=Df~7_T
zlaO<eE)v@5k@*rZu#SW#Zd)6weKzJBgL*MgWU3VL2T_b5<|jnLgCt}`LT%g0*92%q
zIZxY*N6W#%A$|FeRJMOxM@NqQDf(qBtB}eI;Ibg~twS7KXdTB@HQ*FOQW6L;?bGOh
z%`GfswE+7VbEg`3{y;E`%%~eu6+1~zv%mG6OS3RDL+pHZTpx?)b^ZtyF3MRGWKt?B
zDhgW_pt2GXguGF1zuX9-$^gEOejtSqds24E7&$F%*1$AGHpW>g=VUlbZY>`H39M|Y
z&~2zkh|mV~kH~D@!GRrw(==VdAh3#e)OukH_r36V3N8$MWl~6zV$`_+RnIn}{}e1?
z+U$ruF`cI!Y~v#%BR8CFZEQxGXr_0btcHV_RaQ<8xzBa*y3?r@>%*2@D+IAE;z|c5
zQ)$i_WT+SqUs{4H_s5SPM%g7Gr#G97HO;xt$ao!+qOx-IF?qz!&JJ^D#r~MftFcNh
zuqu3p=Q?8DbT=i3?`CMmS2lZcw)$BcG%Gk}AY7Tj{#VR$Uk4ViR0I>ymE3><Om4&M
zBIPp6pX|bWi{D%6^}tr)dUiOE@aK?nH^7F<V%kTGbSICqc|e;6vh>m(kdG->yK?#H
z+$1i0AtiPB9yfQntbK%i14L-V<6KsbF((Ren38~K{Nlxn8$iP_g8%bz#hsF@z?U7B
z2lv5p#qV}BVa17fHjTTJuYv*5Y@{@_iOgf<bXz$VRK38~wsDNR)`8%l9r=s&IIG3O
zFRHE!33VXw#zbbw*AP|*4Gj$<Q}Ol8dT(J8%v?&|L&Aq}evz;@lCLM4vXqncPe&f&
z$^!*_*~<4;q;nh<)KC~Hn(QsqhPC9$29GcZ@qH0}P`SOS;}Pg*Mw>F9DOBzww*w^m
zBw&*J<m4C9dYt>*c0q#T^`&bBad`ItVwxir=AG?r27jJvKAovl&|bnv5S9UIN30IW
zH;ycVo_4+$7itXvRU&pXB4`U-)~ll6Cc|JNmN78qOD@X>hcw7ygZdeQlP1zIU%7N|
z$n+3KV+U|Jfq1+vAisi0Vlpymq+Y@)2hDA5v;|m!4L?DvrCe%tAIjp|G-Tmvur;P@
zTRvVt*qm<AJU%||0aLLqg+C-*6y;cM_sq-;IK*C(pMqnrfe{Oob}LfEpbr6yFx+ij
zf+5jmXv4lBS_y!NUq-p86*k9U!v}ywn+dT+Ko>4AFQ5M@gB7-Ktx3@qIQVW6v$@Vj
zD7~2B+nv>J2q0K1lpEfNA0&Lpe=Q~|N@<0Ut~KU-->_hi?}i33MGHA0dQh*wJi@C*
z(rf&Bf1AZGIG74(8<^8<Bg}>Cp$+J<IuHm5=6x>fsf*zJ*I(>TX-LHt*JF<ZJ0K1c
z^FZ<<OhIq1RuVWJzr))|a4^n+<U)&PHn;{H-xo*=Be2$x9ubRqH&jy-m43xo7~vrT
zk*VqFtH7lomK890|Mgi@BPLj6zySLRNU>ie5!~QYM_TdTZ!gz|z1?8)hhSGMQr>7V
zK2~hAg98Tq9npwxBq>Q6Agc;I4#bKG1JG}MK`xn7e{&jAJ|Wmag}s(V%;gJc0-z2v
zWXzoMrN=p-<pp?H5Am*{%aaIP%?*%sms(FFnFUl{08-+>)~V}d4BHe=W-x?ykQW^9
z&;~&Dud<Cbsow73K^6+`1<|fU8B9XPV358d2Mk4oGQ_eq9BE`R5{9~U0M(?B3iQ)S
zg1Og1-C@wcCnQALfAv8y370NX4Mm_@U}yd@0Z-5!DDWHz?c|s!H~bcnQ>GLnuCPXs
z)CB$hEmLPvLJVgHG=er&8`4ykay1AnI{-^@n6bwpg<B8oeg+ms)~Nv&uVhu#kck{I
zd_N1cI9Om=M;e|+C=o_~$sTx0M7X>go^ZZ`AvmgSh9hHRO%P-*0$+&@q8;Sx%!dk@
z{<_``5@nT@jS(pnSUR$F3I^Z93KFf?oS&&2yT3k{#vlSNLCe3=$U<C~X#Q+0${`2J
zC{($T;YHW;(|yX|sNz#G9i7j}St|$x0TC|K(cY@*36#&2zI>g!%U@xj=F@J-0#y%!
z)G%Ci4oca3AY$r~*M9l>HK>WyrR@NaFIFw&#WJ4p7`0dkrCK6Rt50~Oq-#0T(jG@)
z(0izVfyRkV!sq3?Oc%t`n5<|l`ClsE|9!CXpAN$jD>?#v0>%+a3qkwOF>I!{n3$N#
ziM{=LRiNWZR^@|q@%|3mFE`6>up)o(Y|($byIF@T5Z+t^3PE}h&x1D*9PKkdFvw`F
ziHo5UegrZY5{8Q0=ykmG>!VZ>LK*E)wg2U=k2pf~gV1@X(yq1<w^)u<#(+k&Cqsb-
z0`P}r&1oZ;-RfaU4`9hg58(tCWGD`(0`dFS>NEclBZHicSfA6`Vf;u*MdbsFm#nz7
z4lZ&yuS28E`Ru5_20*U_h_`Tj(LI6GLJsoCciPm2+22BkOC2`H23TT|y^U1IY;!`q
z^oJny7b`-x8ALk`Q^ucfFDZ~+^tVm2Uq<>K1onbAKYNQE7y>x(KxT<4tq1(W;-xT1
z96`HVYhE(y1OE4OIGiG27TlB64<5Wld~d&oACcQsgMM7Bh!eU&NUL_>a_^l6-_c*g
zKR9r-!AyamXsDvaW`k25(>7&*K)BLGA;<c2Bee#y5Nvg%=>e5q^}wc<`{6raZa}0a
zOFL><0D~z}h*kM7ZE%)A0f4Drl^|#O^8y_AvULBY<6eff2<%(W07AFIieG#JNG@hR
z=@lsUc2<W8A71;FQ?;jxBoXkQ{?#Wyg+QG8a702XC@N47sG(0JAW-;H3C?`Q9Ca1)
zD(mUSN`MdZy;OLivB`+ny#b0A8Yrjfcby>Xj-uUt0!5ZiwNM+*8dxg?>uw~iijCB+
zJwWHvN+6e^K=hm*L0Ux?<|~jD67ig&;3h@;kdT1wIcmZ1GO?tj<oICG`^}pcICKQs
zCYYB(A`ucY@kU}o(MKmdmJ}afj+~B@Ga6Kx`=&q-UdAGlp<Fcv4=b3$5g3?~mv=p|
z!5bW|PtsjFXJ=+gLz33yJ?wTf{d(E{M*dqLp9?oJyQPfsm0w5-@Lhpkat2WUyu6l}
z*cJA;>=9nD;6P&&0<e|@#}|74Ue=XG<a~$VVEWL+*jSJNTBf7HQ77ieNeh3zKw3gQ
zprTSNe-9mzyoscS3izx7A+7*4p`a1I#lXM-BcV>ndlMaCGH8NL6PS|lF*;fdgqR&w
z_+az~NtB;_AOvj>l>Hk@@rptQ{k<Sn2b7Yh+vIVyNL_s}h3SKJ+j~&mZqfth(oj$*
zPWQ*>)6){c(v9+j#}fkf)0~>KqZ_*jO6vM5s#Givp#q@RPd@IS#(Nzik(9nq3J#K^
zK^$X{dvQU+@c{g9aB{*H9B(kWyxa<eiVW>KBwA6ZA0U~;Mdm{_r+dV}b{Ii(3)QSt
zc?=c_4jZ`YgG;74i5xS)fv~oJ=0Mpq0p|g*Rp$=F(NPfUI?x#*RkbZiAPG5J3BbY)
zxioNm8A6E$T6+UI7XiMR33@v)ILLe%fq_#r*v>{Jp9(EK2z%L=@(NB5_~dbZCL}7V
zwc%0$eNi`TVx$KHQRe3%i(o&ot9R1zLrr#pp<n^r0b@C>VM#6qNFxm<MaVHGz2K4;
zdXa+6&k^?UU;q9$SNc%e8W!L!;2Xfj{mF#~xJ`MyI>u%o3c0z5s?bX1=`PtDAxC<_
z-6P6NFc2lfLq|05AdnQ5m2vcuE4j*TH35^ruaRmLAUnJrFGUTnw)5zC@kb8%y<85C
z?_VdA7(q{W>eAQe=Km#|s0zyo2d#+c=-h{+IY9Tw4UM60zH-N-+`j~4I${wA!yg<4
zWm56TUJBY6a5@=PH4A~wsNBJ?NteH*Sy_m8IDcpQ0OJ7El0qgD`uYz{`?AuzxTP=I
zO-@e24qJMx5^!M@x?`K2g_kIR?7!M`KjmPVCLS~3XsxDDkjwmkUVE`1<j5?O!fNnA
zJ3b-2b4T^7eZM_85K+*RhHq?a*re&&e|`(A(wiY?x&Q^lw2aceCw+g`%oH0=kn!^q
zOZfDtzdK?1_{%?6{vU;{SEaG#LbL532?KT(s&PyX)^aN~dPHx~&Z+WbvYSHY8Pi?E
zJS@LbZFHW$zAdrQ@cE1~Pn>|IjQVdAz_v*NrMa=UmkKrb1nhdVbLa3+<U}?|t;lIw
zz*}Z~#HN?}D^L)lY=>Zk1jYgu5@1@y;8+B32Yhg`EMv*mtcnG@Bod7X#)23oV$!_`
zn`+s)1UXa$?t!vBn(0l&T6rLFUkLyXk=6(9)ftYy5N{WN0UL5u(A%r{#^AC;G*J-i
zb3hRCtQX50cMmXPo$aIv<ZJ)90KMH{;VXRD<ylrK&{t4GzXK;gC?2SX+8S8g_AE>J
z+N*adiR>c!Mv=C~Z0$`G^omQ@)IYrRG<BqEoQ&V1Vf0FOM$HjvZR9`6!JO7+&oP&J
z$eL08jGV9(U|km+1cgG(lHi{QH>chDSOO$;Fy7{=W`lF}yJ<f?+@jr{TUnbSfLr9u
zEXY_eBPjrXuKUTR=%|2vhK8mlDV%Mj4YQ>+R`6*<WBjX<q6xSaA^~^T{6hF1h369T
zV6_K(FPxGW0;2Id92OEVtohYmV>!+bgO(Q;J>ZC)dd8j4)NFOlbZ@hMvfKzf%iqcH
zZRy1P`rGrp>&EU+`OaXjUXjVIuM@Nn3)F4rwcD`W0^K87Fl#$8ZRI>91QoB<NT)ZF
z`l~ys7Mm^NEJ)46NdWNV9zJ|{kB$x-3@d;W7lw*xAg~c)k)~t>at&M+CJ_5D)r5n5
zpqCYcF77f;@=!{0vJ9Nz0q81!2hNj&LaMpM1!o7*vz<ex?5yA;xVE&nGWZPUFpwd6
zb6<dixpUbyG=MR9W@0E>am%yhc9>RIuBYk$>g+qfxm^GMKS-Sjr9xyx_Q;5cd>WJ@
ziZU{@N603#Wn_~*Gm?-QK6XjS&feLPy|@4S);Ztr`F+Rr|6RZ9I$f6|$N4<>^E~(c
ze!tec__&S|%iAYHTw(4&&+~5BKYV?%jqm~A1+Jc<-W98h*R?6$N||$oz0iJdP<>*|
zH_$0GP(0B4MdeUDG#qf)jzE^x<qG-&EjWpgV+k3~0Yn0s8rZcM`1ylCL<F!m4u?w)
zRfWbM5^V4)z=G3su6+9{Jof-iu>rW60*5sCgA1;2vdL3MFGUtX$@YH>9Jua7HPjde
zK3oXrf*>y_cV7Xn83Dxk;(R4p@e-8V<by@9+K`tRyR`xhNa2t>%^%<>g1QJ+FBf79
ztXu@X2t*_}WgYSU$}dd5#Qf%{A@{|A5ipq(_0)6IDsQ?!oTYd*#Q%7v>9erB?H&DB
zS;cWQ9y#8-m!cKU4<8C=EJ;hfz(`2};5UGFPajeNE}is+s{!D??h=Pw$M<<zS<R&t
zePfnGmLbl}+ECYnpSmd)YKyyf?kqdfLamVOd`v_E9z~Gxofff7Qr?6U2t4{+E77p0
z6OCnhT9Hm6kpRSlU3-3zL`qf`Il(~b$Olj+I<s^1>L6ST<?799epU!u<^h8kL>PdP
zqE)_*4(lQYCpc7`Wm-@!rmNwCpp;a6#{?TT;@d&x3FT8Qv;h#IXfXfd6ErR%4jN`=
zJhCadUDw1shWHse30aAG9aMR5Jkh~%)zfyJba6exBuEC?3+y%SwfgQ(l-D0@m{`b&
zr;u1q#IC*x-r72!Dc5~LjRF^r!kD1a0-4yA>({@kEN!i?H$xkRv7H-GpDhvwe0gnY
zdU~3?Ne;^+PYVp{!omVHUiC5KmN=+JA-rapZ6zA=_D<B1fhR7Ae@W?!h+O$WNPmJ5
zlPYbC8^MSh%@%{=T?lMJDCZEuZOADo4qi4iG>kZAc>Vf8h+rgzP@FT+{2kJQ>K6`a
zrFTr=v#B1l<TGeIy>{t40?FZcg6{dqY2mE4rMI#%5QC+_aW71bO6^e2LphewJ@F?L
zX4T(w>A>M{6;1!4!6;nxpbk;~P*1RzN!oFR7qrp^cerSuvDLGuub+57tI9S^K|ygH
zOiPH~ooI|PXqbB}jPMfJ!XU*45)xkAYERqpxOJNv?h*jnZ=}O`Al=Uw(!E|nWrr$n
zj9KxF5K5GStOz`mr}z_!9aQ8%q?8kA43KN402<noUP}#Cmbspf-rx<&h6V>=qfFLN
z?n8CA8vU)S*zO$Ma`^2HF~Cdx)L^yv-~uHZx(|;s4qld(mCa@!61W{LoZWSw4PbFF
z)hWI+OHIsqmFXeiH?BQOp`D{z(fN$M-cKaQFtRPgnw*Hl=b8xB29J!Qz*}C&h$}IH
z&JVh6m*^91@&^x<=MqcDp0x}-OhQuiNT@aG=JB<kQDI6g)v};m#6c=b#07IYKtEx3
z<Arz>fVre(WONsXt7eVba$xH(ii1bx@#5&s_@Bex0{(S`-}HGR71BS6-YVSli%W#`
z%s@DgkcEnrUR(zJ#434)fm_<1U?Hj>$9@B>hZWs&*R9^c^<SEMw*UkJSZPS4`u3Pe
zX1p1wu_1E-G1<`dLUIMHQhwiL<2U{Y8Rib{XmPH;`^~K6e5pdeBCA<%ofr=#52VBR
z4CeIXg|gRcoM^_2<~s0;esKKu6-A$Ar}-uP%)wS-mpum*%yZv5xd5IGCXquUQapLP
zJAH=qYcI)Jq+4)f8P&&eUC~6NsL0P!e&s@s#6NI!%`YwvgSK3M>{|%);?~9n^76+U
zN_zt509qN8=GLbdU36#oo-^Fs_IBpfGc20txCj`hwo5euEO40xz(C$VL}=Iv$T2JU
zqf{81n`(W>ZhPGV4DFv_WhSAMx?{c4G70b7oxJ&vT(Bj4HU5}m)_-v=p#wl-aFf+S
zc*A%jB|N~EmN{E12xb`khCwlZ5cP4GjI4Ty&DoqER~SCB9se<r?HwmCL;TO`ezwQ(
zI;dE`k%%P@Zt%hB96i5!2rJz}`A68(Uq9wdtm|21@X9sS9`5Y)Dn1eaK)}j(VY#YL
z{3!L-Nx~OyfkA1Unv#gt4eRO{f{MFp=yBm%>e9-J9@MV@4nevn{J+oT=S@IJClfO>
zVQ#k!B<;9ud9TA!(gg_+pf;F*%q*0RJKKK}c&Q{#%}Xz8JBgi4iceCy-#TOiAT;z>
zsFQ%88cBJ1a>RUOi=kQ*6fA}<jF9BuR#0H)H6sIs2o(STU>Re8V*{{U7P(j-u;AtM
zBxshqlmMcM#4J#1A_+*kV%pX^97hpCrqn3@OS3uKb<YL)*^oE{g{O6Y_(u5cBNKsw
zb6kR3eG@(_u|-Tx<wvGOLrszR$5njw-1x7A=vUXCIZ=Li9i1RHIi|@q)DupmLKsfH
zNkM3h<5BZ)8FW&XX!L@eZ!ZJu+_1`?a`4bMBrOIi5tT>4HlgweRPZ3SP&GBBNKNCQ
zZHs8o7vKEJgsgHX(FKsib)dB;2l5ZRa_>W!!D78Y6Ut<TJXZB7-cdk@3Lys*Mg2V%
zJ}tJ|rE^;NaYZQ)KEetRM7ccM9|cLc=wO1KqOaQ7*;xf7#VFwr_{y_POwUoHSnlfT
z+ozu7v%|`_8pnlI(8v2gk~%gaq9XO_^Yi3;m=80jvUaAGmc66`p*PIz`UQeG4bX_?
z#wsa_gSs7loqY$*J8!ik|I6?P`;sR*J@b7_P3@?cs0VC80<f--P=~3dm9`kHY*=vc
zH3&-smaXcbby5wk3_D@-M&drIaY1;P0=#QdN(#gPHDL-VtB`C&YU7_-5JLks=I~Y;
zloq7J-*}!#J$^+!MK<PDhW~Ggb7~i<BXk6e@~rJ%``ugMcK~Ywf)xZDU7t8Qh7}65
zzeG(m^d2`MH^gcF)1k3}FYz2ITohnzbpCYEHqzxJxYaJQv$Cdq#lOf_M4%{ngHQ=v
zf6#FQqzum1?8if&b7R5keHjiEh|@i9+VdI=qW3?1GH8}kKlMaR<cN2-d~d_iz>5gW
zhBx8&N<`SL2IT1J=^@APY~ctg*svWswgJdLLVNMG<1EgDOnlu9Py!3!zR4lK3A)9`
zG;X4)Z&|Y`Bt^$x!iRosQWJ3I$SMBO=%uDeL2rnF0@*m}WvDLJfl+sW%+e;&cdd{Y
z39Ci|dN2}b|JgzTPG$hi)M`MFb3a!t^$b+7ERgx4`~Xx0aL~Z%zr>-szbt=6Wy@cK
z<XsVP-DKy%!vy3Cppzn*Is5IIXIlVN10XcZ(KZ5$#2fH@PN?Vax0`{tzSY<3f~4-z
zK*=e#o*eWgkjHquz*wSV)NU7DC?v@&qFg@bEn5_t3C5j0)Za(jsqP16NX3lK%=lKx
zTRDDxz(oNFAg6UBJj#>5^=tTApvt$NYV(3q7B3(=dYW_|<x8&vyQGq%dzkj3HF`mS
zeVHgJ5WwR-<^C|*j-BqyzKlr<1o*2+4WG(5=3F)oyak{u5Jl+~e{0Q!dv8Z1Mu?sQ
z-F{96hIrY74{QSeH7#zkli%97^{_YC4W(3~RoI2`w@pBaBu~P~n5c{!2XVOjy#P?f
zmpJd#t@rM$lx#Zwmf#^J_}eB4L3IDc5BXQG3n?2XbjRAI)HPg$9$h(EoG79GyB(fN
zD#<(T5>s|CvEn|{N<>M0-j1|Cu~63U``&hwP{l!zq8X2h@WU-%A`mP5I&a+e2Y_IR
z{_w|!mebY)GbRJ!s0+{uLsS)_9zskl`F`t`IE;Wj1gCl|6wtYlaSItV74Rg2z{N0i
zLn%|E0i*;}psIw_H*M&9L6GxI6Sj+z-h@e|R11s59`f(v_-@Csv}YC)hp{#KQ7=2l
zY%A%eh%0uR@`D#gK3Ue~5jq?fWbkwQ`bl}mbToM^AZPa|1Y<nCd+9#}hqP#q>IhTk
zaao9sH1p)23x8m?@z8YZXip9+M0O~r+hSZC<}yc<=*0KFdbd49I5Ft<h7kj+1<$+_
zyb>r%OAM4t%!^H+7XpLFek3{ME2;qa_!$X*L9PpL8s)N7KOlUk__QT(4glKM0nV+)
zG!vLh?CWoN$Xd_q)WB!j<{090SWfP_U$;uv4NIIdQ0mB~FCGW4MYd=e<}R7$c;2XO
zYGhcNkJa`UDBl=OqUW|1+p^!#e2>A`=^dh>xOT<%y5L8(q*;hu)&W-J2o-NdjJ{CR
zDPAoMfG+GS(V$I*N}Cr6DwdQaLj2l%fp7t07*fTRL#jfJASR8<-TfiYiTtDV)TAb|
z<Gbe{Jsn8qksvXnlr@}SlihY2P5(4FIRUGS`7aS9v8vGX;O-=|?n(@QFQ{v<`dPha
z>D2q4D5MxW<h&^N_BbrCZzh$U`0W!ot)#!Y;DsjGTK=UAp1{6WXHC&uT3iIVfk6W$
zD6R-H0y!QE9kbJ+;bpMcrHCAC$4O9_#K(l2JP2Be3clvO7HxIMqjtG^Y4G{B!lyC_
zL4af#Rv{xIsRIm!MwC}0?GhQ*&a(Q?PW&K?vshG?&lck=HBY!(-dj^AQ|X|^d8pIr
z`j!0LKCpv+f1lmXXvxDNxKjjZ2@tIw09S%W2^QAv@NpItJ~6*B5eK3okP?BF5cBvy
zCEqov52PBk!FlTkN9um-jM`?<?DVZ584y&%p~7>oQAYr~7^qi}eiF7BgBj_+`n2~N
z)_cO+I`}s?okSf{MVJM`8D`yuAiaWoV(-;m`&p)}$zBDvdTvboSgE?v%H<2%H@35R
zYR0qT?6xGuo(^%G-P65@YdrDD<;sC7oA~w>xnUZLkw%5^IB5+<4#MtJ_K;LFkWZFj
zW>22m*<Ehvt@da?7E*f%n6s+y{-;H*!daI(fdYWpTqW-bzmtYW6(~l;;Ul{wU@max
z)B<LXgGQVRB2pm@1(1bSP(m**EuDho3*f&|e+tr|p}DyxKxk|0wnlP>;8cRzjugEw
zK(}FYa&k~i22^n<`wIcF$<kq#z<-5x=c6SDmSh2zJQo-j5bAY7vx(b@xM;%xh0*K{
zc;-1bLu*vlJIEMv<hZIYFk5Lu3eM7twMSByN3=+{%~^5m-6uJ)=L9{}&CNgKCJwki
z{2d44Gqma4)|kxxBydF3MXPn%V`Hq}YkPBduq!+3r*vj_-7Whq7tCqpjbl**Pu#fa
zWrkA<=*OCi_V9ktQeA%>9F+Krky13O0X2lM|5>Hr)ObK?0w@2aTbEMc7^_+&K=48R
z_Q|_Llbz7P6A=+niCCTmJ`w}?ASnj;MG(?^OGZJNFAf+m$QraZog*JXM+;t$Kw#-1
z^bZfzRXUB)BC|%F-MI$A!NHYKDSLyN3zJ#8N)6XP0{W&KUK>9Fh~EWa4vJ+*bs7e#
z{7{i<=E@=)T?1@;e`btq3Vt!i_`f8DaAOV>kTA_piVNGtVd97GHyGL9U(!~1Pm*bT
z-Ecz6w!Y_%KJA|YA}%wOBMj-P&KOI-nDj+)FEvBU)>p1?ELlv?U_}Yi<S_YeN8QL<
zvsi7@nTvki$8_OeHF^66qC^n#31S`;g8*X|?17B0`S#6Hlf<#twBT;Ji}PDq?xJVg
zZ#?J{kem_Bte|FE1QA>|2-Bb)&R8;&e)+rz%Kv7#Hh3?q;TQoDlL5FmT6lDl$M-<a
z1|p{m@X%F#2SLkzuB8oVUtVP-g6x0>j`GKctT6!0d>*re=g*HE>^hemYwGF{i8`AH
zN{qF8W-0)B*%rT@F#eo+dDfc6`;>Q;k5>XiqxQKCCZqfZeG?C8Z6p1E9jN$Zz8a9f
zLWOb3eyYoy{2Y(M{2cWW_M$04tN__W?D6*R^biiGyM0u4T3K2{zh&-u;zN^kaY@%O
ztC;GA#lsAFdSre(MfgfCE#zJS<NbXT6BiXQS(}6{!;uLYR4%GJhm9QT;z*8((8Gx>
zqcKihn)o72vnrIj&wj-BG>|k<u%PTcfEQ~Z)4$K+7BoKkOZ3MR0XiIo!)Danb&JMz
z=a(iL3jq4sBxs-}`gi{ip)A8He&#ac>t9DfNK60A>rz)roAXq>;ULb|e+~cDoPq&=
z)I90AO_gl>y@e4P(FEp>%c4+kbX{}K?&6|)$rA)9OHC%OzeaNE{k&~hYy8B=QN8Gj
z`vD$yk6V0hJ=7qDjEzszdQ5NEZlC|@K#QBa!KXUY<w_q@CvM@{c1e<Hj3t*aE`SZg
zWlL|jLGnc7#^Y9du6*nVV{>sQG553pCM@^JPdRCm_i4jdATtY<Fc)3tI`>nR#vZ18
zI49f>B9$iiqdD}X@vXd4d87;r%t??0?OZs>9^j2c?hNp$?oXST`CDj9rA4C+%DMml
zg|;LN{*TZWoEyAM4DW7JK>`UCA<uU^x{3KkXWdz8$XxbnJ6!v#>};V5rPI|Z9lHZT
z6JoOZi)N%d>zbC|Np)6pEa3bc_B48KrzOmCj<cJUG%(-pcr8uYS2ixT0;jZ0g1Pu%
z0XEwkmmqj?C0U)LiQ?WSg|fUnydeK>ONAl?eST0EqeKw^46Z<UH|(8%OBRHj`BSh{
zQ~-63G(is2+120FmV}^4gZ3DN96Cso1To-9`+xy(%ET*G3DP)SL19!4oCBl-Faw(h
zhUT2EfRr{`6hqQC0#CrB@WScnR@W^{%5rC1+P5CoVIx1&4L=>#GR~#KGzS0YSf{U_
z=*2k;qDy>6(zhmPnZF+oD|7DHnUhF*a$oozOs+^*&VB~y86kM2AY=;RWYDu%Pkw(1
z#}C>|Apf;4>|6?fzvw5S6h*UCf*@fUh<9)(x#F1YK;7#Eni!<~1n}}Cs5B544f=q)
z<02s2V1hUnFew4~R-w7ONC{mxXOASXV}Znewawh8URmtt5eCiIOsw|`>M_WNUNKi|
z=xiELMm?D^iqQLe3d%XYcd$kO*^=5jn!I0;nb_PlG?NEd?BA29_q&{9WOq`X_oJze
zubO?eAj%OiN#Z$2@4EK3my&Tm-CCY?O#k#mm>rz_;hIpSb1C-puWl34bQx{lhu=Gu
z^lZAGn)oNi=kqp^KhXzR2y{WTw7#I@2VJzmC00n?0wy2eE|k0sT~wOd^QQQ~BL2RG
z1tw+d3q*nZ4l1TTE)F*#Gf?9F6$#yYrWb~uK}wM%kR64P?f~c3sLB#63kyGBd;7MK
z^&bRq+I@wH+MB>v0FZq#Gapc+%D3E69M+i+U<GmO&|JQNnKIE<yU;!!gJc2_xIRsA
z?1#ZJYbD!@brZC0Mr^m>5|Z4ZYIX--*a>MGs0E;O6`A`-dkEMIG+B4Kxc3!qH*IZK
z9YQkx2aB3|2588Au)POvKL8v6{6s5t|L#GIrQk#c=vgH@53;D5Az*Wf=XZN5#SGzM
zIS};;ioPQ!OjXdA$i5a1mNba3h!Wj{kuQ}X`$TU<SDCY@dC9^-1g_YR2Ht2ahHe^t
zblJtlg=*dG4rON+?~~`~_lRP0;miWxq6A<G7+|c#kT?s1CXmRXm_W=88g1mZhWh6!
ztS+S8T?93!$lQz!M#lXi)r9)1Lw`PVv9TZaAB^`>4sn?xA|gQelT%W90F;%jQ?zEX
zG1oMiw~>Y52bEQ~=8LBc+a>n_2gu(6yY@A;EunCmIi+LA{B>_#x}H_8X?h{Ck?6k%
z*`yx4vQ}Piu+wP1<|1d?Q1HL;J6r#zUR7>Jq$Z8(LVvkZvg5fx_5lLLpZ}(egM;`L
zj8-PGHSApP0yj&zH6GDytpDxVeI1>So?S~KVh%wLB-3dHo*j)^8YmUTf?rfHZ+7~3
z^5wnnfK;A>^#tZvA3$f2NIwNC&;lSkYIASst;O6j!+dYjW@<@1Qk~j57&OPB>#EN&
zH`iUWta<#^*veW?uEjdjd5boJACQtTKkg?u7CNWLFOYrB)y$;l<-xM2ryFW@Kb6Hr
z<XIe3A#`YO=Z2&qbMJu{q|W$pg&*uiK4YX}?Z{H=2Dw6;2C`@$R<c6a0~y8OWfUS{
z*~vj#=Awfeiz?kYOtW#s??dOv*dMHWU+}&wmP&u4Y<ilO)w9WL0C`qNklA&g4nX1k
zKy=jtuiJoTtt6U)v3?j&qj;41>-k?q6`83Xhw~g-#m`1kUQ4KDub)^kP3v>eoU!l~
zjTcqgD_Xv8W<)Lbrml*|fXJ8Ns(oRDp7oK`CO64kUf%b4SVl_fxvs|b0^G9_9moDF
z{Si))D~z_Ewd-v)iR9h}TRkS$FUI1<xQA&L)=E@*?0IXZl~EBojMDadbLX5I>Xi9J
zG4SG=xo@gFT`dl3CDdesil*hCP3G%8J={gZ&HG;nmS1@iXtkL4PG-$BKX6&Z?<U;9
ze2LBfv9@`gwlKqpayPheBf4((#Dh51$gc$R^Zs$I<GGYLmyKe~vaj)H&n}_+ZN!1Z
zN}KQ3x?KF9CxsmhiD2FI>kgkOOUiJGiy^xnB%CVXNN3*IxK+0obP{58AWaAV@sBvB
zZh`ER5VS+(7Ozn-MBAJCzUa)!*Fiy7LC63bQPp=4ML@><W8Z=l*@HI36kPFuRKOl_
zl$!b*AoMlgI35%V18?-;qw!17(17ZB6qG8D<Ls(w?V8J+9id8pn(#*v2$hhd@8fr}
zz{tg&Kncvj3x>6FOmuo%ALV|G=;>7rl|t^0D295SaMATrrk?XRyx5+?pV4)6YURmW
z8!1~Z;KC;8?YlA(-|FYBmQIu_t2H|D+M#bC>(c5^MNBwv%QC&F;exBt7xA(qs=7;-
zb*tvs0~n>*n=q1x$jxUZZrh@y|HqPPnZ4&l$~I<JRsrKSHU$3xMRpw&YJjeN;XNV_
zdo#=J$S+9R4u4?}s%eL*)D&cc!62CUCinv!Nw;cQ<IV#84|H2b{U%%@>^fC+jFges
zOpS`-;!c8Q2yAk9xiU&YF)wY*pX$&Z&buHx7nL2!-M83(XjDmyMD|)a!`f1!NCbw8
zDruC$x#g9}OrhuNM!I$|Xk<+os^M7i=1uzjyI9GyI=s1;)J&wVVTc6$pWn@Ed&98f
zxmBoSC}VS8rSZzY^QGwvt8F+z8)zz#KOC?_Z5?lyn%<8lhmgG`0(MQf*aHk8%0)6F
zP&o2KdjY+)1Dq{LKm>gN8qUV0S$5i^^6q|a4;qBiKFAkny!<$-t}ywgK^Zov=40LX
zvGEzX&mn7;nrRv*wq05z2H&ytt?amVjmfYJF{`S)<uEL$k*wUVt~Q;rvyyVz;z?hj
zZdtBUS9ZexqsW@8nnLkMZcWl)MNx%*E}Fv!1TGSAV4$T#2rPV{PJ73xJ}GtB?tc!J
zV^ItTY;*>|^(Vc38w81<(1;)y40y!Z>cPETJNTOs@p9-P#lU+k1`|)v_?IC7f4K`#
zZck8b?5qm_`0+)>Z1TxYT#B)iVpO@c_3(tTFkkRCoxGv!oM(6|-__dc;tL6adCRq-
zq`}40<v!s*#9gl!q^XGHO=HCm<$UjJsw~}knlPDor<fjdGliBi@cYRmI%{mxKryM?
z_G{-8x~B;$9%SB)m>7>^+f>Ur&}MbWO2cm+`%OIdUVB3b5jx(0n;3^O;30HbXr3Jl
zCug0r;G^5`t**pWX+H##8WC4{X8n;!?+XrvdC<<_^8!-`$L_QX60>dt#Xq~G&O-^<
z^JnesENIAbI}HAGN;6WDYo6m&%q4VNe?o%EQY|s0y?1`ubu*M{S5XxHaZ!vCZRo9g
zm03k}IWO}$cc<7_l;tBKqlzgLDeRB2(@k8(95L-8qCaQiHNq%gG+m07x96wD#1-;j
zl&|T`6R{P^3(eK=U`lB0l55*c&qoxI%zc$q$}`xN&pX)hXz5e=P6+0JHLgQ_HFYHR
zU2FmUKxkytz2$uSyG^~BS0_+QhQ<uZM4tBoJu@mJ;4<$G4|_%4nRu&-2liQbz7qiN
zL_Yz_ihriz=)vKo7b^;$2T+4Fg6IwPwP?OwhD05s;F$gF%`(vv5And;RQ+dT%^va{
zu<!O-=H7g^I6J#i-%}PTS{64a&f0+uFYUx0J~5cmXq?~7|IY9LL-ROE#n+KT8*8z?
zKE~f8l!{0u20||xksl?hcvb1BQs^q}yK+D21CeLByC;n%mrw(>(Bi499%8pvV?UmD
zTB7jdc8aWt(qt)|Q(qS<Kap9%CPQ&<u{G%CZF^!&ZUd#Gwe*^mn-2-G2u_E&jQdIJ
z{3p3u(V}W+3(3tXvYPJlRCh;JEA8#)-9n2Ce>G!@Z&jP_O+5SWT?)Ix+j+?40*`PF
zf{Pj&^q`K1cLW7hA!!tf{Lj+DgVodqi61KIQL+{EA)U6JpbW0jTe2^4#=2MMy?JVW
z6!!t&MYt%gv%tB@!?9_PY_}_fKD#dBbH?rB=Ccg1bwpI0a`p^jd$Ob)+q^LP^oR02
z-&TDuIX(0P9opuY==JfFYTdF|4~qM6C!BhDs=B6YoA3rUR_fvPTiq{dc!KngFi-aw
zSXh^mLX6O&KWkvwl)9hu!rIidV$9JIdZ$kV-XGsaz+4gq*T4eP-mgslVxQr&!afUB
zHIm5`GW|ftc~It~(G1|NFzoz53*X@d@4y*IDF?di^@YK0okMe=a0w0y`V4j_kd+u@
z4DJ&Uo|7wEd>dV98F9c-IX;_pjR@&P6E$r*AcYWef^_~0Nna+Hj)94dK**ov;UNXT
zDMv=<@#tS$3Xn_%XI@o;q+TW&@>U@uD}-$8N!J?1?~9FlfZ1V`S8B%*GefN_ku?0f
z6#K6;PpA6Y_lE(cjMB>`#Fsd4zYkEld@216pP;;w&2Oo=jr{JySzGJ?J`*<O`jmi5
zsnMoCE#)Jv6p#RZI>+<nQO9(irzb0a_q5uQY1Qn410RGhg0vD7&%l%nEE<Lh>+#Iz
zH4&i=r((JmOdfpHmGu<qW5DAGDx$m2zFZ!^lZ2AL_;n)wE~W0<w*(MaBIwfRv`CI(
z?{DrSDG0=^pTS!RYX(z+fT{!c_3x(PzYY;9+Hbj%g3!DQu+(kY0Q(4YR$*k^i}3Ks
z{x8aZ@_`z7E4(VDCd+>0@D;~C<%ENdr$4jWA7-jTPYpUIjr&4#>3@<`x+K1DCb70{
zlaA%!Wu~m;eV;0MGI77lnc^}rtS59l*Ae;l_gxTkyFP`&jF|MT`aK;QP@0kTh!OBm
zr;Y`-q$}s*V3g6xBv?+?>oQfkYZ){8n4|n#)D)LqYeLg{EYSPQ?=T~uBEJ2YO2&!b
zD#O3%x_`!2s<VNuh|xw-e(I)t&Z2SBhbJyV+RuJ{l?>XS)P;J_q_*I^?P^O)A~iwz
zxPq{V8GlWUp^90+S`6UcoQ8Qs0NU+)Ti~2FHrDG#!(QQc%HLAd@Fldyh!8?-DpBKq
z5=S0|12E&QV(4uqo0Fp^xx5>%#(fNFZth&BUnEwamTMT_@hxAF{~T-CRGO`G{`jU)
z^8FXs?v0mD#iB1+%lf_!jIf%#TS0QJcK?d?@*kbsPfAYn6<W%R;MU6FT8Na)s44a~
z57<O`+F8qS?XK`eWSkIQ3(j6qkB-tLRQjayZH=|CT6a5A-Mi{S^fMm<U9&B>Jo=Pp
zca^V|!gSYeM61J)J2dhYFC6?1prdC-!X>0Y@Xb$_^=Sj?0~+EB@@tb~c7v53!I|*K
z4hoSkH*(h`C>s0ldv1u@FTC&Q8%%>kw!bCg%w8FDY$FN&P1|uTzKFsP&gau+%U^q9
z92;ydkK&VCtQbIm(}naaw`=73Uv^LVC+@^~mW*@qs$?A${M52*R&qKUF5&rI&kPNk
ztLL^3y>Zr$PPRUNtP;zVwRBC{sj^m1?n_;Z=MTEzcXn0N)RS?ZYF!Z7h-O~^zxy}^
zlm#GguK=;7amghc#&^X5YSfhY6`j5Q-*=Pe3>Q8Z=uS{NC;D^ikPaC0@vVGQ#6Sec
z(c395(!Y%AEXGHVAJdx1+==Xv8Ij~+fGnr=^V7r?3ojN<Z^jcJ+H*dPS*_t&7c7tc
z9DUD^d*k@lpwDo#N5AmAt-xaI8vc(#%X_%7Eum9>a?e5IyD>;|+Npq+wQP$RgO_3c
ziJ~L>$wd}MI*^LI1epdrNUxG_NJxMvd5GHF|5*{yg9!W6ZfCeFH#S<tr|lEy#TAJ;
zSLqBk?ULonZB3~p!d+Vo`%<v11Oe0aV;(z3m?OrgG<HZRan0efdrQk(k<=sp9|gtv
zc537{Ue<G4y~#2p@!`DSTk<+XQia&6j(eb>B|gDwQ7$NET@a&EjNMAy9`yJ2RAYmr
z3M-si$@;K2aQG*{I4&b|1XSQ^UQRDxzC<I0q3;J(6xh=V1q>+4kEuhw4#L06rY3(F
zqXcMV72t1p>i}aH!bWlUk3V4Y84O@<0<v1#RN+rzC{fH+W1e^}9gDf#D|LwExl;`{
z+h~|BYif7$VBE)t0sdPXE%jImExKB+&1NrDji0E+X2Yk~#QS4ri>M{B^X}gG$Jr5z
zy`SV0%Fm;UZ>;gVI2ZRVa^sO;9*><qbOb{rM`uC5Mr3F&-ar}Xefd9E=7x7r)D!a3
z#y**1&o}G;1{O1A?Hck|&&nYSe;#~5xp1VRE<|&0Qwfps7&O)sw9QEH23}N9FsBbe
z@*ij*@OS}gL;xhXV9|33Z_k1Kpr_k07egv4F0Um%tX$j+z*?&4n9cOL%^jCJD!xI}
zBcm#N-R#WT%Bzu=gDJ{QA*ok!l05wG-j&22fw-A#M7auC7l&&^)-P=s6?^+pQtXlI
zd3&-h7A1$9?&{8L`m_luaNJFG;glTh>wY8zYGjrCM-ie8zz%@=1}2@{9hphZPSXVc
zE*MT(5ZeNhKD>AU-ce!!8gQkok=719h3obeKNv;^7F<Xt*aJ4qtYXp{YV<}Jl_Cjt
zNQ`0ol)oDXOojnIt2swc8Ut^aqJHt;co~C;r!oHRp9P+)f;o46c|sSqTdCS_I`@YU
zyBGK-xh_3Pj<sXp?%fgV8?^0NN?`9!f)=udo=AYqaylw|wv)v>#@&w6_w_`|taZwk
z5Z(2z06RViDBI<a8W|M7^XPqkTaJJWLB+I!e}YzH#qd_OyeDjE_dCdrEvTWfSb*~)
zT-B_<2n~`AfPnHc>U@^=E~$QKA!`~MlEDly@p=#Tb#NfYf{_*VS`g47b?X*B5>o-C
z!Vl6Al-&<<iJP*rwsE3RwShlZ7VRe}F(dYpp1Gj*pJlz?!p8?$4_*!Az-H&&xW?bg
z!jExFFG(O~3EiF~&lk59X(D4M;^?H6vYK-g$Z<Y_xl1Y<_T?bne8XzwVBPi6t6Idr
zJZq;kmp|*W6s{3NG#N&BcAE-wqk8Lb`4l$1#7Q{YlCe6hUx_h3SJL{;u4(xl2tUe_
zw8r}}x?6<_FLj#tqQQ65Y&Ged(4E_Q3o*frbi)v6I|`9q2(W|Ea)BT}%Y|1~-heSp
z)t3;A2NPvx(~17#!@y1HL0lWMBcnhZ5aI`ityMqCgHVAS{RZkMtJAdrS0VE=c6*)x
zgXXb=P5L2{dBV@4b_jI(>FHCA-3!4EP6a7HuX?Et(`1b-)ZN@RagFR)BKXli*x4g^
zcqLJ<5F1Epj#(roiQP7)!E}D~*I)bM@+ju92&p+|HNTa^!;RS`vnWqiB13`Da4c=o
zE;o%A?ic1M?HmDv^N7r%UwSoNdrWhg0B;00l|#h35bc#VU(S9qZ}XN&irjWeRVm^}
zXSiWzzz2COZo<lv)P?Nwm-~CLWZ2{14K89LdiQP*WH$&7K0L<$Ha3pLJTp?2eSAg_
zzMh#r)$|xT++2{x0fwLg1G!+nL8g00jJogbav_rr<f8l&C{{5W04^8?<H^mIu4&jk
zFiJE9Xfy#}6@ZiiK?O7q5g3-W%#x~f=gkMA#~b#_ON%{Bhe=0-cZ<8+gW0<dw0))U
zvd&Ar_mSMH>gvTnLeJ+fa?X-VAGOAlz$bX8uB`m3Etxj}b5)(A=V`(bT521^F$u3b
ziD&UA+m6;m^|9D$kZN1ezB_hC_?-N3G2LL3{?@ePHoX&eaMEE`Cq~@)!R_0(i!4Um
zsi~<$oxeU2v`(_w_4c=3Z{?Wrih2_2`etx%^Tyz={x8|V;GntU-D=PASW4^F3Xjdv
zowd<X&7U5Jbc*Je;=kr_+D+oky*D_1i9d~u?R4(hQ%{6i(x#u!ofmCBGdVWamGCM<
zRF@@mH-LKRr2gfibLV$kAI@xVbDE~zW+(4Wb21w(AF2$yFw~1t&Wwz?OY*VPVy;t=
zL-n1&ZH1Yc8DkTZR#?|OSa<lyMp{~0-OSxRJrDXz92C^m)pNL{foGM!7!#_e5_GmV
z-)5#q5GYWoT3z|d%E~9}^KT-fqCRG2#e-jI5=4&h^NERxw^CY-A3f>|nByzWdbhN8
z&}Mf)tJZW?V|Og0X>igpM|++it8My+X#?p1&qk3z5BFA7``-Pqo<`@`*1_^vXTdm9
zx4pJ$pWV#oiXtH^khFA8<DT&84q)_`mnRjN$q&Z}TWELNW@Kbc{rc4kdxkERdt_v!
zHRyu-%l9lp@13V^Z*C5_C@LCKtNU8&EJ(B1Kj!Ci#5HjS*S^EyRA2yn57gcAsck`D
zi7^<rcs{jGn^K<7ayi}U+S%dDV{82Rn<H^o##Rm{m+Q-($Ha?~YWSLT4Ro=T4KLWh
zXkitFgc0E!Q`Jzt+Vc%0{pLs0ZJecuFc=Kp<#z-a%paJmv>41~%+V?M7QWk^Lm13a
zj2^8KJ_eITw=wNWB>R?YizuS_#K2vJP-oMF=%?qCoopo%W8!MLh4s4Qj`4p?8Edj3
zES{Y;Q^#3KmX?-M)6lfSYheRfSPh((gfEK7Ib2KPUlF;g#eDd~y!}&JZwU_pX5Mc-
zo@3JHa8S~-V0P8`;Ijok-u=++I5{;wJ`Nca;<Q*7wbXmY$J114-+VaW81LHsSOS)y
zv%Rx3t*mU|7ysE7Haz&af4sN|-^MVs!M_~9ydlM47%+yf?VtZD3QBq;^vi-7zI|h%
z@PSyP_Q<iRfq=TPvmiWfr&GemT%I4b6sgkY%dG3-&CSh)Q?MlsFvPU-a?OyCkoT$V
za|}9H`)Pt|lfa0O4#jO9?~%+pJPgB8wa)uIVG8`X)l>FotipT(R62$#&scR94Yanl
zCKeSHDfxq7is`1Yk<r~7H>$6P7I0LKiLt@j1^n}FFlSmsWrcK3*hHtoohUmwaCf9G
ze(Z*IIV}EP7xIVAKW<go<+<_R3>$UytqO%eDT$!-SnQky^X@&hu;Gg#SJyb<*WY71
zAumP;@{6Ri?9{&eoQmg~nj4y$Oe`(s7wpyE+%c@0iBN!>EH+W};cIoZ;H675Htc1`
z{j2ZazyHy=nHeqOAlnI|2Wo1|MMnlbB^c9U-*7ViURz)Ce|(t#_}u*cMpTq@$cEG$
UV5YW){|0kYQdZ)_Rjp_L4}K^{ga7~l

literal 34537
zcmbrm1yok=)-U=ZAWDd!G=hSZAl)fS2?!D*(jg$y-60}^l!%n1q;yHA(k+5?r_$YV
z=W@rm_uTJ{Z-3`7Hsik$cv<gS&ok$5{^ALEtRRhtO^%I1q3|BcNGPLF7aUM1wEjyM
z;a{|LCw$;P0``*X_9|A!_D=e?MyN;n_SWWB_U5JrHyw>^?M$sKxmfvGIazMLu(!9i
z6J%qv`1cR6TG^Vg-8?>PgcrGNE%VF{g~HcIKGD92WtgH+ergXT#8sW+R>qy2RiB)-
zZ0>hH*KoDMxX-AJVfs_@@%!YWn|I!oUoaTeW)@R=roCb$c~xz!spx`Z|C2FoA`D+l
zf-uX^GWnl<cLduyPfrO*FEib~6+FG|n&1*Gus^0ke|tOLu3KdAIjP?(fB1;&>)ahj
z5x|Fl8}^a}e7Jl22qIroCnYJsmyKAAU#`KIn`r<2w^L>1>8~jYZ%U+X@mh?0EHrFU
zs|XQs+tcHWiRLm<i0H|^e0sE-qFv><t6b(U<G}Rgnm|uy3>Q{-TjNNHnOSP}dhKxt
zLxQu<&dv@wQ@&sbrNH;s_@Xk)1G$^~`wTa4db4U*Zgg_>^qW=Pnf>wkf$(PIHS@9Z
z!rE9GnnnJ)BTH}pIL)oCt&Q#DRmB#kNspFrMbWmQLc_~=cz45`MoTTCyGoA7ZMr`3
zXt~TsJU211+wS7)OOqo(4diKe;ILKKZ_Rfc9xdmamEf8D{R@Y4;oR_MLT(5RIN3?e
zQKrLf$!eUoRu#-2x6E&fyyxmFIH+ZZM;!PDkAn9{rNc5|OQOf#vkB+fU=5204n#ym
z)0@sqRb(_X54cb7&;;EOimb1{cZEv0bz{2WGA{1zJ9pk;QM+F)8`Y*+&oo;~aGtqL
zuMl_Xp0Y9_efVX#Hwhwc-apNjlvPw*Hojq*PS%RNdi4qe3ybS{>s7eXs=1DJdn0C)
z!p`qa`_dz@f1?Zx44hXATVzvDPuIN`zvHkqT9#d-rtV2R+sUxac)@2i(P<Q=?N?VP
z7EI1F(?p`n60!KZHzkN#EFn6Yl=2GJ4MBUowXq7itvD^(gP(G293DsZpZKgo#w{1u
zz0QT;31aGvmD^10=hqGpzH6;`MC(9v!FzPo_(zhs{rW_Fl#XjaQy>YK!;&ieS~vCC
z##_!FG7bZby0ar@YWEE^wz?C}u&^+cgrsCdjMreEc4wkk;&OgnLTc)5<f)tG*GNs)
zdJI>)aBXgGK6$xacYbOWr@%0%?WCQyB?gaO7#<8RDZ97W$uz$CXlbz5$t-owvtqvp
z=BEZF8Z|C`KYx-6JG0YD1?kPThPiLGF$kY*G<>z1Ot|&H57ng73ro?kCFI7K$Nu<j
zI=aox7HTfX74@HeeHE|@W&6e4_RP1ow<~5tgbGGiwd}fho*Nl)xNJRNE1$h`wDSl3
zUcOFE){dyZKT6NP)x~IUW%!~VTtIUyuW=_SI(vfaQYN8`ki*h*`#(R^B*^uI&kklM
z236AKIZ%fCYvXB4vhwoW%|T?d9`T8ZR^{Vsgf5Qj6Ba|OdxQ?_6E&l#py6VZr$bQ%
z1?(vg-d-VE5cmVn$gj(o>*5tk^<+FQlWuO*Y;M`aR{HMh=pZVnY|?WyR_*a)y~gHd
z@<AQ9Ts{l9&Wn20$7@v++)*7yJ1fI|mvC{pIk~vxb{P2ihxlE0UIaMiJ}XXF_8=k<
zxb~>-oj=^+@h3A5iJ~4uC1y{9Bdp3sYe*Ez$1Ovs#T4}PZsux~7{IN2!pg?Q6`quo
zBs;HN?aaX%6?nW}SJ&7?0aqdP?b|o^%_b7MEv$<dZDE7C>5iK7!p=~eIX&3wU<(Ki
zHtPSPG{K$ytG#A_Le9~V_xyPBT=t{OYRO=v&^sm?+tTa(0hb=Tqc#3PlU7$JH8nNm
zQG44mJ3Swzn{Lz*Ikf6XLrWX*?%hlawbyN0+LZQrOf0PMh=?bd)o%N1m`P8v)#!d&
z@KV4=-=-2{x^)XJyiJv6b!XoYZtYj|;i7(`T8ojdOiGIn;UnT@Hk2cz(4km}<#Go%
z$LxAI^uJ@U*6b)xc<gHvvud>_O9qy2e8<UlpnC!3=x}#X*QP>?XJTM$0>6PhOFO;q
zPtINi4UI`M)Q9f&KPk_MV1X)@uM~A0EQ+1C1@ld`*1tk0W7oS^!gq;E(4Nu7#pPsK
z_gv6$qOnmT$$7n6r9z7pR=}!=4pBfrfca!?jp{wPX#<1~YN>bD6a8f*;ij}cyL{;q
z>WiwLaiO|wu)W90W(#cm0u6^rhe7Serlw7=^L?)qC}3$J)FQ4QO%hx#dZVFV)GHo7
z7@a&*OU64tSrlt)ZJo->&dahreX8R^_ZFYpc%s^6;%jm;jgHGa(FJt$Kci*qWF24H
zj~Qr`2`o(EZOzZ!prrEHQL?Pwl*&<Kh6TBMkvXICXerxeDKmqV&yupnb$2Es(Q^i#
zXv(u<<6l`qEG$7iqQ^nZ`PCnwd`6b1G#>3PA5Oi-|2<sH@VL6GJ@sL@Gi*#U*r#>J
z<Lg7E7O_tG<tl9NzT^n4mA{5H91KmQ$Z}lp6R$-(L!$fl_6X(}Q4iNN)|nR*)dHVC
zfA0Lm6PT2A^TLG-zb9&{1||b!BA~5=75{|Ao8Wn5YTWgy$p;f>!&!I*zTcU&*5h@?
zB_JSBeSUV_8Os;i9WS&E1#qL4HdM9hY;C-1d%H)B%c%W&?ZFJrgv;WMPGc@uDAQrm
zRMTr!>w60xGXV_#@7@_KbjAp8)T8^peQVcp!)XInOvF%1UJ=x&j>x;S@Z<VWK%W~M
z|IB*AY}}KeZO}LWPCZv`p*y|<-gfzNUKQh=I|fiSps%L3%Qp|_91%gE4ZA3A_CM~!
zFFWzYPQ1H}nH^}|bO%zT@u^&Ye?GGG^Y@2do+%J2dc>F_6`VETp_<JsBoqsKChps7
ztZP(MA#KkMy5j_l7H$YRve1WT_NTylquW{?F3#w5rw<Q+mo1Q(cuVaW{Z!W@qTwz6
zpc;v!#2rbek5|LU3vFhlh}m>Hq0K(c*iBDO-RM_Q>`i@mP0;>#eUVYeoldA7F0lRO
zQs4OcevEN~J&$x<x}ORYjw`A94Socal%133XHH11%yPOXk?IF~FpQi7O7ACb)BB-R
zB1>N<%FFpbgoh(FV|Ap&9zMUHczSw9)b^MycE|Tb51RAZjXMl(?u}Y)!-g~9jJY%J
zND%fZ_0Lm+OoHvjo<*oQY()J?6VJ}C{bW^l8fiIRIc?4VVjzbMzH$uiH!B;3*U$Ya
zU|AO1+1V*O-__RkZQk}t^x`|_JUyrw2*!|Iy-JYz5T4dt&JAd_9>;sV6?Ti9q3c71
zui<HiFz1!~WU6Ek5)(HA;5dXzRqY~7h0nyq6ybWEm{_eUx1yo~y7o0f!u!t7{9Z?W
ziLKq;tX%5=Q655Jz4G>To_5t16qcx%e7PuARE}0f(a0(xfq>>GPol8mQmCJ(4<A0X
z3)fnV@<y`hDnN}!kqX$}4yNEUhJ{qU^HUDBGFp~DwyjuZ|3?`%+tYc+W~d229^k{Z
zn!wfI5VK4n9XyuT!eF?_=-`m({2_O27cM3Ld$YU>KdE5yLbE~E^hdGH?{7Ug*snY9
zdh%5JN%9zcL%(=<sM^IYV)QnRvdZ5|B;<L*kJ8uIH=?o{EzRXiLWg~k43&3fq@*L3
zn1Z4MYUZ6Mx^J)C$okd2nCMm8xJ+SH9freN?k!F2@ljFq&=<ZY@TPzCsla;rJE4mg
zw1K`-3(d3j^Rs-_$5>4siVq)Vs#Qryy!1g<GKZi3$CwG&KFX%>XH(PCTy%vfuUy0>
zeQajI+=GrUT(5h!MFMMb1#hXHkcz6i+-BB!`V9%c^%Uyz<;!#-6n>urZ%d@={4Ej0
z1O#Z0X=H>Vnhcn@PWx~ESaq~p87h?hMFJb6%Q)MBxnsdzFu>pcCJ)bUkP8K$CGNeK
zFJERRv5JVqx3;#Dn}+|RYlIa$+HOybS2;bc3OYUBA3XY0Sp;Aq?8Arqs;d2D+X->3
zq9W_C<Dg@F1&Ag!0=MwS-<8DkvfsZsn4;Kmb>wA~(x}-IKn^ArmfMVsUnGwD6vcw|
zNpY!!hhiP!!Z^{($|hU`16eEm%gf6NTLyA9vHpUPF;-abNZ)zcSfJl<uu|NURa3fR
zKG5G4ClJ1~V}Ip_AV$5{**5hRN`6$m$o|*^8JSB^_0v@{6oSZk+J%bBt*7tTo@~io
z#HZ?nm#Cf>+4=EcWvn9n?DQA`j(|X&p*Od-wdn(5Nn6tW`hwumC+^Si0>56?`O7?F
zC!5bTCnm`BK`<~ZHMJb|Z*);wGmosG$-<tUMetj|2}6crkG*Do`wRG2=+k$%hoIt`
z!WV!3{DBH8;(mU*Pbp}BsciI7An{#N5m#<}(Sw_6*}vSn0V3A}yc#Mp!T~hk>*x0~
zLy?-wZKY7zZ4>qxAt52Dn5VF!*y+x*eMyPbN?4&X!2$wByWK+T{Z1-NohOIe6@SuW
zhv4cmO`j2jRi+|$taIsC_Mg5C{LwNi1`ZAwBvcq&_f{Ul)^Y{}vs?vOBthjd0}Tz$
z47&aL)BkcZ8p&AWrV4!vJ(1x0_0(hjcmDqUR~y({+uLbGMX8FP8OoN3i;tVaW0<R{
zOM$7YKL?gcRHai<KQ5+2hJXvsXZZRWC8btRE;4H2#l*yLTaK0Y=N_dlFn$s0YGWFl
zn$p8`fUDq<my3NFN?#)RpNwjGo_D3AuB;Zku#+IjG%;-nB_Jl(nCyDLEpKm5*-@vf
z3IG3&^5dr=m_0%Ct&=*FU01`lwU&E(3Z}YoU}UjPAUhe!-yO%uyoe_W8^Qko_4guZ
zMDK~mBjY3f<A2^7xZD5Lw||H_(3t2I9bWrU$+lOn7JWmU^bO{9BT9O^2Ct2|beHXz
z8{d5Q&(V!zuX*o8tj`EmrhG5D`>hw_>)Kl$VYK2dYNtILTc;cvugT_-l@F}~x5}4G
zFiNR?bY*bq#VFRsaIvqP-9Kt~ViYdL@%9yS_tt%3o@rO?r6bYqm5^O<@?M8;cCv`1
z^!Gyai8*S;!{(-6Rl$M~tuj^?x3UT8MW-<vtm*0LXnA^G%Q5v_Sr`a9U@Xga$dl3L
z96q_c?j&^9Rb^O;HPbB`l5vy5%-N8oJ4wZk(0!9<!aI+XMO?{uzXne_r8Jtwv%ENT
zmW1$ub<{L}?EaU-ZvQyLm0%&gJ^at?r&r(nd0T+tckM?ybzw?|evzQ;oI~2m=cy^<
z^%Ta{DY6dy7vm)G>SGmTCBHDk1fQgZRbt{0vzv5NM&8wwDmL!&4+_d*KZl(tBP(mr
z;D=|wGSnL@YS@g6+wev0wd9hr-e^Z5hOg2`mybJElg%pcwcYEVqxD%Go@nlUCytOf
zdzEN&`)ju9a0vYkj>BR0F-gkvBch=1!WTw#6z*lqU4AhhT0L&|O+^)_hERYn)&0yq
zK&kkLTnoX-F)>T`)f|FH);G>p^RO~;GQ#VuR3h5Nwl;5%J!A6K;q?0<YG{@<OtYIV
zD?Sd8Dj+a0<!Qc-$rDo<nbv3VY|2kpf>|-@<735t)0OFM;-@h(M2zXyrA^iMpbp+N
z68zY+#r{whR_11r_F6Ff&&MzCuj$$<xDRuTQmQwQD>=Jc1@B7EX&IK|Y}QQ?-ggc9
z{(QfMe1a6^q(J^A^H^>!n0)iLadvoeC+*x><p)?wdgZpvER*pO%4QZMnfMV2o(RI(
zAG5<h-dk;W&e{9bRE3yDy&h`R{TDA7Vb^JS?EZp^*XO1Hb2lnp(BV7$RKBj5)w2wR
zI%639U{(T@=^q$4iwq&Ub)I8h=cg0GfRYZk7o%RNz{}VzbZ*v8)Om@~N{8I$=Ki>`
zVeN-U(G1V*&dO79aWvT1bW}7bdQER9!CzFPgyFwlWw2A5$*|ng&TpWaBxX_bsnQ9|
zxEa;CoA@;I<m7=SyD*hJu^3HWdDKS7xwlugTun<CUAmKGw_>C>`%3VLL7@4&HuKNy
zt5@GnCVEe|Z!c_KpPTBI-DC~E^v8ylgRfR5VA?)x`(E1XN-{2E1fhk!e}B)`meXt?
z=N1FQ+soJJ3mjK9B3YjK44W@}GRylmyNTLpz!yUmIByyN@H*V-Q|w@I`|f)g@W=<G
zL$hdn^(SIhEATu$y5o>qQ1D*-^+hOrjqiw<)hlf2fSZ7pMH<!dWX+{V>|FTPty>e)
zv(3TTs`&@3;fH0qp~;<wl)C({XK_p(e{{h`nG{=Ml&*%-vHNs7B*xpQPT5Ol9=9Ew
zp<}Dydk_1xEwBi9*w5_C%(+LFOpC3!oN?~exuOfi)_o%<R_xzsi&k)-7CaiBU2!{f
zqC8r(Sn-;7E%G`dM@gsP6f$V3Cq5#{w1X9Zclq)yM#f8lP4~^sHzTBAy08ckwyN1B
zR#H+jpQw(39!x+)gn3Uw!goKe&j=p5)0)<V*SW{h{yKjB%5ZT&XlTB&^g>rGGKqHa
zP3|a|btQ<z{`m12wrDd<cOAl3KT{vdN3(xB7Ubemc@>XpnmogEBA^eXmLz-o%^Lmj
zM;p?*DMpsF&lA<RL@Y666*B#H%3t|B&}`E>&bM8*zA+rK-0HY)=5}8wF5pJsa@#=v
z`__kRohhka`&4KaY%Q>n?~R^4w>q&3aBv`{vMpt$EAZHybgP>p#XhJ=&|p#C?|zwB
zxjI_L^|GJw6OWmcwKe;rSYAv$pi?4P4S{L_F#0oBlg~;6PykHxS<0F~A~nK<b_pAs
zhMoOAuv6a~8rph#*f{9i1*^sttgWpB-d;lawN#~dvE9BXZ_FMOn25d>H+#?KrpaEh
zKK!ppy(S};MQz=zu0YBczdUvW&1sfopf$FIJnXlJrQ=0t0MsbS7-<_5s@z{|0o>1x
zP1w;^^HAGXhWJom&mG5>>dA*c)(7VKhSxdKbD`xPxo$(FxL8j-_({sCGp+ZAU13v6
zUhEbB{@2AL8J53)e7;Rjztv1(b(NYrv}N`)a8g7HPHItY&+j-ind?tmJ(V;vF+r%k
z3Agnc<;4U_j_a9OD_xUUiNpMLt+@$auVK~C;5KhKw+@&3eE%pyx1~uNOXxUrr&47<
zlj~W|0j8vDV&e6mX53J9=1dGP2h~5{Y+$qvtC*fFncxfa>om6`<k{(wz)iTPG;*tR
z54BEmM^XHWz^IxYUNGHOR)7R<W$5>>uhrvjsb6S7YaK04@OfQ#J<6!WF!;rcho|z~
zI#m<Z<i+CIvfb>>vYv}KS7QfTPaP!Vcb=rCs|tNV@HJ4Dv%noBDJF`d+>ckw=(dK;
zn_>kj$JcB-Oksdg%Z}hN8~Bm>aFb12O6nrQuSMw`jV5n?`N9ahw<f>LYEr>+&B`oV
z&}pr++GX26BxK-4Y=3m}*Vn0DLQ&qOQHFSCzSFGV9qbu~W9u^eMb%AaWGDQ1ikKpk
zWP*|}U|%_x?H)Lv?iN-?&m|FXa@i_wo2bTc(=t2xKEn}n%Gl0-|LpA1bN79!wN6@6
zNmfoZFW+KQlJysr%fA(DJ)bF07rhNXxj<W(z-fEMSvPj|tmG$K*of+C5-TtS{{EMA
zyiOfvW@d)Utf)hUw`2l@dwYBN;}xJj-@AYR!j-ov{f_kF;^IMMTz&!97y#ER*xPf5
zik<Nr7gsbMrFO8@#l>)${46$69JAzuk(FI-*@~7bMLN{D{UZ}D1#hRDc*{hM8&b(!
zet&)|BO~+0^69+r)9A{?9jiq(R;zZZ5YIh{v_9tTZ?v|eTi55_Zyj<GEw(+PU5)j6
zucpu1I$0JV(xa<(=XB#3KTZ-yl04yMekNCJm;1r=+Z@fZ5U3rS_2~HQM(rO7T~1C;
z5VEBiAu-+Eg`rn77v})(?}aPOt37%N1Is-aD&b%7uC!PQ`6<Q=M!*oz2Y}YH@((?g
z`-b?*sn;?6{Q7zi(9xzpQ!fLJnFPBSVQ5cI+=IsuwAI5!hPN?r1;`9;m?B{G<n%l_
z5X=<<Uhe(-tB%#jm&DKZxDs#DC<khUdiw2p{ZT=%U+2`Ir_pjkv95t_R(N7Q85tS5
zc_y8S!)J*{!v*^2nq`)0@cI1tbEH|IeH1D}z%LCQkAKpYskA5^16Ywmz~<Ti`3?oY
z+u7<Ff{B?1Zu(Rz#0wU}X#JOX@NX4Z&f%qo8H*syPyD||i{2M)K1)2{J}BHTh$h;8
zCv#dfvV!ZmItVWiXOe6<yHs^q84r7@z0k6+XM|D-5N|gb`8z^*$^S%+_)h?`<u$bs
z$9O@5QpN?MOlGFcOkZEW0VggA4FA9Y{nhp-yW_Z-G?)(aq#}!Sl&{d>%dqX{J`d?1
zrx5}01p!*=1Xo&Z?9E?EpAii4e<GS7msu}RbGQMd!9ztwle3e<`Gr4&`P2Ydo&$gD
zFu<BgHpsJD|3<$MbrKiZS+{p`>7IX=c-;^&Hy$>_j~=N?=0v;}{EhKmkqB0y)|ZR;
zxo)#N#TPc#LL62)WivNL{o~9mO>H;~H5_5$BEm7A=D=m4=i=j8@2r=~^|!@%zt%qM
z)g}~EbBeV!E@_gJ;-O9n-<6O3LAqj$lZjd3?qrXjZ~5$5-#CFle*cTdPh)Oo(|1L4
zpf|8DrLtyf-e{PuK^xCJIn5dK6vw}i<A^;bnL4+4brpB+omB3nB0+U$YqPZWI{LyI
zzNBkmKYwbwP_}om#5m9NtTyhQmM1GQkNLfQEkM?Ih-Ig(jHQ7gy)}h@o}OIS?qL|a
z{o*Z}dZ#g`mdi?4_w64qGc;AN!5m_eDIU@foT)kQDA_hJO#R|!sd;aP<8T#TG@Pw)
z`&eALZ?3RjSo*mTLwyx>vOTyi5SjR3(I3N)x8l8}Qt&eed;FL;heVN+LV^LQNtZko
z<kGd(wQ~0WX_g%Q*3mgJHO<_K{1&Fw`-KRVZ>iI^c<h;OVo%d!U-2P?Ao%x<@WwJ}
z$EauN4KppLBs%Y!%oiH8DGnC%vp3@o4A#E>mWym1&Kmesi7*{|tE01j*#y0MZ#4Nh
z{xBeqwhk8)4LM`jW8_&hOL4tn2;{Pzlez!k!DW1Wlc55Az#J`_6R<GIMBVv;N@W2p
z<m<~PCs?xx(F(lQ{s2NgzK@UBUaAhawY4pukJJQ)|28XY2-E|$3^(Oa`DoJK9U5F*
z(edMJ8TgH#*j*A6g&l3b4w!THaWhtb-!hCB*CHPDu^^Lv7F$#&HOe|8chtx`+wrPw
zyVs~}!#wKN*GnN7NoK0otP+T%9nm&Uduy)8MLVvN52xDeGuVg7+#Fg!y{oJ4s#+ly
zT^0+;5W>IJxyG_DOpqCO2g{+Ytqn+qcbZl;5qC993(SW(rGiMm&4!9~w}#Oo)Gz$N
zZ3c$E?5AwlI5=e7FRntZ&rvV*0VRi!l(co~+iT!WCTycvo_#O17*z-A;T&Ww_$_@<
zn|$BADKzdP7d<;LM7YRtr%7$gQIOAC&(HS7M9)v`YR`_$5O@Y`81stazd*RM*HIG>
z!>yYi+4ik&jYm9PI@s$poWKtq)!3ctRuiendZ@X+Idv_?*=Un1rlRzZ<Xok;d+q+>
zC#!1zP-{2gRN^^Xezw3%(kPMS*e%&mvlj%E55Lm>h-IruI&kVoacVWu&h>WJ7rnzs
zvbpzV|BxtHBW$|8#klYXb2?LBB|dHadM6h*jdJZB#X5Qh1{GD+Mt=fY8F_h(Bw!X%
z(EhNovHgQjE$;SLt=9(r18_Dzze8mdBjUyj#5b&%pjYTv7p`8tS|4jw`zPMf1Aszn
zPtW3N*(CkVn<#JSiJbPoAD<l!))8L2rjo2ZJUo1|op?S6H!02Q{0!9@FBIkL>x+H7
zeX`wykf&<97VkB#F5)hHh`!C6Mm3Z1b&hj8wQ9@RsYQb|)aqdzSM<&7_(v`Hy-djj
zpiMeqnz3Lt>fZAe@pf{`qwUdF*Brkpkl|tXLMU<dG$7~u1r%$Dd8clbPH@*uO9B0=
zHAMnv4_8dPJnJ`uBMVCX?mcQ}t3}t5w@n;1E!N4Due1om8^6CEeVIy4vzP*l2ZYEZ
zxKu>8gDHc<dg@+Y)tcb<@88=y>rDITfff^g{P=Oe4m3r_wXr`Q<Fy{ggv-B6&4=#;
z?O2U=18KCI&cdJ&=Bhp}vs+ZqVkH5lZa@9a+qZ5RwjEriKX22^M^Aag$HY8@DTP^B
z`la+&uB*ju@}oaUM|>sF^>{Q<FRqa{qy`>wtuCJ1+~j**=9IIW5H&@8f3bCD>eSg&
zp<=#;KkyuX;(b3s{V^T-l%Mg9;ZHUn6&1DgnQdzY-^VJ=?2ddqeO|tGsZb!m@nR9v
z?)!c9FXIniinG$+9MTZZb>(J@-AcWFvUgl@ca{{T$ra+<r*IbXPG(ot<>;3|VZn=D
zn%bi!=8;m14iHrCXRGB7)l{v#DTId#0^=LtfFqp|-Dk|c@6m*N#ziZKHV}P*Q<}N8
z+wrngulPW}_`n67EwC-9YX&QQhAA$5I=W$IQTuDW%5Q_G12(9c(+3ACu@fhmkE~`+
zy{eL$pL4o&h?21<<n+D#SsF~k$Gfa<@UBEo!e`$gt)b#sjzDRTTaKbU7s^$F!JzDV
zC_&CcCeevIw?7SgSUz`5q%d`uksI-w#j2+DAn}zTI;Wsqy}LeTik9ZD`luw&lP5-g
zXKg$dkQo=qs<1>4fw?RgHvcPy3jG_c<|)_rr$3?qz2$*|LI~fKCd`-s{5pHfU=lMd
z(Rfzl%Jb;cb+-OTa&ohSx?X8kleNG8r1uv-%6uyubOALsxTk!7wVUv`Uwgd#uDBW&
z)O_FJxFHr#XIy;Ul}Y*D0DRo+VfncLeERGPe)ZTY!$$!{*U%eCM~f(uJeDLB$VU$D
zH|zd*ZFo35SA}Cr`B)ey{HbbdhMU)OD@Ss<D;61E?s8}SrF;#9qXg@^CuOHC{SR`4
zZRWEZW~9UG+1Ks74G#sNe3(da*ZAAqt+IKhg)<Y^w7AwLYjijrpTQ?Bk(Jc`;9B<6
z{Dq3e1VbQ9fyWO8GV>!y)V=WT5&a7gZz~i)nTHQQ!_eKEB7KdFtPR*&^+E&8DyMa0
z;Pb~|)S=|JP8xHS3Z)8enAYoz<~V`P&7xiT(QU2bEr>1qY3Vv>An}m_TY8N_VGe4?
zG;s6AKrlylk$Y_N2o;B@Ofk`Xr}D;neItgcbG$b4Zif7<zszoRotX$l2z9*y=QXGi
zGCIV8_-F0flOEqER=81TN3(p@H6C}Rw~GSy@#z~PS~A=`=X%WQa;!rIz3Z%=wNZ?!
z*)a0G_Vs@JQ_ID@bdh20_>2o%#3+C8T3CDLmW(z`svS{mCIEUsH~SDB-P)h2LdN<0
z5{a%m2O_G?1hUbCe3d2_^%IC0P(ZqYr|5*T2~2CUQi{~h`efbrFd;85FGR2fU0)ie
zT`-O;j+o^ejeIqoPe6pdqT*sQkYAz6oJ^jdQUm<$7K`c{s;*WY8N@2v+$AGcS?^Mq
zZNtOq5e#iTTnIeXJTN$q#NIB)!?$zFVd6RR$bzNy`}~)fY#K%V+bZ6z>AYZ*VPyh=
z__iN+$z)qQJ8$2<?JFI5jw7vinzMaPfLdhS3BzNtMimHF0=;Ds75;L=r@@gfTiZOe
zv=5dC%>owa|06>8f3fDMF7xUyzTR&Z^&d`-ewv!cZ|?XHKzfVjo)(#t<p)1Sacu$5
zi2|iqS|Fuc1s*(5LcsigO5grB?C!t5ZNENft2LH+gSgm0hkzh6p}+t7_xlfkKmJ8{
zgQ8_C!fd&UOj!~E^1n-BFJH0l2ZRkynqTfyDlDgFJx7*OC0ctM3rChds;~W$0TBO>
zO-L^i1;e%@hTn00W*gX0i?;ZtS8hY{d`k8(%;!P&yW*h-*0*<>*ixB{&2A?6*kgxS
zrcV8EA|hUu`5jnnGie)Yl9Tu<<7SFf6|HSvp=2pT(yw6k2DW+we9U)Ve6~)f`(^U<
zIInW0r{yXAo5TneJq2B`qQXzNDRG@V58uRE_&Ox|>b}5X31P=oBuJ>9oufX9%(=Vn
zlTqL(#ISo`JigVXUCi{--tQ1pxm~D1(a<K2|2MWN?2YTwvWIVy_R922FY1-!STCmO
zZH6|B*`SOyZ@ReEvX+^LGKVMUpws>A46*K1JD<UiO2!rqlIXtX`T1@^i+yu^I4*-v
zW#1dElbdG&=<K+BYdNpgFU(#tcey+8gE;lfibn6FOG2AL6V>1ir*0)|t~~=AsVR~%
z-}<RocHq{=*i-fd>3606ULjlgWz-u}q1K5b(dg4ltAYgK!|#Tv-rY^UYix!>#fa6s
zVOS?E{jGq(P5n-!l;ZGhZHu*0<g?P{-xsx{`A#DB_GnYLJ(^lVEG{^SuR1EsQNAy;
z;;5*qt)IC>3nabVjeQ)8S6(wH6er)=>#_Jm!_Z}M=k<(2X@`1($=H(zlkj3n(GAl*
zo>)>!Ha>0g{Ye4`n9cfu)dd4bw|C``tIT4wqb8Q-A1ENka4N?MZwhHMPh+=ZJAA9A
zANayf)-t8Bjy{~s%vwc@NlYz^Ra5ot41rj~&GR4oJfj3*b|0tPvo^|UOJnz4&b|?$
zuHVjYwkFXl8Y*nj(XxE}R7;%TmNxwq$p_yivNug+Xx_~8i^W#e+%kAu_XGZaFn(@s
z4z;lmYxNT<G`Oa6HOm%2T&>!#M!+tj)A{(!D50V{Br_D^5oq3?o1K*84GJ?Me%QA7
zS{@rCf8}LVpcl8X+a<q@OHOM9LCM0{eWiwRq)IrJd<jW`YxQD(?yMggYd%g}wr$)+
zC9$tQ6+Ub+D<Gk%?N|xg65XH6NOU%K9!&Xkn$n~0r)#ZZKAl^Gradvy^xV?DBt7;H
zp2TaMC7^=HcwSxy{g7F+)Lb80MGz^+g$8!#^_u-?VHZwhxPWg=L-(0w^E$+_TkOW1
zvhq5n0CqumtCbcdeDpgdH6w!yC~n(jx3j}V&{H=>$@wg`7*4NeTJ4on4JETC!?XEg
zrbA;mJIOJ|N?AB=O~2Qo@?K$JVV|*cDEQT@VPfCqA$r^t=da!D!Vy_O;kO^Z`rze<
z!y=K&DId}8;{K=RwwEg%liYese3TED4eFJJWnNS1Lywo%efa?SuM-h*7&Kl2uz>ag
zd~hg)1p*P$2*5K6TwH}N+r~O>t2`i5J^)D_j9jdMrs)_N{Xyd^22I@O_3Mt@|AnMI
z5G|(eZa#%6tX#QF@A|yK^?BuwCDfYE%sIEZTG7Xbv05iiLiV}k^e*g0ccUVxZ0s6$
zFC;N8qt<dxc_xSmbm!ZI7J%jSooQ3Vl#gC*N;FLvONn6l^8Wj5-TN{xvR5Qp(t68E
z&#jIhn`wGHhUWnmiF@Z~r-+yD^+h~`srpx1PU8;1s{o2^JJ_79INckqIShRF?h@SQ
z+BNXYAVMer0&axgiqf)u_-1!_)p5nqEq7Ns^%iGLZgeLxiUrMvf$*;SNnnN>uG6L?
zR%B9hdGQt0MfyPW^G0UW#`T5A(S_Okih9Cmcm2@c7aNMMtvr};WqEb^0F%V);ZMeR
znO}3%Iu9OK*`WBzcLji+agvXm>B;eFFheWwE1~(i`~a1j?-KU?tHn^V(6o(_(#KkK
zRCv>?yIaXKi^e$g#a&g^_uZx89kLMPfP^~(!^<rA`W7Od>P+`aQ_qW?AL65|sy8nJ
z2xoKXeh9iL;@B%xhbdP&zEpcoOaJ=3Qgj80nKjMcm%t4ibKf?D7IPd71+CB67~ECZ
z;dd`7iqjKIlhOy`HEg`ETE^M9uKUD<Q?rntwx&mM-hchFx6nHOn@r8+EH|1(4Cxa(
z3~hUgXwPF@9CJ4uG<X#HOKLy8oL^OYOir2G3Vge9e(&D)r2b`=GeR}%73nSfT#WiT
zlYN~o%Pr<B2j*d`mtMNztetQRA7-jwFsON2yhyF_MoE2BlHAtvrM37C@3!a3PU{m0
zmTe1X2nXB(E+x<kOaOfZQHfNJ*=5Mbd<V7u5a3r!lKAUrnBWjWq+uX1j>5LIyP@|>
znbWp@gSE;>2XCJSG3WdcX8FlCJTr|nW{L!F{s^^+75F6Qc$lzEu3Ty$pTK6UHBgk0
z<$QL!iDRxx(=~Gm{X4V{R1>Y0<S>>E{grw`(zh8mU!R~ew;mbs&aWzAJ$81-!Ib8k
zA&CR-o5nAr;9!4$3cIMxLp8AG!_jk2g*;cwn`-$wT_5P=NcpXA0t1_)Q}YSHVGO8S
zb{kXTYc=~iI?mIWU=fgqDcWo>kM8c>U_?IBZ@B2X^ZOfO&jL%X%jxbQVpAP1x1k01
zUAF2es9rE-7MS$j_CcpC8Pf#%?nxJxXcJp8TlTuKYKIZNmGN@-YaWe-ww}SDpdjqQ
zO1WL0@~}$Kyb+MM^QW&3AQcBk7{YOa=M{BwdK$xNgbncZHh6-8S+N7(KzL$e4~R0I
z2wejB4uwd&y6&5p;5=A=R_pE(V5)&obh<yMd@4ckDA4=nV;edN(-JZ<8BraBswbP~
z0hNtsfERu)%rEQPs>)yP^GOCc1^x$CXgtfrO23(RLX%AfYMsn9G`|c>1j<*YyLQKe
zUIv5&XRp)C@GyR1ZW+N7Tjr$)JZ9Vd<MGG;L{I-0;`-mn?f?3=eQbg4*X1Fy;$h{<
zE`Byjxkq%|++S;BZ_&^&G&VIBIQ}a8H?w*AQ+NgK6<pf(cP`q>-j_3DfBD<E0*m(O
zzlbbd_x}W%{>wFvXRtjD7V!5UKQ@9C`WqgO2Vcg-jy!J46Z@M4XuNT1T*H`_d#MXs
zb@43^JD4%*&3#0>XnvDY%NI1rhGhHeUNI@Om&Rr-vuPG2Y>3-2B^FP)iltT@U>+5l
zD_#7oVE2WElVa+(z{^G3-@3HhMMRytJQwoD6_v}y3@drlE$nZ8yTW|TwRJF<{+!r6
z`(#}yuoYz-@*=cL`DEO+6N_RfF~MoT1Dp>}Utd~ERo`!9zfb9zI#<qU>vn0othZ3<
zQWq_HbcF#QrX_yENYl?-q8Oq2??27bbWTe3q)#_u3Po2LixItXAZf<1S(UVj?xMHV
zmZa=pYMeGnH+qi3S;~o`!F&37v|)kv8Qsqd7Q~-?57DiK2PF-n!((ViX1{8*wzAFL
zlcMUPz2N+Nix^|*?#Xh?9M^zhakPv5Jvpm;V8gt2y`Sgi@|D!>+0XKvTY)Ij{yXmy
z9yMGpIu5%&tth2)>&DCjlzPD~qu7evqR$tU<x<?eK5H)xX{@687mhJTns@2*UQ)as
zHlG<CZ#gt}!%nzHkkn<66@C`LZJ$~3X6*EjBHhomn%e>N`e)A<NQw>y4aaP44CB;r
z@@VkOs_S0m_|RCc4*14a-DY2yX+BwuqAedwz+OHyTu;C4cUWvLjeV*iF<=}eF(<5?
zp?gh)3ZJd^fab2o*LuscKClHQc<c=;AA&?P+Y-va#B@2M?;)@l(Hw?S0m3lerGV;y
zfq`+AoLm+-n_*}VC~&FG6m_uDLjQ(ZfJhDy6If0SH8E_;Q?<H>Rlv4dmPuOHk|e3>
zy}!HA(3Em{G<5#4^9*{b<eXKD;vkDK$}1Mr`(7;dwIkle&m9`<St_&_R^P2kwiBA~
z8$MosfjT?wAlM2Nj#fOp@|;vEzzcoV^J_@nK|-&{wT6T7H-+v;hMd{^X}=U2B<HGC
zGZGiV=^xF2A~yx~V&Q`v>n#C+7yy2e?Teevpe#)Ru?bt+2T3v@9#(KKN`+89R8k@U
z_v9j&6c-`TP(N?G38SkXC((Jy4KO)@+hB(J`~V9w1i<5feZ9k61!^oNCg#K-RL_#W
z!#kX4+T?F9y#KUpFu19u5x=D$HfxKnhpJ%ZyQRQ&tGnp<(p56OZPE3s)SZRqB>HD+
z8j@IqB<731BHS1mvz+b4=iIle91ACuDtbKcB&cZ@GDn!Py=z%8S%{An4F9l9!!$PO
z)}2{rmVslZ;j&Rgyds>?U^clvZ5#YEKn5?XIEj^+IS?!a>~;(C0I$&RZ7+0H%zcm>
z0#PfV8e~YQ|38lfnV15=0k{dRo&%_CaGE1t<E89;@cJyE!2H%6>jgqI!hb_UME?Oc
zQQOg7U?`%#WZx>#kk@ij6241=-g<{|=JhU?4aIzZB-ayzy`BEvH)%r3>b1-f^q&<h
zS0{Y+-bV2DzA_i?XTP8k=2!ocm!n#n1r=CIy{(h8(-Fo(b^MaKu8#jzy8fB((CLBe
zD96P_(q87RkLOi_pRMdpa*fZdHXX5<ZS=v!_oV;X<1(WTVxU}qPu6**ZPkIVxq5d%
z6JcShtA({#qxRc}z|GMN_8@5B<umV?N!j&Kh|PnQ^@=xiLF-YAvQ4m>vFp{N^6Ji9
ztu~x0ABNK-F7ox|&Q5ti%I=QwA~>&jarhh*Zi+UF%Z3E*58T-BjD9{x;<}AF#PYmq
zEE<(j;;g|#n_rMagVxG&V4<<EC*>L7g?68rbRKvFJaZz+@;baw{f*7$2XrC16H#gg
zOFLV{VG~Jt$;|4p(g(99yxU9udnk2<Ln+?n%E0LJZT*P@Ppe0c4<x#29ZJ)?|JbVu
zu7jnw6&7MEKusj=f;4MDBn|VnF58`8X@E!A3w%5fY%Q21TJ`&rUikHp?D+<<jqhLx
zZQ`4Be*iilLfEzEbo`TNRMJhpU_LfL8+LFN%T2}ZaB-<-!uyJzv~0+}%jnQel!g?h
zo`?<Rc-w67eaf-%Dki6WOIfJps45k^%{bE4K(R2#EnUXiEJX=VJCC={D7E!RhLfhR
zMkH<OPUHQt3pmovHc7sXXI88zbYV_?NVQ%V8wwNPb7bM5GvkQ0*0{@`OKk30cY4`T
z<8I}PhZ5bg@xlEi=EH;;eQkhi%2!KA?{rq#&da*)E;Ye^leQWGuaeySa)0JE*n$S&
z-GGQhIK10LDiK$d^*2ls<P%)p`0QY&02g-`&!Cn{aeS6OpZ&|Z8yv-lF=XQc-`vbk
zU*Qf=cTQ!@$&a*Pa+N!9Yg4=>T+=<#FXZEi%4NuNFZuLH;^Ih64;tro;DNrbpa@|@
zd9~#2RfE@;g_H(+S1#~UvmR$X3)@!8PuHKYkY9-BmBrdPd1BGZw)H-8ndVkwKv$2v
z&ga<74oBy0jJP7Ack_+w(~s73*_67I)RctXT{gF_B6!6lyW2<bZNoQ&8hd~*;@f4M
z0^pQV$T3LQ^DwRdqdAmX7cde$E4RJ{(TD2oZXqN#1GyU%gb%s^t9P$^?hDP#&JLGa
zP=GCRF;dg=78_eANbj~_v<`z#4!VPSu`wRvpwV_H897mSc0f$B!f3NH$2evpMrSA7
zBKh_}@B696T=}~p_uH{nrPSF73;0Ge^uv?K5{pu|Brw$5+Sv>mzEMW5$z)o0CA1zI
z&jlz|-jyyo;HcQq=zHR(T)`;n+x%I<7H8bOjxD%saK&UWKuM4w?C5dE)4H>HHm`3z
zJ)wZO?SB8b1@-~NQZGAS1@8xuvH8ZV(ZQMptl(yg*jZS9?GZK#h{00J@%FI_I{|Q3
z*x1;d{899pflZtG<%>TwCA*HhW!7`81Q0EOtdhj*D$-<@zNmFUL#j};4o51J;qKY_
zh2ikXmTNd=5wd*V5mhSBDui~;KQYa0ubrPnVTXzpJ0FvEsF$yj;*{MGS{A0Ei7?{C
z#p{i`^AAy;>TpnF1&wH(kB(<HhG4vWCoLaIhSAu5FJ&_1FJJD|)ybNfTQrlBlB&tm
z-8ZU~4*tJ@g-HtQj5e$;>)$BYWA7pe&u84yaP%VwqFfUQ++w9q|0`W5<xz@RY31L@
z2b?z-_ZfWi@bSNm|FM4L2U9W%7}^0lIfSZfY3unFo>;8BeZ4=Po|zfP8>|e2|M-@o
zni?_Kzu@mBKIgp788x-(%y0i2(f6X|(~m4e;Jqob<G#+xVS(kw5nFPU&HFFhh#m2N
zgF65Ftl~dO#;`lkzwlw92-qz+_aV_cd3u#oo~(CSKPQ|fT;4*H0*L|AgP2L+_}f_?
z%-<s)y4T;_Q<B38=%gL+Y_`@woPZs$2cXt22DxNwN<Tp|e&C@WczJ_HnFEyS4E{s$
zX9GwmB-37?M`WK8VhQ;*egm%nIB-Jf49UoW6aE`)J{{&=cgPMP!N!PgG0?X-y+q-s
zSR)2!H#dL)|M2mnZ^QIdav<vr)bB6f1_J%LDLo;>8RqvpGb4e9i78V;qsnm=l*HR?
zZ1~x~G^P)czX-9e8-PIao74vKAV39?L;zkF0i125d%?uUqtO$k>)8eY35~S+Zp*T5
zV6`G;a4uhFidaX+_BSLtQvhBvVC-{;5eXTYz%vVBpl0l4Bx;y-z4hrJh!gCw*=MIm
zkQ)%;n{cLetg0c_prcs?3k8i4=|7l!o0t~LE7CX0($zgZMR{X$HPG%kP>Q%-uNPW%
z{m^ABG+R+Jt|lfXW?6f<fW3``8e!_&$`CsZ06p^p(q?<htaK5D7m^a8O$M0-SH$W@
z&aXWiPwF%%R2}jp5IBEv^duF#T|>o5FuYAdLE#FpS-T2&36%+dKujKUUCwQ6Y}6|o
zt$;k2%W`g+?8ib-r6IdY?Rju3RAi47Qjy;$CljG<qmY>xehJvFZ}2amE*9C&w<}v5
z!p-<IQo;<O4QP_vATT`1N(6(>Jj4jGZr-_r9o{x~<DrnpBW+2GyDThoZ48O2aNGGQ
zufVuh4zLN4u^$UZ0M5vFoh%CvZMUu(pV&`fN%T4sc=P7X$!4h6PI`<yJtQZ^tf%VH
z)N(bR=pP#8dU3_bBauZUFA1h27cWnUbTB%28ciOn3|8f8R}n*C6e^tDye1TQSlFQW
z)ZH64Tf%LaiZMxm-G~NdkhN-s(-W4N>~3x~rx>W@d6i3dU=J3Ku5w#V#DTkj9O9_P
zUJAR@Z>fKRht9{x=hBrc=|HA|qBI+{7$CFzB2^(37u-Xssi_}j0>PZzU$I`jeWh&N
zkz%~k0b-MDs3ZjRQ3}}lLOXSS{TeK3h=kc3O08&YOb>0=E1(F;Xqod;AnX>vT_C{V
zu~YcrlHHGD)pmw(8%Jyx0!R>#7kEy8fNv4ydGv<?B9Jnws@EZGj3Fs0i9{>mk;WC`
zgNF*yzZZ&oL}w914Q5m=`(8<+1Qj*SMA+R4US}@r$6+xsZ7@kZtMe3js%3)-8=V2v
zVuRjfNu(V{xO$=4+Sy&ca)q9c4~btw_@znr{OBpDW;HIn=0o@H-9sVaL7TO<eG_nZ
zY1y{XMdWR#BGDp5vlBbp48_32+$^g*4F(TbB-}NqsA#I^HFXCwfrx95+GB^nusMjy
z27IS3U}4~R@gvE$ClMiHg|~5$P#*FL)y4#v!E&v<JNw(IP@%4Z9m4nWHE?(!vYr0c
zTpN^OPVk)B_GrR-f{a%(h~+1{`CgcM(B=`tE@VIRM^*{Kut{lVcpz7W&bX~l>R0Ht
zZ;8$hfM!)>5mqdp!ub>IOPAOI*Q;cmY|ge&Ky#Fmm6Z+HfwI&K93%(a5WEKVAkf~(
zy?b+>aanzWmuZ>h&?q%G#U<x%g@E0}&e)h1EfZ540*|Xrq1y0KUU^&Sw60?{0-FRc
zut_<q#z<f|A2iJfnRN&%Av<|}d28e)sbBBs2UxR6Cb1wv3?|Wtmpu=`3+TTIR%8m8
zG`K_$rqB&x?OoIZo^6zog=J@<YAGu_2=aXqf?^B|EvO)ADwovnuI)z^7m$!Fl+Ni#
zqhC$K(Ma>Fx8CvP-dXp6Ty(%WGYoK?IcnKqpFS-;VrFDy2fs})5ca=f9k<SxAdiB)
zPcRRX^Tu|S!=I2nnLGg3ggJ_Ci80&CD8iIfDl;)tpR_sNX~Oj2_?Y7G@NnWQhhZ}*
zq&6o8_s5+QA=hROOm&2yuEfi?z@M~lz&0;>(VI*dT5Lpm@f;i(q(D*m>A(N*Va5Y)
zA9(w52l91wXX4;>rtAf4^7$;sT7lR?5-a?kC*}JmadDL3Lq`07U}u0hE+TP4QUT0?
z0sALN3Y`C^9j}5F^Y#1|gvuJK^>7RR_1D+|`EPcZ#gODCi0BQlH9CNxMAE6%hme(T
z*UERaee_&MN2g?onqAM^X^seKFEsr8(cq~?i`#YcYQW=rtghY)Wnt;jF^mh~Q(N?-
zuw>k?hLHEHZAJNFf}5;-Zi48M=97#w`QoA?M5G4G>fFLY5hMz2dvNgZ=olC>q&%G@
zQf+~~`uG{9eTZ)-HO`uY(Jcs4N{9<IIXPKY81MtuiKET;hZ(xB+&KBkxs2sAIVaq=
zAHjGAv8yTQxpVXLy)Zf|ufF(|_3PPw8cRkQ(Yt&a1-LFB99G0ZgG`vzUMGdk%bruO
zu-Id}Sahm~Vj<7P#K_1P!6zm*=~gYMfY+}8o+H}3ck!ljCXXSY1NnV3SgVVVV)?A7
zAfT=VRdPO|4A!u>@~)JmBmpt8W3y|=Hq;|)7zVCFmJd{GFpy6Ks&aC2wu`KXdY#ac
zXj!8@@WX{-`wA9TNoUDh8fCk$=?bYK&cakEq_Zh6rzr?}jF0?z)zZe9E+2Cr37)Xf
zva=U->ZO-P5Z%^x{gZ|Su)skdmtS}G2_~*^=+H5cthIq2`0K?ul$6x;bk{zBzUr0s
zOc0GDt&mkvxOh`kv{u6*!l?<?=&w8ZEC>=t<Q)hiB1H)Vs#rk9zlMI5&ua(S9yQ@Z
z_G)K!Gzu6N$YC-eF@G=)K*Sr#%TJt#8U=)e7{hSLqE-GLVDvlW6O!o&SpMj83yzQw
zf$m<Cl!M;|^&R>IG_qe*he86YBw#+#Cxd&#AK%THAyQ^Bs+L=s%`7Y&XEWQ(VL8SR
zN_+)m7F{4wTP?2rXsl#<?yH`v>PNm@s1VhM3$c*XegmMwwGXl+5R0aGIa5<N?MH!R
zh7Skp&LrjKae*E&8GFjc!9l=C)nze|6T^23>tjNxF<c>JEG;7=YN%*o?#1Lo#6JMT
zuKUrS3@x~|6FZ<kJk6r!jSc?%;4Na4f^Z~qga8sjKj_dsQ#M$CK#)lW$;(VkG&D4j
zzM^D`)mgOMR!fl7_6%4iSRJ|{o!qT%39SlABeT`)5P}<LK2pC>ofK?r0P;+L>zJ5D
z6u0T`Z=S3?+Pxv>S>3n{4ttHSFO&>!bAV4PssH#%B>0z#ii%{@AimZ|>h}#0M~x|T
zmWG!lTL7lUV~t5KLImiq$pHduS>u@>KRK^Mb$bADmKa#7P$FqrS@R?wKZ!)r-X$84
zvwbQXL=G9z8r6w#-77$(uRRi~!0Y+r@#;g$d)BDzMmLRwKCeCF;$a~|ZED&!nnm>B
z{(ycTIAQ~degJ+&%*!w$A_qE<=s3SZd>1utPL>dv1s9Us%a?bM2DLfUM3;`=VZa-U
zM1B9-HFTVB!Fox7RffPf_%%Yv`6Bs4RD%1~L&Vk%@wx-(&oV%ZL2<^0SJ%(YF=Lhk
zP64sLYFFCdf@M--HV98T(Hcss+cub-FulRv5(PdmxWVl=A_UzJ*paavIZ9yE=PoQy
zpeH^+%#zD|h}~+!*B4{0?%WgET>ARApp+SbwHEOe?oYVotMRSn2JD|f5z~V%3a|`7
ziC1%QTuO4XGl&;1fN?7Tn^kPJGb7WauE*|^%vEOubwKw(f`*_yG#V6c3}{+GX5<OY
z)zx3wB#=o(JnC?fBmlyMAkm+hxiD&372Myi%360~3mL^Bpf!<qjHHoDx`O=s-@8_=
zsimR`R-`pRSfgE7XZY?T7*by%i`1|04@*gWV;HS;5YiK+;HKhb|9XiW?*LX&KBQn;
zC4`Uqx&ux!Dm8@l(hO!8#8Ka1Sw;cfs2LoV!E|zwA6sD&o%a0xnHC5sUZiJ|!D_nW
zBox^9#6d2+XygdaR3Qb_#s$3)FQd=&tL6BJ^WLxtUR1|?7kCq&BjHRb*qZtfuQu1z
z(D-Bn51RtU2X@e6@T$fwwIFBP3_tWVxifssvJ5#vfefwy>BOwn8~10yb0}gssknNc
ziJDdPI#S83&ajd*AdT=^?+_eh$2Ej?Jh5tD3$mwjOUH<3<EkSC$nB{FBqS!_g-n6r
zs~Jk6mhlZZe1WUdY26Z_VhRv!&A=(Pv)IkIM}Xr=1Q)Kov89CqUu5TfA!J(7*&|uC
zm%5yeBIy&|w*nCVAg!R_h-M4YyTNoCwG`;zj~lMvKeYtT74WW*X#ACYX3+l|h2{VC
zZRKpwgJ~>tUd>F+$MI%xZzW-g*z;i!m=?fK$>h53^h$&F{&9aI)d3OSVf5vso`hIY
z=E<bT{teJ_FQ1*A!9%qLEk3dmZxrem7+$O4NEe4+cm!d0dM&<7qz*s?Jq)_-Hr&Wx
za;LC{*<eZE4RbsSo&&5PgNVtHC+JImGz&o_Bnl1(NW?>}ZjbGmwq}PTNsv%CWPe|`
zJ<ke<Q%h7Mkwi65u}M!l+{PU^T101Q9fGhhC~CiPIfGLou0fI=nk?Y{i%Ia>W1h#W
z?l3AsR{Te-6P&414PX{AmO(G9RvTrdQGSBt)HS%g&fH<OZz-o}AOjH~pM_DYIwVs7
z+YHutGkm2zYk_#SzI^%8X$Tb@Nt6Mg9j5q!HfGy#7vti^Ef_i@S|?%hfGE--@WW=+
z@o)+SI6(aS-@V4dFS42xg+Pne!Y1q_YoMiUnuwmV!*MfU4JC!W1%E$~4`M;L=;+dv
zL*XBl9fBzMLZHWiPa{g(kqideyEZ3)Xy79J1KDcJB`Sd_gT7-vRPb3fCSJ(N7}nG;
z7oa#^MgW9MRWl#2B=Uw~hcgt??!ogNQAl1R$@yh5bTec`firI|djs-_a69`0t4_Ag
z^VEALX$3~Z`F7^~+nk&qx?=gZ;V_+E_`TZWH6cNVB@85qhXSyj)*cubXxQ|QDD7t~
z+@im=gEzpxpALGfGBX^#a^;@7x_V3t6v$<~-zdOIfP9Ex&fiLJYHmiLrR=XkfY9sP
zIyJ5cb-TVidAjC=m<o{jNB$Fc0HjG#@1vryyvHrso+LC(Z%~OnfumL67!p8fGwbWR
znWgXtF_Ax*joZM;RRhZ+?I+wi89BM_T1S3L1g=Sqz>tK4gc{-2y&GL^HYfzRiq<~n
z#)?3y6IF)elx(<LpOqjOTmrN1g(Qjhj&}h18y{tgCylNy!0L|3ViOjojO|j>PXMe3
z_z`mDx8OJ$U;u2%_f)~=0ed(~R#8r_VREvr(_2v@RTN3=YxDv-Ls&}fRnEQjN!?r?
zhxy?WGZGlWp`)T;qI;W|NDWWc2v+BWJDf>_xMF35OyRg4C==;h@o<6zEppz58j_fS
zza3RM^Yu<vRr(rS6>=02e=V4b&%j-d9Ie#3FoSdc;QoDt=|`L-UWXmJ5Rtnu1;yhQ
z_z;1usZ|w0dLZncU#lPdWpe+%Yh!TkyF>9L^;62_dGSkI#R>2;+@(LzB0yJ1qK05o
zy#Zr5=;?0=U69kFfRUC2tbm+0gdA3YSXyB6gB)eK!~_6nm{r<AXU`g=CkPu!2TyA1
z077FRC%<&>J5WG60MMO$EN?I}dLqY~5YS3NtaC(JB?DQ>(4(P<QaMkdAw2_HT@Q#g
z%2{LJo{5IT7k*n3()_#`6ST$p=o&#L8uS(hQPH!w$<bVJD?!-+|4*nj)T&=_0G5Po
zuMF(LEI|ArFx|s+H*sXy{tPjtg9cQ!R+){Q#idLSPy*P!1Ju9%3(E>V{%t&jV0Z@1
zF_885htpfeM<Lvk>PPA)4um0O947GU(~OxQ_zu|Y1)MYG!S?{<cZSu;x<qgifIh9L
z{MhMJc-iF<$lAhTL%U<-P`oGK|HHyv^G`BzVeX@uwRHjFnG7ROPkU4%9X|5^xrl$s
z<$%nA4kX8gQwN}hd+?=vE6~VQXNi=ZfATlpzT3@?dxct5-qqFBgD>qHuz_^Eyn8_;
zF(BgqJ716XGlZI!<*&-FjD(le&rU}KbGM)0Be`)SxUQ})??W2@kFKt+d8o7~IH>1I
zv@UoviG{g7-Sl!p-XOX6KuV(iw|El79=U|1v~v57pq26^1^WAc>8qp=%;>vZ54K;x
z<9XJ1)c#sImGc$yK5y^u=PyQb%QH+yapm%;{=H^ApLDGbwc&lqe@3r<d_qojGcAC8
zxwBS<gL2gHLTHm%s=`thB&^`wr;6Jy8e6YyXV{pe^*{80=wGtm>3kgC3UX!{4GrFh
zmWV(hX3X%e06NBGHER0UYffsiB~q9OG6Hk*yN2l`IM4wm27IE0^Mlz?)A33k*q1QE
zZo@RHwtx4|KSE&KFN~|ijHMs1imv>s+HwT?1LCr;kP~Tkmk0Zxf?dQTpBvX)vqyR(
z5E=$22b)lmUrjf><#Io;_5@&7>=O9TV+el1+{=yNQP|4TP#R-+%!r|vBFdf7LL%^+
z&?$^(8UtYS`$4_>1ZodrU@Nwiix}u9^ZN!H8Sth1eGLtA3SRRhsQyp#AXZ<97~Q<i
zPVOW!D?dZaJtp{(t{c-5((Q$fix08u3mUq9Je&l@BB+#BNu`1Rr?s<ys%rh({YF3(
z5EM`lDN#zLQ$SDzLAtv^x+DZ74OBpB1Vlh36hTtDyFuyBO-RG0^UkH`U-!G`e%~GA
zo^crGaP(~VTCDxPbI#{^ey<Yn392JYzfI!HnjPqDRt+RwfpBa5Qz`^a&_0t)VW-jN
zDut#Z=RJ71vSQjkV=u6^+ZAmiU(Hr_-lJLq^6!Q@f)3`?4CXD$T;<1SYa%rsel#Pv
zTy~GN-|nn;l~SQoXVXCO>y2r9BeI!3@6YJc8OI{c0y}CbG}cpM+RPwOSou{gty$Q-
z=c^jr&yV7Rpi)|0W&EnFSYTLn0%`kUCr5xc(9@ou8H1TO6!ytaD?rdSq_~$yc|%zZ
z-tq8FO9oPfgJXgvF9NtwLf3N=S`Aq{FOifvFMa5Ddo!R5Bn>-06C$qpnS}jfqAD&f
zfPs$i_`<C<0{<(ALq)zv!GZ6|`n?@u403L?8$z7@r4stn>W;{bJE592mq-C6V3SSm
zu#tE_nLtPatwwyV-;VA@gPhXfoddC{Q@<YF+KPTl<?VA~?^eWML|E9<p(RE}#uV|#
z(CR@2!HRhgW2LB=W0-uHkHlZYb%C4!l6Hb!FHSA_^5xkXH3mn!YXYTl>6;jt<XGw|
z%@ftA#(|uO;ic=@_xKpnJ2YX?Yj%8=ndu9Lz#q8`)PN`(oqZp1zcD-A_gT6APw$L_
z?d7UIvgY+Ng*y2gSO`yO4BrKO!WuzPJ-fi5M=OMYi_x^Ewg<#7t`mpXNw>~LJ>)4q
zdpvWCL+Au#g`>1GSW_6DES)?NQtZa4wJ<Yf^D{hpA)zL;|J*SF+vV-P^Ddo|p)WoF
z)en)Bp)9L`(K2WV2&D)%YKJC<D8vXME{+4>C0@A5jrj102%OtsB!%AvIz)#~FWiH&
zUFaGSObWE92>c5oVO?O2NBSOn8X%roS#SY{0MRxKz#E?codgNID$s|ZS1<vpOQbh<
zE&ssO(AmurNq>rUpXVE;rz=7<QNp`gWDj7F{h{r_bG=7Qmq+X?R|}hla%uV|mZslY
z23k&d={q=ZqcUoSIEU*B9I_<}m@&ekm>{t1uF&At+-<H%zzzOfYXp3T7*S7joEh)1
z-)J)Ni-tE752JLx=eqN+V1~c7Kqz|cR($o-$*H#DNcVUtIsezMp{;U2OEqEng1Ub8
z(j}dlo7vgfYRv}cBZ=nCTBEB}6mCgCEEuW;M7M;o7{qrfk!fy@>1ngv7{C~v?C>`j
z43FAi(=UxiqfJGir{-#j=P){PT_Xbx9z1?9CK^Mnm2OZ`E=>)%9a9LB!E7YkxDH=E
zLl=QD!NO5Lea*)Uv0*@g%0M$IL>P>xaEEVs*!e~f-C{QR+1~AZ(kF3u_5zB?QmCsT
z8cs?1a$u&Gx%D&it(brmTltZiMwJk!96n*(x~S^dj-R)J#I3X~5;*%W<xTJOaYJqz
zvk^8RN_MR7<vV$9BcQi?7x$2Fa<*Mgbgi{AJW9z^_-gUSeR|$Tv)rFh2?L{HQV!!L
zd_ybeIUpt8y34~)PzH?$Qi4*^ZrZ2Lu1uIX!F51;T>2B8*Qc*T-vRR{2^f^aI4cGE
z1*mbb&YnA`WQX%EGO`Y20(UUZ{TizfML{3H2!kVOYGj%Krc9VOb$JAThd~c1C1o=#
z9^eWzjG4W4e1Jj7i)vqG+RyCbdDFYjTD`8GB*xWob`tYArU|bj2AqSb?GEvsxQu$r
z3ku6YE5Fm?vNB@D!B@XbsTE>!9T=j@7(eM2+kbo&#kzGu>~Rp|{%mH`?t`_9n0>!`
zQl}aM2;YZR95V&6N2|BJ+w_>TmiIQYd*y=c6)+)n05ne*NC7Dd3GyKCKwpeK4l4CW
zm^cW-8$=BQk_;JXfpX;7qo%P2B1R5CE};0?%w4Y08c{Nac2hQ9IHo5}BOFGs)E`0T
zriIi^n(9TUsXsxy#%{#>5+m{H(;vYKp=Q+rRS=Yjc%WP|bYd_V8QIebc3>A^$Pfc-
zViI%!YF~9I8MpmYd%|@kT_;%@>uk0gCJC!3BcI>eI)Pf>w#9M1jQf&rilw!|#>Ftu
zO5ILzw!LjtgntNomCO{Iv6YVr6|WK$_&(D{={|UQVJAs$!mUfRXH26;9jixmqL0f>
zk6RZw@g1}Htz<S_I*iwd8C>_JgW7y36Tf_@UJuOi9>Qy#qzw>aI?rMB=p;F92eU^4
z$4%F0+at}V^sujwi*!Z);H@Mf^B+*pJDj&GoHz;8hkA5V_)Qq4KqF;Z^%QqB@G@-L
zF9!~WYAt|p=>`voI|zKp2wWD1XfdKb-pNxb0q@lm`C)1XMr?P89li>NUTf_K@XPQ&
z3#GM~azz#!=VxX8*cXO68DX{pgCSCI8~w^iYw?i~$bvJJ*Ur4~X&FV+?hFW>H+f?<
zGosrW7aKG%zdEx?5crvSisNnNDOvBjV_JDT8RJJd=N>mvpmKb)USD^=o{=atUOgH5
zz#}oAi7mpyzsVSvS%i%q75y#wxc=P@?>gejfq7;6mq!<Kd*o*Z74GZtgm*=WYM!{_
zUuJm?W!W-Hx7}yoG?e!utGobfd~C0ilw5jawKu9DZBp+rc_Me`<yqcE_A^JLneZU-
z!<TDJ17t|muPwpwdltSj9dDT{FMU{si6SHm!42m2OvEdo;)&N^ga$s;@bWr&F`soW
zAuukX>9v3iKMKS$B%$|v{^$vFQ5ZmY7(iMrZg_aiyAdhWp&SyIltfZ{BX<784;ft;
z#;dX1*10$W@UwBw+gDAoIEI#BJcMxv+;rWeZRMW|>c8>HU5>`C4yd^8HcUn(kX=$R
z@N~@IN#W#EiC<0VmE+TrQ(HLtJQU`+8L$;%IV3M8k1L-4Y%AfZxTQeQbYyAd|NRd*
zDcv<eP{t5PsZSWNfc4CdM#0u{!>kpy)sH<VoOLDhO>#7|?veT2i(bM)6#^7ui(Tea
zX_n{&_A=uaj6=Fnos26y=k(isVF~vg?Q(WrzR%q@{$|}nG!IQAGy~%x*fPOh4L$N2
zbQBSdq)-`m{X!l($V-ho2uRzSlK@8sgz0F3hxCADf_s61TGE6sw#<48g^bY5+TY)Y
zRqtWR2r``pp5n?HRK!TE3wqL-vJVK}6PfV}1HTMjRJ9<@F{pXh1vluPZ8naC$>~$I
z;j!`)Ylmo?{RCGlKHFW<(}m;8>7jC2IQPa`S`+CW5Z@enoA6M{b*8w(!JM`M>Vvnw
ztE92S_KfB&{1)~%LYq0Y4me^0KBvff^8tb9t$=rm4uA5&wtGM)(b*q%biG+_!$rT)
zc6#!I>pq{x?%o^&l*eDF2qd5P-MTLb)LcXzY%8%n?F%id&;HPu8(=h%B!<`C28$Dg
zBgEB)f#``)iGUf10fJw)&#^aE&{#@CihNvi;lrsp7}h|q1e(}0=n&n5_xJa6J$D^x
zymyQsV2yNAAY&tXAZ3w|!t{!k<WaoK`59OSLW(jL({dEn(J4+FxlGcu9KvWDy?gDw
z4|?pA8r^3qMejwr=|wzz4#0~?^_b?i*@xlf#zeoa&vYviNVMxQCSr{aBIqEHMLA*O
z0_~*1XBTrbGZ=4H2f5N-ZZxo7636)$a^h?%_cMe2$ZsA*y%sS{q6XYtT+(Oq(X+F;
zph5!BP?~T~D!aK{^zRT998mv1wu^x$sIlnd9UK=FGH8ctA7h6ufNh+6R7V{e81Vog
zRAfNo%+T9_w4PEus(XA!NWa(V!22o`>3%|U$R5zJJ|s^VG+iUEV&54qGymNkp0f-I
z;0n(1spn;~!_wHlfi=NXzLM+2m({hA2Cu{;1)Q<ro2ft$i;k<x+vC_BaF|U}cUN`T
zU7seT7L3^0af6x?#{HsGH(`lP1iJwMYGJ}(1;&uC+b84hT>*r|LBEG)Yk<21PCB0u
z+>bo4WB+axfpWnRolex6AF#sam@n+Bq`K+RH;YDtmmbfqY*QV6m=Eg@AmBLz^5`VQ
zpAo{U2^iXj9}U?;(T8ZqR^Jc!{{m$cllsAXR<&phxlDMGE2(qW943;0w1PW>iy%Z0
zh)n6hdW$)Aw<R+anBqx6>DqY#pc+I-T>{ktEV+UUIp=@3@7<wLZIeQFoSv1D@del+
zKrU?mYV#kxyc9K{GpAwD%+LwoW<(8>|BOj*R{!0FDoX-P5AohfC&8@v6;v(pE(67T
zYVgNW@+5N5Do}STQG;dzJy*^CV;IvQQ%>+%0{ycZFpmOmt8^f$^PymjMGm}U^v9Pp
zlQ43hhGF4EVz3DKVaf#-M1FgQLQghL(mrLNqjHtE%G(M_bM(TR&^);N)wvyN!y>@&
zmS!@^Mx!c`)rH%f2+Bt=cXoIPgJBhH%1`lV+<c)K3JD28()8}VLoujm!WxeDVlvt=
z7%KoO30CjH&oC%RdZD_p`wqeeAQj)2mad*~h!9Bes8ByRdM^zaY<S)bcOh!gcsuQK
zrymLO0RC(0v0;64&}IHlT2uI6xRIm_zg_v$17rEp^KYLPJHC#`kOWow!lXm$*{1W@
zIcJ{-#>kAos;bu`rQ4M>iIRtm0bcpP&Z&RHW0R_y9i~HdR5h7gcK5YTXWv{%${75^
zbEixq)S&6G(Uq0<eiYYD{_7MvI|#ddpi;go6M5@&(-+Q%g(bY{>X3eZ<%2q{yy$h7
z-GLs;^PZbM8kXRBfSw+K<bu0UHToF5&MC0=z+88*z#cIWX^j+o!TJ4obbi>~A2^jW
zXuUbHVxFcxgjh+zx`0?Z5ZeU;7>A-8j6dI7e~-cczSWnjkiJ!zVTs2^7>GMQ``z>?
zZ?=J?!PbnxHGf&6aD>+PczNH?rGZ#g)9i1kD0!WDp-yeAtZZhoEg27JN{p*+2*0T6
zwyQ`q5l>WZ@R63AIqJb&zi~WYZ+r2q4;JgwD-ZXyuqa9nu(xKD5^TF9D3S5uL-^!y
zgKUXHn8J`ru7cXS?n<z*^0i}-eGu3<Dbs}&41yWTg|t3wRY-@@{*`rhLuv%7M}T;^
z`MCp!qTz6R%z$YrlzV}KX64LWGxybxZ}>l?5&DQ;FMyaiphX<C(~|(7V(028k*H2+
z)xBpEdt*`ss3|t`qRNv`l(;%R4!<U8*f#qdpFM1+B!|!^UR=+7tIgKATByqguC)eV
zEbLNC=!f8gon@sq;ELAFvr_XvKtbH>8o!TP_uTLI^cQi4SIedddaEZ!6z1*BXT+l0
zn#P~=Eq2IX>7QYCx@-j`1i^belW_9~{p!@d4*z#Z72ZMeP^X^xL=X41&&lU!++7<4
zFVGCDCp(%S<h=j>R5-)J8H-rYd3S(E3H=^0ssHyb|K!cXdB>uxfp*~nY7!D<yl<iJ
z46AEjrH~sLuKpTLpy>Lv!K0YfQ`Pf^9j$10QYPG0RqzY9(YlD}#i<0stu{t(xo6e8
zoSQ!eT*YefHvDbxRTN$@CBKk$3qz4xkZB}oChafmW6>46KeWlfAo_BX1ur`H5U;bH
z<AiG|T>)P#D}AKmw_kn#CH{$c4F1OymnO?@^8~NAuvcj3S>aZKPpB3KW6R%W!6-#1
zBvjJv&NPiiJAkSVE*~jfT`E|xZ<!b%o`=@f8*3r&`q(pWp2qs<S>!|TYuZylH@u%F
zu&EYkK(B^YRf`vwVj7qo$gv!B(rAVUzA|a>S}xq8@iZn=Kzq(^xk<4$PCMvdI2Q!0
z2L9<tn!X?O0-do;V|r53-b5EOgeMQ56&2stIqG%7A(DAUg?%<v113e9H!i@Mu1tRj
z%X@A8+p=zEKE8ZuXf>Pf=z(wnpIKxz245R1LL-5h^D0zp2*L&liNSAxMRsGpC$PF&
z45oVGii&5itEU@+tO5cLm)oi#*y;<0L7vP4p78}%?OK2f8UVxw(avEGi*)zMfEq+E
zCs2T+hVsFIybd~M@IO2e@CRoDgzyi)1hC4JUI;bytG)mGJ3uOtF%IjGc>F4%Gpwbx
z>&mszlsgkQ1hOK#uX#@?xK6Jv2jwlc3kDfc;Akz`2zoaPY0vhOm7N-+8f?!SYc*<J
zCoXiUdnsHxQ-K0h1fg{&k*|-<VbDb2*kr!_?8B)}S2rK?wXqJj<H<3jb1Y5&vmU+K
zpbVn8EI|H-#3}&nPW%9$4ho>*V18rx)CR2vd2mAw>_Glas={)7VAujhRx97;A~qfo
z4jqKukiJ}KX%Dz}A+QC3lqLseG!*QOra;6)oL{WGv~h89ZIH86K+yHS*@e*P!w&W(
zOB4o%#z1oXohrd!w9@=L&`B=$f&l6nsW!L3M|9m8_?NEB!tn+9Q57p4PlgFcN3(nP
z1*=4#w};k=?oaLB<Q=39b^7@PmZZ$SltreRNXb`i_u;iabn`5R-5n}QbvH+SqQ2wO
zr5MVEzEITuJ~e(m8s;%zft!FmJOw6uvI!z_Fw0r#$toCjBtVX(05%fdIOt@F&>_hJ
zAuTPVH~_-0A&4sgx~K#f+OJgIE^|Mg0He4I?jk&5wIF>KEf+7GA|SZVef-a!o#B`c
zsWt)rva$fZZIq@Y683-y?l?$kP<6a%m@<NTGWlsS^k9>YMF>frpH*0RSSV75req{&
zB;!|6u>)5KkfKfNjX6Cu>IRylC)c=jN@q`MEe&^08h_ZqHn+aFCUW}?w%E;(@E|OK
z8{R;T+L?_9wo0GBuM*nxkx6clr;`Pqq{3qB7z2HkM|S7L@C`9r`IEH;Xkbh<I+5F#
z53*+zZ6DOQYoWvkv7<X}nNjvf#qS~|AN4)pM+;A_m@{;qgHJ4<yNg3wmXsvj5mP=o
zHT{xZ;S-o$BZpHUG6h(Pw6YAJ@XHUtP{RykPnw9J`C-qQ&2FHXLiqQ<VKN0X1au<H
z=|cw7AKL%o%YZ0};F6F-01(zccwhC*jusbfkmMC~E*kTiU;)_PO|6REc?$@Pe&<0y
z{O!E52*r}zR<HRFZevCJ?27~u!=Kx2xX%4zrcujEAVKN>pQ$TK=X+=UcfNjn!)|(j
zdrRPLf?Rn?NIh1aK`kZ#f7f{>xS!Sg;G>+qoqf-<XGJk^Lu&;BD_#Sd<3U}h9twD1
z?{F?8TykqR`&Ljee<vZ@j6vk=@!+OafG~~7&^~Fx(qU>w@imdFwKGVGLwQhU03zP9
zW7NyI((-XtP~dLCmsjM%&2X!4p8rD@4WocSVW$fsC%{ZKBs_fT%nl3;{ReE=oHo9J
z90m<3+%!4(n4v&5{WluT`@4_Ks6WeD5T4sY1rYKJaj9rNbR*w)?3^#Lq(4)=U((yf
zpMTT;$wckVXE(#5>eww`x|`1m5}}i(vUXsta?z!Vxn1gWiZ9TwY``*II49P<gimqq
zp@yb+bld$z-!LJP_k{S*oU_B;`<;!&ye}iS)=P6#EO@+W1!Kc%_*@_^uFd%#BnSek
z39@)p`w<`!4`_@)RO0}=Em&o$L1jcbQAn~O-1~pD($g94mte|_r0tQ<tUpsc;=)5r
zAqQYDX6N0+N35wZ4T*uq9bOP=+~*R5-%_kcyfCSsfyb%CVH$&e;TI`?yf@BQo-tXj
zO+OpXzGXiFkACy5_!Tzetu5MUy{bVq-z)A~8^`ms$(;rVmcK4;SCEDds!e}j13w~&
z?%Z$^4W!EM%gUP0cc(&!^Bj7BXxIrJ!b~t67Cvw*BNie6$;`so&kUAXK(YV)qI3mT
zLW7CMI5_l%fSe7DEG(cDgMZTr9A8{u=XLBrGR4r0AT1q;k_d!87~;r|7|`Ed?cj%h
z)ZvOiBu3~3f^Qs?2i^XLA!X@6Xn?MxGDxN(6<22^rn-u{#OizNAF^WdNf!XR50rWE
zgWMuO6-4>1k6d-@@aJ$6yyV~0vt2DNK(~5w!9xGS#f$^N)a$h~bc628qa+Hy)jS?4
zfb39Yn-ZmFlnSP@56_^xj!ZiKRl`3hT)*!!ZN#s0DjVu=_@|W1wA%k;yXt1Y_$-<g
z+STT%tbu{Y9T7+>xcg}#*YdDx0@Vf41BIeK2U`|_nx)ld*wKJ}PF)DH8{}qcFJj=y
ziA9!ckb?uksS*r(JYw7jGY=Af|Jr544ot-z_nP`^BWXzQ-YyDmNh9c3A@X3|djn<y
z(Sk156g@Nd5lc1n6}#!r1pg4oyF+0h%LxoU5=#V}0D70XUjA<^PN<;c&Z?2IeC%kD
zzyD>(+mx2(me7g+2YOOj(%8i0CKM?C-w}0zit2UOFX6O*cDvA@=9o5xkywI2@X5R*
z9$q@b8&2g2J;^$z*KN1}=QWtY>N?to;Fe(O42ORp!B&vaYr%Ggq^hux6jJm}Cq{4}
zokeIHW1g5Zh}#>I=yh7#e`%mnm-+A}_?Y0wXSH-$Lk<4*GM{8*yA~LFz?%}Yy$PKX
zih|pW5X4F(UkVd5_h2SCqzE~w1WBU-^&6%wZs5V{cq~MU3I{i`R@CHO_+UU#;W99V
z_L2U*J0w8F0H)RdeEt}oT^rfYo7|pNrA&oGy+Ti3AH3L2hyCq^6uC6=Kq7K|<kY16
z(r+H;tDR5(B>9|CcLrj-M#i@lGpX||kX`@OXU8b;!77kJLL5E(<2T7?VZi?KL>ili
zo0Av+ov(NP8`O{<Go;b!&%FIihpnJw7uWq@3)5gD>Y5qwXL4{R3;+h-dpaiCWm3cU
zZnR5P7a#9}9R1sPCG`x=(p(3Ckb#aTO&t~L3taNu(=5&392K|0ZUlySGGrfwN74;k
zkaN%tvF3gE{7dHM-rY6H?^*s<EsT;-t7upRBrZ~lAnr;Ku#qwZ42IcIk$y@{1k=%Z
zuyjlkw5q>9?UntapxTDuWX+orI&6*RU@|Ig7j6AZRn*IXx51+`?;|e^N>3~F48JIu
zD3c}OQfWg_&)iz&;>qp49|a_!w}K3g%UKft4i};`IV1$SP=UBeMab0?L&L9G=jmH}
zQNp+v3zfYudu?)`e`~+)MB+9<@%{~4-QuabIFHw*Wq}Bwha!G({;g^?Y(<6OnY#$*
z@1Wokkb%fJ90@gvM2))g=jJue(T*qZ;7BIwYra3{hTq9CjJtzdJD%W%yVakOydF<D
zcTLY5Gd5u}Oq_cQN8t`7qctjMhXyB~MlBpre&)M=`IR(X7i%Kb>}afO{lxXR_}`z7
zGUn1@Y3>hHQuV!ZB<q^bajtjY@`R4Q4%LqbOeUvaU7oAve6D+7*!BocN}F~BCafSR
z)Z=n61Vv8lLKGDTv!x&4BOzWTusd<vXfSBfe%VY^tVF)nG~ZZ+(+Xwy#v<@mF?L3T
zL6(HRW@-*AR(5|P+X*`R4l#V9%qq%9SWmv$*HU(pS-V-|s}%IQC5X2(xbI&G2ObDe
zm_B8C7PJa~)MGSUcM2(R<@~jve=?YWnZ_^e)s=GotuI8Vq;HSIo}znp_-XG+eJ)v3
zD1G{VDM2}+Jkd<(Y9uT`|JjoE9*)(Wys5kmZULlVhk6}0IpkWPWJMv?bQsq@0}BKU
zLk_?_v>opXu_DKx668vXiYPM4A7E3R1c53w%KKG!<fm3IlxB_g+2IK`F~OtL;RDh}
zGy3ZavK8-as&@EG0zNI?t%GyiOwZU(cr6Q-<N8*AH{g_)Ez!Ok+o#Ge-mdlHddsDx
zNZ?t;Y`2gzlKy;VPReZ~G7(k&ih<)}b&f(%{w0ZBva83a+ecOQGSuXj4u8|G8B&}C
z&mm1Xlfuk^c>>bM2?HYk0HT%1;YFx}C3~MM5Umh;c?g?%SbTgt9M;$JE%ToYBPY4P
zVHlxKXcIP==7i)Ps&EWPvJKe~Y>DvWJITB2m)dt1J^oc8rMFJ7$Liq3_bd3J@tyKp
z#iZxF45}w-RYp^z4--+Vg<M#PJMP}^whQaD3If-r`Tt8dTHovLK{>o(4Q(bQfPiN7
z6pqnC!5oUX5@CxX=r}lU4h8$}Z^IQFB?ZDEQl$WqMh2`_r}6QPfi#C07{|xQ|1gc*
z!ES6_VM3qRNYjU{(=9>vVEKy2kYsfF5ssd~1xXuz4q-QZuB;b44Cb>HYv|9tigadQ
zpKt+=i!pH9a(OwkQIr2npwA7ZXS0U+ds8-oA&YG*svk?WEfQ71rjZDz`XJ&hcv-L^
z3aJT#HKHxW6@0G3{TtFHM593bHw6<6V!s?cfJ=S;{9p^K2U7P4CmM-QArpT#h~ncf
z#PV2@kTS}ymbd>AC3(j<LkEqwY6c2vf!3)pf8q5fUHCg7XUl}Z-j6rwqQt;U)T+H8
z#c|3rnfJ?M&SwF+)41pUD5j1)n23y93!4K!S3#Iv9}msa_~nhh{|=eL=))CEb+;{W
z%!sF0q#7n)S56uneF<XPrf_gn9mb)EUNRb;S}sk@>zsK}%iNrtpSi~S)x9qOGV+KE
zm92co7!Eb7N1!fEO>6pRbfPGA70t6Xkw6^c`7!W0<V76P!Ek8R@ot9?3eHjFhmOo6
z)k%KcO&BswB!P9+LDc5@0#*e>5Ln|KIvpx#%B8~@6X=8Y7Xe9Lc7S4eET4)~%jUuR
z6+AQRH7A|3cWix#Z;zHUkfqoqa(FSdGe%(>23ky&DONjAzOUYh%j@DiM>{D!67ZP=
zFP9f(-H)5KA4n?0Ach4(5Kbj!Z@`ZXVyuwz83{miDLI&)Icm$OHO&Ty*}XgKR#b1i
zFJ&<cl%}-rt7x=(A8Ec1t#gg|QcvTk_Tb??<BUnh1|*3A<%0=0&Z$H^qd|MkF|2y3
zOs>%Fcappt3j&qUjLJdr6azXCn5U8a7R-yl$5#iYv6fq|fHMadY9xZcMM6U$&iBcD
zzoD#wP{X*ljnBx}(zu-Se)5JxYcbmP392y2_S1NywxjFOkVnt}jhNKzzHiumcZ{sr
zjrPXL*$a56<@ct#ETTq%Ho>{S=oUH2O+>O8kkv?U_u2`8N%qTgpN6Dmb5G`%`S}uL
zByzZ1|0P;_duyiP^ztg+L9CuXYYw%UqyS?sMCjm{7B4K{rjF&Z57lT5;qefkRiTv?
zL(PtczyhArRybrnpq5-WXxlmqJ9$yn*#0IQr<M`*tGA1{@Pi$6CdelPg5@OC)NK%P
zxcO+20*dkpSaH!H9y0(?5lSo-MMcEwabHHp1jZPMFAVbE7-Us0akWPb&It4oKHQM8
z)ADkaOX@}2#Jx;ZvR|30c`=a!#4g;IK9!N{n1QH@DW_rg)VfP4g4LV>j?4n(ex+WK
z6#?qCQ8DR9JGgw;Ld_mN$1)n-e>-DyOr~JOhQ%s5PTmjfDZ!$qw8p((32wUQSxr_$
z_LfZQ7EDR^3{StViSvkn?YU`mU4O~t>l2QlyUusi8Dc<2m@hf^c@R!6A{Jgnfnk&h
zs#934mWyU->~Ri`_E!)Alh1xs7G^nca>FeA7jvkF5w5EYThVjK*|+J8LT3i4DkHdy
z3@Mf^;eQ$tj|+120+mYEhh&h8?}9SyfNMVofQrvt25MpliZO=&piCxq<GzNV8XSTb
z3sHY6BCoCZjGPsPXO~i~!8VR$NVj+7_(y8tOLZn2K?SG|LdHek#1=XI)$$AFFdXFr
ze<{2CuA}~*(0FDUiVmnP!-tC`7Zx8q2ejRV<2k|9zoOkB)4*N2a$H2D;H@3kZT{%_
zIqr>c(Ares7^3UTp5%dAxn1=1<&9QO4sZ!TjzmgNT`zttl){~R4{w`io8)vadkvm(
z#fk?c7w$zh{CNZt6~Ur{IO1SUL9Dlgv?8Tjl-gL(W@wg7%qk1%Lk}l4!VRB`Z%29B
zFJDep!s&;OjxaOBMyhnc>wnpOq4wlYBVk1eQDx{zQdi7Sf5fU<rhn$r|Idf>A(a-~
zrXPgxwgkT<3PQALAlu~S_9loW2wVfiY#xZ}rM%N%`Ku1)8oJ9NRK1G+tR^(~RoN?u
zqjf9fS{xh%rj3MiR~BHsL?H_%c=uGo+k~$CQB~-)4ifz@?%jV@zHJQ#a}L79x7w<_
ze_paGXGb#Ox8z=|C<a|sEuKYE2Gx?_qk5Zs@RXti-oBHzu(0@TrfnyE$jO<mw`mLC
zeU$<u>b(jcVvpN~Ui{d;=J%WvduogRPy!TeV!8fc0{&Yr02qs)qqve8GA!mA*F9~Y
zcXsl7?|nTu`wBTLibv|AyPt6Gjz-XflKG_aFSjoRpDFy%{A2_&=diA+2p|>|LFaJ*
z#Ua_iyzb_8@O&dL$x`s64VX6aX}CT*OUu-;3)eEP7yQsm{$Tf)dlWKpKOvX|ZlHE(
zFX0r)OFw_Qcrd})T56_w(Q|#mVZA?};0PT!cn|_Pso$CT*~(UunwwooKiy*{Khlea
ziH#;%`)+K3-nZ^)>l;z$>7$1ySGk?;e!i#MEBrK?PDl&au6!uGId*J?VQ89>tFpdL
zkHmuflDO&zYSCQ(8ZF!#!;d3;4kKO<sjMAR#Hn*n74v<x$D9(2Q-{ml+3QOzEm=mH
z3#f_Dhg7tQg_LdGeOqcJi;)M*Hq3D2z(9n>0iEyfLnj|Eu+q4rL*LmM`%%{Vk$7w^
zR+)0aVT<)(w7#g<!}wzK{)9MAzf=or(uc;ovLSI#zJ2=Dd-eM;MvxbOqL^q6h*XVi
z%+)yG-<&b_+-Ah-T`wR8JtkKp->9|jjV0M6`u1ZH`LviZJ&KmVyX>_#Pw!nO@^&@4
zk$s1V$b4pMi{h%^zLQ2~%J7I#YITlWW$xWU2fdeWa5Z7v82{5b-1*j|G{xx>BhHHA
z?UvxPhq@Pfa+LkM0UI;G+k>F|wSA}yj)ZW!AI`yq8VY`Z*<RUNZ7wKbarQu+a*4R5
zYt32ZOF)m+?<o*?@%kgoeDeqyS<&h5`#(QxiI4E;qp89uW|CAA&!Wt{OJavaM{D&G
z;M{|Sx!ZerudZINcop?_Y5d)QsKYFisAE^jf<QzJ@$KWcq37TF+*cN$R){#>skVK6
z=y1iypy-k>+#$=07Zr8%NYBy!Z+a7)gO`EXV*Zl$sm#(H_5C$II=JkeV+X^1O;)q5
zl$rIB_-?Ka@ZyjdUZ>$3^}I^pn|^{HuQO28Q+uxFZW0*;{k`lIg~_cbD%$)XS&~&}
zbnBK6J4RHKIwcRV@p~vcVU=1=!_SA%_dtw4ZTkXyoM!n#;kH{J{=HENamr7EXXYV$
z2paVNe^k~VjMQ)c;`!YbcFBG>4s72!I-$$**Cl9(C36UXE)(G{|7xnh1$8rHY-S->
zPa-)eP%7KYvS?8-9=q~Z6wlP=f)K|-G12H$C|17mEDw>3o8)$Rzh_eQdg6(|j*c75
zlcqba^LxE>xAzBl48+^9|52*5>~7Ee7LJfoeV*qFQk}AwuAV@06mXIx_X@aFQ!mRf
z6CJ@UV+WKY<g{m`7ePkU$jk%<)h?{$@4*_o4mEv_RnIl#@Z}Z$eenFi87GxstU?xj
z2#TLlw;oVQ{P*Lgt}ov0NHKDLMQmoE&BpGX`%I%^<V{-R-2z6VQI_6jE+auF#!yT|
zgRf+HfAuGW;LuP5Dc!8NgfQ-#(MrPF`p$z=x@li->(lfSeeS7V*K3H$H8b$y(E*AQ
zghr8&q-hh0{X38jz~2DapxpK$74y5|T3T9&BL`vc!{UuX$R2RUST#(Rq|tfW-w^NQ
zXdg5si01-I+*e@9Mf?SDbPr<mg@h0uV$v8r0E40l;#{{Iq4Wjo+1sIvf1^pYI${f@
z4b(nhrRR*1wGj!Ev-=o3EXKdziLs!j7SCNpjccZ2V5uwZz?Ls!o<w6*f&&}vl^u13
zD@;qa^LmUY4NQtpWF6P8<NE5$)wnrV;sQG92-9L@mfr7DpQfR(>~Vf_wFZacc0bDy
z1yvi{mu&A>uU6juoo3^b+Wxc{R0Ux1BUsbxk;gq62GcNzFot3b!rBucs`O`HLBTeN
zuu)*b@;k2uO9lc*1F=)_;N%$CSrGgJ(pE#XX7tiQ9Dqnn3N-BA)md?xaFPdS)Zq!C
zR=HzKyLd84$z_2~ccSW%Xg*b&S<@C7(F1ho!nNwL8Ofn&I&YEp5+NgEVipfC3Mo5Y
zQ>$p)MM!*jT<eY3Hw^iz^lrtHxEbp_dmyBOqWCEiB)VZ0(1eruZLp^F%}eRXE<wQ%
zNjCQKPn&-Bj{q>pC%ULG^);Wf|L}&h0^!ttnWgUIW(gVVt>iv$Jc%S_N4-2BN&orY
zhY<8}SnTIyRETedChr$qGE!MNG3Mm=XerJ*3to@g2ImC23Xc|<UL&w67&Ieecc^&F
z+!4SOczaslyy9qZ;UXjwWDqGCOoQNHfVpy9;Od$!7w1aV;8QQMl`UfvE=E~`-6Pdh
zEPt&<;<|pZoNV}4UBWxJ#RepDE+>6BYkA@fg$kD1yx!yNSSu3#QrlO7a#G=n&G`Hl
zsXTs!nho#BzoeZbagchpRLhzD#&v>`{&{I(Ejv|tg3!J&)9lI>hS^!B+1&(#c~A@H
z=jMJu;@Nq5_@?bp{!)?1UHZt9)BA6vgGY%TZx(k%v}D`t(L0%NxQgm0H?UL;DN6A0
zhleytNA1(jk&CAH>hzr7d&7UVz$mK4sBgzq`}kCFtGn#@1NCFG$ax>d<dUE0jg`o&
zhtu^R9@CpVUdWra@?h^PHudECMw{v7Qg5}Mn6Nb$D=y8}!1YaQBJ^qzRZ)q&V9nUv
z*j>&t>ecz0?TXAQP31An^pU*Y{3Uf!gM8mG48@~w+RNRm{b<+u%TbwXEph7<y9F1m
zt*upbbab$;H7ooS)02~X16dn3gYt9;hIa_%=jN)wfv>98)(aaRaD-MoZ$?Ijyo-wq
zgpRWS0rCO-(pHd>4@(*x)b8o+wSwo|%EqP}B<gsf?)#lZ)~=hQgVJSMYL_!xKFac?
zJ$XsGH+y_<vrf^o{jeDC)4NvT-g^FPdGkXvHOFPNeaYoUb+qSaV;)ro&a^Z}dv6pg
zU}dtzCXF^mA3egv3qGa~bIhGp(cgW<cyM3A&S!8cePf@%V7_l9qgSD?{OgB4Rnx`L
zqf!2JQ5Rw|GzQfe++jQSETM0Gu}|Wf_7`qB6_|&1LJu|VESLu=w==}V3t-C0Jz~&N
zQAva><HBTJAikfNpwrCj;^Ja5{OR)0mo-KZSl5O8x)KM4s=a;Y-u`}1cXyJ%|7mIg
zhucDr78RMvIx8!E;eYrJLu!Uys-_JSy?KuhGG4?pRG6374Q^o$*<VazPKu(qy?Xi6
zJNC@RJ_}CQ8_I%kJWFu&zNnr-`Jz|nZl1`mgFW|V9L3ur1#4n>-;A$WvA=;CedC0~
z95la_R8*>TXMeylYWBw3(()&Wz~4d79kAi!<>ULBm#0QeO&xYMxFIn<o(6_((pRor
z!N)l<SmK(gt*sp)|CyJI>r+Y!<4A>%0fYcl@7}F>?MIG6q0~)UL<ZVO1L};|cRlw;
z3^JQHo7a`2uH4`IGNE^fo(fWq!a}*--^k}*Sk^csAa1Sc-sGix)4Z_R-v3sX@jTJX
z^Z0(I?`Fbi`tUD!c)NYVL7{wc&(6b3-|$t$L!k_@o(sY^1a#e4D3s=HC7pNh9r_B1
zPm%?fT7=22>|M3<fga8lH|gOw_~wxLi+1LR>ruur%YV_C;B!9GSV5m(%+M|Rw6*of
z#nsi?!a{1cBcTsw+5{SC;ZrV7I~aRctAU(@gk$wnY*8v@OKr()*r?U8QxW4v8yzHa
zDLjIqVO@6NznB+!Z^pD+IMXLva7Ai-xkO9b2ZI>UfZ2F?c*wG<{m#wLtR^fRzw4no
zheDAh{K{&Xx^W=hV#*k|EwZ_}`86|>{WkBe$;ZFDM8l7@Dmeq`S6MCuMi!EwQ0XS~
z#eyzzb)8qbZ!ypW%Y-3%@}x8hl@;$M%$Cp*)hE*AoN~KI_?WZ4zTO#v-M{)K95nkf
z4KyA+NQMqRS9O{d|Kl=Cm7${R^01PGgddW!ZoxuTlE9xcxmNhJYW&moTFeXe!ZUB=
zlgd>}LOs7ZTw0I0r@yaT_E-%<5Aoh*V4J<K?BtXyA1_=KG0oce_5lqU3KgNq6{9LZ
zrOr$iX8Ddge=Ry|x1%FGIu#p*dXV<DTJ7zfG^*^q^C(p1iBm4{BfD_>3;a?RN{kwX
zVkSr*@yxG_C;$BoOHE_mx%*d17&4NC<>I1JF+Akz6(>->VL3xf>-&cR;`GRac{fvY
z+H8)*|2ci=s{`+f#Yg+4Q@>Q;0g+~lZJUSj)?*N_Qe{|#yekZPdU^;>>ivH4OEI$d
zQdJ+mJ5zCrutwA2g&qD!BIG@3GQ;EapU)QR&#(NKck7FLK>Nak=4mN~ZxLJw>b``$
Kc!rqa)BgpQbd5>?

diff --git a/docs/_static/docstring_previews/flowchart.png b/docs/_static/docstring_previews/flowchart.png
index 358f210c35369a3deda5e722c5c904b7ffadc93c..23eea25564ceb274338f0746d9ba78b264c2e3a4 100644
GIT binary patch
literal 23579
zcmeFZbyU^iw)eXb36VymK}1BlMCncu>5>NN4(Uz_QIwPp=>`euMnOUl5RjHG>E_OL
z&O7e7W88Dzao_)L?Xhu>jmx#3^~9X>H$UI+6Rz|^1_z5A3xPo3$jM5oAP^|N@Kc0|
z4!`MPyfB3S;d7DFa#6K6b8$0rGDRpDxj5L^yVzJ7-*Yu}a<;U$<7VS!<7T;M;o{=p
z%+Jnl`@g?{&ECnJy=PfV72f2QgRHhQ0)c0Q{6WbV$+JWtbl%8GimAD$Z2ftyb;p_l
zeb*~o?`wk6d!;pewL8QIZ&G{gDo`7uJq07SDjHB5bU4e~+BKBK3A#j;p2dV7p9gh#
zmsa0k9@uYSD~<G7Np(vj>+uTv{hJKMBq{)39Mis6wgDlIN&7c4FdbjqJ@kL!&uf8)
z4<D|Mm!((tz2|UVevWzbhWFWl+~ntr*qodP*yN8f&(6;9F>jIzxE5E--~XIr(BjL%
z$r<BHmXqjCbTL(HA0~RQ_t^Y<EaOkModhnwD{o2>-^S<Nf0}&+Lh)##^%`6=71M>$
zj(9jZ=dXNTy?T|AnTaAGAb?HAbrS^zMMO_VyLh(tJT@u{n@PPe*zCur`So8#pPO04
zeM3{KZ$GlHG9L^rC@7FjE-5LAEj;D<kuE}$D&W?g?suhKxTxRkP4;?YEF>X8v*K(#
zUm2C?;d2DS`*aVjt*wpMWflL4{Y;eOVz<%nN>dRv8SN4p#ADmZJ3i+}XeQVM1O!^O
zc2wj~?XYQRXsT^~SB@qe7fm#Ja937Uebz%~d^m&V*B8&KT%^HjneeaAD2kSchd3f4
z!n7tTF_BbDM~9@r@9+1x;RZMR@S2d@Bw6NzX<E(Re1)22(eoW)39d;ZUDMOLS<ARo
z{H-%H)XW;ik2p?LRaI&4VA0a4I$k3Z;6JXna&9thr)fmPmkDRn`zk6gfBV_9XZbp{
zY~J49b!$<atE;Pq7bpL8bzEJaJbwK69y4=faPZAipZWHX_a8q>sHhN*-z6jrEY)ie
zXSpk(tW0qC?%l*EwlCB82?+^@$_zv-EG$Iz^r$Y+k7b;kDkW*`?d>1iO_P58`V~K1
zvb?6I`|NOIKg!hDIPi(><lM#Se!hCq-8;m@6^@J2$USj+Ml9gEIrl>6&S8?w^XJ&;
zxRjVeLPC!oJ!)-hLqYyku0fM0@r?ik%<1mGfAifwhwJK@i_0=Fm<HFa!1(yP%RiFC
z?CseaUVS4REz-oBoSalrQ<MDhiO2C|*C_n-woc1<nL)?M2oVe({wBDl=E+x=wLOW~
z%di?j(b2f~@83sU_~~rUw$NN(?Bg+MmSWS>(|ccCkRcpahiFHOwTWPDhK7g5c4nIX
zZU1QyT)b<Z?s2?Lg^o>%fZ@K8ygxB9VK`M|>(TwP$&<&%#wH^t=ca;!!f2uTZLj0)
zzqx*}b4+@oL&C#vN!}-LdHs60+EP`}eb;|u|B~-YvrJ#?9=_qX5A+k&mPCabC0H@C
z!INLC7vaM^tdQHSxTGXABLktMqeCg+8YTSnkwl}wg5vY%{rj{9(a|pgF0-?<mCN*n
z9xyPN9IOrxjg56jQwa=}>IqN^dy`V}yZDRV6TLh;oL^b#`Ek7Tjp)I{hvL%G=uhpY
zMK{JuP!LyFR}L;N!4VNy<Ac9`5%2Bok#fDleE$47J~ef;YTgT@nMM!7FD_e?#JRb-
zTIGf))PnB*tU5KKnwliHN!XUoiVa$*pFDYDF<wfww6x@~KFa9s>3Ls3fPzjs>OCp@
z<HvV|JP%rjG9@_OcMY|)v<$~f^++FE;HRXfa=z-tM~MD{xtbPcWMs4$%_q=rbPtH3
zlW89vy%UB@CEntD`TqTTe@>I`*6-i3VBy>A92X5OEP7L$2_n1DqIF1#iHTiShfs8C
z?YiJ(8=9D)BDCtA=wTAW^qV|sg@mZik9TAe*z_FE4lM1?NosuEc4oSMBtIQ)@^T|3
zCtvtgq^Z^7D@4v?6%40MOir$#^GkP651p_uHH_?|MLM@RZm>A*!xt%Us;eJkhKSqD
zHvh-Q-<teVVc2#<QbxuE&Mb@zom$u%69)$e(KThB1<O)vrw4mIijt4exPlbkyQ?(-
z_222f`F7ADd_HoyR;TN^yQ9d0Mn}~IJ@%0CYxQ>(1Hy2k+=$ovghM)-@|Lx=b!T_?
z0-PsIT-?x)A5}9qTHJQ@kr7EE3^zpdUTc<A<WPl)Or!VdhjZq(o*s-pUtb5Mrc&Ky
z(eghzamS|Q#gAjwWczpYu&dL(eQGKhrT2J6r@OnmzoR3_OUQLo3*G<p{M^~rpxN69
z7P-H(GuYJBl(U;ur_r7BV_ckRx`>}y6laaqFD5e^n|@fh-%H;<l9@yyUj^UyiQP1>
z9m~_FPjd?kzw5X7vMS3!Tsdw?NFYINg`;wBI1h&U9QJS)D<2=<;Mb@2Gbu?)Nq@B!
z5u4NX&K_sX%*^sOLG8s_6@e&*4Q@NjlRrmBk|HB-t12tc*T`&h!Sd=AJ4UWIc%M3<
z`=|0c23nu{{uzN6m6WhrPuEpQdWmf$%+@r#-nfg}>bgCx{ptr|0_OX7&pR4z2r#!X
zF%7r3x3LjAUtJhYqAHj*OF0nz{r%64J=WINo@;6*{GObwjE%ihdU>!mf{jS#G(jKN
zNbxb7YYpgcZT0toES9Bsy!|KX>sS8T&7-X;ZVHNI@o{-Qz0|j)oLNJ{)YR1d6BCJE
zR0ITV2o76Y+q)zrJzvscs3?YcFH#pKH#avEGc#$=_x|1<cZFz%E^7Q-#naub&>VJB
zox^-!r=rn73LmPux%p#@A%AQ5?rraBC3e?;F$;PfVI$_&*CUFHS*9C3h(rfsnu>~;
zb@lY-cXsd)m{?fi;^H^p|DqshM0|tO(`jJZ{UQI!vS{AJUn($#4MkX{r>93iLD84`
z5OL2jTKKpuITVLt;cs7p?DOZbiL-FxU{|ucTt+%A_miz|Y-G#4qt;DsY!sA!NBtJE
zw=NvNyl!^cZG-2qQ|2d|^^ReCK|E>y^-FyrR??_BRfLWXwW6XTU!#N^(n^F-_gIMz
z?cK=CqM}It*XwiLZz)Jce7)1;HH8EPAH>5!6zc5kY=sbwg@vW-1@Xwp((+sCOO^rQ
z4$=+?14A{oIvm}*Ya<NXc^>=AFwmtj83_qg``P9n#|QFku=#G@yg3JHsB02^c73db
z;xP|T{doQPRbK)-3SwurMaIzZ-r3Ps-(~U*(T-3YBXe`jg}*Ti^Lo2bGi1wsd`Kt~
z;&FM#!@|O{SFNM1?Kpzc`t92-NSC_3E9foB$rKQZ!{B_{z1&Wg-5xNQE{bTrewTJm
z!z3)6X59d%{?lX28*mbkDUs#w;!evoy1#eOwRw~8B{{^F!Tq+S<EgyrzrNC0a6G=G
z3j2^F3yi?P8`ag-_wGhk_+Fj~Zs@&-z1N?}i7G8E4Jj0oPJkARG^}bxkHu`W5SwnD
ze<TU}{d@Ni>B2tkKR?Uw+obQ*`fR&?w>PM&u3pL|F;71|`9}x&x!iK(;hQew@bs(8
zbJX5bU3A1l#WZY)lf3YaJrCDOIgM}HP1n(K&OqeAzjNm%9UUF~mC15{k}j1LqgDD3
zy|Nv}5aLnkXBayJ{}PBjJUm7VR55&geck^ppb2}QAT%{KSJ&2PVYMM-?H(V8eg6Cx
zFMTRLF78Kd>g6$txw|{>^q)Vln&GB>@ol}mnBtlE<^w5hdw=`(b?r3cHFsmbWMte!
zT;pN*`T12_kK@|e*&%BuY;5d%?Cf|@PlV$!-R8C$edO-tWwJR@;o#^P07viU4V14C
z{x;=_VeKOY6A|BCr&h)$p<r@%cVC+7fucxsfWpPX;?awA;kV8!1M|HxbjUY0>rV{O
zt};XEJ+=7;d9Ej#DyXcC{VuzH6x3M@tHaq+`dKeWh2?XC0$VU4I6@%*F+xp9*sd`6
z7Y-isMGIY#$aVSrnI1*~xqc4L9&+@SBcU7iW@hMf*R)4fFH-qKAWx(SdZ;YOZhVhp
zrYC1;8F`Mz`Q%CWM+U|Bsi~}vYRR)ljI45Ua=fmaq)|~(buBE>lzcbP(9lvu{HX88
zyxG{;+DbLQ3LM+no$uJ)-ED;hudjh!7+zHQ{n@rF77mWoV)tA6+>p>vr;*62L0Iqn
z!ovH^%tc!E;@`i2UxlOa_4P*R!GVj*<^;k1@{bN!!upXH=;-LX2M57<c?`zJ#&b(c
z!EmfD58Q~H#y7emNj!4@jw4t7#bK;kkw%Hb$^g~$*VmEO=Vyn^P{`dRA|eXk_z7_&
zCO$rtm{}u3EC^k<weIUzV@T@m*H=Pr(L}wcc<al{p|(>sv|L;&#`PjjlCfSchg*%|
z;o+yD21c=QagkK6B`?3UhvCsU%(bGFvT}116={}*LJdOx#6~O2lmyB!G>Eujy1JCP
z3d!BJCfeHCkV;>fnmYPGT^P6X)aCX|5pC_?f>Tfz{vtGqk9+yWN(@R8iYGRrFJF?6
z`yOp--^9Rh)}$vkPmhU<lY$WJu+$slJ$p~s_oDZbnu3CvQ{mB__Q^@|C{B2cfEt*2
z(2@<ln_8ioT-E@MdVvc4#N?#$%0Mb%0<S@XD^5T_fQXRrHyWycBu4OnGEr+!^RmV!
zd^lT10xE1MT16!#Z^A;}cZ}UwShxYHE*O?40qcf8gb_E^%i1<P1k9><5IPQW=4mBD
zTOsI{={E(wdxu9Y<f)eRGMUF3JyYpTjqQ~2>QJWavuEv4<s!ppety33JJl*P1>Ht>
z0(gXw`1~x+(8?+T!s^nLTkm_?i8@C*C=$Y;+Ue-*ya{s)(e`P%aTiv&^Dh12_8)=G
znI;O+0p%+wlaS@D_sK2_Jlr5Fv-;)|o!ah7)4riBsoMvezlk_GIkg&GSqs&RLKzg(
z?pR+=R$B(aqn8-e{9|Kdb&zG*^cyj4ZMSIj!xlayB=oe0+#YW6^LuPFfe+>09NWrc
zPR=%{iRzlWew$xi9vkS^JE6T^A8ng!3o^2`eOp(@3so?Jgk4_}7TG_=N7}%E=5S*?
zEG>;XFes?P><4+Z-5>J8!a`YDS${~I{BGOW7#J8Z4|8x1>%5LjI}M~oOs+0Yk@0bT
zyzCu>J7m#jJysk5k9M8OD2!4n4Qcr)fAt5TX8$98?&HVDU=kG)7FIQoD!>i#BR3<X
z6<N)!jpXg8wzzIhVk4kVn46qw_US`#(9vakuiV1IGKMO8uDiCp+;AjMaV2*}f8g}r
zVvqMY<Nf=A)&d@T52|Ww%^`sy>n2v+y7#d2SruE_*%GjC-7<o(_FP#x2IbM-&w<rd
zi!qp^-0bWQC@8C{s!V2PW<1W~4xC<GxYd`4-h&0T>hJCj#i16|GM@6ahJ;uQ0sHv_
zLQZ*k`Q@#xAGQq(#RlHqg0<94Oiaaac3B=gcrfd8Ujz}+xXF3!GysLxUx7;~eENrn
zW5Olx=YIa&1#i&>Unx1yK=~ZogzuUSDsw}~Gtc$))7CaNxZT{`Y9MEn4#JK}f~`6-
zHf9DvNWJ7`oWiHa;q(t5a&?o!08BnUR4~TH#eJ@>9>>Vc%ni#?WA}$Y+?3dbZv_@*
z|IZ73eSJb(*n;Nq@$txEL(S*9w<I@KpvquoX{o(>^gu+qmDK0EIOgf$C_G&oSJd^&
zm&$z`=JOSQqh<=nkegdIYYU#(-7ii!p(cNjn&s>4(y53U1JA(vnUJto|1>;BKH0y2
zk8d+kj<UYK4&@J1!l2eksPHs>uxwmz?nA4w;&6CInOp36r=+1F4fFU9^0LZm0s;R5
zLoj|?rt?KGemnP_|Nqbb3p3#Klg<BvoYdq^L_|9@XkubwXt$i)#q{;5LqkJ3I5^-5
z^Svuyu*{>ATMG*d<<^7Ljpz75ESUEg*k#1jx=BrX`+I{~NSQV^vJ}hohkpM2;Hn+W
za5IYY;nw{(2RV6pCV$s?PEumto^2}_S=ihlBjZQaqK0Sme@BPR({(N=R(o8e+*S{Q
z#WSn>?hA@ucX+U|uwc^|FAi0jn5%K}WDl{+aB*?v8(W3OQwc0e*ig2&x98{)&fhvS
z7;t7cXb%1qkNdc~`m-LEy`_bvgn~Q<{4&;+F__`ei3c0BV6dJ&`^**pRtqO#Acd`Z
zaF#C2NNtw7i+W#g$$H|=FMP_ACG$RJ6*+5W1$e2ff<oN)%3Ie%VGTEXSAwQnWN<qY
z2E(<QoDC|%>d6l%*U8BdPB$8~6$B`oy{R-@k95-u_DQl>gf%=bxhRw4+1pC2HStN#
zZ+$&OFE7)+eJLo@RuigPz+RRUF`af`#Lp;m=OSnH&EY}BW_Dq&{(2vd=K%s$phkj6
zp4pvM8)amD`Q%e<F!PFcxqJ8HO<s?h(FAQ>NRlogcTWEf6;P|NTu#=uUle>Se>LV(
zr&f03>4ue7gGIAsy1nAz<04B8#Oib%icP20>UcA;e`B8N1-qW;&e#v7+CqZ$B#MqI
za}r7EJJ-5TO|VC#al$IJ)i?QlsBx0Nh!U%EYyJM+;rvsrc;@h$ZnNB6akxc-f|qP9
zE3KZhvHM;3cutO>iD#K!Na&Ma%7fFS&R68@|C&NOgWaD~ao%M6RhVH$EQGx^Mo`va
z;4t!TYbxlg`df$7X|n1P1B&Rl*SmujiYdJ1Bng9Ot<uHl6cyy#)4S|P{Ddl#k51$&
zQhCeqpgf=O+va-c{-gVh{<h=(`d{z!?c=5!hzZZ72c_sjJrP?C`&$VgQn266(>czy
ztJ?@UVZp|VS{uLp`}gmzLdFNXxAcQ=c62zuYSw&EP+*Nfd}#6QdS6PFky{#o+N@b&
zv+-j3s3F2b_2ILF7Uw@S_WDw-=XH3?WxxH2Fpd^Zu7su!EVD!DR)-~ue}AFz8qmFI
zg5BZi67TLTD<{`IrryWtbxNL6=ZoRTt`&TDOCU>Z@nIco4$YvnZ0Rlju9qU>-*2R@
z({`EJ*`f~0siGjd)r*XF#uVzd+;9%bc?L!_<#jL+8Knj4tBe_Dwo~$rX$?+P<XWHZ
z3qHR2tE4!xc<r#y0pYLBvoC&rN+}dvU9HBV)j;s<`}cfY!^yxd(%}Y*?!iPHDm^`f
z|EEy&+{=cWm6eqoMRvHg4}TEMxe@D~9dvB$lGHhEW4oT~`e!@8aA2P-N89Epc0BF3
zd~839f_R@46R;K;-}ZAXTBli73}tQ2vWuTeT2?!o&R4}|dF%;cS(oS8oAZNzdEMSr
zS2u{l4rKOU9yvTIiOPLZgYeI<wU6NOa};^xOLiCQHif_=MrLQ*=-2DW4pEuD;L2p7
zwwAbtM!Xm6*oeeto+843cf|;rkU>5_UQWB<u-Yc7l|=8ZWWtlE66S4+DdS&KZypa3
zh%i3k^Y92CzY&lzk@pIVE6eREDJi+pSfx7t28F+=d3KX^s_`s)Z;llYqP362sJznc
zq~N80;|68`qBwXXKO;kyX7z&sIn}?9T(sA-B^VHIur@#9lSIBD8RWFNFv#B+!&Lvl
z`abc!Qr2Is$=}(9A@r`rIv;nEhgLdM{`L2>;-5Xn|7OtQG(wF&9SZH{=Y{I*W0i#(
zPi4%^9;QvVFjU}{wWntn)|+PTf4?2l`<=;M2VW%(_HfeBPXY-sWrA^bz0SNuQt#80
z#JqQ(LLWkma<E8vs^5a$V|#6__1MHJJf038lf&&7e<PM(fUxbSj-7UoMi1hEp**pP
zIvY;u!*BJ0`3iZZ_~CbUk9&>WsV}fm#xFUpcjPRlJ`1?{wmgDOmFd_wwo5L8b!<N?
zE~|9!As!93o)79{yWd}jJeVoe#U`rq-njpx{8v%of}wU_9YW9Z-f!QG)*GdG>qaR2
zCX$>{E?CLP$XI?az0G?^Gp|_tRg2e$>7nNj)_;|osaErIT}2Rl(aZj257+qGMs1Ez
zv5N<Bnr^os5K{E?HvI0`;kW*+jujJh<E#epm%ioLnx&_IIX11nVYsM-O6U1ns!T(n
zm9!(&Cye&`k}+$?WqSlK_g7)2^~X>PL(7OTH@gIPp<9fsE^aT<1l)J;-8Uv0u90`M
zOlY^x!`%uisNw(Wy2Z4mf{6z^gEn{<8|}aSlaWYO5W{3-?}HwbwldX3GjUqesuiV*
zqtfn26%xWbm@Sdbt32*nNvw<QYadBR>xJOJwg0Z>@~m;`IqV~$KifQ>3B<+Go3nYD
zw6W+nteK_Qa#T{ML%(v>|A{v~ISsZDt)4(eObMTr+E_qC5b^ShH2)m-^m<_`&m*zj
zk3DsM8)iu*?6jZ51Jh%_Ht9Uc@@T+S(oklsGM#RCjmm~|-FoH|=D51R!#nQQ@n+@^
z{>2jKxJ;RfY1Fy-LtMSa&>)hstQSF{q`zx1Ac)CF{i;_K*>?Q~k;No%+{W&cs5JUp
zoto}Xf6g73k5Kz{ny70pOP@9JLG!YDdlaqVsr{p>L;9S@!wQqK1@N>OxKlu+Rp(nJ
z>7~k`<FYn_7em1Ac(vr1Z)st;dLp+vS$mDgYQEkWLRi*wwHVU3;_c9N%;Hv6Bl4*>
zc9q*~5_@yD8SUN9A1TOBrFvq@6|GbBMb;KsUe{vN8_%TGhnlYMxtHnD50@!d#gmSH
z7E)1F$;ik*k=T%w^~i)kQrMn`KHs+@IkB{#GPSj#*^wJchih;ob9@!VVg8d>(%V^n
z$rUF>xo;w)H9hkW|J}Oj%}pC`g=c|ZL(Q5;swyiTmM4sI9x!xd)>L&qPN&0Ao^8Lm
zrMF>tnvx_d_kzuVkCP_LVR+?^)vqG<VLR0h0_~(E_q{oN6lcP{jSS=vW?FnuA%$8*
zc6sF1^H_#bqoSiC(BD#r=~0X?47?UeymE8v$2rF(@!YR$kIrctFBRV#e^70~e!7f*
zw8`jMV-t8#t@enq$ne*ZP`=r}YXQre^-;7q?6(<3_WtGOVp>n8ZbL*c`FkWpxcHp%
zCynMns+iR?EOH+E@ZWvovd=ZuR01PDk<YY*vy@AV`x3)RRQsHpEBc7=ua}6Y^TxGE
zisU_fQ5{XZidn(l&mWZIXkIR_(P32herrKiA8Ai1KU)2AT>Ad{fzK)S>_>*(<9wqD
z-vt(utJyRwuG*S_=rmz;SDVP*HwkcHyA_fJR}9==e2UPVXD>J2e$JA@Nz8qhH9;#r
zFz`Msk*@_6CZi&96qZ(&bo6{=&??2ha#Ot0b#nU9kc{|%$7ea-O13k5=VGbk(`&u*
zf33>lHQYV89oEy2Sf(nGd%2hUozkOVC?F92Z-eLpsfAj%6*f4X*S>ov812OFyOit%
z85HMXRHRY&7bZ*SNf7x=J?Yu~_mOsNeC6!H!%m6)7MT}>d(7E$44=RM9HU0yQ<L9d
z7C0yTq4QH!NLU8X=q!;ytMVHK6eo|yxFC$jq{VB@9lXyFMex0gW0BoXR`)!+?g@`_
z53e2_B`o8n{VCd}_$)3y8n<}MdQPmhgfPrd?!{q>uoM1>G;T+%$8Dn(wc9k=*(eCC
z2rMOaRSX16IyX@%RHud(Z_4E4cKMtz-Rw_HM%qjePVJK`tF(1ImgqPNtii%w5mF<0
z8S(Ny+e!7lBm;Cui!u}Z{9<z2WbERB6;O-u@{ddXW$M-yf?ymN6CU1aa4Yzt5OPAn
z#%Kfk3+m}G{h-kJ-(0vm)6vkK?oSm&ab{4Qo1Ajn2tT!UD$RR$en6lqgWJ`!iU&gr
zs{g|Jgi0{f^DQ}QuLtGAmbGSu^&&qU1c??E)5n%0)E(W53*AJFObff)Tga#ljdiE9
zYP&EolWw_D=jByem-Rno{bLF5e@fyR&(~MF<IBUAeAzJC`SZt2zH-Bhg16^Io&<zZ
zC{$CeD@Ul?Wp6)ZVN|@|lN|r|Q6(=U{|%b;)BD0~FD5_7M!p-S;qd#>O@r$r{+$A9
z8&R^0;T7H4FQYvd>VMsMl*5N<W^yQp2}A}NA3eJF<-L93hrO6jDcI2YvP=+j4-8CS
zfu8&>-!>89f-$mOr|;}qaGqvJOd2K@%kfs)-A}2n=R5fF#o=hxNK(2{^605S{(Vzs
zs3njg3x2t3ZQTZhbdsJREu+X2^Z)ojf^gc~JDqU~h^6i(<>KbLcRQo9neFUg9gUS0
zXsi_S(-FP5vhZkx=b#ZD;_DS3OAW10QZh0z1B3J#3%UBNx&QXzF{k}DHnyapVMcY|
zy`1)pIR;{VB_;g8z`z_mbXsEc|2`l6&$HG4v;Ov!m9v`zn4X2DC1A}tXpA9|9stR{
zL+B{AMG!ORXqfmy^E8@9M4zpX-}izST1@B=bfc&Q?jqrTXsC2OOS0^l?<|Wo1eIEs
z?P)$FZ_86mmsV4|V`FRUc(AJ6pTwQl>hYc1#ZS!=y%QQ(!}ZSQjUM~x&ddE`wRV5-
z@7_I9H96WC$Md^7HIKQxxVS))1ZeCPv~CTMc0`eJ8yXrSq@<)CTmHN`m@bmAN<>0}
z<b`4i3OMBC<fOa~_db01@Zbzd<w{FSm6VkHV`2#Ivc5dvF4D50|D7H&`Ba_)P2`jn
z+DRR)tqABs1~oSe0|CwW{25JM{TG*kE3{aW*cZ*AnMwQbAv)|S8ED#N6c%E^2d}QL
z-{n>G^z<xXz1|ovb8v8I1!Q@9v&v%le(}pMSjax6s3-!OrNfOL&N;ccPCn4a34wW7
zfL0r?&)H-6QpmZsJ6jn{A1>DBCZ(WQgcg_<Fgvq>lyHUQCvlTsFQTSxU+SMf3?oo!
z+xpXxfc){=nxAh&IEX-#JBqU#I<CEdAUh7ZpFA-GYO>-xXb%TO0M^FY#pM<Bfs#Ib
zs!ZZGuY?lUjvs;S_>941=H}#dHa0d!@EB-mkvhMAZ3?f1h2%6aFhKXG5^!aiJ-qO3
zLX^LGg9$AF=-rrpu^u-Q&wLLZr>bus8RS_ubTmtKzi|%$oJ_2(eX6RdxpD^QjT6~c
z^4K>sGd6CAtpcnLdRoXv60%zd$Rx3$K|uc5Gh=8qokr`#UX1Yx2qYul6#CcBwzqEG
zg1%D)=rMp_dL3<&p|)<%H1!?wc^$D3Vcy*Q)8GbQ8n9FkkRoklr`UXN>05t$d)CmW
z8>pz-k#|`WD=WE1e*H3S@;WvHnzoet)Mn!STXLQVWHZdt%8Cof|MqNZMc{E>V9na`
zxtWy}Q!<CqjUu%ILRA$N<AH$z!l^w#KgF;H&m|;+0Bv&u#$8@qR9)rw0a1;b`z(eZ
z0J~%%uP5A(AHPyoR%UtlFqBineYdBx6S(Bml#VgAh=>T#*zOwp*@5<sjwGNJ+?<?7
z+~xyI-6*Y~B=`+oVl4nIUYBRDWA>3O7A9$VdwT#{$(<nG=%%Eix{Y%M!j9#Ig?4m2
z>i(>(EG;-ief|BffS{SeQ*{5x$VW5`j4F^Qa8glGh2J~VfH}F1gJX)A0RD!2H8lSE
zo}R?T#UbZrA_xO-0=jCyp~sh$l0vCdV?!4C=8d6ZnxK*Oc&RB8AtQgFS8U4qvU+a)
z{E5Tu53~GXVPV{e&H0Wn!Y$~C+5q~7Re-JZ9Wdd8T=Q42IAHI%!qJAk0*!y8t*P3C
zRp-+^Q*{5XuCBj2iqP{UitMskA1z?v<xLa<0R@lkWbEa?9va<rz?fV-JUmZ8Zv;}0
zZr9C;HrP1(hli$7#El|<YZaV(&Ru6IL303<fqH*`--KPi(GYMNvdb9~5&}EL`p;J$
z>&a?Huf3&PxAE}IffgW<2D7pJb-HfSwVaqNAMDSt!a^os5Se9V2E`to#+(RI2?+_~
zG{8zo9sryMfk;eBLPFe=jk1=xzkj<9H-P7#9q$Y_QF9oIX71}j_nwxPme=n}5PFz5
z5s4yx!oU!;S3Pfqh(BUvlpC#s^93o#2y`r<lPa&S?u4o4^w@iago9jMAIx1s#A)50
z$;itW4cW$~rA32w$!Th(EeNBxzh4rtmXVDOHln<$YVnMjiRnHoYsth^(>)qtZv+C6
zom}<C+M44CFueCGqiWlXP4?%BpFS;<(Ybxf%{5CiEz_<-kB*K;4r?5?*Ao~CE%e%!
zmzNO;FE1}Untc5x{;^{1Ao&EgOv{m6v~5Mv8QZg_-d<VYbaXJjsi#P#1c0};sVR#6
zl|h+TuO0xT=(P3X=C1PYVZww|7=mo>0)h;H2DYOutUPqoq@0|%Vq;?wh%X>B81@FW
z6$cmB!j%cCe`I7I-Y-&0N^BAm5`aLZB>+s3WCZesG6Yl_auP(;+qd*LzJqcI5I=l%
z3E*AGkciA4vxt&>Y7yTk0REy<QuKq>ySonIbJy_!0VqJgA}FuVVfd}BqDZ#X_bo0y
z9v#p<q7{H3r}v4&$=O*JvgZu&EX@;3qD@hj7V-jz^%x6aHO|4uP`GTnnVSBk`Lv^X
zeChe^`*&n`uJhP`=<VZE;dyBFJoyO@Fj9}`Vn69^WhWPx$3QsijsOKf`#Wf3-)G3o
z%<MiJTXZr6PMYE2VL~4#1Cj$xInZED*4Rb@cU(?7vSkHb4&eywYN&4`b!&HOEq(hE
z;^Ml1eA5EB;$%JkVKI~;W--&q2NJKjjg3gO+jkAuetst9u~Lia!&oh(7UzZc=>7iv
zgN3t`lQ;lUNWJkZE9a_@=#vIPTw_1eC<e#<=^&iF$YSM3ca{r=;n$E(Z<s{=2k2H)
zS2wb-(3<|m{v<%uA7GOS>;(6dU0R?SAUe`8^btYoc0l1*=dwmPegsk%pz{uXWc77*
z_h@NRj(28P&U8Uw2Q9dK)jZtsp20!Ig=a+|BWi^J+}fg&l9IA}u%?=rn7DLW%gj{W
z5;uFLMhP1d#w`w-q_n?&=Vm|&f{i*t=V6HwA?$)6Pj}N)X_sh^2JocrSC?C`*+n5<
zAPga|P1HLxfCdnmIPdfFlEPW)5xKrlUDcBMF`gqwl$e~{(b0kWjz%Q3r9}i-qed|~
z>Wv#n%?0QdZ^1x;mUbpTwdVj59UL4y*BMT<ySLYV@^A6BkMkJ;EG_b^AT<JT1So|(
z?;!OTadE_&@bsmKPfi{IVnt$iOKVt82+P4BsX#%1?re8;m`+k!`bMvZGa2N{<$=^l
zc>R424t&U5Qn0J*c5Z2DYiCMD-X#@u=h&XE52AQ#X9xgub*4$c4XXs?MEF=C$T}rW
z*e3!N1J@8VNJw!-zy_~zCuV0?S0;QGU@|ty9gbHQ9tR+XkQ7Tp&L^Da)n7#<Ad83u
zjR%b8hTHaZTOiu)xy40P1V}Sk2GT{|LE=UdI>^!US2z)$CumTK_9D-&3ZXdAY4)z~
zq)&<>_EpC-iAojp_@G{-@%G!dZ*u)!aO5;VkCvOC-#a=wnpjoEJq6OXr&MUV;i{wF
znVFfzz>J?iy2It{><l`cdhxRLZ>f29f4%}H`xnN(@``2UJF&N0y{TA&voAzXb0p<@
z&T^MKMz*v6$B*BDhU(X#nk<Is^IS^m^H3k(enZY3#QyH?+}SI0^D2OO+>oIk5{4ry
z1xRIaqzq4<#2>B>2NmS!b3md%Lb#*9eI((+u<m9!<c|%Jw5wD%>GFK11<fu&*T~4I
z7^zr7=58h?F35mSKqVEQn)-vZuYsMjTkHI2i++93-tW78^72!CPfvcNTnL$+;fNvC
zLa$ec0)#yFf;qb<Kz^VNI<h3$2gb$~iHB<=VUXVndJn(T8*V1I;N2?l2Q3k_<A<}u
zVwq`kj5Ab){=bQ)a>Sp>%8Ei=o^}M`7_ernpIETkJ?K7&g9@`99wz=l{UB%y#ga;K
z)O!OOG?9Y$Alu0(C<uqr!_gM>DMO>9!I04JFw-rK%d{ynJ%FxdU41?6g9kT3@{7ze
zva*=d)6@6pBpP^VLo%;vbJ{CwYr_B;!jmFC1;uU9Zca>1$v1#p*aJSg{kzb*BIFuS
zlCD{#zc<dhrq2-v-Wvw8sHMMu|KoRr9Nmf4k@uF&8$t1dFto8@`TGpBfD~{TZLs;O
z8s2~T3>O7j<a__|;0iGx28QMnZNPc!<_%E5iYh9;Yt9pWhK6=+AHf+4nU;2LM(BaB
zocI${-;jrNh*5QQr92JRx`qacd-$>q4^#q_cdv7idY7Ww?be?a57$5Gh$_p4w`T<!
zN6|vthP^}F@o0}HCnqNZ;Pg+nNHxPd7cH3p!RD4z8}ipC@p-JqV_f-i?pRn@wPjS&
z#;U71?B~*4oh;%3aF=ahQ4t%y)iFB+`8*~q4o6<+Ye?HpH*{mxw12x_UDQ|0(-M=y
za980`<~<3csF<?y<mdM)xf%JG9x*o{W9H~#6|_D&@so(EaB4bR9Ki7dthUR?nK>Pw
zl>Q}hIol-r>~AFtupYKlK0UU+vov_x(94wa#X=K#cOm3N3aF}_ps1})^!5*{NuzJY
zYqIhRajuqszaNj@I<TI_dfkr=%|Qee`B%d~AN*qQX&APCB4m#hLUvKnL<S#QOy7j=
z=;Cw}L!zKL3hvf!_a~|S&w3j5+I;p;9^!6e=*TeGO!(boxUe_*S?qy{gVkOTbpH9J
zj;H5WwFJeJpulwV;QIufo}L`}Hkij6wo2nA1a^8&pbi=s=w2q4A1=<NN{MnA&Qumk
z5%JJ9#%{U2GF%u<hy2~n;FvoJ86&4qiz=JI(_$%TvWC}Wd%0%np_-587Y6|+<8<4*
zX*apjd9$BXfDng8VczB?*`+|Wn5<Gp=S8w6TTt?A_0cB<!D}Yc_gHyKhF<&a%~3xo
z(F&BSS~gnfuXk`M4f?JgrCg?LQas&EvDI+AbhZtR>bqaI=9T0qN;dfCm)d+hx+CAc
z?Xq~6mai>^$C_$Aw^z?iII=TS7j5^0+;OAUriO%zo4bAu$quK#5tSQGz8bCNXag!%
zXZv_En*kQVqlS$3b3kBvPa1|uM|K0sdy?<Yw#k|-UEhMvIZXaJrR?H~&SuhabUAZu
z&(z@}-gqA0H90$$&R%APz9f#z_8*eI756yz-{(!WhTE~XR)6NvM4$g@hhDXE*JxC-
zux|#2v4~G+=%tJUbilQy1Pl%HYNpo`oqOIH|5-pI-LECV)o(<n5)y(&yviz<4=zpc
zhf1ukc%&&iE<CSs5Uo<_s#+YjT!NwR+HYM?tuAmPuA`15TVIQWUu`5@L@sX$%&q=P
zcCh}5{V@z@VLdY1^WD6VKu1j|QM7(jI9X)E;_6H)(3O}@QU=`3phl%Pe}qZqzPv)i
zJ<&%)^y_`0hOOjWc8gKLg-T>lcV8&%Wj8t$D3YEp$w5ejgc}I|gX+gN`t^5oaMR~X
zub*;pc6Ipc@YXp~SdH8Jqt56Uy)uqyF<VKY<|@&kG8#-X+F1WfUud#e*|ke5Oz0Ep
z>168~&4YqKBAdCO^IbgLFr^1K5YIpH-~plzg?>FW<waa=Uxq~sooA@-G$Z^u{RUfY
zTZ%%c4_kc6a3zR{di(lFCNL0<<E8ECf4nk#{Os{BLp6Iodbw5jY%z^m&UGJJ&o(i<
zbldz;bz!ML(d2=8gJwFc$V+u|wg&Aij%hJHM4lEO(@^6NB{a{=VRwWornLK<*x6_h
z6E!U_6^JS5Q5g$0-LIvi6@l)Vc?w0H(OH%$^xL}!aI9$=7{UNsq9X3o>EDMa6vThc
z$f+nT&0oCf)C^UW$0J}za?i_@r|KnnI(;wRB;qLpb$c|smZ^vubHQ`#uwUW9<cNyu
z?y|b%3o0X{)pbr$<F68|=F|1t26Il4UiAHaE19c5Ytto@vz@6np@7Z$(SCNFXA~P7
zmzh-<TBk@;&q-;;C-1E@Ib6^7q2=Lx$GKbnD~p?ma@Tz2;!l{23U@99v-5Lqnp71c
z{OgOxGEK5Qc`Lmb_z$-um&*9YL(68pFu$Us6Yl?|_8<uZvV#4A_I#_}7ELNdd`0o<
zjM!Mn$DQW*8#n8*OaD8Lf|odWH|#)BncbUxwebGj4jW44O~RQwa>{aRc^mg~a#krt
z-hJ!sHQb%JrS-)S7uxJfzuF1KV=$F>W?CnT3p?M5nMwfFeSQ0;a3eeCz@nEMy}q^O
z#=R#GUoKz@JXpPT+)x?8#H7jD4a<ExW@@|RugEzJ$h1#2EVst>1=aGhyVG@Ar?=wv
z><@e17aK41-otF3*V+!ad2S!$_ZRBS&8H&5E%A14W)#*-WiI;l6@Nveo^7D3z@eV0
z$1x>VU)3zsU2TMhDeP;eqe@IhMl9bn>zPticucCWpE7$?l%JKp@`xoxn7HS7N9<Ln
zyr9<wrq>hu)_<2?2!!4_PU}X(ycX9!jbx*H^oR_u(}>fr(w&Gp#q{#mhgi_M?VKq|
zp7UD^l%xUnU2bwozByA(jZ_5mp&$}M{aQ2Qxup-c!YMS_=0SuZ*hu6bw{Hft9Tg$E
z)7If=GiygV)j)y7DL>jh+zAL(R=-Y;7r#Zx?l!I%f6S7^L-lSOeeN<@7D@3PQO_1S
zubL44D0Dq@fbJ)Dn$Y004A0Ma>Q<czFQuc{j$F5d4e^B={lWnGFZ|u-&=uYYg3Q!m
za`NVWi>~JHd@Q?j18BQj0F)r{N(#P10*M7`-prm>+KC!7G^BoDwO|CvjwdRe$R@|i
z-e&$0SeOji=O{M5fq-2XOgz)zu-0$sVxxwKQET5ZVfjJxP>rXy^+2kx93~ZhPZNN*
zo!yf6%HryG6G?9WB?*Mdj8tnX_oMZQarO?Br4+8;<<}fm^`jjk*+RpGr9=;L*WMl5
z(`B-bp&_a_hi-LrUx$3I%FB-Ql4$hXNB?t%$KvF}(4Y8LilXNJ?v)Vs+9bgi`Dj}^
znP10rcW6V(NRv9n^VC}WsqqqVfpLD`Y7>g((w7)3XQf>;rJm?<<-R*cuoK@!zg;&P
z>#SIvsA3O|zAozZVEB>tPDisWLk@@8#ZLFalu$+YhV<;^)z-=B>HPGUja4D<SSSok
z4Uf-$Sqzt>K;0Fw{ajXf_050lDvJ$gObe>8JSDVDP-a6)QhKBhV?Z$rWS8+SmaodX
zc;JnBw_%bdbUfxNQ}2sil#ijfqDB1C<-gQ>0xdu8Z@FD{a#lMQ+>(*tymBeJ4Oy5z
zm%F?CY_{IQLY3-;^jFe}0xJEmmiMb4@oB<ZLUDqtEKNsCtdN;?ZsVz9tJ7<HEItpu
z6k+d-1I^g?Tu4^*XNHwxxKR(|QNnx&sMSX@K9>Y#E;6*WceBk3!znWSIfSx0QN)gG
z&>Y>(s$&u>qbP{y+kN=^t0&r#%?xH4J#+cawCN)zMkCqUXhOWaBtaW|{or83zWGJ1
zhLwBUf4&AJ5$*Z1&!2mp;_np!XZXy-%zV2!??r}k)m_3Zg;X2ltXyzU5`9r%YHJGR
z!QDQ%n*Xa-Y8(a1qD-IKG=M>UwS&OchBF+Auu1vQLA=1>x$c6@pdcziI7J`1+3JSD
zu#s|r^74wVz%LX8`?OdI1IvOa%z`X;#@P)zH;Idk462@3jlFK0y?u9avx4r;DSj7g
zk@9}L``IcjMDnX$egYMJB(TErA`fOyCo)p)U>vOeFPLqk`+$Ny;K46ZI9L~ClO-nM
zRvs=p0Py|Ct~=m#{y{-B`dnhA8)wm<<rfxo%oN<Ep8c`5f3i)LUtryOlzixDd4d5M
z^}#l=+P5LhNn^d~7M~(X9>V)LAO!e70o(3Bi@rmvjuT%TS+{kMgDk(SBm4o^B4crD
z1@RqXd9}X0L<@_J^=H+Vk*^6Th%RRjND=<gy!PqSmN1IU&)IaGdM{q!u7CP;axlwV
z?{bKPf2gr%`6~D#0{MlXDJTReJ@`<E=fk#dziq&dkC{<Y6gF|aZGeY}iz<wje`_k0
zli?3Set=oBG01=NbM&Y5Gbb*9Zs5?P&xI%CZMz$ot<Ci}E1~#<1De}fVGWbrjVcqV
z7a3hjZG0`@G)R2Ti9Y$&^IBn$@8|UaC%UnZPq=V=jq5b2W4f?hR7_G*c;|`x>gqFL
z+6`xM!Rl)IYyNvnnvzJY`Pp{h<JWcxiMl*jVol5R^#bKSvf9hC%$;%q-6@crDP?O6
z5OrX`RmjLL3PW65>z2qF2rqsM@|sQ+rm_-PbT+<`ChUtzfgyts^}ptUC+d#o<?D3G
z`1nr|y_y8{`sU&}Q2)g;u~JcKO^vq(`RmZ*wtYxjzlGcHQ76AQUiRkQW$;`p!+R1e
z<sa>Yh#{|D2dK~qA-;s_j4bndZE^qw?(F|ugjh<kTjw4nqDWwsCax-kFd6mhdLX2<
z8D(0h2IakT<Qn?h<W}oxi+-(2$|SO+#HXhBU%IZ|)VsVE`Mmk>vH>4U${}C9vPTJa
z1U#&bzFz-B!hp9eQ6?tQ=famt@pR7D@9c!uJDO6QKAWj-_a3I`fU^^c?L>t7?PbHR
zMs1EKMG?OQj<UYS@jRVvRW^zFa6!!7^R64zd=7<+fp0OQMRRtqdnDMV%hC38pLe!K
zZ{4G#3(B>fG+MuqBK(rV=SgbI@Ag@S@dH-CcF#4%hoeidW2y{IetsBlcO?CG@^y}i
zgh#}?w^G(543v!yIHBDSO`8S3(ukaDL-ok`8Y?0~tK5}tVR0cx?^(V3@iI>?yRHUX
zrlPDRbOLgo$1ZV(5U1MWFod(a6YLdb(_PKJhi6aNQEPTFEtw#Kc$JeU0W1Wbx3HoD
zub;wsx%tZc{<>}3c<09m#`P@<%t~(OtGBmK#gNZ@U`WXRpoM*XdnE>+gn7$JmQn6a
zWXLrnDf~#~dM8uM1PLaNuZ1g5i2;bQPBjuxk>IH-*gks1{MU;DZEwRQiuTZ&tGTmB
zTv47V2>l=ogubB4{Ew$U7BJfZOzQ!P1pWZ@v%Jzo=vad75ty!0ljoseZ9CLI|Kqus
z^1o2%6zqabIkMJ%_E`hAI1@Gm{CT|=Un+R`PH-g)RwH#OqgI(84zUc5#isg4B#t=<
zUBCaA{`qm1&;(l|6gvQnt}uzn-l6ncaw9Op=px&k;FJ@$;#rmf$_lN>zm>(nWU|2m
z@g91@6BQ=Fzg)qO6x&yMigZ|?><@~3)(gFf<-4Z?I(kY#O3|QYr(<Hms;Q~r^*ne4
z-XGOD!M|i*Mb^)smZfvLtSUkCoD6!xZ7F=tLnS&-p|5f|n%IR?4=7L^+71_5FJC@8
zkW*0TdP6{$QBs0~G-9rg-ujOg_I*0K;jgc)0capS86X=MH#1vSERk@GUCJ<mM&(4a
zk5|%|stQtC{Js*Lrnso6s9@rKXFc-sCsTr{`Kwo=`ufef@rT*aJisR+3IZEMw#w&c
zx(!)b#sT{v8kTr;C;lZ47*Ffm|J^@+`QimvG|1O6z;{i_?}7=ca8f>}2mG#^Z=i3A
zK!EIV2nGbMwnca+P%?|pHv7=Bv%fptUrBhGDuT?mU^a$-_LGgCU0N2c7EDGp@U#6T
zEDQs+^%CeRH(;WPa#{**Gwks}aQd+tG^d;kwRn)4Y{DA+54}CW!0&)xzkwdPZVfR3
z9vm&O2Pb{_K)AoR$28^*t;POdzv4idZ)R%R1(1LY0lcchdwl==@J1{2Vo6P+^uXcr
zt5h!;kWg+}*&w*LSSpI`^}y9S$P4XQBXA8o2ixJs3IHZbO3DfNMoHi=2D}MS4V>Yf
zfXS1<Uj}n&3_yg1kue;&>R$mnU#A{6S>TQz6B5k9UAr&|P7C<_jWJfhZeY(cf>CdF
zgyHF?gA;=rI;C*9HRdKj@m&mEDCB1X{;UL_gyR&an*f~CG*<n_M5KiPBoxbQYu_Nx
zk(uzF4<B!QBVDYS4JTqqgkCt(LJBtSY4>_qh6yl(BS)*Mq-1z>bVRs81~y60qs?zF
zE-r~MXfVi-!;I+!Z0yrP1`Rp(*=2a}pd1>(+yL9|5)*d;wjnhM!SDlDCu(bBQxpC2
zKR|k5ov?wnEQ_FE%IVn|7n~6!54591AibT?^?Hl+eLy53;WN<CxLft*iz!UaCUg}y
zfxE-vRSeB3%E}tDv9a0g&b6^HF+~8SuK)?o(b3j_z-~byA#Dc-4zT%<jtmda&-gbd
z#K(7wOGv;BF7*;1NA~mQ&pjMpAhBcMqdA&uLp?hAXL59P>GhC-p(mbIryJIH0{Z+D
z@R7e0*bOG&M+-0`lC0L9VPP4vv$G?)Iv}FNf&xY)I|R7P^YjwzU)YGqckdDs%bvwC
z!My^eu+q7T>1iuJ7Qw)Yjquu?6Kn2MP*gP6(9p1H^+(wR#J-8_o&zWVx*b^b3(UUI
z1FZmZQZYAY=n6h<<VS(4tCQ4(1sddJ<m8p`51co?X}#F(>+7omy+H*Wo^mi+9Kn7N
zY!h+Y)&&>GH*nANAc-_Ud!lg3P3XwOXr!Y^+pI?O+mN&!`85GtO~4rT_0nmipcC7;
z6O_)`9nWsS<~TZ_HQzQ!+V|Au&!0cPVLQ-I!j|fSFa2PdLcrA~d|3eS=*N#AO~7`)
zn42!-RWmd<Xio*_3``tmfOuJe_*?6MK?y9fMt~1bqlsu}(pTOHw#>ox2P-o-$xJpu
z3IWpsHrwV!Um7PgKIdg@q;k{?$fE#+KxaosBlSWxqyJ%THXo#Zcn530U$ep0_GsL|
z50Eega2N0rC3PN@8MJUAHbHc9%H{^E<x!&!j9gVs?G-35R>s;O6quTsIr+fl0_||#
zyPK|jXN_3Q&ruBDQ44*rxOPnj`34uLKwS5i6~cvq)w}|WlwlMnQbqwR;@W$Ioz*2e
z{k^!L?*%Cs^IJj6Qda|3U-)<T`S{3y(<A&LxW`#E!^#1Yl}j;jb96k-Bo7J-ng^#@
zfN1NNUu6a~;3#xIKVk)$%bvvI4gZP?PJq<Mlm8VLCHwU06Fqq?xHG4B;20ut96(e@
z*hb(N-woSf{SF!B-$K`IuoXV5&Vzx#!Sb>Ndp*qe_7|Y1pq7ZFGyn(P%uI9Z8W43+
zAh@5N*fI^LrKRD@$jBgC!Mr#O&L1R30-W6eH*7He(^j_xbyN;Mj5U$R`YuH1MDtH@
z?xCqsxrg`Txai!s3CUvT+tH9t=jP^SEx1_$$XuwuKLS|w)vgn01`O@&qQN_cKpY$%
z#sCBdkwdurbGQou*!<vOz4J<RCvUyx<<ilqK9D!09Z1OQXesGvi(>kiX5cnh3C+z(
z>Zt)N5Zbq%veF=cnkqs06K;7w28zVw1!!bDf(=@?&H>4kk(Swb7HyTKy~s$ml?O{A
z{}oD=k<OjF7MA*9ZfY6=4)=QyLXuYi?nsS2f~_@b0lx+tdjMzmmp^rmsNihlc>I`U
zdz~ctKW}K@1gI?NeS!~8S)lmL<;@EKxfe=1Z}@-Wu@(on4cuX{a0ZUT`#{;x`+B>&
z(7=%RCOmxp3^_`Oe{k={fE(Y77cZbC&j$jL{fma!rS)VQJw;HTz+Dvp9GA75LB}+%
z$Z{8nJ;7<;4WcYC4CfNE7jb~isu!Z_bBDE`^rNtyU@BmjBm&w(+W$bWkO5f#eN0Rn
zYz=}tcY<M>@A7IW@AH6&3Ya3&rwT4G<P8Tvy-T!s(7~DwG8iRgWn>it@v{Qlh*9J`
zSadSck^rDVKE&aBaUwpO;wuSSC%CBud0h*#VgRgq0%qpo;^Mk9B#RsxQnIkNmeA7^
znw8gK0DS>y6WS*xNP%gF!7T%@Um%d<A^gE03^$3q0ec?OkqJ`*`1_8ouC5YXAAk_8
zGW%f&)<a>RGsEF5sgvD#)T`4KnkZ7vKmpGyeehwHyZ<w;wx6X&UJ(P9L%7P|EdbJm
zvX(2RBZ|+Ew2(qFxHbaj5^hkrV=b(%uKqlQmjJA|5^&=I`C|+J&Q8f`SY7yN<aB`0
zObI-ja7oBLR#qHf(91RUmY{bQzp8RsQ$a2pXvC&Z#~_0*!nFgthlf&d7Xb3&2o{58
zN)Tv;M5~W_Lsmu_g;i8k<~BE@0H4ZvQ!BlP<O;rtyI{w}kL-FA6(tFE6>`hMg$fv8
z(j$FhLyDXWKlqL-PrthXjGbFq3DYQf3HnG%<b5acY<k~Qt7st10Yy`SD`~()y#48X
zr^)C1E(9q5kPtLb)nkDiV!GK!XThZp%-h4c3KYmH4@5je<KsOr>-B4p)RB5LSOkbg
zfiq2B*&r@T6!LNnIAmsIY#SWJgFq{WRHncQ2tdWS4~8q4pRO5PnCEVg<S2cyQV;(W
zX#%>LL}4F(xVHiUnucnN;Tyuj!qWv@V2CYH&4V~Y4IhiFa3QwKS4U-GPSpC*5UQYn
zU;z(Nmn|hOJO4lFE&VUvWc1(msQ&NyK5DX-mX>d#qX$N)si@w90i97zQnpJBGH(tX
z9-Em)GO%Ck>KGf}M_ynAzhcyo?k#%<S~OWms30Y%^S$IBcLgaKm;~UK7TF3eabF)K
zd?gH*1n+$tl!Z_S1^|{r1}UW6QTD+HSC8O}pU5ih@8%{pEEBkTL3tqd=uQ@7bl{Bd
z4XA<AmOH&W6Y;x>f@i~#VPV=X=u<E~hFdN`%me@BpJd2MT*}cy(27VrAVi9argwth
zXWb$-c|IwHFOJD5;!+s)Ur&1Jddz*$@^Wz@UG>l-M2-%qEHDe;jf^IDE>zwpCSrm`
zU(n<1PI}7&xduvcUaLZ;^4f|3xISa_*RRidxQuizeCB-e&!2ZLmrUU&$KK#(wg%@N
zEgduSRwbsGtHxbnX}{LKIrAY6<<7YP)JGS7s&_8__QORhSI5+u5;S=Eg~NVKdmE>F
zd;~qNTmCw>GSybUo=yIR<{`~?d=-OYdVBx(PvYqpm1@66jm%8n_K1o7WDAKUz`AmZ
zg<j5JdW)!xrsCvhe3eI3dO~mk5gOc%f_w!E@)fD6IOw7p-&YRC**+n=u<~C*Y5uY8
z@c@m(#Pz=ai$D55Ep*>awUK>7=Z;Ay?Xg-oj%|pq3!6`++O8u{yp}isxb9sv`jTC#
zE>6Dr1%2o>mBJ6PNG^uGF$1jSt#VZaHL`0h#eF61GU{cSfG+B3&m?&U*p?QD{At^-
zj1@u==wFCD)&Sk=+Fi|?-C7ljx1r$>>&Wxd&7tQB_WCF;|5s0~`6dd6ve|&~1!CE=
z3o>Z@#9P_^ZL)uFfQNR?F8PR1tT>P1AIxtg(m9#mEcD=vE%6E2<>d8sM5OPQj?wng
zEhNEP*GV*#9mlm;e{~x{x;0Gn#F%IMb?HZp<IIAOtm}zn#tU=4C!Y>Nl0ljA_6Ku(
z78ov(K12yXIDUJ)PcQ!yGtc=Alju$VQOd5wrM@}+z3N4Mp7{W^z>RmVtn(I+y4N}|
zP!M5mTcT5*U*;?xNmC?zny9s{>TFb`o|ecHA#WX4iYiqgU*_xjILnwIZ~me~y7=L+
z>)#7)7u!kv=ul%MA_b;8T@B2S<>*tW$R>+YpBBxZ<m&)|rx*JyJ<zVT(irZ5*T_b_
z>#IteJ=?vaTE4y{@x-AH1?JRN-^>g+N4muGkkhM+96OyiX=B4{{pKW(gY?_){P6vg
zcap_Hm48LH6+ZbeZce|09K+31{yCo4(H)?~0Uu-7kfu%R3kz?Bv+H9Cr{naueRs3L
zf;3q-Dhc+a-7AF$cVDD`zkDrs64cy!NWtf5Nb~iO21zy%h;{N$4qv!Bz>bCN9fuFz
zooS3{{md<_Rl9OenBjrp#XUS5?RrX++#d*pjef<QJUp7#sIEREz<|ege>(3BwhjFp
z6Wcn&sdJ3~ks2<?z0y83anp*N9Br(|d;2*OT7m}_ENTvCyC}ulym%v5dSK!3x``ni
zFn*$}q^5!dl=IdqgwhA=SO9+v=g)q)Q=DR5zev`dJ3*^o_QEw0^fd0R3x{4A5%j+1
z{e}~gh26eO28V?jetT&D=J3!FzzWh-+Eex?(D9cQ5HL2gD5`*p>;7P`qOm&Pwfkdi
zFO^og2Ga#YqYhlKDfynvu34?~O|7jhh9z?JqNe#NNcD*bEX=10Bno&4j`DmN%9LR*
zy*@fD;<WlMLMrHb!&@=6HAQYwsX-q#QKKk&smy>g#~Ooo)TtX~w&=~}0fGLaw9!m#
zcw&ng>QkU|)=J9iUGW_Z2n4^U=jv#_F#qvZ*g(^HoO{EEh_`{)Bu+n$DXqvT9Ln_m
zBpaXBq;{pvP81`Jb>Buxb4zh6(w5>RqfAY%J{^KT4EWC0Bo*{>jj_${39;gUECBiD
zO4+Vd2>+OHMzA5ax1~-!2=p6WvdkaPcZ~URNXA8}Fjc*%U#87V^U-W@aS|H8V4w+<
z5NP^meyqRC=ZOg#x5O%AR!J<Zd^5|CRS=tc{j4*#$3sk18{dL<YjDiRJV#sTJ|p~r
zM*nYpoOw8uYahqQv6bae$&xK;Jt@n<7-U4vSi%uSDBD4f7)2P0b7BZ-DoIIYOO|Ze
z31ucC>#-jD8T-D*n3-nY-*evUdav_d*ZZD-{558tdFFZU`+n~4=kpz-*A7?EXQceU
z`jv%ncBNncwsux#u$<qOvs#~Th9M+z<KuWyh_i_g<o5lx_WHZTZHAkcwRO+_-ah4$
zGTHgQlze+bp;s!;q;+(act*TzOnB<-<K1Ueom5ax|Gj^Y9G<6VHnKqqiLRyWhZ5dc
zRWLp8y;7CduB!UWE^M)@<pZi*+6H$~9I{H{rtlgweT+lWLi<u6SF7kVF=;-OG&+!q
zD`H+$%c&>np<r|u-J0El+OI}CbpBDiYS#Lz-RnR9oIz=ByOP$L8R=T|#KVdb%{;X|
zZ-dB|Qm*)V*oDqNb^Q3AoL?|+WUfS#>W2oz5AgE(KpZ_y%Ns&NFeRv)cALF$4B2YR
zxgUQf1Q!e?NTuvtMWLm<L8EhlS(6S|SHe?mx2lO!;*kv}4%O|SoN$n+*337Ab=;Oj
zQsq~UMJF&o5K$;>b~SPuc92MMvX38g>I$@avtwbe_kHkcvS6s6mvnT;5f<(vN6-?C
zv7$>ox{ELg(%FK&&$e&r=VWDAFStQ6Y5HDV{{u8bvNWbyD~y_`6fdX!ofh=rapBnN
z76S87*f#;x)6#2u7m|N5-Z<X$I$urg&VlkT<ph~gQwX+^b)A2c2s~}b2Oc|?D^`7f
zS;{6a1|px;RZ7$WG9%W|+<dJr8}(h`?knz>9^DRSv1cg50{T{Xe))8)Q3;8MSKR9g
zhDe0DDB&vc!U-h>v1rk+OTzbBb4*y0)|yJ4w%eVZJ;;v#$otR)57tobQ$(!YRNCR0
zojrsE*a?z&u%I2|Ykdg^^+y(6e7s`^**5EWl;#mRcoMzKq^nxUR=&Zuw6yLDLPBqH
zUG8{D>i~NvF;&#Q@9pP622eFV3!spio(u7v);yPU@%W~6%CkydgI9be*TSQ?%8?!;
z@a6<x*gzphe%JId0k%MjIX7DO_J$}BAfLdHwk`IR1JzQ(Yor->!FQ1JgB}jAsH#wF
zyZ(<!c&LjRn?Q98bN$ln)?ed!>Iw;q2isPwED&8*;K+@-I$Q+LqFi9j*T;5<(l2X?
zESKgUY$(jVBq$JLVv8Y&L*ho;co<HcTTfQ!^|&wBkVIM+l@pHv?Y_vZ=LQNh-96~P
z6qBtOMFU?oh{@RdQnw-sCPP}%GUl(S9B#t1;9ipY!143=f*U%zpe}+6Z;nJjr~rZ;
z83{g@;^u=mhj#<H;5rqd+F$6u3zaguVu_gO^NA;^xhvCJwP7Ng-op{0FVh8OIriS#
zH%UerzasPZ`k2R#ilwOSFbjjhlf#u=y1dyAuLY1ub0%x+E;;VRrke35{6|MF1j3>m
zW<fgbo?>5&UC}tUbFnHwx%WK`OpSTDHXRv}d$rL4@Or$(i~G*wocbpTUphc27FrzH
z)ps&E%6V_Ryn~?e3ucg0?^=&m89{{Y@An&?`2;yQp<GFB6mV;W0b5)Vr$Qpz;-sA>
zSr|#IYZ+B?Uy>d8tv|9(GSUnUC6LqxM*D=Bafi{pU&@xK34R`(vt)*3N0BfPFw>ds
z!1%zSHM$gdUR^xDC!M)@rX!tSMWv<kTsLqD?kDWyhM)JQkIR3B>H#+)SuF6J^DusC
z%4^?;FD*G-aAcr6m$%@$B2XwXYz}XAsiCma@H=h~vh3-QzACty^l;`c53hdou@p;q
zTsS+X3Q$41eKAfFogy+<p;=!NmMDf<{@m2HvSfD9t~)@V+qBC?Io|q~e_-1Ug!-2%
zp8oKfC~aN-?g?~nh|LgJejL&c+gI#y?u`g3%n)fzQP2fm{r~-K{shx@f64n9Q=zb%
z$_9oV{UhgTf>sgM>agZ-fjhT;IZYYXCl}qne6BIu*usJjd5Ez%T(uL{n6PZZZB2!Z
zHE}l)CNp1Wy0kftBTWyG^r~b!5^lvr|Cwo1To`e?b)dh<+3Z=Kpn_W*WQLXhk)5|!
znWJKf?TgIaBJVoIV=F!3%zR7(B*Bj+CQ~&b0!0Brivzog<gqty!@?ydw|2S^8X2JJ
zf)5TvCG*vUM2_~x3<*%GSz<{>JtXFC|06MfWFW7yrdZ-ng^$_ZobuAqe0zF&L~f>|
z7!9=p^<GT^N00tMn;C(A1Q^z?|J-iQ(g@L;tI~`BR~0A-GE8DC^buJfnM7RN!>6L;
z5Oe{@zb3*6pu%HT&KBV-lYVJb?`P1RShr78L*_dD9kyyqJ^~Zppg{ZN=M!Ab*OujW
z?WeK!4j<X|dw<N5`49{tT)Vt6{8!E>>I9OQ$zku9Qm_@8z^ZoN^rDg&U5#XDUEoTL
zjH}Rd?`Z|=Gy*@$QwJLD>pa<|(b{dR!AMch)Iqg)wXf_&X+U&Ll5aF!=;y|voG7&v
zlHMzu)0PjJm*2h-0rJ2JT<2NOS%8}O8Js@^m~EFM;c7fc{RC}c$1P+Bs;ce=?czhx
z#r6mjQy+vFi1*=@o@lj;prD*2h%;rjisn(M6NM};^SqVSezEAa+bYJ+B9dG;<ixEU
zK+^XkNTSAO9QdIi!)dyqlpZ)dWiZ83g~>4^GvhI%BH}SW^3Up}B@Z7Z-`m*O&_y$U
znB!7h+@~cGY<;;nUr<v{%WlkH=S|WO_n1aW2_iJCo<519Hc<8P=;;wAJW_l%vS@^H
zJUX4-PnKUFqG_q>$>9$B<4w=2TCT*axp_Wc&uUodc%V!6Rx$yZcjCywqoOV~$w!<Z
zM6uQBM#wl{Lc;pdPcVz}&>aoQzZs%Xn%`?cNey{0w6Ei>cL++K8asVAn8~M68?rDS
z`@l)vi|G;F8}-1E7x}gAw6zP%3V!3~DmRf~9m4b!`%&y9_dQR6UWnhje)kR9`X1hw
zGG*9UsV7Il(UvVK$e9l?FD?AHkehBu9zmflXx#Ll>ye0cX9?dlGt0`#abb5GXITzj
zq7mzs4>&4n&(%<)K15g_L~g^s=h@fb@{qDsA16?6-Mw=s?LbSOchTf$<G(OI&p%T4
z&|I+2t@Yc;&uD6|b4}Ox3{Aal0wcH|`XkE12#^w|v*nlQ9)GGuyS;z`5E@2Kc1rJ2
zsM{SZYKo!FhuZ>)_V?WSa+`&*g#{kq8ru@*<H>W~2>}?lNIB|SNcucfSXljJO+^S2
zd5I&7=R@tf2hoyh0rdn~#>5u>O>px=*uLgLsR?d>SPPtVsH3+tza>aLzd!;J9N3;I
z#J|;V(=!dl`cob6!}=igh`8e1u)SxZfl(?V;-4yd3SsuykB)5%i`Q1Qm3Wp%e=pZ-
zDwzE?h`xJEP9Q{QFij_3AX;zega=sZ{WTsUFZIH9zA#Bh&CA8(FH9uDMiHt1>D61o
zobNSzT05$3lo?K2@Q%v7a3XgI2~oQ|+C`-$$Z(*ZiD!)3P9>D6JG*)dLoXzdlcs=G
i<NhJ42B^D*gSv73GI3PJ<vkR2qm0j*pUF3Hy7MpXyLznv

literal 23797
zcmeFZbyU{f_xE`L64DaVARr=LBHbWZpfn;aA>G|6A)!)=q=1BUcY}a*Nry;xO3uE%
zznOXFne{wt)~q#u&CI<P-1mI}*Y)|Fv(G;J?AQDK34Ng`kB3c(jX)sq9xKQyBM>Ow
z@Rf&!0k3qjT<F8UgdF8G9aU^h99<0TjS<fc9BnOa94+4(-f=dzcX(@K&BrOo$;*Dn
z%+b--L70on>VN+QP8)kut~=*fZSW>HZ56Z}5C{ST<QK|!iR`xs#B0OHvQnz9Nn0~6
zsw0|I7`vwT`3}WN&{<e%Zpf0LJGSeN&$jCp)z+-ngw2%K9!J)n%;HQGv9YEy_}}**
zB3ZL-l+fH48p1%A?>!d@aY*>`<%<B1I5BUcY!kWBoj}4CD*F;nUqb1wD7j`OEP7Fy
z{~O=0N%{Er4%dd|$BY#c9$Hb}Kt&PvJiW7_-0<|t6Ko7za)gYG4Au?Q2wJgMV@xr;
zL)l7n35kh6Yj7EuW^Y{o{q=>O5DWV<QP3_uzvn!sH<1rBgn*7S-cG<~HX=<X#Bi#@
zYIQVEBgX4jZ*P?kosf`_+1t0x&CMtQ7<kezUJSU5zogwbSr%tvW)680%bNE1<Gm`+
zfA)!7;n6#3cQ#PH%P9Cvnx>~;r~jIr)qOO2k@UAWF(8sgWbtIqhill5OSkTwu+w@|
zf4@@hm7=2J=~9#g0+Ax_C9R-<b-Xnda<nmS)SJMgS?9VZF#1w=0~7I{O5oAXY<)Ai
zjI8Vp^&hW-g7K&p+d_%WX6xJ%xpqv8y)WJW{QcYgpL<02Ir#f)VSSx-gV$R)g6CUy
zwv3im^>o}VGBQkUZS9<Mje^${Vjjnz%)SWN(AK!@L^3^%OLae5Cl_`!(~AG6!_L9M
zWNtlPKtoMKQ|!1l#1J4MCPpKjMrdzOdO~0;Cnu*BFZm>Gar>>rKu<i^V2%p&>guYb
zoE*B-`bcvqF>}>g1kdW)n!)Bo@oQ~sYtFK=vO8Q{QQyAZE%1UXe(~akBuv%#ZA!|p
zzON!O?6=_pDb&@~6GYwY{e(Y!_%Hx-27@l8qC)I-ek^Bho-0e|<Ky$da)R*e?2Ir}
zwy3<keSI_!C&JjoBp3!{0jAN`$tkF$gm<h!2ggs+w?9R~s3(qN`fd0J@q>5a;WF1(
zmj?R!2qq?`k?$|jv@0wZvh*4}DNy|ow{G1Mv|GF(u$`w-7)Ddc-QC?iQDH^p=;+Ap
zact@7=I}cRhb-NCre<Jt6!{6}WBH+%=O<5l^`_6{9zALqA18y0f1$1(Tu{IXGkf^l
z>92#Wt#5KNEdv7s;=)IJbGDw!=lW8BRiglxl$6x-;`A1x$YxG_BwvddK5bxlSf)Q&
zY!6}+b?HfIf1TR_#^u>jfS(_VEW@qM&CQ~>!&#_C9C{7Z&d$y+UcU6JsCf8VN2l0f
zMbY{!TPcOx{m=|zT<R{7l%5_PL`z4#$4QmAUdocs(K;!tklUQE-#KlJrJ0ZB!svZh
z%e!qo*PxRc45QmTHg@~TlP5)$m7Ng%s}2G7ZMBS7fByVIrTehHwnq9};^D*F@V77v
ztdAn8Z|dmiSpF_Sx0)<#zPdagEYJ~w#Ws?sL4bmSGVK{fD<*Ak&kKKUyEVyQX8D^4
zK}b%HWol~rJtxQRQdCSV?fdt4QUMsvy}h^#3k!nIo8%__$s8}AuJnC1u(pn@c34@Q
zt@oHHGC(^#I^waJ6)ot8nKm#r4S8raDJdm|(iuq;P*})i@8G~=Il&#e*V585@IzYw
zmsXS<4HG{wHkP!)YAP574WsGTFFaVYs0h^`ugbf&KE}lbW8hKFw}+FFi+c+8{`yr6
zQvjI+^YP=y<oxC&mX?-9R#OinDEKfsIy$D;sERdT|EX$!siGoN>%3KDxAaKV>x=|u
z0O1RhNd_Yz1-BcWm>4=!>yn|8ttjYyAy|1CuH86QX|r^8w6Qu_&KB|h{oeX0TfS~R
zB_8F2rr~U*wDff4^;g8i#4!m8?_fe6m-QzJ5ySFY>Ub~n=n+PsH2p(JMGao(=@8io
zA}*YUMn-!(jn|8l<rY;w*W$385DJz3Z5<uyieH5Zv2Gx5r_GQD{wD~gWNE20Y8rE6
zW4y4lD~55f*hoHtk{^3*ZO!#?^_IWC|3JNm3!*8SUZ%+B+8dWnoVuc-;$2i!C*-_a
zL`0}$WbuaMJ}?`nhik+b1hlxYnEoBCWI(96{@Z6d(P)7cK9Kp8Oi)lze|x%G;n_1;
zSX(cv?U$8pY#5k2h_9}$a3#|u^mzICjm5>qr#A*O<oOWn+}zO?=kT6<0s^KcCMN8V
zK6yGO{Lyg|>gq(#Pxly&B8rWB;&@b4RQ~?`+i~w%0&Et@-+KG@Z848ghpmyDn;Val
zlhbVq3YM4hkXKG>Q&Q;AnqZ`v2eWl*ou6rFXry!S^OrZ(2v|*i($dr<(Nscg!W5&Q
z<>lr+u?$#b*Qxo0q7U=4GRe!wmsndX0w1?jA%DdS8G&WAis?wF<jqaArbud`mU6H9
z?Zm{y-r3n?e#?m$>Gj`_*es`Js_X(#^tCIkIf_jX`99ZIn~)3GIXNTI(b0KqY-})m
zIbM~$gFDN78c&WeF)*MnuIuf6mQ`5Tzqz%g_39Pnxc9}W9ft4oufnkwA5Bu$<{BmV
zczF%!#66=|hs!NAvT|~Ma@1N)RrL1t6-*orjf}i$4aQ@uh>D6L;N$0C)|Ho)<wn5G
z#xvc$Yr7>MNo|0bC^qV>dWeaMX<}*FEADmHR<6_FnTQfi{_zvd?ALXzYI`O~waj}e
zNgtV&T3}(&($c>E^7%6*tcXC3LfwV6;cNs#Nl6J-?mP@AVn1?IOiXNHZ7sB^NfOZn
zF%|Ra(+&6qMAP}n&MiVhU+>GaG}!-$dPuQupqo?hzIoSX0Y7i+>~zS!iBL$&=)c~`
zf?q;L&4-U4nf?k02*?j-JrCEp{*WZ<Mid+zygK%SYCH=PLJXrk0}~TE>`ea(_QF40
zlfgPksjA=_7vtf%_4DV?-NVCAf=n-yxK{eRyQMunMRV2iFgQ3kaylof>=+^TAy>St
zb+&q4>x^gA8OdS0mJ6fRwh!?cP*o*(dUhuL_;IL(&F>PX{2xE6Rl<ih`676z1#J;O
zv@0=;U=?`Idw+gt753-P8*49EV1zU@I;(5(p08fL!otRGfq==J#L!zG%ET!vFRvc|
zcYXys9}2=|u0hV$mfQVg``6_cR7ptM3-j|Y)|YRs?Q)*ZOivGdf5{k{vM39?M0a<t
zWbOO+cmgzt*RNlvrl+?Ad3x`9xVbIB#3(2!tz<1XhW^>P7}8ZcJn}ibxPB2D8v1&w
z++rO0Cf{>&rJp{<F)SwcxjZ66kWgOCvGMW}!J6gTZT>Spm7Txr^C-Qbps$=p!kZfQ
zVZRR_h+qf1b342kwszDFolj+D51xG$CcSs>9&7=y<pgN5KT=lyFk)0@HFX=-dn>H+
zTO=fb1a#v1gXwbU75(E=;yxmz6@^;b+782+zfETweUO`YzIG+PVX@|7Hrzd8UpyYg
z_a^Z@$>-1UtgWrT>(&cHaT07XUhp_Vz~{<S)cb-QKGN2~p|$MoaFy^A<stDejk<a%
z*lfA)KjF=OUE<;j3k&<MQ~MAV6?J#99jDIY1V1@BnVXwiQeFKvBrN0xVqvjiBVNhD
z#ieH^a=wGsx4UcW>fzBkIY}8##uXeIdK0pyVzOu`lrkh$(*>f1^DH(|bP^c5*S1)=
zxUyd!TFoyn2dU+$H+OYm!RPIsoW!4|mIyLFituf0lz0?Hi|69vVrFhGeTOg=E?!wL
z*UZxL^<wjl7Y$S5jbq7T9xa`n*#qK{k&%e~+0@pS7F$@wN#b6VnE2F95RTkV>#EOE
z#AzgFO`pQ$3p%b6_9csjqvPDxhnh#wbyq(${(D?mKhf{hc`m=492VHOu**V%gYAb8
z+}zy_k-y|~C3<*xsBdM3=O>8}5fQn8fx*DdO$gi9;+eLlCgRlRomU~mH$5XGntT*(
ztl&_M<C=0#2g~LIm6y++Ki^v&WQ02K3H9jh3a@V1)v@tugm1uBv&|I_d!yvz$2U#;
zlKd3zE}os8pOYmtem|OO^pWuN^ei$PWGLVe6VqAD&ENmf_!YUS78%HdzBMo~n09=7
zOi`?f3#k{fgTm3hZ@IZ)oH{kBPvbeedV6KhkGFWBP<r<HL8!>qn>TN6=;`Ufieup9
z#OLJXOn_3;#YceNr(U<t^`-tR1FLVAyW6TtZ_CXFQ7f%yTA*C1sz8=&Gc``!e0(3P
zIA?~3i76;3G;17LL`6lbeD@9y+h9-MJ>JsZre=;lp&RMx!GS52R#pAfX5YB{?*8@a
zdXL-3j~{XH8W&-yJfU%hB-~W%vQuQ*Pg}qtB0>erH3ZV#`goyUu5tWb4i3C}_akX(
zY1Dg4$-P#{tp&;$WZB}SSMRR2D&QFYO2W<UAL^9Xzx`z*qc}3yTUIwVjDCIQ@9OHB
zpHza)HaePcJNGUO4*a#_>HrEv!u><MMc5CfclP%7h=w63Ev>AeY==O$x3|Ym`a3i}
zE}!`%_U`(HdMvA&e|^1Jh4l=pUFbn9n+72@^~KE6$CklcJ~!_vCcT3lEzRxVFS(HY
zU0BIQ#yz)XWo74^{LqNxtGBnek!7@|Cb>(_9fYf!o89=355)zd$nrO@tE+2MdpkOm
zKK|9!Lb~<tLB!08^UVR6nYra~SRup?GMlO4hpj#0(<hRAy+&H7so#Z$#<(Ya=C@e9
z_;&y&4<1&v+#3!K4kHCRwF3?CRxMC`3_@8xUGG5#yY~cy4g4q=Hn}HB0bD=e@T3oM
z0NchsoMghFSXh9f2zj*OhR83lfl|EKb<gmD#TYKUF8}&9wR*wpP^f+upg2(ep_ixr
z16g~)0&IbLRKR+=39?aIUf%hyjXNB0;^_S%K!7|7K!zQJVsG1C5l(?nsReJa{SrUu
zm$;rE%6y7UA{hA8Sa8a@d-v`<l#$34MMOk|__M_a+s${4LRws0TtXrue^NH}R46uH
zXlTeQD-%JHhOCp22Lp3+bL1aEdCajFKX{VBXNo^rVHFNJG!4=n6gzEUq-^>&He7oq
zlHJk01<Ys7`&=t^UgyrmmkkX`A9*@jT3fZCMoo;1yRD+EYzRql^K@@1s1NRu1H%`p
z1Usw2!9iRE9}f?UV{&H{U8(cdq}`IXmR2!~N|x>L0oRJ>^_5rJ2Fa~k&E+&=?w^E&
zg?ZuJ0>j<`M}$tOgH==Y?BbuoE%9=3ab@M__aZ*IzBRS9WZ#&o<Qg6xHj++@77`TX
z_VKw!y?o*!#C%mdt$v4)8xIf9#MHDC2I+pWRbv^HwCeTlPWOmH6Uxi^4^B>2D!xA0
zh)+o9`hi?57L#QxICyv*jw$zXeTld{dK;4*<{G?$6z57!2^^fAO&DU{t*x#W(Tcej
z!3hvPY0{Lu=iG(jp(Q!oodM+8-hqK=HxG|eI6&BW!O^UzuC6Wtj!DF!viUIkmKUdc
z>_S3`JRO=)`BF)Ei{`z|z);c9FwM=)&42Ws(sDFcji`rp-M$YJDM`)Oi;Ii9lg>M{
zzhE0rRJdz+e0;pQI+&3!6N2A6IaxEY^|;}^FVEShAj16vJ9~S7O?J+`4;ISqj*%&Z
z3@Tb$T3vm8ab~ri|9EM|J>Nm`)0OpIkD6wWSo#f|GBnIazTa{`-VBAhhW^0=5{t2X
zzpSi#v=ZJIGU7%HRA;vF@Gg^+li#&U?>)4f@IBd?lTlM6<<PEZleoT+Nz+qYeV0Ng
z?Hb$%?}*j^V)y_5{(n6J_q%Ek`wsdM^!M&zz{=?A?*8B;RG0ewJC>1=5jh2g)E&aE
z8tN8Cl8YNRZ{D1!Fv6Uz3zxya@#2MQ>K#_o2%a~l#xk_Hv3N&=dGZ7hh+Q@8EySoD
z<R6`G$vQ)szqhkbdYTX$?mqk8z|tH;NJtp%6uD0_2=9{?=m=us$q7YJ@=!~sy%;mT
zMVorPAbPU-o3AchN%G*B=4Jb|zw|XGRIoCn4mb~N=5FmQ67Ry#wc=%6{jcH=reSMI
zOUuCIpwZEaZ_$1wTr)5**a|Ni>tHY3#P&{ps-zV1A?!(Fgfa72Q_#oG@U4UKU3&7t
z+AkwU4WVIyUp3L_o1+>k>xg12DoVb6OZPuKbWCLbvP<Aq_0(5dyzA29l+g1mA!1qE
z<*``ra%X>9PA2*tuA(m^j!jqL8U{=AT}O!zOliB9j55387c?v1&|AN*4h%dc7!Y{i
zZ}eQ_aE}JXgE^j8PxozXmDpOV<38HU-H*JC54rCbN7V$IByyeX1$eB^HQ?oHO*Ppp
zC?{xCwyk_3c`fXPij$?-5totO?U^{#{qtuU+qW*I5XB?HoWr+Ms1Enucebpl%<I-;
z*IDKnEMEu_c{PZR4u8u+L!=tX5KM&f4`f^6AkdYvZ|Dk!By)U8Ra7G$DSr@TbE5bz
zEc)uTfD!ITdEB5JIW<cWZ;Yca50HaAAt>15a&eb^?)aK<(^M%X)uvY6mS<32=c&r+
zdz#RG78dl5L%nL;qpjSGf=m{ZD<Sno54@j`LlYiY5>Dr-N&RBQAfL_@V`$a09VXbC
z^b=NlZo7B*xU4_}r!V$s_e4RD`gbWYnbKEZ7<*?!yy=?T_m6~$l*jI$hK@vT{ziu;
z&8E25EuYh0P3u7iwwL?u(&yXa9ybw1drdE2dtXpL6|mtd(|cvk>KTmRnd}f>yw}8o
z_2&<Ft`Y&V08>*j+hwf7!t;yT*_AF-7#bKsMSSzVijev`B_*SR<I$(vvG`uwj&GjV
zM2VVt@Oo@ivd)_n<+_!b+q9h$d2t~wJM&pev^*qHbkm(-R<*@Tf84k`xW8%3hYEz0
zm8n|3uDHP$m){hVSGhYr{m*QyLQItMVG9I@-M>8t!O>if(X(61w5zHFF`qMk(0Fz#
zp&}C0e*}M}8ecGJA|0_BD3a8p(Gf%VW>qK_XZwCGvu4|KA8X8;H?WG_Eqb9^S<F$K
z1<`~+gqz2;AB))8((&>R@whC%`u;sPUD)@h1DU>NKPS0}6UO1;5ekCx)vJVye(A2B
zp4RjlraDS~xfe~wj_aeCF$-L_Yln|5CG7C~tlgT$e12iLiu_(&{JCT{SwVxiBf^HG
z@|P>PydJI6Gv)&ktk-YCG>;VAG8nxX_@8QvgpM1c$=vmHz1A^$6x`TUCFM0ZGJ4DU
zZ|vYiJ1v5c`0`cmzs$?j)8$fb&EG<ja%!;_4h1@x)z#ImM~gz^>c6@v_xJawEsFh6
z79~6ph(HM+4A_MQ#~hclEY?<A^39zBB?y)#NrI#`fB$}}Mn+#EDSshrjvW?8REUo1
zH<JA(j5ptujGW?&{@J22Ar+<H^566yrdBySWN*~VDG~F_;ig4)`FlF$2=wt=s3@8&
z<1ZgSP81(tlHeC!WL}ouYt%+Y1<7w_7JAxST**dV6A~{w^H6eMy`mRl$x-bocD9vO
zW3#Z-Do-B}NQ?UEwxp%`ihlDkx3YxK->eyenw!vtPEgC^94*OJuatFmen>l0&sdCK
z*piZwo9R@1z=|Ez^^?V2n{b~I7Pfs_26OO-65aw`!6*VdyVFCSYDqdg4lA>@RsFe#
zKC%tCoxhsc5+t0IbK)2YvA8XMv({nz`04)m%rHMq0fFdO#2i)ZyoVaG8ta1oG7tYN
zJrT=7;eyPr=pPpU0u22=y`{k#6^-o+@5j&b)MZ-#ax4fq-`Jk5^Bd@FPT+RDbBOnT
zpvbMqaOH=kGJ1cqxKvyCib6zGwn!>bV){G`3H0$kn)G+h{<u2*)%*cxo(m571`EPI
zqHd>%Mvu|vI+l;zeKzVj@yn}=Vf8Cq8DZ7v7aftZ8}8-Ybs8FVSUAmXTGt4~CpFa~
z>mB;L%uMM|KVpy%7m}HKJZ$>ob34vlt2FM1zYZQV?#3X6*sBMqMz|KvTOo_%+3y;L
ziZ$pGH1dM`-mkZ|PEw!P%&;H16Dg~&ZYPOWy6ou6wlQZoWfzoi28~B<P2oA{Ja^ol
z)*8zsq)lb?md0A3!2X|pS2id1!Ze)uFr<U=)5YVVO`r`yjbX?I_5IVwIf3H%kDC6K
zuJYI_RC6p{*1C{ORMEW}aSpC;xc65v`Gr{{pFDnAtjBXfs_?y94VoKCB8)tj(|Z}@
z6FD5~uRh;CBPqPd_8>iW_}fJv+7RNrC0Tp6ZTs+c>u?7u-!(NG&sQFd>)4)w^a{Qt
z){O-LOtyTTI^Qp|1s+_1qVDf%1qyunv*^D)8>n|z$l07Oy8V2rEFkyut7;BLK0Y+z
z8h3J0XVHoREj6>0^GbhiZ(E6oD-X(bfw#fLn8${JN-mp?gk3YEd>lu7NfxykZ@(Px
zQi-HDkq5`VM`oSS3a_b)g+@aP(K;#zt1lXbO>t^Hd!1#8wvKjkkSk~FB0nV9Z0q?f
zo-K!y3wS+bx3Y{$`t<Hko2+%si~F)Y_+N{>E-6PG3(tQ|1xJ@iHY<IL`}i@Sv@=qF
zcKwKG>%GqAV2%cztXy4NVn#?>T4qqQQD=t3PQzSTQn%VOp0|CTh^Dl!L35ox=QX3d
zW)oPK!BMD)h@=PT0tK&I`rdEMSK6{5QlGEX+2AyWhOs;1BF}i0j~i?k1wWTxUQ=n?
zD>KCPe+`YDI!TSxKr5JwcbUWWxg9L_MPSiZpRo73b|<PhN6T-S-)x8)O}Jdn{T1b>
zPq9xYumeBQ36IKrJKp?h1ba1Zt#q<Mt=p0PgpiQYb-SeFmI#8cGfJwBtWv~!X#qn+
zC4<2S!{x1B?kn2qYNlIlh!;BoHlkGy8(+ro5Xqu-Q3IV(0}b*v1zR1<PrH8olC$j9
zE-Tmw3yvHkpPIwuu}QBMkMB?NFV3MTND_O0kDq1$(xHkxfxNPzf7_|EVqe$KzK#qJ
z7s+?*gSRANCZ|@%qvPX43N&Bqf7Zyjw>oBX2ch58#WPXi%w(;DHy`OczB#G4`Tpk<
zj~f^CoVRQAW?nzn$d5X$noxd!kclQz(b6&$=Fx?XsEhOr0=$*4SA#`t+8<C8j1l+i
zYmgt!)Z%<B6`U^L!OtrP3+?`{()a2V-xjJCN;o;W8?iP{brllnV8oL5UGljMnoXnd
zbJ?z_N60x&PR2-6P+hJ6%VBlxvo3RoP*P(Kv8N~cE%U?USUTmU$|p)sTYsXR^ZbdD
zV4Bpj<}4{K;fBc`IsZk4DAM1$H&U4wtgV4IKUj1NxIj!|j9)d|IAr!&s+rzQwt<`&
zG6r4ImDvm5I*+65?99t5QpRtDB#`7Mg=-X^K9@5wxfggnV<VB`@BgRZuJ0eTXOpf~
z3tJoMDY<H;ks@Q;Lie3G+|MxHMTLJf3-_VJ)VZ|rC|me#6cQ5`*X_j_tz@Z>h4^lM
zbbY?HvAQOutr0IW1Lc*kzdtJ3ndI!{wO<``s29Zr6e$B&JH?*@yRgwEVm_DGGa~MY
zvXeZ-kIejBnXhytSY5MVGEaI)%Nt=)Yqw04^m^tAzhXmxNX0^B7Ds9ZGevh_`GMkj
zo%<eja4aht;sX(JG&x<=la70CUeoJ$pwsd478lg4423?YI0#85Vq)w3@9*Nf6qLA*
zDj2`zzuYCxSzBFe9vi%Q;&1&)xbBHZF7x&0^O63P81icrC`425lcZ1j$j^#FU7S+!
ztJrPS-j&}vOeN%aX)oYBp~8$p;E}Q+s$qJ@-jxn6!nS{GbXXv|>NE4cvEcS}Rq$x-
zG1>Q4*i|1`&JsB=D>aqL^`vGML^73THP_WJO^enc5ZgyS0Y5T`vmz*iLrbruID6Xj
zZ~CNK2h82Q!mjc-z<37eRRB}^O;Eu)edbV_sZEk#QRQ^}aB&)@@WaEVk0B>^i%sH=
zmaBegGs5=6UuVpXu0Fq?s2ru+q8e%cZ6sgh*`v5k>5ZszJX4+ZoDo4mbl|OrN8*VU
zx%4~=HKv)Yroo71by(SLd<YedU`<s^r3dqAnH4_t8uDixqCHN-l9H&^w<pk`tUt6K
zrk1I7@|SaBW7k~a;_3Sta6XV`no=?$8t6&pPKeVVdo0Di_V%sWnl5qp*m-M?`ql^2
zI|yI0H2T%?22#c8I!q_LU#P=tQtW@^(;2;$S)X)wC)C<=5D^nS*?uOtV>D;b7gin=
zEKV;$`E`j|lI42UA#X)XG3cUZdgEH%6Z5R~%Y~$4TfYM|^%Rtz$2w8e#V3EphOU5C
zyp*fIf659wJNqur0@??nML2iZFfp|)8MR<)Y7UHw|LIJfUlDrx^eOrJ`us@q<-0J~
z&}ps)N4ahdfGFt<x{L54_sFPUW@cvoQwC8fjUz(npXg6T(zqUQ;tL^2NJw$&InY={
z{&UolqLnn3@j6@|L=zSkF5wK{y@4))5=|xxfVvf+Q}|^t2a5<C{keH~{L<3s>jyv3
zh57$a55*s`xda8t0k4;E21H{A_@}74Lf^Inc!RMH20bYTx&(dTF7E%^zyD+6>EJdS
zxP#>AcL<TaI<tXP6zEF-I^yBuYlYsHvAU4~fLwe!@e7$ar_)^n)jV}PsB6i^+#dig
z#O2`lL04kQu(sh<S_dVd-VgwtVz&bmYQO<JJUmFe3-8L~@VA#?<zGzN`HybiJIEFw
zQkMXNg9IUmd(cbvH|<aM8c*;#J8TBvsB75h>x(F}o}q+ZC98ccUk9?ujXaS9uYXU;
z-*T|h&vXW%k_ye}fq?;EXnCET>_kCZF;pDTsDE}g^QTXr^bHODp_}K^&-&!YgL9R~
z=O$Vi^URIkpob6dl?Gs)`@w@i$du6KOf4wT-R|k@?q)qP_AG=qYHM##1N!}4HHT<q
zbTlnLAJ@z4l40d%Z!df4vd`581;6=7Q+F&|_$D9F26pWVTxi-zy?8;Ws-`wr_ErVt
z4-Q`LhpT?fO3Cxn)ecBT3!nix*w$yzd>j1sm^k6{=f&-rTFoMTl&t4pf}vLx`@6<7
z>cZ+Q<M=r~<2V1QY6oWI*U4mOX9&u?H?-O*j3P!0^-^FI!Zu9J&Aa2+wOTH9CqyH7
zoMCeQZf^Dhx<{f;AYMk$cbaxYAbVN)@aE6t<bwYF`}de1+Gza=Fh<Y}>+R?W{wn4X
zhoT>XPs8dc)~L&cU}s?P2adSh6R!mXAmPRRH_-62Yze$c)N=+A5sw|=ur|c>cW0;n
z0Y87WH~jOTjv2H_dwY8$Zr*?nh!ON;ji5*N4Cohh>bQYV#(ntE0$nt0n##LDKO7(d
z@I83&2In^W{3Hl35}>69tlKbx=lF0<1;cly!RyP0!}<uzJwmJrkc*HTMZl$my|u6?
zjflW4=nuR}oDdtUR?Lr%iCJp$>obp<ni`UHLJ#dH(4zk5g?f$RR6K?^kgdJpu`v@2
z42<bU5hJ-BB{{3%!joUn={WAB^C^Oc8xI`10sP)8=70`-`>S&EcHnRcVPV*%<>kiA
zib?mKzc$D=E3wuU21u~5g#Jq4380{&DoscrzfDHgNzNvIy0o}Rj)3vt;!%>6L<|oM
z=))zOfXZNV^LHt7OrC#v_?`+AF+4tQ3i!Yfe!|YnThjIqdh-d;ia$6$HX)!D)dvh|
z(AU@3j^P8nVFwo%Q(z}=U=gw3yBBP1YFfIwYL*53&eO|F3nV)tKzg9%`VlLg%lOTk
zqM{;h3||^STi^D!wxWoL2<ZSlHMLlNi?Il3d`F3So<`SuoyW1<zh4AGgwl8Ku#N%R
zRM;;+IbIuPMt+S|JD85;Yw;B{aG_P+hdFX}6+~;=o@?Y9++JR82MxwgawAI+3p|6A
zZs-nZ;bVNfDJB8!n+TpU;DKWmR@$NBWL&y$5Sjou(vI%lyT@Cs*?aH#=QjBN36O?p
z!e9TkX}s(>%7OVo4iqv<L4slg6x;yS&leerJnDy6suK@?(e3njIW+mA<ikQ`zjMd8
z-gVEPL$k;?RF>fw+F%f5(EfY_6vqUXJaSAy_G7dOOOnZQ?ONsji3AJQzh;Qgw1NVB
z043y<lu!7hG}$o_pzwLz%jb3G_$*2A76=6p2pk-oUeH4LT>WE)5Uuh?q5^0*SC2zX
zvpF|Fx?zN_EFdf?U?>O#Br_?{I)&?SrM<v{=n19N`0)xIa96%&F)9dq#>U26&yP7k
zN#L8Kk}a*MxF1pQ;{G2+SP82nweO+RQ0%m!Zhbyo>q7AR_iwYg1~FLKC<uIf{MSCl
z#xR?26zXTf_qS(65WaqXezLWo$Pw6rt3p?NzML|2UJ_MZ?dHd`pA8FwO!FAX!2ITB
zgpk8uTPuh!f54K!4KZMv5WX21jFkkcg}O9s>iJEeCmN`-TVgn2PMv>v`ASul<l^%3
z7Aa}_78T4)E2Qp)ot-FnpF6Cq*fhe9^A+_$L2_wEq(S~E(T1X)r-b7LI;{}Y)fG^s
zK#F7J=O<-lWknz~G{%*Gk%5c@3H>A<BC2X@eN^7l)6?6!xP(aFnV*C%cL=0aS$X*k
zSaWj^I1!Njp4?D{L^i**g$D|P2Ue3L&p$tq0)~NuHxRbI&>B+ZxJEQGGJ^dKhkzg+
z>xv8D96cvzLEB#AY?dAlKK>&`Mf}PnPk0$5Qs{^XY9S(M$hZG4H5~+Y4QMP)Qmj2~
zp7|R&3kyqvxR>x`rH!8cXF${^wYj<L0>4g5&)w_)EHB$_PYb<n^rnGrsC|FAhsSOI
z%?)&PQo)id+JEvti;Ii7d3hHnp)I`d=TF<wKVh@d($bjd=%!ybW@hguCNx9E{}qOx
z<=q7>0f7Au5K>`3J1@BnMP2Ay2ngs2F%?XWPn8ud_w@8ELC%LGQG7z&waEkhCIs@v
zA<s0yW%Iafzh-|`h6Q-|zTH2B&ZrEQ{L7cOpbJA}1xynQ2S=vdd=&e_(+fNE!NKo|
ziFw#VfuSw@t%f}}lpi~&1(kmiUha5Lr5|Bu1=AH!QzHzVDO1NQ2U0UqFa;WiPGE)f
zB-nN^LJS+{w;}z(R@esNQB?s|5X1fZ@6LO=W@czWp;YB@!a2UO*dC5N3~o)9SBrz-
zib&(tR}mL{ke|t2#V#%``ok{Q)Z3fi)Tr|~?J~3FnE9=shzKJyGp3!L9eUjL@|Rc$
z%3xdy{{T#a#VCm@q<AJ%7g=RpBMS}q<Xwn%$QY=IEXA*3Fo=d{hik6;e{MqRDDpV5
zzPLP@YmZbL@`Ql#&&*^3cD%d0AOob@aR1Lw<xz4_5Fqs=tsw+(p+O+j(rxs9e~a<a
z9ay;_Rq+GKPDdn-_U4!2N~E$PNyx!=rH@8dPHq9p9@pzDF_2+ZodHq`giPvrx|`PZ
zR2w+j?tC+vsOv5QL3rzyA0#{E%8;D>(gW~^$6}0~TFjjUipbTOS^*%JZC*-E9UU0J
zYFh!9+HH(+AcY}tN%Ale)hcGNf<vI+0J5dMKR=0K9bsNyopT{C>yI!#)k9ps3`l7T
zK*0A<q`)O&0DI(6izcg(c-%TPL?q^ZC<*1R-R`^;63LGJ&_>F%6q5u|5XgG0FG(l>
z5<4=j#B=H_Ug3RXIO=Q)y!h78kPDh8C>0p&9^W%+*BfmmH-gg17zzU;Y;0_zgJf~9
z&!a!I<5}3)_<+?Rbx`o;ozLZOX)Jzvpa;13V0Tv^;2!9RN+5p7Jk*q~L_zLh05ZZk
zPa$R_H;sb%ydIA5nxuoLN-5&Tu;m#W7AHWmHhKH@XB?Ma-}9*Oa3f^82GrTE+ov_b
zl!-_HY0)^dfI$57MBX3@a&qn%g|@o-dQea$v<U*r2FZstq{jDe&CH5Gm{tPs{P@uA
zbeA4^IY0r`3SK8(0)~~dj`u`vBM_fPAf0)prIiTcky2p1rO+|x&}{Y!<>>&P%J$O7
z;9SK37UCXc;ol=UDj>9?P*GKV^XSo|d;s5@@Fx*G9l->2eYxsCx}laL4sXwgz}kda
zWGc`QM={d7@KR7fFS8iuL`Z^;;LYos^rUEC*XaWZlOfgof9*S|GSn*m0RcE=kir>R
zSk^kpWEIe*p<en16%w_mD<K3d+rVvNV#d~PtbhODjDY>KyBix|W|jRih;a!C5mr`K
znp#>vi+H8Iy-6;<Ndxl=h>s_emXV1`PL709EigE^1*8}D_GQ|{%G76c(yZ+A^73XD
z7L4@tC=ezCV`IS?85#LK41coa;iK~}GOl2=@w~c7j6A*>Lv+bSPx={lLK0F^kbx@5
zp9d;D29aM0r+@U$EeeX82yTA<SUaXbRa}w_P>J(7=7u0)>VKUc{}10ENsXaL@5g@m
z5)L^Zw!Z-J&cXyZ9U%`J`udV{DtC<i{jbT|$&m?-&~XL-gGA<?c-}_okHm=2>*n9L
z*0G@y2Ehj~JXUV(E4`7urB1tl9UPh><FYhQ+3x3kf1#=R0>smqS&P6Gl4@MhL4^pi
ziE_9%QgZZbu2oolRo)v9&E3aBRDMjBaE5$B;jp41S9=(o`<b#pkDv+aD-3>qeyw;p
z*89Q+r#S9c7ec%a_6Al;mE5PJno}YSe9UB^dx>{q`G%S?e~s=1r<pJYru!^xJvAiC
z9ne-xyRR($<VjfmQ~LjW<~yEanZm0oCk{ejR-Vrk&kTCld%OQR3^Yj9)0MD`6k~8E
ziC)RX9r43PRU;klNrvrB`w!qDa5#PDF-bf|ng=;+zCpv;cNoQV#xgHom`2e^rBrad
zm}(&}$a-X^F-Mbk`zqjD+YPOF49N-`)<C-Zr*0h?28>ANmAxa{5}4>vY!fb_-kZX>
zQR1E8W7^+T<>#fpv+2)YOE9l6|2!f^oXQ}IZfRn6hW7^tAD>EA5A%gob3dPS43;4p
z8ghJG!qvYojUog%^%k>M{{A(*9Fj9r9!BB3cV;%8%GpOP#?JGoshZF1<rD?HwtC?4
zv@SI4LxS!EOZ*QD_4MNKdfBOFnyG<~Ry@VnqZe}p!Y0p81HK&d2;GGKXGlo7&~z_b
z1~5Y`FKTw4>oXs+)}9Jh^F+0lPcK`eW0GHP)=1R2U~a^5ww<n(@>ox@3)YT4T4v4q
zK9x#;HB(1pRqMQRwtYP3Yf?`AGUEg1t1-Lg!TMV4h<NC<#fZ<mEDb~XTr0G&xb~4~
z*ha55jPp6A?M9_*FkMM$Y4w`dWJMFSv-G$B-285891P6QZp!SNk_*fNwk4l*H!8YB
zlo)|q7-nTa>V4HelVe;ufzYTmSodcVBq9%-4G1wv-;tbda3)0%9^9kMI6IaN$ogeB
zKg932`J3vg$yu!X@7hZN{joNQ0qBNP<5&5SPPrfniGY`-4_%v&YZOd|{@g#dBBy-6
zE1Fjh*+z<V;_uXO{u++xc~5v;VISh+7ESrc6!(Z0`*5kMjt-R1HA!b%n3<y>`cedg
zt3NU~nXY0{->eH&6u(6B(i=K>(W{N_3ld0Xhp*d!U-;-bw?P-8pkX4Rl6<nBar5hI
z+uePqXF5eUFkP(MC)*F0Is{8h$$+XM$)7h)ykY<E9)zz}kNVR7(#o`jbT^SsXFs1t
zUvh9>&MkZU66o5ZK-@mNGNnh{;bTPGxYh~8JCCv*zy6Ytxa2*6w>eXDvwYo6KkE8X
z0=&{s;g%iJo2S-Unypl9_wR%9C=ywn;?hLw26$CJ#9aA|@U0cFxpO+Jb0auV<~|Yh
z0F_eS<1j}pZ5A`GPS<i({Ll_Mp7BVN>7PaTzVvzC{>!(go%Tq+0XL*Lkphxb_$C_S
ziY<QOq(E$7t5M*cxJ=&C;Mb{%I(rIp4_CX(N$0J|xGj-o4xbXbibB1Bc@LL`wHfkS
z0V`jF%<fZ<Sq8{4nyQtu)3xO6$qONAOZV*Q8os<zZ5M1zL^mH87zvzcN^RRx=&DSe
zU!f2n;ChzWNd7h_XQIZ0iG+k?;G2^+LQ=KuDdZtT$<8^ZzmT#6-^<Lppz4Yf1Z0Hg
zwK-<Lvd+w<zq<5js}Gl=DZ~g;B=8`L80lyEZt?csU~)`y@OUA^{<`Zi?pLp~$Il8E
z){&j^@Ye-p8ZC96w84t?=L%&)wWA%!M;3P{^DVD^C?CnGk(#elDxB^nQ)L9_eQB%A
z{5rFK=g0wjw%#iMTFnT=6s4l_sH}okI$X9fDQUthRdht}s|svn`_*qtdk|W--@N;3
zA9^13c?}5gen{ZD?$V}26%yiB2;Piesfhg)8}?P==)m+H9`1tAsjT1If7wP2L0Y5c
zRb-CD&@W-4J$fp$kjDdA-TtYY%3Wj9*qwra=%I9n3(<QZ3ir#nZYrtfi%E_P&Q%{H
z43mB5c0ieRTaC#(f|Uvl!Jey4e|n`8YDYU;&9V1NUV~uIu9!%EA}xx}t}!n+C&Qjp
zWg55YwV6&uB>yhp4k*9S*9XQ%f916f5k+m!#w6?v&P)a2b-w+bMyxE6t$&hr6r|XH
zO^fUTPdAd@<=MLWaYwbhjMe#qOG~w~rB$TzZbkE@wm@2L7JloK*zm6!QvW$ID1$L*
z2*#;8%W6_eSC_W-;g68q&%$S?kxf83ZCy|BkvwZSgrLail9GD19t|;2tt5ejv@LIa
z7#G_es3nTxzr+=)iz(gpJoSCw6VJf9$mDGOI;06_<u1HU$~%rc4SfJ!?{IyZIWQ5>
zF%G6Ium*vNNP$j`ufWTgDjzni>%L^aPJ(R|ggZWUm`rWO(m>yqE9n(A+uvUu;R6GP
z<D<nU9EVgqlsFgnj1~@ATx~_LZ1m(oEMSa8H}cLqHxP$Z;@BWeXtKV>gU|B&m8UU(
z5t^~eXZdXegvyO5t|!NMAZQ4Q?@w7+%pukk^1%mv1lwMZQ|$BHd6RkLx~TVZ_R8ra
zJ$Geakz9yor>6Q^Ut$^zhh-{UUQjSHKW~mM_AzX23?Hn~<-pyZq}AyA>SssF2Oz_U
znEhxj7*=nFS3u1~tsC(prMSA>?A%P!tX{ncoiAhXx;s{b3z?S}lgt6ESZ=#9iHWU7
z6f04S)=GI(7*Nb;m+m*ewl8g+3d0-a&(h^+$Reml$ZRq)QrX5Z5f5y}QDi!<Lgo0f
zQa`$Q)d7msIXmXE^SVwDMmrd{l-->XAs?w0n{*xNsLSzb9_ov%O&YG1V|(^ljNpp~
zhYVH!<V2-0u?dGZdZl}^vW*5>#uLqY2MPb3Ik|HTCmye9v$CC8s({Wbt;M<})81y7
z%Ym++7&KH*8lOpjW70pI+0j#k)|H}Oby%{Dd>51f_!Wh^ktRoCAYWFqH<iYglK)R-
z^k4PSfB*g4>t}~F@0_0A?yiZTW7usGDzkY(07qWH*n~MRK<`6#CP<F5(x2YhM88}S
zak4XOa53+V+-8r@J|~E{NvB=*Q40mCZ}Tr$vUlw0IuY7#N3=P1kjGu~;eghrwOy9(
zQlw1k%_&Z(_~X!<rw$4}Rf!spzUJOb;58=~_u0F_uC&tuvaF`POOI}ExtlDQP)7D{
z{jZ?eTu2;O^uKZD{7q`?Er)A$!VZG%kYh1-z@&L*6Ln0FP8~ZBqO)J8z~~ruo|W%v
zSCP~z>IIgBVPiPG<=V6Td{YKvUe4_P!%c~ZSdrt!zV{Xh$MVfPOrBm`bkg4H@R`q+
z!Nvet-fZAoqO6AznMb~M387Eap&Uo)lU?VMcF-Nn)%|bCxNK{PW}T8U==gVoH9v@G
zq#GA9UW^W5Lpb{VFv8oMuELj`9lWY2&*OI8bVX7vb?0Lwy;+=XpKr;jW7ZFnz-b)~
zU*R!Yvlkr85Blshy*ECMF6gw1fM2ogF&`)sN*1k(0uDD<rx<u~A9;mh9{c1e;`dtn
zer7h6{mP4n|DZVM!brE6LrD<w#zpeVtjayxfV`ZSF%ki_Wn5|;1l!ZBwwMF%jc!ZL
zU$qgwt6dtWgk)?WjFKUa<ScM?6)Eg*ovKVt%RxWq8tbgTvWW&wbd|;nSUjs?M|LNF
z-z$$Qykj4j5qg~HPIZaY#SD;!<2*Ys9|^4Oy!4GRDze&oGF?N7I4~bYL4aOD9$E`A
zMr<r>@wssXcEtmu<MC$0q#KOloX3r8b_j%+U9<SVVez(hX?Z0j!y0JfKPWHXTWGY~
zU!$U=zxQhY?VG@h2oOIcK2cH%cyaeb6@iy{57)LH7a>+^o{jJ6^|l@XBKBGCMKonk
zxWeO`P-Y`Z5D`+JmiZ$iylh1Ejfsi$eUc~7R9-|+&)gvN^|HJ9hKfJ{)@MyuvBj8O
z=cE{BW`Ud%AmeMA>XrZCS+9WWtQDFyA6Gie;<PkbItvd>>j<zm-0c7X=p%bWhQmXq
zl;|TKmRrI#AAD#C_xV7UV%$9`zVldT!60o~rS_ZJShhRhS6jYaY<B{Rj`Eo*`<sdd
zZc4a(5BUR<8&LdF3vJ7@JkQ?ApKRN@oaMjl!K3^5are)H*TkqT2I2;~t`v0FsV%Oq
zAHShkTYKkD`f}arp6qSpU<?foBl+hx`pkVNthi%UB?a5rRvl3LJTkn%47?%XpZ?;V
zx%wEwH#s0UU1{p?M4h=%9i>2fj?DSqX|vZe?J``+*|jVvbdWWVYVmjMCDts>tA6%X
zD$3?sH#US@Wx_GYAH0no8Lqc5b@g?3Y=$lk)ahWZQe9oM7!&Nw2EVv3jW<3d7W{!w
zD;4g&0IDFt8W%EmT=LfiJ!=yK0q2)LTh9gU^yyd=Rhz3lqEP;bF}na})qLo3FRm{c
zUnM+Ov%#1-cWT1^;E0Zg@@-BEkJY9EQvDD=xvg2df{=Z3?IZf9j-QI#V^Q$XW)1@y
z?Pk-tc=vuKa=q*UiaMBUWa|EjA%}3)`=arV)?Ac?7zZB5jD_K#yAM*s_L0*2TidB?
znWIbdtb>NWrMdccPsaJ)#~r!xy^p-v-|{1X^E2?!OO0h-5ZdU<8+O;VOWZxY$$|q%
z$4}uyqFF`hr7rd{r~YL05+cHj*Wh7hWnkm`xPAK{PB%2t)J^U|od?7;r|JHwj}$mn
zB&{DBriVA&OICfxGgwI9^{ek(;NBmqm%G>gVBd^*V#tHq|5Zv^x-P$m-uNyEy8Kr9
zniSX{EqBKn>)_`!t>z}{se-TIq2<23r8L7xxl`J=T;)|OlCJ(Jp5Vj&oYPO_1ZXr?
zxU1u2{{>K0^+_%oi@+l$>T^ZSrPly%A&k|O8bczi8%V2bQIX`)`Y6scHnRHpuf~SM
z|6mXJ))F2W`N+|+EWhW@w-%UD(peBR+a0Z|0U=fhk~&$ZC#BU%c%-m4aO?l&O!faM
znko*BcmnV1OBbzpdO!b^iw}w^;_rWb5m*8}KQw#&02$%aioQ$aGrcX;AB8((b9D4&
z{DfT8m0Q?(^Bp)%jKKN53Z5~b0LMB;PpIqxvjj-qi9Ri^sEAwr2u&4()?g_O4KirO
zK?@iSjGjzq&<agMS`)zL+t4815k(gT4&a?vm~TJ*O<kDtxo!s!Ku>w{kaDyWO5mrE
z|3*<?e~FCjBhYdNA_Ot*(^b--1`Z(T3J(AWRn1imwa`#k=d@3TNW%(h;mvFA>%(8^
zPg&7ryRXnOn80IXyV!PfbE=Zy^XJb<ODQzzqreJ<f`C;!P-6TP$ju_Oi{Z|tpqGU7
z@dBavr1f=<75dB2U57s_;mo}ANCCYPjL8-7wOtN<MYN_%Kn%Pc<atOI2tM2K{*jSS
z99kuT=k-Ny<VKsj3$4Ase<vbal-AaLKwvvcN=mBM5EJ04(%cyPkq8X%;Nak+c>`bp
z9<Uf9{R-fu6Y)4^M{5Fylj^FwcfBq`6U_18AqC}iDbPcnv=?2tczC}7QQnO^0$j>T
zO1=Wvw0CYU<*m87{Us14O=x0Iz%3!0kHlCvpy3Gq`%bWFCIXrm9vn27ot<T8VnWhd
zQekU@{Z7s&&}BomDS@jlO+rhY9O1mwfd>o>8t-ploHn7g{5k~|9<qHK7ZdaH&&11@
zFZ011{7gmVBa)kfd&72$$48{nYAOzi(I_Y>d1+{9{%&maKr58O$nAV^eB&qiQpFsY
zB9XU=q<dgdpLRuJ$_Z#jgVUP==_~;&8qr1~&=2tN_knLL0WKPRG?Y^?-7y;n9I(*@
zojC0>Gcgsz6A8G1nn3rj72<*1s2$S>EI4RQfSs8B-o(Pvr=+B`1eC@uDw>33e=s1$
zK;7+@5c*&avV^cj{sheKae{ildP*xRO<`y@K~aDNeK32~6Z1+>o*05H$r2n&NKlHz
zpPQSOw`pj8sj8|b00_XjdD8&swxPSb`!v<SI4>`+CN#8R2B`&`3=Ivz0rDuVxiB}k
z8*E4&z-BtZqR}<m;MD^s;RygAS}?yJ;9=GVGb_?ajQrY|uI2;aPobgr5@mB`r4#v4
zF){xE?=TalF)=aIhd5-M+z56ao|lFVzy^@m6`<u8UgIZz0#Wf?db|j*=X(Tk3V`<v
zNKO%0R*Ru2Uknb^61WOJHa4~bbvin_thBUdASW$Ih6-+5B$<Y;I1NPH>8UHyc5Bj`
z(EI~B@Xdwazjv*V{pbOMR0lLqf5RffxpAGZRf=@kh<Ken0E}h|F6n1?iIC9>S3?Dk
z83%iN!%CYu6PW8w5Go*TEkOnSm5}mDXUCiS$&H@IvBwCSo93U+n?{gQ*1!Z<0%;%u
zqp+0`j2+Rzds?v?)q}4>4#7yg43;e0VX&#GS?-cm-~-Vg*Yn|ln~D1Z>Ep-i>JK?N
zImJmyNz$E<pFLx@mDSbUWn*QvxA6xF!^yS|hA$Z8atG`h@vGCpuKNM7XFQT#e+5KQ
zU0Yj<97<qRu&$?yjWCI^gV6ux!E*)9fdYUw0*_iSn*Z&Ptl=5lJO#oT*pPHQ($kDO
z#mdMS0DPmA&$Q2WLm#9Q$flY2S79eZof^jlU7z|L?zqZJL=iw9g)*P&=R?E8$P7^H
zeOV)$0{-?=;LtoB<QCvz1Gke++EF~IaHo<257N*B>B<10^lNPhQuw==&!4e@3nP5<
z@+@-xLj3uGr?%~%lZ(rVO)k&fyEh<S!9V|68*VlF!v}u=`>Hw0%U?Rs5nu~{*KWA&
z^graHX+M7G<l2F_pzm^fb@d%s<61{XNIXuq<p3!)R?mJwB3gLPhi^?yjjV^eyD)gd
zs>k8U0e~A<bg6)z3JO-W?FC8G!y`7rldiz^`YkIf-EOIaXbYB}t+O)*0+&XZ1ki62
z;Q9g3s9ao~?+~S^#iE1sI4lOP74{~iCm%fvkSemJk?TF|DN$QE{uEjaNH;G4NxS1s
zEp~Wl1X#Wwfg%Q;c+lu31X!l1tgH=evj_y}NstCkZ~?b5Wg+S6g9m@krfZy@8YW%j
zdGDWlasWsL=|UT*gaD-O<85ycte`_j#=+s~YXTtTp#t+QaRdmUrjk(qFl9jwN4VL9
z!Lk!{+rJCOW=!a_;{s8GHOOxFZz#P-=;ptu^mGPDWp^jR2TgeUb};zBNKa!vvYj<5
zJ^p`+rG9(_Tb~QlZ3OZW8z4rjvRyzOXJ%)|g={>n_Xy@43$Qih!0{DWYe?<Vj~`ZP
zJwQpDL1mCuSg1EYm-#?r&U3>Z2#NGvB6?QV+@}5K&z@Czg9#hd3KI=pB5|BLVMw)*
zyH<M|2*a<8;Atdqr>Wq%M0%{p#yWs(4d$w`K~$mBh&TrVtw)9_=#Gp);vx6nbJQaJ
zW>n;(L14P#aywE}@6@>;VMApwFgWN3=@a>3z)|&2cNc&^2mGHnDpFDaVFDIa#{2iN
zcIFyGVC&GBgW3d!7y};K0oKaf?r%3hb}&%mWPy}CZB13~9vlR~(-&%-HZTwaBO^#6
z%HzDLiIkzq$e<z3(eQ+X^^qJjgxS!yo8V;~1c6C!iUb`L0Ig7qyodV5PQG6c9`S&A
zn_bfvT)KuG5ft1HA3l2X<}R{00$ZHQy1)}~Yu>nV1L>^(FZ20Fb8)Z}M^Fji8+Ak=
z&q-kPjZR96fM1Y;PloYL5O!jPC#~EB8#L%J=Aq=8KIUL^Y*Xs;xy|t^&|#&oxvecw
zWNs36JEX`BY6CgYCcVg3q+!=8xsjEX^&d<PfrhkV4-E~?|NV=6Du{w7?Tb&a(vYn<
zc=82dc>BAsFlmThyZt|pA@*^Qf--nI#z3Qw4-$6IHNI>}PfPPdD&fGf92XaNdcM<G
z41VeDq)VX72FAwd;0;D5i_hnUa-ge#U7@I?L>d+$7+Aw0OFT)k)`F)>Kv5C|)k+f(
zG+21x2)`qs0~Z0o5CU^BJu@>{YN~X&JwwH3(tE$e8xj|C$Uvom6s7<bM_v)YE_l3=
z%=#9$Q3ozEKETo*`(29?l4;u~78P#LElf;Iq-vF#ARlStlKbpFiQ&+<$F`1+fk5&=
zVNncON|__$F6!~}Tu7j*B&fYm5!+@lgvz4eRr&uKHT|#qB~^wK7u4Q3Wt|-z42+C2
z7^Jt9?Vw|Z%>o$%G_CURNSpCE2v10=jO^^sf@5#(AAp9D^aDI?229=sy7k@VI`!_P
z@DvwlnDFLDM5oZ9OXNfYw}yGO13T{<V`F+YHXJaF%7f3Ho+$K<4lTU|Keo5B0kr6(
z?vpS8LIvFc=;z__JaPXWLWOO15d^W0WlHZpA&WYY1z^Af0{kQ;|3gp#h7#_J!ADUF
zSodBkVzDIp@^rk>xj`QoO^3AV{SS?(18QQtZU1FVcW{97FFmx10w6KpK=lLMxE$fH
zhXMHa7CHGlc@R!6Ws_*dg9b)M2QSju8J@-w2<r<Dz{`Q0D;Lf1i^bkc_u8_*HMex!
z2Det%&z_zdcIoIZ9|KN`cXoOIusDu>;wkcxz`oWaLBA$&hJ(h9SvOzjde77Xsq37s
z{*~d*H`J3A=(^=e@azd3_^@siud4F^G57r&w;xuJoc+npEJ(YFL!~Y1d7(L$8Er4!
ze2oTTJb2oT(g5hsomc`%o<a+VPQqK-*|`FF55E4_m~h5NN4vwd<Z#Be(l2$6arEX@
zYo#kO=}my8(*ATm%Oc^~K441xNLP~I!FdeaWtl*0EFKn0Goog3lT5Dp`s3+;5mf98
zXdz!0*lo2@R@v^mQm8$C!nkUV2zK1^t%=L}uGNnCu`$|mk>f-E@b8cOr)YO-1xOh@
zNG>F9MFqdNMcQNNj8sZ(DB1Y<N-e|LP>i0f52FI<SbqI}b8DvLCYo@WAGDR?&TKjd
z^D5^n^;wj~z1#k(^%%dJikcPZ>JRb1W@aD>z4MSe`ZOj!zEiI?L|=Btn#gNKG1*rf
zoD=P!*XQ82XrY5<F6z2gqQQ$5XR}R10tETmFbyE%ZKnp{jA_N;V&>Z9M$=8=r%C&v
zU!WcgKYB`*_V_twRl;fp2$cjK4!H#`$I2yvSQ5Hku?xF-H0PAiBD+;ZivO}r6={SK
zDGz@kyTs7wMM0cbT1jox6o3&a1I$aClV!Swv(OYvZMC6AzUGi|hOW3r;@rL5@^h((
z*g~O9N7w-yjEEtn#pFl%z1&0DA_5#h{D3?xfoG}eGiAEP@SN?&D}BFZunMM|kfKE#
zu0{T73q;-cm+4|N2L?1BWEVT3+>&&9-7OtpPSx0t5Q?W7`m0qpU(bPIDD`~^6?mZF
zX1&sEutA%D>hX;W7_fikM~-9bqxgiY%{Agft8r{V%s|Sp8;cqDgqE-82!77_jB)Dv
zVjRbrE$b2^S?<*M%LD;qqLXo^1%flwajrjI$vUe;l@b*xf|#&t4h3hwM~9w29!&FH
zac$R2R#Y;@<fdF2k3Alin?{(Y`mK#mxy|=g5lv^A;kKUFv`9z_sLF1o!wEX}Sm44O
z$z5rFU+OycAR^@<PDhr>z`zh|Nd-723@#3T7unecDtrrEIzQ2GultQOS%N{9;qK`z
zFvR&nSKNDN5-mkVnLA(Jv}*hs??Am!-f@KOg3nDlW{GUM`}Z*U@-R(6M?~v%>_&8K
zVrbKWY4uv^y+|2X&HJ|t4nigd2+0&_JLfsK>`uG!kS0-i7aODXRRi+NlLxMqW}$gx
zeMZX{)$N=Iy@N7c-MzBc3mG_H3(^uRzapJV5*VjVC;gAjt~@-Z8=Y`U??nV<Rp3uL
zZ8RUv)wv5jdUipDCV%aRbaDMX<@v5Qj`nvnml4<>?)o%7|Ld8el}%%S)&z|_jL1G=
zuU2s|)e?AAzvtu`wO%nolNk-o(d-2QjmL>pXR_CmdY#AZkD9NEtL={;yng<^YCL)N
zees)N^_&%f^6z1G-h4P^K(eJR@3=ADGOGSx3OUnwDAzuY-x{(nagf7Uj?}3Xibi%K
zWJ{xxeaVu2A1V93q);epmLp^*YsN$=$x;oDWMs(}<Je=E^PlHDFV2hS^LbwT823GM
z&*gtzzwh^#S!X9_Li|jJR4C$Jf9QCBIJ)$0@7oTb0&L;My~?f%A~tr8$-C;9YTV+{
z_z4ji+cOL2oK5ZAUmj%Fn!4uRGz<|LVOyZC`J6rWY*cKt9qC#qWxRpWP-rX`sHR?v
z7H%jnGO2r^4u>b1G9gKyzo3wzQ<ibxmlSNaquyC$f!m5qxUahkjW?iYMU<l3ZN~%1
zif%_Lt(5M=`eU<B3K@BEnru)<jT=Oex-#Nl69HV6R&AH3&xl*cTrjXi*<I+gbNx~q
z#IO)Z>>f}(>SZ6R${J1W36+Q-l^gT;En>J3X`AW8Zsx7rjg<#G+RXg3<(oJK1m<Z$
zg^9^kV|2%KGpjUstVuZvsGyx41cBUE-@5hy&}4$w4XuRB%3K&W*;f|<MTu{^9cbG_
zc|GRJ%S*iS!PC4`ccC2)>AjwEA^>qf42d8#iv$km+2*!eyOe<Qf~WMMbDghGy36eV
z!_#Br(md6vjdCV^w6YMKLi;S8KmX_D;kSx2E2}pSOS&(KUAZ#($FP@wn#W8#*N%MC
zTk+9ZqdZYKNjl}ys#EXe;Hai&K8sh#CA2!6jGu?cB!N9dTE`!smC>L3)YkW%lumOK
zkpQ6#{Nw7}i*~%VOU3%ZDO3wsTGW+&yv!>bhgpP|NF|@_YgiP;VEC&bQi&ja95VcE
zof!}3xtbu4+PD}usI|ga-o78<A0Q_3;6(W|otKgf3|jHS2bh}lwyvno=7?Cm^c7jZ
zmvm%@86VX4wEjJ%)_P&K9|@aJ%pee6UoEh{b=v<?8qdQf%l5ld+h*`H*ywe)v(mw}
zQJVC$t{N;Tv^$zLoKOkYf3u4tUs77ghym3s6iY{MQ{2{DE4zE}=f+@cD{t6~T;cKV
z_B3Z!U{{E3;2^(!_wG1xwan}W5Da49Ke$5F{%BjK%*~>#hNZBiiZl3vI0$)HZ77uT
zOi9U#l^^r=H;YPq*OXiCnA-`#+H%+MXV=RVgrPC9xI)?Epw;Pm4nb{gsJapTwzqr0
ze>!i!y1Jg0ZE~+EQ_7O7<ZkUL(x?rKn)-oxNHm&i$fTCzGE3tXS<>ZZ1`8N9R=tx^
z%FO8_2e%?UfU4$jB$n|jR>AOKS<;XFg~F@m&MBpI+e<`y^M@vch0ilFf>Ih6ik>vq
z>hg}UwIyxU)aV#qdgvm2Rn8x(3Bqq<kA7QOM$?|C5>64D?j1S#wmaRp=kKWOyK7Qz
z+c=pQRp$4uf9wAX)@&hAw?HjLka{-~Gko45_SzF#roSJ6pqSd@z-=%8*)KY^mSsOj
z84V_!H>G3zz{XF=$kQ0N>|yz20+|yk1#CKNC6Lv=(#C}mfwUOTH_cPphtNM+?k|5O
z&=%>LAGZ*-{_qQ48<V%Ut1M#PuJDi4*q*B7#XVsRGj_1o&pglnWw_!E07YKAqgUI4
z(b0F*F)~Z{jZN#ERgnDtfkRCKUJT{@A3DodX5nrap_?pLrCmG`P1UPV@QZ=<93R6;
zfSi-5<=3$Dq4%^G4K15LDQT5j6tmDQI>2f`n9c}#Tr^oYQH9Xl{s*f2T?O;l!QywG
z+u5lR=W<#v$&t4Oe7&YtC1ibS)w^SIuSoe$7ac!*X48oFu9*#&M-wc+O!YLMm9Dhg
zyyq%bYA+=M-+3WWhdant%PY6O2tl47=FN*m!T?*l=0S>tH>Uuq8z_oSN8cFch{?pH
zzu@>@VRb)j&1tZ*UC~f@jFY5#On2S8d^Y~IVB_S+dF<SZx$?RziG{GW?DNU#l=(fl
zG`n!p&3eD8_>`Bepbg%UKGWB#UmsI0_=l8L^3vk9!adIos8!Hn6IN{R+E@e`b%?B(
zL}Q8mEKxKg`bU#8zg@;&*CtUTzP^az;d38}z{YaqS2&^JX=3a=rk}oT(|Dxk#0k^J
zEw0ow?!jVR6krOh4a$NxtE#Me**I8u1J@K-KDZFWVjG%)IN%lenzP2()7#tiS>Wuq
z@p@heu6kO8G!oXqYWK)KpDcha8Hjy_7yZBcAk{8SQn+bj(X<|0Xn@o!jlqFIvq@HL
zl!+_^yJOM+w!Fou=LW5*fIU2#YxJE`kPWDz4Lz1w`!U*ENAQobQgV|YhbC)5vjt2&
zwkCPG5a@OrR5?1ZcRy*UcKDFsW6rP9Bgmk79H)qRuJ!5MQhwRy(iT2tC_W1(A$+s&
zB%kjmn8(6?VyO}2ruY$@7}vD`j>BWk%9AAg>`rgi8D>~8X>AiVGjqCdA%tftsn*Y2
zhAS91w>^ty2UA%RBk~@W!xEUS96-<hL^jRY^M~DizzFzu!%{q<N+3P(!G{ct^_N;l
z28GRa_10x&W9U=>qDw>fh5*1`C<G$EBoKrr`vrV2ch{->aFMqkswL705>E2i<Qs3#
z0WJ_<n|${caKn0WlCZP#+4#qkmh?9NemCp7RBTS#*-dLU&Ea;Gv@hkbW?NvNf#<(H
zX`5aS{O$a9OMkYx5x%X6{ZLdrEint;rn#h3qp+M^hDHVc1tVb<wSa4UV~d!H9U!lw
zQfiom?;fJS#%ySl1Ju$ToVy@+r5^9~dx-xi(HZhHsZ~vkNl3>N%iY9al`3dX*lHc=
z;DS4R1)=<7fyeUI!K^#`;|FLMf;;=MBNAPJ)R(BCTc9{A@Gmxm1`s7^h;2nlR)X2$
zohyPs0m2FZRr45nG1Wt`$*Y9ZNYP*7S%Rf?y*UrltqSvrpMs3N83O`sK>0mDw5t7=
zE>Y@Sf@OR|1fd4w<EM{3cP%tR=xM)P@9&qA{v-hurURJW6Z3T9?n+|`%4L-|y3wv%
zLXti$k$MKpbSqFjL-QH7Mfs7*smc73JRCV#IT&@*!Ew^wC$-Bsl(M_U;NhK}oyo*=
zwxv*_G!u4{&9yK3@0i#JnmUJfpGz;2EeT@b0Zzm&61WEDXpNB}uUCM@t$3tt{r(!J
z9a^m2xF3li0dm>-`Oo6A=&t`msX@a-Tzm+7NtN*K_HVNE24+;NvJJ6mCv}2R&!gmA
zVK;P+m(<+F(nvh~3FsJLhGEL2q4uKV#e<ukq=@{ElcspG^l-P+TQ7>79n+8O>~9m3
z3GS}f#ZJlbF12jXBhOr&3ohIr&s5!za&~crsR#}IGqKoHnE2T}=+V&xLq47}oViqo
zc2|o7b(rMkuhGJ&gBf4WK%gWYx?ml5haV;3;9xpBYTp@N_-yIh>y2CEl_>5>RhX>L
z!&i?LI04Xq)h1>8fOLhTXz3PDBY6J`#y!UY=d+7<$Wdw%@fVbpBnt9~sqzGIz31#2
zW~?bHDSldf$3W7I%voaA;9%K@13^(UakMWJ?<j8JGMwUnekMk^Yt^-I5v~n;<=UIE
zN3JQE4`{X3=KAZdY`^YxOQCnBhqX(BLPWumVnS~6uZ7Q4^;G1p_wVjKer#q!E6KT$
zVZ03PyQtc@ie8AaUl$bxJMB(2$&0;NP~EJdQ!{=M9}{k>PoO3iOi&3z=pToEO8(+m
zIB%XeJ-dExN3I|@mM}<YpYAKtDB#3}(o$zPQVq%|7awI++%rJL<QJY;#lp;?e%C@$
zr7alMy1+O0XCY&7^)9?(M)>pRlJty>Q|~^N#2M*m8X6{pFnolZdr4umx~?HQG7|Nl
oG@bkSvHt?xP`mZN0PgRhQUVndgtq9jP(hC9Y8h%)YS={l4e*EBlmGw#

diff --git a/ehrapy/tools/cohort_tracking/_cohort_tracker.py b/ehrapy/tools/cohort_tracking/_cohort_tracker.py
index 14a78c6e..2d84ed5c 100644
--- a/ehrapy/tools/cohort_tracking/_cohort_tracker.py
+++ b/ehrapy/tools/cohort_tracking/_cohort_tracker.py
@@ -1,8 +1,7 @@
 from __future__ import annotations
 
-from typing import TYPE_CHECKING, Any
+from typing import TYPE_CHECKING
 
-import matplotlib.colors as mcolors
 import matplotlib.pyplot as plt
 import numpy as np
 import pandas as pd
@@ -29,7 +28,15 @@ def _check_columns_exist(df, columns) -> None:
         raise ValueError(f"Columns {list(missing_columns)} not found in dataframe.")
 
 
-# from tableone: https://github.com/tompollard/tableone/blob/bfd6fbaa4ed3e9f59e1a75191c6296a2a80ccc64/tableone/tableone.py#L555
+def _check_no_new_categories(df, categorical, categorical_labels) -> None:
+    for col in categorical:
+        categories_present = df[col].astype("category").cat.categories  # unique()  # TODO: use unique()?
+        categories_expected = categorical_labels[col]
+        diff = set(categories_present) - set(categories_expected)
+        if diff:
+            raise ValueError(f"New category in {col}: {diff}")
+
+
 def _detect_categorical_columns(data) -> list:
     # TODO grab this from ehrapy once https://github.com/theislab/ehrapy/issues/662 addressed
     numeric_cols = set(data.select_dtypes("number").columns)
@@ -76,11 +83,16 @@ def __init__(self, adata: AnnData, columns: Sequence = None, categorical: Sequen
         self.categorical = (
             categorical if categorical is not None else _detect_categorical_columns(adata.obs[self.columns])
         )
+
+        self._categorical_categories: dict = {
+            col: adata.obs[col].astype("category").cat.categories for col in self.categorical
+        }
         self._tracked_tables: list = []
 
     def __call__(self, adata: AnnData, label: str = None, operations_done: str = None, **tableone_kwargs: dict) -> None:
         _check_adata_type(adata)
         _check_columns_exist(adata.obs, self.columns)
+        _check_no_new_categories(adata.obs, self.categorical, self._categorical_categories)
 
         # track a small text with each tracking step, for the flowchart
         track_text = label if label is not None else f"Cohort {self.tracked_steps}"
@@ -95,19 +107,47 @@ def __call__(self, adata: AnnData, label: str = None, operations_done: str = Non
         t1 = TableOne(adata.obs, columns=self.columns, categorical=self.categorical, **tableone_kwargs)
         self._tracked_tables.append(t1)
 
-    def _get_cat_dicts(self, table_one: TableOne, col: str) -> pd.DataFrame:
+    def _get_cat_data(self, table_one: TableOne, col: str) -> pd.DataFrame:
         # mypy error if not specifying dict below
-        cat_pct: dict = {category: [] for category in table_one.cat_table.loc[col].index}
-        for cat in cat_pct.keys():
+        cat_pct: dict = {category: [] for category in self._categorical_categories[col]}
+
+        for cat in self._categorical_categories[col]:
+            # this checks if all instances of category "cat" which were initially present have been
+            # lost (e.g. due to filtering steps): in this case, this category
+            # will be assigned a percentage of 0
             # for categorized columns (e.g. gender 1.0/0.0), str(cat) helps to avoid considering the category as a float
-            pct = float(table_one.cat_table["Overall"].loc[(col, str(cat))].split("(")[1].split(")")[0])
+            if str(cat) in table_one.cat_table["Overall"].loc[col].index:
+                pct = float(table_one.cat_table["Overall"].loc[(col, str(cat))].split("(")[1].split(")")[0])
+            else:
+                pct = 0
 
             cat_pct[cat] = [pct]
         return pd.DataFrame(cat_pct).T[0]
 
-    def _get_num_dicts(self, table_one: TableOne, col: str):
+    def _get_num_data(self, table_one: TableOne, col: str):
         return table_one.cont_table["Overall"].loc[(col, "")]
 
+    def _check_legend_labels(self, legend_handels: dict) -> None:
+        if not isinstance(legend_handels, dict):
+            raise ValueError("legend_labels must be a dictionary.")
+
+        values = [item for sublist in self._categorical_categories.values() for item in sublist]
+
+        missing_keys = [key for key in legend_handels if key not in values and key not in self.columns]
+
+        if missing_keys:
+            raise ValueError(f"legend_labels key(s) {missing_keys} not found as categories or numerical column names.")
+
+    def _check_yticks_labels(self, yticks_labels: dict) -> None:
+        if not isinstance(yticks_labels, dict):
+            raise ValueError("yticks_labels must be a dictionary.")
+
+        # Find keys in legend_handels that are not in values or self.columns
+        missing_keys = [key for key in yticks_labels if key not in self.columns]
+
+        if missing_keys:
+            raise ValueError(f"legend_handels key(s) {missing_keys} not found as categories or numerical column names.")
+
     @property
     def tracked_steps(self):
         """Number of tracked steps."""
@@ -118,11 +158,12 @@ def tracked_tables(self):
         """List of :class:`~tableone.TableOne` objects of each logging step."""
         return self._tracked_tables
 
-    def plot_cohort_change(
+    def plot_cohort_barplot(
         self,
-        set_axis_labels: bool = True,
         subfigure_title: bool = False,
         color_palette: str = "colorblind",
+        yticks_labels: dict = None,
+        legend_labels: dict = None,
         show: bool = True,
         ax: Axes | Sequence[Axes] = None,
         subplots_kwargs: dict = None,
@@ -133,9 +174,10 @@ def plot_cohort_change(
         Create stacked bar plots to monitor cohort changes over the steps tracked with `CohortTracker`.
 
         Args:
-            set_axis_labels: If `True`, the y-axis labels will be set to the column names.
             subfigure_title: If `True`, each subplot will have a title with the `label` provided during tracking.
             color_palette: The color palette to use for the plot. Default is "colorblind".
+            yticks_labels: Dictionary to rename the axis labels. If `None`, the original labels will be used. The keys should be the column names.
+            legend_labels: Dictionary to rename the legend labels. If `None`, the original labels will be used. For categoricals, the keys should be the categories. For numericals, the key should be the column name.
             show: If `True`, the plot will be shown. If `False`, returns plotting handels are returned.
             ax: If `None`, a new figure and axes will be created. If an axes object is provided, the plot will be added to it.
             subplot_kwargs: Additional keyword arguments for the subplots.
@@ -151,45 +193,59 @@ def plot_cohort_change(
                 >>> cohort_tracker(adata, label="Initial Cohort")
                 >>> adata = adata[:50]
                 >>> cohort_tracker(adata, label="Filtered first 50 individuals", operations_done="Filtered to first 50 entries")
-                >>> cohort_tracker.plot_cohort_change(subfigure_title=True)
+                >>> cohort_tracker.plot_cohort_barplot(subfigure_title=True)
 
             .. image:: /_static/docstring_previews/cohort_tracking.png
         """
         subplots_kwargs = {} if subplots_kwargs is None else subplots_kwargs
 
+        legend_labels = {} if legend_labels is None else legend_labels
+        self._check_legend_labels(legend_labels)
+
+        yticks_labels = {} if yticks_labels is None else yticks_labels
+        self._check_yticks_labels(yticks_labels)
+
         if ax is None:
             fig, axes = plt.subplots(self.tracked_steps, 1, **subplots_kwargs)
         else:
             axes = ax
 
-        legend_labels = []
+        legend_handles = []
 
         # if only one step is tracked, axes object is not iterable
         if isinstance(axes, Axes):
             axes = [axes]
 
+        # need to get the number of required colors first
+        num_colors = 0
+        for _, col in enumerate(self.columns):
+            if col in self.categorical:
+                num_colors += len(self._categorical_categories[col])
+            else:
+                num_colors += 1
+
+        colors = sns.color_palette(color_palette, num_colors)
+
         # each tracked step is a subplot
         for idx, single_ax in enumerate(axes):
             if subfigure_title:
                 single_ax.set_title(self._tracked_text[idx])
 
+            color_count = 0
             # iterate over the tracked columns in the dataframe
             for pos, col in enumerate(self.columns):
                 if col in self.categorical:
-                    data = self._get_cat_dicts(self.tracked_tables[idx], col)
+                    data = self._get_cat_data(self.tracked_tables[idx], col)
                 else:
-                    data = [self._get_num_dicts(self.tracked_tables[idx], col)]
-
-                # Assign a unique color to each level (i.e. column)
-                level_color = sns.color_palette(color_palette, len(self.columns))[pos]
+                    data = [self._get_num_data(self.tracked_tables[idx], col)]
 
                 cumwidth = 0
                 # for categoricals, plot multiple bars
                 if col in self.categorical:
-                    col_legend_labels = []
+                    col_legend_handles = []
                     for i, value in enumerate(data):
-                        # Use different shades of the level color for the stacked bars
-                        stacked_bar_color = mcolors.to_rgb(level_color) + (0.5 + 0.5 * (i / len(data)),)
+                        stacked_bar_color = colors[color_count]
+                        color_count += 1
                         single_ax.barh(
                             pos,
                             value,
@@ -216,17 +272,24 @@ def plot_cohort_change(
                         single_ax.set_yticks([])
                         single_ax.set_xticks([])
                         cumwidth += value
+
                         if idx == 0:
-                            col_legend_labels.append(Patch(color=stacked_bar_color, label=data.index[i]))
-                    legend_labels.append(col_legend_labels)
+                            name = (
+                                legend_labels[data.index[i]] if data.index[i] in legend_labels.keys() else data.index[i]
+                            )
+
+                            col_legend_handles.append(Patch(color=stacked_bar_color, label=name))
+                    legend_handles.append(col_legend_handles)
 
                 # for numericals, plot a single bar
                 else:
+                    stacked_bar_color = colors[color_count]
+                    color_count += 1
                     single_ax.barh(
                         pos,
                         100,
                         left=cumwidth,
-                        color=level_color,
+                        color=stacked_bar_color,
                         height=0.8,
                         edgecolor="black",
                         linewidth=0.6,
@@ -241,22 +304,27 @@ def plot_cohort_change(
                         fontweight="bold",
                     )
                     if idx == 0:
-                        legend_labels.append([Patch(color=level_color, label=col)])
+                        name = legend_labels[col] if col in legend_labels.keys() else col
+
+                        legend_handles.append([Patch(color=stacked_bar_color, label=name)])
 
-            if set_axis_labels:
-                single_ax.set_yticks(range(len(self.columns)))
-                single_ax.set_yticklabels(self.columns)
+            single_ax.set_yticks(range(len(self.columns)))
+            names = [
+                yticks_labels[col] if yticks_labels is not None and col in yticks_labels.keys() else col
+                for col in self.columns
+            ]
+            single_ax.set_yticklabels(names)
 
         # These list of lists is needed to reverse the order of the legend labels,
         # making the plot much more readable
-        legend_labels.reverse()
-        legend_labels = [item for sublist in legend_labels for item in sublist]
+        legend_handles.reverse()
+        legend_handels = [item for sublist in legend_handles for item in sublist]
 
         tot_legend_kwargs = {"loc": "best", "bbox_to_anchor": (1, 1)}
         if legend_kwargs is not None:
             tot_legend_kwargs.update(legend_kwargs)
 
-        plt.legend(handles=legend_labels, **tot_legend_kwargs)
+        plt.legend(handles=legend_handels, **tot_legend_kwargs)
 
         if show:
             plt.tight_layout()
@@ -324,7 +392,6 @@ def plot_flowchart(
 
         node_labels = self._tracked_text
 
-        # Draw nodes
         tot_bbox_kwargs = {"boxstyle": "round,pad=0.3", "fc": "lightblue", "alpha": 0.5}
         if bbox_kwargs is not None:
             tot_bbox_kwargs.update(bbox_kwargs)
@@ -338,7 +405,6 @@ def plot_flowchart(
                 bbox=tot_bbox_kwargs,
             )
 
-        # Draw operation text
         for i in range(len(self._tracked_operations) - 1):
             axes.annotate(
                 self._tracked_operations[i + 1],
@@ -346,7 +412,6 @@ def plot_flowchart(
                 xytext=(0.01, (y_positions[i] + y_positions[i + 1]) / 2),
             )
 
-        # Draw arrows
         tot_arrowprops_kwargs = {"arrowstyle": "->", "connectionstyle": "arc3", "color": "gray"}
         if arrowprops_kwargs is not None:
             tot_arrowprops_kwargs.update(arrowprops_kwargs)
diff --git a/tests/conftest.py b/tests/conftest.py
index 31a0af4c..c40dc90d 100644
--- a/tests/conftest.py
+++ b/tests/conftest.py
@@ -1,12 +1,16 @@
 from __future__ import annotations
 
-import os
 from pathlib import Path
+from typing import TYPE_CHECKING
 
 import pytest
-from matplotlib.figure import Figure
 from matplotlib.testing.compare import compare_images
 
+if TYPE_CHECKING:
+    import os
+
+    from matplotlib.figure import Figure
+
 
 @pytest.fixture
 def root_dir():
diff --git a/tests/tools/_images/cohorttracker_adata_mini_flowchart_expected.png b/tests/tools/_images/cohorttracker_adata_mini_flowchart_expected.png
index 781c157ef9a177f7e8cd892e2b49787b1b3ec578..281ca1d946198c7e9bdcc982e6df4ae2c4a9d11b 100644
GIT binary patch
delta 41
xcmca+bIoRghn%sFLPkkRL9vy-er{q(K~8>2PG*u`eo?yq@n;di8`D0>0su|p5WWBa

delta 41
xcmca+bIoRghn$g)LPkkRL9vy-er{q(K~8>2PG*u`eo?x<!tZ}aHl}@$1prSq5kdd}

diff --git a/tests/tools/_images/cohorttracker_adata_mini_step1_expected.png b/tests/tools/_images/cohorttracker_adata_mini_step1_expected.png
deleted file mode 100644
index 8fb392b824816c1946b64878106bc822d5de37b5..0000000000000000000000000000000000000000
GIT binary patch
literal 0
HcmV?d00001

literal 15461
zcmd731yogExGlaZK~WJ=kdRbBI;0y3kw&DuySqgckP_(@Bt*KqyA`Cn^U!gq!&}GS
zz4zUB-~GpXW4!S$V;l~9v(MRkul23(n{$412P?=)-n&C|2ZEq`Qm@67AP5!wiQ<QK
z3%pH}Pc?xz9w%`%r?<8yPOkb6#*nPOlbw~Vla;vvxr?!bqq(gOI}<k(8w0tSlarkz
zFEg|Cza7A2>tM=Eeu-!YUvk^-wYnn&-IYVWP<&lOsUV1NSV~Mp*)171>*}U#j6fWm
zl^^T;Krs-F`g)iAokD;g?dN>|3E}ob=kd~@>im4B&xM89^o6CDcpMvde&lDo7iS<B
zrw|hfN<V{8(3Pa*v#$_XMq%jYQ8N28Mcaw=Jv>AF>yGy+9ETiJdAe3y>Slw$PkMZV
zwjX%^BZL9r;^LwyK@V_oS;_x*Hu9>l_wjXgbu+$yZ_xe}6r@ig<db}O=wKU1K|{z1
ziAhRE#l)<_*u=!{Qc+WPPCxIK-*|*PH_8hL{C@HU_3s9u%YE?vlEkK}?t)^y%ZN>%
zGwtTm9eG6uL~(r`C5BzyLqp>sH{!*UR^?Wc56Q^V9DTx(A9fo+Twk&#4YYT58k9}e
zYY#r}fw@FQMYX+0!!(<%J>e0OM-dhit8ZO(P*hQoNaD0E);Mu+Vx^>_>KPd5SanD_
z+Jv6kgA;vz`wBrFmAmGUgI3kIc>qi3$CruKUGJOAHFOiqQS?cX6crO=lCsJ9U`TdD
zC^@Geab|tbR^oe7PAUgFIQ`eJU;SroOc7V-GD)0at*xT|{{CiT`Gm~%`=}{=uAN3~
zk#TX7YHDf)L)IIQAThDb7Q4D+?!)_rw=F;SSRJ@E*~A?sOQOvATw3dR9f&I_DQ)~p
zM}y{lZv+By$g&;R2TAh=H2?XeIab@r+iaP|lX2Z5p99Q|v4SK5`a=e7tk+^<E2eaG
zblh(H<PDx@r!Fs?C6EKq{=r8uw<P`VT%Qw?k&zh~7^u~`zzC<2lN1&8)==HWg)@wZ
z<LCAve`~QLbm-t%<Nn$g%%nr&Y@I9ad}&Qp--BF=d_~7O!~&jeoMKU``x+s&OuVp`
z771*+3R7>nGYmQZEgvu)VHM+SmSYTv)2b$v{oT5~hXe$YBiXVI>Zd&HR*+pyay`QL
z8iY)@`7QqX_k<r%!QZd%MKB5|7fl}>J3~yFO`dxV1irq$yY)wMpFe*t)+lK2>N;8q
zVMg8Wr7vRfxpI5>@F7dmz)SPt034yKU*Da_pSRrt5o7KqV}d=lZ{qCo)X(wUOy9)B
z<mAkC#vT*ABnx0*;uF|S!*hD`T<<`Xl*4-Q3@Svk{O?pz^D%iJf*s$$js}HsH19je
zVe><dx7gj+f5-4&H_8~m?1|X((b@4mCEXi0DlsU%aI2q}jzLD`%O^IH7?(Fb+9}$Y
z-$s|yQvOB<n9bJoSAigYp?tLxe-yM^`<L@Ksf%ra?BG_0M@CR+>FAbnb5acsmO9XH
z-7?@zBILFsXKsX39<2>fJ%9c&Iy!m_p<1Y=zI;BH0g+Qu-upDY*&ONmL|P9$xlP=2
zFrhD)YLCSxXE|8{MWfQHCqq1>dNo!PdMzin_L;tryFsTaiP)z4J*V|lvH1w~)%A65
ze*QBao&>O9hemZVUt)qACL%4%+G-Ub^+>jjuU=2}j2H0agfK8LGPAN+jk>>#Ofl*<
zCWCl58uCTpf$4#)IEirt;p8GBA_Tr9TYndH_mGzRYK)PIi35CJf6n*DTOyn!xWmCC
z2#&bi+Z-z(78Hz|n?Xe`??{bMQh`L*IrqTHGE;V&nQH8<HY2u_@xzNu<Si<Nuo`ym
zZ{-y&t*=XhwQFHLQlQDjX*D4%E-qfM{r&s*o}QlE5(YlLB#AI0_xib`tbaacGLU!x
zV&!mHnmMIpzz+qjdZ)C{?fAFk3wnCA=0&^xPGVaHWo5CURDsD7L&=lS8!!l8qoN4;
z-R!(~TA7D)z_e#B+Vj|ZRySeW4+)%IEqO%swCAMx^rj(>X?R_Bf|{CynBqdfq#`d)
zLE*v8`H(MscF#vrO6oTF2gR#T`y1aQbQA<q|K7{JPdx@5M05Frg=lt6u7v&l{rS%4
z>_hx-udn5W+Q>*q`tyn!d%;U%`{?3qPu{{}>pWV6`JeU9qV8>W0ynLFs3)KR(sksm
z4x^Lb*v<mKo(ca?DSVg&xe63Wm}umxB!$qi{5p(Uw|a>0YA^ju)~QaGncT++l)l41
z=UV2ckP7pF@qF|H87q;~y7!0TcMveAZjATdqx$`!bJ%xdc(2b9yPD%I+z|<bS8R#u
zP5Z=Ozp}$$S1doFrwg|dGSo$}zFZ_5e_AFFh2wS6P95{a@L+Da;jE0)u2t<B-m_FP
zoGgh*%`h4#@_$%JQ$F-m?ZfxnejypePop<3g@T5<=?#0&s<&m;P(^dO8zW|nZ!-BU
z-)isqCx0A)t=UPT$fAiB$Md_!PHno_S7$;w)82B-UPtRgIhs{AURMWQSIcBKX8W$E
zbq}05G0EB4qsDc8ual>Rj{4b0;}p^1M`<_x_3lSo883Mx!6m+c?|$D3q}qE>*{u@b
z)NniQdva^&P>U6Hc~^m&KcywCt;m>C<SFJ7Lw~l}2m!dP+c0lT>Lg!*y|60{Wg}Au
zeq)Q1?^oUTy)>Ua<T$>qwc4@5loZ@!_56ZsNH5Gx(UR<aO40NB8*k+`67sV=U;TU5
zGRu+7dZ*Zh>kX0YCGq`g1F4>1k%7TB8P1f_avpuxpC;54emY!7Bm=pfPMTVx`{lmP
zaad1r9TPWq2nVN7CYhBwPalXTn+na#e{cgQg&x^zim0!7ym84gU`sk^%J(erhIhTI
zV4gYV$+{d<OsqjE8^CFb4L+$PJKSEP8_Y41C+e71QK+}Z7|gH|t-Xgyev{-%I|J9f
z)0h&`V?~j<En9JF|I>R{^(ND<P4v!)P{$5|n8Lc{wmEYq#dNJRJthw6%9Lf%>YV4^
z^=K;U{%eSqk<s|i4@IYqU+>%7#ctt{_J6<;><)j*S3C{7OU!xKFBp$zxWUtP$AwQ-
z+xan{%MP^Af=Pz{O}EKs0_Fk^?3^eu1Svxm0!h$drrO?V>rb#;3h%*A*^r8Do$n1o
zCYd|Z=lTMdj7)bnPK#etnLfidS?0Jn2Wyk(PLw<p+x*k@qA+TGUF-QdWJ-bpclTUZ
zp9)D1v+YCI5NHfgk@pfg9vCt>#p%GuMe|#3FZo%7A&F9)s)cZ@+&{8p?);jZ+#k<s
zSggjN<$HayGhG#7JzeQu#}IMUP2x8sbd!8_<yC1jdq>#DhKZ1nFqBNFvGe3Bjcj83
z@#Y8$eBlEQ{CdxKXg$q$aJ#5!Fkih4zLsQ*g^kTua0BvaILI$-Y;0htvTWxY%WdaV
z6~9ZM6q^o_s^qJQ7#lx-%Il;Do2oe8ot?(3Gf~UB<XU1FY=Qk~7b@iQ8DPScG-hNR
zI!hg!Yr?G5s=dQ5*cKW0F)t(Zi%r{YE7!-ny@w-HDSHKD#LWehS?_=3=Nk<s^GsNA
zf@u<f-n@AO;;ayiV8wnnmau^V<#eU>vTM`jQ?Q1*Ov*;5Y&M6pz`2vasIxk6{UKUh
zh-1*K487cIA}gDe(bFS0><rZh$Bb@Eh7o-nwdk0iC##zsOcn5y+g@Ic)1{`S&c0Z*
z)0s#~56v*@qU$Z=e0#8b|8rFV)J!z>#MFM&yI=rCpQh;(>H@=Hfk#HqT@(5ooekf)
zxx=GV%Jp=;fx1Vw+H#w_%7=}q_x$9)bJIDNLB*e5-fDI@%%j><7{wdFm%-N@-J?-R
zaQ$>z6S!q6nc1RIZiZ`bZ%@MO6x81@!^X}oB`s|T?p#@A?38+Acr_AwKtRwnR-j49
zW`t!qmY@DMM{Z?xb=1spp&5nL2d=+1n9{d4kaP|v&?5Orh%7s;!&ykEX={%WANzjS
zCRb!Ui$1Ely80+io_HFD0G{+~!`6$?PN7BupMqNDG=hR|9^uV7X9!(qj;v|I2EBaq
z!}Mx~lkVu@0#lor)5u$RhYQ*d8?u>mjVtEua9=A`t(ev5r>~1{o<>P0GV+FZR$YgE
z*`>mIQym_Z83@t#P01{N`UnM_)FqK$h6S9Ge4Jj**CBPD71Ow0l=g?cLZQ8Cv~X{4
zp2wXzp$6ObF$8bR&#{?z68>~<x-8nc?=Cvk4Vu@1(wH+KRCjgQXFgZYa|5Dt+d#FU
zlPU))EiJ{u!7-gE*1t>077!i%2!Ns&wD9?LZtOdE07Q9dJ@u$+dZ0gn-SQZhghU1u
z_ktV<09+)L0M$4<Gc$n4agD`%_?dh(Q-xg|!Lkbvmf17`NH(hOxX<gxmCBuYxaoOH
z4w}#_VbXR>C)^>9JSDrYl0QyXoW64*e_G~?g^1G@QD}WYf+6{0>KJ49^RUMpea)Ym
zH-`XNVZogFOq$4(J8B+zol%_UeYNN7?LqA91)SEVbhku+2#8;P@pcGqKawGvW_UZ8
zy1Fk+tlTT;v*Bg;^_Fh+8~Hy+Iz2<`GSrC}0b`TS5Q8bte3?<tSB*H^iQ<;!6SpKu
z)aGp2B)7vJ8vD9gZj$>F5u~zdg5DbwB~&TA&Xi0{;o>3q_S4o?zX3o114Jn3<rXGL
z>U(8>>mHGqurQ?Sb8Z~wEH8BZGD7fz7C`)4dMn+LGalOoskGV_i*(Ex<gjb@Lgv1!
z)YASvOKsI^ijS@LUU0<5?7$PhKGpRs6=NkgwP)%Q^w<jB4jCqR8{Jqv<~*6!D0Z5|
zJgTNrY;qsNa?Pz=qQ*_7?4xO&P98nV7+Tb{Hm2nH!-c)|KEzqtu?rz(4&nZOAA?VI
z5)z;<mSlB!m0s0_YyrmgDL+4dD=kHihR=036XY{BCymn#t?rUP1AhNAs{IGzI&lF+
zVXi=@Ek{!5csYYOFHRhoIDX-0@ef*7u1%h)aUc!cTHu8S&_ol)ZwAXg`Atlyd3k#u
zoSo&=*C#V+*Y1PV_wdP+jf2E`Ddg7@-3{#PNdJnOagAD%y3Xk3TcYdwuIYZ<Hh`*Y
zLunDSHeIU@_RC>hhl6~3`@de{xaX&S68`&1lDwYKj-3{B3j#NZ^o$JuuU{XwwzUbU
zN~5Jq`B73*3b9-Kp#_yiHt}VP73U3IkzgJY0v@u1%tj+1kg~I5WngIN8*`SXh~&9E
zXnCW(j8!wOAW`qal=$V4Cs`q20I1dlHsxM-H(~?{sm$$2Y{oI<XZQ?#*Y!bdgI``Q
z+i{zX{(M~2gs5HrY)<|UmI8zG5(lfGY#UM=T8zR-NJt=^cX{<McyGW;U+0F;`FXR)
zx4(G(^MKX=2WIO{xeo?D;>?m=bH8H4c<<5E+u&MQ!0tk=D<YxT)P1*k*{fG-j%`%y
zk5EbOWBQ)=>AKxqpOt77s1>Mq9B;JDBTl(mTU+^clDKH0prF?u6gM3X4L5I#S~bSY
zP)D3>yVRYR9HYlDt@B{`0ge?nGov+Ds4ea0#%)uT1R^;I?Dz}KYYo^E0O;b99dIX&
zdpy(N$wlS^r@*{V@}u!yagUJ!2zJuE9E!gXG%^h&6Ey3oatb=S-s!AAFaA*z*mhn(
zWs|KR%2Yh+?naKpUw?47GB5d89%&n=U|Z_~>n+r->+b2HU}GDeUj0MZcFS)+2ul#*
zl?9FiNH^Pgdu+y`k8uWEZ2i$7U%<zYi{KvM%MqX)VT0U1$Ml=Z2}#z7NUEm8Mqq~%
zZ7`R#@l(^GR9R5z1UyeO4ZFgT=!CiH>MM{Pc&C;xOeb)WDA~=*1RD(K)($VU?h!?i
zG|*W#B-jUD+XuWvcaS$>&f>+WGB&dxREVyI<Ma$k{0=JAPDHhF{_>^&SAxjX+pNBL
z9?mEjU*MCvclPivIl1?#PqDM5%D_Hm)uZ0FOc$MTw=aL>fnL@0ITA43zmMo}xxPFz
zy@DUyB@?_VcYg8W1(M|IG<b|E#=QIIPVbwFr_{sI-d&G%bhpZ9=H#TEblg}0lKEO%
z`ba_CG#eRk4#UK@^(|;P{7^F}fUF5JUgxWEkW31148WD8mg7vH?mvBVwmU2J`nBF#
zf5PYxSAxvS`a0T|#Kh0DHaS3W2qoqUZ*Lb<0jzV@<z#F0czZko4tFE<+82Rh8Pr<=
z#f;~)CUU7ushac18$5l7q)lRS0B$T5`CijQ(s4{#u6uJxBmq=`GN(-?#Km}%!zHYW
z-E~)Ms>)U#aHR&1lMFCdPJ43=wP~L}-vLanqi(OUk7Hb`?ek-LH*mD2l_g+mNaEbv
z`-azPqj{>rvIcRnO~`5W1f2X}xodp0(E=U0lr!JDUQE}yGWW$YE_V><8l4}mK6?DP
z#CcoA|HFsLQe*6@Sr=mP8A0z01vRy9Fff+KW!{%A`6~3|T8{T1<gbH+1CFUsB91$G
zqn`oG14n}(`ZFY#nu!dW$h6ns{y>Zvju&ccd7q3{&)VQKs0oAXFHr-^fY_JRRG+!)
zM44<Nq094YTegf2gO)`*EvwCSaViLzvLd6S+sn;IHhzD{gwSr?)$5IZ@$C8Y--X(I
zqN1V}ch|u;CC$tv;t@<&SV{tHn5R;JsyF)c-OgOYWpX9i!-t|GBJTrF*ZTj86FvqD
z&F@LL?ffz`pFyBDwF5;AsU4`~#dL%a936C#MSb}q+!g+`=4{UM1tVj@@4%uWW-_mR
zlz*vPYp|;MZ0$N1u(DE|FTYDhtO?zm<7n49<&H@CQCCgF3d`n*i{s;;0NhajSLVCA
z+AAE{G+G}_DYu@+59vVFx)?`+=YqVm6@gx;Srx9|_NfG5TU1n3{>Fl-@YPAPzY5<I
zlRPpI?pN=$Zrio^V*tR72A1rog)u;E&WB6B>3_4T2&?Axf~sB59MJvPTz1rK<Lo68
z(Lg{O2qTLS<*mQ@_1`v-WnlPQ^gG^MyITIaqKJ;rG}w$Fr7LwwNl9jACRkVkZ=(r-
z9J#W#`rCKnWRB?-Q}eENojESSHjHfrb!R?FYD9$<xm#F9nF@dK`rBaDTb0dcYc*Lb
zvC1CxdqEJfx;#$p#)ExJ{RwEniKk|5a^AD<ZVf~%IV0H@><a2e4U|Oc_W8$6-Qcuw
zH%52jiF$)oPV45+C<0%<nXhl?k_#R0K-dIKD-pg|RDyznNaVG$Vl-BytNUH{260U&
z<ip<<h=Zxu9)vq%*LE-TDPOF=KPnP$3=a<jTH>@l76wF;x5~=x-y;NjUFW?qo4X^a
z64;FIK!Bg{1DU98e7vgmlS+X`iOno8HJ~Pdc-ir$$fWWoL{drW93QY)A!}gn02tUO
zTs9;~JnVgWT7P{yy|S{BtV*`r711?6pB5T=x8H=lD}oGjp=LexJE-Yk&l_p!&FLy;
zp!AvUFSN*{38qj<hA-^xA(ya8SpuL(q;3l+AS;Lhp|59fFxzsx5KyB)0Iqfa_*gvr
z78KMi4`ifB5R8|nFqG=c-8!()c!~t^nRWWXrLrdF>6idX2Su&YitudSR|tG-bF*+~
zC}G=hmh|o0x6@^lIA@&4wfhq}LO_way1IHTD{BlEi<axG^91a4>k@BgqGUPa?CsxW
zM$l|gvE;;?=gUM-&$lNiH%mETQ1T@<R>H!9j$XAeQ?*E^{swUgRxeO<URzlF?Aj9i
zmj(JePn8+4wV(6GLn(Z5pFfZ6$^NQ!wrspU)r3xA(_LGmd3odah&e+yM{<@O-gzf3
z{&|61en3G2VDb?G0lVoS2_V8U`uY@-5u{mw=<;75j~v6M<HCr!hI8b}kl#{j-1qZR
zPF}vKBK05L>JSeIuuBD1a|ZDJMZk=KQYWUM@XNgF@Vbo*(F~H|#!L;nyx{pWkdN|p
zxC}e*_V)E{f_)%G^+eI2?H`ZG8<tHXj{-*+ftrkjqmLdvvO5~!8mVzK1)3?KhATjk
zrRL{P2G}*4K*tTq+L1|>>;~SorDxCjPl>rp7LpJIX{LDJ{nh1Qv*#K0^(%b9)(1`a
z&6_(R1kBV7452{7+TsKiIA61>aBMv=AmB@Kve(q%w{PSoL#f_VhahEu*tLM=flvSl
z>neFmknA5A37=#z5)kO#CP4K7QgOVHub^BG$WjkLu&f_=)XV;bevgEbI<^V66ZqX7
zPKvLKV%7kj`L~|I-v)we^2OcyY1pa5^SH*hUHgAn1ql8o3LSmUq=TcQ<zMNqUfRq&
zNg7y#eFf5eRZ?;?i4~{jNet%D$G_bJS&&+}U3Q)T;6`-28jCY2ES6C-b=(n|huST{
zA^@;5-F~?f1j$D~Y(2Jw!R?fr7sz{9a;A>jtRK|v>-Y;8|5w{6+E^P8S+kv86XBT4
zhLs9CKlD6X6dU==KdhnX&rFn}r3)g!IUV|>;cFW@m1|@L6>0bN!w4tx87Vk*lWn-V
zdOxfAxDh0X*PPT#b{?2s>S<tnTsAFHQc&!y-TBZz*^tUfp*S;N$ffA>zS{oCP>LdL
zXI^aT;JV^;gueE1RhR4YRvIX9T1v$pg=}ZGIhOrOv#-N5Z;y5=kh68(f~u-Mpy>YI
z(q5A4*9UXaLau}|csCE42M`%vb2BBeS%I!&Q$q`DFt~SMp4keSSXN(8YAcD(>><Z?
zDbCp5{ot#BE(s_jBcr>&f8}Dk$P^Ty)GZY`UJVa_<3QI#^D&iP?fV{LRT-8>cbTd}
zWNReh#|tWBB(suKH{>?KvdvvqxfNu9H+<S*KzQ2G6FZce;XdgMnh5ZiBR1?|=rx{8
zz4vVIDwfUk`747d_HXLMBsK(j5xc7eDdr3_O$qq#_4%jL$q=zqwn0jZ4u_+QDfKmm
zjC~P1+16$+oA%t^0X45=hnBlDlrax3eWCRE>EjvP`Iscl=L(xH2b(R{SF3?p;0)no
z6pHnKt|d?l1S?M8?mF!&uHnec!k?4v&5Id`v1gxDm!zIaWKu^&Kn&`o?IYTE_B+y_
z!qu(d>iJGsmZIgl4{XTuGX2_~)9jaN1Rvj9c{6StFT2LBqvW_Ayu9<Gzh&GYt6x!S
zPom_cmQ3P!clQKc!yBvo;um}E;{mj`=6iK=MrBnjflwaK0IFzp&+P4xp8kZwPqh>@
zA+qj)2UT!|l^+YA@Skou3x%3ZWj_v>xWcMDXZzvKx8zTS!bMmUIGK1x)xxQ(fgx^Z
zJ58!4W72O5@6tS3CLJT~d)B?x7u?I%S>Bl8<0_VkPe20UGitPFN<{%N02HcB4|gjo
z27p6BPD=GfxPw)Sgb>JbmiaDE1uXw?sVHjurS{v&ax_si@+N>907i|3mvVCVv$C?D
zQBxoOLkYV9*nh(3!T|L9YAq(4+D&TY7t+zwyI*YPs5g4?JbwHbV2W7vG80@<QlA6K
zocHdTuXIu68r!<N1^PcFGeAqPGODQ>>7mDRm}xsh-)1(s4TZF+467SER-tHaIylQw
zAa9|T&L&Ut12Q3~@!XzS_vs{tFqkRo=XcS+^{e;v7v#OLS57TSLy?_kDYTY{<dmaX
z4g1EOwQ8<he>DVL-hztZ%_{gtCeW`Odj@vTc~s%t@BTF7zHju6&j~MUcsSqp1WQ3H
zF>(9ws*tT}he%x)ksJuAWD1<kwE1ITcXxMx1H=WC)jwuYadBOs5>5h!w7I#NnVmgc
zZEtvZbmYG99y6ZHmUMf($P{oHegu3O=vY`G9j$=sbX}ZmPu9D$r>CdapH5jq{mDF0
z$nFi`VDCX4WYup)hk$+|0i?Nk@3W`ii=B375^_>KwkUC=+g|$G@{;XJuPfN2q_ige
zVgkTbz2h926^6<xQNH}iXY`3>H?4zVNG8->da962hNX3Melvlwd{46blNHBRWYoF#
zZbvXuHcg_$DdaQCPYD#n*T!g97XUh*9%v77hu$q_x+G`5Dh|Ebk*4WhYtufME#A0A
z=WNkhg`EJOrQ}>C)qgGHdBJ6UPgr6X7hyXPSRZ}x7>aPNP1H-;BfM0%cOz)hKBBmp
zsZ9hEMeVc|F`DT<u_Qq75*1mcMMY7c@+#jG0Rm<N|6w;3$5cQ-z^`ms;@#S9CZxy=
zl6!D)us#6zJF|6h+l6(75|>Z;Tz-|CzcuNPZvm~E&uxp}zJ2opP#>}zeW!ZRjwcVA
zKJBBby0S`2@4m`ONu`5-V_@S$3W|y%R#rRa98Ewjs|T-9A-<=B*ptphyb8&x*Vo;B
zJI^FI4K-U7-=xIsF!t-)>fb+TQ7pe5wpRM>b>(A{4w=!Xm0C6gIX0#0YU*<O#yp;P
z$~GU31+2YKe|d_t_ImU7-cCG<6+O4g^|fJ2VYY9STO*&3Uu&-e+hO>ZNY{Y&AQPwH
z&~!sLJg^@M=?R!LYF%_#7CbMC)6@~Qalb9JLYgiAT}>lU>^B=rvUg&;x@n^j&+sjl
zhAtKxi^)ikRln*g_D&fl<O{?~rAfs7E|MY<_^w&9voDuVBag|g#Uyt<#HXRs{=1ZN
zjHJ&9Ob@T*%Pd*5m(wqU-`2NNIYJhE^HwElfJ9PpaKyyWs}2JogzVtW*l>ZNea69Y
zQoh~~Akk`HT$#lv9iS_2C!?y!bqiI^IAD8U93w?yWUQ*{xPN>u9lPLC=MF0T6MnaS
z<}q&l)_^TVG)ULCDP(5*0*{PDd_9PogrOJno1_yx6dsGumg(*Z$^xi+SZ4?8r3<gY
zkb+VNRhJk`?nDl#d+A}%uN&Dr=AnnjJ+{v%4W&l3h+SeTF6MhQy6J!=e!<xo`V^xJ
z!UH7%t+<IcqE6E_=Z|_%L>7(N4uMS0em1u@Q|ad;6t{2MwjekMlK6(zT-<0PoHn}B
zm%K{*LW1voCZ}czWs`g^6pQa+i|N}@X!WS?gps$Hb^QVL!f8~_q{|}`Q>As3M?Aee
zYRU34!cV{`FWJ>hcvBMlz0fXP%~nZqvssSwia$s5Z7`tqyyM8Nq4B!zmqdG@nBvlp
z1wF+%W@U1vDkfS8skwXZ)&v4rJ<{my32>7vNR45fA3uJ4gpWU5q$>pSKo<Zd6V>)q
zfSQ@iH+lm)oqwV2<>A2#ng-;|%n?6+(1MwuWnfrIah*#y?u&(XrYb@JQ3u%hRRpQP
zPsv-KaL5Kj3E4LQf=59?0lnm6+j&8Nd)-*OvP3{K{MF4If0`b~nk4f`CARFMt0AAa
zG?1QHS7Q9f6aV~1Ap4Opz0e!@?Y$CBFg~T4Q1*5AHzQUtTw&LtcwLM$&u8XU&33=z
zt(DftDzPJ!EXliyj3x0hE3t83N`3Ljj1kSUjC`!1Mz?xhy>R87(2Rf8ix?nb%JqJh
zU0s?iGBAy6aFo@q*n?9OsQ%>~Eo+c{i0}(`oT}oR?Ju;qURr<Xuv$_x$F+sMUOs{4
z&Wu!IR<%<<SS5a)Fwzox^}T=Y97ZT4=b2HOtQhA^F*BZ)(d(cTVRCqJs{EPcaYZFT
zE)J5)3VN1}FltqQ5kedQB?Hc4)cpW}cJ(G70ie0ifHpfAONr1YBKG#I)KS3zfU|g>
zT8X}Tg-qUfG;+^qXaa!*0ov<Q00ytFt|FBw5`MQY8yjY|&fA@UdP_-3Eh2@7-I*G2
zf(3^f_(&R}9e{d}jr6wu<az<cd6Wt_iduFFh@Dp*7v9&`@yWIOqAuF?JFB>7tSbdg
zZb&|vCb<8|-q&+w=U!Gz$-qVEo{+AOqIY|H`<Iv)=hIJ0%uL*n*4874N&5~I$AsZ0
z9f#Un6p{Vy^M3%z|LU^&A1pWj_czpm3#m%2A74f0Jg}<o3-*JZ<)QbRj=%__<@X?u
ztWrJgFRNx%46E1ncZ=6Y6BcvhrYB%L0E#yyrB3?W|3nHvs$T!53;ZA7P_ZRgV05qH
zxl1<DGi@9cr1h|ssA4JQOz6aga7pCTJ%}mp10mNgQlh7f3+O^@!|t??{3TcaxoU_c
zJ^$tYlRw8KoJp}CQu8^RDmJCE{!olsy=)p)@9;K7#-1)#P}A)YTki0sww1zEnVv$7
ziqp;Maspgc`7SOxTvK(<k4bZsH`7hYHJ{e5&3F&#OPf4`T4ikwZrg|)FBo|dJ9q35
zi<*zo`COU(G9SV=yCx=@XCeD`QB#ro^9BQT$+mrmSmJ!KVkO@zw0FiD64T|vyixhk
z<54G2cv!D1s7LA0WBGsz0t6VSPi2+c^G!kks8$q8OG!~MG7hxG1S=}9SS?R{t@}wP
z&WAt~K9AV`IsLN6oPy%^jpSrYj1jYMe1`QAOM-R9sp;vxtOHr>l8+VC>)SLxy60rv
z_FkYn*RO|i1n8cV43_eLu585i>sbmZTY1vqsWb>DkydcbxP`!0?U=l|5xW;J^|}=;
zdgnU2p2N4Z$X$pCM-y86d4_^PjCkL{oczlsBBg>Ex8|552m&ghoSogi_(?iMPSEzw
zBY7$1Cf5fpMuVzQjM#(V9a>?!NAxfF$`hB<=4B7Uak}M`-hej?7<2I~Ozo=m#nhqg
z9d;cW(^YgxbO&d2Y7T!Ax5>ntcJfkpX3zm5Ij`eEOlm($yGxbBj_VDZ2EGM_de24m
zv)?zR6wpe6&sRL+8<F#pdlRi1%kx_r5M6j-4w&DTr+X+erZ9OiCTr5$kZt|JEgb%R
z2m)4%4kNY$(CkHJao$qSmP<h@_fL48Xn`MyhJxZf@Kvb@%>$oXS{jZp5Lv*S45sh_
zR}c3?JUsOp2V>{y(LB`=P}7!{m)-YUFj3S>jc$kFGa5{ln*(Yw=dsNwzZ3f;`o{Rx
zL9m%52n)J!YpA)(o(p1lFKc$9z*IoW^lj2QcY>1l&r~2|JqHxt`hlitw@8C@toGEp
zk_1Z_#LBvs>BeKcgW#?H1LjVY!m>(Ne6oTml^y2^N<MvTTgTFD_!j4C;7Bn#XW@Ne
z%EN8YCH?{=*;3<o>+*3-+HZiQWO>DJw;*iObnSvboHdmJJq;)sZlFV@R%N3d<^vpd
zPdF?dv77Yk0ksS1VG0cmC9*!<nM6v~69BjCdL3W@*<%$+eSr?c55Omo=^X)9cQ>eN
z|0Y@&hWSnJx^J{^hKUB)7qX3fPaa@wBR<(i(Fc)w+31fr#k&5<(Fu4euD<>!jU%69
zu^~sH*oTNZgv>;96f3bqh^S20s9>l^orly^ohhu)GcNu&akJB5Q`4re?=2GXz@Ha5
zVSG8fVk`f2!BtPX48V;cwRi$uuSkI4kWwPhzi;2aZvxcv8JB5mFbNmOo3DUs1tz=T
z_YhF2KxawIaexCt{sIoXaWiDU4T$9k1OgllZrD+dqY(gRP?wa2NW#mS3?D8ZnE%1J
ziO<XJo_u4zW;)tOf8>g$5ZREh1lv>G_a4oC|8_5<C7@>sn`RwddF@?A)u(dw=7(A*
zqmjO`^Qu*s$y?>OINFR3C^F`C&FzOcQKzp!Q%G>=!PrsHPYDyvcd|5Xl9TU?xZ_EN
z7I6xnZO^Ct;O1#s{6i}_dMb}-V%dD;*Iks-C5VBXR?{@^P7tWo7h8EioMTWc_M2<)
zG%QgAViLK4z;*p4$Mp^lh*{l}GyD*Yz&3liYngbK_qXw=m2Y<TR4-Q;Su20pylbMu
zm>>F4Qtxy_puD!%$u%YSI!?0qqpCOCx8r69=*r;Yw^KV?Z@5=q<oPx2+yTm>s@(G*
zZBOWq&K^GCdgO;EbX7J6Q^DUsYbGLj+t6=>{j3;P02fn9F)2oW`TWPSfHKMSYw<yq
zaHhtBTI<(KAK+CRL&37PO66G=5A~yx;ugppIfQ&1(6TO@o{{fLSmDC#Vo5rfvxN`^
zw-ggkIE>YGRnHtPF2nsB2;9)cSCke@r2GIJLHbci1w0@Yy=Ev|CP7q5=^=2MP(6DV
zP&MyUS9`rXTL-+b1-~VM9~5aE0i6aYU!&qPSk`GHr>_9Aq@~^eH5Y`d?yvb^;=+Wa
zLy~q1p)(KzTqy*zqLY!ztwmm`&G5xXHs@TEy4RIsjasYcB--30k_a8jqSSDCCscRy
zA;oHUmjamdnQ98#k@eKJ?+IKISHKKEUE3GOPhLzj19O0q93i2VH5t8wsJ>XF<%fhj
zN}T9Q+2!UTze$PLOSZv8@Ea^Mj(9x6GVa<9E{;SvCoC9b;>TUS+}j5aBc+po<aFyk
zNjEV6f*uX3-mh;WNTC{R8dPYg*SoO-f!h>-V(t_<piH^#H6;2Pd3$>!%^z+X>4A1N
zT1b^0G=g#Q@m<DX06HfEsT>#|T3_J^WV%BRbIBhud%PI;L$~Q2s}7d^EkRp9_Jre7
z=ODKv|H_UoP-w$ZZo064Qf6|o0=<0}3!ltRWr4ford)97*=65lH==ZH?`o)s(2kBF
zIaq$#W<qqiTbB#wn}aI0qT09OlC~_><)H*z!q<__KtEu$nqUEI?3J*vpJdhYo~hxs
zu5;>JqVTr_RJ}T<ns>Np#xGCvn9+|1SF3v+h7jni(Orn?*ytioU1c&>R#qe9lxZ>5
zT+e&EZ~U<F8T7!t16dI)U&pon19(W&>sQLtw4&a;h)J^|YfZ5&-4EChXqFlS9Ub&T
zKs!yfKt$`d0%&S&#UTaL|6ttyzmFpSpReFAzKkDGA-%M%M|P5Ho&|7Ntim3)x7{U%
zTB#_d&MnH2`*9EMq@ez9JK+E6bN~Oe(3g7DLAE_T16j-O#sRp+jg%I(ih3=0J){L}
z)}%(DwPaU2F}CcA9Qu{v{n@5^?e}i+awz{%_Wm!5=>LbgKn{Np*Y#qeTjz<H)ONRU
zzX{&>Rg<y`tS48;*D$d2YW3o@JowJaez1d%Mq?7!@F^OLv7Ki2T^rT!P0<5F4o7_3
z7PBhFMKQj2_uBX6WQ)~hyejCE0tM&V)BQ^|rZoqsIfZ_wf#%vpePtrN8#0-jNeIZM
z2)LZ?rlVa>vphYBusucj#mwZdPW8~7a4mwEt1g`2s8zKNPP^Pqc-G|oom{7O(V;Lr
zq}_tEj3y@@_eN)^aK}KWx=<kEEFx#5JZ7htNw8>{j`Cibq$C84BvSJ7XGU?9&b>nx
ziz5Y+JYRG@rfR{t1#2(wpnB(r>Q;YT7GJFwq<sJ0I>Vye^EgU&M)1tjChbzdSGQ(8
zblSt%O|sL)w&ju-=b-9w9ifN2;apwWHsN#xY|g*(x5(9UQUlZ;aE|wQqm5p7Z#5##
zhKjPeGc<|U{#9Tnq|7^8Uqulz@15Kt-RC-zE5))mI~k)lKjr8XrKj8Iv+?FuMey29
zmdRl^O_Ev#4JMi8Xxc|r)x9<=wuAK+Yu|J7&)v_Zs2~Hgw{g4Z-(0QGOK?#*#;$eu
zR3;cqO!BAN@+&SUlis!x;2bY4(lamTX-|&A8fV;-uidkv`Ls-D<iM2t);8>BDqKj9
z;6kypu9G+p32C2W2smz|k@FlxYtT?sS~OZ*BSTS3N{~ELZOqjnBgSop=Pea;Ve21^
zeb0mq8D6(|wO(15hut?C_8U_hpzhC}vh*l+rO!b1cJtYkOVzBzizA-XaV_?^j`+Bm
zKNr^*6p+n-(PMJ>hJMK5C8%x=B2<4Hs)v2dG_0(VK!L`6{8${={lhr_;qqA*>$p@B
z`+#Lb3JMC%-e<FtIdUnlWn_$i#S{Xr09F7^d7Zb!z+R-`HZSew>S)Ljm?R9Vrc0(_
zJM+36ck2N3yLI<T^J7&l5r7Th`)xSPMsZ?tBfNB2-#nc3e06R^*5A=A%cn2eIUY6H
z2_&U$3@udtt~h;6K<GDWu(lvCL3Fwvle)vaQGSYx!dc6S^ZeQl#hB)U_{Sq(l(r!Y
z6TKu_Eher$*Bj<gt!bu2!y%o%t-woEmWzh4N6|bM1-Y6bERfCCpMb9ffl61cCtloZ
zcl^B=N3E6E9S<rrs(A%V#{@yfaGs-rK%2!7+#c|segf$2=yci^Jl)}VN`q*JA1nih
z77Y~@TG?bjuuP!1?bfW*GBblcABKmOE>2<6pf3v?@GK)a^6Q|r3`|dNm6ZG^t*C{B
z(lANg{D7W^R98#O$~LDe=z-Xm?tZkkUS~<!)GanpHB5~2cz?eusk@&(-SS8QEaWap
zS6z{QM`<4WPXx1x*D2~+MvzHau7n=<^FKqqU50fdHRf0m%u}KkCORjz<=5w0ml8yA
zTz8h3$E((hUNM_a7Fr!zx}WVj)u6B>$;=$?InRtVJg<88(pcAq)mN?lo?pKIPgs}a
zt>jZL+?w)~Gb~{Z4I<Dg1_oPFKVTvu22TndZ5e`RDu9C+x2(qE==bk;dx=0#>%2aj
zj{vvE!tw<?V^9pNDL_Iq8_B*0ddUQ3CIggQTm(WwLP~GnP69W{xVGz4KxdE+3zwY<
z`zcPO`Om<>;3pYV&XJd^_FWTK7hI^IHBaYC<I^RYvV4$c`6}(rn*pPqN<;o8D+0c7
z0^AkL=Ypjc0nJ?ooo6fO#|=Esxs#bR0?t*>ZjKNZ%P!+rJk)<8Y)Kox`oxLb-T3<G
zb&h*l<`2Ed)!Q6guCMhV%R60pKtR~-H}=jEKi>qeQN{7(#rs*Bsyesr2!<(#Xji%W
zH&wwG^Fi%Bp?*Eb()QwFOya;1kL-kB^@LcdtHan+e0y?*c=1C%mY2T-cL#XcbKt;P
z0>jY-a~T0%N3e5fYCPaJ9&o;jo(ys73Oz!})YU?_{_Y=iKmSh2D`s0Ukk7w1@ih~h
zV?LQ@4nI)sP~5d+EtJHDObQDVoo{~6sush{DC){Wq|As!13OztnCVQr4t_2adk=H8
zOQ74GOVdx-*9W;1<Jg})W{f{ik|SZH%Q5Y<8{6qKUQYKP*_w)bdop;L^PU0>$sfs!
zhj?p4b=x>f%8Po@4TP5t((qkj!J?Tzk$A@UUDhAr5#TmAb1H)Ith`dgY9^+_Bb;u)
zXwv@D0<u{`g^N(caYJZi@W4cjc7KYtAq2M=KiIOa#44CV%q2gJAa%t(XgnfC*%-|u
z0Osz|y;5Lpkro&4Fo$VW*$e=fjNRPmeW}*q;e1jHye+KW7fxD*KOGZu<WPWfCo(xX
zbQ+djT^)~f#k%e5-NDA@@i?ZvzBpOl%t^}zj&G}(>PXOsK&q}F_*7<HJv=-LPJ;fd
zrS3AiAB~AEeg8REA-^-9$Ym>w{&S8Z!t2Zvm`3V(>+!<;z1%mJ@5ZJiL-F=-bgGAm
z2Su6>pE!zR3(ot5TF$@zo|$oKc77Q>P+x~2cXbKVALg)zTV1k(qT^j)?f8ZcE=Zro
zH1H}<5_jo4YA_EWR;Si)|FgZV&P!M4<(9b;h=lp4-tlABexk@`DsN_xn^A2LXBS(-
zhmLinAvGtndoK!{4u+EW@IV?LNwYRxO9YMo2E(>oyK1!>n!3Nq$Jn9HSFqd7CC%&b
zHs}F%j*rKQe7JQ0_^;#sf(USE{?k7^-JR_N&zR8C(|-p*6IikBk!~>H@{vpCZZV7y
zkN^u9wBZ3KIRYI+rT@h30?gQ<PgwnzmJDR5@qpz}9{}WlPoGNl+ip?IB?F7{{g{{-
z@RZ<JU~C2+5g&Upt41GgvflwB=GD7<lV8<AaGdXY;9s?xo^QWd*3-Q&7}G9EC|PYd
z_sL_v@g+l??-_HY*Y%|M2`(PqD;1U4K)SCX&LSw^ii^K2Ea*YNW{0#5PgYtdb%o?f
z-+_jIrPMx9aIwemHZ9COgQbCu^-cOocP#=`HeL)r-8Ae(w0-c*1gkH?3v@^+C@A`q
zxQ0fYgP)|ft!7_K)7MvC_kdwNG@Wj!^v2EABzOc-%cu6Q^h610hXC~V2XFQ+c(U)m
zoH+Q`fA$|9j{Em2X7?5hwZ6lGb>y_p_rC*P+c<JklL_bnA#O`-k<iiqc|8{?{GYnu
z&Mj<jOFKE0BT+wVD4>FGb#!)t1MJz~=LT`9MqU7S-~XzS<KGj2Z#OuKPt#uXsvgpT
QM}8qGaXGO<Q9b|v296GL4gdfE

diff --git a/tests/tools/_images/cohorttracker_adata_mini_step1_use_settings_expected.png b/tests/tools/_images/cohorttracker_adata_mini_step1_use_settings_expected.png
new file mode 100644
index 0000000000000000000000000000000000000000..fad4448596a4a3d7073003f84dfd1f8c0199015e
GIT binary patch
literal 15182
zcmd731yodT+c!FdA|N0lAkv_8Bb_RUgdiZ@E!`ah0)liXT}n#|N)1DI3nDpmNcYfi
z?&1GFXRYUa?>XzN_e<8A1+!;{+55h)_{Ft9zEPCH!+C@Qfk5zNUrDJzAn4#vH2*ud
z!LKQr$wu%?$VFPyMb+Ni#ofrs4D#B@#lgnj#m3T@#?8#h+0x#QmqU<)myPC~i;IJ^
zFej(&|NRCIdnXId&J|5n@K3NEUTHZ)Aa@l}4>YKI1U&>I43m|5_SPe93*qiTY<6|E
zcWk+aT^MwqCg{y$HA&_c6{)rl#s1@xP5aIzp`}bk;a^#cDi$>KZt->Dgch?ik%T>v
zkMQ^Z!UR6zpA&Mp^l6qlpg5L>y&JplvaVG3v2%v!3NKdI^g*AqxdKOGZ*XX6=)<rH
z#4Ygq)uJlo9w8w&%`FHaAt8p!f8j|*9Sq{WcX;@ijja!12LAO+H#dG_V&ZL%7g0Pm
zkUO}z^t`+y|GXk46$F!ynH_ra-+mj<CIvWXDXhz((Vv(`72h2+;d8cnfCHhUTi{pB
zk1sA}!@<FE*_FWMPmQdyom!ZmckJ0<f`E&d+C;!eD<>38O&{~8_9K?l;&nMlYinya
zr>dj>{{5?}uY@KkB{ki$=Axpnuh<zy`CZ#pP?(&SmUd}*dEP}>>D0-8#*vWFG2G<E
z6+?XCAE$gcPGYP6&J?d!aZ<_ZDpTwC<6ReyY}B{Cil6_X@txW7epFbT*Ur?!n*4$-
zXU0=Zsh2O~($W-+jp@Qk*cXCH^|J~J?np>TX#ccouN)MY*VUz9i2rzT;e(EWv7uXC
zh+6kphSueBfv#riCS_$6u?mN<Y>0j1CI3ji5~Gk%a;Y8+eRnZh!fAKz*}}rYqnwV8
z90&~^ok~cdYCezkpW>823CZJa1VU5!ayZgoN=nKvzoH`U^89!N=5FtvIBfU;0uHYK
z+F|75S3@l%UE|{X{4*&j%vtF5CH7QMaBysDs*bgmq9T5Os(}69{q<m@RmjrnvSWX4
zVh!wW>>|b4&m^Sn*J-^Euh`MP1U&ACP9e;meq+q^iIa_IAS5Q1PUO&ojMq5QPfSeQ
zCF2R|<!nrwsI)k^I2gp(9T3^M2j0?~z>d&DGV|I%9E2$^E++>-Xe<dHH#7tUyTjMF
zd8Djr9sj_$IllZtxJ}u5czZh^^b4r!>PC!O<-Q2PSI88n^iS0H`uLHE@jf+m^n^{G
zbT~=<_0_pu>Dg~Ge$=Lnm{e$f_c-Gp_k}1V#8V@AczCQD8rSktYrtdOhV=gZmU_>F
z^`iQh`KTp){b9n2ey@tum9gn%*3E&i2`iE1)T6HCk$lvm-F<}$nKu8M75vwer`T_#
zq*5Lvw%!_0g_d42^QLlX4Ma{I^4~^fk0p;TK4}yu!80`sOh_tr?@$6G9$IW{?5BhT
zr@1CHo2qps`eN;$Vb20@?=@i%aQU1#Fes*V<-JaYFmZ6K>~&IGbjLk8IXg4vPpv&1
zmQNP<6~-lDqXChyw6v6RrBSA@XYK3$3za*T!fB-+gic<#Hs%Xf3{~0ch@URSr)0**
z#r^%2-RZam;$OqSm+#4wCpI(n$(ep<AAeep-KV5{EiEk#|2%W==Cb@OE>8CPl6kTG
zTUws0sH&0)yZm+BQZLo*9!M8CJ#M^?tgaS-fNX<p*WpS?G`=i3bzw&3+r_|>MBV2v
zU*5(dpfa6r366NghZElJqzj{lUc_!AU?9lmKd=l&q5FfKq3nntYn`rjPUf}1Ct^?#
z%obP3t6L>cd*^9#Q*8=c$G@9+o-QZu?eBjz>5Ou%TAP83kn`JQRa8_AjJ3A3boKX_
z{A6Qf3|Z+<9rK3g%cHg?2fsx7r}5e0n&|PgGouafyCIs%YWu;<<5?w7Pa%jKEL~v6
zBXD8C*bBMYjog~DHowq#`!?p&r?wc{7mrLxuVTb+GttpGp)YnuU$>6EW+2M+MUp|!
zr)EPxer%qaBF`V`R@2Z6PD~{A^Ygm^tFgQd+sXU!;{kYK=*x>*5PsyA1d0l%lS#p^
z#J0)14I_Z8#2a`Ic6AjvpT;YRYaLwLOSJCn><oge=mQVkRWFR&cWItHS$1KueSpfy
zVbdhGQGtzL=@9Y^)N(y#W(dc2$+B}u!31Y#EP0c)w{OzkSTY3z6}ob@@G!aj?3wc;
zE}h+q$#wXw^XHq7k}V@yCNGXH_gUmN;5&v&)JTKtHrW@Fpe$x&WO#UbDyyhGS5UYc
z&#08AlqnABOW_Ou`jzhLY^*UuO-wakaXHx{!w3|VAJ${r#b?4qI1rWm6UV_tvU}Rr
zT4!+@Oyls_q0-BDp1;&QPHrhHC%rcBk$JE=<IvKejjRb<*~9qmb*`G=?C(3vmUCk0
z^<w9JS_s>V7VgvES3j3eXi3MuI4NxTaJlbeapHa&cf9h!9Q(3Qgfv@U3aN!tdnC=j
zB93AYj?p>Z(#_p_D@=*!w2ReM#FRU{))XEejn1b1Bebne3L<{?m%g-qUoL?|Pi`<n
ztahUmHc&?b_639)8J8idhPst_-+h{JI~5})HnzD~t5UJ{615@D<kJ3)*OxnNn&nxv
za`CEf6Wn9;4EnyEvIx~<UH?gcC*dI)6~==s(F*Ohp*`zqoFg9pIw1bM?vVED;GLd*
zKeO4foq580&pWNDnikfky1eOEamQj=C^q)38gx%T-&+V+ZlNh~TA0>_m=mppzi%$v
zcBW7zr)JYjooD&hQk#3vhfUKf`0Xsy%O&?x*tYO?-3EtwUlQ6|>M~Fb?nqi$u^?dX
zjNII@<Kvq8(90v$b?4Ll<)DxdUXzZyGk)i`X^IvvK0}|sW`Kwu{Z{>n-t@zAq~>%b
zmHg{dA!)T-8K$mR^qjuzo@>^HEX-M+7r79M6WZiW_M+3HPEPJ*_NUR*nikaa*rQv=
z(8`$2{-Um#irYo`1YbxGEq`3b94O=xjm<mhV)ZysG6gy@t`w}VmWn``TRtm?@xt%K
zPXyKaBA$&<sKKYf)r@7_*CoqMvN2b%^?hV3urxhfX`xhQhwz%-UmM9k*li`Kv>aim
zvYBYX6PR|8y95=9+o%PTqki`}_|g1sEbW`<qjKbgomE^M;kelG-SfcQhMyc15LEMq
zdmqOoCB2%evIXT$$>U@`M61#qPfbk?6BBcHCDneo)_FB${*&M3nbCJm%_<wRXwm&U
zA3uInNE3_&slVNL?f0it54Wj6IXg+%mF10ZcsrGt7dR;i2hdYFV4D>qI8;u?S0y69
zE;-9UUsq0wecAENeHZ60R#pWQfq&Rn3W%r@7y8^O4Eb4lRH<F~S7<<&Y_NwRypEV7
z1Q#q+S}G3LshUJSLd~koPSwo?eFFO>#3~_Cz1sP+TA9Uu<uK$pGycH8SM9hlA^-zx
z|8PVpJ|klw>sjF6sp=is^Pc{Gvxy3`da#8Mg(R+Kc7i+K5+S?Gi3Tcf-ZTNE1L^AR
zrDtc41nuunox%Q6cf9q7H#6QlVpJY;Xbl(p^5*?1?VA&o;2+rExqDZ%^g2J&FC{Dt
zXLxv+iIp|mezw77Z8-Q>whTtO^%#eGiMF(w8nGVCZTWnsG2eY>=IFcihtHK;<UP0*
z4(e({qJH<D{X#fSZq$f-O%DF}bFU~(Xmv(~sVcBFxDd{{qCPV>M62^?rp|REG%`}m
z?zHVAQ7v-PTI;*<E!8{)T7Le7F@5hXg7Xvsdn!<fgMu(26!=&GkS(pO+`oT+nC56K
zSDypz@#Dw42ZIvM#|ZbzcfaUts_Gau*$f)eX5ijL<m9~|tgV6rB`S>Dakgs~@%zkq
z8yXrSC`A&MdlSd%#k~&S6|1w(%*;GyV5qPjD_uPE*lQ>K7#dn|S~0b8P0z3zA80M{
z#J-c@!RAC;!0%|ZUHr(b`Z+F0#?|QcE3wOO*TsT$)23C16`I!Xo4+qp$xL~lU}3z`
z_o#_QN)0fUZb^^6xNb~Xyg0oB$%BxWF05Y<@)2Rei)1jF#D(b$zas5nCcVDjrjqN6
z#zg)^ciFS+*C&k`K+{90y#CgVRoTxfo+Gzv7o$Xyc+7AL3kw;yIGLG05{RGpM?~BO
z5M{g(Dlw+(jCZg)Xdk+-RG^P8`S$Gt5bh>_hralLb!qG9h|A10DC%$UJ+C>Q@lKY2
zitAL_%)^goid_4zU6TCV|Ew)9s~y1GjhpSNsjCNaWaGQ~`if60wAn3w*wXLff6N-$
zl?`IlfB9~#;m4YW@9UxyLt(8{4YZ+et3QL<k@?ccUux2#QkG)_?Q%QC?C+6`Q2Q7E
zSYNh1Lr7{)OSTxFFBY)TudeP)ms>a~;+7k46TL4C<jkuQ*IY<Jo9iiQ`kDU_;!Fxv
zh!sx~Ia_q9bJ%7s)b}%@`D0YVEv;^_Sf)iabywBf0KB2*TUrg0uW!>?=s1E(ECqzd
z*X(RrIk^Q82>BL+8J$19k47~>3eUITaxMW(v}$mDMheOF*~hM)^*eXm8Y|V4S63$q
z#3CsDc>w~Rftk6`$ZZ94efhqmxjDnLXV31E^GAGn5rP8auU_2=4i4tC#lgk>K4J62
zX5xNgZ+lT$gLV-C6bc1LH|jCDJa!G^0B)LXl*rA^?MoGi?v7_1lUY;2^420$^d@G9
z8GLN-4fz)L+z(L`fiVjW8qt^S(76609nESvb#2aqQ#pRr&en54Fgl?zFgR;`fcw>R
z8gijC`%24>;k?n4K*m#+e&=)A0ZZDPgVFY10)F1Nf)3kkrBv7sH4Zj?ozvaa&OU}m
zoshBB%N>PlcinS2M-Lq?nKNUYqvETFb1TmatS*wrRZ>GHo|oq<1tPGTT7CB88;Qj5
zPm*T+Om{wO=oeP@mm9S{<hL1j?B#4YF~+AB_rpAt+9FqhWBLPZe=u$mJ?~UoKD8A=
z#`{6s_v{`K(F+?JHf3dHpZ#veX@^$4wzjr=04DFww^|?V4ZR?6oG?qAUtDahFzdB~
zBio3S9Ih{qiAYJY-YTKjU!84iWV;l;P_v{JN&I~o{fNeU<55L*Y2m?7G1irGi}p0_
zUi^~tPcJsaOU53x`?I78Lc2ufdrRTDg2+F<Lk@OKO|@UL+|IY81*=F$ErL}D>WldS
zU<umB9KWMbO7z-RbkNh=bN}ezEa}TrF%((Y$%C%^i!9V%FLzpu$^a^poSe)jAwq%g
zwmJSRUopL0(?ttulh*BB7=i|h!NV97Gzvw#!L9##vS#TIkXMc%Yp<iw`gL~%LpYW>
zL0P=O(>wp*__WW<IZ8H>*-LHIr=qGPvxdSAE^|4wG(0@_m6Vj~;qbzOf;m)t_6y9m
zjU8Wjg%xt3-7rI^xBNG3F3!D&9vARH<B`1AHi){<0AQ~Sq)*ZC)yAXFrwjY7y1I(C
zw%nUHF);Z;_kXw<FDPmEiHIzJeZ6H~d0Jn8UVQW39SRs`Onm$cQ`05S(+e`xDaI0&
zYnMbk<bKcDcrlE~{Vj5HP@>#vSx(q}dvlv-J^cpV?u%s0M0Wr%3qvTv+_sYVtP^1J
z-)_DR1w5%VeFDih#UTvgow8GArl!Bot5WIyS%N+)eHWiKf7E86xI&I>q4=u}NyN|%
z1^R!|A}kaHeqiw#Be7L&=Q4$jp^sl4e(YU}!uNpUYAw^)o>dF>Y#z;2w@=AknD{C&
zV~%gRUF}6f%N^#QgEFpO^f{cfA!#)en$gtM<W<&BdGoE^4nn#p4CwG_qDY3shSA=8
z%C<T-F|qS_7OM2-%{`99a8xm^uNS>pfP-IvxOjpnov5}q=r{2Ppg#L$q?5DRk$gNv
z8qi~t)(?0to8z%VLvH{*&iaSkb(JSV`pj`hNbKF+$>=&Alk}zqH}u^TX;tJU-(|t%
z-#tatxzdY?rB6@m0?ab4rx?f$WDOkrm65UrXYdia;WjB58D8VIJ9e-wez4E-hK4lY
z)S|`DLUB3kV+snG+!Hl9U9iD2ExE7sCBJz799_&C2^!Z2!0{_=CV1Wl1|syd;oO^d
zP#i!EaDedYpUvACh80PQg(dywi-q|a3*hnXSU3geKX!;8cQ5*A6cC4i8}!d?p1S<W
z?hn0%j5FA<|4#y&52v8`d^G#L9iwuz=;~Zd29+HsT_dsby5v>bvM1Q63DzY3YKt_L
zI}iD+9>g=e{t9qH1i3&|TbmTXsUF%>8)IeEJM{<nTT|5%tzqN|HyJQ8UtkjxDDeid
z!yLg@tOrW~TjO=S?W;cJ{r5`<z+mfIN55kJVV@?*&53E3k=L#bNRUTEKdm+Xlo=R|
zQ~@I5vhgPpV3<eoYf$$`_3Ph?P8PyBnb_EDKGi8D-Xw*6u|7N{L(H41+UM6->1p>_
zUmxUz#o^yyf$Ta}y8tZiAB`DQ8nxcuL{3(7R!!KnwYTqquX(DzmM?v?Kz=}#ir?Mc
z)$RqfGNj$uX!M&J=B-<3s3iJ;Pbmko7gB?2-Tb!NB}dvn-aRunH(z|<m+}-Hu&{k{
zadGiaiHT#{B-GSO`ubErJ`tbyYNWi9$g8iXu(#jwD*G;eqfTAC^zjj$s<tot$wo^X
zpd@km37d!k1@F+g9OfiZ90v3=h^(w^=g|@9g7GmiL%Y(MXg3i(`%uW~v3w#209m(4
zb(@BB<vr&DuucGi{TLp;h-!~<@*H1us_nq3i~yRGf+Qm&6Qsbu`EY!NhanQjE~d0!
zzu9n4&=R({*jeearXnXN2T2lfx9b;!0LI@0$T1nWF($;MBZ3x0IUYIxJt`{wjqJcm
z^Ztcc8LGmAVM*02Syah1O)Am<Q7hJ3;*NfJa9{#L8Nkkv@$qp9oAbZf(fbJZS!Fdf
z=_2)#Br$Kderps(ZwbP-&~tmiU8~EH2n%-z0o9m6gjZE+$>;EV7KR@ZNGakrRH(|x
zMiNfWDsr3-Dz`iYNLBwLz5m95&mN%igLzpEpM`S3YiF$0xTDw2Egp}ao?cc_k%620
zyI+_Sed}%i>3UJ)4;S4Cv##XvbVV1BRv$&pM}KBo@h5+5&u#N8-NmfFvmz0`!gu1<
zQ$!&p3Qwya{^qqKGT!-eg(xxI^^TyyLw?A3I=)}+o-wEe7FoSS5<FsO8mPbL@-EXo
zDh40l{N4SDl-x|@GrmP2#Rb~FTv36Gpb|5L4In`oYTpM01O%vHQc+Rqr}vLWky9c@
zMbnerdC4P6%(8|vtBy!=86zWF0<pt;03d3zk<_{G45SG;qa>lL^Bs$=$)6CQuY>`~
z4}G_fcd*>B8C_IN%-ZVRSSjVkpAr+m2PlM`Fhg3KKQXI<Wc{txNc{m1wSunQJw2%9
zM*&F`0RR9hb$WVwWZh(;?ITdx;u8`+OxQ4ph>-jLr-Xs_#Ro3HIq@llgMbhOz)@1a
zIgns$4hBR{fxKrqQSn4bNC=gy08b17Hc&aK^cwvEA>nD{@854178losb8#WnGhVDz
zRE&+myY}o_6+t1?eh({a&$!4*NNzWU5woIFj4QPt$X+~do0<un28xoB5D*ee2|9L0
zphBP&+T~?wIXN7d`_4)jYbgz&#?66P7!=3lUwOX}qPiik;ZJQxK?**J>np;R{DpC6
zbT$%Pu$nnJIptnQR$fQz?JdE$j<vb~;>5(oso1+xXWx8q_U~vBsimbQpTz`#;Q1B%
zCl*AYlpg?s2r0K1q-wZCPNj@}z#|757O#VFkwm}4=KZF*=0Jd0Tb7-Go`O;pAM%*q
z0lFH6xK9#bqI&=-4T1i{Z#VraQ`~pF(t?0hy|`e_8$bcqKhAak6u)fGgO&QDKRS1*
z#nV7Z<+q=S$jqb;2nhH&DYoBrqZ8KD2vT`(as&NJ>tGVdbO3bFaB!dsaGpXkdJ~b7
zXh6R?Jvai&Adk&BS83yw7l5?2Kv4({52qIw&-nI@#SVU?{{H>@lg-LO4^SXwU%ko#
zTuJooZ#G}^g3aIisB|KuBH?WoueyEyFe_yF_kcH0_4+ceFDxK&@$p&}rr6-DOaVCf
z@i8pS7_>syQtEs6Bvn*~9eXb?E|PdH1{{0o>F8d)eqB;oW{0Tz?zG&a`t981Kg4pg
z%9ztOGrLB@A=a|`mao#`4uTYbzJoY^?^V``xasq<dlg#Kew7)_s(HWfPGy!oMInlR
zLmATx;}&yz9Equ{LdTTe1kv#P{O71L-o_4ip}xL8g`k7~?yiFke=1B~m=V+(Cue5>
z`T5+bM2S%qWPu+f5<M$xI3Qk;m}$3C`(FZ@rDRQss+5l8z&rr;{tf{_=du$sW*S<2
zw~MyLRp&qMLb5{MNSz<Mh0+a(_?H;`4|w>`Wa58&#f3i0xh?FAd+ft$9D^b(t>yN|
zJBSv8F<-gCa<{&9-nc5ifEwKxzsIR-mIiyzUTUXbMuaKMQ6tK-ek$>@q~lkST=XnT
z_%*K^xby8(k;%YhG!{nr2`lqCk@Q**E>E~aX1yQo;EP!bZR6R{+M<zuZp@3nT$5nZ
zK7N*diK#0ZwvwG^f@c%it1R0dP+|&~2p%?27g)b~D4Rb8n@98zRQA7z=fu4)nK1Rs
zh_X*L|8ee2DvAzwKqD5)rO^KTtT~}r_#8XTT5+RS#1MD4ZP`INL9Hbq7sfP`K3gO5
zc%S@;AN}5=M{+<B*3kESg6ef;cFP5s`}q5R&jxFb&!d^XeVdyp^{p*YRr;2OKZAe=
zcsE>!uJV^X<}bK~do~-<*k3u|*F&LJU%QC4EqCglLn$!?^L&hp6=Asq=T6x7O%<&F
zONbA578uy4d(R1rZn~;otk6(CxkSgmT;*J1J7vru{4RB2Mr+P_Eqv~%o)Fu(j)&h5
zkAKf+KTm{64fQ7B+n;$TzxIU=-{*V0Z+P`$fx}R}68UpZtH|E9$H(~w6R;CfKfU+>
z_Zm)Rfi^?rkKYr~l|8b^$-vFa6H=!?)NN(9ykG4iY?t7Kg(WjG5b^7agR^ZF|Byd^
z&tpOa0;pypLL*k+H#z7^S|RoStZp(6wWccew<J*r+$oxfIK`~~ty=*l?<Lr#X<Cu?
zGb1DDK-y1Ja%Z*fD7I{p2emWC_*2pE*|?(B=ziLg%VjeWBgvE$)pnsfT$gFK2_~8z
zfq9JFm#gCF+$8qTew-QeCH|48QUv2cr3>4PI-9=w`-PN6#5Adca>dS_C&|*9@Dh&?
zXLFm>AG>`PbS6ly;#y<VJvBwy0T>TD7K4XWg+2}9#lNA~Ko#yVujF?5I|QVF*x1L6
zjPXfHUA%*$t?zH)?j9D4lk3%caDyx|0X0A&Lo}IQCTh+>5T$GYklkWe?L4Zc0qi7s
zEiE!^Y;2UA1&VqcP`YXy7eP1cSaWgO`Z=}_de|E^wRTXxntML878V!h0KnImA(onz
zl~oCKPf0R(DJoZx$O*x9yL>eqMGdK`ceN0&-Q<VJePbazKY~3_Psl9KoDx1-`m7JV
zO_N_A^s8vba<1peSuLgJ<qI^XSIJ6ODb<kOaCll(d;(;ym}$OgQ^z5JB%?eNgw2Rk
zUlC+H+^NgQ^(}@5$9ICGj`a;R$?c98OT{90Br6r0rDjv^_-79{XGrTfewtDmSNbbm
z@=m0f7lMqFb8Cdv8sU-(edMjfjQRYz{bX<P6u>cAd3kza;S|tbwQctg4wAV|ZUYO$
z5Y&9ogaDpWeD(~@ea4d!7&?ZRXGbWR3y4Ia7+Cm*-!(lzgVt*F69dZ57a(xf@At4m
z>d)4TbbXITFMwQ#@^ED1zYY57d(7YR@%7b$xv{76+tPvB-w7-iw%LR{%cY)cht_YN
z#r-p#e~}+`NAg(Th~aR#tdT2R(31JDiIvg~wVnrEvdzGPSrg&izwzX;_jgxn!%I7?
zhJA)Ud^+uMHN86b>TxT})fyz<ZQOkX>2b@gI1Z-05WYr(w3I>dE0yN3&zaA-li|D{
z;*NgJem}WQ=+}yC`0H)Phac8g=3r@(<1T|SJ3~f_<y~V~XWWA*o%t)SM<^jux`dJK
z8*Nb1TRN4pP;Yvy#GQ@>Tl$5tcU&9u0t@06hpF^@e4kHFJnGy^#33jf5QXV~PJ&J`
z^xZ_pu6CK^!(Fl#;K%?Thsml@Y6^S`UB`CfpOZk|Oiy6f;c;G35I^7K-E<QHn$O|d
z2$sLZ<(ksPQ5iJ1rY7m?@|+haJ*b|>%*M6^Iy@#YnBuy+x)!bcfN<yu9@c(t!W0%2
z8+J$i^9a7#MgfPr)h`X&WV{99b(DMd14T6b9Iz6rxBG|*Nx2Hlo>FfiB1TO+jQLfU
zUFJm3lKL@ngDIHUP1gdpBBJD$7RsyIu&Pa8Y@@kvO?sBOoD1;H%1j>^pr1!&zq^zX
z40C3ls7Q-r@7;@pQCtVH?rjN&em$ZNXb<bQX*MfEPb7kPU!iTDVlhqBe=--dZxUHI
zuAgjE&G_C7#M>Fx0j7de@%22x=XKRNhSvO1!z^LL2TXE<^jWH0{lXy&v$Ub0s2lPd
z@uWHkr8yj~RjJ9a6bP#{$5MJ?wudZE<&lZAUoeSjE@e;>wH76+hSZgMYq604D-f^|
zqoeiF!}ZbdU|iA#r&?f&(E{75#&u;da}eY$N{2^7L#w~o?=9A?z3)#*|8gi_i5lRy
z0idt-$aBKm0GW^KU0wj<lu=g<EmU(6NRw7C@gO$(8G?gOjpV({B_hryzk_p3%?w-G
z#7K3~o0Lwmk|z$dquwS|7BM9v{hIGuO%R^{nz)46>+*!@i0HQ=n5lWZ&yV}AM_${o
ziYl1P(qcE4{C)IL(NXP|Nc7T`GPctPpGA&*t({aR#2zJZ+_QSW1v;HxW!`B#eUfG-
zQy^MT6uqK+;syskt|DlNkROk(83Hal9N%s0EtRFb-hh$3adaS8t3BVdN({BwC>87t
zFBGjGNr!pN;QW1{;P4zb>{#6(&T@tk-M`{RyOb^h+afr9e_BN}5Wcr_FQRuohph<H
z-)+p5_J?=Tx-8V1V93wqu9?=8w<HORBg51?4h9r5W2BCWyf_fxaF_!Z7Azwt3Ze@+
zue{OH(kgnbtV{?X<>AIyG=P-fT{qMLoa{>CHu)Ev6bhgHVMzqS;E9ApCXh4G(b1V$
zSd6!)Yk%14h-onZA_s6g7l70p4JQjJu>td~A-o_%&X8N-(a{6|%%hlFQc_Z-r$Nf6
zwEg?Wf!aLvdTi?P!61k)c(+*dY;`;6_b_VVs+`&$=c-m^roIDy8MF0aTfq~q$j=5^
z9R|P^`#q9KuSU!!QJ!v@n&6zX-sXIJsZS`(X*VLYwQ(wwTdR2dfFP#QRCuCWiO_4*
z6#aZQ;HMU@?&sW37R8Jf-4cbOZ;$ieQ9L=t#H%AOf-*%_&_A%(dg->WD%czKbYGt=
z5T1jxL*j?o5d`A6Kvbpq%P1Meg%tt&SvCFPSM~7=F-LVFrhd_uGHawttllviLf@RR
zSOZg?ntT99L}u7o=wy^Cjd*qDRZg0O0>{`d$N7;T9(%@AzK6=9qM{FlU6??$7aKHY
zpkdsO0iyA{cTAwQfwGMm-z^1(MNr0de+nN;I}*LvZ3P~N3gEZ^Jp#s=_x0<05c=iN
zYrmG3mcPKb_CDSQtSA|nU%b#OAMgbZ!s!=V_Rz(%jj~R=YF^vPr{Y&9iz%_8KxRhG
zIb0iE@^dPsmCYW9*ydQS@3(E#LkkDe2BZZ_)}@B~%W7eFg)o%z>bvr-#YKw50k1>p
z`}tu@HtCKK!7*k?8tegNKo1A9%5ls8AUe-4{QpJu{5J{ofBzFLgq}Z4v!CQENvW$l
znG-`jx`&NL4y-7KcYb~jkZOcpa9Y3|N%J<xv-M;n49QmxTsn!qTR;E?qdd<O0rvmT
zHLia{-~SVz6mN4h>Fpmlk@;B`!+P%#3Nd2x+jmX}XMznYlmQG}F+V+dY|{ZH5UvJc
zNw~V-QklKc0RN5H{Ga^<T%=?U^KRG)5xY#Opq=Mv>|LH!7xbnjLP>{N=-kM>He1HC
zae4f><E;mBBND}xN`ndTJKGW%gF~1U%R&!q{lec<UpX#3(mu{!E-33dU8f4;Z~WNa
zh-dU7Zr5f8uhIfymTY};6xbW3Gst*07-RsylVy&mdpmZ`CQez}yh-+|y?Liw?~+E%
zQv&0jTf1oe6dvut7Kb3~{`CnSQLOWfCoeje`?jv%=`tx0WL67$rZc<GE)r787PhW$
zWi5=k^4V(Z!jk+?u4PWIMwg1~6BjZ`4SA;4HwQN-B6h7wO>iR^jyD=rk}HF=W>ojW
z$)GKd-g9_9lJOL8pPE5yGTm!aPLb{|T_P$q*|UEAH@9rQm`!+1`~$w%Xa5rsX2d1g
z>FgzC*`Lch)JWQ-qCpHCfeUJqGMW@u=x&bh$F5p1e}?IhV5VBJ*vMx<WrPiRP<l-Y
z(I>pzv+~7W#%g$S-mAkraYc2jaD-{%N<aPb)UAOqT;pE5&h-oBNj4`9C3G5^y$s>0
zsprPKR-+-qh|C42w-7_laakm3^&6+B5fp<`<IDD{$?-3{2ih4CGCRl@XkFJx?kJlJ
z36=_eL)?fC7-AXnmP%g<?BRt;2S#2zl96t;ih&3+<FSahU4Nhm0bcWmZc}4mMne=*
z`6Dey@?zuSo*Ns}IXXH5XF345b?RmM{W|vW_Oqi6&)sG$Sp|h9pjo=FmUamZz%JE1
zFx~Umobf(ukwbRx?%?~kSKr#&?#@QQgMZY~-qLXx6Fx7$C5>UoISD-+gpc-OWN*xv
zPsyVd2%cOrk4+SJwt570#KK+B=R_^NWxD6rZKd@C*dLSVyy?v5b1kq`b>t!rxA%Ar
zxgUF=5swW4HlTTX*t&IjZz3nk$^Pj_1td0LdYA)k5(S&NO*&!{5^gtbRnIC$Q;8)B
zIWqv{ywJm1<}&5iRa>TY{1I#h^t*X-?;50OVJa@IZ2OA?;=rTo@3ZlkYQ;jO5!r!)
zXv(0i_T>EEnS5L+AVipTuhAL8NzGHCuD&xyN$}_Ex{U5eQWVRcPa78~={js6BfqY%
z@~$r*m6K8zY&UBCW*()bTP+1^{q^fN==|mkJe8PLoy+&k1x#DXIy!QLZbHw+b+2f&
zrzrpf?V+Fp4W+QlNj8Ty3YPwdr>*4I@wj90*TwUC`ZLvJSY$;6r&WRm@g`;#j|J|N
z#(Mer=+VU6QR&iw*zxT|Q3KcoJ!N+Gc<*ri+Z)f6yqilUIGUD)bA?O%O1QK{?`$6{
zO-5Xk%|-zZRWKNjiS#M1i=9x{8~L@X+_1xnWOSXI^G=L8&D3Ox9<UeW?@@+dW?yqE
z)M3&%HCU@;qCXV%;4<*pyS28qcDOT34V+b=?Y-KWfp6H@3r8BL0Mo#DI7gOG!rOaP
z%u_z1MC*7RN*^pSQ2evrW=G{4j@trR!mFm>V<X&eSEiB^4WA(%(udj0UBsWL&jMbA
zpD9;dioNGZ{$Mcn{Vnlk2je;mB<kXDN4Q!?tp*EH1eNONiSH+!RUBI+RPjx2UKqLG
zn6&rS+%|{DojfFQ`ppPG%{)i_)ZMwR6#?cMZLyE#?XH=Nl)o|!BpOmG;Jv#a>u&2f
zSWx7K?1eq_cZc;d^<Jdx_mx?n303)e&u{bkU`a3O{xW?D$r6Ut@3!FT`s}qs?ml`N
zikiO!WCD0?3Sg1~Ok$!L0)((RU0Z36@b`xR6%L_Nokq=eWxE^MfbZQ+8wT~PMusbj
zgh-G*JN@ziRF~Vc_^r&7XVgv&XY%R3LaaD~^&W!A_lTgw%m5ip(bGaD)gNYx>07->
zb%X7Fn~4qRkw?gVw+WjRS8oU;dQLG&kDojfe{7De_lwJ_44(n=t1kpx{8t{vl5NVT
zQ+HmS;C@MLh*E;lQpl$E&nsQ&7Jann%@CBUEa<yh%h^c`O`8I6h&f7#|2=TF0DJ+S
z00$VrLk**#s%cJcE+HvtCs2)h#g3-~N7ixxw#muM+Xtg3u|OJ#dm`@vOr2dMa%I9I
zQ*2EXV$VrFbyaAjXeq3%-*{W+4I>ghicj%5vcWfccl$@fo;S`Pxa5^P)cSBsegrv}
zpZ@B~s#og>3?4r-Un%T&j{Pu}gWlBrfS7oG$3Set5(o__GasN_;6po4`Z7aasXA|(
zO9`=FJ^z)E^K#a&Khl3i-@j6{>{m7C>s0in(|p-}G{Iy7iHjHj+W>C?J}I#DcOvkw
z=w2&(u8eJZcDLG24PF^XOjWVdMed~GKo&b9O{Qxc!AvT!=hDV<7|_Ci8@Bl7zc32@
zUm2hNAN83c7_I{tg6;{-bo&p?(X_HPt@r*nFR(+h=I+HcPA>c}22VE-$?0G}^mGI@
z4GvcK-UMCH;FZWM!t9k8m<=x%NxSgbL)^SJU5?~)*PuC4_}>rh{wt^vg_PRTsmR6y
zut}?*2)BW)Xloi%)vRQ#UOBR^`*0A=3LE17ofs&?;$gR}IP(gD+w_|@JxMsnx#TLY
z`l83xd2F<n9je57Eg+gO|DC;DOTp)JMynXL?TVz4XZlu&Ig!qluz&q9%US(CJLPj9
zkE1%l1)baXmTc)O%hRCH>X7Sma{|ca*Z0X{JONb!Gc^ve8G0?12Dt~%CchAELlam0
zMBR^W9Y*6cTk(VQ3W~kfQu0#79@PI89CF)*1{pk>v-&CR&p8X-YXf)`1WdG701R-p
zsgm=|Q;XfFYK!X}rzgt0<V7d}_4Fi$0Rbx!sr?gJ&3i5S#o8Yx?@HW`q4mk+tJhtT
z2hb&hBKLF$9ZtMxqiJbd&6+*PVA*818fnnMc0WRK+z%Z5`(UN>+P9M2KUC8Gxv!YV
z+6F@OY+?}ed_2k1mQBMQ4=bx@egT1(JbkJBXs~2j)&>_k1Ft2&%5Yu1Ai2u$@g01k
z`-eMXOqHi*XR*I{;&XV&&$F?X-r4PD8=k#m+VrcF*AHTy<6c7AhVs~F(Ij`e4Cw-W
z!8pscS9Pm7H?MO)hdHz{5mB4V2CpPN)*sF&8+NvuWS@%dpSj>+*&HCU>zp0wNwOnc
z`t;UmP$m_vrrz24OgcDboW=h&Ry0NRKstO<S$oZJFTnuqM4Ub|T>mU@<*3V$n&FB$
ztW6z_ZNe?-H~h4C)<80=I#SW7L4oQ)<fu)H%{N#c%&VII_gJ7s!p__pv4}(6+?MOP
zwf^<bg7@~vOWZsSPrz~pUsQaHuTQ0O;t<Lv^luL*b#A^z8WR)K1jYcd@$jC|(`T!v
z{=-~#4-UFw>61YfhX4g^+V6Y|H9iL38iY-R8ZkEqULZJfAI#RNDv7C8UZKQYU@J=k
zga;<BHW4~vcVTu2;oEqRqySc`0UZxDOfhJhC>2FSxsLUQSGZiD@qs_Mh@X?T=Npo8
z8SprEC}mTOqLC|-n0O&@)Y$ym{^cX&N+RXf;7$#)S)9M_r(H*0VDKkuJd)2**SGF+
z7A4Dzd{}U))2uy!8ZGS)7@3N>3eC<;k@|D**yWT@>s!1=Oktus*xKOA<r_$MS8D8E
zjZ>X?YFka_m@ds*{v2w2@haUnw5xXGhYSSFh$nFwq0jmrQ|i^Z%mbLmARutO#Cp{M
zh7$LXNCtLxJYepCJ5nIPd4CFSO95?7SzZ0*Y=bY#o_z7)t`RWzD1=?^*xTF7$;)%6
z_CI*=z+pAregAYdbELvd-pR@7@1GKK-?PInt{Z>yVb*leE~zcmwMZ<S+T-kg)1}1W
z5?9_l{Ix!;Y9!6V+A@Vu^RsF<C?g42kqMh`#1Eq^SIm~P9Z3ntc>F`eC*<T_`t@5E
zdgInyM84&Cx7{gGy4fO8R_9N=CiLL9BXZlJ4#JhH5OFdn++pxLfpCv~B3`3~Uf%CE
zO=Y>)IY|8|9RsNW+k7$fn3NPbFkJvH9c}K9j(vM;>*9|yP2kyq4#;Khe6-Tq@@+6h
zy;!rE(tSE$!lq+vEb+TZXNAwH6Htu)j(?}Go_4~mv>0UI<Rswa<ODwTINV!6L)Y;Z
zh={vXq7VJSd<S)PKdFIFEO72qIP+T_>tGzWD|z!D`nx?F#cHg2HUr@!S`eS`N*Yt)
zIk{M6%g~A)KyDgeh?cnADWsy^El=7L^%C=Sw_|n-iRTZP;k&$UXsoNz{RxeFx6&?f
ztpB_b**v{Bo0)Waw5?sXnxJ0cBg4-Lg@y3JoIc|0nZ0p2O1o7w5tKiw>nv-y`fH10
z0?C?bBi1*a-Fi*=;#z>t@hQC_&!w_1VXU)qU_Tn@`htGWG3=aW-gyVqLP+^e<L{`Y
z0xoiqMp+)XlY&jR=3Zj2G?<2GW@cvGvIam^A?0b|ibZNZX&7XflO;*<i(k;GddPSY
z$2&!rP)=Fv<Gq*b;ta@e<Auzx1?3{i4vypy{mUw^KbK(N=T=*#H~W}r8_Xbjt(V;#
zrH0;at_~+=`;wsE)XdeY*$i@+H+Nu6PiEtlT?r+$wi6%Pkhi&e?~i+A?bg&Gm((*o
zulc#8o*vLGH#A6LdL0a$T|Us5BugI;6Z6b1Mrbt2k99cYOb-l9-~amFx8{cl6IjlY
zMiR8$jsnaMz26@ieUdshcaC|^&*9T!)neKAwPQ0>Tlx`nkP{X$n0cygikdyt-Odu;
zutT@t2cl=FNP`p12wM$QtPW->nVBs+Zh>jc)A<l;2uf!J^Om;#)<E_GlObe+>7!#$
zqxEi3KhVo+bUDCkCV>H8kU?Jx3PM{$2u8{bBr-e}uz|+d3&t9N1V-SE+%#$rBj)q9
zpZrlc;3c)rbM>{V-qRPI@_EtVBsJANa7PCtWYnT7d*(EDSPqMmAUd};lkwQ+Q%6`f
z5o83P%E;7*RiB)2aGc8aWIvIO14{pSxTh|A8KYar!{XA<ZkwceDS+~HA(724?b;;W
z)>)qQl)KU&S4W_TO#1J=b@{eml3Z0X<Rml0>A}@5tEDrs5K~eOo;(@MN|_F4gIOdf
z&0Vh7S_VX~{5Zw(gEnSQV=Gt~uNyd`Ko!q8Ss>NF<jBr@z0*a*vZbiL(^FVUKq%y%
ztXn!uz>_fz446JJc(HqUxKz>0Ve+R~Yjd_S6PWbCXaMXOWmBLAUHA8$jE#-E1_xI=
z$n8x5+ypZlL0Hs2F<?>x%)687xe(enoHDexw@-sxv%usOiNA98OK#Wow>-SOWMBvg
z49+JhFbKPCc7frLcYlh>fW0&ab_$fcMBwvJR@s6=#RHshXbl0s5+}v6WlPpmJ@-SK
z-c4of2rmmU>Z@jpQ*ybQM#gh$>OL&XkQ&S851<S_WhX86jVv*MI;#5uO)VxlSxbqy
zdqOxH^5*T^pvFcCNOyO)D01B70?d3I>Fd5wfyCr(k4O4D$+4`Rue(P!T$%`=hoyc`
zp`55AjQIrwN$9{k{HFiL?LRA(Z`<8eB-P8%S+3erI$a>m^VtiZs^X->+lWAK0*eT|
zL^<Cvpm5SfN)ce(|L?mk@qZMqMhyu3|GqKu=3dC}+I=qC=PCb)f>Kp?9Nn>12JezX
zp1%*AI6ES}AsTJ${}ZLat|g~h-EEE*N-!vdjf;CTqxc~<JssTr6%6Z*$p+^Fb~@s}
eA8B7naE`a!3#?$O2KTl?WTh3QN}n6P|GxlnK#6Vu

literal 0
HcmV?d00001

diff --git a/tests/tools/_images/cohorttracker_adata_mini_step1_vanilla_expected.png b/tests/tools/_images/cohorttracker_adata_mini_step1_vanilla_expected.png
new file mode 100644
index 0000000000000000000000000000000000000000..aa33f3cbb696f84821806f76b7ad598cffbf8b6b
GIT binary patch
literal 15313
zcmdUW1yodTyYB`>L<x~@1f?59N>EfnN<rz8lpJ#CP*fVE8w^6FyE}$X$)QuaJLbOQ
z|2y9~-#y>D>%_Y2p6gn(hP`LI<9*-fd4BQifagjw_}8hfLlA^7C;LPNf-u3K7(Umo
zf-fY)cpdl>aeS)bs0uT6bbf7b0x7<Bw6%gcTA3R%z)kEO%waYH+``-foD6Rq9c>*%
zd3db<egHSj-i!yaq@fB<f@>?Q=>S1^O6U)Uw{tKP1c^z=J$dxfC23>Q)rEBN;&SI`
ztT5tJ^u338S)blB`E$Or!g#?saDDD7<9ErQ!z#s3&2s|PinmF#im~q*U_W`u%lMVl
z_yr!(O-8ZzcOOEx#m1C-cF&kTQQ;QeO{$;BTkCkrzZ`ES;o#HtpjX_ibz#Rv+#me=
z5?7<^1HN^<u0xEBj91ej21Z81hyRrgH5mB}S8Z+W6KQF;Dm)^hrnauGwCwCIq)8jw
zM|>zeJlxpW*yYbJ{(*tJaNJ$u$}9i=Y9Euaz)mMW#Qcx*h#N-y@w|yV=N@_(51%}#
zZry-8%r)cg?(RBn*If}x3|Z<<oS&a}m|A0nz(b7Ch`Zb@8dGp|6ckGAMlR}vi_;N~
zjg8TWI{K^SXeA%Hp`UAx)TS(PU_eEy%qC>$XH$RrH9|rb78aZ`q~}yD)_!P^zrT7C
z-R}K(g;z0E$%`&JJyg>v?S?H@HKH<<N_BqZ_LNW2ce9FDR+RLs8MnM6(4ra^1Y<ZX
z$V(Bp@<c`^GAio*q>b5Z;|GOg(dc*YFjt4NWFnarT~^X#{eyy<jQPV86Xj)PWlPqE
zozS<WVg|0FC31`f&c!ZG*5g28(32#h5YPW1E$z;D8H}+0Xxh8C|C>xeP*7Gh6%PwG
zL_`#BA%3=A)RiPsv^zuCSK(mBqxQ1Mk%<BPl4>VRLnGl4w>eqmUKOSF4LzwUzFrbP
z3UXDTn?yt>l=SrUU%!6UD6^T=M1l#6NKV!@(Nt3-!>1DbzOifZ{`E4nu(Yt9b$K7m
zWw-#%S-~gEgcdM+;q332q-2xsBakLKYP}DRXo#5=7?_zq!Z*e-@?G~#`2_?hg>4z`
zPkRUynRd@@d85Rjt=cngK0ZD{n~Bjxlhpf7SD`Xwi#`mtw!#0TXM42neu2RR9xV2@
zO3S1G{T>`Qc5<}9Xy@o?dCGk&JwN|u*3h%g7_NAJlj~4k%^}r1B8){sf(}gdH|db$
z?Iyw+or4nrEAU|9l?dYk!yWwWk~1idb6H|NJu_3?%NyLLE%>}da*K;6RX8jUzjTdF
z#stg$7o}+y{@JLy6H9&P!#{IMQZxjgy1y14l&OpXRusi2^kM!OZ0P0nFB|5!K7;?t
zMi+}Sv5e9=+cOn<U8qA@wo*ew0zG+>*0WtMUS2DFb+P0oWW~(sOO^U(QK;(ZbcHj?
z<d3mSgydTcv7a8SA;9Rke<YyoINq6;Q&hzMoS*-x?tGDlQ@1jzzW!mCsldhAky%F+
zTUV-NDp&~-5fSWX`ZaD&(FZk8W6{^zv~{;wo*^W?NLg-^+FR<@EHsu|><tJA(DA-F
zFe~135ubn~tw~5p3!K)}ypCsahV%55v($1}IXJQk1}uHiPuJEfzFm*imGXev=ltA5
zMMZ@|@J%1(hFY#p`&5nl$+9FWpsr350?}qS-})j?uclLL@55iqD4rTt9lpQs7)tk`
z;~n-53JDKkvY^Fa__R0bL9&>0D*}PI@k)>Cuhkk%BNE{}<?1(9WF}zR6&JMHWc<^s
zwqYzrU+GV}8KSsnL6TQ@DeC3rRctd^IW7cFVLF(hC`=P-p&u3;d`C>Is-rMFo0CfL
z&GY#KJ=VX@D`-COo=r7lyMrSysYJ2P>!N%q!P4b?2f@b0g<ZemxVw}HQ_;|n9m&-l
zt8sT;3`P|e^4}sR2FrA2arA?JS{O#4zkgB-7NPidP+p<}v&7lDz~%X}<lvXbIOi7^
zc2$#jbmA0}l9C{d5jgZ7^-N*nQg(KHAdNioy}ENeAEux2Kv)cYH$TL>nDgT;HYA|9
zUeApO4<6KvFM^e)vv?$Vf#?1F`Azimfsb3s+QG`8-@?MejUW4k&(TX)w8{8Kxn|hr
z6^*k$%See&Q{gj~Q_kR>heSj~Sbqkbj{$uVBCG)#^tDu<B`*q3<Wg;oAgmgCEwXmJ
z@?QRxO`=3Q%?}%HEJTrxxJ<vU#h|I(i{HDBfsKs~Vla+~N%~V<oW^{~FYN1(O3rb=
zH{!nV*N(#Cbl9gdZvmnW_x|!u1CP}YDlw^Y(z}MchcHn~{g6C*1>fUZjSFAacG1Q!
zFtX<^f;pwmk8-7A-s1CNzq}|}JmseDEl^S5!sIQ(#gx23B4W2C9<Z#jqIt$&^26^W
z!EMIqW8`9a>EdJfL778zCt?;p)X_lCkeN;4hx|alOaR>==T382?%~v{j`KdBr+2z+
z_agr`yac6XW(qhgb?w=pCLDTSO-U2Y<au9k>@D}^g$yi+M=#KrOfL&?)CGppGkiCv
z&Xse4)HQMahB@nvOZ!%br{~BAzK=^ZA`cjTB50!)&fC7^O(R%@lZNX~ACOJgcUIrG
zFWFAwM3mgl-0h59^RcDp9QhzsL47nV&YQ^VjO*(*lsX@mfNfx|Pp|w)WYhZOra{91
zSM{7UALBzqLq=j2rG<<{8$*!sr?@lKe#JxE=X<^CxKJdE@)Ag5Ie+ewJoOHb=M53_
zwRSmeW!D#U_;Y7iwhL8v_{(yZ5L<QVdu6DFm3RuSYxG}k-@nl|Q5o1}ZJB?1^`@P5
zSZ67{eY9Qq4h>I>L)w+z@LQ}-Mxj51#T>TS>m&7o#d4s?vG*qltG*c`8HN?^+7c;P
zPhAFr_{O^85^!`%CqGqMB0_1=RUr{p8((0A&?||BEbc&2QBk~x&DTJcQ3k)GbN75}
zh|$sJ<md+cBOWDYeWbF4Tr9V4K5}EiaZqu6s)pJJBm#`SEOnlBxEPC4lEMD({_4YV
zONfMwjNi12!qCVlAS7h*WH~i&0-h@2>9X6-vNDuK&ZS+Ny)MtK`+QKy@p7Z?_jwzq
z*5oOx8j9guiVAavwGHEQHp2~Qx+oYrzRa%q_Id7yh8Z>AO_5An`s}iM4ZB&)aj<Jx
z^a2@fq`l@xBw`%nQAr68;W0!$9g8ib8{&6K*W!DI_3-ZU@rA$mCjGiVqx@i4Tf}w-
z5_8_%hil%;mA)+T-f1CUSX=}HKmgg>+m9BS+yDh7&%SOUfS85fdR!2i@<w@jq0XlF
zTFCWz9hMYKyAxVKJ$o!G+eUd&x7~OnDmuErq?7E~vuD4@idZ(rON^TRiJ*04S=g&A
zwHJDNN@s@~nq@Y$lU2@n0lEU&*|gKo6Pz)fT`9qo6OI2Aura$gBExeWKTKs99keW=
zf%iel)|FS6*uNq`vfN@erod2RadELLS&TffyVZTG2InD2Hz2D=Bl5DIh6F>7>!aZy
zIqtSGh2`7MN?DbzvS@JX*Cv5_PfShSxzHY&_1;&%(6Hq?o!jr*ppLwK`!<;BUf^P9
zta_~knC)yej?2qSFvEu%NM~vKyxLRt7cX8!>noXcdp9pw*gIZiF07w~R%893kvpM)
zQ2wmX_Cm2{s7ds8%xVH9JC(TZ$8*JT4tE&Xx>fZ4IFxww8j4z?uTyYmR44A;d>8ly
zMK2GcAJ>O%^kZiJ9Lizw$|F=D5yo}pkZW<@Pm!DZyFP(%W>nyT(h*Z@iE;ZaLqkJh
z`-N*OD=Q$_BO@XjCn_A+czBNBuJ?XQV?I<+z!QV78}xmC<Z-r9*4)|}k(j8R-u>mV
zpUZjyBJMtn4q0kFe{nROXWxyIit~aGuPiUCpVYJ%Hmo?N6!g5uC2nhL`+i=i#YJ9X
z&9qGz_O-hYSFo6n;j5=|!a?N?pT=+INu!n5eVBD&Ec~VSLqrK1=i#ro!#15IM<h>~
z!)e{HYz^gGc1>$G8ToU0!V7BbM4q*cpX87qD&)RtQ0sco$X9~b{7W2?*YSkJ$4p!Z
zNH7g-$ltSxLJn&T;x>evSWihi^KLYVrm<!G4zxUv?lf96s${;Y>IkLS-%oj5B5<aY
zO9STFtV^;A1eE!BF<YwF>5tf-KWXpT<1#QY`S$fa2SM5R`H`>evuBI|&YYj0cL<%F
zoQzrtQQfx*M$Er%4W@i`dhPmkMoG!Kad`+A7xxY`^FWm|EHEf2Ud)*rB&+Ozfpn^#
z2+WBddn)Hz(ja|k@#)EV`H6gJwstUbh2bI#)%W{U+kEOwuVYs^e`_y3YKbnZG%q{Z
zvS8>&`Xe3scL`I9a`3Bub*}X4=KUs992IlX6%C6G6Qb9wOj#Xfr?fTrZq4k@j3)(E
zDBfH;{uCI7VbgUo9e>&DLO?pdTV@bS&`dI`y~9_ZbMkz%w)*Uj**2f{wLSdr0J|N>
zd7mu6#tKb_xTlL(*6E#URfc4t(Ml%{WyuRcx-gu~!S=Rrrj`D0IJBaU3_?ONad8we
zoSKcg&XZ652xtMKH~Zc9<<+UnTmTCNuK|RCf#Gqo82d`keJS1yjST>h#xgkGo}F-A
zPQLe|^5wT`;nW)WY2mfCc;v{Z3Qxt7JX#IDL@W+dbxpWcQzi}4*6FU<&O845n_Koh
zr-jP~e7U|u^nw$KySA{7tZ^@AjND1=dzxQ<L6TV^P0kYE#4z-(GKV3I+9(Aty=px!
zoISgqucja_YXV@I9!miE{w#N2GV(ZXLB#%&dHIrrl9Do7vPy^haDD7?qL6jIhNI@8
zicXjN4?m26f5o2v07O@6V*#$uET>I((Yc2*mX2(}DGHmepr(5N7r3>NVKbjkoMRU~
zlab(wxyu_ooJT8ScFjLDv~^^J+}qo`qobpIvvSR4EvM`nUUg-P;ZOO$v60X}Y$VXW
z$KCifsA_2$8HJu_@ZG&VFka3U>CCV{ctqreU(+Lyn!bGbGR2+V`se1(yZnimK(P?C
z_;syl%;VLRry3erFPJImW43j4Fp7$v{`9yE{)0R<HFDH8%jXGdkLPZ{VpZnRz~_lq
zLq{}+IkGgTtn<6wDFwjVZN0%1e3;;VlayGvw2BRuy5gs}fBd<eZ1U^EV*^~gtUrif
zV%aJ-B}H!@nH%&M6z4rV?msPIswAE<`HUt(PS08%Y2mDnXUO^dd$ucre)i1rPF)wB
z6acD>`CgZ5T2xPSME^CKoA%_|8QCqb=CXHcxh~b_Qa4@oZlm#KS=~hl_6;(F60HKQ
zoC(_|{PT+~Z<afE(poL|uKq!V0^;1tIW*@OI(K^l!?BSS$WU6bn4~0{;T&z5H*Z+!
z!u-M{PW{1-KZrJ~AC0>KuMu(l{e02LN0FJh$&c{J-sGNqB=i&vkRbrY03j#%`(Fd2
zmi1@U@Y?cYp)S)Kk&<H@Xj(PH?~YzyQ{0c!0%!bh!eVX599Mq@6c-B?78T(V5+1=t
zN?P+V7z8<kmFrH2ktY8Bvb2LgUvZ%4qcxPaQ*5gmRJ?28-dcjl9VVSdh=_^jH*Q(J
z=SCBw^RD%=qLtACCfzD{`eQ$WV%T&FPu+P$em*ZOR?P@c3==)tVeN*EbFiF0bCa=W
zVOccha(S}s3O<(Ap_Nj9vFM_1sq?`H|CWMv+v@=eCjSjRA^%*GNVviZP9KQWyv{At
z?QFHvV0Dmn;+6OOFZf%YoKE8<M&)d-6mn#Jc;nlc?@9?3CIBnb9((wY9zCiX%~ej7
z2&R`z0p%lQM@vQJRtVL-ui9lcB`-~|burOIvT5;qw6EmzewcMI=lK&0i&YWA%jJlu
zs3)N6<G*r*|Di0I)hA1d;^*5<(E4ZrR%Cl~tiBg+Jt)8N;;#JW=H`Gm#Bu9Z-lV7R
z4W<-e9|4*IfF2m1l9GZaZCJRuL-~wa`wEO?)bjLn#{h#a*<T$Bis91cx0w)zz;|oJ
zUHLdsCw6#Ma@n+3)$rjfH0yh!Bqu97vly#?)!<}z(Q#`knUq6~*LLP%43B<KCr@41
zr1Nxlwd>xSkzCrAmKMjg;gDS2s)2O**bXlm0vsmk;2Cg4I&8W&TdO2c#9`^>*l?}q
znMR2PDL@_QnRHqkwgWkV-?;527&SCJ8uA>vhUSWFnW_6_-Y74%Y>kE}HdT!Z`@0`L
zd;m}H`}s3NDQvn2WZ(~zHkmSEbR;As`9_FP_KWrLk{f`+PuF^4{6?1Xpkd&EDL_mh
z_yN5tKyCo!w6(jwz($3E)+PepzrP}ZI(HZt82CL|bssFir$J=>fVpv!-FW@_HAqEz
zwsp*^LpVeyCl@gOujmS&eS*VqaVrEuugoV>QkbflN{b(<ZJ^KkbzW$d<j$QteWljg
z?|gl;@RqkGt74OqeiD-d#31kLDq=aDjbYH8aBqFG>ZrPuo`yyNFi`xB?y<k}*%Q>|
z1&G9(ppZg%6udsZID`Xr-rfL0$;Wb=?98<^BM@k+2!cjlPEIab!o$<k6?Jjs`d67T
z0!mJ?H8SSYr%w(x-V1Hv9@`B#I?j_$C8D@}K!&rpj1N-x`Tm`ipP#>4_oKi6>+|!Y
z+|EVA-joN$=7U#*(&zYU)K5Jjub-bkf8J$!OD$w&(Dvyz8gBad_)tx&Ic+UgydeOy
zgO)SsyXh3_*f+0~zv*YpP)-d+%r{t+@H=l#fP}8k`6nBxcBx60U$+?(bK8F-9YzPz
zMb9$_^t!^6MMq?_&tD1UZySQR&w<tzHDFIp5L?xHc2E$vQ<OX{v(UpVDw+U9iFLSp
zH_75tpzw`}h&WQXM3!}a3M!BcQNwSl*8QOK-WQ=OXgR(+fuG)kYe~}TP1cihun`qg
z<?r@dM5JZ$-juV3Pe`!YRUP|l<ysU%knkioeA3eweU`q>_^tSXRT&@+k<X?yS-a#S
zuKfACVnLm!6HdeJj83Zdft_~c6U(HK#K{72irXrihldA{Fm3`IWi?S=@GMcn^OV=?
z+yw|N*X#3NH<fRcA}JoY2?Ee0zSB&cmX>Bvpz(X6LZHa3Hv=dz2zu|NgS8PXUXNqj
z6wgCVNY7*E=~8cMSY95?1Rq_0nm*zyBngKDayMsR5hrMV3#8||g`=i?^zfyLqA5{V
z93L8|7F&*lf&lxJob36?6C@luP{c~D#@>xtF$)L~`;hCoyai$e*drw;hx@?&Fb$+r
zJR0Fo=7Sk_>!W-iJlMFoTdjm5V`C9bgpyr_CY^64%K0=kHPOkO?txn$SR(S{%8HmI
z=*WqRBAJ_?HwSSK0l`pGLPCQ47Xk#nD_~%Ji5vt)K!k;|H45JO(Yb-*tYBck03<X+
zkVUTG;0#teS#DNsNdUU@R7r^d#K0XEmLd@AHWiCxV3rs6aN%21p6zEph=BSK9t7&0
zU4>4xWNb`~o-KVQJ&{#`hOC_2Z4r_9p)B>m9PRsLoEkoed5~iXz?nugCNs^ifB{-y
zsdQRv1=)?gw{W>T(XYvFsR&oA#G=J{+7qY#Ef$VO72JB!^6^K!*W<<J<&&hTrox`M
z47YE;ZEk)71l^}kp8{qS3l3Cy{=9x?9)aeTz2bX#VEl}M1_s!uFBT3#zF|vXN3E)=
zYF5s&)ZZf6_bR>%m=#J<M^=zhbF|A;fSe^jw|JH*c@Y4(8xc7<Vo+I95lBTG`e5CN
znwXe0x3s)2L&k$yhR_!QGFW-Ja3y@B<J0ZO<$FD1?Y+Ir_v=pZ?d<G8B2d-S>j5N0
z8C3OD&qGR{n*Cd}V(>K2vqSNNU#ig!b1J`Y{fSfwGJ)3^rCMhcpTic<d)$d{-8w$V
z^TvM&Y7s~)ckkSJzrVlVAp~SotI_-~;|IRJzH$l*;zq7DZU=GTNhv$Oef<iC`$uV>
zQnF}){bI-7&oe2s3@)2IYgY7WrwhuPzV<uh{*(d0Z{<4NAos;fdV7V8%v=Adl9|V2
zoO*Sdy=iih?0y@0Y6)oB{U4NnJ-btZZMrZ-XSB`<(7iLK2lu|9OY$74X%Qum!$N69
zq(CSXj9E=_uLOcr*4ES0YFhMa%5~f0FFn5aP5(PUH$J7L<c9mm8x)PbnW{d)&iSK9
z>lq0pn*aJvgikHhJio(wMdwOnyW=CX%gjG>H4?9oaih|UD^}?b<oTzW=^xV6KWsQM
zedcHmd^Df$ckg!JYE!^~R0M&|d>FMvx~1_A(?bgBJ-6E~;-BFBk!;$vbKmy4h7JY9
zE1A64JZ>mrt<Kw4jiwwid7oT5sG-WY__xCCid`)3dUbEad1Be6h(BsLqIWpIVA+9<
zMYn#NZYG<e*`7J(3+YluNjH29`?<k#{Bc&0LIT@TR#luE$3&*oVXjGzWJ}9a^9Uh3
zE<TEjVm{zSata60ELiv6G18=Z+r2tCH?pyfyVHafWkTv;^-9zgsq_-5{6+^WVQS)*
zc0LbXtX~B2Q<{@4iR(wpdQ-HwP6DW?(MXKp!GmONSWPq#8%AE1*>@0%$hca}n~9g&
z@L?~Sh!5AY7$F_gCpPA4E)rcFCk^$+DkDbwq`S5}c^(N(Tb!}kMehx#Y*ECX69u@}
zeslMIL$*)p8h_NP@r*Ft-eWdZ_;8b{T>i#2>TNM8H%WhlrqX5Q3C&qAtX=DrzxGAJ
zlM6V8Y29awGdr<oZMp}{h$+{&>U%KazMo<*!nmuu+>JO}5Z8rwPxJ^p3#rxPgRVYE
zAnH$Nqbo~Xr}tWzVW~Q{Qxv_##|o&(!ZOvDFS#i1oHWjr>}Uvc4lmhZJxCZ~S=t;0
zA|2zaq2p`yAO~7h%44Kx)V+UB?=9tVUcn@qV)>YA4nu{Xu#j(!_qN1k@elg;=5jBM
zZqqKaiu9#{CNT{!N5TdnNqgG6=eJ9b7AJ|YL@%eVbiMtefhl#Z){CN6F&8_`FT#^8
zDh<|r>*hP<Z$eKd4oTv2Y-8Nr$TPUp-<;WZc{P*=+T+p-&hl>s>==8*d~BHxZ0UUl
zX6c;z3MO43L2?@FHz=!)blM-9sQJbh-|NMJJlEAsnn?6*8F`o0EZ3Ame%MDAK5pBO
zVbC6N7nBhqAn^b?&mJrDFUHX_(4CAg&rgm|)sjNIK<Y+^5eO4rySXPyDdGVD8w{k$
z;n7Rb02(~g;M?BQlaZG89&n1?VQm-@1%<4Z7S%VIFi^~HmLa=0%BFjQcxq(<(m^LM
z4)xsTS=+u0MQU`SfAi*ddYzb4SFTtJV<XQ^pOdp%7Q0?&Xkwz6Oa5vw_UG>t4d3&o
znWkU34SThgu5?gOdf`7*GbdtFtI_U^md{c@G{6sp7<gRiDlDh2e%4#>7EH}2ft)po
zWK1Xhs)SM57w`AIKD9ELK{K3fj7F{7TDdxgslP+-_2KxecO%-yZApAitsCDs)zLz+
zO)YOfAnl?>d&iZZ#4-!g50F7-;l3AJ);W`nPkUy+a$_V3+r7h~bMXO~5Us(1y73ea
zher!lR8%M=3WX>p3WX;mJhQjw2LpD8hbPRB-Yf28Z@I{BUK$OLiIv40&@PMyXZ6vP
zJHQ9w*2vNqbzYuRF7x;huq7I-%O+b1CC}>tH;d=}GPiEF`s>G)t5*xXF5Fa9Re^3W
z&*Od4dAF9!X)r$;HJkg&K{M6WpJ-(C<W<q(!xq;1&#2n=<E;oN&i=vxtj^x?Uf?a-
zt;NE2)-zDIR_A*;bZwd}20Z%4qjp+hZO6;WQS1837rD+vj+1#1>sV}K!w;+zyQwXV
zH!jN@aadA4<H`+3ax_Uo{=`A3z`MLrADtDkgEDr1Wq9tD7NTl$NN9UlIBRboAYWan
zXnf2vCg3jMCpn4r@E2E3{&Dif79-4^+1x?M!STe4^HS)mjXAdIgLCQX<*!~20Q~@V
zqI`Bpit`yrqPtbo7rD<8XJ!n5JKF7eG>HHOnNr+^7pQ~TIu#MXk8<ebDtl>O=YC`Z
z^ro);mBEwsqF!6k<b~zs!EZ8zkWJ+Z^W`x@5^cF`P1OLZ0x__$1psS_j7#f^Ptz;|
zPym7x^gRz=>3JRub)-&mkF=~;s(U(kcrPAp9nrR5<W%M#&!`o425N$!?F`$TUCOYo
zxVx%#de5e1&R4r2p-ZF7v}2NcB9pGjHBC5`spaJSsLe*Yd9>nop~%kbkg+JwX6-8Y
z5XaPo;^NVZFt3NQz5OIsI>A}IiC?Wd*~Uwg>dO;xS6w*XZk)Zyx8u{uj_V>5@3G%5
z$FN<ZYlwG$|1}4d?ln@i9ae!+XSx%W0oxtU)jhaAxW`@>R>2aw-4oJ|ObN&;PkL8O
zbA7=6nhmk^MJdsh61=-$u;Uctx|qee?<cucTA#G@cPPi$*((!#*}=!PQcLxbK46v5
znn|W5ZpsQ*^;^w#*W@GvwwtVylIg+f5Sp$5HDPAQ1t?fp?d|Pm*#fXBakL$maL@7@
z0m0J2h`!lkM>I5HhoD7+9iZw87`5W*xo_M9sOlqPeimKzXX@Y9Y0LxJlU)WMvhu1e
z=dr8fiA_u_UEGXYla&Y$#U&2Yb7z7i7wAmfMvx*atS0)8ghqbd_@+NUEUetUiQ}@@
zIy*64c_Zu&kKyp9Q&-gr9ejbF<r`~i=xfC{^aGz7cli%fx-QdiEihuYPAXh%9zCyK
z)>dF9{({B+&VFT<{*B9)Wx2f4oNz69>Q%lgrCVW$#kBP{IFum1O?JCR+F?D}ZXq|}
z>7d;nn^>jq?t0U4N&Y;t_`&9Y6b<6Y&4D~mtHHYQ+S;gpqc3T!G1b05=H~^C+b_yW
z8g!S^1gj{fdA;9heVnzcU=w+9<zditG1FIv$Ur&owJ_Ume}bqa*2ZxCXAkEY&Z)1?
ziI&v*m+8EOeoxhi0k03Lf^U2UcpL3}`Zc{sSA6sc?U11<xL%E00g$_dU{hT{W<(nZ
z9BSF$#e<X;71IHfTUc7UMMU(-)|M9>UkxyzAWV->lbMSv1kK$oezDq=&wlI(e+!T=
zzvro=+u^$0!-o){;xxjxJ_NL)Ivr;!na0s%MAYMPU$_hl5;$qtdX^Wt^!Z%(V$a!X
zc?e5aq?u3*fyo3yrnF02+Uf;jl11aS=;BqXOUUSfC8DX<y314*pH!I?xoDOuz(C&n
z>eC0<OPWW?ULl}Y0W9F`bk#Nmw-o;dV%(LBo5QCe+roitiqo>g1F?2AS(mDUxnFgp
z&K|tG$BL=mE_c1mQ_gf?4@{QcQVsuAvlk+lA*)0Kw{HPcY2KI;p_b{epM@p3I{D6@
zTQMK|WRBbUMUq_v{RmAOXVEEEWl^)7aT#kihR}qGl2V2V`rz`;-0E=I3b#1%6xI1g
z^54Kmn*EWP(uhF#0oyb@F0QRLgnDH#lMs+*Kt6e`MtK7R19{^W(9#w#iB&;X7P201
zM*B~2xDb#g1YuK2fN`Q#%lS|-VUJ^8z-JscC*lDm2hvP45G2q>0H8hPlQw7v*8zx?
zbB>~DHHO~%+-^~}xjBHi0wjL#0<XkWo=tDmCOJ}e^4Dg(_ugyoWD(tfP0gW?&F-kh
z4gLj2@z`_r)ZCNxh*Qw7!pOw*?IU}Hyi-5M%(G0$c2NZ?w`YTNDoCJQiG)+G{|!6;
z1CaiovtfJy$Cxs_r|qwqKLz@7TDEU2t<>~S<w|jX3=3M0&_95a+7mTWeQ>S1^5Zq2
z^s}#DTcLI>0qI7={(si4^G65!@3am5r%v$a{v#42yZbH+fVBB!A@JtFXY4J`W6)(J
zi_dQb*;XYhpoZwsH*Ej(xG8fg_n}3?AI0)tx;OsNi~rdhLr84@U)WwR?)STl1`5Q)
zrYf))!KonyiH7$F=hPTnFvJ{nV9{?@5lfaANu|Kq_M~=Z%03eGU=p2U@BV7AhFtJG
zDk~c4D<%)|^L{$49;T9iz{UBf+n*-RALgqR=jqGVuF(`mk1uRg<D*uWF>8GTgMB1N
z>FEUzQQxk5oabX!vC7c(A@9%PW-L&6SyxM4%R~O9;wIggvrbr?zs}BT^EDqJ^L_q2
zleBTVzw#K=uadmGckhNrM5r`H2fPq2bhbAf^NiO2P$da{UPs_ZTT@M5;x_D{!UmeM
z+O<w7K8P(F@gsk3^sZ~Cek}8T$CXAprjH_$85WYNl9aC@AWSOx*hj5j|0+R#Hq;q|
zSJT6eH1+cv{Q=F9f?t-;o}j#E;YnvG<tKHNR}hj4JD0s}7l(Pr_coq3-43ePv7`G`
zw?eORF7;#jgm8IX{z<Nf6C(t~!&@XIj&h!?P+Piftg%Y*vTp}ZYFj9@06Uk2&cC|N
zkL*i73(LN{7j(upej(w?774z*N-nTVYp8?tSwA3Aj~im*jq_5Vhxr%%7D!~H*vv>8
zPRp5O%RD<)eSBGLSMS!h-!CadA_LkImuK3>rTV9odn&>%NANHioHquOCe8^vl&l)b
z2Ck#X?qkge*M{##u3T|)fq0XPB#9~tfz#!(*L~mhc-9XC^g-b;M+N|8{gzLAB#VIS
zt}&2BDF7hTt#rICA(28UVEQr_#vk9-*@?v<B0^a;sidPrW7Hb_1xT;idNq}WKvlN<
zoBM7&QEmruYL^Dzm5<?dT73z$U0B@(TyZ^yA_{fScL&!}UapmjhD8Wt2AMAa#nej|
zy9oM`g)I&)<QiMO-pJZ$A{)OBnKjI+dedcV&<*@NwHDeUd^pzcw5)4(v6kA*!Wi0x
z3dNXMw%TGB)ho#Pt)2e?VR*yZ^Jhj(v~@>lh5@dX2BpuFh1FL~7-&g{l!HAtmHRL2
zTK%NL8+ADWQ0MtH>Vh8VoD6sGegMvUB*5}7R}izd=tcz~YW4tZPAOu4N4wPO$ABr&
zmC-sj!0|iNsLR-Yqg|t6XhcFnhcBUo265HhtFP+zT%g_iwOSQj^KcAsD+(YqBI;(h
zj)f{y+TJdc3n?xOf0OXM2`4j6QuiMIr1l=}9djO{%BOkGi)5GX<Gc)yyJ*bPWpO`n
z(McB(EML_9Ds6)~_ycvBfx4iwC8PWCyUwf2b?7#udM*_%#K%Wk$9cFl-5W^C5eM4W
z0QX7*3L{X2Zjz8R1Mj_JEw>7IWoQ%!+63Y2qc74^0VH`{qO}5)atL6Jbd%1QSL0~^
zD{Cka^kD(r3hYf*5BXQmez~GIuGz(dO2^OpC5&cFj8zU*ZSU$jr3cYyB3Z`|=uW29
zb*O|r$?%yK$n=Hpv3wRU`&x6aTm*{w94MLWx_o@Jx|5`yJnW_jqU_q*0e?5$sGPx?
zKGx0gc@e2f3EloStE}%UFOOLt?34NtYuoQ%P`4i|bB3z!Q7eVnzqCUALfwc2SypGm
zVqF#OIC}t#CoMg_AGppyu?r&O`s#kPsliFs*w`qms8}=Or6_s!6flOr8}#auppVd?
z@jY&|<k{IPNmu#c9Ixd-Z>E!J<-Ee`g&A+QUStp~EqP|m%a5ygE)wIepBm;2`A|@6
zliG8)b+J&c^{4%tN4Pb1{0HM<U&UdP7<|4M!Y56VS(7%dml#m?qZiSg3Z)M*krM&R
zE-fZk<sTZYf}leLbc0-F>#?F6ZS5--+Ek}WuVmvH)VM(!d5!wQl=AXL&B@L@l2zpR
zE(Tv+BiS&S`i|ZC*VCD4L+CE0zG}jblZj+z*;d!Iclg$4e;qapt6K*DM+YKaz)Z>H
ztj7H)A~I4nJ*?CksRBepfbAiTB2#=YP7m@naBy%UQ&aUHyE91}<61}?WKGX-ib~Js
z9{sHS4Y&7br;VbMXWnlm$F&lPdbfePy_PkthK#M^44Y5IYl$yQ-JTlU4hUwszleRk
zv^%%dv%Miu-n2HP>LQO0b&V0H>#JAhG}a99@W(ccqHAwXI416Dl5FYfoGfw*U-6lL
z!XBE<Wo84_TQQZ-!vcPzV3=BX3kxxtH2S-1?1~=L&MmJzFlCkqmp$s9EOe+DF+qSB
zF86(T415-d*PtHCP)w8twh8b<(Y=g7qZI%+ScjV&G_O@3=?&=@f<EDD)a9A@>F;j<
ztBE>qDgZMEEF6*LLzz&T_TT;A<eoj7>zAP)5{D`ba^E$j)e*T2k90b~cjvz*W_jia
z&RY^xFR7Ynj`bUd)=tk?m+Gd}gbaLJyXJpFs^L+;Sns+!L+P2KI2=nDabPm!y(d{`
zR=<*OD<pw0W|T$$c<SR1R22yWqdHe4!Xbs6mM8MvyLTO7S8PV}-`;-wF&{Q9fr*J}
z;V5?LB?Pa|UQ1GOPO7DRF&sFplFzkM2&>zbez>I*R3*4P!Qubqe0EUij%Ufn->~BR
zIdkQ+jVuReOr7F4M1z^ii1{6$mxJ~Tn-Kc*r>kFArHQyRyLx5ZtdWfNCqoqw@88WB
zfQ54{oaU5#fca_f`u_o3LEGbhZ;=Jq#U*IWYLPItbRSO1$f*AQ(^al&-FLruo38rw
z<=aNoj-#6&Kr^@0A20JisqFrP%=;hQ$e!Y^*DiFh|7|<S^*7Ks$(qzF5x2a-pLUfJ
zSWdb&@TqP8vd)`C75v$#*fNe!R2k?_h<|QA|ECb^f1{!Pzj7VWgwHUi=)H=J;JYuF
z`O1aCv?!n!rDWP&yWTP4Jyd(!yQ+jXhY3@}tuR<!ZF8|Z$`=wB%g65}u+mPiOS6{T
zO>`D|S(U?}&w-+#>g*ep9_3`NebzbG-=ZxNp|54_o;B87DtUN2vpk6omYr_m0tt8#
z%XrKiNL^SPFU0Ql!U(s+U<6Pf2XnAyljwkWGd&^Q0Dq9nN|vC+xcz$e*>vus2YMA7
zT5gE<M)wlOVs)GIQC4@jvixIh%n`(%TDQcU?wM;oCEu?Nl$})n#su~E_XFHEY#hs2
z($Q-%TSSWxiBh+3wd88TI{tmRhPz+De;K)9OculEpdo&tE%Va+=oq10;XZXV@nFmS
z5_V8qWV2TCg_dG6o$`hd{p?oOA^-9C$7*L-W>G0U*lEqUwi5<B#JWj3r~KqIVV-AA
zYNiBI4)~HQWe9uivyV_zm3Va-o39KRY&JUDJ;0iz+5HI4*j9F$G|k!;n@3+0j*-K3
z;%qNkcjm%jd1x|klyh9NeL9X6r^tLMn%~V+7XbLkz%?I=@r6Zq=r<j$w4=tu=#fkP
zBeGFW(~P_^M79&oSgLN9H~nU-(S5xpb#u+=joOLG_J)aD9Z8!-HY{x;9c@Z4j7ddY
z9Cy?p8ylpL9^dR};+06Z{RPT3@rfMl76S5At1}`Wr<Dpv*Nw!Ek#9|y6yB<Mhs+I}
zlXgi_z7<yDNizO*AwuDA71l?tXN|odd+Dh-Us=Y$*yl$cyEXfP)ZNzIpyy*#bRJ6M
zdFwcq2wo7ZYn~p)RF-y5vz6hmBgbuVWd=><$1U^H0meidp`Z_CU~n+2_-|xBT9S}j
zC>XSQ<DeVL={&X{fsRmguLyvp80a2%(DiVWnp)n<inCV|<*|6mQ+v!fqUX*9ucED*
z>}l{NZORAW5AEHf+t|@wxd#Bl<VEs~W^YJe5Pr6IoH4NBgRB{`vWk&Ah+2{wM-JrW
zsNY*_#;gqvW^pL1rE4}dSdgZROJqpDMqu0zd-V6OVu>Fzdu$I#IcaiH*m#mz@s~MF
zwp9<ZC(i14?H!P^O5)xEmh@}5^=(H&k@~CIr5QclQc(GWQlb{stE&40IUd5RZmTC+
z?8rNr%QKcGeN`qdC7%9m(QEHyAf+VXhB!W><3l72ZNGwE<=bLnXA>Nk&7ij*yvkq(
zUSNppM!y|_;WGPy9Q5BCfjyx2wy(%c86b4D&KfUdO$AzTo2I8z=jIGS*QTqhE4sM^
zjo5)y5*Z!+G*c-Vltvt&a@X|f5}!O}{+aR&CK1tV6jT<GqOFjbao`F!(3Yq~xqa2s
z=xHyaam+e&hV#v0#C2z|znv{U^-?1X+mzCoH%QSaOkUWZ=onEUShCwoWYa-@rbFij
z9>UrfQgh9PR}P1p_U;gGqQb;ZD!zDMx`=c0{iTOWMBZAJ#6H32fg5Fbgu+IifUxLp
zUwioM+4Y^B9bk78_?&IkfwxOiZ>$=v45W*nEJVh2lO~1*XA4_u6&PX@KG?r$S`-gn
zZ;%e5YS~@v#3dle0GNFOxW8Qq_kutZ6liHgYoP!?OxSmFfhN^kw{AhB05DM^Jr0J{
z0|Enop4(UHkXNo5^12sY$9O}eWw&ifw@Ub2<hD)w;ruy=NV18T^7dSyAl!^hrFZ06
z=XM|wCVdd)wCc#<ks0~CV}&`{oc<W9_be+Ep?a|V{nX22Z!~1Tk~!E<;`r@~{#M?u
z8@#}G(VgGba|2^|y_1<){F7G>`x2){H?v31QinXIyRx<ON+`W%;f3&464fP)cBhUg
zK$Q|BawT>y&|fFO!kWNm$fD&S($~t|{6HYtyM0Aplq~22@aO^X)2wm3H(BXq)bI}5
zMVwC5@%KY}`(2T<1m4$}00@6SA6r`UWAxb%#QmoD<N+fVht&FvC3PRW_^0LrKNo?;
zoAd51yq6gi9Z5E=6Mc9-l$IQ(lirS~gLQGe`l)QjIcl82+SfPZKbWldWM3nwVm`bp
z%IRaNui3KMCA9>TXuKG?Z`7J+B%-GwJ+k!0!=1seTV}WS``SZzOZ<^C*5j9+k`dk`
zk>786PffbSN%=(*9Wv9jv6)_c5!b7#!D5k5Uhn2+Wajii2T|pO6{-O8x$oy}>=<i$
z4(}&3%@Jz@3GGC;*OFqY02ie52_)NfpS%Hnj<T_Aua-kF3nRbJxEZIM8<-&UK!9k7
zHX8%>ebcOMQhzQm<2{ZhoguVhT4Fh(Rx1R0kpQavh}5-H+}pqF1Db@uI~?FGHt=!_
zpYx{9RRS7WAmAdx=rh&x^u+hN1oqd*BD)jsqfH3{8sU0C!Pw7~6cuwy>B3jh>lrpV
z79K-HYPO+#Xx^gj48OwO^U1I7>;cH4d4wV9j~?jL=~jOr&18mbMBE}??<gLR(C!3R
z)>7)Pi#Bw_GM=x0m(aSDFsI(K5$~!={u(!0b}w~Hv|)0uHm}qgte!1_-SS-Xi;frX
zV~E~9NxpvNTQ02i-I-|7Ol?o_8Xf*a(=Wc^BQkuJ8sc?puK3i1%X^6^t8BbGprztf
zbMls#vRp-?_a$@a?|dlb#DiR*Rx!_N@PR0a-(dHur{>ThV42*Jh`qf>tqy14EfjQ*
z0>IVi9wopr?2or!1A-64Rdky-;Gm!_la-4L-K+=nYkM%b0swLXKe3(uavKNeH~C{$
zjw=Hn%BI|c8XKkPfL;n-79y<Qo^5hH95ZWgYcm+le+@?aHGlk@$;x}^ZW*tOQ>*df
z0G3pbi1l@=j?PZ=scI2$#beN!H77CZabbI*M0_=U)9t&ZgDL)2+qU!l!r$Z)o%-5T
zS5NGZd2l_77b%XqrX9!1c$_**)=({nd6IQ$65!P|bhGxCzy%fEmk=u}D+4R*H6I}T
zrg)v2!#5{1=UaxDv7sMpO4HX^v)Uf@Qcmj~Zlu-{J)bBZT5taq;uEZdF|+FM!U012
zdSu)>{+)}A!op}LaKurU@#C_i`<tT<_B?Cvf#s^W{S)pyYyBq{BTun?<RY2*%m>mE
zazUfz6K>)Mc(VVlBkDg0JO3XZ*0#EVS>0m6H)wiMhwdqBBfVYuqP$H06P4*Gbd!#(
z6%6wc8l5xhefhtLb3yy_{0^LGYG!Rs-q<(|jj#EFySfyBfDi5}$#C?~%cjQvZCQCA
bg>n1pZ)2Kte@*auEhP6;=}E4X!Mpzj`ta<k

literal 0
HcmV?d00001

diff --git a/tests/tools/_images/cohorttracker_adata_mini_step2_expected.png b/tests/tools/_images/cohorttracker_adata_mini_step2_expected.png
deleted file mode 100644
index 853042a888893fa9ef951030b8c086513b5602b3..0000000000000000000000000000000000000000
GIT binary patch
literal 0
HcmV?d00001

literal 24087
zcmdSBWn3KJx+UC5kPt!$1lJHexI2LW3GQye-Jx+v&;*O&4#C~sf&~xm?(S~0JO6Xf
z%sF%J{O*}M?}zt8LsL|DRqfjQdDgSmdg`ORtOWW)f`<?Y1pV!s*NPAbJa`M|i-H7x
zO_NPEfnPjMV(LyxwkA%{cMir7nRiZhR<=%7<_2Ug#tx3=wl-{x+>ER*$;_Ob>>PQS
zn5_T41EZ~jDHGZCT^slkRJ%7Cju6NrS=bBC2O3HNf#~?Zef>(=HDPxa>WS@q*LgTz
z`66ETy)X(ndAa`!Ww=*gp1xA%eo*o}%^~R-X2kID@@s(std^g%a<4yQ$@7RID1HwY
z2_C^C`!4*@U)X)Je78vmKJSZvtr5bpn;^+G3g7YUzNZQKjz`mRB7WqoO%;Xz6A=_K
zg}Pb@Ie5n|fC>>oA@_%W3jv$SvERV^g5RmaPa&|~9(;w6f!{Qr5&!J`{sjc=`sCpY
zIB<Uc@?`(?(d{jG?e^K(#Lu6fs6~VQJ}IcH2M?M#IpZ%qYDe+~n;je-b#wy37mBN^
z$K**pX(A8$H;zMl#@;;i_30jg;IC0*6#MXW3AW&ew#!?r{@I;PVuXUXM_S;x32#c^
znoY`%21$`?Y9EpJvs`Qn^pdQt!ggD~bf(tP`C!M&gZOV;*FO%4;!c2X+vNjFS@Gpn
zJzEWp?&ug<;-Bd{(0bm-uM5uGHB}^yuaKm(9&=1Zn|8k6pzug{Ic`%_5MFKEXTc?H
z>h>Tjlz8WJ1z+Hk=ic+M(K}!()Y`RkHc;>Wdhp6$pRp(1s>d@o!pgdwsmU8-Vb~Cx
zKu$<HK;wJ<Jy*w!mzoEzzZqxY6pFBm3n!`9$)H%({q_94thIH)^5y*98CTRx4F?WM
zEZi<dg*TVsMZ_3Vn{S<7?H=s1JdAG-Cg1gt&pX_9#30yDLApLbiRk1rK~WXOjJ)<1
zUfvASYr0n0JdvsAT5^9!*PY319=QCD-x=?`@(?k?6EnW#)7Mi+`J4*>o5g)IglW!+
zAhjwjRJ;?swWa0nRu>qWO;P1U*BB!tg5WDes%6ViAw>;mnhg+U!;VKJt_$z;6ymDi
za2h_#yrIC8+v{|75USKfxXKoUC=<`TwmxIbW&5g-GrBQ-Zc~imWu)WL+PI(7HT72J
zjAVotga7Apyp>SfX}hY5_-P4t?qi#@tp%$5NU@7EB6Q(NIW(4?+wi00=rl!3aJ9KD
zg8h|DlLxksJfaNW=mbo)67JhnaSrODEV|3`Aj&H#wfduCO;p*E@+O8yF=#p5T-ZA}
zIJ|lDhK!#-;TgRuv-6g6rPbtzNvn?W@o2Z>wZR-E>Lk}CR0z^zqG0j#ce_&+H<trG
zROICEyS`B?X=rqn4G8vx6Z1jv_*`BNXUQS!|4w;zvNal#>~j}eU(X+iPU!NsMU-6@
zcP9KZpKgaug?{fqbMZMJyRP&6iS_8Jsdp3{J1_jf_1PF&Iqs6kOzX1Q`A>@-9IVk!
zMH*c;G#!0BNYtkfwaKfO5VL`o_X0Z285Ouo?knFR2e;?*N~)^j?(V!0N=nKf85t~Y
zM@EWDO2oUFGVvX~v21V`M?Fk#do}A9ch~!u8|e`-d{9QA+Y_w2ySw$1`>X9DLw16<
zQc~%enSpQ!NWW_wO*_H}k()~lJ15G`u}sQh3w7#By{=uZPPf-Xb(Hes5*pO=K9y+o
zf?2*iSc|6Ki(am*V=xdsr5mh=KQyLK+t+TA+g`P9R?j=@KWoLGmALivK32!VgS-$K
z+=C~iqysH}$Zkhn)Ec!;w7iLhEiT(*1?hTC-fbNntcD$-?u!eaV?}xc%Mn81ngty_
zJt2q7on<z&yq#C}%N?ns7K7WTFJ8Q`+nFeFyo%wl#CiAbU3gHtMMJ)q%BQVXE#|eo
zPNyqo6=W{BlvB%(6}fnjimdzCWN(Oip5&HpwxXYA&m_jxeAW0(W5bi^6whf><qX>&
zYi6t&LX42+0<jlsU$r6SjSO(V6E4kZ1ssw2Dv%<@)z0^^n4>v|8ei|ZQh#ryg(G8S
z{d#iZ6o^6EUuiuZ92S;QUS6&nzuXydd@?Lo=6Yywwm&}*t>>+=A9J4}f6G|t;p)nL
zmp^D$tp0Mn-lY3G?aJC3FNtQ@T4`kgBEfvr#kmiPm$DxJlrxkFm*W{}7L_A4L{|rC
z5pxi0VB>DWn$x0mE9%;*=^IYEZVN>kwQqWZT$R>f)qa>@8Zu|um#O6(nI^-fxJ<N<
zKUN#=e^#C2)h~-b;Xa{-SU8>=PLvTr?%!i9C)y!<bA`6rl<`I(N5n|h2^F<dzk2}*
zoOSL&2;%=J7ol+RBDBpUKh5DDgK>h5TpM<^TwJ)3lI@Amxh^oRN~U)aA3RXd)!j2w
z($x)@&7xpu$F7_0PvCLfI$c~|Rw}x+MJML{etEoZST&uKn`<zbCia4giZ|V95K>q2
zu&(ceX(!x8Tg2=1^Ngzv7rOJJcJ1#n?Q{F`+x!O~o6vVS>0G=6M&(EJ^(s1*a-`>j
zJE-yYs~6StdT8hGcwCfZu=XRX$0l=oinXLKP8%Xpw5ZchQ8PU|)xO(Y2w6mADVDG4
zjO@)JQVRObnq)AWi>GJBtL|MV4UyuepgpFx+fY*PM*DTYRWxBi7w1~!Z%3Tf3kM4g
zObsVbUFwKK+JA~-2)w?(yZ!j_BRMB$)URJ}RP)ta8qVjWq@^Fddi5%lM4(oel&Ssz
zzBLfNb$Iw&h2{8|*7;Um(a}!Xz)*n}SBlNk1%dOK<<KE91gj3oZly*uCok(W8;de1
zS9O0eHEL_p5BHv2jQcmIi=>GfY4T*2DyR?inua#6x<?|FGYhTkbOv*)Xf;n=8jM$`
zaTFS@zbIK^FKMQgPVc&?;VwRO$ACICx_13ZjxBNn2CKsYuylLM-&FX?mTRv%P4MT>
zpTGM0oE62EB)c+v1ODw4|BtEAp*zF3NX()lkb2F=#Tj2DLA2m}@zPnPkWn<){YI<Y
z-Wm007d9aT9v<Ox?Ya+a2C%^n^}msP|1}=@G2sFC{d*MaM*p<7PGI#T3AXSdPS&5J
z#=$+Gzs1U!)iDEHz$mJ2hZe<jHw1=<TG0wpJe5ClH020RN;|N@kc7f>%kR&7uW!2K
z7ZDVHzNp4=yt)=~(Z~ViUjm+g9F70%+|b~3EIXhZGV~1?HWuhAI9Ng5#~L0-%+~z3
zH5B#q-Zt(#E_{?77T(<hFB)DpwRqQFRT;h64!_u%DVf^E9#Ji2Ve;GQ{z^4Rv3j8;
z@_e@*;o!F!E!Svi*#=@r_}=p=u9!}~Q;LaJG(HzuHNp!{Q&QRh9`v(et{=qbVy9%k
z`>;Z0C2drA^i1{^g(-Z-m%6=J^@$R1HB-j7XjoLs$eH}0=eAUgWZ#>d<$grY;TC^W
z(&(sfoz{qk{%G-jI_<PnQ^++avL7~SvKo^bgYo$0SAQCoOr5C6P2jLN<b$7ISAYLn
z<+RQ6!D4IwmTJNmJ*AebdW!sRnNG8^-_$>d1wQ;Np0wVf=V{Ny+O+B!NDG*YucKJ9
zQ?ex3=`<)S$)btgm;F(e#T{L#dl&jK*m9{sA!?`lX%x|P>kWmFk|hTle?FU^mV{x+
z^h8{87qx_<dQw?3*qMLYnan+`u*F-Nlb-M>oO%Dig)@m)QljZeQb<kC$8&Lw$I-aV
z*UdO{(mE26?QORT0@ZmsL?(4%D%6jU)|j?qe2kEZJ^kG3&WT;kXTj;1RU9~<G4)Q=
zMcF!>VUkY7L9b>8+dL><d&!c*sW-S+l+|H3V^Cu(rw%Tr4i1erJtHTXn|MF7<#3cs
zfAHG^)!}?eT+3sGj)N{Tx5YbL8QY;(KD16;s-RuE2?tZAmQ?a&a!>~x>|m(YUi1y$
zl<TD*F)Q7tU+mb+InL=7Sxo6WC1%PN1ebY+a`ueZlE=R-XR^y<svE4wsb(=K-A(7I
zJe^4{HfHe0GO@l0zVb~h|H?lkCAXaJn3V`6i*VWT6VDKz{iQb-HQYI&`#G#?Eb*Zu
z!!qBA#S}4S8eB9n1Qxx@&4+()Iz1v4oGIad{`|v_A1|Ihed@X##-(0rgu1u42YIR2
znBcNIRpPiV3m}BJnAii=yhxD5VgVvBC{f4bad^?*-aaxi0!!reZpQ~lN6#rJ{7|u|
z<TW%t_L&%tWGmDkb`Zw&ndo|*s&LxO1a}#w8Fq&ECh|r5`ufuQK;vRfYacfs_i}}T
zOE|+`hkvq4*bD9vgjs{ZGyYrPpoMvPGRdoVGh>x#kWm~gJ>z?aNP67;S!Mz~E)0k_
zZ(>)V5<a92xza>)<_W~2z+~-*ZN`*=O?h%l>Y((FD+1(Mrltw$#UY&<V<X`i4cuN5
zB%eR2zjTKka#a<oFWXcHmt7L$YCcZ)$iy)HcIFvoXV0N191Vx+@y`MJLWfn6gn46P
zcp4g-g)b5$?<Px)0kj%S6-E9O60%)E%fMi~)*HJsSr#-hqWtySH}lC-N@HVV={WX}
zlu}W}mg9_L1zMeGO!dtGBMzFed){5H?@U#M++LqAudjCok$R&vgG^a{dp3Q!8m$LQ
zI!c++a0?*0wg9+8%;)ksE>5z_X4Z8zQmyd7+B_-BB=bl`m5w^E`y;xU81+03WJNWy
zWMhM|aX<j2`S%#{;cxb&-N%vZWfvPZoQb@jg2X$E#}eg-e~rs7VwauSnw;G+5Eam*
z`GlajBaP2uGNq>lb199NFdHI6&LvLGobd4VO$Sl$4tw11q9Z?5^@ky+yQi63{SxMG
z|DGJp=G^)4EHc5sRoxURzy(59u!8L~bN0+O%Nnw9IT3$2Si?0nkXireu6&pNF0t_$
zxKEjFk45snx=s9aX0${kqofR|uI8?FDIomdc`_*Bwiau;RdINFO3BL`hfM1ExvlNB
zcsS8pb#**KE}Iu@Y~L^`w&pQ+Ky2|33}p2<vr;Y43}`n<t*=kGyS+xR8<ycZTnb@Q
z1OU8`Yrgk)s`Fc3TH23CM-BkiBl&h2J%9duDIw0ZFRndbql{>v0B3u3r&fx%h|c-O
z&FUI)$xqX<m_L~yFIHNVB{t9?&GW;@GqltO=o93gV}*2)48?b4)5Xbd1ua%>`kY`f
zzs~oe2a9%4albVu?&0`O3_Q_GCcB$UaP{~jbf1U=I$2`EencicFF3(Id|$9hf-NpB
zOz(nCB9X3`s+v<gu`@Anvm1$*N6VsmEcKS;R_pce7Ooh(B1oyx({+`y>)tnn&Aoz9
zRyzK`Q#T13w$b7SviRKxlP9)=N$(4u*O31p-aZab5pd*gHYl?{<6>&6{CYDbS^3Ne
z7A9xmMZx(mrbsVgGcJASfuIxHj+wFl`0?X&LPGCaU;O5D6_cLl;RBbt#F!X7FlPtc
z`j*;5Cadi!v?{GyK(T`%c;OpH$OUhAJ_{Xg^6`O9VK7Nu_Z!YZ0OSB4#1Ogx_o?l8
zjN7;yyE7Zkw{C9v;d$-m>r2F8hT><`bUly_tf9xgHv|Z+(@q^`W%363RLwGS9?4`@
zE`$#%A|W%FB%&hngsLP$&3jAQdltA7@2n}Gt!Zj*dZ7$@7?zGT=yL0M>RTd}TMaB8
z!3(z^6!$|`l%tO9g62ll`w|LsQZj-tO5jq}tvMa?x*;(mIi!9fe9ZfH55d?RNRPiy
z$fryHr7q?VC3~lPH=@~_P{>kkgIr|DRT;ccS*bIr<@xlP-F$4TRZtIF?3gukBn}02
zZa2#D$$H>tyO*4V67-)SF>|-7;uRsQrwx{ubl13dla?<!AGNQH<87>rf1sJy&t;&J
zRgOx^f+sI5?0E>ey*gb6QDX6Ts;JZ6OzZ<R9DUZ<tqK;rM|h8rK-u`K`8@)U-2%kp
zd~Y>1h~zmveaPwdIA3FXR%Yf<g{7LLloaaY$DrQC^=oPV8Gwe0fq_g0LTYw4O-BT&
zT(Y1S5d}iJM1;xN&g3IJW{8}eTy9=oS9kYwDylWfxo8s*`qBb|FhFuoQ_W$H39}MN
z?21zAtS&M3a7PSUEO73$D*LLY(_eY>@>7@!`k9I)$1vXyh9m;6P)@W?x=#}hS9s!D
z%Z^BGt86i0<);r*pmy8@k)NI#a=hiazTmya6%XiSt#)R=vvChqMFXqltHaqhwb_F=
z(vg$Dia8SxMr)L;Szxoer_?94W9lLS&i+zThQsLq-Nep}ouJ~=hd4<=o`jzgL0{c|
z7?SRV2UASUhG^ymAQ)f1<UVpt&l*;Z2@AEHPlP}PB26f|E+;*mXksNpa$3$r*dM?+
zEWPT<nkXi{e!aUR+?kmfJV8(H<Mjd4@j{(P1nj{ej{!u2&t~#SMMWiBnJz6oU87hZ
z&fD7?7Nv8PvJ|zoLnp1MK$Q=64Hc(=&U<lD%O;S$c;TNU-~rZysG{Oi+lEt|5L~8?
zzCQH$z7D<nD>YC!lf1-1V>9Vpn6_<l4t*^h&l!dcSdIGAx@`eh9=A&nRqRSM3`q<x
zsBLR&AKBxp%D4yo@)2-c3{0CG?9*x8+vn0%AO$rlDE4%#Q_Y@&OkC*v6jG@3zr2K4
zAE^8*e(k>kH~)#6<5{)w9e*CC*_yezUk5uO&zTdV<eCW>G+xL%73PA5<@G#?CVyES
z0O&Z`%n8U7yeG|T&+`7zbSR5Y|C`7+G76i`{Y^eNhDfmG2cHm^mp2BH;4j%96A=9}
zjTo;^%s*7&xc_>Sqw|BGlPgdD@A=1n93RCUP~coJL=?xP5SQI0tgd~sa=>*$6kc7E
zTJYsM3?&HY5P(RjkWgyrW%3$o_<7X^olzdF#a1JC8Wu48EZ)cJ_VzA2B$#^i!f6NT
z6DNPbyLo<gL2J~v2@xJ4B-yiGIbE|qH;aN2k9h|k`=qSyO2H8oS-;V_y>DjQdOWgs
zhj(15RDbE#ONk7Jf?tB-RrlIKXg9P^-TjVB8pXcP5{CkUii(Q&C=jiZklpOb+qZAg
z3#B9_pHot{cSWLRFh3*xp?~TZF0s0J52w%HFg%z!n!S3_ar<QoE1q`hf;|#BIz@4t
zU`K9uPohVmXYdl5a?y#)@@$tnW!|aYki@(^(3eoJQ&S5NlBSSs_C57Hv~1p}kqt}s
zIcpS>BT}nOwEH_mj3htXnHRZiY6`Wxwyx#)Z+iHm-`DPW2wx_h54s{Qj2~a3ol*K%
zpK*jh02KKY5i#)WlpIc28&yk&Tt(t;-L*QP4MDzfTL;4b)21W6FKi@fMso2SL&^C<
zs$0=S-ua>j>H2ufiIgA$LWgmKD>@~CZJPw|Yx`3Py>k3P`@&$Y&}i)ZyD9>>;_t*6
zc71-ziKrCoikJ@DSuvl6vUzK7K3r$p3K97Z6)-83oO^{n@c4ZR6`tQyg-Bn5;?8Ip
z>lh`aR^JCqvUnx<=1x+d-XCwjLEgy8Nn|LNxa?{|OnPHlLh;#t{raW9^6SgV-fTQT
zAON>5yEaH>q~+&7@?BXm0_b;Rtbh=UTK0vYU{YT^7vGf4Y;ELw1mvgBp0#7i@xwK5
zjpk{T7(CdoKYk4EohKl!`X?At<*&{t?9p~!%k1GuKk$Iq*-m0X!qn)e1XM)ovU<$D
zY*}lY3(?O~(KaBy?RHDO0|~XOcf<s9CKpd2p>ESmwX#jRO7>`b;)@tqaMKT_wepU(
zE1$8z>&U;X_1$3O5aXG`J68PdTH%C9M&5SXLkp+B@jEqNPY-(IG+XO@vOQijcT6rC
zXgBY5ez;e+UkU)bqN3u(#d5?XAn_W!rKzx%JHiY>(%6}<iU=Zc4+iv<yEULHI{>hs
zXz~#fx;a3YuCiT+6><H1`z!6ZdM`LQB0EVudhzQhvAc`bjUe67K8KfO)A<mydJ)>G
z1MF44`v8wTUH=!l+7rq;3xubW81JNv_ry-KPWOd{n>N4iyCX+-AXF6EAj+g;yxPSa
zS+M-XCD);;B@%*}jCuB(L_YEIRPFX^-Z$gONax=B3$9AglclNE?~o_>8t2JkCMHzi
z($)*A<_y5;E32x0O-c%nh`>O8OtdiPu}=wzR{#JFneo7#E*c+56%G1rQ8DAt!vKN(
zcr7M2YRP6Z!(Fb);NFnr?~e$s*F}M&KaL}$xR|-K)u85piz=1hhM_hcqYhumo|j|N
zlBlQ8SOSju*>2X>@AdEdg?#LGXIkvp^5!#(y$KLUa~+QYTW$zJa&2-{DcX%k#WYi{
zYys3#nScu(`R-d~1W_CP6$S+H<eu-2t6q8hO+jAXru0R&K;wI?<MdzGN8%3mgdv-~
zH-Tm8M9$Mf#~!I)MIj>E`w1>M%>XoLl$oI8<KxeG9`{Y`x&Zhp-sFAz+l(C%3F&KG
zoO{=8uJ<|Vm!P1a<pF{)q>y$8g<I)Z)&N?iOzxy@0fMA@%I|$ern607j%LfY8H&0t
z;T$IwW<;MJ?A|@9^{r(?un4zBY^JaxOEnx*^4Kz!Wgvbty2ylDZI3jk(`m@q)5hlT
z0`jcsrmkf-uBQz0`^ujuD#uCJ)wv$IDy6pd5y5~w2rVtj#kgR+F6sB3%TL36;Jv0_
zDeuHftiG>(>tC>^878=6BJ1P9$H8NH7+Q~@qfebbdg<zqE;4U}$5A7s7=N9fo?JFv
z_1#o=jgQ8^T9w=S=>oWVxAjEm#&8w}#19$6<mzPe5nv5v6L}Z6w|laZy|>JKPBs)I
zB_+2`VN86l2^&DGaQZz_3`7F%Cux=Y+uKrZZZ+S|#A&{CWHIOx@BLJC{?sj+Jg#S6
z_PJ@v#zKw~%Wj#_RV=K-ZlMX=`a;RiYckY$g^4sQ<K?%Mc+(}ErxNDf+jwV8X2o7I
zva{rD=f4}9#?mpUj3%^i-1(7xwe%%K0{hxQfVbihQzpIbI4s5sN6@Ex)2{s8We*85
z!t1$w|N7wbeQTM{^JnIcU3!N>?lLo?w#}M4zBvccXpMTXAO%R1z(@E$0hf)>>qG+q
zl%XLAMvw4W5erm%x+Cdh`Q2E{Ob0qFDxMM%^{yw6-^+~wWsRxn=4)1x`@{G?@qr|P
z&5;}!bB{Z72e`>Lpk=&~lw8WmO8V*sS^BDhPW|idci!OX+E>2>T^CEgZ3{=A7uy0=
z?meMyT`OLb_i2^ZjB|3<b1wW8`;ag~bs2R!hyKkn7BNfGbTjK)pW~(`ljh0d-*{C~
z0)v-gij~KSdKeE-$P2aT>TaK7@N#1MwJdN`+O78|&VU<7#OLw|pwTQFiTzZH+hm6X
zC3aF%CUNERJ2||w0o$WpK1RgyVzkPsuV_ueG~LJZ&M(Yc@94KEUcL;SnF)#Cpet-N
zfl@Xb_e8OP5)W`<ngvl}Avip@t$M+0*{KTtnab5`<k8kzD1#?)I*=IVckozL(sqD%
zL?Lf;F3`|Mm4~U5GE)DD1O7EF_fI(KpVZI3J;i1jv4XLq8>qq`*0K=wj$x_>Mg@LC
zhrqpmrRHf$5015@N%`*zm`h)u8rTpl0yXl<Qrwe4x?q)itsU?e7A+@4J)iUOnH!9t
zByX#D*Xg6J^ZvgjRsPeZV|&o*lSA%2rLeKC{fEhap;c(MGu%h_Mj5t;pF*9iyW*>y
zsdy(s>K{_r+VXdI((*v0<#v9Zw8@@cNlp30)#2<}*=Y&xY0i-n_iRk4Zm`pF8et!f
za9wk2=uYc<6lk{J;UX(2T4&w|X?=WnWYn)S8;o{JDO^NRRUhSJ+p~0=TnfR%!~ZqI
za_E1(U1CU`Ui`_&>es?{xl-7k&pOUry_W~BP52{`JXOd1>Q5~9vtr^3&+hRCV{wth
zFt7VfSU=@5ZhAcvVfT?Gh6@f3Cgb23jn#Q*ui|FLkuN#0?yAwpvz_*V|MF^4*R{qx
zw*CeZ5GgK5fYteVW5Inl>DGnxoMEodD+GGHw`Le-v)wSrBE2yi#dMamq8A^|S}G+a
z=#wImzDs>#Bh~rDcJ7`Q%a_?lphefF{@jqtN5lEEupDujM7x8>coZb2z(sKWThF)o
zABSfh>1~xqi^s_B3(IqM{kE&i{Sk7rzlt}W+%fUQn>NmNf>EwE-@enj(6_{~i2T$<
z>B4^>4`Lrx7mBZZIsSGy<Tumv&%we2?%Q`0MZ~;eag$o%EKrZfvF)Qb%sEVQ^Akzv
z_+sHMmJ!%rl{EbZ4J@7bz|A1p!W37C{Fv1^iQHU}sL9OC8tgyT_+w2a;W+NC*R+Cl
zX{qPEIi%7wdg!_#nP{5=ULe*}=>=+mmqu)ooTMymo$U9;qVYkNwh(haz7&-_`8<Nc
z&5z!@C}$4KVa^h+UbmQ5E*DPgsGV2CNEBLSl6Y9S-DUH8S>+v`@1aU4_6#-on7*q$
z(Zd;1xJ%2+v7FWfEiElZrl#YTY-Yd7zJC2mG|(A}-wMiEI2tCV^^#5!3Lv3eT)0xn
zCZGr#8&kp%7uYf}p|0z(%h9<#@4Ad$>$E8%A_5Ok*2>BVdh6ay%{L$w4(F*c)k63C
z)%3iadFo!Y@$e#gAaIK#J4oiiZQw1b6zN=rRLP%4Oi$b|kv-)LWS>wo`b@1E7Khb@
zZ~lw6fM+=QoM#t7!1=^%S;9h>r^56@K-ZOJ^TF?2{HMgBeI};MgL<vM3KB(VTz6F~
z1qC3!L4xyAZUq@Faqg9;$ok4w^ttl{Z%2h}mh%o94qV2$r(T#wFBFd2IV#Khd4ce#
z6U`4VFE8KR_7Q8c(wZ=l54tuiCzMfLeesj&CJc~(>Eqi#WECw?RZ>?+Y2Ges3Li9k
z!N3p}7e|2f2u~Qq)3?&nMj#&~@H#(i(3Nv`=CG)UD=lTEl8I{r02CD+JtI3im_)$+
zH=t&w%FP{f>Zsb>a<cDL-1oa2lI~1MCo@LD1$Kg^>v?O(^ix+)LX3j6_Vn6P8;piU
zMXs(sch7)m4;PmEFvCbRt|}23Pu)lfBV|^omXK_!6q@Z^eF57&(PY3WzSwt;gWXw+
zGK3V>oa*Z)KDQrGZfIw~R~5zfWHk~*vZ{KUYlT-y>-D-Q_SSVPn7-e?V=pZ&ZOqiL
zK>$g_Y-3{+J<<uwX}r#K7S6M4q9R+390ht!NdTr3RcMAqN8<vS)D+;@i>s@r#Ke7m
z7=piMY8+{4X<J%b{Qxd!Wn;r<F+c=U0RRGQ-_{_E<gz}1lrV8{)?6p@$E0SRQ<)_~
z=-|w7<<A#!Nm7vQq=U8W$&&A1n3i4X!`bE!EZDuHoZHh(`>-XBGQ(~$Ge9Ia@UpPO
zpki=LZ-3+d{FbB5rs$r(cfc<>sqxGET?DrqpC;z(&R+NJkUW`uKE+?iT4fUs`7yu9
zn2?~^@nvA4*p25B2O)+hP40D{4Cixy9{9=DdOz6G6#gcm3ng25J%k>l`azX=gp^T>
zlDjq0%F4>c*#33LGa=$IaZyRhhh1G=R&({Sjb7JB0B-A7O~b@R9-hk&UNV2jbTa~<
zQOJ<Qk%m|%;-3`NKH)2sLgPhrWAW*<7nJ4cUwQvUx{>y2dV5Ekpfy}fS%uBm?KsBy
zl78d+bTO5-fUBX33=9wgALEWa2ofE(A%Fh+8>6l55*Q9C5c3vwtz(aa)Q{-HW}c8V
zSLQ!GOL{4kRhYk2YdhD3{^*@I$qOMiSO84k<$aaW^|RW4%qUX1Y}plmKkQR#+St#*
zCA#Mb0<fjiZn4~KjsPKz2s0Jgue>qWAaJZ&X9n?W;E~zD#Ko8qw#uP!1TzGr4BpDg
zu_Bvm5ZGkB<^!G-Lc1s+Ue{-o3=DmT3?7F|?J?4sj#JJXzuyDj1io*6{>w_6*)4S_
zK!>c3YtmMgwT5w~7;rFj_|KZM0W;pOfdSKn=J)kiTX}#X^KI}r9|Fsyw6s*c$-5za
zAE*^b+n8zO5hssL=L8O4KN$D2Oy<o-INtk+ywyk^acjjjN<~i^8lDkQ#V6<Wt>cOf
z3o?8}hle#DcRxeid7*24c_$B7<((i>k+W!asXgQ^IPLra+1;OIH_B{`h;E^=y^hdI
zOWWE2YJ)#JeZeD+nf9Rpg%1m*<J_VL7vpRt1li;0brTBYA-xR$2z_Xc;gYL~u!l<`
z&3fVeo&)07`5v0wrKABDCBw=?K-*wqZgk~YK4vl1s!W$@3WiAI5XMJ5<CuVzN*Sfo
zK0?1--_3hpGXQkR3RIg%gq)!ff>*&HsDA$ZSr|}RVBr{<n6QFM$imVR%(2w+^6%Gs
zbrCR~MnR$FctA)F%tqjXnQQQ1vz_Coqo;Q}n>3v)0eS!sYm5OB2YeN)NiSY*ZmxWe
z@@ol+GW{0#o12@7GE)qtELjlT@X98ofujMC7_P3a2~bj}JrecgFQ|TLm+m)#c5C5b
zraPu}>Mlu+mqk%gf_n+B#x#0aiA#5s9&w)6X>OaP%R)pK_I=Mz!J-zou_^pU_f&1G
z6yCQ*zqzv$*%uIXy6&5lk4gA9PL>L(P$6kGd>-T}X>?c!YgU#=yU7ht<f~%ic@$}I
z;T3sNA>+wCaR>*CL8PTln~M3W^o@;;FEvWrCY@1JU2)<Q`)zF<CeQSA9fPCCiNL#n
zCu-xy4}4+L7I;lgR=|*C+Q(mgzW=@r?q9;s-^Mx*FWgf`vp1`kh5iO9F$+Fu!^~*F
z8<*1DOy5`*3G{gT0$4o=4*y?s<SDz<&F*bI*KYYweRmLWB_dt_?|8ML+FFR(7Y{fT
z^4G&(J>b}x03s$QCkGP*FrL5_7z#>jwP-SL(Z^ty|J^J|>+%>de8Fq2HdTBiOGAL3
z9odq)7i9WlDeRfB;(Yk<Aq_qKH;jAwl;KSj42;gczTGrEiogE9va*t*n)fj{SU79A
zrZ&Oy>#R*yb#)BLAH$&d5AjYkKoB;6iwZ?2T>+SyyK^ud|8+o>J*W{zmgm#Zlf5qs
zl&;GstwzSiK&A;(&6|jz$U=GeFzv?=#P~j!!5{?e>+oqSRl%!eGC)kW@gNB`1WmTq
z12+K7MzQx}_?n+@rARSni>XSpcsgHpK1s6RRYr}V&{=<DP_zTy>Xj!o7;o)wJE~|8
z=5NrY%#((%$B@c1+jZC&@!)|F1&XxifVaqyiar87i#{{n!P!}+^Y+-yGU<JaR+SAH
z5g-B_o}b$uRP-hOF<$ls&YH9<#szf<VMl5%pLp+bZi_|0dtPm3)t`=Oy`ZEV1j+%#
zWq(daSC<6jQAbVd!soC7dtEk66u6xwm)tKw>a}ulfV?qTmW4*|z*zI-Gni|Kd{o~)
zf-v#BNKZ&D+F|s~9}j54#`!mbc=|u)&bIyPni{YoJI;2e39Yf7J{7UIw{I{O6B8>p
z8(gx8UdR4p6b(kYx!w0M1tumUUZ-CrR{L`e`8sttzFM{oku{F%yklM<oV*qne@Z|w
zwNjz`I2MrxEU!uXGiL~6Rueck>;~Nd1Pp!{76erv5Ig4Fw;3RS;t`;dOOn>sCISEm
zh<=BoYI-~sdwFl+ea+<MmyFnT?@ps4Gdw?Tj*tVRRA=6@@QAx!zDkC2rfsv+mV7HQ
zToN&ITEtG+R3Da<7+mCaW2d&~4c-3xYLbB4O|<g6f-MI6i>qh3Y?X_|P(tt9eTi+X
z4>d0%N*RuG1WPLh8*i=ze1QkbjQ^JTnTbN_-8`P-xz4W66WkO<%il=ljTYl@)iFwF
zWo2c=K8^Qx*EQfpXoZW7-IgX6lAxmdt<D0DjE^sVzv(^+W*$tNx(7qG&g656MnXaY
zx!JY#Ne2;>|DYK`E$_8GT+}C#+{f0Jr*3~YZ8}cU<_<M>velA$d+ntfwKkQg;v^3T
zNx1UQv4}E1Gg3^1J}F}KJJ^N_wF}ip%xVvRucIi$%1pEBuJT@UwfUd+&E-vs?>F_>
z^MdD3j)IyT9p)(1C1n$a|HH${p-~6yR9JTi)W02egnr#NU6+g!#5q9Q_OsbFHD@?P
z3lS#}3#W*DcXq4<uQt*{Z0@Hlixjl9wp|5*z4P(_$`W7lQ0Wdcl=|F|zlKz+0M1L9
z6auqpa(dv7;&)|_Se)#7a7_dWmKHX=n?D~h+WkT0fnkw|OtN?=Q%SmXWspagex?2p
z*!2{g?XbqWpL=?QBc<$Q1>l`|OI!{}DdccmoZ*e?lbTLPTU~Gk3iPo2C=}UU41-o_
z#hWp;U4k<m;e}=it+X}?X4MmNaBxV+Fuw=jB6WDv3@G!R5v0`A)Iop?{Mp(HV{*aR
z1IB>?I21g+ykfa*NfLRT9YUWzBI14$Cf+(#VR?C+d@l_Ok#E#;nWd#)O_RN%0E-A=
zcG*$SluLfxobt+VV>Ay}=>CSy&CLxY-<_`?g3J;atBJ{>^`WyCEiuXO-|a)q>OVk=
zRicwGHIRHe__A9Bz31oL-=B$S2Ya{@?dqzB%^IYED{r?nNkw~v(5$E+e_qB<3fYZa
zZOTHcE(beCBs_(H8RpU5-3{nY<dh#j{L9PPVe7-fA~$^&m^os29H}6B-j|dE0_Q~Y
zKKBB^9*pU@-oFaSvs|dmYMHLjtuxTO%+1XK3S!o8hVVl7daLc1A>1xIc=4RpKdm^2
zK`5!To%hz!t52mJS=j-8=FnLKaneb`h%udJ1Dll9jHG>c>jYBzIBL-;f|nUJ6VesH
ze;cNj6<-`5wq9O0GF)8rWw<iHjxLJ`J~$#`eWAsV|8nKqOP$*CK@Z*{t*UVE+bc{$
z!X7}#Zw2@rYczQaL<rq+TT~D*F);y|bD-9Fn`pZ;j4((wPpoMC+vm?P3u^zS;aI+g
ze?R~a2~OT|1e1<jE{1JOOdaX)W-D+HBl?aNQI6%-F5g(l-LnbZLW)%*M;hcakSe0e
zNKcw<i!Qx9$QfT<Rt{uK&cEp!G?MhuhWKiQyy>N^ftYz|!Kd&X4=cK_J$A7`rwc3G
zo2S-iwEp4s@p#{r`nBU9N*ezy@YaIr!*sgR8u(KcH8nMh`T&RL1|qRWy(`Q0`Tlwj
zQ&T>)_g?Ylj!w#y`Y}9EJYVAU@D#<e8b7S5sR<8KU9OLp*0yDDWH{Wve{z-G=vuzL
z$IWO#CnzhkCJU*ree@HUIg<u&OQx+xj7AyGH@&GD#2Y?Hyq-NR9jH9tbfLdGYZ#O`
zX*S_h^x&`9gi31<JCT1WuRqtjI3w|c&Y-{t2|f_E-tz{=XM_czgLm~<d(HcS_7m2s
zokt6pO6KS56~NPp*j!v^<F@db`f!uwVY2_1iBJ{gz5A929#sC8+^bK7ZEP6f;NW1?
z9&iZ3l+q6$K7jHaT29NvB#||YNkH&v*2WaLJWsC<4iAADqrOEHA{zAA`0nNs#$PL|
zsg;=xJ_q{38qD&z(?Pexi3Y-P%9xuzo2sIh6JO*eVT(nk!9LX0naEe}#u`|1vnMW}
zsfsH2d5x0Di4~lS%{SuC*|(N_UstNxN8X|0K@ip^&~nCi2|;aYJWoA4=+IOV0Va2v
zX;#v@_BVrZ(8FyH(pMh`^cnlo|8Dan3A05JoZTm5zM++VpaJu#SH^v!nxh|lQrGT3
z>%NCuL!U8ZreHpsFt=aP4aAlCMz8vlAt}JX1HBhR;N%G~>rhis1p<Ns_?uv1<9yD8
ziG_s)_+DAuPd41W>JNURfwd1~Ks!~U`rS`Q)nE{sxJnQ3qA$;993$8yanXzM>1pLI
z(*{o~ChKzg=4KaoTa#Wq^7y*NkSBOUF&gCKWb#queZC4Mw&$&6`Zm3inqq8iX^XR1
z-dv{H=rqW6wpy<*D=p87_O>Hu4W6&hVyf-(LO}R%Y<cno6Y~|I3aI6hoQQ_zfL|Ms
z=xa6WNropIfCa8rQB^(arY|aXSd|2~N<MECu+9EebM7I)rS$px_t%PwU(@e_=9SDQ
z(f<sD8u3uPmS9YZevL*4<6FIs^Ld|S@R~ooiB$8xHBJ)rsx>qbf>{)Zea_C#Oj<rX
zK)3NYUh6we^0_@7k5;S&D9@w~INWLbM3<kT;Mxx+g_x8X(Dt?3RA`UTG*6a|?=@cA
zK-LD5<%)D0`U}Qw*3YJGseuOw7Ouy)S>pS8lZ8mr4`)zw`##Q~!SKv+`q+q0b?4g%
zDRis|{-faK5K}id9>5fUDSDIO0_;L78-wXLpdUiKXdD3mNY|y0)Bs>L9-S}J{Y8h~
z^oMcHy#c)wAX<AKw&R8p^Wn%KeCvb)_ykCGd&j!7HJrbY5W&H3QH&D1MG>5r>M%No
zDKBC^xxyuNRR#bMfS(3^!Qww|K_*>TTzuu^<m6}VezwyEECkH4(&&X4sHmyc)puTs
zpFVvu8%zrr-fmF-8(J#2nT>-Debu~o-5@8pPnqD7`LB1%!Y!xOU*ZVfT^NEItY$Hg
zklXX3*s!B%(98&|D&U#sY-db}p22?w8w<lS1E_;fG|X?pv+oZ4*tsfs7u$OGlE9Jd
z-arCe3%|Zl93C7rPfy3gKnE-cgGFgrHVHFK3<zHD7>_PX|2ZXgLb@;<YvJuY7h>O#
z8VY?E7Z<;t`@8cdzPp=g#|tk{&x-{=4A_gna~5R0H?p#g-xcHi{{~Mbx5evl$9_dZ
z;(h<x@=0==igZLqxCBnPVC<l62Il<@cFaB%1Xkp3z;7!=_=Ip4GQQ%D1%O>xpKRW<
zX%a?z`#)hM&w_ghs~dhyV8Tk5zHjX&?H16jx4sZFWxDeI`j+1uvarz)Hfc2|WAz%q
zSW$`QYuF!4@~$6S9t(L0hvfJE_iuU7xX^iXd7P*#<w*7y6n<AR-Id2kJu-|N*6k*d
zQ{Y1AHoB_5dlquFpaq%_1U$}G;%yrjK^KMX8T71<h!yz{f=*Y+8O%*QuO4A$@wxMa
zbyCQ1&9)6?NEHp~)<*rg+$FA|;b9*8B}8yf=|Sne;p~AT{AY$762g2bHy4K>4cOG}
zasn2D+w~wNug3w@s4(3EA2d|}k>haw{w_K=7=_nqL)h8*)S2hupOg0IHf;a&3;<_`
zqm{R%Wo5u+O9jkx-~;bC<8^eBl>X=zgFEWGoG&@o1@c{{#jK%VqQKDPX_WKkPyna(
z)WIxtUMEb)8J`Co2qdq|QWbK{b~mA;g$OAMIsMZnr(2`wjPqrt5yRW|9H18cI;4m+
z>lP35g)EwNC`^fqSHqkm4y<LU*0^h1tZl|9Xf`cJ?RvVWyJn*Vy4&`7+Ni~-G(f+w
zR<&{4<$;;4Tjy9JFa=Lv{dwQ`w&=-CC*~h}^-PUroRyYlg?J6AtjwD=$iJ!)|3i&g
z(*XZ9Q|;xX$C++Vtn#5Px8UIAL2G{XxBG8ZcGSE^nNa_59L{Wcv-9gK3~gq_*RNkM
z?Cpi?7Im7j4>fo|f!{4CIy!pT(t+~NkjJ5i^mTM?If(Gr|JGe%@En*fYADnT>2py?
zhe$;;0!<kXtQ=zKObdu@Zd*B73954cF@@4qat5;#xn*`)BWPlzsetne6$Qmkm!Z!e
z(bux=;&Fp{#(9fOL29?blDwOsr5s6SSe6jfAKE@369FpKHuXNa%h(^$t6URuYWGo(
z%x5w668Ti?5<g5ya;_PF*3lSh^Y}sJBlYG-A{0VGz_6OB>4KA}KRBrK`WorjHOm{(
z!}T2iLKX<~YdaJf%vmkoSApfrjoBoKwQdF_T3rQn>pi^!L}n}dn%XsTNTKOvdT1w_
z#K2w7xng{<ch}eU!2hUf?_>GUHN1-hT~7?um2dkfoP9#hAYPV#{FF*_qk&~QpNI|n
zxaB?Rq=r+PIrNQoIjq%Yd>hQZOw)m+8OH&ELdLGm1e~J(hVk3dD7feUC%$jScCgyt
z%E-V-cd$lbfM)A!LjNWOBy9H0EA3cPS&|OX^j6XIj65HDFs(`uYDQOARMZwyN1v|j
zwTD%nWvf^yhs_);y%8e^=D}*qa$rV0vBVtwah>`)nRqrJdJLVZfRM|-JUuL0Xo4Kr
zae)yvNs~*{<BGMLBt$$Tv&18Ehqm8X&z!_Pj4%=htEy^u3ij#!ooWA~egx7)zSb$H
z=R?#e#otuFO1N*DdphlvA3t6h4EOq-J(5j$*)t7{ebZ>Ahhw!$?+T4QgLGwp|Gw0`
zLKBvSbv+No=Df~z)S{P_|GzMY@5i{se)-Y@H0td9{5F#^7K;&@IUw=?4Zo=IiVo!M
zHY3)-5<`i#{zMNQk|MXvZ3zE>XX_jKE|>!zqR`kfLpq88`8N8qipLzUAWHmt6BE0`
z+PV%RZ_RS3a>k@q<8U~Ko_^?j3|Yt;j<w>BgcAnl(ad&qxRfyq6WJ@(T!yfDM^9iu
z#NwNS2nXauO(J{OcSnZo2MS2~)=@*~BeKd~@}c|QF;K(|@7q8uUQ+2!RXp^YoG=rD
zinSd&+{i;D!$#{5jz8sT*Vsh?f9w#jA*e+!Iz>C9q@<+yN|3nugJSYaRFrtoV`3Uc
z#=`MTK()rQnU<Okkbq9Sx3aRCV2AIKku2szFF>Q_PtZ`OUS&fJ>l*^V?REO#!R2~#
z88D55cCm=$WKvS^3v!6#TJIu&M(co@d<UQ@Fnm1lh0eJlgKEiTXX2HG1sx!LmLt`4
zYkpQ`xM{}4cB^UR0P`y~XUelgPc6u2EtQV6^tubdb25mi7PL)t*tL>RIcZ=iy>8NJ
zi-v@NxK$G>(syb5g|@Ia|296h!3Pa^UpMBTe@JR^hPL%&(=7FkIKS+jfNKWUd;(%|
zqzMhD4?Z`gN%zJXO9hV0E!j!9^$j`ItV-Ux{UgNm6KW2h7T+b}V^6{PmVZ^cv7c6}
zPppLVT{pDtkXWBH6N*1#y%?@6x`1A(dd~94Jmve9Q#Mn?g;R%~*XM>2ayw)K7f#3(
z4(fCRLqi_d12Vv$+^I|{jl;CYzbQ;p6*eeGgA$~xv8G8m<FFphjSVvr&x2P_v9Rz#
zYXY!YTFq2{>Ft#k56$+GC<c`q76CywFo9}RSUkfC2>=H1tMh$~W>`WGVQPE>*e9-e
zPd3q@G0II)?{xR{XjEF^%X1mB{Xg)Jb9*?-_)P}4SgpjQ2CLtzmLDFaXUqjDaXc%W
zWO=}?uDft?d|blCmaT%#_b|ecneZ*WfSBec?Gl34umDYHjIZhr4S%?JhDd?pyOPP0
z<jZM#7F8vK{I%$>{8v@MDSo!hi@I<|T`&6cOwMD!aQeo2?HMV%J_257d6V3C9<MmT
z>ffAC6F|@sa{x^zW9!7PJl5=d+f9~tY<0b@DK9G`t>}N509SwQgTSY|xIbUTwrNDD
z7JwW!uTuBe_w0H9%U}KdE?1TD6relKY{RuoJe?fC5ung<+RjOXnRy9Z_{Gz^z!wgj
z=AQkQI83^|0EmH>2{b^!dmgkPgBybBTesgB4}4p&Hcp^&wtDP0bOKouv=Mq<ECvCP
zn^e^SGc<{#LuO7%)T+m2sJp3SY~B}eD`+4ICx3Q<0Q)uGTqc`%d;qW|!`Zk`T_XaA
zK23H3;zIZV-qR0L#YY!#F1YIsYIWQXE$MTZ$A={MMT6%Ezg6s={=5W3%9dj60s4X<
zbb&$6NMu>1--Gy0GCM5RoMav_2zUmgqqJL;ebG`uYW^X6i-$e+-YrN;JsIi`dR118
zRPe<*=$_XjpdKG(jTx*tg|#^Ixf0ivh#^*0>~_yBearT}?k6(g)7Je}ZsV6zj>n^*
zA|8_^#O3dASTf12sessgvRB2LtMjsGU#CZHvgu$zmx9NHhN`JUn7Rq)^(7K8P}`@K
zp71}T`6gTNt}6w3MbcMhp%h*in893EzDRI8Z9D<GGOXwVQklE;BP0ZXdVx=o)owxf
zF%fq&5OUl$QiFbVcN+jldbsUjPEL-~?o{N^kRr(0pw*&XpSc(mHh=_!1D;4SI=WAw
zlmQ*fq@BN-K?M~n;K7N2^r#v1F>Y;H0Mr`%@ng-`^9+rg${4f7MN!kHsB9I`FGHav
zXIi>C!eV1JnMBg^&ed`jpa%fR%NT@03=CC38L!!Q^4L5is{=i~?`O%U=amV9frDdz
zJmKKe)(!9*#6)0Hr?sieDE13B_TL34ES5a7Ar5PbCrKX_$xMy;k#M`B!o~ko$(WNh
zUxwP#c~4oZ_eSw+XX%b!b_u7TsjzmVq7pjj;KC;&2oo<zOUe8y_c-fdFx7};b0P>N
zITV-kv@wfCVGTjF@SB;9&GKg^6w6ak>Si6ejLA0+-khYwl9@vW7R3!in<E65DK_wQ
zb+a~2c4dwXp#Q1qW|aw$;4GdOcA!)p00A5CFS9NpGBU7&|Cb~&yfATCmX%H5e)Z-J
z3aB(*01gsF7#Pb9)<|f4JRxU%Skv8U5e!^`3J_*j1;ybu-p3f=roaGr-fbH??s4F7
zbksBHyl7C?!QXKpHwSyP0&9NU$-^)A?rT5M#mds->jG!zZR`Km>3Mtl`O~PyDkuY;
zb|x^x#03wE+E(kM|Apy1uByk4$@pO|sA^M%9e_^=R9>G3En2J`?_xE1E0E5Q6P_2a
zAIZW1IVbS_a~0lv9#&f{Ur}|}tDL>-6$^lX9$8R7b>ZrHz6Z2j-sljRP2h@r7Wen}
zPo}j0Cun!5&4kxW^hl3pZz2MnB*PrfVyr>o1_6~54QO1qnr}?>^@RWfn4Zrq2iWje
z(VQoip_ppF6#29u+D>Qzv_KpIF~A=5>~|Wks?a?HI;FFwK;L_5cx+Htc^iUK9>|nf
zL_|AViiLFl@T3ZF@L>`oBclYzZ*|*Yf2-g?<<$%drM~nCK}ih_5cEIVfv^Q+)7naI
z>Oagu+9Zfjpwh|`U_4E$Lc@7U228?)Hr4Mz4^6GhZa3hHQ6D|3m-Lu9!S=xM6)$)<
ze6qcN*DtxSoVk9{+YT##kn{?WYilh3G?%_>VTd#0<E;o(b)WxJcQ<r>md#qO7}tfp
z9Nmfg0zh%BFARxS%ZGK>f~#uPNjIL}9r#md*JFV#anGP(s)E_^zP)hZ&i$v4@8~Us
zu@(6fFDvvpc$y5u2rUYNj;|ZU|0*A?!1jcT7w}rvABV41JrIY^N_O%H;DH<ZVlyil
zaC<zqb+7!8U{4bek09~tuLUz#v%&&!b6!P?k<NdlMIh)|)Y6gO3F|be85%8e0kFe#
zvFJk%^kC3;#GRAYLF<B;QvM@mP}Ja0Z-F)zXb?7B#Ld&z=}<X8>?DPO9vJZfN@*Zb
z1LfA6jD_V3=;DCDT+Oh917(nzAPm-62l#9PNTj5gxj8K`K!U||yz))Xph9zVwvMX~
zy1xc$cvJA?1&9@JZ-eHJ=)WGl&;f)dn12Sw-+`wL7%u({;J-Z?=0BS-R5R9lMhLOu
zL@V3pUi!<McW?tOEKD(nnGAgHj*}tzfAQ8O-nZ@mUuXc(c7L-jRI}fBtq!PP5REfy
zYvX{G!=V+O>vtX0Rze7fYoKsCoHR|g?_sDaHXo)E1c!5<=KW`~z2CZOwgdG0z5PpO
zXw1E)`(rB%W&>;^+^^nVBM`)4QXdS?bazXFAv!!fTm-E|Gaz?@#~;9aKp+QfZf>4l
z@qpX_G&C)nng_n%M?iLVw!1Y*!vDaf$C*$M$nT*1A+bIK<+RkKkARhxwfb@`p12Br
zZ@h?<AS|%CSr~R3YxE)|RrYIwI*YzxauA^7gse(RHN9Ykhdn{%rsMzU+SRh<z6aUl
z`}dLL>WTlOWjE9QuT?&e$7+NEL>g1zZ->B~3<(KDz$yDdHP3jpI})(UTy5>`m{?e?
zuyraUBhw%_Gdp_;d#nXuTQ`AuKiT_=7DzZh=jZP~O#6Twcmb#tiSxZ#Kz_*;b}+*K
z>1cWFm}W1!i|GnT5Z{j754(WcE}pasUk5V}kiMPAK!I$$E?m_3(qxSbT3dm$0=~IF
zk?#^r%Z-T=3P?PcZQ*%sQPbVj;X25*j=*n%w*ULt-{OS4rGVMB`kAx+|L$g8+}xZr
znr?dX7ZvfWVho|sPwGq-XjH(AQC%A<-bDlfu~+1uAGH&UK!&)0>4F7p+sS1!+FxY$
zKGbT|(Q)a)BiDa49id_0gHz8vqM5v69mVe<*clcmXL`>y&?dl5Vrh*T=#?ygb{roz
z{@VojzPaSf7ik||ucaQEOTAh0oY6WF-rbV|tqZa~0y_!gZY~!H47Pd?7hJ7Hx3=z)
zlN6vSF6W&wRf$V|1BQ$|D8j#j#WtoXzh0Z0T6GK^H@yQcC*3kUIPl<<W_@Pu;y?VL
zq@;zNj?%2aTkT}&c4S)h+mvMmw+kN#p4`4uxjBagPourhkFz$DPn?1>t1Xn-Beh7^
zN_E$R4kHducf>SzhhoxV*Qgs@#QQN41O?;^9pfcqQHrs5_z=1<zOgO%+;B>cD}pd6
z;fJ6b<oL~>O+qbx>kEWjAGdl{zS~?Al2{`R9rYWh6ke0aH?DiJlfYe^dft8TLbc}@
zjVn+diA%i8=Flptt^^f*Pwhi}$B~!fkc2}MK(BrI{Ml}FKlx>$TxFa{Zurf>;Y`zh
z9t1Qq;6gw*(V>|mMFn-0;>B%iHoMJ4L5KK%^tN<E2Rc5VVRCeGE$}Qt8d2pG+!lr%
zr?x*B9GK0Lie8z9dovhI&}tu(UAcU`a3Y>>R#o<&xuVTXV}BX=?ZV0Hp#JY|F7Mc)
zmPhR$k)(2^*h1+!H1e2N_kU`C36txk_)aB<ytBS57o3+mW<|Mk>ik_fK5-73k`+Fq
z7POc;%-yhW`9(<_0t=?VF^DJ(L@T<;PvChJ@UW-w0BG?_TpWo)BK!xSyuoT>G3_Vn
z+y*HWbS*<bECq6#FlaZ9;kJJcqRc5{OtJH}DrjANqpXZA$A5_EyL|TXZ~ZT7r`>Ot
zuCi=}o>n!>j;z>#wgAvU93=n;T2<0!l_<+SY7@rJsnXGQru0c`eIG;AA5r(;^yX;o
z&8IF}2V%fA$5h>>ZJ%s&JgZu;!M<b0EY>M*sGw`I(ZZzo{h&P!Em%-#xQL5}Wi2D+
z(Bh-kn;Wrla(}ki{K~weMGSdG>kgxagZ+oiN2@)XLm6n`d|^F#3k&*yEZwfSyv;s|
zVKG#%wj+nEuCBrYJ&^65f&~r8d^o^F7%jDZglz#@q-9aVn(Hm0ZT+igMjhBQfPhCM
zA))r3%Vlq-3n-YIK;xaQgRbWlHP+s6EpjEW!l(S0rv9{@tT)gFCP&<=x=Gb2I<duH
ztuZ$56A$-~aYddAk?;51;qE8}Y}IcMc%x|1IWpyn7x*B<U&&;#U)pA~lVm#!I-hQe
z%6lNb%VQ8+=V*_o9$u=XEXPxT2Deze<Wrg{nn!pUNxgk_9@3^9<E~JFV&6bZL!%E)
z-fFrs48)~lkf#;3J@@uUE~S(B@quSlQ<DH#CVu`Df&44YSK7efIh$$!%1DmV#;H4?
zF%|Mg|3*rv7U{~Enb81;2at{YK$SKO_%8Y>xvWr#Hd>Wj!rvdU<6uuKPZOdQvbWJ{
z)1jOoD|1HB>ENg_E5i>~ZVvAW&uZQFWAubu9CSKm0w^~db+fkSxb8OLW8aRfy{wEK
z_<P-5T&|tMQ=2f!W$}e(pP{oaey(EoaK65GKArc0(OU#LK+Z};C9UY>vNQ0;cGH`C
z^FKN{^Khv5_Ky!PMT8c66izy1$-c&OLXoX=Lc&;!k~P~*h(wOHESc<EvhU1{oscA3
zma&BFgt4y;o_n0<JiqIAo!|9b&-wi`f4Ht0*LV4RKKFgU->)~($4V?QcFf5=I^E%i
z%~MgmSw^92CY``K{T>pM4pKKlk+jl?uR!jceqUD(QrA13#9$-jwixH>)0FiPTZAU9
zw!p=r&}gxzBotuHfrW*l-QC@aply&t2_hp;D=I32yup8jonHqB-8Z~)!+2w0Axb$>
z`glskn36O2Twn&%ylS%Cy)q@7YWMW&!a-Q{auU)Jt=+-+rY(GeJJQ#2+op@UH<A68
zix(e)+k%Rgt2Alowz7-u{6BISd^GCABvy#>+%9+dlOp&ivUaA&PC5`wlBj+AFMTf7
z%2<D&J9!mP-N$9!g?cf;wNCTSujWl~PpcWaFg@RqbY^atA<c43QIN6MyV22&V7p-<
z<FYqAoZWgi#Ab?Xj#9xQzoUiEj%m-mF{>4b!LR4J_7g^GGZfdw$A007R>|$NEI*Jh
zF2Z8BB7eyKV9)G+pst|-=F?b^0DEwbX0<yLbWnnW#tB=3LO!t+043kw9Fl)7FylzZ
zqTKLwntcL-jNYa`8SU2&f~&HxzSgP_bNt3E#nJfe@VByY-IH7QC1N7#yd+~vwEP^B
zVDX;I5Gb@7s7q<sl9F@Fyigf<ARnavU&1y0D-1bqJ0A{%zVRFe+El3uewm4^n_%3G
z!v~J~mRBNfkBhQ;)%oK=lo-Z<FQNsv*c`%YC7$T%=*Pe%zXB#ls#mXHH^U?6S|C=h
z>sWUr%;U4W2LG*D!@&>sM#HiPr=SV|T6=AMon6-H0(m|Ot*|+Gnl_fllEIAHUwF7F
zIep|Kt<clqgh|L&5Ui#=j~`9o?v?;}1q_GafcH8;?;qal993EcDCfO(X*Uo(nw8!Y
z|3m6bfEC`QzkoKNh(qVqe(Ig{h4zcb<!ToDj}iK24%GWSHv?(H(5<Xe%7}WC>g1v7
zCDirNUp;W*xwM*U9dh2%?;b2!oz2fbbUfH5(93eNa7^}>T#~O-|Kwh~zL1LGIhzGq
z5t&kicp?rD4-W_+dx67k%O$IkYg%?0hjTDF<O5|J`k01d2jBx04@v8;3;ib(b#A~F
zlMOL&19B!=8pux2=_F4gkBkgP%<97}ef@x&8N8s@LZ!~8tgH<7tY+9x(BlEDSK{0K
z0|RBA+$3e(8eC#UcNLTGr<d1ve#s@U@hdD>N5PND?K9kb!pJp$7(JPgwkzu?mX@~Z
z^S;REX_VubGfzqYvfxv`VRQM??y`JjnZ@D1oT0?RGst{{%QwQ$rNVDKkaWwOk;3#K
ze16=Gj*f296s>6jN+8iRs#J2r)Kt9`&rt6uwLN_4gmby1WN=sX34-EOi9MZjOrVYm
zF}G1<&$UK#{N1l+8&U`*(uzuq2of6RW*8-cDlS{Yedm!oA3|I0I;0d%Oi1<7_~I2Y
z>!#8nniO<UrKLUmQ>i)<`!KSe4As|Osw(siZm(<;I^>eAWsaC6@vJ^u*i~fFYd;Aa
zV^uSw=*o}`RU<mi&SLV39nRAYA%N3DXWdrzh!F1QM+t}5{hz*_m~)kf`Hbki)KO`S
z4>ihvr=M$3W;b}cYD9$Rn|PL9j^^Hs&mS73Nak<ZBJ(@t_{ZW$|MM^DjKwa9UR&W#
zA69Y_BCto_#?li1WWJf$T$>5yEYLv~=>IL^+ftWGNJM~V<{eqwgKyJ<r`2$?j@;|Y
zMp#n$f+^fQ86p=U^U;LUoz&9DOmivf@AzB)bK+z%hLVNyTPIFctXNl;%MV-)`d`<4
zTJ7Wv77cxF5Mrv)#x`lsYd+>)Spz`FD<P7qexoTIpeDV-y0`n86_3U{j^U;!3#RwG
zi$lZh19Z|YRj!}X^)7uGHC+yDa*qC@k2BN7KI=Un{Oha>x-(z<!tl4uC^zIEw>{e3
z#tH!+MCbb~;@M}eGrIUXZUuMo6S+GrFX(G^6#C{4JRHW`hjAX`GaoO2a@i7`%eFp-
zB>UiBo=QYp+&WvlY{#FT^*wU*4ygIFy{CLZU4o7)aWI4F&g(#f#+(hWa7<r9sfggm
zsKWEa08D&LYa8P&!-rGpV$SqI3<U(@pyRgl*u=<J97<-{L8*TZJIcw`=s#+|Yq_j(
zP(J@NUsc%@KuMBAPr;P$O20a@pDa#6vrag(IQ_v>*?JA5f1GxsSIkpS2<Kr>>pN&W
zyMDDoaaWrI2(X)Wm(K$YbzF&fa+YMEL?rF>hS7&Gj2}jpqGyJ^;Vy^9V~KSEj96<3
zHQ6z?Y+0`hf@UBx6hWgFUreMy?DJ9v(QM@5h%fhtQhTyuYo^%%3V3zbq@6KpZ*3!N
zY4FvrQx9-k>O688Qf$K5>$pjH7^&PAXZ2%I-{ZXzUx%`(sra6&KUmVVHEjkLRWFVU
z-1%+W$;Zq+-34mKezX=V0n*SaqZZjUgn2!``*te2WI>hv;0YI(O7$CQJe5)TiSnHB
zxTHBkxVfmzsG)~HfpEUTI#Aam$th3S{z{2duD_Ubs^X8Or#BycW<|*aFcBi#I4BHj
zz2uOUt~@a0dD!#r)VXt7V2Y@betzsEuYf@7s@LZ?VnQZ(CSMQ{xIGueK&A+K`O*ez
zG6ZnS;?5JQ<iCW4O_OIi)d%1?Nda{kdUdiVUq@%;$|BTl?24!>HBUCvAPvL|97C8=
z&1(tiDn&BBR<0K^%Q$?)c}zElGU`xt`iJ<ieNhJ9Vk`CM-fG9tX?<C8tML&p628Pq
z?4y%=l)7%;cq#q#?ab@jp8SSy^6HM!Gv>$fbT<sxxl42JOND+N67csUINV?8p+cbM
z)@z~(MjUzBHpV`P>&@7-z5EEh{YmLsjxjn%kFa#@$L_R5qmwc>sQ-Wr+<?!Zrl!_P
zkd3e(#&!cy4ug!_5Pu{}dO!9CP*tF-;mGZ^l?n(5KvW;>Ex^F@%a<=%+1VELz7*sj
z6o~YJ1Gg2zH@X%!!Gw}-kSlQcvelQj8hNHQap2bglMGA>j)T!EQ#)Q7Q~Bsx0w#FR
zEMq%KXm;LlD*|bx4HpPYw+ZkTUg&!CB-c`n>*_v`?}kC{cj_0VB*ii>iBy@9Yu;#!
zW{SvCO|FIC9)H4bPM)nSTbXGemS69TKpH0hs(P1dY&2h50Rb3Z)nQfCjVFOF9wtBE
zy{Jz@Pyd2=+_r!{&pd_hSb#SHtI174E{sG(!r^&Qt*kUMPR`7@Yk23GR7Aqnq!|$G
zm413lH17HT6_;e@n*vNeD{{mCF0I>bYkHBaCtS+1PW!r3xSfAGPVofM;*V!OzP79T
zSUQ)HW#z*jJp-%3#8SaU!7K;=dhPTN`=3`gK2O*R8F1R~Z-yWEjG3t{ZtZn8k`#C5
z=Q-I2lwP$QZqt?ee9ptw;)j|l`2xI(q(X#Sf`jPSES&-?<>+9cc;pW`tACq|T`JI0
zZV#r~6;hNSo#!Gg{N5BB>U-0z7SHRCXF-<BA&ekG$x-Rvqeq-c^Il3ehdmH%^&j$u
zL<b1b0=M6Y-w&Rw=oq02LRgyLyTJ9lhITcFt{Aqc6qQ%3>mKK3C!;5|xY=5#;(5i4
z?6ekpSbXJ-7pzBofMH{p;gmJfQ!~jDMH=ssw?HQgq?4@V5(_Wo)%mq8ez+Ya_m(xd
ziG4A0KEGeA`||PRIE7uCpv7|Qn#=c3f(1Wj_J`}4W@dj`lc0hPaDA}gj9#2jsOAH0
z0}UyKaLnpx7Mq~0@HycEUh%FdSUO}HH$fV)vO0j`-*9M4j>#a?n6Q7XeD>W64Rh26
zG?j6KhRR&!da6~t{r-1s@?Ddq-fbl0%K`$}1w}NhfpcRetT5qky(gcE1W|Yg<XsA~
zECU0`a5oTfFG1zlWSy(<4!y-hei>qweUxA*o|7ZUB5b4zeHsWmB$5|7ItW_NPrejX
z163ZPWIy7+qJ#Eyw6*PvxOq6%)R&!RM$YK!IgQ}9vY*ReSx!Qw5INTd03@<6S!<Xx
zG(uIa_sTrXQ<k8(B3$Sv=A6IVIpQ3;+!wRV$n3PxBA^$~5gyZdacQ;`c?p$rgRRFQ
z$htk1l4x9A)qrj8<9DBp>8=S&wmQorI1o}K@yy(-5bL#Su3L*Lx}>`if=DU)(xg)J
z{*3~ExXrOL8$!@y;@+~zjSIUqp=pA|T{{^mYtF;kT+UO^lBz3bm|U>;6^1rKXzK43
z7w;N^>61)P0$jfRJd|~72{#5}hv+1c*imHBxC3dmvf!`+lnHnT7}5L#vLPT7A{~p3
zf%!WS6XZ!OnF(iRW+vCL@HT?|Zw>~+;7LsXt8rTg?5&V5o0^kz0tU&hu97g&ukY-9
z3E~Vn_ZK!d@E<}I8x<W5pDpO^UuKEVHrZpZ{2W)<;z(md96)_|iVe9xD~CC>eK<ea
z5p?Ipn%j^7(X1r16(<^p5m7j3e0H8-@@L$Sd^WFhhR9{aQ6=n!AhY@z7}E(DE$E06
zZDDd4S5?YNDVQTJX5NA2A3GSNqi~C_h~*1-%yS_l3gNV0ktc^B+X3JhX|^q&fJ}tQ
zi<${Y3{$X8@!&jZiyvcuI4u>3Lbn4XZ${(nhBQE+AW{xui7<)C5VAfH_xo3Rbtp9d
zQXC%DuYWJ~8`%%xpT~i$4OciIXK??4vw!TnNOS479fIZref5NbF)dWd1i`-*_<0{U
zyi_3ao*Cx4@UY3~4oxZ>oGWmofSIi0mfaDg60P-r<3<0gAph?!DCO`tXT4hLWZrvr
Q2x&)XsOhK{Tr>CkC%kV#9smFU

diff --git a/tests/tools/_images/cohorttracker_adata_mini_step2_loose_category_expected.png b/tests/tools/_images/cohorttracker_adata_mini_step2_loose_category_expected.png
new file mode 100644
index 0000000000000000000000000000000000000000..045a721950e7e22b5f589f3f44f789cd0d04208d
GIT binary patch
literal 22658
zcmb@ubwHGTw=R5BA|W9qt)TEA4brKUNJ&VCbT`rrC<;hQNQVNF(%s#uba!{h5a%~O
zyUsrEKHt0d`TiJ&fqQ1|_^q|BbzRrGKfjijx`#=I2|>_3>6hY45QGeVMe@0O2RuzZ
zoNNG3d`=STPRh0>PA>Wm#*m!8lbw~Vla;x_LuX?LM{`>nP8MDkPNs)uPEK}?{H(0j
z|F!{(t%E6R$EvzA_z(=cml}=`ge{MFAbGn4(?XEip0xOL71zX_85dU-<C~j<vnVH}
zmq^G+1`kV70|*&DIN{lS)p_njB+ewFriiaNyjT6SIyavwtFXogt2&T|7&|~+G|=F#
zM5mv;&tp6bch<<bi);J`pHa=<V<f}h={Sb-t;MZ*_M35>x%G3d#U{#U65!+GvlEuo
zVBz88hchZ5iHeK2`Vd0m;^NpZ{|7HRH+PYz+j@F}3Jcj>Vjj^LJcx^ni;9V<sln#!
zDnxq7#1tGHjQ#h2z&G!*JeN;XhG=ML?xaHxX=n^YG5^aKe%EAgZRBwp$nENCdhqyK
zei=$=4?XLiF%D26z$0s|Mcb>a!OoWOy9eR;CpAp}o&^6%)NgL6DBN$-*kut*j~6ge
zuyess>T$yS?;mAnAqjGX-kARM+yUm@ymBVv3E4v&t0$#8w#_v-9VNP6-A9~VFV$sd
z-!q%;FxK|ikvXmVk94H`{@FrtrT<7mN{Qu+rPKC{v7MM~{m8A=1dq{@pr&7T73-?%
z&Aj#d#a9$Hn08dm0+xCiswZrE%D94kuiv)R3sDy{f_DZClg3!1K4fMNnJ6($ae<TQ
zy1p+iPVv^bJ&bACZ$hWm4*k6>_;r;+|5mL;Cv0&k`<%zTjSFS3NmsYKsSJxfj=7ZL
zl)l!NY-C*OWkRy}aqZXNiG~zcH?)E%F9L364ApC|^7m&uZ9kPX`gtjbZY}iY`QRxq
zjg6uwpmdZAvL&#&-1BqiqF#!Pzcer>2`Cx#GxQKX*DQ3xMZz8WIzx)jlf#KL-$tVM
zlNid7PwGx`nRVlP6?tUD$ncPrIW93OWI5fcRW(t{7Y7C6t$mc-yJN|FIy4sBOAOAH
znj!T)0o5{*<CYg36?t@yaLu2(^&iW3+}cjXV1`dAc*9~A1hoT?>=>f>*rY~K8jEW1
zOf@k=qh83G5Ug5S;KZ+H<*bD6T<|If*=Kxxd&p*JvFg$0EY><>u|g<nZqBgO9%49E
zZtd#oI#{SLhC?Oz(%qfk-roM1<!JB^nHUbYL!;4rouEDw+J_H)D&}Abqj_3K^S-#q
zgEcO;OvG4oFgUY<&<P=Ui)<oqcxoyQmDjN%x6Sn3#w^7&*PDwqr1hx^R-I}m{T9D_
z6nqXJpZng?dA<Lo<jD4SV88bzN$4x`3-hk@JQ(iYH$O;)a20y38;j;MrXdN{!hY4K
zvXRgKq`gD4M!5L-B%_^oM?SHm$L{fpevT<pqlS{kNXD@4$sX8PyGJHxl20aA*EGkp
z&20jcBwNek3FLWwGPb<3QslTM4>>qEyp)$WS#16M_{o#1Nx3+VPv%1zNc+E}s1L>s
z!wycy^f*0EtyX*ET5v_KL|?xiB)x7p{w)I{H8L?#7M!g*_mfn)1nyueAu^w9_*sej
ziG>w6`OajS;q~Ro@%pbcZs)$jV&@xCjadweatCk)6|SR`>{UkXV8c$|#WKhYz6Pzm
zWPP9ds!)JkXUV^+2zUG>oY{dLaWy(ZDJRM-NpOXZ@q&n%n_F7a+MO9ye}0n+qiSt$
z=Xkd{Y_Tvu_EWnm0>0NkT{01zl0t3R5o+8PNR)q)CKc9hQMS0WB&MxR0YBRlNekN;
zLgcXpcNNu1Rtvq5#m;ut&WW1$wZo5b=wUcM3TCLrZXHPbhq&376}lr0EA%l5y~MJ_
zxr`4IHFw>L>oUFBAmfXUg2CiY+s}>)`Xht4MI0a=AvQGEd8V98FK0Qt<7_hgZ|lne
zKA$(ecjTGR@+?dbv)yO1VdBBxRFOqhR0suJc9+}9ZI?T!y&rRPM^8^rTUdi-*7*^G
zK)vuS(o}`*T3VQJj&uL*i&&Wb@W#&0&gFzb(Q6G2w{}Z9xp;9&Nxv=qGEEkXDam<)
z`_FUnBh4Q01PnUYnN7l-p}^2|JA)|;h|HztQ{lMqvrSK0dH<O*!Q$H)CpT9XeReO1
zH(a}{)_iXY!*VVRY%<RHs%!Ejjr>$L-vUXsS4fPlY`<l=mqZ3G3MEdgfJlMZDh@gb
zH-h*ksWV$X)0S0QT$DXmx3&4eP`pXeVCiP{FvIeSP2zH~l+TWe$-X7@bEQ-sst-uv
zvUCJIbDo?-8FDQfPWSKMe*uHl=gPrg!c<gLAN~E8ocQCoaxbs@udlCj%S2`x-=oCw
z*iy-7MlLQI%zEE?-NB{OCjTLG@2L!yy~^a>J;C7B>-Kq#0^yEwSkL!&)XK3%38i~C
zL2W%TCSJR(8-*pexf4+<RCz@OKXtY`)o=NPn--qcEZ{uFERnugl-o9LQ*$j?&QaTi
zY6u)lMOH(f&l@%s72NFzMNZTp8n0OW;7{u4|Im)#-m3rN;^i$#BDzfB-qXhAz4ujK
z52GfF=uLe4>?K^JTazq|aT?35r(9R#OzrrSR8&<tOnM)<pKeRb%3>E46^Tno6gh1u
z*)&``8yFbe#lX;S4Zt6{bY1*R|KOP!UVVMNCw!-3WK`&E+EMUwJ+0Jsjw&l2)pg3c
zVnlL72}Mg<7#`gxxOLe}Y1uVhRNnqBn{h3&+5uK(EXCv1^h-`}*doM0bOnck+@ZgE
zGEVn{P~9r6Ye%#oh1OwC#hh}0mx8{v>R%3T*!hy=#rnS;9Got~8@4R{l+I+&&xm23
zD>9(MD)mv8(C>GRrRXV9gHmf`WHhWK;gIdb@D{v6QG!6hXMczW{Wo6Ze3e1c3+Xk%
zQjXcScOj)QElfz7X7~62%5gT`!)Ki{PhK)}T0vRV6ul&K8=+31p3Cy1{eP78|GMD+
zZ~yVDBN26$$6@!iwR0hLV8=-xEY>fq$^8=a1vV`|M8`a><EQ;`Ub5-R<rf?jdy8;=
z6wx0o{ksDBj$nKINx?IMf>OcNU5;g%tAp2!y6b#hGsc|~u8;JqYh=D?X8e;N<Uj2b
z-r$JIN(k<KlIaH`x%BkifdLA0{gDav-?Y>nP0lE!OWdFHCVf%*(#6;{cVMpR5ON(H
zxZSBIfcxXK?L#6Q0fEENgSC`;gTA6`HP^7<w2PtKeH&u`iJq|T4+Mla$9x?k@xcl~
zHXp+hf}~z@HcGb@5ppz~Gk!`C&MQjs*^>PAc9>D!H-5fU?KR95onF{2I%38$fBoix
zbS|8KVW^VI1Q+XT8`1Jk=VL=ROqd*s`3;=g^^V&TYIn&i-fI4k+8ug}?MxW_&NI^A
z$YBs`qBH#u`ar&`K3E{mvhD=seLlIo5a^?yUqp0td!c^Ic!{Zk^*cuTYu2^93br>9
z7+WODW{jpUYILWka>%*;lW=jZ78Xu|+^nab8f@__`U<-&?k&7A6#3yyZo4<r=!=oU
zVY_$7{aJvK{v5lSt8MU)?YH+kNQBycol!L_s=nO+CUtkyk-8t$$}N%kb?;m8(PFiO
zz<?)<3RvtO@Uii*wA85S5l6nuU6kd#YutKf54?l@RK{AKb|r_>FD-gIOY;#z<2^gi
z)+eF{^X2ULDU5FS9&E@sJ6c(j3E#U8s1~|Y67!=SM!}69M;`X&=EA0!oBF$MiD>F>
zKwYy3t&&b&7{sJkUt3;7IR80BCVOu~>`NwYg{1R0>(^q5q6Q}X2_-oA%Nt_3jcggE
zdtoD1?N%Hk;{(NF0vh&4tVmR|%c@a_u2w0GHMDn}u3DsHuQv@3?$Lbfd09(VftDHh
z`}%WKhIRWQ`A5?<)sM5s&tmi5=JUmnJYi0A?ZE7vMH&xs#Hdn?8Lpe=sF!V8G;CY6
z{sFGh`LlhLL_%t&Z@g)ExnV>Fuk|RrXxQWAV*3w0Vj2X3K6vq|*D{UEFq)6&PJErg
z4=2-hg4C;>7?d+)n_60Y$!%+Yl?ifiM4X?yE-o&*UaZ7i0EDxDn8Qz|Q*4CMm%w8R
zGP*2)3m-lpZ#xSN2*a7uBtue*iz60V{3DZ+<N!3F6mVYNshDdY7zm4sB5V&Kmr_*3
z$MN^ovZ?wMqX#2R=xcWz5bDTzlbe;5_3cfrYHu8;|K-V+-erw)$9DaR`HQ-1==|-i
zLhn=!lFuVK%@U`<I*4%U=>zz+|FT+CQdx?J);mL!vpJM6Z&Z8NNA`49yML@$dK6vy
zBY%l0^qDTZMf#?@jTbMcdB-QMHnzoaw^=d5dldHJa$4#<Rs&KYHVJ3995&9^Hb_~2
zKQchzvz;f^HR*)v)2P~1S}TA9d3;J5mJle6d6Qh0GMFD??(Ap@Jq=>y-Crp0(W>m^
z(;jB<A)uGH8xT6%pD;}c2cW1Ynz^gLKYx^thbIPrsUqVZ;)sZdmoHy_19%SruY*Pu
zTm%v?bUQLp{i)tWpdi=;PFj&Vljr$@ztwDQ+|E=5!Gi~#;B+9xa#<4d%s3I;!sl+s
z%dN>H>E2LY*WO%ionKrO*)K_gPd1yY7xKP2N?|>)&Q6fbzhtc_fTe`Qp_6*afzmvQ
zlSPJBd5};gJB^3i$f)B-u)N_o3X4v+H65?8J-Ujv&Dz^&u>|dj#c$8S9tOiz>dx~u
z>nN)6mv1Yk4jN8fpr~Di$JZxK6k7WapnJ9+_eyLBHQ)?a49UmlJPrm2dWGjBnF~7w
zz7ooAESc9><@WO})`ZmE^p#>_$Wtn4RjIyc+er;y#q_q(9vYa*FwL6Y+J;xqMn<`m
zI~cVgF$q(NV_Rda#!F#~=YI^k><N#Jed&)wX*QO>iLyTHv9}T~MU_2D2hjHVy4h~^
z77M@xz55F-p92Ekj^w<d;I;GlF6OUl*<d<A1u$Y9yWt%@*u|5tU%xJ|tqp-Z1ld$B
z+)4I4lnx9GTp!BBMM1|weJCuPe0Ak14BzI2o<4mV5F9L+whHher`P3xONqny4oS(D
z%}L7vN`8vSCXZi@mb0<_C3kU;j5>&1#?Wo!myx5`A!ahS`_WADbABU6?FQV2j^p;c
zNe~4~*P5NQ@EpZ#%X$`XK9;4V|GZtf5w-JJ{`KnD?Te-p$JQ2Vf?-RFPYEQ3tlb%|
zaLmApg2sYtyek2mmI57bOdi%dK}3-gA4-1Tkw=HVR{3Z53dtUn+ujASCagHNFjI<R
zY_2t3X4&8-VeXcnl6Etm{H1>Zr0-T;gEY1${>@Po$eL&b<NSoT#3x--;TR?abR`&>
zGe~b5oX0V{i-@9>3wiu6DajVcp5e_&p8Un9w4O{!(31OA%s;;0LxYL<J^_LB%a?Ea
z;<=pmW|KZ)kt{m#fByXW(W6HT!DCG7g+3r)i=1~fk&%&IxAH2~%PfeXidlC?A&*ld
zfWLd=xq^6RU42)2Vh|XQ&ymg-9rtF0j9xB&nT!kC-#=}({|xu~fo=XfULZC9$$~f;
zovh&WsXKHN<MEMn#^(FC)Zpt+6WYqGu{R69RFV^CaM{i-6{jg@zWU!6HtbJ6VX9mR
z+wK*Sm$VnA^d%o+$$BJF?r=(EbgPNs5Pmg@`>3#knqdM?OBeb1(l_UW!^HR7U;F($
zc2UHy?LDcJ@1^cr@<5q&5r-`TkeTPe6zU08$1fM^$1W!n@q*__D^kt|J;EPdj(BV1
zOIs9G#oG?NcBk9se)_03p}JU!E4q&`#N831h5i!K6mONwvx`zwiq`p{R)~AU{K0;b
zGAjA6JKrv-C-Ek$>_^fYLj0>toUlR7%<g1u;hCmR@q>*k)a$nDusnr8pXBSYTjf7f
z;{wtu`RYK5f`o)b8TY48p9qPGe}lf#{cN`{lu}^4`W>TTd+;MJuCHGi)v7}mBSHL+
z<x^HwRe{ob9}iDdU;oiwJ)EU4{*7F;w6yehi6G)&N`dID{F;6)4mN|<yC%J{&kJ<x
z_D@cJq^Exz9##R^lx4*$Te-M9@N*t@f52*yx*Rc1%<_@Np^NOyC9}f2^$xGR>Y>h#
zl!$IjZd1y|y(!fadXbX$dI!c=7s?6AAKfn$-!huEEeOfVX+D3yv+a0id2s@(WIxa>
z%sX?=uvi_jd_j5c4E7EmTu=seTV%9Rl9%z-UoumOzi0Dm9#xoaw^a@A-Qua<zTW<U
zMreH--Fr@AA4ROaP{-ldZ9py@vWZVf;MZl9n))p<&GI+zq0ZBg%V+eZmWF~?b#~+1
zQAUJzR@VG0aa`Wt)S^}Ij|w5@7NCb7zQR;0<^vmt9_HFUR)Pu0%CB|VFJP4#b=^;f
zU78CD2~{1mlOx)W1RmQA+qwGIjt&AcvhL1sS_T%Db|VgaO3J>giz5tdY$K4|Ws?Qt
zwzsW7mkVNp7Nn8GLkF9>6IC2aesOj62T1}hbV5Q&pmOTEtq>72srzJR(u1mZ3vakx
zDxb1R12r;RIrH#hEu}`1CPOx^84w^BxhEd39t-DZv<;Jzn;F~#*9i)D);l*Wjyp#3
zVqMl6{=im}`x!@VldP=SVF3YmIXE~hvT3SRNGPFvXG}<q70o9%2Dx!W25G)RO1VMe
ze@M*z3z>)QL2I8o@nX0+1b}1~KNo0VJIKj}$bnsOr}him_4H19KZXtA_U`@kXi{$I
zBPx2y($wCJj$c4PfYv&p4|qaifWXfOUy%Q~7y7KIQ3;WK0Z2vQ3&odn@P@zf?Xg__
zo-uyO%wYxD@ji_^sHyw=fBWYoyZNFP3Gzt)op1ODJ3@M@)JT(g6k7RyKb&zu-G=cc
zts#sNl1)ZJJ-cMAXSQite3R0>NW1*gBRV2#z+!udu~wkTD`SCZ65m6E_|w#l*sNbe
za-APt9R~FYrN*3)>_T}((K*i(t9`**GN<NkB2n{p%}_7?_299K@++ZxwwkE(?$&ZS
zo44fk%rO56#(+cGtuEao-tt5p=;J*v>XcRAF4xR^p@-9Y-YwufsL@M80D}F+#3X21
zQ||@8Pfe}2R1H=a$j_El9PELC+d7Hru$V2y(ltY<JIval%-pHxJ~IE6NH<R2Bg+xg
zKxZ$Xu5Cp_Z<7L1ve{zPuy4^u-nWm5G^VoWmfSYmrX5GgHuY}rpB8HeI@hE7bS-|W
zTSjezFAZFh$<)z(#5pGk4RpM{`FNTlA?lC*7vKJqDv=wsJ-Rw`*CW%n`TRJYbBA=q
zdzYxru&YF`&z&O$@~A%P>5s3juUWHk(V!~A(BHCwXRvK|SHE|t5512>p#9KwUzf-&
z27%hp-h#ah*8!>JC-OhQ6W8h5*d-3r><Po<fgWb0X=I(reF(`OcWiu?_VJy)b|D77
zeKdIwKIp3(yz^?4B~ZaHoHOn&`5_LGMi=3pmP5bSan-6M<S-^A)-=k)AX0<|j`#c-
zHL_^iLEkv$@BOeUbhe}t5&Wv^kylWxfC~$*;64NaPNvm};{d=_WVUykDu9=K_39O7
zi$89r<ygMc`rvzw5|b)jTaLKSo*tBkj~>0RtmK`nupKx*SlSpbqO~0TX<w#OVMAHx
zah3s6uHgB+FQPR_1cc$?<yHX!2}uV6qfL%8v;UQ1>9<5W9sfJHOeJv~*+;FVA5y;}
z`*pQI5wFmhd9Z`widvCdgd29rjFiPB=yW)DoU`(r*>cvT1PO|w*6^NvcgHy{*woB}
zqk$Bcw|`<**R$3_D_rd0?&r2N*l}09X60lQwRG#^*dURQ_;N+ADsm&>RWz!KH?KAV
zi5LMXCsnqFatC`pRA6_W`t}?>f!msfojttDajh62R*P#D6%{u3V>3`HngGA?fYSuK
ztE+2da(k{JWh7Uv#0&0Rf3ZqFTA()oI=#w^)daW0E_&%#uNLd!ySX7V_kA(91@V2Z
z`i69R5^)3|;s(3odXLyEIq!~KMUv(_DSG42l9mE4hyCr072DlTOz&ApUsDZ4He9@u
zCARIQdP8eQoWj;S`-rQa&ZAF!+nDd!gG(xW!*ZXE<${KdH4&7r68^cDxWW7d{Nl@-
zZy1ptKF!$-YkQA^u3qKvq@eZ?Ggmb)H7hFsB;&=sy#Y`;E3Z$dJ_iM*nGPg_l3MJz
z)^BmWJzGbfJMyXlb`7nM6;Ly4mS&BbfLg4cr|RkH84wg?czbib_GamLV+dePJE5nH
z1759<+m|NhH9cuAUx(L4{w^)h?y+`*Ls^}Z@&~)PUANhr!;RT#W^DS!u1#HhNafQx
zr=vb5-~#8~<n;=4+!n~Mxs90TjXj8Li=3Y=_<fwLb1rd9<#Q>d_b_XU^Gygie6A#H
zYGd?<rNLTaii>r3jvI2BwyukNnUZ`t<=%&9&FO%IXn0q4%IpJ<_;5r`FFYdRMT)Q&
zAfY@@CrrT+paiwOu(M+W0Dxq=V1vXvpti0~M?{N>_$edfKyHe3Bpti?@5f_~)O+#t
zwx5g8obxZeAKzw@jzq^~c>6NjQtRs5qiVzNQk~@1lfHJhBZ6k}L5i!2C{82it<K_U
z^AJwGIZ@6YvtcT)MLCzehyHEe`}U`OP5G4y^z#Nk-<S^Y#hNpIrY}d#?*!TMo*3+Y
zUu;!<TI}NBIkA&pjzme_nrW_wSzagiD>O&?W5`V*MM{7$1^kNc)lfql&!!6qf_7a8
zioJk(5WV8&?PpQOo_m{X!>5c#xZ2ACf6=G~X(piZ+RcmVdSAcGui52?F3<J`js`>q
zz+yL_spdjLLaIIL7sx+_pYG7o(`RZ@=V?{6o0K3>D+DO~9*|D^E#Y7@&?k)jo(;g)
zxJp<YJH%BXrv71e=0Bpw|CF!(ShYKp?)B|u%z-8Qhi1y#CtEclBjYh?f?-QClCR@j
ze7*<|?PG6=+APWt+N9_i!hav<D4Xam%y{Q_&_@+Mcf7lO`~XUB%DYSc41JEcp+zxX
zqN~0wU&?slIrLQiv10gF&(%9nFBPiO6~T9un-JThCc?)93+KP*)GQZ?X771)VLauy
zb?HBo5zn_N<3?tC+Xx}(VsMm52p$n2M#DxQB(P|snYG$mKNBDtAm^H-!NJ$USB00<
z8mz3W!k6o`RIqb>3LJD05*hEdMp4nwk{saxOX^Qo+5^CIwBLejN#&!>=CSbTx5;JF
zyE<l5k`7^VrKRfP;5})tmXgWhLs|@qi>x(3*+kbR71M4Vgj`ap3pVLKH-3Ie>Xj@v
zT5%SBVtsPE=Iz$!qpZJRB}nevPRkZpD_3yq`BYRKRsx`xTe4H_3rs959X*WcQu9CL
zFd*kU#>OgCE0&gan)=n=B`a@+DwEK5YhKV}Rt(fEPPypZU7soE?RMzDMrM!dXfTtw
za+&&97}Rf@nyU?axBJQ`#p{%m)1<d?!^xmb)9v>9%<{TLFP$aFoN(b)_mxHPR5>$^
znr;#ngphc^VI$Cl_%ctPyf(&bAY|q<{kK@o|BV^Vag0H(SL!uk4D<fg3<SjhZK4lk
z&s-2HA~rd+iSDm`n3RY34%L5>=rkHHKr&fd#&c|WjZi@y-UL_?Hpfu20#;Kg2z=zz
z9UNz!voC;B#joPGx{9e`{2!8&|B0z2EWF#Oe67iLQ*Ba;eb8dX{cCRy<L6Auj^=HR
z^Sz6$LT+*YOZ)V1iDRlehBgNL&;YJY+mWx9+7>Ge&VIAMn?Y8wvS+-~hhG#NVExg*
zt<{A=L$KDB-2bS@!pfE_Vf@BO&*7ETabi^{@uT|QOEC}rkLa)GI(&n#IBqTspW5Dh
z!4%!gdb=9D{j%gCbt12H7t^Zk+>d?p!(S!4-U)Y3Tz4|ck3VBJS#Xz#)WnA0&d9z}
zW%nkZ^rOEle|UPrm2O0&Hwb@9^G+`>6ar#NL{yYYf=-~e>Z~f3-mYMOQ|)ig(XzV-
zD?>Fk^K0hDixHGY#9ui*jL`3h0~+CMeQm1Jut{OMXrBJ2$mCX|;-o#`LHj%F?&duI
z3;3oNw(Zmx+`Vn5(bJ`B?jLW|g=Vi7APE1t@W)HH!}c?D*UKIqRJc3aQnb2_#TXux
zFw^C)#o=mZ&ry&!DP=|LEnHhdaHGq$b|rbE>v_9>f*pvnG>2g>czD4*x;xN5q<Lbx
zkoPV-R*vq5qjIF6;WN1Wzwi1ym>eeZKw8Hql{G4~EqixcAmu2?WMcAjW+3rogQAmV
zApyObbK%nRoaIT+$nI2KT>4tHlI{lYmh)BJ&)^Nh=c@K-xU92x8o!L*Tc;#?p}hn}
zgc2qm#G_a9WTrRwN9=S!>M7AOV>~dsyfExj<HdW5tz0;7C;FEjXRH-pyrz0COjU|`
zWccbfTD_01HzVHCy->>W#kHWDTbr(~W)+D)!R&-nXW5OqKHrv-<AaGuiBckS@;2E!
z6nhx~+Hx!wsBgO>o|J<80V*<waW~-$F)?WwncUHT>AGr?!zlP7T;{w;d;mAITeBx@
zU}%WIS4$mX=-)s+KUnG@ASaixvSMPbKTq8{VQskLu%0S^!om^?0NL-MTt9st>q&7?
zwl{wNz#7x_U{3Ko>}uPN=p@zj8%41T!2#z48hqMYpjnfoaGG{*+k#k+cPcP*iwd&v
zfF#Z7FHdg%a)4QALa&Z<5*_Y7m@WNKTI!HNfRyR!wg@`5ZEhUWukXAR@~S&<zF=%~
z%J$DB{)IL#SfO^$%CA?&-DH-V)6eeuXc>?z9(LOW-f{;kv^#%Kj%gdAd&G>+n)^wZ
z`Gv){n0US}i6FAwVk<^_81+Dy$jwWD!d%yqTsr&vh0m}QIPDk30SW=~Y`YbAWLz9Z
zqYP`ExSJavfMQY6(Z6%mSSL%(Tjra5<G8KKfFjz~(cvEvalhAu)6LCIaK8}+fj{^6
z?TU+w*N3wS056jc>OM)eB~3%O_=(s?p8ru?p)0w0)e|R+`<)<!k2Yp>0$k0}&`d7|
z^JeEJBtD%Ra3|!T`6(mU<uOr>eLY=35-f5;$1tzTfA@LA4LdI3<<y;%bC45i`D>XO
zm5V!b7uIKX?U3Jc$5oy*BvfWva@y?R$y>;?D=<%?iAusoPPIOsCm*;2^sTdhSJc3O
zhLYdu8_<uTX|J;xS9f>a*VLfJHauMJitT%ofQyyOZ=new-PxSCh)D^lh=@pgC}j`8
zuy^m?eaO!rU$a{s#OJvBB{I^tqJjql1LI%F!21AiBkChiAqGgN5{13^fqa$k-(u4+
zs6*mA5%UIB^yu;pHLPzF4pq{%GSAp2dOt}o@*Km^)s1UT?xopu(=f+Z;(9S}&c5{S
z(DU;B_I;3X<L_KlUfvpebJ_FfiiF)gUiHTq8-mBb-Mu~c?mfQ|EjyWbLiBSyp-XE^
zNux$2Aq!9Bvf^ia$y=_3{o;6?Ow`k)>D}U_4)y9K-o0dsn=qVfH}ORy$ws5gvPv2n
zwX_~?^hS#^1CbrE@}ECf<N6y`hp8z6mIEgW9NbTh62IM<ni9K(7m4!p?Ck81A3shk
zDfl-uhs$b9h?MT4!j|`93x8FO$iTnCTC@jGWYt=*bGp@6cAK)Tl|1#Y0w<8&GoW5g
zHLk52kOK?1Tpz)U12^!fL~HzRKq@SuFL}zlo%}nj^R3=weR;f4oP=lL37d<HdH2t`
z4u{&TM4P_NBSQ-uE!?#fMKufOk?^yP_m5}>6Fy40<Kr*4cqHH|swj4?SGK3GMim~s
zrLvvAFZNvH3A{MyGv=wWsMB9xozVqCiICY`RHP7@HAPs)r2o=s^{2vGxb7HRj-jDa
zabQ)6GAcCp!>GD^;#r{u9KBe^;>3#t|Ey*n-v^uqU_jXB@74v1IS~<&X{qHz@e^`x
zE9fEM#sI#l8lCYx>XS|2zUQB|nYfy*3<1{HXtvgkH%UG)C<uWLNCjQl0K-Nh0!y5&
z^EjGz9QXx-s@!Jg)j+b4w49vrO83|Ndcd$*{FF+kJ9DybpQ?D5E@3vJ2w$#ec7vCP
zijLqPW}U%OLGsk1H86iY5adrnUvjvN+PF-M7Q8(e>pe61zDOtj@MM<&5iQaZeZLNi
z&Ub1nF7RBO1K+-CFsq;R`XE4jrD8a>9f1k!?p&Ird!MiPk@!1IJP{k!u;!WZf=-f#
ztC}g*Ha>=^i6WX|wK$up5jxezyL$<^D@XpfHqK}fdat^ku9ky68D5=F@itR1Dr6n-
zDofB1$*7t<4*01c@xzx3Uh;R=q02mq@eN)(1q9U5cFDlv%1Ya}=f1H*9^8O*rvaJt
zZ_x)JYY12<GJgIHj*Wfb;OJPj6e{q5$A$t84NdUoeCaVA-QvQ+^Rav#H=trOJbk(d
zOe4ksSwX>MT!CvT-pN4ymkgzlwzp>o3@o7T5wzypw-2?-t@J^B8=0CKmQ0w9{v<s(
zIB@mwxZM<Xb#*iJ7ALs#W4F3Ni*q%2voq_A`)H!TGv?O8e0{I&y9n;`&~n<<fiJU~
zCtRtbbmpAZ+4FYf$q9(==aQ0=%x@mYY^EY3NYI;`8wdewdpQRRK?|?Ry6d0>eW~6Z
z{^U~}$hgk6)xlVQtxSS!A!2>cgY(^^hrnKTzEhRO2XUJA8~sX^7~gRQ)HdkjaV=_C
z@LO`7c+HO{Eh?;UCkmX8w$*B8%DwQbY5{pG2;VM2Jj!%}03D_MsG9W4f1m#PkL1*#
zgbwC;MWY4#ki`;$_zUQK^Qc%c03kn{j6-<$DLlJe%J1XjS7~nilT2IJh7<pc&PM@P
zH#dA@Vjyk<a*&pjGZHYD09?`o1`k1`#r;3?X&N&+J~ivQ5J+XwyJ)W~9sp|a)2Hz8
z@P)s>l7^;v=1HgdKl0UK3!nqD-6l@v(1v^GC-ykm*gg`NU;*@yhQ?;ej+fVd5fzlR
z%l!saU>pHfENlgZfg@N5;=EPL0fUxFNnB^=OF==wX`TKz;CLgFlKPr_(ak4{SKJ!<
zfuoRz-hb|1<t)JF+J$Kyft4BNQ2;zw83AgxxR@g~Ev-oN9^77q@x_Z5on2i`3wDWT
z8&FIA(!ZMowF@zBXPcYRO{;3K!?B3@zH61OW%^yo(J7a$0LzxVhjE_moKcZM`3<)e
zseGM3i33?%A;KHENhS3oB;75tFotBzeU6w_WN|kM){lorN-FK?DF}RGOq!(uV|w06
z&CM@1M{=b=5M4~o-AIDI6`Y9W)z$ligepsxTl|RrB%3qww7chU0rxE{Yd)C<?|`BY
zFwps{>AHOafje1diQcFna@`tED+PqXYvb&LSAQCn>?TBKyggB(Lwy5*9?;oPqD+Pa
z87!lVPl17$2=Y3NN@!ZA+c*tz&Xs~-eAz|$f8Kh5rl6nzte5J5DAAZgT+pN>9cv9$
zRpNVO&nl-^Mlcck6;4*W-;5<2|LikRF>0c%XR5#2He3_?JQDL`Gwr7st3I3c2+PX)
z3n+92u6ZP4kr1G$E$sH{&VWU$1#sZfKpj@i{YuQNx!=uL0PO>d6dS-m{7&mW6IL0D
zX_5fR(c9!-UNJzOZX0a7d#-37Rkfmvt?1m)%kNL-^DOB2F9BoF%0-;fV<x)2#^{n$
zqFH0!d;G^On@bAjqxv{Eg<n!>8Dcz0TE8**k{cZ6-DLLHKO$fI#Ub10#a?lUZi_{z
zwmAOA=g4eA1I|&F<9`wFK5y=aOn$m^@kPhJ%BAVGEHNW7YeSDSB^0?bP3>WBZl0fV
zd)^^(*dcOL)b0%QvD(`kc-fm-B9{qdQB6%UJ@0EST;cP1w=(aAPb99F8<_{Er+4OA
zy)Uo;AF<jmuve@7q#NyNA769$JsF*Wps>uvxpbm2>{|atf(?qR4NJ0jWG{O3y$#?W
z*8b`daF(QgWbJ<c*OBTW?(Sok+n8q?!XMV>gd`7Zlv$=+yd@IRxG6j^y<y2(NRN`$
z^Yg+u3y`H=Jz9;viLJR~sP_&zu)h{FDLb1r+W9@NeAEK4msW*{MDIQtZxFh}J=2;|
z9-BwiIwPQ?hFpuxGDYE2TKddnAlYknN6+&>qPrU=KdYQ6KjQCAY54za*~HFjet+aj
z+TZen7dPfb9Ezig5JXXf1)Q1lFBw{mxb^GX3))$F72otwP95225HHCxHZAM75BHu0
z?)WK!{Zy~@w7C9`d%J02jQXv#G|GOdy~uis8zcg=@xn*wxKxjU7z28PcrMHK-#=vH
z1l>45IcPH}!N<dEoU(0@+ZxT&Nkav^ZM#9yKHz^3hm{rZ0el2R%0DP-R6c+-0-r;|
zZ8hEqVhj*4gJ@{DeBL*5n!hd-ZkrX&X+@ahN=Nn<D2MTU0-GtIZmbe$AJam`kXpCF
zOixjr%Xbvcv?F055kNnhK(<sWU{A%`5+3EBos0&Vc6HJADS3=Eob(@>9t<Bw02koz
zoUyn~jb8^1+9Cf>?#<Ub@!grV|Gas#VIzpAQf#!J8mzhZEvoXqIM>;8#hjttWc^;c
z`)7Iup?f{uBR%1|N?hr-E(X=@xQ>R@g<G^*>8kQbPTS+*IyKG*gJQTvX1`KPW{2E>
zD2&iK$H&zHyROMhaTO0FOb39&47n2yj*bks$BTZd7d4l`{xbt>dd%*12~1)%+F{+V
z;1WBzij))QXnY<#Ut;207g`69_Ax@fG41g}x;KyB+>Yj6R$MJRL{2DiZLAux@s>>=
z8F?t&zJ<CE!}Puj;iuW}GW;aF6jAjvR6~L?AlyWWH{K}P1gKs)lBn8iw{j}CSZ)?o
zwr>O?6-Kt?PiP&=vcE!jL_}@<NdgF=@Zz9@`g2Ig6CR!l%>*E7bLcmt0-3MzbZ6>f
zJuNIJ>8Js(p|<;X_Ge#+IOtP%Hz$1<A}s<qG}dBKrafF9(n<gLgDNfEFE$&z?>p|l
zO3!v6LQ3g{*76bP*{(B|evy~i@P%X|PK~9T(QVDf;l6gW!6jrzQ)S^cG+RM1T2Bo&
zDO8J!DBF_a*^awfMN?-#!)_XG##QFjW7<JMvNMTUl4VA2oNlL9PG?;;^G?K}4;{|w
z%38JbEu5W3RW=&ogN+@Ii^`6j>jxgUnLEx4sxSIc5>z>r-G_j*$QCC%{V(`Aa|rkV
z?p|D6<a2``z!r^1MHM=<VKn-4^NjQC?5tq)#_1nnYdh;GmbLH2rFfwD?f18s&Dvro
zpKLl>ZB<Zcp0R%-t&xWBHh}OoA6=6NDgnG9UQT287a6>JvMe<CBxQ+!LcWB2_ggAk
zLKN$4t8RuZ)M@<spK)zwACN;(zweWA7oxX^XN}Xsc&~QG2}+t)D^qKXP~~e%o@$Sr
zie<ubvcK{{!0EjdXPVr!;iROY(G|;XxIR&WXsv+1_&&JiAk?c6$^kGwJbwHUpH@<3
zYw0t6^4r00&(AL}tMiMUx)B~G;Md)(zFq|W?#i9=DT_2GKHTyM&r{|Ex6zY0fzgO6
zL+g+&CAEI~cPm#GJ)t(2AJP))vqs1yEL3(Iiyp20%uU*OGCpJ}{K3VltEK2kfjeC2
z)3^eh|NF3T=FI2ot+bf6+^<x0Ogw&u#aEZpX#nd0fr{7jyl^Y(6kvvpxdxFiq0<Mn
zlEGQr9eIE%5xKn>VA3kb20b&N1Q2S^S2f)QUtD2?^93LUwOb%s@fn6D3Jkx39`HN;
zCFYMKWOz|*+>;~ZebmR}1|B*Cw^wT-H82=3t?O{Yk!HQlbjAiwHb)F^t~RM@GC@O{
zF?j^w#_@|+B>3UgtqZlDV{Z!iaxX=sAV_;ox8`J23n>;T>%i^KmzFfNv7C`$LjxLj
zK-w%UE&2AD7=o4zxXH&ZS%&FmumD(ECfTi8qZ)f}3OvH#v`@P(g_xFV0e1;NLrW>Q
zSM<Wd!Vp-6`v93M0V4yW=Fz9WValhif`(*Jt3oOGnwvjik#boO68FSKc6eWI(DS?O
z24`kwlIPSu`$KqgAw(^NQQ~NA!1U_;fMleDwV}*8e?=_<0V!YJL+}h*wsoHXt?=vD
zFRt>9=_*dpxqMGeT}^lay2k&AfwE5AS)=}b8ySyHFJnP%vt>b@^C&kk#IMC$<}ZNG
z=$+0~22lQi#HOU8B3@xL1E`5G%XV)-=p3!~<r3%_kL~yXMAW~S?6G?>aspZ=KyF1K
zTm!Upba$w!sR4Hn^vI5>sRZ|v%>$5xjzDr%R8soTvM`UZlY=j*d*NXRyZ`{cxx~uN
z>sMQiAvJpLI_?3G!o`9>AKD@*H<U1MZ|_<6EoSh@BhwqJu+;;;Y?k!jj4%FxRqV(g
zn+8I>EMgE_LcbgSLs~^(0S@PHGh$B~Yja=z4M2TtA8`iD#1a7YEU`1&%b^nlh&Dxg
z<qy@3e$?htjgl15`4xc6h28th&o0TnAvz*bXH*0j&M1CGPK4u;ceZ6-K5u=Jbpk6;
z{j<-eQEw*AE)QguqsBZFR%P(FS9-orB?}bby;fI$>FUZ0x=|CbTFP`~biN^MxJsrf
zyxTD=Ez}P)YnN-KhyijGU+YiOsC8ur?wOb$R4&R0^YX<fx1&{dKsF*=U|^f&T?ZDd
zKZtc`d%7wTEPHmNPQ2Z^Gu<R)gxC24Vsqzhbuh{S<}+vkod611Rn1S$Oaw4?i0Cs}
zx;~p*Yq-5;1)WW))dcY1Ma(V#<%3uOozgNfEKi<34Fs1eGCEok;6Jd45r}ZyirXbd
zE%(pJlzco034q#7%Zx!F`~U=w=Rq5ZP0r@w;5$fMJiUManbzIY!#Z&Nz;Cd!zQdIs
zOYM5&o@mX2ju~*4iya*TNIiWK?0@)>tCkS2)0AWN{duCd<Irnhfrp1~AudRO{znm)
z`PxZ0Euno$>fxA#RS41N{Ub{Tsq=ksjfUky133~;12y-?j1F42H*~hHb@LE=oZfGz
z5*5x-gmJo<sJS<_NFsYo?ms|^OhZe1aCq1Zm`Bh`=GUES&cfi<S65dSb#yMj8UCeP
z3H-ZT>F3)E{{*!0@TjO(K>v0Q46Mz#%+)b*RYm?ea~eQ(eyybR0?duf=)B4sVt~RK
zx4TZSA5eQyGNZg;OXxkm-JJU=R8H+P$7S8oljM(2iPs$-fqYYMP2_m?82K$%oLgCs
z-LBQII0%<@)HLJBY;kS7S3q~k`AVqPwPfn9jX80U|3q_68iQahjNM=6uwcACP&-dD
zj4C873>QR+)9!TKlub?F%#)F7O>(5W_wGFbqg24RSp;<B$xYQ|o{lMll^!txKJ_Ry
zYd5T0JmC#*9+;{E)m-7n-eQ`Y8OO&c&1p!k8A}jV*iK8lA>8#ebHxeEvOVJcQc=9c
z4HqnJ>h5mnZg(?^)&;zL1UtyA2i<G;XHi~&ipwUUDCnxdsshg1wm<tIf>!8x;h<Sz
zqcb%Vcx6%>Rr;YTp?+T7T)%7uw{-VCIGSKe%eWF#A@7MpE{I@K2e+}sx}amhY5x{>
z6~*In(-?8QBYxp;r#$3rYSl8=5>%t<M3eh77UOe}+l4pd_?LHIW<yc~uxvapT$Gl1
znk~v4fSvCUO#al=L}|MGLV63RbOdD&TEu;TjjN)i{zxjQjXkUlvY>ndRh<^FM#sla
z5cmJVy<<SYhKWte@sUNR8jpaW1-K2Q0R4vaQd)We_#!gFv<?`1NJbPZ2t)|NkA;wO
zf#U^HNIU!b>c(g8+b691yoPXy){wsFMtL~6J$Roy`_XNTwjnhbSxaN&axroXrKNwJ
zv0DT+cVcl9Ykt1#Mss9PSG?Z__(L1IP^N<7Q4mZ!qO!ow7tleM+%a9{_#bV$_Q&gr
zfNuYbfuRY|?J+SOD7b=+AnY)3ap@TugXEL=e*q^cBE9=l!+@FhE*Meqr*aR7<1{@y
z9#Y`69A&bp-k?i%TO|ckRXHY~>;K)P3+cLVJh=gt^k@K>K$K=l@GCgu`iqhspYf*E
z1tb<IJ3>I0H=nKL19Y{si*(s8;IIt-=Eikua=v$X?A66cQMsjSCiwV3fU2khcD@II
zy#z^yo`&WFaAMleHzN7rP-^B~CkuJp#l<yk4<_{lwrwz!5|NUk00>Q_$^}1iSI|O(
zt|nC>MHric7taUafWxDs3qS+<<d7zSg=ab}2(WgBLdvfDZsFKX9-Rb>#H;7TP~X+2
z8r((iIq?ONu~9S;{^E+vw@e=J;@J@vf0ZynR`p1Bmr?4pZrA%}tEZIC@yS*wgQdFH
z7E}1yL9)aOOt7Yy9e#)|XpL{`$B!Q)m*rcjnis@A*?wX~|95`-ARY~ixX@zcDNyWX
zL`5Onx)WmH?m(U&)w2Boq7w8BAAmUr0CQVJr~qOdLE!DrH#KN30l_2onR&!km45|E
zsLZr&c+jsJeZoBC7Z^wR)mfc|0p+E~xC=XVpJjtL`dVYjv+PtH6r4tOa!hwL(Hk3F
zp6th1qI1@)y}YlG?kIDlu7u^N?rcT^)!eXsdU^YHk{K2vbP?z2`j*j9%UK^7(H(0`
z18lzcogcV9)hfZQG3}7(_njC@Sm(jndgY_mD1qifCFG7Uix+yF*&yT_U@Oed%DSP>
zEiC*3*dJi?l-rxFn<z3w*Q&5F?u}(n&&VitSa~(3=&7W1AB@$}0#`Xu_5n2sEMdza
zf&KsoFFHcu11cLEs9K2V8S88ez^4UR_70fNYaSbm0kV*ivN9+{RZ(XW%f(gn)r>Au
zF7Jd=rt4XuXcs!ism(Lj?4j4Q+uR>$pP(U<BV$-Pv|q8jclYpcAvDXfs>&Rjycp*Y
zWYlpAt}qj!=m%jKl_e7Ng1yA`!nFzhqu$O{E(*&LZ@gY^`CdBIWI=eu#O>hlF0S{y
z!+-@6n1N{lIWjT{%odyu%P&C|0OXdql$7Jyj0?i<4n_jHmb)U|z<5;QgcV}i60A-H
z42yUoV+QOT`uh6EyE9W<_WMH$BF0T0(12XDXveF)W(Z7_F9433seb1eA`Vig1d{0H
z&x0*ucGyfR*%&ilrSB7Tm*L9()xx85zbAtA`qCj=TAr5`76VvT#=-D~G#N(KKjs8L
zKJL;?H?>S*bu>`(xrhO_jc8%O>(9Vl)r^fPxHP2}BbY&Li|yU{Fp?wPP2e9y%p}Vo
z1qHlJS}dHfpS1M0n>1qAKOKD8!tgh&;k9y((9`g*D=i*{m!GR|?8_D$VIqONcy&_X
zZM@L@30OCNsxpr0xKMfnpFajBX3ojv^z*;LSmunIhVT{A_K&Y`XtAi)Y(OeFi-?SD
z0do@BdJV!L3jz2Ym6%BJ2hq%>dcjW*y698v_UE6ES^LP*5rDA_2S?*FnLm7+d<mFF
z00{)+-o1A&3^893h4){anu!^J)JacIPqdG$nIrofA0faE0Ks1o`KPv45CXx4@K~VZ
zP}tepm;JGdE|VZ=eK4rGF`7pX`leFLu~5MKMbFQ_b%{~6#D+6N|K=kVZ)H0iNDJvQ
zXkIi#U>i1#{#p-fIl$Gp0{G(w9+q{X)y;sceast6IS1SOSJrqj(Ylw4m|t*620=*y
z`~wh)4Tpz%|CWJ}gZVQAbNr8o(GdroE^n_A{k|^sRG{3|pxk{7)}Y3KIJ|65im%2L
zmkN(VixHS+n^JZN{+q?}F7SOVfgImBH+QQxb+R)R^O=C2!=N>Qa}+47Ls^Qn)h>Im
zxSXG<e=w-aI3r7szsT#6y_E+VymCrPzxhP22A+T(tF^UNGiQApTLAgXzZ+Y{H@_fU
zc3|{lwLhr`FbW_j7ZPmi^+CdtP2wl=5!kKt1z+d@s#SMSkK6ov^jJ21WC*ka)6?1?
zK70VBXa2iCRkKaSP?@C~474`0o=0+*r#o1U)$g_d4|&t{H!Xz+YKh6k;R=&Zbrcb!
z>ON4@5ljVms_!W-Kuo_&J>i5^j3LY_4<z@5?oI17^WCH1{i3GlMdSnCF_<QWtD3Lf
z1r%nDA_KI7`pc&-)As0C#7qdmHYzG=u+GB;e5WLkfOP)VG6=>sguv4504=E-(6V!^
zErAyQQ@4&Rp--~Lc?VF)LI6}RttQyg0n7jnRCm;C1OXJoBU4jTA%y6}sFuGBT2*~e
zx(SJh5M#=~NQ%gaVAfgykmbkQ6OkZQI`2+v<fQ<v*|1D=1Gx7ffE&{Qyd0TC%p&KG
z1Oh|{wg#TX+wSdUX+KIN0PeH9fGT{knOgwx$9GT!^R<x9{*zm-OXXTpe0EkuL9$ex
zNkw(^=H3@rh$L8X;_{sTceB}u-_hC5l$5;u$>*T|+G&<DP)ES}N4MxSzk~gt5r`C|
zymk*E-#hnc09A>Ea_4QCrbf9HX%Go(253&eGD48#N}8IQdAR>-<pKlc2=^(Vj>Q0f
z2{H!%+G$5HX$u%B?kyYB?Ll}=dy0Y}aOVcoojJ#Yr}KZDRReqVgP?`ka96MtMGO)D
z$DH<Kle<n0{tl8?Q&(TVd^MjTmw<3~PBeInfJxP|`n>qr9>-r4e*&jh)i!=M<}{)5
z{=8|57l0a^z<LO>7#P^S24LU2=ugie&#qOsz+=}n3piPJn`vH`7<<)t6wpKcr@J?R
zzd0n=(%L-qT<+$Kp2Gay89Rq*De4#>s=#OHb=lBj7nO36q#Nl^{mAR){Q#F(kt>=Z
zYhSvno*<uyERug-JK1I#rP%lvw|}({mNb=pJoZ8vC=87i@?@C4pwgJT@toN&)Tdq?
zDdYrpREm|VsO{SiiF+xn8XB60xB6nUO_>*Odv)IflNp0hWyCfYfqm54l48R>xpunn
ziUp#>`+9<N6>{R6S^VtCcqAlJ`uYoef6~MG`8nB`PIe9bXZtrsdyB1yHQHgRLewa)
zVOGjz&TR}qxn+*yftK5>g+~PE9ibi32`^6LJM2^l@mzOMuckuss-VcYOQD+=68$#7
z`<SO`bkY`Gk@EYPh^xM}#$F-tjETnM#E|z@@!j=(=)|MeZq%!Eh58}SfkbgVoS^)J
z%4S(zWMjU9T3*kLocSaVs<S(GiQ0TY-5Z*i^j?$9Z@sEpVH^q95tc#0!L?br(5S>a
z1JrNav%R3e0;J6Xzm9CHr1%j>xOY8%BA5ktgqCe@q{^I$CIlFybwY<!kGCk}a}Fz*
zU08pht1V0|5(|n=>a+Ib6;&xLhp3~(M<$#{SZZ8^Q!|Jh(%$Usbrh^Ho7S!l3gl_)
zWjaKe{Mug~03J!#(FkxFo0my_CODxzI-Yq--RrY#-W5*I0@6UeErGveldren^bYvl
zR?alKn$P8>`_kUV(2S1(x%$q+jy-HH@-{E`E4~j%p@(3K4k`r60SuVW!L;6e1a|}Y
z|HrnSKShtg?yYTYI-`vApP_SYPKtr~(I}IjKNee~Ew(;X3&||E@I8YMVQikcqh(!u
zP`*gWRIK$TMcUa*+c>W~*X4`?mPh4cTl>B#D{jaKSo{YX28ZfpcF5lmKtKFduQz3y
zZLj@I&L|s2R&8G4!lY)~)wH*|<!z=iFTNcWJYPmu8K?Z4+Gv&hG-I!89mAyY0yJ9)
zGZYxoeDv&DRB36cg*A-tsSp5D2x1NbNWKbCS^*&;KR{a*Of5p?18zP6+(in&F9qh&
zYtI)4GKX|w7siw2)-vFx2p~PCXJ(?jA;I4S^dX?=7A7qVRw5OItIl~Ao|SPRzfGT_
z4}?&ilkbY&tZw8R<7ly0efIuA!-)|&5_36oEBiWzzHnvthgpY!>G8slCtH2E*k97{
z_~zv*RA;w{#{S5^MB-m|y=Uz=+rw9RM_<(49LzbIB(s0dyF=uZD)6i=vcYY#puA7>
z<nE|@$idz|RNsfbMD56fre%u!mNjRxT<YyG+So&fGK2sOTwrS;!=;1rgE0022Z7#e
zYA}5XbbkNkWg|dyf}6-d|4`z<rV@+=*7kp{%Zq<z?jQ}<V5EiAT6PFB6;kBI#XoK-
zYPi7|UN6T8l4}Pkh*JYAvr_!&SIerjXF;{CBn%S`9~XtN7^a0Aa=gsoId|x==z>UM
zbNr~Ur$U;H4!+q<3BJN%N;5k{2RAbDO!G@g`+!yHUeOi_vJ&37InAf3HOIxCsiupg
zTR2^i+8epdY~K2~h-;ZHIoIzw{jnbMQ^#Fn@lK@H7dx+HWz#`vC^a9t;{)nBInctf
zHVr3=jlpaZX-U<lAvY5fw$DG3S3dlE0~yO{io?gp2do#7U%r5{h>jqE!IH>_69Acb
zE#j<XbQKytnWC2m;sv0ns*>TZ9@=)yG|>B`+#8YZTXV^>m7Aio#gR(-6{#^p4v+hI
zXL!(x)P0{EWt5W|5vkdXv99Zb+rpJMUWVT+lyvZJc5>y@-WXXI{B0Cxy1^fJ$c0@n
z*N5y^nfx^y;(<JmKOEuCe)ih=9R2Ch>96&(o`cVo?+=$m6oCmH_KVhjJdi=|`o2%6
zR1V!-M&EG<*>J?AknXqRc^BRt<MWQ-+#Gt&i8sD&4Ghp%dP{tE4uAw^ZjM@;fej98
zcWKx(vCZHSn05I05q0#ZhS=Hej7Fv1BLJg;$pz4$7J0kM7CTyyWR+Lyg9Pv?Fy>oP
z;{(M&XK!`uB#;29-j4w@c!N^I86)pg%q~r?_lr>qmfMUj*FnJk9md5LICxW1=!zCm
zaY0RZwWr5b&MH@MiMhBza3uIa3!mEz=VWqLqb%QEq4`c<Q;j_?<Vj-#HRkG@l3{C!
z`s!od-Qe+!=Vin!I_IJ}zl^!N!$V+MM+ZJ?&L|JoI?^ss+tH<IaWne9)Al*odue%j
zXG&vMFpn7+$hjvsSZpi<q)ABB!s01F299|(poSw1f;dtmSrh+57gS2aheY6^H54Mr
z8I^=As=XG$%FCG%wg>l(0WB&@&5XR6*<aK!^=sVq>?wNTtNj*U3LHSWdp@sFpa7O3
z;P4Ryf1C!+&Na`F&QHZ8R5aEZdLdigIJb?z6fRRv%G^G)hPzB+!!q+?e}2VHJ}Z|Q
zs!uJRe+aI5*VO)4j<t!W!X~*Xb^KM?y9RAN4c9MeLGHH}9u2A$i-knWr5+1zW6WUo
zY495o;F7!_up6N2>gr;XJqrMAdD4N#F`%y4!AuK)O$y-l9Ee^qF9CGOz<9|E=B+!x
z)O)^8&Ee^kEy4&^WrL5+W-<JiP0bEBP{SVZIXrG_Ym1eoLy#}<-5M|oxD4noBj95a
zJZL2V%oWg=okt5=^#b@^1m+pv!Nq5A)bk}zCPN;9QLs>ED#?u|PS#gWd~U<Ga*^!P
z%hv88wI{6FzpQ&~TP~(?D)T+>?H>=Mf{KiwkeqiW-)wP>M9;PE_-tM75J*gCo2|hK
zcRc5qr^;7$AMG$xS6FEUw@kIXXm|AOhi|!)G!GdF?^!~<iF}a&1?yK#Vcfk7BobjD
zPE6<Zhd|m4WB>tI75AH>in8>#)yBMA{L{QT82^Vsy~?)iy+-n5S*JYy6RMlTNxp`R
zemP`9e0)Q;gg%fd9MAXl$+)co{QX%5o!LP<I59B+w3y0#L04e}=4-VE%yvJ3^)F>)
zhF&&4+27y)2KWh*<3A0hJ$cE5Ea*oO%DjB0@}qg0SlJ9L{zDPau#1aN_ZxBO3|?4P
zImM&?^?!gwza*5m+*)o|LBt5xeyix=|7zyEpPJ6|FdlmIV3Ac2fef&~QUt|?qX~|{
z(yI{28Vp6jmPHf@B1i<mLRAq&S^^@nR3QPB0E>Xin!p*FM++;FmRt}iQsv%Xo!ObY
z*}HLX?uY#YzQ25DKA(9$&+~eo5kFGivl;>NrL*GmYyuHxM%0o31N#&>{BMlC+MAv5
zECMpIsp{g3&(ER>p}2EIiF1mdlYu|Bg7Po@07G60$+r9XpO&mwTn`1nj=_G;($W!*
zGtjk@{3R;%yy#&cp2&EDxp`O+ckd4^wKqmQhZCD0slAD$nF%8JlBcI9AyWIzpN*_X
ziy=U=y7_mLSNoJS_l#aH?PN$<|FHj|x370I8%=Jh)^b}sInNC(y4L3a$6|FyX<Fn(
z8&SDkyd#9sE$%BNndF>Zr$6iui8TZ4bY=1~-?{OlVvAGPIof(;sOHS$l1v}7?kCFd
zr-ICEGrVza?a2)Sd+$dGMcXBQHE2S*uPy=t_%ffVT~<?fY<(YUp!6+kDHV<5swc$e
zL|K0wmMY|1zhH8G=J6TJOuLM9vC5|#cUi3(m^z8NvdMdl2S4rDGQZVxqd8|}u+kLD
z1)MAls%FMr<)o>`$Z+MZp4htGr3*OV>iab0jRb3;%#5fyFk*5Hk4sW0l=~$m-luyZ
z+ylJIug5M;sSA{X)Ims9!EZx@97GTe;OT|jy}UZ0#)u#dutp`1k9oDU23}kaH2(mj
zwga6wGvnP}0d#s%+}6kN8CD{Mn^|s_tDL|{2TnX~Tcy~zqB0=!0%Y&Icv_@+A<u4}
zo%Cq4t7V(zg~bgg57idmJ;BoX#;lBjBhj_3(IHfs<t6gr+Xru@aSlfs`g*;NlU5kR
z>=!Ds4l~KC82WHtVHe9RMb1DvVD!pSRLbgE-PPlFj0P6_YV#}HUjEvtO0d5Wj8fh$
zwl`qj*ckl_k>+6kl0s>fS8f+VAlKO{DHVa=SB!=0iKRIvVQP>ff1{-Z=t&oYAv|7^
zeYA*9&jS(Z=;SohUoKA=qJFYua*b0Vc0@+R92_KoF$5P(C2CwC%K+~Ri6j<_4WI41
z4wgF$tY)cEqnl2FMQfDBf%a8rUwSHaWV@M>1ZTCaa&?ifZ$qcX$JfP7W9H^~MTcVc
z<+}6>wrgJ=!|Y7wI<CJ%75F|$xI)DZKIvGrNMg9$CnQ8~MELGLsyvo;M;puOKP$bJ
z)~LC7ZQEpPu8eH!KNPB^Ltj$aw4ezJ&PUwD$cke75T9E7HH5!k%GT@r-`V=IU_3?^
z#pL?rE1tlx#YmYIfv7CLNq6gZmTwDJ*r3`uz%5vIiC#e!FH=-~N4+~=i`kTh@(P8;
zC!fqnFW<s1``pG$H`xhCFSK9b&cwRx{Sdcd6&w>s9$EuL8pC69pc?~?Bp0u1=*=S-
zCbs6DOB9w_by#qs$w*|JIVV$d?D@Vdo=H-})kM`w#m2Ssd3xi6Pj5iHnN5%;^`Gjz
zG<5Y07R%h96PG%}F}45i0mQ!p_8eTaok*_^RVCn(w<<AeSQL6vL1WJn7%mnsI>tPl
z>8dy0JBm$Im#_VtN2@wDuvtnZ1`N(0T0Xv`SJi?P`IpTuJ)ID@VDGc950s#9wzW`W
z2}TzUlUMw?0en6y6Wf~pHp;_v^j`ne^FF0|m2Ruzsy^ON>RY`+);8q7N~ABUslB-R
z!5Cfiu;|vSuv7=8cECiN#aN?8QhcrB@^N{&9lV#MX?LU{rlc$E&wrSf!X3)Aip^p`
zPl(6V)2#PE#X$ktfT$7@jtfH7E3dEGTfN~d%N)z@n_bse%VvwDqmb1PI1hs!%kd&o
zFwkRC><7(^zt;o^p~}k2DvmUC)A(TH1w)EqqU1fL=aw(106rGL3^4TOiNi5qlCKB6
zLv4sD;Ccb;K6k3%V%pK>=tmJ;46>325*56@@TL4u+EPt^qM(Wy1yns%ywdDc;Yfo<
zD0@Jb6pQUliJsGP)Zx*M)UrgE=uuH&hPrO9C}&I0pI=C)t`1fzD-4%t<L4~fG9F*g
z^y*VOfKFz2LkA%&q*VODV)A79s1{P_GOB0lmyMUESX2?VPb=D<ke!`cJE#JmO3lqO
zo;H~+_@~>W43l*fdNtxr)y?18SD$_z^KGHk2C+^zV}o_2b9iMX&Te`awlJqPy)&5j
z9<C$3(Xf&J2X+f<dEe@2&AiA_wY8ZwzTJD_*YiZHt&Q^qvCCLGosQHr!H%8$8^s!I
z1z9xO4Q^F%(ft8+%e-?(R!+|P#~<|(cNNgzVltTouiU(j4jXXBr-x{|;HMh{(SU5k
zof&DAB>@XNzp>DB`{vEIFWjh52@RdYkBUJ-gqMEeDH+$ypBqRqEjBZpZDz(qX3f3F
zyZYS|t(||gJBr{f*`vpf?QAgcDSU?BM6fl45%=>#LEzNEW}}hN@HdUx>E|kmI!7~)
zP$gHqk&I@p&EE*!0>-`w?`P|JU0?W{sRt<c23(&K34+qv6R*@@zy=DpF))Bc^UJ=L
zXI;|nkUNNJBCHcH)pQbuciwa&dZ?pXMLvtw?*!c&rh^ZD(e5h}w*3A)tbBw4i7i;(
z>0F+=b?%c;q~v}-xCg|ZYFis`<)bdrunbfH_4r42W;G!#3Mb0)$KSrS_4U1ruo(MN
y7z}5y?O_Nb@7QYu7YRJ}g<AAK&)P0-3HTg1iK|fYYKA>vl!L9SP1Pa)_<sUdyesVh

literal 0
HcmV?d00001

diff --git a/tests/tools/_images/cohorttracker_adata_mini_step2_vanilla_expected.png b/tests/tools/_images/cohorttracker_adata_mini_step2_vanilla_expected.png
new file mode 100644
index 0000000000000000000000000000000000000000..22f59960f3e300307626b1e0760c6fe1699c6a42
GIT binary patch
literal 23928
zcmdqJby$_*`Y!lUA|N1AA|)W8bhn^_q9P^T4I<qgih|New~7eTor~^9x;rEm-3!*5
zce(fMIs2M(&g}D>nLlQ{T&%plc)$03p8L7)`}saADaaDy-@%6<i15X8nb#1630`9O
zU%v*PrteNQfG1Jsr<%?xcIM7*Mvi8XypglLwVkuIrSV-?Ge;*&J6nD(5iWj?yKkJG
z?VZH9xo!US11>v93+|3JO%<>aJp1QbP7riM0sX-6a|>mHAl=m$GLKc=Q{c01?&M}j
z<o@Y{jXI2X5U=yCCuB;GZvJ>+Y@~5h4~L{@fZT<|MY&<?`yZ~Y?)M*OadgSZW%4kJ
zF<z5N|7ah?i^P5+zuVRIpg{^#mg#XY9$xQNO@R}0qDMb}x@csISDHtEXxFM^^-U5I
z5@}vxXADM0M#sem(5FwIZbxGJgC~st${Q`}8vnqh)zzmmGBVQf42;INCnqOW)zyWZ
zsT|)MVLg(O!Pe5!BGe1~^vOu(=~F>xyaRqQoV$#S#*gqJa0u6aK>wpRrV|kuYwN2H
z!;QJrH8-MEXhgM}%{Wc&fY*00TLKcrQTc&f(wmV+SpLHs3x$==Z~pMfDD3a~3776Q
zC4yZ(Cv@xetokMwDfUz4|6*VNy`u=MyBT>|b5@LMW#@slmy~4|sucMVET$rEO`6<J
zLo%pRm`BGR;o%^J(<=%)cUs=u^5axbhq|4J92ROn#?UC_?&?L#MVdWVc~vinvKIC2
zY;*a6sbXc&3&(1l#>{k05f8Da{Uu^!@bd(jz<r3T$WNxWIH7*My5eFEC+b<o+%t%e
zk1s*cYD}xrZ>M@aJNs-S{<zTZik*OhBRgVfg(G&I)_gv`bS7fRWCOeG&45|JPLr}$
z|2shf-bBu_y2S{0T;bgEh@$Yi&znB%s~)nVdvJvNwZPvyKc$TJ4{95|nBws@l*z*h
zJ$8cF=dT&1X>t+2V`J^l$1xN#+9AP5meVgMXkQp_(0s_P!e6g=vR7`nL3NGpCqXSg
zBVqP$Z2$FizxroT0-qVa+nk4mPYjF2ubjM3??Y7&hQ^}fE7Ed?Fq2w|u@?r3KbzX=
z^!=7MFyhkzLuHfiBLLT)MJ~7zbj{TSZ`)WEd7j^z`*n&lO<dhNkx~_|wjib&zNI9p
zD#F`whP5!G7RsJ+!$O;vr4;YbtQ=R~>D93#I3nGR%qqTa9*+dx@`F0VT85#)%@wjo
z9v-459pPqMWBKmx?t^8PDnv9QkIc<k@bK`;-i`({%f;=rk}4LN_ua|K$+>>x#!8s*
z+*j!!B8jtw0O$xlO&QYGTw>O9c9@QO_2daADa)&`KYo0WrT2+B-Wtc^aa?)MqL_B$
z5ej*=F;^!pb-9NXL`-kKx6rgsCb4pasa=*eO^*r~)_-dBjEyRHu1UP=0aURh0kMTR
zL3?6-*Zs%2r>29QFcid%?>asM<8Egi-m}>3l80&W5`)E$04m_~a!GcSJsuwpg{tD?
z)heB0HKYbs?Ap}qK86Tzx!i*4FE-in@$n~$O>iL(504)|fASeMW50g=dIruSpVV4r
zISjd@u1?Q~(~+$j20m?hTBQjOy@cj{PHCH(nzD{i%>>*^T3RpOzGYioT|L?u4hp9g
z=X*PHA3A|gcjTz$X6NL5cUtey)2R&Cuk##jE1**E58utcFA`pw^Mz#}>$>GMVOCw-
zVa87<TE5YnxhYZ>=2S5gO^eOWiS%o0k_&@66`kkV(JGK$t*hC3f2zdMpfrbe*>z&#
z2<>T3gZdP(SrSUhj+h53Z$AVWlp43;?=+x%d(HWMeSJfz_ybpa;%mnv?g--Njs{0W
z5QB@iJzcpQBl%M?T|ref24Pew-fv|PwzPLK&^djGg|qR1<|UF0O5eNBh98cxKz>vG
zYP8|PYW4FfODRrt+a}R9)_AY9H@j3f@gF^1EG8&pb$s9e=PIDPkkWqVWI+P`d_=6v
zJlZg@gZEx?lnFB?<u(bA@?qg@*J}MJ{FUVBfcRv@Cz6pD(5KIzjXJ_<b^Wf6y}=;2
zwzWm2ryFGSy~DzFKU)k+l=73DuCQ6YI-M_c=tt=$_&SViz~OM@oN=*X<+P$lf9q$`
zuI_F}sXX-_51Z-h-p$|NNO?2LlU^g8t0v||%n?26S`-Mk)5Cdmm}|?xSU}%bqZiS$
zVb2{wXD>Y4F~GoBK@WAP<i=NSdx?|i6)=OJ5b21DiZ|H#Ka391{z~q&%k#?SlBfKj
zZHW0gq+!ckaC8gmrN!cTiZHgN^zYedPSn?~yF4(=6RvC|)AI?Ruan@qK5XWj@ZHBj
zP1J$*zLh^MspV}j)k;cAcoJXDcpvt?N){rF?|G`JdHdtXk2^x(4_IF}si;2ptQxQN
zCY_2CU%#IH8@Yas@J{PYwUa}&ZcJR<!A|WipJ``g&3(PMM%>)(mW7NZ(}ZR9t^y#8
z(#)2G23Sdb6|thEnumq8^~9@R?X<2s7wz_FsjkrfDlbe)wd&G7eIWTQs;p*>$hxii
zhtCOHc<>MbOvCYauJikrkJWn6T0zr-NmF$t_WB8<uBZ@=&OuX4sHo&^U9J@BLtn8W
z6wdfNJ{s{59vpYWom<$eL^NI1N~^H1wfBZ1H%*Be$MSVNE;h2j0Z+hPpGDBg@6OeI
z4hv)E;tEelxE;r#)eNtkSFpC`FfuZ_K`#k=Mjs)C6bZm3ZXX|yH|vQT)jZuU9XR-t
z5iyi0NBPzB)`CsttU6_;9A{29{$%aZ@o>SwTz%j+-0H1feEpLAkOoZ4WHO2{-NCfQ
z!bLmdwo)ePI<ZHV9=w~QB}xlH0&{Z0=<mZ3oZjVXFRrcPVkfK4PSE9wXRsJNbNAYZ
zOC9q{TYA$Wcg^e5FFZi*eceJ^O1IWhD0wCC15#~NY^<6rd6!+L97f20F~=*V>f)an
zn+oY9l(@TrJvibAO}2rZ*n&j+_WBKC+dsE{gi{GYA3g*gObXX9rGouLD`5QJgcSdT
zAx676k;(D0+mTiv<BQTGhTlvJlzq#MJdE_{ecVG~gZ`BGTi+88uiv#G*^e-^T!U(d
zn+n7WtpD(>NH}}DlB!J8*`QxvE*g5JZ9Odh))RTSr#o~M$n|(L;lFa6R3}d!Q6`g*
zOFe(`#_2LIji&HE1FsXcqU~8|E~7So?&)>ybKGUe$J)w5Dk=?&8KEQE2r8=C*ajEs
zmjv^7trc*m3BnKJ<qb+M{?rq|WZQpV{EBt2eA&3Y8dUd3q%B!;bpoNdFucab*wNp_
zlt;eM>?#npJ%D_+9~ki#&brqfB|s%gZe>rPOGF(TRN-{lVsU@E{NnviL{wSBF1~jX
z!sMjLkh@&D=wWgm_t2gdmCV+>ODq2bHKkIbN_&(V6`8fTzLgY%Uee%FjgULG=b~fP
zeJTC!Z-uAIB~qy7oyg9Ck;&_i+}%Z}sHkS_TL`!3>S_yWVo7~_#d}q#rPdc+(v@`8
zx>`~Z_I8NeY)3J7JCu2QA>u=PMZrC;A5;VvvoHLEH1e-v#B<zD28bmRj!2wNFCP`M
z{N~t}y1?8GD^{g>zBut+%4sDv$X=a{RP2o8Jz_SfNB`+!4GxV5B4UN(3Rhp*LFUSV
z!F<<W8{sh3-DMyc#B@ma)vyn>XEPm=y&uB0H+NfZGRm09r#c=h#BDj-dGLpuT(ehM
z<Erx)#IiCU<?hm-q#NEz7bH$z9Om3|`Bx&jDW-)|uhz0^J15<5qtqX_&XBmhJAL-!
zrx-oP*-{AiW{n&Mok7Egn~L=BANo`Yilr$u$va?lO(t6RiasyOB8CSrCXARt>ZMer
zU8+`6a>P4^=Zg=qJ>8@2*w>4zd?{UJZ~U3?BIQMVG~6d-DMDO_CAne7haC8H#aLLk
zE>VYCl%9N;KO;-XH!OG<AOC|R(;4B|JNHiH<sn|Je*8$jomP65(SEV<e#Nb!Ogu^X
zONcTPH~4-ZTl=O-8~Y5UMZ=p{=A=J|FXza2y1%`#qod#^2zL~hN0ikRi8H<qIV6JQ
z;y8?ke!KvI+5YNyyf$7|RyH~z;W>aY-1TRS06RAJ<X*gS-d}1fvz<*Mq!!|_9J&`A
z9BeN_BXQWvA4(&7t9G`($f%{WuP+k@ODWK+J!q%4TUcM0)z-cPU{&UiAJ|#LU7VFO
z%^}?N&(+k(2P6?3>>35F=B2;ueZ0&E(&V2%@3ixCxs#|@DfzUjk{%*Bu`%9Bqlc;r
zp5n&csHgNFVxu;x_lbC-(`JdNuDnu%%^kWyE2=SkZ%4(Cb*syKEQ!bb?IHXxYtapB
ze$ZKmo^0-CIShOCIL9|9dLljeiqner4Tw)P4@n^}2O=$g*)gZpTc_^adA@1Timm9A
z{=gOpyPP-^K8pv5SQ8CV+H@!zN!ds(K1>?kTqA-e<ZQP6?5wURxE!yoyWlUJ?Jv(b
z4LtO|-0uKbYN*0S$I;2js;p~Zz+$S*(s~Y-7!yPG<;#~hQ)LfuaBw8{8nHMui~Pf=
zg-g889qIfo9f)YfKC~Gx)}79oEVqYQPgjI}{d!m5@5%)-?@#Fh2(uI9j82eg5=DwH
z{;aRY8?+A$L;ye;{pHJ(crLxO9s}e7wLx6FWovh3wa;j6-hgCdcF(P<UuG6xJ;@d>
z3fFlSgJhteuJq4}vIZ;>S}E0G)Nyr9D}4r}OB*e64{clD4tWn=#2jDjc3qs>#OxX<
z4ysj>+MmlqF}rdshpP-UsVG(`Y3}5f`TWB0i3;pNRIlZ@(`9sn@jk(~4LEneQx%WR
z;rer(Nz|PBOJa&!cjjZhhL1ijSb0R^H8<1TG;jjl6fEPU*d0dckXa9~!1Fn2^y}$E
zB5;+Zr{7nfkYGg-EW(hE_gGm;>FE{M`$z2_oz8i2M@Sqz4-<O#?p>MHczEqjO$dW@
zAdmF~&-d@&HLD$&5vRKre+It^PJz6e+}Ziuyf>i<)E42hwpm3*k#lukSoVE_<9jR7
ziU8q0Y4E+eE%4?h^!f8=G75@={SJC8!_vu5;o*{*Ya7E^{5}_oZl#W5H0z~XgqyV{
zh}roDKlqy++vxyBzRE8;TI~ldUYfq{#N*|E8V9o#ZhW%tVNKL2PU_X{HP1;QO5vh}
zg1YfVRn<GkgIxL9Sl+QSX_L_3=~|nSCM>s#hn39ES2sl5L18rcw7-V!lhQC($Jpv9
zY*ehCxm1Vy7YHwmdq8c}QWzoR(Ub3c^l`XYp_5DLi9=%Lnq#Z?1vRwk3GDl}REkgA
z*8-=?<@8@Xdxt<INI5g)>D<GaE_H}eQ%;GKb<n1J2bad%ys7Xg`6hmaMZn~TF6-u1
z`zObRv}#XdRms&}m$61fp8Y&4ztLQoC}kEDLT%pBy0^J9_^sklpCaVr%8DtNvYK_C
zf)Bk;Oc$F!R_SsXo0z<K`BH3E^l*K^VzoOK6b6BfjgR!b;DYJ}x_FKGm2)q_RnV&U
z7Uj3-Uy0SwUj)m%an*7EVyloJ;5dVPho#ofJIYuPu^o3xgF^#%%$BzELU=;Y3$e6!
zc%;kXdW4M;=Rt`Mqnu=uY!ezOJ!Z&nV332ZTDN`uMKvN;c3dC2-`_AX-_`H9v_dw9
zH=CHoQVa=}iKYf5zkaROHTQkSfc<fQv_bsV_0414?i25bE>lKRdl#hwkuxUwU2hx<
z50VNKuYK!V0S4>us%Tu<h;4Aco<Fp^IpfA5yY}!oGH0^P-5w)0$W1BM>*nqlo_qP9
z(9-YN)5Ud!)gkeDjVIqexY|5>VSfA|k)<vJx-a$m$w(O7L4E31Cua-$NRYqKw<;Qi
z?udBI!HZK`yxlRF=g|Cxw5||a+`)3@1U~y-*p0@lU-aRm1C*Bc*U2Hz<IT2*r!y@e
zUjCUVX0oe4XAw%yf0n$oWE>n4vN2uB?SnWmL!7{^Crdv5&XA2xOY09NXP=SXfAHXe
z^Uh4-<)wFXOAFqO8$YV55<Cvqe&?tin<kc(34r+1lE7<Hb<oW|%Y@YuL~PL>N(Cx5
zTmu7x?(XjAii)g4LMN3luCSL1PailiMBT8X4H`9FRMKKVJ--TkpcA*her8o&=%bi)
zj;&-~S<JhB7d2B8Jo|C7{3I5nt_WH$(w$XVWi{p=Vz~LsZpYvp1u2V1{plGZxP{qV
z*t+}TuPE5T3|`?#ed)S%_<=ZC1_a;YYl1Ci>=}$EOrmnAckt)%^zRNaYovS&40#xf
zBHbbPCo4j?Th|@pDD@WV_^x?bl1uho;)RDbnhF`Q=d5&Y-XQXeRw(t&$J_l)69I9j
zs1bFS1dG(LM0amb%@4v?T^T)}866)tYFTF8=ZtTe|FN70JA1~sI%MQntq@9(p3xRU
zc{rk?xVW*=bv$k`07{O@DhH-3#3?H;Z`9tN{j<>9k1Z`9uta}$g-`9TCEML1BWqpj
zOZHg$%!*ELIXS`V`Pxw+J(`&@SI&8|;}X-|V`O}9J6jW#nE1ljm{BE5!3=qM?(XTi
z<ScgHfI>*w*20ChUBy*?zFHhslyZrXd6gvaS$r3EsIhgrchY=5FY8+#CZIr@e#K2s
zG!ix3_J=rkivP49B>+e<x-_|@ehVu})W3mrmpP0D<1EO2|20(Q{s7X2kwIJ!lP);_
z6Q1n9$6ft{TQi-Y{;bnSVN6kA2Vk(bJMO{J+@HXLC;dTL0qb9WrdOv5hyesmU4#A0
zB#;m?`u5G*;KT+^<1r51b;bweU8_GC1em=W&oTa^H=RY4nB4<CW@n96fSQw8`F?Ze
zhu|TF@7$o-b6P%Y$X*QITV<(;-h-pE`*vwy!K2Nj{|XTQH>#5A<aHyX)H<yTVHA^k
zN(n6Ien6h^G?wAPjeBJS^LsW<1Qdlgw=;vfbPzYpEa}ImvNTK;pCII0*kSD9!8E~@
z$cXp!NY{%y+DFWeE5a+@{Ao-1c7}LwUbIG5y9-uhmTKHx)v^+~_`RT#wmyC0CQ#=<
z;*Y2<|5bq6J*%k3E7jnM4H}QCz2MC`aK*gId!*;BP0X;BA1=av=y!!f8Wp0SaU0|L
z^XFfys?I{;7pJ?5#>U3^TenC_UnnRPEO)owx2_f!MYief$ad`zL)j8FO|w5IIYaP5
zQGTj4y7K+qK@TRYq3FqsF=Vg)V}Uv|Ub-fmuB($5)}qtbHw39r6J82B(w(E+3)!7n
zrXAu?0*AgxceqLo_kP(x@mb|v?*%{mss2j@siYyxyM1D%3Y6191B!i)^J3^|Euu-+
zheoODT#aLn0o%oEJJjcMXIr5%wm$$on3$NpD)ji!fsRN|>4VwBcZfx*-{k(yVJOT-
z@f`X&&VrvygW4qxHOWJ>yDE_qAHmc3RJuWsc#(UMQpuG~u!==gttTV&4sY4YW|xE_
z#O-h{29CNBpK%NFYSJ%hfQ1Z^8m*|ZFcx?b=qRvddQql3(ED)J()nMOE2@;Os<B1b
zwKG+5O6C`0HM&|esXe8_Zk@wQ=$}acq&u^D59&?i55lE$hx}>9T@CpXH7l&CAh0-%
zx?>*v$jsyg7{l}E&tq?I36Ra4rRGifl$DgAN3UPs0(_G5#?VKH<@Vp@RvPMs`f%G?
zSGe}Y30y8&$i{uI3HNVRpjl#y$K7zr2Y!CJ>xaq>nY|ToP;GrX?Hld=0qp?znK=YY
za>Z&~8V?;?K74a!&(LS5*vwSsl<(vIp$w-{-+Qlhymxc8$;N8lJog=>N~iBO=5o!&
z4|8CT%hc~7k$qZmUB;2-amDwbB@&7t+2^&Sg5?gCza&yZ+X{b)L!Ub{a}_WfzdjDo
zEdG9RxpA=CvoV^h5tfciMAJ#@Hq+FAJmwa%o_OTw$mg`yYt#}*D0opkX&rh;Fqn$p
zyvSus9nEOGefu^{2&5mA(Hu22I4x{ICb?{lIkXT8+aLA`9|CLzutKxX>8u82E%P<<
z5gSbCBXw47e02&D#Niu-UKq`UsMcI9H^#sey$ONT#ZMP2$whVdw(>4YuT?Vt7&pZi
zt`F^63Y*OE#O_-neK#qdH@3b~kmjGk7{MYv&+zuhoG10{SorG5qEX*5y9V_&p)^NY
zhD#riOvc7JD9!EphIDdvbsm>ZRe(iaym)~p?z(;V?p=TQbfpslUKt%3`3PW>2V7j{
zfPNFam@K!VRL_mIDy*aM2Y=#(Y^Fxu@B<PvPn}&-Qu5xtdqsBh^`$uu7e^a_6se|n
zMI=Tfw7N?q>xL_;N`Yl}f8cDW4whrc4OOJr#O!!RG|P`>T~nZY8h{twMxfA(m)%`x
zYe#5}<Inu{BDZeg2FgIMpfqQAbDk!;EqbsX^>y{3_`cPtIP}C*N_J08QA3!|OLg3t
z(OZ?~EHJ=n+Wzb+x64zUX!`RxA*-xtPNm<{Y4Z=DNW|BOA;{2I_3WUB`-PlbhCzcb
zfM=4(GZO$MKDHS%2nfW0e@;8&M<G%&b8~Y$yPOmRuC-Tr7l|ScjHIjz&ZzBWJ;Vwp
zZ^c*JQIssca9?&YOybZg2mfS=JnGTHYQzG`>#HvR11xroz3@%Bl~3fVH0SS_LM~p{
zAbg$_q`miB=T#-&?hrP=4Ly}Y){eo|$$w%Q5`J>z=Jz;T97hUCwB#uAYkD*7N7Ug?
z?{iK#y4hQAz;`O%>Cv}lnfV1*)IaVz*?jS#&liOwseC1Ju@M$y!PO$X#v<`VeBr)$
zpqr{62!}r<wshD^>RnEq91$nYSL^5a!JTd2`&lrE?g)H}H}FklFNEFl$H2PA<8fdL
z*h85hB3d*r55TBn<$SFhH=wu$FF=KWrKY9^5bHI6po<i_?aq}>`5GGD1w^DvnXs`;
zoPXR|`PIXdA*PsZ&%(8rS*a#+wwX)MZ!u^zd@@L0B(w8NPO(F3S{d|rV}^6mh1VHM
z8D0Hcq#{xDJs=>n^C}%XSxp?x>V+|SAyp$3MMwfHNS)i#p}iY6imeNi=x}f0L1gA7
z<sh})h6@$ZDmKMMdv0EXdg-uTub39GP$70(3}l)8YZG7QN8c=wdyt2hEKij9Ni1(M
zGg4NHQDfW|vA+veuHoR<$IuNri3H2xpS<=9kHCzTi{~<H{(#q;z>5{s@)&S1yLB3e
z5qa_Pl)DWm$>olSjt%Fbd>vsi=Rc3^?f0lEr96+`w1?Bm0h|K99hZPaEy9-qOkq22
zQ^#r^;-VM(F69NeofmOO8WL6<xe_U@i>F@cqZPT1mC6q8wDr8g@7hRdgtnh1W*vEm
ziT3O;S!ng44jW+Rja^0dVrlZVizH;yO@Z-#V&r!j<2j;xT;f=1xg!CjJ(a~2ww`G0
z6Q1v__d)C(ALkFl0Wn#$OfNR@AgrS+gk&}EjHzvUO@9ww%Z)8iNZKC^<Cl7TW3_l-
zD*CLiVtzzcRh3Lj%R>0K+NME9?X86ldcR&k``cIRYUZoYfjA`QcSva&`j+}>Rb)v6
z&8*=xMHE5dZyf((lMX)D?Flr-kXiXMcE14oKF|^=<NoV9rXXN`4DOqk#&$?r4G)@q
z`lL(tKDq3_Avmwr#C1M{kMsURT+~@akJa1P!*r7L4Xu85K4qxUx}EOF(*Q_FG1Cdi
z&`o60V^#X^cWp*jes^jxf0_J;%IN<L@d36Mx;&o}3au?6w`sZLNtV||%-<W^?r|w>
z_~oR6bd{E!`iO9-%Kj3szGJ4~$O1(Nm027_5(!?&&A7(Q4@@g<RbayMU}6&Ciw=Ez
zMG0ptoI(5P)~iZJPxImV=A_eU+lu;H<aGuHa>+af=E!=2{it^z&fFu_G@BY46xT<0
zhcNz7lXh>D=vk6)2M|zNPG(+SB@0NteCRlJwL8(89&fuX(~!WEXU>msl%v`|c{b$u
z@*FW3E-}|Y=)hgb!kKlHqOU<>dAaF~eOF8@3B<sf?bko_J#;v|>z8li;&&@*jkEN|
zN9^o`r)S0Dg;YlCz3$2**}q{;sh=^Mk#4&hex6)qf@DFKd(K-{9(V6)wx7J|YUT_&
zThFPxu{9ENx&vD(gLj;~d4ag@r+;ND1t~X4en5B}bXOugHV;!jTzK*<$BEc(&ye}E
z1%;dnS?mw8{(_Y`ay#^4y6^V0U*F4lxGqevw=})4!C$3^jgs$Ph*GR9F*o2!8{3p3
zD9o4du;$wzXMp=%Jtd0@`SJ{E9sJ2LkM*eVkJ5K(AFDisU!C@{5ih?<pQL$VkEx$6
zm4Vb?bWmY^sNIm`3_K3r#lqi^^0M}@Y#MK`gzqBwzT-=qP?RY<H16`|$$SFOtt8wp
zj#YnR)+Gi@1<izTLn>);4E3H#KGZ+&uaguTiN_mnjU1GS%oGogN=Y1O55U1jxDQG;
zN4&tAG8RncNq$?n(WIY`cY3(ew(8|>)JB)W<!;sT`1B&9EjCYnz(mITUW9K=*yTYE
zKP-vSkyo^k#2*w~!@pEmYh2;oKVC!wDUX$jDdSl(@(KYs#y?1p)=wn8$->0`&CO4M
z>La<dwDf1HoE^|%SZDziKrJBn$?EFT3`n71h;uX{nJVVO$iVRK!v}0W{@frVbWj0#
z;17^hL0S-{_dQ!Y-~Q71U@8!SRV9ndci02Hp<mR<m<?ww%sko^E`oO<Z&=bRTN&^g
z)oXH^6CtgCFz`2A%J?C(%7~AT7lJ7s|DfUeu_RMeB<hQQdrqAnYx?7|bB$*8X%>&8
zUeNV$DL+~(`W>vsKc^K+nns1?KW#Y!FshgfsLWJ?lrK)H&+K2hUDk2z2)&^4aM$8I
zvLaQtqGxp67h5Os2Vs}rc1ENgh4d10-R{(}t9xo|yOWrH#t$^(#*U7lkr7oOXaZLR
z1lwyS0M>{vNl7n(JneIF^bK&SH>mhNfCS*M(n;d&?R}q}y~SEcMM<d<i&&zO(!l$%
zsp&mYQBhroHu9sL*<=b%?Ub#8&zEeZg2b7XG;Ax2iWOg?4BE*#-zP}?z@nd>P0gl^
zQ*llbo=v(+)qj=1ZC?88nW4yo&gpJe6R$yLzQ~SvYt{!$^L1RJrT$xM_lAFcmh!9p
z#)P^(_L<yLlx)QDXve+=;z@orb8&KQXXAIU&N9=B0ru4aIX6tO#OtnvjOw!_3}m=$
z*Cj9`Q&L_5U=4sb_5!%v?Zd-|DHobG&JPqN_OZ>Kj@OtNb0U(|^E8`)5*jgSjpjRf
zP1=h9VTy@`WjS6z>o)5`S!yxB!p%)=XJ-fS=R|{_6cFN>!IXF$a1Eo#>vTt@0*sb3
zAPQ%QvsxSqYPqhJ3TI*ak1>qe`Q|_IhGs~Y;;g0G{l4&>8LvC&>qfQjY~Q-9%PYgn
zrDNDDOV}54kn{oT%QYxEu1RtS={IR$qg#+9C5z*6INPEzd|INkgIbU^_pWVdznI~p
z?m0C*oSJe5L6e!KHv+>Zcb!ACj|G-ze)`~D9cx9cbB8i|e=yP(6ujB*PLxO#2~VOg
zXThh=w(Pkz2qZ5WDX-}K>};Q(hlQf&vK_|7=z#(iN3KTUPm6&x4&ADoKt%;=B0#tn
z+g5;0dzB*c`>Hf3UN{&U5@v}xSwza5sRd0Kr6vsu^+r^05%aK~L!=k&Jc~)U+29K8
zdXu;O6RmPBX8oMq++69{&FvH(fH>@Id*_ks3+CF=+<=yX7vr~RH|4V8)r-YzKj7#P
zIk=t1^<L`Hbi8D%&QZHQpS51Rs<imPC39V!-UB7UBZK1+ee|6E`+FaEi{-~JtzDln
z3$Q*c3_ZrhcVCUu3})Nj?&D-;8Hs-fZ!Q*3v6j-l(U=mYM>U<n>~|>=JGsS|SjQDD
zd!gIjS<v^Q=fP?}`*#SWkLkL;xW}W@^XWS3S%tNq1Yk~d7=I>9S%3ZdwUe=Zez<N@
zI;mhe4j6Z^*m0qu;bE;>*L<&8G`0XX279sG0n0QA9G*9Wch~w;LwZ)fmzBl1&HKay
z3Ry{4w->;HKY(3Bix6N{8TG_*qSwmNku#Whi#y`X%$Z}JGH;y}Bh+CIIvP_SH0aBv
zem8L1&5V&}OFp4H?`P?@Ceo*-lO^@z2D7L;n3NP!A*SSItQSR93#ReT>Ku#~Eb9uh
zCi3aPe;+7*o<@()4vaDQ#dyVXk6uT$3G%1*^jh5HM2uSFD(Cx|jO#YdSD(a4uD3he
zbv4)46z>(-N=&fd@!!5holz^x>m?%*aa0`|hD&gMHBCMHZY3V1z-oA^e2j+Z`RQ)$
zP1?<#QiC1iwNt7cp}OzVMU$y{bMY#FqYhuSZUc>}M5&_GRg6WN$2D$l?rX$!y#N!7
zxbNKsMe02kmM@W!jUXgb3BDyGCns-hZ~u~>K7g(ifm`GO8{1NG2OX2BC>1d=F({3j
zsO?~nL8TQTd2xShyb#xLr8DYPnnYq=9>;%7pM9#VJT2S_0P{CoR=`N<P9dW3ZQC}O
z8|VGcB$2TL2jlTSG4yFJd+q`l&`eKZUB56L-<~;qfhxM1=Jr#eS5{UwH8UGketMvj
zssY{S;$pdb7xHgyl}Yoxuyoy-(OPdCWyOIkw-xlS-_II(mYz6oum?|v-BgOV%$<zR
zC%~Xe#ca%6hpj;{<i!@+W&$8Is0`3vsjV8B8v{GezxOlM2rI5E>Sx;c4R+oZ!9$*3
zi4{V7z5D~{$A}By+ZY|+{V{*^KX<16kMY+(zgc&Aupswt)cEJs1)B5&(DdUUb0b2%
z_anS5*ZhkH2j{&l*hxsDLSp|z@+%Pt|GD2n-dxw$*9Uyu-)#m)rKZZ8nVA7LjrF$B
zUn5`Qf4I4SqtE{B&0jy?cQ2sf>uT0eK3X=OfPiE0vAViCnzS>%4K5Qnjmyyg-j0mc
z!9hG<UUaOUAr=*8iN-5EpNRmxxQxt?CXrNyaC#|)$B*C5UHV~xd8b|e_G4Dox|WnI
znp(W`3V5^3OB4V6c_Seqai=CF<V(*gqp)y17B10Xh0S!SZyGQY`s&jO6T)h~_1qL~
zi^^cu9pn26UjJg?kw~aMoX;7SOjTe7(;Wy*M7g6ycCs#v92}uAm_&-gx>h<7&o=vx
zv)I`dCoS4Ss3LJW-^bx?Zw}7uuXCc{P*RcTJo(knJ3Xl<Y`R2VsdfXV#l{u-Nf|d-
z4O*N;!@yYFWn>JC0Isg?na?6@a$7LFka&8$Qw1rL>p**A0hBX9F(5~`_HZ}6a&X{9
zB9Q`ib14m|t6e}e;Y4@I`1(o$^HsTIWw8nv*r;bMt3~5I$x{9qC1swbiyX2TbDmqA
zKNV76f{HFx-0hm-@K1RfN$&(My_ylZFv>6J3g<Z|@SdU9`-~kDnyqoMY7E|t<H82C
z^#OrUav1QW41qio$dx&FA3ZBZz8NUIwa)uX<oY%Z4@YNb-v=6`+msNng08HsjcLY3
zM@vVu$Zyud#J-_V@pZ_0npgN|Qq~fyajw%6PMwM(>q$PSpms+LNV{<T(|+;Yo1~=D
zrlu=6g~#4+Zhb``)K1;0K7cS)hyCI}&uDsEKJ-(b8k`D#BPi`bRZIFQBBH}uXgE)c
zU-ELV<yyhT_5<j7{!#~$w6JmSPR~$Wy+RX`^Ml)toN0uS`$dqFR)W0x5G&dCy!TW+
zV=ZyUi_3CNAI$p$qOI?CDPB8@<k!1B=6W2*AkTX~c3HZ5#y~k9h5stwhqsa#vLlF@
za@gT$cKNYX=94&uFsbjQm*`Vc{b-`YurjsjMY(bVC>j*p4W6|`{#}l1m35(^H%?Fe
z&RXcff25LCZBY>qx(2Ob+kKx~2}MRlHKAQ}0mPE->hN^09nmz;%?Gf$I*hX|1Jp~v
zR-jLOD_2G|-~iS_gEab#SpPp<nct4wS#X7Xo|du~WWWjiOs6j`e%%-Gd`WKvgqKl-
zIzI9CUBCJDvOl6JuEJNJM?9i8;idJCmQLhqi!*wc!vp6+*AH7AJ+2pEQumh`_u63b
zoL+A$J`jmX@q8+(B4@yn8emUX*a!@!P8|@(!xnoEaeTEJ26<^dorlk>+-eH2-7!X{
zY5c)Sn)fBSjKS0WkREJozAv``Q~EpE4=b0A{rg>}SXkNdN@jjmPMJS2_+f|~Ni$AP
zt2;RlYb5kiB()moSkcSZs#i*#%|0oqEMWASzLTe4|G{n3r|CqRi&u!Xz=XVtD0W#)
zm55xSmlp|%xREb+-)~7&4+XlVccMYQUV=$Zym%$~(o62PSmnU3me60ZyExN?q7O?)
zu|nb%Q>Bj93{+A>p_ur1e%Ecy3ftM>pdeiNWTDl@cUY$8<{+~@%2m%tbFyfRX*HJj
z4QMz(zMOMkCIr050v$6N3jpa8n6mcQrBI1rNwU03c=Ufxa8&*vMGn__xuH=95xr#Z
zLQ}xN1NY7ioU%8Hl~PFDth(UlP*pzlx&SwO709~H15%|}hZe|Z^2Q`#zQ5<xbsw6H
zAq3!0`|tOXZiV$Zi0YzktFIn3uGnETs<|#@7tbhPjiLnnmK3<Sb8Xkz3VqYM=7Ymg
zZTA6h?R`9|K}JSaWYR$d9UL4ymzPJAKR<r_sG61siV~;_F!qp$-NP)YtIvh?7r|Pk
z=Dc1f*7b;;+B@T_k{cf>^}m*sd;tYUcTW#cy@>$~Hyz1VhJZ;jLM>NaSxxOJa9wIu
z+Rifj><1DsSai6llF@W%slA9k!}w4J4;<F-Isfg1gl)C$pKDsCq$HY+N?#CB{_MHr
z5kAETnk#wQFPoZziLT(p-F8;K1<@J*{{A$pF6;0g6f~uSgTt#tzRYJLXM*hNc?;lZ
z!3tkqU+(~FrDnZ%wbxa)az>GM{@K}CB!DD<C{oxk`4<&iy9_%Sy-!q#3WYtCQi@D#
z{<flsv4<D&{rkdgNFT#!)Zi+&K|M{=mQN>#bu#7p|Dt<y=Mcs67FjM{JyThe+Ld+T
z!HOW^eJ_Rw@$&L+PZS4i6*i;+UM;PvO61`1(5g(-egPtuy5vF2;=sHKLm?4@{paZ+
z2Kr_jE3J>Akhb@;zMCsiATWN3ja~KC%#S;qsvtT0@btIunM@MP4HlRaz4l2Rt^b!6
zOPcu_eE&6#*(Wl(Qb{hgbn(NB`g9K$Y-=LYrDd$8pzthpZoA1-SH>D$Iq(yj3P(p*
z-+9Q3g|hq+``VmQ!ql13+trXvxtmbI%uC%3W+*}Ynqlxs+YA>JKm3uu&hz`P^Z&1U
zH}QQiAN4C|lG1SX!ZCT=2(Qr-^w%+qR8lT>7cWkHA*M6$-EUfOT%1YJ*kT1;61d&=
z+Z@eg3-7AsZpMh)-&=nYd-#6m%SivKy&7&wjqe|4WkMG|Tti^@mbf|^1|&MrXES6Y
zegb|NfZtnya(MOX2Jl5Ko=n>TsXPwE_CR8K{fr1eIi@rP;4W`%ZPj-Vy+D@`00w6x
zPC}Su!&fS%ZF570GAY!|K5pw>Tv(1<*cnKloCuZ|$QkUjY_QMe6>oP24v3U2mWt&o
z-^d>Nd>8~9E^sxATkOocx;|H34Xv-ePlCHqE}u<y;AJ}4i|UHhi@lbj?>OQys+PPJ
zy*j?m_*7O_MP0r9n{?1jR+&b=b{n8S&d{#yy~P%*UI+FE4;H!o&ItfH+%_;kTsqkc
zB4MJK3mZVTcLC%87}*U4627EaxUO_|G{oclM=6kq*0cK{Q!ak&?0i(iU==)EP?zxy
zmu**9D_zmE{0Iyng0Caz=I7Ie=kv~gIu31)=2~4|{+6XrmD>RNZ-R*4u^VR~n}!R@
z;yUDy7`8&I;}GzJXM&16tIVldcgB4=3~lDg*Qs1uUX}(nF!Z;;E8Oq)jcY?60wB7<
ze&FG06|n>X0Rd1t*MNUSDs#3wmVFWs`Y7aS!~Xt0Mq^7$04}}vO+P;rf%akbYc$|I
z|Cn1?p9GQ>IxIJTB%l&?yhqXVE4oA2W=aOcv%RAuYkxLBY8(ilXzftChe^OLftL6A
zY-g$tH<#nnwZ(@I2wtGE<qryz_Rz)has8I*>15yn7@hj5Q)vtQTvW;VC)6)M=SpO9
zvW|(Cf&vi;yT1-S2V(z$vXX@D`bU1Lcx+_GD@>aMQcBl#(oEvwPsWbbG%yLT^(IP#
zq@tncLSE~(>pE~&n3rd|GgBS5l(AhyTlDEurzP4?)=)EN%C~>Mo_^?uI`spy5sX#9
z@P_F`k&(NX*YfV}066(ZY2t@6T3S>JS<&dI1o{k)FyT!bXlVX?*+}SJ(66}yEdTS~
z^*Me%zPEZvu;J<k)VakO7AB@UynGz}c<ZAN+7Zysp7@vC*U=+mDex+8Gvs(7F9<Re
zO5JCC^z8dz1RLCCES4K`+-YY5c<BE%cdz6%z_c4b3Z%sV=(YV<v;N1vjqF3;hkX>V
zd^(r8-L{v1;d~gr7dfYt`#V_olpXw1=Y!|D<6ecu02?+LB_$~pRTsz#3F2<HspCVp
z|BwMlD{x0BD?@+o661iar{B}m2xJ-Ot+6mg$@33@67u|&_3dw!a@h<F9*F{$x5jmy
z2tz11-~Msj{D{6-odf8A4v??otjoB=Zr$l7daHnnOcjp@e3N0ZX4PUKHSoz;fP_e>
zr=E&N-z1`F=z!)CD(8KZLFqA7VI!xdHOVZB{}nw>j?G_~<$%$Daqt^}x~2dDJ_Z8o
zZasnvNKk}&`l)}Xs;CjT<})7ase*AXn<JkA`rn(hNEcL6JC+_t57M)iVQYqwo=R{5
z7c46--zEdj^ZVFpvansjq34|>P}bP$(-lh>SfVW&+lm=B!-HsxhV&XQE~icz`csJy
zLIkVur@LbaS1`ylFu0Z;-X128HB=-{yIr%~4P=mh=?)8F-g5YF^y?OSqKftEIa1Iy
zB}zQNEZ<-_eNf@K5H^>+mAnRK)PKL>i^>%Z;{_BHba1$W^7|2jS`xfop5q=3l<fbQ
zODPrjLo8*D%1IjQK023AEKj??@dx^YVS~Y7d;9y}02BdbPm1uIJC>AGgIM7&hK+w*
zTCh3vrKQRAuy|Nf^yW7saPUpk`-lV04p7B~`fuMp|La%DGAf+}1Sr$Jdq2L({c;k+
zv5(SpxC*%UL8xP3vui*2)-C?Z{guFhDz@)}rJfOAWBquzD1VTf@e3YPOLjlcFHev9
zt#__d`K@e^uzAh7sBYoD-{cW<Zp02~bTXyMPuKqJmW$o1?>i9^t^(Uo39mz8OWG7L
zI&RZQG;3pUS<4Q0BDd|x?9Q^sd@&1YUPMQ!6kn@sTrE=ecq85+<sDd}m&^SJdHV$d
zQP7xGEoDwjaPpcO*!8+D7zrEV9+x5YWF-RX957-<&1ke@A2(Gjm!FUV8!u;pRe<bQ
zO0J*~;2fj5@T6x(9CB|Hb!kxpf(FRcV0H}xSq~VCDgwEF^^gAl(SThzRmi`a=lG=#
z$&;AA!)UQ=bg6?w;v_qRGe<2-$*PF4va)iMmbMS53U_LLh-s`dL0Pp??=OWS3Jsoo
z$|7z`lbpFjtC+R$v&w&E`<2id&Ao}?pp(ZWuQvIh+{w^M)1313OkB63%RM^Ghp?{7
z*h5X-27+3qYtGwspA)X06rEioo$$fd?-PI5^Ypylw3}bZc%QJnRLiT`&Ktj1oo76)
zNA7jp>lz<X^W8@@ndcw$3gtG(D)R=0i_SEQ-F5IbhOpCFeEma)u%5=4lbakDFOlzr
zlo|+tj<jJ*-(7sSRi!dv;*fY8+3M7vKH7Q{v{8rV<=WN9V>%uOU99Kh4am{K)sbKS
z?|8IpEcAZ@y9+3qUcP*3FP1vv*ux0}u69*5wSCZ<QsTJ!(prcb_$Vw+wkP9P&XfLH
zv)zt4MsN^9ylw-;knZ?*EV*m*1dl9Z!+7WGWxwSwm0itDBiZzK&&44}wwF}po2i}!
zx>NOKxK@uuAe<M6T?U6z(B*PL#%OCMB*HEyDuCchG5eSWXRVs|kqJ-3^RE*L%1dP0
zs&@SB1*24#QiL2UX&5Y8jG*^ZK|?1y*bt2n?nX$(a&}RAMfx6HpQVcvF0`E~bka~+
z?n@9aRtk<A)OCY{W+T;yASPU%Ie=bBI<!mTwTg=6bOmR1b+zEs%KAFL&6I$U-Q25}
zFYy7bE?_ei^Zq?1o%;d?1&3y1dl)s^(28!_0kwLwaq-@q*KV=H@-x6LJyTM;3Cu8J
z5ZdkvobD?C6`BGp`6@-z=jafaGtd>y#qk!{>NkMGE>{5@3*(ks7BF6&o~hr|8?&G{
z95KV287bjcdWdg-HSUzKC(Q_{)DkmI46m16arDo$6ZXs*9;_AO?H?^>(g_<2T1vk0
z>Gr%jHN8Rzb)Ocp{>hoftGZ-?x=)#-&kt^f_oGN5c{?w>QoB7%#Dlq*=WAtKqR1|T
z;<Jx~M<?-3-+xIQTh3qR0}DHQCN2S9nB7aIj?KFoj$c1Lw6k4Yu<(LD-I<~YIo&{U
z0_Sk_$ruF@_k2WKM?m&Bn<ibvr*{G{<&0-hpqvJ39J;y#7}4g>m(3ZEtJUWU3M`V6
zY5l3W^ZsaBht6Xq>W<BSqu59X-Y^0H1~_Vp!R0yGogV->(yDB}&PzBdDhl`)0)W0p
zQ#-3*WW-SCb$V?f$*N$ni!~h-w;z13E_eZx4*~bv0}hUr<p`;3ITqH>WM$J|9|BKn
zqU2Rpt(GXQi(hWn#mQ{()z0pGgR6jBCA<k(w7c>-OhCs4hZHObSdI<fsIp#@saTEf
z&DWGS=P>72Q@~Y-C-9dx`$)sazOp4<OGO^t7)z4>i8##EY3JKttM%m0?##Sw7|MS^
z(&WM18D97o-ItY|S<cgrmo>VwkH)F@wj}(Q!qWZQg1fAn%nC8%L4yI3Vf#Gj{_KU|
zV~yEx#a_^~6Pm@iJXEXUVWRv+c>yW8Myz*zi{eGoi8YOx8d=sV+P-awP<k)+fpn1N
z!+ZE7XsG2&wiiuNy0Vg0Q)UMCZ8CTl)4lo!zMKJjHfxHLi%Wr%#m~A;_gj}Jwe7|+
zc(f!2Cb)pQ+Fv<)d$jADw9R@Q2L(Y_SJype=3>v|x1J~4T>u+{mNhiN9~0x;?X)(K
zJ^<h~U<wF<(;jv*WtFE}eaj!HlCi)-UkYS5bSDCuj|1A`yPnl0V4Ah)jAUrgUIq<`
zvl-VQx;cFjzhz&8`)F%CEpL3Zvji3?OS@am9e?^mV*ceEo%<zfMTH{qJ(to{Hz$b@
zpC_*%rvkdJS-9t((oje5TwVD2_~kp=j~$(CoIkAIIEmp3Xa30fQunmFLH&ISYXjFf
z$;cLQ4E&r9@S&Xxn_qPu_GFXh&hh&X7_B=_mL~6AYv{*v5ZzX#n<9F2=rwn@YF9J7
z*DLZ;t<-XV0h5h*_bc1!-MM_yy|=%qES};48#uKiCASvZDi5Cu{nERs!YA)3i=pKE
z<3l#}^m1OAN|a8ZMm8~q%@YUFlo)z$oUg~b;<x7VJ&u1g90yfdt8%P%$Q6^TFHlqY
zE2>D*u%>YfCO)hK5QhN`wL}V`!CVb)bl3s)!&6{9^8kLi;>ov26LLyQ6<ys|^YuPx
zfA@5yodr;z(Dlea)hEedMG!PVT?FD2DIJ{x5G2~#+5qf{P|wd{?E(J8-+(0rZtfOf
zv)`bjx-ID~!f)RDRWV)i?Be3Jy81IuPqE_S;)YchOH~koKNEa5b>9-nV<7<j3Foq#
z!mlPvLZvzkdVJtsL%pp2e6?X=qPXYlV`|%6y#?ik-a4Xw+Y`z5+)}|gY+Al=gjHV0
zsi?YKh+;V+Y1%~HpSmzC1kW3lXcn*2Gg@q@mXM~idQaQ~{a6u#*;HIQROD3DU0N5!
z*0OI=WMLTs>5&-3MugWpl=Wr0D$Gn-B_m2<e=v7rc_-v@_bk^G?X9svHq=pOmQ}Z~
zKTCE_I<}H`k63fZc3;eMREkq;{VkZNYCet(wB@cO{mqY`TZr5LQv9i9H4T!5)?&3y
z_?8EQYQu}Iu$uZr-2&3(=JhwAPf6l<L<JbI9YH%2zuS(kknPNKbgA>|)jbJ`6cVOq
zO4{1NU8dhbn=$NsEg2gJr`Y4rJk5Ox4|MSS0$t323<ILrYb~uVU?=Ma^*(U&d<8`y
znomqf=mbR`x?wjhEe*&N$XygNV%oNr5cEKS#^kf3VZ~V{5<rhH0oMC+^x`3`&K`Lh
zWnjY;rqW@7>IuSfS#jpQXvsP0tvi$gex8&d;H^HKOd9x?s#WdhDnhe8SsDZyt4)eW
zkR&ff#n%tD)7+q{iGP~C_3<G}ZXEx1xVOHX{1pWW3GZDW-{(LWA-9OdiqF`=&61}e
zH75=eNck%<g$0&&Ki(WE7wJ~?Q_egz+b_O_a!xPB$8fDpZMcHlp#^5pGd=otwVS<g
zB^`Ci3RDLdeI03~f2t1lqkmP08U2SP3l3`)=ms5Nu*(Cj-jR_vb-5@(0ZIZs_>0Tq
zzk?8duoy(|xX_4kYPKrRLdqnouN^-(cJC1=ztbmS9{ljBawZ807{Kis|L~swJ<J1s
z4P6HuOH0eFnU$T_8R(+wfZt}iVs%W%ZWZL;%&%V|D88o+ZTtdUtlw1^CCT5C-oPC!
zunj;vqN{5#{n6bKfX^klb7yKWbCjyZ7r*gs#m!QCAE4f>w}?k(ojwzGCU=ySpp}Rh
zxax)f_>E`dYBYEDih`7a-l|Ryx6xGec&EI4e{XLAl;Z2)QFthJhv2Uor&Pg>r~pC|
zI2B$m%};(c5oosjucY>T+4Q6+Wwgz6rSdlosHjP>{?>fFchN`ruQ&3TWT4J#vn+&y
z@^{25+>I+d$9)<oAN;d!I>9GGa^JCds-Jplb3&B4>O*yW2lLjgTW7#AasB%Bvr$*j
zuV4YFU(i>48RG(angp!ILen04jj6<r{<WLkA^uxJz1#HNGBI7W0QFP<*_s~<_$BDF
z8drXrWidhGe+knKBo4I#y}2@6_)YTK{C11$gnG~FlKCx&sD*7!`%^^GMe@XiW<B!4
zD)0tXGnn2`)pEFa0*?`9M+AKqsNjH)+CPZi_crt|tqLTY{XuEsMSGDyzcX-wFM0_8
zg-kyb63Wx7jq|(OGC=E7U`_E`jd9eUEfJu5r2##hC~QZEW>ZgR-3;osbJ2I{u(W8G
z_X6S@5I&nt(gmGyDBIEQ*A6KpM&myTWxygEf_%gQZVupgxlBLf&_UaqE|m`MOn{Ds
z18EPdH*<2qfdtjkC-=1j8o_-%)vXHTIj!=y<aTw(>R~nj;r=`OO)ETOk0m_s%@4Hk
zR|!1d^iS2YfG8=W0BU;-Ll8*vG>hN=QprLiWbhAtLj?>B4D0#&)M>khR3IOCp6<v4
zH4iwEvwshO?jAsB>w3Y<A@m>MPx$#P86YX6a&5q7F#y{N=<?A^NR?<RVnS13<y4?i
z&A|7Z87zaNt?>aL_62a;1EXTi>x6{N%*?AP2*Jd+bDuh=5m{ZIC_->*>vrt?kE{eR
z|1Oq*YlKUnPjC$^V{Ah}L7{bdV2k<vO&S-#F{>NvyR)_NFjx{Pr8w|}0tyZYA{d5!
zNrKz6wWoO%^tW$62Mz1A@V>uX(x*!_cZlY>!FDkX!R;zQ1mydd_jEN{QIY}RJ}^bl
zs1K+LR!57gt8IwghJlO-$(Nu-(S0kw5^cf-;lyGF2`<F>LI82q<w4K+JPN5nUwi!@
zbPfxJV(;9yV^>;;BhA8^l7b>|c*7YbVcL09w;<hCaR%f(Nkra>R+N<8II!OPUhd06
zz`))(Zr~>m9O@XLVa+9M-RJ1H?9Q$ia43MlIRKMxh(dEc0(Z=Fg*kV7s_qqFpMlOl
zev|eaAYuU*IG}PU4GiW9Dz5mI|HipC>*GMVg)epJTY-SNPpH>!X-R}8hW`CtH{i4D
z3;*6el4}7^lEA`)9}0U<&%AGA$M)JVIvA6b)|%=lKiMFkj2`qI-oxJw&R`5$%131^
z1U01A2U7Tara31Roe#}&-=fyU)V{@Q%B3$~(6~+h++*xcxUaxj%=&r76|KCTF7|@2
ztxe7rO1$32wt1WU)#{?lN`Y@^4lEUtx#2^ZJ|^AinakAQeQDs=B(w5;{QfftL=0ey
zkTj_UkrH_AG`DI9Tu$#*3rsq?IbMd;&P05frH7O}oC9j*OLZtnZ=+DzB2`H#vBWAR
zP8ckuviVnub?@JJ_83dNpZ<b?M^89bkPKPzA&XB_Dgq;O4H}iB6uZLRARaB4eGF}u
zv%!*Oy2-xX<FiK8LRsqhDXVwmE6ip3uhyvb6+e^)Oy?+bqmMqI`a<NZ{b8;qO#1rf
z$pjVR)-OeJHKJg)oXtyR<<+ykazJmVkE=n{8P9kzl_<H*jg1+>LOk|GnN|yk{<EiK
zeIacCQ>kj-0CJZ-ZmDe>Ri)9EJJE16DREzNR86apY9Mkm+$Z{GKYRo9lPLQxP<waS
zvho;7Fi+2mW!A}Ra?JI4slb+L-5O5b*O~fxP)g<p-5m6I#i_$f;xMZ-;rP0klAiJF
zp<Y8xgk-!P@gw9^mV-LQ(zq}7T$IM*@P%0SuvEvtx=W;ehfEufbAb{Y6wHRJ%+0}C
zn3vO!AUZz2yEWP5?xnd2+YM0$GUfj>y(i;$C|_yTx^jZZkqM;Ezc-=TnP#mA>k4!*
zjBLe5uMDj_P4Y-ciz}s-Uz^_VgWLt-*ZkWGsq#D4e_t#<Nt#dlAp2%r8?krYlEA<2
zY~gcZ;%HC6^jLkLIu=H@1Z*d7>&7M$J8e;~6X{8_AYtMqyHUJp^OU%aiGwI2(v|h?
z#l|(}*r4d*Gr>0g?^{~q46oy@i7e`1=U6MTYDP{e=WaNL0E;;Tk`u&0KF~7`niab`
zJF|bU#b*pLJqtCQgLy!p9oG@uVg~K<952wL61Js*K$Hum77hn88*qcVBhS}o2mJQi
zC^trak)qv7_wGd|YF)XRfUcDrw|V?`W^0|!SK|)>-bh1*?D4zsVH*CJX6?@8&Ys4G
z@P!rJxSd3EgNQi)Hl)xE42EMHoRg7ABRkO6_F_fSp9Xq<2l5m8UtH7$wF9PafjNHu
z!mrJAzkQoTNiSUy+kBTNg}gdTg4OL9J87oKpG7)XkynI`K!NOZ*ulSUMac#An0KN?
zr9X!vO0I8zxdVL`^M0s@Kwm}oBQOe|j?Diu;0Jfx0e%M{+qPK?MW?0`HWF~d9)UJO
zU~2>F^>CIVJ=zf($snB|WOGM0oTdr5K}$;bfE)Cm?c?5H6l9(~`w|`9Law3v81%JR
z{{G%(H#Hzeg;90eHeKX@cRki#ZNL6{JH#ToXaDnN5T_gNDwRP^yau`Id}~WwY1ok7
zu5#6{i;>2zj5j<+CGJh1MKRRQN8$PJ@}VMZRljn|ltonT)^^}*KC+KjO9Bs$Al9BM
zC2eqRz`Fb6)Ct?**x7}tRMthFwOO|9ot=nCC1KEfLsDMla#3o+{Qp+Uxkp2}_icO#
zDaw{qPNT!KQDmrxlEW0Da;mK;V&p6`jGPK%?+z4FIc6BPNI9D^&Sz>w*wH4<b{>SB
zGY&J%`@P%Gdf&a@=dku#&maC^)?%&OegE$7_qwjn<wb~nev2(1f-$Z|U)EQ4f(^8Z
zPVXvmFhtGNusyZ3JA$?hrCR@A>XYFCwO^=zWe98hf&cEMy2K<Tgq4;a16>8-e!u_&
zcYm-3TMzdDaWSO3S32}f2pbkt`6J%L`pX?U{WP=w9S%a17VE?eudYbR*4Cr7F96-K
z72U#TyCI0)_iL)K0=94mYW8J~{-evMUO$mNPI-yRtQ?SMP99wGN9$dx4+(O(&vX9k
zNUeLH=cQ{WP~D*l%SU2h=9$Jf9?f0*`yJRp0H-UI8Tbh<G(En5##8F!(y1UTR+j6O
zRIEpVP6M9B#>_pKq`x=AXK49I07Wq*`iaM|SGDT-+DO+emC-8$KRj8=t4xhyl<uCv
zc37ToIu3NlcDk9Ekm8|3?d8!iv)=y^&d#P*m}OJVc=n949uOude8o)#)3Zeh=${>X
zZmU_In3lFCD2f|C6)ayn&G{sp6~JaNgBIylY=XruFOPYzFZ}E+Q^&S-ga<6d;L-vZ
zBw7ikBX>hE-I8e*bNR5&kiIb4<QH)wZ{5*iZ6I7m7-MqDqZ+Z(>rLy0yJb;zep>w2
zrq`!6U0Hhzr)0OtuHb^KMSfQBRj0>u@(sianseY4L62V_k-%Y|7+r$7?5v52udGu?
zDpr;>{o%G)@eM{iM7McDVGr)kx0$l_hk+gu=$1|HnUZ-CcpsMH*Q10BR#q*Qvo)>o
zn84d@42Cag-%4;v4UX3dUMqBJ-GvNaumdYvaK5UJ0tsL>rD5r-&U>CQEtX|MvKV1K
zfS|P<Y|zK0x5I%HI9r`WbTsSN@#d_|!#Y>tJLP9uY9_7erv&42q%4*U@ri<bKNI1G
z3qg+~d)7@WYYyS%4Y4)(-M->i2A*!5Yune*(s<TI4U0-U<S!*2)-xpKW4_ea@>e+-
z`;*`!4iUL*Ss=CPt*sG=!8|)lY`Og&HZd22D9O_>Nrgv6jb`WM6ADPG4g%5ItUWdr
zH|xO)MQh)h0}XNxB-;wSs4!ABh5*N4tL{;?=gZ8@Of;7JM02P2`45<YZV5Ra4I~JI
z2%BI_LAlvvvFrovZn=ahC9@N@h}yu%chSuCQ{lC5GcZh7Ig6SKtN)4AF|YpZ;U#6y
z=-_U06TW2qg$_ion_Jf0(sLSz?BIW0j2}QpWUf6Aj2d+a#S$^+jE$KSS*gz8t@+xT
z+q{~lX~h_L%Y>%7e<?NF`3gge+JE&-FYb|N;f;u7=Jd6F@D6oZ_pbICa#KxRQbQpX
z8Q`1UZ5-OGqVkwuXxE28?QMDQqfl2BB*8+{xV`&2-^uol)vEZYVL*MsnY;Ff(Bj|2
zV^V)S_W9Z86TS;-oKc!jh{<yj0tCIWQzy-{SYseff*8D0M8qa-Ita1>APT@9=!FZM
z^FRV}XYRQ5Rj9%K10;jq0iN0)A#aptMLUr-{`;q@q&8$9dtpKvCY0~wZNKx@@kG`E
zZVVEr%cy^5{KDr1@o;U59M8jla@HUbnb>B5X6Kq9=}E!`LC?+#q>s=UQk6R-S)>AY
zNSqRiIuHJFoBY#i`KNChLxfe1VWK~}--mj|e}qbdHtIQG`hBZG>>F|%j}uY*_F32x
z*X{Gxj7Q%*xc@ir{^8>Ng-@<w_?39)#@!hzGc6Xy+ZF*(@SY~dT&7}rsyAM=o2)Xn
z$++#Fp%0~>v$ZNu(0Ka{WsL45ztubO;WNgYaUleY*UMvUoue#F&K<u`S8HsvA0`z`
zzadiM^zMn%N=lPVJZ@94flimch6EqJ0Wq;vwsMo9z>eXOQx@0wxf?numbFHk)9}U$
zRKuH@mkLFcuVy$UEG*+YR}_|--Ii+|#E7;PMPlpr4TVW<Us?}4H8G@+`1!1!=0X5h
zy{Wn?sW`dQ=a$XsjQN}{QN*)^6?&3l^yJp)4cgr5BL`^(1+0?|;7mFdm6{=B3pu5J
zQ;5-14VsP?+5|-T0R|JNIffqwtZ1cYGB~TRTd~mr$-N7eN)M;zu#V)KoA*~PKI|%y
zg&$5I-~=vJuA}t%+275Kzf~NV8GqH`zhiQ$KGJ%)tjNb4l`XB@&q#i)mzuHA*;3u4
zCNx;BS+KV*hm|~_R2PA#<Jt!OL^;O}-Z9kaYe@<&53tJNDpi%J%TV7o;lhf!OS>=V
zE1!dhLytrW5<x*i{K596yrLQI()FmY*Y6_L*olzDUoUqGT0d!qu30-$bGFsH)Vf5T
z^o~O~G~M3iDPPyx@>wyx`5;YoBPI{5pP#n`J2aeeMW1OIN;fAjp6uw3?Uh#Q_uCoj
zE3x`DU{rH??elq@+*L6>-NlpfGsg`G&N@;eyT1D}a*8d8mD}bT@vBFJM|Sm#&9`ZC
zzjegyPM96KCw|+jVlZ#p=Dw-uU6vGe?P$^YCt7P0>I<zo?Q_?G0A*Ggpmt$wjk8Qs
zkdzt7Z2Ljdi;Rqvyzx6GpUxf!flmeBuZ7wwBvcOWdw4{|bN8Nn7%8`hK7G0aj2(n0
zf+U(2m$NuCUJI5NGNZ!Y4jWVRyIl1BJ~;a}2?(b_aOG;q>Sz!Y^o>)E_Gh_D&4>)5
zWt!h%Hl+pM#YOpPlB?o7YX|B|<k?3vg7x!)U*BD|rzL8vP@mj?$5{|tUrp`Q`OtGm
z(TeX0q3rQPvHFIwm4kKx)@?wZJ?fgUAMH<VX!e!ae|&@En;Tpvf_gr{(7krAx;v?Y
zCM+-*U*;TZB4aQyfFEpB?VDewlrjG1#2j*>{Ze3ZIWHLak*Q=u6a=N)%e+TcAWb)c
z^oUe1?E;I-C1kh*)C(S2Sy}MC;81dOaoGe!(ZA?$gf$yZt2a*$7$UxWZmux3*Z9;a
z9uciM6fS6<9~>I+TD@M2TmDk%kGCpz1*Ar9A;<WGR%}1v<>N4H?>U1I3Cm(Zd%yyN
z053_my=?m4t^cMw%l{TYR*7vdB!3*r*o#+8jhsG}(0me&qoia+1}#@G-|^;?VvMuo
zHaOGwA8{Nm%UfiOo<HCEzOr$Gx6b2;_Y^~_EXz>QGb;LT+6A##Oy8E17T1<ojJvDk
zg&@)LvXn(Z1Y!pv<>nWVv|uohGBfM&nRJ-w%+#815=(6Rl@uNRQs=r=kv;$N01@j<
zmB3=Lh<*+N64IOk8t`wvJvQ&i+9XJRLp~P*G4`PG2+N#^g<As1V!k>UFU@+#z-xN&
z@F4-4yq#V?Q?ch=fAS8LRoKPaRod2h9xt<SYs&f_?%9^Xh!lFD_i4#Ws?ppsrBg2V
z*4?@;`!`cft6$;y`Y2;+CBQCYbDfAZ@AB&#(&EfEe-WI?K&x&!_m1qd`RNVGDW6sk
zrGc0>clWL<R7^sm3Ra*vk4&{wBh68g`m*>~<c~9*GKN>+?=6}Nb4!J|Zu-Oe26c{W
zv@$s5pWh;FvZGmfbH13Aq^Gl<wbza@Wo05>>%xVC$vM}yItjEwir0`Xs34$~f?$_Y
zRPet;GupOaf%nWn;`vWuY_TCr8T&Bb>$(gT>{vdwE}6N>__I{wU3dGJ#hMcSPDUu_
z{5-NR_kCYqZ(wd>wMHhK!=2@BsNV16zO?6W-(D$aV|4Q+CdJn3H6T{IJT#j6`mB#E
z`g=<&zG7xu><daI)8qB-WqpVvTL&}O7wf(gXPR^#2oB%M8cK=Ue6XljBJP*uuT@97
zWfcbFtnCkG7gCRv#wMP}36u!YXS^uuuBs>>13xE^=y=Zj$VCK<WK|fWCwwFhRI1A#
z&WhD62<IIn6?-hJTMJym6N#QzQfg-Qch0p3@}&V;1<hR*(UkFy)aW*6%}g7JajwE5
zi9(V;AW_fwA~GOE>Ki9=b+Q5%zR~v(L{e5xi0JJ60pi{^po{8))(0NXb0`kbVt=8+
zeG&!{c;s0@$fP3sdzG`_=1|dBZQ9ya^KWk0xQW>Yf7sCHmLPml+L~Y>Q<J1X)d;u|
zAZ@~_hP&vHqNy>;So>;I_{tb(ygLBFf-EZ^`Sc|_ws_DCTe07Jj==RPUwo(_XTx}e
z@s}BgSIzYsBx`2}a|@n?Qb9uW{^Bjm2PXCR?&+%jUTAjm4PZs{b_K2|ijha;4EpsK
z(Rr=T@gwSJCFU|7e{=Iw+&Op!miVh*upMO%)XwU{|E(jWJmT4aOBv(q!|nCX{*+rd
z({HCvE0eHJcs&13XZ@==IdDt@5<Ppg0#vBZZNrd>MmF@{`{R(3E(jQ3ym;|DDJdgh
z8T^0q-i#E&y?hx?N8gTwwkg0j5S<ubH7@{rKa!Tq#nqxi;Ve1=VE;c&u_S!WS(suW
z4K?;ap1c=`kL>J2z=oU#l?VzCYAZzG{ymO<fA^GI)l6-+IMvY4iK!{Ix?-u*io+!d
z1bEM3?!XuzNW5IQ`V$@}nP57_ZDiqVIe+fgvE_m8mBi3c6k<CgMW4OpJ}xP{EmL=*
zGKw?<_oXHOAQkK!LaZ+(F6f@Ld6DOAR-(-3cylW%fxMWsX%3Q$R7fsKZ{G|XEY!ZG
z<xlNU)vD||O;4~;`^ZXNCG749D$l7e!R%j}s)D_$L#q+K%zJ2{Ha7F8lX$>#`frN1
zQ+hH(G5i)0&sbC3e*o&~q<he({R3TC3e}*iM_f$F15sPq;x9tvj)d~7`S+F6_XPw7
zf|v0gf!_BR`1z^F#l;~mmfbhn#oqBR{<?q9M1AG*jI2aWMO^CBffjVs+0&-_`MP#L
F{SS!h*>(T`

literal 0
HcmV?d00001

diff --git a/tests/tools/cohort_tracking/test_cohort_tracking.py b/tests/tools/cohort_tracking/test_cohort_tracking.py
index 55d8b380..05493011 100644
--- a/tests/tools/cohort_tracking/test_cohort_tracking.py
+++ b/tests/tools/cohort_tracking/test_cohort_tracking.py
@@ -62,40 +62,86 @@ def test_CohortTracker_call(adata_mini):
     assert ct.tracked_steps == 2
     assert ct._tracked_text == ["Cohort 0\n (n=12)", "Cohort 1\n (n=12)"]
 
+    adata_mini_col_name_gone = adata_mini.copy()
+    adata_mini_col_name_gone.obs.rename(columns={"disease": "new_disease"}, inplace=True)
+    with pytest.raises(ValueError):
+        ct(adata_mini_col_name_gone)
+
+    adata_mini_new_category = adata_mini.copy()
+    adata_mini_new_category.obs["disease"] = adata_mini_new_category.obs["disease"].astype(str)
+    adata_mini_new_category.obs.loc[adata_mini_new_category.obs["disease"] == "A", "disease"] = "new_disease"
+    with pytest.raises(ValueError):
+        ct(adata_mini_new_category)
+
 
-def test_CohortTracker_plot_cohort_change_test_sensitivity(adata_mini, check_same_image):
+def test_CohortTracker_plot_cohort_barplot_test_sensitivity(adata_mini, check_same_image):
     ct = ep.tl.CohortTracker(adata_mini)
 
-    # check that e.g. different color triggers error
+    # e.g. different color should trigger error
     ct(adata_mini, label="First step", operations_done="Some operations")
-    fig1, _ = ct.plot_cohort_change(show=False, color_palette="husl")
+    fig1, _ = ct.plot_cohort_barplot(show=False, color_palette="husl")
 
     with pytest.raises(AssertionError):
         check_same_image(
             fig=fig1,
-            base_path=f"{_TEST_IMAGE_PATH}/cohorttracker_adata_mini_step1",
+            base_path=f"{_TEST_IMAGE_PATH}/cohorttracker_adata_mini_step1_vanilla",
             tol=1e-1,
         )
 
 
-def test_CohortTracker_plot_cohort_change(adata_mini, check_same_image):
+def test_CohortTracker_plot_cohort_barplot_vanilla(adata_mini, check_same_image):
     ct = ep.tl.CohortTracker(adata_mini)
 
     ct(adata_mini, label="First step", operations_done="Some operations")
-    fig1, _ = ct.plot_cohort_change(show=False)
+    fig1, _ = ct.plot_cohort_barplot(show=False)
 
     check_same_image(
         fig=fig1,
-        base_path=f"{_TEST_IMAGE_PATH}/cohorttracker_adata_mini_step1",
+        base_path=f"{_TEST_IMAGE_PATH}/cohorttracker_adata_mini_step1_vanilla",
         tol=1e-1,
     )
 
     ct(adata_mini, label="Second step", operations_done="Some other operations")
-    fig2, _ = ct.plot_cohort_change(show=False)
+    fig2, _ = ct.plot_cohort_barplot(show=False)
 
     check_same_image(
         fig=fig2,
-        base_path=f"{_TEST_IMAGE_PATH}/cohorttracker_adata_mini_step2",
+        base_path=f"{_TEST_IMAGE_PATH}/cohorttracker_adata_mini_step2_vanilla",
+        tol=1e-1,
+    )
+
+
+def test_CohortTracker_plot_cohort_barplot_use_settings(adata_mini, check_same_image):
+    ct = ep.tl.CohortTracker(adata_mini)
+
+    ct(adata_mini, label="First step", operations_done="Some operations")
+    fig, _ = ct.plot_cohort_barplot(
+        show=False,
+        yticks_labels={"weight": "wgt"},
+        legend_labels={"A": "Dis. A", "weight": "(kg)"},
+    )
+
+    check_same_image(
+        fig=fig,
+        base_path=f"{_TEST_IMAGE_PATH}/cohorttracker_adata_mini_step1_use_settings",
+        tol=1e-1,
+    )
+
+
+def test_CohortTracker_plot_cohort_barplot_loosing_category(adata_mini, check_same_image):
+    ct = ep.tl.CohortTracker(adata_mini)
+
+    ct(adata_mini, label="First step", operations_done="Some operations")
+
+    adata_mini = adata_mini[adata_mini.obs.disease == "A", :]
+    ct(adata_mini)
+
+    fig, _ = ct.plot_cohort_barplot(color_palette="colorblind", show=False)
+
+    fig.tight_layout()
+    check_same_image(
+        fig=fig,
+        base_path=f"{_TEST_IMAGE_PATH}//cohorttracker_adata_mini_step2_loose_category",
         tol=1e-1,
     )
 

From e79c0310de417ae5421019302f4047ad44d6da58 Mon Sep 17 00:00:00 2001
From: eroell <eljas.roellin@ikmail.com>
Date: Wed, 13 Mar 2024 11:55:32 +0100
Subject: [PATCH 37/46] remove grid lines, add notebook for testimages
 generation

---
 .../tools/cohort_tracking/_cohort_tracker.py  |   3 +
 ...t_tracker_test_create_expected_plots.ipynb | 164 ++++++++++++++++++
 2 files changed, 167 insertions(+)
 create mode 100755 tests/_scripts/cohort_tracker_test_create_expected_plots.ipynb

diff --git a/ehrapy/tools/cohort_tracking/_cohort_tracker.py b/ehrapy/tools/cohort_tracking/_cohort_tracker.py
index 2d84ed5c..e0a01e78 100644
--- a/ehrapy/tools/cohort_tracking/_cohort_tracker.py
+++ b/ehrapy/tools/cohort_tracking/_cohort_tracker.py
@@ -228,6 +228,9 @@ def plot_cohort_barplot(
 
         # each tracked step is a subplot
         for idx, single_ax in enumerate(axes):
+            # this is needed to avoid the x-axis overlapping the bars, which else would pop up sometimes in notebooks
+            single_ax.grid(False)
+
             if subfigure_title:
                 single_ax.set_title(self._tracked_text[idx])
 
diff --git a/tests/_scripts/cohort_tracker_test_create_expected_plots.ipynb b/tests/_scripts/cohort_tracker_test_create_expected_plots.ipynb
new file mode 100755
index 00000000..bff86a21
--- /dev/null
+++ b/tests/_scripts/cohort_tracker_test_create_expected_plots.ipynb
@@ -0,0 +1,164 @@
+{
+ "cells": [
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "### Create expected plots for CohortTracker Tests"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 2,
+   "metadata": {},
+   "outputs": [
+    {
+     "name": "stderr",
+     "output_type": "stream",
+     "text": [
+      "/Users/eljasroellin/Documents/ehrapy_clean/ehrapy_venv_march_II/lib/python3.11/site-packages/tqdm/auto.py:21: TqdmWarning: IProgress not found. Please update jupyter and ipywidgets. See https://ipywidgets.readthedocs.io/en/stable/user_install.html\n",
+      "  from .autonotebook import tqdm as notebook_tqdm\n"
+     ]
+    }
+   ],
+   "source": [
+    "import matplotlib.pyplot as plt\n",
+    "import ehrapy as ep"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 3,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "%load_ext autoreload\n",
+    "%autoreload 2"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 23,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "_TEST_DATA_PATH = \"/Users/eljasroellin/Documents/ehrapy_clean/ehrapy/tests/tools/ehrapy_data/dataset1.csv\"\n",
+    "_TEST_IMAGE_PATH = \"/Users/eljasroellin/Documents/ehrapy_clean/ehrapy/tests/tools/_images\""
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "adata_mini = ep.io.read_csv(_TEST_DATA_PATH, columns_obs_only=[\"glucose\", \"weight\", \"disease\", \"station\"])\n",
+    "\n",
+    "ct = ep.tl.CohortTracker(adata_mini)\n",
+    "\n",
+    "ct(adata_mini, label=\"First step\", operations_done=\"Some operations\")\n",
+    "fig1, ax1 = ct.plot_cohort_barplot(show=False)\n",
+    "ct(adata_mini, label=\"Second step\", operations_done=\"Some other operations\")\n",
+    "fig2, ax2 = ct.plot_cohort_barplot(show=False)\n",
+    "\n",
+    "fig1.tight_layout()\n",
+    "fig1.savefig(\n",
+    "    f\"{_TEST_IMAGE_PATH}/cohorttracker_adata_mini_step1_vanilla_expected.png\",\n",
+    "    dpi=80,\n",
+    ")\n",
+    "\n",
+    "fig2.tight_layout()\n",
+    "fig2.savefig(\n",
+    "    f\"{_TEST_IMAGE_PATH}/cohorttracker_adata_mini_step2_vanilla_expected.png\",\n",
+    "    dpi=80,\n",
+    ")"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "ct = ep.tl.CohortTracker(adata_mini)\n",
+    "ct(adata_mini, label=\"First step\", operations_done=\"Some operations\")\n",
+    "fig1_use_settings, _ = ct.plot_cohort_barplot(\n",
+    "    show=False,\n",
+    "    yticks_labels={\"weight\": \"wgt\"},\n",
+    "    legend_labels={\"A\": \"Dis. A\", \"weight\": \"(kg)\"},\n",
+    ")\n",
+    "\n",
+    "fig1_use_settings.tight_layout()\n",
+    "fig1_use_settings.savefig(\n",
+    "    f\"{_TEST_IMAGE_PATH}/cohorttracker_adata_mini_step1_use_settings_expected.png\",\n",
+    "    dpi=80,\n",
+    ")"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "adata_mini_loose_category = adata_mini.copy()\n",
+    "ct = ep.tl.CohortTracker(adata_mini_loose_category)\n",
+    "ct(adata_mini_loose_category, label=\"First step\", operations_done=\"Some operations\")\n",
+    "\n",
+    "adata_mini_loose_category = adata_mini_loose_category[adata_mini_loose_category.obs.disease == \"A\", :]\n",
+    "ct(adata_mini_loose_category)\n",
+    "\n",
+    "fig_loose_category, _ = ct.plot_cohort_barplot(color_palette=\"colorblind\", show=False)\n",
+    "\n",
+    "fig_loose_category.tight_layout()\n",
+    "fig_loose_category.savefig(\n",
+    "    f\"{_TEST_IMAGE_PATH}/cohorttracker_adata_mini_step2_loose_category_expected.png\",\n",
+    "    dpi=80,\n",
+    ")"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "ct = ep.tl.CohortTracker(adata_mini)\n",
+    "\n",
+    "ct(adata_mini, label=\"Base Cohort\")\n",
+    "ct(adata_mini, operations_done=\"Some processing\")\n",
+    "\n",
+    "fig, ax = ct.plot_flowchart(\n",
+    "    show=False,\n",
+    ")\n",
+    "\n",
+    "fig.tight_layout()\n",
+    "fig.savefig(\n",
+    "    f\"{_TEST_IMAGE_PATH}/cohorttracker_adata_mini_flowchart_expected.png\",\n",
+    "    dpi=80,\n",
+    ")"
+   ]
+  }
+ ],
+ "metadata": {
+  "kernelspec": {
+   "display_name": "ehrapy_venv_feb",
+   "language": "python",
+   "name": "python3"
+  },
+  "language_info": {
+   "codemirror_mode": {
+    "name": "ipython",
+    "version": 3
+   },
+   "file_extension": ".py",
+   "mimetype": "text/x-python",
+   "name": "python",
+   "nbconvert_exporter": "python",
+   "pygments_lexer": "ipython3",
+   "version": "3.11.8"
+  }
+ },
+ "nbformat": 4,
+ "nbformat_minor": 2
+}

From d24f6777bb02358700b080ffb54397a35694032b Mon Sep 17 00:00:00 2001
From: eroell <eljas.roellin@ikmail.com>
Date: Wed, 13 Mar 2024 12:22:54 +0100
Subject: [PATCH 38/46] prettier docstring demo

---
 .../docstring_previews/cohort_tracking.png    | Bin 34136 -> 36527 bytes
 .../tools/cohort_tracking/_cohort_tracker.py  |  26 ++++++++++++++----
 2 files changed, 20 insertions(+), 6 deletions(-)

diff --git a/docs/_static/docstring_previews/cohort_tracking.png b/docs/_static/docstring_previews/cohort_tracking.png
index 889e867538236954b58418ffe8d31de08c6f0e63..7bb5715231f3a198e06953f326c68e0273ef6005 100644
GIT binary patch
literal 36527
zcmce;2RN4h-#30qQ5mIy5Tztp5h1HVizo_ZMnZ(_O(`Q$l06ziB0I84Mz+XanUTF^
z{a@$r{{5ckIqu_r{?C2i|KT{k$G5_DUFUgzKJWMI{Td(ds|purwli%fkw`R`F3Kp8
zNSmxlB(kQh6!;en?e0hTpO}@bnw7HIT`Sw$7W$+sx2?=g%&bg|bPw3*TUZ*InV#Si
z<vYQ1z|hLd-0~DZ|NZ~*1AJx{2K?niYRY(#ZRQtmT9QaRZWI5JeUOebB9Z)9F3Fs^
zZWl4uWgk)VldXKZCDOU4I^@aiyNuLj3I`7gRmk|qyrjDpq{L;k?0=bRuWY!`^|Qxz
z$yU$@T3=nT=Ft;jPOn>Adk|iApZ-M9zNE$3=`TALUFw#jEDN;DoF>GE9sP<8g_!8@
z<sTk)<)+AK8GI#0Jvv4F6Ze}p3i#`{?MFZC#$Qu6Z677Ro#^&|{I<XPt709Ux!(<?
zmCU<#ot2lD_x>YlH+6u*{h?7~Y~WI-@!J~-Dosh6x$>j*bX@oG!lp`%u~$lTj;Gu%
z@i12EZvSk&Ptrli+uOV4eIECNzHi$+cCab<G6|bMcu@TFXP^fS^UU8r$-zdsPy1U_
z^`5axhJE?+CC&7ySLeI2${aFwrISB?{76a9UD7VXYuV5LRK;r*S5;N9PvUJ;QnL$s
z$Xb%M70dnhU%GN-@RuiBEZ&)9(wg!xtJ}c`A8-0ka-mksNey4Ua=+=(gydbgl6;5x
z${aCv3o50noEAP778b{Bv@TpI|I@nbqP+Y-!OBEQb8}ElxQKG}c^aQNk#0M8F6CGa
zyh!}mn2Cj@WmV2Tc6KJ0y{Sh81O&chXFr(xeJ#go<WYDyvx|!hHJemKL<EaZUT|$Z
zGp@c)a*>7f_U&6nY6@x!a>^a7n`i3c#T^%TN<4RP3JUHL5fPD)koaBymRvp4B>SC>
z>)(+$7D<O!i`|a=+NUXdCnxKw_zO)Qj72|}dZ+T~mguQdyY}q))?4gOwf|jLfwTB0
z^{lee(kI*Z9-FEapB$*NoEf<Ecm9uOVWG5wf<n_L_27}lIQ4<=;fF+BR_%_+hnx;P
zcEjAtYJhmb2mNKFx$%w`r;W;cd~dxv!<v*ib@g#iyq-So%j_`KTV0yw(ad>e+v6O3
z?8e8bZ@U%u*wwBr_S8*u<%Wt`9+@oMkSZuBAT2E~?+tO?SYN9S7kN&}YUfs8e^ntu
zv^egy%jK#d-Zp&DGiT0l+?!opo_%0r6RVzOT3RXXFlVUcJb$A-^M2KCMTufO)TO2V
z=Qj&oB#oOBi@UqKBNjWeEi=q|qE;pgU*j!jyKH*|biZw7KXhmWw@ioWsp-#;dmR@i
z^ebK*eC**NXi&?dtE)>q$oROZv=7oJPX^7+Sy50kx<yAHTI{lA<Kp64opjyMD{>>t
zFl^Z5cDE)h@VAtgm&V>`ug+JxdV0QY57z$959vAoofFtpCZJcQHt<QiupkvHi@u!H
z|KVD6_3Lt9W`oVoC}~+LR8<En1AU2~byZMQ6d4*AP#X|4?{#ay*SiUFN?JxTJrffX
z|CYY-@%xFmYJdBg2nqWizk7Ri@>V8tdE1X0{WyF`XSL6Vzq<TZrisz!XWA~yG0SS_
z&wB_QI8fC3*;sWpb=&SEjAZ>S+OGGdf<i-$OxiQz8)|B5LLwqe9!RpXvX%vKD&ER4
z#s@FWQK9qyBU_TpuJU`joGojq^PMc!!eozYO{J}sm5{Wke?UM<=m`UJOUr17`BCDP
zz72A4a`sP5XeH*!1@m#7I1zwry(y75-KzKe*RNm2U!PF%&g5oiPfwNX2$vQY6|H}N
zX}8bH^lY8<?wvbNCu`+bmyeB)+s}OG#5UB&4WZY!8E^kwuO(0CpZ4yJ!naCMF|k1^
z(Y*Y8G24kw;e<mXBEgG2uCJ%2%vtvDpPrvDt*d*Eh5Yi>tLuF;*bf&kUE)&dKIchG
zb~WF@+NYnEmiF<JC!vkA*Avy~+>3P{j1?~S*S|gY*wb^^@v8_s``35164+&>_Y%A=
z8ewDW>YLoZpX|6F`+hLgAnF&fHBbxgPWP4a+)R6X>+?O$hJZK1_nLGTC%W$3y{k2w
zNpB<l?0o;;;E}O0r_$l4UK;!;X`2sdo2gtwQ4qhCWm?k|uRPM0&WYV1a=#-pzi-Q<
z=;-K6a&rAxo`ojem4V#kv7cx80z*T;KP0ED9^tr_ZK1y~))v5jpex%lPBv1O>-*Yn
zx@^`xP2O8cNl8hh((TpwptJK<HK8Xwqc1$yy4LkvdR_G4!-o<pW9ehv`QqAtzXx#I
zug$lR8Fpmy;zl@46;smR4>@Ih?9!!6otAZy97%<^OcrUEh_-Zt@aOsl2D2!IeM3X%
zvGe%V&tIB&?Nu@VhU1FN=Au0J5;^{SnYt*+)T}JdQ>WTl5;Z<5$0<D9cfzB<dF4mq
z%@djTJIe6atyEOG?{f0<^FzgL`Sy$3yw}QipclW41>=q7^CQDZxhSK$vcmrmQJ0N+
z3M7e&m1{FDn7r3^<Q!e8uuc(4YHdy#+p+b_gF^=o9(2O~OT3xxfyKP%&p|)blitC>
zQqI>-k8*RxaWgt?I_`(La<a#qKNIOis{ispPs?#!SZaA-H$Cqy8B<d}Y`CMxKR8a8
zGKjgZOY}I+(xe;I#h!7e;L-T<f<epf*V8|R<Cz^xcpCdeOj)QH_y!V^3%-_>d3bqk
z#kaS3Q18>N6-U)sU0n?kGTNmmy(T0jB}L76+$~kF{8r*MRgK!(+C;U~M|ib|sG9*1
z5%Tk+KfQf@KdV_~A6-~jkbf-|>3`_L4C+)sWaJeSlSFh<E@6uPpVjpAL=R?uG}Pqn
z<>jSDw#AwqUn-|@HTc?l^(rU+s=Ecx{@;DOg4eVw$68cWbf6<ctG7Dz#Fw0$8SJiL
zYh@W3KDjfI6}%HM!CLQBleJVc3`PTQp+KJVpni*XfS*)~SK`RAoz&{3zpWfA|5m>$
zXlZ4o|M%~UhK5I-R~81*Dwb!Ms694mo8`|3$rfZ=4)5B#SN7`Foqm3P*iZg=^J(lx
z^#N|xB$l<6#o<6zt?bwqes&8<!|B`p>S`JqEc1U_46vn#?ti&_F3Yq_E&VRd4v8s}
z^=Q+trRje8Fd^@zrt9mgvx@gxlJ3`v8`dA-)A(}w&Yi<3;w+-(ZfKz|P<VoP)wgcn
zE;uaANS$Ha;&*}7{@c3(bSvcq72erl6x$T%6`RqP<mbuB{N>NthJE-84jerAP<nmI
zJ;(Y_*z81Ws(YBQahge+;6%=!q(645fI5DUPWh|1Dl1Bs=Go8w8$@~AG^&|v!wanO
zwYa$0$pq_N=KT2|pVUPP*8l!tuj|2z#5enDM0zq>!>V<TAGe#*IdUcJdf&-N(Fc#7
z?&bcGtSwC}J)ew<ii`L}b-~&$H~TDPWiI=8Xv>=3o*k;`Ps*LBWT=1g_;G!_GN;SN
znr)bn(W8KX-M4SwZsGZ0)SMV9dzx}5E33k%Tb}`N%Ju7ZY}-b5KtLe1Xf7i|=tpCt
zTvP1KU{!x#pOQ`}Zf`%@<=LKxvIFO1-zW^wTLW42_4gaqzd7@81Whgm{e!q|s6Py5
zW@gw~OT({StL*_w#GL<HrE1;0c~dsC+)&5@Sj)-TnMz*Ta)_E!G3ri<2MyXhgWJjb
z?E_UJ_uGXuG&PAfjPB<&TO;zMIW#1sB3?PJHS<0*Rsg4%Say*h^GTD7y1MU5Kl^xf
z8vYZI*-Uh9J8fua_;3VUUH9k58~<pc=*+uWSmaGic>2qqQy!3#DyW~`FkYS+%((aS
zWBuH|6L-yVDX1#SIu1Z86%`dE`5niJLWaw!WminUTk{*G4h5V=+~yhT-E9529saLg
z0d7e6Cs%iv6-MH5lJyP^l>rRN7rLy?3{p{1k?X81jMLAq$;g-<CgA5agW3qPqK=N-
zqB$&*F!EC*X=&+?s>!wMYb!P7n>KAyOS^L^bu@SGm-|M0wk4-K<&Icj**R?VB8kG)
z8LE+ykuj{W<iho^+2Ok3Dts;`pStH#e~;2IpByM<&^~<R2!(-xfzMoPtJ=$efVM(c
z>D1KJ_c1Z(s_wqNg9pxcOP(SgMbZtK_dLMd*`<3=Dkdt$TweCGcXYIz=oFhZ;N;{4
z;&Hq&&xVriIFU`MsHjME7&HQ^>Xsa9W6~+B-=~LbBjYu5Pnw#VQtuOdU!S(0g@sY@
z&Q_v3FpHQD_=!)XTr8aP?un9g+(k!6rZeGP9bo(X^6c$hbp9`YL|{u?&$bvO@jS|q
z+e@#Z$+Q{R#2j!=H(&6?iP^xwSFdC|Y4-yu-D!9yOJ5bRC+2eCv8H70*LYeQC(X>w
zkCP1QqId;gXJv`lc3G=F)F0}K{yMbc?bW#zDAB^_<F#lC5=!!yK0#VGDW_7)Q7mY0
zKfk-^s03!jBMAC1fptF1lKXCehG7K|F`@eX`}a=evlG{}6r;Q<d@mBU?c>=e+c#5E
zrUP|w^YvdIZW<gIsE@v|V|}TAhs)}8`B?@jROKTa92`A)h1P$XQ}1`2WObT4Gn<OW
zHAvN|q@)yb{PrfmP$84ndZ*Qa?*Xs)wY%9Ti`^-2W}587Mz3t~^JNz48yPW5bxXzj
zqQg}1CRz^HR!536L^s&U&Xy)wDRiaRNmT)$;v$B|+R~TlnO&BCd6Heca)m=wG=x>s
zAtY;02X+V^xt)W<Et4K$`<XjKHQ^TktN^OnQK|R~mL4rF*%ka9<@T|j2292JqCp2K
z>F5ZTbX+)YJ$gMYEp6sL7e&gQ@4MbbM+?U($e0?oUpqTMpJ_jPmr2O5X!K_SFDv`j
zZQI@gMe(Sm6tTN@PmQ1^C4KBh$xpQytUTK|yRKeYT|KZo{MrEfj77rk1@82tM~?`Q
zz+bqUW>2Ske2Z3pnXeu`afFy9g{rD5%F|abx<fPfCxy`i>G?D&(DDJb-a0MMblML`
z<}LO4piJHx09fHfFT+#JC=wpS_S!>iuLt+rGY%d->XlsZ_hVwX#W|*vCX5LxpLFpi
zqw%}_nZt-W+^OxlChJ4b+nR2$BamB_@6S-@)!GQLb3ld53ms-7sDn2}F37yUh*$i_
zAle-N9}lM1qXV(Q;o()No94E*Esfsn`rizNUcP*J7M<(l$x;5DTvXffGqz)`0<-H;
z>`Iwfh4Q6O_Y(72*5NJI7LAs+wv_UP^1zdoUxeASVh0B9JaTuhk$%pu^mk9w_bdEn
z6+cT#&Yg+8xUX@~R{ykU?;ZcCj;&bu^rxXl9Bt;Ux$6UA$Jl!7o_=<HWqm^CJtLa?
zPc`YWOA-;0#OF|n8S3b-%-P1SBtP1`^S?HL<EQ`I<GUt71DFb2F?Ih-qtj2(no?jj
zO1=Hj?^<&SGM5O|Q`S$tUI_x9WaVUzzqx#)aV6hTr|n18>;8ewG|XoPcbY$VSS>H=
zNTHJC_UhHUx`h4P#Kfi+Gkwa#J1Gvn?H%2I^5lykiVM#}HWfBCz3af(w0&2?4Ux##
zK@=;C#ec0|MBcBo9Ak_)wT0~AP8Mp<9op+mvNEQEEmljyM~-;#Y368DvHR$^XBgSp
z+Xve3Q~d*!f)Tl|ukUh91Zq7U28+vdg^%S{3Y;)k$XvJ(e9Jj2Z}cZWg;nX`z~)73
z;VlMyd@o+z+5WzACH&FHcaK%c9UVn27=JYKvnDfAqq(Kk#y!dvZ}M{Ui{$Hx3i|Zs
znx46(`@kJ@8;?`gZ_iz*a&lY@Sx>x~AQQ~ZZjpY}MlrHj=cu4_(}!#F5u$V`3qO62
z#bevoV+~dr<P~qBqzriR!VTDj#I2I}2Bj!M+;$>ov&x|hes?g!f62|&o=r<`!em`(
zoSa{VjckPCdQ+r4$KLLAx_Ix>1Le~LLn?W*Od<l2-~f15xAAEl-=D=*X8Y~quUP3F
z`!gig`!$p&HKUqsLmj&F@{1ej1O>PA)wQ&|+MzSfv@~q?L1SOL?c+DwcQS3Kq+L}X
zK4{oZ{fPUayL-ItHJNWu_lSRxyLj<<;cH39lN&2N8<-!1mqWl!n3k@^nEQ%}i}#O>
zX&kZEmdnh{+`M`7tIL|x)6=LjZabt`Q?8G4^HEY#64g~|sh50<bxS}o$@cS>XgU}5
z?G20c8=Wb~Vx^Upe?=K8C*QKzWztQ(vOw<=6&cC3m0Z-mQjnLkyvbeV(<xE0M{&U;
zOKs^-D6Mq1*UzxDn*E)vKiHU(fFG6Q{~&i+C~)w%;ph#CcL&wzQu#_*U1+v$Jzgjw
zDJj56O<&&G**QZ~n0fE#)3@h51x;FyV|JWgTr{YB#T6m#8il&mH!-0A{INV3tj#m}
z7m%MI1kf!3ej9Ub%@kfsX&QE{i_3_j)Zsytt7<#TUc9(9Dk^FYwY}nYnS>7m{}Fcf
zO<-W!v&_uss)vS#1n*We9#cv50$G5ft%rv<G*?|MkGbKjj?Mw_Dc6%ULPdI<8Y4`P
zGZI_{@up}_7*3}_O@ub`?>{LCk_@vf1%L>%d+}*`d1?~sVILZHQB{=>igyFh;=*``
zZJTj!Q&x|Q9I$`<)wgF09B%^2YiMYEnJiclwH$gT`xlq_yEFU5`uaL}rUnc|L45h5
zl*ZK|0^r@mwVm-9`vKO;<RxER2@?XRwwIruKTO1|7B{Wh9<ynvw5ybnk<pe+$wP<S
zQT}{_UtKLZ-#fXZFqzMRw$oKiEZjqz?ATgIf1p3vpwku0*|R0vrCYlCb<MOclerq0
zn#P_S++r1{5bChW$6|!xyWB?p0L}RNqR-v#RSCuC@2d9FWpC45<hT;6_}UK}<4^M^
zJ?sRD>2JFwri#dgjGLcb2|xJ&lOOR=_*TUq2LNem+IGo;t7uEPO-|gJZ{EADCcE=x
z<|jI`p4-n3rRo%KqUTX7!aQyN=Tp{GiD6I>&)B4+(0QJM)uUr$tDE{&QZurfQ(*wN
zT_3Q<q$?-z!-u02`HNZ$V*)ilmwVlJgy8CXTvm7qs0#4;`0-=4rs+!V<a%%_nI>&t
z&+U*TCgHYJy$c1IKp?Xqw#Z|%&9-gZMic-yV8Gex6Lp|7DKNNr`1;Zi3{rM>c4>ih
zf@*T?wdnKo<(Mb<F~O9-IC##>$0r&^ArJpVZ?B}RjAnNTBe))B1%ku+{o4>|o~&_d
z#Baa2TH$)1vTwME*<IW+VDn&*i(nDY5@cD9HMi^PG$lC&bqZs*S<?r)wzjs_`4;Uh
znQItWzBM*Bu5PEL<y)|(>iuKjkfO7g<TG3Sv})+35#w;}*Xn(-E!pX@dRsQn-KC{v
z=rPbARMcK%a9Y?_9k%Vl^GB~=$BRr0DobBJjP_SN-5$urZsGoALb0$JTp^~MeaCBl
zIpDTW&Cd^}ma~%gx^AqnvAM1aVEPzD3pe=j{t|%`Q53_ffEf6<($W%`?>ex6+B~s|
zFkj3%rFQ2#rejX)w(IvjTlc-PvM)eU>h0ydjZ;5xs~At*eJ19N=k|<I;1M9~rYXE~
zy3?`+2+AS=IPd4rIVB__#3#OxV-Ep5%>JA%BHvM8=;}%$Sb-p(n_oTITq-WGIp6WG
zYi_1zV`C##2J@RW>~6!;^~uwAn%)9B)}*H(zep?X5z8s7quN(R9`wmzuh(I=-f{Z#
zy`O5iHlJk8?hqSLJu7K>mMGdF9En{+%-6c5Pi16f>6n@20CC8?e0^guTw|+0SQxtr
zAo9B@K44kuf{ct>MoDSu7NB^a?p!%v7O|HGQTAo4Vq#)9A8KA&S_*Nf{`yg4>}LAP
zV&S9u<8~R+PHXzbmbE_`ubhj?%rqaO9&q*-wWP>%mOq#J<a^G03CER3RU%ARO5S&5
z0vj`EU+~+bF%+X%c1m13q9?c|e5r@iV0M#SP26reZmTZG(b1-OJr^@$V>uvxuV>GM
z*k%_ODSL4nNhWP+PX?=k!-s!k*<!uYxb315@W4ipAnjyiG)#&<&d0|$bh^KA_%J7@
zC7=|lhv@?atk@dK`{1{~m6l!u0~h5qbIAo9l#{g(9&XR@BZtwaX$EzyU=U^1)EL1)
z4s{phb*J@Rl#?srnq=rHv>9tHzMs{la>r3wS$Ut7lPJ&(di^hqa9I63qb_5q<sl%D
zF>--^o58xC`W^jzQ*VF&IaJJR*RG-EpTOhtIXEGK0z1^*v0-6l^{u3&q>n?jsJ3>G
zMq7vW7)|<TVdvULE~cDCAzrafu?P5mg^c7YiYc2L>ghS`h>cUF3W#tL*lLw~cxiS@
zh}YL5Ue4Vr*R`X+@7_JR#i-apMG~#{jF}Yw=}#KVx-T4-q@{P;m!!&mR+bq!VvSon
z#=o}Zp^;WwO!<sQ*Ng`(u7(&7#<O~ARaJC!dT;!6%>MQLQ7zZ^#ylD(q28j0<f}7P
z{KNo+skkxozPgXSWy@J&Lf}@ti}tm7%NGCea5+r--+9hVx}#=d(&K*61wFvv`%4a@
z%zeC(&^I<F51xf*6dX-iLxVp^o4(%O(<&<TSga@F%FD`T1_Bh*3>(<*)<xA|D6W3V
zA@^l%p`*sWHcHax^=obK4$jgP$KVm`cuF?G8vB*WQ`W4kvHhHi8WNc<`!`dBI;=WK
z-Z?ciba#8}NwZ^l-D4)A&P<iTA~c_()@%!66Sy-bbw(xSE>RdTG%9iFj2emmHRF%;
z^&K%%`_kHP^H*rBjr`ZI`b4p&kN5AL2zgYO=XtnDNVTq^p{dbsWF8EdH?!&VFA5r_
zDb&^7yLLT_?DB1t*s)^=#uS1;;#JQSV$;SS(7RGAS}^fB%>6!tt2P9ih}lzkN6b3r
z8;O~ULqn^s>&u)Vq4-l2JY8KM3~Hn`5iB;)4FgCo0qOk&0tjAmAeb8TR$f0@8JU6b
z@&)iCx16-_9$3qr;7#1Pwo2Pgy^&=V7Z?9KH*yY?1t?V@f)hQB-|_l!E^B|EVF$-v
z2`AQMh<gqwH2<h5t>4c6f;KaUSA}{qGA=F8#OjP!+*VGp4ed5xm=K6m7TatyF0^0n
zihyMH=pTdadDIljitP=DQjGM9XD9hS_lYz%X8rmpPp0X>G}dEwuP$aL_8f~}%T$7w
zS`0Jiy}vEMzoCNq+aXT%;hM|K%RTcQ7X^)*d89U0Ex`;_CEh2br0Da$?}#})LG?K(
z2*BkK0X+)RWd0&6C}UJ*HfCn#q3Y0cr%$_;I(-qd9IF3t?a;)Re%ia$A;qB3KpWqC
zIDaEaW1o9*(W5O?ztN(wz3dzvm6J5Yq*lgGY$7K|@g07`pmlWmGs<5BN-hMU>czS5
zoUh-bI}<CkHc~v@Zd%u8*@^^(U=_juwyWN8M!Ug4)$L<s<2{9i_DP*!0IJJ(u!t5J
zyml_jU+k8{<VN&j^j!Lf698q~sHoBzw+H_j3*W~f?$Rg96QVuiurPLVi+gc#ti9RD
zNNk*$tESVkrLuBw<MUazx*Fm8A~xB)<ulC{E5ol}W~pPBns*I11zy)OQ&Ui&C75}C
ze}8J`lhi;$CQsA9j9AZMqs*X<PmQJ5?b^5R`#V`50vP=I6{}4@F)@)hTPrSfugR;-
zm)UZv_l(Q>vVo6H;os4O>M&uSQ`VzX0GNW7L-O~U6L0D`?>%<C_ou4%AZ|}lPtWUv
z=R9=u^h&^L6CC$v<CdY3xYu_(zdYb52%_D8N(Wp~D$94GSiIs=4iSC84tOvK8An)p
z;N>vx2tf;ggQMHO|H}32dr&_rR9)=sL~Fu?@t~RLXXgM1TQQ7dB8q`#K@im-HBOlK
zp0<p_@(*KsPsbe)W!lbKU0a)RNY-fEM;)da$*4E;BQoFa?R%!5y5$>hCD=`<E`n{B
zKOKNfba#DqnMX_Gnbyl)f=s>;P)>lX_2In1kFB!)3zViVdO7F~`acl>jKJfU95ys1
z5_RbA`7>Xt$tq8E-dL|1I<diWgk$6vhrggj2~FYUk$s}(2O!RZ-O*_L^i?79R4MKW
z1V>S9AKr#+ASx#y^-|j3nEAiSoay4l2lflrau^8<Rxfa}2U)p{$1_#ycq=u8AjfJt
z4zaUaa9_Qu`qVVge?oom6gIT!VUbSvUBrm4WbrTh;Qy#&|5wK7)<A1=nWVt7-BLx(
z=X;c8l454a*7dtK({UZyVy!FOY}WPE)APbjk#jOK@8v(J<A47hW&J<%iF8z(X?A9-
zFtyN@G?)D?h&!xZ@wKGHJ^GTtg9o<Ue}=pQE9I+-8cTM&QEiKxw#r^I`{Z<WXM)Im
zo`hKMivnC9b|<8+zm=C@q0#SpX3RI`&XVambHJ~wJNn1bo-|WGwRkXFjHz}zN&Kp=
z-u7qq));B}?dMLo>wR&Xnsych6R!$huyg+h<>J5d?c6j|w~L{Mr#64g^0fHXJ4O@F
z+1oprGIL+lw*D38U9!g+6|SbMY*Vr;RV}}?sc!kXRn`=LC$D|&QQ}%05n^o5p~77u
z>{Z%D&k(kLt@DCY-k=^u)4tOQy(-rill<I2k&|c&?_IoWI;8){Jk|2{#kGT-n#-n6
z5l*5mzeHyjYYZxzxds?5liEIedA*_lQDmHRpJYc-G9vjRlD7Nc)yXp-EEwePZQjSl
zFdHRFVo=atb+Q;9tV*^LpNpbe;Cm5u=Tv#|-jB9<L&3VA1T9arS={p}71%+!<E-R4
zGSSGpf5nYdv-h7dJrf}(?OB-Mnp!2IqoWjlNr=`^UPXcag3|M=-}865|Die3Dj5>&
z;yTi<`ONDg7X%{Nr{3PT(U(ZAZEe?5b!A!YddV=kxCaC%@5`DvX0o~~v-Z*ZL$P(&
z^xlr>nD20_4@$8tF3n5me)z@gvWm+bWvrX&ACr)c_gv<i&CPR9ju&o;P!O*vk(P#q
zptM8laMOub@1Gr}9}`*IPjmC+<J}AjQ!-tgjMvSsrdAau{LsW_q%KO?TcT!tXYY<?
zyBGdhz60II7H8%clG`5Glz+}W0C`#7a&#aKKw9za*-a~plf&rm!;Om4`e07<yAl)9
z9ac--Xywm+w6_rCnLVGsjsNxIxXjaYGUoH#IVKzJy{c<HJ17`)$h5pXrKHbfe{6Zb
zoW-tqLGhI82zAkzX8NTn1@hi+Pj;(b@7}+6f^>ja-1p}N74npU`dyX|v)gNuW3$=W
zV-(IrY6M|YCnF=n@YWb3$3PefUUI~0yFdtd3$^U}hpS`^3=E)Q;Z=w&T?Pgw%Ki0E
zxAtDRZ~+43H)gZE+X~W7OUJu%ZOf~ws4zyHvYn_#k)^t?c>X-a_wV27lM{4P9(oAG
zE*k78S`TDsOyfIUDbP5TVC%)Dy%7_u_2ui5hqjY;ocNv4M?20=3e_~uezdY4&zGG~
z?B31b$RKDTxODfzHb(`joxMz?(@N)eoqtA(xFoHou)~TbZ}gOm%;kVbr?~tJH`c?^
zLi&1j!d`%;P=IK2^NaaG*Nrt1jDl}$C%a)FsCB9(G*`@LM6YdyuY%A)*VkjS6`}s1
zU;mnzs6}hN3CsgFgqS<4s~>UZ6fO^iqpF8#Ll0ARIkUi5Z%J|eJ{4E_?YlcFmb1EL
z`M;_M{)&#B(z>FymNfFkcB_olu{?R^4xdR8=WPK@C7iqmM3`B+u2?R<ml?Z1HgYqX
zlSH+^o3gbaiG$qZk5UqZ_xgxa*@;`I=podcx3M|7Z{NPw)>gvcKst8)LlH5u+e34x
z>?`#lxL~6n@3+B@fl=H4QoxahI7K$n$LsN5_uGbe6adv;f#QpnBu3#r!f+2bO-oCQ
zP-b%{3&gb@&-<}_KT#*4w+^$7lm5d<b$h!h^{pSx9CUQgXZa`=Y+~uT7wX!<VmO^R
zl5apa&%vqT<Z2Z#Qupm#Y@Bu%d6Tm3mWfXm`~|fI15B2o1<rqkxn`G^`$=|U4!4Wk
z=nhFKNDa+Bp568&@)*0tIVHtN4HXRP-abD4P>0?rzLp|pb{JMR5wa%c#`DlRAti&S
zNd+(>=6o>Sl(+*KPymY3vOqUrAbwzLTV7K`1Io$C$?2k^qCroA^Oh)xwO*c{o}~%j
zR_}#)Tyq~Sc`JPUbEN+1;Zb7;^S&RH%=fxnr=9x_Hol&Cc>GTO$c_G<?%ZhRD~frK
z;+>9fd6-my#kmlim!I`T>a)D|R+arXOCP=`=a4GQG`l}!8(O!@Hyb~$&g>?9p@r0Y
zXIRGPX>5#f`wdeADNcS(nSr}+qfTY>Kq2_3B79|P12g>x^(;Q9D>2%IF3*@n{JeK_
z3f}&@85`+=wY7e0%I%@I$_56*>{SX^uEapM0V8!tJM49jKI_FVOp*l@$(6B-v=dYH
z5k5xY1&`Ob<rTkWI`6HWprj1;$+wKN`Reyte)9f!aaPz!eq&*=zRjsbJF51=Z8R$(
zK_U+pDVfC^cTSeI{;}0m`EbLWdsgsB<9cWi|Jo;dGrAFG@3`o=<<>mD+2c&Ud0Eqv
zL8bg9-<!W{cm#YJi68Hw-6$*D&203$=f$>9N6bYBj|WQ1ZyceY(Eh-wEjth!{YK&P
zH+TXVw4MAhsC!VdIlT}zZm!R^G+=dU&6p#c2D!GAgf>b7JOW)>ou*&)yQk2#V1|%q
zz#ejG7pw-(f6?R;_w<=0|DzT=u6d8?n5J-ThH0ej(N^)|c+C{cgNdPA+cxaaE)=e>
zR%A4HjVqqcJ{CJ;v$jc2`w3NxfmF!nr}2Bqi<*_y+w64p3g%|`2IB{Ai@o1JX1cwo
zYm7?!?2(p`s$PBfeM)9q$IrIYjj_sqUEb`{zPS7K{k5@nO*b@+Ygair7Ql>5!3iP&
zISMFYc_Mcb%my!{_dJ*~B19j2WoZH%K^Ui^T-WVcSXd0Z^Tgs5BKxqv-r^BU%+;SG
z9y4J5iGO22E5INEA+XtD0*hBm?VHc`+Gfr7dN`})O7gqIWhI7nj~Y+9>N|YD;eB(D
z@}0lYK5Y3W`<N9>4p=zcDd-<gubo}^n`Q5~)}h)i0aJtr3Y-9DE-<NI#V7Mhpqdfl
zDdv1w{*J>cLU3P7N-xDWxRdi9MMmx?xNA|HF-F1(10j%vA4|FS^Vs=kjK%0SikC0L
z@OJ$TxW$+bOZX}*0!}dco`<w_cHkbEjQ)|4OFyV4>_l(%yxjAUZQ#Y!_*ItN{0pDG
zf5Q8x8`B4WOyqnrY06?gSZ(jE(uigb0e{VM9~Y+9Rm+f9@A>br>&rmq`FBQ*{|?;w
z-?%`i&qp_?HT&+&X0m7YR;A+e{Ny|Dtaz#5#S>3Y$_oLteSH%u%{6q~fuTVU<2*S@
z47(KiM=Yj$ZY{q#pkmgpHn40j?Fj>rot>RVOXkyy#s3iKTPpt+F8&FI!K5=SG5H?=
zb%6f;f7b%~p8=}>@|XS-I)D5;dRs4gl1n6gY>vv#!49g}z%H`&^~7aoeF8T>J#!{f
zjV8|2Ot1g9Qgh9Vgcm|F{QZSWqz8xgb?4$o&d5By)W!6llHvXzq0^$IMCJp{3510e
zvotJ8udJ0#x0bYfoJhR?c~j%ZyrmOj+2^}QR6NTLO1-kR{ifrl%(ZVdbN2A=poxQA
zya^REn)^-chbHb>J)?dfk#K0+C8O1pk)r`YlN_8ac9MJ}K@Fmv0Z#`bNTL>}BIg|g
zJ`4=YlpKqC5|!?-a#MkmVcqtwf!+c3*1+vj3Zm^D<_AxBmc;#Nv!0Xq+p)`^`|X?B
z%Kb11Ky3(kLd9S)P(g=bEdJK#V-RYN!*N(#S{joWZdWB15O!v@jf%ef+4;ugX&bur
z_wNnTC$ArqN}|2aH+;Eg=<Va>oFAMv&TsXZsE-TB3S@Z(9y%c&Tw=h#v+qx3<M_vG
z@ex#YF1dp_S^>fd1H*S{fAa<U2UfMclK0W8^~)3$Z{74FDpdVWc2>*O*SeC2of?OR
zPgs~+rg5;h(x`H(pPY~@uwgjBo<+UITEzq=P$9E!7R=vZ`sC0SL1CG6Wct=gtvth~
zH3wM{WPO4v5RQprouaa`oV(k2daPgmvdp9pu~`YPs;DSnq;?Wd;kt6x*^seM`A(>9
z3w7^8iUmnxeg3>lhy6>bnzMHE-M6gnU-jpAkuH)h-*fcUtAuD?>$9C=0WZDN-Ff-^
zGKVsDdxu6`%%C2fcsf_yRvfh3d*A!m;XSuZUt}F}VWlOleirxd_YQr$EUqll<bK(u
zcI?WR(?6b!{jt2puKPLhPgaZ?qmbcF8yg$KP)t}{=;+RX?2Ht*z5DCy6A&KcFG4~t
z;&P{FXTeLl=T7AK@T&7Ij1f*|Fr(8T3g|$4#l>+#`ajIUQ5rD}cLBj?0VdEfGFAY~
zgeOY^BoMA)+?(Ff(Mph}WzhPoM+h%91f%8IVdeC@SKyNbEmP<s^5I!V^8Sk2miF|C
z(v$Mzjb69TunuKB8`Fp~QuVcW51U@`Y8Kiiyv354Hi|c`TIJ^&>0(v&?d7BZ(rOxS
zaYT1oSkLDI>I4H4(~Wu3A(Pf{tG0s;Gf}j5Iql>roWIWc+ppz=sIu|%@86*CD6hRc
zAjd$*wGEGU6xOs=IJT)-B^XszTQ<!?u_X~Bib9mccbE>of<u8>i3~o0Ut?`P55Gab
z3K215rK0E2DSosiQ6u{$eA!hI>*!T*F<geZ;{sS^Rh2}wVi0aTqs%b45XuL9H@ZMu
zgn~o>SlqpkS&c8|522QY*GX?gfDH76Z)(`s_FoV^T-HW`FHT8N)Pm1<hE*ZMdurH|
zGwkMOo3{l?O$h^6WX&4umqsPy7$x+SZ_94lIbg)ZDq(vt!1aQKSfqo%#SguMPQJ<C
z{{H;%`|0ys@&}YkUtV0ByL#-j-HVY)?rA6bxR-)H)qhsao@)<r<fo0?Wa!auKXjB`
zNn`tHViS;^0F&?aWUa2gFCcV*yYYaDc1xJTAgh$afMEz72V5E0GgfR3!tXO{k9kBs
z{G>0;?Tv8hIxJ4q0BMlmH_->9Pw;H;YwZMO1&7l2+5%ChJ#1`>K*E+o)o<akxE`<c
z<X4HOW%+*Foy^R}wZVabrO-oUaQ8mItohJ#z1wN_3W`Qea&K?1vGYoj4_m)q!_<6C
z25aG?%*js!{_(D4y@qCTld*v{%O16z@;@0|WBb4T9FvHW_c50)vx?X3`#bZ<hD(fJ
zRs7=(bB^`vOc!e3Yo9EP^VvTrxu2}nF;pw(>u)nTnWUgcIqKD57zq&!E|oiX?xg5f
z(L)8}Qca2wYqhkr><2eK){#}`qy-%o%-c{MntOzV5z|mdmKj@UXeeO-9kRj*=Hq4o
z*UG@az|~K;gnb-Tk~EC2J$H8}hux5Ch6&jJ5}?Ae_0rU@gaZDUUqvIc^ol7-?UrHb
z!&f9vorwyPG-}H}pUHXngmL6Zd(O!Z7eAO4OD2)ZU&=r9e4apVpQ<6&;&aaK!tTbT
z@4}tOBs9iS1Q<VDt@Zw0<{~4qP*Pev*Q5X9`Jo2iwUqRHMvuYyADR^vzbZzmC?6W_
zi(R(yTVU9rDwtb)^6cwyzqi+(^E)o9#e=ukJ<dod?xoWyDaT0>kyp;h+`D%Smq~Bg
zF$4J{w5Ow^L;d|eY6=TWe(A%3oWjC;)z#Hcns$19`ZQSoGCVwdIX^vpR^KrM8~%Zn
z6^~7q$kTFl4*}_k5HV72^hbWF4GOIbwK3P?-f{BLX0B%U2#wu0Gk#-#Lwk0aZ$s+F
z&8&Bvlh(<UMJIa7QU@>Y;<kPLPBq}NXj{jR90z-%eA4<q9I{lMkh;MXY-nwApL#WG
zM@?Co8x6(mH3LfKV@n|(g^T+OUzklqW+m*WNeBy)YWpJMZ8@AS=Kt!FS0PpANs3K*
zETT~%mtNGBx;Ab&lg`P!7v0lzGoBE)Axy(a`3>f5sz}wAH)qKG*rfdc=wJX=y{-zI
zR7zIXFfl{H6i0%8!Vi;oW17Ak3DmtJC|uYT-(zvBUazZ}_|?HZ>vWyQk0ZuEDD>Q3
zY31XNQisC%n_1pyr7-lfeUR&WBJal$OZR)^*<#_D9_jLztx-B*;=jjq<N{|*hZaeN
z>y{DLl(jz&H+d?unqEYk9E!@f4}U)Ty=B)^=|F{5t}527+&?RQqlU&>aeRtDPcxX9
zP~W*bY%wLX&U#H<oqVF!W@u5z;b@<g#gPJ5+D?_4s=W8xihea3In3OUN%Hd0{?tlr
zI*hyUnd!ppO*ja_*G@yQO}89Y0C#T$PZohx;jSmjCt=pYoImqJj-N;s=>F`-xC~FD
z-9+~A61ZZ*zYH!lRqg3}aJ~dKCt3zUvJ*~PFvy`T^|^cHwF-iTR}}QjjL&hfG58&c
z6|kMX_D*^=GM`hq&06Bg-du&3{*iJ&XC~0U(}rcXzfS7Bkj&*8#^!ePViFx|WM)*W
zFr}!YRKR5^J{CE7y}ln^)UTt3N0S)D)yV_b<ny~J6=VH1AIU~-v<6mmKc>2KXBhXX
z>d7+=8<tBynXK6@N}H4oN{N{raDJb-4G(-U17IhRQBgw6LfWVH%^COOjMS^EvvqlM
zKjes+8B-<9Jw=hW-QI<TuBm^R;M9(fk4HO21fdoVJUQ%b*(j(RTCQu>QO*lD{a7WR
zqN<|!i$|Wa&S?&IITw~aar{KfUa1YUj)ud!Pt|lp>t3(a?V35eep*!cP`IjVdu@XB
z{lbqnmy2DxoKE$mj<Fn~I=M$O!}*k$Pqu5+(W%b+o=UAYM`PU5LBG`M2+_~4X%CbI
zBT(V6-KI6GRrH?cv+$cgUKKc{nyUl3@(JYG$}Nny*|$0UPvCvB&`ZM+Q6K>2|3b3r
zzXGfO72C4W?;`mdbFAgkwYL6ttgNYSuJ!wsuMQyf4|x4QIqHS=-JlZwvm7-g$3U!s
z5EI!&1qHtTAg|7x8JG1L7gpK#b6g(+!nH7Ek-!(}ckZIYs8QV5n4TDZ&_Yg-asKmV
zQwwx2*mbFGZEdM3&cW0-5Oha~2|ovq;Ke_(PhZjfYxL2M`!}oK|Epa6zwXTc6axIi
zo50(_K7^iH<79&98Y8twn(_5*f%=ZK6Fu;$SG{+k^EbW%^C!(cc+c?dP1^i<Q)g4F
zh}8?xJJ*GVR~Fu`-u0D9&I1QX@It{<phyH~R;O@{<7qC?ht5A|)oiZWNymMcVCnte
z`64`lypxX}c6BUjApF3$&CIk1mJ!P!Jlj?yKL$@_&irK-zPCrO?AbvW8xRjsLN)}>
z$Pxe=`IaQjPzZTDq?UAUilkqa-dH^WsYXSIlfA_ku1aPN&}IM-*B%<dgA~)r6o60@
z#`u&}uE^ST*oFLHov$a{?aM>VKBpJ5hokh6bh)FiA9z{)@+d$jaspT@G6VE)eyRG0
zUq53xx?ZJo+<bl29>yJ0)*N4NZ-F>A<jzpb_PDybqWi?(NT7#{;)(1&+>j!ONDw2n
zT$c4g<+=IzP=OgG5vCLXG*(z2AXFK)rEws)^}Qvi$8{qTSg?O!Kn4YWX|7Sx$N5uo
z-pqC|WKM{0Y*~;$4`mTPsM%p$C*ieUMaI#>;vKRGNLb|8&qzaxL}fw1<`A5_ci|RV
zo>>KHNchb`RORL65zZ)ao3WIN0FevkFoH`g^}0jH{0AJ`y}JbN2QnUy23eUTj`=3<
z&R!TcM4kU;lnQ8T_vl8h0GkuaCSi$nBO@n$P>R`3;#7#}#aoo)TJKcsL`Gx@+TZV%
zWHwwYB*p4?i7i@=G*}ZvjuFTc$}#bgtbW%&M3zzTUei%Ty<ig=BvdqbvxJO(Y<5FX
zh_FBP4-c3BZixO~TYDGJ5hgY51tv6$5>PJoa}5_j3>KQK!@`S-UkW(64@ekxcmQtC
zkPyaQyLa1<#>er@xWcRtjqNxDK%@yL^JZxg+>izZFGBeO-6aNzDfm+aV0lAwatR&7
z2t-f#$@}bw4}W`CP2}<rxx!fH(_r`>N;l!Q=a37U!p+cJU_z9_>za(|McA9qzbOiC
zFce}HH80xAsHdk_GyDuD>ZKkwW88ba*HX?ifjxL=dOA99rU6dr%N-ZSjOOO%gtWex
z6KPc6s$hPRO=SQ<vjWi#hSfDS!XBQU%+c@OJww$P>SuuM(Zj=ospGS;fmAb?`{;mx
zfN5ZcAw6J0k@oiXA>&EcH5ly97cF5gGKBKiJ>ItanHv%4Jjqs7T`dG?-1Hmn1$hO9
zlT5x0liKjUJcDbXMzsgAo;y-1$y$<nutxhr3lL!INYnReC@U*lsyOsK<oRIN1_1J)
zF`>HiJAPtkS3QTooX8g>57J@F_G$SEY1C<BZIMv5g_sbo$XMQs%#4!KF5JvC(KztG
zqT=G=(&eS4sj$Hieo1U)J!tjIvmksuKr0X?55O6i3}f+Ru7AA#-N{8pW-}PC;QPki
z$ITu%j1<kCK70B!37WqkTRKn?1op$i!htY~+!~l{h^7RAH&n-aD}$tiA(M3Y@Zp^d
z3>V;z!7dtd0v3J?m_fAG<@PDuDTV}GNXttionj701=rjeZ*BrGWAPB$>Jt9jUQMJ!
zed;!xH`zjjcjId1md7<=07c27V`Yuv$7dPcLLnPGGc`pzz`+qy;badTTLvssfu|@~
znd*_?9HIKn)nwgpr0~o@W0BO>L__1z$`kD=*}0Lg3SkBuPX+H$5fKJAShT*>ze$81
zO#hmHJSI2e84{=0kQm%lg#hBALx=i5w%>v)sYr-_(jn#ZXWpqV4TgDrf*X#>Xv^+s
z7~HD$%0V!4!CQ+Zat>ob@`uhmN<NsnpYoK{)$M@9ZdunI1FO7sn*$K|d2GC8Bp*Ye
zwqp>J9ZJ^77DN%|5D{s4mt_jG&CuM-iYo1^>JY`flatL-gLoPmlh`ftjT9oi0xhkr
zIRpe6`a(W^;>GP91Qr`c$hFqC7V#K9byJ8k@u(HZxd6L=t``Kn^M3kNw>!`NTX}hS
zBi!i4kT-m+Y2Z~nM#xaE8|!yKmCJ3QR^Cj}AtND6H-KJL9VOWj0pn~9?vWS#xp2eq
zs6^Vsei5&l%-R^InDym3A^jAt|K-N7@TiE{KSW*S!!F&6f~J;@l9OgNqSWm;DG5W(
zRvMZ?1a|vBWp(Rf)f1zX#6qiHQ{v44gyhnU8jl3=X;va2=)=<j@|&2V@nEkaclCE+
zTn?ce_+*b_*Or6Gc<u6c+wm>>ux7$5K~D`=0?Qp(ZGGTs)mmVJNv9YYLD>34TJ!8r
z5UY3cXeyjY@JN<{ts+OXmYk^hz_7ProP+Ih3MS1|WNBdTyE;b6$0uysDN1BqKyi$;
zB)_)5v|CZJPIq!=L(1)QB*JU<?%g}Ah8S#T4pSemi*GBJM%3xJWV&A-`O}ije?0$-
zT-aD=cB9>Jq%BiqWTdZK>C6#S@8h(r60M9ZEG8KUtx#;<%-E4(q_>KNUu{1VBm~2R
z!ozWV=JaWl6p<iq)!N_iWP+AROGUb>4CcO@1M1lpF$5|<k20^PrWP^{*CJO+zAS<y
zp{WRrl!13(9v>TfhBaA+-|j~s$IyQg7S=Mb9Zb&7&c0|1WeAP=tXPe#A*!QM;P8*w
zFpnRvW=<2{-ZDBpJ-q>p5##>-_d5%mb7iCQ{)izNQXq(L_@agj|6N|*M?>KrBm%es
zaH>^mqC=p@?1P3_qxky22^0SNZ+G^Afbev8KUd&8u?0cO$Km1XsWz=U>9~yl5$fR2
zpL=QV>G9)bSXd>(&4l*~^##lh7#S7Abu7L50%u3~v8Se|-$H}jLU9WvoRCo;dwBGr
z!vmAU$Tze%xD7w|QRP!*xLJYyjTUZpB_^aHI={lCq?TjFh5KX|-|YBz&Ip2b2|R$1
zn2?5k&LZ{|v%q(v=V6r-E;x{DM9K}mYo9!Px?P9PKYN3%=mb3iWz?I{YoKmwJ_tp;
zdH}LBf<c4r8hOaLDmAO$^19eYB$0gqJOp7z4Z4>Pi(%R0$D7fIgp#xirFv`(f*tz2
zIiJDzREC=!`keP>+PAHY(g>^0recI)93LO|MK%n527N2ju;Hs3q<B3Lf!<X@Ojp$X
z+S=Qhpq@C~XyHNGh>t^5@1$9`1QLmS)D-Te(=T4^gmLH_%7#L~kt^mF7VJDc&%lq(
zgf;y^F7x#1)2TCHh8C?xo4jG>3Vr_hen$|Ja?=?8!k!ad56Ddbwwzf3xU`65JQAn)
zSLw?dM^0mNqY}{~5~=5%tSs|h2wsnD+5LcS$Q?nSx~lUR_j0896EiG^{n&V=@E8F@
zK*(XUjOy&vAe`}uswyhYdp}PDB*FW1RQnb_C$TY>fq}yz@o}KkYZv6ry$lMC!)8d-
z%+35Z4i@kVF^AA`F(dY{`GAm6M$sG|Oe%DlTN#FAr(qo0MoW7HnfvMfa`K&QY}&t<
zH$KDE3s7_8-N_?|4-=--5Psg{$6uf*z4Z5g2zUo?5a7`8AM0xIMQElRd?ysavZBQf
z;nN|=RUp%HfL-Ypv6kMU$&$NeW!XL)fnp~_X+^>=1nm}y%wqWSEt@wZRP|E#r)@SR
z`}UVXK@>W4I)gx!g2P}^gf?zoFTnM1PO?TyIp2bK(3wKKX5X{(dJoI_Bh7-!aig%q
zL#4&V#2QfH{xqLNb0G-|3eLpD#NiADBqsT63i|%>8f#e$S7ipyBi%kg4?CS0$>+eh
z5+W5_SPO-=R)7D_weU3$`v{@lS08-s&%L)l7aM%};>993`#PiP!Tq9+fHR~oR_)Fv
zPDa30cT3L9&+}eeLB<q*0Oa>TD3C<WdyDo4KLFMB=Tux0+`u?#fv8VTMj%L|fntX0
zh!g-E%?3CI1LeNaX%*~>U|euUwFJ_6MJ+81($a;yK5PGLsoeN*^&$AcP43}+Io+0{
zq#T>^N?<rKm>@9gk$BWIpMjE@MF9_=wVvn<MJ6>25lAm+^klJGrHWF^k3m(}!{4<l
zIMe_I9+u#_r|tazd`4dh{{&3bvots9TYgr@ha+c=PB0XwgrhAu{aC~{6#?f>&&_S_
zMWFHyo@j<?S6Ez>!&4-PF*IC&VgS>aaIh*i9eP~Zmk0e7m6gO%6~x&RKND0>=#^3X
z{2J&B+GY+klWE#j2^g3B8xFC{S8gUb@v!R~qp_SX`r<Hu&!-<~>SQ-k^2;;O;D5A&
zU^P~<B@S&k{V*8?L~URVE#4Q#!K{YT($e39gZfUt-+Den`nXQ@u7ZN!1MoRzNFp#J
zfcXrbaY4cdy`@>ZQ;^t#(LO#t0%BszVH#PeKf1@6sO;yi$e3>XcTu^Is5=M&_d&Yk
z{}b!wwdWE6ktPTl3>_AVvM1gl*xJ(cDA9CL7{_`Fwf{7FSI{W_Q{}4fPW}%?u?&UC
z&N(6g46PKLYJ0fy`)5!>k%>gm`fJ(wwJq%Wz+;_rsJM5#_nwxiZ2nc^2}xHMq{cuH
zv_Vq`{GQuLus3LfeOc!RnqMLGZf>5kJbX>&^6N*ZvPTAKvK{Q~FmQ9}g>5>hWFs_v
z=09}45yAJ1@fem4a&spTOhF^KzVhbgOf~zx_*R8As0J~30<j$QyY~2p`vnIxz`PUe
zyc4%W&9}^YR3|}JQ8p<XuDe9FmGT>b{tWej{MxMW@JMoSD?u~$N2s58>LlPKpX%f{
z3W;TZZ!7lw8q>f)o_b2Sht>AR@bJJu2zd?$am{EI?@p$OjkSCm{0=60?SZR_Il{08
z=TwHD8%FH9jF&E3Y;bK%!&K@cK~He)WwO|tEm*qN24NjHTz88Q0-7?F3YWLPj|<jv
zR#14eW>k6^H5W%)_`lqgZg5sXVdSTgF_ol)W_l4H!foHa$)JeZUa90xg7rfH;!(ma
z_Muy^9()EmAwu~GeZ*qpw&W8*w+2w3m;>D%4{lIg{<p9jahd@j*vQAKAMgizz4D!#
z`$6-&=(@ImBLxk`IsfE_=zmz_(V%UdJ2c#oP)h-RQA%FCMTPQ<6e<J-@)X_D?bH+{
z)>Iz|uM9!Q()_OcN<`D}dul*!H-clLaK-+cFj$22U>peFSetYmYyVv3gn11fm8>*F
zCc=6j!)JQ;KM@u~3wUrtAKeh<ND?0FP%wlq3iskZB7Fw?HCT{RCk2E=7?H&QjE47>
z8B9a0eCStbTq7M>{8)#uW(DH>lhSudgc}M76;>h-9AyJD^bm^FlT2(ZA~P5&{rrC;
zvJIxSWW#}jXGa`yf=WVPYyw9Ico!niK~R&JS%8D?DAZL~-@&mln<*#=7=w~vkLkK*
z1P2@etAr5m4HG_EQIGRtq}Up2%)bQ`i4z9?zsq-N-cw*hHYPi6T8XIZ|Cw|bM^5Yp
zjy!=0g67I=sr}m8+JxkUCvFdvW7qqT(y%#kALiCR_(EcV;Gm?DxnY$Oc!luGqsXdd
z7}5bhzrix%_-G4L5>aYl<Vryj3+l^vb+h~%17eukbr9Kc!hs)&w?01w-T)C|s_+_<
z7C0zA%tQugO7NgPfW}M2Lr6r3aH#He9k|;nn577_I&=Vf>%5stZmaS3Kny)oi266A
z8z>?$;-Acm4n)k|ZblR8@D?K9wYis&OcBE(L|7p7+pt?ZFLqtPXqP%!M<gtpKKPfW
zVoLG>p&<a?26)r3D@Pc$HAglIJ>`_mSOuy+F{_Ms5NDo*;e!_mIpye;(>5n-n@Je}
z^bqke;P;m}(j%BplVGw?ga@$>bTG}(SG^(};z(#A*_EuTt$0Rg^%!xxfByfGm<l{b
z!U+J6!SIDN5?u)i?%z0shGbT_b`k@d!Pm#z<f9})aZ&L8WcT}tgK$SoQA~S3aY6@>
z5oPb^7A-wwO9%w*1hR>K8B;4#pJZ<k{HB-mM;ZEDv4U>g3z7=y392y;Zo&XGv&9-u
z;63(`#K!6@2>~x(Fr2sM&SQNA6X$VA7cR6LBSIgH4SWK(9?{8*@UeqH5;E;%bOVQv
z6OArm3lW1H7ne79fY3=x8JWkj;@}IX=H@bf=jZN1cpRgR$A1`NIzBm^4H+?rU)wY)
zdKfwhfr%~)ePi6UOLv-d8vIkX-&c5pLdbnLqHtPEHBxCXH)WTK;RG*{#f;K#-w4tK
zTZtG7mNA3R{%in4<Mf(pG@iQ>((>=02@E)}0}MxqBl)aD)zd?~yfiqX!AB3|VlPH&
z?zDX96NEyKQ)Iw9qO5?|-A`{VjrqDdLX6;IaijwLm1cIhxn!7=4x$YpjEHzYScrF}
zJ8fABmSS=|@SoeX4I-e^Nh0U^zlMsA@bT65ea5ySG6@ME7cEKxwnTz{^3gPKr=S54
zBqirBFd@!~uw*2N(!DFIeNX@Gr~>&=!6(=PRTA=$QE!e3!cUKrmNdNoup)T{OsfwN
zlc<%C^VLM*;p6>ZQbjWup!@>^^&hYy#J!9egD4inu{yx)t4N~I&qqi)(h~?WYf|YS
zQU}*LBIg9q#CiDeX7OC+;g&R5mVjY_2tB|7ggsBjAD{uM4g3L#CFwBt278In@$s@@
zo=R9simdhnbe8L&xdRZLWKfHPwo4KV=COvdy$?)IDbf<vik|Bj(>V&F@FVFs7RE;p
z6Y%m->Y7OFt^anQ$9!cxi#(RK4P52fNx27Iq8ko?GNxxR?h)avxzV3K$U(3|gI<{E
zs=^g5fnQ4nF_oSvx!7q57W64>!&lKlaJ%3JMO)cNB3yRpB1(lW&LDt8P8iUl{n5|<
zB7!)`aeTp^Bq|?jqaGIdmS$Bj|DbPSi+F>T1R}A2eEbUNOM;@?N=*$PX%x;Mdi$+~
zIHw2H9}ty5Pa_UWdI?~00;jkU5F0EL?@%Kd9E!6gJEV8+-|t(j(~EY&l|q>T&S@yy
z0$jWRn>3cqYv+Z1#OX|UMriLouv4gQ&Ow0n2oF!S?$OoDhD$eOE~^G~==AQc|L|bs
z=g2Rz=QNMCc0BZ5@5sF^Di&&Ecl1hTW~_tR0sH*ha1wT+0^^{Ff8?_B@>00^U6U~u
z9Ilg?rH0Fg2MHXWTNz#`3cFcZY29$T6Twtt=Mo^EP(NWiY<t&5Fl6v5yhBwW!Xd;~
zLSl$sO&e^K9eRxYuO|#s$b6$m6E<SV575akVyd;r8F@J1Ny9stAiQCfzsYmK0uT0t
ze&wDc3K16(kt23`&5rB=dvv0yg@srU2H<jc<9=c*%>=UHP#(f2aV^@TR6-K)66e}n
zf(R3@@`(vvYn`dyVrmxA`UnF&!XZkdB#&aXm&KO;%vUi?!kT?Je}j)zC56!--nfJI
z<8|Tuo?OZxUfxsEAue1eO*RLfD=OV0c;eUCKDF$x8S~25h9#F$xKjT9t@7bs5y_@G
zdlpgcSC~&PJTEGvc@cKY&VFH`{>!$_n{kNXh~>RwV;xlLeFwPz=yuy&+euH6pz_sk
z{0TkR@=qs6mX5_qe!=I<!^gR{Axh_yf$BrcB1)XeQ{X+?zVb5%=7OQxNaNHX?VDi|
z_JRm88$b(c$uo&=xQJ{KuD=f_qH!(QV<Yas$u0~qyOu%PE0U01r`GHXbJ?;q$^I;6
zu+Zmt&<&i-@I#hSF8ZHmGF%<#dlloXO&pAb;wJa0KwWEolz}{_&%)x-O3vBwi^`{i
zaU{c6;z))x9LaEEi`E5N^>z0DaUesO%SbbEAj4C>x`u7+f84S}BBaf>_vPiulNof_
z1lcc!HUGphhc7jgHMA!tdWTgV__tVPWk4>o0M|!vd~N8ZsP*U`&=q~)P@yMh$qa*6
zAVOO7O!a{>sl!d0ITy9gpih*7larfks{Z@~$wGhtwb6MkV|&Nm<-dP7#qK58ry6kB
zNpd>Y(Gw3B+l`Y`!F?xm^U3m+l#qvYk(ECWcgTL4+jXt~NUUXD)RQODXSxq9mF_$8
z@g7s!ulh4p{nE9cJIOQVHLrfk;Zn1-di?C!qpGU!9@5QW4&7;K^5={j79Y8iu79Ma
zrH$2+aVwdA5l%JeR9wZpOKyEngjZ+h*w~oeNK9~&eKs<><t;72*ge!{W|O-j!E<*Q
zH~Z;QTDOx={%vHnYq8lO!S(2W9L8{O`{SqFV?Ew^@$u%08S%r1#I~I@J4xe6Usg-I
zzV`L&cjX`>F|n)ZL0m5^EjBCiH%_M;Q0@3#)O7BPwU53Sr()QnoKB6#`N8kC9+n1I
z&S>hq0p*}%o=`oNb1f;owIsQnT+y+&*f*6@>z^Jgey+D3zC}H3F4~%NEMGP(EDA)w
z=P7Y0E2^FM;JSLRgZ6)abb#E~?`=Y6wDM#_F(4e2)+1lr(s+oW?h3W6r812_WYE3l
zxlok*S90oqbu8&>XORkAt`?Gwbz47(y7{0@Ly)Gw^Ja)x$=oU1W~;;x;|}YZuG=fv
zOvn=duhzahoa?^tTNDkYiLy%5Ua8EAhRmoWBT9Cekxi*2Q7U^yX0}M#L{WsS>`fti
zQ=ZqS&g;4F<9@FDxbHun>p0Hyx~?;4{eIu?XT0C9^<G>jxwull^6tyi3X|a@#wLRo
zxHr(zeQ#Z**|u}n3e}WP;rqFSTr`k;T4JpotLV^q=_{-4#R8gk$Q5|PxMdX;Rdp_I
zmXwTPeVxk5;LtEZFTQ@g+v&yNFnJ!s=$WaTwHsrTctG*&=w4atU=|*^Kf*vuB`nXW
zsE^6($@A6!Vbpd~`2Blh{fdy#uFm6XZ%PifE@;mzQ(Ns-yLTBvdka{KGibv0{=9u_
z+M!cKRJ6OR>&p03E;t3(CJ+DQo4McA93(mLW9#}hhaf-Z73Vhkn^#P=$A)km$w(8(
zqp7V;xOYY|pK)cj*5)NU+aSMF@9LtDXSWsnw6GA)jCN3}o!D<D_5MX|9*jUOs%lGe
zMJrfbb6WaZ;PQW#JtAR!F`K5H1Zt%u78VvJyvoX$DO|+vArLKKpIFe9^ppL`_W*5k
zUQDjQMGPE)9eIuDDgB#SZRPy9ub>AY78=wvC<0Z#v)TgjUWGmM|2mQ@TA~Xxo2Zo5
zru~0(BHJ_ed@snmdFW6~?3J$Wv3SpQ896OmpZE6$_+9v6xm+shqOJF8i~RA#toYer
z%6u7!p@&W1-YCwXX+MeGG&ocEY^b;Q%$!lak<qh>K82!YpZ763qk?%|!)>~vv6icc
zn~g;Mf+%Tbg8LWj>;ow?qV}BQ&-!EKo%r@meMazi%cPmUOL^<NPT82ku?i!JdHdaG
z#AzA6Xi9cucB^d2J*Xv}SnIV#_?MhRr&g52Sh6X0?`<>v=fiUzJ*hg*<wH%cTei|t
z(*%b_DkvS6a9T7GP`#z3a9inU{!&`;QclvQ<)w4!hdxg~vt`^Te0P}LFX>&wF!PnN
z4<_{AwtInlW-bF}^${l+X>3s6a`BIPy~&Zf1UU5y3>x686BC24TARLA*qHg}iz_86
z+KVpbRUX``E;cbSpR$dP?%b%3<wPl~6o&)Dm*=L(Xkr$UUl@#`>nTJ8E|SSUcFz&F
z?4cVwP^%z_FQUx79uc|-_>lG(%BRPNfNYHiMH|Pn1-?w@U-k%EJ=;*3D%-r6xfM3<
zuci$WW54Ab3La`|H{KX+)U_D9Bq!VE>bvoL%8A&i(GE5^VPX_RKcP3oGNaT%v;=$z
z$<?vbMH%<wrML6GwCXBe1*~T~l(-fa7V{1Fd3f}k*i7~H(q|J|H4~la;4U!P1=bYQ
z)+F-c%;A#eM@B(mMt2eUn2#TW&JVcAZit*%WIB*XGq<pS!fI7O{<o^&!&RCi0M|(q
z0JIh=iLdZWPv-W2=9$_MbA@yh6Nn`sSu==7BlX!trVAiTLaUL8NbtOXX0Lco<$6&O
zS{+CQY)T|d6U}0n!-xoMK5=Sk?C6N-2UsK>10W;r>Fc}q_%R)1#BTz0LobfNf_5Cn
zXaJjsvmDUO^*aI|L-nx{;0WAJaR`CK)2EQ1EOS8G0%{|~C+`vzd<r^CJ>w-lf%tsy
zvwjW@)SM6=XO9(}<%)SRLhqMF)1KMB2>5*hpQr!fY~klcIs#dbfn`tu=SqU{P&rn>
z2Xni|@ETHB`nB%k?Q+v=f#4U#o+br7sq3@@(Ew#!Z+eP=QbCvBjwCw;fU{ZJ<`c^6
z@&@c6Z$KBmiG-OTs*q&A&}-OI%7=>jy0bYmLL36FPiKQH-!#r0z^MX|hn6~w=6yu@
zg$#`?pt*^VgxEarwV&)>PTn5q>w6N7Bmg1OwE*i6xOFxArtn<=0BI(Tyh#7Jx}lEO
z7Zf{$F(gX>sbj)W4h(?)6s(#?6fdDI2#`i#q!m=u)kvQ~*J_~r1t<~-jy}Kx-Qb<B
zz`2A#*r>Py)r6bxB@SGv{FmLhx8VuH))k~}iHh0-4CD-uD+H18nRkppL(z`7G^x)^
z==n%F0#fpDiuYNt999c`^!8P?NR&bD!OBAP>b8J0F6BGEdv}qfP?DC4DETPF@pl4i
z<RUGCUTl7DZVqXPY%r5clU^0I&Rg>?N~FoLn_En}5<LxCRY8Xt5<oqVmZY)|7N%~*
zr$Q?J7CFKz%<-M@PWWzqv;Ofl8b(3jy*2@d8CHL88KkRioshd_vjeyr<OuzE3z=$m
zr0gc54#&~`kZS@UF+}|v;X>0OBSCUY?~!N)LNS2icQwMj`k3`zvicM9SWm`4_MnVr
z$LiRucIl<>EZhupRV~kceZ4nRFxR~f+%2MGR>>dNfwZ9D*G%w95y0l@wsBP*9UWI+
zy5?3=@wOe%LEsNK{ouoENxv*1(S~0^E+qQ^LE&M2EtR_E`Ki3T3(&=G#AKr(0|lv;
zYir>%yW(M1m#Z)=c@G}Q?850GYh&|fzGyYS*D0WTf^Zc|Re{ClNpD?FZkzw}0UusA
zJi33sJ8~YamgX!xt+lNzElF~XA$S6xzO5H*Aksy`<3T>8Wympj7c?*OgP}$%Q3lZo
z0#!<oR*o-@py*ava7MuIcS8LEM(-)k;70L~fLKU8=JyTz6Ec)^j6C$Q1?Mg3?F&NC
z>I6Q2z7_o7^|TK|c##S~*UrexXQ-Xd-1DF8WEXDKtHmm^fFDu$-adlXEdp(Of7d7o
zJDbsVt?w#2S)%KSA!#d-(i7b*;4bAHn;V`N@!t5vR%$#4rc7effM1avL43JHynxf3
zf|qv+82U#v=eAg7a;=jG!8Su}$~xGPWKuJeyI*zPpdMGUJFER_pogTp1h9q}YZ7zT
z302C$qw>nk!0LJtqa0<@n2a4$l3TCeUOWp$JO0mIz(VPO)5xs^t03>W!6c;C0=i!B
zZyCeFD*G(-WqkWp5?PG$g2&t=8s~jFQk4*)nG(01$f$s;f$9sxqJ<j8XuQtAg2A4z
z0>76U9I(KrD*g5U0)MwI9FK?k44aV%h+CnMK>uD1EfPsnZMB*O!%~Z&ED}tE6a_o5
zo|P@S`PgI!4Xa&&$pF^S+i*8pN8M4))E<B9UC1`|q;s2b-o|min^!VX1dCtACil*^
zYaW`r|CiEE`-KaMr+U_~>6pOz#NMbrPqCbCdgsAb=vUHq3}u<-z4~cPPzHUXRg4;r
z+Aq~V|23hiXj@3GY>K3h${&N)r&IRqiB1fbujNt{nZyiduE(#mL0tTvg9o3JFnVZa
zhAxLc*tZiDOH-M+cJ>OEfWe$QVMSn}s30N7+YGxJPEuHbo8oowS~#&^lp-&oOplzz
zfb-J9)8YZECXzgnHLF(_eo=J0sv`P#LI0a8)Cw7?1GXT;#^7&hY6`@$2haogRS88p
zJ9&9GxFY8s*vx5A8lVXX4U6}sUbk(_7HO<O#-q;S5WYX>y66p6<}+X43+nz{hXtIR
z5QE533_w^8fB<|7qQuSwIzSi7VF2ux?fSO8G3)q(&2<N&Ujj(;@FW#l;@ksi#ZIgb
zn(r(@Nc-Vb2b(Md(z$XT_@E*J+6Wwqa|OyzH{!oRez|YM-p>lM<T1!>B83JnJlbOr
z3!agcy$Zx1`LqPaCem{NH5AC)<j7!laBv{0^pgY7bmC5;%l|c_;rZeV*kMqhSCF!V
z5B_!_%s~y35YS`Xf4AjTY(sMFVi&GJHIMUBf-HCnCGqi<1m(a#CnzeR(#t0HDTT@7
zI}_tjW#u1d6h4KgIL^u9hK7bE5fJ3iMPBaV+G6r&aYSu|Y<e%woEI;4qG?Qh{kjKG
zT|WWg0IPzkF#le?E6jPp2=Dh4pngV4N88(a*Z`g&`)aW}h|LDT_;;`@KnUsQk=M>6
z$wX`jRw*sQ52ycxQ>(b5YQbKSYj>K}!cSMEV9#&Fv;oKvS6~smgzU``?5BJH|7~=t
ziG>r$M3KgqBR(k-o&ya73?3;5`OYMU{7J$7&xchP(pP;N5Y9#31A%G3rS8HP1|I1K
zrU9!$S;r2?34wy3CJVsBCn>5x=#_wN7-@R9QX+Pion7yJQGLdHX`}Qsp?VEk;!(z#
zGEXVWzseec#Yl$oMm?v3qZBAA5#I+2o2cIITaUs5Pb(Z;nR&~NEDPX_xS(%d@ZW98
z;Ra!dpY`q&$*2CJJLmtnF|D2djT`g%UpM9-Gv?KKv&4TgHzt;H!{4J!fGCaoD-I%Z
zbV9v})cZ?eae3M&c@xUQBjf<Hfvt2G95+7*TV%3lA)V#7-{o#cY#biTzI6NyU+O*h
zQm>YDAJ5hNwYPuOg$$V?l?s-v$q%0S-{JN#DG|*9b%}FC;6F6aL>2Tu&iJq<$4<qb
zsD0+VJ(}_L13$kjywK2zuivi{L^i)npp1;di`rbpYyH>!)I&GC*}(lhz0+^qFnn3Z
zQ33!-W=2)3*T(!#@elE#*RA>Coyz*}6oOWQDRBm4Kw@ZFM4v<L$;`q+6d>SOP^oFT
zt6p}%8btPcM4{95Qv0|++~8`QY9vPpES>;vS-gLQ!(9bL1zTAFs$xIuEqtGgif+5Q
zx~7}n*EIM;4{#L!brO&rhRZ_*z91~NQbYi}#|^ut5alEh4j@GyQ4l222F9aGy?=~&
zm!6Bkye=gr#qI3`^>N*+3+oX7K)MRSYY#+zT3RC=?g1KJWK5zoj7E7%RF$qqbM-p>
zu=uSmM5#?W1UQZhF=_+aF8wbdMm2u;!Eq8UgdgrfZW*MUD3kI(<#04UvmfkjsJprP
zycu0zaZwYKic(=w$CbVo^N|-EN)?o=jeM3{=Z<!I&Ybigy|6|-Pe-&PNqi6M{Yvek
zX1YQ{1dHf;-O+SYYrSC+@q?|@spO-E<cHEP_jXOYo8Rio8Mm0|n!DFi(Q!DOLnHQy
zZUu*`mawBS|4S6m*I6ju-WthcmXS19NR~D=1@)qb9LwDC4@_up`sJ#E!^u?&6%{bU
z1E>-VNX?Fui7-Q@$wTB^Arli&Nk;-*g>8{2_fXdkGK$;s<M06Ww-zd39May9lHIn3
zG^Q51Jvg&(g9iXV9|ew)c+)@u1WbaF3?(jmLeSx=a)S~H5p#-&fqLkqrDBg5-lQLI
zR&;Ksjh#JaBWmk@ERQDQlNNvcrfX4}6%I2Lnn#?*C(A$6X1;yy_R*Q@;?Z*=vA3k>
zgQ`tw-_jID`ZV@EO;pe~-KU{=Bkf2(Fqb%k0~E2fin@h`QHP!xGFJyAhH!gFu|_;~
z*h}5WB%kWQ&-d&1SG^&N4C~&X^k%_SqCRpz+j5WsErWCYVZ~vDjZQ8?!bR*StZ*Z4
zUS*ADL6iX7N)pr0=hYy2@zgr)=qQf5WG`w{*VNQhsi9YABPgL>JM6;xpUUjjB%1F%
z`W%YYbaW&*5$YQG<8A9@->0ZV;ZcAEpmtk<3r9*?n$#+1l$4+k(F*m_F1{-h(7JM8
z!@Z;Xg1vKSjv2~*CMhpfEteg67*%gJWsV4~x~cPN`OG@Cu}43wo~`Iq?h3m6FjF+t
zCdl2z`CIq2!5yPGE{@$Es}>o|FGlAWGV3fvAHH77KN5P2Z#2<`#?>LARqn1*iQbT1
z^0?JqyL)MKI$e2pEQbW?UmG}~i6BB|VoF8xi%S&}KUxsL2O8$#(CURy3%}g|aL5+4
z8ktT(gyy(HiEZ<qAH2Q)HWPzqqCWoI$8IZJ)%Z=g8-&mNICplxe?5eQXgt3m5P-~+
zf%YMOKDb4}WZG6n)N<TviH)IS;F?;$m;IK@l53Q1-2U#tK=r1Wk(Qx@$%<z=Bh9Uv
z^8U@fba57mADYoZ6t8bp?n7dX7-#E>W9l|L5*{Z~>y$_A?Y9cLXn6TrAP?7@sMFJ|
z72gEjJLZ=~gEWO@<fJ661=7DD+xt_(gLrB}px3kg!TAUD^@GAjGSkw^8FuRH=>>Km
z#ekv7ySt9q2QXMg3D2epq2O3N-KZ|OQ*%Lb2q3wTL=`~f1(IktC+8<kRaU%a&nl=3
z(8#d2aO$1@Z{ur1N&~OM3HTYbi8OS#5U)j{)+IZ*e+o;iHqYS}cu1nhM2`SICAE46
zcFQO5E{$h%QZpGc>e{=^Pz-C^3GfV2MllPmFR*tHn3&dHJHva=Kg*!Jw5#Q#Ur|p&
zMC~F&ZJS$ia$Z(kSiM8{)lsQaiqsv$bJVUB7i()B&ul-#DDbOT{F(CyN5vN=T_!%h
zR-7LbDlhBJHnrtub>$6xvwRnb-gV=)ZKV+FVJe4X{uGo2gr|Y#f`|f$YYX3j40xiS
zhou6UU~f<AfN?Q!9>v0vk&%gzi`p8y!4(5>u-hz<xK0xW(-&kRDmr%m^g*OqCX<YC
z!;oeuM1%b>u6q@BqK`&9Qd(4WaQT2)`cK5f&f}Kf&aXoO!H7P1pnfZ&et~?U6uA<`
z;I;`(3sQ+jzT-UH2x#aH&~F3ip#B>J96rjV8wL?G1j$O1Tt|+~ec~%?UuFNr_0^4C
zeMc&WSJ(~d^Vn}{e?`kCd~epcc<`3O$2{qsk>|Sf!o*f<x6#n32HJmbrP?(T?3ri%
z?$W(*u2tz-X|V=gVHDiovI^PF7x}2}OV#oUzxr`N^QYGKH2al-@1k>5gIYC7Q3%HX
zwEJE75;(Udz}mXO6XXiTjtam9B*)JIJ7~f_LH6&qaK{OcliP``40CCSAQ!dQV=_De
zagb`gWb_t!0c?(AE8xz!f>yUb>`iwa1WEk#e3vUgseH>Y+JcgKKYL^+$LfFE$t8Db
zJxfQ#1Fs31PI#KTAe6y1OHKgv8&tfCZ;~KZ2YpWNe>~kuU`v!wo<1!G_d-C7@3paX
z7haw3Liq6ukdhP-y$7SCC7Fh}_t8l2gjf`%FDeaX9TMvEe~rCdIs91@)w2e1?KM+{
zB@D{u#*L*@JsNz}wiGS>nvSD(fyXAw>;qVDtX=!@g0RHA@!agkpHpF6CZjyV$gAoe
z8kS2I@c+>g8~m8<s0%X%-)TN8{(+ueT}7oM;b#p4>m6TRG1IMj<SjHeR8(La{e!DN
z<5H;p_)WLilxe*(7h5rV{{<>e6hq{HeGkFXgX81`jyJoRCId2=0Ys#3Q@WlbCKQmh
z34--87JD#oaw1S9M24pzfRs7lhXLT>;M>&0f~2S8qyPlvj>L(vn$v8`5a1K7aF|5Q
zGT*~<r}6$lMI)B_4xZ*Yyjv-@I$Ko4#8waM2<cZ`T}fDqO5`H@VOEJ>oc{Hg7$FHM
z;+OEFRaiHV;kCuZv`T*2;}MIPA5<cKs7js&2A%<r!^<1fRSsb2iI0!+_udYa_o<j1
z)7sX?;~g6t``WA{pRx+aO-W77_#p466l(D)9@`7|rA>4iq@UQ7b517bpN*xt9<m{D
zIhpFh&~MKl9`5ok*)&q&^D(nXwFzA(Kg!j%!M!oiejp^bbEHw>V^#QlDu3Eh*|e$n
zdK%XYr&TR>{^(ZOZQ`9Ap^<5Nq)DVE^6EO|6`joa^ppGRijh}08Q$-prFyZP@=L_7
z=wMAt!v}hX;3erFyfeJtTAByMcpFt;XzW~<&Ha{R#(6z0?KRx0m_HL=I7Ni2vEgr)
z*a_I@Sq0$Z2>yCong)ckrh5|Z^X(8(Y^zuPh6ZbT>t)fYp`M=im>ILSl5Xo<>4N2e
zOcMke{(XakvN*C<)D{PjmxddBYN)YFyvc@uvnK`I*E8v}isrTUc0gQ&83<*7S0|$9
z=e7E8-%Y6LF*%2+ZJitFw7#y(`XTqNfHLxfRLjM#VP|vmci+Tr8SfvSE1zm_V0BCg
zY!|cB<bGMZ81v4m{B)t%R%XpzhnYY;T{3DM%}zXvK)ZfD`IuiqmFz#W-WQ~emz>(M
zaQK>3!Ww2p5qUkj+K!>(U+a*{Mnf?&Gtw_OOdGR2)9hmQu0dyjGt}<ahI3;{)`PN2
zFEl!Nvgi`ZrX#y9av;r^>9XVaN#f*#M~qjj>5#ZMGYl$H`<D*^B7r#IY`&u%bo|ES
z*~t!Do3#Xcx2Vx@9?GAaSS>gmm$buR#~tfwp(Kf0#tezhbTOxn+nnmsqZ(dZFZfi!
zlW9QzeY;*v*Eg2Qln&ub@d4@r6xSs(eZBS{QjhA}VYNN?I={Bra?tDT25W5S_H{W@
zvVC-__*_t+Nu9K96NlJf{IkstkgO3eG}$$zAq2dPu9+SaZ;00gQ{|4MsaOG7{ne5O
zn@Hsfk##j}6#z6#At{k2R%)OYj4&3HtPUW&WXch0*>T*@V<^_o0bU%k+O%a*Q?kk|
zCHV5mGmqSBR;>*8)E3O^eO1c+kVEj7V}o7$<>f<WDK0tH;zvgoeb{Dysa+E(E#J*<
zxf?~xUVQ@tlKD+6D1?DQA-$iIb1g15a3BPOlaZAb9B@HK8)?=_gf=uoyZQO6#Z!=}
zd=Ica$%hyzSaya~Hx9KUgpEb--@i=V2Lnf$%YnFO?68a{752cNN=g1QpZ9ou=+Gfz
zqt2aDDU!Gdyq{Pl3lE|N<Y2A*Z^mf+!L#&L(x08F;>gQn3cVg`kyR_Lb~W^$&aYfi
zf5xY(d{=DUU`DAQ>$XMh+Co!kEWsZ=Dk&*3QSGl6h2Rn0?Dh8CK%gqV)>cu^iDKl$
zrxW&y>haV!DMWgo#~JYN+ZR(Pp?on>?Ka>c0RbPh+G^a{&btv0{*NXV^Q@w_{t<}w
zazgc@Z6jCmRUdc?3=t3`%Ldw$ks!&%C@jY&CInsD)AiOtX!)V7jRi^vveTec<evBf
zX(FH~P$OhWAwuQ)Cue`QoQAOr0|Jn)<^>V?YaEcCqS8`QjPR6@`8`Bh3hxMt_=*r`
z6nZGUK_~U=5)A|}TNHs*@G(QeS_Bnfa+QLQU<64`>fK?Fl|&dEK>S4Oo;y~&?i-Mx
zT0YIUg<x)q5SU&Lf<i`KJ_<K>|I~jcIP|n9pvn$m;}U3i2%AN4A#_2UsS1FK`T(rr
z(DgxDy%+4RgmZ!v(aFj5&hs^Y;iW5oFW^hepUXKMd3DfzkT_rru%p;1F5b#6!NuhX
zbjkwQ#tMW3-NePCmSxI<S|S2UG2Bm$A<IMsIMVvABK#=Htqph#ubK~rr>)?}LTq<g
zb|N4eilNC&qlw;fWCfD+VJp{5xP+dzQomdCPxp>D@<*sP4JLyjXc?^{vAAWx!GdfT
zu8xhdxPR~!4!I;_J`i97)G$>1({=0rtA1|Y{-4-#e#X_SSCg>vRc|<WdA%XcRhwdb
zIQ>U|t+Mh@C>_osv{zCQ^LSL1*y8V@#d?pM?Gz4zK(Sk<X+uo^PUlu$QISzL+d)x@
zHXg02PT=8YH)cyGg2SBZ*}mahv)aJyR*f8PW69s^6=ybtI8Wb%jXm)RzJb|kr@Kin
zcWeEA-Hybq7wJZI3g~-{i3E~dGX!d9V)_gj6wNUtKLB`ofu~PF)Kw^d@$e}K1P{@3
z{}QM(l;~rxngLa@z-+Cks_JJylo;>_8(waI?C!4mkijSYnV;VWxD-w!{+d|%AWKNp
zE(inE5j2VQCUo=rS)!!{_D6<5q0Yz=DO*QJHvocdr>N)-*o08usu<E~TC9gw6p-o@
z!WUWg${_w_9=s4(ku$fg#(Mj&g_3kF%lB64)*t9wRFZ1xHA1eZbexR>kqf(Gm(OX|
z2z*f@F$AnQhMB2Blh9X^H9}mj=-tuvGeOCBD7FA44-D6}m>CE$5L}RvSWOq}lOenx
z2gl?S6huaU8Uf}*M7MHEsIS0(jc7yC1NQtU8RetyFlmu=3!r}3*xX-DGxM~~N_a|p
zKU}P<s*e5(JAG*T@>1T7dy@$+vi|_SqMqZNGW>oandx771OHei{$QH_X5ik?AIz&<
z)aKE!)qrf0q?no6aHfBf8VC9JRraNOv*>CsZeSD=l1oyAF!2KYFBRk4p*ByRe2pT1
zF2|OpfiJ$s$7pmaDsm&kiqMfwV!vW4>s|ab>?ZvfJ=zV(v|k)rXJDmWJKm^9UHc2O
zUm@vO&=TXC<wvMA5g9_)S4`-)<_r#;AUxBD^kCb!54G&UErZW<7wCCUXJ;WW2=X}N
z&&O|O5xuE{{+x`7MTD{kU?_`!;NL%ZLQ&+SQcnhGYH#CB5y|teT}AA_un6$mQ9{{j
zTo+daoBuUMvA^Sd!JLx-7Cd*T6i>x#=Co?U^pJR}UVS<>t!DOl@CBQNRpJsEX^UP5
z3OmAu3sSg;^q+fR4WvV8kNFPkLqbD&zGA)_xpo6Ecn1AO1hxo?tNzI9P{H1T05ie|
zkgf%H7-{FAr)^~ps0<KrBd!B7K#<5hF#(I!VORju>A=-JeiuSGc2I2o_%(zH-2BCN
z*;I7xV2SY&YX>H#)BXe>jELfJ+P++$wEaly_xrP^ml_lf2;t9%3ms$8lyxBQT#2h;
z_x}AfU`bHvg7`pW0HTuc+@)X=q#zp#GERZm$AEvxABG~DjVwe$LtwKTZ>j?~SP^nM
zYUdx^mG`~j!OP?C$lL_i4LkoL^^_z@0Tugc)E?aSbLp>N!{C&0AQlrBF5>JI{u5Tg
zaROB(um;lAfwJY(UI1DTapH%ns+4NMe<AW#PXdC1))qo|^A@u#xhG&YRClzow$6(*
zd7|M?_@4(`BSC=>izKmSAv3@;1JZlabikU6%eN5VDPHDg-_g-gB`6C}nh<23KBc<4
z+Qg2-a@}9B|1M4}C^e6RX9t@`E>a@eg!)!FLTVLxtA!bgZ7@_5V=&x!yf*VhuxXY*
zzmK$(5Dn@>qy`px0#%t-iri+dXmVqIx*H1XEsZ-)md;Ms(5zb?U!YSo0NH_Hf*K$f
zBrXO9VIt87xJCoj1qIPdpz8s~;mhu>yPSag#1UW^es7S0XRzo458Vi|AB-xY{g4p&
zi^Gx%aV1bJKI^>FG5Dh~SnWSjTGsOaO=-DwE$kme{`c#H3ikG*aKgc_c(6l-xq0zI
z@5IA}?b**+pKsjcBdE)^HEO3^?~WScwfUc&wbivL?9tcMTFmcfCaLPE`!cwEfP~Iq
zG+D7{BO9l!SWmak=d_Z8^w$+SYke1wgkQ?_9Yj#z{Cqt9d0vBNmG9B-S^~x|E(zy&
zYbf`aD{=Y>@$5G5y(v00yHB~i%H#e8@pTdg(=)F_6u)q$w_1+OoKNX|m=!F6)TOVl
zEzcYYE_x%)(U|dz-O>c_u~7l<E)G5TmVNREhGH;lElPi*%gmip#4bF@N}zh&zDf3!
z>V`$d=<d<369M5&N^jQePrtL@DK?2ubzWDr@es{T%W2MekwB_!IlXIrmWTG%b_cwE
z$@67H%rDVeAuCVrtUHf)dlgHh2{4wGecHR1hsi;M3HeXoY`!<cRA|bA0JK*4J&AdS
z#1)_RW=2rVGc245U={GOvuSE(sF@jQ<PYt6KXASBmiTTi|HJenoxuegEF^_!PKz<=
zOWKI7=X9veK4m8!@>53fGjIB^13T4^=2Z4|Z<t@6(@OefovF!kBYsQwm%!)pAuQJ|
zEOOr#EqIj3EACOJw!U?8p>tU0vquT7O6|JQOPQIs?IT#fVWUafg#8?k?@gjKGthm@
z>Tmg8G$`Q1yU9TRDqAnx9MQ8ypWoFD=CFbpV8kpttWmC^n6mX44!Era7z=1uL_&Ll
zs=_k#fKwIgy$1nBFX6&$1>vCARw0E^g*yC<scC7+@!>yBXuifPZY(#tbX_c{#6ivm
z%JFN8(fa&))6Q1;Na%aRd?xC+o|&bkS8wmuym`47v0HRUA72~tIq_jRTkd$<1wjh0
zSN5)(4h<`Gdk0CL?B^T(;ToGDLE*5OS94#=?hP|BE{)zZhce7MJ-vzw1D|uhVUgV>
z_VQYAaYrH&pe!2awq3i%(3W7wBhuV7)-tAAx?vboLIWh7?DKobp^RgoG24#)VrjOs
z#>0EKuCJhrK~+trZ{kKJTNE+4m<%Wb!z=}KNJOKBA(|kz-VP<_z&+{tE(G>7(9sfv
zox_uZi-T0Y2>uV5wZbV#r2RlRp<^ebnJOE@9zCKZc_~D=1dR$ADh|;~{}jA&UC^(<
z-4lQ5^*&@omV6utU+0Gd$<i{tbV%PP^x5b5=d|7-axZh;x7=V~n(y+^uUk7&=ARXk
zK*eHALpfIKr+e~CybxpOFV^YQ*8*d8iAU0{G8a4QUn$pYW4lGuy&=kcSI@-p)3#6I
z8^!*F_&prKwAAcxJr~OxJu0s6K7=SUCuSq%S|+g;jh9&?t@9P&wF;;7;-e(o{<%84
zqn$4Gor-$#MK)B#>OO=<zT`$8R3BU7bSs+j?72<DN~%{f_2qi&TOZJh)iDZ8n8IT6
zCC98o66O9!4BAjb^aJMW_7Z_o>`_<=5IgV`Q<Z$kZ@7(1H5CaBo(m)%7yqybfaEt!
zfg%PTd<p^u_KO32?*zQokBgj)YKVZaYrgpx;7l^s(iyYqwCnRYd5(TydD%52XC&dg
z>PCk~`R;Sel!Z0ze(efe8of!qJ;fm(gEx2nC|Phburo5Pafp;`7JuII{?quNQj3WB
z!jg5)eT{qm2`u%TZ#TPedhIs&$`{;GTk1U#=1no2C(p+caA!<h%sa%OsX-{Q&8T^~
zuBpavC)-CZled}-c!`)Or<M!Wl_YQw%nETuSXAREBcFykSv<T8j{*!Q(Wr${RD$y@
zBHtE%btB00K;DsgVXzwoRy83J5Y79Cjz1DXWQ0|Gjh6Gg-qDvI@Hq(+8-1H|Zt=km
z3Z04u^DG94|4r2<m^(@9Ad^mD!vH4Fj<6-1P7vLd;%+yP<b#wIOOGgL(Cr-ucuz)z
zQlLsFDhd1~a8f2<lYx>>*`+R~a7p1BB4g#rPorF5M5A%5)q)0!zJ5m}t`pXbd}@3U
zh9d)J5HWx;3)FU}W6N*I8qGdpz94?ah#~yGX6C>_gXNzBt0!_1mG|N)gB6YN?usg}
zy)8=B4ujhMP7}iH)hw>cHNx`)^#`1t78I?TmqqJ)7xxEgeU?kJXXUlQiFW?Lz=@l>
zM<baQySjB2Mg_}0`XI58KF`Q5?<R8Qk612D9Pf;#Qx1`<OrP758j!R9a14ZnlSsW8
zt2?KZ1{)2T$qhXD14+L38AVUngOYz28ZY#wV;Gtyh-@K(nd4MIyf&O?B_T&4uSBVm
zj%i-dG08(dfe}c&=5_2!7UXpxOv+RMz$lsO4ha$TQnXMAN<mWvg~jVKe?LF{@t*=P
z3!%fF94IV2hPco*0r{8o_4O^$uLjmSIXZGvegch;DZ0{`#(r}4pfl9!eq!Jedh>_5
z)L?6wS<Ldo9{o*1WJHc*!PIEnpeM2%DG-LZqB_EcVstjn{aGGQn#l%!{UV!3%iGUP
z&zkoJ|IVTs`_diC{d?f9NM*gi^5cw`c|2cc7Jl38MfPEUexuekUEM-zxB0l&ZtJ%>
z$}4>w5p%x1yur~Sr@1EemC=u9UJlQW+>#7FS00G|1r_$?<e3opPbEVg^R^W6Mp++1
zvV%*p!{G}Pav8xqK$M$3m=E&;Y6%Lm_+c#=2kIn*parXTzzbhd1yNw79>-F%13&~A
zgR&DnLnPqW{v|+;(P-KGtBBc)Shp~*g}z8TESIkzu=~N~0x&4H<WRwK{)j><xhwPu
zQ|KL0t_xrY@acdlq#t3Wr>Ay({W{*@0A3xQxaYeYcA$(ZMiFhn^A$gZ8Fu08o&^Ox
zMB9gql9GmobcXJ)%iIW4seE7D(EneP7i+D<tIGO4g=r3qj+FJ%EO)f$l^)EDwxC$P
zmE~))E$V=I)UmS%sXda!ZP&C%>QKv^@bD;l>AG)Uq(1x3&8MjD2XuO^esD_hm~{6p
zq1{4sSIZbrZP~lsdk?e0k;}*Kdp=x!#4<kmyzId32;DwT%6pZ|vnw5b=FK!0uI*57
z8`9XZ`p(Kcr&EbDH>JiIgs!{HnFu?ZjNH-WEWA}yn7Ov)c86SX_-~`MNE?sWEUf!)
ztSS_@-cnN+bNXCx-y@Aw^))n1@ghoXFQP|gHm+Pl>-5XewnHe2H~u{3z0is!VTvin
z1?9%CZun)jA3t38$}(9UYmXk1o^<&T_VeYB*RUXJLWH)IhYk;Y&e_YCO$CJDW{+qb
zim3fj5iaN`BBPaSD?p)x5}*F#l)XfDb~ZogFvCpssF;|@sHl}VX{5Ea+3fA@Pld!Y
zjTVOVQ&U^B_jPyQ!3j^=6X1hP<9VW9r`-Jf=*zANqP14~ei>7ZpM36%i;HuWji#h<
zT$<Y?EG$fj8;WB|s^~-_Gc&ir*R>_P>iVteBCaJ<U$G^6uFmXgYBMqWvrTiGn)EW~
zZY}QTws_+<R&bO4kRV4Lr-A3twl~AcLSFKmxBZ6-DdHnC2ZHJ2mJ-sK7OGC!<}?@8
zs8Z(@J@{~lDy*R4&XA@)Jq?ZfE-tN1bsm7mso>mp4YV}>KAO_yPt{$tV#xd>HBC8`
z=Wkt>C3!FBK0wrE@s}?f&|dH}-q&swXumB|xqgXn*HuKvK!HsK?FHZgLS?MJ<BDb7
ziH4ORCUCvn#|F59Ep)`?w>bF1>ekjrPoHiEFo;&mykZd&&u&<;fw$d-43Om6Kwas1
zeh-2`u#bv=|5~*8@pW%py>UinW<+8lEl^V6DWRX(N*|3)@}-PV^6f5Ip6%DA%T@Ay
zbg9br3zM1FL1&i)V>ZjgD#PNoPj8e{Ohy+T6)g4qO4ZA(6s<e-I4|{%J}ce}@&gp@
z)t^mFSi(z&`9ze3$Eq8utND$^@^9%Vb{D-+EP?S-`ND;J*tfexM7Eo?=ibKDNT`QB
z5J1B$V!3j&$4Us+q!Fx$x~#gPfzXbq8fuo5i)@K6TRn*$koVaGTOT}reA>Vuwp3)*
zs#O>iLF5KFJJv<3ctG5TJelk1Y@rVfNl8Kt`t5A|c^tcUcdoWwnBsVqkbsG0l3n)x
z7+TrgeH9i#X?CAZuud=Y2L-oNS#74n${QFN@rK+B*;J2hy`>_^XgTnU_j!~Q6iyl&
zzbZ{jc^xPwDjJ!Zy7At<d;Pj72I(8i`Ye$Ywu^`7#<xZC3`9lkr)b#OgI1srG@l*O
zB3@CXiDAVeU}6;jq$#cG%lVGmaa*btI7>X2-HRB$Q<bG6m+&cQJj$o0n?)m=q7xEK
zc51hcF0U@S-x5}d6fA)<zU>lhY?n?KKMp>7aq()M74;%Y)&2cdS@JPSjYG=X+H@o*
zc(zjiz4zu5LPau8-%Uc|@lF!vipn>*D1EEQ)Arc6#!Zx+Jw1pmWSH^VCWX!Vdq}R$
z<`vY|L`KD4_`(|vpUg_eIsztnP>>&HXWyiA79$-`?4hI`B>mtCS3CT7neolMQ;abd
zr#<WhSIivxYLwIAzhlP^Oe=JI^@<Z3^=8f+J1um{zo{BAB;5Tw<>&YC50#v&)}%UR
zTDIQSLOG2?{T{HQ9$*>tdH|oxDl1Dt!kWW<&SA(?lXd0n>Bxv>tIaEKn{BO7RSBeB
zbJR55M#zxS^jwBOZbXWUItJQ*KfJzmRzrUXu%7by^LLSv^sQ<Aurtmfi^Z86apzyf
z8*|ni#s`TUAI#ErzrIKD#8lqF9qR`N2hVD1K7s(2;+Tbnz<JJXa3)j?`1<*=OGvQP
z*VjLhjV6F`vgSKV1SD=6LNRvg>Q#1t_z+al;IG|6LR(~`EkA{Omblj<EAWIak`7NE
zKYn|DacRjB5S&`b<d53@m^MM_>+5@fi_2~2YCOMVF3V%1v^F<y>Yze6n@}H@n0Ug-
zh@-v0<uNcY`5&*J*jmKkJs4t;@K0$7@OCgUt)S@Z>pQ^7S&YY3IWXLw*U}w?KLhfm
zZeUel@7YugK$O-Q3nM+zz`_eT6UN5IFs~mLrjw2=C9l<x7GD4xzZ6Z}6~VxbJ9dcn
z?*0lW1xei0i1ITuGdo;WjiYouQ=Np<!d5y~ao3kGd!D3cXNyu>PgVX|6JBIIAkt^o
zQXooZ@Ch>U%pqFa#m7g7ftTM$N8MrfK*UDsv{9xyHW}jVbY%i_au$0Jh4rtbwmvu%
zC{Oiy05?Hm>*mOJCaKHoXHE+=V=N3V;60ny5JB1}Nyj-?oGa}3O>vA<xWeK8SUvpI
z%OX_2oj-rB4}7%j{bO-F<$-?Q$cQNgl@ZN@vh3`f95N%<<gl)+Otgx)%kS2uX@7ul
zUARqfn-_vHQBY8TGodbS8ghE_i6Nvdg*G6~yh|$AetPZ6lPBjes_uTPn;|#tpkc%0
z*jOPbj)=1dcKXO9wlPPFlwT8!$jM>CH|@#S50iJ@S(LG#Eo^_VZMVD6*L`d5xJE{@
zGaq$k#&BY_d`A(S9jOrYZ5qN;I%a1lf_=UbfjRB%m+{E1;BKpUcV-K_fQMZ&Z1jnW
zDUC1l^H~A@Am%OeK(JiP-AxL0hECt*{%+#5EFZx#_Mkl>9sli)t;y)j7LWAUZq|W~
zbK=A~SzdoCikqULz8X>o#n#V+7Udt@ob0@pJ(6RuuPk&6cN-d1DO~gNB&=jNd%2Zg
zyQ3H@ZDPU&pzMRZOr<!EgoAzZ`?jd7t5?_5T=^NZu~HPx3yL}>EB+F)q++~k3`+*8
z!jsx)BQHtw`agE~eB@sz$vtvG{jX1y<19FDd{!?pT<`UD{#AH2oC5zjb5ij{vXsu9
F{{uX9&rSdU

literal 34136
zcmbrm1z1+?wl4f4Dk>!+f`XtRpwitSN+~JbUD6%WsHA{Mw?RmQbfdJCgmiaz=NZ$r
z*8cak&ffpo=Q>`O-}mwH#+=U_&lvZ($9+HkGE$=0*X~?Hp-|Y*#e`&0s0;Qe6k6|<
z%kY2HwZ^>RUtD&=%64*=hIWoRHU=n39Xl&COFOeyy7wFmY;0dyTCmV_(zDRsGq$s{
zvgKxAF#nGi&|BIVF*MC7%fW|SwGvaYMWJwYkiTf(1kzriP;VQa3kk|Q#jlJyI>{@X
zHEr&<=^M&Y2g%>Nzccm}@4DcV0Ka>RLYIU!`sryNFJJqTSxWxU^zd_@x4nU|u%kk<
z*hREZlh094u0Or{Am%}G=jkcl?W>PId}!Qo=Ii8bi+5Rj^z7qd2iJU}DKl2x6CwCR
zZ#*|-;pqo|;sq|*lfxg<CxQvc&loZ?)bQgJ+Ra{n_|XaVfBv=XeX213v?TYV_$_vm
zUcY?3CdJZVx8pq>=9p*>OU-wk+1KEw6!YQYBH0o@F@s0n@XMOp;y7@_TI!E>mot+q
z|70p{@AOEAMMf%8&MGv9QiwY3bcwzi{7zY2U43wPh(|`&Ievb&uWH_3#%Z^3U0UFj
zHIBnFXEoKY`BFz_G4VZe!pM~sv&zb|gP#1?yC$g;F+T_L^{!sK_Bhh9Dfn)5`=8@c
z>(O$1b?4a#eM3W=R=SewWMsIt@cqt{?+BIibuR@HGUW8J$A>)g#bGk|b@L9Ji4eTg
zw1SJ=22BnYg<xuR)vd#M)xm|W_+oc=cj;WLX`V8z*-)W>B%RtVp0g3#P9~SV6%+Gk
z1_U>6PS}*0RNN+?5@bDnO#c2Z&n*uR4;)h7W<NZNjs1OUTH5!bWNzU<fBw{Ume!Kn
zhude-`FRmFH8sT;BM~h@wYNHq5w0%hI8}d@)NSYGP?4!`O^pDK@0<J2d^UQd1r#zy
zEvpZyFmOrr8{gliyub~Q^UB#?NezLx%XAQjaZf6?Lq)>Eg7I3#dNp&Jaf!v~uh}TA
z;M`n>-R1uI20|^`(D{X~ukVRDg5jN&RaJ|>KclZ#ZcF9!ruAK^-su*v-1<es`l|b4
z#b^hyWu;($js{xk>E7_@x|Vr1<;_scIFTixi=ID!KD~^6M`(X-)Oj%_O1a$bQBRt5
z+Ve1~lfzCA`rUYVn%`~brOr;*J+`-68RT*``A~4%FWPGy*EMg^t2giVYe}oClZXbB
zOty23Crc;sn~j!7EG#TsD>@8hsIF~l@?FlUSc~5hINj-yh+%$*hJJZtUf?|DE|0^8
z$N9cTmRd=$LZ+flk?N?koSfWkUMEHh(cl>$Vhimb-j{gK4yFrD`-5TKP{MLu5Q+wH
zts@@ir}YgDo10B!EDkHm@I0LsKKlmz5pY~nx3RG)JzlHGA6`|rZD-Th*Jpgy{a~&7
z*y8p68upXUL_V4wem_5ywqLWb!P-dKWo@{Cjs$kYHX?M!MCZ8}U1uKqCH?xQCPH!a
z_bp-66<X49?Cq%c_4V~P>>C;z@8~tYx31}EX<@pCi5WQiTtb3%#la=N93LMayEdA~
z(W1=lDh>`ix7~t_MWS3VDWA!J8N1#@tyja)#*k_5ZO_fUVatl1qn$-PR3kjyWvSNo
zb~bVfiXhY6s=kE1Uo^Qyo||XK>lHm)TU+l*`Cb#9ot^pgEYkLo`PS)>@VzcnuIS5F
zbChx;z~ds8tO@WN=Gz~UTd`jss~W6uw2X+bEEygpl)8EAmd(k5^~ve!tFAAXh&e2>
ztemCN!aO}aZGTnREy}{DI~`2=25MAV_iH##T5%r00}h<s-`}5)RI{+IDIPSlTkgvq
zINtoDq=k3;cFVVyUq@akW=3F;xqfJClc<JO)-}*{SuHXM)(!BBXMkjL;id-4+OQUn
zghax`#3bU+q)nOA;cV2azO362&#!lv@NV5|>=Zbcc>VhIyH;CH5_mrGa$~2*`{=*F
z2nXsevd0BA1`$_;Bz1ohM3<nF%P(XZ4}o9zWT_O~P&z0!`IDiXkOt2_K~7Cr=<zei
zdnoOE<&3J}`6Q3T&FLoFL4zn{^|m1z^YmM83>ChTOQ#W3D>iHCNZ@H{ZOz;&j@)%8
z@_D()?B?qF*e%DedGO&k{LpQiIb%iVTeQlxu+chd&UP6r$``3F<B&9aqM<*o65Pns
zscY`&m<yBU=Q14|7*PIXyk6NeJ3AX19lgM>iI<<BzcN&mDfFZqf>hT&GL3va$xqA~
zGI6u>=Jqx^XIgEWZhkn?;?co$2$$1_cFI!qe4>kAK)@7(#~~(<;~K5?32f;T*lHNs
zwVr4YHfOL2Z<*F17t`I-bDP`F_}y|1>}xL91IF$7PRBGzRU1p0xl`-wu<_hFR7Hfx
z1n@$)jK%TsxERvbCN^fmq-}a6Ibe6)W;cuGQRW&|P||2%>Ex|jEg6@B-2hKV>tNyy
zuJK5zb;=SB4$fRbC;v1AVX|IRaKqZeeCufuWcT-c6~+Bz92en_Pt6Po_A)MM8-zuB
z@Liq}mxcuQm-S`k5D(6<?6-LwXo=a)KE3=ZLaUtT3441JJ}$mTm_#y`_5Q<$ekuiq
zc-OCcCMA)LRyg(_@!+L1Lg?6NmOHHef-i4{N3C3Fj7J|82zyTUO*!0tuD*f6=A<tf
zv&We$x8qt{l$Lu7zBKR4#iI(x^^V+{vtNa;dXV^aUa#6`jB776`4d2$=;VF8UNfGs
zznGRp1u45<(<Mlh)cNg)57&`=kmzxCsO!?3Kto1GcH_nkrzwAGr}+d&EAE6s<1X(q
z+xc6=c4fQ}?~Y^6zHr+%oqYQ;GBULzWwr#p{$lT9-fyey(jev6-n;LATyOtXt>t6L
zJWlZ9+7+in_7Yb<iqlW#H<4zp8v6Mm!I-fP502TS_uGq6(W39nEG#j+&dhHxaH$?V
zxCn)}(s}zuI}c<&*dhL+t&wzrv%@8pTIq)pRQP02Z}KZc1kN~NS-5?1$#Qb&VRzz>
zzI-Z_Q9sunrvgbnQz<v6>$RM$tkYbKk)OZ+XQ-PG9y|z-8@DK7G3todedXoll`?mi
z&E%$%rn5nQp<Kyb9)p98Nlp@+F{f#4BvA%9J`qX|K(6*N6Xac8UER`|kK!hUW8c0#
z{1_JI|Ng!H>Cvv;>QK?hiKeFJylYZF?7$?p+(EOs^5vZAQO9wEU*V6UB~ooID*VJ0
zx_2`D<o(}a5_acm)!Y!2{3;Sux|ET50GoNSFPnI5Y>d@nB-(M@t>nd0(eSE}h=}4s
zJk+mvsd=v1_Fuoq85nX$c$L(xFCFi%8*R_E6PWnI$I3$WpI@E$@#X;~WfweE*$~K+
zaZr<jX>v-v7kkq1Nl4lNZVY}m2;>QAoh7HFoQ)?vJUooBScdd(%}H|c@@0sRpKkl3
zqK1Zs>CR~qG2T$k;ez1iA{qh-qhcBO`ALyTS18vYC!ZJZ$tAZLE4CcRd7`eap5ElR
zGVmjSfM#K|f&;}dVug-P%&rfmrxL0(3a%=``79=qK}%}9+KpzU16GgJWrYc%{C(@J
zZr2y#%FRaN6F?a7d?scUcBs3|*jcP5p5&Cx-C63-?HsFeiDA;a{LBYy66&q?!tc)w
zzp}Mz_#u;KtCc90N|KW+2t%5TFPUx%LBCEyqQ0~V<)Z?f+cPhZQMJ(c2YkZltI*I;
zT9xlJ@k`6UKR-(u;<8`r)y=hp3>eE{*#SAP#rqOw>0)Xe3Zi8)&a(PZ<bI!;Ws3cB
z-^nb4$0NL6!I$F@Pmll5<Ar8?`XnWk@m_~0CAF!(K3ohF2ZxG{?a%4Z@NgWIe|fdx
zUt7%-AsfAN()mz`9A%VZVNY7roH}SaO<r<8UgfyMYAmx-szX#R_Eg9u!kZh9%kuN*
z&!`6P!X6dcNZB4~+zXhPm|PAk16+RyZ`|;1H_X(g5ucx)er9a^kXkC9({#)SoA`p&
zRK5QVK@C;aJG)DNGC#8W`^o$K{jfClbx8?{EW6tnI3(Zp*z8kUnw#%YQohT}%PW}U
zW7BXo9E08VF)B(lQ!zUu8P+i3Gu%^$8btJ~fxP?6{kgezxBpb7(o<9Wo}9Q|xNspL
zMc{a)fXL?ugxcsU0G~8$RxZb@C9lS-`Ch(!*$`|#kcZ}($mgP13NiBck;wZ`*7<G^
z8YL;bHXl69N{3R~FD5QN4Nm}dvR`vPTQcsBKH-HydXJIuL!B>fPr4knNYI^GQk9dr
zILiY_{6v73;1`Zj`vZC*`9__jNBis2c6O}Cd#lq>03XoM1jeu!MSk;W`A7-L^mn>k
z+KVDBq2$NW@4qXu2l~m^rb@;wjFhp^DCNv_i!<1#aX=GLY(BhIbAFnud?weEW>k^g
zUuN9<4G(q|!>+QfE_u)$HgY<;APiC`PcrudFGw9Pm6gSMQYCNmy=D#JKYFN``NypT
zKETqjJ+=WpOiWUe67oL8*>*|IX?DG)E97>7os1SEoQZCGgWGF?|L812AKRMfFIVl4
zWv-_7MnfvFfqGJ4)R}mr=Q}w$+j9tG`IsWp{%(Ld4y!}LW@cvQ9BCpu3texJ51$HT
z=yQ-0njuevXC)+5kG?RLi{$73tvxD~9=)wjPw0ztm;BznC;Qen@$k~geRMvAh6=*s
z85T5#XT0$)XCdLXYj*Px#e?6fLZCdfUe`~*b8V@^YU$Qf__=n{s@KL8zKHit&tkGp
zoe!nOA8G|r!O`1Nl4aZ^IDeZj1qt!Ej1a0KVdQn55Pa|l7ug@k^+EE4A2(6gmf%PD
zb&fmTvwoaS<+jKn3m-rAncx(%CV|?T;<)>SG0n|leOW4U!|flg$;v4Cuyvji`sVI+
z<An|>SV|B>m(d09QP<aey_hm&;Vti-s`Wqm%742e<0S6Tg8Ue!60F*ri`ZkgnjLjG
zx2?@zya{{fF|*CCw@B#KqPU}xZe@S1v1??j)xk-}VzKt78G77pW=yNqmU=O66}MBo
zvL}U><B5vdsz{-aiXt2w93<g)jfURE79c6gqH0%>1V_~07Rij=DdMek0{r}mrKOy(
zOMM$;3kOdC13aXqrQPQZEzfvLVeg}3wLj{J4?Dd@Ul&q)M26kUK>m9orZ2Md0lg37
zZi^=|wgsL3u<&ZT&}>g|NoTBKol`d$*BZla{y{BftAS4RSGQVS@3spPUW&8{Z<yFN
zGpB<6)SV!MhJ=LR0xsU!v2%5Edo@(}h(hFjZENO1bUhS(PS=CH<Iu!J_r9$-cJmi9
zGS}f&{`|QVdPm`NXt{9UEhft`UZ@TLpO_%W&~ThryB+%l2I|3ffJ^L3;J2NVz^eNu
z|3=v4V_5`xSQ4QROVUq;M<13Sb88=+Zt}>nc(my_sMjRakJRpu{0iP|m@#O4$of#-
zhG+}BEt@1-?V`5D;wMHXOm&5E>QKyvsfNJ7pjb$}K{*-~(AG*?SusMou_XiiwGIuo
z%)4Lb&~J+>D&Eqla7a-ru~6u<2py9B{K*3Fo{+F`7wo9*3|2<Qjmi$f^hc>j{3xQq
zcgnB-plf036HE=ocv)}kTkd-)!nc=xEMVF*_`!AAK85D!yWaQM`vaGowmS+i$O&_x
z_CmvpWDlq{pFZ6~NhR_=S5;NbuY_XsqqQ{<nuNWz(N<{FTRJ-#N=~w}vP?|M#8x@y
z%f0~7uvLqW`yo91ZHxXF1S_f*`iV1Cg~JCuefF#jD*r%%QOI#TL<Nf5ehGP%<#_el
zcc~=z_U2}7*z*XtKtRyY%1|Q#rPCRDnNeo|K$wK|w0)`YF@SaE-Q6Z?(JV?QFME4?
zU%7f!2O!MKa7pfn`|RA@lpS{<pAaf9X0#`L?uunz?EUeFvE-|FydMn_(^AoDc;D|=
zC~VBRpx(koM2{ZU@YdkT#mLq=Cbsiw_qtS;R7YWY5ui8#h(G{%51LW&@g+cSQ>k1V
z35CPZkjmNF`QoKZ06L6vs*jdb&5Dp)|70=O7Q?&U#=>g++jC|{A8Il3sGv9-L2;o~
zE5<4qY-8(Jf77bKYP2&QB5)xOY9s)vL%;yybS@_cn^~$w0Swg#-jw2zc{Z~z*eyns
zKL^ju&$Af+z6X^^xy*(V%GS^E@uVMbFv6mg_Kcf))8(#WVc7}XVK-C${JP;y!rm}!
z>__UZ(LWn}nc{`XsOB2aqSbh6$1(dG<(o4n)EE7E)lEb6;+xS6u9_LXM^DqZ_nGn<
zx!K_Ncl}x(2?!{L@WLY^YOeJ{&sWWU()Illm+c%QAaq#xCSX9&{NBHR@c3~CdmO#&
zbcgd18&ok*PZaGfOaL1wb*Od?Ta`dbynXjB-+ozfsMuUwS(yk0g?`PpBRn+p2hVzC
zoiCZY*!kHh8g$n5U0q!SOU);T+ty|3Z80nuVJ$QC)KFO7zP$*2n(~q}KDCrStj}7%
zRGOSu!p#<Yotw9Ul5cmM$gzAS>Ts37p*c_;X<cd?eHM@yeWq+|v_={b8XsipW)h=7
zBx>`BcdM!Vo}3$*-=lJ|pUt%2y4U>Fn*n_f=INlf$FfndvW7!RgNyCcU0%Yf(*`P{
z&}{Gq%ERM)$RtNBn=_YA66#Vz^7Am|3J1DJj~=bzl_w`ZfLn8sGFTa{h&nkr(G8SG
zM@NUAT=?D47H~oNF8daL@x%Pg%+0tP{9&UOO%^3e-{{N4h_p#M8*Z*?+^jz(&5a=u
zrno3#h(=YAz99ZKD!qP@D)4IEmdoo81mdoYsCUG&_gDARzy9v&Yd=(GETKI*WQ#wX
zke8t*WT+UGGBcxvR>{GmxW2KmJVY&9@#)j2KoM#~<|tLChiwDRne{xtAk;lx-grQ1
zC}9>+(Gb$?ABldw@!Qvy08fd)hd}P2;^w{+)haw(Y~Gr%mPH%ZB0m){*wZf$)Z7i=
z`hXsgnHrI!@F6Fd9h$4ZY0@;4r+ZtOb;xWhU%M3ReZuWh*?{w`xTV27b=941ha-M_
zDhB*iROazvdXm0)h|ELQdrg<lG=;siy#jw;%jv}5WQ7ESpmoTtuxaxL&DI!JVRIo1
z3IPTArO6;lL*(^qZhn6LgQHs--*$#AYpf48bWoL*l~-_Z90UYr188#Yk&}NJAUuX-
zgO&$8BjhGE3O9oE#}PP7!DIp*(91^ixv*Ec>|dUM#*rbi$*=$|ydqR@x-kd>93iTr
zqlr+^i4T_9Qb%bx#PDcAz_zrsto2totWM3e2=tNJuZ_eUZqMVBl4>rAgp)^nR4t6~
zhr}<JM{o^WUDVdf-1e!-p;BK0=Cl<~stnIXCqXZ-WigzbRjHuv7GVYsj;I74#~XNf
zUUieTX6#|G4WKz2fB?LG`}WP7H}jmg4FMfLdGZ7W#BF%^O(focQnl;jhS++s(rC4`
zo8HL~iupX{P5H!xHsn(0<?Irm8OR?!@`s-Ii*n+|hLx+UE7D_Q-(jsC9<Hcokb%zw
zVoT@e+pBNizP(OL+A-4-#uV4SR%DHWv@~7RulW_C0XU^Cpf4&2e_a4(D)v=RDwFGh
z)&8Zs01oFNmS>@f{VMq8a>Qze`ij)dq-unXluA%I@3mW`yz@I-eZ&6hZNNJzf7gFr
z=KQ^&K&8m!k4h_WW!IrueeJ));wXQ;?6<<vSgv6~3_wXgXqA2!o2!vJ&)@-4Riq68
zFmZk|FMtpY{=vb9kk3;h+4LH*UhgdYK&TBMYZm@w5@xFw{Tv=1PA~$Vjn!=61HeOD
zc+2*}*N}4G!Osid_X%heujAv_HHT8X`jbHbkC=j*T9m>O;NsM#gvi~>qGynm0k9yS
z05ofk$g%@>0<?lQCvVI-UC}$-z5T@4GB{GPJNeGl%{iXYGc#1}-0=7JS5Tcx@9YOS
zfn6*gwp}jsB?DoZSBFb_veh^sc%XGQLj58cM4(^j)RK@4hVb|`=w!Uw4JiNxMbIcE
z?v4T6B`23nmbz%4z-_Nv=Yw4|iiNx$Q!#}h{5tYq<p}rnUvU1O+}59kG<v67bp8WI
zW+~FZwaNayNJza!qZiXJUa~h(-tW0=9{do#h!6{*5&vfc19<uWhNu61+)#3r=2;qh
zLS8h%c0lh<BgJdMsB;9-qVPgrW~p!M!IeA~6V$jpR()o0vg!`^78=<2i6dw3jcCp{
z;6Sc09?eqzy-aN3|LGv(zq&5^g?3*^aN^JaBB+2x8gRnmKt4ro<<^U!BJm87k#EIs
z3qw#7_0pnZCksj_Gr=(-Y(I)m){KfC#Llm{lCuQrLnEGR$Eyu0`X=<@+BdcKU#rb2
zi?dqeeMxtEz3hXKBbn6cTAesw@!B7rA67cdYJN-1;lCBD@(wZNKCI5v=j@W4OtPpS
zmKkN&36yhnY4}bNb`cds1)YQ~Ca;bPbq9DT**gl+KUTQYn;`x{A;PdD-d`tvI;~V)
z=IQj;5wy|8%9E_Z%3K<_q<3YPW=MGarQeca`>@t6<!uk2J0w(D;j@;1F5>xwANp~t
zfTdYd;I)WJ*Z8mqx<#61Ya$UnA=fL>JPaSa!B_XHb(mHT?C#9ERa~>^i=cY5)~H|n
zAt^y{&;OazpxafmUQ&3L@lo7dvQbe{zU_uQrq#8zwRenzdRf){aIt;asBJy&6hE<q
zg<8}5vHCOri15_R%tMclOSIryKU?fTX!ezwWHeDls?a3OGciG-b!8ICr<|MHp|a?r
zcwRg8Nn+|p4*quIc|=P^D_5U%2jFrYZI@hZwEuLN=WE4k{M;%`DtT>C*(Ao0fjH=B
zETF~gzVoTlN%5Q<qq=ZY)>uj+1^IlAtmzU^wND`JZftJehq??EixEaJW6tw;`Oi;m
zp`bx}1Ym%|A>pjW<`{Vb(Ea}X`&eZB@xafPwGsnM2|$0MY(9~Kn>+RvohoVrkmT=N
zEdfV9AelH~Vqz#ecqpNR9=52!hWXEzFJGeh-Ci4%Qx~<Ek*_tfYd@Jfp`#Sll}r^#
zJa=D_Ck<<mcrra<6S(h>bMGjiG1;82F`vvmvpY4=>ElGIhF^{2t-Fowp9b@MH?RZI
z?lBfu@klSNVWX0%&)#R=!$pajRWP^O)fYClHj$w+bKe9zN~HB@oF^|T;A-#M%PT1f
z(cdA!SpqD08;Ci_5o-*n!4&uJd%={3*?CJJ!MQ-pO4!&iLs8mU84Lqz0vYc>a)$<A
z|4)X(!SOK*aL<r~M`r2Oiea?nfUx9ez&3<NL^OO7d;{guaJ1as^<=Y&R--%|3SmuK
zv1Bx(*LwAFFu>b*AiTmpe5h><O7YO+rR3lsOzTluxnD$G?m=?K<n_R<I-#;R{zg+z
zNt#Tw8tO@_OGWpRmqmTB@gl#4?LaH-us&PQK_mOk_%@?)>fwy)O>)woRs<hCm=n~E
zuCdqL^Pb}Vy>vvqi5>Xk)<zRy+R2KX2LtxC?}61;!{|?c(f0E^>^|i!RFK}XWoC3%
zxoR=uAKB0E(=aDT2aTt)ywedI7gvn8_XUw)lIMU}P^&`(ZY<jnCQw_On+4ph$34#7
zp#8U<Yx}Wt{IC@WU6?hJ!#aAQV_cY+pKprivK{HkL7)!MI}i+n0s;akm~^lK2x=5K
z+r%X&50H*CFf`oC^*9aOANNQ)J3U6<=-^rZ4J&AK%woSJ4^vFpt`flvFUwOD`Ek9C
zs~j^mJJ};PX-yNHim@@Fom#2`5e#knf#-uCB&z7UqT8sjlm<<mmvBu2suCG0<}6Y-
z^xf_4pUrqXxHjRE3|5B!9HB*#Fqs=OO)*4QWxXv~_nmx9dnzQi*wXMlnr>|#!$VnY
ztNeVDZ1lZizhwO!*;ARY4wCWWjK}5oZC++r$@G7IhJi3dux*G~jjy9);LHFMOTk&X
z^C}lc!2ZC;<6plnL(2!Z&F8XLs#RjzpEJN+4%6Q*c&p<RjNh3`46t;-JwL?jT^%g&
zt*qoNcU)hS!n}IbT6#9p02sITWCCNSLyLCZFfp;|3|*SUpA2cyr7AV&6wIzeF{%af
zU-TaK!8@=cb5A_tQ+Y1s&^|cB-^tJ@s9bV{g8Zgpcv$^XxQAcT;?dC!UFP%6N4+zV
zd$65ErbN{;isi@&qelv-w7$hZS8ML<s9nPMq`UX_{kTJUh>9D@Op*kKgp)Jjsr#&9
zxEq19reVghb6;q^?)im*ySo`*Fv$tCuVL8V3VesOPXGgNF=)1B=2rhgNKP2kink}l
z#5{L(<t2075cJp6(a~9(Ps6ZeYq2QND~aMWpk`Z_zvsvu*6_YTw^5&Yl8{(TQI?xX
zdvov0j}L~Z8Kb$$5xE_&so`3|k=<-Gu@fg6Y7d%ND&`nT<rPYgm3Bo;!`zR%0qSPt
z)4o=^O{!bWf$mWx*lK#akB(#$mwBa*&a{kw+de@3kev3Yk@>=BnEQocw&S!0H9?3c
zIzr%h;X|oDj*94wmIHHQpVoIRW{nKII7Z#KPJR-XEq6SpaeJ#N)FGLY<IWZ%v9t6i
zP_yba6U;z$_4E!`3Oe%!%>c!b#jZo&4l^WbUS1O4=@G)8Fj@<6-kyUjFpF~l%+0$&
zGkW`-Y#0SAXDjyQXoyQlba{{U_R2e~jXZ~8)8ih=be$#+DqB}QoBYD!kllLH6*;q{
z@&P=${L4cv;;lbOmjW9Me>}7cEwthhIZLNmY;*EGE=xjRiPvy-OYu;@PPZ0}Q7E0z
z{AhBOK25OGw565P3-aLl#Ftstcg|c8@^9nueL1VAF`lZaKW#JMz#pq7+pD!gG2^V=
zvzP9ASU%Y5BP|ww%yl!U_wK3lZM?g=pWH+6-m<Ry@Z~MEy6tVgNnX>`>KP^2d`M6&
zqYJYL-t&{~EUlVE=*sv{cjN%`i$Jf8@S6hXCoB-x2n`HrQzV2`vg-9N5x*-Zq|OIS
zOrcOrNFgNLkF!@YaRCblm6UJ*WS)d>*x+z$mUp8TedP6WHf5?N8q8d8b6AkzQ%Ouf
zBQlG;0_dvf_VQOgY1V7}f;v)YV><2=%wprt@d>u<p~bVx<{XkU-cgln;_(>8I?OIQ
zj&DQfc{?^&O$Uy;SZ3lmFFu+2=wWTOIzMAT(nLxmCFpQjl$}9mo{)uW7-!-+4gXYa
zMo5dKfRCfq#Y5@@lEYVzhHptUJf!{RMH=9gb1kHoiv&k8OZf*)Zl!n6;sbW}NSJ-F
z=r><4w_Ai>@PbCU-F+68u-VqA9vCGduLSO)1K=_AgrD3`k9O9_YeqtOw6wIU59f$U
zUoYH;d0hYq5@4)G$70y_a62B_fzX(k7TESzu3VWwrkab3iv+7*dhgd%mzE0+S!%6K
zl`yo9>vOU&1!3byB{-6X2DazZja!?wI?m29slOvpJ?8T5bJsLXeE)_=dB2nA1y5WZ
zarNAWPi|*H3b_qc*k?87o6~mOKxRCc4DH-3>F?kc2o3B)z~i!!W!AsI<B9~YOo~C`
zk*lP%bnb5(S}w=fIAD4I6W980#OJ?LvH$$FO#1L!4Xm|;13cPt;&);}o8O0sPPab(
z3(WotN)Hqogt3dl7ldN^_<Z>XnfpKcg8z;R{<n7nlWpK<I)1r8g(=t1Pd@n3F42!O
z0M74UTn)+S{6)sm@4%AmeX}>>{%$^AwCl>>cA@OarO|Wdq}|q>RYUn-Y~y_nn90nn
zHz@%s|KCW2{vzw+UVrsu*D5oKC8tXiy2G;L93Nbj?jN(rIeT=9q@Ks3o@9A2hVkGe
zTguCkSDp?gMf@7(@z6iY9oNwgSJ_JrMvobJr5)2m>~TT@@5(F}Cnvl)eyi&0-J^^A
z)*72OZ_z&Ru9M0})f|da1rqN1oJM@o)XVc>{Z#9GbG|%;>!?MJs7%Em(){~tx${Ka
zjo{VyO4=7iboei&V_4^$hlevHHZ*r;udO|&0~M4_=<C&wrbIniG^m<XQmWr`lE)v6
zHL?7u(ioEtT{yPu6map@rlm#PafW+By4g8Dde~J37;9lndz^e}`;KdvvRT)4+Rz=#
zwS>2{Dk>$&pI$~d=aWUQ(85#9a%P@$Z+$@1QN}TK!P*kPcUcij>AthCqYkE}#KI@h
zjo`)lGmjnguR9D%RGA5$!AhY~-;+k(sBK)!3E?Xz&`}cbZx9HtTR$n|uAnLs5=>|o
z&QKB}K?S-Gcl@Tb-6(bnwb?E>a8mK##hoiJ#d&*tfc|^iBlkPo>J>-!^^UlzocU_S
zs_6V#9{1K43`Hv*t4E>mw+kk0Q^Hfb9_SNQe8g6B^YXUqLQB;<3hSZeCz6#HD2&<N
zwEU;X@)NEs+0&qW1)ThPQ*x)rHy`Vg9x}LRAKm!veDO9hKTVNYe~Vr5=qm2JL*c82
zn3CuD_-amiu5r5=IrPuzdSbal1AF&HZ=ey4e4-QxatdGV=l2SUiXiaaXP)j~XN;3r
ztIQb}MVC5VHYm1o*6*>+F_q}af!mm7R&P7lt$4ihuK8Ui<MKrjE-kB*VO(p0PUZcp
zs2Bk^e;S74{pmNIvV#0C9owiZwp3%=#eUuzBk~NZs21pq6xUFoep7x4^VaVi4IUsd
zu3%#ufL21^fFS#BG>Z|=Kl=VS&~$Az5?hvbR!j%;7{dgvtW4PS5}@9N-gNwuQ3tgc
z`~M(-=`VX|R(U!)=3;y@w|anm>fKC7I}z2ZFT4Hy@TiuZ+5Nm$#rDq|v(h8Hu?6Ss
z;}hfYTHB1KxBkAv>X&QJam)<Twv`F?h3Z+H-?7>WFj$e&z1AhVX_$=C#`fnjy}fs2
zzWHhd+ti8`^(u3Ge&ZNd(M>N!>YJ^g1^{)Ka!&>GtZ4(1hXF66&5RgU9mp~GwY!s_
z^X`w>nhh5RA<~w##Mc49xng*FoNXr|YKVG#X%r&w_!t@42+AQvLpQlJDN07h->1gF
z7R5LV^S0EP(@#fIB^;iKk>im`rLLy+G&gt7B9U{)gVH~Sn&_n6tXgs`sFl%g4|^s-
zW7+C9J!&iHr)`FjlLT=r#;t6^*NahaO_qY1g=Y$1?i38}o8cA*6@(q^i?>MgiOF=T
zU8y)qH!+LvP3100rT=%>U4*W^HPa%I#Ge2S%7yEMgiSzpBh1Zh7DFr$HC3LU9q%mm
z_yUKg={!pSGw|TL$%h5ih{nYI#A)9&ttY)!iZ?N&;TKB+h8}y&q1||A@{D2V$GC>2
zum0+Hdm_)e+R+Xs2i$X%SzF!4D0T8j2B&VI98!~p{e};U-B8N5c4V2sYnemAeolGb
z_!uLdRV0%)lgTcJT@Cbjh&rVH*pW{1^<Kz|My6Fxg7m_~W-sxcl5Djr8ih0=t#d3l
zhwOg>&825Y%P~L^rl+UFBS7Y@AXWlpU(N9*G*Af$WbRz=`ig-JK|A@6X*~{GwMt>i
z-<sWNQ8I1C8NesRn~)ntkU4naiEGtm)Bv4kAx^->djNVoH7A2cOqadQ_$vo?AESb2
z^}nzDrbm_cnKnn7lIpdxg>)5ZZZZh*{nQ&J)HkL}5Npss``&!Y`^mJc8{-`38Pl+9
zK=8S`-AIMCQmn+NNLc#thK0JFaD~>gyHk^cO_JsJ<H_MK#<$<LOucJa(x7nr5UVgR
zvIKOzcqFY346kT4E2EXP+#;xxTt9+dvj)|c$QQQ;94qJi&8XIf424WkOs<SoRfr7J
z+kECsboscBv;J9HA76ySKTS;;3r*93+%2%WH(u3rU!-&S+xqgv6U94{7nUZCie64U
zo*Da-^x^ba!M<9PUG?(2iln492J_<!d9BXxJNM32pH<<PF<z|KdR8CA;C&bpGL;wF
zd7XRa<@f!p*@?K457l$eF1U<H%q+yb-KoJ<bHYb`cgVV@H~kdV#;&!}MK{`s<{Gz8
zwGx~5$ci-O-tP6jjYehH0QVXeb~i5?@nfOaU*KV4Xfwwk{GeQ_gbs}8Kw||^R~UxA
zF8gaKfL|I&t*x!sY||v;8exKQ2op7gCbgaK$Zh~0T!SL2eY!y_vVh|BVHdyG0IlYU
zm7PF|`#xGhaqea9mLGRHw{2hH5lMNyY@T|<LzK55$v=ADs?utC<NB{Z*u$ueUy%{B
z7K*esOw7ledBUs>xB8kB=4P)n-RioTJ8^^}iRbgt!%X|&!Fv_%@EUWlwd<w?P8*Tx
z9H~8udjI4}U-7Iq(ZF)a4}Ipa0uL5Z0q1~m&6>~UP)F+ZiDpmad%a;k{@rYl873lP
zPEMTAz|Df3PiOMyI}%=}4~TY;P}9i^84Sh6A)~qHl9C26GeI~?ar&~ikXjkd1{lX)
zFVi8)Rd&lU3E)OM)7{g0o@;C0OYL88n$gRh8w0v9?OOS{fuiPo#y`=8UXxcwmF#Kn
zyy@%L<&oS9VcZ<2OIDnSJQoPJYcOL4)m*A{Qpbn;!pFM@M@P)&L#$L1(TEZZkoh#|
zyO4fc&7$lbOb40aDo}_{tNLV192pn`Bz<WXtB4HF!+Og?%ZE}!VVa(P?M^z46ovht
zT&bM&zm#gg%l|78v?q7Tp_K;S$<)&}!-h&AUAd^@5=^UiL5n5-!b$%0?=aoFocY07
zrEnYt+;;P&XU|>&Ci*|qp#L>N`|qOc-{wDmUGGo*x^}AitSzSHA{TwzF?^5H7r$m>
z#jHW~e@*=HDN=R6ANM)1&|kY>kgL=6?C>whI~<*eDH^?yj`jaWl`hrK)q!e+f}TED
zGLF3q9>9==RQB!hG$y5@;7`KQ0{5#Gag1g@=FM|<(vCf!@()8f#YmnMd9YY4cY9v;
z;wNcuGx*Wj!dqm_qvHI|DEq79^mVE|Q<44Xd#+ix;H_H?k#)Q$hmlKbiDC)!vwkU)
zcPoDJCLYxlB%f~3_WdrtsI_%-@0&$*)w4%pqryk2G5Qv6`|-=)-`PJnYK(}kTtS`e
zN%3sSImSDZ3C&C?P<wMfvhm2})b(~9u%-=LIeB`=_lPMqkO4i;w`+^6wWXPELW{#;
zrtjeL&2lV<6#XvF`Y2xpQ;DV{oTE<FH#*@OXS;X598Ao0IH|NH>(A!zic$TjR`e;v
zQj(#*Ky;~W=0Pr9q+v0J4(2PgCpwDEeIFEs#tUy0{0aHFGJP|7weeP8aGsYf<1o6l
z#p~Uy-Dujex7g18cYlM>=+Y&7`{wwf=anUs95~4a=#0}#?7XfjdS6J!b}SV`PSXcR
z?u?^6eFe@V`ju9vGy3pctwlMG*wdT**h5|SmM3Lf_li`U^z>aSX$>{itK+C^#-u)H
zKD@9~qGpo%hTL9#vZd}p=eh<hUEw;-z*AM3Xj|{L)pM$E$K#A@r&a_r8^e`-r8aU`
z*AB4+M-%TYd`xRTaAp_Lkt*ALyA$I>mrNPMLpU*hQogkE(NA5-#7;7yb#@B0<;jo;
z2UZ<G(dV>QHveLVhL%<z<e{Ltodktd;4Qjnu7ZQ01(<p-fH2S9-5r^712MF>K8{-p
zv#5H|lu7~T2zvK}1E;=REiDQ+Ub0Sx;aQvol;RrSntP(!gsxbt)&8=yyVT{0R&&wj
z4u)8VyRJi<EZD}1Au<BQsL3-2oD*|YQ!1A3<f?}MM!&I_*sHd&`v)br(VsMSTI$Rg
zfk#2A_2o`OGow06c|~x5q9R-^ReD5*Wc6I=D7C+NY~Q}9P)&$lij7;A!+J`{ursk8
zINr9DD9slDy+AU_5ZXCw(*;sv{<8x;7?r71y=FtmUl44KSk+;}yp2(}{>obUGCiV!
z4U39u0%7ICnI6yU9b*tB)SsV@pX=l+4^%p{C}tM*vID667#FwfzG}yvnYK-!&;COr
zs{TSAHD#bdg=}fLBtYP<wlD0&mm%7^7@K{41o;gj`^nqJPfv;uE}QYi3Qk`a$g#kG
zK7Zs@==KQBNJ@}xS7herjIl?Ei+o4fp*>VRhWf5W!8Xk?R-%fCpKbbCyF;a&`3L<B
z1-yZl7j={>pO58hiJiKWyVF^G+vE3ZZur{hUH<+(77P|z04X*`GYOBo#<9No3G46+
zMydg`MXK2#&2Kp6hTYF*+P9H(e&S1PQHlmRABL(*8ukP*9Y*^E0>`zJ=g*%@02>Sl
z`@(Nf=`atoAyPs6KbNZ6G0B40+9o0=AMn_gzFrHNAt?XVX8Cx#nXIzSV<z^b9Vbz(
zI8c20sj{^e!<~zVR+7VFV=Cur1cjRQ&y>wB+LdB>X04LRI&ED+m9+?1%><fWT{=@T
z`<%^m{PsiHs@nzC@ynhsqqPh}Kc9}*>SInUP`_ckX4~3EO)h(zjr|mON_HDt+r{yk
zbFNvB)7^gmkPz@{U;%lL)U%HtKcWx~2v9LtcUb*Eu*C%89AMK&=VAPt2x@~c#GLRG
zlOYczXUxknoYv$(V5~I(E!>g&b^Z}~t<=vX7aRLHF$a?NneKkY{8ce36h`-%avJ{G
zad>*mTNl;co%CUryLM;RYvt&#P$$vhb5f=^r_>ZQu5VsJ%eMH@%5bC3DzthsLbvJr
zzMJIM$=HeqML4?cRP$I(y&J87y-|$8(9a<q=3&2TYs&on#CwI~%WHaF9Oq9qQaY8K
z1t9ai)Duhyksg$LV{`=bOvG)2uuWirF$8JU2<RtgW@pQ`+Y)%9;9>qQFp{3@Op=CQ
z9YTdgIUOygfz}lhWP7tfP9X|T*S$fbls<!!VGz4EkQSxbzZEbgEe)r~exsbf&&j8C
zI#v~Nwf4)Z=xArIIoYM!-->t#%5)9Ej^;%5eNFU6%X(2gMLO3}<7+RhwTQZA_Y3K<
z(MM-jxg!<KPj8f6=i(N)@Wf%Ne<tmg7S)u-c}U<R+B57cOJ_Gx&X3fZ1<q(t)sjAP
z=fzszJcb#RPri&EN*ZDs#(8wrwGJm(6d&n)AIQiSAfGxkF*_;(`S`)XK^{PywaV=d
zo9!G>!+@;P29)3BJuisOm!ZW&hNi$oBf>yjfs+paY%Y3&Dy65=*%lckgNU*LSpPCG
z0Rf$?uoA|kF-l>G=D<*0Sx|92P}4`XnM)RSU^Q(yGE#<?i)Gmg{dCuZUR+NCmyiN0
z@vewmh<3WVQvjXr;+>Pb3jVwU6-=F2CO?AietkuvD&HSjcboR$FeFqeG%&$(I&)+=
zI@vN>bIPfHqm${H6vX&LO+iT}7w73;o#Y0Cd~3Xh6?GauEwY8F`z0lDgfvYxyV-BC
zPn5ctR^>pzoZjfoZE0yq>TzO?sDsVeiGiG)0@*U?aZ>8T%?b?Loj@UX2)a;EpG|{=
zQV)hWkg?G|L`OHPSynZGT3-@yE-;+h-~}3S7&g~8GD2KS;LhO&9oGRE0+dTE?gE`>
zE@@xXy`t%X74MKiro1x1!4n`d+2~#vGx|tnWGLmn<kM5m$cR>I0qX4%;o;WNuQ(H-
z48>N>F?FRDOsu%u)7r0bw9I*GlPw*>D-5M)bny?4A8K?=lC^LhxcN_0<JZ(zxLaoZ
zXt$m<_z-cxmTr13<^GF-ZB!l)k0VScp6=#1ZcvDVSe4mj&kW}-w~)O2P3VPjp;><H
zAnHFdq7HS5l>c~z1?GJ4KgmH^1OkHC_BamPnPxoXm%v!Jf*r&^AV9Y*nn@y&7nulw
z*QyTM7*3m+E70%g%`}Hf#Ia*rTCNuqcP-l8AR@Z5@A0-J$KH_bzWyFtR@@h!K7YSn
zf;UE&^C>nL23Ztq|4#OIR5g}nT1;5KRa}4gEz=oMZ!X=sWKT%t_VNE^if*#Pu0t66
z;T=9x9Km+q#e%4WA1tSc4UK%O!moMZKY*ykVAkDS5O>aafByUsIL-fFPf8!o`9_Bb
z#hK8dYJtqr=Q(%FUN6}yW+Z~T6|6U?-@w)W^H+!hnvpTDIBqgj!sqFn?tkcGQ6TUc
zZb!J&EB}(n7VddvQ~gV{`qbe6_Zt3<$oy}3*8ejNntTBoGGCB?j#WCZ6@p{&4qTT@
zP{yZ#!hHz$uLiA=FsJ~TA8KK^gaKx}+0~c(@cwL`tE$F;Vp$imJ^{Tcy!jvYJjcGu
zi0rTAji3&I0Lm!B!F!bC%^lhkfAooV6(LvRZMk6D$qTxHtZJV=Ptb25R<l?rn=69_
zh@>VXB{8@9DB!Pmay#TMPjT;g5>&;==4CKFZ8i}42L{%G;2o}dd)?!_;+h%*xvZnp
z9|v@yr7uQ_&OW83rMj<RvPKD1JlG$z6iv&(7k9F)?l@|XC|mESsZ;^PvI4DJJXaW~
zU&?z388p>Xc3pS>C{+RzL2zwl#eoN=n2#cF5fLeqk0TQ-Ooqx@4aZSo*yD#M+nr^Q
z5tw2mM(lbdXZto0_4QLPSA*4oeDs2{sqtJ(=&KN$6FAt|3SS7(3&BHu!Ss)zfb0?z
z?2J1WlA-bOBxdYtDHl9{alYPp=;<=Fdo!-xq0ct6A{&eyBjCrJodZYHJ+KbWgDVUX
z3!zWIFcfhEB3NNLw}v#ZLDu4~rskh>>hIqLV;XcHm7J2>Y<BHrid_mXPF<6LX?<1}
zEi%5}@10m&jzQd(cO6Frg@i7GbBx`5Xaj6>gfuYiMgs#3xG(F+&-OSF`7Vqgq|)#`
zq3vac$@T?w^v&Jf6!>OPNoEMCRiB@^AcG1R)&Pb<Jc_(&?ugSBydCJ*Z{6}~(YFan
zPAE&wsVtrY2_<-&V%bc6;N_r0R0lCNxFRk-jD6t-aAI_QIK8P1bjfx#WGTq%f~<9Z
z%w^p^Ek=Us3{*+mP*C$hA!yUeV163cqCW-R6(i`p3(bf7PR~n>yT06*Pg-{IIC@<I
z$|61H7{qY&V&)jKHuy$QAWYC-XvrB7UG_LX9)I=w6B@`%)!h$gA<56iA1NxvBzc^1
zLE)wZ1s-Vn!l)#jpmaY1p>&Zz+A&}OKL9-U?xB3^1YyJ*4RTDCPW;ecU1AAhk3e|@
z(CtyAVy<SDEjO}drjbVk^57A0wxAHJs3#aNty(CfO5eNzYe@CkktsrN^(@W+ZlG7}
z18-JfyWw<~C>b*Kk85uPdqV4n`_JG@jezHiZa19kND!}aKScznP<_xRz^0`G&PNb<
zZi0AK4$M=aDuYDCX-b9=eq(O?8b}g9TR%U;1&}rbgQqpFl%icQY#A=~rmyYSE!ur|
zKYa}f*Jc7G^&d%RJJe7P>OWG7J3+bcSUkPN>Adv-`iwAY=^NnV0gK{6lfao5P;xw-
zuw^#3x4!~vZRp_JJ^RaWfDS5tnRgac4S(fo@bD<_-TTX_B6?IX<KW%7b4~uNYP~xJ
zGl-aDPT>3y7lwR~9L&y64liP0fW-C_b7vCks~gB8Dg!Sb@^`S*ngWLD55R%+^OV;B
zk_@7=016WZfrk*cG4OhzKcKCOa2|t;Cj$8fm>TEF&JQq9!2IfgGJ|*L_p=C#5Wn!c
z>v$#LQ@Zmc9k-4VY^Or#jIgfD{Ky(N#?-ukehVPH-ivSnuuo7te*8-ed93c>^1BVD
ztwxAzY<M5c@rXUUO0D%AV|9ppT4$Ldstp15HbXvH+kxe}14LINEb62bF_Z4w$<<Z!
zm2oi36@wDD4t$Enman9CmjPfjbOT4L2Q+QdX?a%h9UmRLh2KAjz=~M5I|8lJd~c#$
zqSMsXY4GB50M4{LIUYY}heE4|5`jp252IAaYEUrVfz_n9c2`u4L?e>s-4rU@4c|Mz
zM#LW1K2YUS!m|o#lo>S*AjLjSVWui1D2Q&afEjwWI+LP1VCb<*gVySfprBwQB!x*S
zp521j=xAdiAH)ya3nT4D@XQ3Bnw{+Cda!xw+K2a*bpK+-`dWXUu0zy_giT$4e}8jJ
zi*;JiZI(vApdkAicd!g&fuk0hyl_q}p$uPeOuX1UTFx!dZ~3SjvTZ<gImvbFDroQO
z7_nq!Wk*6l*a%jD4%ifd2{7>_g89QpE{Ek<8wl}BA<=TVpOhYh%AFQ87M)=Df~7_T
zlaO<eE)v@5k@*rZu#SW#Zd)6weKzJBgL*MgWU3VL2T_b5<|jnLgCt}`LT%g0*92%q
zIZxY*N6W#%A$|FeRJMOxM@NqQDf(qBtB}eI;Ibg~twS7KXdTB@HQ*FOQW6L;?bGOh
z%`GfswE+7VbEg`3{y;E`%%~eu6+1~zv%mG6OS3RDL+pHZTpx?)b^ZtyF3MRGWKt?B
zDhgW_pt2GXguGF1zuX9-$^gEOejtSqds24E7&$F%*1$AGHpW>g=VUlbZY>`H39M|Y
z&~2zkh|mV~kH~D@!GRrw(==VdAh3#e)OukH_r36V3N8$MWl~6zV$`_+RnIn}{}e1?
z+U$ruF`cI!Y~v#%BR8CFZEQxGXr_0btcHV_RaQ<8xzBa*y3?r@>%*2@D+IAE;z|c5
zQ)$i_WT+SqUs{4H_s5SPM%g7Gr#G97HO;xt$ao!+qOx-IF?qz!&JJ^D#r~MftFcNh
zuqu3p=Q?8DbT=i3?`CMmS2lZcw)$BcG%Gk}AY7Tj{#VR$Uk4ViR0I>ymE3><Om4&M
zBIPp6pX|bWi{D%6^}tr)dUiOE@aK?nH^7F<V%kTGbSICqc|e;6vh>m(kdG->yK?#H
z+$1i0AtiPB9yfQntbK%i14L-V<6KsbF((Ren38~K{Nlxn8$iP_g8%bz#hsF@z?U7B
z2lv5p#qV}BVa17fHjTTJuYv*5Y@{@_iOgf<bXz$VRK38~wsDNR)`8%l9r=s&IIG3O
zFRHE!33VXw#zbbw*AP|*4Gj$<Q}Ol8dT(J8%v?&|L&Aq}evz;@lCLM4vXqncPe&f&
z$^!*_*~<4;q;nh<)KC~Hn(QsqhPC9$29GcZ@qH0}P`SOS;}Pg*Mw>F9DOBzww*w^m
zBw&*J<m4C9dYt>*c0q#T^`&bBad`ItVwxir=AG?r27jJvKAovl&|bnv5S9UIN30IW
zH;ycVo_4+$7itXvRU&pXB4`U-)~ll6Cc|JNmN78qOD@X>hcw7ygZdeQlP1zIU%7N|
z$n+3KV+U|Jfq1+vAisi0Vlpymq+Y@)2hDA5v;|m!4L?DvrCe%tAIjp|G-Tmvur;P@
zTRvVt*qm<AJU%||0aLLqg+C-*6y;cM_sq-;IK*C(pMqnrfe{Oob}LfEpbr6yFx+ij
zf+5jmXv4lBS_y!NUq-p86*k9U!v}ywn+dT+Ko>4AFQ5M@gB7-Ktx3@qIQVW6v$@Vj
zD7~2B+nv>J2q0K1lpEfNA0&Lpe=Q~|N@<0Ut~KU-->_hi?}i33MGHA0dQh*wJi@C*
z(rf&Bf1AZGIG74(8<^8<Bg}>Cp$+J<IuHm5=6x>fsf*zJ*I(>TX-LHt*JF<ZJ0K1c
z^FZ<<OhIq1RuVWJzr))|a4^n+<U)&PHn;{H-xo*=Be2$x9ubRqH&jy-m43xo7~vrT
zk*VqFtH7lomK890|Mgi@BPLj6zySLRNU>ie5!~QYM_TdTZ!gz|z1?8)hhSGMQr>7V
zK2~hAg98Tq9npwxBq>Q6Agc;I4#bKG1JG}MK`xn7e{&jAJ|Wmag}s(V%;gJc0-z2v
zWXzoMrN=p-<pp?H5Am*{%aaIP%?*%sms(FFnFUl{08-+>)~V}d4BHe=W-x?ykQW^9
z&;~&Dud<Cbsow73K^6+`1<|fU8B9XPV358d2Mk4oGQ_eq9BE`R5{9~U0M(?B3iQ)S
zg1Og1-C@wcCnQALfAv8y370NX4Mm_@U}yd@0Z-5!DDWHz?c|s!H~bcnQ>GLnuCPXs
z)CB$hEmLPvLJVgHG=er&8`4ykay1AnI{-^@n6bwpg<B8oeg+ms)~Nv&uVhu#kck{I
zd_N1cI9Om=M;e|+C=o_~$sTx0M7X>go^ZZ`AvmgSh9hHRO%P-*0$+&@q8;Sx%!dk@
z{<_``5@nT@jS(pnSUR$F3I^Z93KFf?oS&&2yT3k{#vlSNLCe3=$U<C~X#Q+0${`2J
zC{($T;YHW;(|yX|sNz#G9i7j}St|$x0TC|K(cY@*36#&2zI>g!%U@xj=F@J-0#y%!
z)G%Ci4oca3AY$r~*M9l>HK>WyrR@NaFIFw&#WJ4p7`0dkrCK6Rt50~Oq-#0T(jG@)
z(0izVfyRkV!sq3?Oc%t`n5<|l`ClsE|9!CXpAN$jD>?#v0>%+a3qkwOF>I!{n3$N#
ziM{=LRiNWZR^@|q@%|3mFE`6>up)o(Y|($byIF@T5Z+t^3PE}h&x1D*9PKkdFvw`F
ziHo5UegrZY5{8Q0=ykmG>!VZ>LK*E)wg2U=k2pf~gV1@X(yq1<w^)u<#(+k&Cqsb-
z0`P}r&1oZ;-RfaU4`9hg58(tCWGD`(0`dFS>NEclBZHicSfA6`Vf;u*MdbsFm#nz7
z4lZ&yuS28E`Ru5_20*U_h_`Tj(LI6GLJsoCciPm2+22BkOC2`H23TT|y^U1IY;!`q
z^oJny7b`-x8ALk`Q^ucfFDZ~+^tVm2Uq<>K1onbAKYNQE7y>x(KxT<4tq1(W;-xT1
z96`HVYhE(y1OE4OIGiG27TlB64<5Wld~d&oACcQsgMM7Bh!eU&NUL_>a_^l6-_c*g
zKR9r-!AyamXsDvaW`k25(>7&*K)BLGA;<c2Bee#y5Nvg%=>e5q^}wc<`{6raZa}0a
zOFL><0D~z}h*kM7ZE%)A0f4Drl^|#O^8y_AvULBY<6eff2<%(W07AFIieG#JNG@hR
z=@lsUc2<W8A71;FQ?;jxBoXkQ{?#Wyg+QG8a702XC@N47sG(0JAW-;H3C?`Q9Ca1)
zD(mUSN`MdZy;OLivB`+ny#b0A8Yrjfcby>Xj-uUt0!5ZiwNM+*8dxg?>uw~iijCB+
zJwWHvN+6e^K=hm*L0Ux?<|~jD67ig&;3h@;kdT1wIcmZ1GO?tj<oICG`^}pcICKQs
zCYYB(A`ucY@kU}o(MKmdmJ}afj+~B@Ga6Kx`=&q-UdAGlp<Fcv4=b3$5g3?~mv=p|
z!5bW|PtsjFXJ=+gLz33yJ?wTf{d(E{M*dqLp9?oJyQPfsm0w5-@Lhpkat2WUyu6l}
z*cJA;>=9nD;6P&&0<e|@#}|74Ue=XG<a~$VVEWL+*jSJNTBf7HQ77ieNeh3zKw3gQ
zprTSNe-9mzyoscS3izx7A+7*4p`a1I#lXM-BcV>ndlMaCGH8NL6PS|lF*;fdgqR&w
z_+az~NtB;_AOvj>l>Hk@@rptQ{k<Sn2b7Yh+vIVyNL_s}h3SKJ+j~&mZqfth(oj$*
zPWQ*>)6){c(v9+j#}fkf)0~>KqZ_*jO6vM5s#Givp#q@RPd@IS#(Nzik(9nq3J#K^
zK^$X{dvQU+@c{g9aB{*H9B(kWyxa<eiVW>KBwA6ZA0U~;Mdm{_r+dV}b{Ii(3)QSt
zc?=c_4jZ`YgG;74i5xS)fv~oJ=0Mpq0p|g*Rp$=F(NPfUI?x#*RkbZiAPG5J3BbY)
zxioNm8A6E$T6+UI7XiMR33@v)ILLe%fq_#r*v>{Jp9(EK2z%L=@(NB5_~dbZCL}7V
zwc%0$eNi`TVx$KHQRe3%i(o&ot9R1zLrr#pp<n^r0b@C>VM#6qNFxm<MaVHGz2K4;
zdXa+6&k^?UU;q9$SNc%e8W!L!;2Xfj{mF#~xJ`MyI>u%o3c0z5s?bX1=`PtDAxC<_
z-6P6NFc2lfLq|05AdnQ5m2vcuE4j*TH35^ruaRmLAUnJrFGUTnw)5zC@kb8%y<85C
z?_VdA7(q{W>eAQe=Km#|s0zyo2d#+c=-h{+IY9Tw4UM60zH-N-+`j~4I${wA!yg<4
zWm56TUJBY6a5@=PH4A~wsNBJ?NteH*Sy_m8IDcpQ0OJ7El0qgD`uYz{`?AuzxTP=I
zO-@e24qJMx5^!M@x?`K2g_kIR?7!M`KjmPVCLS~3XsxDDkjwmkUVE`1<j5?O!fNnA
zJ3b-2b4T^7eZM_85K+*RhHq?a*re&&e|`(A(wiY?x&Q^lw2aceCw+g`%oH0=kn!^q
zOZfDtzdK?1_{%?6{vU;{SEaG#LbL532?KT(s&PyX)^aN~dPHx~&Z+WbvYSHY8Pi?E
zJS@LbZFHW$zAdrQ@cE1~Pn>|IjQVdAz_v*NrMa=UmkKrb1nhdVbLa3+<U}?|t;lIw
zz*}Z~#HN?}D^L)lY=>Zk1jYgu5@1@y;8+B32Yhg`EMv*mtcnG@Bod7X#)23oV$!_`
zn`+s)1UXa$?t!vBn(0l&T6rLFUkLyXk=6(9)ftYy5N{WN0UL5u(A%r{#^AC;G*J-i
zb3hRCtQX50cMmXPo$aIv<ZJ)90KMH{;VXRD<ylrK&{t4GzXK;gC?2SX+8S8g_AE>J
z+N*adiR>c!Mv=C~Z0$`G^omQ@)IYrRG<BqEoQ&V1Vf0FOM$HjvZR9`6!JO7+&oP&J
z$eL08jGV9(U|km+1cgG(lHi{QH>chDSOO$;Fy7{=W`lF}yJ<f?+@jr{TUnbSfLr9u
zEXY_eBPjrXuKUTR=%|2vhK8mlDV%Mj4YQ>+R`6*<WBjX<q6xSaA^~^T{6hF1h369T
zV6_K(FPxGW0;2Id92OEVtohYmV>!+bgO(Q;J>ZC)dd8j4)NFOlbZ@hMvfKzf%iqcH
zZRy1P`rGrp>&EU+`OaXjUXjVIuM@Nn3)F4rwcD`W0^K87Fl#$8ZRI>91QoB<NT)ZF
z`l~ys7Mm^NEJ)46NdWNV9zJ|{kB$x-3@d;W7lw*xAg~c)k)~t>at&M+CJ_5D)r5n5
zpqCYcF77f;@=!{0vJ9Nz0q81!2hNj&LaMpM1!o7*vz<ex?5yA;xVE&nGWZPUFpwd6
zb6<dixpUbyG=MR9W@0E>am%yhc9>RIuBYk$>g+qfxm^GMKS-Sjr9xyx_Q;5cd>WJ@
ziZU{@N603#Wn_~*Gm?-QK6XjS&feLPy|@4S);Ztr`F+Rr|6RZ9I$f6|$N4<>^E~(c
ze!tec__&S|%iAYHTw(4&&+~5BKYV?%jqm~A1+Jc<-W98h*R?6$N||$oz0iJdP<>*|
zH_$0GP(0B4MdeUDG#qf)jzE^x<qG-&EjWpgV+k3~0Yn0s8rZcM`1ylCL<F!m4u?w)
zRfWbM5^V4)z=G3su6+9{Jof-iu>rW60*5sCgA1;2vdL3MFGUtX$@YH>9Jua7HPjde
zK3oXrf*>y_cV7Xn83Dxk;(R4p@e-8V<by@9+K`tRyR`xhNa2t>%^%<>g1QJ+FBf79
ztXu@X2t*_}WgYSU$}dd5#Qf%{A@{|A5ipq(_0)6IDsQ?!oTYd*#Q%7v>9erB?H&DB
zS;cWQ9y#8-m!cKU4<8C=EJ;hfz(`2};5UGFPajeNE}is+s{!D??h=Pw$M<<zS<R&t
zePfnGmLbl}+ECYnpSmd)YKyyf?kqdfLamVOd`v_E9z~Gxofff7Qr?6U2t4{+E77p0
z6OCnhT9Hm6kpRSlU3-3zL`qf`Il(~b$Olj+I<s^1>L6ST<?799epU!u<^h8kL>PdP
zqE)_*4(lQYCpc7`Wm-@!rmNwCpp;a6#{?TT;@d&x3FT8Qv;h#IXfXfd6ErR%4jN`=
zJhCadUDw1shWHse30aAG9aMR5Jkh~%)zfyJba6exBuEC?3+y%SwfgQ(l-D0@m{`b&
zr;u1q#IC*x-r72!Dc5~LjRF^r!kD1a0-4yA>({@kEN!i?H$xkRv7H-GpDhvwe0gnY
zdU~3?Ne;^+PYVp{!omVHUiC5KmN=+JA-rapZ6zA=_D<B1fhR7Ae@W?!h+O$WNPmJ5
zlPYbC8^MSh%@%{=T?lMJDCZEuZOADo4qi4iG>kZAc>Vf8h+rgzP@FT+{2kJQ>K6`a
zrFTr=v#B1l<TGeIy>{t40?FZcg6{dqY2mE4rMI#%5QC+_aW71bO6^e2LphewJ@F?L
zX4T(w>A>M{6;1!4!6;nxpbk;~P*1RzN!oFR7qrp^cerSuvDLGuub+57tI9S^K|ygH
zOiPH~ooI|PXqbB}jPMfJ!XU*45)xkAYERqpxOJNv?h*jnZ=}O`Al=Uw(!E|nWrr$n
zj9KxF5K5GStOz`mr}z_!9aQ8%q?8kA43KN402<noUP}#Cmbspf-rx<&h6V>=qfFLN
z?n8CA8vU)S*zO$Ma`^2HF~Cdx)L^yv-~uHZx(|;s4qld(mCa@!61W{LoZWSw4PbFF
z)hWI+OHIsqmFXeiH?BQOp`D{z(fN$M-cKaQFtRPgnw*Hl=b8xB29J!Qz*}C&h$}IH
z&JVh6m*^91@&^x<=MqcDp0x}-OhQuiNT@aG=JB<kQDI6g)v};m#6c=b#07IYKtEx3
z<Arz>fVre(WONsXt7eVba$xH(ii1bx@#5&s_@Bex0{(S`-}HGR71BS6-YVSli%W#`
z%s@DgkcEnrUR(zJ#434)fm_<1U?Hj>$9@B>hZWs&*R9^c^<SEMw*UkJSZPS4`u3Pe
zX1p1wu_1E-G1<`dLUIMHQhwiL<2U{Y8Rib{XmPH;`^~K6e5pdeBCA<%ofr=#52VBR
z4CeIXg|gRcoM^_2<~s0;esKKu6-A$Ar}-uP%)wS-mpum*%yZv5xd5IGCXquUQapLP
zJAH=qYcI)Jq+4)f8P&&eUC~6NsL0P!e&s@s#6NI!%`YwvgSK3M>{|%);?~9n^76+U
zN_zt509qN8=GLbdU36#oo-^Fs_IBpfGc20txCj`hwo5euEO40xz(C$VL}=Iv$T2JU
zqf{81n`(W>ZhPGV4DFv_WhSAMx?{c4G70b7oxJ&vT(Bj4HU5}m)_-v=p#wl-aFf+S
zc*A%jB|N~EmN{E12xb`khCwlZ5cP4GjI4Ty&DoqER~SCB9se<r?HwmCL;TO`ezwQ(
zI;dE`k%%P@Zt%hB96i5!2rJz}`A68(Uq9wdtm|21@X9sS9`5Y)Dn1eaK)}j(VY#YL
z{3!L-Nx~OyfkA1Unv#gt4eRO{f{MFp=yBm%>e9-J9@MV@4nevn{J+oT=S@IJClfO>
zVQ#k!B<;9ud9TA!(gg_+pf;F*%q*0RJKKK}c&Q{#%}Xz8JBgi4iceCy-#TOiAT;z>
zsFQ%88cBJ1a>RUOi=kQ*6fA}<jF9BuR#0H)H6sIs2o(STU>Re8V*{{U7P(j-u;AtM
zBxshqlmMcM#4J#1A_+*kV%pX^97hpCrqn3@OS3uKb<YL)*^oE{g{O6Y_(u5cBNKsw
zb6kR3eG@(_u|-Tx<wvGOLrszR$5njw-1x7A=vUXCIZ=Li9i1RHIi|@q)DupmLKsfH
zNkM3h<5BZ)8FW&XX!L@eZ!ZJu+_1`?a`4bMBrOIi5tT>4HlgweRPZ3SP&GBBNKNCQ
zZHs8o7vKEJgsgHX(FKsib)dB;2l5ZRa_>W!!D78Y6Ut<TJXZB7-cdk@3Lys*Mg2V%
zJ}tJ|rE^;NaYZQ)KEetRM7ccM9|cLc=wO1KqOaQ7*;xf7#VFwr_{y_POwUoHSnlfT
z+ozu7v%|`_8pnlI(8v2gk~%gaq9XO_^Yi3;m=80jvUaAGmc66`p*PIz`UQeG4bX_?
z#wsa_gSs7loqY$*J8!ik|I6?P`;sR*J@b7_P3@?cs0VC80<f--P=~3dm9`kHY*=vc
zH3&-smaXcbby5wk3_D@-M&drIaY1;P0=#QdN(#gPHDL-VtB`C&YU7_-5JLks=I~Y;
zloq7J-*}!#J$^+!MK<PDhW~Ggb7~i<BXk6e@~rJ%``ugMcK~Ywf)xZDU7t8Qh7}65
zzeG(m^d2`MH^gcF)1k3}FYz2ITohnzbpCYEHqzxJxYaJQv$Cdq#lOf_M4%{ngHQ=v
zf6#FQqzum1?8if&b7R5keHjiEh|@i9+VdI=qW3?1GH8}kKlMaR<cN2-d~d_iz>5gW
zhBx8&N<`SL2IT1J=^@APY~ctg*svWswgJdLLVNMG<1EgDOnlu9Py!3!zR4lK3A)9`
zG;X4)Z&|Y`Bt^$x!iRosQWJ3I$SMBO=%uDeL2rnF0@*m}WvDLJfl+sW%+e;&cdd{Y
z39Ci|dN2}b|JgzTPG$hi)M`MFb3a!t^$b+7ERgx4`~Xx0aL~Z%zr>-szbt=6Wy@cK
z<XsVP-DKy%!vy3Cppzn*Is5IIXIlVN10XcZ(KZ5$#2fH@PN?Vax0`{tzSY<3f~4-z
zK*=e#o*eWgkjHquz*wSV)NU7DC?v@&qFg@bEn5_t3C5j0)Za(jsqP16NX3lK%=lKx
zTRDDxz(oNFAg6UBJj#>5^=tTApvt$NYV(3q7B3(=dYW_|<x8&vyQGq%dzkj3HF`mS
zeVHgJ5WwR-<^C|*j-BqyzKlr<1o*2+4WG(5=3F)oyak{u5Jl+~e{0Q!dv8Z1Mu?sQ
z-F{96hIrY74{QSeH7#zkli%97^{_YC4W(3~RoI2`w@pBaBu~P~n5c{!2XVOjy#P?f
zmpJd#t@rM$lx#Zwmf#^J_}eB4L3IDc5BXQG3n?2XbjRAI)HPg$9$h(EoG79GyB(fN
zD#<(T5>s|CvEn|{N<>M0-j1|Cu~63U``&hwP{l!zq8X2h@WU-%A`mP5I&a+e2Y_IR
z{_w|!mebY)GbRJ!s0+{uLsS)_9zskl`F`t`IE;Wj1gCl|6wtYlaSItV74Rg2z{N0i
zLn%|E0i*;}psIw_H*M&9L6GxI6Sj+z-h@e|R11s59`f(v_-@Csv}YC)hp{#KQ7=2l
zY%A%eh%0uR@`D#gK3Ue~5jq?fWbkwQ`bl}mbToM^AZPa|1Y<nCd+9#}hqP#q>IhTk
zaao9sH1p)23x8m?@z8YZXip9+M0O~r+hSZC<}yc<=*0KFdbd49I5Ft<h7kj+1<$+_
zyb>r%OAM4t%!^H+7XpLFek3{ME2;qa_!$X*L9PpL8s)N7KOlUk__QT(4glKM0nV+)
zG!vLh?CWoN$Xd_q)WB!j<{090SWfP_U$;uv4NIIdQ0mB~FCGW4MYd=e<}R7$c;2XO
zYGhcNkJa`UDBl=OqUW|1+p^!#e2>A`=^dh>xOT<%y5L8(q*;hu)&W-J2o-NdjJ{CR
zDPAoMfG+GS(V$I*N}Cr6DwdQaLj2l%fp7t07*fTRL#jfJASR8<-TfiYiTtDV)TAb|
z<Gbe{Jsn8qksvXnlr@}SlihY2P5(4FIRUGS`7aS9v8vGX;O-=|?n(@QFQ{v<`dPha
z>D2q4D5MxW<h&^N_BbrCZzh$U`0W!ot)#!Y;DsjGTK=UAp1{6WXHC&uT3iIVfk6W$
zD6R-H0y!QE9kbJ+;bpMcrHCAC$4O9_#K(l2JP2Be3clvO7HxIMqjtG^Y4G{B!lyC_
zL4af#Rv{xIsRIm!MwC}0?GhQ*&a(Q?PW&K?vshG?&lck=HBY!(-dj^AQ|X|^d8pIr
z`j!0LKCpv+f1lmXXvxDNxKjjZ2@tIw09S%W2^QAv@NpItJ~6*B5eK3okP?BF5cBvy
zCEqov52PBk!FlTkN9um-jM`?<?DVZ584y&%p~7>oQAYr~7^qi}eiF7BgBj_+`n2~N
z)_cO+I`}s?okSf{MVJM`8D`yuAiaWoV(-;m`&p)}$zBDvdTvboSgE?v%H<2%H@35R
zYR0qT?6xGuo(^%G-P65@YdrDD<;sC7oA~w>xnUZLkw%5^IB5+<4#MtJ_K;LFkWZFj
zW>22m*<Ehvt@da?7E*f%n6s+y{-;H*!daI(fdYWpTqW-bzmtYW6(~l;;Ul{wU@max
z)B<LXgGQVRB2pm@1(1bSP(m**EuDho3*f&|e+tr|p}DyxKxk|0wnlP>;8cRzjugEw
zK(}FYa&k~i22^n<`wIcF$<kq#z<-5x=c6SDmSh2zJQo-j5bAY7vx(b@xM;%xh0*K{
zc;-1bLu*vlJIEMv<hZIYFk5Lu3eM7twMSByN3=+{%~^5m-6uJ)=L9{}&CNgKCJwki
z{2d44Gqma4)|kxxBydF3MXPn%V`Hq}YkPBduq!+3r*vj_-7Whq7tCqpjbl**Pu#fa
zWrkA<=*OCi_V9ktQeA%>9F+Krky13O0X2lM|5>Hr)ObK?0w@2aTbEMc7^_+&K=48R
z_Q|_Llbz7P6A=+niCCTmJ`w}?ASnj;MG(?^OGZJNFAf+m$QraZog*JXM+;t$Kw#-1
z^bZfzRXUB)BC|%F-MI$A!NHYKDSLyN3zJ#8N)6XP0{W&KUK>9Fh~EWa4vJ+*bs7e#
z{7{i<=E@=)T?1@;e`btq3Vt!i_`f8DaAOV>kTA_piVNGtVd97GHyGL9U(!~1Pm*bT
z-Ecz6w!Y_%KJA|YA}%wOBMj-P&KOI-nDj+)FEvBU)>p1?ELlv?U_}Yi<S_YeN8QL<
zvsi7@nTvki$8_OeHF^66qC^n#31S`;g8*X|?17B0`S#6Hlf<#twBT;Ji}PDq?xJVg
zZ#?J{kem_Bte|FE1QA>|2-Bb)&R8;&e)+rz%Kv7#Hh3?q;TQoDlL5FmT6lDl$M-<a
z1|p{m@X%F#2SLkzuB8oVUtVP-g6x0>j`GKctT6!0d>*re=g*HE>^hemYwGF{i8`AH
zN{qF8W-0)B*%rT@F#eo+dDfc6`;>Q;k5>XiqxQKCCZqfZeG?C8Z6p1E9jN$Zz8a9f
zLWOb3eyYoy{2Y(M{2cWW_M$04tN__W?D6*R^biiGyM0u4T3K2{zh&-u;zN^kaY@%O
ztC;GA#lsAFdSre(MfgfCE#zJS<NbXT6BiXQS(}6{!;uLYR4%GJhm9QT;z*8((8Gx>
zqcKihn)o72vnrIj&wj-BG>|k<u%PTcfEQ~Z)4$K+7BoKkOZ3MR0XiIo!)Danb&JMz
z=a(iL3jq4sBxs-}`gi{ip)A8He&#ac>t9DfNK60A>rz)roAXq>;ULb|e+~cDoPq&=
z)I90AO_gl>y@e4P(FEp>%c4+kbX{}K?&6|)$rA)9OHC%OzeaNE{k&~hYy8B=QN8Gj
z`vD$yk6V0hJ=7qDjEzszdQ5NEZlC|@K#QBa!KXUY<w_q@CvM@{c1e<Hj3t*aE`SZg
zWlL|jLGnc7#^Y9du6*nVV{>sQG553pCM@^JPdRCm_i4jdATtY<Fc)3tI`>nR#vZ18
zI49f>B9$iiqdD}X@vXd4d87;r%t??0?OZs>9^j2c?hNp$?oXST`CDj9rA4C+%DMml
zg|;LN{*TZWoEyAM4DW7JK>`UCA<uU^x{3KkXWdz8$XxbnJ6!v#>};V5rPI|Z9lHZT
z6JoOZi)N%d>zbC|Np)6pEa3bc_B48KrzOmCj<cJUG%(-pcr8uYS2ixT0;jZ0g1Pu%
z0XEwkmmqj?C0U)LiQ?WSg|fUnydeK>ONAl?eST0EqeKw^46Z<UH|(8%OBRHj`BSh{
zQ~-63G(is2+120FmV}^4gZ3DN96Cso1To-9`+xy(%ET*G3DP)SL19!4oCBl-Faw(h
zhUT2EfRr{`6hqQC0#CrB@WScnR@W^{%5rC1+P5CoVIx1&4L=>#GR~#KGzS0YSf{U_
z=*2k;qDy>6(zhmPnZF+oD|7DHnUhF*a$oozOs+^*&VB~y86kM2AY=;RWYDu%Pkw(1
z#}C>|Apf;4>|6?fzvw5S6h*UCf*@fUh<9)(x#F1YK;7#Eni!<~1n}}Cs5B544f=q)
z<02s2V1hUnFew4~R-w7ONC{mxXOASXV}Znewawh8URmtt5eCiIOsw|`>M_WNUNKi|
z=xiELMm?D^iqQLe3d%XYcd$kO*^=5jn!I0;nb_PlG?NEd?BA29_q&{9WOq`X_oJze
zubO?eAj%OiN#Z$2@4EK3my&Tm-CCY?O#k#mm>rz_;hIpSb1C-puWl34bQx{lhu=Gu
z^lZAGn)oNi=kqp^KhXzR2y{WTw7#I@2VJzmC00n?0wy2eE|k0sT~wOd^QQQ~BL2RG
z1tw+d3q*nZ4l1TTE)F*#Gf?9F6$#yYrWb~uK}wM%kR64P?f~c3sLB#63kyGBd;7MK
z^&bRq+I@wH+MB>v0FZq#Gapc+%D3E69M+i+U<GmO&|JQNnKIE<yU;!!gJc2_xIRsA
z?1#ZJYbD!@brZC0Mr^m>5|Z4ZYIX--*a>MGs0E;O6`A`-dkEMIG+B4Kxc3!qH*IZK
z9YQkx2aB3|2588Au)POvKL8v6{6s5t|L#GIrQk#c=vgH@53;D5Az*Wf=XZN5#SGzM
zIS};;ioPQ!OjXdA$i5a1mNba3h!Wj{kuQ}X`$TU<SDCY@dC9^-1g_YR2Ht2ahHe^t
zblJtlg=*dG4rON+?~~`~_lRP0;miWxq6A<G7+|c#kT?s1CXmRXm_W=88g1mZhWh6!
ztS+S8T?93!$lQz!M#lXi)r9)1Lw`PVv9TZaAB^`>4sn?xA|gQelT%W90F;%jQ?zEX
zG1oMiw~>Y52bEQ~=8LBc+a>n_2gu(6yY@A;EunCmIi+LA{B>_#x}H_8X?h{Ck?6k%
z*`yx4vQ}Piu+wP1<|1d?Q1HL;J6r#zUR7>Jq$Z8(LVvkZvg5fx_5lLLpZ}(egM;`L
zj8-PGHSApP0yj&zH6GDytpDxVeI1>So?S~KVh%wLB-3dHo*j)^8YmUTf?rfHZ+7~3
z^5wnnfK;A>^#tZvA3$f2NIwNC&;lSkYIASst;O6j!+dYjW@<@1Qk~j57&OPB>#EN&
zH`iUWta<#^*veW?uEjdjd5boJACQtTKkg?u7CNWLFOYrB)y$;l<-xM2ryFW@Kb6Hr
z<XIe3A#`YO=Z2&qbMJu{q|W$pg&*uiK4YX}?Z{H=2Dw6;2C`@$R<c6a0~y8OWfUS{
z*~vj#=Awfeiz?kYOtW#s??dOv*dMHWU+}&wmP&u4Y<ilO)w9WL0C`qNklA&g4nX1k
zKy=jtuiJoTtt6U)v3?j&qj;41>-k?q6`83Xhw~g-#m`1kUQ4KDub)^kP3v>eoU!l~
zjTcqgD_Xv8W<)Lbrml*|fXJ8Ns(oRDp7oK`CO64kUf%b4SVl_fxvs|b0^G9_9moDF
z{Si))D~z_Ewd-v)iR9h}TRkS$FUI1<xQA&L)=E@*?0IXZl~EBojMDadbLX5I>Xi9J
zG4SG=xo@gFT`dl3CDdesil*hCP3G%8J={gZ&HG;nmS1@iXtkL4PG-$BKX6&Z?<U;9
ze2LBfv9@`gwlKqpayPheBf4((#Dh51$gc$R^Zs$I<GGYLmyKe~vaj)H&n}_+ZN!1Z
zN}KQ3x?KF9CxsmhiD2FI>kgkOOUiJGiy^xnB%CVXNN3*IxK+0obP{58AWaAV@sBvB
zZh`ER5VS+(7Ozn-MBAJCzUa)!*Fiy7LC63bQPp=4ML@><W8Z=l*@HI36kPFuRKOl_
zl$!b*AoMlgI35%V18?-;qw!17(17ZB6qG8D<Ls(w?V8J+9id8pn(#*v2$hhd@8fr}
zz{tg&Kncvj3x>6FOmuo%ALV|G=;>7rl|t^0D295SaMATrrk?XRyx5+?pV4)6YURmW
z8!1~Z;KC;8?YlA(-|FYBmQIu_t2H|D+M#bC>(c5^MNBwv%QC&F;exBt7xA(qs=7;-
zb*tvs0~n>*n=q1x$jxUZZrh@y|HqPPnZ4&l$~I<JRsrKSHU$3xMRpw&YJjeN;XNV_
zdo#=J$S+9R4u4?}s%eL*)D&cc!62CUCinv!Nw;cQ<IV#84|H2b{U%%@>^fC+jFges
zOpS`-;!c8Q2yAk9xiU&YF)wY*pX$&Z&buHx7nL2!-M83(XjDmyMD|)a!`f1!NCbw8
zDruC$x#g9}OrhuNM!I$|Xk<+os^M7i=1uzjyI9GyI=s1;)J&wVVTc6$pWn@Ed&98f
zxmBoSC}VS8rSZzY^QGwvt8F+z8)zz#KOC?_Z5?lyn%<8lhmgG`0(MQf*aHk8%0)6F
zP&o2KdjY+)1Dq{LKm>gN8qUV0S$5i^^6q|a4;qBiKFAkny!<$-t}ywgK^Zov=40LX
zvGEzX&mn7;nrRv*wq05z2H&ytt?amVjmfYJF{`S)<uEL$k*wUVt~Q;rvyyVz;z?hj
zZdtBUS9ZexqsW@8nnLkMZcWl)MNx%*E}Fv!1TGSAV4$T#2rPV{PJ73xJ}GtB?tc!J
zV^ItTY;*>|^(Vc38w81<(1;)y40y!Z>cPETJNTOs@p9-P#lU+k1`|)v_?IC7f4K`#
zZck8b?5qm_`0+)>Z1TxYT#B)iVpO@c_3(tTFkkRCoxGv!oM(6|-__dc;tL6adCRq-
zq`}40<v!s*#9gl!q^XGHO=HCm<$UjJsw~}knlPDor<fjdGliBi@cYRmI%{mxKryM?
z_G{-8x~B;$9%SB)m>7>^+f>Ur&}MbWO2cm+`%OIdUVB3b5jx(0n;3^O;30HbXr3Jl
zCug0r;G^5`t**pWX+H##8WC4{X8n;!?+XrvdC<<_^8!-`$L_QX60>dt#Xq~G&O-^<
z^JnesENIAbI}HAGN;6WDYo6m&%q4VNe?o%EQY|s0y?1`ubu*M{S5XxHaZ!vCZRo9g
zm03k}IWO}$cc<7_l;tBKqlzgLDeRB2(@k8(95L-8qCaQiHNq%gG+m07x96wD#1-;j
zl&|T`6R{P^3(eK=U`lB0l55*c&qoxI%zc$q$}`xN&pX)hXz5e=P6+0JHLgQ_HFYHR
zU2FmUKxkytz2$uSyG^~BS0_+QhQ<uZM4tBoJu@mJ;4<$G4|_%4nRu&-2liQbz7qiN
zL_Yz_ihriz=)vKo7b^;$2T+4Fg6IwPwP?OwhD05s;F$gF%`(vv5And;RQ+dT%^va{
zu<!O-=H7g^I6J#i-%}PTS{64a&f0+uFYUx0J~5cmXq?~7|IY9LL-ROE#n+KT8*8z?
zKE~f8l!{0u20||xksl?hcvb1BQs^q}yK+D21CeLByC;n%mrw(>(Bi499%8pvV?UmD
zTB7jdc8aWt(qt)|Q(qS<Kap9%CPQ&<u{G%CZF^!&ZUd#Gwe*^mn-2-G2u_E&jQdIJ
z{3p3u(V}W+3(3tXvYPJlRCh;JEA8#)-9n2Ce>G!@Z&jP_O+5SWT?)Ix+j+?40*`PF
zf{Pj&^q`K1cLW7hA!!tf{Lj+DgVodqi61KIQL+{EA)U6JpbW0jTe2^4#=2MMy?JVW
z6!!t&MYt%gv%tB@!?9_PY_}_fKD#dBbH?rB=Ccg1bwpI0a`p^jd$Ob)+q^LP^oR02
z-&TDuIX(0P9opuY==JfFYTdF|4~qM6C!BhDs=B6YoA3rUR_fvPTiq{dc!KngFi-aw
zSXh^mLX6O&KWkvwl)9hu!rIidV$9JIdZ$kV-XGsaz+4gq*T4eP-mgslVxQr&!afUB
zHIm5`GW|ftc~It~(G1|NFzoz53*X@d@4y*IDF?di^@YK0okMe=a0w0y`V4j_kd+u@
z4DJ&Uo|7wEd>dV98F9c-IX;_pjR@&P6E$r*AcYWef^_~0Nna+Hj)94dK**ov;UNXT
zDMv=<@#tS$3Xn_%XI@o;q+TW&@>U@uD}-$8N!J?1?~9FlfZ1V`S8B%*GefN_ku?0f
z6#K6;PpA6Y_lE(cjMB>`#Fsd4zYkEld@216pP;;w&2Oo=jr{JySzGJ?J`*<O`jmi5
zsnMoCE#)Jv6p#RZI>+<nQO9(irzb0a_q5uQY1Qn410RGhg0vD7&%l%nEE<Lh>+#Iz
zH4&i=r((JmOdfpHmGu<qW5DAGDx$m2zFZ!^lZ2AL_;n)wE~W0<w*(MaBIwfRv`CI(
z?{DrSDG0=^pTS!RYX(z+fT{!c_3x(PzYY;9+Hbj%g3!DQu+(kY0Q(4YR$*k^i}3Ks
z{x8aZ@_`z7E4(VDCd+>0@D;~C<%ENdr$4jWA7-jTPYpUIjr&4#>3@<`x+K1DCb70{
zlaA%!Wu~m;eV;0MGI77lnc^}rtS59l*Ae;l_gxTkyFP`&jF|MT`aK;QP@0kTh!OBm
zr;Y`-q$}s*V3g6xBv?+?>oQfkYZ){8n4|n#)D)LqYeLg{EYSPQ?=T~uBEJ2YO2&!b
zD#O3%x_`!2s<VNuh|xw-e(I)t&Z2SBhbJyV+RuJ{l?>XS)P;J_q_*I^?P^O)A~iwz
zxPq{V8GlWUp^90+S`6UcoQ8Qs0NU+)Ti~2FHrDG#!(QQc%HLAd@Fldyh!8?-DpBKq
z5=S0|12E&QV(4uqo0Fp^xx5>%#(fNFZth&BUnEwamTMT_@hxAF{~T-CRGO`G{`jU)
z^8FXs?v0mD#iB1+%lf_!jIf%#TS0QJcK?d?@*kbsPfAYn6<W%R;MU6FT8Na)s44a~
z57<O`+F8qS?XK`eWSkIQ3(j6qkB-tLRQjayZH=|CT6a5A-Mi{S^fMm<U9&B>Jo=Pp
zca^V|!gSYeM61J)J2dhYFC6?1prdC-!X>0Y@Xb$_^=Sj?0~+EB@@tb~c7v53!I|*K
z4hoSkH*(h`C>s0ldv1u@FTC&Q8%%>kw!bCg%w8FDY$FN&P1|uTzKFsP&gau+%U^q9
z92;ydkK&VCtQbIm(}naaw`=73Uv^LVC+@^~mW*@qs$?A${M52*R&qKUF5&rI&kPNk
ztLL^3y>Zr$PPRUNtP;zVwRBC{sj^m1?n_;Z=MTEzcXn0N)RS?ZYF!Z7h-O~^zxy}^
zlm#GguK=;7amghc#&^X5YSfhY6`j5Q-*=Pe3>Q8Z=uS{NC;D^ikPaC0@vVGQ#6Sec
z(c395(!Y%AEXGHVAJdx1+==Xv8Ij~+fGnr=^V7r?3ojN<Z^jcJ+H*dPS*_t&7c7tc
z9DUD^d*k@lpwDo#N5AmAt-xaI8vc(#%X_%7Eum9>a?e5IyD>;|+Npq+wQP$RgO_3c
ziJ~L>$wd}MI*^LI1epdrNUxG_NJxMvd5GHF|5*{yg9!W6ZfCeFH#S<tr|lEy#TAJ;
zSLqBk?ULonZB3~p!d+Vo`%<v11Oe0aV;(z3m?OrgG<HZRan0efdrQk(k<=sp9|gtv
zc537{Ue<G4y~#2p@!`DSTk<+XQia&6j(eb>B|gDwQ7$NET@a&EjNMAy9`yJ2RAYmr
z3M-si$@;K2aQG*{I4&b|1XSQ^UQRDxzC<I0q3;J(6xh=V1q>+4kEuhw4#L06rY3(F
zqXcMV72t1p>i}aH!bWlUk3V4Y84O@<0<v1#RN+rzC{fH+W1e^}9gDf#D|LwExl;`{
z+h~|BYif7$VBE)t0sdPXE%jImExKB+&1NrDji0E+X2Yk~#QS4ri>M{B^X}gG$Jr5z
zy`SV0%Fm;UZ>;gVI2ZRVa^sO;9*><qbOb{rM`uC5Mr3F&-ar}Xefd9E=7x7r)D!a3
z#y**1&o}G;1{O1A?Hck|&&nYSe;#~5xp1VRE<|&0Qwfps7&O)sw9QEH23}N9FsBbe
z@*ij*@OS}gL;xhXV9|33Z_k1Kpr_k07egv4F0Um%tX$j+z*?&4n9cOL%^jCJD!xI}
zBcm#N-R#WT%Bzu=gDJ{QA*ok!l05wG-j&22fw-A#M7auC7l&&^)-P=s6?^+pQtXlI
zd3&-h7A1$9?&{8L`m_luaNJFG;glTh>wY8zYGjrCM-ie8zz%@=1}2@{9hphZPSXVc
zE*MT(5ZeNhKD>AU-ce!!8gQkok=719h3obeKNv;^7F<Xt*aJ4qtYXp{YV<}Jl_Cjt
zNQ`0ol)oDXOojnIt2swc8Ut^aqJHt;co~C;r!oHRp9P+)f;o46c|sSqTdCS_I`@YU
zyBGK-xh_3Pj<sXp?%fgV8?^0NN?`9!f)=udo=AYqaylw|wv)v>#@&w6_w_`|taZwk
z5Z(2z06RViDBI<a8W|M7^XPqkTaJJWLB+I!e}YzH#qd_OyeDjE_dCdrEvTWfSb*~)
zT-B_<2n~`AfPnHc>U@^=E~$QKA!`~MlEDly@p=#Tb#NfYf{_*VS`g47b?X*B5>o-C
z!Vl6Al-&<<iJP*rwsE3RwShlZ7VRe}F(dYpp1Gj*pJlz?!p8?$4_*!Az-H&&xW?bg
z!jExFFG(O~3EiF~&lk59X(D4M;^?H6vYK-g$Z<Y_xl1Y<_T?bne8XzwVBPi6t6Idr
zJZq;kmp|*W6s{3NG#N&BcAE-wqk8Lb`4l$1#7Q{YlCe6hUx_h3SJL{;u4(xl2tUe_
zw8r}}x?6<_FLj#tqQQ65Y&Ged(4E_Q3o*frbi)v6I|`9q2(W|Ea)BT}%Y|1~-heSp
z)t3;A2NPvx(~17#!@y1HL0lWMBcnhZ5aI`ityMqCgHVAS{RZkMtJAdrS0VE=c6*)x
zgXXb=P5L2{dBV@4b_jI(>FHCA-3!4EP6a7HuX?Et(`1b-)ZN@RagFR)BKXli*x4g^
zcqLJ<5F1Epj#(roiQP7)!E}D~*I)bM@+ju92&p+|HNTa^!;RS`vnWqiB13`Da4c=o
zE;o%A?ic1M?HmDv^N7r%UwSoNdrWhg0B;00l|#h35bc#VU(S9qZ}XN&irjWeRVm^}
zXSiWzzz2COZo<lv)P?Nwm-~CLWZ2{14K89LdiQP*WH$&7K0L<$Ha3pLJTp?2eSAg_
zzMh#r)$|xT++2{x0fwLg1G!+nL8g00jJogbav_rr<f8l&C{{5W04^8?<H^mIu4&jk
zFiJE9Xfy#}6@ZiiK?O7q5g3-W%#x~f=gkMA#~b#_ON%{Bhe=0-cZ<8+gW0<dw0))U
zvd&Ar_mSMH>gvTnLeJ+fa?X-VAGOAlz$bX8uB`m3Etxj}b5)(A=V`(bT521^F$u3b
ziD&UA+m6;m^|9D$kZN1ezB_hC_?-N3G2LL3{?@ePHoX&eaMEE`Cq~@)!R_0(i!4Um
zsi~<$oxeU2v`(_w_4c=3Z{?Wrih2_2`etx%^Tyz={x8|V;GntU-D=PASW4^F3Xjdv
zowd<X&7U5Jbc*Je;=kr_+D+oky*D_1i9d~u?R4(hQ%{6i(x#u!ofmCBGdVWamGCM<
zRF@@mH-LKRr2gfibLV$kAI@xVbDE~zW+(4Wb21w(AF2$yFw~1t&Wwz?OY*VPVy;t=
zL-n1&ZH1Yc8DkTZR#?|OSa<lyMp{~0-OSxRJrDXz92C^m)pNL{foGM!7!#_e5_GmV
z-)5#q5GYWoT3z|d%E~9}^KT-fqCRG2#e-jI5=4&h^NERxw^CY-A3f>|nByzWdbhN8
z&}Mf)tJZW?V|Og0X>igpM|++it8My+X#?p1&qk3z5BFA7``-Pqo<`@`*1_^vXTdm9
zx4pJ$pWV#oiXtH^khFA8<DT&84q)_`mnRjN$q&Z}TWELNW@Kbc{rc4kdxkERdt_v!
zHRyu-%l9lp@13V^Z*C5_C@LCKtNU8&EJ(B1Kj!Ci#5HjS*S^EyRA2yn57gcAsck`D
zi7^<rcs{jGn^K<7ayi}U+S%dDV{82Rn<H^o##Rm{m+Q-($Ha?~YWSLT4Ro=T4KLWh
zXkitFgc0E!Q`Jzt+Vc%0{pLs0ZJecuFc=Kp<#z-a%paJmv>41~%+V?M7QWk^Lm13a
zj2^8KJ_eITw=wNWB>R?YizuS_#K2vJP-oMF=%?qCoopo%W8!MLh4s4Qj`4p?8Edj3
zES{Y;Q^#3KmX?-M)6lfSYheRfSPh((gfEK7Ib2KPUlF;g#eDd~y!}&JZwU_pX5Mc-
zo@3JHa8S~-V0P8`;Ijok-u=++I5{;wJ`Nca;<Q*7wbXmY$J114-+VaW81LHsSOS)y
zv%Rx3t*mU|7ysE7Haz&af4sN|-^MVs!M_~9ydlM47%+yf?VtZD3QBq;^vi-7zI|h%
z@PSyP_Q<iRfq=TPvmiWfr&GemT%I4b6sgkY%dG3-&CSh)Q?MlsFvPU-a?OyCkoT$V
za|}9H`)Pt|lfa0O4#jO9?~%+pJPgB8wa)uIVG8`X)l>FotipT(R62$#&scR94Yanl
zCKeSHDfxq7is`1Yk<r~7H>$6P7I0LKiLt@j1^n}FFlSmsWrcK3*hHtoohUmwaCf9G
ze(Z*IIV}EP7xIVAKW<go<+<_R3>$UytqO%eDT$!-SnQky^X@&hu;Gg#SJyb<*WY71
zAumP;@{6Ri?9{&eoQmg~nj4y$Oe`(s7wpyE+%c@0iBN!>EH+W};cIoZ;H675Htc1`
z{j2ZazyHy=nHeqOAlnI|2Wo1|MMnlbB^c9U-*7ViURz)Ce|(t#_}u*cMpTq@$cEG$
UV5YW){|0kYQdZ)_Rjp_L4}K^{ga7~l

diff --git a/ehrapy/tools/cohort_tracking/_cohort_tracker.py b/ehrapy/tools/cohort_tracking/_cohort_tracker.py
index e0a01e78..0f49b2f0 100644
--- a/ehrapy/tools/cohort_tracking/_cohort_tracker.py
+++ b/ehrapy/tools/cohort_tracking/_cohort_tracker.py
@@ -188,12 +188,26 @@ def plot_cohort_barplot(
 
         Examples:
                 >>> import ehrapy as ep
-                >>> adata = ep.dt.hepatitis(columns_obs_only=["class", "sex", "steroid", "fatigue"])
-                >>> cohort_tracker = ep.tl.CohortTracker(adata, categorical=["class", "sex", "steroid", "fatigue"])
-                >>> cohort_tracker(adata, label="Initial Cohort")
-                >>> adata = adata[:50]
-                >>> cohort_tracker(adata, label="Filtered first 50 individuals", operations_done="Filtered to first 50 entries")
-                >>> cohort_tracker.plot_cohort_barplot(subfigure_title=True)
+                >>> adata = ep.dt.diabetes_130(columns_obs_only=["gender", "race", "time_in_hospital_days"])
+                >>> cohort_tracker = ep.tl.CohortTracker(adata, categorical=["gender", "race"])
+                >>> cohort_tracker(adata, "Initial Cohort")
+                >>> adata = adata[:1000]
+                >>> cohort_tracker(adata, "Filtered Cohort")
+                >>> cohort_tracker.plot_cohort_barplot(
+                >>>     subfigure_title=True,
+                >>>     color_palette="tab20",
+                >>>     yticks_labels={
+                >>>         "time_in_hospital_days": "Time in hospital (days)",
+                >>>         "race": "Race",
+                >>>         "gender": "Gender",
+                >>>     },
+                >>>     legend_labels={
+                >>>         "time_in_hospital_days": "Time in hospital (days)",
+                >>>         0.0: "Female",
+                >>>         1.0: "Male",
+                >>>     },
+                >>> )
+
 
             .. image:: /_static/docstring_previews/cohort_tracking.png
         """

From a4c021bb44c0335ffaca66dceaa9de77a45d340d Mon Sep 17 00:00:00 2001
From: Eljas Roellin <65244425+eroell@users.noreply.github.com>
Date: Wed, 13 Mar 2024 14:23:28 +0100
Subject: [PATCH 39/46] Update ehrapy/tools/cohort_tracking/_cohort_tracker.py

Co-authored-by: Lukas Heumos <lukas.heumos@posteo.net>
---
 ehrapy/tools/cohort_tracking/_cohort_tracker.py | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/ehrapy/tools/cohort_tracking/_cohort_tracker.py b/ehrapy/tools/cohort_tracking/_cohort_tracker.py
index 0f49b2f0..cf54e1db 100644
--- a/ehrapy/tools/cohort_tracking/_cohort_tracker.py
+++ b/ehrapy/tools/cohort_tracking/_cohort_tracker.py
@@ -25,7 +25,7 @@ def _check_adata_type(adata) -> None:
 def _check_columns_exist(df, columns) -> None:
     missing_columns = set(columns) - set(df.columns)
     if missing_columns:
-        raise ValueError(f"Columns {list(missing_columns)} not found in dataframe.")
+        raise ValueError(f"Columns {list(missing_columns)} not found in DataFrame.")
 
 
 def _check_no_new_categories(df, categorical, categorical_labels) -> None:

From b75cdbb203df67cdc2977a874cc4bd546785235a Mon Sep 17 00:00:00 2001
From: Eljas Roellin <65244425+eroell@users.noreply.github.com>
Date: Wed, 13 Mar 2024 14:23:37 +0100
Subject: [PATCH 40/46] Update ehrapy/tools/cohort_tracking/_cohort_tracker.py

Co-authored-by: Lukas Heumos <lukas.heumos@posteo.net>
---
 ehrapy/tools/cohort_tracking/_cohort_tracker.py | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/ehrapy/tools/cohort_tracking/_cohort_tracker.py b/ehrapy/tools/cohort_tracking/_cohort_tracker.py
index cf54e1db..fe4052b7 100644
--- a/ehrapy/tools/cohort_tracking/_cohort_tracker.py
+++ b/ehrapy/tools/cohort_tracking/_cohort_tracker.py
@@ -54,7 +54,7 @@ class CohortTracker:
     Tightly interacting with the `tableone` package [1].
 
     Args:
-        adata: Object to track.
+        adata: AnnData object to track.
         columns: Columns to track. If `None`, all columns will be tracked.
         categorical: Columns that contain categorical variables, if None will be inferred from the data.
 

From d5cfa23b2b7a9b79394f46dda24e331c00de3dd1 Mon Sep 17 00:00:00 2001
From: Eljas Roellin <65244425+eroell@users.noreply.github.com>
Date: Wed, 13 Mar 2024 14:23:57 +0100
Subject: [PATCH 41/46] Update ehrapy/tools/cohort_tracking/_cohort_tracker.py

Co-authored-by: Lukas Heumos <lukas.heumos@posteo.net>
---
 ehrapy/tools/cohort_tracking/_cohort_tracker.py | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/ehrapy/tools/cohort_tracking/_cohort_tracker.py b/ehrapy/tools/cohort_tracking/_cohort_tracker.py
index fe4052b7..4236b948 100644
--- a/ehrapy/tools/cohort_tracking/_cohort_tracker.py
+++ b/ehrapy/tools/cohort_tracking/_cohort_tracker.py
@@ -55,7 +55,7 @@ class CohortTracker:
 
     Args:
         adata: AnnData object to track.
-        columns: Columns to track. If `None`, all columns will be tracked.
+        columns: Columns to track. If `None`, all columns will be tracked. Defaults to `None`.
         categorical: Columns that contain categorical variables, if None will be inferred from the data.
 
     References:

From 5261de30dd2f04d341d19c43f909aeeac7cf9338 Mon Sep 17 00:00:00 2001
From: Eljas Roellin <65244425+eroell@users.noreply.github.com>
Date: Wed, 13 Mar 2024 14:24:28 +0100
Subject: [PATCH 42/46] Update ehrapy/tools/cohort_tracking/_cohort_tracker.py

Co-authored-by: Lukas Heumos <lukas.heumos@posteo.net>
---
 ehrapy/tools/cohort_tracking/_cohort_tracker.py | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/ehrapy/tools/cohort_tracking/_cohort_tracker.py b/ehrapy/tools/cohort_tracking/_cohort_tracker.py
index 4236b948..6172b330 100644
--- a/ehrapy/tools/cohort_tracking/_cohort_tracker.py
+++ b/ehrapy/tools/cohort_tracking/_cohort_tracker.py
@@ -56,7 +56,7 @@ class CohortTracker:
     Args:
         adata: AnnData object to track.
         columns: Columns to track. If `None`, all columns will be tracked. Defaults to `None`.
-        categorical: Columns that contain categorical variables, if None will be inferred from the data.
+        categorical: Columns that contain categorical variables, if None will be inferred from the data. Defaults to `None`.
 
     References:
         [1] Tom Pollard, Alistair E.W. Johnson, Jesse D. Raffa, Roger G. Mark; tableone: An open source Python package for producing summary statistics for research papers, Journal of the American Medical Informatics Association, Volume 24, Issue 2, 1 March 2017, Pages 267–271, https://doi.org/10.1093/jamia/ocw117

From 61683f5a4599291cfa04cca8e74c37bc93c5d71a Mon Sep 17 00:00:00 2001
From: Eljas Roellin <65244425+eroell@users.noreply.github.com>
Date: Wed, 13 Mar 2024 14:24:50 +0100
Subject: [PATCH 43/46] Update ehrapy/tools/cohort_tracking/_cohort_tracker.py

Co-authored-by: Lukas Heumos <lukas.heumos@posteo.net>
---
 ehrapy/tools/cohort_tracking/_cohort_tracker.py | 1 -
 1 file changed, 1 deletion(-)

diff --git a/ehrapy/tools/cohort_tracking/_cohort_tracker.py b/ehrapy/tools/cohort_tracking/_cohort_tracker.py
index 6172b330..b892a6f7 100644
--- a/ehrapy/tools/cohort_tracking/_cohort_tracker.py
+++ b/ehrapy/tools/cohort_tracking/_cohort_tracker.py
@@ -108,7 +108,6 @@ def __call__(self, adata: AnnData, label: str = None, operations_done: str = Non
         self._tracked_tables.append(t1)
 
     def _get_cat_data(self, table_one: TableOne, col: str) -> pd.DataFrame:
-        # mypy error if not specifying dict below
         cat_pct: dict = {category: [] for category in self._categorical_categories[col]}
 
         for cat in self._categorical_categories[col]:

From 9663cfb93f247505b8f2f91149fa0be724222c50 Mon Sep 17 00:00:00 2001
From: eroell <eljas.roellin@ikmail.com>
Date: Wed, 13 Mar 2024 14:37:36 +0100
Subject: [PATCH 44/46] remove old comments, better variable names, simplify
 adata check

---
 cohort_tracking.ipynb                         | 360 ------------------
 .../tools/cohort_tracking/_cohort_tracker.py  |  21 +-
 2 files changed, 11 insertions(+), 370 deletions(-)
 delete mode 100644 cohort_tracking.ipynb

diff --git a/cohort_tracking.ipynb b/cohort_tracking.ipynb
deleted file mode 100644
index 84582c20..00000000
--- a/cohort_tracking.ipynb
+++ /dev/null
@@ -1,360 +0,0 @@
-{
- "cells": [
-  {
-   "cell_type": "markdown",
-   "metadata": {},
-   "source": [
-    "## Cohort Tracking with ehrapy\n",
-    "Important for many reasons"
-   ]
-  },
-  {
-   "cell_type": "code",
-   "execution_count": 1,
-   "metadata": {},
-   "outputs": [
-    {
-     "name": "stderr",
-     "output_type": "stream",
-     "text": [
-      "/Users/eljasroellin/Documents/ehrapy_clean/ehrapy_venv_march_II/lib/python3.11/site-packages/tqdm/auto.py:21: TqdmWarning: IProgress not found. Please update jupyter and ipywidgets. See https://ipywidgets.readthedocs.io/en/stable/user_install.html\n",
-      "  from .autonotebook import tqdm as notebook_tqdm\n"
-     ]
-    }
-   ],
-   "source": [
-    "import ehrapy as ep\n",
-    "from tableone import TableOne"
-   ]
-  },
-  {
-   "cell_type": "markdown",
-   "metadata": {},
-   "source": [
-    "### Tableone\n",
-    "nice package"
-   ]
-  },
-  {
-   "cell_type": "code",
-   "execution_count": 2,
-   "metadata": {},
-   "outputs": [
-    {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      "\u001b[1;35m2024-03-13 11:17:19,282\u001b[0m - \u001b[1;34mroot\u001b[0m \u001b[1;37mINFO - Transformed passed DataFrame into an AnnData object with n_obs x n_vars = `101766` x `48`.\u001b[0m\n"
-     ]
-    },
-    {
-     "data": {
-      "text/html": [
-       "<div>\n",
-       "<style scoped>\n",
-       "    .dataframe tbody tr th:only-of-type {\n",
-       "        vertical-align: middle;\n",
-       "    }\n",
-       "\n",
-       "    .dataframe tbody tr th {\n",
-       "        vertical-align: top;\n",
-       "    }\n",
-       "\n",
-       "    .dataframe thead th {\n",
-       "        text-align: right;\n",
-       "    }\n",
-       "</style>\n",
-       "<table border=\"1\" class=\"dataframe\">\n",
-       "  <thead>\n",
-       "    <tr style=\"text-align: right;\">\n",
-       "      <th></th>\n",
-       "      <th></th>\n",
-       "      <th>Missing</th>\n",
-       "      <th>Overall</th>\n",
-       "    </tr>\n",
-       "  </thead>\n",
-       "  <tbody>\n",
-       "    <tr>\n",
-       "      <th>n</th>\n",
-       "      <th></th>\n",
-       "      <td></td>\n",
-       "      <td>101766</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th rowspan=\"2\" valign=\"top\">gender, n (%)</th>\n",
-       "      <th>0.0</th>\n",
-       "      <td>3</td>\n",
-       "      <td>54708 (53.8)</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>1.0</th>\n",
-       "      <td></td>\n",
-       "      <td>47055 (46.2)</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th rowspan=\"5\" valign=\"top\">race, n (%)</th>\n",
-       "      <th>AfricanAmerican</th>\n",
-       "      <td>2273</td>\n",
-       "      <td>19210 (19.3)</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>Asian</th>\n",
-       "      <td></td>\n",
-       "      <td>641 (0.6)</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>Caucasian</th>\n",
-       "      <td></td>\n",
-       "      <td>76099 (76.5)</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>Hispanic</th>\n",
-       "      <td></td>\n",
-       "      <td>2037 (2.0)</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>Other</th>\n",
-       "      <td></td>\n",
-       "      <td>1506 (1.5)</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th rowspan=\"14\" valign=\"top\">time_in_hospital_days, n (%)</th>\n",
-       "      <th>1</th>\n",
-       "      <td>0</td>\n",
-       "      <td>14208 (14.0)</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>10</th>\n",
-       "      <td></td>\n",
-       "      <td>2342 (2.3)</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>11</th>\n",
-       "      <td></td>\n",
-       "      <td>1855 (1.8)</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>12</th>\n",
-       "      <td></td>\n",
-       "      <td>1448 (1.4)</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>13</th>\n",
-       "      <td></td>\n",
-       "      <td>1210 (1.2)</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>14</th>\n",
-       "      <td></td>\n",
-       "      <td>1042 (1.0)</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>2</th>\n",
-       "      <td></td>\n",
-       "      <td>17224 (16.9)</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>3</th>\n",
-       "      <td></td>\n",
-       "      <td>17756 (17.4)</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>4</th>\n",
-       "      <td></td>\n",
-       "      <td>13924 (13.7)</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>5</th>\n",
-       "      <td></td>\n",
-       "      <td>9966 (9.8)</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>6</th>\n",
-       "      <td></td>\n",
-       "      <td>7539 (7.4)</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>7</th>\n",
-       "      <td></td>\n",
-       "      <td>5859 (5.8)</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>8</th>\n",
-       "      <td></td>\n",
-       "      <td>4391 (4.3)</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>9</th>\n",
-       "      <td></td>\n",
-       "      <td>3002 (2.9)</td>\n",
-       "    </tr>\n",
-       "  </tbody>\n",
-       "</table>\n",
-       "</div><br />"
-      ],
-      "text/plain": [
-       "                                             Missing       Overall\n",
-       "n                                                           101766\n",
-       "gender, n (%)                0.0                   3  54708 (53.8)\n",
-       "                             1.0                      47055 (46.2)\n",
-       "race, n (%)                  AfricanAmerican    2273  19210 (19.3)\n",
-       "                             Asian                       641 (0.6)\n",
-       "                             Caucasian                76099 (76.5)\n",
-       "                             Hispanic                   2037 (2.0)\n",
-       "                             Other                      1506 (1.5)\n",
-       "time_in_hospital_days, n (%) 1                     0  14208 (14.0)\n",
-       "                             10                         2342 (2.3)\n",
-       "                             11                         1855 (1.8)\n",
-       "                             12                         1448 (1.4)\n",
-       "                             13                         1210 (1.2)\n",
-       "                             14                         1042 (1.0)\n",
-       "                             2                        17224 (16.9)\n",
-       "                             3                        17756 (17.4)\n",
-       "                             4                        13924 (13.7)\n",
-       "                             5                          9966 (9.8)\n",
-       "                             6                          7539 (7.4)\n",
-       "                             7                          5859 (5.8)\n",
-       "                             8                          4391 (4.3)\n",
-       "                             9                          3002 (2.9)"
-      ]
-     },
-     "execution_count": 2,
-     "metadata": {},
-     "output_type": "execute_result"
-    }
-   ],
-   "source": [
-    "adata = ep.dt.diabetes_130(columns_obs_only=[\"gender\", \"race\", \"time_in_hospital_days\"])\n",
-    "TableOne(adata.obs)"
-   ]
-  },
-  {
-   "cell_type": "markdown",
-   "metadata": {},
-   "source": [
-    "### CohortTracker\n",
-    "A visualization aid automated in ehrapy: summarizing tableone information graphically.\n",
-    "Especially useful for cohort processing, as the overview component becomes even more important there"
-   ]
-  },
-  {
-   "cell_type": "code",
-   "execution_count": 3,
-   "metadata": {},
-   "outputs": [
-    {
-     "data": {
-      "image/png": "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",
-      "text/plain": [
-       "<Figure size 640x480 with 2 Axes>"
-      ]
-     },
-     "metadata": {},
-     "output_type": "display_data"
-    },
-    {
-     "data": {
-      "image/png": "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",
-      "text/plain": [
-       "<Figure size 640x480 with 1 Axes>"
-      ]
-     },
-     "metadata": {},
-     "output_type": "display_data"
-    }
-   ],
-   "source": [
-    "# instantiate the cohort tracker\n",
-    "pop_track = ep.tl.CohortTracker(adata, categorical=[\"gender\", \"race\"])\n",
-    "\n",
-    "# track the initial state of the dataset\n",
-    "pop_track(adata, label=\"Initial cohort\")\n",
-    "\n",
-    "# do a filtering step\n",
-    "adata = adata[:1000]\n",
-    "\n",
-    "# track the filtered dataset\n",
-    "pop_track(adata, label=\"Cohort 1\", operations_done=\"filtered to first 1000 entries\")\n",
-    "\n",
-    "# plot the change of the cohort\n",
-    "pop_track.plot_cohort_barplot()\n",
-    "\n",
-    "# plot a flowchart\n",
-    "pop_track.plot_flowchart()"
-   ]
-  },
-  {
-   "cell_type": "markdown",
-   "metadata": {},
-   "source": [
-    "Some nice plotting options are available"
-   ]
-  },
-  {
-   "cell_type": "code",
-   "execution_count": 11,
-   "metadata": {},
-   "outputs": [
-    {
-     "data": {
-      "image/png": "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",
-      "text/plain": [
-       "<Figure size 640x480 with 2 Axes>"
-      ]
-     },
-     "metadata": {},
-     "output_type": "display_data"
-    },
-    {
-     "data": {
-      "image/png": "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",
-      "text/plain": [
-       "<Figure size 640x480 with 1 Axes>"
-      ]
-     },
-     "metadata": {},
-     "output_type": "display_data"
-    }
-   ],
-   "source": [
-    "pop_track.plot_cohort_barplot(\n",
-    "    subfigure_title=True,\n",
-    "    yticks_labels={\"time_in_hospital_days\": \"Time in hospital (days)\", \"race\": \"Race\", \"gender\": \"Gender\"},\n",
-    "    color_palette=\"tab20\",\n",
-    "    legend_labels={\n",
-    "        \"time_in_hospital_days\": \"Time in hospital (days)\",\n",
-    "        \"AfricanAmerican\": \"African American\",\n",
-    "        0.0: \"Female\",\n",
-    "        1.0: \"Male\",\n",
-    "    },\n",
-    "    legend_kwargs={\"title\": \"Variables\", \"bbox_to_anchor\": (1, 1)},\n",
-    ")\n",
-    "\n",
-    "pop_track.plot_flowchart(\n",
-    "    title=\"Cohort flowchart\", arrow_size=0.75, bbox_kwargs={\"fc\": \"lightgreen\"}, arrowprops_kwargs={\"color\": \"black\"}\n",
-    ")"
-   ]
-  }
- ],
- "metadata": {
-  "kernelspec": {
-   "display_name": "ehrapy_venv_feb",
-   "language": "python",
-   "name": "python3"
-  },
-  "language_info": {
-   "codemirror_mode": {
-    "name": "ipython",
-    "version": 3
-   },
-   "file_extension": ".py",
-   "mimetype": "text/x-python",
-   "name": "python",
-   "nbconvert_exporter": "python",
-   "pygments_lexer": "ipython3",
-   "version": "3.11.8"
-  }
- },
- "nbformat": 4,
- "nbformat_minor": 2
-}
diff --git a/ehrapy/tools/cohort_tracking/_cohort_tracker.py b/ehrapy/tools/cohort_tracking/_cohort_tracker.py
index 0f49b2f0..695b9362 100644
--- a/ehrapy/tools/cohort_tracking/_cohort_tracker.py
+++ b/ehrapy/tools/cohort_tracking/_cohort_tracker.py
@@ -17,20 +17,16 @@
     from matplotlib.figure import Figure
 
 
-def _check_adata_type(adata) -> None:
-    if not isinstance(adata, AnnData):
-        raise ValueError("adata must be an AnnData.")
-
-
 def _check_columns_exist(df, columns) -> None:
     missing_columns = set(columns) - set(df.columns)
     if missing_columns:
         raise ValueError(f"Columns {list(missing_columns)} not found in dataframe.")
 
 
-def _check_no_new_categories(df, categorical, categorical_labels) -> None:
+def _check_no_new_categories(df: pd.DataFrame, categorical: pd.DataFrame, categorical_labels: dict) -> None:
+    """Check if new categories have been added to the categorical columns: this would break the plotting logic."""
     for col in categorical:
-        categories_present = df[col].astype("category").cat.categories  # unique()  # TODO: use unique()?
+        categories_present = df[col].astype("category").cat.categories
         categories_expected = categorical_labels[col]
         diff = set(categories_present) - set(categories_expected)
         if diff:
@@ -63,7 +59,8 @@ class CohortTracker:
     """
 
     def __init__(self, adata: AnnData, columns: Sequence = None, categorical: Sequence = None) -> None:
-        _check_adata_type(adata)
+        if not isinstance(adata, AnnData):
+            raise ValueError("adata must be an AnnData.")
 
         self.columns = columns if columns is not None else list(adata.obs.columns)
 
@@ -90,7 +87,9 @@ def __init__(self, adata: AnnData, columns: Sequence = None, categorical: Sequen
         self._tracked_tables: list = []
 
     def __call__(self, adata: AnnData, label: str = None, operations_done: str = None, **tableone_kwargs: dict) -> None:
-        _check_adata_type(adata)
+        if not isinstance(adata, AnnData):
+            raise ValueError("adata must be an AnnData.")
+
         _check_columns_exist(adata.obs, self.columns)
         _check_no_new_categories(adata.obs, self.categorical, self._categorical_categories)
 
@@ -131,7 +130,9 @@ def _check_legend_labels(self, legend_handels: dict) -> None:
         if not isinstance(legend_handels, dict):
             raise ValueError("legend_labels must be a dictionary.")
 
-        values = [item for sublist in self._categorical_categories.values() for item in sublist]
+        values = [
+            category for column_categories in self._categorical_categories.values() for category in column_categories
+        ]
 
         missing_keys = [key for key in legend_handels if key not in values and key not in self.columns]
 

From d638a888682ea0e7d4157559879f387e13ba6f63 Mon Sep 17 00:00:00 2001
From: eroell <eljas.roellin@ikmail.com>
Date: Wed, 13 Mar 2024 14:45:55 +0100
Subject: [PATCH 45/46] fix two doc typos

---
 ehrapy/tools/cohort_tracking/_cohort_tracker.py | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/ehrapy/tools/cohort_tracking/_cohort_tracker.py b/ehrapy/tools/cohort_tracking/_cohort_tracker.py
index bd3e4d8b..c4b21d36 100644
--- a/ehrapy/tools/cohort_tracking/_cohort_tracker.py
+++ b/ehrapy/tools/cohort_tracking/_cohort_tracker.py
@@ -178,9 +178,9 @@ def plot_cohort_barplot(
             color_palette: The color palette to use for the plot. Default is "colorblind".
             yticks_labels: Dictionary to rename the axis labels. If `None`, the original labels will be used. The keys should be the column names.
             legend_labels: Dictionary to rename the legend labels. If `None`, the original labels will be used. For categoricals, the keys should be the categories. For numericals, the key should be the column name.
-            show: If `True`, the plot will be shown. If `False`, returns plotting handels are returned.
+            show: If `True`, the plot will be shown. If `False`, plotting handels are returned.
             ax: If `None`, a new figure and axes will be created. If an axes object is provided, the plot will be added to it.
-            subplot_kwargs: Additional keyword arguments for the subplots.
+            subplots_kwargs: Additional keyword arguments for the subplots.
             legend_kwargs: Additional keyword arguments for the legend.
 
         Returns:

From 3c92f26ca69d84538cce49d741b70d87c2f13f26 Mon Sep 17 00:00:00 2001
From: eroell <eljas.roellin@ikmail.com>
Date: Wed, 13 Mar 2024 15:10:59 +0100
Subject: [PATCH 46/46] identical Returns field

---
 ehrapy/tools/cohort_tracking/_cohort_tracker.py | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/ehrapy/tools/cohort_tracking/_cohort_tracker.py b/ehrapy/tools/cohort_tracking/_cohort_tracker.py
index c4b21d36..900e38d2 100644
--- a/ehrapy/tools/cohort_tracking/_cohort_tracker.py
+++ b/ehrapy/tools/cohort_tracking/_cohort_tracker.py
@@ -378,7 +378,7 @@ def plot_flowchart(
             arrowprops_kwargs: Additional keyword arguments for the arrows.
 
         Returns:
-            If `show=True`, returns `None`. Else, if no ax is passed, returns a tuple  (:class:`~matplotlib.figure.Figure`, :class:`list`(:class:`~matplotlib.axes.Axes`), else a :class:`list`(:class:`~matplotlib.axes.Axes`).
+            If `show=True`, returns `None`. Else, if no ax is passed, returns a tuple  (:class:`~matplotlib.figure.Figure`, :class:`~list`(:class:`~matplotlib.axes.Axes`), else a :class:`~list`(:class:`~matplotlib.axes.Axes`).
 
         Examples:
                 >>> import ehrapy as ep