{
 "cells": [
  {
   "cell_type": "raw",
   "id": "4a51e514-6a29-4607-a2c3-620d9db8b337",
   "metadata": {},
   "source": [
    "---\n",
    "title: \"Data: Main Site\"\n",
    "description: \"Here I talk about the data collected and its implications\"\n",
    "date: \"2026-06-20\"\n",
    "date-modified: today\n",
    "author: \"Zak Toews\"\n",
    "categories: [project, data-mining, data]\n",
    "jupyter: python3\n",
    "execute:\n",
    "  enabled: true\n",
    "  freeze: false\n",
    "format:\n",
    "  html:\n",
    "    other-links:\n",
    "      - text: \"Download notebook\"\n",
    "        icon: downloads\n",
    "        href: MainSiteData.ipynb\n",
    "    resources:\n",
    "      - MainSiteData.ipynb\n",
    "    code-fold: true\n",
    "    code-summary: \"Show code\"\n",
    "---"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "dfa48881-4c48-4b29-b0f0-b728ce37ab6a",
   "metadata": {},
   "source": [
    "<center>\n",
    "    \n",
    "# **Data: Main Site**\n",
    "\n",
    "### By Zak Toews with assistance of AI\n",
    "### 4th Year Software Engineering Student and Hobbyist\n",
    "\n",
    "</center>\n"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "00695d51-ef08-4ef3-a6d3-6e425673c884",
   "metadata": {},
   "source": [
    "# Table of Contents\n",
    "1. [Data Collected](#data-collected)\n",
    "2. [Notes about Data](#notes-about-data)\n",
    "3. [Tools and Techniques](#tools-and-techniques)\n",
    "4. [Importing the Data](#importing-the-data)\n",
    "5. [Visualizing the Data](#visualizing-the-data)\n",
    "6. [Classifying Contact Interest](#classifying-contact-interest)\n",
    "7. [Clustering Sessions into Behavioral Groups](#clustering-sessions-into-behavioral-groups)\n",
    "8. [Association Analysis: What Site Behaviors Go Together](#association-analysis-what-site-behaviors-go-together)\n"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "50d52861-b273-4cbe-89be-915208e00008",
   "metadata": {},
   "source": [
    "## Data Collected\n",
    "\n",
    "All data is anonymous. No personal data is recorded.\n",
    "\n",
    "This is subject to change as the analytics evolve.\n",
    "\n",
    "### Analytics Event Table\n",
    "\n",
    "**id** (string) - Event specific ID\n",
    "\n",
    "**event** (enum)- Different types of events and their corresponding properties  \n",
    "    - **page_load** - Triggered on page load  \n",
    "    - **section_viewed** (enum)- Sections viewed include \"about\", \"experience\", \"research\", \"contact\"  \n",
    "    - **scroll_depth** (enum)- As percentage of the total length of the main page - 25, 50, 75, 100  \n",
    "    - **link_clicked** (enum)- This event is fired anytime a user clicks a link - \"planner\", \"notes\", \"old-site\", \"email\", \"ayuda\", \"youtube\"  \n",
    "    - **nav_clicked** (enum) - This event is fired when a user navgiates to sections using the nav links at the top of the site - \"about\", \"experience\", \"research\", \"contact\"\n",
    "\n",
    "**sessionId** (string) - Randomly generated ID for browser session\n",
    "\n",
    "**createdAt** (time) - A precise time stamp\n",
    "\n",
    "**service** (enum) - We will only be concerned with \"main-site\" for now\n",
    "    \n",
    "### Feedback Table\n",
    "\n",
    "**id** (string) - Feedback specific ID\n",
    "\n",
    "**category** (enum) - The feedback category selected by the user - \"Suggestion\", \"Bug Report\", \"Other\"\n",
    "\n",
    "**message** (string) - Max of 500 characters to explain feedback \n",
    "\n",
    "**contact** (string) - Just a string, any information could be put into here\n",
    "\n",
    "**service** (string) - Currently always \"main-site\"\n",
    "\n",
    "**createdAt** (time) - A precise time stamp\n"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "3ed4d879-6ae6-4335-8245-a29a00613fc4",
   "metadata": {},
   "source": [
    "## Notes about Data\n",
    "\n",
    "The `AnalyticsEvent` table has two different `id`/`createdAt` formats mixed together:\n",
    "\n",
    "- **main-site** events (raw SQL insert) always used `gen_random_uuid()` for `id` and\n",
    "  Postgres's `NOW()` for `createdAt` — a UUID and microsecond-precision timestamp.\n",
    "- **trip-planner** events (via Prisma) used Prisma's client-side `cuid()` and `Date.now()`\n",
    "  defaults instead of the database — a cuid and only millisecond-precision timestamp.\n",
    "\n",
    "This got fixed at the schema level (`trip-planner/prisma/schema.prisma` now uses\n",
    "`dbgenerated(\"gen_random_uuid()\")` / `dbgenerated(\"now()\")` for both fields, so *new*\n",
    "trip-planner rows match main-site's format exactly). But a schema default only changes\n",
    "future inserts — rows written before the fix keep their original cuid/millisecond format\n",
    "forever. So any notebook reading this table needs to tolerate both formats existing\n",
    "side by side in the same column.\n"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "747cbd00-b89a-4bc3-b3ff-12bf3a0a15fc",
   "metadata": {},
   "source": [
    "## Tools and Techniques\n",
    "\n",
    "Everything below builds on a small set of Python libraries and a handful of core data-mining concepts. This section is a quick reference for what each one does and why it's useful - skip ahead if you're already familiar, or come back here whenever a technique further down feels unfamiliar.\n",
    "\n",
    "### Core libraries\n",
    "\n",
    "| Library | Role |\n",
    "|---|---|\n",
    "| **pandas** | Loads, filters, groups, and reshapes tabular data (`DataFrame`s). Nearly every cell below starts by transforming raw event rows into a per-session table. |\n",
    "| **matplotlib** | Draws every chart in this notebook - box plots, scatter matrices, dendrograms, ROC curves. |\n",
    "| **scikit-learn** | Supplies the classifiers, clustering algorithms, evaluation metrics, and hyperparameter search tools used throughout. |\n",
    "| **scipy** | Used once, for hierarchical clustering's dendrogram. |\n",
    "\n",
    "### Why turn raw events into a \"session table\"?\n",
    "\n",
    "Data-mining algorithms expect **one row per observation, one column per feature** - a shape pandas calls \"tidy.\" The raw `AnalyticsEvent` table is one row per *event* (a single session might have 20 rows). The `session_metrics` table built in [Classifying Contact Interest](#classifying-contact-interest) collapses that down to one row per *session*, with engineered columns like `total_actions` and `max_scroll_depth`. That reshaping step is what makes every classifier, clustering algorithm, and association-rule search below possible.\n",
    "\n",
    "### Supervised learning: classifiers\n",
    "\n",
    "These all answer the same question - \"given a session's features, predict whether it shows `contact_interest`\" - using different strategies:\n",
    "\n",
    "| Classifier | How it decides | Why it's here |\n",
    "|---|---|---|\n",
    "| **Decision Tree** | Asks a sequence of yes/no questions about one feature at a time. | Easy to read and explain - the whole \"reasoning\" can be printed as text or drawn as a diagram. |\n",
    "| **K-Nearest Neighbors (KNN)** | Looks at the *k* most similar past sessions and votes. | No training step, and very sensitive to feature scale - a good teaching example for *why* scaling matters. |\n",
    "| **Random Forest** | Averages many decision trees, each trained on a random subset of the data. | Usually more accurate and more resistant to overfitting than a single tree. |\n",
    "| **Bagging** | The same \"average many models\" idea as Random Forest, but with a base model you choose yourself. | Shows that *ensembling* is a separate idea from *decision trees* specifically - it works with any base classifier. |\n",
    "| **Gaussian Naive Bayes** | Assumes each feature is normally distributed within a class, and combines their probabilities. | Extremely fast with no hyperparameters to tune - a useful \"does anything more complex actually help?\" comparison point. |\n",
    "| **Dummy (baseline)** | Always predicts the majority class, ignoring the features entirely. | The floor every real classifier has to beat. If a \"real\" classifier can't outperform this, it isn't learning anything from the data. |\n",
    "\n",
    "### Evaluating a classifier\n",
    "\n",
    "A single accuracy number can be misleading, especially once one class is rarer than the other:\n",
    "\n",
    "| Metric / technique | What it tells you |\n",
    "|---|---|\n",
    "| **Accuracy** | Percent of predictions that were correct. Misleading on unbalanced data - a classifier that always guesses the majority class can still score high. |\n",
    "| **Confusion matrix** | Breaks predictions into true/false positives and negatives, so you can see *what kind* of mistake a classifier makes. |\n",
    "| **Precision** | Of the sessions predicted `contact_interest`, how many actually were? High precision means few false alarms. |\n",
    "| **Recall** | Of the sessions that actually were `contact_interest`, how many did the classifier catch? High recall means few missed cases. |\n",
    "| **F1 score** | A single number balancing precision and recall. |\n",
    "| **ROC curve** | Plots true-positive rate against false-positive rate across every possible decision threshold - a way to compare classifiers independent of any one threshold choice. |\n",
    "| **Train/test split** | Trains on part of the data, tests on data the model never saw, so the resulting accuracy reflects generalization rather than memorization. |\n",
    "| **k-fold cross-validation** | Repeats the train/test split *k* times with different slices, giving a distribution of scores instead of one number that depends on how a single split happened to land. |\n",
    "\n",
    "### Feature engineering and hyperparameter search\n",
    "\n",
    "| Technique | What it does | Why it matters |\n",
    "|---|---|---|\n",
    "| **Feature scaling** (`StandardScaler`) | Rescales every feature to a comparable range. | Distance-based methods (KNN, clustering) would otherwise let whichever feature has the largest raw numbers dominate. |\n",
    "| **GridSearchCV** | Systematically tries every combination in a set of hyperparameters and cross-validates each one. | Replaces manual trial-and-error with an exhaustive, repeatable search. |\n",
    "\n",
    "### Unsupervised learning: finding structure without labels\n",
    "\n",
    "Unlike the classifiers above, these don't use `contact_interest` at all - they look for structure in the session features alone:\n",
    "\n",
    "| Technique | What it finds | Why it's here |\n",
    "|---|---|---|\n",
    "| **K-means** | Splits sessions into *k* groups, each centered on a mean point. | Formalizes the \"user persona\" idea from [Visualizing the Data](#visualizing-the-data) - do sessions naturally fall into a few behavioral types? |\n",
    "| **Hierarchical clustering / dendrogram** | Builds a tree of nested groupings, from every session as its own cluster up to one big cluster. | Doesn't require picking *k* in advance, and the dendrogram is genuinely readable at this dataset's small size. |\n",
    "| **DBSCAN** | Groups points that are densely packed together, and labels sparse points as noise. | A different definition of \"cluster\" - density-based instead of distance-to-center-based. |\n",
    "| **Anomaly detection** (Robust Covariance, One-Class SVM, Local Outlier Factor) | Flags the sessions that look least like the rest. | Useful for catching bot traffic or unusually extreme sessions, not for finding groups. |\n",
    "\n",
    "### Association analysis\n",
    "\n",
    "**Frequent itemsets** and **association rules** answer a different kind of question entirely - not \"predict a label\" or \"find groups,\" but \"which behaviors tend to occur together?\" (e.g. \"sessions that scroll to 100% also tend to click Contact\"). See [Association Analysis](#association-analysis-what-site-behaviors-go-together) for how this gets adapted to session data.\n"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "6d95fc76-b719-42d6-a627-b7552731bacb",
   "metadata": {},
   "source": [
    "## Importing the Data\n",
    "\n",
    "We load the raw `AnalyticsEvent` export, normalize the two timestamp/id formats described above into one consistent shape, then narrow down to `main-site` events only and treat `event` as a categorical column so the rest of the notebook can group and count by event type cleanly.\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "id": "6f5d6cec-6e9f-4e15-9c64-55fee845a5c5",
   "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>id</th>\n",
       "      <th>event</th>\n",
       "      <th>properties</th>\n",
       "      <th>sessionId</th>\n",
       "      <th>createdAt</th>\n",
       "      <th>service</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>cmqqq040x001jqgjo2poi6ym1</td>\n",
       "      <td>page_load</td>\n",
       "      <td>{}</td>\n",
       "      <td>miki8egczbrmqqq03qr</td>\n",
       "      <td>2026-06-23 07:09:06.603-07</td>\n",
       "      <td>trip-planner</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>cmqqqskt2001kqgjoq3j1u21j</td>\n",
       "      <td>page_load</td>\n",
       "      <td>{}</td>\n",
       "      <td>fvr6awmbhy5mqqqsjvo</td>\n",
       "      <td>2026-06-23 07:31:14.717-07</td>\n",
       "      <td>trip-planner</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>cmqqqtror001lqgjoajbjzrhv</td>\n",
       "      <td>page_load</td>\n",
       "      <td>{}</td>\n",
       "      <td>z9a2kc0k9simqqqtr9a</td>\n",
       "      <td>2026-06-23 07:32:10.346-07</td>\n",
       "      <td>trip-planner</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>cmqqs41zx001mqgjox4soj63v</td>\n",
       "      <td>page_load</td>\n",
       "      <td>{}</td>\n",
       "      <td>d4tpym359nvmqqs41w3</td>\n",
       "      <td>2026-06-23 08:08:09.772-07</td>\n",
       "      <td>trip-planner</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>4e55128c-2284-49a9-b8cb-7104c20870c2</td>\n",
       "      <td>page_load</td>\n",
       "      <td>{}</td>\n",
       "      <td>ayv6yrctjgmqqzlb90</td>\n",
       "      <td>2026-06-23 11:37:34.118451-07</td>\n",
       "      <td>main-site</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1062</th>\n",
       "      <td>582c653a-d7bb-40be-b228-99068a3c6b5c</td>\n",
       "      <td>scroll_depth</td>\n",
       "      <td>{\"percent\": 75}</td>\n",
       "      <td>zhavnda0zcgmsodwuh9</td>\n",
       "      <td>2026-08-11 01:15:02.870659-07</td>\n",
       "      <td>main-site</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1063</th>\n",
       "      <td>b00a3874-a6c5-42f9-ba43-6733512cbccf</td>\n",
       "      <td>section_viewed</td>\n",
       "      <td>{\"section\": \"research\"}</td>\n",
       "      <td>zhavnda0zcgmsodwuh9</td>\n",
       "      <td>2026-08-11 01:15:02.943507-07</td>\n",
       "      <td>main-site</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1064</th>\n",
       "      <td>22ca57ea-5c71-45e9-b781-260846f9d84a</td>\n",
       "      <td>section_viewed</td>\n",
       "      <td>{\"section\": \"contact\"}</td>\n",
       "      <td>zhavnda0zcgmsodwuh9</td>\n",
       "      <td>2026-08-11 01:15:10.323054-07</td>\n",
       "      <td>main-site</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1065</th>\n",
       "      <td>0e3e3c47-3f5e-46fb-a56f-ec89a073278c</td>\n",
       "      <td>scroll_depth</td>\n",
       "      <td>{\"percent\": 100}</td>\n",
       "      <td>zhavnda0zcgmsodwuh9</td>\n",
       "      <td>2026-08-11 01:15:10.991559-07</td>\n",
       "      <td>main-site</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1066</th>\n",
       "      <td>16b116d8-7acd-47d2-93c8-ee006966c596</td>\n",
       "      <td>page_load</td>\n",
       "      <td>{}</td>\n",
       "      <td>5ipjwzolpamspc5wss</td>\n",
       "      <td>2026-08-11 17:13:18.824844-07</td>\n",
       "      <td>main-site</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>1067 rows × 6 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "                                        id           event  \\\n",
       "0                cmqqq040x001jqgjo2poi6ym1       page_load   \n",
       "1                cmqqqskt2001kqgjoq3j1u21j       page_load   \n",
       "2                cmqqqtror001lqgjoajbjzrhv       page_load   \n",
       "3                cmqqs41zx001mqgjox4soj63v       page_load   \n",
       "4     4e55128c-2284-49a9-b8cb-7104c20870c2       page_load   \n",
       "...                                    ...             ...   \n",
       "1062  582c653a-d7bb-40be-b228-99068a3c6b5c    scroll_depth   \n",
       "1063  b00a3874-a6c5-42f9-ba43-6733512cbccf  section_viewed   \n",
       "1064  22ca57ea-5c71-45e9-b781-260846f9d84a  section_viewed   \n",
       "1065  0e3e3c47-3f5e-46fb-a56f-ec89a073278c    scroll_depth   \n",
       "1066  16b116d8-7acd-47d2-93c8-ee006966c596       page_load   \n",
       "\n",
       "                   properties            sessionId  \\\n",
       "0                          {}  miki8egczbrmqqq03qr   \n",
       "1                          {}  fvr6awmbhy5mqqqsjvo   \n",
       "2                          {}  z9a2kc0k9simqqqtr9a   \n",
       "3                          {}  d4tpym359nvmqqs41w3   \n",
       "4                          {}   ayv6yrctjgmqqzlb90   \n",
       "...                       ...                  ...   \n",
       "1062          {\"percent\": 75}  zhavnda0zcgmsodwuh9   \n",
       "1063  {\"section\": \"research\"}  zhavnda0zcgmsodwuh9   \n",
       "1064   {\"section\": \"contact\"}  zhavnda0zcgmsodwuh9   \n",
       "1065         {\"percent\": 100}  zhavnda0zcgmsodwuh9   \n",
       "1066                       {}   5ipjwzolpamspc5wss   \n",
       "\n",
       "                          createdAt       service  \n",
       "0        2026-06-23 07:09:06.603-07  trip-planner  \n",
       "1        2026-06-23 07:31:14.717-07  trip-planner  \n",
       "2        2026-06-23 07:32:10.346-07  trip-planner  \n",
       "3        2026-06-23 08:08:09.772-07  trip-planner  \n",
       "4     2026-06-23 11:37:34.118451-07     main-site  \n",
       "...                             ...           ...  \n",
       "1062  2026-08-11 01:15:02.870659-07     main-site  \n",
       "1063  2026-08-11 01:15:02.943507-07     main-site  \n",
       "1064  2026-08-11 01:15:10.323054-07     main-site  \n",
       "1065  2026-08-11 01:15:10.991559-07     main-site  \n",
       "1066  2026-08-11 17:13:18.824844-07     main-site  \n",
       "\n",
       "[1067 rows x 6 columns]"
      ]
     },
     "execution_count": 1,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "%matplotlib inline\n",
    "\n",
    "import json\n",
    "\n",
    "import matplotlib.pyplot as plt\n",
    "import numpy as np\n",
    "import pandas as pd\n",
    "from sklearn.cluster import KMeans\n",
    "from sklearn.decomposition import PCA\n",
    "from sklearn.preprocessing import StandardScaler\n",
    "\n",
    "# Read the csv as a Pandas table\n",
    "# (the data-dump pipeline writes CSVs under notes/notebooks/data/ - see CLAUDE.md -\n",
    "# shared by both draft and published notebooks)\n",
    "events = pd.read_csv(\"data/analytics_event.csv\")\n",
    "# Show the table as it is read in\n",
    "events\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "id": "e24f3353-1c86-41b6-b985-14a918c56cfc",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "1067 events, 3 services\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "service\n",
       "main-site       644\n",
       "trip-planner    303\n",
       "notes           120\n",
       "Name: count, dtype: int64"
      ]
     },
     "execution_count": 2,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Normalize the different time formats to meet  ISO8601\n",
    "events[\"createdAt\"] = pd.to_datetime(events[\"createdAt\"], format=\"ISO8601\", utc=True)\n",
    "# parse the json properties\n",
    "events[\"properties\"] = events[\"properties\"].apply(json.loads)\n",
    "\n",
    "# Print the total amounts of events an service\n",
    "print(f\"{len(events)} events, {events['service'].nunique()} services\")\n",
    "\n",
    "# show the amounts of each type of service... We are only concerned with mainsite\n",
    "events[\"service\"].value_counts()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "id": "9bf077f6-e7c6-41dd-9e9b-1dc40e582178",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "cuid era     : cmqqq040x001jqgjo2poi6ym1 2026-06-23 14:09:06.603000+00:00\n",
      "uuid/NOW era : 4e55128c-2284-49a9-b8cb-7104c20870c2 2026-06-23 18:37:34.118451+00:00\n"
     ]
    }
   ],
   "source": [
    "# Confirm both id/timestamp eras are actually present, and that parsing didn't drop anything.\n",
    "# Not every render has both eras - a fresh/sparse local dev database may only have one\n",
    "# (or neither), so this reports what's actually present instead of assuming a full\n",
    "# production-shaped history.\n",
    "old_era = events[events[\"id\"].str.startswith(\"cm\")]   # trip-planner cuid era\n",
    "new_era = events[~events[\"id\"].str.startswith(\"cm\")]  # gen_random_uuid era\n",
    "\n",
    "if len(old_era) > 0:\n",
    "    sample_old = old_era.iloc[0]\n",
    "    print(\"cuid era     :\", sample_old[\"id\"], sample_old[\"createdAt\"])\n",
    "else:\n",
    "    print(\"cuid era     : none present in this dataset\")\n",
    "\n",
    "if len(new_era) > 0:\n",
    "    sample_new = new_era.iloc[0]\n",
    "    print(\"uuid/NOW era :\", sample_new[\"id\"], sample_new[\"createdAt\"])\n",
    "else:\n",
    "    print(\"uuid/NOW era : none present in this dataset\")\n",
    "\n",
    "assert events[\"createdAt\"].isna().sum() == 0, \"some timestamps failed to parse\""
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "id": "9bc1520f-18f0-4ca3-85c5-fdd2a03c19f1",
   "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>id</th>\n",
       "      <th>event</th>\n",
       "      <th>properties</th>\n",
       "      <th>sessionId</th>\n",
       "      <th>createdAt</th>\n",
       "      <th>service</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>4e55128c-2284-49a9-b8cb-7104c20870c2</td>\n",
       "      <td>page_load</td>\n",
       "      <td>{}</td>\n",
       "      <td>ayv6yrctjgmqqzlb90</td>\n",
       "      <td>2026-06-23 18:37:34.118451+00:00</td>\n",
       "      <td>main-site</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>94f66e36-2ce1-4392-a213-828fa36ae170</td>\n",
       "      <td>page_load</td>\n",
       "      <td>{}</td>\n",
       "      <td>nj17pvtvly9mqr3jd1j</td>\n",
       "      <td>2026-06-23 20:28:00.434174+00:00</td>\n",
       "      <td>main-site</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>17</th>\n",
       "      <td>81e9c854-83db-4f14-b360-6413b4ef530f</td>\n",
       "      <td>page_load</td>\n",
       "      <td>{}</td>\n",
       "      <td>slx8anuwuremqrbc4do</td>\n",
       "      <td>2026-06-24 00:06:19.830965+00:00</td>\n",
       "      <td>main-site</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>18</th>\n",
       "      <td>220eb49f-c8d6-463e-b087-bc33ed2273cc</td>\n",
       "      <td>page_load</td>\n",
       "      <td>{}</td>\n",
       "      <td>9ewxvfuudiamqrcx5vv</td>\n",
       "      <td>2026-06-24 00:50:40.628926+00:00</td>\n",
       "      <td>main-site</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>19</th>\n",
       "      <td>7e505a28-f0ab-4459-941b-576546fd77ed</td>\n",
       "      <td>page_load</td>\n",
       "      <td>{}</td>\n",
       "      <td>q25txt5b7zmqre5eto</td>\n",
       "      <td>2026-06-24 01:25:05.179063+00:00</td>\n",
       "      <td>main-site</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                                      id      event properties  \\\n",
       "4   4e55128c-2284-49a9-b8cb-7104c20870c2  page_load         {}   \n",
       "7   94f66e36-2ce1-4392-a213-828fa36ae170  page_load         {}   \n",
       "17  81e9c854-83db-4f14-b360-6413b4ef530f  page_load         {}   \n",
       "18  220eb49f-c8d6-463e-b087-bc33ed2273cc  page_load         {}   \n",
       "19  7e505a28-f0ab-4459-941b-576546fd77ed  page_load         {}   \n",
       "\n",
       "              sessionId                        createdAt    service  \n",
       "4    ayv6yrctjgmqqzlb90 2026-06-23 18:37:34.118451+00:00  main-site  \n",
       "7   nj17pvtvly9mqr3jd1j 2026-06-23 20:28:00.434174+00:00  main-site  \n",
       "17  slx8anuwuremqrbc4do 2026-06-24 00:06:19.830965+00:00  main-site  \n",
       "18  9ewxvfuudiamqrcx5vv 2026-06-24 00:50:40.628926+00:00  main-site  \n",
       "19   q25txt5b7zmqre5eto 2026-06-24 01:25:05.179063+00:00  main-site  "
      ]
     },
     "execution_count": 4,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Now we can strip out the trip planner data\n",
    "main_site = events[events[\"service\"] == \"main-site\"].copy()\n",
    "\n",
    "# Show the top 5 results from the table\n",
    "main_site.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "id": "156cfefe-3622-483e-81cc-67d42f647858",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "id                            str\n",
       "event                         str\n",
       "properties                 object\n",
       "sessionId                     str\n",
       "createdAt     datetime64[us, UTC]\n",
       "service                       str\n",
       "dtype: object"
      ]
     },
     "execution_count": 5,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# confirm column data types\n",
    "main_site.dtypes"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "id": "89fab3d1-0524-4e46-9285-a7bbb91bbc8b",
   "metadata": {},
   "outputs": [],
   "source": [
    "# change event type to categorical to associate an integer with the names\n",
    "main_site.event = main_site.event.astype('category')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "id": "373617ee-35e6-4d8b-9254-b79e99afd69f",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "id                            str\n",
       "event                    category\n",
       "properties                 object\n",
       "sessionId                     str\n",
       "createdAt     datetime64[us, UTC]\n",
       "service                       str\n",
       "dtype: object"
      ]
     },
     "execution_count": 7,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "main_site.dtypes"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "135ade47-ca9f-4e5c-978f-4a824042d697",
   "metadata": {},
   "source": [
    "## Visualizing the Data\n",
    "\n",
    "A first look at how people actually use the site: how long sessions last, how many actions they trigger, and how far they scroll - each shown as a box-and-whisker plot with the underlying numbers spelled out in prose below it.\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "id": "62af0502-99a3-4350-9c51-06792d92000b",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<Axes: >"
      ]
     },
     "execution_count": 8,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 672x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# 1. Find the start and end time for each session\n",
    "session_times = main_site.groupby('sessionId')['createdAt'].agg(['min', 'max'])\n",
    "\n",
    "# 2. Calculate duration in minutes\n",
    "session_times['duration_minutes'] = (session_times['max'] - session_times['min']).dt.total_seconds() / 60.0\n",
    "\n",
    "# 3. Cap duration for these visualizations only - a single session with an idle open\n",
    "# tab can run for hours and swamps the scale of a box plot / scatter matrix. 100\n",
    "# minutes is a generous cap for someone actually reading the site. This only affects\n",
    "# the charts below - the classifier later in this notebook still trains on the real,\n",
    "# uncapped duration.\n",
    "DURATION_CAP_MINUTES = 100\n",
    "session_times['duration_minutes'] = session_times['duration_minutes'].clip(upper=DURATION_CAP_MINUTES)\n",
    "\n",
    "# 4. Create the boxplot\n",
    "session_times.boxplot(column='duration_minutes')"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "8f90cf98-8890-451e-af73-1b093eec8a6e",
   "metadata": {},
   "source": [
    "**Reading the box plot:** the box spans the 1st to 3rd quartile, the green line marks the median, and the whiskers show the normal range of the data - circled points beyond them are outliers. Duration here is capped at 100 minutes (see the code above) so a rare session where someone left the tab open for hours doesn't flatten the whole chart."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "id": "913aee8d-8c94-4ca0-b7c9-85b98d5349ad",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Therefore, we can conclude that a typical user session duration is 0.00 minutes (the median).\n",
      "The middle 50% of our users stay between 0.00 and 0.01 minutes (the box range).\n",
      "The chart above caps duration at 100 minutes, so the 100.00 minute maximum shown\n",
      "here is the capped value, not the true longest session - a handful of real sessions run far longer\n",
      "than that (people who leave the tab open), which is exactly why the cap exists.\n",
      "\n",
      "We most also take into consideration that the duration is calculated by the first and last event per session.\n",
      "This means that a user could stay on the page for some time after without triggering another event.\n"
     ]
    }
   ],
   "source": [
    "# 1. Calculate the components dynamically\n",
    "stats = session_times['duration_minutes'].describe()\n",
    "\n",
    "median_val = stats['50%']\n",
    "q1_val = stats['25%']\n",
    "q3_val = stats['75%']\n",
    "max_val = stats['max']\n",
    "\n",
    "# 2. Format the dynamic print statement\n",
    "print(\n",
    "    f\"Therefore, we can conclude that a typical user session duration is {median_val:.2f} minutes (the median).\\n\"\n",
    "    f\"The middle 50% of our users stay between {q1_val:.2f} and {q3_val:.2f} minutes (the box range).\\n\"\n",
    "    f\"The chart above caps duration at {DURATION_CAP_MINUTES} minutes, so the {max_val:.2f} minute maximum shown\\n\"\n",
    "    f\"here is the capped value, not the true longest session - a handful of real sessions run far longer\\n\"\n",
    "    f\"than that (people who leave the tab open), which is exactly why the cap exists.\\n\\n\"\n",
    "    f\"We most also take into consideration that the duration is calculated by the first and last event per session.\\n\"\n",
    "    f\"This means that a user could stay on the page for some time after without triggering another event.\"\n",
    ")\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "id": "1d38c8d0-7c92-41c5-be24-ef6fec787ee9",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<Axes: >"
      ]
     },
     "execution_count": 10,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 672x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# 1. Count how many events happen in each session\n",
    "session_counts = main_site.groupby('sessionId').size().reset_index(name='event_count')\n",
    "\n",
    "# 2. Boxplot the distribution of activity\n",
    "session_counts.boxplot(column='event_count')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "id": "926c1fe4-03cf-4266-a858-844f5ed9113b",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Therefore, we can conclude that a typical user session consists of 1.00 events (the median).\n",
      "The middle 50% of our users trigger between 1.00 and 2.00 events (the box range).\n",
      "While the absolute maximum recorded activity was 19.00 events, the whiskers show the normal\n",
      "range of data, meaning extreme high values are likely just click-happy users.\n"
     ]
    }
   ],
   "source": [
    "# 1. Calculate the components dynamically\n",
    "stats = session_counts['event_count'].describe()\n",
    "\n",
    "median_val = stats['50%']\n",
    "q1_val = stats['25%']\n",
    "q3_val = stats['75%']\n",
    "max_val = stats['max']\n",
    "\n",
    "# 2. Format the dynamic print statement\n",
    "print(\n",
    "    f\"Therefore, we can conclude that a typical user session consists of {median_val:.2f} events (the median).\\n\"\n",
    "    f\"The middle 50% of our users trigger between {q1_val:.2f} and {q3_val:.2f} events (the box range).\\n\"\n",
    "    f\"While the absolute maximum recorded activity was {max_val:.2f} events, the whiskers show the normal\\n\"\n",
    "    f\"range of data, meaning extreme high values are likely just click-happy users.\"\n",
    ")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "id": "0fd8b875-fff4-46e2-af31-7e4e2a60af35",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 576x576 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# 1. Filter for scroll events\n",
    "scroll_data = main_site[main_site['event'] == 'scroll_depth'].copy()\n",
    "\n",
    "# 2. Extract the number from the JSON string safely\n",
    "def extract_percent(prop_str):\n",
    "    try:\n",
    "        # If it's already a dict, grab it; if it's a string, parse it as JSON\n",
    "        if isinstance(prop_str, dict):\n",
    "            return prop_str.get('percent')\n",
    "        \n",
    "        data = json.loads(prop_str)\n",
    "        return data.get('percent')\n",
    "    except (json.JSONDecodeError, TypeError, AttributeError):\n",
    "        return None\n",
    "\n",
    "scroll_data['depth'] = scroll_data['properties'].apply(extract_percent)\n",
    "\n",
    "# 3. Ensure it is numeric\n",
    "scroll_data['depth'] = pd.to_numeric(scroll_data['depth'], errors='coerce')\n",
    "\n",
    "# 4. Create the plot! \n",
    "# (We drop 'by=sessionId' here to avoid the ugly X-axis problem)\n",
    "plt.figure(figsize=(6, 6))\n",
    "scroll_data.boxplot(column='depth')\n",
    "\n",
    "plt.title(\"Overall User Scroll Depth Distribution\")\n",
    "plt.ylabel(\"Scroll Percentage (%)\")\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "id": "5b769e88-8111-45ce-9413-8d479fea9556",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Therefore, we can conclude that a typical user scrolls to 50.0 % (the median) of the main page.\n",
      "The middle 50% of our users scroll between 25.0 and 75.0 events (the box range).\n"
     ]
    }
   ],
   "source": [
    "# 1. Calculate the components dynamically\n",
    "stats = scroll_data['depth'].describe()\n",
    "\n",
    "median_val = stats['50%']\n",
    "q1_val = stats['25%']\n",
    "q3_val = stats['75%']\n",
    "max_val = stats['max']\n",
    "\n",
    "# 2. Format the dynamic print statement\n",
    "print(\n",
    "    f\"Therefore, we can conclude that a typical user scrolls to {median_val:.1f} % (the median) of the main page.\\n\"\n",
    "    f\"The middle 50% of our users scroll between {q1_val:.1f} and {q3_val:.1f} events (the box range).\"\n",
    ")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "id": "52b28956-dcbd-4798-b4c0-3ccadaa5b014",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 960x960 with 9 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# 1. Create a clean metrics DataFrame per session\n",
    "# Total events per session\n",
    "session_metrics = main_site.groupby('sessionId').size().reset_index(name='total_actions')\n",
    "\n",
    "# Session duration in minutes\n",
    "times = main_site.groupby('sessionId')['createdAt'].agg(['min', 'max'])\n",
    "session_metrics['duration_minutes'] = ((times['max'] - times['min']).dt.total_seconds() / 60.0).values\n",
    "\n",
    "# Count how many links were clicked in each session\n",
    "link_clicks = main_site[main_site['event'] == 'link_clicked'].groupby('sessionId').size()\n",
    "session_metrics['links_clicked'] = session_metrics['sessionId'].map(link_clicks).fillna(0)\n",
    "\n",
    "# 2. Drop the sessionId column (since it's text) just for the plot, and cap duration the\n",
    "# same way as the box plot above. This clips a copy used only for the chart -\n",
    "# session_metrics itself stays uncapped, since the classifier later in this notebook\n",
    "# trains on the real duration.\n",
    "plot_data = session_metrics[['total_actions', 'duration_minutes', 'links_clicked']].copy()\n",
    "plot_data['duration_minutes'] = plot_data['duration_minutes'].clip(upper=DURATION_CAP_MINUTES)\n",
    "\n",
    "# 3. Generate the Scatter Matrix!\n",
    "# Needs at least a couple of sessions with values in every column, or pandas' internal\n",
    "# min/max range calculation blows up on an empty array - a fresh/sparse local dev\n",
    "# database (few or no main-site sessions yet) can hit that, so this adapts the same way\n",
    "# the classifier evaluation cell further down does.\n",
    "# (The '_ =' syntax just suppresses messy text output in Jupyter notebooks)\n",
    "if len(plot_data.dropna()) >= 2:\n",
    "    _ = pd.plotting.scatter_matrix(\n",
    "        plot_data,\n",
    "        diagonal='kde',\n",
    "        figsize=(10, 10),\n",
    "        alpha=0.6,    # Makes dots slightly transparent to see overlaps\n",
    "        density_kwds={'color': 'red'} # Color the KDE lines red\n",
    "    )\n",
    "    plt.show()\n",
    "else:\n",
    "    print(f\"Only {len(plot_data)} session(s) available - too few to plot a scatter matrix.\")"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "ed223172-6fcc-482f-ba1e-31c772a4afa7",
   "metadata": {},
   "source": [
    "A correlation matrix, visualized as a heatmap - warmer colors mean two metrics move together more strongly. The prose below walks through what these numbers mean for `links_clicked` and `total_actions` specifically.\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "id": "6318ad28-15af-44a0-92ad-699065947991",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 576x480 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "corr_matrix_viz = plot_data.corr()\n",
    "\n",
    "plt.figure(figsize=(6, 5))\n",
    "plt.imshow(corr_matrix_viz, cmap=\"coolwarm\", vmin=-1, vmax=1)\n",
    "plt.xticks(range(len(corr_matrix_viz.columns)), corr_matrix_viz.columns, rotation=45, ha=\"right\")\n",
    "plt.yticks(range(len(corr_matrix_viz.columns)), corr_matrix_viz.columns)\n",
    "plt.colorbar(label=\"Correlation\")\n",
    "plt.title(\"Correlation between session metrics\")\n",
    "plt.tight_layout()\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "id": "59e7af97-c472-43de-862f-009b279608b6",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "--- EXECUTIVE SUMMARY OF USER BEHAVIOR ---\n",
      "\n",
      "Based on our session metrics, a typical user experience lasts for 0.0000 minutes and generates 1.00 total actions.\n",
      "\n",
      "1. IDENTITY USER PERSONAS\n",
      "The relationship between total activity and duration tells us our audience personas are fragmented. We have a mix of rapid-fire clickers alongside users who generate very few actions over long stretches of time (Window Shoppers vs. Idle Tabs).\n",
      "\n",
      "2. DISCOVER FEATURE VALUE\n",
      "When looking at link engagement, there is a slight positive relationship, showing some connection between link engagement and session length.\n",
      "\n",
      "3. SEE YOUR 'AVERAGE' USER EXPERIENCE\n",
      "Looking at the distribution curves, the shape of our traffic shows how varied our audience is. While our median session length sits comfortably at 0.0000 minutes, the longest session shown here reached 100.0000 minutes (this chart is capped at 100 minutes - see the code above). This spread reveals the true 'shape' of our user base—separating the quick bounce traffic from the deeply engaged power users who exhaustively browse the site.\n"
     ]
    }
   ],
   "source": [
    "# 1. Calculate correlations and statistics dynamically\n",
    "corr_matrix = plot_data.corr()\n",
    "stats = plot_data.describe()\n",
    "\n",
    "# Extract key summary metrics\n",
    "median_duration = stats.loc['50%', 'duration_minutes']\n",
    "median_actions = stats.loc['50%', 'total_actions']\n",
    "max_duration = stats.loc['max', 'duration_minutes']\n",
    "\n",
    "# Pull individual correlation coefficients (ranges from -1 to 1)\n",
    "click_duration_corr = corr_matrix.loc['links_clicked', 'duration_minutes']\n",
    "actions_duration_corr = corr_matrix.loc['total_actions', 'duration_minutes']\n",
    "\n",
    "# Dynamically evaluate the relationship between link clicks and time on site\n",
    "if click_duration_corr > 0.5:\n",
    "    feature_value_desc = \"strong positive relationship. This proves that clicking links directly drives user retention and keeps people on the site longer\"\n",
    "elif click_duration_corr > 0.1:\n",
    "    feature_value_desc = \"slight positive relationship, showing some connection between link engagement and session length\"\n",
    "else:\n",
    "    feature_value_desc = \"weak or non-existent relationship. This suggests that link clicks don't impact how long someone stays, meaning they might be clicking an external link and leaving immediately\"\n",
    "\n",
    "# Dynamically analyze user personas based on action vs duration correlation\n",
    "if actions_duration_corr > 0.6:\n",
    "    persona_desc = \"highly consistent. Users are steadily interacting with the site the entire time they are here (Classic Explorers).\"\n",
    "else:\n",
    "    persona_desc = \"fragmented. We have a mix of rapid-fire clickers alongside users who generate very few actions over long stretches of time (Window Shoppers vs. Idle Tabs).\"\n",
    "\n",
    "# 2. Format the dynamic print statement\n",
    "print(\n",
    "    f\"--- EXECUTIVE SUMMARY OF USER BEHAVIOR ---\\n\\n\"\n",
    "    f\"Based on our session metrics, a typical user experience lasts for {median_duration:.4f} minutes \"\n",
    "    f\"and generates {median_actions:.2f} total actions.\\n\\n\"\n",
    "    f\"1. IDENTITY USER PERSONAS\\n\"\n",
    "    f\"The relationship between total activity and duration tells us our audience personas are {persona_desc}\\n\\n\"\n",
    "    f\"2. DISCOVER FEATURE VALUE\\n\"\n",
    "    f\"When looking at link engagement, there is a {feature_value_desc}.\\n\\n\"\n",
    "    f\"3. SEE YOUR 'AVERAGE' USER EXPERIENCE\\n\"\n",
    "    f\"Looking at the distribution curves, the shape of our traffic shows how varied our audience is. \"\n",
    "    f\"While our median session length sits comfortably at {median_duration:.4f} minutes, the longest session \"\n",
    "    f\"shown here reached {max_duration:.4f} minutes (this chart is capped at {DURATION_CAP_MINUTES} minutes - \"\n",
    "    f\"see the code above). This spread reveals the true 'shape' of our user base—separating \"\n",
    "    f\"the quick bounce traffic from the deeply engaged power users who exhaustively browse the site.\"\n",
    ")"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "140bcde6-19da-43cb-bbba-0cf41f628b88",
   "metadata": {},
   "source": [
    "## Classifying Contact Interest\n",
    "\n",
    "`session_metrics` above already looks like a supervised learning setup: numeric features per session (`total_actions`, `duration_minutes`, `links_clicked`) with an obvious target missing. Nothing in this schema hands us a class to predict — so let's engineer one instead of only clustering.\n",
    "\n",
    "**Label:** did the session ever show interest in the Contact section — i.e. a `section_viewed` or `nav_clicked` event with `section: \"contact\"`? That's a reasonable proxy for \"this visitor was interested enough to consider reaching out,\" and it turns the same session-level features into a decision tree classification problem.\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "id": "c17af984-8540-4457-91ff-a53aa47feafe",
   "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>sessionId</th>\n",
       "      <th>total_actions</th>\n",
       "      <th>duration_minutes</th>\n",
       "      <th>links_clicked</th>\n",
       "      <th>max_scroll_depth</th>\n",
       "      <th>contact_interest</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>00bsroi4eesdbmrk9eb4n</td>\n",
       "      <td>1</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>False</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>03gvoia5f6i8msg8xxsy</td>\n",
       "      <td>1</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>False</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>0brdpxf1ghbomqzjqplv</td>\n",
       "      <td>1</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>False</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>0f5pn5nppwhgmqx5ihpv</td>\n",
       "      <td>1</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>False</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>0qbzvkhcrl9bmr1ftn9s</td>\n",
       "      <td>1</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>False</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>268</th>\n",
       "      <td>zia5gcb6nkpmrkzvz7y</td>\n",
       "      <td>3</td>\n",
       "      <td>0.013074</td>\n",
       "      <td>0.0</td>\n",
       "      <td>25.0</td>\n",
       "      <td>False</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>269</th>\n",
       "      <td>zibdt9s92bemqu5usva</td>\n",
       "      <td>1</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>False</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>270</th>\n",
       "      <td>zoarkshcbimqxb7b2e</td>\n",
       "      <td>2</td>\n",
       "      <td>0.000069</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>False</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>271</th>\n",
       "      <td>ztk0odbt8fsmr8hq4ok</td>\n",
       "      <td>1</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>False</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>272</th>\n",
       "      <td>zwl6hocmtnmreaq821</td>\n",
       "      <td>5</td>\n",
       "      <td>0.730516</td>\n",
       "      <td>0.0</td>\n",
       "      <td>50.0</td>\n",
       "      <td>False</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>273 rows × 6 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "                 sessionId  total_actions  duration_minutes  links_clicked  \\\n",
       "0    00bsroi4eesdbmrk9eb4n              1          0.000000            0.0   \n",
       "1     03gvoia5f6i8msg8xxsy              1          0.000000            0.0   \n",
       "2     0brdpxf1ghbomqzjqplv              1          0.000000            0.0   \n",
       "3     0f5pn5nppwhgmqx5ihpv              1          0.000000            0.0   \n",
       "4     0qbzvkhcrl9bmr1ftn9s              1          0.000000            0.0   \n",
       "..                     ...            ...               ...            ...   \n",
       "268    zia5gcb6nkpmrkzvz7y              3          0.013074            0.0   \n",
       "269    zibdt9s92bemqu5usva              1          0.000000            0.0   \n",
       "270     zoarkshcbimqxb7b2e              2          0.000069            0.0   \n",
       "271    ztk0odbt8fsmr8hq4ok              1          0.000000            0.0   \n",
       "272     zwl6hocmtnmreaq821              5          0.730516            0.0   \n",
       "\n",
       "     max_scroll_depth  contact_interest  \n",
       "0                 0.0             False  \n",
       "1                 0.0             False  \n",
       "2                 0.0             False  \n",
       "3                 0.0             False  \n",
       "4                 0.0             False  \n",
       "..                ...               ...  \n",
       "268              25.0             False  \n",
       "269               0.0             False  \n",
       "270               0.0             False  \n",
       "271               0.0             False  \n",
       "272              50.0             False  \n",
       "\n",
       "[273 rows x 6 columns]"
      ]
     },
     "execution_count": 17,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# 1. Add max scroll depth reached per session as another behavioral feature\n",
    "max_scroll = scroll_data.groupby('sessionId')['depth'].max()\n",
    "session_metrics['max_scroll_depth'] = session_metrics['sessionId'].map(max_scroll).fillna(0)\n",
    "\n",
    "# 2. Label each session: did it ever show interest in the Contact section?\n",
    "contact_events = main_site[main_site['event'].isin(['section_viewed', 'nav_clicked'])]\n",
    "# .apply() on an empty Series returns a result that doesn't line up as a boolean mask\n",
    "# (indexing with it collapses to a 0-column frame), so a sparse/empty local dev dataset\n",
    "# needs this handled explicitly rather than falling through to the apply/filter below.\n",
    "if len(contact_events) > 0:\n",
    "    is_contact = contact_events['properties'].apply(lambda p: p.get('section') == 'contact')\n",
    "    contact_sessions = set(contact_events[is_contact]['sessionId'])\n",
    "else:\n",
    "    contact_sessions = set()\n",
    "session_metrics['contact_interest'] = session_metrics['sessionId'].isin(contact_sessions)\n",
    "\n",
    "session_metrics[['sessionId', 'total_actions', 'duration_minutes', 'links_clicked', 'max_scroll_depth', 'contact_interest']]"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "626e4f5e-b5e1-429d-8c6a-d69c37ab8cbb",
   "metadata": {},
   "source": [
    "Each row above is one session with its behavioral features (`total_actions`, `duration_minutes`, `links_clicked`, `max_scroll_depth`) alongside the engineered `contact_interest` label - this is the exact table the classifier below is trained on.\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "id": "cf61d725-c4a3-48ca-81ae-fc632dab3f21",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "|--- max_scroll_depth <= 62.50\n",
      "|   |--- duration_minutes <= 0.00\n",
      "|   |   |--- class: 0\n",
      "|   |--- duration_minutes >  0.00\n",
      "|   |   |--- duration_minutes <= 0.00\n",
      "|   |   |   |--- class: 1\n",
      "|   |   |--- duration_minutes >  0.00\n",
      "|   |   |   |--- class: 0\n",
      "|--- max_scroll_depth >  62.50\n",
      "|   |--- max_scroll_depth <= 87.50\n",
      "|   |   |--- links_clicked <= 3.00\n",
      "|   |   |   |--- class: 0\n",
      "|   |   |--- links_clicked >  3.00\n",
      "|   |   |   |--- class: 1\n",
      "|   |--- max_scroll_depth >  87.50\n",
      "|   |   |--- class: 1\n",
      "\n"
     ]
    }
   ],
   "source": [
    "import sklearn.tree\n",
    "\n",
    "# 1. Pick the behavioral features and the engineered label\n",
    "feature_cols = ['total_actions', 'duration_minutes', 'links_clicked', 'max_scroll_depth']\n",
    "X = session_metrics[feature_cols].values\n",
    "y = session_metrics['contact_interest'].astype(int).values\n",
    "\n",
    "# 2. Fit a shallow decision tree (kept shallow so it stays readable while the dataset is small).\n",
    "# Needs at least one main-site session to fit at all - a fresh/empty local dev database can\n",
    "# have zero, so this (and the two cells below that depend on contact_classifier) adapt the\n",
    "# same way the rest of this notebook does.\n",
    "if len(session_metrics) > 0:\n",
    "    contact_classifier = sklearn.tree.DecisionTreeClassifier(max_depth=3)\n",
    "    contact_classifier.fit(X, y)\n",
    "    # 3. Explain the tree as text\n",
    "    print(sklearn.tree.export_text(contact_classifier, feature_names=feature_cols))\n",
    "else:\n",
    "    contact_classifier = None\n",
    "    print(\"No main-site sessions available - skipping the decision tree.\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "id": "1f269b5c-b4c8-42f0-82bf-3f4363b48736",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 1152x768 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Visualize the same tree\n",
    "if contact_classifier is not None:\n",
    "    plt.figure(figsize=(12, 8))\n",
    "    _ = sklearn.tree.plot_tree(\n",
    "        contact_classifier,\n",
    "        feature_names=feature_cols,\n",
    "        class_names=['no_contact_interest', 'contact_interest'],\n",
    "        filled=True\n",
    "    )\n",
    "else:\n",
    "    print(\"No classifier to visualize.\")"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "da7bed0f-7fb1-458a-add0-05cd2252f05f",
   "metadata": {},
   "source": [
    "**Reading the tree:** each split asks a yes/no question about one feature - true goes left, false goes right. `class: 1` (`contact_interest`) leaves show in a warmer shade in the plot above, `class: 0` in a cooler one. The text dump above and this plot describe the exact same tree; the plot's easier to scan, the text is easier to copy or quote.\n"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "9b6e3f8b-1fc1-427d-9412-599d094add30",
   "metadata": {},
   "source": [
    "### Evaluating the Classifier\n",
    "\n",
    "Accuracy alone can be misleading, so this checks the tree's held-out accuracy alongside a confusion matrix - see [Tools and Techniques](#tools-and-techniques) for what each metric means.\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "id": "2da0a8cf-1a05-4ceb-ba00-2fb6a7701b3b",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 672x480 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "273 sessions available ({False: 248, True: 25} by class) - enough per class to hold out a test set.\n",
      "Trained on 191, tested on 82 held-out sessions.\n",
      "Held-out accuracy: 0.94\n"
     ]
    }
   ],
   "source": [
    "# How good is the classifier? This adapts to however much data actually exists at render\n",
    "# time below a per-class threshold a held-out split isn't meaningful, so we fall back to training accuracy.\n",
    "from sklearn.model_selection import train_test_split\n",
    "\n",
    "if contact_classifier is None:\n",
    "    print(\"No classifier to evaluate.\")\n",
    "else:\n",
    "    class_counts = session_metrics['contact_interest'].value_counts()\n",
    "    min_class_count = class_counts.min()\n",
    "    min_per_class_for_split = 10\n",
    "\n",
    "    if min_class_count >= min_per_class_for_split:\n",
    "        X_train, X_test, y_train, y_test = train_test_split(\n",
    "            X, y, test_size=0.3, stratify=y, random_state=42\n",
    "        )\n",
    "        eval_classifier = sklearn.tree.DecisionTreeClassifier(max_depth=3)\n",
    "        eval_classifier.fit(X_train, y_train)\n",
    "        accuracy = eval_classifier.score(X_test, y_test)\n",
    "        from sklearn.metrics import ConfusionMatrixDisplay\n",
    "\n",
    "        ConfusionMatrixDisplay.from_estimator(\n",
    "            eval_classifier, X_test, y_test,\n",
    "            display_labels=['no_contact_interest', 'contact_interest']\n",
    "        )\n",
    "        plt.show()\n",
    "        print(\n",
    "            f\"{len(session_metrics)} sessions available ({class_counts.to_dict()} by class) - \"\n",
    "            f\"enough per class to hold out a test set.\\n\"\n",
    "            f\"Trained on {len(X_train)}, tested on {len(X_test)} held-out sessions.\\n\"\n",
    "            f\"Held-out accuracy: {accuracy:.2f}\"\n",
    "        )\n",
    "    else:\n",
    "        accuracy = contact_classifier.score(X, y)\n",
    "        print(\n",
    "            f\"Only {len(session_metrics)} sessions available ({class_counts.to_dict()} by class) - \"\n",
    "            f\"too few per class to hold out a meaningful test set.\\n\"\n",
    "            f\"Training accuracy (read skeptically - not a generalization estimate): {accuracy:.2f}\"\n",
    "        )"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "2365e48a-b842-478e-ad92-dc752833eb83",
   "metadata": {},
   "source": [
    "**Note on the accuracy above:** like every other cell in this notebook, this reruns against whatever `AnalyticsEvent` rows actually exist at render time - local dev data when run locally, real production data at each `publish:notes` build (see CI). The evaluation cell adapts to that automatically: below a per-class sample threshold it falls back to training accuracy, which should be read skeptically (the tree is being scored on data it already saw); once enough sessions accumulate per class in production, it switches to a genuine held-out train/test split instead.\n"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "fd573cbb-fa81-4e39-80d1-bba8c2e0e1de",
   "metadata": {},
   "source": [
    "**Precision, recall, and F1** give a fuller picture than accuracy alone, especially since one class may end up rarer than the other once real production data accumulates.\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "id": "59e136f2-6d58-4261-8b07-9f6d187b02e2",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Precision: 0.71, Recall: 0.62, F1: 0.67\n"
     ]
    }
   ],
   "source": [
    "from sklearn.metrics import precision_score, recall_score, f1_score\n",
    "\n",
    "if contact_classifier is not None and min_class_count >= min_per_class_for_split:\n",
    "    predicted = eval_classifier.predict(X_test)\n",
    "    precision = precision_score(y_test, predicted, zero_division=0)\n",
    "    recall = recall_score(y_test, predicted, zero_division=0)\n",
    "    f1 = f1_score(y_test, predicted, zero_division=0)\n",
    "    print(f\"Precision: {precision:.2f}, Recall: {recall:.2f}, F1: {f1:.2f}\")\n",
    "else:\n",
    "    print(f\"Only {len(session_metrics)} sessions available - too few per class for a held-out precision/recall/F1 evaluation yet.\")"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "e94224d9-ab47-44ff-abf0-b638c0bba2b7",
   "metadata": {},
   "source": [
    "### Why One Split Isn't Enough\n",
    "\n",
    "The decision tree above was evaluated on one particular train/test split. But is that split representative? Below, the same idea - picking the best KNN `n_neighbors` - is repeated across two different random splits, to see whether the \"best\" choice actually changes depending on how the data happens to be divided.\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "id": "255a7eae-a19d-4dea-b947-e2bee8b62a65",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "random_state=1: best n_neighbors=1  scores={1: 0.9512195121951219, 3: 0.9512195121951219, 5: 0.9512195121951219, 7: 0.9512195121951219, 9: 0.9512195121951219}\n",
      "random_state=42: best n_neighbors=1  scores={1: 0.926829268292683, 3: 0.926829268292683, 5: 0.926829268292683, 7: 0.926829268292683, 9: 0.926829268292683}\n"
     ]
    }
   ],
   "source": [
    "from sklearn.neighbors import KNeighborsClassifier\n",
    "from sklearn.model_selection import train_test_split\n",
    "\n",
    "if contact_classifier is not None and min_class_count >= min_per_class_for_split:\n",
    "    n_neighbors_range = [1, 3, 5, 7, 9]\n",
    "    for seed in [1, 42]:\n",
    "        Xtr, Xte, ytr, yte = train_test_split(X, y, test_size=0.3, stratify=y, random_state=seed)\n",
    "        scores = {k: KNeighborsClassifier(n_neighbors=k).fit(Xtr, ytr).score(Xte, yte) for k in n_neighbors_range}\n",
    "        print(f\"random_state={seed}: best n_neighbors={max(scores, key=scores.get)}  scores={scores}\")\n",
    "else:\n",
    "    print(\"Not enough sessions per class to demonstrate split instability yet.\")"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "6a406e0d-bc9e-455e-89dd-a04a37cdb270",
   "metadata": {},
   "source": [
    "### Cross-Validation\n",
    "\n",
    "Cross-validation fixes the instability shown above by repeating the evaluation across multiple splits and reporting a distribution rather than a single number.\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "id": "c83771fc-970f-466e-ad0d-765b56cbbc38",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 672x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Mean: 0.96, Std-dev: 0.01\n"
     ]
    }
   ],
   "source": [
    "from sklearn.model_selection import cross_val_score\n",
    "\n",
    "cv_folds = 5\n",
    "if contact_classifier is not None and min_class_count >= cv_folds:\n",
    "    cv_scores = cross_val_score(sklearn.tree.DecisionTreeClassifier(max_depth=3), X, y, cv=cv_folds)\n",
    "    plt.boxplot(cv_scores)\n",
    "    plt.ylabel(\"Accuracy\")\n",
    "    plt.title(f\"{cv_folds}-fold CV Accuracy\")\n",
    "    plt.show()\n",
    "    print(f\"Mean: {cv_scores.mean():.2f}, Std-dev: {cv_scores.std():.2f}\")\n",
    "else:\n",
    "    print(f\"Need at least {cv_folds} sessions in the smaller class for {cv_folds}-fold CV.\")"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "02118f78-d61c-465d-9c80-b639d82ea72c",
   "metadata": {},
   "source": [
    "### Overfitting: How Deep Should the Tree Be?\n",
    "\n",
    "The tree above was capped at `max_depth=3` somewhat arbitrarily. Sweeping over depth and comparing training accuracy to held-out test accuracy shows why: as depth increases, training accuracy climbs toward 100% while test accuracy plateaus or drops - the textbook shape of overfitting.\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "id": "9c851ca9-90d7-47fd-a2bf-2b30477e8b53",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 672x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "if min_class_count >= min_per_class_for_split:\n",
    "    depths = list(range(1, 8))\n",
    "    rows = []\n",
    "    for d in depths:\n",
    "        tree = sklearn.tree.DecisionTreeClassifier(max_depth=d).fit(X_train, y_train)\n",
    "        rows.append({\"max_depth\": d, \"train\": tree.score(X_train, y_train), \"test\": tree.score(X_test, y_test)})\n",
    "    depth_accuracy = pd.DataFrame(rows)\n",
    "\n",
    "    plt.plot(depth_accuracy[\"max_depth\"], depth_accuracy[\"train\"], \"ro-\", label=\"Train\")\n",
    "    plt.plot(depth_accuracy[\"max_depth\"], depth_accuracy[\"test\"], \"bv--\", label=\"Test\")\n",
    "    plt.xlabel(\"max_depth\"); plt.ylabel(\"Accuracy\"); plt.legend()\n",
    "    plt.title(\"Decision Tree: train vs. test accuracy by depth\")\n",
    "    plt.show()\n",
    "else:\n",
    "    print(\"Not enough sessions per class for a train/test overfitting curve yet.\")"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "5265e74e-7495-4037-a849-c9409a854965",
   "metadata": {},
   "source": [
    "### Comparing Classifiers\n",
    "\n",
    "Six classifiers - see [Tools and Techniques](#tools-and-techniques) for what each one does - evaluated with the same cross-validation used above, including a `Dummy` baseline that always guesses the majority class. If the real classifiers aren't clearly beating Dummy, they aren't learning much from the data.\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "id": "17f95400-86e4-462b-b53a-7df10eca5233",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 672x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "import sklearn.neighbors\n",
    "import sklearn.ensemble\n",
    "import sklearn.naive_bayes\n",
    "from sklearn.dummy import DummyClassifier\n",
    "\n",
    "if contact_classifier is not None and min_class_count >= cv_folds:\n",
    "    n_neighbors = min(5, min_class_count - 1)  # can't exceed the smallest class size\n",
    "    models = {\n",
    "        \"Decision Tree\": sklearn.tree.DecisionTreeClassifier(max_depth=3),\n",
    "        \"K-Nearest Neighbors\": sklearn.neighbors.KNeighborsClassifier(n_neighbors=n_neighbors),\n",
    "        \"Random Forest\": sklearn.ensemble.RandomForestClassifier(n_estimators=100, max_depth=3, random_state=42),\n",
    "        \"Bagging\": sklearn.ensemble.BaggingClassifier(\n",
    "            estimator=sklearn.tree.DecisionTreeClassifier(max_depth=3), n_estimators=50, random_state=42\n",
    "        ),\n",
    "        \"Gaussian Naive Bayes\": sklearn.naive_bayes.GaussianNB(),\n",
    "        \"Dummy (baseline)\": DummyClassifier(strategy=\"most_frequent\"),\n",
    "    }\n",
    "    comparison = pd.DataFrame([\n",
    "        {\"model\": name, \"mean_accuracy\": cross_val_score(model, X, y, cv=cv_folds).mean()}\n",
    "        for name, model in models.items()\n",
    "    ]).sort_values(\"mean_accuracy\", ascending=False)\n",
    "\n",
    "    comparison.plot.bar(x=\"model\", y=\"mean_accuracy\", legend=False)\n",
    "    plt.ylabel(\"Mean CV Accuracy\")\n",
    "    plt.xticks(rotation=20)\n",
    "    plt.show()\n",
    "    comparison\n",
    "else:\n",
    "    print(f\"Need at least {cv_folds} sessions in the smaller class to compare classifiers.\")\n"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "247d8ce1-36a5-46ed-a3f5-090aecffef8b",
   "metadata": {},
   "source": [
    "### ROC Curves\n",
    "\n",
    "Accuracy and even F1 collapse a classifier's behavior into one number at one decision threshold. ROC curves show the tradeoff between true-positive rate and false-positive rate across *every* possible threshold - the closer a curve hugs the top-left corner, the better. The diagonal dashed line is what the Dummy baseline should trace, since it uses no information from the features at all.\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "id": "fc7b4dc9-1a5d-4be4-bebf-374d4175b1ab",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAgEAAAINCAYAAACnJKwaAAAAOnRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjExLjEsIGh0dHBzOi8vbWF0cGxvdGxpYi5vcmcvctoD+AAAAAlwSFlzAAAOxAAADsQBlSsOGwAAwRhJREFUeJzs3XdYlXUbwPHvOewhQ3EPFCfuvQfuPXGXq6wcZa5Ss9QsMy1zpGmpqZmmlqW+udLce6a4BUQEJ3sd4Izn/QM5ioKyD+P+XFdX8KxznwPy3M9v3D+VoigKQgghhMh31KYOQAghhBCmIUmAEEIIkU9JEiCEEELkU5IECCGEEPmUJAFCCCFEPiVJgBBCCJFPSRIghBBC5FOSBAghhBD5lLmpAxDZb9u2beh0OgDUajXOzs7UqVMHJyenV5535coV7t69i42NDbVr16ZgwYKvfS1fX19u376NSqWiXLlylC9fHrVack/xeoqisHXrVurUqUP58uVNHU6O5u/vz5kzZ+jRoweWlpYmf90bN25w+/Zt4uLi6NmzJ6dOnUKtVtOsWbNsi02kkiLyHTs7O8XV1VXx9PRUevXqpVSuXFmxtrZWvvvuu2SP37dvn1KpUiWlWLFiSpcuXZRmzZopVlZWyrBhw5Tw8PBkzzly5IhSu3ZtxcHBQWnbtq3SsWNHxcXFRSlXrpyyatWqrHx7Ig10Op3y+++/Kz4+PjnuNbRarQIo33//fba8Xna6c+eO8vvvvyvx8fGZcr3169crgPLkyZNMuV5GXnfixImKg4OD0r17d8XT01OJjo5W2rZtq3Tt2jVbYxOpI0lAPmRnZ6cMGzbM+L3BYFDee+89Ra1WK+fPn09y7M6dOxVzc3Plww8/TPIH6/r160rZsmWVBg0aKLGxsUnO2bNnj2JhYaG89957SnR0tHF7XFyc8tVXXylNmzbNmjcm0iwyMlIBlOXLl+e419Dr9Yqnp6eye/fubHm97LRy5UoFUEJDQzPleqZKAo4cOaJ4enoqERERiqIoSkxMjKJWq5WffvopyXGff/65Mnfu3GyNTaSOdAcIVCoVo0eP5scff2T//v3UrVsXgLi4ON5++22aNGnCokWLkpxTpUoVfvrpJzp06MB3333HtGnTjOeMGDGCJk2asHz5clQqlfEcS0tLpk2bRtu2bVMV19WrV7lz5w6lS5emRo0axm6EiIgI/vnnH1q2bEmRIkWMxz969IijR4/SsWNHChQoQHx8PDt27KBBgwa4urpy/vx57t69S8+ePdm+fTuVKlWievXqSV4zJiaGXbt2UbduXdzc3ICEZmkvLy/u3r2Li4sLDRo0wNz85X86N27c4M6dOxQuXJjq1atjbW2dqveZlveeSKvVcvbsWYKDgylXrtxL7+PF937u3DkePXpEw4YNKVy4MJDws9q2bRsAFy9e5I8//gCgYcOGlClThv/973/ExcUB4OjoSKVKlXB1dU1TzK97jVdRqVQMHDiQihUrZtp7gtf/PF/1e2NmZgaARqPhzJkzREREUKVKFWOMz4uLi+PixYuEhIRQsWJF4zHe3t6cP38egB07dmBrawuAp6dnkn8vL4qNjeXcuXNERkZSs2ZNSpYsmeKxjx8/5siRI0BCl1/x4sWpWrUqjo6OqY4zNftdXV0ZOHAgVlZWXLt2jcOHD2MwGPD29jZ+9gBVq1alePHiL7326z5HHx8fLl68SJ8+fYiNjeXEiRPY2NgYuxVeF7tIBVNnISL7vdgSoCiKcvv2bQVQPv74Y+O2rVu3KoCyefPmFK9VvXp1xc3Nzfj9n3/+qQDK77//nu74bty4odStW1dxcnJS2rRpo9SuXVtp3Lix4u/vryiKonh5eSmAsm/fviTn7d69WwGU69evK4qiKE+ePFEAZdmyZUrXrl2VFi1aKJ6enopGo1E8PDyUZs2avfTav/zyiwIoV69eVRRFUW7duqXUqVNHcXFxUTp06KCUK1dOqVixovE1FEVRoqOjlfbt2yvOzs5K586dlcaNGyvly5dX9u7dazzm1KlTqfpMXvfeFUVRjh8/rpQuXVopXbq00rZtW8XBwUFp0aKF8ujRI+Mxie99xYoVSu/evZWWLVsqVatWVWxtbY1P1qGhoUqvXr0UQKlTp47i6empeHp6KkeOHFEURVFGjBhh3NamTRvFxsZG8fT0fKnl51Uxv+41XuXF7oDMeE+p+Xm+6vdGURJ+R5ydnRV3d3elXbt2SoECBZQ33nhDiYuLS/LzLlGihFKlShWla9euSpUqVZQ2bdoojx8/Vv7++2+lXr16CqD06NHDGKNer0/xs1i7dq1SsGBBxc3NTenYsaPi6uqqfPDBB8b9L7YEXL582Xjd3r17K3Xq1FHs7e1fah15VZyp2f/8627evFnp3LmzAihNmjQxvr6np6dSpEiRl7oDUvM5fv/99wqgHDt2TKlWrZrStWtXY4vC62ITqSNJQD6UXBKQ2Dy5ceNG47YpU6YogHL37t0Ur/XGG28ogBIUFKQoiqJMnTr1tee8SlRUlFKmTBmlefPmSnBwsHH72bNnjTfmtCYBVapUUU6ePJnk2NWrVyuA4uvrm2R7+/btlbp16yqKktC0Wa5cOcXDw8M49iE+Pl7p3bu3UqVKFUWn0ymKoigLFy5UnJ2dk8T74MGDJM3YAwYMUF6Xc6fmvQcHByuFChVSunfvbvxj6e/vr5QpU0Zp37698ZzE916tWjXl1KlTiqIkNK8n/rE0GAyKoqSt6TwgIEApXry48uWXX6Yp5vQ2z6eUBKT3PaX25/mq35sDBw4oKpVKmTdvnnGbr6+vUrhwYeWzzz4zbmvUqJEyYMCAJOceOnTIOE4hLd0B//zzj6JSqZTZs2cb36PBYFDWr19vPCY13QEbNmxQzM3NlRs3bqQ6ztftf/F179y5owDKX3/9leScF8cEpPZzTEwCBgwY8NL4o9fFJlJHhmnnU3fv3uWPP/5g8+bNzJgxgwkTJtCpUyf69u1rPCY4OBiAQoUKpXidUqVKJTk2KCgIIFUzB5KzadMm/P39Wbp0aZJr1K9fn6pVq6brmnXq1KFx48ZJtvXr1w8bGxt+/fVX47YHDx5w4MABhg4dCsCWLVu4c+cOixcvxsHBAQALCwu+/PJLbty4weHDhwG4f/8+NjY22NvbG69VrFgxOnXqZPy+cePGeHp6Zvi9b968meDgYL755hvjaOzSpUszefJk9u3bx40bN5Jcs0GDBjRq1AhIaBZ+++23uXHjBgEBAan45CA0NJQTJ07w119/cfLkScqXL29836mNObOl9z2l9ueZKLnfm/nz5+Pu7s5HH31k3FauXDlGjRrFihUrjNvu379v7KJI1KpVK2MXU1rMnz+fKlWq8Omnnxq7C1QqFW+++eZrz/X19WX//v1s3boVc3NzFEXh+PHjqY4zM9/Hi+8pNZ9josGDBxt/ZqmNXaSOjAnIp+7cucOmTZuIj4/n3Llz2Nra8u2332JhYWE8JrGvUqPRYGdnl+x1QkJCkhyb+P/Y2NgkN8XUSoylVq1aaT43JXXq1HlpW4ECBejZsyfr16/ns88+A2Djxo2oVCoGDRoEwJkzZ7CwsODmzZvcunXLeK5erwcS+sDbtGmDp6cnixYtonr16gwaNIh27drRpEmTJP3M48ePZ/z48a+MMzXv/dKlS9jb21O5cuUk2xs0aADA5cuXqVKlSorvPbEfOSAggNKlS78ynhkzZrBgwQIqV65MmTJlMDc3JyAgIMnPNSt+Xq+T3veU2p9nSq+TeI0aNWqwdevWJNvDw8N58uQJT548oXDhwgwaNIj58+fj7e1Njx49aN++PRUqVEj7myXhM+7Tp88rxwu8KDg4mH79+nH27Fnq16+Ps7OzcVzJgwcPjMe9Ls7MfB/PS+3nmCi5n0VWxZbfSBKQT3l4eLB27Vog4Ybdrl07OnXqhJeXl7FeQOLN5Pr167Ro0SLZ69y5cwcHBwdKlCiR5JwbN27QvHnzNMcVHx9vTCRSkvjHTFGUl85NjouLS7LbhwwZwqZNmzh9+jSNGjVi/fr1dOjQwTjYUKPRYG5uzubNm18619PT03jDadSoEf/99x+rVq3izz//ZPbs2RQvXpy1a9fSoUOHV7/hF+J/3XtP6ZjEbYkD+RK9+PSU2Hrw4nEv2rt3L1988QW7du2ic+fOxu19+/ZN0tqQmpgzW3rfU2p/nomS+73RaDTcv3+fTZs2JXuNRPPmzaNRo0Zs2bKFzz//nDFjxuDh4cGWLVteenp9nfR8xpMnT8bPzw8/Pz9jS55Wq8Xa2jrJv5vXxZmZ7+N5qf0cEyX3s8iq2PIbSQIE1tbWrFy5kho1avDFF1+wYMECALp37864cePYtGlTskmAn58fhw4dYuDAgcYbc48ePfjwww/ZvHlziklAZGQkBQoUSHafm5sbQUFBhISEpNilkHiTTmyFSHTnzp1kj0/pCapjx44ULVqU9evXY2dnx6VLl4yzHBJjiY+PZ926dSm2hCSqWrUq3333HZBQQGXAgAGMGDGCwMDAV573vNS8d1dXV548efLSZ+jj42PcnxYpfTYnT56kYMGCSRIASEgInz8nNTGn5Qk2M6T0emn5eaZ0HTc3NypUqJBk5HtK+vTpQ58+fVAUhf3799OjRw++/vprFixYkKbPxM3NjZs3b6b6eEj4+XXu3DlJV96NGzcwGAxpijM1+9MjLZ8jpPwzzYrY8hsZEyAAcHd3Z9CgQSxfvpxHjx4BCf3948aNY+XKlcbpRokMBgOTJ0/G2tqamTNnGreXLFmSiRMn8uOPP3LgwIGXXufatWuMHDkyxTgGDBiAubk58+bNS7Jdp9MRGhoKJDwVFC9enEOHDiXZ/9tvv6XpPZuZmTFo0CA2b97Mzz//jIODAz169DDuHzRoEGq1moULF750bkhICBqNBuClvugyZcrQpk0boqOjjU9dp0+ffu0fvNS89169egHw008/GfcrisKKFSsoUaLES33Yr2Nra4u1tTURERFJtjs6OhIZGUlYWJhx2+7du/H19U1zzCm9RlZJ6fVS+/N8laFDh7Jnzx6uXr360r7nfw+e/1qlUtG+fXtKlChBVFQU8GzMTGo+kzfffJN///3XOK0w0ePHj1M8x9HRkXv37iXZtmTJEuMUx9TG+br96ZXaz/FVsiq2/EZaAoTRp59+ysaNG/nmm2/49ttvgYQmt4iICDp06MDIkSNp3Lgx4eHh/PLLL/j6+rJjx46X+uHmzJlDbGwsnTt3ZsCAAbRq1Qpzc3NOnjzJL7/8Qs+ePVOMoUKFCixbtozRo0dz584dOnToQEhICJs2bWLFihU0bNgQgClTpjBp0iScnJwoV64cf/31Fz169OD06dNpes9Dhgxh0aJFfP/99wwdOhQbGxvjvvLly7Nq1SreeecdvLy8aNu2LQaDgf/++489e/Zw5swZbGxsWLJkCcePH6dLly6ULVsWHx8fvv/+eyZOnGh8glm4cCGbN29+qQsjre+9Vq1aTJ8+nSlTphAQEED16tXZtm0bhw8fZvv27WkuGatSqWjVqhWrV6+mSJEi2Nvb07BhQwYPHsycOXPo1q0bb7/9Nnfu3OHAgQP07duXixcvpinmlF7jdXUC0iul10vtz/NVJk6cyOnTp2nWrBmjR4/G3d2dhw8fcvDgQQoUKMCWLVuAhO62Jk2a0KhRIxwcHNizZw+PHj3i3XffBRIGitrY2DB16lR69OiBubl5inUCJk6cyJEjR/Dw8GDcuHFUrFiRK1eucPjwYc6ePZtsnOPGjWPIkCF88MEH1K1bl927dydbu+J1cb5uf3ql9nN8layKLb+RloB8qHfv3saBZM+rXLkys2bNIigoyLi2gLm5OStXruTYsWM4OjqyZ88eLl68yIgRI/D29qZ169YvXcfMzIxFixbx33//UblyZY4cOcLhw4cpUaIEBw4ceO0T+7vvvsulS5coX748Bw8eJDw8nA0bNhgTAIAPP/yQ9evXc//+fby8vPjmm2/o0qULnp6exj5jKysrPD09KVu2bIqvVbduXd577z169uzJqFGjXto/dOhQrl69StWqVTly5AheXl7Uq1ePq1evGrsl5s+fz8KFC9FoNOzZs4fw8HB27NjBrFmzjNdJzeyA1L73L774gn379qHT6fj333+pUaMGXl5edOzY0XhMSu/dyckJT0/PJH2m69atY+DAgezdu5dNmzZx9+5dihYtyn///UfLli05cOAAdnZ27Nq1i2bNmr00ziE1MSf3Gq+jVqvx9PQ0JpkZfU+Qup/nq35vzM3N2bp1K5s2bSImJoa9e/cSGhrKhAkTkty4rl+/TpcuXbh58yYHDx6katWqXL9+nXr16gFQvHhx9u3bR8GCBfnzzz/ZtGlTigmihYUFf//9N7/88gshISEcOnSIUqVKJWkJc3V1xdPTEysrKwDeeOMN/vnnH/R6PceOHWPAgAHMmDGD3r174+7unuo4X7f/xde1s7PD09PzpUJGLVu2TNI9mNrPsUKFCnh6er7UgpGa2ETqqJRXPZoIIYQQIs+SlgAhhBAin5IxAUKIbHfy5MlXzpwoW7Ys9evXz8aIhMifJAkQQmS7o0ePcubMmRT3t27dWpIAIbKBjAkQQggh8ikZEyCEEELkU5IECCGEEPlUvhgTYDAYCAsLw9raOttLmAohhBDZTVEUYmNjcXJyMpZ1T06+SALCwsJeuRyuEEIIkRcFBwe/cmn3fJEEJJbKDA4Ofm1ZUCGEECK302g0FCpU6KVS0S/KF0lAYheAjY2NJAFCCCHyjdd1gcvAQCGEECKfkiRACCGEyKckCRBCCCHyKUkChBBCiHxKkgAhhBAin5IkQAghhMinJAkQQggh8ilJAoQQQoh8SpIAIYQQIp+SJEAIIYTIpyQJEEIIIfIpSQKEEEKIfMokCwjFxcURGhpq/L5gwYJYWlq+9jy9Xk9ERAROTk6vXRRBCCGEEK9mkpaAgwcPUrt2bWrXrk3x4sU5ceLEa89ZuHAhBQsWpESJEpQpU4Zdu3ZlQ6RCCCFE3mWSJKBTp048fPgQPz+/VB2/b98+PvnkE/73v/8RExPD9OnT6du3L/fv38/aQIUQQog8zCTdAWm1Zs0aevToQcuWLQEYNWoU33zzDZs3b2bChAkmjk7kRQ+jHxKtjU7VscqjJxCjyeKIhBB5lWIwcOvcKVxcXWnU6c1sfe1ckQRcuXKFgQMHJtlWs2ZNrly5kuzxWq0WnU5n/F6jkT/Qud3fSy9x90qwqcMQQohMp4u7ii7mJCrzezRoPwi1mVm2vXaumB0QExNDgQIFkmxzcHAgOjr5J7U5c+Zga2tr/K9QoULZEabIQpIACCHyIl2cF7qYfwAwsyiXrQkA5JKWAEdHxySzCQBCQkIoVapUssdPnz6dKVOmGL/XaDSSCOQRY1e0yfLX+PnKzyw8v5AR1UYwsf7El/YrisKT7xYSvHIlmJtTaskSCrRpneVxCSFyj9h4HTcunSb0yl4c7h/DPf4Ktqo44/54xZyPTxYhONKG2mVK0GLQMBr26pftcebIJCA2NpawsDCKFSsGQP369Tl+/Lhxv16v58yZM/Ts2TPZ8y0sLLCwsMiWWEX+E/T90oQEwMyMkt8tkARACIHeoHDj9i0eXNyDlf8RKkefp7bquYdXFdw1L0tQkabYV+3A+s37WLxvOWqVil7D3jJJAgAmSgL0ej1PnjwhLi4hKwoJCeHhw4c4OTlhbW3Ntm3bGDRoEIqiADBmzBgaNWrEkiVL6NChA99//z1mZmb079/fFOGLfCxo+XKCfvgB1GpKfvsNDh06mDokIYQJKIqC7/0g7pz/B8XnAGXDT1ONe1RLPEAFISpnAgo2wrxCW1wbdsG1UClcgU8/fJ85S5cDMO6tYXiOet9Ub8M0ScD9+/dp0KABAEWLFmXMmDEA/Pjjj/Ts2RMbGxuKFi1qPL5WrVps376d2bNns2DBAqpXr87+/ftxcHAwRfginwpauZIni5eASkWJefNw6NzZ1CEJIbLRw7AYrp4/SuyNfRQNOkkNww3Kq54NQo/FkrsF6qAr60GJul0oWLYWBV8obPfJB2OZu/QHACa+8xYLflqdre/hRSZJAkqXLs3Dhw9T3N+zZ8+Xmvo7duxIx44dszo0IZIVvHYtTxZ8ByoVxb/6Csfu3UwdkhAii4XHaLnodZmwK3txfHCMWtpLtFVFGfcbVCruWlcmumQLXGp2oki1llQ2t0rxelPHjmbeDysAmPTuSL79cWWWv4fXyZFjAoTISULW/8rjr+cBUPyL2Tj17mXagIQQWSJWq+fibX/uX/wH63tHcI85j4f6wbMDVBBkVpSgok0pULU9Jep0wtUudYPO/920gSUrVwHw0ah3mb/8x6x4C2kmSYAQrxC6aROP5swBoNismTj17WviiIQQmUWnN3D5XjC+Fw+D7yHKRZymAd6YqwwJB6ghRmXLg4INsKjYhuJ1u+JSuAIuaVy75sLuHfz312+807IR9lVrMWfJsix4N+kjSYAQKSh16AYPVyU01xWdPh3nFwpWCSFyF0VRuP0oEq/LF4i9uZ/iQSepz1Xqqp4WlFOBHjUB9jXQlWtNsTqdsXVtSHmz9N8q//zpB+78m7DWzYiPplG7Y9fMeCuZRpIAIZLR6rKB6ruOAlBkyhQKDsneUp5CiMwRGKbh7DVvwq7sw+nBceob/sNTFZSw8+kDfZBlKaJKtcSlZifsq3hQytoxU1574jtvsXj1WgY1qsVHs+dQq32XTLluZpIkQIgXuBy9xuidBlRA4UkTKTRiuKlDEkKkUmh0PKdvPyDA6xDW/oepEXeRHqo7qFUJU85RQbTageCiTShQrQPO1Tvg4lQGl0yOY/xbw1m8Zh0A5Ro2z5EJAEgSIEQSEXv2UOH7XaiAW33r4f7OO6YOSQjxCjHxOs7eCeGW1xnwPUjFqLO0VN14Vp1PDVoseORcG4tKbSlUsyN2xWthp8668rzjRgzl+7XrAZgxYRyff7c4y14royQJEOKpiH37CJw0GZVB4Y9mKux61TV1SEKIF2j1Bi4HhHHh6k1ib+6nVOhpmqq8aKUKSzjg6Yo4QXYVMJT1oFDNjliUa0ZxS7tsie+DYUNY+suvAMyc+CGzFizKltdNL0kChAAiDxwkcOIk0OsJ7N2ILZXPMcLUQQkhUBSFGw8jOXXzHqHXDlHw0XEaKZd5R30v4YCnN/0oi0LElGqBU42OWFZsg0uBYtke6/i3hrH0l19RAbM+msiM+QuyPYa0kiRA5HtRR44Q+OGHoNVScMQITnYpBBfOmzosIfKteyExHL/9GP9rp7C5d4R62osMVt/CSqVLGMyngniVFeFFGmJftT02VdphX6Qq9mmcupeZTv+1Be75YGNpwbQPx/HZ/G9NFktaSBIg8rWo48cJeP8DFK0W5yFDKPLxR3B1janDEiJfCY6K44RPMNeuXwHfQ1TTnKeD+goFE6vzmYEBFaGO1bCo1Ab7qh2wLN2Iwq+ozpedTv25meOb11O6kDP7t26habdepg4p1SQJEPlW9KnTBIwZixIfj9OggRT9ZBoqEz5JCJFfRMfpOHMnhLM37xJz6yDlws/QXH2F7onV+Z6O2YuyLo6+XGscqrVHXa4VzqmszpedRg0eQKTvLeqWLUWn0eOp1qqtqUNKE0kCRL4Uc+4c90aPRomLw6lfX4p99pkkAEJkkXidgf/uhXHi9kOeXD9BkScnaKr2YqLqaXW+p3eiODM7Yks1o0DV9qjLt8G+UHnIwf8u3xnQj1Vb/sBMrWLEpI9zXQIAkgSIfCjmwkXuvfseikaDY+/eFPv8c1RqtanDEiLPMBgUrj+M4PjtJ3jfuIx94FEaK5d4S30NB5XGeOcxYEZE4brYVmmHecW2WJWsh5WZhWmDT6W3+/Xh5z/+QqWCOZ9MpV2/QaYOKV0kCRD5iubyZe698w6GmBgcunen+JdfSAIgRAYpisLd4BiO+wRx6aYvqjtHqK29SBczL0qpgowj+AGi7ctiUakNlpXaoS7bHIdMqs6XXQwGAyP792XN1oQEYN5n0/no8y9NHVa6SRIg8g3Nlav4vz0SQ3Q0Dl06U2LuV6jMsq5giBB52ePIWE76BHPq1n2ibp+giuYczdVXGJRYne/p3SXewhFDuVZYV24H5Vtj51TGtIFngMFg4O3+nqzdug2VCr6Z+RmTZs42dVgZIkmAyBdib9zA/+23MURGUqB9e0rMm4fKXH79hUityFgtp31DOO79hMBbFykdeormai8+Uz+tzvf0n5NeZYGmREOsK7dFVb41lsVrQRZW58suiqKwY8VSNv9vJ2qVim9mzWDijFmmDivD5K+gyPNib93Cf8RbGMLDsW/dmpILvkVlkTv6HYUwlTidngt3wzjhE8SVm7dwfnSCZiovRqm9KKoKg+f+CcU6V8aqUltUFdpi5toEm2yqzpddFEXh2KZf8Dn8D6M8GuPasj0fTv/M1GFlCkkCRJ4W5+OD/4i30IeGYteqJSUXL0JlaWnqsITIcfQGhWv3IzjmHcS52wGo/E/SSLlEF7UXk9T3ktwt4m0KY1ahDWYV2oCbB9YmqM6XXQwGAz/PnU345XOo1GpGff4VlRo3N3VYmUaSAJFnxd25w93hw9EHB2PXrBmllixBLQmAEEDC0+2doGiOewdx4vYTQn3PUUf7H83VXrylvomVmc54rN7MGsW1GeYV2kD51lgWqZqjp+5lFoPBwJs9u7Hp790MalyHmYu+p1KjZqYOK1NJEiDypHh/f/yHj0D/JAjbxo0ptWwpaqucUV1MCFN5FBHLce8gjnsH4+N9nUrR52ih9uJL9VUKqSKNTfwKKrRFa2FRsQ24tcasTGPIIdX5sovBYGBw9y5s3rUXtUpFk55981wCAJIEiDwoPiCAu8OGo3v0CNv69Sn9wzLU1tamDkuIbBeu0XLKN5gT3kFc9L5HkeCzNFd7MUbtRXn1gyT9+roCJZ8+6bdBVa4VFjmwOl92MRgMDOrWmS27/0GtUrHk668Y+/FUU4eVJSQJEHmK9v59/IcNR/fgATZ16lBqxQrUtramDkuIbBGr1XP+bijHvYM46f0Y9f3zNFd50c3Mi89U3phbGozH6i3sUbu1RFU+4WnfPIdX58suBoOBAV068sfe/ZipVSyd9zWjJn9s6rCyjCQBIs/QPnrE3eEj0AYGYl2rJqVX/oSZfd4apSzE8/QGBa/A8IQm/ttPeOJ/nUbKZVqovRilvoaDZYzxWEVlhqFkQ9TlE/r1zUrWg1xSnS+7KIrCsF7djQnAsvnzeG/SR6YOK0tJEiDyBO3jx/gPHYbW3x/ratUos3IlZvb2pg5LiEylKAo+T6I4djuI4z7BXPO9S634hMF88828KGUelOR4g3N51BVag1trVOVaoMpl1fmyk6IoHFz3E4VjIyhgbcW3X83h3QmTTB1WlpMkQOR6uqAg/IePIP7uXazc3SmzehVmDg6mDkuITHE/TJMwgt8nmDO3H1A65got1Jd5X32FGqo7qC0V47EGa2fUbq3g6dO+OhdX58tOiqJwYM2P/Lf3b8oULsTJ/fuo1qyFqcPKFpIEiFxNFxKC/4gRxPv6YlWpEmV+Xo2Zk5OpwxIi3cJi4jnpE8xxnyBO3A7CPOQmLdRe9FB7MUd9A1vLOOOxipkllG4E5ROe9tV5pDpfdjIYDHh2aItVZCiNK5Slx+TpuNVpYOqwso0kASLX0oeF4f/W28Td9sayfHnKrPkZc2dnU4clRJpo4vWc9QtJuOl7B/Pw/l2aqa7QwsyLD9ReFLUKS3K8UqQqKrfWUL41KtemkMeq82UnvU5H7/Zt+N+ho1iYmTFxzrx8lQCAJAEil9KHhyckADduYFm2bEICUCj/TmkSuYdOb+BSQDgnvIM45h3Edf9H1Fau0Vx9hfnqy7hb3UtyvGJfFJWbR0ITv5sHqjxcnS876XU6erVrzd+Hj2GuVrNy8UIatuto6rCynSQBItfRR0biP/IdYq9dw6JMGcqsW4tFkSKmDkuIZCmKwq1HUU+L9ARx5k4QZeJ9aKn2Ypzai/rmN7FSPavOp5jboCrbDBKf9vNJdb7spNfp6NHGg11Hj2NhpmbV90sYOnqsqcMyCUkCRK6ij4rm3jvvEuvlhUWpUriuW4tF0aKmDkuIJAJCYzjhHcyxpwP6LKMCaW7mRS+1F/PVVylkFWk8VkEFxWsb+/VVpRuBhRS3yip6nY7urVux+9gJLMzUrF62lCHvjTZ1WCYjSYDIVe6Neg/Nf/9hXqI4ZdauxaJ4cVOHJAQh0QmD+RJu+kEEBwfRWH2d5movPlR7Ud76QdITHEuDm0fCk345D8jH1fmyk2IwsGHelxw+cxYLMzVrfljGG++OMnVYJiVJgMhVNOfOY160KK5r12JZqqSpwxH5VHScjjN+IZx4Wof/5oNQaql8aK6+wrdml6lt7YMFeuPximUBVOVaPO3Xbw1SnS/bKQYD//z0PU8un2NMuxbU7T2AQSPfNXVYJidJgMjxDHHPpkSZFy6M67q1WJaR+c8i+2j1Bv67F5YwX987mIv3QihheEiLp0/6Tayu4aBKWp2Pkg2N8/VVUp3PpHQ6LQs/moj6/h3MLa34YN5XlKle09Rh5QiSBIgczRAfT8D7H4DZQADKrFuLZdmypg1K5HkGg8KNh5Gc8EkczBeCeXw4zdRX6KX2YoH5FUqrniQ9qWD5Z/365VqAVOfLEbTaeLq2bM7+U2cZ1LQeX//0M6WrSQKQSJIAkWMp8fEEfjie6KNHwSMhCbByczNxVCKv8g+O4fjTm/5Jn2AiomOop75l7NevaX0HNc+q82HjDOVaGW/8OLuaLniRLK02ns7Nm/LvmfNYmpnR692xkgC8QJIAkSMpWi2BkyYRdfAgZo7yRCUyX1BUHCd8gjl+O4jjPkEEhMZQURVIC7UX36i9aGJ9HRuedUWhtoAyjZ/d9KU6X46m1cbTsVkTDp69gKW5GRtWr6Lv0OGmDivHkSRA5DiKTkfgRx8TuW8/agcHSv+8Gr5/ZOqwRC4XFafjzJ1gjt0O5oRPEDceRlKYMJqprzDBzIsW1lcoQmjSk4pUNc7XR6rz5Rra+Hg6NG3MofMXsTQ3Y+PPq/EcMszUYeVIkgSIHEXR67k/dRqRe/agtrenzOpV2FSrBkgSINImTqfnon9Ywgh+n2Au3QvD3BBLQ/UN+qiv0NLKiyoq/6Qn2RV59qTv5gEOMgU1tzHo9fRu15pD5y9iZW7OxrU/0+eNIaYOK8eSJEDkGIrBwINPphPx99+obW0pvfInbGrUMHVYIpcwGBSuPYhIqMznE8zZOyHEarVUVd2lhdqLSeZe1FffwhLts5PMbRKe8Mu3ThjJL9X5cjWDXs+upQuobGPGKXs7Vv30I70GvWHqsHI0SQJEjqAYDDyYMYPw7dtR2dhQeuVP2NapY+qwRA6mKAp+wTFPl9lNqMwXFqOlOME0N/NivvoyLWyu4aREPHeWKqEvP3G+vlTnyzP0Oh27li7g1smjlC1elPOnTuAqgwBfS5IAYXKKovBw9mzC/9iKytqa0itWYFuvnqnDEjnQ44jYhMF8T+vw3w+PxZ4YGquvM17tRWubq7gqgc9OUEhSnY9yHlKdLw+K1Wjo0LQxxcwVWlSrgucnsylRqYqpw8oVJAkQJqUoCo/mfEXYps2orKwo/cMy7Bo1NHVYIoeIiNVy2jfEeNO//TgKM/TUUvnQV30FD+sr1OI2ZonV+RTAsgCUa/FsQF+hCtLEn4fFajS0bdyQE5evYGNpwawfVkoCkAaSBAiTURSFx/PmE/rrr6gsLCi1dCl2TZuaOixhQrFaPRf8Q5/e9IO5HBCGQVFwVT2ihdqLj62u0FR9DTsl+tlJxup8T/v1pTpfvhGr0dCmUQNOel3FxsKCPzZuoGqDRqYOK1eRJEC81t9LL3H3SnAWXb0BeDRI+HJDPGw4kOoz/cL9WHRhEaGxoa8/OA0eRj/M1OuJlOkNClcCwznuk1CO96xfCHE6A45E0VR9lYHmXrS2uEpRw3OzQxSgoNuzfn2pzpcvaWKiadOoIaeuXMPGwoI/N22kU5++pg4r15EkQLxW1iUAaeNa/VlfbpAmiFH7RxEYFfiKMzKmuL1MD8tsiqLg8yTaWI73pE8wEbE6LNBRT32LD9RetLe7RiW9N6rE6nwGpDqfSEITE03rhg04ffU6NpYW/Ll5E5169TF1WLmSJAEi1cauaJMp13nyww8ELfkezMwouWABDp06pul8jU7DuAPjCIwKpFqhakyuPxlVJvf52pjb4F7QPVOvmV89DI99Om0v4Wn/YUQsoFBRFUhftRftba9Rl2tYGTQJJ+h5Vp3PzSPhiV+q84mn9Dota2d/yhVvH2wtLflry2Y69Oxl6rByLUkCRLYK+mllQgKgVlNi3rw0JwB6g54pR6bgFeRFSfuSLG27FBcblyyKVqRHeIyWk74JVfmOeQfh+ySh/96FcJqpvWhnc40WZldw0gUlnGB4emJhd+Oqe1KdTyRHp9Xyv4Vziblzm3GdWtP6rdG079HT1GHlapIEiGwT/PMannz3HahUlJj7FY7duqb5Gt+c+4aD9w5SwLIAP7T9QRKAHCBWq+ecXyjHns7XvxIYjkEBK+JpqL7BEKsrtLO8Rmmtb8IJCqBDqvOJNImOiuLrcaNwig7D2r4AH379JUXLlTd1WLmeJAEiW4T8sp7H8+cDUPzLL3Dsmfbsff219Wy4vgELtQWLWy/GzUlWFDQFnd7A5cDwhHK83sGc9w8lXmdAhYGqqruMNr9CR5vrVNVew1yJTzhJS9LqfG6toWg1mbonUiU6MpJWDepx/uZt3mzRkAW/LKFIWfn3nxkkCRBZLvS333j01VcAFPv8c5w8PdN8jf139/PN2W8A+KLZFzQo1iBTYxQpUxSF24+jjNP2TvsGExmnA6A4wfQy86Krww0aGC5jqwtLOCkejNX53J5O3ZPqfCIdoiIiaNWwHhduemNnZclbU2dIApCJJAkQWSp0yxYefj4bgKKffYrzgP5pvsblJ5eZenQqCgrj6oyjq1vauxFE2gSGaRLK8T6tw/8kMmFJXTs0NFZfo4v9DVqaeVE47ukCPE8f+HEo9XS+vlTnExkXFRFBiwZ1+e+WD/bWVvzvrz/x6NTF1GHlKZIEiCwT9udfPJw5C4Ci06ZS8I20L+RxL+IeHxz4gDh9HJ4VPRlZY2QmRykAQqPjOen7rByvX3AMgLE639u212lvdY1ysddRK7qEPn0dUp1PZJnIiHBa1K/Lpdu+2Ftb8ff2bbTq0MnUYeU5kgSILBH+v//xYPp0UBSKfDSZgsPSvpZ3WGwYY/4dQ0hsCM1KNGN64+mZPhUwv4qJ13HWL5QT3gkj+K89iEBRABKq871tdY2udjeoFv8fVrqohBH8GhKq85Vq+Kxfv1R9qc4nMp02Po6uLZsbE4BdO7bTon3aZhKJ1JEkQGS6iF27uD9lKigKhcePp9Dbb6f5GnH6OD48+CF+EX5Ucq7Et62+xUItN5v00uoNXA4I49jtYI77BHHRPxStPqEYjyNRdDO/Rk+HWzTU/4dD3P2Ek2KenlzQ7Vm/vlTnE1lMGxfLtm++pFGxgvj4O7J5y+80b9fe1GHlWZIEiEwV8c8/BH70MRgMuIwdi8uo99J8DYNi4NNjn3Lh8QWK2BZhWdtl2FvaZ0G0eZeiKNx4GPl0md2EwXzR8QmL7CRU57tNH+ebNFdfoXj09YTqfIk3fanOJ0wkThPDjm/n4H/lEm5lSvPfT+soXKasqcPK0yQJEJkm8sABAidOAr2eQu+9h8v7Y9N1ncUXFrPHbw92Fnb80PYHitkVy+RI86Z7ITFPK/MFc9IniKCoxNF6CdX5ejnepJ3VNcpHX8JcH5PQvA8vVOdrDcVrS3U+ke3CQ0NoUb8elZztaVu3Fv1nzKVQqdKmDivPkyRAZIqow4cJ+HA86HQUfPstCo//MF3991tubuHnKz9jpjLju1bfUblg5cwPNo8IjorjhE/w0zr8wfiHxBj3uRDOULsbdLO/SY24C9jEPoY4Ev6Dp9X5njbxS3U+YWJhwcE0b1CPq3fu4m9rw7e/bpYEIJtIEiAyLOrYcQI+GAdaLc5Dh1Bkcvpq+R8JOMKc03MAmNlkJk1LyrLCz4uO03HmTojxaf/6gwjjPivi6Wh9G0/H29TX/0fBqFsJNfjDnx4g1flEDhUWHEyz+nW55uePg60Ne3bvomyVqqYOK9+QJEBkSPSpUwSMHYsSH4/z4MEUnTYtXQnA9eDrTD48GYNi4N2a79K7Yu8siDZ3idcZ+O9e2NN+/SAu+oehMyQM5lNhoLb5PQYU8qG56jIlIy+h1sc9u+lLdT6RC4QEPaF5/Xpcv3sPR1sb9u7eQ6OWLU0dVr4iSYBIt5izZ7k3egxKXBxO/ftT9NP0TeF7EPWAsf+ORaPT0NWtK+/Xfj8Los35DAaF6w8jOOEdzDHvIM76hRDzdDAfQElVMANdfGhndZUKUeexiAt5dtOH56rztYbSjaU6n8jRQoKe0KxeXW74B+Boa8O+vXtp0LyFqcPKdyQJEOkSc+EC/u+NQtFocOzTh2KzZqJSq9N8ncj4SMb8O4Ynmic0KNaA2U1n55taAIqi4B8Sk7DwjncwJ32DCYmON+63Q8MQ5zt0s79BtdgL2Ef6QhQJ/8HT6nweT6fueUh1PpFrxGtiWDfrEwIePcbJzpZ9/+ylftPmpg4rX5IkQKSZ5tIl7r3zLkpMDI49e1D8i9npSgC0ei0TDk3AO8wbN0c3FnosxNLMMgsizjmeRMY9HciXMJgvMExj3GeGnvYFAvB08qa+4T8KhV5CpdE9G8VvaQ9lWzxbbleq84lcKF4Tw9a5szA8ecDEXp3o8eHH1GvSzNRh5VuSBIg00XhdwX/kOxiio3Ho0oXiX32Fyizt08kURWHWyVmcfnCaQtaF+KHdDzha5b0iNJGxWk77hnDcJ+Fp/+ajyCT7q9sEM9jFh+ZqL0qGnsUsPgKePN2pUkt1PpGnPHn4kDnvv0tpMwMFChXm7Rlf4VRMBqmakiQBItVir13Df+RIDJGRFOjQgRLz56UrAQBYcXkFO3x2YGNuw7K2yyhpXzKTozWNOJ2eC3fDjE/7lwLC0T8dzAdQ1CKGN4vepa3lVSpEnsUy8t6zmz48V52vdcJTv41Ttr8HIbLCk4cPaFa/HrcDHzC8TXMWLlmFU1GpAWJqkgSIVPN/620M4eHYt21LyQXfojJP36/PDp8d/PDfD6hVaua3nE81l2qZHGn20RsUrt2P4PjTm/5ZvxBitQbjfmu1nsHF79PV7gbVYy9gF+yFKujZfqydnhXpkep8Io96/OA+zRrUxzvwAYUK2DP+6wWSAOQQkgSIVNOHhWHfqhUlF36HyiJ9zdKnH5xm5vGZAExtOBWP0h6ZGGHWUxSFO0HRHPcJ5vjtIE76BhOu0T5/BB0Kh9HH8Tb19RcpFHQOVWg0hD7drbYA12ZSnU/kGw8DA2jesAE+9x9SqIA9Bw4coGb9BqYOSzwlSYB4pTjfO8av7Zo3p+SSxagt0zd4zzvUmwkHJ6BTdAytOpRBVQZlVphZ6lFErHEg3wmfIB6ExybZX8MpjsEuvjRTeVEy9BRmkQ/h+a7/xOp8bq0T5u5byToIIn94GBhAswb18X3wiEIF7Dl48CA16tU3dVjiOZIEiBTF372L//DhUDXhyb3U0u9RW1ml61pPYp4w5t8xRGojae/ankn1J2VipJkrXKPllG8wJ55W5vN+HJVkf3FbhcHFA2lreS2hXz/oKgQ8d4Bdkeea+D3AoUR2hi9EjhAbFUUXj1b4PniEi4M9Bw8eonrdeqYOS7xAkgCRrPiAAO4OH4Hu8WN4WsFTbZ2+4jMx2hjG/juWB9EPqFW4Fl81/wq1Ku1TCrNKrFbP+buhxnK8XgFhPDeWDztLFZ4lw+hmd4OqmgvYPTyDKjDu2QHm1k+r87WR6nxCkJAA/DHnU9pXKE1EdBTbd+2mWu26pg5LJEOSAPESbWAg/kOHoXvwAJt6GcvcdQYdHx35iOsh1yldoDRL2izB2ty0lez0BgWvwHBjOd6zfqHE654N1jNXq+hQSkcfp1vU0/1HwUcnUT0ISnoRqc4nRLLCQ4LZMX82j+/4UKliRc7/sgXHwkVMHZZIgcmSAK1Wy86dO7l37x41atTAw8PjlcdHRkayZ88eHj9+TLly5ejYsSNm6ZyeJlKmffiQu8NHoL1/H5tatSj9448w+XS6rqUoCl+f+ZojAUdwsnJiebvlFLQumMkRpy4OnydRHPcO5rh3wmC+yFhdkmPqFTNnQGE/mqq8KBF8CvWT20mn7iVW50tcgMfOJTvfghC5QqD/XZo3bECdEkXo2rwJ/WfOpUAh+beSk5kkCdBoNLRp04aIiAgaN27MV199RdeuXVm1alWyx/v4+NCkSRPc3d2pXr06S5cuZcaMGRw5cgRbW9tsjj7v0j56jP+w4Wjv3cO6enVKr1qJmX36l5hde3Utm29uxlJtyfdtvsfVIfumvz0I1xhv+id8gngUEZdkv1tBK/oWf0Jby6u4RZ7F4v45CHsuMTBW53u63K5U5xPilQL8/GjWqAH+j4PQxMWx4n+7JQHIBUySBPz4448EBgZy7do17O3tuXXrFu7u7owYMYJmzV4uH/njjz9Srlw5Dh8+DCS0CpQsWZKdO3fSr1+/7A4/T9IFBeE/YgTxd+9iVdWdMqtXYVagQLqvt8dvD9+d/w6Ar1p8Re0itTMp0uSFxcRz0ifYWJnPNyg6yX4Xe0t6lI6ji90NqmrOYxtwHHyeW31HpYZSDZ7160t1PiFSzf+OLy0aNcL/SRBFnRw4fPQYRUqWNnVYIhVMkgTs3LmTPn36YG+fMFWqUqVKNG7cmJ07dyabBFhbWyd54reyssLc3BwbG5tkr6/VatHpnj3VaTSaZI8TCXQhIdwdPpx4X1+sKlWizOrVmDmmv4TvxccXmX50OgAT602kY9mOmRWqkSZez1m/Z+V4r9wPR3luMJ+9lTmty1jQy8mHurqLOD04hurO3aQXkep8QmSY/x1fmjdqyL0nwRR1cuTI0WNUql7d1GGJVDJJEuDv70/nzp2TbCtTpgz+/v7JHj9+/HhOnz7NoEGDqFatGgcOHKB///506dIl2ePnzJnD559/nulx50W60FD8R7xFvLcPlhXKU2bNz5g7O6f7en7hfnxw4APiDfEMqDyA4dWGZ06cegOXAsKfTtsL4sLdMOL1zwbzWZqpqV/aDs/C92mq8qJo0EnUARfh3ovV+Vo9u/E7l82U2ITIr+76+tC8USMCgoIp5uzEkWPHqFg191YAzY9MkgQoivLScrFqtRrl+Ue55zx+/JjIyEjUajUPHz4kJiaGoKAgIiMjcUzmiXX69OlMmTLF+L1Go6FQIVlm9UX68HD8336buJs3sSxXDtc1azDPwOcUEhvCmH/HEB4XTstSLZnacGq6lwVWFIVbj6KMffqnfEOIinvWuqNSQfUSBehRMoo2FlcpF3EGs7vH4eFz3QBqCyjT9Gm/vlTnEyIzxYSHsenLGYRHRVHc2Ykjx49Twb2qqcMSaWSSJKBkyZIEBgYm2RYYGEiDBsmXkvzggw+oXr06P/30EwAGg4EGDRrw7bff8sUXX7x0vIWFBRbpLGubX+gjI/Ef+Q5x165j6epKmbVrMS9cON3Xi9XF8sGBD7gXeQ/3gu580/IbzNVp+/UKCI3hhHcwx7yDOOETTFBU0sF85Vzs6OCqprPtDdw157G6ewS87ie9iFTnEyLLxYSHsWX2J5hHhzN1QG8GTpuJW+Uqpg5LpINJkoD27duzdu1avvrqKywtLbl//z4nTpxg+vSEfuSbN2/yv//9j8mTJwMQExODnd2zUepqtRobGxtiYmJMEX6up4+K5t7Id4j18sKidGnKrFuLRdH0z+M1KAY+OfYJl59cprhdcZa1XYatxetnbYREJwzmO/b0af9ucNKfZ5ECVni4FaCrkx/1dP9hH3AUrnolvYhU5xMiW/neusncD8dQxcGGQqXK0O+zOdg5pb8LUZiWSZKA999/n19++YW2bdvSqlUrtmzZQufOnWnfvj0AFy9e5KOPPjImAePGjWPo0KGEh4dTvnx5Tp48yeXLl/nxxx9NEX6uZoiO5t5776G5dAmLEiVwXbsGi2IZW81rwbkF7Lu7jwIWBfih7Q8Utk2+RSE6TscZv5CEfn3vYK49iEiyv4C1OU3KOdOtaDBNuUyhxydQeZ8EfTLV+dyeTt2T6nxCZBvfmzdo3rQpD0JCeadTGxZKApDrmSQJcHBw4OzZs/z6668EBATw+eef079/f+P+KlWqMGnSs9ryAwYMoFatWuzevZvg4GC6devGmjVrKJyB5uv8yKDRcG/0GDTnz2NerFhCC0DJkhm65sbrG/nl2i+Yq81Z2HohFZwrGPdp9Qb+uxeW0K/vHczFe6Fo9c/GfViaq6nv6kyH0npam1+hdOhp1HcOw50XqvMVq5lww5fqfEKYjPf1a7Rs1owHoWGUcinIJ9+vkAQgD1ApKY3Gy0M0Gg22trbExMSkOK0wrzPExnJv9GhiTp7CvHBhXH9dj6Vr6or3LBt1AICxK9ok2X7Q/yDjD43HoBiY03wO3cp158bDSE74BHHcO4gzd0KIjtcbj1eroEZJRzzK2dDRzodKUWcx9zsEQbeSvqBU5xMiR/G+fo0WzZrxMDSMUi6FOHb6NK5u5U0dlniF1N73ZO2AfMAQF0fA+x8Qc/IUZi4ulFm3LtUJQEquBF3h4yMfY1AMtHB5g39Ol+LzjfsJjo5Pclz5wna0LO9Ex4IPqR1/EWv/I3D+DBhSqM7n1hpcKkoTvxA5xO2rV2jRvDmPwsIpXbgQR09JApCXSBKQxynx8QSO+5DoY8cwK1gQ17VrsHIrl+7rBUXFsfPaVZbcGEe8Eos2rB67rlcHHgBQzMGaZhVc6FA8msaKF44PjsKNIxCbTHW+xH59qc4nRI4UGRJEj04deBQWTpnCLhw9fZoy5dxMHZbIRJIE5GGKVkvAxIlEHT6MmZMTZdaswapChdef+Jzn5+Z3WnSEG48fYVt2BWZW4eiiKmAZNgCPakVo7WpBK8vrFH2yD5XvQbjml/RCUp1PiFwlMjiILbOn0bNGJawsLfn730OUKlvW1GGJTCZJQB6l6HQETv6IqP3/onZ0pMyan7GuXOm158Xp9Fz0D3tamS+YS/fCmEDCQLwbj0KxK/MraqvHFLIow5J6w6gefQ6170E4eBEUqc4nRF7w0P8uO7/9krBHD6hWoxYzN+/ApoCDqcMSWUCSgDxI0eu5P2UqkXv3oi5QgDKrVmHt7p7ssQaDwrUHERx/etM/eycEjfbZYD4z9bO++Y6N/8eJMF+KYM4G/0sUuzXs2YWM1fk8Epr4pTqfELnStUv/4dGqFU3KlaJvx3b0/fRLbOzTv5iYyNkkCchjFL2eB598QsTOnajt7CizaiU2NZ4t5qEoCn7BMcZyvCd9ggmN0Sa5RqWi9jQt70LrUioaKJdYszKhJeBE2GlsDQaWPbhHsXgtFK7ybNU9qc4nRK539b8LtG7lwZOISLweBrH+45mSAORxkgTkIYrBwIMZMwjfvgOVrS2lV/6ETa1aPI6M5YR38NMbfzCBYUlXVSzpZEOzCoVoUc6e5pbeOD/YD74H4UJidb6/ADBTFL61dadKl+lSnU+IPObKhfO0bu1BUEQU5YoV5fiZszgULGjqsEQWkyQgj1AUhYefzyZ865+orK15NO0rNgVYcfzQYW4/jkpyrLOtBU3Lu9C0vDOtHR9TPPgUKt8DsPsU6GKfHWhuzYkyteFhwrfTG8+gRZX+CCHylsvnztKmTRuCI6NwK16U42fPUaxkKVOHJbKBJAF5gCZex41PZmL9959ozSyYWX84F8/rAD8AbCzMaFiuIM0qFKJlUS2Vos+i9t0MRw5BTHLV+RKm7t20L8jE/e8y5OmufpIACJHnPJ8AlC9RjONnz1G0RMYqiYrcQ5KAXEhvULh6Pzxh4Z3bQdTYvoYe3kfQqs2Y1Wg4XkUqUr+0E00ruNCijDW1DVew8NsKlw9C0M2kF3Mo+axIz3PV+R5GP2TMrjeI1ka/HIAQIk8If/yQHYu+RqvTUr5EMU6cO0+R4tLNl59IEpALKIqCb1B0wgh+74TBfBGxOlAU3rq2kx7eR9CpzTg9Ygrvd21LY2t/bO8dTejXP3k6zdX5ouKjGPvvWB7HPKZukbrZ/G6FENkh7NFDtsyeho02jlkj3uSNT7+gSPHipg5LZDNJAnKoh+GxT6ftJSy+8zAiNsn+0s7WvH/nX2rePgRmZpQd14MaTnthx/RXVOdrnfD1K6rzaQ1aJh2exK3QW5R1KMuSNkv4dfv5LHqXQghTuHjqJAumTKJOsYKUqOROn2mfY2X7+uW/Rd4jSUAWuxJ0hTvhd1J1bMQ/+4nz9idcozXO1XcBegDmajWOtuY42VjiaG2O450InA75oqjgoUcE3pFrIJKEn2gRVyhaFYpUBZdKYGmX8AK6YPDb88oYjgYe5cT9ExS0LsgP7X7A0cox3e9dCJHznD95nPbtOxAaHYNz72588MnnWNpIApBfySqCWSg0NpTWW1qjV/RJtne+/i6uYdWyLY7M8uIqgkKI3OX8yeO0a9+BsOgYKpcuyfFz5ylUpKipwxJZQFYRzAEi4yPRK3pszW1pU+bZDbTUydyXALhWL2TqEIQQGXDu+FHadehIeIyGKmVKcfz8BQq6FDZ1WMLEJAnIBoVsCjG3xVzj98s2HACee7KOeMD9QU0I97HhjEc3hq34xhRhCiHyqNNHjtCxcyfCYzS4u5bm2LnzkgAIQJKAnOHcamKDEn4Ujyo2MHEwQoi8JDjwHgP79SE8RkNV19IcP38Rp0LSsicSqE0dQL6njUV/8mfiws3RqdTo3CqaOiIhRB4RHHiP32d/wsB6NWhWo6okAOIlkgSYmtfvxAZEAip8HEvi6CSLdQghMs7by4stn08jOiyUuk2acPDseUkAxEskCTAlRYFTy9EEWwJw07kMTrYpz+EXQojUOHFgP/UaNWTX6fOUqV6LXh/PwMLK2tRhiRxIkgBT8jsKj6+iCUtYgvdGQVecbCxNHJQQIjc7tn8fnbp1J0ITy93oOLpN+kQSAJEiSQJM6dQKFAU0YQmFOm4ULIOztAQIIdLp6L69dO7enUhNLDXLl+PY+QvY2NqZOiyRg0kSYEo3d6GNtUYfEUuUtT0PbAvhKEmAECIdDv+zhy49ehIVG0ftiuU5ev4CDo5Opg5L5HCSBJiUgsa2OQDehVxBpcLZVroDhBBpc3jvbrr17JWQAFQqz9FzkgCI1JEkwMQ08a4AXHEqAyADA4UQafLYz5dDq5djrlZRp1IFjp69gL2Dg6nDErmEJAGm5NqMWO/7AFx1LI2luRobCzMTByWEyC0e+/ny+5efYq8yMH/caI6ckwRApI0kAdnN8GwxIUO9d4i9dg1UKm45l8bZ1gKVSmXC4IQQucX+v3cwZmBfYiMjcKvbgBGz52FfQOqMiLSRssHZ7fY/gBUAcfqyKFotims5YixsKC3TA4UQqbBvx3Z69etPTHw8bhUr8eHETzC3kK5EkXbSEpDdTi03fqnx8gIgvlJVQMYDCCFe75/t24wJQIOqVZi1YpUkACLdJAnITo+uwp3Dxm81ly4DEFGuMiBJgBDi1fZs+5Ne/RMSgIbVqnD47Dls7exNHZbIxSQJyE6nVyT5VnPpEgBBpSsAyPRAIUSK9vz5B30GDEQTr6Vx9aocOnNOCgGJDJMkILtEB8PlLUk2aQMCUNna8qBgCQApFCSESFbgreuMHPkOmngtTWpU46AkACKTyMDA7HJ+DehioWIHePhss02NGoTFJswYkJYAIcSLHty+yZ9fzWRokzpcDtfwx74DWNvYmDoskUdIS0B2UBQ4uyrh68ajk+yyqVmT0BgtgKwbIIRI4sLRI/wx51PiNTE0adeB7YeOSgIgMpUkAdlBGw2RD6BwFXBrnWSXTe1ahMXEA+AoUwSFEE/t2PwbTdu2ZfcFLyo3aUHXcR9hZi6NtyJzSRKQHWIjEv7faBS8UAzIpmZNwqQlQAjxnO2/baD/m0OJ0+qINrem09iJqM2kmqjIfJIEZAddLNg4Q80BSTZblCiBeeHChD5tCXCSMQFC5Ht/bVzPgKHDidPpaFW3NvtOnpY6ACLLSBKQXeoOA0vbJJtsatcCIFwjLQFCCNi6fh0Dh40gTqfDo34d/jlxCksrK1OHJfIw6WDKSpGPnn6hgobvvLTbplYtFEUxdgfIFEEh8q+/fl3P4LfeJl6np3WDuuw9fhILC2kdFFlLkoCs5PW0LoClHTiWemm3Ta1aRMXp0BkUbC3NsDKXPj8h8qN717y4+OcG7CwtaV6nKnuOn5AEQGQL6Q7IKvExcOWPhK+tHY2b9VFRxq+t3N2NrQBONtIKIER+dO/qZf78ehaOFuYs/XSKJAAiW0kSkFW8tjybFWD+rE8v9umiQQBqKysZFChEPrbp51WMGTIYXVwc1TzaMXDKp5IAiGwl3QFZQVHg1IpkdyWsF1DJ+L1xeqCdtAQIkZ/8tuonho0ag1avp0GTpnR8bxwqtTyXiewlv3FZwfcQPLkOti4v7UpcOTCRsSVACgUJkW9sXPkjw0aNRqvX06lZY6Yv/kESAGES8luXFRJXC6zZP8lmRVGMKwcmSpweKMsIC5E//PrTCoaPHoNWb6BT8yb8feioVAIUJiNJQGYL9oFbe8DMCmr0TbJLGxCAPiQkybbQaEkChMgvflm+jLfGjEWrN9ClRTP+PnhEEgBhUpIEZLbTPyb8v2a/hCqBz3mxKwAgTJPQHSArCAqRt3mfO83H06ah1Rvo2rIZOw4ckgRAmJz8Bmam2HD4b0PC141Gv7T7xa4AeDYw0FGmCAqRZ925eI6/F33N283rc09tw5q/dmAmawGIHECSgMx08VeIj4KyLaBYdYjwT7I7+SRAWgKEyMsObP+Ty1t+Qa/T0aZPP9qMGIXqhYXEhDAV6Q7ILAb9swGBjV9uBTDExRF7/fpLqwiGyhRBIfKs1YsX0qFPP3Zfuk6dTt0lARA5jiQBmeXmbgjzB+eyUKnTS7vjrl8HrRarCuWTbE9sCXCUKYJC5Ck/LfyW9yZOQm8wYFe8JB7D3pEEQOQ4kgRklsRWgIbvgfrlvr7ErgDrWrWSbA+TFQSFyHN+XPANYyZ/jN6g0Kd9G/7Yux+11AEQOZD8VmaGB5fB7yhYFoA6byZ7SGISYPNcEqA3KMY6ATIwUIi8YcW38xn78RT0BoW+Hdry+559kgCIHEt+MzND4rTAOm+AtUOyhyROD7Sp+SwJiIzVoihQwNocczP5UQiR261c+C1jP56K3qDQr1N7Nu/+RxIAkaPJ7ICMinrydMlgFTR8N9lDCkTq0AYGora1fTomIAB4NihQCgUJkfvdPn2Ce4f+wdnWhnatWrDxf7skARA5nvyGZtT5NaCPTxgMWKh8soeUuRcHgHXNmqiemxss0wOFyBtunT7O34vn4WhtydoFX0sCIHKNDP+WGgyGzIgjd9LFw9lVCV83HpXiYa5PkwCbFwcFGlsCJAkQIrdaPOcLxo18G4NeT8Oefen67vuSAIhcI12/qQEBAQwYMAAXFxc6duxIeHg4b7zxBjqdLrPjy9mu/gVRj6BIVSjXKsXDyhiTgJpJtj9bQVC6A4TIjRbPmc3Ez2ay98pNzCpUo/mgYTINUOQqaU4C9Ho9PXv2xNnZmZkzZ6IoCo6Ojjg7O7Nq1aqsiDFnUhQ4vTzh60ajXioClEhlUCgT8DQJqJk0CUhsCZDpgULkPt/NnsXEz2ZhUBSG9OrOuC/mSgIgcp00JwG3bt0iMjKS5cuX4+7ubtzeqlUr9uzZk6nB5Wj3TsP9i2BT8KUlg59XOgis4hUsSpXC3MUlyT5joSDpDhAiV1nw+Qwmz/ocg6IwrE9P1m7dJl0AIldK82+tRqPB0dHxpYxXq9VSoECBTAssxzv1tBWg/giwsEnxsIr3FeDlVgCQQkFC5EbfzPyUjz7/AkWBEZ69+fn3PyUBELlWmn9z3d3d8fX15eTJk8ZtWq2W1atX07Rp00wNLscKuwfX/wdqc2gw8pWHVgx8mgTUrvXSPpkiKETucunAP3yzcBGKAm/17cPPf0gCIHK3NNcJsLGxYdGiRbRv355y5crx+PFjqlatSrFixXjnnXeyIsac5+xKUPRQrS84lHjlocaWgFovJwGJ3QEyO0CInO/akQP8+9NS3mvViOgipflu9VpThyREhqkURVHSc+KNGzf4+++/CQ4OpmbNmvTv3z/Hro+t0WiwtbUlJiYGG5uUm+5TRRfH3x/9wF3Nyzf1tNpTzRKvwHC2jW1G7dJOGb6eECJr/LFqBX77/kYFNO3/Bk08B5k6JCFeKbX3vTS3BPj5+XH8+HHeeOMNqlSpkuz2PC06KFMSANfqhQiNDgZkiqAQOdlX06bw6dfzaVu1Al988SWN+wwwdUhCZJo0JwHe3t6sWbPmpZv97du3WbduXd5PAp4zdkUb49d/v9WR8if8CRzTg6ZvTUtynK29U7LnT5u5F5CKgULkVF98PJmZ3yxAASrUri8JgMhzUp0EaLVawsPDiYiIQKvVEhQUZNyn1+s5evQoRYoUyZIgcxO1pWWKN/3nafUGIuN0qFUJCwgJIXKW2R9PYtY336EAo98cxA/rN5o6JCEyXarvPocPH6Z9+/bG7wsXLpxkv4uLC3/99VfmRZbHPb+EsFotBUaEyElmTZ7A5wsWATB26BssXferaQMSIoukOglo2bIlDx484MSJE/z444+sW7fOuM/CwoJChQplSYB5lcwMECJnmvfJVGMC8MHwISxZ84tpAxIiC6U6CbC0tKRYsWL06dOHPn36ZGVM+UKY1AgQIse5tG830V7ncLG3ZXC/viz+ed3rTxIiF0t3Z/Tly5c5ffo0jx8/JnGWYbly5VI9MPDcuXN8/fXX3Lt3jxo1ajBr1ixKlSqV4vEajYbFixezd+9ebG1tmTRpEm3atEnx+Jwu1LhugLQECJETXNq3i/2rfsDR1pq/1q6muedAU4ckRJZLVxIwY8YMli9fjpubG35+fhQvXpxbt24xderUVJ1/+/ZtPDw8GDduHGPHjmX58uV4eHjg5eWV7HxGrVZL+/btUavVfPTRR9jb2zNv3jyqVatG0aJF0/MWTE5WEBQi5/jkg7Gc+XcvnapXpvWwd6nXtaepQxIiW6Q5CXj8+DGLFi3i+vXrXL9+na+//pp9+/YxaNAgypYtm6prfP/999SrV4+vvvoKgGbNmlG0aFH++OMPhgwZ8tLxq1evxs/Pj9u3bxuThFatWmEwGNIafo4RbuwOkJYAIUxp6tjRzPthBQBvjpQEQOQvaS56ffPmTWrWrEnJkiWxtLQkLi4OlUrFwIED2bFjR6qucerUKTw8PIzfW1pa0rRpU06fPp3s8X///TedOnVizpw5NGnShL59+3L06FHMzZPPYbRaLRqNJsl/OY2xJUDGBAhhMh+Pfs+YAHw86l2GT/zYxBEJkb3SnATExcVhbW0NQIkSJfDx8cFgMBASEpLqJ/OQkBBcXlhWt1ChQgQHByd7fGBgIL/99htarZZvv/2WBg0a0K5dO06cOJHs8XPmzMHW1tb4X06cuSArCAphWh+NeodvVvwEwJQxo5i3/EcTRyRE9stQlZoKFSpQrFgxmjVrxq1bt5g/f36qzrO2tiY6OjrJtujoaBwcHJI93srKigoVKjBv3jwgofvgyJEjrF27NtmVC6dPn86UKVOM32s0mhyXCCROEXSU7gAhst2kd9/mu5U/AzBt7Gi+WvqDiSMSwjTSnAS0a9eOdu3aGb/fu3cvGzZsoFSpUvTt2zdV16hSpQrXrl1Lsu369esMGzYs2ePd3d3x9/dPss3FxYXIyMhkj7ewsMDCImc/YYfFSEuAEKZw/M/NrN+0GYDp48by5eKlJo5ICNPJ8ELYhQsXZvz48bRq1YpNmzal6pzBgwezbds2YyLw119/4e3tTb9+/YCExKJ27drG4998801OnDjB5cuXAfD392fXrl0yRVAIkSbn/vcnpzavZ5RHY76a+pEkACLfS1NLgKIoHD58mEePHtGxY0ecnJwIDQ3l66+/ZtmyZXTv3p2BA18/t7ZPnz6cOnWKunXrUqxYMUJCQli1ahVubm4AhIaGcunSJePxbdu2ZebMmTRr1ozixYsTGBjIW2+9xciRI9P4dnMOY3eATBEUIlv8OPcLIi+eQqVS0ffDydRs28nUIQlhcmlKAoYPH84vv/yCWq2mVKlS/PzzzwwePBgXFxdWr15N//79U32t+fPnM23aNB4+fEjZsmWT1Afo1KkTFy9eTHL81KlTef/99wkICKB06dLY2dmlJfQcx9gdYCctAUJktQ+GDWHpL7/Sxr0Ci5Yuo0abDqYOSYgcIdXdAQ8ePGDLli1cuHCBmJgY6tWrR5cuXZg8eTKXL19mwIABqFRpWwjH2dkZd3f3lwoEOTk5JekOSGRvb0+VKlVyfQIQq9Wj0eoxV6uwszQzdThC5Gljh7zB0l9+RQW07NpdEgAhnpPqloA7d+5Qp04d6tSpAyS0Cvj4+PDRRx9lWXB5VeIKgk62lmlOnIQQqTfmzUEs37AJFfD5RxP5bP4CU4ckRI6S6iQgJiYGW1tb4/e2trYvLScsUkcKBQmR9UYNHsiPv21GBXwx5SOmf526KcxC5CdpGhNw69Ytxo8fD8C9e/eSfA9QuXJlRo8enZnx5UkyPVCIrPXRu28ZE4Avp33MJ1/NM3VIQuRIqU4CXFxcqFq1Kjdu3DBue/H75Bb/ES8LM7YEyKBAITLbid83YvvQn6IO9oz/4H2mfjnX1CEJkWOlOgmoXbs2e/bsycpY8o3EGgGygqAQmev4ll85tXUTjra27P1rK7VkEKAQr5ShssEifWR6oBCZy2AwMLJ/X+5cuUS3WlXp/P5E3Jt7mDosIXK8DFcMFGknhYKEyDwGg4G3+/Vhzda/OHzLF7fOPSUBECKVpCXABMKkZLAQmcJgMDDCsze/bNuBWqXim1kz6D0891YSFSK7SRJgAjJFUIiMMxgMDOvdk193/I1apWLB7FmM/3SGqcMSIldJdxIQHR2Nl5cXBoOBxo0bo9frc/zKfTlFmLFYkHxeQqSHwWBgaK/ubPjfLtQqFQu//Jxxn3xm6rCEyHXSNSZgz549uLm50aFDB2bMmEF0dDQ1atTg8ePHmR1fnpQ4JkC6A4RIO0VR2Lt6BfsOH0GtUrFozheSAAiRTmlOAmJiYhg6dCirVq3izz//BKBAgQIMHTqUhQsXZnqAeZFxiqC0BAiRJoqicGTDGq7t38WYNk358btv+GDadFOHJUSuleYk4OrVq5QtW5bu3bsn2V69enW8vLwyLbC8SlEUwmVgoBBpZjAYmDtxHGd3bEVtZsbgT2YxcvwkU4clRK6W5jEBjo6OhIaGvrTd39+fUqVKZUpQeVlMvJ54vQErczXWFrKCoBCpYTAYGNi1E7/v2Ucb9wr8uG49FRo0NnVYQuR6aW4JqFChAlZWVsyePZvo6GgArly5wrfffounp2emB5jXJA4KlFYAIVLHYDDQv3MHft+zDzO1iv5vvysJgBCZJM1JgFqt5rfffuP333+nV69eHDhwgDp16vDuu+/Svn37rIgxTwmNlumBQqSWwWCgb8d2bP3nX8zUKn749hvemyTLlwuRWdI1RbBGjRr8999/3L59m+DgYNzd3SlYsGBmx5Ynhcv0QCFSxWAw4NmhLdv+PYSZWs2KBd8wcvxEU4clRJ6S5iQgNDQUMzMzHBwcqFKlSlbElKeFyvRAIV5LURTe7dfHmAD8uGgBb38w3tRhCZHnpLk74OLFixQrVowhQ4Zw8OBBFEXJirjyLJkeKMSrKQYD/65eTildNKUKOrJy8UJJAITIImlOAtq0acP+/fuxtramV69elC9fntmzZ+Pv758V8eU54caSwdISIMSL9Dod+1f/wKV9u3AqYM/hffsY8f44U4clRJ6VroqBTZs2ZeXKlTx8+JDZs2dz+PBh3NzcGDdO/rG+jrElQFYQFCIJvU5HjzYefPHNd5iZW9Dro89wq9vA1GEJkadlaAEhGxsbunTpQlhYGH5+fly/fj2z4sqzZAVBIV6m1+no3roVu4+dwMJMzcxvvqVsrbqmDkuIPC9dLQF6vZ5du3bRr18/ihcvzpo1a5gwYQKbN2/O7PjynMR1AxxlTIAQQEIC0LVVC2MCsOaHZbTt2cfUYQmRL6S5JcDLy4uOHTui0+l44403OHv2LDVr1syK2PIkKRYkxDN6nY4uLZvxz8kzWJiZsXb5Mga/856pwxIi30hzEmBlZcXSpUvp3r27LB2cDs+mCMpnJ/I3rTaeri1bsO9UQgKw7sflDHr7HVOHJUS+kuYkoFKlSlSqVCkrYskXEscESHeAyM8MBj3bF3+L1/XrWJqZsX7lj/Qf8bapwxIi30lVEnD//n0uXbpE586duX//Prt27Ur2uJIlS9K5c+dMDTAvMRgU45gAJxvpDhD5k8GgZ+8Pi7h79gRj27egWq+B9H5jiKnDEiJfSlUSEBgYyO7du+ncuTOBgYGsWrUq2eMaNmwoScArRMbpMChgZ2mGpXm6xmQKkatptfFMGzmcYppwLK1tGDB1FqWqVjd1WELkW6lKAho0aECDBg2MX586dSpLg8qrwo3VAqUVQOQ/2vh4OjRtzKHzF2lbrRJrf99KKXdJAIQwpTQ/jp49ezbZokApbRfPhMbICoIif4qPi6N900YcOn8RK3Nzxkz7VBIAIXKANA8MDA8P59q1ay9tDw4OxtvbO1OCyqtkeqDIj+Lj4mjfpBFHLl7CytycTb+spdegN0wdlhCCNCQBDx48YO/evVy7do0HDx6wdu1a4z69Xs/WrVspX758VsSYZ4RJS4DIZ+Lj4mjXpCFHL17GysKczb+so+fAwaYOSwjxVKqTgHv37rF06VIiIyN58OABS5cuNe6zsLCgcuXKTJs2LUuCzCtCoyUJEPmHXqfjja6dOHrxMtYW5mz5dT3d+w80dVhCiOekOglo2LAh586d48qVK2zfvp3p06dnZVy5jipxRWVVysdId4DIL/Q6HbuWfEN1O3MuFHFhybJldO3b39RhCSFekOYxAdWrV6d6dRnQ8yK1zgCAYm6W4jHGQkGygqDIw+JiY9mzdAHeZ09SqGBBjh47SomKVUwdlhAiGVIsKJOodXoAFIuUP9IwY8lgaQkQeZMmJpo2jRpiFhvNgGYN6Tv9C4pVkAqjQuRUUiwok5hpn7YEvCIJCDXWCZCWAJH3aGKiad2wAaevXsfG0oKFq8dIAiBEDifFgjKJsTvgVS0BGikWJPImTUw0Hg3rc+bqDWwtLflry2YatPQwdVhCiNfIlNq1ISEh6HS6zLhUrmWWqjEBsoKgyHuio6JoWb+eMQHY9vsWOvTsZeqwhBCpkK4k4KeffmLTpk3Gr11cXChTpgx+fn6ZGVuuYpaKMQHPpghKS4DIG6KjovBoUI9z129ia2XJ9q2/075HT1OHJYRIpTQnAeHh4cyaNYs+ffoAsHjxYry8vOjUqRPLli3L9ABzC/VrxgToDQoRsQmtJTI7QOQFuvh4/pj/BYEPHmBrZcn//txKu249TB2WECIN0pwEXLt2DTc3NywtLfHz86Ny5cpUq1aNPn36cPPmzayIMVcwdgdYJN8dEP50PICDtTlm6lcUExAiF9DFx7NjwRyCbl7lw27t2bltG226dDN1WEKINEpznYDChQtz69YtoqKi+PPPP2nWrBmQUFGwVKlSmR5gbvG6gYHG8QB20hUgcreoiAgmDhlEJSuwdXCk32dzKOxaztRhCSHSIc1JQIUKFWjQoAHFihXD1taWS5cuoSgKv/zyS5JSwvlN4pgAQwpJgHF6oHQFiFwsMiKcFvXrcum2Lx1rVWX9jp0ULlPW1GEJIdIpzUkAwPbt2/Hy8sLV1ZWCBQsSGRnJvHnzqFevXmbHl2s8GxOQUneADAoUuVtEeBgt69fjkrcv9tZWTP/mO0kAhMjl0jU7wNzcnFq1ahEaGsr58+dRFIWWLVtmdmy5itlrugNCoxPXDZCWAJH7hIeG0LxeXS55+1LAxprd//sfLdp3NHVYQogMSlcScPbsWdzd3alQoQL169enUKFCTJ06NbNjy1VeNyYgNEZaAkTuFB4aQov69fDyuYODjTV7/v4fzdu1N3VYQohMkOYkID4+nr59+9KlSxcCAwOJjo5m165dbNiwgfXr12dFjDmeotejNigYAMyS/0jDNVIyWOQ+2rhYBnbugJevX0ICsHMnTdu0M3VYQohMkuYk4MaNG9jY2LBw4UJKlCiBra0t7du3Z/bs2ezZsycrYszxlPiEp3ydOaBKfvqfsSVABgaKXEIbG8tf82bTpFhBqpUuwd7du2jSuo2pwxJCZKI0DwxUFAW1+uXcwdzcHEVRMiWo3CYxCdCmXDHYuIywTBEUuUFEaCg7F33N/RtXKVKsGEdO/UzBEvl3CrAQeVWaWwLc3d0JDw/ns88+IyIiAoPBwJkzZ5g1axbt2uXPZsK0JAFSLVDkdCFBT2hcpxbfb9iEnZMz/WfMlQRAiDwqzUmApaUlmzZtYv369Tg6OmJlZUWTJk3o3LkzI0aMyIoYc7wk3QEpCNMkLh4kLQEi5woJekKzenW5fvce1x48wWPMRAqWKGnqsIQQWSRddQJatGjB7du3uXHjBuHh4VSqVIkiRYpkdmy5hiEVLQHPpghKEiBypuDHj2hWvx437wXiZGfL/n3/UKVWHVOHJYTIQmlqCbh//z5vv/02derUoVOnTnh5edG8efN8nQAAKPEJN/hXdwckJAqOMjtA5EDBjx/R9GkC4Pw0AajXpJmpwxJCZLFUtwQYDAa6d+9OfHw83bp1IyAggKFDh1KoUCE6dszfRUNe1x0QrzMQHa/HTK3CwTpdjS9CZJknDx/SvEE9bgXcx9nejn/37aNO4yamDksIkQ1SfUfy9vYmICAAX19f7OzsAKhYsSJbtmyRJED76u6AxPEAjjYWqFKYQiiEKcTFRPPXN7OJjoqioL0d//77L7UbNjJ1WEKIbJLq7oDAwECqVatmTAAAGjduTEBAQJYElps8mx2Q/A0+PEYKBYmcJy4mmq1zZhAZ4M9kz24cPHhQEgAh8plUtwTo9Qmr5MXGxibZptVqjdvMzMywsMh/Nzpjd4AZJPfuZQVBkdM8DAxgwpBB1HVxwKlIUfrPmItjkaKmDksIkc3S1EF98OBBbGxsXtqeuK1t27bs378/cyLLRZ5PApKTOChQZgaInOBhYADNGtTH98EjIuvXZuOyNTgUzt+De4XIr1KdBNSrV4+DBw++8hhnZ+cMB5QbGbsDUvg0w4zdAZIECNN6cO8ezRo24M7DR7g42DPvp9WSAAiRj6U6CXB2dsbDwyMLQ8m9DK9rCdAkriAo3QHCdBISgPrcefiYwg4FOHj4ENVq1zV1WEIIE0rXUsIiqdeVDU4cE+AsSYAwkUD/uzRtUC8hAXB04NCRI5IACCHSVzFQJPW6YkHPCgVJd4DIfpqoSAZ164LfoycUdnTgyNGjVKlR09RhCSFyAEkCMoGiTUgCUioWFCYtAcJENJER/P7lp7QvXwpNrIb1W7dJAiCEMJIkIBO8vjvg6ZgAG2kJENnnwT1/9i6ZT7C/HyVdXTmwegMFCrqYOiwhRA6SrjEBOp2O77//nl69ejFu3DjCw8NZvHhxZseWa7wuCQiTYkEim/nf8aVRvbqs2LoDx2Il6D9jriQAQoiXpKsl4M0338TX15c6depw7do1HB0dOXToEBUqVKBr166ZHWOO92ztgOQrBhq7A+ykJUBkvbu+PjRv1IiAoGC0Oj0dPpyCfcFCpg5LCJEDpbklwN/fn0OHDnHo0CH69etn3N6zZ082bNiQqcHlFq9tCUicIigVA0UW8/O+TfNGDQkICqa4sxPHT56kjFt5U4clhMih0twScP/+fdzc3LC1tU2y3cHBgfinN8P85lULCMVq9cRqDViaqbG1fMVaw0JkkJ/3bZo3bkRgcCjFCzpz7MQJ3CpXMXVYQogcLM0tAZUqVeLatWs8evQoyfY9e/ZQs2baRh1HRERw+/Zt4uLiUn2ORqPhypUrBAcHp+m1spLhFUsJhxqnB8oKgiLr+N66SbOnCUCJQs4cO3lSEgAhxGulOQkoWLAg77zzDo0bN2b16tXcu3ePN998kx07dvD++++n+jrTp0+nSJEieHh4ULRoUTZu3Jiq895//31q1KjBmjVr0hp6lnlVd4BMDxRZLToslN2L5qE2GChZyJljJ0/hVqmyqcMSQuQC6ZodMH/+fGbNmoVGo6FIkSI4Ojpy8eJFChYsmKrzt23bxpIlSzh79iyBgYGsXLmSESNG4Ofn98rzdu3axc2bN3F3d09P2FnmVcWCZHqgyEpRoSFs+XwascGPmTbYk2OnTlOuYiVThyWEyCXSlQSoVCqGDRvGtm3bOHr0KMuWLaN48eKpPn/Dhg307t2bGjVqANCvXz/c3NzYsmVLiueEhITw/vvvs3r1atTqnFXtWHlFd0C4TA8UWcT7+jUGd25PUOA9XMqUZfhX31G2QkVThyWEyEXSPDDQ29ubVatWJbuvYsWKvP3226+9xvXr13nzzTeTbKtatSrXr19P8ZyxY8fy3nvvUbny65s5tVotOp3O+L1Go3ntORnxqu6AZ+sGSEuAyDy3r16hRfPmPAoLx8K8IetXbcTWwdHUYQkhcpk0JwFxcXEEBAQYv1cUhevXr3Pr1i0+/fTTVF0jNjYWOzu7JNvs7OyIjY1N9vg//vgDb29vfv3111Rdf86cOXz++eepOjYzvHJMgKwgKDLZzStetGrRnEdhEZQp7MLC9b9JAiCESJc0JwHVqlVL9mY8fPhwXF1dU3WNggULEhQUlGRbcHAwZcuWfenYmJgYxowZw+LFi40tBXFxcTx8+BAfHx/Kl395DvT06dOZMmWK8XuNRkOhQllXLOVVxYKeVQuUlgCRcUkSgCIuHD99llLJ/LsRQojUyLS1A3r37s3mzZsZNGjQa49t1KgRhw8fNn4fHx/PyZMnGTBgAADh4eHcu3eP6tWrGwcfzpkzx3j8vXv32LBhA35+fvzxxx8vXd/CwgILi+x78n5ld0C0tASIzHH98iU8WrbkcXgErkUKc+zMGUq5ljV1WEKIXCzTkoCLFy+mesDeBx98QJ06dZgxYwYdOnRg2bJlODk50bdvXwB2797NoEGDUBSFQoUKceXKlSTnV69eneHDhzN58uTMCj9DXlUsKEwjUwRFxkUEPWFY3z48Do+gbNHCHDtzlpJlUtfyJoQQKUlzEnDx4kU++uijJNuCg4O5fv06e/fuTdU1KlWqxL///svcuXPZuXMn1atX59ChQ8YqhE5OTlSrVi3F8ytWrIiLS85ZDMVYLCjZOgFPiwXJFEGRThFBj9ky+xN6VKuAna0NG3bsokSZMqYOSwiRB6Q5CShcuDC9evVKss3JyYmWLVtSJg1/mBo3bsz27duT3depUyc6deqU4rl//fVXql8nOxjrBCTzaT5bPEhaAkTaeV+9woEfFhAZ9ISyVaqy+9etWNvZmzosIUQeka61A65cucKKFSuyIp5cSaYIiqxw+dxZ2rRpQ7lCTowb0Ie+n34hCYAQIlOlq2zwuXPnsiKWXEtJoTtAURTCNYndAdISIFIvMQEIjowiNF5Hl4mfSAIghMh0aU4C3NzcsLGx4e+//86KeHIVnzAffMJ8MMQnLID0YndAdLwerV7BxsIMawtZQVCkzqWzZ2jTpjXBkVFUKFmcE+fOU6hIUVOHJYTIg1LdHRAYGMjFixdxcXHh7t27dO/enRIlSuDs7Gw8pkmTJqxcuTJLAs0pYvXPVjzstb0XKAqb4vSoSegOUPGsVoBMDxRp9d+Z07Rt25aQqGgqlirO8bMXKFysmKnDEkLkUalOAq5fv86iRYvYsGED8+fPT/aYYvngj1VYfITxazdHN9R6BTW30auhTrF6NCreyLg/XCOFgkTqXT57hjZt2xIaFU2lUiU4dva8JABCiCyV5oGBRYsWZeDAgVkRS66zvdd2DNHR3JxeH3NLa9Z1Xpdk/7MVBKUlQLxa6MP7HFixiAKWFhQuXVK6AIQQ2SJNYwIMBgOxsbEp/qfVarMqzhwrsUaAyvLlp32ZHihSI/RBIFs+n4Y+KoKZI4dKAiCEyDZpSgIOHjyIjY1Niv917tw5q+LMsRJrBCSfBCSOCZDuAJG8c8eP0r9TeyKCgihZpRpvfv61JABCiGyTpu6AunXrsmDBghT3Pz9IML9ILBmssnz5ad+4eJB0B4hknDl6hA6dOhEeo6FIkaJ8OG0WltY2pg5LCJGPpCkJcHZ2xsPDI4tCyZ0SawSoLV5+2pdCQSIlp48cpmPnzoTHaKhatgzLNv0hCYAQItuluU6ASEp55ZiAp4WCZIqgeM6pwweNCUC1cq4cP3cBpyxc6loIIVKS6paAkiVL0qVLl6yMJVdStK8YE6CRlgCR1MmDB+jUtSsRmliql3Pl2PkLODoXNHVYQoh8KtVJgLu7O+7u7lkZS670qpYA4xRBaQkQQHCAP6PfGk6EJpYabmU5eu68JABCCJNKc50AkdSrkoBw45gASQLyu6B7d/n9i+n0q+NOmZIlWf/3Lhyd8t9AWiFEziJjAjIodS0B0h2Qn104fozfZk0lJjyMyvUasvXAIUkAhBA5giQBGZRSsSCDQTGWDZYVBPOvw//soVW7dqzdf4TSNWrT86NPsbC0MnVYQggBSBKQYc9aApLe6CNjdRgUKGBljoWZfMz50aE9u+jWsxdRsXEYrG3o+uEUSQCEEDmK3J0yKLFioPqFloBQmR6Yrx3ctZNuvXoTFRtH3coVOXr2AnYFCpg6LCGESEIGBmZQSmMCZHpg/nVg19/06ONJdFw8dStX5Mi5C9jZ25s6LCGEeIm0BGSQMQmwSL4lQKYH5i+H9+6m+9MEoF4VSQCEEDmbtARkUEotAYnTA2VmQP7xyNebc7+uooi9HS5u5Th05pwkAEKIHE1aAjLo2QJCybcESI2A/OGRrze/fzkd4uP46sOxHD4rCYAQIueTJCCDUhwTICsI5ht7t/+FZ5eOxERGUqFBY/pPm4GtnSQAQoicT5KADEo5CZBCQfnBnj//oHf/ARy/6cttrZpu46diZi6JnxAid5AkIIMMKdQJCDWOCZAbQl61a+vv9Bk4GE28liY1q/H9b79jZi7DbIQQuYckARkkUwTzp51/bMFz0GA0Wi1Na1bnwKmzWNvYmDosIYRIE0kCMiilYkFhUiwoz/rflk30HfwGsVodzWrV4N9TZyQBEELkStJ2mUGvGxgoLQF5S+DN60ybNIFYrY7mtWvy76kzWFpJKWAhRO4kSUAGpZQEyBTBvCfwxjW2zp3J4Aa1uF6zJqv/3CEJgBAiV5MkIIOSSwJ0egORsTpUKihgLUlAXvDv/7ZzecsvGOLjqOXRlqljJ6I2MzN1WEIIkSEyJiCDkisb/PwSwmZqlUniEpln6/p1dOnjyS+HT1KpSQs6SwIghMgjJAnIoORaAkKlUFCe8ccvaxn81tvE6/QULFaCTmMnSAIghMgzJAnIIIP25ToB4RopFJQX/L5uDW+8NZJ4nZ62Deux5/gJLCzkZyqEyDskCcigxCmCz3cHhEZLoaDcbsua1bz59jvE6/W0a9SA3cckARBC5D2SBGSQkkzFQCkUlLvt+O1X3nznPeL1eto3aciuo8ckARBC5EkyOyCDEpOA54sFPVs3QFoCcpu7Xv9xY8fvlC7oSIUKFdh15LiUAhZC5FnSEpBByQ0MfLaCoDw95iZ3L//HtnmzMTPoWfjZNEkAhBB5niQBGZT87ICnhYLspCUgt/j1pxX069WDuLhYarTtSPexEyUBEELkeZIEZNCrWgIcZYpgrvDL8mW8NWYsZ+/c46GNM+1HjkWlln8aQoi8T/7SZVCySYAmsWSwdAfkdOt+WMrb749DqzfQrVVzvvl5nSQAQoh8Q/7aZYCiKCjap1MELWWKYG6zdukSRn7wITqDge4eLdh+4LB0AQgh8hVJAjLiaQKAuXmSp8dwmSKY463+fhEjP5yAzmCgZ5tWbPv3EGppARBC5DPyVy8DDPEvtwLAs4GBjtISkCP5nD/Dt3PnojcY6NXWgz/3HZAEQAiRL0nbZwYoT0sGqy2e3ezjdHpi4vWYq1UUsJKPN6fxOX+aHQvmMqRJHR7aOrNo3QZJAIQQ+ZbcpTIguUGB4THPxgOoVLKCYE7y+5rV3N27HZVioEn3PngMe0d+RkKIfE0egTLgVSsIyvTAnGXFt/MZNPIdfjl+ltqdukkCIIQQSBKQIcnXCJDpgTnND998zdiPp6I3KLjXqEnrYe9KAiCEEEh3QIa8qiVApgfmDMvmzWXctOkYFIUBXTqy8X+7ZAyAyDBFUYh/+u9fCFOztLRM94ONJAEZkOyYAE3i4kHSEmBq38+dw/jpn2FQFAZ168yv2/+WBEBkmFar5c6dO+j1elOHIgQAZmZmlCtXDguLtD98ShKQAa9sCZAxASa1bukSYwIwuHsX1m/7nyQAIsMUReHBgweYmZlRunRp+Z0SJmcwGLh//z4PHjygdOnSaW4RkCQgAwzGJODZDT9x3QBnO2kJMJWbJ4/y4PAeyhUuSJPGTVj313b5Yy0yhV6vJzo6mlKlSmFjY2PqcIQAoEiRIgQEBGAwGDAzM0vTuZIEZMCrBgbKmADTuH78MLuXLsBCrWblgm9oNXiYJAAi0yR2AaSn2VWIrJL4+6jT6dKcBMhfxwxQnlYMVCezjLCTjbQEZLcFn89g0KBB6HQ6GnsOxOON4ZIAiCwhs0tETpKR30dpCcgAY0uAxcvLCDtLS0C2+mbmp0z5Yg6KAn36eNK03xvyh1rkeV5eXly9ehWVSoW9vT1VqlShfPnyGbrmxYsXsbOzo1KlShk6Ji1Onz7NnTt3kt3XtWtXChQokCmvI14mSUAGJN8d8LRYkCQB2Wb+Z9OZOucrFAXe7teHGYu+N3VIQmSLzZs3s2XLFurWrcuTJ0+4fPkylStXZuPGjZQpUyZd1/Ty8qJw4cKvvMGn5pi0uHDhAocPHwZg//79uLu7U7JkSQBatmwpSUAWkiQgAxLXDkiSBGikWFB2mvfpNKZ99TWKAiP792Xl5t9NHZIQ2apDhw4sXboUgJiYGIYNG8agQYM4fvw4AJGRkZw5cwaAJk2aYGtrazw3MjKSc+fO8fjxY1xcXGjbti01atTAzs7OeMypU6fw9/fHYDDg4eFBsWLFXjomKCiIs2fPUqhQIRo0aGBshTt37hzOzs4oisLNmzdp3LgxhQoVeuk9jB49mtGjRwPQuHFjJk2aRK9evYiMjOTo0aMUKFCA06dPU7FiRVxdXdFqtZw9e5awsDAaNmyIi4uL8Vqv2ideJklABrzYEqAoihQLykZffTKFT+fORwHeGdCXnzZJAiDyN1tbWxYuXEjp0qUJDAzk8ePHDBw4kIoVKxIbG4uvry/Hjh2jRIkSnD9/nu7du1OhQgVKlChBlSpVaNu2LevWraNs2bJUqlSJYcOGceHCBdzd3VGr1bi7u1OsWLEkxxw5coQ+ffpQv3597ty5Q7ly5di1K6Eo16pVq7hy5QparZYCBQpw5coVrl69mmwikJzAwEDeeustypYtS4kSJZg4cSL29vZ06tQJBwcH7OzsGDlyJJs3b6ZFixYEBwenuE8kT5KADHgxCYjVGojXGbA0V2NjkbYRmiJtrhzaz7o1a1CA9wYNYMXGTaYOSeRTZafuzLJr+33dNc3nlCpVCltbWwIDA5kyZQoDBgygatWqAGzcuJGVK1cyc+ZMJkyYwPTp0xk7dmyK19q7dy+nT5/G1dU1xWM+/vhjvvvuO4YOHUp8fDwNGzbkr7/+wtPTE4CyZcvy66+/AtCvXz92797Nm2++mer3ExISwqlTpyhbtiwAH330EZUrV6Zbt24AuLm5MX/+fFq0aMHXX3+d4j6RPEkCMuDFOgGhxnUDZAXBrHTl4D72/riE4U3roS1TkRkLl5g6JCFyjKioKGJiYnBwcODixYs4OTlx69YtIKGloFSpUgBcunSJTZtenTxPmzaNunXrUrFiRXr27MmECROwtrZOcsy1a9do3749kFC+tk2bNly7ds2YBDRt2tR4rKurK8HBwWl6P2XKlDEmAJAwKFGr1bJt2zbjtpo1a752n0ieJAEZ8GJLgEwPzHo/fTOX0DPHMFeraDPkLRr16mfqkEQ+l56n9ay0evVqSpQoQaVKlShWrBijR4+mQ4cOLx1XvHhxbt26RYkSJVK81ocffsjo0aM5c+YMs2fPRq/X8+mnnyY5pkiRIty5c4fixYsD4OvrS+fOnY37X5ymqyhKmt6PuXnS21SxYsWoWrUqn3zyyUvHvmqfSJ4kARnwYp2AcBkPkKVmTZ7A7AWLqFayKCu/XyIJgBDA7du32bRpE0FBQZw8eZLt27fz22+/oVarmTp1KsOHD2f8+PHG2QJNmjTB1dWVyZMnM3ToUCZMmEDx4sUpXLgwbdu2TXLtxJYCjUZDeHg4zs7OL73+6NGjGTFiBBMnTuTWrVucOHGC1atXZ9n7nTBhAu3atSMyMpIaNWqgVqupXLkyderUeeU+kTxJAjLgWZ2AxO4ASQKyysyJHzL7abN/mw4dady7v4kjEsL0atasibe3N9u3b8fOzo46deowd+5c4w1/+PDhuLm58ddff3Hx4kUURaFUqVK4uroycuRIypQpw44dOzh9+rRxYGDdunUpXLgwgLFZ3d7enlGjRjF8+HCAJMdMmjSJ0qVL8++//1KwYEHOnTtnHPjXoEGDJOMJ6tSpQ8GCBV/5ntq3b2/ssnBwcKBr16QtLfXq1eP06dOsXbuWnTt3otfr6dKlC3Xq1HnlPpE8lZLWtplcSKPRYGtrS0xMTIbrfT98+B9bZ4UA0KfIYcK2bKHYzBk4DxrEhtN3mf7XFQY2KM3XntIPlVk+Gz+OLxcnzP3/cMQwFv281rQBiXwrLi4OX19f3NzcsLKyMnU4QgDJ/16m9r4nNVUz4MUxAVIoKPN9+uH7xgRgwsgRkgAIIUQmkiQgA15OAqRQUGZaNGsGc5YsA2DiO2/x3cqfTRyREELkLTImIAOMFQMtEmcHyLoBmeXi3r+J9TpLpaIudO/Zi29/XGnqkIQQIs+RJCADDCm0BDjKFMEMObdzO4d/WYmFuRkbVq+kftdepg5JCCHyJOkOyICUxgRIS0D6fTTqHYa/+x46vYE2b42SBEAIIbKQtARkgKJNuOm/WDHQScYEpMukd9829vvb1mpInY7dTByREELkbdISkAEvFQvSSEtAej0/8G/6uLF88Nks0wYkhMhXfvjhB7RPH+zSsi+3M1kSEB0dzapVq5g5cyZ//PEHBoPhlcffuHGDxYsX880333Ds2LFsivLVnu8OUBRFpgim0/i3hrNw1RoAPhv/AV8uXmriiITIHU6ePMmhQ4eM38fExLB06VK8vb2THLd3715WrFiRZNs///zDxYsXsyPM11q+fDlxcXEp7v/555/Zvn17km1r167l8ePHr732jz/+SGxs7GuP+/jjj1OM4VX7cjuTJAGRkZE0bNiQ1atXExsby9SpU+nbt2+Kx3/22Wf079+fO3fucP/+fbp3784HH3yQjREn7/kkICpOh86gYGtphpW5rCCYWuOGD2XxmnVAQgIwWxYDEiLVdu7cyR9//AFAeHg4HTp0wMfHh/Llyyc5bv369YwZM4atW7cat23cuJHDhw9na7wpmTZtGhqNJsX9s2fPZuDAgfj7+xu3ffnll0m+T8m9e/fQ6/WZEmdeZJIxAUuXLiUuLo5Dhw5hZWXFBx98QLly5fj3339fql0N0KtXL2bPnm1cma9Tp0506tSJGTNmGEtXmsLzScCzQYEyHiC1Tv65mX/27gFg5sTxzFqw0MQRCZE7PXnyhI4dO9K9e3c+//zzZI/p27cvn376Kb169cLM7OUHFR8fH44dO4a1tTVdu3bF3t4eSGgxuHbtGpaWllStWhUPDw/jOdu2baNWrVp4e3sTGhpK//4J5bzPnDnD5cuXKVmyJB07djQuIhQYGMi+ffsICwvDxsaG9957j927dxMXF8fy5cuxsbHh3XffxdbW9qX4unXrxqxZs/j55+TrhaT0mqVLlza+X4PBwM6dO4mKiqJjx45s2rSJ9957z7g/OjqavXv3otVq6d27d5KqkJGRkezevRu1Wk337t2xtHz2t/7AgQP4+voaSxcn+uGHHxg4cCCHDh3CxcWFli1bJvsZmJJJWgL27dtHr169jB9wqVKlaNq0KXv37k32+Hr16iVZmjfxlzOl5Xq1Wi0ajSbJf1khuSTA0Ua6AlLj9LbfObF5PcOb12fZV19IAiBEOgUEBNCyZUuGDBmSYgIA0KZNG0qVKsW6dete2rdlyxZ69OjByZMn+eOPP6hduzYhIQnl0YOCgvDz8+Pq1atMmjSJCRMmGM9bsWIFvXr1YuXKlQQEBAAwefJkJkyYwIULF5g3bx7dunVDURQCAwOpW7cuBw8exM/Pj3v37gEJCYzBYMDf3x8/P78Un9onTZrEnj17uH79+kv7UnpNgOnTpxMVFQVA//79+fTTT9m/fz/dunVjwoQJSfr6+/fvz65du1i8eDH9+iVdoKxPnz7s3r2befPmJVnPYOjQobz//vucOnWK7t27M3/+fOO+jz/+mB49erBr1y6Cg4NT/AxMySQtAQEBAXTv3j3JtpIlSxIYGPjac3U6HTNmzKBPnz64uLgke8ycOXNe+Y8hszyfBIRGPK0WaCdJwOvMmTwBc78bmJub023MBKq3bm/qkIRIv1mOWXjt8Ncesn//fkqXLm18onzy5AkbNmwAoGzZsvTq1ct47Ndff02fPn144403klzjww8/ZMyYMRQoUACAu3fv8tdff/H2228zcOBAihQpwu3btylbtixz585l4cJnSbunpyczZswAElY0XL9+PdOmTQOgUqVKzJkzh1u3bhEQEECFChVYtmyZ8UEOEm6i48aNY+7cuTg5OaX4Pu3t7Zk6dSrTp0/nzz//NG5/1WtWrlzZeJyXlxenT5/G29sbKysrzp07R4MGDZK8xpw5c2jevDmxsbEUKlSI+Ph44xP/Z599RpcuXdDr9dSsWZPDhw9TuHBh9u7di6+vL3Z2dty6dYvatWszfvx443mzZ8+mTZs2APz777/JfgamZJIkQKVSvbSmtKIoL607/SKdTseQIUOIiIhI8kvwounTpzNlyhTj9xqNxriqVWZKkgTEJPxjdZJCQa80avBAfvxtM9VKFmXTr79S3aOdqUMSIlcbOnQoGo2Grl27snPnTuLj4/Hz8wN4aeGYevXq0ahRI5YtW2bcFhoaypMnTwgODiY4OBiApk2bUrZsWSChG+Hu3bs0aNAAa2trwsPD0ev1xib055u/b9y4gZWVlfH1Ad544w0sLS1p2bIllSpVomTJkri7uydZlTC1Ro0axaJFizhz5kyqXvN5vr6+1KpVy9gCXb9+fczNk94Ca9euDYC1tTUFChQgIiLC+LDZsGFDAMzMzKhfvz4+Pj6Eh4dTo0YN7OzsgIQEpECBAgQEBODm5vbS55MZn0FmM0kSULp06ZcGdNy7d48WLVqkeE5cXBwDBgzgyZMn7N+/H0fHlLNvCwsLLCyy/oncYKwTYGmcHijLCKfs3UH9Wbnpd1TA4KHDJAEQeUMqntazklqtZvXq1bz11lt069aNv//+m0WLFqV4/JdffkmrVq1o0qQJAE5OTtjZ2TFs2LCXltzVarXGPnQLCwtu3rzJ4sWLkzzEPT++wNXVlbi4OL788stkn3TXrFnDihUrOH36NP369aNJkyZUrlwZc3Pz184QA7C0tGTWrFnGp/7UvGaiEiVKcOvWLRRFQaVScevWLXQ63WtfM9GNGzdo3rw5ANeuXWPw4MG4uLgYr2Nubs7Dhw8JCwujePHixvOe/3wsLCxS/AxMxSRJQJcuXViyZAlz5szBzs4Ob29vTp06xdy5cwG4fPkyGzdu5OuvvwYSpr307t0bnU7HP//8Y8y6TC4xCbCwIDRakoBXGTmgL6u3bEWlgrnTpzHli69MHZIQeYZarebnn39mxIgRxkQgucF1kPC02qNHD37++WdatmyJSqVizpw5dOnShREjRlCkSBEgoX+8RIkSVKhQgbfeeotKlSqxffv2ZAcVJqpZsyZt2rShefPm9O3b13hTHj9+PDdu3GDPnoSBwPfv30elUhmfsitWrMi0adNwd3dPcWBgojfffJNvv/2Wu3fvvvY1n1e/fn1cXFzo3r07jRs35p9//sHc3Py1LdCJJk+eTJcuXbh06RLx8fG0a9cOMzMzKlWqRMeOHWndujVbtmxh9OjRKS7d+6rPwFRMMjBw9OjRFC5cmCZNmjBq1ChatWrFm2++aWwJuHbtGvPmzTMeP3z4cA4dOkSNGjX44osvmDp1KlOnTjX+EpiSysIClUpFmEZWEEzJ2/36GBOAr6d/IgmAEJmkadOmtG7dGkhIBNasWUP79u1fGmTdqVMnatasafx+5syZjB071vjk//7777Nt2zZsbW3x8/PDz8/POC/+n3/+oXLlylhZWbFp0yYmTJhgvHH27t3b2G2QaOPGjXz++efExsYarwUJI+/9/Py4e/cuBQsW5MSJE8Zu2g0bNlCyZEnu3r2b7MDAt956y3izVKvVLF++nLFjx1K0aNFXviYkdCHY2NigUqnYt28fHTp0QK1WM3fuXJydnY3dBmPHjk3ShfDuu+8ab+Zjx45l48aN2NjY0LRpUw4fPmxMhnbt2sXgwYOJjY1l5syZSVphXrzmqz4DU1EpL3bOZ5P4+Hi2b99OQEAA1atXp337Z4PDrl+/zvbt25k6dSqQMMc1uUGDb775JqVKlXrta2k0GmxtbYmJiUkxQ0uthw//Y+ushFGzbQ6NRW1nR+Xz55iw+T/+uhjIt/1q0bfe62PKDxRFYda4scxeuhyVCuZ9Np2PPv/S1GEJkW5xcXH4+vri5uaWZPqYyB0SpztGRUWxadMmevTowVdf5f6HkuR+L1N73zPZ2gGWlpYvTcFI5O7ujru7u/H7IUOGZFdYafbiCoJOMkUQSEgATmz5FZsHd6haoihvvfMuk2bNNnVYQoh8LHG6o52dHV988QW9e/c2dUgmJwsIZVBiEhCaWCxIpghiMBg4snEd5/+3FUsLC/7Yshn3Zq1MHZYQIp8bPHgwgwcPNnUYOYosIJRBL7YEOObzKYIGg4FhvXsy+uMp6BWFruM+lgRACCFyKGkJyCBjEiArCGIwGBjaqzsb/rcLtUpFCY9OVG7S3NRhCSGESIEkARmksrREb1CMdQLya9lgg8HAmz278dvfu1GrVCya8wVvjjH9Ik9CCCFSJklABqksLYiM1aIoUMDaHHOz/NfDYjAYeKN7Vzbt2oNapWLx3C95f8onpg5LCCHEa+S/O1YmU1tYPhsUmA9rBBgMBgZ17WxMAJZ8/ZUkAELkMQsWLODJkyemDkNkAWkJyKCEFQSfTg/MZ+MBFEXhwNofuXzpImZqFUvnfc2oyR+bOiwh8o09e/awf/9+VCoVhQoVomPHji+V/s0Ma9asoWvXrlm6dPvy5cvx8fExfu/i4mKsFZOVjh8/zpMnT5IstJSfSEtABj2/jLBTPmoJUBSFQ+tWcmnvTka0aMTGn1ZIAiBENjt27BinTp2iaNGiPHz4kGbNmnHw4EFTh5Uumzdv5uHDhxQrVoxixYplacLxvIsXL7J///5sea2cSFoCMkhlYUFoPisUZDAYmDRiCEWjQ7C2sqLXxE8oX6+hqcMSIl+qXbs2kydPBkCv1/P7778bSwn/9NNP3Lp1CxsbG+rUqUPv3r1RqVRAQhN/z5492b9/P6GhoQwZMsRYgVWv17NhwwYCAgLo0aNHktdTFIWtW7dy5coVKlWqxIABA4wldBcsWECXLl04dOgQoaGhjBw5Ep1Ox7p167CxseG99957ZfW6bt26MXDgwCTbnjx5wm+//UZkZCQdO3akfv36QMKS9Fu3bsXDw4OdO3fSs2dPqlWrhr+/P3/++SeRkZF06NCBRo0aARAbG8tvv/3GrVu30Gq19OzZk6pVq7Jt2zaCg4OZPHkyjRo1SrGIXV4lLQEZ9HxLQH6YHmgwGPDs0JZFv2xk4+n/6DFpuiQAQuQQsbGxSZbHdXZ2plixYtjY2LBo0SI++eTZeJ01a9bQv39//P39uXXrFi1btjSuqjdy5EiWLl1KfHw8Y8aMSVK2/b333mPOnDkALFmyhP79+790zcDAQM6cOUOnTp0YMGAAcXFx7Nixg3ffffeV8W/YsIHJkyczefJk9uzZQ3BwMLVr1+bcuXNERUXRsWNHtm3bBsDDhw+ZPXs2kydPRqVSYWVlxaVLl+jUqROPHj1Cr9fzxhtvGJedHz9+PJs2bcLJyYlixYphZ2eHmZkZBQoUwM7OjmLFir1yddq8SloCMuj5MQGOebw7wGAw0LtdG3YcPIyZWs2HH0/BrW4DU4clhEnVWFcjy67tNczrtcccP36cyZMnExgYyMmTJ5MsHtS+fXs2bdrE3bt3cXd3Z9u2bcbVWgFmzJhh7AuvXLkyPj4+FCpUiD///JOAgAAKFCjAhx9+SMmSJQEIDg42thA4Ozvz0Ucf4erqyq1bt6hUqRKQsFRxz/+3d99xTV3//8BfBAIhhD0ieysqiJY4GApaLIha0TqodeACrW39/PyoHbaOqnW21qp1j6oVW63irrbWAVpFHAUtIooMFRREkBVIyPn9wYf7NcrSAAF5Px+PPB7JPeee+84p9b5z7rn3DBoEmUwGPT09/P333/Dy8sL9+/fRtWvt/14YGhqiTZs2AAB9fX389NNP6NmzJ3bs2AEA8PT0xLJly7iYGWM4dOgQN7oQFhYGFxcXyGQyyGQyuLu7Y+fOnRgyZAju3r2L0NBQjBs3TmmVwrfffhu3bt3iRlNaGxoJUJGGtnareFBQhVyO0LcDcOj0WWjxeNj8w/eY8PF/1B0WIa1e1a9YPT09GBsbc9fSKyoq0KNHD5w9exb6+vpo06YNnj59qrSvi4sL997Y2BiFhYV48OABbG1toa+vz223srICADx48ACWlpYwNjbmju3s7Iz79+9z7Tg7OwMA+Hw+hEIh99nIyAiFhYW1fpcBAwZwIwG+vr7cAnNVPD09lY7l6OiodHnh3r17sLKy4uYV+Pn5cZcXVqxYgUOHDkEsFiMwMBA3b96sT/e+8WgkQEUa2m/+LYIVcjlCA3vjyNnYygRg9SqM/fAjdYdFSLNQn1/rjen5OQEzZszAzJkzsWXLFmRmZqK4uBhRUVEAgIMHD9arPXt7e2RmZiI3NxdmZmbIyspCZmYmAMDOzg5ZWVnIysqCpaUl8vPzkZycDCcnp0b5bk5OTjh58iT3+eLFi1xSAYCb31ClXbt2MDIyqvZXvaenJ06cOIGysjIsX74cs2fPRnR0NPh8PncZpDWiJEBFGtr85y4HvHkjAUyhwPSxo3DkbCz4mjxsXv0DxkyZqu6wCCHVmDt3Ltq2bYurV6/Cw8MDfD4fgwYNgqmpKS5fvlyvNoyMjDBp0iR4e3vj7bffxtWrV2FgYMCVTZs2DT4+PggKCsKZM2cwYsQIODg4NMr3CQ8Px+rVq/HOO+/AxsYG0dHR2LdvX431586dC39/fyQkJMDNzQ08Hg+BgYEIDg7GsmXL8PjxY8hkMpw8eRLh4eEAgE6dOmHOnDkQCATw9fVtdRMDKQlQkfLEwDdrJIApFPhj0xpYlRWgs501pn8xG6Mjp6g7LELI//Tr1w9SqZT7rK+vj927dyM3Nxd8Ph+XLl1CdHQ0jI2NsWTJEvz2229c3RkzZsDS0pL7/Mknn8DW1hZA5dB537598eDBA8yePRunTp2ChYUFAOCbb75BSEgIEhMTMWzYMLz99ts1trlw4UJuuF5HRwfffPNNjd9lypQp6Ny5s9I2kUiEq1ev4siRIygqKsKXX37JjTrY2triP//5j1J9Z2dnJCUl4cSJE8jIyIBCoeAua5ibm4PH40EgEGDYsGHw86tc18Tb2xt79+7F9evXW+XEQA3GGFN3EI2ttLQUQqEQJSUltd6eUh/Z2dfx27w8AECfM1NhPu0ThD5rh/tPS3FmRgAczPQaImS1q5DLcWzdKtyOPQ0tbR2EzvoK9h6d1R0WIWpVVlaG1NRUODk5QUdHR93hEAKg+r/L+p73aCRARRra2ih4w0YCKuRyhPTyQ1rqXUzq44vBn86BnbunusMihBDSwCgJUJFCi4/CMjl4GpULCLV0crkMIT398MfFOPA1NdF+UBglAIQQ8oaiWwRVJEXlk7IMdfng8TTqqN28yWTlCOnpiz8uxkFbUxO7Nm1A/xFhde9ICCGkRWr5P13VrPR/eVRLvxQgk5Wjn58PTsVdqUwAtmzCsLHj1B0WIYSQRkRJgIqKFZVJQEu+PVAmK0ewrw/+unwF2lqa+HnLZgwdE67usAghhDQySgJUVMRa9kiAoqICR1d/i0cP7kNbSxO7t23Fe6PGqDssQgghTYCSABUVVVTOA2iJKwgqKipwfO13uHvpPMb39oXHsFHoO2iwusMihBDSRCgJUFFhVRLQwkYCysvKEDF0MNy0GfT0RRjy+dewbtde3WERQl7Br7/+yq2Sp6urCw8PD0REREAkEjXaMXfs2AFzc3P069ev0Y5Bmg4lASp6xiUBLWckoLysDIHe3RBzLQGe9tY4dvIPWLWlBICQliYhIQFPnz7FuHHjIJVKsW3bNly9ehW7du1qtGN6eHhAT+/NeCgaoSRAZQX/SwJaygqC0tJSBHp3x/l/EiHga2HBshWUABDSgrm6unIr5YlEIixfvpwr27dvH/bt2wctLS04OTnhww8/5JbqBYCtW7fi0qVL8PHxgVwuh5OTE3r37l1rWWJiIszNzdG2bVts2bIFVlZWSEpKQlJSEgYOHIh33323Xu2T5oGeE6CiAlnlU5dbwuUAaWkp3u7RjUsA9u3+GQOH03MACGnJTp48ibCwMISGhuLLL7/EkiVLuDI3NzeEhoaiX79+KCsrUxrCX7x4MVauXAkvLy/ExcVhxowZuHv3bp1lV69eRXJyMgDg8uXLiIyMRGlpKTw9PTFu3DgkJibW2QZpPmgkQEVPZS3jcoC0tBR9unfF34k3ocvnY1/Uzwh5r3WtlkVIY0hya7yRtPa3kuqs4+joiNDQUJSVlWHv3r2Iiorifm3b2dnhxIkTuHHjBkpLS5GSkoJHjx5BLBZj8+bN+O2337hFe6pO3gBqLXvR2LFjMXv2bADAzZs38ffff8PDw+OV2iDqQ0mAivIqlw1o1rcIVshlmDJ8CJcA7N+zG8FDhqo7LEJIA3j+csCwYcMgEokwf/58WFpaYuzYsdDU1ERISAiEQiH++usvFBYWQiwWIycnR2kJYEdHR+59bWUvsra25t7r6emhpKTkldsg6kNJgIryyhUAKh8b3BxVyGU4vHIp3HQ00M3FAfOXrUDw4PfUHRYhb4z6/FpvKtnZ2WCMQVu78kfJ5cuXcfbsWTg7OyM5ORlPnz7l6rq5ueH06dMYPHgwysrKcP78efTs2bPOsvpqiDZI46MkQEW5ZQB4gLFe8xsJKC0pxqGVS3A/4RpEBgY4dOIPiJ1c1B0WIaQBVc0JKC8vR2xsLD755BOYmpoCAAYNGoS+ffvC3d0dd+/ehYmJCbffggULEBYWhs2bNyMzMxNCoZBLHmorq6+GaIM0PkoCVFSo0ICWlgb0tDXVHYqS4qIiBHSTIP9JLj4MCsDQLxdSAkDIG2bEiBFwd3cHAOjo6GDRokVo3/7/5iisXbsWMTExKC8vR7du3RATE8PdHRAUFISEhAQkJibC3d0d7777LlxcXOosGzt2LHeL4MSJE2FsbMwdb9SoURAKhXW2QZoPSgJUJONpwVioDQ2N5rOCYHFREQK6eiH+1m0IdbQhGT2JEgBC3kAeHh7w8PCotc7zQ/AhISHc+xs3bmDhwoWQy+W4du0aOnbsCB8fnzrLunTpwrUhkUiUjlU1CbCuNkjzQUmAKjQ0UKHBa1bPCCguLIR/Vy9cSU6BUEcbh/f/hp59g9QdFiGkmTE3N0doaCg0NTXx6aefomvXrvUqa4j2SfNBSYAKGJ8PaGg0m9sDi549g383L1xNvgM9HW0c3r8fvUP6qzssQkgzJBaLubsKXqWsIdonzQclASpgWpUn/+bwoKCiZ8/Qs+tbuH77LkQCHRw+sB8BwSF170gIIaTVoiRABRWald2n7hUEZeVlOPL9ElQUF0Mk0MHRg9Ho9U6wWmMihBDS/NFjg1VQ8b+RAHXeHigrL8PB5Qvx4GYCJgUF4K+TJygBIIQQUi80EqACOa+y+9T1oKCCp3kIHzQQXc31YWxqhmFfLYKZrb1aYiGEENLyUBKgAtn/Lgeo45HBBU/z0FPihcTUNNx3tsfvZ87B1MauyeMghBDSctHlABWUa1Q+IKipbxEsyMuDn9dbSExNg4GuAD9s2koJACGt1OPHjzF37lz0798fw4YNw4oVK7jn9zelNWvWYN++fQ3a5pYtW9C7d29kZmZy27Zv3441a9aoLZ6AgAAEBASgX79+mDVrFnJychr0GE2NkgAVlFddDmjCJCD/yRP4enXBjXvpMNAV4MTxY/Du3afJjk8IaT7u3bsHT09PpKSkYPTo0Xj//ffx+PFjvPde068PEhwcjB49ejRom/fu3UN8fDzmzZvHbUtLS8OdO3fUFo+xsTHmzZuHadOm4fbt24iMjGzQYzQ1SgJUIP1f9zXV5YD8J0/g49UFN9MyYCjUxcnff0cP/95NcmxCSPPz5ZdfIjg4GLt370ZYWBiGDBmCZcuW4ddff+XqbN++HQEBAQgMDERERARSUlK4sq+//hpnz57lPi9fvhy///47gMon/oWHh6NPnz4ICAhAXFxcrdt///13XLx4sc5jfvfdd4iKisKsWbMwcOBAbN68udbvOGbMGBw7dgy3bt16qay241TFk52djcGDB4MxphTD8320a9cuDBs2DMOGDatz9MDa2hoBAQEIDg7G8OHD8fjx4zrjycnJeSmGFStWYP/+/XXGUFN/NxSaE6ACKSovBzTFw4LKpaWIHDYYSemZMBTq4sTx39G9V69GPy4hpPn666+/sH379pe26+vrc+8DAgLg4OCAiooKJCQkYODAgdwJ9d9//0WHDh24usnJyTA3NwdQuS7BhAkTMGbMGPB4PO65/zVtv3PnDuRyeZ3HvH37NlauXIl58+ahW7du+Pjjj+Hp6VnjEwX19fXx6aefYvbs2fjtt9+Uymo7TlU8bdq0QUZGBmJiYtCrVy/I5XKsWLECFy5cAFCZEFy4cAHh4eGQSqVYsGABRCIRgoOrv8vqwIEDuHHjBqRSKbKyspRO2DXFY25ujoKCApw8eRJBQUGQSqVYuXIlrl+/XmcMNfV3Q6EkQAWlqJoT0LgjAeXSUuxfPA9epiI86tgOy9dvQlc/WpKTkOZg7eS/Gq3tqetrv9RXXFwMIyMj7vPEiRO5ofKTJ09CW1sburq6OHr0KG7cuIHS0lJkZGQgKysLlpaWdR5fKpXCxMQEnp6eSuuj1LS9Sl3HjIyMxIQJEwAAZ86cwZUrV2p9rPCUKVOwatUqXL58+ZWOUyU8PBzbt29Hr169cPz4cbi5ucHBwQEAsG7dOpiammLFihVcn548ebLGJKBbt26YNm0aysrKEBUVhW+++QYHDhyoM56pU6diw4YNCAoKwr59+9C3b18u4aorhrr6WxWUBKigTEMTOlo8CPiNt4Lgk8ePcHjlYjxJvQNDcwscWL0Zxm2sGu14hJCWw9nZGYmJiejevTsAYPLkycjJyUFISAgUCgUAYPTo0XB2dsbHH38MoVCIsLAwFBcXA8BLJ5SKigru/aFDh7Bq1SqMGjUKxcXF2L9/P7p06VLj9ufVdkwAMDMz494LBAJIpdJav6eOjg7mzZuHzz//HH5+fvU+TpWRI0di/vz5KCkpwfbt2zFu3DiuLDc3FwsWLOBWVwQAK6ua/42tuhwAAL1794ZAIEBmZiZsbW1rjSc0NBQzZ85EdnY2NmzYgO+//75eMdSnv1VBSYAKqlYQbCy5jx7BV+IFeVkppg0Kxog5i2HUpu7snRDSdOr6td6YIiMjsXDhQvj5+cHNzQ0SiQSFhYVKdZKSkrBq1Sq0b98ecXFxSrPZLSwskJSUBAAoKipCTEwM/P39AVQmGD/88AMAYNasWYiKikKXLl1q3F7fY76u0aNHY/ny5fjzzz+51QvrexxTU1P4+/tj48aNOHfuHHbu3MmV9ejRA4mJia+1zkHVpQcDA4M649HU1MT48ePx3//+F4wxeHl51SuG+vS3KigJUIGMp9Vo8wFysrPh29ULKfcfwlikh4CI/1ACQAhRMnnyZGRlZaFbt26wsbGBoaEh7t+/j88++ww6OjoAgHHjxsHPzw/Ozs6QyWRKvzarJpwdOnQIpaWlsLCw4Mqqfu2WlpYiLS0Nhw8frnX782o75uvi8XhYtGgRQkNDuSTgVY4THh6OYcOGYdSoURAKhdz2NWvWYPDgwdizZw9sbW0BALNnz0bfvn2rbadqTkB5eTlu3ryJhQsXwtDQsF7xTJo0CXZ2dtixY4fS9tpiqE9/q0KDPT9d8Q1VWloKoVCIkpIS6OrqqtRWdvZ1/DYvDwBQkr4LVwZPwp4I74YIk5OTnQVfiRdSHmTBRKSHU6dOoXO37g16DELIqysrK0NqaiqcnJy4k2xzUFxcjJs3b0JXVxeurq4QCARK5SkpKSgvL4ebmxuuXLkCDw8P7t/Cp0+fIjU1FR07dkR6ejoMDQ3Rpk0bnDlzBgAgEonQoUMH7sRZ0/Y7d+5AIBDAxsam1mPevn0bIpGIG+6+e/cudHR0uP2ed+/ePTDG4OTkxG2LiYmBpaUlN0GupuO8GI9cLkdsbCzc3NxeOjkrFAqkpKQgOzsbjDG0b98eYrG42njS09MBVF6icHFx4a7r16evHz16BH9/f9y4cQNaWsq/wWuKoab+fl51f5f1Pe9REvCKnk8CCjL3IPP9SKwb5VXHXvX3OOshfLtKcOdBFkz1RTh16hQ8u3ZrsPYJIa+vuSYBpPmbNWsWduzYgfnz5zf4swVUSQLocoAKGvpyQE52FnwkXrj7MBum+iL89ddf6CSpecYsIYSQlmHUqFEIDw9XuiWzOaAkQAWVSUDDTAyUFhfh+PdLINLUgKm+CKdPn4aHl6RB2iaEEKJenTp1UncI1aInBqpAxtOCUQOsICgtKsK+hV8h595dTBkYhAsXLlACQAghpNFREqACGU9T5VsEszIzMaCXDzKSk2AoboP35y9FW3ePBoqQEEIIqRldDlCBTFO1OQEPMtLh160r0h7loKRUit837oSBmXndOxJCCCENgEYCVKDKnIAH6Wnw7VqZAJgbGmDrr/soASCEENKkKAlQQeUTA199JOB+Whp8unVF+uMcWBgaICY2Fm4ezXPSCCGENKZz587hq6++AgDMmTMHUVFRTXLcRYsWVbv40utatWqV0qqALQUlASqQ8bRg+IpJQMa9VPh264qMx7kQGxngXGws2tEcAELIa1i5ciVcXFzg6uqKLl264P333+eWAm4pPv/8c4wcORIA8PjxYxQUFDTJcXNycpCfnw+gYRKC999/H19++SVa2qN3KAlQQeXdAfW/HFDyrAAfjRyOjJxciI0MERN7nhIAQshre/LkCby9vXH8+HH8+OOP6Nq1Kz744AN8++236g6tXqpW22vfvr1a44iMjMTgwYNVasPCwgL29vY4depUA0XVNCgJUIGmjja0terXhSXPCrB3wWz42Vgg2MsTMefPw7WjeyNHSAh50xkaGsLFxQXe3t6YPn06tm7dinnz5qGsrAz/+c9/lEYGvvjiC27Ies6cOViyZAlGjBgBT09PrFq1CgkJCXjnnXcgkUiwd+9ebr/61s3JyUH37t0hl8u5fT/99NMah/hPnjyptCogAGRnZ2PEiBGQSCRYvHgx98v6hx9+gIuLC9zc3BASEoK4uDhun3PnziEwMBBt27aFi4sLYmJiAFQ+KnjJkiXw9vaGt7e30sp9z9uwYQO3HPCcOXOwatUqjBw5El26dMG8efO4enW15+/vjxMnTlR7jOaK7g5QgY5QUHclAOl37+D4qqUofpQFcxs77Nv0M/SMjBs5OkJIU/h2xIBGa/u/vxx55X369euHkpIS3Lp1C9nZ2SgqKuLKHj9+jGfPnnHvT548iY0bN0JTUxODBw/GoUOH8PXXXyM3Nxfh4eHo378/hEJhveuam5tDLBbj0KFDGDJkCAoLC7Fz50588cUX1caakpICV1dXpW0bN27E5s2bYWhoiMjISFhaWiI8PByjRo1CSEgIKioqkJCQgBEjRuDu3bvg8XgIDw/HkiVL0LlzZ/B4PG5dgnnz5iE/Px/r1q2DVCrFzJkzYWVlheHDhysdMycnB3p6ely/7Ny5E+vXr4ehoSE++OAD9O7dG/7+/nW25+DggAsXLrzyfzN1oiRABbp6dScBaXdS4NejO/gawMwRgzF87mJKAAghjYbP54PP50Mmk9VZd8qUKQgMDAQAdO/eHeHh4fD19QVQuZ59Wloa95jb+tadOnUqvvvuOwwZMgQ///wzhgwZwq2y9yKZTPbSQjqTJk1CSEgIAOCrr75CVFQUwsPD8eTJE8yZM4e7hPDgwQM8fPgQNjY2MDMzw5kzZyAUCtG7d29ukZ1du3ZBQ0ODGw0pKCjA+fPnX0oCquuXoKAgAMCAAQNw8+ZN+Pv719leffu9OaEkQAW6wtoXI0q9nYyePt54+OQprEyN0W/ap5QAEPKGeZ1f640pISEB5eXlcHJyAo/HU5qo9uIJSl9fn3vP5/Nf+lxeXv7Kdd955x1MmzYN9+7dw8aNG7F79+4aY7WxscGjR49qjMnAwAClpaUAgLFjxyI4OBiff/45hEIhevbsCalUCgA4duwYtm7dijVr1mD8+PHYu3cv/P39UVhYiL1798LOzo5rs6aE5HkGBgbVfre62svOzuaWAm4paE6ACoS1jASkJt+Cn3dlAmBtaozzFy/B0bVtE0ZHCGltkpKSMHHiRISFhcHExATW1ta4ePEiACArK6tJJq1paGhg8uTJmDx5MszNzeHm5lZj3Z49eyI+Pl5p2549e1BUVAS5XI4tW7agR48eAIAHDx6gX79+6NSpE5KSkpCdnc3tY2ZmhlmzZuH333/H2LFjcezYMQBAnz59EBUVBTs7O7i4uFS79O+rqKu9uLg4+Pv7v3b76kBJgAqEoupHAu4k/Qs/b29k5T2FjZkJYi9egoOLa7V1CSFEFbt27YKLiwuMjIzg4+MDb29vbNq0CQAQERGBX375BZaWlggJCWmyWfjh4eGIjY3FJ598Umu93r17IzU1FU+ePOG2VU3+Mzc3x+PHjzF9+nQAlUvx+vv7w9raGt9//73Sr/GqE7K1tTUOHDiAcePGAQBWr16NjIwMmJqacnUOHjz42t+rtvbKy8tx9uxZhIaGvnb76qDBWtpNja+hvusq10d29nX8Ni8PAKDhr4UP3++lVJ52JwXe3boh+2k+bMxMEXvpEuydnFU6JiGkeahu3XZ1ysvLQ15eHjQ0NKCnpwexWAwNDQ2lOhUVFcjNzYVYLEZOTg4EAgH09fWRk5MDHR0dbug7KysLhoaG3PX0zMxMiMViaGtrv1JdALh79y4GDRqExMTEl+J50caNG5GdnY05c+ZwxxGJRMjLy4OZmZlS3cLCQigUChgaGiIjIwOWlpbg8/m4c+cOAEAkElXbB8XFxcjOzgZjDG3atIFIJEJubi74fD4MDQ2V3r/4XZ88eQItLS2lYf/q2nv+ezS16v4u63veoyTgFT2fBIj6G2DswP9b7a8o7wn2zPsMPx48htzSMsReioOdo5NKxyOENB/NLQlojqquyW/btg1Dhw6ts35FRQWys7NhbW3dBNE1nuzsbJiYmHCJUFNSJQmgywEqMDAQcu8L83Lx69efo+BRFj4eFoq/L1+hBIAQ0uosWrQIGRkZ9UoAAEBTU7PFJwAA0KZNG7UkAKqiJEAFRoaVSUDyjUQE+XojKyMD5g5OCJu7GNb29mqOjhBCmp6lpSWMjekuqJaCbhF8Rc9fPTEw0ENSwj8I6NULjwueQaAjwNEde6Er0q+lBUIIIaR5UNtIwKlTpxAQEABnZ2eEhoYiOTm5Qes3mooK7u3TzBT49+qJxwXP4CA2x85DRygBIIQQ0mKoJQlITExE//798e677yI6OhpisRh9+vRBYWFhg9RvVOWVz8R+8uwu3n83GDkFhXBsY4ELl6/A2o4uARBCCGk51JIErF27Fr169cL06dPh4eGBtWvXQiaT4ddff22Q+o2porwcT/LvYNWh/4fcZ4VwbCPG+cvxsGxhT4kihBBC1JIExMfHc8+cBgAtLS107979pSdHvW59mUyG0tJSpVdDefwwB2f++R55xcVwshTjQnw8LG0oASCEvLm8vLxw+/ZtdYdBGoFakoCnT5/CxMREaZuJiQmePn3aIPUXLVoEoVDIvUxNTRsmcAAlhcXo59ERwR5v4UL8FbSxtmmwtgkh5FXMnz8fAoEAurq6sLGxwaRJk7jn6TeksrIyKBSKBm+XqJ9akgChUPjS9fxnz55xSzmqWn/27NkoKSnhXs8/klJVPgM+QNjXs7HxwG6IrVr+va2EkJZLJpNh3LhxePr0Kc6cOYNLly5h7dq16g6LtCBqSQLc3d2RkJCgtO3GjRvo2LFjg9Tn8/nQ1dVVejUUnqYmbN06w9a5XYO1SQghr0tTUxMCgQAuLi7csrdVQkJCIBAIYGxsjD59+ijdVeXl5YVvv/0W7u7usLa2xpo1a7iy69evo1u3brCyssKMGTOURgFu3LiBgIAAmJmZwcfHB3FxcUptzp8/Hx07doSVlRWioqKwdu1aWFtbw8XFBefOnavxe1y9ehVdu3bljtmlSxfuEoSLi4vSaoOenp5ITU0FAKSnp2PQoEGwsLBAu3btsH37dq7e119/DRsbG+jq6kIgEKC4uLjW7a0SU4Pjx48zXV1ddv78ecYYY5s2bWICgYBlZmYyxhg7ePAgs7a2rnf9upSUlDAArKSkpIG/CSGkNZFKpezff/9lUqlUaTuAGl8bNmzg6m3YsKHWuq9q9uzZbOrUqYwxxnJycphEImHffvstV15eXs5KS0tZQUEB27x5MwsODubKOnbsyN577z324MEDFh8fzwwNDVleXh6Ty+XMxcWFrV69muXk5LAVK1YwACwpKYnJ5XLWtm1btnz5cpaTk8PWr1/PLC0tWXFxMdfmsGHDWFZWFjt69CjT1dVlH330EcvJyWGbNm1iEomk2u8hl8uZk5MT+/HHH186JmOMWVtbs6ysLK5+u3btWEpKClMoFKx79+7s4MGDLD8/n924cYO5urqypKQklpaWxmxtbVlqaiorLS1lpaWljDFW4/aWrLq/y/qe99SSBDDG2NKlS5menh4TiUTM0tKSHTp0iCuLiop66X+I2urXhZIAQkhDaI5JgKamJtPR0WEAWHBwMJPJZFz5zz//zDp27MhEIhHT0dFhNjY2XFnHjh3ZpUuXuM8SiYRdu3aN3b59m9nb2ysdRywWs6SkJJaSksLatGmjVNa+fXt28eJFrs24uDiuTCQSsYcPHzLGGMvPz2eGhobVfo/bt28zBweHao/JWM1JQGpqKtPQ0GA6Ojrci8fjsaioKPbs2TNmbW3NIiMjWVRUFMvPz2eMsRq3t2SqJAFqe1jQrFmz8PTpU6SmpuL+/fsYOHAgVzZo0CBkZmbWuz4hhKgTq/xBVe0rIiKCqxcREVFr3dcxadIk5Ofn49q1a7h16xY35F5QUICpU6diw4YNePDgAa5fvw6ZTKa0b9UqgEDlZQW5XA6FQgFNTU2lelWfFQoFeDzl04aWlhYqnnuI2vOXXzU1NbnPVe1Xp6Ki4qVjamn93wNtNTU1lY5RNflRLpfDzMwM+fn53Ku4uBjDhw+Hvr4+rl69Cg8PD+zbtw9ubm5IT0+vcXtrpda1A/h8PszNzV/6o6qa6Vrf+oQQ0lpVzQno3Lkz1q5di48++ghyuRwlJSXQ0NCAra0tNDU1sXz58nq15+zsDLlcjl27dkEmk2Hz5s14+PAhAMDJyQl8Ph+bN2+GTCbD3r17kZWVhU6dOqn0HZydnSGVSvHLL79ALpdj27ZtePDgAVfu4OCAo0ePoqKiAr/88gt30nZ2doaZmRmWLFkChUIBgUAAgUDAnSMsLCwwdepU/PTTT7C1tcWNGzdq3d4a0dmUEELeECEhIbC2tsa6detgaWmJyMhItGvXDq6urjA3N69XG1paWti9ezfmz58PY2Nj/P3332jXrh1XtnfvXqxZswZ6enqYM2cOfv31V4hEIpXi5vP5+Pnnn/HFF1/AyMgIsbGxsLW1hYGBAQBgwYIFWLBgAYyNjXHu3Dk4OzsDAHg8HqKjo3HhwgWYm5tzSUB6ejouXLjAfRaLxejQoQP69u1b4/bWSoO97hhUC1LfdZUJIaQ21a3brk5Vw+vPD51XVFRAoVCAz+cDqLxUoaGhAaAy/qq4y8vLwefzubIXPz+/b21lz3ux3vPHq+5zdWQyGQ4cOICZM2ciLS3tleIpKysDAO4YZWVl0NDQUDpmVb0Xt7dk1f1d1ve8R6sIEkJIC/X8yb+Kpqam0vX150+Uz5/0tLW1lfZ78fPz+9ZWVlsbL55kazvpLly4EAsXLoRCoUDbtm2xc+fOl45RVzwCgUBp24ufa6rXmlESQAghRO0+++wzzJgxA3w+/6VJgqTxUBJACCFE7bS0tKod2SCNiyYGEkIIIa0UJQGEEPKKWsF8atKCqPL3SGMvhBBST3w+HzweD7m5uTAzM6t2chwhTYkxhtzcXPB4PO6OkFdBSQAhhNQTj8eDra0tMjMzX1rZlBB1qfq7fJ0H6VESQAghr0AoFMLV1fWlR/ASoi5VI1Svg5IAQgh5RTwe74150Axp3WhiICGEENJKtYqRgKqZk6WlpWqOhBBCCGl8Vee7uu4caBVJQNWyk6ampmqOhBBCCGk6UqlUacnoF7WKBYQUCgXy8/MhEAga5Jae0tJSmJqa4smTJ7Qg0WuiPlQd9aHqqA8bBvWj6hq6DxljkEqlMDIyqnXSYKsYCeDxeDAxMWnwdnV1dekPXkXUh6qjPlQd9WHDoH5UXUP2YW0jAFVoYiAhhBDSSlESQAghhLRSlAS8Bi0tLcydO5dWvFIB9aHqqA9VR33YMKgfVaeuPmwVEwMJIYQQ8jIaCSCEEEJaKUoCCCGEkFaKkgBCCCGklaJZHDUoLi5GTEwMFAoFevbsCX19/Qat3xrIZDLExsaisLAQPXr0gIWFRa31nzx5gsuXL0MoFEIikdTrHtc3HWMMcXFxyMrKQqdOneDk5FSv/e7fv4/Y2Fh07doVzs7OjRxl85eYmIg7d+7AxcUFHh4eddYvLS3FhQsXAAB+fn60WBCAtLQ0XL9+HW3atEH37t3rfPBaUlISUlJSYGJigu7du7/WWvdvksOHD6O4uBgAYGNjAz8/vzr3efz4MS5evAh9fX34+vpCW1u74QNj5CWJiYlMLBYzLy8v1r17d2ZmZsYuX77cYPVbg+zsbNahQwfWrl07FhAQwEQiEdu/f3+N9f/73/8yGxsbFhwczDp37swsLCzYuXPnmjDi5kcqlbJ33nmHWVtbs759+zKRSMSWLl1a534VFRXMz8+P8fl8tm7duiaItHmLjIxkxsbGLCgoiJmYmLCJEyfWWv/EiRPM3NycSSQSFhISwiQSCcvOzm6iaJunlStXMpFIxPr27ctsbGxY7969WUlJSbV1FQoFGzlyJDMxMWEDBw5k7du3Zw4ODiw1NbWJo25ePvroIzZixAjm7OzMBg0aVGf96Ohopq+vzwICAli7du2Ym5sby8rKavC4KAmohq+vLxs7diz3eerUqaxz584NVr81mDBhAvPz82MymYwxxtjq1auZiYlJjf9wbNmyhZWVlXGfIyIimKenZ1OE2mytXLmS2drasry8PMYYY6dPn2Y8Ho8lJyfXut+yZctYeHg4s7e3b/VJwPHjx5lAIGApKSmMMcbu3r3LdHV12ZEjR6qtf//+faavr8+2b9/Obbt58ya7d+9eU4TbLKWmpjItLS12/Phxxhhj+fn5zMnJiS1ZsqTa+jExMUxDQ4PrM4VCwXr27MkmT57cVCE3a5GRkXUmAaWlpczU1JStXLmSMcaYTCZj/v7+LDw8vMHjoTkBL3j8+DHOnz+PKVOmcNumTJmC69evIy0tTeX6rcWBAwcwceJE7p7X8ePHo6ioCGfPnq22/vjx45WGujw9PZGbm9sksTZXBw4cwIgRI2BsbAwACAgIQNu2bXHo0KEa9/n333+xbt06rFy5sqnCbNYOHDiAwMBAuLi4AACcnJwQFBSEAwcOVFt/27ZtcHJywsCBA3Ho0CFcvHgRbdu2hYODQxNG3bwcOXIENjY2CA4OBgAYGhpi5MiRNfZheXk5+Hw+rKysAAAaGhqws7NDeXl5k8Xc0sXExKCgoAATJ04EUPkMgUmTJtXY56qgOQEvuHfvHgAoXXt1dHQEAKSmpr70j8Gr1m8NCgoKkJeXp9QnQqEQYrEYqampde5fVlaGzZs3Y8iQIY0ZZrOXmpqKsLAwpW2Ojo419qFcLsfYsWPx/fffw8jIqAkibP5SU1PRoUMHpW2Ojo64fv16tfWvX78OLS0tdO/eHe3atUNycjJ0dXXxxx9/QCwWN0HEzU9qaupLc1Fq+zsMCAjAkCFDMGTIELz33nu4c+cO4uPjcfjw4aYI942QmpoKCwsLiEQibpujoyP3b2tDroVDIwEvkMvlAKD0q7TqfVWZKvVbg+r6BAB0dHTq7BOZTIawsDAIBAIsXbq00WJsCeRy+Sv14aJFi+Dq6op33323KcJrEV61D8vKypCQkIDDhw/jyJEj+Pfff6Grq4t58+Y1QbTN06v2oUKhgIODA1JSUhAdHY3jx4/DycmJnib4Cmrq86qyhkRJwAuqsv3s7Gxu26NHj5TKVKnfGhgZGUFHR0epTxhjePToUa19IpVKMWTIEOTl5eHkyZOtfjUysVis1IdA5d9ZdX346NEjLF68GD4+PtizZw/27NmD4uJixMfH49SpU00VcrPzKn1YVd/FxQVubm4AAD6fj6CgoBpHDlqDV+3DTZs2YefOnYiPj8fBgwdx9epV2NvbIzw8vAmifTOIxWLk5ORAoVBw27Kzs8Hn87nLgw2FkoAXODs7w8bGBkePHuW2HTlyBKampujYsSMA4OzZs7h06VK967c2mpqa8PX1VeqTc+fOobi4mLst5tq1a/jzzz+58uLiYvTv3x8ymQy///670jBYa+Xv76/Uh48ePUJ8fDz8/f0BVA4Z/vbbb1x5aGgoYmNjER0djejoaJSUlOCff/5BTExMk8feXPj7++OPP/7grkfLZDKcPHmS68Pc3Fzs2bMHpaWlAIDevXvj4cOHKCkp4dpISUmBtbV10wffTPj7+yMhIQGZmZnctiNHjnB9KJPJsGfPHi5RyMrKgoWFhdJt0i4uLi8lEkTZwYMHcfv2bQCVt6VKpVKcOXOGKz9y5Ah8fHwa/lbLBp9q+AbYsWMH09XVZQsXLmSLFy9mIpFIaZZ1UFCQ0t0AddVvjWJjY5mOjg6bPn06W7VqFbOxsWHTpk3jyqdNm8a6d+/Offb29mZt2rRhP/30E4uKiuJeCoVCDdE3DxkZGczMzIyNHDmS/fjjj6xLly4sMDCQ65NNmzYxPT29GvenuwMYKy4uZm5ubiwwMJCtX7+e9e3bl7Vt25YVFRUxxipnsgNgmZmZjDHGysvLmUQiYb1792abNm1i06ZNY7q6uq3+lt+QkBDWqVMn9uOPP7IxY8YwY2Nj7pa/p0+fMgDc3QOJiYlMV1eXjR49mm3bto0tWLCAGRoaskWLFqnzK6hdTEwMi4qKYoGBgUwikbCoqCj2xx9/cOVisZi7G4AxxqZPn86sra3ZqlWr2IwZM5iOjg47e/Zsg8dFF2mqMXr0aFhaWmLv3r1gjOGXX35BSEgIVx4QEABDQ8N612+NfH19ceHCBWzfvh3Xrl3DokWLMHr0aK78rbfeUvqlYGdnBzs7Oxw7dkypneHDh9f5UJI3la2tLeLj47Fu3TpcvHgRY8aMweTJk7n+cHZ2xtChQ2vcf+DAgdys+NZKKBTi/PnzWLt2Lc6fPw8/Pz/s2bMHenp6AABzc3OMGDGCezAVn8/HmTNnsH79esTGxsLKygr//PMPXF1d1fk11G7//v3YsGEDLl68CLFYjPj4eG4CtLa2NkaMGAFLS0sAgLu7OxITE7Fjxw6cOXMGRkZGiIqKQr9+/dT5FdTu0qVLuHz5MkxNTWFqaoro6Gi4uroiMDAQQOVIXrt27bj6K1asQJcuXXDq1Cno6+vj/Pnz8PLyavC4aBVBQgghpJWiOQGEEEJIK0VJACGEENJKURJACCGEtFKUBBBCCCGtFCUBhBBCSCtFSQAhhBDSSlESQAghhLRSlAQQ0gIkJSXB19dX3WEQQt4wlAQQ0gIUFxfjypUr6g7jlX366afYuHGjynUIIY2DHhtMSBMLDAxEfn6+0rYJEyZgypQpaonn/v37CA0NBQDweDxYWlqif//+mDBhAjQ1NVVqOzIyEgKBgPs8Y8YMdOjQAePHj6+xTkNbuHAhoqOjAQACgQD29vaIiIjgFsCpjx49emDHjh1o27ZtI0VJiHpQEkBIE7t+/TqmTZum9Cx1KysrtcUjlUpx5coVHDhwANbW1khMTMT06dORlZWFuXPnqtS2k5OT0uc7d+7AyMio1joNLS0tDYaGhli6dCnKyspw8OBBBAYGIi4uDl26dKlXG/Hx8UorCxLypqDLAYSogaOjIyQSCffS0NDg3vv7++OTTz5BTk5OjfuXl5dj4cKF6Nu3LwYMGICdO3cqlefm5mLGjBl4++23MWzYMPzxxx91xuTu7o6uXbti/Pjx+OSTT7Br1y6ubMuWLRg0aBCCgoLwzTffcEvzAsCVK1cwatQo9OrVC5MnT1Zacvb5of5Vq1bh7NmzWLduHSQSCfr37/9SnRMnTmDgwIFKcclkMvj7+yMuLo7bdvjwYYSFhaFPnz6YOXMmCgoKav1uxsbGkEgk8PX1xbJly2Bubo7Tp09z5f3794dEIoG3tzdGjx6NhIQErmzo0KGoqKjABx98AIlEgsWLF3NxLV++HP3790f//v2xfv160FIspKWhkQBCmgFTU1OsX78eAFBUVITt27ejb9++uHbtWrWrKC5evBj79+/H4sWLoampiT179sDY2BgDBgxAfn4+OnfujIEDB+LLL79Eeno63n//fezevRvvvPNOveIxNzfnTqxfffUVtm/fjhUrVkBfXx9fffUV4uLiEB0djbKyMrz99tv4+OOPMXnyZNy+fRuTJ0/G0aNHAQB3796Frq4uAOC9997DgQMHuMsB2traL9Xx8fHBkCFD8Pfff8Pb2xtA5Qk/OTkZb731FgBg9erVWLZsGRYtWgQ7Ozts2bIFgYGBuHTpEni8un/X3Lp1C0+ePIG9vb1Sf5aXl6O8vBxnz56Fn58fkpOTYWlpifnz5yM6OhqzZ89G27ZtYWFhAaBylcby8nLMnDkTCoUCn3/+ObKysjB//vx69TEhzUKDL05MCKmVqakpc3R0ZF5eXtzr0aNHSnUqKiqYlZUVi4uLY4wxdvnyZaajo8OVR0REsA8//FBpH6lUyhhjbOHChaxXr15KZStXrmRBQUHVxpOSksIAsJSUFMYYY8XFxczHx4f179+fyWQyJhAI2MGDB7n6d+7cYQDYzZs3WW5uLtPQ0GAZGRkvxcEYY++99x6bO3cu93nQoEFswYIFSsd/sc6YMWNYREQE93ngwIFsxowZXL8YGBiwY8eOceUymYxZW1uz06dPV/v9JkyYwIyNjZmXlxdzd3dnOjo6bM6cOUyhUFRbnzHGhg4dyr777jvus6amJrt27Rr3+cyZM8zAwIAVFRVx2y5fvsxEIlGt7RLS3NBIACFqMG7cOKU5AUZGRti8eTOOHDmC7OxsyOVy5OfnIz09HV27dn1p/y+++AIRERHo1q0bfHx8MGDAAG5d8ps3b+LWrVuQSCRc/YKCAigUilpjGjx4MLS1tXHv3j3Y29vjp59+QmZmJqRSqVIMzs7OMDY2xu3bt9GhQwesXbsWQUFBcHNzg5+fHz744AOIxWKV+mbQoEH4/vvv8ezZMxw/fhxLly4FUDmJ8dmzZ/jss8/w1Vdfcfvk5+cjJSUFAQEB1bbZpUsXLF26FAUFBVi0aBFiY2OhUCi4iY9//vkntm7dioyMDEilUmRmZsLa2rrGGG/evAm5XK40ubCiogJFRUXIzs6GpaXla39/QpoSJQGEqEHVnIAqa9euxbfffotvvvkGtra20NHRwbBhwyCVSqvd397eHidOnEBJSQkuXLiA6dOnY9SoUZg1axZMTEzQo0cPpZMkAG74vSYLFiyAra0t2rRpAysrK2hoaODZs2cAKucYVJ3YysrKUFhYCFNTUwDAlClTMGXKFKSlpeGXX36Bu7s7kpOTYWJi8tIxqru08SJ/f3+YmZlh//79yMrKgkQiQfv27QEAJiYm0NDQwLx582Bra/tSn9Skak4AALz11ltwdXXFli1bEBERgWvXrmHw4MFYsmQJJk+eDKFQiEWLFin1/Ytxm5iYwMzMjLuE87yqfiGkJaCJgYQ0A0lJSZBIJAgLC4Ovry9yc3ORkZFRY/21a9ciIyMDQqEQgYGBeOutt/DPP/8AAIYPH47Y2FjweDxusqGenh6uXbtWawzu7u7w8vKCtbU1d9IzMDBAz549sWTJElRUVACovH5uYWEBLy8v3L17F1u2bIFCoYCDgwNGjRqFvLw8PHjwoNpjGBsb1zrhEag84Y4dOxbbt2/HTz/9hHHjxnFlIpEIISEhOHjwINzd3SGRSNC5c2fExcUpTVasjbGxMWbOnImFCxeivLwcycnJMDMzw8SJE9GrVy+YmJjg4sWLtcYdGBiI0tJS7r+bRCKBo6MjTp8+XWeyRUhzQkkAIc1AREQEYmJiYG9vD1dXV3z99de13jrn4OCA3r17w8XFBXZ2doiNjcWsWbMAAL169cKKFSsQFBTElQ8fPhwuLi6vFdvWrVtx69YtWFpawt7eHlu3bsWePXsgFAohFotx6dIlmJiYwMPDA+3bt0dkZCTc3d2rbWvkyJHYtm0bOnXqxN0dUJ3w8HD89ddfSE1NRVhYmFLZli1bUFBQALFYjI4dO8LMzAxJSUmv9Av8o48+QmlpKbZs2YKQkBCYmZnB2toaHTt2RN++feHm5qZUf+LEiRg6dCi8vLywePFimJmZ4cCBA1i8eDGsrKzQrl07uLu7w9zcvN4xENIcaDBG97QQ0pT++ecf2NravjRcXl5ejjt37kAoFMLBwQFJSUkQi8UwMTFBSUkJbt26xc2QBwDGGFJTU8Hj8eDg4PDSkLVCoUB6ejr4fD5sbGxqjKesrAyJiYnw8PCAjo5OjfUePnwIqVQKR0fHl45VVFSE9PR02NjYwNDQkNuempoKgUCg9ByEwsJCpKWlgTGGTp06VVsHABITE8Hn8186IT/fzv379+Ho6Fjrw4bS09NRUVHxUlKVmpqKiooKuLq6cn1ZXl6Otm3b4uHDh2CMwc7OjqufnZ2NrKwsmJqavrS9qKgITk5O9bo7gZDmhJIAQgghpJWitJUQQghppSgJIIQQQlopSgIIIYSQVoqSAEIIIaSVoiSAEEIIaaUoCSCEEEJaKUoCCCGEkFaKkgBCCCGklaIkgBBCCGmlKAkghBBCWilKAgghhJBW6v8D2UTOmnk7fTIAAAAASUVORK5CYII=",
      "text/plain": [
       "<Figure size 576x576 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "from sklearn.metrics import roc_curve\n",
    "\n",
    "if min_class_count >= min_per_class_for_split:\n",
    "    plt.figure(figsize=(6, 6))\n",
    "    for name, model in models.items():\n",
    "        model.fit(X_train, y_train)\n",
    "        probs = model.predict_proba(X_test)[:, 1]\n",
    "        fpr, tpr, _ = roc_curve(y_test, probs)\n",
    "        plt.plot(fpr, tpr, label=name)\n",
    "    plt.plot([0, 1], [0, 1], 'k--', label=\"Random guess\")\n",
    "    plt.xlabel(\"False Positive Rate\")\n",
    "    plt.ylabel(\"True Positive Rate\")\n",
    "    plt.title(\"ROC curves: contact_interest classifiers\")\n",
    "    plt.legend(fontsize=8)\n",
    "    plt.show()\n",
    "else:\n",
    "    print(\"Not enough sessions per class for a held-out ROC comparison yet.\")"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "c318f636-2997-4894-9553-50ef518556fd",
   "metadata": {},
   "source": [
    "### Tuning KNN\n",
    "\n",
    "KNN specifically has two things worth tuning: which distance **metric** it uses, and whether features are **scaled** first. Scaling matters here because KNN is distance-based, and `duration_minutes` (ranging up to ~100) would otherwise dominate the distance calculation over `links_clicked` (ranging 0-5).\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "id": "d17255be-571a-4571-9230-44332616b0af",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "{'euclidean': np.float64(0.9632996632996633), 'manhattan': np.float64(0.9632996632996633), 'cosine': np.float64(0.9303030303030303)}\n"
     ]
    }
   ],
   "source": [
    "if contact_classifier is not None and min_class_count >= cv_folds:\n",
    "    n_neighbors = min(5, min_class_count - 1)\n",
    "    metric_scores = {\n",
    "        metric: cross_val_score(\n",
    "            sklearn.neighbors.KNeighborsClassifier(n_neighbors=n_neighbors, metric=metric), X, y, cv=cv_folds\n",
    "        ).mean()\n",
    "        for metric in [\"euclidean\", \"manhattan\", \"cosine\"]\n",
    "    }\n",
    "    print(metric_scores)\n",
    "else:\n",
    "    print(f\"Need at least {cv_folds} sessions in the smaller class to compare KNN metrics.\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "id": "632b2a44-b55b-4132-bcf4-d85971829df6",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "KNN mean CV accuracy — unscaled: 0.96, scaled: 0.96\n"
     ]
    }
   ],
   "source": [
    "from sklearn.preprocessing import StandardScaler\n",
    "\n",
    "if contact_classifier is not None and min_class_count >= cv_folds:\n",
    "    X_scaled = StandardScaler().fit_transform(X)\n",
    "    n_neighbors = min(5, min_class_count - 1)\n",
    "    unscaled_score = cross_val_score(sklearn.neighbors.KNeighborsClassifier(n_neighbors=n_neighbors), X, y, cv=cv_folds).mean()\n",
    "    scaled_score = cross_val_score(sklearn.neighbors.KNeighborsClassifier(n_neighbors=n_neighbors), X_scaled, y, cv=cv_folds).mean()\n",
    "    print(f\"KNN mean CV accuracy — unscaled: {unscaled_score:.2f}, scaled: {scaled_score:.2f}\")\n",
    "else:\n",
    "    print(f\"Need at least {cv_folds} sessions in the smaller class to compare scaled vs. unscaled features.\")"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "68249ad1-28c4-464c-9cb5-7896ca5f48dc",
   "metadata": {},
   "source": [
    "### Systematic Hyperparameter Search\n",
    "\n",
    "Rather than tuning one hyperparameter at a time, `GridSearchCV` searches every combination of a parameter grid at once, cross-validating each - run here for both the Decision Tree and KNN, on both scaled and unscaled features.\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "id": "d40b2c69-9331-498f-9463-70b6845bf00e",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[unscaled] best tree: {'criterion': 'gini', 'max_depth': 2, 'min_samples_split': 2} -> 0.96\n",
      "[unscaled] best knn:  {'metric': 'euclidean', 'n_neighbors': 3, 'weights': 'uniform'} -> 0.96\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[scaled] best tree: {'criterion': 'gini', 'max_depth': 4, 'min_samples_split': 5} -> 0.96\n",
      "[scaled] best knn:  {'metric': 'manhattan', 'n_neighbors': 7, 'weights': 'uniform'} -> 0.97\n"
     ]
    }
   ],
   "source": [
    "from sklearn.model_selection import GridSearchCV\n",
    "\n",
    "if contact_classifier is not None and min_class_count >= cv_folds:\n",
    "    tree_grid = {\"criterion\": [\"gini\", \"entropy\"], \"max_depth\": [2, 3, 4, 5], \"min_samples_split\": [2, 3, 5]}\n",
    "    knn_grid = {\n",
    "        \"n_neighbors\": [k for k in [1, 3, 5, 7] if k < min_class_count],\n",
    "        \"metric\": [\"euclidean\", \"manhattan\", \"cosine\"],\n",
    "        \"weights\": [\"uniform\", \"distance\"],\n",
    "    }\n",
    "\n",
    "    for label, feats in [(\"unscaled\", X), (\"scaled\", X_scaled)]:\n",
    "        tree_search = GridSearchCV(sklearn.tree.DecisionTreeClassifier(), tree_grid, cv=cv_folds).fit(feats, y)\n",
    "        knn_search = GridSearchCV(sklearn.neighbors.KNeighborsClassifier(), knn_grid, cv=cv_folds).fit(feats, y)\n",
    "        print(f\"[{label}] best tree: {tree_search.best_params_} -> {tree_search.best_score_:.2f}\")\n",
    "        print(f\"[{label}] best knn:  {knn_search.best_params_} -> {knn_search.best_score_:.2f}\")\n",
    "else:\n",
    "    print(f\"Need at least {cv_folds} sessions in the smaller class to run GridSearchCV.\")"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "990d69f9-f407-4106-b633-370c46903e86",
   "metadata": {},
   "source": [
    "### Visualizing Sessions in Feature Space\n",
    "\n",
    "With only 4 features, PCA here isn't really reducing dimensionality - it's mainly a convenient way to check by eye whether `contact_interest` sessions visually separate from the rest at all.\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "id": "2130f509-b6b1-401f-b75c-410a574f8353",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 576x576 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "if contact_classifier is not None and len(session_metrics) >= 2:\n",
    "    X_scaled_for_pca = StandardScaler().fit_transform(X)\n",
    "    pca_2d = PCA(n_components=2, random_state=1).fit_transform(X_scaled_for_pca)\n",
    "\n",
    "    plt.figure(figsize=(6, 6))\n",
    "    plt.scatter(pca_2d[:, 0], pca_2d[:, 1], c=y, cmap=\"coolwarm\", edgecolor=\"k\")\n",
    "    plt.xlabel(\"PC1\")\n",
    "    plt.ylabel(\"PC2\")\n",
    "    plt.title(\"Sessions in PCA space, colored by contact_interest\")\n",
    "    plt.show()\n",
    "else:\n",
    "    print(\"Not enough sessions to plot a PCA scatter yet.\")"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "7477859d-c04b-47d8-8768-378ff635fe35",
   "metadata": {},
   "source": [
    "## Clustering Sessions into Behavioral Groups\n",
    "\n",
    "Everything above predicts a label we engineered ourselves (`contact_interest`). This section instead asks an unsupervised question: without any labels at all, do sessions naturally fall into distinct behavioral groups? [Visualizing the Data](#visualizing-the-data)'s executive summary speculated about \"user personas\" from a single correlation coefficient - this section checks that claim with an actual model instead.\n",
    "\n",
    "As with KNN above, these algorithms are distance-based, so features are scaled first:\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "id": "bf524b58-29f3-49fb-954a-8ecabcabc23c",
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn.preprocessing import StandardScaler\n",
    "\n",
    "cluster_features = StandardScaler().fit_transform(X)\n",
    "n_sessions = len(session_metrics)\n"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "b95eea2d-aa71-4deb-80ad-f1879e7bdecf",
   "metadata": {},
   "source": [
    "### K-Means: Elbow and Silhouette Methods\n",
    "\n",
    "K-means needs to be told how many clusters (*k*) to look for. The elbow method (inertia - how tightly packed each cluster is) and the silhouette score (how well-separated clusters are) both help pick a reasonable value.\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "id": "70fae64f-9941-4bfc-b839-ed1e75dee865",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 960x384 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Best k by silhouette: 4\n"
     ]
    }
   ],
   "source": [
    "from sklearn.cluster import KMeans\n",
    "from sklearn.metrics import silhouette_score\n",
    "\n",
    "if n_sessions >= 4:\n",
    "    k_range = range(2, min(8, n_sessions))\n",
    "    inertia = [KMeans(n_clusters=k, random_state=42).fit(cluster_features).inertia_ for k in k_range]\n",
    "    silhouette = [\n",
    "        silhouette_score(cluster_features, KMeans(n_clusters=k, random_state=42).fit_predict(cluster_features))\n",
    "        for k in k_range\n",
    "    ]\n",
    "\n",
    "    fig, axes = plt.subplots(1, 2, figsize=(10, 4))\n",
    "    axes[0].plot(list(k_range), inertia, 'bx-')\n",
    "    axes[0].set_xlabel(\"k\"); axes[0].set_ylabel(\"Inertia\"); axes[0].set_title(\"Elbow method\")\n",
    "    axes[1].plot(list(k_range), silhouette, 'bx-')\n",
    "    axes[1].set_xlabel(\"k\"); axes[1].set_ylabel(\"Silhouette score\"); axes[1].set_title(\"Silhouette method\")\n",
    "    plt.tight_layout()\n",
    "    plt.show()\n",
    "\n",
    "    best_k = list(k_range)[int(np.argmax(silhouette))]\n",
    "    print(f\"Best k by silhouette: {best_k}\")\n",
    "else:\n",
    "    best_k = None\n",
    "    print(f\"Only {n_sessions} sessions - too few to sweep k meaningfully.\")\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 33,
   "id": "9ce9ab2a-1dd2-48cc-a66c-63d2a827cd39",
   "metadata": {},
   "outputs": [],
   "source": [
    "# Fit at the chosen k and profile each cluster - this is what turns the \"user personas\"\n",
    "# language above into an actual finding instead of an eyeballed guess.\n",
    "if best_k is not None:\n",
    "    kmeans = KMeans(n_clusters=best_k, random_state=42).fit(cluster_features)\n",
    "    session_metrics['cluster'] = kmeans.labels_\n",
    "    session_metrics.groupby('cluster')[feature_cols + ['contact_interest']].mean()\n",
    "else:\n",
    "    print(\"Skipping cluster profiling - not enough sessions.\")\n"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "a3cf7e91-bbce-4a35-b8fd-d31b4ad1d5fb",
   "metadata": {},
   "source": [
    "### Hierarchical Clustering\n",
    "\n",
    "Unlike K-means, hierarchical clustering doesn't require picking *k* upfront - it builds a full tree of nested groupings, visualized as a dendrogram. This is especially readable at this dataset's small session count, where the same dendrogram on a much larger dataset would be unreadable.\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 34,
   "id": "78544e68-88f5-460f-8392-d8d53f6949ae",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAvYAAAHWCAYAAAAcpc2LAAAAOnRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjExLjEsIGh0dHBzOi8vbWF0cGxvdGxpYi5vcmcvctoD+AAAAAlwSFlzAAAOxAAADsQBlSsOGwABAABJREFUeJzs3Xd4FNX/9vF3ek8IgSR0CDXU0HvvAtJBinRBUIqAAoICCqKCoKgUKZHeBKR36b2DSYDQQyAJpJNe9vnDh/0ZBQkK38T1fl3XXmRnZmc+O2yy954556yZwWAwICIiIiIi/2rmWV2AiIiIiIj8cwr2IiIiIiImQMFeRERERMQEKNiLiIiIiJgABXsREREREROgYC8iIiIiYgIU7EVERERETICCvYiIiIiICVCwF5GXLjExEV9fX0aNGsU777zDN998Q1BQUJbUsnXrVvr165clx/4r2bWuv5Kdap4+fTrTp0//zxxXRCQzFOxF5KW6cuUKxYsXZ9q0aeTKlYvy5ctz6NAhvLy8mDVr1v+8nl9//ZWVK1f+z4/7PK+qrg0bNjBgwICXvl/IXudy37597Nu37z9zXBGRzLDM6gJExLSMHj2a1NRUTp06RY4cOQAYNGgQ69at48SJE//zetq0aUOhQoX+58fNKpcuXWL16tUsXLjwpe/7v3YuRUT+bRTsReSlunDhAlWqVDGG+ic6d+5M06ZNMyxLT09n48aNHDx4kISEBHx8fOjbty/29vYZttu5cyf79u0jNjYWb29vevXqhaura6bW37p1iz179tCtWzfj9gkJCSxdupSzZ89iZWVFvXr16Ny5M+bm/3cRc82aNRw4cIC5c+eyceNGdu3ahYuLCwMGDKB48eLPPQ/p6els3ryZAwcOkJycTPXq1enRoweWlk//s/vDDz9w584dpk6dmmH5559/jpWVFaNGjXru8121ahUbNmwgKSmJnj17ApAjRw6+++67TJ/v3z/v9evXs3fvXqpVq0bfvn3/dC5f5ByFh4czf/58bt26Rbly5Rg4cCDffvstAO+///5fnsuIiAjmz5/PzZs3jY991jl/kef3vJozc9y/Ol8AN27cYPny5dy5cwcPDw86d+5MpUqV/rSfM2fOsHLlShITE2nbti1169Zl4MCBDBo0iLp16wLw3nvvUbFiRdq2bcuCBQvw8/Pjgw8+wNvbm+nTp3Px4kUA7O3tKVy4MN26daNIkSIZjvNkH6+//jrz5s3j1q1bNG7cmC5dugBw8eJFli5dSmJiIr169aJ69ep/+X8jItmLuuKIyEtVoEABLl++TEJCwp/W/T7sx8fH06hRI4YMGYKbmxvlypVj8eLFVKtWjaioKON277//Pj169MDBwYHKlStz9+5datSoQURERKbW/7H7yMOHD6lcuTIzZsygePHieHh4MHjwYFq0aEFKSopxu/Pnz7Nu3To+++wztm/fTvny5Tl27BjVqlXjwYMHf3kO4uLiaNiwIf369cPe3p5y5cpx8OBB2rVr98zHHDt2jG3btv1p+d69e9m/f3+mzkfJkiUpVqwYFhYWtGjRghYtWtCwYcMXOt9PnveECRPYuXMn1apVo0yZMk89l5k9R6GhoVSsWJE1a9ZQpkwZHj58SPfu3dm7d+9zu7WEhYVRqVIlVq5cSenSpQkNDeWNN97AYDBk2O5Fn9/zas7scf/qfG3bto0yZcpw9uxZKleuzP3796lSpQpz587NsI+ffvqJ6tWrExISQtmyZfH19WXOnDmsWLGCGzduGLd78qGlQ4cOJCQk0LhxY3LmzAlA5cqVjf/nlSpVws/PD29vb/bs2ZPhWBs3buTQoUN0794dgLx589KrVy+++eYb9u7dy7hx4yhSpAihoaHUq1fP+GFBRP4lDCIiL9H27dsN1tbWhmLFihk++eQTw+7duw2xsbF/2m706NEGe3t7w61bt4zL4uPjDQUKFDAMHz7cYDAYDGlpaQZHR0fDt99+m+Gxjx49MsTHxz93vcFgMEybNs1gY2NjXNevXz+Ds7OzITQ01LjsxIkTBsDw1VdfGZeNGTPGYGVlZfjss8+Myx4/fmxwdnY2TJgw4S/PwYgRIwy2traG69evZ1geHBxs/PmPdfXu3dtQoUKFP+2rcePGhlatWmXqfBgMBsPEiRMNDg4Of9pPZs7375/35MmT/7SPP9ac2XP0zjvvGDw8PAwxMTHGZdu2bTPY2dkZmjdv/qfj/N7QoUMN7u7uhqioKOOyjRs3/umxL/r8nldzZo/7rPMVHx9vyJUrl6Fdu3YZlr/33nsGKysrw+3btw0Gg8GQmJhoyJ8/v6Fv374Zths4cKABMPj6+hqXFSpUyODg4GC4fPnyX50yowEDBhgqVaqUYVmhQoUMzs7OhsDAQOOyDz/80ODu7m7o3r27ITU11WAwGAzp6emGYsWKGXr37p2pY4lI9qAWexF5qVq2bMmlS5do27Yt69evp3Xr1ri5udG1a9cMLaJLly6lY8eOFC5c2LjMzs6Orl27snHjRgDMzc0xNzfn6NGjGa4AuLm5YWdn99z1f2QwGFi/fj1dunTB3d3duLx69epUr16dtWvXZtg+JSWFt956y3jfwcGBGjVq/GUrpsFgYNmyZXTt2pWiRYtmWJc3b95nPi4zXvT5/l5mzvcTKSkp9O/fP1M1ZeYcrVu3jm7duuHk5GRc9tprr+Hm5vbc/a9Zs4Zu3brh4uJiXNa2bVtjS/XffX7Pqzmzx32yvz+er/379/Po0SPefffdDMuHDRtGSkqKsaZjx45x7949Bg0alGG7Xr16PfV8VKxYkbJlyz513d69e5kwYQJ9+/alZ8+eXLx4ET8/P9LT0zNsV716dYoVK2a8X69ePcLCwmjcuDEWFhYAmJmZUadOHbXYi/zLqI+9iLx0JUuWZMaMGcBv3VLWrFnDiBEjCAwM5PTp08TGxhIWFsbFixeNfcGfCAgI4N69exgMBszMzJg5cybvvPMOW7dupWHDhjRr1ozu3bsbA9bz1v9eVFQU0dHRGULNE8WLF/9TtwVHR0dy5cqVYZmbmxvXr19/5nOPiooiPDycUqVKZe5kvaAXeb6/rymz5xvAxsaGfPnyZaqe552juLg4wsLC/tTXG3jqst978lgvL68My83MzDI89kWfX2Zrft5xn3ja+bpz5w7An15rhQoVwtramtu3bwNw8+ZN4M/n4lnn5o81PdGrVy+2bdvGgAEDqF27Nra2tlhaWnL69GmSk5OxtbU1bvv7Dz+A8bXztOWhoaFPPZ6IZE8K9iLySjk4ONCvXz/u3r3L5MmT8fPzM4aTUqVK0aJFiwzb//F+//79ef3119m+fTu//PILH330EZMmTeL48eMUL178uet/z8rKCoCkpKQ/1ZmYmGhc/4SNjc2ftjM3N/9TC+jTjpGYmPjMbZ7G2to6Qx//J2JiYjKEshd5vk88GbCbmfMNvwXfzHreObK2tsbCwuKZ5/yPA6V/73mPdXBwAF78+f3Tmp8c94mnna9nvdZSU1NJTU01rn9ypeWP2z3r9fP7qx5PXLp0iWXLlrF582batGljXH7t2rWn7sPa2jrD/ScfeJ62/K9e6yKS/SjYi8hLdfz4cWrWrPmn5U9aBWNiYnB0dKR48eKkp6f/qYX1aXLnzk3v3r3p3bs3ISEhFCxYkGXLlvHJJ59kav0Tjo6OFCpUiPPnz2dYbjAYuHDhAuXKlfu7TzvDMUqUKPHCU3vmy5ePBw8eZGhZTkxM5Nq1axm6DcFfP19LS8s/hbEXPd8vk5WVFUWKFOHy5csZliclJREYGPiXs65YWVnh5eX1p+4gCQkJBAYGGrvyvOznl9nj/pUn3WXOnz9PiRIljMsvXLhAenq68bVWsmRJAC5fvkyBAgWM2/3xfP2VJwNsK1SokGH5sWPHMr0PETEN6mMvIi/Vm2++yZQpU0hOTjYui4yMZOHCheTOnZvKlSsDMHbsWNavX8/69eszPP7+/fvG/seRkZGsXr2atLQ04/onodXDw+O565/m7bffZsuWLRw+fNi4bN68eVy/fp3Bgwf/k6duNHr0aHbt2pVhBpm0tLQ/9fX+vcaNGxMZGcmWLVuMy6ZNm4azs7Pxfmaeb548eUhISDDOCvREZs73q/Luu++ybt26DEH5q6++ytSVgXfffZeffvqJc+fOGZd98cUXf2q5ftnPL7PHfZbq1avj4+PDZ599ZpyVJykpiQkTJuDh4UGHDh2A32azqVmzJlOnTiU+Ph74rSvQi3wR2JNuX7t37zYuW79+fYYZdUTkv0HBXkReqrfeeosffviB/Pnz065dO1q0aEGhQoWIj4/n559/NnY96NevH19++SW9e/emYsWKdOzYkWrVqlG3bl1jNwQbGxu2bNlC4cKFadGiBe3ataNs2bK0atWKAQMGPHf904wePZo+ffrQtGlTmjRpQu3atRkxYgSfffZZhm4M//QcTJ48mX79+lGpUiXatm1L8eLFuXDhwjMfU6dOHbp3707Xrl1p1aoVlStXJleuXBlaezPzfNu1a4e7uzu1a9eme/fuxsGbmTnfr8rgwYNp2bIltWrVomXLllSvXp20tDS8vb2NgzWf5e2336ZVq1bUqVOHFi1aULVqVczMzIxTSj7xsp9fZo/7LGZmZvz000+kp6dTqlQpXn/9dUqVKoW/vz+bNm3K8KFm4cKFBAUFUapUKdq2bUvNmjXp168fwHPPD4C3tzdjxozh7bffpnnz5tStW5dFixYxfPjwF37eIvLvZmYw/GFSXhGRfyg9PR1/f3+uX79OWloaXl5e+Pj4GLuY/F5MTAwnT54kOjqaIkWKUL58+T/1dQ8LC+P8+fMkJyfj7e39pwGJf7Xez8+PS5cuZfiCKvjti6vOnz+PhYUFNWrU+FML/4ULF7h+/TqdOnXKsPzYsWNER0fTsmXL556H8PBwTpw4QXp6OpUrV84wK86z6jp16hR3796lYsWKFC1alH379mFhYUGDBg0yfT5iY2M5cuQIkZGR2NjY0LFjR+O6553vZz3vp9X8oufo7Nmzxi+oKlmyJCVLlqROnTosWrTouefyyWPLlCmDt7c3v/zyCwCNGjXKsN3ffX7Pq/lZx/2r8wW//S6cPHmSoKAgcufOTa1atZ7axz8xMdH4ZWb16tXj1q1bVKpUid27dxu/2O3nn38mX758VK1a9anHunLlCv7+/uTPn59q1aoRGBjIyZMn6d69u/HL1562j4iICLZv307z5s3JnTu3cfm5c+e4detWhtePiGRvCvYiIvJKJSUlcfjwYZo0aWJctmLFCnr27MnGjRv/8ou7/gvOnTuHp6en8YNfUlISPXr0YP/+/dy9e/dPg3VFRJ5Fg2dFROSVsrCwYMaMGbz33nuUKFGCO3fucOHCBUaPHv2fD/XwW7edBg0akCdPHnLkyMHZs2dJSkpi9erVCvUi8kLUYi8iIv8TV69exd/fH0tLS6pUqUKePHmyuqRsIzk5mbNnzxIUFIS7uzs1atTIMM2piEhmKNiLiIiIiJgAzYojIiIiImICFOxFREREREyASQ6eTU9PJyoqCltb26dOryciIiIikp0ZDAYSExPJkSOHccra5zHJYB8VFZWpr/wWEREREcnOwsPDyZkzZ6a2Nclg/2QmgfDwcOO3XIqIiIiI/FskJCTg5ub2QjNkmWSwf9L9xs7OTsFeRERERP61XqRbuQbPioiIiIiYAAV7EREREREToGAvIiIiImICFOxFREREREyAgr2IiIiIiAlQsBcRERERMQEK9iIiIiIiJkDBXkRERETEBCjYi4iIiIiYAAV7EREREREToGAvIiIiImICFOxFREREREyAgr2IiIiIiAlQsBcRERERMQGWWV2AiGTU1/cU+68+zOoyREREXqmGJXPj27daVpdhUtRiL5LNKNSLiMh/gd7vXr4sbbGPiooiIiKCAgUKYGVl9dT1Dx48oHDhwtjZ2WVBhSJZ5/bnrbK6BBERkVei8NhtWV2CScqSFvvr16/TqlUrSpQoQePGjfHw8GDRokUZthkzZgyenp60aNECDw8Pli1blhWlioiIiIj8K2RJsA8ICGDMmDGEhYVx69Yt5s+fz8CBA7l79y4AGzduZO7cuZw7d447d+6wZMkS+vfvz82bN7OiXBERERGRbC9Lgn2bNm2oV6+e8X7z5s1JT0/n3r17AKxYsYJ27dpRunRpANq3b0+xYsVYt25dVpQrIiIiIpLtZVkf+5SUFPz8/IiNjWXu3LnUrVuX6tWrA3DlyhV69uyZYXtvb2+uXLnyzH2lpqYa7yckJLy6wkVEREREsqEsC/aRkZH06dOH2NhYoqOj+frrr7GwsAAgMTERBweHDNs7ODiQmJj41H1NnTqVyZMnv/KaRURERESyqyyb7tLd3Z0LFy5w48YNdu3axcCBA9m5cycAOXPm5NGjRxm2Dw8Px83N7an7Gj9+PPHx8cZbeHj4K69fRERERCQ7yZJgn5ycnOF+5cqVyZ8/P1evXgWgRo0aHDhwIMP2x44dM3bV+SMrKyvs7Owy3ERERERE/kuypCtOnz59qFixIjVq1CAlJYXVq1cTGhpKq1a/zds9dOhQKlasyIcffkizZs2YM2cObm5udOrUKSvKFRERERHJ9rKkxX7+/PnEx8fz8ccfM3XqVGxtbTl16hTFihUDoHjx4uzfv5+rV68yZswYnJycOHDggFriRURERESewcxgMBiyuoiXLSEhAXt7e+Lj4/VhQP51nnwbn755VkRETJXe657v7+TZLBs8KyIiIiIiL4+CvYiIiIiICVCwFxERERExAQr2IiIiIiImQMFeRERERMQEKNiLiIiIiJgABXsREREREROgYC8iIiIiYgIU7EVERERETICCvYiIiIiICVCwFxERERExAQr2IiIiIiImQMFeRERERMQEKNiLiIiIiJgABXsREREREROgYC8iIiIiYgIU7EVERERETICCvYiIiIiICVCwFxERERExAQr2IiIiIiImQMFeRERERMQEKNiLiIiIiJgABXsREREREROgYC8iIiIiYgIU7EVERERETICCvYiIiIiICVCwFxERERExAQr2IiIiIiImQMFeRERERMQEKNiLiIiIiJgABXsREREREROgYC8iIiIiYgIU7EVERERETICCvYiIiIiICVCwFxERERExAQr2IiIiIiImQMFeRERERMQEKNiLiIiIiJgABXsREREREROgYC8iIiIiYgIU7EVERERETICCvYiIiIiICVCwFxERERExAQr2IiIiIiImQMFeRERERMQEKNiLiIiIiJgAy6w68MWLF9m5cyfJycnUrFmTJk2aZFjn6+ubYfsyZcrw1ltv/a/LFBERERH5V8iSFvsxY8bQr18/IiMjiY+Pp1u3bhlCe2BgIKtWraJw4cLGm4eHR1aUKiIiIiLyr5AlLfY9e/bkiy++MN5v3LgxTZs2ZerUqbi7uwPg6urKiBEjsqI8EREREZF/nSwJ9uXKlctw39raGgALCwvjsqioKD7++GOsra2pWbMmjRs3fub+UlJSSE1NNd5PSEh4yRWLiIiIiGRvWT54Njk5mQkTJvDGG2/g5uYGgI+PD2PHjsXV1ZWIiAi6dOlCv379nrmPqVOnYm9vb7w92Y+IiIiIyH+FmcFgMGTVwVNSUujatSuhoaHs3LkTJyenp27n5+dHuXLlOH78ONWrV3/qfv7YYu/m5kZ8fDx2dnavrH6RV6Hw2G0A3P68VRZXIiIi8mrove75EhISsLe3f6E8m2Ut9gkJCbRt25bIyEh27dr1zFAPv82IkzNnTq5fv/7U9VZWVtjZ2WW4iYiIiIj8l2RJsH/8+DGvvfYaANu3b8fR0THD+kOHDvH7CwlHjhwhPDz8T33zRURERETkN1kyeLZXr14cO3aM/v37M27cOOPy4cOHU6RIEQ4fPsw777xD1apViY6OZseOHXz00UeUL18+K8oVEREREcn2siTYd+3alXr16v1puY2NDQDjx4+nW7duHD58GBsbG7788kuKFi36vy5TRERERORfI8uC/fN4eXnh5eX1P6hGREREROTfL8unuxQRERERkX9OwV5ERERExAQo2IuIiIiImAAFexERERERE6BgLyIiIiJiAhTsRURERERMgIK9iIiIiIgJULAXERERETEBCvYiIiIiIiZAwV5ERERExAQo2IuIiIiImAAFexERERERE6BgLyIiIiJiAhTsRURERERMgIK9iIiIiIgJULAXERERETEBCvYiIiIiIibAMqsLEBEREZF/t76+p9h/9eELP67w2G2Z3rZhydz49q32wsf4L1GLvYiIiIj8I38n1GfHY/zbqcVeRERERF6K25+3eiX7fZGW/f8ytdiLiIiIiJgABXsREREREROgYC8iIiIiYgIU7EVERERETICCvYiIiIiICVCwFxERERExAQr2IiIiIiImQMFeRERERMQEKNiLiIiIiJgABXsREREREROgYC8iIiIiYgIU7EVERERETICCvYiIiIiICVCwFxERERExAQr2IiIiIiImQMFeRERERMQEKNiLiIiIiJgABXsREREREROgYC8iIiIiYgIU7EVERERETICCvYiIiIiICVCwFxERERExAQr2IiIiIiImQMFeRERERMQEKNiLiIiIiJgABXsREREREROQJcE+NDSUUaNG4e3tTdGiRenZsye3b9/OsM2WLVuoUqUKHh4eNGnShEuXLmVFqSIiIiIi/wpZEuzHjRtHwYIF2bhxI9u3bycmJoZmzZqRnp4OwLlz5+jUqRMDBw7kxIkTlC9fnqZNmxIVFZUV5YqIiIiIZHtZEuwXL17M8OHDKVWqFCVLlmTatGkEBgZy584dAObOnUvjxo0ZOHAgRYoUYcaMGZibm7NmzZqsKFdEREREJNvLFn3sT5w4gbOzM/ny5QN+a7GvWbOmcb25uTlVq1bl/PnzT318SkoKCQkJGW4iIiIiIv8lWR7s/fz8eP/995k1axbW1tYAREdHkyNHjgzbubq6PrMrztSpU7G3tzfe3NzcXnHVIiIiIiLZS5YG+8uXL9O4cWMmTJhAv379jMsdHByIiYnJsG10dDSOjo5P3c/48eOJj4833sLDw19p3SIiIiIi2U2WBftTp07RsGFDJk6cyMiRIzOsK1euHBcuXMiw7OLFi5QrV+6p+7KyssLOzi7DTURERETkvyRLgv2hQ4do0aIFM2fOZPDgwX9a369fP7Zs2cLu3btJTU3l66+/5uHDh3Tt2jULqhURERERyf6yJNiPGTOGmJgYhg0bRo4cOYy3EydOANCoUSNmzpxJ9+7dsbW15dtvv+Xnn3/G09MzK8oVEREREcn2LLPioHv27CE1NfVPy52cnIw/DxkyhCFDhpCQkKCuNSIiIiIiz5Elwf5Zg2CfRqFeREREROT5sny6SxERERER+ecU7EVERERETICCvYiIiIiICVCwFxERERExAQr2IiIiIiImQMFeRERERMQEKNiLiIiIiJgABXsREREREROgYC8iIiIiYgIU7EVERERETMDfCvaPHz9m/Pjx1KhRg549exIdHc2HH36IwWB42fWJiIiIiEgm/K1g37FjR86dO0eDBg0ICQnBxcWFoKAg1q5d+7LrExERERGRTHjhYH/jxg38/f3ZvHkzTZo0MS5v0aIF69evf6nFiYiIiIhI5li+6APCw8PJly8fVlZWGZZbW1tjZmb20goTERERkX+vIXuHcDj48Eva2+cAlFtS7iXtD+rmq8ucJnNe2v6ygxcO9qVKleLKlSvcvHkzw/L169dTuXLll1aYiIiIiPx7vbxQD07eY1/avp54mfVlFy8c7J2dnRk7dizVqlWjcuXKXL9+nSZNmnD79m0WLVr0KmoUERERkX+py70vZ3UJf/IyW/6zkxcO9gBjx46lYsWKbNy4EWdnZ8qXL8+GDRtwcHB42fWJiIiIiEgm/K1gD9C8eXOaN2/+MmsREREREZG/6W9Ndzl69GhWrFhhvP/48WMaNWpEcHDwSytMREREREQy74WDfVBQEOvWraNHjx7GZY6OjnTs2JEvv/zypRYnIiIiIiKZ88LBPjQ0lFy5cv1pubu7O/fu3XspRYmIiIiIyIt54WBfsmRJrl27xoULF4zLDAYDy5cvp0KFCi+zNhERERERyaQXHjzr5OTE+PHjqV27Np06dcLV1ZVDhw4RERHB0qVLX0WNIiIiIiLyHH9r8OzYsWP56aefsLGxITg4mPbt23Px4kVcXFxedn0iIiIiIpIJf3u6y5YtW9KyZcuXWYuIiIiIiPxNfyvY37hxg++++45bt26RmppqXO7j48OUKVNeWnEiIiIiIpI5Lxzs4+PjqVevHj4+PtSqVQsLCwvjusKFC7/M2kREREREJJNeONj/+uuvODk5sXXrVszMzF5FTSIiIiIi8oJeePBs7ty5sbOzU6gXEREREclGXjjYFylShPz587Ny5UoMBsOrqElERERERF7QCwf7Q4cOcejQIXr06IGjoyO5cuUy3jp27PgqahQRERERked44T725cqVY9OmTU9d5+rq+o8LEhERERGRF/fCwd7V1ZUGDRq8glJEREREROTv+ttfUBUeHs6lS5cICwsz9rX39PRU6BcRERERyQJ/K9gvXLiQUaNG4ezsTFRUFNbW1sTHx9O7d28FexERERGRLPDCg2ejoqIYOXIkBw4cwNfXl+rVqxMcHEytWrVo06bNq6hRRERERESe44WDfUBAAKVLl6ZixYpYWlqSnJyMra0tb7/9NitWrHgVNYqIiIiIyHO8cLCPi4vD0dERAHd3d4KCggBITU0lLi7u5VYnIiIiIiKZ8rcHzwKUKlUKgB49enDy5EkGDRr0UooSEREREZEX88It9rVr12bp0qW/PdjcnB07duDq6sqAAQMYMWLEy65PREREREQy4YWDfUxMDNeuXTPeL1WqFN999x19+/bl6NGjL7U4ERERERHJnBcO9pcvX2bKlClPXT516tSXUpSIiIiIiLyYFw72z3L9+nVy5sz5snYnIiIiIiIvINODZw8ePEjbtm1JTU0lKSmJHDlyGNelpaWRlJTE2rVrX0WNIiIiIiLyHJkO9j4+PmzdupUzZ86wYsUKZs2aZVxnZWWFl5cXuXPnfiVFioiIiIjIX8t0sHdxcaFOnTpUq1aN3r174+rq+rcPGhoayunTp433a9asiZubm/F+SEgIZ86cyfAYDw8Pqlat+rePKSIiIiJiyl54Hntra2tCQkK4ffs2FStW5ObNm0yePJmSJUsyduxYzM2f323/1q1bzJs3j/T0dHbs2MH+/ftp0KCBcf2RI0fo06dPhmXVqlVTsBcREREReYYXDvbp6el06NABX19fAAYNGkThwoVZuHAh7u7uDBgw4Ln7qFGjBlu3biUxMRE7O7unbpM/f362bt36ouWJiIiIiPwnvfCsOP7+/qSlpVGuXDni4uKIjY1lwYIFTJ48mX379r20wlJSUjh06BAnTpwgNjb2pe1XRERERMQUvXCwT0hIID09HYBjx45Rs2ZNAOzs7EhNTX0pReXJkwdvb2+++OILBg4cSP78+VmyZMkzt09JSSEhISHDTURERETkv+SFu+KUKVOGoKAghg8fzqFDh5gxYwYABw4coFGjRi+lqNq1a2fohrN8+XL69etHnTp1KFq06J+2nzp1KpMnT34pxxYRERER+Td64RZ7e3t7tmzZQlRUFH369KFx48ZER0cTERGRqf71f0fPnj1xdHTk1KlTT10/fvx44uPjjbfw8PBXUoeIiIiISHb1wi328FuLeu3atY33XVxcWLly5UsrKjIyMsN0mvfv3yc6OhpPT8+nbm9lZYWVldVLO76IiIiIyL/N3wr2/1RCQgL79u0jJSUFgOPHj/P48WMqVapE3rx5GTZsGC4uLtSqVYvo6Ghmz55N3bp1qVevXlaUKyIiIiKS7WWqK86BAwdo06aN8WdHR8en3p5s8zyxsbHMmzePRYsW0apVK44ePcq8efO4ceMGAL6+vlSsWJEdO3Zw9uxZRo0axZ49e7CwsPibT1NERERExLRlqsW+UqVKfPnll8af9+7d+9TtXFxcMnVQd3f3v5yj3tLSkv79+9O/f/9M7U9ERERE5L8uU8He2dkZZ2dn4881atR4pUWJiIiIiMiLeeE+9jdv3uS7777j7NmzJCYmUrBgQTp16kSXLl0wMzN7FTWKiIiIiMhzvNB0l+fOncPHx4cNGzaQJ08eKlasSHh4ON26dWPw4MGvqkYREREREXmOF2qxf++99+jUqRNz587FxsbGuPzo0aO0aNGCfv36Ua1atZdepIiIiIiI/LVMt9hHRUVx5swZZs+enSHUw2/z2g8aNIiff/75ZdcnIiIiIiKZkOlgHxgYSJkyZXB0dHzq+ho1anDt2rWXVpiIiIiIiGRepoN9bGyscWacp8mRIwexsbEvpSgREREREXkxme5jn56eTmJiIrdv337q+tDQUNLS0l5WXSIiIiIi8gJeaPDs0aNHKVKkyDPXN27c+B8XJCIiIiIiLy7Twb5WrVoEBAT85TYODg7/uCAREREREXlxmQ729vb2lCpV6lXWIiIiIiIif9MLfUGViIiIiIhkTwr2IiIiIiImQMFeRERERMQEKNiLiIiIiJgABXsREREREROgYC8iIiIiYgIU7EVERERETICCvYiIiIiICcj0F1SZhBWdIXB3Vlch8hwrf/tnkkvWliGSGcWbQY91WV2FiIjwXwv2CvXyL3DbtntWlyCSefq7KiKSbfy3gv0Tk6KzugIRkX8/XVUSEclW1MdeRERERMQEKNiLiIiIiJgABXsREREREROgYC8iIiIiYgIU7EVERERETICCvYiIiIiICVCwFxERERExAQr2IiIiIiImQMFeRERERMQEKNiLiIiIiJgABXsREREREROgYC8iIiIiYgIU7EVERERETICCvYiIiIiICVCwFxERERExAQr2IiIiIiImQMFeRERERMQEKNiLiIiIiJgABXsREREREROgYC8iIiIiYgIU7EVERERETICCvYiIiIiICVCwFxERERExAQr2IiIiIiImwDIrDnr69Gm++OIL4/3JkydTpkyZDNsEBQXx/fffExQURLly5Xj33XdxdHT8X5cqIiIiIvKvkCUt9vny5eONN96gY8eOrF+/nocPH2ZYHxISQrVq1bh16xYNGjRg8+bNNG3alLS0tKwoV0REREQk28uSFvu8efPSqVMnEhMTn7p+9uzZeHp6snr1aszMzOjSpQt58uRh06ZNdOjQ4X9crYiIiIhI9pct+9gfOnSIli1bYmZmBoCLiwu1a9fm8OHDT90+JSWFhISEDDcRERERkf+SbBnsQ0ND8fDwyLDM09OT0NDQp24/depU7O3tjTc3N7f/RZkiIiIiItlGtgz2lpaWJCcnZ1iWlJSEpeXTew6NHz+e+Ph44y08PPx/UaaIiIiISLaRLYN90aJFuXHjRoZlN27coGjRok/d3srKCjs7uww3EREREZH/kmwZ7Nu3b8+GDRsICQkB4MSJE5w/f5527dplbWEiIiIiItlUlsyK8/DhQwYPHkx6ejoAEydOJHfu3Lz33nvUrl2bPn36sH37dipUqEC5cuU4efIkkyZNokKFCllRroiIiIhItpclwd7BwYE33ngDgO7duxuXFypUCAALCwvWr1/P+fPnuXfvHmXKlMHLyysrShURERER+VfIkmBvb29Pp06dnrtdxYoVqVix4v+gIhERERGRf7ds2cdeRERERERejIK9iIiIiIgJULAXERERETEBCvYiIiIiIiZAwV5ERERExAQo2IuIiIiImAAFexERERERE6BgLyIiIiJiAhTsRURERERMgIK9iIiIiIgJULAXERERETEBCvYiIiIiIiZAwV5ERERExAQo2IuIiIiImAAFexERERERE6BgLyIiIiJiAhTsRURERERMgIK9iIiIiIgJULAXERERETEBCvYiIiIiIiZAwV5ERERExAQo2IuIiIiImAAFexERERERE6BgLyIiIiJiAhTsRURERERMgIK9iIiIiIgJULAXERERETEBCvYiIiIiIiZAwV5ERERExAQo2IuIiIiImAAFexERERERE2CZ1QWI/G0rOkPg7qyuQkQmuWR1BSL/XcWbQY91WV2FZBNqsZd/L4V6ERH5r9N7ofyOWuzl329SdFZXICIi8r+nq2XyBwr2IiIiIpLtDdk7hMPBh1/qPsstKfeP91E3X13mNJnzEqr559QVR0RERESyvZcd6l+W7FSXWuxFRERE5F/jcu/LWV2C0cto8X+Z1GIvIiIiImICFOxFREREREyAgr2IiIiIiAlQsBcRERERMQEK9iIiIiIiJkDBXkRERETEBCjYi4iIiIiYAAV7ERERERETkC2/oCo9PZ3k5OQMyywsLLCyssqiikREREREsrds2WK/YcMG7O3tyZEjh/HWtWvXrC5LRERERCTbypYt9gAlSpTgypUrWV2GiIiIiMi/QrZssX/CYDBkdQkiIiIiIv8K2TLYW1hYEBwcjIuLC05OTjRt2pRLly49c/uUlBQSEhIy3ERERERE/kuyZbBv3749sbGxxMTEcOPGDQoVKkSzZs2IiIh46vZTp07F3t7eeHNzc/sfVywiIiIikrWybR/7J9zd3Zk3bx45cuTg0KFDtGvX7k/bjB8/njFjxhjvJyQkKNyLiIhI9rKiMwTufvn7neTy8vZVvBn0WPePdjFk7xAOBx/OcH9Okzn/tDLJhGwf7OG36S/T09MxN3/6BQYrKytNhSkiIiLZ26sI9S/bS6jx96H+affl1cmWwf6dd96hadOm1KpVi+joaD7++GPc3NyoV69eVpcmIiIi8s9Mis7qCp7uZbb8A5d7X6bcknIvdZ/y17JlH/uRI0eyZs0aKlSoQNOmTTEYDOzbt48cOXJkdWkiIiIiItlStmyxL1q0KKtWrcrqMkRERERE/jWyZYu9iIiIiIi8mGzZYv/STPEAK7M/L/99H7KXMPpbRERERCSrqcX+3zBCXURERETkOUy7xR7+euT5Sx79LSIiIiKSVdRiLyIiIiJiAky/xf6felXfEicvj668ZE8avyIiIv9Cf/zm3MzI7Hz9dfPVfaXfwqsW++dRqBf5e/S7IyIi/0Kv8ptyX/W38KrFPrOy67fEiWRHuooiIiL/cpd7X36p+/tffAuvWuxFREREREyAgr2IiIiIiAlQsBcRERERMQEK9iIiIiIiJkCDZ/8OTYEpkjkaRCvy1zQtrAjwYlNMPm8Q6queUjI7U4v936FQLyIiL4PeT0SAlzsN5KueUjI7U4v9P6EpMEVE5O/SFS2RP/mnU0z+L6aUzM7UYi8iIiIiYgIU7EVERERETMB/qyvOswa9/vFyqAYziYiIiMi/zH+rxT6zg5Q0mElERERE/mX+Wy32T/zVoFcNZhIRERGRf6H/Vou9iIiIiIiJUrAXERERETEBCvYiIiIiIiZAwV5ERERExAQo2IuIiIiImID/5qw4IiIiIvJKDdk7xPhzuSXlqJuvLnOazMnCil6tIXuHcDj48HO3K7ek3DPX/dNzpBZ7EREREXnp/hhyMxN6/81exvP7p/tQi72IiIiIvDKXe1/+y1bq7OBpre2/r/lFWtIv9778t2p4GedILfYiIiIi8p/2vJbyf8vVBrXYi4iIiIjw9Nb27H614ffUYv97Kzo//WcRERERkWxOwf73Anc//WcRERERkWxOwV5ERERExASYfh/7FZ3/3Po+yQWKN4Me67KmJhERERGRl8z0g/2zutSoq42IiIj8Vz2t4fOJSS4Z7/9HG0P/OAXmv+FLtkw/2D8xKfp3P7s8ezsRERERU/ciDZz/0cbQp01xmd2nvfzvBHsRERGRl+2vWr6fJbMNjP+LlvLfN3w+db0aQ59MgflvmPZSg2dFRERE/q5X2Zr9H20p/68ZsndIhg8N5ZaUY8jeIX9rX2qxFxEREfmnntfy/cL7U0v5f8XL7PKjYC8iIiIiv3lW16Lff9D4jw6mfdVeRpcfdcURERERkd9kpvuPughlW2qxFxEREZGMntW1KJt0EfrjVJR/9LRW71cxVeXT6nhy7KyYGlPBXkRERET+p54VzH8fyP8qGP+dPuhPe8w/nav+r+rIiqkxFexFREQy4+9Ma5gZL6sFVP2e5V8kM6E3M9s86Zf+PM/qt/6yBq7+sY6smhpTwV5ERCQzsnu/4uxen/z7rej8fz9PcnkpHyafFcz/18H4Zc1V/7QrAPC/65bz3wr2f2xteUkvShER+Q952dMavgzZpN+zmLg/fng0oQ+Tfwzkf5xH/vf3h+wd8sJdhDJzFeCPNQz/ZfhzH/NH2TbYp6en88svvxAUFES5cuWoUqXKP9/p016AJvSiFBEREXnlJkWb3IfJPwbvv7r/ol2EMnsV4I/7PXr/aKYe93vZcrrL5ORkmjRpwqBBg9i5cyctWrRgxIgRL+8Ak6KzZ4uLiIiIiGSZzPbZz641ZMsW+wULFnD16lX8/PzIkSMHv/76KxUqVKB79+5Uq1Ytq8sTEREREcl2smWw37JlCx06dCBHjhwAlC1blqpVq7J58+aXE+z/2Nd+RWf1sxcRkcx5VbPjvAzZrXuExrE93Yu8hv7J/2k2PP/Pm17yr9Y/a874vxqY+k+ns3zec/njfp937Cde1WDabBnsb9++TdOmTTMsK1y4MHfu3Hnq9ikpKaSmphrvx8fHA5CQYvi/jRIS4Ml9/10Zd+C/K+P6pz0mIQGmeGRcP945c09IRET+vSaE/vbvs95D5Nn8d5nOe+WT18Efs8ATmXmeT/bxv3oNPe38//H1/LR8Axkf97THPPl5vDNMCCU9Jen/r0ogPTk9w65+v+zgrYN/OtTBWwdJSEh47vqnrXuyvsyCMn9afqbnmece769qfd6yP+779/f/qt7M1vBkucFg+NPjn8XM8CJb/4+UKFGCIUOGZOhX37NnT9LT01m5cuWftp80aRKTJ0/+H1YoIiIiIvLqhYeHkzNnzkxtmy1b7D09PXnw4EGGZffv36dixYpP3X78+PGMGTPGeD89PZ3Hjx/j5OSEmZnZK61VRERERORlMxgMJCYmGrumZ0a2bLGfPHkya9as4fLly1hYWBAWFkbBggXZsGEDr732WlaXJyIiIiKS7WTLYB8ZGUmlSpUoXrw4jRo1YsWKFXh6erJ79261wIuIiIiIPEW2DPbwW3+ixYsXc+/ePcqWLUvv3r2xtrbO6rJERERERLKlbBvsRUREREQk87LlN8+KiIiIiMiLUbAXERERETEBJhXs27dvz+PHj/nggw9ISEjAYDCQnp5u/BcgLS3tqY9NS0vD39+fR48eGZcZDAauXr1KSEiIcdl3333HkiVLWL9+PQcOHPjLen5/3Cf3/3jMp9XxZLsn/z7Z7lm1//GYmfH77X5f41+JjIw0/pyYmEhSUlKGfTzZ5+9//uMxg4ODn3q8vzo3v19nMBgICwvj8ePHmar5aZ73fNPT03n77be5ceMGn3/+OcHBwc/cdurUqSxZsoRvvvmGc+fOGZdfu3aNmzdv0qZNG5KTk2nRogUrVqxg69at7NqV8YtJ/qqe8PBwbt++bbyfmppKQEAAERERxmXx8fGEhob+aT9/9XqJiooCfvtytydf6JZZCQkJREdHZ2rbxMRE/Pz8gD//P/4Vg8HA3bt3uXHjxp/WBQcHZ+p34Y8iIyNJTk5+6rqIiIgMr+9n2b17N19++SUnT55k+fLlL1zDEwkJCYSEhPzpefz+i/b+SlBQEDdu3CApKQl/f39iYmJ48OAB165dIzk5mYCAACIjIwkLC+PKlSukpKQQEBDAo0ePCA8PJyAggJSUFOPft6ioKPz9/UlOTiYwMJDg4GBiY2Px9/cnMTGRmzdvcvfuXeLi4vDz8yM+Pt74/5OYmIi/vz+xsbEEBwcTGBhIcnIy/v7+REVFERoaytWrV401hIeH8+jRI2MNV65cISwsjMjISGMN165d4/79+3963k973aSnpxMeHk5aWprxd+BZv1NP+/1JT0831vB7v3/tPnH9+nXu3bsHQHJyMo8fPyY8PPyFvjzmaQwGAzExMcb3nD9O9/xXr/clS5YwZ84ctm/fzvbt243Ln/Y7ev/+fa5du5Zh2ePHj437j4uLy/AaTEhIeObvTGpqKuHh4U9dFxUVRUxMTIZlYWFhBAUFPfN5+Pn5kZiY+Mz/u3v37tGjRw9CQ0OZMmVKhmP9/r3oaZ6ci5iYGNatW8dXX33FsWPHnvq9OPBif6sePXrEG2+8wfXr1xk7duxfbhsREWF87f0+H8TGxv7l+4C/v7/xi4yeCAkJ+dPzTklJMe47s6/JJ/9PTx77xFtvvcX169f57LPPePDgwVPru3r1Kh9++CFHjx5lwYIFXLp06bnHzcz7/+3btzP87c/Me058fPyfnsPTPHl9h4eHG98Hn+aP+e3AgQNMmTKFc+fO4evr+8xt/+r5P3z4MMP/Y0xMDP7+/hn+H/943L/DpPrYd+jQAXt7e3766ScGDx7MnTt3CAwM5KuvvuL7779n06ZNtGjRgokTJ1KzZk2SkpLYv38/wcHBfPrppzRs2JAdO3bw5ptvMm3aNBo0aECVKlXYtm0bTZo0wc3Nje+//56kpCQMBgNly5alQIECBAYGkpqayr179xg/fjxjx47Fw8MDLy8vLC0tKViwIC1btmTDhg0sW7aMhIQExo4dy8cff0zv3r2ZOXMmMTExdOnShfLlyxMSEsKxY8e4d+8e06dPZ8iQIWzdupXbt28zffp0XnvtNXLlysXVq1cpWbIkDRs2xN7eHoALFy5w69YtIiMjad26NR9//DGLFi3i4sWL1K1bl7Vr1/Lee+9hZmbG6NGj+frrr9mwYQMJCQm4uLjQqFEjwsLCCA8PJzU1ldmzZ2Nvb0/p0qU5efIknp6evPbaa0ydOpUffviBdevW0aZNG6KiovDz82PEiBHs2bMHg8HAuHHj+Oqrr1i6dCmHDh2iVatWXLlyhfz583PixAnS09P55ZdfiIqKYsOGDfj6+nLhwgW+/PJLUlJSMDMzIyUlheDgYGrVqkWfPn0YPHgwtra2WFlZYWFhwdSpU5k4cSI9evQgV65crFq1ikKFCuHo6EipUqX45ptvCA0NxdfXl2rVqvHJJ59w+PBhrly5QpkyZWjTpg12dnYA3LlzhxYtWrBgwQLOnDmDh4cH169fp1WrVlhYWBAUFISvry8TJ07E3d2djz76CHd3d2xtbXn8+DFFixalZcuWrFmzhurVqxMTE8Ply5dp0aIFixcvxsnJCUdHRywtLSlZsiQGgwFvb29SU1M5c+YMbdu2xd/fn9OnT5OcnIyFhQWpqakkJCQwffp0mjdvTrly5ejSpQs7d+7k448/Zs+ePVy8eBEfHx8OHz7MBx98wJAhQ2jYsCHJyckkJSXh6enJ3bt3sbCw4JtvvmHUqFGcO3eO8+fPU6FCBZo0aYKLiwuJiYk4Ojpy+fJlhg0bRoECBRg4cCATJ07k4sWLBAcHc/fuXa5cuYK9vT3p6em4urqSK1cu7t69y+HDh/niiy+Ijo6mYsWK7N69GysrK/bv30/79u2Jjo7m5MmT1KhRg7i4OJKSkli/fj1XrlxhwYIFbN++naJFi+Lt7c2ePXsoWbIkqamppKam8vjxY/r06cPOnTsJCQkhPT2d9u3bc+nSJUaOHMnjx49ZsGABO3bs4Ouvv2bWrFl4enoCUL58eQ4ePEhERATx8fFUqlSJHDlycOnSJXx8fDh69CgGg4HKlSsTERFBvXr18PPzw8vLi6+++ooPP/yQ6Ohobt++TWRkJHfu3MHc3Jy0tDTq1q3Lr7/+Srdu3RgzZgxFihShZs2aBAcHExISQvHixTlz5gzW1tZYW1tTsGBBHj16hIODA+XKlWPnzp3kyZMHc3NzgoODsba2xsXFhbfeegtHR0dmzpxJQkIC1tbWWFpaUrFiRSIiIkhLS8PDw4OdO3cyYsQIFixYQP78+Slfvjxr1qxh2LBhrFixAicnJxo3bszChQsZPHgwmzdvJj09nY4dO/L999/Tq1cvjh07Rnh4OH369GHWrFl07NiRwMBAbty4wdtvv83XX39No0aNePz4MWfPnmXYsGF8++23VKxYERcXF/bu3cuIESOYN28eXl5elChRgg0bNjB8+HCWLFlCzpw5qVOnDj/++CNDhgzh559/BqBt27bMnTuXfv36ceDAAWJiYujZsyczZswgf/78xlBZo0YNjh49yrBhwwgODqZ+/fp069aNkydPcvv2bT755BPMzc2JioqiSJEiHDt2jEOHDjF37lxiY2P58ssv8fPzY9q0ady6dYvBgwezcuVKRo8ezYABAyhRogTNmzfn2rVrtG3bluTkZJYvX05iYiIxMTH4+PgQERFBcnIyRYoUYf/+/Xh7e5OQkMC1a9coVaoUd+7cYeHChURERLBu3Tpy585NcnIylpaWFC5cmMKFC1OxYkWOHj3KunXrOHbsGAMHDqREiRK89dZbeHl54eXlRVJSEnFxcXh6elKgQAFWr17N8uXLWb9+PfXr16dKlSrUqVOHjz76iDVr1uDk5ERgYKAx0Hh5edGmTRt27drFhQsXiI2NpUSJEkydOpUtW7YQEhJCnjx5OH78OE2aNGHXrl2EhYVRsmRJ4uPjsbCwIC4ujjFjxuDv78+PP/7I/Pnz+f777ylQoADNmjVj9+7dPHr0iDt37uDo6EhiYiL29vY4OjpiZWVFhw4dWLp0Kffv32f27NncvXuXbdu24eLiwrlz5/juu+9wdXVl2rRptG7dmqtXrxIeHs7evXvx8fEx/k1PTk4mPDyct956i27dutGsWTN+/fVXXnvtNbZu3Urfvn3ZvXs3YWFhWFtbU61aNcaPH8/ly5cBWL16NQ8ePCBnzpxERERQoEABTp8+jbW1tbEBLU+ePHTo0IHTp09z9epV0tLSKFSoEBERETx8+JBPP/2UNWvWEBYWRq1atUhPT8fJyYmwsDAOHDiAg4MDAKGhodjb2xMREUHx4sWxtLQkNDSUmjVrcvjwYcqVK8f58+fJlSsX9evX59ChQyQnJxMbG0uVKlUwNzfn+vXr2NnZERUVhZeXF+3atSMtLY2lS5dy584dWrVqhbW1Ne+++y4XL15k6dKlXLp0iUaNGpGens65c+fw8fHh7Nmz9O3bl0uXLlGpUiU6d+6Mr68vNjY25MuXDzMzM9auXcu9e/do0qQJ/v7+XLlyhbJly3L16lUWLlxIYmIi77zzDunp6cTGxtKjRw/S0tI4ceIEZcuWpWTJkjg6OvLll19iaWlJamoq8fHx1K5dG29vb44cOUKjRo3w8vICICkpiePHj9OjRw/i4uLYvXs3np6ebN68mVKlSnHy5EmqVKlCz5492bRpE/7+/tjb22NlZYWHhweXLl3C3d2d4OBg9uzZg5eXF9999x0BAQFs3bqVfPny8fDhQ6Kionj77bc5c+YMrVq1YvXq1QQFBdGkSRPy58/P0qVLyZkzJ9euXeP1118nPT2dLVu2YGNjY5wvvnjx4oSEhODq6oq1tTUhISHUqlWL06dPEx0dTfHixYmIiDD+DerUqRO1a9emQ4cOHD58mJUrV2IwGPD09MTd3Z0dO3bQsmVLYmNjGTp0KAkJCezZs4fbt2/Tt29fRo8ezezZs6lQoQLx8fEEBQVRokQJGjZsSIUKFYiLiwOgR48emZ4V0qRa7Dds2MB3331HmzZtGDZsGEOHDgVg7969XLhwgYEDB2JnZ8fQoUOZO3cudevWNf4n2Nra8s0337B48WLOnj3Lxx9/zPXr12nWrBnW1tZs2LCBmTNn4uLiQsGCBenVqxeRkZHMmTOHPXv2EBkZSbt27QgPD+fChQs8fPiQJUuWALBmzRpy585NgwYNmDhxIn379iUwMJCPPvqI06dP8/bbbzNs2DCSkpJYuXIlZmZmvPnmm+zevZuIiAiuXLnCr7/+yuPHj+nUqRNTp05lx44d/PDDD4SGhnLmzBmWLVuGpaUlb7/9Nl5eXvj7+5Oens6oUaM4ceIE69atw9nZmaNHj2JmZkbdunW5ceMGN2/exNnZmQYNGmBjY8OgQYNYtmwZy5cvJz09HTMzMwYOHMimTZv4+OOPsba25tChQ6SkpHDjxg2uXr3Ktm3bKFKkCDt37mTGjBncv3+f5cuXM3v2bCIiIrh16xbdu3fn5MmTtGzZEj8/P4YNG0bp0qXZuHEjnp6eNGnShI8//pjdu3fj6OjIsWPHcHBwIDo6mq+++go/Pz86duxIvnz5OHbsGJs3byYsLIz+/fsbfwl/+OEHbt++jZ+fH3v37mX58uW0aNGC0qVLM2/ePOrUqcOqVato1qwZERERVKxYkZ07d3L79m1q1KjB0qVLsbCwIFeuXDx8+JDXX38dMzMz2rZtS7NmzUhISOD48ePs2LGDFStWMG3aNOLj49mzZw9du3bFysqKAQMGkJCQQKlSpYiIiCAuLo579+5hZ2fHBx98wMCBA3n77bd55513GDVqFJ07d6Zv3758+umnfPHFF+zevZv69etTo0YNEhMT6dGjB+XLl2fJkiW0bduWIkWK8P3332NlZcWsWbM4ffo0Li4uDBo0iAIFCvDDDz9w//59jh8/zocffkiZMmU4ceIE4eHhXLp0iT179nD27FkA9uzZA0DevHnJnTs3np6e7N+/n0mTJmFhYYGnpyfh4eHMnj2b5ORkGjVqhIWFBe+99x5dunShYMGCFC1alOrVqxMeHs7kyZNZu3Yt58+fZ82aNfz6669ER0fj7OxMaGgoLVu2JC4ujrfffptDhw6xb98+Dhw4QJs2bbC3t+fx48dER0eTL18+UlJSqF27Nt999x1Xr14lOTmZQoUKceHCBSpUqEBwcDBXr17lypUrHDt2jDlz5nDp0iUANm3aRI4cOfj+++/5/vvvCQ8Pp0uXLhgMBnLlyoWfnx+FChWiXr16ODo60q9fP5o2bcrEiRMxMzNj586dxMbGcuHCBVq1aoWNjY3xtXnlyhU2btxo/JsREhLCF198Qe3atVm+fDnJycl8+eWXVK9enYULF7JhwwYMBgNr167lo48+Infu3CQkJLBs2TLmzp2Ls7Mzzs7ONG3aFCcnJ3bu3Em5cuWIiYmhRo0apKam8u233+Lr68vjx4/Zvn07169fZ+bMmfTo0YPExEQmTpyIra0tFSpUoFevXqSlpfHBBx+QM2dOvLy8GDhwIAaDgSFDhpA/f37c3d0ZOXIkaWlpDBgwAG9vbxwcHBg7diypqan06NGD6tWrY2FhweTJk0lMTKRjx440bdqU1NRUJk2aRFpaGk2bNqVz584kJCQwceJELC0tqVq1Kr169SI5OZkPP/wQZ2dnSpQowcCBA0lPT+e9994jT5485MmTh2HDhmEwGHjrrbcoXrw4zs7OjBkzBjMzMxYsWECfPn2wtbVl69atNG/enIIFC1KqVCljA0e9evUYP348VatW5dq1ayxevJiTJ08ybdo0/P39GTBgAH5+fowaNYrRo0czaNAg4uPjGTJkCA4ODmzfvp3g4GDCw8OJiYnh8ePHfP7553z77bcEBQURHByMnZ0dBw8eZPHixfj5+VG+fHn69+/PtWvXuHbtGp07d+bWrVuYmZnx7rvvUrVqVY4cOULTpk1p1aoVHh4efPnllyxfvpy6devy7bffcufOHQCqV6/O119/TZMmTYiIiMDKyopHjx5RtGhRZs2axdmzZ41Xhxs1akRAQADdu3enQoUKXLt2je7duxMQEICfnx+enp6MHDmShw8fMmLECGPL+O7duwkJCSFHjhz4+fkxfvx41q9fD8D3339vDKaDBg3i4cOHdO7c2fgavXPnDunp6UybNo2bN28SEhLCnTt3OHfuHGfPnmXOnDmkpqbSrl0742OfvGZbt25NVFQUR48exd/fn+PHj1OsWDEqV67M119/zYgRI7CzsyMsLMzYimllZcXDhw8pX748gwcPJioqytjINW7cOM6dO4eHhwe9evUyfvlkv379yJs3Lx06dODu3bt89NFHXLhwgaJFi3L69Gk+/fRTYyPN4cOHyZ8/P5UqVaJw4cK0b9+ePHnyMHDgQCIiIvDw8GDJkiXExMTg4OCAi4sLs2bNYuDAgdy4cQNXV1eqVq1KtWrVyJ07N71796Zx48bGc9i8eXPy5cvH2LFj6dOnD3369GHLli20bduWMmXK0K1bN9zd3VmyZAkPHz4kMTGRI0eOsH//fmbPns29e/fo3Lkz9evX59tvv2Xt2rVs3boVc3NzcuTIQY4cOTh48CBz585ly5Yt9OvXj5SUFMaNG0eePHn4+OOP2bZtGxUrVqRgwYKULl2aU6dOsXbtWg4cOEBCQgI///wzmzZt4uzZs8ycOZMFCxbQsWNHIiIiOHDgAG+99RbLli1j+PDhjB8/nuvXr2Nubk7nzp2ZNm0av/zyC66urmzfvh0/Pz8KFCjAlClTCAgIwNPTkzfeeIMJEyawc+dOqlevTkBAANHR0VSqVIkWLVpw4cIFNm7cSGpqqvFD5Jtvvkn37t2JiYnB0tKSbdu2sXnzZpYsWcLx48c5cuQIVlZWlCxZkqSkJAYMGMDjx4+ZNGkSGzZsoHfv3oSFhdGzZ0/y5s2LmZkZ165do3DhwgQGBjJ+/HjOnDlDcHAwTZs2JSwsDHd3dxo3bsyqVaswNzdn6tSpTJ8+HXNzc+7fv09cXBxDhw7liy++IDw8nPXr1xMfH4+DgwOxsbGEhYUxcOBAIiMjKVu2LLly5SIkJITPP/8cJycn8uXLR1hYGL6+vmzbto333nsPX19foqKiqF27NosXL2bHjh3Y2NgwceJE4uLiGDx4MNevX+fXX39l+/btfPLJJ9y8eRNPT09jI1VmZctvnv0ncuTIwfnz53F0dKRMmTIsWLCA6OhoSpQogbOzs/Gyk6OjI3nz5gV+uzTj4uLC119/zU8//URcXBy2tra4urryzjvvYG1tTaVKlTh58iStW7dm27ZtxsvXDx48wM/Pj6ioKM6ePcuXX35pbJHasGEDKSkp5MyZk1u3bmFtbc2dO3eIjo7m7t27FChQgJo1a5KcnEz58uVxdXVl6tSpBAUF4eTkxOzZs6levTqTJk2iWrVqhISEGIP9rVu3+Pnnn4mNjSUwMJC3334bf39/5s+fT1paGiVKlKBAgQIcOHCAvXv34urqSqtWrWjWrBl79+41tg4cPHgQNzc35s2bR0BAAEWLFsXCwsLYgvPo0SOWLFmCj48PBw8eJDU11XhpNiwsjNDQUAYMGEDu3LkpUaIEhQsXpnr16jRr1oxcuXLh6+tL6dKlOXfuHAUKFODkyZM0btyYH374gc6dOxMXF2f84/Hw4UOuXLnC7du3yZcvH8nJyXTs2JH09HTu3LlD69at2bFjB1OnTjV2KXB3d6du3bo0atSIy5cvM336dMqXL4+fnx+lSpXC2dnZ2FLr6OiIt7c3bm5uABw7dowqVapQtWpVAgMDsbe3Z9GiRbRs2ZKlS5dy+PBh3nnnHeMlZ39/f+7fv4+joyNVq1alTp06VKhQgQ0bNnDhwgVj6765uTmrVq3C0dGRAQMG8ODBA1xcXPjll184f/48aWlpODo6kpqaypgxY1i+fDmenp60bduWQ4cO4eTkRMOGDdm7dy+pqanExcVRs2ZNrKysWLJkCVOmTOHBgwe0bt2avXv3Ehsby4EDB4iIiMDc3JyLFy+SmprKw4cPuXjxIkWKFMHFxYX79+/j7OxM1apVOXv2LNevXwd+uxy4YcMGOnToQFJSEnfu3CElJYWlS5eSnJxMiRIljL8Xv/76Kzdu3GDo0KG4u7tz/fp1atWqhbW1tTH4N2/enCJFihi7q9ja2jJu3Dhy585Nrly5mD9/PtevX+fSpUuYm5uTmprKuXPnsLKyIjY2lk6dOrFv3z62bdtm/A4Lg8FA//79qV69Onv27MHe3p74+HgSEhKwtbVl8uTJtG/fnmHDhvHo0SNjq8i9e/coWbIkAQEBxlB35MgRChcuzPHjx0lKSuLAgQOkpqZSqlQpgoKCcHFxwWAw0KRJE+zs7HB1dWXRokXG/7MSJUrg5ubGrVu3cHNzIzo6muDgYGJiYkhKSmLhwoUEBgaycOFC0tLSSEtLIy4ujuPHj3P+/HmSk5Pp2bMnkydPJioqiqpVqxITE0N0dDQffvghZ86c4ezZs/j7+xMWFsbRo0fJly8flpaWJCQk4Obmxttvv42DgwOWlpZ07NiR2NhY9u3bZwx0PXv2JCQkhLCwMMaOHUt6errxjSMxMZGBAwcCMGrUKIKDg3n06BFvvvmmMcw/CbodO3bEysqKmTNnYmFhYWztNxgMLFmyBDc3NywtLenUqROJiYls27YNf39/zM3NeeONN3j06BEnTpzgww8/BKB///7cunWL5ORkhg4disFg4L333uP27dtERUUZpzVevXo1Z86cwcbGhhMnTnD69GkArKysSExM5JdffiFnzpzcu3cPZ2dn0tPTKV++PC4uLlSoUAFHR0cuXbpEVFQUt27dIikpicKFC2Nubs6MGTOIi4vj/v37zJ8/n9TUVC5cuICDgwNvvvkmNWvWZMqUKfj5+RETE4ONjQ27d+/GwsKCc+fOceLECXLmzGlsZc6ZMyeVK1dm/fr1nD9/nkKFCnH06FFWr15NsWLFeP3119m4cSMACxYs4L333uOXX35hzpw5FCtWjKZNm/Lrr78SExNDcnIyd+7cYcCAAeTPn58JEyZQu3Ztdu/ejZOTE3ny5KFXr16sW7eOa9eu4eTkxODBg42/m1FRUbi5uVG7dm02b97Mvn37sLe3JzQ0lIYNG+Lr68vWrVtxdXVl1apVxpb++fPnY25uTlhYGIGBgQQFBXH37l0AmjRpwurVq3nzzTcpXbo0R44c4dGjR7z11lukp6dz69YtoqKiiI+PZ8mSJTRs2JDIyEgcHBywt7enUKFCFClShPT0dCIjI2nYsCERERGsXr2aYcOGsWfPHnx8fHBwcCA8PJxDhw4xc+ZMkpOTSU9Pp0+fPjRs2JCdO3eSmJjInj17yJ8/P4cOHSIsLIyUlBQcHByoU6cOrq6upKamsnXrVqytrXn//fcpV64c/v7+2NraEhsby6NHj6hevTr3798nNTXV2BKfkJDA3r17qVWrFmfOnMHBwYFu3brh4+ODra0t7du3Z/78+XTt2hUvLy9OnDhBixYtWLJkCU5OTly6dIm4uDhatGjBhg0bWLlyJR06dODAgQM0btyYTZs2kStXLpYtW4bBYMDV1ZW3334bGxsbRowYgbW1NTt27MBgMHDq1Cl++eUX9u/fz9ChQ7G0tGTIkCF06NCB/v37U6tWLTZt2oSlpSWbNm0iKioKX19fSpYsyeXLl42NU5aWlgQGBuLl5UVkZCSXLl2ibt262Nra0qtXL6pUqcK5c+ewtrZm3759rF+/njx58pA3b17WrVvH0aNHsbW1xcLCAkdHRw4cOEBaWhr58+cnb968nD59mqCgID799FOio6O5d+8e69at4+rVq1y7do0LFy5Qp04dZs2ahcFgoEOHDhQtWpS0tDSOHz9OfHw8u3fvZv/+/dSoUYNRo0ZRtmxZBg4cSFJSEh4eHkRERFC0aFHy5Mlj7Jlw9epVAEqUKMHZs2cpU6YMd+/e5ebNmxQuXJiYmBimTJmCo6MjCQkJREVFUadOHZYuXcqoUaMYN24c9+7dw9fX13hl7u7duxgMBqKjo3FwcODs2bOEhIRgbW3N6dOnjV1sypQpY/xgUrBgQQYNGsSkSZPIkSMHnTp1Iioqil9++YWrV69ia2tLTEwMKSkplC5d2vjBYeHChXTr1o2YmBg6d+6Mvb093333HU5OTgwbNoxGjRrh6OhI69atKVmy5AvnYJPqivPEoEGDiI6OJmfOnNjY2LB3714KFChAZGSk8Y2gRYsWbNmyhYSEBEqXLo2ZmRnt2rXj9u3beHt7U7JkSWJjYzl06BAWFhasXbuWRo0asXHjRho3boyPjw8eHh5s2LABe3t78uXLx1tvvUXRokVJSkrixx9/xNzcnJMnT7Jz507jf+rIkSNxcnJi9+7ddOjQgYEDB9K+fXtGjRrF999/z61bt6hUqVKGPldFihTh3r17VK5cmWLFivHTTz8ZL1MHBATwww8/MHv2bEaMGMH333/P6NGj2bt3L5UrVyYgIIBjx44xd+5c8ubNi4ODAzly5MDFxQUfHx+6dOnC999/T+7cuQkICCA2NpaYmBgKFChAw4YNefz4MUuWLOHQoUMULFiQ/v3706ZNG06fPs2dO3ewtramZs2a+Pn58cEHH+Dl5cXNmzcpWbIkI0eOZMWKFSQnJ5MzZ07Mzc3JkycP48aNw8rKip9++olRo0bh7e2NwWBg+PDhxm4HxYsXp2jRouzbt488efLw6aefMmbMGH7++WcePHhAvnz5iIyMJDo6mvLly9OvXz/c3Nz46quvCA0NpXHjxpQpUwYnJyfGjRtHYmIirq6uVKtWDTc3NzZu3Mgnn3zCqVOneOONN5gzZw6rVq2iVq1a5MiRg9u3b/PRRx/Rp08f8ubNS9u2bQkODubevXt4enpy5swZY3/mIUOG8NNPP2Fvb0+LFi0oWrQooaGhuLi44ODgQK1atbh58yaFChVi8ODBLF68GDMzM0qUKMHFixeZNGkSU6ZMIS4ujpSUFH766Sfi4+Pp0KED165d4+TJk4wePZqoqCh+/vln3NzcOHfuHBUrVqRAgQKsXbsWd3d3GjRowIMHD7hx4wY+Pj5cvnwZg8FAly5d2LRpE5s3b+brr7/G3d2dtWvXYmZmRvHixfnmm2+Ii4tjxYoVvP/++xQpUsR4+fXo0aN88803+Pr60rJlS7755huGDh3KxYsXjSH9yeVZNzc3zp8/z+HDh6lVqxYADRs2JCwsjPfff9/YxeHJpdAnrW+BgYHcv3+fwMBAypYtS82aNVmzZg2//PILd+/epVChQsbLk8eOHcPMzIy+ffuyevVq8ufPj6WlJWPGjGHu3Lmkp6dja2tLhw4duHjxItOmTWPp0qUcO3aMRYsWsW7dOnLmzImHhwfbt28nLi4Oa2trIiIicHNzo1KlSmzYsIFatWqRlpZGYmIiaWlpWFtbk56ejqenJxYWFowaNYrbt2/TtGlTrl+/Ts2aNRk6dCjz5s2jatWq2NracubMGczNzSlXrhx+fn6sXbuWXr168eabb1K1alVatWpF/fr1uX37Nra2tri7uxMYGMjDhw8xGAxMnTqVnj17YmFhQWhoKDly5MDR0ZGAgABy585NtWrVePz4MdbW1tja2hrfkJKTkzEzM8Pe3p6oqCjs7e1JS0sjNTUVJycnY+iC3y6ROzk5ERERgYODg7EVLUeOHISHh2dY5uLiQkREBPb29tjY2BATE4OjoyOxsbHY2tpiY2NDdHS0sXuGubk5Dg4OxuOlpKSQnp6Oo6OjcZnBYCA5ORlnZ2fj8dzd3bGyssLd3d3Yr//hw4dMmDCBhQsX0qtXL06ePImbmxs//vgjVatWJSAggObNm3P8+HE++ugjvvvuO06cOEHfvn3Jly8fzs7O2Nrasnr1ary9vbGzs+PQoUNERUVx+/ZtY9CrUKECFStWxNvbm+PHj7N7926qVq2Ki4sLZ86cIX/+/Ny/f58yZcoQHx/P4cOHGT58OFZWVqxZswaDwcDs2bP56aefqFmzpvFqj62tLdWqVaNYsWKEhYWxe/duALp27Wq8qnns2DFsbGzw8PAgPT0dCwsLbG1tCQsL4/z588YrrR06dODChQv4+flRuHBhTp8+zcOHD/H09KRUqVLs2bMHV1dX8ubNi8FgYMWKFdjZ2eHk5ES1atXw8fHh3r17FC9enAIFCnDq1CmuXbtm/NA7YcIEbt68yaFDhxg1ahS9evXizp07tGnThurVq5M/f36+/fZbrl69ypEjR/D19eXy5cvkzJmTmJgYgoODcXFxoWLFivzyyy+kp6fzxhtvcOLECQICAhg2bBjw29iAsLAwKlasSJ06dZg3bx41a9Zk+/btPHr0iA8++IAHDx5Qo0YN4LfuLk+ugE2ZMoXg4GC++OILfvjhB+MHve3bt3PgwAH69OnDzJkzsbS0ZNq0aRw8eJCYmBhsbW157bXXOHLkCIsWLaJkyZJcuXKF0qVLExgYyPvvv4+trS0uLi7cuXOH0qVL4+bmZhwrlZKSQlhYGKNGjcLf3x93d3dq167NrVu32Lx5M0eOHGHKlCncvn2bXbt20a9fPywtLfH29iY+Pp7Y2FgsLS3x8fFh/fr1BAcHU69ePU6fPk316tUzdOepWLEiNjY2HDt2jP3799O/f388PT05dOgQEydOpH79+jx48IC+ffuSnp7O+fPnjWMbBg8ezLp16+jatSt58+ZlypQpxiu0uXPnxsvLi7CwMMqVK4eXlxe2trbGczh8+HCsra2ZP38+Dx48oE2bNvj5+bFt2zZu3rzJhg0bKF68OCkpKYwaNYrKlStjYWFB7ty5cXNz49GjR8auxOXLl+fw4cPkzp2bI0eOUKBAAZydnfHx8eGDDz5g69atHDhwgOLFi1OpUiXCw8O5cuUKxYoV4+jRowQFBREVFUXx4sWZNm0aa9asoWbNmnh7ezN+/HhWrVpFyZIlyZEjB6+99hoFChTg+++/5969e4wcOZJ69erx+PFjFi9eTGRkJCdPnqRu3bo0aNCAxMRELCwsjB9Wk5OTKVasGFeuXKFUqVLcuHEDa2tr8uTJw7Fjx/D09OT+/fscPHiQ+Ph4ChQoQEhICObm5jRq1IidO3eSN29eLly4QKFChWjdujURERGEhIRQunRp3N3dCQkJYcSIETg7O3Py5EmWLVuGj48P8NvYg5w5c3L+/HkKFy5MREREhvVLly7F3DxznWxMMtifOXOG1NRU0tPTGT16NPHx8djb2+Pk5MSDBw+M32Lr5OTEhAkTMDc3Z9SoUTRt2tTYx6xAgQI0b96cTp06sWHDBqpXr86KFSto3ry58Q12/PjxXLlyBV9fX65fv27s116+fHlWrVrF9u3bqVOnDvnz56d06dJMmDCBDRs2MHv2bB4+fEi9evWM3QyaNWvGxo0byZkzJ+np6bRr144zZ87QokULdu3aRWhoKOPHj6d27dqUKVOGXLlyGVt+z507R6FChXj06BHbtm2jfv369O3bl9TUVHx8fHj06BFLly6lQYMGxsFvaWlpWFhYUKNGDQ4ePEjx4sVJSkrCwsKCEiVKsGnTJr766ivq1q3LiBEjiIqK4s6dO8Z+4g4ODnz99dd4e3tz6NAhVqxYQWhoKDNnzqRDhw5UqlQJX19fevbsaWyZTU5O5ubNmwwcOJDx48dTs2ZNfvzxR+rVq2ccdJQvXz4GDhxI69atCQ0NpWrVqhw8eBB3d3esra25e/cuVatWxd3dnTt37uDi4sLDhw+ZNGmSsaWqSJEixk/JZcqU4fLlyzg4OBAUFGS8dLh06VJq165Neno6n332GU2bNmXHjh2UKlWKKlWq8ODBA+7evUtsbCzTpk2jWbNmVKhQgRIlSpCcnExMTAx2dnacPn2aESNG4ODgwO7du3njjTeYNWsWiYmJrFq1iiFDhpCYmEju3LmJjo4mLi7O2EWnR48e/Pzzz1StWpVffvmFxo0bY2ZmRtGiRXFxcSEtLY1PPvmETz/9lJ9//hkfHx8SEhJ47bXX+Oqrr+jTp4+xhbRgwYIULFgQGxsbtm3bRv78+Zk/fz4TJ06kS5cuvPvuu4wYMYLU1FSOHj1K/vz5jQMn8+fPT0hICK+99hqJiYns2rWLli1b8ujRI86ePUvOnDmNV728vLxYv349PXv25OrVq7Rv397YDWzu3LlMmjSJJk2a8N577xl/B578Qf/pp59wd3cH4IMPPsDX15cHDx4wbtw4pk+fzrp16xg0aBAFCxbkwYMH2Nra4uDgwOuvv06TJk24cuUKx48f5/Dhw0RGRlKqVCnu3r3Lm2++yaNHjzh58iQeHh4cP36c119/nbt373Lx4kXmzZvH5s2bOXXqFAaDgeLFizN37lymTp3KsWPHSElJwdLS0nh1b/Hixfj4+NC9e3fq169PQkIC3bt3Z+/evYSHhzNkyBA2bdpEXFwcM2bM4OLFizRq1IjZs2eTP39+Y3/mKVOmGFsQV61axfLly1m2bBnw20BeR0dH3n33XerWrcu4ceNo3bo1o0ePpkuXLly+fJklS5bQqlUrypcvT1xcHCEhIXTv3p3Dhw8TFhaGj4+PSfax/+abb+jcuTPXrl3D39+fQoUKERcXx6BBg9iwYQOlSpXC29ubn3/+GRsbG+7cuUPBggWJi4vj0aNHdO7cmXPnznHx4kXjYxo2bEipUqWYPn06b775Jnfv3uX06dNUqVKFzZs3Y2FhQVpaGnny5KFAgQIAVKtWzXilLikpidDQUJo0acLVq1c5ceIElStXJjg42NiFpkCBAsYuHw8fPqRSpUoEBQWxYsUKxo4dS+fOnY3jtUaOHMm2bdtYvHgx9vb2JCQkEBwcTEJCAm3btuX06dPY2dnRvn17mjVrxvXr19mxYwd37txhxIgRrFq1ivLly/PBBx/wxRdfkJqaSlpaGg8fPmTevHl0796dIkWK4O3tzdSpU6lcuTLx8fFcu3YNCwsLEhMTGT16NJcuXeKnn36iWLFiVK1alYEDB9KvXz+aNWtmHEtgZmbGrl27aNCgAeXKleODDz7g888/5+HDhxw8eJDy5cuTO3dupkyZwvDhwzl79ixDhgzh3r17bNmyhaioKGxsbChVqhRbt27F2dkZDw8PAgMDjV1M3nzzTVq1asVHH31Ely5dmDFjBidPnjS+pz948IC33nqLs2fPGsf0LFu2jO+++45cuXLRq1cvgoKC8PDwIDo62tjAkD9/fho3bsyxY8cICQlhwoQJDBs2jIIFC7Jv3z7c3d3JnTs3oaGhJCcn4+LiAoCrqyt58uRh69atLFu2jMGDB7N//34cHByIiori7t27ODk54e7uzpo1a9i7dy9lypQxZgU7OzuuXbtG6dKlOXToEF5eXjg7O3P+/HkaNmzIwYMHsba2xs3NjcjISDw8PMiZMye9evVi1qxZhIaGkpaWRu7cuXn99dfJmTMnBoOBQYMGsXLlSkaOHEmRIkUIDAxk6NCh+Pn5YWVlxeeff87hw4fp1asXPXr0oGDBgkybNo3XX3+dQoUKsXHjxgxXbYcPH27s/lKrVi2mT59O7dq1sbGxITIykp07d7JixQrjoPeDBw8ax3/lzp2bkJAQIiMjadKkCdevX6dGjRo0bdoUCwsL41WEJ11TT5w4YfyA4+bmhr+/P3fv3qVChQrGHgTp6enY2dnx2muvkSNHDq5du0Z8fDxhYWFs2bKFkSNHUqFCBUaNGoWXlxfjxo2jb9++7N+/n7CwME6ePEl4eDilS5dm8eLF9OzZkyJFihi7hZYsWZImTZpw4sQJrl69yvr164mNjeW1117Dz88PFxcXfvrpJz766CPmz5/P5cuXqVixIqNHj2b79u20b9+eK1eucPnyZUaMGMHnn3+Oubk5efPmJSAggNDQUIYMGcK8efOoUqWKcZygra0t77zzDrNnz+bOnTtUrVqVhIQEnJ2d8fX15dtvv2Xfvn1cvHiRKVOm8M4772BjY0NQUJDxPfNFmFQf+0GDBtGzZ0/ef/99OnXqRM+ePXnw4AG//vorcXFxxkEIt2/fZtKkSbi6ujJ37lzmzZvHhAkTCA0NpX///vTo0YNPPvmEtLQ0SpUqRYsWLdizZw+TJk3i6NGj1KlTB4C+ffvSrl07Nm3axPLly6lTpw558+alevXqjBkzhvfee4/ExEQsLS1ZvHgx8+bNY/r06bz77ruEhYWxYMECli5dyqJFizh//jyDBg3C19fX2Fp/8+ZN8uTJw6FDh0hNTWXAgAH07duX8PBwPv74Y4YMGWIcNHP48GF27drF5s2bgd9aNwICAkhNTTVe4q1SpQr+/v6MGjWKFStWEB8fz4ULF/j4448pU6YMzs7O/Prrr4wcOZKxY8fy448/MnfuXM6fP8+iRYuoV68eaWlpdO3alZs3bzJ37lzq1avHuHHjyJkzJ3fv3iUgIIC4uDhOnTrF4cOH2bdvH7lz5za2VD95XosWLcLW1pbBgwczb948ChcujIeHB4sXL2bYsGHGDxLnz5/HwcGBkSNHUrZsWVxcXKhbty5ly5alc+fOuLm5kZqayueff06rVq1ISkpi6NChuLi4ULZsWdavX0+uXLkoUaKE8erExIkTOXv2LHv37mXLli3kzJnTeAn7SZefBw8eMH36dCIjI3n//ffp0KEDt27dIjEx0dgftlChQiQlJXH16lV27dpF4cKFKVKkCDdu3MDc3Jw5c+YQGhpK69atSUhIMPapnzVrFhs2bGDGjBncvXuXfv360bZtW0aPHs2oUaPo1q0btra2HD9+nNu3b/Pzzz9TpEgRevToQWpqKl9//TXFihVjxowZ9O/fn/j4eHLmzEm/fv2Mb045cuRg8ODBxMfH4+rqyrfffsupU6e4dOkS7777Lrlz56ZJkyZYW1uzadMm2rVrR/PmzTl58iSLFy/Gzs4Oc3NzzMzMOH78OL1796ZmzZp07dqVixcvcuHCBWxtbfn888957733uHnzJlu3bqVw4cLMnTuXzz//nAEDBtC9e3du375N4cKFqVKlCqVLl6Zx48bs27eP6Ohovv32W77++msuX75MQkICjx8/xtfXl/r16xsHlrdo0YJPPvmEDRs2cP/+fUqWLElKSgqffvopBoOBcuXK8csvv1C3bl0ePXpEwYIFuXr1KoMHDyZv3rxcvXqVQoUKUadOHeLj4wkPD2fz5s1cvHiRGjVqGD9MfP3115w7d44aNWoY+/UmJCTQv39/lixZYhzI++2337Jo0SLu3bvH6dOn+fnnn5k7dy5HjhwhMDCQ6dOn8+2332JnZ4eVlRUBAQEcOnSIK1eu8NZbbxEXF8eCBQsIDw/n4sWLDB48mPDwcDw9PZkzZw4FCxakfv36nDhxgrS0NDZt2sSuXbuIiIigWbNm3Lp1i6lTp5psH/vU1FTefPNNSpQogbW1NQsXLsTLy4vExERmz55Nq1atmDRpEu3atWPy5MksWrSI48eP06JFCxo2bMjcuXNp3bo1DRo04NdffzWew3Xr1vHo0SPKlStHUFAQI0eOZPLkyeTNm5d79+4xbNgwvvzyS0aNGsWoUaNo0qQJS5cuZezYsZw6dYoHDx6wcOFCKlWqRGpqKmfPnqV58+bcunWLX375BX9/f+O4qm+//RZ7e3t69Ohh7Kc7bdo0RowYwY0bN4wfup/0CQ4LCyMsLIzRo0ezZs0aFi5ciJWVFa+99hqTJk3ihx9+wNHRkXbt2nH69GkCAgKYM2cOY8aM4cyZM9y4cYOAgAAGDRoEwMqVK+nduzfnz58nJiaG0qVLc+3aNUaOHMm5c+eIiIjg+PHjxtdd3bp1cXJy4uOPP6ZatWrUrVvX2NXg4cOHPH78mIIFC1K3bl2OHz+OjY0N8+bNo2fPnnTo0IGbN29Sp04dypYtyy+//IKDgwMTJkygZ8+expb65cuXkyNHDgYOHMhHH31EamoqgYGBmJmZ0aRJE2P3va1bt3L16lWqV69OoUKFqF69OoMHDyY2NpZChQoxb9486tWrx61bt6hZsyYWFhbMnj2b+fPnU7lyZS5evEjevHlp2rQpdevWpUaNGjRp0oT79++zfv167OzsmDhxIuPHjychIcHY8DZnzhxWrlxJcnIys2bN4v79+xgMBh4/fkxqaipTpkzh9ddfZ+3atWzevJkpU6YwY8YMlixZQqNGjVi0aBHVqlVj7Nix5MmTh2bNmrFp0ybi4+OZPHkygwcPpnfv3mzcuJG0tDTjgO/Y2FgiIyO5ePEi06dPZ/DgwTx8+BAPDw8WLFhAoUKFuHjxImZmZqxcuZJ58+Yxa9Ysli5dipubG61btzbOzHbmzBl27NhB+/btjV2uPv30U65evcqIESPYunUrly9f5qOPPuLu3bsUKVKE6tWrs3v3bszMzIiNjTV2AcmXLx++vr7s2rWL2NhYrKyssLOzo1y5cnTo0IHIyEi8vb3JlSsXn3zyCR4eHnTo0IE7d+7Qq1cvrK2tjQ2rAwcOpG3btri7u9O6dWt8fHwoX748BQoUoE6dOnh4eNCiRQtq1qxJ8eLFadq0KWlpaTRp0oSuXbtiY2PD8uXL2blzJ/Pnz+ebb77h4cOHrFy5ksaNGzNhwgSsra3x9vZm9uzZjB8/np07d+Lj48PGjRvZtm0b586d48yZMyxdupTp06dz9epV7t27x+PHj1m1ahVr1qzhzJkzTJs2zZhVzp49i8Fg4MyZMyQnJ7Nq1SpiY2O5efMm8+fPZ/HixQQGBrJixQoiIiJYs2YNO3bsAH6bdScxMREXFxeGDx/OoUOH6N69O5GRkbz77rusXr2aN954gw8//JBJkyYZx1Z9+OGHTJs2DQsLC8aNG8e8efOYN2/eC828ZVJ97OfPnw9A1apVmT17Nm5ubnTr1o3p06eTL18+HB0dOXr0KC1btsTS0pISJUpw/vx5li5dyuXLl3F2dubUqVPkypXLeMn45MmTxkG0lpaWWFtbExgYSJUqVdi1axeXLl1i8eLFXL16lQMHDrB7925mzpxp/OP/ZF9WVlacOXOGiIgISpYsibm5OV988QVBQUF07dqVevXqsWXLFooVK0ZISAinTp0iNDSUPXv2YG5uTteuXVm3bh3r16/H1dWVixcv8uDBAy5evMiJEyeYNWsWuXPnNl6qsbCw4PLlywwYMIATJ04wbtw4atSowdq1a439/JOTk2nTpg3Lly/Hzc2NgIAAHj9+TNu2bWnZsqXxcueSJUs4d+4cQUFBlCxZki5dutC9e3fgtzeQe/fu8fnnn9O0aVMuXLjAokWLjP1Rq1SpgpubG7NmzcLR0RFzc3NiYmKIioqiffv2BAQE4OPjw507d4wDWV1dXQkICMDDw8PYFaBTp06kpaVx8OBBY6vn/fv3iY+Px93dnXPnzrFx40ZGjx5NYGCgcUq648ePc+vWLUqWLImzszM//PADbdq0Yd68eeTPn59SpUpx7tw5mjZtypw5c7hx4wYDBgwgLi6OnDlzkpiYyLRp02jXrh3FihUjMTERT09PGjVqRIsWLYiLi8PLy4vHjx9Tq1YtWrRogZubG/Hx8dy8eRM3NzcuXLhAaGgoS5YsISoqiuXLlzNo0CDGjBljbKVv3rw5n332GYGBgWzatIk8efJw+/Zt1qxZY5yp5siRI5w7dw6DwUDHjh25fPkyoaGhWFtbc+PGDSZPnkxYWBgWFhbkzZsXFxcX2rZti7W1NTExMdy6dYtPPvkEGxsbSpQowYEDB0hMTOTYsWMcPnwYMzMzwsPDuXXrFm3atGHQoEHExcWxbds2Vq1aRZkyZbh48SLh4eEEBweTP39+hg4diqurK0lJSRQoUIBcuXJx+vRpChQogKenp7F/4JMxAJ9//jlr166lY8eOfPrpp9SsWZPSpUuzY8cOAgICjF1JDhw4YDx2yZIlOXv2LMWKFWPZsmWMGzcOa2trPD09GTVqFIsWLWLmzJlERUURGBjIr7/+ire3Nzt37iQoKIh3333XOI7g3Llz1KxZkypVqlC4cGFWrlzJmTNnuHfvHitXrjR+CD979iwVK1YkLS0NX19fypYta5yOsHr16sycORNra2u++OIL7t+/z6lTp8iRIwf9+vWjfv36zJ49m9y5c2NhYUF8fDxTpkyhSpUqWFhY4OrqSnh4ODY2Nvzwww9069aNWrVqYWVlRZUqVZg8eTIDBw5k+PDhbNiwgbZt25IzZ07MzMz46quvSEhIwNfXF1dXV+zs7OjcuTPJyckcO3bM2K+2b9++REZGEhsby6RJkzA3N2fEiBHcu3fP+OHX0tKSCRMmEBQURExMDP369cPa2povv/ySqKgoUlNT6dq1K3Z2dsybNw/4rY97ly5dsLS0ZN26dbi6umJvb0/nzp1JS0tj//793Lx5ExsbG3r06EFsbCwXL15k8uTJWFpaMnjwYGPf5lGjRmFubs6HH37IrVu3iI6OZuDAgVhbW/P555/z4MED4wfPU6dO8fjxY06ePMndu3fJmzcvJ06c4IsvvgB+mw3m+PHjvPnmm3z33XfcvHmTpUuXEhwcjKenJz169MDHx4fOnTsbZ9GZMWOGcQrUiRMncv36dXLlysXQoUOpXbs2SUlJpKSk8PjxY+zs7DAzM+O9997j4sWLANjZ2bF27VqcnZ3p1asXDg4OfPfdd+TJk4e1a9eyfPlyKleuzIoVK3jzzTcZP348+/btM3ZXKVmyJDdv3iQoKAhzc3OsrKwoU6YMhQoVwsHBgQcPHjB//nxcXV0ZOHAgfn5+zJkzh/Lly1OrVi0CAwNxdHSkWrVqbNu2jfLly7Nu3ToCAgK4desWgYGBWFlZMWHCBCZNmoSnpyezZs1i9erVJCQkGLs5Xrt2DS8vL0JCQli5ciUrV66kcOHCxlbtFi1acP36dRYsWMDx48c5deoUnp6e5M2bl06dOnHv3j3jGI7169cTHR3NuXPnyJ07N/fv3zfOULVkyRIePXrEpk2buHv3LikpKXzzzTe4urryySef0KJFC2JjY419293d3Vm/fj2+vr7kz5+fmTNnsmzZMoYOHUpsbCwJCQmcPHmS+vXrc/LkSRo0aMCKFSvYs2cPqampJCUlYWNjw4MHD8iRIwetWrWiVKlSxpm1tm3bhq2tLXnz5sXf359FixYZr7D5+PhgY2MDwN27d3FxcSEyMpIuXbpQsWJFfvzxR65fv27snrdlyxasra05evSosWvV2bNn6d69O3Z2dixYsIC8efOyfv16bG1tsba2JjExEWtra5ydnYmMjMTe3p4bN25w7949LCwsCAsLY9u2bTRs2JDXXnuNvn37EhcXZ/xQ/OR9Z+LEiQQHB2NpaYm7uzvFixdn7dq1xm6Kx48fp3Dhwqxbt45NmzYZx0zBb1Nm7t+/3zgjzJM++bdv3yZ//vz4+fnRpk0bxo4dS9WqVbl9+zZt27Zlzpw5FClShIcPHxIbG0toaCg3btzg9OnTnDp1CkdHR+rXr8/EiRNJT0+nWrVqfPTRR4SHhxv7nhcvXpzo6GiSkpKM42Ry5crFzZs3efjwIR06dGDevHkcOXKEmJgYihYtipeXF+bm5tSpU4dt27ZhbW2Nr68vTk5OtGzZkhUrVvDw4UNy5szJ3LlzuX//PlZWVrRq1YqjR49y+PBhWrRogbOzs7HbW0hICKVKlWLq1Kk4OTlhbm5OoUKFuH79Oj/99BO2tracPXuWQoUKMWHCBNasWUNaWhoPHjygffv2TJ8+nfDwcAoXLmycwe/OnTsUKlQIf39/ihcvzscff0xUVBRpaWnG2RWLFi1KYGAg3t7e5MmTB1tbW8qWLYuFhQWNGzc2ds16IrMz4oCJdsVp1KiRMdAMHz6cunXrEhAQYOz3OXbsWEaPHs3169f58ssv8ff3x8HBgffee49cuXIRFhZGvXr16NWrF/Hx8RgMBgICAihSpAh3795l8ODBBAcHc+fOHWbMmMGoUaPYt28fu3btom3bthw/fpzq1avTv39/3n//fTZs2MDcuXOpWLEiY8aMwdXV1TiwaujQoQwcOJB3332XMmXK0KVLFwYNGoSrqys//vgjX3/9tfGyjLOzMwMGDOD+/fv8/PPPbN68mV27dvH9998zYsQIY0tVxYoVjS12UVFRpKSksHjxYooXL46/vz82NjaULFkSMzMz1q1bx/jx45k3bx7Dhw9n/fr1xkGuQ4YM4cCBA4SFhTFo0CAcHBxYsmQJQUFB2NvbG990OnToQFhYGK6urlhYWBAbG8tbb71Fnjx5AIzdkerXr0+PHj3w8PCgXr16TJ06leDgYAwGA/fu3WPJkiXky5cPgCNHjvDjjz/y4MED8uTJQ82aNTl9+rRxysvu3buzdu1a2rZty61btxg6dKhxMKmfnx+1a9c2TsGZK1cuvvnmG5ycnFixYgWXLl1ixowZ1KlTh/r16zN06FDjKPeqVatSoUIFypQpY7zsHRYWZpz54eTJk7z22muULFmS999/n5iYGMqXL8/JkyepVKkSzZo1Y9asWXTp0gUvLy/j2IEnf2i3bNnCjBkziI2NJWfOnBQsWNA41d2jR49wcXFh69at5M2bl4YNGxpn8Ll48SJNmzZl7969lC9fHg8PD/bs2UOTJk2Mb6I2NjbGFvocOXIQExNDsWLFuH//PufPn+fatWvGUFm2bFm8vLxISEgwTiVWuXJljhw5Qo0aNShQoABmZmbMmDGDihUrUrNmTYoVK8bp06d5/fXX2bJlC2XKlMHOzo7o6GgKFy5Mhw4d+PXXX41XA54Mhvzhhx+oUqUK3t7eFC5cmICAABwdHZk3bx43b94kNjaWChUqUL58efLly0dAQACVKlUyzgryJEA9mZkkJCSE3bt3M27cOLZv346Pjw+rV68G4Pz58/z8889MmDCBTZs28emnn/L666+TL18+Dhw4YOwvfvnyZeMsU7Vq1aJhw4bkzZuXnTt3YmFhQY4cOYyDsdzc3Gjfvj07duwgOjqayMhILly4gL+/P7169eKHH36gSJEieHp60rp1a9q2bcv+/fsJDAykQIEC+Pv707hxY8aMGUOZMmVo3bo1c+fOpXDhwowfP94Y8p9MS3rr1i0WLlzIG2+8wcaNGxk+fDhbtmzh+vXrjBw5kgULFuDo6EizZs2M852bm5sb+2U/mZP8yc+/X/a07Z5MV/e0ZZaWlsapZ5+sf/IGYzAYjNv9ft+/X/a0Gp78/GT7Z9VQoEABrly5wr1795g7dy4RERF899139OnThxw5cnDo0CGWLVvGvXv3CA8Px8rKivr16xvnVm/UqBE5c+bE3t4ec3NzQkNDmTp1Kt9//z0GgwEzMzPjB/BChQphaWnJV199RXh4ONeuXSMsLIyNGzfy5Zdf8tlnnxEUFIS3tzcAFy9eNP6dadasGQsXLqRDhw7Gfr5XrlwBfmt86NevHz/99JNxJq8lS5aQnp5O4cKF8fb2NvaFz5kzJ+XKlePKlSu4ubkxZ84cLCws+PXXX40Dt48dO8bUqVMJCwtj0qRJDB8+3Ni91NzcnD179jB9+nS2bNlCy5YtGTNmDLVr1+bmzZsEBwfz+uuvU6BAAdq1a0eePHlo0aIFb7zxBu3bt6dWrVrMmjWLkSNHsnr1asLCwhgwYAAbNmzgyy+/xNvbm8mTJ/Paa69x+vRpY3CvXbs2efPmZcGCBaxdu5bbt28zcuRI4uPjWbt2LeHh4caGFQsLC/bu3UtQUBAdOnTA3d3dOHg+Ojqa0NBQDAYDsbGxJCYmUrp0adLT0xk7dizHjh0jNjaWOXPm0KlTJx4+fMj333/PRx99ZPy7Z2VlxaZNmyhatCiXLl3CxcWFvn37GmfDsra2Nn7wefIht1u3bmzbts3YRWXfvn3GFvrLly+zZ88eypcvz8qVK2nQoAFNmzbl0aNHjBw5kg8//BALCwvy5ctH3rx5adOmDT179mTo0KE0aNCA8ePH4+TkRMGCBXFycmLjxo2Eh4fj5eVFgwYNuHr1KtHR0cZxTQ8ePKBatWpcuHABCwsLSpUqReHChTl27Bi5c+c2Trd55coVXFxc+PDDDzl//jxTpkzh+PHjTJ8+HYB3332XhIQEatSoga+vL2+88QZr167lnXfeYf78+QQGBtKnTx+qVKlCp06dWLVqFV26dOHu3bvMnDkTc3Nz1q5dS3R0NEePHiU+Pp7evXvz4MED9u7daxyHkJKSgoWFBQ8fPiQuLo569erh7e1t7Fb54MED2rVrx7vvvsvVq1dp164dP/74I35+fmzZsoWNGzdy9epVmjRpQu3atdm6dSsODg4UKVLEOLPUpk2bSE1NJTg4mEOHDrF69WqcnZ2JjY2lbNmyXL9+ndWrV1OrVi1OnTrF2LFjadeuHU2aNOGzzz7DwsICGxsbfHx82L59O6+//rpxPEHJkiX58ccf6dGjB6tXr+bu3bs4ODhkGHfUpEkT4zigwMBAfP8feW8VVfW6xm1fIJ2S0iUhCCigIhiIiiiK2N2KYrDsWnZ3LhvF7kQRO1ABaUGwaJDukI7vYA2e8e7xnuz3dH/z0LF1s3TO/3ye+/79ruviRSZOnEhqaqogN/n5+XHx4kVx4O/atSsvX75EXV0dFxcX7t+/z6xZs/Dx8RHfKZmZmURHR9OpUyf09PR4+fIlaWlpDBgwQDw/QkNDRVS8ndrz37z+Jw/21dXV3Lt3j6ioKFJTU9HQ0EBLS4tLly5hYWHB6NGj+fnzJwkJCaxatYpDhw6RlpbGtGnTePPmDQ0NDSgqKmJsbIybmxsVFRUYGhpSWFjI8ePHRQFKU1OTAQMGcO3aNQwNDfHx8eHy5ctkZGQgKysriB8yMjKEhobi6upKU1MT69at4+LFi7i4uPD8+XP27NlDaWkpO3fu5PTp0zx+/FgUBrOyspCWlqa2thZra2uMjY3FKj4yMpJ58+YxYMAAoqKiGD16NDk5OZSXl7Ns2TJev37NkCFD6NixI48ePSI2NpaOHTuSkpLC2LFjefnyJQEBAcybNw8TExO6du1KYGAg9vb2lJWVoaKiwqZNm5gyZQq9evWitbWVpKQkHB0dcXFx4ffv38TExKCgoICHhwcpKSkoKipSUlKCubk5EyZMICQkBA0NDQIDAykrKyMmJoZx48ahoaFBTk4Oly9fxsLCAi0tLd6+fcu0adNoamrC0tKSN2/ecP78eV6/fk1kZCQ1NTWiDCgtLU1eXh4+Pj5kZ2djY2NDz549ARgyZAjr1q1DR0eHnTt3snv3bvT09EhISMDc3BxfX1927NhBS0sLCQkJqKurIy8vz58/f/Dx8eH+/fsUFRWRkJDAihUrGDBggChYnjt3DkdHR37+/MmMGTPYtGmTyCtaWFjg7+9PQEAAd+/eZfny5YSHh6OsrIyTkxPOzs4EBgYKFOX/STsyMzOjc+fOpKWlUVRUxIABA7h79y7du3cnPDwcGRkZZGVlRcGmfT3r6OjIp0+fMDIyIjMzk+bmZrS1tenbty+DBw/m2bNn4oHas2dPgoKCkJaWpmPHjvz8+ZPs7GwhdOvTpw9jx44lIyODN2/eUFNTw/bt2zly5AhJSUlUV1cTExNDQEAAo0aN4t27d8jJydGvXz8KCgowMjLC0tKSfv36ISUlxa5du4iPj0dGRobRo0fz8uVLsWU6ffo0Dx48QENDgw8fPrBgwQJevXpF165d2bt3LxcuXODDhw98+/ZNbJBWr16NlJQUHTp0wM3NjYkTJzJu3DgxGdfW1mbgwIFcu3YNeXl5nJycqKioELnOqKgocnNzxcG0vr5erOEbGhpYuHAhWlpaTJs2DQcHB4KDg0X0oLS0lPz8fGxsbDA3N0dXV5fS0lIOHjwo0JtRUVGi39PY2Mj06dM5duwY6enpxMTEsG/fPkaMGMHo0aNxc3PDxcWFr1+/MnjwYHFZtbGxITY2lsmTJ3Pp0iV0dHTw9vYW5IvVq1ezZ88ebGxscHV1JTMzk4EDBxIfH4+0tDT29va8fv2afv36kZ6eTm1tLc7OzoSEhNC7d29KSkooKSnBzc2NR48eia1Eeno6gwcP5uHDh9jb26OgoEBiYiLDhg3j/v37WFpaoq2tTUxMDMOGDSMkJARdXV1MTU359OkTgwcPJiIiQvD5X716hZubGz9+/KClpYWePXsSEhKCi4sLhYWFlJWV0bdvX548eUKPHj1oaGggMzOTQYMG8fDhQ7p160ZjY6PAyZ0+fZqWlhYx0XJxceHmzZsYGBhQVVVFVlYW2tra7N27FycnJ27evElFRQV//fUX06ZN4+vXr8yZM4fr168zZswYIiIi8PLyoqysDDk5OUJCQsjMzMTU1JSRI0eybt06xo0bR15eniA13bp1Cy8vL+zs7Dhw4ACWlpYkJiayb98+wsLCuHDhAiUlJYwbN44TJ06QkZHBkCFDWLNmDYCYQrcPLZ49eyb8BnZ2djx8+BAPDw8KCgqws7MTiMXZs2czbdo0nj9/zr59+5gyZQoZGRns2bMHAwMDRowYQVhYGFJSUv9BSfv69Supqal07NgRZ2dnMjMzkZWVZcWKFVy8eJGBAwfy8uVLIiMjkZKSQkpKCiUlJdLS0sjOzmbDhg1cuXKFZcuWERgYKKglEydOFMKgbdu20bFjR+DfiGt2djYqKipCJmRqakp5eTl1dXUYGxuTnZ1NTU0NL1++ZObMmWzevJk5c+bg7u7OvHnziI+P5+rVqwIpq62tze/fv3FwcKB///6EhobS1NSEs7MzGhoa7N+/n2nTppGcnExCQgLu7u6kpqaK7bqJiQmDBg0iPj6eV69esW3bNpYtW8agQYNQU1Pjx48fWFpaMm/ePNLT09m8eTMKCgrMnTtXZPhlZWU5fPgwXbt2pUePHsTExDBo0CCioqK4fPkyKioqhISEIC8vz9ChQ7Gzs6NLly7iwv/s2TP+/PnD1q1bOXDgAD179iQ7Oxs3NzcSExNpbm6msrISc3NzTp06xZw5c8Rlv6amBiUlJUF1k5WVJS0tjZCQEIyNjTEzM8PHx0f4TnJycvj58ydVVVWEh4fTp08fOnfuzO/fv+nduzcNDQ3k5OQwfvx4bt26hYODA9nZ2ZSWltK1a1dSU1PR19dn0KBByMvLc/jwYTEc6tSpE/Bv4d7T05OEhARSU1Opq6tj1apVhIWFYWJigq+vr4jHwL+X23v37jFixAhSUlIICwvD0dGR4cOH8+LFC5SUlDA3N+fw4cMsXrwYSUlJwsLCSEpK4uTJk8yePZumpiaKi4txc3PD1tYWgDdv3nDp0iVOnDhB//79Wbp0KadPnyYuLg5vb2/mz59PYGCg8BGYmJiQlZVFaWkpKioqqKqqir/Pzp07U1dXx4wZM6ipqSE5OZl58+YxduxYdHR0WLduHR8+fCArK4svX75gZWXF2LFjyc/PJygoiF69ehEZGYmHhwf29vZcvXqVmpoaqqqqMDEx+Y/z26NHj9DS0hJAgsbGRnJzcykuLiYrK4v+/fuTlJSEsrKy2OrIyMjQ0NDwXx/s/6cy9vAvTuzy5cuC7PDz50+0tLT+w8i3adMmrl27Jgq2tbW11NbWcvfuXczNzWlra2PDhg0i4xgREcH+/fs5duwY6urqNDU1ERAQQGNjI83NzWhpaTF27FjS0tJE7ldFRYWOHTsiISFBfX29IB3Av4a9xsZGrKysMDMzw8XFRax3O3bsSFpaGt26dePRo0ecPHkSLy8vMWUbNmwYERERVFZW4u7ujpqaGg4ODpw9e5YlS5awd+9ewSe2srKioqKCgwcP8ubNG/Ly8sRE7e3bt8yePZuEhATWrl2LlZUV2dnZbNy4kTVr1tDU1ISVlRWnT5+msbGRJ0+e4O/vj5aWFhISEhgbGwt5Vm5uLhcvXuTHjx9kZ2cTGRlJcXExPXr0wNXVFSsrK06cOEFwcDDLly+nqqqKs2fPYmVlxcqVKzl69ChycnLMnj2bgoICpk+fTk1NDVpaWmzdupU3b94watQoXF1dmT9/Pq6uruzevZuzZ8+yfPlyQVm4dOkS9+7dY9myZcTGxtLQ0MDmzZv59OkTtra2mJqa8u3bN1FUuXLlCjk5OWRkZGBubo6GhgY/f/4kKyuLc+fOAXD//n02btzIpk2bWLNmjSjxhIWFiX+T9kPPkCFD+Pz5M0+ePGHBggWkp6fz/ft3vL29qa+v59evX8yaNYvg4GD69OmDr68v6enpGBkZIS0tzYYNG5g3bx7nz5/H2dkZfX19Ll26hK6uLt27d+fGjRu0tLSI1V+PHj1oaWkROeYDBw5gZ2fHwYMHOX/+vGBd79q1i7q6OgYNGkRGRgZlZWVUVlbi4uJCRkYGO3fuZP/+/ezYsQNbW1vKy8u5fv26EEWdOHGC3bt3Exsby9ixYyksLOT69etoaWnRt29fli5dyqJFi0T5tL0kGxoaipOTE15eXly+fJmFCxeiq6vL1q1buXXrFiNGjMDOzo7W1lbS0tKws7Nj/PjxuLi48OXLF5ydnUlLS2PUqFF06NABdXV1Ll26xN27d7l06RKurq68fPmSv/76C1VVVRYvXsybN28wNjZGVVUVGRkZtLS00NXVpa6ujubmZioqKkQES05ODl9fX7Zu3cqIESN4/PgxkydPxsbGhmvXrjFnzhy8vb3/w8RZVVXFmTNnWLVqFZ8/f2bHjh1is2VnZ0dqaqq4PKampjJmzBi6detGQ0ODKNYbGBggJycnJqtfv35l+/btWFhYsHLlSlxcXASmbe/evezfv5+bN29SVVUlplqrVq1CUlKSrKws9PX1KSsro2PHjtTW1opfKy4uRklJiYaGBnJzczE0NKS0tBQJCQkaGxspLCwUm6K6ujokJCQoKyvD2NiY4uJiysvLkZGRobq6WlDF/s+Dm4GBAdXV1WRmZtKpUycqKirQ1NSkurqarKwsDA0NxdSrtraW379/Y2JiQmlpqZCvFRQUYGJiQlFR0X8YIWtrawkMDOTOnTukpKRQU1PD0KFD8fT0ZMiQIfTu3ZufP38SEBCAvb09KSkpaGpqsmvXLqSlpZk+fTq/fv0SyOGNGzeSm5tL7969sbGxYdq0ady9e5czZ85QU1NDTk4O/v7+NDc307dvX86dO0eHDh3o168fBw8epH///pw/f54pU6Zw6tQpSktLuXr1Kjt37iQhIYHjx49z+vRpfv/+zfXr1/nz5w8nTpxg6tSp9OrVSzyjPTw8uHjxIjk5OQQGBqKpqUlwcDCfPn2iR48exMbG8uXLF+7evUtcXBxr1qz5j07S/PnzxRRZUlKS/Px8Xrx4wc+fPzlz5gyZmZnCwO7t7Y2Kigp5eXnU1NSgoKCAu7s7jx49Ij4+HiMjI4HuW7hwIbKysixatEiI+ZqamgReuXv37gJkMGXKFL58+cLgwYPJyMggNzeXoqIinJ2dGTRoEOnp6SgpKeHt7Y2xsTFpaWkkJSWxf/9+oqOjMTU1Fazx9ufb79+/OXr0KAEBAYwYMQJVVVVmzJjBqlWrhBBq0qRJpKenc/z4cWJjY8nLy2PYsGGcP38eX19fjIyMmDVrFv369RNbxBEjRtC7d29sbW2xsbGhQ4cOVFVVoa2tjYODA05OTnz8+JGwsDBWrFjBjh07RHSnqKiIS5cu8fbtW+7fv09ERAQbN27k5MmTJCQkiM3K169fcXd3x8rKirdv3/LkyRMOHDjA0aNHRSYb/rWRNzc3Y2Jigq2trcBfJicn8/jxYz5+/MjYsWNJTk5m3LhxpKSkMGHCBJKSkhg/fjxJSUkiRfD+/XvU1NSwsLDg06dPaGhooKamxrhx46iqqqJDhw7s2bNHUICsra159uwZQ4YM4eDBg1y4cIFLly4REhKCn58fRUVFvHr1irS0NIyMjHjy5AnZ2dkoKChQXl7O2rVrhReknVxVWloqwBHa2tqsW7eOxYsX8+rVKyQkJHj27Bnp6enIyMhw4sQJcZBt753k5OTQsWNHOnXqRH19PXV1dcTGxpKUlISzszNxcXE4OTnRs2dPTE1NuXXrFuXl5WRkZDBixAg0NTVZu3YtOTk5HD9+nMuXL6OoqEjv3r1JTk4mNjYWPT09goKCUFNTY+zYsSgpKSEjI4O6ujo2NjbizyorK2Pt2rWcOXOGt2/fIiMjw549e9DQ0BAD3cjISExNTenYsSOenp4sXryYzMxMLl26xMWLFwWg49WrV7S2tnLs2DF0dXUZPnw4dnZ2+Pj4sGLFCqSkpPjw4YPoSL59+5aSkhI6dOjAjh07KCkpEc6TU6dOcfr0afr06fNfH+rhfyxjD//m69u/IMrLyzEyMmLt2rU8ffqUZ8+eidY7wPPnz1FQUKC+vl487Dp16oSGhga2trZs27aNs2fPCpnNiRMnGDJkiODMp6WlYWpqiqysLCUlJcTExJCUlMTWrVuRkJAgLCyM8ePHk52dja6urmCT9+jRg0+fPhEfH4+cnJxgNMO/PYHq6mr09PTo1asXVVVVqKqqiijEunXrqKiowNjYmF27dlFdXc2TJ08ICAjAyMgISUlJTExM0NDQQFJSUkxz9+zZw6JFi/D392fChAm0traydu1aFBQUhASi/WE4f/58bt++zZkzZzA1NcXBwYH9+/cjJycnSDLnz5+noqKC3r17c+vWLVJSUoQXICoqivv373P8+HFBD9HW1iY9PR07OzsxHTA2NmbAgAH06tWL5cuX8+vXL9LT07l27RqmpqZYWFjQqVMn3rx5Q3p6OpqamhgZGQkOemNjI2/evMHV1ZX79+/TtWtXPDw86NOnD6tWrSIrKws7OzskJSVJT09HS0uL5ORksrOz6dq1K/369cPV1VXw+NPT03FwcKC6uprp06dz8OBBCgsLOX/+PJqamqJU5uXlRWBgIGPGjOHbt29cvHiRtrY2vn37hqqqKnp6ehgZGYkPsqSkJBkZGZiYmIj30Lp16/Dw8GD9+vX4+fkxatQoFi1aREREBE+fPqVLly58+/aN8PBwSkpK0NXVJSEhgerqarS0tPDx8SEkJITBgwcTHx9PSkqKIJno6emhpKRE586dcXBw4M6dO9y+fZvW1la6du3Ko0eP6NGjBxs2bGDUqFH8888/vHjxgl+/fuHo6Ejnzp358OEDsrKyqKqqigNOdnY2q1atoqysTEzDIyMjiYqK4smTJxQWFpKYmIiysrLIcPbp04etW7cKTm9ZWRnh4eEMHDgQDQ0N5s2bR1xcHPn5+ZSWliIlJUVoaCgTJkwQEYL2/83QoUNZv349FRUVODs7s3jxYq5fv46qqipqamo0NTXRrVs3qqurefr0Kb9+/WLXrl0UFBTQt29f3r59S58+fejWrRv9+/cnKyuLmJgYqqurSUpKoqioCGtra379+sWaNWt4+PAhkydPFg/UhQsXoqqqyv79+5GQkBDFNfg3lnLmzBmio6PJz88nNDQULy8vFi9ejKurK7169aK5uZnQ0FB+//5Nly5duH37NqGhoUhJSaGoqEhYWBjDhw9HTk6OT58+UV9fz7p16wQNpH213n4QaKdu3Lp1CwkJCQoLC/n27RtdunQRB9rq6mpevXqFubk5wcHB1NXVCcGMmZkZL168oKamBnV1dYKCgtDX1yc0NFQ8g+7du4empiYxMTFUVFRgaWnJzZs3UVFR4du3b5SWltKlSxdu3bqFtLQ0v3//pqSkRKzj26eRb968wdLSkkePHomf4dmzZ5iZmYmJpoaGBkFBQRgYGJCWloaWlhYuLi7Cd3Hw4EFhG5WVlRX876CgILp06UJlZSXh4eEcPnwYBQUFrKysWLZsGdOmTePx48fCy3H//n1Gjx7N79+/RaxLUVGRmpoaoqKiuHLlCidPnqSgoIBDhw6xadMm6uvrBVWsnXU9adIk8WzR09Nj165dInajqKgoXAfv3r3jwoULgm6jpKQkOPCNjY3IyckhJSWFp6cnly9fZu7cufTp04e8vDzKy8spLS3l0qVLNDQ00NDQwLhx45gzZw7//PMPe/fupaCgAHNzc5YsWYKFhQUPHz7k9u3bNDU14enpycuXL9m7dy8vX74kNzeX1tZWmpubmTlzJgYGBigpKVFZWYmHhweurq6CKmZkZMShQ4fYu3cvycnJFBUVUV5ejrq6OqmpqYJzr6mpKQgltbW1AhEcFBTEz58/aWhowMLCgvPnz5OYmIi8vDyBgYFUVVVRUlKCkpISV65cQUtLi7a2NhHHkJSUZNiwYZw5c4bk5GQ8PT2RlZWlvr6e1NRU1q5dy6ZNmxg+fDi6urpkZ2dTXFzM7du3UVFRIT8/n+3bt4vvnJSUFIGCffXqFTdu3MDa2pqCggL2798vPl/q6uro6+sjKSlJly5dxCYyODgYgPHjx4vLXadOnbCysmLWrFmsXr2aYcOGkZSUxNKlS5k+fTrr169HVlaWlpYWdHR06Ny5M4AoqVZWViInJ4e5uTnS0tJoaWmJC2/7NhPg4sWLyMnJcefOHeTl5YmIiBDiTVVVVdFDaOeft0utJk2aRFVVFePGjePQoUMiGlVVVcXp06c5dOgQzs7OtLS0sG/fPtLS0igvLxcleykpKTw8PAgODubs2bOoqqoyc+ZMtm/fLvCxubm5PHv2jKCgIHR0dKipqcHPz4+goCCsrKy4ffs2t2/fFsbtdu9Aa2srL168oG/fvnTu3BkJCQl0dHR48+YNXbt2JSQkhPv372NiYoKqqipKSkqcPHmSqqoqJk2axL59+1i/fj1OTk4MGTIEZWVl0tPTefr0KQMHDsTLywt3d3daWlooLy9nx44dNDY2CqNs+9D306dPAoVdVFREaWkp1tbWdO/end69e3Pt2jW+fv3Khg0b+P79O3JycgQGBvLx40f09PRYuHAhf//9t3B0PH/+nKFDh7Ju3ToKCwvp1asX7u7u2Nvbiwvjzp07RSpETU0NbW1tysrKqK+vx9zcXAAj7ty5I+y4/y+v/8kozvr169m8eTNVVVUsW7YMOzs7vn79ipOTk5BT6OnpcevWLSIjI9HU1CQ/P19Y6kpLS/Hy8iI+Pp65c+fy+/dvfH19UVNTE2/ic+fO0dDQwNChQzl58qQQjrTrjJ2cnLhy5Qpnz57l8OHDWFtbM3nyZAYMGEBLSwvdu3fnx48fFBYW4uHhQVtbGx8+fODs2bNs3ryZBw8eMGLECP7++28UFRWxtbWlubkZb29vjh8/zuHDh/n27Rtv3rwRkZvg4GD8/Py4e/cud+7c4dWrV4SFhaGkpCRKqUVFRQwZMoT4+HjxwCkoKMDCwgJzc3ORY9TS0hK5y/bX27dvUVBQQEFBgStXruDr68vq1atxcnJi5syZ9OzZk2nTptHc3Mz27dtRU1MTG4/nz58zZcoUtLS0iI2N5dy5c/z+/ZusrCw0NTUpLS1lyJAhLF68mBs3bnDlyhV69+7NoUOHyMvL4+vXr4wYMYJp06YxevRo5s6dS8+ePfH09CQ8PJzm5mZWrVqFgoICtra2/PjxgydPniAhIYG5uTmWlpZERUXx/ft3ZsyYIda/kZGRgsaioaFBbm6u6Ehs376dr1+/oqGhgYeHB9OnT+fp06fcv38fGxsb3N3dsbOzw9ramqioKMGqjYiIwNTUFEAo5UtKSmhra6OkpISuXbsyaNAgTp8+jYKCgliBhoWF0bFjR2bPns22bdswMjJi3bp1dO3alZqaGmHM+/btm8DMjR07Vkz9IyMjMTMzo1evXuzYsYNfv34hLS2Nnp4eM2bMYMSIEfj6+hIaGsru3bv58eMHwcHBdOjQgd+/f7Nu3TpGjBhBdnY279+/F0p3GRkZ7t+/z5YtW9i+fTvV1dVs2rQJaWlpUlJS6NChA5mZmSJu1traSklJCUOHDqW2thY9PT2qq6uRlZXF0dGRsrIytmzZgoWFBUpKSuTk5LB//34aGhpYvXo1gDjsKCgo8OzZM5YsWUJ0dDQ6OjpUV1czbdo0IiIikJGRwd3dnXv37qGvr4+VlRWKiookJCTw6NEjLl68iJmZGb6+vvz+/ZslS5YAiFhT+2bszJkzWFhY4OTkhJubG62trWRnZ6OlpYW9vT3Xrl0TlIb6+np+/PjByZMncXNzIz4+niFDhnD//n2mTJmCpqYmdnZ2eHh40NTUxLx588jIyEBNTY2KigrU1NTIy8tj7ty5pKSkMHLkSGE1jImJoaSkhNzcXGbOnMm1a9cYO3YsBQUF7Ny5E39/fz5+/MiUKVPQ09MjJycHHx8fzp07J362b9++MXr0aK5du0aPHj1QUlLi48ePjB8/ngcPHmBtbY2hoSGPHj1i0qRJvHz5El1dXezt7bly5QqTJ08mIiICeXl53NzcOHr0KFOnTiU5OZmmpiaGDh0qsJjt01pvb29OnTrFsGHDqKmpESjU9s2KrKwsERERjBs3jnv37tG1a1c6derE8+fPmTBhAs+fP8fAwABra2tu3LjB5MmT+fnzJ6tWrcLc3Jzbt2/z6tUrHB0dMTAw4MGDB8yfP5+bN29y6NAhvn37Jt43LS0tmJubM3bsWExNTXn48CFOTk68e/dO/BntUcoZM2YAYGdnx5YtWzA2Nuavv/7C0tISOzs7Pn/+zLZt2xg+fDju7u4kJSXR1NTEy5cvxca3nRzTHhUaP3484eHhNDQ0YGtri7u7O/X19SgpKfHr1y+mTZvGx48fRdnu7NmzohSZlJREREQEFhYWfPnyhTVr1rBu3TrMzMxobm7m9OnTGBkZkZSUhIKCAt+/f6dbt244OzvTs2dPAgMDmTNnDhkZGXTu3BlpaWnk5OQYMmQIDx48YNSoUSgrK1NbW8vhw4fF4WTmzJkUFhZy6tQpFBUVUVJSEgfAS5cuAf8iJ5cuXYqFhQXR0dFMmjSJS5cuoampSe/evdHX1xdDiClTpgjMpImJCdra2vz584dTp04RFxdHx44dsbKywt3dnStXrmBkZET//v3x9fVly5YthIeHEx0djaurKydOnCAiIoL3798TEBDA06dPOXfunMBD+/j4CKnchQsXUFVV5dChQ5iamjJnzhzOnj3LoUOHWLNmDc3NzXz9+pXdu3dTVVUlZJPXrl3j2LFjmJiYiIJjuwivU6dO/P79GzMzMzw8PMjOzkZaWpoBAwYQGBjIwoUL6datG0OGDGHu3LnEx8ezdu1axo8fz507d9DR0eHjx49ISUkxefJkXr9+zePHj5GWliY7O5vhw4czb948Pnz4gKOjI3/+/CE7Oxt1dXWqq6uFJOnDhw90796d4cOHc/nyZbZu3SoKxR4eHrx8+ZJHjx7x9u1btm/fLszlgYGBDBw4EGlpaSoqKmhsbMTGxoaYmBhBQzt//jxtbW00NTUxffp0rKysqKur4+TJk/Ts2ZPk5GSOHz/O+fPnef/+PRoaGpw/fx5tbW3Gjh3LvHnzCAgIIDMzU3Tr2jPkurq6PH78mOrqan78+EFBQQFz5szh/v37xMbGoqSkhJWVFUVFRSQlJbF7926Cg4NFzFNPTw9TU1NCQ0Pp3LkzI0eO5P3795SXl+Pp6Ym8vDwHDx4UXpzZs2czb9483N3duXz5Mr179+by5cu8e/eOSZMm0adPHzIzM/ny5QsZGRk0NzezaNEinjx5wufPn9m7d6+wL7e0tKCsrCzOQH/+/GHJkiVcunSJQYMG0bdvXyIjI5k9ezYxMTFs3bqV6Ohorl27hoSEBEZGRgK/qqSkBEBkZCTa2to0NTUJIWh8fDxqamr8+fOHgQMHiiFhZWUlkpKSDB8+XET6/pvX/9zEHuD8+fOsWrWKJ0+ecPfuXVRVVamurubXr1+iHHjt2jXOnTvHpUuXmDx5MpcvX2bBggXo6uri5OTEtGnTmDBhAhoaGqxatQpjY2NOnz5N9+7dhV1w48aNlJSU8O7dO3R1dYW978iRI9TW1rJu3TpSU1NZsGABo0aNIjg4mPT0dJSVlcnIyCAqKgoVFRWuXbvGvXv36N69O4WFhaxZswZVVVUWLFggbs6vX7/GwsKCvXv3iqnnpUuXBDUmPT0dfX196urqKCsrY8+ePUIq1dDQgJ2dHU+fPsXFxYVTp06JFd+7d+/w8/MT/391dXV8/PiRCRMm0KFDB/bt24ecnBwjR47k1atXFBQUUFJSQrdu3Zg5cybBwcE0NzcTFBQk0Fzv378Xa3QDAwO8vLyQl5fn6dOngtYTFxfHvHnzuHjxIr9+/WL+/PmcOnWKuXPnsnPnTiFhGjt2LEVFRZw+fRpXV1fBgW+PLLWjFMPDw/n27Rs2NjbIyclhb29PaWkpHz9+5NKlS2KKHR0dLX6uUaNG0dTUJLCgY8eO5e3btyxYsICmpiZOnTrF4sWLUVdXJzw8nH/++QdnZ2eBL/v69Stqamps3boVLy8vtLW1OXz4MCNGjCAkJIRdu3bx7t079PT0ePbsGTY2NhgYGLBo0SK0tbXFir9Xr16EhYURFhaGvLy8EK/Av5fU5cuX8/LlS1RVVcnMzKSmpoauXbvi4uLCwIEDyc7OFlIUKysrtm/fTk5ODvfu3WPv3r1iyzB58mQaGhrYuHEjYWFhJCQkoKamJoQb1dXVeHp6Cqa7ra0tRkZGzJgxg+LiYvLz8zE0NOTTp09MnTqVlStXcuXKFQICAnj16hV+fn6C6NGO/erataug09ja2vL48WNxWejcuTMHDx7Ezc2NwMBACgoKSEpKorGxUVxQnJychFilpaWFz58/C1JJ7969UVZW5vTp04wfP57du3cjJydHbW0t9fX1GBsbM2PGDFRVVYVYp32KuHr1ajp27MjTp0+Jj4+nV69eaGpq8vLlS7p06YK5ublwPmzfvl0YOr9//w6Arq4uxsbGODk5YW9vj5KSEgUFBdy5c0eUyWRlZencuTOxsbEkJCTw7ds3Pn36xOvXr5GVleXbt2/IyMgIDn97eTslJYUfP35w7949nj9/jqSkpODQt2fUe/XqJTjiPj4+FBcX8/37d7p06SIK3qWlpSQnJ9OvXz++fv3K0KFDqaqqIjY2FjMzM379+iWmRBERETg6OpKZmUlBQQEtLS18+PCBQYMGkZ+fL0gdISEheHt7U1BQQGZmJkpKSrx7946RI0dSXFzMz58/MTMzIzw8HG9vb8rKykhOTqZ3797Ex8fj5eVFRUUF8fHxjBw5kh8/flBeXi6IN/b29mRkZFBYWMiTJ0/Q1NTk+vXrpKWlkZaWxrlz57C0tOTdu3fU1tayceNG5OXlkZKSQl9fn82bNwP/EiTmzp0rhjoREREMHDiQsWPHMn78eC5cuCBoZUePHhWOk48fPyIvLy8EM+2Xu/be0+/fv5k5cyaamppoaGjw8uVLoqKiCAkJEaVNCQkJVq9eLeRKXl5eYmPy+vVrMYXMzs7G0dGRvLw8rl69SmFhIT9//mTSpEns2bNHIDXbKSNTp06lpaWFwsJClJWVkZKSon///qSkpIjJeztJqqamhhcvXqCgoEBzczNnz54lOTmZ169fU1BQQGxsLBs3bhSW8XZrMsDHjx95/vw5b9++5c6dOwKnqaCgQHFxMRUVFSgrK3P16lURzQwJCaFjx440NTWJyWlTUxP+/v7k5OQgIyPDqFGjCAsLQ1lZGUtLSz5//oyvry9jx47l6NGjVFdXC7b5rl278Pb2pkuXLjQ0NLB7926Sk5Pp0KED586d49GjR3h7exMcHExhYSGKioqcOnWKrKwsGhsb+fr1KxYWFly/fl18BtvpalJSUnTq1AlHR0fev3/PgAEDaGhooLq6mk6dOlFeXs7jx4/R1tZmypQpTJ48mWnTprFnzx4+f/6Mj48PRUVFHDhwAG1tbTZu3MifP38YMGAAly9fFp0wd3d3iouLBfa5pKSE6OhoQWtJSEigvLycDRs2cPfuXVpbW0U5NiIiAj09PSGmy8/PZ8yYMSxdupRJkyaRkJBAZWUlfn5+wL8bw9DQUGxsbCguLqaurg5bW1s2bNhAcnKyQEu2XxJKSkqQkJBg27ZtnD59mszMTEaNGsXbt28ZPnw4Bw8eFB2B2NhYysvLKS4u5vDhw+zYsUNcmidOnMju3bspKCgQz9n2KXZ75CgsLAxdXV1GjBjBsGHDhMCwoqKC5cuX079/fxQUFNDV1eX9+/fAv7S98vJyABHTaadvHT58mCNHjvD333+jpqYmEN/tv7cdMZqfn/8fZm1paWmSkpJQVFQUFnYbGxvxjLSysmLDhg10796dy5cvU1NTg5SUFIWFhSgoKODo6Ehqaqoo47fb2fX09Dh48CC/fv0SpMWwsDC8vLw4efIkkpKSmJqasmfPHnr06EGnTp349esXmzZt4t69e8TFxWFqakpCQsL/lbEvKSkR8sL/3x/s3dzcOHv2LIWFheKh7+zsTHp6ungw6OrqCr53Xl4e169fF6ZaJSUl5OTkePjwIbNmzUJfX59jx46xbt068vPziY6OFhnwvLw8xo4dy4wZM1i/fj0zZ87k5s2bAvU4adIk7t+/z/bt29HT02POnDmCdqOgoICenh65ubksWbKEsrIy9u/fzz///MO+ffsEGaVnz57s27eP6Oholi5dysuXL/H29ub9+/eMGDGCJUuWMGLECPr27Ss+EGPGjCE/P5/Kykr09PSIj4/n8+fPIsu/Y8cOXF1dkZKSYtu2bZw/f56NGzcyevRovn37JjCBvr6+nDhxgh07diArK0tdXR3z58/n4MGDjB49muTkZOrq6qivrxd5QRUVFaytrTl16hS2trYiKnT79m0hrli7di0/fvzg6tWrPH36FAUFBdra2rhx4wYABgYGLF68mCdPnlBaWkpcXBxKSkrs3buX7du3i4vE2LFjiYuL48OHD7x7946nT59y5coV0tPTGT58OOPHj0dfX58JEybw9OlTUTy9du0aqampJCYmUlZWhpubG2/evKGoqIgVK1aQlZXFvXv36NevH7W1tZw4cYK5c+cSGhqKtLQ0GRkZNDY2Mn/+fLp06cLSpUt5/fo1d+7cobi4mPDwcEFl0NHRoaWlBTU1NWxsbAA4c+YMN2/e5N69e5w+fZqBAwcSGxvLxYsXhRRHR0eH+fPnExoayuTJk5k1axZOTk7U1NSI7KqPj48wLX78+JGBAweKg+2mTZvQ19enra0NPz8/vn37xqRJkygpKWHz5s2cOXOG8vJympubefXqFZGRkZSVlaGsrMzMmTPFl5yfn5+gGty+fZtevXoxb9483r59C8CXL1+QkpISl6j2LLuNjQ21tbVYWloKoceDBw/w9/dn165d9OjRg4sXL5KRkcGsWbPIz8/H2NiYTZs2sWPHDvr374+VlRV//fUXra2tVFRUCDrHhAkTOH36NC9fvmT+/Pm8fftWGHUXL15MRUWFcChoamqSkZGBvr4+jo6OqKioMHLkSEJDQ+nRowc9evSgpqYGS0tL5OTk2L59O+Hh4cTHx1NXV4e3tzenT58mMTERS0tLvn37Rm1tLQMGDODly5d8//6d+/fvY2BgIDjkBgYG1NfX8+DBA0xMTAgICKC6upq0tDR+/vzJ0aNHBd2j/Qv9w4cPeHp6MmbMGKFhnzdvnsgR9+jRg4yMDP755x8MDAyQkZEhICCAuXPncvbsWbp27cr8+fM5f/48w4cP59KlSygpKaGlpcX58+fp2bMn169fp7S0FAcHBwICAjA3N+fFixfiMnDmzBnU1NTIycnh1q1bzJ07l5MnTwKgrKwszNYnTpyguLiYXr16cfbsWRYsWCCcF1OnTuXChQuMHz9eTPVUVVWFC+PatWvU1NTQpUsXAgICRHQoKSmJQYMGcfbsWbS0tIQV+cSJE7i7u4uewLRp0wQ17Pbt2zx48IChQ4dSUVFBSEgIixcv5sOHD1hYWHDz5k0AcSi5ePGiYL5PnjxZiIzi4uKQlJSkrKyMmpoaJCUlcXFxISsri1+/fmFoaEh6ejqnTp1ixIgRHD9+XHzWtm3bhr29PbKysrx8+RILCwusra3JyMhgypQpaGhosHTpUqGzl5eXp6mpicGDB7Nx40aOHTvG7du32bt3L3fv3sXMzIykpCQ6d+5MVFQU27Zt4/jx48JN8e3bN7p168ayZcv4+PEjnp6euLm5ISsry9WrV3F2dsbGxobZs2czYMAAhgwZgqSkJPPnzycgIIDTp08zY8YM5s+fj4ODAytXrsTGxob8/HxMTEwYMmQIq1atwsvLiwkTJgD/xm2kpKS4cuWKKI8PHDiQkSNHoqSkxKNHj/j7778pKioiJSWFQ4cOMWfOHOGQCA0NRU9PT3ynnT9/Hnl5eW7dusXq1avx8fHBzs6OXbt2ER0djYaGBhoaGnTr1o1Zs2YB4OHhQa9evZCUlKShoYHg4GC6du3K79+/Wb9+Pc+fP8fT05O4uDjh33jw4AFOTk6Eh4czdOhQgcHU0NAQJckBAwZw69Ytli1bxoYNGzAzM6OyspLx48cLyVJDQwPbtm0jKSmJAQMGkJGRgby8PG/fvqVLly4CX9y/f38+fvyIr68vkydP5smTJ2zbto2bN28SHx8vuhgPHz7EzMyM8vJyPn78iLW1tYBsKCoq8vfff9OzZ0+BOW7v6bULKX19fRk3bhwjR46ksrISQ0ND1q9fz7Fjx+jduzeRkZEEBAQwZswYtLS0aGpqIjExkdu3b7Nnzx5evXqFuro6X7584fXr16K7Fhoayt27d3n06BHW1tY8ffoUWVlZ9u/fT0pKClFRUejq6qKoqEiXLl2wt7cXfw/tZ6J2DLOGhgaJiYmMGTOG4cOHi8uVlJQUjo6OHDhwgLq6OvLy8sTZYtCgQSgoKJCSksKXL184f/688AJ8//6dq1ev0trayqVLl6iqqiImJgZ1dXXgX3dPfX0969evp7q6Gg0NDUJCQliwYAFBQUGoq6vTrVs3zMzMePToEfr6+mKompSUxOfPnzl48CBWVlacOXOGq1evYmpqir29PRMmTODy5cv4+/uL94iPjw9du3bF09OTsrIydHR0qK+vJyMjg4EDB5KRkcHcuXPx8/NDU1NTWKSLi4sxNDRk5MiRXLp0iaKiIrKysv6vjL26ujq3bt0SGMz/l9f/5MH+6tWr3LlzB3V1dQ4cOEBTU5OYptjZ2ZGdnc3ff/+Njo4OLi4utLa2snPnTlxdXXn48CGRkZHIyckhLS2NlZUVnTt3Fqgub29vfHx8ePz4MbNnz8be3p7z588zceJERo0ahZ2dnbiF6+vrk5KSgqqqKtu2baO0tJTXr1/ToUMHjI2NUVJSYsqUKRw+fFjo5JOSksjKyuLOnTvk5uby6dMnXFxcqK6uxtHRkcbGRtavX8+sWbMYPXo0eXl53LhxgxMnToiybHV1NSNHjiQjI4OOHTuSm5vL6NGjhY757du3HD58mKqqKgYMGEBBQQFKSkqCid5O3hg5ciRycnICw9jY2EhjYyOjR48mLi5OFNn09PSwtbVFVlaWNWvWoKSkxIwZMygtLWX37t3U1dX9B6qpubmZI0eOUFRUxIULFwRj3NXVlU2bNtHY2IiFhYU4CPz8+ZP4+HhKS0uZPn06/fv3Jz09ndWrV2NqasqGDRtYtmwZXbt25enTp8C/NIZ79+4JXvX69evZs2cPp06dYvPmzQwaNAgNDQ3BIW/nzR8/fpy9e/dSUVEh8pDtLwkJCRwdHcWBXVFRkX79+uHt7U15eTn9+vVjz549Yso2Y8YMxo4dK1bmqamphISEiIPN+/fvycvLIygoiMGDB+Po6EiHDh2wtramW7duzJ49mylTplBRUSEQjRISEkhKSvLz50+cnJxYv349KSkpTJw4kePHjxMTE4OWlhaLFi3i48ePoh9iY2PD8+fPRTkqLS2NFStWoKmpSWhoKAkJCSgrK4uHzj///ENDQwNDhgxh5syZoiBmYWFBYWEhR48eZeXKlTx48ICnT58K5nf37t3Jy8sT8apnz56Jh1pwcDCPHz/GwMAAV1dXHj16hLKyMosWLRJEh9evX6OkpERwcDBNTU2CoqGoqCgmae0xmdbWVrZv305ERATm5uaUl5eLPKShoSFBQUHU1dWJaNqsWbPIy8tj+PDhjB07ltraWvr06YOOjg4vXrwgJSUFFRUV3NzcePz4sUAQ3r59G4AXL14gIyPD2rVrBa3g169f4j1eX1/PyZMn0dPTIysrCyUlJWJiYlBWVhZZz/fv3xMTE8PDhw8JCAhg6dKlAivZzk5uf82cOZPExEQiIiIAOHnyJDt27EBCQgJXV1du376NhIQECgoKNDU1oaWlRXp6OhISEqipqVFXV4e+vr7IQmtrawtOeDsm08DAQBA42jGRhoaGREVFISUlJXC/RkZGQpLT0tJCW1ubyMLDv9Gm5uZmOnXqRHp6OgCqqqrU19ejpqZGVlaW+BnaSR+5ubnAv9uP8vJy8axpbm6mpKREULYcHR3Ztm0bCQkJ7Nq1i379+tHU1ISCggIrV67k4cOH3Lx5E0lJSaKjo5kwYQImJibk5ORw6NAhamtrycjIoKamhlWrVgnk35MnT1BWVkZWVpZx48aJbZm5uTlRUVFs3ryZDx8+UFFRwd27d0XeeMuWLbS1tdHQ0ICqqir29vZISkqyfv160b84fvw4enp6PHjwgCVLllBfX4+ioiJBQUG4uLgI8tCtW7cwNjZGQUFB9L9iYmLo1KkTDQ0NHDhwgBcvXlBUVCTIIrm5uURGRjJ+/HgCAgKwsbGhvLyc2NhYAgICxPOr/cAlKSmJpqamIGLt3r2b1tZWlixZQnh4OJMmTaJXr158/vyZu3fvoq6uzpMnT3j8+DFjxozh8OHD3L17V4iELl68yPHjx3ny5Am+vr5oamoyefJkMYUOCwtjzJgxBAUFsXXrVrp3787y5cu5desW7u7uAt1ramrKokWLsLS0REdHhx49erBixQqGDRsmytPPnj0T7pfMzEw6duwotlzGxsa8fv1aPHeePXsmDnMaGhoMGTKEQ4cOoa+vz4EDBzAxMWH69OmEh4eLAZ66uroo5l+5coWCggJhfldRUcHY2Jjnz59z6tQppk+fjp+fn3Cd7Nq1i0uXLlFXV8fz589RUVERF46bN2/S1NQk3B6ZmZnU1dUhKSmJvr4+ra2tzJgxg5KSErZs2YKGhgaKioocPHgQOzs7YmJiWL58udjiSklJ8f37dz59+kR1dTUZGRn8/fffIhb04MEDevXqhZaWFnFxcfj7+3Po0CGGDRuGnJwcOjo6BAUF8ebNGzp06ACAlpYWW7ZsQV9fH1lZWfF3LScnh6SkJNnZ2Tx58oS2tjb69evHo0ePeP/+PRMmTKChoYH379+L6X9NTQ2DBw+mX79+xMfH4+rqyufPnzExMeHatWs8fvyYyspK/vrrL+rq6njw4AEGBgbAv9uGxMREPD09SUxMpLS0lLNnz3L8+HHOnTsn3BZLly7l27dvDBkyBFVVVebPn8+8efPo1q2b2ET16tWLw4cP4+LiQl5eHm/fvqWxsZGWlhaqqqro2LEjdnZ2REdH8/HjR9GT+fXrF1paWty8eZOTJ08KH0t7JLX9eXbz5k3evXsnsJ+HDh0S392qqqoUFRVhaWnJjh07uHPnDps3byYuLg4VFRW6dOmCuro6WlpalJWVsWbNGiwsLBg2bBiysrIiCmVtbc2IESPYsmUL9+7dE12E/5fX/+TBfu/evaxcuZJx48ZRWlrKtm3bSE1NFbdsaWlpfH19aWtrQ1dXl9zcXPEF3M7ynjt3LhoaGqSmpnL48GEaGxt5/vw51dXV/PXXX8LElpKSgrq6umiVv337VuQqv3z5wsqVK/H19eXMmTP06tWLvn370tLSwtSpUwUJZM6cOYJKsWDBArGaW7hwIWFhYQCMHDmSs2fPsmnTJt6+fUtGRgazZ8/m1atXDB8+nG3btrFo0SI8PT3x8PAgIyODAwcO0L17d6Kionj16hUyMjLk5eWxcOFClJWV6d27tyBuVFdXEx4eTm1tLadOneLHjx9MnToVU1NTfv36hYeHB3V1dbi5uTFnzhz69+/Pw4cPSU9PZ9++fRw+fFhMJ9szwufPn+fPnz9s27aNjRs34urqKv5sDw8PBg4cyPfv3/nnn394/vw51tbWrFixgoaGBmpqaujRowe1tbX4+PjQ0NCAjIwMa9asISQkRNz8Dx48yPfv38nJyWHQoEEkJydjbGxMSUkJo0aNYuvWrWzfvp0OHTrg5OQEIOgA/+chJDc3l3PnzrF69Wq0tbVZvnw5EhISzJw5k9OnTwvSw6xZs9i2bRv5+fmkpqYyY8YMCgsLGT9+PI8fP2bXrl1kZ2cLdGpRUZHQgc+aNUtEjI4dOyaKRO0s/xMnTvDz50+eP39OVFQU8C8qrD2fGhkZyfr16zEzM8Pd3Z2OHTtSX19PSEgIampqPHz4EGVlZVauXCkmtmVlZWhra+Pv74+Pjw9XrlzB1NSU8+fPo6CgQHV1tbBUKikpERAQwJAhQ4RPoL6+HgcHB6ZNm0ZbWxve3t7iMHr37l127tzJ0aNHGTJkCAUFBaioqAi739SpU5GXl2fUqFE8f/6cwsJCNDU1MTU1FSt7Q0NDIScxMzOjR48efP/+XcTfvn37hpGRkegd3L17l9jYWOzt7WloaGDdunUCwff69WsGDRpEt27duHLlCvfv36dbt24EBASgoKDAmDFjUFRUxNLSkpcvX4pL6q1bt5gxYwYKCgrIyMgwbNgwQVX5+fMnkpKS9OrVi06dOnH//n06duzIs2fPBO/+w4cPODs7C+zjhg0bRElxxYoVODg4CDTeuHHjuHbtGsOHDwfgwYMHvH79mq9fv/L8+XMqKyuRkZFh6NCh9OnTh969e6OoqMjXr1+pr6/Hy8uLnTt30rVrVyGAMjExYc2aNejo6ODq6kphYSGdO3dm3bp1tLS0MGzYML59+4aVlRVr164lPz8fNzc3YZ1dtWoV8fHx9OzZEz8/P7p164a+vj5dunTB2tqaJUuW0KNHD3r27ImKigqmpqYsW7ZM8MubmpowNTVl9erVdO7cWWTYzczMWLduHdLS0owYMYIfP35gbm7O2rVrKS0tZdCgQfj6+mJtbc3q1asFDWPgwIFiANKpUyfxvh4zZgw2NjbY29tjZ2cnBiJPnjyhW7du/Pz5E4CtW7ciKyvLypUrxQV/3LhxdOvWjXfv3vHgwQMCAwPp06cPhw4dIjExkezsbB4+fIiPj4+gf8XGxnLz5k1GjhzJ/fv3RTTDyspKxBILCwvJy8vj48ePuLi4MHPmTK5cuUJbWxtZWVnIyMgQHBxMfn4+d+7cITIyklOnTuHv7y+6Nbdv30ZDQwMvLy/+/vtv4ZhISEhg3LhxZGZm8vz5c1RVVXn8+LGIZvTr14/MzEw8PDw4evQoe/fuZevWraxevZqVK1dSVFQkZDjR0dFERESwZs0a5OXl6d69O5aWlmzdupWoqCjh/Pjz5w/Tp0/nwYMHnD17lhkzZogJuIODAxcuXCAwMJDPnz8D8OPHDzp37kxTUxMbNmwQtKJjx45x48YNli1bRn5+PvPmzWPevHnCYt3O7peUlKS4uJiAgAA+ffpEUVERw4YN48aNG/j4+ODn50dAQAAhISFcuXKF3bt3M2vWLPz8/IiNjaVbt24MGzaM2NhYNm/eTFRUFNOmTWPr1q2cOnUKT09Pvn//zogRIygtLUVNTQ0JCQkh4Hr69Cmtra3U1dUxe/ZsXrx4wYIFCwgJCSEuLg4DAwPy8/NZsWIFampqvHv3joaGBgYNGoS0tLQwyGtqalJfX09KSoo4wLu6uuLv709ZWRlPnz5FQ0OD27dvM2rUKK5fv05GRobwRKioqHDz5k0xfdbT0yM8PJzFixfj6enJ2bNnxYCo3S67a9cuzp07J7oTAQEBjBs3DnV1dUJDQ8nIyKBfv34MHDiQ5cuX8+7dO6KiojAzM+PEiRMiztWOw26P+Dk7OzN06FDu3LlDTk4OoaGh2NraMnv2bN6/f4+cnBweHh7ExcUJ+3t7Dt7X11dcNCZOnIijoyMnT54UVJ+WlhZWrlzJjRs3ePPmDSdPniQ1NZWjR49SV1dHY2Mjs2fPRldXFwkJCeLi4tizZw+3b9/m+PHjJCQkUFtbi6GhIVJSUly8eJGwsDC6detGbW0tsbGxnDp1imnTpvHp0yciIyPp27cv06dPR05ODnV1dczNzenQoYN4pgNMnToVDQ0NmpqaiIuLo2/fvuTl5SEpKcnkyZNpa2tj/vz5SEpKUlpaiqSkJCEhIaxevZrCwkIePnxIz549KSoqEvbiefPmoaGhgY2NjUBdduzYkYsXL7J7924UFRUxNDRk0KBBfP78mdTUVCIiIpg/fz5r164F/qUkGhgYEBYWRmBg4H9NxvmfPNi7u7szZ84cfv36JVbogwYN4vHjx2zatIm///6bvXv3snnzZmRkZMjIyODHjx+MHj0aRUVFWlpaxHS0uLgYbW1tFixYwObNm1FTU2PNmjWkpKRw8uRJtm3bxpkzZ5CRkWHw4MEEBwfTq1cvgoODkZCQECXDCxcu0NLSwrVr15CUlOT27duYm5sTExODi4sLb968wdTUFFdXVwICApCUlOTSpUu0tLTw+/dvTp06JW77v379ElNBMzMzoqOjaWxsJC8vjxMnTpCamkpWVhZPnjzB2tqagIAAli1bxsWLFwV/WFlZmZ8/f1JbWytutaampvz+/VvkYydOnMisWbOwsbGhU6dOvH//nri4OJYsWUJERASjR4/m9OnT7Nixg4EDB+Lu7s60adOYOnUqBQUFDBs2jMrKSkaNGsXevXuJiopiw4YNAqtnb2/P48ePWbNmjeCER0REUFhYKCIk+/btQ0VFhUmTJlFfX8+hQ4cwNjYWB+LRo0fTrVs3HB0d6d27tygPFhYW8uvXLzEJabeZqqqqUlBQwJMnT3j//j36+vqMGTMGIyMjbG1tuX//PhISEjQ3N+Pv709bWxunTp3i58+f+Pv7C8vrgAEDqK6uFpnvb9++MX36dGxsbKisrKSlpYWamhoeP34s5DqPHz/Gzs6OBw8e8PDhQ1avXo23tzepqal07tyZiIgIbt26RXFxsZCRZGVlceHCBcLCwjh37hxtbW107tyZ/fv3Y29vj4uLCxs3bkRHRwd3d3fu37/Pnz9/UFFR4c+fP1y9ehU/Pz8cHBz4/fs31dXVxMfHk5yczLNnz1i7di0HDhzg6dOn3L17l1+/frFjxw727NnDnz9/yMrKEmbV9i+89gnwtWvX+Ouvv5gzZw7Tp0/H19eX7OxsYTF2c3OjqqqKa9euMXLkSM6dO4enpyfx8fGkpaWhoqLCx48f6d69OwUFBUIUZGNjw8WLF3FwcEBdXZ3nz59z//59IWwxNjYWQrRFixbR0tLCs2fPmDRpEo6OjkyZMoU3b96IzkpOTg4aGhqYmZnx5s0bJCQk2Lp1K9XV1UhISLB27Vpu3LghvvSzs7MJCQkhJiaG3r174+Pjw4ABAwgNDeXSpUt4eHgwevRoevfujZGREd7e3uTm5pKQkEBbWxsZGRmEhoby8eNHsrKyMDc3JyAgAC8vL65cuUJmZiZXrlwR5srJkycjLy9PQUEBiYmJ/Pnzhy1btrBjxw7y8vLYsWMHffv25fjx40LI1j6pDQ0Npbi4mMzMTCorK3F0dOTdu3eCEJKfn09kZCTv3r3j169f1NXVCQzru3fvSE5ORlJSUsTvPnz4QExMDKqqqsTFxVFcXExUVBRhYWHo6+sTERFBRUUFiYmJPH/+HCsrKz58+EBJSQmpqan8+fOHbt26iW1Ufn4+RUVFhIeHCyttTU0NmZmZfP36VfwMbW1t/Pjxg8zMTH7+/Im+vj4NDQ0YGBiwfPlyZsyYQceOHTl06BATJ04UfoZ2uWBqaira2trs37+fHj168OzZMz59+kRpaSmJiYmYmJjQuXNnHj58yNChQ5k6dSpLly5FUlKSKVOmMGTIECIjI8nKymLAgAFoaWnRq1cv3r59i52dHbm5uWhqaooNr5ycHFFRUWzfvh0PDw/WrVtHly5d6N+/P/n5+Vy+fJmNGzcKzOOHDx/EVi86OlpMcMPDwxk2bBiJiYns379fUFIkJCRQV1cnODiY8PBwQQlLT0/HyspKxGaUlJRQUFDAwcGBpKQkTExMuHjxIn///TfFxcViCjhu3DgaGxvp378/BQUFREREYGNjg7GxsSghampq8vXrVy5evIiFhQX//POPwBMeO3aMd+/e0bFjR6SlpYmLi+Po0aP4+/vj6+vLypUrGT9+PKqqqixbtgw3NzeWLl2Kp6cnsbGxvHv3DgUFBWxsbPj8+TPGxsbExcUJZPGVK1cwMDBg3LhxvHz5krKyMj59+kRaWprouxw+fJjm5mZ0dXWRlJQUEr23b9/i7u4utvJaWlpERkbi4ODAvn37RAm/vSje2NiIgoICffr0oampiZaWFvr06UO/fv0IDg5myJAhfPnyhcjISPz9/fH29gb+3Ur9+PGD69evc+7cOVRUVLh9+zZ5eXmMHz9eoISbm5uJjIxEVlYWZ2dnCgoKRG9r06ZNAGzevJmpU6fi7OzMly9f0NfX5/r16xw6dEjEErt3786ZM2eYOnUqnTt3ZuHChcyfP5/g4GDc3d0pLy9n+/btmJqaEh8fT1ZWFsePH8fLy4vg4GAGDBiAmpqasIYPHz5cRF5ycnJwcnKirKyMyMhIIiMjcXZ2JiAgQKAu582bR2trK4cOHaJHjx7ifd/uP7GysuLWrVvcu3cPQ0NDZGVl6du3L9+/f2f69OkoKChgYGDAhg0b8Pb2ZtSoURQUFBAWFkZMTAweHh6EhIRga2tLfn6+2EYrKCiwdetWOnfujKSkJOXl5Tx48IAdO3YIP1H//v35+vUrsrKy+Pv7c+bMGQYMGEBqaipHjhwhIyODr1+/8vv3b7EJNDIyoqysTMRRc3JymDt3LpcvXxZCtHaTfbuLZPDgwXh5eREaGkpgYCBtbW1YW1sTHR3N06dPeffuHbm5uTQ1NQnYSXh4OPfu3SM/P18w/4OCgpCRkeHJkyckJydjYGDAo0ePcHFxITAwEFNTU75+/YqLiwurVq0iLy9PIIs1NDT+n8/A/1MH+/YsHCBoHVlZWXTt2pXTp08DICMjw8iRI7lw4QJ1dXW0trYyYcIE9PT06N+/P0uWLCEzM5MnT54wefJkrl69KtZjFy5cEKSQqqoqIdxQUFBgzZo1vHjxAk1NTWpra/n27RtRUVGUlJTQu3dvnJ2defr0KceOHRM6aT8/P1auXMmaNWvo37+/OGCvXLkSIyMjwsPDmT17NgYGBjQ0NIiHka2tLTU1NVy8eJHKykr+/vtvpkyZItBtra2tWFlZoaurS0lJCaNHj8bIyEiYc/v06cOwYcOwt7fny5cvQjd/69YtkpKSsLOzQ0tLi8TERO7duyekK8uWLWPixIk8evSIhoYG8cWipqaGo6MjgYGBQsHc1NSEtrY2EyZMEAfqwMBASktLef78Oc3NzaSlpeHm5sa+ffsYO3Ystra2ZGRkMH/+fFJTU3n//j3+/v7CBJeTk8M///wjqCBfvnxh1apVrF69mosXL/Ls2TNyc3MxMjIiICBAKMGDgoL48uUL8fHxmJiYICEhQWxsLEFBQezdu5d79+5x6dIlxo8fLzCJN27cEDZcCwsLkpOT+fnzp4idODk5sXjxYiZMmIC2tjY+Pj4sXLiQzMxMqqqqhDyqPefd7hKwtbUVxtX2B01ERARWVla4uLhw/fp1zpw5Q1NTE7Nnz6akpISPHz+ye/du1q9fz5IlS1i8eDGzZ8+mra0NJycncnNz0dXVxcfHh0+fPtHY2MjQoUNpbGxk+PDhouCbkJCAtbU1NjY2ODg4YGtri6WlJcuXL0dWVpZJkyYxdOhQ5OTkyM/PJzExkTt37rBp0yaGDBmCg4MDv3794sWLF1y7do3Lly8TGxvLnDlz+PLlC0ePHqWgoEAg2168eEFMTAxXr17l6NGjWFtbExERwYkTJ4iMjGTFihWUl5czdOhQ1qxZw7x587h8+bLwNrS/14OCgsTFFiAuLo4tW7YI4oC0tDSSkpIcP34cRUVF4uLiaGhoEJSRxsZG8vPzhcCooqICDQ0NIXzS0dFh8uTJnDhxAkNDQ2xsbNDS0mLSpEls2LBBkKqKiop4+fIls2bNYs+ePbS1tfH161fs7e2prKxk0aJF2NnZsW/fPmpqasQhyM3Njf79+9OxY0dBZVFXV+fu3bv8888/4uHdblZ98+YNMjIyWFpa8vjxY5qamkQn5MuXL1haWvLq1SvRl7l79y6urq7cvHmTP3/+UFpayqNHjxg5ciSbN28W9ttnz54xY8YMFi5cyIABA4Rlc8WKFYwaNYpZs2ZhZGREeno6e/bsoVevXmzZsgU1NTW+fPlCYGAgZmZmHDlyBDMzM8H/lpeX5+rVqzg6OnL//n0KCwtFvG7IkCHs3bsXNTU1fv36xdOnT5kyZQp//fUXPXv2ZNCgQbx69Yq//vqL8ePHC9pP3759gX/RufPnzyciIoI7d+4wYsQIzp49KzYQkydPZtWqVezZs4dVq1ZRW1tLRUWF+P1NTU2sWLECPz8/1q1bh7a2NiUlJcTFxdHS0oK3tzfFxcXMmDGDwYMHs2/fPqKiopCRkUFSUpIrV67g4ODAyZMn0dbWJiEhgeLiYgBWrVpFbGwsGhoarFu3DmNjY3R0dDh58iQjR47k69evuLq6MnDgQFxdXenTpw+mpqaUlpYyf/58NmzYgL+/PwsXLsTS0pLw8HDc3NxoaWmhsbGRyspKvn//jpSUFGPGjGHy5MmkpKSwdetWqqqqSE9Px9bWVuB7O3TowLRp00TZddWqVTQ1NbFjxw72799P//79mTNnDlOmTOHdu3ci2jFp0iTKysrw9PRkwYIFSEhIoKKiwrlz50Qvyt7enlevXjF79mwqKip48eIFYWFhmJubk5WVhaOjI4qKihw7doyuXbvy4cMHcahycnISYsIbN25gaGiIu7u7wM/26dOHNWvWMG7cOAYOHEivXr3YtWsXOTk5qKurY2FhgZeXF0lJSQwfPhxbW1vCw8NRU1Nj/PjxaGpqMnv2bDZs2IC9vT2urq6cPXuWvn374uHhwZgxY3Bzc0NJSQkvLy+Ki4sZM2aMsGcvXryY/fv3c+bMGVHA3r17N8rKygQFBdHY2CioWe3F1MWLF3P69Gn++usvNm7cSEJCAps2bWLQoEHcvHkTbW1t4uPjGTx4MLm5ubS0tNDa2krv3r359OkTQ4cOpaioiEGDBtGnTx8A9u/fz+HDhwVOddu2bbi4uPDp0ydevnxJW1sbnz59okOHDkJWdvToUXr27ElLSwv//PMPkydPxtTUlA4dOmBmZsbVq1cpKCigqamJ27dvC4oPQFpaGoaGhvTs2RNHR0cKCwvR09PjypUr9OjRg+DgYA4cOMDBgweZNGkSJiYm2NjYEBYWhoqKCq6urhw8eJAZM2Zw48YNqqqqxLmhoaGBp0+f4ujoiJqaGo2NjRQUFPDlyxfu37+Pg4MD9vb27N27V2Ce2y3PW7Zsobi4mPPnz4tNS3l5OQoKCgDU1tYCCPLahw8fkJKSEpfl5cuXCwHV9+/fOXnyJM3NzXh5ebF8+XK6d++OjY0NGzZsIDo6mvDwcHFo/+uvv0RRODQ0lDNnzrBs2TLevHnDu3fvhDfj69evvH//npaWFgICAlixYgWOjo5ERUXR2tqKlpYWEydOJCYmhpMnTyInJ8fRo0f58eMHBw8eZMiQIfj5+TFr1izq6+sJCgrC2NiY69evCywwwIoVK3BxcQEQn83/5vU/dbC3s7MTDPv2iZylpaVAVfn7+1NTU0NlZSUNDQ2sXbuW7Oxszp07x6xZszhz5gzFxcWCDayhoUFCQgLDhg0TD92EhARu374tdNTtpsTKykqkpKQICwtDTk6Oq1evMmTIEFxcXDAwMGDFihUoKytz4MABsrOzyc3Npbq6mpMnT7J7926kpaUJCwvDzc2N+/fvU1NTg4ODA8eOHaNz585iiq2trY2NjY2Yug4fPpzu3bvTpUsXkZ0DRPbP3t4efX19XFxc2LNnD0OHDmXcuHGUlJSwd+9eTExM+P37N8uXL8fe3h4ZGRkmTpwoRCEJCQkYGRmxYcMGMR3T0NBg/vz5ODs7s2XLFlasWCGa6bm5uXTs2BFbW1tiY2MpKipCWlqa1tZWVq1aRZ8+fURJuD1mpKKigq2trZjkxsfHIy8vT21tLT169MDLy4sVK1Zga2tLdXU1/fv3F1/gkydPFiKgcePG4enpSWFhIb6+vnh6egqObktLC7W1tTQ3NzN+/Hiio6O5desWnTp1QkVFhd69e4vtiLKyMvPmzUNZWZlx48YRGBgoHpr+/v6EhobSqVMnEhIS6NOnDw0NDRw9epSamhra2trQ0dHBwsKCadOmISsry7t372hpaUFWVpYtW7YgISHB0qVL2b59Ozt37qRjx474+fnRu3dvZGVlOXXqFLNmzRKX1EOHDgn2tJycnHjAdejQgSlTprB582bq6uo4evSoMH8aGhpy5MgRxowZw/nz58V/g7+/P6dPnxbWwJEjRzJjxgxSU1P5/v27yEK/e/cOLy8vrl27RlxcHAcOHGDYsGHs27eP69evU1VVxY0bN+jbty8xMTE0Njayf/9+oStfvXo1jx8/Jjk5mZycHMzMzMjPz8fa2prjx4+LHLyRkRFnzpzB2tqa27dvo6OjQ3p6OlJSUty6dUsUtfz8/DA0NERLS4vLly/j4ODA0aNH2bVrlzigf/v2TWBSnz17JgQ8vXr14vTp06irq4t4VHvOUVJSEg0NDezs7Ni5cydnzpxhxowZTJ48WRzqVFVVSU9PZ86cOdy8eZMHDx4gISHBoEGD2LlzJz4+PkJXX1tbS2JiIidOnCAwMJDOnTvj4uKCi4sLV69eRU5OTmAF1dTUmDhxIteuXSMyMhJFRUV69OhBeHg45ubmwuh58uRJfH19effuHceOHROl6nbb4f79+8WktN0AvX37dmRkZNi4cSNjx46lsrJSRHN27drF79+/kZeXF5nXf/75h8bGRnR0dJg3bx4KCgoi4mVmZsaMGTMwNDTk4cOHmJqaoqenh6+vLw4ODoSGhnLw4EE0NDRYvHgx7u7ufPv2jV27diEvL8+aNWsYNWoUJSUlbNiwgba2NrZv387UqVNpa2tjxYoV/Pnzh2PHjvHnzx8MDQ1RUFCgoqICBwcHbt++zaBBgwThwsvLi5aWFvG5GDp0KDt37sTFxYWlS5cyZswYzp49i5OTE+bm5rx8+RJra2skJSXF9jErK4vo6Gh69+4N/DsUev36Nba2toL0snDhQu7evcvdu3eJjIykublZbGmysrJwdnYmIiICAwMDOnfuzPHjx+nZsyfr16+npaWF69ev8+XLF8zNzQWd7fDhw2zcuJEuXbpQW1vLgwcP6N69O42NjURHR1NRUYGOjg7S0tJISEgI0aKxsbGIsjg6OlJfX4+Wlha1tbU0NTVRV1dHSkoKGhoaeHp6oq6uTkFBAQoKCsJg2bt3b6Kionj27BkAd+7cwcPDg9evX+Pk5ERERAQfP36kqqoKS0tLIiIiCAkJIT09nerqai5dukSHDh3Q1NQkMjKS2NhYZGRksLCwQE9Pj9u3b6OmpiayzM7OzpSWlvLnzx+Ki4uJiIggKSkJMzMzwsLCcHZ25syZM+Tl5VFYWEhubi4+Pj5MnjyZsrIyunTpQn19vegbjRw5EgcHB4F19fLyQl9fnzNnztDY2PgfXoLMzExyc3NJT08nKiqKurq6/zD6Pnr0CAMDA/Ly8jA2NsbFxUX0ZNrJX0ZGRri4uKCqqkpLS4vAsGpra/PhwwdSU1Npbm7G0NBQ9PXS09Px8fHhwYMHGBoa8vnzZ37//i3+TuLi4ujTpw8BAQHo6uqKTlRubi41NTWMGTOGGzduUF1dTXBwMBcuXMDU1FQ889sHloaGhlhZWfHnzx/Cw8PR0NDgx48faGlp8fLlS3r27ElqaqpwfAwbNgx/f38SEhKwtLTk/fv3GBsbs2LFCurr69mwYQNPnz7l/PnzZGRkUFtby9ChQ3n79i1VVVU8fvyYzMxMEeesq6sTnYK8vDxR2Le2tub58+c0NDQQGBjIxo0bGTVqFEuXLkVTU5OxY8fSuXNn4uPjBcpRX1+fgQMHkpaWJr7rMzIyKC8vp6ysDD09PZSVlbGwsMDPz4958+bh5OTE33//LaJFo0ePpq6ujurqanbt2kVLS4uQAP748YOEhARUVVX5/fs3hw4dQl5enkmTJok+ZV5eHpmZmSgrK4tY7vXr12loaMDX15epU6fy+/dvVq9eLaKVO3bsoKamhtOnT7Nw4UKePn2Kmpoa3t7ePH/+nFevXuHg4MCyZcv4/PkzjY2NbN++nfj4eBQUFDh9+jSKiorie6pLly5s2LABd3d3HBwc2L17N/3796ekpEQ8k/7b1//Uwb5Hjx4A/PXXX2zbto0/f/7g7OxM165dCQ8PZ9q0ady8eZOGhga+f/9OUFAQCxcuJDw8XDBljx8/zsmTJ1m5ciU+Pj4oKytz69Ytxo0bR79+/ejevTv6+vpkZmaira0tGMIVFRV8/vyZEydOiAcyQL9+/Xj//j27du0Sb5CioiKxbtPR0eHr1694eHhw69YtevXqxZQpUxg3bhwTJkzgypUr7N27l7S0NJ4/f87y5ct59uwZycnJDBs2jK1bt3L58mXKysrYsWMHo0aNYtiwYbi7u2NkZMSyZcuIiYkhJyeHmpoali1bxtOnT0UR8NGjR6xdu5aXL19iampKeno6srKy/Pnzh6ioKGGyjI2NZcuWLTx69IiWlhZiY2MZPXo0mzZtwtLSkoKCAkaMGMGqVatYsmQJlpaWPHnyROBEnz9/zsiRIxk6dKiIyURFRWFra4urqyt6enqcOnWKffv2ERwczL1793j27BlTpkzh8ePH+Pj48OPHD2bPnk1NTQ2bN2/GyspKHLp+/PhBYGCgQJlNmzYNV1dXOnToQEJCguA+AygpKWFsbExpaSnBwcGYmJhw//597O3tGTFiBDNmzGDevHnY29sTFxeHubk59+7dQ0ZGhvnz5xMfHy8swCdOnCAlJUU8kNtjQs3NzWRnZ3PixAkGDRrEmDFjOHDgAGPHjuXBgwf4+vqirq5OdHQ0UlJSqKuri/x2XV0dJ06cEFSFvn374uTkxNSpU8VBq700uGDBAlxdXbGzsxNTnq5du3L79m0xKVu7di3dunVDXV2drl27EhcXx5MnT6ivr2f8+PGi1G1nZ4eZmZlg8I4dO5bq6mr8/f05evQoOTk5LFq0iKamJoGVrK6uZujQoSgoKJCTk0NBQQFFRUVi2hQVFUVLSwv6+vpoampy6NAhpKSkyMvLIz09nd+/fws0nI2NDd7e3mRlZeHg4IC0tDQtLS3MmDFDRNfaYwrr1q3j/fv3zJs3j3v37hEeHo62tjZjxoxBV1cXKysrHB0dxSFkzpw57N27l/79+1NVVYWTkxNVVVWCFtUuI3rw4AFGRkYMHjyYpKQkPDw82LhxI1lZWfTr1w83Nzf+/vtvcnNzuX//Phs2bODWrVucO3cOW1tbnJychJGxvaTZ3tFp/zlmz57NmTNn8PPzo6SkhDNnzohiob6+Ps7OzkyePBkZGRnKy8txcHDA2dlZXKZVVFRQV1dHRkYGY2NjrK2t0dHREfxxPT09LC0thfRGXV1dSISKiooYMWIEurq6rF69GhMTEyoqKnBzc8PY2BhlZWUMDAyoq6ujR48eWFpaChRdS0sLNjY2dO3aFSUlJUFKMTU1FVsOFRUVFBQUhFPA2NgYWVlZURozNzenoKAAHx8fOnXqhLq6OqamppSVlTFw4EDRuTA0NOT169fcvHmTKVOmkJCQQEVFBT4+PhgYGPDz50+6d+9OSkoKO3bsYPjw4aSkpNCjRw/69OnD69ev6dWrF/v372flypWUlZWJQ5GCggLjxo0jKioKNTU1NmzYQJcuXcTmoR2ZOH78eBITE/Hz82PBggW4ubkBMHjwYPLz8wkMDMTW1pY3b94gLy+Ph4cHI0aM4N69eyxatAiA169fExQUJDK9L1++FLHI4uJiVFRUKCwsJCQkhHfv3rFx40aRBW6XGx47dkxQewYOHMjq1asZN24cRUVFREdH4+zszJUrV7h58yYLFiwgLS2Nmpoabt++jZaWluiehISEMG7cOO7evYuVlRXp6emkpaUxdOhQkR2ura1FTU0NOTk5Fi1axK1bt8T7ISoqikePHqGoqIiTkxONjY14e3tz8OBBBg0axJUrV4iMjOTBgwcAHD9+nB49eoi8dLsR1dHRkWnTpomNdXV1Nfr6+pSWllJYWEhFRQWLFi3i3r174nNeWFjItGnTmDZtGmFhYcJJ06VLF6SkpJCTk8PV1RVlZWU+fPjAgAEDBAll0aJFqKiosHDhQoyNjcnJyUFSUpKPHz9SW1vLnTt3+Ouvv1BTU0NXVxdLS0taWloICgoiMzOTVatWUVJSwtq1a5GRkSElJQU3NzeUlZUZO3Ys+vr6TJs2jW7dugl6zZ8/f7C0tGTZsmVCUCkhIYGnpyebNm3ixo0bLFiwgBcvXmBubk5cXBzOzs5IS0vz8OFDLly4QElJCV5eXowePRolJSV+/vzJtWvXmDZtGk5OTtTV1SEvL8+DBw/Q1NQE/h3oSUlJ0dbWxtu3b9m7dy8NDQ1cuHCBKVOmoKyszLNnz1i9ejXZ2dm8efPmPwzG48eP5+HDh8yZM4clS5bw8eNHFi9eTMeOHfny5Qvy8vIUFxezcOFCbty4wbt37xg4cCDv3r0jODiYtrY2zp8/z9GjRwWpp51J39zcjJOTE126dMHb21t0AdasWYOioiLfvn3Dx8eHd+/esW/fPrHJbZezbd68mYKCAiIjI5k/fz6ysrIoKCjw/v17goODaWlpwczMjJaWFo4ePcqUKVOwtrZm+PDhrF+/ntDQUIYMGSKwmjk5OVy7do2ysjIePHjA79+/GTlyJAcPHsTT0xMJCQmBp2w/fNfX19OnTx+GDx9OSUkJeXl5eHh4oKamRnZ2Nps2bRIUGyUlJTp16iTiv8ePH8fc3JzXr1+TlJTE3Llz8fX1BRBizaioKJ4/f05eXp7YcJSWlqKuri4m9//N639SUJWens6CBQsoLS1FR0eHkJAQ1q5dS1RUFNXV1axdu5bNmzdTVlYmREseHh4kJycLvrSNjQ2pqalYWlqSm5uLlJQUf/31F6tWraKmpobDhw+zfv16LCwsyM3NpVOnToJF/fDhQ96+fYu5uTna2trk5eWJouzdu3dZsmQJ8vLyjBgxgqysLF68eCHU1qNHjxaFvUGDBrF//34WLlxIREQEkpKSTJs2jejoaOLi4ujatSteXl7iIGJiYoKamhrJyckiJzh27Fi6dOnCnz9/2LBhAz179mTu3LnY2NgwcuRILl++zN69e6mtraW8vBwNDQ1aWlrw9/dn/PjxVFZWoqury+nTp4mKiiIiIoLc3FwkJSXx9PTk7du3tLa2Ymtry5AhQ6ipqSEjI4MPHz5QW1uLiooKysrKSEpKkpmZiYWFhZj8q6qqUlxczLt37/Dw8GD27Nn8/PlTyMTKy8tJSUnB2NiYuro6MjIyhFb9169fmJqaUlBQwPnz59m8ebPI6LeLcdoPP79+/cLKykpEkdq5x3JyclRUVDB48GAOHDiAl5cX9vb2YsrSjsWzsLDAwcEBU1NTrK2tefToEW1tbWKaf+TIEaKjo/n+/TtxcXFcuHBBUGaOHTtGbm6uwDqmpaXR2trKx48fGTJkCG1tbXTq1IkXL16QmpqKmZkZNTU1aGtrIy0tzZUrV5g+fTr79++na9eu9O/fnxEjRnDmzBl8fX3ZuHEjGRkZdO3alYkTJ+Ln54e9vT3GxsacPXuWmTNn0r9/f0JDQwkODiYsLIxHjx6xcuVKwsPDefHiBbm5ufj5+fHo0SN69uxJTEwM06dPR15enpycHDp37sz9+/dRVFQUedFt27bR1tbGvHnzSE5OJjMzk8WLFxMUFMSCBQuYNGkS8+fPJzs7m/LycpYvX87+/fvJyMjAy8uLxMREzMzMxFpz27ZtNDc306dPH8aNGyemge1fCO3vjXZBXEpKCvX19QwePJiWlhZxoJWWliY5OZnRo0cjJyfH5cuX8fHx4cuXL/z48YOWlhYkJSXp3r07+fn5uLu78/XrVzp16sSGDRvYsGEDX79+pa2tjba2NuTl5fnz5w/Tpk0TD+Xfv39ja2tLeno6FhYWxMXFCStrWloa8+fP5+vXr1hbW4s89/fv31mxYgWXLl1i1qxZgs708uVL8vLyGDhwIAkJCQwaNIjQ0FAKCgqExK579+5ERkZSXFxM//79BbNeSkqKiooKYS2Ef7df7Z/jqqoqgVktLS2lY8eOYsKroaFBaWkpSkpKNDU1UV9fL2Iq8vLyolDY/msyMjJISUlRXV2Njo6OUKDLyclRWVlJp06dRKlMUVGRiooKNDU1qaiooK2tDVVVVcrKyoSEpbm5GXV1dUpLS1FRUaGuro6mpiY0NTWprKzEwsKCpqYmdHR0cHR0xNnZmUuXLqGtrY2LiwvHjh3DyMiIfv36ce/ePdzc3JCUlOT06dMCVjBkyBCkpaV5+fIl/v7+jB49mlmzZnHz5k0+fvxIQECAkDUZGhqSmZnJ8OHD0dfXR0VFBTk5Ofz9/TExMaGhoYFfv35RX1+Prq4uTU1NPHr0iL59+/LgwQOWL1/O69evMTExYdy4cTg6OvLz508SEhK4e/cuHh4e9OjRg9DQUKGSz8/PJy4uTmTYHR0d+fTpE66urvTo0UMUKZOTk0lLS6O5uZnCwkLRYXB0dCQrK4v6+nqGDx9OcnKywHN269ZN5PU7duxIUlISvr6+3Lhxg0OHDjF79mxB2rl8+TKGhoZ8+fKFyZMnc/PmTXR0dPj+/TsKCgq0trZSU1PDp0+fBLv86dOnTJ06le7du3P9+nVSUlIYMmQIycnJLFu2jB8/fnD37l2mTZuGvb09Dx484M6dOwwcOFBgS798+cK8efPYt28fkydP5uPHj1hZWREWFsbgwYNJTEzkypUrLFq0iJ49e3Lnzh0cHBz48uULRkZGXLlyBS8vL/78+SMM8mvWrCExMZGoqCiWLl3Kpk2b8PHxQUpKivv377Ns2TKuX7/OzJkz6dq1KxYWFjg6OpKSksLly5cJCQmhrq4ODQ0NFi5cSG1tLVVVVQQGBjJp0iR69OhBenq6iDxKSEjw4cMHpk+fzp49e7CxseHHjx/ExMQwadIkunXrxrdv35CSkqKgoEAMK5SVlamursbIyIjk5GTS09OxsbHByckJY2Nj7t27x7Jly7h8+bLoE7T7AJycnJCWliYnJ4evX7+KKBX8m1TYtGkTVlZWHDlyhOLiYvr06UNOTg49evQQW18rKyvi4uLYt28fFRUVZGVlCUdHSkoK8C/B6vHjx7i4uJCdnU1GRgadOnWioqICBQUFysvL2bdvHxcvXhRW2zdv3uDg4EBiYiJXr14VNutu3bqJuFp+fj7FxcU0NjZSUlLC+fPn+f37NzExMZw5c4b58+eTnJyMtLQ0JSUlqKur8/HjR4FWffHiBbNnz0ZCQoIBAwbwzz//iCRAbW0tjx8/Zvz48fz8+VN8ZnR0dIQDx8/Pj+/fv/Po0SOkpaUFqrutrU34M9rjy3V1dYJNv2DBAu7du4e6ujrfv3/H1NQUTU1NJCQkRGa/W7duxMfHCxZ9SUmJEHJqaWmxcOFCHB0d0dfXF3FcHx8fNm/ezJMnT/Dw8KC6uho1NTVBMPv58ycmJibU1dX912fg/6mJffvr+vXr9OvXj6ysLB4+fMiyZct4/fo13bt358OHD0ILXFRUxIcPH/D19aVDhw6MGjVKmOzMzMzIyMhgwoQJfP/+nTt37rBjxw6MjIzw9fWloqKC0aNHIyMjg56eHk5OTgQFBXHx4kWKioqYOnUqzc3NXL9+HTc3N9TV1ZGSkqKkpIRz584xatQo5s6dK8obV65cQVlZmb1791JZWUllZSUAjo6OhIWFMWHCBN69e0deXh4ZGRniRtnW1saAAQO4f/8+X79+JTU1ld69ezNw4EA+ffqEnJwcra2tHD9+XLxp7O3t6datGw8ePODRo0cUFBSwZ88e/Pz8SE9Pp3fv3iKHV1RUhIGBAZWVlSgpKbF9+3bKy8uxsLCgubmZuXPncvr0aWRkZHBxcaG2tlaIvB48eEBWVhZWVlYC8/bx40cWLFggYi7JycmYmprS1taGu7s7I0aMYMeOHdy9e1f8e96/f585c+bw9u1bEhISiI2N5cCBA6IU29zcjISEBPPnz2fOnDmC7DJmzBjU1dXZsGEDxsbG9OvXj9evX2NgYMCaNWvQ09MTeDhJSUkmTpxISEgImpqaglv8/v17sSaVkpIiMTGRtWvX8ujRI5YsWSImf+32u/HjxwP/ascXLVpESUkJ3t7ePHv2jO7du2NoaEhCQgIPHjxg06ZNtLa2ChNje9b46NGjNDc3M3XqVAwNDTE3N8fX15cRI0YIO2tmZibXrl2jra2Nu3fvsmHDBtTU1JgzZw46OjpoaWnh7e0tyAKrVq3ixo0bjBkzhtbWVgoKCsQ0oV0m5eLiwpYtWzhx4gRNTU24u7ujqqrKli1bWLZsGX5+fnTo0IGHDx+iqKiImZkZISEhoiD3/PlzZGVl2b59u+hnJCQkkJCQwKJFiygoKGDmzJncvXuXHTt2CM392bNnuXnzJlJSUjQ3N3Pt2jVMTU1FEdjd3Z1evXrRu3dvampqMDAwQEJCgoKCAiGPOnPmDG/evOHXr188fPiQFy9eCArRli1bOHXqFM3NzWzbto3o6Gjs7OyQkpLi06dPmJiYYGxsjLGxMatXryYgIEAcPDMyMoTt1tbWliNHjlBSUsKLFy84f/484eHh9OjRg7KyMpqamli/fj36+vpYW1uTlJTEkSNHyMzMJDAwkKKiIjw9PfH09ERWVpaBAwfi6OiIv78/YWFhIi7S2tqKiYkJd+/eRV5ensmTJ3Pt2jUGDBjA9OnTWbFiBefOnUNDQ4ORI0cK+ZenpyezZs3C1dWVtLQ0PDw8MDY2Zv/+/VhaWoopUUNDA/fu3UNHR4fIyEj27dtHbGwsnz9/Rk1NjWfPnhESEiLEUO0Svby8PDZv3kxOTg6dOnVi69atGBsbM3HiRHJycrCwsGDSpEmMGjUKNzc3MjMzRU9i8+bNdO3alR8/fjBlyhQkJCS4evUqhoaGxMTEsGnTJn79+iWslnPnzuXnz5+iS3DlyhW2b9+OoaEho0aNEvImHR0dbt68ibe3N8OGDRPCo/aSdjs2rz0WUVRUhISEBHfu3BGHlD9//pCQkICXlxenTp1CR0dHxBONjY3R19fH09OTXbt2iZKxu7s7e/bsoV+/fgQFBYlM7ezZs3Fzc+Pjx4/Exsbi5+fHlClTMDY2xsfHR3QIfv/+jZOTk8AnKisrk5iYyJ49e7h16xbfvn2jsrKSHTt2YGFhwZ07d3jx4gW6urpUV1ejra0tStRHjx6lrKwMd3d3goODiYyM5OfPn0ycOJGkpCRiY2ORl5cnJiaG6upqYdKcMmUK1dXVoqNja2uLt7c3qqqqorw8duxYbty4gba2Nl26dKGyspJDhw5RUVGBrq6u6Fvp6uqKaWlkZCQHDhygpqaGpKQk0tLSeP/+vbDJbt68GS0tLQwMDBg4cCD79u0jMTERAwMDpKWluXbtGvn5+fj7+9OhQwf8/f2ZM2cOEydOpK6ujm/fvqGpqcm1a9dYvHgxL168EPx6ExMTXFxcmDhxooiOvX37Fi0tLYYOHcrff/+Nqakpubm53L17FyMjI168eEFoaCheXl5s2bKFkpIS+vTpQ3h4OGvXruXJkyf8+vWLX79+ceHCBRoaGnBwcKCwsBBLS0v09PTEpa68vJzz58/z8OFDOnTogKmpKY8fP+avv/6isbGRy5cvc+TIEW7cuIGenp7wz+jq6qKjo8Phw4cpKyvj1q1brF27locPHyIlJcXgwYMJDAxESUmJlJQUZs6cKUqbI0eOFHGhiIgIunfvjp2dHUpKSvTp04ekpCSGDRvG8+fP+fTpE15eXpw9e5bRo0dz5MgRJCQkePr0Kfb29mRmZmJsbEzPnj3F9+X37985c+YM/fv3R1dXlxUrVoifadu2bQwePBhzc3OCgoLw9PREW1ubwsJCvL29qampYf78+UydOpXY2FhevXrF+vXrOXv2LEOHDuXmzZu8evUKSUlJ3rx5w7Fjx0SczNDQEICysjIRUystLcXX1xcfHx9ycnJ48OABmZmZxMTEsGjRIioqKvj777+FJ2bYsGHY2NiQlpaGp6cnHz9+FMPLYcOGoaamxtGjRwkKCsLU1JSBAwcSHx/PkydPUFdX59KlS7x8+RJFRUXs7e1FMX3WrFls2bJFEBEjIyPR09Pj8uXLLFmyhOvXr7N8+XJ27dolHDLtBf4vX76wbds2Ro4cKeKo1dXVtLa20qtXL3E5iomJoU+fPqxatUpQ1P7b1//kwd7T0xMjIyPU1NTo0qULGRkZLFiwQHzgQ0ND0dfXZ/LkyQwYMEDYC0+dOoWRkREODg5oa2tTXl5OQ0MD5eXlmJmZMWfOHOBf9nNsbKzglPr7+7Ns2TIePXrEP//8w+bNm4mOjkZTU5Nly5bR2tqKnZ0dLS0t9O3bl3v37tHQ0MDNmzcxNDRk2bJldO7cmdraWq5fv86bN28YMWIEgYGBAAwYMAAlJSUsLCy4evUq+vr63Lp1CwkJCVRVVdm1axd1dXV4eHgQHR1NaWkptra2gj+ro6MjGv0ZGRkcOnQIZWVlbt++zdy5c0lLSyM/P5+mpiYaGxu5ceMGgwYNIjc3F3V1dRYuXMjKlSs5ceIEu3fv5q+//uLq1askJSWRnZ2NjIwMXl5eLFy4kBcvXrBx40YCAwO5cuUKd+/eZc+ePVRWVjJlyhS6du1KY2MjY8aM4c2bN/j4+JCYmIiHhweLFy9GW1ubHz9+oKenh7a2tjAntpsBz507R2FhIQBWVlZISkrSs2dPNDU1BfrLwMCAHz9+8P37d3r16iXoJu2Fs7Fjx/L582d69eqFoaEh9fX1AAQGBtKzZ0927tzJ+PHj6dy5M9+/f2f58uWsXLmS/Px8XF1duXHjBnFxcWRmZrJ06VJ0dHT48uULMTExHDhwgPv37zNq1CiGDx+Oubk5ly5dQlZWlpSUFMaNG8ezZ8/Yt28fbW1tLFu2jH79+uHq6oqrqyt1dXXs2LEDJycn6uvrBVJRRkZGUFz++ecfUlNTmT59Or169eLy5ctER0eLh+vMmTP58OEDEhISXLhwARkZGcLDw/Hy8hIm458/f+Lo6MigQYOIj49n165dHDlyhDdv3hASEiJoE7a2tmhqanLkyBHKy8v58eMHFRUVArVnamoq3juDBg1CU1OT8PBwPn36hKysLOHh4Vy7dg17e3sRUXN0dOTKlSsoKiqKOJSkpKQwnrYXklevXi0oBI6Ojjx58oTg4GCio6P5/PkzhoaGfPjwgfnz51NZWcnx48cpKyvj2LFj+Pj4MGTIEE6cOEFiYiLjxo2jubmZ3bt3k5WVJboqPj4+Ar3m5eXFs2fPsLCw4OHDh0KKEhsbi4WFBWFhYdTW1uLp6cn+/fupqakhICBAbO9ycnLw8fHBycmJnj17EhQUhLe3N7t27SI8PBwZGRm6detGZGQkX758YcaMGZw6dQoDAwNxufX390dSUpKgoCAhpnFzc6NTp05ERUWJONLJkydRVVWlqamJyMhIKisrSUlJITIyksbGRj5//kxxcbEQ9TQ2NhIREUFeXh5NTU2Eh4eLX2unebTnQKOiovjx48d/UJ+io6MpKSmhvr6eyMhI6urqSExMpKioSPwMVVVV/Pz5U/w5ERERlJaWkpmZ+R8/QzsB4//8tczMTEpKSsTvjY+P59u3b3Tq1ImUlBSWLl2KjIwML1++FFNPd3d3GhoaSExMFOXBdohAuzxLSUkJDQ0NXr16JSg5q1atoqCggIqKCmpra5GTk8PQ0JD8/HxUVFSoqqrC2dmZX79+0aFDByIjI0lNTRWmz3Z+fXJyMs3NzSIO1765PHbsGC9evMDNzY1NmzYJvK6JiQlv3rxh4MCBtLa2EhkZyfbt21FRURFbqbNnzwoLb3thMTc3l/Hjx9OnTx8+fPgg6B1r1qyhZ8+eImLYHpP5/PkzgwYNorGxETs7OzQ0NESJt6GhAXd3dx4/fkyfPn0YO3YsgYGBhIeHc/XqVZYvXy5INAMGDKB3795IS0vT1NTEpEmThObe3Nycmzdv0tbWhoeHB3PmzEFKSoqdO3dSVVWFmpoajx49Yvv27YIod/ToUfr37090dDQuLi5YW1vj7OyMtbU1a9euZdu2bcTExJCYmIiioiI5OTlISEjg7e1Nnz59+Pz5M5s2bUJHRwdfX1+ampqora3F2dmZ8ePHExsby8ePH2ltbWXq1Kn07t2bYcOGYWZmxvTp00X3oKSkREQCg4KCsLKyErlpPz8/pKWlsbOzo3///owdO5adO3diYmLC0aNHmTdvHvn5+WzcuJHZs2cLgdORI0cwNTVl//79Ynru5+dHeyAiOTmZ7OxsZGVl0dTUxN3dHQkJCcLCwpg9ezaLFy9m4MCB7Nq1S9C5pkyZQnl5uTCmy8jIICMjQ1NTEzY2Nrx//57q6mpBwmtHThoYGODv74+1tTWWlpaoqqqKQvmcOXMwMjLi7NmztLa20tjYKLp47RLHjh070trayufPn5GSkqJ79+4EBwezY8cOKioquHr1Kjk5OdTX1/Pp0ydSU1MZP348MjIyvH37lsrKSnR0dMjPz2fGjBlcuHCBd+/eoampiZ+fH6GhoVhZWWFhYQH8i0xt/zdsd/esWrWKrKwsNmzYwOjRoxk6dCg9evQgLS1NGM7nzp3LggULiImJoaWlBWlpadTU1MjIyCArK4vFixejqKiIrq4unz9/Ft2tnTt30tTUxKhRo9DR0QH+dR8ZGRmhrKzMxIkTOX36NH/+/CElJYWqqiqUlJSYOHGisAu3i0WDg4MpKCjg/fv32NvbY2lpKYZFN2/eJD8/n8GDBzNixAgUFBSEhXfgwIFISUkRFxfHP//8Q1NTE3/+/BHxpylTprBz506ysrLQ19cXMID/9vU/d7A/c+aMKNfFxcXx9etXJk+ezOfPn9m2bRsmJiY4OTmJW5a8vDxfv35l3759mJmZISsry+fPn9m3bx/Dhg0TRrU1a9bw+/dvXr9+jZGREXl5eejq6nLkyBE6derE3r17yc7OJjk5mc2bN7Nlyxb+/PnD6NGjyczM5MWLF3Tt2hU5OTmWLl3KrVu3qK+vJz8/n1OnTlFUVMTo0aMpKCgQK9CSkhLq6upITU0lPT0dDw8Ptm/fDoCOjg7Pnz+nqqqKS5cuERMTg7OzM1lZWdjY2Ahc0v/X3nvGRZGt69sXOYOACAhKEhQDwUEQFMGEqJjHjDlnDGMaHcxZMeecRlHMCTGAIoKRJKMCiqKCSs5I6PeDv17vONE5Z/bZ+z+7rk/QVV1d3dVd9dRa93PfW7duRUNDgy5duvDp0yeCg4NxcHDAxMSE9PR0li9fLqZ7hg8fTmhoKHv37qW4uJhTp05RVVXFTz/9RI8ePQgODub58+ckJSVx/fp12rZtS05ODh4eHkI28OjRI2QymZi2rFmzJjNnzqS6uprJkyejpaWFn58fSkpKZGRkcOfOHS5evEhCQgJLlizB1dWVR48esXv3brZu3YqhoSHv3r1j9uzZJCQkMH78eDG1u2/fPmbMmEHNmjUZM2YMWlpatG7dmoqKCrS0tGjevDlNmjRh/vz5vH37VtxIfPjwAYD8/HyKi4tRUlLizp07Qj7Ru3dvEQkdERGBra0tK1euZObMmWzYsIGBAwcCn5Pu5EXHjRs30NfXp169eiK059mzZ1y6dIk7d+6gqakpPOwdHR1p3LgxS5cuRVtbm/Hjx1NYWEj37t2Bz85NRkZGGBsb880331BRUYG5uTn3798XTZtpaWkEBwfz6NEjdHR0SElJoVmzZqxatYpOnTqRlpZGRESEmAKUO+k4Ojry8eNHbG1thXWo3Nmjb9++tGjRgg0bNuDn50dubi7dunXjwoULjBw5kh07djB9+nR27dolJCrr168XWt+LFy/SsmVLdu/ezeXLl9myZQt37tyhTp06DB48mHr16rFv3z6aN2/O+fPnuXXrFsOGDWPv3r0iJdfe3h4nJyeUlJSoVasWtWrV4vDhw3h5eZGRkYGrqyvq6uosW7YMT09P9u3bR0REBJ6ensTFxYnZCfnodVJSErNnzxa6fbndW79+/USyoFyP7+Xlxfnz5wkNDRXTwc+fP2fVqlUEBARw6NAhBg0aROPGjXn27Bm9e/fG09NThMfUqFGD/Px8NDQ0KCsro6KigtzcXKKjo9HT0yMpKQklJSXGjBmDsbGxSPc9e/asSCw2MjISqYYPHz7kyJEjtGjRgnnz5oliTt4gVlJSwpkzZzh37hxr1qwhOzubt2/fcv78ea5du8b8+fNFM++lS5eIiIhg3LhxKCkpcfv2bS5fvizkf7q6uly5coXz588TExNDixYt6N27t9i3+Ph4rKysOHDgAKdPn+bgwYMkJyeL0f+zZ8+yefNmMjIyxM1aaGgoixYtorCwkKdPn4p9mDx5MjKZjLt373Lx4kUiIyMZOHAgmpqaXL16VVg8mpmZoaWlxZ07d7Czs2Po0KEoKiqipqZGWlqacPCxtLSkfv36JCQk8PjxY06dOoWRkRGNGzemffv2fP/99xw6dAgnJyeUlZXx9vZGSUmJu3fv0rFjRw4dOkTDhg2prKxEUVERBQUFMbott2KNiIjA29tb6LO7d+9OXFwc+fn5nD59mnfv3mFtbS0GEgwMDFi3bh0ODg4UFRWhqqpK//79efbsGfn5+eI3oKyszPnz54XkwNvbG1tbW9asWcPcuXOZMWMGoaGh9O/fn3Xr1hEbG0u3bt14+/atyB7Yt28f58+fx8vLiw0bNhAcHExhYSHfffcd69atw97entjYWI4cOYKjoyMLFizg2bNnogiSZ2V4enri7+9Pp06dSEhIEJkru3fvZteuXQwaNIgJEyZgaGjInj17hMSuadOmvHnzBgsLC3JyclBRUWHz5s2UlZWxevVqzp07x8ePH1m/fj2PHz8Wn+vMmTOJjo4mLS0NLS0tjI2NSU1NRUtLi8GDB7NgwQI0NTV59+4dBw4c4Pnz53zzzTc0adKEu3fvUlRURFxcHPv27ePWrVu8e/dO+Lhv27aNjx8/0qBBA0aOHMmiRYswNzcnKyuLhg0bEhQUJGZ+582bR4sWLdi3bx/du3dnw4YNaGhocOfOHUaOHEmdOnXENmNjY9HV1SU4OJi8vDxq1KhB27ZtWbRoEba2tiJXRm6HKbfaVFBQIDk5GQ8PD1auXCmaIeWzlOXl5URGRrJ27VoAzMzMMDY2plu3bsDncCRTU1MePXqEs7Oz0Kr36dOHfv36MXfuXJKSklixYgVHjhyhbt26ZGdni4EyAwMD+vbty71797h9+zb37t2jZs2axMXFYWtrS0pKCpqamuTk5ODu7k6nTp2AzyFWAwYMYPv27bx584Z169YJF7v169eLXgi5bXB2djZLly4VA6o1a9ZEQ0OD9evX8+zZM96/f0+LFi1QVVVl/fr1uLq6EhUVJa6zqamp9OnTh8ePH3Pu3DkUFRWJjY2lSZMmvHnzhuDgYCwsLAgMDOTVq1fo6uoK+aH8JmvJkiXixql+/fooKyuTm5tLrVq1RD9A3bp1GT16NGZmZuL4nDx5Em9vb7Zv387Jkyfp168fHTt2REdHh0OHDhEQEEBaWhqOjo6oq6uTlpbGx48fKSgoQE9Pj2bNmiGTyUhJSaFdu3YoKipiZGSEtrY2t27dwtzcnIkTJzJ48GC0tLTo06cPqqqqqKioAJ+D+j5+/Ejt2rVZvHgxhoaGDB06lKysLOLi4pgwYcJ/pysOwPLly7l16xYWFhYcOnSIwYMH8+7dOxITE6moqODhw4fCjaZ79+7igNeuXRsPDw+aNm2Ks7MzlZWVzJs3DyMjI27cuMHr169p27YtkZGR2NvbY2BgwN27dzExMeH9+/dCI56UlISJiQnGxsZ8+PCBO3fukJOTg6WlJdXV1axatQpra2v8/f1ZvXo19erVY+/evSLtNS4uTjRaPX36lJiYGExNTcWFdPTo0cKBRu44oqamRp8+fXB2dsbR0RFNTU0uXrzIq1evRAd7SkoKLVu2JCMjg8uXL9O0aVPevn3LmjVr2L9/P6qqqjRo0IDatWvj6OgopDBymcK9e/eEVeaxY8fIycnh7NmzqKurk5iYiJ2dHfHx8QQGBuLv78+QIUNEMqienh4JCQkcOXKEiIgIatWqhZeXl4iItrOzo1atWpw9e5aoqCiqqqro3r27GDHYt28fZmZmpKamEhERgZWVFVVVVaiqqrJ//36ysrJYvHixcJlo06YNgYGBFBQUcPv2bXJzc8XU77Zt22jYsCFKSkp4e3vToUMHnj59yocPH/Dx8WH48OHMnz+fM2fO0LFjR3bs2EGtWrV4/fo1Bw4cEJZfDx8+JCQkRPRCwGf7LX19fdGorKWlRXFxMTNnzqSkpITRo0dz584dtLW1MTU1RSaTiRPpoUOHOHr0KBcvXiQnJ0cU/Fu2bGHJkiWsWLECHx8fIbtJSkoS7gG3b98mLy+P2NhYgoKCKCws5MyZM1RWVuLi4sLVq1eJi4vDxcVFpFRevnyZiooKCgoKMDQ0RFlZWTQIGxoacv36dYqLi3n69KkYhVdXVyc4OJiysjJUVFSwtbVl2rRp4qbWysqK8PBwAPLy8nj9+jVTp05ly5Yt9O/fnxYtWhAWFoa+vr442YaEhFCzZk22bNlCSUkJ7969Iysri6ioKO7cucOnT5+4du0a33//PXPnzsXNzU3Y+b1584bly5djZWUlAlMyMjIwNDRk165dxMfHo6qqiqamJs2aNSM2NhZjY2NkMpnIbVi1ahWTJk1i48aNwrWjS5curFy5EgUFBSoqKkSoz/Lly6mqquLJkyfcuHGD8PBw4YAk96JWV1fn48ePlJeX07x5cxQUFESzoY+PD9euXWPs2LFkZGQwY8YMdu/ezcWLF2nfvj1dunTh5MmTFBYWMmzYMCIiIrh27Zo4LpcvX/7iImhpaUnfvn0JCAhg2rRpmJub4+DgwLfffits+HR0dGjbti09evRg3LhxrFq1iurqanr16kWPHj0YPHgw586dIzs7m6FDh/Ltt9/Sq1cvkpOTxeiYv78/bdq0oaioiBUrVjB58mRGjBghHINWr15NQEAAkyZNwtraGjs7O9asWcOUKVOYM2cOBgYGtGzZkh49ejB+/HiWL18OQLdu3ejZsyfDhw/nzJkzFBQU4O/vT69evejduzeZmZm4u7uzYsUKnJycAFBUVOTmzZu8fPkSAF9fX+rVq4eysjIFBQWUlJQwfPhwUVTt2LEDJSUlZs2aRWpqqvDsDw0NJScnh6ioKNG06eTkhK+vr5Bffvfdd2zfvp2HDx+io6NDRkYGbm5uXLhwgezsbFRVVSkoKCAgIEA4brVo0QIHBwd0dHQICgoSM2KmpqYizVf+HdLS0uLevXu4uLhw4MABtm3bRt26dTl8+DB5eXls374dDQ0NJkyYINLLHz58SElJCZmZmfj7+wtf/rp166KoqCjSxOPi4pg6dSrKysoYGhqKBlg1NTVSUlKEkcL8+fOFI0xWVpZwjNPX1+fSpUu0bdsWFxcX1q1bJ9xz5DNCs2fPxs7Ojvfv33Pu3DnU1dVJTU2lS5cuNG/eXPiby8OYPDw8mD59OhUVFSgoKGBubs7p06f54YcfuH79OosWLaJfv58ItmkAAEk9SURBVH5oa2sLI4SysjL69u3L8OHDGTt2LObm5uTm5jJ//nyhTbe2tsbIyEjokN++fYuTkxPff/+9GKWW91N9/PiRpKQk4ax2/vx5srOzWbduHZmZmXz69EkYLqSnp2Nvb8/ChQtp2bIlLVu2xNvbm1u3blFUVMTNmzdp0KABrq6uYkDE2tqay5cvk5mZSWZmJs2aNSM/P58nT55QVFTE3bt3SUpKEmnvcgtl+c3rTz/9JByCLl68KG62Lly4gKOjI3Xq1GHlypU0atQIAwMDkpKSuHXrFtHR0aipqTF8+HBhiymX9v700084ODgwYMAAkVHi7u7Oy5cvKS8vx8PDgydPnlBYWMiIESPQ1dVlxYoVhISEsGbNGlq0aEHXrl3Jzc3l6NGjVFdXY29vT0JCAqqqquTk5DBx4kTevXvHpEmTuHfvnnAyU1FRwd3dHV1dXU6cOIGSkhI3b95EX18fJSUl8vLy8PHxoaSkBAUFBZEob2ZmRkREBCUlJeJan5KSQuvWrTl8+DDNmjVDX1+fFi1aEBwcTEBAAG3atGHy5MmcOnUKXV1dUlJSGDhwIIsXL0ZTU5MffviBwYMHU11djYaGBoMHDxYy3iZNmog8Dg0NDQoKCggLC6OoqIjnz59z+/ZtwsLCmDt3Lq1atcLZ2ZkhQ4Zw/fp17OzsePHiBR8+fCA1NRUXFxdMTEzYsWOH6C/68OEDw4cP58SJE5w+fZpNmzZRXV1NZGSkaNZeu3YtISEhDBgwgF27dv2qYfa/1hUHPodTLV68mE6dOnH79m0UFBR4//49JSUlaGtrU1BQQEZGBjVq1EBbWxtXV1ehc3v9+jXNmjWjsLCQgQMHkpSUxMOHD1m2bBmRkZF07twZc3Nzxo8fT0FBAX369MHDwwMfHx+UlJSYPHkyZmZmDB8+nO3bt3Pq1CmioqJ4+PCh8BiOiYkR+7p//37KyspEc52enh6LFy9mxowZKCsrs2XLFkxNTYHPXdOKioosWbKEx48fiwCFqqoq9uzZw9q1a4XtmJGREZ8+fRJTWenp6eTl5WFubo6qqirjx4/n1q1bfPr0SXhqy227/P39cXNz4+rVq3z48EE4R6Snp9O7d29GjhyJtbU1K1asQFVVFU9PTxQVFYmLi8PJyYlu3bqxadMmcnNzgc/6uOfPn/PgwQNsbW3p3Lkz165dY/v27YwYMYLvvvtO3F1PmjSJ3NxcqqurOX36NElJScKXe/To0ezevRsvLy8iIyNp1KgRTZs25datW8TFxRETE8OLFy9ISEigvLycQYMGMWLECAAWL16Ms7Mz5eXlKCkpsXz5cvr378+jR4+Ijo4mISGBZs2aoaCgwPnz5zE0NGTixImYmZmRkZHBsWPHqK6uZuvWrQQGBqKtrU2PHj3w9/dn4cKF2NjYMGjQICZOnIiBgQGFhYVcunQJe3t7jIyMOHXqFN9++y329vb4+/ujrKxMcHAwjRo1ok2bNpw5c4Y5c+YwZMgQmjRpQnx8PGFhYVRVVZGens7EiROFzlkuv3BycmLHjh0Aojm5W7dubN68mdDQUHJzc3n8+DGqqqq0b9+e/Px8Ll++THp6OklJSbi6upKRkcHr168pLy/HwMAAHR0dRo0aRWpqKgMGDEBJSUnITE6ePImvry979+7F09MTdXV1+vXrx759+4iNjRUzTMrKn08p8uAhAGdnZ65evSrcbeSBTqqqqsTExNCgQQNiY2OFl/SJEydQV1fnyJEj9OzZk7S0NBo3bsycOXNYsWIFbm5uHDp0CBcXF5YuXYq3tzcrV64kISEBPz8/rl+/LkZhWrVqhbq6OiNHjmTYsGFiilfuKlJZWUm9evVwdnamurqahQsXUr9+faytrQkPD2fDhg2sWLGCK1eu8PTpUy5fvkyXLl0wMzPj5cuX1KhRg+TkZEaNGkWLFi148+YNL168IC0tjT59+mBra8vNmzcJCgoSLjcTJ07k4MGDNGrUiB9++IHFixdTq1Yt6tWrh0wmIzk5Wfjjy2QyCgoKqKiooFOnTvz444/o6+vj7++PhYUFZ86cITAwkEOHDuHo6MigQYM4evQoM2fO5PLly9SpU4fRo0eza9cuxo8fz6NHjygrK2PatGls3LiRkSNHiu/K7NmzWb16NQMHDuTy5cukpaWxcOFCVqxYQa9evXj37h0xMTEsWLCA9evXCw3/+fPnCQwMZPfu3TRr1oxu3bpx7Ngx5s6dy+nTp7Gzs2P06NHs3buXqVOninPg5MmT2bp1K6NGjeLt27ckJycza9Ys1q5di4eHB9u2bRP9DgsXLqRPnz7Y2Niwe/dudu/ezdq1a1m6dKmwIGzXrh3u7u60adMGXV1devbsib6+Pps2bcLV1ZXc3FzOnj1LzZo1ad++PVVVVaxevRo/Pz+GDBki0k6HDBlCUVERI0eOZOrUqaIJV01NjeDgYBISEujUqRPe3t5YWlqSkpLCnDlzhE98w4YNmT9/PgoKCujo6KCjo8O0adN4/Pgxo0eP5unTp6IhuEaNGtjY2FBWVibWW7ZsGT179iQqKgpFRUUsLS3x9/dn165dPHv2DDc3N2bOnCmsEidNmkRiYiJPnjzB2tqaGjVq8ODBA5o1a0b9+vXZu3cvly5d4vz580La9+HDB+Hyk5+fT5cuXYiKiiIhIYFr167x5MkTDh06JBqvy8vLcXR05OHDh1y4cIGUlBQMDQ0JCQnB0tJSXNN8fHy4dOkSHz9+5M6dO9SoUUNIFSMiIlBWVmbs2LEkJiZy5coVkfAqN0KYNm0aAEVFRRgaGmJkZMSgQYPIzs5m9uzZAFhYWLBy5UqUlJRo164dHTp04PDhw0RFRfH+/Xu0tbXx8fHh8OHD7Ny5k6lTp2JpacmLFy+Eu9arV6+oWbMmampqtG7dmqZNm7Jz507Wrl3LmDFjUFNTo6KiAg8PD7y9venRowcPHz4kIyNDNI+GhoZiZGRESkoKioqKPHr0SLhFubq60rBhQ548ecKFCxfo1q0btWvX5ty5c3Tt2pUlS5bQv39/rl+/TkhICLVr1xafYVlZGY8fPyY0NJSoqCjevHkDQP/+/VFSUmL06NE4OTlRXFyMtrY2qqqqtGjRAjMzM5YuXYqPjw+hoaHCKnTXrl1888036Orq4uDggIODA/r6+gQFBaGiosK2bdtYu3YtdnZ2jBgxgrVr14qMDDs7O9Fv9c033/D69Ws6dOiAnp4eY8aMoUaNGnh6etKkSRO6du3Ku3fv8PHxETbj6urqZGZmMnz4cC5fvszEiRMZMWIE27dv55tvvsHZ2Vl8X01NTTl06BBXr16lqKiIMWPG4OzszI8//sibN2/Q1tZGQ0ODpk2bcuXKFWrXri1UDu/fv8fZ2ZmysjJ8fHyoWbMmR48eZfny5ZSUlLBixQpu3ryJubk5enp6FBUV0bFjR1JSUkRdd/ToUTp37sz27dvZunUrgwYN4unTpyKQKjs7m9LSUjQ1NVFSUuLhw4fMmzePhg0bsmbNGqqrq3nx4gU2NjYEBwfTrl07Tp48iYeHB3PnzuXHH3/k+fPntGnTBm9vb+zt7VmwYIEY7e/QoQMHDx5k+PDhXLx4kaSkJHH9/LmV+dfwjyvs9+3bR2RkJGlpaaKBMSwsDEtLS1q1asWJEydwc3PD0dGR6dOns3PnThITE4mOjsbe3p6MjAz09fUpKyujcePG7N27l9jYWLp06YKNjQ1du3bFwMCAN2/e8PjxY44ePcrp06cxNjbmwIEDpKWlsWXLFmQyGf379wc+jzKZmJjQuHFjvv/+e54/f05GRoaYYq+urqaqqoq4uDiR1vb+/XuOHj3KDz/8QI8ePXjz5o3QW5mYmJCfny8kJfKRQrnForq6unBW0dTUxNLSEltbW4yNjYmJiWHnzp1Ck/b8+XN8fHzYvn07SkpKbNmyRUg0vLy8UFJSwsjIiODgYJGyWVlZiUwm4+TJk2zZsoXIyEjq1KmDTCZj8ODBfPz4kePHj2NpaYmWlhaFhYWcPHmSDh06iC/rgQMHiI6OZsiQIbi6umJvby8ucAMHDhTONfKpqmXLltG4cWNiY2N58+YNZWVlhIeHo6Ojg0wmEy4E8s/24cOHwl4vPj6e1atXc+rUKS5evMjChQuprq5m2rRpvHjxQtjceXl5YWVlxZAhQ6isrGTt2rUcPHgQDw8P0tPTad68ORs2bBCOAjExMaiqqtK6dWt++uknxo8fz9ixY+nTpw/Dhg3D0tKSsLAwLl++jIqKCsOHD2fnzp1cvXqVsLAwEc6xY8cOpkyZIqba8/PzRaDazp07efnyJd999x337t2jXbt25OXlMXLkSH744Qdev35N9+7dCQ8Px9ramvj4eI4cOcKFCxc4f/48jRs3Zty4cRw8eJC+ffuyaNEiXFxcePLkiRhNlHs9T548mcDAQM6fPy9C0eRWkMnJyVhbW+Ph4SGm1bOzs0US8ubNm1mwYIFoHr179y42Nja4urqKNNqkpCSMjIz48OEDSkpKqKqqMmLECPLz89m/fz+5ubmUlJQI6dKLFy84c+YMCgoKnDlzRvjk+/r60rp1a7S0tJg+fTrm5uakpKTg6OjI2LFjRVKhhYUFr1694rvvvmPp0qWoq6vz3Xff0bBhQwIDA/Hy8mLLli2EhIQQHx/Prl27GDNmDM2aNaOgoIB79+4xffp0wsPD0dTUFL9pNzc3Ecj2ww8/8PLlSzF7tHz5ciGbSUxMFB7J8pmDgoICYmJiKCoqYtOmTVy7do3u3btTUVGBr6+vcMGJiooSXtxZWVnUqVNHuHspKSmxadMmzMzMUFZWFrak169fJz09HQB/f39hPTp79myqq6sZN24cKSkplJWVMXr0aADhDZ+VlcWgQYNQUlJi4cKFFBQUUFRURK9evVBRUWHdunUoKSmJ0X6ZTMaBAwfEbM+3335LWVmZuCApKiqKhMfo6Gjmzp0LwIgRI3j58iWfPn0Syc5Tp04lLS2NvLw8hgwZgqqqKps2bUImk4lU2W7dunHgwAHevXtH48aNgc/pnXIdctu2bWnbti1ZWVnAZwMF+QCC3Ns8OzsbNzc3AgIChFPWvHnzKCoqYs+ePVRXV9OgQQOqqqpwdnbmw4cPFBcXs2rVKvbu3YulpSVTp04lKiqKpk2bAp/NBSwsLACYPXs2HTt2ZNq0adjY2IhZ2z179vD+/Xt0dXWJj4/n2bNnjB07lvj4eFJSUnBxcUFNTY3c3FzmzZuHhYUFkZGRlJaWMn/+fG7fvs3NmzfZuXOncGWJiIigbdu2hIeH8/HjRxYsWMDixYtp27YtW7Zs4ciRI1RXV6OqqoqJiQlt27YVftozZ87k2rVreHp60r59exEqtHPnTnbs2EGvXr2wsrIS7mwxMTHcunULRUVF7t+/z7x583BycqKkpAQnJyf09fWxsrJi0KBBJCUl4ebmRkhICBMmTODcuXP4+fmRmJhIXFwco0aN4vnz57x69YqJEyfy7NkzsrOzqVevHsbGxtSsWZM9e/bw6tUrMQMql1A0a9YMJSUlunbtysGDBzExMSEqKopmzZrh5uYmchQqKyuZOHEijx8/Ji8vT4zEFhYWUr9+fQBatGhBQkICDx48QF1dnZiYGJo3b8727dtRUVHBxsaGe/fuiZkOXV1dwsPD6dixI2FhYejo6LBy5UqcnZ2ZOnUqRkZG9O/fX9i8JiQkEBYWRseOHYVEr7CwEGNjY9q0acP69euFRGnKlCl07NiRT58+cePGDdFELg+0Gj16tHAI8vDw4NixY/j5+WFiYsLLly/x9PTk9OnTLF++XEhbV69ezapVq7h9+zbnzp2jVatWLF68GB0dHfz9/enduzcTJ05k79693L9/n9atW3P16lVsbGzo0qWLcGmT20Tr6ekRFhYm8iKqqqpQUFDA2tqa4uJi2rdvz7Nnz6hbty5ZWVnCItnFxYVXr15Rp04djI2NuXPnjnDX0dfXJy0tDQMDAxo0aMDcuXNZvXo1ZWVlpKWlkZycTIcOHahduzbq6up07NiRBw8e4ODgIFyvzp8/z/Xr1/Hy8uLly5ecPXtW9L/Z2dmxd+9evLy8uHXrFrt27SIrKwtDQ0NKSkowMzNj1KhRREREYGdnB8C5c+dEonnTpk2ZM2cO+vr6TJs2jeXLl/Px40dCQkLIz89n6NCh4rrUuHFjwsPDRXhYRUUFrVq1Ijs7m0uXLjF37lyOHDnC5MmTRdZGSEgIu3fvxsjIiOjoaAICAnj8+DGHDx/m5MmTJCQkEBkZyZo1a4TsUh7S+Gf8I+0u5URHR/Pq1Sv69u0r/n78+DFnzpzBzMxM6NWWLFnClClThG1ZWVkZ2dnZ+Pr60qFDBwIDA7GwsGDChAm4u7uzefNmQkJCuHXrFps2beLmzZs4OzujqanJ9evXMTExwdvbm4CAAAwMDNi4cSMDBgwQ0gZNTU26detGYWEhDRs2ZMWKFVRWVpKYmIitrS1HjhwhIyMD+HyB2rRpE69evRKyiJcvX2Jubk5OTg7Nmzdn6tSphIeHc/fuXW7evIm2traY7q9RowYBAQHUr1+f3r17c/78efT09Dhw4ADV1dWEh4ezevVqYmNjWbBgAY6Ojjg6OorCTH7RevDgAaNHj6a0tFQ4knTu3JkLFy4wdOhQpkyZwsSJE+nUqROKioqkp6fj7e1NcnIyubm5YsbBwsKCNWvW0LhxY3R1denQoQP37t1j48aNpKen06hRI65du0Z1dTX16tWjpKSEGTNmEBQUxJkzZ/j2228ZP3483t7edOvWjXr16vH48WMaNmxIcnIyr1+/pmPHjoSEhLB48WJMTU2F/eCSJUu4ceMGS5YsYfny5fTt25fKykpOnjzJ2LFjhQ1av379OHHihIiFnjFjBtnZ2axfv55Dhw6hp6fH+fPnOXr0KP7+/mzbto0BAwZw4sQJdHV1xfHX1dXl6NGj5Ofn07JlS2rWrElmZiYlJSWEhIQwf/580ZyspqbG8ePHuXLliri4APz4449kZ2dTUVEhnHmcnZ1JSUkRja3r1q3Dz88PNzc3ERy1ePFihg0bJqRNJiYmpKamEhgYyLZt28jJycHKyop3797h7e1NREQEurq6bN68maCgIFauXElERATnzp1j3759wuXEzc2N1NRUevbsSWJiIq1atRIzTLdu3aJJkybo6+uTlZXF2bNnGTFiBJs3b+bw4cOYm5vz/PlzVFRUaNCgAbt370ZDQ4PIyEguXrxIvXr1MDMz49mzZzRu3Jj09HSioqIoKSnh8OHDeHh4UFBQQLt27bhx4wZBQUEMHz4cZ2dnzp49i4+PDwYGBoSFheHt7U1+fj4PHz6ka9euwpZQHtQ2evRoGjRoIGwd5bK1tm3bYm5ujqamJqmpqVy4cIGgoCCmTZvG3bt3cXNz4+3bt4SHh+Po6IiNjQ2pqals27aNgwcPYmlpiYqKCpMmTeLYsWMizXTIkCGoqamhoKBAdHQ03t7eqKqqsnLlSs6dO8e9e/d48+YN/v7+1KpVi4qKCuCz3E5dXR1tbW1htSkvgGQyGenp6ZiZmVFaWkppaSk1a9YUmumKigrhoJKZmSma8rKzs4WeU1tbGxUVFRGAl5OTg6qqqmhelOumlZSU0NXV5dWrV1haWpKfn091dTU1atQgPT0dc3NziouLKS8vx9DQkLdv32JiYsKnT5++2AdDQ0Oqq6vJzc3FxMSEjx8/oqOjg5KSktCXKigocPr0aRo1aoSZmRlz5swRshM3NzeuXLmCn58fZ8+e5f3790LGN2vWLBo0aEBGRgZPnjzBwMBANFjq6enRpEkTbty4QWBgoCgKtm7dSlRUFElJSUydOhUFBQWqqqrIy8ujqKhIFI3u7u4EBAQQEhJCRUUF27Zt4/Xr17i4uIgRXXnwma2trZARNGzYEF1dXb777jvKy8spKiriwoUL3LhxgyFDhjB69Gg6dOhAdnY2ISEh4ppQXV3N4sWLGTBgAJcvX+bp06c0a9YMPz8/LC0t2b17N8rKypw6dYo+ffqIxsHNmzfj7e1NTk4Or1+/xsjICDU1NTIzMzEzM6OsrAxDQ0PS09Px8vIiOjqaVq1akZmZyfnz5wkJCWHw4MFcunSJGzduEBsbi76+Pvr6+ty/f5+BAwcyffp0YdvaqlUrNm/ezMuXL1FUVOT58+ccOXKENWvWCPciNTU1ioqK8PLyIioqCnd3d1xdXfHx8cHV1ZVvv/1WSIS0tLR48OABjRo1wtDQkLFjx7Jt2za2bduGhYUF9erVw8rKipSUFMaPH09QUBCXLl3i7du3zJ8/nyFDhhAeHi7MFDIzM8WIN0BlZSUVFRVYW1tjZmYGQHV1NXfv3uXkyZPk5+dTVFREVlYWKioqotE+JiaGvXv3snHjRt68eYONjQ23bt3iwoULJCYmUlxczPLly/Hx8eHOnTts2rSJyZMnM3LkSPE919PTY+LEiSKBu2nTphw+fJiePXvSq1cvLl++jL+/P87Ozly6dInq6mocHBzo1asXZWVl4mbSw8ODRo0a8eDBA968eYOdnR0ymYyMjAw6dOhAcnIyGhoatGvXjnfv3nHt2jWRimxiYsK8efOwsrKiuLgYRUVFPn78iLKyMhkZGfj7+6OkpISenh7Kyso0b96c3r178/TpU2bPns2yZctYtGgRRUVFqKurs3nzZhFC+fLlS+Lj44WffUVFBTo6Orx//54+ffoQFhbG/PnzkclkPHz4EJlMRsOGDfnhhx+QyWRcuHCBtWvXcvv2bUpLSykvL0dNTY1r167RunVrXFxcyM7OpkOHDqSmpqKqqip6BuUWv8rKyqIxuqqqiuTkZAwMDAgODsbAwIDY2Fjq1KlDYWEh9vb29OjRA/icbTNv3jysra1ZtGgRU6ZMoXHjxtSpU4d79+5RWVlJamoqBw8eFMGObdq04datWwwdOpRBgwbRsmVLFi9eTPPmzdHX12fRokUYGxszefJkhgwZwuzZs3F0dOT169esX78eW1tb6tWrh5+fH9ra2hw6dIi7d++ybds2Bg4cSHx8PN7e3mhoaODm5kbXrl2/uvb9x43Y/5xnz54RHh5O3759iYuLIzY2luLiYjZt2kT79u0xNjYWQRWpqamEhIRw8+ZNmjZtirKyMhs3bkRZWZn79+9Tt25dPn36hIWFBaqqqtjY2KCqqkpiYiK3bt3C0tKSK1eu4ODgwKpVq1i0aBHu7u6Eh4djZWWFqakpDx8+pH///lRXV6OpqcmIESM4ffo0r1+/5sqVK8IFITExEVVVVfT19TEwMKBjx45UV1dz48YNIiMj+fjxI1ZWVrx584awsDAGDRrErVu3yM/Pp1atWlhZWXH37l0KCwtFaqFcItS/f3+qqqooLy/n/PnznD9/nvfv32Ntbc2xY8c4duwYhw8fpk2bNgBUVVVx48YNTp48SXV1NSYmJnh6erJ7926mT5+Ol5cXp06dYvbs2dy7d4/atWuTlpbGiBEjaN++PRoaGrRs2ZLVq1fj5eVFWFiY0Bvb2dlx6dIl4X97//59MjMzuXnzJoGBgezZs4cuXbpQv359ysvLGTx4MM+fPycgIIDatWvz/PlzatasSUlJCQsWLGDatGl4e3tjYmLCsWPH2LVrF05OTlRWVrJx40a+//57goOD8fX1FSM3OTk5IkSloKCA9evXs3z5cmrVqsWePXt4+PChuFmaMWOG8Ppt2LAhxcXFxMTEYGlpSVpaGg4ODsyfP599+/aRlZXFrVu3cHR0pEaNGrx8+ZJLly5hYmKCoaEhpqamvH79mvz8fPbs2cPgwYOZMWMGM2bM4P79+3h4eADg4OBAaWkp7969IzIykqqqKuFO0alTJxYtWsTp06d58eKF0A+Xlpby7Nkzbty4IRqAIyMjCQsLIykpia1bt+Lu7s6PP/5IdHQ0mzZtYt68eezYsYOwsDAWLVpEaGgoFRUVpKWlsXPnTjQ1Ndm5cyc6OjrExMQwduxY9uzZw9y5c/H29mbBggUsXbqU4uJiJk2aBCAsOd+/f0+HDh0ICgqiS5cuDB06FBcXFzQ0NACEfvXKlSsATJkyhR07dpCYmEijRo04ePAga9asISsri3r16lFYWMinT584dOgQBQUFtGzZkrNnz3L37l1OnTrFsWPH0NPTE17LOjo6REREkJyczKRJk3B1dRUSttzcXGQymdBhy0df4uLi0NDQEMmp9evXp23btqLpu2fPnuI8c/PmTSoqKkT4UXFxMevWraNJkya8fv2asrIyWrVqhYGBgXAr0tfX58KFC3z69ImzZ8+K2PZt27bx9OlTpk6dSkhICKmpqdja2nLixAkxMycf8d2xYwcWFhYYGRlx7949Ro0axcqVK9HX18fBwYHr168zbNgw1q1bh4qKCm3btuXs2bP0799fxKzL7QzlLlz5+fkMGTKEvXv34u3tzfHjx8nIyGD06NEEBgbyzTffcOHCBaG/X7RoETY2NmhraxMbG8vYsWPF78fW1pZbt26J6X0NDQ1atWrF+fPnGTBgABs3bkQmk9GtWzeOHz9Oz5492bVrl7DIu3LlCqNGjSI3N5ekpCT27t3LqVOnePbsGU2aNEFdXR1DQ0Pgs0yhd+/eHDhwgP3797Nnzx7s7OyIiYlh0qRJpKamsmfPHpycnDAwMGD06NE4OjpSXFxM7969mTNnDlZWVtjY2DB27FjU1NTo168frq6utGzZkvnz56OoqEhUVBQ7duxAXV2dGTNmcOnSJaytrXFxcREuWIaGhsyYMYO3b99ia2tL/fr1mTVrFh8/fhS9CmfOnMHQ0JA5c+aQkJAgDAO2b98uUoNfvnzJs2fPGDNmDDNnziQlJUXowQsKCti8eTPnzp3j6NGjXL16lWnTpgnpztu3b8nOzmbNmjVMmzYNTU1N1qxZw/3797l9+/YX7j3Tp08Xv8WrV69SVlZGbm4uRUVFmJiYiPN0rVq1iIiIEGmjtWvXJjk5mbCwMNTU1Bg2bBht2rRBQ0ODs2fPEhgYiJOTk5CVnDlzhh9++IE6deqgoqIiJF21a9dm9+7d6Ovr07JlS4yMjGjSpAmPHz9m8+bNtGvXDnNzc2ExqaWlRdOmTdHW1qayspLatWuLFOfy8nK8vb3Zt28fmzdvRllZGVNTU+bMmUObNm3Izs4mLi6O3bt3Y2Zmxo8//iispd+/f09xcTEdO3ZkwIABBAQEEBAQQMeOHalfvz5paWl06tSJffv2CSeXzp07c+bMGXr37o2GhoaQWpw7d45GjRpRWVnJtWvXmD59OpmZmURHR2NlZcX06dNFQ21FRYVw6erYsSOtWrWidu3aREREcP36dY4ePUpVVRUaGhr06tWL7OxsLC0tuXPnDhYWFixcuJDExEQhTyotLSU6OprTp08LeeWuXbvo0aMH0dHR1KxZk3v37rF582Zat24NfE6Ml5uJDBs2TLh+RUdHo6CgwIgRI6isrGTx4sXiJk7u8qKgoED37t2xs7MTvvB37tyhRYsWuLm5iWttp06dSEpK4sCBA7x+/RoVFRUsLCxo3749WlpaXL16lePHj7NhwwYWL16MkpISS5cuJTMzk/Lycnr27MmkSZMYO3YsCxcuJCQkhA8fPjBmzBjxPd+8ebMYQGzdurVo2r9z544Y6JDnOvTv3x8XFxd69+6Nrq4unp6ebNy4kZcvXxIVFUVUVBS6urrA5/45MzMzkTPQqFEjHB0d0dDQICYmRty0JyYmsn//fsaNG0eLFi3Yu3cvWlpavH//nsTERExNTendu7fo+3v//j35+fkoKyvz4sUL9u/fT/v27dm/fz9Lly4lOzsb+Gv6eviHj9g/f/4cPz8/du3axZIlSxgxYgQ3btxAV1cXX19fBg0axIEDB9DV1SU9PZ3k5GQ2bNggGmTk4S0TJ07k8OHDxMbGCvmHjo4OGhoa7Nmzh6KiInx9fTly5AgWFhYsWLAA+FwUz5w5k+zsbLy9vYmMjMTY2JiCggK0tLRE9PHNmzdp3rw5I0eOJDs7m4kTJ4pAkIsXL2JqaoqmpiZ16tRh1apVnDt3Dm9vb2rWrMmNGzeIiIggNDSUpk2bYmFhgYaGBvfu3ePRo0c4ODgICzf53ancHUIeES0/qS9dupTVq1fj5uZGcXEx+/btQ0lJiWvXrtGnTx90dXU5d+4cr1+/pl27djRs2BCA0NBQ4RP+zTffUFJSgqmpqTh5vXr1CltbWzHlFhUVxYIFCwgKCqJ+/frMnTuXSZMmERoaio2NDQkJCfz444+cPHkSKysrkpOTyc/PZ9q0aaJIU1FRwdLSku7duzN58mRyc3Px8/PjypUrnD59WqStLlq0iM6dO4uGvtWrVwut5i9xcnLi8OHDnDlzhpo1azJy5Eg+fPjAiRMnKCgoQCaT0bJlS3R1ddHT00NLS4vy8nLxuRQUFKCmpgZ8ltC0bdsWbW1tOnToQElJiQiYUFRUxNnZmcDAQPr164eqqipr165lxowZKCoqYmpqKmZK5MjtDOWMHDmS2bNno6mpSZ8+fbh27RoHDx7EysqKNWvW4O3tzZ07dxgwYADbtm1jzZo1omG4QYMGANy9e1dIelauXIm5uTnu7u5YWVkRFRVFkyZNePfuHR06dAA+B5ZYW1uLfZA7DNy9e5eJEyeyc+dOMaoktwxMSkoSo9nDhw9n7ty5NG/enBo1anDv3j3s7e1JTk4mKCiIBw8eUFZWxuLFi5k6dSqlpaUMHTpUvN7GjRvFdG9mZibr1q1DJpNRq1YtwsLCOH78OFu3bqVRo0b06NGDjRs3EhkZSe/evQkJCaG6uprdu3fz6tUr8vPzxQigvCF83LhxIrny1atXqKqqoqGhIRoQt27dyokTJ0Qh1q5dO/z9/dm7dy8//fSTkBfZ29ujoaFBSUkJt2/fZsWKFSKsKzAwkM6dO4vUX7n/ubyZtVatWtjY2NC9e3fht21kZMTw4cM5ePAgampqDBkyhG3bttGpUydCQ0PR0tISMyNNmzYVUhi5ltXU1JTy8nLy8vIYN26cmBGqXbs2KSkpTJgwgV27dlFeXo67uzuRkZGMGTOGgwcPUlJSQq9evTh+/DijRo0SqcQDBgxgz5499OjRg3PnzqGpqcnw4cPZvHkz7u7uPHjwADU1NUaNGsWWLVvEyH9paSljx45l+/btIqQuPT2dMWPGsHfvXj59+sQ333xDXFyc+P7Gx8cDnwsvY2Nj8Vv9ubb77NmzYpq9tLSUT58+oaenR3Z2thh9lNO5c2fGjx9P586dKSkpwd7enrFjx+Lr64uzs7NYr7KyUhgnTJgwgcTERBwcHITM5OXLl2RlZZGcnIyKigoJCQm8fPkSMzMzGjduLGYhjIyMaN68OampqcyaNYv79+/z7bff8ujRI8zMzJg4cSLXr18nOTmZY8eOERkZSbNmzejatSuzZ88mPT2dWbNm0bRpU7KysvDx8WHkyJEYGRlx8eJFunTpIq538qbvH3/8kZKSEq5fv46vry+1atUSNnzV1dWcP3+eVq1a0bt3b+bNmyccoV6/fo2joyNxcXEYGRnRs2dPVFRUePjwIY0aNeL+/fvk5OSQnp4ublTU1NRQVlYWye/Lli2jurqa2NhYKisryc3NJTY2litXrhAYGCiaFsPDw1m+fLmwvDx+/Dg1atSguLhY2CY2btxYNE6XlZWJxsvmzZvz4sULTpw4IQIPjx8/jqenJ0+ePMHGxgYfHx88PDwICwvj/fv36OnpCQcTFRUVAgMDiY2NFXbHP8fZ2ZmhQ4eKpvjbt2/TokUL0dcwduxYunfvTlRUFCEhIRgYGPDx40e6d+/OsGHD2Lx5M6ampnh7e4vvbHV1NfPmzeP69es0aNCAhQsXcu7cOZKSkqhRowafPn3iwoUL2NrakpeXx44dO5g6dSrdu3cXVr5nz54lPDxc6MLv3LlDgwYNePbsGTdv3hQ3dtOmTcPd3Z2HDx+KWZP8/HzevHkjZh1/SWJiIgMGDCAvL4+BAwcKxzG5256Wlhb79+8nMzMTOzs76tatK57bqFEjQkND0dPTo0WLFsKCVU5FRQVmZmZ4eHigq6tLRkYGcXFxYjZSJpNRWlqKk5MTP/30E9bW1ujq6nLgwAEuXLjA+vXrhWf+qFGj2LlzJ/C5eT4gIICysjKcnZ0ZM2YMAQEBuLm50b17d1q0aEFMTAw2NjZMmDCB+fPn4+Pjw40bNzA3N2f27NmMGzdOzFTm5+czYcIEcb2TH7umTZvi5OTEoUOHxKx0cHAwenp6wOdQr1WrVhETE8O1a9c4duwYCxcuFIW5HLkFu5GREaNGjWLHjh1s374dZ2dnbty4wb59+7CxsRH9WkZGRty/f5+ioiJOnTr11VKcf/SIvZ2dHSEhIYSHhzN79mzatm1Lz549OXXqFE+fPqV///4cPXqUDx8+UFhYyP79+/nuu+94+vQp/fr1IyQkhAcPHuDk5MT+/fuBz57yP6ddu3bA51FHebAJwLZt24iJieHt27d06dKFgIAAQkNDcXNzAyAyMlL4/Xbo0IHjx49jYGBAq1at2LZtGwEBAUyfPp0bN27w4MGDL15THh4lH+XW1NQUBWJkZCRv3rzBx8cHIyMjBg4cyLt37ygvL8fV1RUAY2Nj6tSpQ/PmzQG4cOECeXl5aGlpoaWlJWyfdHR0yMvLE1Z+kyZN4sOHDxw/fpxFixbRvn173N3d6dChAy1btqRLly4YGRnRsGFDtLS0gM8/jk2bNomR/KlTp3L79m0yMzMZO3YsUVFR+Pr6Mm3aNJG0KP/hhIWFERoayp49ewgICKBFixacOHECKysrvvnmG7Zs2YKBgQGvXr2iTZs2dOrUiQMHDjBq1Ciys7PJzc0lMjKSmJgYsrOzGTFihJC4/BYtWrRg0aJFqKmpsWfPHiIjI7l9+zatW7fG398fAFdXV0pLSzl+/PgXz5X70MqP79atWwFQU1P7olj4Od26dePu3bu8fv2amTNniuPxW+jp6YkLJ8Do0aN5+fIl+vr6GBkZ8cMPP6CpqUlkZCTTpk0jLy+Pzp07Y2lpKT4vQIxwAri7u3PhwgXi4+M5fvy48BWOjY1l8eLFPHv2jMzMTHFD4urq+kVh7+7uzs2bNwkLCyM7O1vcMMDnm5eioiKuXr0qLOLkIUkmJibs3LkTGxsbFi5cyJAhQ0Sj65QpU/Dx8SEiIoINGzZ88ZmMGzeO48ePY2dnh7GxsXCAAkTYS506dTh58iRNmzbF3Nycfv36AdCnTx9mzZqFr68vjx49Qk9PjxMnTvDmzRsiIyPR1NRk9erVVFZWMnr0aC5evEiTJk0oLS0lPz8fd3d31NTUaNy4Me3ataO6upozZ85QUlJC27ZthS+7XEI3bNgwIddKSUnh3bt3rFu3jo0bN9K5c2fU1dU5d+6csLmU23vC/z868+nTJ2rWrIm/vz9GRkYiNbJmzZqiMVtdXR0NDQ0MDQ2xtrYWSddKSkoYGBhQp04d9PX1KS0tFSPcRkZGKCkpUbNmTRHEpa2tjbq6OkZGRtjZ2WFoaIiKigqKiooinKhWrVpUVlairKyMkZER9vb2GBsbo6ysjIqKCgYGBsLhSkdHB3V1dQwMDKhbt66IvK+srBT7oKmpSY0aNVBQUMDQ0BA9PT0qKysxMjIS7igFBQXi+MsvsvC5CHnw4IEo2OUJq4CYafnl911O69at2bx5M6WlpaLBb9KkSWhra3+xnrKyMj4+PmRlZTF79mzq1KkjLPoKCwsBRFCSvDhcvXo1s2bNYujQoaipqZGamsrly5d58OABJSUlaGlp4eHhITJUTp48yfbt24HPMz/q6uqcPXsWBQUFcnNzcXd3FwmuN2/exM/Pj+joaIKDg7l69aoo6uHz9W7Pnj2cP3+en376SeRJ1KtXDwcHB0aNGiUcy+Q2ih4eHqxZs4agoCCio6Np3LgxBw4coH79+syZM0ecL+XnD09PTwBKS0tZtGgRXl5e4m/5+UneT1GnTh0hJZPnMMycOZN27dqRkJCAkZERpqamdO3albt37xIUFERqairjx4/n1KlTeHh4oKKiwowZM1izZo3QWQNEREQIwwkLCwt2796Nq6srK1asYPXq1fTq1UtYNhYXFzN58mQxgp6QkMCoUaMAxE1DUFAQp06dwsbGBoBVq1Zx8OBBTp06xfPnz/nw4QM3btxg5MiRHD16lCdPnqCjo4OmpiZxcXEsWbKE6upqMWgmn5Vevny56JGRyxgnT57MggULyMvLw93dneDgYKysrMQMura2NgkJCTRu3JjXr1+Tk5ODjY0Nz58/x8zMTHy+gJBnyBtCIyMj2b9/Pz4+PlhYWDB06FCmTp1Ku3btOHXqFKGhoWKU/pcYGxuzc+dOYQ9dUVFBnTp1MDExITMzk/r16zNy5MjffC7At99++7vLlJSUMDMzw8fHhwsXLlBQUMDOnTs5cOAAenp6eHt7c/fuXXbs2MHMmTNp0qQJgwYNAj7nEnl4eGBqasrRo0dp1aqV2O6OHTswNDQUv9158+Zha2uLuro6vXv35vXr11hbW3P//n2Rst62bVtSUlKwtLSkoKCAlJQU1NXV0dTUxNXVleLiYp4/f058fLxwm4LPDd0LFixgyJAhGBoa0q5dOyFhg8/nnaCgIG7evEn37t2JjY2lXbt26Ovrc+zYMZo0aUL9+vWxsbHh1atXbN26FQMDAx48eMCDBw9EYJh8rP3Zs2diluJrC3o5/+gR+7+Ck5MTYWFhGBkZUV5eTnp6OvXq1QMQDW/t27f/6u3du3dPFNJpaWkcP34cNTU1+vfv/8UFSs6jR4/EaOezZ8/EdH2nTp3w8/PD2NiYWrVqfSEBuHbtGpcvXxbx5GPGjEFVVZVTp07x4sUL0tPTGTBgAJaWlpw8eZL27dvTq1cvMb0kJzo6WngC6+rqUl5ezvLlywkNDWX//v2EhITg6+tLcHAw+vr6TJkyhaCgINq3b09ycrLYXmxsLPPmzfvixx8TE8PHjx8xNzdn5cqVHD58mIEDB7JgwQLRwDNmzJhffR5yazl5yFVZWRlOTk6YmpqKbICPHz8Cn6ePQ0JC0NHR4cWLF8ydO5ewsDBh3zl//nw+fPiAurr6Hx6zx48fU15ezrRp0/D392fkyJFMmzaN1q1b06tXL7FeSUmJaAL+OVpaWtjb2//ha8Dn0fzq6moATp8+jZmZGa6urowZM+arp9x27drFxo0bqVu3LkVFRfj4+GBsbIympiYWFhbMnTsXbW1tcfGUWwx+LZWVlZw9e5aUlBSGDBkCgKam5hffHXlxm5GRQffu3YVmVc7SpUs5e/YsKioqfPjwgQYNGvDq1Su2b99OSkoKhw8fZtGiRYwbN46NGzfi6elJTEwMMTExuLm5iZukw4cPU1RUxMOHD8nPz8fS0pLY2FiRgPu1bN++natXr5KamsqWLVtYvHgxhYWFeHl5kZeXJ6b2zczM6NSpE6ampmRmZpKSkoKfnx/h4eHCg/nn7Nu3D1tbW+rUqYOZmRne3t6sWLGC7777jufPn1NdXS1CYeQNjD/n3bt37Nq1iyZNmnD79m1++uknAgICaNmyJdra2mRlZQmHJPn7lfdIPHjwQNg9ZmRkYG1tzZMnTzA3N0ddXZ309HQhFTMwMBA2cHIXEi0tLWrWrCkSoAsKClBQUMDExIS4uDisrKyoqKigrKwMCwsL7t69KxpMCwoKsLKyIiYmBjs7O5SUlPjw4QNWVlbEx8djYWGBmpoab968Ee4x8qI/NTVV9Hjo6Oigr68votNzc3MxMzPDwMCAiIgIUcz9nJ+P3v9PiI2NJTExET09Pdq0afOH3yO5Vld+4/rp0ydCQ0NFUT127Fjhr96+fXtOnz5N69atUVJSIjc3V+RayDEwMPjiBlnOp0+fePv2LVZWVly6dElY6wYHBzN9+nQcHR3JycnBzc2NwsJCPDw8RGjiLyktLSU4OJgXL16Ql5eHiooKZmZmdOnSRVzXABISEoR0zNHRkalTp9K2bVuOHTv2q1nDn1NcXCwSmeU39fLzsZx+/foJ3bM8JfS3iI+PJz09nTt37hAaGiosDOU5HpWVlRw8ePCL58jtBeWEhoZy/vx5lixZwvfff8+aNWu+KGCfPHlC7dq1RXK7n5/fF9vz8PBg06ZN4gbm5/z8WrRz505ycnIYP348tWrV4v79+8hkMmxsbFi8eDEaGhpcunRJ2EoGBQWRkpJCw4YN8fT0pEGDBqioqPDo0SNGjx5NeXk5ixYtokuXLqKR3cDAgIcPHxIeHs6ePXu4d+8eixYtIiEhQcy6/5LTp0+jqalJgwYNCA0NZeDAgWhpaVFSUiJmuDMyMkQC+x9dZwoLCzl9+rSwR5Y3Mf9VZ5ZfMn/+fC5evIiuri7a2tpcuHCBwsJClJSUiIyM5Pvvv2fz5s3MmjVLpLr/GXJ3ro8fPzJr1iwMDAy4cOECSkpKpKam/uo7CVCvXj3i4uLQ09PD2dmZcePGkZGRQa1atXB0dBQy2LVr19K8eXOmTJkCfJ4xOnHiBEVFRXTv3l3Id3+rFtDT02P+/Pk4Ojoyfvx4xowZQ79+/fj2228pLy/n6tWrhIaG4unpKdyp9PT0xIAEfL5pkddGgYGBBAYGSs2zX0NoaCgfP34kLy+PZcuWER8fL0aW/i6SkpI4dOgQhYWF9OzZkzFjxohQoT+juLiYhIQE8b+2trZwhIDPuq+bN28SGRnJs2fPmDNnDo6OjoSHh3P79m0SEhLw8PAgLi4OHx8frKys2LlzJ+3atWP8+PG/er0HDx6Ql5eHqqoqTZo0QVVVlevXr3P9+nURV+7l5UXDhg0xMTEhMDBQSIgAxowZg5ub2+9ebH7J3LlzRdy1fHr/54VjcnIyV65cEcX8751Y5D/g0NBQLl68yIgRI9iwYQMHDx7E2dlZXGDkoyl/NOoAMHPmTMLDw2nTpg0pKSk8fvxY+MkuW7bsiynI/wkxMTH81s/uj0bsf0nPnj0ZMWLEb0ZNZ2dn4+Dg8JvTrf8Oli1bho6ODpMmTaJLly4EBgbi4uIibPWaN28uivjfQn7D9UuaNm36qyL7j5DJZJw5c0YUcwUFBURHRwvdfWFhIampqWzcuPF3R7V+SY8ePVi9ejVHjhzh1q1bFBQUoKqqipWVFdHR0axatYrly5fzww8/0LJlSxISEr4YcZITFhZGaWkprVq1EiPPEv/vcPnyZQoLCxk1ahSvXr36W49hamoqc+fOxdTUFDs7Oxo0aMC5c+ewtLQkICDgD58bHx/PoUOHqK6upk6dOvj5+X1R2D979kw0oCopKYkgvjt37vzqRv2XuLi4MGHCBDIyMnj06BEnT5781TqxsbHIZDKcnJx+t5hcs2YNV65cwdramgMHDjB79mx+/PFH/Pz8GDZsGPb29l9IqX6P69evk5iYiJub21efS+Ua+ydPnvDmzRshO/zaa5G/vz++vr5ERUXRsGFDLl26xPjx45HJZHTt2pWxY8dy7do1Tp8+/cW1e8GCBSgpKVFYWIiBgQGzZs0iNjaW8vJy3r9/z4QJE9izZw9btmyha9euf3rN+k9HPvuYk5Pzu8fyzp07PHjwAEdHx18pI34PuRRn79699OzZU8yK/pVr6ePHjwkPD+fevXtUV1f/ajb+f8q3335Lv379aNCgARMnTvzqmxU5ysrK3LhxA09PT9G78DW/A/gvL+zv379Pfn4+qqqqODg4/EsuqFu2bKFnz57Cj14eNf5no8cAb9684dKlS3h7ews7pp9z6dIlZDIZnp6e4iR048YNCgoKcHFxISIigsuXL6OlpSW067q6uowZMwYHB4f/9XsrKChARUXli7vMv8Iv9Y3yZND/DfJisUWLFri4uCCTyb7IDlBUVBQzKb+H3IUGPks4AgMDRQjVv5uzZ8+SkZFBWloajx49okePHqJ5SV1dnWPHjgEQHByMu7u7kGL07dv337bP69at4+7du0ydOpVhw4Zx8uRJmjRp8m/bn1+Sm5tLjRo1yMzMJC8v76tmXeR4eHiwYsUKZs6cycaNG+nTpw8hISFiNFnuqCEPlPktnj9/zvDhw+ncuTP3798X8o2fe1tL/Ocjt6kNCAj4qvP7XyEzMxMVFRUhLYqLi0NLS+uLIv23KCoq4tatW0RERJCQkEC/fv0YPHiwWN6+fXvGjx/PwoULef36NfPnzxc9SX9GXl4ewcHBqKmp0bdv3//xe27dujUrV67E1dWVunXrMnbsWBwcHH41qv6v4I809l9zLfL39yc/P5+MjAwCAgK4cuUKvr6++Pv7c/v2bR4/fkyrVq2E5EfOz2for169ipeX1xev9/TpU8LCwrC1tcXX1/d/9yb/AygrK8PPz0/IUTQ0NMRM8P8GX19f9PT0SExMZO3ataxfv56AgICv/swqKyvZsmUL3t7eODg4/OUm1T/i1KlTNGrUiPr16xMWFiZmBL4WZWVlSkpKiImJoXXr1lJh/0+hsrKSqKgowsPDuX//Pg0bNmTlypVf9Vx581Hnzp2JiooiJycHPz+/vzTC+TWsX78ee3t7OnTo8MXf/xR++OEHHj16JC4yvXv3/k3d7v8VSUlJX+iO5TRq1AgVFRViY2N/tUxDQ+MvnVD+bmQyGRcvXuTFixd4eXn9W/fl7+bQoUMsX75c9FE8e/aMmJiYvzRtHRgYiEwmIzMzk9q1a5OZmUnTpk2F17yExP+UmJgYXr16hbe3t+jh+Dny/rG5c+eKULI/kuDA59FXefOiHHV1dVq3bi006l9DcHAwOTk5JCUlkZaWhqOjIxEREZw9exZ9ff2v3s6/k9zcXIqLi4HPEquffvoJY2NjzM3N/8179p9FZWXlF72CKioqvyl7+qtkZWWJGzMjIyMePXqEjY3Nb37X/68JCgoiMTGRHj164Ovr+9VFuZxVq1bx3XffoaCgQFBQEFOmTJGkOP8EsrKyWL9+PaWlpdSoUQMfH58/lCz8X3Pz5k2+//57rKyscHd3Z9u2bWzatElYZf4TkGtQ5Tg4OPyq41/iv5vU1FQePnxIjRo1aN269R82aP8WM2bMwMTEBDc3N7S0tDh69CgmJibMmDHjX7THEhKfSUtLw9TUlPT09D/UcP+cqqoqHj9+/MVjr169Ys6cOTx//vyrXzs+Pp6SkpJfPS63EZSQ+H+doqIiTp8+zZ49e1i2bJnQ7/+rkQr7/2CysrJYt24dZWVlGBgY0L59+/+owl4e6mNmZoajoyM1a9b809EeCQmJL/mlzCIhIQE1NbWvkkNISEhISPzn8f79e44dO0ZUVBS2trZMmjTpf9Xw/1eQCvv/YCorK4mOjiY8PJyHDx/SoEGDv+xuIiEhISEhISEh8X/H4cOHsbCwoGXLln+rdv9rkAr7/2DevHnDlStX8Pb2/tMmKQkJCQkJCQkJif9upMJeQkJCQkJCQkJC4h/AX4uzkpCQkJCQkJCQkJD4j0Qq7CUkJCQkJCQkJCT+AUiFvYSEhISEhISEhMQ/AKmwl5CQkJCQkJCQkPgHIBX2EhISEv+lXLlyhZkzZ/67d+P/nIcPH9KuXbu/dZtVVVV07tyZd+/e/a3blZCQkPgrSIW9hISExH8grVu3Ji4u7l+2fZlMxowZM/Dz8/uXvcZ/KoWFhcTGxv6t21RSUsLd3Z0FCxb8rduVkJCQ+CtIhb2EhITEH5CRkcHMmTNp3749nTt3Zs2aNRQXF//LXzcoKAgbG5t/2fbDwsIoLy+nVatW4rFWrVqRmJj4L3vNv4P/i33s2rUrp06d+uJ/FxcXmjVrhpeXF4MGDWLv3r2UlpZ+8byhQ4dy5MgRcnNz/6X7JyEhIfF7SIW9hISExO9QUlJCy5Ytefr0KdOnT2fKlClkZWXRp0+ff/lrOzk5oa2t/S/b/rFjx+jatesXjz169IiioqJ/2Wv+Hfxf7GN8fDwfPnz44n8fHx+2bdvG0qVL8fT0ZMOGDTg5OX0hvTE3N8fe3p4zZ878S/dPQkJC4veQCnsJCQmJ3+HRo0e8ePGCo0eP4uvri4+PDytWrODkyZNincrKSoKCgujSpQudOnVi8+bNVFdXi+XHjx+nS5cutGnThoULF1JWVvZVy34pxSkuLmb+/Pm0b9+enj17cvz48S/2tUWLFly7do2xY8fi7e3NuHHj+Pjx4+++t8jISFxcXMT/AwcOpLS0lGHDhuHi4sLChQvFdq9evcqYMWPw9PQkPj4eFxeXLwrfe/fu4evr+9X7UlFRwfr16+natSt+fn6cO3dOLOvRowcuLi40b96cgQMH8ujRoz/dxz87Bnl5eUyZMoXWrVszbtw4MjMzf/dz+T3q1q2Li4sLLVu2ZPTo0URHR1NVVcW0adO+WK9Zs2bcvn37L29fQkJC4m9BJiEhISHxm7x8+VKmrKws2717t+zTp0+/uU63bt1knp6esgsXLsguXbokc3Z2ls2cOVMmk8lkkZGRMh0dHdnBgwdl4eHhssWLF8tmz579p8tkMplMT09Pdvv2bZlMJpNVV1fL3N3dZa1bt5aFhobK9u/fLzM0NJRt2rRJrK+mpiaztraWHT58WHbjxg1Zq1atZD169Pjd96aqqioLDw8X/z958kSmoaEh27dvn+z+/fuyly9fiu3WqVNHduDAAdm9e/dkHz58kAGy9PR08dywsDCZsbHxV+9L586dZU2bNpWFhITILl68KOvatassKSlJJpPJZAkJCbL79+/LoqKiZKtWrZJpa2vLXr169Yf7+EfHoLq6Wubq6irr2LGjLCwsTLZ9+3aZoaGhzNDQ8Hc/GwsLC9m2bdt+9385a9eulWlqasqqqqrEY4sWLZK1adPmd7ctISEh8a9EKuwlJCQk/oCTJ0/K7O3tZZqamjI3NzfZnDlzZJmZmTKZTCaLjo6WaWhoyPLz88X6CQkJMjU1NVllZaUsODhY1rBhQ1llZaVYXlZWJpPJZH+4TCb7srAPDw+Xqaury7KyssTyHTt2yExNTcX/ampqsh9//FH8f/PmTZmBgcFvvqeqqiqZgoKCLCoq6ovHtbS0ZHfv3v3iMTU1Ndnu3bvF/4WFhV9V2P/evkRFRcnU1NRkGRkZX7xOeXn5b+6rv7+/bNmyZb+7j392DG7evCnT0dGRFRQUiOWLFy/+Wwr7M2fOyABZdna2eGzlypWyFi1a/O62JSQkJP6VKP+bJwwkJCQk/qPp1asXvXr1IjMzk9jYWKGtTkxM5MmTJ8hkMtq0aSPWl8lklJeX8/r1a7p37861a9dwcHDA1dWVNm3a0LdvX4A/XPZLnj9/jrW1NYaGhuIxFxcXMjIyKCoqElr8evXqieU1atQgPz//N7enqKiIoaEhOTk5X/UZNG7c+KvW+zm/ty9PnjzB2toaExOTL9ZXVVUFIDw8nN27d/Pq1StKS0t5+/YtWlpav/s6f3YMkpOTsbW1RUdHRyz/5ptv/vL7+S2ys7NRUFD4Yv9ycnIwMjL6W7YvISEh8VeRCnsJCQmJr8DExARfX19atWqFtrY2t27dwsDAgBo1arB9+/ZfrW9qaoqKigo7duygqqqKhIQEVqxYwaFDh7h69eofLvslhoaGZGdnf/FYVlYW6urqf1j0/hFNmzblyZMndO7cWTymoKDwm+sqKSmJv5WVP182KisrxWOFhYVf/bpGRkZ8/PgRmUz2q9d78uQJfn5+LF26lNGjR6Opqcnq1au/6D345XP+7BgYGBj85mf3dxAaGoqjoyNqamrisfj4eDw8PP6W7UtISEj8VaTmWQkJCYnf4fHjx2zZsuULW8OLFy8ik8mwtLTE29tbrOfi4oKLiws2NjZcu3YNdXV1Lly4wO3bt1FSUsLJyQk/Pz/REPtHy36Jp6cnpaWl7Nq1C4DS0lJWrVpF586df7cY/zM6d+5MeHj4F4/p6+v/YcMtgLq6Oubm5ly7dk3sy86dO7/6db28vFBUVGTp0qXIZDIALl++zIsXL0hOTkZPT4/Ro0fTqlUratWqxd27d/9wH//sGLRq1YqcnByOHj0KfL4J2bx581fv729RXl5OUFAQJ06c+MK3/tOnT9y9e5dOnTr9r7YvISEh8T9FKuwlJCQkfgcLCwuePXuGqakpDRo0oG7duowZM4Y1a9bg7OxMjRo1OHPmDBs3bhTr2NvbY2BgAICdnR3z58/HxMQEe3t7Jk+ezKpVq/502S8xMjLiyJEjzJs3DxsbG8zMzKioqGDTpk3/4/fm7+/PnTt3vnC3GTlyJAMHDqRp06bCcea3WLVqFVOmTMHe3h5bW9tfyWr+iBo1anDq1CkOHDiAiYkJFhYWbNmyBSMjI9q3b4+FhQXm5uY0atQILy8v7Ozsvnj+L/fxz46BkZER+/btY+zYsdja2lKvXj3s7e3/4qcFy5cvx8XFhSZNmmBoaMjRo0c5d+4c3bp1E+tcvnwZGxsbmjZt+pe3LyEhIfF3oCCTD5lISEhISPwmlZWVpKamoqqqirm5OSoqKr9a5/379xQUFGBjY4OiouKvluXm5mJlZfWFbOOPlsXGxlKvXr0vvOyrqqp4+fIlOjo6GBsbf7GdR48e0aBBAzQ1NYHPI+lJSUl/qCefO3cuMpmM5cuXi8c+fPjA27dv0dfXx9LS8lfblVNYWMjbt2+xtramvLycFy9e4Ojo+Jf2JT09HSUlJWrXrv3F4y9evKC8vBxbW1vev39PRUUFlpaWv7uPP/8sf+8YlJeXk5aWhqWlJZ8+ffpif39JQkICJiYmQiufkJBAeXk5CgoKaGhoUKdOnS80+3I8PT2ZO3cuHTt2/M3tSkhISPyrkQp7CQkJif9SSkpKSEtLo2HDhv/uXfl/nqqqKuLj43F2dv5374qEhMR/MVJhLyEhISEhISEhIfEPQNLYS0hISEhISEhISPwDkAp7CQkJCQkJCQkJiX8AUmEvISEhISEhISEh8Q9AKuwlJCQkJCQkJCQk/gFIhb2EhISEhISEhITEPwCpsJeQkJCQkJCQkJD4ByAV9hISEhISEhISEhL/AKTCXkJCQkJCQkJCQuIfgFTYS0hISEhISEhISPwDkAp7CQkJCQkJCQkJiX8A/x8tUNktE1mTGgAAAABJRU5ErkJggg==",
      "text/plain": [
       "<Figure size 768x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "import scipy.cluster.hierarchy as hierarchy\n",
    "\n",
    "if n_sessions >= 3:\n",
    "    linked = hierarchy.linkage(cluster_features, method='ward')\n",
    "\n",
    "    plt.figure(figsize=(8, 5))\n",
    "    hierarchy.dendrogram(linked, labels=session_metrics['sessionId'].str[:8].tolist())\n",
    "    plt.xlabel(\"Session (truncated ID)\")\n",
    "    plt.ylabel(\"Distance\")\n",
    "    plt.title(\"Session clustering dendrogram\")\n",
    "    plt.xticks(rotation=90)\n",
    "    plt.tight_layout()\n",
    "    plt.show()\n",
    "else:\n",
    "    print(f\"Only {n_sessions} sessions - too few for a meaningful dendrogram.\")"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "e136bf46-a004-4b34-805d-a4c77c8a9840",
   "metadata": {},
   "source": [
    "### DBSCAN\n",
    "\n",
    "DBSCAN groups points by density rather than distance-to-center, and needs an `eps` (maximum neighbor distance) hyperparameter, estimated below via a k-nearest-neighbor distance plot and then used to fit the model. With this few sessions, don't be surprised if DBSCAN finds one giant cluster or labels most sessions as noise (`-1`) - that's a legitimate result given the data density, not a bug.\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 35,
   "id": "03517ccf-c7c6-40e7-81e7-9aa44d0b7ad6",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 576x384 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "from sklearn.cluster import DBSCAN\n",
    "from sklearn.neighbors import NearestNeighbors\n",
    "\n",
    "if n_sessions >= 5:\n",
    "    n_neighbors_for_eps = min(len(feature_cols) * 2, n_sessions - 1)  # lab 8's own rule of thumb: 2x dimensionality\n",
    "\n",
    "    neighbors = NearestNeighbors(n_neighbors=n_neighbors_for_eps).fit(cluster_features)\n",
    "    distances, _ = neighbors.kneighbors(cluster_features)\n",
    "    distances = np.sort(distances[:, -1])\n",
    "\n",
    "    plt.figure(figsize=(6, 4))\n",
    "    plt.plot(distances)\n",
    "    plt.xlabel(\"Session, sorted by distance\")\n",
    "    plt.ylabel(f\"Distance to {n_neighbors_for_eps}th neighbor\")\n",
    "    plt.title(\"DBSCAN eps estimation\")\n",
    "    plt.show()\n",
    "else:\n",
    "    print(f\"Only {n_sessions} sessions - too few to estimate a stable DBSCAN eps.\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 36,
   "id": "e9ad08ef-dc47-4fa3-a394-bbbe45663a6e",
   "metadata": {},
   "outputs": [],
   "source": [
    "eps = 1.5  # replace with whatever value the elbow above actually shows\n",
    "\n",
    "if n_sessions >= 5:\n",
    "    dbscan = DBSCAN(eps=eps, min_samples=n_neighbors_for_eps).fit(cluster_features)\n",
    "    session_metrics['dbscan_cluster'] = dbscan.labels_\n",
    "    session_metrics['dbscan_cluster'].value_counts()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "be8ff561-02ee-48eb-93b3-2e0efc221a78",
   "metadata": {},
   "source": [
    "### Anomaly Detection\n",
    "\n",
    "A different unsupervised question: not \"what groups exist,\" but \"which sessions don't look like the rest.\" Three algorithms - see [Tools and Techniques](#tools-and-techniques) - each flag their own outliers; sessions flagged by two or more are the more trustworthy calls, given how easily any single algorithm can be thrown off at this sample size.\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 37,
   "id": "266e8950-8354-442c-a7a3-8094c5b144f6",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Skipped Robust Covariance: The covariance matrix of the support data is equal to 0, try to increase support_fraction\n"
     ]
    }
   ],
   "source": [
    "from sklearn.covariance import EllipticEnvelope\n",
    "from sklearn.svm import OneClassSVM\n",
    "from sklearn.neighbors import LocalOutlierFactor\n",
    "\n",
    "if n_sessions >= 10:\n",
    "    outliers_fraction = 0.1\n",
    "    n_neighbors_lof = min(20, n_sessions - 1)\n",
    "\n",
    "    anomaly_algorithms = {\n",
    "        \"Robust Covariance\": EllipticEnvelope(contamination=outliers_fraction, random_state=42),\n",
    "        \"One-Class SVM\": OneClassSVM(nu=outliers_fraction, kernel=\"rbf\", gamma=\"scale\"),\n",
    "        \"Local Outlier Factor\": LocalOutlierFactor(n_neighbors=n_neighbors_lof, contamination=outliers_fraction),\n",
    "    }\n",
    "\n",
    "    results = pd.DataFrame({\"sessionId\": session_metrics[\"sessionId\"]})\n",
    "    for name, algorithm in anomaly_algorithms.items():\n",
    "        try:\n",
    "            if name == \"Local Outlier Factor\":\n",
    "                # LOF only supports fit_predict in this mode - no separate .fit() then .predict()\n",
    "                labels = algorithm.fit_predict(cluster_features)\n",
    "            else:\n",
    "                labels = algorithm.fit(cluster_features).predict(cluster_features)\n",
    "            results[name] = labels  # -1 = outlier, 1 = inlier\n",
    "        except ValueError as e:\n",
    "            # Robust Covariance needs real spread within its \"cleanest\" data subset to\n",
    "            # estimate a covariance matrix at all - real traffic with many near-identical\n",
    "            # bounce sessions (single pageview, no scroll, no clicks) can make that subset\n",
    "            # exactly degenerate. Skip just this algorithm rather than failing the cell.\n",
    "            print(f\"Skipped {name}: {e}\")\n",
    "\n",
    "    results\n",
    "else:\n",
    "    print(f\"Only {n_sessions} sessions - too few to fit anomaly detectors reliably (need >= 10).\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 38,
   "id": "6f13eaba-7f8d-4545-a050-c402af7f8d38",
   "metadata": {},
   "outputs": [],
   "source": [
    "if n_sessions >= 10:\n",
    "    algo_cols = [name for name in anomaly_algorithms if name in results.columns]\n",
    "    results[\"flagged_by\"] = (results[algo_cols] == -1).sum(axis=1)\n",
    "    results.sort_values(\"flagged_by\", ascending=False)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 39,
   "id": "f352c2ac-e6e1-4b2f-9633-76dead41ef68",
   "metadata": {},
   "outputs": [],
   "source": [
    "if n_sessions >= 10:\n",
    "    flagged = results[results[\"flagged_by\"] >= 2][\"sessionId\"]\n",
    "    session_metrics[session_metrics[\"sessionId\"].isin(flagged)][feature_cols + [\"contact_interest\"]]"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "b548a358-6c43-4c06-8f5c-c32792444d04",
   "metadata": {},
   "source": [
    "## Association Analysis: What Site Behaviors Go Together\n",
    "\n",
    "A third kind of question, distinct from both prediction and clustering: which site behaviors tend to co-occur within the same session? Each session becomes a \"basket\" of the distinct events it triggered (e.g. `section_viewed:contact`, `scroll_depth:100`), borrowing the same asymmetric-binary-attribute idea used for classic market-basket analysis (which products get bought together).\n",
    "\n",
    "### Building the Baskets\n",
    "\n",
    "Each session's events, deduplicated to presence/absence per distinct `event:property` combo - \"did this session ever do X,\" not how many times.\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 40,
   "id": "631ed1da-c2a6-4d81-b693-eec306feb447",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "273 sessions (transactions), 20 distinct items\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "['link_clicked:ayuda',\n",
       " 'link_clicked:email',\n",
       " 'link_clicked:exact',\n",
       " 'link_clicked:notes',\n",
       " 'link_clicked:old-site',\n",
       " 'link_clicked:planner',\n",
       " 'link_clicked:youtube',\n",
       " 'nav_clicked:about',\n",
       " 'nav_clicked:contact',\n",
       " 'nav_clicked:experience',\n",
       " 'nav_clicked:research',\n",
       " 'page_load',\n",
       " 'scroll_depth:100',\n",
       " 'scroll_depth:25',\n",
       " 'scroll_depth:50',\n",
       " 'scroll_depth:75',\n",
       " 'section_viewed:about',\n",
       " 'section_viewed:contact',\n",
       " 'section_viewed:experience',\n",
       " 'section_viewed:research']"
      ]
     },
     "execution_count": 40,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "def event_to_item(row):\n",
    "    event = row['event']\n",
    "    props = row['properties']\n",
    "    if isinstance(props, dict) and len(props) > 0:\n",
    "        key = next(iter(props))\n",
    "        return f\"{event}:{props[key]}\"\n",
    "    return event\n",
    "\n",
    "main_site['item'] = main_site.apply(event_to_item, axis=1)\n",
    "\n",
    "# one basket per session: the *set* of distinct items it triggered (order/repeats don't matter)\n",
    "transactions = main_site.groupby('sessionId')['item'].apply(lambda items: set(items)).tolist()\n",
    "all_items = sorted(set(item for t in transactions for item in t))\n",
    "\n",
    "print(f\"{len(transactions)} sessions (transactions), {len(all_items)} distinct items\")\n",
    "all_items"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "a8680396-9632-44ae-8b35-173c8ad990d0",
   "metadata": {},
   "source": [
    "### Frequent Itemsets\n",
    "\n",
    "With only a handful of distinct item types, a brute-force check of every combination up to size 3 is fast enough that a specialized algorithm like FP-growth isn't necessary.\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 41,
   "id": "8883fc7b-c54c-4f03-8ace-f664654fe9e4",
   "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>itemset</th>\n",
       "      <th>support_count</th>\n",
       "      <th>support</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>(page_load,)</td>\n",
       "      <td>269</td>\n",
       "      <td>0.985348</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>(section_viewed:about,)</td>\n",
       "      <td>101</td>\n",
       "      <td>0.369963</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>(page_load, section_viewed:about)</td>\n",
       "      <td>97</td>\n",
       "      <td>0.355311</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>(scroll_depth:25,)</td>\n",
       "      <td>57</td>\n",
       "      <td>0.208791</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                             itemset  support_count   support\n",
       "0                       (page_load,)            269  0.985348\n",
       "2            (section_viewed:about,)            101  0.369963\n",
       "3  (page_load, section_viewed:about)             97  0.355311\n",
       "1                 (scroll_depth:25,)             57  0.208791"
      ]
     },
     "execution_count": 41,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "import itertools\n",
    "\n",
    "min_support_count = max(2, len(transactions) // 5)  # tune this - see note below\n",
    "\n",
    "def itemset_support_count(itemset, transactions):\n",
    "    return sum(1 for t in transactions if set(itemset).issubset(t))\n",
    "\n",
    "frequent = []\n",
    "for size in [1, 2, 3]:\n",
    "    for combo in itertools.combinations(all_items, size):\n",
    "        count = itemset_support_count(combo, transactions)\n",
    "        if count >= min_support_count:\n",
    "            frequent.append({\"itemset\": combo, \"support_count\": count, \"support\": count / len(transactions)})\n",
    "\n",
    "frequent_df = pd.DataFrame(frequent).sort_values(\"support\", ascending=False)\n",
    "frequent_df"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "f41d6daa-108b-4e69-8c73-ca316bc00f94",
   "metadata": {},
   "source": [
    "### Association Rules\n",
    "\n",
    "For every frequent itemset, every way of splitting it into a left-hand and right-hand side becomes a candidate rule, kept if its confidence (how often the right-hand side follows given the left) clears a minimum threshold.\n",
    "\n",
    "**Read any rules found here skeptically** - confidence on a 15-session dataset is extremely noisy (one session flips a ratio by ~7 percentage points), so treat this as \"here's what the mechanism produces,\" not \"here's a confirmed finding,\" until real production data accumulates.\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 42,
   "id": "f12b7f7b-6c35-4562-8666-6bc8ac3cafe0",
   "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>left</th>\n",
       "      <th>right</th>\n",
       "      <th>support</th>\n",
       "      <th>confidence</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>(section_viewed:about,)</td>\n",
       "      <td>(page_load,)</td>\n",
       "      <td>0.355311</td>\n",
       "      <td>0.960396</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                      left         right   support  confidence\n",
       "0  (section_viewed:about,)  (page_load,)  0.355311    0.960396"
      ]
     },
     "execution_count": 42,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "support_lookup = {frozenset(row['itemset']): row['support'] for _, row in frequent_df.iterrows()}\n",
    "\n",
    "def proper_subsets(s):\n",
    "    s = list(s)\n",
    "    return itertools.chain.from_iterable(itertools.combinations(s, r) for r in range(1, len(s)))\n",
    "\n",
    "min_confidence = 0.6\n",
    "rules = []\n",
    "for _, row in frequent_df[frequent_df['itemset'].apply(len) > 1].iterrows():\n",
    "    itemset = row['itemset']\n",
    "    for left in proper_subsets(itemset):\n",
    "        left_key = frozenset(left)\n",
    "        if left_key not in support_lookup:\n",
    "            continue\n",
    "        right = frozenset(itemset) - left_key\n",
    "        confidence = row['support'] / support_lookup[left_key]\n",
    "        if confidence >= min_confidence:\n",
    "            rules.append({\"left\": left, \"right\": tuple(right), \"support\": row['support'], \"confidence\": confidence})\n",
    "\n",
    "rules_df = pd.DataFrame(rules).sort_values(\"confidence\", ascending=False)\n",
    "rules_df"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 43,
   "id": "0bc6bb76-2ea9-4a4d-ad5d-2b134cbaab55",
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3 (ipykernel)",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.11.15"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}