FinalReport.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "id": "227e441f-7475-4036-91a7-97e535953155",
   "metadata": {},
   "source": [
    "# Final Report\n",
    "#### Wei Kong\n",
    "#### Predicting Commercial Success in the Movie Industry"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "8b95b553-4162-43b1-930e-f4cd2ae3e5c3",
   "metadata": {},
   "source": [
    "Note: The code to produce the results below and later in the report are hidden. For the original code please refer to the code on github repository (https://github.com/kongwei981126/Movie-Industry/blob/Wei/Final%20Report.ipynb)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "49258a95-db12-4ca0-a85b-096de707dec4",
   "metadata": {
    "tags": []
   },
   "source": [
    "## Executive Summary:\n",
    "<!--TProvide the decision to be impacted, Movitation, Business value.-->\n",
    "We all know that movie industry is risky, but it can also bring enormous profit at the same time. That's why countless companies and people have been attracted to this industry. However, how risky is it? We all saw super hits like Titanic, and we all have seen big productions fail. There hasn't been an universal formula to make a popular movie, but it is always to our interest to choose the right movie to invest in. Personaly, I would also like to find out how to prevent the emergence of bad movies: that is how to identify them before they even started shooting.\n",
    "\n",
    "In this analysis, I aim to produce a study to figure out specific reasons that lead to the success of a movie economically, and give advices for companies or individuals who want to enter this industry. In other words, I decided to explore how to select a movie that will be successful economically to invest in for companies and individuls who want to make a profit in this field."
   ]
  },
  {
   "cell_type": "markdown",
   "id": "f52231ac-91ec-438c-9c90-dde2395e0b9b",
   "metadata": {},
   "source": [
    "The whole project was carried out in the following procedue:\n",
    "- Collecting data\n",
    "- Data Cleaning\n",
    "- Outlier Removal\n",
    "- Feature Engineering\n",
    "- Hyperparameter Tuning / Model Selection\n",
    "- Detailed Tuning of model\n",
    "- Analysis of results"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "8c13b270-e50a-4a4e-a7b8-2ac7f3a4bc70",
   "metadata": {
    "tags": []
   },
   "source": [
    "## Data description/preprocessing"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "a45dd9aa-59ed-4e1f-a4f1-850ffc10527a",
   "metadata": {},
   "source": [
    "For the first half semester, I have been using \"The Movies Dataset\" on Kaggle (https://www.kaggle.com/datasets/rounakbanik/the-movies-dataset?select=movies_metadata.csv). These files contain metadata for all 45,000 movies listed in the Full MovieLens Dataset. The dataset consists of movies released on or before July 2017. Data points include cast, crew, plot keywords, budget, revenue, posters, release dates, languages, production companies, countries, TMDB vote counts and vote averages. This dataset also has files containing 26 million ratings from 270,000 users for all 45,000 movies. \n",
    "\n",
    "However, this dataset consists of large amount of missing entires for budget and revenue. There were only around 6000 entries with both of these features, and a large amount of data with problem, new data was chosen for the study. For detailed information on this old dataset, please refer to https://github.com/kongwei981126/Movie-Industry/blob/Wei/midReport.ipynb in the github repository."
   ]
  },
  {
   "cell_type": "markdown",
   "id": "e6540512-165d-42cd-b1e3-7132c3a75f12",
   "metadata": {
    "tags": []
   },
   "source": [
    "### New Dataset\n",
    "The new dataset was gathered from The Numbers (https://www.the-numbers.com/), a website operated by Nash Information Services, LLC., consisting of detailed information (especially on the business aspect) about various movies. Code to get and assemble the dataset were recorded in file \"scraping info\" (https://github.com/kongwei981126/Movie-Industry/blob/Wei/scraping%20info.ipynb) and \"Scraping more\" (https://github.com/kongwei981126/Movie-Industry/blob/Wei/Scraping%20more.ipynb) in the github repository."
   ]
  },
  {
   "cell_type": "markdown",
   "id": "0025fb77-106a-4bb8-b83e-955fcf6eb90f",
   "metadata": {
    "tags": []
   },
   "source": [
    "### Overview on the dataset"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "id": "d30bf893-3075-4f89-abc0-b7ed9bc4b00e",
   "metadata": {},
   "outputs": [],
   "source": [
    "#Preparations\n",
    "#Import the libraries\n",
    "import pandas as pd\n",
    "import numpy as np\n",
    "import matplotlib.pyplot as plt\n",
    "from sklearn.model_selection import train_test_split\n",
    "from tqdm import tqdm\n",
    "import re\n",
    "import warnings\n",
    "\n",
    "warnings.filterwarnings('ignore')\n",
    "\n",
    "#Settings\n",
    "pd.set_option('display.max_columns', None)\n",
    "pd.set_option('display.float_format',  '{:,.2f}'.format)\n",
    "pd.set_option('display.precision', 2)\n",
    "\n",
    "#import the dataset\n",
    "df = pd.read_csv('metadata.csv', low_memory=False, thousands=',')\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "id": "e0106c04-83dd-47f0-bbb6-0e1dfe4ff062",
   "metadata": {
    "tags": []
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Rows: 6382\n",
      "Colums: 35\n",
      "---------Info of dataset--------------\n",
      "<class 'pandas.core.frame.DataFrame'>\n",
      "RangeIndex: 6382 entries, 0 to 6381\n",
      "Data columns (total 35 columns):\n",
      " #   Column                            Non-Null Count  Dtype  \n",
      "---  ------                            --------------  -----  \n",
      " 0   moviename                         6382 non-null   object \n",
      " 1   Budget                            6382 non-null   int64  \n",
      " 2   link                              6382 non-null   object \n",
      " 3   Domestic Box Office               6382 non-null   float64\n",
      " 4   International Box Office          6382 non-null   float64\n",
      " 5   Worldwide Box Office              6382 non-null   float64\n",
      " 6   Domestic Release                  5967 non-null   object \n",
      " 7   Dom Year                          5967 non-null   float64\n",
      " 8   International Release             2942 non-null   object \n",
      " 9   Int Year                          2942 non-null   float64\n",
      " 10  Rating                            5983 non-null   object \n",
      " 11  Runtime                           5399 non-null   float64\n",
      " 12  Franchise                         1333 non-null   object \n",
      " 13  Keywords                          5242 non-null   object \n",
      " 14  Genre                             6224 non-null   object \n",
      " 15  Production Method                 6212 non-null   object \n",
      " 16  Creative Type                     6096 non-null   object \n",
      " 17  Production/Financing Companies    4050 non-null   object \n",
      " 18  Production Country                5887 non-null   object \n",
      " 19  Languages                         5388 non-null   object \n",
      " 20  Leading Cast                      4644 non-null   object \n",
      " 21  Supporting Cast                   4381 non-null   object \n",
      " 22  Production and Technical Credits  5769 non-null   object \n",
      " 23  Director                          5769 non-null   object \n",
      " 24  Est. Domestic DVD Sales           2418 non-null   float64\n",
      " 25  Est. Domestic Blu-ray Sales       2061 non-null   float64\n",
      " 26  Total Est. Domestic Video Sales   2672 non-null   float64\n",
      " 27  Lead Ensemble Members             704 non-null    object \n",
      " 28  Cameos                            583 non-null    object \n",
      " 29  Uncategorized Crew                1757 non-null   object \n",
      " 30  Narrator(s)                       81 non-null     object \n",
      " 31  Extras                            7 non-null      object \n",
      " 32  Documentary Subject(s)            26 non-null     object \n",
      " 33  Interviewee(s)                    26 non-null     object \n",
      " 34  Revenue                           6382 non-null   float64\n",
      "dtypes: float64(10), int64(1), object(24)\n",
      "memory usage: 1.7+ MB\n",
      "None\n"
     ]
    }
   ],
   "source": [
    "#some more information\n",
    "print('Rows: ' + str(df.shape[0]))\n",
    "print('Colums: ' + str(df.shape[1]))\n",
    "print('---------Info of dataset--------------')\n",
    "print(df.info())\n",
    "# print('---------Number of missing values----------')\n",
    "# print(df.isnull().sum())"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "id": "6544a309-714b-401f-a910-45e7d732f7c4",
   "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>moviename</th>\n",
       "      <th>Budget</th>\n",
       "      <th>link</th>\n",
       "      <th>Domestic Box Office</th>\n",
       "      <th>International Box Office</th>\n",
       "      <th>Worldwide Box Office</th>\n",
       "      <th>Domestic Release</th>\n",
       "      <th>Dom Year</th>\n",
       "      <th>International Release</th>\n",
       "      <th>Int Year</th>\n",
       "      <th>Rating</th>\n",
       "      <th>Runtime</th>\n",
       "      <th>Franchise</th>\n",
       "      <th>Keywords</th>\n",
       "      <th>Genre</th>\n",
       "      <th>Production Method</th>\n",
       "      <th>Creative Type</th>\n",
       "      <th>Production/Financing Companies</th>\n",
       "      <th>Production Country</th>\n",
       "      <th>Languages</th>\n",
       "      <th>Leading Cast</th>\n",
       "      <th>Supporting Cast</th>\n",
       "      <th>Production and Technical Credits</th>\n",
       "      <th>Director</th>\n",
       "      <th>Est. Domestic DVD Sales</th>\n",
       "      <th>Est. Domestic Blu-ray Sales</th>\n",
       "      <th>Total Est. Domestic Video Sales</th>\n",
       "      <th>Lead Ensemble Members</th>\n",
       "      <th>Cameos</th>\n",
       "      <th>Uncategorized Crew</th>\n",
       "      <th>Narrator(s)</th>\n",
       "      <th>Extras</th>\n",
       "      <th>Documentary Subject(s)</th>\n",
       "      <th>Interviewee(s)</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>The Super Mario Bros. Movie</td>\n",
       "      <td>100000000</td>\n",
       "      <td>Super-Mario-Bros-Movie-The-(2022)#tab=summary</td>\n",
       "      <td>260,823,700.00</td>\n",
       "      <td>248,398,143.00</td>\n",
       "      <td>509,221,843.00</td>\n",
       "      <td>April 5th, 2023</td>\n",
       "      <td>2,023.00</td>\n",
       "      <td>April 4th, 2023</td>\n",
       "      <td>2,023.00</td>\n",
       "      <td>PG</td>\n",
       "      <td>92.00</td>\n",
       "      <td>NaN</td>\n",
       "      <td>['Ensemble', 'Prince/Princess', 'Royalty', 'Go...</td>\n",
       "      <td>Action</td>\n",
       "      <td>Digital Animation</td>\n",
       "      <td>Kids Fiction</td>\n",
       "      <td>['Universal Pictures', 'Nintendo', 'Illuminati...</td>\n",
       "      <td>['Japan', 'United States']</td>\n",
       "      <td>['English']</td>\n",
       "      <td>['Chris Pratt', 'Anya Taylor-Joy', 'Charlie Da...</td>\n",
       "      <td>['Keegan-Michael Key', 'Seth Rogen', 'Fred Arm...</td>\n",
       "      <td>['Aaron Horvath', 'Michael Jelenic', 'Matthew ...</td>\n",
       "      <td>['Aaron Horvath', 'Michael Jelenic']</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                     moviename     Budget  \\\n",
       "0  The Super Mario Bros. Movie  100000000   \n",
       "\n",
       "                                            link  Domestic Box Office  \\\n",
       "0  Super-Mario-Bros-Movie-The-(2022)#tab=summary       260,823,700.00   \n",
       "\n",
       "   International Box Office  Worldwide Box Office Domestic Release  Dom Year  \\\n",
       "0            248,398,143.00        509,221,843.00  April 5th, 2023  2,023.00   \n",
       "\n",
       "  International Release  Int Year Rating  Runtime Franchise  \\\n",
       "0       April 4th, 2023  2,023.00     PG    92.00       NaN   \n",
       "\n",
       "                                            Keywords   Genre  \\\n",
       "0  ['Ensemble', 'Prince/Princess', 'Royalty', 'Go...  Action   \n",
       "\n",
       "   Production Method Creative Type  \\\n",
       "0  Digital Animation  Kids Fiction   \n",
       "\n",
       "                      Production/Financing Companies  \\\n",
       "0  ['Universal Pictures', 'Nintendo', 'Illuminati...   \n",
       "\n",
       "           Production Country    Languages  \\\n",
       "0  ['Japan', 'United States']  ['English']   \n",
       "\n",
       "                                        Leading Cast  \\\n",
       "0  ['Chris Pratt', 'Anya Taylor-Joy', 'Charlie Da...   \n",
       "\n",
       "                                     Supporting Cast  \\\n",
       "0  ['Keegan-Michael Key', 'Seth Rogen', 'Fred Arm...   \n",
       "\n",
       "                    Production and Technical Credits  \\\n",
       "0  ['Aaron Horvath', 'Michael Jelenic', 'Matthew ...   \n",
       "\n",
       "                               Director  Est. Domestic DVD Sales  \\\n",
       "0  ['Aaron Horvath', 'Michael Jelenic']                      NaN   \n",
       "\n",
       "   Est. Domestic Blu-ray Sales  Total Est. Domestic Video Sales  \\\n",
       "0                          NaN                              NaN   \n",
       "\n",
       "  Lead Ensemble Members Cameos Uncategorized Crew Narrator(s) Extras  \\\n",
       "0                   NaN    NaN                NaN         NaN    NaN   \n",
       "\n",
       "  Documentary Subject(s) Interviewee(s)  \n",
       "0                    NaN            NaN  "
      ]
     },
     "execution_count": 7,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.head(1)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "1c95b3ed-325d-45a1-94fd-82baee2b8d8a",
   "metadata": {},
   "source": [
    "In this dataset, we have 6382 observations across 34 features. The output above also mentioned the amount of null-values in each feature. However, some of them need futher processing before any other action can be taken. The following part will go through each of the feature in the dataset."
   ]
  },
  {
   "cell_type": "markdown",
   "id": "6122933c-f56d-42cf-bae5-a9f9ee88db89",
   "metadata": {},
   "source": [
    "#### Features removed"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "0cdf7ecc-ad9c-4702-8aac-58b1b7dfb4ce",
   "metadata": {},
   "source": [
    "These features were removed due to inability to incorporate NLP analysis in this analysis. \n",
    "- moviename: the name of the movie.\n",
    "- Keywords: keywords of the content of the movie."
   ]
  },
  {
   "cell_type": "markdown",
   "id": "02fed4e2-ab83-4888-8872-02da7f383bf4",
   "metadata": {},
   "source": [
    "These features were removed due to not sufficient information included in them.\n",
    "- link: the link to the information page of the movie\n",
    "- Production Country: Most of the movies were made in the United States, causing this feature to be not useful. (Refer to table below)\n",
    "- Languages: Most of the moview were in english, causing this feature to be not useful. (Refer to table below)\n",
    "- Production/Financing Companies: The large number of companies in this category made it hard to use.\n",
    "- Supporting Cast\n",
    "- Production and Technical Credits"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "id": "fad98beb-8104-4f46-a34b-f19e405952fa",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Top of Frequency table of Production Country\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "['United States']                      4325\n",
       "['United Kingdom']                      203\n",
       "['United Kingdom', 'United States']     179\n",
       "['France']                               97\n",
       "['Canada']                               59\n",
       "Name: Production Country, dtype: int64"
      ]
     },
     "execution_count": 20,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "print(\"Top of Frequency table of Production Country\")\n",
    "df['Production Country'].value_counts().head(5)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "id": "f0342494-12e1-4825-96ea-c05ff0d0f0f5",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Top of Frequency table of Language of the movies\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "['English']               4620\n",
       "['French']                  72\n",
       "['English', 'Spanish']      64\n",
       "Name: Languages, dtype: int64"
      ]
     },
     "execution_count": 19,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "print(\"Top of Frequency table of Language of the movies\")\n",
    "df['Languages'].value_counts().head(3)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "3c63ae69-8699-44fc-b52f-14554bc34e6d",
   "metadata": {},
   "source": [
    "These features were removed due to large amount of null values (refer to the data overview outputs above) and inability to impute them. \n",
    "- Est. Domestic DVD Sales: Estimated domestic DVD sales\n",
    "- Est. Domestic Blu-ray Sales: Estimated domestic Blu-ray sales\n",
    "- Total Est. Domestic Video Sales: Total estimated video sales\n",
    "- Cameos\n",
    "- Uncategorized Crew\n",
    "- Narrator(s)\n",
    "- Extras\n",
    "- Documentary Subject(s)\n",
    "- Interviewee(s)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "a35beb63-0a62-4a75-892a-15dfd3132d34",
   "metadata": {},
   "source": [
    "#### Features merged"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "2b26d73e-12ec-4689-912a-de37d288bb6a",
   "metadata": {},
   "source": [
    "The following features were merged into two new features: Realease year and month. Here, whichever earlier between domestic and international release (if differs) is taken as the final result:\n",
    "- Domestic Release: Domestic release date\n",
    "- Dom Year: Domestic release year\n",
    "- International Release: International release year\n",
    "- Int Year: International release year."
   ]
  },
  {
   "cell_type": "markdown",
   "id": "3bc1de59-08e9-46ad-8b4d-93af7bb04673",
   "metadata": {
    "tags": []
   },
   "source": [
    "The following features were merged into one: leading cast. These two names were used interchangably for different movies:\n",
    "- Leading Cast\n",
    "- Lead Ensemble Members"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "d0ca7517-0c73-4d43-bafb-7d7f8796dac1",
   "metadata": {},
   "source": [
    "The following features were merged into one: Revenue (The total boxoffice of a movie, which is what we aim to investigate in this report)\n",
    "- Domestic Box Office\n",
    "- International Box Office\n",
    "- Worldwide Box Office"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "425ba772-b689-43cf-bb7f-2cf8381a69f5",
   "metadata": {},
   "source": [
    "#### Features changed\n",
    "- Franchise: Due to the large amount of franchises out there, this feature is turned into a binary True or False indicating if a the movie is in a Franchise."
   ]
  },
  {
   "cell_type": "markdown",
   "id": "686157f6-20a3-4cb4-a135-64d35106e39f",
   "metadata": {},
   "source": [
    "#### Features unchanged\n",
    "- Rating: Motion Picture Association film rating system rating of the movie\n",
    "- Runtime: Length of movie (in minutes)\n",
    "- Budget\n",
    "- Production Method: 'Live Action', 'Digital Animation', 'Animation/Live Action', 'Hand Animation', 'Stop-Motion Animation', 'Rotoscoping', 'Multiple Production Methods'\n",
    "- Creative Type: 'Contemporary Fiction', 'Historical Fiction', 'Dramatization', 'Science Fiction', 'Fantasy', 'Kids Fiction', 'Factual', 'Super Hero', 'Multiple Creative Types'\n",
    "- Franchise\n",
    "- Director"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "id": "4b574eee-e6bc-4e74-a148-e817f0fb7717",
   "metadata": {},
   "outputs": [],
   "source": [
    "##################################################################\n",
    "###                 Code Accomplishing the above               ###\n",
    "##################################################################"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 59,
   "id": "69fcf657-217f-4b6d-a40d-4ad5abdafd0a",
   "metadata": {},
   "outputs": [],
   "source": [
    "# Fix some entires\n",
    "df['Domestic Box Office'] = df['Domestic Box Office'].fillna(0)\n",
    "df['International Box Office'] = df['International Box Office'].fillna(0)\n",
    "df['Worldwide Box Office'] = df['Worldwide Box Office'].fillna(0)\n",
    "\n",
    "df['Revenue'] = np.maximum(df['Domestic Box Office'] + df['International Box Office'], df['Worldwide Box Office'])\n",
    "for index, row in df.iterrows():\n",
    "    if pd.isna(row['Leading Cast']):\n",
    "        df.at[index, 'Leading Cast'] = df.at[index, 'Lead Ensemble Members']\n",
    "        \n",
    "# Subsetting rows with both Budget and Revenue\n",
    "df_sub = df.loc[(df['Revenue'] != 0) & (df['Budget'] != 0)].copy()\n",
    "# Compute ROI\n",
    "df_sub['ROI'] = (df_sub['Revenue']-df_sub['Budget'])/df_sub['Budget']*100"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 60,
   "id": "9e4876f6-83c6-42e1-887a-2c5b005894ce",
   "metadata": {},
   "outputs": [],
   "source": [
    "mon = ['January', 'February', 'March', 'April', 'May', 'June', 'July', 'August', 'September', 'October', 'November', 'December']\n",
    "dom_rel_month = pd.Series([])\n",
    "for index, row in df_sub['Domestic Release'].to_frame().iterrows():\n",
    "    d = \"\"\n",
    "    if not pd.isna(row[0]):\n",
    "        d = re.findall(r\"([a-zA-Z]+)\", row[0])\n",
    "        if d != [] and d[0] in mon:\n",
    "            t = d[0]\n",
    "        else:\n",
    "            t = np.nan\n",
    "    else: \n",
    "        t = np.nan\n",
    "    with warnings.catch_warnings():\n",
    "        warnings.simplefilter(action = 'ignore', category = FutureWarning)\n",
    "        dom_rel_month = dom_rel_month.append(pd.Series(data = t))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 61,
   "id": "3ea20b9c-0d4e-469c-a2f7-ee9e23788259",
   "metadata": {
    "tags": []
   },
   "outputs": [],
   "source": [
    "int_rel_month = pd.Series([])\n",
    "for index, row in df_sub['International Release'].to_frame().iterrows():\n",
    "    d = \"\"\n",
    "    if not pd.isna(row[0]):\n",
    "        d = re.findall(r\"([a-zA-Z]+)\", row[0])\n",
    "        if d != [] and d[0] in mon:\n",
    "            t = d[0]\n",
    "        else:\n",
    "            t = np.nan\n",
    "    else: \n",
    "        t = np.nan\n",
    "    with warnings.catch_warnings():\n",
    "        warnings.simplefilter(action = 'ignore', category = FutureWarning)\n",
    "        int_rel_month = int_rel_month.append(pd.Series(data = t))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 62,
   "id": "dd90c2a9-c1ac-448c-b49e-53a4bae0e43b",
   "metadata": {},
   "outputs": [],
   "source": [
    "df_sub['int_rel_month'] = int_rel_month.to_list()\n",
    "df_sub['dom_rel_month'] = dom_rel_month.to_list()\n",
    "df_sub['Year'] = ''\n",
    "df_sub['Month'] = ''\n",
    "\n",
    "from datetime import datetime\n",
    "\n",
    "for index, row in df_sub.iterrows():\n",
    "    if pd.isna(row['Dom Year']):\n",
    "        df_sub.at[index, 'Year'] = row['Int Year']\n",
    "        df_sub.at[index, 'Month'] = row['int_rel_month']\n",
    "    else: \n",
    "        # dom year not na\n",
    "        if pd.isna(row['Int Year']):\n",
    "            df_sub.at[index, 'Year'] = row['Dom Year']\n",
    "            df_sub.at[index, 'Month'] = row['dom_rel_month']\n",
    "        else:\n",
    "            #both year not na\n",
    "            if row['Dom Year'] == row['Int Year']:\n",
    "                df_sub.at[index, 'Year'] = row['Dom Year']\n",
    "                # month\n",
    "                if pd.isna(row['int_rel_month']):\n",
    "                    df_sub.at[index, 'Month'] = row['dom_rel_month']\n",
    "                elif pd.isna(row['dom_rel_month']):\n",
    "                    df_sub.at[index, 'Month'] = row['int_rel_month']\n",
    "                else:\n",
    "                    #compare month\n",
    "                    i = datetime.strptime(row['int_rel_month'], '%B').month\n",
    "                    d = datetime.strptime(row['dom_rel_month'], '%B').month\n",
    "                    if i <= d:\n",
    "                        df_sub.at[index, 'Month'] = row['int_rel_month']\n",
    "                    else:\n",
    "                        df_sub.at[index, 'Month'] = row['dom_rel_month']\n",
    "            elif row['Dom Year'] < row['Int Year']:\n",
    "                df_sub.at[index, 'Year'] = row['Dom Year']\n",
    "                df_sub.at[index, 'Month'] = row['dom_rel_month']\n",
    "            else: \n",
    "                df_sub.at[index, 'Year'] = row['Int Year']\n",
    "                df_sub.at[index, 'Month'] = row['int_rel_month']\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 63,
   "id": "a85b060a-68a3-477c-95cb-fe174bfdd687",
   "metadata": {},
   "outputs": [],
   "source": [
    "df_sub['int_rel_month'] = int_rel_month.to_list()\n",
    "df_sub['dom_rel_month'] = dom_rel_month.to_list()\n",
    "df_sub['Year'] = ''\n",
    "df_sub['Month'] = ''\n",
    "\n",
    "from datetime import datetime\n",
    "\n",
    "for index, row in df_sub.iterrows():\n",
    "    if pd.isna(row['Dom Year']):\n",
    "        df_sub.at[index, 'Year'] = row['Int Year']\n",
    "        df_sub.at[index, 'Month'] = row['int_rel_month']\n",
    "    else: \n",
    "        # dom year not na\n",
    "        if pd.isna(row['Int Year']):\n",
    "            df_sub.at[index, 'Year'] = row['Dom Year']\n",
    "            df_sub.at[index, 'Month'] = row['dom_rel_month']\n",
    "        else:\n",
    "            #both year not na\n",
    "            if row['Dom Year'] == row['Int Year']:\n",
    "                df_sub.at[index, 'Year'] = row['Dom Year']\n",
    "                # month\n",
    "                if pd.isna(row['int_rel_month']):\n",
    "                    df_sub.at[index, 'Month'] = row['dom_rel_month']\n",
    "                elif pd.isna(row['dom_rel_month']):\n",
    "                    df_sub.at[index, 'Month'] = row['int_rel_month']\n",
    "                else:\n",
    "                    #compare month\n",
    "                    i = datetime.strptime(row['int_rel_month'], '%B').month\n",
    "                    d = datetime.strptime(row['dom_rel_month'], '%B').month\n",
    "                    if i <= d:\n",
    "                        df_sub.at[index, 'Month'] = row['int_rel_month']\n",
    "                    else:\n",
    "                        df_sub.at[index, 'Month'] = row['dom_rel_month']\n",
    "            elif row['Dom Year'] < row['Int Year']:\n",
    "                df_sub.at[index, 'Year'] = row['Dom Year']\n",
    "                df_sub.at[index, 'Month'] = row['dom_rel_month']\n",
    "            else: \n",
    "                df_sub.at[index, 'Year'] = row['Int Year']\n",
    "                df_sub.at[index, 'Month'] = row['int_rel_month']\n",
    "            "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 64,
   "id": "882324f1-5b1f-4c22-b638-43ccc58f4987",
   "metadata": {},
   "outputs": [],
   "source": [
    "df_sub['is_Franchise'] = np.where(pd.isna(df_sub['Franchise']), False, True)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "fbb659be-4d73-433d-8c3c-6b462f2faebb",
   "metadata": {},
   "source": [
    "### Outlier detection"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "fc74c91d-a762-403b-b292-2b4d41fb4d61",
   "metadata": {},
   "source": [
    "In this part, a new variable was introduced to measure the economic return of movie. ROI was conputed by (Revenue - Budget)/Budget * 100%. Following are some information on distribution of ROI"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 46,
   "id": "d48740ff-a7e8-4ed3-9814-4ab60c5b5db2",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Descriptive Statistics of ROI\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "count     6,053.00\n",
       "mean        391.31\n",
       "std       2,895.68\n",
       "min        -100.00\n",
       "25%         -36.71\n",
       "50%          82.06\n",
       "75%         292.78\n",
       "max     179,900.00\n",
       "Name: ROI, dtype: float64"
      ]
     },
     "execution_count": 46,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "print(\"Descriptive Statistics of ROI\")\n",
    "df_sub['ROI'].describe()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 48,
   "id": "a04b96fa-e0f3-498e-8146-7a0254003a4e",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Lowest ROI movies\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>moviename</th>\n",
       "      <th>Budget</th>\n",
       "      <th>link</th>\n",
       "      <th>Domestic Box Office</th>\n",
       "      <th>International Box Office</th>\n",
       "      <th>Worldwide Box Office</th>\n",
       "      <th>Domestic Release</th>\n",
       "      <th>Dom Year</th>\n",
       "      <th>International Release</th>\n",
       "      <th>Int Year</th>\n",
       "      <th>Rating</th>\n",
       "      <th>Runtime</th>\n",
       "      <th>Franchise</th>\n",
       "      <th>Keywords</th>\n",
       "      <th>Genre</th>\n",
       "      <th>Production Method</th>\n",
       "      <th>Creative Type</th>\n",
       "      <th>Production/Financing Companies</th>\n",
       "      <th>Production Country</th>\n",
       "      <th>Languages</th>\n",
       "      <th>Leading Cast</th>\n",
       "      <th>Supporting Cast</th>\n",
       "      <th>Production and Technical Credits</th>\n",
       "      <th>Director</th>\n",
       "      <th>Est. Domestic DVD Sales</th>\n",
       "      <th>Est. Domestic Blu-ray Sales</th>\n",
       "      <th>Total Est. Domestic Video Sales</th>\n",
       "      <th>Lead Ensemble Members</th>\n",
       "      <th>Cameos</th>\n",
       "      <th>Uncategorized Crew</th>\n",
       "      <th>Narrator(s)</th>\n",
       "      <th>Extras</th>\n",
       "      <th>Documentary Subject(s)</th>\n",
       "      <th>Interviewee(s)</th>\n",
       "      <th>Revenue</th>\n",
       "      <th>ROI</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>5892</th>\n",
       "      <td>Kirikou Et Les Hommes Et Les Femmes</td>\n",
       "      <td>7500000</td>\n",
       "      <td>Kirikou-Et-Les-Hommes-Et-Les-Femmes-(France)(2...</td>\n",
       "      <td>17.00</td>\n",
       "      <td>0.00</td>\n",
       "      <td>17.00</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>July 3rd, 2015</td>\n",
       "      <td>2,015.00</td>\n",
       "      <td>NaN</td>\n",
       "      <td>88.00</td>\n",
       "      <td>['Kirikou']</td>\n",
       "      <td>NaN</td>\n",
       "      <td>Adventure</td>\n",
       "      <td>Digital Animation</td>\n",
       "      <td>Kids Fiction</td>\n",
       "      <td>NaN</td>\n",
       "      <td>['France']</td>\n",
       "      <td>['French']</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>['Michel Ocelot']</td>\n",
       "      <td>['Michel Ocelot']</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>17.00</td>\n",
       "      <td>-100.00</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>16</th>\n",
       "      <td>Capricorn One</td>\n",
       "      <td>5000000</td>\n",
       "      <td>Capricorn-One#tab=summary</td>\n",
       "      <td>401.00</td>\n",
       "      <td>0.00</td>\n",
       "      <td>401.00</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>December 10th, 1977</td>\n",
       "      <td>1,977.00</td>\n",
       "      <td>PG</td>\n",
       "      <td>123.00</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>Action</td>\n",
       "      <td>Live Action</td>\n",
       "      <td>Science Fiction</td>\n",
       "      <td>NaN</td>\n",
       "      <td>['United Kingdom', 'United States']</td>\n",
       "      <td>['English']</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>['Peter Hyams']</td>\n",
       "      <td>['Peter Hyams']</td>\n",
       "      <td>NaN</td>\n",
       "      <td>114,814.00</td>\n",
       "      <td>114,814.00</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>['Hal Holbrook', 'O.J. Simpson', 'James Brolin']</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>401.00</td>\n",
       "      <td>-99.99</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4990</th>\n",
       "      <td>Pandaemonium</td>\n",
       "      <td>15000000</td>\n",
       "      <td>Pandaemonium#tab=summary</td>\n",
       "      <td>1,438.00</td>\n",
       "      <td>0.00</td>\n",
       "      <td>1,438.00</td>\n",
       "      <td>June 29th, 2001</td>\n",
       "      <td>2,001.00</td>\n",
       "      <td>July 15th, 2005</td>\n",
       "      <td>2,005.00</td>\n",
       "      <td>PG-13</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>['United States']</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>['Andy de la Tour']</td>\n",
       "      <td>['Julien Temple', 'Frank Cottrell Boyce', 'Wil...</td>\n",
       "      <td>['Julien Temple']</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>['Samantha Morton', 'Dexter Fletcher', 'Emily ...</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>1,438.00</td>\n",
       "      <td>-99.99</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6012</th>\n",
       "      <td>Perrier’s Bounty</td>\n",
       "      <td>6600000</td>\n",
       "      <td>Perriers-Bounty#tab=summary</td>\n",
       "      <td>828.00</td>\n",
       "      <td>0.00</td>\n",
       "      <td>0.00</td>\n",
       "      <td>May 21st, 2010</td>\n",
       "      <td>2,010.00</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>R</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>['Ireland']</td>\n",
       "      <td>NaN</td>\n",
       "      <td>['Cillian Murphy', 'Jodie Whittaker', 'Jim Bro...</td>\n",
       "      <td>NaN</td>\n",
       "      <td>['Ian Fitzgibbon', 'Mark Holt']</td>\n",
       "      <td>['Ian Fitzgibbon']</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>['Brendan Gleeson']</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>828.00</td>\n",
       "      <td>-99.99</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>15</th>\n",
       "      <td>Born of War</td>\n",
       "      <td>5000000</td>\n",
       "      <td>Born-of-War#tab=summary</td>\n",
       "      <td>671.00</td>\n",
       "      <td>0.00</td>\n",
       "      <td>671.00</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>R</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>['Direct-to-Video', 'Revenge', 'Terrorism', 'I...</td>\n",
       "      <td>Action</td>\n",
       "      <td>Live Action</td>\n",
       "      <td>Historical Fiction</td>\n",
       "      <td>NaN</td>\n",
       "      <td>['Cyprus', 'Jordan', 'United Kingdom']</td>\n",
       "      <td>['English']</td>\n",
       "      <td>[\"Sofia Black-D'Elia\"]</td>\n",
       "      <td>['James Frain']</td>\n",
       "      <td>['Vicky Jewson']</td>\n",
       "      <td>['Vicky Jewson']</td>\n",
       "      <td>6,397.00</td>\n",
       "      <td>NaN</td>\n",
       "      <td>6,397.00</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>671.00</td>\n",
       "      <td>-99.99</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                                moviename    Budget  \\\n",
       "5892  Kirikou Et Les Hommes Et Les Femmes   7500000   \n",
       "16                          Capricorn One   5000000   \n",
       "4990                         Pandaemonium  15000000   \n",
       "6012                     Perrier’s Bounty   6600000   \n",
       "15                            Born of War   5000000   \n",
       "\n",
       "                                                   link  Domestic Box Office  \\\n",
       "5892  Kirikou-Et-Les-Hommes-Et-Les-Femmes-(France)(2...                17.00   \n",
       "16                            Capricorn-One#tab=summary               401.00   \n",
       "4990                           Pandaemonium#tab=summary             1,438.00   \n",
       "6012                        Perriers-Bounty#tab=summary               828.00   \n",
       "15                              Born-of-War#tab=summary               671.00   \n",
       "\n",
       "      International Box Office  Worldwide Box Office Domestic Release  \\\n",
       "5892                      0.00                 17.00              NaN   \n",
       "16                        0.00                401.00              NaN   \n",
       "4990                      0.00              1,438.00  June 29th, 2001   \n",
       "6012                      0.00                  0.00   May 21st, 2010   \n",
       "15                        0.00                671.00              NaN   \n",
       "\n",
       "      Dom Year International Release  Int Year Rating  Runtime    Franchise  \\\n",
       "5892       NaN        July 3rd, 2015  2,015.00    NaN    88.00  ['Kirikou']   \n",
       "16         NaN   December 10th, 1977  1,977.00     PG   123.00          NaN   \n",
       "4990  2,001.00       July 15th, 2005  2,005.00  PG-13      NaN          NaN   \n",
       "6012  2,010.00                   NaN       NaN      R      NaN          NaN   \n",
       "15         NaN                   NaN       NaN      R      NaN          NaN   \n",
       "\n",
       "                                               Keywords      Genre  \\\n",
       "5892                                                NaN  Adventure   \n",
       "16                                                  NaN     Action   \n",
       "4990                                                NaN        NaN   \n",
       "6012                                                NaN        NaN   \n",
       "15    ['Direct-to-Video', 'Revenge', 'Terrorism', 'I...     Action   \n",
       "\n",
       "      Production Method       Creative Type Production/Financing Companies  \\\n",
       "5892  Digital Animation        Kids Fiction                            NaN   \n",
       "16          Live Action     Science Fiction                            NaN   \n",
       "4990                NaN                 NaN                            NaN   \n",
       "6012                NaN                 NaN                            NaN   \n",
       "15          Live Action  Historical Fiction                            NaN   \n",
       "\n",
       "                          Production Country    Languages  \\\n",
       "5892                              ['France']   ['French']   \n",
       "16       ['United Kingdom', 'United States']  ['English']   \n",
       "4990                       ['United States']          NaN   \n",
       "6012                             ['Ireland']          NaN   \n",
       "15    ['Cyprus', 'Jordan', 'United Kingdom']  ['English']   \n",
       "\n",
       "                                           Leading Cast      Supporting Cast  \\\n",
       "5892                                                NaN                  NaN   \n",
       "16                                                  NaN                  NaN   \n",
       "4990                                                NaN  ['Andy de la Tour']   \n",
       "6012  ['Cillian Murphy', 'Jodie Whittaker', 'Jim Bro...                  NaN   \n",
       "15                               [\"Sofia Black-D'Elia\"]      ['James Frain']   \n",
       "\n",
       "                       Production and Technical Credits            Director  \\\n",
       "5892                                  ['Michel Ocelot']   ['Michel Ocelot']   \n",
       "16                                      ['Peter Hyams']     ['Peter Hyams']   \n",
       "4990  ['Julien Temple', 'Frank Cottrell Boyce', 'Wil...   ['Julien Temple']   \n",
       "6012                    ['Ian Fitzgibbon', 'Mark Holt']  ['Ian Fitzgibbon']   \n",
       "15                                     ['Vicky Jewson']    ['Vicky Jewson']   \n",
       "\n",
       "      Est. Domestic DVD Sales  Est. Domestic Blu-ray Sales  \\\n",
       "5892                      NaN                          NaN   \n",
       "16                        NaN                   114,814.00   \n",
       "4990                      NaN                          NaN   \n",
       "6012                      NaN                          NaN   \n",
       "15                   6,397.00                          NaN   \n",
       "\n",
       "      Total Est. Domestic Video Sales Lead Ensemble Members Cameos  \\\n",
       "5892                              NaN                   NaN    NaN   \n",
       "16                         114,814.00                   NaN    NaN   \n",
       "4990                              NaN                   NaN    NaN   \n",
       "6012                              NaN                   NaN    NaN   \n",
       "15                           6,397.00                   NaN    NaN   \n",
       "\n",
       "                                     Uncategorized Crew Narrator(s) Extras  \\\n",
       "5892                                                NaN         NaN    NaN   \n",
       "16     ['Hal Holbrook', 'O.J. Simpson', 'James Brolin']         NaN    NaN   \n",
       "4990  ['Samantha Morton', 'Dexter Fletcher', 'Emily ...         NaN    NaN   \n",
       "6012                                ['Brendan Gleeson']         NaN    NaN   \n",
       "15                                                  NaN         NaN    NaN   \n",
       "\n",
       "     Documentary Subject(s) Interviewee(s)  Revenue     ROI  \n",
       "5892                    NaN            NaN    17.00 -100.00  \n",
       "16                      NaN            NaN   401.00  -99.99  \n",
       "4990                    NaN            NaN 1,438.00  -99.99  \n",
       "6012                    NaN            NaN   828.00  -99.99  \n",
       "15                      NaN            NaN   671.00  -99.99  "
      ]
     },
     "execution_count": 48,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "print(\"Lowest ROI movies\")\n",
    "df_sub.sort_values(by = ['ROI']).head(5)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 51,
   "id": "b9687374-20aa-44e4-bed9-f6c067f64b9c",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Highest ROI movies\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>moviename</th>\n",
       "      <th>Budget</th>\n",
       "      <th>link</th>\n",
       "      <th>Domestic Box Office</th>\n",
       "      <th>International Box Office</th>\n",
       "      <th>Worldwide Box Office</th>\n",
       "      <th>Domestic Release</th>\n",
       "      <th>Dom Year</th>\n",
       "      <th>International Release</th>\n",
       "      <th>Int Year</th>\n",
       "      <th>Rating</th>\n",
       "      <th>Runtime</th>\n",
       "      <th>Franchise</th>\n",
       "      <th>Keywords</th>\n",
       "      <th>Genre</th>\n",
       "      <th>Production Method</th>\n",
       "      <th>Creative Type</th>\n",
       "      <th>Production/Financing Companies</th>\n",
       "      <th>Production Country</th>\n",
       "      <th>Languages</th>\n",
       "      <th>Leading Cast</th>\n",
       "      <th>Supporting Cast</th>\n",
       "      <th>Production and Technical Credits</th>\n",
       "      <th>Director</th>\n",
       "      <th>Est. Domestic DVD Sales</th>\n",
       "      <th>Est. Domestic Blu-ray Sales</th>\n",
       "      <th>Total Est. Domestic Video Sales</th>\n",
       "      <th>Lead Ensemble Members</th>\n",
       "      <th>Cameos</th>\n",
       "      <th>Uncategorized Crew</th>\n",
       "      <th>Narrator(s)</th>\n",
       "      <th>Extras</th>\n",
       "      <th>Documentary Subject(s)</th>\n",
       "      <th>Interviewee(s)</th>\n",
       "      <th>Revenue</th>\n",
       "      <th>ROI</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>1117</th>\n",
       "      <td>The Blair Witch Project</td>\n",
       "      <td>600000</td>\n",
       "      <td>Blair-Witch-Project-The#tab=summary</td>\n",
       "      <td>140,539,099.00</td>\n",
       "      <td>107,760,901.00</td>\n",
       "      <td>248,300,000.00</td>\n",
       "      <td>July 14th, 1999</td>\n",
       "      <td>1,999.00</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>R</td>\n",
       "      <td>81.00</td>\n",
       "      <td>['Blair Witch Project']</td>\n",
       "      <td>['Found Footage', 'Found footage horror']</td>\n",
       "      <td>Horror</td>\n",
       "      <td>Live Action</td>\n",
       "      <td>Contemporary Fiction</td>\n",
       "      <td>NaN</td>\n",
       "      <td>['United States']</td>\n",
       "      <td>['English']</td>\n",
       "      <td>['Heather Donahue', 'Joshua Leonard', 'Michael...</td>\n",
       "      <td>NaN</td>\n",
       "      <td>['Daniel Myrick', 'Eduardo Sanchez', 'Robin Co...</td>\n",
       "      <td>['Daniel Myrick', 'Eduardo Sanchez']</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>['Bob Griffith', 'Jim King', 'Sandra Sanchez',...</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>248,300,000.00</td>\n",
       "      <td>41,283.33</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1409</th>\n",
       "      <td>The Gallows</td>\n",
       "      <td>100000</td>\n",
       "      <td>Gallows-The#tab=summary</td>\n",
       "      <td>22,764,410.00</td>\n",
       "      <td>18,892,064.00</td>\n",
       "      <td>41,656,474.00</td>\n",
       "      <td>July 10th, 2015</td>\n",
       "      <td>2,015.00</td>\n",
       "      <td>July 9th, 2015</td>\n",
       "      <td>2,015.00</td>\n",
       "      <td>R</td>\n",
       "      <td>81.00</td>\n",
       "      <td>NaN</td>\n",
       "      <td>['Theater Actor', 'Accidental Death', 'Found F...</td>\n",
       "      <td>Horror</td>\n",
       "      <td>Live Action</td>\n",
       "      <td>Fantasy</td>\n",
       "      <td>['New Line Cinema', 'Blumhouse', 'Tremendum Pi...</td>\n",
       "      <td>['United States']</td>\n",
       "      <td>['English']</td>\n",
       "      <td>['Cassidy Gifford']</td>\n",
       "      <td>['Pfeifer Brown', 'Ryan Shoos', 'Travis Cluff'...</td>\n",
       "      <td>['Travis Cluff', 'Chris Lofing', 'Travis Cluff...</td>\n",
       "      <td>['Travis Cluff', 'Chris Lofing']</td>\n",
       "      <td>994,152.00</td>\n",
       "      <td>391,997.00</td>\n",
       "      <td>1,386,149.00</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>41,656,474.00</td>\n",
       "      <td>41,556.47</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1207</th>\n",
       "      <td>Paranormal Activity</td>\n",
       "      <td>450000</td>\n",
       "      <td>Paranormal-Activity#tab=summary</td>\n",
       "      <td>107,918,810.00</td>\n",
       "      <td>86,264,224.00</td>\n",
       "      <td>194,183,034.00</td>\n",
       "      <td>September 25th, 2009</td>\n",
       "      <td>2,009.00</td>\n",
       "      <td>December 3rd, 2009</td>\n",
       "      <td>2,009.00</td>\n",
       "      <td>R</td>\n",
       "      <td>96.00</td>\n",
       "      <td>['Paranormal Activity']</td>\n",
       "      <td>['Found Footage', 'Exorcism', 'Guerilla Filmma...</td>\n",
       "      <td>Horror</td>\n",
       "      <td>Live Action</td>\n",
       "      <td>Fantasy</td>\n",
       "      <td>['Blumhouse', 'Solana Films']</td>\n",
       "      <td>['United States']</td>\n",
       "      <td>['English']</td>\n",
       "      <td>['Micah Sloat', 'Katie Featherston']</td>\n",
       "      <td>['Mark Fredrichs', 'Amber Armstrong', 'Tim Pip...</td>\n",
       "      <td>['Oren Peli', 'Oren Peli', 'Oren Peli', 'Amir ...</td>\n",
       "      <td>['Oren Peli']</td>\n",
       "      <td>17,763,786.00</td>\n",
       "      <td>4,346,772.00</td>\n",
       "      <td>22,110,558.00</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>194,183,034.00</td>\n",
       "      <td>43,051.79</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1342</th>\n",
       "      <td>Mad Max</td>\n",
       "      <td>200000</td>\n",
       "      <td>Mad-Max#tab=summary</td>\n",
       "      <td>8,750,000.00</td>\n",
       "      <td>91,000,000.00</td>\n",
       "      <td>99,750,000.00</td>\n",
       "      <td>March 21st, 1980</td>\n",
       "      <td>1,980.00</td>\n",
       "      <td>April 12th, 1979</td>\n",
       "      <td>1,979.00</td>\n",
       "      <td>NaN</td>\n",
       "      <td>93.00</td>\n",
       "      <td>['Mad Max']</td>\n",
       "      <td>['Post Apocalypse', 'Revenge', 'Vigilante', 'G...</td>\n",
       "      <td>Action</td>\n",
       "      <td>Live Action</td>\n",
       "      <td>Science Fiction</td>\n",
       "      <td>['Roadshow Films']</td>\n",
       "      <td>['Australia']</td>\n",
       "      <td>['English']</td>\n",
       "      <td>['Mel Gibson']</td>\n",
       "      <td>['Joanne Samuel', 'Hugh Keays-Byrne', 'Steve B...</td>\n",
       "      <td>['George Miller', 'James McCausland', 'George ...</td>\n",
       "      <td>['George Miller']</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>99,750,000.00</td>\n",
       "      <td>49,775.00</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1486</th>\n",
       "      <td>Deep Throat</td>\n",
       "      <td>25000</td>\n",
       "      <td>Deep-Throat#tab=summary</td>\n",
       "      <td>45,000,000.00</td>\n",
       "      <td>0.00</td>\n",
       "      <td>0.00</td>\n",
       "      <td>June 30th, 1972</td>\n",
       "      <td>1,972.00</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NC-17</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>['United States']</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>77,509.00</td>\n",
       "      <td>77,509.00</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>45,000,000.00</td>\n",
       "      <td>179,900.00</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                    moviename  Budget                                 link  \\\n",
       "1117  The Blair Witch Project  600000  Blair-Witch-Project-The#tab=summary   \n",
       "1409              The Gallows  100000              Gallows-The#tab=summary   \n",
       "1207      Paranormal Activity  450000      Paranormal-Activity#tab=summary   \n",
       "1342                  Mad Max  200000                  Mad-Max#tab=summary   \n",
       "1486              Deep Throat   25000              Deep-Throat#tab=summary   \n",
       "\n",
       "      Domestic Box Office  International Box Office  Worldwide Box Office  \\\n",
       "1117       140,539,099.00            107,760,901.00        248,300,000.00   \n",
       "1409        22,764,410.00             18,892,064.00         41,656,474.00   \n",
       "1207       107,918,810.00             86,264,224.00        194,183,034.00   \n",
       "1342         8,750,000.00             91,000,000.00         99,750,000.00   \n",
       "1486        45,000,000.00                      0.00                  0.00   \n",
       "\n",
       "          Domestic Release  Dom Year International Release  Int Year Rating  \\\n",
       "1117       July 14th, 1999  1,999.00                   NaN       NaN      R   \n",
       "1409       July 10th, 2015  2,015.00        July 9th, 2015  2,015.00      R   \n",
       "1207  September 25th, 2009  2,009.00    December 3rd, 2009  2,009.00      R   \n",
       "1342      March 21st, 1980  1,980.00      April 12th, 1979  1,979.00    NaN   \n",
       "1486       June 30th, 1972  1,972.00                   NaN       NaN  NC-17   \n",
       "\n",
       "      Runtime                Franchise  \\\n",
       "1117    81.00  ['Blair Witch Project']   \n",
       "1409    81.00                      NaN   \n",
       "1207    96.00  ['Paranormal Activity']   \n",
       "1342    93.00              ['Mad Max']   \n",
       "1486      NaN                      NaN   \n",
       "\n",
       "                                               Keywords   Genre  \\\n",
       "1117          ['Found Footage', 'Found footage horror']  Horror   \n",
       "1409  ['Theater Actor', 'Accidental Death', 'Found F...  Horror   \n",
       "1207  ['Found Footage', 'Exorcism', 'Guerilla Filmma...  Horror   \n",
       "1342  ['Post Apocalypse', 'Revenge', 'Vigilante', 'G...  Action   \n",
       "1486                                                NaN     NaN   \n",
       "\n",
       "     Production Method         Creative Type  \\\n",
       "1117       Live Action  Contemporary Fiction   \n",
       "1409       Live Action               Fantasy   \n",
       "1207       Live Action               Fantasy   \n",
       "1342       Live Action       Science Fiction   \n",
       "1486               NaN                   NaN   \n",
       "\n",
       "                         Production/Financing Companies Production Country  \\\n",
       "1117                                                NaN  ['United States']   \n",
       "1409  ['New Line Cinema', 'Blumhouse', 'Tremendum Pi...  ['United States']   \n",
       "1207                      ['Blumhouse', 'Solana Films']  ['United States']   \n",
       "1342                                 ['Roadshow Films']      ['Australia']   \n",
       "1486                                                NaN  ['United States']   \n",
       "\n",
       "        Languages                                       Leading Cast  \\\n",
       "1117  ['English']  ['Heather Donahue', 'Joshua Leonard', 'Michael...   \n",
       "1409  ['English']                                ['Cassidy Gifford']   \n",
       "1207  ['English']               ['Micah Sloat', 'Katie Featherston']   \n",
       "1342  ['English']                                     ['Mel Gibson']   \n",
       "1486          NaN                                                NaN   \n",
       "\n",
       "                                        Supporting Cast  \\\n",
       "1117                                                NaN   \n",
       "1409  ['Pfeifer Brown', 'Ryan Shoos', 'Travis Cluff'...   \n",
       "1207  ['Mark Fredrichs', 'Amber Armstrong', 'Tim Pip...   \n",
       "1342  ['Joanne Samuel', 'Hugh Keays-Byrne', 'Steve B...   \n",
       "1486                                                NaN   \n",
       "\n",
       "                       Production and Technical Credits  \\\n",
       "1117  ['Daniel Myrick', 'Eduardo Sanchez', 'Robin Co...   \n",
       "1409  ['Travis Cluff', 'Chris Lofing', 'Travis Cluff...   \n",
       "1207  ['Oren Peli', 'Oren Peli', 'Oren Peli', 'Amir ...   \n",
       "1342  ['George Miller', 'James McCausland', 'George ...   \n",
       "1486                                                NaN   \n",
       "\n",
       "                                  Director  Est. Domestic DVD Sales  \\\n",
       "1117  ['Daniel Myrick', 'Eduardo Sanchez']                      NaN   \n",
       "1409      ['Travis Cluff', 'Chris Lofing']               994,152.00   \n",
       "1207                         ['Oren Peli']            17,763,786.00   \n",
       "1342                     ['George Miller']                      NaN   \n",
       "1486                                   NaN                      NaN   \n",
       "\n",
       "      Est. Domestic Blu-ray Sales  Total Est. Domestic Video Sales  \\\n",
       "1117                          NaN                              NaN   \n",
       "1409                   391,997.00                     1,386,149.00   \n",
       "1207                 4,346,772.00                    22,110,558.00   \n",
       "1342                          NaN                              NaN   \n",
       "1486                    77,509.00                        77,509.00   \n",
       "\n",
       "     Lead Ensemble Members Cameos  \\\n",
       "1117                   NaN    NaN   \n",
       "1409                   NaN    NaN   \n",
       "1207                   NaN    NaN   \n",
       "1342                   NaN    NaN   \n",
       "1486                   NaN    NaN   \n",
       "\n",
       "                                     Uncategorized Crew Narrator(s) Extras  \\\n",
       "1117  ['Bob Griffith', 'Jim King', 'Sandra Sanchez',...         NaN    NaN   \n",
       "1409                                                NaN         NaN    NaN   \n",
       "1207                                                NaN         NaN    NaN   \n",
       "1342                                                NaN         NaN    NaN   \n",
       "1486                                                NaN         NaN    NaN   \n",
       "\n",
       "     Documentary Subject(s) Interviewee(s)        Revenue        ROI  \n",
       "1117                    NaN            NaN 248,300,000.00  41,283.33  \n",
       "1409                    NaN            NaN  41,656,474.00  41,556.47  \n",
       "1207                    NaN            NaN 194,183,034.00  43,051.79  \n",
       "1342                    NaN            NaN  99,750,000.00  49,775.00  \n",
       "1486                    NaN            NaN  45,000,000.00 179,900.00  "
      ]
     },
     "execution_count": 51,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "print(\"Highest ROI movies\")\n",
    "df_sub.sort_values(by = ['ROI']).tail(5)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "7ca3da2b-9153-432a-bcd1-02d2e98af56f",
   "metadata": {},
   "source": [
    "Here, we do have some strange values, both with super large and small ROI. That being said, I decided to fact check the movies:\n",
    "For movies with low ROI, with evidence of ROI of -95% existing, -95% is used as a cutoff. \n",
    "For moview with high ROI, the one movie \"Deep Throat\" with incredibly high ROI is removed. \n",
    "No statistical outlier detection method were included to remove outliers since all movies are our subject for prediction."
   ]
  },
  {
   "cell_type": "markdown",
   "id": "5f8cfac3-c1a3-4f62-906c-090c631d9fb0",
   "metadata": {},
   "source": [
    "### More Feature Engineering"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "a5329bb6-506b-488d-a33b-eecff347f2dc",
   "metadata": {},
   "source": [
    "In analyzing Genre trend, we can see the following: The hotness of Genres change over time, so as the trend with actors and directors. Like shown in the following graph, the hotness of different genres rise and fall over time. "
   ]
  },
  {
   "cell_type": "markdown",
   "id": "e0e67ece-a19f-47f5-a432-ed28485a7714",
   "metadata": {},
   "source": [
    "For creation of the graph below, refer to the the Descritive.ipynb in the repository (https://github.com/kongwei981126/Movie-Industry/blob/Wei/Descriptive.ipynb)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "320475af-eeb8-4775-b3c9-b20811c6004c",
   "metadata": {},
   "source": [
    "<img src=\"img2.png\" width=100% />"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "eb367ff6-9540-4330-a244-3a239a4855ff",
   "metadata": {},
   "source": [
    "Thus, genre may not be a good feature to predict the performance of a movie. Instead, I propose the following method:"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "42a45e1e-40cc-447c-a19a-baf6fd4748a2",
   "metadata": {},
   "source": [
    "To make the explanation clear, use a movie released in 2023 as an example. As shown in the graph below, the production period of a movie usually takes 1 to 2 years, thus, we have to decide by 2020 if we want to make a movie that is gonna be released in 2023. Thus, only information till 2020 are considered available when considering the investment of a movie released in 2023. "
   ]
  },
  {
   "cell_type": "markdown",
   "id": "707b730f-272b-453c-bb77-52464aedbc86",
   "metadata": {},
   "source": [
    "<img src=\"img3.png\" width=100% />"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "bf7af1da-67b6-4306-a0d5-6975483bf3ff",
   "metadata": {},
   "source": [
    "Then, we consider a time window of y years till the time of decision. This time window of y years is used to measure the current trend in actors, directors, and genre. \n",
    "\n",
    "In each time window, the following three things are measured:\n",
    "- The total revenue of each actor in this y year period.\n",
    "- The total revenue of each director in this y year period.\n",
    "- The total revenue of each genre in this y year period.\n",
    "\n",
    "Then, from the information above, we define the top n actors and directors the \"star actor\" or \"star director\" at that time.\n",
    "After that, five new features were made:\n",
    "- The total revenue of lead actors in the time window beforehand.\n",
    "- The total revenue of directors in the time window beforehand.\n",
    "- The total revenue of the genre of the movie in the time window beforehand.\n",
    "- If there is a star actor in the leading cast by the time of production.\n",
    "- If there is a star director at the time of production."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "14b83760-1190-4786-9f35-9e1700821b37",
   "metadata": {},
   "outputs": [],
   "source": [
    "##################################################################\n",
    "###                 Code Accomplishing the above               ###\n",
    "##################################################################"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 65,
   "id": "844dabc3-89ee-4139-88b3-7016c7f09305",
   "metadata": {},
   "outputs": [],
   "source": [
    "dataset = df_sub[['moviename', 'Budget', 'Revenue', 'ROI', 'Year', 'Month',\n",
    "            'Rating', 'Franchise', 'is_Franchise', 'Keywords', 'Genre', 'Production Method',\n",
    "            'Creative Type', 'Production/Financing Companies', 'Leading Cast', 'Director']].copy()\n",
    "dataset = dataset[~dataset['Year'].isna()]\n",
    "dataset = dataset.loc[dataset['ROI'] < 50000]\n",
    "dataset = dataset.loc[dataset['ROI'] > -95]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "da4ade7a-2895-4126-b9cd-7d69fb38b96b",
   "metadata": {},
   "outputs": [],
   "source": [
    "import ast\n",
    "dataset['Leading Cast'] = dataset['Leading Cast'].fillna('[]').apply(ast.literal_eval)\n",
    "dataset['Director'] = dataset['Director'].fillna('[]').apply(ast.literal_eval)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "3c83a6fe-0fec-45f9-98d4-5bedb897825c",
   "metadata": {},
   "outputs": [],
   "source": [
    "def castinfo(timewindow_actor, star_actor_num, dataset):\n",
    "    cast_table = pd.DataFrame()\n",
    "\n",
    "    for y in range(1980, 2024, 1):\n",
    "        # get reference set for movie to be released in year y \n",
    "        subset = dataset.loc[dataset['Year'] <= y - 2].copy()\n",
    "        subset = subset.loc[subset['Year'] > y - 2 - timewindow_actor].copy()\n",
    "\n",
    "        # create table of total revenue of all actors in the past timewindow\n",
    "        cast_year = pd.DataFrame(columns = ['name', 'year', 'Total Revenue', 'Star Actor'])\n",
    "        for index, row in subset.iterrows():\n",
    "            lead = row['Leading Cast']\n",
    "            if lead != []:\n",
    "                for name in lead:\n",
    "                    if name in cast_year['name'].values:\n",
    "                        rownum = np.where(cast_year['name'] == name)[0][0]\n",
    "                        cast_year.at[rownum, 'Total Revenue'] = cast_year.at[rownum, 'Total Revenue'] + row['Revenue']\n",
    "                    else:\n",
    "                        cast_year.loc[len(cast_year)] = [name, y, row['Revenue'], False]\n",
    "\n",
    "        # Find stars in this timewindow\n",
    "        topactor = cast_year.sort_values(by = ['Total Revenue'], ascending = False)['name'].head(star_actor_num).reset_index(drop = True)\n",
    "        for name in topactor:\n",
    "            rownum = np.where(cast_year['name'] == name)[0][0]\n",
    "            cast_year.at[rownum, 'Star Actor'] = True\n",
    "\n",
    "        # Append to the main table\n",
    "        cast_table = pd.concat([cast_table, cast_year], ignore_index = True)\n",
    "\n",
    "    # Create two new features\n",
    "    dataset['has_star_actor'] = False\n",
    "    dataset['actor_revenue_before'] = 0\n",
    "    for index, row in dataset.iterrows():\n",
    "        lead = row['Leading Cast']\n",
    "        if lead != []:\n",
    "            temp = cast_table.loc[cast_table['year'] == row['Year']].copy()\n",
    "            star_of_year = temp.loc[temp['Star Actor']]['name'].values\n",
    "            namelist = temp['name'].values\n",
    "            for name in lead:\n",
    "                if name in star_of_year:\n",
    "                    dataset.at[index, 'has_star_actor'] = True\n",
    "                if name in namelist:\n",
    "                    dataset.at[index, 'actor_revenue_before'] = dataset.at[index, 'actor_revenue_before'] + temp.loc[temp['name'] == name]['Total Revenue'].values[0]\n",
    "    return dataset"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "fc6e98ef-0b36-46c7-a913-5fac44759c1e",
   "metadata": {},
   "outputs": [],
   "source": [
    "def directorinfo(timewindow_director, star_director_num, dataset):\n",
    "    director_table = pd.DataFrame()\n",
    "\n",
    "    for y in range(1980, 2024, 1):\n",
    "        # get reference set for movie to be released in year y \n",
    "        subset = dataset.loc[dataset['Year'] <= y - 2].copy()\n",
    "        subset = subset.loc[subset['Year'] > y - 2 - timewindow_director].copy()\n",
    "\n",
    "        # create table of total revenue of all directors in the past timewindow\n",
    "        director_year = pd.DataFrame(columns = ['name', 'year', 'Total Revenue', 'Star Director'])\n",
    "        for index, row in subset.iterrows():\n",
    "            lead = row['Director']\n",
    "            if lead != []:\n",
    "                for name in lead:\n",
    "                    if name in director_year['name'].values:\n",
    "                        rownum = np.where(director_year['name'] == name)[0][0]\n",
    "                        director_year.at[rownum, 'Total Revenue'] = director_year.at[rownum, 'Total Revenue'] + row['Revenue']\n",
    "                    else:\n",
    "                        director_year.loc[len(director_year)] = [name, y, row['Revenue'], False]\n",
    "\n",
    "        # Find stars in this timewindow\n",
    "        topdirector = director_year.sort_values(by = ['Total Revenue'], ascending = False)['name'].head(star_director_num).reset_index(drop = True)\n",
    "        for name in topdirector:\n",
    "            rownum = np.where(director_year['name'] == name)[0][0]\n",
    "            director_year.at[rownum, 'Star Director'] = True\n",
    "\n",
    "        # Append to the main table\n",
    "        director_table = pd.concat([director_table, director_year], ignore_index = True)\n",
    "\n",
    "    # Create two new features\n",
    "    dataset['has_star_director'] = False\n",
    "    dataset['director_revenue_before'] = 0\n",
    "    for index, row in dataset.iterrows():\n",
    "        lead = row['Director']\n",
    "        if lead != []:\n",
    "            temp = director_table.loc[director_table['year'] == row['Year']].copy()\n",
    "            star_of_year = temp.loc[temp['Star Director']]['name'].values\n",
    "            namelist = temp['name'].values\n",
    "            for name in lead:\n",
    "                if name in star_of_year:\n",
    "                    dataset.at[index, 'has_star_director'] = True\n",
    "                if name in namelist:\n",
    "                    dataset.at[index, 'director_revenue_before'] = dataset.at[index, 'director_revenue_before'] + temp.loc[temp['name'] == name]['Total Revenue'].values[0]\n",
    "    return dataset"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "eda9d268-6ad0-4faa-906d-100a7fc43d11",
   "metadata": {},
   "outputs": [],
   "source": [
    "def genreinfo(timewindow_genre, dataset):\n",
    "    genre_table = pd.DataFrame()\n",
    "\n",
    "    for y in range(1980, 2024, 1):\n",
    "        # get reference set for movie to be released in year y \n",
    "        subset = dataset.loc[dataset['Year'] <= y - 2].copy()\n",
    "        subset = subset.loc[subset['Year'] > y - 2 - timewindow_genre].copy()\n",
    "\n",
    "        # create table of total revenue of all actors in the past timewindow\n",
    "        genre_year = pd.DataFrame(columns = ['genre', 'year', 'Total Revenue'])\n",
    "        for index, row in subset.iterrows():\n",
    "            g = row['Genre']\n",
    "            if pd.notna(g):\n",
    "                if g in genre_year['genre'].values:\n",
    "                    rownum = np.where(genre_year['genre'] == g)[0][0]\n",
    "                    genre_year.at[rownum, 'Total Revenue'] = genre_year.at[rownum, 'Total Revenue'] + row['Revenue']\n",
    "                else:\n",
    "                    genre_year.loc[len(genre_year)] = [g, y, row['Revenue']]\n",
    "\n",
    "        genre_table = pd.concat([genre_table, genre_year], ignore_index = True)       \n",
    "\n",
    "    dataset['genre_revenue_before'] = 0\n",
    "    for index, row in dataset.iterrows():\n",
    "        g = row['Genre']\n",
    "        temp = genre_table.loc[genre_table['year'] == row['Year']].copy()\n",
    "        genrelist = temp['genre'].values\n",
    "        if pd.notna(g):\n",
    "            if g in genrelist:\n",
    "                dataset.at[index, 'genre_revenue_before'] = temp.loc[temp['genre'] == g]['Total Revenue'].values[0]\n",
    "            else:\n",
    "                dataset.at[index, 'genre_revenue_before'] = 0\n",
    "                \n",
    "    return dataset"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "c18880e2-5ce3-427b-a9f1-b94081814678",
   "metadata": {},
   "source": [
    "### Finalizing\n",
    "Due to the fact that the time window technique was used, it is mandatory for all data to have \"year\" feature. Thus, observations with NA in year were removed. In fact, due to the low number of observations before 1980, the dataset was further subsetted to only observations after 1980 to ensure better fitting. This left enough data for model and should not hinder validation of the model.\n",
    "Below is a histogram of numbers of movies versus years. It can be seen that few movie were produced before 1980. Refer to the the Descritive.ipynb in the repository for the production of the graph below. (https://github.com/kongwei981126/Movie-Industry/blob/Wei/Descriptive.ipynb)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "570621a3-4d72-4b53-b233-825359736e61",
   "metadata": {
    "tags": []
   },
   "source": [
    "<img src=\"img.png\" width=100% />"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "5417f7c7-c2c9-4ff5-b081-0eb4685c2862",
   "metadata": {},
   "source": [
    "## Modeling approach"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "8d5a78ed-4de1-4e7e-8429-41c31e246b5c",
   "metadata": {},
   "source": [
    "### Deciding on the objective\n",
    "At first, the project aimed to predict ROI of a movie, due to the fact that ROI should counter the problem of inflation and give a better idea of relative return instead of absolute return. For instance, ROI for a big budget movie and a small budget one making the same revenue will be greatly different while the difference cannot be directly estimated by their revenue. (For further details on predicting ROI, please refer to the following report: https://github.com/kongwei981126/Movie-Industry/blob/Wei/passMidReport.ipynb)\n",
    "\n",
    "One major reason is the fact that ROI has low correlation with all the features in the dataset, which can be seen from the correlation plot below. Therefore, instead of ROI, the project switched onto predicting Revenue of a movie yet to be made."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 450,
   "id": "7dbb0923-b18d-48ea-8f9a-b383178284cf",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<AxesSubplot:>"
      ]
     },
     "execution_count": 450,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAA/MAAAOLCAYAAAAcjdMPAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjUuMiwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy8qNh9FAAAACXBIWXMAAA9hAAAPYQGoP6dpAAEAAElEQVR4nOzdd1yV9d/H8RdDVPQAgjgjRI2jJgiOCHeoKZjb3JDirDRHmprlytzmwExz517gNjUrFStXKmYhZuYeOZEcKHD/4e35eQIHclDA9/N+XI+H57q+1+f6XNeh3/34nOs7rJKSkpIQERERERERkUzD+nknICIiIiIiIiKpo2JeREREREREJJNRMS8iIiIiIiKSyaiYFxEREREREclkVMyLiIiIiIiIZDIq5kVEREREREQyGRXzIiIiIiIiIpmMinkRERERERGRTEbFvIiIiIiIiEgmo2JeREREREREJJNRMS8iIiIiIiKSyaiYFxEREREREclkVMyLiIiIiIiIZDK2zzsBkYwgKSmJiIgIIiIiOHLkCHFxcRQoUICqVavSuXNn8ufP/7xTFBERERERMbFKSkpKet5JiDxPCQkJvP/++/z666906dKFKlWqkCtXLo4cOcKUKVM4ffo0K1euJG/evM87VREREREREUDFvAgzZsxg/PjxLF26lFdffdXs2M2bNwkKCiIwMJCPPvroOWUoIiIiIiJiTsW8vNCSkpIICAjg9ddfZ8SIESm2OX36NK6urtjZ2T3j7ERERERERFKmMfPyQjt16hRnzpyhYsWKD21TuHDhZ5iRiIiIiIjI46mYlxfaxYsXAXB2djbb36VLF3bu3Gn6XKhQIdatW/dMcxMREREREXkYFfPyQsuTJw8AV69eNds/ZMgQbt26BcC8efP4/vvvn3VqIiIiIiIiD6V15uWF5ubmhqurK7t27TLbnz9/ftzd3XF3d8fR0fE5ZSciIiIiIpIyFfPyQrOxsSEkJISVK1cSHR2dYpuzZ88+46xEREREREQeTd3s5YXXoUMHfv/9d1q1akWnTp2oXr06uXPnJiYmhvnz57Njxw6aNGnyvNMUEREREREx0dJ0Iv9vw4YNrFixgt9//53Y2Fjy5s1L+fLlad68ORUqVHje6YmIiIiIiJiomBcRERERERHJZDRmXkRERERERCSTUTEvIiIiIiIiksmomBcRERERERHJZFTMi4iIiIiIiGQyKuZFREREREREMhmtMy+SzmKvX7doPAeDwaLxREREREQk89GbeREREREREZFMRsW8iIiIiIiISCajYl5EREREREQkk1ExL89VQEAA1atXJy4uLtmxfv36ERwcbPp89+5d5s6dS+PGjfH19cXPz4927drx888/p+qaq1atIiAgINm1jEZjitvkyZOf7uZERERERETSiYp5ee7Onj3LyJEjH9kmPj6ed955h7lz5xIcHExERARz586lePHihIaGsnLlyie61vr16xkwYECy/QMGDCAyMtJsa968OS4uLrz99ttPc1siIiIiIiLpRrPZy3Pn5ubGsmXLqF27NlWqVEmxzaRJk4iOjmbdunUUKFDAtH/AgAHcuHGD4cOHU6tWLXLlypXi+XFxcQwePJgNGzZQvHhxrv9nhnmDwYDhgVni9+7dy7Jly5g2bRr58+e3wF2KiIiIiIhYjt7My3NXv359/P39+fTTT1Psbn/nzh2WLVtG06ZNzQr5+7p3786MGTPIkSPHQ69x6tQprl27xvLly6lZs+Yj80lMTOTzzz/nzTffpGrVqqm/IRERERERkXSmYl6eOysrKz7//HNiY2MZMWJEsuMnT57k6tWr+Pj4pHh+vnz58Pb2xsbG5qHXKFGiBNOnT6dkyZKPzWfz5s1ER0fTq1evJ74HERERERGRZ0nFvGQIhQsXpk+fPixfvpzt27ebHbt27RoAjo6OzySX2bNnU6dOHdzd3Z/J9URERERERFJLxbxkGC1atEixu72zszMAV69efWyMqVOn4uvra9oGDhyYqhxOnDjBvn37aNasWarOExEREREReZY0AZ5kGPe729erV8+su72bmxt58+Zl3759BAUFJTvv77//ZujQofTt25cWLVoQGBhoOpY7d+5U5fDdd9/h6urKa6+99vQ3IiIiIiIiks70Zl4ylMKFC/PRRx+xfPly9uzZA4C1tTVNmzYlPDyc8+fPJztnxowZ7N+/n8KFC+Pk5IS7u7tpc3FxSdX19+7dy2uvvYa1tf7TEBERERGRjEsVi2Q4LVq0oGLFipw8edK0r0uXLri7u9OiRQtWrlzJiRMnOHjwIAMGDGDFihV89tlnqX4Ln5Lo6Gg8PT3THEdERERERCQ9qZu9ZEjDhg2jXr16ps85c+Zk/vz5zJo1i+nTp3PmzBmyZ8/Oq6++yty5cy3WLf7ixYs4OTlZJJaIiIiIiEh6sUpKSkp63kmIZGWx169bNJ6DwWDReCIiIiIikvmom72IiIiIiIhIJqNiXkRERERERCSTUTEvIiIiIiIiksmomBcRERERERHJZFTMi4iIiIiIiGQyms1eREREREREJJPROvMi6ezcpSsWjVfAJQ9Xrll2uTuAPI5a8k5EREREJLNQN3sRERERERGRTEbFvIiIiIiIiEgmo2JeHiogIIDq1asTFxeX7Fi/fv0IDg5+4lhJSUlERERw6dKlR17PaDSabd7e3tSuXZvJkyeTmJho0es9iYCAAMLCwtIUQ0RERERExNJUzMsjnT17lpEjR6Y5zu7du+nXrx83b958ZLvQ0FAiIyNNW0REBPXq1SMsLIzZs2db/HoiIiIiIiKZkYp5eSQ3NzeWLVvG9u3b0xTnSRdNsLe3x9XV1bQVK1aMrl274ufnx7p16yx+PRERERERkcxIxbw8Uv369fH39+fTTz9Nsbv9fVevXmXIkCFUq1YNb29vWrZsyZ49ewDYuXMnISEhANSoUYPw8PBU55E9e3asrf/353rkyBHee+89/Pz8KF26NLVq1WLu3LmPvN6vv/5K69at8fb2pnr16gwZMsTsnq5fv07fvn0pX748/v7+zJkzJ9V5ioiIiIiIPAsq5uWRrKys+Pzzz4mNjWXEiBEptklISCA0NJQ9e/YwatQoIiIiKFGiBG3btuXgwYP4+vqaxp0vW7aMoKCgJ75+fHw8K1euZMeOHTRo0ACAmzdv0q5dO+zt7Vm4cCHr1q0jMDCQ4cOH88cff6R4vejoaNq2bUulSpVYvXo1Y8eO5dChQ4SGhpre4vfo0YOoqCimTp3KrFmz+OGHHzh9+nRaHp+IiIiIiEi60Drz8liFCxemT58+DB48mDp16lClShWz45GRkRw6dIg1a9bg6ekJwMCBAzlw4AAzZ85kwoQJODo6AuDs7EyOHDkeeq1p06Yxa9Ys0+ebN2/i4eHBgAEDaNWqlWlfSEgIrVq1Infu3AB07dqVadOmcfjwYUqWLJnsejNnzsTf35/33nsPgCJFijBu3Dhq1qzJrl27cHV1JTIykjlz5lC+fHkAxo0bxxtvvGGJRygiIiIiImJRKublibRo0YKNGzfy6aefsnbtWrNjMTExGAwGUyEP997oly9fPtVj7Vu0aEFwcDB3797lp59+Yvz48dSpU4fWrVub2jg7O9OqVSvWr19PdHQ0x48f548//gB46Iz3v//+O8ePH8fX1zfZsaNHj3LlyhUAvLy8TPvz5s2Lm5tbqvIXERERERF5FlTMyxO5392+Xr16ybrbJyUlYWVlleycxMREbG1T9yfm6OiIu7s7AMWKFcNgMNC3b1/s7e3p2LEjABcvXqRZs2bkyZOHGjVq4O/vj5eXF9WqVXto3MTEROrVq0eXLl2SHXN2dmbHjh2mdg9Kbf4iIiIiIiLPgsbMyxMrXLgwH330EcuXLzdNbgdgNBqJjY0lJibGrP3evXspXrw4QIrF/pNo2LAhderUYeLEiRw+fBiANWvWcPXqVRYvXsx7771HrVq1uHbtGvC/Wez/e71XXnmFI0eO4O7ubtoSEhIYMWIEZ8+epVSpUsC9SfLui42N5cSJE0+Vt4iIiIiISHpSMS+p0qJFCypWrMjJkydN+ypVqoTRaOTDDz9k586dHD16lCFDhhATE8M777wD3FtyDiA6Opp///03VdccOHAguXLlYsCAASQmJlKgQAFu3rzJhg0bOHPmDJGRkfTq1Qu4N2FeStcLDQ3ljz/+YODAgfz5558cOHCA3r17c+zYMYoUKcLLL79MnTp1GDp0KD/99BMxMTF89NFHpngiIiIiIiIZiYp5SbVhw4aRK1cu02dbW1tmz55NyZIl6datG02aNCEmJoY5c+bg4+MDgKenJ9WqVaNHjx4sWbIkVddzcXGhf//+HDx4kG+++YY6derQvn17Ro0aZZrFvmnTplSoUIGoqKgUr+fj48OMGTOIiYmhcePGdOrUCTc3N2bPno2dnR0Ao0aNonr16vTs2ZPWrVtTvHhxSpcubZmHJiIiIiIiYkFWSff7JYtIujh36YpF4xVwycOVa9ctGhMgj6PB4jFFRERERCR96M28iIiIiIiISCajYl5EREREREQkk1ExLyIiIiIiIpLJqJgXERERERERyWRUzIuIiIiIiIhkMprNXkRERERERCSTsX3eCYhkdbHXLbuMnIPBYPGY6RXXwaDl7kRERERE0oO62YuIiIiIiIhkMirmRURERERERDIZFfMiIiIiIiIimYzGzEuGlpCQwJIlSwgPD+fo0aPY2NhQvHhxmjdvTsOGDbGysnreKYqIiIiIiDxzKuYlw7p79y7vvfceBw8epGvXrlSqVImEhAR27NjB8OHD2bJlCxMnTsTGxuZ5pyoiIiIiIvJMqZiXDGvq1Kns3buX8PBw3N3dTfuLFSvGa6+9RtOmTZk5cyadOnV6jlmKiIiIiIg8exozLxlSUlIS8+fPp1GjRmaF/H0lSpSgQYMGzJs3j5MnT2I0GlmzZg3169fH29ubZs2a8euvv5qds2LFCgIDA/H29iYwMJC5c+eSmJgIwKlTpzAajWzYsIG3334bLy8vatSowfLly5/J/YqIiIiIiKSGinnJkI4dO8aVK1coW7bsQ9v4+/tz4cIFkpKSAPj888/p1KkTERERFC1alHbt2nHy5EkAlixZwqhRo3j//fdZt24dPXr0YPr06YwdO9Ys5siRI+nSpQsrV67E39+fTz/91BRDREREREQko1AxLxnS1atXAciTJ89D29w/dvnyZQA6d+7MW2+9RbFixfjss8/IkycPS5cuBWDKlCmm425ubtSuXZuePXsyf/58bt++bYrZrl07atSoQbFixejbty+JiYkcOHAgne5SRERERETk6WjMvGRITk5OAFy/fv2hba5duwb8r6h/7bXXTMeyZcvGq6++SkxMDJcvX+bcuXNMnDiRyZMnm9okJiZy+/ZtTp06Rfbs2YF74/HvMxgMANy5c8cyNyUiIiIiImIhKuYlQ3J3d8fV1ZVdu3bx5ptvpthm586duLq6mpans7U1/3NOTEzE2traNC6+f//+VKxYMVmcggULcuHCBQDs7OySHb/fjV9ERERERCSjUDd7yZBsbGwICQlh+fLlHDlyJNnx6OhoVq5cSatWrbC2vvdnfPDgQdPx+Ph4Dh06xKuvvoqLiwsuLi6cOHECd3d303bo0CEmTJjwrG5JRERERETEYvRmXjKs9u3bc/DgQdq0acMHH3xA5cqVAYiMjGTSpEn4+fnRqVMnzp07B8CECRPImzcvbm5ufPXVV9y8eZNmzZphZWVFhw4d+OKLLyhUqBDVqlUjJiaGIUOGUL169RTfxouIiIiIiGRkKuYlw7KxsWHSpEmEh4ezbNkyxo8fT1JSEq+88gq9e/emadOmpi72AC1atGDEiBGcO3eOMmXKMG/ePPLlywdAaGgo2bNnZ968eYwaNQoXFxcaN25Mz549n9ftiYiIiIiIPDWrJA0Ilkzu1KlT1KhRg2+++QY/P7/nnU4ysY+YxO9pOBgMFo+ZXnEd/n8SQRERERERsSyNmRcRERERERHJZFTMi4iIiIiIiGQyGjMvmd5LL73E4cOHn3caIiIiIiIiz4zezIuIiIiIiIhkMpoAT0RERERERCSTUTd7kXT299l/LBqvSEFXLl2NtWhMABcnB6JPnLVozBIvF+TKNcvOkJ/HUTPki4iIiIiom72IiIiIiIhIJqNiXkRERERERCSTUTEvIiIiIiIikslozLxkSQEBAZw+fdr02dramly5clGyZEm6d+9O+fLlTce+++47lixZwu+//861a9dwcXHB39+fTp06UbRo0eeRvoiIiIiIyCPpzbxkWaGhoURGRhIZGcnWrVtZuHAhuXLlokOHDpw7dw6AoUOH0qtXL7y8vJg+fTqbNm1i1KhRXLp0iaZNm/Lnn38+57sQERERERFJTm/mJcuyt7fH1dXV9DlfvnwMGTKEqlWrsmnTJlxdXVmwYAFTpkyhRo0apnaFChXitddeo3nz5oSFhTFx4sTnkb6IiIiIiMhDqZiXF4qt7b0/eTs7O7755hv8/PzMCvn7rK2tmTx5MgaDlkETEREREZGMR93s5YVx/vx5hg4dir29PVWrVmX//v1UrFjxoe3z58+Pvb39M8xQRERERETkyejNvGRZ06ZNY9asWQDcvXuX+Ph4ihUrxoQJE7C1tSUxMRFnZ2ezc4YOHUpERITZvn379j2znEVERERERJ6EinnJslq0aEFwcDBwr9u8k5OTqdt8fHw8VlZWXL161eycrl278s477wCwadMmxo4d+0xzFhEREREReRIq5iXLcnR0xN3dPcVjdnZ2eHl5sWvXLjp16mTa7+zsbHpb7+Li8kzyFBERERERSS2NmZcXVtu2bYmMjGT79u0pHj979uwzzkhEREREROTJ6M28vLDq1q3Lb7/9xrvvvss777xD7dq1cXFx4fjx4yxdupQNGzbw+uuvP+80RUREREREklExLy+0vn37UrlyZRYvXsz777/PlStXcHJywsfHh6+++oqAgIDnnaKIiIiIiEgyVklJSUnPOwmRrOzvs/9YNF6Rgq5cuhpr0ZgALk4ORJ+w7NCCEi8X5Mq16xaNmcfRYNF4IiIiIiKZkcbMi4iIiIiIiGQyKuZFREREREREMhkV8yIiIiIiIiKZjIp5ERERERERkUxGxbyIiIiIiIhIJqPZ7EVEREREREQyGa0zL5LODv19xqLxXi1SiHOXrlg0JkABlzz8/Mcxi8b0L+nB5WuWXUbP2dGB2Vt2WzQmQLsaFSweU0REREQkvaibvYiIiIiIiEgmo2JeREREREREJJNRN3t5rgICAjh9+rTps7W1Nbly5aJkyZJ0796d8uXLm4599913LFmyhN9//51r167h4uKCv78/nTp1omjRok98zVWrVjFx4kS+//57s/0//fQT48ePJyYmBkdHR+rVq0f37t2xs7NL+42KiIiIiIhYkN7My3MXGhpKZGQkkZGRbN26lYULF5IrVy46dOjAuXPnABg6dCi9evXCy8uL6dOns2nTJkaNGsWlS5do2rQpf/755xNda/369QwYMCDZ/ujoaDp16kSVKlVYu3YtI0aMYPXq1YwePdqi9yoiIiIiImIJejMvz529vT2urq6mz/ny5WPIkCFUrVqVTZs24erqyoIFC5gyZQo1atQwtStUqBCvvfYazZs3JywsjIkTJz70GnFxcQwePJgNGzZQvHhxrl+/bnb8zJkzNG7cmA8++AAANzc3AgMD2bFjh4XvVkREREREJO1UzEuGZGt770/Tzs6Ob775Bj8/P7NC/j5ra2smT56MwWB4ZLxTp05x7do1li9fznfffUdERITZ8YCAAAICAgBISkoiKiqKzZs3U6tWLQvdkYiIiIiIiOWom71kOOfPn2fo0KHY29tTtWpV9u/fT8WKFR/aPn/+/Njb2z8yZokSJZg+fTolS5Z8ZLuEhAR8fHxo1qwZTk5OdO3a9anuQUREREREJD3pzbw8d9OmTWPWrFkA3L17l/j4eIoVK8aECROwtbUlMTERZ2dns3OGDh2a7O36vn370pxLYmIi8+fP58qVKwwfPpyOHTuyaNEirK31u5eIiIiIiGQcKubluWvRogXBwcHAvW7zTk5Opm7z8fHxWFlZcfXqVbNzunbtyjvvvAPApk2bGDt2LABTp05l2rRppnb16tVj6NChT5xLtmzZ8PLyAsDR0ZFmzZqxd+9eKlSo8NT3JyIiIiIiYmkq5uW5c3R0xN3dPcVjdnZ2eHl5sWvXLjp16mTa7+zsbHpb7+LiYtrfokULAgMDTZ9z5879RDn89ttv3Lhxg9dee82075VXXgHudfsXERERERHJSNR3WDK8tm3bEhkZyfbt21M8fvbsWdO/nZyccHd3N20PFvqPsmrVKvr160dCQoJp34EDBwAoXrx4GrIXERERERGxPL2Zlwyvbt26/Pbbb7z77ru888471K5dGxcXF44fP87SpUvZsGEDr7/+epqu0bp1a5YvX87gwYMJDQ3l5MmTDB48mNq1a1OiRAkL3YmIiIiIiIhlqJiXTKFv375UrlyZxYsX8/7773PlyhWcnJzw8fHhq6++Mi0r97SKFCnC3LlzGTNmDI0bNyZXrlzUq1ePnj17WugORERERERELMcqKSkp6XknIZKVHfr7jEXjvVqkEOcuXbFoTIACLnn4+Y9jFo3pX9KDy9diLRrT2dGB2Vt2WzQmQLsamuRQRERERDIPjZkXERERERERyWRUzIuIiIiIiIhkMirmRURERERERDIZFfMiIiIiIiIimYyKeREREREREZFMRrPZi4iIiIiIiGQyWmdeJJ3tP3rSovF8irlx5p/LFo0JUMjVme/2x1g0Zk0fTy5dtezSdC5ODoxZudWiMQH6NKxGwOCZFo35/eD2Fo0nIiIiInKfutmLiIiIiIiIZDIq5kVEREREREQyGRXzIiIiIiIiIpmMxszLCy84OJhdu3aZ7cuWLRv58uWjRo0afPjhh+TIkeM5ZSciIiIiIpKcinkRIDAwkAEDBpg+37hxg8jISEaMGEFCQgIDBw58jtmJiIiIiIiYUzEvAuTIkQNXV1ezfe7u7vz222+sW7dOxbyIiIiIiGQoGjMv8gjZs2fH2lr/mYiIiIiISMaiKkUkBXfv3uXHH39k1apVNGjQ4HmnIyIiIiIiYkbd7EWANWvWsHHjRtPnW7duUahQIdq3b0+XLl2eY2YiIiIiIiLJqZgXAQICAujduzeJiYkcOHCAESNGULFiRbp06YKtrf4zERERERGRjEVVigiQK1cu3N3dAfDw8KBAgQK0a9cOGxsbBg8e/HyTExERERER+Q+NmRdJweuvv067du1YtGgR27Zte97piIiIiIiImFExL/IQ3bt3p0iRIgwaNIh///33eacjIiIiIiJiomJe5CGyZ8/OZ599xtmzZxk/fvzzTkdERERERMREY+blhTdv3ryHHnvttdeIjo5+htmIiIiIiIg8nt7Mi4iIiIiIiGQyKuZFREREREREMhkV8yIiIiIiIiKZjIp5ERERERERkUzGKikpKel5JyEiIiIiIiIiT06z2Yuks0t/HbFoPJeir3D92lWLxgQwODrx5+nzFo1ZvHB+Ll+LtWhMZ0cH/vkjyqIxAVxLenP13BmLxnQqUIg/jp+1aMyS7gUtGk9EREREMid1sxcRERERERHJZFTMi4iIiIiIiGQyKuZFREREREREMhkV81mQ0WgkPDzcIrFOnTqF0Wh86PbHH39Y5DpPIzw8HKPR+Mg2AQEBhIWFPaOMREREREREng1NgJcFRUZGYjAYLBozLCwMX1/fZPvz5Mlj0etY2vLly8mePfvzTkNERERERMSiVMxnQa6urhaP6ejomC5x05uzs/PzTkFERERERMTi1M0+C3qwm/2lS5f44IMP8PPzw9vbmxYtWrBr1y6LXu9+V/wpU6ZQqVIlAgICiI2N5ciRI7z33nv4+flRunRpatWqxdy5c03nhYWFERwczPTp06latSpeXl6EhITw119/mdrcuHGDYcOGUblyZXx9fWndujVRUebLkoWHh1OrVi28vLxo3LgxBw4cMB17sJv9zZs3GTBgAJUqVcLLy4uGDRuyadMmU9ukpCSmT59OjRo1KFOmDA0aNGD16tUWfVYiIiIiIiKWoGI+ixs8eDC3bt1i/vz5rFmzBg8PD9577z1u3Lhh8WutXr2auXPnMnHiRLJly0a7du2wt7dn4cKFrFu3jsDAQIYPH242zn7fvn3s3r2br7/+mjlz5nDmzBmGDBliOt6zZ09++OEHhg8fzsqVK/Hw8KB9+/ZcvnzZ1Gbx4sWMGzeOFStWYGdnR48ePVLMb+LEiRw+fJivv/6a9evXU7VqVXr27MmpU6cAGD9+PAsXLuSTTz5hzZo1hISEMHjwYBYsWGDxZyUiIiIiIpIW6mafxZ04cQJPT09efvllsmfPzoABA6hXrx42NjapitOxY8dk5wwcOJBGjRqZPrdq1YrixYsDcPnyZUJCQmjVqhW5c+cGoGvXrkybNo3Dhw9TsmRJAO7evcvo0aNxcnICIDg4mDFjxgBw7NgxfvzxR2bMmEGVKlVM18yVKxdXr141XXf48OGm67Zv356uXbty6dIlXFxckj2L3Llz8/LLL2MwGOjevTvly5fH0dGRGzduMGfOHEaPHs0bb7wBwMsvv8zp06eZOXMmrVu3TtXzEhERERERSU8q5rO4rl270qdPHzZv3kz58uWpXLkyQUFBqZ4UbtiwYZQpU8Zs33+LZXd3d9O/nZ2dadWqFevXryc6Oprjx4+b3sgnJiaa2uXNm9dUyAMYDAbu3LkDwOHDhwHw8fExHbezs6N///4A7N+/HwAPDw/TcQcHBwBu3bqV7B46duxIly5d8Pf3x9fXl0qVKlG3bl0MBgNRUVHcvn2bvn37muLDvR8b4uPjuXXrFjly5Hj0QxIREREREXlGVMxncbVq1WL79u1s376dn376iRkzZjBx4kSWLl3KK6+88sRx8ufPb1asp+TBYvfixYs0a9aMPHnyUKNGDfz9/fHy8qJatWpm59jZ2T00nq3tvT9PKyurR143pV4GSUlJyfb5+vqydetWduzYwc8//8zy5csJCwtjxowZ2NvbAzBhwgSKFi2a7NxH5SkiIiIiIvKsacx8FhYfH8+IESM4efIkQUFBDBs2jM2bN2Ntbc2PP/6Yrtdes2YNV69eZfHixbz33nvUqlWLa9euASkX2ikpVqwYAAcPHjTtu3v3LtWrV2fdunWpzmnSpEns3buXGjVq8Mknn7Bx40bc3NzYuHEjRYsWxdbWljNnzuDu7m7atm7dysyZM7G21n8qIiIiIiKScahCycLs7Ow4cOAAn376Kfv37+fUqVOEh4fz77//prhmvCUVKFCAmzdvsmHDBs6cOUNkZCS9evUC7v3I8CQ8PDx48803GTJkCD///DPHjh1j4MCBxMfH4+/vn+qcjh8/zqBBg/j55585ffo03377LWfOnMHX1xeDwUCLFi2YMGECK1eu5OTJk0RERDBmzBjy5s2b6muJiIiIiIikJ3Wzz+ImTpzIiBEjePfdd7l+/TpFixZl3LhxlC9fPl2vW6dOHQ4dOsSoUaOIi4ujcOHCvP3222zZsoWoqChatmz5RHFGjBjB6NGj6dmzJ7dv36ZMmTLMmjXrqdaPHzJkCKNGjaJPnz5cvXqVwoUL07t3bxo0aABA//79cXZ2ZtKkSVy4cIECBQrQtWtXOnXqlOpriYiIiIiIpCerpCft8ywiT+XSX0csGs+l6Ctcv3bVojEBDI5O/Hn6vEVjFi+cn8vXYi0a09nRgX/+iLJoTADXkt5cPXfGojGdChTij+NnLRqzpHtBi8YTERERkcxJ3exFREREREREMhl1s39BnT9/njp16jyyTalSpViwYMEzykhERERERESelIr5F1TevHlZuXLlI9ukdi16ERERERERS/tz4bQ0xyjeqrMFMslYVMy/oGxsbB67bryIiIiIiMjzZmVl9bxTyJA0Zl5EREREREQkk9Fs9iIiIiIiIpJhHV30dZpjFGuZ9ZabVjd7kXQWe/26ReM5GAxcuWbZmAB5HA2cu3TFojELuORJl/s/t/dni8YEKFDOn6tnTlo0plMhN/YeOWHRmOVeeZkzO76zaMxClWpaNJ6IiIiIRVmrm31KVMyLiIiIiIhIxmWl0eEp0VMRERERERERyWReyGI+Li6OMmXKULFiReLj41N17s6dOzEajZw6dSqdsoM7d+4wZ84c0+ewsDACAgIsfp0LFy7g6+tLfHw8wcHB9OvX76Ft0yuH//rwww8xGo18913quxHfuHGDBQsWmD7369eP4OBgS6YnIiIiIiLPmJWVVZq3rOiFLObXrVuHi4sLcXFxbN68OVXn+vr6EhkZScGCBdMpO1i7di0jRowwfQ4NDWX58uUWv87WrVt5/fXXsbOze2zb9MrhQdevX+e7777Dw8ODRYsWpfr8WbNmMXPmTNPnAQMGEBYWZskURUREREREMoQXsphfsWIFlStXxt/fn8WLF6fqXDs7O1xdXbGxsUmn7OC/CwzkypULZ2dni19n27ZtVKtW7YnaplcOD1q7di3W1ta8//777Nixg5MnUzcZ2X+fm8FgwMnJyYIZioiIiIjIM2dlnfYtC8qad/UIR48e5cCBA1SqVIk6deqwa9cujh49ajoeHBzMqFGj+Pjjjylfvjxly5alb9++/Pvvv0DybvYBAQHMmzePbt26UaZMGapWrcqyZcvYt28fDRs2pEyZMrRo0YITJ/43o/XevXtp164d5cqVo3Tp0rz11lusXbsWgPDwcPr37w+A0Whk586dybq4nz17lt69e1OpUiV8fHxo3749hw8fNh3v168fffr0YdSoUfj7+1OmTBnee+89/vnnH1ObO3fu8NNPPz1xMf9gDsHBwfTo0cPs+N69ezEajRw/fhyAH374gcaNG+Pt7U2tWrWYMGHCY4c0hIeH4+fnR82aNcmZM2eKP7T89ttvtGvXDl9fXypWrMjAgQO5ceMGYWFhTJ48mdOnT5u+n/92sz969ChdunTBz8+PcuXK8cEHH3DmzBnT8cd99yIiIiIiIhnFC1fML1++HHt7e6pWrUrNmjWxs7NL1qV73rx55M2bl2XLljFs2DDWr19vNob9v8aNG0eVKlVYu3Yt1atXZ/DgwQwaNIh+/foxf/58/vnnH8aOHQvA+fPnCQ0NpUSJEoSHh7Nq1Sq8vLzo378/Fy9eJCgoiI8//hiAyMhIfH19za4VFxdHy5YtOX/+PF999RWLFy/G3t6eNm3amBWmGzZs4OrVq8yfP5/Jkyezd+9exo8fbzq+d+9eChUq9FTDBRo1asQPP/xAXFycad/q1aspW7Ys7u7ubNu2je7du/P222+zdu1aBg0axIYNG+jTp89DY/75559ERUVRu3ZtcubMyRtvvEF4eLjZDwCnTp0iODgYZ2dnlixZwuTJk9m5cycDBw4kNDSU0NBQChQokOIwiNOnT9O8eXPs7OyYO3cus2fP5tKlS7Rp08bsPlL73YuIiIiISDqztkr7lgW9UMX83bt3WbNmDW+88QY5c+bEYDBQrVo1Vq1axc2bN03tihUrRq9evfDw8CAoKIiqVavy66+/PjRu1apVadasGW5uboSEhHD37l2Cg4N5/fXX8fLyIjAwkJiYGADi4+Pp2rUrffr0wd3dnWLFitG5c2fi4+P5+++/yZEjBwaDAQBXV9dk49lXr17NlStXmDhxIt7e3pQoUYKxY8eSI0cOs8nfcufOzdChQylWrBhVqlShQYMG7N2713R827ZtVK1a9ameY506dbC2tjbNNxAfH8+3335L48aNAZg6dSpNmzalZcuWvPzyy1SuXJkhQ4bw7bffPnTiwBUrVmBnZ0fNmvfWu65bty6XL19m06ZNpjZLly7F0dGRkSNH4unpSdmyZRk2bBgeHh7kypULe3t7bGxsUhwGsXDhQuzt7Rk7diwlSpTA29ubSZMmcenSJVavXm1ql9rvXkRERERE0peVlXWat6zohVpnfuvWrfzzzz8EBQWZ9gUFBbF582bWrVtH06ZNgXsF3YMMBgOxsbEPjevh4WH6d44cOQB46aWXTPuyZ89uesPs5uZGkyZNmD9/Pn/++Sd///03f/zxBwAJCQmPvYeYmBiKFCliNn49e/bseHt7m3W1d3d3J1u2bGb3cOfOHdPnbdu28emnnz72eimxt7enTp06rFmzhkaNGrFt2zZu3bpFYGAgAL///jtRUVFERESYzrk/nv3o0aNmzwbu/ciyevVqqlSpYvoho0qVKjg4OLBo0SLeeustAA4fPsyrr75qdl8VKlSgQoUKj805JiaG0qVLm/044uLigoeHh9lzS+13LyIiIiIi8jy8UMV8eHg4AB988EGyY4sXLzYV808yu/uDbG2TP0Zr65R//Tl69CgtW7akVKlSVKpUiRo1apAnTx7efvvtJ7pWUlJSiksrJCQkmOXxqHs4c+YM586do2zZsk90zZQ0btyYd955h3/++YfVq1dTs2ZNcufODUBiYiIdOnSgUaNGyc5zdXVNtu/HH3/k4sWLfP/995QqVcrsnvbs2cOff/5J8eLFsbW1feplJR713B78cSC1372IiIiIiMjzkDX7G6Tg8uXLbN26lcaNG7Ny5UqzrWnTphw8eJBDhw6lex6LFi3CxcWFOXPm0LFjR6pVq8bFixeB/729flTB6unpybFjx7h06ZJp3+3bt/ntt98oXrz4E+WwdetWKlasaFbEplaFChUoXLgwK1eu5McffzQr3F955RX++usv3N3dTdv58+cZPXp0ipPJrVixgjx58iT7Xr766isA00R4xYsX5/fffzfrwbB582aqVq3KzZs3H/vcoqKizMbgX7x4kePHjyd7Gy8iIiIiIhmIlVXatyzohSnmV61axd27d+nQoQOenp5mW5cuXbCxsXmqtc1Tq0CBApw7d46tW7dy+vRpNm3axODBgwFMhaa9vT1wb+b2W7dumZ1fr149HBwc6NGjB1FRUURHR9OnTx9u3LhB8+bNnyiHrVu3pjhe/vz582zbti3Z9t8l3+5r2LAhX331FU5OTlSsWNG0v2PHjmzatImwsDCOHTvGzz//TP/+/YmNjU32Zv7SpUts27aNZs2aUaJECbPvJSAgAD8/P1auXMnNmzdp1aoVV65cYdCgQRw9epQ9e/YwduxYKlWqRM6cObG3t+fatWscO3bMbEgBQMuWLYmLi6N3795ER0cTFRVF9+7dyZMnD3Xr1n2i5yYiIiIiIs+exsynLGveVQrCw8OpWLFiim9h3dzcqFWrFuvWrTOb2Tw9hISEEBgYyEcffcRbb73FV199Ra9evShcuDBRUVEAvP7666Yl7X744Qez8x0cHJg/fz4Gg4G2bdvSqlUrbt68yaJFi3Bzc3vs9ePj49m5c2eKxfxPP/1Ex44dk20PG8vfqFEjbt68SYMGDcyGFdSpU4fx48ezZcsW6tWrR+/evfH392fy5MnJYqxatYqkpCRatmyZ4jXat2/P9evXWbt2Lfnz52fWrFkcO3aMRo0a0aNHD6pVq8agQYMAePPNN3F1daV+/fr8/vvvZnHc3NyYN28esbGxNG/enPbt2+Pq6sqiRYtwcHB47HMTEREREZHnRG/mU2SV9LDXriJiEbHXr1s0noPBwJVrlo0JkMfRwLlLVywas4BLnnS5/3N7f7ZoTIAC5fy5euakRWM6FXJj75ETFo1Z7pWXObPjO4vGLFSppkXjiYiIiFjS36vT3oO6SP2UXx5mZi/Mm3kRERERERGRrOKFms1eREREREREMhnrrNlNPq1UzIuIiIiIiEiGlVUnsEsrPRURERERERGRTEZv5kVERERERCTjyqKz0aeVZrMXERERERGRDOv4+mVpjuEe9LYFMslY9GZeJJ2d3r7ZovEKV6mVbkvTXboaa9GYLk4O6bI03Z8Lp1k0JkDxVp3ZE3PcojHLe7qz7bc/LRqzauniXIjaY9GY+bzLc/aiZZclBCiYN4/FY4qIiIjIPSrmRUREREREJOPSBHgpUjEvIiIiIiIiGZaVxsynSD9xPCMBAQEYjUbTVrp0aapXr87QoUO5csW8e6vRaCQ8PPyJ4p46dQqj0cjOnTsBuHHjBgsWLHjivP57/sPs2LEDo9HI+++//8SxHxQWFkZAQMBTnfukrly5wrJl/xtPExwcTL9+/dL1miIiIiIiks6srdO+ZUF6M/8MhYaGEhoaCsCtW7eIiYlhzJgx7N69m0WLFpE7d24AIiMjMRgMTxSzYMGCREZG4ujoCMCsWbMIDw+ndevWFs09PDwcDw8PfvjhB86fP0/+/PlTdX5oaKjFc/qv0aNHc+rUKd5++97kFmFhYdjY2KTrNUVERERERJ6HrPkTRQZlb2+Pq6srrq6uuLm5UaNGDWbNmsWpU6eYOXOmqZ2rqys5cuR4opg2Nja4urpiZ2cHQHosThAbG8vmzZvp0qULuXLlYunSpamOkStXLpydnS2e24P+e+9OTk5P/KOIiIiIiIhIZqJi/jkrVKgQtWrVYu3ataZ9/+1mP3fuXAICAvD29qZt27ZMnjzZ1GX9wW7yYWFhTJ48mdOnT2M0Gjl16hTx8fGMGzeOmjVrUrp0afz8/OjVq1eyrv2PsnbtWu7cuUOVKlWoWbMmS5cu5e7du2ZtjEYjS5cupV27dnh7e1OlShWmTfvfjOMPdrO/n/PWrVtp3LgxXl5e1KtXj/3797Ns2TLeeOMNypYty4cffsjt27dNMVasWEHDhg3x9vbGx8eH4OBgDh06BEC/fv2IiIhg165dGI1GIHk3+3379hESEkK5cuXw8/Pj448/5tq1a6bjAQEBfP3113Tr1g1fX1/8/PwYPnx4snsVEREREZFnx8rKKs1bVqRiPgPw9PTkxIkT/Pvvv8mOLViwgC+++IL33nuPVatW4efnx5dffplinPvd+AsUKEBkZCQFCxZk9OjRrF27ls8//5yNGzcyatQoduzYwVdfffXE+a1YsYLy5cvj4uJCUFAQFy5c4IcffkjWbvTo0TRs2JBVq1bRpEkTvvjiC/bsefgSWkOHDqV3796sXLmSHDly0KlTJzZs2MDUqVMZOXIkGzduNI2B37x5M4MGDaJt27Zs2LCBuXPncuvWLQYMGADAgAEDCAwMxNfXl8jIyGTXioqKIjg4mOLFi7NkyRImTZpEVFQUoaGhJCYmmtqFhYVRoUIFIiIi6NatG998843ZDy0iIiIiIvKMWVmlfcuCVMxnAA4ODgDExcUlOzZz5kxCQkJo2rQpHh4evPvuu9SsWTPFOLly5cLe3t7U9d7GxgYvLy9GjRqFn58fhQsXpnr16lSuXJnDhw8/UW4xMTH89ttv1K1bFwB/f3+cnZ1ZvHhxsraNGjWiQYMGeHh40KNHDxwdHdm7d+9DY7dr146KFStSrFgxGjZsyLVr1xg0aBBGo5E333yTUqVKERMTA9zrMj9s2DAaNmxI4cKFKVOmDG+//bbpPgwGAzly5CBbtmy4uromu9asWbMwGo0MHDiQ4sWL4+fnx7hx4/jtt9/Yvn27qV2VKlUICQmhSJEitGnThhIlSvDrr78+0bMSEREREZGsJzExkUmTJlGlShXKlClDaGgox48ff2j7I0eO0KlTJ/z8/PD39+eDDz7gzJkzFs9LxXwGcP36dQDTBHj3XblyhdOnT+Pj42O2v1y5ck8cu0GDBiQkJPDFF1/wwQcfULduXdavX2/2NvpRVqxYga2tLW+++SYAtra21K5dmx07dnDixAmztsWKFTP7nDt3bu7cufPQ2B4eHqZ/58yZEwA3NzfTvuzZsxMfHw9AhQoV8PLyYsqUKfTt25dmzZrx2WefPfF9xMTEULZsWbN9RqMRBwcHsx82/nsPBoPhkfcgIiIiIiLpzMo67VsaTJkyhcWLFzNs2DCWLFmClZUVHTt2NNUqD7py5Qrt2rUjV65czJ8/n+nTp3PlyhU6dOhgNoTYElTMZwCHDh2iSJEi5MqVy2y/re29xQbSMqnd4MGD6d69O7du3aJ69eqMHTvW9Jb9ce7cucPq1au5e/culStXplSpUpQqVYolS5aQlJSU7O38/Un4HvSo3O/f34OsH7JsxLp166hfvz7Hjx/H29ubPn36pGrZuaSkpBTHyiQmJpItWzbT59Teg4iIiIiIZF3x8fHMmjWLbt26Ua1aNUqUKMH48eM5f/48mzdvTtb+u+++4+bNm4wcOZJXXnmF0qVLM2bMGI4ePWrxHr9amu45O3fuHFu2bKFjx47JjhkMBgoXLsz+/fvNutZHRUU9NN6DBeuVK1dYtGgR48ePJygoyLT/r7/+wt7e/rG5/fjjj1y+fJlBgwZRvnx5s2O9e/cmPDycHj16pFgAW9rUqVNp2rQpQ4YMMe3bsmUL8L9C/VETW3h6eiYbvx8dHU1cXFyyt/EiIiIiIpJxWFk/vzHv0dHR/Pvvv7z++uumfQ4ODpQqVYrdu3cne1Hq7+/Pl19+Sfbs2ZPFenDybUtQMf8M3bhxg3/++Qe4t8784cOHmTBhAi+99BLt2rVL8ZyOHTsyatQoihUrRtmyZfnhhx/YsGEDBQsWTLG9vb09165d49ixY7z00ksYDAa2bNnCq6++yq1bt5g/fz6HDh2iTJkyj813xYoVFChQgGbNmiV7ix4aGkrfvn359ttvqV+/fiqfROoVLFiQX3/9lUOHDmEwGPj++++ZP38+cO/XsuzZs2Nvb8+FCxc4efKkWXd9gLZt29K6dWuGDh1K69atuXTpEkOHDqVUqVL4+/une/4iIiIiIvL81KhR45HH778o/K9z584BJKu/8uXLx9mzZ5O1f+mll3jppZfM9k2bNo3s2bNToUKF1KT8WOpm/wzNmjWLypUrU7lyZerXr8+4ceOoUaMGCxcuTNbF/r6WLVvSpUsXxo8fz1tvvcX27dtp1KiRWdfwB7355pu4urpSv359fv/9dyZOnEhMTAz16tWjQ4cO3Lx5k169enHkyBFu3Ljx0FwvXrzI9u3badmyZYrd4evWrUv+/PlTnAgvPXz66afkzZuXNm3a8Pbbb/PDDz8wevRoAA4cOABAw4YNuXnzJm+99RYXLlwwO9/X15fp06fz22+/0bBhQ7p3746vry+zZ89+6LMUEREREZEM4DmOmb958yaQfDhu9uzZn2gM/DfffMPChQvp1asXLi4uT51HSqySNCA4Q9u2bRuvvPKK2S9Bn376KSdOnGDu3LnPMTN5Uqe3Jx9LkxaFq9TiyrXrFo0JkMfRwKWrsRaN6eLkQOx1y+bqYDDw58JpFo0JULxVZ/bEPHxW0qdR3tOdbb/9adGYVUsX50LUw5d8fBr5vMtz9uIVi8YEKJg3j8VjioiIyIvn1NZv0xzjpWp1nuq8jRs38sEHH3DgwAFy5Mhh2t+9e3fi4+MfuuR3UlISEydO5KuvvqJz58706tXrqa7/KHozn8GtWrWKd999l/3793P69GlWrlzJ6tWradCgwfNOTUREREREJEu7/1L1vz1/L1y4QIECBVI8586dO/Tp04epU6fy0UcfpUshDxozn+F9+umnjBw5kvfff5/Y2FhefvllPv74Yxo3bvy8UxMREREREcnSSpQoQe7cudm5cycvv/wyALGxsfz++++0adMmxXM++ugjNm/ezLhx4554JbGnoWI+g3NycmLkyJHPOw0REREREZHnwiqN68SnhZ2dHW3atGHs2LE4OztTuHBhxowZQ4ECBahVqxYJCQlcvnwZg8FAjhw5CA8PZ/369Xz00Ue89tprpgnQAVMbS1E3exEREREREcm4rKzSvqXBBx98QNOmTfnkk09o2bIlNjY2zJw5Ezs7O86ePUvlypVZv349AGvXrgVg9OjRpsnP72/321iK3syLiIiIiIhIhvU815kHsLGxoU+fPvTp0yfZsZdeeonDhw+bPs+aNeuZ5aXZ7EVERERERCTDOrPjuzTHKFSppgUyyVj0Zl4knVl6GbXirTpbfLk3uLfk2/Urll2ezJAnT7osTXfl5N8WjQmQx60If5/95/ENU6FIQVeiT5y1aMwSLxck9uKFxzdMBYe8+Sy+LCHcW5rwnyvXLBrTNY+jReOJiIiIZFYq5kVERERERCTjeo4T4GVkeioiIiIiIiIimYzezIuIiIiIiEjGlcbZ6LMqvZl/zoKDg+nXr1+Kx/r160dwcHC652A0GgkPD39km7///huj0UjDhg2f6hrh4eEYjcanOvdJ3bhxgwULFpg+P6vnJyIiIiIi6cfKyjrNW1aUNe9KLC48PBwPDw/++OMP9u/fn+rzg4KCiIyMtHxiD5g1axYzZ840fR4wYABhYWHpek0REREREZHnQcW8PFZCQgIrV66kcePGvPLKKyxevDjVMXLkyIGrq2s6ZPc//11l0WAw4OTklK7XFBEREREReR5UzGciR44c4b333sPPz4/SpUtTq1Yt5s6dazoeFhZGcHAw06dPp2rVqnh5eRESEsJff/1lanPu3DneffddfH19qV69OuvWrXvsdSMjIzl//jwVK1akTp06rF+/nmvXzJebCggI4Ouvv6Zbt274+vri5+fH8OHDuXv3LpC8m73RaGTt2rWEhITg7e1NrVq1+P777/n++++pXbs2Pj4+dOjQgcuXL5vO+f7772nRogW+vr54eXnRtGlTfvrpJ9O9T548mdOnT2M0Gjl16lSybvZHjx6lS5cu+Pn5Ua5cOT744APOnDljOh4cHMyoUaP4+OOPKV++PGXLlqVv3778+++/T/oViYiIiIiIpVlbpX3LglTMZxI3b96kXbt22Nvbs3DhQtatW0dgYCDDhw/njz/+MLXbt28fu3fv5uuvv2bOnDmcOXOGIUOGAHD37l06dOjAlStXmD9/PuPHj2f69OmPvfaKFSt46aWXKF26NEFBQdy+fZuIiIhk7cLCwqhQoQIRERF069aNb775hrVr1z407rBhw2jdujVr166lePHifPjhh3z11VeMGTOGqVOnEhUVZcrvt99+4/333+fNN99k9erVLFu2DBcXF3r37k18fDyhoaGEhoZSoEABIiMjKViwoNm1Tp8+TfPmzbGzs2Pu3LnMnj2bS5cu0aZNG+Li4kzt5s2bR968eVm2bBnDhg1j/fr1zJkz57HPSERERERE0omVddq3LEiz2WcAa9asYePGjcn2x8fHU7ZsWeBeMR8SEkKrVq3InTs3AF27dmXatGkcPnyYkiVLAvcK9tGjR5u6lwcHBzNmzBgAfv75Z44cOcLmzZt5+eWXARgxYsQjJ7W7evUq33//Pe3atQOgaNGilCpVisWLF9O2bVuztlWqVCEkJASAIkWKsHz5cn799deHxm/UqBG1a9cGoEWLFnz//ff07NkTb29vACpVqkRMTAwANjY2fPLJJ7Ru3dp0fkhICKGhoVy6dImCBQtib2+PjY1Nit35Fy5ciL29PWPHjsXOzg6ASZMmERAQwOrVq2nVqhUAxYoVo1evXgB4eHiwbt06fv3114c+HxERERERSV9Wms0+RSrmM4CAgAB69+6dbP/YsWO5evUqAM7OzrRq1Yr169cTHR3N8ePHTW/kExMTTefkzZvXbJy4wWDgzp07AMTExODo6Ggq5AFKlixJzpw5H5rb6tWruXPnDkFBQaZ9QUFBjB07lp9//hl/f3/T/mLFipmd++C1U+Lh4WH6d44cOQBwc3Mz7cuePTvx8fGmPB0dHZk+fTrHjh3j77//Nt1/QkLCQ69xX0xMDKVLlzYV8gAuLi54eHhw+PDhR95DbGzsY+OLiIiIiIg8SyrmM4BcuXLh7u6e4v77xfzFixdp1qwZefLkoUaNGvj7++Pl5UW1atXMznmwWE3JfyeJA7C1ffifwf0l65o0aZIsxuLFi82K+ZSundL1HnXdh/3qtnv3bkJDQ6lWrRrly5enbt263Lx5k/fff/+h8f+bR0qxExISyJYtm+nz456fiIiIiIhIRqBiPpNYs2YNV69eZePGjabi8/4b5UcVzA8qVaoUsbGxHDlyhFdeeQWAY8eOcf369RTb//HHH/zxxx906dKFunXrmh0bPXo0W7Zs4Z9//kn3WeoBZs6ciZ+fH5MnTzbtmzdvHvC/+39U9xtPT0/WrFlDfHy8qWC/ePEix48fN3WxFxERERGRDCiLjnlPKz2VTKJAgQLcvHmTDRs2cObMGSIjI01ju+93RX8cPz8/ypQpw0cffcT+/fs5ePAg/fr1w9o65T+DFStWkDNnTkJDQ/H09DTbOnfuzJ07d1i+fLnF7vFRChYsyOHDh9mzZw+nTp1ixYoVTJw4Efjf/dvb23Pt2jWOHTuWrHt/y5YtiYuLo3fv3kRHRxMVFUX37t3JkydPsh8qREREREQk47CyskrzlhWpmM8k6tSpQ/v27Rk1apRpFvumTZtSoUIFoqKiniiGtbU106ZNo2jRooSGhtK5c2eCgoJwdnZO1jY+Pp41a9ZQr149HB0dkx2vUKEC3t7eLFu2zGzMfnr54IMP8PHxoUuXLjRs2JBly5YxfPhwcuTIYbr/N998E1dXV+rXr8/vv/9udr6bmxvz5s0jNjaW5s2b0759e1xdXVm0aBEODg7pnr+IiIiIiIglWSU9aR9tEXkqfy6cZtF4xVt1JvYhQyPSwsFg4PqVKxaNaciTx+K5OhgMXDn5t0VjAuRxK8LfZ/+xaMwiBV2JPnHWojFLvFyQ2IsXLBrTIW8+Ll21/ESPLk4O/HPlmkVjuuZJ/uOiiIiIZG3n9+9Mc4z8Pn4WyCRj0Zt5ERERERERkUxGE+CJiIiIiIhIxqUJ8FKkpyIiIiIiIiKSyejNvIiIiIiIiGRYWXU2+rTSBHgiIiIiIiKSYV04+GuaY+TzKmuBTDIWvZkXSWcLt+2zaLxWVX2JvXzJojEBHJxduHzNsjOaOzs6cP3aVYvGNDg6MW7VNovGBPiwQVUuHj5k0Zh5ja9y4ttwi8Z8uU5jvtsfY9GYNX08qTdinkVjAqzpH8wbg2ZaNOYPQ9rzxertFo3Zq34Vi8YTEREReRY0Zl5EREREREQkk9GbeREREREREcm4rDVmPiUq5kVERERERCTDstLSdCnSUxEAAgICCAsLe+jxfv36ERwcnO55tGjRAqPRyB9//JHqc69cucKyZctMn4ODg+nXr58l0xMREREREckQ9GZensiAAQNISEhI12scO3aMffv24eHhwaJFixg6dGiqzh89ejSnTp3i7bffBiAsLAwbG5v0SFVERERERJ4VLU2XIr2ZlydiMBhwcnJK12usWLECDw8P3n77bdasWUNcXFyqzv/vKotOTk4YDAZLpigiIiIiIpIhqJiXJ3K/m31SUhI1atRgzJgxZsdXr15NmTJlTAX4ihUrCAwMxNvbm8DAQObOnUtiYuJD4yckJLBq1SoqVapE7dq1uXHjBqtXr07WbseOHbRo0YIyZcpQtWpVxo0bR0JCAv369SMiIoJdu3ZhNBqB5N3s9+3bR0hICOXKlcPPz4+PP/6Ya9eumY4HBATw9ddf061bN3x9ffHz82P48OHcvXs3Tc9ORERERESenpWVVZq3rEjFvKSKlZUVDRs2ZN26dWZvwlevXk2tWrXInTs3S5YsYdSoUbz//vusW7eOHj16MH36dMaOHfvQuNu3b+fChQvUrl2bl156CR8fHxYvXmzW5sCBA3To0AEfHx/Cw8MZPnw4y5YtY9KkSQwYMIDAwEB8fX2JjIxMFj8qKorg4GCKFy/OkiVLmDRpElFRUYSGhpr9yBAWFkaFChWIiIigW7dufPPNN6xdu9YCT05ERERERMRyVMxLqjVq1Ihz586xe/duAC5evMjPP/9M48aNAZgyZQqdO3fmrbfews3Njdq1a9OzZ0/mz5/P7du3U4wZHh5Ovnz5KF++PAB169bl8OHD7Nu3z9Tmm2++wdvbm379+lGsWDEqV67MZ599Rr58+TAYDOTIkYNs2bLh6uqaLP6sWbMwGo0MHDiQ4sWL4+fnx7hx4/jtt9/Yvn27qV2VKlUICQmhSJEitGnThhIlSvDrr79a7NmJiIiIiEgqWVmnfcuCsuZdSbp66aWXqFChAmvWrAFg7dq1uLq68vrrr3P58mXOnTvHxIkT8fX1NW1Dhw7l9u3bnDp1Klm8y5cv8/3331OnTh2sre/9SQYGBmJtbc2iRYtM7Q4fPkyZMmXMzq1VqxatW7d+bM4xMTGULVvWbJ/RaMTBwYHDhw+b9hUrVsysjcFg4M6dO4+NLyIiIiIi6cTaKu1bFqTZ7OWpNGnShOHDh/Ppp5+yevVqGjRogLW1tanLev/+/alYsWKy8woWLJhs35o1a7hz5w7z589nwYIFpv2JiYl8++23fPzxxzg5OWFra/vU412SkpJSPDcxMZFs2bKZPtvZ2aV4roiIiIiISEaiN/PyVGrXrs3du3dZsmQJhw4domHDhgC4uLjg4uLCiRMncHd3N22HDh1iwoQJKcYKDw/H09OTVatWsXLlStM2ZMgQbt++TUREBHDvrfnBgwfNzp0zZw6NGjUCeGSh7+npyZ49e8z2RUdHExcXl+xtvIiIiIiISEanYl5Mjh8/zrZt28y2nTt3ptg2Z86c1KlTh/Hjx+Pr64uHhwdwr6Du0KED8+bNY968eZw4cYLvvvuOIUOGYGdnl+zN96FDh4iOjqZNmzZ4enqabc2aNcPd3Z3FixeTlJREhw4d2L9/PxMmTODYsWNs3bqVadOmUaNGDQDs7e25cOECJ0+eTJZv27ZtiY6OZujQoRw9epRdu3bRu3dvSpUqhb+/v4WfpIiIiIiIWIqVlXWat6xI3ezFZM2aNaZx8Pflz5+fbdu2pdi+cePGrFixwvRW/r7Q0FCyZ8/OvHnzGDVqFC4uLjRu3JiePXsmixEeHo6DgwP169dPdsza2pp33nmHoUOH8ssvv+Dv78+UKVOYNGkSM2bMwNXVleDgYLp06QJAw4YN2bx5M2+99RabN282i+Xr68v06dOZOHEiDRs2JHfu3NSsWZMPP/zQrJu9iIiIiIhkMFl0abm0skrSgGCRdLVw277HN0qFVlV9ib18yaIxARycXbh8LdaiMZ0dHbh+7apFYxocnRi3KuUfmNLiwwZVuXj4kEVj5jW+yolvwy0a8+U6jfluf4xFY9b08aTeiHkWjQmwpn8wbwyaadGYPwxpzxertz++YSr0ql/FovFERETEsi79dSTNMVyKvmKBTDKWrNnfQERERERERCQLUzEvIiIiIiIikslozLyIiIiIiIhkWE+7PHVWp2JeREREREREMi5rdShPiZ6KiIiIiIiISCaj2exFREREREQkw7py4liaY+R52cMCmWQs6mYvks5ir1+3aDwHg8HiMdMrbmaJeT/ulWuWjZvH0WDxZQQdnF3SJ88X/G/qzD+XLRoToJCrs8VjioiIiNynYl5EREREREQyLk2AlyKNmRcRERERERHJZFJVzAcEBGA0Gpk9e3aKxwcOHIjRaCQsLOyJYwYHB9OvXz/T571797Jnzx4ATp06hdFoZOfOnU8Ua+fOnRiNRk6dOvXE1/+v+/d4fytZsiTly5cnODjYlJclWSLnByUlJREREcGlS/e69oaHh2M0Gi0S+2H69euH0WikS5cuKR5ft24dRqOR4ODgJ455584d5syZY/ocFhZGQEBAWlNNJr3iioiIiIiIhVhZp33LglJ9V9myZePbb79Ntv/u3bts2rQpzWsAtmrVihMnTgBQsGBBIiMj8fX1TVPM1AoNDSUyMpLIyEi2bt3KwoULyZUrFx06dODcuXPPNJfU2r17N/369ePmzZsABAUFERkZme7XzZYtGzt27CAuLi7ZsfXr16f672Lt2rWMGDHCUumJiIiIiEgmZWVtleYtK0p1Me/v78+BAwc4e/as2f5ffvkFe3t7ChYsaLHkbGxscHV1xc7OzmIxn4S9vT2urq64urqSL18+PD09GTJkCDdv3mTTpk3PNJfU+u/iBDly5MDV1TXdr1u6dGly5MjBli1bzPbHxcWxfft2ypUrl6p4WmRBRERERETk4VJdzHt7e1OoUKFkb+fXrVtHYGCg2RvYlLp4P6pb+f22/fv3p1+/fsm62QcHBzN8+HA++ugjfHx8qFq1Kl9//fVDC7+kpCSmT59OjRo1KFOmDA0aNGD16tWpvWUAbG3vzRV4/4eFgIAAhg8fTlBQEH5+fvzyyy8kJCQwZ84cateujZeXF7Vr12bp0qVmcfbs2cPbb7+Nt7c3DRs25PDhw2bH/zvsAO51Y3+wi/rly5fp27cvfn5+lCtXjo4dO/L333+zc+dOQkJCAKhRowbh4eHJvoOrV68yZMgQqlWrhre3Ny1btjQbPhAWFkZwcDDTp0+natWqeHl5ERISwl9//fXI55MtWzZq1KjBhg0bzPZ/9913GI1G3NzczPafP3+enj17Ur58efz8/OjSpQt///03cO/vpn///gDJhllMnz7dlHtwcLDpnCe5N4AlS5ZQq1YtvL29ee+997h27ZrZ8a1bt9K4cWPKlCmDv78//fr1S9ZGRERERETkeXuqwQOBgYFmxXx8fDxbtmyhbt26aUrmfnfwjz/+mAEDBqTYZuHCheTMmZMVK1bQs2dPvvzyS6ZPn55i2/Hjx7Nw4UI++eQT1qxZQ0hICIMHD2bBggWpyuv8+fMMHToUe3t7qlatatq/aNEiPvnkE2bMmEHZsmUZOXIkU6ZMoWvXrqbrDR06lHnz5gFw8uRJQkNDKVmyJBEREbz77rt8+eWXqcrl7t27hIaGEhMTw5dffsnSpUuxsbEhNDQUX19f03wFy5YtIygoyOzchIQEQkND2bNnD6NGjSIiIoISJUrQtm1bDh48aGq3b98+du/ezddff82cOXM4c+YMQ4YMeWxugYGBybrar1+/PtnfxY0bNwgODiYhIYH58+czb9488uTJQ7NmzTh//jxBQUF8/PHHAGbDLE6fPs3evXuZNm0a8+fP559//jH9nTzJva1bt46hQ4fStm1bVq1ahY+Pj9nfwuXLl+natStNmjRh/fr1TJ48md27dzN69Ogn/n5ERERERMTCNGY+RU+1NF1gYCAzZ87k7NmzFCxYkB07dpAnTx5KlSqVpmTudwc3GAwYDIYU34gWLVqUwYMHY2VlRbFixTh69CjffPMNHTt2NGt348YN5syZw+jRo3njjTcAePnllzl9+jQzZ86kdevWD81j2rRpzJo1C7hXPMfHx1OsWDEmTJhAoUKFTO2qVatGxYoVgXvdyRctWkS/fv2oV68eAEWKFOHkyZNMnTqVNm3asHTpUvLmzcugQYOwsbGhWLFinD17NlVjw3/55Rf++OMPNmzYQNGiRQH47LPPmDlzJrGxsTg6OgLg7OxMjhw5zM6NjIzk0KFDrFmzBk9PT+DepIUHDhxg5syZTJgwwXTPo0ePxsnJCbjXW2DMmDGPza1ixYrY29uzZcsWGjRowLVr1/j555/57LPPiI6ONrVbt24dV65cYdy4cWTLlg2Azz//nJ07d7J06VK6deuGwWAAMBsiYGtry5gxY0zHWrRowfjx45/43r755huCgoJM332nTp3Yv3+/Kbfz588THx9PoUKFKFy4MIULF2bq1KkkJCQ8yVcjIiIiIiLpQUvTpeipivnSpUvj5ubGt99+S7t27Vi/fj1vvfWWpXNL0WuvvWbWld/Hx4fp06dz5coVs3Z//vknt2/fpm/fvqYu2/C/4vzWrVvJit37WrRoYerWbm1tjZOTk6mAfJC7u7vp33/99Rd37txJNja8fPnyzJ49m0uXLhETE0OpUqWwsbExHS9btmwq7h4OHz6Mg4ODqZCHewXv/a75R48efei5MTExGAwGU7ELYGVlRfny5dm+fbtpX968eU2FPNz7ceXOnTuPze3BrvYNGjRg06ZN+Pj4kD9/frN2v//+O3Fxcbz22mtm+2/fvv3I/PPmzWv2PTg4OHDr1q0nvreYmJhkvQR8fX1NxXzJkiV566236NKlCwULFqRixYpUr15ds92LiIiIiEiG81TFPPyvq32rVq3YsmULy5Yte2jbpKQkUwF+9+7dp70k8L+x6w/GBswK5Af3T5gwwazwve9Rk+o5OjqaFeoP8+CPAfev999Z2xMTE83y/u/4/v/eT0ptHiykbW1tn3rFgAe/h//m+GAeaZlwMCgoiHfffZe4uDg2bNiQrKv//et5eHjw1VdfJTtmb2//0Nj//Y4f9KT39t9ne79nwH3jxo3j/fffZ9u2bfz000/06tWLsmXL8s033zz02iIiIiIiIs/aUw8eCAwM5MCBAyxfvhw3NzeKFSuWrM39Qun69eumfcePH3/aSwKYje0G+PXXX3nppZdM3cvvK1q0KLa2tpw5cwZ3d3fTtnXrVmbOnIm1tWXHTdy/3n8nXNuzZw+urq44OjpSsmRJDh48SHx8/EPvJ1u2bGbPCzAt1QdQvHhxrl27ZvYcL1++TIUKFdi7d+8jC32j0UhsbCwxMTFm+/fu3Uvx4sWf/GYf4fXXXydXrlxERESwZ88eateunayNp6cnZ86cwWAwmL6XwoULM27cOHbv3g0k/1HkcZ7k3kqWLMnevXvNjj/4/Pfv38/w4cMpWrQobdu25euvv2b48OHs3LmTS5cupSofERERERGxDCsr6zRvWdFT31XJkiVxd3fniy++eOjEdz4+PlhbWzNhwgROnjzJjz/+aBqL/jD29vYcPXo0Wbf5+/bs2cOkSZM4duwYy5cvZ8GCBXTo0CFZO4PBQIsWLZgwYQIrV67k5MmTREREMGbMGPLmzZv6G34Mg8FAs2bNmDRpEmvWrOH48eMsWLCAhQsXEhoaipWVFS1btuTmzZt8/PHHHD16lB9++IHJkyebxSlbtiw//fQT33//PSdPnmTSpElmBaq/vz+lS5fmo48+4sCBAxw5coT+/fvj4uKCl5eX6c12dHQ0//77r1nsSpUqYTQa+fDDD9m5cydHjx5lyJAhxMTE8M4771jkOdja2lKrVi0mTJhAhQoVcHZ2Ttamfv36ODo60rVrV/bv38/Ro0fp378/W7du5ZVXXgH+94b+t99+M3Wlf5QnubdOnTqxefNmZsyYwd9//828efPYuHGjKUbu3LlZuHAhY8aM4fjx4xw+fJh169ZRpEgR8uTJY4nHIyIiIiIiqWVllfYtC0rTTxSBgYHExcWl2JUawM3NjaFDh7J161YCAwP56quvTLOUP0xoaCjz589/aLsaNWpw5MgRGjRowNSpU+nXrx8tW7ZMsW3//v1p27YtkyZNIjAwkC+//JKuXbvSrVu31N3oExowYABt2rRh3Lhx1K1bl4ULFzJw4EBCQ0MByJ8/P3PnzuXcuXM0atSIkSNH8u6775rFaNu2LbVr16ZPnz40atSIixcv0rZtW9Nxa2trpkyZQqFChWjfvj0tW7bE1taWmTNnYmdnh6enJ9WqVaNHjx4sWbLELLatrS2zZ8+mZMmSdOvWjSZNmhATE8OcOXPw8fGx2HMICgoiLi7uoT/yGAwG5s+fj4uLCx06dKBp06amiQnvF/Ovv/46ZcqUoUWLFvzwww+PveaT3Fv16tUZN24cK1asoF69emzatMn03cC9Xg9hYWH88ssvNGzYkFatWmFra8v06dMt3pNDREREREQkLaySHrZIewYUHBxM4cKFGTly5PNOReSJxf5n2ERaORgMFo+ZXnEzS8z7ca9cs2zcPI4GYi9bdoiGg7NL+uT5gv9NnfnnskVjAhRyTd4zSURERFIv9uKFNMdwyJvPAplkLHrdKCIiIiIiIpLJPPVs9iIiIiIiIiLpLotOYJdWmaqYnzdv3vNOQUREREREROS5y1TFvIiIiIiIiLxgsuhs9GmlYl5EREREREQyrKy6TnxaZarZ7EVEREREROTFcv3KlTTHMOTJ89TnJiYmMnnyZJYtW0ZsbCzlypVj0KBBuLu7p9j+ypUrDBs2jG3btgFQp04d+vfvj729/VPnkBIV8yLpLDMszZVecdMrpqWXZoN7y7NdvhZr0ZjOjg5cj7VsTIODQ7rk+aL/Tf199h+LxgQoUtA1XXIVERF50TzvYn7y5MksXLiQESNGkD9/fsaMGcPJkydZu3YtdnZ2ydoHBwdz+/ZtBg0aRGxsLAMGDKBChQqMGjUqLbeQjPoriIiIiIiISMZlbZX27SnFx8cza9YsunXrRrVq1ShRogTjx4/n/PnzbN68OVn7ffv2sWvXLkaMGMGrr76Kv78/Q4cOZdWqVZw/fz4tTyEZFfMiIiIiIiKScVlZp317StHR0fz777+8/vrrpn0ODg6UKlWK3bt3J2u/Z88eXF1dKVasmGnfa6+9hpWVFXv37n3qPFKiCfBEREREREQkS6tRo8Yjj2/ZsiXF/efOnQOgYMGCZvvz5cvH2bNnk7U/f/58srZ2dnY4OTml2D4t9GY+CwsODsZoNJptpUuXJiAggM8//5xbt249cawffviBP//8Mx2ztazw8HCMRuPzTkNERERERNIoycoqzdvTunnzJkCysfHZs2fn9u3bKbZPaRz9w9qnhd7MZ3GBgYEMGDDA9PnGjRtERkYyYsQIEhISGDhw4GNjnD59mi5duvDNN99QvHjx9EzXYoKCgqhSpcrzTkNERERERDKAh715f5wcOXIA98bO3/83wO3bt8mZM2eK7ePj45Ptv337tsVns9eb+SwuR44cuLq6mjZ3d3dat25NvXr1WLdu3RPFyIwLHty/bxERERERydwSEtO+Pa37XeYvXLhgtv/ChQsUKFAgWfsCBQokaxsfH8/Vq1fJnz//0yeSAhXzL6js2bNjbX3v67916xYTJkygRo0aeHl50bBhQ7777jsATp06ZRpfEhISQlhYGABHjx6lS5cu+Pn5Ua5cOT744APOnDljiv/333/Tvn17ypUrh6+vL+3bt+fw4cOm4zdu3GDYsGFUrlwZX19fWrduTVRUlOn4vn37CAkJoVy5cvj5+fHxxx9z7do10/GAgACmTp1K586d8fb2platWixbtsx0/L/d7I1GI0uXLqVdu3Z4e3tTpUoVpk2bZvZM1qxZQ2BgIF5eXjRt2pS5c+eqq76IiIiIyAusRIkS5M6dm507d5r2xcbG8vvvv1O+fPlk7StUqMC5c+c4fvy4ad/9c8uWLWvR3FTMv2Du3r3Ljz/+yKpVq2jQoAEAvXr1YuXKlQwYMIDVq1dTs2ZNunbtypYtWyhYsKCpSA4LCyM0NJTTp0/TvHlz7OzsmDt3LrNnz+bSpUu0adOGuLg4U8x8+fKxYsUKli1bhrW1NV27djXl0bNnT3744QeGDx/OypUr8fDwoH379ly+fJmoqCiCg4MpXrw4S5YsYdKkSURFRREaGkpi4v9+Vvvyyy/x8vJi5cqVtG7dmoEDB7J+/fqH3vvo0aNp2LAhq1atokmTJnzxxRfs2bMHuDcnQN++fWnatCmrV6+mSZMmjBs3zuLPX0REREREUifJAv/3tOzs7GjTpg1jx45ly5YtREdH07NnTwoUKECtWrVISEjgn3/+Mc1HVqZMGcqWLUvPnj2Jioril19+YdCgQTRs2NDib+Y1Zj6LW7NmDRs3bjR9vnXrFoUKFaJ9+/Z06dKFo0ePsmXLFqZOncobb7wBQNeuXTl8+DBTp06lRo0aODs7A+Do6EiuXLmYMmUK9vb2jB071jS5w6RJkwgICGD16tW0atWKEydOUKlSJV566SVsbW0ZPnw4f/31F4mJiRw/fpwff/yRGTNmmMa1Dxw4kFy5cnH16lVmzZqF0Wg0jecvXrw448aNo379+mzfvp1q1aoBUKlSJdMPBEWLFuXAgQPMnTuXoKCgFJ9Fo0aNTD9g9OjRg4ULF7J3717Kly/PzJkzqVOnDu3btwfAw8OD48ePM3v2bIt+HyIiIiIikjrPe9TvBx98wN27d/nkk0+4desWFSpUYObMmdjZ2Zl6Mo8YMYLGjRtjZWXF5MmTGTJkCO+88w7Zs2enTp069O/f3+J5qZjP4gICAujduzeJiYkcOHCAESNGULFiRbp06YKtra2p63u5cuXMzitfvvxD30zHxMRQunRps1kaXVxc8PDwMMXr2bMnw4cPZ9GiRbz++utUqVKFwMBArK2tTW18fHxM59vZ2Zn+wGNiYqhUqZLZNY1GIw4ODhw+fNhUzPv5+Zm18fHx4ccff3zos3hwrUeA3Llzc+fOHQAOHTrEm2++mewZqJgXEREREXmx2djY0KdPH/r06ZPs2EsvvWQ2nBju1UaTJk1K97xUzGdxuXLlwt3dHbj3trlAgQK0a9cOGxsbBg8e/NDzEhMTsbVN+c8jKSkJqxSWd0hISCBbtmwAtG7dmjp16rB161Z+/vlnvvjiC8LCwli5cqUpbkoxHhU/MTHRFB9Ill9SUpJpHoCUpLRExP3J/Wxtbc268IuIiIiIiGRkGjP/gnn99ddp164dixYtYtu2bXh6egKwd+9es3Z79uwxLUP338La09OTqKgosyUXLl68yPHjxylWrBgXL15k6NCh3Llzh8aNGzNmzBhWr17NP//8w65du0xvyA8ePGg6/+7du1SvXp1169bh6elpGst+X3R0NHFxcWZv1x88H+DXX3+lVKlST/VcSpQowYEDB8z2/feziIiIiIg8e4lJSWnesiIV8y+g7t27U6RIEQYNGkTBggWpVq0aQ4YM4YcffuDYsWNMnjyZLVu2EBoaCmBaDzEmJobr16/TsmVL4uLi6N27N9HR0URFRdG9e3fy5MlD3bp1cXJy4scff+STTz7hjz/+4OTJkyxcuJBs2bJRunRpPDw8ePPNNxkyZAg///wzx44dY+DAgcTHx+Pv70/btm2Jjo5m6NChHD16lF27dtG7d29KlSqFv7+/6T7WrVvH/Pnz+fvvv5kxYwabN2+mQ4cOT/VMOnbsyMaNG5k9ezbHjx8nIiKCefPmpf1hi4iIiIhImiQlJaV5y4pUzL+AsmfPzmeffcbZs2cZP34848ePp0aNGnzyySfUr1+f77//nrCwMOrUqQNAnjx5aNKkCaNHj2bixIm4ubkxb948YmNjad68Oe3bt8fV1ZVFixbh4OCAra0t06dPx9ramrZt21K3bl1++eUXvv76a15++WUARowYwWuvvUbPnj1p3LgxZ86cYdasWTg7O+Pr68v06dP57bffaNiwId27d8fX15fZs2ebdbNv2LAhmzZtol69eqxatYoJEyaYxtOnVtWqVRkyZAgLFiygbt26LFmyhBYtWphdT0REREREnj29mU+ZVVJW/ZlCsrSAgAAaNWpEt27dLBJv165d5M2bl6JFi5r2TZ06leXLl/Pdd9+lKXbs9etpTc+Mg8Fg8ZjpFTe9Yl65Zvn7z+No4PK1WIvGdHZ04HqsZWMaHBzSJc8X/W/q77P/WDQmQJGCrumSq4iIyIvm3KUraY5RwCWPBTLJWPRmXgTYsWMH7du355dffuHMmTNs2bKFuXPnmpayExERERERyUg0m70I8P777/Pvv//y0UcfcfnyZQoWLEjbtm2fegy+iIiIiIhYhvqSp0zFvGRK33//vUXj2dnZ8cknn/DJJ59YNK6IiIiIiKSNRoanTN3sRURERERERDIZTYAnIiIiIiIiGdapC5fSHOOlfC4WyCRjUTd7kXR29expi8ZzKlg43WYezwyzpDsYDFw6etiiMQFcihm5cPmqRWPmc3bi7MW0z776oIJ586TL95ReKwRYOm4ex/SZzX7Hob8sGhOg0qtFM81s/iIiIpL5qJgXERERERGRDEt9yVOmMfMiIiIiIiIimYzezIuIiIiIiEiGpWneUpbl3syvWbOG5s2b4+vri6+vL02aNGHx4sVmba5cucKyZcvSPZewsDCMRiP16tVL8fj+/fsxGo0EBAQ8ccykpCQiIiK4dOneJBDh4eEYjUaL5Ps4H374IUajke++++6pzg8ICCAsLMzCWZnbu3cve/bsAeDUqVMYjUZ27tyZrtcUEREREZH0k5iUlOYtK8pSxfzy5cv59NNPadKkCeHh4axYsYLGjRvz+eefM3nyZFO70aNHs3r16meSU7Zs2YiJieGvv5JPrrR+/XqsrKxSFW/37t3069ePmzdvAhAUFERkZKRFcn2U69ev89133+Hh4cGiRYueKsby5csJDQ21cGbmWrVqxYkTJwAoWLAgkZGR+Pr6pus1RUREREREnrUsVcwvXLiQpk2b0qxZMzw8PChatCitW7embdu2fPPNN6Z2z7KbRr58+ShevDjffvut2f6kpCS+/fZbypcvn6p4/809R44cuLq6pjnPx1m7di3W1ta8//777Nixg5MnT6Y6hrOzM7ly5UqH7FJmY2ODq6srdnZ2z+yaIiIiIiIiz0KWKuatra359ddfuXbtmtn+jh07smTJEgD69etHREQEu3btMnVPT0hIYM6cOdSuXRsvLy9q167N0qVLTefv3LkTo9HIli1bePPNN/Hx8aFt27YcPXr0ifKqU6cOGzZsMNu3Z88eEhMTqVChgtn+q1evMmTIEKpVq4a3tzctW7Y0dRvfuXMnISEhANSoUYPw8PBk3ewfdT7c6/ofHBzM9OnTqVq1Kl5eXoSEhKTYc+BB4eHh+Pn5UbNmTXLmzJls6EJ4eDgBAQFERERQq1YtSpcuTZMmTdi3b5+pzYPd7MPCwkw/slSuXBkfHx969erFP//8w0cffYSvry/VqlUjIiLCdH5sbCyDBg2iWrVqvPrqq1SqVIlBgwZx69YtANNz6N+/P/369UvWzf5Jv+etW7fy1ltvUbp0aerWrcsPP/zwyGcjIiIiIiLpJ8kCW1aUpYr5jh078scff1C1alU6derE119/TVRUFAaDAQ8PDwAGDBhAYGAgvr6+pu7pI0eOZMqUKXTt2pU1a9YQEhLC0KFDmTdvnln8zz//nAEDBrBkyRJsbW0JCQnh+hOs9xsUFJSsq/369eupU6cO1tb/+woSEhIIDQ1lz549jBo1ioiICEqUKEHbtm05ePAgvr6+pmJ42bJlBAUFmV3ncefft2/fPnbv3s3XX3/NnDlzOHPmDEOGDHlo/n/++SdRUVHUrl2bnDlz8sYbbxAeHk58fLxZuwsXLrB48WLGjBnDkiVLsLa2pm/fvg/tCbFnzx727NnD3LlzmTBhAhs3buStt96iZMmSrFixgqpVqzJw4ECuXLm3Tnffvn2Jiopi0qRJbNy4kf79+xMeHm76oeb+9/nxxx8zYMCAZNd70u95zJgxDBgwgPDwcNzc3Ojduzf//vvvQ5+PiIiIiIikH42ZT1mWKuZr167NkiVLePPNNzl48CDjxo3j7bffpk6dOuzduxcAg8FAjhw5yJYtG66ursTFxbFo0SI++OAD6tWrR5EiRWjdujVt2rRh6tSpZoVov379qFatGkajkbFjx/Lvv/+ybt26x+ZVrFgxPD09TV3tExIS2LRpE3Xr1jVrFxkZyaFDhxg3bhyvv/46xYoVY+DAgXh6ejJz5kzs7OxwdHQE7nVZz5EjR6rOv+/u3buMHj2aEiVKUK5cOYKDg03PJyUrVqzAzs6OmjVrAlC3bl0uX77Mpk2bzNrduXOHwYMH4+Pjw6uvvkrnzp05fvw4//zzT4pxExMTGTZsGMWKFaN69eqULFmSokWL0q5dO4oWLUrbtm2Jj4/n+PHjAFSqVIkRI0ZQpkwZXnrpJd566y1effVVDh8+DGAabmAwGDAYDGbXSs333KNHD/z9/fH09KRHjx7ExcURExPz0OcjIiIiIiLyrGWpYh7A29ubMWPG8NNPP7F69Wo+/PBD/v33Xzp27GiaAf5Bf/31F3fu3KFcuXJm+8uXL8/FixfNznnttddM/3ZycqJIkSLExMSwZ88e0+z5vr6+yYp0MO9qv2vXLrJnz55sYraYmBgMBgOenp6mfVZWVpQvX95UsD7Kk56fN29enJycTJ8NBgN37txJMebdu3dZvXo1VapUMRXIVapUwcHBIcWJ8IoVK2YWF3hobBcXFxwcHEyfc+bMiZubm+lz9uzZAbh9+zZwb3K7U6dOMWrUKN5//31q1arF/v37SUxMTPmBPCA133PRokVN/86dO/cj70FERERERNJXUlJSmresKMusM3/u3DmmT59Op06dyJ8/P1ZWVhiNRoxGIzVq1CAoKIjdu3dTp04ds/Puf7H/nVX+foFoa/u/R/Tgv++3sba2pnTp0qxcufKh7eBeV/tJkybx119/sX79+mRd5O/nktLs9omJiSnGfNrzUzMh3I8//sjFixf5/vvvKVWqlGl/QkICe/bs4c8//6R48eKPjP2w/3iyZcuWbN+Dww7+G6NLly4cPnyYevXqUbt2bXr16sWnn376RPeRmu85NfcgIiIiIiLyPGSZYt7Ozo4lS5ZQoEABOnbsaHbs/tvVvHnzAuYFXdGiRbG1tWXPnj2UKFHCtH/Pnj24urqaurUDHDx4EH9/fwAuX77M8ePHadeuHTly5MDd3f2R+Xl4eFCiRAk2bNjApk2bmDNnTrI2RqOR2NhYYmJizN6u792711QwP2opuyc5P7VWrFhBnjx5mDNnjlmhferUKd59910WL17MJ5988lSxU+P3339n69atLF26lDJlygD33pafOHHC7G3+w6TmexYRERERkYxD79VSlmWKeWdnZzp06MCECROIi4ujTp065M6dmz///JMpU6bg5+dnWgbO3t6eCxcucPLkSdzc3GjWrBmTJk3C0dERb29vIiMjWbhwIb169TIrnocMGcJnn32GwWBg9OjRuLq6JnvT/yiBgYFMnz4dV1dXSpYsmex4pUqVMBqNfPjhh3zyySfkzZuX+fPnExMTw6BBg0y5A0RHR5MnT55Un58aly5dYtu2bbRv396sAAbw9PTEz8+PlStX8uGHH6Y6dmrlzZsXW1tbNmzYgLOzM1evXmXq1Kn8888/ZhPx2dvbc/ToUdOkefcZDIYn/p5FREREREQyuixTzMO9icuKFCnC0qVLWbBgAbdu3aJgwYIEBQXRuXNnU7uGDRuyefNm3nrrLTZv3syAAQPIkycP48aN4+LFi7i7uzNw4ECaNWtmFv/tt9+md+/exMbG8vrrr/PNN9+QM2fOJ84vKCiI8ePH07Zt2xSP29raMnv2bEaNGkW3bt2Ij4/n1VdfZc6cOfj4+AD3iuhq1arRo0cPevXqZTb2/UnOT41Vq1aRlJREy5YtUzzevn17OnXqxNq1a7GxsUl1/NTInz8/I0eOJCwsjAULFuDq6kr16tVp27YtW7ZsMQ0xCA0NZcaMGfz111/JZrR/0u9ZREREREQyjqw6G31aWSVpMPBj3V/ffcuWLbz00kvPOx3JZK6ePW3ReE4FCxP7BEsippaDwcDla7EWjens6GDxXB0MBi4dffyEkKnlUszIhctXLRozn7MTZy9eeXzDVCiYN0+6fE9Xrln+byqPo8HicfM4GtLlb2rHob8e3zCVKr1aNF1yTY+YIiIiGVn0ibNpjlHi5YIWyCRjyXKz2YuIiIiIiIhkdSrmRURERERERDKZLDVmPr34+fk90TrvIiIiIiIiYlkaGJ4yFfMiIiIiIiKSYWkCvJSpmBcREREREZEMS3O2p0yz2YuIiIiIiEiG9duxtK8OVdqjsAUyyVj0Zl4kne0/etKi8XyKuRF76aJFYwI4uOTl3CXLLqNWwCWPxXN1cMnL2t2/WzQmwFsVShF78YJFYzrkzcfp7ZstGrNwlVocPWPZPIsVyscPUUcsGhPgDe9X+HZvtEVj1ilXwiL/D/1BpT0K8+fCaRaNCVC8VWcuXbXsMoIuTg7pEvPk+UsWjQnglt/F4jFFRETkf1TMi4iIiIiISIaVqL7kKVIxLyIiIiIiIhlWEqrmU/JCrTN/9+5d5s6dS+PGjfH19cXPz4927drx888/W/xaZ86cYd26dRaPm5UEBwdjNBpT3D7//HNTm379+j1RvBs3brBgwQLT5379+hEcHJwuuYuIiIiIiDxPL8yb+fj4eNq1a8fZs2fp1q0bvr6+3Lp1ixUrVhAaGsqIESNo2LChxa7Xt29fChcuTN26dS0WMysKDAxkwIAByfbnzJkTgLCwMGxsbJ4o1qxZswgPD6d169YADBgwgISEBMslKyIiIiIiz5zmbE/ZC1PMT5o0iejoaNatW0eBAgVM+wcMGMCNGzcYPnw4tWrVIleuXM8xyxdPjhw5cHV1fehxJyenJ4713//IDQbD06YlIiIiIiKSob0Q3ezv3LnDsmXLaNq0qVkhf1/37t2ZMWMGOXLkAODq1asMGTKEatWq4e3tTcuWLdmzZ4+pfVhYGMHBwUyfPp2qVavi5eVFSEgIf/31F3Cva/iuXbuIiIggICAAuNczYMyYMVSpUgVfX1+aNWtGZGSkKWZ4eDi1atVi/fr1BAQE4O3tTfv27Tl//jyff/45FSpUoGLFikyb9r8Zl/v160evXr0YPnw45cqVw9/fn5EjRxIfH29qc/bsWXr37k2lSpXw8fGhffv2HD582CxG165dCQ0NpWzZskybNo2kpCRmzJhBYGAgpUuXply5cnTu3JmTJ/83K7vRaGT8+PG88cYbVKpUia+//hpfX19u3rxpapOYmEjVqlX55ptvnvq7+283+99++4127drh6+tLxYoVGThwIDdu3CAsLIzJkydz+vRpjEYjp06dStbN/ujRo3Tp0gU/Pz/KlSvHBx98wJkzZ8yuNWrUKD7++GPKly9P2bJl6du3L//+++9T5y8iIiIiIpIeXohi/uTJk1y9ehUfH58Uj+fLlw9vb29sbGxISEggNDSUPXv2MGrUKCIiIihRogRt27bl4MGDpnP27dvH7t27+frrr5kzZw5nzpxhyJAhwL1i39fXl8DAQJYvXw5A//792b59O2PGjCEiIoLAwEC6dOnCjz/+aIp59uxZFi1axJQpU5g9ezYHDx6kfv362NrasnTpUlq0aMEXX3xBTEyM6ZxNmzZx5swZFi1axLBhw1i5cqVpvHlcXBwtW7bk/PnzfPXVVyxevBh7e3vatGljVsRu3ryZihUrsmLFCurXr8/cuXOZNm0affr0YePGjUyZMoVjx44xcuRIs+e2ZMkSJk2axJdffknTpk25c+cOmzZtMh3/6aefuHz5Mm+99dbTfXH/cerUKYKDg3F2dmbJkiVMnjyZnTt3MnDgQEJDQwkNDaVAgQJERkZSsGBBs3NPnz5N8+bNsbOzY+7cucyePZtLly7Rpk0b4uLiTO3mzZtH3rx5WbZsGcOGDWP9+vXMmTPHIvmLiIiIiEjqJSalfcuKXohu9teuXQPA0dHxsW0jIyM5dOgQa9aswdPTE4CBAwdy4MABZs6cyYQJE4B7k+mNHj3a1A08ODiYMWPGAPe6hmfLlo0cOXLg7OzM8ePHWbt2LcuXL8fLywuAdu3aER0dzcyZM6levTpwrwfBp59+arquv78/+/fv56OPPsLKyorOnTvz5ZdfcuTIEVMbR0dHxowZQ86cOfH09OTChQt8/vnn9OnTh9WrV3PlyhXCw8NxdnYGYOzYsdSsWZMFCxbQp08fU4wOHTqYnsHLL7/MyJEjTb0KChcuTGBgYLIJ/Ro0aGC6H4CAgABWr15NgwYNAEw9E+5fOyVr1qxh48aNZvt8fX2ZNWtWsrZLly7F0dGRkSNHki1bNgCGDRvGrl27yJUrF/b29tjY2KTYbX/hwoXY29szduxY7OzsgHtDL+7n3KpVKwCKFStGr169APDw8GDdunX8+uuvD81fRERERETSl8bMp+yFKObvF5NXr159bNuYmBgMBoOpWAawsrKifPnybN++3bQvb968ZuO5DQYDd+7cSTHm77//DkBISIjZ/jt37uDg4GC2z8PDw/TvnDlz8tJLL2FlZQVA9uzZAbh9+7apjZeXl2myOLhXCN+5c4djx44RExNDkSJFzIrp7Nmz4+3tbdbV3t3d3SyHgIAADhw4wKRJkzh+/DhHjx7lyJEj5M+f36zdf89r0qQJXbp04fz58+TKlYvvvvuOiRMnpvhMHrxW7969zfbdH+7wX4cPH+bVV181FfIAFSpUoEKFCo+8Btz7XkuXLm0q5AFcXFzw8PAwexbFihUzO89gMBAbG/vY+CIiIiIikj5UzKfshSjm3dzcyJs3L/v27SMoKCjZ8b///puhQ4fSt29fkpKSTMXzgxITE7G1/d/jerAofJz7f3wLFixINsGetbX5SIcHC9WUjv/Xf9snJiYCYGNj89B7SUhIMLuX/xbP06dPJywsjMaNG/Paa68RHBzMli1bkr2Z/+95lStXxtXVlXXr1uHk5ITBYKBKlSqPzD9XrlzJfhR4GFtb2xTv50k86lk8+AxT872KiIiIiIg8Ly/EmHlra2uaNm1KeHg458+fT3Z8xowZ7N+/n8KFC2M0GomNjTUblw6wd+9eihcv/lTXf+WVVwC4cOEC7u7upi08PJwVK1Y8Vcz7Dh06ZLb82r59+8iZMyceHh54enpy7NgxLl26ZDp++/Ztfvvtt0fey1dffUXXrl0ZPHgwzZs3x8fHh7///vuxv4jZ2NjQsGFDNm3axKZNm2jQoMETLyv3JIoXL87vv/9udr+bN2+matWq3Lx585GFvqenJ1FRUWaTA168eJHjx48nexsvIiIiIiKS0b0QxTxAly5dcHd3p0WLFqxcuZITJ05w8OBBBgwYwIoVK/jss8/InTs3lSpVwmg08uGHH7Jz506OHj3KkCFDiImJ4Z133nni6+XKlYvTp09z7tw5XnnlFd544w0GDRrEli1bOHnyJDNnzmTatGm4ubml6b5Onz7N4MGDOXr0KJs3b2bSpEm0adOGnDlzUq9ePRwcHOjRowdRUVFER0fTp08fbty4QfPmzR8as2DBguzYsYM///yTv/76i/Hjx7Np0yazQvhhmjRpwoEDB/jpp59o3Lhxmu7tv1q1asWVK1cYNGgQR48eZc+ePYwdO5ZKlSqRM2dO7O3tuXbtGseOHUs25KFly5bExcXRu3dvoqOjiYqKonv37uTJk4e6detaNE8REREREbGcxKSkNG9Z0QtTzOfMmZP58+fTpEkTpk+fToMGDejYsSPnzp1j7ty5poLO1taW2bNnU7JkSbp160aTJk2IiYlhzpw5D50NPyUtWrQgJiaG+vXrk5CQwPjx46lduzaDBg0iKCjI9ANCkyZN0nRfPj4+WFlZ0aRJEz777DNCQkJME7g5ODgwf/58DAYDbdu2pVWrVty8eZNFixY98keE0aNHc+vWLZo0aUKbNm2IiYlhyJAhXLp0iVOnTj0yH3d3d3x8fChZsqTF33jnz5+fWbNmcezYMRo1akSPHj2oVq0agwYNAuDNN9/E1dWV+vXrm+YpuM/NzY158+YRGxtL8+bNad++Pa6urixatCjZvAUiIiIiIpJxJCWlfcuKrJI0m0Cm1a9fP06fPs28efOedyomSUlJvPnmm3Tq1Im33377eaeTIew/etKi8XyKuRF76aJFYwI4uOTl3KUrFo1ZwCWPxXN1cMnL2t2/P75hKr1VoRSxFy9YNKZD3nyc3r7ZojELV6nF0TOWzbNYoXz8EHXEojEB3vB+hW/3Rls0Zp1yJfjt2GmLxiztUZg/F06zaEyA4q06c+mqZSfQdHFySJeYJ89fenzDVHLL72LxmCIi8mLaGf13mmP4lSiS5hgZzQsxAZ6kvzt37vD999/zyy+/EBcXp67rIiIiIiJiEVm1m3xaqZgXi8iWLRvDhg0DYMyYMdjb2z/njERERERERLIuFfOZ2MiRI593Cma2b9/+vFMQEREREZEsRiPDU/bCTIAnIiIiIiIiklXozbyIiIiIiIhkWHoxnzLNZi8iIiIiIiIZ1vbfjqY5RpXSll02OyPQm3mRdHbh8lWLxsvn7GTxpakg/Za8ir1+3aIxHQwGrlyzbEyAPI4Grl+x7NJ8hjx50iXmz38cs2hM/5Ie6bY02d9n/7FozCIFXdNnubc/LbuEHoBL8RLp8vefHjHT639TJq3dYdGYH7xVyaLxREREMjONmRcRERERERHJZPRmXkRERERERDKsJDQyPCUq5kVERERERCTD0ixvKVM3+0wqICAAo9Fo2kqXLk316tUZOnQoVyw8RtdSjhw5wo8//mj6bDQaCQ8Pt0jspKQkIiIiuHTp3rjf8PBwjEajRWKLiIiIiMjzk5iUlOYtK1Ixn4mFhoYSGRlJZGQkGzZs4NNPP+Wnn34iJCSEuLi4551eMp07d+bgwYOmz5GRkQQFBVkk9u7du+nXrx83b94EICgoiMjISIvEFhERERERyWjUzT4Ts7e3x9XV1fTZzc2NkiVLUrduXWbOnEn37t2fY3aP92DuafXfFRZz5MhBjhw5LBZfREREREQkI9Gb+SymUKFC1KpVi7Vr1wL3urKPHz+eN954g0qVKvHXX39x7tw5evfuTcWKFXn11VepVq0a48ePJzExEbjXRb1WrVqsX7+egIAAvL29ad++PefPn+fzzz+nQoUKVKxYkWnTppmuGx8fz7hx46hZsyalS5fGz8+PXr16mbr8BwQEcPr0aSZPnkxwcLApt/DwcE6dOmU2ZODBLSIiArjXRf+9997Dz8+P0qVLU6tWLebOnQvAzp07CQkJAaBGjRqEh4cn62Z/9epVhgwZQrVq1fD29qZly5bs2bPHdDwsLIzg4GCmT59O1apV8fLyIiQkhL/++iu9vioREREREXkCSUlJad6yIhXzWZCnpycnTpzg33//BWDJkiVMmjSJL7/8kqJFi9K5c2cuX77MzJkz+fbbb+nQoQNTp07l+++/N8U4e/YsixYtYsqUKcyePZuDBw9Sv359bG1tWbp0KS1atOCLL74gJiYGgNGjR7N27Vo+//xzNm7cyKhRo9ixYwdfffUVAMuXL6dAgQKEhoYSFhZmlm/BggVNwwUiIyPZtm0b5cuXx9PTk1q1anHz5k3atWuHvb09CxcuZN26dQQGBjJ8+HD++OMPfH19TTGXLVuWrOt+QkICoaGh7Nmzh1GjRhEREUGJEiVo27atWbf/ffv2sXv3br7++mvmzJnDmTNnGDJkiOW/IBERERERkTRSMZ8FOTg4AJjGzTdo0AAvLy98fHy4desWDRo04LPPPqNkyZK4ubkRHBxMvnz5OHz4sCnGnTt3+PTTTylRogTlypXD39+fHDly8NFHH+Hh4UHnzp2Be2/MAby8vBg1ahR+fn4ULlyY6tWrU7lyZVNMZ2dnbGxssLe3x8nJySxfGxsbXF1dTduMGTM4duwY/8fenYdVVbX/H3+rYIoMAiIqIjiSOZImmpmKmkmaU5kTZKhpZplTgZazaA5pUg4phnNOOA/p18zUx5zKOSQRERxQcAAfB1D4/eGP84Sgou6jYJ/XdZ3rgr3Xvve99jlY99lrrzV9+nSsra25ceMGfn5+DBs2jLJly+Lm5kbv3r0BOH78OPnz58fOzs50nnuH1+/YsYOjR48yceJEateuTdmyZRkyZAgVKlQgJCTE1O727duMGzfO1GdfX1/2799v1NsiIiIiIiKPITXtyV/PIz0z/xxKSkoCwNraGgA3NzfTvgIFCtC5c2c2btzInDlziI6OJjw8nAsXLpiG2acrXbq06eeCBQtSsmRJ8uTJA8ALL7wAwK1bt4C7Xxjs2rWLb775hlOnThEZGcnJkyepWbPmI+W+YMECfvrpJ+bMmYOLiwtwt0Dv2LEj69evJzw8nOjoaP766y+ATDlnJSIiAhsbGypUqGDalidPHmrWrMn27dtN24oUKZLhiwYbGxtSUlIeKX8RERERETHW8zpM/knpzvxz6OjRo7i7u1OoUCGADHeqb9y4Qfv27Zk2bRrW1ta0bNmSBQsWUKxYsUxxLC0tM/yeN+/9Py7Dhg2jT58+3Lx5kwYNGjBhwgTeeuutR8r7t99+IygoiFGjRvHyyy+btsfHx/P222+zePFiihQpQvv27R9pSbu0tDTTlxD/lJqaioXF/77Pyp8//yPlKyIiIiIicuvWLYYPH06dOnXw9PTk008/NS2ZfT9//PEHvr6+1KhRg3r16jF48GCuXLnySOfVnfnnzPnz59myZQvdu3fPcv/27ds5evQoO3fupEiRIsDdyeESEhIe+xuvy5cvs2jRIiZNmpThefWTJ09iZWWVrRjHjx+nb9++dOvWjZYtW2bYt2bNGq5cucLPP/9s+oIhffh+es5ZFevpPDw8SExMJCIiIsPd+f3791OuXLnsdVJERERERCQLw4YNY//+/QQHB5M/f36GDh1Knz59mD9/fpbto6Ki6Nq1K++88w7Dhw/n0qVLDB8+nE8//ZS5c+dm+7wq5nOx69evc/HiRQBu3rzJ8ePHmTx5MiVLluSDDz7I8pj0O/CrV6+madOmnDt3jm+++YaUlBSSk5MfKw8bGxtsbGzYsmULlSpV4ubNm8yfP5+jR49SrVo1U7tChQpx6tQp4uPjTV8kAFy8eJGePXtSu3Zt3n//fVOf4O6ogmLFinHjxg02bNhAzZo1OXnyJGPGjAEw5Zz+pUF4eDj29vYZ8qtbty4eHh7079+fL7/8kiJFijB//nwiIiIYOnToY/VZRERERESejpw8zD4uLo6VK1cyY8YM0yPG33zzDW+++SYHDhygevXqmY5ZuXIlRYsWZdCgQeTJk4cyZcowdOhQOnXqRExMDK6urtk6t4r5XGz27NnMnj0buFvMFitWjDfeeAN/f3/TEPt7Va1alcDAQEJDQ5k8eTLOzs74+PhQvHhxDh48+Fh5WFhY8O233zJ27FhatGiBnZ2daWm66dOnc/36daysrPD19eXrr7/m77//ZvXq1abjt2/fztmzZzl79iz/93//lyF269atGTNmDEePHuXrr7/m2rVruLi48O6777JlyxYOHTpEhw4dqFChAvXr1+ezzz6jX79+GZ59t7Cw4Mcff+Trr7/mk08+ITk5mUqVKhEaGprlH5eIiIiIiOQcRkxg16hRowfu37Jly2PFTZ8w28vLy7StdOnSODs7s3fv3izrjbfffpuGDRtmObr4ypUr2S7m86Tl5K85RJ4DFy5dMTReUYfCJFxJNDQmgGNhW8PjOha2JfH/T8hoFFsbGy5fNTYmgL2dDUmXLxsa08be3iwxd/0VZWjMOhVLExP34Oe6HoersyOnzl18eMNH4F7cySyf04QT4YbGBHAs96JZPv/miGmuf1OmrN1paMxPm9c1NJ6IiOQOG/b99cQxJnzR+4H7H7eY//HHH5k5cyb/+c9/Mmx/5513qFq1KkOGDMlWnB49enD06FG2bt2aae6y+9GdeREREREREXmuPW6xHhsb+8C7+n369MlyIu0XXnjBtPLXw4wdO5Zt27YxZcqUbBfyoGJeREREREREJEvOzs6sX7/+vvu3bduW5dxjt27domDBgg+MnZKSwpAhQ1ixYgVDhw7ljTfeeKTcVMyLiIiIiIhIjpX6DJ8Mt7S0pGzZsvfdf/z4ca5cuUJycnKGO/QXLlzIcvnvdNeuXaN3797s27ePiRMnPvKy3qB15kVERERERCQHSzPgZS41atQgNTXVNBEe3F2iOy4uzjS7/b2Sk5Pp0aMHhw8fZtasWY9VyIMmwBMREREREZEcbO3eY08co/krLxmQSdb69+/PgQMHCAoKomDBggwdOhRra2vmzZsH3C3er169ip2dHfnz5yc4OJjvv/+eiRMnUqtWrQyx0ttkh4p5ETO7HHPK0Hj2ru4kxl8wNCaAbZGinN35fw9v+AhK1G1sntm8Lxk/87qtgyNnL14yNGYJJwezzOZ++ny8oTFLFStittnMzycYO5t/MUd7s3ymrpw/a2hMgMLFSuSa2ezNtUKEOVZeiL1g7N9/yaKOhsYTERHjrdlz9IljtKhVyYBMsnb9+nWCgoL4+eefAXj99df58ssvsbe3B2D37t34+fkxd+5cvLy8aNq0KadOncoyVnqb7NAz8yIiIiIiIiKPycrKilGjRjFq1Kgs93t5eXH8+HHT7+lF/5NSMS8iIiIiIiI51rOcAC8n0wR4IiIiIiIiIrmMinl5Knx9ffHw8MjyNXr06CeOn5KSQmho6JMn+g9hYWF4eHgYGlNERERERB5NWtqTv55HGmYvT02zZs0YPHhwpu0FCxZ84thr165lzJgxdOnS5YljiYiIiIhIzqE527OmYl6emgIFCuDk5GSW2PoDFxERERGRfxMNs5ccITExkaFDh1K/fn0qVapE3bp1GTp0KDdv3jS1iYmJ4eOPP6ZGjRp4eXnRt29f4uPjCQsLIzAwEAAPDw92795NcHAw3t7eGc5x77D58+fPM2DAAF599VUqVapE/fr1mTRpEqmpqU+n0yIiIiIiIo9Jd+YlR/jiiy84f/48U6ZMwdHRkQMHDhAYGEiZMmV4//33SUpKomPHjpQrV47Q0FAsLCwYOnQon3zyCT/++CNJSUkEBQWxY8cO7Ozs2LNnz0PP2aNHDxwdHQkJCcHa2ppff/2VUaNGUaVKFRo3bvwUei0iIiIiIg+j2eyzpmJenpo1a9ZkWlPR09OT2bNnU7duXWrWrMmLL74IQMmSJZk/f75pPcb169eTlJTEpEmTKFy4MACjR49m1apV5MmTBxsbG4BsD+O/efMmLVu2pGnTpri4uAB3J+n74YcfOH78uIp5EREREZEcQo/UZk3FvDw13t7eDBgwIMO2AgUKANCxY0d++eUXVq1axenTp4mIiCAmJgZ3d3cAjh8/jru7u6mQByhfvnymeNlVoEABOnfuzMaNG5kzZw7R0dGEh4dz4cIFDbMXEREREZEcT8/My1NTqFAh3NzcMrycnZ1JS0ujZ8+ejBw5knz58tG0aVOmT5/Oyy+/bDrWwsKCPHnyPNL57v0G7/bt26afb9y4Qfv27Zk2bRrW1ta0bNmSBQsWUKxYsSfrpIiIiIiIGCo17clfzyPdmZdn7tixY2zbto0lS5ZQrVo14O668adPn8bV1RWAcuXKsXTpUpKSkkxD6o8dO8YHH3xAWFhYpkLf0tKSa9eukZaWZtoXHR1t2r99+3aOHj3Kzp07KVKkCABXrlwhISFBw3hERERERCTH0515eeaKFCmChYUFGzZsICYmhsOHD/PZZ59x8eJFkpOTAWjRogV2dnYMHDiQ8PBwjhw5wrBhw6hQoQIuLi5YWVkBcOTIEW7evMnLL79MYmIiP/zwA7GxsaxZs4awsDDTOdPvwK9evZozZ86wb98+evXqRUpKiumcIiIiIiLy7KWlpT3x63mkYl6eOWdnZ8aOHcsvv/yCj48Pffr0wdnZmS5dunD48GHS0tIoWLAgISEh3Llzhw4dOtC1a1fKli3LlClTAKhduzbVqlWjffv2bN26lVq1atG3b1/mz5+Pj48PK1eu5IsvvjCds2rVqgQGBjJ37lyaNWtGYGAgr7zyCs2bN+fgwYPP6lKIiIiIiIhkS5605/VrCpEc4nLMKUPj2bu6kxh/wdCYALZFinJ25/8ZGrNE3cYkJiUZGtPWxobESwmGxgSwdXDk7MVLhsYs4eTAqXMXDY3pXtyJ0+fjDY1ZqlgREq4kGhoTwLGwLecTLhsas5ijvVk+U1fOnzU0JkDhYiXM8/k3Q8zLV42NCWBvZ8Ouv6IMjVmnYmliLxj791+yqKOh8URExHg//fbnE8do/7qnAZnkLHpmXkRERERERHIsrTWVNQ2zFxEREREREcllVMyLiIiIiIiI5DIaZi8iIiIiIiI5lqZ5y5qKeREREREREcmxVMxnTbPZi4iIiIiISI41d+v+J47h17CGAZnkLLozL2JmRi+jZuvgyKWrxi8j5mBna/jyZI6Fbc2yjFZC5HFDYwI4lvUwS64nzsQZGrOcizMX/zpkaEynilU5v2+noTEBitWsmyuWJrR1cOSv6HOGxgSo6FY81yxNd+XcGUNjAhQu7sKPW/YaGvODRq8QftrY9+rFUsUNX5YS7i5NKSIiYk6aAE9EREREREQkl9GdeREREREREcmx9GR41lTMi1n4+vqyZ8+eLPf5+fkxePDgJz7H9evXWbFiBZ06dXriWCIiIiIikjOlqpjPkop5MZtmzZplWbQXLFjQkPizZ88mLCxMxbyIiIiIiPzrqJgXsylQoABOTk5mi6/hNiIiIiIi8m+lCfDkmUhMTGTo0KHUr1+fSpUqUbduXYYOHcrNmzcB2L17Nx4eHmzbto3mzZtTuXJl3nrrLbZu3QpAcHAw3333HWfOnMHDw4PY2FiSk5OZOHEijRs3pnLlynh5edGvXz8uX75sOu/KlSt56623qFKlCvXq1WP06NEkJyeTkpJCnTp1+O677zLkuWjRIl599VVSUlKe3sURERERERGTtLQnfz2PVMzLM/HFF19w6NAhpkyZws8//0xgYCBhYWEsXrw4Q7vx48czePBgwsLCcHV1ZcCAAfz3v//F398ff39/ihUrxo4dOyhevDjjxo1j7dq1jB49mp9//pmvv/6anTt3Mm3aNADCw8P58ssv+eSTT/j5558JCgpi1apVzJo1C0tLS95++21Wr16d4fyrVq3i7bffxtLS8qldGxERERER+Z+0tLQnfj2PNMxezGbNmjX8/PPPGbZ5enoye/Zs6tatS82aNXnxxRcBKFmyJPPnz+f48Yzrh3/22WfUqVPH9HPLli2JiIjA09MTKysr8uXLZxrKX6VKFd544w1q1aoFgIuLC6+99popZmxsLHny5KFkyZKUKFGCEiVKEBISgrW1NQDvvPMOoaGh/Pnnn3h6enLq1Cn+/PNPhg8fbr6LJCIiIiIi8hhUzIvZeHt7M2DAgAzbChQoAEDHjh355ZdfWLVqFadPnyYiIoKYmBjc3d0ztC9Tpozp5/Si+35D3lu2bMmuXbv45ptvOHXqFJGRkZw8eZKaNWsCUK9ePTw9PWnbti3u7u68+uqrNGrUiMqVKwNQvnx5qlSpwsqVK/H09GTFihVUrlwZDw8PQ66HiIiIiIg8Os1mnzUNsxezKVSoEG5ubhlezs7OpKWl0bNnT0aOHEm+fPlo2rQp06dP5+WXX84UI3/+/Jm23W+YzLBhw+jTpw83b96kQYMGTJgwgbfeesu0/4UXXmDu3LmsWLGCtm3bEhkZSffu3Rk0aJCpTdu2bdmwYQPJycmsWbOGNm3aGHAlREREREREjKU78/LUHTt2jG3btrFkyRKqVasG3L3bfvr0aVxdXbMdJ0+ePKafL1++zKJFi5g0aRI+Pj6m7SdPnsTKygqAbdu2cfjwYXr37s1LL73Ehx9+yLRp05g+fTpjxowBoHnz5owdO5bQ0FAuXrxI8+bNjeiyiIiIiIiIoVTMy1NXpEgRLCws2LBhAw4ODly5coXp06dz8eJFkpOTsx3HysqKq1evEhUVRcmSJbGxsWHLli1UqlSJmzdvMn/+fI4ePWr6wsDCwoLvv/8ea2trGjVqxJUrV9i6dSuenp6mmDY2NjRp0oTvv/+exo0bY2dnZ3j/RUREREQk+zTKPmsaZi9PnbOzM2PHjuWXX37Bx8eHPn364OzsTJcuXTh8+HC2Z5t84403cHJy4u233+bYsWN8++23RERE0KJFC7p168aNGzfo168ff//9N9evX6du3bqMHj2aZcuW0bx5c7p164a7uzvffPNNhrht2rTh5s2bGmIvIiIiIpIDpKalPfHreaQ782IW8+bNe+D+Fi1a0KJFi0zbAwMDAfDy8so0s33JkiUzbCtZsiQbN27M0GbNmjWZYvbo0cP0c5s2bR5apF+4cIHixYtTt27dB7YTERERERHze16L8SelYl7k/zt69CgnT55k8uTJdO7cmbx5NXBFRERERERyJlUrIv/fgQMH+PLLL6lWrRrvv//+s05HRERERETkvnRnXuT/69SpE506dXrWaYiIiIiIyD9kd06tfxsV8yIiIiIiIpJjqZbPmobZi4iIiIiIiOQyedI0ZkFERERERERyqClrdz5xjE+bP38rVWmYvYiZXbqaaGg8Bztbki5fNjQmgI29veFxbeztSUxKMjSmrY0NsRcSDI0JULKoo1neqyvnzhgas3BxF45EGRuzcmkXYuKMv6auzo6cOnfR0JjuxZ1ITIg3NKatYxESTv5taEwAxzLlzfL5N0fMhCvGfvYBHAvbMm7FNkNjft66PgdPxhoas1qZkmb7/P8Vfc7QmBXdihsaT0REcjcV8yIiIiIiIpJjpaHB5FnRM/MiIiIiIiIiuYzuzIuIiIiIiEiOpVnesqY780+Zr68vAQEBWe4LCAjA19fX9LuHhwdhYWHZinv58mWWLl36RLnFxsbi4eHB7t27nyhOunv7k9V+Dw+PLF+9evXKVox/SklJITQ01PR7cHAw3t7eT9QHERERERF5tlLT0p749TzSnfkcbMeOHdjY2GSr7bhx44iNjeXdd9997PMVL16cHTt2YGdn99gxHpWnpyfBwcGZtr/wwgsADB48mDt37mQr1tq1axkzZgxdunQBwN/fn06dOhmWq4iIiIiISE6hYj4Hc3JyynZbI1YYzJcv3yOd0wiWlpYPPGd2v8yAzNegUKFCFCpU6LFzExERERERyak0zD4H++cw+4SEBD799FO8vLyoWrUq7du3Z8+ePcDdoegrVqxgz549eHh4AHDnzh1CQ0Np2rQpVapUoWnTpixZssQUe/fu3Xh4eDBz5ky8vLxo3bo1p0+fzjTMft68eTRt2pSqVavi4+PDqlWrTPv279/PBx98QI0aNahcuTLNmzdn7dq1hl6De4fZx8TE8PHHH1OjRg28vLzo27cv8fHxhIWFERgYaLpuu3fvzjTM/ty5cwwYMIC6detSvXp1unbtyvHjxzOca+DAgXz99dfUqVOHatWq0atXLy5eNHZpLRERERERyb60tLQnfj2PVMznEsOGDePmzZvMnz+fNWvWULp0aXr16sX169cZPHgwzZo1w9PTkx07dgAwduxYpk6dSu/evVmzZg1+fn6MGDGCefPmZYj766+/snjxYoKCgsibN+PHISQkhAkTJtC1a1fWrl1Lp06dCAwMZOfOncTFxeHv78+LL75IWFgYq1atokqVKgQGBhIfb+wa0OmSkpLo2LEj169fJzQ0lNDQUM6cOcMnn3yCj48PgwYNAu4+nuDp6Znh2GvXrtGhQwfi4uKYNm0aP/30E1ZWVnTu3JmzZ8+a2m3YsIErV64wf/58vvvuO/bv38+kSZPM0h8REREREXk4PTOfNQ2zfwbWrFnDzz//nGl7cnIyL7/8cpbHnD59mgoVKlCqVCleeOEFBg8eTIsWLciXLx9WVlYUKFDANGT92rVrLFq0iICAAFq0aAGAu7s7MTExTJ8+nc6dO5vi+vv74+7uDtydAO+fQkND8fPzo127dgB06tSJmzdvcufOHZKTk+nduzddu3Y1fQnQo0cPwsLCOHXqFEWKFMnWtdi3b1+mwrto0aJZXp/169eTlJTEpEmTKFy4MACjR49m1apV5MmTxzQkP6th+6tXr+by5cuEhYXh4OAAwIQJE2jcuDELFixg4MCBAFhbWzNixAgsLS0pW7YsLVu2ZNu2bdnqi4iIiIiIGO85rcWfmIr5Z8Db25sBAwZk2j5hwgSuXLmS5TG9e/dm4MCBbN68mZo1a/Laa6/h4+Njmijun06ePElKSgo1atTIsL1mzZr8+OOPJCQkmLalF/L3unTpEhcuXKBatWoZtnft2tX0c9u2bZk/fz4nTpzg1KlT/PXXXwDZnrAOoHLlykyYMCHDtnz58mXZ9vjx47i7u5sKeYDy5ctneS3vFRERgbu7u6mQh7uT7FWtWjXDUHs3NzcsLS1Nv9vY2JCSkpLd7oiIiIiIiDwVKuafgUKFCuHm5pbl9vsV802aNGH79u1s376d//znP8yaNYtvv/2WJUuWUL58+Qxt058JyZMnT4btqampAFhY/O9tz+rLAID8+fNnGSNdZGQkHTp04KWXXqJu3bo0atQIe3v7R55Nv0CBAllei6xYWFjcN5+HSUtLy/LYO3fuZLge6f0WERERERHJyfTMfC6QnJzMmDFjiImJwcfHh1GjRrF582by5s3Lr7/+CmQsusuUKYOFhQX79u3LEGffvn04OTlla+k5a2trihYtyuHDhzNs//TTTxk1ahSLFi3C0dGR0NBQunfvTv369U3Pyptrgoly5cpx6tQpkpKSTNuOHTuGl5cXZ86ceWChX6FCBaKiojKMSrh16xZHjhyhXLlyZslXRERERESenCbAy5qK+Vwgf/78HDx4kK+++ooDBw4QGxtLWFgY//3vf03Pm1tZWXHhwgViYmKwsbGhXbt2TJkyhTVr1hAdHc2CBQtYuHAh/v7+2b67/eGHHzJnzhxWrlzJ6dOnWbBgAVu2bKFx48YUK1aM8+fPs23bNs6cOcOmTZsYNmwYcPfLB3No0aIFdnZ2DBw4kPDwcI4cOcKwYcOoUKECLi4uWFlZAXDkyBFu3ryZ6VhbW1s+++wzDh06RHh4OAMHDuT69eu89957ZslXRERERETEXDTMPpf49ttvGTNmDB999BFJSUmUKVOGiRMnUrNmTQBatWrF5s2bad68OZs3b2bw4MHY29szceJE4uPjcXNzY8iQIabJ7LKjc+fO3Lp1iylTpnDx4kXc3d2ZNGkStWvX5uWXX+bkyZN8/vnnJCcn4+7uTr9+/ZgyZQqHDh3i9ddfN/waFCxYkJCQEMaOHUuHDh3Inz8/3t7efP755wDUrl2batWq0b59e8aPH5/hWFtbW+bPn8/XX39Nly5dAKhRowaLFi3C1dXV8FxFRERERMQYz+ts9E8qT9rzOuZAJIe4dDXR0HgOdrYkXb5saEwAG3t7w+Pa2NuT+I/HIoxga2ND7IWEhzd8RCWLOprlvbpy7oyhMQsXd+FIlLExK5d2ISbO+Gvq6uzIqXMXDY3pXtyJxARjl7+0dSxCwsm/DY0J4FimvFk+/+aImXDF2M8+gGNhW8atMHY1kM9b1+fgydiHN3wE1cqUNNvn/6/oc4bGrOhW3NB4IiK5RdCyrU8cY9A7DQ3IJGfRMHsRERERERGRXEbFvIiIiIiIiEguo2fmRUREREREJMfSk+FZUzEvIiIiIiIiOZYmwMuaJsATERERERGRHGvEki1PHGNIu0YGZJKz6M68iJmdWDjD0HjlOvYwfNZ1uDvz+vkEY2ezL+Zontnsjb6mcPe6bj1k7IzmDauWZ+XvRwyN2ap2Zc7t3WFozOKvvEbcgd2GxgRwru5lls//6U0rDY1Z6o1WvDkq1NCYABu/7JJrZrM31/s/d+t+Q2P6NazB8Zjzhsb0cC3GuXjjVwgpXsSeE2fiDI1ZzsWZ8NPGzpD/YinNkC8ikltpAjwRERERERGRXEbFvIiIiIiIiORYqWlpT/wyp1u3bjF8+HDq1KmDp6cnn376KQkJCdk+ftq0aXh4eDzyeVXMi4iIiIiISI6Vlpb2xC9zGjZsGDt37iQ4OJg5c+YQExNDnz59snXsoUOH+O677x7rvCrm/4W8vb0JDg7OtD0oKIgXX3yRJUuWPLBduuDgYLy9vR87D19fXzw8PLJ8jR492tQmICAgW/GuX7/OggULTL8HBATg6+v72PmJiIiIiIg8SFxcHCtXruTLL7+kZs2aVK1alW+++Ya9e/dy4MCBBx57/fp1Bg4cSM2aNR/r3JoATwAYM2YM8+fPZ+zYsbRq1QqAZcuW8cILL5j1vM2aNWPw4MGZthcsWBC4+4VBvnz5shVr9uzZhIWF0alTJwAGDx7MnTt3jEtWRERERESeOiNurDdq9ODZ7LdsebwZ8/fvvzvZq5eXl2lb6dKlcXZ2Zu/evVSvXv2+x44ePZoKFSrQsGFDfv/990c+t4p5YezYscyfP5/x48fz1ltvmbY7ODiY/dwFChTAycnpvvsLFy6c7Vj3Dp+xsbF53LREREREREQeKi4uDnt7+0w3QYsWLcq5c/dfgWTz5s1s27aNNWvWsHXr1sc6t4r5f7n0Qn7y5Mk0adIkwz5vb29at27NJ598AsDixYuZNWsWcXFxvPbaa5QoUSJD+23btvHtt98SGRmJlZUV9evXJzAwEDs7u8fOz9fXFxcXF8aOHQvAkSNHmDhxIgcOHKBgwYI0btyYgIAAQkJCTM+aeHh4sGXLFr777jvOnDnDvHnzAIiMjGT8+PH8+eef3L59m7p16xIQEGDqh6+vL5UrV+bq1ats2rSJ1NRUmjRpwpAhQyhUqNBj90FERERERB6fERPYPe6d99jY2Afe1e/Tpw/58+fPtP2FF17g1q1bWR4TFxfHV199xbhx47C3t3+svEDPzP+rff311/z444/4+/tnKuTvtW7dOkaMGEGXLl1YtWoV1atXz/B8+qVLl+jduzdt27Zl/fr1fPfdd+zdu5dx48YZlm9sbCy+vr44ODiwePFivvvuO3bv3s2QIUPw9/fH39+fYsWKsWPHDooXz7hu7pkzZ3jvvffInz8/c+bM4ccffyQhIYHOnTtz7do1U7t58+ZRpEgRli5dyqhRo1i/fj2hoaGG9UFERERERHIPZ2dn1q9ff99XgQIFSE5OznTcrVu3TI8O/1NaWhoBAQE0a9aM119//Yly0535f6klS5aQmJjIyy+/zIIFC3j33XdxdXW9b/u5c+fi4+Njeh79ww8/5MCBA4SHhwN3v11KTk6mRIkSuLi44OLiwvTp0x/6zPqaNWv4+eefM2zz9PRk9uzZWeZsZ2fH2LFjsbS0BGDUqFHs2bOHQoUKYWVlRb58+bIctr9w4UKsrKyYMGGC6ZuzKVOm4O3tzerVq+nYsSMAZcuWpV+/fsDdZ13WrVvHH3/88cA+iIiIiIiI+Zh7NvoHsbS0pGzZsvfdf/z4ca5cuUJycnKGO/QXLlygWLFimdqfPXuW//znP/zxxx+sXLkSgNu3bwN366AePXrQs2fPbOWmYv5f6tq1a/zwww+8+OKLtGjRgn79+rFw4UJTkXyviIiIDM/Tw90PW3oxX7FiRZo3b07Pnj0pXrw4r776Kg0aNHjobPfe3t4MGDAgw7YCBQpk2fb48eNUqlQpQ46vvPIKr7zyykP7GxERQeXKlTP8gTk6OlK6dGmOHz9u2nbvH6qNjQ2JiYkPjS8iIiIiIubxDGv5h6pRowapqans37+fOnXqAHDy5Eni4uKynKXe2dmZTZs2Zdi2adMmJkyYwMqVKx/pEWUNs/+X8vPzw8vLy3Sn+/Dhw0yePPmBx9z7jdi9hf/EiRPZsGEDXbp0IT4+nn79+uHv7//AmIUKFcLNzS3Dy9nZOcu2FhYW5MmT5+Gdu0/uWR17586dDP3I6nkXERERERGRrDg7O/PWW2/x5Zdfsnv3bg4dOkT//v2pVauWaSb75ORkLl68SHJyMhYWFpnqH0dHRwDc3NweaQJwFfP/UhYW/xuU8eqrr+Lr60tISAg7d+7Msn3FihVNyy6kO3z4sOnnAwcOEBQURJkyZejSpQs//PADQUFB7N69m4SEBENyLleuHMeOHcswdH/z5s28/vrr3Lhx44GFfoUKFTh06FCG51ni4+OJjo5+4LAZERERERGRBxk5ciR16tShd+/edO3alTJlyjBlyhTT/j///JPXXnuNP//809DzqpgXAAYMGEC5cuX4/PPPiY+Pz7T/ww8/ZPPmzcyaNYtTp04xb968DM+6W1tbs3DhQsaPH090dDTHjx9n3bp1uLu7P9EMjf/UsWNHLl++zNChQ4mMjGTfvn1MmDCBunXrUrBgQaysrLh69SpRUVGkpKRkOLZDhw5cu3aNAQMGEB4ezqFDh+jTpw/29vaZHh8QEREREZGcIzUt7Ylf5mRlZcWoUaPYu3cve/fuZeLEiRlqIC8vL44fP55hLfp/atOmTYZHf7NLxbwAd5dOGD9+PFevXuWLL77INKS+QYMGTJw4keXLl9OiRQs2bdqUYQh9uXLlCA4O5vfff6dVq1Z07NgRCwsLZs6cSd68xnzMnJ2dmT17NlFRUbRu3ZrPPvuM+vXrM3ToUADeeOMNnJycePvttzl27FiGY11dXZk3bx6JiYm89957dO3aFScnJxYtWoStra0h+YmIiIiIiPHS0tKe+PU80gR4/0K//PJLltsrVqzIkSNH7tvOx8cHHx+fDNvSZ34HaNiwIQ0bNsx2Hunrvz9KG09PzwxL4v1TyZIl2bhxo+n3atWqZdhfuXLlBy4zl1U+6evbi4iIiIiI5CQq5kVERERERCTHSn0+b6w/MQ2zFxEREREREclldGdeREREREREcqw0dGs+K7ozLyIiIiIiIpLL6M68iIiIiIiI5FjP62z0TypPmq6MiIiIiIiI5FB9f1z3xDEmffCWAZnkLLozL2JmiUlJhsaztbExPKa54por5qWriYbGBHCwsyXhirFxHQvbcvVCnKEx7Yo6k5RobJ42trb/+s+U0e893H3/TyycYWjMch175Iprmh73wqUrhsYs6lCYy1eNzdXeznz9N0euuSFmelwRETEvPTMvIiIiIiIiksvozryIiIiIiIjkWHoyPGu58s68t7c3Hh4e/Pjjj1nuHzJkCB4eHgQHB2c7pq+vLwEBAabf9+/fz759+wCIjY3Fw8OD3bt3ZyvW7t278fDwIDY2Ntvnz8rt27eZM2cObdq0wdPTEy8vLz744AN27dr1RHGz6/LlyyxdutT0+73XyEjp7+n9Xr6+vmY5r4iIiIiI5GypaWlP/Hoe5do785aWlmzcuJEPPvggw/bbt2+zadMm8uTJ80TxO3bsyJgxY6hZsybFixdnx44d2NnZPVHMR5GcnMwHH3zAuXPn+OSTT/D09OTmzZssX74cf39/xowZQ6tWrcyaw7hx44iNjeXdd98FIDg4mHz58pnlXMuWLePOnTsA/Pnnn3zyyScsXbqU4sWLA3ffbxERERER+ffRnfms5dpivk6dOmzfvp1z586ZCj6A33//HSsrKwoWLGjYufLly4eTk5Nh8bJjypQphIeHs27dOooVK2baPnjwYK5fv05QUBBNmjShUKFCZsvh3j+awoULm+1cDg4Opp/TvzRxcHB46tddREREREQkN8iVw+wBqlatSokSJdi4cWOG7evWraNZs2YZ7syHhYXh4eGRod2DhsKntw0MDCQgICDTMHtfX1+CgoL4/PPPqV69Oq+//jo//PDDfb8xSktLY+bMmTRq1Ihq1arRsmVLVq9efd++paSksHTpUt55550MhXy6Pn36MGvWLAoUKGDKd9KkSTRs2JC6dety8uRJkpOTGT9+PPXq1cPT05N27dqxY8eODHGWL19Oq1atqFq1KtWrV8fX15ejR48CEBAQwIoVK9izZ4/peqQPs//vf/+Lp6cnCxcuzBBv2rRpNGjQgNTU1Efu88OEhobi6enJjRs3TNtSU1N5/fXXmTt3run93LJlC2+88QbVq1enS5cuREZGmtpnJ6eQkBAaN25M5cqV8fb25vvvv9c3gSIiIiIikuPk2mIeoFmzZhmK+eTkZLZs2cJbbz3ZGoLpRe+gQYMYPHhwlm0WLlxIwYIFWb58OX379uX7779n5syZWbadNGkSCxcu5Msvv2TNmjX4+fkxbNgwFixYkGX7mJgYrly5QvXq1bPcX7RoUapWrZphyPvixYuZMmUK33//PWXKlCEwMJDt27czfvx4VqxYQbNmzejZsye//vorAJs3b2bo0KF06dKFDRs2MGfOHG7evGnq7+DBg2nWrBmenp6ZvgQoVKgQTZs2Zc2aNRm2r1mzhpYtW5I3b95H7vPDvP3226SkpLBp0ybTtv/85z9cunSJ5s2bm7aNHj2awYMHs3jxYiwsLPDz8yPp/y859LCcfvnlF6ZPn87w4cPZtGkTAwYMYNq0aU/0JYSIiIiIiDyZtLQnfz2Pcu0we7hbzIeEhJiG2u/cuRN7e3teeumlJ4qbPrTbxsYGGxsbrl69mqlNmTJlGDZsGHny5KFs2bJERkYyd+5cunfvnqHd9evXCQ0NZdy4cTRs2BCAUqVKcebMGUJCQujUqVOm2Onne5Rn9Fu2bEmVKlUAiI6OZu3atSxbtsy07YMPPiA8PJyQkBAaNGhA4cKFGTVqlOm5excXF959912GDh1q6nuBAgWwtLTMcqh7mzZt8PPzIzY2lpIlS3LkyBEiIyOZOnXqY/X5YRwcHPD29mb16tW0bNkSgBUrVuDt7Z1hiH5AQAD169cHYMKECTRo0IB169bx9ttvPzSn06dP88ILL1CyZElKlChBiRIlKFq0KCVKlHjkfEVERERExBjP6wR2TypXF/OVK1fG1dXVNBHe+vXrM9ylNadatWplGMpfvXp1Zs6cyeXLlzO0O3HiBLdu3eKLL74gMDDQtP327dskJydz8+ZN03D5dOnF6ZUrV7Kdj5ubm+nnY8eOAeDn55ehTUpKCra2tgC88sorODg4MHXqVKKjo4mKiuKvv/4iNTU1W+d75ZVXKFmyJGvXrqVnz56sWrUKT09P3N3dOXTo0CP3OTvatm1Lz549iYuLo1ChQvzf//0f3377bYY2tWrVMv1cuHBh3N3diYiIyNb78Pbbb7N8+XLeeOMNPDw8qFu3Lk2aNFExLyIiIiIiOU6uLubhf0PtO3bsyJYtWzIspXavtLQ0UwF++/btJzqvhUXGS5f+XPW9s72nb588eTJlypTJFCd//vyZtrm6ulKkSBH+/PNPfHx8Mu0/deoUI0aM4IsvvjA9z/7P4jj9nAsWLMg0QV7evHefrFi3bh2ff/45zZs3p2rVqrzzzjtEREQwYsSIB3f8/8uTJw+tWrVizZo1dO/enfXr19OnT5/H7nN2vPbaazg5ObFu3ToKFy6MjY0N9erVy9Dm3vclNTWVvHnzZiunAgUKsGrVKv7880927tzJjh07mD17Np988gm9e/d+rJxFREREROTJaA6rrOXqZ+bhbjF/8OBBli1bhqurK2XLls3UJn1Zs/Rnp+HuUPQncfjw4Qy///HHH5QsWTLT0PgyZcpgYWHB2bNncXNzM722bdtGSEiIqbj+p7x58/LOO+8QFhZGXFxcpv2zZs3iwIEDuLi4ZJlb+fLlAbhw4UKGc4aFhbF8+XIApk+fzjvvvMPXX39Np06deOWVV4iJiQH+98fysOX9WrduTWRkJAsXLiQpKYlmzZo9dp+zI1++fLRq1YpNmzaxadMmWrZsmenLk3++L5cuXSI6OppKlSplK6dVq1axaNEiatSowaeffsqSJUt49913Wb9+/WPlKyIiIiIiYi65vpivWLEibm5ufPPNN/ed+K569erkzZuXyZMnExMTw6+//srs2bMfGNfKyorIyMhMw+bT7du3jylTphAVFcWyZctYsGAB3bp1y9TOxsaG9u3bM3nyZFauXElMTAwrVqxg/PjxFClS5L7n79mzJ25ubrRv356VK1dy+vRpDh8+zODBg1m+fDkjR47E2to6y2PLly9Pw4YNGTp0KFu2bCEmJoaQkBBmzJiBq6srAMWLF+ePP/7g6NGjnD59mtDQUObPnw/cnUgw/RpcuHDBVOTfy8XFBS8vLyZNmkSTJk2wsbF5oj5nR9u2bTl48CD/+c9/aNOmTab9w4cPZ+/evYSHhzNgwACcnJx48803s5XTrVu3+Prrr1m5ciWxsbHs27ePPXv24Onp+UQ5i4iIiIiIGC3XD7OHu3fnp02bluWQdLg7bH3EiBFMnz6dJUuWUKlSJQYNGsRHH31035j+/v7MmjWLkydPZjmjfaNGjfj7779p2bIlRYsWJSAggA4dOmQZKzAwEAcHB6ZMmcKFCxcoVqwYvXv35sMPP7zv+QsWLMj8+fOZPXs2M2fO5OzZs7zwwgtUqlSJOXPmZHg2PCuTJk1i0qRJDB06lKtXr+Lq6srIkSNp27YtAF999RVDhgyhc+fO5M+fnxdffJFx48bRt29fDh48SK1atWjVqhWbN2+mefPmbN68OcvztGnTht9//900kd6T9Dk73NzcqF69OqmpqVmOwnj33XcZMGAAiYmJ1K5dm7lz51KwYMFs5dSuXTuuXr3K1KlTOXfuHHZ2djRt2pQBAwY8Uc4iIiIiIvL4NMo+a3nS9ADCI/P19cXFxYWxY8c+61T+ddLS0njjjTf48MMPeffdd03bd+/ejZ+fH1u2bKFkyZLPMMPMEv/xeIcRbG1sDI9prrjminnpaqKhMQEc7GxJuGJsXMfCtly9kPlRmSdhV9SZpERj87Sxtf3Xf6aMfu/h7vt/YuEMQ2OW69gjV1zT9LgXLl0xNGZRh8JcvmpsrvZ25uu/OXLNDTHT44qIGOXDaSueOMYPH7U2IJOc5bm4My/Pv5SUFH755Rd+//13rl27dt9HKkRERERE5PmShu4/Z0XFvOQKlpaWjBo1CoDx48djZWX1jDMSERERERF5dlTMP4Z58+Y96xT+lbZv337ffV5eXhw/fvwpZiMiIiIiIvLsqJgXERERERGRHCtVo+yzpGJeREREREREcizN2Z41zWYvIiIiIiIiOZb/98ufOMbsj9sakEnOojvzImZ2Lv6yofGKF7E329Js5xOMzbWYo71ZltEyOk+4m2v88aOGxiziUYnYCwmGxixZ1JH48MOGxizyYhXiDu41NCaAc7VXuHj5qqExneztzBLT6PcJ7r5X5vj8Xz4dZWhM+1KlzbY039QNuwyN2atZHY6eOmtozEruJcz2/h+POW9oTA/XYhyJOmNozMqlXfg71tglNAHKl3Q2yzJ6IvLvpPvPWcv7rBMQERERERERkUejO/MiIiIiIiKSY2kCvKzpzryIiIiIiIhILqNiPodas2YN7733Hp6ennh6etK2bVt++umnbB+/e/duPDw8iI2NNWOWTyY9x/u9EhMTH7kf+/fvZ9++fQDExsbi4eHB7t27zdkNERERERExo7S0tCd+PY80zD4HWrZsGaNGjWLQoEG88sorpKWlsWvXLkaPHk18fDy9e/d+aAxPT0927NiBg4PDU8j4ySxdupTixYtn2m5jY/PI/ejYsSNjxoyhZs2aFC9enB07dmBnZ2d0yiIiIiIi8pQ8r8X4k1IxnwMtXLiQd955h3bt2pm2lSlThvPnzzN37txsFfP58+fHycnJnGkaxsHB4b65Pkk/8uXLl2uugYiIiIiIyKPQMPscKG/evPzxxx9cvZpx+aXu3buzePFi0++3b98mODgYb29vqlWrRps2bfjtt9+AzMPsk5OTGT9+PPXq1cPT05N27dqxY8cOU6ywsDC8vb1ZsWIFTZo0oXLlyrRt25Y///wzW+cDiIyMpHv37nh6evLaa6/Rv39/Ll68+ETX4t5+PCgHDw8PAAIDAwkICMg0zP7OnTuEhobStGlTqlSpQtOmTVmyZEmmc23bto3mzZtTuXJl3nrrLbZu3fpEfRARERERETGaivkcqHv37vz111+8/vrrfPjhh/zwww8cOnQIGxsbSpcubWoXFBTEggULGDBgAGvWrKF+/fr06tWLEydOZIoZGBjI9u3bGT9+PCtWrKBZs2b07NmTX3/91dTmwoUL/PTTT4wfP57FixeTN29evvjiC9OwlgedLy4ujo4dO+Lq6sqyZcuYPn06165do3379ly/ft2wa/OgHNK/nBg0aBCDBw/OdOzYsWOZOnUqvXv3Zs2aNfj5+TFixAjmzZuXod348eMZPHgwYWFhuLq6MmDAAP773/8a1gcREREREcm+1LQnfz2PNMw+B2ratCmLFy9m3rx57Nixg23btgHg7u5OUFAQNWrU4Nq1ayxZsoQvv/wSHx8fAPr06UNqamqmwjM6Opq1a9eybNkyqlSpAsAHH3xAeHg4ISEhNGjQAICUlBSGDRtGxYoVAejRowcff/wxFy9exMrK6oHnW7t2LUWLFmXIkCGm806ePJnatWuzceNG2rRpc9/+Nm/enDx58mTYNn36dLy8vDJse1ify5UrB9x91t7GxibDyIZr166xaNEiAgICaNGihel6xsTEMH36dDp37mxq+9lnn1GnTh3Tzy1btiQiIgJPT8/79kFERERERMxDz8xnTcV8DlW1alXGjx9PWloaERERbNu2jblz59K9e3c2b97M2bNnSUlJoXr16hmO69u3L0CGGdyPHTsGgJ+fX4a2KSkp2NraZthWtmxZ0882NjamdlFRUQ883/fff09kZGSmgvfWrVtERkY+sK8//PADzs7OGbbd+zvw0Bwe5OTJk6SkpFCjRo0M22vWrMmPP/5IQkKCaVuZMmVMP1tbWwN3r4GIiIiIiDx9qSrms6RiPoc5f/48M2fO5MMPP8TZ2Zk8efKYlmpr1KgRPj4+7N27F3d392zHTP8ma8GCBRQqVCjDvrx5Mz5pkT9//iyPt7S0fOA5UlNTqV27NkOHDs20L/1LgfspUaIEJUuWfGAb4KE5PEj6Nbh3BEBqaioAFhb/+1O43zUQERERERHJKfTMfA6TP39+Fi9ezOrVqzPtS79LXKRIEdzc3LC0tOTw4cMZ2rzzzjvMmjUrw7by5csDd5+Jd3NzM73CwsJYvnx5tvJ62PnKly9PZGQkxYsXN8W3s7MjKCiIiIiIbPf/SXJ4kDJlymBhYWFagz7dvn37cHJy0vJ1IiIiIiKSq+jOfA7j4OBAt27dmDx5MteuXePNN9/E2tqaEydOMHXqVLy8vKhZsyYAnTt35ttvv8XBwYHy5cuzfPlyTpw4QcOGDYmPjzfFLF++PA0bNmTo0KEMGTKEChUqsGnTJmbMmMHo0aOzlVfBggUfeL70LyH69evHxx9/TJ48eRg/fjzHjh0zfZnwpB6WA4CVlRWRkZFcvnw5w7E2Nja0a9eOKVOmYGdnR9WqVdmxYwcLFy6kX79+me7Yi4iIiIhIzqAxsllTMZ8DffbZZ7i7u7NkyRIWLFjAzZs3KV68OD4+PvTo0cPUrl+/flhYWDBs2DASExPx8PDghx9+oGzZshmKeYBJkyYxadIkhg4dytWrV3F1dWXkyJG0bds223k96HwA8+fPZ+LEiXTs2JF8+fJRvXp15syZg6OjozEXJhs5+Pv7M2vWLE6ePJlpRvvBgwdjb2/PxIkTiY+Px83NjSFDhtCuXTvD8hMREREREXka8qTpYWARszoXf/nhjR5B8SL2XLqaaGhMAAc7W84nGJtrMUd7EpOSDI1pa2NjeJ5wN9f440cNjVnEoxKxFxIe3vARlCzqSHz44Yc3fARFXqxC3MG9hsYEcK72ChcvX314w0fgZG9nlphGv09w970yx+f/8ukoQ2PalypNwhXj/01xLGzL1A27DI3Zq1kdjp46a2jMSu4lzPb+H485b2hMD9diHIk6Y2jMyqVd+Ds2ztCYAOVLOnP5qrGff3u7B8/BIyLPr3YTFz1xjCX9OxiQSc6iO/MiIiIiIiKSY2k2+6xpAjwRERERERGRXEbFvIiIiIiIiEguo2H2IiIiIiIikmNplH3WNAGeiIiIiIiI5Fhtxi144hhhn3cyIJOcRXfmRczM6FmiHQvbGj5DNphnlnhzzWZvrpm3z+83dubtYjXq8Ff0OUNjVnQrTsyWtYbGdG3UnIt/HTI0JoBTxapmmc3aHJ+pmDjjZzN3dTbPbPZmmSHf4PcJ7r5XO4+eNDRm3UplzLJChNErJMDdVRLOXrxkaMwSTg6Gf1ZdnR0NzxPu5mqOz6o53n8Ryfk0AV7W9My8iIiIiIiISC6jYl5EREREREQkl9EwexEREREREcmxNM1b1nRn/l9qzZo1vPfee3h6euLp6Unbtm356aefnnVa9xUQEICvr2+W+4KDg/H29n7KGYmIiIiIyNOQmvbkr+eR7sz/Cy1btoxRo0YxaNAgXnnlFdLS0ti1axejR48mPj6e3r17P+sURURERERE5AFUzP8LLVy4kHfeeYd27dqZtpUpU4bz588zd+5cFfMiIiIiIpJjaJh91jTM/l8ob968/PHHH1y9mnEpnu7du7N48WLT797e3gQHB2do4+vrS0BAAAC7d+/Gw8ODLVu28MYbb1C9enW6dOlCZGSkqX1aWhozZ86kUaNGVKtWjZYtW7J69WrT/vQYM2fOxMvLi9atW3Pnzp0n6l9ycjLjx4+nXr16eHp60q5dO3bs2GHaHxYWhre3N6NHj6ZmzZr07NkTgMjISHr27ImXlxc1atTg008/5ezZs0+Ui4iIiIiIiDnozvy/UPfu3fnss894/fXX8fLyombNmtSuXZsqVapga2v7yPFGjx7N0KFDKVasGOPHj8fPz4+NGzdiY2PDpEmTWLNmDUOGDKFs2bLs3buXYcOGkZSURKdOnUwxfv31VxYvXsyNGzfIly/fE/UvMDCQv//+m/Hjx1OsWDG2bt1Kz549+e6772jQoAEAZ86cIS4ujhUrVnDz5k3OnDnDe++9x6uvvsqcOXNITk7m66+/pnPnzqxevRpra+snyklERERERB6P7sxnTcX8v1DTpk1ZvHgx8+bNY8eOHWzbtg0Ad3d3goKCqFGjxiPFCwgIoH79+gBMmDCBBg0asG7dOt5++21CQ0MZN24cDRs2BKBUqVKcOXOGkJCQDMW8v78/7u7uDzzPvn378PT0zLQ9JSWFokWLAhAdHc3atWtZtmwZVapUAeCDDz4gPDyckJAQUzEP0KtXL1xdXQEYP348VlZWTJgwgfz58wMwZcoUvL29Wb16NR07dnykayIiIiIiImJOKub/papWrcr48eNJS0sjIiKCbdu2MXfuXLp3787mzZtxdHTMdqxatWqZfi5cuDDu7u5ERERw4sQJbt26xRdffEFgYKCpze3bt0lOTubmzZumbQ8r5AEqV67MhAkTMm2fN28ev/zyCwDHjh0DwM/PL0OblJSUTKMO/nnOiIgIKleubCrkARwdHSldujTHjx9/aG4iIiIiImIez+ts9E9Kxfy/zPnz55k5cyYffvghzs7O5MmTBw8PDzw8PGjUqBE+Pj7s3buXN998E8g8pCUlJSVTTAuLjB+j1NRU8ubNazp28uTJlClTJtNx/yycX3jhhYfmXqBAAdzc3DJtt7OzM/2cfs4FCxZQqFChDO3y5s04RUSBAgUyHJcnT55Mse/cuYOlpeVDcxMREREREfNIQ9V8VjQB3r9M/vz5Wbx4cYZJ6NKlPxdepEgRACwtLUlKSjLtT01NJTY2NtNxhw8fNv186dIloqOjqVSpEmXKlMHCwoKzZ8/i5uZmem3bto2QkJBMxbURypcvD8CFCxcynDMsLIzly5ff97gKFSpw6NAhkpOTTdvi4+OJjo6mbNmyhucpIiIiIiLyJFTM/8s4ODjQrVs3Jk+ezKRJk/jrr7+IiYlh69at9O7d2zQhHsDLL7/M+vXr2bt3L1FRUaaJ6+41fPhw9u7dS3h4OAMGDMDJyYk333wTGxsb2rdvz+TJk1m5ciUxMTGsWLGC8ePHm74wMFr58uVp2LAhQ4cOZcuWLcTExBASEsKMGTNMz8dnpUOHDly7do0BAwYQHh7OoUOH6NOnD/b29rz11ltmyVVERERERHK/W7duMXz4cOrUqYOnpyeffvopCQkJDzzm2rVrDB06lNq1a1OjRg169uxJTEzMI51Xw+z/hT777DPc3d1ZsmQJCxYs4ObNmxQvXhwfHx969Ohhate3b1+uXr1K9+7dKViwIO+++y4+Pj6Zht6/++67DBgwgMTERGrXrs3cuXMpWLAgcHdmeQcHB6ZMmcKFCxcoVqwYvXv35sMPPzRb/yZNmsSkSZMYOnQoV69exdXVlZEjR9K2bdv7HuPq6sq8efOYMGEC7733Hvnz56du3bqMHz/+sWb4FxERERERY+T02eyHDRvG/v37CQ4OJn/+/AwdOpQ+ffowf/78+x7zySefcObMGYKDg7GxsWH48OF89NFHrF69OtsjmPOk5fQrIznW7t278fPzY8uWLZQsWfJZp5NjJVxJNDSeY2FbErMYIfGkbG1sOJ9w2dCYxRztDc/V1sbG8GsKd6/r+f27DI1ZrEYd/oo+Z2jMim7Fidmy1tCYro2ac/GvQ4bGBHCqWJXLV419/+3tbMzymYqJe/C354/D1dnRLLmaI6bR7xPcfa92Hj1paMy6lcoQe8HY96pkUUcuXr5qaEwAJ3s7zl68ZGjMEk4Ohn9WXZ0dDc8T7uZqjs+qOd5/Ecn5moz48YljbB7ygQGZZBYXF0eDBg2YMWMGr7/+OgBRUVG8+eabLF68mOrVq2c6Zvfu3bz//vusWrUKDw8PAE6cOEH37t0JCQnJcr6xrGiYvYiIiIiIiMhj2L9/PwBeXl6mbaVLl8bZ2Zm9e/dmecz27dupUKGCqZAHKFeuHFu3bs12IQ8aZi8iIiIiIiI5mBGDyRs1avTA/Vu2bHmsuHFxcdjb22danato0aKcO5f1CM1Tp07h5ubGwoULWbBgAYmJidSoUYPAwECcnZ2zfW4V8/LYvLy8tAa7iIiIiIg8t2JjYx/4RUCfPn0yLLmd7oUXXuDWrVtZHnPt2jWOHj3K5cuXGT58OAATJkzAz8+P1atXZ2vZblAxLyIiIiIiIjlYqgF35n95zDvvzs7OrF+//r77t23blmF563S3bt0yTQp+L0tLS27dusX333+PnZ0dAN999x316tXjl19+oVmzZtnKTcW8iIiIiIiISBYsLS0pW7bsffcfP36cK1eukJycnOEOffpKXlkpVqwYzs7OpkIeoEiRIhQuXJjY2Nhs56ZiXsTMHAsbv7SdrY2N4THh7uzzRjNHrua4pnB39nmjVXQrbnhM10bNDY/pVLGq4THh7ozmRjPHZ8rV2TwzWpsjV3PENMf7BHdnnzeaOWYfd7K3e3ijx1DCycHwmOb4rJojTzDPZ1Wzz4v8O+Xk9ddq1KhBamoq+/fvp06du/8vefLkSeLi4qhZs2aWx9SsWZOwsDAuXLhA0aJFgbvF/+XLl3Fzc8v2ubU0nYiZ5aZlpHLL0nSJCfGGxgSwdSzC1Qtxhsa0K+pslqUJr5yNMTRm4RKuXI45ZWhMAHtX91xzTXPT0lz/9n9TLl019v13sLM129J0SZeN7b+NvT2J8RcMjWlbpKjhMU1xzfBZTbp6xdCYNnaFuXL+rKExCxcrYWg8EYGGQ0OeOMbW4V0NyCRr/fv358CBAwQFBVGwYEGGDh2KtbU18+bNAyA5OZmrV69iZ2dH/vz5SU5OpnXr1tja2jJo0CDy5ctHUFAQCQkJrFq1Kstn8LOiO/MiIiIiIiKSY5mzEDfCyJEjCQoKonfv3gC8/vrrfPnll6b9f/75J35+fsydOxcvLy/y589PaGgoY8eOpUuXLqSlpVG3bl0mTpyY7UIeVMyLiIiIiIiIPDYrKytGjRrFqFGjstyf1SpgTk5OTJw48YnOm/eJjhaz8vDwICws7Fmn8UBnz55l3bp1zzqNh0pJSSE0NPRZpyEiIiIiImIIFfPyRL744gu2b9/+rNN4qLVr1zJmzJhnnYaIiIiIiIghVMzLv4LmeRQRERERkeeJivkcLioqig8++ICqVavy2muvMWPGDNO+tLQ0Zs2aRbNmzahcuTI1atSgR48exMT8b6brbdu20aZNG6pVq0adOnUICAjg6tXsz9p76tQpunbtSo0aNfD09KRr166m5z18fX3Zs2cPK1aswNvbG4Dz588zYMAAXn31VSpVqkT9+vWZNGkSqampAISFheHt7c3o0aOpWbMmPXv2zFYeiYmJDB06lPr161OpUiXq1q3L0KFDuXnzpqlNTEwMH3/8MTVq1MDLy4u+ffsSHx9PWFgYgYGBwN1HF3bv3g3Ar7/+Srt27fD09OS1115j7Nix3Lp1yxTPw8ODSZMm0bBhQ+rWrcvJkyezfd1ERERERETMScV8Djd//nxatmzJunXr6NixI9988w27du0CYM6cOcyYMYOBAwfy888/M3XqVKKiohg7diwAly5donfv3rRt25b169fz3XffsXfvXsaNG5ft8/fr14+iRYuyfPlyli5dSt68eU2zNAYHB+Pp6UmzZs1YtmwZAD169ODSpUuEhISwceNGunXrxvTp0/nll19MMc+cOUNcXBwrVqygf//+2crjiy++4NChQ0yZMoWff/6ZwMBAwsLCWLx4MQBJSUl07NiR69evExoaSmhoKGfOnOGTTz7Bx8eHQYMGAbBjxw48PT35v//7Pz766CPq16/P8uXLGTlyJBs2bGDAgAEZzrt48WKmTJnC999/T5kyxq+XLCIiIiIi8jg0m30O16FDB1q1agVAr169mD17NkeOHKFOnTqUKlWKsWPHmu6Ku7i40KxZM9OEdHFxcSQnJ1OiRAlcXFxwcXFh+vTp3LlzJ9vnP336NHXr1qVkyZJYWFgQFBTEyZMnSU1NpXDhwlhaWlKgQAEcHBy4efMmLVu2pGnTpri4uAB3797/8MMPHD9+nMaNG5vi9urVC1dX12znUbduXWrWrMmLL74IQMmSJZk/f75plMD69etJSkpi0qRJFC5cGIDRo0ezatUq8uTJg42NDXB31kiAGTNm0KRJEz7++GMAypQpQ1paGh999BGRkZGULVsWgJYtW1KlSpVs5ykiIiIiIvI0qJjP4UqXLp3hd1tbW9NQcG9vbw4ePMiUKVOIjo4mMjKSv//+G2dnZwAqVqxI8+bN6dmzJ8WLF+fVV1+lQYMGpuI/O/r27UtQUBCLFi2idu3a1KtXj2bNmpE3b+ZBHQUKFKBz585s3LiROXPmEB0dTXh4OBcuXDANs0/n7u7+SNehY8eO/PLLL6xatYrTp08TERFBTEyMKc7x48dxd3c3FfIA5cuXz3SnPV1ERARvvfVWhm2vvPKKKVZ6Me/m5vZIeYqIiIiIiDwNGmafw+XLly/TtvTJ3GbOnImvry+XLl2iVq1aDBs2DH9//wxtJ06cyIYNG+jSpQvx8fH069cvU5sH6dSpE7/99huDBg2iYMGCfPPNN/j4+BAfH5+p7Y0bN2jfvj3Tpk3D2tqali1bsmDBAooVK5apbYECBbKdQ1paGj179mTkyJHky5ePpk2bMn36dF5++WVTGwsLC/LkyfNIMe9tnz5iwcLif99xPUqeIiIiIiIiT4vuzOdi06ZNo3fv3nz44YembSEhIaZi/8CBA6xfv55BgwZRpkwZunTpwurVqxk4cCAJCQk4Ojo+MH58fDxTp07lww8/pE2bNrRp04a4uDhef/119uzZg4+PT4b227dv5+jRo+zcuZMiRYoAcOXKFRISEp5oNvljx46xbds2lixZQrVq1YC768afPn3aNFS/XLlyLF26lKSkJNOQ+mPHjvHBBx8QFhaWqXCvUKEC+/fv5/333zdt27dvH4DprryIiIiIiEhOpTvzuVjx4sXZuXMnJ06c4OTJk0yaNIlNmzaRnJwMgLW1NQsXLmT8+PFER0dz/Phx1q1bh7u7O/b29g+NX7hwYX799Ve+/PJL/vrrL2JiYli4cCGWlpZUrlwZgEKFCnHmzBnOnz9vugO/evVqzpw5w759++jVqxcpKSmmnB5HkSJFsLCwYMOGDcTExHD48GE+++wzLl68aIrbokUL7OzsGDhwIOHh4Rw5coRhw4ZRoUIFXFxcsLKyAuDIkSPcvHmTrl27smnTJr7//nuioqLYunUrI0eOpGHDhirmRUREREQkx1Mxn4uNGzeOmzdv0rZtWzp37kxERATDhw8nISGB2NhYypUrR3BwML///jutWrWiY8eOWFhYMHPmzCyfeb/XP9t26dKFt956i99//50ffviBUqVKAdC+fXsiIiJ4++23qVSpEoGBgcydO5dmzZoRGBjIK6+8QvPmzTl48OBj99PZ2ZmxY8fyyy+/4OPjQ58+fXB2dqZLly4cPnyYtLQ0ChYsSEhICHfu3KFDhw507dqVsmXLMmXKFABq165NtWrVaN++PVu3bqVZs2ZMmDCBjRs30qJFC4YOHcpbb73F5MmTHztPERERERGRpyVP2pOMfxaRh0pMSjI0nq2NDZevGhsTwN7OhvMJlw2NWczR3iz9T0zIPGfDE8d1LMLVC3GGxrQr6kzClURDYzoWtuXK2RhDYxYu4crlmFOGxgSwd3XPNdf07MVLhsYEKOHkYJ7P/7/835RLV419/x3sbLl4+aqhMQGc7O1Iumxs/23s7UmMv2BoTNsiRQ2PaYprhs9q0tUrhsa0sSvMlfNnDY1ZuFgJQ+OJiNyP7syLiIiIiIiI5DKaAO9faubMmUydOvWBbQICAnjvvff+FXmIiIiIiIjkJirm/6XatWvHG2+88cA2Dg4O/5o8REREREREchMV8/9SdnZ22NnZPes0ckweIiIiIiIiuYmemRcRERERERHJZTSbvYiIiIiIiEguo2H2ImZmjmW0jF7uB+4u+WP08kxO9nbmWZrrUoKhMQFsHRy5fDrK0Jj2pUoTE2dsrq7Ojlw5d8bQmIWLu5BwItzQmACO5V40yzJi5ohp9BJqYMalGXNBzPS4UecuGhqzdHEnsyz3ZvS/03D332pzLM14Ne68sTGdi5nt/TfLMooGL6Np7+pulv9OXzj6p6ExAYpW8jQ8pojkbhpmLyIiIiIiIpLLqJgXERERERERyWVUzIuIiIiIiIjkMirmzcTDw4OwsLBnncYDnT17lnXr1j2Tc/v6+hIQEADA7t278fDwIDY21mznS0lJITQ01GzxRUREREREniYV8/9iX3zxBdu3b3/WaeDp6cmOHTsoXry42c6xdu1axowZY7b4IiIiIiIiT5Nms5dnLn/+/Dg5OZn1HFqBUUREREREnie6M29GUVFRfPDBB1StWpXXXnuNGTNmmPalpaUxa9YsmjVrRuXKlalRowY9evQgJibG1Gbbtm20adOGatWqUadOHQICArh6NftLh506dYquXbtSo0YNPD096dq1K8ePHwfuDnPfs2cPK1aswNvbG4Dz588zYMAAXn31VSpVqkT9+vWZNGkSqampAISFheHt7c3o0aOpWbMmPXv2zFYeycnJBAUFUadOHWrWrMnEiRNNMSHzMHtvb2+CgoLw8fHBy8uL33//nbS0NGbOnEmjRo2oVq0aLVu2ZPXq1RnOExMTw8cff0yNGjXw8vKib9++xMfHExYWRmBgIHD38Yfdu3cD8Ouvv9KuXTs8PT157bXXGDt2LLdu3TLF8/DwYNKkSTRs2JC6dety8uTJbF97ERERERERc1Ixb0bz58+nZcuWrFu3jo4dO/LNN9+wa9cuAObMmcOMGTMYOHAgP//8M1OnTiUqKoqxY8cCcOnSJXr37k3btm1Zv3493333HXv37mXcuHHZPn+/fv0oWrQoy5cvZ+nSpeTNm5fevXsDEBwcjKenJ82aNWPZsmUA9OjRg0uXLhESEsLGjRvp1q0b06dP55dffjHFPHPmDHFxcaxYsYL+/ftnK49Ro0axfv16xo4dy6JFizh79iz79u174DGLFi3iyy+/ZNasWbz88stMmjSJhQsX8uWXX7JmzRr8/PwYNmwYCxYsACApKYmOHTty/fp1QkNDCQ0N5cyZM3zyySf4+PgwaNAgAHbs2IGnpyf/93//x0cffUT9+vVZvnw5I0eOZMOGDQwYMCBDHosXL2bKlCl8//33lClTJnsXXkRERERExMw0zN6MOnToQKtWrQDo1asXs2fP5siRI9SpU4dSpUoxduxY011xFxcXmjVrZpqQLi4ujuTkZEqUKIGLiwsuLi5Mnz6dO3fuZPv8p0+fpm7dupQsWRILCwuCgoI4efIkqampFC5cGEtLSwoUKICDgwM3b96kZcuWNG3aFBcXF+Du3fsffviB48eP07hxY1PcXr164erqmq0crl27RlhYGEOHDqV+/foABAUFme6O30/9+vV59dVXAUwF+rhx42jYsCEApUqV4syZM4SEhNCpUyfWr19PUlISkyZNonDhwgCMHj2aVatWkSdPHmxsbABMw/lnzJhBkyZN+PjjjwEoU6YMaWlpfPTRR0RGRlK2bFkAWrZsSZUqVbLVVxERERERkadFxbwZlS5dOsPvtra2pmHc3t7eHDx4kClTphAdHU1kZCR///03zs7OAFSsWJHmzZvTs2dPihcvzquvvkqDBg1MxX929O3bl6CgIBYtWkTt2rWpV68ezZo1I2/ezAMyChQoQOfOndm4cSNz5swhOjqa8PBwLly4kGFIPIC7u3u2c4iKiiIlJSVDQfzCCy9QsWLFBx7n5uZm+vnEiRPcunWLL774wjRcHuD27dskJydz8+ZNjh8/jru7u6mQByhfvnymO+3pIiIieOuttzJse+WVVwA4fvy4qZj/Zx4iIiIiIiI5hYbZm1G+fPkybUufiG3mzJn4+vpy6dIlatWqxbBhw/D398/QduLEiWzYsIEuXboQHx9Pv379MrV5kE6dOvHbb78xaNAgChYsyDfffIOPjw/x8fGZ2t64cYP27dszbdo0rK2tadmyJQsWLKBYsWKZ2hYoUCDbOdyPhcWDv0f65znSr9nkyZNZuXKl6bV27Vo2bdpE/vz5sbCwIE+ePNk+f1paWqb26aMe/pmbEX0VERERERExmor5Z2TatGn07t2bYcOG8d5771G9enVOnTplKlwPHDhAUFAQZcqUoUuXLvzwww+m4ekJCQkPjR8fH8+IESNISUmhTZs2jB8/ntWrV3Px4kX27NmTqf327ds5evQo8+bN49NPP8XHxwdra2sSEhKeaCb4smXL8sILL7B//37Tttu3bxMeHp7tGGXKlMHCwoKzZ8/i5uZmem3bto2QkBDy5s1LuXLlOHXqFElJSabjjh07hpeXF2fOnMlUuFeoUCFDToDpOf70u/IiIiIiIiI5lYbZPyPFixdn586deHt7kzdvXlatWsWmTZsoUqQIANbW1ixcuBBLS0vatWvHzZs3WbduHe7u7tjb2z80fuHChfn11185ffo0/fv3x9rammXLlmFpaUnlypUBKFSoEGfOnOH8+fOmO/CrV6+madOmnDt3jm+++YaUlBSSk5Mfu59WVlZ07tyZKVOm4OTkRNmyZZk9ezZxcXHZjmFjY0P79u2ZPHkyhQoVokaNGuzbt4/x48fTvXt3AFq0aMHUqVMZOHAgn332Gbdv32bEiBFUqFABFxcXrKysADhy5AjlypWja9eu9O3bl++//x4fHx9OnTrFyJEjadiwoYp5ERERERHJ8VTMPyPjxo1jxIgRtG3blkKFClGtWjWGDx/OsGHDiI2NpVy5cgQHB/Pdd9+xcOFC8ubNS+3atZk5c2aWz7zfy8LCgpkzZ/L111/TpUsXbty4QcWKFfnhhx8oVaoUAO3bt+eLL77g7bffZteuXQQGBhIaGsrkyZNxdnbGx8eH4sWLc/DgwSfqa//+/XnhhRcYMWIE//3vf2nWrNkjPfsPEBgYiIODA1OmTOHChQsUK1aM3r178+GHHwJQsGBBQkJCGDt2LB06dCB//vx4e3vz+eefA1C7dm2qVatG+/btGT9+PM2aNePOnTvMmDGDadOm4eDgQPPmzfn000+fqK8iIiIiIiJPQ560JxlDLSIPlXAl0dB4joVtSfzH4wRGsbWx4eLlq4bGdLK3MzxXWxsbEi89/FGTR47r4Mjl01GGxrQvVZqYOGNzdXV25Mq5M4bGLFzchYQT2X/0Jbscy73IpavGfv4d7GzNEvN8wmVDYwIUc7Q3z+c/F8RMjxt17qKhMUsXdyLpsrHvlY29veH/TsPdf6uvXsj+KLTssCvqzNW488bGdC5mtvffHJ/VyzGnDI1p7+pulv9OXzj6p6ExAYpW8jQ8pojkbnpmXkRERERERCSX0TD7XGjmzJlMnTr1gW0CAgJ47733/hV5iIiIiIiI/NuomM+F2rVrxxtvvPHANg4ODv+aPERERERERP5tVMznQnZ2dtjZ2T3rNHJMHiIiIiIiIv82mgBPREREREREJJfRnXkRM7tyNsbQeIVLuJpt5u3d4acMjen1onlmCY47sNvQmADO1b04tXqRoTHd3+7Akh1PtrTjvdq9Vo0TC2cYGrNcxx6Gx0yPG7N5laExXZu0JCnR2M+Uja0t367dYWhMgD7NX+PyVWNn87a3szFLzKSrVwyNCWBjV5jF2w8YGvO9etU5ZfAM+e7FnTgSZewKEQCVS7sQ//cxQ2MWKf+SWd7/2G0bDY0JULL+m2b5Wz3z2yZDY7q8/oZZ/k09s32zoTEBXOo14cKhfYbGLFq1pqHxROTp0mz2IiIiIiIiIrmMinkRERERERGRXEbFvIiIiIiIiEguo2Jenhpvb288PDxMr8qVK9O0aVNmzZr1RHFjY2Px8PBg927jnqNOSUkhNDTUsHgiIiIiIiJG0gR48lT5+/vj7+8PwM2bNzl48CBffvklBQsWpFOnTs84u/9Zu3YtY8aMoUuXLs86FRERERERkUxUzMtTZWVlhZOTk+l3V1dXdu/ezfLly3NUMa8VG0VEREREJCfTMHt55goWLGj62dfXl4CAgAz7AwIC8PX1Nf0eERGBn58f1atXp2nTpvz++++ZYs6ZMwdvb2+qVq1Kly5d+O677/D29jbtj4uLo2/fvtSsWRMvLy969uzJqVOnAAgLCyMwMBDA8OH7IiIiIiIiRlAxL8/UoUOHWLNmDe+991622iclJdGlSxesra1ZunQpQ4YMYerUqRnaLFiwgG+++YZevXqxatUqvLy8+P777037r1+/jq+vL3fu3GH+/PnMmzcPe3t72rVrR1xcHD4+PgwaNAiAHTt24OnpaVyHRUREREREDKBiXp6qGTNm4OnpiaenJ5UrV+bdd9/F1dUVHx+fbB2/bt06bty4wddff0358uWpW7euqfBOFxISgp+fH++88w6lS5fmo48+onHjxhliXL58mYkTJ/Liiy9SoUIFRo8ejbW1NUuWLKFAgQLY2NgA4OTkRP78+Y27ACIiIiIiIgbQM/PyVLVv3940ZP727ducOnWKSZMm0bFjR5YvX/7Q4yMiInB3dzcV20CGO+eXL1/mzJkzVK9ePcNxNWrU4OjRowAcO3aMa9euUatWrQxtbt26RWRk5ON2TURERERE5KlRMS9PlZ2dHW5ubqbfy5Yti52dHZ06deI///kPkHnyuZSUlAy/37vfwsIi088PmsAuNTWV0qVLM23atEz7rKysstkTERERERGRZ0fD7CXHSE1NxdLSkqSkpAzbT58+bfq5YsWKREVFcenSJdO2w4cPm362sbHBxcWFAwcOZIhx6NAh088VKlTg7Nmz2NjY4ObmhpubGy4uLkycOJG9e/cCkCdPHiO7JiIiIiIiYigV8/JUXb9+nYsXL3Lx4kUuXLjAvn37CAoKomjRotSpU4eXX36Z//znP/zyyy/ExMQwZcoUIiIiTMe/9dZbODo60r9/f8LDw9mzZw9BQUEZztG9e3fmz5/PihUriI6OJjQ0lA0bNpj2v/3229jZ2dG7d28OHDhAZGQkgYGBbNu2jfLlywP/u0N/5MgRbt68+RSujIiIiIiISPapmJenavbs2bz22mu89tpr1K9fn08//RQXFxfmzJlDwYIF6dKlC02bNmXgwIG0bt2a+Ph4unTpYjreysqKuXPnYmlpSYcOHfj888/p3r17hnN06NCBnj17MmnSJJo3b8727dtp3bo1lpaWwN279/Pnz8fR0ZFu3brxzjvvcObMGUJCQkzFfO3atalWrRrt27dn69atT+36iIiIiIiIZIeemZen5pdffnloG2tra77++usHtilZsiQ//PBDhm2tW7c2/fzbb7/RsmVLevbsadr21VdfUaxYMdPvrq6uBAcH3/ccdnZ2LFmy5KH5ioiIiIiIPAu6My/PnVWrVvHRRx9x4MABzpw5w8qVK1m9ejUtW7Z81qmJiIiIiIgYQnfm5bnz1VdfMXbsWD7++GMSExMpVaoUgwYNok2bNs86NREREREREUOomJfnTuHChRk7duyzTkNERERERMRsNMxeREREREREJJfJk5aWlvaskxARERERERGR7NMwexEzO59w2dB4xRztSbiSaGhMAMfCtoSfPmdozBdLFefy1SRDY9rb2XA17ryhMQHsnIsRd3CvoTGdq71C1LmLhsYsXdyJuAO7DY3pXN2LM9s3GxoTwKVeEy4eO2hoTKeXqnH24iVDY5ZwcmD7kUhDYwLUq1yWxPgLhsa0LVKUpKtXDI1pY1fY8Jjpcc3xb0pSorH//tnY2hr+dwp3/1bNkWvipQRDY9o6OJJw8m9DYwI4linP1Qtxhsa0K+pslr+p2G0bDY1Zsv6bnFg4w9CYAOU69jBL/43OtVzHHobGE5H70zB7ERERERERkVxGxbyIiIiIiIhILqNiXkRERERERCSXUTF/H5cvX2bp0qXPOo3nnq+vLwEBAU8U4+rVq3Tt2pUqVapQr149UlNTDcpOREREREQkZ9IEePcxbtw4YmNjeffdd591KvIQK1euZPfu3cyfPx9nZ2fy5tV3VCIiIiIi8nxTMX8fWrEv90hKSsLJyYnq1as/61RERERERESeiuf6Fubff/9Nr1698PLyonLlyjRp0oQ5c+aY9u/cuZP27dtTrVo1Xn/9dSZOnMidO3cICAhgxYoV7NmzBw8PDwDu3LlDaGgoTZs2pUqVKjRt2pQlS5aYYu3evRsPDw9mzpyJl5cXrVu35s6dOw/NMSAggN69e+Pv78/LL7/MjBl3lwfZunUrbdq0oWrVqjRp0oTJkyeTnJxsOubeEQPnz5+nYsWK7Nq1C4A//viDTp06UbVqVRo0aMDw4cO5du2aqb23tzc//PADn3zyCZ6ennh5eREUFMTt27cBCAsLM/X93j7GxsYCd7/wmDlzJo0aNaJatWq0bNmS1atXZ+/N+Yfr16/Tv39/qlevTr169QgNDc3wZUpkZCTdu3fH09OT1157jf79+3Px4kXTtQgODubs2bN4eHgQHBwMwJ9//omfnx81atTAy8uLQYMGcfXq1Qz9DwoKwsfHBy8vL37//XfD+iMiIiIiImJuz+2d+Rs3bvDBBx9Qu3ZtFi5ciIWFBcuXLycoKIhatWqRnJxMt27deP/99xk9ejTnzp1jwIAB5M2bl8GDB3Pz5k3Onz9vKg7Hjh3LqlWr+Oqrr6hSpQo7d+5kxIgR3Lp1C19fX9N5f/31VxYvXsyNGzfIly9ftnLdvHkzAwcO5KuvvqJAgQL89ttv9OnTh8DAQOrWrcvp06cZOXIkUVFRfPvtt7Ru3Ro/Pz+io6Nxc3MDYPXq1Tg7O+Pl5UV4eDhdunShZ8+ejB49mvj4eMaNG4e/vz+LFy8mT548AAQHBzNw4ED69+/Pjh07GDVqFC+99BKtWrXKVt6TJk1izZo1DBkyhLJly7J3716GDRtGUlISnTp1yvZ79fPPP+Pr68vy5cs5evQoQ4cOBaBLly7ExcXRsWNH3nrrLQICArhx4wbBwcG0b9+eNWvWMHjwYOzt7Vm/fj3Lli3DysqKQ4cO4evrS7t27RgyZAgJCQmMHDkSf39/li5dahqGv2jRImbMmIGNjQ0eHh6G9UdERERERMTcnuti3s/Pj44dO2JtbQ1A7969mTFjBsePH2f79u1UrVrVNPla2bJlGTlyJBcuXMDGxoYCBQpgaWmJk5MT165dY9GiRQQEBNCiRQsA3N3diYmJYfr06XTu3Nl0Xn9/f9zd3R8pVzs7O7p162b6vX///rzzzjt06NABgFKlSjF8+HDef/99YmNjqVWrFq6urqxZs4bevXsDsGbNGlq2bEnevHkJCQmhTp069OrVy5TrxIkTady4MXv27MHLywuAevXq4efnZ2qzbNky/vjjj2wV89evXyc0NJRx48bRsGFDU55nzpwhJCTkkYrfl156iS+//BK4+z5ERkYye/ZsunTpwqJFiyhatChDhgwxtZ88eTK1a9dm48aNtGnTBisrK/Lly4eTkxMAs2fPxsPDw3RMuXLlmDhxIm+//Tbbt2+nfv36ANSvX59XX33V8P6IiIiIiIiY23NbzDs4ONCxY0fWr19PeHg40dHR/PXXXwCkpqZy/PhxUyGXrkmTJlnGOnnyJCkpKdSoUSPD9po1a/Ljjz+SkJBg2vaohTxgurue7tixYxw6dIgVK1aYtqUPO4+MjKRkyZK0atXKVMz/9ddfREREMGXKFNPx0dHReHp6ZjpXZGSkqZgvW7Zshn02NjakpKRkK+cTJ05w69YtvvjiCwIDA03bb9++TXJyMjdv3qRAgQLZinXvda1atSrTp08nMTGRY8eOERkZmakvt27dIjIyMst4ERER1K1bN8M2Dw8PbG1tOX78uKmY/+d1N7I/IiIiIiIi5vbcFvPx8fG0a9cOe3t7GjVqRJ06dahSpYqpkLOwsDANN3+Y9EL63vbpS6BZWPzvMr7wwguPnOu9RWJqairdunWjdevWmdqm331u3bo13333HYcOHWLDhg14enpSunRp0/EtWrSgZ8+emY53cHAw/Zw/f/5M+++d+C8tLc3U7/Tn6f/ZbvLkyZQpUyZTnKxi38+9s8+npqaSJ08eLC0tSU1NpXbt2qah9/9kY2OTZbx/5nxvXEtLS9Pv/7zuRvZHRERERETE3J7bCfDWrFnDlStX+Omnn+jVqxdNmjQxTYCWlpZG2bJlOXz4cIZjQkNDTQX0P4vBMmXKYGFhwb59+zK037dvH05OTtjZ2Rmae/ny5Tl58iRubm6mV1xcHOPGjeO///0vAC4uLtSqVYuNGzeyfv36DIV/+fLl+fvvvzMcf+fOHcaMGcO5c+eylUN60ZuUlGTaFh0dbfo5/ZqcPXs2w3m2bdtGSEjIIy0Pd/To0Qy/79+/n5IlS1KwYEHKly9PZGQkxYsXN53Dzs6OoKAgIiIisoxXoUKFTO9VeHg4165dyzQawRz9ERERERERMbfntkIpVqwYN27cYMOGDZw9e5YdO3bQr18/ANPkdwcOHGDy5MlERUWxbds2ZsyYQaNGjQCwsrLiwoULxMTEYGNjQ7t27ZgyZQpr1qwhOjqaBQsWsHDhQvz9/bN9hz+7unfvzqZNmwgODiYqKopdu3YRGBhIYmKi6c48QJs2bfjpp5+4fPkyPj4+pu3+/v789ddfDBkyhBMnTnDw4EEGDBhAVFRUth8DqF69Onnz5mXy5MnExMTw66+/Mnv2bNN+Gxsb2rdvz+TJk1m5ciUxMTGsWLGC8ePHU6RIkUfq7x9//MH48eOJjIxk6dKlLFy40PS8f8eOHUlKSqJfv3789ddfhIeH079/fw4dOkT58uWzjNelSxfCw8MZMWIEkZGR7NmzhwEDBvDSSy9Rp06dLI8xsj8iIiIiIiLm9twOs3/zzTc5evQoX3/9NdeuXcPFxYV3332XLVu2cOjQITp06MDUqVOZMmUKs2bNwsnJCV9fX9PQ9FatWrF582aaN2/O5s2bTbOmT5w4kfj4eNzc3BgyZAjt2rUzS+6TJk1ixowZzJgxAzs7Oxo2bMjAgQMztGvatCkjRoygcePGGYacV69enVmzZvHtt9/Spk0bChYsSO3atfniiy+yPVzc1dWVESNGMH36dJYsWUKlSpUYNGgQH330kalNYGAgDg4OTJkyhQsXLlCsWDF69+7Nhx9++Ej9fffddzl16hStW7fGwcGB/v3706ZNG1Me8+fPZ+LEiXTs2JF8+fJRvXp15syZg6OjY5bxPD09mTlzJt9++y2tWrXC2tqaxo0b079//wzD7O9lVH9ERERERETMLU/avQ9Ji4ihzidcNjReMUd7Eq4kGhoTwLGwLeGns/cYRna9WKo4l68mPbzhI7C3s+Fq3HlDYwLYORcj7uBeQ2M6V3uFqHMXDY1ZurgTcQd2GxrTuboXZ7ZvNjQmgEu9Jlw8dtDQmE4vVePsxUuGxizh5MD2I1lPqPkk6lUuS2L8BUNj2hYpStLVK4bGtLErbHjM9Ljm+DclKdHYf/9sbG0N/zuFu3+r5sg18VLCwxs+AlsHRxJO/m1oTADHMuW5eiHO0Jh2RZ3N8jcVu22joTFL1n+TEwtnGBoToFzHHmbpv9G5luvYw9B4InJ/z+0wexEREREREZHn1XM7zP5ZW79+PYMHD35gGz8/P/r27fuUMnq6Zs6cydSpUx/YJiAggPfee+8pZSQiIiIiIvL8UDFvJvXr12flypUPbGNra/t0knkG2rVrxxtvvPHANv9cJk9ERERERESyT8W8mRQqVIhChQo96zSeGTs7O8OX7BMREREREZG79My8iIiIiIiISC6j2exFREREREREchkNsxcxM6OXkXMsbGv4cm9wd8m33LI03ZWzMYbGBChcwpXz+3cZGrNYjTociTpjaMzKpV3MsoRe9IblhsYEcGvW1vAlrxzLlDfLco9bDxm/NFfDquW5fDrK0Jj2pUqTmGTs35StjQ1Jl429pgA29vbsOX7K0Ji1PNzNstxZTJyxy70BuDo7Gv5vVeESriQmxBsa09axCJeiThgaE8ChdDmz9N8c/6bEhx82NGaRF6tw7vdthsYEKF67vlmWkcsNMdPjikhGGmYvIiIiIiIiksuomBcRERERERHJZXJNMe/h4UFYWBjBwcF4e3ub9VyXL19m6dKlZj2HgK+vLwEBAU8U4+rVq3Tt2pUqVapQr149UlNTDcpOREREREQk58p1z8z7+/vTqVMns55j3LhxxMbG8u6775r1PPLkVq5cye7du5k/fz7Ozs7kzZtrvp8SERERERF5bLmumH8a67drgv/cIykpCScnJ6pXr/6sUxEREREREXlqcuRtzPPnz/PRRx/h6elJgwYNWLdunWnfP4fZx8bG4uHhwdSpU6lbty7e3t4kJiaSlJTEV199Re3atalRowZ+fn4cPpxxptKdO3fSvn17qlWrxuuvv87EiRO5c+cOAQEBrFixgj179uDh4QHAnTt3CA0NpWnTplSpUoWmTZuyZMkSU6zdu3fj4eHBzJkz8fLyonXr1ty5c+eh/QwICKB37974+/vz8ssvM2PG3Zk/t27dSps2bahatSpNmjRh8uTJJCcnm465d8TA+fPnqVixIrt23Z2J+48//qBTp05UrVqVBg0aMHz4cK5du2Zq7+3tzQ8//MAnn3yCp6cnXl5eBAUFcfv2bQDCwsJMfb+3j7GxscDdLzxmzpxJo0aNqFatGi1btmT16tUP7fO9rl+/Tv/+/alevTr16tUjNDQ0w5cpkZGRdO/eHU9PT1577TX69+/PxYsXTdciODiYs2fP4uHhQXBwMAB//vknfn5+1KhRAy8vLwYNGsTVq1cz9D8oKAgfHx+8vLz4/fffDeuPiIiIiIjI05Dj7szfvn2bbt26YW1tzfz580lOTmb48OEPPGb16tXMmTOHGzduYGNjQ4cOHbC0tGTGjBlYW1uzatUqOnTowJIlS3jppZc4ePAg3bp14/3332f06NGcO3eOAQMGkDdvXgYPHszNmzc5f/68qTgcO3Ysq1at4quvvqJKlSrs3LmTESNGcOvWLXx9fU15/PrrryxevJgbN26QL1++bPV38+bNDBw4kK+++ooCBQrw22+/0adPHwIDA6lbty6nT59m5MiRREVF8e2339K6dWv8/PyIjo7Gzc3N1H9nZ2e8vLwIDw+nS5cu9OzZk9GjRxMfH8+4cePw9/dn8eLF5MmTB7j7pcjAgQPp378/O3bsYNSoUbz00ku0atUqW3lPmjSJNWvWMGTIEMqWLcvevXsZNmwYSUlJj/QYxM8//4yvry/Lly/n6NGjDB06FIAuXboQFxdHx44deeuttwgICODGjRsEBwfTvn171qxZw+DBg7G3t2f9+vUsW7YMKysrDh06hK+vL+3atWPIkCEkJCQwcuRI/P39Wbp0qWkY/qJFi5gxYwY2NjZ4eHgY1h8REREREZGnIccV87t27eLvv/9m8+bNlCpVCoAxY8Y8sMjs2LEj5cqVMx3/559/smvXLhwcHADo168ff/zxB3PnzmXs2LHMnTuXqlWrmiZfK1u2LCNHjuTChQvY2NhQoEABLC0tcXJy4tq1ayxatIiAgABatGgBgLu7OzExMUyfPp3OnTub8vD398fd3f2R+mtnZ0e3bt1Mv/fv35933nmHDh06AFCqVCmGDx/O+++/T2xsLLVq1cLV1ZU1a9bQu3dvANasWUPLli3JmzcvISEh1KlTh169eplynThxIo0bN2bPnj14eXkBUK9ePfz8/Extli1bxh9//JGtYv769euEhoYybtw4GjZsaMrzzJkzhISEPFLx+9JLL/Hll18Cd9+HyMhIZs+eTZcuXVi0aBFFixZlyJAhpvaTJ0+mdu3abNy4kTZt2mBlZUW+fPlwcnICYPbs2Xh4eJiOKVeuHBMnTuTtt99m+/bt1K9fH4D69evz6quvGt4fERERERGRpyHHFfMRERHY2dmZCnmAihUrUrBgwfsek36HGuDo0aMANGrUKEOb5ORkbt26BcDx48dNhVy6Jk2aZBn75MmTpKSkUKNGjQzba9asyY8//khCQoJp26MW8vfmDnDs2DEOHTrEihUrTNvSh51HRkZSsmRJWrVqZSrm//rrLyIiIpgyZYrp+OjoaDw9PTOdKzIy0lTMly1bNsM+GxsbUlJSspXziRMnuHXrFl988QWBgYGm7bdv3yY5OZmbN29SoECBbMW697pWrVqV6dOnk5iYyLFjx4iMjMzUl1u3bhEZGZllvIiICOrWrZthm4eHB7a2thw/ftxUzP/zuhvZHxERERERkachxxXzkPUEdBYW90/1n4VWamoq1tbWhIWFZWqXP39+U6z04ebZzeXe9ulLoP0zrxdeeCFbMf/p3iIxNTWVbt260bp160xt0+8+t27dmu+++45Dhw6xYcMGPD09KV26tOn4Fi1a0LNnz0zHp49UgP9di3+697qnpaWZ+p3+PP0/202ePJkyZcpkipNV7Pu5d/b51NRU8uTJg6WlJampqdSuXds09P6fbGxssoz3z5zvjWtpaWn6/Z/X3cj+iIiIiIiIPA05bgK8l156icTERP7++2/TtqioKJKSkrJ1fIUKFbh27RrJycm4ubmZXjNnzmTLli3A3bvS906IFxoaaiqg/1kMlilTBgsLC/bt25eh/b59+3BycsLOzu6x+nk/5cuX5+TJkxlyj4uLY9y4cfz3v/8FwMXFhVq1arFx40bWr1+fofAvX748f//9d4bj79y5w5gxYzh37ly2ckgvev95zaOjo00/p1+Ts2fPZjjPtm3bCAkJeaTl4dJHUqTbv38/JUuWpGDBgpQvX57IyEiKFy9uOoednR1BQUFERERkGa9ChQqZ3qvw8HCuXbuWaTSCOfojIiIiIiLyNOS4KsXLy4tq1arx+eefc+DAAQ4fPkxAQEC2C6p69epRsWJFPvvsM3bt2kV0dDRff/01y5cvNxVz3bp148CBA0yePJmoqCi2bdvGjBkzTEPzraysuHDhAjExMdjY2NCuXTumTJnCmjVriI6OZsGCBSxcuBB/f/9s3+HPru7du7Np0yaCg4OJiopi165dBAYGkpiYaLozD9CmTRt++uknLl++jI+Pj2m7v78/f/31F0OGDOHEiRMcPHiQAQMGEBUVle3HAKpXr07evHmZPHkyMTEx/Prrr8yePdu038bGhvbt2zN58mRWrlxJTEwMK1asYPz48RQpUuSR+vvHH38wfvx4IiMjWbp0KQsXLjQ979+xY0eSkpLo168ff/31F+Hh4fTv359Dhw5Rvnz5LON16dKF8PBwRowYQWRkJHv27GHAgAG89NJL1KlTJ8tjjOyPiIiIiIjI05DjhtnnzZuXGTNmMGrUKPz9/SlQoAA9evQwLYn2MPny5WP27NmMHz+evn37cuPGDcqWLUtwcLCpmKtYsSJTp05lypQpzJo1CycnJ3x9fU1D01u1asXmzZtp3rw5mzdvNs2aPnHiROLj43Fzc2PIkCG0a9fO8P6/+eabTJo0iRkzZjBjxgzs7Oxo2LAhAwcOzNCuadOmjBgxgsaNG2cYcl69enVmzZrFt99+S5s2bShYsCC1a9fmiy++yPZwcVdXV0aMGMH06dNZsmQJlSpVYtCgQXz00UemNoGBgTg4ODBlyhQuXLhAsWLF6N27Nx9++OEj9ffdd9/l1KlTtG7dGgcHB/r370+bNm1MecyfP5+JEyfSsWNH8uXLR/Xq1ZkzZw6Ojo5ZxvP09GTmzJl8++23tGrVCmtraxo3bkz//v0zDLO/l1H9EREREREReRrypGX1gLqIGCbhSqKh8RwL23L5avYeO3kU9nY2hJ/O3qMY2fViqeKG52pvZ8OVszGGxgQoXMKV8/t3GRqzWI06HIk6Y2jMyqVdiDu419CYztVeIXrDckNjArg1a0vCyb8f3vAROJYpz/mEy4bGLOZoz9ZDxuYJ0LBqeS6fjjI0pn2p0iRm87Gz7LK1sSHpsrHXFMDG3p49x08ZGrOWhztXL8QZGtOuqDMxcQkPb/iIXJ0dDf+3qnAJVxIT4g2NaetYhEtRJwyNCeBQupxZ+m+Of1Piww8/vOEjKPJiFc79vs3QmADFa9fnxMIZhsYs17FHroiZHldEMspxw+xFRERERERE5MFy3DD758H69esZPHjwA9v4+fnRt2/fp5TR0zVz5kymTp36wDYBAQG89957TykjERERERGR54uKeTOoX78+K1eufGAbW1vbp5PMM9CuXTveeOONB7b55zJ5IiIiIiIi8mhUzJtBoUKFKFSo0LNO45mxs7MzfMk+ERERERER+R89My8iIiIiIiKSy2g2exEREREREZFcRsPsRczsyvmzhsYrXKyE4cvdwd0l74xensrG3t4sy2hFrZhnaEyA0q19zbKM2Mrfjxgas1XtysT+usHQmCUbNOPkslBDYwKUeaeLWZY8Soy/YGhM2yJFCZi30dCYAGN93yQp0di/VRtbWy5dNTamg50t8X8fMzQmQJHyL5nl82+Oa/qfYycNjQnw6ktlzPJvSuwFY5fRK1nU0WzLiJnj339zLE0Xu83Yv/+S9d80/L2Hu+9/0tUrhsa0sSucq5ami1o539CYpVt1NjSeyNOmYfYiIiIiIiIiuYyKeREREREREZFcRsW8iIiIiIiISC6jYl6emeDgYLy9vZ84zujRo6lRowaenp4cOWLs85kiIiIiIiI5kSbAk1wtPDycuXPnMmTIEBo0aEDRokWfdUoiIiIiIiJmp2JecrXE/z+r8euvv46Li8szzkZEREREROTpeC6G2V+6dIm+fftSs2ZNvLy8GD9+PH5+fgQHBwOwdetW2rRpQ9WqVWnSpAmTJ08mOTnZdLyHhwdLlizhgw8+oGrVqtSrV48ZM/63pEZwcDDt27enX79+vPzyywwfPhyAP/74g06dOlG1alUaNGjA8OHDuXbtWrbzDgsLw9vbm9GjR1OzZk169uwJQGRkJN27d8fT05PXXnuN/v37c/HiRdMxVapUMRWx6Zo2bcrkyZMBiIuLy3A9evbsyalTp0xtAwICGDhwIF9//TV16tShWrVq9OrVy3SO2NhYPDw82L17d4ZzeHh4EBYWZvr9Ydc1u6ZOnUrt2rWpUaMGQ4YM4fr166Z9SUlJfPXVV6b9fn5+HD582HQtfH19AWjcuLHp53PnzjFgwADq1q1L9erV6dq1K8ePH8/Q/969e+Pv78/LL79seq+N6o+IiIiIiIi55fpiPjU1lR49ehAdHc3MmTOZPXs2hw4dYs+ePQD89ttv9OnTh3fffZe1a9cydOhQNmzYwMCBAzPEGTduHK1atWLVqlW0bduWb775hn379pn2//nnnzg6OrJq1Sref/99wsPD6dKlC3Xr1mX16tVMmDCBo0eP4u/vT1paWrbzP3PmDHFxcaxYsYL+/fsTFxdHx44dcXV1ZdmyZUyfPp1r167Rvn17rl+/zptvvomFhQU///yzKcbBgwc5deoUrVq14vr16/j6+nLnzh3mz5/PvHnzsLe3p127dsTFxZmO2bBhA1euXGH+/Pl899137N+/n0mTJmU77+xe1+z0f9euXcyePZtp06axe/du+vfvD0BaWhrdu3fn1KlTzJgxgyVLllC9enU6dOjAsWPH8PHxMX1hs3TpUoKDg7l27RodOnQgLi6OadOm8dNPP2FlZUXnzp05e/Z/671v3ryZV199leXLl/P2228b1h8REREREZGnIdcX83v27OHQoUNMmDABT09PKlWqxOTJk8mfPz8A06dP55133qFDhw6UKlWK1157jeHDh7Nx40ZiY2NNcVq3bk3Lli0pXbo0n332GXZ2duzfvz/DuT799FNcXV1xd3cnJCSEOnXq0KtXL9zd3alZsyYTJ07k4MGDpi8SsqtXr164urpSvnx5Fi1aRNGiRRkyZAhly5alcuXKTJ48mfj4eDZu3IiVlRVvvvkmq1evNh2/evVqXn75Zdzd3Vm3bh2XL19m4sSJvPjii1SoUIHRo0djbW3NkiVLTMdYW1szYsQIypYtS7169WjZsmWm/j5Idq/rw+TPn59Jkybx0ksvUatWLYYMGcIvv/xCdHQ0v//+O3/++Sfffvst1apVo2zZsvTr14/q1aszd+5cChQogJ2dHQAODg4ULlyY1atXc/nyZb799luqVq3Kiy++yIQJEyhQoAALFiwwndfOzo5u3bpRunRpihcvblh/REREREREnoZc/8z8sWPHsLOzo0yZMqZtjo6OlC5d2rT/0KFDrFixwrQ//c55ZGQkJUuWBKBs2bIZ4lpbW5OSkpIhpo2NTYbzRkdH4+npmSmnyMhIvLy8st0Hd3f3DHEjIyMzxb116xaRkZHA3S8e/Pz8OH/+PEWKFGH9+vX07dvXdPy1a9eoVavWfY8HcHNzw9LS0vS7jY1Nhv4+THav68O4u7tTpEgR0+/VqlUD4O+//zY9GtCoUaMMxyQnJ3Pr1q0s40VERODu7o6Dg4Np2wsvvEDVqlUzDLV3c3MzS39ERERERESehlxfzOfLl4/U1NT77k9NTaVbt260bt060z4nJyfTz+l38v/pn8PlCxQokCluixYtTM+5/9M/C8ns+Gfs1NRUateuzdChQzO1S/8y4ZVXXqFkyZKsXbuWsmXLcv36dZo1a2Y6vnTp0kybNi3T8VZWVqafs+rvvf7Z/3sL/exe14fJly9fht/v3LkDgKWlJampqVhbW2d4Tj/d/fJPS0sjT548mbbfuXMHC4v/fdyzej+N6I+IiIiIiMjTkOuH2b/44oskJSVluOt85coVoqOjAShfvjwnT57Ezc3N9IqLi2PcuHH897//fezzli9fnr///jtD3Dt37jBmzBjOnTv3RHEjIyMpXry4Ka6dnR1BQUFEREQAkCdPHlq1asXPP//MunXraNy4sanQr1ChAmfPnsXGxsZ0vIuLCxMnTmTv3r3ZyiH9jv0/J/M7ffp0pjyNuK6nTp3KcJ79+/eTJ08eypUrR4UKFbh27RrJyckZzjNz5ky2bNmSZbwKFSoQFRVFQkKCadutW7c4cuQI5cqVu28e5vqciIiIiPw/9u48Lsb9/R/4a1odCp1s2UolISlLZF+yRpu9VZJ9KeqgiFTaFSKhyBaiHcdykMRB9pBIEpFT0YbW+/dHv+5vY4qZe+amfN7Px+M8ju6Zrt7NfTdzv7frIgiCYEOT78wPGjQIWlpa+Ouvv3D//n2kpaXBwcEBX758AYfDga2tLc6fP48dO3YgMzMTN27cwLp161BUVCTUjOu8efPw9OlTuLi44MWLED4SPwAA0QZJREFUF3jw4AEcHByQmZnJtWxeUKampiguLsaqVavw9OlTpKWlYfXq1Xj48CG6d+9OP8/Y2BiPHz/GP//8wzWbbGBggFatWmHZsmW4f/8+MjIysG7dOiQmJnJ9//e0a9cOXbp0wf79+/HixQs8evQIGzZs4JoNF9XrWlZWBjs7Ozx58gTJyclwc3ODkZEROnXqhOHDh6Nnz56ws7PDjRs3kJWVBW9vb5w6dYpnW0StqVOnomXLlrCzs8PDhw+RlpYGR0dHfP78GbNmzWqwHWxdJwRBEARBEARBEGxo8p15ANi+fTs6dOiAuXPnwsrKCn369EHHjh0hKSmJiRMnIiAgAP/88w+mTp0KBwcH6OrqIigoSKifqaWlhX379iE9PR0mJiZYsGAB3QHmZwl7Q7p06YLDhw/jy5cvMDU1hbm5OTgcDsLDwyEvL08/r1OnTtDR0YGMjAyGDBlCH5eVlcXhw4chLy+P+fPnY/r06Xj79i1CQ0P57sxzOBz4+vqirKwMRkZGWL16NUxNTdGhQwf6OaJ6XTU0NNCzZ09YWlrCzs4OI0aMoLcYiIuLIywsDJqamrC3t4eBgQFu3ryJHTt2QFdXt954LVu2xOHDhyErK4u5c+fC1NQUX758QUREBLp06dJgO9i6TgiCIAiCIAiCINjAoQSpo9YIFRQU4MGDBxg2bBi9PLy8vByDBg3Cxo0bYWRk9GsbSPzP+/Q+58dPEkDrDh2R/6lIpDEBQL51SxR//CjSmLJycigqLhZpzJayssiMPiTSmADQzdgCH19nijSmXNduiPk3VaQxjQZr4M2VsyKN2XnUJLw8eUCkMQFAefpcvDgaItKYqqYLUZT3QaQxW7Zph7WH/hZpTADwspiI4iLR/q3KtmyJgkLRxvyzVUvkPX8i0pgA0KZ7L1aufzZe0+tPXoo0JgAM6aXMynvKmw/5P36iADq3kxf53ynw//9WWXj/z3/5XKQx5ZW7402iaP/+O4+cKPJzD9Sc/+LCTyKNKduqNSvv02xdU5kxh0Uas5uRuUjjEcTP1uQT4ElISMDe3h6zZ8/GnDlzUFFRgdDQUEhJSWHEiBG/unkEQRAEQRAEQRAEIXJNvjPfsmVL7N69G4GBgTh+/Dg4HA769++PgwcPCpxVXlRyc3MxceLE7z6nV69eXHXPfyf37t3DvHnzvvscPT09+Pr6/qQWEQRBEARBEARB/F6afGceAAYPHoxjx4796mbQ2rRpg5iYmO8+R1pa+uc05hfo1avXD3//umXyCIIgCIIgCIIgCMH8Fp35xkZcXByKioq/uhm/jLS09P/0708QBEEQBEEQBMG2Jp8AjyAIgiAIgiAIgiD+15CZeYJgGRuZlws/5Io0JgC0ateelSzZbGQzFvVrCtS8ru/zRZvNv4O8HApz34s0Zqv2HZCX9kikMduo9xH5eQJqztXHQtHGlWsly841JeJKDgB71RyaQkw6bn6eaGPKt2Hl75StzONsVF74UPBJpDHb/dmatQopbFyrbLynNIX3KYCd9ypZOTm8eCvaewrVTu1F/ncK1Pyt/vexUKQx28q1wqt3/4k0ppJCW5HGI4jv+S3qzBMEQRAEQRAEQRDE/xLSmScIgiAIgiAIgiCIJoZ05gmCIAiCIAiCIAiiiSF75okmz8LCArdu3eI6JikpiXbt2mHs2LFYvXo1mjVrRj92+vRpRERE4OnTp6iurka3bt1gaGgIU1NTSEpK0s9bu3Yt3r59i0OHDv2034UgCIIgCIIgCIIfpDNP/BYmTZoEZ2dn+uvPnz/j2rVr8PT0RFVVFVxcXAAAGzZsQEJCAhYvXoxNmzZBQkICKSkp2LFjB86ePYvQ0FC0aNHiV/0aBEEQBEEQBEEQfCGdeeK30KxZM7Rty509VFFREampqTh9+jRcXFwQExODU6dO4fDhw+jXrx/9PCUlJQwfPhwGBgbw8fGBq6vrz24+QRAEQRAEQRCEQMieeeK3Ji0tDTGxmsv84MGDGDlyJFdHvlb79u1hZWWFqKgoFLNQToYgCIIgCIIgCEKUSGee+C1VVlbiypUriI2NhaGhIb5+/YqnT5/W25Gvpauri/LycqSmpv7ElhIEQRAEQRAEQQiOLLMnfgvx8fE4d+4c/fXXr1/RsWNH2NjYYNGiRcjPz0d1dTVat27dYAw5OTkAQEFBAdvNJQiCIAiCIAiCEArpzBO/hTFjxsDBwQHV1dV48OABPD09MWTIECxatAgSEhJ0J76kpKTBGEVFRQD+r1NPEARBEARBEATRWJHOPPFbaNGiBRQVFQEA3bp1Q4cOHWBtbQ1xcXFs2rQJ0tLS6NOnD27evAlra+t6Y/z777+QkpJC7969f2bTCYIgCIIgCIIgBEb2zBO/pcGDB8Pa2hoRERG4evUqAGDevHlITEzEjRs3eJ6fm5uLAwcOwMDAAK1atfrZzSUIgiAIgiAIghAI6cwTv62VK1dCSUkJGzduRGlpKSZPngwzMzMsWrQIISEhyMjIQHZ2NmJiYjB79mwoKChg3bp1v7rZBEEQBEEQBEEQP0Q688RvS1paGm5ubnj37h0CAgIAAOvXr4e/vz9u3LgBU1NTGBgYIDw8HBYWFoiIiICMjMwvbjVBEARBEARBEMSPkT3zRJN36NChBh/T0dFBWloa1zE9PT3o6en9MK6Xl5fQbSMIgiAIgiAIgmADmZknCIIgCIIgCIIgiCaGdOYJgiAIgiAIgiAIookhnXmCIAiCIAiCIAiCaGJIZ54gCIIgCIIgCIIgmhjSmScIgiAIgiAIgiCIJoZDURT1qxtBEARBEARBEARBEAT/SGk6gmDZx9eZIo0n17Ub8j8ViTQmAMi3bomnWe9EGrOnogI+FhaLNKZcK1l8eHxPpDEBoF1vbbxNuiDSmJ2Gj8PZlKcijTlpQE+8OBoi0piqpgtFHrM2bvY/CSKN2WXsFBR//CjSmLJycvCPvSrSmACw2nAEiopFe/23lJVl5W9K1O0Eatp66UG6SGOO6auGjJwPIo2p0rEdXr37T6QxAUBJoS0Ksl6KNOafisp4lyfa61+hjRxyrl8SaUwA6DhkDCvXf+6D2yKN2b7vQFbeU9/dEv17ioLOCLxPSRZpzA4Dhjapz5Sm0FZV04XIf/lcpDEBQF65u8hjEk0fWWZPEARBEARBEARBEE0M6cwTBEEQBEEQBEEQRBNDltkTjFlYWODWrVtcxyQlJdGuXTuMHTsWq1evRrNmzX5R6wiCIAiCIAiCIH5fpDNPCGXSpElwdnamv/78+TOuXbsGT09PVFVVwcXF5Re2jiAIgiAIgiAI4vdEOvOEUJo1a4a2bdtyHVNUVERqaipOnz5NOvMEQRAEQRAEQRAsIHvmCVZIS0tDTKzm8iovL4evry+GDx8ObW1tzJw5E9euXQMAlJaWQltbG0ePHuX6/uDgYIwaNQrV1dWgKAp79+7F2LFj0bdvXxgaGiIuLo5+7s2bN9GjRw8kJiZiypQp0NDQgL6+Pi5fvkw/x8LCAmvXruX6GWvXroWFhQX9dW5uLuzt7TFgwAAMGjQIixYtwqtXr0T90hAEQRAEQRAEQQiNdOYJkaqsrMSVK1cQGxsLQ0NDAMC6deuQlJQEX19fREdHY9KkSVi0aBGuXLmCFi1aYMKECYiPj+eKEx8fD0NDQ4iJiSEgIABHjx7F+vXrER8fD0tLS2zatAlHjhzh+h5fX184OzsjKioKXbp0gYODA0pLS/lq9+fPn2FhYYGqqiocPnwYhw4dgpycHGbOnInc3FzRvDgEQRAEQRAEQRAiQpbZE0KJj4/HuXPn6K+/fv2Kjh07wsbGBosWLUJWVhYSEhJw8uRJ9OnTBwBgbW2NtLQ0hIaGYtSoUTAxMYGlpSXevHmDzp07IzU1FRkZGdi1axc+f/6MAwcOwMfHB6NHjwYAdO3aFW/fvkVoaCjMzMzon21nZwddXV3634aGhkhPT4e2tvYPf4/Tp0/j48eP8Pf3h6SkJADAw8MDN2/exIkTJ7B8+XKRvWYEQRAEQRAEQRDCIp15QihjxoyBg4MDqqur8eDBA3h6emLIkCFYtGgRJCQk8OTJEwCApaUl1/dVVFSgZcuWAICBAweic+fOSEhIwKJFixAbGwttbW0oKSnh4cOHKCsrw5o1a7Bu3Tr6+ysrK1FeXo6vX7/Sx5SVlel/y8jI0D+HH0+ePEFJSQl0dHS4jpeVlSEjI0OAV4QgCIIgCIIgCIJ9pDNPCKVFixZQVFQEAHTr1g0dOnSAtbU1xMXFsWnTJlAUBQA4cuQIWrRowfW9tXvqORwOjIyMEB8fD1tbW5w5cwYrV64EAPr7AwMDuTrrtaSkpOr9d63a7//23wB3R7+6uhrdunVDcHAwT4zmzZt/5xUgCIIgCIIgCIL4+cieeUKkBg8eDGtra0RERODq1avo3r07AODDhw9QVFSk/4uKisKpU6fo7zM2NkZGRgaOHj2K4uJiTJo0CUDNbLuEhARycnK4vj8xMRGhoaH0gMCPSEpKori4mOvY69ev6X+rqakhJycHsrKy9M/o1KkT/P39cfv2bWFfFoIgCIIgCIIgCJEinXlC5FauXAklJSVs3LgRHTt2xOjRo7Fx40b8888/yM7ORmhoKEJCQtClSxf6ezp16oRBgwYhICAA48aNg6ysLABAVlYWs2fPRmBgIGJiYpCdnY3o6Gj4+vqiTZs2fLepX79+uH79Oi5duoTs7Gxs374d6enp9OMGBgZo1aoVli1bhvv37yMjIwPr1q1DYmIiPSBBEARBEARBEATRWJBl9oTISUtLw83NDZaWlggICKD/27hxIwoLC9GlSxe4ublh2rRpXN9nYmKCf//9F0ZGRlzH161bhz///BPbt2/Hhw8f0KFDByxbtgwLFizgu01z585FdnY2HB0dweFwMHnyZMydOxd3794FUDNocPjwYfj4+GD+/PmoqqpCz549ERoaSjrzBEEQBEEQBEE0OqQzTzB26NChBh/T0dFBWloa/bWTkxOcnJy+G8/Q0JAuZ1eXhIQEli5diqVLl9b7fYMGDcKzZ8+4jnXu3JnrmIyMDLy9vb/787t06YIdO3Z89zkEQRAEQRAEQRCNAVlmTxAEQRAEQRAEQRBNDOnMEwRBEARBEARBEEQTQzrzBEEQBEEQBEEQBNHEkM48QRAEQRAEQRAEQTQxpDNPEARBEARBEARBEE0Mh6Io6lc3giAIgiAIgiAIgiAI/pGZeYIgCIIgCIIgCIJoYkhnniAIgiAIgiAIgiCaGNKZJwiCIAiCIAiCIIgmhnTmCYIgCIIgCIIgCKKJIZ15giAIgiAIgiAIgmhiSGeeIAiCIAiCIAiCIJoY0pknCIIgCIIgCIIgiCaGdOYJgiAIgiAIgiAIookhnXmCIAiCIAiCIAiCaGJIZ54gCIIgCIIgCIIgmhjSmScIgiAIgiAIgiCIJoZ05gmCIAiCIAiCIAiiiSGdeYIgCIIgCIIgCIJoYkhnniAIgiAIgiAIgiCaGIlf3QCCIAiC+F9WUFCAP//881c3o15fvnxBeno6KioqQFEU12MDBw78Ra0iCG5sXaefP39GXFwcnj17BgkJCXTv3h2TJ0+GjIwM45ivXr2CkpIS4+//WTEJgmgaONS373oEQbCmoKAAYWFhWLlyJSQlJTF16lR8/vyZfnzIkCFwc3P7hS0EgoKC+H7usmXLWGzJ/6YvX74gICAATk5OQsUpLy+HlJSU0G05cOAA7ty5U+9N8sGDBwWOWV1djYSEhAZjenp6/tZtLSoqgq+vL8zNzaGqqgobGxvcvHkTSkpK2LNnD7p06SJwTLZcuXIFjo6OKCkp4fndORwOnj59KnBMNs5TU1NVVQVxcXGRxszPz0dAQECDr+s///wjcEy2zlV8fDwGDhyIDh06YNeuXThz5gz69esHZ2dnSEtLCxyPjesUAN69ewdzc3Pk5+ejW7duqKqqQlZWFuTl5XH06FF06NCBUdxhw4Zh165d0NTUZPT9PyvmzyCKzym2Y+bk5CAjIwMDBw5EaWkp5OXlRRabIESBzMwTxE/y4cMHTJs2DZKSkjAzM4OCggLevHmDadOmoXXr1sjJycHJkydhZGSE/v378x13165dmD59Otq1ayeSdkZFRXF9/e7dO0hKSqJLly6QkJDA69evUVFRAQ0NDaE682/fvsWDBw9QXl7O85iRkRHjuGxKSUnBhw8foKKigh49egj8/WVlZfD19UVCQgLExcVhaGgIBwcHiInV7Hi6du0aXFxc8P79e8ad+YiICOzduxfv37/HuXPnsG/fPrRt25bRuXJ1dcWZM2cwYsQItG3bllF7vuXt7Y2DBw9CXV1dqNmtbzWVtnp6eiIlJQVz587FpUuXcOfOHfj4+OD06dPw8fHBjh07RPJzRMHPzw8DBgzAypUrISsrK5KYbJynpmbEiBEwNDSEiYkJVFVVRRLTxcUFKSkpMDIyatTnateuXdi9ezcOHDiAd+/eYfv27ZgxYwZu3rwJPz8/ODs7CxyTjesUALy8vKCgoIDIyEh65UxeXh5WrlwJX19f+Pv7M4orJSUFCQnR3n6zERMAbt++DW1tbZ7YZWVluHLlCiZMmMAorig/p9iKWV5ejjVr1uDs2bMQExPDuXPn4O3tjeLiYgQFBYn0WiMIYZDOPEH8JCEhIejUqRMOHDiAZs2a0cetrKzo2bjc3FwcP35coM789u3bcfToUXh7e2Po0KFCt/PSpUv0v8PDw3H58mX4+/vTo9FFRUX466+/oKamxvhnnDp1Ci4uLqiqquJ5jMPhCNyZz8jIQFBQEDw9PdGsWTNoa2vj69ev9OO6uroICwvjO155eTn8/Pxw48YNDBs2DEuXLsX8+fNx//59uo0TJkyAr68vJCUl+Y7r7++PY8eOwcDAAFJSUjh27BhkZWWxcOFCuLu7IyIiAl27dkV4eDjfMeuKj4+Hv78/rKyssG/fPgCAiooK/Pz8IC0tDVtbW4HiXbhwAd7e3pg0aRKj9tQnNjYW69evh5mZmchiAk2nrYmJidi5cydUVFQQFhaGoUOHYurUqVBTU4O5ublQsdXV1cHhcOp9TFJSEh06dIChoSGWLFnS4PPqysrKQmBgoMg6nAA756nWq1ev4OrqSs8kf4vfGdrvvY5MY9a1bNkyxMbGIiwsDH369MG0adOgr68vVOcgOTkZO3fuFMlnQC02ztWpU6fg7e2Nfv36wdvbG1paWnBzc0NKSgrs7e0ZdebZuE6Bmtd0//79XFtg2rRpgzVr1gj8XlqXgYEB5s+fD0NDQygqKnLdDwDMBrPZiAkAlpaWSE5O5tkG9OLFCzg6OjLqzIv6c4qtmMHBwUhLS0N4eDgWLVoEoOb1cHJygq+vLzZv3sx3LEtLS76fK8zqpA8fPuDEiRN4+fIlnJ2dcevWLaipqUFFRYVxTKLxI515gvhJrl69ig0bNvB8yNZlZmYGd3d3gWOPHTsWtra2mDhxIuzt7UW2VHfPnj0IDQ3lWlbWsmVLrFq1ChYWFli1ahWjuMHBwTAxMcGaNWuEnvHMzMzEzJkzoaqqiqKiIvr1Xb16NeTl5ZGTk4MdO3YgMTERI0eO5Cvm9u3b8ffff2PSpEk4d+4cbt68iS9fvuDEiRPo3r077t+/j/Xr1yMwMBCOjo58t/XixYtwdnbGnDlzAACjRo2Ch4cH3r17h5MnT2LevHlYuXIl4yWCYWFhcHZ2hrGxMT14YWlpCVlZWQQHBwt8QyMmJoZevXoxaktDysrKMHz4cJHGBJpOWz9//gwFBQUAwPXr12FtbQ0A+OOPP+od3BLEunXrsHXrVpiamtIDgg8ePMDhw4cxe/ZstGrVCgcPHoSUlBRf14KSkhIKCgqEatO32DhPtTZu3IicnBw4ODgI1THesmUL3515JubMmYM5c+bg1atXiI6Oxt69e+Hp6Qk9PT1MmzYNQ4YMEThm8+bN6etKVNg4Vx8+fIC2tjaAmut/3LhxAAAFBQUUFRUxisnGdQoA4uLi9X5eS0tL17uijF+7d+8GAOzfv5/nMSaD2aKOeeDAAXh7ewMAKIpqcICI6ZJ+UX9OsRXz9OnT2LRpEwYNGkQf09HRgZubGxwdHQXqzHfq1Engny+orKwszJw5EzIyMsjNzYW9vT3Onj0LJycnhIaGol+/fqy3gfhFKIIgfgoNDQ3q3bt3XMcWLlxIffjwgf76zZs3lIaGhkBxe/ToQeXl5VEpKSmUsbEx1bt3b2r16tXUrVu3hG7zgAEDqDt37vAcv379OqWjo8M4bp8+faiXL18K0zTa2rVrKVtbW6q6upo+pqWlRb1+/Zr+2s7Ojlq+fDnfMUeOHEldv36doiiKun//PqWurk4lJSVxPScxMZEaNmyYQG3V0NDgaldlZSXVs2dPavDgwdS///4rUKz69O3bl45f9zV4/fo11adPH4HjOTk5UQEBAUK3q67ly5dThw8fFmlMimo6bTUxMaEiIiKoK1euUD169KAyMzMpiqIoX19favr06ULFnjFjBnXs2DGe41FRUdTs2bMpiqKoixcvUuPHj+crXmJiIjVlyhTqn3/+oTIzM6m3b99y/ccEG+epVp8+fai7d++yEptN5eXlVEhICNW3b19KXV2dGj16NBUWFkZVVlbyHcPb25tycnIS6Ht+hI1zpaenR12/fp3KysqievToQZ+vmJgYvq/Lb7FxnVIURS1atIiys7OjysvL6WPl5eXUypUrKWtra8ZxG7uKigoqOjqaOnXqFNWjRw8qPDycioqKov+Ljo6mzp07RxUWFjKKL+rPKbZiampqUtnZ2TwxMzMzGcdk06JFiygnJyequrqabm9lZSXl4OBAmZmZ/ermESwiM/ME8ZPIyMigtLSU61jtaHqt4uJitGrVilH8/v3749SpUzh//jwOHDgACwsLtGvXDgMHDkSPHj3QunVrzJw5U6CYY8aMwYYNG+Di4gINDQ1QFIU7d+7Azc0NU6dOZdROoGYZa1ZWFrp168Y4Rq0bN27Ax8fnuzNpM2bMwJo1a/iO+enTJ3p1Q9++fSEtLQ05OTmu53Tt2pXnfP5IRUUFmjdvTn8tLi4OaWlprFu3jmv0n6k2bdrg5cuXPCsz7t69yyinQsuWLREWFobExEQoKyvzrBhgkgCuT58+8PHxwY0bN6CiosKzTYHp/sam0tYVK1Zg+fLlqKiowJQpU6CkpARPT08cOXIEO3fuFDheXWlpaRg8eDDP8f79+2Pjxo0AgF69euHdu3d8xVuwYAEA8CzLpyiKcWIxNs5TLTk5ObRo0YLx9zfk0qVLePbsGdfKifLycjx48IDxlhigZtVETEwMzpw5g/LycowbNw4mJibIzc3Ftm3b8OjRI2zdupWvWHl5eTh79iwuX76Mrl278ryuTJbusnGuZs+ejZUrV0JaWho9evSAtrY2jhw5Al9fXyxfvlzgeAA71ykAODg4YPbs2Rg3bhw0NDTA4XDw8OFDlJSU4NChQ4xi1vXy5Us8e/YMkpKSUFFREcnnoShiSkhI0DP5HA4H+vr6Ik0oJ+rPKbZiqqio4Pr16zz3TQkJCUJt6SgpKUFqaioKCgogJycHDQ0Nkey/v3fvHg4fPsz1NyAuLo5FixYJfO9HNC2kM08QP4mqqiqSkpK+u3cpMTFR4GWNdd+4a/dyT5gwAa9fv8bFixdx9+5dnDhxAvn5+QK/oW/YsAErV66ElZUV/XMoisLEiRPx119/CRSrrnnz5sHV1RXZ2dn13iQKUkooLy8PXbt25TpmYmLCtXxfSUkJnz594jummpoaYmJi6M7a3bt3UV1dTT9OURTCwsKgoaHBd8zv6du3r0jizJo1C66urli7di2Amhu7pKQkbNu2DXPnzhU4XmpqKt22Dx8+iKSNERERkJeXx5MnT/DkyROuxzgcDuPOfFNp68iRI5GYmIjc3Fyoq6sDACZPnoyZM2cKva+xc+fOuHz5Ms+5vnTpEp15+/Xr13yXwWMjszwb56mWhYUFtm7dCl9fX5ElpwoICEBISAjatWuH//77D+3bt0deXh6qqqqgr6/PKOauXbsQGxuL169fQ0NDA/b29pgyZQrXe5a4uDhcXFz4jikuLo4pU6Ywak9D2DhXNjY26NatG7Kzs2FgYAAAkJWVhbOzM2bMmMEoJlsVEFRUVBAbG4sjR47g+fPnoCgKU6ZMwezZs4XaylZeXg4HBwecP3+ePsbhcDB69GgEBgYy6jizERMAjI2N6X3YmZmZcHJyEnoftqg/p9iKuXz5ctjZ2SE9PR1VVVWIjo7Gy5cvcf78eQQEBAgc78uXL/Dx8UFUVBTKysro49LS0jA0NMTatWu5BvsFVVVVxXWfUqukpETk1TOIRuZXLgsgiP8lp06donR0dKinT5/W+3haWhrVv39/6u+//xYobu0yeza9fPmS+vvvv6m///6ba5k4Uz169GjwP3V1dYFiDR06tMHXtNbDhw+pESNG8B0zOTmZ0tDQoDZu3Mjz2K1bt6gxY8ZQ/fr1ox49eiRQW3v06EHl5+dzHft2S4Cw/P39KU1NTfr17N27N+Xm5kZVVVWJ7GcQjVNcXBzVq1cvatWqVdShQ4eo8PBwatWqVVTv3r2pkydPUi9fvqRGjx5NeXh4CBw7Pz+f8bLan8Xc3JzS0NCgevbsSQ0bNowaM2YM139MjBw5kt5qMXLkSOrt27fUx48fKVNTUyowMJBRTF1dXcrT05N6/vx5g895/PgxK9tRGpOysjKRxxTldbpo0SLqxYsXIolVl5eXFzVy5Ejq0qVLVHFxMfXp0yfqwoUL1MiRIyk/P79GE5OiKOrVq1eUjo4ONWbMGKp3797U69evqeXLl1NaWlr1bsHjFxufU2zETExMpExNTSktLS1KU1OTmjZtmsD3aBRVs3Vh9uzZlI6ODrVr1y7q3r171KtXr6gHDx5QISEh1JAhQ6gZM2ZQFRUVjNtqZ2dH2dnZUZWVlfR9RUFBAWVmZibQNkOi6SF15gniJ1q0aBGuXbsGIyMj6Orq4s8//8THjx9x+/ZtxMTEYPTo0Xwvq6y1bt06ODs7i7TM17du376NjIwMTJkyBe/fv4eioqJAWdy/9fbt2+8+LkiymAULFkBTU/O7s6R+fn54/fo1tm/fznfctLQ0vHv3DqNHj+Y6/vjxY8TGxsLMzAyKiop8xwNqthdMnjyZq5ZyfHw8xowZw7M8WJjlxl++fMGLFy9AURSUlZWFujY+f/6MuLg4PHv2DBISEujevTsmT54s9PWWlJTEFXPw4MFCzx401raOGTOG74RqTOqB13X58mWEhYXh8ePHkJCQQI8ePbBgwQIMHz4ct2/fxrVr17Bs2TK+/36PHDmC4OBg5OfnA6hZzmpjY8N4tgtg7zwFBQV993EmKyk0NDTw999/o3Pnzli0aBGMjIwwceJEpKSkwNnZGefOnRM4ZmVlJStlxN6/f48jR45wva6zZs1Cx44dGcdk41yxUZaMjet04MCBiI6ORufOnRnHqM/w4cPh7u7Ok5D18uXLcHV1xZUrVxpFTABYvHgx/vzzT7i7u6Nfv36Ii4tDx44dsXbtWrx79w6HDx8WOOatW7egra2NyspKkX1OsRFzz549MDQ0RPv27RnHqHXw4EGEhYXh+PHj9cbLzc3FnDlzYGlpyfiazc3NhaWlJT59+oTi4mIoKyvj7du3aN26NQ4fPvxTkvARvwbpzBPET1RdXY2wsDAcPXoUOTk59PG2bdvCwsICtra2rGZRFlRJSQlsbGzw4MEDcDgcnD9/Hh4eHnj16hUOHDhAL91lqry8HG/evEHXrl1BURSjAYKLFy/CwcEBgYGBGDVqFM/j165dw5IlSxASEgJdXV2h2issCwsLvp/LdE9mTk4OWrZsCRkZGfz77784f/48+vXrx2gJ7rt372Bubo78/Hx069YNVVVVyMrKgry8PI4ePcro/BcVFWHevHlITU1Fy5YtUV1djZKSEvTu3Rv79+9Hy5YtBY7Z2Nu6Y8cO+u+6sLAQR44cwejRo+n6zY8ePcL58+cxb9482NvbC9xOtkRGRmLz5s0wNzfHgAEDUF1djdu3b+PYsWNwcXHB9OnTBY7Jxnli09ChQxEWFoYePXrA09MTsrKyWLZsGXJycjB58mS6XKUgKioqEBoaikmTJkFRURHOzs44c+YM+vXrBz8/P578HPxIT0+Hubk5mjVrBk1NTVRVVSE1NRVfvnxBREQEunfvLnBMNs5VfHw8XF1d6RJiCQkJuHz5Mvz8/LB8+XJGWcfZuE6BmqoGHz58wNKlS6GoqCiyfePa2tqIjo6GkpIS1/FXr17BwMAADx8+bBQxAWDw4ME4fPgwVFVVoa2tjbi4OHTp0gUZGRmYOXMm7ty5wyhmaGgoevfuzahNPytm//79ERUVJfCgfX1MTExgbm4OExOTBp8TExODgwcPIioqivHP+fLlCxISEvD06VNUV1eje/fuMDQ0ZHWyh2gEft2iAIL43/b69Wt6qVVjXQLt6upKzZo1i3r9+jW9bCsjI4MyNjam7O3tGcetrq6mfH19KU1NTapnz57U69evKXt7e2rt2rVcmYMFaWePHj0oKysrKiQkhDpx4gS1Z88eytbWlurZsyfl5eXFuK1Nyfnz56nevXtTSUlJ1OvXr6nevXtTEydOpPr27ctoye6KFSsoMzMzrq0B//33H2VqakqtWrWKURudnJyoKVOmUM+ePaOPPX36lJoyZQrl4uLCKGZTauuSJUuoPXv28Bw/ePAgZWlpyShmXSkpKVRISAgVFBRE7dixg+s/QU2YMIE6cuQIz/HDhw9TU6ZMYdQ+Ns5TXampqZS9vT01efJkysDAgFq1ahX14MEDxvFWr15NzZs3j8rJyaHi4uKoyZMnU/n5+VRYWBg1evRoRjE9PDwoHR0dKjU1lbp27RrVs2dPKjg4mJo5cya1du1aRjFtbGyo5cuXcy1d//r1K7V8+XJqwYIFjGKyca6MjIyoqKgoiqK4txhFRUVR48aNYxSTjeuUoihq9OjR9Nav+v5jaubMmVRwcDDP8Z07d1IGBgaNJiZF1VS0qX3/q3u+7t+/Tw0cOJBRzMmTJ1M3btxg3KafFXPevHlUWFiYSGJpa2tTWVlZ331OdnY2pa2tLZKfR/xvIQnwCOIX6dKli8jqwbPl8uXL8Pf352qnsrIyNm7ciEWLFjGOe+jQIcTGxmLjxo10rVY9PT24urpCXl4eDg4OAsVzcXGBjo4ODh48iMDAQFRXV4PD4aBPnz7w8/PD5MmTGbe1Kdm1axdsbGwwZMgQ7N27Fx07dsTp06dx9uxZBAUFwczMTKB4ycnJ2L9/P1fCtDZt2mDNmjWMZtCAmmXkO3bsgJqaGn1MXV0dGzZsgL29PVxdXRnFbSptTU5Orjd55IgRI+Dn58eonbX27NmDrVu3olWrVjzbNpgk7MvJycGwYcN4jg8fPpyuQy0oNs5TrZSUFFhbW0NNTQ3Dhg1DVVUV7t69C1NTU4SHh6N///4Cx3R0dMTChQtx7tw5mJqaYv/+/XTd7dpkW4L6+++/sXXrVvTu3RubN2+Gjo4OFi1ahKFDh9KZ2QV1584dHD9+nGv2WFpaGkuWLIG5uTmjmGycq8zMTAwYMIDn+IABA/D+/XtGMdm4TgEwzq7/I4sXL8aSJUuQlpaGfv36gcPhICUlBRcuXGD8HsBGTAAYNmwYgoODuWJ8/PgRvr6+9VbO4DfmwoULMXLkSCgqKnJtOwOYbYdhI2bz5s3h4+OD3bt3Q0lJiSemIIkXxcTEUFFR8d3nlJWVCbX6w9LS8ruPs5Uokvj1SGeeIJo4NvfjFhQUoG3btjzHZWRk8OXLF4Fi1XX8+HG4uLhg3LhxcHNzA1CT0VtKSgoeHh4Cd+Zzc3ORl5cHIyMjbN26FeLi4mjdujXjff0/c4+zKGVkZCAoKAhiYmK4du0aRo4cCTExMWhra/8wT0F9xMXF0axZM57j0tLSKC8vZ9TGysrKerOpy8vLo6SkhFFMoOm0tV27drh+/TrP0s2LFy8Kvafx8OHDWLx4MVauXClUnFodO3ZEamoqT7WIhw8fok2bNoxisnGeam3duhUzZszgyQLv6uqKwMBARltX2rdvj5iYGPpG++jRo7h69So6dOgATU1NRu389OkTnQk8OTmZXgYuJyeHr1+/MorZokWLel8/YV5TNs4VGyXE2LhOgZpM7mwYNWoUtm/fjj179uDKlSugKApqamrYunUrJk6c2GhiAjUDVpaWlhgyZAjKysqwePFieh8204GSCxcuQF5eHqmpqUhNTeV6jGmVEDZiysjI0CX6hNWzZ09cvnz5uxUA/vnnH4GrGdX17edHRUUFXr9+jfT0dKFyRxCNH+nME0QTt3LlSqxfvx6qqqoYO3asSGP36dMHZ86cwcKFC7mOHzx4UKgPnTdv3qBnz548x3v06IG8vDyBYqWkpMDW1pYeXGjRogW2bdtW70wNv9h8TdnUsmVLFBcXo6SkBPfv38e8efMA1JQja926tcDx+vXrh127dsHHx4ceGKmoqEBwcDC0tbUZtbF3796IiIjA+vXruY4fPXq03mvid2urjY0N3NzccP/+ffTp0wcUReHOnTtCz6ABNfvxRXXzCdTUBHd1dcWnT5+4Zvu2b98uUP6Hutg4T7UeP34Md3d3nuPm5uaM903Xqp2Va9asGcaPHy9UrK5du+LRo0coKChAVlYWhg8fDqBmQIdpsrXBgwfDx8cH27dvp//WCwoK4Ofnx3gGlY1zxUYJMTau01qJiYkIDQ3Fy5cvcfz4cZw6dQpdu3YV6u/s9u3bGDVqFPT09LiOl5WV4dy5c5gwYUKjiAn832BW3X3Yc+bMEWof9qVLlxh938+OKUwS2m/Nnj0bGzZswMCBA+stRfvvv//yrIAQVEPt3b59O50Ykvg9kc48QTRxhoaGkJSUxF9//QVPT0+6drUorFq1CtbW1rh37x4qKysRHByMFy9e4MmTJwgNDWUct1OnTnj48CHPjWtiYqLAWw+2b9+OwYMHw9XVFeLi4ti8eTO8vLyQkJDAuH1svqZsGjlyJFxcXCAjIwMZGRkMHToU169fx6ZNm+pNDvgjDg4OmD17NsaNGwcNDQ1wOBw8fPgQJSUljBP02dnZwdLSEg8ePOC68U5LS8PevXsZxWxKbZ01axZkZGRw6NAhnD9/HhwOBz179sSuXbt4MlELqn///nj06JFIEjYBNcs23759iy1btqCqqgoURUFCQgIzZ87EkiVLGMVk4zzVkpOTQ35+PpSVlbmO5+fnM16+qq6u/t1VOk+fPhU45vz587Fq1SqIiYlh8ODBUFdXx86dO7Fz505s2bKFUTtrX9fRo0dDSUkJHA4HmZmZaNmyJaOM43VjivJc2draori4GI6OjigrK8PChQshISGB2bNn8wwa84uN6xSoWTWxbNky6Ovr4/79+6iurkZVVRWcnJxQVVWFadOmMW5vcnIyz6qfFy9ewNHRkVHHm42Ytf744w/MmDGD8fc3hI2KJqKOKaoKEfr6+vj3339hZmaGkSNHol+/fmjdujWKi4tx69YtJCYmwtTUlJXJA2NjY0ybNo3xFjai8SPZ7AniN+Hg4IDi4mKEhISING5aWhrCwsLw5MkTOjvqvHnz6h1d5tepU6fg4+ODRYsWYdu2bXByckJWVhYOHTqEdevWYc6cOXzHGjhwICIiIqCqqgqgZsn9qFGjcPv2baEzuLL1mrLl69evCAwMRHZ2NmxtbaGlpYUdO3YgKysLrq6uPPuo+fH27VscPXoUz58/p5duzp49W6h8Dw8fPkRYWBhXzLlz50JLS4txzKbWVjZERkbC19cXJiYmUFZW5unAMp1NLCkpwcuXLwFA6HJPADvnCahZTn/nzh0EBATQy1lfvHiB1atXo2fPnvDy8hI4ZlRUFFdnvrKyEq9evUJ0dDTWrl0LAwMDRm1NS0vDmzdvMGLECEhJSeHq1auQkJDAkCFDGMUDgNLSUsTGxnK9rlOnToWsrCzjmGydK1GWz6wl6ut09uzZmDhxIubOncuVyT00NBTR0dECDRgfOHCAXpZOUVSDA0Sampo4fvz4L4v5rYa2nHE4HEhKSqJDhw4wNDQU6L2FjYombMRko0JEbcb6tLQ0VFdXA6hZAWltbc1abp/Lly/jr7/+wu3bt1mJT/x6pDNPEL+JkpISvHnzRqSzyDExMfRe9ro+f/6MEydOCLUP6/jx4wgODqaTHsnLy2P+/PmwtrYWKE6vXr1w9epVrr2RmpqaOHv2rNB7kNl4TQni8ePHCA0NpWd7VFVVYWVlxXgPdq3vXaccDoevWeTbt2/TJfN+dPM3cOBAgdvIpsLCQlhbW+Pp06eQlZUFh8NBUVER1NTUeBK5CSs+Ph6xsbHYt2+fyGICNQNy9e1T/52UlJTgzJkzSE9Ph5iYGHr37o2JEyfyJBj7np9xnWprayM2NhZdu3bl6sxnZ2djypQpePDgAd+xKisrkZCQgOrqajg5OcHJyYlrkIXD4aB58+YYPHgw3x1PNmJ+q3bFiJ6eHp1A8sGDB/j7779hYmICMTExxMfHw8nJie/Ze2dnZzx8+BD+/v50ctG0tDQ4OjqiX79+jGaQ2Yg5f/58NG/eHH5+fvQ9UFlZGb2qRJhB/rKyMhQWFqJ169Z07C9fviAgIABOTk6MYq5bt47nWHFxMZKTkzFx4kSRbhsgGhfSmScIgktBQQGdhGns2LE4efIkT93jJ0+eYNWqVYxr13778yiKgry8PKPvV1dXR3JyMtf3173x+t0FBQXBxsYGf/zxB4KCgr77XH6SAFlaWiIoKAgtW7aEhYXFd5cY85sdd926dXB2doaMjEy9Nxx1CXLD0ZTaWqtuxvUBAwbQGdfT09MZZ1wXpbp/T7VLzOu7TeB3cABg5zw1pLq6GklJSVwzycOGDRN6+e633r59Sy+/FlRhYSGCg4Px7NkzVFVVAaiZWa2oqMDz58/5rt1d9/35R0k7+U3Uyfa5ysjIgJWVFUpLS6GkpITq6mpkZWWhffv2CA8P57t2PRvX6bdGjBgBf39/DBw4kOsz5dKlS9i0aROuXr3KKG50dDT09fUBgO7IvXv3DgoKCozisRUTAObNm4dBgwbxbIE4cOAAkpOTsXfvXsTExCA0NBTx8fF8xRw8eDB27NjBM8hy69Yt2NvbIzk5WeB2shFTW1sbx48f56pmAtQMEpibmyMlJYXvWGVlZfD19UVCQgLExcVhaGgIBwcHiImJAQCuXbsGFxcXvH//Hk+ePBG4rQDqzQ8hJSUFbW1tWFtbM1qZRzQNZM88QfymNm3ahBUrVgg8G3X16lWsXbuWvjmqL3EURVFC7+99+/YtTpw4gWfPnkFcXBy9e/fGzJkzhco+zDamrymboqKiYGZmhj/++ANRUVENPo/fjL6dOnWibzA6derEd1b/73nz5g29pPDNmzdCx6vVlNpai42M66L0zz//0IN3oqrUwMZ5aoiYmBhGjhwp9PvTj5w+fRqtWrVi9L2bN29GcnIyhg0bhjNnzkBfXx8ZGRn0ICm/jI2N6Vl8ExMTRm35Ftvnyt3dHT179oSfnx/9+hUUFMDBwQHu7u4/HJCsxcZ1+q2pU6fCw8MDHh4e4HA4KC0tRWJiItzc3IRaEj1ixAhYW1ujf//+9Pk2MjJCz549sW3bNkbXFRsxgZoqAxs3buQ5PmbMGAQEBAAAdHR06n1OQ9ioEsJGTFFWiPD398exY8dgYGAAKSkpHDt2DLKysli4cCHc3d0RERGBrl27Ijw8nFFbAfzyzw7i1yEz8wTxm+rXrx9iY2MZzU7fvn0b1dXVsLKywo4dO7huBGqX7qmpqTEu/Xbv3j1YW1tDTk4OvXv3RnV1NR4/foyvX7/i8OHDAu1FU1dXp2ema4WEhGD27Nk8NzBMytPUJcxr2hTl5OSgQ4cO9M19rcrKSjx58kToZeHfysvLYzyY01Ta2rdvX0RHR/MkacvIyMD06dNx7949geL17NkT165d45qhbIigM5R1VynU9enTJzg7O2Pnzp0CxQNEf57Y/P0B3j3DFEWhtLQURUVFsLe3Z1QXftCgQfDx8cHIkSMxZcoU+Pn5QV1dHRs2bEBZWRl8fHwEjnn79m1oaWnxvCeXlZXhypUrjBKgsfE3pa2tjRMnTvC8x6elpcHU1BR3794VOCYb1ylQk7l/7dq1OH36NADQA9yjRo3Ctm3bBNoWUNeqVavohH21uR2ePHkCV1dXqKqqwsPDo1HEBGpWfyxbtoynTF9MTAwCAwNx5coVPHjwAEuXLsW1a9f4imllZYXu3bvzVAlxc3PD48ePcezYMYHbyUZMBwcHfPjwgadChJ2dHVq2bMn3wBNQ8z5ia2tL5wO6cuUKPDw8oKuri5MnT8La2horV64Uqs48ULNUPz09HRUVFTwrVRrbtihCdMjMPEH8poQZp6t90z948CD69esHCQnRvlV4eXlh0qRJcHNzo2NXVFRg3bp18PDwwIEDB/iO1bFjR5w9e5brWNu2bXlma5jWmq2rqYx9FhQUoKysjKe9gmbgHTt2bL0Zkt+8eQMLCwuB9ozW6tmzZ4Mxp06dKnBntqm1VdQZ17ds2ULvk92yZYvQM6l37txBdnY2gJob9t69e/N0kjIyMnD9+nVG8UV9nur+/mzsCa1vxltSUhL9+vVjfHNcWlpKL91VUVFBWloa1NXVYW5uzmhwAGAnmzkbf1Pt2rXD+/fveTrzJSUlAu3rZvs6BWrOs7+/P1asWEGXZVNTU6OTrTKVnJyM8PBwrprjvXr1woYNG2Bra9toYgI119XmzZvx6tUraGlpobq6Gg8ePMChQ4ewePFivH//Hps2baLLK/KDjSohbMRcvXo15syZI5IKEf/99x9Xudzhw4fj7du3uHDhAvbv349BgwYxamNdV65cgaOjI0pKSng++4XZbkI0fqQzTxBEg3R0dJCWlob09HR66TFFUSgvL8eDBw8Yl1FKS0uDp6cn1yCBpKQkFi9eLHA9aDbqyzZVDx8+hJ2dHd69e8d1vDbTMT8f5keOHEFYWBj9fdOmTeOZmSsqKhJoYODkyZOIi4ujYy5dupRnBvHDhw8CJ2lqSm2tNXr0aLi5ufFkXPfw8MDo0aMFjld3xkwUS605HA5dA5zD4dRbt7158+awsbHhOyYb56lW3d+fw+F8N2EnE4MGDWpwxptp/W4FBQW8ffsWCgoKUFJSQlpaGoCaEmCFhYV8x/k2m/nQoUPrfZ4gM+hsnisAWLNmDV1nXkdHBxISEnj06BFcXV1hZWWFnJwc+rnf+xlsXKcNUVRUFFm5RwCoqqqiP0/rkpCQQFlZWaOJCdTMeEtISCA0NJRO+NaxY0esWbMGs2fPRlJSElRUVH6YX6QubW1tHDlyBPv378e1a9fo3Bbr169nXCVEVDGfPHmCHj16QFxcHAoKCjh9+jRXhYjp06czqhBRUVGB5s2b01+Li4tDWloa69atE0lHHgD8/PwwYMAArFy5UqgKFkTTQ5bZE8RvKiQkBHPmzGHc6QBqZuZrO+x1EwxxOBwMGDCA8R4tQ0NDLFy4kGffYWJiIjw9PfH3338zbjObRPGasmnatGmQkpKCjY1NvW3U0dH5YYwvX74gNDQUFEVh586d9SbOadGiBcaPH893tYBPnz7RnY7o6GhMmjSJJ2N3ixYtYGRkBA0NDb5iNrW21mI74/rZs2cRHh6O9PR0iIuLo1evXrC1teWaFeKXuro6rl27JnQeCzbOUy1+EnY+ffoU9vb2jBJ2NrQ64/Hjx5gzZw6jmP7+/jh9+jR8fHxQUVEBOzs7uLq64uLFi3j+/DliY2P5isNGNnM2zxXAXXHh2+0LtccEGXysjfltElSmfpRIsC6me/WXLl1KlxGtPV8lJSX466+/UFVVxShLOhsxv/Xp0ydISEgwKvk3cuRInDhxAu3bt+dK2ioMNmL2798fp0+fRocOHbiSQQqroUS9MTExIhso6tOnD6Kjo4VeOUI0PWRmniB+E1++fEGzZs3oG5GFCxfi+fPnkJCQ4BoRFsThw4excOFCLF26FKNHj0ZUVBQ+ffqE1atXY+zYsYzbamtrC3d3d/z3339cszMBAQGYM2cOV6mhxrTP69uMvo3N8+fPERUVJdSH+R9//EFvR+BwODw3SGVlZQLvFW3dujXXEuj69rey0VYm2GprrVatWuHkyZOsZFw/efIkXFxcMHHiREyePBnV1dW4e/cuFi5ciG3btkFPT0+geLUzxsJi45qqxUbCTjZmvOtavnw5vn79infv3mHq1KmYNGkS7OzsICsri+3bt/MdR0JCgq7vzeFwRJLNnM1zBQhfraA+aWlpyMjIwNu3b+lzEhYWhtGjR6Nbt24CxTI2NuYZZKjNwVK7b1pYa9euhampKUaMGEG379WrV2jdujVCQ0N/eUxB6pHz+/n88eNH5Obmon379ti5cydMTU2Ffq9mIyaHw8GVK1cwfPhw3Lp1C9nZ2TyDg7UEXZ1S3yDRt6tehKGkpISCggKRxSOaDjIzTxC/gZiYGHh6emLfvn3o06cPfdzGxgYPHjyAm5sbJk2aJHBcDQ0NnD17Fl26dIGNjQ3mzJkDPT09XLt2DV5eXkhISGDUXn7rtv/KfV4ZGRkICgqCp6cnmjVrBm1tbXoGEAB0dXXp5aiNxfjx47F161ZGM8b1+fLlCzZu3Ihu3bph8eLFAGpmQ4YPHw4XFxehk/XUVV5ejocPH2LAgAGMY1RWViI/P5+r3FftlpDaTo8gqqursWPHDrRt2xampqYAapayjxs3jn49GpMJEyZgzpw5mDt3Ltfxffv2IS4ujt4+wK/y8nIcP36cq4Ra7fFHjx7h/PnzAreRjWtK1Ak72a7fXV9iuU+fPkFGRoZxYrn8/HysWLGCK5v5oEGDhMpmzsa5Onz4MKZMmSKyjjEAJCUlYenSpZg3bx7s7OwAANOnT0dGRgb27t0r1HsKwE6p05KSEpw+fRrp6emQkJCAqqoqpk6dyrMK6FfErFvqr6HVE7X4/Xy2sbHB7du30aZNG+Tk5EBBQaHBjiy/Kx7YiOnh4YFDhw59d3WGoCtHgJrXdPLkyVwDYfHx8RgzZgzPyhemuT+uXr0KX19f2NvbQ1lZmefvk+nWGKLxIzPzBNHE3bhxA05OTjAxMeGZhXFxccHevXvh4OCAtm3bCnxT06JFC1RWVgKoGfV98eIF9PT0oKKigrdv3zJu84ULFwDULJGVk5MDh8MR6Qi1sDIzMzFz5kyoqqqiqKiIvhlavXo15OXlkZOTgx07diAxMZH1EliCWLx4Mdzd3bFlyxZ069ZN6GRonp6eePDgAWbOnEkfW79+Pfz8/BAQEIA1a9YIHPPp06dwdnbGs2fP6t3jyXTw5saNG3B0dER+fj7PY82aNWPUmQ8MDERkZCTc3NzoYwYGBtizZw/ExMQYrdTIysrC+vXrkZqayjU4VEuYwav3799j1KhRPMfHjRuHHTt2CBxvy5YtiIqKQu/evfHgwQNoa2sjKysL+fn5PAMG/GLjmhJ1wk42Zrzrqi+xXOvWrZGVlcU4sZyHhwcqKythaGhIH9u/fz9cXV3h4+PDKJs5G+dq79698PHxwejRozFt2jQMHz5c6PepgIAAzJ8/HytWrKCPnTx5EgEBAfDz82OUyZxtMjIymDVrFisxy8vLISkpyfh1rdvx/ffff7Fz5044OTnRf1sPHz6Ep6enQIn1AgICEBMTg6KiIgQFBWHixIlC1z1nI6azszOmTZuGoqIiWFpa8gwOMjVw4ED8999/XMe0tbXx8eNHfPz4Uej4AOjkmUuWLOEZhCEJ8H5vZGaeIJq4efPmQVVVFU5OTg0+x9nZGbm5udi3b59AsZcuXQoZGRm4uLjgzJkzOHHiBI4fP47Tp0/D29ub71I0tSiKQmhoKA4dOoQPHz7Qx9u0aQNzc3PY2to2ik79unXrkJ+fj5CQEPpD8dvZGXt7e1RVVQm0LJYN35bj+nY2pS5BP8yHDh2KnTt38iQQSklJgb29PZKSkgRur4WFBcrKyjB9+nS4u7tj7dq1eP36NY4cOQIfHx9GK0gAYMaMGZCXl4eFhQWWLVsGPz8/5OTkYPv27fD09BR4iTlQMwvp6emJIUOGcB1PTEyEq6sro+SLFhYWyMvLw6xZs+q9Sfy2BJQgFi1aBE1NTSxZsoTr+JEjR3DhwgWBqkQAwLBhw+Dk5ITJkydj/Pjx2L17N7p06QJ7e3t06NCBpwwUP9i4puoSdcJOUc14100sV5v8rr7Ecm3atOGpzsGPQYMGITw8nGfVU2pqKmxtbXHjxg2BY7JxriiKQnJyMmJiYnDx4kXIysrCyMgIxsbGPBUe+KWlpYX4+HiemfPs7GwYGBgwrpBRS9Qz82yseAGAiIgI7N27F+/fv8e5c+ewb98+tG3bVqgqLhMnTsTGjRuhq6vLdfzWrVtYu3Yto/fAhkoJCoONmNHR0dDX1xfpCjQ23bp167uP85Mzh2iayMw8QTRxT548obP6NmTOnDmMSh7Z2dnB2toaERERmDNnDoKDg6Gjo4MvX74wyhK8YsUKXLlyBYaGhtDV1YWcnBwKCwvx77//Ijg4GPfu3cPu3bsFjitqN27cgI+Pz3dnNmbMmMFoZkrURFGOrCGfP3+uNyuunJwciouLGcVMTU1FeHg4NDU1cerUKaipqcHU1BQdOnTAiRMnGHfmnz17hsjISPTo0QO9evVC8+bNYWFhgebNmyM0NJRRZ76wsLDeGVhFRUXk5eUxaufDhw9x5MgRkW2FqEtbWxu7du3C48ePMXDgQEhKSuLRo0dISEiAsbExV11kfm7wP336RHfk1NTU8OTJEygrK2PhwoWws7Nj1Jln45qq9aOEnUyIasbbxMQEHz9+pBPL1TeLWJtYjgk2spmzca44HA6GDRuGYcOGobS0FOfPn8f58+dhYmICdXV1zJgxA/r6+gItDf/zzz/x5MkTns728+fPG2WyUjZWvMTHx8Pf3x9WVlb0oL2Kigr8/PwgLS3NuDxdbm4u2rVrx3O8ZcuW+PTpE6OYnp6e+Pr1K2JiYpCRkQEbGxukp6dDVVWVcQJQNmIaGxsjLS0N4eHhyMzMxLZt23Dx4kWoqKhg8ODBjGKyqW5nvby8vMkMQhDCI515gmjiysvLf3jj06pVq3qX9P5I9+7dcfHiRXz+/BktWrRAZGQk4uPj0aFDB0ycOFGgWDExMbh58yYiIyN5Zo8mTZqEOXPmwMrKCqdOncK0adMEbqso5eXloWvXrlzHTExMuEb9lZSUGN/MiNK35cg+ffqET58+QUlJCQBw5swZeuBEUNra2ggJCYGnpyednI2iKISHh3PlZhBEdXU12rZtCwDo1q0b0tPTMWDAAIwdO1aojMvi4uL0+VFSUkJ6ejp0dXUxePBgOpmZoNTV1REZGYm//vqL63hsbCxPnWx+ycnJCZVA7HuOHz8OeXl5PH36lGsVRrt27ZCcnEx/zeFw+OrMt2nTBvn5+ejYsSO6du2K9PR0ADW/A9PBDDauqVpsJOwUVf1uNpI11jVo0CD4+/vzZDPfvn074ySibJ4roGawoLCwEB8/fkR5eTnExMQQEhKCrVu3ws/Pj2c2uCHGxsZwdXVFUVERNDU1weFw8OjRIwQGBgq10oUtFy9ehJeXF73ixc3NjV7xUlFRwShmWFgYnJ2dYWxsTK8AsbS0hKysLIKDgxl35jU1NREYGAgvLy968OnTp0/w9fVlPNObl5eH2bNnIy8vD+Xl5Zg5cybCwsLw6NEjhIeHM0rgykbM1NRUzJkzB1paWkhNTUV5eTmePn2KLVu2ICgoiFEpUbaxsTqDaPxIZ54gmrhu3brh3r17PJ3Puu7evcuojBBQs9+4drBAXl6enjkIDw+HlZUV33GOHz+OFStWNJj8Tl1dHStWrGgUnfnWrVvj06dP6NChA31sw4YNXM/Jz88XuoyYqD18+BC2trYwMTGhVw34+vqioqIC+/fvF7gDumrVKlhYWCAlJQW9e/cGh8PB48eP8enTJ8bJ/5SVlXH79m0YGBhAUVERjx49AgAUFxejvLycUUyg5vq5cOEC5s6di27duuHOnTuwsrLC+/fvGcdcvnw5bG1tcffuXWhpadGdhPv372Pnzp2MYpqbm9OdFWH3d36LyZLX7xk5ciQ2btwIT09P9OvXDx4eHhg3bhzOnDnD9bchCDauqVo5OTmYPn06pKSkoK6ujkePHkFPTw9r166Fl5cXo1lPNma8ly1bhi9fvuDBgweoqKjAt7sdmXS+2ciQzsa5Kisrw/nz5xEbG4sbN26gTZs2MDIygo+PD/0ZVluHPjExka+YS5YswcePH7F582ZUVlaCoihISEjAwsICy5cvF6h9lpaW9bbZwcGBZxCOaWZ+Nla8ZGZm1rv6ZMCAAUK9B65fvx5z587FiBEj6BJqmZmZkJeXZ/z7e3l5QVVVFfHx8fQWJm9vb6xatQo+Pj7Ys2dPo4jp6+uLefPmwd7eHtra2gAAd3d3yMrKNsrOPFurM4jGj3TmCaKJMzAwwPbt26Grq1vvcrgPHz5g27ZtAnWQ9+/fj4SEBIiLi8PQ0BBmZmb0Y8+fP4ezszMePXokUGf+xYsXDZZ4qjV8+HAEBgbyHZMtvXr1wsWLF7+bdf/cuXPo27fvT2zVj/n4+GD8+PGwt7enj124cAEuLi7w9PQU+AZcQ0MDCQkJOH78OJ0hecqUKTAzM6v3WuOHubk5nJ2dAdRk3zc0NESzZs3oDjNTtra2WLZsGaSkpKCvr4/t27djwYIFePbsGeMlkUOHDkVERAQOHjyI5ORkSEhIQEVFBSdPnuS7IgPAW7v67du3GDRoENq2bcuzb5pp7eq6kpKS8OzZM0hISKB79+4YPHgwo7J3Dg4OWLNmDVJSUmBqaooTJ05gxowZkJCQYLzaQUNDA/Hx8Thx4oTIrqlabCTsZGPG+8qVK3B0dERJSQlPR55poqouXbrg7NmzXNnM58yZI1SGdDbOla6uLiorKzFq1Cjs2rULw4cP5/kb0NXVFejvQFxcHC4uLli9ejUyMzMhISEBJSUlRr93fYPeTAfCG8LGipc2bdrg5cuXPFsN7t69K9TfVffu3XHu3DkkJCTg+fPnAABTU1Po6+szXlny77//Ys+ePVzf36pVKzg6OtY7mPKrYj5+/BibNm3iOT5nzpxGmVSRrdUZRONHOvME0cSZm5vj/Pnz0NfXx/Tp06GlpUXvZ7t//z6ioqKgqKjI9x73oKAgBAUFYdCgQZCWlqaXWM6ePRuhoaEIDAxE8+bNBS6fUllZyVeHgq3934KYOXMmHBwcoKGhUW928GvXruHgwYNCLQtnw+PHj+Hp6cm1V05CQgILFizgWY7Pr06dOtGJv2qVlZUxXkExbdo0tGrVCq1bt4aKigq8vb0REhICBQUFntUPghgzZgwiIyMhLi4OBQUFhIaGIiwsDGPHjuXKci2ovn37wt/fn+f4v//+y/cgwbe1q9lSVFSEefPmITU1FS1btkR1dTVKSkrQu3dv7N+/X+D9w7Kysti1axf99Z49e/DkyRO0adNGqA5C586dea4pURgwYAB2794NFxcXqKur48SJE1iwYAFSUlIYr4JgY8bbz88PAwYMwMqVK+vdk84UGxnSRXGuLC0tERQUhJYtW2LlypUwMDD47rafMWPG/DB3QG0pMg6Hg5ycHPp47WqpuvW2BSnJxbQsmCDYWPEya9YsekUDALx8+RJJSUnYtm0b4334tWRkZDB79myhYtRVWlra4EBA7WBcY4gpKSmJkpISnuM5OTki3SIjKmytziAaP9KZJ4gmTlxcHPv378f27dsRGRmJ/fv304+1adMGpqamWLx4Md+zFPHx8VixYgWdETsmJgZ79+7Ff//9RyducnFxEXiJuaqqKq5fv/7d7QBJSUlce1N/FT09PZiYmGDRokUYPHgwhgwZAjk5OXz69Am3b9/GtWvXYGVlxfeezp9FRkYGr1+/5pmdef/+vVD1i2tlZGTg2LFjiI2NRXFxMePtEHWT0enr69Olv4TVu3dv+t8DBw5kPHPakKKiIkRFReHYsWPIysriewa17lLf27dvQ0tLi6fmeVlZGa5cuSJU+7y9vVFWVoa4uDioqakBqMnu7ujoCH9/f7i6ugocs7CwEFlZWVxLyktLS5GVlcX361s30/S6deu++1xhOlPfS9g5b948RjHZmPHOyspCYGAgo328ddXtJP9o9pHfJdFsnKtbt27Re8H5Wc3FT2nBsWPH4tq1a5CXl+dZ+VKrsZbkYmPFi62tLYqLi+Ho6IiysjIsXLgQEhISmD17NqMSmrVEdV3VNXDgQBw5coRrO0FFRQV27tyJfv36CRyPrZh6enrw9/dHQEAAfSwjIwMeHh71DvL/amytziAaP9KZJ4jfgJSUFBwcHGBnZ4fs7GwUFhbizz//RJcuXQSeEXz//j1XRvHJkydj3bp1CA8Ph5eXF6N63UDN7OSOHTswdOjQekv8vHjxAkFBQTzJxn4VFxcX6Ojo4ODBgwgMDER1dTU4HA769OkDPz8/TJ48+Vc3kceECROwadMmuLq6ciWC2rx5M8aNG8coZmVlJc6fP4+IiAikpKSAoigMGjSIceeovLwckZGReP78eb17jpl25j59+oQ9e/Y0GJfp/k4AePDgASIiIvD333/j69evUFRUZLyKwNLSkqfGOFBz/Ts6OmLChAmM2/nPP/9gx44ddEceqMklsGHDBtjb2wvcmY+JicHGjRtRXl4u1HLwN2/e0PvO37x5I1AbBCHKhJ11iXrGW0lJiWvmmKlOnTrRS9RFtRT8Z50rYYWHh9NlAYX52/4V+F3xEhkZiUmTJvFVbu3WrVtYvnw5Fi9ejBcvXoCiKCgrKwtdqu3b66qiogKvX79Geno64xn/NWvWwMzMjB7k2bRpE16+fIni4mIcPny4UcWcP38+hgwZAoqiYGJigpKSEqirqzea+5S62FydQTRupM48QTRx/O4H43A4CA8P/+Hz1NXVkZycDHl5efqYtrY2HBwcuPbOC6q6uhqLFi3CvXv3YGJiAm1tbbRu3RolJSW4efMmTp48ieHDh//yuu21cnNzceHCBUhJSWHEiBEQFxdH69ateWZUG5MvX77Azs4OiYmJXIM448aNg6enp0BLjd+8eYPjx48jKioKBQUFkJWVRXFxMYKDg4WalVi9ejXOnz+PXr161Vs659ChQ4zi2tra4uHDhxg6dGi92eIFHST4/Pkz4uLicOzYMTx79owudebq6ooZM2YINEh24MABesatdrawPpqamjh+/LhA7axrwIABOH78OM/qloyMDJiYmODBgwcCxRs1ahRGjBiBuXPn1vuainovsbBiYmIwfvx4NG/enOt4SUkJPDw8GA0UsVET/OrVq/D19YW9vT2UlZV5/g4EWRbeFKirq+PUqVN8VdTg93cvKSkRaU3xxqhfv36IjY3lq7794MGDERoayrU6iU3bt29Hfn4+o9U+QE0un6NHj+Lp06eorq5G9+7dYWpqis6dOzNuExsxgZpStU+ePEF1dTXU1NTqzfPQWGzduhXh4eH0gHbt6gwnJ6dG22ZCeKQzTxBN3I+WQqakpCA7OxsyMjJISUn5YbyGOvNRUVH0nlGmqqqqEBwcjCNHjuDjx4/08TZt2sDKygo2NjaN4gMnJSUFtra2+PLlC4CaxFrbtm3DsGHDfnHL+JOZmYlnz55BUlISKioqdJk6fly6dAkRERG4du0amjVrhjFjxkBfXx/Dhg2DtrY2YmNjhVoe3L9/f3h7ezOq+/49tWW0mJZLqpWWloZjx44hLi4Onz9/hpaWFvT19TFx4kSMGjWK0e9fWVmJhIQEVFdXw8nJCU5OTlx7pTkcDpo3b47BgwcLVRfbysoK3bt358mG7ebmhsePHwuctElbWxvR0dECXT/18fT0xMqVK3k62aKmrq6OHj16ICQkhGvvcV5eHoYPH85oufWmTZu+WxO8tmqEoO2sVXdgR9Bl4cuWLYOXl5dIO7VsnCt1dfUfDn4J+rv37NmTXmZf6/bt2+jbt6/I6mt/b+DtZ9DW1kZcXBxfnXl9fX1s2LDhp9U/z87OxrRp03Dr1i2h4pSUlEBSUlKk5TpFHTM7Oxvp6emQkpKCqqoqFBQURBKXLV++fBHp6gyi8SPL7AmiiWtotqmkpAReXl7Izs7GkCFD4O7uLtTP4Wcf44+Ii4tj2bJlWLZsGTIzM/Hp0ye0bt0aioqKjaITX2v79u0YPHgwXF1dIS4ujs2bN8PLywsJCQm/uml86datG+OBlyVLlkBZWRl+fn4YO3asSPba19WqVSu6xJEotW/fXiSl3oyMjKCsrIzly5dj/PjxIpl9lpCQoLencDgc6Ovri6zDUZednR0sLS3x4MED9OvXDxwOBykpKUhLS8PevXsFjjd+/HgkJiYK3Zk/ePAgFixYwNVBtLGxgaenp8j3csrKymL69OnYuXOnSKpNsFETXFTLwv/55x+UlZVx3axPmjQJoaGhjGf32TpX27dvp5fFi0J981ALFy7keyabH9OmTcOWLVsEqlzxqwwbNgwLFy7EyJEjoaioyNORFXWd8drOIlPh4eHYv38/cnNzweFw0LlzZyxZsoTxNj42Yn758gUODg64dOkS/buKiYnByMgIGzduFOkAhCiMHj0axsbGMDY2Rp8+fX51c4ifiHTmCeI3lJycjA0bNqCoqAiurq4C7/cMCwvjytZaWVmJgwcP8tyMCXODIOwsP5uePn2KiIgI+ubVyckJo0aNavRLOzMzM7F582bcuXOn3o4GP7Ne+vr6+Oeff+Di4oLTp09jwoQJ0NPTE1lN9MWLF8PLywubNm0S2U03ULO/cfPmzbC3t0fnzp15Bof47dxoaWnh/v37iIyMRE5ODiZMmFBvhmCmjI2N8eHDB5w4cQKZmZlwcnLCrVu3oKamJnTyR21tbRw5cgT79+/HtWvXQFEU1NTUsH79ekZl/xwdHaGvr4/z58/Xm3+D32Xr9d303717l3Gd9oZwOBwEBARgz549sLS0hIeHB6ZMmSJUTDZqggu7eqRWfa/r+/fvubYDiCKmKM5Vv379uGbR2SDqhaZv375lfTWJqFy4cAHy8vJITU1Famoq12McDofxZ3V9K/+Ki4uRnJzMOA/FgQMHsG3bNlhaWqJv376orq5GSkoKNm3ahJKSEpibmzeKmJ6ensjIyEBoaCg0NTXpmO7u7vD29oaLi4vAMdk0c+ZMJCQkIDg4GNra2jAxMcHEiRMb9T0LIRqkM08Qv5HS0lJ4eXkhMjISurq68PDwEHiGpmPHjjh79izXsbZt2/LU/RXmBqGxKy0tRevWremv27dvD0lJSRQWFjbqD8ZNmzYhJycHDg4OjEte+fv7o6SkBPHx8YiOjsaaNWsgLS2N4cOHg6IooW+Y1dTU4Ofn12DpKWEyTz9//hzW1tZcxwRdvnvs2DG8evUKJ0+eRFxcHA4fPox27drRN67CLrvNysrCzJkzISMjg9zcXNjZ2eHs2bNwcnJCaGgo48zLtbp27YqVK1fSs+lnzpxhvBLC09MTpaWlKC8vZ1yn/WeiKApiYmJwdnaGkpIS1q5di4yMDFhYWDCOyUZNcDb24ROiZ2trC2dnZ9jY2KBr1648q5QaU26DS5cusRK3viSIUlJSsLGx4Xmv5df+/fvh5ubGNdCmp6cHFRUVBAcHM+p4sxHzzJkz2L17N9dg7pgxYyAtLY1Vq1Y1us784sWLsXjxYqSmpiI2NhaBgYFwd3fHuHHjYGxsjCFDhvzqJhIsIZ15gvhN1M7GFxYWYtOmTYzrwrJ1U9CU1Gaur0tcXJzO8txY3bt3D+Hh4dDW1hYqjoyMDObMmYM5c+bgxYsXOHXqFOLj41FdXQ0rKyvMnDkTc+bMQfv27QWOvX79eigqKsLIyEiktXo9PT0xePBgzJo1S+i4SkpKcHBwwKpVq3D16lVERUXh6NGjqKqqgr29PSwsLGBgYMBomaWXlxf09PTg7u5Od9wDAgKwdu1abN26lXHmZQB4+PAhbG1tYWJiQu/l9vX1RUVFBcLCwriy3PPj0qVL2LlzJ0aOHMm4TT9T3b9ZMzMzdOnSBatWrcKLFy8Yx2SjJviWLVu+uw//d9OxY8dGtY2KX35+fgBq9uILk9uALTk5OVBQUACHw0FOTs53n8t04KFuQtKKigpcv34dFEVBV1eX8TLzoqKiehP19e/fH/n5+Y0mpqSkZL2D93/++WejvhfQ0NCAhoYG1q1bh6NHjyIgIAAJCQm//Hol2EM68wTRxJWWlsLb25trNr6xJ2gh2CEnJyey5fC1VFVVsWbNGjg4OODy5cs4deoU9u3bh3379vEs5+RHVlYWYmNjRb7NIjc3F6GhoSJdui8mJoZRo0Zh1KhR+PjxI+Li4hAdHY0NGzbAz88PN2/eFDjmvXv3cPjwYa7Ogbi4OBYtWoSZM2cK1V4fHx+MHz8e9vb29LELFy7AxcUFXl5eCAsLEyheixYt0LVrV6HaVOtnJBL7dtXIiBEjcOTIESxatIhxTDZqgotqHz6Hw2HldRV1TLYGiN+/f8+z/D83Nxfi4uJcx4TJH9CYjR07lk4COGbMmHrPG9OBh6NHjyIqKgpATcmzSZMmwczMjF6Z0r59exw4cIBRPo2xY8ciIiICTk5OXMejo6MxYsQIgeOxFdPS0hJbtmxBYGAgXUr08+fPCAgIgKmpKaOYP0NOTg4SEhIQHx+PjIwM6OjowMTE5Fc3i2AR6cwTRBM3depUvHv3Dl26dEG/fv1w6tSpBp/7uy6LZ8PPyBsgahYWFti6dSt8fX0ZL7NviLi4OPT09KCnp4f8/HzExcUxitO9e3fk5uaKvDOvpaWFZ8+eibQzX5ecnBysrKxgZWWFx48fIzo6mlGcqqqqemd1SkpKeDohgnr8+DE8PT25kutJSEhgwYIFjG7mFi5ciMDAQHh4eAi9vcTd3Z1rJq+iogK+vr48g09MysfV+ueff+ib7lo9evRAZGQkkpKS+I5TN6M7vzXBBSGqffgURWHp0qVc5TLLysrg4ODAM2sqSMf0Z5wrUZg+fTrX1xRFcW2pEHYGvW5ug/LyclaSVgojPDyc/jwKDw8X2SBMaGgogoKCMHXqVPzxxx8ICAjAqVOnUF1djcOHD4OiKHh6eiIgIADbtm3jK2bdvffl5eU4dOgQUlJS0L9/f4iLi+Px48e4e/cuZsyYwXc72Yj57aDI27dvMXr0aCgpKUFcXByZmZkoKytDYWEh3zF/lmPHjiE+Ph737t1Dp06dYGRkBGNj40a1HYRgB+nME8RvQEFBAZWVlfRIen1+5z3uotZU8wYkJibi/v37GDRoEOTl5XluPr9tP1Py8vKM90uuWLECGzZsgLW1Nbp168ZTJWHgwIGM4s6cORMuLi64d+8elJSUuDo4AITKkvyt3r17M67nPGzYMAQHB9NLeAHg48eP8PX1FbqslIyMDF6/fs0zoPH+/XtGVQkuXbqElJQUDB48GPLy8jznit/raeDAgfjvv/+4jmlra+Pjx49cJSqZiImJweTJkyElJYXbt283+DxBOjo/yujeq1cvodosqn34xsbGPMeErb7A5rkSpZ81ax4REYG9e/fi/fv3OHfuHPbt24e2bdsK9b6/ePFiODg4/DDh5dKlS7lyt3yr7mDDoEGDGLfnWydOnICHhwcmT54MoCYp6syZMxEcHIz+/fsDqOlI29nZ8R3z2733tfvQ09LS6GP9+vXDy5cvf2lMY2PjX1qOUBje3t6YOHEi7OzsGH+OEk0TqTNPEATxmwgKCvru441h4OF7ZZ6EmUVjK66o5ebmwtLSEp8+fUJxcTGUlZXx9u1btG7dGocPHxaqM+bu7o7ExES4urpCU1MTHA4Hjx49wubNmzFw4EC4uroKFK+pXE/JycmQl5cX2TVQN2YtQWp+/8jGjRvx6NEjeHp6Ijs7Gx4eHti+fTvOnDmDS5cu4dy5c0L/DEJ48fHxcHV1hZWVFfbt24eEhARcvnwZfn5+WL58OWxtbRnFHThwIKKjo9G5c2eh2ldfpvmGCLKKok+fPvj777+53os0NDQQFxcHZWVlAMB///2HUaNG4fHjx/w3WECRkZGYNGmSSJPOshFzzZo1+Ouvv1iv1vAjnz9/bjLVFwjRIjPzBEE0qO6sV12fP3/GiRMnfstkTU1ZY+hc/YioVgd8q+5sTGPWvn17xMTE0AmJqqurMWfOHBgaGgp9g7l69WpkZ2dj3rx59OwSRVEYP348/vrrL4HjNYXrqe55byrXABv78Juixrh0va6wsDA4OzvD2NiYzjdhaWkJWVlZBAcHM+7MGxsbw8/PD0uXLoWioiLj16DuzDRFUUhJSUGbNm3Qq1cvSEhIIC0tDbm5uRg7dqxAcSsqKnhW8khKSnKtduJwOKwngatNairKjjcbMS9cuIBly5b98s588+bNkZiYiNDQULx8+RLHjx/HqVOn0LVrV5GuTCMaH9KZJwiCS0FBAb5+/QqgZuS/e/fukJOT43rOkydPsHXrVtKZb4TS0tKQnp5O32hRFIXy8nI8ePAAW7ZsYRQzJycHGRkZGDhwIEpLS4W6aRF2GfCvUF5ejjdv3qBr166gKIpnCb+g/vjjD4H2cQoSNyQkBK9evcKzZ88gISEBFRUVgZJUBQUFwcbGBn/88cd3Z+Y5HA6WLl0qglaLjqWlJXbu3MmTLyI/Px82NjaIiYn5NQ0D+/vwmxI2lq6zITMzk6ssWa0BAwbg/fv3jONevHgROTk5Da7A4HcFSd1M81u3bkX79u25cmZUVVXBxcWlyS4bZ2PhcFOJyURycjKWLl2KKVOm4P79+6iurkZVVRWcnJxQVVWFadOm/eomEiwhnXmCILhcvXoVa9euBYfDAUVRPEmGgJoPr6ZSrup/ycGDB+kOe+35q/13fTelP1JeXo41a9bg7NmzEBMTw7lz5+Dt7Y3i4mIEBQXxnWRv7NixOHnyJOTk5BrMulxLkJn7nj170tmc1dXVvxuXyTJ7iqLg7++PQ4cOoaKiAufOnUNAQACkpaWxefNmvjv1lpaWfP9MUewFVlJSgpKSEgoKCnDr1i1ISEjwvaQ3KioKZmZm+OOPP36Yg6MxdOYTExPx6NEjAMCtW7cQHBzMs9Q0KysLb9++FSiuqDtAbO/Dbyri4+Ph7+9PL10HABUVFfj5+UFaWprxbDcb2rRpg5cvX/Jsrbh7965Qgy7Lly8Xtmk8jh07hoiICK5ZfnFxcdjY2GD69Olwd3cXKN6PEsB+/vxZNA0nRGbHjh1wcHDA3Llz6YEie3t7tGzZEvv37yed+d8Y6cwTBMHFyMgInTp1omuKb9++nSuDO4fDQfPmzQWuWU2w7/Dhw1i4cCGWLl2K0aNHIyoqCp8+fcLq1asFXmoJAMHBwUhLS0N4eDhd3svS0hJOTk7w9fXF5s2b+YpjbGxML9sUZYKhLVu20AMKW7ZsEXkH7NChQ4iNjcXGjRvp31VPTw+urq6Ql5eHg4MDX3Fu3boFMTEx9O/fX+h9st+Tnp6O5cuXw93dHerq6jA0NMR///0HKSkp7Nmzh68Ee3XLiLFVUkyUOnXqhM2bN9OZy8+cOcNV07z2/UrQbQaizuhe3+zd3bt3eUqr/e7YWrrOhlmzZsHV1RVr164FALx8+RJJSUnYtm2bUKvS6ktcKCwJCQnk5OTwJNXLyMgQeB81vwlgSQncxuXZs2fw8fHhOT5+/Hhs3779F7SI+FlIZ54gCB61mVAPHjyIfv368WSxJhqnnJwcTJ8+HVJSUlBXV8ejR4+gp6eHtWvXwsvLS+Ab0NOnT2PTpk1cmZJ1dHTg5uYGR0dHvjvzdZfPinJWqu5NMRt1dI8fPw4XFxeMGzcObm5uAEDnkPDw8OC7M+/v74+zZ88iKSkJ5eXlmDx5MiZNmiTyJdXe3t5QVFSEsrIyzp49i4qKCiQmJuLo0aMIDAzEsWPHGMXNy8tDRUUFT4eUScmjw4cPY+rUqTwlHplSVVWlOxljxozByZMnecrTCaqpZHT/FQoKCoR6fdlauv7161f8/fffyMjIgI2NDdLT06GqqipUW21tbVFcXAxHR0eUlZVh4cKFkJCQwOzZs7Fw4ULGcYGagbJnz56hqqqKPla7HSo8PFzgeFOmTIGzszPs7OygoaEBiqJw584d7NixQ+Ca6E1hEI/gJSsri9zcXHTt2pXr+PPnz0X2fks0TuQOnSCIBuno6ODx48cIDQ2l9+CqqqrCysoKmpqav7p5xDdatGiByspKADVLrV+8eAE9PT2oqKgIvMwYQL03BkDNjExRURHjdt65cwd37tzh6SAKs3S7uroaCQkJDcZlki/gzZs36NmzJ8/xHj16CFRCTF9fH/r6+igpKcHFixdx9uxZbN26FX379sWUKVMwfvz475ag4te9e/cQGRkJeXl5JCUlYeTIkWjfvj2mT5/OqINw//59rFmzBq9fv+Y6Lkz97r1798LHxwejR4/GtGnTMHz4cJGtqPheJ+Tr1698l+eruxe5MWN7+0ZRURF8fX1hbm4OVVVV2NjY4ObNm1BSUsKePXsYZfZnY+l6Xl4eZs+ejby8PJSXl2PmzJkICwvDo0ePcPDgwR+WgPueVatWYfHixXjx4gUoioKysrLQydMCAgIQEhKCdu3a4b///kP79u2Rl5eHqqoq6OvrM4rp4OCAr1+/YuPGjaisrARFUZCWloa5uXmjy0VAsGPq1Knw8PCAh4cHOBwOSktLkZiYCDc3N7rMIPF7Ip15giAalJKSAmtra6ipqWHYsGGoqqrC3bt3YWpqivDwcLrmLNE4DBgwALt374aLiwvU1dVx4sQJLFiwACkpKTzLg/mhoqKC69evY+bMmVzHExISoKqqyqiNe/bswdatW9GqVSueNgnTmff29sbBgwehrq4uskzFnTp1wsOHD3mWxicmJjLqyMjIyMDIyAhGRkYoLCzE+fPn8ffff2PLli0YOHAg9PX1hco6LCYmBikpKVRVVeHff/+Fs7MzAKC0tJRRnXl3d3e0atVKoPwIP3LlyhUkJycjJiYGK1asgKysLIyMjGBsbEyXvWKqsLAQwcHBXDOeFEWhoqICz58/x507d0TxKzAm6m0gbCeT9PT0REpKCubOnYtLly7hzp078PHxwenTp+Hj44MdO3YIHJONpeteXl5QVVVFfHw8hgwZAqDm/WDVqlXw9vbGnj17GMUFgJKSEpw5cwbp6ekQExND7969MXHiRK4tGIKKjY3Fhg0bYGZmhlGjRuHo0aNo3rw5li5dyrj0oZSUFDZv3ow1a9YgMzMTQM37d9197wA7pdn+1zWWBIN2dnZ4//49vTfe2NgYFEVh1KhRsLe3/8WtI9hEOvMEQTRo69atmDFjBlxcXLiOu7q6IjAwsMnMYP2vsLOzg7W1NSIiIjBnzhwEBwdDR0cHX758gY2NjcDxli9fDjs7O6Snp6OqqgrR0dF4+fIlzp8/j4CAAEZtPHz4MBYvXoyVK1cy+v6GxMbGYv369TAzMxNZTBsbG7i6uiI3NxcUReHGjRs4duwYDh06JFCN5/q0atUKM2bMwKRJkxATE4PAwEBcu3ZNqM68lpYWdu/ejTZt2uDLly8YMWIEcnNzsXXrVmhpaQkc79mzZzhx4kS9qxOY4nA4GDZsGIYNG4bS0lKcP38e58+fh4mJCdTV1TFjxgzo6+szGnzYvHkzkpOTMWzYMJw5cwb6+vrIyMjAkydPsGrVKpH9DkyJeh++ILXDmUhMTMTOnTuhoqKCsLAwDB06FFOnToWamhrMzc0ZxWRj6fq///6LPXv2cHVcW7VqBUdHR4FWL3wrIyMDVlZWKC0thZKSEqqrq3HixAns2rUL4eHh6NChA6O4eXl5dAJZdXV1PHz4EBMnToS9vT2cnZ2Fem9s0aIFNDQ0GnycjdJs/+saSzZ7SUlJ+Pv7Y8WKFXTZUzU1NXTs2BH+/v5wcnL61U0kWEI68wRBNOjx48f1ZsE1NzevN8s98WspKCjg4sWL+Pz5M1q0aIHIyEjEx8ejQ4cOmDhxosDxRo8ejR07diAkJATi4uIIDQ1F9+7dERAQgAkTJjBqY2FhISs1b8vKyjB8+HCRxpw2bRoqKysRHByMr1+/wsXFBfLy8rC3t8ecOXMYxy0tLcWlS5dw9uxZJCcno0WLFvQ+emFs2LAB9vb2yM7OhpOTE/7880+4ubnhxYsXdOZwQSgoKKCiokKoNn3P58+fUVhYiI8fP6K8vBxiYmIICQnB1q1b4efnB11dXYHiXbt2DT4+Phg5ciTS0tJgY2MDdXV1bNiwAS9evGDpt+DPz9iHX1BQgMzMzHrLUjJZ8fL582c6ydn169dhbW0NoKYEYt293oIS9dL10tJSnhnoWrXbjphwd3dHz5494efnR+85LigogIODA9zd3b9buvF7WrVqhdLSUgCAoqIifW127NgRubm5jNvLj8bS8axVXl7OlYG/rqVLlzLafiTqmD8qzbp//360b99e4HaKQllZGXx9fZGQkABxcXEYGhrCwcEBioqKAGreExcsWID379+TzvzvjCIIgmjAyJEjqVu3bvEcv3nzJqWjo/MLWkR8z5gxY6jU1FSRxXv9+rXIYtWytram4uPjRR53+fLl1OHDh0Uas7S0lP53fn4+lZeXxzhWSUkJFRcXRy1evJjq06cPpaOjQzk5OVHXrl2jKisrRdHceuXl5TGOHx0dTc2aNYtKS0ujysvLRdKer1+/UnFxcZSNjQ3Vq1cvasSIEdTWrVuprKws+jmbNm2iRowYIXDs3r17Uzk5ORRFUdSKFSuo6OhoiqIoKi0tjVG8piQhIYHS1NSkevToQamrq9P/V1dXp8aOHcsopomJCRUREUFduXKF6tGjB5WZmUlRFEX5+vpS06dPZxRz1KhR1LZt20T63mJra0u5ublRFEVRWlpa1OvXr6ny8nJq2bJllI2NDeO4WlpaVHp6Os/xp0+fUtra2ozjrl69mpo3bx6Vk5NDxcXFUZMnT6by8/OpsLAwavTo0Yzj8qP29fnVjh49So0ePZrq2bMn9fr1a8rFxYXasWNHo4pZVlZG2dnZUT169KBjLl26lLK0tKSKioqEaquoeHh4UL1796bWrVtHbdy4kdLW1qZ27dpFVVVVUa6urpS6ujo1fvz4eu/jiN8HmZknCKJBo0ePhpubGwICAugkQi9evICHhwdGjx79i1tHfKusrIzR8uSGjBs3Dv3794eJiQkmTZokcImj+kyaNAmbN29GamoqlJWVeWZQmM7a9+nTBz4+Prhx4wZUVFR4asAzSQI1ZMgQTJgwAcbGxnyVdWvIkiVLkJycjGbNmmHMmDHYuXMndHV1WakS8fnzZ8TFxdEJK7t3747Jkyczmvncvn07Pnz40OA5YZIAT1dXF5WVlRg1ahR27dqF4cOHc5WSq33Ot2Ww+KGgoIC3b99CQUEBSkpKSEtLA1Azk1xYWChwvKZk9+7dmDJlCmxtbekEcB8+fICrqyvjChIrVqzA8uXLUVFRgSlTpkBJSQmenp44cuQIdu7cySjmzJkzkZCQgODgYGhra8PExAQTJ04UamZ+zZo1MDMzw61bt1BRUYFNmzbh5cuXKC4uxuHDhxnHbdeuHd6/f4/u3btzHS8pKUHLli0Zx3VwcMCiRYtw7tw5mJqaYv/+/Rg6dCj9u/zu4uPj4e/vDysrK3rFkIqKCvz8/CAtLc2oPCEbMUVVmpVNFy9ehLOzM71SbNSoUfDw8MC7d+9w8uRJzJs3DytXrmxwpQLxe+BQVCNbc0MQRKNRWFgIa2trPH36FLKysuBwOCgqKoKamhr2798vdAkoQrSCg4MRHx8PMzMzdO3aladjX1tykF937txBXFwc/v77b5SXl0NPTw/GxsZ0kikm1NXVG3yMaYZ0oKYs2ffiMukcxsXFIT4+Hjdu3EC7du1gaGgIIyMjegkjv9TV1SEhIQENDY0f3lQxyTpe6927dzA3N0d+fj66deuGqqoqZGVlQV5eHkePHhV4j290dPR3H2dSL3vv3r2YPn065OTkGnxOZWUlo4EOf39/OjlbRUUF7Ozs4OrqiosXL+L58+eIjY0VOGZT0adPH8TGxkJZWRlWVlawsbHBiBEjcP78eezevRtRUVGM4n78+BG5ubn03+2DBw8gIyMjVIZ4AEhNTUVsbCzOnj2LkpISjBs3Tqj3lg8fPiAiIgJPnjxBdXU1unfvDlNTU57klYK4dOkStmzZgrVr10JHRwcSEhJ49OgRXF1dMWPGDK6tRkzKNJaVlUFaWhpfv35FUlIS2rdvjz59+rCaUE1bWxtxcXGME+2JgrGxMSwtLWFsbMzVnujoaAQHB+P8+fONIub48eOxadMmDBkyhCvmjRs34OjoiGvXrgkcU9T69OmDM2fO0OezqqoKffr0QatWrRAYGMhVVpb4fZGZeYIgGtSqVSucPHkSSUlJeP78OSiKojPbi4uL/+rmEd/Ytm0bANA10eti0lHu378/+vfvj/Xr1yMxMRFxcXFYvHgx5OTkYGhoyChDbu1sqah9rywZ0zFrAwMDGBgYID8/H/Hx8YiPj8fu3buhpaUFExMTzJgxg684RkZGPyXjsZeXFxQUFBAZGUkPtOXl5WHlypXw9fWFv7+/QPEa6qyXlZUhISGBURvDw8MxaNCg73bmma5YWL58Ob5+/Yp3795h6tSpmDRpEuzs7CArK4vt27czitlUSEtL06tRlJSU8Pz5c4wYMQIaGhrIyspiHFdOTo7rXPXt21fotgKAhoYGNDQ0sG7dOhw9ehQBAQFISEhgPJhXXFyM0aNH08njwsLChM73sGTJEgA1q3rq/v1SFAVvb2/4+PgwKtM4duxYnDp1it673axZM4wbNw65ubkYPHgwbt68KVS7G7vMzEwMGDCA5/iAAQPw/v37RhOTrdKsolRRUcG1Yk5cXBzS0tJYt24d6cj/DyGdeYIgvktMTAwjR46ks+8SjReT2Wd+SEpKQk9PD9ra2vRMx759+0RW7qaiogLnzp3D8ePHRVohIT8/HydPnkRkZCQuXrzIOI68vDzmzp0LMzMzHD9+HAEBAXBxceG7M+/l5cX4ZwsiOTmZZ8VMmzZtsGbNGkbLTL+VkZGBY8eOITY2FsXFxXQJJEFISUmxsr2gNnZtOT4A2LRpE92Z/90HHzU1NXHs2DE4OjpCVVUVly9fho2NDV68eMGz5eR7xowZw/fAkzDvNzk5OUhISEB8fDwyMjKgo6MDExMTRrGSkpKwdOlSzJs3D5qamgCAs2fPYseOHdi7d2+9nTx+CLNK5ltnzpxBUlISAODt27fYvHkzT3m7t2/fNpoyZ2xq06YNXr58ybM64O7du2jXrl2jiclGadafRVSDbkTTQDrzBEFw4beUD4fDQXh4OMutIX4kLS0NampqEBMTY6Xu9OfPn3H+/HnEx8fj5s2b6NSpE2xsbBgtsf5WdnY2jh8/jqioKBQUFDAu9fStmzdv4tixY7h48SIqKiroG3ymUlJSEBcXh3PnzqGqqgoTJ05k3PFgk7i4eL05E6SlpVFeXs4oZmVlJc6fP4+IiAikpKSAoigMGjQI8+bNYxTPwMAA8+fPh6GhIRQVFXnaK2ylgxcvXiA9Pb3e35eNKgqNxdKlS2FjY4M///wTJiYmCAoKgr6+Pt69e4fJkyfzHcfY2JjuUBYWFuLIkSMYPXo0tLW16WXm58+fZ3z+jx07hvj4eNy7dw+dOnWCkZERjI2NGS1TrxUYGIj58+djxYoV9LHIyEgEBATAz88Px44dYxRXR0eHcZu+pa2tjWPHjtGrhHJycrgGWTgcDpo3bw5vb2+R/czGatasWXB1dcXatWsBAC9fvkRSUhK2bduGuXPnNpqYbJRmZUN9A0Df5iEhfm9kzzxBEFx+VD87JSUF2dnZkJGRQUpKyk9qFdGQnj174tq1a1zlcmJiYjBu3Die+tWCsre3x5UrV8DhcDBhwgRMmzaN8SxXrerqaly6dAnHjh3D9evX6a0b1tbWmDJlCuNZ2+LiYkRFReH48ePIzMwEAAwbNgwLFy5k3ObaPdjv3r3DwIED6WRdokwyKEqLFy9Gs2bN4OPjQ3cUKioq4OjoiKKiIoSFhfEd682bN1wDLbKysiguLkZwcDBGjRrFuI1s5UwAgD179mDr1q2sxG4KcnNzUV5eji5duuDly5c4evQoFBQUYGFhwSgB1tKlS6GlpcWzquPQoUO4ePEio8FcbW1tejBM0BweDdHS0kJ8fDzPrGx2djYMDAxw7949RnF/9Fno6enJKK6FhQWCgoLocneisHjxYjg4OPwwl8G+ffswa9YsyMrKiuxnM7F161aEh4ejrKwMQM3WmtmzZ8PJyYlxR5SNmFevXkVISAhXLgZbW1vGpVlFTV1dHZMnT+Za5REfH48xY8bwfP4zvV6Jxo905gmC4EtJSQm8vLxw8uRJDBkyBO7u7kLNphCioa6ujuTkZK7OfL9+/RAbGyt0kiMLCwu6A9tQHWd+ffjwASdOnEBkZCRyc3Px559/YsKECThx4gRiY2MZL1t8+PAhIiIicPbsWXz9+hW9evXCpEmTEBgYKFRcANDT06NnDtlY9SBqGRkZmD17Nlq0aAENDQ1wOBw8fPgQJSUlOHToEHr16vXDGJcuXUJERASuXbtGZ9/X19fHsGHDoK2tLfRryqahQ4di9uzZWLhw4f9k9uYbN26gqqoKw4YNAwB4eHhg3LhxjGeYtbS0EBsby5PwMSsrCwYGBnjw4IHAMT9//iySqhh1jRkzBmvWrOHpYF26dAmurq5ITExkFNfCwoLr68rKSmRnZ6O0tBSTJ0+Gh4cH4zbXVVBQgFu3bkFDQ4Nxwr6BAwciOjpaqIR/P9uXL1/w4sULUBQFZWVloSoasBmzMfv2Gv0eUW5hIxoXssyeIIgfSk5OxoYNG1BUVARXV1fMmjXrVzeJ+A5RjdGK6sN/+fLluHz5Mpo3b46xY8dCX18furq6EBcXx4kTJxjHNTExwdOnT6Gqqor58+dDX18f3bp1A1Cz9FZYwuyz/xVUVFQQGxuLI0eO0Akrp0yZgtmzZ/M9sLNkyRIoKyvDz88PY8eO/amrEHJycoQaIKyoqICBgcH/ZEc+Li4OTk5OWL16Nd2Zz83Nxbx58xAYGAg9PT2BY7Zr1w7Xr1/n6cxfvHhRoMGtdevWwdnZGTIyMvUm56yLyeyhsbExXF1dUVRUBE1NTXA4HDx69AiBgYFCbQeq7/2Poihs3LjxuwkcfyQ9PR3Lly+Hu7s71NXVYWBggLy8PEhJSWHPnj2MymAaGxvDz88PS5cuhaKiYqP+G8jJyaH/XTsIXVRURCeVY/IewEbMmJiY7z7eGLbtkA46AZDOPEEQ31FaWgovLy9ERkZCV1cXHh4eZDb+Nzd27FicPHkScnJyP0yGxW8CrAsXLkBZWRmLFi3CsGHDRFbS8MmTJ1BWVoaBgQGGDh1Kd+SFYWlpiaCgILRs2fKH+SNEmSBLFIKCgmBjYwNHR0eu4yUlJfDw8OBKDtcQfX19/PPPP3BxccHp06cxYcIE6OnpCb1lo9abN2/g7e2NZ8+eoaqqCkBNB6m8vBwFBQV48uQJ49iGhoY4ceIEz+//v2DPnj1wcnKCqakpfWz79u04cuQIduzYwagzb2NjAzc3N9y/fx99+vQBRVG4c+cOLly4AD8/P77jvHnzBtXV1fS/RW3JkiX4+PEjNm/ejMrKSlAUBQkJCVhYWGD58uUi/VkcDgfz5s2DmZkZ4wSg3t7eUFRUhLKyMs6ePYvKykokJibi6NGjCAwMZLTH/+LFi8jJycG5c+fqfbwxbTH50ecKk7ayEbN2//23pKWl0aFDh0bRmScIgHTmCYJoQO1sfGFhITZt2oTZs2f/6iYRP4GxsTE9G1s3GZYwQkNDERUVhQ0bNqCyshIDBgzAlClTMG7cOKHinjt3DlFRUTh8+DC2bt2Kzp07Y/LkyZg0aRLjmJ06daL3V3bs2FHk2aVfvXoFV1dX3Llzp97SWYLedGZkZKCgoAAAsHPnTqirq/PsxU1PT8eJEyf46sz7+/ujpKQE8fHxiI6Oxpo1ayAtLY3hw4eDoiihV324u7sjMzMTkyZNQmhoKObNm4fMzExcuHABmzdvFir2/PnzYWBggDNnzqBz5848566xDb6IUnZ2NoYPH85zfMSIEfDx8WEUc9asWZCRkcGhQ4dw/vx5cDgc9OzZE7t27RKouknd2cPvzSTm5eUxaqe4uDhcXFywevVqZGZmQkJCAkpKSqytKsnLy8Pnz58Zf/+9e/cQGRkJeXl5JCUlYeTIkWjfvj2mT5/OOKmsqAct2PTt32FlZSVevXqF/fv38/Ue9bNifltGtbKyEllZWXBxcYGZmRmjmATBBrJnniAILqWlpfD29uaajVdQUPjVzSIaoK6uzpNMydbWFh4eHjwleUSVcAqoSWQnaFKh4uJixMfHIyoqCqmpqZCQkEBVVRU2bdqEGTNmME5SVF1djaSkJJw6dQqXL19GZWUlgJqkUNbW1r882VNdVlZWyMnJgYWFRb3tEnRZ8JUrV7Bo0SK649rQR/q0adMY7fF98eIFTp06hfj4eOTl5eHPP//EzJkzMWfOHLRv317geAMGDEBwcDAGDhxIL4/W1NREQEAAXrx4gZ07dwocs5atrS0ePnwIXV3denM8/M4JoCZOnIj58+dj+vTpXMdjY2OxY8eORrNlpGfPnkhOTuZZnfPmzRtMnTqVcbI6oKaTXVFRwfM3wHQ1WVBQEM+x4uJinD59GlpaWvU+zo8BAwYgOjoaHTt2xODBg+Hs7AwjIyOkpaVh7ty5+PfffxnFbequXLmC3bt3M64+8LNiPnr0CA4ODg2ugiCIn4105gmC4DJmzBi8e/cOXbp0gYGBwXefu2zZsp/UKqIh6urq4HA4P5wxFWU276SkJCxYsECoeOnp6XQnsaCgAAoKCpgzZw4WLFggVNsKCwsRGxuL6OhoPH36FH/88QcMDAzg6uoqVNy6bt26hb/++gtXrlwR+Hs1NTURHh4ObW1tkbUnJycH1dXV0NPTQ2RkJFdHqbbkVevWrYX6GVVVVbh8+TJOnTpF18tOTU0VOE6fPn1w/vx5KCgoYPXq1Rg8eDBmzJiBzMxMWFhY4Nq1a4zb2LdvXxw4cECkr21TcejQIWzduhWWlpbo27cvvW88PDycrsHOj9qtGn/88ccPO6v8vv+fPHkScXFxAGr+drS1tbnKsgE1CTK/fPnCKFnd/fv3sWbNGrx+/ZrrOEVRQr3vjRkzhueYpKQk+vXrh1WrVqFt27aM4s6fPx/t27dHmzZtEBoaiqtXr6KiogIbNmyAmJgYdu/ezSjupUuXuLavAEB5eTkePHjQJMrIvnnzBpMnT8bDhw8bdcyMjAwYGxuLNCZBCIMssycIgoeCggIqKysRFRXV4HM4HA7pzDcC/O5bb2zU1NSwbt06ODo60p3E7du3C92Zb9WqFSwtLWFpaYm0tDScPHkSCQkJIu3Ml5WVITc3l9H3ysnJiWz/ea3amcd//vmHla0BQM1SZj09Pejp6SE/P5/unAmqS5cuSE9Ph4KCApSUlOiOVnV1NUpLS4VqY5s2bUT+2jYVFhYWKC8vR3h4OEJCQgDUJLCzt7eHubk533GioqJgZmaGP/74Q2Tv/3p6erhz5w79dYcOHXiWwKupqTHeg+zu7o5WrVohKChIpKtwLl26JLJYdW3YsAH29vbIzs6Gk5MT/vzzT7i5ueHFixfYt28fo5gBAQEICQlBu3bt8N9//6F9+/bIy8tDVVUV9PX1RfwbiF5JSQkOHDjAaLUPWzFv377N9TVFUSguLsaBAwe+W2KTIH42MjNPEATxP8bCwgJ+fn6Mb3JEMTNfn/z8fK4Se6JSUVHBMxMoDGF+/3379iElJQW+vr6sLP+/fPkydu/ejWfPnkFCQgKqqqqwsbEROj+BqOzbtw979+6Fl5cX5OXl6SRlycnJ+PLli1DLYc+cOYNjx45h48aNUFJSgri4uAhb3nR8/PgRkpKSjbIsV93M9qLSp08fnDhxAj179hRZzG8VFBQgJSUFbdq0Qb9+/UQePz8/H61bt+a6ZiMjIzFp0iS+XqtRo0bB1tYWZmZmGDVqFI4ePYrmzZtj6dKl0NHRwcqVK0XeZqZqV5N9i8PhwM3NDdOmTWtUMb/tJnXp0gV+fn7o27evwDEJgg1kZp4gCOJ/TGpqKsrLy391M3iw0ZEHINKOvLASExNx//59DBo0CPLy8jwlpIRZaXHx4kUsX74c48aNg76+Pqqrq3H79m2sXLkSO3bswNixY4VtvtDmz58PCQkJcDgcaGpqYtmyZQgODoaCggJ8fX2Fih0YGIicnBxMmTKl3scbU0ZvUbh9+za0tbUhISHBM4v4LX7zZURGRmLGjBnffU5BQQHWrFmDvXv38t3WWg3lLSgvL8fDhw8xYMAAgWMqKCjUm0ySqZ07d+LgwYM4ceIEFBUVcffuXSxYsAAlJSUAAF1dXQQHB4s0wV59732enp4YPHgwX535vLw8Oimhuro6Hj58iIkTJ8Le3h7Ozs6NqjO/ZcsWno63pKQktLS00Llz50YTs773YklJSZ5cNATxq5HOPEEQBEH8JIMGDcKgQYNYib1z504sW7YMS5cupY/NnTsXQUFBCA4ObhSd+T179sDQ0JBeFWJrawtbW1uRxF68eLFI4jQVFhYWSE5Oplc4NJQ7Q5B94y4uLpCWlm4wX8qNGzfg6OhI1+8W1NOnT+Hs7Ixnz57R5eq+fVxQS5YswZYtW+Dq6gplZWWhBu+OHz+OkJAQzJ07l+5gOzk5oXnz5jh+/DhkZGSwfPlyhISEsN5BFmThbKtWrehtKoqKinjx4gWAmi04TLcEscXExKRJxOzUqZPIYxIEG0hnniAIgqCtW7fuh89pbDeHosRPhuqsrCzG8dnMM5GRkYHAwECe41OmTGE0i8qGkJAQTJgwgZXYglYCaOr++ecfOtnhhQsXGFeDqMvCwgJOTk6QlJTkKvFYVVWFwMBAhIaGQkVFhfHe7i1btkBCQgIbN26Eu7s71q5di9evX+PIkSOMS+ht374dHz58aHDPvSADBJGRkVi7di1MTU0BAA8fPsSrV6/g4OAAFRUVADWDRl5eXo1qtltXVxc+Pj5wd3eHhoYGdu/eDVNTU5w7d46ncsCvVl1djYSEBLo857eDFkyqTrARMz8/HwEBAQ3GbKr5aojfD+nMEwRBELQ3b97w9TxBlsP+aAlwXYKUz4uJieH7ufwm1/pe0q+6hCnX+PjxY4SGhnLta7eysoKmpibjmEBNwrNXr15BUVGR6/irV68Y7c9/9eoVXF1d6ZvZbzGZRdXS0sKlS5dgbW0t8Pfyo6ln9BZE3ZnDlStXYsuWLUIn5nJycgKHw4GjoyOkpKQwduxYvH37FqtWrcLDhw9hYWEBBwcHnu0h/EpNTUV4eDg0NTVx6tQpqKmpwdTUFB06dMCJEye4BhD4Jcoa6xkZGRgyZAj99b///gsOh0MvYQcAVVVV5OTkiOxnioKDgwMWLVqEc+fOwdTUFPv378fQoUMBAGvWrPnFrePm7e2NgwcPQl1dXWS5E9iI6eLigpSUFBgZGTWq8qYE8S3SmScIgiBo+/fvh4SEaD8avrcEuC5By0itXbuW5/spisIff/wBCQkJFBcXQ1xcHHJycnx35tnKYF0rJSUF1tbWUFNTw7Bhw1BVVYW7d+/C1NQU4eHh6N+/P+PYU6ZMgaurKzZu3EjHuXPnDjZv3oyJEycKHG/jxo3IycmBg4ODyG5mmzdvDh8fH+zevRtKSkqQlpbmevzgwYOMYzf1jN7CePv2LZo3by6SWOvWrQOHw4G9vT2sra1x5MgRtGjRAmFhYdDV1RUqdnV1NV3SrVu3bkhPT8eAAQMwduxYOgu/oES9IqPu3us7d+7gzz//RPfu3eljpaWl+OOPP0T6M4XVoUMHxMTEoKysDFJSUjh69CiSkpLQvn179OnT51c3j0tsbCzWr18PMzOzRh0zOTkZO3fupAdFCKKxIp15giAIgqajowNdXV2MHDkSI0eOFEmpILaWI6alpdH/PnPmDPbu3QtPT096djIzMxPr1q1rVB25rVu3YsaMGXBxceE67urqisDAQBw6dIhx7MWLFyM9PR0LFy6kOyQURWHkyJFYvXq1wPHu3buH8PBwkdZtl5GRYVyC7EdiY2OxYcOGejN6d+nShZWf2VjY2trC2dkZNjY26Nq1K09yttryhfyqHSgLCQlB//79ERwcjJYtWwrdTmVlZdy+fRsGBgZQVFTEo0ePAADFxcVCJeVMTExEaGgoXr58iePHj+PUqVPo2rWrwNdajx49cPv2bSgqKqKoqAg3b97k2RZy9uxZqKmpMW4rG8aOHYtTp06hdevWAIBmzZph3LhxyM3NxeDBg3Hz5s1f28A6ysrKMHz48EYfs3nz5kKtwCKIn4V05gmCIP7HfK8O+blz55CYmIirV6/Cx8cHHTt2xMiRIzFixAj069ePUbkvfhMJff36VeDYtfz8/BAQEMC1zLhbt25wdnbG4sWLYWFhwTi2KD1+/Bju7u48x83NzTF9+nShYktLS2PXrl3IyMhAeno6KIpCjx49oKCggICAADg5OQkUT05OTuR125nsXeVXU8roLWp+fn4Aara01P37pihK4BUvtdauXQsOh4Njx47RM+jCMjc3h7OzMwBg/PjxMDQ0RLNmzXD37l1oaWkxipmcnIxly5ZBX18f9+/fR3V1NaqqquDk5ISqqiqBypKZmZnBxcUFz549w71791BeXk6/d3z48AHx8fEIDQ2Fh4cHo7aK0pkzZ5CUlASgZmXG5s2beVa6vH379rvv97/C8OHDkZSUJNJZdDZiGhkZITQ0FJs3b/6fLXNJNA2kM08QBPE/5nvL3du2bYvp06dj+vTpqKysxJ07d3D16lVs3rwZubm59Kz9iBEj0KZNG4F/dmFhIYKDg7n2NVMUhYqKCjx//hx37txh9Dt9+vSJ50YWqFnWK8wggajJyckhPz8fysrKXMfz8/MZ7UMuKyuDr68vEhISIC4uDkNDQ65kXdeuXYOtrS3ev38vcGfewsICW7duha+vr0j3jL5//x5HjhyhcwZ0794ds2bNEnj2+FtNKaO3qAmzPaGubxNAtmjRArKysliwYAGsrKy4OjVMkjlOmzYNrVq1QuvWraGiogJvb2+EhIRAQUEBGzZsYNTmHTt2YPXq1Zg7dy7OnTsHALC3t0fLli2xf/9+gTrzU6dORVlZGSIiIiAmJobAwEBoaGgAqKnEcOzYMdja2sLQ0JBRW0VJW1sbx44do9/Pc3JyuDL5czgcNG/eHN7e3r+qifXq06cPfHx8cOPGDaioqPBUH2ByXbERMy8vD2fPnsXly5fRtWtXnvdnUf3NEYSwOJQgtS8IgiCIRu3r168QExODlJQUMjIycOXKFWhra6Nfv370cz58+IA2bdoInP06JycHV69eRWJiIm7dusWo47169WokJydj2LBhOHPmDPT19ZGRkYEnT55g1apVWLBggcAxgZol5oWFhfDx8aHrCmdkZMDR0RHKysr0zOWvVptQLiAggO5wv3jxAqtXr0bPnj3h5eUlULwtW7bg6NGjMDAwgJSUFOLi4mBra4uFCxfC3d0dERER6Nq1K9zd3QVKLgjUdObv37+PqqoqyMvL89zMMtk+kZ6eDnNzczRr1gyampqoqqpCamoqvnz5goiICK69yYJycHDAx48f4e7ujpSUFOzevRuHDh1CbGwsDh06xHo+hMaioKAAEhISjJbFjxkzhq/ncTicRpPNW1tbG7GxsejatSu0tbURFxeHLl26IDs7G1OmTMGDBw9E8nNyc3MhJSUFOTk5kcSrq6CggCfr/L59+zBr1iy+BtIsLCwQFBSEVq1aibxtova9a4zpdcVGzB9VdmFzlRFBCIJ05gmCIH4Tt2/fxtKlS7Ft2zaoqqpiwoQJEBMTw+fPn+Hv788oU3RDKioqGNVzHjRoEHx8fDBy5EhMmTIFfn5+UFdXx4YNG1BWVsa4PFVubi5sbGyQkZFBd2KKioqgqamJPXv2NJqb3MLCQlhbW+Pp06eQlZUFh8NBUVER1NTUsH//foHLSI0ZMwa2traYM2cOAODKlSvw8PCArq4uTp48CWtra6xcuZLRrP+PyvQxme2aP38+mjdvDj8/P7pNZWVlcHR0RFlZGeMkaEDNNbBw4UIYGRnB1NQUs2fPppeXr127FlZWVoxjNwVHjhxBcHAw8vPzAQBt2rSBjY0N5s6d+2sb9g1R7W+vNWLECPj7+2PgwIFcnflLly5h06ZNuHr1qmh/ASEVFRXB19cX5ubmUFVVhY2NDW7evAklJSXs2bNHJPkdCgoKcOvWLWhoaNCDmwRB/J7IMnuCIIjfxNatWzF27Fj06dMHp06dgoyMDM6fP49Tp04hJCRE4M789+rscjgcXLx4UeA2lpaW0smjVFRUkJaWBnV1dZibmzOelQeA9u3bIzY2FtevX8fz588BAD179sTgwYMb1Z7RVq1a4eTJk0hKSsLz589BURSd2Z7Jvsz//vsPw4YNo78ePnw43r59iwsXLmD//v0YNGgQ47Yy6az/yJ07d3D8+HGuwQVpaWksWbIE5ubmQsX+8uVLgxm9hS3719hFRkbCy8sL5ubmGDBgAKqrq3H79m1s3boVMjIyQudjEBVR7m+vNXXqVHh4eMDDwwMcDgelpaVITEyEm5sbJk+ezMJvIRxPT0+kpKRg7ty5uHTpEu7cuQMfHx+cPn0aPj4+2LFjh8Ax09PTsXz5cri7u0NdXR0GBgbIy8uDlJTU/2vv3uNyvv//gT/edeVUMmVb5VBTWoaUxZwjh0lCWDMTkhyGyRzj842cckglWURRTk2bU8hhbEbGSoZWdJBTTUbSQZTq94eb6+fadbG6uq6u6+Jxv90+t1ve19vzevappef79Xo9nwgLC0Pnzp2V8JnUTEJCAjIzMzFo0CDcu3cPpqamcj0gVmbM+/fvY8+ePcjKysKCBQvwxx9/wNLSUryrikgdsJgnInpLpKSkYPXq1dDT08PZs2fRq1cv1KtXD7169ZLr3KQy5uwaGxsjOzsbxsbGMDMzE3ekr1+/Ph4/flyj2Nra2ujRo4fCuxormpaWlnhaQE2VlZVJjCTT1tZG3bp14e3tXaNC/qW//voL4eHh4vPtFhYWGDt2rNzFsa6ursyu5TXpZP7S6NGj8f3334tze9nR+10QHh4Ob29vjBo1SnytX79+MDU1RWRkpNoU84o83/6Sl5cX7t27J/67Li4uqKysRK9evTBz5kyF5q8Ip0+fxoYNG2Bubo6IiAh069YNzs7OsLS0lPuB1qpVq2BqaoqWLVsiLi4Oz58/x+nTp7Fr1y4EBQUhOjpawZ+F/IqKiuDh4YHLly9DEAR069YN/v7+uHnzJrZt2wYjIyO1iHnr1i24urpCT08Pubm58PLyQlxcHBYsWIDw8HCJo2tEqlS9A5NERKS26tevj9LSUpSWliIxMVE8E/rBgwdyFePx8fEICAiAt7c3pk2bJvU/eQwYMABz585FYmIiOnfujH379uHo0aMIDg6GqampXDGBF2Po3N3dYW1tjdatW0v9T5Vat24t3vpsZWUlMz9F59m+ffsax0hMTMTIkSNx69YtdO/eHR07dkRWVhZGjRold6PCzp07Y/Xq1cjPzxdfy8vLg7+/f41XD+vUqQOR6N1co8jJyZHYofFSjx49cOvWLRVkJNv169dlnm/u378/7ty5I1dMHR0drF27FsePH0dQUBDWrl2LQ4cOYePGjTKbYqrakydPxCPPzp07h65duwJ48fP7ZVPQ6rp06RLmzZsHQ0NDnDlzRjxWdMSIERIjPNVBQEAABEHAiRMnxCMU586diwYNGsh9zEoZMVeuXIm+ffvi559/Fq/uBwYGom/fvggICJArJpEyvJv/6hERvYU+++wzrFmzRnw+vEePHkhNTcWyZcvkWqVVxpzd6dOn4+nTp/j777/h7OwMR0dHeHl5oWHDhggODpY77uLFi5GTk4PZs2crtPO6IqxYsUKckzKaJsk6RlDd5oayBAQE4IsvvoCPj4/EdV9fXwQFBWH79u3Vjjl79myMHDkSvXv3hpmZGQRBQFZWFvT19bFjx44a5Tt48GBMmDABQ4YMgampqdSsdWXNt1cHJiYmSE5ORosWLSSuX7lyRa6pE8rSsGFD5ObmSuWZnp5erb4WOTk5MDY2hiAIyMnJAQCIRCKJHSMvr9d0SoKimZub49dff4WxsTH+/vtv9OzZEwCwZ88eubdvv2x6Wl5ejvPnz4vH/xUXF0v9d6Bqv/zyC9auXSvRG6Bly5ZYtGgRJk+erDYxL126hB07dkj8fNXW1sbkyZPh6uoqV0wiZWAxT0T0lli0aBEWLVqE69evY82aNdDT08OBAwcgEon+szOvLMqYs/vgwQN4e3uLi83FixfDy8sLenp6SElJkTvupUuXEBkZCVtbW4XkqUguLi7ijwVBwMCBA6Ua0j158gR79uyRK/6yZcskViDLysqwZs0aqRnx1X2Q8Ndff2HZsmVS10ePHi33tm0jIyMcPnwYBw4cEPcMGDFiBJydnWv8EGbjxo0AgK1bt0q9JgjCW13Mjxw5Er6+vsjPz0eHDh0gCAISExMRHBwsnpOuDhR1vr1Pnz44e/YsDA0N4eDgIPOBVmVlJQRBEDdBVBfffvstpk+fjrKyMgwaNAhmZmbw8/PDzp07sWHDBrli2tjYYOPGjWjSpAlKSkrQs2dP5ObmIiAgADY2Nor9BGooLy8P77//vtR1PT09lJSUqE3M8vJyVFRUSF0vKiri3HlSKyzmiYjeEgYGBlLNk7777ju5OpkDypmz26dPH8THx0t0bX/vvfdw69YtuLm5yT1GqnHjxlLFq7rIy8sTz7r39vZGq1atpMZbpaamIiAgoNqdxzt27Ih//vlH4pqtrS0ePXqER48e1Sjvxo0b4+HDh2jZsqXE9YcPH8r9PQW8ODc/dOhQZGVlQVtbGx999JFCtkNfu3YNT548QWFhIfT19VG/fv0ax9QUY8aMQXZ2NlasWCHeqq2trQ1XV1dMmTJFxdn9f4o63x4ZGSleyde0ed/29vY4ffo0cnNzYWVlBQAYOHAgXF1d5V6Z/7//+z/MnDkTd+7cwYIFC2BgYIClS5ciIyMDW7ZsUWT6NdauXTscOXIEkyZNkrgeFRWFTz75RG1idu/eHaGhoRJjTR89eoQ1a9aoZUNBendxNB0RkQbbv3+/eKV3//79b7y3uiuTipqzu3PnTkRERACAuPndv7eBFxQUoEmTJoiLi6tWji9t2bIFiYmJWLNmjdpts9+/fz/mz58PQRDEq4X/VllZCXt7+xqNZlM0X19fXLx4EYGBgeIiIyMjA7NmzULr1q2xcuXKascsKyvDihUr8NNPP6GsrAzAi0Z1Y8aMkbtZWXFxMSIiInDo0CHcvn1bfN3U1BSDBw+Gu7v7O1PYFxUV4caNGwBebDPW09PDgwcP1Gar/c2bN2FmZobbt28jJSUFFRUVsLS0hIWFRY3iZmZmori4WLzNPiIiAr169ZJ6EKXucnJyFHYs4OHDh3jvvfckVpFjYmLg6OgIPT09hbyHPJKSkuDu7o4uXbogPj4ezs7OyMjIQEpKCsLDw+U6EqaMmLm5uRgzZgzy8/NRWFiIli1bIjs7G++99x527NiBpk2bVjsmkTKwmCci0mBWVlaIj4+HoaGheJVHFlVuNy0pKUF4eDgqKyuxYcMGuLu7S62i6+rqon///nL/guTm5oY///wT5eXlMDQ0lFo5PnnypNz5K0JCQgIqKiowduxYrF+/XuJ8sCAIaNCgASwtLWs8mkmRHj9+DHd3d6SmpqJhw4YQBAEFBQWwtLTE1q1bJXZXVNWqVauwf/9+zJgxAzY2NqisrERSUhLWr1+PcePGVft8a35+Ptzc3JCdnY1+/frB0tIS+vr6KCwsRHJyMk6dOoXmzZtj165daveQR5Fat24tteMFAO7evQtnZ2dcunRJRZlJ6t69u8TEAUU4c+YMpk6divHjx8PLywsAMGLECGRmZmLz5s2ws7NT2Hspwt27d7Fq1Spcv35dvIuisrISpaWlyMvLq9Fxo//SoUMHHDhwQCGz7Gvi2rVriIiIED/QadWqFcaPH1+jxp3KiFlSUoJDhw4hNTVVHHPIkCEqfRhC9G8s5omI6LXu3buHnTt3ikeTtWrVCl9++aXcq0chISHw8PBQ+EppSEjIG19Xxsx0efzxxx/o0KEDCgoKxIXXpUuX0LZtW7Uq5F+qqKjAmTNnxOfbLS0t0b17d7nPjHbt2hXLly9H7969Ja6fOHECy5cvx6+//lqteEuXLsW5c+cQEREhs1njvXv34Onpib59+2LGjBly5ayufvzxRxw8eBDAi+8rW1tbqe+h+/fvo6SkBKdPn1ZFilIcHBwQEhIi99ZnWYYNG4ZevXrh22+/lbgeGBiICxcuqNVYNgCYPHkysrKyMGDAAISHh2P8+PHIysrCiRMnsGTJEqU2V7O1tcXBgwdVWsyfP39e4dvUlREzODgYLi4uKn/wQfRfWMwTEZFMaWlpGD16NOrVqwdra2uUl5cjOTkZJSUl2L17N1q1alWlOAkJCbC1tYVIJEJCQsIb7+3YsaMiUldbt27dgoeHB/r164d58+YBADp16oT3338fW7ZsUfj0AHVjZ2eHH374Qeps8I0bNzB8+PBqryD37t0bPj4+Ug8HXvXzzz8jMDAQhw8flitndZWfn49Vq1YBAPbt2wdHR0epzuUv+xO0bdtWFSlKCQoKwp49exQ6ccDGxgaxsbFSRdedO3cwePBgtdmV8JKdnR1CQ0PRsWNHuLi4wNfXF9bW1ggMDERGRobcTfCqQh2K+U8++QRGRkYYOnQohg4dKjXZQF1ifv7557h9+zY6dOgAFxcXDBgwgCvypJZYzBMRkUwTJkxAgwYN4O/vL962/uzZM8yZMwfPnj2r8vnufx8FeHl2/N9qehTg2rVrSEtLE3cgfrl19fLly1ixYoXccRVp4sSJ0NLSwrJly8TnmPPy8uDt7Y26devWaDyfIrRu3VrcJfzl1+p15Pla+fn54e+//5b4nqqsrMSiRYsgCAJ8fX2rFa9du3Y4evToG49n3L17F4MHD0ZSUlK189UU3t7eWLhwodoXG8o4CuTg4IB58+bh888/l7h+6tQp+Pr6qs2uhJfatWuH48ePw9jYGLNmzULnzp3xxRdfICsrC25ubjh79qzS3lsdivnc3FwcOHAAsbGxyMjIgK2tLVxcXGp0ll8ZMQHg8uXLOHjwIOLi4lBSUoI+ffpg2LBh6Nq1q9wxiRSNxTwREclka2uLH374AZaWlhLXr127htGjRyMxMbFKcbKzs2FiYgJBEJCdnf3Ge+U9Mx8VFSUu2F99WCAIAuzs7OSaia4MdnZ2iImJwUcffSRxPT09HV9//TX++OMPFWX2wr59++Dk5IQ6depg3759b7z31ZF7VTVr1iwcP34cjRo1grW1NUQiEVJSUpCdnY327dtL9DqoSpfyVx8Uvc6DBw/Qo0cPtRtRpmhPnz4VzxvPzMzEr7/+CltbW3To0EHVqSnV+vXrsXv3bsycORPW1tYQBAFXr15FUFAQXFxc8N1336k6RQkDBw7EvHnzYG9vj/Xr1+PRo0fw8fFBZmYmRowYodSdBOpQzL8qNTVVXCzn5+ejb9++Et3j1SVmeXk5zpw5g8OHD+PkyZPQ19ev9pEgImXhaDoiIpJJV1cXpaWlUtdlXXuTVwt0ZXUA3rFjByZNmoSpU6eid+/e2Lt3L/Lz8zFr1iz06dNHKe8pD5FIhEePHkkV8/LOQla0Vwt0QRDEkxJe9eTJE+zZs0eu+HXq1MGgQYMkrnXs2PGtP16hbAkJCZg6dSrWrVsHCwsLfPHFF9DS0sKTJ0+wdu1aODo6qjpFlJSUoF69ehK7PdLT09G0aVM0aNBA7rjffPMNHj16hCVLluD58+eorKyESCSCm5ub1Dl6dTBs2DDMnTsXK1euhL29Pdzc3GBiYoL4+Hh8/PHHqk6vVrVu3Rrl5eXQ0tJCdHS0QgpkZcR88OABbty4gVu3buHZs2dSP7+JVInFPBERydS5c2esXr0awcHBeO+99wC82BLu7+8vd7MhBwcHmVu3BUGAjo4OjIyMMGTIkGqfnc3JycGIESNQp04dWFlZ4erVq+jbty/mz5+PlStXVnt+u7LY29tj2bJlCAwMhKmpKYAXZ3tXrFiBHj16qDi7F1/fp0+fAnixdbtVq1Zo3LixxD2pqakICAiQ6//Tqo4zrI6IiIg3NlR88uSJwt9T3QQEBKBPnz5o164dfvrpJ+jp6eH48eP46aefsGnTJpUX8/v374efnx+2bNmCdu3aia+vXLkSly9fxtKlS+XOUVtbGz4+Ppg1axaysrIgEolgZmYmdR5fXUyYMAEikQiCIMDa2hrTpk1DaGgojI2NsXr1alWnVyvu3LmDgwcPIjY2Frdv30anTp3g4+MjdVRClTGLiopw9OhRxMbGIiEhAU2bNsXQoUMRFBSksPGBRIrAYp6IiGSaPXs2Ro4cid69e8PMzAyCICArKwv6+vrYsWOHXDGHDx+ODRs2oG/fvvj0008BvDiXePToUQwbNgxaWlrw9fVFWVkZvvjiiyrH1dXVxfPnzwEAZmZmyMjIQN++fWFubv6fW/tr07x58zB+/HgMGDAA+vr6AICCggK0adMG8+fPV3F2wG+//Yb58+eLjyqMGDFC6p7KykrY29vL/R4ZGRlIS0uTucOjug9xTExMEBcX95/3ve2NBVNSUrB69Wro6enh7Nmz6NWrF+rVq4devXqJm+Spyu+//44FCxZg2LBhUl8HHx8fbN68GbNnz8b7779fozFyurq6MDExQWJiIp48eaLWxwtefRDm6ekJT09P1SVTy1xdXXH16lU0a9YMQ4YMwbBhw2pcHCsjZteuXaGjo4P+/fsjMjKSu4dIbbGYJyIimYyMjHD48GEcOHBAPJpsxIgRcHZ2lntm98WLFzFjxgxMmjRJ4rq1tTXi4+OxefNmdOjQAeHh4dUq5u3s7LBx40b4+PjAysoKe/bswcSJE5GYmCg1016VDAwM8NNPP+H3339HWloaRCIRLCws0KVLlzc2m6stQ4cORdOmTVFRUYGxY8ciODgYjRo1Er8uCAIaNGgg1UehqsLCwhAQECDzNUEQql3Mnzp1Sq483jb169dHaWkpSktLkZiYKO4f8eDBA7n/W1WUzZs3Y/To0ViwYIHUa6ampli2bBkqKyuxceNGbNmypcpxN2zYgKioKOzZswempqZISkrCxIkTUVRUBADo0qULQkND1XKF/sKFC0hOTsbTp0+lmoEqaoxmXl6eePzlS1OnThXvslIVc3NzzJ49G506dVLrmL6+vhgwYIDCx6gSKRob4BERUa2xsbHBgQMHxFvMX7p9+zacnZ1x+fJl5OTkwNHREZcvX65y3PT0dLi7u2PcuHH46quv4OzsjIKCApSUlMDDw0PtmmBpgj/++AMdOnSASKS45/7dunXDyJEjMWnSJKmz+CS/GTNm4NmzZ2jUqBGOHz+OM2fO4M6dO/Dx8UHz5s1f+wClNnTu3BlRUVFvfACUnJyMiRMn4ty5c1WK+cMPP2D58uUYN24cJk6cCD09PQwYMABPnjzB1q1boaenh+nTp6Nbt26YMWOGoj4VhXj5QKthw4ZSD1oEQcDJkyerHbOgoABr1qzB6NGjYWFhAQ8PD1y4cAFmZmYICwtTm4Z3r8rJyUFmZiY6duyI4uLiNzaxVFXMp0+f4ujRo8jMzISHhwfS0tJgYWEh9ZCESJW4Mk9ERGJ9+vTBjz/+iMaNG7/2fPtL8vzSaWhoiKSkJKliPikpSXw2+59//qn2amKrVq3w888/48mTJ9DV1UVMTAwOHjwIY2NjDBgwoNp5KsuYMWPe+HpVOrjXlk6dOuGvv/5CeHg4rl+/Lt5FMHbsWFhbW8sVs6ysDIMHD2Yhr2CLFi3CokWLcP36daxZswZ6eno4cOAARCIRvL29VZpbaWnpf66ON2rUSNyroSpiYmIwf/58jBo1CgBw5coV3Lx5E7Nnz4a5uTkAYMqUKVi5cqXaFfPbt2/HjBkzMGXKFIXF9PPzQ2JiIsaNG4dTp07h4sWLWL16NQ4fPozVq1dj/fr1CnuvmiorK8PcuXMRFxcHLS0tHDt2DKtWrUJhYSFCQkLk2kmijJgPHjzAyJEj8eDBA5SWlsLV1RURERG4evUqoqKixN9nRKrGYp6IiMRcXFzEv3gPGzZM4fHHjBmDJUuW4ObNm7CxsUFFRQUuX76M7du3Y8qUKbh37x4WL14sVzO4evXqiXM3NDSEu7u7otOvsX938y8rK8Pt27eRlpamNk36XkpMTIS7uzssLS3RvXt3lJeXIykpCaNGjUJkZKS450F1DBkyBHv27MGcOXOUkPG7y8DAQKpg++6779TioclHH32ES5cuoUWLFq+9JykpqVqTLjIzMyVmfZ8/fx6CIEj0crCwsEBOTo58SStRQUEBnJ2dFRrz9OnT2LBhA8zNzREREYFu3brB2dkZlpaWGD16tELfq6a+//57XLt2DZGRkZg8eTKAF/8uLFiwAGvWrMGSJUvUIubKlSthYWGB2NhY8ffaqlWr8N1332HVqlUICwurdkwiZWAxT0REYq+e1/zss89gY2MDHR0diXuePXsm97ifsWPHQiQSITw8HJs2bQLwoonZvHnzMHLkSJw5cwbm5uZVWk38r1XuV6nLivfrurkHBwfj4cOHtZzNmwUEBOCLL76Aj4+PxHVfX18EBQVh+/bt1Y45YcIEDB48GEeOHEGzZs2kdn6oy9dJE+zfv188OnD//v1vvLe6vQgUafDgwQgODkaXLl3wwQcfSL1+//59rFu3DsOHD69W3Fe/dy5evAgDAwO0atVKfK24uFgtzzt/+umnuHTpEpo1a6awmE+ePBE3Fzx37pz4QWb9+vVRXl6usPdRhMOHD2Px4sX47LPPxNc6deqEpUuXYs6cOXIV3sqIef78eYSFhUl8DzVq1Ahz5syp1r89RMrGYp6IiGQaM2YM4uPjpc4HZmRkYM6cOXKP/Pn666/x9ddfIz8/HyKRCHp6euLXevToUeVVeVkrebGxsXBwcFCrpndV4eLiguHDh8PX11fVqYj99ddfWLZsmdT10aNHy+xyXxX/+9//AADt27dXy0JLk8yfPx89evSAoaHhGychyNNYUJFGjx6N48ePw8nJCSNGjICNjQ309fWRn5+PP//8E3v37oWpqSk8PDyqHPPjjz9GQkICTE1NUVBQgAsXLkj9PIqLi5O7UaOivfqwpW3btli0aBHS0tJgZmYGbW1tiXvl+VqZm5vj119/hbGxMf7++2/07NkTALBnzx612w6em5src5eGsbExCgoK1Cbmmx4GvZycQqQOWMwTEZHYtm3bxKOsKisr0a1bN5n3VefMtLJWEGWtch89ehRz5sxRy4ZPb5KRkSHV1VrVGjdujIcPH6Jly5YS1x8+fCj39u0//vgD27Ztg62trSJSfKddu3ZN5sfqRltbG1u3bkVwcDBiYmKwdetW8WtNmjTBqFGjMGXKlGp1nf/666/h4+OD69ev49KlSygtLYWbmxuAFyv9sbGxCA8Px/LlyxX++chD1sOWzZs3S12T98HLt99+i+nTp6OsrAyDBg2CmZkZ/Pz8sHPnTmzYsEGelJXG3Nwc586dg6urq8T1Q4cOwcLCQm1iduzYETt37hQ/gAReHIvasGGDWo89pHcPi3kiIhIbPXo03nvvPVRUVGDBggXw9vaWaB70cjRZ586dqxxTU1YQa4Os4wOFhYWIj49Xq0Z9ANC7d28sXboUgYGB4tW9jIwMLF++HL1795YrZpMmTTRu1wTVXJ06dTB79mx4eXnhzp07ePz4MQwMDNC8eXO5RjI6Ozvj2bNn2L17N7S0tBAUFIS2bdsCeNEtPjo6Gp6enhgyZIiiPxW5KPthi729PU6fPo3c3FxYWVkBAAYOHAhXV1e1W5mfPn06vLy8kJaWhvLycuzbtw83btzA8ePHERgYqDYx582bh6+//hp//PEHysrKsHjxYty4cQOFhYXYsWOHXDGJlIGj6YiISKZ9+/bByclJLZpoVZWtrS0OHjyotivzL1cPX1WnTh3Y2trC3d1drQrdx48fw93dHampqWjYsCEEQUBBQQEsLS2xdetWucYzHTlyBNHR0Vi0aJHMLcZUdW5ublUuhN+lXgS5ubmoU6eOeDrGuy4nJwcmJiaqTkPCb7/9hk2bNiElJQUVFRVo1aoVPD095T66payY9+/fx65du5CamiqOOWrUKIX2OyCqKRbzRET0Wnl5ecjKykJFRQWAF1vvS0tLcfnyZUydOlWumCUlJSgsLETDhg0Vfm5a3Yt5TVNRUYEzZ84gPT0dlZWV4s728hbh/fv3R05OzmubcqWmptYk3XeKrF0er+sZ8brGi/R2uHv3LlatWoXr16+L/9t6+bM6Ly8PKSkpKs7w/wsLC8OQIUPw4YcfqnXMKVOmSIw6JFJX3GZPREQyHT58GAsWLMCzZ88gCAIqKyvFK4FNmzatVjFfXFyMiIgIHDp0CLdv3xZfNzU1xeDBg+Hu7v7WNkSrzngsdVtB09LSgr29vcTIr5pQ5Gztd93b1DOCambZsmXIysqCo6MjwsPDMX78eGRlZeHEiRNydXJXpk2bNtVotby2YiYmJqJu3boKjUmkDCzmiYhIpo0bN2LQoEHw9PSEq6srIiIicP/+ffj6+mL69OlVjpOfnw83NzdkZ2ejX79+cHV1hb6+PgoLC/HXX38hLCwMcXFx2LVrl8T5/P8ia2WyrKwMa9asUauVSQcHhypvh1b1ynSfPn2qfO/JkyerHd/FxaXaf4eI3iwxMRGhoaHo2LEjfvvtN/Tt2xfW1tYIDAzE6dOnpRrDqZKNjQ1OnTolHp+nrjFdXFzg7++PqVOnwtTUVKOOm9G7hcU8ERHJdPPmTaxbtw5mZmZo3bo18vLy4ODggOfPn2Pjxo1Vbi61fv16PH/+HIcPHxbPQn7VvXv34OnpiYiICMyYMaPK+d29e1fqmq2tLR49eoRHjx5VOY6yvXpe+fr16wgJCcE333wDW1tb6Ojo4MqVK9iwYQO++eYbFWb5QnZ2NgRBgJ2dHTp16iRXc7L/cvr0aYSHh+PGjRv44Ycf8NNPP6FFixZvffNDklRaWsoCSUGePXsmPsfdsmVLXL9+HdbW1hg6dKjMPh2q1KBBA6xevRobN26EmZmZ1Oq3PP0dlBHz559/Rk5ODo4dOybzdVU/eCV6icU8ERHJVLduXejo6AAAzMzMkJ6ejp49e6Jt27a4detWleOcOnUKPj4+Mgt5ADAyMsKMGTMQGBhYrWJ++/btVb5XlTp16iT+eOXKlVi2bBn69esnvta6dWt88MEHWL16NUaOHKmKFMV++OEHHDlyBEePHsWtW7cwYMAAODk5oX379gqJHx8fj2nTpsHJyQl//vknKioqUF5ejgULFqC8vBzDhw9XyPuQ+tq9ezc2b96Me/fu4dixY9iyZQvef/99TJs2TdWpaazmzZsjLS0NxsbGMDMzExeaFRUVKC4uVnF2kvT09BT+4E4ZMauz+4xIlVjMExGRTNbW1oiOjsacOXNgYWGBX375BR4eHsjIyBAX+VXx4MEDWFpavvEeKysr/P333zVNWe1lZmbKnHvcokULtfj827dvj/bt28Pb2xuJiYk4cuQIpk6dinr16mHgwIEYOHCgePSVPNavX49Zs2Zh3Lhx4hWvmTNnQl9fH1u3bmUx/5aLjY3F2rVrMXbsWGzZsgXAixnh/v7+qFu3Ljw9PVWcoWYaNmwY5s6di5UrV8Le3h5ubm4wMTFBfHw8Pv74Y1WnJ6GqR56CgoLg7u6ORo0aqSRmVY8Eubm5wd/fX6HN94iqg8U8ERHJNHXqVHh4eMDAwADDhg1DSEgInJyc8Pfff2PgwIFVjlNWVoZ69eq98Z569eqhpKSkpimrvY8//hhRUVHw8fERb2F//vw5Nm3ahHbt2qk4O0l2dnaws7PD//73P1y4cAFxcXEYN24cDAwM4OTkJNc0g+vXr2P16tVS1/v374/g4GBFpP3O0JSeEa+KiIjAwoUL4eLigoiICADAmDFj0LBhQ4SGhrKYl9OECRMgEokgCAKsra0xbdo0hIaGwtjYWOZ/b5ogKioKw4cPr1LhrcqYycnJKC0tVVg8oupiMU9ERDJ9+umnOHbsGEpLS9G4cWPs2rULu3fvhrGxsdqdw9QUc+fOhYeHB86cOYNPPvkElZWVuHr1KkpKShAZGanq9GTS0tJCly5dYGxsDCMjI0RERCAkJESuYr5hw4bIzc1FixYtJK6np6cr9Bfsd4Gm9Ix4VVZWFuzs7KSu29nZ4d69eyrI6O0xbtw48ceenp4a/2BEGZOzOY2b3kYs5omI6LVe3Tpobm6O//3vf3LFiYiIeOPouSdPnsgVV9PY2dnh0KFD2LNnD9LT0wG82M751Vdf4YMPPlBxdtJu3ryJo0ePIi4uDmlpabCysoKnp2e1dma8ytnZGcuXL8fy5cshCAKKi4tx+vRpLF26VO6Y7ypN6RnxqiZNmuDGjRtSo/OSkpLU8vtfk1y4cAHJycl4+vSpVNHKfgREby8W80REJNOdO3fg7++P9PR0PHv2TOr1qo4mMzExQVxc3H/e97oGeW+b5s2bY9asWcjLy4NIJIK+vr6qU5LwsoA/cuQI0tLS0KpVKzg6OoonG9SEl5cX7t27Jz4b7+LigsrKSvTq1QteXl41T57U2pdffglfX1/Mnz8fAHDjxg2cOXMG69atk1hZpuoJCwtDQEAAGjZsKDXeUxAEFvNEbzEW80REJNPcuXPxzz//wNHRUWrUT3WcOnVKgVlpvp07dyI0NBQPHz4E8GK10sPDQy2KmSFDhiAtLQ0WFhb4/PPPERQUhJYtWyosvo6ODtauXYtvv/0WqampqKiogKWlJSwsLLgF9h3g6emJwsJCzJkzB8+ePcOkSZMgEokwcuRITJo0SdXpaazt27djxowZmDJliqpTIaJaxmKeiIhkSk1Nxc6dO9GmTRtVp/LWiImJwcqVKzF69GjY2dmhoqICCQkJCAgIgJ6eHkaMGKHS/K5fvw6RSITi4mLs27cP+/bte+29Vd2Z8ao+ffrgp59+gqmpKUxNTcXXc3NzMXjwYFy4cEGuvElzfPfdd5gyZQoyMjJQWVmJli1bQk9PT9VpabSCggI4OzurOg0iUgEW80REJNNHH330zpxlry3h4eHw9vbGqFGjxNf69esHU1NTREZGqryYV8Z23CNHjuDMmTMAgOzsbCxZskRqp0d2dra4uz+9XXJycmReNzQ0BPCiEC0oKADw4kgOVd+nn36KS5cuoVmzZqpO5Z3Dn1ukaizmiYhIpkWLFmHx4sVwc3NDs2bNoKWlJfF6x44dVZSZ5srJyUH37t2lrvfo0QOrVq1SQUaSlFHM29raIjo6WryNPicnBzo6OuLXBUFAgwYN1OLzJ8VzcHD4z4KnsrISgiAgNTW1lrLSfPv37xd/3LZtWyxatAhpaWkwMzODtra2xL1Dhw6t3eTe4OXXWt1jVue9iVSJxTwREcmUnp6OjIwMLFy4UOo1/uItHxMTEyQnJ0uNZrty5QqaNGmioqz+m7OzM8LCwuRqUmhsbIyoqCgAgJubG0JCQjiG7h3y8mtPivWyieCrNm/eLHVNEAS1KuaHDx+OFStWwMrK6o33LV++vMo/E5UR86WEhARkZmZi0KBBuHfvHkxNTSUeRh47dkytf3bT24/FPBERyRQSEoLhw4djzJgxqFevnqrTeSuMHDkSvr6+yM/PR4cOHSAIAhITExEcHAw3NzdVp/dad+/exfPnz2scRxPHqVHNdOrUSeb1/Px8aGtrS3Vfp6q5du2aqlOQS3Z2Nho0aPCf91VnVKUyYhYVFWHChAn4888/IQgCunXrBn9/f9y8eRPbtm2DkZERAHCkIqkci3kiIpLp8ePH8PT05DlMBRozZgyys7OxYsUKlJeXAwC0tbXh6uqKb775RsXZvZ6itrDevHkTvr6+uHjxIsrKyqRe526Pt9+WLVsQFRWFf/75BwDQrFkzeHp6wtXVVcWZUW3w9PTEwoUL4eHhgRYtWkg9KJanb4IyYgYEBAAATpw4gcGDBwN4MeFl9uzZWL16tfh1IlVjMU9ERDL17NkT58+fV3lTtrdJYmIi5s6dixkzZuDGjRsAoBHdvBV1LnTRokXIycnB7NmzuSL7DgoLC8P3338PNzc32NjYoLKyEhcvXsSKFStQWVmJL7/8UtUpkpL5+/sDeLF9/dWHhDXpm6CMmL/88gvWrl2L5s2bi6+1bNkSixYtwuTJk6sdj0hZWMwTEZFMnTp1wvLly3HmzBl89NFHEIkk/8lQRrO0t923336L8PBwtGnTBtbW1qpOp8o2b96MDz/8sMZxLl26hMjISNja2iogK9I0O3fuxOLFiyXOcPft2xfm5uYICwtjMf8OUEYPBWXEzMvLw/vvvy91XU9PDyUlJQp/PyJ5sZgnIiKZIiIi0LhxY1y9ehVXr16VeE0QBBbzcjA0NERhYaGq06iyoqIiHDlyBGlpaTh+/DjatGmDAQMGSI2Wq6rGjRtDV1dXwVmSpigoKED79u2lrtvZ2WHp0qUqyIhq26s9FPLy8iASiaCvr692Mdu1a4cjR45g0qRJEtejoqLwySef1Cg2kSIJlZypQEREVCv8/PwQHR0Ne3t7mJqaShXF6vSAJDMzE2PHjkVxcTHMzMxQUVGBW7du4cMPP0RkZKS4AVR1bNmyBYmJiVizZg232b+D5s2bB11dXfj4+EhcDwkJQVpaGoKDg1WUGdWmnTt3IjQ0FA8fPgQANGnSBB4eHhg3bpzaxExKSoK7uzu6dOmC+Ph4ODs7IyMjAykpKQgPD8dnn30md65EisRinoiIxHJycmBsbAxBEJCTk/PGe+VpKvSuc3BweO1rgiDg5MmTtZjNm7m7u0MkEsHf3188Si4vLw+zZ89GgwYNEBISUu2Ybm5u+PPPP1FeXg5DQ0PUqVNH4nV1+vxJ8fz8/LBr1y6Ym5ujY8eOEIlESE5ORmJiIvr06SPxgMfPz0+FmZKyxMTEYMmSJRg9ejTs7OxQUVGBhIQEREdHw8fHR64eLcqICQDXr19HeHg4UlJSUFFRgVatWmH8+PEyd5cQqQqLeSIiEmvdujXOnj0LQ0NDWFlZyexiXpOmQqQ5bG1tsWfPHrRq1Uri+rVr1zBq1CgkJSVVO+Z/PQBQp50JpHjVGb/IMYZvpwEDBmDMmDEYNWqUxPWdO3ciOjoasbGxahFz6dKlGDt2LFq0aFHtv0tUm3hmnoiIxCIjI8WrsMpoKvSuys3NxYkTJ1C3bl307NlTIc3klO2DDz7AvXv3pIr5oqIiuc+jslh/t7FAp5ycHHTv3l3qeo8ePbBq1Sq1ibl//364u7vL9XeJahOLeSIiEnu1kdCrH5P8EhMT4enpKe6ArKuri3Xr1sn85VOdzJs3D76+vpg/fz46deoEkUiEq1evwtfXF2PHjpU4hlGdIxd//fUXwsPDcf36dYhEIlhYWGDs2LEa1d2fqi4hIQG2trYQiURISEh47X2CIMDOzq4WMyNVMDExQXJystSK95UrV9CkSRO1idmrVy/s2LED06ZNU/vRofRu4zZ7IiKSqaSkBNu2bcPFixdRVlYmNWucK/dVM2bMGOjq6sLX1xfa2tpYsmQJMjMzcejQIVWn9kZWVlbij/89u/nlteoeuUhMTIS7uzssLS1hZ2eH8vJyJCUlIS0tDZGRkfj0008V+0mQyllZWSE+Pl7i6I6sXz15dOfdsG3bNoSGhmLGjBno0KEDBEFAYmIigoOD4ebmJtfuHWXE/Oqrr3Dp0iUIggBDQ0OpZqXs70HqgsU8ERHJNH/+fBw5cgQ9e/aU2XmcDaqqpmPHjti9ezcsLCwAvNhy36tXLyQkJKj1is8ff/xR5Xuruotj1KhRsLKykupm7uvri4yMDG7DfgtlZ2fD2NgYWlpayM7OfuO9TZs2raWsSFUqKirg5+eH3bt3o7y8HACgra0NV1dXLFy4ENra2moRk/09SFOwmCciIpk+/fRTLFu2DI6OjqpORaN98skn+O233yS2e1pbWyMuLk5jihdFzW5u37499u3bh5YtW0pcz8zMxIgRI3Dp0qUaxSf15u3tjYULF0o9xMrPz8fChQuxYcMGFWVGta2oqAg3btwAALRs2RJ6enp48OCB3NvilRWTSN3xzDwREcmkpaWFTz75RNVpaLyKigqpqQDa2tqoqKhQUUZVp+jZzY0bN8bDhw+livmHDx9Kjamjt8PFixdx584dAC+airVp00aqmM/MzMS5c+dUkR7VstatWyM+Ph4GBgYSfTLu3r0LZ2dnuR7oKSMm8OJ7V9YxM0EQMHXqVLliEikai3kiIpKpf//+2LdvH7y8vFSdCqlATEwMVq5cKTW7OSAgAHp6enLNbu7duzeWLl2KwMBAmJubAwAyMjKwfPly9O7dW9GfAqkBQRAwf/588cfLli2TuqdBgwbw8PCo7dSolvz44484ePAggBc9N6ZOnQodHR2Je+7fv1+tnT/KiPmqsLAwBAQEoFGjRtDV1ZV4jcU8qRMW80REJJO+vj4iIiJw+vRptGzZUmrllGfmqy4iIgL169cX//n58+eIiooSjwF8SZ3OYYaHh8Pb21tidnO/fv1gamqKyMhIuYp5Ly8vuLu7Y9CgQWjYsCEEQUBBQQEsLS0xd+5cRaZPaqJDhw64du0agBfN8M6ePcttz++Yvn374uLFi+I/GxkZoV69ehL3WFpaYujQoSqN+aodO3ZgypQpmDFjhlx/n6i28Mw8ERHJ5Obm9trXiouLsXfv3lrMRnM5ODhU6T5BENSqQ7K1tTUOHTokNe7p9u3bGDRoEK5cuSJX3IqKCpw5cwbp6emorKyEpaUlunfvLleTKiLSLK/rm6BuMdu3b4+DBw/C1NRUYTGJlIEr80REJJOszuIpKSnYvXs3Dh8+rIKMNNOpU6dUnYJcFD27uaSkBPXq1YOWlhbs7e1hb2+P9PR0NG3alIX8W8zb27vK93K3z9tv+fLlWL9+Pd5//33xrp9hw4ahX79+mDJlitrE/PTTT3H16lUW86T2WMwTEdEbPXv2DIcPH0Z0dDSuXr0KLS0t9O/fX9VpkZKNHDkSvr6+yM/Plzm7uTr2798PPz8/bNmyBe3atRNfX7lyJS5fvoylS5dyasJb6u7du6pOgdRIUFAQYmJisHTpUvG1wYMHIywsDFpaWpg0aZJaxHR0dMSSJUuQnJws85iZvNv3iRSN2+yJiEimGzduIDo6GgcOHMDjx48hCAKGDx+OyZMno1mzZqpOj5RMUbObf//9d3h4eGDYsGHw8vKSWNW/desWNm/ejH379iEyMhJ2dnZK+VyISD3Y29vDz88PXbt2lbh++vRp+Pr6yrWTSRkxraysXvuaIAhITU2tdkwiZeDKPBERiT1//hzHjx9HdHQ0EhISoKOjA3t7ezg6OmLu3LkYN24cC/l3hJaWFhYuXIgZM2bUaHbz5s2bMXr0aCxYsEDqNVNTUyxbtgyVlZXYuHEjtmzZotDPgdRLTk7OG183MTGppUxIVR4/fgxjY2Op66ampnjw4IHaxHzZtJFI3bGYJyIisV69eqGoqAidO3eGn58f+vbtK24qNGfOHBVnR7VJUbObU1JSxKPJXuerr77CxIkTa5QvqT8HBwcIgvDa17na+fazsrJCTEyM1PSKAwcOoFWrVmoT86WcnBxkZmaiY8eOKC4uhqGhYY3iESkai3kiIhIrLCyEoaEhjIyMoKurKzW3lxSntLRU6hymqiljdnNpaanUyKh/a9SoEZ4+fVr9hEmjREVFSfz5+fPnuHnzJrZu3YqFCxeqKCuqTdOnT4enpyeSkpJgY2MDQRBw9epV/Pnnn9iwYYPaxCwtLcW8efMQFxcHLS0tHDt2DKtWrUJhYSFCQkLQsGFDueISKRqLeSIiEouPj8eRI0fw008/ITo6Gg0aNICDgwMcHR3fuKJGVbd7925s3rwZ9+7dw7Fjx7Blyxa8//77ajFjXhmzmz/66CNcunRJqiv+q5KSktC0adNq50uapVOnTlLXunbtChMTE2zcuBG9evWq/aSoVnXr1g27d+9GVFQU4uPjIRKJYG5ujh9//PGN59RrO2ZoaCiuXbuGyMhITJ48GQAwZswYLFiwAGvWrMGSJUvkikukaGyAR0REMmVmZuLHH39EbGwsHjx4IG6AN2HCBJiZmak6PY0UGxsLX19fjB07Flu2bMGhQ4fwyy+/wN/fX7y6pC4UNbs5MjISUVFR2L17Nz744AOp1+/fv4+RI0di+PDhmDp1ao3eizTT3bt3MXDgQFy5ckXVqVAtuHbtGtLS0lBRUQHgxS6g0tJSXL58GStWrFBxdi/0798fixcvRteuXWFra4uDBw+iefPm+P333zFnzhycPXtW1SkSAWAxT0RE/6G8vBy//vor9u3bh19//RUVFRXo2rUrm5XJwcXFBWPGjIGLi4vEL4j79u1DaGgojh8/ruoUZcrLy0NiYiKaNGmCDh06VOvvlpeXY8yYMUhLS8OIESNgY2MDfX195Ofn488//8TevXthamqKHTt2/Od2fHr7FBUVISgoCKdPn8aJEydUnQ4pWVRUlLhgFwQBL8sQQRBgZ2eH7du3yxX35Sp6VlYW1q1bh59//hnm5ubo3LmzXPHat2+Pw4cPo1mzZhI/q2/evInBgwfzwROpDW6zJyKiN9LW1kafPn3Qp08f5OXl4cCBA9i7d6+q09JIWVlZMsev2dnZ4d69eyrISNqGDRsQFRWFPXv2wNTUFElJSZg4cSKKiooAAF26dEFoaGiVC29tbW1s3boVwcHBiImJwdatW8WvNWnSBKNGjcKUKVNYyL8DrKysZB7XEQRBYkY4vb127NiBSZMmYerUqejduzf27t2L/Px8zJo1C3369JErZnJyMr766ivY2NggOTkZpaWlSE1NxYoVKxASEoLevXtXO6a5uTnOnTsHV1dXieuHDh2ChYWFXHkSKQOLeSIiqjIDAwO4u7vD3d1d1alopCZNmuDGjRto3ry5xPWkpCSZW9Br2w8//IBNmzZh3Lhx4q7NCxYsQIMGDfDDDz9AT08P06dPx6ZNmzBjxowqx61Tpw5mz54NLy8v3LlzB48fP4aBgQGaN2/OXgzvkBUrVkh9vXV0dGBjY8ORl++InJwcjBgxAnXq1IGVlRWuXr2Kvn37Yv78+Vi5ciXGjRtX7Zj+/v4YP348Zs6cCVtbWwDAsmXL0LBhQ7mL+enTp8PLywtpaWkoLy/Hvn37cOPGDRw/fhyBgYHVjkekLCzmiYiIasmXX34JX19f8ai2Gzdu4MyZM1i3bp1cv8QqWkxMDObPn49Ro0YBAK5cuYKbN29i9uzZMDc3BwBMmTIFK1eurFYx/5JIJMJHH32k0JxJcwwbNkzVKZCK6erq4vnz5wAAMzMzZGRkoG/fvjA3N0d2drZcMZOTk7Fo0SKp61999RWio6Plitm7d2+EhIRg48aN0NbWRnh4OCwtLREUFIT+/fvLFZNIGVjMExER1RJPT08UFhZizpw5ePbsGSZNmgSRSISRI0di0qRJqk4PmZmZ6Nq1q/jP58+fhyAIsLe3F1+zsLBATk6OKtIjDRQSElLle9VhogMpl52dHTZu3AgfHx9YWVlhz549mDhxIhITE6GrqytXTB0dHfExoFfl5OSgfv36csV8+X3bpUsXdOnSRXw9PT0dN2/ehJGREXr27In33ntPrvhEisJinoiIqBZ99913mDJlCjIyMlBZWYmWLVvWuGO8Ir26DfrixYswMDBAq1atxNeKi4vl/gWZ3j0hISHQ0tKCkZHRG+8TBIHF/DvAy8sL7u7u2L17N7766iuEhoaiU6dOKCkpgYeHh1wx+/bti7Vr10psf8/MzMTy5cvlHneYkJCAhIQE6OjoiHcT3bp1C0+fPoWxsTHy8/NRt25dREVFSfx8JKptLOaJiIiU6HWr2C/PpBcUFKCgoAAAYGJiUmt5yfLxxx8jISEBpqamKCgowIULF/D5559L3BMXFwdLS0sVZUiaxtXVVdyl3snJCU5OTnLP/ibN16pVK/z888948uQJdHV1ERMTg9jYWBgZGWHAgAFyxZw3bx4mTJiArl27orKyEsOGDUNRURGsrKwwd+5cuWK2a9cOFRUVWLduHQwMDAAA+fn5mDNnDqytrTF58mT4+PjA398fmzZtkus9iBSBo+mIiIiU6HUdvF9VWVkJQRCQmppaS1nJFhsbCx8fH4wYMQKXLl1CSkoK9uzZg7Zt2+L+/fuIjY1FYGAgli9fjiFDhqg0V9Ic5eXlOH/+PA4fPoyTJ0/CwMAAgwYNgpOTE8zMzFSdHmm4oqIi6Onp4ffff0dKSgoqKipgaWmJHj16QEtLS66YXbp0QUREBFq3bi1x/dq1a3B3d8fvv/+OjIwMjBw5EomJiYr4NIjkwpV5IiIiJYqKilJ1ClXm7OyMZ8+eYffu3dDS0kJQUBDatm0LAAgLC0N0dDQ8PT1ZyFO1aGtro1u3bujWrRvKyspw9uxZxMXFYfjw4WjRogUGDhwIJycnle9MIc00ZMgQBAcHS51vr4nnz5+jrKxM6vqzZ8/w9OlTAC+mdHBNlFSNK/NEREQqkJ+fD21tbTRs2FDVqVRJbm4u6tSpg8aNG6s6FXpLlJaWIiYmBoGBgSguLlb5zhTSTN27d0dkZKR44oYizJw5E7dv30ZAQABMTU0BAFlZWZg7dy6MjIwQFBQEf39/XL58Gbt27VLY+xJVF4t5IiKiWrRlyxZERUXhn3/+AQA0a9YMnp6ecHV1VXFmRLUjNzcXcXFxOHr0KC5fvgxTU1M4OjrKNe6QKDQ0FLGxsfj666/RokUL1KtXT+L1jh07VjtmXl4eJk2ahOTkZOjr66OyshKFhYVo37491q9fj7/++gszZ87Epk2b0KlTJ0V9KkTVxmKeiIioloSFheH777+Hm5sbbGxsUFlZiYsXL2L37t3w9vbGl19+qeoUiZTi3wV88+bN4ejoCEdHRzbEoxp50/dPTXqRVFZW4sKFC0hNTYW2tjasrKzEhfujR48gEok0ZmcVvb1YzBMREdUSe3t7zJw5E0OHDpW4/uOPPyIsLAzHjx9XTWJESrJt2zYcPXoUV65cgYmJCRwdHTFgwAC0adNG1anRWyI7O/uNrzdt2rSWMiGqfSzmiYiIaomtrS327t0rnlv80s2bNzFkyBBcvnxZRZkRKYeVlRV0dHTQtWtXtGvX7o33cs48EVH1sJs9ERFRLenfvz+2b98OHx8fieuHDh2Cvb29irIiUp6XHerT09ORnp7+2vsEQWAxT1XWp08f/Pjjj2jcuDEcHBzeOP7z5MmTtZgZUe1iMU9ERFRL3nvvPezatQtJSUno2LEjRCIRkpOTkZiYiD59+sDb21t8r5+fnwozJVKMU6dOqToFegu5uLiIG90NGzZMxdkQqQ6LeSIiolqSkpICGxsbAMC1a9fE1+3s7PD48WM8fvxYRZkREWmOZs2a4dixY+KPid5VPDNPREREREQa498d7AVBQGVlJerXrw+RSITCwkJoa2ujcePGOHv2rIqyJFI+rswTEREpUUJCAmxtbSESiZCQkPDa+wRBgJ2dXS1mRkSkmV7d2XTkyBFs3rwZfn5+4iI/KysL3t7ecHJyUlWKRLWCK/NERERKZGVlhfj4eBgaGsLKykq8gvRvNZmHTET0rnJwcEBgYCDat28vcf3q1auYMmUKV+bprcaVeSIiIiU6efIkGjduLP6YiIgUJz8/H3Xr1pW6XlFRgadPn6ogI6Lao6XqBIiIiN5mTZs2hZbWi39uQ0JC0KhRIzRt2lTif7q6ulixYoWKMyUi0jyfffYZlixZgrt374qvZWZmwtfXF7169VJdYkS1gCvzRERESnTx4kXcuXMHALB//360adMGenp6EvdkZmbi3LlzqkiPiEijLV68GB4eHujXrx/09fUBAAUFBbC2tsb//d//qTg7IuXimXkiIiIlSkpKwqhRowDgteflGzRogPHjx2PatGm1nR4RkcYrLy/HuXPnkJ6eDgBo3bo1OnfuDEEQVJwZkXKxmCciIqolVlZWOHv2LJo0aaLqVIiIiEjDsZgnIiIiIiIi0jA8M09ERKRE3t7eVb7Xz89PiZkQERHR24TFPBERkRK92mGZiIiISFG4zZ6IiIiIiIhIw3BlnoiIqJbk5OS88XUTE5NayoSIiIg0HVfmiYiIaomVldUbRyWlpqbWYjZERESkybgyT0REVEuioqIk/vz8+XPcvHkTW7duxcKFC1WUFREREWkirswTERGp2K+//oqNGzciOjpa1akQERGRhtBSdQJERETvOgsLC6SkpKg6DSIiItIgLOaJiIhUqKioCNu2bcOHH36o6lSIiIhIg/DMPBERUS15XQM8QRCwdOlSFWREREREmopn5omIiGrJ3r17pYp5HR0d2NjYoFmzZirKioiIiDQRi3kiIiIiIiIiDcNt9kREREoUEhJS5XunTZumxEyIiIjobcKVeSIiIiWysrKClpYWjIyM3nifIAg4efJkLWVFREREmo4r80RERErk6uqKEydOAACcnJzg5OQEKysrFWdFREREmo4r80REREpWXl6O8+fP4/Dhwzh58iQMDAwwaNAgODk5wczMTNXpERERkQZiMU9ERFSLysrKcPbsWcTFxeHkyZNo0aIFBg4cCCcnJ5iYmKg6PSIiItIQLOaJiIhUpLS0FDExMQgMDERxcTFSU1NVnRIRERFpCJ6ZJyIiqmW5ubmIi4vD0aNHcfnyZZiamsLNzU3VaREREZEG4co8ERFRLfh3Ad+8eXM4OjrC0dGRDfGIiIio2ljMExERKdG2bdtw9OhRXLlyBSYmJnB0dMSAAQPQpk0bVadGREREGozFPBERkRJZWVlBR0cHXbt2Rbt27d5477Rp02opKyIiItJ0LOaJiIiUyMHBoUr3CYKAkydPKjkbIiIieluwmCciIiIiIiLSMFqqToCIiIiIiIiIqofFPBEREREREZGGYTFPREREREREpGFYzBMRERERERFpGBbzRERERERERBqGxTwRERERERGRhmExT0RERERERKRhWMwTERERERERaZj/B7hcKf3IiyYJAAAAAElFTkSuQmCC\n",
      "text/plain": [
       "<Figure size 1100x900 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "from string import ascii_letters\n",
    "import numpy as np\n",
    "import pandas as pd\n",
    "import seaborn as sns\n",
    "import matplotlib.pyplot as plt\n",
    "\n",
    "sns.set_theme(style=\"white\")\n",
    "\n",
    "temp = pd.concat([\n",
    "               pd.get_dummies(dataset['Rating']),\n",
    "               dataset['is_Franchise'].astype(int),\n",
    "               pd.get_dummies(dataset['Production Method']),\n",
    "               pd.get_dummies(dataset['Creative Type']),\n",
    "               dataset['has_star_actor'].astype(int),\n",
    "               dataset['has_star_director'].astype(int),\n",
    "               dataset[['Budget', 'actor_revenue_before', 'director_revenue_before',\n",
    "        'genre_revenue_before', 'ROI', 'Revenue']].copy()\n",
    "              ], axis = 1)\n",
    "\n",
    "# Compute the correlation matrix\n",
    "corr = temp.corr()\n",
    "\n",
    "# Generate a mask for the upper triangle\n",
    "mask = np.triu(np.ones_like(corr, dtype=bool))\n",
    "\n",
    "# Set up the matplotlib figure\n",
    "f, ax = plt.subplots(figsize=(11, 9))\n",
    "\n",
    "# Generate a custom diverging colormap\n",
    "cmap = sns.diverging_palette(230, 20, as_cmap=True)\n",
    "\n",
    "# Draw the heatmap with the mask and correct aspect ratio\n",
    "sns.heatmap(corr, mask=mask, cmap=cmap, vmax=.3, center=0,\n",
    "            square=True, linewidths=.5, cbar_kws={\"shrink\": .5})"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "7a445d8d-35ed-4125-b6ba-fdbd68e3de81",
   "metadata": {},
   "source": [
    "### Model Selection"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "3c6e77d0-6479-4616-b2a4-a02102d03403",
   "metadata": {},
   "source": [
    "From the histogram below, we can see that the response is hightly skewed. This ruled out most methods based on normality. Thus, tree based methods were selected for this project: mainly random forest and XGBoost."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 73,
   "id": "d5394cc1-17d8-4bcb-b1a9-bd367620aaa6",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Histogram of distribution of Revenue\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "print(\"Histogram of distribution of Revenue\")\n",
    "plt.hist(dataset['Revenue'], bins = 50)\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "a678d7a5-a789-415c-b032-68f054a0cae6",
   "metadata": {},
   "source": [
    "### Model Traning Process"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "b0699dab-48bb-473f-a097-96f4abd9f9f8",
   "metadata": {},
   "source": [
    "The following parts describe the procedure of model training:"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "664300cd-18d5-48d3-ae26-edadf63e36d5",
   "metadata": {},
   "source": [
    "#### Tuning Hyperparameters:\n",
    "In this part, there are in all 5 hyperparameters for tuning:\n",
    "- Length of time window to consider for actors, directors and genre: Here these three time windows are not required to be equal length.\n",
    "- Number of star actor and directors to be considered for each time window: these two are also not required to be the same.\n",
    "\n",
    "A grid search was performed while 3, 5, 10 were allowed for each of the 5 hyperparameters. In each iteration of the search, a 5-fold cross validation search on hyperparameters (max depth and features) of the model (random forest/XGBoost) was performed to find a model to estimate the performance with that set of hyperparameters. In this process, RMSE was used for comparison. "
   ]
  },
  {
   "cell_type": "markdown",
   "id": "6ebd50d0-42ab-4acf-94a0-20bb829a03f1",
   "metadata": {},
   "source": [
    "The following shows the example result of hyperparameters of both grid searches:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 85,
   "id": "5a4b988a-eba9-4a0c-8600-27f5397622f8",
   "metadata": {
    "tags": []
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Random Forest Grid Search Results\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>timewindow_actor</th>\n",
       "      <th>timewindow_director</th>\n",
       "      <th>timewindow_genre</th>\n",
       "      <th>star_actor_num</th>\n",
       "      <th>star_director_num</th>\n",
       "      <th>params</th>\n",
       "      <th>score</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>53</th>\n",
       "      <td>3</td>\n",
       "      <td>5</td>\n",
       "      <td>10</td>\n",
       "      <td>10</td>\n",
       "      <td>10</td>\n",
       "      <td>{'max_depth': 10, 'max_features': 'sqrt'}</td>\n",
       "      <td>-1.41</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>59</th>\n",
       "      <td>3</td>\n",
       "      <td>10</td>\n",
       "      <td>3</td>\n",
       "      <td>5</td>\n",
       "      <td>10</td>\n",
       "      <td>{'max_depth': 10, 'max_features': 'sqrt'}</td>\n",
       "      <td>-1.41</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "    timewindow_actor  timewindow_director  timewindow_genre  star_actor_num  \\\n",
       "53                 3                    5                10              10   \n",
       "59                 3                   10                 3               5   \n",
       "\n",
       "    star_director_num                                     params  score  \n",
       "53                 10  {'max_depth': 10, 'max_features': 'sqrt'}  -1.41  \n",
       "59                 10  {'max_depth': 10, 'max_features': 'sqrt'}  -1.41  "
      ]
     },
     "execution_count": 85,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "print(\"Random Forest Grid Search Results\")\n",
    "t = pd.read_csv('CV_rfc.csv').iloc[: , 1:]\n",
    "t.sort_values(by = 'score', ascending = False).head(2)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 86,
   "id": "3b8b9e4c-4c93-48bc-b491-9d6b8ef3dc99",
   "metadata": {
    "tags": []
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "XGBoost Grid Search Results\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>timewindow_actor</th>\n",
       "      <th>timewindow_director</th>\n",
       "      <th>timewindow_genre</th>\n",
       "      <th>star_actor_num</th>\n",
       "      <th>star_director_num</th>\n",
       "      <th>params</th>\n",
       "      <th>score</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>54</th>\n",
       "      <td>3</td>\n",
       "      <td>10</td>\n",
       "      <td>3</td>\n",
       "      <td>3</td>\n",
       "      <td>3</td>\n",
       "      <td>{'max_depth': 6}</td>\n",
       "      <td>-1.38</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>43</th>\n",
       "      <td>3</td>\n",
       "      <td>5</td>\n",
       "      <td>5</td>\n",
       "      <td>10</td>\n",
       "      <td>5</td>\n",
       "      <td>{'max_depth': 6}</td>\n",
       "      <td>-1.38</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "    timewindow_actor  timewindow_director  timewindow_genre  star_actor_num  \\\n",
       "54                 3                   10                 3               3   \n",
       "43                 3                    5                 5              10   \n",
       "\n",
       "    star_director_num            params  score  \n",
       "54                  3  {'max_depth': 6}  -1.38  \n",
       "43                  5  {'max_depth': 6}  -1.38  "
      ]
     },
     "execution_count": 86,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "print(\"XGBoost Grid Search Results\")\n",
    "t = pd.read_csv('CV_xgb.csv').iloc[: , 1:]\n",
    "t.sort_values(by = 'score', ascending = False).head(2)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "c5d4485f-4ada-4161-80ec-3965860e89f5",
   "metadata": {
    "tags": []
   },
   "outputs": [],
   "source": [
    "### Grid search with random forest\n",
    "my_super_search = pd.DataFrame(columns = [\"timewindow_actor\",\n",
    "                                          \"timewindow_director\",\n",
    "                                          \"timewindow_genre\",\n",
    "                                          \"star_actor_num\",\n",
    "                                          \"star_director_num\",\n",
    "                                          \"params\",\n",
    "                                          \"score\"])\n",
    "\n",
    "import itertools\n",
    "tuning_list = {\n",
    "    \"timewindow_actor\": [3, 5, 10],\n",
    "    \"timewindow_director\": [3, 5, 10],\n",
    "    \"timewindow_genre\": [3, 5, 10],\n",
    "    \"star_actor_num\": [3, 5, 10],\n",
    "    \"star_director_num\": [3, 5, 10],\n",
    "    }\n",
    "\n",
    "# tuning_list = {\n",
    "#     \"timewindow_actor\": [3],\n",
    "#     \"timewindow_director\": [3],\n",
    "#     \"timewindow_genre\": [3],\n",
    "#     \"star_actor_num\": [3],\n",
    "#     \"star_director_num\": [3, 5],\n",
    "#     }\n",
    "tuning_grid = list(itertools.product(*tuning_list.values()))\n",
    "\n",
    "for s in tqdm(tuning_grid):\n",
    "    timewindow_actor = s[0]\n",
    "    timewindow_director = s[1]\n",
    "    timewindow_genre = s[2]\n",
    "    star_actor_num = s[3]\n",
    "    star_director_num = s[4] \n",
    "\n",
    "    dataset = castinfo(timewindow_actor, star_actor_num, dataset).copy()\n",
    "    dataset = directorinfo(timewindow_director, star_director_num, dataset).copy()\n",
    "    dataset = genreinfo(timewindow_genre, dataset).copy()\n",
    "\n",
    "    final = dataset.loc[dataset['Year'] >= 1980].copy()\n",
    "    X = pd.concat([pd.get_dummies(final[\"Month\"]), \n",
    "                   pd.get_dummies(final['Rating']),\n",
    "                   final['is_Franchise'].astype(int),\n",
    "                   pd.get_dummies(final['Production Method']),\n",
    "                   pd.get_dummies(final['Creative Type']),\n",
    "                   final['has_star_actor'].astype(int),\n",
    "                   final['has_star_director'].astype(int),\n",
    "                   final[['Budget', 'actor_revenue_before', 'director_revenue_before',\n",
    "            'genre_revenue_before']].copy()\n",
    "                  ], axis = 1)\n",
    "    Y = np.log(final['Revenue'])\n",
    "        \n",
    "    \n",
    "    rfc = RandomForestRegressor(random_state=42)\n",
    "    param_grid = { \n",
    "        'max_depth' : [4, 6, 8, 10],\n",
    "        'max_features' :['sqrt', 'log2', None]\n",
    "    }\n",
    "    CV_rfc = GridSearchCV(estimator=rfc, param_grid = param_grid, cv= 5,\n",
    "                         scoring = 'neg_root_mean_squared_error')\n",
    "    CV_rfc.fit(X, Y)\n",
    "    \n",
    "    my_super_search.loc[len(my_super_search)] = list(s) + [CV_rfc.best_params_, CV_rfc.best_score_]\n",
    "\n",
    "    my_super_search.to_csv('CV_rfc.csv')\n",
    "    \n",
    "    best_rf = my_super_search.sort_values(by = \"score\", ascending = False).head(1)\n",
    "    \n",
    "    best_rf = best_rf.values.flatten()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "b15dc681-3461-4217-b03d-e51b7dd5955f",
   "metadata": {
    "tags": []
   },
   "outputs": [],
   "source": [
    "### Grid search for XGBoost\n",
    "import xgboost as xgb\n",
    "my_super_search2 = pd.DataFrame(columns = [\"timewindow_actor\",\n",
    "                                          \"timewindow_director\",\n",
    "                                          \"timewindow_genre\",\n",
    "                                          \"star_actor_num\",\n",
    "                                          \"star_director_num\",\n",
    "                                          \"params\",\n",
    "                                          \"score\"])\n",
    "\n",
    "import itertools\n",
    "tuning_list = {\n",
    "    \"timewindow_actor\": [3, 5, 10],\n",
    "    \"timewindow_director\": [3, 5, 10],\n",
    "    \"timewindow_genre\": [3, 5, 10],\n",
    "    \"star_actor_num\": [3, 5, 10],\n",
    "    \"star_director_num\": [3, 5, 10],\n",
    "    }\n",
    "\n",
    "# tuning_list = {\n",
    "#     \"timewindow_actor\": [3],\n",
    "#     \"timewindow_director\": [3],\n",
    "#     \"timewindow_genre\": [3],\n",
    "#     \"star_actor_num\": [3],\n",
    "#     \"star_director_num\": [3, 5],\n",
    "#     }\n",
    "tuning_grid = list(itertools.product(*tuning_list.values()))\n",
    "\n",
    "for s in tqdm(tuning_grid):\n",
    "    timewindow_actor = s[0]\n",
    "    timewindow_director = s[1]\n",
    "    timewindow_genre = s[2]\n",
    "    star_actor_num = s[3]\n",
    "    star_director_num = s[4] \n",
    "\n",
    "    dataset = castinfo(timewindow_actor, star_actor_num, dataset).copy()\n",
    "    dataset = directorinfo(timewindow_director, star_director_num, dataset).copy()\n",
    "    dataset = genreinfo(timewindow_genre, dataset).copy()\n",
    "\n",
    "    final = dataset.loc[dataset['Year'] >= 1980].copy()\n",
    "    X = pd.concat([pd.get_dummies(final[\"Month\"]), \n",
    "                   pd.get_dummies(final['Rating']),\n",
    "                   final['is_Franchise'].astype(int),\n",
    "                   pd.get_dummies(final['Production Method']),\n",
    "                   pd.get_dummies(final['Creative Type']),\n",
    "                   final['has_star_actor'].astype(int),\n",
    "                   final['has_star_director'].astype(int),\n",
    "                   final[['Budget', 'actor_revenue_before', 'director_revenue_before',\n",
    "            'genre_revenue_before']].copy()\n",
    "                  ], axis = 1)\n",
    "    Y = np.log(final['Revenue'])\n",
    "        \n",
    "    xgb_model = xgb.XGBRegressor(seed = 42)\n",
    "    \n",
    "    param_grid = { \n",
    "        'max_depth' : [6, 10, 14],\n",
    "    }\n",
    "    CV_xgb = GridSearchCV(estimator=xgb_model, param_grid = param_grid, cv= 5,\n",
    "                         scoring = 'neg_root_mean_squared_error')\n",
    "    CV_xgb.fit(X, Y)\n",
    "    \n",
    "    my_super_search2.loc[len(my_super_search2)] = list(s) + [CV_xgb.best_params_, CV_xgb.best_score_]\n"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "43bee9ff-fea3-498b-928a-00bb1e196fb2",
   "metadata": {},
   "source": [
    "Due to the fact that XGBoost out performed random Forest model for all set of parameters, It was chosen as the final model. Then, for the best of hyperparameters, another 5 fold cross validation was used to fine tune each of the model (XGBoost). Here, RMSE is again used for comparison."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "9f8e9f77-b680-4660-9e97-45c08a6295f8",
   "metadata": {},
   "outputs": [],
   "source": [
    "timewindow_actor = 3\n",
    "timewindow_director = 10\n",
    "timewindow_genre = 3\n",
    "star_actor_num = 3\n",
    "star_director_num = 3\n",
    "\n",
    "dataset = castinfo(timewindow_actor, star_actor_num, dataset).copy()\n",
    "dataset = directorinfo(timewindow_director, star_director_num, dataset).copy()\n",
    "dataset = genreinfo(timewindow_genre, dataset).copy()\n",
    "\n",
    "final = dataset.loc[dataset['Year'] >= 1980].copy()\n",
    "X = pd.concat([pd.get_dummies(final[\"Month\"]), \n",
    "               pd.get_dummies(final['Rating']),\n",
    "               final['is_Franchise'].astype(int),\n",
    "               pd.get_dummies(final['Production Method']),\n",
    "               pd.get_dummies(final['Creative Type']),\n",
    "               final['has_star_actor'].astype(int),\n",
    "               final['has_star_director'].astype(int),\n",
    "               final[['Budget', 'actor_revenue_before', 'director_revenue_before',\n",
    "        'genre_revenue_before']].copy()\n",
    "              ], axis = 1)\n",
    "Y = np.log(final['Revenue'])\n",
    "\n",
    "xgb_model = xgb.XGBRegressor(seed = 42)\n",
    "param_grid = { \n",
    "    'max_depth' : [4, 6, 8, 10],\n",
    "}\n",
    "CV_xgb = GridSearchCV(estimator=xgb_model, param_grid = param_grid, cv= 5,\n",
    "                     scoring = 'neg_root_mean_squared_error')\n",
    "CV_xgb.fit(X, Y)\n",
    "\n",
    "print(\"Best parameters: \", CV_xgb.best_params_)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 274,
   "id": "6c99e297-511d-4e61-b699-2f66ccf118be",
   "metadata": {},
   "outputs": [],
   "source": [
    "timewindow_actor = 3\n",
    "timewindow_director = 10\n",
    "timewindow_genre = 3\n",
    "star_actor_num = 3\n",
    "star_director_num = 3\n",
    "\n",
    "dataset = castinfo(timewindow_actor, star_actor_num, dataset).copy()\n",
    "dataset = directorinfo(timewindow_director, star_director_num, dataset).copy()\n",
    "dataset = genreinfo(timewindow_genre, dataset).copy()\n",
    "\n",
    "final = dataset.loc[dataset['Year'] >= 1980].copy()\n",
    "X = pd.concat([pd.get_dummies(final[\"Month\"]), \n",
    "               pd.get_dummies(final['Rating']),\n",
    "               final['is_Franchise'].astype(int),\n",
    "               pd.get_dummies(final['Production Method']),\n",
    "               pd.get_dummies(final['Creative Type']),\n",
    "               final['has_star_actor'].astype(int),\n",
    "               final['has_star_director'].astype(int),\n",
    "               final[['Budget', 'actor_revenue_before', 'director_revenue_before',\n",
    "        'genre_revenue_before']].copy()\n",
    "              ], axis = 1)\n",
    "Y = np.log(final['Revenue'])\n",
    "\n",
    "X_train, X_test, y_train, y_test = train_test_split(\n",
    "    X,\n",
    "    Y, test_size = .2, random_state = 10) #split the data\n",
    "\n",
    "xgb_model = xgb.XGBRegressor(max_depth = 4, seed = 42)\n",
    "xgb_model.fit(X_train, y_train)\n",
    "\n",
    "y_pred = xgb_model.predict(X_test)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "cda0cb8b-aca8-4e33-9133-9cb4a4f67959",
   "metadata": {},
   "source": [
    "## Results and Insights"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "6bdc0ff3-f98d-49d1-b080-aad9a3c61eae",
   "metadata": {},
   "source": [
    "Firstly, below provides the descriptive results for the model, and this part will discuss further interpretations. "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 275,
   "id": "d348f808-356b-43d4-afb1-dbe5bbc08710",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "r2 score:  0.5738890076632135\n",
      "MSE:  1.3877082380000316\n"
     ]
    }
   ],
   "source": [
    "from sklearn.metrics import mean_squared_error\n",
    "print(\"r2 score: \", r2_score(y_pred, y_test))\n",
    "print(\"MSE: \", mean_squared_error(y_pred, y_test))"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "51f77186-46bf-4e08-abfb-b48b8c4a8ab0",
   "metadata": {},
   "source": [
    "### Comparison with alternative models:"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "31ee587b-18d5-46fe-8809-38999f960972",
   "metadata": {},
   "source": [
    "In order to compare and measure the progress in this model, alternative models were produced."
   ]
  },
  {
   "cell_type": "markdown",
   "id": "ba4e8a1f-625c-4809-8dee-f46f31bba8f8",
   "metadata": {},
   "source": [
    "### XGBoost model without feature engineering"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "e3ead9d6-c7ef-4f91-935c-73169bd772e8",
   "metadata": {},
   "source": [
    "This part below shows the result of an alternative model: This model is trained with XGBoost without all the features engineered in this project. Instead, the original ones were used in training. The great boost on r2 score and decrease MSE validates the importance of the features created in this analysis. "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 283,
   "id": "8b9731db-ff1a-4572-85d1-7a6253a77f84",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "r2 score:  0.49796386164252404\n",
      "MSE:  1.6210515813363064\n"
     ]
    }
   ],
   "source": [
    "final = dataset.loc[dataset['Year'] >= 1980].copy()\n",
    "X = pd.concat([pd.get_dummies(final[\"Month\"]), \n",
    "               pd.get_dummies(final['Rating']),\n",
    "               pd.get_dummies(final['Production Method']),\n",
    "               pd.get_dummies(final['Creative Type']),\n",
    "               pd.get_dummies(final['Genre']),\n",
    "               final[['Budget', 'Year']].copy()\n",
    "              ], axis = 1)\n",
    "Y = np.log(final['Revenue'])\n",
    "\n",
    "X_train, X_test, y_train, y_test = train_test_split(\n",
    "    X,\n",
    "    Y, test_size = .2, random_state = 10) #split the data\n",
    "\n",
    "xgb_model = xgb.XGBRegressor(seed = 42)\n",
    "xgb_model.fit(X_train, y_train)\n",
    "\n",
    "y_pred = xgb_model.predict(X_test)\n",
    "\n",
    "print(\"r2 score: \", r2_score(y_pred, y_test))\n",
    "print(\"MSE: \", mean_squared_error(y_pred, y_test))"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "ccec8d5b-5fc2-4a7f-987f-daabf18b8413",
   "metadata": {},
   "source": [
    "### Discussion on feature importance"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "03ea2beb-13fb-4946-ae8d-dab8439abf35",
   "metadata": {},
   "source": [
    "As shown in the graph below, certain features are deemed important by the tree predictor. And the only one categorical feature being if the movie belongs to a franchise. In order to validate its importance, interaction analysis were conduct around this feature."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 276,
   "id": "01a83246-589b-4487-b207-0ab2ba418155",
   "metadata": {
    "tags": []
   },
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "from sklearn.inspection import permutation_importance\n",
    "\n",
    "result = permutation_importance(\n",
    "    xgb_model, X_test, y_test, n_repeats=10, random_state=10, n_jobs=2\n",
    ")\n",
    "forest_importances = pd.Series(result.importances_mean, index = list(X_train.columns.values))\n",
    "fig, ax = plt.subplots()\n",
    "forest_importances.plot.bar(yerr=result.importances_std, ax=ax)\n",
    "ax.set_title(\"Feature importances using permutation on full model\")\n",
    "ax.set_ylabel(\"Mean accuracy decrease\")\n",
    "fig.tight_layout()\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "7b0fe52f-7a9a-4a2c-8975-98025f800e7c",
   "metadata": {},
   "source": [
    "The graph below provides comparision for the two different groups. Revenue were plotted against the important numerical features in this project, and the two colors denote different groups. It is easy to see, in all four plots, movies that belongs to a franchise have a line above the ones that don't: this means that franchise movies produce a much higher revenue given the same circumstances compared to non-franchise ones. "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 291,
   "id": "b707c4da-15ed-48da-9b96-652bfd2feff5",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1095.11x250 with 4 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "import seaborn\n",
    "seaborn.pairplot(final, x_vars = ['Budget', 'actor_revenue_before', 'director_revenue_before',\n",
    "        'genre_revenue_before'], y_vars = ['Revenue'], kind = 'reg', hue = 'is_Franchise')\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "a860c090-32a4-4796-8272-b0d16988f2fc",
   "metadata": {},
   "source": [
    "To further support this argument, a linear regression was run on the whole dataset, Franchise movies, and non franchise movies to compare the results. "
   ]
  },
  {
   "cell_type": "markdown",
   "id": "f7debdab-1a13-48d0-b84a-bcc4a4714717",
   "metadata": {},
   "source": [
    "On the whole dataset:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 303,
   "id": "80ea31b9-1c31-4a68-81c6-9e4ab2f507c8",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "r2 score:  0.1260607791334808\n",
      "MSE:  2.023095142953555\n"
     ]
    }
   ],
   "source": [
    "final = dataset.loc[dataset['Year'] >= 1980].copy()\n",
    "X = pd.concat([pd.get_dummies(final[\"Month\"]), \n",
    "               pd.get_dummies(final['Rating']),\n",
    "               pd.get_dummies(final['Production Method']),\n",
    "               pd.get_dummies(final['Creative Type']),\n",
    "               final['is_Franchise'].astype(int),\n",
    "               final['has_star_actor'].astype(int),\n",
    "               final['has_star_director'].astype(int),\n",
    "               final[['Budget', 'actor_revenue_before', 'director_revenue_before',\n",
    "        'genre_revenue_before']].copy()\n",
    "              ], axis = 1)\n",
    "Y = np.log(final['Revenue'])\n",
    "\n",
    "\n",
    "X_train, X_test, y_train, y_test = train_test_split(\n",
    "    X,\n",
    "    Y, test_size = .2, random_state = 10) #split the data\n",
    "\n",
    "\n",
    "regressor = LinearRegression()\n",
    "regressor.fit(X_train, y_train)\n",
    "\n",
    "y_pred = regressor.predict(X_test)\n",
    "\n",
    "    \n",
    "print(\"r2 score: \", r2_score(y_pred, y_test))\n",
    "print(\"MSE: \", mean_squared_error(y_pred, y_test))"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "c7232bc4-410c-4ce6-a2ee-2f0bbfab1657",
   "metadata": {},
   "source": [
    "On Franchise movies:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 299,
   "id": "ff00d1c6-e282-4cb9-a724-30d4e38ce693",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "r2 score:  0.37567420925092965\n",
      "MSE:  0.844808891064326\n"
     ]
    }
   ],
   "source": [
    "final = dataset.loc[dataset['Year'] >= 1980].copy()\n",
    "final = final.loc[final['is_Franchise']].copy()\n",
    "X = pd.concat([pd.get_dummies(final[\"Month\"]), \n",
    "               pd.get_dummies(final['Rating']),\n",
    "               pd.get_dummies(final['Production Method']),\n",
    "               pd.get_dummies(final['Creative Type']),\n",
    "               final['has_star_actor'].astype(int),\n",
    "               final['has_star_director'].astype(int),\n",
    "               final[['Budget', 'actor_revenue_before', 'director_revenue_before',\n",
    "        'genre_revenue_before']].copy()\n",
    "              ], axis = 1)\n",
    "Y = np.log(final['Revenue'])\n",
    "\n",
    "X_train, X_test, y_train, y_test = train_test_split(\n",
    "    X,\n",
    "    Y, test_size = .2, random_state = 10) #split the data\n",
    "\n",
    "\n",
    "regressor = LinearRegression()\n",
    "regressor.fit(X_train, y_train)\n",
    "\n",
    "y_pred = regressor.predict(X_test)\n",
    "\n",
    "print(\"r2 score: \", r2_score(y_pred, y_test))\n",
    "print(\"MSE: \", mean_squared_error(y_pred, y_test))"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "67c1baa4-b275-44be-8d7c-0ce8e207a611",
   "metadata": {},
   "source": [
    "On non-franchise movies:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 302,
   "id": "5c9a1b83-3e63-432f-ae73-f1200f47b68f",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "r2 score:  -0.28570693372472533\n",
      "MSE:  2.367045898228619\n"
     ]
    }
   ],
   "source": [
    "final = dataset.loc[dataset['Year'] >= 1980].copy()\n",
    "final = final.loc[final['is_Franchise'] == False].copy()\n",
    "X = pd.concat([pd.get_dummies(final[\"Month\"]), \n",
    "               pd.get_dummies(final['Rating']),\n",
    "               pd.get_dummies(final['Production Method']),\n",
    "               pd.get_dummies(final['Creative Type']),\n",
    "               final['has_star_actor'].astype(int),\n",
    "               final['has_star_director'].astype(int),\n",
    "               final[['Budget', 'actor_revenue_before', 'director_revenue_before',\n",
    "        'genre_revenue_before']].copy()\n",
    "              ], axis = 1)\n",
    "Y = np.log(final['Revenue'])\n",
    "\n",
    "\n",
    "X_train, X_test, y_train, y_test = train_test_split(\n",
    "    X,\n",
    "    Y, test_size = .2, random_state = 10) #split the data\n",
    "\n",
    "\n",
    "regressor = LinearRegression()\n",
    "regressor.fit(X_train, y_train)\n",
    "\n",
    "y_pred = regressor.predict(X_test)\n",
    "\n",
    "print(\"r2 score: \", r2_score(y_pred, y_test))\n",
    "print(\"MSE: \", mean_squared_error(y_pred, y_test))"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "64b42032-b6d0-4a52-9132-b05fc7102b8f",
   "metadata": {},
   "source": [
    "The great difference in the latter two models validates the fact the two population performs really greatly. The boost in performance on Franchise movies means there is a linear correlation between features and confirms our findings. The bad result on non-franchise movies suggest they need more study: either they don't follow the linear trend line franchise movies, or there are other factors that plays the important part in their successes. Either way, this confirms our hypothesis that these two population behaves differetly."
   ]
  },
  {
   "cell_type": "markdown",
   "id": "ac19998e-ba95-4b9c-8247-a4476ba3a76c",
   "metadata": {},
   "source": [
    "## Conclusions\n",
    "In all, this project tried to find out important factors that lead to the success of a movie, and tried to predict its revenue. Some conclusions can be drawn from these results:\n",
    "- The current trend in movie industry is that large budget movie and franchise movies tend to get larger success. Larger budget also usually means more money is used to hire more famous actors and/or directors, which are also important in our predictions.\n",
    "- Franchise and non-franchise movies behave very differently in this industry. From the information this project has, it is hard to find better explanation or predictions on this matter. However, other information not included in this dataset (for instance, amount of money spent on advertisements, etc.) or other techniques (for instance, NLP analysis on keywords or titles, etc.) may introduce new insights into this project. \n",
    "- The feature engineering in this project was successful and proved to be useful. It shows that it is possible to predict (to some extent) the success of a movie based on past performance of its cast and directors, and the average performance of movies of the same genre. Further analysis may dig deeper into how the trends shift through years, and may introduce deeper understanding of this matter."
   ]
  },
  {
   "cell_type": "markdown",
   "id": "0934a02d-3898-472d-aaa9-ab6a0c9d05a4",
   "metadata": {
    "tags": []
   },
   "source": [
    "## Bibliography"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "d9fb355a-3ecf-4369-a629-17834e1dea08",
   "metadata": {
    "tags": []
   },
   "source": [
    "Banik, Rounak. “The Movies Dataset.” Kaggle, ROUNAK BANIK, 10 Nov. 2017, https://www.kaggle.com/datasets/rounakbanik/the-movies-dataset?select=movies_metadata.csv. "
   ]
  },
  {
   "cell_type": "markdown",
   "id": "b1b469da-f951-4510-91d7-700813a81e00",
   "metadata": {},
   "source": [
    "De Vany, Arthur, and W. David Walls. “Uncertainty in the Movie Industry: Does Star Power Reduce the Terror of the Box Office?” Journal of Cultural Economics, vol. 23, no. 4, Nov. 1999, pp. 285–318., https://doi.org/10.1023/a:1007608125988. "
   ]
  },
  {
   "cell_type": "markdown",
   "id": "0adbfb81-521a-44d2-940a-a63c958fb0a4",
   "metadata": {},
   "source": [
    "De Vany, Arthur, and W. David Walls. “Bose-Einstein Dynamics and Adaptive Contracting in the Motion Picture Industry.” The Economic Journal, vol. 106, no. 439, 1996, pp. 1493–514. JSTOR, https://doi.org/10.2307/2235197. Accessed 6 Mar. 2023."
   ]
  },
  {
   "cell_type": "markdown",
   "id": "aa5b639c-ba1e-4af0-afb8-112e2e6a7e4b",
   "metadata": {},
   "source": [
    "Walls, W. David. “Modeling Movie Success When 'Nobody Knows Anything': Conditional Stable-Distribution Analysis of Film Returns - Journal of Cultural Economics.” SpringerLink, Kluwer Academic Publishers, https://link.springer.com/article/10.1007/s10824-005-1156-5. "
   ]
  },
  {
   "cell_type": "markdown",
   "id": "edb96fa0-d69c-4304-a486-6561a3fe148e",
   "metadata": {},
   "source": [
    "“Movie Budgets, Most Expensive Movies, Most Profitable Movies, Biggest Money-Losing Movies.” The Numbers, www.the-numbers.com/movie/budgets. Accessed March 2023. "
   ]
  },
  {
   "cell_type": "markdown",
   "id": "4653514f-6966-4efe-b0ee-470944f25c9c",
   "metadata": {},
   "source": [
    "## Appendix\n"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "73b3598d-ecfa-4e3d-a456-8b002fd4149d",
   "metadata": {},
   "source": [
    "All code used in this project are stored on github: https://github.com/kongwei981126/Movie-Industry/tree/Wei\n",
    "\n",
    "For all code used in this report, please access through the original file: https://github.com/kongwei981126/Movie-Industry/blob/Wei/Final%20Report.ipynb\n",
    "\n",
    "For references mentioned in this report, please access through the following files:\n",
    "- For discussion on the original dataset: https://github.com/kongwei981126/Movie-Industry/blob/Wei/midReport.ipynb\n",
    "\n",
    "- For creation of the dataset: https://github.com/kongwei981126/Movie-Industry/blob/Wei/scraping%20info.ipynb and https://github.com/kongwei981126/Movie-Industry/blob/Wei/Scraping%20more.ipynb\n",
    "\n",
    "- For descriptive analysis: https://github.com/kongwei981126/Movie-Industry/blob/Wei/Descriptive.ipynb\n",
    "\n",
    "- For discussion on ROI: https://github.com/kongwei981126/Movie-Industry/blob/Wei/passMidReport.ipynb\n",
    "\n",
    "- Other code references for this anaysis: https://github.com/kongwei981126/Movie-Industry/blob/Wei/Goodbye_ROI.ipynb"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3 (ipykernel)",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.9.13"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}