notebooks/eda-on-the-evalrs-dataset.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "nkF_4ePZHNzg"
   },
   "source": [
    "# Exploring the EvalRS dataset\n",
    "\n",
    "This notebook provides some basic EDA on the dataset of the [EvalRS 2023](https://reclist.io/kdd2023-cup/) hackathon.\n",
    "Please refer to [our paper](https://arxiv.org/abs/2304.07145) for all the details.\n",
    "\n",
    "With the goal of building a *all-around* recommender system, we likely want to understand how users, tracks and listening events are distributed, with a special case to *long tails* and rare (or new events). Yeah, you can thank us for the tip later.\n",
    "\n",
    "Before we start, we need to check we are inside the right virtual environment"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {
    "execution": {
     "iopub.execute_input": "2022-08-04T18:34:13.645302Z",
     "iopub.status.busy": "2022-08-04T18:34:13.644435Z",
     "iopub.status.idle": "2022-08-04T18:35:00.150666Z",
     "shell.execute_reply": "2022-08-04T18:35:00.148978Z",
     "shell.execute_reply.started": "2022-08-04T18:34:13.645133Z"
    },
    "id": "dljiNmSJjr-u"
   },
   "outputs": [],
   "source": [
    "%%capture\n",
    "# check we are using the right interpreter with the right RecList version\n",
    "!which python\n",
    "!pip install -r ../requirements.txt"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {
    "id": "Mv-L2X8Fj0fw",
    "tags": []
   },
   "outputs": [],
   "source": [
    "import os\n",
    "import sys\n",
    "sys.path.insert(0, '../evaluation')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 89,
   "metadata": {
    "id": "_-FxCXfaoFNU",
    "tags": []
   },
   "outputs": [],
   "source": [
    "from EvalRSRunner import ChallengeDataset\n",
    "\n",
    "import gc\n",
    "import pandas as pd\n",
    "import numpy as np\n",
    "import os\n",
    "import matplotlib.pyplot as plt\n",
    "import seaborn as sns\n",
    "import pyarrow.parquet as pq\n",
    "\n",
    "sns.set_theme(\"notebook\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Next, we have to download the dataset (~1.6GB) -- actually, you'll download it just the first time!"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {
    "id": "hij5x-hOoHKx",
    "outputId": "857b78b1-e056-4836-8442-877deb0e721f"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "/Users/mala/Library/Caches/evalrs/evalrs_dataset_KDD_2023\n",
      "Downloading LFM dataset...\n",
      "Downloading to /Users/mala/Library/Caches/evalrs/evalrs_dataset.zip...\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "evalrs_dataset_KDD_2023.zip: 100%|██████████████████████████████████████████████████████████████████████████████████████████████████████████████| 1.60G/1.60G [14:55<00:00, 1.92MB/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Loading dataset.\n",
      "Generating Train/Test Split.\n",
      "Generating dataset hashes.\n"
     ]
    }
   ],
   "source": [
    "dataset = ChallengeDataset()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "And get a sample fold (see details on our Bootstrapping Cross-Validation evaluation strategy to know more about folds)."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {
    "id": "K-UJp05VoJVD",
    "outputId": "6d6f97fc-dcf8-4904-86b5-5d6853945ac0"
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "((6902066, 6), (29725, 2))"
      ]
     },
     "execution_count": 4,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "train, test = dataset.get_sample_train_test()\n",
    "train.shape, test.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {
    "id": "n5z3WQ1iQF_Z",
    "outputId": "71918ba2-f965-44fd-9068-7bd28cec7d53"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "<class 'pandas.core.frame.DataFrame'>\n",
      "Index: 6902066 entries, 751 to 37926116\n",
      "Data columns (total 6 columns):\n",
      " #   Column            Dtype\n",
      "---  ------            -----\n",
      " 0   user_id           int64\n",
      " 1   artist_id         int64\n",
      " 2   album_id          int64\n",
      " 3   track_id          int64\n",
      " 4   timestamp         int64\n",
      " 5   user_track_count  int64\n",
      "dtypes: int64(6)\n",
      "memory usage: 368.6 MB\n"
     ]
    }
   ],
   "source": [
    "train.info()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "So. We have ~7M training listening records. Each tells us who (*user_id*) listened to what (*track_id*), when (*timestamp*), and how many times (*user_track_count*).\n",
    "\n",
    "Let's now run some preprocessing to get started."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [],
   "source": [
    "# convert the timestamp in a more convenient format\n",
    "train[\"timestamp\"] = pd.to_datetime(train[\"timestamp\"], unit=\"s\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [],
   "source": [
    "# add a column for Hour Of the Day\n",
    "train[\"hod\"] = train[\"timestamp\"].dt.hour"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "We are ready to do some EDA."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "yxlmpAMSQAYD"
   },
   "source": [
    "## EDA on Tracks\n",
    "\n",
    "Let's begin with a few statistics about tracks. We can inspect how many *unique* tracks are there! "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [],
   "source": [
    "percentiles = np.arange(1, step=0.1)\n",
    "gb_track = train.groupby(\"track_id\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**Q0: what's the Track distribution?**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "count    228997.000000\n",
       "mean         30.140421\n",
       "std          41.823156\n",
       "min           8.000000\n",
       "0%            8.000000\n",
       "10%          10.000000\n",
       "20%          12.000000\n",
       "30%          13.000000\n",
       "40%          15.000000\n",
       "50%          17.000000\n",
       "60%          21.000000\n",
       "70%          26.000000\n",
       "80%          36.000000\n",
       "90%          59.000000\n",
       "max        1585.000000\n",
       "dtype: float64"
      ]
     },
     "execution_count": 9,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "track_records = gb_track.size()\n",
    "track_records.describe(percentiles)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "There are ~230K distinct tracks. But wait. We know there is a `timestamp` feature, does that mean that a user can play the same track multiple times? Programmatically, we should check if the pair <`user`,`track`> does appear multiple times."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "True"
      ]
     },
     "execution_count": 10,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "len(train.groupby([\"user_id\", \"track_id\"]).size()) == train.shape[0]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "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>user_id</th>\n",
       "      <th>artist_id</th>\n",
       "      <th>album_id</th>\n",
       "      <th>track_id</th>\n",
       "      <th>timestamp</th>\n",
       "      <th>user_track_count</th>\n",
       "      <th>hod</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>751</th>\n",
       "      <td>50931921</td>\n",
       "      <td>3887</td>\n",
       "      <td>1168550</td>\n",
       "      <td>21882383</td>\n",
       "      <td>2014-01-08 17:07:16</td>\n",
       "      <td>1</td>\n",
       "      <td>17</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>752</th>\n",
       "      <td>50931921</td>\n",
       "      <td>3887</td>\n",
       "      <td>1168550</td>\n",
       "      <td>123627</td>\n",
       "      <td>2014-01-08 17:07:13</td>\n",
       "      <td>3</td>\n",
       "      <td>17</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>753</th>\n",
       "      <td>50931921</td>\n",
       "      <td>3887</td>\n",
       "      <td>630816</td>\n",
       "      <td>5065782</td>\n",
       "      <td>2013-12-06 18:17:02</td>\n",
       "      <td>1</td>\n",
       "      <td>18</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>754</th>\n",
       "      <td>50931921</td>\n",
       "      <td>279</td>\n",
       "      <td>4451359</td>\n",
       "      <td>37612</td>\n",
       "      <td>2013-12-01 11:33:44</td>\n",
       "      <td>1</td>\n",
       "      <td>11</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>755</th>\n",
       "      <td>50931921</td>\n",
       "      <td>4166</td>\n",
       "      <td>29770</td>\n",
       "      <td>85498</td>\n",
       "      <td>2013-11-17 11:53:59</td>\n",
       "      <td>1</td>\n",
       "      <td>11</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "      user_id  artist_id  album_id  track_id           timestamp  \\\n",
       "751  50931921       3887   1168550  21882383 2014-01-08 17:07:16   \n",
       "752  50931921       3887   1168550    123627 2014-01-08 17:07:13   \n",
       "753  50931921       3887    630816   5065782 2013-12-06 18:17:02   \n",
       "754  50931921        279   4451359     37612 2013-12-01 11:33:44   \n",
       "755  50931921       4166     29770     85498 2013-11-17 11:53:59   \n",
       "\n",
       "     user_track_count  hod  \n",
       "751                 1   17  \n",
       "752                 3   17  \n",
       "753                 1   18  \n",
       "754                 1   11  \n",
       "755                 1   11  "
      ]
     },
     "execution_count": 11,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "train.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "The two sizes correspond: we can assume there are no <`user`,`track`> duplicates. Indeed, the `user_track_count` value lets us know the play count!\n",
    "\n",
    "We also obtained a new information: `track_records` contains the number of distinct users that played a given track. Let's now observe the distribution graphically."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Text(0, 0.5, 'Track Count (log)')"
      ]
     },
     "execution_count": 12,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1200x800 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "fig, ax = plt.subplots(figsize=(12,8))\n",
    "ax.hist(track_records)\n",
    "ax.set_yscale(\"log\")\n",
    "ax.set_xlabel(\"Distinct Users\")\n",
    "ax.set_ylabel(\"Track Count (log)\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Let's look at the distribution in percentiles again. "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {
    "id": "nQk3NbVYpsE2",
    "outputId": "1bd40fd4-a366-43da-e316-6449648beca0"
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "count    228997.000000\n",
       "mean         30.140421\n",
       "std          41.823156\n",
       "min           8.000000\n",
       "0%            8.000000\n",
       "10%          10.000000\n",
       "20%          12.000000\n",
       "30%          13.000000\n",
       "40%          15.000000\n",
       "50%          17.000000\n",
       "60%          21.000000\n",
       "70%          26.000000\n",
       "80%          36.000000\n",
       "90%          59.000000\n",
       "max        1585.000000\n",
       "dtype: float64"
      ]
     },
     "execution_count": 13,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "track_records.describe(percentiles)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Tracks are listened by 30 distinct users on the average, and 90% of them has less than 59 distinct users that played it. Further, there are some outliers: the maximum number of users is ~1600. Is this a very popular track or a dataset artifact? Let's find out.   "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "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>track</th>\n",
       "      <th>artist_id</th>\n",
       "      <th>artist</th>\n",
       "      <th>albums_id</th>\n",
       "      <th>albums</th>\n",
       "      <th>urlSong</th>\n",
       "      <th>arousal_predicted</th>\n",
       "      <th>valence_predicted</th>\n",
       "      <th>lastfm_id</th>\n",
       "      <th>has_emotion_tags</th>\n",
       "      <th>has_social_tags</th>\n",
       "      <th>isClassic</th>\n",
       "      <th>distinct_user_count</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>track_id</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>4868</th>\n",
       "      <td>Rolling in the Deep</td>\n",
       "      <td>504</td>\n",
       "      <td>Adele</td>\n",
       "      <td>[1518, 18942, 64876, 66600, 80986, 93573, 9761...</td>\n",
       "      <td>['21', '19', '21 (Deluxe Edition)', 'monday_pl...</td>\n",
       "      <td>http://lyrics.wikia.com/Adele:Rolling_In_The_Deep</td>\n",
       "      <td>0.390623</td>\n",
       "      <td>-0.263898</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "      <td>True</td>\n",
       "      <td>1585</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>26142</th>\n",
       "      <td>We Are Young (feat. Janelle Monáe)</td>\n",
       "      <td>3151</td>\n",
       "      <td>fun.</td>\n",
       "      <td>[6907, 9619, 53139, 162664, 238945, 246098, 27...</td>\n",
       "      <td>['Some Nights', 'We Are Young', nan, 'The Best...</td>\n",
       "      <td>None</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "      <td>1500</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>11393952</th>\n",
       "      <td>Somebody That I Used to Know</td>\n",
       "      <td>3085</td>\n",
       "      <td>Gotye</td>\n",
       "      <td>[20107, 33321, 90034, 103479, 107493, 158828, ...</td>\n",
       "      <td>['Making Mirrors (Deluxe Edition)', 'Somebody ...</td>\n",
       "      <td>http://lyrics.wikia.com/Gotye:Somebody_That_I_...</td>\n",
       "      <td>-0.383686</td>\n",
       "      <td>-0.720571</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "      <td>True</td>\n",
       "      <td>1402</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>16110</th>\n",
       "      <td>Get Lucky (feat. Pharrell Williams)</td>\n",
       "      <td>320</td>\n",
       "      <td>Daft Punk</td>\n",
       "      <td>[5331, 10200, 11728, 40809, 84112, 95892, 1237...</td>\n",
       "      <td>['Random Access Memories', 'Get Lucky', 'Homew...</td>\n",
       "      <td>None</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "      <td>1349</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>14034</th>\n",
       "      <td>We Found Love (feat. Calvin Harris)</td>\n",
       "      <td>355</td>\n",
       "      <td>Rihanna</td>\n",
       "      <td>[4961, 6808, 14253, 18206, 18254, 19802, 40014...</td>\n",
       "      <td>['We Found Love', 'The Official Uk Top 40 Sing...</td>\n",
       "      <td>None</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "      <td>1248</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>51176</th>\n",
       "      <td>Titanium (feat. Sia)</td>\n",
       "      <td>1935</td>\n",
       "      <td>David Guetta</td>\n",
       "      <td>[16065, 18345, 18845, 32658, 40236, 80698, 875...</td>\n",
       "      <td>['Best of...', 'Nothing but the Beat 2.0', 'No...</td>\n",
       "      <td>None</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "      <td>1134</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>28847</th>\n",
       "      <td>Get Lucky (Radio Edit) [feat. Pharrell Williams]</td>\n",
       "      <td>320</td>\n",
       "      <td>Daft Punk</td>\n",
       "      <td>[10200, 19277, 136209, 188202, 269754, 361728,...</td>\n",
       "      <td>['Get Lucky', nan, 'Random Access Memories', '...</td>\n",
       "      <td>None</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "      <td>1110</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>572665</th>\n",
       "      <td>Set Fire to the Rain</td>\n",
       "      <td>504</td>\n",
       "      <td>Adele</td>\n",
       "      <td>[1518, 64876, 93573, 98525, 116649, 143113, 14...</td>\n",
       "      <td>['21', '21 (Deluxe Edition)', 'Rolling in the ...</td>\n",
       "      <td>http://lyrics.wikia.com/Adele:Set_Fire_To_The_...</td>\n",
       "      <td>-0.428821</td>\n",
       "      <td>-0.415065</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "      <td>False</td>\n",
       "      <td>1081</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>29375</th>\n",
       "      <td>Lights</td>\n",
       "      <td>4091</td>\n",
       "      <td>Ellie Goulding</td>\n",
       "      <td>[9086, 9174, 10388, 10389, 12804, 19124, 35773...</td>\n",
       "      <td>['Halcyon (Deluxe Edition)', 'Halcyon Days', '...</td>\n",
       "      <td>http://lyrics.wikia.com/Ellie_Goulding:Lights</td>\n",
       "      <td>-0.272432</td>\n",
       "      <td>-0.724689</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "      <td>False</td>\n",
       "      <td>996</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2498</th>\n",
       "      <td>Blue Jeans</td>\n",
       "      <td>402</td>\n",
       "      <td>Lana Del Rey</td>\n",
       "      <td>[1066, 4509, 12831, 20462, 20475, 20735, 79820...</td>\n",
       "      <td>['Born to Die', 'Born to Die - The Paradise Ed...</td>\n",
       "      <td>http://lyrics.wikia.com/Lana_Del_Rey:Blue_Jeans</td>\n",
       "      <td>0.278431</td>\n",
       "      <td>-0.109800</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "      <td>False</td>\n",
       "      <td>990</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                                                     track  artist_id  \\\n",
       "track_id                                                                \n",
       "4868                                   Rolling in the Deep        504   \n",
       "26142                   We Are Young (feat. Janelle Monáe)       3151   \n",
       "11393952                      Somebody That I Used to Know       3085   \n",
       "16110                  Get Lucky (feat. Pharrell Williams)        320   \n",
       "14034                  We Found Love (feat. Calvin Harris)        355   \n",
       "51176                                 Titanium (feat. Sia)       1935   \n",
       "28847     Get Lucky (Radio Edit) [feat. Pharrell Williams]        320   \n",
       "572665                                Set Fire to the Rain        504   \n",
       "29375                                               Lights       4091   \n",
       "2498                                            Blue Jeans        402   \n",
       "\n",
       "                  artist                                          albums_id  \\\n",
       "track_id                                                                      \n",
       "4868               Adele  [1518, 18942, 64876, 66600, 80986, 93573, 9761...   \n",
       "26142               fun.  [6907, 9619, 53139, 162664, 238945, 246098, 27...   \n",
       "11393952           Gotye  [20107, 33321, 90034, 103479, 107493, 158828, ...   \n",
       "16110          Daft Punk  [5331, 10200, 11728, 40809, 84112, 95892, 1237...   \n",
       "14034            Rihanna  [4961, 6808, 14253, 18206, 18254, 19802, 40014...   \n",
       "51176       David Guetta  [16065, 18345, 18845, 32658, 40236, 80698, 875...   \n",
       "28847          Daft Punk  [10200, 19277, 136209, 188202, 269754, 361728,...   \n",
       "572665             Adele  [1518, 64876, 93573, 98525, 116649, 143113, 14...   \n",
       "29375     Ellie Goulding  [9086, 9174, 10388, 10389, 12804, 19124, 35773...   \n",
       "2498        Lana Del Rey  [1066, 4509, 12831, 20462, 20475, 20735, 79820...   \n",
       "\n",
       "                                                     albums  \\\n",
       "track_id                                                      \n",
       "4868      ['21', '19', '21 (Deluxe Edition)', 'monday_pl...   \n",
       "26142     ['Some Nights', 'We Are Young', nan, 'The Best...   \n",
       "11393952  ['Making Mirrors (Deluxe Edition)', 'Somebody ...   \n",
       "16110     ['Random Access Memories', 'Get Lucky', 'Homew...   \n",
       "14034     ['We Found Love', 'The Official Uk Top 40 Sing...   \n",
       "51176     ['Best of...', 'Nothing but the Beat 2.0', 'No...   \n",
       "28847     ['Get Lucky', nan, 'Random Access Memories', '...   \n",
       "572665    ['21', '21 (Deluxe Edition)', 'Rolling in the ...   \n",
       "29375     ['Halcyon (Deluxe Edition)', 'Halcyon Days', '...   \n",
       "2498      ['Born to Die', 'Born to Die - The Paradise Ed...   \n",
       "\n",
       "                                                    urlSong  \\\n",
       "track_id                                                      \n",
       "4868      http://lyrics.wikia.com/Adele:Rolling_In_The_Deep   \n",
       "26142                                                  None   \n",
       "11393952  http://lyrics.wikia.com/Gotye:Somebody_That_I_...   \n",
       "16110                                                  None   \n",
       "14034                                                  None   \n",
       "51176                                                  None   \n",
       "28847                                                  None   \n",
       "572665    http://lyrics.wikia.com/Adele:Set_Fire_To_The_...   \n",
       "29375         http://lyrics.wikia.com/Ellie_Goulding:Lights   \n",
       "2498        http://lyrics.wikia.com/Lana_Del_Rey:Blue_Jeans   \n",
       "\n",
       "          arousal_predicted  valence_predicted lastfm_id has_emotion_tags  \\\n",
       "track_id                                                                    \n",
       "4868               0.390623          -0.263898      None             None   \n",
       "26142                   NaN                NaN      None             None   \n",
       "11393952          -0.383686          -0.720571      None             None   \n",
       "16110                   NaN                NaN      None             None   \n",
       "14034                   NaN                NaN      None             None   \n",
       "51176                   NaN                NaN      None             None   \n",
       "28847                   NaN                NaN      None             None   \n",
       "572665            -0.428821          -0.415065      None             None   \n",
       "29375             -0.272432          -0.724689      None             None   \n",
       "2498               0.278431          -0.109800      None             None   \n",
       "\n",
       "         has_social_tags isClassic  distinct_user_count  \n",
       "track_id                                                 \n",
       "4868                None      True                 1585  \n",
       "26142               None      None                 1500  \n",
       "11393952            None      True                 1402  \n",
       "16110               None      None                 1349  \n",
       "14034               None      None                 1248  \n",
       "51176               None      None                 1134  \n",
       "28847               None      None                 1110  \n",
       "572665              None     False                 1081  \n",
       "29375               None     False                  996  \n",
       "2498                None     False                  990  "
      ]
     },
     "execution_count": 15,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "K = 10\n",
    "top_by_user = track_records.iloc[track_records.argsort().values[:-(K+1):-1]]\n",
    "top_by_user.name = \"distinct_user_count\"\n",
    "\n",
    "# Retrieve the task information from the `dataset` handler variable\n",
    "dataset.df_tracks.loc[top_by_user.index].join(top_by_user)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Well, these song seem pretty popular, right? But something more came up... look at position **3** and **6**: there are two different versions of *Get Lucky*. Is this something you want your model to factor in at training time? "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**Q1: How many times does a user play the same song?**\n",
    "\n",
    "We can use the handy `user_track_count` to measure how frequently users get engaged by songs (remember to take a look at all the columns in `dataset.df_users`, there are very userful variable there)."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "count    228997.000000\n",
       "mean          2.577950\n",
       "std           2.499551\n",
       "min           1.000000\n",
       "0%            1.000000\n",
       "10%           1.250000\n",
       "20%           1.416667\n",
       "30%           1.590909\n",
       "40%           1.772727\n",
       "50%           2.000000\n",
       "60%           2.250000\n",
       "70%           2.615385\n",
       "80%           3.176471\n",
       "90%           4.286157\n",
       "max         231.333333\n",
       "Name: user_track_count, dtype: float64"
      ]
     },
     "execution_count": 16,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "avg_play_count = gb_track[\"user_track_count\"].mean()\n",
    "avg_play_count.describe(percentiles)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "On the average, a user repeats a given track ~2.6 times. Again, let's inspect our tail."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "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>track</th>\n",
       "      <th>artist_id</th>\n",
       "      <th>artist</th>\n",
       "      <th>albums_id</th>\n",
       "      <th>albums</th>\n",
       "      <th>urlSong</th>\n",
       "      <th>arousal_predicted</th>\n",
       "      <th>valence_predicted</th>\n",
       "      <th>lastfm_id</th>\n",
       "      <th>has_emotion_tags</th>\n",
       "      <th>has_social_tags</th>\n",
       "      <th>isClassic</th>\n",
       "      <th>avg_play_count</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>track_id</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>87233</th>\n",
       "      <td>Final Hour</td>\n",
       "      <td>14527</td>\n",
       "      <td>X-Ray Dog</td>\n",
       "      <td>[37565, 129840, 159281]</td>\n",
       "      <td>['Night Hounds', nan, 'Harry Potter']</td>\n",
       "      <td>None</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "      <td>231.333333</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>190112</th>\n",
       "      <td>Asleep</td>\n",
       "      <td>15738</td>\n",
       "      <td>Agua de Annique</td>\n",
       "      <td>[44908]</td>\n",
       "      <td>['Air']</td>\n",
       "      <td>http://lyrics.wikia.com/Agua_De_Annique:Asleep</td>\n",
       "      <td>-0.270782</td>\n",
       "      <td>0.108086</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "      <td>False</td>\n",
       "      <td>146.454545</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>583120</th>\n",
       "      <td>Rubik's Cube</td>\n",
       "      <td>10017</td>\n",
       "      <td>Athlete</td>\n",
       "      <td>[194694, 206508]</td>\n",
       "      <td>['Black Swan', 'Singles 01-10']</td>\n",
       "      <td>http://lyrics.wikia.com/Athlete:Rubik%27s_Cube</td>\n",
       "      <td>-0.214444</td>\n",
       "      <td>-0.143472</td>\n",
       "      <td>TRGPGZL128F9330871</td>\n",
       "      <td>False</td>\n",
       "      <td>True</td>\n",
       "      <td>False</td>\n",
       "      <td>122.928571</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>11741477</th>\n",
       "      <td>Dear Darlin'</td>\n",
       "      <td>9699</td>\n",
       "      <td>Olly Murs</td>\n",
       "      <td>[152172, 680866, 699780, 1075749, 2277719, 448...</td>\n",
       "      <td>['Right Place Right Time (Deluxe Edition)', 'R...</td>\n",
       "      <td>http://lyrics.wikia.com/Olly_Murs:Dear_Darlin%27</td>\n",
       "      <td>-0.226224</td>\n",
       "      <td>-0.783971</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "      <td>False</td>\n",
       "      <td>122.300000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>100848</th>\n",
       "      <td>Slow Me Down</td>\n",
       "      <td>15692</td>\n",
       "      <td>Anneke van Giersbergen</td>\n",
       "      <td>[33336, 7100104]</td>\n",
       "      <td>['Everything Is Changing', 'Everything Is Chan...</td>\n",
       "      <td>http://lyrics.wikia.com/Anneke_Van_Giersbergen...</td>\n",
       "      <td>-0.119751</td>\n",
       "      <td>-0.049463</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "      <td>False</td>\n",
       "      <td>122.062500</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>102757</th>\n",
       "      <td>Entombed in Butchered Bodies</td>\n",
       "      <td>15795</td>\n",
       "      <td>Cerebral Bore</td>\n",
       "      <td>[33636]</td>\n",
       "      <td>['Maniacal Miscreation']</td>\n",
       "      <td>http://lyrics.wikia.com/Cerebral_Bore:Entombed...</td>\n",
       "      <td>0.519996</td>\n",
       "      <td>-1.059996</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "      <td>False</td>\n",
       "      <td>121.181818</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>764985</th>\n",
       "      <td>Flesh Reflects the Madness</td>\n",
       "      <td>15795</td>\n",
       "      <td>Cerebral Bore</td>\n",
       "      <td>[33636, 2707275]</td>\n",
       "      <td>['Maniacal Miscreation', 'The Dead Flesh Archi...</td>\n",
       "      <td>http://lyrics.wikia.com/Cerebral_Bore:Flesh_Re...</td>\n",
       "      <td>0.866363</td>\n",
       "      <td>-0.560911</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "      <td>False</td>\n",
       "      <td>119.818182</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>764984</th>\n",
       "      <td>Maniacal Miscreation</td>\n",
       "      <td>15795</td>\n",
       "      <td>Cerebral Bore</td>\n",
       "      <td>[33636]</td>\n",
       "      <td>['Maniacal Miscreation']</td>\n",
       "      <td>http://lyrics.wikia.com/Cerebral_Bore:Maniacal...</td>\n",
       "      <td>0.742891</td>\n",
       "      <td>-0.777889</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "      <td>False</td>\n",
       "      <td>119.545455</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3130123</th>\n",
       "      <td>Everything We Touch</td>\n",
       "      <td>2001</td>\n",
       "      <td>Say Lou Lou</td>\n",
       "      <td>[174522, 5112993, 8791457, 9090677, 13248338]</td>\n",
       "      <td>[nan, 'Everything We Touch', 'Everything We To...</td>\n",
       "      <td>http://lyrics.wikia.com/Say_Lou_Lou:Everything...</td>\n",
       "      <td>-0.228184</td>\n",
       "      <td>-0.180018</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "      <td>False</td>\n",
       "      <td>116.421053</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2645787</th>\n",
       "      <td>Main Menu</td>\n",
       "      <td>2566</td>\n",
       "      <td>Jesper Kyd</td>\n",
       "      <td>[82574, 83385, 121124, 127995, 2035355, 242793...</td>\n",
       "      <td>['Borderlands 2 Original Soundtrack', 'Borderl...</td>\n",
       "      <td>None</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "      <td>114.181818</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                                 track  artist_id                  artist  \\\n",
       "track_id                                                                    \n",
       "87233                       Final Hour      14527               X-Ray Dog   \n",
       "190112                          Asleep      15738         Agua de Annique   \n",
       "583120                    Rubik's Cube      10017                 Athlete   \n",
       "11741477                  Dear Darlin'       9699               Olly Murs   \n",
       "100848                    Slow Me Down      15692  Anneke van Giersbergen   \n",
       "102757    Entombed in Butchered Bodies      15795           Cerebral Bore   \n",
       "764985      Flesh Reflects the Madness      15795           Cerebral Bore   \n",
       "764984            Maniacal Miscreation      15795           Cerebral Bore   \n",
       "3130123            Everything We Touch       2001             Say Lou Lou   \n",
       "2645787                      Main Menu       2566              Jesper Kyd   \n",
       "\n",
       "                                                  albums_id  \\\n",
       "track_id                                                      \n",
       "87233                               [37565, 129840, 159281]   \n",
       "190112                                              [44908]   \n",
       "583120                                     [194694, 206508]   \n",
       "11741477  [152172, 680866, 699780, 1075749, 2277719, 448...   \n",
       "100848                                     [33336, 7100104]   \n",
       "102757                                              [33636]   \n",
       "764985                                     [33636, 2707275]   \n",
       "764984                                              [33636]   \n",
       "3130123       [174522, 5112993, 8791457, 9090677, 13248338]   \n",
       "2645787   [82574, 83385, 121124, 127995, 2035355, 242793...   \n",
       "\n",
       "                                                     albums  \\\n",
       "track_id                                                      \n",
       "87233                 ['Night Hounds', nan, 'Harry Potter']   \n",
       "190112                                              ['Air']   \n",
       "583120                      ['Black Swan', 'Singles 01-10']   \n",
       "11741477  ['Right Place Right Time (Deluxe Edition)', 'R...   \n",
       "100848    ['Everything Is Changing', 'Everything Is Chan...   \n",
       "102757                             ['Maniacal Miscreation']   \n",
       "764985    ['Maniacal Miscreation', 'The Dead Flesh Archi...   \n",
       "764984                             ['Maniacal Miscreation']   \n",
       "3130123   [nan, 'Everything We Touch', 'Everything We To...   \n",
       "2645787   ['Borderlands 2 Original Soundtrack', 'Borderl...   \n",
       "\n",
       "                                                    urlSong  \\\n",
       "track_id                                                      \n",
       "87233                                                  None   \n",
       "190112       http://lyrics.wikia.com/Agua_De_Annique:Asleep   \n",
       "583120       http://lyrics.wikia.com/Athlete:Rubik%27s_Cube   \n",
       "11741477   http://lyrics.wikia.com/Olly_Murs:Dear_Darlin%27   \n",
       "100848    http://lyrics.wikia.com/Anneke_Van_Giersbergen...   \n",
       "102757    http://lyrics.wikia.com/Cerebral_Bore:Entombed...   \n",
       "764985    http://lyrics.wikia.com/Cerebral_Bore:Flesh_Re...   \n",
       "764984    http://lyrics.wikia.com/Cerebral_Bore:Maniacal...   \n",
       "3130123   http://lyrics.wikia.com/Say_Lou_Lou:Everything...   \n",
       "2645787                                                None   \n",
       "\n",
       "          arousal_predicted  valence_predicted           lastfm_id  \\\n",
       "track_id                                                             \n",
       "87233                   NaN                NaN                None   \n",
       "190112            -0.270782           0.108086                None   \n",
       "583120            -0.214444          -0.143472  TRGPGZL128F9330871   \n",
       "11741477          -0.226224          -0.783971                None   \n",
       "100848            -0.119751          -0.049463                None   \n",
       "102757             0.519996          -1.059996                None   \n",
       "764985             0.866363          -0.560911                None   \n",
       "764984             0.742891          -0.777889                None   \n",
       "3130123           -0.228184          -0.180018                None   \n",
       "2645787                 NaN                NaN                None   \n",
       "\n",
       "         has_emotion_tags has_social_tags isClassic  avg_play_count  \n",
       "track_id                                                             \n",
       "87233                None            None      None      231.333333  \n",
       "190112               None            None     False      146.454545  \n",
       "583120              False            True     False      122.928571  \n",
       "11741477             None            None     False      122.300000  \n",
       "100848               None            None     False      122.062500  \n",
       "102757               None            None     False      121.181818  \n",
       "764985               None            None     False      119.818182  \n",
       "764984               None            None     False      119.545455  \n",
       "3130123              None            None     False      116.421053  \n",
       "2645787              None            None      None      114.181818  "
      ]
     },
     "execution_count": 17,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "top_by_user = avg_play_count.iloc[avg_play_count.argsort().values[:-(K+1):-1]]\n",
    "top_by_user.name = \"avg_play_count\"\n",
    "dataset.df_tracks.loc[top_by_user.index].join(top_by_user)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Here it is a new piece of information: the top 10 tracks by number of play counts is completely different than the one by number of distinct users."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Cleanup..\n",
    "del gb_track\n",
    "del avg_play_count\n",
    "del track_records\n",
    "del top_by_user\n",
    "_ = gc.collect()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Statistics of WASABI features\n",
    "\n",
    "The new features coming from the WASABI dataset are not available for all of our tracks, but they still cover a good part of them! Let us see what the coverage is for each of these features:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {},
   "outputs": [],
   "source": [
    "# NOTE that we only show statistics on train, but as train/test were uniformly\n",
    "#      sampled statistics for the test set are very similar (you can check by\n",
    "#      using \"test.join(...)\" below here\n",
    "train_tracks = train.join(dataset.df_tracks, on=\"track_id\", rsuffix=\"_data\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "'arousal_predicted': 0.51 tracks, 0.79 events\n",
      "'valence_predicted': 0.51 tracks, 0.79 events\n",
      "'isClassic': 0.51 tracks, 0.79 events\n",
      "'has_social_tags': 0.17 tracks, 0.31 events\n",
      "'has_emotion_tags': 0.17 tracks, 0.31 events\n"
     ]
    }
   ],
   "source": [
    "for feat in ['arousal_predicted', 'valence_predicted',\n",
    "             'isClassic', 'has_social_tags', 'has_emotion_tags']:\n",
    "\n",
    "    perc_tracks = np.sum(dataset.df_tracks[feat].notna())/len(dataset.df_tracks)\n",
    "    perc_events = np.sum(train_tracks[feat].notna())/len(train_tracks)\n",
    "    print(f\"'{feat}': {round(perc_tracks,2)} tracks, {round(perc_events,2)} events\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Similarly, topics features are available for ~69% of samples in train_tracks and song/verse embeddings cover ~78% of the samples:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.6915584406176354"
      ]
     },
     "execution_count": 21,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "train_topics = train_tracks.join(dataset.df_topics, on=\"urlSong\", rsuffix=\"_data\")\n",
    "np.sum(train_topics.topics.notna())/len(train_topics)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.7794887791568496"
      ]
     },
     "execution_count": 22,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "train_embs = train_tracks.join(dataset.df_song_embeddings, on=\"urlSong\", rsuffix=\"_data\")\n",
    "np.sum(train_embs.song_embedding.notna())/len(train_embs)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Using topics\n",
    "\n",
    "Topics are 60-dimensional descriptors that represent a song wrt its level of membership wrt 60 different topics. We expect songs with similar contents to have similar descriptors. Here is a simple example that shows the quasi-perfect matching of different covers of the same song:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 53,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "http://lyrics.wikia.com/Ayreon:Space_Oddity\n",
      "http://lyrics.wikia.com/Chris_Hadfield:Space_Oddity\n",
      "http://lyrics.wikia.com/David_Bowie:Space_Oddity\n",
      "http://lyrics.wikia.com/Helloween:Space_Oddity\n",
      "http://lyrics.wikia.com/Powerman_5000:Space_Oddity\n",
      "http://lyrics.wikia.com/Tangerine_Dream:Space_Oddity\n"
     ]
    }
   ],
   "source": [
    "descr = []\n",
    "for urlSong in list(dataset.df_topics[dataset.df_topics.index.str.contains(\"Space_Oddity\")].index):\n",
    "    print(urlSong)\n",
    "    descr.append(dataset.df_topics[dataset.df_topics.index==urlSong].topics.iloc[0])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 54,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "x = range(len(descr[0]))\n",
    "for d in descr:\n",
    "    plt.plot(x,d)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Here is, instead, the descriptor for a random song in the dataset plotted in black together with the above:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 57,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[<matplotlib.lines.Line2D at 0x7fef3c9c4dc0>]"
      ]
     },
     "execution_count": 57,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAi8AAAGgCAYAAABi2ofUAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjYuMywgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy/P9b71AAAACXBIWXMAAA9hAAAPYQGoP6dpAAD3zElEQVR4nOz9eZhkV33fj7/P3Wqv3maVRgsIoUFikQUjLIyw7DjgYEww0fMDDIEQAz87TgSEKPHClwj42TGGhCAwNnaEiZ0I+WdjFoNsZBbL8iKNNkuAGO2aRTPdM9NbdVXd/Z7vH+ece29V13arq7uraz6v59Ez6upb1VW37vI+78/GOOccBEEQBEEQOwRtu98AQRAEQRBEFki8EARBEASxoyDxQhAEQRDEjoLEC0EQBEEQOwoSLwRBEARB7ChIvBAEQRAEsaMg8UIQBEEQxI6CxAtBEARBEDsKY7vfwGbAOUcUbU7vPU1jm/bakwjtr2zQ/soO7bNs0P7KDu2zbAy7vzSNgTE20LYTKV6iiGNpqTHy1zUMDTMzJdRqTQRBNPLXnzRof2WD9ld2aJ9lg/ZXdmifZWMj+2t2tgRdH0y8UNiIIAiCIIgdBYkXgiAIgiB2FCReCIIgCILYUZB4IQiCIAhiR0HihSAIgiCIHQWJF4IgCIIgdhQkXgiCIAiC2FGQeCEIgiAIYkdB4oUgCIIgiB0FiReCIAiCIHYUJF4IgiAIgthRkHghCIIgCGJHQeKFIAiCIIgdBYkXYkNwTmPiCYIgiK2FxAsxNH/+53+KH//xH8X3vvfQdr8VgiAI4hyCxAsxNHff/Q+o1Vbx4IP3b/dbIQiCIM4hSLwQQ+N5HgAgCIJtficEQRDEuQSJF2JofN+T//rb/E4IgiCIc4nM4iWKItx888249tprceWVV+Ld7343jh8/PtDz3vWud+HTn/70ut/95V/+JV772tfixS9+Md7whjfgH//xH7O+LWIbUM4LiReCIAhiK8ksXj772c/i1ltvxUc/+lHcdtttsShRN7JOeJ6HX/u1X8Ndd9217nd33303brzxRrz5zW/Gl7/8ZVxzzTV4z3vegyeffDLrWyO2GBIvBEEQxHaQSbx4nofPf/7zuOGGG3Ddddfh4MGD+OQnP4n5+XnccccdHZ/zwAMP4I1vfCPuu+8+VKvVdb//gz/4A/zUT/0U3v72t+OSSy7Bf/kv/wVXXHEF/vf//t/DfSJiy6CwEUEQBLEdGFk2PnLkCBqNBq655pr4sWq1issvvxz33nsvXve61617zp133olrr70Wv/zLv4zXv/71Lb+LoggPPPAAfuVXfqXl8Ze//OVdxdCgGMbo03l0XWv591zH90WibhgGHfc37a9s0P7KDu2zbND+yg7ts2xs1f7KJF7m5+cBAPv37295fM+ePfHv2nn/+9/f9fVqtRqazSb27ds38OsNgqYxzMyUhn5+P6rVwqa99k4iDIXjouvoub9pf2WD9ld2aJ9lg/ZXdmifZWOz91cm8WLbNgDAsqyWx3O5HFZXVzP/ccdxur6e67qZX08RRRy1WnPo53dD1zVUqwXUajbCMBr56+80HEd8R/V6E8vLjXW/p/2VDdpf2aF9lg3aX9mhfZaNjeyvarUwsGOTSbzk83kAIvdF/T8AuK6LQiG7ysrlcvHrpRn29dIEweYdZGEYberr7xTU9+Z5Xs/9QfsrG7S/skP7LBu0v7JD+ywbm72/MgWlVLjo9OnTLY+fPn0ae/fuzfzHp6enUSwWR/Z6xNZC1UYEQRDEdpBJvBw8eBDlchn33HNP/FitVsMjjzyCQ4cOZf7jjDFcddVVOHz4cMvj99xzD172spdlfj1iayHxQhAEQWwHmcJGlmXhbW97Gz7xiU9gdnYW559/Pj7+8Y9j3759ePWrX40wDLG0tIRKpdISVurFO9/5TrznPe/B5Zdfjle96lX40pe+hB/+8If4jd/4jaE+ELF1BIEQLSReCIIgiK0kcy3TDTfcgOuvvx4f/OAH8Za3vAW6ruOWW26BaZo4deoUXvnKV+L2228f+PVe+cpX4jd/8zfxxS9+ET/3cz+Hu+++G7/3e7+HSy65JOtbI7aQKIrimUY024ggCILYShjnnG/3mxg1YRhhaWl99ctGMQwNMzMlLC83zvnELcdx8KM/eiUA4Morr8IXvnDrum1of2WD9ld2aJ9lg/ZXdmifZWMj+2t2tjRwtRF13SGGIl0hRmEjgiAIYish8UIMhRoNIP6fxAtBEASxdZB4IYYiLVhU4i5BEARBbAUkXoihaA0bUcIuQRAEsXWQeCGGIi1eyHkhCIIgthISL8RQUM4LQRAEsV2QeCGGojVs5PXYkiAIgiBGC4kXYiioVJogCILYLki8EEORFiwkXgiCIIithMQLMRTpUFEQBJjARs0EQRDEmELihRiKdNiIc44wDLfx3RAEQRDnEiReiKHwvNZQEYWOCIIgiK2CxAsxFO0VRiReCIIgiK2CxAsxFOmwEUCN6giCIIitg8QLMRTtTgs5LwRBEMRWQeKFGIr1zgvNNyIIgiC2BhIvxFCsz3mhLrsEQRDE1kDihRiKdueFwkYEQRDEVkHihRgKEi8EQRDEdkHihRgKKpUmCIIgtgsSL8RQtIsVStglCIIgtgoSL8RQUNiIIAiC2C5IvBBDQeKFIAiC2C5IvBBDQTkvBEEQxHZB4oUYCnJeCIIgiO2CxAsxFOsTdkm8EARBEFsDiRdiKChsRBAEQWwXJF6IofA8IVZ0XQdAzgtBEASxdZB4IYZC5bwUCkUA5LwQBEEQWweJF2IoVNioVCrJn0m8EARBEFsDiRdiKJTzQuKFIAiC2GpIvBBDocRKsUjihSAIgthaSLwQQ6Gcl2JR5LzQbCOCIAhiqyDxQgyFynlR4oWcF4IgCGKrIPFCDEXivFDYiCAIgthaSLwQmeGcU8IuQRAEsW2QeCEyk25Ip5wXalJHEARBbBUkXojMpF0WynkhCIIgthoSL0Rm0hOlk2ojEi8EQRDE1kDihciMEi+apiGfLwAg54UgCILYOki8EJlRQsWyLJim2fIYQRAEQWw2JF6IzCjnxTRJvBAEQRBbD4kXIjNp58UwSLwQBEEQWwuJFyIzifNiwjQNADQegCAIgtg6SLwQmVGjAUyTnBeCIAhi6yHxQmRGOS+UsEsQBEFsByReiMwk4sWMxQv1eSEIgiC2ChIvRGbSYSNyXgiCIIithsQLkRnq80IQBEFsJyReiMykq41Uwi5VGxEEQRBbBYkXIjOdE3a9Xk8hCIIgiJFB4oXIDOW8EARBENsJiRciM1QqTRAEQWwnmcVLFEW4+eabce211+LKK6/Eu9/9bhw/frzr9svLy/jABz6AQ4cO4eqrr8aHP/xh2Lbdss03vvENvO51r8NLXvISvPa1r8VXvvKVzB+E2Dq6iRfO+Xa+LYIgCOIcIbN4+exnP4tbb70VH/3oR3HbbbchiiK8613vim9o7dxwww04evQovvCFL+BTn/oU7rzzTtx0003x7++++2785//8n/G2t70NX//61/HWt74Vv/qrv4o777xz6A9FbC7KZREddo348TAMt+stEQRBEOcQmcSL53n4/Oc/jxtuuAHXXXcdDh48iE9+8pOYn5/HHXfcsW77Bx98EIcPH8bHPvYxXHHFFbjmmmvwkY98BF/96lexsLAAAPj2t7+Nyy67DG9+85txwQUX4K1vfSsOHjyIu+66azSfkBg5Sc5L0qROPE6hI4IgCGLzySRejhw5gkajgWuuuSZ+rFqt4vLLL8e99967bvv77rsPu3fvxiWXXBI/dvXVV4Mxhvvvvx8AMDc3h8cffxx33303OOe455578OSTT+LFL37xsJ+J2GTSYSNVKg2QeCEIgiC2BqP/Jgnz8/MAgP3797c8vmfPnvh3aRYWFtZta1kWpqencerUKQDAv/7X/xoPP/ww3vGOd0DXdYRhiF/8xV/E61//+kwfpB3DGH0usq5rLf+eq6hRAPl8Dvm8FT/Oediy32l/ZYP2V3Zon2WD9ld2aJ9lY6v2VybxohJtLctqeTyXy2F1dbXj9u3bqu1d1wUAnDp1CsvLy/jQhz6Eq666CnfffTc++clP4oILLsD111+f5e3FaBrDzExpqOcOQrVa2LTX3gkwJhJzq9USZmfLsCwLnuehWDQ67vdzfX9lhfZXdmifZYP2V3Zon2Vjs/dXJvGSz+cBiLCB+n8AcF0XhcL6N5rP5zsm8rqui2KxCAD4D//hP+B1r3sd3vrWtwIAXvCCF2B1dRUf//jH8cY3vhGall29RRFHrdbM/Lx+6LqGarWAWs1GGEYjf/2dQr0u9m0YMiwvN2AYJjzPw9mzqygUpuLtaH9lg/ZXdmifZYP2V3Zon2VjI/urWi0M7NhkEi8qBHT69GlceOGF8eOnT5/GZZddtm77ffv24Vvf+lbLY57nYWVlBXv27MHS0hKeeuopvOhFL2rZ5sorr8Tv/u7vYmVlBbOzs1neYkwQbN5BFobRpr7+uOO6QpDquo4giOKKI9f1Ou6Xc31/ZYX2V3Zon2WD9ld2aJ9lY7P3VyZb4+DBgyiXy7jnnnvix2q1Gh555BEcOnRo3faHDh3C/Pw8jh49Gj92+PBhAMBLX/pSTE1NoVAo4NFHH2153qOPPopqtTq0cCE2l3TCLgBqVEcQBEFsKZmcF8uy8La3vQ2f+MQnMDs7i/PPPx8f//jHsW/fPrz61a9GGIZYWlpCpVJBPp/HS17yElx11VV4//vfj5tuugnNZhMf+tCH8IY3vAF79+4FALz97W/H7/7u72L37t146Utfivvvvx+f+9zn8Mu//Mub8oGJjZMeDyD+JfFCEARBbB2ZxAsgms4FQYAPfvCDcBwHhw4dwi233ALTNHHixAn8s3/2z/Df/tt/wxvf+EYwxvCZz3wGH/7wh/GOd7wDuVwOP/3TP41f/dVfjV/vve99L2ZmZvC5z30Op06dwoEDB3DjjTfizW9+80g/KDE6yHkhCIIgtpPM4kXXddx444248cYb1/3uwIED60JAc3NzuPnmm3u+3jvf+U68853vzPpWiG2CxAtBEASxnVDhOpGZ9HgAAHGjOhIvBEEQxFZA4oXITHo8AIC42igIgm17TwRBEMS5A4kXIjMUNiIIgiC2ExIvRGZIvBAEQRDbCYkXIjPtOS9KvKiZRwRBEASxmZB4ITJDzgtBEASxnZB4ITKjRIoSL6raiJwXgiAIYisg8UJkpr3aiJwXgiAIYish8UJkIoqiuCSaxgMQBEEQ2wGJFyITKt8FoJwXgiAIYnsg8UJkopN4oQ67BEEQxFZC4oXIRDopl3JeCIIgiO2AxAuRCeW8GIYJxpj8fxoPQBAEQWwdJF6ITCQ9Xsz4MXJeCIIgiK2ExAuRifYGdQCJF4IgCGJrIfFCZCLp8bJevFCTOoIgCGIrIPFCZIKcF4IgCGK7IfFCZKJ9NACQJOySeCEIgiC2AhIvRCaU86LcFvH/QshQ2IggCILYCki8EJnolfNCzgtBEASxFZB4ITJBOS8EQRDEdkPihciE53XKeSHxQhAEQWwdJF6ITFDYiCAIgthuSLwQmaBqI4IgCGK7IfFCZKJztZFqUkezjQiCIIjNh8QLkYne4oWcF4IgiJ1OVF8C9+ztfhs9IfFCZELlvFC1EUEQxOTBnToaf/Jf0Pz6x7b7rfSExAuRCSqVJgiCmFyi2mkg9BGtzm/3W+kJiRciEyReCIIgJpc4XDTmaQAkXohMKIGSLpVWfV4o54UgCGJnw72m/J8QPAq39830gMQLkYmkz8v6hF1yXgiCIHY46UTdcHyv6SReiExQ2IggCGJySVcZcRIvxKSQlEp3Fi+c8215XwRBEMTGicNGABB42/dG+kDihchEp1JplfMCUKM6giCInQynsBExiXQaD2CaRvz/lLRLEASxg/FTYaMxvp6TeCEy0anDbtp58X1yXgiCIHYq5LwQE0mnqdJqMCNAzgtBEMROpiVhl3JeiEnB89aHjRhjVHFEEAQxAbQk7JLzQkwKnUqlASqXJgiCmARaS6XJeSEmhKTayGx5nMQLQRDEBNBSKj2+13MSL0QmOo0HAJKkXRIvBEEQOxfu2YjbdVHYiJgUOlUbpX+mhF2CIIidCQ8DhIGPO5tlPOgUxjph1+i/CUEkUM4LQRDEZMJ9G2uRhibXYIeMnBdicujUYRcg8UIQBDEuOM01PPDdL2HlzMlsT/RsLOo6ACAEG2vnhcQLkYlOs40AynkhCIIYF44deQCPP3gnHr3/O5mex70mGkzIAo1TnxdiQuCcU9iIIAhizPFdUTHkNNcyPY97NuyULPA9Z6Tva5SQeCEGJj10sV28qC67NJiRIAhiewll4YTnNPts2Qr3mvDA4p+d9KiAMYPECzEwKt8F6F5tRM4LQRDE9qKqPn03o/jwbDhRcm23yXkhtgunuYbHHvibzAq8EypkBKzPeaFSaYIgiPEgdl4yihfu2fCDpAjZ9t2Rvq9RQuJlwnn47/4aD/7Nn+MH9/zNhl9LiRdN01qGMQLkvBAEQYwLTlM4Jp7TBI87zvWHezZ4kMgC16OEXWKbmD82DwA4+fSJDb+WEibt+S4AiReCIIhxobZSBwDwKESYoWKIe02ktY47xk46iZcJp7m6CABYWzq94ddSOS/tISMgXSo9vkqdIAjiXMCur8T/nyllwLOBKPnR8cPRvakRQ+JlwomiQP67cVHheWqukbnud0nOC1UbEcS44p09i9Pf+S54OL43JWLjBG6SaOs5g+e9iInSifXikXghto94wtaGX6lbjxeAwkYEsRNYuO2LePxTn0H9oX/a7rdCbCKcJ/aJ5w7uvHCvCaSeG/iD58tsNZnFSxRFuPnmm3HttdfiyiuvxLvf/W4cP3686/bLy8v4wAc+gEOHDuHqq6/Ghz/8Ydh2qxJ8+OGH8da3vhUvfvGL8eM//uO4+eabEUVRl1cksiH2I8PGHZHBwkbbL16adQ/f+cYRLJysbfdbIYixYvFZET5eOn5qm98Jsbkk988s5dLcs8FSC10/HN/7cGbx8tnPfha33norPvrRj+K2225DFEV417ve1VJGm+aGG27A0aNH8YUvfAGf+tSncOedd+Kmm26Kf//000/j7W9/Oy655BJ87Wtfw6/92q/hC1/4Am655ZahPxSRwNRBzPxMWeed2CnOyxPfP4VHv7eAf/r7p7b7rRDEWNGsi1X4wrNnt/mdEJtLIjpce3DnJXSbYCy5T4yxdskmXjzPw+c//3nccMMNuO6663Dw4EF88pOfxPz8PO6444512z/44IM4fPgwPvaxj+GKK67ANddcg4985CP46le/ioWFBQDA5z73OTzvec/Dhz/8YVx88cV4zWteg3/zb/4NHnjggdF8wnMdeSBqLIBtb6xmPxEv3XNexkG8rD19FABQP0GrS4JIo8mctLBe3+Z3QmwmLCVe7Azftd3m0kTjm/KSTbwcOXIEjUYD11xzTfxYtVrF5ZdfjnvvvXfd9vfddx92796NSy65JH7s6quvBmMM999/PwDg7/7u7/C6170OjCUtiW+44Qb87u/+buYPQ3QiOfqWTm2s4kgJk85hI6Nlm+0ksEWymu+P8bJhgvje9x7Cf//vv4Vms7Hdb4Xogx6K85M3SLxMNsm1r15fHfhZdb+1o240vikvMPpvkjA/L3qG7N+/v+XxPXv2xL9Ls7CwsG5by7IwPT2NU6dOoV6v48yZM6hUKvi1X/s1/O3f/i2q1Sre8IY34Bd+4Regy9Hcw2AYo89F1nWt5d8dAUsO4pPHjuLiyy4e+qXCMOnz0r5/czkhaKIojH+3XfsrCMQZF0RsU46DzWJHHl8Afv/3P4u77roTL3rRi/Ha175uS//2Tt1n24Uuqw9hN3bUubGd7MxjLLnuN+q1gb5rHkVotFktPMp+L92q/ZVJvKhE2/ach1wuh9XV9erOtu2O+RG5XA6u66Iu7ayPfexjePvb344/+IM/wA9/+EP8xm/8BprNJt73vvdleXsxmsYwM1Ma6rmDUK0WNu21R03aPlxbXtzQfjFN4Y6VSoV1r1Otip81jXf43dbuL5XaE3B9U4+DzWInHV8AUK+LxOgwdLdtf++0fbZdKPGiO80deW5sJzvlGOOcg6UWrW6zPtB3Hdp1NFNDGcVrMUxP5cG07EbCZu+vTOIln88DELkP6v8BwHVdFArr32g+n++YyOu6LorFYhxqeMUrXoF//+//PQDgBS94AZaWlvA7v/M7eO9739sSThqUKOKo1TY+y6cdXddQrRZQq9kIxzmTqYVESa+cXcLy8vDW/sqKGK/OmL7udZTbUa83499t1/5yXXGB9rm2oc+71ezM4wtYWxOLkKWl1S3f3zt1n20HnHMYSry49o46N7aTnXaMhW1dcZv1+kDfdVg7C5u3SoIIwPLZFTAz3/lJHdjI/qpWCwM7NpnEiwoBnT59GhdeeGH8+OnTp3HZZZet237fvn341re+1fKY53lYWVnBnj17MDMzg1wuh+c///kt21x66aVoNptYWlrC3NxclrcYEwSbd5CFYbSprz8qwiBoyRz3HXtD79txRMKvYZjrXkfXxaHkef663231/gpCGTZiBnw/HEoAbyc75fhSNBriwthsbuz42gg7bZ9tB5Hvxetqy9u+72qnslOOMdduzVsJBvyuQ7sBm5tIN6kDZ/BdFxpbH0Hp+3qbvL8yBaUOHjyIcrmMe+65J36sVqvhkUcewaFDh9Ztf+jQIczPz+Po0aPxY4cPHwYAvPSlL4Wu67jqqqvw0EMPtTzv0UcfRbVaxfT0dJa3R7Thua3VRdEGh2ztlFLpuHko0+B7Y5wuPyEo8dLev4kYLwInuR4YPETkju/EYGJ42p0XPuBsI+7ZcHlreIhzABlmI20lmcSLZVl429vehk984hP49re/jSNHjuD9738/9u3bh1e/+tUIwxBnzpyB4wjl95KXvARXXXUV3v/+9+Phhx/G3XffjQ996EN4wxvegL179wIAfumXfgl33XUXPv3pT+PYsWO4/fbb8fu///t4xzvesaGEXQJorK21/Mz5xhrV9RrMqJrUBWMwyCst9t2G031DYiQ0myJEq857Yjxxm63fj992fSAmg3bxgmjAa7Jnw4/EPTfiqh0GB8Ltv6Z3InM68A033IDrr78eH/zgB/GWt7wFuq7jlltugWmaOHXqFF75ylfi9ttvBwAwxvCZz3wGBw4cwDve8Q68733vw6te9aqWJnUvf/nL8bnPfQ7f/e538drXvhYf//jH8Z73vAf/7t/9u5F9yHOV2spyy8+cb8yFUM5Lr9lGY+G8REmYyK1RXH8z8TwvFqyuS+JlnPHaxEt9cWV73gixqawTLwN2V+deEyEX106OvPwX4GOwIO1EppwXANB1HTfeeCNuvPHGdb87cOAAHn300ZbH5ubmcPPNN/d8zWuvvRbXXntt1rdC9GFtXQXYxuKPO2U8QMgBFdx366NP3CYSVMgIIOdl3PHbciEaS6sYLqOQGGd8T4QDI25CYz4YIoSBD91Yv+hMwz0bUcigAYh4Hjpbg3BeJiBsROws7LW2RlRsNM7L2Oe88OSwdut0Q91M0o3pnAzTa4mtx3Nac1yclcGblxE7B1fmnnHk4rYR3gDzjbhnx13pIq6qh/nYOi8kXiYY224NmTAWINxA9rfndc95GSvxkjqs3SYlJW4mreKFhOIoCOYfg/2t30FUXxrp67Y7L84KDS6dRBqyezLnBjjkddkZwIH2mnGTrAiqNJqcF2IbaE/QY8zD2srwSXqDjAcYh4TdEDpUuZ9H4mVTaQ0bkfMyCvzvfwvBU/cieHr9yJWNELY5L3793EjY5ZzjzJmNjUbZSTTXhCjl0GPx4rn9xQv3bDApXlioimUi8ElJ2CV2Dr4nxAuXjYcYPJw6sX6Mw8Cv5/cfzBgEG6toGgnaCqbNv0BOexqeM56rhkkh7bzYNjkvo4B7TfnvaMWg3yZewnOk2ujWW/8Y//yfvwp/9Vff2O63siU040GMOjgXC81BJktz3waTTU31UCQNMsaBMViQdoLEywQTyMQtdQBrzMfCyYWhX28nVBtFEYehLUFjPkxtHp4znifepKDKpAEKG40KLofjjVq8BG3iJWqcG5V4jz12RP77aJ8tJ4NYqHAWOy/OAENTI7cZixfTV43qoskplSZ2DnHJXJSEeVbPLg79ejshYdd3fTAm3B8GH547Bk7QBENho01AiRZ/tGIwbGtKx+xzQ7yo5onnShNFVy4iWARw2a+lOUCI0HMaSQdmX103o4Gb3G01JF4mmCgWEhq4rMBx6vXuT+hDr1LpsREvTRcM4j0w+PCow+6mQgm7o2eznBfVYdvRxLmqDZLEOQEoUX2uHJ++K75nBgbIjrm1Wv/KsoY8HjjXkZMzsBjjJF6IrYer8eYc8UHcbh1nYUc4L44DBnXi+fB93ucZxEZoDRudGyvbzSYWLyN2XtQ4gDWzDAAwBiifnQQS5+XcEGuhFBuMM2iRuMWvrfUXL2vquIOJQqoD3CDJvtsBiZcJJpJDfhgYmPyqN1KzP8h4gG0XL7bbEjbyAxIvm0k6bOS6Ljin/b0ROOebFjZSzotbqAAArBG//riiHJdzxXkJ5TWYRQCT3cbtAXJeGtJZj7iFYj5RL/aIHcBRQeJlkuFKvMQNZ8GjjfR56Z+wu92l0n7Taw0bkXjZVNJhozAMt/373/GEPsDFOTrqsBH3PEQM8KoFcABm5KdCy5OLclzOlZyXKFT5Kgy6NN/9Pi4b5xEaobhWcpgol3NxqoE9po4qiZcJhvNEqBhMHZjD38wHzXnZztW373ipsBFHsAGxRvQnHTYCqFx6o6QFy6jDRtz3cHx/HqvlU1grijCyvzb5jeoS52U8b8KjhkvxwhmDHorrX9hPCPsubJncy7mB0nQRHOIYsb3xPKdJvEwyUrwwBphqNIC2ceelV9gI2N5eLyJZLfn7wQaHURK9abbZ0efKDWLTSAuWUTsvvg/HEuGAelGcr/XFyR8RcK5VG6kBvBwMhmz3z/t0yeWeDVf1A+M6clNVQIkXfzwbfZJ4mWSUeAGQ06QaRzS0MzJIwi6wvaGjwPXBWPL3ow0OoyR602i0i5fxXKXtFNJuy6idF+Z7aFolAEC9UAUghjNOOkm10TkiXlShBtNi54X16dXCPRueTO5lkQZrqhI7L45L1UbEliMOXI1FKGsyA535aNaHU9K9xgOkxct2Ju36rh+HjQAgZCReNhMSL6OlJc8lCkbbmt33wdW0denATPpwRs75Oee8xDlTYNDUApb3OY68JgJZkapFTIgX+bM7pu0mSLxMNHJOBQfKmgoheVh4drgRAcl4gPXiRdf1+P+3N2zUKl7AQoQhCZjNor389FxZ3W4abW7LKN0XFvrxZPlAF9eGSR/O6Ps+Ipn3du4Ia3m9Ywy6mlWEMK4+7QT3bDVQGlrEYFSnoOSBT+KF2HrEQayzCEVNHJkaPMw/O9yIADVVulO1EWNsLHq9+K4n5nFIGAtoOOMmQs7LaFknVkaY96IFvqifBcA1cUOa9ITdtLg+15wXpmkwNBY/3KtfC/ea6mnQIw6tWARktVEQjOfij8TLRKPEC4cpb+iMeVg5uzLUq/VyXoDxaFTntWXGM3hw1sazydIkoBJ28/k8AHJeNkp7efQonRct9AGofAjhTgZrw3fc3gmkxXQQ+Nveh2prkGJD02CYWjyYt1e5NPdsQLo0esCh5QtQDTZIvBBbjnIgdBbCUuIFPhq17DNNoiiKw0Gdcl7E49svXpTAUjD4cEm8bBqqVHp2dg4AlUpvmPaw0QidFxYE8TWBsQARA/iED2dcH9Y8F47PxHmxLA0RxPXa67Ww8G0wGV4zoxBaoRA7L2E0nr2ySLxMNGKVZbAIpsp/gQe/kT17XFUaAb2cF/H4doqXoF28MB9undyAzcD3/fi4mJvbBQBw3XPh5rB5cL/tWG3/eUO0npeewYAJH87YLlbOBWcwFqi6gVzejIczuj2+a+7ZYDJuZPghtFwOynmJSLwQW46MbxssTIWNgMDLLi7SgqSbeDEMaU/621da5wcdnJcG3VA3g3SPl9nZWQDnysp28+Beu/MywoTdtrYBrmVN/HDG9jyXcyPvRVaZGjqKpTy4dF56TZbmXhOaLHQwGAfTNDBZmsYj1vV52wmJlwlGXawsFkJjgKqT5ENU36QFSaeE3fTj21pt1FZayuDDpYTdTUGFjCzLQrks5uWcCyvbTWUTq40i1noTsq3cxA9nbBcr54a4Fo67ZlooTVXjobyrtZWuzwicBph05y2mZIH4NxrTeWUkXiYacRDn9GRAIxDnZWVChQcMwwRjnZX4OOS8hO3ihfnw7PFssrTTUZVGpVIplbB7LtwcNo/NChvxMARPtTMAAMeyYI40LDV+tIvpSZ8sHUVRkuuYM1GsVMAiKV5Wu/f0sW3hynAO5A0pC9Ril5PzQmwhYSo5L5cTokJX841YdvWSdNft7LoA4zFZOoxUJ2EBgw/PORcqDLYeFTYqFIrI5wsAyHnZKLHTYgirf1Rho9BzEWit4sUzdeRCD7xH/4+dTnsC+aSL6yi1eLNyBRjFYhz+afQIG9UccS5zWCjk5JgApsQLOS/EFuK6Sagkn8uB5cpx3gtn2Q/IXqMBFOPgvERciJdItrbWmA/P3b4w1iSjwkZp54WqjTaIFCusNANgdGEjr+kiaHNefLnC9ta639R2Ou1Oy6TnvISp0SxmIQ8tX4AWChHSK2F3TYo6DhOFglqgKgcmNXJgjCDxMqHUpUXIOVDMG2ClaeTi4Yxh5ht6r4nSiiTnZfvEC5ehslCTczrgwyfxsiko56VYpLDRqOC+jaYX4oF5W5SojqhU2m3a65wXXyyw0VhcGcnfGEfOtWojVW3JuYZ8sQitUIhn8QY9jqU1Xw5z5CZysXiRzgsAjHJMxYgg8TKhrC4tyf/TUcznwYrTKGoq98XD4sJiptdTbkov5yWpNhov8eL549lkaafTOedlsm8Omw33Hfz+Pcdxwx/egb95cnF9DsyQ+LaDUBfnhMphiGQuXH2ChzOea9VGynnh0JEr5qEVCjBC4bKHPaZD24F05bkBsyDOZcYSecCD8csbJPEyodRrMgELOqxcAaw4jbwMG2nMw7PHT2Z6PRU26lZplP7ddlYbqTLBSLbFZiyEN6azOXY6SrwUi+mcF3JeNoTn4NlVsQ9PrbmjCxvZbizoI4jJ0pCT5p3lyRUv7WJ60sW17ymBoqNcLknxIhdvYWcBwjmHqxovcwOaEi/yeOHg5LwQW0ezIePYXINm5KAVp1u67C6fWerx7PXslJwXxKGx5KH28mliNKh8gtaw0WTfHDYb7tuoyzCn7Ucj67Dr2w4ieTMKpHhRXXad1cmdb3SuOS9OU3w+Dh3FcglavgBDtvdn3a6DoQcvkk51xORoAIClw4zbmArQDRIvE4rTVIlqGpiZAytOpxrVuagvZ5tpkiXnZXvnhwjxomsRIhmz9UPKedkMOoWN0oniRDZ46ANRiDXpFNp+uK7vy7D4joNQupGBpoPL3h+eweDVJle8rHdeJtsZXJO9XDjXUalWoRXyMOX1j/HO12Xu2QjkKADGNTEaAICmKXnAxbE5ZpB4mVA8tcLgGmDkwIpTsfOiwYebcURANudlG+OjcuCcrkXgstQv4CReNgMlXqhUejQol0U5L00/HJnzErheHErleoQI0imzcgjrkzucUTktKh9v0vu8NOPvUkO5UoaWL8CUi0mGAJyvz//jXhOhzIPSI8TOi6bLjG5EXUNO2wmJlwnFj1fAGphhQSvNtEyWDtxseSCeN0jC7vY6L6L8WyYlaxFUV+uIUcLuZqBuBNSkbkRIl6UunRfHj0aW8xI4LmRkABrjCJEDANhWHlFjcsWLOh5nZmZafp5UmnVZDs015CwLmmnCkr2vGAC/0+wxz0akBlGHiHNe1PUc4OAUNiK2imRAIQNU2Cg1nDEKsn312Uqlt8fp8L0ATDovzOBqKCoikHjZDFrDRmK1Nuk5BZsJ9x14YQRX5ijYfgh4zkiahIUp8aJrGkJNrKpdywSa3ft/7HSUwJ6ZmZU/T/bx6cbpAixuMmfoSMKEHWZZcc+Ou3oaIY+dF11e6xkj54XYQkIlXrgG1hY2YvAQhXqPZ69HuSm9qo2223nxbRdMTc41dTXSA5xFY9slcieT7rBbKJDzslG4Z6ORckRrXgTwcCSVHqHrqm7vYLqGSIkXUweb4OGMifOiBodOtnjx3FS6gMQwRPM5AHC7ihchmPUwhCZdVCunFqoR5bwQWwePk1SZyHnRTZh5VWUARBm/+p1QbeQ3HDA5GVWzLKjuTIwFNCJgE2jtsEul0hvGd7DmJa5lQ/YnGkXoKPI8RJqceWMY4DL/xTc0mBM8nFE5LeeK85KkCyTziCxLR8TFddvpECLkXhNMihcziKDLhF1TihggomojYuuIUhU2TM5JMUrT8SGtLl6DkqXPy3aJF6fRiOc5MbMA6KqvjQ+3PtkXre0g6fNSQi4ncigcxyaXa0i476Cecl4cGT4aRZfdyPPimWaGZUKT50ZocJgjnFw9biin5VxxXkJ17U05L/liPnZearUOPX08GxoXx50Z+HHYqCBFDGOcmtQRW4catsa4yHkBAFaahiGDm5EGhOHguSAq52WcnRdnLSn5NIw81CgXBg9ObXLj+ttFMh4gqTbinMdCl8gG95IeLwDgqJbtIxAX3PMAKV7MXA6m7K4baSFygQMeTWZemHJaZmfPDedFddhlaeelVIonS6+sLK97TuQ2Y8faCoK4VLpYqsTbdMqV2W5IvEwsMmQCgBlSvBSnYcnKGw0eVpYG7++QZTzAds02UslqnOuwjBw01WUXATkvm0CnPi/A5K9uNw3fiSuNgJR4GYHzwlPOi5UvoKjJGTh6IKpQJrRc+lyrNlJTpdUICAAwS0Uw6cR0cl68Zi2WOrnAi3NeSuVyvE3dGb/jg8TLhBKvpBgAGTbSitPIxeXSPp49+uzAr5cl52W7qo08lWnPdVh6DoaczcGYB69BN9RRk+6wa5pmLF4n/QaxWXDPRi3lvLhBKEJwo3BefB+QC5d8qYSqJnIjVJfd+uJkjgg416qN1PTndFKAViiAyVKzZoecl7WmWMRyrsNkHEyex+WpErgUPTY5L8TWkXJezLTzoiqOXJydPzPwqynxktT+r2fbw0ZO0hrb0vMwZdyIIYDboM6voyQIgliklEoiEZx6vWwM7jtYdRLnJYo4/IiPZjij70H1QMoXi5gzg/jG5Bka6ksrG/8bY0YQBPG16FzJeVHpAjx1a9fyyWTpTuGfVbshn2MiZyXPK5SK4BDXUHtEzRJHCYmXSYUr8cJTYaOppFEdfNSW1wZ+uZ2Q8+KnygRzZh6WJt4PgwfPpjyMUdJsJhfBYlGJF6o42hC+g9W25pH2qLrsBonzUiyWsCunI4LszZPLw55A8ZI+Ds8Z5yW+7ifei14oQKY4IfDWn5urjljYRdxELpe00DALOUCKl6Yzfos/Ei8Ti8xtYTwJG6W67GrMg7M2+E1mR5RKpyaq5qwC8nGTJR8uiZeRoux4wzDjY4KGM24M7tmotU1Ad/xoNPONfBdMDi0tV6cwbeURQnxvtpWDszr4QmanoI5DxhimpqbkYw6iCU1OBiD6AgHxaBRAhI30SFz3O1UNNeJQpQkjn4sfN4v52HlxnfG7fpJ4mVjUHKP2hN2kUV3gDD4iICmVHmQ8wPYc6Eq8cK4hbxZRMGWpH3z47vj1KdjJJMm6xfgxChttDO47WHNb88VGNd8oPVG4XJ2CkSshlM6kaxrwJ3A4o3JZ8vlCXPYLAG6nFvmTgppdlBYv+QIMVVnaoVOurUryIz2uNAJkrywVWvTG7/pJ4mVCYcp5QZiUSqfDRszL1LhTVRCNs/MS+EmDpmKuiFJONuWDD9eh4YyjJN3jRUHDGTdIW58XALBHNd8oSs7JqekZsFwJkSZbxpsGgvrkOi+FQgG5XFINZ9sTLF5UriNL5bwUCjADmcgbrb82q76I6YnSAMA0DUoieF62WXhbAYmXiUW2e2YckK3AmW7CsoSQ0eCBZxgRMEjYSDkv21VtFEhLlDMNxUIZlaLoU8AYqPfIiElXGinIedkY3G/t8wKonJcRdNiVTiznGiqVCliuBKguuyZD1Ji8PkiJ85KHpmnx8TnZk6Vlg0g97bzkYchrssbXixdlvGiRFpdJJy8nJELgk3ghtgrpsOga4gFdAGDlRe0+Yx4i3r1yqJ2d0GE3UD0OwFAqVDBdnIJq9up645dwtpPpFDZSq9tJT4rcLLjnoNG2wm16ITCCaiMe/2uglC+A5UpJl12dT+RwRiWiC4Wi/PdcSCiXzotcsAKAVizCCtRk6aClAzYPfYSyJ4wWocV5AZKqpSwNTbcKEi8TizjYjLZv2CpXAYiclxC59id1JctU6e0SL2EsXoB8oYRCoQIu7VNvDGdz7GSSoYyJ80LDGTeI76ApPfxCTpxLjRGFjVSDOtFGwASzirB08bciPYI2hn08NopyWJTjcm5MPpfpAnriqmv5AnLSlWbgqfC6SBJXaTJaaqK0gklhE4bjN/KDxMukIisLjLbIUK40LX4ND4GWG3gOjUrYGqxJ3fYIhYiLizFngGbloedL4LJkMOCU8zJK0kMZFVQqvTG4b8OW4qVSFDfchh+NJGE3kiKeQwdjDFqulHTZ1XzoEzicMXFeWmf1THJOFpPl8KrRHABohTzMwEv6+qSFqpfMIjNCxEMZE+RxE5F4IbYIlbBrtlkvueou8XsmDvT6gG3zB8t5ESfMdjkvkYrnMlFhxcxCXDIYkXgZKUnCbqdqo8m9OWwWPAzguh4Cac9XisIVbfjhhkulOefxnD7lRLJcEdNMvC5jAYxRNMIbM9LVRuLfcyGsKXMdU9dpLV+AFXnxcEY3lfPDPRsqtq6HUYecF3X93Mz3PBwkXiYW4bxYZutXbJRnoSHpsnvy2KmBXi2ZbdQ/52W7EnZVjwMxEiEHWHl17iHE+MVsdzLJUMb1CbuuS/lFmfEdNLzkvFHipTkC5yXyfYRa4rwAAMuVMMua8WqcwZu4aeAqbHQuOS9xuoCZpARopgmDheCyr089VRbPvSaYjBuZYbgu50UNGojGUL2QeJlAwiAAkzHugtUaN2LFqaTXC/Ox8OxCz9fioUjwUjkvhjG+OS8cqkETB9M0MKsIaDLWz0i8jJL0UEYFlUoPD/dtrMm+SznTgGEJF7M5gpwX33YRyrLoxHkpYVfURATpluXy8Ces4kiFjc6VnJcoDOPrvplrzWc08nlwLo6ppeWz8ePcs8Hkos8MgnU5L0q8jKOuJfEygdjp1u351oM43WWXwUNtcaXr60S106j/0b+H+/d/nAobjW+1kXKblGBhVgGQFRUqB4gYDSrnhUqlRwP3HazISqO8ZUDLKfEiqo024op4to1Ql85LKmxU5hwhE9cH28qh3uNasBNRIqW92mhSxUuY6p5rFltFiF4ogEVCwC6vpIZwek1oEI6fFQTrnBfWMuJxvCDxMoHUVlYACLVcKraKl3SjOo15aNa6VxkEzzwA+A6CEz/YEeMBwKTtLsULrAI0XXWcDMeyV8FORYWNWjvsTvbNYTPhnoMV2aAuZ5nIQ/Qosv1InMgdOqMOitt0EMmeLkg5LwxAyIRIci0T9aXJmiytHMB252VSncEwVSiRLxYRNVdi0asVCnHPlvpaEjYK7HqcH2kF3vqcFyURxtB6IfEygawtr8j/01tWxoAcEYDEefHt7kIjnH8cAMCbKwMOZhS/2w7xIk5SGTaSkTJmJuKFwYc7YHIy0Z/OHXbJeRka38aKChtZFkwjyXkBsKG8F992ECrxoiXOCwDwuMuuPnHDGRPnpTXnZVLFdSCv0ZzrKAeLaPyf98G7/ysARKM6LRLHgN2sx89pNFfkcwAr8NeXSsuCh/GTLiReJhKlrDk0WLm2g1E3YcqqIAav64KOc45w/jHxQ+AOONto+6qNojDJ84EhxZluQLU7YPDhrk1WTH87se314uXcSIjcHLjnoKacFzMHU5fhHNX+dANddn3bRaTcfyVeNB0w8/EdIDAYnNXJmm/UrdpoUsW1L6/RHDpKgfgug+PfAyAa1SnxErg2vvOdv8ZNN/06FpYW5XNMmKELrdDqvCRjBjh4NF7OdWbxEkURbr75Zlx77bW48sor8e53vxvHjx/vuv3y8jI+8IEP4NChQ7j66qvx4Q9/uKvy9TwPP/uzP4tf+ZVfyfq2iBTNulTWXAczc+t+H48IYB6ioPMhwFfnwR0x70Qk7Gbr87LVlQt++uKuJXFaQ12smQ9njW6qo6LRWN/nJSeTBCf15rCZcN9OxIuVh2m2ipeNJO36tgMuzwmmJ+c7y5VicR/oHH5tsuYbpWcbiX+F2zSp4wGa8XwqHUVDNiBcOg4ehdDyeehSe4S+i9/7vc/gK1/5Eg7/4EkAAIcFI/LWJ+yqRG/xxM3/EBnILF4++9nP4tZbb8VHP/pR3HbbbYiiCO9617u6zo654YYbcPToUXzhC1/Apz71Kdx555246aabOm7727/923jssceyviWijaSOXxMlw21YeXESi/LIzgm4wXzyPYQRj8VIr/EAarYRsPXl0oEUL5zr0FJvMSfbZGvw4TVIvIyKzn1eqEnd0PgOajJh17KKMA3Zk0TmafEN9GHxXReRvNKnO68yqwhLk40d9RDB2mSJFzWAMcl5mWznpbYqcpY411HQYqWCaGUeWiGZLM0DD4uLwnFZVG40N6BpDKxtcapp6njh4MF4zYfLJF48z8PnP/953HDDDbjuuutw8OBBfPKTn8T8/DzuuOOOdds/+OCDOHz4MD72sY/hiiuuwDXXXIOPfOQj+OpXv4qFhdYS3bvuugt/+Zd/iUsvvXRjn4iAq2x7roF1KG22ZEt3xrqPCAhT4sVL1fgP4rwAW99l15Nj7jlMaKmhZJYu3hNjPtzGZF60toPegxlJJGaFew7W5FDGnFmEIcWLq5LMNxA2Chw3Hg/QIl5yRZR00ZOHaz7CRr3j83cqjqP6vJwb1UaNWHxqyCERGtHZZ6DlC9BVi//QxcrKMgCgpq6JXIeWL7TMwQNE6F1usLOdlyNHjqDRaOCaa66JH6tWq7j88stx7733rtv+vvvuw+7du3HJJZfEj1199dVgjOH++++PH1taWsKv/uqv4qMf/ShmZmaG+RxEiiB2wTo7L7niFACRB6JKJdtRybosX4GfGso1yGwjYOvzXlxZusthQLOSwzonxRuDD7cpLtTPPPMUvva1LyOKqPfLsPTu80IiMSvct7GmnBezFDsvrh/Evx+WMCVe9FTbeJYrtXTZ5RM2nFE5L2rm1qRXGyWJuBq0lEsSLh6DXijADMWx5Dk2wlAcaw1HXBNZpK3LdwHSxwsHxmw+nNF/k4T5+XkAwP79+1se37NnT/y7NAsLC+u2tSwL09PTOHUq6ez667/+6/iJn/gJ/ORP/iT+8A//MMtb6orRPpFwBOgyXqzr453nHMaDtxj0XH7dvshXpwEAGnMRaSYCP0C+kIiSqLECXjsNMAbzOVfBv++vxfaahny+u3jRdROMMXDOEUXhlu4vT564nJswLCP+zOVcAcAqGHz4rg/D0PCbv/kRHD58Ny644AIcOnT1pr+3Qdkpx1cURbHzUqmUk31dFitc13U25fzrxE7ZZ/1ggRtPlLasEgxD3Gg96bxooTv0Po08D1y2DzBknyZd16DlS9iFBp7leTAWIfSbW/a9bQVKpJTLJRiGFpf1O06243OzjrGzZ8/g//l/fhVvetPP47rrfnLDr+c5KnTOWiaR88WjMCpXw5JC2HESkWo7cmBjpEEvFNftF8M0pekXQUMw0H7bqnMyk3hRdlt76CCXy2F1dX2PANu2O4YZcrlc3EL8tttuw5NPPon//t//e5a30hNNY5iZKfXfcEiq1fYuhOMFl6oanKEyPYVS276o7zsPgMh5AUS76P3nXZT8fv4hAIC152KUz7sodl4sy+q7X03ThOd5KBaNeD9txf7SoiTT3srn4vc5U6oCmAdjIaIgxMxMCWfPngYANBorm3qcDMu4H1/1ehJeOP/83bEdv3fvLABxc9jq/Tru+6wfHvPRdGVSvFmGaYgbrR+ECCOOvB5iesh9avBQdJ0GkJer62q1AH9qGrsiFxGmoaOJAOFYng/D4nniHrN79wxmZkrYvVu4+r7vDvU5R32M/dVffQ133fW30DSGn/u5n93w60WyVJpxDXrkqpadCBePoXLxP4cZKPGSJCw78pjTIgarUlq3XwrFApprYg5epagjn2G/bfY5mUm8qJi253nx/wNilklh3UwEsX2nRF7XdVEsFvHUU0/h4x//OG655ZaWxL+NEkUctR7N14ZF1zVUqwXUajbCcHxDDqGv9jlDw43gLbfawZGVhI0Ajsd++BR27d0V/775+MPi93sugcOK8GWs1LIsLC/3tpYNQ4iXxcUaKpXZLdtfK2dXxP9wHQYz4/dZtJKTrbZaw/JyA6uyJHR+/mzfz7OV7JTj6/RpIf50XYdth/FKznXFcWLbNpaW6uvi55vBTtln/fDqdTgy58W0KjD05HrqBhGaq6vgQx6rTr2JeGCfTKqv1Wy43EQeHBEsIV4YH6vzYaOo0GYQMCwvNxCG4nis1xuZPudmHWOnTp0BACwtLY9kvyed1TUEdpJ8zd0mmvU15OR9QbktAOB6QrzoEQc311/ftTjnJUJteQ12sf/73Mj+qlYLAzs2mcSLCgGdPn0aF154Yfz46dOncdlll63bft++ffjWt77V8pjneVhZWcGePXtw++23o9Fo4J3vfGf8e8dx8MADD+Cb3/wmHnzwwSxvr4Ug2LwLWRhGm/r6GyUKQ9nUmSHULLC292pWdovfMg4gwNn5pZbP458Uybpsz/PB8xV48gA0DLPv5zZNcUg5jhcfuFuxv2wVr+c6DC0X/71yvgzONTAWoek0EQQR6rKksFarjeX3OO7HV02W1BYKRYQhh2phlc6HajTslgXOZjPu+6wfoduEIwczWkYVmp4HY6J5mB2ECF176M8Xui4gZ3tZOfGdhGEELt2dkJkwAYQa4PvhlojOrUDlvJimuB7E5ef2cPty1MdYTQ5IXFtbG8nrJrmODJErhYxZAHwbob0CK/TAuYGm20G8hADL5de9DzPuExYh9Jx195JebPY5mSkodfDgQZTLZdxzzz3xY7VaDY888ggOHTq0bvtDhw5hfn4eR48ejR87fPgwAOClL30p3va2t+Gb3/wmvvKVr8T/vfCFL8RP/uRP4itf+cqQH4lQYSPGGViHhF2jujueLK3BQ3M1CQNwz0a0dAwAoO+7FKw41eK89GO7uuz6ygrlGqxUbxsjXwKXh7nne/B9Lw5Z1uuTVRq6VXRK1gWAXC4RK5OaFLlZBK4TixfTmgKYBdMQlUG2H20oYZd7fjzbK19IvjPVZTeSFUiBqcNvTsb3FkVRjz4v4/EZ1fUnHYbdCPF4gFTOi77/+QAAZi/CDF1wmLCd5NqsxIsRRh0mSgOFoqpM5YiC8ZoWn8l5sSwLb3vb2/CJT3wCs7OzOP/88/Hxj38c+/btw6tf/WqEYYilpSVUKhXk83m85CUvwVVXXYX3v//9uOmmm9BsNvGhD30Ib3jDG7B3714AwPT0dMvfyOfzKJVKuOiiizq8A2IgeAQwMQ+0U6k0001YGuBEolzaa6Qy0xeeADgHq+yGVpoB93LwZVVOrx4vimS+0db2BPBdVfKnIW+kVvxmAWoyahD6WFtLLhTp/ycGRw1lbBcvhmHANE34vk8VRxlZa9QRyvPMMisAAliGAc8PRa+XDZRKc98DLCFeiuVy/DjLyRuToQEh4BsM9cUVzJZGF8LfLtQCBejU52U8xMvamhIvo1lERXLRysHiWUTG/oMIjz0E3liAzgNwbrY4L56vnJeww0RpoFRJznHXbqD/8nXryJwOfMMNN+D666/HBz/4QbzlLW+Bruu45ZZbYJomTp06hVe+8pW4/fbbAYi5CJ/5zGdw4MABvOMd78D73vc+vOpVr+rapI4YDZyrYYToWCoNAKZcbTF4CN3E2lP9XfR9z5cbFuDJYUGWoaMf2zUiwPfUxUpDzkxOQmYVwKUNHkYhGqleFo0J62uxVaihjGolm0a5L+Nyg9gpLKdGV1hmHpwbKecl3FCHXe678eiMYrkaP67Eiy4nr4c6n5jhjGl3RZVIp/u8bHUH8E4ox8VxnNFcL2UpdDwJmjHo+0TfNF47KR7lBpodcl6sIOzovJQqU/H/N+zxul5mcl4AkaR344034sYbb1z3uwMHDuDRRx9teWxubg4333zzwK//x3/8x1nfEtEOV8MIecfxAABgGSbgu3JEQHIYqP4u6qBnjCEwxU3KHCCRarsmSweqPJwz5HPJTZVZBSHiOBDxsGWVQ2Gj4egWNgJET416fY2cl4ws1YWblbNMMU+GM1imEC8Nb2NhIwTJc8uV6eRxS5wnOS0ABxDpIex4qOvORolny7Kgy4WaEjFRFMH3/YHC4JtJ67WovuEeZ1EUCjeCS/FiFqDNXQAwBuYnI2NUVRuQOC9mEHSYKA1UpkpxzmBjzMYqTE5RP5FCiRcAHcJGAGDK+UYMHngkxAsPA4SnxawLFSsFgED2nDD1/ol8akTAlo8HCJJktUKqwghWMZnqztrFy3itJHYKKmzUPrEcSG4QKlmS6A+PQqza4vjNmRYYD5EL3NjpbGwwbBQGMpzAGcqVtPMixEsZUtxoPpyVyRjO2D5RGkBLAvk4zDdKX39GspCKHXdZrmEVwIwctOn9YBoAXQeLtJZqI0/2frECr6PzkisVwCGPw+b277M0JF4mEnEQa4wDWmdzLRfPN/IRykhmdPYZIPTBcmVoU0lzwUBOuDUHKELYLucljFtXMxSLqbi+lY8tc46oJc+FnJfhUGGjTs7LuOUV7Ah8ByvxUMYccoENHTye/i6cl+HFS8Rll17oKKVCfSpsNKOJ74oxH26Hfl07EXX85VN5HKZpxourcXAGR+4C89bKHmaJz67NifxRzdKhca0l5yUII4RhhJzfYSgjAKuQB6R4GbcFCYmXiURWEjF0LXtU84005sYjApJ8l0tbnudrUrxo/ePE6cnSW0kYqiZ1DOV8JX6cWUVomlqRtDovaxM2iG6r6DSUUaFuFumESaI33Hew4iUTpXNwobMwPpcaQQTuDS8GI6ibmh7P+gJE4j50C7t5A5yLW0Gztjz03xknOjkv6Z/Hw3nZHPGirt1M5v7pu6R40Rm0EGg6rddmx4tghZ2dFz2fi50Xxxmvc5rEywTClPPSYxurWJXb+gi1PMIwSvJdUiEjICVe2ODiZaudlyhSq0uOfNp5MfPQ9ES8pAULhY2Go9NQRgU5L9nhno1VJV7MAgpGBEOLRF4aIMYG+M7QSaYcynnU1w/eyxUxxz1EEN9bc8ySModFOSv5NjdhXCZLc85brj+jqXxU4kX+qJwXKV4YAkR+AF+GEdWx4PgMRuR1zHlhmgZIYet3aDi7nZB4mUik89Lj282VpwEADEJNn5lfTCXrtoqXQBMXUZP1bzi0XdVGyhoHA3SrtVTaNGQWPvOxmkpIbDTqI686+J3f+RR+/uevj0Mrk0ivhN1xuTnsKHwnGcpoFVCwAEMDTJmv1vQjiMF4w618uaw+USvoNCxXhAExjR0A/GAyvjclsLs7L9srrh3HackLHE0IO5XriCRspM+JhrKMhfBkHx9N01ApSZfUE0dIp7ARgJR42do8xn6QeJlEpMjoNUMrV50T28hkvYf+/iFwtw7oVhwjVfhMipd4WkZ3krDR1h3oQoAo5wUtjfmYpsHQVb9hH6tLiS0eRdHIRcaXv/xneOSR7+MHP/j+SF93nFDipVOp9KRP7t0MuO+gIUcDWGYJxaIBy9RgSudFiBcMHTqK5N2Md7jcM5ncHknxEvL+5/hOIHFeWt2EJKF8e4/PdrEySvGi6a1hI5YrgVV2g+mAqyZt5yzk5JBOVwrnTlOlBeK4UYnf4wKJl4lEipce1UG5KTEiQIe4ER19MkTAdeh7LwHTW5N8fSZWbAbvL0i2I2wUhQGU2wSNA6m4PgDk5Odh8FFbWWn5Xb0+OvEimjQuApjsfBq1qu3lvIxbct84wz0bTSleclYZpXIOOUuL29k31T1jgKRd13XxhS/cgqeeejJ5fRke4KzD5V5WHHGZ2B8N4K7uBJKcl1aBrZyX7RbX7SHr0YSwxTVQV+H9lAOt77oImg44tlx45C1Y8lrtyT5f3Z0X2SdrzGaHkXiZRGQrcKNHU7mczHnRdB95v46Q5fGw/fy4v0uaQDapMxGCB73jntvRYddPlZEyxtfF9WPxwvx1omKUFUcrK8uIZJfUtbXJKDntRO+w0XjcHHYUvpMMZTTLKM2UkM+bcdioEQzuvHz3u9/C//yfH8enP/3J+LGoh3hJuuzqctvxukENS1Jt1M152V5xvZnOiy4LK1TYCBB5L0wHXNmJvJizUtdqcex1ynmRzwYAhGM2O4zEywSiEnYtq/vXa8lS6QjAlP8IAOAx+0XA7vXixZPHrKVr4HbvUkpViriVzksgu+tyroNr60+wgpy3xOBjbVMuGoKzZ8/G/6+Grk0ivTrsFgqU85KVlqGMZgWVXVVYOROGbFFgq7DRAM7LmTNiUvHZs+JfHobgmuq42ilsJL5DXQ5U5fp4hQaGZdyrjdqdllEk7DKVLiD/Vd8tAOhzQrzYcgZcMWfAkt+56/pgubxIzu38ygCAaAy6Eqch8TKRyOQ/s3sHSSsvVlwcDHueD+SCBgJWwuEj6x0T5aKYOgNv9hYv2zGYUTkvHCaYtv7iW1KxX/jrRgKMsuJocfFM/P/nqvNC4wGy4zZqcJV4sSqo7JlBLm/ANJR4kU3mBuiyq467Wk2cp77jIJTihWudEnbFd5jLa3KbEN4EhPy2qtoochw0Hz0CHmVzJdoXTaMZVSKOE9XSgpmJk6LJsJEtz8ti3kROihff83vkuyAOG/GIxAuxiYRBEDdlK+S7D1LUDROatJOfry9hlycSTH/44Jl1sU0lRExdQ9Rc6fn3t6PaKHZeYAD6+otIVTXkY0BDugaaXGVslvMyyeKl22BGgKqNhsG163DkjJm8kUNx9yysggVT3nwcKV4G6bKrRIty/jzbRaSclw4ra9Vlt6Lu8Zo/EfONulcbjXay9Nk//1Oc+PhvYe2+w5metz7nZWPXIVG0IB13JnMT02Gj4hS0QgG2ChvlrbiDs+f50LvluwBQzsuYGS8kXiaNRuokKBW7H5CMsSRha2UBB3Y9AStoIkQBd9/5g5ZtPTW8S9PAGys9//52VBvFzgs3wDo00qsWquBy9dCUF7U9e8RU81Em1i4upsXL5Cbs9gobJTkvJF4GxbUbcdiopGvQy2XkSnkYuhAvrsxJGMR5UaJlba0Gzjm8ho1IXuVZh9lkynmZS7cTWDizbrudRvdqo9E6g+6JE+LfY8cyPU+JFXUObVS8RGEQ93dR4oWZrdd/bXoX6nIGXDFnImck/Vs6NahTsFi8jJd6IfEyYag+JpwD5XLv0faWnG/kc4ZLch5mvR8CAB45fKLFfVFhI0Nn4H2cl1FVGzmOgyNHfjjQCROkwkbokPNi5cuIW1y74qK1b58YfzDasFE652Xnr147wTkfyHlRiYFEfxy7EU/3rZqGWFgUc7HzoubP8AGcF+X4BUEAx7GF86JSXvQOYSNZKr0HzbjL7tL8qY19oDGgX7XRqJyXYHlJ/CurDAdFLW727x/NdShMdTTPM3G9TifsAoAxuw9rcrti3kLeEgeG22U0QEJcaz9WkHiZMNbiUmAdxVK516awcuKA9TmDxoDKj+yDGdgIUMThux6Jt/NkZ0VLZ4j65ryMRrz81m99FG9+88/hnnv+se+2Sc6LAWasP8OYVQCHhiiK4MmTNxEvmxU2mkznxXHsWFD2qjba7mqOncRqo45Q5hNMyVBvrlyAIQeierIPxyCl0ulE8VqtBs9xVI8xaB3EiyqVngoa4BCLmaWzp4f6HONEt2qjUZZKc84RLIu+UX7q3B8EJVbUdWij14tALjA5B/IQx0m7eNF3n4+adMQLORM5ObVcOC89cl5koveYaRcSL5NGQ668ODSYuT7OS0GIGw8MrLoXP/H6N2HGOwIA+P7dxxHJC6oSL6augdsrPV9zVLONVJ+K733vob7b+nG1kQHWobeNsE8ZXD9J5t0M8dLqvExmzotK1mWMrUuGBNLVRpSwOygLstKEMYapsth/+WoJphIvQQDO+UCl0ulcq1ptFb7tgsscOL2T8yKvEZprg3NZmr22MvyHGRO6VRuNskldWF8Dl2LAX8zmvKjrzqgc4MTpNJDn8rO1Oy97LmpxXlTCrhv4PZ0XFjsv4yVfSLxMGE3VdI3rgJnrua2ab+RzBn3fpdB1HaWrzocZOghQxH1/J8JILdVGfXJeRpWwq8IuR48+03fbdNhI6yReLCleVFJkPo/p6RkAmxc2mtSE3fRQxk5DPylhNztLcp9apoVKWZyzuWoRhi7bGUQcfsQHKpVud14C20nES4fqQxU24m4TnKvqk50vPDtNlQZGW8qvXBcACFdXEGXobaWuO/v3nwdAJBhvJE+wLq+XHDoKTFzn2nNe2Mw+rMm/UcylxIvfJ+clTvQm8UJsIm68otDAjO6l0gCQq0wDEOLFkPOMfupn34gZ91EAwEP/8Iy4cEohYunaluW8rK6Kk/HYsaN9t02HjTSzQ1dhqwAwxM5LqVRGuSwu2qMUL+dC2KjXUEaAmtQNQ60pnMOcmUepIm6uZqkEU0vOX8ePgIESdldT/19D4CbOizo306iEXfAQgLiZhUPOUBonuvd5UdVGG+/zEiwt9fy5F8p52bt3X/zYRsqla6uJ455DJEI9bdf/RuDH8qOQN5GzxPftBL1LpZkssR8v6ULiZeLwZEIquAYYvZ2XXKEinsM16OcdBAAYuo7iS/bCCF0EKOHBu48kYSNNA3fWwKPuK4RRVBtxzmPn4vjx/uLFc5NqI8My1v2eWQUwMLhSUJXLZZTL4rOPKmzkeV7LjcO2m1s+nHIr6NXjBSDnZRjqrji/clYe5RkZxjFNGEyHISuEbD/sGzZyXTc+VwHh/gWOG886M3MdFjOGBaj+LzKsxAcYAzLubEWfl7TzAmTLe1HXnenpmfg9bWQh1azJ6xjXxUw7q7DOGV1VbpRpQte02HlxgqB32CjuD8TBo/FpYkjiZcIIXXXx0sD6hY1Ul929l0Kr7okf/6k3/KvYfXnwrqfjC6IaN8Cb3UMio3Bems1GLH6Wl5f75o94chXFYcLIdVhdWgUwjceNwMrlSkq8jMZ5USEj1WEYmEz3JR026gQ5L9lpOtLZNIso75qKH9c4YMpzrumHfcNG7RVutVpNXA+keLE6tH9njMWdWLV4JtjOFy/9O+xu/PhUlUYKf3Fw8aI66lYqybVoI9cLO3ZtNDDW2qBOsSpfvyirTFWHXTsKwbqOBgC0+JrGgXB8FmQkXiaMILZ82TrbsB0lXvy2w8DUDeRfPAczWoGpPYEVaYfmiiLBt1foKBkPMPxsIxUyUvQLHfnyRsm5DiPXYQVhFaBpUdwvQ4gX8VlG5bwsyoS9Xbt2xa+9tjZ55dKqx0u3sFEuJy6M5LwMBo8i2K7siG0VUdkzG/9O44gbiTW9COjjvLSL/FptFaHrxrPOct3yGmToyFLCX9v54kWJ50JbOGSU4jp2XqTDEWRI2lXXHeECb/xaFH8eWVqWHg2gWFkR77ckxYsKG4Wcw+9UiSbRlXhhUd/ZdlsJiZcJI5Q3aHAG1idspEYEeE5r/DcMA1x8YA6V3HeQ15+A3RAXRaskEnx7ddkdxXiA9hXksWPP9NxehcoY12DlOqwuzQJ0LYqdl0pl9GEjNRpgbm4XKhWxn2q1yXNeevV4AZKVreu68ZDKzeRP/uSL+O53v7vpf2ez4L4NVzovOaOI4p6Z+Hc6GEwzi/PSKl7W1mqIPBeqbXyh0Ll1gqo4KpbE9YKx8blBDQPnPHZW8vnN6/PiS+fFOu988fMQYaNRucB+LMaEkGovkwaAJbkILctrtJUa3Gv3qCTSLbUIjsbKeVmfIEDsaHioVk0DiJfCevFy+vjjuP87f4ra4nzcmyiUcc5caQrgjT7Oy8arjdovwv2cFzUegHGGnNFBvFhFmFoYOy+lUhmVSnLB4Hz9JOqsqGTdXbt2wfd9nDp1ciKdFxU26tRdF2jtq+G67jrbfpScOHEcH/3of8WuXbvwN3/zD5v2dzYTr74Si+qCmYee2q86Ayx5PtX9qK94aT/earUawqIbD+wrlruJF9llt5pH7YzoshsGAXRjZ94egsBHGCrB1jlsNJKcF+lkFJ53KbxnTwzcqM513fj6OCoXOFC5TqqpT4ewkXJeKmpQLWPImSZc34fdY6FhWmp7DmywBcYoIedlwohS4qVfqXROrko8uwGnUcM9f/V/8N0//TRqi/PIFcoo7RFJvGoFbZVFPL7XcMZRJOyurq60/NxXvKiLOteQ77DigGHB0oP4JlEsluILhuhEuvELmcp5Ec7LxmPY44oKG3VzXnIp52uzQ0dnzohmamfPnt3ScRSjxK0vx6MBim2lzIYOmFJANPwQ8B1w3v0m0368ra3VEKZCTRVZXdiOKpfeN5WPW3ksntm5IwLSrkq3sNFGq43SDeoKl14KYHDnRYkUxhhKpdJIXOAwDtN3d16W5futpERpTl6v7bB7Im7iZofgY+S8kHiZNKRLwjjrWyodJ+xGIW7/w/8fnnnkMACGS178Y/gX/+bXcf7zXwoACOSoALMsLO1ezssoEnZVzotycfqJl1DGYRkH8h1yXsQcp2RGTMHMo1gsxW7LKCa6Js7LblSrQuRNYqO6fmEjXddhyZXaZiftLqeqPdrzpHYKbn0l7j9UzrUuNgyDwZL5Bg1Piha/exmzCreqvLNarYYoVV5dqsx0fJ4KG83ldTFiA8DJjLN6xgnV3dkwjDiMrVDOi+/7GxK8kd0Ed8V3UbhEiJdgZTluWtcLJVJKpRI0TUstdoa/DsXjAXgv8SLCRlN6Il7i46vH+87HM/IiIByfkCKJl0lDrswY0LdUWjesuGW47zmY2XMAP/WW/4iX/dSbkCuUsP+iC8E5EErnJScvfr1zXjbeYVddhC+9VPSe6Ste5GqAcaCQ62yNFy0rdl5yhilXPaNL2m3NeVEXo0kUL73DRsDWDWdMO3TKEt9pNFPOS6VtkKplavH51AyEJdKrXFqJZdX4rFarxTc1znWUu4wLUcmdeugAXPy9xYX5oT7POODIMHj7aADxWCG13fDHp3JdtHIZxq5dYKYJcB7nwfRCiRTluIwibBQFKccdAMz14kWdIzNakuuSV85LjwKLgjxuGOMIx2hmGYmXCYOr2CUDmN47Zs0Yw54Dl8LMFXDVT1yPn/r5/4S5/RfFv9+9aw84rHhIY64iKiE223lRF+ErrniR/Hm1680pDPzESucMxW7iRbdi58Vi4j2OYsWjUM7L3NwcqlWVsDt54qVfnxdg9JN7u7ESz/HaueJleXUJjjxXZqYqLb/L50yYcmVcl65+r7wXdbydf75IIF1bW42FPYcOS1/fRgBIcl642wBkl93G6s7cn0DivHTKt7IsC5rsGLuR41OVSZszM2CMwZidE48PUHGUrjQS/248bKT6r7CeYSPxnmdS6X352NnrLl5KleSa6rija+q5UXZmRhbRA+W8DNYP8dqf+0UAHJq2vlRO13XwSEckA+G5qjhBB8l5GUXYaO/efdi7dx8WFuZx7NjRuKV/miBtozPEbko7JasQOy+GPMFHWS69JJP15uZ2x9VGk+m89O6wC2xdo7pW52Wl63bjzEoqYXfX7HTL7woFC6YM/TZUSkKPLrvqeDtw4AIAQsyomxqH3j0pXYaNuNcE5+I64EmRuhNJKo06h5Dz+TyazeaGKo6CJSHujBmxoDN37YK/MD9Q3ku60ij970aqjXgUIv3tdhIv6hyZ0TWscQOMBcjLBW6zh5CrVqfj/1+r11Ad+l2OFnJeJg4hNAYtntE0raNwUYRhcohY07vFX7BXE4enjdFUG60AACqVKi68UDhB3UJHftxdVwcQoZivdNyumi/GzgtCsY9GWS6drjY6lxN2gdEOv+vFJIiXpl2Pw0Zzu2dbfpcv5hLx4quw0SDOixAvruvCTTkv3WhxXuQtIejxd8adpMdL50o3FfLciPOiwkPNfB7f+c5fg01Pi8cHaFSnRMoow0ZoS+Tu5bxUTQNc+hYFKV56CadytQwuq5jWmhsfqzAqSLxMHMp52VjpryKdx5WrzAlVxDm409lVGE21kXBepqam+osX6bxwGAALYHZpc13Ml+MVbhSI1Why0diYFdpsNuLqBRE2Ugm7OzOJtBf9OuwC2ZyXMAxx//33DlX9kRYs7RVqOwXbacYJu3vO29fyu1zBhCnz1ppyLhfv6byI4+28886LXZamn3Tc7kbc0MxtAlLkjFMzsqwkzkvnrrHq8Q05LzLn5QuH78Z//I//AXedPCEeH0C8qEWNuv6MZLEjF5Pxt9yW82Lbdnw+Vg0zniBekjmPva6BuXIhFr/NEYTYRwWJlwlDhYu00WgXBFHyQqZlgeWFadgt72WUOS+t4uWZzu9POS8wwVjYNUlZsxLnhclVyqicF+W6FItF5IMmiq64sE2i89JvMCOQLeflm9+8Hb/wC/8an/nMpzK/l0lwXmzHRRiJc3bfxQdafmcVcjDk8WxL8YIBcl6q1ak4dNmUuQy8l3hRzovXBJgULz3ml407ifPSWWCPoteLynl5Rua4HJMNKf0MOS9KtIxmVElrukC786JcF8swkNc0aM4UmtFzMSsXcr2ugXo+ByVqnU12U7NA4mXikAp8ROpFlf+L8JIGVpwG0Eu8jK7DbrU6jQsvvBhAD+dFTZTmBsDC1BCxVriRT4kXlfMymvBO0uNlN+xv/Q5y3/+a/ByTl/OSLWG3/83hmWeelv8+lfm9TELCrr0mu0MzhtkLz2v5Xa5ciJ0XW7qGg1QbVSrV+MZoSweFs17iRea8uI0kyX8HD2fs77xsvNeLcl5Orwix8uSpkwC2MWykxAvrLF7U+TFVqYIxhkvWTqOmOajKBoaNHjlOmq7HYSNvjMZ+kHiZOKTzMqJvNpSvp8sXZKVpAEDUJWk3mW208YTdarU154V3aGEdpMJGjHVvtNQMEzFnyH1TqYwmbHT2rCiT3rVrF6LVBZQtFUeeXPHSO2ykRgT0v9CpFeHyACWm7UyC8+La4vjNmXkYbaMtcuU8TNkpVQnvXuJFJexWq9W44q2hzsOe4kUK0dCHFvdFGZ/pwVnpNpRRMUrnxZYuz8rasnx8GbxHwzegtc+L+Hd04kVTzovZ7ryI9zc9JULau00Db3/pj6IYh436/W0pXtzxCSeSeJk0pJI29NF8taFU3Or1tKLssttY6bi9KSeVBoHfUWz0w/e9eEU0NTWFAwcuAGMM9Xq94w3OT4WN0EO8NAIhXnSNxSe4umhstEmdcl5mZ2cB30YlJ/ZBrVYbah+MK5zzvoMZgWzOSyJesjknnPMW8bJTc15ceTPIdeiGnasUYehSvEjnpVvYyPf9uBJMiBdxnjqqz0sP8SJayYvzw8yp+UY7V7yosFGnaiPx+MZyXiLHRmTbCDhDUzaqs+0GoOtAGCLoI6Tbq42US9ZsNuOxBtmRi1YpYmC1CmF1nk1PTQMAQscGXHeghF3x8jKR2x8fR47Ey8QhDl7dGM1Xy6XjEjsvfcJGhpH0khimUZ2yvhljKJcryOVy2LdvPwDg6NH1oSM1GoBzA0zrJV5kubdlIOLJdGlg4zkvaqL03JRY7ZZzYjUThuGG25CPE+lhi6OqNlKiJat4aTQaLUnhO9F5CQM/TtYtdJhFU5gqw5SzujzlvHQRL+nQpwgbyZyXQZwXpsXl0vmifB9sfNrAZ6W/87KxaiMVMjpTqSCS+Uq264DJXi/9Qkf1ulgAtOe8ABtZSMnrvly8rs95kQ3qZGl3ZDuIHDt2Xvr/XSlegvERtSReJg25YrLM0bTwYTIBV9cHEy8qYRcA/CFUugoZVSpV6PLESvJenlm3feK8GLHr1ImGL+czmQY4V9VGo0iUSzWoqwonJ29oMGTO0STlvSjXBejXYTeL87Ist7UzrYTbnZadKF5cu54MZbTWOy9mqRjnvHiq2qhL2EhVGpXLZei6HoeN4s6pfZxYVXE0XRWilGE453QcUMfdZjkvvjxmT6bcDdvxsDy9B0D/iqP2JnWWZcUjNYbPv5OOO4tE0rXeOhYhblA3J3t1uQ7CZgOFAaqNxBPE8ROSeCE2CyYPYsvq3E0zK7q0kXV5M1bipVvOS6t4yb56S+e7KHqVSyel0ibQw3mpe+J3edOAiuernJeNJ+zK0QAVcbFkjKEsQ0eT1KguPVFa65FUNUzYqP3/+6ESEFVfodXVlR13s3XtOhzpvJQ6uATMMJDTxPkXLwS6OC/pZN30v0q8sH7iRea97JqV4oVxrO3QUn/ldvbLeRlWvKh8l7M8ub41XQ+nDXGN7Fdx1B42Sv9/r8TZ3qScFyu/riGhOl9md+2OHwuWl1M5L/0WcOL1oh7Tp7caEi8Th7gx53KdS4azkpPhAV1niKIIWh/nRdeTTp7DiBfVoE7F7AHgwgsvBNBZvMSl0twA07vfvBqueC85ywBHu/MymlLp2VKyzysydDQJzktk12DfcTPWnnoQQO+QEZBOiOx9c4iiqMVBySZexPNUN9kgCDZw4d8elleX4kTcqXLnztCWXEH7QYiI8x5ho1bxkjgv4rjXjO5N6oBEvOybKsZddnfqcMbEeeldbbTRsNFKygWLIo6zsgdPvy677dVG4v83lrTLWOK89JooPbtrN5gU/Gnx0mjU+wgTJV7GZ4FA4mWCCAI/LpUrlEYjXsoyO93QNKytrICphN3majJTKAVjbEO9XtI9XhSDOS8GoPUIGzlSvJgmsE68jChht5C4TmVLXBQmoddLcPRBBM88gNXv3Qmgd8gIGNx5qdVWWy6YWUI/atu9e/fGf2+nlUufXjwdd9ednZ7quI2VyiFzgqhr2Cjp8aLEi0zYlc6LZvR2YtUNL898cDmc8czJkwN9jnEjyXnp3edleOdFHGdNr1Us1x3hVPWbb9QeNhL/v9GFlHTcWbiu0ghIzo3p6WloUrwFy8soyNYSnPOe+XmqvQQn8UJsBo3VZJVfSan6jaDEi65rePboMbCCvMjyENzpfNPfyIiAdHddRbrXS3toQLUx5zDRY8oB6o4sSbWMOC9oJNNcoyhO2J3NJ6dTUnG0M633NGqWVWNtBUB/52XQlW17km4W52V1VV2MZzAzI2Ze7bS8l5XaYpywO7N7ruM2lmbGoz5sP+zaYTfdoE78K0SMEi9Gnxy4lkZ1cjhjbSl7+fo4kFQb9evzsrGwkd02pNBxVsHRO2HX9z24skJJJeym/3+YIbGcR7HzYrGw91DGmVloBbFfgpVlWJoW5xb2XsRJ8TI+2oXEyyShbgacA+XqaMSLJq+chqbh7PxJMN0Ak/ODug1oTEYEDOO8qITdRLwcOHAAmqbBtptxTxVFukkdM7o35qvH/TSSfjBqteN5HrweU1X7vV/1OWcseWbrRirnZec7L9wRn6EpKxJKpd7OiwpZ9nNe2sXKMGGjqanpWLzstHLptdXFZK7Rvr0dtzE0DZZcDNh+BHSZOZScN6r8VooXKY4M0+r4PEUy36gZt4J3d2ifosH7vAzvvDiGBcdtfb7j1mCbBQRLi11nv6XFSXqI7EYWUmGq6i6nBUAP8TI9PQNNOVKcgzEW51sNIl7GSb2QeJkg1mLnRUexNBrxom7quq5hTVqPScVRZ5t+I1121UU47byYpoX9+0X30fbQUeCpqdIGtF7ipSEsUeG8BAiCsMVBGDZ0pFyXqakpGL74G/qeS1CJw0Y78waQRonUZpywOyrnZXjxooTK9PT0jnVe/NVaEjY677yO22iMwZT5Kg2vu/OiRLJyXBLxIs5fs18Cf6rLLmTOi7+BwYXbyeDVRsM1qQuWl7FaqqDptC54bNfHSmUKPAgQdnFclTgpFoux4wFsLGykGnUCQI4F68JGQRDEztzs7Cy0NkeqPEC/K5XHOE5J8SReJoiGvFFyaNBznU/crCgBomsaHHlipfNeOrGRnJdOYSOge96Lnwob6Vb3uJG6KORMAww+vIYDXdfjTrHDho7SowGUQ6HvvTR2XiYibGSL46opb7T9w0Yq58Xtud36sNHKwO+pk3hRoaSdAnfsOGw0JacSt6PreiJe/BAIvI6r+q5hIylerFznEIqCWfI7dRtQc2wif3y6qWahf7WR6vOSvQdT5HsI62uo5/NoyiKAeAim42FVOijdKo46VRqlfx5mEWXLhRnnDAU9AGtrUFerrcaio1qdgta2X9T53PsaqKQCiRdiE3BUHw6ug3UZUJiVtPMSyJVYUi690vE5GxkRkMw1Gky8JDkvBjSje1xfrSpylhAvzmprxv+w4iU9GiAWL/ueF+e8rK3szLyBNGqCuC175fTqrgsMbssr8aLKrjcaNtppzgvzvdh5UU5JO5ZpwDJlToInRUsH90X1eUmqjcT54wUhoihCvk+SdTzfqGU4485sVNe/2mh45yWQAtvOm7Hzct555wMAmq4POycrLbtUHHWqNBI/Dx82UtdMDh0FFiRTwiXqPJuamoJhGHHCbvK31TWwe7Ve7LxkfnebB4mXCcKNm4hpXacrZyUWLxqLV2L9yqVH4bysFy8XA1jfqC7JeTFh5rrH9VucFwasnhEro41WHCXOyy5wWzpT5TlUpHO0unSm63N3CrwpnRfZKK1fzsug1UZKrKhy542GjXZazgvjIVx5jqSTN9NYlgFTLgYaMrWhU7l0e7VR+vUcL0C+T6gvnfMSDzfl49OQLAuDVhsNk/OiknU9i8OWzstznnMJAOG8hLpwG7s1qutUaSR+Hn5IbH1Vubs6LBbJcQ8J6WRdANCKbeJFHjO9hFMy8HZ85AuJlwnCk1ns4BqY0TtBb1BUMqqha/FKLM556TLfaCPVRp1yXoDOzksYBojkLBAOA1ahu2BTiXKWFFZLsrHcRiuO4u66c3MiXwAAy1dQnRXNoNZ2WPluOzwKwWVVhRIv/ZyXJOdlMPHy3Ode0vLzICiXJV1ttNPEC6Kgr/OSL5hxuXQjEpdr3iFpt71JnWEYyMuuvY7no1Ds3EdGoVbr3G2AxWXVO1O8bGa1kRIvke7Hzstzn/tc8XquB401EELrWnGkxEm7WFU/D7OIqqu8Oq7D0LpPlJ6eFudJi/PCGMrymOk5IkCjsBGxiQRKvEAbfdhI0+KVmMp5iezNqzbqFjY6fvxY3BskSdaV4qXYPc8ndl7kfqnVxAmdlChuMOdlqgp1YrN8GdVdYh5TbW1n57yoUBgANL1BxYtyXnrfHNRFVa1cs4R9dnrCLucRgtCLm351c16KxRxMuRhYU0UlHcNGrc4LABRknovjBSi3nU/tpEuldfk8xsZnCN+ghGEYX7M2x3lZRggGzutxg8HEefGhMR+1QrVHzsvow0aOTKTn8nbeLWyknBc9lfOiFQoDuc+aTs4LsYmEsdPBgA5TaochnfOiVmKbFTaKoqhjkzpAxJV1XYfjODhz5rR4/ThkpEOLIhTy3W+q6sS0ZDJbs96e8zJc2EjlvMxW5AUjVwLTDEztOSBfd2d1fW1HJesCos8IMHifF8/zek7JVRdV5bzUaqsDHTO+78UTlKenZ8Q0b+ws58VzmnHISGOs6422VC7E1Xtrcrho77BRct4U5bwcx/VRqs70fD9KvMB3kI9LeHeeeEkLkm7OS9KkbpiclyWsFcpw5EJR07R4YdV0xf5aLZcRdM156R02Gka8uMpBkvOHuk6Ujp2X5PdavpASTr3Ei8wnZCReiE0gUk4HZyMLG7U4L/JixkrT4s80O8+TGVa81OtJi+p258U0TZx3nhAEKu+lJVmXB8hbnW+qnPMkYVfuF09e5DYaNlpakhOly3K1KnvgTO2/GABQG7Icc1xIi5dmnLA7WM4LALhu98+vLqoXX/ycOCFwEAGiHBZN01CpVHak81JfW4mHMhZzuXWzaBS5Uh6mPGYbgdj/7V12oyiKj9+085KXosfxAlTKncNSMalQQ3lKHMNsB06WToeC+okXx7Ezz+rxl5exViyi6Yrr4tTUNObksEMVRmrkLfhLix2vjZtRbeTH4VnpvLSVSifOixQvbc6LWoz0ChvpcZ+gCDwaj3AiiZcJgocp52VEYSPfT5wXZSPHXXbDQJZWtjKseFHWdz6f7zibSa1wjh4VeS9+u3jp4rzYdjN2AHIyD8CXIadkINqwzoscDVAUn1mT4qV6noyDeyH8IUoyx4VW8TJY2Cj93XXLe+Gcx+JldnYO07JUeJAW/2qbqakpaJq2I3Ne5hcX4gZy5UL3MuZcMSVeZNiufThjvb4W3yjTuTMFeR7aXoh8HyeWaXqc6LlrVpzfjIVwBxiuOU6ke7x0E4St4rp3OX87wfIymvlcLFRmZmZiR8PzfYRhBC8HcM9D2CEU3S1spIbEDrOIUtcyyBb+7TkvScKuEi/J4kPL5wdyfYx46jkHwvEQtSReJggeKpuXgendy4azoJwXQ0uJF8MCpM3cabr0sAm76ubT7roo0nkvQJLzwrkJhhClfOekRJWsq2saLE28tygQn0s5L8O05Q6CILkB58XrsoIUL7uTpmOrp57O/NrjAld5TVZx4LCRpmmxgOmWFGnbzfjYSt8ABknaTZdJq+cDQLPZHLpTclaefepJ/MUffgpnTj471PMXTs/HybpTpe65WvlKEaZciKj93+68qJBRPp+HZSWOa16eh44bdr2Rp4mHM+6eihupLgz5+baLfj1egNbmdVmTdoPlJbiWFlcazczMiN4pMqG16XoIje4VR0nCbuewUaPRyOwGhXE/HpXz0jlhN642SoeNCoOFjcy8yoOKwIPx6P9D4mWCUM2r+l+mBiduUiedF7US0+JGdSvrnjOs89KtQZ3iootaK47SzgvjIYrFzkmPakVRKhbiiwyPxI1jI2Gj5eUlcM6h6zqqptz30nkxTRMFSzaqO7lzxUsky6T12QNxwm6/wYxA/3JpZWXncjkUCsX4wrq01N95USJXiZdqtRp/r1s1nPG7f/nnaC4/iW9948tDPb92OhnKWKl2rwTKVQowpPNiy+3bc17aJ0rHz5XzjNTf6Yfq9VLN6eAQf3P++PGBnjsu9OvxArSK6yxJu6Jzbg2BGcTOy/T0DDRNi49F2/WhsTocPd+x4qhf2EiEuLPlyUVB2/fbNWwkxUs6bJQfLGHXis/5iJwXYhOQU57VBNBREIeNNPGaCydPib9RFKvdXuIlaD+p+pBUGnWOz7eXS7eIFxag0kW8qJBQOZWrwXn7ZOns4iWeJj07Cy1VJq2oyuqn2vz6adg7BdWgTps9kAobDSJexGfvlvOytJT0nmCMxe5JFudFhZrEzUP21dmi0BGzxTEV1YYTS+7qStxdt1ruJV5KMA0pBGVCaDfnpd2xzMvzcGDxInPGtMCJhzMun+k+ZHAc6dfjRTHMZGl/ZQXgHFxz4+6609UqglOPxsdv3QnAGMdKaapjxVG3sFEul4uvm1mvRZF03FmXsFF6ojTQWiqtFfID5bwU4lBxCAxRRboZkHiZJKR4GaX1omx4TRMn1vwJEbKJy6U7ho2Gc16Si/B0x9+rRnUnTohy6SDVoI4hhJHrfMGKVzulYnLArxMv2cNGSY+XZDSAChsBQEXelFZP7yzrPY3KedFmD8Qddktm/8tGP+clsbJn5L+zLY/3ot15AZDKmVnp+/xRoEGcFywazkIPbTtxXnqIF7NcjHNeVIJve85LN9GfNzKKl1SXXS7Fi73Dxlv06/GiGHT+VppgeQmubkLT3Nh5qdgLsP/iv2EqJ86JNU/+Wyp07LLbrdpIPDbcQorHFX1MdEfWkzlW6dyyTs6Lni8M1GOmWBbihTEO3x2PHD4SLxOF7DMywldMqo3ECbFyRpQGaz2GMyaDGbNd2JObUuew0b59+2EYJlzXxcLCfJyoxmGAIQAMs+PzVD5LuVSGHpf6KfEyfNhIlUnPzc0l4iXlvFTkzaS2OJ/5tccFJV6CXAWB7ElSYP2rDfrdHNrLN7PkvKR7vCimprY4aZdJ8TJsRY7vxc5LtwZ1AKAZBixdVsjJxUD7cMakQV3raj6vSqW9Ad+jlXTZVcMZPXt0pf6+5yLa5EqVfhOlFUM5L0tLWJUDb5uO+BxTmrgGTcnjYc0X50gzZ3TMeUk7L/U1B1/746/j8R8cjR9LbzMoqvqHQbgu6fymZrPRklsGdKo26p/zUk4dW3ZzuMrMUZNZvERRhJtvvhnXXnstrrzySrz73e/G8R5x0eXlZXzgAx/AoUOHcPXVV+PDH/5wywETRRH+1//6X3jNa16DK6+8Ej/zMz+DP/3TPx3u05zzyLDRAMl5g5KEjYQwaMoLZa/hjMPmvHRrUKcwDAMHDqhy6aNJwi4MMIRgrPPhHK92KhXomnKnxAmfrDqyn5BxmXR6NEAhuRFVZ0QJZW3p7FhNY81CPJQxShLAcwPcsPvnvLSuBtW/7cMaO9GesAtsrfPi2DY0pobyDZm8mO6u20WsK2Lx4mULG+XkeegOmMTM0pOl5XDG0M9WjdMNu7aKL336V/Gl/3nTSF6v69+xlfPSW7wMOsIiTbC8jLp8XkOG8KY0+S8Tr+PIy0tgBV3CRonz8tdf+v/DPnMH7r3jVvmYEI9ZG2Zynsp1XBcyWgEgPq8KpWmmCSaTubV8Il7SVZntTE0nfYJq9fFw4zKLl89+9rO49dZb8dGPfhS33XYboijCu971rq5Z/jfccAOOHj2KL3zhC/jUpz6FO++8EzfddFP8+8997nP43Oc+h/e+97342te+hre//e246aab8JWvfGXYzzSW+I/9PYITP9jkvyJvkCMNG6mp0uJCqFZivXJehq826p3zArTmvaTnGqGHG5CIlyrMeDtxwiczRYYPG4mhjFLUpZ2XmV3itW2na0O/cYZzHosXB+L7zxkadK//ajxpwd47YTcRL9mdl07iZSucl6OpERWMBVgbQvgyBHGH1m5hUkVOV05m57BRt4TdxHkZVLzIvAYvES98RMmZD9z1t2AsQMhXh+q8PSjK6evvvIgbuapOGgR/eQl2Xi7i1DRwXSbuymrDSA211G24Z1t7vfi+H4slyyqgefYx8YtwY0NiWUq8tE+UVrll0ynxASR5L1oh3xLC6pb3Upoqg8smeLXaDnRePM/D5z//edxwww247rrrcPDgQXzyk5/E/Pw87rjjjnXbP/jggzh8+DA+9rGP4YorrsA111yDj3zkI/jqV7+KhYUFAMAXv/hF/Nt/+2/x2te+FhdeeCHe9KY34V/+y385Ue5LePpJOH/zB7C//dmhV+B8oPI5sY2ujS4aGFcbycFcoXQ7euW8DJ+wq7rrTnfdJhEvz7Q0qWM9xYs4ISvVaVhyO1X2rU5cx7Ezi62WoYyO+BstCbvyprTmBIhWTmV67bHAawLSkm7K3Vs09aR8ugeFQu8RARvJeWlP2AWSY2YrxMvxJ55s+fmptp8HI4zDOZWp3t1v1WwjL1Cl0p3FS7voz2UMGyXOSxNQg/hGFOY5c0om+jPg2SeH2V+DocRy/5yXIZyXpWX4prh+244MF1niWjtdEOLFlC0qdGajxixEqcqhdFjm+/c8AEOTCx7mgnM+fNgoznXk6xrUtZ9nCk2en1qhAMuy4hL7bn/bLOTBpaB1huxGPmoyNQM5cuQIGo0GrrnmmvixarWKyy+/HPfeey9e97rXtWx/3333Yffu3bjkkkvix66++mowxnD//ffjp3/6p/Gxj30Mz3nOc1qep2lafCMbFsMYfTqPaJGf/Dso3tOHxf+4DeihA61HG/tONL7zB/CPPozqm38TWqHSY0txYmna6D5/0qROXEB56MMwNLDqLGyInBddZy2hqpyc7hzKLPhB99eanAM0MzPd9f1ffLE4Vo4fP4bgEun+wASY3fU5ajVRnZ5BIVoVGo8FMAwNU1PJBd/zbBR6DHdsR4mX3TNTQENcQMzKFJj8vOq1614A1OZhXPTCnq837PG1WYTKjbIKsUtQMDUwt973+FIrX89zO26rLqpzc7MwDA27dokQ2/Lyct/XrtVWAIgqL7WvZmdn5O9WN+XcT7N2ulWILhx7GsbLrhr4+ZxzMObHYaPpubme79mSDeZ8tRgInJbtlXiZnm49b/JWEjbSNHFd7XWMhQW5AvebYjhjBADBSPan10hW68efeByXXHH5hl+zE64rxHKpVOr5vlXFnOs6fT+f2lfB8jJCIwDjPBY903kD0HRMSUcmdGwEmIYBF6ulKqKVJeSmxXXAlq51oVDAyUcfhMp715iHxtoaqlWZT9Psf361Iq49GgAtV2x5rprhNjvbeowZU9Pwz5yBJa+15XIFS0uLsO1G579tWBBunA+32f1am95fm30dyyRe5udF4uH+/ftbHt+zZ0/8uzQLCwvrtrUsC9PT0zh16hQ0TWsRQgBw8uRJfOMb38Cb3/zmLG+tBU1jmJnJJhCyUK32tiTTcB6h9vT98c9lrYnczJ5Mf2/1mQfBnToKjWdRPO+lvf4aAMAw9JF9fiVADN0EEAI8wMxMCVHpPNQAIPAwXdKgpSp9qlX1t1Wr/8H2l7JLzz9/b9f3f8UVlwEQFUeIxAWQc+G8dHuOJ3ME9uzdhdKZp4FA2PbT00UwxpDP5+E4DjSt+2ukCQMfumFieVnEtC/YNw2cEMlys7um4+327ZOTpd0Qpn1m4O8ky/E1Kr7yla+g2Wzi53/+5+PH7DUXNQBmeRqaJsukTR1m2Oz7WabiFvOd96kSIBdccB5mZkq4+OLzAQhRo76Xbqjw4kUXnRfvq/379wIAGo21TT33AcBvtDpPzcWFTH/TsZtgLIqrh8674Pyezy/IzrdRxOGHEfKB07K9uinu27er5fFcLklgN4wo/k6AzsdYc24WTYhSaSs/h6AJAIOdE/0IAwey2wJWz57exO9IHKfT05Wef6Mqe+v0um60460sge2O4PpB3EhuumDC2n0h5lZ0AE9g9ewCPP0SGKGLesFCzqnHr//ss/IcKhRhsNb75cKJZ7Brl3Afg8DNtH+YWrSCI1du/dyOI46NvXt3tzxu/vL/F2s/PIJ9V/8ImKahWhXipec1kGsAA4LQG+j9bfZ1LJN4UclQ6S6OgKhRVxeU9u3bt1Xbd2rLfPbsWbz73e/G3NwcfumXfinLW2shijhqtdGXc+m6hmq1gFrNRhgO1gUxmH8cYS3JOl89+SxMa/fAf5P7DiIZkqjNn4S762D3jZmMfWoalpdHUyWgbFgx28IGeJC8tlUAPBtLz56EPpOI1EAOkGs0xPEy6P5S8VlNy3V9/3Nz+wCIEQGNODZsAFrU9Tmq8Zmu51DJaYATgbEIx54+ierMNMrlChzHwbPPLqBSmev5Hv/mT27Ds0f/DpW5F+H0aTEgUp2iLF9peQ+G7IxadwM054/1/U6GOb5Ggeu6eO9734cg8PGc5zwfBw++AADgLYjPF1llnD4tvpuipcNZWez7WZiM/S8v1zpue0b2D7GsIpaXG9B1sReDIMDRo6e6VpxFUZSabZRDrWajWi0gJ/M1zpzp/942Sug5LfF2p9b5M3bjqSceF8+T4RxmlHo+P2clbqDtRzBdu2V7dXwbRr7lcV3XYegagjDCsWOncOCA0fMYC3zxnQXNNRj5CxA0RXh1JPsz8lUaDRqrq5v2HS0vq1CM0fNv6DKPaGmp/3vRdQ2VkoWlpgfGGJqOdCFzOeQMDTw/hblLdgH4WywuLsLPl4HGMtwcw/LRE9AOCsf12WfF+YQwEmHuII9IY9A0G0/88FGYUqSeObOUcf9I54VxeDBbnnvypBBJpVK19TVn9iD/ij1YWRXXaDXy49SpM13/tppa7TScPvt2+OtYtVoY2LHJJF5UnNDzvHXzITolSOXz+Y6JvK7rrmt09dRTT+E973kPwjDEH/3RH/VM2hyEINi8i38YRgO/vvP44Zaf/dpZsAzvLVxJhE9QOwu953PF7wzTGNnnV9+flS8AsAGE8WtrhSlEng1/bQm8sjd+jibj5eq5g+4vFSosl6tdt9+1ay8sy4LneTh9dgk5yLCR1v1vqGTcYrGEKccCX3HAGHDy+LMoVqool8s4e/YMVlZqPd9nGAQ4/vT90DRgcf7xuCrAMGReQL7c8vxSSRzDa26AcPnUwN9JluNrFJw4cSJOovza176K5z1PuFtBYwUAwPLVOEmvYGoIm6t93188vbvZ7LitStitVqcRBBE0zUCpVEKj0cDZs4solTqHR2u11XjVWy5X44ujqrRZWVne9H3HQ3Ej5lwHYyGiwMv0N4898SQ458lgxmKl5/NN04xFSNMLUQ08+J4HJkddqIVjqdT6Opxx5C0TddvF8vIK9u07P/5dp2MsNMQ1PHIaKFSn4CwBYP6I9meSdxN69qZ9R2raeC6X7/k3VIfdZnOw9+KtrKKWLwBwsObKPJeyvOcVZzB70YsAfBorDRtmdRpoHEdoeHDPnIlff3VVXN8sU1hQAStAi0JAs7G2eDYWELVa7+vQepTzEgFG6+deXBSLDnWedUNVHK2u9vjbMmHX9wY73jf7OpYpKKVCQGrFqTh9+jT27t27bvt9+/at29bzPKysrGDPniR0cv/99+PNb34zCoUCbrvtNlxwwQVZ3tbYwnmE4Kl7AQCsLFb0vN6/mqLlNRrJ9lF9feldC9J5sczO/U6GQeW85IqqSVGSwMfiXi8rLc8ZplTacZw4jtytVBoQcfsDBy4EAJw+K/YNhwGmdU+EjscDlMooFMqArJxZOiOOzUET5e768pehaWKlYrvi5ssMDZ84+9f40u4KTrYN2IvLsL0QvLG0rsR1XDh1Ksnh+Mu//HpcLqkqjVihimZTrLREwm7/fLQkIXK9w+r7XvydpKsgBun1okRPsViM+wmJ504D2KI+LzLZm4fSOufZEr1XFhbgByEimbyvch26YZoWTEMsCBqyyzFSSbudEnajIABnQF6OqBik/FZ12IVnY0rmIDH4o+nNki4pjzav2mjwPi/Zqo28xUU080LwrMmPMl2U835KM5i9WISww4hjr/zTTKujfjo5ltUxnzcZONew/xU/E3cy9pvN4bt9y+u+zqIOpdKdE3bbUeKl94BaKbqCHThV+uDBgyiXy7jnnnvix2q1Gh555BEcOnRo3faHDh3C/Px8PAUYAA4fFk7ES18qcjcefvhhvOtd78Kll16K//t//29HEbRTiU4/JcSHmYf5/FeKxxrZxEtasPA+z2Uy3mtlSDrthxIgZWXjp3p8sNK0fF+tFSLDVBupHi+apvUd/Kcqjs5Iu5xzA0zvJV6SUkRmFeLuoWvyRjdoo7pTRx+Sfw+o2+KmXJgqwkeEe6cK+J/mWXzivs/gn858H0ByM1lzxfcSrS70fP3tYn4+ES9nzpzGvfeK87tVvIiL/ODipXuTOiVAdF1vEaqD9HrpVCYNpMXLaubBdllh0kXQQrlI4Nmq6uzV5TgBWtdY354kVi4HS4qXeihnc/mqTQBPNalL9qXvOIAWxUm7tQE65bJCBdAMABz7d4ubHWMbHxHguy4YS4lY5m9a36P0VOleZK028hYX4cjKorpsRDclK4y00gzy+TyKslDh/LVT4JxBYwFOpcIrSkDmLAM8KOBVP/oyub8BHjgtwxmzIR13Fq2rNlKh+NnZ2Z6vMNCAWum8RDtRvFiWhbe97W34xCc+gW9/+9s4cuQI3v/+92Pfvn149atfjTAMcebMmfiAeMlLXoKrrroK73//+/Hwww/j7rvvxoc+9CG84Q1vwN69exEEAf7Tf/pPmJubw2/91m/BdV2cOXMGZ86ciXf6TsaXrotx0ZXQpoQoy+y8pMRLX+dFipd+ZYID/23O49DP7G7x/hmSC49W3tXxfQ0zHiDd4lzrU+qtxMviqjjROMyBnJdKpSzmfsjuoY0V8b4HWfE8dOffQNNq4FxD6O1Hwxb75ZLznotftp6Hl6w50MHwdO0Y/uB7f4THlp+M+27UvRCcc0Sr41kufbJtcvDXv/5VAK3iRV1QC6YO+E7fybJJqfT6m4MSJ1NT0y3f9SC9XjqVSad/jqIodiI2A1EpJG7EluponbHLbuA0kgZ1eatvU8lCMQdT9k5qyGNXddltNhuxU5Z2cJymCw6OfE4OBx2gepMxLXaId5c0cPm3jh99ZtCP1pFjh7+Dlo/IPARP3tN1+40wyFTp9O8H7bDrnl1EIIevOrLT9LQcCcBKrV2im8ceQcDE66+kFnBP/lCUiOdMA+zCy2DomqziATj3h+/2zZR4CXvMNertvKi/3ct5YdJ5icLxaLiZuZbphhtuwPXXX48PfvCDeMtb3gJd13HLLbfANE2cOnUKr3zlK3H77bcDEJ1eP/OZz+DAgQN4xzvegfe973141ateFTepe/jhh3H06FEcP34cP/VTP4VXvvKV8X/XX3/9SD/oVpMOGRnPvTq+KGR3XpLteWO5a78X13XBZOv7UqW3DT0oQRDEQmXvBSJUwxhHTd5cWEWKl7XWlVkyHiCLeOncJbQTSrwsy6RsDgOa0fkGwDlPxgNI54WFwjJunHkMQeDH4Z1eq45H7r0TABAFs7jk5f88dl7K+SKe43G8ZaGG/zrzCrxkt0jO+87xu2LnJQgjuEE0tr1elPPyylf+OADgO9/5a9h2E1GnsJGsYOnnvvRyXhIru3U1OEivl27Oi2lacR7dZnbZXVleBpM3i+myEAaMZWt7z8NkNEB5AJe0VCnClBOi1wJ5nMuwUZJ3Zba4DV7TBljaeRlM0GkyYd1wVkXzRwBLCxtzDE/94D4AiMUQYx4a/3AbeDCa7r1pBpkqLX6fbbaRt7gIrov3G8nbZtWQ4lWKl5ldIhVitWHDl8M0bYshlK7lsScfBQDkzBxe/bo3AQCMoqp6Sl+HMjapk86LpYXrwkbtzSC7MVjIShx7cV+ZbSZTwi4grN4bb7wRN95447rfHThwAI8++mjLY3Nzc7j55ps7vtZVV121bvtJIR0yMg68MG6jzxtLcvU2WBvctPOCKAS3V+OTJU0jdXFK28cbIT2b6KLnXowj39XAWITjzxzD1OwcNCle+Fqr8zJMzotKOuzVoC5+LxcJ8bJUa8qujxq0Ln0HXNeNk1HL5QpgFnDAOoPjPAeNNXHHn/yfvquO40eOgDH5Gfeej5f9+FX44z+UvR/qjXiuUbU0h9effw0eOvN9fP/sD1G/xIau6wjDEGtuiPKYi5fXvOZf4JlnnsKJE8fx3e9+G69S4qU4lYSNpEDgdg2Q338netnyyVyj6ZbHB8l56ea8AOLYaTabm5r38sTjoitqxE0897LnYuF7p8EYx8L8Avafd96Ar+Ilzkuhv0taqRTjRnWrciGvnJdE9Fdbrim+7YKzCLkMYSNAuKkhgKh+FuAmAAf11eEmZwNAeOYZ1Oo2oAFRVIKmrYExjjOrLgr/dDtyL/u5oV+7E5s1VXp14Qw0TVZeymqYaZlypZWEMFDH74rtw8jpQAD4VoRgcRErNQ+uJ6vCilXMVsTfr8ztwmpDCOBBZgx1RqYLoNV58X0/1QNosJyX3n9bhiyjHeq8EIPhPylye4yLfgTMsKTgYEAYxDe7QYjawky8S+hoRc7Z4RyozoxGvKQrxSrliqjqAbAob3aacpPqrbN7hhkPoG44g1SZqenSq3UbYSQbIpl6x23VSoIxhmKxCGYV8OLyKnRHNjWbfxiGDF10W3X8w+1fBmNAGMzgdW96CwxdQ9MRn9cywmSuUb6CfaU9eMHs88HBcdfJf0wl7Y5vl12VsHveeefjta/9WQDAN77xF8lE6XzKeZEXucGdl+7ipZvzMkzOC7A1841OH3tG/A+3sP/FLweHcE6eeuyxgV+DaV5cJl0u9hcvxUoJphQvavCfynnpNlHasx0wFmZK2AUAJp0XvnY27qjq1Ycva/Ye+gYcmWMGboJzsb/OBEV4D90+QCg8G5s1Vfr4GfE+I25BkxVuUwUDMPOxYFBhzxUnwG5dnB9cd+CeOYO//fI34EvB+ZwX/2j8uvsuFg1cNeZBk+0FGo36wDlBURTGjrulhy05L+pc0TSt73U1WcD1+q6l8wISLxML5xGCp4VVajxXJDIz3YiH9vVLvE1eh4M3xEmjKnu6hZ3WVtXNREehvLEyc4USH5qmCUEibeQ4bCTFC3wHcJODfhjnpd9QxjS7d++BaZqIOEetIccX5HqLl1KpDE3TRNgIwAufZyAMp8BYiGNHvteybZqVhQVwLmzzqDiLqlzRRbIUuFzkWFoRroQaDfATF4jk7H84eW88jXXNDRCtzg845iE7vu/hf/yPj+Hw4bszPS+KIszPnwQgpnb/zM+8HgBw991/j6VV5bwkOS8laS9HfUYEJM5L94Td9eKlv/PSS7xsxYgAe1FWT0YmzF0XAJEU9M8eG+j5juNAY27ivJR6OwQAkKsUUuJFjQhodV7a5xr5jgsgjMNGqnt1P1ry2GSYJ3AHb6GfJlpdQPD0fQi4vOkxA5wLu2JFmwZCH+7dfzLUa3djs6ZKL8sct4AVkqnmeROavC4DyfG76mt4DhfHuMbqOHX0NOr1E3BkkvZlz7kofs5zLjsYzww6K3PPoiiKFwv9CFLueIEFLbON1Hk0NTUFXe98fVQMlG+jNMuYDJkl8bIJRAtPtoSMFKwsLtbtbko3uLMGhAEABn3v88RjXVYqTXnQcWhgxmiqjeIeL5ZKKpMrMRleYYYFVpAzjupJ3ktSbZQ956Vbc7I0mqZhnyy1X16TyZNW58+s8liUA6JiwhdPmdC1GXAO5PVmy7Zpvv1nt4KxCGFYxXVvekv8eKQLEVIqWDiyKPYPk6MbXjD7fOwp7oITOmB5sc/WfAjXrb6xyo1u/N3f3YU/+qM/xP/4H7+d6XlLS4vwfR+MMezZsxcXXXQxXvSilyAMQ3z78UVAM7BwuoGnHxcX1qIMSSq3qRu9w0adyzdVRUQv56VX2GgrnBffkUKVa2CGFZe6OiuDndNPPvJDsb2sQKv0qawDgFy5FOeRNdRwRq+38+I3GmAsypSwC6Tz2BYRhwnC4SZnew//FcA5QhnNYnqyv+o8DzCG4KnDCE6NLnVgM6qNeBTBkXlOvp5PkmALRnxNBxIxXtOr2BsFsg8Qx4NHTsDUTsV9fVR+CQBUy8XYjTp78mTsWg8aOmqktsszH7ASMdwtt6wTg5RKM+lQb1alWFZIvGwCSZWRCBkpVGx00IojtR0rTkGript1N+HjyJwEcB3MHI14UTkvqnooXomlVtOdknY3kvMyiPMCAHv3iC7FK3Vx8TG72O/qZFQrizgm7Nl49S+8E9zfg5ypLhitN2TPdeA5JwAAkTaFA7uT3kRLcq5RqZCD+kZYXtxANKbhugPCfXENsQ8bmrhJbVbo6KisCDl69JlMFxcVMlJuFoDYffnmY2fBClXc9fU74cljoVyV86Q25LyosFGreBkk5yVxXtbH8LfCeeFSkDN56WRy1Ry6A4YfnhDhJccT31F6om838lNFmPI60nSk8+K3Juy2Oy9OXYiVvJkxYTfuR7UIqAnJYbZScEAMbPUfu0v+JMvC8wVwJmcABT7MgyJB3P2HW0fiSHLOh5gq3f97C+t1eJb4vsJ8OcnZyptgxfV9ilYjEzoDIohzIDSPiXEQ0jWrVJLvnDEWu3f15aVUyfJgYb7VpRUAAOcaSloAZibXQdV5uV+PF/Ge+hctJGGj8YDEy4hJh4zM517d8jul0gcNG6l4MCvPJs/tIl7cpjoJNWBkzou4UCcjHuSFOlUmG1/sNihesoSNAGC3bKK1UheiLV/sbL+nw0YA4pgw92zMTM+geODCuC33/LMnWp57xxf/LzTmI4oKeIm8oQPiIqmGMpbzFri+hrWw2CIaX77vKuT1PEK56xqa7F66SeLlxAkRtrDtJpaWBs8jOHVKhIz270+STV/zmtdC13U8eqaBo3UOt3YWYSDHA8TipV/Oi6y2sEeb86JclZmZ6XW/i28em9moTvZ00eSgHi1OYhzsWK/LsJPjqfLm/iFes5SIF9sVf0cl7CYN6lrPm1i8xM7LYGEjVpoBmAZEAVgcasguXvwffAsIA2h7nhs3qCvMzIGpAa/ch/WyNwJWAdHi0ZTQGR7XdWPhPrjz0l+8+MtLiAz5vZenYldkqmBASxVPqON31fbBcmVoMoHa0FbE3wpUZ+i2atC4Ud1a5kZ1NXmsc+iwDAboSYPSQcukgQGdF6bkwnjIFxIvIyYdMtIPXNHyOy1r2EiKF600m0qO7Xxj8t1EvKTdno3QHjYCk4l3UXIx00bmvKwAGCxsBAC75RThlTXZf6T9giBJN6gDEudFXfxf95a3IsfEvm0219CUN4MoCrF2VlaWRNN40cHL4tdsNBqx3Vyw5sAY8EijtfImb+TxivMOwSjInAOZLxStrB9gOgqOH09yLo4dGyz/AkCc75IeoDozM4NXXCmO3W/+8BTAm/BkYypuif04aMKu2yFfonvOi/jZceyuK+LtTthlsruuKl1WImbQRnW+7EOiQgjtjkknNF2HJc9pxx0sbOTJ8FZOCvOBE3Y1Pa5mNGLxkq0pGfdseD/4tnjvV7w67ouz+8ILoOfFIoMhgFaoIvfSN4j3e/jPwL2NzaNLHzNZcl76OZXe2UVAdtYuS8dPYwyVnNFS+Rk7hyvLMC7+EeSN5JjgXEMYKret7VoVN6rzBu72rbDj7XToVutA026LhE6oxqCDhI1IvEwo3UJGAMBK0qUY1HlpJImxLA45dRYvcVId14ARh41UvJ3pqmogOSk7hY2GqTbKkvMCALtkRdVqvQ4WBSh2ES/qoh3b81bivACi9P+yV10HAHB9H1//358DANz1F1+DptmIuIkDP9o6+Xxx8Uz8mkVNfKdLHWzvHz/wili8nPZVl93NcV7S4kW5MIOgwkb79rWW+f701S8GANzx0JPgcOHJ9396Ra78B3RefN9f12m5Wyy+VCrFwrdbr5c4WbLDanIrwkaxi1ASx5ElE2IHbVTHQ3EjVwuDyoBOoylX1MoNbS+VrrT1dlKVLWqo49pabeBwolqQmJbqy5JNvPhH7gS8JtjUPpzyymAM4Jzh0stfiJz63uR+NK/4Z9Cm94M7a3Af+Fqmv9OOclEsy+qboKrCRmEY9s3Ne/aZY2AsBOcMB/aIZp3VggWNsTgVAABm1YJqZRnGc16GPVYiGG29FL+/deJF3Se4l7lRna3EBte6jgYYxHlRx4/rui0tMtKQeJlgRMhIiJf2kBGQdl4Gs/Vj56U8l4RnnLWO3U3D+IBj8cC2jaLEh3JemMp9Sa3E4l4vqc+0kfEAg4aNZqfEybZSr0HnIYr5zrkDSXdd5bzI8JKXrLiu+bEfEw/5IQL7GJ764fdx8vH7xefyZ/HjUtwozp4VQm1ubhcO7JXiTV9D7Wzr97qrMIcDuw4AAJ6syVLLTQgb+b7X0uI/LWT6oZ6XDhsBwI9ddgAlS8f88hqOnT6GMFLD5WRb+gGdF6DVfYmiKCVApluewxjrmfdi23bseG2H8xKGIRiE+JiVOVfFiky4HVC8MIjz1JXnVqVD7k4nLDkFWY0VSEqlO4eNAimO8rL6JAiCgef4qCrCoqlW8YOfxzwM4H3vDvGeX/zTOPrkE+JxnseumWnsPk/MrWPMRRgEYJqB3DUiEd7/3l9v6PxQzku/kJHYJskN6Zf3ckyGkyMUYMhrxrRclHRyXur1OsLdz8Nzcklhzr4feVW8XXuek1FIjqGsYSNX5TpCW9ddd9AGdUAyVVr87c6VTppavJJ4mTxEyGgZMAvrQkYAEveksTJQglo65wW5UtJKuoNzE6nVAx+s+d0gqBuFWl0aOXFysJR4STsvSgwM02E3a8LudFWIENdz4XprKBYGDBuphDYeAbKKQsV7AcAPA9xz+x/LUQAMpec9f91rpsXLC1+4G1FUBmMc//j129dt+6IDYmDb/OoibI2B2zVwJ2sTqt6cPHmyZZ7PiRPHB35ukvOyv+XxXGjjuueK4/Whx4/EjweO2GfcqYP36CqrpvYCrRUdtdpq3M6+UyJhr7wXJXoMw+g4/2qznZdjR4/FLsLFl4rjYpfcbxrz4HYYQrkO6Th4vti2Oqh4iUduqLBRe85La9golAsHQzdjJzRr0m7VEsdUlvEHwZN3gzeWwApTMC99BVbkYF7Vrfd5L7wcnDMwxnH8CSFsjAteDP2CFwM8hPfIdwb+W+0MmqwLyEndcp92ystKU2sIIRHxpNJoKieus2q+GyBCgMrxWV2rY/qiK7Fr2oexdw4vukRcB/L5fLy4U5TnxDWUMW+wGUMpfFfNuOokXlQzyP7HmGEYsRvVTThp8jhS4wi2GxIvI8R/SjWmu7Jj3gkrTotkOB72rdYAkuRcrTwHxlgq72W9eEkqAkYnXp59VtwEVUjBKqsVQirnJe71YgMyZp11tlEYhvEJM2jYSAdHKS/2cb1+GuVS59wB9brxasfMI86alzeAXC4XX1Acj0Nj4nEezOFn3vDGda+5tJSIFyNooOiLG/XymafWbfu8vaIJldd0cd8u6bytjjbvpV2sZAkbJT1eWp0XbtfwmsvERfXx4+L96poGJkv3Ad5ThDHGOjaqU6KkXK60TIVW9BoRoByVqanpjh2qNzth9+jjoqSXcwuzF10KAHjOZS+I+3Q8/dSTPZ/vecmAQlc6I9WZuYH+tiWT8D2/W8Ju6/EfyesBY0acV5O1XHrWdOVrhC0lud3gPIL3kBDw5oteDWZY8OSNX/WI2jM3F5cGH33i8fi55nNeJt738smB3mMnEudlsNluSdJub0fKD8U+j7RcS6URmB737gJECwcloJeXRejo5aGN1/Bl1I6LEvl1ISMA+5/zHPF85sbnzKDOS+DJ7wgMMDuHjQapNhLvrXfeix6LLnJeJoqWKqNL1oeMABEzVM3m+uW98DCIRwooG5elyxjbiVfeoxMvTz/9NADgOfLkqsgbS3olxoxcfAKrvJd0n5dB4uz1+lq83SDVFwDgew6mZYvtWuMMcvnO/TIS8SKdF8YA1cjJS+zislzJu43EhTFm98Ew1sfOlfOya9cucKeOiw0pRLVlNFZbRan6PKHt4x/KJiKMPnR0/LiY2n7eeefLnwcTL7bdjAXBvn2tzgu3a3jJeVVUS+X4u7FMHeA+WH6wLrtqOGPalu/Xe0JdaDsNZu2VrJt+3HGcgZuPZWHllBSJ3IJWFN9r5fyL45uxKoPuxtOPPQbGOKKIoSlzVyrTA4oXs9V54V5r2Kh9JEgk589wsGS6+YADK1Wjul18GVw6uSeeeabv88JjDwnxYeZhveA68T7U9Gum5kCxuFHd8kIi4rVpcfxt5NyIk+j7jAZQJEm7vZ2XSJPl8YVSLL6n8wZYaTpVgSNIi2/jgheCFargzRWc/avfAwCUcua62UDPff4VsQD25WcYWLwoMdvDeek3UVqhrpHdkrvN2E0l52WiCFtCRi/sut2gjep4cxkAB3Qj7tyqksM6PVeFDdgIxcszzwgn4aKLhHjZtX+f+BssgOsmFjmLu3KKm7qyqYHB3BcVMioWix1X450IPAfTZXGRqjUWk2TiNlT8Nr3iUXkv/mN/D/tvbkH9izeixMVF4zzzBEJ/D7i7Hz/9pv9Px9dUZdJzc7vAnTVcVF5CFJXAWIS720JH6qYSOgGWtQiPlKxNEC/ipnrNNSJ3Z3l5eaBqhfl5cfMol8vrRCO3a9AYwwukwwBI8aL5cWPCYUYEJBUQ0x2f06vLbq8GdYBI+FXH3ma4L566+XMjvmkxqxi7CvXTvR21Y3KOm+ebUONhKrPd50OlyckkfE/lkbWNB2hP2E1ukFocis06nNF0VuKRIKfbJo93QlUYmS/4CbCcWAyoEnKmp85rWRrspm7QSrzwxhK4P9zARpXTM7jz0r/LLucckDONSrv2JOKlYHacMaeOzeXlJTAjh+Lrfx3G869FQ5ZJF/1VNP/01+EduRNcOjrVaikWwIl4GSxsFMnjgYG1iBfOOZaXV+R7Gsx5ScqlO+e8mDJtACzqGTLeKki8jIhAhYwu/hEw3QTnHIHvwbUbaK4to7a0gDAMBm5UpwQKK83FFnnsvDQ6JfyOXg0/84xyXp4LADhfzhQCgFMnkp4oyYDGVucF6C5ennz0GD7/21/Ht//i7zLnuwCtzstqx/0hSBJ2U42hpL3q/dPXETx2F/jaGZRkZcXMRZfjFbtO4FWzj6IYdL6AJM7LbnB7DboO6IG4+Z859UTLttWqLC12xPfzD1PFkYsXFSa67LIXxCu/QUJHp06JG9I61yUK45DQCy5+bvy4ZRhgzIZjqnLp7I3qlKPS3Xnpn/PSzXlhjMW/24yk3VC6HaoxnfqbiFTn6d6r5dWzqseL+NnQGPLlwY75vLxx+Kpk3bfhum5ctdR+7kSxtZ/MtRm414sKBQdeLMzWOjhhaTjnCM+I64X5vGR2jyotVyXSAOJGdcqVAQCWL8eLtGHDqonz0j/nJb1dr14va2fPgjHxuhdffnmS85Jv7fGiaA97alN7UbjuF+BfJRZCpZxYvLh/+4do3Pqf4D74dZF8LfczZO7ioM5LOl0g3aCuXq/HVVTZxUvn655VVGXukTg2tpnRlKUQ+Nv7HsOStxt44FHggfeBdUhqingFP/djLxL/3y9sFFcaJRf5nn1ipLU/6LTqfjQadZw+LWb6XHyxcF5m5+bAuQnGfCycOIGLL5FDxdrKpdeLl9YEteNPn8K3v3QEoVbB0YfPojCnQkaDi5fAczFdlmGjHkKwvUkdIHKSvNoCtLkLYey/DPr+g5g6/AngzGG45/8IdnsM0eknEa2chD57YN1rqlLpubld4CfE6+/Zcz4WVk4BbBF2o46C/Hsq38BtOmARcDRn4vjKs1ifBjw8ynk5cOACXHDBBVheXsLx48dw8ODlPZ/XrUxaDA7lABh2TZexd7aChaU1WKYOndVxrL4Hz0N/5yWXW9+CvX/YqHvOS7cqpTRTU9NYXDy7Kc4LjwJAB7Q2d1M1qouC3o6B36xDA+C4sllZzoCm9S7pVSiXwA9CRJxDCwOsSndK07R1CcxxtJax2JUZOOdFN8GK0+DNlbirtt3nZsrtVTHfjLHYRREvJkvLU6GLpFFd6w1Qm9qH0FlDtHIK+q6L0A7nvOv1LXLW4hEpg1Qbie3WhzXbeeTBh2SStokXHXwBflflvBTMuAAjTbewZ9OXwxwvvQq5H30dvO/dAd5YhnfvnyFaORm7UaqaaVDxEkVh4kB0GA1QKBQHdqLUAq9b2CjppRWChz4YBtvPmwWJlxHAwwCLngNN63zx4lwDYxE0toZamEMB3fu1KGLnpTyLJ7/3T1g48QwOvUhY+B2fq8TL8B+jBeW6zM7OtYgKzg0w5mPl7Jn4MdbWZVfXdRnb5vB9PxkvAGD+5Bn85RcfQig7zvqsErexHjRZFxDOy0xFnKyrje4dWdurjQAgd/X1sA79q5YLYVl+xnq9Dm3PeUK8dEkebMl5eUKc6C99zbX4+hd/AE2zcfdf/BV+4s3XA2i186+6J0Kp4OGBsIJLA28kzQSjKIoTqy+44EIcOHAhHn74oYEqjrqVScdzi3JlMM3FC597HhaWHkUpnwNjEebXOJ6nD97rJb2y7Ve+OUjYqJvzAmx2ubR0Edp6iDCmgUOKmx7wwAV0xD1zynmz5/ZpiqkO0k4AFE2gdkYcn5VKBZrWaqJzOWkYTItDl4PmvADinE6Ll6BPmbU6V1hlT3xcB56XNKg7kIgRPV8EHNGoLo02vR/hwuMdnck/+x+/gYCfBqDLG70OMAOMmdB1A6bfwDHpNg7uvIh92st5OXlMXAd5lEchl2tzXqbXbd/NOYxnrFWnYb34X8C84p/D+95fwTv8ZyLlgInrhKWzlu37IsM3jLOOQxkHTdYF+jsvJbkQY4zDd23kCtWO220VJF5GQO3Ek2CyQiUIzkeUs8AMHTANaFYOWt5CdOJ70DQHP3hmES9DFudlDoe/eSs0zYEG4AoIYbN+FSKdnhE5L8/IBD2VrJu8MXHINFMWdOy8yJwXxhhM04TnefA8D4YhLhKLp5fxF1+4D6FWRMlbgm1UEWkGnnpc5NZkES+B5yYJu81VBEHQkmujWFdtJGlfwaWbQ+mXXYAAnSsfoiiK2+/PTlXj3IPyrj1gwQxg2Th94ghWF8/gwW/dgYUTR2AaGvwgghesoIQioDXgnn4G+fM27r+cOXMGrutC13Xs27cfBw6IPhqDdNntViatwkGLmIHGGnjpwQtw0RWvgL4kEoPrdgSUgWhA8ZLOj0rKN6c7PqdXn5d+YaP07zbDeVEhECvXKjp1Q5e34X79UITTYMs8hWph8GaSpZJq9gY4+RkUwxWsnBTnTacuvRyJeKlWhSuTRbxolV2ITj8JxK5S7zBBtCxCkPpMIoSPPi4SlDlneP7liQtYmJ6BPY/YlYn/5rTIqWsXL6HvI+BnwBgHELRUOwJAGAKhBjyzpJyXwZyGTgnl7bi1JbEHZE5KS85LsXvOS7tz2H4dYroB8/nXis7CtdNgulgA5gytZfteOI6D1bUaCjkfOfA4HC7e5+Bl0op+3X2rqfOuVlvG7um9A7/2ZkDiZQR8//A/gDEg4hb+1Xvfi3xu/clz62//V0BzcHZpFcgNkvMibpC1wIImE8aOPfE0rsgBCFxh0bY0Zhtt2Egl616cynkAhPMCAH4zWYm1TqIVKPGicl5Wl2r48v/6BwRaGQV/FbuutHD6oVXY2hyOPvUMgPUVE92IwhBh6KNcyEHXDIRRgIWFeZx/fmuIx/e9+MbZntDYTvrE1eQFuJN4WV1N+pRMFy1xO9J0wCpiZt/zsLJ0Epydxl/974/KnQPkLRN+4KKxthvTZQeMhfjhvffgR/5lNvFi/80tiFZOovi6/xKvbtOVRqZp4sILxQp3sJwXVSa9vtIIAI47JWhsBWAafvHf3YAvffo3ANSSvIsBE3bTN4d+Lct75bwkc426X5A3s1yayZttebr1OM0VcgjsRNz0e/6yvDFdvGfwG0uxXIZlGHD9AI3cLGabK1idF999rwo9prFUzksG8aIqHFVbgbB34r06V7SZ8+PHjj4pSsc5z2Fudjp+fNf5F+D4/D2iUV0YQJfJ9nHFUVsX6h/cdxiMReBcR630PGi+B90PoYUh9IhD4xF0ax5BmK3aqFNCeTs8tAEdgJEH57zFeUlPlFYkx2/r9b296hEAWKEKlq+AO2uwciYCF1C6uF/CruM4eP3rXxOH9hlj+N1v/AWq03OoVqvxZxqkQZ1ChR67/e2p2eS4r62uYPfAr7w5kHgZAadPiQMIUaGjcAEAMGmlurYQL81V8Cjo2g1XlVJ/75nT8WOBWwebEgd7VF+EnhYvMsem3T4elqefVuKlzXlRwxlTFQGqtBJeE9xtgOVKLSMCGmtN/Nnv3QlfqyLvr2H2BSFe+/rX448f/mMAc/Fk1EGdF88Vwokxhkp5F1Zq8zhx4vg68ZK2XtM5L51InJeUeFmdB49CsFRegsp3mZmZgRHY8ACwfAWMMVzzhp/GN/7XvdA0G5wDUTQL5lVRyD+EteYCfuz6n8VTd90OXa/hxLGT+JGBPq2AezYCObwuPP0kjPNeACDp8aIcF/VvtrDR+S2PK1GyZMupwtyAlSuIlV1YA+eDipfhc15qtfVu2iDOy2aFjRr1RtwiYO+FF7b8bmpmBg37KBhzu+ZlhGEQh1BWZLfl51184brtulGqlGEauhAvhsxhOS3cjk6iX/WqZLqeqjYaLGEXSBYkGpPLoj6zm5TzoqWcl1VZfaVKoxXPe+HlOHafCD8cf/wJXHzwoHhuXC69AM6juKLr6Ud+IB6PinjPL/7yur+99vWP4etH8vBDcQ0c3HlR4rp7SEw56qVdu7C2thZ3DZ8umNA6Oi/JiIA0yXT7lHhhDNrM+QhPHUG1ZGLJBfI5Hm/fK8fn0Ud/GAsXQOQDra7VsdoWbrrgggu6frZ2+o0mKFUqcQpEc2XwY2mzIPEyAjxXxLLjjPEOsFxJuMrcESv1KARvrMQXiXaU87J4JjkJGHdELNpZE+KmJalNHPSaPhrxcvSoiPW2Oy/igwI8TCxfZubiFURUX4SeK8Ulz2urdfz5rffA06ZhBU1UntvA669/k3iv+QAIgGZDXCAGFS+1xeSiWC0l4uXlL2+dQaROwmKx2HfWSfrEZeU50c048MBrZ8CknQ20dtdVuSGqSqI6XUa1chXqZ0+huvsCXPkz1+G88+dwxxP34/TSgrDt5YXc7rHa60S4lFR3RWePAVK8qJ4uBw6IG+EFF4h/5+dPwfO81ETwttcLQywsiIvf+rCRECWOz0WKgbTM89UZuMsLgOa1bNeNTtUc/XJepqam4nyp1dUVzM0l50e/UmnxfPG7UTsvjx8RTcY413HRC1oToc97znNx8uQ/gbEQS0tLmJtb37vlmceekO4Bw2mZL/b8y3onVKeZmq7CNAwALpYhvo/Vs+I86Oi8MHU90DMn7ALJgkRnXAbDulczcs4RdhAvXjx3p/U2s2duFzjyYHDwzBOPxeKFVXaLa2PogdeX4muj/f+29+ZBcl312fBz7u29Z7p79kWjkYSk0TIz2oxsyRbeXn9+/RKTAIEQCgIfBuyqUDiJE2KoYn0pUyQYhyUBx4mBBLPkAxMnfg0vgUCgwJtkG8m29l2aRbNPz0zv957vj7PcpW8vd2YsecJ5qlwedd/uvn363HOf8/s9v+c3O8VsEal3mo3MjABmQkYE69W81Iq8zE5PQeMNGTft2mGJYAMawgHN4a4rUEvz4k5fa82MvHQ0AFNTQCTIxtkwDGSzGYdtvx3Hj7Oy+9d0d+HNN25FMdOA627/X1jQG5FOzyKdTqNUKuGmm/5HzXEQEMSqkuZFCwZBoYPARKZGZd3lgCIvywATJWgAtEDlHHaiowvzQ6dBtCwQawLmJ2AuTEm9iB20kJUGasV8BoKPEG0BNLqWfWaZaJctVvoykBfDMHDu3FkAVpm0hIhCuOr8SSPzPDHnJqC39MqKo2//3f9FonkQQSOH+KoJvOUdfyRfE22JIX0JyOa9yz0rYXqM38jNGBr5IitEq3Z4hWorwd5ThBANWqoL5sQ5GDNDMhcP2D1e2nhVDkBsrQl+5663lb23uHnMzc2B8DliUH8+CeaklQYypuxNGIVYl+2wmptbEI3GkM1mMDx80YN8MkxMTKBUKkLXdbS2OgPAQstSMikCOgCNnXNH7xqcnz4KQnIomBpC2XTV3aE78sK8J6oLCXVdRzKZxMzMDKanp1zkRQi7U56vBV65yMvoGdGjJ4xYq1Pg3Lt5APt/9e8gpISTRw6jZd/ryl5/9hgjP/mijuFJNr6bdnibWXqhMdUoO1lP8gxOeoatAV7khcIEAaAHA/K68iXY5V4vIVIC81SuHHmhuTmWxoaz0sgo5qBrVmm0A2YQ0HOYuWRFD4imQ0t0wJwZhjk7alkwlLIgAYAEykkJzc2DZmagoUmSl3qrjWppXl749ZPsVGkI1+y7Fr/61TMAuN4l0iirpuywV8vZrw13jzUBrZlHTAPzOEwJQkEdmqbBNE3Mzc1XJC/H+HxqSyURDOiIxqLYsHELtOTidSi1BLv8jAEAuUx1AfflgPJ5WSJoKS/bpcfa2ise13/NNbynRxFjNMVeW0H3Ikuhw3EQavNCICZOzwe9X8vTRrqHaNUvRkaG5a7dXYliXbDOm6/mqjiS4X5yHLqRQ6R1BG/7f9/peE3XBhYlyBYFealPvT4zznZ51Iwh0cBuvBdtvjMCXqHaSnCL1bSUt+5lgu+aW1vLIy+VYDcJizTxm7GWk+0J6oGdvNj/dkdeCCGSyIgSai+ItgDt7R1lQmcZUeEajgBPT/ZtY4kujSxgtNAAmCXWFqICrJ0tOyabzUgNUjXdipduoFQqyZvvlYi8ZKZ4dR0NlKV6A4lmGRWYuHDG8/Uz4yxKMjGdBwXQHAuidV1lM0s3ookG2d9oOpMDAiHMZXlnag/Broi86KGgb4ddwIq8RIWotkpnaZEyIo2tkpyzJzwM6jhEr6P8vPOcPEW7XPMX8milYEyxOR7WDN+RF6GNqZQ2Gj9/lv1hRhEMBp16F48yacCa14ZhOMbbq+oRAHSuEQotXAKlYRBCZCfwaqLdY8dYv7F2bvaoE7Osq7RfWKXSVcgLrz4r1HAlvhxQ5GWJGD7yAjR+gfft3lPxuLVr14JSdrGcTrMJUMllV5ZCR5tANHZhmTzdcG4yW+G1jLwEKqQJ/ECUSff2rilLt+j8wnLvxKRoV5y7KN2mU4gljuEdd767bIe+7apt0M0i8rwnUrUdtR0i8mKaDUjE2ed6aTwqhWq94M73CuGhm7yIsenq6rYiLzXIixV5SWPdzl3s/bUcJk+9XPO8BAw7eZkeluZU7sgLYBEZIeb1QqUyaYCRF0oB8NL/xlZ2Q2lq6eQE3MRIKcWOzVS+IbojLyKUHg6Hq4oqLfIyIx+zpzyqReheqchLMctdR2n55oAQDTDZ41mPKikAKPB5NTbN/r+hPSnbLNSDSLIBQU5eZmdmoTWtwlyeVy253ZEplZuZYDjqu7cRYKWCY9wav1rXbEvv4tROSYM6j99atAswXW667jYBsxMTMnWzdnCw/LN5OjUZ11Hi5CVUn3WOpybLjiKf24KYCu8W5vGS8nxNKBSSwld76qhS1aPWzMaMLkzLlLJoBVGJvBiGgRMnWCuKjia2tuigZe0B/MJP5EV0Lb+SUORliTj83PMAANOMYmPfpqrHUpNdBDNZ3nekgjOsKKM+W2gEIQYo1UANtgBlFtjFXub1whcrdxnnYmCJdctTDkERxnTtxNwuu0W+wzZME9feeJVnaiEeiyNszCJfYDeGetJGRqmI9BTbxZpmAo088lI9bVT7JmFP7QBW7t6ccZKXl19+EQCwdesAaI4tbqRCR2vrvcXOdw6D27bDpGzRfOnA/prnBTDHW7FIs7I2A+b0EGZnZ6QIUwh1AUv3Ui3yYhnUdZU9R7NpzBlhaISRynW8zFXTNFkymjbYdzCruOxWIi9NTc1Vq+K8vF5EJKWhodGzJF7glYq8iK7tWqWu7cIPJe8diaLcwG6Cu0lv6C0njdUQjEUkeVmYTUNv7sF8gV2DbvJSzBdA+XoQjkbkdZXP5x1l67VAGlqQCHBXYRRRrHDDEgTfXibNXsSOj6XKoxQiGkOpkxRpSWfk5YVf/xoA8PyxETz2b4+iWHSeg8kjL32bX4MCJy/TF87W/nKwC3YrRA8pe1zj7sAi8pKKervrCniV+1dKYZNQzIri8DkU5vO7UtXPhQvnkMtlEYlE0JRga5umUc80lh/YU+eVIHpduX+HKwFFXpaIWcGuzUjtSh/CFn6hWK+UNhLE5PwsW4BMMw7oUf43mzRunxjC0zjhSOUdbb0QZdJlehcAMUEwXGWhIswsXHYJ18QYJsXUaOWyXULmkZORl9rkZXZyBNQ0EYrEQM0IEvxzZ2dny3aWi9O8sAVDl2mjEdknJpvN4tQppn3o7x+sO21k7XxnoWsaKL/xT01UL5cXMGcvAUYRCIShdzCjQnPqgiQnbW1tjkiGiMJUK5eu2BqAUtBsGhcLCWi8OuY1fZawlHKxaIFb4lcT7VoOu+wmUK/3hNfiX49Y1/68qFZaLogKK1LhGtd4ZQwtVYpQsMfH+Q1w4/qNFY7zhqbrMm2UWZiH1twjIy/uaqNCJgcRiY01NCAej0uy6KfiSGtoQStP2RACjAx5mzZ6RV7sBnXtruosANB5uwMCF3lxRV7GL5yDaVL87Lkj+D+PP4annvq143ghZG9fswm8ZyWmbA0fq8Gd1nSD8IhPooOdk4i8JCPe7roClu5lhp2jYch+QV6WDRp38Sa8GMKKvHiTl2O8R9bGjZugaVwusAwOGWKTt7CwULGhLpHVpqq30YqH6B9BSG2tiR7nE5df1JWM6kTqZT7Hr0YaRkRUZ/A+G3Rh2tUci+d7Y8tBXkSlkbtMGki2sUgHgbNjNHEZ1Yk13jRMTFYhL2YwD9P0Dn97YYaXh6baemCSAIKBMOL8xu2Ovlh55vrTRrJEMdEGaAFW+cD9a44ePQLDMNDW1oaOjg6YdaaNLM0BP56IDsG1m1YClsZFa+6BxivMjInzkpyINJGA+He1cmmRNurudu2WCxnALGGiJBrrBRG0OXeSAI+4oHa5tGW/LiIv9bl+emle6imTBpzROz9pkloQaVK7W7Qd1sbFe1EnpABKKcZ5KqxvwE+hPIMgL7lMBlrzaszlvCMv+UwWhEdGY40JaJq2uNRRYyviAVN6O41WMD708ng5e+KkzaCuXNsTE3OgzKiON2jMzDB7gGwaE7PzKHIieujQQXkspSbMKU6cmnuk5sUo1HddVYu8nDt1ChopgFJgYA+TAyw28mJvdOi1kRItSHSNMZBwUERevCMgQqzb17cJUi6wDOxFpI1KpVLlCB2PvJiGIi8rHpTvHPRwbdLQumYtAMboDaNa5IU9Lm7q0CNYN8DyvZqWxbwRBKjJ7LsBZHNZ7j5ZbqC1GFQjL52r2AJFiIkZW05XCHaRX4CRm5cVUoZJMTsxglKFMCNpYLsNjWgVlfV2CL1LU3sPKC+Wa+bf2X2z9hd5YReuaZrIZjOs8kGKB9kCKVJG/f3st7CqjaqTLittxHa9Gp8rZg1TMwFBXvSWXugtvfIxe08jO0Ta6OLFC9JQz42KfY14NCljirC+s4IuxBsJUo1HD6uQF3FzyOcZeanl8SLg3rnaX1sr8hIIBOTvvaypI675iDR46woCIUFqyn9T0zRBSA5zmTyyhQJ0jWD94G7fpyDSRoVcDlpzD+ZF5CXq9DXJL2TleQhL90UZ1YlUMBfXzkyMlR1jZtPyOrBXGl04eYK/NoympvI1qVUKzPNyAwgwSwlxPbEGjTkMjVvRohdftJGXuQlm2KkHoCU6UOJzXa/zRl6tVPrl/c/y84/hNbyHmzPyUk1w7vR6EetQKBTytC4QpC+i8+h5UHe8zg1RJr1p0xYpFwgEln4rj8ViZdVR5eDkxVz+RsB+ocjLEmDk5qSgNtlTHhp1Y9fe67jJj4ELpQbm1+Jhuy1FrzzKEkk1o3/XTqmVOGmwqiZBctK2Rb7Rh6OiF9LpWVkO7EVeulb3gPKuukNnrMoKEoqChBkBOPzMryF0vlogBEpNq7zZhVQ3t8UOxZHL1c7Hi0qjVPsqmDzM2sx3Ou6KI9lPpIa7LsAWMiFOFq/TbKkjwIO81F1t5Iy8JFbxhVvLoDhfuS+TgBDrai290Dh5MSatyIsgKwIdHZ0IBAIoFosOIys7KrnrCg1LSWg7NOdi27pa3HRyKJmkrsiLl+alGqppXuoRdS+3aJeRDzY3U66ycoGoiO55CFsvnDkDQkwp1l3TFEW4tX7zMIEQd1Uu5PPQognMcc1LA3VGDjLpadklRKSU/HaWBqyeZV4tQQSsnkatIEGL6M6Ms3nnNqgTWL91KyjlRnWnnJ3YBQkqTQ2BaFkMj8/I51566ZAk5KLSSGtaBaLpKPLHg0GzYtrDjmomdelLvNrJtH0nW+SlHvIi5m+lSiMBkTZq5PeSSFhzvM4Ne+SF8MiLIDxLgb3BZ2XRLk+PKvKysnHs2Se5oJZg+74bah7f1NwEStnkGCqyxYS6mgpSaoIuTGOhpFsq+4FtCAWCoAa72CYK7IKSLQS4Qy0AJD3KCf1ARF3a2to9XWnDoTAo7xI97soti9TRqZcPQ+dh9EZeGjx1ybvypa2TPR8Ox/HScy9WPTdKTUlemtp6YBJ2wba2sJthedqofsEuIcSj4oiRF4Mv0BZ5GWBN+LhWp17Brtj17rj2el61U8Lp556qeW4y8tLay3Zpmg4UMrhwjmmT3JGXQCAgXXO9dC/z8/OyjLOSQR3laQc94oyGbdnOq6VIBmOlhhrkxakpWIrmxQ95WW7R7sjwsOwSv6aCKL+Vk0B3JAEATh1hVWVj04zEre9uq6idqQYZeSkWUSwWkeW6g3jJ+Rss2NxdG5JsPRAkZjGRF/DNSsHjJi8E7e5Ko4IwMfOozgKA9tZWqZ86e8JFXrho99yxoyCkiKEJizQtLCzgzBnWdsCcZJsVcfMXaaNgoID08Nma369atZHJiwgosciL6GmWjASh1aF5EWS91jqkpboAQtCusd8mHKwc/ZiamsL4+BgIIdiwoQ8iTbkchRrsHEXhQiXywttFmLXJ4SsNRV6WgNNHuOumGUdHe2WPFzsoD8fPGez/bt0LzaYBs4RTxRR7noYxcBXPj/MWA3mDTRxRLj3PKxgo1RGO144yVIMgL15iXesk2YK04HKRFKmjXGZe5m+jjex7VBLtZrnZUTgYw4UT1S3t52cmUCrmoetBNDa3w+Q6o/YOZsy0lLSR/Tg3eTFnhpBOp3H+PCNgrNKIX9yEMPfkKrCXSgNAV1eHLJs/dfRo1deamRk2JwiB1twDogdkROgCP5/Vq9eUva63t3LFkdC7JBLJMoIqyQt30W1ocZpetXZ0S4vw4UKjz8hLfZqX5uZyl9J6+hoJVLJoXyzOHjkMADBpEB3rt3ges3ZLv4wknDtz1vHcDB/vsSk2ZzZ6RDTrgdC8FItFGcUDgGjOuYaICAmlBPEYm3uWy64/wS7/CwBgFMojo1JzknKmH2UJtJdBnTyIrWczrk2QiLycPn8J2XwRk7OMSGzcyHqBHTz4G/7ZbG7rzT0oFgswuQYwFNRx8Ne/qPX1bD4vXoJdNmcDtvVUpMlTjfGqZckW+XaTF+91iARC0BId6NBzoJQgEmLrmv03Fjh+nK0XPT29CAWDMsIWjS0PealdLi1Scoq8rGhkxOSqEBr1hM4W9BIPa7p1L+Lf40IwaUTloiV6GVHenVaUWmfm+WIFzRG6XQwqNWR0niSLeORcOwMic+Q5KWCMNKQAVCYvYicYDseRmaru2ihST8nWLgBEkpduHj0YGnKmjfyY1NmPs4zqLK8XEXXp6VmNVKrJ0ruEG2QPlkoQItJCoWC5zfJw9MLcQsXXATaxbqJDGoBpLb3Il0yMT7HF0at/idXjqHzcBXmpVCa9YAZlmfSaTZsdzxOiSR3MrBGGmasn8uJMGwlyUglul1JgcWmj5Yq8TA3zMaQhBCpU8zWvfo2MJFw45vTvyQuPF94Ppm9z/W0B7AjxlhuFYlHqp2JBHVra2cgwn2XzlyIgr8PFaF5IOA4Eo9CqNGcUkRd3mTTlBnWoIHAGLC1NwWU1L7Rm89kiRnjUZfXqXlx//U0ALN2LsA/QmldLUTgABAM6xkdqVxxZ5NpJXorFAjSewmlbw4hmPp/HPBfeNjd7t3QRsM9foL70tdbcg1CAaYTCocql0iJltGnTZszbTPDidegF64HVnNFb8yIbddaRlnulocjLEmBwQS2ptrtwIZRIsT+4AZjb5l/8O2+KngAWGUmtYjckomVhmhbRkSSC6qwnzxJw5owQ666teAzl5MVweVpoja0oGAREW7AiLw1s0ZyfGUchV05OZnnUKByKw8hVr9ia4eQl1d4Do1CC2HasXsPGZWRk2FEe66fayH6cjLwkOwCiAcUcXnqBCfjK9C41UkYAE8KJm4iIvhCuJSmZ1UW7dr2LgN7Si+F0jp9zo+cN3TKqK4+8CL2LO2UEMPJyIZ+QxosbNnu5wPKbqBmo06ROpI3q07yInStz1WXjXG+pNLD8aaPcHH8fs/J1rgWC8vnZS04yQYs5lAwDU2n2Xfq2198WwI4Q35gUi0VJQhrDuoxACOTlzdjSQSymRQDArmkpf/WYq5UN6hh5CUQq31SlUV3BmbYRUZwSpTJlNDi4Hdu2bQfAKo5oqQAzzXQ1WnOPjJ4QQqBrRPpMVYPQvBQKBYew/aXnXpD+Wruu2wfAmrs6ARpd0Ug3xBy1NC+109dy/GiwarWRXaw7yzcvlAKNda5xtSAIVsWu1stQkr1cUORlKeCh9WCdO3sA6NnEws6EZJE3iOweLSArjXi1gGZz4dy+91qplRgqRiXRsUiBVrFLdb2o3JDRDk5eXLl9rbEVpwpJEGJC5+dBQdCQErqX8htpOj0DAAiHYijR6hfgtNC7tK9CMWMRp5616xAKhVAqlXDJFoIWN756yYvbqI7oAWgJlg58+dALAJjeBUDd7roAeKmqszleiJM6aIWquxhzkosSWy3yorX2YmiWLfirV6/2NHyzKo7KIy8WefF2150osegCNUMIe7ijEh49NACgmPUUnQNekZf6NC/hcBgxXvIvXnMlBbtGXny/GsslT6fmy0LuRUzMMO+MRDiAzo3bFnUeYV6yXrC1SmgMB0CzadmPCgBKIr1DrfNdTOQFYKJdnYj56axco7l5mTbUKhrUVf6tKxnVkYZWQAuAagUp1t22bTsGBxl5OX36JGYvHAcoZT2GoklJkIOBAGvsWcURWMDeA8kefTl7mEVZqRlDks+lyUneRyoShN5Qf+QQqC99LXQ7oIE6Iy+bMCfJuW4JxpeIWmkj4fOiIi8rGJnJMRAeWu/s886De2H71btBaQCEUJwupGC6BLvm/CRME7KKqam7Rz7X2dUpBb/ni0lJdEoiZEqX9nMWi0Wc514OVTUvwtPGtRMjja0YKXLHR67PKRaLaO5gN1Iv0a4VeYmhEGhEerby4iojL209mJ/l5IGaSCSbsWoVGye77qWWyt8Nd9oIsHZEh7kxlN9KIwGrXJp9v+4tjAQRkkGmQiUW4CyTFtCbV2MozW5QPd2rPF8n0kYXLpwvW2istJE3eVkwvMukBYKceMly6Zx3iFk0viuVSshmM3IRr0e34r4BXMnIC+XzXLNFMrxAeETSdNkCEFLA2Az77us7krJU3i8ifDyLpZIVeYlz80pb13GxqaDwIi/1a14AQGtsQVC6aTvJi+gkTRpaQIJWuXaxkLcM6taU67EEdJ6CcxvVEU1DId4GYAHDPPKybdsONDe3SFL+4v5fsfNr7gEhREZeRF8gaFkYNVxgw+GwJP523UuWN7y0z3+rNUB1jxfAmt8LCwsoFAp1rUPC60WjBOEK7QHy+bx0P9+0aQsW5kV6UJeSgqXCij5XIC9SaK7Iy4rFoSd/zk2YAth57b66XxeNRGGa7KK9VIyVa14WpjBcioKQEqtiuvY65/NcKzFfCoHm50GLeZRkf5Cl/ZzDwxdRKhURiUTQ0dFZ+UCdkxdXZ2StoRU57ryqcavqQqGAli5OXjx0L2IRjvF016EDhzw/MruQRi4zB4Ag1daNOX4z02gJjQ0JrFolNB6MvIgbJlA/efEqE9SaujGxUMDY9Cw0TcOWLUyv4CfyApTvfHdcc40ksYcqiAtpMQ9zhkWS7GkjEmnA0AJbPFa1eHvMCPIyPz9fFoGw+hqVp43MbBolQYKJdwqyWRBqkoNhAjTjfUMUDrvsM9n30HW9rjYQ9ooNSqn8Dlci8iI8U7QaHduF9onaSD3rLJzH+DQX6672Jpv1IBpl87NYMuQ8EvPKQV5kCsQ638WY1AEsmhrWRHrcuVnxMqcD7AZ1QF9/f+XvI6IyHlGSk7kkpucWkCuUEAqHsXEjq/IS0ZdDh37DPpvf9EXkJN7YCEoJNFLEqeefrPrdCCHeXi8GT0EFrPlrebxUL5MG2FgL24Xp6em60kYk0Q7oAQSIaYu8OMnL6dMnYRgGkskk2ts7kBPrFNWW3NdIoBZ5AZcDgCjysmIxeo7diKkZQ6zOLqYSnNFnzUBZtZE5P4nzBb6zpXF0drlIBA/Zi8vdXJiEIXUeS0tIikqjNWvWVW11oAe8DblIKApTpNJs+fmWTrb78iYv7MYX5+WBo6dHyo4BrKhLY3M7AsEw5vkirNESIpEG9PSwRUyUS9sJSP2al/LeHlpTN46Osfdat269NNKzDOr8Rl7Y66KRKKggsUNDnq8xpy8CoCDRBEjUecMfznDykvCOjkQiEbS3iyos57jXShuZvCy4UoRgk61ceqIUlT2e3AiFQnJnKz4zmUzVbqMBp1dGJrMgy4+vRORFzPNguLoY3urobpH64YsXQUhJerxs3OivLYAdwnCuUCrJ66YxyUienbxQQV5oOXmZrRLZ9AJpaEWUkxd3SxBL7+KcR+eFQR3CSFUhm22cYHuVl4+kKYZ4yqh/6wCCPBohdC8vHWfl1YK8iMhJY2OjrOQ7cch7I2SHl9cL4f5a4aSVHrI3ZaxWJs1eTxzl/lb6uvJawUwxuxHWSjbNy7wjaio6SW/atAWEEJu2SQMJLhd5qd7fSNNE9FGRlxWLvEzV+G+GJSaaQUwgvwBq66xK56cwx3vfULN8sYyIC4qTBDo/JTsML5W8CLHuunXVSzn1iOhL4oy8zM3OylRaKsUuApY2Wg2AIDs/g+yCc/EUaaNYlFcwzXgL7abHubNuG9vlZeeZ8l+jBoim2SIv7Dixc4hEInLhqwXL48BGXlLdODrGPkvoXQCrpHixaSP2Jrz0vYI5nyH0Li29ZbqWoRk2zt2RymZRogrpwgWLvJRKJWlcV9bXqFQAilk5t+IVqiq6Vq3m5dIUw8VERdGufWcryEstvYuAffEXEZRQKOTQKVR+bQrAMpIX3lk5nqzupByK8OvVFkk49dJLAIAxEXnp37Ho00hyJ2nTpJYGo5mZ5hk28mLS8sjLogW7DS1o1Nl8cLcEsSqNnJGX9Dh34jWrFw/0SaM6E+dPnnY8lyuUHCkjAfH3S+dHQSmF3szmuIicRCJRaSy3UAdRc5fzp2dn5Bq2ZqsVNRLjXU/kBXC67NZbOKA1rUKDXpDkpVQqOSJCgrz0ca+honxOA2wtPJYCS/PiXQWpBRR5WfEwuchMW0R1T0y4dMoeRzzHahRBs7MoigVCL5+QvVvYBUVIBpmSBnN+0rbTWmrkpY4yaQBh4Q3i2ont/+V/gRAKk4bR3cKOKRaLCIYjSHCF/tSoU/cibjCxJNstGXnv8bRXGgFAnvvDiJ4zVmkwu+HXs9txo7GxPGSqpbpwhEdetvZZu2a/aSMv8qLx6JQJb3Ghl94FYIvayOQMAKBLr1xq7dXjaGJiHIZhIBAIotXlFktzc8gZOghhO7rVG70N2Vi5NJubM6VIjXJpdpwgL/XoXdhxlubFLtat1o1aQERepqdnliwszOZzIGDXaVtPZf0GADQ0l/frmRy5iIVsHplcAQTAxh17Fn0uTS0WmbzEU3/JNt7IcHpINhC1vrM1VosW7Da2okkXlTwmpies6kjp8eKKvIjSZ2FmWQktNqO6cyeOOZ6jKMrIi4i2AKwZYTgUxlyuhAszeZmyEpGTaDQqqz+NGpV87Hin18sLTz7JGrfTIAauukoeN8Vdx1PRIEg8VfN97eS7XrNMrbkHrYEMQja3XHsE5PhxIdZlGsuS6PJNNZDQ0nvaAbVLpTUhGSDKYXdFglIK8G6rUY9277WwYTszndO0HNKlgNS9SLddvvONJMvfe3D3VTBpCIQAp7ho17JqXp60kVdbADsS/MbizoGPc/IDI4oIvyGL5oNStOtKHYmbeUcvS48VifcFPj0mnHXZYiUiXyL6404b+XHXFfAMmepBHB1nC+OWHuvmUW9fIwGrRYB184i1cmNDknc12WTwKpMGgEuXRmEYBkI6QYsx5dBY2CHIizDXAywS0dHRUZa+oZlZXCw2yjLpvsHtqAgeNcrVWS4tdDa1yqQFxHFTU1O+xLr240qlIjKZ6j46tXDm6BEQwgzfNvRXrxLq5qRfI0XMcZKQn0vLqEtPUxSx1p6Kr6+FlrZ2BLjuZph3BU+2dTMNWinPev0AMKuQl0xmoe6GoAAj58kAZEuQC3yNYJVGvFdXJYM61FH5KIzqXG0s8qV5jM+wcRvYbJHoYDCILRvZOB9OE+lrZUVeIgiLDYtWu1zaXc4/JlqemFFL/Atggp9fKhKq65q3k+96zTL15h506RkAmiN1BLB7jugmLciLIX9Hsmyal1ql0gHZm0mRlxWJsbMnoXHysn7Ha32/fnN/P0yuezldSFnl0fOTyBia3PmKKIsd4VAY1ORtAopxmPNTcqdFlkxe6ou8tEihZxHFgrXLNPgOXIeOQImNT4E/39zJbqTTtnLpYrEoL5JtuxmhK+pxjI05vW+KhRzmZ8YBWJGXUoGLCDl5EdVGs7OzSKfTvg3q7MfaL9yhoYtI54oIaATrm6zFzH+1kbNUGgA2X30tAEZih44cdBxPqWmVSbc4TehEGqgrEYVmGlLU64ZIG9kjL1X1Lrk0xkrcCNEMSYGoJ3Q2FgZIRc0LYGkKRvjNtn7yYrmU+imTBljqQDTAW6pod+jkcQCs8iTeVkXEDmBt/zZ5kz/5MtNbmMWc1Lts6G6vK3JUCS2tzQjysL2wBEgkU5I8GNLvhZMXm3mi/TrwZVRHCEKpVhlFmbzEbuLGDCOjJN5cduO0DOpqR6WlUd28dU5D589hbHoMlAKt8TBag05iP7CGrT+HJ621R0ROotEYurgJICEZzE969/YScHeWLmaE4aczZT85ztefVLKmKSU7zpq/9VY9ehvVsbkzPDyE+fk5BAJBmdY3hU6IakBwudNG3uQlqMjLysbBX/0SALPu37x9h+/X67ouCchkKSpFu3R+CqcLSZ56CWJw91XebyAFvzpz2aVLj7xMT0/LhX4N735dCav5xUMIMHTeiqQI0hXX8tANRl6syAsX7V46J8mWXVuytX8rQiW2S37pOafQbmac3XCjDUlEuN15ib+vRhmJicXiaG5mdubDwxd9G9TZj7VHXoSz7oaWGALzLJdPqQmaZ+9fr2DXS3OwcdMmmCZbdI7wLrYCND3GO+YGZa8XAUFeVrWxBVKkl9zwShtVddfNpLFQql4mLRDkvwPVSnW1CBge9ps2ssLugrzUG3khhCybaHd+kus3aLAm8Yg0Nslxu8T7ToEWLbHuurVLOpeG5iSCXBQsrtXGxoQUrYo0jgDRrPMNBAIyJeCX0JHGVulhMz/DN1oVxLpAfQZ11sHsfQ2b7u/Fp5+2UkadcZgzThF/fyebey8NWcUOInISiUSw89rXgdIgCKF44Rc/q/rxQkMlS6Upj+i60jBTXPOSaq6vd5ydfIv1RKSlK4HEmxkJsRnViTVS6F3Wr9+AIHdatrSOWMZqo+qRl6AoTlFpo5WJaRHiNKxOxP7BFrm8qTsiL+NFXs1ixhxhSztEszyTGI7IC5awqxMpo66ubrkbqYRkKgXKF7NRfmMcGx2VltprIzPQi+xvQV5Sbd3QNB357AIyafZ9hcV5Q0MDgsEgApTdBCfOOXdLdn8XAaPILx5i7crs3iaL0bx4pY0Eednc3iBLQ2l+gdlagpUt1wOvyAt7M64dmXJWncmUUXMPiOacY4K89HR3O451Q0ReJibGpSZgZESUSZffdMxcGkWhmyLVyUuKN34EycOsmjZic0mIhBejebHSRvW91v45S428FBd4FUqFBoNl4JGEDDfXI6RglUkvsi2AQDAeRYiTF3HNJxIJ6REiKo4oj7y4IwSCQPsul25okS1BCnxHXslZl32wMKirI8rGjepgM6qbHR3C0DhbG7Z2NpaRly0p9v9TQ6MyLWhFXqIIh63o9Lir35kbYq0T5IfwZriN7c4NwxQnwc0t3l3F3RDzb2pqQopfa61FRPQvo3qZUZ3oadRnawxqL9Qguv/CES80NAjNizd5Ccca+Ccq8rIiYVlPL37CaGFBQEqOyEuW+6Sgys430cVv4loW5twkRJh4KUkjK2VUX9M4Ee6dnWB59ud+8XMAgGlG0RPMIGCwMRJpIz0QZPl5AFOX2M1WVBqJRVUPstfkZ50aDllp1G4jLyX2nTUP8nLx4sVFaV7s+V5xc7DIS9wiLzxlhFCsbkdjd6m0hRD/Pk5DLXPCW6wLWJGU1Ty9J9JLbiSTKTm24jVV00aZNAzu36DV2Mn18aoPjWQwMVtZVyIiL8J7xK/mZTFpI/uxS428mLyfD6l3qeQ3+SK/GRo0i4lZdiPYtG1xbQEENE0rq5xLJJK2yAuvOOK/obtztVUu7c+ojjS0yu9VyrPIhOXx4pxHxWJBGtR19Jb33HLDqly0rvlSbh7DEzMAgP6OBgd5oaaJluI02htCME2Kl19m1VxW5IXPW67JKuTLO0bbIeZnNpvDudNnofFz37r7GuszKcVMmv2GTa3VU4cCYv4ODQ3JtaSejZTe1APi0Lw4Iy9C7wJAah2XKhewQ5zjwsI8TLOcoMRko0rDU6d3OaHIyyIgrKcDFSIj9SApbh5azoq8LExKjw29SsiVtQlgwsDhnA5U2Gn5gXBurKV3keA70Szfxc2M8F2fGYUWjiLIhYX2XkNu0a5YRJNJdoMNJkQI2ZkrtyIv1i7PNIQo0brABHkZGrqwqLSRyPeWSiXk83kYhoHDh1lH4S3tDaDZWdD8gm+PF6Cyw2mQu4y67cwtsW75DUD4tvRuYDt5c7LcRVfAikYx8jI6ym46lZoyil1zpAbJWL12nSyXvjgftInGnXCXNvslL9lsBpe4zqLetBGwfOSF8rJjQuqLsBLuwktLJYyNjmJmbhaGSREN6li1eeeSzgWAjLwIJBIJaLxc2JwdZRWL/Jpwm+qJOeiXvGiNLfIGKXQWIvLiLpM+e+yENKjbNDhY871ldMZWoZWen0ImV4SmadjYGoc5a2m66NwYYBTQ38m+i2jSaI+8AIDOhby0QiWfgFVtlMHL+59i382M4DUbNshj5ubmUDLYmDZ1lJN+L4goobhWA4EgwjV8ggBAa16FAMqN6kTkZZO9USpdfvIg1kBKqcP7Rj7Pr0FCKHIL/iJ4yw1FXnyCUlNWGiU6ym8A9WLbHkFAChiemgOllJEYHrZMeOyMBbp7Vsk2ARcKKYgbuD3H7RdWT6M6Iy98kS5k2a6bFvnuW49Ca2xFkJ+LvbJBiHZFubS4kSd4s8qWXlZ9UySWmt80DMxO8koVW+RFyHyIZt007S0C6lX42xGLxaSuYX5+DmfOnEY2m0E0GsPaHqvDtF+PF6By5KW5l+uHtAwKWStUK5rtaS3O8lxKqSQivZt3AEQDzc2BZmY8P1fYqV+4wMbcctctn1+5uRkZNu9+TXUzNWe5dFhqgNxwL9j1pn7i8TgC3AxRRAX9RF6Wz2WXkW+9SndkO4SJF6UGjh06JPUu69uT0JfBiyPoOo/GxgRILAWE4wA1WUSER17c5yzmoN8x0RpszRmNEiPwfL65Iy/nTlkGdfU4KbfyKkFC8igWiygWCxieYnO0p6sb4YAGc/aS3OULP5uB1zDCJpx27dVGANDYxpsnatmq5fJ2n5c0b6hJqZNwi/5a0aCGSKq95ncCLPItNlGNjQ11ibW15h6ENcNRbZROpzE0xL63k7wsT6GGHZFIREohvFJHjU2W5mfWleq+3FDkxSdeevJX0EgBlAID+25Y9PusWt0jnSDPZaJAIYPhyTn53tv2Xlf19cLAbtYIW2HiRetvrMhL1Z5GjhPgzRm50E4YO0Vb2qE1tCKgswuqYKtGkqLdsQug1LSRF7aoDlzFdmolPYoL59jFmp4ahWkYCIYiiNtKxynvuk2IV+Tl4qLIi6ZpcucxPz8nU0ZbtmxFsIWRF2NmWEZeNB/kRXzH+fk5Rzh21w038giGgReFEDyblmXzQs8gMDU1iWw2A0IIVvWug5ZiYezKol2r4iidTssFqbOzPPx9bnwBGo8A9e/YVftLSeF4UJbNulEeeamPvBBC5LGi39aViLwIL6NIvD5BpC7SOqSEyeGLlt6ld/FtAewI2QhJJBKRLsZO3QuP3obKozTAItJGja22rk4lmcYh8aYyYWt6rD6DOoG+ftHjy8T5U6dw+MDzGB5nN8Xde64F9CBglqwycJ4iHdzC0ieHDh3kUQKr2ggABvZdLzeHp36zv+Ln26uNzDzfgLn0XqKjdCoaBKnRlFHATdLrXYe0plWI63lEQux3npubwwnugdPV1e0ihCJttHwghFRtEdDUan3/9PTEMn6yfyjy4hMHn3oaAEBpDKtWl+sR/EAQkLQRgTE9hLOZqPXePTX8IDTe9I4CIm20WPFwsViQzL7eyAuIFR4/c+I4NC0HSoHBvdeCNLYi4BF5SbR0QA+EUCrkMTc1VpY26uzqRLjEiMGR37wMwOoknWpb5UiLmbwk1R55ETfqkZFh+d5+0kaAvbP0vMynDwwMynJUc3p4UWkjseullDoWhVSqySKxPK9tcL0LSXSUVRGIG3lnZxdCoZD0gKks2hWRl/M2r5Umucjbwb34YJphxOJ1jJvG9Tqw6YBcEDtbgXrJCzuWLZTCOv5KRF6EfiPZWp9QM8oFjyBF5NMzy9IWwA575EXMKQDQmth6YUxdlJUgwbBz7BdtVBdLIiCvM0M2ZHT7uwBAIVOfQZ1AS0sLTK5POX/iBE4ffhFD3Fl39+5roCVZBMWcZXNX6Hq2DO5CIBDE1NQkhoeHHNVGANODiQjK8Rcqkxert1EW4GaEAdd1PTXFK40iAWgxf9VyAvWuQ1o0gZYwlZGXdHrWU6wLQEZellKo4YVq5dKRaEzaAczPqLTRikJ6khvJeVj3+4bG3qNACYzRE0jLtgC1w8vhJLs4qFaAIC+aKx9eLy5cuADDMBCLxdDWVl9YVJAX0BIOPflr9ieNY8OmTTxtxKaWnbxomo6mDrbITl46XybYBYAA2OI3dZEtGG5nXQFxAdk1iW1t7QgGgyiVSjjJ+6sIMlIv7OXSIvLS3z8ow+PmzLDN46U+gzpAWNuz37XMop3PpRwnNYYU61bWuwiiJgS99URequpdTBPzpdpicTt0XnlAiVEx8iI6SwNs9ynKPOuBWx9zuSMv4yOXpBFjdwW3YTeaeENTQgowCpbHS98S2gLYYY+82K8brVmkNa3ISyjiJKjieN+RF6JBuMITGBUbMgI+Deo4xCZuZmwUc5NjGJtiYzY4uB1aijsI82iPMc3Wg2jXa7B5s4i+/KZM88JexObefBUCK47PZBZktWSryypiepxpbpKR+tx1AXa9i9J0wF8EuLurBSFOXsZGRz3FugxCmL285KVWfyMhGcgteD9/uaDIi0+UeFWIsKBeCgJxtphQUoIxegJFkZrVapOXHj6RCcnIcuF68/Ju2M3p6jXRIvKzDGS4EZQgXcSWNnK7edpFuyJtJCIvAKCF2PgW59h3EuSlqQJ5sesoNU2Tupdx3l/Fz6JhP356ekrueBh5sWlefLYGELCiOk7yovGKJYObe1Vy1gWsMmkRUakdeWGpupGRYamV6ez0qDTKz6PIx5SiPoKREuJFkq/o9WJPG/mJurDjU45/X+7Iy+nDTAxKqY7VfZW7I9uxdotIgxjI5mYxl2E38407r130edgRdJAXW+RFiHanLsr1INbgFP2L+beYMYmIFBQpVfV4AZ/DJOBjc8fF/4X5OYyOD8OkFPFoFN3dq6THkTkzClrMg86y61pr7pF9jg4d+k15tRGsNdooVRbtig3FxKUxEFICpQRX7dvnOGbqEu/L1RD1VZJsJ9++LBs61iDMTeempiYqR15EZeAyR16sDVylKkK2TuQ8BL2XE4q8+AXXBAQ9wu5+0bbGEmvmh0+A8gZw4UTtRX7HnmtAaQCEUKk3CdWhZveCb70LYBMfGqAG7zMUZItlJcEuADTLDtPnbJoXi7yEU9y5tRQFpdSRNrJDsH93pbJo0CjgN20kjn/hhedRLBaRSqWwalUPdL4DpPOTMOeY26aftBEANDZ6N8cLc8GysDMv1VMm7SIvdHYMtFheFtrW1oZwOAzDMPDCCwcAVOomPQtTVKnU2aF2Pa8mISSLSZe9u4A9beSfvFiLPyHEkSapheWIvEwOMZEzpeG6r62OdRuljcAEr5JJNcSQaFsmzUvQnjay5p8uIi/z09J1OibmlTx+cZEXAIg38JJmUqwaeREGdXrEx/ooSEYxh5FJNmaru3uY74kt8sJIE+uyrkUTsufRoUMHPSMvYRH50Jw2BHaI9OkMF+VSGkPSpVeZnuAeRQl/17td9+JnHdKaexDm1+BcOi2jyOWRF1FVtnitoxdq9TcS3cqLFRrKXi4o8uIDhewCCA8ttq1du+T323XddaCUgBADZxZ0SUJ6Nm+u8UrRJoBdeIJ4B6O1IzZeqLenkR2hmPhsQ46J8J/RGltlDxa7YBewIi8z40M2/w6LvHSsY4tVkSSQnhpDMZ+FpulItDgFpoK86EHnhSuM2QQWG3l55pknAQBbtw6AEAISaZA9TcxxdlNbbOTFrTlYM7CDvR/JYvz8Gevm4ElenGkjLcqrTUAtnw8bNFvH7QMHmItvV5dH2ig7J8u1I8n6SMZrNmwCpToIoTh1fszzGGfkxV8fMPvxiUTCl6bLirxM+/pMO4QNgJ/O8ZqmgXINxziPcPS0Ni2pLYAdIZs9g530k1AMpKEFhRKRIvZEi9MNdrGCXQBoamV9vQgpIZvmYnKvyAufQ3E/v7WIZtAChifYxmDbTuYuLrVms6Nyfoso0yDvvXXs2BH5nezzrWNDHz/nDDJz1dOaBR658UrZi6aMTT5MEgEnWfezDulNqxDmkavxqUkUCgXE43EZVbbAIy/LTF7sXi/e4OSlUN1D55WGIi8+8MIvfw5CDFCqYecNtyz5/ZLJpCx5Pp1LgRATlAawY099nWfd/TcisTrsuD2wGPIS5bs4QjK8QorgqhtuYk+GYgjyyIzbJKoh1YpgOArTKGGal9rZF+HBq7YD1ERJD+P4oRfY861d0HVniMWsQF6WK/Iimhn291teFXKnKbw/fEZerOaMzh3NwO6rYXI78yf//QcANUHCDSDx8sXSnTYC6kkdsTER6QIvzUs+PQGNl0l3rqtvHhBC5GI/Neu9C3NGXhZPXvykjOzHLywsoFisvPOuhpIwo6Q+bw48DSIqjTatWXwzRjfs5MUdidKae7BgO9dUa4fj+cUKdgGgq6dX6kOnjBBILAUSdq43ToO66h247QjwKI2mzWGYi3VvvuVW9hgX7NJsGqURVnUjTPm6u1ehpaUVpVLJKme2RV52XX8zJ9cmfvOL//T8bHF8QZhEBsrJywyvNmqqszWAgH3++tHeaU2rEOG/syjz3rhxU1kjVSHM1gPLTV4qVxuxk2LnYfho8PlKQJEXH7h4nDdpM2O+QtjVIMRqRb5jqdYWwA0t5Fw8Yg21fRXKPp/SRaWNmtqZsFfs8qgZR2dXB3+MIJhgF3rRFXkhhMjoi1fkpak5hbDBbu4XTrJcb5NHyF1EXgIux1ERkRBYbORFwEFeXNUV/iMvFYzqgkGAR9HGhljYXGvtLdutz8/Py7JN0bcIsIl2J6r3OBLo6iofz9MnzvKcP9B/1e66v5NwCM4XvQ2z7DthP/b+7PiU59/1oLExIcdvMZEGADBN0fzT5zLJCYQQ6+7cXqU7t0+EbWTQrnkBWFl9xmTXA6VAoytttFjBLgDE27plBdGsEfHUu5w+cZJvwIC+gfo0QoBlVDeXmcNcJg9CCLbznnEkFJUkvnSObWZEWTghROpeBOxkORKJgppsjRzla5wbYn4KM02vlP00H6+mOivOBBabNiKhKGIh57rm8HeRYGvvUsxSvVCrv5EozjYqXPOXC4q8+EBOhNFo/RUTtUB4d15N42ZvdVZ6AEDCJbyM+1zgAWBycgLz83PQNE2KO+tBp6tM3G3sFGpkC5Jb8wIwszpKKeZ5zw+3mVUAbJwL3HnYXWkEAJRXM4SizvHqsZWYB4P1uVra4V5k+rkPBVAuUPRLXipFXgDI1ES+xPPYzZUrjZqamhznKVx4janqkRcBr7TR0AgL11Ma8UfMCS+XrmAEtlyaF7/ER9d1SYoXLdrlETZN91fFR4gO06QYn2HzeOtVexf3+R6I2CIL7utGa+5BVqa4dA9DOytt6WX9Xg2avTmjGfbUu8jNXZ0GdQJtfMMhWgK0pVKOUn7ZmLTA0tOazftI6F4EyiwARNdqnhZyQ0Reipy8rNlS3n9qZo79js3t/kxJFyvYBYC4a13r66tMXoKR5SUvQvNSOW3E3ZaNFUZeTNPEl770Jbzuda/Djh078P73v19WMXhhenoaf/7nf47du3fj6quvxqc+9SmrgyfHj370I7z+9a/Htm3b8MY3vhFPPfWU/29yGWByJb2uLU8TLAAIJZ2hdN1Hd9CBa5yLYrKl1ffni5RRd/cqXzf6rlXdoNSKDOiuBoWhJDuXYql8gjd39KJQMuQCWrYIh3lLAZMtVk1t5eTF5N1ow64mkva8sN8Fw/2a9vYOR+m4g7wEIyABfyS2UuQFAHS+ezJF5VgdlUbytbJc+qKnTb+dlIZCIc/0zcICT+/5IM+AJcykMDydTK9U2sj+mkWLdvlvEQz7+511PYDp+QxKhomArqN3ax2Gf3Uibpuf7lSE1tyDHCcYFOWpBHGdUUq9CXQVkHiTJC85M+Dp8TIz4c+gTkBEaUQzxlUu53Ih2mUnQhzXodC9CERduj+dX6OV2gSIyEvRMEBpAIO7y6OO0/PsftXc4S/9t1jNCwAkm5xrYrlY14p6h+rp3u0DtUqlRdqoUkuQywXf5OUrX/kKvv3tb+PTn/40vvvd78I0Tbzvfe8rE2YK3H333Th37hy+8Y1v4Itf/CJ+8Ytf4JOf/KR8/umnn8aHPvQh/OEf/iH+9V//FXv37sWdd96JU6dOLfpLvWLgqvWIj11FLfRudk7Kxo76qxJ6166BaVo372RrnR4tNtjLpP0gGAw6jKha1jhfH06ycykZ5Te15s5e5PJsMbH7nwjEWiIgyIPwfiepNo8OyLxGOhxz7rRisTiaeW56qeTFnjICnGkjv1EXwF4qXb4oJNuFnTkjEVpr5UojdxqIJDqAQAgwCjDTo2Wvs6fSOjq6ynPnYL8Tgz9i3ih0FVoe8Kh2Wi7y4jdtBCwHeWFzNOrzeg9EwhjnKaO2ZBKB4PLtjBtsqaAy0p/sQl5oXjx0OlW9hmqAaAGrszTVpa+MHUXu+0F9CJwBpiUxaUiSlw19zjXRTl5IosNRht3fP+AQcoddxnwNYkNHvMWlVuTFgGFEy1L2hUIBmQLbTLV0+zMldZIXf9q7tq5uaVSnaRo2bPAyOWTXbMTne9dCTc0Lj7zQV6C3kh/4Ii+FQgFf+9rXcPfdd+PGG2/E5s2b8Td/8zcYHR3Ff/zHf5Qd/8ILL+DZZ5/FX/3VX6G/vx979+7F//7f/xv/9m//Jput/cM//ANuueUWvOtd78L69etx7733or+/H//0T/+0PN9wmTAzMSargdYMDNQ4un5su3o3qG2hGdjr0w+CCyYp1REK+682OnOGRV7W1SnSdEDs8qiGPTfd5HgqZOsBUnL5LEQbUjC5O2tjQ3nPj64NvdDJDAAgnmxBMFwejRKRl7hH+aJIHTU2+r+o7a+xp4wAsGojLlJcDHkRNxsvweTAdazVhEbyGCnFnbtNDhF5cet6iKZZPh8eupfu7m5JWLq7vXtmmTJF4u/Gs27AKpeeHin/7KVoXpLJpJwbi4m8LMXrpVCwxKctrvGuhYamJoxxsW67z1RZLTTZKojcmheiB1DQxHh7L+1LEe2K9yyZGnSPyIs0qCP+zTKNUgCjk+yc/tcbfs/5qbZrwd0uIxqNYSM3EIxEImXEfPNe5tmiaTmcPfxS2efayXXJKD9vYVCnEyDhkb6uBvt892uWuX5glyQvXV3dZRs8Jo5mkY+GRRD7arBapNQgL2blnlGXA75m2dGjR7GwsIC9e610RSKRwNatW7F//37cfvvtjuMPHDiAtrY2rF+/Xj529dVXgxCC5557Drfddhuef/55fPjDH3a87pprrvEkQ1cSL/ziP0GpgWIpgI07XisbgS0HCoUwNH0OphlFe2eHr/cumQSkZMCk2qLO6cwZFuFas8Y/eaGcvJhmAxpcC2koZVU6pKenEHI3pQuxCyQSDqGYc5oh9fdvwNGnf8M+Q4thSvRL4cjn86C8VUCDxw1x1arVOHTo4LJHXggh0JtWwRg97rvSCLAWsNnZmbLfqqWjC/lCCJqWxdFsCvGh8gjKqVMnAQCpZBMmxsYdz5Ui7TCNs8icPQYtVn6zbW9rx+ilUaQSSYwPD5WfHPcYCvkcs/WbNuM3P9ZBiIGDTz2NgaCTMGZsKTJaLHh/dhU0NjYinU5DBzxfqwc05BdimJ3NwCg5w9jhELedP3MaFysINivh0rnTKBlsTFb39fu6tpo7e3CJu8R2tvlP5VZDm02vVCoUy+ZBTujxaCXyksTY2BgmJsbR2+vvmi+VTBAYKJYMDI/PAJhxPF8sZGEGDJjE/1o0PJ6GYZqIhILY+Vpn6kZqXuCtBRsc3I6jRw873XU5XtO3Bc+YEWhaDi8++UvEXeuFYdNt5E0N2XknqRs9yyqcktEgtEgchlH/DXspmpdU7waEQkEgk0ObxzWbtkUTG5r9R9yrQURe5ubSnr9jqVSCTkxoWEHkZXSULahuwV97e7t8zo5Lly6VHRsKhZBKpTAyMoJ0Oo1MJlPWJK7S+/lBILC8WuSXD/4G//yjn6NQNPD57/zfZX1vOx747v9Z/GuXcF4bNmzwP2YyYhQpe2201fpNb7j5+opvQYwcfvCVj5Q9HuMzc2I8gX/52pGKr29ubyn77N5eFt5tbGz0/Z2SSYuEbdu2rez1enM3jNHj0GMJ3+8tHGNffPEg9uzZUf3gbz5a8amRF/8D/3nJq19LHDj0HPvPhQhPwS1cOo6fffevyp4XG9aO1b2+vlcgEAalURAyj7Hh5/Cz7zo/ey5jLX4Hf/I1HAn625UHue7k/Is/w8/Sh329dnaY3Xi+8c2v4xvf/Lqv19px/7e9y2zrwbo1a5d1LepdZ6UuDv7k6zgTrxRt1Tw/V0Re/viP71zSedz//32/6vMPfOufF/W+3a2tCLrsD2iqlaVFSwUE21aXfa8dO3bge9/7DiKRqPdYmxFAyyGTPoqfffdo2dO6psEwTXzle9/BV773Hc/zSsWjCAR0RyPYWmhttaJkqZTP9SIQRpi30khFC57XrPycrq5lnWNiDTx37mzFdUrXNPzuvus8P1fnHl/i/68UfK0kQmgbCjkFWeFw2LP8LpvNlh0rjs/n85LVeb1fXngsLAKaRtDUtLwipkRbK64w0XzFsGbNGlx33dWIufQjtRCIpmAW0mhb31c23mYyit0burH/5HDV91i/qvLOlNIAirSywj9eHEVP7/9kpcY2vP71/xPf/vY3ccstN/ueB9u3b8XatWvR19eHtWvLQ+Phwetw6eTTSG1+LRp8vvc111yFzs7OJRHzVGMU7U3+oz59ve0YnpzFuu7K422aDfgfv/c7aEj6+14aYgC8Q8zxSBgdTY2IhAMILsKPYkNPGzK5ArpbU75fu7arBc8dvQDjCgkLUw1x/OH737esa1E8vhG9HW0oFHNoiHpraSgFdC3i+bm33vr/4Pnny8ntqwEaIdi2ZZvneZe2Xovs6UNo3bILusvP6vbbb8OXvvQAbr75Js/XhqONKBRmQYj3Ar6+pxXHK5gsCtx0zS4kEvUXUwBAKhXDnj17kMlksH59r+/Gueu7ezAxM4uNPZVLtE0jid61y+PeLLBr1yDWrFmDc+fOVT0uFIlWndt+x8svCPUqEaiAH//4x7j77rtx8OBBRw7uT/7kT1AoFPDVr37VcfynP/1pHDp0CN/73vccj+/duxd33XUXfu/3fg979uzBQw89hBtuuEE+/61vfQsPPPAAnntucReZYZhIp71L45aCXC6DRCKGdDrru9SwFig1HV2TfZ1XPouIhy6kXoTDkUV3pM4Xcgi7U0IchAA6zWJuLu89XpQiWGUnPpueh1lleja3dZQRFwHTND2FqfWg1muX8lsZhoF83jukrmkaCvkFTE3MwKiQTw6FQhV/K0pNgHuTeKHW92rp7EQkvLi2F2PDQygWvTccpmkyl+JFusxWO29d0xCPhbCQKXiSlEKxAMOj4q0eaERH52p/ehf5uYUiQuEAYtHlFVOy9y5g/NKliuMZCAbR3tHh+ZyuawgGgZmZhUWtYbMz08gsVF5bE6mkoyGhH5SMEhKNlcXR1a67WnN74tII8hXKpSmliMWjFdexQCCAzp4upNNZGIa/MaOUglK66LVo9OJ5GEbla7qtq8czQLBUVFunAKBkFJFoTHk+p+saEonoosYrkYjWHbHxFXkRKaCxsTEZmhf/3rSpvOtqZ2cnfvrTnzoeKxQKmJmZQXt7O1KpFGKxGMZcmoaxsTF0VLj46kWptPy7rUgkhng8jkLhlXn/xSLko7zaC5Qu/vvoWqjiawMBDY1NrShhYVHv39RWO8JQ7X2XQjBrv3ax700q/l6BgIaurlYkkq1XbH4t9nP9emAsFwIBDU1NcUxPL26OvVIQv/ErcU6aFkCHh9GgHdU+N5GII5czF3Vube2v3G46hHrGa3HXe6pl8fcTkRoxjMWNGbD4tai1s7ZI+JWZ95XXKQAIIVrzc5cyXvXAFx3cvHkzGhoa8Mwzz8jH0uk0Dh8+jN0e9fG7d+/G6OioI/z07LOsv8pVV10FQgh27dolHxN45pln8NrXvtbXF1FQUFBQUFD47YCvyEsoFMI73/lO3H///WhubsaqVavwuc99Dp2dnbj11lthGAampqbQ2NiISCSC7du3Y9euXfizP/szfPKTn0Qmk8HHP/5xvPGNb5SRlfe85z248847sXXrVlx//fV49NFHceTIEdx3332vyBdWUFBQUFBQWNnwnYi7++678Za3vAUf/ehH8fa3vx26ruPhhx9GMBjEyMgI9u3bhx/+8IcAWGnp3/7t36Knpwfvfve78ad/+qe4/vrrHSZ1+/btw2c+8xl85zvfwZve9CY8/fTTePDBBx3l1QoKCgoKCgoKAr4EuysFhmFiamqh9oE+8WrNr79aocbLH9R4+YcaM39Q4+Ufasz8YSnj1dwcr1uwqxozKigoKCgoKKwoKPKioKCgoKCgsKKgyIuCgoKCgoLCioIiLwoKCgoKCgorCoq8KCgoKCgoKKwoKPKioKCgoKCgsKKgyIuCgoKCgoLCioIiLwoKCgoKCgorCoq8KCgoKCgoKKwo/Ld02KWUwjRfma+l65rvNt+/zVDj5Q9qvPxDjZk/qPHyDzVm/rDY8dI0AkJIXcf+tyQvCgoKCgoKCv99odJGCgoKCgoKCisKirwoKCgoKCgorCgo8qKgoKCgoKCwoqDIi4KCgoKCgsKKgiIvCgoKCgoKCisKirwoKCgoKCgorCgo8qKgoKCgoKCwoqDIi4KCgoKCgsKKgiIvCgoKCgoKCisKirwoKCgoKCgorCgo8qKgoKCgoKCwoqDIi4KCgoKCgsKKgiIvCgoKCgoKCisKirzUAdM08aUvfQmve93rsGPHDrz//e/HhQsXrvRpvSrx93//9/ijP/ojx2NHjhzBO9/5TuzYsQM333wz/vmf//kKnd2rAzMzM/j4xz+O66+/Hrt27cLb3/52HDhwQD7/1FNP4c1vfjO2b9+O2267DU888cQVPNtXByYnJ/GhD30Ie/bswc6dO3HnnXfi1KlT8nk1xyrjzJkz2LlzJ37wgx/Ix9R4lePSpUvYtGlT2X9i3NSYleOxxx7D61//egwODuJ3fud38KMf/Ug+d/HiRdx1113YtWsX9u3bhy984QswDGP5Ppwq1MSXv/xles0119Cf//zn9MiRI/SOO+6gt956K83n81f61F5VeOSRR+jmzZvpO9/5TvnY1NQUveaaa+hHPvIRevLkSfr973+fDg4O0u9///tX8EyvLN7znvfQ22+/ne7fv5+ePn2afupTn6Lbtm2jp06doidPnqSDg4P0gQceoCdPnqT/+I//SLdu3UqffPLJK33aVxRve9vb6Fvf+lZ68OBBevLkSfrBD36Q7tu3j2YyGTXHqqBQKNA3v/nNtK+vjz766KOUUnVNVsJ//dd/0cHBQXrp0iU6NjYm/8tms2rMPPDYY4/RrVu30kceeYSeO3eOfuUrX6GbN2+mzz//PC0UCvTWW2+ld955Jz127Bj9yU9+Qq+++mr6xS9+cdk+X5GXGsjn83Tnzp30W9/6lnxsdnaWbtu2jT7++ONX8MxePRgdHaV33XUX3bFjB73tttsc5OXBBx+k+/bto8ViUT72+c9/nt56661X4lSvOM6ePUv7+vrogQMH5GOmadJbbrmFfuELX6Af+9jH6Fve8hbHa+655x56xx13XO5TfdVgZmaG3nPPPfTYsWPysSNHjtC+vj568OBBNceq4POf/zx917ve5SAvary88dBDD9E3vOENns+pMXPCNE1600030c9+9rOOx++44w764IMP0scff5wODAzQmZkZ+dx3v/tdumvXrmXb9Ku0UQ0cPXoUCwsL2Lt3r3wskUhg69at2L9//xU8s1cPXn75ZQSDQfz7v/87tm/f7njuwIEDuPrqqxEIBORje/bswdmzZzExMXG5T/WKo6mpCQ899BAGBwflY4QQEEKQTqdx4MABx1wD2Hg999xzoJRe7tN9VSCZTOLzn/88+vr6AABTU1P4xje+gc7OTmzYsEHNsQrYv38//uVf/gWf/exnHY+r8fLGsWPHsH79es/n1Jg5cebMGQwNDeENb3iD4/GHH34Yd911Fw4cOID+/n4kk0n53J49ezA/P48jR44syzko8lIDo6OjAICuri7H4+3t7fK533bcfPPN+PKXv4zVq1eXPTc6OorOzk7HY+3t7QCAkZGRy3J+ryYkEgnccMMNCIVC8rEf//jHOHfuHF73utdVHK9sNovp6enLfbqvOnzsYx/D3r178cQTT+C+++5DLBZTc8wD6XQaf/mXf4mPfvSjZWuXGi9vHD9+HFNTU3jHO96Ba6+9Fm9/+9vxy1/+EoAaMzfOnDkDAMhkMnjve9+LvXv34q1vfSt+9rOfAbg846XISw1ks1kAcNxsACAcDiOfz1+JU1pRyOVynmMHQI0fgOeffx4f+chHcOutt+LGG2/0HC/x70KhcCVO8VWFd7/73Xj00Udx++234wMf+ABefvllNcc88MlPfhI7d+4s2xkD6pr0QqlUwunTpzE7O4sPfvCDeOihh7Bjxw7ceeedeOqpp9SYuTA/Pw8AuPfee3H77bfja1/7Gq677jr88R//8WUbr0DtQ367EYlEALAbh/gbYD9ANBq9Uqe1YhCJRMpuumLyxmKxK3FKrxr89Kc/xV/8xV9g165duP/++wGwC9w9XuLfar4BGzZsAADcd999OHjwIB555BE1x1x47LHHcODAATz++OOez6vxKkcgEMAzzzwDXdflOj8wMIATJ07g4YcfVmPmQjAYBAC8973vxZve9CYAwJYtW3D48GF8/etfvyzjpSIvNSBCrmNjY47Hx8bG0NHRcSVOaUWhs7PTc+wA/FaP3yOPPIIPfvCDuOmmm/Dggw/KXUlXV5fneMViMTQ2Nl6JU73imJqawhNPPIFSqSQf0zQNGzZswNjYmJpjLjz66KOYnJzEjTfeiJ07d2Lnzp0AgE984hN43/vep8arAuLxuGODCgAbN27EpUuX1Ji5IL6z0KEJbNiwARcvXrws46XISw1s3rwZDQ0NeOaZZ+Rj6XQahw8fxu7du6/gma0M7N69G88995yjvv/pp5/GunXr0NLScgXP7Mrh29/+Nj796U/jHe94Bx544AFHePW1r30tnn32WcfxTz/9NHbt2gVN++28XCcmJnDPPffgqaeeko8Vi0UcPnwY69evV3PMhfvvvx8//OEP8dhjj8n/AODuu+/Gfffdp8bLAydOnMCuXbsc6zwAvPTSS9iwYYMaMxf6+/sRj8dx8OBBx+PHjx9Hb28vdu/ejcOHD8v0EsDGKx6PY/PmzctzEstSs/TfHA888AC9+uqr6U9/+lOHz0uhULjSp/aqw7333usolZ6YmKC7d++m9957Lz1x4gR99NFH6eDgIP3BD35wBc/yyuH06dO0v7+ffuADH3B4SYyNjdF0Ok2PHz9O+/v76ec+9zl68uRJ+vDDDyufF0rp+973PnrrrbfSZ599lh47dozec889dPfu3XRoaEjNsTpgL5VW41UOwzDo7//+79PXv/71dP/+/fTkyZP0M5/5DB0YGKDHjh1TY+aBv/u7v6M7d+6kjz/+uMPn5emnn6a5XI7ecsst9L3vfS89cuSI9Hn58pe/vGyfr8hLHSiVSvSv//qv6Z49e+iOHTvo+9//fnrhwoUrfVqvSrjJC6WUHjx4kP7BH/wBHRgYoDfddBP95je/eYXO7srjq1/9Ku3r6/P8795776WUUvqLX/yC3n777XRgYIDedttt9IknnrjCZ33lkU6n6Sc+8Ql63XXX0W3bttE77riDHj9+XD6v5lh12MkLpWq8vDA+Pk4//OEP0+uuu44ODg7St73tbXT//v3yeTVm5fja175Gb775Ztrf309/93d/l/7kJz+Rz509e5a+5z3voYODg3Tfvn30C1/4AjUMY9k+m1D6W2oeoaCgoKCgoLAi8duZRFdQUFBQUFBYsVDkRUFBQUFBQWFFQZEXBQUFBQUFhRUFRV4UFBQUFBQUVhQUeVFQUFBQUFBYUVDkRUFBQUFBQWFFQZEXBQUFBQUFhRUFRV4UFBQUFBQUVhQUeVFQUFBQUFBYUVDkRUFBQUFBQWFFQZEXBQUFBQUFhRWF/x/jG0n9wEhzKQAAAABJRU5ErkJggg==\n",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "for d in descr:\n",
    "    plt.plot(x,d)\n",
    "plt.plot(x,dataset.df_topics.topics.iloc[np.random.randint(len(dataset.df_topics))], color='k')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Using song embeddings\n",
    "\n",
    "We can do something similar with song embeddings: here are the song embeddings for a few covers of the same song:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 85,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "http://lyrics.wikia.com/David_Bowie:The_Man_Who_Sold_The_World\n",
      "http://lyrics.wikia.com/Lulu:The_Man_Who_Sold_The_World\n",
      "http://lyrics.wikia.com/Midge_Ure:The_Man_Who_Sold_The_World\n",
      "http://lyrics.wikia.com/Nirvana:The_Man_Who_Sold_The_World\n"
     ]
    }
   ],
   "source": [
    "descr = []\n",
    "for urlSong in list(dataset.df_song_embeddings[dataset.df_song_embeddings.index.str.contains(\"The_Man_Who_Sold\")].index):\n",
    "    print(urlSong)\n",
    "    descr.append(dataset.df_song_embeddings[dataset.df_song_embeddings.index==urlSong].song_embedding.iloc[0])"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "... and in the following cell, we add embeddings from 10 random songs and plot all the pairwise distances:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 86,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.image.AxesImage at 0x7fef3d53c6d0>"
      ]
     },
     "execution_count": 86,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "for i in range(10):\n",
    "    descr.append(dataset.df_song_embeddings.song_embedding.iloc[\n",
    "        np.random.randint(len(dataset.df_song_embeddings))])\n",
    "    \n",
    "descr = np.array(descr)\n",
    "\n",
    "dists = np.linalg.norm(descr[:, None, :] - descr[None, :, :], axis=-1)\n",
    "plt.imshow(dists)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Using verse embeddings\n",
    "\n",
    "With verse embeddings we can do something more, i.e. measure signals *within* each song. For instance, the following code loads lyrics embeddings from Guns'n'Roses' \"Paradise City\" and plots pairwise distances between each pair of verses (note that we'll have to download the lyrics file manually here as due to their size they were not included in the zipped version of the dataset)."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 87,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "--2023-07-21 12:33:49--  https://evalrs.object.lga1.coreweave.com/evalrs_verse_embeddings_G.parquet\n",
      "Resolving evalrs.object.lga1.coreweave.com (evalrs.object.lga1.coreweave.com)... 216.153.56.63\n",
      "Connecting to evalrs.object.lga1.coreweave.com (evalrs.object.lga1.coreweave.com)|216.153.56.63|:443... connected.\n",
      "HTTP request sent, awaiting response... 200 OK\n",
      "Length: 1284882026 (1,2G) [application/octet-stream]\n",
      "Saving to: ‘evalrs_verse_embeddings_G.parquet’\n",
      "\n",
      "evalrs_verse_embedd 100%[===================>]   1,20G  3,61MB/s    in 8m 40s  \n",
      "\n",
      "2023-07-21 12:42:29 (2,36 MB/s) - ‘evalrs_verse_embeddings_G.parquet’ saved [1284882026/1284882026]\n",
      "\n"
     ]
    }
   ],
   "source": [
    "!wget https://evalrs.object.lga1.coreweave.com/evalrs_verse_embeddings_G.parquet"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 92,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(58368,)"
      ]
     },
     "execution_count": 92,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "table = pq.read_table('evalrs_verse_embeddings_G.parquet')\n",
    "df = table.to_pandas()\n",
    "d = df[df.urlSong=='http://lyrics.wikia.com/Guns_N%27_Roses:Paradise_City'].verse_embeddings.iloc[0]\n",
    "d.shape"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**NOTE** that if you load embeddings for another song you will very likely have a different vector length! How so?\n",
    "\n",
    "For each song, we have a verse embedding for every single line (we used `\\n` as a separator). For parquet to properly store these embeddings we had to concatenate all of them into a single vector, so we have to reshape them to *n* embeddings of size 768. "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 94,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(76, 768)"
      ]
     },
     "execution_count": 95,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "descr = d.reshape((-1,768))\n",
    "descr.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 96,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.image.AxesImage at 0x7fef3f300d90>"
      ]
     },
     "execution_count": 96,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "dists = np.linalg.norm(descr[:, None, :] - descr[None, :, :], axis=-1)\n",
    "plt.imshow(dists)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "You can see the verse pairwise distances above, and compare them with the text of the song below:\n",
    "\n",
    "\n",
    "```\n",
    "Take me down to the paradise city\n",
    "Where the grass is green and the girls are pretty\n",
    "Take me home (Oh, won't you please take me home?)\n",
    "Take me down to the paradise city\n",
    "Where the grass is green and the girls are pretty\n",
    "Take me home (Oh, won't you please take me home?)\n",
    "Just an urchin livin' under the street\n",
    "I'm a hard case that's tough to beat\n",
    "I'm your charity case so buy me somethin' to eat\n",
    "I'll pay you at another time\n",
    "Take it to the end of the line\n",
    "Rags to riches or so they say\n",
    "You gotta keep pushin' for the fortune and fame\n",
    "You know it's, it's all a gamble when it's just a game\n",
    "You treat it like a capital crime\n",
    "Everybody's doin' their time\n",
    "Take me down to the paradise city\n",
    "Where the grass is green and the girls are pretty\n",
    "Oh, won't you please take me home, yeah, yeah?\n",
    "Take me down to the paradise city\n",
    "Where the grass is green and the girls are pretty\n",
    "Take me home\n",
    "Strapped in the chair of the city's gas chamber\n",
    "Why I'm here, I can't quite remember\n",
    "The surgeon general says it's hazardous to breathe\n",
    "I'd have another cigarette but I can't see\n",
    "Tell me who ya gonna believe\n",
    "Take me down to the paradise city\n",
    "Where the grass is green and the girls are pretty\n",
    "Take me home, yeah, yeah\n",
    "Take me down to the paradise city\n",
    "Where the grass is green and the girls are pretty\n",
    "Oh, won't you please take me home, yeah?\n",
    "So far away\n",
    "So far away\n",
    "So far away\n",
    "So far away\n",
    "Captain America's been torn apart\n",
    "Now he's a court jester with a broken heart\n",
    "He said \"Turn me around and take me back to the start\"\n",
    "I must be losin' my mind, are you blind?\n",
    "I've seen it all a million times\n",
    "Take me down to the paradise city\n",
    "Where the grass is green and the girls are pretty\n",
    "Take me home, yeah, yeah\n",
    "Take me down to the paradise city\n",
    "Where the grass is green and the girls are pretty\n",
    "Oh, won't you please take me home?\n",
    "Take me down to the paradise city\n",
    "Where the grass is green and the girls are pretty\n",
    "Take me home, yeah, yeah\n",
    "Take me down to the paradise city\n",
    "Where the grass is green and the girls are pretty\n",
    "Oh, won't you please take me home, home\n",
    "Oh, I want to go, I want to know\n",
    "Oh, won't you please take me home?\n",
    "I want to see how good it can be\n",
    "Oh, won't you please take me home?\n",
    "Take me down to the paradise city\n",
    "Where the grass is green and the girls are pretty\n",
    "Take me home\n",
    "Take me down to the paradise city\n",
    "Where the grass is green and the girls are pretty\n",
    "Oh, won't you please take me home?\n",
    "Take me down, take me down\n",
    "Oh, won't you please take me home?\n",
    "I want to see how good it can be\n",
    "Oh, won't you please take me home?\n",
    "I want to see how good it can be\n",
    "Oh, oh take me home\n",
    "Take me down to the paradise city\n",
    "Where the grass is green and the girls are pretty\n",
    "Oh, won't you please take me home?\n",
    "I want to know, I want to know\n",
    "Oh, won't you please take me home?\n",
    "Yeah, baby\n",
    "```"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 97,
   "metadata": {},
   "outputs": [],
   "source": [
    "del train_tracks\n",
    "del train_topics\n",
    "del train_embs\n",
    "del table\n",
    "del df\n",
    "del descr\n",
    "del dists\n",
    "\n",
    "_ = gc.collect()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## EDA on Users \n",
    "\n",
    "Let's move to users now."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {},
   "outputs": [],
   "source": [
    "gb_user = train.groupby(\"user_id\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {},
   "outputs": [],
   "source": [
    "user_records = gb_user.size()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "count    29717.000000\n",
       "mean       230.488508\n",
       "std        108.455895\n",
       "min          9.000000\n",
       "0%           9.000000\n",
       "10%         75.000000\n",
       "20%        126.000000\n",
       "30%        168.000000\n",
       "40%        204.000000\n",
       "50%        237.000000\n",
       "60%        268.000000\n",
       "70%        299.000000\n",
       "80%        332.000000\n",
       "90%        372.000000\n",
       "max        492.000000\n",
       "dtype: float64"
      ]
     },
     "execution_count": 21,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "user_records.describe(percentiles=percentiles)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "There are almost 30K distinct users in our dataset. The average number of songs they played is 232 with some binge listener that made it to 494.\n",
    "\n",
    "What about their demographics?"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {},
   "outputs": [],
   "source": [
    "train_users = train.join(dataset.df_users, on=\"user_id\", rsuffix=\"_data\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "metadata": {},
   "outputs": [],
   "source": [
    "gb_train_users = train_users.groupby(\"user_id\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "gender\n",
       "n       0.533028\n",
       "m       0.331359\n",
       "f       0.133795\n",
       "None    0.001817\n",
       "Name: proportion, dtype: float64"
      ]
     },
     "execution_count": 24,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "gb_train_users.first().gender.value_counts(normalize=True, dropna=False)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Only half of our users are annotated with (binary) gender, and there is a significant skew towards males."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "country\n",
       "None    0.540835\n",
       "US      0.084194\n",
       "RU      0.041390\n",
       "DE      0.039338\n",
       "UK      0.037992\n",
       "          ...   \n",
       "AS      0.000034\n",
       "KI      0.000034\n",
       "TD      0.000034\n",
       "GI      0.000034\n",
       "KY      0.000034\n",
       "Name: proportion, Length: 160, dtype: float64"
      ]
     },
     "execution_count": 25,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "users_by_country = gb_train_users.first().country.value_counts(normalize=True, sort=True, ascending=False, dropna=False)\n",
    "users_by_country"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "country\n",
       "US    2502\n",
       "RU    1230\n",
       "DE    1169\n",
       "UK    1129\n",
       "PL    1044\n",
       "BR     965\n",
       "NL     378\n",
       "FI     332\n",
       "ES     304\n",
       "SE     293\n",
       "Name: count, dtype: int64"
      ]
     },
     "execution_count": 26,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "top_countries = gb_train_users.first().country.value_counts(dropna=True, sort=True, ascending=False)[:10]\n",
    "top_countries"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "fig, ax = plt.subplots(figsize=(12,8))\n",
    "ax.bar(x=top_countries.index, height=top_countries.values)\n",
    "ax.set_xlabel(\"Country\")\n",
    "ax.set_ylabel(\"Count\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "There are 158 distinct countries but only half of the users are annotated with that information. Most of the users were logged from the United States."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "metadata": {},
   "outputs": [],
   "source": [
    "del gb_user\n",
    "del gb_train_users\n",
    "del top_countries\n",
    "_ = gc.collect()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Analysis on listening events\n",
    "\n",
    "We have explored relationships between users and tracks. But what about the playing time and hour? \n",
    "\n",
    "For example, one might search for any correlation between listening events and hour of the day. "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**Q0: Is there any listening pattern across hours of the day?**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Text(0, 0.5, 'Count')"
      ]
     },
     "execution_count": 28,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1200x800 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "fig, ax = plt.subplots(figsize=(12,8))\n",
    "ax = sns.countplot(data=train_users, x=\"hod\")\n",
    "ax.set_xlabel(\"Hour of the Day\")\n",
    "ax.set_ylabel(\"Count\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Remember that the `timestamp` attribute is in Unix time, i.e. it is absolute in terms of timezone. The chart is showing us that events are mainly located in a single timezone, as the events decrease in the morning and increase in late afternoon.   "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**Q1: Is there any listening pattern across hours of the day, separately by gender?**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Text(0, 0.5, 'Count')"
      ]
     },
     "execution_count": 29,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAABAoAAAKrCAYAAAB4NG9dAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjYuMywgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy/P9b71AAAACXBIWXMAAA9hAAAPYQGoP6dpAABhC0lEQVR4nO3de5yMdeP/8fdcs9buYg/EOvRNUkjROqy45dBW7u6iO6RSqERulI6k6KASdyg5ZR0KRVSk0kl136FuZN2l7pAzWXY37MGyBztz/f7w26u2tXZ2zcw1s/t6Ph77eNhrPnPN2xhzeM/n+lwO0zRNAQAAAAAASDLsDgAAAAAAAAIHRQEAAAAAALBQFAAAAAAAAAtFAQAAAAAAsFAUAAAAAAAAC0UBAAAAAACwUBQAAAAAAAALRQEAAAAAALCE2B2gMjNNU263aXcMAAAAAEAlYBgOORyOUsdRFNjI7TZ17NgJu2MAAAAAACqBmjWryeksvSjg0AMAAAAAAGChKAAAAAAAABaKAgAAAAAAYKEoAAAAAAAAFooCAAAAAABg4awHAAAAAICgd/r082653S67o9jC6QyRYXhnLgBFAQAAAAAgaJmmqZycbGVnZ1bakqBQeHh1RUbWlMNR+ikQz4aiAAAAAAAQtLKyjiknJ1thYdUUFhYhw3Ce8wflYGOapvLz85SdnS5JioqqdU77oygAAAAAAAQlt9ulnJwTql49WtWrR9kdx1ahoVUlSdnZ6apRI+acDkNgMUMAAAAAQFByuVySTFWtGmZ3lIBQWBa4XAXntB+KAgAAAABAkKtchxqUxFuHXFAUAAAAAAAAC0UBAAAAAACwUBQAAAAAAOBj48c/q1tu6WF3DI9QFAAAAAAAAAtFAQAAAAAAsITYHQAAAAAAAF8rKCjQ3LmvafXqT5WZmam4uNbq1u16vfDCM3r33Q9Vr159bdnyvebOfU3btv2s0NCq6tixk4YPf0gxMTGSpE8++Uj//OcLmjVrvqZNm6KdO39RTExN9e59m+64o791W1lZWZox4xWtW7dGpmnqppt6yu12F8u0bt3XWrBgvvbu3a3q1WsoIeE6DRkyXOHh4ZKk+fMTtXr1p7r++hv1zjtvKzS0it58811FRkb69L6iKAAAAAAAVHgvvTReX375uQYOvE+XXNJUX3zxmV56abx1+Q8//FcPPTRMbdq003PPTVRWVqbmzZutESOGaN68RapaNUyS5Ha79fTTo3XbbXfqvvuGadWqDzRr1qtq3PhiXXllB7ndbj366ANKSTms++9/SFFRUVq8eJG2bftZ551X27q91as/03PPjVW3bn/T4MFDlZJySImJs7R37x5NnTrTOtVhSsph/ec/3+i5515UZmamz0sCiaIAAAAAAFDBJScf1KefrtLw4Q/q9tv7SZKuvLKDjh07pu++Wy9JSkycoQsuaKiXXnpFTqdTknTZZS3Uv/+tWrXqQ/XufaskyTRN3XPPIHXvfrMkqUWLK7Rmzb/1n/+s05VXdtCGDf/Rtm0/a/LkaWrf/i+SpDZt2qlPn98XMjRNU7NnT9eVV/5FTz/9vLX9/PMv0EMPDdP69d/qL3+5SpLkcrl0//0P64or4nx6H/0RaxQAAAAAACq0//43SaZp6uqrry2y/brr/ipJys3N1c8//08dOlwl0zRVUFCggoIC1a/fQA0bXqikpI1FrnfZZS2tP4eGhio6Olo5OTmSpC1bvleVKlV05ZUdrDHh4eFq376j9fuBA/uVlpaqq67qbN1WQUGB4uJaq1q1atq0qejtXXJJE+/cER5iRgEAAAAAoELLyEiXJMXE1CyyvfD348ez5Ha7tXjxQi1evLDY9atWrVrk97CwsCK/G4Yh0zQlnV6fIDIy0jp0oFCtWudZf87MzJAkTZkyUVOmTCx2e0eO/Fbk94iIiBL/br5AUQAAAAAAqNBq164jSTp27Jjq1q1rbU9PPyZJqlatuhwOh2699Q5rlsEfFa5P4Ino6GhlZGTI5XJZhzBIUlZWpvXn6tVrSJKGDXtQrVu3KbaPGjV8vw7B2XDoAQAAAACgQmvR4go5nU6tXfvvItsLf4+IiFCTJs104MA+NWvW3Ppp1OgizZ+fqO+/3+zxbbVpEy+Xy6V16762tp06dUrffbfB+r1hwwsVE1NThw8fKnJ7551XR6+9NkM7dmw/l7/uOWNGAQAAAACgQmvQ4HzdcMNNSkycoYKCU7r44iZau/bf+vbbdZIkh8PQkCHDNXLkgxo3bqy6dbteLpdbS5e+pa1b/6e77hrk8W21bdtO7dp10MSJLyg9PV1169bVu+8uU0ZGunWog9Pp1H33DdOkSS/K6TTUsWMnHT+erYUL5yktLU1Nm17qk/vBUxQFAAAAAIAK7+GHRyoiIlxvv/2WTpw4obZt43XXXffqjTfmKiIiXO3atdeUKdP1xhtzNXbs46pSpYqaNr1Ur7wyS5df3qJMt/Xii5P02mvTNH/+bOXl5euaa67TTTf1KjLLoEePmxURUU1LlizShx++r/DwcLVocYWefvoF1a/fwKt/97JymIUrLsDvXC63jh07YXcMAAAAAAhKp07l6+jRw6pVq56qVAktcVxWVqY2bPiPrryyg6Kioq3tM2e+qo8//lCffPKVH9L6Xmn3R82a1eR0lr4CATMKAAAAAAAVWtWqYZo6dbIuuaSpbr21r8LDw/Xzzz9p+fJl6tfvbrvjBRxmFNiIGQUAAABA2RmGQ4bhKHWc223K7ebjTkXm6YwCSdq58xfNnfuafv75J+Xk5KpBgwa6+ebe6tXr1mKnMgxW3ppRQFFgI4oCAAAAoGwMw6HomHA5DWepY11ulzLScygLKrCyFAWVAYceAAAAAKh0DMMhp+FU4ppFOpSZWuK4+lGxGtJlgAzDQVEAlBFFAQAAAICgcygzVfuPHrQ7BlAhlT7nAAAAAAAAVBrMKAAAAAAAL2GhRVQEFAUAAAAA4AUstIiKgqIAAAAAALyAhRYDj6czPLwt2GeMUBQAAAAAgBex0GJgMAyHoqMjPDodoLe5XG5lZJwM2rKAogAAAAAAUOEYhkNOp6GZb3+r5LRMv91ugzpRGt63Y1DPGKEoAAAAAABUWMlpmdqXnG53jKBCUQAAAAAAgM2uuqqtRo58Up9//om2b9+qevXqa/Top7Rnz24tXDhf2dnZat/+Lxoz5hlVrRrm0yz+P1gDAAAAAAAUM3fuLN1xxwAtWLBE1apV16hRD+vrr7/S5Mmv6sknn9a6dV/ro49W+jwHRQEAAAAAAAHghhtu0lVXddYFF1yov/71Bh0/nqVHHnlcF110sbp2vUYXX9xEe/bs9nkODj0AAAAAUIynp5UL9tPAAYHk/PP/z/pzeHi4JKlBg/OtbVWrVtWpU6d8noOiAAAAAEARhuFQdEy4nIaz1LEut0sZ6TmUBYAXhIQU/4huGP4/EICiAAAAAEARhuGQ03Aqcc0iHcpMLXFc/ahYDekyIKhPAwegOIoCAAAAAGd0KDNV+48etDsGAD+jKAAAAAAAVFgN6kRV6NvzBYoCAAAAAECF43abcrncGt63o99v2+Vyl/lwnG++SSry+w039NANN/Qosm3GjDnnnM0TFAUAAAAAgArH7TaVkXHSo7N3+OK2g3ndDooCAAAAAECFFOwf2O3i//MsAAAAAACAgEVRAAAAAAAALBQFAAAAAADAQlEAAAAAAAAsLGYIAAAA2MQwHB6vyM6ibAD8haIAAAAAsIFhOBQdEy6n4fRovMvtUkZ6DmUByoVSCmVBUQAAAADYwDAcchpOJa5ZpEOZqWcdWz8qVkO6DJBhOPgAhzKjlEJZURQAAAAANjqUmar9Rw/aHQMVWGUupcoyk8Kbgn1WBkUBAAAAAFQCla2UMgyHYmLCZXg4k8Kb3G6X0ss4K2P79q16/vmndehQsnr3vk333/+Q7wKWgqIAAAAAAFDhnJ5N4NTeVXOVc/Sw3243vFY9Neo+uMyzMhYtekMhIVX01lvvqnr16j5MWDqKAgAAAABAhZVz9LByUg/YHaNUx49n6ZJLmqhBg/PtjiLD7gAAAAAAAFRmt9zSQ99/v1mfffaxrrqqrQ4fPmRrHooCAAAAAABsNHfuIl1+eUslJFynDz74THXqxNqah0MPAAAAAACwUUxMjEJCQlS1alXVqnWe3XGYUQAAAAAAAH5HUQAAAAAAACwUBQAAAAAAwMIaBQAAAACACiu8Vr0KfXu+QFEAAAAAAKhw3G5TbrdLjboPtuG2XXK7Tb/frrdQFAAAAAAAKhy321R6eo4Mw2HLbZe1KJgxY46P0pQdRQEAAAAAoEIqzwd2sJghAAAAAAD4A4oCAAAAAABgoSgAAAAAAAAWigIAAAAAAGChKAAAAAAAABaKAgAAAAAAYKEoAAAAAAAAlhC7AwAAAAAA4AuG4ZBhOPx+u263Kbfb9PvtegtFAQAAAACgwjEMh6JjwuU0nH6/bZfbpYz0nKAtCygKAAAAAAAVjmE45DScSlyzSIcyU/12u/WjYjWkywAZhoOiAAAAAACAQHMoM1X7jx60O0ZQYTFDAAAAAABgYUYBAAAAKjxPFzQL9gXIAASvq65qq9Gjn9IXX3yun37aoho1quvmm2/RPfcM9nsWigIAAABUaGVZ0CzYFyADENxmzJiqhx8eqccfH6Mvv/xcc+bMUqtWbRQX19qvOSgKAAAAUKF5uqBZRViADEBw+9vfuuuvf71BkjRgwEAtWfKmfvppC0UBAAAA4AssaAYg0DVseGGR36tXr65Tp075PQeLGQIAAAAAEABCQ0OLbTNN/89woigAAAAAAAAWDj0AAAAAAFRY9aNiK/Tt+QJFAQAAAACgwnG7TbncLg3pMsDvt+1yu4J6UVTbi4KMjAy9/PLL+vrrr5Wdna2mTZvq0UcfVdu2bSVJ69ev16RJk7R7927Vq1dPDzzwgG688Ubr+nl5eZo4caI+++wz5ebmKiEhQWPGjFHNmjWtMf7YBwAAAAAgcLjdpjLSc2QYDltuu6xFwTffJBXb9t57H3krUpnYvkbBI488ou+//14vv/yyli9frksvvVT33nuv9uzZo927d2vIkCHq1KmTVqxYoT59+mjUqFFav369df1nn31W33zzjaZPn66FCxdqz549GjFihHW5v/YBAAAAAAgsbrepggK333+CeTaBZPOMgv379+vbb7/VkiVL1KZNG0nSU089pXXr1umjjz7S0aNH1bRpUz388MOSpMaNG2vr1q2aN2+eOnTooNTUVK1cuVKzZ8+2ZiC8/PLLuv766/X999+rVatWWrhwoc/3AQAAAABARWFrURATE6M5c+aoRYsW1jaHwyGHw6GsrCwlJSXp2muvLXKd9u3ba/z48TJNU5s3b7a2FWrUqJFiY2O1adMmtWrVyi/7cDjKP5UlJMT2SR0AAAAVmtNZtvdbZR1fXuW5nUDN5q9c5bktsgX2Y+1cud3+P6wgGDidjnP6rGlrURAZGakuXboU2fb5559r//79evLJJ/X++++rbt26RS6vU6eOcnJylJ6ertTUVMXExKhq1arFxqSkpEiSUlJSfL6PP65lUBaG4VBMTLVyXRcAAAC+ERkZbneEEgVqtkDNJZGtvAI52x/l5jp15Ihxzh+MKwq32yHDMBQVFaGwsLBy78f2xQz/6L///a+eeOIJdevWTV27dlVubq5CQ0OLjCn8PT8/Xzk5OcUul6SqVasqLy9Pkvyyj/Jyu01lZZ0s9/UBAABQOqfTKNOHnqysHLlcbh8mOq2suaTAzeavXBLZyiOQH2vnKj8/X263Wy7X6bUBKrvT6yO4lZmZo5wcV7HLIyPDPZotEjBFwZdffqnHHntMrVu31uTJkyWd/rD+5w/ihb+Hh4crLCzsjB/U8/LyFB4e7rd9nAsezAAAAIElkD9wBGq2QM0lka28AjlbUQ5JDuXl5apKlaqljq7o8vPz/v+fjHP69wuIouCtt97S+PHjdf311+uf//yn9W19vXr1lJaWVmRsWlqaIiIiVKNGDdWtW1cZGRnKz88v8o1/WlqaYmNj/bYPAAAAAID/GYZT4eHVlJ2doYKCUwoLi5BhOM9pHblgZJqm8vPzlJ2drvDw6jKMczsMw/aiYMmSJXr++efVv39/jRkzpsg/aNu2bfXdd98VGb9hwwa1bt1ahmGoTZs2crvd2rx5s3X2gb179yo1NVXx8fF+2wcAAAAAwB6RkTVVpUpVZWdnKDf3hN1xbBUeXl2RkeVbQ++PbC0K9u7dqxdffFHXXXedhgwZoiNHjliXhYWFqX///urZs6cmT56snj17as2aNfrss880b948SVJsbKxuvPFGjR07Vi+++KLCw8P1zDPPqF27doqLi5Mkv+wDAAAAAGAPh8OhiIjqCg+vJrfbLbe7+LH5lYHTGeK1L7JtLQo+//xznTp1Sl988YW++OKLIpf17NlTEydO1KxZszRp0iQtXLhQ559/viZNmmR98y9Jzz//vF588UXdf//9kqTOnTtr7Nix1uWXXHKJX/YBAAAAALCPw+GQ0+mU0+m0O0rQs7Uo+Mc//qF//OMfZx3TuXNnde7cucTLIyIi9MILL+iFF16wdR8AAAAAAFQEHGAPAAAAAAAsFAUAAAAAAMBCUQAAAAAAACwUBQAAAAAAwEJRAAAAAAAALBQFAAAAAADAQlEAAAAAAAAsIXYHAAAAAABUXobhkGE4Sh3ndptyu00/JAJFAQAAAADAFobhUHRMuJyGs9SxLrdLGek5lAV+QFEAAAAAALCFYTjkNJxKXLNIhzJTSxxXPypWQ7oMkGE4KAr8gKIAAAAAAGCrQ5mp2n/0oN0x8P+xmCEAAAAAALBQFAAAAAAAAAtFAQAAAAAAsFAUAAAAAAAAC0UBAAAAAACwUBQAAAAAAAALRQEAAAAAALBQFAAAAAAAAAtFAQAAAAAAsFAUAAAAAAAAC0UBAAAAAACwUBQAAAAAAAALRQEAAAAAALBQFAAAAAAAAAtFAQAAAAAAsFAUAAAAAAAAC0UBAAAAAACwUBQAAAAAAAALRQEAAAAAALCE2B0AAAAAFYNhOGQYjlLHud2m3G7TD4kAAOVBUQAAAIBzZhgORceEy2k4Sx3rcruUkZ5DWQAAAYqiAAAAAOfMMBxyGk4lrlmkQ5mpJY6rHxWrIV0GyDAcFAUAEKAoCgAAAOA1hzJTtf/oQbtjAADOAYsZAgAAAAAAC0UBAAAAAACwUBQAAAAAAAALRQEAAAAAALBQFAAAAAAAAAtFAQAAAAAAsFAUAAAAAAAAC0UBAAAAAACwUBQAAAAAAAALRQEAAAAAALBQFAAAAAAAAAtFAQAAAAAAsFAUAAAAAAAAC0UBAAAAAACwUBQAAAAAAAALRQEAAAAAALBQFAAAAAAAAAtFAQAAAAAAsFAUAAAAAAAAC0UBAAAAAACwUBQAAAAAAAALRQEAAAAAALBQFAAAAAAAAAtFAQAAAAAAsFAUAAAAAAAAC0UBAAAAAACwUBQAAAAAAAALRQEAAAAAALBQFAAAAAAAAAtFAQAAAAAAsFAUAAAAAAAAC0UBAAAAAACwUBQAAAAAAAALRQEAAAAAALBQFAAAAAAAAAtFAQAAAAAAsFAUAAAAAAAAS4jdAQAAAAAACESG4ZBhOEod53abcrtNPyTyD4oCAAAAAAD+xDAcio4Jl9NwljrW5XYpIz2nwpQFFAUAAAAAAPyJYTjkNJxKXLNIhzJTSxxXPypWQ7oMkGE4KAoAAAAAAKjoDmWmav/Rg3bH8CsWMwQAAAAAABaKAgAAAAAAYKEoAAAAAAAAFooCAAAAAABgoSgAAAAAAAAWigIAAAAAAGChKAAAAAAAABaKAgAAAAAAYKEoAAAAAAAAFooCAAAAAABgCbE7AAAAADxnGA4ZhqPUcW63Kbfb9EMiAEBFQ1EAAAAQJAzDoeiYcDkNZ6ljXW6XMtJzKAsAAGVGUQAAABAkDMMhp+FU4ppFOpSZWuK4+lGxGtJlgAzDQVEAACgzigIAAIAgcygzVfuPHrQ7BgCggmIxQwAAAAAAYKEoAAAAAAAAFooCAAAAAABgoSgAAAAAAAAWigIAAAAAAGChKAAAAAAAABaKAgAAAAAAYKEoAAAAAAAAFooCAAAAAABgoSgAAAAAAAAWigIAAAAAAGChKAAAAAAAABaKAgAAAAAAYKEoAAAAAAAAFooCAAAAAABgoSgAAAAAAAAWigIAAAAAAGChKAAAAAAAABaKAgAAAAAAYKEoAAAAAAAAFooCAAAAAABgoSgAAAAAAAAWigIAAAAAAGChKAAAAAAAABaKAgAAAAAAYKEoAAAAAAAAFooCAAAAAABgoSgAAAAAAAAWigIAAAAAAGAJqKIgMTFR/fv3L7Jt7Nixatq0aZGfhIQE63K3261p06apU6dOiouL0+DBg/Xrr78W2ce2bdvUr18/xcXFKSEhQYsWLSpyuTf2AQAAAABARRAwRcHixYs1derUYtt/+eUX/eMf/9A333xj/bz33nvW5bNmzdKSJUv0/PPPa+nSpXK73Ro0aJDy8/MlSenp6brnnnt0wQUXaPny5Ro+fLgmT56s5cuXe3UfAAAAAABUBCF2B0hNTdUzzzyjjRs36sILLyxymWma2rVrl+677z7Vrl272HXz8/P1+uuv67HHHlPXrl0lSa+88oo6deqk1atXq3v37nrnnXdUpUoVPffccwoJCVHjxo21f/9+zZkzR7179/bKPgAAAAAAqChsLwp+/vlnValSRR9++KFmzpyp5ORk67IDBw7o5MmTuuiii8543e3bt+vEiRPq0KGDtS0yMlLNmzfXpk2b1L17dyUlJaldu3YKCfn9r9q+fXslJibqyJEjOnTo0Dnv47zzziv33z8kJGAmdQAAgADndJbtfUNZx58LspVdeW4nULPx71m+2wrUf8/yXqc8AvU+K89t+TObr9leFCQkJBRZc+CPduzYIUl68803tXbtWhmGoc6dO+vhhx9WjRo1lJKSIkmqV69ekevVqVPHuiwlJUVNmjQpdrkkHT582Cv7KG9RYBgOxcRUK9d1AQAAShMZGW53hBKRrXwCNVug5pLIVl6Bmi1Qc0mBna2sbC8KzmbHjh0yDEN16tTR7NmzdeDAAb300kvauXOnFi5cqJycHElSaGhoketVrVpVmZmZkqTc3NwzXi5JeXl5XtlHebndprKyTpb7+gAAoHJxOo0yvRHNysqRy+X2YaLfka3syppLCtxs/HueFqjZeKyVTyBnK6/IyHCPZj4EdFEwdOhQ3XHHHYqJiZEkNWnSRLVr19att96qn376SWFhYZJOr1VQ+Gfp9If38PDT/6BhYWHWooR/vFySIiIivLKPc1FQENgPJAAAELxcLnfAvtcgW/kEarZAzSWRrbwCNVug5pICO1tZBfRBFIZhWCVBoUsuuUTS6cMBCg8XSEtLKzImLS1NsbGxkqS6deue8XJJio2N9co+AAAAAACoKAK6KBg1apTuvvvuItt++uknSdLFF1+sZs2aqXr16tq4caN1eVZWlrZu3ar4+HhJUnx8vDZv3iyXy2WN2bBhgxo1aqRatWp5ZR8AAAAAAFQUAV0U/PWvf9X69es1Y8YMHThwQGvWrNGTTz6p7t27q3HjxgoNDVW/fv00efJkffXVV9q+fbsefvhh1a1bV926dZMk9e7dW9nZ2RozZox27dqlFStWaMGCBRoyZIgkeWUfAAAAAABUFAG9RsE111yjqVOnas6cOZo7d65q1KihHj166KGHHrLGjBgxQgUFBRo7dqxyc3MVHx+v+fPnq0qVKpKkWrVqad68eRo/frx69uyp2rVra9SoUerZs6dX9wEAAAAAQEUQUEXBxIkTi23729/+pr/97W8lXsfpdGrkyJEaOXJkiWNatmypZcuW+XQfAACg4jAMhwzD4dFYt9uU2236OBEAAP4TUEUBAACA3QzDoeiYcDkNp0fjXW6XMtJzKAsAABUGRQEAAMAfGIZDTsOpxDWLdCgz9axj60fFakiXATIMB0UBAMCvPJ39Vp6ZbxQFAAAAZ3AoM1X7jx60OwYAAMWUZfZbeWa+URQAAAAAABBEPJ39Vt6ZbxQFAAAAAAAEIV/NfjO8vkcAAAAAABC0KAoAAAAAAICFogAAAAAAAFgoCgAAAAAAgIWiAAAAAAAAWCgKAAAAAACAhaIAAAAAAABYKAoAAAAAAICFogAAAAAAAFgoCgAAAAAAgIWiAAAAAAAAWCgKAAAAAACAhaIAAAAAAABYKAoAAAAAAICFogAAAAAAAFgoCgAAAAAAgIWiAAAAAAAAWCgKAAAAAACAJcTuAAAAoPIyDIcMw1HqOLfblNtt+iERAACgKAAAALYwDIeiY8LlNJyljnW5XcpIz6EsAADADygKAACALQzDIafhVOKaRTqUmVriuPpRsRrSZYAMw0FRAACAH1AUAAAAWx3KTNX+owftjgHARzjECAg+FAUAAAAAfMIwHIqOjpDTWfoa6i6XWxkZJykLgABAUQAAAADAJwzDIafT0My3v1VyWmaJ4xrUidLwvh05xAgIEBQFAAAAAHwqOS1T+5LT7Y5RDIdFAGdGUQAAAACg0uGwCKBkFAUAAABAkPP0m3GJb8cLcVgEUDKKAgAAACCIleWbcYlvx/8sUA+LAOxEUQAAAAAEMU+/GZf4dhyAZygKAAAAAA8F8uJ3fDMOwFsoCgAAAAAPsPgdgMqCogAAAADwAIvfAagsKAoAAACAMmCKP4CKzrOlUQEAAAAAQKVAUQAAAAAAACwUBQAAAAAAwEJRAAAAAAAALBQFAAAAAADAwlkPAAAAEFAMwyHDcJQ6zu02Of0gAPgARQEAAAAChmE4FB0dIaez9ImvLpdbGRknKQsAwMsoCgAAABAwDMMhp9PQzLe/VXJaZonjGtSJ0vC+HWUYDooCVEjMrIGdKAoqME+fXCSeYACgIuPNJoJRclqm9iWn2x0DsAUza2A3ioIKyjAcio4Jl9NwejTe5XYpIz2HJxgAqGDK8nrAawEABAZm1sBuFAUVlGE45DScSlyzSIcyU886tn5UrIZ0GcATDABUQJ6+HvBaAACBh5k1sAtFQQV3KDNV+48etDsGAMBmvB4AAABPlX7QCwAAAAAAqDQoCgAAAAAAgIWiAAAAAAAAWCgKAAAAAACAxSdFQUpKii92CwAAAAAAfKxcRcGll16qH3/88YyXJSUl6W9/+9s5hQIAAAAAAPbw+PSIr7/+uk6ePClJMk1T7777rtauXVts3Pfff6/Q0FDvJQQAAAAAAH7jcVGQl5enGTNmSJIcDofefffdYmMMw1CNGjU0dOhQ7yUEAACA1xmGQ4bhKHWc223K7Tb9kAgAECg8LgqGDh1qFQDNmjXTO++8o5YtW/osGAAAAHzDMByKjo6Q01n6Uagul1sZGScpCwCgEvG4KPij7du3ezsHAAAA/MQwHHI6Dc18+1slp2WWOK5BnSgN79tRhuGgKACASqRcRYEkffvtt/r3v/+tnJwcud3uIpc5HA69+OKL5xwOAAAAvpOclql9yel2xwAABJhyFQWvv/66XnrpJVWtWlU1a9aUw1H0+LY//w4AAAAAAIJDuYqCt956Sz169ND48eM5wwEAAAAAABVI6SvYnMGRI0d0yy23UBIAAAAAAFDBlKsoaN68uXbu3OntLAAAAAAAwGblOvTgySef1EMPPaSIiAhdccUVCg8PLzamfv365xwOAAAAAAD4V7mKgr59+8rtduvJJ58sceHCbdu2nVMwAAAAAADgf+UqCp5//nnObAAAAAAAQAVUrqKgV69e3s4BAAAAAAACQLmKgk2bNpU6Jj4+vjy7BgAAAAAANipXUdC/f385HA6Zpmlt+/OhCKxRAAAAAABA8ClXUbBo0aJi206ePKmkpCR98MEHmj59+jkHAwAAAAAA/leuoqBdu3Zn3N61a1dFRETotddeU2Ji4jkFAwAAAAAA/md4e4dt27bVd9995+3dooIxDIdCQoxSfwyDs2sAAAAAgD+Va0bB2fzrX/9StWrVvL1bVCCG4VB0TLichrPUsS63SxnpOXK7zVLHAgAAAADOXbmKggEDBhTb5na7lZKSouTkZA0ePPicg6HiMgyHnIZTiWsW6VBmaonj6kfFakiXATIMB0UBAAAAAPhJuYqCP57toJBhGGrSpImGDBmi3r17n3MwVHyHMlO1/+hBu2MAAAAAAP6gXEXBm2++6e0cAAAAAAAgAJzTGgVr167Vd999p6ysLNWsWVNt2rRRp06dvJUNAAAAABBADMPh8YLjbrfJIcRBqlxFQX5+voYNG6ZvvvlGTqdTMTExSk9PV2Jiotq3b6/ExESFhoZ6OysAAAAAwCaG4VB0dIScTs9OnudyuZWRcZKyIAiVqyiYPn26Nm/erJdeekk33nijnE6nCgoKtGrVKo0bN06vvfaaHnzwQW9nBQAAAADYxDAccjoNzXz7WyWnZZ51bIM6URretyMLkwepchUFq1at0v3336+bbrrp9x2FhOjmm2/W0aNH9fbbb1MUAAAAAEAFlJyWqX3J6XbHgA+Vqyg4duyYmjdvfsbLmjdvrtTUkk95BwBAReTpMZscrwkAAAJduYqCCy64QJs3b1aHDh2KXbZp0ybVq1fvnIMBABAsDMOh6JhwOQ1nqWNdbpcy0nMoCwAAQMAqV1Fw++23a+LEiQoLC9ONN96o8847T0eOHNGqVas0d+5c3X///d7OCQBAwDIMh5yGU4lrFulQZsmz6upHxWpIlwEcr1mJsDo4ACAYlaso6Nu3r7Zu3arJkydrypQp1nbTNNWzZ0/dd999XgsIAECwOJSZqv1HD9odAwGC1cEBAMGq3KdHHD9+vAYOHKjvvvtOmZmZcjgcuvbaa9W4cWNvZwQAAAg6rA4OAAhWZSoKfvnlFz355JO69tprNXToUDVu3FiNGzdWVlaW2rdvr08++URTp05Vo0aNfJUXAAAgqLA6OAAg2Hg2F07SwYMHNWDAAB05cqRYEVClShWNGjVKGRkZuuOOOzjrAQAAAAAAQcrjomDOnDmKjo7W+++/r+uvv77IZeHh4br77rv13nvvqWrVqkpMTPR6UAAAAAAA4HseFwXr16/XoEGDVLNmzRLH1K5dWwMHDtS3337rlXAAAAAAAMC/PC4K0tLSdOGFF5Y6rkmTJkpJSTmXTAAAAAAAwCYeFwU1a9ZUWlpaqePS09MVFRV1TqEAAAAAAIA9PC4K4uPjtWLFilLHrVy5Us2bNz+nUAAAAAAAwB4eFwX9+/fXxo0bNXHiROXl5RW7PD8/Xy+99JLWrl2rO++806shAQAAAACAf4R4OrBFixZ64okn9OKLL+qDDz5Qhw4ddP7558vlcunQoUPauHGj0tPT9eCDD6pTp06+zAwAAAAAAHzE46JAku688041a9ZM8+fP11dffWXNLKhWrZquuuoqDRw4UFdccYVPggIAAAAAAN8rU1EgSW3atFGbNm0kSceOHVNISIgiIyO9HgwAAAAAAPhfmYuCP6pZs6a3cgAAAAAAgADg8WKGAAAAAACg4qMoAAAAAAAAFooCAAAAAABgOac1CgAAAAKBYThkGI5Sx7ndptxu0w+JAAAIXhQFwJ/wZhMAgothOBQdHSGns/SJki6XWxkZJ3n+BgDgLCgKgD8wDIeiY8LlNJyljnW5XcpIz+HNJgDYzDAccjoNzXz7WyWnZZY4rkGdKA3v21GG4eC5GwCAs6AoAP7AMBxyGk4lrlmkQ5mpJY6rHxWrIV0G8GYTAAJIclqm9iWn2x0DAICgR1EAnMGhzFTtP3rQ7hgAAAAA4Hec9QAAAAAAAFgoCgAAAAAAgIWiAAAAAAAAWCgKAAAAAACAhaIAAAAAAABYKAoAAAAAAICFogAAAAAAAFgoCgAAAAAAgIWiAAAAAAAAWCgKAAAAAACAJcTuAAAAAAAAnCvDcMgwHKWOc7tNud2mHxIFr4CaUZCYmKj+/fsX2bZt2zb169dPcXFxSkhI0KJFi4pc7na7NW3aNHXq1ElxcXEaPHiwfv31V7/vAwAAAABgD8NwKDo6QjEx1Ur9iY6O8KhQqMwCZkbB4sWLNXXqVLVt29balp6ernvuuUcJCQkaN26cfvjhB40bN07VqlVT7969JUmzZs3SkiVLNHHiRNWtW1eTJk3SoEGD9NFHHyk0NNRv+wAAAAAA2MMwHHI6Dc18+1slp2WWOK5BnSgN79tRhuFgVsFZ2F4UpKam6plnntHGjRt14YUXFrnsnXfeUZUqVfTcc88pJCREjRs31v79+zVnzhz17t1b+fn5ev311/XYY4+pa9eukqRXXnlFnTp10urVq9W9e3e/7AMAAAAAYL/ktEztS063O0bQs70o+Pnnn1WlShV9+OGHmjlzppKTk63LkpKS1K5dO4WE/B6zffv2SkxM1JEjR3To0CGdOHFCHTp0sC6PjIxU8+bNtWnTJnXv3t0v+zjvvPPK/fcPCfHN0R9OZ9n3W57rlEdZb8dfucpzW/7MBiBwBfJzhzezORyeH/tpmqV/S+OtbN6+/731GuqL1+JAvc+8uc/KlI3HWvnGk63s43mslW98ZXl9PxPbi4KEhAQlJCSc8bKUlBQ1adKkyLY6depIkg4fPqyUlBRJUr169YqNKbzMH/sob1FgGA7FxFQr13V9ITIy3O4IZxSouaTAzgYgcAXyc8fZsrndpsdFgS+O/fTW/eaL+z9QswVqLm/uk2z27sfb+6tM95k391mZsgVqLl/t01vKms32ouBscnNzFRoaWmRb1apVJUl5eXnKycmRpDOOyczM9Ns+ysvtNpWVdbLc1z8bp9Mo84MhKytHLpfbJ3n+qKzZ/JVLCuxskve/zQPgHYH83OGtbIX78fTYT0/+jt7Odq77Ke/+StqnL16LA/U+I1v5svFYK3suspUvG4+1sufy1T69pbzZIiPDPZpdENBFQVhYmPLz84tsK/xgHhERobCwMElSfn6+9efCMeHh4X7bx7koKPDfh8zSuFzugMpTKFBzSf7NZhgORceEyWk4Sx3rcruUkZ7DAi1AgArm5zVPj/30xd/RW/usTNkCNZc390k2e/fj7f1VpvvMm/usTNkCNZev9uktZc0W0EVB3bp1lZaWVmRb4e+xsbEqKCiwtl1wwQVFxjRt2tRv+wD8wTAcchpOJa5ZpEOZqSWOqx8VqyFdBrCSKwAAAIByCeiV2OLj47V582a5XC5r24YNG9SoUSPVqlVLzZo1U/Xq1bVx40br8qysLG3dulXx8fF+2wfgT4cyU7X/6MESf85WIgAAAABAaQK6KOjdu7eys7M1ZswY7dq1SytWrNCCBQs0ZMgQSafXFejXr58mT56sr776Stu3b9fDDz+sunXrqlu3bn7bBwAAAAAAFUVAH3pQq1YtzZs3T+PHj1fPnj1Vu3ZtjRo1Sj179rTGjBgxQgUFBRo7dqxyc3MVHx+v+fPnq0qVKn7dBwDA9wzDswU9pdOLenL4DQAAQNkFVFEwceLEYttatmypZcuWlXgdp9OpkSNHauTIkSWO8cc+AAC+dXpBz3CPFvSUWNQTAACgvAKqKAAAoCSeLugpsagnAADAuaAoAAAElcIFPQEAAOAbAb2YIQAAAAAA8C+KAgAAAAAAYKEoAAAAAAAAFooCAAAAAABgoSgAAAAAAAAWigIAAAAAAGChKAAAAAAAABaKAgAAAAAAYKEoAAAAAAAAlhC7AwAAgN8ZhkOG4Sh1nNttyu02/ZAIAABUNhQFAAAECMNwKDo6Qk5n6RP+XC63MjJOUhYAAACvoygAACBAGIZDTqehmW9/q+S0zBLHNagTpeF9O8owHBQFAADA6ygKAAAIMMlpmdqXnG53DAAAUEmxmCEAAAAAALBQFAAAAAAAAAtFAQAAAAAAsFAUAAAAAAAAC0UBAAAAAACwUBQAAAAAAAALRQEAAAAAALCE2B0AQPAzDIcMw+HRWLfblNtt+jgRAAAAgPKiKABwTgzDoeiYcDkNp0fjXW6XMtJzKAsAAACAAEVRAOCcGIZDTsOpxDWLdCgz9axj60fFakiXATIMB0UBAAAAEKAoCgB4xaHMVO0/etDuGAAAAADOEYsZAgAAAAAAC0UBAAAAAACwUBQAAAAAAAALRQEAAAAAALBQFAAAAAAAAAtnPQAAAAAAwIcMwyHDcJQ6zu02A+I04hQFAAAAAAD4iGE4FB0dIaez9An9LpdbGRknbS8LKAoAAAAAAPARw3DI6TQ08+1vlZyWWeK4BnWiNLxvRxmGg6IAAAB/83T6nxQ4UwABAEBwS07L1L7kdLtjeISiAABQqZRl+p8UOFMAAQAA/IWiAABQqXg6/U8KrCmAAAAA/kJRAAAoJthW5i2PYJr+BwAA4E8UBQCAIgzDoeiYcDkNZ6ljXW6XMtJzgrYsAAAAQHEUBQCAIgzDIafhVOKaRTqUmVriuPpRsRrSZQDT8gEAACoYigIAwBkdykzV/qMH7Y4BAAAAP/NsyWcAAAAAAFApUBQAAAAAAAALRQEAAAAAALCwRgGACq0ynOYPAAAA8CaKAgAVFqf5AwAAAMqOogBAhcVp/gAAAICyoygAUOFxmj8AAADAcxQFAGAT1k8AAABAIKIoAAAbsH4CAAAAAhVFAQDYgPUTAAAAEKgoCgDARqyfAAAAgEBj2B0AAAAAAAAEDooCAAAAAABgoSgAAAAAAAAWigIAAAAAAGChKAAAAAAAABaKAgAAAAAAYKEoAAAAAAAAFooCAAAAAABgoSgAAAAAAAAWigIAAAAAAGAJsTsAfmcYDhmGo9Rxbrcpt9v0QyIAAAAAQGVDURAgDMOh6OgIOZ2lT/JwudzKyDhJWQAAAAAA8DqKggBhGA45nYZmvv2tktMySxzXoE6UhvftKMNwUBQAAAAAALyOoiDAJKdlal9yut0xAAAAAACVFIsZAgAAAAAAC0UBAAAAAACwUBQAAAAAAAALaxQAAHyG074CAAAEH4oCAIBPcNpXAACA4ERRAADwCU77CgAAEJwoCgAAPsVpXwEAAIILixkCAAAAAAALRQEAAAAAALBQFAAAAAAAAAtFAQAAAAAAsFAUAAAAAAAAC0UBAAAAAACwUBQAAAAAAAALRQEAAAAAALBQFAAAAAAAAAtFAQAAAAAAsITYHQCBzzAcMgyHR2PdblNut+njRAAAAKiInM7Sv8f0ZAyAc0NRgLMyDIeioyM8fkJ2udzKyDhJWQAAABDASntv5+8P41E1wmS63YqMDPfr7QI4M4oCnJVhOOR0Gpr59rdKTss869gGdaI0vG9HGYaDogAAACAABeoH8mphoXIYhvaumquco4fPOjaq0eVq0LmXn5IBlRNFATySnJapfcnpdscAAADAOfD0A7ldH8Zzjh5WTuqBs44Jq1nXT2mK4rAIVCYUBQAAAEAlU9oHcrs+jAeiQJ2FAfgSRQEAAAAAlIDDIlAZURQAAAAAQCkC+bAIwNs4iAYAAAAAAFiYUQAAAAD4QKCdghAAPEVRAAAAAHgRi98BCHYUBQAAAIAXBfopCAGgNBQFAAAAgA9wCkIAwYqiAAAAAACCGOthwNsoCgAgyBmGQ4bhKHWc223K7Tb9kAgAAPgD62HAVygKACCIGYZD0dERHn1T4HK5lZFxkrIAAIAKgvUw4CsUBQAQxAzDIafT0My3v1VyWmaJ4xrUidLwvh1lGA6KAgAAKhjWw4C3URQAQAWQnJapfcnpdscAAABABUBRgKDGsdkAAAAA4F0UBQhaHJsNAAA8eR/Aiu8AUDYUBQhaHJsNAEDlxWrvAOA7FAUIehybDQBA5ePpau8SK74DQFlRFAAAACBolbbau8SK7wBQVhywBQAAAAAALBQFAAAAAADAQlEAAAAAAAAsFAUAAAAAAMBCUQAAAAAAACyc9QAAAAAA4BNO59m/my7tctiDogAAAAAA4FVRNcJkut2KjAy3OwrKgaIA8BHDcMgwHKWOc7tNud2mHxIBAAAA/lEtLFQOw9DeVXOVc/RwieOiGl2uBp17+TEZPEFRAPiAYTgUHR3h0VQql8utjIyTlAUAAACocHKOHlZO6oESLw+rWdePaeApigLABwzDIafT0My3v1VyWmaJ4xrUidLwvh1lGA6KggDHDBEAlZknxTfHGQNAxUFRAPhQclqm9iWn2x0D54gZIgAqK44xBoDKiaIAAErBDBEAlZWnxxhLHGcMABUJRQEAeIgZIgAqq9KOMZY4zhgAKpKgOJgsNTVVTZs2LfazYsUKSdK2bdvUr18/xcXFKSEhQYsWLSpyfbfbrWnTpqlTp06Ki4vT4MGD9euvvxYZ4419AAAAAAAQ7IKiKNi+fbuqVq2qdevW6ZtvvrF+brjhBqWnp+uee+7RBRdcoOXLl2v48OGaPHmyli9fbl1/1qxZWrJkiZ5//nktXbpUbrdbgwYNUn5+viR5ZR8AAAAAAFQEQXHowY4dO3ThhReqTp06xS5buHChqlSpoueee04hISFq3Lix9u/frzlz5qh3797Kz8/X66+/rscee0xdu3aVJL3yyivq1KmTVq9ere7du+udd945530AwYQV/AEAAACUJCiKgl9++UWNGzc+42VJSUlq166dQkJ+/6u0b99eiYmJOnLkiA4dOqQTJ06oQ4cO1uWRkZFq3ry5Nm3apO7du3tlH+UVEnJ6UkdZTylU2vjynKLoTNfx1n7OZZ8ljff2febNfQZyNofDoRo1wjxewf/48VyZZsllgS8eI95SGf49K1M2ntfKN55sZR9v52OtShXnWa/nScnrSYZAOJUh2crHW481bwu2++xs2/0p2LIFaq6zbS/rfnyxz0DOVpKgKAp27NihmJgY3Xnnndq7d68aNmyooUOHqnPnzkpJSVGTJk2KjC+ceXD48GGlpKRIkurVq1dsTOFl3thHeRiGQzEx1cp1XV+cpshb+/R2tkDN5c192pHN0xX8o6MjvB0tYE+zFcz/nv7ejy/2WZmyBWoub+6TbGVTeBrC6tXDvJTotEB9vpXIVl6Bmi1Qc0lkK69AzVaRXwu8vZ9z2WfAFwUFBQXas2ePLr74Yo0ePVrVq1fXxx9/rPvuu09vvPGGcnNzFRoaWuQ6VatWlSTl5eUpJydHks44JjPz9Ickb+yjPNxuU1lZJyWdbnjK8o+XlZUjl8td4uVl3V9J+/TWfs5lnyXtz9v3WWXL5ukK/v56rPlCZfr3rAzZeF4rey6ylS+bHY81T09DWNZTEHrrPvMFspWPt57XvC3Y7jOJbKWp6I+1YH6dOpd9RkaGezS7IOCLgpCQEG3cuFFOp1NhYadb9ssvv1w7d+7U/PnzFRYWVmxBwby8PElSRESEdZ38/Hzrz4VjwsNP37He2Ed5FRSU78OSy+Uu93V9vU9vZwvUXN7cJ9kCQ2W4z8hm7368vb/KdJ95c5+BnK200xCW9RSEgfp8K5GtvAI1W6DmkshWXoGarTK8FgRCNvsPMvFAtWrVinxAl6RLLrlEqampqlu3rtLS0opcVvh7bGysdbjAmcbExsZKklf2AeDcGYZDISGGRz9lPU4XAAAAgGcCvijYuXOnWrdurY0bNxbZ/r///U8XX3yx4uPjtXnzZrlcLuuyDRs2qFGjRqpVq5aaNWum6tWrF7l+VlaWtm7dqvj4eEnyyj4AnBvDcCg6OkIxMdU8+omOjqAsAAAAAHwg4A89aNy4sS666CI999xzGjdunGJiYvTOO+/ohx9+0PLly1WrVi3NmzdPY8aM0aBBg/Tjjz9qwYIFGjdunKTT6wr069dPkydPVs2aNdWgQQNNmjRJdevWVbdu3SRJvXv3Pud9ADg3huGQ02mUusii9PtCi4bh4PSNAAAAgJcFfFFgGIZmz56tKVOm6KGHHlJWVpaaN2+uN954wzpTwbx58zR+/Hj17NlTtWvX1qhRo9SzZ09rHyNGjFBBQYHGjh2r3NxcxcfHa/78+apSpYokWWXDuewDgHd4usgiAAAAAN8I+KJAks477zxNmDChxMtbtmypZcuWlXi50+nUyJEjNXLkSJ/uAwAAAACAYBfwaxQAAAAAAAD/CYoZBQAAAAAAeIvTWfp35p6MqagoCgAAQKVS2hu/yvzGEAAquqgaYTLdbkVGhtsdJaBRFAAAgEqBN4cAgGphoXIYhvaumquco4fPOjaq0eVq0LmXn5IFFooCAABQKXj65rAyvzEEgMoi5+hh5aQeOOuYsJp1/ZQm8FAUAACASqW0N4eV+Y0hAAASZz0AAAAAAAB/QFEAAAAAAAAsFAUAAAAAAMBCUQAAAAAAACwsZggAALzO6Sz9uwhPxgAAAP+jKAAAAF4TVSNMptutyMhwu6MAAIByoigAAABeUy0sVA7D0N5Vc5Vz9PBZx0Y1ulwNOvfyUzIAAOApigIAAOB1OUcPKyf1wFnHhNWs66c0AACgLDg4EAAAAAAAWJhRAABAkGLBQAAA4AsUBQAABBkWDAQAAL5EUQAAQJBhwUAAAOBLFAUAAAQpFgwEAAC+wIGLAAAAAADAQlEAAAAAAAAsFAUAAAAAAMDCGgUAAAAAAASIQDj9MUUBAAAAAAA2C6TTH1MUAAAAAABgs0A6/TFFAQAAAAAAASIQTn/MYoYAAAAAAMBCUQAAAAAAACwUBQAAAAAAwEJRAAAAAAAALCxmCABAKUo7V7Gvz2UMAADgTxQFAACUIJDOZwwAAOAvFAUAAJTA0/MZ+/pcxgAAAP5EUQAAQClKO5+xr89lDAAA4E8cVAkAAAAAACwUBQAAAAAAwMKhB0GKFbgBAAAAAL5AURBkWIEbAAAAAOBLFAVBhhW4AQAAAAC+RFEQpFiBGwAAAADgCxzIDgAAAAAALBQFAAAAAADAQlEAAAAAAAAsFAUAAAAAAMBCUQAAAAAAACwUBQAAAAAAwMLpEeF1TufZ+6fSLgcAAAAA2IeiAF4TVSNMptutyMhwu6MAAAAAAMqJogBeUy0sVA7D0N5Vc5Vz9HCJ46IaXa4GnXv5MRmAYMBsJAAAgMBAUQCvyzl6WDmpB0q8PKxmXT+mARDomI0EAAAQWCgKAAC2YjYSAABAYKEoAAAEBGYjAQAABAYO+AQAAAAAABaKAgAAAAAAYOHQA1QanqyYzqrqAAAAACo7igJUeKyoDgAAAACeoyhAhefpiuoSq6oDAAAAAEUBKo3SVlSXWFUdAAAAACgKgADA+gkAAAAAAgVFAWAj1k+Av1FKAQAAoDQUBYCNWD8B/kIpBQAAAE9RFAABgPUT4GuUUgAAAPAURQEAVCKUUgAAACgNRQGAsyrteHWOZwcAAAAqFooCAGfEMe0AAABA5URRAOCMPD2mnePZAQAAgIqFogDAWZV2TDvHsxfHKQgBAAAQzCgKAAStQFs/gcM1AAAAUBFQFAAIOoH6gZxTEAIAAKAioCgAEHQCff0ETkEIAACAYEZRACBosX4CAAAA4H2spgUAAAAAACwUBQAAAAAAwEJRAAAAAAAALBQFAAAAAADAQlEAAAAAAAAsFAUAAAAAAMBCUQAAAAAAACwUBQAAAAAAwEJRAAAAAAAALBQFAAAAAADAQlEAAAAAAAAsFAUAAAAAAMBCUQAAAAAAACwUBQAAAAAAwEJRAAAAAAAALBQFAAAAAADAQlEAAAAAAAAsFAUAAAAAAMBCUQAAAAAAACwUBQAAAAAAwEJRAAAAAAAALBQFAAAAAADAQlEAAAAAAAAsFAUAAAAAAMBCUQAAAAAAACwUBQAAAAAAwEJRAAAAAAAALBQFAAAAAADAQlEAAAAAAAAsFAUAAAAAAMBCUQAAAAAAACwUBQAAAAAAwEJRAAAAAAAALBQFAAAAAADAQlEAAAAAAAAsFAUAAAAAAMBCUQAAAAAAACwUBQAAAAAAwEJRAAAAAAAALBQFAAAAAADAQlEAAAAAAAAsFAUAAAAAAMBCUQAAAAAAACwUBQAAAAAAwEJRAAAAAAAALBQFAAAAAADAQlEAAAAAAAAsFAVl4Ha7NW3aNHXq1ElxcXEaPHiwfv31V7tjAQAAAADgNRQFZTBr1iwtWbJEzz//vJYuXSq3261BgwYpPz/f7mgAAAAAAHgFRYGH8vPz9frrr2vEiBHq2rWrmjVrpldeeUUpKSlavXq13fEAAAAAAPAKh2mapt0hgsGPP/6oPn366LPPPlOjRo2s7X379lWTJk00bty4Mu/TNE253afvfodDMgxDmdm5crncJV4ntIpT1SOq6tSJLJluV4njjJBQhYRXU1bOcRWcZZwkhRhORYbXkNvt1p8fDZ7m8kW2s+UqSzZPc5GtfNl4rJU9F9nKl43HWtlzka182XislT0X2cqXjcda2XORrXzZeKyVPVdFzWYYDjkcjrPuW6Io8Njq1av1wAMPaMuWLQoLC7O2P/jgg8rNzVViYqKN6QAAAAAA8A4OPfBQTk6OJCk0NLTI9qpVqyovL8+OSAAAAAAAeB1FgYcKZxH8eeHCvLw8hYeH2xEJAAAAAACvoyjwUL169SRJaWlpRbanpaUpNjbWjkgAAAAAAHgdRYGHmjVrpurVq2vjxo3WtqysLG3dulXx8fE2JgMAAAAAwHtC7A4QLEJDQ9WvXz9NnjxZNWvWVIMGDTRp0iTVrVtX3bp1szseAAAAAABeQVFQBiNGjFBBQYHGjh2r3NxcxcfHa/78+apSpYrd0QAAAAAA8ApOjwgAAAAAACysUQAAAAAAACwUBQAAAAAAwEJRAAAAAAAALBQFAAAAAADAQlEAAAAAAAAsFAUAAAAAAMBCURDk3G63pk2bpk6dOikuLk6DBw/Wr7/+anesIhITE9W/f3+7Y1gyMjL09NNPq3PnzmrdurX69u2rpKQku2NJko4ePaqRI0eqffv2atWqle677z7t3r3b7lhF7N27V61atdKKFSvsjiJJSk1NVdOmTYv9BEq+lStX6oYbblCLFi1044036tNPP7U7kjZu3HjG+6xp06a65pprbM1WUFCgV199VVdffbVatWqlO++8Uz/88IOtmQplZ2frmWee0VVXXaV27drpscce09GjR+2Odcbn2G3btqlfv36Ki4tTQkKCFi1aFDDZJGn//v2Ki4vTwYMHbUh15lz/+te/1Lt3b7Vq1UoJCQn65z//qdzc3IDI9sknn6hHjx5q2bKlrr32Ws2dO1d2nN26tNfzsWPHKiEhwY+JfnembGPHji32HOfvfGfKlZaWpkceeURt27bVlVdeqUcffVTHjh3za64zZevfv3+Jrw0rV660NZsk/fzzz+rfv79atWqlrl27avLkycrPz/drrpKyrVu3znr+6NGjh1atWuWXLKW9p12/fr169eqlK664Qtdff70+/vhjv+TyJFshO14PSsu2fPly9ejRQ3FxcerWrZvmzJkjl8tle64333xT3bp1s95TLl++3LeBTAS16dOnm1deeaX573//29y2bZs5cOBAs1u3bmZeXp7d0UzTNM233nrLbNasmdmvXz+7o1juueces3v37uamTZvMPXv2mOPGjTNbtmxp7t692+5o5m233Wb26dPH3LJli7lr1y7zgQceMK+66irz5MmTdkczTdM08/PzzV69eplNmjQxly9fbncc0zRN8+uvvzZbtGhhpqammmlpadZPTk6O3dHMlStXms2bNzffeustc//+/easWbPMZs2amf/9739tzZWXl1fkvkpLSzNXr15tNm3a1HzvvfdszTZt2jSzY8eO5rp168x9+/aZY8aMMdu0aWOmpqbamss0TXPgwIFmly5dzK+//trcsWOHOWzYMPOGG26w9fn2TM+xx44dM6+88krziSeeMHft2mW+9957ZosWLfz+b1vS8/+uXbvMhIQEs0mTJuavv/7q10wl5dq0aZN56aWXmq+99pq5d+9e8+uvvzY7d+5sjh492vZsa9euNS+99FJz0aJF5oEDB8zPP//cjIuLMxcsWGB7tj/64osvzCZNmphXX321X3OZZsnZbrnlFvPll18u8lx39OhRW3Pl5eWZN954o3nbbbeZP//8s/nDDz+YN9xwgzlo0CC/5SopW3p6epH7KjU11bzjjjvMG2+80czOzrY127Fjx8x27dqZTz/9tLlv3z5z7dq1ZocOHcx//vOffstVUrakpCSzadOm5nPPPWfu2rXLXLVqldmqVSvz/fff93mes72n3bVrl9miRQvz5ZdfNnft2mXOmzfPbN68ufmf//zH57lKy1bIrteDs2X74IMPzMsuu8xcunSpuX//fvPjjz82W7dubU6fPt3WXEuXLjVbtmxpfvjhh+aBAwfMZcuWmZdeeqn5xRdf+CwPRUEQy8vLM1u1amUuXrzY2paZmWm2bNnS/Oijj2xMZpopKSnmkCFDzLi4OPP6668PmKJg3759ZpMmTcykpCRrm9vtNq+99lpz6tSpNiYzzYyMDPORRx4xf/nlF2vbtm3bzCZNmphbtmyxMdnvpkyZYg4YMCCgioI5c+aYPXr0sDtGMW6327z66qvNiRMnFtk+cOBAc/bs2TalOrMTJ06YV199td8/FJ3JTTfdZE6YMMH6/fjx42aTJk3Mzz//3MZUprl161azSZMm5po1a6xt2dnZZtu2bc0VK1b4Pc/ZnmNnz55tXnXVVeapU6esbVOmTDG7desWENni4uLMnj17+v2N4dlyPfroo+bdd99dZPz7779vXnbZZX4pgs6Wbfny5eYrr7xSZPywYcPMwYMH+zxXadkKpaammu3btzf79evn16LgbNncbrcZFxdnrl692m95PMm1fPlyMy4uzvztt9+sbWvXrjWvueYa8/jx47Zm+7M333zTvPzyy/32RcrZshUWUX+8j1588UWze/futmcbOnSo2adPnyLjZ82a5fP/C6W9p33qqafMW265pch1HnnkEXPgwIE+zeVJNtO07/WgtGy33367OWbMmCLXmTFjhtmlSxdbc82dO9dcuHBhkev8/e9/N5999lmfZeLQgyC2fft2nThxQh06dLC2RUZGqnnz5tq0aZONyU5PD6tSpYo+/PBDXXHFFbZm+aOYmBjNmTNHLVq0sLY5HA45HA5lZWXZmEyKiorSlClT1KRJE0nSsWPHtGDBAtWtW1cXX3yxrdkkadOmTVq2bJkmTpxod5QifvnlFzVu3NjuGMXs3btXycnJ6tGjR5Ht8+fP15AhQ2xKdWazZ89WTk6OHn/8cbujqFatWvr3v/+tgwcPyuVyadmyZQoNDVWzZs1szbVv3z5JUtu2ba1t1apVU8OGDfXdd9/5Pc/ZnmOTkpLUrl07hYSEWNvat2+vffv26ciRI7Zm+/LLLzVhwgRbHmtnyzVw4MBimQzD0KlTp5SdnW1rtl69eumhhx6SdPpww//85z/atGmTOnbs6PNcpWWTJNM0NXr0aP39739Xu3bt/JLJk2wHDhzQyZMnddFFF/k1U2m5vvnmG7Vv317nnXeeta1Tp0768ssvVb16dVuz/dGxY8c0depUDR061G/34dmy1axZU5L09ttvy+Vy6eDBg1qzZo3f3mOeLdv+/fvVpk2bItuaN2+u5ORkHTp0yGeZSntPm5SUVOQzgnT6tWDz5s0+P3TJk/fbdr0elJbtscce07333lvkOoZhKDMz09ZcgwYN0oABAyRJp06d0ieffKLdu3f79LUgpPQhCFQpKSmSpHr16hXZXqdOHesyuyQkJNh2nOLZREZGqkuXLkW2ff7559q/f7+efPJJm1IV99RTT+mdd95RaGioXnvtNUVERNiaJysrS6NGjdLYsWOLPd7stmPHDsXExOjOO+/U3r171bBhQw0dOlSdO3e2NdfevXslSSdPntS9996rrVu36vzzz9fQoUMD6v9GYSH16KOPKjo62u44GjNmjB588EFdc801cjqdMgxD06dP1wUXXGBrrjp16kiSDh8+bBVTLpdLKSkpqlWrlt/znO05NiUlxSocC/0x/x8/oPg727vvvivp9DoZ/na2XM2bNy/y+6lTp7RgwQJdfvnl1gcUu7IVOnTokK677joVFBToqquuUt++fX2ey5NsCxYs0G+//abZs2crMTHRL5kKnS3bjh07JJ0+pnft2rUyDEOdO3fWww8/rBo1atiWa+/evWrbtq1mzpyplStXWv+eI0eOVGRkpE9zlZbtj+bOnauwsLBiH5h86WzZWrduraFDh+rVV1/VK6+8IpfLpfbt2+vpp5+2PVudOnV0+PDhItsKj7c/evSo6tev75NMpb2nff/991W3bt1iWXNycpSenu7T5zZP3m/b9XpQWrY/lz7Hjx/X22+/rU6dOtmaq1BSUpL69+8vt9ut3r17+3RtKWYUBLGcnBxJUmhoaJHtVatWVV5enh2Rgs5///tfPfHEE+rWrZu6du1qdxzLXXfdpeXLl6t79+4aPny4fv75Z1vzPPvss9YCPYGkoKBAe/bsUWZmph544AHNmTNHcXFxuu+++7R+/XpbsxV+C/n444+re/fuev3119WxY0cNGzbM9mx/tGTJEtWoUUO33Xab3VEkSbt27VKNGjU0c+ZMLVu2TL169dJjjz2mbdu22ZqrRYsWuuiii/TMM88oNTVVubm5mjJlitLT03Xq1Clbs/1Zbm7uGV8XJPHa4IGCggKNGjVKO3fu1DPPPGN3HEtkZKTeffddTZ06Vdu3b9eoUaPsjqTt27drxowZmjRpUrHHnN127NghwzBUp04dzZ49W6NHj9Y333yjYcOGye1225YrOztbK1eu1C+//KIpU6boueee0+bNmzVs2DBbFqg8k+zsbL3zzju69957recOu2VnZ2vPnj2688479e677+rVV1/Vvn379NRTT9kdTX//+9+1evVqffjhhyooKNC2bdv0+uuvS5JfXx/+/J72TK8Fhb/7exHIQH2/LZ0924kTJzRs2DDl5eX5/Tm3pFyNGjXS+++/rxdeeEGffvqpJk+e7LMMzCgIYmFhYZJO/2cv/LN0+o1geHi4XbGCxpdffqnHHntMrVu39ul/svIoPNRg/Pjx2rJli9566y1NmDDBliwrV65UUlKSPvroI1tu/2xCQkK0ceNGOZ1O6//A5Zdfrp07d2r+/PnFptz5U5UqVSRJ9957r3r27ClJuvTSS7V161a98cYbtmb7o5UrV+rmm28u8hxil8OHD+vRRx/VggULrCn+LVq00K5duzR9+nTNmjXLtmyhoaGaMWOGRo0apc6dO6tKlSrq0aOHrr76ahlGYHXuYWFhxd4EFhYEds9OCnTZ2dl66KGH9N1332nGjBlq2bKl3ZEs1atXV/PmzdW8eXO5XC49+uijGjlypBo0aGBLnry8PD322GMaOnSo7YcGncnQoUN1xx13KCYmRpLUpEkT1a5dW7feeqt++ukn2w6LDAkJUUREhKZMmWK9TkRFRalPnz766aefAuIx9+WXXyo/P1+9e/e2O4pl0qRJyszM1LRp0yRJl112maKionT33Xfr7rvv1qWXXmpbtptvvlnJycl66qmn9Pjjj6tevXoaPHiwnn32WZ/PXil0pve0VatWLfZaUPi7Pz8nBPL77bNl++233zRkyBAdPHhQ8+fP1/nnnx8QuWrVqqVatWqpWbNmOnbsmGbMmKEHH3zQJ2VtYL27QZkUTgFPS0srsj0tLU2xsbF2RAoab731lh544AFdffXVmj17dkA05seOHdPHH3+sgoICa5thGLr44ouL/Rv70/Lly3X06FF17dpVrVq1UqtWrSRJzzzzjAYNGmRbrkLVqlUr9iH3kksuUWpqqk2JTiv8P/jnKeAXX3yxbaeE+7Pt27fr119/DZiZIlu2bNGpU6eKHJ8nSVdccYX2799vU6rfNW7cWMuXL9fGjRu1YcMGTZgwQSkpKbYfFvFndevWPePrgiReG84iLS3NOh3n/Pnzi00BtUtSUpJ+/PHHItuaNm0qqfjrvz9t2bJFO3fu1IwZM6zXhsTERB06dEitWrWy/bTDhmFYJUGhSy65RJJsPTyzbt26atSokVUSSL/nCpTXhi+//FJdunTxy6EQntq8efMZXxuk39eQsdPw4cP13//+V19//bW+/PJL1a9fX06n02eHHfxRSe9p69Wrd8bXgoiICL8VGIH4frvQ2bLt3r1bt956q44eParFixcXe+zZkWvt2rXatWtXkbFNmzZVfn6+MjIyfJKFoiCINWvWTNWrVy9ybE9WVpa2bt2q+Ph4G5MFtiVLluj555/XnXfeqZdffjlgpkseOXJEjzzySJFp6adOndLWrVttXaxv8uTJ+uSTT7Ry5UrrR5JGjBih8ePH25ZLknbu3KnWrVsXO77tf//7n+0LQF522WWqVq2atmzZUmT7jh07AuaDZVJSktVKB4LCYyl/+eWXItt37NihCy+80IZEv8vOzla/fv20fft2RUdHq3r16jp48KC2bt3qt0XlPBUfH6/NmzcXOefzhg0b1KhRI1vWUwgGmZmZuuuuu3Ts2DEtXrw4oF5DFy1apBdffLHIti1btigkJMTW/xctW7bU6tWr9cEHH1ivDbfffrvq1KmjlStX6vLLL7ctmySNGjVKd999d5FtP/30kyTZ+voQHx+v7du3Kzc319pWuJ5Cw4YN7YpVxJkWwbNbbGxssdeGwt8bNWpkRyTLW2+9peeff15Op1OxsbEyDEOff/65WrVqpWrVqvn0ts/2nrZt27bFFtvdsGGDWrdu7ZeZcIH6fls6e7Zff/1Vd911l8LDw7V06VKryLM719SpU4vNrNyyZYuio6N9tvYQRUEQCw0NVb9+/TR58mR99dVX2r59ux5++GHVrVtX3bp1szteQNq7d69efPFFXXfddRoyZIiOHDmi3377Tb/99puOHz9ua7YmTZqoc+fOeuGFF7Rp0ybt2LFDo0ePVlZWVrE3O/4UGxurhg0bFvmRTk99svvbycaNG+uiiy7Sc889p6SkJO3evVsTJkzQDz/8oKFDh9qaLSwsTIMGDdLMmTO1atUqHThwQK+99pq+/fZb3XPPPbZmK7R161brm8lA0LJlS7Vp00aPP/64NmzYoH379mnq1Klav3697rvvPluzVa9eXaZpavz48dq5c6d++uknDR06VO3btw+4N9S9e/dWdna2xowZo127dmnFihVasGBBwJ1tI5BMmDBBv/76qyZNmqSaNWtarwu//fZbkcLFDnfffbd+/PFHvfLKK9q/f78+/fRTTZo0SQMGDCj2jbk/hYWFFXttiIqKUkhIiBo2bGj74Ux//etftX79es2YMUMHDhzQmjVr9OSTT6p79+62lu+33367nE6nHn30Ue3cuVObN2/W2LFjdeWVV+qyyy6zLVehw4cPKz09PWAK5EJ333231q1bp6lTp+rAgQNav369nnjiCXXt2tX2rI0bN9bSpUu1cuVKHTx4UHPmzNGHH36oBx980Ke3W9p72v79++vHH3/U5MmTtXv3br3++uv67LPP/DIbNJDfb5eW7cknn1R+fr5efvllhYSEFHk9sDPXoEGD9Mknn+itt97S/v379c4772j+/Pl64IEHfFb8sEZBkBsxYoQKCgo0duxY5ebmKj4+XvPnzy8ypQ2/+/zzz3Xq1Cl98cUX+uKLL4pc1rNnT9tP/ffyyy9rypQpevjhh3X8+HG1bdtWixcv9svUtWBkGIZmz56tKVOm6KGHHlJWVpaaN2+uN954o9iUfzsMGzZM4eHheuWVV5SamqrGjRtr+vTpuvLKK+2OJun08XeBcKaDQoZh6LXXXtPUqVP1xBNPKDMzU02aNNGCBQsC4jSrL7/8sp5//nn17dtXoaGh6tatm0aOHGl3rGJq1aqlefPmafz48erZs6dq166tUaNGWWtloCiXy6VPPvlEp06d0l133VXs8q+++sqvx6b+WevWrZWYmKipU6dqwYIFqlmzpgYOHKjBgwfblikYXHPNNZo6darmzJmjuXPnqkaNGurRo4d1qkm71KxZU4sXL9aECRPUp08fhYaG6tprr9Xo0aNtzVWo8MNQIL02SKdPIZmYmKiZM2dq4cKFiomJ0XXXXefzD+Oe6NChg8aNG6dZs2YpNTVVF198sV577TWfny7Uk/e0s2bN0qRJk7Rw4UKdf/75mjRpkl/K7UB+v322bB07drRmYfz9738vdt0/z2rxV67C++zUqVOaO3eu/vnPf6p+/fp66qmn1KdPH59lcpiBssQqAAAAAACwHYceAAAAAAAAC0UBAAAAAACwUBQAAAAAAAALRQEAAAAAALBQFAAAAAAAAAtFAQAAAAAAsFAUAAAAAAAAC0UBAAAIWqZp2h0BAIAKh6IAAIAKZPTo0UpISCjx8oSEBI0ePdqPiconJSVFd955p1q0aKEOHTooJyen2JidO3eqb9++RbY1bdpU06dP90mm/v37q2nTptZPs2bN1KpVK/Xq1UuLFi1SQUGBT24XAAB/C7E7AAAAwJ8tXLhQP/zwgyZNmqTY2FiFh4cXG/PZZ5/p+++/92uu5s2b65lnnpEkuVwuZWZmau3atZowYYKSkpI0depUGQbfwwAAghtFAQAACDgZGRmqU6eObrjhBrujFFG9enXFxcUV2ZaQkKCLLrpI48eP16pVq3TTTTfZEw4AAC+h8gYAoBJzuVxavHixevTooZYtW6pr166aPHmy8vLyrDH9+/dX//79i1xv48aNatq0qTZu3ChJWrFihZo3b653331XHTt2VLt27bRr164z3ubx48c1YcIEXXvttWrRooW6d++u9957z7o8ISFBK1as0KFDh0o8lGD69OmaMWOGpOKHG2RnZ2vMmDFq166dWrVqpREjRujIkSNFrv/ll1+qV69eatGihTp27KgXXnhBJ0+eLOO997t+/fopNjZWS5cutbbl5uZqypQp6tatmy6//HK1bt1a99xzj7Zt2yZJ+vrrr9W0aVN98803RfaVlJSkpk2bavPmzeXOAwDAuWBGAQAAFZCnx8s//fTT+uCDDzR48GC1bdtWW7du1cyZM7Vt2zbNmzdPDofD49t0uVx6/fXXNX78eKWnp6tx48bFxuTm5uqOO+7Q0aNHNWLECDVo0EBffvmlxowZoyNHjugf//iHZsyYoalTp2rr1q2aMWOG6tatW2w/ffr0UUpKit577z0tW7asyJhFixapR48eevXVV7V7925NnDhRkjRt2jRJ0kcffaTHHntMPXr00EMPPaTk5GS98sor2rVrl954440y/Z0LGYahDh066OOPP1ZBQYFCQkI0atQoJSUl6ZFHHtEFF1yg/fv369VXX9Wjjz6qjz/+WJ06dVKdOnX0wQcf6KqrrrL2tXLlSl144YVq06ZNmXMAAOANFAUAAFQwycnJuuyyy0odt2vXLr333nt69NFHdd9990mSOnbsqDp16mjUqFFau3atunTpUqbb/sc//qGuXbuWePmKFSu0Y8cOLV26VK1atZIkderUSQUFBZo1a5Zuv/12NW/eXDVr1lRoaGixaf6F6tata5UDfx7TokULvfTSS5KkDh06aMuWLVqzZo2k02dJmDx5sjp16qTJkydb17nwwgt19913a82aNWfNfzbnnXeeTp06pYyMDEVGRurEiRMaO3asdfhEu3btlJ2drYkTJ+rIkSOqXbu2evbsqTfffFMnTpxQtWrVlJubq08//dT69wAAwA4cegAAQAVTu3Ztvffee2f8qV27tjXuu+++kyTdeOONRa5/4403yul0WocVlMWll1561su/++47NWjQwCoJCt10003Ky8vTli1bynybf/bnb+LPP/98ZWVlSZL27NmjlJQUJSQkqKCgwPqJj49X9erV9e2335b7dgtP1ehwOBQaGqr58+frhhtuUGpqqjZs2KClS5fq3//+tyQpPz9fktS7d2+dPHlSX3zxhSTpiy++0MmTJ3XzzTeXOwcAAOeKGQUAAFQwoaGhatGiRYmXFcrMzJSkIuWBJIWEhCgmJkbHjx8v821HRESc9fLMzMxityed/jZekvWB/lz8OYNhGNaH+IyMDEnSuHHjNG7cuGLXTUtLK/ftpqamKiwsTNHR0ZKkdevW6cUXX9SePXtUrVo1NWvWzMpWmKdhw4Zq166dVq5cqZtvvlkrV67UX/7yF8XGxpY7BwAA54qiAACASioqKkqS9Ntvv6lBgwbW9lOnTik9PV0xMTHWNpfLVeS65V34LyoqSvv37y+2/bfffpOkIrfpC5GRkZKkUaNGqV27dmfMVx4FBQXauHGjWrduLafTqQMHDmj48OG69tprlZiYqP/7v/+Tw+HQ4sWLtW7duiLX7d27t5588knt3r1b69evL3JIBAAAduDQAwAAKqnCD8off/xxke0ff/yxXC6XNYW/evXqSklJKTKmvCvyx8fHKzk5Wd9//32R7R9++KGqVKmili1berwvwyj725iLLrpItWrV0sGDB9WiRQvrJzY2VlOmTNHWrVvLvE9JWrZsmX777Tf17dtXkvS///1PeXl5uu+++3TBBRdYCyQWlgSFMwok6a9//avCw8P17LPPqlq1arr22mvLlQEAAG9hRgEAAJXUxRdfrJ49e2ratGnKyclRfHy8tm3bphkzZujKK69Up06dJElXX321/vWvf2nChAlKSEhQUlKSVq5cWa7b7NWrl5YsWaLhw4drxIgROv/88/Wvf/1Ly5cv1/3332994++JwrGrVq3SFVdcof/7v/8r9TpOp1MPP/ywnn76aTmdTl199dXKysrSrFmzlJqaWuoikNnZ2frhhx8kSW63W+np6frmm2+0bNky3XTTTerWrZsk6bLLLlNISIgmTZqkgQMHKj8/XytWrNDXX38tqeiMjPDwcN14441atmyZ+vbtW+TwEAAA7EBRAABAJTZ+/Hg1bNhQy5cv19y5c1WnTh0NGDBAw4YNs76x7927tw4cOKD3339fS5cuVXx8vKZNm2Z9e14W4eHhevPNNzVlyhS9+uqrys7O1kUXXaTx48frlltuKdO+unXrpg8++ECjR4/WLbfcomeffdaj6/Xp00fVqlXTvHnztGzZMkVERKh169aaPHlyqWXD1q1bddttt0k6vWhhtWrV1KRJEz377LPq06ePNa5hw4aaMmWKZsyYoaFDhyoqKkpxcXF688031b9/fyUlJalp06bW+K5du2rZsmXq1atXme4DAAB8wWH+ce4bAAAA/O6ZZ57Rli1byj1TAwAAb2JGAQAAgE0WLVqkPXv26J133tGkSZPsjgMAgCSKAgAAANskJSVp3bp1uuuuu9S9e3e74wAAIIlDDwAAAAAAwB9wekQAAAAAAGChKAAAAAAAABaKAgAAAAAAYKEoAAAAAAAAFooCAAAAAABgoSgAAAAAAAAWigIAAAAAAGChKAAAAAAAAJb/B6qJ3FOZeghgAAAAAElFTkSuQmCC\n",
      "text/plain": [
       "<Figure size 1200x800 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "fig, ax = plt.subplots(figsize=(12,8))\n",
    "ax = sns.countplot(data=train_users, x=\"hod\", hue=\"gender\")\n",
    "ax.set_xlabel(\"Hour of the Day\")\n",
    "ax.set_ylabel(\"Count\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Judging by the chart, there seem to be no evident difference. Well, beware!"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3 (ipykernel)",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.9.5"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 4
}