notebooks/2.1_gb_fullname.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "id": "43bdd7b0-bff8-428d-9025-0593bd0330d5",
   "metadata": {
    "tags": []
   },
   "outputs": [],
   "source": [
    "import numpy as np\n",
    "import pandas as pd\n",
    "import joblib\n",
    "\n",
    "from sklearn.ensemble import GradientBoostingClassifier\n",
    "from sklearn.feature_extraction.text import CountVectorizer\n",
    "from sklearn import metrics"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "id": "6f9bb41e-594e-43a6-ab47-cb2dfc35f40f",
   "metadata": {
    "tags": []
   },
   "outputs": [],
   "source": [
    "train_df = pd.read_csv('./data/fl_2022_FullName_train.csv.gz')\n",
    "val_df = pd.read_csv('./data/fl_2022_FullName_val.csv.gz')\n",
    "test_df = pd.read_csv('./data/fl_2022_FullName_test.csv.gz')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "id": "08cc4254-d07e-4b01-9e55-14ca25bdc558",
   "metadata": {
    "tags": []
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "(7214894, 9)\n",
      "(901862, 9)\n",
      "(901862, 9)\n"
     ]
    }
   ],
   "source": [
    "print(train_df.shape)\n",
    "print(val_df.shape)\n",
    "print(test_df.shape)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "id": "e83c6439-a067-4707-bbc1-46629517b395",
   "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>full_name</th>\n",
       "      <th>asian</th>\n",
       "      <th>hispanic</th>\n",
       "      <th>nh_black</th>\n",
       "      <th>nh_white</th>\n",
       "      <th>other</th>\n",
       "      <th>total_n</th>\n",
       "      <th>race</th>\n",
       "      <th>race_code</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>Veit Karen</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>2.0</td>\n",
       "      <td>nh_white</td>\n",
       "      <td>3</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>Barahona Octavio</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>hispanic</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>Albert Jurieal</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>nh_black</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>Foco Kevin</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>nh_white</td>\n",
       "      <td>3</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>Ewen Deborah</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>nh_white</td>\n",
       "      <td>3</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "          full_name  asian  hispanic  nh_black  nh_white  other  total_n   \n",
       "0        Veit Karen    0.0       0.0       0.0       1.0    0.0      2.0  \\\n",
       "1  Barahona Octavio    0.0       1.0       0.0       0.0    0.0      1.0   \n",
       "2    Albert Jurieal    0.0       0.0       1.0       0.0    0.0      1.0   \n",
       "3        Foco Kevin    0.0       0.0       0.0       1.0    0.0      1.0   \n",
       "4      Ewen Deborah    0.0       0.0       0.0       1.0    0.0      1.0   \n",
       "\n",
       "       race  race_code  \n",
       "0  nh_white          3  \n",
       "1  hispanic          1  \n",
       "2  nh_black          2  \n",
       "3  nh_white          3  \n",
       "4  nh_white          3  "
      ]
     },
     "execution_count": 4,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "train_df.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "4e2ed379-51d3-445d-995e-258c724d3abb",
   "metadata": {},
   "source": [
    "## Extract features"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 63,
   "id": "56a1d629-37ea-435f-b99e-f48e285335d6",
   "metadata": {
    "tags": []
   },
   "outputs": [],
   "source": [
    "vectorizer = CountVectorizer(analyzer='char', ngram_range=(1, 2), min_df=50, lowercase=False)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 64,
   "id": "accd0e12-5cb2-4f64-b46e-fd0e65b4843c",
   "metadata": {
    "tags": []
   },
   "outputs": [],
   "source": [
    "X_train = vectorizer.fit_transform(train_df['full_name'])\n",
    "X_val = vectorizer.transform(val_df['full_name'])\n",
    "X_test = vectorizer.transform(test_df['full_name'])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 65,
   "id": "d18ba028-c270-49fc-8921-18b996754fa5",
   "metadata": {
    "tags": []
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(7214894, 1267)"
      ]
     },
     "execution_count": 65,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "X_train.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 66,
   "id": "84e32c72-e1c4-48dc-bda2-6348b9114e5c",
   "metadata": {
    "tags": []
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(901862, 1267)"
      ]
     },
     "execution_count": 66,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "X_val.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 67,
   "id": "0af4b540-2f0b-4af3-8e48-931c105419af",
   "metadata": {
    "tags": []
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(901862, 1267)"
      ]
     },
     "execution_count": 67,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "X_test.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 68,
   "id": "7078aec6-e9ef-4393-9086-e6c6b919a8a6",
   "metadata": {
    "tags": []
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([' ', \" '\", ' -', ..., 'zw', 'zy', 'zz'], dtype=object)"
      ]
     },
     "execution_count": 68,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "vectorizer.get_feature_names_out()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 69,
   "id": "8f8729a2-b6f0-4265-abbd-d9878879f14a",
   "metadata": {
    "tags": []
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "1267"
      ]
     },
     "execution_count": 69,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "len(vectorizer.get_feature_names_out())"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 70,
   "id": "19714c3f-3aa9-4e2f-8145-1aeda563f13c",
   "metadata": {
    "tags": []
   },
   "outputs": [],
   "source": [
    "vocab = vectorizer.get_feature_names_out()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 71,
   "id": "550526c7-8ac1-4e4f-b1ca-3922832c6119",
   "metadata": {
    "tags": []
   },
   "outputs": [],
   "source": [
    "y_train = train_df.race_code\n",
    "y_val = val_df.race_code\n",
    "y_test = test_df.race_code"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 72,
   "id": "9b0c0894-2832-4d63-a09b-0063c7d06207",
   "metadata": {
    "tags": []
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "3"
      ]
     },
     "execution_count": 72,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "y_train[0]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 73,
   "id": "c5747cde-b751-4037-9012-95ec8e7b4566",
   "metadata": {
    "tags": []
   },
   "outputs": [],
   "source": [
    "#Create a Gaussian Classifier\n",
    "clf = GradientBoostingClassifier(random_state=42, verbose=6)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 74,
   "id": "27b2c2e0-561e-42c9-965f-72cfcfd2caa8",
   "metadata": {
    "tags": []
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "(7214894, 1267)\n",
      "(901862, 1267)\n"
     ]
    }
   ],
   "source": [
    "print(X_train.shape)\n",
    "print(X_test.shape)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 75,
   "id": "fdfccc76-a48e-4666-967b-919e3e13a55f",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "scipy.sparse._csr.csr_matrix"
      ]
     },
     "execution_count": 75,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "type(X_train)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "f4a954e6-adb7-4ed1-a998-ff259bb63d59",
   "metadata": {},
   "source": [
    "## Training"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "3eaf206d-5323-4b62-b449-6ee4df96390c",
   "metadata": {},
   "source": [
    "## Doing it for 1 million with 100 iterations"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 79,
   "id": "d17343e3-38ff-4d22-afcc-29eaef58f443",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "CPU times: user 4 µs, sys: 0 ns, total: 4 µs\n",
      "Wall time: 16.2 µs\n",
      "      Iter       Train Loss   Remaining Time \n",
      "         1           1.0630           63.65m\n",
      "         2           1.0401           61.77m\n",
      "         3           1.0209           61.61m\n",
      "         4           1.0073           60.39m\n",
      "         5           0.9962           60.00m\n",
      "         6           0.9851           59.38m\n",
      "         7           0.9754           58.55m\n",
      "         8           0.9669           57.99m\n",
      "         9           0.9604           57.35m\n",
      "        10           0.9539           56.75m\n",
      "        11           0.9483           56.19m\n",
      "        12           0.9430           55.66m\n",
      "        13           0.9381           55.23m\n",
      "        14           0.9335           54.77m\n",
      "        15           0.9297           54.02m\n",
      "        16           0.9257           53.36m\n",
      "        17           0.9220           52.68m\n",
      "        18           0.9185           52.01m\n",
      "        19           0.9152           51.39m\n",
      "        20           0.9124           50.69m\n",
      "        21           0.9094           49.99m\n",
      "        22           0.9066           49.34m\n",
      "        23           0.9039           48.60m\n",
      "        24           0.9015           47.93m\n",
      "        25           0.8994           47.24m\n",
      "        26           0.8968           46.60m\n",
      "        27           0.8949           45.98m\n",
      "        28           0.8931           45.24m\n",
      "        29           0.8910           44.64m\n",
      "        30           0.8888           44.00m\n",
      "        31           0.8868           43.37m\n",
      "        32           0.8852           42.69m\n",
      "        33           0.8836           42.03m\n",
      "        34           0.8817           41.40m\n",
      "        35           0.8799           40.82m\n",
      "        36           0.8782           40.21m\n",
      "        37           0.8768           39.57m\n",
      "        38           0.8754           38.93m\n",
      "        39           0.8737           38.30m\n",
      "        40           0.8719           37.75m\n",
      "        41           0.8704           37.14m\n",
      "        42           0.8689           36.50m\n",
      "        43           0.8675           35.93m\n",
      "        44           0.8663           35.33m\n",
      "        45           0.8650           34.70m\n",
      "        46           0.8638           34.09m\n",
      "        47           0.8625           33.48m\n",
      "        48           0.8611           32.89m\n",
      "        49           0.8603           32.24m\n",
      "        50           0.8591           31.60m\n",
      "        51           0.8579           30.98m\n",
      "        52           0.8568           30.33m\n",
      "        53           0.8558           29.68m\n",
      "        54           0.8546           29.02m\n",
      "        55           0.8535           28.37m\n",
      "        56           0.8526           27.72m\n",
      "        57           0.8516           27.08m\n",
      "        58           0.8508           26.42m\n",
      "        59           0.8497           25.78m\n",
      "        60           0.8487           25.12m\n",
      "        61           0.8478           24.47m\n",
      "        62           0.8470           23.83m\n",
      "        63           0.8461           23.20m\n",
      "        64           0.8453           22.55m\n",
      "        65           0.8444           21.91m\n",
      "        66           0.8436           21.27m\n",
      "        67           0.8427           20.64m\n",
      "        68           0.8420           20.01m\n",
      "        69           0.8412           19.38m\n",
      "        70           0.8402           18.74m\n",
      "        71           0.8395           18.12m\n",
      "        72           0.8387           17.49m\n",
      "        73           0.8380           16.85m\n",
      "        74           0.8373           16.23m\n",
      "        75           0.8367           15.60m\n",
      "        76           0.8360           14.97m\n",
      "        77           0.8353           14.34m\n",
      "        78           0.8344           13.70m\n",
      "        79           0.8338           13.08m\n",
      "        80           0.8331           12.45m\n",
      "        81           0.8322           11.83m\n",
      "        82           0.8314           11.21m\n",
      "        83           0.8308           10.59m\n",
      "        84           0.8304            9.96m\n",
      "        85           0.8298            9.34m\n",
      "        86           0.8290            8.71m\n",
      "        87           0.8283            8.09m\n",
      "        88           0.8275            7.47m\n",
      "        89           0.8270            6.84m\n",
      "        90           0.8264            6.22m\n",
      "        91           0.8258            5.60m\n",
      "        92           0.8252            4.98m\n",
      "        93           0.8247            4.35m\n",
      "        94           0.8240            3.73m\n",
      "        95           0.8235            3.11m\n",
      "        96           0.8229            2.48m\n",
      "        97           0.8222            1.86m\n",
      "        98           0.8217            1.24m\n",
      "        99           0.8212           37.25s\n",
      "       100           0.8207            0.00s\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "<style>#sk-container-id-1 {color: black;background-color: white;}#sk-container-id-1 pre{padding: 0;}#sk-container-id-1 div.sk-toggleable {background-color: white;}#sk-container-id-1 label.sk-toggleable__label {cursor: pointer;display: block;width: 100%;margin-bottom: 0;padding: 0.3em;box-sizing: border-box;text-align: center;}#sk-container-id-1 label.sk-toggleable__label-arrow:before {content: \"▸\";float: left;margin-right: 0.25em;color: #696969;}#sk-container-id-1 label.sk-toggleable__label-arrow:hover:before {color: black;}#sk-container-id-1 div.sk-estimator:hover label.sk-toggleable__label-arrow:before {color: black;}#sk-container-id-1 div.sk-toggleable__content {max-height: 0;max-width: 0;overflow: hidden;text-align: left;background-color: #f0f8ff;}#sk-container-id-1 div.sk-toggleable__content pre {margin: 0.2em;color: black;border-radius: 0.25em;background-color: #f0f8ff;}#sk-container-id-1 input.sk-toggleable__control:checked~div.sk-toggleable__content {max-height: 200px;max-width: 100%;overflow: auto;}#sk-container-id-1 input.sk-toggleable__control:checked~label.sk-toggleable__label-arrow:before {content: \"▾\";}#sk-container-id-1 div.sk-estimator input.sk-toggleable__control:checked~label.sk-toggleable__label {background-color: #d4ebff;}#sk-container-id-1 div.sk-label input.sk-toggleable__control:checked~label.sk-toggleable__label {background-color: #d4ebff;}#sk-container-id-1 input.sk-hidden--visually {border: 0;clip: rect(1px 1px 1px 1px);clip: rect(1px, 1px, 1px, 1px);height: 1px;margin: -1px;overflow: hidden;padding: 0;position: absolute;width: 1px;}#sk-container-id-1 div.sk-estimator {font-family: monospace;background-color: #f0f8ff;border: 1px dotted black;border-radius: 0.25em;box-sizing: border-box;margin-bottom: 0.5em;}#sk-container-id-1 div.sk-estimator:hover {background-color: #d4ebff;}#sk-container-id-1 div.sk-parallel-item::after {content: \"\";width: 100%;border-bottom: 1px solid gray;flex-grow: 1;}#sk-container-id-1 div.sk-label:hover label.sk-toggleable__label {background-color: #d4ebff;}#sk-container-id-1 div.sk-serial::before {content: \"\";position: absolute;border-left: 1px solid gray;box-sizing: border-box;top: 0;bottom: 0;left: 50%;z-index: 0;}#sk-container-id-1 div.sk-serial {display: flex;flex-direction: column;align-items: center;background-color: white;padding-right: 0.2em;padding-left: 0.2em;position: relative;}#sk-container-id-1 div.sk-item {position: relative;z-index: 1;}#sk-container-id-1 div.sk-parallel {display: flex;align-items: stretch;justify-content: center;background-color: white;position: relative;}#sk-container-id-1 div.sk-item::before, #sk-container-id-1 div.sk-parallel-item::before {content: \"\";position: absolute;border-left: 1px solid gray;box-sizing: border-box;top: 0;bottom: 0;left: 50%;z-index: -1;}#sk-container-id-1 div.sk-parallel-item {display: flex;flex-direction: column;z-index: 1;position: relative;background-color: white;}#sk-container-id-1 div.sk-parallel-item:first-child::after {align-self: flex-end;width: 50%;}#sk-container-id-1 div.sk-parallel-item:last-child::after {align-self: flex-start;width: 50%;}#sk-container-id-1 div.sk-parallel-item:only-child::after {width: 0;}#sk-container-id-1 div.sk-dashed-wrapped {border: 1px dashed gray;margin: 0 0.4em 0.5em 0.4em;box-sizing: border-box;padding-bottom: 0.4em;background-color: white;}#sk-container-id-1 div.sk-label label {font-family: monospace;font-weight: bold;display: inline-block;line-height: 1.2em;}#sk-container-id-1 div.sk-label-container {text-align: center;}#sk-container-id-1 div.sk-container {/* jupyter's `normalize.less` sets `[hidden] { display: none; }` but bootstrap.min.css set `[hidden] { display: none !important; }` so we also need the `!important` here to be able to override the default hidden behavior on the sphinx rendered scikit-learn.org. See: https://github.com/scikit-learn/scikit-learn/issues/21755 */display: inline-block !important;position: relative;}#sk-container-id-1 div.sk-text-repr-fallback {display: none;}</style><div id=\"sk-container-id-1\" class=\"sk-top-container\"><div class=\"sk-text-repr-fallback\"><pre>GradientBoostingClassifier(random_state=42, verbose=6)</pre><b>In a Jupyter environment, please rerun this cell to show the HTML representation or trust the notebook. <br />On GitHub, the HTML representation is unable to render, please try loading this page with nbviewer.org.</b></div><div class=\"sk-container\" hidden><div class=\"sk-item\"><div class=\"sk-estimator sk-toggleable\"><input class=\"sk-toggleable__control sk-hidden--visually\" id=\"sk-estimator-id-1\" type=\"checkbox\" checked><label for=\"sk-estimator-id-1\" class=\"sk-toggleable__label sk-toggleable__label-arrow\">GradientBoostingClassifier</label><div class=\"sk-toggleable__content\"><pre>GradientBoostingClassifier(random_state=42, verbose=6)</pre></div></div></div></div></div>"
      ],
      "text/plain": [
       "GradientBoostingClassifier(random_state=42, verbose=6)"
      ]
     },
     "execution_count": 79,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "%time\n",
    "# 1 million\n",
    "samples = 100 * 100 * 100\n",
    "\n",
    "tr_df = train_df.sample(samples)\n",
    "X_tr = vectorizer.transform(tr_df['full_name'])\n",
    "y_tr = tr_df.race_code\n",
    "\n",
    "clf.fit(X_tr, y_tr)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 80,
   "id": "b0e02b4b-c6fb-45db-b25d-4a8ca6b1b1d9",
   "metadata": {
    "tags": []
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Test Accuracy: 0.7080296098516181\n",
      "CPU times: user 12.8 s, sys: 1.44 s, total: 14.3 s\n",
      "Wall time: 19.8 s\n"
     ]
    }
   ],
   "source": [
    "%%time\n",
    "y_pred = clf.predict(X_val)\n",
    "# Model Accuracy, how often is the classifier correct?\n",
    "print(\"Test Accuracy:\", metrics.accuracy_score(y_val, y_pred))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 81,
   "id": "706b6bbd-7bf7-4f7f-b3ca-42f425ee8da0",
   "metadata": {
    "tags": []
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Test Accuracy: 0.7075339686116058\n",
      "CPU times: user 10.4 s, sys: 1.2 s, total: 11.6 s\n",
      "Wall time: 15.1 s\n"
     ]
    }
   ],
   "source": [
    "%%time\n",
    "y_pred = clf.predict(X_test)\n",
    "# Model Accuracy, how often is the classifier correct?\n",
    "print(\"Test Accuracy:\", metrics.accuracy_score(y_test, y_pred))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 82,
   "id": "2899fe5b-1699-444d-988a-981a7053940f",
   "metadata": {
    "tags": []
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "['models/gb_vec_fullname_1M.joblib']"
      ]
     },
     "execution_count": 82,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "joblib.dump(vectorizer, \"models/gb_vec_fullname_1M.joblib\", compress=3)  # compression is ON!"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 83,
   "id": "7d836a71-0053-4159-819e-b78f7fbbd27c",
   "metadata": {
    "tags": []
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "['models/gb_model_fullname_1M.joblib']"
      ]
     },
     "execution_count": 83,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "joblib.dump(clf, \"models/gb_model_fullname_1M.joblib\", compress=3)  # compression is ON!"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "e273dbc0-f400-4915-9f73-fcf0532c0155",
   "metadata": {},
   "source": [
    "## Doing it for 2 million with 100 iterations"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 84,
   "id": "76f1c55a-8ff1-4286-8670-699c2bba71d5",
   "metadata": {
    "tags": []
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "      Iter       Train Loss   Remaining Time \n",
      "         1           1.0630          172.90m\n",
      "         2           1.0416          168.42m\n",
      "         3           1.0223          164.00m\n",
      "         4           1.0091          160.43m\n",
      "         5           0.9963          157.65m\n",
      "         6           0.9854          155.09m\n",
      "         7           0.9760          152.11m\n",
      "         8           0.9688          150.05m\n",
      "         9           0.9613          148.25m\n",
      "        10           0.9551          146.52m\n",
      "        11           0.9491          144.67m\n",
      "        12           0.9439          142.53m\n",
      "        13           0.9389          140.55m\n",
      "        14           0.9342          138.75m\n",
      "        15           0.9303          136.80m\n",
      "        16           0.9262          135.04m\n",
      "        17           0.9225          133.29m\n",
      "        18           0.9187          131.63m\n",
      "        19           0.9156          129.74m\n",
      "        20           0.9128          127.36m\n",
      "        21           0.9094          125.07m\n",
      "        22           0.9070          122.60m\n",
      "        23           0.9041          120.46m\n",
      "        24           0.9016          118.40m\n",
      "        25           0.8988          116.26m\n",
      "        26           0.8966          114.10m\n",
      "        27           0.8945          112.06m\n",
      "        28           0.8926          110.04m\n",
      "        29           0.8905          108.11m\n",
      "        30           0.8888          106.12m\n",
      "        31           0.8869          104.39m\n",
      "        32           0.8849          102.57m\n",
      "        33           0.8830          100.87m\n",
      "        34           0.8813           99.15m\n",
      "        35           0.8798           97.77m\n",
      "        36           0.8780           96.13m\n",
      "        37           0.8767           94.40m\n",
      "        38           0.8750           92.75m\n",
      "        39           0.8736           91.05m\n",
      "        40           0.8721           89.36m\n",
      "        41           0.8707           87.93m\n",
      "        42           0.8692           86.30m\n",
      "        43           0.8682           84.59m\n",
      "        44           0.8668           83.00m\n",
      "        45           0.8656           81.42m\n",
      "        46           0.8642           79.82m\n",
      "        47           0.8630           78.16m\n",
      "        48           0.8616           76.57m\n",
      "        49           0.8603           74.96m\n",
      "        50           0.8593           73.38m\n",
      "        51           0.8585           71.78m\n",
      "        52           0.8573           70.21m\n",
      "        53           0.8561           68.65m\n",
      "        54           0.8551           67.08m\n",
      "        55           0.8539           65.52m\n",
      "        56           0.8529           63.97m\n",
      "        57           0.8519           62.41m\n",
      "        58           0.8508           60.86m\n",
      "        59           0.8499           59.32m\n",
      "        60           0.8491           57.78m\n",
      "        61           0.8478           56.31m\n",
      "        62           0.8469           54.78m\n",
      "        63           0.8460           53.28m\n",
      "        64           0.8452           51.89m\n",
      "        65           0.8444           50.41m\n",
      "        66           0.8436           48.90m\n",
      "        67           0.8428           47.45m\n",
      "        68           0.8419           45.98m\n",
      "        69           0.8409           44.54m\n",
      "        70           0.8402           43.05m\n",
      "        71           0.8393           41.61m\n",
      "        72           0.8385           40.14m\n",
      "        73           0.8377           38.67m\n",
      "        74           0.8370           37.21m\n",
      "        75           0.8363           35.76m\n",
      "        76           0.8357           34.30m\n",
      "        77           0.8349           32.86m\n",
      "        78           0.8341           31.43m\n",
      "        79           0.8334           29.98m\n",
      "        80           0.8327           28.55m\n",
      "        81           0.8321           27.12m\n",
      "        82           0.8315           25.69m\n",
      "        83           0.8307           24.25m\n",
      "        84           0.8301           22.82m\n",
      "        85           0.8295           21.38m\n",
      "        86           0.8289           19.95m\n",
      "        87           0.8282           18.51m\n",
      "        88           0.8276           17.08m\n",
      "        89           0.8269           15.65m\n",
      "        90           0.8263           14.22m\n",
      "        91           0.8258           12.78m\n",
      "        92           0.8252           11.36m\n",
      "        93           0.8247            9.93m\n",
      "        94           0.8242            8.51m\n",
      "        95           0.8236            7.09m\n",
      "        96           0.8230            5.67m\n",
      "        97           0.8224            4.25m\n",
      "        98           0.8217            2.83m\n",
      "        99           0.8213            1.42m\n",
      "       100           0.8207            0.00s\n",
      "CPU times: user 2h 18min 8s, sys: 4min 46s, total: 2h 22min 54s\n",
      "Wall time: 2h 23min 18s\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "<style>#sk-container-id-2 {color: black;background-color: white;}#sk-container-id-2 pre{padding: 0;}#sk-container-id-2 div.sk-toggleable {background-color: white;}#sk-container-id-2 label.sk-toggleable__label {cursor: pointer;display: block;width: 100%;margin-bottom: 0;padding: 0.3em;box-sizing: border-box;text-align: center;}#sk-container-id-2 label.sk-toggleable__label-arrow:before {content: \"▸\";float: left;margin-right: 0.25em;color: #696969;}#sk-container-id-2 label.sk-toggleable__label-arrow:hover:before {color: black;}#sk-container-id-2 div.sk-estimator:hover label.sk-toggleable__label-arrow:before {color: black;}#sk-container-id-2 div.sk-toggleable__content {max-height: 0;max-width: 0;overflow: hidden;text-align: left;background-color: #f0f8ff;}#sk-container-id-2 div.sk-toggleable__content pre {margin: 0.2em;color: black;border-radius: 0.25em;background-color: #f0f8ff;}#sk-container-id-2 input.sk-toggleable__control:checked~div.sk-toggleable__content {max-height: 200px;max-width: 100%;overflow: auto;}#sk-container-id-2 input.sk-toggleable__control:checked~label.sk-toggleable__label-arrow:before {content: \"▾\";}#sk-container-id-2 div.sk-estimator input.sk-toggleable__control:checked~label.sk-toggleable__label {background-color: #d4ebff;}#sk-container-id-2 div.sk-label input.sk-toggleable__control:checked~label.sk-toggleable__label {background-color: #d4ebff;}#sk-container-id-2 input.sk-hidden--visually {border: 0;clip: rect(1px 1px 1px 1px);clip: rect(1px, 1px, 1px, 1px);height: 1px;margin: -1px;overflow: hidden;padding: 0;position: absolute;width: 1px;}#sk-container-id-2 div.sk-estimator {font-family: monospace;background-color: #f0f8ff;border: 1px dotted black;border-radius: 0.25em;box-sizing: border-box;margin-bottom: 0.5em;}#sk-container-id-2 div.sk-estimator:hover {background-color: #d4ebff;}#sk-container-id-2 div.sk-parallel-item::after {content: \"\";width: 100%;border-bottom: 1px solid gray;flex-grow: 1;}#sk-container-id-2 div.sk-label:hover label.sk-toggleable__label {background-color: #d4ebff;}#sk-container-id-2 div.sk-serial::before {content: \"\";position: absolute;border-left: 1px solid gray;box-sizing: border-box;top: 0;bottom: 0;left: 50%;z-index: 0;}#sk-container-id-2 div.sk-serial {display: flex;flex-direction: column;align-items: center;background-color: white;padding-right: 0.2em;padding-left: 0.2em;position: relative;}#sk-container-id-2 div.sk-item {position: relative;z-index: 1;}#sk-container-id-2 div.sk-parallel {display: flex;align-items: stretch;justify-content: center;background-color: white;position: relative;}#sk-container-id-2 div.sk-item::before, #sk-container-id-2 div.sk-parallel-item::before {content: \"\";position: absolute;border-left: 1px solid gray;box-sizing: border-box;top: 0;bottom: 0;left: 50%;z-index: -1;}#sk-container-id-2 div.sk-parallel-item {display: flex;flex-direction: column;z-index: 1;position: relative;background-color: white;}#sk-container-id-2 div.sk-parallel-item:first-child::after {align-self: flex-end;width: 50%;}#sk-container-id-2 div.sk-parallel-item:last-child::after {align-self: flex-start;width: 50%;}#sk-container-id-2 div.sk-parallel-item:only-child::after {width: 0;}#sk-container-id-2 div.sk-dashed-wrapped {border: 1px dashed gray;margin: 0 0.4em 0.5em 0.4em;box-sizing: border-box;padding-bottom: 0.4em;background-color: white;}#sk-container-id-2 div.sk-label label {font-family: monospace;font-weight: bold;display: inline-block;line-height: 1.2em;}#sk-container-id-2 div.sk-label-container {text-align: center;}#sk-container-id-2 div.sk-container {/* jupyter's `normalize.less` sets `[hidden] { display: none; }` but bootstrap.min.css set `[hidden] { display: none !important; }` so we also need the `!important` here to be able to override the default hidden behavior on the sphinx rendered scikit-learn.org. See: https://github.com/scikit-learn/scikit-learn/issues/21755 */display: inline-block !important;position: relative;}#sk-container-id-2 div.sk-text-repr-fallback {display: none;}</style><div id=\"sk-container-id-2\" class=\"sk-top-container\"><div class=\"sk-text-repr-fallback\"><pre>GradientBoostingClassifier(random_state=42, verbose=6)</pre><b>In a Jupyter environment, please rerun this cell to show the HTML representation or trust the notebook. <br />On GitHub, the HTML representation is unable to render, please try loading this page with nbviewer.org.</b></div><div class=\"sk-container\" hidden><div class=\"sk-item\"><div class=\"sk-estimator sk-toggleable\"><input class=\"sk-toggleable__control sk-hidden--visually\" id=\"sk-estimator-id-2\" type=\"checkbox\" checked><label for=\"sk-estimator-id-2\" class=\"sk-toggleable__label sk-toggleable__label-arrow\">GradientBoostingClassifier</label><div class=\"sk-toggleable__content\"><pre>GradientBoostingClassifier(random_state=42, verbose=6)</pre></div></div></div></div></div>"
      ],
      "text/plain": [
       "GradientBoostingClassifier(random_state=42, verbose=6)"
      ]
     },
     "execution_count": 84,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "%%time\n",
    "#Train the model using the training sets y_pred=clf.predict(X_test)\n",
    "clf = GradientBoostingClassifier(random_state=42, verbose=6)\n",
    "\n",
    "# 2 million\n",
    "samples = 2 * 100 * 100 * 100\n",
    "# 1 million records took 2h 10min 13s\n",
    "# 2 million records took 3h 50 mins\n",
    "# 8 million records will take ~17 hours\n",
    "\n",
    "tr_df = train_df.sample(samples)\n",
    "X_tr = vectorizer.transform(tr_df['full_name'])\n",
    "y_tr = tr_df.race_code\n",
    "\n",
    "clf.fit(X_tr, y_tr)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 85,
   "id": "d915e779-d055-4c74-b129-5864ff3dd982",
   "metadata": {
    "tags": []
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Accuracy: 0.7082735496118031\n",
      "CPU times: user 18.2 s, sys: 1.16 s, total: 19.4 s\n",
      "Wall time: 26.7 s\n"
     ]
    }
   ],
   "source": [
    "%%time\n",
    "y_pred = clf.predict(X_val)\n",
    "# Model Accuracy, how often is the classifier correct?\n",
    "print(\"Accuracy:\", metrics.accuracy_score(y_val, y_pred))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 86,
   "id": "574338bf-b610-4619-bd23-ddc2eec5c899",
   "metadata": {
    "tags": []
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Accuracy: 0.7077767995546991\n",
      "CPU times: user 10.2 s, sys: 957 ms, total: 11.1 s\n",
      "Wall time: 13 s\n"
     ]
    }
   ],
   "source": [
    "%%time\n",
    "y_pred = clf.predict(X_test)\n",
    "# Model Accuracy, how often is the classifier correct?\n",
    "print(\"Accuracy:\", metrics.accuracy_score(y_test, y_pred))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 87,
   "id": "ce7677d6-3723-4d2b-9e88-823ed900dc6b",
   "metadata": {
    "tags": []
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "['models/gb_vec_fullname_2M.joblib']"
      ]
     },
     "execution_count": 87,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "joblib.dump(vectorizer, \"models/gb_vec_fullname_2M.joblib\", compress=3)  # compression is ON!"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 88,
   "id": "ba7d6ce8-ed66-4d5e-86a0-f75a66c35bc2",
   "metadata": {
    "tags": []
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "['models/gb_model_fullname_2M.joblib']"
      ]
     },
     "execution_count": 88,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "joblib.dump(clf, \"models/gb_model_fullname_2M.joblib\", compress=3)  # compression is ON!"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "5f97bc2d-13a9-430d-970d-4929d539f8f5",
   "metadata": {},
   "source": [
    "# Now lets do for the whole train dataset with 20 iterations"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 89,
   "id": "f88669a8-23d8-473f-a377-287a26139330",
   "metadata": {
    "tags": []
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "      Iter       Train Loss   Remaining Time \n",
      "         1           1.0629          119.29m\n",
      "         2           1.0399          108.12m\n",
      "         3           1.0232          100.65m\n",
      "         4           1.0072           94.18m\n",
      "         5           0.9965           88.46m\n",
      "         6           0.9851           82.43m\n",
      "         7           0.9755           76.54m\n",
      "         8           0.9671           70.51m\n",
      "         9           0.9608           64.74m\n",
      "        10           0.9540           58.86m\n",
      "        11           0.9481           52.99m\n",
      "        12           0.9429           46.99m\n",
      "        13           0.9381           41.08m\n",
      "        14           0.9336           35.17m\n",
      "        15           0.9296           29.38m\n",
      "        16           0.9256           23.52m\n",
      "        17           0.9218           17.64m\n",
      "        18           0.9182           11.79m\n",
      "        19           0.9151            5.90m\n",
      "        20           0.9119            0.00s\n",
      "CPU times: user 1h 54min 15s, sys: 4min 28s, total: 1h 58min 43s\n",
      "Wall time: 1h 59min 13s\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "<style>#sk-container-id-3 {color: black;background-color: white;}#sk-container-id-3 pre{padding: 0;}#sk-container-id-3 div.sk-toggleable {background-color: white;}#sk-container-id-3 label.sk-toggleable__label {cursor: pointer;display: block;width: 100%;margin-bottom: 0;padding: 0.3em;box-sizing: border-box;text-align: center;}#sk-container-id-3 label.sk-toggleable__label-arrow:before {content: \"▸\";float: left;margin-right: 0.25em;color: #696969;}#sk-container-id-3 label.sk-toggleable__label-arrow:hover:before {color: black;}#sk-container-id-3 div.sk-estimator:hover label.sk-toggleable__label-arrow:before {color: black;}#sk-container-id-3 div.sk-toggleable__content {max-height: 0;max-width: 0;overflow: hidden;text-align: left;background-color: #f0f8ff;}#sk-container-id-3 div.sk-toggleable__content pre {margin: 0.2em;color: black;border-radius: 0.25em;background-color: #f0f8ff;}#sk-container-id-3 input.sk-toggleable__control:checked~div.sk-toggleable__content {max-height: 200px;max-width: 100%;overflow: auto;}#sk-container-id-3 input.sk-toggleable__control:checked~label.sk-toggleable__label-arrow:before {content: \"▾\";}#sk-container-id-3 div.sk-estimator input.sk-toggleable__control:checked~label.sk-toggleable__label {background-color: #d4ebff;}#sk-container-id-3 div.sk-label input.sk-toggleable__control:checked~label.sk-toggleable__label {background-color: #d4ebff;}#sk-container-id-3 input.sk-hidden--visually {border: 0;clip: rect(1px 1px 1px 1px);clip: rect(1px, 1px, 1px, 1px);height: 1px;margin: -1px;overflow: hidden;padding: 0;position: absolute;width: 1px;}#sk-container-id-3 div.sk-estimator {font-family: monospace;background-color: #f0f8ff;border: 1px dotted black;border-radius: 0.25em;box-sizing: border-box;margin-bottom: 0.5em;}#sk-container-id-3 div.sk-estimator:hover {background-color: #d4ebff;}#sk-container-id-3 div.sk-parallel-item::after {content: \"\";width: 100%;border-bottom: 1px solid gray;flex-grow: 1;}#sk-container-id-3 div.sk-label:hover label.sk-toggleable__label {background-color: #d4ebff;}#sk-container-id-3 div.sk-serial::before {content: \"\";position: absolute;border-left: 1px solid gray;box-sizing: border-box;top: 0;bottom: 0;left: 50%;z-index: 0;}#sk-container-id-3 div.sk-serial {display: flex;flex-direction: column;align-items: center;background-color: white;padding-right: 0.2em;padding-left: 0.2em;position: relative;}#sk-container-id-3 div.sk-item {position: relative;z-index: 1;}#sk-container-id-3 div.sk-parallel {display: flex;align-items: stretch;justify-content: center;background-color: white;position: relative;}#sk-container-id-3 div.sk-item::before, #sk-container-id-3 div.sk-parallel-item::before {content: \"\";position: absolute;border-left: 1px solid gray;box-sizing: border-box;top: 0;bottom: 0;left: 50%;z-index: -1;}#sk-container-id-3 div.sk-parallel-item {display: flex;flex-direction: column;z-index: 1;position: relative;background-color: white;}#sk-container-id-3 div.sk-parallel-item:first-child::after {align-self: flex-end;width: 50%;}#sk-container-id-3 div.sk-parallel-item:last-child::after {align-self: flex-start;width: 50%;}#sk-container-id-3 div.sk-parallel-item:only-child::after {width: 0;}#sk-container-id-3 div.sk-dashed-wrapped {border: 1px dashed gray;margin: 0 0.4em 0.5em 0.4em;box-sizing: border-box;padding-bottom: 0.4em;background-color: white;}#sk-container-id-3 div.sk-label label {font-family: monospace;font-weight: bold;display: inline-block;line-height: 1.2em;}#sk-container-id-3 div.sk-label-container {text-align: center;}#sk-container-id-3 div.sk-container {/* jupyter's `normalize.less` sets `[hidden] { display: none; }` but bootstrap.min.css set `[hidden] { display: none !important; }` so we also need the `!important` here to be able to override the default hidden behavior on the sphinx rendered scikit-learn.org. See: https://github.com/scikit-learn/scikit-learn/issues/21755 */display: inline-block !important;position: relative;}#sk-container-id-3 div.sk-text-repr-fallback {display: none;}</style><div id=\"sk-container-id-3\" class=\"sk-top-container\"><div class=\"sk-text-repr-fallback\"><pre>GradientBoostingClassifier(n_estimators=20, random_state=42, verbose=6)</pre><b>In a Jupyter environment, please rerun this cell to show the HTML representation or trust the notebook. <br />On GitHub, the HTML representation is unable to render, please try loading this page with nbviewer.org.</b></div><div class=\"sk-container\" hidden><div class=\"sk-item\"><div class=\"sk-estimator sk-toggleable\"><input class=\"sk-toggleable__control sk-hidden--visually\" id=\"sk-estimator-id-3\" type=\"checkbox\" checked><label for=\"sk-estimator-id-3\" class=\"sk-toggleable__label sk-toggleable__label-arrow\">GradientBoostingClassifier</label><div class=\"sk-toggleable__content\"><pre>GradientBoostingClassifier(n_estimators=20, random_state=42, verbose=6)</pre></div></div></div></div></div>"
      ],
      "text/plain": [
       "GradientBoostingClassifier(n_estimators=20, random_state=42, verbose=6)"
      ]
     },
     "execution_count": 89,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "%%time\n",
    "\n",
    "clf = GradientBoostingClassifier(n_estimators=20, random_state=42, verbose=6)\n",
    "clf.fit(X_train, y_train)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 90,
   "id": "3e216450-1c30-45cc-85c5-833c05fb2f36",
   "metadata": {
    "tags": []
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Accuracy: 0.6771745566394859\n",
      "CPU times: user 4.21 s, sys: 2.93 s, total: 7.14 s\n",
      "Wall time: 10.3 s\n"
     ]
    }
   ],
   "source": [
    "%%time\n",
    "y_pred = clf.predict(X_val)\n",
    "# Model Accuracy, how often is the classifier correct?\n",
    "print(\"Accuracy:\", metrics.accuracy_score(y_val, y_pred))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 91,
   "id": "b2a64a28-7aae-4156-bb57-76db1774680b",
   "metadata": {
    "tags": []
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Accuracy: 0.6768762848418051\n",
      "CPU times: user 4.33 s, sys: 737 ms, total: 5.07 s\n",
      "Wall time: 7.34 s\n"
     ]
    }
   ],
   "source": [
    "%%time\n",
    "y_pred = clf.predict(X_test)\n",
    "# Model Accuracy, how often is the classifier correct?\n",
    "print(\"Accuracy:\", metrics.accuracy_score(y_test, y_pred))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 92,
   "id": "fce9d2a2-df49-43fe-b477-fa51dc64f01e",
   "metadata": {
    "tags": []
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "['models/gb_vec_fullname.joblib']"
      ]
     },
     "execution_count": 92,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "joblib.dump(vectorizer, \"models/gb_vec_fullname.joblib\", compress=3)  # compression is ON!"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 93,
   "id": "e391f3db-c17b-4f01-bf1d-4317ff363572",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "['models/gb_model_fullname.joblib']"
      ]
     },
     "execution_count": 93,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "joblib.dump(clf, \"models/gb_model_fullname.joblib\", compress=3)  # compression is ON!"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "39450453-38bd-4b6f-a792-6ecd875a106b",
   "metadata": {},
   "source": [
    "## Confusion Matrix"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 94,
   "id": "575f426e-e2c1-40bb-8c88-bf0dcec9bab8",
   "metadata": {
    "tags": []
   },
   "outputs": [],
   "source": [
    "race_id_df = train_df[['race', 'race_code']].drop_duplicates().sort_values('race_code')\n",
    "race_to_id = dict(race_id_df.values)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 95,
   "id": "4d186d8d-ef16-466a-9b94-132d4d79ab6f",
   "metadata": {
    "tags": []
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "{'asian': 0, 'hispanic': 1, 'nh_black': 2, 'nh_white': 3, 'other': 4}"
      ]
     },
     "execution_count": 95,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "race_to_id"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 96,
   "id": "547e0fce-052e-4835-8a2a-fb8435b2b395",
   "metadata": {
    "tags": []
   },
   "outputs": [],
   "source": [
    "id_to_race = dict(race_id_df[['race_code', 'race']].values)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 97,
   "id": "324feaf6-2b00-4a27-9170-681248a23c01",
   "metadata": {
    "tags": []
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "{0: 'asian', 1: 'hispanic', 2: 'nh_black', 3: 'nh_white', 4: 'other'}"
      ]
     },
     "execution_count": 97,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "id_to_race"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "9cf922af-de73-4254-93b2-a4e6e9726e6c",
   "metadata": {
    "tags": []
   },
   "source": [
    "### For 1M model"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 98,
   "id": "f21a2c41-faa1-49ba-b785-5a81e951d36f",
   "metadata": {
    "tags": []
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Defaulting to user installation because normal site-packages is not writeable\n",
      "Requirement already satisfied: matplotlib in /home/distill1994/.local/lib/python3.10/site-packages (3.7.1)\n",
      "Requirement already satisfied: cycler>=0.10 in /home/distill1994/.local/lib/python3.10/site-packages (from matplotlib) (0.11.0)\n",
      "Requirement already satisfied: contourpy>=1.0.1 in /home/distill1994/.local/lib/python3.10/site-packages (from matplotlib) (1.0.7)\n",
      "Requirement already satisfied: python-dateutil>=2.7 in /usr/local/lib/python3.10/dist-packages (from matplotlib) (2.8.2)\n",
      "Requirement already satisfied: fonttools>=4.22.0 in /home/distill1994/.local/lib/python3.10/site-packages (from matplotlib) (4.39.4)\n",
      "Requirement already satisfied: pyparsing>=2.3.1 in /home/distill1994/.local/lib/python3.10/site-packages (from matplotlib) (3.0.9)\n",
      "Requirement already satisfied: packaging>=20.0 in /usr/local/lib/python3.10/dist-packages (from matplotlib) (23.1)\n",
      "Requirement already satisfied: kiwisolver>=1.0.1 in /home/distill1994/.local/lib/python3.10/site-packages (from matplotlib) (1.4.4)\n",
      "Requirement already satisfied: numpy>=1.20 in /home/distill1994/.local/lib/python3.10/site-packages (from matplotlib) (1.24.3)\n",
      "Requirement already satisfied: pillow>=6.2.0 in /home/distill1994/.local/lib/python3.10/site-packages (from matplotlib) (9.3.0)\n",
      "Requirement already satisfied: six>=1.5 in /usr/lib/python3/dist-packages (from python-dateutil>=2.7->matplotlib) (1.16.0)\n"
     ]
    }
   ],
   "source": [
    "!pip install matplotlib"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 99,
   "id": "27ace3e7-d206-448f-8fd0-8baa003bc4a0",
   "metadata": {
    "tags": []
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "              precision    recall  f1-score   support\n",
      "\n",
      "       asian       0.75      0.11      0.20     25756\n",
      "    hispanic       0.78      0.56      0.65    163525\n",
      "    nh_black       0.75      0.09      0.16    133471\n",
      "    nh_white       0.70      0.96      0.81    552737\n",
      "       other       0.38      0.01      0.01     26373\n",
      "\n",
      "    accuracy                           0.71    901862\n",
      "   macro avg       0.67      0.35      0.37    901862\n",
      "weighted avg       0.71      0.71      0.64    901862\n",
      "\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "<sklearn.metrics._plot.confusion_matrix.ConfusionMatrixDisplay at 0x7f24c24fbb80>"
      ]
     },
     "execution_count": 99,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 800x800 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "from sklearn.metrics import confusion_matrix, classification_report, ConfusionMatrixDisplay\n",
    "import matplotlib.pyplot as plt\n",
    "%matplotlib inline  \n",
    "\n",
    "model_1m = joblib.load('models/gb_model_fullname_1M.joblib')\n",
    "preds = model_1m.predict(X_test)\n",
    "predicted = [id_to_race[p] for p in preds]\n",
    "actual = [id_to_race[p] for p in y_test.values]\n",
    "\n",
    "print(classification_report(actual, predicted))\n",
    "\n",
    "cm = confusion_matrix(actual, predicted, labels=list(race_to_id.keys()))\n",
    "cmp = ConfusionMatrixDisplay(cm, display_labels=list(race_to_id.keys()))\n",
    "fig, ax = plt.subplots(figsize=(8,8))\n",
    "cmp.plot(ax=ax, xticks_rotation='vertical')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 100,
   "id": "5ef7ae84-84be-4131-816d-2d801a1b7b62",
   "metadata": {
    "tags": []
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Accuracy for class asian: 0.11\n",
      "Accuracy for class hispanic: 0.56\n",
      "Accuracy for class nh_black: 0.09\n",
      "Accuracy for class nh_white: 0.96\n",
      "Accuracy for class other: 0.01\n"
     ]
    }
   ],
   "source": [
    "for class_id in id_to_race:\n",
    "    correct = np.sum((y_test == class_id) & (preds == class_id))\n",
    "    total = np.sum(y_test == class_id)\n",
    "    print(f'Accuracy for class {id_to_race[class_id]}: {correct / total:.2f}')"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "2d4ffcdb-0dfa-4135-92f0-4ee9bc1cca04",
   "metadata": {
    "tags": []
   },
   "source": [
    "### For 2M Model"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 101,
   "id": "775560d2-5583-4faf-8451-75ab19d09b18",
   "metadata": {
    "tags": []
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "              precision    recall  f1-score   support\n",
      "\n",
      "       asian       0.76      0.11      0.19     25756\n",
      "    hispanic       0.78      0.56      0.65    163525\n",
      "    nh_black       0.77      0.09      0.16    133471\n",
      "    nh_white       0.70      0.96      0.81    552737\n",
      "       other       0.37      0.01      0.02     26373\n",
      "\n",
      "    accuracy                           0.71    901862\n",
      "   macro avg       0.67      0.35      0.36    901862\n",
      "weighted avg       0.71      0.71      0.64    901862\n",
      "\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "<sklearn.metrics._plot.confusion_matrix.ConfusionMatrixDisplay at 0x7f24c6137d90>"
      ]
     },
     "execution_count": 101,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAtsAAAKbCAYAAAAkKEbEAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjcuMSwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy/bCgiHAAAACXBIWXMAAA9hAAAPYQGoP6dpAACpoklEQVR4nOzdeZxN9R/H8dedfd8wM4ZhyL5nCVEoTKVFEUqFtBAKyZKsZUkppH350SIpUSglosVEtuz7MJiV2feZe+/vj8nlZjDKnblzvZ+Px3k83HO+95zvudfc+7mf8/l+j8FsNpsREREREZGrzqmsOyAiIiIi4qgUbIuIiIiI2IiCbRERERERG1GwLSIiIiJiIwq2RURERERsRMG2iIiIiIiNKNgWEREREbERl7LugIiIiIj8d7m5ueTn55fJsd3c3PDw8CiTY9s7BdsiIiIi5Vxubi41qvsQn2gsk+OHhoYSHR2tgLsYCrZFREREyrn8/HziE40c3xqBn2/pVgmnZ5io3uIY+fn5CraLoWBbRERExEH4+Trh5+tc1t2Q8yjYFhEREXEQJsyYMJX6MeXiNBuJiIiIiIiNKLMtIiIi4iCMZhPGUk40G82lm0kvb5TZFhERERGxEWW2RURERBxEUc126aa2VbN9acpsi4iIiIjYiIJtEREREREbURmJiIiIiIMwlfrEf5TBEcsXZbZFRERERGxEmW0RERERB2E0mzGaS3fAYmkfr7xRZltERERExEYUbIuIiIiI2IjKSEREREQchObZtj/KbIuIiIiI2Igy2yIiIiIOwoQZozLbdkWZbRERERERG1FmW0RERMRBqGbb/iizLSIiIiJiIwq2RURERERsRGUkIiIiIg5Cd5C0P8psi4iIiIjYiDLbIiIiIg7C9PdS2seUi1NmW0RERETERhRsi4iIiIjYiMpIRERERByEsQzuIFnaxytvlNkWEREREbERZbZFREREHITRXLSU9jHl4pTZFhERERGxEQXbIiIiIiI2ojISEREREQehebbtjzLbIiIiIlIqJk+ejMFgsFrq1atn2Z6bm8uQIUOoUKECPj4+9OjRg4SEBKt9xMTE0K1bN7y8vAgODua5556jsLDQqs369etp3rw57u7u1KpViwULFlzQlzfffJOIiAg8PDxo3bo1mzdvttpekr6UhIJtEREREQdhwoCxlBcThivqY8OGDYmLi7Msv/32m2XbiBEjWLFiBV9++SUbNmwgNjaW++67z7LdaDTSrVs38vPz2bhxIwsXLmTBggVMnDjR0iY6Oppu3brRqVMnduzYwfDhw3nsscf44YcfLG2++OILRo4cyaRJk9i2bRtNmzYlMjKSxMTEEvelpAxms1ljSEVERETKsfT0dPz9/dm2NwQf39LNpWZmmGjeIIG0tDT8/Pwu2Xby5MksX76cHTt2XLAtLS2NSpUqsWjRInr27AnA/v37qV+/PlFRUbRp04bvv/+eO++8k9jYWEJCQgB45513GDNmDElJSbi5uTFmzBhWrVrF7t27Lfvu06cPqamprF69GoDWrVvTqlUr5s+fD4DJZCI8PJxhw4YxduzYEvWlpJTZFhEREXEQJnPZLFAU8J+/5OXlFdvHQ4cOERYWRs2aNenbty8xMTEAbN26lYKCAjp37mxpW69ePapVq0ZUVBQAUVFRNG7c2BJoA0RGRpKens6ePXssbc7fx9k2Z/eRn5/P1q1brdo4OTnRuXNnS5uS9KWkFGyLiIiIyH8WHh6Ov7+/ZZkxY8YFbVq3bs2CBQtYvXo1b7/9NtHR0dx0001kZGQQHx+Pm5sbAQEBVs8JCQkhPj4egPj4eKtA++z2s9su1SY9PZ2cnBxOnz6N0Wgsts35+7hcX0pKs5GIiIiIyH924sQJqzISd3f3C9rcfvvtln83adKE1q1bU716dZYsWYKnp2ep9LO0KbMtIiIi4iBKe3Dk2QXAz8/Paiku2P6ngIAA6tSpw+HDhwkNDSU/P5/U1FSrNgkJCYSGhgIQGhp6wYwgZx9fro2fnx+enp5UrFgRZ2fnYtucv4/L9aWkFGyLiIiISJnIzMzkyJEjVK5cmRYtWuDq6sratWst2w8cOEBMTAxt27YFoG3btuzatctq1pA1a9bg5+dHgwYNLG3O38fZNmf34ebmRosWLazamEwm1q5da2lTkr6UlMpIRERERBzE+Znm0jxmSY0aNYq77rqL6tWrExsby6RJk3B2duaBBx7A39+fgQMHMnLkSIKCgvDz82PYsGG0bdvWMvtH165dadCgAQ8//DCzZs0iPj6eF154gSFDhlgy6YMGDWL+/PmMHj2aRx99lHXr1rFkyRJWrVpl6cfIkSPp168fLVu25IYbbmDOnDlkZWUxYMAAgBL1paQUbIuIiIhIqTh58iQPPPAAZ86coVKlSrRv354//viDSpUqAfD666/j5OREjx49yMvLIzIykrfeesvyfGdnZ1auXMngwYNp27Yt3t7e9OvXj6lTp1ra1KhRg1WrVjFixAjmzp1L1apV+eCDD4iMjLS06d27N0lJSUycOJH4+HiaNWvG6tWrrQZNXq4vJaV5tkVERETKubPzbG/cU7lM5tm+sWFciebZvhYpsy0iIiLiIExmAyZz6ZaRlPbxyhsNkBQRERERsRFltkVEREQchL0PkLwWKbMtIiIiImIjymyLiIiIOAgjThhLOZdqLNWjlT/KbIuIiIiI2IiCbRERERERG1EZiYiIiIiDMJfB1H9mTf13Scpsi4iIiIjYiDLbIiIiIg5CU//ZH2W2RURERERsRMG2iIiIiIiNqIxERERExEEYzU4YzaU8z7a5VA9X7iizLSIiIiJiI8psi4iIiDgIEwZMpZxLNaHU9qUo2LZDJpOJ2NhYfH19MRg0wldERKQ8MJvNZGRkEBYWhpOTigekiIJtOxQbG0t4eHhZd0NERET+hRMnTlC1atUyObam/rM/CrbtkK+vLwDt6YaLwbWMeyNiZ8y6XGkvnHx9yroL8jdTRmZZd0GAQgr4je8s3+MioGDbLp0tHXExuCrYFrmAgm174WRwK+suyN9M+q6wD39/PKkEVM6nYFtERETEQZTN1H9KglyKqvdFRERERGxEmW0RERERB1E09V/plrGU9vHKG2W2RURERERsRMG2iIiIiIiNqIxERERExEGYcMKoO0jaFWW2RURERERsRJltEREREQehqf/sjzLbIiIiIiI2osy2iIiIiIMw4YRJNdt2RZltEREREREbUbAtIiIiImIjKiMRERERcRBGswGjuXTv6FjaxytvlNkWEREREbERZbZFREREHISxDG5qY9QAyUtSZltERERExEYUbIuIiIiI2IjKSEREREQchMnshKmU7yBp0h0kL0mZbRERERERG1FmW0RERMRBaICk/VFmW0RERETERpTZFhEREXEQJkr/JjOmUj1a+aPMtoiIiIiIjSjYFhERERGxEZWRiIiIiDgIE06YSjmXWtrHK2/06oiIiIiI2Igy2yIiIiIOwmh2wljKN7Up7eOVN3p1RERERERsRMG2iIiIiIiNqIxERERExEGYMGCitOfZLt3jlTfKbIuIiIiI2Igy2yIiIiIOQgMk7Y9eHRERERERG1FmW0RERMRBGHHCWMq51NI+XnmjV0dERERExEYUbIuIiIiI2IjKSEREREQchMlswGQu5an/Svl45Y0y2yIiIiIiNqLMtoiIiIiDMJXBAEmTcreXpFdHRERERMRGFGyLiIiIiNiIykhEREREHITJ7ISplO/oWNrHK28UbF+GwWBg2bJldO/evay7Uup6D02g3e2phNfKIz/Xib1bvPhwehgnj3gAEFI1j4837Sv2uS89GcGvKwMAaNY+g37PxRFRL5fcbCd++jKI/71cGZOxaPRy1etyeXrmSarVzsXb18iZBFd+Xh7Ip6+FYizUCGe4/HsB8PTLJ7i+fQYVQgrIyXZi3xZvPpwWxonz2tRpms2jz8dSu3E2ZrOBAzu8+HBaGEf3egLg6m7i6ZknqN04h2q1c9n0kx9TBtYs9fMtb3oPTaDdHWnW78+0yhe+Pzdl/uP9qcyJw0VtfAMLGTs/hhr1c/ANNJJ2xoWoH/z434zKZGc6l9Wp2Y1eT5ygXdczVK2ZU/Qab/flo1cjOBXtZcOjmnn46Rhuuz8ebz8je7f5Mn9yLWKPe17Q0tXVxOtf/sV19bMYck8zju73sWG/yodGrTO5/6kkajfOpkJoIZMfjSBqtb9le0DFAgaOj6NFhwy8/Y3s/sOHN1+oQmy0ezF7M/PSp9G0uiXjgv2I2DsF25cRFxdHYGBgWXejTDRpk8mKhRU5uMMLZxfoPzaO6YuO8HjHeuTlOJMU60afZg2tnnNH3zP0HJzIn+t8AajZIIcXPz7K4nkhvPJMdSqEFvD0zBM4OZt5/8UqABQWGPjpy0AO7/YiM82Zmg1yGP7KCZyczPxvZlipn7c9utx7AXBopyfrvg4k6ZQrvgFGHno2numfH6FfmwaYTAY8vIxM++wIf/zoz/znq+LsbObhUfFM++wID7VqiLHQgJOTmfxcJ775qBLt70gt25MuR5q0zWLFgrPvj7no/fn8KI93qHve++P19/vjhm9gIQ89m8D0z4/Sr3V9TCYDZhNE/eDHgpdDSTvjQliNPIZOP4VvwElmDqlexmdY9hrfkMaKzypzcJcPzs5m+o88zrQP9/Bkt+aW1/hK9R16nJAqebw2rk6x2+9//BR3PxzL7LF1iD/pwSPPHOelD3fz5B0tKMi3zuQ9Ojqa5EQ3rquf9a/64og8vEwc3ePBD58HMemjY//YambSR8cwFhqYPKAG2ZlO3PdEEjO/OGL1d3PWvY+fxmwuta6Xa0YMGCndRFVpH6+8UbB9GaGhoWXdhTIz/qHrrB7PHl6NJbt2U7tJDrs3+WAyGUhJcrVqc+PtafyyIoDc7KIPyg53pxC9z4PP5hS9jrHH3PlgWhjj3z7Gp6+FkpPlTHyMO/Ex5zIZiafcWLcsk0Y36EvrrMu9FwDff1bRsj3hJCycVZl3fjpASHg+ccfdCa+Vh1+gkY9fDSUp1g2AT18L5d21Bwipmk/sMXfycpx5Y1w4AA1aZeLjZyylMyzfxve1zv7PHl6NJbv3/OP9qWDZnnDSjYUvh/LO2oOW9yczzYWVH597DxNPubFiYQXuH5xUOidh5yY81sjq8Wtj67D4j03UbpjJ7i1FWU5v30IeGxNN21vP4Opm5uBuH96bXoPoA/8my2ym+yOnWPx2OH+sLXrvXh1dh883buLGzmfY8F0lS8uWNyfTvF0q04bVp1WHlH99jo5my89+bPnZr9htVWrm06BlNk90rMvxg0VXd94YW5XFf+2l072prF507u+lZsMcejyZxLDba7P4r72l0neRq+maKLJZvXo17du3JyAggAoVKnDnnXdy5MgRAPLz8xk6dCiVK1fGw8OD6tWrM2PGDMtzDQYDy5cvtzweM2YMderUwcvLi5o1azJhwgQKCgos2ydPnkyzZs345JNPiIiIwN/fnz59+pCRkVFq52sr3n8HXhmpxWeRajXOplajHH5YfO5D0tXNTEGe9X+z/Fwn3D3N1G6SXex+wiLyaNkxnZ1/6DLsxVzuvXD3NNK1dzJxx91Iii36QXTyiDtpyc5E9jmDi6sJNw8Ttz1whuMH3Yk/4VZqfb8W/Jv355+CQgpod3saO6O8bdbP8szLtxCAjLRzOaPn5+4noEIBEx5vyLD7mnFkjw8zFu7Gx7/gYru5qNCqeQQFF7B9Y4BlXXamCwf+8qXe9emWdQEV8nnmxcO8OroOubnXxFfqVeHqZgIgP+9cRtRsNlCQb6Bhq3OJFndPE2PfPM6b46tckNwRKS+uiU+GrKwsRo4cyZYtW1i7di1OTk7ce++9mEwm5s2bx7fffsuSJUs4cOAAn332GRERERfdl6+vLwsWLGDv3r3MnTuX999/n9dff92qzZEjR1i+fDkrV65k5cqVbNiwgZkzZ150n3l5eaSnp1st9sZgMDNoyil2b/bm+IEL6xUBS+C2d8u54GDLel/qt8yi4z0pODmZqRCaT9/h8QAEBRdaPf/1bw6y4shf/O/3feze7MPHr1y7VxUu5VLvxZ39TrP84E6+PbyLVp3SGffAdRQWFP2Z52Q581zPWtx6XwrfHtnJ8oM7adkxgxceus5SPy//3bn3x6v49+fQLr49sptWt2Qwrk9Ny/tz1ti3jvPNkZ18vn0v2ZnOvD4qvDS7Xy4YDGaefP4oe7b6cfxQ0edNwxZp1G2SwfSn63Foty+xxz35YFYNstJduCnyzBUfI7BSPgApZ6x/iKaccSOw4tng3czImYdYtTiUQ7t9/9M5XWtOHPYg4aQrj46Lw8e/EBdXE72GJFIprICgkHM/jp6cfIq9W7yJ+kE12iV1doBkaS9ycddEGUmPHj2sHn/00UdUqlSJvXv3EhMTQ+3atWnfvj0Gg4Hq1S9dG/nCCy9Y/h0REcGoUaNYvHgxo0ePtqw3mUwsWLAAX9+iD9+HH36YtWvXMm3atGL3OWPGDKZMmfJvT69UDJ1+kup1c3j23trFbnfzMNGpewqL5loHyNt+8eODl8J4euYJRs87TkG+E5/NCaFxm6wL6u+mD47A09tEzQY5PDYhlp6DEvny7RBbnVK5dan3Yt3XgWz7xZeg4AJ6Dkpk/DvHGNG9NgV5Trh5mBj56gn2bPFmxpAInJzN9ByUyIsfH2VYtzrkKyt3VQydforq9XJ5tnutC7ZZvT+Dkxj/7nFG3FPL6urPu5PC+Oy1EKrUzOPRcXE8OSmW+c9XLc1TsHtDJh0honY2ox5sYllXo24WHl5Gvtj0h1VbNw8TlavlAEUB+Yvv77Fsc3E1YzBA+8jTlnVvTKrFzyuCS9SPux+Ow8vbyJJ39YPoShkLDUwdGMHI106wdN8ejIWw/VdfNq/1xfD3b/82XdNo1i6Tp7oWX1MvUl5cE8H2oUOHmDhxIps2beL06dOYTEWXr2JiYujfvz9dunShbt263Hbbbdx555107dr1ovv64osvmDdvHkeOHCEzM5PCwkL8/Kxr0iIiIiyBNkDlypVJTEy86D7HjRvHyJEjLY/T09MJD7efD+8hL52kded0nr2vFqfjii83uKlbKu6eZn76MuiCbV+/F8zX71UiKKSQzDRnQqrmM/D5OOKOW484P1tHHHPIAydnM8/MOsHSd4MxmZR1Pety70V2hjPZGc7ERruzf5sXS/fupt1taaz/JpBO3VMICc9n+N21MZuLXtOZQ6qzdO9u2nZNY8O31+ZA4KtpyLSTtO6SzrP3Xley92ffHtrdnsb65ede+5QkV1KSXDlx2IOMVGdeW36ERXNCSE7UJXSAwROOcEPHZJ57qAmnE859hnh6m0hJcmP0w40veE5WRtFX3aHdPgzpfr1l/T0Px1IhJJ+PXo2wrEs9U/Q6pyQVvX+BFfIt/z77+Mj+omx60zap1GuWzre7frc63rylO/h5RTCzxypIvJTDu7x4qktdvHyNuLqaSUt2Ye7KQxzcWXRFqFm7TCpH5PP1/t1Wz5vw/jF2b/JmdM8Lf9AKGCn9AYsa3XNp10Swfdddd1G9enXef/99wsLCMJlMNGrUiPz8fJo3b050dDTff/89P/30E7169aJz58589dVXF+wnKiqKvn37MmXKFCIjI/H392fx4sXMnj3bqp2rq/WXosFgsAT4xXF3d8fdvbipjsqamSEvneLG29J47v5aJJy4eB8j+5zhjzV+pCVf7L+UgeSEotelU/cUEk+5cnhX8eUoAE5O4OJixuAEXPylu4aU/L04y2AADGZc3YteQHdPEyYTVlcUTCYDZnPR6y3/hZkh0/5+f3pe4fvjdvEpFs5m+C7V5tphZvCEo9zY5QxjHm5MwkkPq62H93gTWDEfo9FA4imPYveQn+dMXMy5z52MNBe8fIxW686KP+lOcqIrzdqmWqbx8/IupG7TDFZ9XhmAd16qycdzzl0NrRCcz7SP9jBjRD0O/KWykpLKziga2xBWI4/aTbNZ+HcJ4Rfzg/l+kXUC572fD/Lu5DD++LH4gZci9sjhg+0zZ85w4MAB3n//fW666SYAfvvtN6s2fn5+9O7dm969e9OzZ09uu+02kpOTCQqy/iPfuHEj1atXZ/z48ZZ1x48ft/1JlJGh00/SqXsKkx+tSU6mE4GViurosjKcrUoOwiLyaNwmiwkPFz8fc89BiWxZ74vZBO3uSKPXkESmDapuyVh3ujcZY6GB6H2eFOQbqNM0mwFj49jwbaDm2f7b5d6L0Gp5dLg7la0bfEk740KlsAJ6DUkgP9eJzWuLvpS2/+LL4y/EMnT6Sb75qBJOTmZ6DU3EWAh/bTw3GLVa7Vxc3Ez4Bhjx8jFRs2HRQNaje2w5n3H5NnT6KTrdm8LkATUu//4ku1CpcgG9hiaSn+PE5rVFQVmrW9IJrFTIgR2e5GY5U71uLo9NiGX3Zi8STmoA65BJR+h4ZxJTn2pATpYzgRWLaqqzMpzJz3Nm+8YA9u3wY+Kb+/jolQhOHvOkQnA+N3RIZuNPFf5FTbWB5R9Xoc/gE5w67knCSQ8efuY4ZxLd2PhT0SDwpDjroD7n71mY4mI8rLLu1yoPLyNhNfItj0PD86nZMIeMVGeSTrlx052ppJ1xIfGUKzXq5zJo6imiVvuzbUPRe3X2Ks8/JZ5yK9EP2muVbmpjfxw+2A4MDKRChQq89957VK5cmZiYGMaOHWvZ/tprr1G5cmWuv/56nJyc+PLLLwkNDSUgIOCCfdWuXZuYmBgWL15Mq1atWLVqFcuWLSvFsyldd/UrGlT06tLDVutfHRHOmiXnZhyJ7HOG03GubN1Q/JdZq1vSeeDpeFzdzBzd58nkR2tYTQdlMhro9VQiVWrmYTBA4kk3vl1Qka/fr1Ts/q5Fl3sv8vOcaHRDJvc+loSPv5HU0y7s+sOHEffUJu3vy+InjngwqX9N+o6MZ863BzGbDBze48n4h66zKlF48ZMjhIafG6D09o8HAYis0szGZ1l+3dX/7/fn6yNW618dHs6aJUFF70/rLO59/PR57483I+6pZXl/8nOduL3vGZ6cnIurm5mkWFd+/96fL+Zr3ALAnQ8WDaye9ekuq/Wzx9bmp2UhgIGJTzSg3/DjjJhxCP/AAlJOu7F7ix8pp//dj5Uv36+Ch6eRp6cexsevkD1b/ZjwWKML5tiW4tVpmsMrS8/9TQyaEgvAj18EMntENYJCCnhyciwBFQtJTnThpy8DWTRH/9/F8RjMZsefJv6nn37i6aef5ujRo9StW5d58+bRsWNHli1bRlJSEm+99RaHDh3C2dmZVq1a8corr3D99UV1ff+8g+To0aP56KOPyMvLo1u3brRp04bJkyeTmpoKFE39t3z5cnbs2GE5/pw5c5gzZw7Hjh0rUX/T09Px9/eno6E7LgbVaYpYcfyPrHLDyVelEvbC5ADTyzqCQnMB6/mGtLS0C8Zz2drZ2OGFP7ri4VO6sUNuZgEvtfmxTM67PLgmgu3yRsG2yCXoI8tuKNi2Hwq27YM9BNvjom4rk2B7RtvVCrYvQtfCRERERERsxOFrtkVERESuFWYMmEp56j9zKR+vvFFmW0RERETERhRsi4iIiIjYiMpIRERERByE0eyEsZTnvS7t45U3enVERERERGxEmW0RERERB2EyGzCZS3fAYmkfr7xRZltERERExEaU2RYRERFxEEacMJZyLrW0j1fe6NUREREREbERBdsiIiIiIjaiMhIRERERB6EBkvZHmW0RERERERtRZltERETEQZhwwlTKudTSPl55o1dHRERERMRGFGyLiIiIiNiIykhEREREHITRbMBYygMWS/t45Y0y2yIiIiIiNqLMtoiIiIiD0NR/9keZbRERERERG1FmW0RERMRBmM1OmMylm0s1l/Lxyhu9OiIiIiIiNqJgW0RERETERlRGIiIiIuIgjBgwUspT/5Xy8cobZbZFRERERGxEmW0RERERB2Eyl/5UfCZzqR6u3FFmW0RERETERhRsi4iIiIjYiMpIRERERByEqQzm2S7t45U3enVERERERGxEmW0RERERB2HCgKmUp+Ir7eOVN8psi4iIiIjYiIJtEREREQdhNBvKZPm3Zs6cicFgYPjw4ZZ1ubm5DBkyhAoVKuDj40OPHj1ISEiwel5MTAzdunXDy8uL4OBgnnvuOQoLC63arF+/nubNm+Pu7k6tWrVYsGDBBcd/8803iYiIwMPDg9atW7N582ar7SXpy+Uo2BYRERGRUvfnn3/y7rvv0qRJE6v1I0aMYMWKFXz55Zds2LCB2NhY7rvvPst2o9FIt27dyM/PZ+PGjSxcuJAFCxYwceJES5vo6Gi6detGp06d2LFjB8OHD+exxx7jhx9+sLT54osvGDlyJJMmTWLbtm00bdqUyMhIEhMTS9yXkjCYzWZNRW5n0tPT8ff3p6OhOy4G17Lujoh90UeW3XDy9S3rLsjfTBkZZd0FAQrNBaznG9LS0vDz8yvVY5+NHR5c9yBuPm6leuz8zHwW3bLois47MzOT5s2b89Zbb/HSSy/RrFkz5syZQ1paGpUqVWLRokX07NkTgP3791O/fn2ioqJo06YN33//PXfeeSexsbGEhIQA8M477zBmzBiSkpJwc3NjzJgxrFq1it27d1uO2adPH1JTU1m9ejUArVu3plWrVsyfPx8Ak8lEeHg4w4YNY+zYsSXqS0kosy0iIiLiIM5O/VfaCxQF/OcveXl5F+3nkCFD6NatG507d7Zav3XrVgoKCqzW16tXj2rVqhEVFQVAVFQUjRs3tgTaAJGRkaSnp7Nnzx5Lm3/uOzIy0rKP/Px8tm7datXGycmJzp07W9qUpC8loWBbRERERP6z8PBw/P39LcuMGTOKbbd48WK2bdtW7Pb4+Hjc3NwICAiwWh8SEkJ8fLylzfmB9tntZ7ddqk16ejo5OTmcPn0ao9FYbJvz93G5vpSEpv4TERERcRAmDJj+w4DFf3tMgBMnTliVkbi7u1/Q9sSJEzzzzDOsWbMGDw+PUutjWVJmW0RERET+Mz8/P6uluGB769atJCYm0rx5c1xcXHBxcWHDhg3MmzcPFxcXQkJCyM/PJzU11ep5CQkJhIaGAhAaGnrBjCBnH1+ujZ+fH56enlSsWBFnZ+di25y/j8v1pSQUbIuIiIhIqbj11lvZtWsXO3bssCwtW7akb9++ln+7urqydu1ay3MOHDhATEwMbdu2BaBt27bs2rXLataQNWvW4OfnR4MGDSxtzt/H2TZn9+Hm5kaLFi2s2phMJtauXWtp06JFi8v2pSRURiIiIiLiIMxlcAdJ8xUcz9fXl0aNGlmt8/b2pkKFCpb1AwcOZOTIkQQFBeHn58ewYcNo27atZfaPrl270qBBAx5++GFmzZpFfHw8L7zwAkOGDLFk0wcNGsT8+fMZPXo0jz76KOvWrWPJkiWsWrXKctyRI0fSr18/WrZsyQ033MCcOXPIyspiwIABAPj7+1+2LyWhYNuemc2Apjkra+Ybm5Z1F+Q8ho1/lXUX5KwaVcq6B3LWzv1l3QORq+b111/HycmJHj16kJeXR2RkJG+99ZZlu7OzMytXrmTw4MG0bdsWb29v+vXrx9SpUy1tatSowapVqxgxYgRz586latWqfPDBB0RGRlra9O7dm6SkJCZOnEh8fDzNmjVj9erVVoMmL9eXktA823bIMs8292iebTugYNu+KNi2H05N6pV1F+RvJgXbdsEe5tnu8VM/XL1Ld57tgqx8lnZeWCbnXR6oZltERERExEZURiIiIiLiIM6/yUxpHlMuTq+OiIiIiIiNKNgWEREREbERlZGIiIiIOAiTuQzuIFnKxytvlNkWEREREbERZbZFREREHISpDG5qU9rHK2+U2RYRERERsREF2yIiIiIiNqIyEhEREREHoQGS9keZbRERERERG1FmW0RERMRBKLNtf5TZFhERERGxEWW2RURERByEMtv2R5ltEREREREbUbAtIiIiImIjKiMRERERcRAqI7E/ymyLiIiIiNiIMtsiIiIiDsIMmCjdTLO5VI9W/iizLSIiIiJiIwq2RURERERsRGUkIiIiIg5CAyTtjzLbIiIiIiI2osy2iIiIiINQZtv+KLMtIiIiImIjCrZFRERERGxEZSQiIiIiDkJlJPZHmW0RERERERtRZltERETEQSizbX+U2RYRERERsRFltkVEREQchNlswFzKmebSPl55o8y2iIiIiIiNKNgWEREREbERlZGIiIiIOAgTBkyU8gDJUj5eeaPMtoiIiIiIjSizLSIiIuIgNPWf/VFmW0RERETERhRsi4iIiIjYiMpIRERERByE5tm2P8psi4iIiIjYiDLbIiIiIg5CAyTtjzLbIiIiIiI2osy2iIiIiINQzbb9UWZbRERERMRG7C6z3bFjR5o1a8acOXOK3W4wGFi2bBndu3cv1X5dyuX67MicnMw89Gw8t/ZIJbBSAWcSXFmzJIhFc4Lh79u3trs9lW6PnKF24xz8gowM7lKHo3s8y7bjds7To4B+fXbQrnUMAX65HD4WxNsfteLgkYoAtGt9nDu7HqR2zTP4+eYzaNSdHD0WZLWPwIAcHn94K82bxOLlWciJWD8+X9qY3zZVt7T5+K2lhAZnWT3vw0+v54vljQFo0jCe++7cS91aZ/D2LOBUnC9fftuQdb/WtPErUP5UCC1g4PhYWnXKwN3TROwxd2aPCOfQTi8APLyMDBwfR9vIdPwCC4k/4cY3H1Zk1ScVLfuY9dVhmt5o/X6s+rgC88ZWLdVzsVcLFq4gJCT7gvUrVtTirTdb4Opq5PEndtChQwyuria2bg3lzfktSE31sLStU+cMAwbspFbtFMxmOHiwAh9+0ITo6EBLm+Yt4nj4od1Uq55OQb4zu3ZX4v33m5GY4G1p06nTMXrev5+wsEyys13Z8mdlPvigKRkZ7rZ9Ecq5Rq0zuf+pJGo3zqZCaCGTH40garW/ZXtAxQIGjo+jRYcMvP2N7P7DhzdfqEJstF5XKb/sLti+nLi4OAIDAy/fsBR9/fXXuLq6lnU3ykSvIYnc2e8Mrz5TjeMHPKjdNJtnXz9BVoYT33xYCQAPLxN7Nnvzy4oARrx6sox7XD6MGLyRiGqpzJrXnjMpntx681FenriGx0bcw5lkLzzcC9m9L5gNGyMYOTiq2H2MHvYb3l75THr5FtLS3bnlpmjGj/yFoWPv4Eh0BUu7hYub8d1PtS2Pc3LOfSw0qJtE9PFAlixvREqqJ61bnOS5ob+Tle3Gpq0KAM/y8S/ktW8OsXOjDy88VJPUM85UqZlPZpqzpc2Tk2Np1i6TWcOqkXDCjeYdMhg24yRnElz548dzwcZ3nwbx8Suhlsd5OboAedYzT3fByclseVw9Io0ZMzbw66/hADz55HZa3RDH9Gk3kpXlylNDtvHChN8Y9WxnADw8CnjxpV/4448w5r/ZAmdnMw8/tJuXpv3CIw/fhdHoREhIJpMm/cbXX9dl1qy2eHnn8+QTO5gw4TeGDY0EoEGDJJ4dtZn33mvGpj/CqFgxh6HDtvDM8D956cX2pf/ClCMeXiaO7vHgh8+DmPTRsX9sNTPpo2MYCw1MHlCD7Ewn7nsiiZlfHOHxDnXJy3EubpfyD+YyGCCpMpJLK3fBdmho6OUblbKgoKDLN3JQDVpmEfWDP5vX+gGQcNKNTt1TqdvsXPZp7dKi1yekan6Z9LG8cXMr5KY2MUx6uRO79oUA8MmSZrRpcZK7uh5gweLrWfvLdQCEVMq86H4a1Eli3vutOXC4KHO6aGkT7rtzL7VrJlsF29k5LqSkFn+lYfHXja0eL/+uPi2axtKu9XEF2+fpNSSR07FuzB5RzbIu4YR1Jq5By2zWfBnEzigfAL7/rALdHj5D3WbZVsF2Xo4TKUnX5o/3y0lL87B63KvXPmJjfdi1sxJeXvl0jYxm1stt+Ouvor+b12bfwPsffE+9eqfZv78i4eEZ+Pnl88nHjTl9uuiKw2efNeTtd34gODiLuDhfatdOwcnJzMcLG1sCiKVL6zJx0m84O5swGp2oX/8MiQlefPtNHQASEnz4/rvruP/+/aX4apRPW372Y8vPfsVuq1IznwYts3miY12OHyx6r98YW5XFf+2l072prF5Uodjnidg7u0yZmEwmRo8eTVBQEKGhoUyePNmyzWAwsHz5cgDy8/MZOnQolStXxsPDg+rVqzNjxgyrtm+//Ta33347np6e1KxZk6+++srqWGPGjKFOnTp4eXlRs2ZNJkyYQEFBgWX75MmTadasGZ988gkRERH4+/vTp08fMjIyLG06duzI8OHDLY/z8vIYM2YM4eHhuLu7U6tWLT788MOr+yLZib1bvGnWPoMqNfMAqNkgh4Y3ZPHnuuI/TOXynJ3MODubyS+wzuLk5TvTsH5iifez92AlOrQ7hq9PHgaDmY7tonFzNbFzT4hVu97dd/PV/xbz1isruP/u3Tg5mS65X2+vfDIydUn3fG26pnPwL0/Gv3uML3bu4c0fD3D7g2es2uzd4kWbrmlUCC0AzDS9MZMqNfPYusHXql2n+1JYsns37647wIBxcbh7Xvr9uFa5uBjpdMtxfvyhBmCgdu0UXF1NbN9+7v/3yZN+JCR4Ua/+mb8f+5KW5kbkbUdxcTHi5lZIZORRYo77kfB3icihQ4GYTQa6dI3GycmEl1c+t9x6nB3bQzAai74y9+2rQMVKObRqFQuYCQjIpX37k/z5Z+XSfhkciqtb0f/1/LxzWVKz2UBBvoGGrbIu9jT5BzNgNpfyUtYnbefsMrO9cOFCRo4cyaZNm4iKiqJ///60a9eOLl26WLWbN28e3377LUuWLKFatWqcOHGCEydOWLWZMGECM2fOZO7cuXzyySf06dOHXbt2Ub9+fQB8fX1ZsGABYWFh7Nq1i8cffxxfX19Gjx5t2ceRI0dYvnw5K1euJCUlhV69ejFz5kymTZtWbP8feeQRoqKimDdvHk2bNiU6OprTp09f9Hzz8vLIy8uzPE5PT7/i16ysfDE/GC9fIx/8sh+TEZycYcHMUH5eZl+lPuVJTq4rew5Uom/PncSc9Cc1zYNO7Y5Rv85pYuN9L7+Dv700uwPjR25g6YIvKCw0kJfnwpRXOhIbf+6H0Dff1edQdBAZme40qJvIow9uJygwh3cXtip2nze3PUadWmeY+27b/3yejqRytXzufOQMX79XicVvBFOnaQ6DXzxFQYGBn74surLz1gtVeGbWSRZt20thAZhMBuY+V5Xdm3ws+/l5WSCJJ105k+BKjfq5DBwfR9Xr8njxsYgyOjP71bbtKXx8ClizpgYAgYG5FOQ7kZXlZtUuNdWDoMBcAHJyXBkzuhMTJ/3OAw/sBSA21ocXxnfAZCoKpBMSfBg/vgPjnt/I009vwdnZzN69FZg44WbLPvfurcSsWW0YOy4KNzcjLi5m/vgjjDffbFEap+6wThz2IOGkK4+Oi2PumKrkZjtx3xOnqRRWQFBIweV3IGKn7DLYbtKkCZMmTQKgdu3azJ8/n7Vr114QbMfExFC7dm3at2+PwWCgevXqF+zr/vvv57HHHgPgxRdfZM2aNbzxxhu89dZbALzwwguWthEREYwaNYrFixdbBdsmk4kFCxbg61sU6Dz88MOsXbu22GD74MGDLFmyhDVr1tC5c1GdYM2alx5MNmPGDKZMmXLZ18Ue3Xx3Krfcl8rMIUU129c1zGHQlFjOJLhaggy5crPmtefZpzay+P2vMBoNHDoaxPrfI6hdM7nE++jXZzs+3gWMntKF9HR3brzhBONHbmDkhNs4FlP0Y2jpygaW9tHHAyksdOKZJ/7go8+aU1BonVlv2jCeUUM2Muedthw/GXBVztNRGJzg0E5P/jezKLN5ZLcXEfVy6fbwGcvfwT2PnqZei2wm9osg8aQbjdtkMWT6Kc4kuLL916LPlu8/O3eZ/Nh+T5ITXZj15VEqV88j7riuJpwv8rZotvxZmeTkkg+2dnMrZPiIP9m7pyIvz2yDk5OZHj0OMGXqLzzzdBfy810IDMzh6Wf+5KefarBhfTU8PQt4+JHdjH9hI8+P6wAYqFYtjUGDtrFoUUO2bg0lKCiHxx77i2FPb2HO6zfY7qQdnLHQwNSBEYx87QRL9+3BWAjbf/Vl81pfDCoJlnLMboPt81WuXJnExAsvn/fv358uXbpQt25dbrvtNu688066du1q1aZt27YXPN6xY4fl8RdffMG8efM4cuQImZmZFBYW4udnXQIRERFhCbQv1R+AHTt24OzsTIcOHUp0rgDjxo1j5MiRlsfp6emEh4eX+Pll6fEJcXwxP5gN3xQFb8f2exJctYA+wxIVbP8HcQm+jJoUiYd7AV6eBSSnevH8iA3EJfhc/slA5ZAMut9xgMeH320JjI8eD6JR/QTuvu0A895rU+zz9h+shIuLmZDgTE7Gnqsjbtwgnqlj1/HOgpb8tOG6/3x+jiY50cVSY3rWiUPutL8jFQA3DxP9x8YzdWCEZXxD9D5PajbMoeegJEuw/U/7txXVFYdFKNg+X3BwFs2aJfDSi+0s61JSPHB1M+HtnW+V3Q4IyCU5pei96dgphpCQLEaO6Gypx3755UC+/GoZbdvGsmFDNe686zDZ2a589GFTyz5emdWGTz5dQb16Z9i/vyK9eu9j796KLP2qHgDHogN4M9eFV2evY+HCxqRcwQ8AsXZ4lxdPdamLl68RV1czackuzF15iIM79ZqWlAkDBkr5DpKlfLzyxi5rtv85s4fBYMBkurBusXnz5kRHR/Piiy+Sk5NDr1696NmzZ4mPExUVRd++fbnjjjtYuXIl27dvZ/z48eTnWw/kK2l/ADw9r/wDwd3dHT8/P6ulvHD3MGH+x0thMoLBoAquqyE3z5XkVC98vPNo2SyWqD9L9iPM3b0QANM/3gaTyYDTJd6b62okYzQaSD1vIFqThvG8NG4dH3zWnO9+qnPlJ3EN2PunN+HX5Vmtq1Izj8RTRUGfi4sZVzcz//zYMBnB4HSJ96NRUflDcqIGTJ6vS9do0tLc2bz5XI30oUOBFBQ40axZgmVdlarphIRks39f0RUDD/fCv2/4cW5fJlPRDUDOfma5uxdiMlkHDsa/H5/Nrrq7Gy/e5uqc4jUvO8OZtGQXwmrkUbtpNlE/+F/+SSJ2yi4z21fCz8+P3r1707t3b3r27Mltt91GcnKyZYaQP/74g0ceecTS/o8//uD6668HYOPGjVSvXp3x48dbth8/fvw/9adx48aYTCY2bNhgKSNxZH+s8aPP04kknnIrKiNplMN9Tybx4+JzWW3fgEIqVSmgwt81d+HXFQUQKYkumnXhIlo0PYXBACdj/QgLzeDxh7dy4pQ/P/xcCwBfnzwqVcyiQmDRrC/hYWkApKR6kpLqyYlT/pyK82X4k3/w3sctSc9w58YbYmjeJI4JM24BoH6dJOrVTuKv3aFk57jSoG4Sg/pvYd2vNcjMKsqiNm0Yz4vj1rHsu3r89kd1AgNyACgsdNIgyfN8/V4lXv/2EH2GJfDLigDqXp/NHQ8lM+e5ohlbsjOd+WujN49PiCM/14mEk640aZtF554pvDclDIDK1fPodG8qm9f6kpHiQo0GOTw5OZadUd5E71NW7yyDwUyXLtH8tCbCUmcNkJ3txo8/1ODxJ3aQkeFGdrYrg5/axt69Fdi/v2hGnm3bQhn42F8MGbKVb7+tg8HJTK9e+zAaDfy1MxiAPzeHce+9B3nwwT2sX18NT68C+vffRUKCF0eOBACwaVMYzzzzJ926HbaUkTw5aDv79wddUVnLtcjDy0hYjXMJrdDwfGo2zCEj1ZmkU27cdGcqaWdcSDxVNG5h0NRTRK32Z9uGko9XudbpDpL2p1wH26+99hqVK1fm+uuvx8nJiS+//JLQ0FACAgIsbb788ktatmxJ+/bt+eyzz9i8ebNlZpDatWsTExPD4sWLadWqFatWrWLZsmX/qU8RERH069ePRx991DJA8vjx4yQmJtKrV6//tG979NYLVeg3Op6hM04SUKGQMwmufPdJBT57/dyMAG26pjNqzrmBq8+/EwPAJ7ND+HS2/U3laA+8vQp4tO82KlbIJiPTnd/+qMb/Pr/eMhtCm5YneG7oRkv78SN/BeCTJU34ZEkzjEYnxk+7lYEPbWPq2HV4ehRyKt6XV+a348/tRQFgQYETHdsd4+Fef+HqYiI+0YevV9Zn6YpzddxdOh7Bw6OQB+7bzQP37bas/2tPCM9NiiyNl6JcOPiXF1MH1mDAuDj6jkgg/oQb70wMsxooPGNwdR59Po4x84/jG2Ak8ZQbC16uzMqPi7KuhQUGrr8pg3sfS8LDy0RSrCu/fefP53NCLnbYa9L11ycQEpLNjz9eOBbm3Xevx2Q28MKEjbi6Gi03tTnr5Ek/Jk+6ib4P7eG113/CbDZw5HAAE1642VL68ddfIcx6uS09799Hz/v3k5fnzL59FXhhfAfy84u+Mn9aUwMvzwLuuvsQjz2+g6wsV/76K4SPPmxyQZ/EWp2mObyy9Ijl8aApsQD8+EUgs0dUIyikgCcnxxJQsZDkRBd++jKQRfobkHLOYDab7ep6f3F3Y+zevTsBAQEsWLDA6g6S77//Pm+99RaHDh3C2dmZVq1a8corr1gy1waDgTfffJPly5fzyy+/ULlyZV5++WWroHf06NF89NFH5OXl0a1bN9q0acPkyZNJTU0Fiqb+W758uVWd95w5c5gzZw7Hjh0rts+5ubk8//zzLF68mDNnzlCtWjWef/55BgwYUKLXID09HX9/fzpyDy4GZX7LmvnGppdvJKXGsPGvsu6C/M2pSb2y7oL8zbRTc3zbg0JzAev5hrS0tFIvCT0bOzRa8hzOXqV75dGYncfuXq+UyXmXB3YXbF9N9nhr95JQsG1fFGzbFwXb9kPBtv1QsG0fFGwr2C6OXQ6QFBERERFxBOW6ZltEREREzjl7V8fSPqZcnEMH2w5cISMiIiIi5YBDB9siIiIi1xJN/Wd/VLMtIiIiImIjCrZFRERERGxEZSQiIiIiDkJlJPZHmW0RERERERtRZltERETEQZjMBgylnGk2KbN9Scpsi4iIiIjYiDLbIiIiIg5CN7WxP8psi4iIiIjYiIJtEREREREbURmJiIiIiIMoKiMp7an/SvVw5Y4y2yIiIiIiNqLMtoiIiIiD0E1t7I8y2yIiIiIiNqJgW0RERETERlRGIiIiIuIgzH8vpX1MuThltkVEREREbESZbREREREHoQGS9keZbRERERERG1FmW0RERMRRqGjb7iizLSIiIiJiIwq2RURERERsRGUkIiIiIo6iDAZIogGSl6TMtoiIiIiIjSizLSIiIuIgzOaipbSPKRenzLaIiIiIiI0o2BYRERERsRGVkYiIiIg4CN1B0v4osy0iIiIiYiPKbIuIiIg4CrOh9KfiU2b7kpTZFhERERGxEWW2RURERByEpv6zP8psi4iIiIjYiIJtEREREREbURmJiIiIiKMw/72U9jHlopTZFhERERGxEWW2RURERByEbmpjf5TZFhERERGxEQXbIiIiIiI2ojISkcswRO0s6y7IeQzu7mXdBflbThXfsu6C/M1dH1NyPg1YtCvKbIuIiIiI2Igy2yIiIiIOQgMk7Y8y2yIiIiIiNqJgW0RERETERlRGIiIiIuIodAdJu6PMtoiIiIiIjSjYFhEREXEYhjJaSubtt9+mSZMm+Pn54efnR9u2bfn+++8t23NzcxkyZAgVKlTAx8eHHj16kJCQYLWPmJgYunXrhpeXF8HBwTz33HMUFhZatVm/fj3NmzfH3d2dWrVqsWDBggv68uabbxIREYGHhwetW7dm8+bNVttL0peSULAtIiIiIqWiatWqzJw5k61bt7JlyxZuueUW7rnnHvbs2QPAiBEjWLFiBV9++SUbNmwgNjaW++67z/J8o9FIt27dyM/PZ+PGjSxcuJAFCxYwceJES5vo6Gi6detGp06d2LFjB8OHD+exxx7jhx9+sLT54osvGDlyJJMmTWLbtm00bdqUyMhIEhMTLW0u15eSMpjNZlXa2Jn09HT8/f3pyD24GFzLujti0JRG9sTg5lbWXZC/5d7SpKy7IH9z//7Psu6CAIXmAtbzDWlpafj5+ZXqsc/GDuFvT8bJ06NUj23KyeXE4Mn/+ryDgoJ45ZVX6NmzJ5UqVWLRokX07NkTgP3791O/fn2ioqJo06YN33//PXfeeSexsbGEhIQA8M477zBmzBiSkpJwc3NjzJgxrFq1it27d1uO0adPH1JTU1m9ejUArVu3plWrVsyfP7/oHEwmwsPDGTZsGGPHjiUtLe2yfSkpZbZFRERE5D9LT0+3WvLy8i7Z3mg0snjxYrKysmjbti1bt26loKCAzp07W9rUq1ePatWqERUVBUBUVBSNGze2BNoAkZGRpKenW7LjUVFRVvs42+bsPvLz89m6datVGycnJzp37mxpU5K+lJSCbRERERH5z8LDw/H397csM2bMKLbdrl278PHxwd3dnUGDBrFs2TIaNGhAfHw8bm5uBAQEWLUPCQkhPj4egPj4eKtA++z2s9su1SY9PZ2cnBxOnz6N0Wgsts35+7hcX0pKU/+JiIiIOIoynPrvxIkTVmUk7u7uxTavW7cuO3bsIC0tja+++op+/fqxYcOG0uhpmVCwLSIiIiL/2dkZRi7Hzc2NWrVqAdCiRQv+/PNP5s6dS+/evcnPzyc1NdUqo5yQkEBoaCgAoaGhF8wacnaGkPPb/HPWkISEBPz8/PD09MTZ2RlnZ+di25y/j8v1paRURiIiIiLiKMyGsln+A5PJRF5eHi1atMDV1ZW1a9dath04cICYmBjatm0LQNu2bdm1a5fVrCFr1qzBz8+PBg0aWNqcv4+zbc7uw83NjRYtWli1MZlMrF271tKmJH0pKWW2RURERKRUjBs3jttvv51q1aqRkZHBokWLWL9+PT/88AP+/v4MHDiQkSNHEhQUhJ+fH8OGDaNt27aW2T+6du1KgwYNePjhh5k1axbx8fG88MILDBkyxFK2MmjQIObPn8/o0aN59NFHWbduHUuWLGHVqlWWfowcOZJ+/frRsmVLbrjhBubMmUNWVhYDBgwAKFFfSkrBtoiIiIiUisTERB555BHi4uLw9/enSZMm/PDDD3Tp0gWA119/HScnJ3r06EFeXh6RkZG89dZbluc7OzuzcuVKBg8eTNu2bfH29qZfv35MnTrV0qZGjRqsWrWKESNGMHfuXKpWrcoHH3xAZGSkpU3v3r1JSkpi4sSJxMfH06xZM1avXm01aPJyfSkpzbNthzTPtp3RPNt2RfNs2w/Ns20/NM+2fbCHebarzp9SJvNsnxw6qUzOuzxQzbaIiIiIiI2UqIzk22+/LfEO77777n/dGRERERH5D8pw6j8pXomC7e7du5doZwaDAaPR+F/6IyIiIiLiMEoUbJtMJlv3Q0RERET+q6swFd+/OqZc1H+q2c7Nzb1a/RARERERcThXHGwbjUZefPFFqlSpgo+PD0ePHgVgwoQJfPjhh1e9gyIiIiIi5dUVB9vTpk1jwYIFzJo1C7fzpuBq1KgRH3zwwVXtnIiIiIiUnMFcNotc3BUH2x9//DHvvfceffv2xdnZ2bK+adOm7N+//6p2TkRERESkPLviO0ieOnWKWrVqXbDeZDJRUFBwVTolIiIiIv+Cpv6zO1ec2W7QoAG//vrrBeu/+uorrr/++qvSKRERERERR3DFme2JEyfSr18/Tp06hclk4uuvv+bAgQN8/PHHrFy50hZ9FBEREREpl644s33PPfewYsUKfvrpJ7y9vZk4cSL79u1jxYoVdOnSxRZ9FBEREZGSODvPdmkvclFXnNkGuOmmm1izZs3V7ouIiIiIiEP5V8E2wJYtW9i3bx9QVMfdokWLq9YpEREREfkXNEDS7lxxsH3y5EkeeOABfv/9dwICAgBITU3lxhtvZPHixVStWvVq91FEREREpFy64prtxx57jIKCAvbt20dycjLJycns27cPk8nEY489Zos+ioiIiEhJmMtokYu64sz2hg0b2LhxI3Xr1rWsq1u3Lm+88QY33XTTVe2ciIiIiEh5dsWZ7fDw8GJvXmM0GgkLC7sqnRIRERERcQRXHGy/8sorDBs2jC1btljWbdmyhWeeeYZXX331qnZORERERK6AykjsTonKSAIDAzEYzs2hmJWVRevWrXFxKXp6YWEhLi4uPProo3Tv3t0mHRURERERKW9KFGzPmTPHxt0QERERkf+sLG4yo5vaXFKJgu1+/frZuh8iIiIiIg7nX9/UBiA3N5f8/HyrdX5+fv+pQyIiIiIijuKKB0hmZWUxdOhQgoOD8fb2JjAw0GoRERERkbJhMJfNIhd3xcH26NGjWbduHW+//Tbu7u588MEHTJkyhbCwMD7++GNb9FFEREREpFy64jKSFStW8PHHH9OxY0cGDBjATTfdRK1atahevTqfffYZffv2tUU/RURERORyymIqPmW2L+mKM9vJycnUrFkTKKrPTk5OBqB9+/b88ssvV7d3IiIiIiLl2BVntmvWrEl0dDTVqlWjXr16LFmyhBtuuIEVK1YQEBBggy6KPWnUOpP7n0qiduNsKoQWMvnRCKJW+1u2P/RsPB3vSaVSWAEF+QYO7/LkfzNDObDd29LmgacTuKFzOjUb5lCYb6BH/cZlcSrlSu+hCbS7PZXwWnnk5zqxd4sXH04P4+QRD0ubWV8eoumNWVbPW/VJBeaNDbda16XXGe57PImqNfPIznTml5UBvDm+qmX7zXel0GdYAlVq5pF2xoVv/1eJr94Jtu0J2rFGN6TT84l4ajfKokJIAVOeqE3UmnPjU9pFJnNH30RqN8rCL9DIU3c05Oi+c//fQ6rksfC3v4rd97Qhtfj1uyAABk86ToMWGVSvk8OJI54M6dbIqu3F9jP83gbs3+FzNU7V7jSpE0efyJ3UiThDxYBsXpjfmd+2R1i239Q8mrs77qdO9dP4++Tx2OR7OXyigtU+wiqlM7jXJhrXTsDVxcjm3VWZt6gtKelelja+3rk8/WAUNzaNwWw2sGFrBPM/b0tOnisAbi6FjHzkd+pUP031yqlE7azGC/O7XLTfjWrFM3f0KqJPBfLYlPuu7oviQHoNTWDg8/Ese78i70yqgm9AIQ+Piqd5h0yCw/JJS3Zh42p/Fs4KJTvDuay7K/KvXXFme8CAAfz1V9EH/tixY3nzzTfx8PBgxIgRPPfcc1e9gwaDgeXLl/+r565fvx6DwUBqaupF2yxYsOCq/kgoyTHLMw8vE0f3eDD/+arFbj911J03x1fhyVvq8Gz3WsSfcGPG50fxDyq0tHFxM/PLigBWLaxYWt0u95q0yWTFwooMv6s24x64DmdXmL7oCO6eRqt2331agT7NGlqWD14Ks9p+3xOJ9B8dz5I3Q3jilnqM7XMdW9f7Wra37JTOmDeOs+qTijx5Sz3mP1+V+x5P5O7+SaVynvbIw9NE9D4v3pxYvfjtXib2/OnLRy+HF7s9Kc6NB1o1s1o+fq0K2ZlO/Lne36rtj19W4pdVQZfsz9i+da32dWi31yXbl2ceboUcOVmBOZ/eWPx290J2HQrhva9aXeT5Bbwy8nvMZhjxyh0MnXEXri4mpg9bg+G8EV0vPL6eGmEpjJp9O+PmdqVpnXiefeQ3y3YnJzN5+c4sXduQrXurXLLPPp55jBu4ga37wi7Z7lpXp2k23R5K5uiecwmDoJACKoQU8v7Uyjx5S11eHR5Oy47pjJx9ogx7KvLfXXFme8SIEZZ/d+7cmf3797N161Zq1apFkyZNrmrnxP5s+dmPLT9ffHrHn5dZz0jz3uQwbn8wmRoNctjxW1FQ98mroQB06ZVsu446mPEPXWf1ePbwaizZtZvaTXLYvelcVjMv10BKkmux+/DxL6Tf6Dgm9a9peS8Aovd5Wv7duUcyG3/wZ9UnRT+E4mPcWTw/n15DEvl2QUXg2rtxwZYNAWzZEHDR7WuXFb1WIVXyit1uMhlIOe1mte7GyBR+XRVEbva5bN3bU4qCef+gAmrUy7no8dJTXC7Yn6PavDuczbuL/xEDsCaqNgChFTKK3d6odgKhFTN5fMq9ZOcWvWYzPuzAinkf07xeLFv3VaFa5RRaNz7Jk1Pv4cDxSgDMW9SWmc/8wNtf3sCZVG9y8115/dP2ADSulYCPV36xxwMY+cjvrN10HSaTgfbXH/9X5+3oPLyMjJl/nDnPVeWBZxIs648f8OTFxyMsj+OOu7Pg5cqMfiMGJ2czJuO19/kjjuGKM9v/VL16de677z4F2nIBF1cTdzx0hsw0J47u9bz8E6TEvP2KMtoZqdaXVjvdm8KSXbt4d+1+BoyNxd3DZNnW/OYMnAxQMbSA99fv49Mtexj/zjEqhZ0LHFzdzOTnWX8s5Oc6USmsgJCqFw8wpORqNcqiVsNsVi+p9K+eP/n9Qyz+cxuzl+ylTeeUq9w7x+LqYgQzFBSe+zvJL3DGbDbQuHY8AA2vSyQjy80SaANs3VsFs9lA/RpXdkXntnYHqVwxg4XfNr86J+Cghk4/xea1fmz/1feybb39jGRnOinQvgIGymDqv7I+aTtXosz2vHnzSrzDp59++oo60LFjR5o0aYKHhwcffPABbm5uDBo0iMmTJ1vanD59mnvvvZcffviBKlWqMHv2bO6+++4SH+P3339n3LhxHDx4kGbNmvHBBx/QqFGjYtseOXKEkSNH8scff5CVlUX9+vWZMWMGnTt3trTJy8tj4sSJLFq0iMTERMLDwxk3bhwDBw68YH/Z2dn06NGD9PR0Vq1aVWzJSl5eHnl557Ji6enpJT43e9S6czrj3j6Ou6eJ5AQXxvW5jvTk/3T/JDmPwWBm0JRT7N7szfED537E/Lw8kMSTbpxJcKVG/RwGjo+j6nV5vPh4DQBCq+VjcII+wxJ4e2IVsjKc6T86jhmfH2FQ57oUFjixZYMvgybHsqZ9Bn/97kNYjTx6PJkIQFBIIQkn3cvknB1JZK8kjh/yYN+2ywca58vJduK9l8LZs9UXswna3ZbCxHcPMfXJ2vzxk+5xUJy9R4LJyXPhyZ6bef/rVhgw80TPP3F2NhPkX3T1IMgvh5QM62SA0eREepY7Qf7ZJT5WleA0nuixmadfvguj6T/nsRxWh3tSqNU4h2F31L5sW7+gQh4cnsD3n1a4bFsRe1aiCOj1118v0c4MBsMVB9sACxcuZOTIkWzatImoqCj69+9Pu3bt6NKlaADKlClTmDVrFq+88gpvvPEGffv25fjx4wQFXbq28aznnnuOuXPnEhoayvPPP89dd93FwYMHcXW98HJ7ZmYmd9xxB9OmTcPd3Z2PP/6Yu+66iwMHDlCtWjUAHnnkEaKiopg3bx5NmzYlOjqa06dPX7Cv1NRUunXrho+PD2vWrMHLq/jayhkzZjBlypSSvlx2b8fv3jzVpQ5+QYXc3jeZ8e8e5+lutUg7U3x5g1yZodNPUr1uDs/ea/1l9f1n52rgj+33JDnRlVlLjlC5eh5xx91xcirKXL81oQrbfikqBZrxVASf79hN0xsz2brBj+8/q0BY9XymLjiKi6uZ7Axnln1YiUdGxWMyIf+Rm7uJTvecYdEbV17Pm57iytcfVrY8PrjThwoh+fR8Ik7B9kWkZXoy+Z1bGfHQ79x36x7MZgNrN1/HgWMVMF/FqcqcDCYmPPEzC75pwckE/8s/4RpVKSyfwVNjGdenJgV5l/5B4uVj5MWPo4k56MEns0NLqYcOwmwoWkr7mHJRJQq2o6OjbdqJJk2aMGnSJABq167N/PnzWbt2rSXY7t+/Pw888AAA06dPZ968eWzevJnbbrutRPufNGmSZV8LFy6katWqLFu2jF69el3QtmnTpjRt2tTy+MUXX2TZsmV8++23DB06lIMHD7JkyRLWrFljyXafnQrxfPHx8fTu3ZvatWuzaNEi3NwuXmM5btw4Ro4caXmcnp5OePjF6xTtXV6OM7HHnIk95s7+bd589Ns+bnsgmS/mh5R118q9IS+dpHXndJ69rxan4y5dt7t/W9GPu7CIomA7OaHozz3m0LkBSWnJLqQnuxBcpeDvNQY+nB7G/2ZWJjC4gLQzLjRrnwlA/HFltf+rm+5Ixt3DxNqvr87g4AM7fGjevnxfCbO1LXuq0ndcb/x9cjEaDWTmuPP1a5+xbnPRD87kdE8Cfa1r5J2dTPh555GcVrLBp14eBdSrcZra1c7wTN+NQNEVKCcnWPveh4x67Xa279eAyVpNcgisVMibPxy0rHN2gcZtsrh7wGnujGiCyWTA09vItEVHyclyYsrACIyFCuSkfLOLa/v/rPeuXLkyiYmJxW739vbGz8/PavvltG3b1vLvoKAg6taty759+4ptm5mZyeTJk1m1ahVxcXEUFhaSk5NDTEwMADt27MDZ2ZkOHTpc8phdunThhhtu4IsvvsDZ+dJTFrm7u+Pu7riBjMEJXN014/1/Y2bIS6e48bY0nru/FgknLv//5bqGRQFEcmLRFYU9W4qmo6t6XZ4lUPcNKMQvqJCEk9ZXHUwmA2fii9p06p7C3i1epKkU6D+L7JXEH2sDSEu+Old5ajbItry/cmlpmUU/Mq+vF0uAbw4bdxRdqdxzJBhf73zqVD/NweNFP4Kurx+LwWBmX3TJ6uqzct0YMNF6ir97Ou2jeb1YJr19K3FJV1Yy5Kh2/OrDE53qWK179vUTnDjswZI3K2EyGfDyKQq0C/INTOpf47IZcJHywC6+Pf9ZzmEwGDCdd836ctuvplGjRrFmzRpeffVVatWqhaenJz179iQ/v2hwmKdnyQb6devWjaVLl7J3714aN3aceaQ9vIyE1Tg3UC40PJ+aDXPISHUmPdmZB59JJOpHP5ITXPELKuTuAaepGFrArysCLM+pVCUf3wAjwVXycXKGmn8HhbHRblazM8g5Q6efpFP3FCY/WpOcTCcCKxVlorMynMnPdaJy9Tw63ZvC5rV+ZKQ4U6N+Lk9OPsXOKG/LbCOnjnqwcbUfg6ecYu6YcLIynHh0XBwnD3vw18aiYMAvsJCb7kxl50YfXD1MdO2VzE3dUnmuZ60yO/ey5uFlJKx6ruVxaHgeNetnkZHmQlKsOz7+hQSH5VEhpOg9qVqzqG1KkqvVrCGVq+fS6IYMJgywDjbO3+7pZSSwUgHuHiZq1i+aMz3msCeFBU50vi+JwgInDu8pyra2uy2FrvcnMWdsDZuctz3wdC+gSvC5zH1oxQxqhZ8hPcudxGQffL1zCQnKokJAUW11eGgqAMlpniT/PY/2be0OEhMXQGqGBw2vS2DoA3/w5ZpGnEgIACAmLpBNu6oyqt+vvPZJO1ycTTzz4EbWbb6OM6nn5kuvXjkFVxcTvt55eHkUUCv8DACHT1TAbDYQfcq6rDE13YP8AucL1l/LcrKcrcaZAORmO5GRUrTey8fI9M+P4u5pYtawCLx8jHj5FA0GTzvjgsmkDHeJ6A6Sdscugm1b++OPPyz11ikpKRw8eJD69esX2/b333+nf//+3HvvvUBRpvvYsWOW7Y0bN8ZkMrFhwwarQZP/NHPmTHx8fLj11ltZv349DRo0uHonVIbqNM3hlaVHLI8HTYkF4McvApk3tipVa+Ux4f5j+AUZyUhx5uBfXjx7by2OHzxXuvDIqHi69j43i8Lba4ouKT7X4zp2RjnmzTn+q7v6FX2xv7r0sNX6V0eEs2ZJBQoLDFzfPoN7H0vCw9NEUpwrv30XwOdzrUt3XnmmOk9OPsXUhUcxm2FnlA/jH6ppdZm28/3JPD4hFoMB9m314rn7a3FghzfXqjqNs5i1eL/l8ZMTiq5yrfmqIrOfq0nbzik8++q5Urvn5xf9fXw6J4xP556bjz7y/iROx7mx7dfia3pHzIymSZtzU9i99d0eAPq1b0rCqaIrGQ8MO0VIlXyMhQZOHPVgxrBa/Pa94wZzdSOSmDP6O8vjoX02AbD699rM/KgD7ZrFMPbRc3cunjToZwAWfHM9C75tAUC10FSe6PEnvt55xJ/24dNVzfjyR+sB8i+935FnHozitVHfYzLBL9tq8MaitlZtXh7+A6EVMy2PP5i8DICOAx+7imd8bavVOIf6LYp+OC2I2m+17ZEb6pNw8tqY8lIcj8FsvprDRK5cx44dadasGXPmzLGs6969OwEBASxYsACDwcCyZcvo3r27ZXtAQABz5syhf//+l9z3+vXr6dSpEw0bNmTu3LmEhIQwfvx4duzYwaFDh3Bzc2PBggUMHz7cchOa++67j+joaP73v/9hMBiYMGEC69ev59FHH7X0ccCAAaxdu9YyQPL48eMkJibSq1cvyzFTUlIICAhgxIgRfP7556xfv5569eqV6DVJT0/H39+fjtyDi0GXiMucQdkUe2K4xPgHKV25t2jKV3vh/v2fZd0FAQrNBaznG9LS0vDzu/g9KWzhbOxQffo0nDw8Lv+Eq8iUm8vx58eXyXmXB9dEMdTMmTN55plnaNGiBfHx8axYseKiAxZfe+01AgMDufHGG7nrrruIjIykeXPrOVPffvttevbsyVNPPUW9evV4/PHHycrKKnZ/r7/+Or169eKWW27h4MGDxbYREREREcf0rzLbv/76K++++y5Hjhzhq6++okqVKnzyySfUqFGD9u3b26Kf1xRltu2MMtt2RZlt+6HMtv1QZts+2ENmO2Ja2WS2j41XZvtirjizvXTpUiIjI/H09GT79u2Wm7GkpaUxffr0q95BEREREZHy6oqD7Zdeeol33nmH999/32qWkHbt2rFt27ar2rnLGTRoED4+PsUugwYNKtW+iIiIiIj80xXPRnLgwAFuvvnmC9b7+/tbBhmWlqlTpzJq1Khit+kyhoiIiFxzNPWf3bniYDs0NJTDhw8TERFhtf63334r9k6KthQcHExwcHCpHlNEREREpKSuuIzk8ccf55lnnmHTpk0YDAZiY2P57LPPGDVqFIMHD7ZFH0VERESkJMxltMhFXXFme+zYsZhMJm699Vays7O5+eabcXd3Z9SoUQwbNswWfRQRERERKZeuONg2GAyMHz+e5557jsOHD5OZmUmDBg3w8dGd/0REREREzvevb9fu5ubmMLcgFxEREXEEBnPRUtrHlIu74mC7U6dOGC5xk49169b9pw6JiIiIiDiKKw62mzVrZvW4oKCAHTt2sHv3bvr163e1+iUiIiIiV8psKFpK+5hyUVccbL/++uvFrp88eTKZmZn/uUMiIiIiIo7iiqf+u5iHHnqIjz766GrtTkRERESulKb+sztXLdiOiorCw8Pjau1ORERERKTcu+Iykvvuu8/qsdlsJi4uji1btjBhwoSr1jERERERkfLuioNtf39/q8dOTk7UrVuXqVOn0rVr16vWMRERERG5Mpr6z/5cUbBtNBoZMGAAjRs3JjAw0FZ9EhERERFxCFdUs+3s7EzXrl1JTU21UXdERERE5F/TAEm7c8UDJBs1asTRo0dt0RcREREREYdyxcH2Sy+9xKhRo1i5ciVxcXGkp6dbLSIiIiIiUqTENdtTp07l2Wef5Y477gDg7rvvtrptu9lsxmAwYDQar34vRUREROTyymCApMpILq3EwfaUKVMYNGgQP//8sy37IyIiIiLiMEocbJvNRT9bOnToYLPOiIiIiMh/UBYDFpXZvqQrqtk+v2xEREREREQu7Yrm2a5Tp85lA+7k5OT/1CEREREREUdxRcH2lClTLriDpIiIiIjYCZWR2J0rCrb79OlDcHCwrfoiIiIiIuJQShxsq15bRERExL4ZymDqv1KfarCcKfEAybOzkYiIiIiISMmUOLNtMpls2Q8REREREYdzxbdrFxERERGRklGwLSIiIiJiI1c0G4mIiIiI2DFN/Wd3lNkWEREREbERZbZFREREHISm/rM/ymyLiIiIiNiIgm0RERERERtRGYmIiIiII1FZh11RZltERERExEaU2RYRERFxFJr6z+4osy0iIiIiYiPKbIuIiIg4CE39Z38UbItchpOXV1l3Qc5jysoq6y7I39Z/+H5Zd0H+FhnWrKy7ICIXoTISEREREREbUWZbRERExFFogKTdUWZbRERERMRGlNkWERERcRAaIGl/lNkWEREREbERBdsiIiIiIjaiMhIRERERR6EBknZHmW0RERERERtRZltERETEUSizbXeU2RYRERERsRFltkVEREQchKb+sz/KbIuIiIiI2IiCbRERERERG1EZiYiIiIij0ABJu6PMtoiIiIiIjSizLSIiIuIolNm2O8psi4iIiIjYiIJtEREREREbURmJiIiIiIPQPNv2R5ltEREREREbUWZbRERExFFogKTdUWZbRERERMRGlNkWERERcRCq2bY/ymyLiIiISKmYMWMGrVq1wtfXl+DgYLp3786BAwes2uTm5jJkyBAqVKiAj48PPXr0ICEhwapNTEwM3bp1w8vLi+DgYJ577jkKCwut2qxfv57mzZvj7u5OrVq1WLBgwQX9efPNN4mIiMDDw4PWrVuzefPmK+7L5SjYFhEREZFSsWHDBoYMGcIff/zBmjVrKCgooGvXrmRlZVnajBgxghUrVvDll1+yYcMGYmNjue+++yzbjUYj3bp1Iz8/n40bN7Jw4UIWLFjAxIkTLW2io6Pp1q0bnTp1YseOHQwfPpzHHnuMH374wdLmiy++YOTIkUyaNIlt27bRtGlTIiMjSUxMLHFfSsJgNpuV/Lcz6enp+Pv705F7cDG4lnV3rnlO3t5l3QU5j+m8D2QpWz/E7ijrLsjfIsOalXUXBCg0F7Ceb0hLS8PPz69Uj302dqg/ZDrO7h6lemxjXi773nz+X513UlISwcHBbNiwgZtvvpm0tDQqVarEokWL6NmzJwD79++nfv36REVF0aZNG77//nvuvPNOYmNjCQkJAeCdd95hzJgxJCUl4ebmxpgxY1i1ahW7d++2HKtPnz6kpqayevVqAFq3bk2rVq2YP38+ACaTifDwcIYNG8bYsWNL1JeSUGZbRERERP6z9PR0qyUvL++yz0lLSwMgKCgIgK1bt1JQUEDnzp0tberVq0e1atWIiooCICoqisaNG1sCbYDIyEjS09PZs2ePpc35+zjb5uw+8vPz2bp1q1UbJycnOnfubGlTkr6UhIJtEREREUdhLqMFCA8Px9/f37LMmDHjkl01mUwMHz6cdu3a0ahRIwDi4+Nxc3MjICDAqm1ISAjx8fGWNucH2me3n912qTbp6enk5ORw+vRpjEZjsW3O38fl+lISmo1ERERERP6zEydOWJWRuLu7X7L9kCFD2L17N7/99putu1amlNkWERERkf/Mz8/ParlUsD106FBWrlzJzz//TNWqVS3rQ0NDyc/PJzU11ap9QkICoaGhljb/nBHk7OPLtfHz88PT05OKFSvi7OxcbJvz93G5vpSEgm0RERERB2Eoo6WkzGYzQ4cOZdmyZaxbt44aNWpYbW/RogWurq6sXbvWsu7AgQPExMTQtm1bANq2bcuuXbusZg1Zs2YNfn5+NGjQwNLm/H2cbXN2H25ubrRo0cKqjclkYu3atZY2JelLSaiMRERERERKxZAhQ1i0aBHffPMNvr6+ltpnf39/PD098ff3Z+DAgYwcOZKgoCD8/PwYNmwYbdu2tcz+0bVrVxo0aMDDDz/MrFmziI+P54UXXmDIkCGWbPqgQYOYP38+o0eP5tFHH2XdunUsWbKEVatWWfoycuRI+vXrR8uWLbnhhhuYM2cOWVlZDBgwwNKny/WlJBRsi4iIiDiK8wYsluoxS+jtt98GoGPHjlbr//e//9G/f38AXn/9dZycnOjRowd5eXlERkby1ltvWdo6OzuzcuVKBg8eTNu2bfH29qZfv35MnTrV0qZGjRqsWrWKESNGMHfuXKpWrcoHH3xAZGSkpU3v3r1JSkpi4sSJxMfH06xZM1avXm01aPJyfSkJzbNthzTPtn3RPNv2RfNs2w/Ns20/NM+2fbCHebYbDC6bebb3vv3v5tm+FiizLSIiIuIgDOaipbSPKRenAZIiIiIiIjaiYFtERERExEZURiIiIiLiKOx8gOS1SJltEREREREbUWZbRERExJEo02xXlNkWEREREbERBdsiIiIiIjaiMhIRERERB6F5tu2PMtsiIiIiIjaizLaUWO+hCbS7I43wWnnk5zqxd4sXH06rzMkj524Le3vfM3S6N4VajXPw9jVxX71GZKU7W+2nSs08Hp8QS4NWWbi4mone58HHsyrz10af0j4lu9WoVRo9H4ulVsNMKoQUMHVwXaJ+qmDZ7uFlZMCo49zYJRnfgEISTrrzzceV+e7zUKv91GuWQb+Rx6nXNBOTycCRfd68MKA++XnOBFfJ5cEhJ2naJo3ASgUkJ7qy7ptKLH67KoUF536HN2+fwsPPnKBarWwK8p3Y9acf78+IIPFU6d4O2F6U5O/A1d3EE5Ni6Xh3Kq7uZrau9+WNcVVIPe16wf58Awt5e81BKoUVFPv3AtCgVRavLj3MsQMePNWlrk3Pz5588moon75m/X+66nW5fPjrfgDmjq7K9l99OZPgiqeXifotsxg4PpZqtfMs7d96oQp7/vTm+AEPwmvl8fZPBy57DAB3TyPfHtkFwI9fBDF7RDWr7a7uJlZG7wSgsAAWvFyZP9f5EXfcDW8/E9fflMHA52OpEFr4318IB+LpbaTf6HhuvD2NgAqFHNnjydsTqnDwLy8Ann09hq69U6yes+VnX8b3rVkW3S2fNPWf3bHrYNtgMLBs2TK6d+9u82NFREQwfPhwhg8fXuz2Y8eOUaNGDbZv306zZs1s3h971KRtFisWVOTgDi+cXcz0HxvH9M+P8niHuuTlFAUIHp4mtqz3Zct6XwY+H1/sfqYuPMqpaHfG3H8deblO3Pt4ElM/jqZ/23qkJF0YjFyLPDxNHN3vzY9fBTPhrQMXbH9i3DGatk1j1rO1STjlTov2qQyZfJQzCW5sWhcEFAXaL320ly/eqcLbU2tiNBqoWS8Ls9kAQHjNHAxOZt6YeB2xxz2oXjubZ6YdwcPTxAcvRwAQUjWXSe/s5+uPwpj1bB28fAp5cvwxJrx5gGHdm5ba62FPSvJ3MGhyLDd0TuelJ6uTle7MkGmnmPjhMUbeU/uC/Y2cfYLofR5UCiso9njefkaemxvD9t98CKx07QVu1evmMPOLI5bHzs7nvtVrN8nhlvtSqFSlgIwUZz6dHcrzD1zHwk17cT7vN0tkn2T2b/cieq/nBfvvOTiRbo+ctlo3ptd11G2WY7XOy9fIh7/uszw2GM5ty8tx4vAuLx4cnkDNBjlkpjnz9sQqTOpfk/mrD/7bU3dII2afIKJuLrOGVSM5wZVbeqQw84sjPN6xHmfiiz7//1zny+wR4ZbnFOQbLrY7kXLBroNtexIeHk5cXBwVK1YEYP369XTq1ImUlBQCAgLKtnOl5J+ZhdnDq7Fk9x5qN8lh96airPSyDyoB0KRtZrH78AsqpOp1+bz+bDjR+4q++D6aVpm7+58hol6ugu2/bfklkC2/BF50e/3m6fy0rBK7NvsD8P0XodzeJ4G6TTMtwfaT46P55uPKfPleVcvzTkWfCza2/hrI1l/PHSP+hAdLP8yh24PxlmC7dqNMnJzg49erWYL0pR+GMfHt/Ti7mDAWXnuVaJf7O/DyNRL5QDIzh1Tjr999AXhtZDgf/HKAes2z2L/N2/LcOx85jbefkc9eD+GGWzOKPd7TL5/k52UBmEwGbrwtzXYnZqecnSEouPgfGXc8dMby79Bw6DcmjsGd65Fwwo2wiHwAnnrpFABpZ0KLDbY9vU14epssj4/s8SDmoCdPv3zSqp3BcPF+ePuZrH4QAAyZdpKn76hL4klXgqsW/0PqWuPmYaL9HWlMHlDD8p3x6exQ2nRJ585HTrNwVmWgKLjWd8G/p5pt+3PtfVP+S87OzoSGhuLiot8nZ3n7GQHISL3wsvfFpCc7c+KwO53vT8Hd04iTs5luD58hJcmFQzsv/CKU4u3b5kebW5KpEJIHmGnSOo0qETls+60o+PYPyqdes0zSzrgy+4tdLIr6k1mf7aZhi/RL7tfb10hG6rn/44d2+2A2Q5ceiTg5mfHyKeSWe5LYsdH/mgy0i/PPv4PaTbJxdTOz/VdfS5sThz1IOOlK/RbZlnXVaufy4IgEXnmmGmZT8Zm7rr2TqVwtr9gyh2vFqWg3Hri+If3a1GfmkGokniw+CMvNduLHL4IIrZZ30asEJbF6UQWq1sylcessq/U5WU483KoBfVs0YFL/Ghw7cOkyqqx0ZwwGM97+xn/dF0fj7GzG2QXy86z/v+flGmh4w7nXu0nbTL7YuYcPft3PsBkn8Q289q7oiGMp02/Ljh078vTTTzN69GiCgoIIDQ1l8uTJVm1Onz7Nvffei5eXF7Vr1+bbb78t0b5btmzJq6++anncvXt3XF1dycwsyriePHkSg8HA4cOHLW2ys7N59NFH8fX1pVq1arz33nuWbceOHcNgMLBjxw6OHTtGp06dAAgMDMRgMNC/f38ATCYTM2bMoEaNGnh6etK0aVO++uqrS/Y1Ly+P9PR0q8XeGQxmBk05xe7NXhw/cCVBsoGxvWtyXaMclh/azcrondz3RBLj+9YgM00/ZErq7RdrEHPYi09/28qKvX/w0kd7eWtKTXb/WRRsV65WVLPad9gJVi8JZsLA+hze482Mj/cQVj2n2H1WrpbD3Q/H8f3ic4FdwkkPxg9oQP9nY/h2TxRLt2+mYmg+05++duqGL6W4v4Og4ELy8wwX1F6nJrkQFFwUBLq6mRj31nE+eDGMpFNuxe47rEYejz4fx8vDqmMyXpuX0es1z2LUnBimfXaEYTNPEh/jzrP31iY789xX14oFFbinVmPuqdWEP9f5MWPxEVzd/l2aLT/XwLplgUQ+kGy1vup1uYx8LYbJ/4tmzPzjmE0w4u7aJMUWH/jn5xr4cFoYHbun4O1rKrbNtSgny5m9W4rKbYJCCnByMnPLfSnUb5FNUEhRQL1lvS+vPFONMb1q8uG0yjRum8m0T4/i5KTUqZRfZZ6aWrhwId7e3mzatIlZs2YxdepU1qxZY9k+ZcoUevXqxc6dO7njjjvo27cvycnJl9hjkQ4dOrB+/XoAzGYzv/76KwEBAfz2228AbNiwgSpVqlCrVi3Lc2bPnk3Lli3Zvn07Tz31FIMHD+bAgQvrZcPDw1m6dCkABw4cIC4ujrlz5wIwY8YMPv74Y9555x327NnDiBEjeOihh9iwYcNF+zpjxgz8/f0tS3h4+EXb2ouh009RvV4uMwZXv8Jnmhk6/RSpp1149t5aPN2tNhtX+zNlwTFLICKXd/fDcdRrlsHkJ+sx7N4mvD8jgqcmHaXZjalAURAI8N3iENYsDeHIXh/em16Dk0c96doz8YL9VQjJ46WP9vHr9xVYvSTEsj6wYj5Pv3SEn5ZV4pkeTXjuwYYUFhgY/8YBNCLm3/8dDBgXR8xhD9Z9XXypkJOTmbFvHueTV0M4ddT9anS1XGp1SwY335VGzQa5tOyYwUufHiUz3Zlfvg2wtLnlvhTe+vEAr359iKo185j2ZAT5uf/ux8nv3/uTk+lMl17W3zENWmbT5f4UrmuUQ5O2WUz8MBr/CoV892mFC/ZRWADTnowAMwybefKC7de6WcOqYTDA59v3svLYTroPTGL98gDMf/8m2fBNIH/86M+x/Z5ErfZn4iM1qHt9Dk1uLL40UYphLqNFLqrMg+0mTZowadIkateuzSOPPELLli1Zu3atZXv//v154IEHqFWrFtOnTyczM5PNmzdfdr8dO3bkt99+w2g0snPnTtzc3Ojbt68lAF+/fj0dOnSwes4dd9zBU089Ra1atRgzZgwVK1bk559/vmDfzs7OBAUV1cUGBwcTGhqKv78/eXl5TJ8+nY8++ojIyEhq1qxJ//79eeihh3j33Xcv2tdx48aRlpZmWU6cOFGSl67MDJl2ktZd0hnd8zpOxxWflbuYZu0zuaFzOjMGV2fvn94c3uXF/Oerkp9roHOvy/+IEnBzN9JvZAzvzYhg07ogjh3wZsWnlfnlu4r0GBgLQHJS0fsSc9jL6rkxRzwJDsuzWhcUnM/MT/awd5sv8164zmrbnQ/Fk53hzEezIjiy14fdf/rzyqjaXN8ujXrNru0vv4v9HSQnuuDmbraUl5wVUKmQ5MSiTGiz9pncdGcq38X8xXcxfzFzSVG975e7d/PwqHg8fUzUbZbDkGmnLG36jkjguoa5fBfzF03bFV/f7eh8/I1UrZlH7LFzP0C8/UxUqZlP4zZZvPD+MU4cduf37/3/1f5Xf16B1p3TLjsQ1cUVajXKITba+ofQ2UA74ZQbMxYfUVa7GHHH3XmuRy3uvq4RD7VswNPd6uDiaibuePHfJfEx7qSecbbU4IuUR2V+3b5JkyZWjytXrkxiYmKx2729vfHz87PafjE33XQTGRkZbN++nY0bN9KhQwc6duzIzJkzgaLM9nPPPXfRvhgMBkJDQ0t0rLMOHz5MdnY2Xbp0sVqfn5/P9ddff9Hnubu74+5eHrJXZoZMO8WNt6XxXM9aJJy48j67exZ9+Zj+8R1kMhtwujavlF8xF1czrm7mC+p8TSaD5VJrwkl3Tse7UbWmdclI1Rq5/LkhwPK4QkgeMz/Zw+E9Prw+tpZlEORZ7h5GTP9YZ/y7pMFwzY6IufTfwaGdXhTkG7i+fQa/fRcAFJUhhFQtYN/Woh8/Lz4WgZvHuT+Cus1yePb1Ezx7by1ij7mRneHEE53qWO33rn5naNY+kxcfr058zJX9yHUUOVlOxB5349YexV8FM5sBs4GC/CvPI8XHuPHX7z5MXhB92bZGI0Tv8+CGW8+V/J0NtE9FuzPrq8P4BalW+1LycpzJy3HGx7+QFh0y+OClsGLbVaycj1+gkeTEMg9Xyg0NkLQ/Zf6/19XVuubNYDBgOi8Su9z2iwkICKBp06asX7+eqKgounTpws0330zv3r05ePAghw4duiCz/W+PddbZevBVq1ZRpUoVq23lI5i+tKHTT9Hp3hQmD6hBTqYTgZWKvvCyMpzJzy36cgusVEBgcCFhNYqypzXq5ZCd5UzSKVcyUl3Yt9WbzDRnnpt7gs9eDyEv14nb+54hNDyfzWv9yuzc7I2Hl5Gw6rmWxyFV86hZP4uMVBeS4tzZucmPgWOOkZfrRGKsO41vSOfW7km8PyPi72cYWPphGA89fYLo/V4c2etN5/uSqFozh2nDiuqtK4Tk8fKne0g85c4HM6vjH3QugEk5XRTM/bk+iHsHxPHg0BOsX1ERT28j/Z+NIeGkO0f2nptV41pyub+D7Axnfvg8iCcmx5KR6kJWhhNDpp1i7xYvy0wkccetPw/8/w7MYg55WGq9/zkWIvWMC/l5hiscI1G+vTcljDZd0wiuWsCZeBc+ebUyzk7Q8d4U4o67seHbAFp0yMA/qJCkOFeWzA/BzdNkFQSfinYjN8uZ5CQX8nMNHNld9PpVq5NrVdv9w+IggkIKaHXLhWNmPn0thPrNswmrkUdmmjNfvR1M4ik3bnuwaDaUwgJ48fEaHN7lydSPj2IyGizBoW+A8V/XkDuiFh3SMRjgxBF3qtTI57EJsZw47MGPXwTh4WXkoWcT+G2VPymJrlSOyOOxF+KIjXZj63rfy+9cxE6VebBtSx06dODnn39m8+bNTJs2jaCgIOrXr8+0adOoXLkyderUufxOLsLNrSgYMRrPZS8aNGiAu7s7MTExFwTyjuCu/kVfLK9+bT3F1avDw1mzpKisptsjZ3j42QTLttnLj1i1SU92YfyDNek/No6XlxzB2dXM8QMeTB4QwdFipuW6VtVulMmsz/ZYHj85/hgAa76uxGtjajNzeB36jzrO6NmH8A0oJPGUOwtfq8aqRefqrZcvCMPVzcQTzx/D17+Qo/u9Gd+/AXExRbMoXN8ujSoRuVSJyOXT37ZaHf/22jcC8Ncf/swaWYeej5+i52OnyMt1Yt92X14YWHRjnGtRSf4O3pkchskME94/hqu7mS3rfZk/rsoF+5JLOx3nyoynIshIcca/QiENW2UxZ+VBAioYMRYY2L3Jh2XvVyIzzZmAioU0bpPJ698cIqDiuTKQOaOqsTPq3A2znupa9GNz4aa9hIYXlSaYTEU3runSK9lqfu6zMtOcmfNcOClJLvj4G6ndJJvXvzlE9TpFSYXT8W788WNR6cpTXepZPXfWV4dpqnpjC28/EwPGxVGxcgEZqc78/p0//5tZGWOhAWcXAzXq59Dl/hS8/YycSXBh2wZfFs4K/VdXK0TshUMH2x07duSNN96gUqVK1KtXz7Ju/vz53H///f9p39WrV8dgMLBy5UruuOMOPD098fX1ZdSoUYwYMQKTyUT79u1JS0vj999/x8/Pj379+l2N0yozkWGXv4nJp7ND+XT2pacpO7TTi/EPXnfJNte6XZv9LQFvcVJOu/H62AtvkPJPX75X1Wqe7fP99HUwP30dfNl9bFhVkQ2rKl623bWiJH8HBXlOvPl8Vd58vvjX/p92Rvlcdr8l+dtyNM+/c/yi2yqEFvLSp0cvu49Xlh6+bBsnJ/hs696Lbh80JZZBU2Ivuj00PJ8fYndc9jgCv6wI4JcVAcVuy8910nfD1aA7SNodh/6peNNNN2EymayyzB07dsRoNNKxY8f/tO8qVaowZcoUxo4dS0hICEOHDgXgxRdfZMKECcyYMYP69etz2223sWrVKmrUqPGfjiciIiIi5Y/BbDbr94idSU9Px9/fn47cg4tBd9Eqa07e12Ztsr0yZWVdvpGUCmVz7UdkWLOy7oIAheYC1vMNaWlp+PmV7jiks7FDk/7TcXa79E2XrjZjfi47FzxfJuddHjh0ZltEREREpCyV22B70KBB+Pj4FLsMGjSorLsnIiIiIlJ+B0hOnTqVUaNGFbtNlzBERETkWqR5tu1PuQ22g4ODCQ6+/EwKIiIiIiJlpdwG2yIiIiLyD5r6z+6U25ptERERERF7p8y2iIiIiIMwmM0YSnlW59I+XnmjzLaIiIiIiI0o2BYRERERsRGVkYiIiIg4Cg2QtDvKbIuIiIiI2Igy2yIiIiIOQje1sT/KbIuIiIiI2IiCbRERERERG1EZiYiIiIij0ABJu6PMtoiIiIiIjSizLSIiIuIgNEDS/iizLSIiIiJiI8psi4iIiDgK1WzbHWW2RURERERsRMG2iIiIiIiNqIxERERExEFogKT9UWZbRERERMRGlNkWERERcRQaIGl3lNkWEREREbERBdsiIiIiIjaiMhIRERERB6IBi/ZFmW0RERERERtRZltERETEUZjNRUtpH1MuSpltEREREREbUWZbRERExEHopjb2R5ltEREREREbUbAtIiIiImIjKiMRERERcRS6g6TdUWZbRERERMRGlNkWERERcRAGU9FS2seUi1NmW0RERETERhRsi4iIiIjYiMpIRERERByFBkjaHWW2RURERERsRJltEREREQehO0jaHwXbIpdhys4u6y6I2KU76t1c1l0Qi/Sy7oCIXISCbRERERFHYTYXLaV9TLko1WyLiIiIiNiIgm0RERERERtRGYmIiIiIg9AASfujzLaIiIiIiI0osy0iIiLiKHRTG7ujzLaIiIiIiI0o2BYRERERsRGVkYiIiIg4CA2QtD/KbIuIiIiI2Igy2yIiIiKOQneQtDvKbIuIiIiI2Igy2yIiIiIOQjXb9keZbRERERERG1GwLSIiIiJiIyojEREREXEUuoOk3VFmW0RERETERpTZFhEREXEQGiBpf5TZFhERERGxEQXbIiIiIiI2ojISEREREUdhMhctpX1MuShltkVEREREbESZbRERERFHoan/7I4y2yIiIiIiNqLMtoiIiIiDMFAGU/+V7uHKHWW2RURERERsRMG2iIiIiIiNqIxERERExFGYzUVLaR9TLkqZbRERERERG1FmW0RERMRBGMxlMEBSie1LUmZbRERERMRGFGyLiIiIiNiIykhEREREHIXuIGl3lNkWEREREbERZbZFREREHITBbMZQylPxlfbxyhtltkVEREREbETBtoiIiIiIjaiMRERERMRRmP5eSvuYclHKbIuIiIiI2Igy2yIiIiIOQgMk7Y8y2yIiIiJSan755RfuuusuwsLCMBgMLF++3Gq72Wxm4sSJVK5cGU9PTzp37syhQ4es2iQnJ9O3b1/8/PwICAhg4MCBZGZmWrXZuXMnN910Ex4eHoSHhzNr1qwL+vLll19Sr149PDw8aNy4Md99990V9+VyFGyLiIiIOApzGS1XICsri6ZNm/Lmm28Wu33WrFnMmzePd955h02bNuHt7U1kZCS5ubmWNn379mXPnj2sWbOGlStX8ssvv/DEE09Ytqenp9O1a1eqV6/O1q1beeWVV5g8eTLvvfeepc3GjRt54IEHGDhwINu3b6d79+50796d3bt3X1FfLsdgNiv3b2/S09Px9/enI/fgYnAt6+6IwVDWPZDz6SPLbjj7+ZV1F+RvxvT0su6CAIXmAtbzDWlpafiV8t/H2djh5vYTcXHxKNVjFxbm8stvUzlx4oTVebu7u+Pu7n7J5xoMBpYtW0b37t2BokxyWFgYzz77LKNGjQIgLS2NkJAQFixYQJ8+fdi3bx8NGjTgzz//pGXLlgCsXr2aO+64g5MnTxIWFsbbb7/N+PHjiY+Px83NDYCxY8eyfPly9u/fD0Dv3r3Jyspi5cqVlv60adOGZs2a8c4775SoLyWhzLaIiIiI/Gfh4eH4+/tblhkzZlzxPqKjo4mPj6dz586Wdf7+/rRu3ZqoqCgAoqKiCAgIsATaAJ07d8bJyYlNmzZZ2tx8882WQBsgMjKSAwcOkJKSYmlz/nHOtjl7nJL0pSQ0QFL+tV5DExj4fDzL3q/IO5OqEFI1n4837yu27UtPVOfXlQH4BhYydn4MNern4BtoJO2MC1E/+PG/GZXJznQu5TMoP3oPTaDd7amE18ojP9eJvVu8+HB6GCePnMtezPryEE1vzLJ63qpPKjBvbLjlcaWwfIbNPEnTGzPIzXJmzZeBfDQjDJOxKHvf7vZU7nzkNDUb5uDqZub4QQ8+nR3K1g3KYF5KhdACBo6PpVWnDNw9TcQec2f2iHAO7fQCil7Xbo+coXbjHPyCjAzuUoejezyt9lG5eh6PT4yl4Q1ZuLqZ2fqzL2++UIXU07q6dVavJ05wY5fTVK2ZQ36uE/u2+/HR7AhORRe9zj7+BTw07DjN26VSqXIeacmuRK2twCdzq5Odee7r7rv9v16w75kj6/LLd8EAjJhxgC73Jl7Q5vghLwbf1eKC9fc/foIBzx5j+cIw3ptx3dU63XKv99AE2t2RZv25Na2y1eeWq7uJJybF0vHuVFzdzWxd78sb46z/3/8Q+9cF+54+uBobvgkslfMod8zm0r8C+PfxistsX6n4+HgAQkJCrNaHhIRYtsXHxxMcHGy13cXFhaCgIKs2NWrUuGAfZ7cFBgYSHx9/2eNcri8loWD7Ijp27EizZs2YM2dOWXfFLtVpmk23h5I5uufch2ZSrCt9mjawanfHQ2foOTiJP9f5AmA2QdQPfix4OZS0My6E1chj6PRT+AacZOaQ6qV6DuVJkzaZrFhYkYM7vHB2gf5j45i+6AiPd6xHXs65HynffVqBj18NtTzOyzl38crJycyLHx8lJcmFEffUJii4kOfmHsdYaOB/M8MAaNwmk22/+PK/mWFkpjsT2fsMUxZE88ydtTmyx6v0Trgc8fEv5LVvDrFzow8vPFST1DPOVKmZT2bauffFw8vEns3e/LIigBGvnrxgH+6eRqZ/fpSjez0Zc39RsNZvdDxTFxa99mazSpkAGrVKY+WiMA7u8sHZ2Uy/EceY9sFunryzBXk5zlQIzqdCcD4fzKpBzGEvQsLyGDrlMBWC85j+jPVn02vj6rD113PBWmb6ua/Dd6ddx4LZ576knZzNvPnNNn77oeIFfardKIPbe8dxdL+3Dc64fGvSNosVC85+bpmLPrc+P8rjHepaPrcGTY7lhs7pvPRkdbLSnRky7RQTPzzGyHtqW+3r1eHhbPnZ1/I4M13JGXvk5+dX6uUz5cE1H2yvX7+eTp06kZKSQkBAQFl3p1zw8DIyZv5x5jxXlQeeSbCsN5kMpCRZZ+FuvD2NX1YEkJtd9MGYmebCyo/PfWElnnJjxcIK3D84qXQ6X06Nf8g6WzZ7eDWW7NpN7SY57N7kY1mfl3vhe3BW8w4ZVKuTy9g+DUk97crRPfDxK5UZ+Hwsn8wOpbDAiXcmVbV6zv9mhtG2azptuqQr2L6IXkMSOR3rxuwR1SzrEk5YZ3PWLg0CIKRqfrH7aHhDNiHh+QzpWsdyheeVZ6qxdN9umrXPZPuvvsU+71oz8fFGVo9fG1eHxVGbqN0wk91b/Dl+yJtpT58LquNPeLLw9eo898oBnJzNlis4AFnpzqScdqM42ZkuZJ83qUHbW0/j41fImq+ts1seXkZGv3qAeRNq02fwiatwho5lfN+aVo9nD6/Gkt17LJ9bXr5GIh9IZuaQavz1e9H/8ddGhvPBLweo1zyL/dvO/YDJTHe+6GebWDOYi5bSPubVEhpalDBKSEigcuXKlvUJCQk0a9bM0iYx0frqU2FhIcnJyZbnh4aGkpCQYNXm7OPLtTl/++X6UhKq2S5F+fnFf9GWN0Onn2LzWr/LBgC1GmdTq1EuP3wedNE2QSEFtLs9jZ1RygpdCW8/IwAZqdbZnU73prBk1y7eXbufAWNjcfc4d1uvBi2yOLbfw+ry7Jb1vnj7mahep/hR1QaDGU8f4wXHkXPadE3n4F+ejH/3GF/s3MObPx7g9gfPXNE+XN1MYIaC/HPBYEGeAbMJGt6QdYlnXtu8ff/+O0i7eN7I29dIdqazVaANMHjiET6PiuL1Jdvpcl88l5pOoWvPBHZEBZAYaz3o7KmJh9m8PpAdUSpnKIl/fm7VbpKNq5vZ6rvkxGEPEk66Ur9FttVzh047yZLdu5m36iBd+5zhiqe/kHKjRo0ahIaGsnbtWsu69PR0Nm3aRNu2bQFo27YtqampbN261dJm3bp1mEwmWrdubWnzyy+/UFBQYGmzZs0a6tatS2BgoKXN+cc52+bscUrSl5K4JoLtvLw8nn76aYKDg/Hw8KB9+/b8+eefHDt2jE6dOgEQGBiIwWCgf//+lueZTCZGjx5NUFAQoaGhTJ482Wq/qampPPbYY1SqVAk/Pz9uueUW/vrrXG3Z5MmTadasGR988AE1atTAw6N0RwfbQod7UqjVOIePZlS+bNvbHkjm+EF39m65MJAe+9Zxvjmyk8+37yU705nXR4UXswcpjsFgZtCUU+ze7M3xA+fqfn9eHsisYdUZfX8tFs8P5taeKYx+47hle2ClwgsyQ6l/Pw4MLiz2WD0HJeLpZWLDioCrfyIOonK1fO585Ayx0e48/2ANVi6syOAXT9H5/uQS72P/Vm9ys50YOD4Od08T7p5GHp8Yi7MLBAUXXH4H1yCDwcyTzx9lz1Y/jh8q/se6X0ABDwyO4fsl1p9Xn8ytzszh9Rj/aGN+/7EiQyYd5u6HY4vdR1BwHi1vSuaHL0Ot1t98RyK1GmSy4LUaxT5PrJ373PKyfG4FBReSn2cg6x8lIalJLlb/7xfOCmXaoAjG9anJb98FMGz6Ke4ZeLpU+y9XV2ZmJjt27GDHjh1A0UDEHTt2EBMTg8FgYPjw4bz00kt8++237Nq1i0ceeYSwsDDLjCX169fntttu4/HHH2fz5s38/vvvDB06lD59+hAWVlQW+eCDD+Lm5sbAgQPZs2cPX3zxBXPnzmXkyJGWfjzzzDOsXr2a2bNns3//fiZPnsyWLVsYOnQoQIn6UhLXRBnJ6NGjWbp0KQsXLqR69erMmjWLyMhIDh06xNKlS+nRowcHDhzAz88PT89zwcvChQsZOXIkmzZtIioqiv79+9OuXTu6dOkCwP3334+npyfff/89/v7+vPvuu9x6660cPHiQoKCibO7hw4dZunQpX3/9Nc7OxWcH8/LyyMvLszxOt9MpnCqF5TN4aizj+tSkIO/Sv9PcPEx0ujeFRXNCit3+7qQwPnsthCo183h0XBxPTopl/vNVi20r1oZOP0n1ujk8e691TeP3n50rzzm235P/t3fncVHV6x/AP4d9lUUWQVkUxBVUMk1TwbQS9+W64pZiXXHLXStUIMTIpej+bnpdqTSXMKIkU0lJUq+aiJqKgiKo4Mq+w5zfH+ToxJLemDnDzOf9es3rxZxzmPPA6PCc5zzf7/fxfX1E7E2Dg0sZsm69+CCVvsNzMHHBPaya1hJ5j3j7ti6CDnD9gjG2r6lO6NIumcC1bSkGTXqEI/vqvqvzrLzHevjwHVfMCb+NYdMfQpRVXzxdv2AMUcZ+7doErkiFS+siLJrQqdb9xqaVCN70OzLSTLDzX84K+77+/OnzG1fMYGRchVHTbiP2y+Y1Xqf/8PsoLNDDyfim8m02zcrwzns38P40T1SUa0XN6m+bvfoOXNqWYuFw9xf+3mf/jqRdMoGRiQyjZz7Ad1ttGzJEzSHhAMnndfbsWXmxE4A8AZ4yZQp27NiBJUuWoKioCG+//TZyc3PRq1cvHDx4UKFouXPnTsyePRv9+vWDjo4ORo0ahcjISPl+CwsLHDp0CLNmzcJLL70EGxsbrFixQmEu7p49e2LXrl344IMP8N5776F169aIiYlBx45PW9aeJ5a/ovHJdlFRET7//HPs2LEDfn5+AIDNmzfj8OHD2LZtG15++WUAgJ2dXY2ebS8vL6xcuRIA0Lp1a/zrX/9CfHw8Xn/9dSQmJuL06dO4f/++fLTt2rVrERMTg2+++Ub+ZpaXl+OLL76ArW3dHwrh4eEIDg5u6B+9wbl7lcDKthL/99M1+TZdPcDzlSIMfeshBrt6QfZHYtB7UC4MjcU6k42cB/rIeaCPzFQjFOTqYn1MGnZ9Yo/H95nU1WfWh7fRvX8+Fo50x8Os2vtNn7h6rrrH2tG1OtnOeaCHNl0UWxIsbaurRzn3FT8KfIbm4N21GQh7x5X9wn/h8X093Lqm+KGbed0QvQbmvtDrnEswx1s926GJdSWqKqurfV+f/x1ZGfW/z9poZlAquvk+xpKJnfDoXs0LSWPTSoRuuYTiIl2Ezm6Pqsr6E+KUC00wYVYm9PRlqKx49lgRr4/Mxs/f2Slsb92hAFY2Ffhs/zn5Nl09oGPXPAzxv4thXr3kn4UEzAq7je6v52PhCDeFz63H9/VgYCjCtEmVQnXb0ray3r8FV8+ZwH/+PegbyHix00j5+vqivmVeBEFASEgIQkJC6jzG2toau3btqvc8Xl5eOH685uxDzxo9ejRGjx79t2L5KxqfbKelpaGiogKvvvqqfJu+vj66deuGK1euyJPt2nh5eSk8d3BwkDfkJycno7CwEE2bNlU4pqSkBGlpafLnLi4u9SbaALB8+XKF2xr5+flwclK/torzx83wdl8PhW0LN2QiM9UIe//PVuGPy5vjH+PUoSbIe/zX/8SerBmjb8AevLqJmPXhHfQckIfFo91rDMCrjVuHEgCQ/9G6/Jspxs29B4umFfJKtXefAhTl6yDj+tNk0XdYDhasy0B4oCtOx1so4WfRLJfPmMLJrUxhW/NWZbh/539LkvP/+D/T6dUCWNpU4tQhjux/SsTMoDT06P8IyyZ74d6dmpUlY9NKfLj1EirKdRAS2P65krFWbQtRkKv3p0Qb8OyWh+aupTgUrdhCcv6UJWYO8VbYNn/1Ndy+YYJ9W1ow0ZYTMSvsj8+tf9T83Lp+wQQV5QK69CpAYpwlAKCFWynsW1Tgym91D8h261CCghxdJtp1EGTVD1Wfk+qm8cn236Gvr3hlLQgCZLLqf1GFhYVwcHDAsWPHanzfsxVyU9O/Hvj3PCssqYOSIl2FHmEAKC3WQUGO4nZH1zJ4vlKEoIk1exlffi0fVraVSDlvjNIiXbi0KUVA0F1cOm2Ce7dZwavL7NW30Xd4DlZNa4WSQh1Y/VGRLirQRXmpDhxcytB3RA5OxzdBQY4uWrYrxTur7uDCSVPcvFL93pxLMEfGNSMsiczA1jBHWNlWYOqSbHwfZSP/o9V3eA4WfXILn69sgatJJvLzlJXqoLiAgyRrs/8/ttgQex3j5tzDL99bok2XYgyc+BifLH7aFmVuWQnb5hVoal/9+3Ryqx6QmnNfT95H/8bYx8i4boi8R3po91IxZobcwbf/sVWYk1jbBa5Ig+/g+wiZ1R4lRbqwsqkedF5UoIvyMl0Ym1YibOslGBpX4ePFbWBiVgUTs+pBeXmP9SGTCejW9xGsmlbgarI5yst00KVnDsa+k4no7TXb2N4cdQ9Xz5vX6AkvKdLDreuKfz5LS3SRn6tXZ/+4Npq9+g76jsjBqrda1vq5VVygi5++tsbbq+6iIFcPRQU6mBV2B5fPmshnIun+eh6sbCtx5TcTVJTpwLtPAcbNvY9vNrKFhBoPjU+23dzcYGBggF9//RUuLtXzOFdUVODMmTN499135SsLVVVVvdDrent7Izs7G3p6enB1dW3osBu1N8c9xsMsffyWULP9oLxUB37+j/DOqlLoG4h4cFcfv/5ogT3/qr23m6oNmVI9u8Xa6FSF7WvnO+Hw3qaorKiuDo0IeAAjYxkeZOkjMc4SX3/69PcqkwlYMaUV5oRnYkPsNZQW6+DIPmtEffx08Jif/0Po6QNzVt/GnNVP54M+tNcK6+ZzHvTaXEs2Qcj0lnhreRb8599DdqYBNq5wxNFvn85Q8cob+Vj0ydOp4d7bmAEA+HKdPb5aV101beFWireWZ8Hcsgr3MvXxdaQ99v+n5rzO2mzwhCwAQMSXFxW2r1/ugSPf2sO9QyHadi4AAGw7fFbhmKn9Xsb9O0aoqhAweMJdzFheCgEi7mYYY/NHrXBwr2L12sSsEj3feIhNqxWnr6PnN2TqH59b+9MUtq991wmH91a3GG5c5QiZCARtToe+oYizx8zxr+VPe+erKgQMmfoQ76wqhyAAd9MNsGmVI37c+XzjIbRSI+jZ1jaCWF/TjIZ49913sW/fPmzduhXOzs6IiIhAbGws0tLSUFxcDCcnJ2zfvh0DBw6EsbExzMzMal3UZvjw4bC0tMSOHTsgiiL69OmDgoICREREwMPDA3fv3sWBAwcwYsQIdO3aFatWrUJMTIx8tO3zys/Ph4WFBXwxDHoCe5glJ/CWsFrR/I+sRkOXi1eojSo1HVivbSrFChzDd8jLy1P54i7y3KHb+9DTU+0dscrKUhw7HSbJz90YaEXD05o1azBq1ChMmjQJ3t7eSE1NxU8//QQrKys0b94cwcHBWLZsGezt7eXTvfwVQRAQFxeHPn364K233oKHhwfGjRuHW7du1VjWk4iIiIi0k1ZUthsbVrbVDCvb6oUfWWqDlW31wcq2elCLyvbLElW2z7CyXRetqGwTEREREUlB4wdIEhEREWkLQRQhqPgOoKrP19iwsk1EREREpCRMtomIiIiIlIRtJERERESagvNsqx1WtomIiIiIlISVbSIiIiJNIQKQSXBOqhMr20RERERESsLKNhEREZGG4NR/6oeVbSIiIiIiJWGyTURERESkJGwjISIiItIUIiSY+k+1p2tsWNkmIiIiIlISVraJiIiINAUXtVE7rGwTERERESkJk20iIiIiIiVhGwkRERGRppABECQ4J9WJlW0iIiIiIiVhZZuIiIhIQ3AFSfXDyjYRERERkZKwsk1ERESkKTj1n9phZZuIiIiISEmYbBMRERERKQnbSIiIiIg0BdtI1A4r20RERERESsLKNhEREZGmYGVb7bCyTURERESkJEy2iYiIiIiUhG0kRERERJpCBkCQ4JxUJ1a2iYiIiIiUhJVtIiIiIg0hiCIEFQ9YVPX5GhtWtomIiIiIlISVbSIiIiJNwan/1A4r20RERERESsJkm4iIiIhISdhGQkRERKQpZCIgqLitQ8Y2kvqwsk1EREREpCSsbBMRERFpCg6QVDusbBMRERERKQmTbSIiIiIiJWEbCREREZHGkKCNBGwjqQ+TbTUk/vGfpBIV/PerFgSpA6BnsTdQbYhiudQh0B+qxAqpQyD88XcbT/+OEwFMttVSQUEBACARcRJHQgB4wUNUl3ypAyBSTwUFBbCwsJDm5BwgqXaYbKshR0dHZGZmwtzcHILQOKuq+fn5cHJyQmZmJpo0aSJ1OFqP74f64HuhPvheqA9NeS9EUURBQQEcHR2lDoXUCJNtNaSjo4MWLVpIHUaDaNKkSaP+4NQ0fD/UB98L9cH3Qn1ownshWUX7CZkIld+S5aI29eJsJERERERESsJkm4iIiIhISdhGQkphaGiIlStXwtDQUOpQCHw/1AnfC/XB90J98L1oQKKs+qHqc1KdBJHz0xARERE1avn5+bCwsEB/50Do6aj2oqVSVoYjGf9GXl5eo++5VwZWtomIiIg0Baf+Uzvs2SYiIiIiUhIm20RERERESsI2EiIiIiJNwXm21Q4r20RERERESsLKNhEREZGm4ABJtcPKNhGRkuXn59e5LzU1VYWREKmHiooK6Onp4dKlS1KHQqR0TLapwRQVFSEoKAg9e/aEu7s7WrVqpfAg1dq+fTv27dtXY/u+ffsQFRUlQUTaa9CgQSgrK6uxPSUlBb6+vqoPSMsdP34cEydORI8ePXDnzh0AwJdffonExESJI9Me+vr6cHZ2RlVVldShECkd20iowQQEBCAhIQGTJk2Cg4MDBEGQOiStFh4ejk2bNtXYbmdnh7fffhtTpkyRICrtZGZmhhEjRiA2NhZ6etUfu1euXMFrr72GMWPGSByddomOjsakSZPg7++PpKQk+UVQXl4eVq9ejbi4OIkj1B7vv/8+3nvvPXz55ZewtraWOhzNIUKCNhLVnq6x4QqS1GAsLS1x4MABvPrqq1KHQgCMjIxw9epVuLq6KmxPT09Hu3btUFJSIk1gWqikpAT9+/dHixYtsHv3bvz+++/o168f/P39sX79eqnD0ypdunTB/PnzMXnyZJibmyM5ORmtWrVCUlIS/Pz8kJ2dLXWIWqNLly5ITU1FRUUFXFxcYGpqqrD/3LlzEkXWOMlXkHR4B3o6Bio9d6WsHEeyNnEFyTqwsk0NxsrKitUJNWJnZ4cLFy7USLaTk5PRtGlTaYLSUsbGxjhw4AB8fX0xZswY/PLLL5g8eTI+/vhjqUPTOikpKejTp0+N7RYWFsjNzVV9QFps+PDhUoegmThAUu0w2aYGExoaihUrViAqKgomJiZSh6P1xo8fj7lz58Lc3FyeXCQkJGDevHkYN26cxNFpvj8PitTR0cGePXvw+uuvY9SoUQgKCpIfw0qQ6jRr1gypqak1LkITExM5tkTFVq5cKXUIRCrBZJsazLp165CWlgZ7e3u4urpCX19fYT9vCapWaGgo0tPT0a9fP3mfsEwmw+TJk7F69WqJo9N8lpaWtY5bEEURGzduxKZNmyCKIgRB4CAxFZoxYwbmzZuHbdu2QRAE3L17FydPnsSiRYsQFBQkdXhaJzc3F9988w3S0tKwePFiWFtb49y5c7C3t0fz5s2lDq9xkskAyCQ4J9WFyTY1GN4SVC8GBgbYs2cPQkNDkZycDGNjY3h6esLFxUXq0LTC0aNHpQ6BarFs2TLIZDL069cPxcXF6NOnDwwNDbFo0SLMmTNH6vC0yoULF9C/f39YWFggPT0dM2bMgLW1Nfbv34+MjAx88cUXUodI1CA4QJKIiLROeXk5UlNTUVhYiPbt28PMzEzqkLRO//794e3tjYiICIXBqidOnMCECROQnp4udYiNinyApF2ANAMk72/hAMk6sLJNpEEWLFiA0NBQmJqaYsGCBfUey1kwVGf79u0wMzPD6NGjFbbv27cPxcXFnIZRhaZNm4ZPP/0U5ubmaN++vXx7UVER5syZg23btkkYnXY5c+ZMrdOTNm/enLPC/B0cIKl2uKgNNZiqqiqsXbsW3bp1Q7NmzWBtba3wIOVLSkpCRUWF/Ou6HufPn5c2UC0THh4OGxubGtvt7OzYP69iUVFRtU57WVJSwrYFFTM0NKx1ddVr167B1tZWgoiIlIOVbWowwcHB2LJlCxYuXIgPPvgA77//PtLT0xETE4MVK1ZIHZ5WeLZPmD3D6iMjIwMtW7assd3FxQUZGRkSRKR98vPzIYoiRFFEQUEBjIyM5PuqqqoQFxcHOzs7CSPUPkOHDkVISAj27t0LABAEARkZGVi6dClGjRolcXSNGCvbaoeVbWowO3fuxObNm7Fw4ULo6elh/Pjx2LJlC1asWIFTp05JHR6RZJ7Mef5nnPNcdSwtLWFtbQ1BEODh4QErKyv5w8bGBtOmTcOsWbOkDlOrrFu3DoWFhbCzs0NJSQl8fHzg7u4Oc3NzhIWFSR0eUYNhZZsaTHZ2Njw9PQFUL0+dl5cHABg8eDCn1JJAUVER1qxZg/j4eNy/fx+yP03NdOPGDYki0z6c81x6R48ehSiKeO211xAdHa3Q2mZgYAAXFxc4OjpKGKH2sbCwwOHDh5GYmIgLFy6gsLAQ3t7e6N+/v9ShETUoJtvUYFq0aIGsrCw4OzvDzc0Nhw4dgre3N86cOQNDQ0Opw9M6AQEBSEhIwKRJk+Dg4FDrnM+kGpzzXHo+Pj4AgJs3b8LZ2Zn/H9RIr1690KtXL6nD0BwyEYCK2zpkbCOpD5NtajAjRoxAfHw8unfvjjlz5mDixInYunUrMjIyMH/+fKnD0zo//vgjDhw4gFdffVXqULQe5zyX1oULF9CxY0fo6OggLy8PFy9erPNYLy8vFUZG8fHxdd5948wwpCmYbFODWbNmjfzrsWPHwtnZGSdPnkTr1q0xZMgQCSPTTlZWVpwFRs14eHjAw8ND6jC0TufOnZGdnQ07Ozt07twZgiCgtiUmuJqnagUHByMkJARdu3bl3bcGJIoyiKJqV3RU9fkaGy5qQ6ShvvrqK3z33XeIioqCiYmJ1OFovdu3byM2NhYZGRkoLy9X2Mc5z5Xr1q1b8taRW7du1Xss7zaojoODAyIiIjBp0iSpQ9EITxa16Wc1RZJFbeJzorioTR1Y2aa/JTY2Fn5+ftDX10dsbGy9xw4dOlRFURFQPdI/LS0N9vb2cHV1hb6+vsL+c+fOSRSZ9omPj8fQoUPRqlUrXL16FR07dkR6ejpEUYS3t7fU4Wm8ZxNoJtPqo7y8HD179pQ6DM0jiqrvoWbdtl5MtulvGT58uPz27PDhw+s8jrdnVa++94NUa/ny5Vi0aBGCg4Nhbm6O6Oho2NnZwd/fHwMGDJA6PK1z/fp1HD16tNY+Ya4JoDoBAQHYtWsXZ6sijcc2EiIiJTM3N8f58+fh5uYGKysrJCYmokOHDkhOTsawYcOQnp4udYhaY/PmzZg5cyZsbGzQrFkzhT5hQRB4x0fJFixYIP9aJpMhKioKXl5e8PLyqnH3je1VL0beRmI5GXqCittIxHLE537BNpI6sLJNSpWbmwtLS0upwyCSlKmpqbxP28HBAWlpaejQoQMA4OHDh1KGpnU+/PBDhIWFYenSpVKHopWSkpIUnnfu3BkAcOnSJQmi0VCiBFP/sW5bLybb1GA++ugjuLq6YuzYsQCA0aNHIzo6Gg4ODoiLi0OnTp0kjlC7VFVVYcOGDdi7d2+tg/IeP34sUWTa55VXXkFiYiLatWuHgQMHYuHChbh48SL279+PV155RerwtEpOTg5Gjx4tdRha6+jRo1KHQKRyXK6dGszGjRvh5OQEADh8+DCOHDmCgwcPws/PD4sXL5Y4Ou0THByM9evXY+zYscjLy8OCBQswcuRI6OjoYNWqVVKHp1XWr1+P7t27A6h+X/r164c9e/bA1dUVW7dulTg67TJ69GgcOnRI6jAIwLRp01BQUFBje1FREaZNmyZBRBpCJpPmQXVizzY1GGNjY1y7dg1OTk6YN28eSktLsWnTJly7dg3du3dHTk6O1CFqFTc3N0RGRmLQoEEKPcORkZE4deoUdu3aJXWIRCoRGRkp/7qoqAjr16/HoEGD4OnpWaNPeO7cuaoOT2vp6uoiKysLdnZ2CtsfPnyIZs2aobKyUqLIGid5z7a5vzQ92wU72bNdB7aRUIOxsrJCZmYmnJyccPDgQXz44YcAAFEUOROJBLKzs+Hp6QkAMDMzQ15eHgBg8ODBHP1PWmXDhg0Kz83MzJCQkICEhASF7YIgMNlWgfz8fIiiCFEUUVBQACMjI/m+qqoqxMXF1UjAiRozJtvUYEaOHIkJEyagdevWePToEfz8/ABUD4hxd3eXODrt06JFC2RlZcHZ2Rlubm44dOgQvL29cebMGRgaGkodnsazsrJ67hXx2D+vXDdv3pQ6BHqGpaUlBEGAIAi1rqgqCAKCg4MliExDcICk2mGyTQ1mw4YNcHV1RWZmJiIiImBmZgYAyMrKQmBgoMTRaZ8RI0YgPj4e3bt3x5w5czBx4kRs3boVGRkZmD9/vtThabxPPvlE6hCoFjdu3ECrVq2kDkOrHT16FKIo4rXXXkN0dDSsra3l+wwMDODi4gJHR0cJIyRqWOzZpgZ3+fLlWme/4AqS0jp58iROnjyJ1q1bY8iQIVKHQyQJHR0dtGjRAj4+PvD19YWPjw/vvEnk1q1baNKkCbZt24YrV64AADp06IBp06bBwsJC4uganyc926+ZjJOkZ/vn4t3s2a4Dk21qMDdu3MDIkSNx8eJFANW92gDkt9LZt03arKqqCt9++608qWjfvj2GDRsGPT3eYFSlO3fu4NixY/Ke7evXr8PR0RE+Pj7o27cvAgICpA5Ra5w9exYDBgyAkZERunXrBgA4c+YMSkpK5G1v9PyYbKsvJtvUYIYMGQJdXV1s2bIFLVu2xOnTp/Ho0SMsXLgQa9euRe/evaUOUeukpKTgs88+kyd47dq1w5w5c9CmTRuJI9Muv//+O4YOHYrs7Gz57/7atWuwtbXF999/j44dO0ocofa6fv06wsLCsHPnTshkMhYFVKh3795wd3fH5s2b5RedlZWVCAgIwI0bN/DLL79IHGHjIk+2jcdKk2yX7GGyXQcm29RgbGxs8PPPP8PLywsWFhY4ffo02rRpg59//hkLFy6ssXIYKVd0dDTGjRuHrl27okePHgCAU6dO4cyZM9i9ezdGjRolcYTao0ePHrC1tUVUVBSsrKwAVC+uMnXqVDx48AAnTpyQOELtUVxcjMTERBw7dgzHjh1DUlIS2rZtC19fX/j6+mLYsGFSh6g1jI2N5b//Z12+fBldu3ZFcXGxRJE1Tky21RfvX1KDqaqqgrm5OYDqxPvu3bto06YNXFxckJKSInF02mfJkiVYvnw5QkJCFLavXLkSS5YsYbKtQufPn8fZs2fliTZQPVtJWFgYXn75ZQkj0z6WlpawsrKCv78/li1bht69eyu8L6Q6TZo0QUZGRo1kOzMzU/63hEgTcAVJajAdO3ZEcnIyAKB79+6IiIjAr7/+ipCQEI7+l0BWVhYmT55cY/vEiRORlZUlQUTay8PDA/fu3aux/f79+xycp2IDBw5EVVUVdu/ejd27d2Pfvn24du2a1GFppbFjx2L69OnYs2cPMjMzkZmZid27dyMgIADjx4+XOrzGSyZK86A6sbJNDeaDDz5AUVERACAkJASDBw9G79690bRpU+zZs0fi6LSPr68vjh8/XiOZS0xMZP+8CuTn58u/Dg8Px9y5c7Fq1Sq88sorAKpbekJCQvDRRx9JFaJWiomJAQBcuHABCQkJOHToEIKCgqCnpwdfX1/s3LlT2gC1yNq1ayEIAiZPnixfLVJfXx8zZ87EmjVrJI6OqOGwZ5uU6vHjxy+0uAc1nI0bN2LFihUYM2aMQoK3b98+BAcHK8xjy2kZG56Ojo7Cv/s/z87z7HMOylM9URSRlJSEo0eP4ujRo/jpp58giiKXCJdAcXEx0tLSAABubm4wMTGROKLGSd6zbTAaeoK+Ss9dKVbg5/J97NmuA5NtIg2lo/N8XWJM9pTjz0uB18fHx0eJkdCz1q9fj2PHjiExMREFBQXo1KkT+vTpA19fX/ZvU6PGZFt9sY2ESEPJZDKpQ9Bq/0sCHRgYiJCQENjY2CghIgKAr7/+Gj4+Pnj77bfRu3dvLp5CRErHAZJEWiQ3N1fqEKgeX331lUKvNzW8M2fOYO3atRg8eHC9iXZgYCAePnyowsiIGoYoEyV5UN2YbBNpqI8++khhYOro0aNhbW2N5s2by2eNIfXCrj71wQsfImooTLaJNNTGjRvh5OQEADh8+DCOHDmCgwcPws/PD4sXL5Y4OiL1xgsfarREmTQPqhN7tok0VHZ2tjzZ/uGHHzBmzBi88cYbcHV1Rffu3SWOjoiISDuwsk2koaysrJCZmQkAOHjwIPr37w+gumLH2UeIiDQTe7bVDyvbRBpq5MiRmDBhAlq3bo1Hjx7Bz88PAJCUlMRVC4mIiFSEyTaRhtqwYQNcXV2RmZmJiIgImJmZAahexj0wMFDi6Kg2EydO5By1REQahovaEBGpQG5uLk6fPo379+/XmAN98uTJEkVFdZk5cyZCQ0M55zk1Gk8WtfHFMEkWtTmG77ioTR2YbBNpkNjYWPj5+UFfXx+xsbH1Hssl2lXn+++/h7+/PwoLC9GkSROFZdwFQcDjx48ljE778MKHNNGTZLsXBkIPKk62UYFExDHZrgOTbSINoqOjg+zsbNjZ2dW7XDuXaFctDw8PDBw4EKtXr4aJiYnU4Wg1XviQpiotLUXLli2RnZ0tyfmbNWuGmzdvwsjISJLzqzMm20RESmZqaoqLFy+iVatWUoei9XjhQ5qstLQU5eXlkpzbwMCAiXYdOECSSIPFx8cjPj6+xu1yQRCwdetWCSPTLm+++SbOnj3LZFsN3LlzB3PnzmWiTRrJyMiICa8aYrJNpKGCg4MREhKCrl27wsHBQeF2OSnfsz3zgwYNwuLFi3H58mV4enpCX1+xn5L986rDCx8iUjW2kRBpKAcHB0RERGDSpElSh6KV6uuZfxb755Xv2QufBw8eICQkBG+99RYvfIhIJZhsE2mopk2b4vTp03Bzc5M6FCJJ8cKHiKTEZJtIQy1duhRmZmYICgqSOhQC++eJiLQVe7aJNMiCBQvkX8tkMvznP//BkSNH4OXlVeN2+fr161UdntZi/7x64YUPEakSk20iDZKUlKTwvHPnzgCAS5cuKWxnsqdaGzduxI4dO9g/rwZ44UNEqsY2EiIiJWP/vPrgwGEiUrXnGzVCRET/s4CAAOzatUvqMAhAeXk5evbsKXUYRKRF2EZCRKRkpaWl7J9XE08ufDhwmIhUhck2EZGSXbhwgf3zaoIXPkSkauzZJiIirdG3b9869wmCgJ9//lmF0RCRNmCyTURERESkJBwgSURERESkJEy2iYiIiIiUhMk2EREREZGSMNkmIiIiIlISJttERH8ydepUDB8+XP7c19cX7777rsrjOHbsGARBQG5ubp3HCIKAmJiY537NVatWyach/F+lp6dDEAScP3/+b70OEZE2YLJNRI3C1KlTIQgCBEGAgYEB3N3dERISgsrKSqWfe//+/QgNDX2uY58nQSYiIu3BRW2IqNEYMGAAtm/fjrKyMsTFxWHWrFnQ19fH8uXLaxxbXl4OAwODBjmvtbV1g7wOERFpH1a2iajRMDQ0RLNmzeDi4oKZM2eif//+iI2NBfC09SMsLAyOjo5o06YNACAzMxNjxoyBpaUlrK2tMWzYMKSnp8tfs6qqCgsWLIClpSWaNm2KJUuW4M/LD/y5jaSsrAxLly6Fk5MTDA0N4e7ujq1btyI9PV2+aIqVlRUEQcDUqVMBADKZDOHh4WjZsiWMjY3RqVMnfPPNNwrniYuLg4eHB4yNjdG3b1+FOJ/X0qVL4eHhARMTE7Rq1QpBQUGoqKiocdymTZvg5OQEExMTjBkzBnl5eQr7t2zZgnbt2sHIyAht27bFv//97xeOhYiImGwTUSNmbGyM8vJy+fP4+HikpKTg8OHD+OGHH1BRUYE333wT5ubmOH78OH799VeYmZlhwIAB8u9bt24dduzYgW3btiExMRGPHz/Gt99+W+95J0+ejK+//hqRkZG4cuUKNm3aBDMzMzg5OSE6OhoAkJKSgqysLHz66acAgPDwcHzxxRfYuHEjfv/9d8yfPx8TJ05EQkICgOqLgpEjR2LIkCE4f/48AgICsGzZshf+nZibm2PHjh24fPkyPv30U2zevBkbNmxQOCY1NRV79+7F999/j4MHDyIpKQmBgYHy/Tt37sSKFSsQFhaGK1euYPXq1QgKCkJUVNQLx0NEpPVEIqJGYMqUKeKwYcNEURRFmUwmHj58WDQ0NBQXLVok329vby+WlZXJv+fLL78U27RpI8pkMvm2srIy0djYWPzpp59EURRFBwcHMSIiQr6/oqJCbNGihfxcoiiKPj4+4rx580RRFMWUlBQRgHj48OFa4zx69KgIQMzJyZFvKy0tFU1MTMQTJ04oHDt9+nRx/PjxoiiK4vLly8X27dsr7F+6dGmN1/ozAOK3335b5/6PP/5YfOmll+TPV65cKerq6oq3b9+Wb/vxxx9FHR0dMSsrSxRFUXRzcxN37dql8DqhoaFijx49RFEUxZs3b4oAxKSkpDrPS0RE1dizTUSNxg8//AAzMzNUVFRAJpNhwoQJWLVqlXy/p6enQp92cnIyUlNTYW5urvA6paWlSEtLQ15eHrKystC9e3f5Pj09PXTt2rVGK8kT58+fh66uLnx8fJ477tTUVBQXF+P1119X2F5eXo4uXboAAK5cuaIQBwD06NHjuc/xxJ49exAZGYm0tDQUFhaisrISTZo0UTjG2dkZzZs3VziPTCZDSkoKzM3NkZaWhunTp2PGjBnyYyorK2FhYfHC8RARaTsm20TUaPTt2xeff/45DAwM4OjoCD09xY8wU1NTheeFhYV46aWXsHPnzhqvZWtr+z/FYGxs/MLfU1hYCAA4cOCAQpILVPehN5STJ0/C398fwcHBePPNN2FhYYHdu3dj3bp1Lxzr5s2bayT/urq6DRYrEZG2YLJNRI2Gqakp3N3dn/t4b29v7NmzB3Z2djWqu084ODjgv//9L/r06QOguoL722+/wdvbu9bjPT09IZPJkJCQgP79+9fY/6SyXlVVJd/Wvn17GBoaIiMjo86KeLt27eSDPZ84derUX/+Qzzhx4gRcXFzw/vvvy7fdunWrxnEZGRm4e/cuHB0d5efR0dFBmzZtYG9vD0dHR9y4cQP+/v4vdH4iIqqJAySJSGP5+/vDxsYGw4YNw/Hjx3Hz5k0cO3YMc+fOxe3btwEA8+bNw5o1axATE4OrV68iMDCw3jmyXV1dMWXKFEybNg0xMTHy19y7dy8AwMXFBYIg4IcffsCDBw9QWFgIc3NzLFq0CPPnz0dUVBTS0tJw7tw5fPbZZ/JBh//85z9x/fp1LF68GCkpKdi1axd27NjxQj9v69atkZGRgd27dyMtLQ2RkZG1DvY0MjLClClTkJycjOPHj2Pu3LkYM2YMmjVrBgAIDg5GeHg4IiMjce3aNVy8eBHbt2/H+vXrXygeIiJisk1EGszExAS//PILnJ2dMXLkSLRr1w7Tp09HaWmpvNK9cOFCTJo0CVOmTEGPHj1gbm6OESNG1Pu6n3/+Of7xj38gMDAQbdu2xYwZM1BUVAQAaN68OYKDg7Fs2TLY29tj9uzZAIDQ0FAEBQUhPDwc7dq1w4ABA3DgwAG0bNkSQHUfdXR0NGJiYtCpUyds3LgRq1evfqGfd+jQoZg/fz5mz56Nzp0748SJEwgKCqpxnLu7O0aOHImBAwfijTfegJeXl8LUfgEBAdiyZQu2b98OT09P+Pj4YMeOHfJYiYjo+QliXaOAiIiIiIjob2Flm4iIiIhISZhsExEREREpCZNtIiIiIiIlYbJNRERERKQkTLaJiIiIiJSEyTYRERERkZIw2SYiIiIiUhIm20RERERESsJkm4iIiIhISZhsExEREREpCZNtIiIiIiIl+X8+kTVMKdaRrgAAAABJRU5ErkJggg==",
      "text/plain": [
       "<Figure size 800x800 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "model_2m = joblib.load('models/gb_model_fullname_2M.joblib')\n",
    "preds = model_2m.predict(X_test)\n",
    "predicted = [id_to_race[p] for p in preds]\n",
    "actual = [id_to_race[p] for p in y_test.values]\n",
    "\n",
    "print(classification_report(actual, predicted))\n",
    "\n",
    "cm = confusion_matrix(actual, predicted, labels=list(race_to_id.keys()))\n",
    "cmp = ConfusionMatrixDisplay(cm, display_labels=list(race_to_id.keys()))\n",
    "fig, ax = plt.subplots(figsize=(8,8))\n",
    "cmp.plot(ax=ax, xticks_rotation='vertical')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 102,
   "id": "91ce0abe-eb80-40da-998d-90e47fefc052",
   "metadata": {
    "tags": []
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Accuracy for class asian: 0.11\n",
      "Accuracy for class hispanic: 0.56\n",
      "Accuracy for class nh_black: 0.09\n",
      "Accuracy for class nh_white: 0.96\n",
      "Accuracy for class other: 0.01\n"
     ]
    }
   ],
   "source": [
    "for class_id in id_to_race:\n",
    "    correct = np.sum((y_test == class_id) & (preds == class_id))\n",
    "    total = np.sum(y_test == class_id)\n",
    "    print(f'Accuracy for class {id_to_race[class_id]}: {correct / total:.2f}')"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "4a95f50f-487a-411b-b3e5-eaf6af672206",
   "metadata": {
    "tags": []
   },
   "source": [
    "### For Model trained  with complete dataset (20 iterations)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 103,
   "id": "7401ff7d-a24f-488b-b9f7-58ac0b45f5d2",
   "metadata": {
    "tags": []
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "              precision    recall  f1-score   support\n",
      "\n",
      "       asian       0.86      0.04      0.08     25756\n",
      "    hispanic       0.82      0.40      0.54    163525\n",
      "    nh_black       0.77      0.01      0.03    133471\n",
      "    nh_white       0.66      0.98      0.79    552737\n",
      "       other       0.38      0.00      0.00     26373\n",
      "\n",
      "    accuracy                           0.68    901862\n",
      "   macro avg       0.70      0.29      0.29    901862\n",
      "weighted avg       0.70      0.68      0.59    901862\n",
      "\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "<sklearn.metrics._plot.confusion_matrix.ConfusionMatrixDisplay at 0x7f255458ead0>"
      ]
     },
     "execution_count": 103,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAtsAAAKbCAYAAAAkKEbEAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjcuMSwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy/bCgiHAAAACXBIWXMAAA9hAAAPYQGoP6dpAACdpUlEQVR4nOzdd3iT5dcH8G+a7pEuOumGMip7l40CRYayh8gGBRkCMmWDjBdEpoCKCqLIFH4CsgTKLKtQ9qa0pXunO2mS949ASugWMhq+n+vKdZE8d57nJCHpycm570egUCgUICIiIiKit85I1wEQERERERkqJttERERERBrCZJuIiIiISEOYbBMRERERaQiTbSIiIiIiDWGyTURERESkIUy2iYiIiIg0xFjXARARERHRm8vNzYVEItHJsU1NTWFubq6TY+s7JttEREREFVxubi58va0RlyDTyfFdXV0RHh7OhLsITLaJiIiIKjiJRIK4BBkiQn0gstFul7A4Qw7vhs8gkUiYbBeByTYRERGRgRDZGEFkI9R1GPQKJttEREREBkIOBeSQa/2YVDyuRkJEREREpCGsbBMREREZCJlCDpmWC80yhXYr6RUNK9tERERERBrCyjYRERGRgVD2bGu3tM2e7ZKxsk1EREREpCFMtomIiIiINIRtJEREREQGQq71hf+ggyNWLKxsExERERFpCCvbRERERAZCplBAptDuhEVtH6+iYWWbiIiIiEhDmGwTEREREWkI20iIiIiIDATX2dY/rGwTEREREWkIK9tEREREBkIOBWSsbOsVVraJiIiIiDSElW0iIiIiA8Gebf3DyjYRERERkYYw2SYiIiIi0hC2kRAREREZCJ5BUv+wsk1EREREpCGsbBMREREZCPmLi7aPScVjZZuIiIiISEOYbBMRERERaQjbSIiIiIgMhEwHZ5DU9vEqGla2iYiIiIg0hJVtIiIiIgMhUygv2j4mFY+VbSIiIiIiDWGyTURERESkIWwjISIiIjIQXGdb/7CyTURERESkIaxsExERERkIOQSQQaD1Y1LxWNkmIiIiItIQVraJiIiIDIRcobxo+5hUPFa2iYiIiIg0hMk2EREREZGGsI2EiIiIyEDIdDBBUtvHq2hY2SYiIiIi0hBWtomIiIgMBCvb+oeVbSIiIiIiDWGyTURERESkIUy2iYiIiAyEXCHQyaWs5s+fD4FAoHapUaOGantubi7Gjh0LR0dHWFtbo1evXoiPj1fbR2RkJLp06QJLS0s4Oztj6tSpyM/PVxsTHByMBg0awMzMDFWrVsWWLVsKxfL999/Dx8cH5ubmaNq0KS5fvqy2vSyxlAWTbSIiIiLSmvfeew+xsbGqy7lz51TbJk2ahAMHDmD37t04ffo0YmJi0LNnT9V2mUyGLl26QCKR4MKFC9i6dSu2bNmCuXPnqsaEh4ejS5cuaNeuHcLCwjBx4kSMHDkSR48eVY3ZuXMnJk+ejHnz5uHatWuoW7cugoKCkJCQUOZYykqgUCh43h8iIiKiCkwsFsPW1hanb1eGtY12a6mZGXK0qRWN9PR0iESiEsfOnz8f+/fvR1hYWKFt6enpcHJywvbt29G7d28AwP3791GzZk2EhISgWbNmOHz4MLp27YqYmBi4uLgAADZt2oTp06cjMTERpqammD59Og4dOoTbt2+r9t2/f3+kpaXhyJEjAICmTZuicePGWL9+PQBALpfD09MT48ePx4wZM8oUS1mxsk1EREREb0wsFqtd8vLyihz36NEjuLu7w8/PDwMHDkRkZCQAIDQ0FFKpFO3bt1eNrVGjBry8vBASEgIACAkJQe3atVWJNgAEBQVBLBbjzp07qjGv7uPlmJf7kEgkCA0NVRtjZGSE9u3bq8aUJZayYrJNREREZCBkMNLJBQA8PT1ha2uruixdurRQfE2bNsWWLVtw5MgRbNy4EeHh4WjVqhUyMjIQFxcHU1NT2NnZqd3HxcUFcXFxAIC4uDi1RPvl9pfbShojFouRk5ODpKQkyGSyIse8uo/SYikrrrNNRERERG8sKipKrY3EzMys0JgPP/xQ9e86deqgadOm8Pb2xq5du2BhYaGVOLWNlW0iIiIiemMikUjtUlSy/To7OztUq1YNjx8/hqurKyQSCdLS0tTGxMfHw9XVFQDg6upaaEWQl9dLGyMSiWBhYYFKlSpBKBQWOebVfZQWS1kx2SYiIiIyEAodLPunKMfSf6/LzMzEkydP4ObmhoYNG8LExAQnTpxQbX/w4AEiIyMRGBgIAAgMDMStW7fUVg05fvw4RCIRAgICVGNe3cfLMS/3YWpqioYNG6qNkcvlOHHihGpMWWIpK7aREBEREZFWTJkyBd26dYO3tzdiYmIwb948CIVCDBgwALa2thgxYgQmT54MBwcHiEQijB8/HoGBgarVPzp27IiAgAAMGjQIy5cvR1xcHGbPno2xY8eqKumjR4/G+vXrMW3aNAwfPhwnT57Erl27cOjQIVUckydPxpAhQ9CoUSM0adIEq1evRlZWFoYNGwYAZYqlrJhsExERERkIGQSQ4b9Xmv/rMcvq+fPnGDBgAJKTk+Hk5ISWLVvi4sWLcHJyAgCsWrUKRkZG6NWrF/Ly8hAUFIQNGzao7i8UCnHw4EGMGTMGgYGBsLKywpAhQ7Bw4ULVGF9fXxw6dAiTJk3CmjVr4OHhgc2bNyMoKEg1pl+/fkhMTMTcuXMRFxeHevXq4ciRI2qTJkuLpay4zjYRERFRBfdyne1jt7xhpeV1trMy5OhYO6JM62y/i9izTURERESkIWwjISIiIjIQMoURZArt1lJl7JEoESvbREREREQawso2ERERkYGQQwC5lmupcrC0XRIm23pILpcjJiYGNjY2EAi0O6OYiIiI/huFQoGMjAy4u7vDyIjNA6TEZFsPxcTEwNPTU9dhEBER0X8QFRUFDw8PnRxb35f+excx2dZDNjY2AICW6AxjmOg4GiKioglFNroOgV6QiTN0HQIByIcU5/CP6u84EcBkWy+9bB0xhgmMBUy2iUg/CQWmug6BXhDwb4V+eNG6zBZQehWTbSIiIiIDoZul/zhBsiTs3iciIiIi0hBWtomIiIgMhHLpP+22sWj7eBUNK9tERERERBrCZJuIiIiISEPYRkJERERkIOQwgoxnkNQrrGwTEREREWkIK9tEREREBoJL/+kfVraJiIiIiDSElW0iIiIiAyGHEeTs2dYrrGwTEREREWkIk20iIiIiIg1hGwkRERGRgZApBJAptHtGR20fr6JhZZuIiIiISENY2SYiIiIyEDIdnNRGxgmSJWJlm4iIiIhIQ5hsExERERFpCNtIiIiIiAyEXGEEuZbPICnnGSRLxMo2EREREZGGsLJNREREZCA4QVL/sLJNRERERKQhrGwTERERGQg5tH+SGblWj1bxsLJNRERERKQhTLaJiIiIiDSEbSREREREBkIOI8i1XEvV9vEqGj47REREREQawso2ERERkYGQKYwg0/JJbbR9vIqGzw4RERERkYYw2SYiIiIi0hC2kRAREREZCDkEkEPb62xr93gVDSvbREREREQawso2ERERkYHgBEn9w2eHiIiIiEhDWNkmIiIiMhAyGEGm5Vqqto9X0fDZISIiIiLSECbbREREREQawjYSIiIiIgMhVwggV2h56T8tH6+iYWWbiIiIiEhDWNkmIiIiMhByHUyQlLN2WyI+O0REREREGsJkm4iIiIhIQ9hGQkRERGQg5AojyLV8RkdtH6+iYbJdCoFAgH379qF79+66DkXrajXNRJ8vEuFfOxuOrvmYP9wHIUdsVdtbfJiGLoOT4V87ByIHGcZ0qIandyyK2ZsC3/wejsbvZ6jtx8Y+HzPWR8K3Zg5s7GVITzZGyFERfl3qhuxMoRYeZcVR2uvx6VdxaPtxGpzcpZBKBHh8ywK/LnPFg+tWAAAXDwk+mRSPei0yYe8kRXK8CU7+ZY8/1zgjX/rqB6UCvUcn4sOByXD2kEKcIsTBrZXw51oXLT/iiqvfuHi06JwOz6p5kOQa4e5VS/y82A3Pn5irxnw4MBnteqSiau0cWNnI0bNGLWSJ+X++JH0/i0LzDknw8MuBJNcI966L8MtKH0SHW6rGjFvwCPUD0+DgLEFuthHuXhfh12998fzFGN/qmejz2XO81yAdIvt8xEeb4fAON/xvW2XVPmo3ScP//Xar0PEHtmyK1CTTMsdChZX2OfaqCcueo8vgZGya6459m520HCnR28NkuxSxsbGwt7fXdRg6YW4px9M75jj6pwPm/fKsyO13LlvhzAE7TPr2eYn76jEqCQpF4dsVciDkqAhb/s8V6cnGcPfNw7gl0bCxe45lY73f0iMxDKW9HtFPzfD9rMqIjTCFmbkCPT5LxNI/n2JY85pITzGGZ9VcGBkpsGa6B2LCTeFTIxcTVzyHuaUcPy10V+1nzKIYNGyTgZ8WuSP8njls7GQQ2cu0+EgrvjqBWTiwpRIehllCaKzA0BmxWPLnU4xqUx15OcqE2txCjqvBNrgabIMRX8fpOOKKoVbjdBzc7o6Ht6whFCowZNIzLN58G593bah6Xh/fsUbwAWckxJrBxjYfA8dF4Jufb2N4+8aQywWo+l4m0pNNsGJadSTFmqFm/QyMX/gIMrkAB/9wVzveqE4NkZ1Z8GcyLdmkXLFQYaV9jr3UvFM6ajTMQlIs05TykkEAGbS7FJ+2j1fR8H9xKVxdXXUdgs5cPSXC1VOiYref2OsAQFkxLYnfezno9Xkixn/ojx037qpty0w3xsHfKqmuJ0Sb4sBWR/QZk/gGkRum0l6PU/vUvxT+ON8dH36SAt+AHISds8HVYBGuBhfcPy7SDHuq5KHr4GRVsu1ZNRddByfh8/erq6qw8VEaeDAGbtZAP7XrKyd6YdftO/Cvk4Pbl6wBQFWpqxOYqfX4Kqq5o2qpXf9uZjXsCLkE//cycfuqsjp6ZJebantCNPDbah9s+PsanCvnIi7KAsf/Uv9Mj3tugZr1xGjRIalQsp2WbIqsjKL/TJYlFiqstM8xAHB0leKLb6Ix6xM/LNz2VEuREWnOO9Fkc+TIEbRs2RJ2dnZwdHRE165d8eTJEwCARCLBuHHj4ObmBnNzc3h7e2Pp0qWq+woEAuzfv191ffr06ahWrRosLS3h5+eHOXPmQCqVqrbPnz8f9erVw7Zt2+Dj4wNbW1v0798fGRkZWnu8+sTMQo4Z30fg+1mVkZpoUup4BxcpWnyYjpshVlqIznAZm8jR+dNkZKYb4end4lp7ACsbGTLSCqpwzTqKERtphqbtxdh68R62XrqLid9GwcYuXxthGywrkfKXgVefa3pzVjYvntf0ohNiMwsZOvSMQ2yUOZLizIrdj6VNfpH7WL//Gn4/cxGLf76FgPrpbxQLlY1AoMC0tZHYs9EJEQ/NS78DUQXwTnwqZGVlYfLkyahTpw4yMzMxd+5c9OjRA2FhYVi7di3+/vtv7Nq1C15eXoiKikJUVPGlPBsbG2zZsgXu7u64desWRo0aBRsbG0ybNk015smTJ9i/fz8OHjyI1NRU9O3bF8uWLcPixYuL3GdeXh7y8vJU18Vi8dt78Dr2+fxo3L1qhZCjJVd6ZmyIQGBQOswtFAg5JsKqKZ5aitCwNG0vxsyNETCzkCMl3hgz+1eBOKXot7m7Tx4+Hp6k1kLi5iWBS2UJWnVNx4oJnjASAp8viMbsHyMwvW8VbT0MgyIQKDB6QTRuX7ZExIPiv/hQ+QgECnz+9VPcCRUh4pH6l/MuA2IwfEo4LKzkiHpqgVnDa702L6FAzfpitP4wCfNGv6e6LSXRFOvmVcWj29YwMVUgqHcclv12C5P61cOTu9blioXKp+/YBMhkwP6fK5U+mIrECZL6551Itnv16qV2/ZdffoGTkxPu3r2LyMhI+Pv7o2XLlhAIBPD2LrlPePbs2ap/+/j4YMqUKdixY4dasi2Xy7FlyxbY2NgAAAYNGoQTJ04Um2wvXboUCxYs+K8PT28165iOei0y8UXHaqWO/WGeO/74zgWV/fIwfGYsPp8Xg/Vfe2ghSsMSdt4KX3SoBpFDPj4cmIJZP0RgQpeqSE9W/1XB0VWKxX88xZmDdji83VF1u8BIAVNzBVZ86YXop8pK4KqvPPH90UfwqJKrNsGPymbckmh418jFV92r6joUg/LF3Mfw9s/ClE/qFtp26oAzrl+wh4OTBD2HP8fM1fcxZUBdSCXqCYG3fxbmfn8H27/3wvXzBW1Y0eGWahMd710Xwc0rBz2GROPb6dXLFQuVXdXa2eg+Mgljg6oB7AEmA/JOfBV59OgRBgwYAD8/P4hEIvj4+AAAIiMjMXToUISFhaF69eqYMGECjh07VuK+du7ciRYtWsDV1RXW1taYPXs2IiMj1cb4+PioEm0AcHNzQ0JCQrH7nDlzJtLT01WXkirrFUm9Fplw85Hgr/u38U/kDfwTeQMAMOenZ1i+57Ha2NREE0Q9NsfFY7ZYM90D3YYmw8FZWtRuqQR5OULEPDPD/WtWWPWVJ2T5QKcBKWpjHFykWL77Me5etcKaqepfaFISTJAvhSrRBoDIR8oE27kyX4/yGrv4OZp2EGNa7ypIijXVdTgGY8ycx2jSNgUzBtdBcnzh9pDsTGPERFjg9lVbLPmyJjx9s9G8Q5LaGM8qWVjy6y0c3uWGHZu8Sj3mg5s2cPPOKXcsVHa1m2bBrlI+fr9yV/U3w9VTilHzYrD10t3Sd0AAABkKJklq70IleScq2926dYO3tzd++uknuLu7Qy6Xo1atWpBIJGjQoAHCw8Nx+PBh/Pvvv+jbty/at2+PPXv2FNpPSEgIBg4ciAULFiAoKAi2trbYsWMHVq5cqTbOxES9iigQCCCXy4uNz8zMDGZmhvchvXO9Mw5vd1C77cdTD/HDfHdcPFb8BBnBi4KGiWkRy5dQuQiMABOzgufR0VWZaD+6ZYmVkzyhUKhXj+5csYKxCeDmnYfYCOX/SQ8/ZYtT/HMmi2WnwNjF0WjeKR1Te1dFfJThvb91Q4Exc54gsH0yZgyug/joMv7SIlD/PPGqmoWlW27hxH4X/Lbap0y7qFIjCykJr74H/mMsVKx/99rj2ln1Np0l25/ixF57HNvpUMy9iPSfwSfbycnJePDgAX766Se0atUKAHDu3Dm1MSKRCP369UO/fv3Qu3dvdOrUCSkpKXBwUH9zX7hwAd7e3pg1a5bqtoiICM0/CB0xt5TB3bdgpRFXTwn83stBRpoQidGmsLHLh1NlKRxdlBVPzyq5AIDUBGOkJpqoLq9LiDZVJR+N3xfD3ikfD8IskJslhHf1XIycE4Pbly2Z3L2mpNdDnCLEJ18mIOSYCCnxJhA55OOjYUmo5CrF2QN2AJSJ9oo9j5EQbYqfFrrD1rFg0uPL1+n6GWs8ummByd9FYdM8dwgEyjaI0NPWatVuKtm4JdFo1yMV84f5IifTCPZOyvdIVoYQklzlD4r2TlLYO+fD3Vf5Zca3Rg6ys4RIjDZBRprBfzT/J1/MfYK2XROwcGwAcrKEsK+kfD9kZQghyRPC1SMHrTsn4dp5O6SnmKCSqwR9RkVBkmeEK6eVbSLe/spE+9o5e+zbUlm1D5kMEKcqP3M+HhyN+OfmiHhsCVMzOYJ6x6FOszTMHlGrzLFQ0Ur7u5KRqv5/Pz9fgNQEE7awlQN7tvWPwX+i29vbw9HRET/++CPc3NwQGRmJGTNmqLZ/9913cHNzQ/369WFkZITdu3fD1dUVdnZ2hfbl7++PyMhI7NixA40bN8ahQ4ewb98+LT4a7apWNwcr9j5RXR+9IAYAcGynPVZO8kKzjmJMWV3Q8vL1JmU7zbaVLvh9ZdmWTJTkGuHDgcn4fH4uTEwVSIwxwfnDtti5nidQeV1Jr8faGR7wqJqHOX2eQeQgQ0aqEA9vWOKrHlVVM/obtM5AZT8JKvtJsP2a+k+yQe7KXlOFQoC5Q3wx9ptofPvXE+RmG+HqKRv8uEB9STQqWbehyQCAb/96onb7txM9cXyX8kt8l8HJGPRVvGrbyv1PCo0hdV0/iQUALN+mfsKZ72ZWw7/7XCCRGOG9hun4eHA0rEX5SEs2we2rtvhqQF2kpygT6ZZBSbBzlOL9jxPw/scF7X3x0WYY9kETAICJiRwjpz+Fo4sEeblGCH9ghVnDa+PmJbsyx0JFK+3vCpEhEigURZ1qxLD8+++/mDBhAp4+fYrq1atj7dq1aNu2Lfbt24fExERs2LABjx49glAoROPGjbFixQrUr18fQOEzSE6bNg2//PIL8vLy0KVLFzRr1gzz589HWloaAOXSf/v370dYWJjq+KtXr8bq1avx7NmzMsUrFotha2uLtvgYxoLSl8sjItIFoajk9ZJJe2QGtIpVRZavkCIY/0N6ejpEWn5/vMwdZl/sCHNr7eYOuZlSfNPsmE4ed0XwTiTbFQ2TbSKqCJhs6w8m2/pBH5LtmSGddJJsLw08wmS7GGyyISIiIiLSEIPv2SYiIiJ6VygggFzL65QruC56iVjZJiIiIiLSECbbREREREQawjYSIiIiIgMhUxhBpuV1r7V9vIqGzw4RERERkYawsk1ERERkIOQKAeQK7U5Y1PbxKhpWtomIiIiINISVbSIiIiIDIYMRZFqupWr7eBUNnx0iIiIiIg1hsk1EREREpCFsIyEiIiIyEJwgqX9Y2SYiIiIi0hBWtomIiIgMhBxGkGu5lqrt41U0fHaIiIiIiDSEyTYRERERkYawjYSIiIjIQMgUAsi0PGFR28eraFjZJiIiIiLSEFa2iYiIiAwEl/7TP6xsExERERFpCCvbRERERAZCoTCCXKHdWqpCy8eraPjsEBERERFpCJNtIiIiIiINYRsJERERkYGQQQAZtLz0n5aPV9Gwsk1EREREpCGsbBMREREZCLlC+0vxyRVaPVyFw8o2EREREZGGMNkmIiIiItIQtpEQERERGQi5DtbZ1vbxKho+O0REREREGsLKNhEREZGBkEMAuZaX4tP28SoaVraJiIiIiDSElW0iIiIiAyFTCCDT8tJ/2j5eRcPKNhERERGRhjDZJiIiIiLSELaREBERERkILv2nf/jsEBERERFpCCvbRERERAZCDgHkWp6wyKX/SsbKNhERERGRhjDZJiIiIiLSELaREBERERkIhQ7OIKlgG0mJmGwTlUJY01/XIdArZPce6ToEekFax0/XIdALRufCdB0C0X+ybNkyzJw5E19++SVWr14NAMjNzcVXX32FHTt2IC8vD0FBQdiwYQNcXFxU94uMjMSYMWNw6tQpWFtbY8iQIVi6dCmMjQtS2+DgYEyePBl37tyBp6cnZs+ejaFDh6od//vvv8eKFSsQFxeHunXrYt26dWjSpIlqe1liKQ3bSIiIiIgMhFwh0Mnlv7hy5Qp++OEH1KlTR+32SZMm4cCBA9i9ezdOnz6NmJgY9OzZU7VdJpOhS5cukEgkuHDhArZu3YotW7Zg7ty5qjHh4eHo0qUL2rVrh7CwMEycOBEjR47E0aNHVWN27tyJyZMnY968ebh27Rrq1q2LoKAgJCQklDmWsmCyTURERERalZmZiYEDB+Knn36Cvb296vb09HT8/PPP+O677/D++++jYcOG+PXXX3HhwgVcvHgRAHDs2DHcvXsXv//+O+rVq4cPP/wQixYtwvfffw+JRAIA2LRpE3x9fbFy5UrUrFkT48aNQ+/evbFq1SrVsb777juMGjUKw4YNQ0BAADZt2gRLS0v88ssvZY6lLJhsExERERmIlye10fYFAMRisdolLy+v2DjHjh2LLl26oH379mq3h4aGQiqVqt1eo0YNeHl5ISQkBAAQEhKC2rVrq7VyBAUFQSwW486dO6oxr+87KChItQ+JRILQ0FC1MUZGRmjfvr1qTFliKQsm20RERET0xjw9PWFra6u6LF26tMhxO3bswLVr14rcHhcXB1NTU9jZ2and7uLigri4ONWY13umX14vbYxYLEZOTg6SkpIgk8mKHPPqPkqLpSw4QZKIiIiI3lhUVBREIpHqupmZWZFjvvzySxw/fhzm5ubaDE9nWNkmIiIiMhC6nCApEonULkUl26GhoUhISECDBg1gbGwMY2NjnD59GmvXroWxsTFcXFwgkUiQlpamdr/4+Hi4uroCAFxdXREfH19o+8ttJY0RiUSwsLBApUqVIBQKixzz6j5Ki6UsmGwTERERkVZ88MEHuHXrFsLCwlSXRo0aYeDAgap/m5iY4MSJE6r7PHjwAJGRkQgMDAQABAYG4tatW2qrhhw/fhwikQgBAQGqMa/u4+WYl/swNTVFw4YN1cbI5XKcOHFCNaZhw4alxlIWbCMhIiIiMhByHZzUpjzHs7GxQa1atdRus7KygqOjo+r2ESNGYPLkyXBwcIBIJML48eMRGBiIZs2aAQA6duyIgIAADBo0CMuXL0dcXBxmz56NsWPHqqrpo0ePxvr16zFt2jQMHz4cJ0+exK5du3Do0CHVcSdPnowhQ4agUaNGaNKkCVavXo2srCwMGzYMAGBra1tqLGXBZJuIiIiI9MaqVatgZGSEXr16qZ1I5iWhUIiDBw9izJgxCAwMhJWVFYYMGYKFCxeqxvj6+uLQoUOYNGkS1qxZAw8PD2zevBlBQUGqMf369UNiYiLmzp2LuLg41KtXD0eOHFGbNFlaLGUhUCgUijd4PkgDxGIxbG1t0RYfw1hgoutw3nk8g6R+4Rkk9Ye8ZT1dh0Av8AyS+iFfIUUw/of09HS1iYLa8DJ36HZsBEysTLV6bGmWBAc6/qyTx10RsLJNREREZCDe5IyOb3JMKh4nSBIRERERaQgr20REREQGgpVt/cPKNhERERGRhrCyTURERGQgWNnWP6xsExERERFpCJNtIiIiIiINYRsJERERkYFgG4n+YWWbiIiIiEhDWNkmIiIiMhAKAHJot9LMU5GXjJVtIiIiIiINYbJNRERERKQhbCMhIiIiMhCcIKl/WNkmIiIiItIQVraJiIiIDAQr2/qHlW0iIiIiIg1hsk1EREREpCFsIyEiIiIyEGwj0T+sbBMRERERaQgr20REREQGgpVt/cPKNhERERGRhrCyTURERGQgFAoBFFquNGv7eBUNK9tERERERBrCZJuIiIiISEPYRkJERERkIOQQQA4tT5DU8vEqGla2iYiIiIg0hJVtIiIiIgPBpf/0DyvbREREREQawmSbiIiIiEhD2EZCREREZCC4zrb+YWWbiIiIiEhDWNkmIiIiMhCcIKl/WNkmIiIiItIQVraJiIiIDAR7tvUPK9tERERERBqid5Xttm3bol69eli9enWR2wUCAfbt24fu3btrNa6SlBazIavVNBN9vkiEf+1sOLrmY/5wH4QcsS1y7IRlz9FlcDI2zXXHvs1OWo604nGslINho26hUZM4mJnlIzbaGqtWNMKjhw4AgEnTrqBDUITafa5edsHcma1U1+cuOg+/Kmmws89DZoYpwq4545efaiMl2UI1xscvDV9MuI5q1VORnmaGA/urYs/O6qrtzVtGo98n9+FWORPGQjmio62xb3c1nPzXW8PPQMVR2vvA3FKGEbNiERgkhsg+H3FRpvjfz5VwaFsl1Zjlex6jbvMstf0e+s0Ra2d4aO1xVAQW5lIM7XcdLZpEws42F4/DHbBhSxM8fKJ8Ls3NpBg5MBTNG0dBZJOHuARr7D9cEwePF/yfNjGRYfTgK2jb/BlMTGS4esMdazc3Q1q68n3h552C/t1v4b3qCbAV5SE+wRoHj1fDvsMBqn20bBKBrh0foIpPCkyM5Yh4bodtu+vi6o3K2n1CKrjy/A0hqqj0LtkuTWxsLOzt7XUdhpq//voLJiYmug5DJ8wt5Xh6xxxH/3TAvF+eFTuuead01GiYhaTYCvdfTiesrSX4ds0p3AxzwtwZLZGebgb3yhnIyDBVG3f1sgtWLW+sui6Vqv9YdTPMCTu310BqsjkcK+VgxOib+HpeCKZMeB8AYGEpxTf/dxZh11ywflUD+PiJMXHKVWRmmuDIIT8AQEaGCXb8UQPPo2wglRqhaWAsJk27irQ0M1y76qrhZ6JiKO198Pn8GNRrkYnl470QH2WKBm0yMH7pcyTHm+DisYLE4p/fHfDbioLnNC+HPz6+bvLoC/DxTMX/rW+J5BRLfND6KZbPOYYRkz5GcqoVRg+5gnq14rBsXSvEJ1qjYZ0YTBh5EckpFggJ9QIAjBlyGU0bRGPRd22QlW2KcSMuYf5XpzBxbmcAgL9fMtLSLfB/61ohIdkK71VPwMTPQiCXC/C/ozUBALVrxuPaTXf88mcDZGWZIqjdYyycfhLjv+6MJ88cdfb8VDRl/RtCZafQwQRJtpGUrMJlPq6u+vfH3cHBQdch6MzVUyJcPSUqcYyjqxRffBONWZ/4YeG2p1qKrGLr3f8BEhMtsGpFQSIdH2dVaJxUKkRqqnmx+9m/t5rq3wkJVtj9Zw3MWXgBQqEcMpkR2n0QCRNjOVavaIT8fCNERtjCr0oaevR+pEq2b91wVtvn//7yxwcdI/BerSQm2y+U9j4IaJSN47sdcDPEGgBw+A9HdBmUjOr1stWS7bwcI6Qmvptf3MvC1CQfrZpGYO7y93HrnvL/3rbd9dCsYRS6dXyALTsbIKBaIo6froKbd5Xb/zlRDV06PED1qkkICfWCpYUEnd5/jKVrWiHsjhsA4NsNLfDL6v2o6Z+Ie4+ccPSUv9px4xJsEFAtES2aRqqS7Y1bm6iN+eXPBghsFInAhs+ZbJdDWf6GEFV0elk2kcvlmDZtGhwcHODq6or58+ertgkEAuzfvx8AIJFIMG7cOLi5ucHc3Bze3t5YunSp2tiNGzfiww8/hIWFBfz8/LBnzx61Y02fPh3VqlWDpaUl/Pz8MGfOHEilUtX2+fPno169eti2bRt8fHxga2uL/v37IyMjQzWmbdu2mDhxoup6Xl4epk+fDk9PT5iZmaFq1ar4+eef3+6TVEEIBApMWxuJPRudEPGw+KSQ1DVrHoNHD+wxc24Itu85gHWb/kVQ58JfVGrXTcT2PQfw45YjGPvlNdiI8ordp7WNBO0+iMS9O46QyZRv/ZoBybh9ywn5+QUfBdeuusDTKwPW1pIi9qJA3frx8PDIwO1bbAUqq7tXLdGsYzocXaUAFKjbPBOV/fIQetpGbVy7nqnYdfs2fjj5AMNmxsLMQq6bgPWUUKiAUKiAVCpUu10iMUatGgkAgLsPnRDYMAqO9lkAFKj7Xiw83MQIvekOAKjmlwwTYzmu3XJX3T8qxhbxiVaoWS2h2GNbWkqRkWlW7HaBQAFLi3xkZJoWO4ZIGxQAFAotX3T9oPWcXla2t27dismTJ+PSpUsICQnB0KFD0aJFC3To0EFt3Nq1a/H3339j165d8PLyQlRUFKKiotTGzJkzB8uWLcOaNWuwbds29O/fH7du3ULNmsrqhI2NDbZs2QJ3d3fcunULo0aNgo2NDaZNm6bax5MnT7B//34cPHgQqamp6Nu3L5YtW4bFixcXGf/gwYMREhKCtWvXom7duggPD0dSUlKxjzcvLw95eQVJklgsLvdzpq/6jk2ATAbs/7lS6YNJxdUtC10+eop9e/yxc3sNVKueitHjwpCfb4QTx3wAAKFXXHHhbGXEx1nBzT0TQ0bcxsKl5/DV+Pchlxf8pDds1E10+/gJzC1kuHfXAfNntVBts3fIRdxrFfOXlXJ7h1xkvkgcLK2k2LbzIExM5JDLBfh+TX1cD3XR8LNgODbMrowvlz/H9mt3kS8F5HIB1kz1wO1L1qoxp/bZI+G5CZLjTeBbMxcjZsXCo0oeFo300V3geiYn1wR3HjhhYK8biIy2RWqaOdq1DEfNaomIiVN+cfn+l6aY+HkIdvywB/n5yp/TV/3QXFUJt7fLgURqhKxs9aQ4Nd0cDna5RR43oFoC2gaGY/ayD4qNrU+3OzA3l+J0iM/bebBEZDD0MtmuU6cO5s2bBwDw9/fH+vXrceLEiULJdmRkJPz9/dGyZUsIBAJ4exeesNWnTx+MHDkSALBo0SIcP34c69atw4YNGwAAs2fPVo318fHBlClTsGPHDrVkWy6XY8uWLbCxUX6YDxo0CCdOnCgy2X748CF27dqF48ePo3379gAAPz+/Eh/v0qVLsWDBglKfl4qmau1sdB+ZhLFB1QCwn6s8BAIFHj20x9afawMAnj62h7ePGJ27PVUl22dOearGPwu3RfhTW/zy+xHUrpuAG9cLEuG9O6vj2GFfOLtk45NBd/HV9CsvEu6yvyY52cYY91kHWFjko26DBIwacxNxsVaFWkyoaB8PT0KNhtmYO8QHCc9NUbtZFsYuiUZyvAmun1V+rhz+o6D14Nl9C6QkGGP57qdw885DbETxFdV3zf+tb4kpYy5gxw+7IZMJ8CjcAafO+6KabzIA4OMP76GmfyLm/N/7iE+0Qp2a8Rg/4iKSUy1w/ZVqdln5eKZiwbST2LanLkJvFj35sV2Lp/i09w3MW9EOaWKLIscQ0btLb5PtV7m5uSEhofDPe0OHDkWHDh1QvXp1dOrUCV27dkXHjh3VxgQGBha6HhYWprq+c+dOrF27Fk+ePEFmZiby8/MhEqn3j/n4+KgS7ZLiAYCwsDAIhUK0adOmTI8VAGbOnInJkyerrovFYnh6epZwj4qhdtMs2FXKx+9X7qpuExoDo+bFoPuoRAxpGlDCvd9tqSkWiIpQ/38YFWmDFq2fF3ufuFhrpKeZwr1yFm5cL7hdLDaDWGyG6Oc2iIywwbad/6BGQAru33VEaoo57O3VW0/s7XNfxFDQ9qNQCBAbo6zCPn1iBy8vMfoOeMBkuwxMzeUYOiMOC0f44PIJ5Wsafs8Cfu/loPfoRFWy/br71ywBAO4+TLZfFRsvwlfzO8HcTApLCylS0iwxa+JpxCZYw9QkH8MHXMf8Fe1w+bpyFZfwSAdU8UlFn253cP2WO1LTLGBqIoeVpUStum1vm4uUNPVWN6/KaVg+5xj++bcatv9Vt8h42jYPx+TRF7Dou7b/KZknetvkEECg5QKXnAW1Eullz/brK3sIBALI5YV7Fxs0aIDw8HAsWrQIOTk56Nu3L3r37l3m44SEhGDgwIHo3LkzDh48iOvXr2PWrFmQSNR7VcsaDwBYWJS/qmFmZgaRSKR2MQT/7rXH6A+qYUyHgktSrDH2bHTCrE9Krva/6+7edkRlzwy12yp7ZCAh3rLY+zhWyoaNSIKU5OJ7441evONNTGQAgHt3HVGrdiKEwoL/z/UbxiMq0kbVQlIUgaBgH1QyY2MFTEwVeP0jQy4DBEbFdzpWqaX80pOSwAmTRcnNM0FKmiWsrfLQqG40LlzxgrGxHCbGcihee1plcgGMBMobHz51hDTfCPVrx6q2e7ilw8UpC/ceFnx59PZIxbfzjuLY6Sr4dUeDImNo1+IppnxxHkvWtFYl90REr9PLynZ5iEQi9OvXD/369UPv3r3RqVMnpKSkqFYIuXjxIgYPHqwaf/HiRdSvXx8AcOHCBXh7e2PWrFmq7RER6usWl1ft2rUhl8tx+vRpVRuJITO3lMHdt+DLiaunBH7v5SAjTYjEaFNkpKr/F8vPFyA1wQTPn3CyZEn27fXHyrWn0PeTezgb7InqNVLwYZdwrF3VEABgbp6PTwbfxfmzlZGaYg4390wM/+wWYmOsEXpV2UJSvUYy/Kun4u7tSsjMMIGbexYGDbuDmGgr3LurbFkIPumFTwbfxcQpV7F7R3X4+IrxcY/H+HFjQRWv74D7ePTQHrExVjAxkaNR0zi83yEC368pOgF5F5X2PrhxwQqj5sRCkmuE+OcmqBOYhfa9U/HjAmUl1M07D+16pOHyCRtkpBrDNyAHn8+Pwc0QK4TfY1vCqxrVjQYAPI8Rwd01A58NuoqoaFscDa4KmcwIN+64YNSnociTGCMh0Qp1AuLRoc0TbNraCACQnWOKIyerYvTgK8jINEV2tinGDr+EOw+ccO+RctKvj2cqls89htAb7th78D3Y2+YAUPbap2coP7vatXiKaWPPYcOWJrj/yEk1Jk8iRHYOJ0mWVWnvHSo/nkFS/1ToZPu7776Dm5sb6tevDyMjI+zevRuurq6ws7NTjdm9ezcaNWqEli1b4o8//sDly5dVK4P4+/sjMjISO3bsQOPGjXHo0CHs27fvjWLy8fHBkCFDMHz4cNUEyYiICCQkJKBv375vtG99VK1uDlbsfaK6PnpBDADg2E57rJzkpauwKrxHDxzwzbxADB1xG58Muoe4WCv8sKEugk8on1O5XABfv3S07xgBK2sJUpItcO2qC7ZteQ/5L1ZqyMszRotW0fh06F2Ym+cjJdkcoVdcseOPZqox2VkmmD29Fb6YcB1rN52AON0M23+vqVr2D1Am9l9MuI5KTtmQ5AkRFWWDb5c2wZngit/q9LaU9j5YOsYbw7+OxfT1EbCxkyEh2hRb/s8NB39TfunJlwpQv1UGeoxMhLmlHIkxJjj3jy3+XM1JqK+ztJRixIBQVHLMRkamGc5d8sIvfzZQrbCzeHUbjPgkFDMnnIGNtQTxiVb49c/6aie12bi1CRSKK5j7VTBMjOUIfXFSm5daNYuAvW0u2rd+ivatC1YBikuwwqBxyl9Pu7R/CGNjBSaMvIQJIy+pxhwLroIVG1pq+FkwHPwbQu+CCp1s29jYYPny5Xj06BGEQiEaN26Mf/75B0ZGBd0xCxYswI4dO/DFF1/Azc0Nf/75JwIClL3CH330ESZNmoRx48YhLy8PXbp0wZw5c9SWGvwvNm7ciK+//hpffPEFkpOT4eXlha+//vqN9qmvboZYI8i96F7GorBPu+wuX3TH5YtF94BKJELMmdGqyG0vPQu3xcwppc8dePbUDtMmtit2+2+/1sJvv9YqdT/vstLeB6mJJiUmDokxppjaq6omQjM4Z0J8cKaEFT9S0y3w7caSk12pVIh1PzfDup+bFbl92+562La7Xon7mLKgU2mhUhmU928IlU6uEECg5Uqztk+iU9EIFIrXu9sMhz6e2r0sxGIxbG1t0RYfw1jAfk1dE9b0L30QaY3s3iNdh0AvyFvW03UI9ILRuTBdh0AA8hVSBON/SE9P1/r8q5e5Q61dUyG01O6kall2Hm73XaGTx10R6OUESSIiIiIiQ1Ch20iIiIiIqMDLszpq+5hUPINOtg24Q4aIiIiIKgCDTraJiIiI3iVc+k//sGebiIiIiEhDmGwTEREREWkI20iIiIiIDATbSPQPK9tERERERBrCyjYRERGRgeAZJPUPK9tERERERBrCyjYRERGRgeBJbfQPK9tERERERBrCZJuIiIiISEPYRkJERERkIJRtJNpe+k+rh6twWNkmIiIiItIQVraJiIiIDARPaqN/WNkmIiIiItIQJttERERERBrCNhIiIiIiA6F4cdH2Mal4rGwTEREREWkIK9tEREREBoITJPUPK9tERERERBrCyjYRERGRoWDTtt5hZZuIiIiISEOYbBMRERERaQjbSIiIiIgMhQ4mSIITJEvEyjYRERERkYawsk1ERERkIBQK5UXbx6TisbJNRERERKQhTLaJiIiIiDSEbSREREREBoJnkNQ/rGwTEREREWkIK9tEREREhkIh0P5SfKxsl4iVbSIiIiIiDWFlm4iIiMhAcOk//cPKNhERERGRhjDZJiIiIiLSELaREBERERkKxYuLto9JxWJlm4iIiIhIQ1jZJiIiIjIQPKmN/mFlm4iIiIhIQ5hsExERERFpCNtIiEohu/dI1yEQ6SWJvamuQ6AXzHUdAOkXTljUK6xsExERERFpCCvbRERERAaCEyT1DyvbREREREQawmSbiIiIiEhD2EZCREREZCh4Bkm9w8o2EREREZGGsLJNREREZDAELy7aPiYVh5VtIiIiIiINYWWbiIiIyFCwZ1vvsLJNRERERKQhTLaJiIiIiDSEbSREREREhoJtJHqHlW0iIiIiIg1hsk1ERERkKBQC3VzKaOPGjahTpw5EIhFEIhECAwNx+PBh1fbc3FyMHTsWjo6OsLa2Rq9evRAfH6+2j8jISHTp0gWWlpZwdnbG1KlTkZ+frzYmODgYDRo0gJmZGapWrYotW7YUiuX777+Hj48PzM3N0bRpU1y+fFlte1liKQsm20RERESkFR4eHli2bBlCQ0Nx9epVvP/++/j4449x584dAMCkSZNw4MAB7N69G6dPn0ZMTAx69uypur9MJkOXLl0gkUhw4cIFbN26FVu2bMHcuXNVY8LDw9GlSxe0a9cOYWFhmDhxIkaOHImjR4+qxuzcuROTJ0/GvHnzcO3aNdStWxdBQUFISEhQjSktlrISKBQKdtroGbFYDFtbW7TFxzAWmOg6HCKiIuV2a6LrEOgF8wOXSx9EGpevkCIY/0N6ejpEIpFWj/0yd/D8fgGMLMy1emx5Ti6ixs5DVFSU2uM2MzODmZlZqfd3cHDAihUr0Lt3bzg5OWH79u3o3bs3AOD+/fuoWbMmQkJC0KxZMxw+fBhdu3ZFTEwMXFxcAACbNm3C9OnTkZiYCFNTU0yfPh2HDh3C7du3Vcfo378/0tLScOTIEQBA06ZN0bhxY6xfv175GORyeHp6Yvz48ZgxYwbS09NLjaWsWNkmIiIiMhAKhW4uAODp6QlbW1vVZenSpSXGKpPJsGPHDmRlZSEwMBChoaGQSqVo3769akyNGjXg5eWFkJAQAEBISAhq166tSrQBICgoCGKxWFUdDwkJUdvHyzEv9yGRSBAaGqo2xsjICO3bt1eNKUssZcXVSIiIiIjojRVV2S7KrVu3EBgYiNzcXFhbW2Pfvn0ICAhAWFgYTE1NYWdnpzbexcUFcXFxAIC4uDi1RPvl9pfbShojFouRk5OD1NRUyGSyIsfcv39ftY/SYimrMiXbf//9d5l3+NFHH5UrACIiIiJ6S3S49N/LSY+lqV69OsLCwpCeno49e/ZgyJAhOH36tIaD1J0yJdvdu3cv084EAgFkMtmbxENEREREBszU1BRVq1YFADRs2BBXrlzBmjVr0K9fP0gkEqSlpalVlOPj4+Hq6goAcHV1LbRqyMsVQl4d8/qqIfHx8RCJRLCwsIBQKIRQKCxyzKv7KC2WsipTz7ZcLi/ThYk2ERERkQ7p+dJ/RZHL5cjLy0PDhg1hYmKCEydOqLY9ePAAkZGRCAwMBAAEBgbi1q1baquGHD9+HCKRCAEBAaoxr+7j5ZiX+zA1NUXDhg3Vxsjlcpw4cUI1piyxlNUb9Wzn5ubC3Fy7M16JiIiIqGKaOXMmPvzwQ3h5eSEjIwPbt29HcHAwjh49CltbW4wYMQKTJ0+Gg4MDRCIRxo8fj8DAQNXqHx07dkRAQAAGDRqE5cuXIy4uDrNnz8bYsWNVPeKjR4/G+vXrMW3aNAwfPhwnT57Erl27cOjQIVUckydPxpAhQ9CoUSM0adIEq1evRlZWFoYNGwYAZYqlrMqdbMtkMixZsgSbNm1CfHw8Hj58CD8/P8yZMwc+Pj4YMWJEeXdJRERERO+AhIQEDB48GLGxsbC1tUWdOnVw9OhRdOjQAQCwatUqGBkZoVevXsjLy0NQUBA2bNigur9QKMTBgwcxZswYBAYGwsrKCkOGDMHChQtVY3x9fXHo0CFMmjQJa9asgYeHBzZv3oygoCDVmH79+iExMRFz585FXFwc6tWrhyNHjqhNmiwtlrIq9zrbCxcuxNatW7Fw4UKMGjUKt2/fhp+fH3bu3InVq1eXezkUKozrbBNRRcB1tvUH19nWD3qxzvaahbpZZ/vLuTp53BVBudfZ/u233/Djjz9i4MCBEAqFqtvr1q2rWi6FiIiIiIj+QxtJdHS0agbpq+RyOaRS6VsJioiIiIj+Ax0u/UdFK3dlOyAgAGfPni10+549e1C/fv23EhQRERERkSEod2V77ty5GDJkCKKjoyGXy/HXX3/hwYMH+O2333Dw4EFNxEhEREREVCGVu7L98ccf48CBA/j3339hZWWFuXPn4t69ezhw4IBqJikRERER6UAFXGfb0P2ndbZbtWqF48ePv+1YiIiIiIgMyn8+qc3Vq1dx7949AMo+7oYNG761oIiIiIjoP+AESb1T7mT7+fPnGDBgAM6fP686V3xaWhqaN2+OHTt2wMPD423HSERERERUIZW7Z3vkyJGQSqW4d+8eUlJSkJKSgnv37kEul2PkyJGaiJGIiIiIykKhowsVq9yV7dOnT+PChQuoXr266rbq1atj3bp1aNWq1VsNjoiIiIioIit3ZdvT07PIk9fIZDK4u7u/laCIiIiIiAxBuZPtFStWYPz48bh69arqtqtXr+LLL7/Et99++1aDIyIiIqJyYBuJ3ilTG4m9vT0EgoI1FLOystC0aVMYGyvvnp+fD2NjYwwfPhzdu3fXSKBERERERBVNmZLt1atXazgMIiIiInpjujjJDE9qU6IyJdtDhgzRdBxERERERAbnP5/UBgByc3MhkUjUbhOJRG8UEBERERGRoSj3BMmsrCyMGzcOzs7OsLKygr29vdqFiIiIiHRDoNDNhYpX7mR72rRpOHnyJDZu3AgzMzNs3rwZCxYsgLu7O3777TdNxEhEREREVCGVu43kwIED+O2339C2bVsMGzYMrVq1QtWqVeHt7Y0//vgDAwcO1EScRERERFQaXSzFx8p2icpd2U5JSYGfnx8AZX92SkoKAKBly5Y4c+bM242OiIiIiKgCK3dl28/PD+Hh4fDy8kKNGjWwa9cuNGnSBAcOHICdnZ0GQiR91nVwEroMToaLp3KibMQDc/yxygVXT4lgY5ePQVPi0KBNJpzdJUhPMcaFI7bYutwV2RlCHUde8dRqmok+XyTCv3Y2HF3zMX+4D0KO2BY5dsKy5+gyOBmb5rpj32YntW1NPhBj4KR4+NbMgSTPCLcuWmHBcN9C+7Cxz8fG4w/h5C5Fzxq1kCXma/ZSaa+FuaUMI2bFIjBIDJF9PuKiTPG/nyvh0LZKAAAXDwl+u3yvyH1/85k3zh60Q4e+KZiyOqrIMX1rByA92eTtPzA9U8c/FgM63kQ1ryRUssvGrA0dcO6Gj2r70K6heL/xEzjbZyE/3wgPIith8/7GuPfMWTXGxjIXX/a/gOZ1IiFXCHDmmg/W7WqOnLyC58+vcjImDTiP6j5JSM8wx1+n3sOfx+qqtguN5Pj0wzAEBT5EJbtsRMXZ4od9TXD5jqdqjJFAjqHdrqFj00dwEOUgKd0SRy5Uw2//1AfAZdFK021oEnqPSYCDUz6e3rXAhtmV8SDMUtdhEb0V5U62hw0bhhs3bqBNmzaYMWMGunXrhvXr10MqleK777576wEKBALs27fvP50sJzg4GO3atUNqamqxXwS2bNmCiRMnIi0t7Y3iLM8xDUlirAl+WeKG6HAzCARAhz4pmP/rM4ztWA0QKODoko+fFroh8qE5nD0kmLDsORxdpPjmMx9dh17hmFvK8fSOOY7+6YB5vzwrdlzzTumo0TALSbGF394tO6dh4orn+HWZK8LOe0EoVMCnRm6R+5m8Mgrh98zh5C59Ww/BYJT2Wnw+Pwb1WmRi+XgvxEeZokGbDIxf+hzJ8Sa4eMwWiTEm6F83QO0+nT9NRu8xibhy0gYAcPpvO1w9ZaM2ZsrqKJiYyd+JRBsALEzz8fi5A/45Xw3fjPm30Pbn8bZY82cLxCTZwMwkH33a38a3E//BJ7P7IT3TAgAwZ8QpONhm46vVnWEslGPGkNOY8ulZLPr5fQCApbkE3355GKH3K2PlHy3hVzkV04ecRmaOKQ6crQkAGNn9Cjo0eYwVv7dCZJwdmgQ8xzejj2Ps8o/wKEr5BeqTTjfwcZu7WPprWzyLtUd170TMGHIGWTmm2HuqlpaesYqpzUep+GxeDNbN8MD9a5boMSoRi7c/xYhW1d+Z/+tk2MqdbE+aNEn17/bt2+P+/fsIDQ1F1apVUadOnbcaHOm/S8fVK6tb/s8NXQcno0bDLBz90xGLRvmotsVGmGHL/7lh2rpIGAkVkMtY7SmPq6dEuHqq5KU1HV2l+OKbaMz6xA8Ltz1V22YkVGD0whj89I0bjv7pqLo98pF5of10HZwEK5EMf6xyQZMPMt7OAzAgpb0WAY2ycXy3A26GWAMADv/hiC6DklG9XjYuHrOFXC5AaqJ6EtH8w3ScOWCH3GzlLwiSXCNIcgs6/Wwd8lG3RSZWfeWhgUekny7d8cSlV6rHr/v3SlW169/vboauLR+gikcKrt2vDG/XVDSt9RyfLemOBxHKX3jW7GyO/xt3BBv2NEVyuhU6NHkME2M5/m9ra+TLhHgW64Cqnsno0/6WKtnu2PQxth2uh0u3vQAA/zsTgIY1o9G3wy0s/qUdAOA9v3icD/PGxRdj4pJt8EHjJ6jhmwiceutPjUHp+VkSjmx3wLGdDgCAtdM90OQDMYIGpGDXehcdR0f05srds/06b29v9OzZk4k2wchIgTYfp8LMUo57V62KHGMlkiE704iJtgYIBApMWxuJPRudEPGwcALtXzsHTu5SKOQCfH/sAbZfv4Nvfn8K7+o5auO8/HPxyaR4rPjSCwo5X6f/4u5VSzTrmA5HVykABeo2z0RlvzyEnrYpcnzV2tmoWisXR/90KHaf7fukIC9HgLOH7DQTdAVnLJShW6v7yMg2xZMo5ZfJ9/wSkJFlqkq0ASD0XmXIFQIE+Ca8GBOPG49ckS8raJO6cscD3q7psLbMAwCYGMsgkaq3UeVJjVG7Spzq+p2nLmhQIwYezmkAgCoeyahdNR6Xbhf/ZYEAYxM5/Otk49rZgveGQiHA9bM2CGiYrcPIKi4BdLD0n64ftJ4rU2V77dq1Zd7hhAkTyhVA27ZtUadOHZibm2Pz5s0wNTXF6NGjMX/+fNWYpKQk9OjRA0ePHkXlypWxcuVKfPTRR2U+xvnz5zFz5kw8fPgQ9erVw+bNm1GrVtE/6z158gSTJ0/GxYsXkZWVhZo1a2Lp0qVo3769akxeXh7mzp2L7du3IyEhAZ6enpg5cyZGjBhRaH/Z2dno1asXxGIxDh06VGRrSV5eHvLy8lTXxWJxmR+bPvCpkYPVBx7D1EyOnCwjLBzhU2S1VOSQj08mxuPw745F7IXeVN+xCZDJgP0/Vypyu6u38v/Yp1/F4cf57oiLMkXv0YlYsfcJRrSsgYw0Y5iYyjFzQwQ2L3JHYrQp3LwkRe6LSrZhdmV8ufw5tl+7i3wpIJcLsGaqB25fsi5yfKcBKYh4aIa7xXxJBYCgASk4tc9erdpNQGDtCMwdeRLmpvlITrfElNWdkZ6l/PxxsM1GaoaF2niZ3AgZWWZwEOW8GJOD2CT1L0EpL+7jKMpGZrYZrtz1QN/2t3DjkRtiEkVoWCMareuHw+iVxYX/OFIPluZSbFuwG3KFAEYCBTb/rzH+vaxefSd1IgcZhMZAWqJ6OpKaZAzPqnnF3IuoYilTsr1q1aoy7UwgEJQ72QaArVu3YvLkybh06RJCQkIwdOhQtGjRAh06dAAALFiwAMuXL8eKFSuwbt06DBw4EBEREXBwKL4K9KqpU6dizZo1cHV1xddff41u3brh4cOHMDEp3AuWmZmJzp07Y/HixTAzM8Nvv/2Gbt264cGDB/DyUv48OHjwYISEhGDt2rWoW7cuwsPDkZSUVGhfaWlp6NKlC6ytrXH8+HFYWhY92WPp0qVYsGBBWZ8uvfP8iRm+6FANljYytOqajilrIjG1Z1W1hNvSWoZFv4Uj8qE5tq101WG0hqlq7Wx0H5mEsUHVUFyNwehFjvbnGhec+8cOALBykid+D72LVl3T8c/vjhg2MxaRj81x8i+eoOpNfDw8CTUaZmPuEB8kPDdF7WZZGLskGsnxJrh+Vj2xMzWXo12PVGxfXfzP5TUbZsG7Wh6Wj/fSdOgVzvUH7hj5TU/YWueia8v7mP/Zvxi9rDvSXkuy38TanYGYOugsti3YDYUCiEkU4fCF6ujc/IFqTLuGT9GhyWMs+vl9PIuxR1XPZIzrG4KkNEscvVjtrcVCVCqFQHnR9jGpWGVKtsPDwzUaRJ06dTBv3jwAgL+/P9avX48TJ06oku2hQ4diwIABAIAlS5Zg7dq1uHz5Mjp16lSm/c+bN0+1r61bt8LDwwP79u1D3759C42tW7cu6tYtmIW+aNEi7Nu3D3///TfGjRuHhw8fYteuXTh+/Liq2v1yKcRXxcXFoV+/fvD398f27dthampabHwzZ87E5MmTVdfFYjE8PSvOT4/5UiPEPDMDADy+ZYnq9bLRfWQi1k5XPgYLKxkWb3+KnCwjLBjhA1k+35RvW+2mWbCrlI/fr9xV3SY0BkbNi0H3UYkY0jQAKfHKL5eRj8xUY6QSI8RFmMG5srKCXa9lJnxq5KJVlzTlgBcv1e7bt/HnWhds+5ZflEpjai7H0BlxWDjCB5dPKPu6w+9ZwO+9HPQenVgo2W7VJQ1mFgr8u7v44kGnT1Lw+LY5Ht/i6gyvy5WYIDrRFtGJtrgb7oI/Fu5ElxYP8MeRekhJt4S9jXqblNBIDhurPKSIlcl4SroF7EXqYxxe3CdZrHy+0zMtMHtjR5ga50NknYekNEt83vMyYl6piI/pdQl/HK2Lk1erAACexjjAxTEDAz8MY7JdAnGKELJ8wM4pX+12+0r5SE0s97QyIr2kF/+TX+/3dnNzQ0JCQpHbraysIBKJ1LaXJjAwUPVvBwcHVK9eHffuFb3sVmZmJubPn49Dhw4hNjYW+fn5yMnJQWRkJAAgLCwMQqEQbdq0KfGYHTp0QJMmTbBz504IhSUvmWZmZgYzM7MSx1QkAgFgYqr8edXSWploSyUCzBvqC2kefwLXhH/32uPaWfUWhSXbn+LEXnvVpKNHNy0gyRXAo0oe7lxWjhUaK+DiKUH8c+WXwUUjfWBqLlfto3q9HHy1Kgpf9aiKmGfFf2GkAsbGCpiYKiCXq98ulwECo8JnfggakIKLx0RITyn649jcUobW3dLw61I3TYRrcARGCpgYywAAd546w8ZKgmpeiXgYqezbrl89BkYCBe6GO78Y44KR3a9CaCSHTK78fGoUEI2IOFtkZqt/LkvyjZGUZgyhkRyt6z9DcGhBocXMNL/QHAe53Eit1YQKy5ca4dFNS9RvmaFaPlMgUKBey0z8vYUth2QY9CLZfr2dQyAQQP7KX6rStr9NU6ZMwfHjx/Htt9+iatWqsLCwQO/evSGRKCt/FhZl+2myS5cu2Lt3L+7evYvatWtrJFZ9MGxmLK6ctEFitCksrGVo1yMNdZpnYtYnfrC0lmHJn09hZiHH8vE+sLSWwdJa+UcwPdkYck6+KxdzSxncfQt6qF09JfB7LwcZaUIkRpsiI1X97ZyfL0BqggmeP1G282RnCnFomyMGfRWPxBhTJDw3Qe8xiQCAsweVf+RiI9STC1sH5esV+cic62y/orTX4sYFK4yaEwtJrhHin5ugTmAW2vdOxY8L3NX24+6Th9rNsjDn08LrnL/U5uM0CIUKnNj77rX2WJhJUdmpYA6LW6UMVPVIhjjLDOIsMwzqHIbzN7yQnG4JW+tc9Gh7F5XsshEcqnw+I+Lscem2B6YOOouVf7SEsVCOiQPO4+TVKkhOV/bH/3u5KoZ0vYbpg09j+9G68K2cil7v38b3u5upjlvTJwGV7LPwOMoRTnZZGNrtGowECvx5tKAQdOGmFz7tHIb4FGs8i7WHv2cS+ra/hX8usKpdmr9+rIQpq6Pw8IYlHlxXLv1nbinHsR1laxWl1/AMknpHL5JtTbt48aKq3zo1NRUPHz5EzZo1ixx7/vx5DB06FD169ACgrHQ/e/ZMtb127dqQy+U4ffq02qTJ1y1btgzW1tb44IMPEBwcjICAgGLHVmR2lfIxdW0kHJzzkZ0hRPg9c8z6xA/XztigTmAmar6YTb4l5L7a/QY3qamqplLZVKubgxV7n6iuj14QAwA4ttMeKyeVrZf3p0XukMkEmLY2Eqbmcjy4bonpfaogM/2d+Ch4a0p7LZaO8cbwr2MxfX0EbOxkSIg2xZb/c8PB39QrdUH9U5AUa1LsKiWAcvLk+cO27+SXnereiVjz1SHV9XF9LwIADl/wx3d/tISXaxqCmj2ErXUuxFnmuP/MCRNWdMOz2IIkbdHP7TBxwAWsmvQP5ArgzDVfrN3ZXLU9K9cUU9Z8iEkDzuPHWfuRnmmOrYfqq5b9AwBTExlGfnQVbk4ZyMkzxqVbnlj8S1tk5hR8OV2zozlGfByKSZ+ch72N8qQ2f5+tga0HG2jyKTIIp/+2h62jDIOnxsHeKR9P71hg1kBfpCVxjW0yDO/EX9iFCxfC0dERLi4umDVrFipVqlTsSXL8/f3x119/oVu3bhAIBJgzZ45aFd3HxwdDhgzB8OHDVRMkIyIikJCQUKgH/Ntvv4VMJsP777+P4OBg1KhRQ5MPUydWfVV8b/nNEGsEudctdjuVT3mfzyFNC3/Bk+UL8NNCd/y00L2Ie7z5Md8VpT0vqYkmZfoC9OsyN/y6rOT2kEkf+Zc7PkMR9tAdbT4fVez2OZs6lLqPjGxz1QlsivM02hHjvy1+hasbj9wwZEGfEveRk2eK9bsCsX5XYInjqGh//1oJf/9a9EpKVE6sbOudd6KBdtmyZfjyyy/RsGFDxMXF4cCBA8VOWPzuu+9gb2+P5s2bo1u3bggKCkKDBuqViY0bN6J379744osvUKNGDYwaNQpZWVlF7m/VqlXo27cv3n//fTx8+PCtPzYiIiIi0l8ChUJR7u8jZ8+exQ8//IAnT55gz549qFy5MrZt2wZfX1+0bNlSE3G+U8RiMWxtbdEWH8NYwJ/RiEg/5XZrousQ6AXzA5d1HQIByFdIEYz/IT09HSJRyWf8fdte5g4+ixfDyLzwuS40SZ6bi2ezZunkcVcE5a5s7927F0FBQbCwsMD169dVJ2NJT0/HkiVL3nqAREREREQVVbmT7W+++QabNm3CTz/9pLZKSIsWLXDt2rW3GlxpRo8eDWtr6yIvo0eP1mosRERERESvK/cEyQcPHqB169aFbre1tUVaWtrbiKnMFi5ciClTphS5jT9jEBER0TuHEyT1TrmTbVdXVzx+/Bg+Pj5qt587d67IMylqkrOzM5ydnbV6TCIiIiKisip3G8moUaPw5Zdf4tKlSxAIBIiJicEff/yBKVOmYMyYMZqIkYiIiIjKQqGjCxWr3JXtGTNmQC6X44MPPkB2djZat24NMzMzTJkyBePHj9dEjEREREREFVK5k22BQIBZs2Zh6tSpePz4MTIzMxEQEABra2tNxEdEREREVGH95zNImpqaGuwpyImIiIgqIoFCedH2Mal45U6227VrB4FAUOz2kydPvlFARERERESGotzJdr169dSuS6VShIWF4fbt2xgyZMjbiouIiIiIykshUF60fUwqVrmT7VWrVhV5+/z585GZmfnGARERERERGYpyL/1XnE8//RS//PLL29odEREREZUXl/7TO28t2Q4JCYG5ufnb2h0RERERUYVX7jaSnj17ql1XKBSIjY3F1atXMWfOnLcWGBERERFRRVfuZNvW1lbtupGREapXr46FCxeiY8eOby0wIiIiIiofLv2nf8qVbMtkMgwbNgy1a9eGvb29pmIiIiIiIjII5erZFgqF6NixI9LS0jQUDhERERH9Z5wgqXfKPUGyVq1aePr0qSZiISIiIiIyKOVOtr/55htMmTIFBw8eRGxsLMRisdqFiIiIiIiUytyzvXDhQnz11Vfo3LkzAOCjjz5SO227QqGAQCCATCZ7+1ESERERUel0MEGSbSQlK3OyvWDBAowePRqnTp3SZDxERERERAajzMm2QqH82tKmTRuNBUNEREREb0AXExZZ2S5RuXq2X20bISIiIiKikpVrne1q1aqVmnCnpKS8UUBERERERIaiXMn2ggULCp1BkoiIiIj0BNtI9E65ku3+/fvD2dlZU7EQERERERmUMifb7NcmIiIi0m8CHSz9p/WlBiuYMk+QfLkaCRERERERlU2ZK9tyuVyTcRARERERGZxyn66diIiIiIjKhsk2EREREZGGlGs1EiIiIiLSY1z6T++wsk1EREREpCGsbBMREREZCC79p39Y2SYiIiIi0hAm20REREREGsI2EiIiIiJDwrYOvcLKNhERERGRhrCyTURERGQouPSf3mFlm4iIiIhIQ1jZJiIiIjIQXPpP/zDZJiqFwMRU1yHQKxRSia5DoBdO//CjrkOgF4IO1NN1CERUDLaREBERERFpCCvbRERERIaCEyT1DivbREREREQawso2ERERkYHgBEn9w8o2EREREZGGMNkmIiIiItIQtpEQERERGQpOkNQ7rGwTEREREWkIK9tEREREhoKVbb3DyjYRERERkYawsk1ERERkILj0n/5hZZuIiIiISEOYbBMRERERaQjbSIiIiIgMBSdI6h1WtomIiIiINISVbSIiIiJDwcq23mFlm4iIiIhIQ5hsExERERFpCNtIiIiIiAwE19nWP6xsExERERFpCCvbRERERIaCEyT1DivbREREREQawso2ERERkYFgz7b+YWWbiIiIiEhDmGwTEREREWkI20iIiIiIDAUnSOodVraJiIiIiDSEyTYRERGRoVDo6FJGS5cuRePGjWFjYwNnZ2d0794dDx48UBuTm5uLsWPHwtHREdbW1ujVqxfi4+PVxkRGRqJLly6wtLSEs7Mzpk6divz8fLUxwcHBaNCgAczMzFC1alVs2bKlUDzff/89fHx8YG5ujqZNm+Ly5cvljqU0TLaJiIiISCtOnz6NsWPH4uLFizh+/DikUik6duyIrKws1ZhJkybhwIED2L17N06fPo2YmBj07NlTtV0mk6FLly6QSCS4cOECtm7dii1btmDu3LmqMeHh4ejSpQvatWuHsLAwTJw4ESNHjsTRo0dVY3bu3InJkydj3rx5uHbtGurWrYugoCAkJCSUOZayECgUCnba6BmxWAxbW1u0xccwFpjoOpx3nsDEVNch0CsUUomuQ6AXjsaE6ToEeiHIvZ6uQyAA+QopgvE/pKenQyQSafXYL3OHml8sgdDMXKvHluXl4t6Gr//T405MTISzszNOnz6N1q1bIz09HU5OTti+fTt69+4NALh//z5q1qyJkJAQNGvWDIcPH0bXrl0RExMDFxcXAMCmTZswffp0JCYmwtTUFNOnT8ehQ4dw+/Zt1bH69++PtLQ0HDlyBADQtGlTNG7cGOvXrwcAyOVyeHp6Yvz48ZgxY0aZYikLVraJiIiIDIRARxdAmfC/esnLyys13vT0dACAg4MDACA0NBRSqRTt27dXjalRowa8vLwQEhICAAgJCUHt2rVViTYABAUFQSwW486dO6oxr+7j5ZiX+5BIJAgNDVUbY2RkhPbt26vGlCWWsmCyTURERERvzNPTE7a2tqrL0qVLSxwvl8sxceJEtGjRArVq1QIAxMXFwdTUFHZ2dmpjXVxcEBcXpxrzaqL9cvvLbSWNEYvFyMnJQVJSEmQyWZFjXt1HabGUBZf+IyIiIjIUOlz6LyoqSq2NxMzMrMS7jR07Frdv38a5c+c0GZ3OsbJNRERERG9MJBKpXUpKtseNG4eDBw/i1KlT8PDwUN3u6uoKiUSCtLQ0tfHx8fFwdXVVjXl9RZCX10sbIxKJYGFhgUqVKkEoFBY55tV9lBZLWTDZJiIiIjIQAoVuLmWlUCgwbtw47Nu3DydPnoSvr6/a9oYNG8LExAQnTpxQ3fbgwQNERkYiMDAQABAYGIhbt26prRpy/PhxiEQiBAQEqMa8uo+XY17uw9TUFA0bNlQbI5fLceLECdWYssRSFmwjISIiIiKtGDt2LLZv347//e9/sLGxUfU+29rawsLCAra2thgxYgQmT54MBwcHiEQijB8/HoGBgarVPzp27IiAgAAMGjQIy5cvR1xcHGbPno2xY8eqqumjR4/G+vXrMW3aNAwfPhwnT57Erl27cOjQIVUskydPxpAhQ9CoUSM0adIEq1evRlZWFoYNG6aKqbRYyoLJNhERERFpxcaNGwEAbdu2Vbv9119/xdChQwEAq1atgpGREXr16oW8vDwEBQVhw4YNqrFCoRAHDx7EmDFjEBgYCCsrKwwZMgQLFy5UjfH19cWhQ4cwadIkrFmzBh4eHti8eTOCgoJUY/r164fExETMnTsXcXFxqFevHo4cOaI2abK0WMqC62zrIa6zrV+4zrZ+4Trb+oPrbOsPrrOtH/Rhne33PtfNOtt3fvhv62y/C9izTURERESkIWwjISIiIjIk7FnQK6xsExERERFpCJNtIiIiIiINYRsJERERkYEo77rXb+uYVDxWtomIiIiINISVbXojXQcnocvgZLh4Kpdji3hgjj9WueDqKRFcPCT47fK9Iu/3zWfeOHvQTouRVhy1mmSg9+ex8K+dDUcXKRaMqoqQY/YaPWa3wfHo/Vkc7J2keHrPEhvmeeHhDesiRiqwaOsjNG6brpW4Kop+4+LRonM6PKvmQZJrhLtXLfHzYjc8f1Kw/NbyPY9Rt3mW2v0O/eaItTMKTlPsVFmC8Uufo26LTORmCXF8tz1+WeIGuUygtceiT7Z964rfv1M/JbJHlVz8fPa+2m0KBTD7Uz9cPSXCvJ/D0fzDdADAkzvm2LXeBbcvW0GcagwXDwm6DE5Cj5FJqvsmxxvjxwWV8eimBWLCzfDxiCSMWRittv9nD8zx2wpXPL5pifjnpvh8QTR6jkpUGzO4SQDinxdeJrTbkESMWxpd6HZS121oEnqPSYCDUz6e3rXAhtmV8SDMUtdhVUwKaH+CJCvbJdLrZFsgEGDfvn3o3r27xo/l4+ODiRMnYuLEiUVuf/bsGXx9fXH9+nXUq1dP4/FUFImxJvhliRuiw80gEAAd+qRg/q/PMLZjNUQ9NkP/ugFq4zt/mozeYxJx5aSNjiLWf+aWMoTfs8SxXU6Y++PjN95fh95J6NA7CdP61yhye+uuyRg1OwrrZnnjQZg1ug+Px+JtDzGyXW2kJ6uv895jRDy4Mn9hdQKzcGBLJTwMs4TQWIGhM2Kx5M+nGNWmOvJyhKpx//zugN9WFCSPeTkFPy4aGSmw6LdwpCYaY9JH/nBwlmLq2kjIpAL8usxNq49Hn3hXz8GynU9U14XCwv8B9/3kBEER30ce37SEXaV8TF8fASd3Ke5etcKaqZ4wMgI+Hq5MuKUSI9g55mPAl/HY96NTkTHk5Qjg5iVB665p+GF+5SLHrD38QO1L0bP75pjZvypadUsvz8N9J7X5KBWfzYvBuhkeuH/NEj1GJWLx9qcY0ap6oc8goopIr5NtfeLp6YnY2FhUqlQJABAcHIx27dohNTUVdnZ2ug1Ohy4dt1W7vuX/3NB1cDJqNMxCxENzpCaqf1A2/zAdZw7YITdbCCra1WA7XA22K3a7iakcQ6Y+R9uPUmAtkuHZAwv8sswDNy/+txMJ9BwZjyM7nHB8tzLRWPe1N5q8n4agvknYtbEgyfMLyEbPUXGY0O09/Hk17D8dy1DNGuindn3lRC/sun0H/nVycPtSwS8EeTlGhd4TLzVokwGvarmY0S8AaUkmeHrHAr8td8WIWbHYttIF+dJ3s+tPKAQcnPOL3f7ktgX2/uCEdYcfYkC9WmrbggakqF1385bg3lVLnD9sq0q2XT0lGLNIWXk+tsOxyGNUr5eD6vVyAAC/LHEvcoydo0zt+s71tnDzyUOdwMwSHh0BQM/PknBkuwOO7XQAAKyd7oEmH4gRNCAFu9a7lHJveh17tvXPu/np/R8IhUK4urrC2JjfT4pjZKRAm49TYWYpx72rVoW2V62djaq1cnH0TwcdRGc4vlgYgZoNsrB0XBWMCXoPZ/+xxzdbH8LdJ7fc+zI2kcO/dhaunytI1BUKAa6fE6Fmg4Ikwcxchulrn+D7Od7FJotUwEqkTLwy0tS/VLbrmYpdt2/jh5MPMGxmLMws5KptAY2y8ey+OdKSCp7fq8E2sBLJ4V29/K+toYgON8WA+u9hSLOaWDbWCwnPC56f3GwBlo31xtjFz0tMyF+VlSGEjZ2s9IFvQCoR4OReewT1Ty6y4k4FjE3k8K+TjWtnC37tVCgEuH7WBgENs3UYGdHbo9Nku23btpgwYQKmTZsGBwcHuLq6Yv78+WpjkpKS0KNHD1haWsLf3x9///13mfbdqFEjfPvtt6rr3bt3h4mJCTIzlQnE8+fPIRAI8Phxwc/02dnZGD58OGxsbODl5YUff/xRte3Zs2cQCAQICwvDs2fP0K5dOwCAvb09BAIBhg4dCgCQy+VYunQpfH19YWFhgbp162LPnj0lxpqXlwexWKx2qUh8auRg/6NbOPjsJiYse46FI3wQ+ajwqWI7DUhBxEMz3C0iEaeycXLPQ8c+SVg8pgruXLFBbKQ59v7ohjtXbdCxT1LpO3iNyD4fQmOoJXiA8rq9k1R1/fO5UbgXao2Lx9mjXRqBQIHRC6Jx+7IlIh5YqG4/tc8ey8d5YVrvKtixzhkf9ErFtHWRqu32TlKkJqp/mX/5utg7lS2RNDQ1GmRhyupILP7jCcYve464SDN81cMf2ZnKP10/zK+MgEZZaN6pbJ+Zd65Y4vTf9ug8MFmTYePCEVtkioXo2Del9MHvOJGDTPkZ9Nr//dQk43f2/z0ZHp2Xabdu3YrJkyfj0qVLCAkJwdChQ9GiRQt06NABALBgwQIsX74cK1aswLp16zBw4EBERETAwaHk6mibNm0QHByMKVOmQKFQ4OzZs7Czs8O5c+fQqVMnnD59GpUrV0bVqlVV91m5ciUWLVqEr7/+Gnv27MGYMWPQpk0bVK9eXW3fnp6e2Lt3L3r16oUHDx5AJBLBwkL5R3Xp0qX4/fffsWnTJvj7++PMmTP49NNP4eTkhDZt2hQZ69KlS7FgwYI3eRp16vkTM3zRoRosbWRo1TUdU9ZEYmrPqmoJt6m5HO16pGL7av4k+CZ8a+RAaAz8HHxL7XYTUwXEqcoqqpN7Hn7897Zqm1CogNBEgX13Q1W37fjeDTu/L/rn8Nc1a5+Kus3FGNv5vbfwCAzfuCXR8K6Ri6+6V1W7/fAfBS0Kz+5bICXBGMt3P4Wbdx5iI8y0HWaF0Pj9DNW//QJyUaN+NgY1CcCZv+1g65iPsPM22HDsQZn29ey+ORYM88Onk+PQsG1G6Xd4A0f/dEDjdmI4ujJZJB3gBEm9o/Nku06dOpg3bx4AwN/fH+vXr8eJEydUyfbQoUMxYMAAAMCSJUuwdu1aXL58GZ06dSpxv23btsXPP/8MmUyG27dvw9TUFP369UNwcDA6deqE4ODgQslv586d8cUXXwAApk+fjlWrVuHUqVOFkm2hUKhK9p2dnVU923l5eViyZAn+/fdfBAYGAgD8/Pxw7tw5/PDDD8Um2zNnzsTkyZNV18ViMTw9PUt97vRFvtQIMc+UycLjW5aoXi8b3UcmYu30gsfQqksazCwU+Hc3W0jehLmlDLJ8YHzXAMheW6HiZR98crwpvviwIDFu0SkVLT9Mxf99WdBXnJGmfOuLU40hywfsKknV9mVXSapqF6nbPANu3nnYe+ua2pjZmx7jzmWbYidevovGLn6Oph3E+KpHFSTFFl6Z4lX3rylXWnD3USbbqYkmqF5f/Wfzl6/L6xXvd5W1rQwefnmIeWaG8PsWiH1mip41aquNWTTKB7WaZmHF3oJfLSMemmF63yr48NMkfDIxXqMxxj83wfWzNpizOVyjxzEU4hSh8jPotSq2faV8/r8ng6Hz/8l16tRRu+7m5oaEhIQit1tZWUEkEqltL06rVq2QkZGB69ev48KFC2jTpg3atm2LZcuWAQBOnz6NqVOnFhuLQCCAq6trmY710uPHj5Gdna36ovCSRCJB/fr1i72fmZkZzMwMp7IlECgrra8KGpCCi8dESE/R+X+5Cu3JHSsIjQFbx3zcuVL0ii5ymQCxEQW/KqQnm0CSa6R220v5UiM8umWFei3EqmX8BAIF6rUQ48BW5a8Quza64ciOSmr3++H4Hfy40AsXT9i9pUdW0SkwdnE0mndKx9TeVREfVfr7uUotZR92SoLyS83dq5boPyEeto5S1QoMDVpnIktshMiHhV+7d1FOlhFiIkzxQS8pWn+Uhg8/UW8H+fz9Gvh8fjSadSxoK3n2wBzT+1RBhz4pGDYjTuMxHtvhCLtK+WjavmK1A+pKvtQIj25aon7LDIQcUU64FwgUqNcyE39vKXrCKpWMEyT1j84zHxMT9V5RgUAAuVxe5u3FsbOzQ926dREcHIyQkBB06NABrVu3Rr9+/fDw4UM8evSoUKX5vx7rpZf94IcOHULlyurLQxlSMv2qYTNjceWkDRKjTWFhLUO7Hmmo0zwTsz4pqKK6++ShdrMszPnUV4eRVhzmljK4++Sprrt65sEvIBsZaUJEh5vj5D4HTF31FD9944nHd6xg6yBF/RZihN+3xOWTduU+3l+bXTBlZTge3bTCgxtW6DE8HuaWchzbrUywUxNNipwUmRBjWqak8l0wbkk02vVIxfxhvsjJNFL1u2dlCCHJNYKbdx7a9UjD5RM2yEg1hm9ADj6fH4ObIVYIv6dsQbt22gaRD80xbV0kfv7GHfZOUgydHocDWypBKnk357L/uMAdzTqmw9lDiuQ4Y2z71g1CI6Btj1TYOcqKnBTpXFkKVy/luv/P7ptjWp8qaNQ2Az0/T0RKgvJPnpFQobZ6yJPbytcgJ8sI6clCPLltAWNTObyrKd+HUolA9YVHKhUgOdYET25bwNxKhsq+EtV+5HLg2E4HtO+TAqHO/7pWHH/9WAlTVkfh4Q1LPLiuXPrP3FKOYzv4SygZBoP+OGjTpg1OnTqFy5cvY/HixXBwcEDNmjWxePFiuLm5oVq1av9536amyp+IZbKCD+yAgACYmZkhMjKy2JYRQ2NXKR9T10bCwTkf2RlChN8zx6xP/HDtTEHVNah/CpJiTRB6mmtrl0W1OllYvrOgD/XzuVEAgOO7HbFyih9WTvHFgPGxGDU7Co4uUohTjXH/uhUu/ccq85mDjrB1zMegydHKk9rctcTswdUKTZqk4nUbqqywfvvXE7Xbv53oieO7HJAvFaB+qwz0GKlMIhJjTHDuH1v8+cocBrlcgLmDfTF+2XOsOvAIudlG+He3A7auUD+py7skKdYES7/wQUaqELaO+XivcRZWH3xYaJm94pw9aIf0ZBOc2OuAE3sLEjflCbfuqq5/0bGgVfDRTUuc2uegNiY53kRtzJ5NztizyRl1AjPV2lWun7FBQrQpgvpzYmR5nP7bHraOMgyeGgd7p3w8vWOBWQN9+RlEBsOgk+22bdti3bp1cHJyQo0aNVS3rV+/Hn369HmjfXt7e0MgEODgwYPo3LkzLCwsYGNjgylTpmDSpEmQy+Vo2bIl0tPTcf78eYhEIgwZMuRtPCy9suqr0nvLf13m9k6flKO8bl4UoZN342K3y/KN8Puqyvh9VdEn13jd8T2VcHxPpRLHHNjqomobKYuS4nsXBbnXLXF7YowppvaqWuIYAEiINsWcQX6ljntXfL0polzjj8aEqV0fNCUOg6aU3jry+v1e5+opKXUMADRsm1GmcVTY379Wwt+/lvw5RWXECZJ6x6B/m2zVqhXkcrlalblt27aQyWRo27btG+27cuXKWLBgAWbMmAEXFxeMGzcOALBo0SLMmTMHS5cuRc2aNdGpUyccOnQIvr5soSAiIiJ61wgUCp58Wd+IxWLY2tqiLT6GsYA/o+mawKTkVSVIuxRSSemDSCtYxdUfQe71dB0CAchXSBGM/yE9PR0i0X87q+9/9TJ3qDN0CYSm2p1ULZPk4uaWr3XyuCsCg65sExERERHpUoVNtkePHg1ra+siL6NHj9Z1eEREREREFXeC5MKFCzFlypQit/EnDCIiInoXcZ1t/VNhk21nZ2c4OzvrOgwiIiIiomJV2GSbiIiIiF7Dpf/0ToXt2SYiIiIi0nesbBMREREZCIFCAYGWV3XW9vEqGla2iYiIiIg0hMk2EREREZGGsI2EiIiIyFBwgqTeYWWbiIiIiEhDWNkmIiIiMhA8qY3+YWWbiIiIiEhDmGwTEREREWkI20iIiIiIDAUnSOodVraJiIiIiDSElW0iIiIiA8EJkvqHlW0iIiIiIg1hZZuIiIjIULBnW++wsk1EREREpCFMtomIiIiINIRtJEREREQGghMk9Q8r20REREREGsLKNhEREZGh4ARJvcPKNhERERGRhjDZJiIiIiLSELaREBERERkQTljUL6xsExERERFpCCvbRERERIZCoVBetH1MKhYr20REREREGsLKNhEREZGB4Elt9A8r20REREREGsJkm4iIiIhIQ9hGQkRERGQoeAZJvcPKNhERERGRhrCyTURERGQgBHLlRdvHpOKxsk1EREREpCFMtomIiIiINIRtJERERESGghMk9Q4r20REREREGsLKNhEREZGB4Bkk9Q+TbaJSKKQSXYdApJc6B7TRdQikkq7rAIioGEy2iYiIiAyFQqG8aPuYVCz2bBMRERERaQiTbSIiIiIiDWEbCREREZGB4ARJ/cPKNhERERGRhrCyTURERGQoeFIbvcPKNhERERGRhjDZJiIiIiLSELaREBERERkITpDUP6xsExERERFpCCvbRERERIaCZ5DUO6xsExERERFpCCvbRERERAaCPdv6h5VtIiIiIiINYbJNRERERKQhbCMhIiIiMhQ8g6TeYWWbiIiIiEhDWNkmIiIiMhCcIKl/WNkmIiIiItIQJttERERERBrCNhIiIiIiQyFXKC/aPiYVi5VtIiIiIiINYWWbiIiIyFBw6T+9w8o2EREREZGGsLJNREREZCAE0MHSf9o9XIXDyjYRERERkYYw2SYiIiIi0hC2kRAREREZCoVCedH2MalYrGwTEREREWkIK9tEREREBkKg0MEESRa2S8TKNhERERGRhjDZJiIiIiLSELaREBERERkKnkFS77CyTURERESkIaxsExERERkIgUIBgZaX4tP28SoaVraJiIiIiDSEyTYRERERkYawjYSIiIjIUMhfXLR9TCoWK9tERERERBrCyjYRERGRgeAESf3DyjYRERERkYawsk1ERERkKHhSG73DyjYRERERac2ZM2fQrVs3uLu7QyAQYP/+/WrbFQoF5s6dCzc3N1hYWKB9+/Z49OiR2piUlBQMHDgQIpEIdnZ2GDFiBDIzM9XG3Lx5E61atYK5uTk8PT2xfPnyQrHs3r0bNWrUgLm5OWrXro1//vmn3LGUhsk2EREREWlNVlYW6tati++//77I7cuXL8fatWuxadMmXLp0CVZWVggKCkJubq5qzMCBA3Hnzh0cP34cBw8exJkzZ/DZZ5+ptovFYnTs2BHe3t4IDQ3FihUrMH/+fPz444+qMRcuXMCAAQMwYsQIXL9+Hd27d0f37t1x+/btcsVSGoFCwa52fSMWi2Fra4u2+BjGAhNdh6OmVtNM9PkiEf61s+Homo/5w30QcsRWtb3Fh2noMjgZ/rVzIHKQYUyHanh6x0JtHyZmcnw2LwZtP0qDiZkCocE2WDezMtKS9Oux6pvSnvuvVkWiY79UtftcPWWDWQP9VNcHTIhHk/Zi+L2Xg3yJAL1q1i72eDb2+dh4/CGc3KXoWaMWssTCt/+gDJiFlQxDpsWh+YfpsHPMx5M7Ftg4pzIe3rCE0FiBodNj0fj9DLh5S5AlNsL1szb4eYkbUuIrzvtAaGdb+qC3qO+oSDRvnwQPvxxIco1wL0yEX1b6IvqZZRGjFVj4w200apWKReMDEHKiktrW9t3j0GNINCr7ZCM70xjnjlbChm/8C+3FzSsH6/Zeg1wG9G3WQm3bx4Oeo0v/WDi55UGcaoxzx5ywZZUvpBLt17FkaelaP+bbUtJ7paLJV0gRjP8hPT0dIpFIq8d+mTu0bjEHxsbmWj12fn4uzpxf9J8et0AgwL59+9C9e3cAykqyu7s7vvrqK0yZMgUAkJ6eDhcXF2zZsgX9+/fHvXv3EBAQgCtXrqBRo0YAgCNHjqBz5854/vw53N3dsXHjRsyaNQtxcXEwNTUFAMyYMQP79+/H/fv3AQD9+vVDVlYWDh48qIqnWbNmqFevHjZt2lSmWMqCle1itG3bFhMnTtR1GHrH3FKOp3fMsf5rj2K337lshZ+XuBW7j9HzY9CsgxjffO6NKT2rwMFFirk/P9NQxIajtOceAK6ctEH/ugGqy9IvvNS2G5sqcOaAHQ5trVTMHgpMXhmF8Hva/cA2JJNWRqFB6wwsH++F0R9UR+hpGyzb+QSOrlKYWchRtXYOtq92wdggfywc6QOPKnlYsCVc12HrtVqN0nHwT3dMHlAPs0bWhtBYgcWbb8HMQlZobPfB0SiulNRjyHMM/vIZdm/2xOiPGuHrEbURet6h0DihsRzTV9zDndDCyUPbLgkYNjkc2zd44/OujbB6TjW0/jARQyfyNSyvkt4rVLGIxWK1S15eXrn3ER4ejri4OLRv3151m62tLZo2bYqQkBAAQEhICOzs7FSJNgC0b98eRkZGuHTpkmpM69atVYk2AAQFBeHBgwdITU1VjXn1OC/HvDxOWWIpi3d+gmRwcDDatWuH1NRU2NnZ6TocvXf1lAhXTxX/rfXEXuUfLBcPSZHbLW1kCBqQgmVjvXDjvA0A4LvJnth85gFqNMjC/WtWbz9oA1Hacw8AUokAqYnFV0a3fesKAOjQN6XE/XQdnAQrkQx/rHJBkw8yyh/sO87UXI6WndMxf5gvbl+yBgD8vtIVzTqI0XVwErYud8PM/lXU7vP9rMpYd/gRnCpLkBhtWtRu33lzP1f/Jea7r6thx/mL8A/IwO1QO9XtfjUy0XPoc3zZtwH+OHNR7T7WIikGTXiGBWPfw42L9qrbnz20LnS8wROe4Xm4JcIu2qFmPbHatpr1xLh73RbBh5wBAAkx5jj9jxOq1+b7pTzK8l6h8hEolBdtHxMAPD091W6fN28e5s+fX659xcXFAQBcXFzUbndxcVFti4uLg7Ozs9p2Y2NjODg4qI3x9fUttI+X2+zt7REXF1fqcUqLpSze+WRbmyQSido3rHeRf51smJgqcP2sjeq2qMfmiH9ugpoNs5lsv6E6gZnYefMOMtKFuHHOGluWuyIjtXxvcy//XHwyKR5fdvWHm1fRX5qoZEKhAkJjQJInULs9L1eA95pkFXkfK5EMcjmQlc52nbKyslFWtDPSC75gmpnLMG3FfWz4pipSkwp/3tZvngYjIwUcnSXYdOAKLK1kuBcmwk/L/ZAUV/BLTt2mqWgZlIRxPRugRYekQvu5FyZCu27xqFZbjIe3RHD1yEGjVik4ecCl0Fgq3n95r5D+ioqKUmsjMTMz02E0+uOdaCPJy8vDhAkT4OzsDHNzc7Rs2RJXrlzBs2fP0K5dOwCAvb09BAIBhg4dqrqfXC7HtGnT4ODgAFdX10LfztLS0jBy5Eg4OTlBJBLh/fffx40bN1Tb58+fj3r16mHz5s3w9fWFuTl/kndwzockT1Co/zct0RgOzvzJ8E1cDbbBii+9ML2vH35e7IbagZlY/PtTGBmVvcRhYirHzA0R2LzIndXVN5CTJcTdq5b4ZGI8HFykMDJS4P2eqajZMBsOLvmFxpuYyTFiViyC99shO5PJdlkIBAp8PuMJ7oSKEPG44Ev6qBlPcO+6CBdPFt0q5eqRA4ER0O+zSPy4rAoWTwyAtW0+Fm++BWMT5TmnbWylmLTkIVZ9XQ05WUV/WQ0+5Izf1/tgxe838PeNs/jl2BXcumKHXT96FTmeilbe9wrpN5FIpHb5L8m2q6vyF9j4+Hi12+Pj41XbXF1dkZCQoLY9Pz8fKSkpamOK2serxyhuzKvbS4ulLN6JZHvatGnYu3cvtm7dimvXrqFq1aoICgqCjY0N9u7dCwB48OABYmNjsWbNGtX9tm7dCisrK1y6dAnLly/HwoULcfz4cdX2Pn36ICEhAYcPH0ZoaCgaNGiADz74ACkpBT/RP378GHv37sVff/2FsLCwIuPLy8sr1OdEVF6n/2ePi8ds8ey+BUKO2GLuYF9Ur5+DOs0zS7/zC8NmxiLysTlO/mVf+mAq0fLxXhAIgD+v38XBZzfRfUQigvfbQSFXHyc0VmDWDxGAAFg3o/h+fFL3xZzH8PbPwrIpNVW3NW2XjLpN0/DDsirF3k9gBJiYKLBpSRVcO++ABzdF+L8pNeDunYM6TdIAABMWPkTwQWe11pTX1W6chr6fRWLDwqqY0LsBFo0PQOM2KRgwOuJtPcR3RlnfK1RGCoVuLm+Jr68vXF1dceLECdVtYrEYly5dQmBgIAAgMDAQaWlpCA0NVY05efIk5HI5mjZtqhpz5swZSKUFhbzjx4+jevXqsLe3V4159Tgvx7w8TlliKQuDbyPJysrCxo0bsWXLFnz44YcAgJ9++gnHjx/HL7/8gsaNGwMAnJ2dC/Vs16lTB/PmzQMA+Pv7Y/369Thx4gQ6dOiAc+fO4fLly0hISFB9c/v222+xf/9+7NmzR7X8jEQiwW+//QYnJ6diY1y6dCkWLFjwth+6XkpJMIapmQJWIpladdvOKR8pCRVnFYaKIC7SDGnJQrj7SBB2rmz3qdcyEz41ctGqS5ryhhe/7O6+fRt/rnVR9XxT6WIjzDC1V1WYWchgZSNHSoIJvt70DLERBb8YKBPtZ3CpLMG0vlVY1S6jMbMeo0mbZEwbXBfJ8QWVs7pN0+DmmYvdF8+rjf969V3cCbXFjKF1kZqofP4jnxRUw8WpphCnmsDJLU+1n2btktFrWJRygAAQCoEDN89g7fxqOP6XKwZNeIaTf7vg6F5lT/GzR1Ywt5Rh/PxH2PGDFxQK9bYIKl5Z3itkWDIzM/H48WPV9fDwcISFhcHBwQFeXl6YOHEivvnmG/j7+8PX1xdz5syBu7u7asWSmjVrolOnThg1ahQ2bdoEqVSKcePGoX///nB3dwcAfPLJJ1iwYAFGjBiB6dOn4/bt21izZg1WrVqlOu6XX36JNm3aYOXKlejSpQt27NiBq1evqpYHFAgEpcZSFgafbD958gRSqRQtWhQs2WRiYoImTZrg3r17qmS7KHXq1FG77ubmpvrZ4saNG8jMzISjo6PamJycHDx58kR13dvbu8REGwBmzpyJyZMnq66LxeJCkwwMxaOblpBKBKjfMgPn/rEDAHhUyYWLhxT3QiveMk/6rJKbBCJ7GVISyv42XzTSB6bmBeWk6vVy8NWqKHzVoypinvEP33+RlyNEXo4Q1rb5aNgmA5u/Uf4heJloV/aVYFrvKuXurX83KTBm1hMEtk/CjKF1ER+tvqzo7s2eOLpH/Qvhxr9D8dP/VcGlU8rJ23evKftJPXyzVYm6ta0UInspEmKU17/6pL5a+1WzD5LRZ0QUvvqkHpITlO8DM3N5ocqrXKZMsAWCt1roe2cU916h8hHIlRdtH7M8rl69qmrjBaDKgYYMGYItW7Zg2rRpyMrKwmeffYa0tDS0bNkSR44cUWvH/eOPPzBu3Dh88MEHMDIyQq9evbB27VrVdltbWxw7dgxjx45Fw4YNUalSJcydO1dtLe7mzZtj+/btmD17Nr7++mv4+/tj//79qFWrlmpMWWIpDT/dS2Biol5pFQgEkMuV/6MyMzPh5uaG4ODgQvd7tUJuZVX6hD8zM7MKM4nA3FIGd9+CSXOunhL4vZeDjDQhEqNNYWOXD6fKUji6KH+28ayiXPQ9NcEYqYkmyM4Q4uifDvhsfgwy0oyRlWGEsYujcfeqJSdHlqKk5z4jVYhPv4rHuUO2SE0wgZtPHkbOjkVMuClCgwsmozpVlsDGTgbnyhIYCQG/93IAADHhpsjNFiI2Qv3/oa2DcgJa5CNzrrNdTg3biCEQAFFPzFDZV4KRc2IQ9dgcx3Y6QGiswJyfnqFq7RzMHewLI6EC9k7K90xGmhD50neiw6/cvpjzGG27JGDhuPeQkyWEfSXl+yErQwhJnhCpSaZFTopMjDVTJebREZYIOeGIz2c+wbp5/sjONMbQSeF4Hm6Jm5ftAABRT9W/+PvXyoBcDrXe8MvBDugxJBpP7lnjwU0buHvlYtCEZ7gc7Ai5nFXt8ijpvUKGqW3btijpNC8CgQALFy7EwoULix3j4OCA7du3l3icOnXq4OzZsyWO6dOnD/r06fNGsZTG4JPtKlWqwNTUFOfPn4e3tzcAQCqV4sqVK5g4caJqdRCZrPA6rSVp0KAB4uLiYGxsDB8fn7cdtt6qVjcHK/YWVO5HL4gBABzbaY+Vk7zQrKMYU1ZHqbZ/vSkSALBtpQt+X6msOG2a7w65Apjz0zOYmClwNdgG62dW1uKjqJhKeu7XzfSAb80cdOiTCiuRDMnxxrh22gZbl7uqnWBj8JQ4tRPfbDz+EAAwtVcV3AwpvPQZ/XdWIjmGzYxFJTcpMtKEOP+PLX5d5gZZvgAuHhIEBinnZmz896Ha/fhaFK/rgFgAwPLfbqrd/t3X1fDv/rK3OH07ozo+m/EE8zfegUIB3Lpiizmf1YIsv+xfcv7c5A2FQoDBXz6Do7ME6akmuHzKAVvX+JZ+Z1JT0nuF/oO33ENd5mNSsd6JM0hOnDgRu3fvxs8//wwvLy8sX74cf//9N548eYLs7Gx4enri119/RefOnWFhYQFra2u0bdsW9erVw+rVq1X76d69O+zs7LBlyxYoFAq0bt0aGRkZWL58OapVq4aYmBgcOnQIPXr0QKNGjTB//nzs37+/2ImRxdHnM0gSEb2k7TNIUvEq8hkkDYk+nEGybZNZOjmDZPDlxTp53BXBO/Fb5bJly9CrVy8MGjQIDRo0wOPHj3H06FHY29ujcuXKWLBgAWbMmAEXFxeMGzeuTPsUCAT4559/0Lp1awwbNgzVqlVD//79ERERUWjxcyIiIiJ6N70Tle2KhpVtIqoIWNnWH6xs6we9qGw31lFl+wor28V5JyrbRERERES6YPATJImIiIjeFQKFAgItNy1o+3gVDSvbREREREQawmSbiIiIiEhD2EZCREREZCi4zrbeYWWbiIiIiEhDWNkmIiIiMhQKAHIdHJOKxco2EREREZGGsLJNREREZCC49J/+YWWbiIiIiEhDmGwTEREREWkI20iIiIiIDIUCOlj6T7uHq2hY2SYiIiIi0hBWtomIiIgMBU9qo3dY2SYiIiIi0hAm20REREREGsI2EiIiIiJDIQcg0MExqVisbBMRERERaQgr20REREQGgmeQ1D+sbBMRERERaQgr20RERESGgkv/6R1WtomIiIiINITJNhERERGRhrCNhIiIiMhQsI1E77CyTURERESkIaxsExERERkKVrb1DivbREREREQawmSbiIiIiEhD2EZCREREZCjkAAQ6OCYVi5VtIiIiIiINYWWbiIiIyEAIFAoItDxhUdvHq2hY2SYiIiIi0hBWtomIiIgMBZf+0zusbBMRERERaQiTbSIiIiIiDWEbCREREZGhkCsAgZbbOuRsIykJK9tERERERBrCyjYRERGRoeAESb3DyjYRERERkYYw2SYiIiIi0hC2kRAREREZDB20kYBtJCVhsq2HFC/eJPmQ8v8vEekthUKi6xDoBZlCqusQCC/+bqPg7zgRwGRbL2VkZAAAzuEfHUdCRFSCNF0HQKSfMjIyYGtrq5uDc4Kk3mGyrYfc3d0RFRUFGxsbCAQCXYfzn4jFYnh6eiIqKgoikUjX4bzz+HroD74W+oOvhf4wlNdCoVAgIyMD7u7uug6F9AiTbT1kZGQEDw8PXYfxVohEogr9wWlo+HroD74W+oOvhf4whNdCZxXtl+QKaL0HlSe1KRFXIyEiIiIi0hAm20REREREGsI2EtIIMzMzzJs3D2ZmZroOhcDXQ5/wtdAffC30B1+Lt0ghV160fUwqlkDB9WmIiIiIKjSxWAxbW1u09/oCxkba/dKSL8/Dv5EbkJ6eXuF77jWBlW0iIiIiQ8Gl//QOe7aJiIiIiDSEyTYRERERkYawjYSIiIjIUHCdbb3DyjYRERERkYawsk1ERERkKDhBUu+wsk1EpGFisbjYbY8fP9ZiJET6QSqVwtjYGLdv39Z1KEQax2Sb3pqsrCzMmTMHzZs3R9WqVeHn56d2Ie369ddfsXv37kK37969G1u3btVBRO+uLl26IC8vr9DtDx48QNu2bbUf0Dvu7Nmz+PTTTxEYGIjo6GgAwLZt23Du3DkdR/buMDExgZeXF2Qyma5DIdI4tpHQWzNy5EicPn0agwYNgpubGwQCga5DeqctXboUP/zwQ6HbnZ2d8dlnn2HIkCE6iOrdZG1tjR49euDvv/+GsbHyY/fevXt4//330bdvXx1H927Zu3cvBg0ahIEDB+L69euqL0Hp6elYsmQJ/vnnHx1H+O6YNWsWvv76a2zbtg0ODg66DsdwKKCDNhLtHq6i4Rkk6a2xs7PDoUOH0KJFC12HQgDMzc1x//59+Pj4qN3+7Nkz1KxZE//f3r2H1ZTvfwB/79JVWxfpRhdKroXGiDlDGc2QW+RgiHKS8zsal0Ouz0yoJrch5JwzOa7NGaYQxm0aSUWDgyMxLkWkjHIPFUp7/f7w2D9b8uPM3nvVXu/X8+znaa+17PXGXD7ruz/f7/fJkyfiBJOgJ0+ewM/PDy1atEBSUhLOnz+PPn36ICgoCHFxcWLHk5QuXbpg2rRpCA4OhlwuR25uLlq1aoWcnBz4+/ujtLRU7IiS0aVLF1y5cgXV1dVwdnZG48aNVc6fPn1apGQNk3IHSfv/QSM9Q63e+7miCgdL1nAHyTpwZJvUxtLSkqMT9YiNjQ3Onj1bq9jOzc1F06ZNxQklUSYmJti3bx98fX0xYsQIHD58GMHBwfjmm2/EjiY5eXl56NWrV63j5ubmKCsr034gCRsyZIjYEXQTJ0jWOyy2SW1iYmIwb948JCYmwtTUVOw4kjdq1ChMmTIFcrlcWVxkZWVh6tSp+Pzzz0VOp/tenxSpp6eH5ORkfPrppxg2bBgiIyOV13AkSHvs7Oxw5cqVWg+h2dnZnFuiZfPnzxc7ApFWsNgmtVm+fDkKCgpga2sLFxcXGBgYqJznV4LaFRMTg8LCQvTp00fZJ6xQKBAcHIyFCxeKnE73WVhYvHHegiAISEhIwJo1ayAIAmQyGSeJadGECRMwdepUbNiwATKZDDdv3sSxY8cwY8YMREZGih1PcsrKyrB9+3YUFBRg5syZsLKywunTp2Fra4vmzZuLHa9hUigAKES4J9WFxTapDb8SrF8MDQ2RnJyMmJgY5ObmwsTEBB4eHnB2dhY7miRkZGSIHYHeYM6cOVAoFOjTpw8qKyvRq1cvGBkZYcaMGZg8ebLY8STl7Nmz8PPzg7m5OQoLCzFhwgRYWVlhx44dKCoqwnfffSd2RCK14ARJIiKSnKqqKly5cgXl5eVo3749zMzMxI4kOX5+fvDy8sLSpUtVJqsePXoUo0ePRmFhodgRGxTlBEmbMHEmSN5exwmSdeDINpEOmT59OmJiYtC4cWNMnz79rddyFQzt2bhxI8zMzDB8+HCV49u2bUNlZSWXYdSi0NBQrFq1CnK5HO3bt1cer6iowOTJk7FhwwYR00nLyZMn37g8afPmzbkqzO/BCZL1Dje1IbWpqanBsmXL0K1bN9jZ2cHKykrlRZqXk5OD6upq5c91vc6cOSNuUIlZtGgRrK2tax23sbFh/7yWJSYmvnHZyydPnrBtQcuMjIzeuLtqfn4+mjVrJkIiIs3gyDapTVRUFNatW4eIiAh89dVX+PLLL1FYWIhdu3Zh3rx5YseThFf7hNkzXH8UFRWhZcuWtY47OzujqKhIhETS8+jRIwiCAEEQ8PjxYxgbGyvP1dTUYP/+/bCxsRExofQMHjwY0dHR2Lp1KwBAJpOhqKgIs2fPxrBhw0RO14BxZLve4cg2qc3mzZuxdu1aREREoFGjRhg1ahTWrVuHefPm4fjx42LHIxLNyzXPX8c1z7XHwsICVlZWkMlkcHd3h6WlpfJlbW2N0NBQfPHFF2LHlJTly5ejvLwcNjY2ePLkCXx8fODm5ga5XI7Y2Fix4xGpDUe2SW1KS0vh4eEB4MX21A8fPgQADBw4kEtqiaCiogKLFy9Geno6bt++DcVrSzNdvXpVpGTSwzXPxZeRkQFBEPDJJ58gJSVFpbXN0NAQzs7OcHBwEDGh9JibmyMtLQ3Z2dk4e/YsysvL4eXlBT8/P7GjEakVi21SmxYtWqCkpAROTk5wdXXFgQMH4OXlhZMnT8LIyEjseJITFhaGrKwsjB07Fvb29m9c85m0g2uei8/HxwcAcO3aNTg5OfHfh3rk448/xscffyx2DN2hEABoua1DwTaSt2GxTWozdOhQpKenw9vbG5MnT8aYMWOwfv16FBUVYdq0aWLHk5yffvoJ+/btwx/+8Aexo0ge1zwX19mzZ9GxY0fo6enh4cOHOHfuXJ3Xenp6ajEZpaen1/ntG1eGIV3BYpvUZvHixcqfR44cCScnJxw7dgytW7fGoEGDREwmTZaWllwFpp5xd3eHu7u72DEkp3PnzigtLYWNjQ06d+4MmUyGN20xwd08tSsqKgrR0dHo2rUrv31TI0FQQBC0u6Ojtu/X0HBTGyId9f333+PHH39EYmIiTE1NxY4jeTdu3MDu3btRVFSEqqoqlXNc81yzrl+/rmwduX79+luv5bcN2mNvb4+lS5di7NixYkfRCS83teljGSLKpjbpDxK5qU0dOLJNv8vu3bvh7+8PAwMD7N69+63XDh48WEupCHgx07+goAC2trZwcXGBgYGByvnTp0+LlEx60tPTMXjwYLRq1QqXLl1Cx44dUVhYCEEQ4OXlJXY8nfdqAc1iuv6oqqrCRx99JHYM3SMI2u+h5rjtW7HYpt9lyJAhyq9nhwwZUud1/HpW+97290HaNXfuXMyYMQNRUVGQy+VISUmBjY0NgoKC0K9fP7HjSc7ly5eRkZHxxj5h7gmgPWFhYdiyZQtXqyKdxzYSIiINk8vlOHPmDFxdXWFpaYns7Gx06NABubm5CAgIQGFhodgRJWPt2rWYOHEirK2tYWdnp9InLJPJ+I2Phk2fPl35s0KhQGJiIjw9PeHp6Vnr2ze2V70fZRuJRTAaybTcRiJUIb3sO7aR1IEj26RRZWVlsLCwEDsGkagaN26s7NO2t7dHQUEBOnToAAC4e/eumNEk5+uvv0ZsbCxmz54tdhRJysnJUXnfuXNnAMCvv/4qQhodJYiw9B/Hbd+KxTapzZIlS+Di4oKRI0cCAIYPH46UlBTY29tj//796NSpk8gJpaWmpgYrVqzA1q1b3zgp7/79+yIlk57u3bsjOzsb7dq1Q//+/REREYFz585hx44d6N69u9jxJOXBgwcYPny42DEkKyMjQ+wIRFrH7dpJbRISEuDo6AgASEtLw8GDB5Gamgp/f3/MnDlT5HTSExUVhbi4OIwcORIPHz7E9OnTERgYCD09PSxYsEDseJISFxcHb29vAC/+Xvr06YPk5GS4uLhg/fr1IqeTluHDh+PAgQNixyAAoaGhePz4ca3jFRUVCA0NFSGRjlAoxHlRndizTWpjYmKC/Px8ODo6YurUqXj69CnWrFmD/Px8eHt748GDB2JHlBRXV1fEx8djwIABKj3D8fHxOH78OLZs2SJ2RCKtiI+PV/5cUVGBuLg4DBgwAB4eHrX6hKdMmaLteJKlr6+PkpIS2NjYqBy/e/cu7Ozs8Pz5c5GSNUzKnm15kDg92483s2e7DmwjIbWxtLREcXExHB0dkZqaiq+//hoAIAgCVyIRQWlpKTw8PAAAZmZmePjwIQBg4MCBnP1PkrJixQqV92ZmZsjKykJWVpbKcZlMxmJbCx49egRBECAIAh4/fgxjY2PluZqaGuzfv79WAU7UkLHYJrUJDAzE6NGj0bp1a9y7dw/+/v4AXkyIcXNzEzmd9LRo0QIlJSVwcnKCq6srDhw4AC8vL5w8eRJGRkZix9N5lpaW77wjHvvnNevatWtiR6BXWFhYQCaTQSaTvXFHVZlMhqioKBGS6QhOkKx3WGyT2qxYsQIuLi4oLi7G0qVLYWZmBgAoKSlBeHi4yOmkZ+jQoUhPT4e3tzcmT56MMWPGYP369SgqKsK0adPEjqfzVq5cKXYEeoOrV6+iVatWYseQtIyMDAiCgE8++QQpKSmwsrJSnjM0NISzszMcHBxETEikXuzZJrW7cOHCG1e/4A6S4jp27BiOHTuG1q1bY9CgQWLHIRKFnp4eWrRoAR8fH/j6+sLHx4ffvInk+vXraNKkCTZs2ICLFy8CADp06IDQ0FCYm5uLnK7hedmz/Ynp56L0bB+qTGLPdh1YbJPaXL16FYGBgTh37hyAF73aAJRfpbNvm6SspqYGO3fuVBYV7du3R0BAABo14heM2vTbb78hMzNT2bN9+fJlODg4wMfHB71790ZYWJjYESXj1KlT6NevH4yNjdGtWzcAwMmTJ/HkyRNl2xu9Oxbb9ReLbVKbQYMGQV9fH+vWrUPLli1x4sQJ3Lt3DxEREVi2bBl69uwpdkTJycvLw+rVq5UFXrt27TB58mS0adNG5GTScv78eQwePBilpaXKP/v8/Hw0a9YMe/bsQceOHUVOKF2XL19GbGwsNm/eDIVCwUEBLerZsyfc3Nywdu1a5UPn8+fPERYWhqtXr+Lw4cMiJ2xYlMW2yUhxiu0nySy268Bim9TG2toahw4dgqenJ8zNzXHixAm0adMGhw4dQkRERK2dw0izUlJS8Pnnn6Nr167o0aMHAOD48eM4efIkkpKSMGzYMJETSkePHj3QrFkzJCYmwtLSEsCLzVXGjRuHO3fu4OjRoyInlI7KykpkZ2cjMzMTmZmZyMnJQdu2beHr6wtfX18EBASIHVEyTExMlH/+r7pw4QK6du2KyspKkZI1TCy26y9+f0lqU1NTA7lcDuBF4X3z5k20adMGzs7OyMvLEzmd9MyaNQtz585FdHS0yvH58+dj1qxZLLa16MyZMzh16pSy0AZerFYSGxuLDz/8UMRk0mNhYQFLS0sEBQVhzpw56Nmzp8rfC2lPkyZNUFRUVKvYLi4uVv6/hEgXcAdJUpuOHTsiNzcXAODt7Y2lS5fil19+QXR0NGf/i6CkpATBwcG1jo8ZMwYlJSUiJJIud3d33Lp1q9bx27dvc3KelvXv3x81NTVISkpCUlIStm3bhvz8fLFjSdLIkSMxfvx4JCcno7i4GMXFxUhKSkJYWBhGjRoldryGSyGI86I6cWSb1Oarr75CRUUFACA6OhoDBw5Ez5490bRpUyQnJ4ucTnp8fX1x5MiRWsVcdnY2++e14NGjR8qfFy1ahClTpmDBggXo3r07gBctPdHR0ViyZIlYESVp165dAICzZ88iKysLBw4cQGRkJBo1agRfX19s3rxZ3IASsmzZMshkMgQHByt3izQwMMDEiROxePFikdMRqQ97tkmj7t+//16be5D6JCQkYN68eRgxYoRKgbdt2zZERUWprGPLZRnVT09PT+Wf+9dX53n1PSflaZ8gCMjJyUFGRgYyMjLw888/QxAEbhEugsrKShQUFAAAXF1dYWpqKnKihknZs204HI1kBlq993OhGoeqtrFnuw4stol0lJ7eu3WJsdjTjNe3An8bHx8fDSahV8XFxSEzMxPZ2dl4/PgxOnXqhF69esHX15f929Sgsdiuv9hGQqSjFAqF2BEk7b8poMPDwxEdHQ1ra2sNJCIA+OGHH+Dj44M///nP6NmzJzdPISKN4wRJIgkpKysTOwK9xffff6/S603qd/LkSSxbtgwDBw58a6EdHh6Ou3fvajEZkXoICkGUF9WNxTaRjlqyZInKxNThw4fDysoKzZs3V64aQ/ULu/rqDz74EJG6sNgm0lEJCQlwdHQEAKSlpeHgwYNITU2Fv78/Zs6cKXI6ovqNDz7UYAkKcV5UJ/ZsE+mo0tJSZbG9d+9ejBgxAp999hlcXFzg7e0tcjoiIiJp4Mg2kY6ytLREcXExACA1NRV+fn4AXozYcfURIiLdxJ7t+ocj20Q6KjAwEKNHj0br1q1x7949+Pv7AwBycnK4ayEREZGWcGSbSEetWLECkyZNQvv27ZGWlgYzMzMAL7ZxDw8PFzkdvcmYMWO4Ri0RScLf//53uLi4wNjYGN7e3jhx4oTYkTSGm9oQEWlBWVkZTpw4gdu3b9daAz04OFikVFSXiRMnIiYmhmueU4PxclMbXwSIsqlNJn58501tkpOTERwcjISEBHh7e2PlypXYtm0b8vLyYGNjo4XE2sVim0iH7N69G/7+/jAwMMDu3bvfei23aNeePXv2ICgoCOXl5WjSpInKNu4ymQz3798XMZ308MGHdNHLYvtj9EcjaLnYRjWysR/FxcUqxbaRkRGMjIxqXe/t7Y0PP/wQf/vb3wC82ITN0dERkydPxpw5c7SWW1tYbBPpED09PZSWlsLGxuat27Vzi3btcnd3R//+/bFw4UKYmpqKHUfS+OBDuurp06do2bIlSktLRbm/mZkZysvLVY7Nnz8fCxYsUDlWVVUFU1NTbN++HUOGDFEeDwkJQVlZGX788UctpNUuTpAk0iGvjtJxu/b647fffsOUKVNYaNcDERERCA0N5YMP6RxjY2Ncu3YNVVVVotxfEASVh1cAbxzVvnv3LmpqamBra6ty3NbWFpcuXdJoRrGw2CbSYenp6UhPT6/1dblMJsP69etFTCYtffv2xalTp9CqVSuxo0geH3xIlxkbG8PY2FjsGPQaFttEOioqKgrR0dHo2rUr7O3ta404kGa92jM/YMAAzJw5ExcuXICHhwcMDFT7Kdk/rz188CESl7W1NfT19XHr1i2V47du3YKdnZ1IqTSLPdtEOsre3h5Lly7F2LFjxY4iSW/rmX8V++c179UHnzt37iA6Ohp/+tOf+OBDJBJvb29069YNq1evBvCi7dHJyQmTJk3iBEkiajiaNm2KEydOwNXVVewoRKLigw9R/ZKcnIyQkBCsWbMG3bp1w8qVK7F161ZcunSpVi+3LmAbCZGOCgsLw5YtWxAZGSl2FAL758XEycJE9cvIkSNx584dzJs3D6WlpejcuTNSU1N1stAGOLJNpFOmT5+u/FmhUCAxMRGenp7w9PSs9XV5XFyctuNJ1v/XP79z506RkkkTH3yISJs4sk2kQ3JyclTed+7cGQDw66+/qhznZEntSkhIwKZNm9g/Xw9w4jARaRtHtomINIz98/UHJw4Tkba926wRIiL6r73snyfxVVVV4aOPPhI7BhFJCNtIiIg07OnTp/jnP/+JgwcPsn9eZJw4TETaxmKbiEjDzp49y/75eoIPPkSkbezZJiIiyejdu3ed52QyGQ4dOqTFNEQkBSy2iYiIiIg0hBMkiYiIiIg0hMU2EREREZGGsNgmIiIiItIQFttERERERBrCYpuI6DXjxo3DkCFDlO99fX3x17/+Ves5MjMzIZPJUFZWVuc1MpkMu3bteufPXLBggXIZwv9WYWEhZDIZzpw587s+h4hIClhsE1GDMG7cOMhkMshkMhgaGsLNzQ3R0dF4/vy5xu+9Y8cOxMTEvNO171IgExGRdHBTGyJqMPr164eNGzfi2bNn2L9/P7744gsYGBhg7ty5ta6tqqqCoaGhWu5rZWWlls8hIiLp4cg2ETUYRkZGsLOzg7OzMyZOnAg/Pz/s3r0bwP+1fsTGxsLBwQFt2rQBABQXF2PEiBGwsLCAlZUVAgICUFhYqPzMmpoaTJ8+HRYWFmjatClmzZqF17cfeL2N5NmzZ5g9ezYcHR1hZGQENzc3rF+/HoWFhcpNUywtLSGTyTBu3DgAgEKhwKJFi9CyZUuYmJigU6dO2L59u8p99u/fD3d3d5iYmKB3794qOd/V7Nmz4e7uDlNTU7Rq1QqRkZGorq6udd2aNWvg6OgIU1NTjBgxAg8fPlQ5v27dOrRr1w7GxsZo27Yt/vGPf7x3FiIiYrFNRA2YiYkJqqqqlO/T09ORl5eHtLQ07N27F9XV1ejbty/kcjmOHDmCX375BWZmZujXr5/y1y1fvhybNm3Chg0bkJ2djfv372Pnzp1vvW9wcDB++OEHxMfH4+LFi1izZg3MzMzg6OiIlJQUAEBeXh5KSkqwatUqAMCiRYvw3XffISEhAefPn8e0adMwZswYZGVlAXjxUBAYGIhBgwbhzJkzCAsLw5w5c977z0Qul2PTpk24cOECVq1ahbVr12LFihUq11y5cgVbt27Fnj17kJqaipycHISHhyvPb968GfPmzUNsbCwuXryIhQsXIjIyEomJie+dh4hI8gQiogYgJCRECAgIEARBEBQKhZCWliYYGRkJM2bMUJ63tbUVnj17pvw1//rXv4Q2bdoICoVCeezZs2eCiYmJ8PPPPwuCIAj29vbC0qVLleerq6uFFi1aKO8lCILg4+MjTJ06VRAEQcjLyxMACGlpaW/MmZGRIQAQHjx4oDz29OlTwdTUVDh69KjKtePHjxdGjRolCIIgzJ07V2jfvr3K+dmzZ9f6rNcBEHbu3Fnn+W+++Ub44IMPlO/nz58v6OvrCzdu3FAe++mnnwQ9PT2hpKREEARBcHV1FbZs2aLyOTExMUKPHj0EQRCEa9euCQCEnJycOu9LREQvsGebiBqMvXv3wszMDNXV1VAoFBg9ejQWLFigPO/h4aHSp52bm4srV65ALperfM7Tp09RUFCAhw8foqSkBN7e3spzjRo1QteuXWu1krx05swZ6Ovrw8fH551zX7lyBZWVlfj0009VjldVVaFLly4AgIsXL6rkAIAePXq88z1eSk5ORnx8PAoKClBeXo7nz5+jSZMmKtc4OTmhefPmKvdRKBTIy8uDXC5HQUEBxo8fjwkTJiivef78OczNzd87DxGR1LHYJqIGo3fv3vj2229haGgIBwcHNGqk+p+wxo0bq7wvLy/HBx98gM2bN9f6rGbNmv1XGUxMTN7715SXlwMA9u3bp1LkAi/60NXl2LFjCAoKQlRUFPr27Qtzc3MkJSVh+fLl75117dq1tYp/fX19tWUlIpIKFttE1GA0btwYbm5u73y9l5cXkpOTYWNjU2t09yV7e3v8+9//Rq9evQC8GMH9z3/+Ay8vrzde7+HhAYVCgaysLPj5+dU6/3JkvaamRnmsffv2MDIyQlFRUZ0j4u3atVNO9nzp+PHj//9v8hVHjx6Fs7MzvvzyS+Wx69ev17quqKgIN2/ehIODg/I+enp6aNOmDWxtbeHg4ICrV68iKCjove5PRES1cYIkEemsoKAgWFtbIyAgAEeOHMG1a9eQmZmJKVOm4MaNGwCAqVOnYvHixdi1axcuXbqE8PDwt66R7eLigpCQEISGhmLXrl3Kz9y6dSsAwNnZGTKZDHv37sWdO3dQXl4OuVyOGTNmYNq0aUhMTERBQQFOnz6N1atXKycd/uUvf8Hly5cxc+ZM5OXlYcuWLdi0adN7/X5bt26NoqIiJCUloaCgAPHx8W+c7GlsbIyQkBDk5ubiyJEjmDJlCkaMGAE7OzsAQFRUFBYtWoT4+Hjk5+fj3Llz2LhxI+Li4t4rDxERsdgmIh1mamqKw4cPw8nJCYGBgWjXrh3Gjx+Pp0+fKke6IyIiMHbsWISEhKBHjx6Qy+UYOnToWz/322+/xR//+EeEh4ejbdu2mDBhAioqKgAAzZs3R1RUFObMmQNbW1tMmjQJABATE4PIyEgsWrQI7dq1Q79+/bBv3z60bNkSwIs+6pSUFOzatQudOnVCQkICFi5c+F6/38GDB2PatGmYNGkSOnfujKNHjyIyMrLWdW5ubggMDET//v3x2WefwdPTU2Vpv7CwMKxbtw4bN26Eh4cHfHx8sGnTJmVWIiJ6dzKhrllARERERET0u3Bkm4iIiIhIQ1hsExERERFpCIttIiIiIiINYbFNRERERKQhLLaJiIiIiDSExTYRERERkYaw2CYiIiIi0hAW20REREREGsJim4iIiIhIQ1hsExERERFpCIttIiIiIiIN+V8vBR1zj0TaqgAAAABJRU5ErkJggg==",
      "text/plain": [
       "<Figure size 800x800 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "model_full = joblib.load('models/gb_model_fullname.joblib')\n",
    "preds = model_full.predict(X_test)\n",
    "predicted = [id_to_race[p] for p in preds]\n",
    "actual = [id_to_race[p] for p in y_test.values]\n",
    "\n",
    "print(classification_report(actual, predicted))\n",
    "\n",
    "cm = confusion_matrix(actual, predicted, labels=list(race_to_id.keys()))\n",
    "cmp = ConfusionMatrixDisplay(cm, display_labels=list(race_to_id.keys()))\n",
    "fig, ax = plt.subplots(figsize=(8,8))\n",
    "cmp.plot(ax=ax, xticks_rotation='vertical')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 104,
   "id": "80bf6183-fda0-4b73-8952-82c03d044580",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Accuracy for class asian: 0.04\n",
      "Accuracy for class hispanic: 0.40\n",
      "Accuracy for class nh_black: 0.01\n",
      "Accuracy for class nh_white: 0.98\n",
      "Accuracy for class other: 0.00\n"
     ]
    }
   ],
   "source": [
    "for class_id in id_to_race:\n",
    "    correct = np.sum((y_test == class_id) & (preds == class_id))\n",
    "    total = np.sum(y_test == class_id)\n",
    "    print(f'Accuracy for class {id_to_race[class_id]}: {correct / total:.2f}')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "92c1af21-c6b3-4816-bc80-3f55be6994f7",
   "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": "3.10.6"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}