evaluation_stratify_HR.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 96,
   "metadata": {},
   "outputs": [],
   "source": [
    "import numpy as np\n",
    "import matplotlib.pyplot as plt\n",
    "from lifelines import KaplanMeierFitter\n",
    "import matplotlib.patches as mpatches\n",
    "from lifelines.utils import concordance_index\n",
    "import scipy\n",
    "from scipy import stats\n",
    "from scipy.stats import linregress\n",
    "from scipy.special import gamma\n",
    "from scipy.cluster.hierarchy import linkage, cophenet, dendrogram, fcluster\n",
    "from scipy.spatial.distance import pdist\n",
    "from sklearn.manifold import TSNE\n",
    "from sklearn.cluster import AgglomerativeClustering\n",
    "\n",
    "fontsize = 14\n",
    "SMALL_SIZE = 8\n",
    "MEDIUM_SIZE = 10\n",
    "BIGGER_SIZE = 12\n",
    "legend_size =14\n",
    "plt.rc('font', size=MEDIUM_SIZE)  # controls default text sizes\n",
    "plt.rc('axes', titlesize=MEDIUM_SIZE)  # fontsize of the axes title\n",
    "plt.rc('axes', labelsize=MEDIUM_SIZE)  # fontsize of the x and y labels\n",
    "plt.rc('xtick', labelsize=MEDIUM_SIZE)  # fontsize of the tick labels\n",
    "plt.rc('ytick', labelsize=MEDIUM_SIZE)  # fontsize of the tick labels\n",
    "plt.rc('legend', fontsize=MEDIUM_SIZE)  # legend fontsize\n",
    "plt.rc('figure', titlesize=BIGGER_SIZE)  # fontsize of the figure title\n",
    "plt.rc('xtick', labelsize=20)\n",
    "plt.rc('ytick', labelsize=20)\n",
    "\n",
    "font = {'family': 'normal',\n",
    "        'weight': 'bold',\n",
    "        'size': 24}\n",
    "\n",
    "plt.rc('font', **font)\n",
    "params = {'legend.fontsize': 'x-large',\n",
    "          # 'figure.figsize': (15, 5),\n",
    "          'axes.labelsize': 'x-large',\n",
    "          'axes.titlesize': 'x-large',\n",
    "          'xtick.labelsize': 'x-large',\n",
    "          'ytick.labelsize': 'x-large'}\n",
    "\n",
    "plt.rcParams['pdf.fonttype'] = 42\n",
    "plt.rcParams['ps.fonttype'] = 42\n",
    "\n",
    "plt.rcParams.update(params)\n",
    "import seaborn as sns\n",
    "import pandas\n",
    "sns.set_style('white')\n",
    "sns.set_context('paper')\n",
    "sns.set()\n",
    "seed=31415\n",
    "np.random.seed(seed)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 97,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "analysis/framingham/alpha_01/CSA-INFO/\n",
      "pred_t0_f:  (455, 200)\n",
      "pred_t0_cf:  (60, 200)\n",
      "pred_t1_f:  (60, 200)\n",
      "pred_t1_cf:  (455, 200)\n",
      "pred_c0_f:  (455, 200)\n",
      "pred_c0_cf:  (60, 200)\n",
      "pred_c1_f:  (60, 200)\n",
      "pred_c1_cf:  (455, 200)\n"
     ]
    }
   ],
   "source": [
    "#\n",
    "model = 'CSA-INFO'\n",
    "is_non_param = True\n",
    "is_stochastic = True\n",
    "\n",
    "if 'SR' in model:\n",
    "    is_stochastic = False\n",
    "    \n",
    "if 'AFT' in model:\n",
    "    is_non_param = False      \n",
    "    \n",
    "data = 'framingham'\n",
    "\n",
    "if data =='framingham':\n",
    "    alpha = \"01\"\n",
    "else:\n",
    "    alpha = '0'\n",
    "\n",
    "if alpha:\n",
    "    main_path = 'analysis/{}/alpha_{}/'.format(data, alpha)\n",
    "    path = 'analysis/{}/alpha_{}/{}/'.format(data, alpha, model)\n",
    "else:\n",
    "    path = 'analysis/{}/{}/'.format(data, model)\n",
    "    main_path = 'analysis/{}/'.format(data)\n",
    "\n",
    "\n",
    "#data = 'sythentic'\n",
    "#path = 'analysis/{}/alpha_{}/{}/'.format(fld,alpha,  model)\n",
    "\n",
    "print(path)\n",
    "has_cf = True\n",
    "if 'simulated' in data:\n",
    "    time = 'days'\n",
    "    is_simulated = True\n",
    "else:\n",
    "    is_simulated = False\n",
    "    time = 'days'\n",
    "if is_non_param:\n",
    "    pred_t0_f = np.load(path + 'Test_pred_t0_F.npy')\n",
    "    pred_t0_cf = np.load(path +  'Test_pred_t0_CF.npy')\n",
    "\n",
    "    pred_t1_f = np.load(path +  'Test_pred_t1_F.npy')\n",
    "    pred_t1_cf = np.load(path +  'Test_pred_t1_CF.npy')\n",
    "    print(\"pred_t0_f: \", pred_t0_f.shape)\n",
    "    print(\"pred_t0_cf: \", pred_t0_cf.shape)\n",
    "\n",
    "    print(\"pred_t1_f: \", pred_t1_f.shape)\n",
    "    print(\"pred_t1_cf: \", pred_t1_cf.shape)\n",
    "    \n",
    "    pred_c0_f = np.load(path + 'Test_pred_c0_F.npy')\n",
    "    pred_c0_cf = np.load(path +  'Test_pred_c0_CF.npy')\n",
    "\n",
    "    pred_c1_f = np.load(path +  'Test_pred_c1_F.npy')\n",
    "    pred_c1_cf = np.load(path +  'Test_pred_c1_CF.npy')\n",
    "    \n",
    "    print(\"pred_c0_f: \", pred_c0_f.shape)\n",
    "    print(\"pred_c0_cf: \", pred_c0_cf.shape)\n",
    "\n",
    "    print(\"pred_c1_f: \", pred_c1_f.shape)\n",
    "    print(\"pred_c1_cf: \", pred_c1_cf.shape)\n",
    "    \n",
    "else:\n",
    "    pred_t0_f = pandas.read_csv(path + 'Test_pred_t0_F.csv')\n",
    "    pred_t0_cf = pandas.read_csv(path +  'Test_pred_t0_CF.csv')\n",
    "\n",
    "    pred_t1_f = pandas.read_csv(path +  'Test_pred_t1_F.csv')\n",
    "    pred_t1_cf = pandas.read_csv(path +  'Test_pred_t1_CF.csv')\n",
    "    \n",
    "    print(\"pred_t0_f: \", pred_t0_f.shape,pred_t0_f.head())\n",
    "    print(\"pred_t0_cf: \", pred_t0_cf.shape, pred_t0_cf.head())\n",
    "\n",
    "    print(\"pred_t1_f: \", pred_t1_f.shape, pred_t1_f.head())\n",
    "    print(\"pred_t1_cf: \", pred_t1_cf.shape, pred_t1_cf.head())\n",
    "\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 98,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "covariates:  32\n",
      "sub_covariates 16\n"
     ]
    }
   ],
   "source": [
    "if data == 'framingham':\n",
    "    covariates = ['age6', 'age61', 'age62', 'ascvd_hx6', 'bmi6', 'bmi61', 'bmi62', 'bpmeds6',\n",
    "                  'chol5', 'chol51', 'chol52', 'dbp6', 'dbp61', 'dbp62', 'diab6', 'female',\n",
    "                  'gluc5', 'gluc51', 'gluc52', 'hdl5', 'hdl51', 'hdl52',\n",
    "                  'pad_hx6', 'sbp6', 'sbp61', 'sbp62', 'smoke6', 'stk_hx6', 'mi_hx6',\n",
    "                  'trigly5', 'trigly51', 'trigly52']\n",
    "\n",
    "    print(\"covariates: \", len(covariates))\n",
    "\n",
    "    sub_covariates = ['age6', 'ascvd_hx6', 'bmi6', 'bpmeds6','chol5', 'dbp6', 'diab6', 'female', 'gluc5',  'hdl5', \n",
    "                  'pad_hx6', 'sbp6', 'smoke6', 'stk_hx6', 'mi_hx6',  'trigly5']\n",
    "\n",
    "    print(\"sub_covariates\", len(sub_covariates))\n",
    "else:\n",
    "    covariates =  ['age', 'wtkg', 'hemo', 'homo', 'drugs', 'karnof', 'oprior','z30', 'zprior',\n",
    "                   'preanti', 'race', 'gender', 'str2', 'strat','symptom', 'treat', 'offtrt', \n",
    "                   'cd40', 'cd420', 'cd496', 'r', 'cd80','cd820']\n",
    "    print(\"covariates: \", len(covariates))\n",
    "    \n",
    "    sub_covariates = covariates"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 99,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "a:  (515,)\n",
      "y_cf:  (515,)\n",
      "e_cf:  (515,)\n",
      "y_f:  (515,)\n",
      "e_f:  (515,)\n",
      "a=1 60.0\n",
      "y_f[a==1.0].shape (60,)\n",
      "x:  (515, 32)\n"
     ]
    }
   ],
   "source": [
    "if is_simulated:\n",
    "    dir_path = 'data/{}/'.format(data)\n",
    "    path_factual =  dir_path + 'event_pairs.csv' \n",
    "    path_covariates = dir_path +  'covariates.npy'\n",
    "    path_treatment = dir_path + 'treatment.npy'\n",
    "    print(\"path:{}\".format(path_factual))\n",
    "    data_frame = pandas.read_csv(path_factual)\n",
    "    print(\"head of data:{}, data shape:{}\".format(data_frame.head(), data_frame.shape))\n",
    "\n",
    "    idx = np.load(dir_path + 'test_idx.npy')\n",
    "\n",
    "    # Preprocess\n",
    "    a = np.load(path_treatment)[idx]\n",
    "\n",
    "    ## Factual\n",
    "    y_f = data_frame[['y_f']]\n",
    "    e_f = data_frame[['e_f']]\n",
    "    t_f = data_frame[['t_f']]\n",
    "\n",
    "    y_f = np.array(y_f).reshape(len(y_f))[idx]\n",
    "    e_f = np.array(e_f).reshape(len(e_f))[idx]\n",
    "    t_f = np.array(t_f).reshape(len(t_f))[idx]\n",
    "\n",
    "    ## Counter Factual\n",
    "    y_cf = data_frame[['y_cf']]\n",
    "    e_cf = data_frame[['e_cf']]\n",
    "    t_cf = data_frame[['t_cf']]\n",
    "\n",
    "    y_cf = np.array(y_cf).reshape(len(y_cf))[idx]\n",
    "    e_cf = np.array(e_cf).reshape(len(e_cf))[idx]\n",
    "    t_cf = np.array(t_cf).reshape(len(t_cf))[idx]\n",
    "\n",
    "    print(\"a: \", a.shape)\n",
    "\n",
    "    print(\"y_cf: \", y_cf.shape)\n",
    "    print(\"e_cf: \", e_cf.shape)\n",
    "    print(\"t_cf: \", t_cf.shape)\n",
    "\n",
    "\n",
    "    print(\"y_f: \", y_f.shape)\n",
    "    print(\"e_f: \", e_f.shape)\n",
    "    print(\"t_f: \", t_f.shape)\n",
    "\n",
    "else:\n",
    "    path_factual = 'data/{}/{}_test_idx.csv'.format(data, data)\n",
    "    data_frame = pandas.read_csv(path_factual)\n",
    "    ## Factual\n",
    "    y_f = data_frame[['time']]\n",
    "    e_f = data_frame[['event']]\n",
    "    a = data_frame[['treatment']]\n",
    "    x = data_frame[covariates]\n",
    "    \n",
    "    y_f = np.array(y_f).reshape(len(y_f))\n",
    "    e_f = np.array(e_f).reshape(len(e_f))\n",
    "    a = np.array(a).reshape(len(a))\n",
    "    \n",
    "    ## Counter Factual\n",
    "    y_cf = data_frame[['nn_cf_y']]\n",
    "    e_cf = data_frame[['nn_cf_e']]\n",
    " \n",
    "\n",
    "    y_cf = np.array(y_cf).reshape(len(y_cf))\n",
    "    e_cf = np.array(e_cf).reshape(len(e_cf))\n",
    "    \n",
    "    \n",
    "    print(\"a: \", a.shape)\n",
    "    \n",
    "    print(\"y_cf: \", y_cf.shape)\n",
    "    print(\"e_cf: \", e_cf.shape)\n",
    "\n",
    "    print(\"y_f: \", y_f.shape)\n",
    "    print(\"e_f: \", e_f.shape)\n",
    "    print(\"a=1\", np.sum(a))\n",
    "    print(\"y_f[a==1.0].shape\", y_f[a==1].shape)\n",
    "    print(\"x: \", x.shape)\n",
    "  \n",
    "    \n",
    "\n",
    "    \n",
    "\n",
    "\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 100,
   "metadata": {},
   "outputs": [],
   "source": [
    "def hr_dist(hr, ms):\n",
    "    perc_50 = np.round(np.percentile(hr, 50),2)\n",
    "    perc_2_5 = np.round(np.percentile(hr, 2.5),2)\n",
    "    perc_97_5 = np.round(np.percentile(hr, 97.5),2)\n",
    "    print(ms, \"{} ({}, {})\".format(perc_50, perc_2_5, perc_97_5))\n",
    "    "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 101,
   "metadata": {},
   "outputs": [],
   "source": [
    "def compute_grad(surv, time):\n",
    "    slope, intercept, r_value, p_value, std_err = linregress(x=time, y=surv) # \n",
    "    #print(\"slope: \", slope, \"std_err: \", std_err)\n",
    "    \n",
    "    #gradient = [(y2-y0)/(x2-x0) for x2, x0, y2, y0 in zip(time[2:], time, surv[2:], surv)]\n",
    "    \n",
    "\n",
    "    return  slope\n",
    "    "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Causal Metrics"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 102,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "T_0_pred:  (515, 200)\n",
      "T_1_pred:  (515, 200)\n",
      "pred_t0_f:  (455, 200) pred_t1_cf:  (455, 200)\n",
      "pred_t1_f:  (60, 200) pred_t0_cf:  (60, 200)\n",
      "(455, 32)\n",
      "(60, 32)\n",
      "(515, 32)\n",
      "all_t_range:  (516,)\n"
     ]
    }
   ],
   "source": [
    "\n",
    "# T_0_pred #pred_t0_f, pred_t0_cf\n",
    "# T_1_pred # pred_t1_f, pred_t1_cf\n",
    "\n",
    "T_0_pred = np.concatenate((pred_t0_f, pred_t0_cf), axis=0)\n",
    "T_1_pred = np.concatenate((pred_t1_cf,pred_t1_f), axis=0)\n",
    "\n",
    "print(\"T_0_pred: \", T_0_pred.shape)\n",
    "print(\"T_1_pred: \", T_1_pred.shape)\n",
    "print(\"pred_t0_f: \", pred_t0_f.shape, \"pred_t1_cf: \", pred_t1_cf.shape)\n",
    "print(\"pred_t1_f: \", pred_t1_f.shape, \"pred_t0_cf: \", pred_t0_cf.shape )\n",
    "    \n",
    "\n",
    "x_0 = x.iloc[a==0]\n",
    "x_1 = x.iloc[a==1]\n",
    "\n",
    "print(x_0.shape)\n",
    "print(x_1.shape)\n",
    "\n",
    "x0_x1_conc = pandas.concat([x_0, x_1], axis=0)\n",
    "\n",
    "print(x0_x1_conc.shape)\n",
    "\n",
    "zero = np.array([0])\n",
    "all_t_range = np.concatenate((zero, np.sort(y_f)))\n",
    "print(\"all_t_range: \", all_t_range.shape)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 103,
   "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>age6</th>\n",
       "      <th>age61</th>\n",
       "      <th>age62</th>\n",
       "      <th>ascvd_hx6</th>\n",
       "      <th>bmi6</th>\n",
       "      <th>bmi61</th>\n",
       "      <th>bmi62</th>\n",
       "      <th>bpmeds6</th>\n",
       "      <th>chol5</th>\n",
       "      <th>chol51</th>\n",
       "      <th>...</th>\n",
       "      <th>pad_hx6</th>\n",
       "      <th>sbp6</th>\n",
       "      <th>sbp61</th>\n",
       "      <th>sbp62</th>\n",
       "      <th>smoke6</th>\n",
       "      <th>stk_hx6</th>\n",
       "      <th>mi_hx6</th>\n",
       "      <th>trigly5</th>\n",
       "      <th>trigly51</th>\n",
       "      <th>trigly52</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>count</th>\n",
       "      <td>455.000000</td>\n",
       "      <td>455.000000</td>\n",
       "      <td>455.000000</td>\n",
       "      <td>455.000000</td>\n",
       "      <td>455.000000</td>\n",
       "      <td>455.000000</td>\n",
       "      <td>455.000000</td>\n",
       "      <td>455.000000</td>\n",
       "      <td>455.000000</td>\n",
       "      <td>455.000000</td>\n",
       "      <td>...</td>\n",
       "      <td>455.000000</td>\n",
       "      <td>455.000000</td>\n",
       "      <td>455.000000</td>\n",
       "      <td>455.000000</td>\n",
       "      <td>455.000000</td>\n",
       "      <td>455.000000</td>\n",
       "      <td>455.000000</td>\n",
       "      <td>455.000000</td>\n",
       "      <td>455.000000</td>\n",
       "      <td>455.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>mean</th>\n",
       "      <td>57.650549</td>\n",
       "      <td>7.005656</td>\n",
       "      <td>1.094300</td>\n",
       "      <td>0.083516</td>\n",
       "      <td>27.836624</td>\n",
       "      <td>2.551863</td>\n",
       "      <td>0.507544</td>\n",
       "      <td>0.257143</td>\n",
       "      <td>202.338462</td>\n",
       "      <td>20.669323</td>\n",
       "      <td>...</td>\n",
       "      <td>0.013187</td>\n",
       "      <td>127.369231</td>\n",
       "      <td>10.467159</td>\n",
       "      <td>1.931534</td>\n",
       "      <td>0.193407</td>\n",
       "      <td>0.024176</td>\n",
       "      <td>0.032967</td>\n",
       "      <td>144.261538</td>\n",
       "      <td>30.468226</td>\n",
       "      <td>11.491322</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>std</th>\n",
       "      <td>9.594583</td>\n",
       "      <td>8.983994</td>\n",
       "      <td>1.965585</td>\n",
       "      <td>0.276966</td>\n",
       "      <td>5.097342</td>\n",
       "      <td>4.287990</td>\n",
       "      <td>1.174040</td>\n",
       "      <td>0.437540</td>\n",
       "      <td>33.118659</td>\n",
       "      <td>29.014455</td>\n",
       "      <td>...</td>\n",
       "      <td>0.114200</td>\n",
       "      <td>18.309741</td>\n",
       "      <td>15.490517</td>\n",
       "      <td>3.913355</td>\n",
       "      <td>0.395404</td>\n",
       "      <td>0.153764</td>\n",
       "      <td>0.178747</td>\n",
       "      <td>96.985919</td>\n",
       "      <td>71.851921</td>\n",
       "      <td>31.109896</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>min</th>\n",
       "      <td>35.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>17.676632</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>105.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>...</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>88.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>33.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>25%</th>\n",
       "      <td>51.000000</td>\n",
       "      <td>0.532778</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>24.342611</td>\n",
       "      <td>0.130127</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>180.000000</td>\n",
       "      <td>1.784340</td>\n",
       "      <td>...</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>114.000000</td>\n",
       "      <td>0.480000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>83.000000</td>\n",
       "      <td>0.244709</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>50%</th>\n",
       "      <td>56.000000</td>\n",
       "      <td>2.286160</td>\n",
       "      <td>0.008325</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>27.235345</td>\n",
       "      <td>0.859652</td>\n",
       "      <td>0.016579</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>201.000000</td>\n",
       "      <td>9.211875</td>\n",
       "      <td>...</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>126.000000</td>\n",
       "      <td>3.840000</td>\n",
       "      <td>0.060000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>120.000000</td>\n",
       "      <td>3.481385</td>\n",
       "      <td>0.111383</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>75%</th>\n",
       "      <td>65.000000</td>\n",
       "      <td>11.056664</td>\n",
       "      <td>1.369880</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>30.549053</td>\n",
       "      <td>3.053826</td>\n",
       "      <td>0.419032</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>221.500000</td>\n",
       "      <td>26.452390</td>\n",
       "      <td>...</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>138.000000</td>\n",
       "      <td>12.921905</td>\n",
       "      <td>1.593333</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>164.500000</td>\n",
       "      <td>18.200893</td>\n",
       "      <td>4.027889</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>max</th>\n",
       "      <td>84.000000</td>\n",
       "      <td>46.451613</td>\n",
       "      <td>11.425598</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>49.154229</td>\n",
       "      <td>28.365909</td>\n",
       "      <td>8.197843</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>311.000000</td>\n",
       "      <td>168.866667</td>\n",
       "      <td>...</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>189.000000</td>\n",
       "      <td>90.133333</td>\n",
       "      <td>24.663333</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>813.000000</td>\n",
       "      <td>670.626866</td>\n",
       "      <td>298.601790</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>8 rows × 32 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "             age6       age61       age62   ascvd_hx6        bmi6       bmi61  \\\n",
       "count  455.000000  455.000000  455.000000  455.000000  455.000000  455.000000   \n",
       "mean    57.650549    7.005656    1.094300    0.083516   27.836624    2.551863   \n",
       "std      9.594583    8.983994    1.965585    0.276966    5.097342    4.287990   \n",
       "min     35.000000    0.000000    0.000000    0.000000   17.676632    0.000000   \n",
       "25%     51.000000    0.532778    0.000000    0.000000   24.342611    0.130127   \n",
       "50%     56.000000    2.286160    0.008325    0.000000   27.235345    0.859652   \n",
       "75%     65.000000   11.056664    1.369880    0.000000   30.549053    3.053826   \n",
       "max     84.000000   46.451613   11.425598    1.000000   49.154229   28.365909   \n",
       "\n",
       "            bmi62     bpmeds6       chol5      chol51  ...     pad_hx6  \\\n",
       "count  455.000000  455.000000  455.000000  455.000000  ...  455.000000   \n",
       "mean     0.507544    0.257143  202.338462   20.669323  ...    0.013187   \n",
       "std      1.174040    0.437540   33.118659   29.014455  ...    0.114200   \n",
       "min      0.000000    0.000000  105.000000    0.000000  ...    0.000000   \n",
       "25%      0.000000    0.000000  180.000000    1.784340  ...    0.000000   \n",
       "50%      0.016579    0.000000  201.000000    9.211875  ...    0.000000   \n",
       "75%      0.419032    1.000000  221.500000   26.452390  ...    0.000000   \n",
       "max      8.197843    1.000000  311.000000  168.866667  ...    1.000000   \n",
       "\n",
       "             sbp6       sbp61       sbp62      smoke6     stk_hx6      mi_hx6  \\\n",
       "count  455.000000  455.000000  455.000000  455.000000  455.000000  455.000000   \n",
       "mean   127.369231   10.467159    1.931534    0.193407    0.024176    0.032967   \n",
       "std     18.309741   15.490517    3.913355    0.395404    0.153764    0.178747   \n",
       "min     88.000000    0.000000    0.000000    0.000000    0.000000    0.000000   \n",
       "25%    114.000000    0.480000    0.000000    0.000000    0.000000    0.000000   \n",
       "50%    126.000000    3.840000    0.060000    0.000000    0.000000    0.000000   \n",
       "75%    138.000000   12.921905    1.593333    0.000000    0.000000    0.000000   \n",
       "max    189.000000   90.133333   24.663333    1.000000    1.000000    1.000000   \n",
       "\n",
       "          trigly5    trigly51    trigly52  \n",
       "count  455.000000  455.000000  455.000000  \n",
       "mean   144.261538   30.468226   11.491322  \n",
       "std     96.985919   71.851921   31.109896  \n",
       "min     33.000000    0.000000    0.000000  \n",
       "25%     83.000000    0.244709    0.000000  \n",
       "50%    120.000000    3.481385    0.111383  \n",
       "75%    164.500000   18.200893    4.027889  \n",
       "max    813.000000  670.626866  298.601790  \n",
       "\n",
       "[8 rows x 32 columns]"
      ]
     },
     "execution_count": 103,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "x_0.describe()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 104,
   "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>age6</th>\n",
       "      <th>age61</th>\n",
       "      <th>age62</th>\n",
       "      <th>ascvd_hx6</th>\n",
       "      <th>bmi6</th>\n",
       "      <th>bmi61</th>\n",
       "      <th>bmi62</th>\n",
       "      <th>bpmeds6</th>\n",
       "      <th>chol5</th>\n",
       "      <th>chol51</th>\n",
       "      <th>...</th>\n",
       "      <th>pad_hx6</th>\n",
       "      <th>sbp6</th>\n",
       "      <th>sbp61</th>\n",
       "      <th>sbp62</th>\n",
       "      <th>smoke6</th>\n",
       "      <th>stk_hx6</th>\n",
       "      <th>mi_hx6</th>\n",
       "      <th>trigly5</th>\n",
       "      <th>trigly51</th>\n",
       "      <th>trigly52</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>count</th>\n",
       "      <td>60.000000</td>\n",
       "      <td>60.000000</td>\n",
       "      <td>60.000000</td>\n",
       "      <td>60.000000</td>\n",
       "      <td>60.000000</td>\n",
       "      <td>60.000000</td>\n",
       "      <td>60.000000</td>\n",
       "      <td>60.000000</td>\n",
       "      <td>60.000000</td>\n",
       "      <td>60.000000</td>\n",
       "      <td>...</td>\n",
       "      <td>60.000000</td>\n",
       "      <td>60.000000</td>\n",
       "      <td>60.000000</td>\n",
       "      <td>60.000000</td>\n",
       "      <td>60.000000</td>\n",
       "      <td>60.000000</td>\n",
       "      <td>60.000000</td>\n",
       "      <td>60.000000</td>\n",
       "      <td>60.000000</td>\n",
       "      <td>60.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>mean</th>\n",
       "      <td>64.150000</td>\n",
       "      <td>12.930808</td>\n",
       "      <td>2.320905</td>\n",
       "      <td>0.316667</td>\n",
       "      <td>28.993009</td>\n",
       "      <td>2.976886</td>\n",
       "      <td>0.552253</td>\n",
       "      <td>0.633333</td>\n",
       "      <td>233.650000</td>\n",
       "      <td>56.843871</td>\n",
       "      <td>...</td>\n",
       "      <td>0.100000</td>\n",
       "      <td>131.266667</td>\n",
       "      <td>13.767726</td>\n",
       "      <td>2.836248</td>\n",
       "      <td>0.166667</td>\n",
       "      <td>0.066667</td>\n",
       "      <td>0.133333</td>\n",
       "      <td>200.016667</td>\n",
       "      <td>62.346172</td>\n",
       "      <td>24.433698</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>std</th>\n",
       "      <td>8.313406</td>\n",
       "      <td>10.740757</td>\n",
       "      <td>2.537664</td>\n",
       "      <td>0.469102</td>\n",
       "      <td>4.397932</td>\n",
       "      <td>3.337312</td>\n",
       "      <td>0.864667</td>\n",
       "      <td>0.485961</td>\n",
       "      <td>45.367361</td>\n",
       "      <td>54.002275</td>\n",
       "      <td>...</td>\n",
       "      <td>0.302532</td>\n",
       "      <td>20.984956</td>\n",
       "      <td>23.428731</td>\n",
       "      <td>6.461287</td>\n",
       "      <td>0.375823</td>\n",
       "      <td>0.251549</td>\n",
       "      <td>0.342803</td>\n",
       "      <td>149.939817</td>\n",
       "      <td>139.867447</td>\n",
       "      <td>62.606721</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>min</th>\n",
       "      <td>49.000000</td>\n",
       "      <td>0.224766</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>19.285225</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>125.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>...</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>102.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>67.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>25%</th>\n",
       "      <td>56.750000</td>\n",
       "      <td>2.713059</td>\n",
       "      <td>0.023153</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>26.688324</td>\n",
       "      <td>0.653533</td>\n",
       "      <td>0.005135</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>203.750000</td>\n",
       "      <td>11.106476</td>\n",
       "      <td>...</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>117.000000</td>\n",
       "      <td>0.937500</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>109.000000</td>\n",
       "      <td>2.162470</td>\n",
       "      <td>0.017379</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>50%</th>\n",
       "      <td>65.500000</td>\n",
       "      <td>11.819128</td>\n",
       "      <td>1.558462</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>29.242462</td>\n",
       "      <td>1.981591</td>\n",
       "      <td>0.175769</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>230.500000</td>\n",
       "      <td>37.201621</td>\n",
       "      <td>...</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>127.500000</td>\n",
       "      <td>4.611250</td>\n",
       "      <td>0.118750</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>172.500000</td>\n",
       "      <td>22.308789</td>\n",
       "      <td>5.523850</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>75%</th>\n",
       "      <td>71.000000</td>\n",
       "      <td>21.341406</td>\n",
       "      <td>4.143695</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>31.914649</td>\n",
       "      <td>4.429522</td>\n",
       "      <td>0.782845</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>266.250000</td>\n",
       "      <td>92.793687</td>\n",
       "      <td>...</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>140.250000</td>\n",
       "      <td>15.403571</td>\n",
       "      <td>2.206562</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>227.000000</td>\n",
       "      <td>60.695171</td>\n",
       "      <td>21.358363</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>max</th>\n",
       "      <td>77.000000</td>\n",
       "      <td>32.903226</td>\n",
       "      <td>7.492196</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>39.183163</td>\n",
       "      <td>14.135882</td>\n",
       "      <td>3.743467</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>327.000000</td>\n",
       "      <td>196.066667</td>\n",
       "      <td>...</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>214.000000</td>\n",
       "      <td>130.133333</td>\n",
       "      <td>36.913333</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1149.000000</td>\n",
       "      <td>1024.179104</td>\n",
       "      <td>459.645453</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>8 rows × 32 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "            age6      age61      age62  ascvd_hx6       bmi6      bmi61  \\\n",
       "count  60.000000  60.000000  60.000000  60.000000  60.000000  60.000000   \n",
       "mean   64.150000  12.930808   2.320905   0.316667  28.993009   2.976886   \n",
       "std     8.313406  10.740757   2.537664   0.469102   4.397932   3.337312   \n",
       "min    49.000000   0.224766   0.000000   0.000000  19.285225   0.000000   \n",
       "25%    56.750000   2.713059   0.023153   0.000000  26.688324   0.653533   \n",
       "50%    65.500000  11.819128   1.558462   0.000000  29.242462   1.981591   \n",
       "75%    71.000000  21.341406   4.143695   1.000000  31.914649   4.429522   \n",
       "max    77.000000  32.903226   7.492196   1.000000  39.183163  14.135882   \n",
       "\n",
       "           bmi62    bpmeds6       chol5      chol51  ...    pad_hx6  \\\n",
       "count  60.000000  60.000000   60.000000   60.000000  ...  60.000000   \n",
       "mean    0.552253   0.633333  233.650000   56.843871  ...   0.100000   \n",
       "std     0.864667   0.485961   45.367361   54.002275  ...   0.302532   \n",
       "min     0.000000   0.000000  125.000000    0.000000  ...   0.000000   \n",
       "25%     0.005135   0.000000  203.750000   11.106476  ...   0.000000   \n",
       "50%     0.175769   1.000000  230.500000   37.201621  ...   0.000000   \n",
       "75%     0.782845   1.000000  266.250000   92.793687  ...   0.000000   \n",
       "max     3.743467   1.000000  327.000000  196.066667  ...   1.000000   \n",
       "\n",
       "             sbp6       sbp61      sbp62     smoke6    stk_hx6     mi_hx6  \\\n",
       "count   60.000000   60.000000  60.000000  60.000000  60.000000  60.000000   \n",
       "mean   131.266667   13.767726   2.836248   0.166667   0.066667   0.133333   \n",
       "std     20.984956   23.428731   6.461287   0.375823   0.251549   0.342803   \n",
       "min    102.000000    0.000000   0.000000   0.000000   0.000000   0.000000   \n",
       "25%    117.000000    0.937500   0.000000   0.000000   0.000000   0.000000   \n",
       "50%    127.500000    4.611250   0.118750   0.000000   0.000000   0.000000   \n",
       "75%    140.250000   15.403571   2.206562   0.000000   0.000000   0.000000   \n",
       "max    214.000000  130.133333  36.913333   1.000000   1.000000   1.000000   \n",
       "\n",
       "           trigly5     trigly51    trigly52  \n",
       "count    60.000000    60.000000   60.000000  \n",
       "mean    200.016667    62.346172   24.433698  \n",
       "std     149.939817   139.867447   62.606721  \n",
       "min      67.000000     0.000000    0.000000  \n",
       "25%     109.000000     2.162470    0.017379  \n",
       "50%     172.500000    22.308789    5.523850  \n",
       "75%     227.000000    60.695171   21.358363  \n",
       "max    1149.000000  1024.179104  459.645453  \n",
       "\n",
       "[8 rows x 32 columns]"
      ]
     },
     "execution_count": 104,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "x_1.describe()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Estimate Hazard"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 105,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Text(0.5, 0, 'Time (days)')"
      ]
     },
     "execution_count": 105,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "ex_i = 100\n",
    "# 250 = negative, 500 = positive , 400 = neutral for framingham\n",
    "# 90 = negative,  3=positive,  100=neutral actg175\n",
    "sns.distplot(T_0_pred[ex_i], label='$T_0$')\n",
    "sns.distplot(T_1_pred[ex_i], label='$T_1$')\n",
    "plt.legend(loc='best', fontsize=legend_size)\n",
    "plt.xlabel(\"Time ({})\".format(time),  fontsize=fontsize)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 106,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Text(0.5, 0, 'Time (days)')"
      ]
     },
     "execution_count": 106,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "#t_emp_range = np.sort(np.concatenate((T_0_pred[ex_i], T_1_pred[ex_i]), axis=0))\n",
    "t_emp_range = all_t_range\n",
    "kernel_t0 = stats.gaussian_kde(T_0_pred[ex_i]) \n",
    "kernel_t1 = stats.gaussian_kde(T_1_pred[ex_i]) \n",
    "\n",
    "surv_t0  = []\n",
    "surv_t1  = []\n",
    "#print(np.sort(T_0_pred[1]))\n",
    "for i in t_emp_range:\n",
    "    surv_t0.append(1-kernel_t0.integrate_box_1d(low=0, high=i))\n",
    "    surv_t1.append(1-kernel_t1.integrate_box_1d(low=0, high=i))\n",
    "    \n",
    "surv_t0 = np.array(surv_t0)\n",
    "surv_t1 = np.array(surv_t1)\n",
    "\n",
    "plt.plot(t_emp_range, surv_t0, label='$T_0$')\n",
    "plt.plot(t_emp_range, surv_t1, label='$T_1$')\n",
    "plt.legend(loc='best', fontsize=legend_size)\n",
    "plt.xlabel(\"Time ({})\".format(time),  fontsize=fontsize)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 107,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "CSA-INFO mean: HR EST Slope  -2.43 (-3.27, -0.35)\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/plain": [
       "<Figure size 432x288 with 0 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "s_0_prime_est = compute_grad(surv=surv_t0, time=t_emp_range)  \n",
    "s_1_prime_est = compute_grad(surv=surv_t1, time=t_emp_range)\n",
    "\n",
    "\n",
    "constant = 1e-8\n",
    "surv_ratio = (surv_t0 + constant)/ (surv_t1 + constant)\n",
    "surv_prime_ratio = (s_1_prime_est + constant )/(s_0_prime_est + constant)\n",
    "HR_est_slope=  np.log(surv_ratio * surv_prime_ratio)\n",
    "hr_dist(HR_est_slope, model + \" mean: HR EST Slope \")\n",
    "\n",
    "colors = sns.color_palette(\"husl\", 4)\n",
    "plt.xlabel(\"Time ({})\".format(time),  fontsize=fontsize)\n",
    "plt.ylabel(\" Log Hazard\",  fontsize=fontsize)\n",
    "\n",
    "plt.step(t_emp_range, np.log(-s_1_prime_est / surv_t1),  label=model+' ($h_1$(t))', lw=2,\n",
    "         color=colors[1])\n",
    "plt.step(t_emp_range, np.log(-s_0_prime_est / surv_t0),  label=model+' ($h_0$(t))', lw=2,\n",
    "        color=colors[0])\n",
    "\n",
    "\n",
    "plt.legend(loc='best', fontsize=legend_size)\n",
    "plt.show()\n",
    "plt.tight_layout()\n",
    "\n",
    "\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 108,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.step(t_emp_range, HR_est_slope, label=model, lw=2, color='grey')\n",
    "plt.tight_layout()\n",
    "plt.legend(loc='best', fontsize=legend_size)\n",
    "plt.xlabel(\"Time ({})\".format(time),  fontsize=fontsize)\n",
    "plt.ylabel(\"Log HR\",  fontsize=fontsize)\n",
    "plt.show()\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 109,
   "metadata": {},
   "outputs": [],
   "source": [
    "def compute_hr (t0_pred_i, t1_pred_i):\n",
    "    #t_emp_range = np.sort(np.concatenate((t0_pred_i, t1_pred_i), axis=0))\n",
    "\n",
    "    kernel_t0 = stats.gaussian_kde(t0_pred_i) \n",
    "    kernel_t1 = stats.gaussian_kde(t1_pred_i) \n",
    "\n",
    "    surv_t0  = []\n",
    "    surv_t1  = []\n",
    "\n",
    "    for i in all_t_range:\n",
    "        surv_t0.append(1-kernel_t0.integrate_box_1d(low=0, high=i))\n",
    "        surv_t1.append(1-kernel_t1.integrate_box_1d(low=0, high=i))\n",
    "    \n",
    "    surv_t0 = np.array(surv_t0)\n",
    "    surv_t1 = np.array(surv_t1)\n",
    "    \n",
    "    s_0_prime_est = compute_grad(surv=surv_t0, time=t_emp_range)\n",
    "    \n",
    "    s_1_prime_est = compute_grad(surv=surv_t1, time=t_emp_range)\n",
    "\n",
    "    #HR_est_slope= surv_t0/surv_t1 * (s_1_prime_est /s_0_prime_est)\n",
    "    constant = 1e-8\n",
    "    surv_ratio = surv_t0 / surv_t1\n",
    "    surv_prime_ratio = s_1_prime_est / s_0_prime_est \n",
    "    HR_est_slope=  np.log( surv_ratio  *  surv_prime_ratio + constant)\n",
    "    \n",
    "    return HR_est_slope\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 110,
   "metadata": {},
   "outputs": [],
   "source": [
    "all_HR = []\n",
    "all_HR_median = []\n",
    "for i in np.arange(T_0_pred.shape[0]):\n",
    "    hr = compute_hr(t0_pred_i=T_0_pred[i], t1_pred_i=T_1_pred[i])\n",
    "    all_HR_median.append(np.median(hr))\n",
    "    all_HR.append(hr)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 111,
   "metadata": {},
   "outputs": [],
   "source": [
    "all_HR = np.array(all_HR)\n",
    "all_HR_median = np.array(all_HR_median)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Median Covariates  Statistics"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 112,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "positive:  0.6893203883495146\n"
     ]
    }
   ],
   "source": [
    "print(\"positive: \", len(all_HR_median[all_HR_median < 0 ])/ len(all_HR_median))\n",
    "pos_median = x0_x1_conc[all_HR_median < 0]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 113,
   "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>age6</th>\n",
       "      <th>ascvd_hx6</th>\n",
       "      <th>bmi6</th>\n",
       "      <th>bpmeds6</th>\n",
       "      <th>chol5</th>\n",
       "      <th>dbp6</th>\n",
       "      <th>diab6</th>\n",
       "      <th>female</th>\n",
       "      <th>gluc5</th>\n",
       "      <th>hdl5</th>\n",
       "      <th>pad_hx6</th>\n",
       "      <th>sbp6</th>\n",
       "      <th>smoke6</th>\n",
       "      <th>stk_hx6</th>\n",
       "      <th>mi_hx6</th>\n",
       "      <th>trigly5</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>count</th>\n",
       "      <td>355.000000</td>\n",
       "      <td>355.000000</td>\n",
       "      <td>355.000000</td>\n",
       "      <td>355.000000</td>\n",
       "      <td>355.000000</td>\n",
       "      <td>355.000000</td>\n",
       "      <td>355.000000</td>\n",
       "      <td>355.000000</td>\n",
       "      <td>355.000000</td>\n",
       "      <td>355.000000</td>\n",
       "      <td>355.000000</td>\n",
       "      <td>355.000000</td>\n",
       "      <td>355.000000</td>\n",
       "      <td>355.000000</td>\n",
       "      <td>355.000000</td>\n",
       "      <td>355.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>mean</th>\n",
       "      <td>57.535211</td>\n",
       "      <td>0.101408</td>\n",
       "      <td>28.535491</td>\n",
       "      <td>0.312676</td>\n",
       "      <td>209.070423</td>\n",
       "      <td>76.504225</td>\n",
       "      <td>0.121127</td>\n",
       "      <td>0.487324</td>\n",
       "      <td>103.008451</td>\n",
       "      <td>46.171831</td>\n",
       "      <td>0.014085</td>\n",
       "      <td>127.839437</td>\n",
       "      <td>0.188732</td>\n",
       "      <td>0.022535</td>\n",
       "      <td>0.039437</td>\n",
       "      <td>162.659155</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>std</th>\n",
       "      <td>9.138963</td>\n",
       "      <td>0.302295</td>\n",
       "      <td>4.778960</td>\n",
       "      <td>0.464238</td>\n",
       "      <td>38.343056</td>\n",
       "      <td>8.671458</td>\n",
       "      <td>0.326735</td>\n",
       "      <td>0.500545</td>\n",
       "      <td>34.024541</td>\n",
       "      <td>12.163031</td>\n",
       "      <td>0.118006</td>\n",
       "      <td>17.504628</td>\n",
       "      <td>0.391848</td>\n",
       "      <td>0.148626</td>\n",
       "      <td>0.194906</td>\n",
       "      <td>91.827395</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>min</th>\n",
       "      <td>35.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>19.285225</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>105.000000</td>\n",
       "      <td>53.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>66.000000</td>\n",
       "      <td>16.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>91.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>38.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>25%</th>\n",
       "      <td>51.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>25.284703</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>181.000000</td>\n",
       "      <td>70.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>90.000000</td>\n",
       "      <td>37.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>115.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>101.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>50%</th>\n",
       "      <td>56.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>28.021534</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>204.000000</td>\n",
       "      <td>76.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>95.000000</td>\n",
       "      <td>45.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>125.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>140.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>75%</th>\n",
       "      <td>64.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>30.888693</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>232.000000</td>\n",
       "      <td>82.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>103.000000</td>\n",
       "      <td>53.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>138.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>197.500000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>max</th>\n",
       "      <td>84.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>49.154229</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>327.000000</td>\n",
       "      <td>105.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>315.000000</td>\n",
       "      <td>95.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>192.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>813.000000</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "             age6   ascvd_hx6        bmi6     bpmeds6       chol5        dbp6  \\\n",
       "count  355.000000  355.000000  355.000000  355.000000  355.000000  355.000000   \n",
       "mean    57.535211    0.101408   28.535491    0.312676  209.070423   76.504225   \n",
       "std      9.138963    0.302295    4.778960    0.464238   38.343056    8.671458   \n",
       "min     35.000000    0.000000   19.285225    0.000000  105.000000   53.000000   \n",
       "25%     51.000000    0.000000   25.284703    0.000000  181.000000   70.000000   \n",
       "50%     56.000000    0.000000   28.021534    0.000000  204.000000   76.000000   \n",
       "75%     64.000000    0.000000   30.888693    1.000000  232.000000   82.000000   \n",
       "max     84.000000    1.000000   49.154229    1.000000  327.000000  105.000000   \n",
       "\n",
       "            diab6      female       gluc5        hdl5     pad_hx6        sbp6  \\\n",
       "count  355.000000  355.000000  355.000000  355.000000  355.000000  355.000000   \n",
       "mean     0.121127    0.487324  103.008451   46.171831    0.014085  127.839437   \n",
       "std      0.326735    0.500545   34.024541   12.163031    0.118006   17.504628   \n",
       "min      0.000000    0.000000   66.000000   16.000000    0.000000   91.000000   \n",
       "25%      0.000000    0.000000   90.000000   37.000000    0.000000  115.000000   \n",
       "50%      0.000000    0.000000   95.000000   45.000000    0.000000  125.000000   \n",
       "75%      0.000000    1.000000  103.000000   53.000000    0.000000  138.000000   \n",
       "max      1.000000    1.000000  315.000000   95.000000    1.000000  192.000000   \n",
       "\n",
       "           smoke6     stk_hx6      mi_hx6     trigly5  \n",
       "count  355.000000  355.000000  355.000000  355.000000  \n",
       "mean     0.188732    0.022535    0.039437  162.659155  \n",
       "std      0.391848    0.148626    0.194906   91.827395  \n",
       "min      0.000000    0.000000    0.000000   38.000000  \n",
       "25%      0.000000    0.000000    0.000000  101.000000  \n",
       "50%      0.000000    0.000000    0.000000  140.000000  \n",
       "75%      0.000000    0.000000    0.000000  197.500000  \n",
       "max      1.000000    1.000000    1.000000  813.000000  "
      ]
     },
     "execution_count": 113,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "pos_median[sub_covariates].describe()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 114,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "negative 0.3106796116504854\n"
     ]
    }
   ],
   "source": [
    "print(\"negative\", len(all_HR_median[all_HR_median > 0])/ len(all_HR_median))\n",
    "neg_median = x0_x1_conc[all_HR_median>0]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 115,
   "metadata": {
    "scrolled": true
   },
   "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>age6</th>\n",
       "      <th>ascvd_hx6</th>\n",
       "      <th>bmi6</th>\n",
       "      <th>bpmeds6</th>\n",
       "      <th>chol5</th>\n",
       "      <th>dbp6</th>\n",
       "      <th>diab6</th>\n",
       "      <th>female</th>\n",
       "      <th>gluc5</th>\n",
       "      <th>hdl5</th>\n",
       "      <th>pad_hx6</th>\n",
       "      <th>sbp6</th>\n",
       "      <th>smoke6</th>\n",
       "      <th>stk_hx6</th>\n",
       "      <th>mi_hx6</th>\n",
       "      <th>trigly5</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>count</th>\n",
       "      <td>160.000000</td>\n",
       "      <td>160.000000</td>\n",
       "      <td>160.000000</td>\n",
       "      <td>160.000000</td>\n",
       "      <td>160.000000</td>\n",
       "      <td>160.000000</td>\n",
       "      <td>160.000000</td>\n",
       "      <td>160.000000</td>\n",
       "      <td>160.000000</td>\n",
       "      <td>160.000000</td>\n",
       "      <td>160.000000</td>\n",
       "      <td>160.000000</td>\n",
       "      <td>160.000000</td>\n",
       "      <td>160.000000</td>\n",
       "      <td>160.000000</td>\n",
       "      <td>160.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>mean</th>\n",
       "      <td>60.343750</td>\n",
       "      <td>0.131250</td>\n",
       "      <td>26.719658</td>\n",
       "      <td>0.275000</td>\n",
       "      <td>199.143750</td>\n",
       "      <td>71.881250</td>\n",
       "      <td>0.056250</td>\n",
       "      <td>0.575000</td>\n",
       "      <td>96.712500</td>\n",
       "      <td>57.337500</td>\n",
       "      <td>0.043750</td>\n",
       "      <td>127.787500</td>\n",
       "      <td>0.193750</td>\n",
       "      <td>0.043750</td>\n",
       "      <td>0.056250</td>\n",
       "      <td>124.350000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>std</th>\n",
       "      <td>10.543141</td>\n",
       "      <td>0.338734</td>\n",
       "      <td>5.355192</td>\n",
       "      <td>0.447916</td>\n",
       "      <td>29.679387</td>\n",
       "      <td>11.176329</td>\n",
       "      <td>0.231127</td>\n",
       "      <td>0.495895</td>\n",
       "      <td>19.212937</td>\n",
       "      <td>15.802988</td>\n",
       "      <td>0.205181</td>\n",
       "      <td>21.059779</td>\n",
       "      <td>0.396476</td>\n",
       "      <td>0.205181</td>\n",
       "      <td>0.231127</td>\n",
       "      <td>128.238732</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>min</th>\n",
       "      <td>35.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>17.676632</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>118.000000</td>\n",
       "      <td>49.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>48.000000</td>\n",
       "      <td>22.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>88.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>33.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>25%</th>\n",
       "      <td>53.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>22.988095</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>180.000000</td>\n",
       "      <td>63.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>89.000000</td>\n",
       "      <td>47.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>111.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>65.750000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>50%</th>\n",
       "      <td>60.500000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>25.768858</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>199.500000</td>\n",
       "      <td>70.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>95.000000</td>\n",
       "      <td>58.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>127.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>89.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>75%</th>\n",
       "      <td>69.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>29.605862</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>219.250000</td>\n",
       "      <td>81.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>101.000000</td>\n",
       "      <td>67.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>139.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>119.250000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>max</th>\n",
       "      <td>80.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>45.992112</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>298.000000</td>\n",
       "      <td>105.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>239.000000</td>\n",
       "      <td>101.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>214.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1149.000000</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "             age6   ascvd_hx6        bmi6     bpmeds6       chol5        dbp6  \\\n",
       "count  160.000000  160.000000  160.000000  160.000000  160.000000  160.000000   \n",
       "mean    60.343750    0.131250   26.719658    0.275000  199.143750   71.881250   \n",
       "std     10.543141    0.338734    5.355192    0.447916   29.679387   11.176329   \n",
       "min     35.000000    0.000000   17.676632    0.000000  118.000000   49.000000   \n",
       "25%     53.000000    0.000000   22.988095    0.000000  180.000000   63.000000   \n",
       "50%     60.500000    0.000000   25.768858    0.000000  199.500000   70.000000   \n",
       "75%     69.000000    0.000000   29.605862    1.000000  219.250000   81.000000   \n",
       "max     80.000000    1.000000   45.992112    1.000000  298.000000  105.000000   \n",
       "\n",
       "            diab6      female       gluc5        hdl5     pad_hx6        sbp6  \\\n",
       "count  160.000000  160.000000  160.000000  160.000000  160.000000  160.000000   \n",
       "mean     0.056250    0.575000   96.712500   57.337500    0.043750  127.787500   \n",
       "std      0.231127    0.495895   19.212937   15.802988    0.205181   21.059779   \n",
       "min      0.000000    0.000000   48.000000   22.000000    0.000000   88.000000   \n",
       "25%      0.000000    0.000000   89.000000   47.000000    0.000000  111.000000   \n",
       "50%      0.000000    1.000000   95.000000   58.000000    0.000000  127.000000   \n",
       "75%      0.000000    1.000000  101.000000   67.000000    0.000000  139.000000   \n",
       "max      1.000000    1.000000  239.000000  101.000000    1.000000  214.000000   \n",
       "\n",
       "           smoke6     stk_hx6      mi_hx6      trigly5  \n",
       "count  160.000000  160.000000  160.000000   160.000000  \n",
       "mean     0.193750    0.043750    0.056250   124.350000  \n",
       "std      0.396476    0.205181    0.231127   128.238732  \n",
       "min      0.000000    0.000000    0.000000    33.000000  \n",
       "25%      0.000000    0.000000    0.000000    65.750000  \n",
       "50%      0.000000    0.000000    0.000000    89.000000  \n",
       "75%      0.000000    0.000000    0.000000   119.250000  \n",
       "max      1.000000    1.000000    1.000000  1149.000000  "
      ]
     },
     "execution_count": 115,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "neg_median[sub_covariates].describe()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 116,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "neutral 0.0\n"
     ]
    }
   ],
   "source": [
    "print(\"neutral\", len(all_HR_median[all_HR_median == 0])/ len(all_HR_median))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Cluster Map"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 117,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "preserved_distance:0.6062902891662171\n"
     ]
    }
   ],
   "source": [
    "linkage_data = linkage(all_HR, 'ward')\n",
    "c, coph_dists = cophenet(linkage_data, pdist(all_HR))\n",
    "print(\"preserved_distance:{}\".format(c))\n",
    "\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 118,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "\n",
    "plt.figure()\n",
    "    # plt.title('{} Hierarchical Clustering Dendrogram'.format(name), fontsize=title_fontsize)\n",
    "plt.xlabel('sample index or (cluster size)')\n",
    "plt.ylabel('distance')\n",
    "dendrogram(\n",
    "        linkage_data,\n",
    "        truncate_mode='lastp',  # show only the last p merged clusters\n",
    "        p=12,  # show only the last p merged clusters\n",
    "        leaf_rotation=90.,\n",
    "        leaf_font_size=12,\n",
    "        show_contracted=True,  # to get a distribution impression in truncated branches\n",
    "    )\n",
    "plt.tick_params(axis='both', which='major', labelsize=10)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 119,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/Users/paidamoyochapfuwa/.pyenv/versions/3.6.4/envs/csa_prod/lib/python3.6/site-packages/seaborn/matrix.py:624: UserWarning: Clustering large matrix with scipy. Installing `fastcluster` may give better performance.\n",
      "  warnings.warn(msg)\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "<Figure size 360x288 with 0 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 360x288 with 4 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "fig =plt.figure(figsize=(5,4))\n",
    "g=sns.clustermap(all_HR,  cmap=\"jet\", metric= 'euclidean', figsize=(5, 4),\n",
    "                   yticklabels=False, xticklabels=False, col_cluster=False)\n",
    "g.cax.set_position([1, .2, .03, .45])\n",
    "g.ax_row_dendrogram.set_visible(False) "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 120,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "clusters:  (515,) [1 2 3 4]\n"
     ]
    }
   ],
   "source": [
    "if data == 'framingham':\n",
    "    num_clusters = 4\n",
    "else:\n",
    "     num_clusters = 2\n",
    "clusters = fcluster(linkage_data, num_clusters, criterion='maxclust')\n",
    "print(\"clusters: \", clusters.shape, np.unique(clusters))\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 121,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "c = 1  hr_c:  (17, 516)\n",
      "c = 2  hr_c:  (251, 516)\n",
      "c = 3  hr_c:  (45, 516)\n",
      "c = 4  hr_c:  (202, 516)\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "<matplotlib.legend.Legend at 0x12cff2390>"
      ]
     },
     "execution_count": 121,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 360x288 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "fig=plt.figure(figsize=(5,4))\n",
    "model_colors = sns.color_palette(\"husl\", num_clusters)\n",
    "handles = []\n",
    "for i in np.unique(clusters):\n",
    "    idx = clusters == i\n",
    "    hr_c = all_HR[idx]\n",
    "    print('c =', i, \" hr_c: \", hr_c.shape)\n",
    "    mean_hr_c =  np.mean(hr_c, axis=0)\n",
    "    handles.append(mpatches.Patch(color=model_colors[i-1], label=\"{} ({})\".format(i, len(hr_c))))\n",
    "    plt.plot(all_t_range, mean_hr_c, c=model_colors[i-1], linewidth=2)\n",
    "\n",
    "handles.append(mpatches.Patch(color='black', label=\"{} ({})\".format('Average', len(all_HR))))\n",
    "handles.append(mpatches.Patch(color='grey', label=\"RCT\"))\n",
    "plt.plot(all_t_range, np.mean(all_HR, axis=0), c='black', linewidth=2)\n",
    "plt.axhline(y= np.log(0.75), color='grey', label='mean', linestyle='--', lw=3)\n",
    "ylabel =  '$\\log$'  + ' ' +  'HR' + '$(t|x)$'    \n",
    "plt.xlabel(\"Time ({})\".format(time),  fontsize=fontsize)\n",
    "plt.ylabel(ylabel ,  fontsize=fontsize)\n",
    "plt.tight_layout()\n",
    "#plt.legend(handles=handles, fontsize=legend_size, bbox_to_anchor=(0.95,1.05))\n",
    "plt.legend(handles=handles, fontsize=legend_size, loc='best')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 122,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>age6</th>\n",
       "      <th>ascvd_hx6</th>\n",
       "      <th>bmi6</th>\n",
       "      <th>bpmeds6</th>\n",
       "      <th>chol5</th>\n",
       "      <th>dbp6</th>\n",
       "      <th>diab6</th>\n",
       "      <th>female</th>\n",
       "      <th>gluc5</th>\n",
       "      <th>hdl5</th>\n",
       "      <th>pad_hx6</th>\n",
       "      <th>sbp6</th>\n",
       "      <th>smoke6</th>\n",
       "      <th>stk_hx6</th>\n",
       "      <th>mi_hx6</th>\n",
       "      <th>trigly5</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>count</th>\n",
       "      <td>17.000000</td>\n",
       "      <td>17.000000</td>\n",
       "      <td>17.000000</td>\n",
       "      <td>17.000000</td>\n",
       "      <td>17.000000</td>\n",
       "      <td>17.000000</td>\n",
       "      <td>17.000000</td>\n",
       "      <td>17.000000</td>\n",
       "      <td>17.000000</td>\n",
       "      <td>17.000000</td>\n",
       "      <td>17.000000</td>\n",
       "      <td>17.000000</td>\n",
       "      <td>17.000000</td>\n",
       "      <td>17.0</td>\n",
       "      <td>17.000000</td>\n",
       "      <td>17.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>mean</th>\n",
       "      <td>66.941176</td>\n",
       "      <td>0.352941</td>\n",
       "      <td>30.911996</td>\n",
       "      <td>0.647059</td>\n",
       "      <td>199.235294</td>\n",
       "      <td>78.058824</td>\n",
       "      <td>0.117647</td>\n",
       "      <td>0.352941</td>\n",
       "      <td>97.235294</td>\n",
       "      <td>45.764706</td>\n",
       "      <td>0.117647</td>\n",
       "      <td>140.647059</td>\n",
       "      <td>0.176471</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.235294</td>\n",
       "      <td>228.588235</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>std</th>\n",
       "      <td>8.613003</td>\n",
       "      <td>0.492592</td>\n",
       "      <td>7.949390</td>\n",
       "      <td>0.492592</td>\n",
       "      <td>19.501697</td>\n",
       "      <td>12.577115</td>\n",
       "      <td>0.332106</td>\n",
       "      <td>0.492592</td>\n",
       "      <td>19.350483</td>\n",
       "      <td>18.081515</td>\n",
       "      <td>0.332106</td>\n",
       "      <td>13.181147</td>\n",
       "      <td>0.392953</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.437237</td>\n",
       "      <td>193.054739</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>min</th>\n",
       "      <td>51.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>22.343918</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>169.000000</td>\n",
       "      <td>61.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>48.000000</td>\n",
       "      <td>22.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>121.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>48.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>25%</th>\n",
       "      <td>59.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>23.869369</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>190.000000</td>\n",
       "      <td>68.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>90.000000</td>\n",
       "      <td>30.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>133.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>79.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>50%</th>\n",
       "      <td>71.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>28.261316</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>196.000000</td>\n",
       "      <td>75.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>94.000000</td>\n",
       "      <td>40.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>140.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>119.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>75%</th>\n",
       "      <td>74.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>33.377669</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>204.000000</td>\n",
       "      <td>87.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>105.000000</td>\n",
       "      <td>60.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>146.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>323.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>max</th>\n",
       "      <td>76.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>45.992112</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>244.000000</td>\n",
       "      <td>98.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>145.000000</td>\n",
       "      <td>84.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>180.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>678.000000</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "            age6  ascvd_hx6       bmi6    bpmeds6       chol5       dbp6  \\\n",
       "count  17.000000  17.000000  17.000000  17.000000   17.000000  17.000000   \n",
       "mean   66.941176   0.352941  30.911996   0.647059  199.235294  78.058824   \n",
       "std     8.613003   0.492592   7.949390   0.492592   19.501697  12.577115   \n",
       "min    51.000000   0.000000  22.343918   0.000000  169.000000  61.000000   \n",
       "25%    59.000000   0.000000  23.869369   0.000000  190.000000  68.000000   \n",
       "50%    71.000000   0.000000  28.261316   1.000000  196.000000  75.000000   \n",
       "75%    74.000000   1.000000  33.377669   1.000000  204.000000  87.000000   \n",
       "max    76.000000   1.000000  45.992112   1.000000  244.000000  98.000000   \n",
       "\n",
       "           diab6     female       gluc5       hdl5    pad_hx6        sbp6  \\\n",
       "count  17.000000  17.000000   17.000000  17.000000  17.000000   17.000000   \n",
       "mean    0.117647   0.352941   97.235294  45.764706   0.117647  140.647059   \n",
       "std     0.332106   0.492592   19.350483  18.081515   0.332106   13.181147   \n",
       "min     0.000000   0.000000   48.000000  22.000000   0.000000  121.000000   \n",
       "25%     0.000000   0.000000   90.000000  30.000000   0.000000  133.000000   \n",
       "50%     0.000000   0.000000   94.000000  40.000000   0.000000  140.000000   \n",
       "75%     0.000000   1.000000  105.000000  60.000000   0.000000  146.000000   \n",
       "max     1.000000   1.000000  145.000000  84.000000   1.000000  180.000000   \n",
       "\n",
       "          smoke6  stk_hx6     mi_hx6     trigly5  \n",
       "count  17.000000     17.0  17.000000   17.000000  \n",
       "mean    0.176471      0.0   0.235294  228.588235  \n",
       "std     0.392953      0.0   0.437237  193.054739  \n",
       "min     0.000000      0.0   0.000000   48.000000  \n",
       "25%     0.000000      0.0   0.000000   79.000000  \n",
       "50%     0.000000      0.0   0.000000  119.000000  \n",
       "75%     0.000000      0.0   0.000000  323.000000  \n",
       "max     1.000000      0.0   1.000000  678.000000  "
      ]
     },
     "execution_count": 122,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "c1_cov = x0_x1_conc[clusters==1]\n",
    "c1_cov[sub_covariates].describe()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 123,
   "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>age6</th>\n",
       "      <th>ascvd_hx6</th>\n",
       "      <th>bmi6</th>\n",
       "      <th>bpmeds6</th>\n",
       "      <th>chol5</th>\n",
       "      <th>dbp6</th>\n",
       "      <th>diab6</th>\n",
       "      <th>female</th>\n",
       "      <th>gluc5</th>\n",
       "      <th>hdl5</th>\n",
       "      <th>pad_hx6</th>\n",
       "      <th>sbp6</th>\n",
       "      <th>smoke6</th>\n",
       "      <th>stk_hx6</th>\n",
       "      <th>mi_hx6</th>\n",
       "      <th>trigly5</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>count</th>\n",
       "      <td>251.000000</td>\n",
       "      <td>251.000000</td>\n",
       "      <td>251.000000</td>\n",
       "      <td>251.000000</td>\n",
       "      <td>251.000000</td>\n",
       "      <td>251.000000</td>\n",
       "      <td>251.000000</td>\n",
       "      <td>251.000000</td>\n",
       "      <td>251.000000</td>\n",
       "      <td>251.000000</td>\n",
       "      <td>251.000000</td>\n",
       "      <td>251.000000</td>\n",
       "      <td>251.000000</td>\n",
       "      <td>251.000000</td>\n",
       "      <td>251.000000</td>\n",
       "      <td>251.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>mean</th>\n",
       "      <td>56.944223</td>\n",
       "      <td>0.091633</td>\n",
       "      <td>28.601880</td>\n",
       "      <td>0.338645</td>\n",
       "      <td>207.741036</td>\n",
       "      <td>77.250996</td>\n",
       "      <td>0.119522</td>\n",
       "      <td>0.454183</td>\n",
       "      <td>101.737052</td>\n",
       "      <td>45.254980</td>\n",
       "      <td>0.007968</td>\n",
       "      <td>127.820717</td>\n",
       "      <td>0.171315</td>\n",
       "      <td>0.023904</td>\n",
       "      <td>0.035857</td>\n",
       "      <td>165.131474</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>std</th>\n",
       "      <td>9.039739</td>\n",
       "      <td>0.289084</td>\n",
       "      <td>4.381290</td>\n",
       "      <td>0.474195</td>\n",
       "      <td>40.415649</td>\n",
       "      <td>8.204678</td>\n",
       "      <td>0.325050</td>\n",
       "      <td>0.498891</td>\n",
       "      <td>29.466499</td>\n",
       "      <td>11.956535</td>\n",
       "      <td>0.089086</td>\n",
       "      <td>16.472879</td>\n",
       "      <td>0.377537</td>\n",
       "      <td>0.153057</td>\n",
       "      <td>0.186304</td>\n",
       "      <td>91.672300</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>min</th>\n",
       "      <td>35.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>19.632587</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>105.000000</td>\n",
       "      <td>55.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>75.000000</td>\n",
       "      <td>16.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>98.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>38.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>25%</th>\n",
       "      <td>51.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>25.647857</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>180.000000</td>\n",
       "      <td>71.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>90.000000</td>\n",
       "      <td>36.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>116.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>112.500000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>50%</th>\n",
       "      <td>55.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>28.249814</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>202.000000</td>\n",
       "      <td>77.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>95.000000</td>\n",
       "      <td>44.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>126.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>142.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>75%</th>\n",
       "      <td>63.500000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>31.033133</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>228.500000</td>\n",
       "      <td>82.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>104.000000</td>\n",
       "      <td>52.500000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>137.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>199.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>max</th>\n",
       "      <td>84.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>45.135681</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>327.000000</td>\n",
       "      <td>102.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>289.000000</td>\n",
       "      <td>95.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>192.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>813.000000</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "             age6   ascvd_hx6        bmi6     bpmeds6       chol5        dbp6  \\\n",
       "count  251.000000  251.000000  251.000000  251.000000  251.000000  251.000000   \n",
       "mean    56.944223    0.091633   28.601880    0.338645  207.741036   77.250996   \n",
       "std      9.039739    0.289084    4.381290    0.474195   40.415649    8.204678   \n",
       "min     35.000000    0.000000   19.632587    0.000000  105.000000   55.000000   \n",
       "25%     51.000000    0.000000   25.647857    0.000000  180.000000   71.000000   \n",
       "50%     55.000000    0.000000   28.249814    0.000000  202.000000   77.000000   \n",
       "75%     63.500000    0.000000   31.033133    1.000000  228.500000   82.000000   \n",
       "max     84.000000    1.000000   45.135681    1.000000  327.000000  102.000000   \n",
       "\n",
       "            diab6      female       gluc5        hdl5     pad_hx6        sbp6  \\\n",
       "count  251.000000  251.000000  251.000000  251.000000  251.000000  251.000000   \n",
       "mean     0.119522    0.454183  101.737052   45.254980    0.007968  127.820717   \n",
       "std      0.325050    0.498891   29.466499   11.956535    0.089086   16.472879   \n",
       "min      0.000000    0.000000   75.000000   16.000000    0.000000   98.000000   \n",
       "25%      0.000000    0.000000   90.000000   36.000000    0.000000  116.000000   \n",
       "50%      0.000000    0.000000   95.000000   44.000000    0.000000  126.000000   \n",
       "75%      0.000000    1.000000  104.000000   52.500000    0.000000  137.000000   \n",
       "max      1.000000    1.000000  289.000000   95.000000    1.000000  192.000000   \n",
       "\n",
       "           smoke6     stk_hx6      mi_hx6     trigly5  \n",
       "count  251.000000  251.000000  251.000000  251.000000  \n",
       "mean     0.171315    0.023904    0.035857  165.131474  \n",
       "std      0.377537    0.153057    0.186304   91.672300  \n",
       "min      0.000000    0.000000    0.000000   38.000000  \n",
       "25%      0.000000    0.000000    0.000000  112.500000  \n",
       "50%      0.000000    0.000000    0.000000  142.000000  \n",
       "75%      0.000000    0.000000    0.000000  199.000000  \n",
       "max      1.000000    1.000000    1.000000  813.000000  "
      ]
     },
     "execution_count": 123,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "c2_cov = x0_x1_conc[clusters==2]\n",
    "c2_cov[sub_covariates].describe()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 124,
   "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>age6</th>\n",
       "      <th>ascvd_hx6</th>\n",
       "      <th>bmi6</th>\n",
       "      <th>bpmeds6</th>\n",
       "      <th>chol5</th>\n",
       "      <th>dbp6</th>\n",
       "      <th>diab6</th>\n",
       "      <th>female</th>\n",
       "      <th>gluc5</th>\n",
       "      <th>hdl5</th>\n",
       "      <th>pad_hx6</th>\n",
       "      <th>sbp6</th>\n",
       "      <th>smoke6</th>\n",
       "      <th>stk_hx6</th>\n",
       "      <th>mi_hx6</th>\n",
       "      <th>trigly5</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>count</th>\n",
       "      <td>45.000000</td>\n",
       "      <td>45.000000</td>\n",
       "      <td>45.000000</td>\n",
       "      <td>45.00000</td>\n",
       "      <td>45.000000</td>\n",
       "      <td>45.000000</td>\n",
       "      <td>45.000000</td>\n",
       "      <td>45.000000</td>\n",
       "      <td>45.000000</td>\n",
       "      <td>45.000000</td>\n",
       "      <td>45.000000</td>\n",
       "      <td>45.000000</td>\n",
       "      <td>45.000000</td>\n",
       "      <td>45.000000</td>\n",
       "      <td>45.000000</td>\n",
       "      <td>45.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>mean</th>\n",
       "      <td>58.088889</td>\n",
       "      <td>0.133333</td>\n",
       "      <td>25.281231</td>\n",
       "      <td>0.20000</td>\n",
       "      <td>199.200000</td>\n",
       "      <td>70.088889</td>\n",
       "      <td>0.044444</td>\n",
       "      <td>0.688889</td>\n",
       "      <td>94.555556</td>\n",
       "      <td>64.088889</td>\n",
       "      <td>0.044444</td>\n",
       "      <td>125.022222</td>\n",
       "      <td>0.244444</td>\n",
       "      <td>0.066667</td>\n",
       "      <td>0.066667</td>\n",
       "      <td>110.866667</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>std</th>\n",
       "      <td>10.175332</td>\n",
       "      <td>0.343776</td>\n",
       "      <td>4.455141</td>\n",
       "      <td>0.40452</td>\n",
       "      <td>27.682945</td>\n",
       "      <td>11.745061</td>\n",
       "      <td>0.208409</td>\n",
       "      <td>0.468179</td>\n",
       "      <td>11.073218</td>\n",
       "      <td>16.697608</td>\n",
       "      <td>0.208409</td>\n",
       "      <td>18.288706</td>\n",
       "      <td>0.434613</td>\n",
       "      <td>0.252262</td>\n",
       "      <td>0.252262</td>\n",
       "      <td>169.896840</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>min</th>\n",
       "      <td>37.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>18.191485</td>\n",
       "      <td>0.00000</td>\n",
       "      <td>137.000000</td>\n",
       "      <td>49.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>77.000000</td>\n",
       "      <td>27.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>94.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>33.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>25%</th>\n",
       "      <td>50.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>22.238275</td>\n",
       "      <td>0.00000</td>\n",
       "      <td>180.000000</td>\n",
       "      <td>61.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>86.000000</td>\n",
       "      <td>54.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>111.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>59.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>50%</th>\n",
       "      <td>57.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>24.541343</td>\n",
       "      <td>0.00000</td>\n",
       "      <td>198.000000</td>\n",
       "      <td>67.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>95.000000</td>\n",
       "      <td>64.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>122.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>78.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>75%</th>\n",
       "      <td>67.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>28.241633</td>\n",
       "      <td>0.00000</td>\n",
       "      <td>212.000000</td>\n",
       "      <td>77.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>101.000000</td>\n",
       "      <td>79.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>138.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>91.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>max</th>\n",
       "      <td>77.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>39.148276</td>\n",
       "      <td>1.00000</td>\n",
       "      <td>290.000000</td>\n",
       "      <td>97.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>132.000000</td>\n",
       "      <td>101.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>165.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1149.000000</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "            age6  ascvd_hx6       bmi6   bpmeds6       chol5       dbp6  \\\n",
       "count  45.000000  45.000000  45.000000  45.00000   45.000000  45.000000   \n",
       "mean   58.088889   0.133333  25.281231   0.20000  199.200000  70.088889   \n",
       "std    10.175332   0.343776   4.455141   0.40452   27.682945  11.745061   \n",
       "min    37.000000   0.000000  18.191485   0.00000  137.000000  49.000000   \n",
       "25%    50.000000   0.000000  22.238275   0.00000  180.000000  61.000000   \n",
       "50%    57.000000   0.000000  24.541343   0.00000  198.000000  67.000000   \n",
       "75%    67.000000   0.000000  28.241633   0.00000  212.000000  77.000000   \n",
       "max    77.000000   1.000000  39.148276   1.00000  290.000000  97.000000   \n",
       "\n",
       "           diab6     female       gluc5        hdl5    pad_hx6        sbp6  \\\n",
       "count  45.000000  45.000000   45.000000   45.000000  45.000000   45.000000   \n",
       "mean    0.044444   0.688889   94.555556   64.088889   0.044444  125.022222   \n",
       "std     0.208409   0.468179   11.073218   16.697608   0.208409   18.288706   \n",
       "min     0.000000   0.000000   77.000000   27.000000   0.000000   94.000000   \n",
       "25%     0.000000   0.000000   86.000000   54.000000   0.000000  111.000000   \n",
       "50%     0.000000   1.000000   95.000000   64.000000   0.000000  122.000000   \n",
       "75%     0.000000   1.000000  101.000000   79.000000   0.000000  138.000000   \n",
       "max     1.000000   1.000000  132.000000  101.000000   1.000000  165.000000   \n",
       "\n",
       "          smoke6    stk_hx6     mi_hx6      trigly5  \n",
       "count  45.000000  45.000000  45.000000    45.000000  \n",
       "mean    0.244444   0.066667   0.066667   110.866667  \n",
       "std     0.434613   0.252262   0.252262   169.896840  \n",
       "min     0.000000   0.000000   0.000000    33.000000  \n",
       "25%     0.000000   0.000000   0.000000    59.000000  \n",
       "50%     0.000000   0.000000   0.000000    78.000000  \n",
       "75%     0.000000   0.000000   0.000000    91.000000  \n",
       "max     1.000000   1.000000   1.000000  1149.000000  "
      ]
     },
     "execution_count": 124,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "c3_cov = x0_x1_conc[clusters==3]\n",
    "c3_cov[sub_covariates].describe()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 125,
   "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>age6</th>\n",
       "      <th>ascvd_hx6</th>\n",
       "      <th>bmi6</th>\n",
       "      <th>bpmeds6</th>\n",
       "      <th>chol5</th>\n",
       "      <th>dbp6</th>\n",
       "      <th>diab6</th>\n",
       "      <th>female</th>\n",
       "      <th>gluc5</th>\n",
       "      <th>hdl5</th>\n",
       "      <th>pad_hx6</th>\n",
       "      <th>sbp6</th>\n",
       "      <th>smoke6</th>\n",
       "      <th>stk_hx6</th>\n",
       "      <th>mi_hx6</th>\n",
       "      <th>trigly5</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>count</th>\n",
       "      <td>202.000000</td>\n",
       "      <td>202.000000</td>\n",
       "      <td>202.000000</td>\n",
       "      <td>202.000000</td>\n",
       "      <td>202.000000</td>\n",
       "      <td>202.000000</td>\n",
       "      <td>202.000000</td>\n",
       "      <td>202.000000</td>\n",
       "      <td>202.000000</td>\n",
       "      <td>202.000000</td>\n",
       "      <td>202.000000</td>\n",
       "      <td>202.000000</td>\n",
       "      <td>202.00000</td>\n",
       "      <td>202.000000</td>\n",
       "      <td>202.000000</td>\n",
       "      <td>202.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>mean</th>\n",
       "      <td>59.579208</td>\n",
       "      <td>0.108911</td>\n",
       "      <td>27.539670</td>\n",
       "      <td>0.247525</td>\n",
       "      <td>205.886139</td>\n",
       "      <td>73.212871</td>\n",
       "      <td>0.089109</td>\n",
       "      <td>0.564356</td>\n",
       "      <td>101.970297</td>\n",
       "      <td>52.198020</td>\n",
       "      <td>0.029703</td>\n",
       "      <td>127.371287</td>\n",
       "      <td>0.20297</td>\n",
       "      <td>0.029703</td>\n",
       "      <td>0.034653</td>\n",
       "      <td>135.232673</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>std</th>\n",
       "      <td>9.945431</td>\n",
       "      <td>0.312301</td>\n",
       "      <td>5.332181</td>\n",
       "      <td>0.432646</td>\n",
       "      <td>33.052828</td>\n",
       "      <td>10.036710</td>\n",
       "      <td>0.285609</td>\n",
       "      <td>0.497073</td>\n",
       "      <td>34.697687</td>\n",
       "      <td>13.529419</td>\n",
       "      <td>0.170188</td>\n",
       "      <td>21.214746</td>\n",
       "      <td>0.40321</td>\n",
       "      <td>0.170188</td>\n",
       "      <td>0.183355</td>\n",
       "      <td>86.938077</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>min</th>\n",
       "      <td>35.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>17.676632</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>118.000000</td>\n",
       "      <td>50.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>66.000000</td>\n",
       "      <td>22.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>88.000000</td>\n",
       "      <td>0.00000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>40.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>25%</th>\n",
       "      <td>53.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>23.929236</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>183.000000</td>\n",
       "      <td>66.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>89.000000</td>\n",
       "      <td>42.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>112.000000</td>\n",
       "      <td>0.00000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>80.250000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>50%</th>\n",
       "      <td>60.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>26.505319</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>205.500000</td>\n",
       "      <td>71.500000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>95.000000</td>\n",
       "      <td>51.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>124.500000</td>\n",
       "      <td>0.00000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>103.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>75%</th>\n",
       "      <td>67.750000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>30.300379</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>226.000000</td>\n",
       "      <td>80.750000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>100.000000</td>\n",
       "      <td>61.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>138.750000</td>\n",
       "      <td>0.00000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>163.250000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>max</th>\n",
       "      <td>80.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>49.154229</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>298.000000</td>\n",
       "      <td>105.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>315.000000</td>\n",
       "      <td>99.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>214.000000</td>\n",
       "      <td>1.00000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>525.000000</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "             age6   ascvd_hx6        bmi6     bpmeds6       chol5        dbp6  \\\n",
       "count  202.000000  202.000000  202.000000  202.000000  202.000000  202.000000   \n",
       "mean    59.579208    0.108911   27.539670    0.247525  205.886139   73.212871   \n",
       "std      9.945431    0.312301    5.332181    0.432646   33.052828   10.036710   \n",
       "min     35.000000    0.000000   17.676632    0.000000  118.000000   50.000000   \n",
       "25%     53.000000    0.000000   23.929236    0.000000  183.000000   66.000000   \n",
       "50%     60.000000    0.000000   26.505319    0.000000  205.500000   71.500000   \n",
       "75%     67.750000    0.000000   30.300379    0.000000  226.000000   80.750000   \n",
       "max     80.000000    1.000000   49.154229    1.000000  298.000000  105.000000   \n",
       "\n",
       "            diab6      female       gluc5        hdl5     pad_hx6        sbp6  \\\n",
       "count  202.000000  202.000000  202.000000  202.000000  202.000000  202.000000   \n",
       "mean     0.089109    0.564356  101.970297   52.198020    0.029703  127.371287   \n",
       "std      0.285609    0.497073   34.697687   13.529419    0.170188   21.214746   \n",
       "min      0.000000    0.000000   66.000000   22.000000    0.000000   88.000000   \n",
       "25%      0.000000    0.000000   89.000000   42.000000    0.000000  112.000000   \n",
       "50%      0.000000    1.000000   95.000000   51.000000    0.000000  124.500000   \n",
       "75%      0.000000    1.000000  100.000000   61.000000    0.000000  138.750000   \n",
       "max      1.000000    1.000000  315.000000   99.000000    1.000000  214.000000   \n",
       "\n",
       "          smoke6     stk_hx6      mi_hx6     trigly5  \n",
       "count  202.00000  202.000000  202.000000  202.000000  \n",
       "mean     0.20297    0.029703    0.034653  135.232673  \n",
       "std      0.40321    0.170188    0.183355   86.938077  \n",
       "min      0.00000    0.000000    0.000000   40.000000  \n",
       "25%      0.00000    0.000000    0.000000   80.250000  \n",
       "50%      0.00000    0.000000    0.000000  103.000000  \n",
       "75%      0.00000    0.000000    0.000000  163.250000  \n",
       "max      1.00000    1.000000    1.000000  525.000000  "
      ]
     },
     "execution_count": 125,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "c4_cov = x0_x1_conc[clusters==4]\n",
    "c4_cov[sub_covariates].describe()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# TSNE"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 126,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[t-SNE] Computing 151 nearest neighbors...\n",
      "[t-SNE] Indexed 515 samples in 0.000s...\n",
      "[t-SNE] Computed neighbors for 515 samples in 0.020s...\n",
      "[t-SNE] Computed conditional probabilities for sample 515 / 515\n",
      "[t-SNE] Mean sigma: 8.792211\n",
      "[t-SNE] KL divergence after 250 iterations with early exaggeration: 53.170387\n",
      "[t-SNE] KL divergence after 4100 iterations: 0.227719\n"
     ]
    }
   ],
   "source": [
    "perp=50\n",
    "tsne = TSNE(random_state=seed, perplexity=perp, verbose=1, n_components=2, init='pca', n_iter=5000)\n",
    "tsne_all_HR = tsne.fit_transform(all_HR)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 127,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.legend.Legend at 0x134f1ea58>"
      ]
     },
     "execution_count": 127,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAVwAAAEUCAYAAABnHY4tAAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADh0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uMy4yLjEsIGh0dHA6Ly9tYXRwbG90bGliLm9yZy+j8jraAAAgAElEQVR4nOy9d3gc13WH/d6Z7WiL3gtRCRIgQbBTFKus3ot7kyU5dhIn8ec4Shw7luPY8ed8lmVbLnFs2bEt2bJ6I0VRpCT23kECIBrRe8e2Kff7YyhQEKnKBeWQ8z4P+QC7M3PPLHZ/e+65554jpJQSGxsbG5tpR/mgDbCxsbG5VLAF18bGxuYCYQuujY2NzQXCFlwbGxubC4QtuDY2NjYXCFtwbWxsbC4QtuDa2NjYXCAcH9TAQ0MTmOaFSwFOTo5lYGD8go0XLWy7Lyy23ReWi81uRREkJsa85XkfmOCapryggvv6mP8Xse2+sNh2X1guJbvtkIKNjY3NBcIWXBsbG5sLhC24NjY2NhcIW3BtbGxsLhAf2KKZjY2NzV8SasMpXEdrEeEIekEOkXmzwe2K6hi2h2tjY3PJ49p5AN/mHXiqyvCsXYynf5CYR56BiBbVcWzBtbGxuaQRgSCuXQdx/+3HcCyqRJ1VhOuzN6EmxOI8dDyqY9khBRsbm4sbKa1wwfGTYBhoRfnoFaWgqgAonb0oGamIhLjJU4QQOBZW4Hp5N9qiuVEzxRZcGxubixr3q7twNbXiWLUQ4Xbh2H4Qva6J4K1XgSmRqoIcGkFKiRBi8jxzcBQxPhFVW2zBtbGxuWgRQyM4j9Th/to9CJ8HAGVOGeK/HkK9/9cASJ8XwhH0zbtxrF6EUBTMzl6MLftAiW7U1RZcGxubixbHqQ6UmTMmxRZAqArqkjnI1/ahVJQgG9uQLgfGnqMY2w8iYn3I4TFEahKRtOSo2mMvmtnY2Fy0SK8HOXKmyIyUEuNwHfruo+BxY+49ihwYxvnpmyAUQWSnoxRkQ3oysquX8NLqqNpjC66Njc1Fi16Uh+wdwDhUi5QS/enN6Bu24VyzGOfVyxFZaaAIRFIC7q/eiVqcC0KgFuVixvjA63nnQd4DdkjBxsbm4sXhIHD7NfDnFxDPvgKhCO5vfAHhdQOgVJQQ/v6v0R58GFGQg5KWhLp6Edqf1qNVlkXdHNvDtbGxuagxM1KZ+OQtyPEgSkXxpNgCyOFRCIQQackouRmY7d1E/vN/0ExJJIrpYK9je7g2NjYXPTIpgcjiOXjau6c8rr/wGuqK+Tg/tMx6YFkV2qt70GtbJvN0o4nt4drY2FwSRJbNxxwcwTh4Aimt4uFmTSOOZfOmHOdYWoXa0Aoy+oXRbQ/Xxsbm0uB0PFc+sxFlw3ZwWvInQ2FEjPfMcaEwOKPv3YItuDY2NpcQZkYqE5//GEpXH8I0cJxoRFm3BecnrkcoCtKUaOu2os8uhTfsOosWtuDa2NhcWgiBmZUGgJGWgvrUBszv/BJRkI081YkRH0vo1qunZWhbcG1sbC4p1JZ2nMfqEZqOVphL4PZrUHoHUPqHMOeUY2amTYt3C7bg2tjYXEK4duzHfbQOx8oFCJ8H564j6CcaCNxxrSW004wtuDY2NpcEYmwC1+7DuP/5LkR8LADKvHLkTx7BUdeEPqsEDCuu62hpR7qcaJUzMTNTo2aDnRZmY2NzSaC2daIU506KLYBQFBwLZuNoaQfDwPfYOrzHavHOLMCb5sf3xPqoFiG3PVwbG5tLAul2I0fPrm8rR8eRHjeOmpM4hMD11x9FnC7LqFbNhB/+Hq28OCr9zWwP18bG5pLAKMhGjoyj76s5s/Ghux99x2G0ijKczW04FldOii2AkpqEyElHbeuKig22h2tjY3NpoKoEbr8G71MbUDbvBq8b2T1AaO0yzLRkpNOBOR7grC0PI+PgdkbFBFtwbWxsLhnMtGRr40NHDyKiYeRkgMsSUzM+DuOVPTjmlSP8Vn8zfV8NjIwh3e63u+y7xhZcGxubSwshMHMyzn44GELJzyT8Xw+hzMhGjgVgIggl+agd3ZhR6P5gC66NjY0NIGO84FJx/8s9mE3t4HGjFOUS/vmjyBhfVMawF81sbGxsAK2iDONQLWZnH+qcUpTiPIzTLXj0oryojGF7uDY2NjaAjI8leN0a+O3TCIcKUmI6HATvuCZqtXFtD9fGxsYGwDBw7zmMkpeJ8/YrcdywCsXtwnn8ZNSGiIrgfuUrX6GsLPr9f2xsbGwuFI7aJlQhcX3+DtQ5pTgWVeL+0sdxHjyBGBmLyhjnLbjr1q3j+eefj4YtNjY2Nh8YjtZOHNWzEMqZSmEixotSkv+XsfGhp6eHb33rW5SUlHDyZPTcbhsbmwvL2BCEAxCfDK7odgaPCk3jI2zr70QCy5IzKYnzR30M0+tGDo0iNR3z5CnQdJSSfOTwKHJmdPJwhZTvr3GPlJK7776bmJgYSkpKePDBB6mrq4uKUTY2NheG4IRk63MhAmOSOL9goMdk5jwnlcuciGmqCfte+V1dDY821nF9fiECwQutTdyQX8TnZ82J6jhm7yCRB34HqorISEG4nFZ6mEPFfd/fINTzj8C+bw/34Ycfpqamhqeffpo//vGPAEQiEVyud1fgYWBgHNOMfpO2tyI1NY6+vujEYS4ktt0XlkvN7oObFFIyHVR92IWiCIITkpcfCyKcYTJmTIOhb+Kd7O4IjvOH+uM8csX1pHitvmMfLS7j4y+/wDxfMgUx8dEzRodYIXB98nrUMuvmzb5Bwj9+mMG6NszUpHe0W1EEycmxZz0++fz7tW3Dhg0MDQ2xcuVKfvGLXwBw9dXT05bCxsYm+oQCMNwPcy+zxBbAGyOYu8xFZ+NfRgLTnoEelqZn0TA6TOv4KAB+t4ercgvYNdD9Dme/NxyNrYicjEmxBat4jWPJHBw19dEZ4/2e+M1vfpPx8XEAHnvsMR5//HEefPDBqBhlY2Mz/egRK16rOqaGDrwxAi3yARn1Bgwp2TvYQ8P4MAPhEI2jw8xOSuFbC5YRMgziRZS3EUS0qd17TyN8XsRIICpDvG+Li4uLJ3/eunUrALNmzTp/i2xsbC4IvngwdOjrNEjNOpPY31Sjk5Rx4cJ9b8XznU2gCp679hZinS400+C7B3bz3QO72NPbzf1Vl0d1PGNGDubmHciRMUSCVbxG6jr63mPoKxZFZYyozBu+9KUv2QtmNjZ/QYwNQvMxaKuDSOjcxygKlC0wee3pEMd2RWit19n2QoiuUzr5sz54wX2pp5UvVc4j1mmtCzkVlb+rrOa1rnY+mVdGhicmquPJuFgiS+YRfuAPaC/vQt92gPAPf4+e5McotLf22tjYvAkpoW6foLdVkFfqIDAs2X7IoHK5SUr22cen54M3zqSjPkJXmyAhxWTRNeCMThbUeTGua6R4pk7xE1xuVCFYkZYzLWNGlsxDz8nEefwkQjfQl81HLymIWhdfW3BtbC4iBjphqFvhxju9ON2WSPR2GLz2dIjLbzVRz/GJj0+C+CUS+OC92jcy15/C+tZm7iqvnHzstc42cn1xxDqiUxD8nAjrP8Mfb9XLjWJ6nC24NjYXET2tgrIq56TYAqRlqyQkCQa7IXV6HMNp4WO5Zfzz0e30h4IsSc+kbniQJ5pOcm/Z/OkZUEq8jzyLo6sXkRCHHBuH3QcJXLMKY2ZRVIawBdfG5mJCctpDm8obt6v+XyHTG8MPq1awrrOFJ07Wk+728r3Ky8jxvXWe6/ng2rIHx8gYrq/eiZKSiAxH0B57Ce+LrzFelA/O85dLW3BtbC4i0vIk9Yc0Cmc7cLoske3vMhjuM6lcceHtMU3oaoaBDgVFkWTMkCRnvf0s3ZAmT7Y38lL3KYa1CC5FAQRxSc7TP08PrhMNOG9eg5KSCIBwu3De9iHM++pRT7ZgzCp+hyu8M7bg2thcRKRkQ3+7yfO/DZBf5iAUlLQ3GMxedu747XQiTTj8qgBToWSuC02TnNinkdZrUjzvrePFv2g8Rq8WYEaCH4nkC7Oq8LvdPNPSwD8f3c6PqlYS5zz/luVnoRuIlKk1GoTXDW4XIhKdxOS/jO0kNjY2UUEIKF8iqbzcJKJHcPk0lt1okpZ74W3pawc9onDFR7zMmOWgdK6Tqz/upa1eEBw/9zn94SDb+zv5UmU1tcOD/L9LVjIzMYkMXwx/NWsu1SnpvNTTOi32GmnJGIdqpzxmtnVBREMvP3/vFmwP18bmoiQhxfr3QRAOwKnjgp5WgcsDHY0GuSWW1Li9gqwZKoNdJtklZ5/bGhijJCGRzolx5iSn4n5Tp4XF6ZlsaWubFrv13EwcWw+AKVErijG7B9DXbUFPSwZ3dDxq28O1sbGJGuEg7HlRweVycvkNHioWuzi0NULNnjNT8uC4xPEWeb5ZnhgaR4dJ8/qoGx7EkOaU508MDZDhiU5DxzfjOtGA48NXgW6gPfMKZl0zjhtXow6NWgnOUcD2cG1sbKJGW50ge4aDhWstRU3NgtQshXW/C1I610nXKYORQZPKc2zCAMjwxlARn8wfG2pJ8/r4/sG9/PXsKmKcTja0tfBS+ykeqJqe1T8xMmYVrplZCOEI+K3tvdqfN4Bu2FkKNjY2f1mM9AvmLJkaBoiJV/DGCl74fQDTgKpV5tv2ZPyH0ip+23KC+qFB6ocHeaG1CYCyuES+OWsxKe6zC8xEAzMtmcivnkD29IPTifC6URZWIpMSwBGdJpK24F5k9PTvpbHtSYLhPpLiyykp+CixvrdwJ2xsoozbKxkdNMmaAbomOVWnM9xvMjEiKVtkklUI4h0CmR7VwReKKrmnsAJDmggEhjTxTHOahdB0lIJsHF/4MLicmLXNaP/7DJGVi+ytvTZn09KxnhNNvyEjtZJwRNI3vJ/WnZtYvfhnxMcWfNDm2VwC5JRKjmzRiEtS2L85QlyiIC1HJSVLofkoJGWYeN/lvgVVCFRheZbO97jcpJsmf247ycaeVsa0CHMSU/h0fvlbFixXegdQgmEct12BOJ3rq5YXYq6YjzIcvYL09qLZRYJhaByp/QlS6mh6kOUL72XV4m+QllLOln1/T2Pr0+h68IM20+Yix58KJfNMtr8QJrdEZfWtXmYvcrH2Di9FFU5OHrgwO95+1niExtAIP758Dc9feysrcnL5+rGd9IbOXddWjIxBWtKk2L6OkpWGMvYWOWzvA1twLxKO1P8Uh8ONz5PMtat+RFb6fLIzFnHdmp/iUF109m3itb1/i6ZF781jY3MuMgutHWblC6amUs2c76S3VURrwf8t6QsH2TXQzfeWrKAw3k+cy8UdRWVcm1fIC13N5zzHzEhBtnYhA1NrWeo1DRgZaVGzzRbci4BgqJ/27o0Uz7iGGbmrEW8IkqmKk4KclRTmrSElqZiG1ic+QEsvfrTwW9efvZRQBJjGVGU1jXeO30aDjsA4xfF+fG+qKFadmkZr4NwOR8Dn4VhBGmM//yNGXTNmZy+RpzdhNrYTqSqPmm12DPciYHCkhozUavxxeXT1Hjjr+eHRU2SmVZOaVM6O/T+hvOgzH4CVFzfBcTixW2G411pfifFD+SKTuKR3PjcaBMZMmo+BrgmSMyWJ6VGtKvieySiQHNmpsfhDLoQQSCk5siNCRoGcdruyvDE0jA4TMvQpC21HB/rI9p5dtFxKybeP7yF9bg5/G3CgrNtCaHyCvQluqj5xIw5v9PrG2x7uRYDblcjoeAdF+VfS0bOP+uZ1SGlimgY19X9mZKyVgtxVhCIjhML9HK79MaPjLR+02RcNpgkHNilkFTi57Ys+bv8bH2VVLg5sUtDC0z9+bxu88LsgUncRE+Okbo/Kse3TP3WXEoZ6ofEwtNZO9eyL50kGunVe+F2QPZvCrPt9kJ52nZLq6a+5m+bxMT8xjW/s3k7nxDgRw+C5lkaeaW7kusyzWxEfHRlg3ND490XLyVi9lNgvf4bkr3+RZ+bmsSM8GlXbbA/3IiDZX4GUktrGp7lm5QO8tuc/2HngfnQ9RHxcLteteRBDD7P38M8pzFuD0xnDln1/z6LKfyMteZpqi15C9HeAxyeoXHImZllc6aS71aCzySQ/ejPSszB0OL5TYfWtHlIyrRX9isWSDX8M0nPKIKNgesaVJhzbIRgbsDpLTIxKdjxrMGeFSVKG1TFi4dUmg10wPmJQOBeSMy+c1/2lkrk8fKqOT29ez4QWYY4/hW/OXkzmOTzcpokRFqSmo7zBOCEES9IzaRodYUVq9NIqbcG9CBBCYWnVd9l79D84UvsITkcchmmgqB4ikTG27f0+vQPHmFVyG0vm/QNCCNJTKtix/0esXfIQ4oOce/4fZnwYelphpA8SU86eLCalKQz2C6azk8JwL8QnikmxBasLb2mVk9aTJhkF0zN2dwuEJxSu/6x3sutv9ymD7etDLL/FRFEscU3Osv5daFyKyp0zZvHZgnIkTBHTN5Ph8bGto4FfnThC+8Q4JQmJXJ9fyPGhASpjkqNqly24Fwkx3kxWLfopYxNt6PoECXHF7Dj0NZISsmjpeI2rV95PdsaZzqN5WcvZvOMbhCODeNzRfVNdCjQfg9YTCjNmOYiJlbQ3GSw0JIpqfbCllLQ3GmQUTvMUWlghjTdjmtO7QNXbqjBzvmtKi/WMfBWPTzDSD4nRW9g/L4QQ56rHPoVkl4eW8VHKg0EWpKazp7eb39fVAPA3BXOiao8tuBcZcTFWHT5NG2dg+Cg3rP0h/UO1vLkNQEQbxzR1VHV6tklezEyMWGJ73ae9eGMVpJRsfjzE5idDVF3mQnXAiQMa4ZBJenSavb4liWlQMybpajHILLC83EhYUrtfe9uas+eNsL5U3oyU52w48YESMnQ6gxMkuz0knKM75q+aa/jHuQu5ocBqo3N9fhE/PXaQxsEhfI7oSqQtuBcp8vQ0VgiVmUU3s+fwg1yb9CBuVxymqbPr4I/JTL0Mp2N6Ki9daHTNmsJeiCLbvW2QX+bAG2u5kEIIVt3iYeOjIbY8F0JRITVHUr1GMtJvpUP502Bi1AoBuL1WbzElCtvzFRUqLjfZvi5ESpaCN0bQ1mCQkS/P2aUXzhS+Otcse2wQRgfBG8vbZjqk5ZnU7tfIK3HgcFoHdTbrhIOS+HOUhTRN63VwOK3xB7us/mseT5iENEiahviulJLH2xt4sr2BFK+PvmCApSmZfLGoEtfpF//UxCh1o0Nckzd1Me22wlI+27I+ugZhC+5Fi8sZR1J8GScanmR26R0Mj7bwyDM3kJhQyPDoKfxxJSys/LcP2szzJjAKtXsVhnqs35MzJTMXSTxnr41EDXGOabzqEPhTBCk5JvmzYHTAKlPodAscDsnooPVlkF2kMtgpqd9vUr3WJCbh/O1JSoeb7vJRc3AcPQLzr4CYBEs8TRPik0FRrC+lhkOCrkaBrkNKFpRUm8T6LTE8uk0wOiBIz1VprzORSOatMTlXrZiMAhjoNHnuN4HJRbOeNoO5K6347esYBpw8IOhsFEjTsssTIwmMKJRUORAI6vapJGealC2Mrkf+Sm87Wwc6+f0V15Lpi2Vc0/j2vp081HycLxRZnYB/f6oWVShMaBoJ7jPe70gkjHcavr1twb2ImTvzH9h24Ct09OwjJbGMuJhcxsf7WDL3OyT7Z3/Q5p034SDs3aBQMsfB2tusDIHjezX2v6yx9IapH/xokpYHe9brzFroJM5vDTI2ZNJ60mDJdZZ4HXpFYcEaN/llDk4e1mg6rrP2ds+kN1h/KMKx7RqLrz1HAPZ94HQLsk83lh0bhJ3PKSBBdQrCAUn5UpP2eoHP5+C6z7hwewUNRzX2b9RYcr1JRwMoQuWmuz2oqpU3e2hrhNo9GnNXni2EQkDFZZKRfslgV4S4ZCiutrIT3siJXQIFlRs+68YbK+hoNNi+LszqW92k5VheZlGlg+d/EyCzSBIfxbzl57ta+LvKajJPN52MdTr54uy5fHrzOub5U6n2p7JvsIel6Vn85NhBvla9GEUIIobBj44cYPU0tDi2BfciJj62gA8t+x1t3ZsYHeumKPfDJMaX0da9ifbul0nyV5KdtgJF+b/3NmithYZDCh6voP6wzuiQZMmVbuYsc9HdqtPXBun50zO2Lw6KqkzW/yFITpElGu2NBiXVVmGW3laIT1LIL7Ne11N1OhWLnZNiC1A8x8nRnRqBMet60cIw4OBmhXkrXMyY5UAIQW+7wStPhXA44Yrb3JMLezOrXYz0m3ScNOluUbjsWhfq6eeEEFQscfHkL3R0TfKmTVuTvLmzhDShpQbaTyqEg5ZnfesXPJMNLXOKHcxebNJUo08KbkeTgaIK9r+k4E+DGZUm/tTzex36wkHaAqMcGeijbWKMlZk5bO3q4BfHD1GemMyTnY38vPEoUkr29/WQ7PVy64ZnKElI5EB/D6aUfL1s4fkZcQ7+733SbN4TTkcMhTk3AtA3eJBX9nyBorwrSE0poKn1GRpbn2B59f+Hw/HuF8/GA+0crf85w2MnkVLHHz+TWUWfwx8Xnb5P70RvG7TVKlzzSS/xiQq6Jtm3OcLujWGWX+8hOUMlOG5Mqw25pZCabdLbZnmoS69nMoyhRcAbc0ZcDYNJIXsdIaz4qxllM3tbIS5RoXD2GYVMy1FJyVBOjznVjtQclZY6HdNgUhRf5/UZtXwPTnjdPkFoXGXVLS5CAcmhLZGzrpuSqdJ9yuoAUX9Io/aAxqIrXPhTFTqbDQ6+EmHeGvO8WgR978ReQNATDNAfCvKLY4dwqSq/W3MtWTGWx/tKRyv37dvBnTMr+FTpLGqHB2mfGGN1di4/OnIAdRrSJW3BvUSQ0uTA8f9i7bJvk5d9GQBzZn6cDVu+SmPbU5TN+Pg7nC8ZGq2lf+goxxsfwuXwsrT6H0hJKqe1cztb932Z5fN/QGJ86bTfS0e9QtVyN/GJ1nTe4RTMX+3i6V8GCE6YdJ0yKK56d9eSJrSfhO5mBcOAvJIIqQXwbprCemIgb+bZjydlQP1+g1BA4vEJcopU6g5qpOcpkznPHU0GQiEqMdzXaauH+n0KOUVnC4U/VaGpRsd8Q+oaQE+bwdiQFZ6pPaCx+ENnYgItJ3RiE88OE7wVkRB0NQluuseD2yMIByUTo3LydRjuNxnqM+ls0nF7oe2kxuHtEdxeQetJgzi/QnGlE9OQNB2J4E+TjA4IFNUq+/huU826ghO0BsZ55IrrJsXVkCZZMbGTvwOszs7jVyeOkuT2IISgPDGZ8sRkpJTcf3g/o1oEv+td3vy7xBbcS4SxiVZAkpu1bPIxIRQqyj7Ma7u+Q1HuzTjeImNheLSRHYfutWoEeNNRBFy54vtkplUDkOQvQlVd1DX/jiVz/2Pa7yUcglj/VFFxugQuj2DXhjCKKknKfOfrhALW9NvpVKi63InTKag/rNG8TiElV+KLk2QUvL346pq1QOZwQVyi5bl6YyEtT/LcQwGS0hUMXTLUL3nhf4PMmOVgdNCkvdFaYIqWE9XTZtByTGHFjW52vhhBC0ucbuvipinp6zCJT1LY+nyY6hVnYritdQaFFQ5mVrt49ckQm58IkV2oMthr0N5oUL3m3O5tJGRtfjA0SMwAbwyMj0JcksDtscZ1ewVFlU5eeTKIyy0YGZCkZCn0d5loYUlve4SFa90kJFmZFS/9KcSVH/OQme/g0FYNLawwPiKJT1Q4tNkkLklStUpyMjTIzxuOMKiFcQqFy5IzubNwFurpxOPt/Z1cmZs/RVyFECS6z66J4He7aRufWu92IBTClDLqKWFgC+4lg6I4McwIUpoIcSYfSddDSAy2H/wnViz40ZTnAFo7X+ZQ7f3MnfUp5lfcQzgywiPP3EhG6rwpx83IWcWBYw9NeayrbwetnS+i6ROkJi2gMPemqKSh+VMlrfX6lN1VQ70GwQlJcpbJzCXvXCAlOAE7n1UQClz7V2cWs1KyFDY+GsIIC8YjsOOwQfUVJnGJljc80GXl4cYkWNdoOKgQnygIBSSqE+asMOlsFAx0CgorVFrrTFJzFOavdtDdatJwTMM0JEuul0SzF2LDUY3Zi5xkzXCQV2qw8dEQ5QucOJxQd0jD5RbMnK9yeJtm1V2QkJAsiU0QLFxjCdE1n/LSckLn+N4ILp/JwqsgJt4Ke4wNWSldMQnWVuaj2xQyCxRCAWg6ak6muJmGJBI2cbmtUI+U0hLaTMFNd1u70qSU7Hk5QnBcUjDTkqDENBWkteiZnqvidAsURXDTXV48PoGhS3a9FGbnznF+4trFsvQs/qV4JkFd57+PH+YfDm4h0eUhYhq4hEJBop/uwAR+txuP6mBpeha/ravhpoJiHKdXU7sC4xwb7KcnMMGNM4rIjoljJBLmvn07WJWaPZk6Fk3et+BKKfn+97/PU089hZSSG264ga997Wso07U0bHNexPqy8bpTOX7ycSrKPgKApgc5WPMb5lfcxfGTT9HVt4ustMsmz+kZ2Muhuh+iKCrVs+9CCIHTEYMQChOBHmJjMiaPHRxpxOWIm0yG37b3+5zq2Mb8yrvwepI4fvIpXt29gcsX/BCPO/G87qVgtmTPizpSQm6xg9EhkyM7IsxcZJL9LsLIugb7Nlg5q4lpypTFrPERiarCyIBJ9gyVmfOdnNilUbXa5OBmBUURpGapNB0xmBiRrL7NQ2qWatWy2K9xYJOGqcN1n/HR02Yw0K1x2bVuhBAkZ0D5AifPPRSkZruJlIL4ZEle+fmLbzhg9Q4DWLDGRWu9QcsJnYFug/yZDpwu2L1Ro2SugxlOwcnDGoYhyci1JCAwZjLcb5KarVIYdFKzW2P3OojzSwJjAk+MIBKSuDxWZbRVN3sYGzY5sU/j6k948acojA2bbHkmxIZHQqy6xc2ejRFcHuvceSvP7EoTQjDvchdP/TKAocvJx7OLVXa8EKa1Xsflhqrlbjw+6znVIVi4xs23f99MYbGf/1i0fDI8U5WSxg3rn+T6wmJK4v388PA+alqb2NLVTsQwuHlGMTfmFdExPsbnXnmRm2eUMBwJ8efGeoriE2gZG+XTm9eT5PbQEwjgdajcPjP6C2ZwHoK7bt06HnroIe699140TeP++++noqKCm2++OZr22USRBRX/ypZ9f0dd0+vBtXEAACAASURBVHMk+Yto69pJXtZlzCy6mYlAH0OjtVMEt7H1cdKTKxmb6EQ5/W2vqk7Ki2/llV33sXbZd/B5kxkaaWbLnu8yEejjld1fQDcCBEI9fOKm5whFRnl15330DR5HoPDS9o+TnrKY6vJ/xOl8l71W3oQnBhZdbXLqhMbezTpun2TWUknyuwgjgJXhkJSuUlrl4PA2bfLxwV6DzY+HKJztYOZ8lb4OK7nfMKB2ryAt28HCtWfKDe55OULjMZ3ULBUhBDPnW5kHxXMceHyCwR6TrBnqlFoViiLIyFMJhwSlVS7aG3T2rNdZeNW7bz1zLtJyFU7V6uQUWZkJ+WUOktKtbrm5xQqvPBnBFytoqzcoKHdw1cc8PP/bIF0tOns3QUutTlK6wki/RGJlJXh8gpFB+NBH3KRkODBNye6NYQZ7TNJyVPZuCrNwrRv/6ToScX6F5dd72PDHIC/8LoTTJbjl8x6e/h8rpPBGXt8AYZrwuh85MmASCkmkBCkFvrg3hY3c0O8b4orc/CmvqVtVWZ6RQ6zDyVA4hAn8dvU1FMYn0BsM8K97tvFEUz0pHg/1I0P8b90xlmdm84OlK5mdlML27g7+bc82uibGWZqRzcH+HjKmKZH7fbujGRkZfPnLX+Zzn/scn/jEJwDo6OiImmE20ScuJpeK0r/GMHXSU+dy/Zqfs2rJNxFCYWC4Aa9nai7ORLCbvKxlTAR76Rs8Mfn4gjlfIBDo4+Gnr+N3T17FMxvvZu7MT/LxG58mGO4hJbGYzLRqhFB5buM9lMy4hrs+sp1P3/YSJTOuY2j0OHuOfvu87sUTA2ULJIuuMZm78t2LLcBAp0LZPCcZeSq6JqnZHcEwrLzTuZe5qF7pJrvQQdXlLqoudyGAvjZB5VLn5AddCMGcpU5O1emT1xVCIDmzKSImXjDcd3YMdGTAZEa5Nf6CNW4KZzloqTm/YG7pXCf93QbbXwjR0axz8rDGxkdDuH3w6lMRcopVll3npnqVi85mnT0vRyie6yAUgv4uk5vu9rH2di833+MlM18lKUOhepWbOL9g3yaNA69F6D5lkFeiTm7dHR+V+FOnSkh8ksDQYN5yJ1kF1owga4ZKwxF9ynHNJ3RcnjOv1VCvwaGtEWYvNcktkwghaT6hTTmn+5SBX/hoGR056/5bx0dJ9Xr5U0Mt/8/cBRTGW6uRaV4f31qwDENCiT+JOcmpfLVqEV+tWsTsJCsNYll6FvEuN39XWc3+vh7K45JIPke8Nxq8bw93/vz5zJ9vlfb72c9+BsCyZcve7hSbvwBy0lZwouFX6FqQhPg8TFOnruk5unsPUVnylSnHJsaXoRsRPK4Ent/0ReaUf4qE2BzqW9YxNtGF0xnLTR/6FbG+dFTVWllaOOevaGnfwsREN7WNz5KTuYTKso8B4FDdXL7wn+ns2cvg6AnGJlqJi5nmYgPnQHWAFpYoimD1bR52bQhzfJ+GFobLb5j6kZhR7mD3SxErpUqZKoqKKqakTE2Mmgig+bjOzPlOCmY6OLZL4+RhjaIKB/J0jDIUlGQXnokP5pc52PaCzvlUFXN5BAuvMmmrMzm6y8DQJC63oKzaScsJg2XXuCe/LFIyPTz76yBCUVBVqF7lmvRAFVUwf7Wbp38ZoHCWSXBcklui4vLAwS0RYuIFY8OSvg6D5HQrjWtG+ZnXrLvVJNZvZRb0dRlIKZmz1MnGR0OMj5hk5qv0d5m01FqZCk/9dwCH0yozmV1ikppj7RYcHYDa/TrhAOQUqwz2mtQd0PjEFUV8pX4DS9IzWZWViyElj5w8wUgkzOK0TL69fxeFcfEc7O+hZWyUwvgEKhNTMKXJV+Yu4Fv7dqKbJoOhEI821nKov5dEt4eQYbAgLZObAgGCwekrYizkuSpQvAcefPBBfvKTn/CRj3yEf//3f4+WXTbTyPBoK6/t/jYDw/UIFPzxeaxY9HWSE0umHDc43MgzG+8iI3Uew6MtjE90oygOXK5YstMX0j9Ux4eve3TKOc1tr3L85OMEQv04VR8lM65hdukdU47ZvOMbdPUeJC2lgkVz/prBkUbiYrJISSqb9nsHSxBr92us/bBnUmgObwtzYr/ONZ/0kpCsMDFqUrNHo/uUQTgocbggf6ZK9Yozns/BrWFa6wyqV7oITpgc36vjjRF4fNDbYZJT5CAUMOnrNDH007m3Clz5MQ/+lDOC21qvc+C1CLklKoWznSSmnv86iBaxGjn2dZlULnExs3rqzoUd60O0nTRwugVrbnOjha1p/use66M/DiAErL3dQ0qWZathSF7+c4jBbhOHE5IyVAa7rftPy1UZ6DbZ+3IYXYfYeMHEmBWaSMtRKJztoOmYQVergZRwxR1W/d5ISNLZrLPzxQhOD2QVqMQlKvS0GkTC0vKAVclhrYO7r5lBeoaTw/29/MvurURME1NaC6R3FJZyz6w5fHn7K3QHAihCUJmUwuGBPnwOB23jY7x8wx08d6qJJ5vqLYFOz+SK7Hw6Jsb5xfHD3F1eiQCODw3y9flLzvtvcC7OS3B/+ctf8oMf/IBbb72V73znO+9pwWxgYBzTnP7q76+TmhpHX1/02h1fKKbT7lB4EInE+4byjFJKBoaPMjB8jNGxJrr6d5CbtQxdD9DRs5/8zKuJ8WXS2fsaYxOtzJ31aebNuhNFsRaO1r/692SkVlFccDXPbryHJH8R167+8eT1TVPnkWduIhQZweWMR9PGyEidx9BII15PBkvm/jsu57lbWUcLKa0E/e5mq27A2JCJpkmSMyVayJrmb3osRH6ZyoxyBxNjksPbIgTGJf4UhfQ8hfYGg4lRSVaxSWeDQnqOSmmVE7dPsOmxIHOWuZBSEglD9ymdmHiFBavd7NkURlFh8RVuVIdgfNTk5UdDZBaoeGME9Yc0SuebZBa+t3t6/X1imtBwUNBeL07HQiExVbD2du9kmhjAM7+awDBMQBAJCeKTBFrYSq8rqlQ5sU/H4ZJc/5mpsczmEzrHdoWZt9akp8XKXpgYVQiMWl8oQsDVn/QQE2eJ6bYXQlZ2R4+JJwa8MZLkdCfVK6fmt7785yBZM1RaanUrXLPMiaIIag9EOBBpY0PMYUa0CNkxsazMzOHVjlZ+sGw1cS4XrWOjfHXXFrwOB5phsDwzh2/MX4IQAlNKvrF3O1s723jh2tvwOR18ZtN6yvxJfGPB0snxT42Ncs9rLzEjLp41yTmsSc99V6/3m1EUQXLyWwfj1fvuu+++t73yW7B3717+6Z/+icLCQu6++246OjowDAO/3/+uzg8GI9PeAuSNxMS4CQQiF27AKDGddjsc3ilpWqapsfvIv9HatR5/QgYTwXYi2jirFn+DueWfJCdjMfuO/hin08H8yjvJz15BTf2fqW9+AZczlh0H7qen/xij420kxOWSk7mYYyf/TCQyjj+hgPFAL1v3fJfxiS5SEstwOb3cfu0jlBffREXZR+kfOkZt06PkZ107rUXRhYCUbCtXVgiT1FxJabUkOQsmBh3s3RQht9jBwrVuvDEK8UkKuSUqdQd1ghOS/g4Tb6wgNl6hpxVcbpi9yEVWocP6wuqWNB3XaW+wvNu8MgfzV7lxugSZ+SpNNTr7NkdoOaFzbJdG+QIn81a4Sc9VySxwsHejQW6ZfE+1IF5/n9TvEwTHrMwLX5xCUaWDSAgObo2QWaDicgt2rg8z0CPJK3WSle8gOC6JiVP40EfcCAGHtmrMXuxgpA9K5k71jAe6DQLjOtnFkJBqtUXvbLRyoDPzHETCkrYGk9wSB26PIC1H5cgOjdW3ummuMUhIg9g4x+S23tfpbDFwOGGgS3Ltp7wkJKvE+RVa4rp4eOAw/7H4Mr4xfykVSSn8praGgvh4PlZSTozTSVZMLEcH+sjw+jg1PsoPlq3C53Se/lsLyv1JPNXcwN6+bgrj/ezr6+ajJeVkvyFP1+9282xLIxHd4PNFFZM5ve/0ep/93hL4fG+duP2+Y7gPP/wwUkqampr43Oc+B8DnPvc57r333vd7SZsPmKa2ZxBC4yM3PIaqWG/Y2sZneGXXN7nt6oeZCPQQH5fDNavun+wMnJO5mD88dQ17j/ychLg8VMVFRJtg/7FfoeshUpNmcbLlRY7W/RFFqDidscyd/RnqGp9j+YKv4nJaHpSiqMybfSd/evYWak7+morSu6f9fn1xU+sYCGDJVW6G+nVyS6cKgjdGwZ+sMNBjsuY2D+m51vPD/SYv/SnI7pfDqC7YszFCXonKkiu9BIMmLz0SZuY85+TWXqdbMGuBi93DIYQiWfQhF4WzzohaYqpCbIJVxDspg/eErkFHo6BwtooWhqVXuya/uOoPRdj4aMhyciSsvcNKZwMrVe3lx0K01BqUVrloOGqQkKwSCRu0NejkFlsyEQlLTuzXKKw84ymdPCBIz3WwcM3U7I3D2yIsusKNL05gGOBPUVEdEJcoaayxYtyvp+MFxky6mg3i5zvJLlQJTkj6Og28PsHvm47zLwsWsiDNejEqklL43pLL+dutm5BSTt7ffQuX8amX1xExzCmNIwG8DgcSWORP59v7djIYCdExMc4bE78002A4HOJ7lctwTkP+7eu8b8F94IEHeOCBB6Jpi80HTEfvqyye94VJsQUoK7yBPYd/yuh4O919hygpuHpKG3anw8uM3DWkJpUzu/QOBkeaeerFT5GWPJu87OX0D9YxOHwSJCyc+0XmlFsZLTV1j+L1WPm4UprsPvQTjp98koT4PBrbHycQ6qB69r041DMx01AA2moF40MCT6wkt0wS++4mVO8Jj08y3GeS/YYSqYYhGRkwSUoXk2IL4E9RmFHuYHgwwrbnwiSmKcxbYU2XfXEKxZUGuzZYNR6cbkFgzGT/q2Hyyk36OwTm1MV7pJQEJ95bXV/DgOC4Sf0+gSLgVK1O8ZypFyiudHJwq8bCtS5q92mTYgvWQllplZNTtTpFFU5iEwSRECy/3s0rT4SoTdWI8wurg0WBJDX3dVuhu0Vw8z2uKdkbFYudrPt9kEVXuOlsNohLVABJJGS12xnuM1n3+yBFFQ60iKTpmEHFUhceHzTX6Jyq1UnPUxkfljTHjlKZNDV7psyfxISuUTc8RHGCny1d7ezs7mQoEqYqMYXHm+r5TNmZaniPNdaxKDmdm7OLuDm7iENDfTxYe5h5KWnkx8WjmQYPHj1IaZyfvJjpDWfZO81sJrE8hrOnUgKrq4GiOBkcbjjr+dHxdvKzlwNQU/8oZYU3snzhP00+n5E6l617/pP9x36NqrrJz1lBYkIhxxue4rL5X6Gm/jE6e/bzsRufxutJRNODvLLzPo7U/ZyCpC/T2yYY7oOJYSsumZSu4PUo7HtJp/Jyc0pKWCQEA53A6bDBu6mJ8EZaTuj0tgn62jVSMlXSchT0COx7xQqBvV6O8Y24vVbOqMcnScmY+pGqXuli/R+CPPnfAWISBIFRSXaJSXwy9HdKju7SyC5yTBa7aTiio0feXaxtbMiqnTDYDYojSH6ZkzW3OwgHJYe3a2hha+ssnM55NWDf5shZObEAhi5RVAgFJN2tBgtWuwiHQHFAbJKO0wMLrjy79oM0zy6krjosr7b2QISa3RqLr3Sz/1UNp9taNJy1RNLbZtBzymR8WOBPVSic5aDtpIaiwo13+yZtfGR9PAf7e1mVfSamenxwgFink7tfeRGfy0mWL5YFaRnkx8VTMzJI7dgQJ4YGmJeSxoG+Xk4MDfCflWcyqKoSU7ktu5i7X91AutdHXyhIcayfL5dO3T05HbzvGO75Ysdw3x0X0u6INkZz2zpKCq6ZFN6GUxto795FKDxKXeMzDI40kJI4k4S4XEByvOEJTrW/xvIF96IoKtv2fo/LFvzjpPcKkJhQyMGaX7NkzrdobHuR/cd+iaFrdPXuZXC4gaa2TSxfeC9JfmulSFWcZKRWs//VIKMdJZiagsersPImD3MucyFNyYl9Okmp0N0syCmz3kjHdkDdHoXeNquQdtMRQUz8u/eCA2Owb6PJlR/zkparsndThJq9Gkd3aSAMKi6T1B+Aoooz02Fdk+zZFCGv3ERKGOkTk2UZwYoXnzysUzLfJNYvCU1YecBDPQqj/QJvjKBmt0Z/l8mJvRo9bQbxyQK3VxL7NhvywgGrwHn5fCfpuSqqIlh+nQdfrEJcokJBmYNdL4XJL3Pg8giO79WIhE3mrTHpahLExCkkJFt/40hYsmtDGH+KwsEtEXQd2k7qNBzRKZ5n7d5TFCtW7XjTF9hQL0SCU73+ozsjDPeZdJ0yUR3QWm/gcEiSM1WajlrNN5uOKEhTEAoAmBzZodPTZjJ/tZuk9DPXSo3x8J/791IQH0+a18eRgT6+fWAnd8+cg1NVqExK4ftLV1IU78elqIxrEXxOB7VDg4TCGnMTUvjrojlnFaEpifNzXWYBpbF+bsoq5LqsGWeFIt6OCx7Dtbn4mJF9Aw2tj/On526huOBqBocb6Oo9SHxcNk2tG5mRu5rSGdeyZc932Sbl6b5oGquXfRtVtcIQiqIS0aau3upGCFMaHGv4JXPKvkRKolVtv7VrM4dO/ACQxMVM3bkwPhCP11zJqts8vPxoiA993jspcjOrXYwNSdobdUIB6DkFHSdBC6usvtVNfJJCW4O1MHVsu4I31vIo34nuZigod+BPUfCnKOQUqUyMSvZsDJGca5ULzCmVvPhwgNIqJ6rD2iIbn2ySmA5xSbDzWYPj+yKUznVi6HBkRwSJZLBL0NsqSEgRrL3NgzfWSj3b/GSIiiVOYuIV3F5Bcobg2V8HKTpHtTMpYagHupqt6ma5JSqlVS62rwuRUzzVzXS6rW3LuzeGMTQYHjBJTJf44qDycpM9L4c5eVjDF2eFChxOGB7UKJkviUsCPQy+eGg5DtueVIhLFIyPSBLTJbOXSVQH1O4RhMatzIr+ToP0XJWuVoOxYRNfnGR8WJCSYcVkx4Yl1auctJ00cLlVbrrLg8dnhVheeyZE4RyDvjblLO/78qwcrhgK8YMD++iNTJAXF8+dZRVcm1/IA0f38/iVN7Kpo5XvHdjN8sxsCuITeLGtmeUZ2bSOjnJtZsFb/r09qoPSuPPbZv5esQXXZpKDtfeTnFhKYe5KhkdPIaWJYUQIBocwzQiXzf9H3O54Pnbj0wyNNNM7cIz9R3/Dqzvv43j945jSZDzQz+5DP+Xa1T/GobqRUrL38M/wx+Uzu+x2dh36OqsW/YxYXzZ5mWtwqG4O1d7PyZb1VFfcNWnLicNtlMzJJjAuz6p3AJCRp3KqXiczX+X4Lmtqe8Odnsl6AsWVTsJByak6nYObFZbfYr5jXFTXBL6EM+MIIYhNEHhiFfSIVbi2uEqSlCHpbokQDoApBd0tgsEuAQJ8CdB0XOPQVg1FgfR8SWqu5NRxBQEMdkue/U2QGeUOqle6WLTWzY71Ya78qIdwSLLt+TD+NInTDfUHYGxQweWRZBZK2uusDIScYpWRWGNypd8XqzAyMHW6KKVkYsTaYJGWo5Kep/DC/wYZHZQkpMDyW0z62k20MFaRmjeXiYyBzkbob1O5/rMefHFWMZo9L4ep3aOTVSQZ6hZc/1kf4nTcuL/bpL/DZOZik6YjCjfe5ZsMlTTVaGx9LoyuweIrz9RI8MUpLFzrZtsLITILTeoPTy1j2d1qUBjKJC0nyL6xXu5bcBlJnjNx/TEtwn8e2M3PLl9Lqd9qF/H5WXP4zOb1DISCb/8H/wCwBdfGaund8yrdfTu5+arfkOw/UwFm845v4nUnUtv0NC6XtaQvhEKSv+h07q3O1Zc/Su/AfgDml3+dQ3U/4g9PXUtaSgXDIy2oqpO0pNkcPPZrivI/RHP7s1SWfhGArLTLiIvJZcu+vycUHiU3ayn9gydo61aJj/8w8YkKQ33m6a4DZ8Swv8sk3q+gOq2Fq+C4JCZeIRSQHNsdobPZQAhrR1mc36p5m1/+9q9DSrakfq9+usqWNVYoIOloNFh0zZnjkjJAUSWHXlEoqnDQVKMzd7mLtGyVvg6Dg1sjVK2yYsv1BwStJyyxXXK1m9xilVBAsv+VCDtfDLNgtYtISLLuD0EcDsgoNDF12P60Qma+SsUiBxOjJoe3aPhTFa7/rAdFFfjiNLpaDErmOCmudLDhj0HSchWyClRM47RnbcL81WcWtDLzHRzfqVG20Kotm5INHQ3QeNgS9dRcyUgfTIxaoZi+dqtzhC/uTN3hBavdPPU/BqoDCiuck8XFiyqdFFVa4YmOk5L0XJWdL1o7tvJKVGbMVjm+V8PQITZh6pdnfKJVdSxvJuzfaLDp8RD5pVZRoqYaHXN2Lw+31SGE4PaXnmVhWgZfq15MdUo69x/ex7yUtEmxBUh0e7ihoIgnGuvf1fv/QmIL7iVOODLMzkP/SkQbJCOtiude/ivys5ezcvE3UBQHuZlL2Lr3+0ip0969m9zMMztw6pteABROdawnP/vayZzeBbP/hRdeu4W05ArmzbqTjNS5CCHYsf8HjIy2oRtTP3BxMXmsXvQLGtqeYN/hX+PzZDKr8iM07DEorbK82R3rwlSvduH1CZpP6DQe08guUgmMydNpVCZjwyavPBkiM1/l8hs89LQZHN0ZYWQAhvoUhnok5YvlOZsigtWlNjlT5cWHg5TMdWLokvpDGjml8qw2OC01grnLnTTXGCy6wk1eqfVRik9ScHkE+18LY5ogpCCvVMUXJ8grsY7xxgiWXOXm6f8JcPKIRlKmZO4KSWcjtBxXMHTIKVK57LoznlzLCYPZi5yTxcMLZzuo3a9xcEuY0ionc5e72Lk+PJn25YsXCFXy0h9DeHyCwgoHve06oYBg74uCmAQrRJGUplA4y8nYsMnh1zSS0xUKKxz0tRuMDxtnNct0ukFVrTEM7exFGD1iEpoQjDsk5QtPpxbu1+g6ZSAUidNjFV9/PdUMrHhxYpq1023BVSbdzSYdpwxcHtAWdPDrnsP8aPkaKpJSCOk6Pzyyn49ufJ6woaOZJvNS0s+yQ0FQFM3q7lHCFtxLnEO1D5CVPpdl87+MEAqaHmT9q3/P0bo/Mbf8k/T0H8UfV4xujrF5x9eZM/OTJPtLaO3aTn3T82SlLWA0cJxX9zzHigU/xu1KYDzYgdsdx4LKe6aMVZR/FS9t+Ufys64/yw6fN505pX89+XvPKaud+LrfBcnMUxjsM3nuoSCmCRm5CouvdLNrgyUwa+/w4PEKNj8ewp9qTVHHhk2O7Yqw9Co32UXqZDz18Ks6C68+d+FvIazc1eMHx+lqNRH/P3vvHR9Hfef/P2e272pXva56r1Zx7w1jY5sWCCUJCS0hCRd+KZfkLqTne8mlEXJpFxISOFJoMRjjjnu3ZcmyZMnqvfe2fWZ+f4y9QsgG2xiDQa/Hw48HzO7MfGZW+9r3vMvrJULWPOWC/bBjgwJR8VqKd3sn6SIA2JM17H8N7Mmimtdslol7S45VqxMICBSpKvYx60aFM0fUfKjPo1rkJGZP/mq+NR2iNwisutvAludc1Jf7kBWIihfJX6gHBHY87yQuVUtStpbxUZkTuzxotAJr7zNiCRQ4uNmNACy+eYLUYxI17H/NTWKGluRsHbZQtZ/2zeTY0yaj0YE9XaF0l4/UfB0B59I4fZ0S3W3qD9oNdxknZBeTNGx+1oljFGxhCke2uhlfKBMeo6G7VebMcQ8Fy1Vm12jAngr2VAWfLPP946d5MCuP3HNCM0atln8vmMUbbc2E6Ix8KjObX5eX0DAy7BesGfG4ebWxlq+lF0394N5nTBPuRxhe3zhdfce4ccl2f1eCTmti9owvcODET9DrLNQ2bSMp9lY0GhdZqbdRUfMi7V3HCQ/NIiv1DkRBZG7hl9h79EfUNr9Abtrn0GttuD0jeH1OdG/yShsd78AneUg657F2ISgKNJwWaK4SVJJRQJJVm5r6conwGAFRI3BosxtBVAcFQiJEgsP19LQ7sSepxFZf7iM5R0fsObLQaBUCw0TqK2DviyJBYZA0Y6pZoSAIRMRBRNzbt9CYrdDfJWOxqtYxoVEThDrUp06imQJEQCA4XKSrRcKePPF1c7sUhvpkkvNlqo6Bc0wgMlZkoFsmMl5tH3szYlM1VBz1Ep2g8Ue5vR0KJovAus+YeP0ZJ/ZkLUaLyKHXXdiTNMxbfb4yryHCrmHrc06MFlXY2zWmkDtfN8lyJyxag06vFthCIjSkzdBRss/LiV0u4lJ1DPZJnDnmJWuujC0EknJltjzrJDpJg+RV6GmXiYiTCQqZ0L5V7/05m6FTPlJy9Cgy1JT5KD/iJSpBw8xVqsD7W1E+3IdeoyHROrk3VidqiLYE0DI2QlF4JN8smMMj+3ew0h6PWatjU1M9i8JiyAm8hErpNca0WvhHGJLkQRDESaQIYDQEMjrWQVXdZhYW/ZzosHm0dBwkODCZZfO+w/qVv2dO/qN0dJ8gMnwGAFmpt9LTf+Lc/sHERs7hyMlfIUlq68zoWAdHS/+HrOQH31aAvLkShrpF1n/GxJ1fsLD2MyacYwoBNg2L1hno61IQtF5soQooKrlt/6eT6hLVCLG/W42UxkdlgsImvvTVpT4qj3tZtN7IzQ+YSM7Rc2qPyEj/ld27hGyZkn1uYpI0HNvpZnRIPe/YsMzhrW6yZ+lISNdSe9pLUpaGprMSZ457cIzK9HZI7N3gxGBSaDwtEhyuY8FNqm243giKrHDmuNd/TFlWcI7KjA3LbHrGycm9bvZtdHFyj4d5q/XUV/hwOxVO7Paw4Q8O+rtlErMmx1IBgWq72GCvjMuh4BhX2Peqmxf+x8GeDS5GBmXVrcGFv1PA5VTHixsqJY7tctHVqkaiEecE3qKTIWehjNbkxe2WEDXQ1y7SVq/6ub0ZwwMKOXN0ZBTqyZyp55YHzMSlaggMFi9ItgAuScKo0bK/o23S9h6ng9axEYyihl1tzSy3x/PcirXEWqy4fD5E4JGU3Cv78Lm6vAAAIABJREFUYN9jTEe4H2EY9EEEmO3Ut7xBasKN/u2Vta8QF72KwqyvAmpRLTAgjc27H6Uw535EUUt59T/R6czEx6gDD07X4CRPtKVzv8O2fd/iuVduIsAczchYK+mJnyQl/ta3XVPLWZEVHzP4uw0CbCJzVunZ+YILjQghUTLdzSIzFmhJzNQx3C+z/zU3deVe4lI11JX7iIrXEBop0t4gkZyjQ5JUzdsb7jL5e09T81SvscYKD/lLL78hPDQGMmbJ1J3y4RyDzc860erV4QJFgehEVYMhNlnD7g1uouJE6srVqE5vUJ19RY1AYpZ2knFjZLyW7f9wEpsmsvVvTgICBcZHVEeI2TfJOEeh5axEf4capR/crBbH7EkaXA5VRD3wnPvCmyFLCuMjMkYT7NngIjZZQ94CPRqN2ie860UXiVkabMFqqkOW1MKeRqv24MpesASpHQ6KogrktFYLGC3gGoeYJA3zV+sRNaoE5ba/O1l/v5pWaKuX6G6VWLh2ci9sYqaO8qMeoi/iO5oXGEq/y8mR7g5+dfokq2MT6XSO89vyEtICg2kYGebZ6kqera5khT2O+ZF2Xq6v4b7ErHfUQni/ME24H2EIgkB+xmMcOP44XT1lhIVk0NJ+iJ7+SpbM/s2k983O+w4NbRs5duopxp0dhNiSuGnZrxFFDW7PKMWnnyIhZoJMDQYbCwp/wrizE5e7H1tAEjrt26voK4r65T1PiucheRWQISJBJCBQy/iQRF+HQnoBOMcVZAlS87TEpmqxJ2s5sdvD6KCMLMG//jCOJKmk8dbjRidoqSr2cqU6tCHREOuQGRsUMAUohNrBGgTNVXBoi5v5a4wULtVjDVHlF6MSZWbeACW7RCw2EY97wtPrPILDRcxWgaEeGVlWiTJ9lkzMOfUwk0XtkvB5ZYZ61UmzeWtU40hbsED+Qj3WYDVHGxGrJThc/WEp2ac+aXS1SMiyMql7IWuWju5WiZpTPjRa2PWyk4FumeAIkY89YmZ8RGH/ay4ay0USs2RqSqC7WSA5R0tgmEBNqY9F6yf0duesNLC9x8mrf3ai0+O//9rJOjiMj8gYTJOT6T6v2jkx2CWiMxh4ILiQ54bLqBzoY39HGw6fF1lRGPN5+eWCpRSERjDgdvHDk0f46alj3G5PYXVUwhV9ntcC05NmH3C81+s2GyOIjVzBwHA9vf1nsQVkUJj1FQz6yeNZiiKj0waQEHMTibHraWrfRkXNP2ntPMKRkl8RFbaA9MRP+L9059et11kxGyPQiO88YysI0N+h5j7Pk6OiKOx83s3cGw3kLzQQk6glPV9LTZmP1lqJ+nIfeoM68nv6kAdZgiW36hkfVR/FF6w1kjVTS02pj+Rs7SSJwo4mieF+iZiUiT9Es1lPe5OH0YFzFfmLhCSucTi+VUSn0xKdqMMxLFBbAuGxChFx4PUqnNovceqgl+F+ifSZMkm54BiBnmYBW6iIz6dgCxEJDp/I/0o+hdOHvSgK5C+VSStULvjILWrUPLLHpdDfqUbBXi9kFOkoP+LD44La017qTvs4fdiL26lgCYS6cpmoeA3x6ZPZzzmuoDH4SCtSqCmG8BhVrMdkFQgO1xAUplr4eD3q0EXaDC0arcCZc6PJsSmTb5TboaAgkVqoYAtVGOoVaKryYbYKWIPUIYqjO9xEJWqpPKZw9rhAd7MaNWsQSczQgSIyfiaApdHRHPY00etyYNbqQIAvZBew3B6vTnZpdSyKsvNcTSWfTc4l8FJ93d8FpifNpnHFMJsiyUl96KKvt3a+QXntH9BqjLg9w4SHFLKg8GeMO9txunrJSX0MszHiqqwlJV/m+E43LodMhF1LU5VKPnFpE6Sk0ar+YUe3u0mboVrhCIKAz6uw+18uqopVAZTbPmv2R1DphToObHKzYK2BgECB7laZk3vcxKRNPHo7RmDzZieKorZxnTkkk5irEuVbUXdKjfAKFp37cufrqC71UFPspegGmcRsSMiSp2gN+HzqeGx4jIb+TonyI14iYjUE2ERkWaH0gAdrkNoedSkKlfFZULxd8hepKo97QFEdchUZhgdk6iu8NFZKuBzqj1pvh+xX2lIUhbY6iepTqtZBd7OILQTi07W4nAp7X3EzY4GO+HQtkqSOUq+/3+wfXLAGi1Qe90xS7gLo7ZAIDFdoPasOa+TN0yFLqpoaqD8OkfEy9eU+5q5SpSmPv+Gmp12iq1Whs8WLVqda8AxXBjAY5ybEYOQLOQX8tbqC9KDJv0JWvZ4QgxHpWkZxV4Bpwp0GACPjzbR37UVBwR6xmEBrCgD9QxWU1/6Bm5Y9SURoNl6fk8PFT3Co9OskxNyExzNC70AJZlMU8dGr37Ujb2gM5C+TaT7joarYi96kIGqnMo/PoyD7IG/+xKOxVicwY76e/ZtcmK3CpMfV/IU6jm6X2fysOn1ksQpoNBByroVTUeD0fpH0fB3phaoRo2NUlV60hchYAqG5UmCoV0BvhKEemLtycpSYmqfj5F4vsqSSrCDAedd5x6hqL97VKDDcD1UnvPi8ahS++ZlzudpRtd+3cMWlkS2o4jyz18gMthuoOOals1lh4VqDagckQkejREejTNFSPUazQH25l94OmUNb3OTNU00vB3tkcuboUBR1XRFxGr8ObmKGlm1/dyIAeqNqB3SebAESMzWc2q+mLPLmqTncsyVeultl+rpEdHqBdZ+eGMtOzNLy2tMOim6QOXtMZO4qPfZzOsJdzRL2FC2Fi/Vo9dBcLXFil5vQZAmPR+aPS28kymyhpK+HI92dfk8ygI7xMQY9LqKupvf8e4Bpwv0IYdzRSUfvIQRBICZiMWZjBD6fg1Nnf0NHzz7Sk9ah0xo5VPoNkuy3kJXyGRraNlKU8wARodkAOF0DdPQUI4oaRsYraO86TpAtAY3Wx64jD7Cg6GcE2y5SBblEBIVD0DK1g19R4NBGtfByvh9UklRdVkG8QH+qSc0XuhzKpOk0URSwhYgkZqoeXi01Pk4dUAtYRzcLOEYEtDr8ZAvq2Gn2HD3NNW5G+gUSM7VkrtIyOqxwfKcb71ueKL0e9dy8iSwVBepPCbTVqgIvOp2CwSiz+BY1Mt77igsFtYqfmi+TlHf590urg+zZekLj3Rx9XfQPK7gcClUnvKy734Q5QE3R2JM17N7gwu1U2PmCC0GAWx82++9TUpaWTX91kpYnERyh2t0EhYuc3OMhKllGmuzrqDocRInUnfZRXeIDAaITNNx0n5Hjb3gIj5k8lm0wCsSlaRjukRkbwj+ePNAjgyAw54aJH9CkLC29HRIH6vuIT7ISZVZrAJ9Ky+Jz+3di1GhYbo+jeXSE35SXohEERryeyxKhudb44K5sGpcFRZHp7D1CV+9hRFGLPXIFTncPbV27URQvWm0AvQMlJMWtQJZ97DryV+KiV9PauZ3gwCSiwmdQ17yNJXMepyD707z4+t3ERCzG5e4jKDDRf549R75PVurtFGR/GgCfz8Xru79IZFge9qjZlFU9ybI5v79q1yUIkLtQ5uh2N01VPgICBVrrJEwBMqJmMhED1JzykZCpxeNSOLTZzewb1MiuvV7tIbUGC2z+Pxcep0LmXJninargd/4iHU1VvilOE0azSsYJ6aprA0BYDPS0SZTsc7P0NiOiqD6alx3yEJU42amhvwN6WkVuedCEqFE1BzoaJfZscHHbZ03MWKCjsUoie7aW0v0eEnMvPbp9K0QR4jNlKk94iEnU0NcpERqt8ZOtej8FkrN1tNX5yCjU4fVMHpnW6dXJuI4mlXBBze8m5KqFu6ObVfHw8zn2oT6Znja1Da2x0qe2hnVI1J0WibSLjI9NXafLoWCyqfoNfR0S0YlaxoYUQiLFKfc/LEokqsdEn9OFT5bRiiLxVhu/X7yS/3fyKM9WnyHeauPTGTk8U1XOqM/L1UluvTeYJtzrHEOjdfQNlNHWswdJGic3/U68PhfHTn8HizmCopwHcbmHOVb2G+5a96JflWtopIWXttzNqkU/JTF2CQB9gzVseuMRPr7uedKSbqKj5wDBtmwaW/cSFz2fMUc3g8P1rF85QaharZGi3IcoqXiaW274EweLf4bbMwRYL7TcK0JQOCy8VaarScbtgozZ6hhuX7vC4S1ukrIlQqNEmqt99HXI3PKQGZ0etjznZNNfnCiK2qEw/yY9jhGF2jIvHgHOHFZda1fcYUKrOycr2CcTFDZRsKs95cXlgOCICdLq65RUc8h2mVf+6CDcrmGgW0ZvVMhfpjDQda7bIkwV6M6aqcPjVtj1kovgcJHQaBGnQ2HL/7koWqpHECRiU7Qc3urB51GLdVeKmFQY7JZ57S8OgiM0jAzIU/Kr57sDnA4Fj2tqztM5rhAYol5v01kfzjGFpBw1TZIxS2b7P5xExGoQRehulUjJ1dJSK7HqbiOhURocozJHtrsJCFQLbWkzNDjH1fSG26XQ3SKTMRtEjcyxnR7mrxEIChfo2yVN0czobJYoSAliU7OV/ykv4d9yC9FrNHhlmR6ngycWLCMvNJyaoUGGvR7izRf3E/sgYJpwr1MoikxJ5S/oGThOfMwidFoNTpdqyCiKGsrP/oPbb/wrOp2ZkoqnyUi+ZZIEYpAtnuS4FYyNd/i3hQWnkxy/kvrmHciyD0EwkRp/B3uOfx6NqCcqogBB0CAKk/9sdFozkuRGkj2gyAjC5DHWqwGdAeLeYuobHgvzb5ZpOeul8SyIWgVJEhkZkGipkUCBwFCBuTcaGOiWOb7DgyVQxOVUI1efR8FsFf05yZnL9Ox62UVGgRazTaDhjA+3QyF7lpaTez0EhohIEhzY5CJvvo4ZCww0n/VSXapGdkWr1IhZOtdj6/WorrP2RIni3T7S8rXkzFEr2Llz4cRuN+VHPUTFac7pwqqC3+8Gogh5ixWG+xTaamXcTpGqYtU3TRAEBntlqoq9hEaL9HXIoKgFrvMOEN2tEu11Eh6nQu1pLyMDCqmFsr/wF5MCYXaZfS/B7JV65t5o4MAm9Yfj/LSd2Soyf42B1//qRFbgjZfcmCwCaTN06Bzqsbpb1BFea4CeoztcjA2p+ejdL7soWqbHYBKoK/fR2y5TtETPDUdn8rJxD5ua6wnQ6hjxesgICmHY4+H52iqeq6nkgcRset1O9vd24JUl5oREkWG7tvKL74Rpwr0OMTxaT3PHdsYcDdx7y6v+SbHaxq3sOvw42WkfJyF2MTqdWkBQFAXxAiQoCKrT7pth0NsYG++mtmkbS2f/DqMhhGWzf09t8/OcOPUUsuyjuX0/ibFL/cc+U/MSCfYllJ55hvCQAvS6qxfdvhNMAZAxS70GRYaBTtizwU1GoZai5Xq6WyR2PO9Cq4NV96iDD7KsUHHUy9lSL6ODE0pkyTk6AsNE9vzLRWCoSGKWlqQsrd+Q8eQ+D6IIs1YY/P2zQWGqmtapgx5O7hQxGAUMRoHs2ar4zamDHor3qOSv0QqMj8j+oY6sWTrqy30sWqeleLeb6CRFFYa5CggMU/9FJcpUHvZSVezFYBJUuctImcBwNcoc7lPzyLZgAQUYHVQIjlIYG5HRmxSKblCVxd4MvRECgtQOBYNJbUk7/1RwHuYAEQT1/M4RDas/YfL7uqXk6tj2dycR8WoqwhLmpLFCbTfzeRX2vuJC1AjYkzTceI+R+jM+4iIMfDtnDq2OUSIMJmJNVnZ0t/D3s5UE6w18M3MWbY4xvl52kNVxSZiNen5ec5K5IVE8nJTznpqSXg6mCfc6gtszxLHT38Ph6gJgQdFXJo3lpiau4XjZ7/H5nIyOd/q3J8UtZ9Ouz1OQ/WnMJrWyOzrWQUPLG8zI/CRuzyjDoy0Y9DaqGzbh9TrJTfscVkssACZjGDMy/g2AvsEy9hz5LinxNxAcmERt01bGHN30D9cjSzILC396rW7HFAgiWGyQNkNPap5aZbcnaenrkolL0fjzjqIokDdfR3WpFwU4vNXN7BV6jBa1M8HtVAVxRHHiSxqfpuXYTg8osPxjKit2NqtVdK9HHb5wjYMiKdzykJqv3f53F7HJWvLm6xFFqDrp4Y0XXaz7jFq1lyU1R73tHy7CYhSy51/9lqbQaFj0MbVA5fOCLZRJpB4eC4vvkBnsAgS1a0ONZt9+LYm5Mke3u1hwk5GQSJGORh+2kIn+04EeSc1lKwKpMyZMNEFVVAuNFhnskomxq4MRzWdE1nzSxECPxLGdHgwmtX3u0BY3Az0yiTkKydZgMt7UlHxP/ERxdtjj5gdnjvHX5WuIP6e9cF96Np/etZV5IVHkBU10NLyfmCbc6wgllb8gKjyH+UV/YsueLyGeM3scGWunvesEDS1vMDreyfGy36HVGKhp2Exa0lpCglJItC/hn5tuJyNJVeqqa95OVHgBr+/+Ij7JhdUSzehYBwZ9MCvm/war2X7BNYQF57Ny3p9p7thGV08tYUFziQm3EmCJJSJk5gU90a42FAVGB9SugMCwiQmm844ISW/REdDpBCyBk9clCAIWq0BkvEhjlcSrf1bbxXR6tfNhbEjBFjJBEiODMibLub7QfhmdQeDQZhcL1hqQztnseF2q0eWO513EpakR3qwVE1X3/IUGBrqdNFf7SM7RUn7Eg8GsMPMGBePbD+G9KwgCF9UrAJWAwy78cV8U0UmgyDKHt7kYH4b2RgkEsCdrGepVjTJTCmRG+wW87qnk7fUo/jSFa1xNGVmDRQ5vdTN/tR6dXmSgRyK9QIfJohJv0tsUFE8O9jArPMpPtgABOj23JKZwdKBrmnCncXlwe4boHSjhpuX/jShqSIxbRnn18zS376epbQ8hQekMjzQTFZbH8gU/orTiLxw48d8cLf01en0AXp+Tmbmf5Wz9q4DAbTc+Q2vnEVyeYdYu+zVmUxge7xi7Dn2bxtZXmZHx6EXXYjKGk5l837W7+DfBMQJl+0Vkn6orOzygkFYkE5umEovOCOOjCrbgiW9mYKhAQ4XPr0cLauFoZEBh5cf1FC0V6GqWqC710N+tIIhwdLubJbeqNjDOMZljO9y43WA0wYHXXUQnaEnJ1RGTqGXnC06sgSLz7jZgDRJob5A4tNlNco52yqNsuF1DdamXsyU+UBT0RgGj5YPdrH8xxKRATIo63DE6CI0VHipPeDEFQGqRTGQ8DNoUzhzykpytPaeeBu0NPsaGJ2QvDSZwu9TuheF+mYg4LQajQFTCRCjuGncjSxef/NMIAl5ZmrLdK8to+GCkE2CacK8beH0ONKLObxuelXI7FdUv4PM5+OStm9HpzCiKzKGTv+BE2e+4YdFPGB5tAUFgftGXCQvOQBS15GbczXMb1iAIApW1L7Nkzrf8aQa9LoDFc/6TF1+/i9y0RxCvsIIzMtbI6ZoX6e6twGyKJDnuTqLD573zju8ARYETO0QEQY2KRA3kzNFxtsSLNVj1HLOnKhTvdrP4ZiM6vVq4GuiWGB5QOPC6i+QcLc5RhdNHvGj1MNyvCsMc2uImZ46emcs1DPdLHNvhYeOfHRgtAs5RhegkDctu19HfpdrMNJ/1MWuFHq9bYaBH5taHJqavYlO0JOf46GqRpnQIdDZJaDSQv1BPT7uPoX7vxS73uoEgqqkKVQRo8o9HcCTY01WVs6gEDW6nwnCfTP7SiUKcVgf2FIVDm10EBAn0dUyWsuzvljCYp7oDvxmzQiL53/pyqgb7yQpWZRn7XU42NtXxrczZV/uSrxjThHudQKsxIckeOrpPYI+ag0ajQ1F8zC/6sr84Jggic/K/yHMb1uDxjmOPms3gcAMRoTlvOo4BW4Cd8rP/xOHswxYQO+k8FlM4kuxBkj0oikRzxzZ6BorRaswkxKwhPOTtraSHRxs4cPIrFGR/msKcBxgYquNI6ZN4PPeTYF/zru5BbSloNAKLbzYQGqVW2Q9vcxNh19BepxAYppAyQ6HquMQrTzkIDBUY7leIiFeYt06hvVam/IiERqf4rbz3verCFiqQXqgja5aam7AGiQTeq2HLc05EUWHOjXqSs9XXzFbVoHHfqy6azvoIjdJgtgiTpq9AdWRoqHRzdLuH/IWqU0PlCQ+DvTI3P2Sku0mhulQVH/+wIykXopNl+jtltDrIWzw1Uk2bqVB/SqK3Q+DINjeL1glExIkMdKmfcWLO2/cnW7Q6vpxeyJcO7GJOZDRmrZZ9HW3cZk8hzXqJts3XANOEe51Akt0IgpadB/+T3Iy7MehsOJz9U0ZpdVozgqjB53PR2nkUvW5yctDjHWN4tAWNGACCSH3LTmZkfsL/ekv7QayWeAREDpZ8DaPBSlbqzThdA5ys/CkpcR8jLeGui66zpunvFOZ8hoLszwAQEpRCoDWOrXu/Slz0DZcdNcsyjPSr6YLOBpGltxgIi1ZDnXC7hrmrDBzb4QZBYN9LAh43hEZB3iI1gjLb4Py0Z1IeJOWpY1iKojA6AP2d0Fot+oXLz8MaJGKywPgIU4RZIuNEfF7V/aDyhAenQ5nUfQDQ1SpjDVEYGvDy2l98KIr66CzL8MofXASFqyPMl2rhfr3DaAZ7ysVfF0VIK1JIKVDobIDju1yMDqqfX2K2jD3tnc8xNzSKP85aydH+TtyyzBMFi4l6L5PjV4Bpwr1OYDZGYtAHkpV2G63thxgYrscWEMfZ+o3ML/qy/32NrXsIMEdxsuJpRsfacXtGKT3zDOnJ6xh39HKk5FfERq2kMOurDI/Wc+DkV3G5h4mLnktPfyWlFX9lVu7jtHa9gcFgYd2K3/gfiRPjlvHS5ntIiFmDXme74DoHhqtYMOuLk7aFh2ajIOPyDFyWyE1/B5w5orZaSZKC2wFhMZOLX2HRIuMjCgFBAqvuNmKxqapUpQc8zFkjc7HRekFQH4NtoTDSrzDYM9m1we1ScI2rEe1At6qwdR5DvaoDcGiUOiSg1Qns2eBi7irVor2tThU7n7nqvNLXRBSrKOo/8b2vLV6XEMXzFjsyisJlT93ZdHpu/ADLM35kCXfM0UZb1x5k2Ut0xEKCbRnvvNP7jCBbJiXlfyImciZajYGkuBVUN76Gw9lLXMwCegeqqKrdgKzIuN2jhAUVYo9aTlvXG5yqfBa9zkZCzDrSE9UINdCawtLZv6O+5SVa2n+NxRTDgqKfEmzL4Hj5D8hIWjsp/2i1RBMekk1D62uYjOEEB2Zis0z+4zYbIxgYqifIlujf5nD24/O50GsvvT/X7YTygyKLbzYSFa/2C2/4Xwc9bTKRcRPk19MmneuxNWIynxMXn6FjbESmpdpL5ux3fmSPy1QoO+DBGiISYVedf4/tcBOZoBAUoXBsp5vF6w2ERGoY7lcdHfQmiErQUV3iJW2GFo9bHYhwu8AWAgXLL2wbIwiXTyIfVXwY79NHknAb2zZxpu7PpCXehEFv5GjZt4mNXEle+uff76VdFF19xxgereaTt23GZAymrmk7p8/+nTvWPMfZ+tdoaT+EyRSKKOpYOfcpzMYIPN5R9DorMRELL3pcqyWWgqyvTNmu01oZd/ZO2qYoCiPjbYyMtRERlseZ2j8SETqbouyv+1MFyXEf40jJkwRa4wkNTsPpGmTv0R+SELMarfYidrkXQGej6uN1PrIUBIG8eToObFJbscJjNPS0SRzd7kanx0+25xFh19Dd6uNSxMVDIiFjtszhrS58HvWxPyZFIb1IbV2SJIk9r7jwedUWqoRsmaAIOLXHS0yShr4umdFBBUmCohUyIdHveMppfETxkSNcl7ufito/csdNfyPQGgdAYfZneHHLvcRELCY0KOcdjvD+oLVrJwXZ92EyqmFTUvwKztS+zI4D32BG5icxm8IoPfMMCTGqBkJt8/MoioQgaEhLuIe0hLsua9omIeYmjpZ9h6S4FQTZElAUhcral5F8bj51+1bEc3nizXu+RH3rq6Ql3AmAPXIJbu8Qm3c/iiBq8XjGSIhZTV76Fy7rer0ugQDrZBJNL9RTW+7j4CY3kk/t3fT5QBTUvk6dfuL6etolzIGXXpCKTICIeBlbQABDI2OThgPi0iE2TcbnUSvq51uN562TaatVELWqi609Ve3jncY0LoaPHOF29R0lLnq+n2xBtYPJTL6Zjp4DH1jClSQXel3Auf/20tVXRmHOA+w/9l+cLH+K4dE2EmLW4vU56Bvawa2rniI4MJnB4QZ2HvgPNBoDKXG3XfL5QgKzSE+4h5e23IPREIwiS3i8Y9x2418Qz/XnaLVGZuV9jkPFv/ITLkBy7C3Myr+L1tZG9PpAdNrL1ygNjlKoK/WRO1fnd5X1eRXcToXClTJetU5GSBScPS5wYJOLWcsNmK0CzWd91Jb5mL3m8joABEHVWNCMX/i1t4rKGC2QWvDh7zKYxtXDR45wBTTIF2iQlhXfNZmSulJEhc2jsm4Der2NvUe/h8UciaLIuNyDGI3BKIrCmLOegaEa1i7/LcGBqglWcGAyS+d9h50HHr8swh0Za6Sm+Z/ERy8kNCSd5rb9eEccWAMmPy8b9FZ8kmvK/hpRh8V85c/WodHQVi3zxktOMgr1yDJUFXsIiVamWJtnzlVoOC2x43knHpfqNVa4QsZy4breNKbxvuFdEe7GjRt54okn8Hq9PPjggzz88MNXa13vGaLC53O65rf0DdYQFqzOYo87ejlb/xoLCv77fV7dVPgkN2OONoKs6VQ1PMv2/V9l7bJfY4+aA0B71wm27H2MlQt+THL8cv74j1mEhUwuAIYFZzLu7LzQ4S+K0qonmJX3OXLS70SSvUSE5FBc/hS7Dn2btcv/x/++ipqXiQxVhxrGHO14vCN+t4h3A0GAGUsVOuolqk+5EUSITZeJSlJfVxR16kyjnYg0UwuUSZXt87o8H8biyzSuT1wx4XZ1dfH4449z++23ExwczM9//nNmz55Nfn7+1VzfVYes+IiLvpGNOx4iwb4YrdZIY+se0hLuIch2Cc1+1wiKolDT9E/q9r2ARjTi9gxhj5qDTmv2ky23i+SfAAAgAElEQVSAPWo2CfbFOFxqgSssJIu2jiMknNO4BWjtPHxBF4bR8RbqWl5mzNGK1RxPSvwdWC3xuD1DDI81kJl6K30D1Wzb9xUs5ghMxmDauo7xyvYHyUheS2PbAYZHWpmT9z0OFH+F0fFmTMYQxp29LJj5VUKsS9/VPRBFiE1T86dvRn8HVB0XQVEFWQKCIGeBjClAJdfxEagtEehtU210opMUUouU6fzqNN53XDHhHjlyBK/Xy2c/+1kiIyP505/+xIEDBz7QhHui/L/o6juCRtSh11loattHUuwtLJ/zx3f1+PteoKl9Mx09u7h99bO0dhyio6eEsOAM3J6RKe8NMEfhdqvbZ+U9wr6jP2S+9yuEh2ZTVbuBqrpXKMz62qR9BoarOFz6H+Rm3E122ho6e0+x78SXWFD43wSY7aDIyJKP7Qf+nbmFj5GWqE6JjTt6+de2T1FW9SJmYyRZyQ9QWvVLEmLnMSvvfxFFLf2DtWze82/MyQu76jlxx6jaLjZ3laruZQ4QaD4rUbrby/yb1cLWyR0imTN1LLtVh8+jSiSW7ZWYuerK3RSmMY2rgStOWnZ3dwMQFBSEwWDAYrHQ09Nz1RZ2tXHwxM/pHy5j2bzvsn7lH8hKvR2t1kBr5xuYjOHvfIBrjPrWDSya/U0CrXEMjjQRHV5IbPQ8Glp24fU5/e/z+pzUNm1hZKyNTbs+T3XDa2Sn3cH+Ez/hpS330NFTQkRYHqVnn6CpfYt/vzN1TzG/6MvMnvEIsdFzmT3jEeYXPUZl3Z/R62yEBGZzpPRX6HUBfrIFsJjDyc/6FJLsIDI8jeqmZ3E4O5iVN6G9EBqcRn7WfTS1v37V70t7rYA1WODYDg8le71s/ZuLwT4ZjUZVCuuoh8h4DTlz9Oj0quX6vNUGPE5V+3Ua03g/ccUR7nnh6isV9g0NvXZWGD6fi+rdG7lz7T8ItMar5w9Owye5qG54HY9cTVzkuxdXuZpwuXsnCl+2JDp7SsjLvBd71Bw27niI3Iy7AIHys/9Ekn24PaPkZ97HuLOX42W/R1Ekblr6JLHRcwEYGmnm1R0Pkpo0myBbEr0DZaQm/m7SOVMSVrP/+I8JD7eycuF3eWXHA5NcIs7DaAgiMiyXeYWPYY+ay/FTv/F3LpyHLcBOe9dxwsOvrhh5yYATyaew/gETJosqWH14qxuXA+pLNWh0AgkZU6UYI2K1iLJIeLjuIkeewNVe87XC9LqvLa5k3VdMuOHhalQ4NDSEwWDA4XAQEXEZY5v9Y8jytWmpGR6tx2gI8pPteSTFLaeydgPdvR0YtaPXZC2XimBbJs3t+8hIvpXQkAxOlv+J42W/Y17BY9S3vEHpmWfxesexR83FZo1j9ZKf+/f1SW7qm7f7yRYgyJZAZsotlJ3ZSE7awxj0QYyNd06aCBsb78SgD6K3dxQIZMW8p9lx8BMMjTT53yfLEmfrN5KZonY8RIRmMzzawth4FwGWKP+x6pt3YrPknDvW1YNjTGT+aiMmi0qqWp3A7JUGXnvaQe48HTVlPqqKfaQX6PwC4rKs0N3mIyxeprd3akfFmxEebr3qa74WmF73tcXF1i2KwtsGk1dMuPPmzUOj0fDUU08REhKCJEksWbLknXd8H2A0hOJ0D+Lxjk8ScxkYakCvs+DzXsBa9H1GZvL9HD75TUoqnkaWJczmcMqq/kZJxTNoNDpMhjA8XgeDw/UU5T40aV+d1oDJGDLlmAa9jbExtVshyX4zB078jBsX/xSD3orbM8qBEz8j0X6z//1GfTAzMr7Exh2fJTv9TszGECprX8ZoDCE1YfW5Y1rJTruDDds/w5z8RwkwR1LTuJn+wWoWzfy3q35fZB8EBL7VWVctlkXEaolN1bL5Wec5FwcDXo9C2UEPRosqVj6NabyfuGLCjY2N5Sc/+QlPPvkkbrebr3/968yYMeNqru1dQ1EUegdL6ejZj14XwN6jP2DpnG9jMNjo6T9DcfkfCQvOVM0PP2AICczGaAghO/U28rPvQxBE6pp3svfI90hJWE14aBbNbfvp6i2jf7CW5PiV/n3tUXM5eOLnjDt7sZjUJxGf5Ka6YTO5aaqwTGbyfZSd/TV/f3U9gdYEhkebiY1aQWbSpyatI9G+liBbOi0d2+gfaEWWBaIjCtFo1EdzRVFwuoYItGZQ27gbj3eE8OBCbrvxr4yMXP0KVXCUQvNZHzlzJ1oOOpskzFYBg+m8DbiW2nIPrzylas9GJSvkL1WmC2bTeN8hKG91EbxGuBYphdPVv6Wr/wjZqbczONxAc/tBZMXnn9ialfcIxaf/yLz8/0fQBdqmriaGRmqoaXqe4bF6AsyxpMZ/nPCQgou+f2C4ipNnfsy9t7ziz5Nv2PZpctPvIj15vf99+4/9F3XNO7jtxr8QEpSCLEuUnvkLlbWvAAo56R9HpzNTVfsKAeZEZuU+Pinv7nL3M+boIMBsx2iYGhW/FePOTg6e/CpBtnjCQ3No6zqOz+dlUdEvJplHXs6j4rizE4ezC1tAEgZ9EE53P5V1f6az9xCioCU2aiXZKQ+g1ZoZH4biHSLJOVpikrQMdEtUFXuZt8aAPUmNH45udyHqvSTlXtLpJ+HD9oj7QceHbd3vWUrhg47BkWo6evZz1/oXMeitKIrC9n1fZWislezUj6HXWSivfoGIkNnvOdkODFVy+NR/MjP3IeZFfY7egTMcK/0hBVlfu6iwjNsziNEQ5CdHl3uIoZEmUhNvmvS+3Ix7qGvewcadDxNgjsbp7sdsjGLJ7N/gdPXQ2vUGsuQlI+kBosPnTylyGg2hGA2hl3wtFlM0N8x/hvae/Yw7OkiNu4eosHlX5A7h8zkoPvMT+ofKCbQmMDhcT0LMGjp7D5OScAOLZv8Tn+Sm+PRTHCl7nEVFT2AJFJhzk0xLlZeS/T7GhyCjSEtM4jljxyYfrXUS89e/w8mnMY33AR9awu3sPUJa0hoMejXqEgSBG5f8nNd3P0pl7WtYTNGkxN1NbOSy93wtVY3PMq/wMbJS1UJTaHAaFnMkB0/8kujwBRfs9HC7h+kfqvEXo0RRi6LISLIHUZxQ3fJ4xwCFEFsOBkMIWckPExmmWoqYjRGEBqlhntfnoKbpBXoHilFQ0GpMWEwx2COXXXavrEZjID561du+Z2CojobWYxgNIUSFzfUbXr4ZZdW/xWwK4KblW9BqDDhdg7z2xufQanTML/r//O9bufBHPL/pTvqHygkLnqFao89W7VxGB6HioI/6Ch8arToIkbdYxnD58g3TmMZ7jg+ueMC7hNc7zsBwI16vw79NFLWYjWEk2dczL/+HxEWtuCb6CQNDZ0iMnTx1FRs1j9HxNiTZfcF9+oZOkRCziI1vfJbK2pdp7yrGaAii+PQf/S15Pp+H3Ye/i0ZjICoykwCLlRMVP6KpffOkY/l8TvYXP8a4q46ivE+Tnbae0fEGnO4WTlT8kMq6p6/atSqKxMkzP+X13Y/icNfR2P4yOw7dx+h4y+Q1SS7au/ewcNa/o9WoqjAmYzCLZn0DSZnw+ersKWXv0R+CAHXNL+N7Uw8yqG6089bLFCyXyVkgs+g2mdAP1gzLNKbhx4cuwnW4ejh++ge43L3o9QE898pNzCl4lNz0u+jtr6Sl4yA3Lrx2urdjjnZAYXC4AZNxpn/7yFgbOq0ZjXjheVONxkB4aAZZaR+jun4jbs8o6cnrqah+npaOQ4SHZNHScQiNqOPeW17x56VzM+5mw7b7iA5fiEGv+rfUND+P1RLOjYt/SnffaTq6i0mwL+Zsw2vcsebvbNz5MLFRK7EFJF7SNXX1Haeu+XlGHW0EBiSTnvhJwoLzAGhq34rD1ca9t7yK7pz+bUXNixRX/Jjlc//XfwxJcoEgYtAHTjp2gCUSj0fNjZ2peZmSM08zI/NTxMcspLphE/tPfpklM5+cpK0rCHxkrGqmcX3jQ0e4x0//gKS4RczMexhBEBkebeHVHQ9S0/A6QyPNzMz5Dz8RXQvUNb9IVHgBh4p/xuolv8RmjcXh7GPXocdJtK+bEmG3de2loW0DY+NtdPYe4M6b/skNi34CQHP7AXw+Nwn2xec0amWCA5P8ZAsQZIsnJnI2XX3HSIhZzdBIDQ2tr7Bo1jc4WvokDS27yEi+GVHUohF1lFX9HykJq+jsPXJJhNvevY/TNb9hQdFXiQjLpaP7BEdLvsOcGT8gPCSftu5dFOU+4CdbgOzUOzhZ/mfGHO3q2DCg1wViNoTT0nGIBPsi/3trG7cCAiUVT1NW9RwfW/N//v7p5Pgb2LbvqzS1byb1TXKQ05jG9YIPFeGOjDXicvdSlPuQn8gCrfHMzP0s9S1vsGbxC+i019ZUbnC0msWzv05X7yk2bP80Op0Ft2cErWgkM2k2zR3b8XhHiQgppG/wNA1t/2J+0VcIDkzm8Mlf8o/XbsEeOQef5GRwuJElcx/nxOk/oBF1+CQ31gD7lHNKkgdRUD/aito/EhacQWfvKVraD/DxdS/489o56Xfx4usfxx45B1tA5Dtei6IoVNb/hZULfuQX0LEF2BFFPRXVzxAe8itk2eu3cj8PQRDRiHpk2fumbQJ56Y+y5/D3yMu8l/CQLFo7D1PfspMbFv03R0uexGKOnDSsIggCGcnrqah+dZpwp3Fd4kOVw/V4RzCbwqeMmQZYotBq9NecbAHMxigGhmopzLmfT92+lfUrfsfd617C63NwrOx79AwcxCd1cqj0G1TW/Zk1S58gKW4ZQbZ41i7/NXkZ99LZcxIUWD7/h2pkKhhQZC0hgclU17+KwzkhEtDTX0lXbwlRYXNRFJmegVJm53+R2sYtpCWt85MtQIA5gsTYZbR0HMR+CcVDWfYw5mgnJnL2pO3xMQsZHK4BICpsAeXVL6AoEwpfLR2HABGrZfKkX2TYbBbN/CWtHcXsPvI9PF4HH1vzN4IDkxBELW7PKG/tWhx39KDTXrux8GlM42riQxXhBlnTGB5tYWikhSDbxJe7tnHrFJK4VkiJu4Pjp79HcGAy0RFF6HVW9h//MaKoY9m875IUtwyA+UVf5uWtn2BwuIGQoAk92bTENTS07CEwIJdjpX9ArwsgK+Uhhkbr0OsloiNm8uLmu0iMXYbXO05z+wESY29BpwtAURR0WjMWUxix0XPPdTRMhtszTHz0jZguoTVMFPXodVaGRhr9Og8A/YPVWMzqWG9K/Mc4VPJ1Xt3xEElxyxkebaGhZTdzZ/zgggXKQGsKC4t+TmPbJqoanqGl/SCS7CEhZh0+bzGnq/7GjKxPIQgCI2PtnKp6jpnZ37rcj2Ea0/hA4ENFuFqtmZzUh9n0xiMU5NyP1RJFTeMW+gfrWTL3cUaGr/2awoLzyM/8MrsPfx+vbxxZ9hIROguTMcRPtgA6nZmi3IepadxMSsJEy1X/UC1mUzTZqQ+QzQP+7baARPafeIzbVj9NauJqWtoPMDrehSjqyE65H1AfwRPsazl08pfMnvF5Nu36PPmZn8RmjQWgb+AsnT2l3Ljwb5d0LYIgkBp/J7sPf59Vi36MzRrLwFA9B47/lNT4ewDQaowsnvkrxlzFNLacwGgIY+W8pzEZLzxXqygS/UNnMBkjWTnvr8iyG73OhkZjYMzRwbHT3+VM7cuYTeEMDNWSlfIg4SEfXAnQaUzj7fChIlyA5LhbsVoSaOrYjMc7TFhQAbmzvnzuUfr9mWixRy4hJmIRLs8go+MtNLdvw+0ZnRKJa0QdA0N1OJz9mE2hdPdVcPzU7yjI/Pcpx7Ra4slJe4QN2z5DVHgBPslN/2ANc2Z8f1LqJCflIY6c+jYbdz4MKLy89RME2hKxmMLp6C6mKPsbl1VETE+8F1n28a9t952LWEUykj5Fgn1CwlEUtaQkrMJmfnsFtqHROo6VfRe9zoxBb6N/qIa89EdJtKvDHQHmGFbM/RNDozV4vCOE5Gah002nE6Zx/eJDPdr7ZrzfI4SKonC6+rd09h0iI2ktbs8odc3bmFf4ZbJSb0OSvLz2xudQZB2DI1WIohZB0DK/8DFMhjwGhisx6kMICcyeNCjh8Y7S038CQdQRGTp7SsFqeLSBAye/QlHugyTYlzA4XM+h4l8QGpTPjIxH0euuzPhLlr24vSMYdIGTpsxk2UfvYCkWi4BekzFp3Hfy/j62H/oE8wq/RHrSWgAGhxvZuPOzLCr6xVWx6bkSvN9/J1eK6XVfW3xkR3v7hyqoa3kZh6uHsKB80hLuvKxR1WuFvsEyuvuPcte65/2Fq7zMT/Dy1nsZHGmipf0gZqMdjc6ARmMgIjQXh7Ofo6d+jU/yEBVWwMhYG4KgY37BjzEbVSlMvc6KPXI53X3HKK/5PaKgITZqpX/CrKbpHxTmfIb8LFWUJsgWjy0glk27vohWc+XjWKKom5L3HRw+y9Gy72Ixh2M02OjuKyc37RGSYm+esn/PQAkWU7ifbAGCA5PISbuT5o6tzMi4+kpj05jG+43rmnCb2rZQXvcHFNmHQW+jqX0TTe2buWH+0x84F4eO3oNkpt7q13XweEaxWqKJiZjJmernSUm4E61opqHtX0iSm+6+MkAl1LvWvYTpnDNvScWfKa74MUtmPQmokXNJ5c8ZGq0kK/V2JMnNifIfkmi/mczk+xgaqWZe0ecmrSU0OA2NqMXp6r0sayFFkRgYrkKSPIQG5aDRTPiGy7KXI2XfZsmcb5IUtwJA7YHe/hDBtowpehVe3xhm09S8rtkURv9g2yWvaRrTuJ5w3RKuLPsor/0DCfZFLJ7zH+h1AfQOVLFlz5c4dfZ/mF/wo/d7iZMgoEGWvDS3H+Bo6a8ZG+9GEATMxlAKcx6govolfLILjailKPchggOTaWjZRVvnUWTZpx5DECjIuZ/y6ucZd3ZiMUXTN1jGwPBpPr7uef+wQWbKbbzw+p3ERd+AyRTJwFDdpM4Hh7Mfj3d8ypTX22FopIZjp7+PVmtErzMzPNpKQeZXiI1aBkB3/0lsATF+sgW1Bzo7/U5aOndMIdzw4HxOVT0xSUJSliVqGjeTEH3rFd3jaUzjg47rlnB7+kuQZQ9L5j7uJ5rwkCzm5D/K8bLfv8+rm4q4qOUcOvVNxDqRFQt+RGzUPBzOPvYc+R7DY20IggYBgTVLnyA6ogiAxNilHCr+Bacqn2HhrK8DamHNoLfh86kaEV19R0hPWjdpsstsCiXBvoTuvuOkxN3JkdJfYbPGERGafe6cPyAhZjVa7aWlFCTZw+FT32LhzK+SmqgKj/cNnGXTri8SaE3FaonFJzkwGoKn7Dvu6KGr7wT9wxWEBRWSlvBxjIYQjIZQ0hPv5ZVt9zMj65MY9Taq6jeCosMe+e7cfqcxjQ8qrlvC7eo/jtEYPIloAIJsSVOa5T8ICA7MxGKMIjN1PXHR8wHVkHHV4p/y91fX4/W5MOoD/GR7HulJa9lz9Pv+/+/qLcPrdWK1JACg0ZjweKcm792eUWwWI9Hh8/B4H2Db3q8hK15k2UtCzBpy0x655LV39hwm0BrvJ1uAsJBMMlJupqVzOzmpDxEeXMCpqicYc3QTYFan1k5WPE1b5xHmFj5GgDmK2qat7DvxKMvm/AG9LpDU+DsJtmXR2rUTn+QkJnwV8dE3XJHU4zSmcT3guv3L9vrGcLkGGR5tmTT+2di6C73ug6lkIggawkOyJ20z6K2YjMGYFB1jzjZGxjroHzyLQR9IdEQh485e3O4RVQtitJXKmpcpyP6an5TiIpez98S/kZ12h993rLOnlM6eEvIzvgFAQswa4qNX4XIP+HtcLwX9Q2corfolw6MNJMROtU+ymMIZH2sAwGgIITP507yy7X5yM+5Go9FTUvE09968we91ZtAH0NZ5jG0H70UUNHh9DiymGLJTHiAueuWU409jGh82XJeE295zgI7ufZiMIWx64/PMK/oygdY46pt3Uln7L4qyv/l+L/GCCLSm0tZ5lOiICaeHMUcPY44eVs1/hj3Hv8CLmz9OTMRMxp09eH1OZMlLcOAMapv2YdCHsHjWr7AFJAEwOt7CkbJvodNZeHnLJ4gIzUFWZAaHG5kz47uT+nEFQXNZhcS6lg1U/P/snXd4VFX6xz93SiaTmWTSe++EhBISIPQqgmDFtvayVsCGXcSyLroW1rL+7KuuCoIoigpSQwg1oYSQ3hvpvUwmU+7vj5HBmFASQiQ6n+fhIbn33HPOvZm8Ofec836/ee/h55WAWuVKRdUBOnVN2CrMf8xMJgN5xZsI9bveck1YwDU4a6IordxMl74aZ02IJdg2t5axYds9jIq6jZCA2bS2lbPn4EocHQLIKPgAqdT2lGLsVqz8WRhyAddkMpCWtZJLZrxFcuqryKQKdqf+C6OxC5PJiESq6JGzf6EQ6n8NSSmLsLFRE+I/i+bWUnYffIPwwOtp6yhHLldyzZzVqJRuiKJIRu5aDmV8THz000gk3dNiRVFk/9HljBp+C8PDrqJT10xB6RZS0t4nNuoxPFz6n8rc1lFBVsHHXD3va0tiRuK+F1j78/WMib4TuVxFRu43mIwi1fWpVNbuxdEhHH+vWbg4RuPiGI1c0cD3W+7CZDIgkcg4mv0lUaFXMXLYDYBZx2HutH+z+scrmRT3BEezVlkDrpU/PUNOvKapNQ9bWye83GOZO+1NmlpKGD38Dm5duJ3br0kiYfQD7Dn8GHpD+x/d1R7Yq3yZHLeSsuOHWbfxZpJT3iDQ+1KGBd9CSeUmRkfdalmxFwTB7EcmU9Lclt+jrqaWHETRQFTolQDYKjQMD1tI/Ii7qajeftZ9MpkMPea8y6u2Exo4t1sW3NRxy3BQ+3Lo2EccTP+U5tZSTKIOGxsTDc1HKCj9ml+Sb6C+KQMw27Lbq/zZc/ANSxbcCYWxE9gpXdCofbG1caC1veys+2zFylBlyI1wZVIlXfo2RNFEZc0h3F2iGBF58rV2WOjllFYkU161vduG+05dMx3aWpS27r1a2gwWDuogxo14rsdxg6EDpW33VX5BELBVOGLo5Y+H3tjRzfPsBLa2jugNHT3K/57ahsMcy/+AxuYcbORqgnwXMCz4NiQSGSaTHsXvPGoEQUBj709DUyG2CiWBwfMZP3oJYN7OtXX3kwAczFjB7An/A2BszLNs3XsbuUU/IZHIqanPwNvjpAh7l76dlrYKWtqP4/gHZZZZsTKYDLkRrr0qABuZhmM5X9PWXoWTpucvqrNTKB2dNQDouprYl/YMX32/gB0H7mHr3luork8d7G6fBQLHcr/uNtpsaCqgubUUJ82wHqWdNVE0t5bQ0HRy9CuKJrLy1+PhMrZH+d/S1JLH/qPPExdzG3+/bh9XXfw/2rSFpOW8BYCX2wTyis16Dyfo0NZRWLoFldKLhuYsYqPvsJyTSKTERt9JXUM2YKKlrQgAhY0jIX5X4uocRUTwAg6mf8CxX6Ub2zpq2L5nGa7OkaQefZ/wwO727Fas/BkZciNcQRAYO+JZ9hx+EgERESPjRy+xrNqLooni8iQiAm9BFEX2pT2Ll3s0F097GZnUlrLKPWzbvYyp8W9fMHO9za3F1NQfRC635fstdxAZcjntHdWkZf2PEZE39tBHALMq14jwRfyw9R6iw69GrfIgt/BnuvRdBEbN66WVkxSUrWNU1M0E+5t3BjjY+zJ70gq+XD+fqJDbcdJE4u0xjbU/XUdk6GWYTHoycteiUYcydsRyNiVfh/R3ppAyqQKTSY8oQmHZetJyXgVscHOKpa4hk05dA/4+kziQ9i77Dr/169yuHEf7EOKin8bdJbaXnlqx8udiyAVcALWdL7MnfEpN/REy8t9j486HiR1+G4IgIS3rCwQUeLqOo7k1j05dLRPGPGTRYvX3nkhU2FUUlW9gRMT9g9Jfk8mA3tCOjVyNIHQXRxdFIwfSn8XbI5aYiOspq9pLavr7yKS2BPrNQBSNlrK6riZKKzfT0VmDs0MkPp7TcFAHUXJ8IzX1RXi5zcTPa9YpfdJO0NZRzki3q7odU9jYo1Z506GtQmHjSEzYvXi7TSK76AuaWnLQdbWgs2mkuj4VldKLzLx1xPxmKudYzmocHYKoqc9AJjcyYcxSdF3NJO57gdjoOxkVdRMABkMnP2y9B2+3GYT4X3Guj/Yvj1bbTltbEzU1Jkwm05kvuMCoqZEMqX5LpTLUakegd1GmMzEkAy6Ytzl5uI7B1ekd8ku/IWn/K4iY8HabwojRjyIIUjo6a3HUBPUQvnZxDKW2fvN576MoiuQWryK/ZA0m0YBUaktE0I2E+F1uKVNZuw+ZTMG86W8hCBL8fSYyftQSVv94FQaDFonUPJJsbMlhz6HH8fOeiKtzKIWl33E09x2CfC5jWMhtp1Tl6g0HVRAVVandkiy0nY20tR9H9avnmCAItHWUo+2sZO60lXi4xlBZc5hfkh5FIVdzKOMTjtccxMN1BCUVu2hsLkBv6MLbfTTTE55HEARa2ysBus2xy2S2jIm5k9S0j60B9xzRattpbW3E0dENpdIWo/HCS/g5EzKZBINhaARcURTR67toaqqluVlJf2Zkh2zAPYFUqiAi6AYigm7occ7RIYyDGenoulq7WcsUle/E0T7ivPctr2QNVXVJXD7nYxwdAqlrzGXLrieQSW0J8L4YvaGdiuqdhATM6vZHQSq1wc9rAjmFGxg36n66OuFI1htMiHuY8KBLABg57Ca27XmG6vrdVFTvYOrYd85aajEkYCFJKUuwU7oQ7D+LlrYKklP+RaDPJd0Cd27xV8yc+AKebmbBb2+PMchkCmZO+ieODgHkFv5Edd1RJBIpJpOIj/tEgvwTLAt5RqMemVTRY1QvlykxmrrO6dlagba2Jhwd3bCxUfz6zIdewB1KCIKAjY0CR0c3qqtrcHLy7HMdQz7gng47W3f8vS7ix233M3bUvSgVzmQX/EB1bTrTx51f+T9RFMkvXcv8Ge+gsffnePVBOrR1xMX8ndSjH6PtrKcLI1cAACAASURBVCOvZDUymRKZ3Njj+rrGbPw9L0Jj709pUxHt2kpCA06KfAuCwIjIG9ixdzkebtEUlH7HsJBbzqpvDqoAJsa+SlbBx+w9tBKFjTNBvpcSFnD1b/pvorW9DA/XmG7XdmjrcHWKQCqVExN5HTFch9Gk56NVCdgq3GhqKbGU1dj7IZMpKKlIItB3quW5HMtdg6drQp+ep5WeGI0G5PLTTx9ZGXjkchv0ev2ZC/bCnzrgAoyIuJ+iih9JPfoe2s5W3F3imRr/dp9ewfuDKBro1DUgl6tY+9N15m1VDgFUVKVgNOoor97CNZd8TVtHNT/tWERB6VaC/WaadxoUfEdbWzUTRr4OmB0UTKLx1/nck6NFo7ETqUROePA8Uo58BJxdwAVwcghnwuhXTnleECQ4qAOorDncbSuXo0MAZZV7LAEUoLgsEaWtO20dZRSVH8LXaxy+nuMwmQx4uo1ia/JThAddgqMmkKLSHei6Opg05uGzf5hWTskfucXxr8q5PHOr48N5ZMueW5DLbAgNnMOoqFsRBAFdVyvf/XILft4TcHOKZM+hlfh5JXC8JhWDQQeIqJTejBn+JA7qQOSKevYf/i9VdXtQKd2ZFP8Ynm4jMRr1bNr5ID6eY1HYaCgo3sm4kS8MaP9LK7eQVfgJ08Ytw8NtBJU1h9i2exkAE8c8jKf7aMqP72Xv4TfxcI0hNPAiSo/vo+x4MjY2agwGHRr7UIaH/p2a+lQ6dfU4Ow7Hx2PqGRf2/iiGkgNBVVUJnp5mEaOhNBf6W4Zqv2tqynB39+tx/E/v+HChIoomVLY+1DUdYeSwmyx/FRU29owdeT9HMj8nt/BHLr/oE5w0wYiiSENTIVuTnyAs4Doc1IE0txaQvPMRosOvZdTwN37V+12Cm/MwWtrKcXWOJMB3Gj9vX8SIiAcG/B78vWYjICFx3z9+Ha2r8Xafirf7RDLz1rHvyH+QCjb4e09k1qQVCIJAZMjlVFSlsG3308wY9wF2SvM8l3Mve4mt/LU5dCiVJUvu4bLLruTRR0/vxPzaaytIT0/js89Wdzu+aNFdHDlyiO+/34SLi1nQftKkuF7rmDt3Pvfd9wBXXDGXZ5/9BzNmzBqYG+kDf5mA29ZRw9Hc/9LYnImtwpVg38twcx593trLLV5NZ1cVSlunHnKDCoWGdm0Nbs5RFrtxQRBwcQohJvI6Ssr34us5nazCz4iNvp0RkeYFQQ/XGDT2/mzZ9RiCIKe9o57vNt1MZPDNeLqOOy/3odM3I4oG4kfei52tC9mFG8gp+oJJsa8hlSpIPHA/UWFXdXvN8vGMRy63uyDTq/8KqN76FEm7dtDaM6mUtC+5tU/XVFVV8s9/Pn9WZUtKitmwYT2LFj3U7fjHH7/PkSOHepSfPLm7nnJWViZ1dbWEhITi5OTEhAmTeffdN5kyZRoy2eCGwH63lpyczIoVK6ioqGD48OG89NJLBAYGDmDXBo6Ozho27bqfYP/ZTBizhMaWYlLT/0lUyJ0EeM85cwV9RBRN5Jd+w4JZ77Ix8UEqqlLw8Yz/9ZxIZt46OnUtdOoae1yr13cg+fV1u6Epg+kJ3ZXPfD3HIYomJox6EaOox8khopsq2EDSpW8lM/9jrr5kFQ5q83ax0MA5/LjtfkortxDkOx+5TImuq6XbdSaTgS592zl5plnpP4MZbPvT3oYN6/m//3ublpbmsyq/Zs1XmEwmZs6cDZiD9cqV/2L37l29ll+x4nXL1zU11dxww0JiYkZyzTV/A2DmzNkkJe0gKSlx0Ee5/Urt7erq4oEHHsDLy4t//etflJeXs3z58oHu24CRV7yKsKC5TIx7BG+PMQwPu4p50/5NRt77FvuagcRo6qJL34qzJpQp8U+xJflx9h1+k6z87/hh698pqdiFaDLQ0lZBWtaXlus6tPUcyfoMVyezfKOtwpmm1pJudbdrawEJTppI3J1jz1uwBbNBp5tLlCXYgnkxLSJkPrUN5vRoP685pKZ/iE5nDrqiKHIk83Mc7QP65Jdm5a/DZ599jFwu5/LLrzpjWZPJxPbtW4mIGIazs9m09NChVHbv3sWll16Bm5v7aa//z3/eRKvVsnjxQxbFvfj48UgkEjZt+vHcb6aP9GuEq9Vqufvuu5k2bRrh4eH88MMPZGdnD3TfBoy6pnRGRD3b7ZircyQymZI2bQUOv7onDBRSiQKV0pPKmkP4+0zkijmfkpX3HfuOvA2iCXuVJ4G+01DbebD30EqKyrahtvOkrHIPbi7RlB7fiL/XLIJ8L2fvoZVcPHUl9iovdF2tJO1/iUCfuUh+l1p7AlEUaW7Np0vfgpMmsltANhp1NLZkI5Op0KhDzrjaKpep6dDWI4pit7Id2jpkMvPCgJ/nLJpa8/jy+0vxco+lubUMUYT5M96kc3AHWr1iMhkoOf4LlbVJAHi5TSHAe47VVeIP5MYbb2XmzIvYtSuR9evXnbZsTk4Wra0tREREWo4FBATx1lvvERsbx8KFPR2hT3D8eAU7dmwlLm4sUVHRluMODg54eXmTlnYYo9GIVCo9ZR0DzRk/denp6Rw7dqzH8bvuMjvBZmRkkJSUxOWXX96jzOk43UreQGOvcqWltRx3l+GWY3p9Bzp9M96ePihtB36L2LjR97FtzzNMGPMI7i7DqahJxd7Ok3GjFiGXq8jMW0fp8d1EBF+KtrMeX69xjB+9BKWtM59+MwN7ByPxo65FKm/jm5+vR6V0p01bTYj/bCbFPYJU2nOVv7W9ks1Jj9LZ1YzazoP69DziY+4hJvJ6cot+Ys/B13FQ+9Kpa0IuV3HR5Fe6uWX8HlfX8aTlGMnK/5ZhoVciCAItreWkZ69i9uRXcHMzP7eZ7o/T2n4z1XXpqJSueLqNQhAk2A/ej7hXRFHk58QldOqaGB11KwBHs7+gue0wsya9fMo/OCfu60KnpkaCTHbyJfW3Xw82fWl74ULzfm+JRLD8f6rrS0rMQkh+fn6WMiNHjui1/d/X8cMP32Iymbjqqqt7nPP3D6Ciopy6ump8fHzPuu+/pT+fkzMG3MTERN55550ex6+//noyMzO5/fbbcXZ2ZsmSJX1qeDC3hfl6zOfA0Xdxc4lCY++H3qBl98HX8HCOp61VRlvrwG8D0tglMDIC0jJW0dxWjCgauOnKTRYPNg/XEfy0435MohFBkBAZYnaqNZkMmEQj6VmbCfSZy5joO/F2WUC7thJbhSsKGw0NDTpA1609URRJPLCU0MAZjP5VV6KlrYLvt9xFR4eRrMJPWTDzfVycwszJBzmr+Wn7g8xK+G+P1OffEh/zAvuPLONo9mqUtk7UN+YQFXonUoJ/t31KjcbOnMxQV9d+QWyvyiv5hvrGXG64bIPlD1SAz2RWb1hIdt4eXJ16/uJeCP0+W0wmk2VL1R+9vao/bZ/4/TeZxFNe39DQAIBSqTptGwaDqcf5zZs3oVAoGDs2occ5Ozvz+kJdXQMeHt597jvQ6+fknLeFLV68mMWLF/c4npeXx2233YZKpeKzzz7D1dW1j90dPLzdJyFIG1i38SbUdh60a2twdRrFmKjHepTt1NWj62pCbed71t5fp8LLLQEvtwSKK36mqfVIN8NLQRAI9J1GRu4aQgPnWo5nF/6Axt6f/NK1BPqYj8tkdmjOoBfb0laETt9kCbYADmofRg+/hfTsNcREXIeLU5il7eiI68jM/476pmO9Bp4TOKgCmDXhUxqaM9Hr24gfHo1c/gcPXc8Cs47Fl0QEz+/2NiCV2hDsP526xrTT3reVC4MTaQJ9FbgpLy+jpqaasWMTUCh6qu2d+B2RSgf3raBfE1kGg4HFixfT0tLCI488QllZGVVVVcTH99/W5XwzIvIG3DSzae0oxdbGuYe/l97QweHMV6muT8VO6YK2s5Fhwbeek8BKS1sxx/Leo7J2Pxr7npukGxrzaWktp7hsB4gi9U15VNUe5pLp77D25+u7KYWdiS59K3a2Lj1Gq2o7D4xGncVb7ASCIKCy80CnP/NKsSBIcHGMPmM5cz9ayC1eTV1KCgIK/DwvIsh3/mlH0eeDLn0Len0bbe1VPc41thTjojm9ZrCVCwMnJ2cAmpp67ug5HWlphwG6zf3+lra21m71Dxb9+i3YvXs3RUVFmEwmli1bxm233caiRedXm2AgkMmUODlE9GqmeDjrdZRKFTdduZHrFqzjijmfUFC2lsraPf1qq1PXSPLBhwnyn8htVycCcDD9I4xGPaIoUly+k9yiH5FKlTg7haPXd+DjEc+189chiibslO49RF9Oh5NDOM2tpTS1FFuOiaJITuFPaNSh5BX93E3cvF1bS03dUVwdY3qprX8YDFqSUh8AoY0pY59m3Ki7qajZwpGslQPWxtkik9oiCFIqqg9QVJaIKIqIokhRWSLHq1Lw9Zg+6H2y0nfCwswiU7W1NX26rrra/Ic2MDCo1/NlZaU4OGjw8Oi7AM250K8R7tSpU8nJyRnovgw6JwJQl76Z6rr93HTFRstrv6NDIGNH3kdm3vd4uU3oc93FFT8R4DuZEZHmvX/zZ75L4t7nOJz5XySCHDAhkcjxch9FReV+bOQq5s98lw5tLdv3PkdYwLV9ak+nbwIR1m++ndHDb8de5UV2wXoam8uYEvcme9OeZtPOh4gMuQytrpHDGZ8SFnAdCpuBs5QvrdyCg70P08Y/a1mQ8nQbxRfr5xMWeC1qu/4tTvQHqVSBr+d0jKZm9h56g/1H3kIURbSdDQT6LjjvWhpWBoaQkFA0Gg0ZGel9uq6x0Twi1mh6fr7b29uorDzOlCnTBqKLfeIvuTemo7OG9Nx3OV6zG0EQ8HCOR2GjQS7vvlHfwd6PTl1Dv9pobS/G3t6JfYffQm3nQVjgXBbMep8ftt5NZ2cTNjZ2zJv+H/T6diQSGUcyP2P1hiuRSBSEBVxLiN+VfWovq+C/ODuF0NJWQWHpFozGLtraq5FIFNgqnJkc+wZFFT9xNGstMqkdI8KXDHh2WmNLFsF+07qt/svldnh7xNPQnDWoARdgRMQiUo/9g66uDtQqD5paSwjwupgR4fcNaj8GG5NKOeiZZv1h3rwFXHrpZaddDJNIJMyePZd1676mpaUZBwdNt/PffLOh1+seeeRxHnnk8V7PHTyYislkYvbsub2eP5/85QKu0ahjV+qDhAfPY/bk52hrr2Ln/pfo6KwjLfN/xEReb9mjWVi6DWfH4WeosSd6Qwd1jWm0dzoR7D+Tqto0Uo6+h5MmmLqGLGQyW9xcpvDD1jtpa6/CZDLg7hqDyWTkookfo7BxOnMjv+N4zR7kMhuunb8OhcKsi9ulb+PL9Quoa0zH1SmGsICFhAUs7HPdZ4utwpWGpsJux0RRpLG5ED+PU++XPF/IZSoSRq2gtb2Mjs4qHNTBKBUug96PwaavabYXOldeuZDvvltLUlIi8+dfds71JSXtwMfHl4kTJw9A7/rGkDORPFfKq3bgpAlk7Mh7qWvIZsO2u3FxCiN+xN3kFG5g7c/XUVGVwp6Dr5OR+zWO9uF9biOveDUebtEsnPsVY6LvJCbyekRMRAQv4LpL1zMhdikFJZsZNewWbrlqKzdfuRkP1xEIgoDR2D9hbkEQCPKfZQm2ADZyNRHB86kZJNPMQJ955Bb9ZM6kE0WMxi5S0z8AUcDVaeDmivuKvcoPD5f4v0Sw/TPi7x/IvHmX8v33p0+SOBtaWprZsWMrf//7vYOuowBDZIQrikaq6g7Q0laMgzoAT9dxfVpQ+i2tHaV4uo9EFE0k7n+BmRP+gZ+3eY525LCb+Gn7InbsXU6Q/wxmTPgHu1Jext/roj5lJlXWJjN9wsl5zMMZnzBu1BKGhZqTQ9q11YQGziU00KzjIJPZEhdzF/nFm2htL8VO6dHn+3LWRKHt7Dn90alrxtZmYDPpToVK6cXYEcvZnfoGScaX0Bs6cdJEMGH0y4O+S8HKn4vHH396QOpxcNCwbdvuAamrP1zwvwW6ria277+b3JL/IZG2klfyJdv3342uq6lf9TmoAjlefYiGpgIkghRfr5POA4IgYcSwG7BXezNxzFKC/WegsNHQ1JrXpzYEQYrJdFIRvq4hBz+v8ZbvW9urcP11T+zJawTcXYaj1fVtNfYEsVGPUFqRTE19Rrd2i8uT8PWY0a86+4O7cyyzEj7jstkfMXvCp0yKfa3XXSFWrPwVueBHuMfyPsDHI5ZJ8Y8hCAKiKLI79VWO5b1PbNRjNDRnUFW3H5nUDj+vmdjZnl7MwsdjKtlF/+No9ioMhk7MPlAnF3mMRp1FHFsUTegNHb3alP+eDm01x/I+oK7pCLquZlKOvs8lM95BKpHjYO9LbUOWRQTG3SWK/JLNDA+/1jIKNhh1VFSnEDCqf3OsSlt34mOW8eO2+3BxCkciyKhtyCQ26lGUtoOblCIIAg72vug6h0bGlhUrg8UFP8KtqE4kNuZ2S2ASBIHY6Dsor9rBoYxXOZixAqVSjlGsZdveO6ioTjptfVKpgslx/0ar7aBL30Zm3reWc3qDlsMZnxEaaPYOy8xbh1ymxl4VeNo6swo+Y+ve2+jsOs4l09/ipis20tXVxpfr57Nz/z9o76hh14EVVNWmIYoibi7Dqa3PYMfeZ6lryOZ49UF+3vEALpoRZ8woOx3e7hOZO3kNAV6X4+c5n4snr8HHY+qZL7RixcqgcMGPcEXRhETo3s0THl9NbdlcM/9ry97ZqLAr2bD1Xjxc4pHJTr1VRalwIT7maZpbi9hz5HHyijehsfejuHwnJtFAScVuMnK/QdvZwITRr5xWVau6PpXSyk0IgsC86W+htDVnrlw19wuSDrzE8ao0tLpGEkY/xPY9y9B2NiKVyrFXeVPfWMrmXU8hldjg6zmLUP8zy9WdCZnMDh+PKedcjxUrVgaeCz7gentMJi3rf8RG305pxW5MooHa+ixUdl5Ehy/spk/g5jwMV6cIahoO4e0+8ZR1GgxaMvI/oOT4L+gNWmzkGhzVGqbE/RuJREF901G83Zxwdx5zxsWyssrNhATMprxqryXYgnkkHhG8gJq6HNxUoyk9nsy86W8jldpQULKVI5mfM2P8h9aVcytW/kJc8AE3Ouweduy7i2M5qy2yf8drDqJRh/QqHm4SDWfcwXAg/XlUdhquXfANtjYasgq+52D6hwT7X45S4YLa7uzVg4wmHQ5qb1rbKyk9vpv07K+obchCrfJEY++P2s6P0cMeJrPgv3z3y23oDR14uo5l8pg3rMHWipW/GBd8wJUIMowmHZfO/tCiZ9vUUsK3m27maPZXRAQvsOw9raw5RGNzEe4jTu1V1txaQHNbAQtm/WgZvcZEXEtjUwHF5T8yLOTsrcYBPFzGkVf8Cz4e49iy6wkmxj3K9IQXaGjKJ3H/C3i7TUcqVRATfg8x4ff08ylYsfLnoayslLffXsnRo0dQKm2ZMmU699yzGKXy1NOAS5cuQSqV8sorKzEYDHzyyQf88svPtLa2Eh4ewQMPLCUszLxnvqKinGuv7anPvW3bbhSKkwqAP/+8gX/+83nuvnsRN910KwBvvfU6Bw+m8MknX54XYfILPuAer0nG12t8N/FwR4cAwoPmU1lzjFUbriTYbwZaXRMVVQcYG/PsaWUVW9vLcHOO6jFV4Ok+kvyinX3un7/XbCqqd1Bdm8b40YuJDLkUADulC/Omv80PW+5ieOgdVocBK4PGj4mXo+vqm7rWuaCwcWL+tPVnVVan0/Hoow9SXl7K8OExNDTUs27dGjo7O3nyyWd7vSYlZR/79u3h5ZfNXmX//e+HfP75J7i7exAaGsaRI4d46KH7+eqrb3Bw0FBQkA9AeHhEN3Ga3wbQ3Nxs3nrrjR5tzZt3KWvWrOKnn37g0kv7rxR4Ki74XQomUxdyWU8zQhu5HW5Oo5g8ZiUyqQdO9rHMmfQVHq6nl4h0UAdSU38Mo1Hf7fjx6oNn3I3QGxKJnIRRKxABH8/x3c45a4IRBAmdXf3TY7BipT8MZrDta3vp6WmUl5cyadIU3n//v/zvf6uwsVGwZcsvGI29y5GuWvUlarWa8ePN6zIbN/6ITCbj44//x7vvfsSsWXNoampk3z6zsl9hoTng3nXX/axY8brl34nMss8++5h7773DItH4W0JDwwgICGTNmlV9egZnywUfcD1cx1FcnkiHts5yrEvfRm7Rz3i6JeCgDiQs4BqCfC85KwUoB3UgzpootiQ/QVNLCZ26Zg5nfEZJ+S4Cfeb3q48SiQwHdSC1DZndjre2V2I0dqGQa05xpRUrfy38/QN49tkX+dvfbgZApVKjUCjo6tKh1fYU3GlubiI1dT9xceOQyWSYTCYefPBRHn30KYuWrbPzCc1cczLUiRFucvJOli9/ijVrVmEwnFzv+fDD/8PDw5NZs3p37B43LoHi4kKyszN7PX8uXPDvuWo7b0L9r+GbjTcSFXoFEomcrIL1eLom4NIPYRmA+OhnyCr8lPWb70RvaMfTdRyT497EVtF30ZgObTUlxzciEeTsOrAChUKDn+d4mlvLSNz3PMF+l5+zc4QVK38W3N09uOiikypdW7duprW1BT8/f9Tqnk4iJ5S9IiKGAWb1sN/KKra0tLB9+1YAoqLM8eDECPeEQeW2bZvJy8vh6aefA2DJkke47LIr+OKLz3rt44m2Dh1KJTIy6hzuticXfMAFiAy+EXeXMZRX7UAUjYyOXGqxEu8PUqmC6LC7iQ67+5z6VdNwiANHnyc0cA4hAZOBTrbuegKDUYdcZkeI35VEBt94Tm1YsfJnJSsrgxUrXgTg+utv6rVMUVEBQK9GjzqdjqeeWkpdXS0xMSOJjjZbJoWFReDt7cPixQ8BsGjR3Wza9BPXXXcjISGhXHPN9aftl5+f2Vj1xEh5IBkSARfAWTMMZ82wQW9XFE00txUiIMFBHWRJghBFE4czX2fmhBfx9zHPLQ0Pv4atyU8ik3owPPS2fgvsWLHyZ6ewsIClS5fQ0dHB1KnTWbCgd9fvE9MEvx/9GgwGli17giNHDuHo6MgzzzxvOffccy91Kzt9+kzWrVtDVtYxQkJCz9g3lUrdre2BZMgE3D+CusZ0Dma8jEQiIIomBEFOXPTTODmE09ZRjojRojQG5mSH4eFXk5z6BoJw5x/YcytWLlxqa2t45JHFNDc3Exc3luXLXzplNucJV5bfL6j9618vsWfPLlQqFa+99rZlBKzX6ykrrUDbYSTALwRBArJfTUQNhrPzCDzRl/NhMDnkAq6uq5GKmmRsbVzwdB173rZb6bqa2Jf2DNMTlhPgY06VLSjZTNKBRxgZsQQnTSRGYyeiaET4Tepxl74NqcQ6Z2vFSm+Iosjzzz9DbW0NERHDePXVlcjlNqcs7+RkXlf5rYnkjz9+z88/b0Amk/Gvf/2byMiTb761NQ3cfMtC3Nw8+OLzdQiCDampKQBnPR97Pg0mh1TATUlfwfGanUilNhiMncikShJGrcDFcWAntgHKqrbh55VAoO9J8ZfQwDlkFawnr/QrdLpmlAo3jmZ/xago84qr3qDl4LFP8HCZQEHZeiSCHG/3iQPqG2bFylAmJWU/R44csny/fPnTllHsU089h719951G4eFmE8m6ulrAPNL95JMPAFCr7Vm9+gtWr/4CgIsvvoTYETOYOHEqu3fv5PY7r8PW1paCwnziYid0C8yno6ysFMCSSDGQDJmAm1/yLdX1e1kw6308XKPp1DWRuO9F9h55knlT1g34SFfX1YSDfc+JemdNMH5eCWjsfUna/zIZud+SX7wZR4cAyqv2obbzI6foK4L8pmEwdnIs7/+IjXrMKihjxQqwZ88uy9c5OVnk5GRZvu/q0gHdA+6IEaOQSqUcO3YUgPz8PGpqqgHzqHfXrpPJSlFRMYyOhuXLX+Cdd/7Nrl2JNDY2cvHFl3DbjUsRRTiNDpWFvLxcAGJj4/p7m6dkyATcwvLvSIh9CA/XaABsFY5MT3iOz7+9iMravfh4DKw/kavjCDLy/48xMXcilcgBs2ZtcUUSsyb+Ew/XGFKOfkB06H2YTAa0ujrchk3icNbrXDv/a9Qqc4ZLXUM26zffgZ3SGyeHM0/YW7FyrihsnAY90+xsefDBR3nwwUct38tkktOaSGo0jiQkTCQt7civ28MiSU4+tWVUSz2o7Ox56qllwDIAjAZoru/Zxh133M0dd/TcqZSaeoCQkDCCgwf+93XIBFyDUYuzY/cHoLCxx1bh2G+XhNPh7jKG/FJXftx2PyOH3YAoGknL+gIPlxjcXcxBXy5T0mVow8s1AYlERkbeR0SGXGYJtgCuzpH4eMazK/VBZk/4r9X9wMp552zTbIcKV111LcnJSRw+fJAxY06fSSqRQXurCXtHCYIAogjtrSIS6dmNbisqysnPz+WJJ5YNUO+7M2QCrr2dP8XlO3F3OTlfa84Ua8LBLnDA26us3UtDcza2CnuSDryEwaAjIngBCbEPIQgCdQ3Z1DVkUVP/HBKJjADvi5FKbJBKegpwyOUq3FyGkV+6zipgY8VKH4mPH0d8/DjWr193xoBrZw/tzdBYY0IqB6PeHGhVZ5ns+f333xIYGMzcuf3LOj0TQybghvpfzYH055BKZAT7z6SppYTk1FeQy1SU1+zE3XXMgLWl7azlYMbLXDLjHcsURkV1Kht3LEFh40Cnromcwh+Qy+3w9RxPfVMulbVJqJT+HK+pZETk3ywKZi1tFZQeT2bq2Gc4knl+8rPPRF1jGjlFX9DUmo/azofQgGvxcR98i2grVvrLypX/OatyEgmoHUGvM08lyFUgV5zd6BbgvvuWcN99S86hp6dnyARciVSOVKoku2AD6TmrkctVRIZcTtnxPdQ3HR3QtsqqthPsP8MSbAF8POLw9UqgpGI/bR0VuDpFcMnMd5FJFYiiyP4j75CZtxZP1wRWbbicqNCrMJq6yCn8kXEjF9GhrUN5Br+180FtsmysKgAAIABJREFUQxoH0peTEPsQPp5jqa3PZHfqqxgNWvy9Lxr0/lixcr4RBLA5sw3hH8KQCbhKhTsGQzs3XvFjN5cHT7dRJO59YUDbMhjaUSp77sGzU7phbzeMvJJVjB11P7JfNRIEQWBMzJ2k53yFn+dstLpaSip24ec9kctmf4Suq4WNiQ8SH31+5oVOR07R50wY8zDhQZcAoLZzR2nrxNbkZ/Dzmn1a+yArVqwMLBe8WtgJZDJbpFIbZFJbTCYjFVUp5Bf/gmgyYjB2nlPdJ/YB6vVttLQV4eIYQ37xZvSGk+pFuq5Wisq24+ESDwjYyLunGsqkCgQElLbuJIz6J4JgS3r2l/yw9S42Jz2KkyaYtJy30HU1n1Nf+0pTaz6+v5ON9HAdQaeuHoOxpzqTFStWzh9DaITrBghk5X/H0ewvkEltsVd7U165D1E00alrwlZx6gQDk8lIedUOKmuTEQQpPh4z0OvbyC35gubWYmwVzugN7ajtPOnQ1qFW+fLdL7cRHX41JtHIseyv8XGfhsY+GA+XcaTnrGba+JOCyQUlW7CxccBBHYjJ1EVreykLZr2PndIVtZ0nEomU7XuWU1D6LVGhtw3CEzOjsvOiriHbovcA0NRSjExmZxmhW7FiZXAYMgFXEASkEiV7D68kLuYuRkTeiCAIdOqaWLfxJqrq9hDoM6/Xa0VRZPveZ2lqLmZ4+NWYTHpS019DEGDmhH/g5R5Ldd1RduxdTmz0bfh4jmXjjoewV4VTXJ6CgISo0HvwdB0HmG1/isp2oO1swN97EvWNOeQW/UR44A0IgkBzWyEqpRuebiO79SM8aB77j7x/3p/Vbwnzv4ZdKa8w2/Zl3F2iaGopYfueZwn1X2gV17FiZZAZMgEXzGLkVXXJjIi8wTL3aKtwJH7kvWTnbzxlwK1tPExdQxYL562yjOoy8taSEPswPp7mbSZe7qOZnvACifueJyJ4AZPHPs72Pc8ze0J3zUyDsZPy6h0snPcVFVX7qKnPwl7lxcS4x8jM+55h3IyNXENHZz1Gk96SNAHQ1lGFjdzhfDyaU+LrOQO9QcumnY+gN7QjFeSE+F9FeODpJeqsWLEy8JzzHG5ycjKRkZF8++23A9Gf0+LvNRO53A5B6N5tW4XmtPORNfWphARc1O0VurmltMcI1NNtBM2tJZRUJFNTf4wObXWPuvT6NqRSGzT2vkSFLWTa+GWMibkTL/dYtJ3mfG+1nTcO6iBSj76PyWRWKGppO87B9I8I9L6k3/ffX4J8L+HiSau5eNJq5k75hsjgm3o8QytWBouCgnwWLbqLmTMncs01V7Bhw5kTNZYuXcLjjz/U43hSUiKTJsXx0kvPWY51dXUxdeo4Jk2K6/avpKSYXbsSmTlzIlVVVQN5S2fNOY1wm5ubefLJJy2LTucbF8cYDPpOquuO4uFqFhsWRZHMvG/xcBl3yuskgpyOjtpux5w0wVRUHcBW4YiNjT1ODkEUlm7DRq7mYPoHaOz9kckUHEh/kbjhT1q0GmwVTkgkcmrqM7oZWxaXJ+LkYBbH6OisYXjo3zmW+3/kFM5DbedFU0sRkcE34+nWfQFrsBAEyVlZEFkZ+ty47xea9LpBa89RruCL8b3b1fwenU7H0qVLqKurZeTI0eTn5/LKK//A3t6eadNm9nrN700kT9DR0cG///1qj/IlJUUYjUb8/QMICAi0HLezsyMhYRJ2diref/8dli//x9nf5ABxTgH3ueeeQ6PRUFMz8Km1vSGRyBg57CE27niIYWFX4KD2Ib94M9rOFkaGP9brNV36VsqqtqLramBY2JWWIKmy82RL8hM42gfQ2dWMjVxNe0cto6JuITb6DsCsnfDz9sUUlK0nLGAhAIIgJSrkDn5JepRxoxbh4hhG6fHdHMn8nNiox0g8cB/t2uMISFAonIkJX4RC7oijQxhymWpQntMJRFGkoTmTTl0djg4RqJSeZ77IypBnMINtX9vLysqgpaWZK65YyMMPP86OHVtYtuxJduzYdsqA+3sTyRN8+OH/WYRsfssJp4Zrrrmeyy9f2OP81KnT2bBhPfff/yCurq5n3feB4IwBNz09nWPHjvU4rlKp2LVrF++99x433HBDnxt2cenpX3Q2uLnNI8B3GNkF31NZc5iwoDmEBs495Yr74Yx1eLrHEOI/m593LMbJIQitrpkObQ1XzvkcF6cwRNHEgSPvkp7zFSOH3Wy5ViZVMCbmTvYfeYfxsTcjkUh/7cNCPNy8OZr9FYfbK3FziWL+zHf5ZefDjB5+G8NCr0QQBPKLf2HPoTe4bv46S+ZZ/+657yPTDm0dG3c+hF7fjqNDEIezXics8GImxj06aNMJ/en3hcBQ6XdNjQSZ7OTP8rdfDzZn23ZcXBzbtyfT1dWFTCahocHsaO3oqOm1jhMmktOmzcDW9qRubnZ2Jt9+u4bQ0DDy8/MQBMFy/QlbnmPHjpKWdpigoGCuv/5GlErz/v0JEyayfv06tm3bxA033NyjzbOlP5+TMwbcxMRE3nnnnR7HHRwceOyxxyyOmQaDAZPJhERydg++vr4Nk6l/UxEmk4YunYSyyv0UlG4ht3AbUSF3orEP7lG2tOIAI4ZdQ5DfNPy8xlNZc4hjuWsJD5qLi1MYYH7dHhZ2JcfyvrYE1RPIZHY0t5bxwaqx2Kt8CQ+8gUCfudjKYxgbs8JSrqxiOxr7QIaHX205FhY0l+LyJA4fW0+If/887t3c7Kmt7WnnfCb2HH4WX89xjB15P4Ig0KVvY8PW+0g5spog3wX96ktf6G+//2iGUr9NJpNFaetMqlvnm761LSCXK3jyycdIStqBl5c3f/vbLb3WsX//AUwmE2FhkZbzRqORFSteQqNx5Pbb7+app5YiiqLlfH6+eYS7adPPlnpSUg7w9tvmHUKhoWaN3dTUFK69tv+eg719TiQS4bSDyTNGx8WLF5OTk9Pt34oVK2hpaWHZsmXMnWt24Fy2bBkpKSn97nxfOJz1Oq3aXBbM+j9uvnIzQX4TST74cK+LXAobJ1raKgBz8oSf9wQkEikO6u5at/YqL6QSOYWl2yzHRFHkaPYXBPhM5u6/pTJz4gvklXxJScWmHu3UNR5Fq6vnpx2LOZzxX3Rd5h+Gs2MwWl1tj/Lnk05dI/VNGYyJ+btlN4eNXE38iLsprfxlUPtixUpvmEwmdu1KxGg04ujoSEdHR6/lejORXLfua3Jysli06MFenX4DAgKIjY3jww8/47vvfiYwMJjDhw+ye7dZi9fd3QOFQnFeTCLPRL/eQaZNm8bXX3/N119/bRn93nvvvQwf3j/b8r7Qoa2msnYPc6a8irMmGIWNPTGR1xEWNI+Csp6rnYE+8zmS+TkNTeYfnMlkQG/Qkl34Q7fFvnZtLQZjF7tSVrB9zzKOZn/F+s23Ud+Y++truICn20imJywnt+Srbm0UV2yksnYXEcELGB62kPqmfNb/ciuduiaKyhJx1pz/5/JbDEatOTNP0t26xNbWEb2h9w+2FSuDiSAI/PTTVl5//U2ys7N4+ulHey33exPJmppqPvzwPWJj47rZrf+WJUse4a233mPYsOG4ublzySWXApCZeXJqVKVSnxeTyDPRr0UzZ2dny1RCeXk5AP7+vfvKDzSt7aW4OIZ301MA8PGM42jW2h7lXRyjiQq5gx+23oWd0g2tth6VnQ96Qzu/JD3CsNAr6NDWczjjE8ZE30lE8Hy2JD9JfWMJ9U0Z3H71TmxszPcliiakEhta2koxmYxIJFKMRh3H8t7jstkf4uwYAkCg71S27X6G9ZvvRC61x2uQdyaolF5IJUrKKvfg731yoSErf/1pd3NYsTIYNDY2IpfLUavVTJw4GTc3d0pLS2hqasLRsXu26O9NJFNTD6DVdnDoUCqTJp10ZNi48UcOHz7ImjXfc/x4Be3t7URERAJgYyPvVgf8mkh1Hkwiz8Q5Jz74+vqSk5MzEH05K9QqX+qb8jAYOpHJTkoCHa8+iCj2LsQS6DOX0TGXUVCUho3cHrWdL3pDB6nHVpCc8gpuLlFMinvckv7qoPZFIffFaOqkuv7Yr3O/h9mxdzkgYKd0Y+veW4iLfhpEEZWdhyXYniA8+BKq6zKYEP+vQc/oEgSBkZEPsG33MiJDLsPZMZiiskQaGguZEv/WoPbFipXf8v333/Lqq//k6quv54EHHqG2tpbGxgaUSiUODj0Xln9vIunh4cnkySd9BpuamkhPT8Pd3YP4+PEYjUZuvdWc1LN69XpcXV1JSdkP0M3TrK2tFReXwd2hAEMs0wzMozd351h+SVrK5LFPoVK6klf8MzmFGxBFEy1txSht3Sir2kaHtgpH+3C83Schk9rjrDn5wOUyO2KjlrJ5943ERFyHl3ssANV1xygq28GsCZ+iVLiTuO8FJsQ+yK6Ul5me8DwBPpMRRZGisu3s3P8kE2NfRautw2QydPNVa++oxkEViExqiyiaqK5PobE5C6WtB74e05DJegqVDyQeLnFMG/sfisp/oqApGSdNDCPCH0Muszuv7VqxcjqmTJnGRx+9x9q1q8jNzaa8vAy9Xs8NN9zS64L7700kx4yJ7yZCfuhQKkuW3MOYMfE8/vjTACxYcAVr167ijjtuxN3dnaysTIKDQ5g8eRpgnpbQ6XTnxSTyTAy5gAvg53URadkrWbfxb3Tp2/D2iGP+jP9QULqVlGMvoetqxNM1BlfnSIqPf0du8VdcPueDHvUobDTEhN3LxsSHEBGRSmwwGLTERT+DUuGCu8sYnGuiSdz3Ij6e8QT4mEW7BUEg2H8mWfnf0tB0DLXKn9T0D4iLuRuJRPprVtnHjIh4gNqGNFKP/QODsQOlrTNyuT2ZBR8zKfY1HNSBA/5sOnWN5JV8TW3jQWxk9gR4z8M37C6rDKOVCwInJ2dWrvwPb7+9kqysDJycnLj77vv529963571exPJs+Heexcjl8vZsmUTRUWFTJw4mUceeQKp1PymeT5NIs/EkAy4BqMWN+dhzJn6OiBa9pXWNmTRoa0kfuS9xERcB0Bs9B3s2Ps8h459RKjfnd3q6dBWk1HwEaOibiY0YA6t7ZXsPfRvGlsycXcexc6URbi7DMPRIZDa+kw+X3cR3h5xxI24G0eHADQOAZRX72DsiOUcOLqcnIINqOw8aWopJDL4ZmzkDiQfWsq4UYvw95pAXWMOew69ga/XOA5nvc7U+LfP+Vno9W3o9M3Y2XpgMGpJSl2En9d4po1/ivaOWlKOvk9rewlRoXecc1tWhgaOcsWgZ5r1hdDQMN58812g7yaSvx8Fx8bG9TCVtLGxOa1zQ2rqAaRSKTNmDL4A/5AMuK6OMRzOfI1OXRNKW/McjyiayCvehN6gJSr0KktZQRAYOewGfkl6pEfAzS9bR3jQPEtmmYO9L5fMeJtVP1yBRGKDq3MYdkpnWtrKmTXpZexVnuQV/8x3v9zC9IQXKSlPwmg0oFS4MDX+HVraiujsasTJPhy5XM3eI08xftRiy95cB3tfVHYebEl+HJ2ulU5dA7aKnkLnZ4PRqONI9ptUVO/ExsYek7ELZ00MHq4xTBn3lKWcl/toVv1wBSH+V6GwObV8pZU/D2ebZjtU6IuJ5JkQRZFduxKZNWuOZX54MBmSCiZKWzfcnEezbtONZOWvp7B0Gz8nPoDJZEAQJIiisVt5k2jAaNKTU/QVtQ1HLCufza15+HtP+F3dzmjsA6ipT8HPawLZhT8wb/pbeLmPQq3yZPTw2wkLmse23WZ9BYnk5MKdgzoId+dY5L+Kkze25HbbJQDg4RqNXt8OnJv+xJHsfyORdHHj5T9y0xU/ccmMd2hoOYqDvU+P+3F1iqSxJfec2rNi5Y/ityaS58q+fXtoaKjn73+/dwB61neGZMAFGD1sKZ2dTRSWbSO74HsCvCcxc+JLyKQ2pGV/YSlnMhlJOfoeMqktEmk7aTkr2XvkSYymLuxsvahtyO5Wr96gpaWtDKXCjZr6Y7hoQrFVdLf89PeaiJvLcPQGLS6Op95jq1J6Ut+U1+1YS9txTKIRjTq436PbusY0KqqTaGjOZ/veZ6moTsXVOYKE0Q9SXJbYrazJZKS5rQylwqVfbVmxciGwcuV/ePHFl8+5noSEiWzfvgdPT68B6FXfGZJTCmBW7fL1nEV51RZCAmbT2FxIavoHOKj8OJLxGYUlW/F0H01JRRI6XQt6Qwd1jdlMHf8Mh459QkHpd4T4XcHuQ4/i4hiKn/cEOnWN7Ep5BZlUSXNbHtX1TUgkkh5b0GobsnB0CMDXcxx1DWWn7GOo/9XsTn0NtZ0Hrs6RtLZXsm33UwjIGD28943eZ6KxJYe9ac8wJvoO/LwTqG3IYtvup5gc/wROmmDaOmqoqk3D020kBqOOlLT3UCo80NiHnLlyK1asnFeGbMAFaOsoYfLYp+jSt2Ey6ZkwZin7D7+J0tYZEZHcwg3Y2DhwxZxP0dj7UVC6hc1JSxk3ajHHctYRHngt0eH3kbjvRbr0rYiYkEvtSBjzEC5O4RSWbiM9+0u27n6KKWOfRmnrRHH5TtJzVnHprA8oqUg6rRCMj8dUuvSt/LTjAUymLkwmPa7OsVw8eVW/lcOyCz9j3Mj7LfPCLk7h2Ku82ZWygmC/2ThphrF51xNIJTJ0Xa24OA5n3Ijn+9WWFStWBpYhHXD1hlZcHENxdY5E19XKqh8uY+aEf+D367xsRXUqm5MeRWnrjFRqQ3jQJbS1V5FXvInOrkZyi9eQU/QFoYFzsFd5cDD9Iy676EOcNGYRHFencIxGHRl5a/li/TwEQYKTQyCzJ72Mys6dzLzvGD3s9CPVIN/5BPrMRdfVhFxu3yPdtq80NGcx0/fZbse8PeJoa68iI28t08b+B5XSi9b2MuRye+tUghUrFxBDdg4XwM1pDDmFPwFwJPNTPFxHWIItgI9HHAE+k8kvMQu2iKKJ2vpMauszcXeJorxqI3K5LaOG3Uig71RUdu6WYHuCE04RJpMeQZCi1bVSXL6Tr39ciJfbJFydurtG9IYgSLFVuJxzsAVQKlxpbC7sdqytvRIQmBL/Fmo7XwRBioM60BpsrVi5wBjSI9zwoL+xM+V+6htzqalPJzLk8h5l7JRu1NQdI7n1VVraymhrr+amKzdhIze/0qdlfcHmXY/h4TaCTl0jyamvMjLyBuzV3gA0Nhfi6BDIZbM/prYhky3JT1BbX0TCqJdxtA8d1PsFCPa7kuTUV5k77Q009v5oOxtI3PciIf5X4qAKGPT+WLFi5ewZ0iNcpcKFafHv0dhSRELsQxSXJ6I3nPQ2Mxh15BX9SGXNYVRKN5pbyogfcY8l2AJ4uY0y24ZLbZme8DwCAus23UR9Yy51DTkcOPKORebQ3WU4c6a8TnNbPhp1T+3dwSDA+2L8POfw7aZb+XL9Alb9cAV2tv5Ehdz+h/THihUrZ8+QHuECCAIYjV1EhS2ktiGL7zffQXTEdUgECYczP8VoMnDD5d8jl9lSXJ6Ijby7otmBtP8wMe5RhoWaBcIDfaeisffjhy13YTR1MW7UYvy8EizlXZ3C6dK3YDR2IvsDdAkEQSA88DpC/K5Aq6tFYeNs1UewMmR57LGHSUpK5K233jttqu1rr60gPT2Nzz5bDcCaNV/x7bdrqa+vIzAwiHvvXdLt+s8//4S1a1fT2allypRpLF36lMXxob6+jnfe+TcpKfsAGDs2gSVLHsHR0ZE1a1bx5ZefsmrVt9jZDbwl1pAPuDKpElE00tJWgcrOg4qqAxz4//buPS6qOv/j+IvLDJcBGQdQEFAThJVQUQQFxVtJ6y3NbTfNbNVWcVV0LddLrSUuuaKbm5efFlLh+vttW22bloGSFxJME7TwBgQoICT3uyAwzPn9Mc0oiZqEw0Df5+PBH3znzMz7jPKZ8/iec76f1F3IZQrq6ksZNnAhsh8u6erjEszlzP/Qq+cwTExMkCSJgqJkfj32zRav6dlvCifP/h1zMysU1j1aPFZcdhkLuRIzs4e7+Mz9mJlZYGPtev8NhV+chA9NaLxpuLUz5JYSY3/34DfynDiRwIkTCffdLjc3h88+28/SpdquvQcP7mf79q0olUq8vX349ttzrFy5nL1738fNrTfx8XFERe1CpbKnZ08nDh+OQy63YPXqvwCwbt0azp//Fg8PT9TqJuLj4ygrK2Pbtl2EhExk9+7t7NsXQ2jokgfep/vp1FMKoC08bk6PceCLF6ioymZ8UASPBf0VK8vuyMytUDff1G/r4zWTmhvX2R8/j4vffUhSSiSmpnd29L1RV4yF3A4HpS9ffv1XcvK/pEldT0FRCl8krkLSaNBoGg29q4Lwkxiy2Lb1/e7Wcbc1H374LzQaDY89NgGA2NiDAGzdupNt23bz7LPP09jYwLFjXwDo70jbvv0tYmLex97egcOH46ivr6egIJ/z57+lf39P3nvv/4iJeZ8ePXpy9uwZSkqKUSqVDB3qz4ED/6Whof3Xo+j0BRfAobsvVpYqJozahHMPX1yc/JkyfjcaqZkL6e9z44eCKpcp8Bu0kPKqbC6k/5sreSdQWDqTmLIZtVpbmJvU9SQlR9Lc3IhaU0GTuo6TKVuI+c94TiZvZtjAUFTKfuQXJXTgHgtC57Znz25KSopxd7/3iWeNRsOxY0fw8hqASqW96ub3v3+BF19cjbu7tidh9+7aOzarqirRaDSkp1/GxsaWvn0fwdzcHG9vHxobG8jOzsTGxob1618nNHTpD40nzbGz095JqusAMXx4INXVVXz1VWK773enn1IAqKzJwr334y1uQjAzk9HXdQw5+Um8/+k0XJ0DaW6+SUl5Oj3tB3G9+Bs8H5mEg8qLby7FsPfjCTiovCktv4y1tSNznopDLtdOS3x+fCl+AxfyK3dtq476hnLKyrOArrVIiCAYQnp6Gv/974c8+eRTNDU13bO3WEZGGjU11fruDaAtiDpNTU3Exn4GwIABj1JVVUVjYyMODo76bXQLm5eUFOPjM4jHH7/1d3vx4gWysjKxtlbQp09fALy8tOtmnzt3lnHjHv/5O3ybLnGEa2XpQHnllTvGK6vzGOy1lHEjomhorKKg6CwNDTV8X5SCtZWKK/nHSMv6hGkh7+Co8gbJkubmRmY8sRe5XDthbm3lQMDgpaRlfaJ/3ZKyNBRWvQy2f4LQVTQ3N7Nly0ZsbbsRGrr0vttfvar9u769iaSORqMhIuI1srK+o1cvF8aMGU9jo3YawNz81rGkbh3cH08RFBTk88orf0aSJH7zm98hl2uvk+/duzcAV660f5PJLlFweztPIL/wazKvxiFJGjQaNefT/kVlVQ4WFipsrFyYNiGaEYM3IJfbIJfb4NYrCDen4VTV5PH5sSUM/NUsKmvSMTe3oqGxBkm6tUanrcKZmw0VNDc3ciHjA74vOoubc/t+8wnCL4Gu4+6SJctbbanzY7rWOgrFnf0St26N5OjReCwsLHj11QjkcjlyuXZtXo3m1t+vWq0GwMLi1rq9paUl/OlPSygrK+XRRwcyd+6tpVt1Vyfo3rs9dYkpBbmsG0FDNnHu4t85efYNmpsbf2gSJ+fspddpaKzBf9AiMnOPYm4u5+mJ7+vX0S0qvcinRxZQVJJKk/oGkkbNJ4eeRyazJnDoCh5xG0fG1YPU1pXw3kdjUSkfZZTfVuQy2w7ea0HofHRXJbz++npef329fnzZskXMm7eAF14IbbG9binV2wsowDvvvM3+/R9jbm5ORMRmfHwGAtrpA5lMRk1NtX7bmpoaABwdewJQV3eDl14K4/r1AtzdPdiyZVuLYqxb5Ly1lj8/V5couADdu3kxbngUtXUFfJm8hOFDljLA/Smqa/M5d/Fdvv5mO6ZmFvgOeF5fbEG7Pq2jagCXMv9DX5dgggNeRi6z4XrxOeIT/0xmThx5BUmMDdiFwspZXPMqCD/DoEG+2NreOlj57rsMiooKGThwsH4O9Xa6E2K3H22mpJwhJiYagL/8JZzAwFtrTpuZmeHh0Z+0tMvk5FzF1dWN9PTLyOUW+hN0b7wRSXZ2Fk5OzvzjH/9zx5F2ba22QOtO0rWnLlNwQXtTQHXtFWwUPfH2mEFmziFOpmymf9+JDHl0PmnZn5BTkMBQn3ktTrBpNGokNIwNXI+5mfabrldPP3y95/Ht5RhcnR5DKZY3FISfbeHCxS1+37gxnNjYz1iw4I+t3vjQv7+2iWRJSbF+LDr6LSRJwsrKiqNH4zl6NB7QnkybPv1ppk37DWlpl1m2bBHdunWjpKSYqVOfwsrKiry8HOLj4wCwtLRky5aN+tf94x+X0bt3H/Ly8gDw8Gj/JpNdquAC1NYVYGfjSmNTLUnJm5g2IRqVUvvN5uv9PB9+/gzZufF49P01AOWVWZSWp9PN1k1fbHUcunsiSRK+A/5k8P0QBAHc3T2ws7Pj0qULAFRXV+kbStbX15OY+KV+W90R6ZQp06ioKOejj/5NYeF1QkImsmzZiwB89VWSfpoiJ+cqOTlX9c+fPXsuAFlZGcDDaTLZ5Qquk0MAX6bsI7cgEUd7b32xBTA3t2TQr57lxJmN5BYkIQHXvk9i5LDVnP7mTW7UlaCwvnU5ydVrx7CxdmuXVb4EwVDklpLB7zRrq1dfDefll1+76+OmpqZMmDCRjz/+gOrqKrp1s7ujaWRr5syZx5w58+4YnznzOWbOfO6ez01JOYNSqWxx+Vl76XIF187WHVvr3pw+tx0bRc87Hlc3N6CRmikuu0hdfSn93B5jgMc06hvK+PTIQoL8XsRW0YvMq7F8d/Vzxg5/uwP2QhDaTnub7c/rmWdMZsx4mk8++YgTJxKYMmXaQ32vuro6UlLO8Mwzs1tcWtZeusRlYT82xn8n3Ww8KK3IoKAwWT9ef7OC8+n/y+NBG5n15H5mT/+cmrrrfHM5Bj+fBbg6BXD8VDifHVnE90WXGDd8D90UvTtwTwRB6N3M77prAAAImUlEQVS7L5MmPcmBAz+/ieT9HD4ci4WFJc8+O+ehvL6JpJvQMLCyslo0mof71kWlySRf/Cs9HQZjZankSt4RPB+ZTHDAWv025ZVZfH48jDlPxXH8VDhmpg4M6Pf8Q831IBwdbSkpqenoGA9M5H74CgtzcXLSroFsbm6KWq25zzOMT2fNXVx8jR493O4YNzU1wd7+zmuG9Y8/zFAdraeDP0+M+jc9VcE42nvTpK4nyO+lFtvY2rhQX19GcupbXPv+FP1cn+ygtIIgdHVduuACyMyt6d0rBB/P3+GoGkx2bnyLx7Nz45HJrCmruMZo/+1YyJUdlFQQhK6uy500u5dHPRZw8uwr1NQV4ezoy/Xibzif/i+CfCNRKb07Op4gPCATJElzz87RQvvTfuZte26b/6Vqa2tZuXIl/v7+TJ06lbS0tLa+lMHYK30YPWwbpWV5fHV2B6Xl1xg9bJsotkKnJJdbUllZilrdRAedivlFkSQJtbqJyspSFIq2dYNo80mzlStXcubMGVatWkVUVBRWVlZ88MEHP/n5hjhpdrvOdDLkdiK3YXWm3JIkUVtbRX19LSDdsd5AZ2BqatqpcpuammFlZUPfvi6Uld1o5fF7nzRr05RCc3MzX3zxBXPnzmXy5MkMG9b+d2QIgnBvJiYm2NoqsbVVdqovitt11txtXdjmvgX3woULXLx4scVYdXU1N2/eJDU1lYCAABQKBRs2bMDJyalNIQRBEH4J7julsGPHDnbu3NnqY/3792ft2rXs3r2bjIwMTp48qV/EVxAEQWjpvke4YWFhhIWFtRhrampiyJAhjBo1ipEjR5KXl0dycjLFxcW4uv60TrJiDvenEbkNS+Q2rK6W+6HM4cpkMsaOHcuhQ4fw9fUlNjYWe3t7MaUgCIJwD22+DnfDhg2sW7eOl19+GRcXF7Zv3/5Aiz2Ymhq2lXNHvWd7ELkNS+Q2rK6U+3770mFrKQiCIPzSiFtUBEEQDEQUXEEQBAMRBVcQBMFARMEVBEEwEFFwBUEQDEQUXEEQBAMRBVcQBMFARMEVBEEwEFFwBUEQDKTLF9xTp04xZcoUhgwZwuLFi6msrASgoqKCRYsWMXToUGbPnk1+fn4HJ20pKSmJyZMn4+vry+zZs8nJyQGMPzdAQ0MDkyZNYtmyZfqxzpD7wIEDjBkzhqCgIKKjozs6zn2dOXMGLy8vjh8/DsDJkyd54oknCAgIYNOmTUbVBUKSJCIjIxkxYgTDhw8nIiICjUZj1Jl1NBoNW7ZsYeTIkYwaNYqYmBigbZ93ly64jY2NLFu2DFdXV9atW0dCQgJvvfUWAG+++SapqamEh4dTXFxMeHh4B6e9pbGxkeXLl+Ps7MzmzZvJz8/ntddeA4w7N0Bubi5/+MMfyM7ObjFu7LkLCwt55ZVXGD16NE8//TRbtmwhNTW1o2O1SpIkDh06xNKlS/VjDQ0NrFixAnd3dxYvXsx7771HXFxcB6ZsKTY2lnfffZeFCxcyf/589u3bx6effmrUmXUOHTpEdHQ0oaGhjBw5kr/97W+kp6e3KXuXLrhyuZyDBw8SGRmJt7c3MpkMmUwGQGJiIqNGjWLq1KlMmjSJU6dOoVarOzixVn19PaGhoaxatYqQkBAGDhxIQUEBYNy5AZYvX46FhQX29vYtxo0996lTp2hqamLBggUsWbIEU1NTEhMTOzpWqzIyMli5ciXBwcH6sdTUVKqqqpgzZw5z586lR48eRpXfycmJFStWMH/+fGbPng1Afn6+UWfWmTBhAocPH2bmzJm4ublhYmJCdXV1m7J3ma69rXWmAJg1axaXLl1ixowZuLi4MH/+fACKi4tRKrUt0bt3705TUxPl5eX06NHDKHIvXLgQgEuXLnHixAmmT58OGH/uyMhIvLy8GD9+fItxY8l9N0VFRQAolUosLCxQKBQUFxd3cKrWOTs7c+TIEXJzczl48CDQMj+ASqUyqvx+fn74+fkBsGvXLgDMzMwA482sI5PJ6Nu3LxEREezbt4/nnnuuzZ93lym4CQkJrXammDVrFs7OzkRHR7N+/XrWrFnD22+/jSRJmLS113E7ulfuy5cvM3/+fFQqlX4+1NhzZ2RktLq9seS+G938mzFn1LGzs8POzo7c3Fz9WGfJv3PnTt555x2eeeYZXFxcAOPPrPPb3/6WPn36EBERgY+PD/Dg2btMwW2tM8WNGzeIj49n8ODBBAcHExwczIEDBwBwdHTUn0CrrKxEJpOhUqmMIjdAZmYm8+bNQ6FQsHfvXhwcHADjz303xpL7bhwdHQFtNgsLC+rq6ozm6Pun0OWvqKgAtPvh7e3dkZHuEBUVxY4dO5gxYwbr16/n66+/Bow7M0B2djbZ2dmEhITg5eXFG2+8wZ49e4AHz95lCu7drF69moCAAKZPn86RI0cYOnQoAEFBQRw7doyDBw8SFxdHYGDgAy2g/jCp1WrCwsKorq7mpZde4tq1axQWFuLv72/Uue/F2HOPGDECMzMzoqKiUKlUNDc3M3r06I6O9ZMNHjwYGxsb/vnPf5KZmUlhYWGLOd6OlpyczNatW+nXrx9Tpkzh9OnTqFQqo86sc/78edasWcPatWupra2lvr6eRYsWER4e/uDZpS4uMTFRmjhxouTr6ystWLBAKiwslCRJksrLy6XQ0FBpyJAh0qxZs6Rr1651cNJbEhISJE9PzxY/AQEBkiQZd+7bjRs3TgoLC9P/3hly79+/Xxo7dqwUGBgo7dmzp6Pj3Nfp06clT09P6dixY5IkSVJSUpIUEhIi+fv7Sxs3bpQ0Gk0HJ7xl+fLld/yf3rRpk1Fn1tFoNNKmTZukgIAAKTAwUNqxY4ckSW37vEXHB0EQBAPp0peFCYIgGBNRcAVBEAxEFFxBEAQDEQVXEATBQETBFQRBMBBRcAVBEAxEFFxBEAQDEQVXEATBQP4fJU8cQbfO1woAAAAASUVORK5CYII=\n",
      "text/plain": [
       "<Figure size 360x288 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "fig=plt.figure(figsize=(5,4))\n",
    "size = len(all_HR)\n",
    "\n",
    "plt.scatter(tsne_all_HR[0:size, 0], tsne_all_HR[0:size, 1], \n",
    "            edgecolors=[model_colors[i-1] for i in clusters],\n",
    "            cmap='prism', c=\"w\",  marker=\"o\", label='Z')  \n",
    "\n",
    "plt.tight_layout()\n",
    "plt.legend(handles=handles[0:num_clusters], fontsize=legend_size, loc='best')\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 128,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "r1_f:  (60, 100)\n",
      "r1_cf:  (455, 100)\n",
      "r0_f:  (455, 100)\n",
      "r0_cf:  (60, 100)\n"
     ]
    }
   ],
   "source": [
    "r1_f = np.load(path + 'Test_r1_F.npy')\n",
    "r1_cf = np.load(path + 'Test_r1_CF.npy')\n",
    "\n",
    "r0_f = np.load(path + 'Test_r0_F.npy')\n",
    "r0_cf = np.load(path + 'Test_r0_CF.npy')\n",
    "\n",
    "print(\"r1_f: \", r1_f.shape)\n",
    "print(\"r1_cf: \", r1_cf.shape)\n",
    "\n",
    "print(\"r0_f: \", r0_f.shape)\n",
    "print(\"r0_cf: \", r0_cf.shape)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 129,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[t-SNE] Computing 151 nearest neighbors...\n",
      "[t-SNE] Indexed 515 samples in 0.000s...\n",
      "[t-SNE] Computed neighbors for 515 samples in 0.018s...\n",
      "[t-SNE] Computed conditional probabilities for sample 515 / 515\n",
      "[t-SNE] Mean sigma: 0.029039\n",
      "[t-SNE] KL divergence after 250 iterations with early exaggeration: 64.677704\n",
      "[t-SNE] KL divergence after 1200 iterations: 0.680303\n"
     ]
    }
   ],
   "source": [
    "perp=50\n",
    "tsne = TSNE(random_state=seed, perplexity=perp, verbose=1, n_components=2, init='pca', n_iter=5000)\n",
    "r_factuals = np.concatenate((r1_f, r0_f), axis=0)\n",
    "tsne_factuals = tsne.fit_transform(r_factuals)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 130,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.legend.Legend at 0x133eccbe0>"
      ]
     },
     "execution_count": 130,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAaQAAAEUCAYAAABkhkJAAAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADh0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uMy4yLjEsIGh0dHA6Ly9tYXRwbG90bGliLm9yZy+j8jraAAAgAElEQVR4nOy9d5wdZ3no/51yeq/be1OXLEuWXOSGHWMDxlQnBgd+4dJCCVxMnAuElhtCCyYmEHK5cBMSAjGY2NjGGAwykq0YW72tVqvt/Zyze3o/M/P748grrXdlrXZXWlma7+ejj/bMeed9n5kzM8887/sUQdM0DR0dHR0dnWVGXG4BdHR0dHR0QFdIOjo6OjoXCbpC0tHR0dG5KNAVko6Ojo7ORYGukHR0dHR0Lgp0haSjo6Ojc1GwaIX0wgsv0NHRwfbt2wH45je/SUdHx/S/L37xi4sWUkdHR0fn0kde6I6apvHUU0/x2c9+dsb2gwcPsnXrVt73vvchCAKVlZWLFlJHR0dH59JnwRZSV1cX9913H9u2bZvepmkahw8f5siRI7zvfe/je9/7Hm63e0kE1dHR0dG5tFmwQqqqquLpp5/m7W9/+/S2aDRKbW0t99xzD1//+tfZv38/3/jGN5ZEUB0dHR2dS5sFT9m5XC5cLhcDAwPT27xeLz//+c+nPz/xxBO88MILi5NQR0dHR+eyYMEKaS6GhoZ4/PHHueuuu6iqqqJUKiHL5zbE5GQKVT17er1AwEE4nFyoqMuCLvOFQZf5wvBqlBlenXK/mmUWRQGfzz6vfZZUIcmyzLe//W2OHDnCzTffzK5du/gf/+N/LOUQOjo6OjqXKEsah1RVVcXXvvY1Ojs7+cIXvsDtt9/OBz7wgaUcQkdHR0fnEmXRFtKWLVvo6uqa/nz77bdz++23L7ZbHR0dHZ3LDD1Tg46Ojo7ORcGSriHp6FxsaBrkM1AqgsUO0hxXvKqWiCa6yGTHMBic+NyrMci2Cy+szpKQzaZJpWIoSmnO70MhEVVVL7BUi+NillmSZOx2NxbL4u8ZXSHpXLIUCzDSLaApIkYzjPVoeKpU/NWn2mRyYQ51fweREmazF6PsJDy1m6bau7Bbq8/cuc5FSTabJpmM4nYHMBiMCIIwq40si5RKF+fD/UxcrDJrmkaxWCAWCwMsWinpCknnkmX0hICvQqayQUAQBAo5jeP7FEwWBYcHSqUsR058l7rKjTTWvoaSkqF/+BkMhiIjE0/T3njvnA80nYuXVCqG2x3AaDQttyiXBYIgYDSacLsDxOORRSskfQ1J55Ikn4VSQZhWRgBGs0BVo0g8XP4ciR7A46yjvekN2Kx+XI56VrffTaEYJ1+IUiyllvMQdBaAopQwGIzLLcZlh8FgPOMU6bmgKySdSxJVAYNRmGXhGEwCqlL+O5sP4XI2onFqKkSWTDhsVRRKKQRBupAi6ywRulV74Vmqc64rJJ1LEpMV8jmNTGpm1o+pCQ2Lo7zNIDsplfKkMuNoWlkpaZrKVPwENnM1Btl6weXW0bmc0RWSziWJKEKgTqV7f4mJIZVYWKP3iEIiWsJ7siKK37OeyWgPsfgQ4ckjTMW62X/0/5FKT9JS/5blPQCdy5K9e3dz3XWb+NrXvnTWtl//+t/xrnf98aztH/7w+7juuk1MTkamt1133aY5//3t336eaDTKjTdu5Xe/e3pJj2Uh6E4NOpcs7gAYzSqxsIpSFLA4NBpWnXL9Nps8NFS/gZHQdgYLuyiVMtgsdWxa8xndOrqEsD34L4jp7AUbT7VZSH/03ee83/j4GF/60hfm1XZgoJ/HHnuED3/44zO2f//7/8z+/Xtntd+27YYZnzs7jxKJhGlpacXj8XDNNdv4znf+geuvv/Gc848uJbpC0rlkyWchFQVBAHdAw+4p/306DlsdHY33UlKyiKKMJOoL4pcaF1IZLXS8xx57hH/6p2+RSMTn1f6hh/4DVVV5zWtuBcrK7IEHvspzz+2cs/3f/d3fT/8dCk3wjne8lbVr1/P2t98DwGtecys7dmxnx45nuPnmW85Z/qVCn7LTuSSJR2DomIjJZMDlNhCdkBg+LqDNkUheEAQMslVXRjrLxr/+6/cxGAzcddfZp4pVVeV3v3uajo6VeL0+oDzV99xzO7nzzjcRCARfcf9vf/sfyGazfOQjH0cUyypg8+atiKLIr371+OIPZhHoCknnkkMpaYQGRDqukKlpEQnWiay8UkJEIh45+/4vR4gnkY71IHf1IsRfXSUAdF4dvPOd7+bf//2nrFq15qxtu7o6SSYTdHSsmN7W0NDEgw9+l7/8y08jSWf2Dh0dHWH79qfZtOmqGWM5nU6qqqo5cGAfiqIs7mAWgT5lp7PsKApEhiEVK78f2T0q/hp4hfvqFUnGNKwOEYv91PycIAr4awTCYwLuwNnrbb1EtvMFJiLPknLkkAUTgf1e/FU3oLU3L0w4HZ05mI9l9BJ9fb0A1NTUTm9bvfrsigzgkUceRlVV3vSmt876rq6unpGRYcbHx2b0fSHRLSSdZUXTYOiYgCQaaF8v07ZeRsTA0LG5p9fmgyiCqszeWVVAFOffaW5qiIHYb6jZeifX3vAFNmz9ENkOD2Njv0FIpRcmnI7OIonFogDYbPMrenc6Tz/9FCaTia1br5n1nc1mO9l/bHECLgJdIeksK6kYSJJIwwoBi13AahdoXCkgiiKpBd4XDo9APqcRi5xSPsWCxvigitM3/34iw7uoab6BYHAdoihhMXtZ2fZmou4kytgwqgov5bvUNEjHIRktW3w6OucL7eSb2rkmWx0eHiIUmmD9+o2YTOZZ3wtCWR1I0vKpBX3KTmdZyaXB5RNnRHoLgoDLK5LLlHPOnSuCIFDTptLfWcJqF5GNEJ9UcQVU7O7595MXkjitVTO2ybIZo1TBQMhBIV6+cQ1GjWJBwGoTkGSB8V4Vf52K55XXlnV0FoTH4wVOWUrz5cCBfQAz1p5OJ5VKzuh/OdAVks6yYjBBOjF7Gi2T0rC6Ft6vxQ7N61VSURVVgcbV5bHOSTZnNbFID+7KlWAs3yqFdJLoyCra1ldT2SSDBgPHIBZWad8oIkkCuYxI194SFpuKZMowGT1IrjCJyejF71mnl7bQWRRtbR0AhMOhc9pvYmIcgMbGpjm/HxoaxOl0UVFRuTgBF4E+ZaezrDi9kE6qTAypqIqGqmhMDKpkkioOb7mExNQ4REbLdY3OBVEEpw/cwXNXRgDByq0Mp44w1vUMxbFxUoNd7Nv5Czz+BmpaTUiSgKoIVNSKeCtEYqGyYjVbBYK1EqGxKY73/xuIGSoC7YhSjq6+H5LNLcDVT0fnJC0trbhcLo4cOXRO+0WjZYvK5Zo9TZBOpxgbG2XjxiuXRMaFoltIOsuKKEFdh8rEgMZIT3nazmzXqFuhkZyC0KCIJyAhSTB8XMHhUwnWXRjZzCYvja1/wsj47zne9SyyYMZgvRmv/9QbpKaBKAnYnAL57ClLTzZANP0sja1bqKnYAkDAuwarJcBYeAfNdW++MAehg2qzXPBMDQvljjvewB13vOEV24iiyK233s7DD/8niUQcp3PmVMLPfvbYnPt94hP384lP3D/nd3v27EZVVW699faFCb5E6ApJZ9kxWaB+hYZy0jNOkkApQWhAZMWV8rT7dlWTwNEXFexuBavjwshms1TS0nT39OdUDKZGNTRNQxAEZCNkkiqxyKnCf5qmERnTKIl9VPhmPlyCvnX0Dv5men+d88/L0/hcrMXuzoU3v/mt/Nd//ZQdO57h9a9/46L727FjOzU1tVx77bYlkG7h6FN2OhcNolhO7aNpZW81p3dmLJFsEAhUiySmlk9Gmws0VPo7NbJpjUJOY3KsPOVYyJb/Pr5XRUPBZDZSLM2cZyyVsoiCYZmk17lUqK9v5I477uTRRx9edF+JRJzt25/mve/94LLmsQPdQtJZRkpFmByDTEIkl9FAExBFkI1gtKgY53huCwKwwPikpUAQoLZDIzJcpGufCBrY3CrN6yERA1URsHk1XD4YDa9icPQZ2hrvRBRlNE1lYGQ7bucq3TrSWTT33//pJenH6XTx298+tyR9LRZdIeksC6UiDBwprw+5vCCoUFEnYvdoqAr0HFaIh1Sqm0UsNuHkPhrhEZWKuZ2ELhiSBBUNUNEwc9rH6YXTtWVV4Fr6R55g96F/wmmvJpkewyj7aKy9+cIK/CqjWEqTy09hNLiQJCOiICOKr/yoiiW6CU29gFlaTy4/iSzbkKXZsTY6Fze6QtJZFqIT4PJJ1HeIHHleof0KA2abQDSk4vRqNK2SOL6vxLHdJRz+LKqWIjJWxOiIY7Q2ABf/w0YUZZrr3kg2FyFXmMTn2oLFHFhusS5aNE1lePy3xFPHMRrtTMW70FQVm6UKj3MVNRU3IoqzzeZo/BjjkztpabiNqbCG0Wgnn08AIArmsuOJODvTu87Fh76GpLMs5NICnoCApmkU82B1lC0P2VB2aLA6ABFMFc8Sij4LQonWdRLuyiFODPwERckv9yHMG4vZj8fZoSujszAReQFFS7Bh9Z/RVLeNG676NA2126jwr0WSFQZGfzX3fpN/oL3p9XhdbWVHE8mMyeQin09TyIFSKIcMFAssOB2VzoVBV0g6FwxN00gUC6RKRSSDRi5b9jSTjAXGxyaIJfrI5sfRhCLJGMjGPInsPjZvvYn21fUEK2ppbXwddrufyfiR5T4cnSVmKnGYptrXUCqlsFkCWC0BWupvZSreRUv9HaRzQxSKM7Ota5pGrhDBaW+YsV3ECJSw2gTMtnJaKlUpv+zoXLzoU3Y6F4RQLsMz4WGymoKqadgEI+19DYimOAltB8XOW2no8GEwJRgbGWRysBpbYBSjoQ6DPDOuw+fuYDzUBd6Ny3Q0OucDRcliNDrJ5saQ5XL6GqPBgaIWEEQRq9lHoRjHaDjl8y8IAiaDm1RmDIetenp7SSkiSTIIL7UDo0mgkNdNpIsZXSHpnHdySoknxvu5tbaBDpcHFTgwFWZnsYfcof20tWxCKvoY69FQFDMlYQRsO3H52hkOTc2K2cnmpvQS48tBSUEcDyMUCqhuF5r/3LNNvxI2Sx2R6FGctmoKxSQmo4NItBO7tQqllCOTjWA2zs6zFvBuorv/cdqb7gRAUYoUCnFM5pnyCaI+ZXexoysknfNOVzJKs8vFCnf5YSIBG31BjsemmCpEqalZiSgJ1LaVlU6h4GHvkSNYLdeDJjM8/hw1FVsRRZlEapCx0D6a6962jEd0+SEkUpj2HkLyuREcNtTObpTJMLQ0lz0GloCqwLX0DP2MCv9qRBFGQ7sJTx6jrmobnT0/w+NcgyhYiYWhVACLo7zW6HOvQ9Pg6PGfYjddQy5vAc2OyEzLWimWLaV8VkNRymJLhovD2WFoaJBvfesBDh7cj9ls5oYbbuIDH/gIFsuZsz7cd99HkSSJr3zlAUqlEj/4wf/hqad+STKZpL29g7/4i/toa2sHYGRkmLvvvmtWH7/97XOYTKfyav3yl4/xpS99gfe//8Pce++7AXjwwb9nz54X+cEPfvSKxf+WgkUrpBdeeIF7772X7373u9x0000899xzfPGLXyQajfLmN7+Z+++/X4+5uMxJl4r450in4jdbGENGJYssnko4WiilkEQTgiDQVPtGBsd+xejEbmTZjFIqUlNxCxbTOdSR0Fkcmobx0DHk9R1QV85+Lq5sgd2HkIbGUBpqlmQYizlAW8M9hKf2ksoOk82FEQWR0YkD+FxrcZg20HtQxOYUsdoFwoMqoqxS2w4B73r8nnWMjw9iNvlRlXIJEoNR4Mln7yJfuHDR1Cajh9ff+Mi82+fzeT75yY8xPDzI6tVrmZqa5OGHHyKXy/G//tdn59znxRef5/nnd/HlL/89AP/v/32PH/7wBwSDFbS2trF//14+/vEP8R//8TOcThc9PScAaG/vmJE89XQFc/z4MR588Buzxrrjjjt56KEf88QTv+DOO9807+NaCAtWSJqm8dRTT/HZz546Yfl8no9//ONs2rSJq666ir/7u79j3bp13HHHHUsirM6rk6DZyuH4FFf5K6dfTkqqSm8yzkpnPf3Dv6O14XZEUUZRCgwMP4PHuRooryG01r+NQjGJouQxm7zTdVt0LgxCOoukqlB7WhZoSURsb0Ta17VkCgnAZHRRW3nTnN/1HxGobZHwV5d//+pmgRMHBabGS/iry+tJglD+X5LLlk+pqF1QZQSQL5xbWYhDhw4wPDzIddddz5e//A1SqRR33nkbv/nNU2csSf7jH/8Iu93O1q3XAvDkk48jyzLf//6/4fF4+fznP83TTz/F88/v4o/+6HZ6e8sK6X3v+9Ccxfn+9V+/zw9/+APy+dneq62tbTQ0NPLQQz8+7wppwXd2V1cX9913H9u2ncp9dODAAeLxOPfeey/vfve7CQaD7Ny5c0kE1Xn10mxzoZRUHh/qYySdYiCV4L8GTuCRTKytuZFcLsPuQ9/myPH/5MVD30bASoX/qhl9GA0OLGa/royWBW3ueS3h5HcLRFXLAdLzWdcp5qFUEPBVzaybVdkgkpqa+5oQJTBe/OFq1Nc38NnP/g333POnQLlyq8lkolDIk83OTgobj8fYvfsPbNq0BVmWUVWVj33sk3zyk5+armXk9b5UM6lc5fIlC+nZZ3/P5z73KR566MeUSqdcDr/3vX+ioqKSW265bU4Zt2y5mv7+Xo4dO7p0Bz4HC7aQqqqqePrppxkYGODxxx8HYGJiAgC3u5ze3Ov1EgqdW80On2/+C6WBwAXKsLmEXK4y/5lvHS+ExtkRGUEUBFb5fVwZqEASRKoq7yWTnSSXj2K1+DGbzqGK3nmU+UJzscqs+e2Uuk4gTUURT1pJmqqhdA9gb69HOke5VUVjsFshMqrASe+3ujYZT+DMLxv5rMa4sYDFYpyxBFDMqVisAoGAEYBQSESWl/+l5VxkqK6uorr6ddOff/Obp0gmE9TXN+B2O2f1uX//HlRVZdWqVSe3idx886nsH4lEgu3bnwZg7do1yLI4bSE98kg5991vf/trTpw4zmc/+wUAPv7x+3jjG9/Mv/3bvwAgScKMY1i5ctX02GvWrJnzOERRnHUNn+s1vWCF5HK5cLlcDAwMTG97qbTuYtaMJidTqOrZX5kCAQfhcPKs7S4mLneZO2QnHb5TN9hUJH3at0aggmQRkixuvEDAQSiUoFRKI4oGJGkBxZAuMMt1baiahsDZ71mhpQnTroNIFSMIdivqaAiT286U0w3zlFtRC4yFnmO8T8LtaKVhhYTDWUkmbqBrf5bqVhXLGd5HNQ1KmsBIn4avSjy5TWOwW0U0lRifSBKa3Eup4CSVDiNJFmTJsmzr1wvNJt7ZeYQvfelvAPjjP37ndD+nZyg/caKsXCora2aNk8/n+cu//J+Ew2HWrl3PypVrKZVU2to6qK6u4SMf+TgAH/7w+3nyyce5++530NLSylve8scA089eRdFm9F1TU6750t3dfcZjU1V1xjX80jUtisK8DY0l9bILBMqR6C8VgorFYqxatWoph9DROSuxxABdfY9RUtNoqordWk9d5S3Iuqv4NOPZNP89Nc5EPoNBFOmwedjqq0Q+g8ec5naSu24T0lgY8gW01iasK+ogkpr3mH3Dj2EyOHGbb2LDFgOF0hTJdA8udzvVTRJT4yo1rXPvKwhQ1aQxdFwhNqlhsQnEIioaKjW1JXoGf4bN6kGWgxiNDorFFAW1gMm4eGv7QtHb28N9932UbDbDDTfcxBveMNsrDk5Nw9ntMx/ypVKJv/7rv2L//r243W4+85kvTH/3+c//7Yy2N930Gh5++CE6Ow/T0nKGk34aNpt9xtjniyVVSOvXr8dut/PDH/6Q7u5uxsfHZ6wx6ejMB0XJE010USgmsVkqcdqb5r12lC/E6B58mOb62/C4WlHVIoNjO+kb+QWt9XdfFh6fqlokluimUEphs1Rht9bOOO5YIc8vJwZ4TU09K9xe0sUi28cG2R4e5taK+jN3bDCg1J8KPj2Xc5nOjpNNKdhNryWZg1Rcwu0PUlJy5AtTWOx+SiMicGbLwmyDpnUqiYhKLgeeSrC7IZY4gcEg0970Rnp7TyBLJmTJSDobRlGL85ZxOQmHQ3ziEx8hHo9z5ZVX8bnP/e0Zz+9LM1GKoszY/tWv/i27du3EZrPx9a9/i5qaWgCKxSKjoyMoikJzcwsABkN5irNUmtnHmXhJFkk6v9OhS9q71WrlwQcfpLe3l3/8x3/k3e9+N7ffvrwVCHVeXWRzEY71/Qu5wiAWi4FI7A+cGHwIRS3Ma/9I9CCVgfV43eW8ZpJkpLHmZhQ1QyY3cZ6lX36yuQidvT8gletGNhQZC/+O3qH/QlVPLWAfjEfYGAiy2uNDEgScRiN31DUznEkRTRfOS/DoxFABKX0LLp9MsEZmpEel76iCQbajqDkSUxpGy9mnuSQJPBUQrAOHp2w5pTIjiKKBY70PUSylKRbLU8GyZEJ9FSgkTdP4whc+QzgcoqNjJV/+8t9jNBrP2N7j8QAQi53y5nv88Uf55S8fQ5ZlvvrVb7Jixcrp76LRKd7xjreetL6yKIrC3r0vArBixfxmsFKp5MmxZwcmLyWLtpC2bNlCV1fX9Odrr72Wp556arHd6lxmaJpKNNFFz9DDNNZuo7byWgyyjdrK6zje9ygTkRepDl571n6KpSQ268y3fEEQsFkCFIoJbJbKM+x58aGqJZLpAUpKDoetfkbKnDMxOPYUjbXXU+HfAEBD9Y10nvgZ4al9VPg3AxAv5llhPRXHpWkaoX5oGWtidMrApCjirVbxVizNcRTzUIhX0bxulMqKGrKpspPC4HEVcSSPUrIzOaLQMMezUdPUs1rHsWQ3skGhvfENhCZKKGqBYjaPpkhIy1xwbj68+OIf2L9/7/TnL37xM9N/f+pTn8fhmPm7t7d3ABCJhIGypfSDH/wfAOx2Bz/5yb/zk5/8OwCvfe3ruOGGm9m27QZ27vw973rXH2M2m+nt7WHLlmtmKK5XYmhoEGA60PZ8cfH/WjqXPJqm0jv8KKqawmgwUF1xBenMECajD4s5QE3FVRzreXxeCslirmAqdgKnrWN6m6IUiKeGCHqvX7CMqgqJScgmy9H97sD5dSnO5EL0DT+C1eLFaLAxFn4Gn/sKqgKzY0heolBMUFKSBH3rprcJgkhN5RZ6Bn47rZA8BjODqSTNDhcAY/0amRQ0bNRo9xopZQR6DitIkoLLv/hjScbAYjUzNeokPTVBdb0Xs03GZCtw9MUCNU026leoM87nVLyTicnnKRSiGAwOAp7N+D3rZ01jZXIhRFGjqfYW7LZKIsI4RoOHTDaCIBhAPbOlcbGwa9ep0Jiurk66ujqnPxcKeWCmQlq3bgOSJHH48EEATpzoJhQqW/+xWJSdO38/3XbVqrVAWbF95zv/wM6dvycWi3HbbXfwsY99ct4ydncfB2Djxk3ndnDniK6QLnMUBeJhyKXBYAJ3EAwX+B6OJU8AOVa1vZ2DXT/Aag5gMfmIRDsxGT1oZ4l1yRdiRKIHKRQTmIwekpkB+tSnCfrXUSplGRzdgcvWhsnoWpB8qgL9R6FfiDJijpBOl7BMWLi+PkhTcOkdJTRNpX/kF7Q03ILfU36DLRbTHOz6N2yWapz2xjPsp50MF3r52oMw4xyuc/t5eOQELqORDoeHkSEN39oCHrMRgyhisENdm8TQCRWXf3Hzd6oKkWEBURCpqK0km8lx8PkkgYYJVMFJXbuV2uaZNY6iieOEpnbR3vR6nPY60tlxuvseBzQC3itmtE1lhvF7V+BytJBMjaCqCpqqYjSa0DQNk1HAZPRe8EwN58LHPvbJc1IOLpebq6++lgMH9qOqKh0dK3j22d2vuI/D4eD++z/D/fd/5hXbvec97+c973n/rO27d79AS0sbzc1nd4BYDLpCuowpFWCgU8TuFPEGBDIp6D+sUNOmlusRXSCS6X6C/rWYjE6s5gDj4b1UV2zGZHRQKCYYGX8el2PuqYJUZpj+kV+U1408K4gl+tA0lVQ6QST6MKJoxONcRcCzYcHyTY7DCTlEzp/mzTXNeE1mOiNRHj82wJ84m/Cbl9ZUSmdHMcjmaWUEYDDYqKnYzGT06BkVktHgRBJtRKJHCHjLsSKapjIaegGX/dT5cxqM3FnVzPNT4+wYGqEl00SDxUTAfCq9k8VWDkRdbL34WAjsLolgjYjbJ2Ew2QhUWeja50DVVOpXzF43Ck+9SEv9bbgc5alXu7WK9qY7OXz8IfyeDTOsJINsI5HqxyDbcDvbCU90I0oipaKGIJQzHNx58yNkUxoGUzlY9nQX6lcrb3nL3Tz77A727dvDlVduPq9jjYwMc+LEcf7qr/76vI4DukK6rAmPgDcoUdd2ao7e7oLRfo3G1RcuLbIoGCiVMgA0172Wzp6HmIr1oKpFkukwRtlHc93sm07TNEYmttPWeDs+zwoAAt7VjIV3Eo3FWdn8Z4uSqzsZ42hyivHJIiEpwSdqNlJjK2vqKyuDTMaK7AmHua2ublHjvBxVLSHLs2OnJMk0wznh5QiCQF3VbfQMPMxk9DhWi5+pWDdgoqZ2ZqkOn8nM66oa0TToTYlY8zLCaXo1MaVhsi7+GkjFBKobRexOiE+pyAYBQRDI58BXpWKynjpmVSshiSbyhSj200pJAFgtQRQli6YpCMKpx5bL3sJo6BlCk4emlbCi5CkpOazm0+YbLzHnys2bt7B58xYeeeTh866QHn305zQ2NnP77a8/r+OArpAuazJxkdrmmXeqJygw0CVQKmrIs6tFL4iym6p2xsVpr3s1/SM/x+9ZjcXsY8PK99I3/Bv6h3/PiqZ3nXT7nv1EUZQsxVIcr7tjxvbKwAaGRv91UTLviYY4kYlzfWUNqkPmQH6cRwd6uLdtFQ6DEU2DGouDnlxkUePMhc1azeBYmHQmhM0aBMqWznh4P077Ky9CW81BVjb/GdHEMXK5FAHPNWc8f1D2UvPXqvQeVqhrE7HYBZJRjZEehZr2pXkpUYoakkHA6dMoFTRUVcNs1XAFyspjeGI7iVR3OWuD7AYkEqlBfKf9rqnMKLJsm7Z6XqHhOGcAACAASURBVEIUZZpr38zAyC8ZHN2JVd5KNmfBanFNX29KqRxYe6llnXrggW9fkHH+/M8/yp//+UcvyFi6QrqMEaTZFTRVFTR1aW5eRckzGtpBNHkMTStht9RTHbwBi3nmSrnVHCTovZb9nf+C015NqZQjk41S4dtCJLqPWPIYPvc67NbamfKLMpqmoah5ZOnU632xlEEUzz07g6ZpJFK9TESP8FzKwntWXknQ7iJnFChl/YRLaXaHJ7ipuo5cRiNcyOA1Lf2CmyQaqQ7exOHjP6YysB6jwU5o8jBoJryus3tFSZIJv2f9vMdz+UGUFMYGVIp5AZNVo6ZdO2PWhPmgaRAZhcSkQCauIYhgtZdjiabGNRA0TFboHXoCm83FpuYPIUtmpuLHOdr9M/Ye+j/Y7UFEUcJkcJLNJagO3jSnYrWYA3Q0/Sm5whTRyTiyEKCYA9VQvpZLxfJ03WUQgvaqR1dIlzFOn8pon0rrWhFREtA0jbE+DatTYynKnvSN/AK71cOmtR9ElkyEJg/QM/RT2hruoVhKI4oyFlMAQRDwe9bicbaTzAyhqQr5wrOYTAZ8njXkCwmGxp4k4NmK37N2un9JNOKwNTM4soOmulsQBBFFKdA//Awe57lnCBkL7yKV6cbo3US1DDa5QCLVh9PejD1nJK7mORaNsl6upjcVZ09yhNdVNZy94wXgda3Eag4yGTtCJjOOz3UlLkfreUsu6/CAw1O2ZBdKIVf+Z7JAMgrZhMyGbQLhERjuVjCaBEqlckqa2naVfGGSfDHM2vq3I560fHzuDir8KxkL7aOp9sZyDaz0KInU+KwXktMRBAGLyUdcTGE0lx1RVEUrW12WJSvZpHOe0RXSZYyvEkZ7Sxz6bxGHWyST0tBQqetY/FRNJjtBSUnQ0nDP9EO0MnAlsUQ/B459A4+7DVUtoqoCjdWvx2L2I0km3I5WxsK78LoaaG08lXDSZa/n4LF/x2ldSSoqo5TA6oSa4M0MjD7GnsPfxWYJEE8NU1OxGrf93MqbF4pJpuL7uXLN+8lhYEeiE7utkWSql3wxRr3Hw4nRKY5Ew0SyGXwWE7cEa6kwn790RGaTj5qKhbuqXyhUBUZ7IZcSsdpF0kmVXBrWXStitgrUtUGgRiMyqjHap7Jii4ooQjwZx2LykMuFUbUismTFYHBgMJgJ+FZTGdgCQNB3JUbZRWhqL3WVN59FmrIlJMnlfzqvLvSf7DJGEKGmFfIZlVxGxe4Di31ppjZyhSkctuoZb/SKWkA2yAT9a1jT/k40TSM8dZjeoZ+zsvnPEMXy5ZjODtFYOzPllNUSwCA56T44STBYg9FSLtAmmyy0tL6dbD5EoRAn6L2e2pq66SSPSgny2bJL+yu5s6ezo7icDRgMNgxAjdXO06MDXON1UiqliJaMdMWmeH/raoJm6xlzvl2OhAbBaDTQfo2AKAkoJZHnf6XMuI7MVoGaZhgfPHV9ldQc45HdNDfcgNnkJJePk86Ok0qP43bO9Kp02uuYij03Y5umaSTTA+QLUSzmADbL0tVl0lkedIWkg8nKtLfTUmEx+ZmYfBZVU6anY/L5KPl8HI+rvFgtCAJB31omIgeIp3oBSKUHSWcmyOancNM03Z+qKSTiaVavsBOsKvdX1SjQvU8gHi7hDgaxmoPT7TUNwsMQD5ff0vMZDYtDpaq57Pr7cmTJQj4fn/782ppGnh4d5NtHDmAQwGjwcrW3imrrIhZWLkE0DRKTImuuLisjAEkWcPqLjA1N0djhms62Hp8Cs60cK6VpGhOR56kKbmZw5Fka627CbPIyGtrN4NguWhtmFoJLpUcxGk4lSi2WMvQOPYwkCTjsNYyEDiCJNmyG8+txpnN+0RWSznnBYg5gMgTo7nuU+uobkCUzIxN/IBrvo73pbTPamowOhsZ+jc3qJ+BbiaLV0NXzU+yWIA57HZqmMjD8HELJR7DqVC4tURQI1omMDwm4gzOnGWMhyKVk1mwVMRgFVEVj4JjGxGCRqiZmYbfWMjyeZzy8hwr/RkySxA0+C87wYaoq30DQXoWor4rPQtPK/wyGlz5rDI3tJCHtZTJsZTydIehbh8d8PSO9GhWN5d8pX4ghCAqrWv+Eichejvf+kpKSw2R0YTR46B95mqbaWzEaHUxGjzE88QItdXdPjzsysR2vp4nGmpsRBAFNU+nufwylML+chzoXJ7pC0jlvNNa8nrHwLg4e+xGqWkCS7FjNlcin1ScqKTlGxv+A37uSNe3vQBBEqoNX0TPoYde+r1Dp30S+kERQnXhMd53MRiBAvoBQUqA09yUcD4s0dJSVEYAoCdS2waFdIhUN6qxFbkEQaaq9i/7Rxxke/wMG2UI2F6ex4iY8juo5Rri8KXskjjAZ60MwNBEN+/BWCIyHd5NMD3DFyvdSKthR1BTdA48QyT5Hc/NWbCfLYYmifDITt0ZVcDNVwbJlE42foFAoIuJm39EfoCp5rJZqGqvfiMXkOzm2QiJ1ghUtH532uhMEkbqq6+jrG5hL3Iuenp4TPPDAV+nsPEIwWME99/zpGctPvMR9930USZL4ylcemLF9x45n+NSn7uP221/Ppz/9eQAKhQK33rptVobwH/3oZwwO9vP5z3+aH/3oYSorlzfXo66QdM4boihTU3H99MK8pqn0DD3M0RMPURm4AlUtMjz+PLJsozq4ecZ6U0v97cQTQ9itbVQHmzAbffQeEImHVXzEEDQNVTYQOZLBo4ahrWGGK1WpODvXnGwor5upytxeV2aTl47Ge8nmw6hqEau5Ynpd62KmVCxbhMV8eerVFWBJvCTPhKqW6Bv+BaKUwmKuRbH0cXjvlTS3NzAS30Nbw5so5uzY3RqSbGOt/bUc6Pw3TNYNTMV6UbUiTnszZmOA0dAL1FRcjSAIKEqeobFdeF1r8HvWUR3cxlzxay/Ftb38tyl/nu2Q887nnyJWzJ+/E/Iy3AYT/7517lLgc5HP57nvvo8SiYRZv/4KTpw4zle+8r9xOBzceONr5tznxRef5/nnd/HlL//9jO2ZTIZvfvNrs9oPDPShKAr19Q00NDROb7darVx99XVYrTb++Z//kc997n/PW+7zwcV/t+lcMgiCSHPtm5iMHWJ4dA+iKON3b0ESjs2qW6NpGhoadmvt9JtxVYvKwPNx4lVmTH47U2EVS5VKMD5Ivm8IpeWUC7bFoRINaVQ2nJpmS0bP7n0lCMKMtaiFoqkQC5czFQgCOLwaTt/Sx8Lk0jDUJeINSnj8kIhq9B9SqV+lnrechOOR5zGZjWxY9UFyuRJavUZP33MMDUfIaAmMshurXZteqzObPOQKU3T2fh+3qwFJNHC8/1k8zrWMh44QnjyKxeIjnhjAaWvF5y679petn9knTBRlbNZ6xiN78dg2U8yXnXHGw3sQBd+s9hdSGS1kvM7OIyQScd70prfyP//n/fz2t7/mc5/7FNu3//aMCunHP/4RdrudrVtnJhz+3vf+aTrR6un09JSrzL797X/CXXe9ddb3N9xwE4899ggf+tDH8PuXIKPuAtEVks4FRRRlAt4rZiTJFICR8efxutqQ5bJZE4keQVU1rOZTNRCsVoWV6mESlddS0lQaOgSsTgkh1ozhD4dQWhrQNI1YYoCS+SjHu63k8mvwBexkUxqj/QoVjep5D5DUNBjuFhBFicp6EU3VmBjUyCRKVDUvzRi5fJTxyC5CEyPYbFZMvg34/BvxV4uM9AqEh1Wql2islxNLHmN129umLRdBEGhq2Ewo9iBeUwVZpRu7tGa6/WSsi0IhzpVr3o/DVvaEKxSS7O/8Fxqq7kTVihRLKQL118y7wmul50Y69/VjNY5jd5qITaZR5AKVdecxBft5YsOGjfz61zsonFz/mpoqJ4J1OudOBhyPx9i9+w9cf/1NyKeV1zh2rJOf//whWlra6OnpnrFPb29ZIR06dIB9+/bS1NTM3Xe/A4ulnL9wy5areeSRh/n1r5/knnvuXfJjnC+6QtJZdlyONlKZUfYc/i4eVzP5QpJMdoqm2rtmRuarGpKg4q+RQTxtu8EAinoyS/YTCFIcp7MNQ3OU/rH/y2TkDhy2RmraWFT2gfmSjoOmiLRfISKI5bd8l1/j8H+L5NIqZtvi+i8Uk/QM/SfVFZsJOF6D3ZOif/h35ApxGmtuxl8tEHphdvVVpQRTE5BNCkiShjsItgUkQNc0ZdZ0mSBIoGlU+K6mZ/AJiqUMTnsdyfQoPf1P4vOsnFZGAEajg6rgBmLJLmoqbjhnGaZGfLQ0+/HXJVDVHPWqjcGjr0FT+879gC4CJEnCYrHw13/9V+zYsZ2qqmre+c53zdl2z57dJ7N8n8raoSgKX/val3C53LznPe/nU5+6b8Y+PT09ADz11JOn9fMi3/rWPwNM97Vv325dIelc3giCQG3ljQQKG0imh7CaLTRUNc5evzHIqHYb0vAYnFZKm/4RSj4P0fgxNDJsWvNe8vnyw7gyuIaj3T+nsum981oP0jSVRKqPbC6MyeTBZW8553WkdBy8FS8pozKSJOAJSKQT566QsilInqye4PDCVHofQd9qaiuvIRERcNq9rG6/mxcPfIea4BaUknXWGplSgoGjAg63TE2TQCGnMTag4g4qeM9xHdtpa2EsvAev51TgcmjyIFZLNU57A001b2Ww9wQ9iVEkyYrZ9lospuFZ/YiCjKrNr4T26ZSKkEsLdGyQEaVTU3Q1zRpjoXPu7qJBVVV27nwGRVFwu91kMpk52/X1lZXLSyXKAR5++D/p6urks5/9G+z22W9dDQ0NFIsFPvjBj+D3B/j4xz/Mvn17eO65nVx77TaCwQpMJtP01N5yoUf3nQc0TUFbwI12uWMyuvF71uJ2tJ5RCRRWtKAcOI627xj0j6D94SClwXFKrQ3EUyeoCm6csa/TXo/F7CKdHT/r+CUlx/H+/yASewGDUSWePERX3w8pFJPzkr9YShOa3EM8fZRMOjfr+3zu3BPWhodh9ISI2WLAZJYZPi4yOSrjdjYhCCCbNDJJDVmyYLcGyWQjjPWpOHwzraPoBDjcMk2rRFw+gUCNSPsGickREfUcL9XKwDVMRns41PUTxkIvcrzvF/QP76AmeBMAibEgDvl61l6xkTUbVuCyrGfiRAXZbGy6D0UpMB45iMvRcm6DU3ZKkaTZ+RYNpsUWy1heBEHgiSee5qtf/SbHjnXy6U/PXSMpFiufx5cUTyg0wfe+9102btzEH/3R7XPu89GPfoIHH/wuK1euJhAI8rrX3QnA0aOHp9vYbPbpvpcL3UJaQgrFJIOjTxFNHEEQJDyutdRV3oTR4Fxu0S4ZNJeD7NUbkYbHEccnUR12lI4WkGWgHI8yax9NnZWUU9PKTgeJSQFNFbC5VDLswu2sprn+tdPtB0d3MDKxnabaO19RrkSqj8GxJ/F72nFXqAwfHcHmtlFRXV1e1wpDKq5S0Tj/Y81nIBERWXWVTGJSY3wAREQKkasY6BrCvVnDYhdIxzVy40Ui4QjZMTsWq0LNy1LsZZICNY0zz4HZKmCxC2TTTLtjzweDbKWj8Z1o4hCj4/2YDH5WNF2PLFvIZyCTEFlztYh0MlDW4ZHJZVawe88vqW+qQpJMTEQOYjHV4nhZufl5jW8q/36pODhOW3KaHNdetTnrotEoBoMBu93ONddcRyAQZHBwgFgshts9c12t7GXItAv37t0vkM1m2Lt3N9ddd6qi65NPPs6+fXt46KFHGR0dIZ1O09FRLtNiNBpm9AFlhShJy3sCdYW0RKhqiYPHHkSUVKqC68kXkyRTxzjWO8Satg+8KtyHXzWYTSitDbz8xd7t7GBk4gVqq06V8I7GeygUMtgsVTPajveDkpepaRIRRYiMaoSGfGy5sX2G8qqp2MLw2DdPKrW5b1ZVLTE49itWtb0Vp71cG8lnL3LwxV7G+uLIkg1F0ahtV+fMEnEmklHwVkgkoxqhYWhbb8DmFAiPihw74Kana5iWjlrM9jQnBn6HyVhFfbMds222nSDJGoW8xulea5qmUchpyAu4NEXRQCCwHomZnhOZJLh8p5QRlB90FTUupMht5PMH0LQU1YFbsFtrz1gW45UQBAg2qPQcKlFRJ2G2CcQjGvFJBWvF2fe/EBTygFbOCiLJr+xd+eijP+drX/sSb3vbn/AXf/EJIpEw0egUFosFp3P2m4LHU65IG4tFAaioqGTbtlPrcLFYjEOHDhAMVrB581YUReHd7/4TAH7yk0fw+/28+OIfAFix4tQ6VCqVxOdbPg870BXSkjEa2okkC1y76VOYjI5y5uzQHg4f/0+i8S58ntXLLeIlj9vRRjI9wIsH/wmHrZVcPkpk6hjVwRtntMtnIR0TWXWVSCErUiqCJ6gRmQwQC0vM/dJ+5smgVHYEq8UzrYwAXD4D7ZtyjA7voqLyZsy2c3f5FoRyOZDwsEbjChmbs9yB2SrRusbFkX2DhDLfQEDD5eigo+21nOkF1x2EsX4Vh0fAbC1ndh8f0JAM2pKmjZKNkJ5j1ieXBYvFTiB47ewvF4DDAwaTSnRCJT4lYLJpNKyGyOSSdL9oJKns7l8qaKgKr1j+4vrrb+T//t/v8tOf/pjjx48xPDxIsVjkHe94F+IcJl97ezn1ViQSBuDKKzfPKNK3d+9uPvrRD3DllZu5//5PA/CGN7yJn/70x7znPe8kGAzS2XmU5uYWtm27EShP++Xzedra5q7MfKHQFdISMRU/SmPtDZiM5YqigiBQFbySEwO/IpHu1xXSBUAQBOqrbsVsTbP30M9IZ/qp8K8nne2ls/cwzbV3YTb5yKbA6RHJJUWszvIDGg1iUTuDvf1U1a2ZfnMfDb2Iw9Y8qzDcjHERpqdRZqJhsGQW7Nnn9EH/EQVBE7E6yvKUk8VqePxm7OZmmhrfgywbzmqB25zgrlDofFHDbBMo5svKqLr13FZdpjNlnAG7C0IDKuERAX91uV0yCpNjClUdCQpFAaPBcU5jngmzlZNpoC6+laOXYsBkg0AuXVZKZ4p/83i8PPDAt/nWtx6gs/MIbreb97//Q9xzz5/O2X7dug1IksThwwfnLc8HP/gRDAYDv/nNr+jr6+Xaa7fxiU/8FdLJCOru7uMAbNy46ZW6Oe/oCmmJkCVTOZhTO/1NqPzHUt2AlxOaVs4+kIy+tA5RdlNOZ0cJT+2mUIxhMnoJ+q6aFchaKCSQJY3rNn0ag6Hs0jYR2U//yON0NP0pskFgKqZR3ShgsZ38sQQwyl4Sqb3sP7Ifn7eBZHqUdCZKa/3sQMLTsVmqyeXjxJP9uByNQHnRfjS0m4B764LPgcEEgTqVvkNl68blFynkNCx2jWyqbI0YDJZ5W17eCnD7y6UhJPncEupG48eYmPwDuUIEk9FDwLMJv/+aWe0EEWo7VEZ7NEb7yglXC4U0BfNTDE6MoWoqJoOXusrbMJs88xdgEbgNpgueqeF0JIOAomivGJDd2trGP/zDd+bVv8vl5uqrr+XAgf2oqjrLitq4cRPPPrt7xjaj0fiKlV93734BSZK4+eY/mpcM5wtdIS0RPvd6QpN78XtWYLdVAAKR6DGSqXFWNV+13OK9KigWU4Sj+8lkJ8hF1uK0tVLdUA50nBhSCYdPUDT/ioaabThsNSRSg/QNP0xD9Z3YradiXCYih6it2jqtjACCvvUMjz9PNh/G5gqSTmpEQyp2d/lmTkxpTE1Ac/1NFLQu8vkIDusK6irazmp9iKJMfdUddJ74LzyuJowGO+GpTqzmelyO1kWdE3cAmterDJ8oIogS7qBAJgWDxxX8Nece5CtK5TpS50IscZzxyWdpb3odTnsDqcwo3f1PMBYyYxBnT/GYLNC0RiOf1SgWs/SN/wttTbfhc78NDZWJ8F56hx9mZfP/94qW51Lx8jQ+sixSKs12fpkvmqaRL0whijIG2YqGRrGYQlMlRFxli3vGDnPlm1gcb3nL3Tz77A727dszY7puIWiaxs6dz3DLLbdNr08tF7pCWiL83vVMJTrp7HkUt7OeQjFFKHKE1vo/RpJefdHjF5p8Ic6JwZ8Q8K7A77yacNRJRWs/VlstRoMDp0/k9zueY03rHQT8bQDYrBXIsoWR8V201p/KIK6oxRnnvJwIdJBsLkw8eQKLyU9FEwx2l5icEMuOBppGwwqBgS6RisZ2rI5zm0t32OpZ0fxnRONHCU3up1BKUip1cXwgQlVgG07bwivLunwgyyqToxqj/QIGk4a/VsPpPfu+ZyObCzMZO0RJyWG31uF1rZylgENTL9La8Npp689hq6G96Q30Dv2Cpuq2M07hmSwQyxzB723F7ykvngtIVAU3E4l2EU/24naWf0tVLRFPnqBQTGK1VGG31izI4eFCoCg5BEFAls2oahFBlDGbvGSyYRSlhFIyTFtDqgrFoobR9Mp9niubN29h8+YtPPLIw4tWSM8/v4upqUne+94PLpF0C0dXSEuEJBrpaLyHqfgRUplhJNHByub3YrVcJG4/r4CmlV2MS8VyJoPlqLQ5EflvqoIbqK++nvEBleo6EY/LTCI5gtHgQBA0RGMYu3lm3IrH1Up3/69mbPO6WhgPHcDrakdVixzr/SmFYhynvYpMboCuvuM01byV5KQTm0vDWyFgMAonF/nVBa/5yJKZbC6E0xGgue4dGA0OovFuuvufpKn2zVjNQYqlNPlCFJPRg0Gef4SszQU21+JKjL+cqXgnY+HfU11xJSZjDeHJI0zFD9Fa/zZE8VTAVL4wNSPLAoDNUkm+kELTFAThzBdMsZjC4ZjtuWW1+CmWUgDk8lP0DD2M3erDagkwFv4tkuigqfbOi9I7VVHzKGoBiiBJRkrFFEVAkgwgFsjnZE4m6EBVylOrogRFVSFdKlHSVCRBwCrJmBZxsz3wwLeX5Hiuvvpafve7XUvS12K5+H7tVzGiKOP3rMfvWb/cosybYr6cdw1VxGgRGOtV8VSq+BdQcUHTIBWDVBQQyovy841vSWWHaaq/Dihn5U4nNIwGJyoDqGoJQZDQVDP5UhQLp6Lzs7nJWQ/2yuAGBob3c+T4f6CoeWSDgYaaG7FZqjEZ3QyO/p7R8HYaVr6B8LBK31HxZAJUldr2hSdALZbSJNI9bF73oemidF53O7VVUUKTe5AkA/FkF1aLj0x2Eqe9ldqK1yzLQ1dVS4yGtrO24x5sJ32lA961HOv9GZHoISrkNqSJCAD2oodEagivu216/1RmFJPRcdYpN5ulkqnYQaqDW6YtHlUtEY33UldZzvQwNP5r6muupipwJQCNtSrHen5GaGovlf6Lb7pbUQvIkhmz0VO+WAwa+UKCQjGFyWjBYGA62Pgl77qCqpAoFnAYjBgliaKqkCwW0QCzXmt9mldpGJmOpqnk8lFKpeyi+hk5IRCokllztUTHFRKrr5JJTkoko+cqD4z3wdSohMtrwOkyEOqXCA3Ob39JMk9nRHAHBBJTKtFQCU0tx87EJ0HMXcHQ+K8oFNPkFIVQKkJn/5P4XBtm9iUaaKl7Kw7bSiZj3VT5NuIy1WOSytqxpmIryXQvolxOdtp6hUrLBpVg/dxlKV5OLh9lNPQcw+PbSaT6pz3sCoUEZpNnWhm9hN1axVT8EIoaZ9PaP2fdinexed2HgNz/z957h1l2lWe+vx3OPjnHCqdyV+dWqyWhjAySMEFIQmDAYA022MNge7j2nWsYP/LMZZ5r7IfhGpuxsQ1OMkMwwiJIXAmJJJRDq9XqXNVdOZ+c49573T92dVVXV5c6qBtjibef/qPO2fmsvb61vvV+78ti5tlze0AXgEYzSyq3j2zhMIaxdlG/1ljE6QisBCNYrhcKX4Y8OoLz5SM4HDYcDpXNS11kn/s++eLYcmptmtGJB+npvOGsabWAbxPtts7oxHcpVWYplic5euKb2G1R3M4E7XaFZjtHPLL6G0qSTFfiGorl0Yv7QC4ChBAI00SR7QhOruFJqIoDYeoosoYkrarKn3w8NV3Ha9NwqiqKJOFQVPyaRlVvb8DQfH3ikoTmP/iDP+CBBx5Y+fsTn/gEH/nIRy7FqV6XyBWOsJB5AkVRaOsNvK5+kh23osjn5zfQrIPRloj3SCsdi+aQ6OyTSS+YeIPn/qLUK9CoKmx7g7JSFBnukDj0rMAfNbE7X3n/kG8nk3OPsW3ofdhsTvq2mRx6cQZFCeJyCAxDZ/O2q8hWazy093PIqhPJbJJW+rApPt5m6NBQycxJzB5toZs2grFteHQvvpKGvd0EvYrpsGP6L1zdNFc4wnzmp8QjO7GpPhazPyVXjNDb+XbsWoB6M0e7XV1DqCiWJ9GNOv3JW1bUzBXFTn/yFvYf/Wc6Yzdc8PWcCUII5lOPUygfJRIcptaospD+Kb2dt+N1W/pnsqzR1uvraNxmsUCk5Ea+7RqwW+1JG0jS832TY6M/piLnsNuCRENvoCO2m3T6lWWVJElhMPluUrkXOT7xfSRJIeDdTDRkBSCBQFr+d/p+ggsnHlxKCCGhqHZM0bJmQgJMIUCybRigdWFiP82kSpMVTGv3i056+PeKSxKQDhw4wG233cZdd92FJEn09fVditO85qEbDTK5/VTqM6iKk1BgJxIyi9kn2T78K3hcHRhGk/GZHzCz8Ch9Xbed1/ENHWyatO4lstmlDfXNDKNJqTKBECY+Tx+qanGHK3kIJ9ZW6Ks2iWBMoVo8e0CKBHfRahfYe/Cv8bji1BtZXB0JEoG3oygGdpc1cn65tRkSvdyciOLQfMiyjR/MT/P0XIpkppvuQYV4l418VjB7sI5WTjDfN8Fg1xYwTeRsgaWJJ3C7kmtSZdX6IvniUYQw8HkG8Hn6kSQJ09TJ5F+mVBlDAPnSMa6+7Pdwu6IAdMbewIFj/0yxfIKAbxMh3y6Ojt1Pf/Jm7FqAbP4o80v7sKluHNpaaW275sM0Ghd9hFyqTFCpT3DFjo+uBMBCaYKR8QfYOvCbyLKK0x4FKIRxFwAAIABJREFUbCxl9hGP7EGSJFrtKvmRvfQNXLcSjABw2NH6BhhWhjH6k2c+6StAUex0RK+jI7qeJq7ZvKiKj3TuELGwpbAhhGB+6QX8nlfHULwUMA2BEBL1RgZVcWJTHciKDaNVQRIbN3JZkmiba4OSIQRndnyynoFptjGFjiypK7/jax0XPSAVi0WmpqbI5XI88sgj3Hrrrfzpn/7pxT7NJYNpQCkL7Za1wO/2n3lNQQhBrbGErtdwORMW/dO0UlfnIw+zEXSjwfGpr+P3dtHbdS3NdomZ+R/QbjcZ7H0zHpclhaModgaSb+GFA1+grVfPa6Hc4bYEP+sVSxPtJLILAqdvfSdZLI8xs/gIPm8XsqQyn/4JicgbiQR3IilgtNftgt4W2M9BUFSSJLriNxELX0W9mUZTfWesUzleLfKrQ1tw21df0BvinXzj2DxXb1KIdErYnRL+sIRroMHh47eTK3+H2rGvEvD3Uy7PURzdR9+Vq4yiVPZFMoUX6YhdjqJoLKWfpFA+RjLxy4zPfhvNptHbfS3NVoFWe4HFzF4GeywRS1lWSUR3k82PEfBtojN2I+ncSxwb+x6GUcPl7GYw+R7mU4+TyR9dk5rKFo7hcp6ZTWYYTbLFw9Tqi9hUD+HAznOu27EsHa5c04kFfP047H4q9TlsopdSVsKtv5vxsYdZSO3HYfdTLE+TdA6hSmfo/AwT1Eszjk923Mr49P3ki+O4nBFyhROYpsRA8szmdP9WEMKgpeexqXZUVUYIk1ojiywpmKaEKm/8+7gUlVK7RUCyY5NlDCEotpo4FfUMOosmzVYBMFEUjbZewzCr2FT/hvJVrxVc9IC0tLTE8PAwd955J16vl//23/4bw8PDfOxj//aUwrOhUYPZERmPX8bpkcjOmeQWTbqH14o2ttplJma/C1Ibu+Znen4JW/MtyK1NCEDV2oQ6ywTCvvNuQK0GFLOQz0/h9A0y3H/rynd+Ty9P7v1/0E4baSuKhl3znndAkmWr8HJ0v06iR0FzSuSXTColg95ta7fVjQYzi4+wffh9eN0W46HeyHHg2JfxuLrwh0NMHTGIdK0Wm1aKglLWZKD79DNvDJvqwqZuTJHWhYn9tIhvk2TcJQ+L04L5CYHH1yLSJYhoBpozQH/vh6g0R6jVU/g8/fQZPpqyVazcbldI5Z7l8u2/iV2z1pgSkT3sP/qPzKeeRJJ0tg59cHkGUcbh8DI+/QNq9TQupzVLOkm6ACuwxsJ7iIX3rLnGjugNjM98i2artOwTNMfc4gv0dd25/h71Osen/wWvJ0ossolaPcuJma/Tk3g7Pk/fWZ+hpbu3flQkyyqFRTvtskysW8HtCWDMvxcaGdyBDInIm7G1FMznXkIZ6gHPcuVsqYo5t4Rx3RVnPfeFwOWIsXXgw+SKx6jXy4QDV+H3DPzcdb5tvYaqatgUP0ggSQaKYqfRKCCLGKp94+t1KCpCQL7ZWPOZ6wyEhna7giyrOOyr73mrXaLdrqBpr22h5osekIaHh9esH33ta1/j+eefP+eAFA6fO+c2Gr14CghCCA4/32Zou0K0S1n5bHS/jlGXiPetPqr9R75NsmvHyqLu0X0V6nKKwSsbGKJIZqHF1IhJquNhhoeux+2K0W7X0A3tFa85NWewOKYTTiiIQIxKbpD0jEzPsHVupzNBKNBLrrCfzvjmlf3qjTxIdbo7eyzq6XkgGoWOLpPUnEElB6GYypbLZZRTRsPRqJfF9Djx6CCxSN/K505ngmTn5ZhMkUz24rAZTBzU8fglTAG1smDblU4CkYvXsWyrRjlazXNjx2qUe+Fgmk6vm6EdNsIJhVJOMH5YRwk6MSsNFK2NRw3g90VxVlQkTxNvdxhJkqz7ig0T8J9KTdbo6dzD9PzT9HRdj8tlkRQcjiCN5hx+bxdLuefwOONompd0/iW2DN5BwPdK7dFLLPabzC2+QCr3Ai5HmKsv/wgu5+p5T7aNE1PPkXH0MCINUM/r9HoG2THYQ3bhJwz07TgrkQB5FzMLz5Dsuhxlmb5drszTbNRw1pLsudGO5rCO0TMkOPhsB52RJN7lImHT3Ibx1ItICevaxGIG+zU78STX07cv3jvoJZF49WtpqZSMqp69vZ3LNqej1W6j2XzIspXSFoaKLKlABc0u0F4hIAF4VQ0PGqYQyEgbsjk/8Yl7ePrpZ/jCF77EFVdYUj6a5qVWS6Gqq8rfn/nMn3DgwH6++tX7APiXf/kq//qv95HNZujvH+B3fuf/WNkf4N57/4FvfOPrNBp1brrpTXzyk/esOMZmsxk+//k/5/nnnwEsB9nf//3/i0AgyDe+8TW+/OV7ue++b+N2bzzglWV5XXs43/Zx0QPSsWPH+OEPf8iHP/xhXC4Xuq6vsdk9G7LZCqZ59px6NOo964Lq+aDVgHJRZmCnoF5fXUAJJgRTIzo2t8Vma7YKFEtLDPe9l0ajTaMmqBRUhq/yUiydwONO0D80iF0W5Ap2ntn3/+J1J3E7Y+hmAb97D/Hw+kI2vQ0Th2S2XqnicJmUqy06+lpMHNBx+gw8fol207QUBrIjSOIHxKNbaDTzTM49RtB7BblcA1jvw3Mu8FqDfYQQTM3OUW8sodl8DPZfRjZbI5crY5gy9XprzX6GIVMul3HZy6BA11ZLXFOSINQFbdEinT7/66npbV7Mp5hrVLHLClu9QTZ7g2zXAnxnbozFUoUej5e5apXaCTc3XWkpajfqJr6QTOegxJHnbAQzT2MecqJ1d2PmU+Re3k9r8zDOjFUDUyq1qdWq6+6rVq/RbqkUS2kigdXv7FqCuaWvABKJyGWkckcwDYl6zUW7ebb2aMPvvo6TnIpqBaoVa59T2/P3J6dwhnfzlq5evDaNI4UsD8zl2dpoMjc/e1abbyG6MHQ3z774RaLhbbTbNVLZw/iUd+IKCAzRpn4KOdMXNpk6USV+cmLq8cOVu1FSWZDAuHK3taZ02vt2sd/BiwHTNNepMDx2n0SrcTHSjRspYa+VrtIcgl967yv3YcYG9WSPP/4YTz9tBQVrHcnazlrPFLTbBpIkMTU1yQMPfJvf/d3fR9dNvve97/AXf/FnBAIBtm3bwf79+/j93//P/PM/f51ksodHH32Yv/3bLxAKhYnHE3z/+w9hs2l88pN/BMAf/uEnOHBgP0NDw+h6m0ceeZhMJsvnP//X3HzzW/mrv/o89977j3z0o7+z4T2ZprmmPZxsH7IsnfNE46LPiRuNBn/5l3/JZz7zGb785S8zOjrKW9/61ot9mnNCswYLEzB9TGJpyqq5ebUwjCaq6lpJJ7Qa4PKALAkEBl53J5JkiXZmy48z1HMrO4Z/hV1b/gNX7PhN8qWDFMvj645bKYA/JK/IjjjsYZqtJUIJg2JGUCubjE8+jd0WZUvPR5mbqPLiS99lZn4v8dD1xMKvPp1imjpjM/ezkP4RyFXy5f28ePBLtNolfJ4+MvnjtNqVNc9iKXNwjcmaolj1R97QmdfSDMNi95mvQKBqGDr3z41ht9u4rW+A6zo6OVzN82x2EZ9N473dmwhIGpOFIi5DocfuJxJRcbgFpZxJZsGk3RQ02xW0qxs4lCDq0QVsRQnpDZczIfatEAl8nn5KlXmK5VV+eqNZYDH9Et0db2YxtZ9U9gD6MiNtIfUcIHPVzt+jp/NNXLP7k3TGr2AhvVpYKAQUMzB9VGLykMzSlDXgOBfU9DZLUpA7e3vpcnvwaRrXxDrYGgpzoq6uKVjdCJIk09v5djqib6JWq2OaDob7fg2fJ4lxBrKKoZ/ht7JrGMkOjO6OtQSHf4e4OMHo0p+vVqvxF3/x2TN+125XURT7yuz4vvu+hmma3HyzldJ/6KHvAfC5z/0Vn//83/CBD/wHWq0mP/7xDwD4znfuB+B//a+/5d57v044HOGRRx6mXq8zNzfLgQP72bRpmH/6p69y771fJxaL8+KLz5NOpwgEAuzZcxXf/e63aDYvrSbgRZ8h7d69mz/6oz/ii1/8Ig8++CAf/vCHueuuuy72ac6KagkWTsjEexRcnRLlvGDysEHP1jMzvmx2UGyC3KIg3GH96EIIlqZNPIHV3tNhD9NqVanUFvC4OnC4rSLOcjWL7ZRAlU3lwZalp/t2arX08r4BejqvZzF1EL93YN01nEq2sqluHPYoqfIs6fQEx+ePEI4GuHzXHWial46et3HkOYN4yMB1LrNiIZCKZaRaA+HzIDzrlTWXss/jcDjY3P+rK/eRyj3PzOIPGUzeRSR4JfuP/BMdscuRZZXF9Mu4Hb24nWevohUmLE5BJSejahJ6WxBKmIQ61pNGDhaz9Pn9vLlzmdHlhA6Xm787dojLAlFcqsruYHTlmY3PStTKApdXQnMINE1maVZHckwR7b4cXImV8ahXCCj8hFa7hF3zo8gavZ3v4OiJb+F1x1EUjUJpiljoGrKFg5hC5/Dxr1mMPGGtm11/xT04HadYZ8ev4cCxrwBvBiAzB7WiQueAgqZBdkkwdcSgd5t5VsfYbKtBlzeG3kxjOnzIkoIQEFdbHLZFz3mNUJIkvO4evO5VLw0lBOkZk2pJXrGyaNYFmQWD5JZzOuwFQSpXkRdTSKbAiIURAd+FVx+/hvF3f/c3pNMp+vsHmZgYo9WuWIQGo4Up2mjL6TrTNPnxj3/I5s1bCYWsdvihD32EublZBget4uVg0NKVKhYLmKbJsWNH8Hi89PX1A7Bt2w6eeOIxxsaOk0z28KlPfRqPx7ssiaTi9/tJpZYoFApEozGuvvpannvuaZ5++gne9KZbLtkzuCS077vvvpu77777Uhz6nJGelujdqhKMWg3fH5awaZCZNenatH57SYKOAcHMiEE+bbHOCmkTSTaJ9a1uJ8sqXbE3ceT4fXTGr8JhD1KnysTBGPE+nbqjSSGtklmq4ol4aLcrKMpq52/XfBjm+mJWbxDS0ya18qrVgCKFqOe9eD0yDsdOhrdEV7ZXFIlwh0wlfw4BqdXG/tJhFF1HCvoQx8dpB3y0d25ZUwlaLI+wZeiONYvJ3R1XMzb5OIbRJBF5Ax5XknzxGEiCzuib8biS56Q5lpoBTJUd18rYNIlGTTB20ECxGQSia7dNN+tcHlubBnGpNhIuF5lmnR519YYlCUIdJuOHDXq3KLi8llDq9KiOPTBFu72WAGIKHcNsraytAHidvSTcv0V2qYIQgkjkVprtF9A0meuv+K9Ikky1tsTU3E9YSO9bITOsXoO8MuPS21BYktlxrYJNs56L22+tOeRTJtG1Cjzr4LNplA0bptBI546g2TwYeoPpUonuwOZX3vksUFRIDJiMvqTjDVrGhMWcSaTr7LT8Cz7n1Bz2iRnk/i5QZMzDo7TCQfQtg78ISqfg2LGjfOtb93H77e+i1WoxMTGGhIqhW1p5bkcAw7Da2MjIUcrl0or7K1hrPifRbrd56KEHAdi6dTvFYpFWq0UkstpuTxr/pdMpduzYxS23rArQHjp0kBMnjuNyuent7QNg82ZLi3Dfvhf//QWkf2uYBjQbEoHTUr7BmMT8pAwbFNw5XDCwy6SUNWm1INJtKSOf/t4E/Vuw20Nk8wfIF+YId3UhNULMjBQZOzBOOOalZ7PJkckxsoXjhAPb0Y0mYHUybuf6Wg5FhXifycg+nUBURlYk8ikDf0yg2hK0a+t/KqMN5yKWbDt6AjUSQNpl6eJIhoH27AHMiRmMwVVGm0CsY2dJyIBEo1UglX2OUnXcKm70bMJpj54xGAkBxTQUMxKmIeH0mhRTMjuvk1c6aYdLIrlJYfq4IBBd+3u4FJVss8Gp4wZTCHKNBp7A+ilGMA6SbDBxRNBuQiAsEes1aZJkev5xPO5ObKoTIUym53+K055YEV8VAubGQMbBpu0uJAlSsyZLJ1xc9cbLVmqVPO4ONvXfwezSs8yn9pLsuG55f8Hc0nMrNTONKri80sp9nkQgKrEwJXE2LTq/zU5Cc/Fkrs0bE70oos1M28axSpU7O8+DrrgBvEFw7Tap5E2EgEFbFsfsPOYJnYozDH0J7EGNRitLs5nDYQ/jsF+gimujiTY2jXzL1eCyIp482IP2o2cxCiVE0H+WA7w+YBgGn/3sn+D1+vjoR3+Xv/zLzwFWScdJVp31nlltZ2LCSvl3da1vD6Zp8sd//H9z4sQonZ1d3HTTm8nncwBr1vJP+iCdnoKbm5vlnnv+ACEE7373e9E0K13b02PNtMfHT1ys2z4jXpMBSZKsYjO9vWqUBdYa0tlko2TFctc8G1yOGK6OtSOFWHeEfClHtvA8kwtlGo0ao+MP0tdt0TirtXmq1SrDfR884zF9YXD5TMo5E9OEnq2gOZYJDzMm1aKM2291dPWqILO4np69DoaBmskhXXXjamRVFKRtA9ieO7gmIPncgyykrBqbk4FmMb0fuy3E5NwDdCX2sG3T7RimzszCk4zPfotNvR9YF5SWpqBdV+kekFE1SM8JqiUDWVlbk+5wgb6WSwDADn+Y7y1O0uX20O3y0DZNnlyax6dqhDQrkBhmy5JqUZxIkkQgykpgCwQN0qkKbmU7jWaWFw/+DV53F8XyBJXaIm5XB0fGvkQ0eBUebQ/tusr2a2Rk2bo2T0Amm4tTL7lwn5LZVBU7TnuE+aV9lCoz+Dyd5IuTtNoNhpKW2riqWWmw0xUQmjWBaju3Atib40mezi7wjyMjmAiCNju/HOtZufdXC0UBfwSUiVns84tkY5tZMD24pDbNAw0K6jRK5Cn8gSBzqXlcji56O9923pp7SiaH3BFeCUYA2FTkvk6UpQz6LwISAPff/w1GRo5yzz2fOqNl+ek4aV3udq8nCnzuc5/hRz96FLvdzn//73+Mpmloy1Lj5ikLt7quA2C3r8pcZTJpfu/3fodsNsP27Tv59V//zZXvXC73mnNfKrw2A5IM3rDJ7HFB7xaQFWvNYnbMxB+5tHIkQd8wQd8wU/MPEQ4OEgluJZV7mUYzR62Rwee9DFXdOD+i2qwR/+mfJQaseiG33xICrRRMYr0msgLpWaiVJRTVsuJ2n/qemwJJAKczHW02q9jxFMQjVzM2/U0OjX6VoH+Aai1Fo7WAyzmEKUp0xq5DlleLcfcf/UfKtek11grtJpRzMjuvlVFtyykrn0QlL5gbs36PkyhmBQ73+k46YndyU6ST/29qHAG0TJNOu4u3xHvQ9TqzSz+iXJ1AkiRsqp+u+JvxuLpotcvMLPwAfSpFu21iU7x0x28hHr6S+fRTKIqd66/4Q9yuOLV6mpGJB6hk/QQiW1eCEVij0UjCzfzcFJHE1pXPU9kD+D2D9HffSaE0Sr2eJ+jbTcA7uDKzdListci5MUFnv9X2qkXB4rRB5zkKD9hkhZui3dwQ6cIQJtrFqLQ+HbqBNjFN9cprSM/Y2XGjit0p0V5aYnY2SEX/MJsHFYQwGJn4DouZZ89b4khI0ro2BlifnYto4OsEjz/+GACf/vSn+PSnP7Xy+cc//p/4jd/4LT7ykY+u2f5ketg8jRn0D//wRb7znftRVZU//uP/yY4dOwErPWez2SiXSyvblssn2Z1WZ1OrVfkv/+U/s7Awx+DgEJ/97OfXBKuTJoBnslS/mHhNBiSAeA/Mj7c58LSM0y1Rqwg8QWsR/VLBMJoUysdptUqk8/u5fs9/xWZz4vNaKTrdzPLSofuAN67sYxpWQCnnZIQAT8Ak2m2NtE+FNwju3SaVoolUa9DhTSFyMDHVhS/mIDko0WoIFiYMQt4ygYSMcDnBpmJ4XMizi9Bj3XxNb2OOTdL0ezhVBlRVHGzq/VUWM88yNfssiqKSjL2JEyOzdCWTlHNW0HP5BLIs4fN00Wzm4JSAVK+AN7AajE4i2i1x4qBFX3d6Jco5wfykQXLzmWcN/W4/fS4fpXYLTVFwKipCCE5M34/fm2DL0O+iyHZyhRGOTz7AUM/7mZx7kHh0C4N9H6DZMMkWjnFi8jts7rubWmOBrUPvXhETdTmjbOp7O/v3/RSHfIYVfT1AofIkxyfH8Ht7KFfnSWdHGEy+G1lWCQU2npp2bRLMj+lknpJQbRK6Loj1nL+thSJJKJfIwE6qVJE8LrIFOx19CnbnshK3U6K3Q+PguEyjBk63Sm/nL3Fw5GvnHZDMWBhzZBwlX4Lg8si/3sCcnMPYs+Ni39K/W+zatRuvd3VddGTkGKnUEjt3XrayhnMqThIWTp2t7N37PPfe+/cA/NEf/Q+uvfb6le8URWFoaBNHjx5hcnKC7u4kx44dQdPsDA5ao6Q/+7PPMDZ2gkSigz//8y+sm6lVlssTTpIoLhVeswFJVqB7E7QaJu0mxPvWd/IXE9X6AhOz3yHgS2K3+0E0GZt5mM39d66QBFTFMvQ6CSFgdlTC6VLZcoWEJEN6VjB9TKdvh1g3iJQVCJZmsU/OIvV0sJBzErLVSXYLhNcNmQKh8lGqsyaOaQPTYae1czOtrUPI+w5BrsiiU0ZK5TEyeR4diOBfmOTWRA/KSXZg4RCF8mF6u69BkV3MTO9Hcacx1V7CiWupVwSVAniDJsXyDB2RtcN+mx0atfUpK0OHYMwkvSBoT0rYnSbJLdaMYiNIkoT/FGezeiOFYVboT96y8kzDwS2Uq/PMLv0EVZXpTtyAItuQpBaR4FaK5SmyhUO020XczsSa47udCdCmqJUNMvPSCruykIZiVmb79pspVA6Rzo6h2UJs7r97DctNN00mayXapknS6cGznB9WbdCzRdBuCUwdNOfP3/q90DSoNTDaglMdtyXdQFI1bPZVKSibzW35/5wvbDaa24exP/EicjxskRrmM7QGkgjvBZpOvQbxH//jb6/5+9Of/hQPP/w9fuu3PsaePVeu237TJovckk6nVj77+7//W4QQOJ1OfvSjR/nRjx4FLLLDnXe+hzvueDdHjx7h4x//T/h8PtLpFO9857twOp1MT0/y6KMPA+BwOPjsZ/9k5bgf+9jH6enpZXraKosYGjo/48rzxWs2IJ2E5rD+X0oIIZief5hNfW8lHLRG2wFfD5OzjzF2Yj+R4OX4gpDK78frWp1N1MpgGjJ921YFTruHJOpVhVJWX8c+k6o17JMzyDdfA04HtZcNOpMySjmFDkgvHES7agc1JYRwCWxzs0gvHqJxw5XUr93DwYMvE87LbEp2I11zBR9SZB6YHmNfPs1VoTi60WAp+xS7t30Yhz1AvQKRHXsYmf4nsvlRpud/SiJ6JXqpxej4k0hoeFxrF1YdbhCYLEzKJHqs9Gk5b6lQdG82aRqTNFo5nPYwdmcvJ0vhKnqbTLOOz6ZtuFbSahdxO2PrJGXcrhhLmaOEAj3r1rPczij5wgIOe4RieWqNp0+xPInDESTZJ5gf15kbt/aVZOjaZGJ3OIk7zuzGuVCv8v2lKeIuN05V5Zm5RXb6wlwVWs232jTg57WEx+VA93pw52fJ2pIEIhJSS0cq1ig6FRpVgXOZ7ZnKHljTbs8HZjxCPeRHXsoiCRPj2h5wXmT71NcZBgeH8Pv9HD58EIBSqcihQwcAqNfrPPHET1e2PTmjue22O8jnc3zzm//C4uICb3nL2/j4x/9PAJ5++smVNODk5ASTkxMr+3/wg78OwIkTIwBnDJAXE6/5gHQ2mCaUc9biusMDLu/5j2YbzQySDKFTaLl2uRefdhVzC/tolQY58lIBLTLNQN+qNl2zBr6gvK4T9QUlyiXgtIAkL2WRuuKQzsF8mnAadCmB6HUinZhGTiYgEcVcMi323UASeXoROZOnFQnybMTFb2/bjbTMsJFNnasjIR6cmuGqUJxKdRqfpxuH3ap3ECbYbDIdsT0osotqLc/eg39NrSTjdW1iaPBd665dkqB7WLA4obM0I6EoFjcokqwxtfRNVFXF5+kmnX+GhfRTDCTv4tl8ntFqkQ6Xm0yjTlDVuDXes864zOmIMZf6EYbRQlFWe/p8cRyvq5d8ccJS564bHM/mcCgyrfwYPucAPs8gJ6YeYkC8BZ+nh3J1lvHpR0lEfgm7C/q2m+Tz8xTLYyi2Noa8GTgzq80QJo+mpnlH7wADXmvBrtpu85UTR+l0OHG0ZiiWRxEI/J5NhAPbz0uXzTBblCuTAHjdvev8lS4WWrs24ztwnMmDcCLnIxQyqashThxJ442NkS96KFamyeRGGEy+98JPZLNhdifOvt0lguYQP9PiWM3x6tTb77nnU9xzz6c2/F6WZW699W3cf/83KJWK+Hx+nnxy71mPe/fdv8Hdd//Gus/f//5f4/3v/7VX3Hfv3ucJBAJr6OWXAq/rgNSsw8wxGbdXxuGWSE+bKDaTrk3r02WvBIsuvTY9pTcdhCIeKjr09JSIljykJn8FVV4tldcckF8wEWKtBUS1JDae1U3OI1dryEM9BDsEtZenaOVVVLsCAS+1skBWVtmEkteF1GxiLlNGVdlSKa7UZmnrZQxDJlseJZWt4bCHaeu1lVMpNksNXNfr2O0+BpK/jGkIDjytk0yaKBssb9g0SG4W6C2BYVj3ObPwGMFAL/3dtyBJEkIIxmce5YWp75Nz7+ajW3fiUFRMIfjh/DRPZOa5Nd6z5rh2LYDX1c/RE9+kp+uN2FQXqewB8sVpNvffzcxCg6+8/C0qzn4GvAGylQyTeZn3DfQScfmQpLcyO/88jdYj2LUgndGbV1Qm5pYeo1KfoCN2OQjBXOoRvO5humI3rru/uVoVv92+EowA3DYbV0TjPD39JFe4SnQm3oCExHzqBcrVCfq63nlO9VonFdW9ng5AYmbpB3THbyXouwSpEpsN84ptdNVaFBdM5mouFBv0bQlS1adYSM3gsIcY7lubqqzWF8gWDpEpykgiTtC35efSavwkLBmf1SChqvI6eaF/b7jrrvfw7W9/k8cff4zbbrvjkp6rVquxd+/zvO99HzwvGbgLwc9vK/oZYGFcomtAIdplRZ/uQYkTByVyi/p5WXg77VEMQ5ArHMdmc5JOT6GM7tshAAAgAElEQVTaHJRrR4hH9uBxJfG4oJaXqBQMfMvrgm6/VTk/Py4TX3YrzcwLSnmT/p3rz2PaNeRmE/na3UgODRtgD4eof+txcr4uQpklCHTjOtlP6gZiMYu5pwNNVohoTo4WsvTZ6yiKQtC/gyMLc+xO7CBffoKYeg2NZpls/hhebzdNI0+tKHNi5hk29d1FKSdYmDRx+c+tkFLVrAYmhKBQOc5V/R9b6ZQlSSKZuI59M5/jxsHbVmZDsiTxxkQXX9z/GMdqz9BoLiBJ4HJ0kIhcT7LjLaRy+xid+D6m2cTj6mNT7/tRFQe6/3rqxiTv7w4gmya2UB8zsU38aGGODyS9+Dx9Z1TLrjVSlKrH2bP9t1YsG2KR3bx0+EuE/Ntw2tcu5OobMN8U0aall9kx/MGVDjroH2L/kX+kUpvF635lL6G2XmNm8RF2DL8Pz7KierW2xMGRr+J2dqDZLp6Y8JrrdmmEBmG12siNn+vPuG0m/zKp3HN0xa/E4wkwM/8SueLhFbLHL/CzQU9PH29/++1897v3X/KA9MgjD2G3O/jABy692MHrtgW1W6A3JSKdp9B9ZYlEr8zUMZlI57mPoKzO9S3sO/wF3K4g0eBOcsVF8qUxuk5RMFbVtRpukmTNJJam2iw+JYOwGGzJLeKM9VJyrQbDfbCUQSwXyNirNeThMLqQUIptHKMHkAa6wTAQI5O0Q36L8ADcEOnkgdkTdNtrDMe3MpUZZ6lW5X39m2lWZGYW9tLfdQcHR/8BzWbD6+6gUl2i3ehm9OUSAX8Tf0RaR0s/J5zBEkGSZExh4j5l1CWEyezM9+mq/hhD7WCo7yZUxYVuGEzNfo++rjuIh68kHl6fyx6rlLmhc5iuaMeKWOpmh+Cp1CKpZp34BgyKYnmceGT7Gv8gm+okEtpGqTy+LiB1uzz8JDNLtlEn7LAis26a7E3PMByMr+mYZVklEtpCuTp91oBUKI0QCgyuBCMAtytOJLSVfGnkjPf8s4RhNFnIPMnurb+O0xHC6dTwubdw+PjXyRWPEgmeYRT1cwIhBEIYy6nT1wbl/JOfvOdncp53ves9vOtd7/mZnOt1G5A28g2WpLWacueKtl4mGt7BcN87aLV0wh43Az0pjk8+yJ4dv4vekilmBR2nZV5UDbo2gTCtpNorpgptNsxmExGLINUtVW8Ri8D0IlrQjrF1F42pOdQDxxGShNERwzgldx+1O7kr3sFjkw+z4O6ly+nm1o4e7IqCZA+iGzU0zY/d5mWw9y0oskY8OkS7rXD0xDdx2iuEQrs3uroNIUkSPs8g86nn6elcpbwvpPeiOXo5Usxxo8PS1EnnDpKvLxL2dHDFjo/idASpN7JUaml6u24klX2B/u7bz3geAxNFPn1NS1o2RNt4gCHLKvopqcqV4xkNVHn9rESTFa4PdfD1sRF2hSM4FZXD+Sw2vUVcW69+3WgWUNXous9Ph2nqZ6xRs6l2Wq1zVGe9hKjU5vC4Ejgdq3MpSZKIR3axmPr5DEhCCNp6FcOowXKq2BRuZMl1TinUX+Bni9fGUOECsCKmurQafYQQpGZMPMHzzy8Xysfpir8BWVYxKSErClK7H0PXGDl0mKMv6HQOqGuUI06FJJ+9VtDoiGHOLUG+iPC6l6neOYx0HjMRAVXBGOyhefVuWm+4DCO5VrlUCHCYIQZMhWsDGpeHYyuWytnCCC5HB+XqDB53nFh4F+HgFmw217Lp3A4qtemNLg2wZp0L43Biv8z4AZnMnEWMAOiM3cRi+hBHjt/H7OLTHD7+DRbTR7iy520czKR5ZHaK48U8+2efB9cQHs2F02E5cDrsYRAGbleCRiuz4fn7XT72Z9MYwkSYFlFlplyh1GxuODsCCPo2k84dpVpbpdFWaotk88cJbLB2s8UX4vaOfpoNnVS5xpX+GHf27KZUniVXGF1hLeWLY+QKY4R8W894nFPh8/STyR1F11ctRAyjSSp7BJ+n/6z7XwwIYQkTL07A4qTFBD0JRbGj67V1luu6XkeRf37ohKden27UMEUTpzOC2xnD7YxgGC10vfpveIWvPZzeJi4Ur98ZEtDRL5geMShmBU63JaZqCpPkOWhYCiGo1ucoV2dRFQfCbGOaOrXGIpHgFmRZo90EydSYXzzM0KYQHb0DF+QNtAK7RvOyrWjPHUR22EEIzFab5u7t65UYTkOjCvNjMrKk4DbuZN/j8/RszhOO+cgXjzO39CIue4LJ/LdptNKoikZf9y04nVYxbbNdXtF/OxMMHaaPyEQ6FJJDEkYb5iZM5sd1uoZAs3nZ0v8h8qVjVKs5PM5NJOObkWWV93T5OFzK8lIqhaPVJumNM1PcT7NVWnFxRZKo1Oax2za2id7sCzG+UOKLLx2i1xagZrY5mM5wlZJE6DJsoLSt2bx0xW7mwMhXCHiTCCEoVmZJxm99RXXtoM3BLq2DahEUAwhDX9cdnJh6CFV5DCQJXW/S23k7suyklLWCttNj/T99gO50RPB7trL/6D/REduDJEkspF7C4xxYV0N1qbA0BfXy8rqqgMUJE2/IINoNbmcnut4mnTtILLwLgGaryOzi83THVy1mDKNJqTqFxKVlCZ4JiqLSbrdW5HJ0o47DHkBecfRVcNj9VGsZbLZf1EJdLLTbLZSz6bKdAyRxsULbRcLP2qDP0C3LcL0JTi94AmenfQthMjX/EM12mkhwM81WibmlF3DaA+zY/G6CfqtYNF8cY2T8ISLBbTSbBpdtf9vFMTQzTaRi2UpB+L0YCISwGHTr0Gojj03TGM3jDqpomxJIgz3kcjWOvphFiz+O2x2iXB0nEd1Od+I6iuUxqvV5UtkRrtnz25RKOQ6OfI3ejnfidp1Z6iK7AEZLZWD76jqRaQgOPqPTvXktCaLezDKfepxKbQpF1gj6ttMRvR5ZVlnK7qXZmsPtilGtz7Kp7x3Iso3ZxWdZTB2mp+M2vO6NRUaLGcFoqomRKOF1qmzxhShN2ahWdLqHz2KaZjQpVSZAknDZ47SNKnab/4wdlxAwPwbthkIoLtNqQD5l0DFo4vYLao0lQOByxGk1ZGZHZBwuGadHopgxsTksKaGTP9nJ9iyEoFKbIV8aQQL83k143b1r0kt6C1otnYY+gSnqeFxJHPaNA/W5ol6BhTGFbW9QVpQ22i3B4ed0klus37DezDIx+100m4bHEySTnSQWumbFjytfGmV26Qf4Pd0IBOXKPN2JtxDwnqNu0qu9h3qVcjlPIBDFZtOoN1O4nfE1z0+WJUrleZyO+M80bSeEpUtpLhNtFdX6fy6X8PPKDBRC0G63KBTSeL1BnM7VAdyFGPS97gPShSBbOEyxcogdwx9YWcDOFcd49qX/ic/TSW/3G2k2yxRKkwz330m+eIJmEy7b/taLes1Vvc0TmTmm65ZtQpfTww3hTgIn1Q0MA/uzL0EwhN7Zg9djYB4dx7TZ4MrtTI+YmLSx+cbJFJ5j15YPIUmSVQdTnWJs6vsYZgPTUElEbnzFNYL5ExCO2wh3rA2K44cM7B6dkw7h7XaFkcn/TW/X9cTCu2jrdabmfkyzZTDQfTumqTM++22gjSRBvjhKo1XA7xmmp+OtKz5SptkmXxql1S7iciTwefqQJJnpoxL9m504fav0etMQvPyUTv/Os/sRCWEyt/QY+fIRXI4w9UYOr3uQZOKWNWSFYgZKGZXNl1vK7ADlguDEAZ3B3eaa9OvUEYlop0qs+6RFuGDsgInm0gkv8xeiUS9LS2WEubEAsGlYKdFCIUNF+jYedwivz0e5PobfM0xX/E2vqoNNzYBds9E1uPY3nB4xEXKbcMfqM6rU5vD5VdqNwMq6V6tdZnTyf7Nz8wdWJJoqtQUOjnydLf2/fs5eTq8W9XqVSqWAYei0dcvY7lS7ESF02nrjZ3Y91jmtYCTLkmWGKMDQBZJ8dsFnsGqPTteu+3mBoqh4PIE1wQguLCC9rlN2F4pi5TidsSvXdFAh/yCx8GWUyguU8m2c9iS7hm9FSG0W0wcZ6L64LBVTCB5cGGdzKMw7+4dQJIn9uTQPLIzz/uQwmqygzKdQvW5aO7YiCRkpICGHdiMefhJRqqA5XVTKYLYLy063yz5MskbAu4loeJ5WK000+Mtr1giEsEbThm4VEisqqHZrvSF8ygRKCEGtLPCcQlLLFA4QC2+jI2YpICiKnU19t7P34BdoNLM47GEGk++mWBmnUp2lOz5EKLAFzbaqrdVo5hib+Ve8njgeV5xM4VmWss8x2PNuDN2B5ljbKcuKhKpKGDpnDUip7Iu0jQxX7vxtbKoTw2gyOvkgC+mn6IrftLJdOScR75ZWghGANyDhcErUK+Bevtx2C9oNiWjn2hF6oldmalQm3Gmit+HEwTZLs1Yg0JyCeK9Yp323OAGaXUWLPcSe5M2E/TsoZk0Ue52Rqa9QKB9/VfVKssyZHWWNtaxPSZLxupNEgmsHhfnSCJHQ5pVgBOBxdRAJbqJQGiUauvyCr+184HS6VzrHcnWW6YXv0dt1w3JB9BwL6aeIBW/G6+6lXrEK1DXnhRXFnwsMo8nU5AiNWo2hbV0EfYNIkiX4fPAZnb4d5oZryyfx82gXfynwuiU1XAro9QiOyseopQcpLSV48fEMz7/wbaLBq3A6Imc/wHlgqlbCYbNxY7wLu6KgyjJXRuJ0ejyMlgsAyKUyUiKMqkGrYaX1JEVBioUQ+RK5JRO3z6qjKpQm1y1MlisLxCI71gSjRg3GD8ikpxXKGZXxl2VySxCMQWbRILdkIoTAaJvMjeiAvqZjbTRzeD1rXepkWcXr7qLRzAJWhxfwDtGd+CUS0TesCUYA03M/xSffile8B49yIzuGP4THHWQp8zxOr0l2ce1IsloSGMYrFBufglzpIP3JW7Atj/otZfNbyZUOrX0+G7Ax1310ks15WkcnyRbhQwiYO26JsO66XuXym1Q6e1VmR2Xap8jHGTpUijLh7hSSLIiGtqOo4PbJmG0n3YlryJeOnf0GXwG+MGQXDerV1buolQWFtInvHCyRLJbg+oesKA4M89xZgifVUwppSz3+1cDr7qav6w7S2QmOHP8WS5njbNv0HjzOfmaOSaQmFYyWSmZGYeqIdM5W8+eKeiPNsYl7MaUZ4t0yc0tPcOTE1zDMNqpNwhuQqb/248w54xczpAtAwDPMfOoFQoHhlVlSJjtOM7ObPdclcPtMWu0i1ZKdycPvJeS14n67XWNuyVo7kWWNkH87If/2lZlJttlgulpA0vP0eYL4nRvnuIutFh2u9SmHDpebYtViaZl2O6JUtWYwmqCQEbjcApGuMFOOgsvEEwDoRpacHJ98kO7EdciyykJqL7V6gUhoK7nsMsVcwNxxme4BhVDCUpdo1GRG9uk43Sbdwybzk4KpfU3kWh0vRTYp8wg1gr55AGQZuxakXJkjGtq+cs2mqVOuzhENXnfWZ1/KNyjN7iI+PIzDLZNfMknPCRJDVzM6+W2Gum8gNW5Sr5sEo5Yz7cKkQTRpbjj6rVcgtwCtpkylNITZ44NTSHmazYtptpbrWKzf2xcSLM2aBGISyvIsqZQTtBpijYOvqoGiCvIpQSguLT9HQWrWUp9vVEEYMr2bFRoNK5CGEhLVEhRSlvI7WAFJVUFSzGXx2OXZ7HJtmyzbEEI/6/N7JWgOiCZNju3V8YUs9fly3iTef/ZUJ4DfO8DU/HfpTlyHukyA0fUGmdxR+rvefU7XUK9YbczplrFpVuF4ILb6HC4EbmcH/d3vPOU6vUyPlHG61BUdSSEEM8dlUtNtOgcv/FynY2bxUfqTN+HRLsNmkxn03MTRE99iIfUCXfFradQEvrNXBLxu8IuAdAEI+rdQqk6w7/CXiIS20GyWWJxp0Nn9TrwBq/DOYQ/jiEI5blLMtgnEmuw/+k2cWifDA2+j3a4xPf8k9UaGrvhNPJVZYDJ/mG79CHabixfSVQI2Lzt733VGx86w3cGzhcU1qtpCCCbLJYYcllSD0Z3AfGYfSiSIqytGq9am+fIU9ZyOfIWXRLjGQnof1fo0kmSjUs1zcOTrCGHi8wwy1PO+5dy7FZBqZbBp0pp1IodLIt6jUEiZdAzAQHARW2Ee+eadKIEYNHoQLx6B45PomweIBHcxOvlVnM4wsdBO2nqNqbnHcNg7LXr3K0AISE9r9G/V6N2kIEky8aTM+CGD3IKKhGVTv/1qG6OHKsyOSaiaoGOQDW3eKwVYnJDpGlBweSU8i9sZeanBtitcuH3Wc80VRnHaY2tStN4QVIsGh5+FYEym1RSUciadQ2sDnyRBol8wdSqbMyPQdYOeLZbwrMsrrRt4uLyQXVr922ZfLqpux2i1q5Qqs/g83TQbAkUxmFnah8/96nvSQNSyQKkUrPuI9rKhRNTpcDlieF2DvHz0XjqilyOAhdQ+/J4t55QhEALmTsj0blEJRk+SKmSO7TVweY21Pl+vEpWcbCnsn6Ic0tkPLz8pIwY2HrycD1rtMm29RDS0E9OAakFgs8t0d1zN6PjDqM1rEOL8bUley/hFQLoASJJMb+fbqdYXqFSnsds66Y5vRVPOUNSoQatt2Tq4nTH6u9+28p3Pk2Tvwb+m4RxmrrbAHnWMndvvxufpptRqcnD6KcZmvsW2wQ+vE+fsdnpQ8zIPzU5yTTSBIsu8mFmi3GwyFF1+cx12Gru3ox05jvLSURRTQMCHdssOfLYmx6f+haAvyUDPmyxzu/mnCPl30RG95oz3beqss+YGy+vPNC2LZdv0PLbdQxBwr1yDtGcrth88g76pD83mZTD5HubTTzA+9QMUxU7Qt52+zjNL1ZyKdhMQKoGITK2RWVb+loh1y+x/Pk+ob9PyM5eIJeFsduEAmVmJvq0qgYh1X0l3EKQFjh0sMrRLplydZ27xeXo63km53WK0UqBlGPS4vHT2u2nWDEr5BnUxgj2RpqYHcRhb11CdnR7o32lSzJhUK+CLWP5Wkmypo2dmTUxj7bWWcgL7KbM0SbJmL+OHZIKJX+blw/cR9u1AEj7KrcNIqIQT29es7zk9Z18zOxNUG+uU5s8VXfE3Ua5NUyiNIiHRFbt1nSL8RqiVQbNLK8EIrN8y0SuTz1zcgCTM9YH2UnjPieU2qKjg9ApKWZNqQyezYKJ6LObnL+pzV/GLgHSBkCQJj6sTj8uiSTU0mD9h0NkvoajLZmeGILdkEu+HpeICvcm1pmSq6sDn7eZofowhW43O8G58Huvl9Wl2YtE9LJWPUyyPU2ssUSgfwTTbeN19JCLX846OPl7ILfGNsREMIeh3+bi9Y3DF2whABH00r90DzZb1xmlWD5XNvIDf28FQ3ztWtvV7ennx8JeIBs/sauv0wuKkSashrxAHhBBkFwVuv/XiSY0WnO5147BbI1HdAE3G6YgwmHzXOs+ksz9zaxTtdnRRrk3QapVRVSe5Qh3dqBP0vIH0LDQKOqZy5lqfUyFMaNQk/KdMzFTFSWdPN/NjGSZmfohmCzLQ/SvM6zZ+OnecrcEIbk3jidw8YZuDm0IhUpXvYlc6CPqjNPQZjk08z2DyV3D8/+y9SZRc133m+btviHmeh4ycM5FIjARIkaJEUgMtybIsSrIG25Jd5apydfXp3vTCpxfeeOeFN150nTptu901dEmyZcmWLMmUKNkaKFIcRJAghkTOc0ZGZMxzxBt68YBMJDITSCQBTuZ3Ti7wEO+9+268uP/7n77PHkTrXs+FdC3tp2ifpWt1A3YnONwG069pRNImirrDZ5iZsMrpex3LcPnCoNgMSpsDONpfptR9HZc/TzR4Hr93lG5bYm1GIMsSqk2QXTAIJYztSr43A0IIfO6BXSrCh4VlJPZ+YZK8s+G5V3D7DfLrJqmhnfttrZu4fJaB0PUOueIF6s0FhFAI+iYJ+Sfv6n21qV5sSpDNrVdJRM+h2kGx6axM/5J03zh9qbdVgfPbAu9Kg9RsbdJoZVEkL1pjkHZdRpZN/LGd6qf9YBgajVYWSZJxORJ39fI53OD0GUz9SieesQxCbs3A7tZxekBtuGm2CvhuSvuYpkmrXQS5D2EUcdh3s1sLAXZ7gNXsj/H7k5wY+yKK4iBXeI3Z5b/h2ODv8YFIig/ciQlWCHDsbk5stNdIx8/sOmazefG4kjTbm/uSkCoqhJLWMyb6JRSb9SPu9XR81yMyut+Dks3D8E28bcUKhqKAuvt1u3l+C502l6sFWrpG3O5i0h/aQ2Cq2kG1mxSzKtG+Y3R7ZXpaj/JGmGAgwdqUTDgpIyTYXJBxeHQSQ7cxSsLauXZau4UCtY6K1xdluP9LAPQMg59mr/LFkWMkrldvvS+a4KuzU7xwYZmw+Sn8sQj1NXD7z5GOv8Ra7l9IBT/H6rREKC4TCFmS7YubOv3HzV2eS2oEek3B3CUNQ7cWy9ggLF2RCEQkvD5BpWBSylrn9o2ZgBfYyblZ4S5BckAhkrJCUd22xLVXdBwe/bbv/dsFLi9szBm0GpbKM1i/ka01E0/o3i7e0QwsX9Vp1U18IUG9YlIpGGQmTAxDY3b5G3jcYYb7P4Smd1jdeI5mK0sm+dG7uk8m+THmV75JsTyL2xWhUJpFktxk+s7f0+d5t+BdZZBMU2dx7Xu0uzkC3kGy2av0ej9nYviLCMPHxoJOMGHsSw5q0f7/EKfDj6730HSdwdRv4HIenkk0MQj1kk4hZ9XOBmJWrkEICAdPs5b7e5z2ND5PP6aps5p9DoGNkcAIr+UKRApXSEStDv2mptHo1GlU5xBC4tjQZ7dzGJnkY7Q7ZQrlS8Qj+wvI3QmK7KLdLd8yfwadbhlFOZhmJ5wEp1unvKVj6AK33yQR2Ql3aCMDKL96HdkwIBaGcg3j0gzdsYMtw3y9ws8K65yLxhi2OZiulvjm6iyfTY/s0URKDpusXNMp5gROt4/yloHNadCuS0w+pOBwCZxOhUBM5+rLVp7HKtzYCyHAHzVYvmYwdMLyKrodk5VpA390p1JvvVUn6nRtGyOwmpDHOim69joPPRrCZlMxdJP5yzpG/RyN5k/IVkwGjlmNswDRNCxfE2yt9UgM3jQOCdJDCjZPa/vY4mVB/5i8na+LZcztc+P7OB/tBkhC2jZGADbH9XBX7p1hkCQZov0G117RiKZlVBsUsiYm+i4v9l5AtcPg9TBquWgVdAydsjYohfIUDruL8aGntucy4B3k5df/C53ueey2A16ofeC0hzk+/AeUqzN0ezWS0Q/jcWXe49E7AO8qg5QrvIIk65w/+b/QbSlkohLl9nNslX/A5OiX8AYFV1/S8EeMXWGTTrfCSvYHnBj/Il532gpDlaeYW/oHjg//+0PT6gthGSDvPiWyTnuYY8O/yetT/4gkgaZ3cdriDKWfQlHcLDTGeCX/fer6/yQUPk25XaVTfA2nLY7L5dszBr93gHxh7shzFQ6cZmn9H/B7B/G4EhimzurGL5CEC6f99gkEl8/62y+EYvo8tB88hTK/gjy3iuGwo02OY0T2ZxIwTJNnC+t8ZnCUPrcV6psIhHh6dZFXy3keCe9mhrA5YPi0Qb1sMRakxywPx2aTcLh2fuSyIoimJKqlgw0SWCG0zSWNS7+0DFKvY+KPGgTjOtXG6nUyPj/6PjXeWlElOtpDki0mad3skBwyuPSCjmZX0Doy7ohBodO13g3FRrRPMPOqBBzc5Kj1oNcRhBI3MdELQbQPZi/uf66hW/nKWxc6Rb19uMs0rdxNs2p91hc5fBHD/UAgCk63QXnLoFkHf3RnU3evIcsQ2me/2WiuEg5N7JpLWbbj9wwwO30R2RjA5VFIZZIo6p0TT5KkEgpM3suhv2vxrjJI5doUY0Mfs9ibewK3V5D2P8zy+nP0tCYOlwuHW9Bu3FhQLRQrV4hHTuF1W/0xQggiweNk8xeo1ucPJNi8W4QCo0wM/Ts63SKK7NhFSfNkfJB175eZyb1IcflZwnYP/ZFHURUfq5vfxzSNXYUN9UYWm3r4nRpYSqT1xgpCSHhcGRKRJ7h07evYVBc9rYXdFmbokEJy+6HTraEbLRzuEOaZ4xympaPc66DI8rYxIldATM3z4a0Sy0YP+biB3p/atSIJYRUF3EC3vVvW4wZM884LmRCWZxvtM+h1rZ1zs73K1fnv4bBbG4FaI0fVGGWpnmbAY704bV2j1Owy6rJRq6/hcoZQZAd2l0qzVaAjSujdNvN1jZDTgYnJXKOCv+O6Y/b8Rq7MNOBmxQ5zn6bVG3B6YH3WpN00tw3zjfyey3eAMTJgdUbQ6/YwHK9QLy7QXVTJ9B8nkZh4y3bxdhfcos34pkJRPLTbpV3Hel2dzewGXtt5EqkwxXyL1345z/HzKVyugyMK7+Hu8K4ySKapI12nCBHCKioQwiIUNQ3d4l3q7NUa0vUWLtdut0bXTIQWp93qwj0IdxiGxvzyj1lcfQnD6OJypEjFHsd9nbxUCEHaHSA99LFbnslEkX3ML/+A/tTjyLKdfPESucIVxgd3ZIc1rclW6SKtTg5V8RIJntlVLl6qTLGa+zFedxLD0FjeeJqB1KeYHPlD2t0ismTHbjtaGdPWRoeV+U30noSktsH5HAPDlnT3nWATEh1dQzMMlGIFceEK8gPH0UOnceUK2K+u0NF19OGDVyh3ADYXDRoVCbd/h4Mtt6qTGD7cM9zgFdP1Dovr32Fi5CmCPuvkenOD7pX/j+/MBenzJ3GrKjOVEsPuPqRanIbxyjaJZ35dQ6g5fLGzXF3OMl4eIR1WrE2Ozcmzl6sk/QqmKVNvrlBvrmNTPQRD53aNxemxepUSA9eNi2GSXTZw+wx0bS/djCRDpM8KdyX6ZVS7Fe7qtHViB9QXZJegvKXRUr9OxBtjcvDD6HqTmbln0UWevsTj+594B5imScfQUSUZ+R0YmgoHTjKz/FWC/mH83kFM02Bu4WcoNo2H3v8IkiTRPwyzlzeZu7rIqfPveT/3CvKf/Mmf/MlbPYib0ZKKxVUAACAASURBVGp1D6VH5HbbaTa7u451ulWarVWC/jEkySp/rTavUm9u0NNaTE8/S70ziz9sx27b2WIbpk6h9CrxiJXk31wxWbis06oJutVh2nUJd+CNlYUurf8TQnQZH/wsg+kPoSgyCyv/hN87tt1EuB+EEPg9o5Qqc8yv/oDVjefodJr0pz65bXC6vRrTS1/D5XQTDR8D0WNp4xmc9jh2m59Ot8zi+nc4fewr9CUeIR45g8cdZ275O0SCZ7Hb/LvGoGlNOt0KsmTD43HumeebUc5BdrXMsTNOTj2YIZ6I09oaYbP4HF6fZw/Lwq2wyTIrzRp1XaN/bh1lbAC9L85Kq0HA68WZiCK9OoU2kD7Q3ZEki25n6apJs271Fi1c6eGLGQTukiCjXJ1GlnUyyZ0ydJvqxdAaDKgCpz2FbAreF4wzGHKzPN1F1yTcrjTFrMTWmoOT5weY13TsnibuUoL8ukm9YrI+Y5LXaxiROo2tH1JtXMPrCdHsbLC8/hMc9j56LTelTetRC1mL+LdZg6Upg1LOQNclSptW46zLtzu85vSAw21V6NUrJi6vQXxgd1XfDbQbsHJVItR/FX+szeT45zA6ATzuKD7XcVY2v0fQf/y2TN37/QanayWe3lziQiXPa5U8bU0n7fS8rXIm+437ZiiyA4ctwsLq02S3XmFl4znWlnN84NGv4LBbnrwQ4PbYmb9aIjkQvO/Pd6cxvx1xY8xCCFyuw8mTvKs8pHjkfcwu/y2XZ75GKDBKpbzFyspVhOihmkP0pz6MZCuzkv1nOt1z29xafs8whdJFLk9/Dbf8fmr5COnjeZxOJ067g+Vpk+xCj/QRCYs73QqN1jLnTv8fdDuWtY2FT9PulMgXL9CX+PBtz5dlO/2pT5AxP4ZpGnvySdn88ySiJxlIf8g6EDyOx5VkfuUnHBv8PYqVKWKRk7hdse1zAr4h/N4MldrcdnzbMHqsZJ+h1ljApnro9hpo5kewyRMHji2/1iM91iKeHLNCgX4YPenk8isfYqv0i0P1oHw0luH72UVia2u4xtPUykWCqh2/agcVhCRZzVyOgxdHbxBcZwxqRQO3z0b/pHEoqqBboRsdbOpeBgxVddHrtZi4mUPHBn3H21y9/Bqq+CQOl50TDwnsTkFvq4NsExw7L1EvW9RN4ZRgudSgXFsgoEqcHv0P22HYamOKiy9N4VMTRFIypmEVZCBp6AZ0OhLHzlsFEqZhbZpWrmkMndzdx+Ly3mgCvv2urrAOqWGFrn2NWGQCu0NCUaCUM5AlJ35vP43Wxl3Jpi81qrxQ3uTTAyOkXG6qvS5Pry7yy8IGj96pEvRtBp9niOPuf0+7UwQUmqsV7MruELmmgckbY8d4D7txX7jsvv3tb/PEE0/w6KOP8ld/9Vf34xb7QpEdjA/8Lj7XcSrVAjabj1T6GIND55kYf5JkOkM8eooTY19is/A8xnV+LSEkhvo+jcc1zvpSg0hfDZ83gtuVQpIFmTFBsyKhH/Hd63RLuF2xXYzDAB53ik6vRK2xwtzKt7gy95fMr3yLWmN13+sIIe1bYFFvLRML72biDvpH0bQamt7CMLrY9qmcU1UXurFDFraa/TGqKvHQ6f+Ncyf/I2eO/z5rmy9Rrk7vOx7TtBZaj0/sym+5/QJDcx9aBM2tqHw+PUokECJYbTHq9u8I6jWamIax3T91O8iKVdkYz8hHMkZg6fdslaZ3ieQZhka+cAWvZ2/sy+32EkhUsUWeJzV83RhpLRz1S0xV27R0DW/QYrfo2rpMV8oE9GXS8fftmjOXcgKpN8qxB01SQxLpEatqsFWT0HXrmcIJGSEsQtfEgEBRJBqVoz1npyURjICpuWhdz5fIivWdGoZJp1e8azbs1ypbfDiZIe22PCK/zc4nM0NcrZfovU2Zqm8HIayeOacjgNNbZ36qtM1naBgmC1NlXIHGHu/IkiKBamH/3OZ7OBj33EPKZrP88R//MZ/97GcJBoP82Z/9GQ899BBnzpy588n3AJKkEApMEsLa9c8sfY1M5CO7Yu5ORxi7zUerU9gWPpMkhUjwNFWnRCCoYLPtrtiSFfaN3R8GDnuYenMTXd/tcldrKxi6zkr2ewxlPorXnaZaX2Fx9bv0xT+xby/QfpAlO91eA6djpzZWNzrb3pTXM0A2/xNS8Ye3DVpPa1EoTTOS+W3r83qHSn2Gh07/79thGqcjzFDmw8wuPL9vYYcQ4HApVEs6QX8XWbbc8lrJxJByuJzpPeccBCEEtrFhHJemkBxOiIehUse4cJVuf+qettFrhkFV6+KSlT1l5Q57iID3OK9N/TfS8Qev8/q9gk0N43Xtn8fqT36ChdVvW8quhh9tbZ0R9yBSIs7/O32ZE8EwumlyuVjg4WAcW2mv91LKG0RTEoqy8382hyAYlShvGaRvyYUJIXB5xC4C1ruBajfRNJOQ/wxzK1/F6xrG7eij09Zpms9jmAK38+68mprWJebcvfHxqjZUSaaja6hvI1XZu8XoiUEu/2qeQq5FIOSitNWkrS1x4vyOmqdpWhIetS2JQFSi1YbcskF67D16oMPinhuk559/nl6vxx/+4R8Sj8f5y7/8S37+85+/aQbpViiym3aniN+7s5gYhkanV913B+j0GJTzJi7PjkFqVCymbPWIwpc21YvPNczlmb8jFXscu+ojX7xENn8RWbIzOvhJQn6L9sZhDyLLdpZWn7utQTJNK39T2pTo1j/D5dw6x8/ECEYcloDg2k/xuoeRJRteVz9FNcLFqf9GInYO09BY23yZoO/UtrCbprdQFMcetmanI3xbTyeSltiYDyPEEtF4mHbNxsylGl3lRaKhj9/VPJnhAO3JcdTLs8jPv4rhsNPrT6MP7DZsN+hx2g2rVNkTPJy9Mk2TC+U8F/NFAvUgckfB65J5aMSH27NzgVTscar1DIXSFKZpEvKdI+AbxzB7FEtXaLTWUGQn4cBpnI4INtXL+OCX0S+/inOziDpwCmFIpK6tMTbUx1TPRBGC30qN4LfZ2dTGWdt8Eb93YCdkV1/CMKU9ORtDt6rOylsmkdQOs4Vh3ODOu6sp3kYoYbI6pzN8IsRw5uNcmvoWWk/FMFt4fQGG+z5z13mRsM3JUr1KyL7zDm21W+iGgfMoHEZvIzgcbs4+cpzNzUWa9XXC/S5isbNIkkK3V2Or9BrVogaNc5x6xI/dbiXuSnmJ5Wsaw2fuDT/eux333CBtblqMkIFAALvdjtvtJpfL3evbHBrh4BmW15/B607jckYxDI3F1X/GZU/tGx8Pp6wObkMHf9hii15fuD1b9GGQSf4are5FLl37G3Sjg0MeJWz/HdYrf03Au5sUM+gbYar997e9XnED6iWZ4RMyTneY7LrOpZdX8CSuoUmLqEqAwbTFcCyEYCD1SSq1WQrFGYSQScU+uovexab6MAyDenMDz02qsIXiFC7n/iqxAP4ICOFhY8Fg4VIZ5AreaJ2TQx9DvU2D7UEwoiE60dCBNduGYYkB9toy/ohErWCSXzH2qNLuh6laiflSg0ebk/QNqbj8sJBrcOmSxslJ23bzqBACv3d4WwwQLA9yZvlvcDsDxKPHaXdKzK/+HenYRwj4xpGqddylLtInngS75QnIo/3EfvIi3g8+tCvkGA2eZWF1mVev/D+Eg+M0W1vU2xVE5Ut02ib267RMzZrFHjB4ClamdJanBdH0daG+BQOby8BxQFStXobihqDXEdicJpH0bn0lt9/ixlu4qqF1h7DxH/AHc0T77DjsR6u2PB+M8d3sApIQDHv95Nst/nl9mfOB2Duy2u5WyLJC6pYdQKtTYG75G8QikzjFaYJDEo32LLI8hKI4CUYF6/OWRtZBBL/3GjdUh8vVa5gY+D1j+DxDb6vCkoNwzw3SjRjrUR/+sMqCYIlW3fkzJ3G5ekwv/i2q4qLbq+PzZjh3+vPbmjd7x2Aw9UqPtVkD04RYWiXdr+5qvDwaHqc//RgrMzpb6zpuj4QwghRyBZLp1DZxabWex+eLHPh8hmGycqXHiYcUnG5rhz16LIPX1WZ9JcT4qcdwu/Y2t8Zi54Bze47fgCl9jLnlbzPY9wQuZ5RCaZps/gJnJ/8NTsfBcx2NwsjEm/Nr21jScToNzrxfQZKs+cou6ZRyBn0P3rTo7zN33yss8YA5xOAxB4l+awfrj8hcsJeo5m0MjhzsAi+vXSQcSnJ85LPb73YiNsGVmb9jZOgs5sYGHBtEDtz0/jptaP0JXN0WUnp3W0Es9vuUq4sWa7f3ONHjx9laU5h7VSMQFZiGRTU08YCDUFwmFrM2RqvTlhptNKkS73dsz8HNKGR1yhs6g+Mybp9EpWiwNqsxdla9zkZ/Y45g+JiJoVuVeELc/Xd48zxH8RIMufjFxjovl3J4VRsf7R9kMnSPaRauo1pbZWn9WWr1dRx2P+nE+4hHDlY1PmjcbwSXp59mbPhx+hIPM93q4fdJOP0q7XYOr9eKejhdEPA78IffWNj5sGNeWPkXtspXSMUfRAjBeu55DLHK2OAn33SjdLfzfM8NUjRqLYTlchm73U6z2SQWi93hrB3cDwlzRYwynB6k3S2iyE4rJLZ8DRB43f17CgXWZ0GSFU59QEJRoZA1uPBsk8GTh9OFud2Y56frlDZkjp2TUVQT4T3PWu77YD5FNBmg1S4wNf8dAp6TBz5frwvdrgQStHbYZrC7BZ2Gj2bDoNk43Nx0ezV0o4vDFkQSQ0SDEksrF+jpdZz2OGcn/y31mkK9dv9VxHQNSpvQrAokBQJRcw/Lwuq8JU/Q6ey03foiJgtTGutrbVTbwe/GVr2JXpXwn7bkHwCECapfpzzTY3Oze2Dob3XjCkOZx2m3d+6rylF0XWF5dR5/tYXDZYfW7qSO2dVolZt0s3lMQ0NR3DctChFcNqsuXZHtyM4ayTGola2G2NQx0KUu+bz1aWfQ+ruBQmHvOE0T5i9JjJxQcPl1THR8Yasva+rVJpG0iWrnjt7kfjBNk063iBAydltg33lWgCd8yZ3ePZ37onTaaG2wuPZthjIfoj/5cZqtHLMLP6RQKN1RmfZeqq9m8zNkkh+j1eri8hpsLOscC/sotZdotTo0a1ApaUS0Dt380e9z2DE321ssrb7MuZP/cXvD7fdMcuHKX6FK09tk0G8G3hYS5o888giyLPMXf/EXhEIhdF3n8ceP1mB3LyFJCi5HjHJ1mtXNH+Fxxy1RruwP6E/++na+ptuGZk3i9KMSkmwJd3mDgsKGxPq8QWb8jdGYVAvCIiZVr++yow+i6z0uXvuvuNYFQjKJBM8TCZ498Bo3RNk6LRO786ZcV9VEdRyOhLLbq7G88TSd7haybMfQNVLxjxDwju4K5Tkd3jfHGOmwdEXg9Sukhy0Kn41Fg05Lv0UWHfY4BTcUWa8/erVosHLterjKYRJOWeGqmM1Jgy7dtm3bG21oPURXQpYFQhw8d7Jkp6c1dx0zTQNNayNLdvS4C+P1q0jDfdshO6oN2hurzLrKNObWEZKEKntJxz98cDm8sPJhkrR//9CdYBig99gj1aDaBKWswGaTaTUMVIdJesw8NE1QrbHCSvYZJMmwlGFlHx7vF4C3Jje0WXiJgb7HiYWt3LRNHWJi5HNcnPqfRIKn6WkN1nM/p9aYRwiZgHeCZPQDu3J0G60Gr5bzlLUOQcXB2WCExEEx0AMgSzZ6vQaq4iQUF5S3DC6/0EXxeGhtGWytG8QHjPsibXEzuobOc1sbLOR/yUl/ivWuRlzoOGQZWbYRC5+kWl94Uw3SUXDPDVJfXx9/+qd/yp//+Z/T6XT4oz/6I06fPn2vb3MkdHs1Vjd/xMljv4PHZVXXVevLXJn5JhPD/w5FdtBugsdvGSNDN5m/ZNDrCjx+iWJOYeGSTv8xE+WoBUPm7oVGCEFf8lGqq+fwRev4As47cudJEgRiBguXDfonJJxuS+htZUYnMXRng2SaJgur3yEWGSMd/20kSaFaX+Hq7Lewq4E7iqndKH8/SsXhQShtgscvM3Tixi9X4A0KrrxoEogZ2wunJ2iQWzUZ9O2EhUs5E1m1vpNaEVayGrF+BZdXUC+brM3opEYNHgrF+WF2Ddelfk6ctdGTddZqDXorTvyR2+cIg/4TrG48R8A7iKq6MU2T9dyLKLIXhz2IaYdOIo79R88jMgnQDPSVLFdjs4SDJzmR+C0kSaVYvsbM4ncYH/zKnqbhrTWrSMUfltB6sLlkVWjdTe5Bkiyy1psZzNtNaDUEkbTE+AMypiGxPG2yuXg4ddRur8bS+j9ybPg3CfhGAJNs/hVen/oag6mvbBdmdLpleloTpyOyS/b+fqDdyeH37GbedjmjSJJEu1NmfvVbpOJnODb8cXSjx8rGsyysfpuR/i8AsNKs8eP8Ko8n06RcHlYbdZ7OLvNr0Qxp1+HTBkHfSZbW/pljw59DkhQGJw0uvf5zqlthImGN/kmO3IJwWJgm/GB1GZ/bzqcHJuh21gg6HCw1qgy7/aiSRE9rIkl3Z2zfCtyXxtinnnqKp5566n5c+g2hVJkiGj6+bYwAfJ5+Av5BytUZIsFT2BywVTMwTYnskqXwOHFeplmFWAZKOUF2qUff2NHG4Pab5NdMAtGdiqlG1aTTkvEFvIfeSUXSUNzQmL5gLV6qA9wBA8SdOdya7SyILn2JD26PwefJkIqfo1C+SF/iI/ue121DdlHQrlvnONwm8UHzSOGfW9GqCZIDux/e7hS4vIJ2fWfHH0rCypTGtVdkAhGr6KSUN+gbtwxxflXi2BkFu9uymk63QJYhu2LSf9zJxyfjvHatyMY/O/G4Few9B+GwJUdwOwS8Y7TaOX516f/G783QapcwTMFQeuc918cGaSWjSLkCqAqVB1JQnaE/9aHteQ4HJ6jWV9kqvU4qtsMGUSsbVAsyJx+WUe3WZysFweVf5ZCi38Qw2rhd/aSij912wyCEtVlZmjIYmrR0q2olyK8ZpIavq6NKgvQIXPyFhGHcefdeLF8mGp4k6L+R0BckYw9SqU9TrS/gdqZYWv8+7W4eh91Ps10kHnqEWPj+SSzYVD+NZhanYyc31+lWMAydWmMRvzdFJvkYYPlwowO/wSuX/4JGax3w8UIxy8czg4z5rJhwxOHEIcu8uJnls3dhkBKRh1ne+AEvXfy/8LiSNFpZnI4UE8OP3HevyDStjdz6qoGvlyDj8iInY5TrP6cvsUXQHqDU7eA2auQLV3dRjb1d8a5iargTrFzJPg2iyk6DqMMFit1kacqkVjKZOKfQbUGraYXuEgOQffZwP+T94I9Crahx5SWTcFym15XZ2tBJDN7d9YSwKgJDSYNiFoobEoamsLVqovVM+sYMZFsLhNhDTdTTGjjtoT0JToc9RLmyf0WkYcDKlESiXyb6gABhaSGtTOkMnzaOFF66GbJi0m2bWLE3C6Zp0m3v9sRkGQaOm9RKGo06KDZLNsBitbbE7PxhQXunrxVfWLB4zYrphe0OPnLagdazDKzNzqG8XSEEqdgHiQbP0mitE/S5cTtTe+bQ9LjRPdZOtFu5issZ2fMZtytKvjC/69jWhkG8T9o2RgCSYxWXVyGV+D1CMSf54mXmVr5x3bs62G2KpCG/onH5BQlJtuiHRk9Zukzb86jc4Hu0/l0rWmX0Nsdexu+e1iDg3lsk43KG6Wl1ljaexueNcSr9RSRJod0pc3nm69htAfzeQ7hgR0As9CALqz/CZvPgdWdod0rMLn2fkP80nW6JgP9WbTGB39tHu7OFaZrkuy1GvLvjmsNeP99bnqPdKaEozttSet2AJCkMpn+DTrdMu1MgHn5iu5XifqHZ6FIqzdFpqojOCCNnZTIOlaSiMndJEHR+isvT38BUgzQ0HXuvSF/8yTvSeL0d8K41SIZhyRMotp0eFZ9nkNXNp0nHH9lu4tS0NoXSFEN9X9g+Nz1mklvqUcpJlPIGLo+l+2MpV17/0F3qhRkGbK7qLM6WKdUug1wlP6Nis7kZHj+Fx3O0zrlWHSo5icn3KdgdVs5rfTnHa5d+gDNgZVFdjiR9iSe3yVPdjgSr2R/Q01q7Kg2L5WvbZK+3olYEp0ci3r9jNWN9gmrRkmUOHL5uZV8EYrAxp+MLCRwu6zk2l02EbOyS8gYrJOULW3+7jgsrHNpusouD5Ea/0s1Q1KPJe6uqh4B6OPZ3lyPBRv6n6EZvF0tHqTKH07F7nk3DRL5pPLrRRdPr+LzHsCkSiixIRs/TbpfIl14lHXvswPsKAbF+i2xV1yC3DOKWHUNly9p4maalveRwyPhCgkbNZOGiQWZip4ze7UqxVbpEIvrgtnHVjR7F8hyx4JO0OhucOvb57VCzwx6gP/VBNjYv3jeD5PMMoeuPMTX3PXS9gRAy4cBZEpFHyJdepdZYI8mD2583TZNqbY1k1GIxd8sqhU6bqGPn/V+vb9JrzrO4/hI9rYXPPUxf4slDhR/ttsBd6SQdBboO81dXKTS/SyiUpJE7z8DECjabh5LmRPXA0KTC1K+GOPu+/8Q/zT2PXYEHMp+7LSfh2wnvOoNkmlYsvrwpIatWgjcQN4ikwe1M4bKnuXjtv5OMPoBpGqznfoXfcxynfWd1k2VIDl8XrGsZxDI7q1thw8TpNu/KIzB0WJ4SuDwdfNE60cQjFDbsxDMGbemXrOX/nnH3V45UklnJQ7xf3u5d0Y0O2do3GBr6IOn0KRTFZCP/MnMr32Ri6PeRJAVV9RD0neLS9FfpT34AVXWR23qdaj3H+MCT+96n1wG3d+/4XN7d3shR4fJa3t7VlzQcbquoQVJM0qPmoYtIboSrFq70SI+a2BxWSG95WicYf/M5XBz2IB7XIFdn/pZM6jEUxcHm1mtUa5uMD+6e50BEYuFaj3BSIEkCXe9g9DzUK9C/QwaAz9PH+ubFQ91fkkCyWZpPy1d1eh3whQTNukluRSc5arK1BoGQQmZ8R9gvtyrYXDTpP27tugLeMbZKF5he+HsS0fPoRpe17C8JBUZQFCc2xb0n72m3+dD01p4x3UsE/RMEfMcwjC6SpG7nskL+Sa4tvMza5i9JRM5ZOaT1nwE2PC6ryfqkL8wPVxf59MAIXtXGVqPAPy1d4vHMGR5MT6DpbeaXn2Fl4xkG079xX5/jsMjOa1T17/HII58l6B/ilZ/0SCSg3p7Bocgs1WsknE46XZOXtwos9/x8IT2GfC+TvfcZ75yR3gGmadJobVDY0BGdPiYftjyGTttk4ZJBcUMjnBJkkh+nWp+nUJpBCEEy8mG8N1WV3YxYPyxf1Wi3rN1js2pSKRpkjt2de1TOg8Mp03esiGF48Lq9xJIml1/qMfnwB8gXrtJorR2KiPRW6LpAvWkDt1W6it+bIR44hzCtMGA6/gjF8hyblRlk5wABm51U7DFK1SlWNl7GMLp4XIOM9f/2gTsphxuK61Ye4sbCZZomlYL5hr2jGwjGwR8xaDetTcGtntFhEElDpypx+YXOdTkJCCbeuAd3VPQnP0a+eIHZxWcwzC5e1yCjA1/aM8/BmMTSnM7Uy1i8dx0bS7M9jp0yUNSd3U+tsY5NPVxIyDB6ZLdepFqfxfCabFbGqNffh82ukjlulX5vzElkzotdm6FIUrAyaxX1SLIVlhrp/zz5wivMLf0LQkgEfROMDz1KLleh22vSaOZ2kfduFadwO+/+fb5bCCH2zKUiOxjJfIH13M9YXP0JQigEvOMMZ3Z6yB4IROkVDf762iXsskyptsID4TDvTx3buUb/x3jp4n+mpzWP1OR9L6FrJoXiKpH+CEH/EAAun6DdkHE4I9ho0cTD1dUas602arXGZ5LDuJR31hL/zhrtAej16syv/gNC6LRzTzJ4Ypme4cZOGrtD0H9MYuY1iXDKuB5LHjlUKEFRYfCkSXVLo1bZLXN8N2hWBfGMhGH0UGTrxXa4LS6yVk3gckbo9epHeXTcPpOtDZNgzCqS6HQqOO1xej0T5/U0g4lJR/Hxs/UruIMqlW6HU94wD4UmCPmPH+4+fiisGyxckUj0AwJyKya6puO5hyFzSX5jHe1CQGZUweZt7QnZvhUQQiIWPn/HBL8QgvQo1Ms61bKOJIM7eZVC51W8nSexqe5tuqnxwS/f8b6maTK38i1cTi8TI58CIVjLvkCz9U0y/V/c9iaEuC6Ke/O5+1xPlmwkoo+QiD6y69kkSSEZfYzLM39Df+pRHPYQhdIUW6U5xgd/947jvF9w2IMMZ546sFFfCMHD4QTngjGaWo817WWOJ87tUYm1qR60t4FBMnQQQke+Ka6bHpKYfV0jlFRw+EDVHBgLKo8dd+AL39/w4f3Cu8IgLW/8kGh4jEzyMV4rGvSlJcr1edqdAg57GKfbaiY9jILorbBKrN/Y+IRkonVNbKqbZquM0xG2xAK7gOhRqS0TDX7gjtfZD/4oVAs6M69aO9tONc5G/WXOTj6MuMFk0KpTrM7zpYmniPgHqfW6fGtxFle1yEn/wV30hgH5dZ21JYEkW/QzjUrvuow2uIMGmeP3R176jSCbf42ZpWfp9SzjHI+8H6/7DmV0byEsVgaDeglc/h013KjxYTbyv+DVK3+NbnTwOPvJxH+LXE2lo1SIu1x4DkiE1RpLIHqMD+1w0o0PPcXFqf9Ktb64TYvkCRlsLJkMTe4s2rkVE5d3JyzdbVsLos25v3EPB05iUwNsFV9D067gcqYZH/zyW76Iw50ZY1RJwm+zU3UkKFVm8Lp3+nRa7QLdXuO+54YOA8UGTns/xeIPaba2cDkjeIMSySGNS69ew2Ufwe3RiA/ubSZ/J+Edb5B6WtNKqia+YDEge6FSlPAEEtTqGzjsYSoFq0T5rVo4A1HILhnEUkE0LUu5ukKzGKLTazK/8V187tFDvfRa10ro3+yhSRJkJkwqWxq5DYEkDSDk51nJf4+UeAjD1Lm88COi7ihhnxWa9Ko2PpzM8KOVpQMNklVVJ3C7GNk2VgAAIABJREFUDSJxhV4Xsgs6oaTB8Jl7m4/RdavC8V4kXvPFCzQ6lxkb/DUctiCFyhTzK3/HcN/nj2SUmjWroAOsIop7zdrcrMH6rIQvoKHrCtkFg2i/QSBqhcrS8SdIxR5H12B51uTiahfZBnrPznOuNYZSNj4QSe5ZeJutLCH/8K7jQgiC/hGa7ey2QYqmYXVa48qLEr6QRKNq0u3oZCZMeh1YnxNoXYGiCnpdk1i/saeYxDRNTNPy/m1qgJB/8m1hjO4GsfBDzCx9FRCEA8dodYosrv6EePj9d+wLfDMghCAxoDI3/SFefOW/k0qcBsPJ2sbr+PvCjAycvm1j9zsFb/1Mv0HsJDSt7VxyUGLxqkZySEEXgl7NYHXucA2j9wtuP/giOpdf0JFtQ9TrZerNadzxy4TdE0SC+zcOm6ZBvniBfG6JdnECm0hiUwO4fBKJIXM7dyRJEIxBMGYCMnH982S3XuDK7LfQTbjWcvGfHvoChgHlnEG3A6rTTlPv7XtfsKqwVFVm4rxCu309uR0VXH1JwxcxDt3hfzt0umVWsz+i2d4AwOlI0Bd/8shls6ZpkCu+wIOnf59Wu06zs47fmyQZO8XVub/igcn/864WytwyNMoykZTlFmzM6XjDBtF7lBoxDFibkRg+oZDos9FqdWk3JaZ+peF071QXCiHILgi69i7nn7CTcrtpN0yuXnDzamWBKXuJ477dXHk21UetuVfHqtHcxOXYIY2VZGtD06zptBs6vghIiiXgV1gX9I0qpIasHFOjajJ7UcPm2CF1NU2DpfXv09OKxMIn6WktZle+TiryxLbw4z1Hp4vo9jDdB7hsR4Dd5mds4HfYLLxErvBtVNlNIvI4Ae8R6dTvAzwBOHbyOJtrcTaWplDsZQb6HicQGHhHEKceBu94g2RT/QihUq7OE/SP4A0Khk9IXHl9Hq3jJhjUSI2abxrT7kEIp2B00sbyQpdg0o/T40eIY7c9Z2Xjh/S6HWytTzH+gAOHr0CztUSrOMLqNYPBk/t7fbJsJx1/nDSPY5omr69MM7/VojPvxuOXcHoEyws6mWYfRsYKZZrm7jLoRlkQS+9OdjtcApdPolUz3nBYwDA0Zpe/QV/yIZLRLwGQ3XqFuZVvcHz4D5Cku6/Jtqq6LONpohENTSKEhN0WZDX7AqvZHzHU9+lDXavdgFpR4sTD8jbNUyQluPyChi98Z2ZxsOh2tkqvoul1nPYEsfCDu/qH6iVw+yT84d1zHE3LVLYMYtdbaXpdaNTAc7ZH0u1DCHB6BJlhlfZKiqna0h6DFPCNs7H1LBu5l4hHHgAhyG29SrWeJRX7xK7PCgFun5W7yy5Cuy7j8QtsdvB4Jbotq8DE7RPEMzKlnEHSyqtTrs2iG1XOHP+DbU8iFj7Fa1P/A7935N6WG/d6qJemkbaqdE0bvVqPreAQynicyMEK94eG3RagP/lr92as9wkOFwyMhYBH3+qh3Be84w2SEIJ0/CNcW/gOicgZXM4IxfIM+PIc7//SLsGztxqKKrbzA3dCu1Oi1lwk4/9fwWkjlpaBNIbZw5kqUy0EaNb0bcmEgyCE4P3hJC9fbPD+CQepjKCudfEG2hxbCDD1komiCIQEqs1iXnB6QFJMtFscKNO0cmFvtAkWoFybweOOko4/vH0sFXsf5eoipeo04cCJu76mIjssnajqIqHAxHbivtHcIOQfp1ZfRtc7h1okayWIJHeMEVh8cKG4TL10Z4NULF8hW3iWgfRjOB0hiuVZZpa+ymj/b2/3gxnG/r1QigrNmySotC4odiy+PW42XqDoMt0b3a03QZIURjJfYDX7DIurP0MIUJUAbmc/i2vfxab4iATP7mJ9aNagU5eZfEimXrZ6o4JxywCpDhNJsgp79OIOcWC1Pks8enZXWMvljOJ1J6k1V+6ph6G+Po3idlF/9CyeoIKvWyfw81dYzLnI9rzbRvI9vHPxjjdIAD73AGP9v8tW+XUazXmcjhTjA792T3dnVgntC1Rq1zBNHa9nlGTk/SgHSFi8UbTam/i9fWhdBW9gZxGy23y02lWcniC9zt7zum1rR2137ix2faqPumIih7qsNdrYJZkht49yUKZR0DnzmIQkWZxwS9d0Bk8a+COQXTRIZHYMejFrqYzeizxKp1vB44rvOe5xJeh0jqbLLYRMJPAAM4tPc2Yigaq4qdSWmFt+hqG+T3Ct8W0MU0Pmzu+FkNhXst7Q4U7Om2nqbGz9nBPjX9jWlvJ5+pEkmc3Ci9u7cLcP8isGva6E8/prZBgmhQ2D4E19s3YX6B1BrwV1Txfv9VhtKW+SFVUyB3whDnuQ0YEvomlNur0aC2vfJuC0E/RN0mznmV/9OzKJHWXiWhH8sQarm7+iWtugmHMRqTyEzd6P1rGIe8t5q+BhZ84VDGNv6NcwekjiHuxcbqDVQa1UaZ8+g9Oh4HALcHtRT40wsLrOa6XjaGnj6ByTbxK6bahXrLYGb/BoBLrvZrwrDBJY7vbtutePCtM0abW3WFz/R3zeOJNjn0eSZNZzLzG7/A3GB3/3viQ9baqfXGmLcNCqwIr1Wbt9TW8hUKkWDdI3EQYYupWAbjcspoNW3cQX2Qn72GWZPpd9OwSnda0Qh8tn7bwBQnFBvQzlvEE0Df6IzqXnu6hOg17XpNcxSY/fWahQ61nknqrtYGJJtzNOrvgc/akndvU1lSrzRALvO/K8xSMP0+xc48WL/xmHzY/dHmQg9VEMo4eq+LbL7u8EXwiWrujEMmJbB6tVNynldQZP3v7cTreKLMu7hA4BQoFj5La+s/1v1W41bU+9rJMZ1dE0g/y6iaTou0KikgSRtEH2moOpeoOIv0e7IDOz1GAlmuMzgdu7BoriYj3/LMnYafpTT1hjYQy3K8Hs4jN43f8GIQS6WWMp/zcMDBxneOBxQr4Cr77+D0S9HyIWPUF1waRe1Rm4qTYk6JtgbfMZosETqKqVWCpXF2i2inhS1ge7bYtzrdcV2J0mwfjds2SIbhdcDgyUbaZ2ADwu5F4Xp1vQbR+OBuqtQm4ZqgWJQESi24PckkFqzLhjlONfE941Bul+oNMts7j2XdrdPIbZIhp6DEWxYbcFGM58nMvtr1GuzRy6l+du4HImENioaT+lVfsgS1Pgi9apVGvUc/043AbYNbqGwCbJbC6B3aEwfkYgyQKtZzJ70aC0qRFKgM1hsrVuEk1fZ3TQTIpZk2Bst3VxeqB8XWcnnIKR4zaW5+vIimW8bmeMTBPyK1DZknC6LZYEu8skNbpX5sDrHmSz8AIzi98hHX8EEKxtvoCum/g8R4+9CCF44OS/5bmX/wq/N0bQP0q1vsrm1iUG058+dPLX5rAUVa++rOEPSZimJWsRGzB2NSLvB0V20NPae8KDnU4Z5ZaiimgaXF6ddsOkWunhi1ynRDKhWrSooVS7dUy1CzY2FK4stairTSLD8FvhYeyHqDBpNFcY7PvirmMB7xC60UTTGqiqh470IiH/JAPpJ5Fl8HsHcTuT/PKFv0VrjOAJSAxMWrv7UnWa5ewFSpUNNK3Ns7/6U/oSD9PrtajW1xlI/SaSpNCsWYUb0ZRMKArVEixe0uk/btwVC7bpcWE2WkitBprDg+0655+5lkP3+2mVzfvOqv1G0KhAoyJz8hEZRRFoGtTKVgHWyNn7L0/xTsF7BukAmKbJ/Orf05c4jwloWoVwcJxSZQ5ZtqPIToL+Yer1HNwHgySEYKjvM6xkf0RN+i/kF4eRZofxuJKooS6vySvkly3OnqTdRV8xw8OP2pCuezuKKkiPSCxelQglDBJDFhFqpWDidAuKOYNK3mT07C0cZwVzlyy2ahN7ynwPQjkPnYbCyUckVJvAMEyWr+0vcyCEYDjzue1qQACfe5SRzG9t536OCpvqYWLoKxQql9kqLqAqfsYHv3zX5JKBKHgCBvWyVeY+1He4nb2iOPG6hlhY+RFD/R9DllTanTKLaz8lHtqbjHb7IBpVtoX4dN0quVcUi9G81TBZeN2gb9xgfFJmHBdwZ09vpVnjxdImuU4To5OgmV/jwaSPmtYjaHMgmz1MQ98uIOkaa6QTn6KUM67zIoLQEwRjCsmBEo7r9Frl6jTZrZ9y8thTTAwnqTVWuTb/HVqtFkHfMdLxX9/mf8stCwaOyYTi1ncajFmEtlurBqm7SS/JMt3hfmwXLtDOjCDHXKiFHL3FTRZi5/AEDh+uM02TVt2KKjg9B4fNWnWoXt+ceUO3b9iulaCYtTS47E5Lg+vmz1cKEM9ImIagWhBIMiiywNQlSjmDcOLga/9rwnsG6QDUm2soskIi+iCl6iwblSlUxYnLGaHTKaG4nNQbWWzq/mSk9wKq4mK479MWr5mpocguOobO11em+WC0j5PBMCbwYi7LLzbmeb9ygpvZsm32nTyI3QnDpw2qRavsO5qxdt7zlyVSQ5YHk183adZ1YvszKd0RlbxE/5i0HVKRJEHfKFx8TsLQ9zKCy5KNdOyx+xJqlWU7sdDBcu2HhaJahulukUk8uS1LYFd9dLpVoqH3EfDdmZi1sA4uj8LQ5E6VY3FTsLZgMnjicP10G60GP86v8vG+AYa8ftaqKb6xNMOPtl7kWCBMpdMmoa8x7hra9uJU2YUplfH6kxaXoAC3o4spWsjyTq50s/giowO/TigwQqvVxe8dZGLkM1yb+x5B/8T25wwdOi2xxwsPJwSbKxJwd/1s+mAfpsuJmFujfKVHGT/1yFk8bpnEIUvxu2249Mse7ZaMYoP1ukmkzyB4Szozvwq1gkQ0bb20mwsG7oC+HQK/GZUtKKzJZMZkXF6olUxWZy0Nrm2jZAAmNGsCf1hCtVkRBZdXkJ2XCMbe85LgPYN0IHS9icPut5oJfSMsr/+EpbWfEQ6Mo+kdNvK/olxdZmJof+2gewlZtm8n4qdrZQZ8Ps6Gd1bJDyRSXN2ocmmtwpnMThlfKWfi8u786CV59+Lq9kEhq7FwRcI0wOU36D9+9ESrru3NGcmKlQPR9bdxAlfTUGYWUTa3EIaBFgvTGx0Ex9GLYmTZzlDfp+n2amhak063RKk2xczSDG5nhljo3J7w3Q00yhLDk7tL7oMxwfK0QOtaEuR3wivlPI8n04z5rffB7gjw64MTPL00zW9EbTR6Ot9f81Bx7STEwoEzLK39DI8ricMZwDA0FlZ+jNvZv6t/q9Mp4PPsbjL2ujN0e2VM09jxcC2VEnRtt2fZ6x79XTBiYYiFsQFRE2KHi8AClgFYnRYMjMr4ruuRtZsm117RsLt2jEenZW2uTjy8k6+KpgWXXzDxhY1dEQTThMKaxPBJq1QeLB0vIWBzzaR/wioA8QRhY9Fk/AG2Q76tukm3Y1os62XLC/vXjvcM0gFwu9Ksbj5Dt9fAprqZHPlt5ld+yNW5b6FIbvzecUYyn3/Tad1rWpe4+8YvwsQwDSQhkwm5mJ5tETf8OD2WLEQhq5OZOPhaQoJICiKpe8O84PQalHImycGdVaJWsu5zFKmHW5EvXmAl+2M0vUXAO8ZA6te3S6iPDNPE9solVJ8H8aGHQJaQ51aQX7pI+9FzvNEOYJvqpVSZoly/RH/qg9hUL/niZaaXvs74wO/sW6UpBBi3ditc7xWzBBhN6s01ur0yTkcMl2Mvt1Wl19lWPjWBcq/LiVCS5zZzaFKApL+PT9oHeXp5gbMhy8sP+Mbo9CpcuPLXOOx+Ot0qLkcfA8mP77q23Rai1lhDksP8/+y9d5Ad53nu+etwcs5hcgAwiCQEgqQoSkxKlizZlixZtu91uNrr63Wt1lvrWtfWVnnL/7jKrnJd75bDXt/y2uW4smhLlkRRFCVTFLNIAiQyBpPDmTlzcg59uvvbPxqYweEMgAEJkqKE5y+ycTpMh+/9vvd93ueZWfg+9eYKQpj09A5Xr9Bl2ZImWp0VjExZaVrDEGTmTPyRt/7O3WzfUbsBsiwTG5RYXxZUCyaSLOELylRyWwGpXoZwQukjT6g2iUhSoV7uD0imYZF43khM8Ecklme26PG+MKzO6My8DqlRBb0HhTWDoT0S1SLoOzA6fxJxOyBdAzbVQyR4lDPT/8BQ6jK9WwK/e4rJ4Z/HZrt10iiGqaFpNWw273VNwYQQ6KbJyeIGaYeOw6yhSJY6w3y9wv0Te6hXepTyVh575AB9s2khBIutGpfqFXRhMuTyccAfRr1FuYJo2rI50DXrg+w0BetLBonRGzPzboS5lX8jX3qJkcEP4XQEyWRf5rXzf8TRA/87DvubpylJpSqqYSIdO7A1wh3ag1JrIK/nMQffWnLfMLrkyy9z9OAXN4NnwDfCjHiMQvlUn1jpFXjDJtklE69f3tQjzK8J7E6BJLeYWfoaoOP1JMmVXsRhTzCa/kQf2zNoc5BpNYg4XZa0D9Ds9eiZgpArhCLLBO0yrTdw2xORu4iGjtDpFrGp3h2NAOPh41yc/xqybDCcfpCh9L2UytOsc5q13LMMJD60dbwRyMzonH5ewuOXaVRM3EGTyJZkHJ1umWzhBRqtZRTFQdh/mHjk2FuuJQoTCutW6s00QFZNVEUwd8ag24H4oIppwMqMTqcjkZ6wgocsww5MdmuV/4ZLkpXtlvFwhYxyNT0eBiYhvyLodgwUVWLv+2RUGyxNX3/i+JOE2wHpOkjF7sPtTLCRP4dhdvG4xkgHP4rWtqOqW+NXq5MjW3iJdmcdVfUQDd5JOHDwhowuIQQbhR+Sr5zAYfPS7dUJevcxkHhoG5XcECaPry/SwaDQqfFMQWYyECPt9vJqbplu4xKRaAT/uJdruQe+XNpgoV3jnkQKp6xwqlRgbr3Kp9NjKG/x4wcrXTdy0KScNcnMS9gcgsG99M0o3wx6epP1/DN86Pj/QcBvJfFHBx7kpdf+hKXMt9g79otv+thyo4kUC22bbkuxMHKleZNVju1YrK5w0UxRyZUZ85ns8QeRJYlIaB8rmVd23CechMyMwdmXBIGoTLsh6LRMBvcJVjeeIhQYZnTwESRJwhQG0/NfZaP4MqnYFmHifcEYT6wv41RUxn0Bat0uX86ucTQc35yAzNTKpHbw+FBkOx5Xip5mDawOV3+KLRSYYqP0Kh6PE5vqpNutEAlbQeTE2b8kETm+ufJTFBieEnRagl7XJJzuT+v2eg3mVv6ZgeRxRtMfptVokS0/zWqvwtBbVE1YnwcJlYlD1sBfWBPMvG4gSyYH71WRZYu4oagqF0+adNsCh8tazSyc1ggmOzg9Arvqo9tWKG0YjF7u19Y6UMlZ7ROKzWThnMH4YcuXrNW47MGV7P8O/RGrpaLVUIgPSDSrguyyiTe0O+WPnwTcDkg3wBWrinIOCssywi+j64JsVzCwx0SoReZX/pWRwQ8SDnyUdqfEwsr30PU2iejx6x67WDlLvT3L0QNfxOkIWDpgi99kPf88A4kH+n57tlrEZlP4/MgeNioXmO0FeXFjjbV2k08PjnP3xBEKhdfwe3eougKNnsbZeon/PHUI9+X82aQ/yJfnp5mtV9nnvzUeEjY7lwu/t04ho1g5T9A/vBmMwFIiGErfz4WZb1xnzxtDuF2IhTzSG6TgRamK8L01vanXy3ler7XYH50g4fNysrTBhUqRTw9PXKaA7xypZRkG9wraDYN2w8AXgdQkgE6tOc/U5Jc2JzuypDCc+hDnZ7/aF5CSLg+PxAZ5cX2NbyzNWTRyXeOOYITVZoPlZo0TuQ0+tYO8gWHA+pxEpyFhc0pobUE4bRK5ir+jKjbGBh/EaR/o29flDNPRynjfkIp0uvtXEFeQL58iEjhAr3gvi/Ngd3mh+Vk2tMeJR+o47G/uGXTb0KrLHLlP3mSeDkxIVPJWP123bd3nTst6T2NpmWbNCgz19gXqylnOv/IAgYiKrpfp1IMMTHiwOSwK9/qcTDStEAhZDcqFdYMzz4vNFVNkwGouvxqSBEP7BJWcztqShCyDP96jLV7l4vw0AgO/Z5Jk9J73jMPrrcbtgLQLdJpQWpPZf1zdbJIsbZgsXwJb/BWG0veSill+Nw57gP2Tn+P1839DLHz0uk2zxcrrTI5+GKfDSuXYVBfjwx/n5Lm/Ih2/f1MwFmChWeO+9AAIA7fNziOJUR4eGOG/XzjD3kAYt6mylt15xg2w1mky6vNvBiOwcvoHQhGWKrcuIL0dsKkeDKPbXzAHdL2Ncp0U525gRkMYMwtIZ2aQpsZAlmBhFaNQwdj35u23W7rOiWqOX997B0JbwWE3ODy5j3+aneZsbp5O9iWGUz99zf0lyaIZX00dNkwThNim86co9h3VEobcPoauOkCu0+J0tcC5UpGwzcnPpScI2rcPfNl5cLm3etq6bWtlYXMYXJHMs6sBGq1sX0AyDI12p7xjmu9a6HRzqJ0H8Hkk7rhfQVYktK7KyeePU1ivMzDyJgNSC3zBrWB0BeGkRG4ZtI6JaUrY7JZvWHYFPB7QenXWck9x/K7/iF2NUisKOt0qi8rf4w7+EkL4yC7KjB1UNzUIw0kwDYlcxsDts1aTtmvUTGXZ+n04KazWkpVvYHeo7Jv4JLKkbjbc7xn5wo+Eyvg7jZ+8v/gmIUxYuQT+oIrWkRGmwOGGcEImt2pSqecYHeqvAzgdQWw2N1qvutm7sRN0o4nT8UaVZh8gMMweqqJYrqz1GdTq86zrLpzR/djtTgyjiyzbMYRABir1BVyOrSmZEAYgb86knbJKTduuNVTraTh+ZOlvFkL+vcws1VnNvsRg8h4kSaHdKTG3/F1S0Z0t13cNSaJ77DDm9Bzq488gCehFQujHj8BbcNvMtBsMe/347XZMdYRGa4V25yJDtjbPL7zGJwY/gMuethTFqzI5j4bqtlxzr5XpVWQ7LmeCfPEMieidm9vXcq+gKn56vQY227V1neJONx927ryCvgK9B62azORhaXMwd7isnrb1ZRN/2FpRRMPvY3X9GyhSgKB/gp7eYmHlSbzukZsKSKocol5UOHhU2TyfzS6Ij7Qprb4Jvv1l2BxQrFuD/tWp814XOm2BwPINMk2djbUqG9k6ce8SjbJBNDyF22WdO5yUgBBt9lCuTRPy3AUCrtaybdUt3UO9J9h/XKZRgflzOqrdvK7MVrO9hm5WOTL+G5sTrcmRT3L20j9Src/2Ueh/UnA7IN0Aa3PQ6yoEQjKBsEy7KWhUBL6QQLVJyL0AzXYWz1W6bD29haY1qRe9lDuXJWJ2+LZczjSlyiVS8bs2t1Xri6iKF0W2Zq5r+edotuc5PnI3HWTqrTnaxSKKYidj+FCEjlY7x8rai0wMfZ5S9QIbhRfRehVsNh+x0HGioTsYdHv5QSHDmXKBQ8EIkiSRa7c4Vcjz6R9xVUpZVtk//uucu/Q3LK89h8Pup1CaJug9QGoHUsBNw2Gnd2Q/vcvuorfCOMsmy3QuEwZkWcXvHcMwNWiuMRJ/gJB/iMVzEv6gyvhBCZuqsnDRoNPUtzURX42BxMPMr3yVenMdrzvOWu5l8qVzJKPHmF78WwK+/QwmHrpm/VLr1ajW55AkmYBvzzY7DouiLaGoErpp0tB7IIHqtGH0tlhjbmecfRM/w7npb9Pr/RtCCDyuQSKhO9D19q41HsP+OynpdXqGC0UNIIRBvZnB6bSDcHGzvUpX4PSAbDNZuSSTHrfaD8o5QXHDYOqYg0uvdbG7dBrNdUxDZnh/hx5tVtZeIBG7Y9vxFMWG3tMtwoNpTVQlxfpvXZM2tSMlyRJQTo0qlLImA9dp/m21Nwj5x/tW/ZIkEQ5M0Ghu/EQGJOX3f//3f//dvoir0W5riF2UHzweB62W9rZeS6cJlQ2ZkSmFSl4QH5SxOyV6Xeh1TNaXDJLDHlay38XrSeKwB9B6dWYWHqeeG8Dn3E8wotDTFNYXIBxX6V3lQeSwh1hc/Q5IFnW7VJtlfvlJ0vEHcTkiaL06mY0nufPAF4n4h9FkL4ZrlHJtmleLNV7eyLBPP4ts6AylPkpXK5MrPc++8Z9mcuSTBP3DrG08i2kKvO4UQ24vz2yscrKY53y5yIn8BvdH0gy6rz2Neyfu827gdIRJRT+AhBMJF2MDnyIZu3fHgfdNX7Mk3TL7W5/Nxg+LG4QcdjxyB02rUdM0vr+R595wCr1sA6EyftB6p7x+FV/YJDMn8PjFNWnyNtVDKHCAdqfC8tr3URQ777/zfyOduJtk7Ci54qv0dA2Pa3vDdq54guX1b+NyutHNGqvZp7Cp/j7Fb0WB4rqE6euxbjSQZYmuMJlf6NIVGono1oVFw0lctin83gma7TUM0aCnl1jPP4cpBF73jbtVbXYntYINxV6mq6/Q6hSQZRftcgqBsWt1/Cu4UgqUJEu8tFowWZkRZJdMWk2T9IRgYMSF3dclV36acELh8LEpgsEY4cAkNpuLuaVvEwnsJZN7iVzhdRrtDdZzr5OKfQiH002zCnpPtlZYhmSZVy6ZhBMSbp/1/pgmlHPmdZuqe0aTamOWRLQ/AGbzJ7Gp4b5n+KPyHd4MrlyzJEm43buT0bi9QroO2k3wh2XCcYlyzuTiSZ1YSqHVEKzOGZazZ2AQpIe4NP84utkCJKTOIcaHH2Ro0kqFRQGv36J3xka3ju92xpkY/jz54itsFC5gVwMMpz6F123l5RutVYL+UWyXZ5uDbi9tQ0cED6IV5/nE/p/FdhUPdSX7JBPDH9skNnjdKfaMfYrzlx4lGrqTsN3JLwzuJd9t0xMmCYf7llG+3wkoioN45K2rL9xKLDZrXKiXaBsGaaebI4EYblVFkWQeDof4yvnvkfDF8Tg8LFSyTEpFwrYx8k0IRvubX2VFwh+WaTe2zPl2gk11Ew+/j3zph9yx/z9tvh+q6mRk4AEuLXxnm0pFu1MgX37lMv3cosk3WzlOT/8DXs9W46skgz3R4bWTBscPBvD7FCoFgVzu8rL3InuMSZxXWRZLkkQm931ikUmGUh9EkiQ0rc6ZS/+E0xG5of2EJEEUNSTpAAAgAElEQVR6zEVuwUNyRMLlkankobBuMLxLRS4hLPHWclamp1lW69G05WybngTTNC023VWZaVmGrnSW4ZHf6nsGqdhdnLv0//Hq2T9jcvRjuN0jrK2fpFSoola9BCMQHxGsz+qUNmRcbonssiA1JhFJbX1L1aKl43g9BLxjrOWeYTX7Aun43UiSTL50hlJlganxB3f3x/+Y4XZAug5sdqgXBUgwfkimsN4hu9amkgfJkSOUSAMOgv69BHx7MIwOsmxj/pSd+ED/rQ0nJNYXTMJG/4fhckQYTvcbpl2BqrjoarW+bS5FxUmHAXe0LxgBaFoZn6ef9eRxJegZLYQwkCQVSZKI70R3uo2bxuvlPOebJT6QGMBvt3OhUuKrmVk+OziJS1HRKi/wK2Oj6J4D6KbJz4zeydra98jmX8Bhe5h2s/94QgjaTYHrOm1VQghmG1XOVtbwdFrkNYOYbG6+C3abH8PobNuvXJsmGTvc17PlcccJB8ap1ueIhg5vbl93lHCP2GmW/FSyOh6/xB3H7WQ2PCw2a0xdVUDpahW0XpHBlEW9r9YXKVdnsdtcZAsv7coPyRcG1W5S3oBi1sThFgzvF3SUHloPvDdQsy1loVlWmDxyRboHFi8YSLK1wrrWnEuRbeh6uy9tqesthNA5dvBL2Gwe2k04tOcI0cizGL2TyK0HyC3pjB6yWJC9rsUq7TRV6mVLxLiSF+QzBiM3MMyVJIWJoc+ymv0eK2svIkkSdluY8aHPXrcf8ccZtwPSdeAJQG7ZJLskE0hWUb1ZQlKSRkXClZhlZukHTI58AVVxIknSZt5c3sFLx7jsoXYzGSGfZ5jVjafIFl4jEbkTSZJoNNdYz73OxNDntv3eYY9QaywTDm7ppdWbGeyqr4+x906j2c6SK75Ct1fGaQ8TC9+Fx/XeVpPUTIMT1Ry/uvfgJlNtyOPjCSE4UylwVyhCs7XCoX0/j3IVK24gcTenL/4Te4ceZvGcgT8kEYhaPkgbywK9Z+K5jvjEiXKO+XaN+9OjlIwkndYiC+YgY24/NlkmXzqL1z20bT+BiSRtH9hlWUGI/jqNLkx8AcFYvH8kt8ky+hvy6brRQVXdSMjML3+bRitDInYEpCSF5X8nX3qNWPjoNf+ent5EQsLldV8mAAhynRb/ll+lLQxMIfArdh6ODxLaQc5bCGtltPeogttrfVz+MAzvU8jMm/hC116lBH0HWFp7mr2jP4MsqwhhMr/6PWx2P6HgJO0G+L0yXp9EXDrE7OKTHJqSOPOCTLdlMerwWf1F1bzO8oyE0ZNweU2GpgSKTUMI+3X7ER32ABPDn0XX2wjEtpreTxpuB6TrQJJgaEqwPqcxc76E3zeKy21n6qiMx//TzCw+Rr70+rbCui9isr5oMnZQoqMV6XbLZBcdaHIT3Yhjl330NEtEs9OUUVRBMC625cslSWZ88GdZXHuM1fUXURQH3W6dwcQjfey9nmaZqzmNh5iefZJ9kxIB3zD15hqzS98mHrln17YLtxr15grL699ieOB+/J4PUmuuspj5N4ZTn8Tn2T5wvldQ7HYIO5zbaNP7AiFeWF/DStTCGwuiQggkLKLLwKTJ8oyOmAabCig6Q/uuLZ6qmQanagX+075D+Gx2qvLHmFn8Oo7gQRZbcZTOOvnSRSaHf2HbvgHvJPNLj+G3340iu1DtAtQqxfIMe0ff3/fbMY+fp4sZjkUSm9YW9Z7GbLXKsYH+oojLEUXTWmTzJ2i01zh68Isoio1qfZmAb4zpuW8S9O/bNtC2OjlW1r9LT68gELgdSQaTH0EoHr6VXeQjgyPsC4QQwKlSnsfWF/ml4b3bGrhN05LvuRKMrsDjB62zRcLYCanYfSyuPc6rZ/4f/N4B6s11FNmD0xHBNHVMw4bjcgzUtPrmxNPjt7yXrjR8SxIE4xCMC4QwKZRPMbf2MqZpMWFjoePEwkev+w2+XUaf7zXcDkg3gM0OkdEsbcdzHNz3K9gcWy9VLHyQxdVngf6AFEnD2qzOK0+X8AUNzN4QsmySGFljZukpxtP/gcy0h2hKITlkNR5m5k16XYPwGxYOTkeYfaP/kXY3j2nquJ3xvv6ESg7yqzKhmILHOUyz+AtcOPMCkufrOGxB4uH7ruvXZJpbTLC3gnZjS7cuENnqxl/PP8fEyEeJhqxr8LgT2FQ3K2vP4/N84S2d8wqEEID5jq4CXYpKradhCIFy1UBT0bq4FRVZVvG6R1jPvcJg6gOb15nZeAm/d69VfJchOmj5K6XSdiq16xetS5eD4BXH2IBvhAOTv8S51ec5vz7N0eg+9o788o7U7/pGCqlzgEurf00ydphuqcfq+hmGh+7bRtNOOT0M2D383cx5Doej6MLkVDHPsUBsW/pMllXS8Qc4P/sPTI5+lJ7epNGqoGktAt5Jgv5Rao0F/J4R8uXX6XRzyLKLauMSe0Z/ilj4IEKYrOdeYX7lX9HCn2Q8EGAqaKUFJeBoJM6lapmFZo3Jq50LsbIRsgKtutgkFIDlymp3Xr+GI8sq44Ofpt0p0OkWCAeO43bGmVv+Kstrz5AIPojWVZGUBouZZxhM3I9pChpVk+A1FvjFylnK9VMc2vt5PO4EzXaOSwvfQJIUYuHt7L3b6MctD0gvvvgiv/Zrv7b5/3v27OGxxx671ad5R6HITgxRtWaVV4lHar06srx9ZiPLEButUJr5FrHUr+Jyq3j84HY/QKtbZGV+nVhyH4OTl2d7AQm3X+LCq4LgDjL0kiTtKKDZ06xgtP+urYbd1Iif8698mKGhh68rR6LrbTK571NtzIIQuF1pBuIP9TGudgMhYGMJWjWFaFLG0C2n1diQxTBqd7OEA3v69gkH9jA9//WbOs/O5zZYy79AuXoGw+zicsRJxT5ILHbwLR/7RgjaHYRUB89mM9yfSKPKMoVOm5c21nk4ZrHLBpMPM7v8L1TqS/g8ScrVRUxTYjj+WRbPSkhYDLtmzUSVDGw3kFjyqDYqWhfdNDfJKG5XFDX8fuxylfRO3ghYMjfNsszd9z1EW9tPqTKD02Mn4flFbPr2/KAkSXwoNsBau8lCs4YswScTo0Sv8UKFA/uJhu6m223QbBWxqR4CvjSSpGCaOrrR5tLSPxKL7GcwfBcr2WdxOT2EAhblWZJkBpLvp1iZJVefJxrbt+0cEYeLRm97868kQThtSfeMTFk1pFoZlqd1EqO7UwtxOaN97/1w+uMsZR4jV/gLTC1ET+QYSh3Hqe5l/qyJ03NtqZ98+QT7xj+x2QbiccXZM/IJzs9+7XZA2gVueUA6deoUwWCQ//pf/+tlut97PyfqdISwqUFWsy8wmPwAkiTR1WqsrL9AMvKhHfdpdXKEoiFiqX7+btA/Sn5OJjTVv3x3uiXsDgmtvXvtt3oJQjFlMxgB2J2WKnGtaBK7Buv2ivlgMDDI3vH/CUW2kSueZm7lUabGfvWatgg7/p11aNcUDt6toKiXpfoHJS68qlumaYqHdqeEx70VUNudIjblLQrcAavZpzBpcueBX8dh91OqzjC7+DjxWATY+fhar856/nnqzXkkSSHg3Ucqft+modzN4MPxYf49t8x/K53Go9po9DTuDScZuNwNabf5mRr7FaqNObpalXj4PnyeUZbOy8TSKolhi2WndWUWz5n4YlyX5uyz2Una3XxvbZmHUkM4FIW1VoMXN9b4aOzaDa+dJvhCMooq4VVTm9bqRUyKGzo7pbUkSWLA7d1UDL8RUrH3s5j5GiMDH94kTtQaK9QaGTwuiYHkXQwm77u8fRlFkWi21wj6tiYrHneMQFtnplrh7mhyM8Wlmybz9QoPR3Z+ocMJkCWD+fOCXhccbkF8RPCGxdSuYVPdTI58nna3SKvZoFmIU17wUF028EVM4qPX3lfrVbZZ13vcSXq92rYm3dvYjlsekE6fPk232+U3f/M3GRgY4I//+I9v9SneFYykP8Fi5ptsFE7jsPtptnJEQ8cJ+HbuYnTYg2wUN7bJ3TRaG6i2QbptgTe49XKahqWxpdykTcNO77csA8a192m0MkCPcGCSmcWv0+4UcTnDuJ1BipVzN9Tguxr1EsQG5M1gBFZw9YcsZedI8E7mlp9gavznsNt9aFqdueXvEAleu9C9G+h6i2rjEncd/i1U1coPRoL76KYqrGZfJuJ/aNs+hqkxu/woiegUk6NfxDR7rKw/z8Lq15kY+vmbHizcqsqn0uPUehodQydsd26uXHS9TaFyhk43j90WIBo6gt3mp9sGU5c2gxGA3SGRGlXILErXLcIDPJIY4uncKv/twinssqXkcV84RdJ17QCv2qHd3K5a0G6BYr81moNuZ5xo8C5eO/dXhAITGGaXWj3DcOoTrGS/w57QRzd/63WnyJVO4XbFMYWBLCmYwqBcnWci8VPMl1t8Y3meY9EEhjD5YS5LSHHg6bhZXZUwDfD4BaHEluirVcO5NTYqV+ByRHA5IkTCsNsGXZcjRqW+0JcVqNaXcDiit4PRLvCmA9KZM2c4e/bs9gOqKg8++CCf+cxn+KM/+iN+53d+hyeeeGLXDyMS2d2MDCAWe2vilzcHH+nUb9Bs5+j1mng9qc3+j2v9vtKIktn4PqNDD6IqDnLFc1TrF9l/5P1kZmXCcRW3V8YwLHXgSFImPbD7iOT3Cs7+UEPGhsNl3d9eV1Ar9pg65sHt3ZnvaqDhrDtYzDzO+PBHCPqGqTZWOT/zL6h2hVis33Tweve5XdaRFXC5+l8lu71HOGzHG76H2cUyp6b/Oy5HEE1vko4fY3Twwbf0gdYbdfy+KD5fP0c6GhlhfnmO2MT2a17PnSQSTjE1ueXvEwp+hldP/yUOV5WA782RLN7Y+9jpVnj9/JcJBcaIxw5Sb66znP0Kh/b+Ak5bkopX32wUNEwdYRrY7OB0OojFbvz8fzkRpK3rdAydgN2BfCNV+aiglutRWpdJjylIEtRKglqhx8G7vX0r7JvF1e9GLPYgk9oxytV5ZNlGJPh5FMVOuf4CstLBdVmOZyB1iGzhJVbWn8XvS9IzDWZmTmEzDhLzT/JrA4JXCxs8V1hDliT2xyKkG1EKa4KBUQWbHfLrJvklwf67bCjKzV//2zV2yOojXFr4FnabRMA/TK2+ytLad9g/+TGi4bd2znd2vLs1uNlrloTYjS7Cdvzpn/4pf/Znf7Zt+/T09OZ//93f/R1/8Ad/wDPPPEMikdj2251QLDYwt7mTbUcs5iOfr+/+gt8F6EaH1ey/U2/OAxAKpgh678PrTlPegELGsvvuaQKXV5AcE5dVii0ZEn+kv2fJNKFRhm7HUk72Bq2GwNK6TCRpDTTFrIE/eu10HVg07DOX/m+OH/ktouGtfP1S5gfMLT3D0f2/s7ntRve53bCUj/cf33LXbNUFF09oOFLfo96axuUM0+oUcDuHGUl/bNc9FqYJtSK066BcthK/QpboaBVOT/9fHNr7WVTVhU314XGlWMu9gqBJyLs9lbqSfYqAP0I6fnff9rmlbyNLkVuW419aexy/L8pQasuaPV86y+r6CSaGvsDc6zJ7j0qYcoae3kBCYW3WjZB1RvcMXOfIbx49DdbnJbSWhKxYSbrEiLmrtFatsUi28CIdrYDdFiAWOkYkeHDX32CueIJmZ479E59DVZ2Yps6lxW9SKs+jykHauffh9wwTigRoVCzfosG9YnP109Ng+hWZ2ICMwykRTki4vDB3xsTu1rcRgW6Et3vsqDUW2Ci+TFcr4bCHiIfvJuAb3/x309QxTA1Vce16UvZOjHeWHc4rrOWfRjdaOGxRwoGDuJxRQv69N61AfuWaZVna9ULjTa+QvvSlL/GlL32pb5umafz1X/81hw4d4u6778a43HyjvgWRyvcyVMXJ6MAnLQqp0EklY5svVSgBgZiJ1raCj6JCZhZ0TSEUk+k0BfOrJoN7LYHGngYrFyUcTssqubJhUli1+h08AZNayQQBA3t3lvm/Gh5XEiEEsiyj6x0kSabVKeB0hJHYntq5Hlxei+Z+7oc6obiCqQsqBRMl/AJCqm6m1HS9w8X5r5IvneyzSbgWDAOWL1h/bygm0W3D8nmD5LiJJyBYynwLrztBrniW8eGPIkyd+ZVvUyjNcc/RL9JsbD+mwx6k0Vzv2yaEoNZcIxF588reb0SjtczkaP8qMxo6wMzC4wg0YkN2Tr2UZ2DMTSg0TCkHGA2a8t/SbP/0jrI/bxU2u+VL1NMEpmEF9t084lpzidWNJ5gY/hhB/xiNVpa5pScwTZ1Y7P5dnTsWPko3W+HVM3+B15Ok1c7jtKc4uOc3yM7bSY3bGJy0UphCCBbPy+RXeyRGLNLM0nkJp0shHFfQNcHCOZPEsNVsnluTCCdvTdrxVsHvHcPv3a4PKYRBZuNZyvVzSJKELNlJRj9wXRbsO4mF1W+QK73IUOo+bDYX+eJZ1vLfIRG5i2zhWUbSn3rbWzVuaaSw2+08+uijPProo/z2b/82X/7ylzl+/DiRyLUVr38SIMsq8g63Wpa3CAylDZAklQN3y8iXXULLOYmVWRg7bLKxBNGkSnrcSsOlhcTKJZn8So/UOLhD62SLL7KRyWJTfURDR7eZBGodyK9Cqypjb/4GKzOCxOgckmRit/lQFR82W4CuaeCQld3P3gbBHzFpVExUJ4wdgemlMxwe+qXN+o6qOhkf+ihnpv9xVwGplAWPT2XswFatJRCRmD+nE52YR1EERw/+z2QLrzI9/016epN2p0Iycj9uV5RmY/tsMhw4wPTCK6zlXiYRPYpp6qysP4dpgM99fRXsm4Es2+np7T57dd3ogCQjIeMKVBC+74D+KxTWTbxBib1Hwixl7qFQfu1tCUhXcDVru6tVyBZepNlaQVHchAOHiYaO9D33XPFlxoc+QiRkCX0GfCPsG/9Zzl76CvvFB3Z1TkmSGUo9QrJ3D+1ugWQkgNMRQgholGUmDm49Y0mSSI3BxRMyiRGTWhFsNoWhSZlwQrZYdQmZcy/3iKRAUd79YHQlyXSj7yWz8QMMUePYwf+Mzeal3swwPf9vqIoLv3f0HbjSa6PXa5AtPMfRg79OJLiHRnudfROf4tL8Y3S1FlMTP8v0/Dc5MPE/vK3tFbd86fInf/In/N7v/R6/+7u/y6FDh/jDP/zDW32KH0s0ShIDY1vBCCAYk1idlei0oFmR2XN4698kSSIxAudflml1siysfY2xwYcIBT5Nu5Nnbvm7NLUGg7F7kCUJQ4flCzKJIYXxAxKtVoKV2Sb5+RhT7/PT7hQ5Ofso57QoLy5PY5dl3heM89Auc8AOF31UWN1o4XT054OcjiC60d7V8VpVmeE9/VpvvpCEokg0ajX8vmFkWSYdv3szBbeafYFms3mtQ6IqTiaGPsda7mkWV74PSAR8e5gY+swtLTgHfQdYyvyA/ROf3VQAWM78gIB3D7Ks0tOb+IIqI1Nb9SJFlXA6I/T06esc+dZB69WZXf4yA4m7GB/+EFqvxuLq99F61T4L8k63hN4cZm7NsIJBXCYQjWMKzQqyNwGbzdvXI3WlWHClzUHTG7Q7G3RaGrWGSqnSoFM5QGLIUkvtNAVOj4TTLeH2SmTmDMYO73CidwjNdpb1/LM0WyvIioOw/xCp2Ad27OkzjC6V+gWOHfov2C7z+/3eQUYHH2Rt4+S7HpAa7QyKaiMZu5N6M4PHFUeWFAaT93Dy7F8zNf45nA4/jdba27pKuuUBaWpqikcfffRWH/bHHgKubnHa+TdC0GhtYJo6XncSCasYsFF4mZH0B0lE70QAZUPClniEU9P/wNMtL8fDaeLNMIGwTGrU+voD9jC2gyavP1fg+R/+MzVKFGzjfO7wJ4m63GTbTR5bnidW8pLg5t0rPa5BCuXzxCNHNrcVyufxuPqLW4bRpdqYRwgDv3d8S+RTEtvkl4QQGIbA6/BTbZzdllqsNzK4neNcD05HmPGhz1z2i5L6GJC3CsnocZbWnuCV03+O3zdIs7WBqvgZG/j05WuI0GoX0bQ69qscUcuVWVzOm5dU6moVNgov0epkURU3kdCdhPx7r7tPvvQa8cjBzaZdlzPMgcnP91mQCwG94r2UTRgYVREmZFcMirk6Egqq4gD6H9JKq86FWpmuqTPk8nEwEMZ2Db8tWQaXT1BYF4SSTZqtFfzeQVp5D/F0kULlKaSOj4A+jjcgaFYttqAkQ6UgCCWv7zf0dqKrVVhY/Rrjww8RDX0BrddkMfM9ltefYHRgu/FiT2+hKs7NYHQFblecXq+27ffvNOy2AIauYZhWG8CVz6p3WZ8TrEmweJN2ILvFbfuJdxDXu2bTgGreKtheGWQrBaiWTGKDglo1z8L6V6h3pq28/vrz1AtBXK4ADeO5y3lfD7lOi64w2BNKodcvcOfgXTxXLOCqeUjHnX3d7DabG73rw+fZw4lukofT95AMuJBlCa/NTsTp4qXCOvvcN9/Q4bCHWVh5HDARCAqlcyxlnmMo+bFNZYBqfY65lUeRZR3drJPJfh9ZduB2JRFAcc26H1dWjflVQbtlkhrxky+doqOV8LqTCGGS2XiRYmWeoeQjeL2uG74bVkPm20PDlSSZoH8vfs8EiuQiHLyDROTuzZmzLKsYpklm4wc47D5MUydbeJX13DmGUx+7qZ6orlZhZunLRCMTDA/cj8cVJpN7FtMw8bjT19wvV3qZZPwILsdW45Oi2ClVZ7DbolZrQxX0VoiBfRkCYTfegJNAVGP6dIFQNEQqNdp3n18r53i1mmd/yEPaJVholHi9XGSfP7JN8ucKnB6rdloplbApMUprXuplwcQhD6HgECsbT9GrHCKckHH7rHproyKoVw2G928NnI2eRrbTAok+NfKd4PE4yK5oZBclihmZVt2qqak30YqWzb9EMDDAQPJeJElGVZ2EA3tZzHyPgHfPNuKOLNvYKL5CJNTvQZUvnUXXDYI3mEC83eOd3eZlo/AyPb1KJLiPdreIzebh0sLjBP2TyLKNtY2TDMQf3PUk7rb9xHsYoTisVg3OvSwIx2U6bUG1aJEahNBpql8lHXqIWPgwNodEuZLh3KV/5uC+X6RdCtJoZXG5opR7XaaCYUy9ia53GPBGeSjl4pVShYGan+hVY5QQgloJOh03o/YBzIKT8wsmA5MS4YRMwuWm3N3uMrsbeFwpJod/gXzpJPnSLDbFi8uRZCHzNSRJxusep1K/yB1Tv4zXY11Up1vm1IW/w+seJBCN0G7onHlRxh+W6bYEmiYua73JTAx9lkzuGV45/RcITPyeCSaHP7/N3vvdhNMRwunYudM1Gb2HXFHhzPQ/0TOaBHxJBuIfw6beXMNwrvgq6cRRhlIWwcDjiuN2xTl14W+Jhu64piSU3ean1c4R8o8jhElXK9PVqpSqlwgHLMPIRkUQSzlwuWJUa6sYZhdZVoklk9jlftZsx9A5WcnxuUE/XpuG0xFiwufmX+anObFxnntTO+fWHC6rRnrq1Iu43T+FLygY2iejKBJ2EsiOIl53h3MvO/EFZfSeoNMSDO41kSQwheAH+VUW23XiLjf5Qpukw8Uj8eFtavib92zVoJhRGNqj4PRYFPiVaYOhfeaum9K1Xpl49FjfNllW8bjjdLUKDvsbJY5U4pF7uDD7L4wNPYLLGaVcnWVl7QUmhj6/u5O+zTi493/kzKU/I5s/jcsZpNZYxeEIoyp+zs88ylDy42+7rfrtgPQjAkmGwb2CVs2gVTewOWH8iMW+q9aX8HpCjIweotcVGLogFEwxNn6YavMc0fAxFlefQFE9SHIIXasys/gt4pE7UBQ7MZeg6lmhlLMUiiNJCcOAtXlBrSQ49H6Voktm0Aty18bFEz3cPsFSr07C9eaVNpyOCEOpj2CaPaYX/p5YdB9TsU9gmjrzK99BN2p4rupqdzpCJGKHKVUvko5/gNQYaB2TVt3EHbDU16/MiFXVzUj64whh9RS915oOu1qZfPkEo4MPEvSPY5hlZhf/HSRpV5YNV9DqZhlM39m3zeUMY7d56GplXM6dHeKioaMsrH4VlzOCJAkkBNnCa7idETIb36XRXCZX6BHVj+AMN1AVNz7vKIpsp2AaqLb+VF220yJiN/HZVUKBic3ncTwt89TF73JP8sA1i+GKCr5ok0Aq06dU3+4UkVCJpVXCcZNm1URRLeHUK5P0E6V1WkLjv0wdxq6o6KbJ46uLvFhc50Ox7RR6IWB1zmDiyJY6+JU+rOKaycCebbvsCEtZf4VwcGsHw9BotDZIx8I77hMPvw9VcbGw8iw9vYHbmWR86OdvWq7r7YLLEeH4of+Tan2WVidPyHcPutFCkVzsG/2VHTUSbzVuB6QfIUiSNei+0X5AN9o47H4kaasPB8Bp99Np5/B7RkjHHmJ+8UmWGuvkPX5GEnczmLTqA4v1GlG3g+GUycayYHlaAhlsdpPYgEUzl3sulht1BjxewimZM/N1XpaX+OyeffDmFkmbKNem8bgjjA5sUaHHhz5KrbFIvbmyaSgIlkeNEFvFcruz/2/efs/eW4HoCjaKLzGYOr4pp+NyTaLIQabnHyfgndj132VT/bQ7hT4fLMPoovUaqDustoSAZhUalThu8TFOn/8KyA3czhiR0H6OHfoMldoCJ879OXce+F/IzgziVhQke456cwk6kzSqJonR/uM6FYVSu4rLua9fDUIoOBSFZnsD73VSiPHwceaWn0RRnPi9Q7Q6eWYXv0UsdMxKidkgcNW4bZo6a7lneDFf5XOT+2k2ZzAdCZyOMA+lBvl/L57lg9H0tvuoa9Z39kZ1cH9YIrsks1tFhlj4Ti4t/iMOR4BY6CBar87C6lP43OPYbdc2tAoH9v/I0Lx3giRJBP17CPp3GZlvMW4HpPcAvO5BssvPouudTRq1ECb50gVCfkuC54pJoFnd4FS9TMIzSrXXY7Fe49nsKp9MjOJwWr0oV2iq1Txo7cuUapsdCdhoNSlpsFCu8fC+QXYzIBcAACAASURBVMb8gbfckNfu5An4R/q2OewBXM4o1drSZkAyjC4bhTMMJnY2LPxxQrOdYXz4gb5tfu8wpuigG61dp+6ioTtYyjyJ2xXH606h6x3mV76DzzO2zfJBCFi73OsWTcm4enup5v4DydEuY3vGNn2bGq01BpN3EwoncUzJzJw2cDijVGotZKqMTLl4I08h4XCj95qcKBb4QDqIJEkUOm1eyWUZt7VuWHcI+CYwzR6X5r+FptdQFRfR0DHi4WM7/n4t9wymqOML7GEkfhg7OpXaArKs4lZ96MLERKC8gSmkqJdlujSw2SV0vU27m6ecFzS6DVod/45Cxm+E3eZnYuhzrOefY2H5KRTFSch/cJsVzc2iVoTimozWAZsTIimzLxD/uON2QHqXIYRAN1rIknrNTmiHPUjQd4Az03/PQPJeFMVONv8aXa1Ls7VGtT6L25UkEjjIwWASl+ripfU16rpG1O7ik4nRPpfYK5NGtx8KqyY9zVKM8Nvs+BQb7ZrBB0cd+G6RLq7DHqLeyMBVZQdJUtG0FouZpxEIZFllo3Aaj3MEj+vaM+kfF6iKm45WweXcSu/oegthGjdFavB7RtDDH+DcpX9BliV0vUPAu4eh5Ee2/bZeBkNX2H+XgqxImKZOS15gdQbi6QB+n/WAulodpz1gic9GJA7eK9OqQWflAsHwAC7vdjVuSZL4eHyIJxZPcKbSwGNzUOq0ucOlYe/0cDtvrNQSCkwR9O+77G587T440+xRrp3n2OHfZHY9y/lykeOxJD5PmmYrz0qjR8Lh3pFIISsQTSssXugxMNmhaywjG0mqaw7iQ1UWVv+VkfSn8LqvI3VyGS5nlPGhn73h73aLWhEKGYXRKQVvAJo1y/lWCIPYzpnXHzvcDkjvImqNRdZyP0A3mwjTwOcZZzD5yI7SOun4h6jWZ9jIX8QUOorsRjcyOJ123K40pcolLi2dYs/wFxj3Bhj3Xsd29DLsTvDHTC6eMEgOy8gy5NcEkmy8aaXknRAO7OfiwstkNn5IMvY+hGmwkn0eWfawd+TnqNQvIUSTdOzDeN2DaG2J3IpEq2bJ3PgjJrGha1tRX41GBUorPaoVGYfHJJLiujYc7xYiwTtYXP0+7kmL0WYYGvMr3yXom7ppYkY4eIBQYAqtZ60sFMWBEJYEVbdlPWe337o3sbSMrEg0Wzkuzj+KyxkkMuDl/MxXiEbGGR/6KbRejXqzwviwNUGSZQmXr0vLuMiAa+cVC8BQ+AAf766xUn8a1RFkqLdOI7+M3zNBo7WCz3Pj5mNJkpCk6w9LutFBUWzYbR4+kBjg0cVL1DSNAbeDS7kcM60qn0iOXHP/oT0KtZNtXn1mA68njd3uJT0uEUkeIVCSWV1/nj0j200O324U12RGpxT8YSsQ+0IwdkBh/ryAG9ih/7jgdkB6l9DuFlnJfpu9Y5+yitpGl6XM91nKPM7E8Gc2f9frXnaWbSmotn3EE3tx+wUX5v+GqYmfI+S3+m7ikcPMLX2bjeIrfY2NN0J8CJp+g3LBQJgSvojAH7k5q/UbQVEcTAx9jkzuaZZWnwYk/N5JJoZ+HpvqxuPe6r3pabB8UWZgXGHvHRJ6DzJzJmuzBoN7r98PUMlBKaswcUAmOqhSLQqWLxgM77+2f827hXDAqjucPPdXOB0BBE3syiDDqQff1PEkSd5kdpkGrM5IGD0Zf0iiWBTkVkzsTks2SAjB7NI3GB96iHj0MMWNDh1jnoXMN1jPncRhj2KaJrNL3yYRPYKut1lef46AZ1+f+sT2a5AYTn+E5tzfoEgdBoc+TiR0gGp9kUuLjzE59AWcjp0L/jcDm+qxamGtDSLuBL84OsWrMxVey1UwDRuf2DtJ0tm/yuy0oJCBTkNmw9/D4RUQfpS9+/9XFNlSqxcCIsEpLs2/O/5tWgfeOI/0BKztb1Jy9D2H2wHpXUKh9DqR0F56eotaYxm/d5ix4Y/y6uk/p9Mt4XSE0TqWukIsrZAclui2BJl5A1+sCugEff16WfHoHUzPfwvYfUCCq4kUb99L73SEmBj6OUxT3zRlu4Ke3gQkbKqbSg4iCYX4oPXvigqjB2TOvGDSbQkc10gjCmGJ1e69UyESV2i3Ddw+yRKcXTNJ3zqpulsCSZJIxd5PPPw+OlqJoC/J8ozE2qyEwyMIxa2emzeDfAacLpWxY1v6cKuzMuVCj9yqiTNYwhQ6scghtC4I004kMoHJw6ysvcLU+K9hGB02Sq9yaf7byLKNkP8QkeChG5671cmBZHD0wG9sPuNwcA/p+PsolE8zmHxw2z7CtIR0ZWV3EyFJkklE7uPC3L8ykn6Y/NwgadXOxKiEQz1IflWlqBlELhM4u21YuSgzMKYQ3C+BqTJz1kansJ92q0ogEEfXoNMQ4Ci+I2yynWB3Wk63/qtidrNm2d1fnb4UwqTeWsEwOnhdA+/a9b4duB2Q3gUIYZIrvYzD4UKWBBuFHEIIpsY/h9sVRetVcTrCFNcgPqiQHrM+bG9AwuWVuPiaB8OrX861bz1CXW+/KaO5dwpCCFqmwC4L7JJFdljJfhetVwYETkcCW+tTpIb6R2JTdFAcNeaXThOJe4kED26rt/W6Vkrv6sZfgEBUIpfZPXvqnYaiOBDdFPPzdoIJSxqnWhAsnjMYOWBiexOPs16S2Xf0Dfpwo5DLyIQSBhdPaJhemVoJepqJxy9QZBW73Y/N5kWSJFTVxUD8g8AHr3uuN6LXq+NxxbaRGNyuONVav12NaUJuGWoF67c2B8SGdqdAHg0dRlVdzE4v4ZJcjB+QcLsGUBUn4bjg/MuCYMyiiZeykBxWiA9Z53G5ZIYmZIob97JR/R7R5M/g8XupVhqcvfgkkcidNzj724NwymTposHogf4aUiS19e52ukXmV7+Ow+7EbveT2fgekeDRXelDvhdwOyC9C8iXXsPriTM1/tMEA2MIIVhee4aZpW/SaGYZiFssn05TZmCsf4B1+yRsqoqQh1nNvsBQ6oNIkoRhdFlZf46Q/0cz2bzYrPF8cZ2eMNFNkyGni0TjSfYMPUA8cgQhTLKFk8yefxZv5eOEE9bgoWlV6q01Os046T1D1LvnmFk6zeTIF/pqbYoNdB30noCr0nOdJqi2H910h2UBL7P/qA272+rtCcUkVmehmDFJbheNvvExTbax4GQZEBAdgGAswrlZjWZ3lmh0HEmyJknruZP4vW9tKelyxlnZyGAY3b5JQ7k6h+sNxIbsAqiKyqH3y9jsUCvBwnmdgT27kwQK+iZpOPcQH1DxX+X95XBKeHwyrbqJLwTdlkx6+A1LL0kmEotgVwZ59fRfYrd70bQG7fJhwmN3vaV7cDMwdKu2JwSXA7HB/FmB1gG7CyJpk+BlQoMQgsXMNxkZeD+JqBU0e70mp6f/Hrcz2Wdx8V7F7YD0LqBcv8DE8MfR9CaN5jpOZ4hE9AgXZr9GMvrAJuVXUS1/JPdV+qaGLtB7MDzwAMvZr1MoXcRh91GqzhH2HyESfBfVJq+BfLfN04UMnxoZZ9jjQzNNnlw+zwVGuf/yhyVJllBqvvAVsqtVXJ4QkRRUqlkauVHCUTsDAxFgipnFx7bZWCgK+EImy9Mm++60AlC7Kf5/9t4zOLLsPNN8rkvvfQIJDxQKZbqq2juSzTYkh05s7dCMRqJmJc1oVrsdDO0GY0MhhrSK0B/+2AgFgxppRVEz8hJFSqREI47Ituxmd1e78qgqeI8E0vvMe+/ZH7caKDTKAFWoArqZTwQiCrcyzz2ZyLzvPef7vvdjbtwg3Ll3BUlvAsKqg6lf5lUaSkicPKGzlMthCEGv20fYvrU+Uu6Ayeq8oHNg/SKcWRI4vQJZBptDZrDvI4zPfYdCdT8Ou59M7jzC1OiK33GNka+PTfPid+/j7Ng36O54P5rqZiV7kmx+iuHeX9zwuisFmTseXO827A9DR69CbnnrHnWKarUtvxwhBM2G4G0HIdUmqJYFbv/6+2HogmYDhnseQpLvodEsYlO9nMorSNLtWU3nVwzS0wq+kIokQXrKREjW30i7pOWXb9vW6mkkWRALr/ft0jQ3qcT9pFfPtgWpzY0hTB1NdeN0dFGtL1PNjwMyNs1PIvLA2uMCccH8uIHTo+BwSRiGYPaiwOU3cTq9DHZ/lvHZfyRbuIjLGadYHWNxxUMy+uCeKhg9XVjl3liCHo9VMGhXFD4Qi/BnpSz5RoOAff1O2ucNYVfGySwfZXK0Sb0pM7DPTsdlK8VY+DATM88AG7cp4r2wPKXz1k+amMJAbwnCHeaGPfm9hqxYvZ/Md7ScP5PN8Kw8y51yCE2W+e7yJPvdQe4LX998NZqCmXMGtYrAF5KoFC0bqq7hdWH2ursY7vtlsvmzVCplIoH78Hn6dsRsNpV4jNXcW4xN/wjTaOBx9TDU8znUyzost5pgd0prYtSsWzVxhiGTW5YJJc3r9vUC8EdhcdzAH7FcwIUQrC4ITHNd1EIJwcK4gd0p4Q1Csy5YnDYuNQIVFFY0DCNMriGwOYxtxe4azQKr+ZM0mzkc9giR4JHr1pAZRoOZ+RfITA6y75jA7VVxOZMUlt1MnjU4+KCE3SVRysHEaZ2uEUsgDbOFqjg2fbctI9x3l6/n1WgL0i7gdfcxt/QiydhRTLOFqjip1nO4HMk141GwgputhsG54wKbQ6LZELi8JslLN0ILKy/gcrk5MvJ/oyh2ms0SZ8e+SSbvIRLcme6nO0FZb3HQsTHNzaG58CgGhea6IAlhki2M0xF9HG+noN6oMj77fToG/g/kDUkQVWR5c82WLEOyH4IBG0uLZTT71lLFdxNFBZdXMDtmEOu23MuLtSY/np/l8yP76QhY79t90QR/fvEsvW4f8etcqTUb9B0yKayaFHJWsLz30OYkCU11E4/cc82xao0Mq9m3aOlFHPYY0dDR615wJUkiGjpGNHTsqo+xOaBRu9QwUJdo1iXcfplS3sDjk5gdleneb141ieVtXF4IJk3OHddxeSVaTTBNk86hdeNVlxdiPSZTozpGC2x2sLl16lWJV34I4YSCLAtWFgSRrq2vpiu1JSbn/ol49DDh4CEKpWkuTP0VA12fuaqHIcDU/PcQtR76Bwbo7rbRaBbIF6exOYbo6LOyQ+NuCV8IIp0K+bRJqttqrFmr56hU07hd1ra+EILllRN43Tewt7sHaQvSLqCpXsZmv0fLKBENjVDInmJy9lmGe//TprufcBKCMZNm3bqgvO1IbJo6+eJZ7jr862t79Tabl76uRxmb+rc9JUhRu5OJYoF+72UN62Qn6UqFQvoZysq9CGEyt/RTFNmLx2XZ4DjsfmxakPmln5JKPIQkSbRaFWYXfkIkeN9Vz6dq0p5L874WiT5BflFw6iUdh0vmdD7LYNK3JkYALlXjSCjKeDl/XUECa+UVvH4t6jUpVqaZXfw+HfG7cTsHyRUnuDD11wx2f/aa6d9bQVHBHzG5eMIkGFGIJCVyaYPVRYN9xxRyK4LM4tayI0Nxa6xayRrX4d6crecNgidgYugQT9hYWWlSzkqM3KPgdFt9tvoOSYy+BpWCscm+60ospJ+jv/vRtRYrkeAIDnuAxZWf0Jf6xBWfU2tkaLRW6Y59GrOlIkkSDnsAp6NKo5pBURPUKiZLMyathtV9uVKUmXC20DwqHbEPcvrC35KMHcNu87GSOUNL10nF995W/Y3QFqTbjBAmq7nXuP/I/0mtkaNcWcTl6GBk8LMUCpNEgodpNawtDJvD2kuWFTa5EJtCB4lNd6tWE7zqtuZUzEA+LWHoEg63Sbjj2v5x2+UOf4Rvzo9hVxVG/CFKrSYvLM1zV2Q/DmY4N/bPSJKM37OPztTdG0S5p+OjTM59h3TmDA57gFJ5npD/CAHvte36302oGozcpTE7XUdvmiRDgpUrPO52RsKEECwsP8u+vo8T9Ftmr6HAPjTVxXLmFbqTH9r2mOXqPOnMK9Qaq9g0P9Hg3QgGmDhjklmScfskBo/I2J3W6iA9d+XsyIZhMF8ro8kyHU4PiiShKFw3O0+SrPdaliUqBXB5ZULxy5VLIt4lU8hcX5BMU6daXyAS/IUNx6Ohw8wsvHTV5zWbedyuOIGwwtgJk44+yyXFprko1nPMjRmoNoioMqWsSbUo09FrbeFNj8qk9o3Qn4qSyZ+iVM7idY8Q9A3fchfu28V741W8i9CNGgIDn7cHn7eHOFZQv97Is7h8kvkxqBVlnB7JCsQGTJK96+7Gb6PIdlTFQ740uVYcC5DJjeJybL0FdmYRSqsKnYMKdofV+Gzm3KWU4+335bsiLlXjUx0DvJ5b5luZCzgUlRFPkBFfCEnqpiP28FWfa9O87Ov9j1TrS+h6lY7oE9tu0fBuweECXDBo+Hh1dpFMvUb40lZnVW9xIrPCR2I712r9WuhGFd2sEPBtXKJEQwdZWvnbbY9Xqswxs/hd+roewefppVJdZGL2R0SDTVzeg+y/W0bVrOLUehWyi1bac37FMlV9+x7lXDHLS9lFki4PTdPgmZU5PhTvJrHVvhFvI668nSvJ651sr4UkyciSim7UscnrGRitVuWapRcOe4TS8jyavUU4qXL2VZ1op0y1VmduzINoCu5/n41axUpVr9dMJs7oDB1WEEJheU7QvT9CKvHB7b3edwltQbrNKLIdU5ibuoVWayvohQOoIZU7HrLsXQxDMHHaZHVBJ/oOay1JkuiIvp/RsX+hM/YQXnecUm2SheU3GOj69JbmYhqQXZA5cK+C3Wl9411eCWFCZtEk0btTrxp8mo0Pxm6s9bEkSbidWxfZ7aI3rdRbJGtr5zr93W45TkXl4VAHfzV2juGAldRwLpflgCe4wZPwViLLGsI0MIzGmqEvQLNVRpG3vx+6nPkp/d2PEQ1ZxbUOux9NczE6/j2Cwf1Mj8p07ZNpVmUadcguGXQNKpRWZWplnWQfZBp1Xskt80tDBwhdyjgcK+b515lJfqln/1WbAF4Jlx+WJk1qZevmDyzT1ZV5k+AWmvZKkkzAO8LU3NMM9n4MWVIwTZ2p+WcI+g5e9Xl2mx+fe4Bz49+kp/MR7G4Pc9Nz5ArTBHzvJ5yQqFWt7FpfSMLhlnF7JQpZQSAqMXVe4vaulW8vbUG6zciySsh3iIvT32Oo9+PYNA+VWprxmR+h6p8lNWiJEYCiSKQGZc6/IRNNbdy6EAIq6X6U0pPMN96k2TyLJCIM7/8cTsfVA6qXY20LSmti9Db+iMTCiQw58zucmygjSyHikfu31afnaghhUixPUW9kcNhDO5bZdaPkli2HB39IxjRhZcYk0W/Vr+wmw74gnU43Y+UChiH4ZLKP0E7uo14HRbbh8wwwNf8M/d0fQpYUWnqN6fnnCPmv79jwTuqNlU3OIl53F7pRJtLfYnVW4q3nZYQp4/ZDap/VJNLQBad+KtOomYxWshyNRNfECGDQFyDsdDFdKW3Jv3Ht9SlWosPoGzqRpIKqQWbJxOYw8Gzxb98Rfz9T89/jtZN/iMedoFRewO3sJhG5enwToCvxBOnM65wf/2cMs4nH18Wh/gfJzKnYHRLeoMA0LJeRy2NhjRqo7/Er9nv85e1NOmIPM7/8Aq+f/pNL/X9MosEHyK56N22T2exW8dw7yS6BMFXufrgLWelGCMHipCA7r+Me3todlKpBsyEwDIGirH/yC/k8jdY0RwY+SSTUzVL6Ahenvo8sKfg8N57No+s1xme/harK+LzdZAqvsbjyIgPd/35Tq4TbQaNmGVoeuEddE+VKUebCmzrOO8wbtu7ZKTyajaPB3bN5TsUfW7vgupwRypUlAt4DN5Qwo2k+KrU0AW3981NvZJElDVVRSfZZ2XFen0yyd729vKJKBCJWkWsTk8gVlq9uVaVlbr92yB8Bp8fKRmw2IdptZeRdr2KiWQe9BXaXjYGuJ61EhWaOeOgDmzrFXglJkomF78bn6UNg4rRHkSQrJT09ZxKKS5dshASyIqiWBE4PjL5u4o/tTceRnaItSLuAJCmkEo/QEXsI3ahfuhjLrE7NMztpdc9UFRcuR4Lcsh3NISzXZuf6l6WUkenZv76akiSJRA8szUroLbGli6lqA5fXZPaCoGvI+vJXS4Kp8zn2HR7C644gywpB/wAD3U8ws/DqTQnSwspPCPg76e/68JrH2tT80yykn6Wn46M3PO6NUsxAtEPZsEJ0+yR8YZlyziRw/bY472kUxc5A989Tb2Rptop0RiM37JsWDd7F+MwP2d//87hdMeqNPBenvkckcGxthWzdfEmbMk3rVYHfAz2al7dyqxwNx1AuPabcajFZKnJnLM7iBFSLsuUQHzEJJa4vLjYHm7bDr4ahw8K4RKNq7SrUK4JgwiTSGcZpD2/5vajW08ws/AAkHUlSMAydrsSH8QS6KOcNzrwCobhMpWiSnjNx+QRnXm5h9+pr/nzvVdqCtIvIsobtUquB+eXnUXxFsgsfwa50IVxlLo4vkZ6K4w9pLIzJIJl0DArsTssHTHmHPYwkgyxtLSj7Nsl+WJpscfJF+ZL9joHueIlY8hc3PM7rSdFo/utNvd5C+QJ39vzaBo+1VOIBjp/4KkKI217Ma17BYges99V8727TXxHdNDlVWGWqWkKWJIbcfvb7QsiShMMe2rZLtyEE5VYTh6JiVxRC/hEMo8npC3+HwAAB4cAx4pdtb/kjMH3WIBCR8AatG5bMoqBRE3gC4MbHuVKWvx0f5Y5QhKZp8vrKMkfcUVYv2IkmFboGJfQmzE+YLNX0tZq9a5Fr1jmeXWaxUcUpqxzwhTjoC236PC5OgsujMnxMQpYlmnXBxRMGmt3YchM909SZnPsn+rsfJRK0Yk350iTnx7/DcN8vk+xzUysblPMGDi+M3G9ZQXWkbOTy743i12vRFqQ9gG7UyRVPcdehX8fUXazMC8qLHsrFBUIDL3H40OOXvpwyc+cN+o+YuHwmqwuC7uH1L01+1fJ0285Wk6xAxyAYuonestLMz01kNxTfAZTKc9htN2l5IEwkaaMCSMiIXQrSeoOwPGkSS13mGNAQ5FZMeq8el37PYQrBdxcncdk1Hu7oxBCCV9KLLNYrPBbfflbf2UKGV3LLaIpMwzDod/l4X6STaOgIkeBhdKOOqtg3fRZsDkj2W1lliiphGgJJEaSGrSJXCYl/l+hlvFxgPJdHlSQeiXRiz3kQIYXOgUuxSBcMHpE59aJMs25es4Sh0Grw7YUJ7o8neczfQ77Z4LnFOSp6a4Mrht6ysl+H7rDEyJqvRGe/wsKUiT+ytc9woTSGxx1bS+4ACPr6iYSGyRbOEg/fg9PDJuskVds7ziu3krYg7QEazTx2u5+SUCiaJaQOCX/TQbDlZTG3CFiriUiHxOqiSTlvmWTOnNNpnVLwRySKhSKzc2/ijeZYWg0TCd6xrbiMor6dXSYRDd3DhcnvMNj7Uez2LnKFccZn/o1UfHMX0u3g8wyxmH6Fns5H144trhzH5x7YFasjpwccXoOzx63GdaYJK/MGocTOpby/G5isFEGReLJncO3v0Ovx8bXzp1ht1Ihso8p4qlLkzeIqnx0YJuZ0UdN1/uf8NC+szvPBWBeSJF/zc+kJgPuISaNqrfhtjo3bbrIkMeQNMORdj9XMzUnE3tFkWFEkPAGZeuXagvRWbpWj0Rj3RC3x8dvs/HzvEF8/f4ojgQiOSzErQ7du9i6PtYI1P13feuZby6jisG+OMznsASqV8pbGuBmEEJQq02SL56hU51Fkm9VtOnD4lmaybpW2IO0BbKqX6cIiwVaZuDuIEJCptWjW5jdZkNgcEoZurYJ6DwryKzqLC2my9W+RGjqIz7uPQmmGC1N/eV0Lk6sRDR5DlhRGx/8ZJirIUpDO2GM3FT8C6Ii9j7GZb1CuLOP3dlMoz1GpZhjs/sxNjXujSBIkeqFaMihlDSQJOoc2FyG/11msVdjnD264KVBlmQFfgIVaZVuCdKqwygeSKWJOS3ScqsqHOnv4k9GTPGgY2N+5z3wFJGl7fwPNJqiWLMeGt2k1BcWswHGdkFe2VeMOz8bEEY+mEbQ7KFzacgRLeEwDKoWNJq25tMDp2XqigceVYnrhdXpTLZRL2/VCmGRy54kErp2dtxPMLz9DuTZFSy/idobxeDoxDYPphX8hFrqfSPDmzHVvlrYg7QEWmgYltYNG+hlsPR/BbvNRFhOcmXyRu+9cv1gbuqCYMem+9MWTFQglINt8jv09j6xZ0kdDh3DYX2Jp9UV6Oz++7flIkkQkeIRI8AjRqJeVldKOvE5NdTPY9UusrFykWFzF4x6iK/7xXa0ylyRw+6yfn0UqBailVRZbDRqejSuSXKNO3LW9G5qKoW9IywZLlByKQs3QtyRI2yUYt2JPbr+EPwSlnMTyjMBoQW5JoZg1SQ2JK8YLvaqN5VqVLvd6TWDTMMg1Gngv2/uWJKtX09gpnWSvZXZcWBVklg16ttHxxeWI4XKkOHPhb+lM3IcsKSykX0eY2k3f8F2Paj1NsTJGZ+JuytV5Dg59BjBZzY4Sjxzj9IW/J+jfv6s91dqCtAdYqFXo734CpzHBm2f+DN1o4HLEMY0nmL0QRusVGLpgadrEE9q4BWGaOpXaHNHQ5zaMGQsfYXbx5dv8Sq6OEFYr9tySDbvrEM2aQPYIAt6fseyBPcTqguXScSwV4zvZUcbzQbrsPtwBwel8hkytRk+ki5ZeRZFtW7pxiNmdTJQKayskgJV6jaZh4tVuTR69zQGdgyZzYy3OFyQkZGJdcOwRBUWFybMyK3Mt4j2bn3uHP8L3l6cJ2x30enxUdZ0fLczQ4/Tgekcw1h8BzW6SXzbRWxIOt6D34Lq/5FbpTn6YbOEscwuvIYSJzzNIV/yOW16PVyxPEguPUCrPk4gdvrQiVnA4gpimhMcVo1JdwOfpvaXzuBZtQdoDuFSVQrPFQ10foCv5foQwkGWVC9pFKuUK8xMKkgz+mMD3sKf+0wAAIABJREFUjuxSqx24im40sF12wdD1Koq0y4U0l1HMQCWvcPA+BZvDCljPXhQsTbbovPl62zbbRG9BblHm4P0KNrvKz4f7+eH8NNUFkBQDt13mfV64OPnnGGYNIUwC3hE6449cU5iOBaJ8e2ECgH6vn2yjzvNLc9wbjG/LSeGdNOtQWLVadXj84PZvjC25fNC5TzD2hsQdD0s4nOvn6uiD0ddl4j2bt9ZiDhePRDr58dwMZaMFAobcfh66Si64y2v9WFmhVm2dYSrbWlVIkkw4sLWW8DuJLGu09CKK4qDVWve7NE0dhI1mq4Ki3L7i6ytxU4LUaDR48sknGRwc5Ctf+QoAuVyO3/qt3+LVV19lZGSEL3/5y6RSW0z0/xllnzfI385eYNyfp9/rR5IUzuYzLNXKPNaXQpOvvoqQJJmgdz/T888w2PNRJEnGNHWmF54lcA0Lk9tNPi3ROWCJEYCsSKQG4cSLMoZu7rpdz88a1SJ4gzI2u/X36HJ7+dWhg4zNVsmvGKRCq8wu/QvD/T+Hz9NDS68yOftDZpd+RE/HR646btDm4FMd/byRS3Mqs4Jb0XggmKDvMrdSIQSV2iKV2jya4ibgG0KWr37zVMxYXXUjSQWn0+qkm08bdA5ZotSowtKURKMmUa9INGsymrZuAaWoVor/1eh1++hxeWmYBposo0gyQkCrAbK6Xl4hhFWQnl+WqdQWaWg/RnNlUVSBx9VLKvHYrhR4b5Wgb5jzk39OT+fDzC29QigwhKJo1Oo5ms0awgSX4yYt4m+SG74MTE9P86UvfYnx8XEGB9dvcf/gD/6AEydO8Hu/93t85Stf4fd+7/f42te+tiOTfa/iVFQ+Eu/mR3PTyLKMKQQKEh9L9qFtoaFPR+yShcmp/4bHlaBUmcflSNGZvPVB0q1i6NKmbCdFlVDVSxlMbUG6rSiqleJ+OZIk4RVOsLXI5N+kq+Nh/N5eAGyam8Gej3H85B/S0qvXvPAGbY6rposLYTI1/z0arTThwD7K9QUWV1+gL/UkLsfmSmTTgOUpmeE7VVxeSzxjXRLn35AoZnS8QZg9L9PZb7WwGD9tUikAQsIbslYxqwsCt//aiQeSJK0lMBQzsDJrfe8Mw2pbkei1jIjrZZWeAxXG5r/NQOIJHOzH4WmykPkJk3PfZqjnP+yp5piXo6luuhIfZnr+h5iiyXOv/D+4HBFk2YkkOejr/OSuz/2GLwNf+MIXiEQihMMb95BeeOEFHn74YT7xiU8wNjbG17/+dXRdR32vmzDdJB1OD7/QNUymWUeWJIKafcsfjrcr6mv1FRrNHNHgQ9suZLzVOD0m+bTAdcnIsl4VFFYFus7PVIr1XsHlg6VJq+dQKGG5I9SrguVZg45BmF0t4XZuvFtWFDt2m5+WXr7hlcBq7hSS1ODOg/9lbesvnTnFzMK/Mtz7S5s+89WSZfj7thiB1T4i1imxuiQhTIHHLxPttAQkNSgzdsKgmJWolKw29sWsSffI1uZXLVliNHBYxeO3XE/mxgTzYy3qZZkD98qk8yeJhQ+Q6jhEvSqoVRz0dj5KLv81KrVFPK6O659ol/B7B/C4f41SZQbDaCBJCjbNi9vZsetiBFsQpFOnTnH69OlNx7/85S8zPDzMo48+uuF4Op0mELDy7IPBIK1Wi2w2Syz2M+7DsgWEMPAr4lLR4PY/HE5HFKdj97zPrkW4w2qr3WoKKkWoVwAJjBakZyDWvdnmxTRbzC68xupiAWHK+ENOunuOYr/MJb3NjSFJkNpnMncRlmYkVA2qJUGky2r97SxFyRcn8HnWt9sbzSL1ZgG7duPN+fKlUXpTD26IQ0VDh5ief45GM7fpRkqSrrzdZpogyYJmgw1i5XBJ7L9b5sIbJoszOpEO6Du89RV4bhk6+hQ8l1K7VU2iex+8+ZzlZGJ3SjTSBYJ+awWo2awW8ZIk4XbFqNXTVGuLNFoFXI4YQd/+PderSJFtO2KUfCu47jv17LPP8tWvfnXT8fPnz1/x8TdrARMOb90rKxp9912YrjRnIUym519gYfk1wAo+9qY+QCK6N7q+7tT7HI0KTrzUxO6Q6L9Hxu1TkGUYfaOFaCrEUhvzcl955QfUVkYY6oui2RTmZ3KcP3mG9z/xMJp67eDre+Wzcavp7BaUCwLTAI9/3bHC432Et87+D1xOF5HQCLVGjsmZHzPQ8zCJxLpPznbnPL9iw+Gw43SuJwEIIbDbNUJhF27nxvHCYUFuoUWjohKIWKsgvSXILbfoGXYjBMxP6Dgc2tp1RzgEqlYjFBsFxxK6FCMcvANNXa+nutq8V6ZauLwNmnoeAKc9gNPpwR+yBFsSGqFAikJpklTyLloNE7tDxmY3KVUnEeZFYpFDxPwJsoVxZpdPcmTkl9C0m48t7fTnQwiTfHGKll7D7+nCbt/52oftzvm6gvTUU0/x1FNPbWMCUfJ564+Zz+fRNI1QaOvbR5lMGXMLRmI7WR9zu7janBfSL9JoLTAy8Hkcdj/l6iKj4/9EqagT8O1uZ9SdfJ9NA5pNmf13q6iaiWGaGCbEugTT5+vI9vVb4VJ5idxcFw98sBO3z7oQJbqjHH9uiBNvnKKnb/9tmfPtYi/MOZu7/DeVztiTLKVfZnz6p2iqm3DgCE7t4No8b2TOmtzN5MxPcdg619KcM/nzNBqCSslOtbx5vFAKzr7ewuOX0WyQXzXxhk2aZgOAWk3i3Osm8W6rwd/CVJ359Bt0hUt4tQSZ7DRjky8y1PM5bJr3mvNeLY7SmArQP2JV5mZyE0hmiFIhiC9qcvY1QWpomFzhZc6c/yEe9Qiaq8bY6RcollY5OPTptXrAkP8oE7M/5PTov5FKPHrF822Vnf581BtZJua+jd3mwG7zc6b4LwR9h+mIvW/Htu7enrMsS1teaOz4WvLBBx/k6aef5rvf/S4/+MEPeOCBB9rxo2sghEG2cIKjB/5XHHZrK8TjStLf9QRTcy/uuiDtJKZhbcG8c/tEu0KLjVymRDAcWRMjsDLzkj0uZqaAW1tD2AZw2sP0dn5sR8eMBI8wMTfJW+f+O5HgPmr1HNn8BL2dP3fVC6HLCwNHTEo5E8OA7hE2JMh07xesLrQYO2V9Vkr6S4zcGSERewiAeOQoMwvPsbTyIt3XyBCs1tMY9p/SLP0K+SUboZiMYoQ4d2Iaf0Qm1uUlu6QzcVpBb3yG+ZWforj/Bs2u4fMM4LAFiIYPr40nSRLJ6N2cvvB3wM0J0k4ihGB64bt0d9xHInonAC29xunzf0WhlNjVa86OK8UXv/hFstksv/M7v8P+/fv53d/93Z0+xXsKw2gAYpO/ldsVo6UXd2dStwjlUipuMQv+y3JhcmmB07cxUGC3+ai16gixMbak6xs7mLZ5dyHLKgNd/wvF8iSV2gKqGmV/3/tR1WvbE8kKV3XUlhWIdUGsy8Qwm5y5+Dqx6P+14TGx8FEWlv/HNc9RLE2QiA+RDGssz5hcOGGg2cAZWkZxzyNJ9xBOQjhpAi7gsUs/1vc4m38LYRobbPhN0dpkIrvb1BsZBC3ikWNrxzTVSSr5AIvpc+9uQXr66ac3/B4MBvnjP/7jmx32ZwZFcSJLGuXKAh73enZOvjiJ0/7eSgSRJIh1m0ye1Ul0KzjdVmvm7LKxKQsqEosxPzbL4vwiiY4okiRTKmWZmyzTPbiFHtNt9iySJOH39uP3bqE3xHbHRgZJwjCayJeJnGHUL/OOExgGyPLGmx1JVjGMKnantMFFf2wqgyxfO+ygKHbczi7ml1+mu+P9l85jMrv4IgHv1beXdwPTbKHImxOnFMWOeWkbdLdo76XtMpIkEY88yOjEt+nvegK3K0a+OMnk3LP0dXxqt6e343gCkBo2yS2Z5FYl7M4r268oisTgwRjn3yowNzGBqkExayMU9+MPt1dI73WqRShkAGF9ZjzBazfb0/Ua9WYWm+bD7xlkZuF5+rs+hCRJmKbOzOLz+L0jpGdh5kyTes3KmoumzDX3k6BvmAtTf0Eydjcup5WtWqmmWc2dZ7j3lzecr9kqUShNIEkyAe8AquoilXiMidlvkStM4HHFyRUn0ZQgnal7b9G7dGM4HVEazdLaTbAQAt2os7TyJj73zt8kbIe2IO0BwoFDKLKd6fmXaLaKOOxR+jo+hdu1u3bwhtlEN3b+jsnh4lLjtGsnr/hCDg7f76CYMTF0QUePwjaMp9u8y6hXoNWESh6qJYV4yur+ujJvUsoaJAc2i5IQgsWVF8kWTuByhqnWszhtCQrFPG+c+f/wuJIUSjM47Qk0/T5aLZWD99owgXIBJk7rKKqJ2w82zUtH7FFOjP4lQV8PAsgXp0nFH9/QKXcl+ybLmZcIB4cwTYNzk8+Tij9O0DfMcN/nKVWmaTSLdMVHcDkTe6K+53JkWSUVf4wzF/8Bp8NPqTJLo1nAMAz6Oj+1K80y36YtSHuEgG+IgG9ot6cBWHd/s0s/olKbxT6joshRUvFHcWyjTfNOoSgQjN1a08k226dWX2Vh+SIz82doNNKoqptw4Cgh/4FtX8z0FiyMSegtK4suuwiHH1Twha1xwgmJs8ehWjRwv6MEKpM/Sa0xw52Hfh2b5sYwW4xPf59Wy00kcAeNVp6Q727sWoTxt2QO3i/jcEnUahLeAHQNKSzPCdx+6+Yo5B/B5+6lWJ4EINn/OOpl/m61RoZ09hWOHvjVtSSkSjXNyfN/hcfVhaa6brlr904Q8O2j1lhlJfcKQ70fw+ftw9BbXJz6LnLWRix8567Mqy1IbTYghMH47DdJRA9xaN/P43I5mJo7zvjsN9nf959QlLatws8yrVaZqYXvUm+sUG8uk4zdTU/nv0MgMzX3NC29TCKyPcuq5Snw+FW6hiQyiwK3R0JRFBpVE7vLyq6MJGVKuSsJ0ikGex/Dpllp2oqs0Zd6nNdO/RFdicfWdhmadSuh5m3vvrdxeiRajY0N9lTVSShw5Z4S+eJ54tHDa2IEVgJSyN9PoTS26/2EtkOhPMbBof+A33vJBt0GQ70f4/SFbxANHduVVVJbkNpsoFCexG5z05V8GLCW98noXRRKM2QLo0RDe6NYt83t4/JMx+nFHxAO9tHSY3jcD5GM3ku2MIbbEWNk4NO8efbrREPHtux+bRhQKcoMHrbsi2TFSjhw+yRKeQm7yxIKvcWGfkamaZUKtPQadttGlVJVF5IkY5jNNZcE1Wadq14VOC/b9i3nBTbn1lugCGEiX8FFX5ZVhNh6o769QLNVwONKUKtnmF38CcXyDKrqpFieRAgDSbr98tDeC2mzgWazgMe9OYvN44rTbBV2YUZtdot6FWbPS1w4LnPxdZnZ8TyNZoZU8iGqtTQh/wCKYsPjSlBv5HDY/dhtXhrN/JbPIQyQpXWx8UckygWTUsFEiLfnIVhdMPBHLHFcnYfxt2Rmzsro6U8wNrqAEOuiki9NoihuVGVdeWQZQkmT8VMGhYxJqylYXTSZHzcIJ7cuSH7vAMurJ9H1+tqxRrNIJncBv3dgy+PcKoSw6v3EFl6SwxZhOXOKE+e+ht3hZGTwE6QSd+JyBllIv3DrJ3sF2iukNhtwOmIsrjyNSJlrlfRCCPKFSXyevdPOos2tpdWE2VGZ1IBC6KiE3oLxUR2j6UGWFOw2P+XqMi5nEkXREOjoRp1ms4RN3br9l6KBrFkO2/6IlV3Ze0Dm3KstHG4JpwfKeZNot4ndabV/qBZVRu62YkHVcoLTry9w+sQbdPWHqNTSzC2+TCr+oU1bTqEEKKrB7EWdYkHH4RZ0DAmcW58ubmcSn3uYN89+nXjkMELoLK+eIha6H5u2u3ZVuTRkF2R03Yq9hpImwfjVsxMTkft56+z/y0D3o/R2vA/DbFJv5OnueJDJuReJR+677e002oLUZgMeVwpZdnFh8p9JJR7AxMHkzEvUG1W6k3sj6aLNrSefhnBCWXPRVhTYdyjGsy+UKZbSJKJ3MzH7HWxaAJBAyIxNfw+f20qB3iqSBPFuk8lzOrFOBafHSvfWnIJg0kBRINaz7u6RW5IZOmKJEYDLY2fkSIpTxxWmbM+hqT76Op/E5dzc10eSIBCF6AEbKys3nj3aGX8/gdoQ+dIYEjJ9nf8ep+MqVbtYq5VKwerbZHNaaew7HZ7Jr0BhWWHwDgW3T6JaFkydNQCD0FXK9hz2GLpex+WKkS9NoSgOQoEhPO4EU/MvUG9k2oLUZneRJIn+1KdYWn2Fs2P/iN0uY1O6Gez+9J5zLW5z62jWJKLv6KKgqhph9yOcPv/39HbfRyR0gOMn/xjdqOO0xwn5DpFKfGDb53L7oWu/SW7ZpJgXNJlBDp2lpEsE3QdR1C7AurC3muB0v+P5PhW7kmSw+zM3+nK3P2dnErfz+mUZhg4zoxKKIuMNyOSXTVbnTLr2C9QdbOicW5Lp3W+JEYDLI9E7onDxhCCUuHJsq95cRZZVNNVN7HLLI2Qq1SVs2s6brV6P9hWmzSZkWaMj9jAdsYf3hOlnm9uPzSko5yF02ULDNAVy8wCpbh+5/EnsDoPO2IcI+PZh13w3dcNid0K8x2R89p9QFJNE5AiGUWd++X8S8B4mEbkXSQKHW1DIQuCyBUkxC3b3xqBJs27VGckyeIPbawDZrFtbiEKAN2TVzd0o6RnwBVW691lJG0JIzI3JpGdadOxgyMlqw7HxmNNjCfg77bfW/t8eQZZUZhdfwu2MEA4O09IrnBv/R0xTbEoWuR20BalNmzabCMZg8rSBwwXhpBVDmh83sbsEoVAHITp2/GYlX7yALOsc2veLa/HLcGA/b5z5GuHAQTTVTaRTMD2qYwxaPYvKBcHcmEG8d12QVuahsCwTiMq0dFiZMUkOmHgCVzvzOrk0rM7JhBMKsgLzFwx8EZNo6vrPvRLlnMzB+6S1eJYkSSR74cRPZES/eanfk06hNEatkcFhDxHwDl1V3OuNLPniBQRW7aLzUm2g3Sko5TcKdblgmdBebXtQU90EfQfQ9ToTM08zOv4ddKNBubJIX+fuuMS0BalNmz1IrQzlPCDpVM2XKdVPYphNvK5eOmLvu+VFyqoNuvebpGcFs2MSsgK+sEnHLezrVqpMEQsfXBMjAJvNS8DXQ6kyQ8g/YsVf+k3S84K5MQmbU5DoF7gv7S5VS1BalTl4v4pms67E5YLMxbd0Bo6aG1LH640My5nj1OpLaJqPgOcY2bkBRu5W12JU8W6Js6/qeALmtpIf3kYIkN6Ry3y5QLT0CmMz38Dp8OH3dlEsnWN59acMdH96U5JEOvM6K7njl94jiYnZfyASuIt45B7CHZZQi2EVb8CKWU1fMAh3bN6uE8JyxTAN2N/3nxmd+jor2VE0m4tWs04i8gFSiQ9u/8XuAG1BatNmDyEELE1BvaQQiluGssWVYfoP3EMkYSedOcnYzDcY7vs8muq+7ng3g90FXcOC61k87RSybKelVzcdb7YqeF3rBdluH7h9V55XMQOxlLImRmA1HvT4Zcr5dd+6WiPD+Mw3SCXvpTf1ANV6mosTP8btbuBwrdfaaTaJaIdCKbtRkIQQFMsTFEoXQVII+kbwujcvozxBS9Q7B9bnk54TuAPW6mgh/QLR8D56Oy+1p0jCzMJzLKSf39D6o1bPkc69wrEDv4rdZqlvR+xe3jz7Z/i9A3iDISTJZGGyRbMmoTkEkZTA9w5P2EYN5i9KyLKMqkGtLNPV9V9x95VptIo4bMHrOq/fStqC1KbNHqJSgGZF4cC9CkJq4Krk6Rs+wIU3dCIJmWTsHqq1VVZzp0hG79/t6e4oocBBpub/kUjwAE6HpRyruXPU6gV8nT1bGkOYm1ckYB27vDZnefVlujoeoDNuuUq4XTEUwpw+9y2EOLxhlfbOmh4hBDOL/0pTz5CIHsU0debTP8TvGSEcOMxq7iTNVg6bFiSQuIPFMT/VksAbtLYYK0Vzzd2+WL7IUN//tmH8jth9zC1+ZYOnXCZ3gWhoZE2MwFo9RsMHyJcukrDfZ5nQBtaF2hLNKfLF8wgEPs8Q2ckhOvtUwklrG7FWEVx4U8fu8uDx3MAScIdpC1KbNnuIUhaiKRlFlWg0G9g0Nx6fjNsvUcpZMQKft4vllQu7PdUdx+WIEQ8/zFvn/hyvO4Gu12g0a/SlPrXlnkLeEKzOm0Q6JBTFupjXq4JS3iScEmTz55hLn2Nx5VXi0V9BNxqol+ywAoEkhikoFkr4A1ZAv9UQrC4adF5W8VCuzlJvLnPswK+uxXpioUMcP/lVljPH6YgfIxQYoVieY2Lhr+jb9xla5TCVCjg8Vhr7+tahZKnoZQiukBUnceVqVyEu/edm5pefo1KfoiN+NxISM9NnUIwQ4eS64avTLRHvUsiv3NiW5E7TFqQ2bfYal647imKnWq8ghNhwLSqW57Dbbr/R7e0gHDhEwDtEuTqPLGt4XJ0bVivXw+2HUtbg3KsQTsoYOqwuGERTJpnicUqVUfYPfphGo4Yh6hTL4/g9AyiKHcOooGhNxk6ohGMmsgzZtEEgZuK4bHe0WJ4kHj60IfFA09y43RFU1U1/1xMIAR77AWiFOD/6Ewa6fo5Y1+YEA793iLnll+lLPX4pC08wt/RT/N6hDYW9keB+Lk48S2fifhz2IAD1RoF05gxDPb+w6X2oNTIUK6PcefDX1xpa2sRhpsfn0Y3qhu1ezQ6mvtHPb7doC1KbNnsIXxjSMyahhISqOVBkJ4sLc5TyIbqHJRbTJ1nJjDLc9/ndnuotQ1HsN9y8T5Ig0QfVkkE5ZyAr0DUCqr3J3PhrHD3wKwQDUVLJGrOLz9KbeoRaYxW7LcTYzA+IhveTCKuUsi2ECV372dTyRJY1Wnpt07nrjSy94bsQAso5CVWT6Oo6wvTsKebOy6gOk94RKzb3Nh3R9zE2+03Klb/E701RKM3TbFYZ6P70hrEddj+JyEO8dfa/EwntByRWs6PEww9gt21OHyyWJ4kERzZ0V/YFVZpVF+VigboxSjY/CkjUVoeJxQ6wF5zk2oLUps0ewuUDp8/gzCuCUEyh1UqxNJ/GcP89r55cxOvuY6D7M7c8oWEvI4SgpZdRZNsV3ecl6e3Eh/Vj1Xoem+Zec+kOB4ZptSqMjn+HRqOAzRYi4BmmM/6I5Xt3jabEQd8IY7N/RzxyFJfTyrMulGbIFaY4NNRJq2FZIHkCcP7NFh75EfoOaeRWBDNndaI9JgGr/x+q6mK49xcplCeoNzKEfMfweweuuCqMBI/g8/STL10EAUM9//GqtUKKbKfeSm84pmoSNv8Ur79xlnini474gzRqJlXby+SbkwTFx3e9d1NbkNq02UNIEsR7wB81KedNVAccuCeMqn1ut6e2J8iXxlhIP48QTUyzhc8zQCr+2HXbothUD81WGd2oA5YTeSJ6JyYmS8tnGOh68ppjmIZladSogM0ZJB78ACdG/wKfpxPT1KlUV4iH7mdu6QU6Q5/E6bKRXdKpVIocuteFz6vg9glaLRg/peMNmSiX4khW19lB8F4/p96meYmFrt+rKOAbYnTiBQqlGfzebgDK1UWKxvfxRTro6/g8EgqeoCCe6OHNs39CpbaIx9VxnZFvLW1BatNmD+Jw3ZxDwHuRSm2JhfSPGO7/FD5PN4bRYHLuR0wv/pD+1Cev+VxVdeFzDzA+/X0O7PsEQqiUKrPMLbxEV+Jja2Jkmpb4KOp6vEdvwvQ5GZdHxh+SqBQFhcVDDAwN0DSnkSSFnqSVBTi18H3eHP1DfJ5Oiukoff3DeD3WcsswBE6PhNsnUy2aeIO37r1SFQfdyY9ybuwfcbsigEyluozPPUA41IPHL7MeM1IJB/ZRqc63BalNmzZttsJq7i1SifvXGsqpqoOB7o9w/ORXaTQL17W6SSUeZz79LK+e+G+0WiaKZCcZ/SBedwphQnoWiquy5Z5ABVdkgWDUTX4+hhFqshwqkpVlhuMh3GmN1TkH3fv3bThHf+qTFLwZZs7nCXlTuOw+QKZWEZgGaDZLmOTbEK7xeXrpi/9nFqdLtOo2PC4nknaCemMZQ7dqkoQpoWiCWj2H27n77TPagtSmTZt3BS29hMu5sSOrLKs4HUFaeum6giTLKl2JxwmFPs5yOoOquNZiJsszIAyVA/dJLGd/wsLSGxi1JIW5LOOZJuXk/dyhdFM3Df5mYpQPRDupl/yWv987xMXvD9M7FGb2vMxYzgDJKrB1+wX5VUGjJnBdFt/S9RrZwjmarSIuZ/ya1kHboVaGxTEbnQMxfCGJWgWmRg+Szr2MjQvEwsMoisTy8hlmJha469hHbvqcN0tbkNq0afOuwGmPkyuME/D1rR1rtspUaxkc9qu3f3gniqJtSAoxDChlZA49IFOsnCdfvMh9d/5XRMtFrtjCcP8ID+PclXwQgKPhGH95/ix34eZqmWneIIzcZ7I4bjJ5ziQYlWlNQ61s0jlkrm0HVutpJma/RSjQj9cbI5s/zUr2dQa7P33duNj1WJ2XSA0qRDqsOdqdMHSHm5effpJ52w/JVZ5BUkCWNLoin2XypJ0D9+9u6ndbkNq0afOuIBo6xsXpv0FRbESCB2g0C0zPP0fIfxRVcVx/gKtgtKxGgZpNIj17ku6Oh7BpbnSgZcC+7vcxMfp1avU8TkeAsN1JTPFSdBeQ5asHgiQJOgah1TDJrKxQa01hjxdpSf04RDeSJDG39G/0dz9KLGyt/Dpi9zE2/T2WMq/SGXvfDb8mgEZVwhvamDWnahIOtZNjI/8F1ZVFQsLpCFMpCo7/uEmjKjakpd9udj/xvE2bNm22gE3zMtj9WcqVPGcu/gOTs88T9B0hGX3wpsZVbVbfolpFYBgNNM1aPbUaAqEYyKqCqjjJLNco5QX5VZNGw3KD2ArZ0nGyjW8RThj4fH4WV55mbunHNFtlmq080dChtcdKkkQydjel8vhNvSYAzS6olTce01sCowWaXcLtjOJyRpAkCb0JNrtEo37lsW4P0OFeAAAPW0lEQVQX7RVSmzZt3jXYbQG6kx/e0TFlGcIdJuOnDLRQL4vpk9iVHiolgdcrM5kdR6dGKBLB1E1yUoWsVKLP13ndsZutIiu549x58New2Sz37kT0Tt48+2dUa32XXDjMDXVHpqlf2ZBvmwTjgtkLBja7gssr0awL5scFsmYyP2YyeMTys9NbgvkJw+qBtHu+qkBbkNq0afMep9WA1QXL2VpVBbJhIOSNNj6hBCiqQWbpCOnK31EufJNkcj+tbI752Z8woYyQX5mlYehMFAs8Gkthk6/vr1coTRIODq2JEVhOFPHIYSrVeZyOOIsrr9EZt4xyTWEwv/QSAc++qw25ZXxhMAyDC29dMls1rfq24bth9FWd/KqJLyRTzJmYpsAfM9uC1KZNmza3ilYTps/KRDsUOnslmnWYnzDQXBB5xwLHHwF/xE6P8Tky+TOkM2dQFTd39v8Ch2Q/U5UifkXjga4Eji22oJVlBaPV2HTcMJtIkkJX4kOMz36LTO4ibleMXGECTQ3SGbt7J14+wRgEoiZGC2SVtYzAkftNFscF6XkTzQHBmEnkBpsQ7iQ3JUiNRoMnn3ySwcFBvvKVrwAwMzPDE088sfYYl8vFm2++eXOzbNOmTZsbILsE4bhC54B1JXZ5IRTTeP3ZFsHEulsCWHGkZh00u51Y+E5g3RHBBRy1Ra95LsNokC2MUm+s4rCHCPkP4PcOsrDyHKXKPF63pYD1Ro7llVMMdH0Gu83PSP8vUyhP0mwVScU/gtuZ3FELH0my4mSX43BB3+Hb1+tqq9ywIE1PT/OlL32J8fFxBgfXLS9OnDiBLMv80R/9ETabDUXZmm18mzZt2uw0jYpEdGDjxd3ukHC4JRpVS6CEgPQMFDMydodEoy7wBk3iPVsP5TRbJcZm/h6fJ0HA302xPMf5yeMMdH+G7sRHOHPhG/i8nciSSq44STLyPhx2q3ueJCmWdVCbGxekL3zhC0QiEcLhjTb4J0+eRFEUfvM3fxOv18vv//7v3/Qk27Rp0+ZGUDSrEPXynkGmaR1TNev3zCLodZVD98toNivIP3nGZGVOJ9a9tfMspJ8nGbuDrqSVqp2M3cPc0kvMp5+nP/VJPK5fpVCeQDcadCcO4HLGdviVvje4riCdOnWK06dPbzr+5S9/meHhYR599NFN/3fffffx+c9/nj/90z/li1/8Ij/+8Y/3RDfCNm3a/GwRjMPCuIHTI3C4W4CNpSkDu0tgu1S6VEjLDB2R19qeq5pE1z6Zc8dlol3rRaz1iuV+oNrAE9iYFFGsTDA8sNHpIBm9i+m55xFCIMs2Gs0cq/k3cdj9NJYKeN19dCU+tCOuDO8VrvtOPPvss3z1q1/ddPz8+fNXfPxv//Zvr/1b13V+4zd+g/Pnz3PXXXdtaULh8NaFKxr1Xv9Be4z2nG8P7TnfHvb6nEVEUKo+z6tvHUeSdYThJBF8P0fuO4qqWYoyrTTwB21rvwM4HAJFgWjUCr5MnNYp5gT+sEQtJyitwP47NezOS51XZ+zYbBIO+3qwptlq4nA6iEa9LK2coGXM8NDd/zt2mxfDaHJ+4l8o1l5hqHdrlj17/b2+Etud83UF6amnnuKpp57a8oB//dd/TTwe5/HHH8cwDAA0Tdvy8zOZMqZ5/UBbNOplZaW05XH3Au053x7ac749vBvmvLjyEnV9jgfu+0U0JUilPs/0wveYmpXwey0zUUmTWJwVRJLrAaNcWiBrLVZXm2SXoFZS2XenvNZ6fHnG5NTxCj0j1rXKoQ1wceJpBno+utb5dWL2aZy2AVZXy4xP/5SB7kcxDTu1WhOAVPxRXjv1R/ic9113lfRueK/fydtzlmVpywuNHV8rPvPMM4yOjtJsNvmLv/iL/7+9+4+put7jOP48P/CoQBwPAh4F64odkqlJDkQzUypbhrvMuXkd/VBmYBYxZrrKtZJZg9pcS2vlDyharY2tH47UPwy9omG0u9shKdBZIHiD41U4eiPP4XA+9w/y6OGAghXfL/Z+bGfw/QBfXuf93Tnvfb/n+/1+SEhIIDk5+Y/+N0IIcU1+v49znd8yK3k1oy29s6pGhcWTOPkBTvxYHWhIMfG9F5D6vBBhNfCLW/Fzcw/2xN5mc+Gckcm3GzGaruxBxcQb+LnJQLdXETYKJsbO58fWT/l3/U6iIhNw/68VlJkpCcsA8Pm6AlOPX2Y2j8VgMNLj98phu9/84bcOKioqYurUqTz//PN0dXXx5ptvYjZLsYUQw6vH7wUDgWZ0WUS4HW/3hcDymAiYPM3PRXc3zY0+3B3dxDv8gRlnlR/6nixsMPSegaf8vcsmk4Wpk1dgj7kPoyEa+/hF3H7rysA99sLHTOJcZ0PQOtwXmzCZxmI2aXw1qo787k5RVVUVtDxx4kTee++937taIYT4Xcym0Rgw80tXO+Fj4wLjHe6fGGMJvqbIMgbsUwD8IesJt/pxnVHcdguB64Pc/wWDCcKuuiG3wWAgMjyeyPDQK0wnjE/nVEsFPT0erLf8jV+6XLT8fJRJsQ9oPm24nsiuixDipmQwGImLTqfhx0+ZkvAA4WNi6bz4E/9xVTMhesmg1xNth9MNPk5+a8Ia0zuv0Pn2nqBpJK5ntCWaqZP/gevcvzjXUUWY+RZutf+d8LH2G3x2NydpSEKIm9b4cTMxGUfxU8s/6fZdZPSoGJIdy/Fesl7/j39jMsOtyYoL53y4O3pnfb1teu/XobCMspJgv2+Iz+CvRRqSEOKmNi7qDsZF3RFYjoqM5OyloZ2xZjSCNab3If48Mh+SEEIIXZCGJIQQQhekIQkhhNAFaUhCCCF0QRqSEEIIXdDdWXZG4+AvEhvK7+qFZB4eknl4jMTMMDJzj9TMQ8ltUErpa8pAIYQQf0lyyE4IIYQuSEMSQgihC9KQhBBC6II0JCGEELogDUkIIYQuSEMSQgihC9KQhBBC6II0JCGEELogDUkIIYQujKiGdOTIER5++GFmzZpFdnY2TU1NAHR0dLB27VruuususrOzaW1t1TZoHx6PhyVLlvDMM88Exk6fPk1SUlLgkZKSomHCUP1l1nudL6upqQmqbWZmptaRrunzzz/n3nvvZd68eezatUvrOIOyYcOGoBrv3r1b60gDqq2tJSkpiYMHDwJw9OhRHnzwQdLS0iguLkaPN6vpm/mNN94IqndRUZHGCa9QSlFSUkJ6ejpz5sxhy5Yt+P3+G6rziGlIXq+XgoIC7HY7r732Gq2trbz00ktA78ZyOp1s3rwZl8vF5s2bNU57RXNzM2vWrOHUqVNB406nE6PRyLvvvktZWRnvvPOORglDDZRZz3W+mtPpxGq1UlpaSllZGVu2bNE60oDa2trYtGkTCxYsYPny5bz++us4nU6tY11XXV0dmZmZgRo/9NBDWkcKoZRi//79PP3004Exj8dDYWEhiYmJrFu3jrKyMvbt26dhymD9ZYbeeqenpwfq/cgjj2iUMNTevXspLS0lNzeXnJwcPvjgA/bs2XNDdR4xDenXX38lLy+PjRs3snjxYmbMmMGZM2cAqK6uZv78+SxdupQlS5ZQU1ODz+fTOHGvgoICLBYL0dHRQeN1dXWYTCYKCwt57rnn8Hg8GiUMNVBmPdf5anV1dXg8HtauXUtRURFms+7uIRxQU1NDd3c3TzzxBE899RRGo5Hq6mqtY12T2+2mubmZw4cPk5eXR0VFBTabTetYIRobG3n22We55557AmNOpxO3282jjz7KqlWriI2N1VW9+8uslOL48ePU19eTm5vLzp07sVqtGqYMNmHCBAoLC8nJySE7OxuA1tbWG6qzLl+p3333HcePHw8Zz83NBaC+vp7Dhw+TlZUFgMvlCmygcePG0d3dzfnz54mNjdU8c0lJCUlJSWRkZIT8bM6cOTz22GPs2rWLDRs28OWXXxIRETEccYGhZ9ZDna82UH6z2czChQtZtmwZJSUlrF+/nv3792Mw6O9uye3t7QBYrVYsFgvh4eG4XC6NU11be3s7DoeDrKwsIiMjefHFF3E4HDz55JNaRwtit9s5cOAAzc3NVFZWAsH1BrDZbLqqd3+ZOzo6iI+PZ/78+UybNo0XXniBrVu36mbPf/bs2cyePRuAt99+GwCTyQQMvc66bEiHDh1i+/btIeMrV67k+++/JycnB5vNFvh8Qyml+ZvNQJkbGxv7/f1NmzYFvvf5fKxbt47GxsbAhh0OQ82shzpfbTD5m5qaeOWVV3C5XMTFxQ1nvEG5fFxdT3W9HofDwZ49ewLLH330EbW1tbprSFFRUURFRdHc3BwY03u9+8tss9n45JNPAstffPEFtbW1WsS7pu3bt7N7925WrFjBpEmTgKHXWZcNKT8/n/z8/JDxkydPsnr1asLDw3n//fcZP348ADExMXR2dgLQ2dlJWFjYsB9CGCjzQD788EPi4uK4//776enpASAsLOzPitevoWbWQ52v1l9+r9dLaWkp06dPJy0tLVBbvR62i4mJAXrrabFY6Orq0myPc7AaGho4cOAAOTk5jB07Fp/Pp9v69nW53h0dHUBv3ZOTk7WMdF0tLS1UVlaSlZWF3W7XZb137NjBtm3bWLZsGS+//DJff/01MPQ66+tZXYPP5yM/P58LFy6wfv16WlpaaGtrIzU1lXnz5lFVVUVlZSX79u1j7ty5uttgfR08eJCGhga8Xi/l5eUkJCTo/oUxEuo8atQoKioqqKiooKCggI8//pjU1NSQz8P0Ij09HZPJxI4dO7DZbPT09LBgwQKtY13TpUuX2LZtG2fPniUxMZETJ06watUqrWMNyp133klERATl5eWcPHmStra2oM9r9MhsNvPWW29RX19PRkYGX331FWvWrNE6VsA333zD1q1bmTJlCpmZmRw7dgybzXZjdVYjxKFDh5TD4Qh6pKWlKaWUOn/+vMrLy1MpKSlq5cqVqqWlReO0oRYtWqTy8/MDy2fOnFGPP/64mjlzplq6dKmqr6/XMF3/+mYeCXVWSqkffvhBLV++XM2YMUPXOS/77LPP1MKFC9XcuXPVzp07tY4zKOXl5eruu+9WKSkpqri4WPn9fq0jDejYsWPK4XCoqqoqpZRSR44cUYsXL1apqanq1Vdf1WX2vpn37t2rMjIy1MyZM9XGjRuVx+PROOEVBQUFIe/NxcXFN1RnmTFWCCGELoyY076FEELc3KQhCSGE0AVpSEIIIXRBGpIQQghdkIYkhBBCF6QhCSGE0AVpSEIIIXRBGpIQQghd+D8svhhoK6J+UwAAAABJRU5ErkJggg==\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "\n",
    "size = len(tsne_factuals)\n",
    "plt.scatter(tsne_factuals[0:size, 0], tsne_factuals[0:size, 1], \n",
    "            edgecolors=[model_colors[i-1] for i in clusters],\n",
    "            cmap='prism', c=\"w\",  marker=\"o\", label='$r$', alpha=0.6)  \n",
    "\n",
    "plt.tight_layout()\n",
    "plt.legend(handles=handles[0:num_clusters], fontsize=legend_size, loc='best')\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Median LogHarzard Ratio"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 131,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "25:  0.024 75:  1.916 50:  0.0 3\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "<matplotlib.legend.Legend at 0x12cfda748>"
      ]
     },
     "execution_count": 131,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 360x288 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "fig=plt.figure(figsize=(5,4))\n",
    "\n",
    "perc_25 = np.round(np.percentile(all_HR_median, 25),2)\n",
    "perc_50 = np.round(np.percentile(all_HR_median, 50),2)\n",
    "perc_75 = np.round(np.percentile(all_HR_median, 75),2)\n",
    "\n",
    "print(\"25: \", np.round(np.exp(perc_25), 3), \"75: \",\n",
    "      np.round(np.exp(perc_75), 3) , \"50: \", np.round(np.exp(perc_50)), 3)\n",
    "\n",
    "xlabel =  'median ' +'$\\log$'  + ' ' +  'HR' + '$(t|x)$'\n",
    "sns.distplot(all_HR_median)\n",
    "plt.xlabel(xlabel ,  fontsize=fontsize)\n",
    "\n",
    "plt.axvline(x=perc_25 ,color='grey', label='25', linestyle='--', lw=3)\n",
    "plt.axvline(x=perc_50 ,color='black', label='50', linestyle='--', lw=2)\n",
    "plt.axvline(x=perc_75 ,color='grey', label='75', linestyle='--', lw=3)\n",
    "plt.tight_layout()\n",
    "plt.xlim((-20, 20))\n",
    "plt.legend(fontsize=legend_size, loc='best')\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 132,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.hist(all_HR_median,cumulative=True, density=True, bins=100, histtype='step', linewidth=2)\n",
    "#kwargs = {'cumulative': True}\n",
    "#sns.distplot(all_HR_median, hist_kws=kwargs, kde_kws=kwargs)\n",
    "plt.xlabel(xlabel ,  fontsize=fontsize)\n",
    "plt.ylabel('Probability' ,  fontsize=fontsize)\n",
    "if data == 'framingham':\n",
    "    plt.xlim((-20, 20))\n",
    "else:\n",
    "    plt.xlim((-20, 3))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 133,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "50 -1.68\n"
     ]
    }
   ],
   "source": [
    "print(\"50\",  np.round(np.percentile(all_HR_median, 50), 2))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 134,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "thresh:  -3.75\n",
      "positive:  0.2504854368932039\n"
     ]
    }
   ],
   "source": [
    "thresh = np.round(np.percentile(all_HR_median, 25), 2)\n",
    "print(\"thresh: \", thresh)\n",
    "\n",
    "print(\"positive: \", len(all_HR_median[all_HR_median < thresh ])/ len(all_HR_median))\n",
    "positive = x0_x1_conc[all_HR_median < thresh]\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 135,
   "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>age6</th>\n",
       "      <th>ascvd_hx6</th>\n",
       "      <th>bmi6</th>\n",
       "      <th>bpmeds6</th>\n",
       "      <th>chol5</th>\n",
       "      <th>dbp6</th>\n",
       "      <th>diab6</th>\n",
       "      <th>female</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>count</th>\n",
       "      <td>129.000000</td>\n",
       "      <td>129.000000</td>\n",
       "      <td>129.000000</td>\n",
       "      <td>129.000000</td>\n",
       "      <td>129.000000</td>\n",
       "      <td>129.000000</td>\n",
       "      <td>129.000000</td>\n",
       "      <td>129.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>mean</th>\n",
       "      <td>55.558140</td>\n",
       "      <td>0.069767</td>\n",
       "      <td>29.326328</td>\n",
       "      <td>0.317829</td>\n",
       "      <td>202.147287</td>\n",
       "      <td>75.922481</td>\n",
       "      <td>0.116279</td>\n",
       "      <td>0.465116</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>std</th>\n",
       "      <td>9.412348</td>\n",
       "      <td>0.255748</td>\n",
       "      <td>4.800124</td>\n",
       "      <td>0.467448</td>\n",
       "      <td>40.368053</td>\n",
       "      <td>7.184632</td>\n",
       "      <td>0.321809</td>\n",
       "      <td>0.500726</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>min</th>\n",
       "      <td>35.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>20.777429</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>118.000000</td>\n",
       "      <td>55.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>25%</th>\n",
       "      <td>50.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>25.995640</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>176.000000</td>\n",
       "      <td>71.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>50%</th>\n",
       "      <td>54.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>28.835150</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>196.000000</td>\n",
       "      <td>76.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>75%</th>\n",
       "      <td>61.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>31.847777</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>225.000000</td>\n",
       "      <td>80.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>1.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>max</th>\n",
       "      <td>84.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>45.135681</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>312.000000</td>\n",
       "      <td>100.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "             age6   ascvd_hx6        bmi6     bpmeds6       chol5        dbp6  \\\n",
       "count  129.000000  129.000000  129.000000  129.000000  129.000000  129.000000   \n",
       "mean    55.558140    0.069767   29.326328    0.317829  202.147287   75.922481   \n",
       "std      9.412348    0.255748    4.800124    0.467448   40.368053    7.184632   \n",
       "min     35.000000    0.000000   20.777429    0.000000  118.000000   55.000000   \n",
       "25%     50.000000    0.000000   25.995640    0.000000  176.000000   71.000000   \n",
       "50%     54.000000    0.000000   28.835150    0.000000  196.000000   76.000000   \n",
       "75%     61.000000    0.000000   31.847777    1.000000  225.000000   80.000000   \n",
       "max     84.000000    1.000000   45.135681    1.000000  312.000000  100.000000   \n",
       "\n",
       "            diab6      female  \n",
       "count  129.000000  129.000000  \n",
       "mean     0.116279    0.465116  \n",
       "std      0.321809    0.500726  \n",
       "min      0.000000    0.000000  \n",
       "25%      0.000000    0.000000  \n",
       "50%      0.000000    0.000000  \n",
       "75%      0.000000    1.000000  \n",
       "max      1.000000    1.000000  "
      ]
     },
     "execution_count": 135,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "if data == 'framingham':\n",
    "    end = 8\n",
    "    positive[sub_covariates[0:8]].describe()\n",
    "else:\n",
    "    end = 12\n",
    "\n",
    "positive[sub_covariates[0:end]].describe()\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 136,
   "metadata": {
    "scrolled": true
   },
   "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>gluc5</th>\n",
       "      <th>hdl5</th>\n",
       "      <th>pad_hx6</th>\n",
       "      <th>sbp6</th>\n",
       "      <th>smoke6</th>\n",
       "      <th>stk_hx6</th>\n",
       "      <th>mi_hx6</th>\n",
       "      <th>trigly5</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>count</th>\n",
       "      <td>129.000000</td>\n",
       "      <td>129.000000</td>\n",
       "      <td>129.000000</td>\n",
       "      <td>129.000000</td>\n",
       "      <td>129.000000</td>\n",
       "      <td>129.000000</td>\n",
       "      <td>129.000000</td>\n",
       "      <td>129.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>mean</th>\n",
       "      <td>102.201550</td>\n",
       "      <td>43.953488</td>\n",
       "      <td>0.007752</td>\n",
       "      <td>123.782946</td>\n",
       "      <td>0.170543</td>\n",
       "      <td>0.015504</td>\n",
       "      <td>0.038760</td>\n",
       "      <td>164.139535</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>std</th>\n",
       "      <td>34.450912</td>\n",
       "      <td>11.543979</td>\n",
       "      <td>0.088045</td>\n",
       "      <td>13.923879</td>\n",
       "      <td>0.377575</td>\n",
       "      <td>0.124027</td>\n",
       "      <td>0.193774</td>\n",
       "      <td>78.358625</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>min</th>\n",
       "      <td>75.000000</td>\n",
       "      <td>26.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>99.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>46.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>25%</th>\n",
       "      <td>90.000000</td>\n",
       "      <td>35.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>114.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>119.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>50%</th>\n",
       "      <td>95.000000</td>\n",
       "      <td>43.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>122.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>143.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>75%</th>\n",
       "      <td>103.000000</td>\n",
       "      <td>50.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>131.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>200.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>max</th>\n",
       "      <td>289.000000</td>\n",
       "      <td>95.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>170.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>468.000000</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "            gluc5        hdl5     pad_hx6        sbp6      smoke6     stk_hx6  \\\n",
       "count  129.000000  129.000000  129.000000  129.000000  129.000000  129.000000   \n",
       "mean   102.201550   43.953488    0.007752  123.782946    0.170543    0.015504   \n",
       "std     34.450912   11.543979    0.088045   13.923879    0.377575    0.124027   \n",
       "min     75.000000   26.000000    0.000000   99.000000    0.000000    0.000000   \n",
       "25%     90.000000   35.000000    0.000000  114.000000    0.000000    0.000000   \n",
       "50%     95.000000   43.000000    0.000000  122.000000    0.000000    0.000000   \n",
       "75%    103.000000   50.000000    0.000000  131.000000    0.000000    0.000000   \n",
       "max    289.000000   95.000000    1.000000  170.000000    1.000000    1.000000   \n",
       "\n",
       "           mi_hx6     trigly5  \n",
       "count  129.000000  129.000000  \n",
       "mean     0.038760  164.139535  \n",
       "std      0.193774   78.358625  \n",
       "min      0.000000   46.000000  \n",
       "25%      0.000000  119.000000  \n",
       "50%      0.000000  143.000000  \n",
       "75%      0.000000  200.000000  \n",
       "max      1.000000  468.000000  "
      ]
     },
     "execution_count": 136,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "positive[sub_covariates[end:]].describe()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 137,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "thresh:  0.65\n",
      "negative 0.2504854368932039\n"
     ]
    }
   ],
   "source": [
    "thresh = np.round(np.percentile(all_HR_median, 75), 2)\n",
    "print(\"thresh: \", thresh)\n",
    "print(\"negative\", len(all_HR_median[all_HR_median > thresh])/ len(all_HR_median))\n",
    "negative = x0_x1_conc[all_HR_median>thresh]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 138,
   "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>age6</th>\n",
       "      <th>ascvd_hx6</th>\n",
       "      <th>bmi6</th>\n",
       "      <th>bpmeds6</th>\n",
       "      <th>chol5</th>\n",
       "      <th>dbp6</th>\n",
       "      <th>diab6</th>\n",
       "      <th>female</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>count</th>\n",
       "      <td>129.000000</td>\n",
       "      <td>129.000000</td>\n",
       "      <td>129.000000</td>\n",
       "      <td>129.000000</td>\n",
       "      <td>129.000000</td>\n",
       "      <td>129.000000</td>\n",
       "      <td>129.000000</td>\n",
       "      <td>129.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>mean</th>\n",
       "      <td>60.666667</td>\n",
       "      <td>0.139535</td>\n",
       "      <td>26.375823</td>\n",
       "      <td>0.286822</td>\n",
       "      <td>197.581395</td>\n",
       "      <td>71.627907</td>\n",
       "      <td>0.046512</td>\n",
       "      <td>0.604651</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>std</th>\n",
       "      <td>10.185263</td>\n",
       "      <td>0.347855</td>\n",
       "      <td>5.325558</td>\n",
       "      <td>0.454041</td>\n",
       "      <td>28.029197</td>\n",
       "      <td>11.657260</td>\n",
       "      <td>0.211411</td>\n",
       "      <td>0.490832</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>min</th>\n",
       "      <td>37.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>17.676632</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>118.000000</td>\n",
       "      <td>49.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>25%</th>\n",
       "      <td>53.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>22.687889</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>180.000000</td>\n",
       "      <td>62.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>50%</th>\n",
       "      <td>60.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>25.285077</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>198.000000</td>\n",
       "      <td>70.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>1.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>75%</th>\n",
       "      <td>69.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>29.230393</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>217.000000</td>\n",
       "      <td>81.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>1.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>max</th>\n",
       "      <td>78.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>45.992112</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>290.000000</td>\n",
       "      <td>105.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "             age6   ascvd_hx6        bmi6     bpmeds6       chol5        dbp6  \\\n",
       "count  129.000000  129.000000  129.000000  129.000000  129.000000  129.000000   \n",
       "mean    60.666667    0.139535   26.375823    0.286822  197.581395   71.627907   \n",
       "std     10.185263    0.347855    5.325558    0.454041   28.029197   11.657260   \n",
       "min     37.000000    0.000000   17.676632    0.000000  118.000000   49.000000   \n",
       "25%     53.000000    0.000000   22.687889    0.000000  180.000000   62.000000   \n",
       "50%     60.000000    0.000000   25.285077    0.000000  198.000000   70.000000   \n",
       "75%     69.000000    0.000000   29.230393    1.000000  217.000000   81.000000   \n",
       "max     78.000000    1.000000   45.992112    1.000000  290.000000  105.000000   \n",
       "\n",
       "            diab6      female  \n",
       "count  129.000000  129.000000  \n",
       "mean     0.046512    0.604651  \n",
       "std      0.211411    0.490832  \n",
       "min      0.000000    0.000000  \n",
       "25%      0.000000    0.000000  \n",
       "50%      0.000000    1.000000  \n",
       "75%      0.000000    1.000000  \n",
       "max      1.000000    1.000000  "
      ]
     },
     "execution_count": 138,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "negative[sub_covariates[0:end]].describe()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 139,
   "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>gluc5</th>\n",
       "      <th>hdl5</th>\n",
       "      <th>pad_hx6</th>\n",
       "      <th>sbp6</th>\n",
       "      <th>smoke6</th>\n",
       "      <th>stk_hx6</th>\n",
       "      <th>mi_hx6</th>\n",
       "      <th>trigly5</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>count</th>\n",
       "      <td>129.000000</td>\n",
       "      <td>129.000000</td>\n",
       "      <td>129.000000</td>\n",
       "      <td>129.000000</td>\n",
       "      <td>129.000000</td>\n",
       "      <td>129.000000</td>\n",
       "      <td>129.000000</td>\n",
       "      <td>129.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>mean</th>\n",
       "      <td>96.023256</td>\n",
       "      <td>58.550388</td>\n",
       "      <td>0.054264</td>\n",
       "      <td>128.007752</td>\n",
       "      <td>0.217054</td>\n",
       "      <td>0.046512</td>\n",
       "      <td>0.054264</td>\n",
       "      <td>119.937984</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>std</th>\n",
       "      <td>16.391912</td>\n",
       "      <td>16.147253</td>\n",
       "      <td>0.227420</td>\n",
       "      <td>22.184417</td>\n",
       "      <td>0.413847</td>\n",
       "      <td>0.211411</td>\n",
       "      <td>0.227420</td>\n",
       "      <td>133.107261</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>min</th>\n",
       "      <td>48.000000</td>\n",
       "      <td>22.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>88.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>33.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>25%</th>\n",
       "      <td>88.000000</td>\n",
       "      <td>49.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>111.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>63.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>50%</th>\n",
       "      <td>95.000000</td>\n",
       "      <td>59.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>126.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>87.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>75%</th>\n",
       "      <td>101.000000</td>\n",
       "      <td>69.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>140.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>109.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>max</th>\n",
       "      <td>228.000000</td>\n",
       "      <td>101.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>214.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1149.000000</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "            gluc5        hdl5     pad_hx6        sbp6      smoke6     stk_hx6  \\\n",
       "count  129.000000  129.000000  129.000000  129.000000  129.000000  129.000000   \n",
       "mean    96.023256   58.550388    0.054264  128.007752    0.217054    0.046512   \n",
       "std     16.391912   16.147253    0.227420   22.184417    0.413847    0.211411   \n",
       "min     48.000000   22.000000    0.000000   88.000000    0.000000    0.000000   \n",
       "25%     88.000000   49.000000    0.000000  111.000000    0.000000    0.000000   \n",
       "50%     95.000000   59.000000    0.000000  126.000000    0.000000    0.000000   \n",
       "75%    101.000000   69.000000    0.000000  140.000000    0.000000    0.000000   \n",
       "max    228.000000  101.000000    1.000000  214.000000    1.000000    1.000000   \n",
       "\n",
       "           mi_hx6      trigly5  \n",
       "count  129.000000   129.000000  \n",
       "mean     0.054264   119.937984  \n",
       "std      0.227420   133.107261  \n",
       "min      0.000000    33.000000  \n",
       "25%      0.000000    63.000000  \n",
       "50%      0.000000    87.000000  \n",
       "75%      0.000000   109.000000  \n",
       "max      1.000000  1149.000000  "
      ]
     },
     "execution_count": 139,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "negative[sub_covariates[end:]].describe()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 140,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "-0.616186139423817 -0.2231435513142097\n"
     ]
    }
   ],
   "source": [
    "print(np.log(0.54), np.log(0.8))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3",
   "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": "3.6.4"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 4
}