Dice-Plots.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "import pandas as pd\n",
    "import numpy as np"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [],
   "source": [
    "import seaborn as sns\n",
    "sns.set(style=\"whitegrid\")\n",
    "tips = sns.load_dataset(\"tips\")\n",
    "ax = sns.boxplot(x=tips[\"total_bill\"])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [],
   "source": [
    "df_ucl = pd.read_csv('results_ucl_tig.csv')\n",
    "df_albu2d_pretrain = pd.read_csv('2d-pretrain.csv')\n",
    "df_albu2d_nopretrain = pd.read_csv('2d-no-pretrain.csv')\n",
    "df_albu3d_pretrain = pd.read_csv('3d-pretrain.csv')\n",
    "df_albu3d_nopretrain = pd.read_csv('3d-no-pretrain.csv')\n",
    "ucl_tc_dice = df_ucl['Dice_TC'].iloc[0:66].values\n",
    "albu2d_pretrain = df_albu2d_pretrain['Dice_TC'].iloc[0:66].values\n",
    "albu2d_nopretrain = df_albu2d_nopretrain['Dice_TC'].iloc[0:66].values\n",
    "albu3d_pretrain = df_albu3d_pretrain['Dice_TC'].iloc[0:66].values\n",
    "albu3d_nopretrain = df_albu3d_nopretrain['Dice_TC'].iloc[0:66].values"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [],
   "source": [
    "dices_data = np.vstack([ucl_tc_dice, albu3d_pretrain, albu3d_nopretrain, albu2d_pretrain, albu2d_nopretrain]).T"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [],
   "source": [
    "df_dices = pd.DataFrame(data=dices_data, columns=['UCL-TIG (baseline)', 'AlbuNet3D (pretr.)', 'AlbuNet3D (no pretr.)', 'AlbuNet2D (pretr.)', 'AlbuNet2D (no pretr.)'])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "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>UCL-TIG (baseline)</th>\n",
       "      <th>AlbuNet3D (pretr.)</th>\n",
       "      <th>AlbuNet3D (no pretr.)</th>\n",
       "      <th>AlbuNet2D (pretr.)</th>\n",
       "      <th>AlbuNet2D (no pretr.)</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>0.94066</td>\n",
       "      <td>0.93747</td>\n",
       "      <td>0.92770</td>\n",
       "      <td>0.92105</td>\n",
       "      <td>0.93299</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>0.95684</td>\n",
       "      <td>0.94147</td>\n",
       "      <td>0.95118</td>\n",
       "      <td>0.95390</td>\n",
       "      <td>0.94478</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>0.93997</td>\n",
       "      <td>0.93049</td>\n",
       "      <td>0.93531</td>\n",
       "      <td>0.93727</td>\n",
       "      <td>0.92257</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>0.93035</td>\n",
       "      <td>0.94043</td>\n",
       "      <td>0.93331</td>\n",
       "      <td>0.93619</td>\n",
       "      <td>0.93133</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>0.89697</td>\n",
       "      <td>0.87612</td>\n",
       "      <td>0.52061</td>\n",
       "      <td>0.27447</td>\n",
       "      <td>0.37945</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   UCL-TIG (baseline)  AlbuNet3D (pretr.)  AlbuNet3D (no pretr.)  \\\n",
       "0             0.94066             0.93747                0.92770   \n",
       "1             0.95684             0.94147                0.95118   \n",
       "2             0.93997             0.93049                0.93531   \n",
       "3             0.93035             0.94043                0.93331   \n",
       "4             0.89697             0.87612                0.52061   \n",
       "\n",
       "   AlbuNet2D (pretr.)  AlbuNet2D (no pretr.)  \n",
       "0             0.92105                0.93299  \n",
       "1             0.95390                0.94478  \n",
       "2             0.93727                0.92257  \n",
       "3             0.93619                0.93133  \n",
       "4             0.27447                0.37945  "
      ]
     },
     "execution_count": 6,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_dices.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.8541972727272727"
      ]
     },
     "execution_count": 20,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "np.mean(ucl_tc_dice)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.13911081585017526"
      ]
     },
     "execution_count": 21,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "np.std(ucl_tc_dice)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.7968918181818182"
      ]
     },
     "execution_count": 22,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "np.mean(albu_pretrain)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.2360952614037399"
      ]
     },
     "execution_count": 23,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "np.std(albu_pretrain)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.7775474242424242"
      ]
     },
     "execution_count": 24,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "np.mean(albu_nopretrain)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.2349440409732071"
      ]
     },
     "execution_count": 25,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "np.std(albu_nopretrain)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "metadata": {},
   "outputs": [],
   "source": [
    "np.random.seed(66)\n",
    "x = np.random.normal(loc=0, scale=0.2, size=66)\n",
    "y = np.random.normal(loc=0.07, scale=0.2, size=66)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "-2.1058174564248917 0.037143505619409004\n",
      "Samples are likely drawn from different distributions (reject H0)\n"
     ]
    }
   ],
   "source": [
    "from scipy.stats import ttest_ind\n",
    "\n",
    "# calculate the significance\n",
    "value, pvalue = ttest_ind(x, y, equal_var=True)\n",
    "print(value, pvalue)\n",
    "if pvalue > 0.05:\n",
    "    print('Samples are likely drawn from the same distributions (fail to reject H0)')\n",
    "else:\n",
    "    print('Samples are likely drawn from different distributions (reject H0)')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 49,
   "metadata": {},
   "outputs": [
    {
     "ename": "NameError",
     "evalue": "name 'rcParams' is not defined",
     "output_type": "error",
     "traceback": [
      "\u001b[0;31m---------------------------------------------------------------------------\u001b[0m",
      "\u001b[0;31mNameError\u001b[0m                                 Traceback (most recent call last)",
      "\u001b[0;32m<ipython-input-49-e153aa3c815f>\u001b[0m in \u001b[0;36m<module>\u001b[0;34m()\u001b[0m\n\u001b[0;32m----> 1\u001b[0;31m \u001b[0mrcParams\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mkeys\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m",
      "\u001b[0;31mNameError\u001b[0m: name 'rcParams' is not defined"
     ]
    }
   ],
   "source": [
    "rcParams.keys()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [],
   "source": [
    "import matplotlib.pyplot as plt"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAzsAAAGKCAYAAAArJl/WAAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADh0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uMy4xLjEsIGh0dHA6Ly9tYXRwbG90bGliLm9yZy8QZhcZAAAgAElEQVR4nOzdd1RU19oG8GdogiBNg6gYEBWNgh0VFMGGDWO9NkRjCYkKosYbjXqj0diuGo3GqBFbgkbUWEEjRUEUggV7byiI2CkiIMzM9wd35gMZYICBYY7Pby3Wup7ZZ5/3nJm5Oe+8Z+8tkkqlUhAREREREQmMlroDICIiIiIiqghMdoiIiIiISJCY7BARERERkSAx2SEiIiIiIkFiskNERERERIKko+4AiD52EokEGRkZ0NXVhUgkUnc4RERERFWGVCpFTk4ODA0NoaVV+joNkx0iNcvIyMCdO3fUHQYRERFRlWVnZ4caNWqUej8mO0RqpqurCyDvS6ynp1ehx7p27Rrs7e0r9BhUMfjeaSa+b5qL753m4nunmYp6396/f487d+7I75dKi8kOkZrJHl3T09NDtWrVKvx4lXEMqhh87zQT3zfNxfdOc/G900zFvW9lfdSfExQQEREREZEgMdkhIiIiIiJB4mNsRERERCQogYGBSExMVHcYKpWWlgYAMDY2rpTjWVlZYfjw4ZVyrIrEZIeIiIiIBCUxMRH34h/CwNxM3aGoTOarNwCAdEgq/liv31T4MSoLkx0iIiIiEhwDczM09uip7jBU5m5QKABUyjnJjiUEHLNDRERERESCxGSHiIiIiIgEickOEREREREJEpMdIiIiIiISJCY7REREREQkSEx2iIiIiIhIkJjsEBERERGRIDHZISIiIiIiQWKyQ0REREREgsRkh4iIiIiIBInJDhEREVExYmJiEBMTo+4wiKoETfs+6Kg7ACIiIqKqLDo6GgDg5OSk5kiI1E/Tvg+s7BARERERkSAx2SEiIiIiIkFiskNERERERILEZIeIiIiIiASJyQ4REREREQlSpSU7EREREIlEsLGxKbGtSCSCSCRCfHy8wtfT0tKwadMmdO3aFVZWVtDT04OFhQX69u2Lffv2Kdzniy++gEgkwoIFC8p+Eh/Ep+xf/mOWdG5SqRSRkZGYMGECHBwcYGZmBgMDA1hbW2P48OE4cOAApFJpmeJOS0uDra0tZs+eLd9mY2MDkUiExMTEMvWpTgsWLIBIJMKPP/5Y7LaK9vTpU9SoUQMbN26stGMSERERUck0burpyMhIjBo1CklJSTAzM4ODgwM6dOiA+Ph4/P333zh27BgGDRqEwMBA6OrqVkgMnp6ehbYdPHgQGRkZ6NGjB2rXrl3gtRYtWijVb3JyMoYPH45Tp04BAJo0aQIXFxfo6enh3r172Lt3L/bs2QMXFxccPHgQ5ubmpYp72rRpyMrKwrx580q1HxWvTp06mDt3LmbMmIGePXuiYcOG6g6JiIiIiKBhyc7p06fRvXt3AMDy5cvh4+OD6tWry1+/ePEihg4digMHDmD8+PH4448/KiSOgICAQttsbGyQkZGBuXPnws3NrdR9pqamwtXVFXfu3EGnTp2wfv16tGzZskCb27dv48svv0RUVBTGjBmDoKAgpfuPiorCtm3bsHHjRhgZGZU6Pk3h4+ODESNG4JNPPqnU406bNg2//PILJk+ejOPHj1fqsYmIiIhIMY0Zs5ORkYHRo0dDLBZj8+bN+PbbbwskOgDQunVrHD9+HHp6eggICMClS5fUFG3pfffdd7hz5w5cXFwQHh5eKNEB8io9f//9N2xsbBAcHIzY2Fil+581axZq1aqFcePGqTLsKqdWrVpo2rQpatasWanH1dfXx9SpUxESEoLw8PBKPTaVTWpqKlauXInU1FR1h0JEZcDvMBEpQ2OSnf379+PRo0do165dsTfsjRo1wqRJk9CvXz88f/68EiMsu7S0NGzbtg1aWlrYvHkzqlWrVmTb6tWrY+HChfDz81O6QhMVFYWYmBh4eXlBT09PYZv09HT4+fnB0tIS1atXR4cOHRRWsAAgMTER06dPR7NmzWBkZAR9fX3Y2tpi0qRJePLkSaH2R48ehbu7O+rUqQN9fX00bNgQkydPRkJCgsL+d+3aBRcXFxgbG8PQ0BDt27eHv7+/UmOVihvHExQUhEOHDsHZ2RlGRkYwMzPDwIEDceXKFYV9xcbGYtCgQfjkk0+gr68POzs7zJs3D+np6Qrbf/HFF9DW1saKFStKjJPULygoCPfu3UNwcLC6QyGiMuB3mKqS7NxMxCWFITs3U92h0Ac0JtkJDAwEAIwYMaLEtmvWrEFQUBDc3d0rOiyVOHDgALKystC+fXs0adKkxPZeXl5Ys2YNmjdvrlT/27dvBwAMGDCgyDaDBw/Ghg0b0LJlS7i5ueHy5cvw8vLCt99+W6DdjRs30KpVK6xZswY6Ojro3bs3OnfujOfPn2Pjxo1wdnYukAwcOHAA/fv3x6lTp2Bvbw8PDw8AwIYNG+Do6Ijk5OQC/U+cOBGenp64ePEiHB0d0b17d/nje15eXmWenAEA/P39MXDgQKSmpqJXr14wNTXFoUOH4OLiUmjCiO3bt6NTp044fPgwbG1t4eHhgaysLCxevBidOnXC69evC/VvYWGB9u3bIyQkBElJSWWOkypeamoqYmJiIJVKER0dzV+GiTQMv8NU1cS/uYbUrBeIf3NN3aHQBzRmzM6tW7cAAI6OjmqORPXOnTsHAOjatWuF9B8aGgpdXV04OTkV2ebp06f4559/0KZNGwDA9evX4ebmhhUrVmDQoEHyff/973/j1atXWLNmDfz8/OT7P3/+HM7Ozrh//z4OHz4sn8Rh5syZ0NLSwqVLl9C0aVMAgFgsxujRo7F7925s3LhRPlvdli1bsGXLFrRq1QqHDx9G/fr1AQAvX76Eh4cHdu7ciS5dusDb27tM1+HQoUPYsGEDvv76awDA+/fv0adPH5w4cQIbN27EsmXLAAA3b97EV199BSMjIwQHB6NTp04AgJycHPj6+mLTpk3w8fHBrl27Ch3D1dUVMTExCAsLw5gxY8oUJ1W8oKAgSCQSAIBEIkFwcDBGjRql5qiISFmV/R1OTU1FWloaVq1aVWHHINVJT09HSkoKxHqVc5ubnZuJ5LcPAQDJbx/Cxswe1XQMKuXYFSUnMxMJqekKP/MJCQkwNjZWQ1RlozGVnadPnwJAoZnOhEA27XO9evUUvj59+nSMHj260J/s5rw4Dx48QEJCApo2bVrkI2wA8O2338oTHQBo3ry5PAnZsGGDfLu1tTUGDx4MX1/fAvtbWFhg0KBBAIDHjx/Ltz99+hS6urqwtLSUb9PW1sbixYvx66+/on///vLtsse/duzYIU90gLxxOFu2bAGAcv2HplOnTvJEBwD09PTw5ZdfAshL7mR+/vlnvH//HgsXLpQnOgCgq6uLn3/+GfXq1UNgYKDCR/ZkY60iIiLKHCdVvLNnz0IsFgPIS75LM/6NiNSP32GqSuLfXJM/eSKVSlndqWI0prKjo5MXak5OjpojUb3c3NxiXz9w4AAePXpUaHv37t0LrJmjiCzxKGl9o5EjRxba5uHhAR8fH0RFRcm3/frrr4XaPX36FBcvXpRPCPH+/Xv5a126dMHx48fh6OiIMWPGoG/fvmjTpo18jE/+Pm7fvg1zc3OFU3U3b94c9erVw507d5CcnFwgeVJWhw4dCm2T9ZORkSHfdvLkSQCKK23VqlWDm5sbdu7ciaioqEKPVTZo0AAAihyPRFVD+/btcebMGYjFYmhrayv8bBBR1VXZ32ETExOYmJjgm2++qdDjkGpcuHABEREReJKWUuHHklV1pMirNEohEUR1R9fAAPVqmyr8zGtahbPSkh1tbW0AKHHMheyXGiBvhiuZOnXqIC0tDS9evKiYAPO5efMmFi9eXGj7Z599hrlz56r8eLKKzofjV2Q+HE8SERGh9CNvsutlYmJSZBttbe0ClRQZKysrACg0/uTSpUv49ddfcfbsWdy7d0+eKIhEIgAF3+PffvsNAwYMwKVLl/D999/j+++/h4WFBfr374+vvvpK/liiLDl4/fq1vJ+iJCQklCnZMTU1LbRNlkTLHofIH0tJ6yMpSmhk11lTJsf4WHl4eCAmJgZisRhaWlro16+fukMiolLgd5iqivxVHRlZdafJJ8IbeqGJKi3ZqVGjBoCCv6Arkn9we/6b07Zt2+L27ds4e/ZsiTf6Fy9exLFjx9CjRw+0b9++1LE+e/YMO3fuLLTd1dW1QpKdVq1aAQBiYmJU3resEpY/ifyQjo6O/KY/P9mXN//irMuXL5dXkxwcHDB48GA0a9YMHTp0wMmTJ7Fo0aICfXz66afyX1gOHTqEsLAw3LhxA1u2bMHWrVuxbt06TJkyRR6fhYUFevbsWew5yT5LpVVSEiUji2XUqFHF7tOoUaNC22RJU0nVOlIvExMTODk5ISoqCs7OzsX+GEBEVQ+/w1RVpGW/lFd1ZKSQIC37pZoiog9VWrIje4zq9evXSE9PL/KG9e7duwDybnrzV3YGDBiAXbt2Yf/+/Zg1a1axx9qwYQM2b96Mf/75B4cPHy51rG5ubuWa9au0/vWvf8HPzw8RERF49OgRrK2tVda3ubk5gLxB/kXJzs7G69ev5W1lZI/OySo8Dx8+xJw5c2Bubo6///670GQRRS1yqqWlhW7duqFbt24AgCdPnmDdunXyxMnb2xt16tQBkJfIFDXldWWpU6cOHj16hOXLl8vPXVmyStqH15KqHg8PDzx9+pS/CBNpKH6HqSpwtOqj7hCoBJU2QYGpqSmaN28OqVSK/fv3F9nuyJEjAPKqKPkNGDAADRs2xNmzZ/H7778Xuf/ly5flVZnJkyerIPKKV6tWLXz99dcQi8UYM2ZMgTEvisTFxSndt52dHYDCj6J9KDQ0tNC2v/76C0Be8gfkzRonkUjg7u5eKNGRSCTyxTRl1Y27d++iRYsW6Nu3b4G29erVw7Jly1CrVi28ffsWb968gY2NDerXr4+HDx/i5s2bhWJ5/vw5mjRpgh49euDt27dKnHnZdenSBUDe+kCK9OrVC05OTjh79myh12TXWXbdqeoyMTHBzJkz+YswkYbid5iIlFGps7HJHn/65ptvFM5WFRwcjFWrVkFLS6tQ9aZatWpYv349tLS0MGHCBPz000/Iysoq0CYmJgb9+/fHu3fvMHToUPTu3bvCzkXV/vvf/6J169Y4deoUOnbsiOjo6EJtbt26hZEjR8oHi1lYWJTYr62tLSwtLXHz5s1i1yH45ptv8ODBA/m///nnH/z444/Q0dGBj48PAMjH9Zw5c6bAOjNZWVnw8fHB5cuX5f+WHfv58+c4fvw4Dh48WOB4oaGhePnyJaytreXnMW3aNEgkEnh5eRWY0e3du3cYN24c7ty5gxo1aii9mGpZTZ06FVpaWvjuu+8KTM4glUqxcOFChISE4OHDh/KZ1/KTPYro7OxcoTESERERUckqdTa20aNH4/z58/j555/RtWtX2Nvbw87ODhKJBNevX8fdu3eho6OD3377DW3bti20f69evbB37154enrim2++weLFi9GmTRuYmZnh9u3buHLlCgBg4MCBRVZ/Vq5ciY0bNxYZY3h4uNKLdaqSvr4+Tp06hUmTJiEgIACdOnWCtbU1mjVrBn19fdy9exfXruVNZWhsbIzZs2fj3//+t1J9e3h4wN/fH6dPn1ZY7tfV1UXt2rVhb2+P7t27Izs7GydPnoRYLMa6detgb28PIG/2G2dnZ0RHR6Nx48bo3LkzxGIxoqOj8ebNGzRr1gw3btyQT7Sgra2NjRs3YvDgwRg0aBDatWsHGxsbPH36FNHR0dDW1sYvv/wij2PatGmIjo7GX3/9hc8++wyOjo4wMTFBTEwMXrx4gcaNGxf73qlKu3btsGrVKsyYMQOurq5o06YNrK2tcfXqVdy9excGBgbYt28fqlWrVmjfqKgoaGtrF6pmEREREVHlq/R1dtasWYPw8HB4enoiMzMTx48fR3h4OHR0dDBp0iRcvHgREyZMKHL/wYMH4/r165g+fTrq1auHf/75BwcOHMCzZ8/Qv39/HDx4EAcOHICBgeLp/jIyMvDs2bMi/9Q5tbWRkRH++OMPXLhwAdOmTYO5uTnOnTuH4OBgvHnzBn369MHatWuRkJCA7777TuGkAopMnDgRAIp8fFBLSwuhoaEYMWIEoqOjcfr0aTg5OeHYsWOYMmWKvJ22tjaOHDmCqVOnwtTUFMeOHcOZM2fg4OCAnTt34tSpUxCJRDh27Jh8gP7AgQPx999/o3fv3njw4AEOHjyIe/fuYciQIYiNjYWHh0eBOPbs2YOtW7eidevWiIuLQ3h4OGrXro158+YhNja20tZZmjZtGk6ePIn+/fvj0aNHCA4OhkQiwdixY3Hp0iV07ty50D7379/H5cuX0bdvX/kYJCIiIiJSH5G0Mkfik9q4u7sjOjoaSUlJGrXqrSaZNWsW/vvf/+Ls2bOFxjQVJzs7G9euXYO9vb3CapEqXbhwQWHVlKo+vneaie+b5sr/3snWFeE6O5oh/zo7jT2Kn+FVk9wNyhtfXRnndDcoFPWMi19nR9Xfh6L+/7K890mVXtkh9fjhhx+QkZGBrVu3qjsUQXr37h22bduGfv36lSrRISIiIqKKw2TnI+Hk5ISJEydi8eLFxU5UQGXz3//+F5mZmVi7dq26QyEiIiKi/2Gy8xFZvXo1TExM8OOPP6o7FEFJSkrCihUrsHr1atja2qo7HCIiIiL6n0qdjY3Uy8jICPfu3VN3GIJTt25dZGRkqDsMIiKqIFxOgOj/adr3gckOERERUTGcnJzUHQJRlaFp3wc+xkZERERERILEZIeIiIiIiASJyQ4REREREQkSkx0iIiIiIhIkJjtERERERCRITHaIiIiIiEiQmOwQEREREZEgMdkhIiIiIiJBYrJDRERERESCxGSHiIiIiIgESUfdARARERERqVrm6ze4GxSq7jBUJvPVGwColHPKfP0GMDat8ONUBiY7RERERCQoVlZW6g5B5dL+90CWsbFxxR/M2FQw15DJDhEREREJyvDhw9UdAlURHLNDRERERESCxGSHiIiIiIgEickOEREREREJEpMdIiIiIiISJCY7REREREQkSEx2iIiIiIhIkJjsEBERERGRIDHZISIiIiIiQWKyQ0REREREgsRkh4iIiIiIBElH3QEQERERERUlMDAQiYmJSrdPT09HRERExQVUBaSlpQEAjI2NVdKflZUVhg8frpK+qhomO0RERERUZSUmJuLx44eoV89UqfbVqwNi8ZsKjkq9UlJSAACGhuJy9/XkSUq5+6jKmOwQERERUZVWr54pfH27qjuMKmPdupMAoJJrIutLqDhmh4iIiIiIBInJDhERERERCRKTHSIiIiIiEiQmO0REREREJEhMdoiIiIiISJCY7BARERERkSAx2SEiIiIiIkFiskNERERERILEZIeIiIiIiASJyQ4REREREQkSkx0iIiIiFYiJiUFMTIy6wyCqUJr2OddRdwBEREREQhAdHQ0AcHJyUnMkRBVH0z7nrOwQEREREZEgMdkhIiIiIiJBYrJDRERERESCxGRHAKRSqbpD0Gi8fkRERETCpHHJzrVr1yASiSASibB8+fIi24lEIujo6JS4TZW2b98OkUiEGjVqID4+vti2Ojo6EIlE5TqeWCzG+vXrMWPGDIWvrV27Fi1btoS+vj7Mzc3h7u6O4OBghX3Jrmn+Py0tLRgaGqJBgwYYPnw4QkNDyxxrWloabG1tMXv27DL3oWp3795F79698ejRo3L3tXr1alhYWCA5OVkFkRERERGRKmhcsrNt2zYAgL6+PjZv3lwlf5V/+/Ytvvzyywo/TmBgIHx8fPDmzZtCr40bNw5+fn6Ij49H9+7d0bZtW0RGRsLDwwOLFi0qss+BAwfC09MTnp6eGDFiBLp16wZ9fX3s2bMH7u7umDRpUplinTZtGrKysjBv3rwy7V8R+vXrh+PHj6ukL19fX9SsWRPjx49XSX9EREREVH4alezk5uYiICAAVlZWGDJkCO7fv4+wsDB1h6VQWFgY/P39K/QYEolE4fY9e/bgjz/+QJMmTXDnzh0EBwcjNDQUcXFxMDExwYIFC3D37l2F+65evRoBAQEICAjArl27cOTIEdy8eRMRERGoU6cONm7ciIULF5YqzqioKGzbtg3z58+HkZFRqc+zohR1/cpCR0cHS5cuxbFjx7B3716V9UtEREREZadRyU5wcDCeP3+OHj16YNiwYQCATZs2qTmqwqysrAAAM2fOxJMnTyr9+AEBAQCAZcuWoXbt2vLtzZs3h6enJyQSCUJCQkrVp6urK/bt2ye/qX/69KnS+86aNQu1atXCuHHjSnVMTTNgwADY2dlh7ty5yM3NVXc4RERERB89jUp2tm/fDgAYOnQo+vTpA3Nzcxw6dKjU4ySePXuGsWPHombNmjAyMoKbmxuOHTtWqJ2NjQ1EIhESExMLvTZx4kSIRCJ5TPn16tULAwcORGpqKry9vUsVW2ZmJpYsWQIHBwcYGBjAzMwMffr0QWRkZIF2bm5u8PLyAgDs2LEDIpEICxYsAADs27cPV69eRZ8+fQr1n56eDgBlGrvk7OyMIUOGICsrS+F5KxIVFYWYmBh4eXlBT0+vwGs2NjYwNTVFRkYGfHx8ULt2bdSoUQMdOnTAn3/+WagvNzc3iEQiXL9+HV26dEG1atVgZWVV4FG02NhYDBo0CJ988gn09fVhZ2eHefPmyc8bACIiIiASiXD//n0AQIMGDeTjp+Lj4yESiTB06FDs2rULVlZWMDAwgKOjI3Jycoo9V5FIhPHjx+Pu3bs4ePCgUtenMiUkJGDbtm0KP89UcVJTU7Fy5UqkpqaqO5QqQ8jXpKznpinXRFPiJCKS0Zhk58WLFwgODoaFhQV69eoFXV1djBw5Erm5udi6davS/UilUri4uGDfvn1wdnZGhw4dcPr0afTt2xe//vqryuLdsGEDzMzMcPToUfz+++9K7ZOSkgIXFxfMnTsXL168QI8ePdC6dWuEhYWha9euBapYPXv2hLOzMwDA1tYWnp6eaNGiBQBAT08P9vb2qFatWoH+g4KCsHfvXhgZGWHgwIFlOi8PDw8AQHh4uFLtZUnRgAEDFL4ukUjg4eGBTZs2oUWLFujcuTMuXbqEUaNGYc6cOQr3GTx4MB48eIB+/fpBS0sLrVu3lh+rU6dOOHz4MGxtbeHh4YGsrCwsXrwYnTp1wuvXrwEAtWvXhqenp/yROtk4pfwuXryIMWPGwNraGl26dIG1tTV0dXVLPN/+/fsXOO+qZMuWLcjJycGWLVvUHcpHJSgoCPfu3StycpCPkZCvSVnPTVOuiabESVSUlKxMLDkTgZSsLHWHQpVEY5KdgIAA5OTkYPTo0fKqhGww+ObNm5UefyGRSJCdnY1r167hyJEjCA8PR3h4OPT19TFjxowSZ1FTlqWlJVavXg0AmD59ulLVJ19fX1y4cAFeXl54+PAhjhw5ghMnTuDs2bOoWbMmfH19cf36dQDA3Llz5ZMFuLi4ICAgAIMHDy7UZ2ZmJoYOHYpmzZqhf//+sLCwQHBwcIHH20qjadOmAIDbt28r1T40NBS6urpwcnJS+Hp6ejouXLiAyMhIhIaG4tixY4iNjYWZmRmWLVuGc+fOFdonJycH169fx/79+xEfHw8LCwvcvHkTX331FYyMjHDq1CnExsZi3759uH//Pr766itcvXoVPj4+AIDPPvsMAQEB8msgG6eU34MHDzB16lScOXMGx48fx549e5Q632bNmuGTTz5BREREiZWgypSQkCB/9DApKYnVnUqSmpqKmJgYSKVSREdH89dwCPualPXcNOWaaEqcRMU5dOcm7rx6iUN3bqg7FKokFTcPs4rJfinPP+6jTZs2aNmyJS5fvoyQkBD07t1bqb6WLl2KBg0ayP/t6uoKHx8frFy5Elu3bi31APyijB07FoGBgTh27BgmT56M/fv3F9n2yZMn+PPPP1G3bl1s2rQJBgYG8tdat26NBQsWwMfHB2vXri3VOKXHjx/jr7/+KrDtypUr6NKlS+lPCICpqSkA4OXLlyW2ffDgARISEuDg4FDoEbb8vv/+e3mVCgBatWqF+fPnY9q0adi0aRMcHR0LtB87dixMTEwAAFpaefn6zz//jPfv32PFihXo1KmTvK2uri5+/vlnBAUFITAwECtWrEC9evWUOlc/Pz/5/5YdRxktWrRAeHg4zp07V+C81OnDas6WLVswf/58NUXz8QgKCpL/ECORSBAcHIxRo0apOSr1EvI1UXRuTZo0KdN+VfGaaEqc6pSamoq0tDSsWrVK3aEISkJCAmrUKP8ta0pWJqIex0MKIOrxIwywawZTff3yB6jh0tOzkJ6eoPTnNiEhAcbGxhUclepoRGUnLi4OV65cQdu2bWFvb1/gNVl1R9kEQFtbWz65QX6yx7OioqLKGW1Bv/32G4yNjXHgwAEEBgYW2e7UqVMQi8Xo2LFjgURHplevXgBQaOxOSaysrPDy5Uu8fv0ae/bsQU5ODnx9fYtdo6g479+/BwCl1gh6/PgxgLyxOcUZMWJEoW2yx94UnW/Lli0LbTt58iQAoGvXroVeq1atGtzc3CCRSJR+f01NTWFtba1U2w/JEumEhIQy7V8RPpxQIikpSU2RfFzOnj0LsVgMIG/tq9jYWDVHpH5CviZlPTdNuSaaEidRUQ7duQnZiiVSqZTVnY+ERlR2ZGvrJCcnw83NrcBrb9++BZD3i1NSUhLq1q1bbF/16tVTODhfNoOaqm8CrayssGLFCnz11Vfw9fVF9+7dUatWrULtZDfG+/fvLzaRKO0NtKGhIQwNDQEA//rXv1C/fn04OztjyZIl8PPzg34pf9GQjXsxNzcvse2LFy8AQF6FUUQ2ycCH6tevD0Dx+6Ho2LLrIhu3VBRlr5+ZmZlS7RSRne/z58/L3Ieq1alTp0DCU9L3hFSjffv2OHPmDMRiMbS1tdGhQwd1h6R2Qr4mZT03TbkmmhKnOpmYmFZdj1YAACAASURBVMDExATffPONukMRlFWrVkEsLrymYGnIqjq50rzqZK5UwurO/9SooQ9TUzOlP7eaVrms8slOdnY2du3aBSDvUa+ipnLOzc3Fli1b8J///KfY/hRVTfJTZhA6APmvW8rw9vZGYGAgTpw4AR8fH+zevbvI/hwcHIq9YVemolKcjh07omHDhrh37x4ePHiAZs2alWr/S5cuASg5qQAgH7NS3LXS1tZWuF22WKyi1xU9UiY7xqhRo4q9Ro0aNSo64BKOoSzZYx5VafrpCRMm4Mcffyzwb6p4Hh4eiImJgVgshpaWFvr166fukNROyNdE0bndu3evTPtVRZoSJ5Ei+as6MrLqztgWbdQTFFWKKp/sHD58GK9fv0b37t2LXED0wIEDGDx4MPz9/TF37txib1SLWh9GNjFB/iqDrB9FN60pKSnKngIAwN/fHw4ODggMDFT42FadOnUA5P1yVp7FSKVSKWbNmoXHjx8jICBAYRVLNktbWQbQBwUFAQC6d+9eYltZBaa48T3v3r1DSkqKfCyQzKNHjwD8f4WnJHXq1MGjR4+wfPlyhZWiyiSraClT/aos9evXl1d36tatq/Zr9LEwMTGBk5MToqKi4OzsXGyV82Mh5GtS1nPTlGuiKXESKXLv9St5VUcmVyrBvdev1BQRVZYqP2ZHNjGBogRBpl+/fjA3N8fjx48VrpeTX1pamsLnjGWD+PM/JiebmvjDmdTEYrHCWcKK06BBAyxduhQAMGnSJHnlQsbFxQUAcOLECWQpmA7x6NGjaNq0KSZPnizfpqiCIRKJcPDgQQQGBipcOPThw4e4ffs2DA0NlRo4m19sbCxCQkJgYGBQaKpmRezs7ACU/Gjg0aNHC207dOgQAMDd3V2p2GQTLijqC8gb8+Tk5ISzZ8/Kt5W3SlYU2fnKzr+qmDBhAnR1dVnVqWQeHh5o1KgRfwXPR8jXpKznpinXRFPiJPrQIree2PH50EJ/i9x6qjs0qmBVOtl5+vQpjh8/Dl1dXQwZMqTIdnp6ehg+fDgA5SYq8Pb2lv/6DuRVj2QTCcgmPADyHikDgHXr1smTE4lEgjlz5hT5OF1xfHx80LlzZyQnJxeaKrthw4bo378/Hj58iK+//hrv3r2TvxYfH48pU6bg9u3bBRIU2XibD6f/lC1kOnXq1AJTDD958gQjRoxAbm4upkyZUqrxOmfOnMHQoUMhlUrx/fffw9LSssR9bG1tYWlpiZs3bxY7Rens2bPx4MED+b/Pnz+PhQsXQk9PTz69dkmmTp0KLS0tfPfddwUmIZBKpVi4cCFCQkLw8OHDApMbFHX9ipOTk4Nbt27h1q1bCit+skRYX18fbdpUrbJ4/fr1MW7cOFZ1KpmJiQlmzpzJX8HzEfI1Keu5aco10ZQ4iYhkqvRjbL///jvEYjF69+5d4oBxLy8vbNiwAUePHi12DZE6deogJSUFdnZ2cHNzw6tXrxAVFQVdXV3s2LEDFhYW8ra+vr7Yu3cvdu3ahYsXL6JZs2aIi4tDYmIihg0bpvTaKzIikQhbtmxBy5YtFVZvNm/eDDc3N+zYsQNHjx6Fo6MjcnNzERkZiezsbAwaNEi+VgwANG7cGEBesvb555/Dw8MD3t7e8PPzw8mTJ+XVoM6dOyM3NxexsbF4+/Yt+vbti0WLFimMcfr06fIJDaRSKdLT03H79m3cuXMHQN50zLNnz1b6nD08PODv74/Tp08X+UugVCqFg4MDunXrhvfv3+PEiROQSCTYsGGD0tWRdu3aYdWqVZgxYwZcXV3Rpk0bWFtb4+rVq7h79y4MDAywb9++AgutNm7cGNeuXcPQoUPRsmVLpRanffToET777DMAeZMdfJg4XLp0Cenp6ejfv3+hRV2JiIiIqHJV6cqOMo+wyTg5OaFx48YQi8XFjnkxNjZGVFQUunXrhvDwcFy6dAm9e/fG6dOnMXDgwAJtO3bsiJMnT6Jnz55ISEhASEgIGjdujDNnziic4lgZdnZ2RSYatWvXRmxsLObPnw8LCwucOHEC58+fR6tWreDv7489e/YUGLDfokULLF26FBYWFggJCcGZM2cA5E2ycPjwYaxduxZ2dnaIjIxETEwMmjdvjo0bN+LIkSNFrntz8OBB7Ny5Ezt37sSuXbsQFhYGqVSKMWPG4PTp01izZk2pznfixIkAUOwaQ0FBQRgxYgSio6MRGxsrf29kFSplTZs2DSdPnkT//v3x6NEjBAcHQyKRYOzYsbh06RI6d+5coP2KFSvQuXNnJCYm4sSJEypZUFb2OKTsvImIiIhIfUTSDwePEKmYu7s7oqOjkZSUVGARKhsbGzx69EhhhUQT5eTkyCcCiIuLU3pMUHZ2Nq5duwZ7e/sKrwZduHABbdu2rdBjUMXge6eZ+L5prrK8d7IpeTn1tGrJpp729S3bD81CtG5d3vqCqrgm69adhLZ26aeeVvXnvKjvXHnvk6p0ZYeE4YcffkBGRoZSj4lpsn379uHZs2eYP39+hU1+QERERETKY7JDFc7JyQkTJ07E4sWLSzUZgCZ5//495s2bh88//7zQ45BEREREpB5MdqhSrF69GiYmJgUWthSStWvXIj09HZs3b1Z3KERERET0P1V6NjYSDiMjo0IriatiQoCqYubMmZg5c6a6wyAiIiKifJjsEBEREamAs7OzukMgqnCa9jlnskNERESkAk5OTuoOgajCadrnnGN2iIiIiIhIkJjsEBERERGRIDHZISIiIiIiQWKyQ0REREREgsRkh4iIiIiIBInJDhERERERCRKTHSIiIiIiEiQmO0REREREJEhMdoiIiIiISJCY7BARERERkSDpqDsAIiIiIqLiPHmSgnXrTqo7jCrjyZMUAFDJNXnyJAWffmpW7n6qKiY7RERERFRlWVlZlap9eno6atSoUUHRVA2mptoAAG1t43L39emnZqW+xpqEyQ4RERERVVnDhw8vVfsLFy6gbdu2FRQNaRqO2SEiIiIiIkFiskNERERERILEZIeIiIiIiASJyQ4REREREQkSkx0iIiIiIhIkJjtERERERCRITHaIiIiIiEiQmOwQEREREZEgMdkhIiIiIiJBYrJDRERERESCpKPuAIiIiIjo4xYYGIjExESV9JWeno6IiAiV9FVVpKWlAQCMjY1V1qeVlRWGDx+usv6qKiY7RERERKRWiYmJePgwHmamn6igNy2kvMlQQT9Vx5uU1wAAiVhbRf29UEk/moDJDhERERGpnZnpJ+jZ/V/qDqNKCg3fCwAquz6y/j4GHLNDRERERESCxGSHiIiIiIgEickOEREREREJEpMdIiIiIiISJCY7REREREQkSEx2iIiIiIhIkJjsEBERERGRIDHZISIiIiIiQWKyQ0REREREgsRkh4iIiIiIBInJDhEREVWomJgYxMTEqDsMIlIhTfle66g7ACIiIhK26OhoAICTk5OaIyEiVdGU7zUrO0REREREJEhMdoiIiIiISJCY7BARERERkSAx2VERqVSq7hCoDPi+EREREQlXmZKda9euQSQSQSQSYfny5UW2E4lE0NHRKXGbKm3fvh0ikQg1atRAfHx8sW11dHQgEonKdTyxWIz169djxowZCl9bu3YtWrZsCX19fZibm8Pd3R3BwcEK+5Jd0/x/WlpaMDQ0RIMGDTB8+HCEhoaWOda0tDTY2tpi9uzZZe5DSHbv3o3Ro0eXux+JRII2bdqopC8iIiIiUp0yJTvbtm0DAOjr62Pz5s1V8tfxt2/f4ssvv6zw4wQGBsLHxwdv3rwp9Nq4cePg5+eH+Ph4dO/eHW3btkVkZCQ8PDywaNGiIvscOHAgPD094enpiREjRqBbt27Q19fHnj174O7ujkmTJpUp1mnTpiErKwvz5s0r0/5CEh0djZEjR+LJkyfl7ktLSwtr167Fzp07sWfPHhVER0RERESqUOpkJzc3FwEBAbCyssKQIUNw//59hIWFVURs5RYWFgZ/f/8KPYZEIlG4fc+ePfjjjz/QpEkT3LlzB8HBwQgNDUVcXBxMTEywYMEC3L17V+G+q1evRkBAAAICArBr1y4cOXIEN2/eREREBOrUqYONGzdi4cKFpYozKioK27Ztw/z582FkZFTq8xSaot63surcuTMGDBgAPz8/pKWlqbRvIiIiIiqbUic7wcHBeP78OXr06IFhw4YBADZt2qTywMrLysoKADBz5kyV/HpfWgEBAQCAZcuWoXbt2vLtzZs3h6enJyQSCUJCQkrVp6urK/bt2wcdHR0sXboUT58+VXrfWbNmoVatWhg3blypjknK+/bbb5GcnIzVq1erOxSFEhISsG3bNiQmJqo7FPrIpKamYuXKlUhNTVV3KGVSWfGr+jiaft2JiFSh1MnO9u3bAQBDhw5Fnz59YG5ujkOHDiE5OblU/Tx79gxjx45FzZo1YWRkBDc3Nxw7dqxQOxsbG4hEIoU3aBMnToRIJJLHlF+vXr0wcOBApKamwtvbu1SxZWZmYsmSJXBwcICBgQHMzMzQp08fREZGFmjn5uYGLy8vAMCOHTsgEomwYMECAMC+fftw9epV9OnTp1D/6enpAFCmsUvOzs4YMmQIsrKyFJ63IlFRUYiJiYGXlxf09PQKvGZjY4NatWohMzMTc+bMQYMGDVCtWjU0aNAAc+bMQUZGRqH+cnNz8csvv6Bdu3YwNDSEkZER2rdvj19//RW5ublKxRQREQGRSIRp06bhwoUL6Nq1K4yMjGBpaYkRI0bg1q1bBdrHx8dDJBJh6NCh2LVrF6ysrGBgYABHR0fk5OTI41q/fj3atWsHIyMjGBsbw9XVFfv37y/Q1xdffAEXFxcAQGRkJEQiEb744gsAwIIFCyASibBv3z6MGzcOhoaGqFmzJhYvXlziOTk7O+Ozzz7DunXrkJWVpdR1qExbtmxBTk4OtmzZou5Q6CMTFBSEe/fuFTlesaqrrPhVfRxNv+5EqpAhzsD+N/uQIS58P0Mfh1IlOy9evEBwcDAsLCzQq1cv6OrqYuTIkcjNzcXWrVuV7kcqlcLFxQX79u2Ds7MzOnTogNOnT6Nv37749ddfS30SRdmwYQPMzMxw9OhR/P7770rtk5KSAhcXF8ydOxcvXrxAjx490Lp1a4SFhaFr164Fqlg9e/aEs7MzAMDW1haenp5o0aIFAEBPTw/29vaoVq1agf6DgoKwd+9eGBkZYeDAgWU6Lw8PDwBAeHi4Uu1lSdGAAQMUvi4Wi9G3b1/89NNPsLGxQc+ePZGcnIylS5di5MiRBdpmZWWhZ8+e8PX1xZ07d9CtWzd07doVN2/exJQpU+Dh4YH3798rfS5Xr16Fq6srbty4gd69e6N27doIDAxEx44dce7cuULtL168iDFjxsDa2hpdunSBtbU1dHV1kZOTg/79+8PHxwcPHjyAi4sLnJ2dcfbsWQwZMgRz586V9+Hs7Ax3d3cAgIWFBTw9PeXvo8zcuXOxZ88e9OzZE5aWlmjWrJlS5+Ph4YFXr14hKChI6WtQGRISEuSVwKSkJFZ3qNKkpqYiJiYGUqkU0dHRGldlqKz4VX0cTb/uRKpyLuMsknKScC7jrLpDITUpVWkhICAAOTk5GD16tLwqMX78eKxfvx6bN2/G7NmzoaVVcv4kkUiQnZ2Na9euoUGDBgDyfmHv3bs3ZsyYgb59+8LGxqb0Z/MBS0tLrF69Gl988QWmT58Od3d3WFpaFruPr68vLly4AC8vL2zatAkGBgYA8m6y3d3d4evri86dO6N58+aYO3curK2tER0dDRcXlyIrLZmZmfDy8sKNGzdw8+ZNfPrpp/jjjz8KPN5WGk2bNgUA3L59W6n2oaGh0NXVhZOTk8LXU1JS8PDhQ1y+fBlNmjQBkDfjnqOjI44cOYJbt27JjzlnzhxERETA2dkZhw4dQq1atQAAz58/h4eHB44fP47vv/8ey5YtUyq2EydOwM3NDYcOHYKxsTEAYOHChZg/fz4mTpyIixcvFvhMPXjwANOnT8dPP/0E4P/H3ixcuBB///03evbsid27d8Pc3BxAXkWoR48eWLJkCVxdXeHu7g5vb280a9YMISEh+Oyzz+SPHOZ3//59nDt3Dq1bty5wnJK4urpixYoVCAkJwdChQ5XapzJ8WM3ZsmUL5s+fr6Zo6GMSFBQk//5IJBIEBwdj1KhRao5KeZUVv6qPU9Wue2pqKtLS0rBq1Sq1xVBa6enpiIiIUHcYH42EhATo6OirtM8McQZuZt0AANzMugFHw/Yw1DZU6TE0VWbWO6QnvC7XdzIhIUF+71aVlaqyI7uZzz/uo02bNmjZsiXi4+NLNQZl6dKl8kQHyLtJ9PHxQXZ2dqmqRCUZO3Ys+vTpg9evX2Py5MnFtn3y5An+/PNP1K1bt0CiAwCtW7fGggULkJOTg7Vr15YqhsePH+Ovv/7CzZs35duuXLlSuhPJx9TUFADw8uXLEts+ePAACQkJaNq0aaFH2PKbN2+ePNEBAHt7e7i6ugIArl+/DiAvadu4cSN0dHSwe/dueaID5FVIdu/eDW1tbfzyyy9KP8alp6eHnTt3Fviy/Oc//0Hr1q1x5coVREdHF9rHz89P/r+1tLSQnZ2NdevWQV9fH3/88Yc80QHyHtP7+eefAaBUX2gXFxd5oiM7jjJatmwJAFXuP5Afju9KSkpSUyT0sTl79izEYjGAvCpybGysmiMqncqKX9XH0fTrTqQK5zLOQoq8GYOlkLK685FSurITFxeHK1euoG3btrC3ty/w2vjx4+Hn54dNmzahd+/eJfalra0tn9wgPw8PD6xcuRJRUVHKhqWU3377Dc2bN8eBAwcQGBiI4cOHK2x36tQpiMVidOzYsUCiI9OrVy8AKDR2pyRWVlZ4+fIltLS0EBYWBj8/P/j6+iIjIwOzZs0q9fnIHhNTZo2gx48fA0CJlbIOHToU2iargsnG7Zw/fx6ZmZno3Lkz6tevX6i9ra0tHB0d8c8//+D8+fPo3LlzifG5uLigbt26BbaJRCJ8/vnnuHjxIiIjIwv0Y2pqCmtr6wLt4+LikJqaijZt2iislnXv3h06Ojo4ffo0xGIxtLW1S4xLlrSUlpWVFXR0dKrcY2J16tQpkPB8eM2JKkr79u1x5swZ+XdP0f/XVGWVFb+qj1PVrruJiQlMTEzwzTffqDWO0rhw4QLatm2r7jA+GqtWrULKG9WNq5FVdST4X4UTElZ38jHQrw5Ts0/K9Z3UlEqt0pUd2do6ycnJcHNzK/AnGw8TFBSk1C/G9erVUzg4XzaDmqp/dbayssKKFSsA5D2mVlRFJCEhAQCwf/9+hQt8Nm7cuEA7ZckGuZuZmeFf//qXvP8lS5aUaSD769evAaBABaMoL168AJD3H5riyKpF+cneI9mjELL3pbjESVatU3bCikaNGincLkumPvwsmJmZFWorez/i4uIUvm8GBgbIzc3Fu3fv5NeuJMpc26KYmJggMzNTPhFFVTBhwoRi/01UUTw8POSVUS0tLfTr10/NEZVOZcWv6uNo+nUnKq/8VR0ZVnc+TkpVdrKzs7Fr1y4AeY96FTWVc25uLrZs2YL//Oc/xfanqGqSn66urjJhyUv0yvD29kZgYCBOnDgBHx8f7N69u8j+HBwc5BMNKKJMRaU4HTt2RMOGDXHv3j08ePBA6cHvMpcuXQKAYmOUkc1UVtK1UuaclFk8VnacDydmKEpRVRbZsT58XdHjZLJjNmjQoNBEAx9S9r1T9rE1RWTJobIz01WG+vXry6s7devWlf+wQFTRTExM4OTkhKioKDg7O5f4w0tVU1nxq/o4mn7dicorOeepvKojI4EEyTnKL9tBwqBUsnP48GG8fv0a3bt3L3IB0QMHDmDw4MHw9/fH3Llzi71ZLGp9mPj4eAAocCMm60fRjWNKSooy4cv5+/vDwcEBgYGBGDFiRKHX69SpAyCv/F+exUilUilmzZqFx48fIyAgQGEVS5YMyJKR0pDN9NW9e/cS28oqFMqM7ymJ7NGnhw8fFtnmwYMHAKD05AtFJc6PHj0CAIWPy31I9r7Z2toqnGygMuXm5iIlJQU6OjpVbtDehAkTsHz5clZ1qNJ5eHjg6dOnGltdqKz4VX0cTb/uROUxoqbmTIRCFUupn69lExMoShBk+vXrB3Nzczx+/Fjhejn5paWlKRws+ddffwHIW79GxsjICEDhx6LEYrHCqYmL06BBAyxduhQAMGnSpEKVCtnaKydOnFD4eNnRo0fRtGnTAhMdKKoUiEQiHDx4EIGBgQonbXj48CFu374NQ0PDApMCKCM2NhYhISEwMDCAp6dnie3t7OwAqObRwHbt2qF69eqIjY2VjwXK7/79+4iLi4OJiQlatWqlVJ+RkZF49+5dgW1SqRSHDx8GAPkU0cVxdHSEgYEBzp07J39sL7+rV6+iUaNGGDJkiPw9L291rihPnz6FVCqFra2tUmODKlP9+vUxbtw4VnWo0pmYmGDmzJkaW12orPhVfRxNv+5ERKpQYrLz9OlTHD9+HLq6uhgyZEiR7fT09OQD//OvRVMUb2/vAjemhw8fxm+//QZjY2OMHz9evt3BwQEAsG7dOvmNqkQiwZw5c4qsChTHx8cHnTt3RnJycqHphBs2bIj+/fvj4cOH+PrrrwvchMfHx2PKlCm4fft2gQRFXz9vmsQP1zCQLWQ6derUAoPVnzx5ghEjRiA3NxdTpkyR76+MM2fOYOjQoZBKpfj+++9LnEYbyKt2WFpa4ubNm+VeZ6F69erw9vZGbm4uRo4ciVevXslfe/HiBUaOHAmJRAJvb+9iZ37LLyUlBZMnT5ZX7mTnduXKFbi5uSk1UYChoSEmTpyItLQ0jBkzpkBcr169wrhx43D//n18+umn8iSnqPetJC9fvsStW7eKHLcVExMDACU+TkdEREREFa/Ex9h+//13iMVi9O7dW+Hg8Py8vLywYcMGHD16tNjZqOrUqYOUlBTY2dnBzc0Nr169QlRUFHR1dbFjxw5YWFjI2/r6+mLv3r3YtWsXLl68iGbNmiEuLg6JiYkYNmwY9uzZU4rTzftFf8uWLWjZsqXC6s3mzZvh5uaGHTt24OjRo3B0dERubi4iIyORnZ2NQYMGwcfHR95eNmnB4cOH8fnnn8PDwwPe3t7w8/PDyZMn5dWgzp07Izc3F7GxsXj79i369u2LRYsWKYxx+vTpMDTMmylEKpUiPT0dt2/fxp07dwDkTb08e/Zspc/Zw8MD/v7+OH36dLkfZ1iyZAni4uJw6tQp2NrayqenjoiIQHp6Otzd3Ys8L0Vq166NXbt24dSpU2jXrh2uX7+OGzdu4NNPPy3Vo4TLli1DXFwc/v77bzRs2BDt27eHrq4uoqKikJ6eDicnJ/z444/y9g0aNICOjg4uXboEd3d3uLq6Flh4tChr1qzB4sWLi3ykUzaTYP/+/ZWOnYiIiIgqRomVHWUeYZNxcnJC48aNIRaLi71RNTY2RlRUFLp164bw8HBcunQJvXv3xunTpzFw4MACbTt27IiTJ0+iZ8+eSEhIQEhICBo3bowzZ86ga9euJcakiJ2dXZE35LVr10ZsbCzmz58PCwsLnDhxAufPn0erVq3g7++PPXv2FHg8qUWLFli6dCksLCwQEhKCM2fOAMibZOHw4cNYu3Yt7OzsEBkZiZiYGDRv3hwbN27EkSNHiqx+HDx4EDt37sTOnTuxa9cuhIWFQSqVYsyYMTh9+jTWrFlTqvOdOHEigLxZ5srLwMAAoaGhWL16NRo1aoSwsDCcOnUK9vb22Lx5M44dO6b05ARA3no+oaGhsLS0xJEjR5Ceng5fX1+cPXsWDRs2VLqf6tWrIzw8HD/99BMaNmyIM2fO4PTp02jUqBFWrlyJsLAweQIJ5I1l2rx5M6ytrREZGVnkWLTSEIvFOHToECwtLeHh4VHu/oiIiIiofERSZabYIo3n7u6O6OhoJCUlVYmB8xEREejatWuxk15omiNHjuDzzz/HTz/9hOnTpyu9X3Z2Nq5duwZ7e/tSJYplwXUjNBffO83E9y2PbD0OrrNDRZGts9Oz+7/UHUqVFBq+FwBUdn1Cw/fC1MxQJevsqOp7XdR3rrz3SWWfX5c0yg8//ICMjAxs3bpV3aEI1i+//AJLS0t8/fXX6g6FiIiIiMBk56Ph5OSEiRMnYvHixeWeqIAKCw8PR0hICNavX1/iOlJEREREVDmY7HxEVq9eDRMTkwID9an8JBIJZsyYgbFjx2Lw4MHqDoeIiIiI/kepRUVJGIyMjHDv3j11hwEgby0loQwX09LSwuXLl9UdBhFRlcXp+ImER1O+10x2iIiIqEI5OTmpOwQiUjFN+V7zMTYiIiIiIhIkJjtERERERCRITHaIiIiIiEiQmOwQEREREZEgMdkhIiIiIiJBYrJDRERERESCxGSHiIiIiIgEickOEREREREJEpMdIiIiIiISJCY7REREREQkSDrqDoCIiIiI6E3KC4SG71V3GFXSm5QXAKCy6/Mm5QVMzQxV0ldVx2SHiIiIiNTKyspKZX2lp6ejRo0aKuuvKtDSFgMAjI1Vk6CYmhmq9JpXZUx2iIiIiEithg8frrK+Lly4gLZt26qsP9JsHLNDRERERESCxGSHiIiIiIgEickOEREREREJEpMdIiIiIiISJCY7REREREQkSEx2iIiIiIhIkJjsEBERERGRIDHZISIiIiIiQWKyQ0REREREgsRkh4iIiIiIBElH3QEQEREREQUGBiIxMbHc/aSnpyMiIqL8AalZWloaAMDY2LjCjmFlZYXhw4dXWP9VAZMdIiIiIlK7xMREPL53F5bV9MrVTzUA71NeqyYoNXqT/R4AoJ/6pkL6T/5f/0LHZIeIiIiIqgTLanqYaG2h7jCqBP9HzwGgwq6HrH+h45gdIiIiIiISJCY7REREREQkSEx2iIiIiIhIkJjsEBERERGRIDHZZ+jyKwAAIABJREFUISIiIiIiQWKyQ0REREREgsRkh4iIiIiIBInJDhERERERCRKTHSIiIiIiEiQmO0REREREJEhMdoiIiEgtYmJiEBMTo+4wiKgEmvxd1VF3AERERPRxio6OBgA4OTmpORIiKo4mf1dZ2SEiIiIiIkFiskNERERERILEZIeIiIiIiASJyQ4REREREQkSkx0BkEql6g5BY/HaEREREQmXxiU7165dg0gkgkgkwvLly4tsJxKJoKOjU+I2Vdq+fTtEIhFq1KiB+Pj4Ytvq6OhAJBKV63hisRjr16/HjBkzFL72yy+/wNHREUZGRjAwMEDz5s2xaNEiZGVlFWgbHx8vv6b5/7S1tWFkZAQ7OzuMHz8eZ8+eLXOsCQkJMDc3x8aNG8vch6qdP38eHTt2VElffn5+aNq0KTIyMlTSHxERERGVn8YlO9u2bQMA6OvrY/PmzVXyl/m3b9/iyy+/rPDjBAYGwsfHB2/evCmwXSwWY8CAAfD19cWtW7fQsWNHuLm5ISkpCd9//z3c3Nzw7t07hX16enrK/4YNGwZXV1fk5uZi27ZtcHJywpIlS8oU6xdffIFPP/0U3t7eZdq/Ijg7O5crgcvvhx9+wKtXr/Dvf/9bJf0RERERUflpVLKTm5uLgIAAWFlZYciQIbh//z7CwsLUHZZCYWFh8Pf3r9BjSCQShdv9/f0RHByMFi1a4NatWwgLC8OxY8dw9+5dODk5ITY2FosWLVK4b0BAgPzvzz//RHBwMB48eIB9+/bB0NAQc+fOxe+//16qOAMCAnDixAksX74cWlpV5yNX1PUrC1NTU8ydOxcbN25EbGysyvolIiIiorKrOneeSggODsbz58/Ro0cPDBs2DACwadMmNUdVmJWVFQBg5syZePLkSaUff/v27QCANWvWoF69evLttWrVwoYNGwAAu3fvLlWfQ4YMkSdvs2bNQmZmplL7vX//HvPmzYODgwN69epVqmNqGm9vbxgbG2P27NnqDkWhhIQEbNu2DYmJieXuKzU1FStXrkRqaqoKIvv4CPH6CfGcKooQrpXsHBITEzX+XIhI2DQq2ZHdxA8dOhR9+vSBubk5Dh06hOTk5FL18+zZM4wdOxY1a9aEkZER3NzccOzYsULtbGxsIBKJFN4cTpw4ESKRSB5Tfr169cLAgQORmppa6se2MjMzsWTJEjg4OMDAwABmZmbo06cPIiMjC7Rzc3ODl5cXAGDHjh0QiURYsGABgLykpmnTpmjfvn2h/u3s7AAASUlJpYoLAIYNG4YOHTogOTn5/9q777Cojv1/4O8FlrqionQQewN7gkJEMYoCYou/q8YSe4lCEhPzTWL3GvUaTUxii4ot0SixYERQCSKKYoktsWIFQUVQBKS4wO75/UF2Q1lggV2R5f16Hp4b98zO+cyZu3A+O2dmcODAAbXeExQUhPj4eEyaNKnEMZFIhI4dOyIlJQWjR4+GhYUF6tWrh169euHo0aMlyjdu3BgNGzbElStX0LFjRxgZGaFp06b4+++/lWWOHDmCvn37on79+jAxMUH79u2xcuVK5ObmKsso5lbJZDJlHI0bNwYAREVFQSQSYdasWfjuu+9gaWkJMzMzDBgwoNy2mpqa4v3330dUVBQuXLig1vV5nTZv3oy8vDxs3ry5ynUdOnQId+/eRWhoqAYiq3108frpYpu0RReulaINgYGBNb4tpBvS9WRY1SAF6Xqy6g6F3jA1JtlJSUlBaGgorKys0K9fP4jFYrz//vvIz8/Hli1b1K5HEAR4eHhg7969cHd3R9euXXHq1Cn4+vpi3bp1Got3/fr1qF+/PsLCwtR+7CstLQ0eHh6YM2cOUlJS0KdPH3Tq1AkRERHo1atXkVEsLy8vuLu7AwCaNm2KUaNGoX379gCAkJAQ3Lx5E2ZmZiXO8eeffwL4d/Spovz8/AAAx44dU6u8IhkcNGiQyuOZmZno2bMn9u3bB3d3d3Tu3BknT56Ej4+PylG7V69eoX///sjOzoaPjw8MDAzQpk0bAMDixYuViaGLiwt8fHyQlJSEzz//HN7e3pBKpQCAZs2aYdSoUcoFIkaNGoUhQ4YUOU9ISAhmzZqFjh074q233kLz5s3Vaq8iKVKVBFenhIQEPHnyBEBBoluV0Z309HScOXMGgiAgJiaG3+hWkC5eP11sk7bowrUq3IYnT57U6LaQ7jgseYl7hrk4LHlZ3aHQG0Z7S5Np2I4dO5CXl4fRo0crV1SbMGEC1q5di02bNuHLL79Uaz6IXC6HVCrFtWvX0KRJEwDAiRMn4O3tjU8//RS+vr7Kb/mrwsbGBqtWrcK4ceMwc+ZM9O3bFzY2NmW+JyAgABcvXsSYMWOwYcMGmJiYAAAuX76Mvn37IiAgAN27d4ezszPmzJkDJycnxMTEwMPDQ62ba0EQMG/ePAAFj6VVRuvWrQEAsbGx5ZbNzs7G6dOn0ahRIzg5Oaksc+/ePdjZ2eHKlSto1aoVAOCPP/6An58fZs6cCV9fXzg6OirLZ2VlwcXFBdHR0RCLxZDL5dDT00NERATmz5+PRo0aISwsDM7OzsryI0eOxMGDB7Fo0SIsXboUHh4e8PDwwO7duyGTybBjx44Scd2+fRs//PADPvroIwDqz+/p0aMHRCIRwsPD1Sr/uhQfzdm8eTMWLFhQqboOHTqkvB5yuRyhoaEYOXJklWOsLXTx+ulim7RFF65V4TYoVLYt6enpyMjIwLfffqvJEDXi5cuXiIqKqu4wapWEhASY5Vd8ZCZdT4azZlkQRMBZsyz4ZNZBXbm+FiLULZn5MmQlJKj1+UtISIC5uflriErzaszIjuJmfvz48crXOnfujA4dOiAuLq5CN5fLli1TJjoA0LNnT/j7+0MqlVZolKg8Y8eOhY+PD1JTUzF9+vQyyz569Ai7du2CnZ1dkUQHADp16oSFCxciLy8PP/74Y6XjmT17Nk6ePAlra+tKrxpWr149AMCzZ8/KLXv27FlIpVJ06NChzHI//PCDMtEBCkatpk+fjpycHGzfvr1E+Q8//BBisRgAlAnuypUrAQCrV69WJjoAYGZmhsDAQJiYmGDt2rXK0Z3yGBkZYerUqcp/q7uwgkQiQdOmTXHnzp1KPSqoLYpRHYWqxHb+/HnlI4AymYwLMlSQLl4/XWyTtujCtSrcBoWa2hbSDYclL6FIv+X//JtIoUaM7Fy6dAl///03unTpAhcXlyLHJkyYgI8//hgbNmyAt7d3uXXp6+srFzcozM/PDytXrkR0dLTG4gaAjRs3wtnZGcHBwQgKCsLw4cNVljt58iRkMhm6detWJNFRUEzuLz53R13z58/H//73PxgZGeG3336DpaVlpepRzH1RZ4+ghw8fAkCZI2XGxsYqH3EbNGgQvv/+e5w4cQJz584tcqx48iSTyXDy5EkAQK9evUrUZWlpic6dO+P06dO4fPmyWnvrtG7dGkZGRuWWU6VJkya4d+8eEhISYGdnV6k6NM3W1rZIwlOVuFxdXXH69GnIZDLo6+uja9eumgix1tDF66eLbdIWXbhWhdugUNm21K1bF3Xr1sVnn32myRA14uLFi+jSpUt1h1GrfPvtt8hNiK/QexSjOrJ/bktkHN1Rm8RAHxaOjmp9/t7E0Vd11YiRHcXeOklJSfD09Czyo5gPc+jQIbW+rba3t1e5sahiDoumv413cHDAihUrABQ8plbaiEhCQgIAYP/+/So3+GzRokWRcurKz8/H1KlTsXjxYhgbGyM4OBg9evSodHtSU1MBABYWFuWWTUlJAVDwx6w0Tk5OylGawhSPrqnqj+Lnfv78uXJ1OHNzc5XX7/Tp0wDUv37qtK80ivYmJydXug5NmzhxYpn/rgg/Pz/lSJeenh769+9fpdhqG128frrYJm3RhWtVuA0KNbUtVPMVHtVR4OgOFfbGj+xIpVL8+uuvAAoe9SptKef8/Hxs3rxZOSelNKpGTQpTdeOtSvEh/LJMmTIFQUFBiIyMhL+/v8plnxX1tWvXTrnQgCrqjKgoZGZm4j//+Q+OHDmCevXq4ffff69SogMAV65cAYAyY1TIy8sDUPa10tdX/a2LYrNYVceL/5FV1G9qalpioYHiyps3Vdo5KkLxLHt+fn6l69A0R0dH5eiOnZ1dpReoAAqSOTc3N0RHR8Pd3b3MZJZK0sXrp4tt0hZduFaF22BjY4OkpKQa2xaq+R4YSpWjOgoyUcHrREANSHYOHjyI1NRU9O7du9QNRIODg/Hee+8hMDAQc+bMKfNGtfjcBYW4uDgARVcpU9Sj6qY1LS1N3SYAKNjos127dggKCsKIESNKHLe1tQVQ8HiAJjYjffHiBby8vHDx4kU4OjoiLCysxCOAlaFYXrR3797lllWMjpQ1v6e0kbT4+IJh7MKLE5SmQYMGEIvFyM/Px/bt20tNoF4XxYhWVUaHtGHixIlYvnx5lUZ1FPz8/PDkyRN+k1tJunj9dLFN2qIL10rRhhEjRmD37t01ui1Us331zLq6Q6A33Bv/GJtiYQJVCYJC//79YWFhgYcPH6rcL6ewjIwMlZMo9+3bB6Bg/xoFiUQCACX28ZHJZMolnNXVpEkTLFu2DEDBBHvFyIWCh4cHACAyMhKvXr0q8f6wsDC0bt26yEIHpY3y5ObmwtfXFxcvXkTbtm0RExOjkURnz549uHXrFmxtbZVLUJdFnT190tLSEBMTU+L133//HQDQt2/fcs9jaGiIbt26ITc3V+WS2FKpFF26dIGHh4cyqQUqNkpWEYr2Ktr/pnB0dMT48eOrNKqjULduXcyaNYvf5FaSLl4/XWyTtujCtVK0wcHBoca3hYh02xud7Dx58gRHjx6FWCwuc6lkQ0ND5cR/VXuzFDdlyhTlt+9AwejRxo0bYW5ujgkTJihfb9euHYCCFb4UyYlcLsfs2bNLfZyuLP7+/ujevTuSkpJKLNvZrFkzDBgwAA8ePMC0adOQnZ2tPBYXF4cZM2YgNja2yKplxsbGAFBib4P58+fj7NmzcHR0RFRUlEZubg8ePIjJkycDAL777ju1Ju+7urrCwMAA586dK5HcFTZ9+vQioz+HDx/GTz/9hAYNGmD06NFqxffJJ58AAKZNm4Zr164pX8/Pz0dAQAAuXbqEzMzMIosllHb9ypKdnY1bt26VuvR2cnIy7t+/j8aNGytH64iIiIioerzRj7H9/PPPkMlk8Pb2Rv369cssO2bMGKxfvx5hYWFlbphoa2uLtLQ0tGzZEp6ennj+/Llyz5bt27fDyspKWTYgIAB79uzBr7/+isuXL6Nt27a4dOkSEhMTMWzYMPz2228Vao9IJMLmzZvRoUMHlaM3mzZtgqenJ7Zv346wsDC8/fbbyM/Px4kTJyCVSjFkyBD4+/sryysWLTh48CAGDhwIPz8/DB06VLk8taWlJWbOnFlqPKr2lymcXMjlcmRkZODq1at4+PAh9PT0sHLlyjJH2QqTSCTw9PREREQEbty4UWRJ6MLXJCUlBS1atECvXr2U/WFkZIRt27ahQYMGap3rvffewyeffILvv/8enTt3xltvvQUbGxtcuHABCQkJsLS0LDFXqkWLFrh8+TJ69uwJZ2dn7Ny5s9zzxMTEwMvLC/r6+iofb1Ss5qfYXJSIiIiIqs8bneyo8wibgpubG1q0aIE7d+6UOefF3Nwc4eHhmDlzJv744w8AgLe3NxYtWgRXV9ciZbt164bjx49j0aJFOHPmDBITE+Hm5oagoCBcvHixwskOUPBo0+LFi1Xuc2NtbY1z587hu+++w969exEZGQlTU1N07NgRkydPxtixY4vMR2nfvj2WLVuGH374AeHh4ahfvz4aNmyoXJns0qVLuHTpUqmxqEp2Ct/wi0QimJqawsnJCdOmTcP06dOVo13qmjRpEiIiIrB//36VyY6enh5iYmLw2WefITw8HIaGhhg8eDDmz5+Pjh07Vuhcq1atQs+ePbF27VpcuHABly9fhpOTEwICAvDFF1/A3t6+SPkNGzZgypQpuHHjBh4/fqxcaa4qFI9DamJeDBERERFVjUgo6/kioiqSy+VwdnZGTk4O7t+/X2TxCJFIVOoISU307NkzODg4oF+/fso5R+qQSqW4du0aXFxcKr23j7q4b0TNxb6rmdhvZVPs3cF9dgj4d5+dSU5W5ReuBQLjC7aw0Nb1CIxPhqGjU4X22dHmZ7W0z1xV75Pe6Dk7VPPp6elh/vz5iI+Px8GDB6s7HK3atGkTcnNzsWDBguoOhYiIiIjAZIdegxEjRqBv376YPXu2zoziFPfs2TN888038Pf3R+fOnas7HCIiIiICkx16DUQiEbZu3YqkpCS1VsuriRYuXAgbGxssX768ukMhIiIion+80QsUkO6ws7MrsQCALk0XW7NmTXWHQERU47i7u1d3CESkhpr8WWWyQ0RERNXCzc2tukMgIjXU5M8qH2MjIiIiIiKdxGSHiIiIiIh0EpMdIiIiIiLSSUx2iIiIiIhIJzHZISIiIiIincRkh4iIiIiIdBKTHSIiIiIi0klMdoiIiIiISCcx2SEiIiIiIp3EZIeIiIiIiHSSQXUHQEREREQEAEnSXATGJ1d3GG+EJ9JcANDa9UiS5qKRVmp+szDZISIiIqJq5+DgoJF6Xr58iTp16mikrupUPyMDAGBobq6V+htBc9f8TcZkh4iIiIiq3fDhwzVSz8WLF9GlSxeN1EU1H+fsEBERERGRTmKyQ0REREREOonJDhERERER6SQmO0REREREpJOY7BARERERkU5iskNERERERDqJyQ4REREREekkJjtERERERKSTmOwQEREREZFOYrJDREREREQ6yaC6AyAiIiKimiUoKAiJiYnVHYZKL1++RFRUVHWHUaaMjAwAgLm5uVbP4+DggOHDh2v1HG86JjtEREREVCGJiYm4ez8OhuYW1R2KSi+kGdUdQpmkGc8BAGm52jtHbkaq9iqvQZjsEBEREVGFGZpbwKFr/+oOo0ZKPBcKAFq9fopz1Hacs0NERERERDqJyQ4REREREekkJjtERERERKSTmOwQEREREZFOYrJDREREREQ6ickOERERERHpJCY7RERERESkk5jsEBERERGRTmKyQ0REREREOonJDhERERER6SQmO0RERDrizJkzOHPmTHWHQUQ6rKb9njGo7gCIiIhIM2JiYgAAbm5u1RwJEemqmvZ7hiM7RERERESkk5jsEBERERGRTmKyQ0REREREOonJDhERERER6SQmOxoiCEJ1h0CVwH4jIiIi0l2VSnauXbsGkUgEkUiE5cuXl1pOJBLBwMCg3Nc0adu2bRCJRKhTpw7i4uLKLGtgYACRSFSl88lkMqxduxaffvqpymNr1qzB22+/DYlEAhMTEzg7O2Px4sV49epVkbJxcXHKa1r4R19fHxKJBC1btsSECRNw/vz5SseakJAACwsL/PTTT5WuQ1eU1W8VlZmZCXt7e8ydO1cDkRERERGRplQq2dm6dSsAwNjYGJs2bXojvx3PzMzE5MmTtX6eoKAg+Pv748WLF0Vel8lkGDRoEAICAnDr1i1069YNnp6eePz4MebPnw9PT09kZ2errHPUqFHKn2HDhqFnz57Iz8/H1q1b4ebmhqVLl1Yq1nHjxqFRo0aYMmVKpd6vS0rrt8qQSCRYvnw5li1bVqPWnSciIiLSdRVOdvLz87Fjxw44ODhg6NChuHfvHiIiIrQRW5VFREQgMDBQq+eQy+UqXw8MDERoaCjat2+PW7duISIiAocPH8adO3fg5uaGc+fOYfHixSrfu2PHDuXPrl27EBoaivv372Pv3r0wMzPDnDlz8PPPP1cozh07diAyMhLLly+Hnh6fXiyt3ypr1KhRaN++PaZOnQqZTKbRuomIiIiocip81xsaGork5GT06dMHw4YNAwBs2LBB44FVlYODAwBg1qxZePTo0Ws//7Zt2wAA33//Pezt7ZWvN2zYEOvXrwcA7N69u0J1Dh06VJm8ffHFF8jJyVHrfbm5uZg7dy7atWuHfv36VeicpB6RSITPP/8cV69exY4dO6o7HJUSEhKwdetWJCYmAgDS09OxcuVKpKenV3NkVFHsu3+97muhrfOxT4mItKPCyY7iJv7//b//Bx8fH1hYWOD3339HUlJShep5+vQpxo4diwYNGkAikcDT0xOHDx8uUa5x48YQiUTKG7TCJk2aBJFIpIypsH79+mHw4MFIT0+v8GNbOTk5WLp0Kdq1awcTExPUr18fPj4+OHHiRJFynp6eGDNmDABg+/btEIlEWLhwIYCCpKZ169ZwdXUtUX/Lli0BAI8fP65QXAAwbNgwdO3aFUlJSThw4IBa7wkKCkJ8fDwmTZpU4phIJMJbb72FFy9eYMaMGbC3t4exsTHatGmDb775Bvn5+SXek52dja+//lp5fczNzdGjRw/s2rVL7XYo5lZ9//33CA8Ph6urK0xNTeHo6IipU6eWSFCjoqIgEokwa9YsfPfdd7C0tISZmRkGDBigLKOJfhs3bhxEIhFOnjyJgQMHwsTEBNbW1spHN8sydOhQ1KtXDytXrlT7OrxOmzdvRl5eHjZv3gwAOHToEO7evYvQ0NBqjowqin33r9d9LbR1PvYpUUm5QjZu5oYhT1D92D+ROiqU7KSkpCA0NBRWVlbo168fxGIx3n//feTn52PLli1q1yMIAjw8PLB37164u7uja9euOHXqFHx9fbFu3boKN6I069evR/369REWFqb2Y19paWnw8PDAnDlzkJKSgj59+qBTp06IiIhAr169ioxieXl5wd3dHQDQtGlT5aNMABASEoKbN2/CzMysxDn+/PNPAP+OPlWUn58fAODYsWNqlVckg4MGDVJ5/OXLl3jnnXfw888/w8XFBT169MCdO3fwxRdflJjA/+zZM7i6umLevHlISkqCt7c33NzccP78eYwcORITJkyoUFv++OMP+Pr6Ijk5Gb6+vjA2NsbGjRvRrVs33L9/v0T5kJAQzJo1Cx07dsRbb72F5s2bA9BcvylMnjwZp0+fho+PD8zMzNCxY8dy22JkZAQvLy9cu3YNFy5cqNB10LaEhAQ8efIEQEGSffPmTZw5cwaCICAmJobfJtcg6enp7Lt/vO5roa3zsU+JVHucfwWZwlM8yr9S3aFQDVahZdF27NiBvLw8jB49Wrmi2oQJE7B27Vps2rQJX375pVrzQeRyOaRSKa5du4YmTZoAAE6cOAFvb298+umn8PX1RePGjSvemmJsbGywatUqjBs3DjNnzkTfvn1hY2NT5nsCAgJw8eJFjBkzBhs2bICJiQkA4PLly+jbty8CAgLQvXt3ODs7Y86cOXByckJMTAw8PDxUjjAVJwgC5s2bB6BgJKAyWrduDQCIjY0tt2x2djZOnz6NRo0awcnJSWWZ27dvo3PnzoiNjYWdnR2AgvlOXl5e+Omnn7BkyRLUqVMHADBlyhRcv34dQ4YMwS+//KJM5u7evYt+/fph69atcHV1xbRp09RqS1hYGN5//31s27YNhoaGkMlkmDZtGgIDA/HRRx/h0KFDJWL94Ycf8NFHHwH4d+6Npvvt6dOnuHr1KhwdHSGXy9We59SzZ0/s2bMH4eHheOutt9R6z+ugGM1R2Lhxo/LayeVyhIaGYuTIkdURGlXQoUOH2Hf/eN3XQlvn02S96enpyMjIwLffflvluGq7ly9fIioqqrrDeGMlJCRApmeotfpzhWw8k98FADyT34W90BFikanWzqeLZNIcJCSka/z3QUJCAszNzTVapzZVaGRHcVM4fvx45WudO3dGhw4dEBcXh/DwcLXrWrZsmTLRAQpuEv39/SGVSis0SlSesWPHwsfHB6mpqZg+fXqZZR89eoRdu3bBzs6uyA0zAHTq1AkLFy5EXl4efvzxx0rHM3v2bJw8eRLW1tb4/PPPK1VHvXr1ABSMspTn7NmzkEql6NChQ5nlli9frkx0AKBPnz5o1aoV8vLycPv2bQAFy2MHBwfDwsICP//8c5FRq+bNmyv7bcWKFWq3xcrKCoGBgTA0LPiFqa+vjzVr1sDGxgahoaElHl80MjLC1KlTlf/W09PTSr8NGTIEjo6OynOoS3Gd37Q/kIpRHYXs7GzlQgoymQznzp2rjrCoEs6fP8+++8frvhbaOh/7lKikx/lXAChW+xU4ukOVpvbIzqVLl/D333+jS5cucHFxKXJswoQJ+Pjjj7FhwwZ4e3uXW5e+vr5ycYPC/Pz8sHLlSkRHR6sbllo2btwIZ2dnBAcHIygoCMOHD1dZ7uTJk5DJZOjWrVuRG2YFxeT+4nNA1DV//nz873//g5GREX777TdYWlpWqp7c3FwAUGuPoIcPHwJAuSNlXbt2LfGajY0NYmNjkZWVBQDKfvH19YVEIilRvmfPnrCxscH9+/eRmJio1mN6AwYMgKlp0W9qjIyM0K9fP2zfvh0nTpzAqFGjlMdat24NIyOjIuW10W/lJYelUSTwCQkJlXq/ttja2hZJeExNTSGVSiGTyaCvr6+y/+nN5OrqitOnT7Pv8PqvhbbOp8l669ati7p16+Kzzz7TSGy12cWLF9GlS5fqDuON9e233+Lhswyt1K0Y1RFQMOIpQM7RnUrQNzKBY0Nrjf8+qGkjx2p/Za2YoJ2UlARPT88iP4r5MIcOHVJr0r29vb3KjUUVN8eVmbhfFgcHB+VoQ0BAQKkjIoob1P3796vc4LNFixZFyqkrPz8fU6dOxeLFi2FsbIzg4GD06NGj0u1JTU0FAFhYWJRbNiUlBUDBH8DS6OnpKR9TK0zRR4rHKxT9UlbipLjZV3fBCsWcm+IUoyrF/7+gqs3a6Dd1rq0qiuucnJxcqfdry8SJE4v8e8qUKcoRKz09PfTv3786wqJK8PPzY9/943VfC22dj31KVFTRUR0Fju5Q5ag1siOVSvHrr78CKHjUq7SlnPPz87F582blnJSpGzC5AAAgAElEQVTSqPr2vTCxWKxOWBXaz2TKlCkICgpCZGQk/P39VS77rKivXbt2JSasF6bOiIpCZmYm/vOf/+DIkSOoV68efv/99yolOgBw5UrBh72sGBXy8vIAlH2t1G2POpvHKs5TfPSlNPr6+mWeq/hxVY+UaaPfKrsXkSIxVLWKXXVydHRUju7Y2dmhTZs2cHNzQ3R0NNzd3ctMhunNUrduXfbdP173tdDW+dinREVlCSnKUR0FAXJkCSnVFBHVZGolOwcPHkRqaip69+5d6gaiwcHBeO+99xAYGIg5c+aUebNYfP6AQlxcHICiq5Qp6lF185iWlqZO+EqBgYFo164dgoKCMGLEiBLHbW1tARQ8UqCJzUhfvHgBLy8vXLx4EY6OjggLCyvxCGBlKJYm7d27d7llFSMU6szvKY9iTs+DBw9KLaM4Zm1trVadpSXO8fHxAP4d4SmLpvutKhQjaZUdGdKmiRMnYvny5cpRHj8/Pzx58oTfItdA7Lt/ve5roa3zsU+J/uVsqHr1WKLKUOvra8XCBKoSBIX+/fvDwsICDx8+VLlfTmEZGRkqJ2Du27cPQME+KAqKuSHFH4uSyWTKJZzV1aRJEyxbtgwA8OGHH5YYqfDw8AAAREZG4tWrVyXeHxYWhtatWxdZ6KC00YLc3Fz4+vri4sWLaNu2LWJiYjSS6OzZswe3bt2Cra2tcgnqslRlT5/iPDw8IBKJcPjwYWRmZpY4fvz4caSkpKBNmzawsrJSq84jR46U6IdXr14hPDwc+vr6aiV0muy3qlJcZ8V1f5M4Ojpi/Pjxyi8T6tati1mzZvFb5BqIffev130ttHU+9ikRkXaUm+w8efIER48ehVgsLnOpZENDQ+XE/8J7mpRmypQpym/BgYLRo40bN8Lc3LzIXi3t2rUDAKxevVp5UyyXyzF79uxSRwXK4u/vj+7duyMpKUn5yJFCs2bNMGDAADx48ADTpk1Ddva/m1jFxcVhxowZiI2NRatWrZSvGxsbA0CJfRHmz5+Ps2fPwtHREVFRUZXeU6ewgwcPYvLkyQCA7777Tq1HxVxdXWFgYIBz586p9RhaWZo0aYJBgwYhNTUVH3zwgXLhAgC4f/++ctNSf39/teuMjY3FvHnzlLHl5eXhww8/RHJyMsaMGaPWCIkm+608T548wa1bt0qdk3TmzBkAUO7jQ0RERETVp9zH2H7++WfIZDJ4e3ujfv36ZZYdM2YM1q9fj7CwsBJLBhdma2uLtLQ0tGzZEp6ennj+/Dmio6MhFouxffv2IqMCAQEB2LNnD3799VdcvnwZbdu2xaVLl5CYmIhhw4bht99+q0BzC77R37x5Mzp06KByFGDTpk3w9PTE9u3bERYWhrfffhv5+fk4ceIEpFIphgwZUuRmXjH5/eDBgxg4cCD8/PwwdOhQ5TLHlpaWmDlzZqnx7Nixo8Rro0ePVv63XC5HRkYGrl69iocPH0JPTw8rV64sc5StMIlEAk9PT0RERODGjRtwdnZW632l2bBhA27fvo3g4GA0btwYHh4eyM7ORlRUFKRSKcaOHYsPP/xQ7frs7e2xZMkSHDhwAG3btsWff/6JuLg4dOjQAStXrlS7Hk3025QpU8o9z+eff46dO3di4sSJKh+ZU6xYN2DAALVjJyIiIiLtKHdkR51H2BTc3NzQokULyGSyMudOmJubIzo6Gu+++y6OHTuGK1euwNvbG6dOncLgwYOLlO3WrRuOHz8OLy8vJCQkIDw8HC1atMDp06fRq1evcmNSpWXLlli8eLHKY9bW1jh37hwWLFgAKysrREZG4sKFC+jYsSMCAwPx22+/FZk03759eyxbtgxWVlYIDw/H6dOnceLECeTk5AAoWLJ7586dpf6oUvj47t27ERUVBYlEgmnTpuHKlSsVXkJQMeKyf//+Cr1PFSsrK5w9exYLFy6ElZUVDh8+jD///BPdu3fHnj17sG3btgo9Iubt7Y19+/bBwMAAISEh0NfXx9y5cxEdHY0GDRqoXY8m+q2q0tLSEBkZiU6dOqFz585Vro+IiIiIqkYkVPXZJnrjyeVyODs7IycnB/fv36/0SmOatG3bNowfP77UEZKaaPXq1fjoo4+wf/9+DBkyRO33SaVSXLt2DS4uLmqvYldZ3Dei5mLf1Uyvu98U+19wn52q42eubIp9dhy6clGNykg8V7DYlDavX+K5UDRqaK61fXY0XW9pn7mq3idV/10vaZ2enh7mz5+P+Ph4HDx4sLrD0UlyuRzr169Hx44dS4xOEhEREVH1YLJTS4wYMQJ9+/bF7Nmz37g9YHTBli1bcOfOHWzYsEFrK70RERERUcUw2aklRCIRtm7diqSkJLVWyyP1ZWZmYt68eZgzZw5cXV2rOxwiIiIi+odam4qSbrCzs0Nqamp1hwEAGDduHMaNG1fdYWiERCIpdaNcIqLXicveE5G21bTfM0x2iIiIdISbm1t1h0BEOq6m/Z7hY2xERERERKSTmOwQEREREZFOYrJDREREREQ6ickOERERERHpJCY7RERERESkk5jsEBERERGRTmKyQ0REREREOonJDhERERER6SQmO0REREREpJOY7BARERERkU4yqO4AiIiIiKjmyc1IReK50OoOo0aSZjwHAK1ev9yMVKChudbqrymY7BARERFRhTg4OFR3CKV6+fIl6tSpU91hlCnDsOB/zc21mIw0NH+j++l1YbJDRERERBUyfPjw6g6hVBcvXkSXLl2qOwx6Q3DODhERERER6SQmO0REREREpJOY7BARERERkU7inB2iaiYIAgAgNzf3tZxPKpW+lvOQ5rHvaib2W83Fvqu52Hc1k6p+U9wfKe6XKkokVPadRKQRL1++xO3bt6s7DCIiIqI3VsuWLSu1yh6THaJqJpfLkZWVBbFYDJFIVN3hEBEREb0xBEFAXl4ezMzMoKdX8Rk4THaIiIiIiEgncYECIiIiIiLSSUx2iIiIiIhIJzHZISIiIiIincRkh4iIiIiIdBKTHSIiIiIi0klMdoiIiIiISCcx2SEiIiIiIp3EZIeohouIiMC7776LBg0awNzcHL169cKRI0cqVMft27fx/vvvw9HREaampmjfvj1Wr14NuVyupahJE/12+PBh9OvXDxYWFjA0NISTkxOmTZuGxMRELUVNgGb6rjgfHx+IRCJERUVpJkhSSRN9l5mZiQULFqBNmzYwNjaGhYUFBg4ciAsXLmgpatJEv509exZ+fn7K35fNmjXDzJkz8eLFCy1FTYVt27YNIpEIp06dqtD7Hj9+jKlTp6Jp06YwMTFBy5YtsXjxYkilUvUrEYioxtq6dasAQDAyMhJ8fHyEvn37CmKxWAAgbNiwQa06rly5IpibmwsAhHfeeUcYMmSIUK9ePQGAMGrUKC23oHbSRL8tW7ZMACDo6ekJ3bp1EwYNGiQ0atRIACBYWloKN2/e1HIraidN9F1x69atEwAIAITjx49rNmBS0kTfPX/+XGjfvr0AQLC3txeGDBmi/LeRkZFw/vx5Lbei9tFEvx04cEAwMDAQAAhdu3YVBg0aJNjZ2QkAhObNmwvJyclabkXtFhMTI0gkEgGAEB0drfb7EhISBAcHBwGA0KlTJ2Ho0KGCjY2NAEDw9PQUcnNz1aqHyQ5RDfXo0SPByMhIqFu3rnD16lXl6+fPnxfMzc0FY2NjITExscw65HK58g/1L7/8onw9OTlZ+frevXu11obaSBP9dv36dUFPT0+QSCRCTEyM8vXc3Fxh+vTpAgChW7duWmtDbaWJvivuzp07gpmZGZMdLdNU333wwQcCAGHEiBGCVCpVvr5ixQoBgNC+fXutxF9baaLf8vLyBGtra0FPT0/Yt2+f8vWcnBxhwIABAgDB399fa22o7fbu3SvUqVNH+TuuIsmOon8WL16sfC0zM1Po06ePAEBYuXKlWvUw2SGqob766isBgLBw4cISx5YsWSIAEObPn19mHUePHlV+Q1Lc6dOnBQBCjx49NBYzaabfvvzyy1LL5ebmCpaWlgIAIS4uTmNxk2b6rrD8/HzB3d1dqF+/vuDi4sJkR4s00Xfx8fGCSCQSmjZtKrx69arE8S5dugjNmjUTUlJSNBZ3baeJfrt48aJyZKC48+fPM0nVkoSEBGHMmDECAMHU1FSwtrauULJz69YtQSQSCc2aNRNkMlmRY/Hx8YK+vr7g5OSkVl2cs0NUQymeVx48eHCJY4rXDh8+XOk63N3dYWVlhVOnTuHly5dVDZf+oYl+MzQ0RPv27dGjR48Sx8RiMZo0aQKg4Fln0hxN9F1h33zzDWJiYrBmzRpYW1trJkhSSRN9t3//fgiCgBkzZsDIyKjE8QsXLuDu3bto2LChBiImQDP9pqdXcKubnJyM/Pz8IseePXsGALCwsKhyrFTU3Llz8csvv+Ctt97C2bNn0bp16wq9/+jRoxAEAQMGDFD2oUKjRo3QuXNnxMfH48aNG+XWxWSHqAYSBAE3btyAnp4e2rRpU+J4y5Ytoaenh+vXr0MQhFLruX79OgDAxcVF5fFWrVpBLper9cuEyqepflu0aBH++usv9O7du8SxrKwsZX85ODhoLvhaTlN9p/DXX39h4cKFGDp0KEaOHKmNkOkfmuq7S5cuAQBcXV2RmZmJwMBAfPjhhwgICMCBAwfU6ndSn6b6zdnZGY6Ojnj06BHGjBmDe/fuITs7G8eOHcP06dOhp6eHTz/9VJtNqZVat26N7du349y5c2jXrl2F31/e/Ykiebp69Wq5dRlU+OxEVO1evHgBqVQKS0tLGBoaljhuYGCAhg0bIjk5GS9fvoS5ubnKep48eQIAsLW1VXlc8frTp081FHntpql+K8vy5cuRmZmJt99+G46OjpoIm6DZvsvNzcWYMWNQr149/PTTT9oMm6C5vrt79y6AgtEAFxcXxMfHK4+tWbMGvXv3RnBwMOrUqaOdhtQymuo3sViMvXv34r333sPu3buxe/du5TE7OzscOXIEXl5eWmtHbfXll19W6f2avD/hyA5RDZSVlQUAMDU1LbWMiYkJgIJlUitbjzp1kPo01W+lCQsLw9KlS6Gnp4dvvvmmckGSSprsu3nz5uHq1avYsGEDH3l6DTTVd+np6QCA8ePHo2HDhoiJiUFGRgZOnTqF9u3b49ixY5g6daoGI6/dNPmZa968OUaNGgV9fX24urrCz88Ptra2ePz4Mb755hukpqZqLnDSCE3enzDZIaqB9PX1AQAikajcsmXtlaNuPdxvRzM01W+qhIaGYujQoZDJZFi6dCk8PT0rEyKVQlN9d/r0aaxcuRKjR49WOQ+BNE9TfafY18PQ0BARERFwc3NDnTp18M477+Do0aOoU6cOdu/ejdu3b2sm8FpOU/32/PlzvPPOO1i/fj0iIiJw7tw5hISE4MGDB5g0aRIiIiIwZMgQjcVNmqHJ+xMmO0Q1kJmZGQAgJyen1DKKYxKJpNL1qFMHqU9T/Vbcli1bMHjwYLx69QoLFizAF198UbVAqQRN9F1WVhbGjh0LW1tbrF69WvNBkkqa+twpvmEeOXIk6tWrV+SYjY0NBg4cCEEQuDGshmiq31asWIFbt25h3rx5Rb4EMjIywrp169CqVSucPHkS0dHRmgmcNEKT9ydMdohqIHNzc5iZmeHZs2clVpcBgPz8fDx79gzGxsYl/igXZmdnBwBISkpSeby8Z2apYjTVb4XNnTsXEydOhEwmw6pVq7Bw4UINR02AZvpu/fr1uHfvHho0aAB/f3+MHj1a+aOYjLtkyRKMHj2aN14apKnPnaWlJQCgcePGKo87OTkB+HeFL6oaTfWbIvlUNS9HLBajT58+AIDLly9rJnDSCE3enzDZIaqBRCIRnJ2dIZPJVD4yERsbC7lcXu4KKIpVTlSttiYIAm7dugV9fX20bdtWM4HXcprqN6CgfyZNmoQlS5bAyMgIu3fvxieffKKNsAma6TvFs+V///03du7cWeRH8Qc9IiICO3fuxL1797TTkFpIU587xfHSlnRX9KGVlVUVIyZAc/2WlpYGoGBBA1UUr+fm5lYxYtKksu5PAODmzZsAoNbfSyY7RDWUt7c3AODAgQMljile8/X1rXQdMTExSElJQffu3bm6kAZpot8A4LPPPsPmzZthbm6Oo0ePYtiwYZoNlEqoat8tXLgQQsFm3iV+FMuIHz9+HIIgYNy4cZpvQC2mic+dj48PACA4OLjESENubi6OHz8OAPDw8KhyvFRAE/2mWKI4LCysxDGZTIbIyEgAQIcOHaoUK2mWou9DQkJKzMt5+PAhLl++DCcnJ/W+jFVr61EieuM8ePBAMDY2FszNzYULFy4oX//zzz8Fc3NzwcTERHj69Kny9bt37wo3b94U0tLSlK/JZDLB2dlZACBs3LhR+XpycrLQvn17AYBw8ODB19OgWkIT/Xb48GEBgGBgYCCcPHnytcZfm2mi70rTu3dvAYBw/PhxbYRe62mi7/Lz84UOHToIAIQZM2YI+fn5giAU/B4NCAgQAAheXl6vr1G1gCb6LTQ0VAAgmJubC6dOnVK+npeXJ8ycOVMAIDg7Oyv7k7SjZ8+eAgAhOjq6xLH4+Hjh5s2bQkpKSpHXvb29BQDCnDlzlK9lZmYKXl5eAgDhxx9/VOvcTHaIarC1a9cKAASxWCz069dP6NevnyAWiwWRSCT88ssvRco6OTkJAIStW7cWef3cuXOCRCIRAAhdu3YVhgwZItSvX18AIEyePPk1tqb2qGq/vf322wIAwd7eXhg1alSpPzdu3HjNLdN9mvjMqcJkR/s00Xc3btwQbGxsBACCk5OTMHjwYKFZs2YCAMHR0VF48ODB62tQLaGJfvvyyy8FAIJIJBLc3NyEQYMGCY0aNRIACNbW1sL169dfY4tqp7KSHcWxBQsWFHn93r17ys+bi4uLMHToUMHW1lYAIPj4+Ah5eXlqnZvJDlENFxISInh4eAhmZmaChYWF0KtXLyEiIqJEubJuvK5fvy4MHTpUsLCwEMzMzISOHTsKP/30E7/p0qLK9tvz588FAGr98MZZOzTxmSuOyc7roYm+S0pKEgICAgQnJyfB0NBQaNSokTBjxgzhyZMnr6EFtZMm+i0kJETw8vIS6tWrJ4jFYqFx48aCv7+/8Pjx49fQAqpMsiMIgvDw4UNh3LhxgrW1tWBsbCy0bdtWWLZsmZCTk6P2uUWCIAjlP+xGRERERERUs3CBAiIiIiIi0klMdoiIiIiISCcx2SEiIiIiIp3EZIeIiIiIiHQSkx0iIiIiItJJTHaIiIiIiEgnMdkhIiIiIiKdZFDdARAREdUUiYmJ6N27t8pjYrEYpqamcHBwgIeHB0aMGAFbW1uVZVu1agUACAkJQcuWLbUWb2Xk5eVh//79CA8PR2xsLNLS0iCRSIq0y9raurrDJCJSCzcVJSIiUlPhZMfFxQWGhobKY3l5eUhNTcXjx48hCALMzMywdOlSeHt7l6jnTU12kpOTMXHiRNy+fRt6enpo1KgRzM3N8ezZMzx9+hQymQympqZYsmQJfH19qztcIqJyMdkhIiJSU+Fk59ixY3BwcChRJikpCd988w1CQ0MhFouxadMmuLm5FSlz7949AICjo2ORhKm6DRs2DH/99RfeffddLFq0CFZWVspjycnJWLFiBQ4ePAixWIzdu3fDxcWlGqMlIiof5+wQERFpkI2NDb799lv069cPeXl5mDNnDvLy8oqUadasGZo1a/ZGJTrnz5/HX3/9BSsrK3z//fdFEh0AsLKywvLly+Hq6oq8vDxs2rSpmiIlIlIfkx0iIiINE4lEWLBgAYyMjPDo0SMcOnSoukMq140bNwAALVu2hJGRkcoyenp6GDp0KADg6tWrry02IqLK4gIFREREWtCgQQP06tULR44cwfHjxzFkyBDlsbLm7ERFRWHXrl24fv060tLSYGVlBQ8PD8yYMaPEaAsA/Pnnn9i2bRsuX76MjIwMNGjQAO+88w6mTp0KJycnteM1MCi4Jbh27RqeP3+OBg0aqCzXt29ftG7dutTjFY0/PDwcQUFBuHr1KrKzs2FpaQk3NzdMnjwZTZo0KVJ29erVWLNmDf7v//4PJiYmWL9+PdLT09GoUSOsXbtW2d6EhARs2rQJp06dQnJyMszMzNCxY0eMGzeuxCOFRKTbOLJDRESkJZ06dQIAXLhwQa3yixYtwtSpUxEVFQV9fX20aNECqamp2L17N4YOHYqkpKQi5detW4fRo0cjIiICcrkcLVu2RHZ2Nvbt24dBgwbhxIkTasfarVs3AEBaWhqGDx+OXbt2ITU1tUQ5U1NTtG7dGpaWllWKXy6XY9asWQgICMCpU6dgamqKVq1aIT09XRl/eHi4yljDw8OxaNEiiMVi2NvbIzs7G46OjgCA6OhoDBw4EEFBQUhNTUWLFi1gbGyMqKgojBs3DmvWrFH7mhBRzcdkh4iISEvs7e0BAM+fPy8xb6e4/fv349dff4WpqSlWr16NEydOIDg4GJGRkejatSuSk5Px1VdfKcuHh4fjhx9+gEQiwapVq3D27Fns378fMTEx+Pjjj5GTk4NPP/0Ujx8/VivW5s2bY+zYsQAKRkYWLlwId3d3DBw4EF9//TWOHTuGrKwsjcW/bt06hISEoE6dOvjpp58QFRWFffv2ISYmBuPGjYNUKsWsWbNw+/btEue6cuUKxo4di2PHjuHw4cPYu3cv9PT0kJiYiE8++QTZ2dmYPn06zp8/j+DgYJw4cQLr1q2DRCLB6tWrERERodY1IaKaj8kOERGRlpiZmSn/Oz09vcyyGzZsAAB8/vnn6Nu3r/J1CwsLrFy5EgYGBjh79iyePn0KAPjxxx8BALNnzy6yDLRYLMb06dPh4+ODzMxMbNu2Te14v/rqK3z55ZfKuAVBQGxsLH755RdMnz4d7u7umD17NlJSUqoUf3Z2NrZs2QIA+O9//4tevXopyxsbG+Orr75C7969IZVKsW7duhLnEovF+PjjjyESiZTnAIAtW7YgMzMTgwcPxscff1xkAYjevXvjs88+AwCO7hDVIkx2iIiItKTwaI7ixlyVuLg4xMXFwcDAAIMHDy5x3MrKCsHBwThz5gysra3x8OFD3LlzB3p6eqXud+Pn5wcAOHnypNrxikQijB8/HidPnsTy5cvh4+OD+vXrK4+/evUK+/btg6+vb5EFCioa/4ULF5CVlQULCwuV+xABwJgxY5Txy2SyIsdatmxZJJFUiIyMBAD0799fZZ39+/eHSCTCzZs3VSZsRKR7uEABERGRlmRmZir/u06dOqWWe/jwIYCCx95MTU1Vlim8kMHdu3cBFKyONmHCBJXlX716BQCIj4+HIAhlJlvFSSQSDB48GIMHD4YgCLh9+zZOnTqF33//HbGxscjIyIC/vz/++OMPGBoaVjj+uLg4AAULNejpqf7e1dnZGQCQlZWFZ8+ewdraWnlM1XyhzMxMPHnyBACwatUqrF+/XmW9+vr6yM/Px4MHD1TWQ0S6hckOERGRlty/fx8A4ODgUOaeOmlpaQBQaqJQnCKJys/Px6VLl8osK5fLkZWVBYlEolbdxYlEIrRq1QqtWrXChAkTsHnzZqxYsQJJSUk4efIk+vTpU+H4FXN/VI3OKBSuKzMzs0iyo2pp7MLziRTLaJfl5cuXasVKRDUbkx0iIiItuXz5MgCgY8eOZZYzMTEBAOTk5KhVryIRaNGihUb28MnMzMQHH3yAZ8+eYc+ePUUSi8JEIhEmTZqEkJAQ3Lp1SzlCU9n4C498FVc4GSkrKVJQxAAAZ86cUc7jIaLajXN2iIiItODx48c4c+YMAMDHx6fMso0bNwYAPHr0SPn4WXHz58/HlClTcP78eeV+MomJicjNzVVZ/tmzZ7hw4YJyQYOySCQSxMXF4enTp4iJiSm3fMOGDQEA9erVq1T8iv1zYmNjIZfLVZa/fv06gIIkRtX+PMWZm5srExzFiFpxMpkMMTExiI+PLzEPiIh0E5MdIiIiLfj6668hl8vRvHnzIquNqdK8eXPY2toiLy9P5UjNixcvEBISghMnTqBOnTpo3rw57O3tkZOTg99//11lnd9++y1GjRqFmTNnqhWvIiFbv3698rE0VZKTk3Hp0iXo6+sr9+apaPxdunSBRCLBixcvcOTIEZXn2blzJ4CC/X9Km9dTXM+ePQEAu3fvVnk8JCQE48ePx+DBg5Gdna1WnURUszHZISIi0qC4uDh89NFHOHbsGMRiMRYvXgx9ff0y3yMSiTBlyhQAwP/+9z+cOnVKeSw1NRWzZs1CdnY2unbtijZt2kAkEmH69OkAgKVLlyI0NFRZPj8/H4GBgdi/fz8AlLqAQXEffvghLCwsEB8fj+HDhyMyMrLIanJyuRxnzpzB2LFjkZ2djREjRsDBwaFS8ZuZmWH8+PEACkZ8oqKilOWlUimWLVuGyMhIiMVifPTRR2rFDwCTJk2CkZERQkJCsGrVKkilUuWx6Oho/Pe//wUA/Oc//ylzwQgi0h0iQRCE6g6CiIioJkhMTETv3r0BAC4uLkUWHZBKpUhOTlYuaSyRSLBixQq8++67Jepp1aoVgIKRhsKrlM2dOxd79uwBULCogUQiwYMHDyCVSmFvb48dO3bAzs5OWX7ZsmXKfXQsLS1hbW2NxMRE5cjMjBkzKpQsXL9+HTNnzkR8fDyAgrkyDg4OMDAwwKNHj5T1Dh48GEuWLIGBQdGpvxWJXyaT4fPPP1cmanZ2dmjQoAHu37+PrKwsmJiY4Ouvv1YuoQ0Aq1evxpo1a9CvXz/lPkPFHT58GP/3f/+H3NxcmJmZoUmTJnjx4gUePXoEAHB3d8eGDRvKXDCCiHQHkx0iIiI1FU52ijMwMECdOnXQtGlTeHh4YPjw4aVOki8t2QGAo0ePYteuXbh+/TpycnJga2sLLy8vTJ06FXXr1i1R16lTp7Bz50789ZJJM94AAAD2SURBVNdfSE9Ph0QigYuLC0aOHFlqrGXJzc3FgQMHEBUVhdjYWKSmpkIul8PS0hJdunTBe++9h65du5b6/orELwgCDh8+jD179ijL29jYoHv37vjggw+Uc3sU1El2AODevXvYsmULzpw5g+TkZIjFYjRr1gwDBgzAyJEjIRaLK3xdiKhmYrJDREREREQ6iXN2iIiIiIhIJzHZISIiIiIincRkh4iIiIiIdBKTHSIiIiIi0klMdoiIiIiISCcx2SEiIiIiIp3EZIeIiIiIiHQSkx0iIiIiItJJTHaIiIiIiEgnMdkhIiIiIiKdxGSHiIiIiIh00v8HEZkLk46nAxoAAAAASUVORK5CYII=\n",
      "text/plain": [
       "<Figure size 864x432 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# sns.set_context(\"talk\")\n",
    "sns.set(rc={'figure.figsize':(12,6)})\n",
    "sns.set_style(\"whitegrid\", {\"text.color\": \"0\", \"xtick.color\": \"0\", \"ytick.color\": \"0\"})\n",
    "sns.set_context(\"notebook\", font_scale=2)\n",
    "# sns.despine()\n",
    "#my_pal = {\"UCL-TIG\": \"b\", \"AlbuNet2D (pretr.)\": \"r\", \"AlbuNet2D (no pretr.)\":\"y\"}\n",
    "# palette=my_pal,\n",
    "ax = sns.boxplot(data=df_dices, palette=sns.color_palette(\"Set3\"), orient=\"h\", width=0.8, showmeans=True).set_xlabel('Dice Score')\n",
    "# .set_title('BraTS \\'18 Tumor Core Dice Scores')\n",
    "\n",
    "plt.tight_layout()\n",
    "plt.savefig(\"dice_boxplot.pdf\")\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 864x720 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# sns.set_context(\"talk\")\n",
    "sns.set(rc={'figure.figsize':(12,10)})\n",
    "sns.set_style(\"whitegrid\", {\"text.color\": \"0\", \"xtick.color\": \"0\", \"ytick.color\": \"0\"})\n",
    "sns.set_context(\"notebook\", font_scale=2)\n",
    "# sns.despine()\n",
    "my_pal = {\"UCL-TIG\": \"b\", \"AlbuNet (pretr.)\": \"r\", \"AlbuNet (no pretr.)\":\"y\"}\n",
    "ax = sns.boxplot(data=df_dices, orient=\"v\", palette=my_pal, width=0.8, showmeans=True).set_title('Dice Scores on the BRATS \\'18 validation data')"
   ]
  },
  {
   "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": "3.7.3"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}