31_EMA_FeatureScoring.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "#Workbench imports\n",
    "import pandas as pd\n",
    "import matplotlib.pyplot as plt\n",
    "\n",
    "from pathlib import Path"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "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>Housing_market</th>\n",
       "      <th>SLR</th>\n",
       "      <th>implementation_time</th>\n",
       "      <th>transient</th>\n",
       "      <th>Mayor</th>\n",
       "      <th>scenario</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>395995</th>\n",
       "      <td>rational</td>\n",
       "      <td>1</td>\n",
       "      <td>4</td>\n",
       "      <td>SurgeHeight\\five_hundred\\five_hundred_430.csv</td>\n",
       "      <td>Lawkeeper</td>\n",
       "      <td>98995</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>395996</th>\n",
       "      <td>rational</td>\n",
       "      <td>3</td>\n",
       "      <td>13</td>\n",
       "      <td>SurgeHeight\\five_hundred\\five_hundred_453.csv</td>\n",
       "      <td>Lawkeeper</td>\n",
       "      <td>98996</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>395997</th>\n",
       "      <td>boundedly_rational</td>\n",
       "      <td>3</td>\n",
       "      <td>10</td>\n",
       "      <td>SurgeHeight\\five_hundred\\five_hundred_157.csv</td>\n",
       "      <td>Lawkeeper</td>\n",
       "      <td>98997</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>395998</th>\n",
       "      <td>boundedly_rational</td>\n",
       "      <td>5</td>\n",
       "      <td>11</td>\n",
       "      <td>SurgeHeight\\five_hundred\\five_hundred_32.csv</td>\n",
       "      <td>Lawkeeper</td>\n",
       "      <td>98998</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>395999</th>\n",
       "      <td>rational</td>\n",
       "      <td>6</td>\n",
       "      <td>4</td>\n",
       "      <td>SurgeHeight\\five_hundred\\five_hundred_308.csv</td>\n",
       "      <td>Lawkeeper</td>\n",
       "      <td>98999</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "            Housing_market SLR implementation_time  \\\n",
       "395995            rational   1                   4   \n",
       "395996            rational   3                  13   \n",
       "395997  boundedly_rational   3                  10   \n",
       "395998  boundedly_rational   5                  11   \n",
       "395999            rational   6                   4   \n",
       "\n",
       "                                            transient      Mayor scenario  \n",
       "395995  SurgeHeight\\five_hundred\\five_hundred_430.csv  Lawkeeper    98995  \n",
       "395996  SurgeHeight\\five_hundred\\five_hundred_453.csv  Lawkeeper    98996  \n",
       "395997  SurgeHeight\\five_hundred\\five_hundred_157.csv  Lawkeeper    98997  \n",
       "395998   SurgeHeight\\five_hundred\\five_hundred_32.csv  Lawkeeper    98998  \n",
       "395999  SurgeHeight\\five_hundred\\five_hundred_308.csv  Lawkeeper    98999  "
      ]
     },
     "execution_count": 3,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "from ema_workbench import load_results\n",
    "\n",
    "target = Path(\"temp\",\"experiments\")\n",
    "experiments, outcomes = load_results(target / \"2021_2_23_EMA_experiments_five_hundred.tar.gz\")\n",
    "experiments2 = experiments.drop(['policy','model'],axis=1)\n",
    "experiments2.tail()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Some preprocessing to make plots more appealing"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "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>Housing market (R1)</th>\n",
       "      <th>SLR scenario (X1)</th>\n",
       "      <th>Implementation time (X3)</th>\n",
       "      <th>Transient storm surge (X2)</th>\n",
       "      <th>Management strategy (L1)</th>\n",
       "      <th>scenario</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>0</td>\n",
       "      <td>4</td>\n",
       "      <td>12</td>\n",
       "      <td>119</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>0</td>\n",
       "      <td>4</td>\n",
       "      <td>14</td>\n",
       "      <td>90</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>0</td>\n",
       "      <td>2</td>\n",
       "      <td>10</td>\n",
       "      <td>220</td>\n",
       "      <td>0</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>0</td>\n",
       "      <td>9</td>\n",
       "      <td>9</td>\n",
       "      <td>141</td>\n",
       "      <td>0</td>\n",
       "      <td>3</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>1</td>\n",
       "      <td>5</td>\n",
       "      <td>13</td>\n",
       "      <td>7</td>\n",
       "      <td>0</td>\n",
       "      <td>4</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "  Housing market (R1) SLR scenario (X1)  Implementation time (X3)  \\\n",
       "0                   0                 4                        12   \n",
       "1                   0                 4                        14   \n",
       "2                   0                 2                        10   \n",
       "3                   0                 9                         9   \n",
       "4                   1                 5                        13   \n",
       "\n",
       "   Transient storm surge (X2) Management strategy (L1) scenario  \n",
       "0                         119                        0        0  \n",
       "1                          90                        0        1  \n",
       "2                         220                        0        2  \n",
       "3                         141                        0        3  \n",
       "4                           7                        0        4  "
      ]
     },
     "execution_count": 4,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "experiments3 = experiments2.copy()\n",
    "\n",
    "map_house_market = {'rational' : '1', \n",
    "             'boundedly_rational' :'0'}\n",
    "\n",
    "map_house_market_invert = {'1':'R', \n",
    "                           '0':'BR'}\n",
    "\n",
    "map_mayor = {'H. Economicus': '2',\n",
    "             'Sentiment' : '1',\n",
    "             'R. Active' : '0',\n",
    "             'Lawkeeper' : '3'}\n",
    "\n",
    "map_mayor_invert = { '2' : 'Economic',\n",
    "                    '1' : 'Sentiment',\n",
    "                    '0' : 'Reactive',\n",
    "                    '3' : 'Proactive'\n",
    "                    }\n",
    "\n",
    "\n",
    "rename_cols = {'Housing_market' : 'Housing market (R1)',\n",
    "               'SLR' : 'SLR scenario (X1)',\n",
    "               'implementation_time' : 'Implementation time (X3)',\n",
    "               'transient' : 'Transient storm surge (X2)',\n",
    "               'Mayor' : 'Management strategy (L1)'}\n",
    "\n",
    "experiments3['Housing_market'].replace(map_house_market,inplace=True)\n",
    "experiments3['Mayor'].replace(map_mayor,inplace=True)\n",
    "experiments3['transient'] = experiments3['transient'].apply(lambda s : int(s.split('_')[-1].split('.')[0]))\n",
    "cols_convert = ['transient', 'implementation_time']\n",
    "experiments3[cols_convert] = experiments3[cols_convert].astype('int')\n",
    "experiments3 = experiments3.rename(columns=rename_cols)\n",
    "experiments3.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "<class 'pandas.core.frame.DataFrame'>\n",
      "RangeIndex: 396000 entries, 0 to 395999\n",
      "Data columns (total 6 columns):\n",
      " #   Column                      Non-Null Count   Dtype   \n",
      "---  ------                      --------------   -----   \n",
      " 0   Housing market (R1)         396000 non-null  object  \n",
      " 1   SLR scenario (X1)           396000 non-null  category\n",
      " 2   Implementation time (X3)    396000 non-null  int32   \n",
      " 3   Transient storm surge (X2)  396000 non-null  int32   \n",
      " 4   Management strategy (L1)    396000 non-null  object  \n",
      " 5   scenario                    396000 non-null  category\n",
      "dtypes: category(2), int32(2), object(2)\n",
      "memory usage: 14.2+ MB\n"
     ]
    }
   ],
   "source": [
    "experiments3.info()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "metadata": {},
   "outputs": [],
   "source": [
    "remap = {\"Area_A_HousePrice_2200\":\"HP_hp_2200\",\n",
    "         \"Area_A_timing_SETP\":\"HP_first_SETP\",\n",
    "         \"Area_B_HousePrice_2200\":\"CC_hp_2200\",\n",
    "         \"Area_B_timing_SETP\":\"CC_first_SETP\"}\n",
    "\n",
    "\n",
    "outcomes2 = {}\n",
    "for newlabel, oldlabel in remap.items():\n",
    "    outcomes2[newlabel] = outcomes[oldlabel]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "{'Area_A_HousePrice_2200': array([0, 0, 0, ..., 0, 0, 0], dtype=int64),\n",
       " 'Area_A_timing_SETP': array([2081, 2080, 2040, ..., 2051, 2055, 9999], dtype=int64),\n",
       " 'Area_B_HousePrice_2200': array([282357.9820158 , 286542.15584763,      0.        , ...,\n",
       "        350017.41980036, 350036.70600789, 350040.12775439]),\n",
       " 'Area_B_timing_SETP': array([2096, 2125, 2106, ..., 9999, 9999, 9999], dtype=int64)}"
      ]
     },
     "execution_count": 27,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "outcomes2"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Feature scoring\n",
    "### Regression \n",
    "The timing_SETP is set to 9999 if there is no tipping point in any of the years (<2300)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "from ema_workbench.analysis import feature_scoring\n",
    "import seaborn as sns\n",
    "\n",
    "x = experiments3\n",
    "y = outcomes2\n",
    "\n",
    "preferred_order = ['SLR scenario (X1)',\n",
    "                   'Transient storm surge (X2)',\n",
    "                   'Implementation time (X3)',\n",
    "                   'Management strategy (L1)',\n",
    "                   'Housing market (R1)']                   \n",
    "\n",
    "fs = feature_scoring.get_feature_scores_all(x, y)\n",
    "fs = fs.reindex(preferred_order)\n",
    "sns.heatmap(fs, cmap='viridis', annot=True)\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "The timing_SETP is set to 2500 if there is no tipping point in any of the years (<2300)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 42,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "{'Area_A_HousePrice_2200': array([0, 0, 0, ..., 0, 0, 0], dtype=int64),\n",
       " 'Area_A_timing_SETP': array([2081, 2080, 2040, ..., 2051, 2055, 2500], dtype=int64),\n",
       " 'Area_B_HousePrice_2200': array([282357.9820158 , 286542.15584763,      0.        , ...,\n",
       "        350017.41980036, 350036.70600789, 350040.12775439]),\n",
       " 'Area_B_timing_SETP': array([2096, 2125, 2106, ..., 2500, 2500, 2500], dtype=int64)}"
      ]
     },
     "execution_count": 42,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "import numpy as np\n",
    "outcomes3=outcomes2.copy()\n",
    "for key, values in outcomes2.items():\n",
    "    if key in [\"Area_A_timing_SETP\",\"Area_B_timing_SETP\"]:\n",
    "        outcomes3[key] = np.array([2500 if y == 9999 else y for y in values])\n",
    "outcomes3"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 45,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAdkAAAFsCAYAAACAQMPEAAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADh0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uMy4yLjAsIGh0dHA6Ly9tYXRwbG90bGliLm9yZy8GearUAAAgAElEQVR4nOzde5xNVf/A8c/3zAxyCWOGYRC5y/0yCpVEoaK7UgohiS5SRPX0dFUkKl2QeLqR6umRSHJJRbmX3OXWGMOMcacxM+f7++Ns48z9zJgzM6ff9/16ndecvfdaa6+1zpnz3Wvtfc4WVcUYY4wx+c9V2BUwxhhj/qksyBpjjDF+YkHWGGOM8RMLssYYY4yfWJA1xhhj/CS4sCtg/jmOxVS3S9Udt9ZoX9hVKDJ2ftSwsKtQZBQvnlzYVSgyNvZ4Vs63DHdsXZ8/c1wR2857f3lhI1ljjDHGT2wka4wxJiAlqe8zA8X9WI/sWJA1xhgTkNwU/TNUFmSNMcYEJDfuwq5CjizIGmOMCUgpAfCzwBZkjTHGBCSbLjbGGGP8JMWCrDHGGOMfSWrnZI0xxhi/KPoh1oKsMcaYAGXTxcYYY4yfpBT9GGtB1hhjTGCy6WJjjDHGT5K0UH7zP1csyBpjjAlIKViQNcYYY/zCgqwxxhjjJ26bLjbGGGP8w0ayxuSj5StdvPZWCO4U6HFdCn16pb2X5LHj8PyrxYiOEYoVg6efOEPtmsruvcKo54qlpovZLwzsm0SvW1MKugnnpdU1TXhg/D24XC6+/WAJs8Z+nSHN4PH30LpLMxJPn2Hcfe+yY/1uAIZNHsil3ZpzJO4YA5uPSE0/4OVeXHp9C5LOJLN/5wHG9X+Pk0dPFVST8t0VERfzVLNrCRLhs13reW/L8gxp2oRfxOhmnQlxBXE48RS9ln5YCDX1j/YVazOycReCcPHF3rVM3f5ThjStK9RgZOMuBIuLw2dO0efn6USUuJCXW9xEhRKlUVVm71nDRzt/LfgG5FIKrsKuQo6Kfg3zQERGi8hGEfldRNaLSBtn/VIRaZUubQcROSoi60Rki4iMK5xa+0ZEpopIw1zm+VxELhaRMiLyp4jUcdaHiMgGr/6ZJiIHReSPdPnHiUjH/GtF7qWkwKsTQ5g45gyfTU/ku0VB7Nyd9ij2g4+DqVvbzafvJ/LvJ8/w2pshANSornwyNZFPpiby4XuJFC8OV7UPhIv/z3G5hCET+zL6hlcZ0PRxOvRsS/UGkWnStO7SjMjaEfRtOIwJD0zlobf6pW5b+J9ljLr+lQzlrl20gQHNnmBQy5FEb9/PHSO6+70t/uIS4dkWXbnvx0/psuBdrq9+CbUvDEuTpkxIcf7dogv3//wZXRe8x5AVXxRSbfOfC2F0k24MWvEx3RdPoltkI2qVCU+TpkxwCZ5ueh1Dfv2UHkveZtiq2QAkq5tXN35H98WTuPPHqdxZMypD3qIoSV0+PwrLPy7IishlwPVAC1VtAnQC/soh24+q2hxoDlwvIu38XM08EZEgVe2vqptykecSIEhVd6rqceBJYJKzeTiwXFXPHrJOB7pkUsybwMi81/z8bdzioloVpWoVJSQEOndM4Yefg9Kk2bXbResWnuBZo7qy/4BwKCFtOavWuqhaxU3liAD4FruXeq1rE/PnAWJ3HSQ5KYUfPltB2xtapknT9oaWLPz4RwC2rNxBqXIlCY0oB8CGn7Zw/PCJDOWu+X4D7hRPn235dQfhkRX83BL/aRpahT0nEvjr5BGS3G6+2buRTlXqpknTvXojFuzbyv5TxwBISAzcUXt6jctH8tfJBKJPHSZJU5i37w+uiqiXJs11VRvzfcxm9p8+CkDCmZMAxCeeYPPR/QCcSj7DzuNxVCxRpmAbkAcp6vL5UVj+cUEWqAzEq2oigKrGq2qMLxlV9TSwHohMv01ELhGRlc7I+Hev0eA9zvJvIvKhsy5cRL4QkVXOo52z/llntLhURHaKyENe5X8lImucEfhAr/UnROQ5EfkVuMx7NC4idzoj0T9EJOMwxeMu4H9ebfwMcIvIE8AgPEH37LZlQEL6AlR1D1BBRCJ86Ud/iIuHShXPBcZK4UpcfNqRbJ1abpYs8wTejZuF2FjhYFzaNN8tDuLaqwNrmhggLLI8cdGHUpfj9iVQoUpomjQVqpQn7q9zL198dAIVqpT3eR/X9unAqgXrz7+yhaTSBWVSgydA7OnjVLogbaCoUSaUssVK8HGH3nzV6T5uvKhxQVfTbyqVuJD9p8+1/8DpY1QqcWGaNDVKV+DCYiX4oF0fPrtyIN2rNc1QTpULytGgbGV+P7zP73U+X25cPj8Kyz8xyH4HVBORbSLytohc6WtGESkP1AGWZbJ5EDBRVZsBrYBoZ5Q4Guioqk2Bh520E4HXVbU1cAsw1auc+sC1QBTwLxEJcdb3U9WWTtkPicjZIUUp4A9VbaOqqSdYRKQK8ArQEWgGtBaRGzOpdztgTbp1jzh5X1DVDEE1C2udstIQkYEislpEVn/wUcaRUn7J7N7Mku6ah3t7JXPsBPTqX5xZ/w2mbh0lyGuwm5QEy5YHcfWVgRdkMzQW0HSdIpmk8fWnXe8c2YOU5BQWffJzXmpXJEgmF8Gkb36wuGhUPoL+P86k77JPGNLwcmqUDs2QLyBl+vKn7YEgcdGwbBUG//IxA1d8xKC6V3BRqXOzFyWDijEh6nbG/PEtJ5MT/V3j85aC+PwoLP+4C59U9YSItAQuB64CZonISFWdnk22y0Xkd6AeMEZVYzNJswIYLSJVgS9VdbtznvJzVY139n02YHUCGnp96F0oImcPqb9xRtmJInIQqARE4wmsNzlpquEJ9oeAFCCzE0etgaWqGgcgIh8DVwBfpUtXGYhLt64LsB9olGWPZHQQqJJ+papOBiYDHIup7rc52IrhcODguX+UA3FCWIW0uytdCv41IsmpF/S4szhVKp9Ls/xXF/XruqkQgJ+p8dEJhFc992EYHhlKwv7DadPsSyC82rnGhVUN5VC6NJnp3Pty2nRrwYhrX8y/CheC2NPHqFzy3Mgt4oIyHDx9PF2a4xyOPc3plCROpySxKm4vDcpVYvcJX481i64Dp49R+YJz7a90wYUc/Dtt+w/8fYzDB0+ltn/1oT3UK1uJPScPESwuJkTdzjfRG/h+/+aCrn6eFOY0sK+Kfg3zQFVTVHWpqv4LGIJnNJmdH53zt42BB0SkWSZlfgJ0B04DC5wAK2Q+VnABl6lqM+cR6ZwPBfA+PEwBgkWkA57AfJkzIl4HlHDS/K2qmQ29fD00O+1V1tkR8EN4RtLdRKSJj+WUcMoqFA3ru9m7T9i3X0hKgoWLg7iibdpuOX7CM1oF+OqbIJo3cVO61LntCxYHcU3HABzFAltX/0lk7QgiaoQTHBLElbdfxoq5aScoVsxdQ+e7LgegflRtTh49TULskWzLbXVNE24ffgP/unkciafP+K3+BeH3hBguKh1K1VLlCHG5uK76JSyK2ZYmzff7ttIqrBpBIpQICqZphSrsOBZfSDXOX38ciaF6qQpElixHiATRLbIRS2K3pkmzeP8WWlaoTpC4KBEUQpPyVdl53NP+55r3YOfxeGb8uaIwqp8nbsTnR2H5x41kRaQe4FbV7c6qZsAeX/Kq6jYReRkYAdyZrtyLgZ2q+obzvAmwEPiviLyuqodEJNQZzX6HJ7iPdfI2U9XsTnaVBQ6r6ikRqQ9c6kN1fwUmikgYcNip75uZpNsM1AZ2O8uvAy+parSIDAMmicgVmn7uMaO6wGwf6uUXwUHwxENJPPREMVLc0L1rCrVqKl/M8cwH39I9hV17XDz7cgguF9SsoTz9+Lmg8fffsHJNEKOGJRVWE86LO8XNW49M56VvRuJyuVgwYyl7Nu3jugFXA/DNlEWsnL+eqC7NmL75dRJPJzKu/3up+Z/8cAhNrmhA2bAyfLzzTT587gu+nb6UByf0oVjxEMbM95ya3/zrDt4YMq1Q2ni+UlT599pv+eCKOwkSF7N3rWf7sXjurNUCgE//XMufxw+xLPZPvrlmIG6Uz3auZ/ux9BM9gSlF3bz4+zwmX9Yblwj/3buOP4/HcXsNzxcqPtu9mp0n4vnp4A7+e9UDuFX5Ys9adhw/SIvQ6vSo1pStRw/wRYdBAEzYtIgfD27PbpeF7owW/RAmOX+2BhZnqvhNoByQDOwABqpqvIgsBRoAZz9pV+C50na4ql7v5L/AydNeVXd5lfskcLeTNxbopaoJInIv8DieUek6Ve3jBL5Jzr6CgWWqOkhEngVOqOo4p8w/8FwJvR/PNG8ksBUIB55V1aUickJVS3vVY6lT39Ui0gvPhUsCzFPVJzLpj95APVV9SkQ6A88Bbc8GVRGZA3yhqjNE5FOgAxAGHAD+parvO+eNfwcaq2py+n2c5c/p4kBza432hV2FImPnR7n6xtk/WvHiWf77/L+zscez5z28/N/OZj5/5vS4eH2hDGf/cUHWpOUcNCwB2mUx7exLGTfh+UrU09mlsyB7jgXZcyzInmNB9pz8CLJf/tnc58+cm2utK5QgW/TH2ua8qOppEfkXnlHy3jwWEwy8ln+1MsaY8xcIv/hkQfb/AVVdcJ75C+1crDHGZMUdAFcXW5A1xhgTkM5oUM6JCpkFWWOMMQGpMH/JyVcWZI0xxgSkQPgxCguyxhhjAlJh/siEryzIGmOMCUg2kjXGGGP8xL7CY4wxxvhJkl1dbIwxxvhHIHxPtujX0BhjjMlEft9PVkS6iMhWEdkhIiMz2V5WRL4Wkd9EZKOI9M2pTBvJGmOMCUj5OZIVkSA8N3bpjOce36tEZI6qbvJK9iCwSVVvEJFwYKuIfKyqWd4n0oKsMcaYgOTrCNVHUcAOVd0JICIzgR6Ad5BVoIyICFAaSMBzt7csWZA1xhgTkHIzkhWRgcBAr1WTVXWy13Ik8JfXcjTQJl0xbwFzgBigDNBTVd3Z7deCrDHGmICUm6uLnYA6OZskmQ2L099K71pgPdARqAUsFJEfVfVYVoVakDX55taL2hZ2FYoMTUkq7CoUGTXv3FDYVTBFUZ7ubp2uiPy9ujgaqOa1XBXPiNVbX2CMem7EvkNEdgH1gZVZFWpXFxtjjAlIbhWfHz5YBdQRkZoiUgy4A8/UsLe9wNUAIlIJqAfszK5QG8kaY4wJSPn5i0+qmiwiQ4AFQBAwTVU3isggZ/u7wPPAdBHZgGd6eYSqxmdXrgVZY4wxAcnHEarPVHUeMC/dune9nscA1+SmTAuyxhhjApL9rKIxxhjjJ/k9kvUHC7LGGGMCUiD8drEFWWOMMQEpn3/xyS8syBpjjAlINl1sjDHG+IlNFxtjjDF+kmRB1hhjjPEPG8kaY4wxfmLnZI0xxhg/cdvVxcYYY4x/2EjWGGOM8ZNkt/2sojHGGOMXNl1sjDHG+IlNFxuTB62uacoD4+/FFeTi22mLmTU2/X2TYfDr99K6S3MSTycy7r532LFud7Z5L25SnYcm9eeC0iU4sDuOMfe8xanjp2lxdWPue+lOgosFk3wmmSkjPmb90o0F2dwMWl3bjMET+uIKcjH//UXMeuWrDGkGT+xLVNcWJJ5KZGzfSexYtyvbvGXKl2b0zEeJqBFO7O44Xug5nhNHThIUHMSwKYOo0+JigoJdLPzwB2aO8eTp+8KddOp9BWXKl6b7hb0LrgOy0Orapgx+vY/TtsXMevV/GdIMntCHqK7NPf3S7x2vfsk6b48Hu9DjwWtJSU7h13nrmDryY+q1rsWj7w70JBDhw+dm8/NXqwqknb7wV18A3Drseu4f25tbKvbn2KHjtOjUmPte6kVIsWCSziQzZcRHrF9SuP8jZwVCkM32S0YiUkFE1juPWBHZ57VcLD8rIiLPiUinPOZtJiLdcpG+nIgMzsu+Ao2IVBaRuc7zm0Vkkde29s5rGSwid4nI785juYg0ddIUE5FlIlIgB2QulzDkjX6MvmEMA5o8Roc72lG9QWSaNK27NCOydmX6NniECQ9M4aG3+ueY99H37uf9UZ9yf/Mn+Pl/q7jtsRsAOHroOE/fOJb7mz/B2H5v88T0BwuimVlyuVwMfes+RnV7kf6XPMpVd7SjeoOqadJEdW1OZO3K9Kk7lAn3v8dDbw/IMW/PkTeybvEG+tR7iHWLN3DHyBsBuOK2ywgpHsLApo8xuNUIrhvYmUoXhQPwy9erGdrmyQJsfdZcLmHom/0Ydd3L9G80zGlb2vdFVNdmRNaJoE+9h5kwaAoPTbovx7xNO1xC2+6tuL/Z4wxoMpzPX/sagN1//MXgqCcZ1HIEo7q9xMPvDMAVVDS+k+mvvgAIr1qBlp2bcGBPXOq6o/HHeabHqwxs9jhj+77NiBlDCqahPnCr+PwoLNm+a1T1kKo2U9VmwLvA62eXVfVMfn7wquozqvp9HrM3A3wOskA5IFdBVkQK/Ax7Pu1zGDAFQFW/BP4WkV7Oa/c2MFhVk4FdwJWq2gR4Hpjs5DkDLAJ65kNdclQvqjYxf8YSu+sgyUkp/DBrOW1vaJUmTdvurVj40TIAtvy6g1JlSxIaUS7bvFXrVmbDj5sBWPv9BtrfFAXAn+t3k7D/MAC7N0ZTrEQIIcUKb4KnXlRtYnacbUMyS2f9TNseadt/WY/WfP/hDwBs/nU7pcuVOtf+LPK27d6ahTOWArBwxlLa9vC0H1VKlCqOK8hFsQuKkXwmmVPHTqeWnRB7pGAangPPa3sg9bVdOms5bbu3TpPmsu6t+f5Dz/siQ79kkfeGQZ2Z+er/SDqTDMCRuGMAJJ4+gzvFDUCxEiGgWlBNzZG/+gJg0Ph7mDLiY9SrvX+u382h1P+Rvwr9f8RbwAfZzIjIdBEZLyJLgFdEJMoZ+axz/tZz0vURkS9F5FsR2S4irzrrg5wy/hCRDSLyqFe5tzrPW4rIDyKyRkQWiEhlZ/1SEXlFRFaKyDYRudwZUT8H9HRGZT3T1fcSJ/16Z5RWBxgD1HLWjRWPsV516unk7SAiS0TkE2CDs/yDiHzm7H+MMwJc6eSrlUl/Xek1+l8nImWccuZ6pXlLRPo4z3eLyDMi8hNwm4h0E5EtIvKTiLzhNSotJSLTRGSVU26PLF6yW4BvvZaHAi8A/wZWqepyAFVdrqqHnTS/AN7Dp6+Au7J6T+SnsCqhxEUfSl2O25dAhcjQNGkqpEsT76TJLu/ujdFcdkNLAK64tQ3h1Spk2PflN7dhx/rdqR+4hSEsMl3bohMIi0xb17AqoRz8yzvNIcIiQ7PNW75S2dSAmRB7hHIVLwRg2ee/8PfJRGbFTOHjPe8w+7WvOX74hN/al1dhkaHEebd53yHCIsunS1M+637JIm/VOpVp3L4+byx/gdcW/4u6rc79C9ePqs2U38cx+bdxTBw8NTXoFjZ/9cVlN7Tk0L4Edv6+J8t9X35LG3asK9z/EW/J6vL5UVjyejhSF+ikqikiciFwhaomO9O9L+H5YAfPCLM5kAhsFZE3gYpApKo2As/UrXfBIhICvAn0UNU4J+C9CPQ7W2dVjXKmh/+lqp1E5BmglapmNo8xCJioqh87ATkIGAk0ckboiMgtTl2bAmHAKhFZ5uSPctLuEpEOTpoGQAKwE5jq1OdhPAHskXT7Hw48qKo/i0hp4G8f+vdvVW0vIiWA7U7/7hKRT73SjAYWq2o/pw9Xisj3qnrSqy9rAodVNfHsOlXdKSKzgCFAhoMCx33AfK/lP4DWmSUUkYHAQIAGrlZUdWVVpI8yOeDUdKMIyeygVDXbvOMHvMvg1/tw91O3sOLrNSSn+5C4qGFV7nupF092eymvNc8XmbUtY/szJlL1LW969aNq405xc0fkQMqUL8X4Zc+z9vvfid11MFf19res2pxzGs02rys4iNLlS/FQ26eo17oWT818hHtqDwVgy8odDGgynOr1I3n8g8GsnL+epMSk82/MefJHXxS/oBh3PnkTI7u8mOV+L2pYlf4v92Jkl8L9H/EWCOdk8xpkZ6tqivO8LDDDGSEqEOKVbpGqHgUQkU3ARcBG4GIn4H4DfJeu7HpAI2Ch84YIAvZ7bf/S+bsGqOFDXVcAo0WkKvClqm7P5I3WHvjUadMBEfkBT1A5BqxU1V1eaVep6n6nTX961X8DcFUm+/8ZGC8iHzv7j87sjZ7OLOdvfWCn1/4/xQlowDVAdxEZ7iyXAKoDm73KqQzEeS0jIi6gE3ACz+sRn277VXiCbPuz65yDqTMiUkZVj3unV9XJOFPL14Tccd5zavH7Egivem7kFh4ZSkLM4WzThEWGcijmMMHFgrPM+9fWmNQAGlmnMlHdmqfJ/6/Zj/Fqv0ns33ngfJtwXuKi07WtaiiHYhLSptl3iIrVKrAxNU0FDsUkZGi/d97DB44SGlGOhNgjhEaU48hBz7Rox17tWb1gPSnJKRyJO8bG5Vuo26pWkQuycdGH0sw+hEVW4FC690VcdEIm/eK8L7LIG7/vED/9dyUAW1f9ibrdlA0rw9H4c2/zvVv28ffJRGo2qsa2NTv91ELf+aMvKteqRETNiry37lXAc272ndVjGHLpKA4fOEpYZCjPfvEYr/Z5u9D/R7wFQpDN6xj6pNfz54Elzsj0Bjwf9mclej1PwTMKPYxnNLgUeBCYmq5sATZ6nfttrKrXZFJmCj4cJKjqJ0B34DSwQEQ6ZpIsu1fqZLpl7za5vZbdmdVHVccA/YELgF9EpD6QTNq+L5Eu29l9ZlcvAW7x6qfqqro5XZrTmZT9IJ6R6X3AJPGK+CLSBM/r0UNVD6XLVxzfRuHnZeuqP4msHUFEjXCCQ4K4smdbVsxdkybNiq/X0PnuKwCo36Y2J4+dIiH2SLZ5y4V7pkdFhF6jbuKbyZ7T/6XKluT5OSOY9tSnbFq+zd/Ny9HWVTuIrFOZiBoVCQ4JpkPPdqyYszpNmhVzVtOp95UANGhTh5NHz7Y/67wrvl5N53s7AND53g4sn+O5Uvbg3niaXdUIgBIli9OgTV3+2rKvgFrru/SvbYeebVnxdbp++Xo1nXp73hdp+yXrvMv/t4rmV10CeA6+gosFczT+OBE1wlMvdKpYPYxq9SoTuzvN8Wqh8Udf7P7jL26vPJDetYbSu9ZQ4qIP8UCrkRw+cJRSZUvywtcjeX/0p2xcvrUwmpylQDgnmx9nr8sCZ/8r++SUWETCgDOq+oUzEpyeLslWIFxELlPVFc70cV1Vze6a8eNAmSz2dzGe0eAbzvMmwG/p0i8D7heRGUAocAXwOJ6R5HkRkVqqugHPOd3LnDLXAA1FpDieIHg18FMm2bfgGfXXUNXdpL34aAEwVESGqqqKSHNVXZcu/za8RvsiEoHnQqgoZyp+AJ4DgCkiUh3PLEFvVU0TbUSkAhCnqn6fK3OnuHnr4Q946ZtRuIJcLJi+hD2borluoOfC828mf8/K+euI6tqM6Vsmer7C0//dbPMCdLijHd0HeY7VfvpqJQumLwWgx+BriaxVibtG38xdo28G4MmuL6VeAFPQ3Clu3hr6Pi9/O9rThg88bbj+/s4AzH1vISvnraVNt+bM2P4miafOMK7fpGzzAswc81+enjWMrv06cnBvPM/fPh6A/01awOPTBjNlw3hEhAXTl7Brw14A+r9yNx3vbE/xksX4ZO+7zH9/ER/+e3Yh9IrTtoem8fJ857X9YKnTL573xdz3vmflvHW06dqcGdsmevrlvneyzQvw7bQlPPb+A0z+bRzJZ5IZ2/dtABq1r0/PJ3qQkpSC2628MeR9jh06nnnlCpi/+iIrPR7sQpXalbh79C3cPdpzJnBklxcL7X/EmwbASFZyOmeTmlDkWTxTjI2Auar6ubP+MmAGnmnJxXg+pGs4F/Kknid1LtgZBxwGPuDcSO5JVZ0vItPPlisizYA38ATwYGCCqk4RkaXAcFVd7QTr1c6+QvEEnRDgZVU9O92KiDwJ3A0kAbFAL1VNcC5maoLn3OMTwKtAVzxT3i+o6iznHOxwVb3eKSv9snd90mzz2v+beKaRU4BNQB9VTRTPhWA98JxzPQPMUdXpIrLb6bd4J/8NwFg807orgUqqepeIXABMANriGdXuTr9vJ/8i4H5V3eG0+UdVfcfZVg34EWjhtP8W4OxVD8mq2spJdytwmao+lr58b/kxXfxPoSkpOSf6/0KKxldfTNGyMGXWeUfIKxY97vNnzrKrxxZKRPY5yJrCISKlVfWEM607Cdiuqq/nIv9NQEtVfeo86vAlnoOhbOeKLMieY0HWiwVZk4n8CLLtFo7w+TPn586vFEqQLRpfdjLZGSAi9wLFgHXAe7nJrKr/daZ788S5IvurnAKsMcYUtEC48MmCbBHnjFp9HrlmUUb6i8tyk/cM8J/z2b8xxvhDIJyTtSBrjDEmINlI1hhjjPGTQLikyIKsMcaYgJRSiD+X6CsLssYYYwKSTRcbY4wxfmLTxcYYY4yf2NXFxhhjjJ9YkDXGGGP8xM7JGmOMMX7idhf9IFv0r382xhhjMqEqPj98ISJdRGSriOwQkZFZpOkgIutFZKNz7/Fs2UjWGGNMQMrPi4tFJAjPTVg6A9HAKhGZo6qbvNKUA94GuqjqXhGpmFO5NpI1xhgTkPJ5JBsF7FDVnc5vts/EcztSb72AL1V1r2f/ejCnQi3IGmOMCUyai0fOIoG/vJajnXXe6gLlRWSpiKwRkXtyKtSmi40xxgSk3HyFR0QGAgO9Vk1W1cneSTLbRbrlYKAlcDVwAbBCRH5R1W1Z7deCrMk3dqNyk5nkq1sUdhWKjOL7jhV2Ff5RcnN1sRNQJ2eTJBqo5rVcFYjJJE28qp4ETorIMqApkGWQteliY4wxgUnF90fOVgF1RKSmiBQD7gDmpEvzP+ByEQkWkZJAG2BzdoXaSNYYY0xAys/fLlbVZBEZAiwAgoBpqrpRRAY5299V1c0i8i3wO+AGpqrqH9mVa0HWGGNMYMrnGwSo6jxgXrp176ZbHguM9bVMC7LGGGMCkv12sTHGGOMnGgA/q2hB1hhjTGCy+8kaY4wx/mIjWWOMMcY/bCRrjDHG+IkFWWOMMcZP7OpiY4wxxj/UXdg1yJkFWWOMMYHJRrLGGGOMf4idkzXGGGP8xIKsMcYY4yc2XWyMMcb4icOqWA0AACAASURBVI1kjTHGGD+xq4uNOT+trm3G4Al9cQW5mP/+Ima98lWGNIMn9iWqawsSTyUytu8kdqzblW3eAa/25tLrW5J8JpmYPw8wrt8kTh49VaDtyo4/2lymfGlGz3yUiBrhxO6O44We4zlx5CRlQkvzzOzHqNe6Nt/NWMpbQ99P3ceVt7el16ibcQW5+HXeWqaO+KhgOuA8tG5VkyEPXE2Qy8U33/7Gp7N+TbO9U8eG3HF7GwBOn05iwpsL+HNnXGFU1S9atq/DAyOvwxXk4tsvVvPZ1GVptletGcZjL9xCrYZVmDFxIV9M/wmAsIiyPP7yrZSvUBpVZd7sVfzvoxWF0YTcCYDpYldeMonIifzYuYh0EJG5+VFWHvZdQ0R65TadiLQSkTfyqQ6j0i0vz49yM9lPcxGZ6jwfJiLve227S0S+cZ4PEpENIrJeRH4SkYbO+nDnRsUFyuVyMfSt+xjV7UX6X/IoV93RjuoNqqZJE9W1OZG1K9On7lAm3P8eD709IMe8axf+xoDGw7i/2XD2bY/hzidvKuimZclfbe458kbWLd5An3oPsW7xBu4YeSMASX8nMf2ZWUx+/D9p9lEmtDQDX+3NE52eY0DjYZSvWJbmHRsVQA/kncslPDykMyNHz6bPgKlc3aEhF1WvkCbN/tijPDL8E/oP+oAPP1nOY490KaTa5j+XS3hw9A08NWgGA7tPpEO3JlSvFZ4mzfGjp3nn5bl88cFPada7k91MeXU+A7tP5JE73+WGOy/NkLcoEvX9UVjyFGT/IWoAOQbZ9OlUdbWqPpRPdUgTZFW1bT6Vm9l+3nSevwG0FJF2IlIOeAEY6mz7RFUbq2oz4FVgvFOvOGC/iLTzU/0yVS+qNjE7YonddZDkpGSWzvqZtj1apUlzWY/WfP/hDwBs/nU7pcuVIjSiXLZ51yz8HXeKZ55p8y/bCYtM+0FcmPzV5rbdW7NwxlIAFs5YStseUQD8fSqRjT9v4czfSWn2UfniSkRvi+Fo/DEA1i3aQPtbLvVn089b/XqViYk5wv7YoyQnu1n8w2bata2TJs3GTfs4cSIRgE2b9xEWVqYwquoX9RpXZf9fCcRGHyY5KYUf5v3OZVc1SJPmaMJJtv2xj5TklDTrE+KPs2NzDACnT53hr51xVKh4YYHVPc80F49Ccl5B1hmJ/iAin4nINhEZ44yMVjojolpOuuki8q6I/Oikuz6TskqJyDQRWSUi60Skh7O+j4h8JSJfi8guERnijMbWicgvIhLqpKslIt+KyBpnP/W99v2GiCwXkZ0icquzyzHA5c6o7VFnxPqjiKx1Hm2zSJc6+haRUKduvzt1aeKsf9Zpy1JnnxmCsoiMAS5wyv3YWXcil/0aLiJfOH22KrMgKCJlgCaq+huAqiYDg4FJeALpNFXd6Ww75pW1FGnfml8Bd2X3fshvYZGhxEUfSl2Oj07IEBDDqoRy8C/vNIcIiwz1KS/AtX2vYtW36/xQ+7zxV5vLVypLQuwRABJij1Auhw/QmB2xVKsfSaWLwnEFuWjbozXhVYvOwUhmwsLKcDDu3Fs4Lu44YRVKZ5m+W5emrFy1syCqViAqVLqQuP1HU5fjDxyjQqWyuS6nUpVy1GpQma2/R+dn9f7fyo9zsk2BBkACsBOYqqpRIvIwnhHSI066GsCVQC1giYjUTlfOaGCxqvZzRlgrReR7Z1sjoDlQAtgBjFDV5iLyOnAPMAGYDAxS1e0i0gZ4G+jo5K8MtAfqA3OAz4GRwHBVvR5AREoCnVX1bxGpA3wKtMokXQevOv8bWKeqN4pIR+A/QDNnW33gKqAMsFVE3lHV1OGCqo4UkSHOqDGv/ToReF1VfxKR6sACJ4+3VsAf3itUdbmIbAY6pU8vIg8Cw4BiXv0HsBrPqJd06QcCAz0NbkFVuTiL5uSeZHK6RVXTpcmYSNW3vL1G3UxKsptFH/94XvXMT/5us69OHDnJG4OnMHrmo6hb2bRiKxE1K+WprIKS2dm5rJrfrGl1unVpwkOPFv3zzL6STHogt69/iZLFeGpCL94b8w2nTibmV9X8Rv6f3LR9laruBxCRP4HvnPUb8ASZsz5TVTewXUR24glC3q4BuovIcGe5BFDdeb5EVY8Dx0XkKPC11z6aiEhpoC0w2+sDqLhX2V85+94kIll9UoQAb4lIMyAFqOtD29sDtwCo6mIRqSAiZw8dv1HVRCBRRA4ClYDcHBr60q+dgIZebb5QRMo4fXVWZSDNlR1Of7XC0+Zw73qp6iRgknjOQz8F3OtsOghUSV9JVZ2M5wCHzq7b8nVSJi46Ic3oKaxqKIdiEtKm2XeIitUqsDE1TQUOxSQQXCw427yd77mSNte15IlO/87PKp83f7X58IGjhEaUIyH2CKER5Thy8Bg5+WXuGn6ZuwaAbgM6kZJStC/ljIs/TsXwcyP08PAyHErIePnIxTXDGf5oF0aOns2x438XZBX9Kv7AUcIrnxu5hlW6kAQfXuezgoJdPD2hF0u++Y2fv9/kjyrmvwD4Ck9+nJP1Ptxxey27SRvE03dH+mUBblHVZs6juqpu9nEfLuCIV95mquo9QvPOn9Whz6PAATwjyFZ4RnI5yfTgOZN9ppD7Axpf+tUFXObV5sh0ARbgNJ4DFm//Bj4CXgRez2L/M4EbvZZLOGUVmK2rdhBZpzIRNSoSHBJMh57tWDFndZo0K+asplPvKwFo0KYOJ4+eIiH2SLZ5W13bjJ5P3MgzPV4h8fSZgmxSjvzV5hVfr6bzvR0A6HxvB5bPWZVjXco5Aat0uVJ0f+Ba5k9dlI8tzX9btu4nMrI8ERFlCQ520fHKBixfsSNNmorhZXjumZt4+dVviN53uJBq6h9b/9hHleoVqBRZnuCQIK7s1oRflmzxOf+jz93M3p0H+XLGz36sZT4LgHOyBfkVnttEZAZQE7gY2Ap4X0mxABgqIkNVVUWkuar6dLJMVY8552tvU9XZ4hnapZ6HzMJxPFO5Z5UFolXVLSL3AkFZpPO2DM95yuedaeR4py6+VBsgSURCvKeRc+k7YAgwFkBEmqnq+nRpNgOPnV0QkcbAdXimtc8A/USks6ouFJE6qrrdSXodsN2rnLqkm3b2N3eKm7eGvs/L347GFeRiwQdL2LMpmuvv7wzA3PcWsnLeWtp0a86M7W+SeOoM4/pNyjYvwJA37yOkeDCvfPc0AJt/3cbEB6YUZNOy5K82zxzzX56eNYyu/TpycG88z98+PnWfH+6cRMkLSxJSLJi2PVoz8toX2Ls5msET+nJx0xoAfPT8bPZt31+wnZFLbrfyxlsLefWl23G5hPkLNrB7Tzw3XOc5I/P1N+u55+52XHjhBTwy1NOfKSluBg35T3bFBgx3ipu3X/yaFyf3weUSvvvvWvb8eZBut3sucpv32UrKh5XmjVmDKVm6OOpWbuzdlvu7T6RmvQg69WjOrq2xTPpiCADTJ3zHqh+3FWaTchQIv10seTlnIyInVLW0E1i8z1cudZZXe28TkenAYTwjxErAMFWdmy7NBXjOrbbFM0Lc7azvA7RS1SHOPnY7y/He20SkJvAOnunREGCmqj7n7Huuqn6eru4hwLdAGDAdmAt8AZwClgBDs0i3zqvOocAHeA4cTgEDVfV3EXkWOKGq45x9/gFcr6q70/XjK0B3YK2q3pWHfg3DcwFTAzwHTMtUdVAmr9cGp19PAD8CL6rqfGdbK86dSx6LZwo6yXm9hqjqRifdcCBRVd9MX/5Z+T1dbP4Zkju1yjnR/xPF9/k+fftP9+3GF8/7hGqt18b7/Jnz52PDCuUEbp6CbK53ki7QmYIlIo8Cx1V16nmUsQzooapZzrFZkDWZsSB7jgXZc/IlyI7LRZAdXjhB9v/z92T/P3mHtOd4c0VEwoHx2QVYY4wpaOIWnx+FpUDOyapqn4LYj8mcqv4NfHge+ePwfE/WGGOKjgCYO7PfLjbGGBOQAuHCJwuyxhhjApMFWWOMMcY/bCRrjDHG+EvR/hEywIKsMcaYABUII1n7Co8xxhjjJzaSNcYYE5gCYCRrQdYYY0xACoTpYguyxhhjApMFWWOMMcY/JACuLrYLn4wxxgSmfL6frIh0EZGtIrJDREZmk661iKSIyK05lWlB1hhjTEAS9f2RY1kiQXhuHdoVaAjcKSINs0j3Cp57oOfIgqwxxpjAlL8j2Shgh6ruVNUzwEygRybphuK59/hBXwq1c7LGGL9ynUkp7CoUGSnlSxZ2Ff5ZcnHhk4gMBAZ6rZqsqpO9liOBv7yWo4E26cqIBG4COgKtfdmvBVljjDEBKTdf4XEC6uRskmR209n0e5gAjFDVFBHf7lFrQdYYY0xgyt+ri6OBal7LVYGYdGlaATOdABsGdBORZFXN8n7bFmSNMcYEpHz+MYpVQB0RqQnsA+4AenknUNWaqfsWmQ7MzS7AggVZY4wxgSofg6yqJovIEDxXDQcB01R1o4gMcra/m5dyLcgaY4wJSPn9s4qqOg+Yl25dpsFVVfv4UqYFWWOMMYHJflbRGGOM8Q+7QYAxxhjjLxZkjTHGGP+wkawxxhjjLxZkjTHGGD+xIGuMMcb4h00XG2OMMX4SCDdttyBrjDEmMNlI1hhjjPEPmy42xhhj/MWCrDHGGOMnFmSN8U2ra5sxeEJfXEEu5r+/iFmvZLx71OCJfYnq2oLEU4mM7TuJHet2ZZu3TPnSjJ75KBE1wondHccLPcdz4shJ6rWuzaPv3e8pVODDf8/m569WAnDl7W3pNepmXEEufp23lqkjPiqYDsiEy+Vi0qoxxO9L4OnuY9Jsq1avCsOnPUjtFjX54KlP+fy1r3Nd/h0jb6RLv6txp7h5++FprP7uNwDGLX6W0MrlOXP6DAAjr32eI3HHzr9BftI66mIeHNIZV5Aw75vfmPnJijTbq1WvwBMjrqN2nQimvf8Ds2f9mrrtlltb0+26Ziiwa+dBXn1lLklnUgq4BfmnVZuLGfzwNbhcwvy565n1Uca+GD7qemrXjeCDKUv5/FNPX1StFspTz92cmi6iSjlmTP2B/85eVaD1z61AmC525ZRARFREPvRaDhaROBGZ69+qFR4RqSEivXJOmSbPqDzu6xERKZmXvLncz40i8ozz/FkRGZ5JmmkiclBE/ki3fpyIdPRX3VwuF0Pfuo9R3V6k/yWPctUd7ajeoGqaNFFdmxNZuzJ96g5lwv3v8dDbA3LM23PkjaxbvIE+9R5i3eIN3DHyRgB2/7GXwa1HMKjF44zq+iIPvzsQV5CLMqGlGfhqb57o9BwDGg+jfMWyNO/YyF/NztFND3dj7+Z9mW47nnCCSQ9Py1NwBajeoCoderZjQKNHGdX1RYZO6o/Lde7jYMzdExnU4nEGtXi8SAdYl0t46OFreXLELPrdO5mOHRty0UVhadIcP3aat95YmCa4AoSFleamW1rzwP0f0L/vFFwuFx07NizI6ucrl0sYOqwLo4bPpP/d73FVp0uoXiNjX0ya8B2fz0zbF9F/JTCo71QG9Z3K4PveJ/HvJH5etrUgq58n4lafH4UlxyALnAQaicgFznJnPDe0/SerQbqb9fog0yArHtn18yOA34Ms8ATwdg5ppgNdMln/JjAyvyt0Vr2o2sTsiCV210GSk5JZOutn2vZolSbNZT1a8/2HPwCw+dftlC5XitCIctnmbdu9NQtnLAVg4YyltO0RBUDi6TO4UzzX/hcrUQzU8w9Y+eJKRG+L4Wi8J6isW7SB9rdc6q9mZyssMpQ23Vow//1FmW4/EneMbav/JDkpOcO2q++6nDd/eZl31471HEC4Mr792vZoxdJZP5N0JpnY3QeJ2RFLvaja+d4Of6tfvwr79h1m//4jJCe7WbJ4E23b1UmT5siRU2zdup/klIzf9wgKclG8eDCuIKFEiWDi408UVNXzXb0GVYiJTiA2xtMXS7/fRNv2ddOkOXLkFNu27Cc5OevRevOWNdi/7zAHDxTdg6tUmotHIfElyALMB65znt8JfHp2g4hEichyEVnn/K3nrO8jIl+KyLcisl1EXvXK846IrBaRjSLyb6/13URki4j8JCJvnB0ti0gpZ5S1ytlPD699fCUiX4vILhEZIiLDnDS/iEiok66WU481IvKjiNR31k939rNcRHaKyK1OVcYAl4vIehF51LsjRKSyiCxztv0hIpeLyBjgAmfdx85IeLOIvA2sBapl1mYReQioAiwRkSXOumtEZIWIrBWR2SJSOqu+ERGX07fhThqXiOwQkbB0da4LJKpqfHYvsqouAxIyWb8HqCAiEdnlz6uwyFDiog+lLsdHJxAWWSFtmiqhHPzLO80hwiJDs81bvlJZEmKPAJAQe4RyFS9MTVc/qjZTNoxn8u+vMfGBKbhT3MTsiKVa/UgqXRSOK8hF2x6tCa+ath4F5YHX+zJlxEe43bn7ImD1+pFceXtbHmn/FINaPI47xU3Hu9pnSBcWWYE4r/6M25dAWGRo6vLwaQ/y7tqx3PXULXlvRAEICy9DnNdIOy7uOGHhZXzKGx9/gtmzfuXTz4Yw+4uHOXEikTWrd/mrqn4XFl6GuIPHU5fj44753BfeOnS6hCXfb8rPqvmNqO+PwuJrkJ0J3CEiJYAmgPdcwxbgClVtDjwDvOS1rRnQE2gM9BSRas760arayinrShFp4pT9HtBVVdsD4V7ljAYWq2pr4CpgrIiUcrY1wjPqjAJeBE45dVkB3OOkmQwMVdWWwHDSjugqA+2B6/EEV/CM2n5U1Waq+nq6vugFLFDVZkBTYL2qjgROO+nvctLVA/6jqs2dIJWhzar6BhADXKWqVznB8Smgk6q2AFYDw7LqG1V1Ax8BZ/fZCfgtk2DaDk+wPx9rnXLSEJGBzsHD6mjdmaeCRTKuU9V0aTImUvUtb2a2rNzBgMbDGBI1kjtG3kRI8RBOHDnJG4OnMHrmo7y+7HkO7IkjJbngv+3e5roWHIk7yva1ue/P5lc3pm7Li5m0cgzvrh1L846NqXxxpQzpsuu3l+9+g4FNH+PRK56mcfsGdOp9Ra7rUZh8ePkBKF26BG3b1eGuO97m9lve4IILQujU+RL/Vs6P8vq/4C042MVl7erww5LN+VQrPwuAkaxPFz6p6u8iUgPPKHZeus1lgRkiUgdPU0K8ti1S1aMAIrIJuAj4C7hdRAY6+68MNMQT8Heq6tlDyU+Bgc7za4DuXucRSwDVnedLVPU4cFxEjgJnT1JtAJo4I8G2wGyvD+riXnX8yglWm0Qk46dRRquAaSIS4uRdn0W6Par6i9dyZm3+PV2eS531Pzt1LYbnYKE+WffNNOB/wASgH/BBJnWpDMT50LbsHMQz6k5DVSfjOYihs+u2PL2V46IT0owYw6qGcigm7YA6bt8hKlarwMbUNBU4FJNAcLHgLPMePnCU0IhyJMQeITSiHEcOZpz+2rtlH3+f/Juajaqxbc1Ofpm7hl/mrgGg24BOpGQyxehvl7Srz2U3tCKqa3OKlShGyQsvYMR/hvLKPW/mnFngu//8wLRRn6RZ3e7GKHo/cxsA4we8Q1z0IcKrneu38MhQDsUcBkjtv9Mn/mbxpz9RP6oO33+4LJ9al7/i444THn5uhiI8vAyH4o9nk+OcFi1rELv/CEePngLgx2VbaXhJVb5fuDGHnEVT3MHjhFc8N3INC7+QQ7mc/m59aW12bIvlyOGT+V09v/hHXPjkZQ4wDq+pYsfzeAJdI+AGPAHwrESv5ylAsIjUxDOavFpVmwDfOHkyOQ5LJcAtzkixmapWV9Wzh1re+3B7LbvxBDQXcMQrbzNVbZBFHbOrA5A6pXoFnvPSH4rIPVkkTX2XZtPmzNq50KueDVX1vuzqpap/AQecC5Pa4JnaT+90FvvLjRJOOflu66odRNapTESNigSHBNOhZztWzFmdJs2KOavp1PtKABq0qcPJo6dIiD2Sbd4VX6+m870dAOh8bweWz/FcKRlRoyKuIM9bv2L1MKrVq0Lsbs8xSDnnA7t0uVJ0f+Ba5k/N/JyoP00b9Qm9qg+i98UP8uKdr7N+8R++BVhg3aI/uOKWS1PbUaZ8aSpWD+Pnr1amXsi0bc1OVsxZTYee7QgpFkxEjYpE1qnM1pU7cAW5uLCC54M6KDiINte1ZPcfe/3W1vO1ZWsMkVXLExFRluBgF1d1bMjy5dt9ynvw4DEaNIykeHHPWKNFixrs3XMoh1xF19YtMURWCyWisqcvOnRqyIqft+WqjKs6NWTJ94FzkCFu3x+FJTdf4ZkGHFXVDSLSwWt9Wc5dCNXHh3IuxBOAjjojx67AUjzTzheLSA1V3Y1nmvmsBcBQERmqqioizVV1nS+VVtVjzvna21R1tniGiE1U9bdssh0HMj2ZISIXAftUdYozZd0C+A+QJCIhqpqUizZ77yse+AWYJCK1VXWHeK46rkr2fQMwFc+08YeqmtkVDZuBu7Npry/qArPPs4xMuVPcvDX0fV7+djSuIBcLPljCnk3RXH9/ZwDmvreQlfPW0qZbc2Zsf5PEU2cY129StnkBZo75L0/PGkbXfh05uDee528fD0Cj9vXpOeJGUpJScLvdvPHgVI4d8ox+Bk/oy8VNawDw0fOz2bd9vz+anCfe/VG+UjkmrRpDyQsvQN3KzQ9fR/9LHmXv5mg+eHomYxY8jbiE5KQU3hoylYN7055B2LMpmmWzVzB14+ukJLt5c8hU3G43JUoW5+VvnyI4JAhXkIt1izYwb0rBH2j4yp2ivDnxO14Zewcul4v5839jz+54ru/eHIC5c9ZRPrQU77zXl5Ili6Oq3HJra/rdO5ktm2NY9sMW3p1yHykpbnZsj+WbuT59rBRJ7hTlrfELeHn8nbhcLhZ88xt7dsVzfY8WAMz931rKh5Zi0tR+lCxV3PO+uS2K/ne/x6lTZyhePJiWrWsyYWxmx+lFVACMZCWnOXsROaGqpdOt6wAMV9XrReQyYAae6cjFQG9VrSEifYBWqjrEyTMXGKeqS0VkOp5R1048I8k5qjpdRG4AxuIJOCuBSqp6l3iubJ6AZ9pXgN3OvtPvY7ezHO+9zRlJvoNn2jQEmKmqzzn1mKuqn3u31ZkK/hYIA6Z7n5cVkXuBx4Ek4ARwj6ruEpFXgO54zl2Odspt5JUvqzYPBR4E9jvnZTsCr3BuSvspVZ2TVd84ZYcAh4AoVd2SyWtYEs80dyPnIOVZPFc1p84lqWpVEfkU6OC0+wDwL1V93yn/d6Cxqma8nNWR1+li88/mvqJ5YVehyHAl27/IWQt/Gp3jzGFOLr17vM8d+stHw857f3mRY5AtSCJSWlVPOKPNScD2TC48+n8pu74RkVbA66p6eTb5JwJfq+r3edj3TUALVX06u3QWZE1mLMieY0H2nHwJsne95nuQ/fixQgmyuTknWxAGiMh6YCOeaej3Crk+RUmmfSMiI4EvgCdzyP8Sef8+bjDwWh7zGmOMXwTCV3iK1M8qOiMzG7lmIqu+UdUxnPvqUXb5D+C5eC0v+/bLuVhjjDkvATAxUKSCrDHGGOMrCYCfmbYga4wxJiAFwvdkLcgaY4wJTEXowt2sWJA1xhgTkGwka4wxxviLBVljjDHGP2wka4wxxvhJYd6M3VcWZI0xxgSmoh9jLcgaY4wJTDZdbIwxxviLTRcbY4wxflL0Y2yRu0GAMcYY4xNxq88Pn8oT6SIiW0Vkh3PzlfTb7xKR353HchFpmlOZNpI1xhgTkPLznKyIBOG5jWhnIBpYJSJzVHWTV7JdwJWqelhEugKT8dwnPEs2kjXGGBOYNBePnEUBO1R1p6qeAWYCPdLsTnW5qh52Fn8BquZUqI1kTb5xFS9R2FUoMubv+rWwq1BkdKkeUthVKDKCIioWdhX+USQXv10sIgOBgV6rJqvqZK/lSOAvr+Vosh+l3gfMz2m/FmSNMcYEJrfvSZ2AOjmbJJJZtkwTilyFJ8i2z2m/FmSNMcYEpNyMZH0QDVTzWq4KxGTYp0gTYCrQVVUP5VSonZM1xhgTmNzq+yNnq4A6IlJTRIoBdwBzvBOISHXgS6C3qm7zpVAbyRpjjAlI+Xl1saomi8gQYAEQBExT1Y0iMsjZ/i7wDFABeFtEAJJVtVV25VqQNcYYE5jy+abtqjoPmJdu3btez/sD/XNTpgVZY4wxAUlyceFTYbEga4wxJjDl80jWHyzIGmOMCUh2P1ljjDHGX2wka4wxxviJnZM1xhhj/COff4zCLyzIGmOMCUwWZI0xxhg/sSBrjDHG+IekWJA1xhhj/MNGssYYY4yfWJA1xhhj/MSCrDHnp1Xnxgwa15ugIBfzpy/ls3FzM6R54LXeRF3blL9PJfLawMnsWL+H8KqhPD71fspXKou6lXnTlvDVpO8AuPzmKHqPvolq9avw0OXPsn3troJu1nn78Vd46U1wu+HW62DAXWm3Hz0Oo8fAXzFQvBi8MALqXuzZ9p/PYfZcz+fTbdfDvbcVfP3PV6trmvDA+HtwuVx8+8ESZo39OkOawePvoXWXZiSePsO4+95lx/rdwP+1d99hUpV3G8e/96LGgiK4KIiCBRWsgKJYooBdURONxhLbq2gS65toLNErxWh8jUmsiVFjibHHGKNGRFGxYi9osCACgiICIhYswO/945xlZpcVV2XmmXJ/rmsu5pSd/Z3h3nnmec4zZ+Anlx7BgF36MvPdWRzR96T5+w/97f4MGNKPzz+bw9vj3uHcw//CR+9/XK5D+to23qYXP/zlnjS0E8NuGMXNfxqxwD4//NWe9B/Um09nf87vf3odr784iW5rrMgpFx88f5+u3Vfgmj/cxb/+OpLVe6/MMWftw5LLLMHUSTM459hr+PjDT8t5WG1TBZ+Trdjvk5X0YYvlQyRdtIh/x38kLb8oH3NR+KrHKul4SUsvZPs/JK2R3x8vabSkFySNlNSjaL8rJE2V9GKLnz9X0uCvcyzfREODOOq8YHI7UgAAHQ9JREFUgzltj98xtO9JDNp7c7r3WrnZPv133Ihua67EoeufwPlHX8ExFxwKwNw5c7n05OsY2vdkjtvmV+x25Hbzf3b8S5P49b7nM/rhV8p9SIvE3Llwxnlw6Tlw+9Vw5wgYO775Ppf+HXqvBbddCWefCr+9MFv/6risgb3pEvjXX+GBx2D8pLIfwjfS0CCOPv9Qfr7bOQzd6EQGfn8Luvfu1myf/jv1oVvPLhy67k8470eXc+xF/zN/2z1/e5BTh/zfAo/7zIjRDO3zM3648clMeu1t9j1p95IfyzfV0CCO+s33OP3gv3DktmczcPd+dF9rpWb79B/Um5VX68xhW5/JBSffyNFnZu+qJo+bytE7/46jd/4dx+56Lp/M/oxHh70AwPHn7MuVZ9/Oj3c4h0eHjWavI8v+598mimjzLZWKbWTLISJ2iYiZqesoJunrjC4cD7TayEpaD2gXEeOKVg+KiA2BB4DTitZfBezUysNcCJz8Ner6RtbpvyZvvf4OU8a/y5zP5/LAzaPYfMjGzfbZfEg/7r3uYQBefuJ1lumwNJ26dGDGlPcZ+9wEAGZ/+AlvvvwWjSt3AuDNV95i0mtTynswi9ALY6B7N1h1ZVhicdhlMNz3cPN9xo6HAf2y+2v0gMlTYNoMGDcBNloXlloSFlsM+m8E9z5Y9kP4Rtbp3zPLxRtTmfP5XEbe9Bhb7NY8F1vstjH3XPsQAC8/MZZlll+aTl2y99OjH36ZD977cIHHffre0cybm3WNXn58LJ27rVDiI/nm1u7Tg7fGT2PKxOnZc3H7swzYYYNm+wzYYQNG3PIkAC8/O4H2yy1FxxWXa7ZPny3X5u2J05g6+T0AVlljRUY//joAzzz0ClvtslEZjuZrmDuv7bdEqrKRldRD0oi8NzYi/7Z6JF0l6XtF+32Y/9tV0oOSnpP0oqRv5+vHS2qUtJqkMZIuk/SSpOGSlsr36Z//nsck/a5lLy/fZ2DeK7xJ0quSzpZ0gKQn8l7jmvl+u0l6XNKzku6VtFK+/peSLpU0HPhbi8feNf/djZJ2yO8/I+lmSe0lHQusDNwv6f5Wnq4DgNu+4Kl8DJjfBYiIB4EZLXeKiAnACpK6fMHjlMQKK3fk3UmFcqZNnkFjt47N9mlsZZ8V8sa0yUrdG1mzTw9efnJsaQsuk6nToMuKheWVOsM705rv02tNuCdvPF8YA2+9A++8C2utDk89D++9D7M/gQdHwZSp5at9UWjs1pF3J02fv/xuK//nK6zckXffLMrFpBmssHLz7CzMjocM5Mm7n/vmxZZYY5cOvPvWe/OXp709kxVW6tBsnxW6dGDa20X7TJlJY5fm+2yzez9G3vbM/OXxr7zNgO3XB+Dbu/ahsWvFDfhlItp+S6SSG9ml8kbxOUnPAb8u2nYR8Le8N3YtcMGXPNb+wN0R0QfYCGjtr2ct4OKIWA+YCeyVr78S+GFEbA7MXcjv2Ag4DtgAOBBYOyI2BS4Hjsn3eRgYEBF9gRuAnxX9/MbAHhGxf9MKSd8l60Hukq86DdguIvoBTwE/iYgLgLfIeqeDWqlrS+DpL6h5J+BfCzmmYs/kj9WMpCMkPSXpqUlzXmvjQ7WNpAXWRcs/li/ZZ8llvsXp1x/LJSdey8cffLJI60ultdeLls/C0ANg1gfw3cPg77dA757Qrh2suRocvj8c9lMYeiL06gntqm1mRhty0Vp2aOPr7H4n78HcOXMZcd0jX6e68mrlMFsGpPVdCvsstng7Ntt+PR66s/Cy+McTr2e3g7figjt/ylLtv8Wczxf20pdQFTSylfznNTtvFIHsPCWwSb64ObBnfv8a4JwveawngSskLQ78KyJaa2TfKFr/NLBafr522Yh4NF9/HTDki35HRLyd1/o6MDxfPxpoavxWAW6U1BVYAiiecfPviJhdtDyI7Hh3iIhZkoYA6wKP5C8gS5D1RL9MV+DdFuvuz3vRU2k+XLwwU8l6zM1ExKXApQA7LnXgIk3ytMkz6LxKoYfS2K0T09+a+aX7zMjftbdbrB2nX38s9934KI/c9tSiLC2plTo3732+8y6s2Nh8n/bLwFmnZPcjYLt9YZWu2fL3ds1uAH+8NHu8ajJt0gw6r1IYyu1c9H8+f5/JM+i8alEuVunE9Bb7tGb7A7/NZrv046Qdz1x0BZfQtLffp3NRD72x6/JMnzqr+T5T3qexa0eaXm4auyzP9HcK+2wysDevvziJmdMKQ+iTXp/Kz39wCQDdVu/MpoPXLeFRfANVMLu4knuyX0XTMz2H/JiUtURLwPxh0K2BycA1kg5q5TGKp87NJXsD0tqbwC9S/PPzipbnUXgzcyFwUURsABwJLFn0Mx+1eLxxwLLA2vmygHsiok9+WzciDmtDXbNb/B7IGvAewEs0HyFYmCXzxyqbV54aR7eeXVipR2cWW7wdA/cewKg7n2m2z6g7n2G7/bcCoNema/LxrI+ZMeV9AH5yyeG8+cpb/POCYeUsu+Q26AUTJsGkt+Gzz+E/98GgFmMMsz7ItkE20WmTDbOGF2B63ta89Q7c8xDsul35al8UXnnqdbr17EKX1bJcbLPP5jx2R/PBmsfueJrtD/g2AL027clH789mxpSFT7/YZIcN2eeE3fjFnufy6ezPSlb/ovTq8xNZefVGVlq1U/Zc7NaXUfc0P6M16p4X2Xav/gD06tuDjz6YzXtFDfHAPfrxwG3N/646rNAeyEYE9j12B/7z90epSPOi7bdEKrknuzCPAvuS9WIPIBuGBRhPNux6E7AHsDhk53CByRFxmaRlgH60OPfZmoh4T9IHkgZExKj8d34THcgaeoCDF7YjMAE4AbhV0t7AKOBiST0jYmw+m3iViHgV+ICsQZ7WyuOMAXqSPTfzRcRsSccDoyX9JiIWOBfbwtrAzV+yzyI1b+48Lv7fv3HW7SfS0K6B4Vc/yIQxk9n18Gym452X38cTw56n/459uPKlc/n048/4/ZGXAbDeFmuz3QFbMW70RP406jcAXPmLm3ny7ufZYveN+fEfDqJD47Kc8c+f8voLE/j57r8r56F9I4stBqcdD4efkH2EZ89dsnOtN+Rn3vfdA16fACeflQ8R98g+wtPkuNNh5qzscU4/Hjosm+Y4vq55c+dx0fFXcdadJ9PQ0MDdVz/AhP9OZteh2wJw52UjeOKu59h0pz5cNeaPfDr7U849/C/zf/6Ua45mw61706FxWa4ddyHX/PoWhl31AEeddwhLfGtxzr4rGwIY8/hYLjj6iiTH2Fbz5s7jz6ffwm+u+SHt2jUw/MbHmfjqFHb5wRYA/Ofvj/Lkff+l/6DeXPHQaXwy+zP+eML183/+W0suTt9vr8MFp9zU7HEH7tGPIQdlb14fHfYCw296vHwH9VXMq9Bh7CJa4BxXhZD0YUS0L1o+BNgkIo6WtBpwBdBINhR6aERMzIdAbyPrzY4AjomI9pIOBk4EPgc+BA6KiDckjScbkm0P3BER6+e/6wSgfUT8UtJmwGVkPc0HgK0jolm/QdJA4ISIGJIvP5AvP1W8TdIewB/JGtpRQP+IGCjpl8CHEXFuK8fal+y8825kvc//A76V/+rTIuLfko4BjgLebnleVtKBwDoRcVq+PD5/7Gn58oXA1Ig4Q9L1wMD8eX0H+EVE/DUfZn8B2CAi5nzR/9miHi6uZne9UaEvSgns1H2TL9+pTrQrnrFW5+6aeN5XGSls1c7dj2/za86i+H1fR8U2spVCUvuIaJqlfDLQNSKOS1xWm+WzpO8HtoyIr/W2L5+A1S8iTl/Yfm5kC9zIFriRLXAjW7BIGtlVj2t7I/vm+Uka2WodLi6nXSWdQvZcTQAOSVvOV5MPC/+C7KM6E7/mwywG/H7RVWVmtghUQSfRjeyXiIgbgRtT1/FNRMTd3/Dny3ou1sysTdzImpmZlci8yr94sRtZMzOrTm5kzczMSiTh51/byo2smZlVpQj3ZM3MzErDPVkzM7MS8exiMzOzEvHEJzMzs9KIuZV/7WI3smZmVp08XGxmZlYinvhkZmZWIv4Ij5mZWWmEe7JmZmal4YlPZmZmpVIFw8X+0narOZKOiIhLU9dRCfxcFPi5KPBzUT4NqQswK4EjUhdQQfxcFPi5KPBzUSZuZM3MzErEjayZmVmJuJG1WuRzTQV+Lgr8XBT4uSgTT3wyMzMrEfdkzczMSsSNrJmZWYm4kTUzMysRN7JmZmYl4kbWqpoym0naU9J38/tKXVe55cf9vKQPJT0mad3UNaXkXGSci/Q8u9iqlqQdgD8BrwGT89WrAD2BH0fE8FS1lZukp4BTgAeB3YHDI2LHtFWl4VwUOBfpuZG1qiVpDLBzRIxvsX514D8R0TtJYQlIeiYi+n3Rcj1xLgqci/T8LTxWzRYDJrWyfjKweJlrSW15SXt+0XJE/DNBTak4FwXORWJuZK2aXQE8KekG4M183arAvsBfk1WVxkhgty9YDqCeXkydiwLnIjEPF1tVyydy7A50A0TWg/l3RPw3aWFlJmlP90oKnIuMc5GeG1mzGuBzbdYa5yI9f4THqpakDpLOlvSypOn5bUy+bvnU9VkazoVVEvdkrWpJuhu4D7g6Iqbk67oAhwDbRsT2CcsrK0kfA2Nb2wRERGxY5pKScS4KnIv03Mha1ZL0SkSs81W31SJJLwG7fNH2iJhQxnKSci4KnIv0PLvYqtkEST8j67G8AyBpJbIey5sL+8Ea9JlfMOdzLgqci8R8Ttaq2feBFYCRkmZImgE8AHQC9klZWAKPpC6ggjgXBc5FYm5krWpFxHsRcVJE9IqITvmtd75uRur6ymxO0x1JxxVvkHRV2atJyLloxrlIzI2sVTVJvSRtK2mZFut3SlVTIlsX3T+4xba6m9ziXMznXCTmRtaqlqRjgduAY4CXJO1RtPmsNFUloy+4X3eci2aci8Q88cmq2VBg44j4UNJqwD8krRYR51N/LygNkjqSvXFuut/0HLRLV1YSzkWBc5GYG1mrZu0i4kOAiBgvaSDZC2oP6u/FtAPwNIXjfqZoW719Ts+5KHAuEnMja9VsiqQ+EfEcQN5zGUJ2gfgN0pZWXhGxWuoaKohzkXMu0vPFKKxqSVoFmNN0VZ8W27aMiLr5+ELeS5sZEe/ny4OA7wDjgYsj4rOE5ZWVc1HgXKTniU9WtSJiUmsvpPm2unkhzd0ELAMgqQ9wMzAR6AP8KWFdZedcNONcJOZG1qqWpA0kjZL0pqRL80kdTdueSFlbAktFxFv5/R8AV0TE74FDgU3TlVV+zkUzzkVibmStmv0Z+CXZebZXgYclrZlvWzxVUYkUT+gZDIwAiIh5acpJyrkocC4S88Qnq2btI2JYfv9cSU8DwyQdSP3NnLxP0k3A20BHsm+hQVJXoN7OuzkXBc5FYp74ZFVL0vPA1k2TOvJ1GwK3AJ0iYoVkxZWZJJFds7crcFNETM7X9wVWjIi7U9ZXTs5FgXORnhtZq1qS9gfGRcSoFuu7A6dHxNA0lVUuSY9FxOap6ygl5+Krq4dcpOJG1mqepAsj4pjUdVQCSc9GRN/UdVQC56LAuSgdT3yyerBl6gIqiN9VFzgXBc5FibiRNTMzKxE3smb1pd6u3Wtt41yUiBtZqwd+ASk4MHUBFcS5KHAuSsSfk7WaIWmZiPiolU3nl72YRCR9wILn194HngJ+GhEvlr+qtJwL5yIlzy62qidpC+BysosQdJe0EXBkRPw4cWllJ+lXwFvAdWQ9tX2BLsArwI8iYmC66srLuShwLtJxI2tVT9LjwPeAfzd9DEHSixGxftrKyk/S4xGxWYt1oyJigKTnI2KjVLWVm3NR4Fyk43OyVhMi4s0Wq+YmKSS9eZL2kdSQ3/Yp2lZ376idi/mci0TcyFoteDMfGgxJS0g6ARiTuqhEDiCbxDIVeCe//wNJSwFHpywsAeeiwLlIxMPFVvUkNZJNYtmO7HzTcOC4iJietDBLyrmwSuBG1qyGSOoMDAVWo+jTAxHxP6lqsvSci3Q8XGxVT9LVkpYvWu4o6YqUNSV0G9ABuBe4s+hWd5yLZpyLRPw5WasFG0bEzKaFiHgv/yqverR0RJyUuogK4VwUOBeJuCdrtaBBUsemBUmdqN83kHdI2iV1ERXCuShwLhLxOVmrepIOAk4B/pGv2hs4MyKuSVdVGvmVfZYBPgU+J5vwExGxXNLCEnAuCpyLdNzIWk2QtC4wmOzFY0RE/DdxSVYBnAtLzY2sVS1Jy0XErHwYcAERMaPcNaUiqVdEvCypX2vbI+KZcteUinNR4Fyk50bWqpakOyJiiKQ3aH7VmqahsDUSlVZ2ki6NiCMk3d/K5oiIwWUvKhHnosC5SM+NrFU1SQJWjYiJqWuxyuFcWKVwI2tVT9LTEbFx6joqgaR2wK4seNGBP6SqKRXnosC5SKdep7NbbRklqX9EPJm6kApwO/AJMBqYl7iW1JyLAuciEfdkrepJ+i+wDjAe+IjCubcNU9aVgqQX6vG4W+NcFDgX6bgna7Vg59QFVJC7JO0QEcNTF1IBnIsC5yIRN7JWtSStCJwK9CQbBvttRMxKW1Vyo4BbJTVQpxcdcC5aVfe5SMXDxVa1JA0DngYeBIYAy0bEIUmLSkzSOOA7wOio0z9u52JBzkU67slaNesSET/P798tyR+sh9eAF+v8hdS5WJBzkYgbWatmyi8Ar3y5XfFyPV3Zp8jbwAOS7iK7Ti1Qdx/VcC4W5Fwk4kbWqlkHsmFBFa1r6rUEUDdX9inyRn5bIr/VI+diQc5FIj4na2ZmViLuyVrVyy+hdwCwRkT8WlJ3svNyTyQurWwknRcRx0u6nebX6wUgInZPUFZSzoVzUQnck7WqJ+nPZFexGRwRvfPzb8Mjon/i0spG0sYR8bSkbVrbHhEjy11Tas6Fc1EJ3JO1WrBZRPST9CxARLwnqa7OO0XE0/ndPhFxfvE2SccB9fhi6lw4F8k1pC7AbBH4PL8AegBI6kz9Xp/14FbWHVLuIiqEc1HgXCTinqzVgguAW4EVJZ0JfA84LW1J5SVpP2B/YHVJ/y7atCwwPU1VyTkXzkVyPidrNUFSL2Bbso9tjIiIMYlLKitJPYDVgd8CJxdt+gB4ISLmJCksMefCuUjNjaxVPUlrApMi4lNJA4ENgb9FxMy0lVUeSY9FxOap6ygH56Lt6ikX5eZzslYLbgHmSuoJXE72zv26tCVVrCVTF1BGzkXb1VMuysqNrNWCefmw157A+RHxv0DXxDVVqnoaunIu2q6eclFWbmStFnyeT/A4CLgjX7d4wnqsMjgXlpwbWasFhwKbA2dGxBuSVgf+nrimSqUv36VmOBdtV0+5KCtPfDKrYZK2BPaPiKPy5fUj4sXEZVlCkhqB6cVfe+dclI4/J2tVT9IbtH5d1nr8thUk9SH7bOQ+ZN+88s+mbfX0QupcgKQBwNnADOAM4BqgEWiQdFBEDIP6ykW5uZG1WrBJ0f0lgb2BTolqSULS2sC+wH5kFxm4kWykalDSwtKq+1wAFwGnkn39333AzhExKv/88PXAsJTF1QMPF1tNkvRwRGyVuo5ykTQPeAg4LCLG5uvG1VOvrS3qMBfPRUSf/P6YiOhdtO3ZiOibrrr64J6sVT1J/YoWG8h6MMsmKieVvch6svdLGgbcQJ1PZnEugObXap7dYpt7WGXgnqxVPUn3Fy3OAcYD50bEK2kqSkfSMsB3yIaNBwNXA7dGxPCkhSXgXICkucBHZG+4lgI+btoELBkR/khTibmRNatRkjqRnYf8fkQMztd1jIj30lZmlca5KB03slb1JHUAfgFsna8aCfw6It5PV1VlkvRMRPT78j2rn3PRdvWUi3LzxSisFlxB9q0i++S3WcCVSSuqXPV0nta5aLt6ykVZeeKT1YI1I2KvouVfSXouWTWVrZ6GrpyLtqunXJSVe7JWC2ZLmv+xjPwqRy1nUlr9cS4sOfdkrRb8CLg6PwcnsqvbHJK0ospVT8OCzkXb1VMuysoTn6xmSFoOICJmpa4lNUkrUvQdoRExMV/fKSJmJCssAeeiwLkoPzeyVrUk/WRh2yPiD+WqpVJI2h34PbAyMBXoAYyJiPWSFlZGzsWCnIt0fE7WqtmyRbcTWizX25V9mpwBDABejYjVgW2BR9KWVHbOxYKci0Tck7Wa4OuwZiQ9FRGbSHoe6BsR8yQ9ERGbpq4tBeci41yk44lPViv8bjEzU1J7si8LuFbSVLJLCtYr5yLjXCTinqzVBF+xJpNfu3g22amgA8i+4uzaiJietLBEnIuMc5GOG1mrWpJGU+ip9ATGNm0CIiI2TFJYYpJ6AGtFxL2SlgbaRcQHqesqF+eidfWei1Q8XGzVbEjqAiqNpKHAEWRfTr4m0A24hGyiS71wLlpwLtJxT9ZqnqTHImLz1HWUQ37ZwE2Bx5sm/EgaHREbpK2s8jgXzkU5+CM8Vg+W/PJdasanEfFZ04KkxfDkny/iXFjJuZG1elBPLyYjJZ0KLCVpe+Bm4PbENVUq58JKzsPFVvPqaYapJAGHAzuQTfS5G7g8/Ie+AOfCuSgHT3yyelAXFz+X1AC8EBHrA5elrqcKOBdWch4utpojaUtJFxetOjBZMWUUEfOA5yV1T11LpZHUmPfmijkXVnLuyVpNkNQH2B/YB3gD+GfTtoh4MVVdCXQFXpL0BPBRvi4iYo+ENZWVpAHA2WRfbXcGcA3QCDRIOigihoFzQZ3lIhU3sla1JK0N7AvsB0wHbiSbZzAoaWFp/arovoCtyJ6fenIRcCrZVY3uA3aOiFGSegHXA8NSFpeIc5GIJz5Z1ZI0j+xarIdFxNh83biIWCNtZWm11quPiAvTVlU+kp6LiD75/TER0btoW91+YUC95yIV92Stmu1F1pO9X9Iw4AbqZDJLS+7VNzOv6P7sFtvqqlfhXKTnnqxVvfzi598heyEZDFwN3BoRw5MWVkbu1RdImkt23lHAUsDHTZuAJSNi8VS1lZtzkZ5nF1vVi4iPIuLaiBgCrAI8B5zctF1Sx2TFlc9ewBSyXv1lkralTnv1EdEuIpaLiGUjYrH8ftPy/AbWubBycE/Wal6dXXSg7nv1beVcOBfl4EbWal69TnaR1AnYG/h+RAxOXU+lcS6ci3JwI2s1r556LNZ2zoWVg8/JmpmZlYgbWasHnuhhrXEurOQ8XGw1Q9KKFH1HaERMzNd3iogZyQqzpJwLS8k9Wat6knaX9BrZVWxGAuOBu5q2+4W0PjkXVgncyFotOAMYALwaEasD2wKPpC3JKoBzYcm5kbVa8HlETCf7lpWGiLgf6JO6KEvOubDkfO1iqwUzJbUnu3zctZKmAnMS12TpOReWnCc+WdXLr2Yzm2xk5gCyrzi7Nu/FWJ1yLqwSuJG1miCpB7BWRNwraWmgXUR8kLouS8u5sNR8TtaqnqShwD+Av+SrugH/SleRVQLnwiqBG1mrBUcBWwKzACLiNWDFpBVZJXAuLDk3slYLPo2Iz5oWJC1GnX05t7XKubDk3MhaLRgp6VRgKUnbAzcDtyeuydJzLiw5T3yyqidJwOHADmTXo70buDwc7rrmXFglcCNrVU1SA/BCRKyfuharHM6FVQoPF1tVi4h5wPOSuqeuxSqHc2GVwld8slrQFXhJ0hPAR/m6iIg9EtZk6TkXlpyHi63qSdqmeBHYCtgvItZLVJJVAOfCKoGHi63qRcRI4H1gV+Aqsm9buSRlTZaec2GVwMPFVrUkrQ3sC+wHTAduJBudGZS0MEvKubBK4uFiq1qS5pF9w8phETE2XzcuItZIW5ml5FxYJfFwsVWzvYApwP2SLpO0Ldm5N6tvzoVVDPdkrerlX2n2HbLhwcHA1cCtETE8aWGWlHNhlcCNrNUUSZ2AvYHvR8Tg1PVYZXAuLBU3smZmZiXic7JmZmYl4kbWzMysRNzImpmZlYgbWTMzsxL5f2hY3KEx78zyAAAAAElFTkSuQmCC\n",
      "text/plain": [
       "<Figure size 432x288 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "x = experiments3\n",
    "y = outcomes3\n",
    "\n",
    "preferred_order = ['SLR scenario (X1)',\n",
    "                   'Transient storm surge (X2)',\n",
    "                   'Implementation time (X3)',\n",
    "                   'Management strategy (L1)',\n",
    "                   'Housing market (R1)']                   \n",
    "\n",
    "fs = feature_scoring.get_feature_scores_all(x, y)\n",
    "fs = fs.reindex(preferred_order)\n",
    "sns.heatmap(fs, cmap='viridis', annot=True)\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Timing set to 2300"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 55,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "import numpy as np\n",
    "outcomes4=outcomes2.copy()\n",
    "for key, values in outcomes2.items():\n",
    "    if key in [\"Area_A_timing_SETP\",\"Area_B_timing_SETP\"]:\n",
    "        outcomes4[key] = np.array([2300 if y == 9999 else y for y in values])\n",
    "        \n",
    "x = experiments3\n",
    "y = outcomes4\n",
    "\n",
    "preferred_order = ['SLR scenario (X1)',\n",
    "                   'Transient storm surge (X2)',\n",
    "                   'Implementation time (X3)',\n",
    "                   'Management strategy (L1)',\n",
    "                   'Housing market (R1)']                   \n",
    "\n",
    "fig, ax = plt.subplots()\n",
    "\n",
    "fs = feature_scoring.get_feature_scores_all(x, y)\n",
    "fs = fs.reindex(preferred_order)\n",
    "ax = sns.heatmap(fs, cmap='viridis', annot=True,fmt='.2f')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 58,
   "metadata": {},
   "outputs": [],
   "source": [
    "from datetime import datetime\n",
    "today = datetime.date(datetime.now())\n",
    "date = \"{}_{}_{}\".format(today.year,today.month,today.day)\n",
    "target = Path('Visualisations') / date\n",
    "\n",
    "if not Path.exists(target): Path.mkdir(target)\n",
    "\n",
    "fig.savefig((target / \"FeatureScoring_regression_EMA_2021_2_23.png\"),bbox_inches=\"tight\",dpi=200)"
   ]
  }
 ],
 "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": "3.8.2"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 4
}