main_candidate_search_YS_quaternary.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "id": "6d917da6",
   "metadata": {},
   "source": [
    "## Searching best candidate in quarternary alloy compositions for optimal YS value"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "id": "0fa4e4d8",
   "metadata": {},
   "outputs": [],
   "source": [
    "import pandas as pd \n",
    "import numpy as np\n",
    "from cbfv.composition import generate_features\n",
    "import matplotlib.pyplot as plt \n",
    "from sklearn.metrics import r2_score,mean_squared_error,mean_absolute_error\n",
    "from sklearn.preprocessing import StandardScaler\n",
    "from sklearn.preprocessing import normalize\n",
    "import seaborn as sns\n",
    "from scipy.stats import norm\n",
    "from sklearn.model_selection import cross_val_score\n",
    "from utils import bootstrap_estimator,Expected_Improvement\n",
    "rng_seed = 20"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "1b84f9b8",
   "metadata": {},
   "source": [
    "### Reading train and test data .csv files and storing in dataframe"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "id": "93ab2959",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "size of train data 115\n",
      "size of test data 25\n"
     ]
    }
   ],
   "source": [
    "df_train = pd.read_csv('data/model_input/train_yieldstrength.csv')\n",
    "df_test = pd.read_csv('data/model_input/test_yieldstrength.csv')\n",
    "print('size of train data', len(df_train))\n",
    "print('size of test data',len(df_test))\n"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "7ff6e219",
   "metadata": {},
   "source": [
    "### Featurization using CBFV and considering only relevant features "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "id": "8b50cb04",
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Processing Input Data: 100%|███████████████| 115/115 [00:00<00:00, 11404.03it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\tFeaturizing Compositions...\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Assigning Features...: 100%|████████████████| 115/115 [00:00<00:00, 5894.98it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\tCreating Pandas Objects...\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Processing Input Data: 100%|██████████████████| 25/25 [00:00<00:00, 7219.61it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\tFeaturizing Compositions...\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Assigning Features...: 100%|██████████████████| 25/25 [00:00<00:00, 4235.81it/s]"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\tCreating Pandas Objects...\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\n"
     ]
    }
   ],
   "source": [
    "# FEATURIZATION \n",
    "X_train_unscaled, y_train, formulae_train, skipped_train = generate_features(df_train,elem_prop='f3_revised',drop_duplicates=False,extend_features=True)\n",
    "X_test_unscaled, y_test, formulae_test, skipped_test = generate_features(df_test,elem_prop='f3_revised',drop_duplicates=False,extend_features=True)\n"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "143fd956",
   "metadata": {},
   "source": [
    "### Selecting the features in the form of weighted average for each alloy\n",
    "$$\n",
    "  \\sum_{i=1}^{n} c_if_i \\\n",
    "$$\n",
    "  where $c_i$ represents the concentration of each element i (Nb, W, Co etc) in the atom percentage present in   the alloy and $f_i$ represents its elemental property (such as atomic radius, melting point etc.). \n",
    "\n",
    "\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "id": "970de825",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "shape of the features space: (115, 15)\n"
     ]
    }
   ],
   "source": [
    "X_train_unscaled_avg = X_train_unscaled[['avg_Atomic_Radius','avg_Pauling_Electronegativity','avg_number_of_valence_electrons','avg_Cohesive_energy_ev_atom',\n",
    "            'avg_Bulk_modulus_RT_Gpa','avg_Elastic_modulus_RT_Gpa','avg_Shear_modulus_RT_Gpa','avg_Melting_point_(K)','avg_rate_shear_mod_Mpa_perK',\n",
    "            'avg_Solid_Solubility_atpct','avg_lattice_constant_A','avg_BEC_percm3','avg_Av.Valence_bond_strength_ev','avg_EngelZ_e/a','T']]\n",
    "\n",
    "X_test_unscaled_avg = X_test_unscaled[['avg_Atomic_Radius','avg_Pauling_Electronegativity','avg_number_of_valence_electrons','avg_Cohesive_energy_ev_atom',\n",
    "            'avg_Bulk_modulus_RT_Gpa','avg_Elastic_modulus_RT_Gpa','avg_Shear_modulus_RT_Gpa','avg_Melting_point_(K)','avg_rate_shear_mod_Mpa_perK',\n",
    "            'avg_Solid_Solubility_atpct','avg_lattice_constant_A','avg_BEC_percm3','avg_Av.Valence_bond_strength_ev','avg_EngelZ_e/a','T']]\n",
    "\n",
    "\n",
    "\n",
    "train_data_avg = X_train_unscaled_avg.join(y_train)\n",
    "test_data_avg = X_test_unscaled_avg.join(y_test)\n",
    "total_data_avg = pd.concat([train_data_avg,test_data_avg])\n",
    "total_data_avg.shape\n",
    "\n",
    "# Correlation Plot \n",
    "# plt.figure(figsize=(20,20))\n",
    "# cmap = sns.color_palette(\"coolwarm\", as_cmap=True)\n",
    "# sns.heatmap(total_data_avg.corr(),cmap=cmap)\n",
    "# plt.show()\n",
    "\n",
    "# train_data_avg.corr()\n",
    "\n",
    "print('shape of the features space:',X_train_unscaled_avg.shape)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "dfe37359",
   "metadata": {},
   "source": [
    "### Scaling using standard scaler "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "id": "8f5e09d8",
   "metadata": {},
   "outputs": [],
   "source": [
    "# Instantiate the standard scaler\n",
    "scaler = StandardScaler()\n",
    "# zero mean and unit variance\n",
    "X_train = scaler.fit_transform(X_train_unscaled_avg)\n",
    "X_test = scaler.transform(X_test_unscaled_avg)\n",
    "\n"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "5c2394a5",
   "metadata": {},
   "source": [
    "### Machine Learning Models Performance Evaluation Helper Functions "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "id": "6600d95d",
   "metadata": {},
   "outputs": [],
   "source": [
    "################# EVALUATION MATRIX #######################################\n",
    "def model_evaluation(y_actual,y_predict, label = None,model='None'):\n",
    "    df = pd.DataFrame([{'mae':mean_absolute_error(y_actual,y_predict),\n",
    "                        'rmse':mean_squared_error(y_actual,y_predict,squared=False),\n",
    "                        'r2score':r2_score(y_actual,y_predict),\n",
    "                        'model':model\n",
    "                        }])\n",
    "    if label:\n",
    "        df.index = [label]\n",
    "        \n",
    "    return df\n",
    "\n",
    "def model_evaluation_result(y_Train,y_train_predict,y_Test,y_test_predict,model='None'):\n",
    "    result = pd.concat([model_evaluation(y_Train,y_train_predict,'Train',model),\n",
    "                    model_evaluation(y_Test,y_test_predict, 'Test',model)])\n",
    "    return result\n",
    "\n",
    "\n",
    "def plot_model_performance(y_train,y_train_predict,y_test,y_test_predict,r2_train,r2_test):\n",
    "    plt.figure(figsize=(8,8))\n",
    "    plt.scatter(y_train,y_train_predict, s=160,alpha=0.7, label='Training Data ($r^2$ ={0:.3f})'.format(r2_train))\n",
    "    plt.scatter(y_test,y_test_predict,s=160, alpha=0.7, label='Test Data ($r^2$ ={0:.3f})'.format(r2_test))\n",
    "    # plt.plot(y_test,y_test,'--',dashes=(5, 6),linewidth=2,c='grey',label = 'Ideal Fit')\n",
    "    plt.axline((0, 0), (1200, 1200), linewidth=2, color='k',linestyle='dashed',label='ideal fit')\n",
    "    plt.xlabel('Actual Yield Strength (MPa) ', size=16)\n",
    "    plt.ylabel('Predicted Yield Strength (MPa)',size=16)\n",
    "    plt.tick_params(labelsize=14)\n",
    "    plt.xlim((0,1000))\n",
    "    plt.ylim((0,1000))\n",
    "    plt.legend(prop={'size': 14})\n",
    "    plt.minorticks_on()\n",
    "    plt.tick_params(labelsize=16)\n",
    "    plt.tick_params(direction='in',which='major', length=15, bottom=True, top=True, left=True, right=True,colors='k')\n",
    "    plt.tick_params(direction='in',which='minor', length=5, bottom=True, top=True, left=True, right=True,colors='k')\n"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "042897fd",
   "metadata": {},
   "source": [
    "### Gradient Boost Regressor with grid search group kfold"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "id": "1a66a7e6",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "GradientBoostingRegressor(criterion='mse', max_features='auto',\n",
      "                          min_samples_leaf=4, min_samples_split=5,\n",
      "                          n_estimators=200, random_state=20)\n",
      "r2score_train: 0.9952670043760236\n",
      "r2score_test: 0.7687336260644045\n"
     ]
    }
   ],
   "source": [
    "from sklearn.ensemble import GradientBoostingRegressor\n",
    "from sklearn.model_selection import GroupKFold\n",
    "from sklearn.model_selection import GridSearchCV\n",
    "\n",
    "base_model = GradientBoostingRegressor(random_state=rng_seed, criterion='mse')\n",
    "\n",
    "\n",
    "param_grid = { 'n_estimators' : [50,100,200],\n",
    "                'max_depth': [3],\n",
    "                'max_features': ['auto', 'sqrt', 'log2'],\n",
    "                'min_samples_split': [5,6,7],\n",
    "                'min_samples_leaf':[4,5,6],\n",
    "             }\n",
    "\n",
    "# Instantiate GroupKfold with no of splits = 5\n",
    "gkf=GroupKFold(n_splits=5)\n",
    "\n",
    "\n",
    "gbr_gs= GridSearchCV(\n",
    "                     base_model,\n",
    "                     param_grid,\n",
    "                     scoring='neg_mean_squared_error',\n",
    "                     cv=gkf\n",
    "                    )\n",
    "\n",
    "# Fit training Data \n",
    "gbr_gs.fit(X_train,y_train,groups=formulae_train)\n",
    "\n",
    "# Best Estimator\n",
    "print(gbr_gs.best_estimator_)\n",
    "\n",
    "gbr_gs_best=gbr_gs.best_estimator_\n",
    "\n",
    "\n",
    "gbr_gs_best.fit(X_train,y_train)\n",
    "\n",
    "# predict y_train and y_test\n",
    "y_train_predict=gbr_gs_best.predict(X_train)\n",
    "y_test_predict=gbr_gs_best.predict(X_test)\n",
    "\n",
    "r2_train = r2_score(y_train,y_train_predict)\n",
    "r2_test = r2_score(y_test,y_test_predict)\n",
    "#print(gbr_gs_best)\n",
    "\n",
    "print('r2score_train:',r2_score(y_train,y_train_predict))\n",
    "print('r2score_test:',r2_score(y_test,y_test_predict))\n"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "db897afb",
   "metadata": {},
   "source": [
    "### Performance Evaluation Gradient Boost Regressor"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "id": "635d673a",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "             mae        rmse   r2score   model\n",
      "Train   7.718396   10.050712  0.995267  gbr_GS\n",
      "Test   73.427359  107.466433  0.768734  gbr_GS\n",
      "****************************************************************************\n",
      "                       feature_name  importance\n",
      "14                                T    0.778554\n",
      "11                   avg_BEC_percm3    0.105806\n",
      "1     avg_Pauling_Electronegativity    0.029670\n",
      "5        avg_Elastic_modulus_RT_Gpa    0.025325\n",
      "8       avg_rate_shear_mod_Mpa_perK    0.022040\n",
      "10           avg_lattice_constant_A    0.019107\n",
      "0                 avg_Atomic_Radius    0.010403\n",
      "2   avg_number_of_valence_electrons    0.002952\n",
      "3       avg_Cohesive_energy_ev_atom    0.002826\n",
      "13                   avg_EngelZ_e/a    0.001504\n"
     ]
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAiAAAAHsCAYAAADrSxZOAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjQuMiwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy8rg+JYAAAACXBIWXMAAAsTAAALEwEAmpwYAAC4lUlEQVR4nOzdd3xUVdrA8d+ZPumBEAhFmihVECKKHQRBIKIiih0buqtr77piZddd3bWvgqwFG7IKGlCKINhFqhTxlSYogRQgdTLtnvePOwnpGSDJpDzfj/NJcu+ZO89MMPPMKc9RWmuEEEIIIRqSJdIBCCGEEKLlkQRECCGEEA1OEhAhhBBCNDhJQIQQQgjR4CQBEUIIIUSDkwRECCGEEA2uwRMQpVRHpdQLSqnvlFJFSimtlOpSRbtEpdRrSqlspVShUupzpVS/Ktq5lFL/VEplKKU8oeueXkU7i1LqfqXUDqVUsVJqnVJqfD09TSGEEELUIBI9IEcDFwH7ga+qaqCUUsAnwCjgL8B4wA58oZTqWKH5DOB64GFgLJABLFRKDajQ7nHgEeBF4Bzge2C2Umr0ET8jIYQQQhwS1dCFyJRSFq21Efr+OmA60FVrvaNMm3HAXGCY1vqL0LF4YDvwttb6ltCx/sBa4Bqt9euhYzZgI/CL1vrc0LFkYBfwd631lDKPswRoo7U+rj6fsxBCCCHKa/AekJLkoxbnArtLko/Q/XKBdGBchXZ+YFaZdgHgfWCkUsoZOjwScABvV3ict4F+Sqmuh/o8hBBCCHH4Gusk1D7AhiqObwSOUkrFlGm3XWtdVEU7B+ZwT0k7L7ClinYAvY84YiGEEEKEzRbpAKrRCthRxfF9oa+JQEGo3f4a2rUq8/WArjzeVLFdKZfLpa1W6yGEfFBSUhJt2rSp9nxubi7x8fGHde2aNLXrZmVl1fg6HYmm9lrU13Wh/l7npvhaNLV/y03xtWhqrzE0vdeitutmZWWRnZ0d9vW01hiGgdVqpaioyKu1dtVFnGE9cKRuwHWABrpUOP4r8F4V7a8Pte8U+nkx8F0V7UaE2p0W+nk6kFFFux6hdldUPBcVFaXry/XXXy/X1VoPGjSoXq6rddN7LerrulrX3+vcFF+LpvZvuSm+Fk3tNda66b0WdXnd33//XR977LHa6XTqL774QgOFuoFygMY6BLOPKnolMHs+4GCvR23t9pX5mhhaXVNTOyGEEKJF2LlzJ2eccQa//PILxx57LH369GnQx2+sCchGzHkbFfUGdmqtC8q066qUiqqinY+Dcz42Ak6gexXtADYdccRCCCFEE7F9+3bOOOMMtm7dysCBA1m6dGm9DXFVp7EmIJ8AHZRSZ5QcUErFAWmhc2Xb2YEJZdrZgIuBRVprb+jwAsyE5LIKj3M5sEFrvb3On4EQQgjRCP3666+cfvrp7Nixg8GDB7NkyRJat27d4HFEZBKqUurC0LeDQl/PUUplAVla6+WYicV3wNtKqbsxh1zuBxTwj5LraK3XKqVmAc8qpeyYdUL+BHSlTLKhtc5USv0buF8plQ+sxkxShlF+WW+DSEtLk+vWs6b2WshrXP/Xre9r14em+Fo0tdcYmt5rcSTXDQQCjB49mt9//51TTjmFTz/9lLi4uDqMLnwNXogMQClV3YMu11qfGWrTCngaOA9wYSYkd2it11W4lht4ErgUSADWAfdqrZdVaGfFTGKuB9oBvwCPaa3/V1Ug0dHRurCw8NCfnAhbamoqK1eujHQYzZ68zvVPXuP6J69x3VmyZAlPP/00s2fPJiYmptw5pVSR1jq6IeKISALSFNSWgOTl5ZGZmYnf72/AqJqXjIwMUlJSIh1Gs9dSXufo6Gg6duyIxdLwI8vy5lj/5DU+MkVFRURFHZwuqbWm8rqMhk1AGmsdkEYtLy+PvXv30qFDB9xud5W/RFE7rTW9evWKdBjNXkt4nQ3D4I8//iA7O5vk5ORIhyNEo7Jy5UrS0tKYMWMGo0eb2581hvetxjoJtVHLzMykQ4cOREVFNYpfohAtncVioW3btuTm5kY6FCEale+//56zzjqLPXv28NZbb0U6nHIkATkMfr8ft9sd6TCEEGXY7XYCgUCkwxCi0fjqq68YMWIEeXl5TJgwgZkzZ0Y6pHIkATlM0vMhROMi/08KcdDSpUsZNWoUBQUFXHrppbz77rvY7fZIh1WOJCBCCCFEM7Jo0SLGjBlDUVERkyZN4q233sJma3xTPiUBEUIIIZoRl8uFUorJkyczY8YMDndj1fomCUgLMmnSJMaOHVtjm7FjxzJp0qQ6fdxHHnmEvn371thm8+bNDBkyBJfLRZcuXQCzS/1//6uyTIsQQohqnH766axatYpXXnklIsvSw9X4+mQaiaSkpEiHUOeee+45GlPdl7L7Djz00ENERUWxefNmoqPNJegZGRkkJpr7Be7YsYOuXbvy448/kpqaGpF4m6qG3t+hJZo8eXKkQ2j25DWu2QcffIDL5eLcc88FOJKl91l1FlQtJAGpRnP8ox0fHx/pEMop+xpv2bKFcePGlfZ+ALRr1y4CUTU/zfHfcmMjb471T17j6r399ttcddVV2Gw2fvrpJ4499tgjuVx2XcVVm8bbNyPqXMUhmJIJSjExMbRt25apU6dWuo/P5+Pee++lY8eOREdHc8IJJ7Bw4cLS88FgkGuvvZauXbvidrvp0aMH//jHPzAMI+y4lFKsW7eOxx57DKUUjzzySOnxkiGYrl27AnDCCSeglOLMM888jFdACCGal9dff50rr7wSwzB44IEHOOaYYyIdUtgkAalDSqlqb9OmTSttN23atBrbljVo0KCw2h2Ou+66i8WLF/Phhx+yZMkS1qxZw5dfflmuzdVXX83y5ct59913Wb9+PVdddRVpaWmsW2duyWMYBh06dOCDDz7g559/5sknn2Tq1Km8/vrrYceRkZHBsccey5133klGRgZ33XVXpTYrVqwAYMGCBWRkZPDRRx8dwTMXQoim79VXX+Waa65Ba83UqVOZMmVKk1qOLkMwLVRBQQEzZszgv//9LyNHjgTMTLpjx46lbbZu3cp7773Hjh07OOqoowC4+eab+fzzz3n11Vd5+eWXsdvtPPbYY6X36dKlC6tXr+a9997j2muvDSuWdu3aYbPZiImJqXbYpWQYoXXr1jI0I4Ro8V544QVuueUWAJ5++mnuvPPOCEd06CQBqUPhTvCcPHly2OOZq1atOpKQqrV161Z8Ph9DhgwpPRYTE0O/fv1Kf169ejVaa3r37l3uvl6vl2HDhpX+/Morr/Daa6/x22+/4fF48Pv9dO7cuV7iFkKIli4jI4P7778fgOeff56//OUvEY7o8EgC0kKFkywZhoFSih9//LFSBb2SUvSzZs3itttu4+mnn+bkk08mLi6Ol156iTlz5tRL3EII0dKlpKQwb948fv31V66//vpIh3PYJAGpRm5ubqVeirS0NNLS0iIUUd06+uijsdvtfP/993Tr1g2AwsJCNmzYQPfu3QE4/vjj0VqzZ88ehg4dWuV1vv76a0488URuvvnm0mNbt26t83gdDgdgTnoVQoiWRmvN5s2bS5fXnnnmmUc8GT89PZ309PSKhxtsuaQkINWIj48vN3G0uYmJieHaa6/l3nvvpU2bNrRv357HHnus3Bv8Mcccw2WXXcakSZN45plnGDhwIPv27WPZsmV069aNCy64gGOOOYY33niDzz77jKOPPpr333+f5cuXl9bvqCvJycm43W4WLlxIly5dcLlcjW5ZsRBC1AetNQ888ADPPPMMH374YZ19EK7qQ/X06dMbbEtpWQXTgj399NMMHTqU888/n6FDh9K3b19OP/30cm1ef/11rr76au655x569uzJ2LFj+fLLL0vneNxwww1cdNFFXHrppZxwwgns2LGjXiZD2Ww2nn/+eV577TXat2/PuHHj6vwxhBCisdFac+edd/L3v/8drTXFxcWRDqnOqMZUGbMxSU1N1StXrqzy3M8//3wkVeaEEPVE/t8UzYlhGNxyyy289NJL2O12PvjgA84777x6fUyl1CqtdYOUm5YhGCGEEKKRMQyDG2+8kenTp+N0Ovnwww8ZM2ZMpMOqU5KACCGEEI3MLbfcwvTp03G5XHz88cecffbZkQ6pzskcECGEEKKRmThxIsnJyXz66afNMvkA6QERQgghGgWtdWkp9VNPPZXt27cTFRUV4ajqj/SACCGEEBHm9XqZMGFCuSKOzTn5AElAhBBCiIjyeDycf/75fPjhh/zpT3+isLAw0iE1CBmCEUIIISKkqKiIcePG8fnnn5OUlMSCBQuIjo6OdFgNQhIQIYQQIgIKCgpIS0tj2bJltG3bls8//5y+fftGOqwGIwmIEEII0cDy8vIYPXo033zzDSkpKSxdupSePXtGOqwGJXNAhBBCiAa2detW1q1bR6dOnfjyyy9bXPIBkoAI0STs2rWLM888k969e9O/f38++uijSIckhDgCxx9/PAsXLmT58uUcffTRkQ4nImQIRogmwGaz8eyzzzJgwAAyMzMZNGgQo0aNavbL9IRoTrKysvjxxx8ZPXo0ACeffHKEI4os6QERdWLSpEmMHTu23u/T3Ozfv5+2bduydevWGtulpKQwYMAAAJKTk0lMTCQ7O7sBImwcLrzwQv71r39FOgwhDtuePXs488wzGTduHAsWLIh0OI2CJCAR5A0E+Tkjj9U79/NzRh7eQLBeH08pVeNt0qRJh33t5557jrfffrve73M4Jk2aVPoc7XY7ycnJDB06lJdeegm/33/I1zvzzDO5+eab6yS2qVOnMnr0aLp37x72fVauXInf76dTp051EkOJl19+ma5du+JyuRg0aBBfffVVje3z8/O57bbb6Ny5M263m5NPPpkff/zxkNs88sgjlf4ttmvXrlybKVOm8MQTT5Cbm1s3T1aIBvTHH39wxhlnsGnTJo499liOP/74SIfUKMgQTATkFHiZvz6Deesy8AUNFKABh9XC2P4pjOmXQusYZ50/bkZGRun38+bN4/rrry93zO12V7qPz+fD4XDUeu34+PhDjudw7nO4hg8fzsyZMwkGg2RlZbF06VKmTJnCzJkzWbJkSUTW3RcVFfHaa6+Rnp5ea9tAIIDNZiMnJ4crr7ySGTNmlJZsrguzZs3i1ltv5eWXX+bUU0/l5Zdf5pxzzmHTpk0cddRRVd7nuuuu46effuLNN9+kY8eOvP322wwfPpxNmzbRoUOHsNsAHHvssSxbtqz0Z6vVWu6x+vXrR7du3Xj77be56aab6ux5C1Hfdu7cybBhw9i6dSv9+/dn8eLFtGnTJtJhNQ5aa7lVcRs0aJCuzqZNm6o9V5ttWQX60unf6bHPf6kvn/69vub1FaW3y6d/r8c+/6W+dPp3entWwWE/Rjhmz56tzV9/eWeccYa+8cYb9Z133qmTkpJ0amqq1lrrzz77TJ966qk6ISFBJyYm6rPPPrvc63DVVVfpMWPGlLvOn/70J33//ffr1q1b6zZt2ug777xTB4PBI7pPQUGBvuKKK3R0dLROTk7WU6dO1WPGjNFXXXVVtc+14uOUWL9+vbbb7frhhx8uPRbO88TMF0tv27dvr/V+VZk9e7Zu1aqVNgyj3PFdu3ZpQL/33nt66NCh2ul06hkzZuji4mJ92mmn6bfeeqvG6x6OwYMH6+uuu67csaOPPlrfd999VbYvKirSVqtVz507t9zxgQMH6gcffDDsNlprPWXKFN2nT59aY3z00Uf1KaecUmObI/l/U4i6tnXrVt25c2cN6NTUVJ2TkxPpkGoFrNQN9D4rQzANKKfAy0Nz1+P1GyTHunDYyr/8DpuF5FgXXr/Bg3PXk1PgjUicb7/9NlprvvrqK9566y0ACgsLue2221ixYgXLli0jPj6etLQ0fD5ftdd55513sNlsfPvtt7z44os8++yzzJo1q8bHru0+d955J8uXL2fOnDksXbqUdevW1TpUUJ2+ffsyatQoPvzww9JjtT3P5557jiFDhnD11VeTkZFBRkYGnTp1OqzX56uvvmLQoEGVejLWrl0LwFNPPcVdd93Fxo0bGTduHJMmTWLYsGFcccUVtT63qVOnEhMTU+Ot5HXz+XysWrWq0o6bZ599Nt9++22V1w8EAgSDQVwuV7njbrebr7/+Ouw2JbZt20aHDh3o2rUrEydOZNu2bZUec/DgwaxYsQKPx1Pr8xci0gzD4Nxzz+W3337jpJNO4vPPP6dVq1aRDqtRkSGYBjR/fQZ5Hj/Jsa4a28W77WTmFzN/fQZXDunSMMGV0bVrV5555plyx8aPH1/u59dff524uDhWrFjBqaeeWuV1evfuzWOPPQbAMcccw/Tp01myZAmXXHJJtY9d030KCgr473//y1tvvcWIESMAmDFjBh07djzs59q7d28+//zzsJ9nfHw8DoeDqKiocvMUDuf1+e2330hJSal0fN26dbhcLmbPnl26PO/rr79m1qxZHHfcccydOxeAmTNn0q9fvyqvfeONN3LRRRfV+NxLhkCys7MJBoO0bdu23PmSyoxViY2NZciQITzxxBP07duXdu3a8d577/Hdd9+VxhxOG4ATTzyRN954g549e5KZmckTTzzBySefzMaNG2ndunVpu/bt2+P3+9m9e/chzZkRIhIsFgvTp0/niSee4P333yc2NjbSITU6koBUIzc3l8mTJ5c7lpaWRlpa2mFdr9gfZN66DBLctc+nAEhwO5i3LoOLT+iE02at/Q51aNCgQZWObd26lb/+9a/88MMPZGVlYRgGhmGwc+fOaq9z3HHHlfu5ffv2ZGZm1vjYNd1n69at+P1+Bg8eXHo+Ojr6iEoX6zLbX5c8xqE+z8O9n8fjqfSmD2YPyOjRo8u9SZ966qkYhhH282rVqtUhf9qq2BNT8bWpaObMmVxzzTV07NgRq9XKwIEDueSSS1i9evUhtTnnnHPKXfekk06iW7duvPnmm9xxxx2lx0vmKEkPiGjMCgoKiImJAWDIkCHMnz8/whFVLz09vao5aA02OU8SkGrEx8czbdq0Orve9uxCfEGDWFd4L7nDZiHfG2BbViG9UuLqLI5wVDUhMy0tjQ4dOvDqq6/SoUMHbDYbvXv3rnGIwW63l/tZKVXrm2hN9zGHJyu/UR6JTZs20a1bt9KfD+d5Hu79kpKS2L9/f6Xj69atO+JVNlOnTmXq1Kk1tvnss8847bTTSEpKwmq1smfPnnLnMzMzq0yQSnTv3p3ly5dTWFhIXl4eKSkpXHzxxXTt2vWQ2lQUExNDnz59+PXXX8sd37dvH4BM4BON1po1azjnnHN48cUXufDCCyMdTq2q+lA9ffr0BltqJnNAGojHH+RQ3zYVZs9JpOXk5PDzzz/zwAMPMHz4cHr16kV+fj6BQKBB4zj66KOx2+2sWLGi9FhRUREbNmw4rOtt2LCBBQsWlP6hCPd5OhwOgsGDv5fDfX2OP/54Nm3aVO5YYWEhW7duZeDAgYf1nErceOONrF27tsZbampq6fMZNGgQixcvLneNxYsXh1UoKTo6mpSUFPbv38/ChQsZN27cYbUpUVxczObNmysNT23YsIH27dvXmBQJESk//vgjw4YNY+/evbz77rulH5hE9aQHpIG47VYO9Z+jBlz2hh1+qUpiYiJJSUlMnz6dTp068ccff3D33XdjszXsP5+YmBiuueYa7r33XpKSkkhJSeGJJ57AMIxae0W8Xi979uzBMAyysrJYsmQJU6dOZdCgQdx1111A+M+zS5curFixgh07dhATE0OrVq0O6/UZOXIk9957Lzk5OaVzHX766SeA0qJjh+tQh2DuuOMOrrjiCgYPHswpp5zCK6+8wu7du7nxxhtL27z44ou8+OKLbN68GYCFCxdiGAY9e/Zky5Yt3H333Rx77LFcffXVpfcJp81dd91FWloaRx11FJmZmTz++OMUFhZy1VVXlYvxq6++YtSoUYf7kghRb7799lvOOecc8vLyOP/883n//ffrtKe2uZIekAbSrU00DqsFXyC8cXxfwMBhtdCtTcPXp6jIYrEwa9YsfvrpJ/r27ctNN93E448/jtNZ97VKavP0009z2mmnce655zJ06FCOO+44UlNTK620qOjzzz8nJSWFo446irPOOotPPvmEKVOm8OWXX5YOOYX7PO+66y4cDge9e/emTZs27Ny587Ben379+jF48GDef//90mPr1q2jR48epWPIDeXiiy/m2Wef5YknnmDAgAF8/fXXfPrpp3Tu3Lm0TXZ2Nr/88kvpz7m5udx888307NmTK6+8klNPPZVFixaVG0YLp83vv//OJZdcwrHHHssFF1yA0+nk+++/L/fYxcXFzJkzh+uvv76eXwkhDs2XX37J2WefTV5eHhdddBGzZs0Kq3aSACXdRFVLTU3VK1eurPLczz//TK9evQ75mm99t4PZK3fVugoGIDO/mAmpnSKyCqYp8Xq9dO7cmbvvvps777wz0uEcsgULFnDrrbeyadOmSsW3xEEvvfQSH3/8MYsWLaqx3eH+vynE4Vi6dCljx47F4/Fw+eWX8/rrrzd4z3BdU0qt0lqnNsRjSQ9IAxrTL4U4t51cT83lv3M9fuLcdsb0q7xEs6Vbs2YN7777Llu2bGHNmjVcddVV5Ofnc/HFF0c6tMMyatQobrrpJn7//fdIh9Ko2e12XnjhhUiHIUQ5MTEx2Gw2rr76at54440mn3w0NHm1GlDrGCdPntePB+euJzO/mAS3o1wxMl/A4IDHR5zbzpPn9auXcuzNwb/+9S9++eUXbDYbAwYM4MsvvzyiWiCRdsstt0Q6hEav4pJ4IRqDwYMHs2rVKrp3747FIp/nD5UkIA2sS1I0z088vnQvmHxvoNxeMBNSO9XbXjDNwfHHH091Q2NCCFHfPvroI3w+HxMnTgSgR48eEY6o6ZIEJAJaxzi5ckgXLj6hE9uyCin2B3HZrXRrE93gRceEEEKE5/333+fyyy8HoFevXvTv3z/CETVtkoBEkNNmbfAiY0IIIQ7dW2+9xdVXX41hGDzwwAOVqjaLQyeDVkIIIUQNZsyYwaRJkzAMg0cffZQnnnhC6nzUAekBEUIIIarxn//8hz//+c+AucXB/fffH+GImg9JQIQQQogqZGdn88ADDwDwzDPPlNscURw5SUCEEEKIKiQlJbFgwQLWrFlTblsCUTckARFCCCHKWL9+Pf369QPgxBNP5MQTT4xwRM2TTEIVQgghAK01Dz/8MAMGDCi3R5OoH9IDIoQQosXTWnPffffxj3/8A6vVimGEt3GoOHySgAghhGjRtNbcfvvtPPfcc9hsNt577z0uvPDCSIfV7EkCIoQQosUyDIObb76Z//znP9jtdmbPns24ceMiHVaLIAmIEEKIFuuee+7hP//5D06nkzlz5nDOOedEOqQWQyahCnGYdu3axZlnnknv3r3p378/H330UaRDEkIcoksvvZSUlBTmzZsnyUcDkx6QauTm5lbaAjwtLY20tLQIRSQaG5vNxrPPPsuAAQPIzMxk0KBBjBo1iqioqEiHJoSogda6tJT6wIED2bp1K263O8JRNbz09HTS09MrHo5vqMdXWuuGeqwmJTU1VVe37fvPP/9Mr169jvxBAl7I/hX8HrC7IakH2JxHfl1xRPbv30/Pnj359ttv6d69e9j3O+6445g3bx5HHXVUPUbXOFx44YWcfPLJja4yZJ39vymaLZ/Px+WXX865555burOtOEgptUprndoQjyVDMJFQmA0rpsFb50H6rbDwfvPrW+eZxwuz6+VhlVI13iZNmnTEj3HmmWdy880319pu0qRJpY9rt9tJTk5m6NChvPTSS/j9/np5zHBNnTqV0aNHH1LysXLlSvx+P506daqzOABefvllunbtisvlYtCgQXz11VfVtu3SpUuVv9cxY8aUa5eRkcFVV11FmzZtcLlc9O7dm+XLl5eez8/P57bbbqNz58643W5OPvlkfvzxx3LXmDJlCk888QS5ubl1+nyFqE9er5cLL7yQ2bNnc9ttt8m/3wiTBKSh5WyFD6+FNe+YvR3RSRDV2vxqc8Lad8zzOVvr/KEzMjJKb9OnT6907Lnnnqvzx6zJ8OHDycjIYMeOHSxatIi0tDSmTJnCaaedRmFhYYPGUqKoqIjXXnuNa6+9tsZ2gUCg9PucnByuvPJKZsyYUac7ZM6aNYtbb72VBx54gDVr1nDyySdzzjnnsHPnzirb//jjj+V+n6tXr0YpxUUXXVTa5sCBA5xyyilorZk/fz4///wzL7zwAsnJyaVtrrvuOhYuXMibb77J+vXrOfvssxk+fDh//PFHaZt+/frRrVs33n777Tp7vkLUJ4/Hw3nnnUd6ejqtWrVi4cKFxMc32GiDqIrWWm5V3AYNGqSrs2nTpmrP1aggS+s3z9X6tRFav3NR9bfXRpjtCrIO73HCMHv2bG3++sszDEM/9dRTulu3btrlcum+ffvqmTNnlmuzfPlyfeKJJ+ro6GgdFxenBw8erNevX6+vuuoqDZS7bd++vcrHv+qqq/SYMWMqHV+/fr222+364YcfLj322Wef6VNPPVUnJCToxMREffbZZ5f+Dmp6zJruV9Pr0qpVK20YRumxXbt2aUC/9957eujQodrpdOoZM2ZorbUuLi7Wp512mn7rrbdqvO7hGDx4sL7uuuvKHTv66KP1fffdF9b9n3jiCR0fH68LCwtLj91///365JNPrvY+RUVF2mq16rlz55Y7PnDgQP3ggw+WO/boo4/qU045JaxYGsph/78pmrWCggJ91llnaUAnJSXptWvXRjqkRgtYqRvofVZ6QBrSxo/AcwDciTW3cydC8QGzfQN76KGHmDFjBi+99BKbNm3i/vvv54YbbmD+/PmA+cl/3LhxnHrqqaxbt44ffviBW2+9FavVynPPPceQIUO4+uqrSz+FH+qQRN++fRk1ahQffvhh6bHCwkJuu+02VqxYwbJly4iPjyctLQ2fz1fjY9Z0v+p89dVXDBo0qFxPxtq1awF46qmnuOuuu9i4cSPjxo1Da82kSZMYNmwYV1xxRa3PberUqcTExNR4Kxli8fl8rFq1irPPPrvcNc4++2y+/fbbWh9La82MGTO4/PLLy02KnTt3LieeeCIXX3wxycnJDBgwgBdffBEdmgsWCAQIBoO4XK5y13O73Xz99dfljg0ePJgVK1bg8XhqjUeISMnPz2f06NEsWbKEtm3bsmzZMvr37x/psASyCqbh+Ithw5zak48SrkTYOAcGXtVgE1MLCwv517/+xaJFizjttNMA6Nq1KytWrOCll15izJgx5OXlceDAAdLS0krnSPTs2bP0Gg6Hg6ioKNq1a3fYcfTu3ZvPP/+89Ofx48eXO//6668TFxfHihUrOPXUU6t9zNruV5XffvuNlJSUcsfWrVuHy+Vi9uzZHH300aXHv/76a2bNmsVxxx3H3LlzAZg5c2bpJlYV3XjjjeWGQ6rSoUMHwNwGPBgM0rZt23Ln27ZtW+61qc7ixYvZvn071113Xbnj27Zt4+WXX+b222/nvvvuY+3atfzlL38B4OabbyY2NpYhQ4bwxBNP0LdvX9q1a8d7773Hd999V+65A7Rv3x6/38/u3bsPab6MEA1p165drF+/nvbt27N06VKOPfbYSIckQiQBaSg5WyDoA1dceO1tTvDmm6tk2vWt39hCNm3aRHFxMaNGjSrXA+D3++nSpQsArVq1YtKkSYwcOZKzzjqLs846iwkTJtTp5EtdZokcwNatW/nrX//KDz/8QFZWFoZhYBhGtXMhjuR+Ho+n0pv+2rVrGT16dKU34FNPPfWQ9oto1aoVrVq1Crs9UGlOScXXpjrTp0/nhBNOYMCAAeWOG4ZBamoqf/vb3wA4/vjj+fXXX3nppZdKJ/LOnDmTa665ho4dO2K1Whk4cCCXXHIJq1evLnetkmWL0gMiGrPevXuzePFiEhISJFFuZGQIpqH4PXCoExSVgkBx/cRThZI30/T0dNauXVt627hxI4sWLSpt9/rrr/PDDz9w+umn88knn3DMMcewcOHCOotj06ZNdOvWrfTntLQ0srKyePXVV/nhhx9Ys2YNNputxqGUw71fUlIS+/fvL3ds3bp1nHHGGUf2pDi0IZikpCSsVit79uwpd43MzMxKCVJFmZmZfPzxx1x//fWVzqWkpNC7d+9yx3r16lUuKevevTvLly+noKCAXbt2sWLFCvx+P127di13v3379gHQpk2b8F8EIRpATk4Oc+bMKf150KBBknw0QtID0lDsbjjUmitag81Ve7s60rt3b5xOJ7/99hvDhg2rsW3//v3p378/9957L+eccw5vvvkmI0eOxOFwEAwGDzuGDRs2sGDBAh566CHA/EPy888/89JLLzF06FAAVq9eXW4VSlWPGc79qnL88cfzxhtvlP5cWFjI1q1bGThw4GE/pxKHMgTjcDgYNGgQixcvZsKECaXnFy9eXGloqaI33ngDp9PJxIkTK5075ZRT+OWXX8od+7//+z86d+5cqW10dDTR0dHs37+fhQsX8o9//KPc+Q0bNtC+fftaEyIhGlJmZibDhw9nw4YN/O9//+OCCy6IdEiiGpKANJSkHmB1mMXHwpnTEfCCzWHer4HExsZy1113cdddd6G15vTTT6egoIDvv/8ei8XC5MmT2b59O6+++irnnnsuHTp0YNu2bfz000/86U9/AsxaFCtWrGDHjh3ExMTQqlUrLJaqO9q8Xi979uzBMAyysrJYsmQJU6dOZdCgQdx1110AJCYmkpSUxPTp0+nUqRN//PEHd999NzbbwX+6VT1mOPerysiRI7n33nvJycmhdevW/PTTTwCVhjIOx6EOwdxxxx1cccUVDB48mFNOOYVXXnmF3bt3c+ONN5a2efHFF3nxxRfZvHkzYA7RvPbaa0ycOJHY2NhK17z99ts5+eSTefLJJ7n44otZs2YNzz//PFOnTi1ts3DhQgzDoGfPnmzZsoW7776bY489lquvvrrctb766itGjRp1qC+DEPUmIyODs846q7Qg3ZAhQyIdkqhJQy23aWq3elmG+8OrWr9yes1LcEtur55utq8nNS3Dff7553WvXr20w+HQSUlJevjw4XrRokVaa6337Nmjzz//fN2+fXvtcDh0p06d9N133619Pp/WWutffvlFn3TSSdrtdte6DJfQslmr1apbt26tzzjjDP38889rr9dbru2SJUt0nz59tNPp1H369NELFizQ0dHR+vXXX6/xMWu7X3VOOukk/eKLL2qttf7Pf/6jjz322DBf1br30ksv6c6dO2uHw6EHDhyoly9fXu78lClTyv0ely5dqgH9ww8/VHvNefPm6eOOO047nU7do0cP/dxzz5Vbdjxr1izdrVs37XA4dLt27fRNN92kDxw4UO4aHo9Hx8XF6e+++66OnmndkGW4LdeuXbt0jx49NKD79eun9+7dG+mQmiQacBmulGKvRr2UYi/MNouM+T01r4bx7DeHbMbPMAuUiQa1YMECbr31VjZt2oTVao10OI3SSy+9xMcff1xublBjIKXYW6aSYeNt27YxYMAAFi9eTFKS/O08HFKKvbmKToIx/zKTi4I95jBLWQGvedzuNttJ8hERo0aN4qabbuL333+PdCiNlt1u54UXXoh0GEKgtWb8+PFs27aNE044gaVLl0ry0URID0g16nUzusJss8jYxjkQ8JmrXbQ253z0OR/6XCDJhxCHQXpAWqa1a9fy0EMP8c4770h59SPUkD0gMgk1EqKTYPBks8hY9q/mUlubS3bDFUKIMOXl5REXZ9ZVGjBgAPPmzYtwROJQNdohGKXUKUqpRUqpTKVUnlJqtVLqmgptEpVSrymlspVShUqpz5VSlcpQKqVcSql/KqUylFIepdR3SqnTG+7ZVMPmNIuMdUw1v0ryIYQQtfrpp5845phjyi2ZF01Po0xAlFLHAZ8DduB6YDzwIzBDKfWnUBsFfAKMAv4SamMHvlBKdaxwyRmh6zwMjAUygIVKqQH1/mSEEELUmdWrVzN06FD27t3L7NmzkWkETVdjHYKZCFiBNK11QejYYqVUf+BK4D/AucCpwDCt9RcASqnvgO3APcAtoWP9gUuBa7TWr4eOLQc2Ao+FriOEEKKR++GHHxg5ciS5ubmMHTuW2bNnh7U1gWicGmUPCOAA/EDFTSYOcDDmc4HdJckHgNY6F0gHxpW5z7mha80q0y4AvA+MVEod1riHZN1CNC7y/2Tz9s033zBixAhyc3M5//zz+fDDDyvt2iyalsaagLwR+vq8Uqq9UipBKXU9cBbw79C5PsCGKu67EThKKRVTpt12rXVRFe0cwNEcIrvdLhtwCdHI+P3+Wivdiqbpyy+/ZOTIkeTn53PxxRcza9YsHA5HpMMSR6hR/t+qtd6glDoTmAP8OXTYD9yotX4/9HMrYEcVd98X+poIFITa7a+hXZW1sbOyskhNrXolUu/evZk0aRLJyckopSp1AbZp00Y26BKiARmGwd69e2UJZjOVmJiIy+Xiggsu4L///a8kmrWYNm0a06ZNO9y7N1gNiEb5W1RK9QA+xOyluBFzKGYc8IpSqlhr/Q6gMEt5V7p7FT+H066cNm3aUF0dEDCXgGVmZuL3+yudy87OJjs7u6bLCyHqWHR0tBSgaqb69evHihUr6Ny5s1QnDsPkyZOZPHnyYd1XKdVgb16NMgEBpmL2eIzVWpe8wy9RSrUGnlNKvYfZg1FV70VJjfOSXo99wFE1tNtXxblaxcXFla5BF0IIUbfS09PJzMzk2muvBaBbt24RjkjUtcY6B6QfsK5M8lFiBdAaSMbsHelTxX17AzvLrJ7ZCHRVSkVV0c4HbKmzqIUQQhyxDz/8kAsuuIDrr7+eFStWRDocUU8aawKyBxiglKo4y+hEoBiz1+IToINS6oySk0qpOCAtdK7EJ5j1QSaUaWcDLgYWaa0rbMgihBAiUt577z0uvvhiAoEAd999NyeccEKkQxL1pLEOwbwIzAbSlVIvY84BORe4BPi31tqnlPoE+A54Wyl1N+aQy/2Yczv+UXIhrfVapdQs4FmllB2zTsifgK7AZQ34nIQQQtTgrbfe4uqrr8YwDP7617/y6KOPSp2PZqxR9oBorf8HjAacwGuYE1JPBW4C7g61MTCrmi4GXsZcMRMEhmqtd1W45NXA68ATwHygEzBKa7263p+MEEKIWs2YMYNJkyZhGAaPP/44jz32mCQfzZzshluNmnbDFUIIUXdyc3M55phjyMzM5KmnnuKee+6JdEgtluyGK4QQosWIj49n8eLFfPPNN/zpT3+KdDiigUgCIoQQIiLWrl3LgAEDADjuuOM47rjjIhuQaFCNcg6IEEKI5u3xxx/n+OOPZ8aMGZEORUSI9IAIIYRoMFpr/vrXv/Lkk09isViw2+2RDklESNgJiFKqC3AS0B5wA9nAL8D3WuvieolOCCFEs6G15p577uHpp5/GarXy9ttvM3HixEiHJSKkxgREKZUAXBe69aDq/VNKanK8rLVeVtcBCiGEaPq01tx22208//zz2Gw23n//fcaPHx/psEQEVTsHRCl1J7ANuANYCFyEuXV9POY29u2AIcC9QAKwWCn1uVLq2HqOWQghRBPz4IMP8vzzz+NwOPjoo48k+RA1TkK9DLgG6Ki1vlVr/aHWepvWOl9rHdBaZ2qtf9BaP6e1PhvoDPyEWQpdCCGEKHXZZZfRqVMnPv74Y9LS5G1C1DAEo7UeeCgX0lrvxuwtEUIIIdBal1Yz7dOnD7/++itOpzPCUYnGQpbhCiGEqHN+v59LLrmEV155pfSYJB+irENehquUSgZcFY9rrXfWSURCCCGaNJ/Px8SJE5kzZw4LFizgwgsvJCkpKdJhiUYmrAREKWXB3MjtBswJp1Wx1lFMQgghmqji4mImTJjAvHnzSEhIYOHChZJ8iCqFOwRzG+ZOtM9gLsWdipmQbAe2AtfXR3BCCCGaDo/Hw7hx45g3bx6tWrViyZIlDB48ONJhiUYq3ATkauAx4KnQz3O01lOAXsAfwFH1EJsQQogmorCwkDFjxrBo0SLatGnDsmXLGDjwkNYyiBYm3Dkg3YCVWuugUiqAWQkVrbVfKfUs8ALwSL1EGCG5ublMnjy53LG0tDRZPiaEEFXYs2cPmzZtol27dixZsoTevXtHOiRRi/T0dNLT0ysejm+ox1da69obKbUbuFprvVAptQN4TGv939C58cCbWuuYeo20gaWmpuqVK1dGOgwhhGgyNm7ciN1u55hjjol0KOIwKaVWaa1TG+Kxwh2CWQOUpLMLgUeVUpcopSYAfwNW10dwQgghGq99+/bx3nvvlf7cp08fST5E2MIdgnkWcxgGYAowEHgn9PNvwM11G5YQQojGLDs7mxEjRrB27VoCgQBXXHFFpEMSTUxYCYjWenGZ7/copQYD3YEo4Gettb+e4hNCCNHI7N27l+HDh7NhwwaOOeYYhg4dGumQRBNU4xCMUmqSUmqtUqpAKfW7UuoZpZRDm7ZorX+S5EMIIVqO3bt3c+aZZ7JhwwZ69erFsmXL6NixY6TDEk1QTbvhXgL8F7OXYz6QgVkPZGqDRCaEEKJR2bVrF2eccQabN2+mX79+LFu2jJSUlEiHJZqomnpAbgPmAL201hdrrU/ArAVyk1JKqp4KIUQLorXm0ksvZcuWLRx//PF88cUXJCcnRzos0YTVlIAcA0zXWgfLHHsZcCKFx4QQokVRSjF9+nTS0tJYsmQJrVu3jnRIoomrKQGJB/ZVOFbyc2L9hCOEEKIxOXDgQOn3PXv25JNPPiExUd4CxJGrrQ6IRSlVeuPghnPljofOCSGEaEY2btxIr169ePHFFyMdimiGakscvgH8ZW6e0PEfKhz31VeAQgghGt66des488wz2bNnDx9//DHBYLD2OwlxCGqqA/Jog0UhhBCi0Vi1ahUjRoxg//79jBo1io8++girVdYeiLpVbQKitZYERAghWpgffviBkSNHkpuby7nnnssHH3yA0+mMdFiiGZK5G0IIIQD49ttvGTFiBLm5uYwfP57Zs2dL8iHqTbU9IEqpYYdyIa310iMPRwghRKS0adOG6OhoxowZw8yZM7HZwt0uTIhDV9O/rs8BHfpeVdNGh85pDq6QEUII0QT16NGDH374gQ4dOsicD1Hvaktv84EPQ7fC+g9HCCFEQ/r000/ZsmULt9xyCwBHHSV1JkXDqCkBGQpcCYwHJmCWZX9ThlqEEKJ5+Pjjj5kwYQJ+v5/+/ftzxhlnRDok0YJUOwlVa71ca30t0A64EUgGFiqldiql/qaU6tVQQQohhKhbs2fP5sILL8Tv93P77bdz+umnRzok0cLUugpGa12stX5Xa30O5h4wzwGjgQ1KKSmPJ4QQTcy7777LxIkTCQQC3HfffTzzzDMoVd1UPyHqx6Euw80BdoRuGtkTRgghmpQ33niDyy+/HMMwmDJlClOnTpXkQ0REWGuslFKnAFdgzgVxAh8DY4DF9RdaZOXm5jJ58uRyx9LS0khLS4tQREIIcWi8gSDbsgrx+IO47VbaRsHDDz+M1ponnniCBx98MNIhighKT08nPT294uH4hnp8pbWu+oRSR2MmHZcDXYAvgbeA2VrrgoYKMFJSU1P1ypUrIx2GEEIcspwCL/PXZzBvXQa+oFFaK8FhtZCa6IE/1nPPHbdGOkzRCCmlVmmtUxvisWrqAfk/IA/4CLgO+C10PFkplVyxsdZ6W92HJ4QQ4lBszy7kobnryfP4SXA7iHXZyN6+iaSuvfEFDL7KtBMXdyI7sgvpkhQd6XBFC1bbHJA4YBJmUbJfa7kJIYSIoJwCLw/NXY/Xb5Ac68Jhs7Bu3ht88tgkNi56D4fNQnKsC6/f4MG568kp8EY6ZNGC1dQDcnWDRSGEEOKIzV+fQZ7HT3KsC601az9+jTUfTwelsLsO9nbEu+1k5hczf30GVw7pErmARYtW0264bzZkIEIIIQ5fsT/IvHUZJLgdaK1Z9eF/+Gn+Gyhl4bTrHubok0eXa5/gdjBvXQYXn9AJp03KrouGJ7vhCiFEM7A9uxBf0MBuVayY9ZyZfFisnHHD45WSDwCHzYIvaLAtS3bZEJFRbQKilLpdKeU6lIsppQYqpUYdeVhCCCEOhccfRAFrP36NjQvfxWK1MfRPU+l24ohq76Mwe06EiISaekCuBHYopf6ulOpfXSOlVKJS6gql1CLga8yJq0IIIRqQ225FA91PPofYNh0YdvNTdEkdWuN9NOCyy/CLiIyaJqEOxKwDcidwj1IqD1gPZAFezCqo3YDuoZ9nAb211jvqM2AhhBDlaa3p1iYah9WCs1V7Lpj6AVabvcb7+AIGDquFbm1kKa6IjJo2o9Na67e01v2BIcC/gXzMpON4IBb4CrgGaK+1vlqSDyGEaFiBQIArrriC5//9L8b2T+GAx1dr8gFwwONjbP8UmYAqIiasUuxa6x+AH+o5FiGEEIfA7/dz+eWX88EHH/Dxxx/z/ZoJLHTbyfX4iXdXn4TkevzEue2M6ZfSgNEKUZ6sghFCiCbI5/Nx8cUX88EHHxAXF8eiRYvoc3QXnjyvH067hcz8YnwBo/x9AgaZ+cU47RaePK8frWOcEYpeiDB7QIQQQjQexcXFXHjhhcyfP5+EhAQWLVrECSecAECXpGien3h86V4w+d5Aub1gJqR2Yky/FEk+RMRJAiKEEE1IUVER559/PosWLaJ169Z8/vnnDBgwoFyb1jFOrhzShYtP6MS2rEKK/UFcdivd2kTLnA/RaEgCIoQQTUhOTg4///wzycnJLFmyhL59+1bb1mmz0itFKiOIkIAXsn8FvwfsbkjqAbbI9YRJAiKEEE1Ip06dWLp0KX6/n169ekU6HNEUFGbDxo9gwxwI+kAp0BqsDuh7PvS5AKKTGjwsmYQqhBCN3IEDB3jzzYPbcx199NGSfIjw5GyFD6+FNe+YvR3RSRDV2vxqc8Lad8zzOVsbPLRD6gFRSrUDjgIqlWjXWn9ZV0EJIYQw7du3j7PPPptVq1ZRXFzMDTfcEOmQRFNRmA3z7zCHXGLbVT5vc0JMO/DsN9uNn9Gg4YWVgCilOgBvA6dXdRpzgrXMbBJCiDqUlZXF8OHD+emnn+jevTvnnHNOpEMSTcnGj8BzoOrkoyx3IhTsMds3oHB7QP4D9AXuwSzH7q23iIQQQrBnzx7OOussNm3axLHHHsuSJUvo0KFDpMMSTYW/2Jzz4U4Mr70rETbOQZmdCg0i3ATkNOAWrfXM+gxGCCEE/PHHH5x11ln88ssv9O7dmyVLltCuXS2fYoUoK2eLOeHUFeYqKJsTvPlE2StPsagv4SYgHiCzPgNpbHJzc5k8eXK5Y2lpaaSlpUUoIiFESzFp0iR++eUXjjvuOD7//HPatGkT6ZBEU+P3mKtdapC+ejfpq3cfPBAoJqgbbkd7pbWuvZFSjwLdtdaX139IjUNqaqpeuXJlpMMQQrRA27dv5/bbb2fGjBm0bt060uGIpmjPBki/9dCW1xZmE3PT0k0FPt2n/gI7qNoeEKXUNWV+/B24Qim1FPgU2Fexvdb6v3UfnhBCtAz79u2jVatWAHTt2pW5c+dGNiDRtCX1MOt8BLzhFRsLeMHmoMhPcf0HZ6ppCOa1Ko51Ac6s4rgGJAERQojDsHnzZoYNG8Ytt9zCfffdF+lwRHNgc5pFxta8U/sqGIDi/TDgMjQLah8WqSM1FSLregi3bvUbphBCNE8bNmzgjDPOICMjg4ULFxIIBCIdkmgu+lwA7gSzzkdNPPvBlWC2b0DV9oBorX9ryECEEKKlWbt2LcOHDycnJ4cRI0Ywd+5cbDbZIUPUkegkGPMvs8hYwR5zqW3Z4ZiA1+z5cCWY7Rq4HHtYpdiVUkGl1OBqzg1SSgXrNiwhhGjeVq5cybBhw8jJyWH06NF88sknREVFRTos0dy07m5WOB1wGQS9ZnXUohzza9BrHh8/w2zXwMJNtWtay2PFnAMihBAiDCtWrGDEiBHk5eUxbtw4Zs2ahdMZuV1JRTMXnQSDJ8PAq8zdcAPFYHM17t1wlVIWDiYfltDPZbmBc4DseohNCCGapbZt25KYmMjZZ5/Nu+++i91uj3RIoiWwOaFd30hHUaqmZbhTgIdDP2rgmxqu83JdBiWEEM1Z586d+fbbb0lOTpY5H6LFqulf/rLQV4WZiMzArAdSlhfYBMyr88iEEKIZWbhwIatXr+b+++8HoH379hGOSIjIqmkVzHJgOYBSSgOvaa3/aKjAhBCiuZg3bx7jx4/H5/MxYMAA2dVWCMKchKq1frS+AxFCiOZozpw5XHzxxfj9fm666SZGjhwZ6ZCEaBTCSkCUUjVVOTWAXGAV8JHWusHKuAohRGP2wQcfcOmllxIMBrnjjjt4+umnUbVsECZESxHu7KehQDyQAAQwV70khe5/INTmdmCrUmqo1rriXBEhhGi2vIEg27IK8fiDuO1WurWJZvb773HVVVdhGAb3338/Tz75pCQfQpQRbgJyKfAucC3wsdbaCC3JPR94JnTeD3wE/A24oh5iFUKIRiWnwMv89RnMW5eBL2igMJcMWo0A3/zzrxiGwSOPPMLDDz8syYcQFYRVCRX4N/CU1nqO1toA0FobWusPgaeAZ7XWP2ImHyPqKjil1Gil1JdKqQKlVJ5SaqVSaliZ84lKqdeUUtlKqUKl1OdKqX5VXMellPqnUipDKeVRSn2nlDq9ruIUQrQ827MLueX9NcxeuQunzULraAetoh20jnYQ5XbR59p/MOiSO7jqprsk+RCiCuEmIP2BrdWc2wqUVDbZBCQeaVAASqkbgI8x55acD0wAZgNRofMK+AQYBfwFGA/YgS+UUh0rXG4GcD3mcuKxQAawUCk1oC5iFUK0LDkFXh6aux6v3yA51oXDZv4pzdy6HgCHzUKXzl3octp4Hpy7npwCbyTDFaJRCjcB2QNcWM25CcDe0PdxQC3b7tVOKdUFeBa4W2t9u9Z6sdZ6odb6Ka11Sc2Rc4FTgSu01u9prReEjlmAe8pcqz/mENHtWuvpWuslwEXATuCxI41VCNHyzF+fQZ7HT7z7YAXT9QveYd4T17L2kxmlx+LddvI8fuavz4hEmEI0auEmIM8C1yul0pVSVymlzgl9nYc5L+TfoXanAWvqIK5rMFfXvFJDm3OB3VrrL0oOaK1zgXRgXIV2fmBWmXYB4H1gpFJKNmAQQoSt2B9k3roM4lx28r0Bcj1+VsydwY+zngPAFVu+EzjB7WDeugy8AdmzU4iywq0D8pxSqhBzCGNMmVO/A9drrUtS/pcATx3EdSqwGZiolPor0BnYAfxba/1SqE0fYEMV990IXKmUitFaF4TabddaF1XRzgEcHfpeCCFqtXbnfnbneigoDhA0DDKXv03m8ncARe+J99Dt9PPKtXfYLOR7A2zLKqRXSlxEYhaiMQp7EwKt9WtKqRlARyAFcx7F71prXabNjjqKq33o9k/gAcx5JhOAF5VSNq31c0ArzKSkon2hr4lAQahdVcNCJe1aVRVAVlYWqamphxX85MmTmTx58mHdVwjReG3PLmTqZz+TU+DDZVNkf/EGmV/NAmWh8/l3YTl2KBv+yKVnuziiHNbS+ynMnhMhGsK0adOYNm3a4d49qS5jqckh7YIUSjZ2hW71yQLEApO01h+Fji0NzQ25Xyn1PJSueKuo4nTzcNuV06ZNG1auXHlIQQshmq+SiaeBoMZuVWR98z/2fDULLBa6Xng/rfqeAYDf0Gzek0ffDvE4rOYotwZcdmsNVxei7hzJh2ClVIPtbh92AqKUigNGA0cBrgqntdb68TqMKwfoASyucHwR5qqXFMwejKp6L0oGYEt6PfZhxlxdu31VnBNCiHJKJp62jXWx+0AxCf3OIHvNAjqOuJaEXqeUtrNbFN6gwd68YjolRuELGDisFrq1iY5g9EI0PuGWYj8Fc3JnQjVNNFCXCchG4KSqQgl9NUJtzq6iTW9gZ2j+R8m1zldKRVWYB9Ib8AFb6iZkIURzVTLxNN5pQylIjnOSodvQ56ZpKGvlP6N2iyIzz0uHBDcHPD4mpHbCaZMeECHKOpRVMDuAEwCX1tpS4VbX/2fNCX2tuGvTSMx5J3swa4B0UEqdUXIy1EuTFjpX4hPM+iATyrSzARcDi7TWskBfCFGj7dmFeP1+fnjrSdbMnUbbWBc2iyJQzZ9Qi1IYWrMnr5g4t50x/VIaOGIhGr9wh2B6ARdprVfVZzBlfAp8AbyqlEoCtmHWITkbuDrU5hPgO+BtpdTdmEMu92P2kvyj5EJa67VKqVnAs0opO7Ad+BPQFbisYZ6OEKIpy/d4Wf3m4/y+6nNsDhfHnHYuPdu1YfOefLxBA7tFYSlT7dTQGn9QY7daePK8frSOkdX+QlQUbgKyE2iw/4O01lopdR5mafdHMedrbAYu01q/G2pjKKXGAk8DL2POS/kOGKq1rjhJ9mrgSeAJzGGkdcAorfXq+n82QojGpKqN42oaHvH7/fz1luv4fdXn2F3RjLj938QkmT0afdvHsze/mMw8LwHDKL2PRSlaRTu4/5yedEmSuR9CVEWVWUVbfSOlLgbuAEZorfPqPapGIDU1VcsqGCGaj+o2jnNYLYztn8KYfimVeiq8Xi8XXXQRn3zyCXZ3DMNufZYOxx5X6dqGoSn0BTG0xqIUdqvCH9TMvG6wzP0QTYpSapXW+vBqUByicHtAxgJtge1Kqe+ovHJEa62vqtPIhBCijmzPLuShuevJ8/hJcDuIdR380+cLGMxeuYuFG/fw5Hn9SnssPB4P48eP57PPPiMxMZFbn3mTlQUJVV7fYlHlrpmZXywTT4WoRbiTUE/F/LCQh1lZ9LQqbkII0ehUt3FcCYfNQnKsC6/fKLdxXG5uLr/++itJSUl88cUX3DzhbOLcdnI9/hofL9fjl4mnQoQh3FLsXes7ECGEqA8l9TuSYyuWLyov3m0nM7+Y+eszuHJIF9q1a8fSpUvJy8ujT58+ADx5Xj8enLuezPxiEtyOcsmML2BwwOMjzm2XiadChCHcHhAhhGhySup3JLgdYbWPws9zL/6HYn8AgE6dOpUmHwBdkqJ5fuLxTEjthDdgkFPoY1+hj5xCH96AwYTUTjw/8XiZeCpEGA6lEmo05s63pwOtgcla61+VUhOBtVrrzfUUoxBCHJbt2YX4gka5+RnV8RbmsfTZW8netpEpPeJ56pEHqmzXOsbJlUO6cPEJndiWVUixP4grjNU0Qojywq2E2glYhrkR3WagL+ZeLQBDgeHAdfUQnxBCHDaPP1jzpk8hxQUHWPjPv5Cz8xeiWqdw2vDRtd7HabPK7rZCHIFwh2CeAbyY+7MMovxGbssxe0WEEKJRcdutVe5EWZYnbx+fPfVncnb+QmxyR0659QW6d+/WIPEJ0ZKFOwQzAnPIZadSqmIf4x9Ah7oNSwghjly3NtE4rBZzQzhb5c9bRQeyWfDPmziwezvx7Tpz1h0vYI1Nko3jhGgA4faAOID8as7FAzWvSxNCiAhw2qyM7Z/CAY+vyvPfvPk3DuzeTkL7rpxz33/wuRIY2z9F5nII0QDCTUB+AsZXc+4coKH2iBFCiEMypl9KtfU7Tr7yXjoPGso5976C3xEv9TuEaEDhJiD/BK5VSk3n4HyP3kqpRzFXxvyzPoITQogj1TrGyZPn9cNpt5CZX0ze/n2UbEERnZjMaTf+jXzlxmmXjeOEaEhh7QUDoJS6Efg75uqXkkmo+cDdWutp9RNe5MheMEI0LzkFXmZ8+i2P3DiRjieMpHfaZFCqxr1ghGhpGt1eMEqpeOB1YCYwBEgGcoBvtdbVzQ0RQohGI3PXNp697TI8+zOJPbCF+0ceTVxMlNTvECJCak1AlFI2zGTjfK11OvB5vUclhBB1aMOGDZx11llkZmYydOhQ0tPTiY6WlS5CRFKtCYjWOqCU2gsEGyCeRiM3N5fJkyeXO5aWlkZaWlqEIhJClPAGgmzLKsTjD+KupQrpmjVrGDFiBDk5OZx99tnMmTOHqKioBo5YiMYnPT2d9PT0iofjG+rxw5oDopR6Cuihtb6g/kNqHGQOiBCNT06Bl/nrM5i3LgNf0EBhbtNd3TyO1atXc9ZZZ3HgwAHGjBnD//73P1yumjelE6Ila3RzQIAdwKVKqR+Bj4EMKF9gUGv937oNTQghDtqeXchDc9eT5/GT4HaU29/FFzD44MddfLT6D64/vSvdkmLo1iaa9u3bk5yczNChQ3n//fdxOMLblE4IUf/C7QExammitdbNahaX9IAI0XjkFHi55f01eP0G8W57uXO+oMHevGIy87z4ggYWpejSOoooh42x/VMY3M5K947tsNvt1VxdCFGiMfaAdK3XKIQQogbz12eQ5/GTHFt++KTIF2TznjwChsZuUUTZreT8uopN327k5ItuYvbKXSx023nyvFZ0SZIERIjGJKwERGv9W30HIoQQVSn2B5m3LoMEd/nhE1/QYPOePAwNTqtZUzH315X89t4j6ICPLsf0pesJw8j1+Hlw7nqen3i81PkQohEJqxKqUiqolBpczblBSqkWtUJGCNFwtmcX4gtW3kxub15xac8HwIHN37H13SnogI9Wg0bTus+pAMS77eR5/Mxfn9HgsQshqhduKXZVwzkr1LrjtRBCHBaPP1jpD1BQazLzvKXJx/5NX7P1/cfQQT9tThxH+zF/QauD90pwO5i3LgNvQD4rCdFY1JiAKKUsSqmSyaWW0M9lb9GYm9Fl13ukQogWyW23VvqEU+QLYmiNRSn2rf+CbR88AUaQtqdcSKfRf0YphaVMAuKwWfAFDbZlFTZs8EKIalU7B0QpNQV4OPSjBr6p4Tov12VQQghRolubaBxWC77AwWEYwzBTEh0MsufL98EwaHf6JbQ/axIasChFtKP8wjyFOZ+kyQl4IftX8HvA7oakHmCTuSyi6atpEuqy0FeFmYjMAH6v0MYLbALm1XlkQggBOG1WxvZPYfbKXaWrYCyhoRdltdLjyqns3/Q1ySeOA8zJqSnxrtI2JTTgsjehagGF2bDxI9gwB4I+UAq0BqsD+p4PfS6A6KRIRynEYas2AdFaLweWAyilNDBda727oQITQogSY/qlsHDjHnI9fuLddgp+24ByHIWhNfbY1qXJhz9oYLMo2lZYrusLGDisFrq1aSL7v+Rshfl3gOcAuBPBFXfwXMALa9+BzfNhzL+gdfeIhSnEkQhrEqrW+tGyyYdSKl4plaqU6lh/oQkhhKl1jJMnz+uH027h+/SZfPa3yeR9+Sb+0FCMoTXeoIHFoujZLrbSipkDHh9j+6c0jV1vC7PN5MPvgdh2lYdbbE6IaWeen3+H2V6IJqjaBEQpNVIp9fcqjj8IZAI/AL8ppd4N7ZgrhBD1pktSND32fMGmj14AILp1O7Q2V8kYGlLiXfRtH0+Uo/yfo1yPnzi3nTH9UiIR9qHb+NHBno+auBOh+IDZXogmqKbE4UYqLK9VSo0AHgfWA68BvYAbgFXAM/UUoxCiBahth9snnniCR//6V5RSvPSfVzjz3Ilszy5k2pdb8QcNWkU5y/V8+AIGBzw+4tx2njyvX9MoQuYvNud81JZ8lHAlwsY5MPAqmZgqmpyaEpDjMZONsq4GioGRWus9AMpc6nYpkoAIIQ6RNxBk9W/7Wbo5k++25WANLZ8tu8Pt6L7teP6fU3n88cexWCz897//5aqrrgKgV0ocJ3ZtVbpDbr43UG6H3AmpnSrtkNuo5WwxJ5yWnfNRE5sTvPnmKpl2fes3NiHqWE0JSDKwtcKxEcDXJclHyHzgiroOTAjRfOUcyOObFd/z1YadbM012GKkYFjs2CwWkuOctI1zgYbZK3fx3HPPsXrWs1itVmbOnMkll1xS7lqtY5xcOaQLF5/QiW1ZhRT7g7iq6EFpEvwec7XLoVAKAsX1E48Q9aimBCQfKJ0yrpTqAbQGvq/QLg+zGqoQQtSsMJvsH94n6/v36BH00TmoUUoTsNpZpE5mEUPIyI0jK99Lz3ZxJMe60AOHk/DNPP7998e55JKJ1V7aabPSKyXMnoPGyu42l9oeCq3B5qq9nRCNTE2rYDYD48r8PA6zZ3NRhXZdgb11HJcQornJ2Ypv1tXkffcGXm0nV8Wzn1gOEI8XO2n6C/6mn6O7ZS/BoMHPGbn4ggZtk5M5+c5pWLqdFOlnUP+Seph1PgLe8NoHvGBzmPcToompKQH5N3CdUup/SqmXgEcxJ59WrIh6PrCunuITQjQHoaWl+/PyyCYBrE68AQNFaAmttpGlE3Dg457Aq+TM+ye/f/YKe3I9ALSKiWoZe7nYnGaRMc/+8NoX74c+58sEVNEkVZuAaK3nArcBJwBXYg69TND6YP9gqA7IUODTeo1SCNG0bfwIo2g/v3uc2K0WfMEgQUPjD2oCxsFbpt/N3+b+zO7VS9m3+jN+374NQ+uWtZdLnwvAnVB7EuLZD64Es70QTVCNhci01s9rrTtrrWO11mdprX+tcP53rXWC1npa/YYphGiyQktLi6zxGNpMNPKLAxhao6D0FgwaPDX3Z+at34/bbuHoS6dgb92BQp/Z69Fk93I5VNFJZoVTuxsK9lQejgl4zeN2t9lOyrGLJkoKiFUjNzeXyZMnlzuWlpZGWlpahCISookKLS0NWtwYhqc0oSjLHzR4cs7PfL05myiHlRcu7cbsdgns0vrgxnM0sb1cjkTr7jB+hllkbOMcc6ltyV4wNgcMuEz2ghFHLD09nfT09IqH4xvq8ZU+1BnXLURqaqpeuXJlpMMQounb9SMsvJ98Szyrd+4nEDRK30s14A8YPP7hJr77NYdop5W/X3ocJ3dQPBa4kg2qBwM7J+K0WvAGDGZeN7jpLa09UiW74QaKzdUushuuqEdKqVVa69SGeCzpARFC1K/Q0lKHTREIGqGDJeXCwOMLknGgmFi3jacuPY5jUmJB5VGMg4ChcdosHCjyMSG1U8tLPsBMNqTImGiGJAERQtSv0NLS3PxClFKU9LqqUA4SF2Xnn5cdx/4iH92SY7Djx6/tbNMpKOCP/UWkJLibzl4uQoiwhLUbrhBCHLbQ0lKr9wBKmds3FPkCzFnxB2BORE2IcdA1OQaABFXIPOMkfNgBsCjVdPZyEUKETRIQIUT963MB2hlPHAV4fEHue+cnXly4hZlf/WYmJZi3OArJ1dHMD56E1aKwWRTXn9aNLknRtT2CEKKJqXYIRil15aFcSGv91pGHI4RojrzORDJOm0rWzD9z37ur2Li7iOQ4J2f1bQuAQwWIp4A8Hc3j+joCrlbEoAlqxejjZOhFiOaopjkgb1T4uWS5jKriGIAkIEKIUt5AkDW/7WfJ5ky+25pDwJPHN29nUrCniPYJTqZfeTQdE4NAPj7szGMoS9TJ5FoScAAF3gBn9WpDrMse6acihKgHNSUgXct83xF4F3Pn2/cx935pC1wCnBP6KoRoQbyBINuyCvH4g7jL7D6bU+Bl/voM/rfqd7ZlFRI0NHhy2THzfoozt2NPbI/rkkd5McZPlN+PxeFml+qAXx1MNDz+IE67hduHHxPBZyiEqE/VJiBa699KvldKPQe8r7W+t0yTX4AvlVJPAfdg7gkjhGjmShKMeesy8AWN0gW1DquFId1b8d3WHPKLA+zO9WC1KFw2C9vm/ofizO04kzrR+Yq/E3DGszm0ItfqVcS6bViAgGFQ7Ddw2i28dOlAjm4bG8FnKoSoT+Euwz0LeLGac4uBP9VNOEKIxmx7diEPzV1PnsdPgttBrOvgn5ACb4DXvtqORUFSrBNDg9Nqjth2GnOT+XX0n8EdjwbsVsUBT4CAoSkoDmCzWLBYFGf1Sub24cdI8iFEMxduAuIFUoHPqzh3AuCrs4iEEI1SToGXh+aux+s3SI51VTq/v8iHxaJQwK59HlyBAnRMAkop7NEJdLvowdK23qBB6xgnfVPi2ZvvpTgQ5PYRxzCsZ7LM+RCihQh3Ge4HwCNKqbuVUl2UUu7Q13uAKcCs+gtRCNEYzF+fQZ7HT7y7coIQNDSZeV7sFoVFKbz7dvN/0/7Crk9fpqrtHuwWRWaeF4tF0SHRTUKUg2PaxkryIUQLEm4PyJ1ALPA34O9ljmvMyal31nFcQohGpNgfZN66DBLcjirPF/mCGFpjs1gozNpJxrv3EczPoWj3r2i/F+Uo32NiUYqAYVDoCxLrsrWcnW6FEKXCSkC01h7gCqXU48CJQAqQAfygtf6/eoxPCNEIbM8uxBc0ys35KMsI9XJ4Mn9j2+v3ECzcj6tTX7pe/jgWR+Xhmor3a1E73ZZVstGc32PumSMbzYkW5JD2ggklG5JwCNHCePzBcgWAKrIohWfvNnbMvJ9AYS7uzv1pe+HDWJxRNV7XohS+gIHDaqFbmxZU7bQwGzZ+BBvmQNBH6fbAVgf0PR/6XADRSZGOUoh6VVMl1KMO5UJa651HHo4QojFy262lVQcNQ1MYGnKxKEW0w4p37za2v3kvQU8+cUen0v7Ch/BiqzZpKXvf7EJvy9rpNmcrzL8DPAfAnQiuuIPnAl5Y+w5sng9j/gWtu0csTCHqW009IDsoX+m0Ni3kr4cQLU+3NtEoYEd2ITmFvtKhEzB7MRKtcbgT2qA69ab7xL9iWOz4vYFQu8ppiN/QpMS7yPcGiHPbW85Ot4XZZvLh90Bsu8rnbU6IaQee/Wa78TOkJ0Q0WzUlINdwaAmIEKKZ2n2gmFyPn90HPEQ5rNgsBxfQGVqTHXBw1BV/x+JwE1Q2DK3plBjFAY8Pb9AoXR0D4A8aWJXCqsBpt7SsnW43fmT2fFSVfJTlToSCPWb7wZMbJDQhGlpNlVDfaMA4hGh5msgExJL6H7EuG26HlaAGi4L87T9xYPO3dBx1A06rBb87Dg0EtUZrSI5z0rGVm715xWTmefEFgwQNsFoU3VpHcWFqR8b0S2k5yYe/2Jzz4U4Mr70rETbOgYFXNcp/F0IcqUOahKqUsgC9gdbASq11Yb1EJURz1sQmIJbU/0iOdRHlsLF5Tx45v67kt/cfRfu9RLU/mtb9h2O1KIp8QdrGOTlvQEe+22aWZI922Ojc2kpQa4Z0a82wnskM7JzYcuZ8lMjZYv6+y875qInNCd58M0lt17d+YxMiAsJOQJRSN2EWHUvCHJo5AVitlJoLLNVaP18vEQrRnDSxCYgV639EOawkZG9k3XtT0AE/CQPOJrrX6fiCBhalaJ/oIsHt4M/DuvPnYd3ZllVIsT+Iq8xmdS2W32Mmm4dCKQgU1088QkRYWAmIUup64Dngv8AizMqoJb4CxgPNKgHJzc1l8uTyY69paWmkpaVFKCLR5DWSCYiVdrFNtOE8sK3KoaCy9T98AYP13y3hpzcfQQf9tBo0hrbn/BltsRDttJIU48Rtt1Ecun6vlDh6pYT5ab+uNOZhLbvb7Ok6FFqDrfo6KkIcifT0dNLT0ysejm+ox1dVlUmu1Eipn4FPtNb3KqWsgB9I1VqvVkqNAWZorWuZVdW0pKam6pUrV0Y6DNGcrJgGa96pfQIimBMQB1xWpxMQK+5iG2/kcrL3S07zfkWbKEVClAO7RZUbClqdY+XR9I1oDWu/XMCuD/8ORpDkk86j/agb8QU1xf4gJWtd3HYrWsGlg4/ixjO6N9z8jqYwrBXwwlvnmQlROElRwAtBL1wxt/EkUaLZU0qt0lqnNsRjhTsE0xVYWM25QiChTqIRormK8ATEirvYdrfu4fKc54ky8imwxLK9yIrNa6Fnu1iiLEFY+w6+DemsbHULv+4JUOzzk/n1bDCCJJw4nsSzriHfayYeFqVAmathvAEDh02x7JcsNmXk8eR5/eiSYKvfXommMqxlc5rJULhJaPF+MwmV5EM0U+EmINlAl2rOHQv8USfRCNFcRXACYsVdbGOCB7g8+3nshpdca2sAHBYIBA0278mnb/t4Ao4kdu3ezQm7HydO/xmvJZb2Fz9K/qblxA0cS3HA7Dm1W1RpmQ+LUhiGxjAUHRPdqKJsvpj5JJfHrsKuA/XTK9FIhrXC1ucCMxny7K85GfXsB1eC2V6IZirc3XDTgYeVUt3KHNNKqSTgdmBuXQcmRLMSwQmIFXexPaFwGVFGPkXWWNDmTrb+oGH2YPiDbM8u5Oc9+eQTwy879jJKfYcGrFFxJKQenAOlgIBRfghXh0609f3Onfn/5NSCBWR6MN/0o1qbX21Os1fiw2vN3osjUVJXo7aeJXciFB8w20dSdJLZE2N3m8NsAW/58wGvedzuNttFethIiHoUbgLyEOAFNgCfY/6deR74GQgCj9VLdEI0FxGagFhxFYvN8HFC4XIKVCzF/iC5xT7yPH7yPAHyiwMU+YL8vr+IPI+f97/bySUz/o8Nn72PC78ZvgZDH6xtqjFLs5eEa1WKNpYCLs0xe1gK7EnsKdAEyz73kl4Jv8fslSjMPrwnd7jDWhXf9Bta6+5mT8yAy8w5HoXZUJRjfg16zePjZzSKVVBC1Kdwd8PNUUqlArcBI4Gtofu+CPxba51XbxEK0Rwk9TCHHgLe8Ccg2hzm/Y5AxV1s2wV+x2L42ed3oHUQw9BopUKdMwqtNRr44LtdvLLY7J3o2tqBxZXFGm8HrNpHd5WBGy/FONmmU/BpO2jz00yMy8ZI41tzbok9CQUYhkGRN1h5J90jrfbZlOtqRCeZz3ngVWY8gWIz2WxMq3aEqGdh1wHRWucDj4duQohDEaEJiBV3sVV+D0X+IBqNoQGlyp9X8N43O5mxdDsAt43uwSUDY3mNfVzj+JnhxjdY8YNWoDR+bWeecRJLLCdT7EjEpf2M5DsKLeWTAqO63p8jmWzbHOpq2JyRT4aEiJBwh2CEEEeqzwXgTjAnGNakDicglt3FFiDDA6BL/qtk5pe/MWPpdhRw59hjOTe1Ay7lY3LwA8bxBX6Lg/3Ec0DFcYB4fMrOBOuXPGN9gaP0Hjrp3TgIoC2Octe1VJco2JwQ8Jm9AIdK6moI0aRVm4AopZYqpXqW+b6m25KGC1mIJioCExC7tYnGYbXgCxgEDc3qgjYEsGPV/kp71Kav2s0by3dgUXDPuT0ZfXwKbu2hg8omoCGbRALYy93Pj51sEnDh435jGok6B5vVUjpJRGuNRSminDVUQD3cXomyw1rhqKNhLSFE3aipB6Ts3xlL6OfqbtKTIkQ4GngCotNmZWz/FA54fBT5gnixsZCTSaCgUtsze7fh2Pax3H9eL87u3xaATmRSqNzkEls6jGJRlXtP8okhmkJS+RmH5eCfDn/QIDnOibWmoZLD7ZUoGdaqrUepRPF+6HO+zLEQopGothKqUqqD1rrF1veQSqii3pWUDa/nCYg5BV5ueX8NOQU+dh/w0Io8njSexYmPXB19sJgY5pJcm9Ws55GoCuiu/uAX1YUi7Sr9SKKAoGF+b97X7MRwEKCVPYhFgc/iotiwYrEo+raPx2Gr5jPKkVb7LMw2l/P6PbXX1bC7I18HRIhGriErodbUc7FRKXVdQwQhRItUMgGxY6r5tZ4+mbeOcfLkef1w2S34gwb7iOPhwLV4tJ1XP9vAs/N+Ll0m67BZcKkAbVQuCs3vOgkPbpSCaIcNl80KKLMXRGvQGqXAZbPidrmxqyCbHX1x+g9gsSh6toutPvmAI++VkLoaQjRZNSUgc4BpSqnFSqnODRWQEOLIeANBfs7IY/XO/fyckYc3EKRLUjT/uWwQybEuNLA9mMQFnziZvSqHxeuzyM3KJMmSTyJ5OAgwzzKUF4Lj8WoHhHZ6sVstuOxWYl02nHYrMS4b7RPcxDhsWC2KgKHxG5qf7H2Jjk+iXytNlKOGhXZ1NdlW6moI0SRV+9dBa321UuodYBqwQSl1v9b6xYYLTQhxKCpuNqcwUweH1cLY/imM6ZfCtad15f3vt7Mq/Wmyf1qGsjnpdtnD/K99G5z48OLgNzrgV3Y66W3YlcLQ4LAq/KFxF4syS623jXXhsFkIak2RN4ihNXavn3vGnYnDNcYsMlawx1xqW7aHI+A1ez5cCXXXKyF1NYRocmqsA6K1/lwp1QeYCjyrlLoo9L2virZL6ydEIURtKm42V7boly9gMHvlLhZu3MNtQ7ux8b2pZK7+HIvDRZdLHye6W3+2VLieP2jwh6UjyuqglRWOSorDYlGlK1rKTiq1KmU+XsALFje0PdZ80x8/wywytnGOWQCsZC8Ym8PslaiPHWqlroYQTUathci01h7gdqXULuBpYD7lKzGXfNCqYZ3dkVNKLcCswvqk1vqhMscTgX8C5wFu4Dvgdq31+gr3d2EWUbscc/fetcC9Wusv6zNuIepbxc3mKnLYLCTHutifX8QFEy7i9zXLsbmiGHj9Pylu3R1v0MAeSi4MbQ6jWJUiIS6GHzmTSx1f4Yx2VPHIFVQsnia9EkKIGtSagCilooGngBuBr4AnqKIHpD4ppS4B+ldxXAGfAF2BvwD7gfuBL5RSA7TWv5dpPgMYA9wNbANuAhYqpYZordfW7zMQov6UbDZXVfIB5l4thb4gQb+P3Ow9RMXEMTd9PhnOjsxd8weZ+V72F/rQGpQFEtwOkmOdnD+wA2O734RzwcYj271VeiWEEFWoMQFRSp0DvAIkArdqrV9qkKjKx5AA/Btz1913K5w+FzgVGKa1/iLU/jtgO3APcEvoWH/gUuAarfXroWPLgY2YG+mdW+9PRIh6UHGzubJ8AYO9+cVk5nlLa3h0mPgE/rwsdtrac16/FC4+oRPbsgrJL/aTne8jKdZBrMtOtzbROG2hTs0x/2rY+RxCiBah2gREKfU2cAmwBLhea/1bg0VV3j+AjVrr95RSVSUgu0uSDwCtda5SKh0YRygBCbXzA7PKtAsopd4H7lNKObXWEd4iU4hDV3GzuRJFvgCb9+TjK/ZwYNV82p08HmWx4IiNp9gdw/srdrF0cyZPntePXim1bOZWssqkoedzCCGatZp6QMYAk7XWMxoqmIqUUqcCV1LF8EtIH2BDFcc3AlcqpWK01gWhdtu11kVVtHMAR4e+F6JJqbjZHJg9H5v35OP3FLLzvYcp+G0DwaJcOp5tlvWxKEW8247Xb/Dg3PU8P/F4WsfUMidD5nMIIepYTQlIH6317gaLpAKllB14FXhaa/1LNc1aATuqOL4v9DURKAi1q6pec0m7VhVPZGVlkZp6eMXgJk+ezOTJh7G9uBBh8gaCbN6Tz3dbcsgp9GK1KGKdNiwWxd78YrxFBex8968U7tqEPS6J1sePJGAYaG1WO0VBvMtOZn4x89dncOWQLuE9sMznEKLRmzZtGtOmTTvcuzdYV2ZNdUAilnyE3Iu5quXJGtqUrMCp6vjhtCvVpk0bpBS7aGxyCrx88OMu3v9xF9kFXoKGxhsw+H2/B4dVkZLgZm9WDr+9/SBFu/8PR0Jbjrr8b3hj2uL1BtCYQye/7s2nbZyLxCgH89ZlcPEJnQ7O+RBCNGlH8iFYKZVdx+FUq9ZVMJGglDoKeBC4DnAqpcr28TpDE1PzMXswKvVeYPZ8wMFej33AUTW021fFOSEale3Zhdz5wVo278knEDTM8umhtNrQUBzQbN2Zwd5Zf8WXuQ1HYgopl0zFiGmDQqOUQmtw2c06Hhm5xWTle2kX52JbVmHtc0GEEKIONdZdbLsBLuBtzCSi5AZwV+j7fpjzNvpUcf/ewM7Q/A9C7boqpaKqaOeDSnWYhGhUcgq83Pu/dWzek483ECRgaAzDTDzKdu3tW/Y6vsxt2Ft1IOWyv2ONb4PFokqTDwvgtFuwKIXTasEwNLv2e9ibVxyppyaEaKEaawKyFhhaxQ3MpGQoZtLwCdBBKXVGyR2VUnFAWuhciU8AOzChTDsbcDGwSFbAiMZu7to/2JpViMcXJBhKPEooDo4lthp2HdG9zyT5kr+holubu9xqs70CYlw2LGVGHu1WC0FDs2K7dAIKIRpWoxyC0VofAJZVPG7WHeM3rfWy0M+fYFY+fVspdTcHC5EpzOW7Jddbq5SahVlO3o5ZJ+RPmAXMLqvHpyLEEckp8DJ37R/8e/H/UeANljtXkoMECw9gccdisVixuGJISrur9HxJ4uGyWcyejwrTngytsVsVK3bswxsIyjwQIUSDaaw9IGHRWhvAWGAx8DLmDr5BYKjWeleF5lcDr2NWcp0PdAJGaa1XN1zEQoRve3Yht7y/hvdW7MQwqm4TyN1Lxsw7yfnsBQxduZHbbiXebcdtt1ZKPgD8hqZdvItAULMtq7Cun4IQQlSrpkJkBlWvHKmS1rrePzpprSv9BdVa7wOuCd1quq8HuCN0E6JRK7u/S4Lbwe/7PZXa+PfvZu97DxLMz8IfFYf2FaOcFac56WqXevkNjc2i6BBrIb5wO5Y/ikG1ldoeQogGUdMQzGMcTEAU5hu8G0gH9gLtMHsfPJj7rAgh6kjJ/i5J0U6yCrwEguV7N/w5u9j7/oMEC/bh7NCL5AmPYKmUfIAvqHFYtTkXJKRkw7k2Ko/LY9ZwStaXEPTRaUU0rLGA1QF9z5fqpkKIelVTHZBHSr5XSj0E/AaMLFtNNLRR3UIgUI8xCtGiFPuDzFn9B0XeIGv2HSBoGATK5B++rB3snfUQRuEBnJ36kjz+4SqTD4DEKLPiaaDMGI5FKQbF7OMu/6tEFxdQoGLxWaNwxieYJdYDXlj7Dmyeb+7v0rp7PT9jIURLFO4k1BuAmyqWMtdaFyqlngZeoOaCYUKIMH39aza/ZhagFNgtCqvVgiKIBnzZO9n73gMYnjxcnfvT5oK/YnFUvQsuwFGtoohx2ij0BTEMjcWiaGfN54bsadi1j1xra3yBICkJTqwlvSQ2J8S0M3e4nX+HuQ+M9IQIIepYuJNQkzD3TKmKA2hdN+EI0bLlFHh5fumvGFrjtJr1OoJaYynJDWKTsCWk4Oo6iDbjH64x+bBbFUHDHH6JddqId9uJddoYXLScKCOfImssgaCBzWqhbWwV13EnQvEBcxM6IYSoY+EmICuBR5VSHcoeDP38CPBjHcclRIs0f30GRb4ANsvBORtam0vQLQoszijaXvwYyRc8hMVe/URRpWB033a4HVYy84vxhcZwbIaPEwqXm8MugSAWi6Jnu1gctmr+FLgSzR1wA1IqRwhRt8JNQG4B2gNblVLLlFKzlFLLgK2Yk1Fvraf4hGgxiv1B5q3LoG2sC4tSGNqcA160ayNZC17AElpma3FGo2z2Kq+hAKuCKLuVSwYfxfMTj2dCaie8AYOcQh/u/O0Q9OHDTkqCm77t44ly1DASa3NCwGfugCuEEHUorDkgWus1SqmjMZewnoRZBj0DeBr4t9Y6p/5CFKJl2J5diC9oEOtykBznJCO3GN/On9j2zl8xfMXYko8mdsCoKu+rAItFYbMonDYLXZKiOb5zIk6blSuHdOHiEzqxLasQyx/FdFoRjTM+4eCcj9ooBQEp1S6EqFthV0INJRkP1mMsQjQ9Aa/ZO+D3gN19RDU0PP5gac2OtnEutq39jm3vPoIOeInuexYxx40oPV+xQI/LbsFhs9I2zonVorhwUMdyVU2dNqu52Zxqay61DTf5AHMMyFb9XJMmqQ5/b0KIw3NIpdiVUkmYPSCtgXSt9T6llAvwhaqSCtEyFGabkzM3zIGgz3xD1/qIami47dbSxGLvhu/Y/u4UdMBHTP+zSR51MwYWNAf3filp67JZsFst9G0fhzdg4LRbGNMvpeoHSephxhjwhveGG/CCzWHerzmoh9+bEOLwhDUHRJn+CfyOubHbf4EuodMfIz0joiXJ2QofXgtr3jHfxKOTIKq1+dXmNGtofHit2e4QdGsTjcNqYcuPX/D5C3djBHwkDBpDx7G3oCzW0pUwOnSzKLBZFNFOG0Gt2Z5TSLQtwD9OtdJ6/0+wZ0PlyaM2p/lG69lf8eGrVrwf+pzfPHoH6un3JoQ4POH2gNwP3IxZHXUx8EOZc+nAFcDjdRuaEI1QYbZZG8Pvgdh2lc/XUkPDGwiyLasQjz+I226lW5vo0qGSguIASTF25s9/Bx0MEH/CeSQOuxa/oXDYLNgsCg0EAgZ+Q2Nos5S6P2iQpPI5L7iCP/MTzq8CNX+y73OBWWTMs99calsdz35wJZjtm7oj/L0JIepeuAnIdcBjWuu/KaUq7vmyBZBSiaJl2PgReA5U/SZWljsRCvaY7QdPJqfAy/z1Gcxbl4EvaKC1pthvYFWKM3u24cxj2/Dckl/ZV+Cj6yVT2LduKe7+52CxKLTWeP1B/EoR47ThdNrwBYJopeiWFE1n9nBD/ks4A/n4aY8zOuFgHFVVNY1OMr+ff4cZoyuxfA9HwGv2fLgSzHbN4Y34MH9vQoj6o7Sufb85pZQXc+fYL0IJiB9I1VqvVkoNA+Zrrd31HGuDSk1N1StXrox0GKIx8RfDzNBwRLjzJ4Jeto9+h4fS/488j58oh439RT4y87wYWqM1+A2DvG1rOarnQHqkJBAwNBt251LoDZgrVUJDL4bWKBROuwWH1ULPdrEkWwuYnDUVu+Ell2iOaRtLvLuKJbqe/eZky7Kf7EvmQ2ycYy61Lek1sTnMYZfmMh/iMH9vXDG3eQw9CXEIlFKrtNapDfFY4faA/AH0Bb6o4lx/YHudRSREY5WzxZy46IoLr73Nid+Tx6sfLsCruhDjtLF5Tz4BQ2O3KGwWcwpWzg/zyZj/Avl9huK58F56t4+jR3IMm3bnETB06WxThdkbEuWw0rNtHA6bhRPylhFl5JNrbQ1Bo9ymc+VU9ck+Osn8fuBV5oqQQLG52qW5rQg5jN8b3nzzNWnXt35jE6IFCzcBmQ08rJRaDXwfOqaVUscAdwLT6iO4SMrNzWXy5PJdsGlpaaSlpUUoIhFxfk+Vy1eDWlPkDWJos+x5lNNaWmPjQJEPvy7CHWdlw+5cDMMssV5i7w8fkzH/JQAcbbviDRhs3pNP75Q4XHZraadE6eoXBf6AOfejpKppoSUOXeaxq1VS1XTgVeUTDJuzeb/RVvN7q5HUPhEtQHp6Ounp6RUPxzfU44ebgDwCnAx8ibkrLphJSSfgW+DvdR5ZhMXHxzNtWrPLq8SRsLvNbCDEFzDYm19cOpxSwqIUyXFO2sQ42V/kx5kYzd78YgIVk49v/sfvC81/Y0kjbiAh9VwChiYQNMgq8JYWIyt7HwCfYVDoC3KU5Xds2o/HEo0/ECQlwV1zcbGW+sm+wu8tLM2x9okQFVT1oXr69Om5DfX44VZC9SilzgQuBUZiTjzNwVz58o7WOlBfAQrRaJSpoVFkWM3hlKCB3WrBajmYJGityTjgIftAPjbDSqazE5lZRdjL7O+S8eV77P78dfOyI28iYdDo0J3NfV8y87z0aR9HVr4Xf2jIpixDaxxGMaBq3lCuopb4yb6l1z4RopE6lEqoQWBm6CZE41Pf1S1DNTT8K99mc54Lw9A4bJWHPJRSOGxWonz7+SBwOjnFKrRkNjTnY80iM/lQio5pt+Psc1al+xta4w9qeraLY/OePLxBA7tFlc7xsChFoeHAbxhYbLVsKFdWS/xkX1L7ZM07ta+CAXMF0IDLmtc8GCEaobASEKVUEBiitV5RxblBwAqtdQ2Dz0LUo4asbtnnArK+/x8ufx4+R/VDpVHBfIqtsXzqG4K1oHwxsITepxKzegFJqaNp1X84uR7/wUkeHKxyamhNrMNG3w7x7M0zh3p8wSAaKA4E2WFtT0JsNN3iXDhq2lCuREv+ZN8Sa58I0ciFuxtuTTO4rFTemkKIhtHA1S2LHYn8zXIDhs1FfHAfNu0HIBiau6GCPuKDOfgtTt5JupV8azzZBT78gSB+fwANWJ1RHHP107TuPxwFOG0WDCj9v8hqOdjLAeCwWuiUGMXxRyWQEu/m0sFH8dT443j9+lNpN+QSHL4wh2ybU1XTQ1VS+8TuNlcDVawQG/Cax+3u5lP7RIhGrsYERCllKVN4zBL6uewtGjgHyK73SIWoqGJ1y4pvrCXVLf0es13hkf8z3Z5dyC7VjunJD/Jl1AgMfzEWTw427z7s3n34vB7e9Z7GQ9bbWJmfSCBo4PEF2L3gFbb+7+8cKCzG4zeLiJVw2ixYMFfTOG2W0hUt0Y7ynYr5xQGS45zceEZ3eqXEmRVU+1wA7oTaS6vLJ3uzCNv4GebwStBr/nsoyjG/Br3m8fEzzHZCiHpXbb+tUmoK8HDoRw18U8N1Xq7LoIQISwSqW5bsWJsZjOXZvDMhOITOxm6shpdi7GzX7fFqG8EcUBQS5VDsW/QS+WsXgNWGb+82SOmBL2AQ47JhVea8Dpfdal5bmatr2ie6sYR6QnwBgwMeH3FuO0+e14/WMWUSrZZY1fRItJTaJ0I0ATUNHC8LfVWYicgMzM3oyvICm4B5dR6ZEDXxF5tzPmoazy+ruhoYh8htt+IPGuzIKSQQ1BT7raynk9mVWKZiKYBhBNk55wUKNnyOsjloe8GDONodjSU0RaWgOFC6kZzNqhjQLoG9ecXkevw4rBb2FfrQmEMwE1I7MaZfSvnko0TJJ/uSqqbe/PJVTQdc1nyqmtaV5l77RIgmoNoERGu9HFgOoJTSwHSt9e6GCkyIGkWoumW3NtHkFwfwBoL4A7p0V1oAw9AEQ/M4tBEke/6/KNq0HGV3cvKNT2HrdBy/7/cQNDQKMACPL0i7BBfJsU6K/UGOah3FlLTeBA0o9gdxVdiwrlryyV4I0cSEuwz3ZaDKj5qhaqj7tNYyD0Q0nAhVtyypZxUIagzMSVSGNiehlrYJBshOf5qiX75GOdwkXziFnPhjGBDvJjnWya+ZBeR6/KX7wGTmeckp8HHGMUncPvwYjm4be/gByid7IUQTEe4qmJcxS65X5XZkDohoaBGqbrk9u5Bop5WgoTEMTcDQ5ZIPAG0ECBbuRzmiSJ7wGK5OffEGDDb8cYD/21tAwNDEu2zEu2247Fa6t4mmf8d4du4rYkr6RnZkFx5RjEII0RSEm4CcCiys5twi4JS6CUeIMJWtbhmOOqqB4fEHKfYZpfuzVJUCWewuki+cQrvL/o6rY6/SNeyF3iDFgSDOUOVUm8WCzWJOQHXZrSTHuvD6DR6cu56cgjCflxBCNFHhJiCJQHXFBvKA1nUTjhBhKqluWdvy0xJ1VAPD6w+SkedBKShbeNTwF5P73QdoIwiAxRmFI7lb6fmSRCUQ1JWSlrI72Ma77eR5/Mxfn3FEcQohRGMXbgLyO3BiNedOBOSvpWh4EaiBsSkjzxz5UQpQ5mRSn4fM/z3KgS/fYv+Sqjcw1PrgZNWSIRujmnofCW4H89Zl4A0EjzheIYRorMJNQP4HPKCUGlP2YOjn+4AP6jowIWrVwNUti/1BFm7YS0qCC63NngzDW0TmB1Pw7lyPNaYVscePrXS/0h4PVf5nv6FJjnOW1vso4bBZ8AUNtmXJXBAhRPMV7iqYx4DTgU+UUnuAP4AOQDvge+DR+glPiFo0YA2M7dmF+IIGHROj2JtbTGFBHntmTcGX8QvW2CTaTXwSe6sOlYZYStILpRRamz/7gwY2i6p2B1uFmfAIIURzFVYCorUuUkqdAVwBjMCc87EFcwLq21rrQP2FKEQtGqgGRkkVVIfVwtHxikWvPYRvzxasccm0u2Qq9gSzImvJZnIADvwcbcnAjRcvTrbqFIKGBZvVUuMOthpw2WV/RyFE8xVuDwhaaz/w39BNiMannmtguO3W0sRiy2f/xbtnC/bEFJInPoktLrlc2ySVxzmW7xlr+R6nJYBhKJQyCFjtfO8+g58Tz8JbzQ62voCBw2qhW5voensuQggRaWEnIEI0B95AkG1ZhXj8QdzhVhkN6dYmGofVgi9gkHrRzQR8HnqOuZ5fC+14A0Zpu64qg8dsrxOnCslXMQQccXj8QQJBTWs3jDW+YNj+H3m79a1k2jtUepwDHh8TUjuFHZcQQjRFNW1Gtw04X2u9Tim1napLHpTQWmvZQlI0WjkFXuavz2Deugx8QaN0mMRhtTC2f0r1+6yUcSAnm1F9kpizdg/JsW5Ou/ZhfEGDdlYPO/cXojW0Io9Hra/jxEeWTsRltWBoTZTDitbg05BrbUVUMJ/Lc55jWpsHKLAmlD5GrsdPnNvOmH4p9fp6CCFEpNXUA7Ics8ZHyfeHWHZSiMZhe3YhD81dT57HT4LbQazr4D97X8Bg9spdLNy4hyfP60eXpKqHPXbu3MmwYcPo068/MaNuJ9fjx261sHlPHgFDE+u04w8ajDG+J14VkqkTAG0mKPFRdEyMImAYbN6Tjy8QRFtjSDD2c0LhMr6IO6/mHW+bioDXnIPj95grj2QfGiFEDWrajO7qMt9PapBoGpHc3FwmTy6/dXtaWhppaWkRikgcjpwCLw/NXY/Xb5AcWnFiaE2hL4hhaCwWRVKMk/ziAA/OXc/zE4+v9Oa/fft2hg0bxo4dO4iPj+f14V15YvFv/PT7ASwWhdNqTiSNUkHONX4gV8dgtSii7DaUggMePx0TIcpho2/7ePbmF5OZ5+WAjmZA3jLmWkZgsTlr3vG2MSvMNlchbZhjbhBYsgrJ6jCLxclOvEI0Sunp6aSnp1c8HN9Qj6/0oe6n0UKkpqbqlStXRjoMcYTe+m4Hs1fuIjnWhS9osDfPfPM3yvy7V0BClB2lFKP7pXDr8B6l8y+2bNnCsGHD2LVrFyeeeCILFiwgISGB/yzbwszvdlDkM0qv1YPfmMI0imwJOG0WVKjCqS8QJCXBTafEqNLHDGpNkTeI8mSTecY/6NAztWnO+cjZCvPvAM8BcCeW7/EIeM0KtK4Esw5LaxmlFaKxU0qt0lqnNsRj1TQH5MpDuZDW+q0jD0eIulPsDzJvXQZxLjtZ+V62ZxcQ1OC0KhyhuRnegIE3YJDvDaCA17/ZwYrt+zh3QHuOceRxQdooMjIyOOWUU/j000+Ji4uj2B/ki81ZdG8Ti82izN4UrTnWbyOqwIZhLZ9I2K0WMvO8tE9wYw0lJValzKEgw05MghWaYvJRmG0mH34PxLarfN7mhJh2ZiXa+XeY9VqkJ0QIEVLTHJA3Kvxc8pFRVXEMQBIQ0ais2bmfPw54yC/2U+gNAAqlNL6AwqaC+IJlekGUWSjMFzTwBoK8/um3fP/czRTn7+fMM88kPT2dmJgY4GBBspK5JCVfbdZoVEHlHkWlFIZhUOQNlpt/AtTJDr0Rs/Ejs+ejquSjLHeiWZF240dmvRYhhKDmUuxdy9xOw9wP5lXgTKBX6Os0YBfmbrlCNBrbswv522eb2VfoIxA0QCksFvP9PmBoioMaA0pvwdDxoKHZV+jjqI6diG3XhXa9B/PWrI9Kkw84WJCsoj32jgSUHZv2VxmTUXG4s4526I0If7E558OdGF57V6JZqTbc3YuFEM1etQmI1vq3khtwN/C+1vrPWusvtda/hL7+CZgF3NNQAQtRm5wCLw/M+Yn8Yj9KYdboMDSBoCYYxpSnjAPFFGNj1B3/5vhrn+SLreU3gi5bkKysgHLwY/QZRBt5VZwtv+stUGc79EZEzhZzwmm4sducEPCZq2SEEILwN6M7C1hczbnFofNCRFxOgZeHP97Aul25ZOZ5KfYHCWoIUvs6cu8fm8me/y+CwQAbd+ei7S5ax8VW2pm2bEGyin6MPpMiSyxRwfzSYzq0622Us8w8jzrcoTci/J7QjsCHQCmzTL4QQhB+AuIFqpsVewLgq5twhDh827MLuWHmKhZv2ovXH8QbMDDCXORVvGsDez/4K4UblpK/5lO8AXPFTFU70zptVsb2T+GAp/I/+wJrAm+3vhW/xUl8cB827ccfNEiOc5oTUOt4h96IsbvN8axD0ZTnuwgh6ly4CcgHwCNKqbuVUl2UUu7Q13uAKZjDMEJETE6Blzs/WMumDLMwmEUpwl1i7vltHZmzp6B9HqJ7n0nswDEYhmZvbjGG1lXuTDumXwpxbju5nsrzPTLtHZjW5gG+iTkbFfSSSD7tbIXmqpGg19yhd/yMpr0sNamHWecj3DkdTXm+ixCiXoS7F8ydQCzwN+DvZY5r4N3QeSEi5t0fdvJzRh5amx+0/eEmH9tWkTXnSXTAR3Tf4bQ+5y8oi9UcLTDMgmVV7UzbOsbJk+f148G568nMLybB7Si3s+0+HccsNYrFySOZepobe4yqlx16I8bmNIuMrXmn9lUwYM53GXBZ83juQog6EVYCorX2AFcopR4HTgLaARnAD1rr/6vH+ISoUU6Bl7lr/+DFL7bgCxiHtF9A0ZYVZM2dCsEAMf1H0Wrkn1GqJIlQaMDrD+K0WavcmbZLUjTPTzy+dI+ZkloiJXvMNNnKpuHqcwFsnm/OZ6lpNUxTn+8ihKgXh7QbbijZkIRDNAole7z8vq+o3G604dBaU7DmUwgGiB04lsThN5RWLi05r4AiX5DxgzpWW6W0dYyTK4d04eITOrEtq5BifxDXIe6y22RFJ5nzWObfYc5rcdVSCbWpzncRQtSLsBMQpVQ0cC1wOtAKuEFr/atSaiKwVmu9uZ5iFKKSkj1ecov8ZBUc+hxopRRJ591H4YalxAw4p1zyUUprkmKdYe1M67RZ6ZUSd8hxNHmtu5vzWTZ+ZNb58OYf3AvG5jCHXWQvGCFEFcJKQJRSnYBlQEdgM9AXc04IwFBgOHBdPcQnRJXmr89gX4GXPXlejHCXugCeHWtxdeqDstqx2F3EHj+62rZt49387fwmujNtQ4pOMiucDrzKrPMRKG5e812EEPUi3B6QZzCX4vYAdlN+2e1y4JG6DUuI6pXs8eILaAKGxmYJrx5FwU+LyPnsBdzHnESbcfehLNUPkcS6bUy7YhBdkirP/RDVsDmhXd9IRyGEaCLCTUBGAJO11juVUhX/av8BdKjbsISo3vbsQooDQXIKfdgtikAY98lf8yn7Fr0MgLNdjxqTD4cVjmoVTafWUdW2EUIIcWTCrQPiAPKrORcPVL35hRD1IK/YT16RH1/QKK3TUVMfSN7KT0qTj8Sh1xA/5KJq27psioFHtcKiVLniY9UKeGHPBtj1o/lV9joRQoiwhNsD8hMwHlhQxblzgFV1FpEQZXgDQbZlFeLxB/H6g2zKyGPOmj/IyCsmEDTwBaCm9CP3h484sOy/ACQOv4G4QWnVtnXbLQzolEiUw4o34KtUfKycwmxz4uWGOeaeKCUTL60Osz6GTLwUQogahZuA/BP4X2ilwLuhY72VUuMwV8acWw+xiRYsp8BbWl/DFzTw+oPs2u/B0Jq2cU5sVkXAUKH3/aonoRZuWl6afLQaeROxA86p9vHsVkX/jglEOcyhmaqKjx0Mbqu59NRzwKx/4Sqz+iXghbXvmPUxxvyraVc7FUKIehRuIbKPlFJ/xqyCek3o8FuYwzI3a62r6hkR4rCU1PfI8/hJcDtw2iz8llOIVZkFvrILfBiGRmtNyQCM1WJWLgVKi4G5e5yEq8vxRPc6nZjjRlT5WBbAYlG47FbcoeTDFzBwWC1VFh+jMNtMPvyeqiuA2pwQ084svjX/DnOJqvSECCFEJeEuw40HXgdmAkOAZCAH+FZrXd3cECEOWUl9D6/fIDnW3Lhs1/4iAobGaTWnLNkt4AuAocGo0PuhtUYbQZTVhsXuJPmiR8tUNz1IARZl9nI47RYCQbPseqzTxgGPjwmpnaouJLbxI7Pno7by4+5EszjXxo/MJapCCCHKqTUBUUrZMJON87XW6cDn9R6VaLHmr88gz+MvTT6ChiYzz4s9tNQ2qDX5Hj/BKkZdtNbs/2IGgX1/0Ob8B1BWO0pZSiep2qwKi1I4bBYsSmG1qDKzRzSGocn1+Ilz26suPuYvNud81FR2vCxXolmca+BVUg9DCCEqqDUB0VoHlFJ7gRpm5DU/ubm5TJ5c/pNrWloaaWnVT2IUR6akvkeC21F6rMgXxNAam8WCEUo+AlUmHwb7P59G/up5YLHhzdiCu2MvwBxi6dshjii7lc178kO75ZafumpoM/lIjjM3mauy+FjOFnPCqSvMiqc2p1kZNPtXqY8hhGh00tPTSU9Pr3g4vqEeP9xJqG9jVjr9tB5jaVTi4+OZNm1apMNoUbZnF+ILGsS6Dv6zLDvEUuwPVpt87Fv4EgXrFoLVTpvzH8DVsRcaiHZYsVoUO3OK6Nshnr7t49mbX0xmnpeAYYTuD6CYOLgT5w3oUH3lU7/HXO1yKJQyK4MKIUQjU9WH6unTp+c21OOHm4DsAC5VSv0IfIy5E265twKt9X/rNjTR0nj8wUoLai2hN3wNePyVN5zTRpCcz56ncMMSlM1Bmwsewt11YOl5h82CVSm8QYO9ecV0SoyiU2IUHeLdFIZ6Vw54fFw6+CiuOfX/27vzMDmqcvHj37eqt9lnkslOQggEAyEGNCIgiiugIUAQBERFEOO+cV1AubII4opexJ+Ky0UUBWTTEERAEb0KQlhDBAQSyEKWmcmsPb3Ucn5/nJpJT6dnyyzJzLyf5+mn6apT1VUnQ/fbZ3nPvL4vMF7WFa0MnDE2LblSSqkeBhqA/DB6ngW8tsR+A2gAooakLO5S/PVekXARgUze32WfCQMaV11F578fQOJJpr77YlL7vrpHmY6sR01ZgrgjbG/LMau2DEcExxGqUjFaMx7TqlOcdOgAkvnWz7d5PvzcwMZ0+Dm7IFv9/P7LKqXUBDPQAGS/Eb0KpYB5UypIuI6dBhuzz9vas+S9kIy/a+sHYUiYaUcSZUw97RJS+yzs3tU1FdcPbTKzuGvHkHTNdMn7IS2ZPNVl8d7HfBSLJW2Sscdv6H8WDNil6A89SwegKqVUCQMNQNJAhzFGO7PViEnGXE5YPIPfrd5IZTLWPWA0GXfIB+EuM18kFmfK8q/gN79CYurOGLmwGyfhCpMqkrRmPPzA0JLOd+f5OG3JbJYumjG41W4XnmKTjGWa+54Nk2mGVK0tr5RSahe9rgUjIq6IXCIiLcA2oE1EbhWR2tG6ODXxLF00g7KEy9OvtBFGuT9ijkN5wsbKxs/T8o/fYny7/JATT/YafAC4jjClKslhs2uZPamcj71lf648ZRG/Ou9wPnDk3MEFH2CTii29yo4H6di669ovfs5uj5fZcpqETCmlSuqrBeSjwFeBvwKPAPOA5UAbcM6IX5makCZXJjlyv3rWbm4DRwiNsbk7XMF4WbbfejnZl5/Ab9lG/dLP9ji2MPgw2ERjrmPHe/ihoaYszrsWzSidYGxQF7m/zXC69jab5yPXvnMtmFjCdrvoWjBKKdWnvgKQDwM/NcZ8pGuDiHwEuEZEPmKMyY/41akJoXDBOVeE/3uhgXlTKmjNeDR3egiGIJ+h4ZZLyW5Yg1NRS83hy3ucozj4AEi6gus4VCRcGjtyvWc33R0V9TbD6WvOtnk+/Kyd7VI/X8d8KKXUAPQVgMwDPl+07SbgR8C+wPMjdVFqYihecC4IQ7a15WjsyJFwHeKuAwJ+Ns3mmy4ms+Fp3MpJTDvjCuKTZ3cPNC3U9TrmCI7rMLU6SXvW7z276VDFkppkTCmldkNfAUgltrulUNe6L1Ujczlqoli/bQc/ufVu/GyaA1KVbIzN4ultWXJ+iABeaAjCgJTJ8PKvv0Ln5ueIV09hyce/xzapJQx7Bh9d/y3YbpdU3MV17H8n487AZ7oopZQaFf3NgpklIoXZmdyC7S2FBY0x64bzwtQ4lW6k49Gbyf7fDZwb+riOYDKGVk/4E0dxr3MkG6UcEZsFdeNffkPn5ueI1Uxj+plfZ5vUkYo55P2gx6wYVyRK2S5IlLxs38kVnPrafQY/00UppdSI6y8AuaWX7XeU2DZMnetq3Gp6EVadT7ZxO5kwgROvBGwG1MBkOVHu5y08zEWcy8vhNEIDNW98H2G+k0lHn0G8egqhAYMhGXcJAgNiWz/m1JWTiDtg4Mj9J/O2BVM5bN+64RvzoZRSalj1FYDoTBc1fNKNsOp8glyGl3NVODG7em3WD+nMB0Cc7aaWajr4bPZH/LfzSVpi9TixBJOP+yQx17ZqOEDeN9SUxfAdgx+EnLZkH046dBapuMu8KRUadCil1BjQawBijPnlaF6IGufW3gaZFjrj9YSmDRA6cn602JxBonksL7bF+dKvHiNedwnm5O8hsThgZ7iK0J3iNAhN97os06pTHDanj6RgSiml9jq9JiJTath4WXj6diirww9D/NDQnvUwxuAI3cHHtpYsn7v+cV7ekUM6thPz2rDhSU/GGPJBiOMIB02v5v5nG8j5wajfllJKqd23VwYgInJqlHX1ZRHJiMhzInKliFQVlasTkZ+JSKOIpEXkPhFZVOJ8KRH5tohsic73oIi8afTuaGLJ+QHPbGnjsQ3NPLOljfz2/+B5OTa2Bzy/rZ1MPiAIDaEx+KENKF5p7uRz1z/OluYsB86o4ifv35/55enucxpjMMY+A0ypTHLIzBpqyuPkg5B1DeneLkcppdReaKBrwYy2zwMbgC8Dm4DDgEuAt4jIUcaYUOxUhz9gF8r7FNAMXAjcLyKHGmM2FZzv58BS4AvAOuATwJ9E5EhjzBOjc0vjX3Fej648HfNzz/DBjg46HIeY62AIumeqCIYNOzr54q+eoqE9x0GzqrjyzFdTk8pQ5ue7z1GRjHX/t2AXrnO6zwFZT1tAlFJqLNlbA5BlxpiGgtcPiMgO4JfAm4G/ACcCRwNvNcbcDyAiDwLrgS8Cn462LQbeC5xrjPnfaNsDwFrgsug8aojWN6a56I41tGU8qlNxwK4+G4aG9W0hngnoDAISoUEwhMamSn+lOcPnr3+Spo48i2bXcMUZh1CWjIEYsiS6U6o7IsQcIReETK9JdQcfYIOSVFwHniql1FiyVwYgRcFHl0ei51nR84nAK13BR3Rcq4isBE4iCkCich42i2tXOV9EbgQuEJGkMaZoRTE1GE0dOS66Yw0dWZ+cH7L2lbZocGnUMhFMJReP4ZAnE8a7jwsNTKpMsv/0Smb7IZe95xDKEi4JPPImzotmBkK0zAo2OVnMEaZVp7rP0bWy7bwpFaN810oppYZirxwD0otjoudnoueFwNMlyq0F5ohIZUG59caYzhLlEsABw32hE83vn9jMxh2drGtMs6k5gwgkXIcwNOQDQ544d4ZHUCe7jtNIxBwuPnUhl59ugw+AGkmzKjwCjzgxx0598YIQR2DB9GoS7s4/25ZMnhMWD8MCc0oppUbVXtkCUkxEZmG7S+4zxqyONk8CXipRfEf0XAd0ROWa+yg3qdR7NjQ0sGTJkt263hUrVrBixYrdOnYsaerIcccTm7nqnv90j8EQEXJegAB+wfSVVcERHOuspoY0/9occMu/NvHFExeQiDkkYjsDihrStJkKVgVH2Gyo0aiPWbVlTK9J9Qg+WjPeyK3xopRSY9S1117Ltddeu7uHj9oy3nt9ABK1ZPwe8OmZHK3UWmRd24tfD6RcD1OmTGH16tV9FZnQusZ8bGvL4gUhAI4jGGOibKU97aCai/xzOX3L97ngt/8mnQ85YFolZ7xhDgBxPGolTbup4GvmPFI1U6nI+eT9gJm1ZcyZVN59rrwf0pLJU10W1zVelFKqyFB+BItI4zBfTq/26gBERFLYmS7zgGOKZrbsoHTrRVdGquaCcnP6KLejxD7Vh64xHzkvpDoVJwiNHadhDEFYOtoDeG5DI5/63fOEXsjbDp7Eh4+sIU47iCFv4twSHMOq4AhapJqKnM/cyeWkcz615Qma0jtnxCRch9OWzNY1XpRSagzbawMQEYkDtwKHA283xqwpKrIWOLbEoQcDG4wxHQXllotIedE4kIOBPPDC8F75+LdqzRbaMh5Tq1Jsa89GM1WEIDS9Bh+Zl56g4davYfwcFQvfwsvv+gQXh9tJhXmyJFhnZpAnjmAHJqWzPgbDLz54ODNqU6xrSJP1Ak23rpRS48ReGYCIiAPcALwNWGqMeahEsT8A54jIMcaYB6LjqoFlwG+Kyl0KnIadxouIxIDTgXt0BszgZL2AO5/cQm1ZArCtEYIQhrbrpZTMukdpuP0KjJ+nYtHbmXz8p/Acl2fNrg1TInYciePAYfvUMbfezm45aEb1iN2TUkqp0bdXBiDAD7EBwxVAWkSOKNi3KeqK+QPwIPBrEfkCOxORCfCtrsLGmCdE5Cbg+1GrynrgY9gEZmeNxs2MJ+sb0+SDkIqkS3vOJ+sFOA70lQesY819GD9P5aHHM+nYj2Pjy105ApXJGHHXIe+HPL6xhZwfaGuHUkqNQ3trAPLO6Pkr0aPQpcAlUTbUE4DvAP8PSGEDkrcYYzYWHXMONpi5HKgFngSON8Y8NjKXv5fzc9D4PHgZiJdB/XyIDWwsxbbWLI0dOV5qTHcnGvOC3jperPql59Ox72IqFx/XnQG1lKpUjLhjE5i5juA4wrqGtLZ+KKXUOLRXBiDGmLkDLLcDODd69FUuA5wfPSaudKNdlfbp2yHIRxm+DLgJOGQ5LDwFKkrPwMr5AX9/vpFv/vEZGtqzlMddEq5DJgxKTjPKvPQEyVkLcOIpJBan6tDj+7y0mCPEHdsy4oWGGTUpXBFNsa6UUuPUXhmAqBHQ9CKsOh8yLVBWB6mCVgU/B0/cAM+ugqVXweT9dx4Wre9y++ObeX5bh11ALjC0hwGJmCHnBbjSM+dHx9N/oemu75PadzFTT7sEcfrvQulKQuYFoc12WpWiPedrinWllBqnxlImVLW70o02+PAyUDV91+6WWBIqp9v9q8635bHjPT594+P8bvVGWjN5RKA87pKKu4TGBh9BtDpcVy6xjqfuoWnV98CEJPc5GHoZ71HIFbrXeXEcYcF0u+ixplhXSqnxSwOQiWDtbTtbPvpSVgfZFlh7W49cH5Mrk+zo8Ig7dvxGMubgFozliPKQkX78Lpr+eDVgqD3mbGrfcGafYz7ADjyNuw7GwIyaFIfMrKE8EdMU60opNc5pF8x452XtmI/+go8uqTpYezt/9N7cneujPecTGkMsGqPhiFCZitGe9cHY3B/ND/+eHX/+KQB1bz2P6tedDGAHk0anDkJDiJ2mVJVwCMVBMMyfVkVNKo4TBTiaYl0ppcY/bQEZ75pesANOBzjLhViS0MvxxGOPdOf6CEsk+HBFqErFcAQ6n3+oO/ioP/aj1B1+Mq7Y1g1jbNBhMDiOkIwJlUmXfAiuA4fMqqWuPIHjCHk/ZHt7lmTc0RTrSik1zmkLyHjnZexsl0HI+gYJst2LxHW1TBRzRSiLu3DAa0nPP4KaA1/H1NctxRXBCw3zplTQnM7TlM5jjG0BmVyZZGpVkpm1Kba0Zsl4AVkv0BTrSik1wWgAMt7Fy+xU20EITYgnyei/TfcCc/kgJO46dtqtMZjAJxmLkfcTTDvlK1Sl4sSiPB6OCHXlCSZXJJlnDFvbssRdhwvfuYDX7FtHMuaS8wNNsa6UUhOUBiDjXf18m+fDzw2sG8bPIW6S9f5MWho6aOrIY4zBDw1eECICcUdo/ut1ZF55ngPe9zUqUwnaMh5eaHAd053HwxHpXrl2UkWCK05e1J1aHSAZczXJmFJKTVAagIx3saRNMvb4DXYKbj+89A7+WfYOnn45Q2jAcWwDStcwEBMaGu65lvZH/wCOS/umZ6iau5iKZIyp1Um2tmYBIRFzaErntVtFKaVUSRqATAQLT7FJxjLNfc6GybY18p9m4dItryLmCH5oCEOb5VSwXTM77v0xHY/fBU6MKSdfgJm+kJwfMr06RVncZdE+NZx39DwmVSS0W0UppVSvNACZCCrqbYbTVedDx1Y71bawO8bPke/YwbPNwtfMeeygmlRCaMv4O4OPMGDHn35Ix1P3gBtn6vKvULb/EhukeAHlSVdbOpRSSg2YBiATxeT94d0/t0nJ1t4OuXYQwQtCmrKG33a8npszS9geVAIBmfzO9V3CMKDpru+TXns/Eksy5d3/TfncQ7sn1+xTl2L5YbP4wJFz99DNKaWUGms0AJlIKurh8BXwmrOh8XleaWzm6r9t5Bmmsd7zIQYS+kDRxBljMF4OiaeYeurFpOYswgAp18F1hfqKFHc/vY2zjthXu1uUUkoNiAYgE1EsSVPlfL5w5+PkZC7JpABt0TiPnQNOu4gbo/7EL+A1bSIxdb/u7V5ocF2xScSCkHUNaZ3VopRSakA0AOlFa2srK1as6LFt2bJlLFu2bA9d0fDoyr1x22Ob2NaWZXZtOe052+oRUjDbxfdoffAmql9/Kk4ihbjx7uCjKy1ZaOzUXEcEwZD1gtG/IaWUUrtl5cqVrFy5snhzzWi9v5hBJqmaKJYsWWJWr169py9j2DR15Fi1Zgt3PrmFrB/w4vYOHBFcR6gpi9PYkSObDwiA0MvRcPvXya5/lPJXHc2Uky8oeU5XABGOnjeZ1pzPlacs0hYQpZQaw0TkUWPMktF4L20BmQDWN6a56I41tGU8u76L2EXiEq7NWrojnSPjBbYFJJ+l4bavkX35SZyyamqOek/vJ46SkrXnfBKuw7wpFb2XVUoppQroYnTjXFNHjovuWEMmH1CWiJHxAjqyfvd+R4R4zLHJxnKdbL/lEht8VNQy7cwrSUyd1+u5HYS469CazXPC4hk6AFUppdSAaQvIOHfzIxtZtz1NR84njLrbDJD1Aruui4AXhAS5NNtvvpjcK8/iVk5i2hlfJz55nz7PnUq4hKGhKhVn6aIZo3A3SimlxgsNQMaxf61r4n/+/Dw5P4xWtDU90qrn/LC7bNvDd9jgo2oK0868gnjdzF7P64gdiOpHi9N9692LNfmYUkqpQdEAZJx6fEMzH/rlI+T8kJhj56344c7kYsVqjjqdMNtO9eHLidVMK1lGiIIPEQyGylSMc4/ajwOnV43MTSillBq3NAAZh5o6cvzXzU/iBQa3O/gw3WnVu4KQoLMViSVwEmWIG2PSOz7a4zxdZQWoLYtjotfGGILQcMDUKt7zutmjdVtKKaXGER2EOg7d8cRmtrZlScYcwBCancFHF79jB9t+cwHbb7mU0Mv2OF7oGag4juAIxBzBFcj7IdOqU1y5fJF2vSillNot2gIyzmS9gNsf34wTTZEFITCmZ/DR1si2G7+M3/wKcXEwXo5E3GWebKGMHBmSvCwzCCQRZUY1ZP0Q1xFCY5hZm+La97+WufU67VYppdTu0QBknFnfmCbnRdlJRYg7duxHF691uw0+WrYSnzqPBad/kZOq/sUJzkPExQMjiBgCJ8F9chT3ylFs9SuZUZMCgSlVSa5c/moNPpRSSg2JBiDjRFeK9Sc2tuAFYXf3Scx1wA8wgN+yla2/vZCgrYHEjPksOf0TfKPyemokTbOpIE05rmtnySSMxwnczzHmYS415xGL7897XjebpYtmaLeLUkqpIdMAZIwrTLGeD0Ky+YDtbTlyfoDnOriO4ABeWyNbb/gSQUcTyZkLeNV7Ps83Kn9BijzbTS1gu2zsTBdDMllGjnJq/DaujP2C+e+7icpJmutDKaXU8NAAZAwrTrFelYqRTQS8vCNNaCBb0PeSqqykfp998dO1HHrqCo4ue5waSXcHH24UfFhCzLHjkzucSuaVpal8YSUc3nNxPqWUUmp3aQAyRnR1sWS8gLK4S21ZfGeK9bhNsd6aybNhRydhNOjUAJNoY6n7ECfEH4JTQvygiorE9cxjC1upJY6HTxxXbPgRYkhFKdX9ICTmOlTXTYG1t8NrzoaYdr8opZQaOg1A9nLFXSxdgUVjR462TJ6uybXGQGfervESd4UQmLztIcr++UOWLp9F1qnGc8rAAddkwIFptDCZdl5w5pCjLJqqa9d3yfsBMddhwfQqEokYpNPQ+DxMP2QP1YRSSqnxRAOQvVipLpYwNDSmc7zSnCEEXIHqVJx8GIIIGEPWNyRfeYz//O5KOrIBP/xnNWcfM7n7vC4hBiFPjKQTcCAbWWvm4hOnLGFbP2bUljGtKkUiFqWKEQE/W+IqlVJKqcHTAGQv1bWKbc4LmVqVIu+HbGzuZGtr1i4sF5ULDLRkPAQ7jsM3kN30b7b87mvk8wFHL6jnrKPnADsTkQVR/jkBfGIk8DiwPI1Tty/JmEt50u3ukulmDMRSo3HrSimlJgANQPZSq9ZsoS3jMbUqRWfe59mt7fihIQhs6NEVHpiCRxgashvWsP2WSzGex1sPrueLJx1EPOZijE3LLiJ4lGGMQ0zACJSnUlTSDOUHgFMiOa6fg1gC6ueP0t0rpZQa7zQV+97Gz5Hb9BTPPfIXFrmbCL0sz25tJwwNCdchHxgcke6go1DmpSfY/rtLMF6WpYvq+OLJB+O6DsYYiBaRs2NIHBqkjrj4CEJoHDAheOnS15RthoXLdQCqUkqpYaMtIHuLdCOsvQ2evp0gm+H9bRniDqQDl5Xh67nfeQPbgypCYwh7WdI2/e8HMH6OOa9+PRefkKc1WojOAFJ0TAN1TKMVFx/T9WcQBrueNNMMqVpYeMqw3apSSimlAcie5OfszJKm5+Gf10CQh/LJeMky2p124o5DZz7NSfyVtwaP8OX8ObSZ6b2ebvLxnyQ56yAWLD4Q1/35Lk0kxhgkGtvhEWdTYj9m5dfjhDnbp+O4Pa8tGwUfS6+Civrhv3+llFITlgYgvWhtbWXFip6Jt5YtW8ayZcuGfvKC1g68TtjxIoShHWcReLiJKRhj6Mz7ZIIYGaml2nRwWewXfMb7JDuo7j5VZv3jJGe+CidZjjguVYuPZT0enokTx8Mjble3FSE04MjOXB85SbE+fgCvrmqHjm3gZWwQZIy9lkPPsi0fGnwopdS4s3LlSlauXFm8uWa03l+M6aU9f4JbsmSJWb169fCfuOlFWHU+ZFqgrA46tkLrZju+woQQePji8mh2Bp0mZbtbjMEAU6SFW4Jj+FVwLGC7XBrv/C7JWQuYdvoVSCze/Tbvd+/hNPcBtptanIIZLU40FqQ6FcMLQmbUljE73gaLTof93mSn2sZSdsCpjvlQSqkJRUQeNcYsGY330kGooyndaIMPLwNV08GJQdtWcKPAQRxCN0E253EgG4njdQcfAC2mgqXOQyTw6FjzZxrv/C6YkNScV4PbszFrVXAEraaCGnYOLO0ajFqZjBGEhpjrMD2etd0sr36PTTK2zxL7rMGHUkqpEaQByGhae9vOlg+w3S+EIDv/GXJ+iIdLQkKmSnOPwz3iJMQj8dQtNN31fTAhNW98H7VvfF/32I4uO6jmIv9cciSop4U4Ho4I5XGXIAxJiM/CqizxVLmO8VBKKTXqdAzIaPGydsxHV/ABu8w6McYGII4IgXGZIa1sZDKFceLNjzSy9u4bAKh98znUvP7dJd9OgHTFHC70PsOx5p8cLw+RIEOKGJPK49RWVRJ/9Sk6xkMppdQeoQHIaGl6wQ7wTO0cQNpj1gkQhCaaMwtGHJzQo1JytJsyAB55cQffvHsjAHVv+zDVS04q+VYCpOIO8+orqUjUsCY/m390vIvT98tx1pKpJJIVOsZDKaXUHqUByGjxMnY9lUKJCsDBhCGBEbwwJIwGBZto8KlrfCrJ4BDy1v1iHLtwEo/POo3yw3qfjeMIJFyHioSL4wihMUyurebE4w4jUalBh1JKqT1PA5DREi+Looqd8gGk3ckkOrfgESc0hsDYVhCHkIR4zJdNZAMh6QrxmM+Pl0/iF2GSVUFbj+m4QPdKuY4jTKtJ4YeGlnSO6rI4V5y8iMkafCillNpL6CDU0VI/H9yETfAFdOZ9nn6llfX5KgJxieF3z3ZxCaiik1jocdn9nSy7vpn2vCFLkq1M5lT3Af4nfg37ylYA4o7gys68Y3FHSLgOOT/ktCWzufqMw5hbXzH696yUUkr1QltARkssCYcsh8dvIF82tXt9l1gsyWZ3HjOz64jjESBUkEVMyOfv87n6wU5cgX9t9Jk5bz86SZI2KWpIc3nsF5zvf5LmqCVEgKlVSS4+8WD2q69k3pQKkjG37+tSSiml9gBtARlNC0+Bslqad2zHD0Jirq3+nKRYy1y2MZmkeDgm4Pw/Zbn6wU5iDlz17n2Ytv8CspLsXgW3lQqqJc07nYcwBlxHmDOpnBvOez3vWjSTg2ZUa/ChlFJqr6UByGiqqCd33HfY1inU00LMeICd/eIRo4kaAiN85C6Pax7OE3eFi09dyMKDDsAzMQzSvaKtADm3mlNTjzB/coLXzp3EL889nAOmVe3RW1RKKaUGQrtgRtm6cBrfrvwCx3j/4MjsA5STxhhDnAAJAz68MstNT2RIuA5fO30hS/afbBeRcwQxEGIHmQIEboKU5Pj4qw1vOOowHWSqlFJqzNAAZDh0rWrrZexsl15ybDR15LjtsU081ujypHMMP+Io5vIKSfJ0+DEmO52kw6+TjDlceeYiXjtvEgChgVTcJe46djE5A14Qsl99BfWu4cCDJ4EGH0oppcYQDUCGonBV2yBv83wYY2e7HLK8R5bR9Y1pLrpjDdvasjhiZ6lAkpfZj8AYWgOPg8wGLj9pX045ah77T6vsfhsRiLsObtTy4WCIuw71VUnczg67eJxSSik1hugYkN3V9CLc+iF4/Abb2lFRD+WT7XMsCU/cYPc3vUhTR46L7lhDzguZXVuO6wiBMfihIZfP88p9/0siyPJiOAOPGAumFbdmCG5BEjMvCJlancQN8hBL2BYXpZRSagzRAGR3FK9qW9zdEktC5XS7f9X53PfoWtoyHjVlcXxjiMeEtoxHezrDuhsvY/vfb2Tjbd8kT5w7gyOoMh3dpwoxJGMOXdNfumbPTKtKQbYZFi7XlOpKKaXGHA1Adkfxqra9Kasj7GyhffUtVKfiNHTkeGJjCx1ZH+Pn2XLLZXS+8DBOWRWT3vQ+BLgzOIKWsJwqOjB23gvJmIMxhrwf4DjCgulVJLxWSNXabh6llFJqjNEAZLBKrWobCYyhPevTmvFoz/r4Qcg2v4zXdNzP2o0NPL25lXTOJ9vZyZbfXUJm/WM45TXMfO+VpGYcgOMIbU4NF/nnkgnjTDItVLoBfmgIDcyoLeOQaSnK8412sOvSq3QlW6WUUmOSDkIdrBKr2ub9kG3tWba35QiNwRiDFxjyQYgxhlpyTPc20WLmEOY62X7LpeQ2rcWtqGPaGVeQmLovxkTL4GLYwHT+d8ZFXDjnWeZuuAvXeCTjLi6dQAIOPavHAFellFJqrNEAZLCKVrXtzPs8u7UdPwiJuw4gdOQDjDGEoYnWZxFS5AFof+IuG3xUTmbamV8nPmkWGEN5KoYYe2o/MGzxKph13KdJ8nE7xdfP2tkuvUzxVUoppcYSDUAGq2BV27wfdq/pkoi5eEFIOu+DXdC2e3E4xJAlAUD165YTdrZRedi7SNROxwCBsUFHedwlNIaYKziOsK4hzUEzqmH6IXviTpVSSqkRowFIL1pbW1mxYkWPbcuWLWPZO4/tXtV2W3uA54cYIJ33CENDaGwzhomClAQe29PwfFAFKRDHpe4t5wI2QJHoOesFlMVdvNAwoyaFK0LWC0b1npVSSk0cK1euZOXKlcWba0br/aXri1L1tGTJErN69erSOx++lvCxX7O6KUnG86NAQgjCnXUZRvXqpBs441dbaY3XM+30y3GS5T1O1dWZY4CKhEsi5nDIzBracz5XnrLItoAopZRSo0BEHjXGLBmN99JZMLtj4Sm0UUEs3wIIDtJjd9fIj3x7M+dcv56Gxh2YfAbj5UqezhQ8L5huF5NLuA7zplSMzPUrpZRSe5gGILujop7fTPksWZOknmbieD12x/EJ2raz4vr/sKEpQ3zKXKadeSVu5a5Td7u6YQTYp7aM8kSMlkyeExbPIBlzR+V2lFJKqdGmAchuyHoBd20u5wL5DCvlLSTwqaONWtqoo42mlg7OuH4zm5uzJKbtz7Qzv45bUVvyXK7YhyNQUxanNeNRXRZn6aIZo3tTSiml1CjSQai7YX1jujtp2M1yPL+XtzOHzfhehtaOTh745Xfx29pIzDiQqe+5DDdVWfI8gp12Gxrb5dLp+dSWJ7ji5EVM1tVtlVJKjWPaAtKLhoaGXvdlvABXhKnVSbzQ4EmcF2Uuz7oHsq78EMrmHkZy1kFMO/3yHsGHAHFHcAuGjJhoym59VZLTXzeHq884jLn1E2Psx7XXXrunL2FC0HoeeVrHI0/reNSMWoZLnQXTi4qKCpNOp0vue2ZLGxfetoaqZIynX2klDA1x1yGMUrGHYQC+RxhLUli7roDjCKGxa7zEXUFE2HdyOb/44OuYUVM2Oje3l1iyZAm9zjRSw0breeRpHY88rePRISKdxphR+RWsLSCDkPMDntnSRlvWww9CQmNYML2K3PZ1PHf9l/EzHVSmYjiOC4kUTs/JMTbpWGibPFIxh7qKBIfMrOZHZ712wgUfSimlJjYdAzIATR05Vq3Zwp1PbiEfhAjQkvFY15imumMD63/5JfKd7Wz9241Mf8eHSMVtVlQvhKRjF6lzRHAdh8qkS01ZnNqKBCcunsnSRTN0vIdSSqkJRwOQfqxvTHPRHWtoy3jUliWoStkqq0zF2Pr8Uzzy64sIc2lmLT6at37g02SNTafuiFAWc2hI54i5wheOXUB1WZzQGFJxl3lTKnSarVJKqQlLA5A+NHXkuOiONeS8kKlVqR77drz4FC/f8BXCXCdVC97ApGUXYNw4VTHbq5X3Q3Zk8tRV2FktE2VgqVJKKTUQGoD0YdWaLbRlvF2Cjy3PPsq93z8fP5dh38PfwQGnfYlX2myXzNQqO/A04TqctmS2drEopZRSJWgA0oc7n9xCbVlil+0vP3o/fi7D/ke9kzd+6Ks4jsu0moDWrMeF71pAVSquXSxKKaVUHzQA6UVoDPkg7B7zUej1Z57P5H0XsP9R77QzXoBU3CWdD6hKxXUBOaWUUqofOg23D4WzaDev/Re5jla73XGYf/QJ3cFHYfmsF4zeBSqllFJj1IQKQERktojcIiKtItImIreJyJzeynclEVv/yH3c873Pcvd3P4Wfy/Z6foNtCenPypUrB3vpAzLWzjuSxlpdaB2P/HlH+twjYSzWxVirYxh7dTEW67iUCROAiEg58BdgAXA28H5gPnC/iOwyRcURIeE6PPePP/LXH12ECQJmHnw4bqL0gNK8H5JwHeZN6X+2y1j7oxyLf+xjrS60jkf+vCN97pEwFutirNUxjL26GIt1XMpEGgPyYWAe8CpjzAsAIvIU8DzwEeCq4gNqNv2D3/38EjCGQ088j8NO/jAiUlwMgJZMntOWzNaBp0oppdQATJgWEOBE4KGu4APAGLMe+AdwUnFh3/f52RVfAGNYuGwFr1m+otfgozXjUV0WZ+miGSN17UoppdS4MpECkIXA0yW2rwUOLt6Yz+cBuODiy1l84jlsb8+S98OeZfyQ7e1ZknGHK05epPk+lFJKqQGaMKvhikgeuMoYc0HR9suBC4wxsaLtWaBrSovguC7i7NplZUKfMAigx8K3DUBjH5dTA7QO/i76NdbOW0/f9TQUY60uRuq8MHL1PBbrYqz9LY/FuhhrdQxjry76O289MGU3z+0aY1L9Fxu6iTQGBHoGCV1K9quM1j+AUkopNRFNpC6YZmBSie110T6llFJKjZKJFICsxY4DKXYw8O9RvhallFJqQptIAcgfgCNEZF7XBhGZC7wh2qeUUkqpUTKRApCfAi8BvxeRk0TkROD3wEbgJzD4TKnKEpFTReRWEXlZRDIi8pyIXCkiVUXl6kTkZyLSKCJpEblPRBaVOF9KRL4tIlui8z0oIm8avTsaG0TkbhEx0UDqwu1az0MkIu8Skb+JSEf0WbBaRN5asF/reAhE5A0ico+IbI/q9zERObeojNbxAInIPiLyg+jeO6PPhbklyg1rnYqIIyIXishLIpIVkSdF5N0DvnBjzIR5AHOAW4E2oB24A5gb7SvHJiV7GjgZmxtkDfAiULGnr31vfgAPATcDZwHHAJ8FWqLtTlRGgL8Dm4AzgeOBB7Cj2vcpOt8N0fEfBt4G3AZkgEP39L3uLY+oDrdgB1ZfXrBd63nodfsRwAO+B7wDOA74EnCC1vGw1O+rozq4P/qcfQf2R6ABPqZ1vFt1+mZgG3AX8KeoLucWlRn2OgWuAHLA54G3RP+OIfCuAV33nq64veUBfAY77faAgm37AT5w/p6+vr35AUwpse0D0f8Eb41enxS9fktBmRpgB3B1wbbFUblzCrbFgOeAP+zpe90bHkAtsDX6ECkOQLSeh1a3c6MP2s/2UUbreGh1/HUgD1QWbX8IeFDreLfq1Cn47/MoHYAMa50CU6Pg49Ki9/kz8NRArnsidcH0Z1CZUtVOxpiGEpsfiZ5nRc8nAq8YY+4vOK4VWEnP+j0R++vzpoJyPnAjcJyIaLY3+Baw1hjz2xL7tJ6H5lzsL7gf91FG63hoEth6yRRtb2HnsACt40EwxoT9lxr2Oj0O+2/566L3+TWwSET26++CNADZaVCZUlW/jomen4me+6rfOSJSWVBuvTGms0S5BHDAcF/oWCIiR2Nblz7eSxGt56E5GngWOENEXhQRX0ReEJFPFJTROh6a66Lnq0VkpojUikhXU//3on1ax8NvuOt0IbYF5IUS5WAA35sagOw0idL5QHZgc4WoARKRWcBlwH3GmNXR5r7qF3bWcX/lSuVymRBEJI7tY/2OMea5XoppPQ/NTOwq2d8GvgEcC9wLXCMin4nKaB0PgTHmaeyYhZOAzdg6+iHwUWPMjVExrePhN9x1OgloMVG/Sx/lejXRMqH2Z8CZUlVpURT9e+zYmXMKdzGw+h1ouYnoS0AZduBXb7Seh8YBqoAPGmNui7b9JZpRcKGIXI3W8ZCIyHzsZIC1wEexXTEnAT8Wkawx5ga0jkfCcNfpkOteA5CdmtFMqUMiIilsTpV5wDHGmE0Fu3fQe/3CzjregZ2t1Fu5HSX2jXtip4N/BTvALFnUt50UkVrszC6t56FpwraA3Fu0/R7srIEZaB0P1dexYwxOMMZ40bY/i8hk4H9E5LdoHY+E4a7THUCdiEhRK8iA6167YHbSTKlDEHUP3Aocjp2CtaaoSF/1u8EY01FQbj8RKS9RLs+u/Y0TxTwghR3g1VzwADsFrhlYhNbzUK3tZXvXr7oQreOhWgQ8WRB8dHkYmIydXaF1PPyGu07XAklg/xLlYADfmxqA7KSZUneTiDjYeeNvA04yxjxUotgfgFkickzBcdXAMnrW7x+AOHBaQbkYcDpwjzEmN/x3MCY8gZ1nX/wAG5S8BfvBoPU8NLdHz8cVbT8O2GSM2YrW8VBtBQ4VkUTR9tcDWewvZ63j4TfcdXo3NiA5q+h93gc8Hc0i7duenr+8tzyACuwH+Bpsf+SJwJPAOormq+tjl7r7EVE+CuCIosc+URkH+Cc28+wZ2A/0v2I/bGYXne9G7C/687BBzS3YD6bX7Ol73dseXfVe8FrreWj1KcBfsF0xH8UOQr02qucPah0PSx2fGtXnn6LP2mOBa6JtV2kdD6leTy34PP5Y9PqYkapT7EDtLHA+dmDxj7CthMsGdM17utL2pgd9ZErVR5/19lL0B1/qcUlBuUnAL6I/+E5swprFJc5XBlyF/aWUBf4FvHlP3+fe+KAoANF6HpY6rcbOytiG/YX3FPBereNhreN3Rl9+DdFn7RPYqeWu1vFu12lvn8F/Hak6BVzgIuBl7JTcp4BTB3rNEp1EKaWUUmrU6BgQpZRSSo06DUCUUkopNeo0AFFKKaXUqNMARCmllFKjTgMQpZRSSo06DUCUUkopNeo0AFHjnoj8TESMiFy1m8fXisglIvKa4b62Eu91nYi81Mf+w0TEE5Gv97L/zyKyQUSqROQlEbluN67hEhHpd36+iLw5qtc391POEZFzRORhEdkhIuloqfsbReTwgnIni8j5g73ePaG3ay2ok7cP4dwVIrJFRN5dsO2S6LwZEakpccwHo/1GRA4o2H5dwXYjIg0i8jcROX6Q11QWXdNp/ZdWamA0AFHjmoiUsTOl8FlRSuHBqgUuBkY8AOmPMeZx4JvAF0RkceE+ETkPeCuwwhjTDiwHvjb6V7mL7wA/Bf6GTdN8MjbJUT02/XaXk7EZFceCkxm5a/0voBG4rcQ+D5vdstgHsAm9SmkAjoweH8Zme71LRN420AsyxmSAbwFXRus+KTVkGoCo8W45NrPlXdhFrgb1y28vdRnwH+DnIuICiMgM4NvAdcaYu8EGK8aYF/fcZXYHgJ8AfmCM+bwx5i5jzL3GmB8aY96OzTi6O+dN9l9q7InWR/kU8GNTOkvkbcD7i46ZDRyDzeJcSt4Y81D0uANYis32/JlBXt51wGzs/1NKDZkGIGq8Oxu7psEHgQz2l+IuRGS5iPxDRDpEpC3qLjgxWpCwa1GlnxY0ZX8wOq5kN0dU5pKC1weIyK9EZH3UjL5ORH4kInXFx/bHGJMHzgEOxf5aBvtFngE+V/Ceu1ybiOwnIjdETfE5EXlCRPr9QhGRKSLym6huWkTkemzLUH8qgAQ2pXOpewmj81+H/beaVVDHL0X7uro1ThGRn4pIAzZNete1fVhEnhSRrIg0isjPRaTHsuPR8ZeLyKejf4N2EXlARBYWlXOjcltEpFNE/iIiCwr/Pfu61gLlInJNdD0NIvJrERlIfS3Hpsu+qZf91wNvEpF9C7a9H9iAbWHqlzGmDRvAHhDdzxnRfTZEf/+Pi8jZJY5rxq7fct5A3kep/uxOc7RSY4KIzATeDlxrjGkQkTuAU0SkLvow7Sr3KeBq7No/ZwMd2O6WudgP3FOwvzyvZOeqkYNtWZgJbAI+iw2I5gFfxrbMHDnYezPGPCwi3wcuEbsa8XLgZGNMS2/HRL+U/wVsxwYqDdgVLm8VkZONMX2t+nwbsDi65uej434wgOtsFJH1wOdFpBW4yxizoUTRrwFTgNdhF4IEu7ZEoR8Af8R+4aaie/oGNgi7GvgCMAu7KOIhInKUMSYoOP59wHPYX/4JbIvR70VkgTHGj8pcGt3jt4H7sH8HxfUykGv9H+BO4L3Aq7DdFwH276svxwPPGGMae9n/d+zaS2cBXeOA3o9dEXlA62pE3ZCz2RlYz8MuNvYN7EJibwJ+JiJlxpgfFx3+N+AKEUkZY7IDeT+lerWnF9DRhz5G6gF8CfuhfGT0+rjo9UcLylRj+85v6+M8c6Pjziux7yVst0fx9h4L8ZXYHwOOjsodVrD9OuClAd5fGfaXrAF+29+1AT/HBh2Ti8rdCzxR8PoS+9HQ/fod0XucUXTcH6Ptb+7nOo+g54KFm6NrObyo3HXYJe+Lj39zdNztJf5dAuCrRdvfEJU/uejf43kgXrCta1XWo6LXddjg8/8Vne/84n/PAVzrL4u2X4Nd1Ev6qatngBtKbL8kOm8M2wX3TLT98Gj7fGwrnwEOKL7O6LgYsA/wk6jcZ0u8jxOV+ynwZIn9byusM33oYygP7YJR49kHgOeNMQ9Gr+8DXqFnN8xRQCV2yfURIyIJEfmyiDwrIhnsYMK/R7tftTvnNHZg4HeilwMZbHo8tsWlVURiXQ9sK89iEanu5bgjsV/0xWMMbhzgdT6Evcd3At/FBiNnAw+KSMkusV7cXvT6HdgvzBuK7udf2DEObyoqf68xxit4vSZ6nhM9L8J2Gf2u6LhbBnGNXVYVvV4DJIFp/Rw3Exsk9uV6YIGIvA77t/yQMeb5PsrPwv69edil2N8LfBXbaoSIzBeR34rI5oJy51H677Lr2mb2c41K9Uu7YNS4FH04Hwx8s6jv/TbgkyJyoDHmP8DkaPumEb6kK7GDCy8D/oltddknup7UEM6bL3ruy1TsF1ZvX/qTsV/cxWYAzUVf3lAwDqM/xpgccHf0QEQOBO7Hzoa5foCn2VL0emr0/EIv5ScXvd5R9Lqr26Sr/mdEz9uLyg34PgfxXr1JsWt3Tg/GmBdE5EHgQ9hWnP/u55zbsQNPDdAEbDRR15SIVGJbwDqBC7Bdi3ngY8C5Jc6ViZ7L+nlPpfqlAYgar7r62r8UPYp9ALgIO90R7K/Ep3fjfbLY8QTdigdARs4ArjfGXF5QrnI33m8omrCtLt/sZf8rvWzfAtSJSLwoCOnv13yvjDH/EZGbgM+JyFRjTPGXfsnDil43Rc/HYsfVFGsqsa0vXQHOVGBtwfbdvs/d0ITtCurP9diBxz69D1jt4hljVvey70hgX+CNxpj/69oovU9X7/rb7m2MilIDpgGIGnfETmU8A9sUf0GJIt8D3i8i/41tjegAVmC7Ikrp+kVa6lffy8AhRdtOKFGuHNu0XeicXt5vpNyN/cJZG3XfDNSDgAu8m57dLmf0d6DYnBGTjDGlWhEWYH9Rt0avcwzul/W92EGTc4wx9w7iuN6sAdLYvDH3F2wvlXxrsNc6UM9iB4X25ybsmKanjDHFrS2DUR49d/9tRjOzTuql/H7R83NDeE+lAA1A1Ph0Arb5/b+MMX8t3ikiPwF+hB08eb+IXAj8QERuBW7Ado8cCmSNMT/ANsE3AWeIyFPYL6n1xpgm7BfyL0Tke9hZD4uxgwGL3Q2cLSJrsF0Gp2DHn4ymrwIPA38TkWuwYzHqsAHUPGNMqSZ3jDH3isj/AT8RkXp2zoIpDrxKqQE2iMjN2PEnm7D/Nmdgx4R8K+qeAfg3MElEPgasxtb/mhLn7LquF0Xkm8A1IvIq4AFsi9Rs7PiQnxlj7u/t+BLna45mFn1ZRNrZOQvmQ1GRsKD4oK51EP4GfFZEHBNNUe7tWhmefBz/xHa7/VBELsaOgelqGdwl4yo2cdxmY8y6YXhvNcHpIFQ1Hp2NDSKKBxN2+S32l/fZAMaYa7C/cvfBBiC3YvvW10f7Q+ygvDrsl9IjwLLoXL/EZkk9BViJ/VVa6ovhU9jpnFdgf71WAWfu/i0OnrHTX5cAT2KncN6LDcSOAf7Sz+GnYAOIK7HXHwM+OYC3bcN+oU3DTm39M7b7YF/gI/RsofoZNqD7OjZQWjmAe/oytvXqTcDNwO+xXW7N2EBpsC7G3uPZ2H+vd7IzoGwtKDfoax2gm7Bf/G8cpvP1yRjTgP17dbGDba/E3tuvezlkKQMcfKxUf8SYAU0dV0qpCUns+ic3A28yxvy9v/LD8H5/BV4wxuxVCb9E5PXYFpODogHcSg2JBiBKKRWJvmSXYscPZYHXYltpnsPmvhjxD0wReQO2pe0AY8zmkX6/gRKR27GzoUp21Sk1WDoGRCmldurAdud8Apukbju29ePC0Qg+AIwx/xCRz2G7qfaKAEREUsDj2ARlSg0LbQFRSiml1KjTQahKKaWUGnUagCillFJq1GkAopRSSqlRpwGIUkoppUadBiBKKaWUGnX/H3W6i2XPdAoYAAAAAElFTkSuQmCC\n",
      "text/plain": [
       "<Figure size 576x576 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "############## EVALUATION ##########################################################\n",
    "\n",
    "print(model_evaluation_result(y_train,y_train_predict,y_test,y_test_predict,'gbr_GS'))\n",
    "print('****************************************************************************')\n",
    "feature_imp_df = pd.concat([pd.Series(X_train_unscaled_avg.columns),\n",
    "                            pd.Series(gbr_gs_best.feature_importances_)], \n",
    "                           axis = 1)\n",
    "feature_imp_df.columns=['feature_name','importance'];\n",
    "print(feature_imp_df.sort_values('importance',ascending=False)[0:10])\n",
    "feature_imp_df.index = feature_imp_df.feature_name\n",
    "\n",
    "#feature_imp_df.sort_values('importance', ascending=False)[0:12].plot(kind=\"bar\", figsize=(16,4))\n",
    "\n",
    "plot_model_performance(y_train,y_train_predict,y_test,y_test_predict,r2_train,r2_test)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "92e7816c",
   "metadata": {},
   "source": [
    "### Estimating mean and standard deviation using the helper function bootstrap_estimator"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "id": "c3c7554a",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "           0           1          2          3          4           5   \\\n",
      "0  415.269036  180.824056  89.908165  79.091237  56.524688  472.403367   \n",
      "1   29.468263   17.860470  16.576380  19.061268  16.393786   28.578031   \n",
      "\n",
      "           6           7           8           9   ...          15  \\\n",
      "0  355.930648  330.968793  215.305404  112.018705  ...  272.770369   \n",
      "1   29.405448   28.127500   18.008821   12.549282  ...   22.005491   \n",
      "\n",
      "           16          17          18          19          20          21  \\\n",
      "0  509.592252  443.660928  426.185052  406.767886  401.786412  355.800119   \n",
      "1   41.485972   45.222944   49.230465   52.761693   60.878875   55.184804   \n",
      "\n",
      "           22          23          24  \n",
      "0  339.567293  289.378877  235.055600  \n",
      "1   49.957911   44.311239   23.572647  \n",
      "\n",
      "[2 rows x 25 columns]\n"
     ]
    }
   ],
   "source": [
    "## Using the hyperparmeters of the best estimator of GBR model \n",
    "from sklearn.ensemble import GradientBoostingRegressor\n",
    "# parameters = dict(criterion='mse', max_features='auto',\n",
    "#                           min_samples_leaf=6, min_samples_split=5,\n",
    "#                           n_estimators=50, random_state=20\n",
    "    \n",
    "#                  )\n",
    "\n",
    "\n",
    "# parameters = dict(criterion='mse', max_features='sqrt',\n",
    "#                           min_samples_leaf=6, min_samples_split=5,\n",
    "#                           n_estimators=200, random_state=20)\n",
    "\n",
    "# gbr = GradientBoostingRegressor(**parameters)\n",
    "\n",
    "\n",
    "mean , std  = bootstrap_estimator(gbr_gs_best, X_train, y_train, X_test, n_iter=1000)\n",
    "\n",
    "\n",
    "print(pd.DataFrame([mean,std]))\n",
    "\n",
    "mean_train , std_train  = bootstrap_estimator(gbr_gs_best, X_train, y_train, X_train, n_iter=1000)\n"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "4f11671b",
   "metadata": {},
   "source": [
    "### plotting the uncertainties for the test data "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "id": "929e6dde",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "r2 score test data: 0.7289222817846034\n",
      "r2 score train data: 0.9832952356284047\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 576x576 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.figure(figsize=(8,8))\n",
    "\n",
    "plt.errorbar(y_test,mean,yerr=std,ecolor = 'grey',mfc='None', mec='red',mew=2,elinewidth= 3,ls = 'none',marker='o',markersize=8,solid_capstyle='projecting', capsize=5,label='Test Data')\n",
    "\n",
    "plt.xlabel('Actual YS in MPa',fontsize=16)\n",
    "plt.ylabel('Predicted YS in MPa with standard deviation',fontsize=16)\n",
    "plt.axline((0, 0), (1000, 1000), linewidth=2, color='grey',linestyle='dashed',label='Ideal Fit')\n",
    "plt.legend(fontsize=16)\n",
    "plt.xlim(0,1000)\n",
    "plt.ylim(0,1000)\n",
    "\n",
    "plt.minorticks_on()\n",
    "plt.tick_params(labelsize=16)\n",
    "plt.tick_params(direction='in',which='major', length=15, bottom=True, top=True, left=True, right=True,colors='k')\n",
    "plt.tick_params(direction='in',which='minor', length=5, bottom=True, top=True, left=True, right=True,colors='k')\n",
    "\n",
    "print('r2 score test data:',r2_score(y_test,mean))\n",
    "print('r2 score train data:',r2_score(y_train,mean_train))   \n",
    "# mean_squared_error(y_test,mean,squared=False)\n",
    "# mean_absolute_error(y_test,mean)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "2cd36cee",
   "metadata": {},
   "source": [
    "### Plotting the uncertainties for the train data "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "id": "04bcd892",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "r2 score test data: 0.7289222817846034\n",
      "r2 score train data: 0.9832952356284047\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 576x576 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.figure(figsize=(8,8))\n",
    "\n",
    "plt.errorbar(y_train,mean_train,yerr=std_train,mfc='None',mec='green',mew=2,ecolor = 'grey',alpha=0.8, elinewidth= 3,ls = 'none',marker='o',markersize=8,solid_capstyle='projecting', capsize=5,label='Training Data')\n",
    "plt.xlabel('Actual YS in MPa',fontsize=16)\n",
    "plt.ylabel('Predicted YS in MPa with standard deviation',fontsize=16)\n",
    "plt.axline((0, 0), (1000, 1000), linewidth=2, color='grey',linestyle='dashed',label='Ideal Fit')\n",
    "plt.legend(fontsize=16)\n",
    "\n",
    "plt.minorticks_on()\n",
    "plt.tick_params(labelsize=16)\n",
    "plt.tick_params(direction='in',which='major', length=15, bottom=True, top=True, left=True, right=True,colors='k')\n",
    "plt.tick_params(direction='in',which='minor', length=5, bottom=True, top=True, left=True, right=True,colors='k')\n",
    "\n",
    "\n",
    "plt.xlim(0,1000)\n",
    "plt.ylim(0,1000)\n",
    "print('r2 score test data:',r2_score(y_test,mean))\n",
    "print('r2 score train data:',r2_score(y_train,mean_train))  \n",
    "\n",
    "# mean_squared_error(y_test,mean,squared=False)\n",
    "# mean_absolute_error(y_test,mean)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "4813af3a",
   "metadata": {},
   "source": [
    "### Reading .csv file of composition (candidate search space)  and converting weight percent to atom percent"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "id": "c94194fe",
   "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>formula</th>\n",
       "      <th>T</th>\n",
       "      <th>target</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>Nb80.9B7.2C6.4N5.5</td>\n",
       "      <td>24</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>Nb76.4B6.8C6.2N10.6</td>\n",
       "      <td>24</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>Nb72.4B6.5C5.9N15.2</td>\n",
       "      <td>24</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>Nb68.7B6.3C5.7N19.4</td>\n",
       "      <td>24</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>Nb65.3B6.0C5.4N23.3</td>\n",
       "      <td>24</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "               formula   T  target\n",
       "0   Nb80.9B7.2C6.4N5.5  24       0\n",
       "1  Nb76.4B6.8C6.2N10.6  24       0\n",
       "2  Nb72.4B6.5C5.9N15.2  24       0\n",
       "3  Nb68.7B6.3C5.7N19.4  24       0\n",
       "4  Nb65.3B6.0C5.4N23.3  24       0"
      ]
     },
     "execution_count": 12,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "from convesrion_weightpct_atpct import wtpct_To_atpct\n",
    "df_test_composition = pd.read_csv('data/compositions_test_24degc_allcombinations.csv')\n",
    "df_test_composition['formula']=[wtpct_To_atpct(i) for i in df_test_composition.wt_pct]\n",
    "df_test_total = df_test_composition[['formula','T','target']]\n",
    "#df_test_total['T']=1300\n",
    "df_test_total.head()\n",
    "\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "id": "97eb9c81",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "1666625\n"
     ]
    },
    {
     "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>formula</th>\n",
       "      <th>T</th>\n",
       "      <th>target</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>1666620</th>\n",
       "      <td>Nb60.8Ta14.2W14.0Re11.0</td>\n",
       "      <td>24</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1666621</th>\n",
       "      <td>Nb59.8Ta14.3W14.1Re11.8</td>\n",
       "      <td>24</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1666622</th>\n",
       "      <td>Nb58.9Ta14.4W14.2Re12.6</td>\n",
       "      <td>24</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1666623</th>\n",
       "      <td>Nb57.9Ta14.5W14.3Re13.4</td>\n",
       "      <td>24</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1666624</th>\n",
       "      <td>Nb56.9Ta14.6W14.4Re14.2</td>\n",
       "      <td>24</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                         formula   T  target\n",
       "1666620  Nb60.8Ta14.2W14.0Re11.0  24       0\n",
       "1666621  Nb59.8Ta14.3W14.1Re11.8  24       0\n",
       "1666622  Nb58.9Ta14.4W14.2Re12.6  24       0\n",
       "1666623  Nb57.9Ta14.5W14.3Re13.4  24       0\n",
       "1666624  Nb56.9Ta14.6W14.4Re14.2  24       0"
      ]
     },
     "execution_count": 13,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "print(len(df_test_total))\n",
    "df_test_final = df_test_total\n",
    "df_test_final.tail()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "fec66806",
   "metadata": {},
   "source": [
    "### Generate Features for the search space using CBFV"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "id": "6c52e6d8",
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Processing Input Data: 100%|███████| 1666625/1666625 [01:21<00:00, 20347.04it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\tFeaturizing Compositions...\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Assigning Features...: 100%|███████| 1666625/1666625 [01:47<00:00, 15547.93it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\tCreating Pandas Objects...\n"
     ]
    }
   ],
   "source": [
    "X_test_comp, y_test_comp, formulae_test_comp, skipped_train = generate_features(df_test_final,elem_prop='f3_revised',drop_duplicates=False,extend_features=True)\n",
    "#X_test_comp[110000:160004]"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "9d8af442",
   "metadata": {},
   "source": [
    "### Using only weighted average values "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "id": "5ba2fe6b",
   "metadata": {},
   "outputs": [],
   "source": [
    "X_test_comp_avg  = X_test_comp[['avg_Atomic_Radius','avg_Pauling_Electronegativity','avg_number_of_valence_electrons','avg_Cohesive_energy_ev_atom',\n",
    "            'avg_Bulk_modulus_RT_Gpa','avg_Elastic_modulus_RT_Gpa','avg_Shear_modulus_RT_Gpa','avg_Melting_point_(K)','avg_rate_shear_mod_Mpa_perK',\n",
    "            'avg_Solid_Solubility_atpct','avg_lattice_constant_A','avg_BEC_percm3','avg_Av.Valence_bond_strength_ev','avg_EngelZ_e/a','T']]\n",
    "\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "id": "3825d116",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "avg_Atomic_Radius                  0\n",
       "avg_Pauling_Electronegativity      0\n",
       "avg_number_of_valence_electrons    0\n",
       "avg_Cohesive_energy_ev_atom        0\n",
       "avg_Bulk_modulus_RT_Gpa            0\n",
       "avg_Elastic_modulus_RT_Gpa         0\n",
       "avg_Shear_modulus_RT_Gpa           0\n",
       "avg_Melting_point_(K)              0\n",
       "avg_rate_shear_mod_Mpa_perK        0\n",
       "avg_Solid_Solubility_atpct         0\n",
       "avg_lattice_constant_A             0\n",
       "avg_BEC_percm3                     0\n",
       "avg_Av.Valence_bond_strength_ev    0\n",
       "avg_EngelZ_e/a                     0\n",
       "T                                  0\n",
       "dtype: int64"
      ]
     },
     "execution_count": 16,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "## to test if there is any null value \n",
    "pd.DataFrame(X_test_comp_avg).isnull().sum()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "0744ad28",
   "metadata": {},
   "source": [
    "### Normalization and Prediction\n",
    "### for predicting mean and std deviation of candidate space we have used entire dataset for training keeping the hyperparameters same for the model "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "id": "9b7899f5",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "maximum value of the YS in the experimental data: 985.99\n"
     ]
    }
   ],
   "source": [
    "X_train_all = np.concatenate((X_train_unscaled_avg,X_test_unscaled_avg),axis = 0)\n",
    "y_train_all = np.concatenate((y_train,y_test),axis=0)\n",
    "\n",
    "scaler = StandardScaler()\n",
    "X_train_all_scaled = scaler.fit_transform(X_train_all) ## fit and transform training data and \n",
    "\n",
    "X_test_comp_avg_scaled  = scaler.transform(X_test_comp_avg) # transform test data\n",
    "print('maximum value of the YS in the experimental data:',y_train_all.max())\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "id": "4fab5259",
   "metadata": {},
   "outputs": [],
   "source": [
    "# Estimating mean and standard deviation for all the available experimental data\n",
    "mean_train_all , std_train_all  = bootstrap_estimator(gbr_gs_best, X_train_all_scaled, y_train_all, X_train_all_scaled, n_iter=1000)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "id": "a98288bf",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "r2 score train data: 0.9843081481757276\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 720x576 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.figure(figsize=(10,8))\n",
    "plt.errorbar(y_train_all,mean_train_all,yerr=std_train_all,mfc='None',mec='green',mew=2,ecolor = 'grey',alpha=0.8, elinewidth= 3,ls = 'none',marker='o',markersize=8,solid_capstyle='projecting', capsize=5,label='Experimental Data')\n",
    "#plt.errorbar(y_train,mean_train,yerr=std_train,ecolor = 'grey', elinewidth= 3,c ='red',ls = 'none',marker='o',markersize=8,solid_capstyle='projecting', capsize=5,label='UTS (MPa) train')\n",
    "\n",
    "plt.xlabel('Actual YS in MPa',fontsize=16)\n",
    "plt.ylabel('Predicted YS in MPa with standard deviation',fontsize=16)\n",
    "plt.axline((0, 0), (1000, 1000), linewidth=2, color='grey',linestyle='dashed',label='Ideal Fit')\n",
    "plt.legend(fontsize=16)\n",
    "\n",
    "plt.minorticks_on()\n",
    "plt.tick_params(labelsize=16)\n",
    "plt.tick_params(direction='in',which='major', length=15, bottom=True, top=True, left=True, right=True,colors='k')\n",
    "plt.tick_params(direction='in',which='minor', length=5, bottom=True, top=True, left=True, right=True,colors='k')\n",
    "\n",
    "\n",
    "print('r2 score train data:',r2_score(y_train_all,mean_train_all))   \n",
    "# mean_squared_error(y_test,mean,squared=False)\n",
    "# mean_absolute_error(y_test,mean)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "998eef01",
   "metadata": {},
   "source": [
    "### Calculating EI"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "id": "f17253d9",
   "metadata": {},
   "outputs": [],
   "source": [
    "ei,predicted_property,std = Expected_Improvement(X_test_comp_avg_scaled, X_train_all_scaled, y_train_all,gbr_gs_best,xi=0.01)\n"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "6702d8a5",
   "metadata": {},
   "source": [
    "### Saving the Predicted value and EI in .csv file\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "id": "89b8f941",
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>formula</th>\n",
       "      <th>T</th>\n",
       "      <th>target</th>\n",
       "      <th>ei</th>\n",
       "      <th>predicted_property</th>\n",
       "      <th>std</th>\n",
       "      <th>wt_pct</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>1546644</th>\n",
       "      <td>Nb87.0Zr1.1Hf0.6W11.3</td>\n",
       "      <td>24</td>\n",
       "      <td>0</td>\n",
       "      <td>16.546617</td>\n",
       "      <td>874.145183</td>\n",
       "      <td>138.917143</td>\n",
       "      <td>Nb78Zr1Hf1W20</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1562644</th>\n",
       "      <td>Nb87.0Zr1.1Ta0.6W11.3</td>\n",
       "      <td>24</td>\n",
       "      <td>0</td>\n",
       "      <td>15.852908</td>\n",
       "      <td>877.303511</td>\n",
       "      <td>134.208063</td>\n",
       "      <td>Nb78Zr1Ta1W20</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1635044</th>\n",
       "      <td>Nb86.8Hf1.2Ta0.6W11.4</td>\n",
       "      <td>24</td>\n",
       "      <td>0</td>\n",
       "      <td>13.612348</td>\n",
       "      <td>870.504778</td>\n",
       "      <td>131.027135</td>\n",
       "      <td>Nb77Hf2Ta1W20</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1562683</th>\n",
       "      <td>Nb86.4Zr1.1Ta1.7W10.8</td>\n",
       "      <td>24</td>\n",
       "      <td>0</td>\n",
       "      <td>13.266260</td>\n",
       "      <td>860.988987</td>\n",
       "      <td>136.311686</td>\n",
       "      <td>Nb77Zr1Ta3W19</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1563064</th>\n",
       "      <td>Nb85.2Zr2.3Ta1.2W11.3</td>\n",
       "      <td>24</td>\n",
       "      <td>0</td>\n",
       "      <td>13.082206</td>\n",
       "      <td>861.990487</td>\n",
       "      <td>134.922511</td>\n",
       "      <td>Nb76Zr2Ta2W20</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1562684</th>\n",
       "      <td>Nb85.7Zr1.1Ta1.7W11.4</td>\n",
       "      <td>24</td>\n",
       "      <td>0</td>\n",
       "      <td>12.884086</td>\n",
       "      <td>867.109776</td>\n",
       "      <td>130.654809</td>\n",
       "      <td>Nb76Zr1Ta3W20</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1635464</th>\n",
       "      <td>Nb85.5Hf1.8Ta1.2W11.5</td>\n",
       "      <td>24</td>\n",
       "      <td>0</td>\n",
       "      <td>12.849483</td>\n",
       "      <td>867.312176</td>\n",
       "      <td>130.384330</td>\n",
       "      <td>Nb75Hf3Ta2W20</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1635444</th>\n",
       "      <td>Nb86.2Hf1.8Ta0.6W11.5</td>\n",
       "      <td>24</td>\n",
       "      <td>0</td>\n",
       "      <td>12.849483</td>\n",
       "      <td>867.312176</td>\n",
       "      <td>130.384330</td>\n",
       "      <td>Nb76Hf3Ta1W20</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1635064</th>\n",
       "      <td>Nb86.2Hf1.2Ta1.2W11.5</td>\n",
       "      <td>24</td>\n",
       "      <td>0</td>\n",
       "      <td>12.845456</td>\n",
       "      <td>867.334318</td>\n",
       "      <td>130.353824</td>\n",
       "      <td>Nb76Hf2Ta2W20</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1634644</th>\n",
       "      <td>Nb87.5Hf0.6Ta0.6W11.3</td>\n",
       "      <td>24</td>\n",
       "      <td>0</td>\n",
       "      <td>12.803306</td>\n",
       "      <td>867.864948</td>\n",
       "      <td>129.828923</td>\n",
       "      <td>Nb78Hf1Ta1W20</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                       formula   T  target         ei  predicted_property  \\\n",
       "1546644  Nb87.0Zr1.1Hf0.6W11.3  24       0  16.546617          874.145183   \n",
       "1562644  Nb87.0Zr1.1Ta0.6W11.3  24       0  15.852908          877.303511   \n",
       "1635044  Nb86.8Hf1.2Ta0.6W11.4  24       0  13.612348          870.504778   \n",
       "1562683  Nb86.4Zr1.1Ta1.7W10.8  24       0  13.266260          860.988987   \n",
       "1563064  Nb85.2Zr2.3Ta1.2W11.3  24       0  13.082206          861.990487   \n",
       "1562684  Nb85.7Zr1.1Ta1.7W11.4  24       0  12.884086          867.109776   \n",
       "1635464  Nb85.5Hf1.8Ta1.2W11.5  24       0  12.849483          867.312176   \n",
       "1635444  Nb86.2Hf1.8Ta0.6W11.5  24       0  12.849483          867.312176   \n",
       "1635064  Nb86.2Hf1.2Ta1.2W11.5  24       0  12.845456          867.334318   \n",
       "1634644  Nb87.5Hf0.6Ta0.6W11.3  24       0  12.803306          867.864948   \n",
       "\n",
       "                std         wt_pct  \n",
       "1546644  138.917143  Nb78Zr1Hf1W20  \n",
       "1562644  134.208063  Nb78Zr1Ta1W20  \n",
       "1635044  131.027135  Nb77Hf2Ta1W20  \n",
       "1562683  136.311686  Nb77Zr1Ta3W19  \n",
       "1563064  134.922511  Nb76Zr2Ta2W20  \n",
       "1562684  130.654809  Nb76Zr1Ta3W20  \n",
       "1635464  130.384330  Nb75Hf3Ta2W20  \n",
       "1635444  130.384330  Nb76Hf3Ta1W20  \n",
       "1635064  130.353824  Nb76Hf2Ta2W20  \n",
       "1634644  129.828923  Nb78Hf1Ta1W20  "
      ]
     },
     "execution_count": 21,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_predict = pd.DataFrame({\"ei\":ei, \"predicted_property\":predicted_property, \"std\":std})\n",
    "df_predict_final = df_test_final.join(df_predict)\n",
    "df_predict_final['wt_pct'] = df_test_composition['wt_pct']\n",
    "\n",
    "df_predict_final_sorted = df_predict_final.sort_values('ei',ascending=False)\n",
    "#df_predict_final_sorted.to_csv('data/model_output/paper_YS_quarterany.csv')\n",
    "df_predict_final_sorted[0:10]\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "dd8bc285",
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3 (ipykernel)",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.9.6"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}