tb_bm_model_sel.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "id": "f190735c-7f51-43fc-8909-c95bdbed2a0b",
   "metadata": {},
   "outputs": [],
   "source": [
    "import pandas as pd\n",
    "import re\n",
    "import seaborn as sns\n",
    "import matplotlib.pyplot as plt\n",
    "from matplotlib.patches import Patch\n",
    "from matplotlib.lines import Line2D\n",
    "import matplotlib.ticker as mtick\n",
    "import matplotlib.gridspec as gridspec\n",
    "import matplotlib.cm as cm\n",
    "import matplotlib.dates as mdates\n",
    "import numpy as np\n",
    "import statsmodels.formula.api as smf\n",
    "import statsmodels.stats.api as sms\n",
    "from scipy.stats import gaussian_kde\n",
    "import itertools\n",
    "from statsmodels.compat import lzip\n",
    "from tqdm.notebook import tqdm\n",
    "import pickle"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "id": "de9c94ad-6793-481f-8dc1-a3dd9bc3aa55",
   "metadata": {},
   "outputs": [],
   "source": [
    "\"\"\"\n",
    "Define figure parameters for RSE\n",
    "\"\"\"\n",
    "# image widths (inches from cm) for Elsevier\n",
    "one_col = 9 / 2.54\n",
    "half_pg = 14 / 2.54\n",
    "full_pg = 19 / 2.54\n",
    "\n",
    "# font sizes\n",
    "font_xsml = 6\n",
    "font_sml = 8\n",
    "font_med = 10\n",
    "font_lg = 12\n",
    "font_xlg = 14\n",
    "\n",
    "# save all figures?\n",
    "save_all_figs = False\n",
    "\n",
    "\"\"\"\n",
    "Define input and output files and directories\n",
    "\"\"\"\n",
    "# output directories\n",
    "outDIR_figs = 'C:/Users/sean.kearney/OneDrive - USDA/Documents/Manucripts/Kearney_et_al_Biomass_from_HLS_VOR/Figures/'\n",
    "outDIR_tables = 'C:/Users/sean.kearney/OneDrive - USDA/Documents/Manucripts/Kearney_et_al_Biomass_from_HLS_VOR/Tables/'\n",
    "outDIR_figs_R1 = 'C:/Users/sean.kearney/OneDrive - USDA/Documents/Manucripts/Kearney_et_al_Biomass_from_HLS_VOR/R1/Figures/'\n",
    "\n",
    "# variables used in model creation\n",
    "var_names = ['dfi', 'ndvi', 'ndti', 'satvi', 'ndii7', 'nir', 'swir1', 'swir2',\n",
    "             'bai_126', 'bai_136', 'bai_236', 'bai_246']\n",
    "\n",
    "# load csv of ground data with extractions\n",
    "df_full = pd.read_csv('data/bm_extract/TB_all_bm.csv', parse_dates=[1])\n",
    "df_ext = pd.read_csv('data/bm_extract/TB_all_bm_veg_idxs.csv', parse_dates=[1])\n",
    "df = pd.merge(df_full, df_ext[['ID', 'Date'] + var_names], how='left', on=['ID', 'Date'])\n",
    "df['bm'] = df['bm_gm2'].astype('float') * 10.0\n",
    "\n",
    "# linear model results\n",
    "suffix = 'strat_projyear_both'\n",
    "#suffix = 'byproj_both'\n",
    "log_xfrm = False\n",
    "project = 'PDOG'\n",
    "df_results = pd.read_csv(\"data/bm_modeling/tb_vor_lreg_results_\" + suffix + \".csv\")\n",
    "df_results_full = pd.read_csv(\"data/bm_modeling/tb_vor_lreg_results_\" + suffix + \"_full.csv\")\n",
    "if suffix.split('_')[0] == 'byproj':\n",
    "    df_results = df_results[df_results['project'] == project]\n",
    "    df_results_full = df_results_full[df_results_full['project'] == project]\n",
    "    #df_results = df_results[df_results['model'] == 'Project']\n",
    "    #df_results_full = df_results_full[df_results_full['model'] == 'Project']\n",
    "    \n",
    "if suffix.split('_')[-1] == 'both':\n",
    "    df_out_all = df.groupby(['ID', 'Date', 'Project', 'gps_E', 'gps_N']).mean().reset_index()\n",
    "    df_out_all.loc[df_out_all['bm'] < 1, 'bm'] = 1\n",
    "else:\n",
    "    df_out_all = df[df['Method'] == suffix.split('_')[-1]].copy()\n",
    "    if suffix.split('_')[-1] == 'vor':\n",
    "        df_out_all.loc[df_out_all['bm'] < 0, 'bm'] = 1\n",
    "if suffix.split('_')[0] == 'byproj':\n",
    "    df_out_all = df_out_all[df_out_all['Project'] == project].copy()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 33,
   "id": "245962a8-c5c5-412d-a9b1-0d0e2f8965af",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Index(['Unnamed: 0', 'model', 'kfold', 'numb_vars', 'formula', 'R2', 'R2_adj',\n",
       "       'AIC', 'rmse_kg', 'pe', 'rmse_pct', 'rmse_orig', 'rmse_orig_pct',\n",
       "       'rmse_kg_low', 'rmse_pct_low', 'pe_low', 'mae_kg', 'mae_pct',\n",
       "       'mae_orig'],\n",
       "      dtype='object')"
      ]
     },
     "execution_count": 33,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_results_full.columns"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 35,
   "id": "9265102d-faed-4d7a-b2a5-745c42758e71",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Adjusting MAE of full results - likely bm was not transformed.\n"
     ]
    }
   ],
   "source": [
    "if df_results['mae_orig_mean'].isnull().all():\n",
    "    print('Adjusting MAE of results - likely bm was not transformed.')\n",
    "    df_results['mae_orig_mean'] = df_results['mae_kg_mean']\n",
    "    df_results['mae_orig_sd'] = df_results['mae_kg_sd']\n",
    "    \n",
    "if df_results_full['mae_orig'].isnull().all():\n",
    "    print('Adjusting MAE of full results - likely bm was not transformed.')\n",
    "    df_results_full['mae_orig'] = df_results_full['mae_kg']"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 36,
   "id": "e54a8e56-b7e2-4549-ae47-0543a2527952",
   "metadata": {},
   "outputs": [],
   "source": [
    "#df_out_all = df_out_all[df_out_all['bm'] < 1500]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 37,
   "id": "5ddfab17-359e-45f5-ad15-55fb6940fdfe",
   "metadata": {},
   "outputs": [],
   "source": [
    "\"\"\"\n",
    "Define necessary lists and dictionaries\n",
    "\"\"\"\n",
    "\n",
    "# formatted names for input variables\n",
    "in_var_dict = {'dfi': 'DFI',\n",
    "               'ndvi': 'NDVI',\n",
    "               'ndti': 'NDTI',\n",
    "               'satvi': 'SATVI',\n",
    "               'ndii7': 'NDII7',\n",
    "               'nir': 'NIR',\n",
    "               'swir1': 'SWIR1',\n",
    "               'swir2': 'SWIR2',\n",
    "               'bai_126': 'BAI_126',\n",
    "               'bai_136': 'BAI_136',\n",
    "               'bai_236': 'BAI_236',\n",
    "               'bai_246': 'BAI_246',\n",
    "               'pct_c3': 'MG-plot',\n",
    "               'pct_c3_ras': 'MG-map'}\n",
    "\n",
    "# grouping names for input variables\n",
    "input_group_dict = {\n",
    "    'DFI': 'Veg. Indices',\n",
    "    'NDVI': 'Veg. Indices',\n",
    "    'NDTI': 'Veg. Indices',\n",
    "    'SATVI': 'Veg. Indices',\n",
    "    'NDII7': 'Veg. Indices',\n",
    "    'NIR': 'Bands',\n",
    "    'SWIR1': 'Bands',\n",
    "    'SWIR2': 'Bands',\n",
    "    'BAI_126': 'BAIs',\n",
    "    'BAI_136': 'BAIs',\n",
    "    'BAI_236': 'BAIs',\n",
    "    'BAI_246': 'BAIs',\n",
    "    'MG-map': 'Veg. Community'\n",
    "}\n",
    "\n",
    "# directory in which to save each model type\n",
    "mod_dir_dict = {'Temporal': '/Main/',\n",
    "                'Spatial': '/Supp/'}"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 38,
   "id": "26366e0c-fd92-4d70-be5e-aeeb2c6af9ed",
   "metadata": {},
   "outputs": [],
   "source": [
    "\"\"\"\n",
    "Define necessary functions\n",
    "\"\"\"\n",
    "\n",
    "def abline(slope, intercept):\n",
    "    \"\"\"Plot a line from slope and intercept\"\"\"\n",
    "    axes = plt.gca()\n",
    "    x_vals = np.array(axes.get_xlim())\n",
    "    y_vals = intercept + slope * x_vals\n",
    "    plt.plot(x_vals, y_vals, '-', c='black')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 39,
   "id": "a720e96d-222d-4d70-b004-0c6b7f5aa1d5",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Explained variation per principal component: [0.64392248 0.22446749 0.07465578]\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 538.583x430.866 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "\"\"\"\n",
    "Finish prep of dataset\n",
    "\"\"\"\n",
    "df_out_all['Year'] = df_out_all['Date'].dt.year\n",
    "df_out_all = df_out_all[(df_out_all['bm'].notnull()) & (df_out_all[var_names].notnull().any(axis=1))]\n",
    "\n",
    "from sklearn.preprocessing import StandardScaler\n",
    "x = df_out_all.loc[:, var_names].values\n",
    "x = StandardScaler().fit_transform(x) # normalizing the features\n",
    "\n",
    "from sklearn.decomposition import PCA\n",
    "pca_vars = PCA(n_components=3)\n",
    "principalComponents_df = pca_vars.fit_transform(x)\n",
    "\n",
    "pca_df = pd.DataFrame(data = principalComponents_df,\n",
    "                      columns = ['PC1', \n",
    "                                 'PC2',\n",
    "                                 'PC3'])\n",
    "print('Explained variation per principal component: {}'.format(pca_vars.explained_variance_ratio_))\n",
    "\n",
    "pca_outliers = (pca_df < (pca_df.agg('mean') - pca_df.agg('std')*3)).any(axis=1) | (pca_df > (pca_df.agg('mean') + pca_df.agg('std')*3)).any(axis=1)\n",
    "df_out_all = df_out_all.reset_index()[~pca_outliers]\n",
    "\n",
    "\"\"\"\n",
    "Figure (Supplement): Correlation matrix for all input variables\n",
    "\"\"\"\n",
    "# define variables for correlation matrix\n",
    "cor_vars_all = df_out_all[var_names + ['bm']]\n",
    "\n",
    "# create and save correlation matrix\n",
    "from scipy.stats import pearsonr\n",
    "corrMatrix = np.round(cor_vars_all.reset_index(drop=True).corr(), 2).rename(index=in_var_dict, columns=in_var_dict)\n",
    "pvals = pd.DataFrame([[pearsonr(cor_vars_all[c],cor_vars_all[y])[1] for y in cor_vars_all.columns] for c in cor_vars_all.columns],\n",
    "                     columns=cor_vars_all.columns, index=cor_vars_all.columns)\n",
    "pvals_mask = pvals > 0.05\n",
    "mask = np.zeros(corrMatrix.shape, dtype=bool)\n",
    "mask[np.triu_indices(len(mask))] = True\n",
    "mask = pvals_mask.values | mask\n",
    "fig = plt.figure(figsize=(full_pg, full_pg * 0.8))\n",
    "ax = sns.heatmap(corrMatrix, annot=True, vmin=-1, vmax=1, center=0, mask=mask, cmap='coolwarm',\n",
    "                 annot_kws={\"size\": font_sml})\n",
    "ax.set_yticklabels(ax.get_yticklabels(), rotation=0, ha='right', fontsize=font_sml)\n",
    "ax.set_xticklabels(ax.get_xticklabels(), rotation=45, ha='right', fontsize=font_sml)\n",
    "cbar = ax.collections[0].colorbar\n",
    "# here set the labelsize by 20\n",
    "cbar.ax.tick_params(labelsize=font_sml)\n",
    "if save_all_figs:\n",
    "    plt.savefig(outDIR_figs_R1 + '/Supp/corrmatrix_inputs.png', dpi=600, bbox_inches='tight', pad_inches=0.2)\n",
    "#plt.close()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 40,
   "id": "f6260fc8-a807-40fa-beeb-a6d76a1682fc",
   "metadata": {},
   "outputs": [],
   "source": [
    "\"\"\"\n",
    "Identify top models based on MAE, % MAE, R^2 and AIC\n",
    "\"\"\"\n",
    "# identify best models for each metric\n",
    "df_results_best = pd.DataFrame(\n",
    "    columns=['model', 'numb_vars', 'min_MAE', 'min_MAE_pct', 'max_R2', 'min_AIC'])\n",
    "for m in df_results['model'].unique():\n",
    "    for n in df_results['numb_vars'].unique():\n",
    "        df_best_tmp = pd.DataFrame({'model': m,\n",
    "                                    'numb_vars': n,\n",
    "                                    'min_MAE': df_results['mae_orig_mean'][(df_results['model'] == m) & \\\n",
    "                                                                           (df_results['numb_vars'] == n)].min(),\n",
    "                                    'min_MAE_pct': df_results['mae_pct_mean'][(df_results['model'] == m) & \\\n",
    "                                                                              (df_results['numb_vars'] == n)].min(),\n",
    "                                    'max_R2': df_results['R2_adj'][(df_results['model'] == m) & \\\n",
    "                                                               (df_results['numb_vars'] == n)].max(),\n",
    "                                    'min_AIC': df_results['AIC'][(df_results['model'] == m) & \\\n",
    "                                                                 (df_results['numb_vars'] == n)].min()},\n",
    "                                   index=[0])\n",
    "        df_results_best = pd.concat([df_results_best, df_best_tmp])\n",
    "\n",
    "# identify top n models for each metric\n",
    "ntop = 10\n",
    "df_results_tops = pd.DataFrame(columns=df_results.columns)\n",
    "for m in df_results['model'].unique():\n",
    "    for n in df_results['numb_vars'].unique():\n",
    "        df_results_sub_tmp = df_results[(df_results['model'] == m) & \\\n",
    "                                        (df_results['numb_vars'] == n)].sort_values(by='mae_orig_mean',\n",
    "                                                                              ascending=True)[:ntop]\n",
    "        df_results_tops = pd.concat([df_results_tops, df_results_sub_tmp])\n",
    "df_results_feats_best = pd.DataFrame(columns=['model', 'features', 'count', 'oftotal'])\n",
    "for m in df_results['model'].unique():\n",
    "    df_results_sub_tmp = df_results[(df_results['model'] == m)].sort_values(by='mae_orig_mean',\n",
    "                                                                          ascending=True)[:ntop]\n",
    "    for f in var_names:\n",
    "        df_results_feats_tmp = pd.DataFrame(dict(model=m,\n",
    "                                                 features=f,\n",
    "                                                 count=np.sum([f in n for n in df_results_sub_tmp['formula']]),\n",
    "                                                 oftotal=ntop),\n",
    "                                            index=[0])\n",
    "        df_results_feats_best = pd.concat([df_results_feats_best, df_results_feats_tmp])\n",
    "    df_results_feats_best['pct'] = df_results_feats_best['count'].astype('float') / df_results_feats_best['oftotal'] \\\n",
    "        .astype('float32')\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 41,
   "id": "5092049c-c3b7-4811-ad09-eb956cc68ab6",
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/home/spkearney/miniconda3/envs/py_geo_env/lib/python3.9/site-packages/seaborn/categorical.py:1296: UserWarning: 10.0% of the points cannot be placed; you may want to decrease the size of the markers or use stripplot.\n",
      "  warnings.warn(msg, UserWarning)\n",
      "/home/spkearney/miniconda3/envs/py_geo_env/lib/python3.9/site-packages/seaborn/categorical.py:1296: UserWarning: 10.0% of the points cannot be placed; you may want to decrease the size of the markers or use stripplot.\n",
      "  warnings.warn(msg, UserWarning)\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 538.583x396.85 with 4 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "\"\"\"\n",
    "Figure Supplement S3 and S4: Plot model performance of top models using temporal and spatial cross validation\n",
    "\"\"\"\n",
    "for mod_type in df_results_best.model.unique():\n",
    "    fig = plt.figure(figsize=(full_pg, half_pg))\n",
    "    ax1 = fig.add_subplot(2, 2, 1)\n",
    "    pplot = sns.pointplot(x='numb_vars', y='min_MAE',  data=df_results_best[df_results_best.model == mod_type],\n",
    "                          alpha=0.5, scale=0.5, ax=ax1)\n",
    "    plt.setp(pplot.lines, zorder=10, linewidth=1.0)\n",
    "    plt.setp(pplot.collections, zorder=10, label=\"\")\n",
    "    vplot = sns.swarmplot(x='numb_vars', y='mae_orig_mean', \n",
    "                          data=df_results_tops[df_results_tops.model == mod_type],\n",
    "                          dodge=False, alpha=0.45, size=2.5, ax=ax1)\n",
    "    plt.xlabel('')\n",
    "    plt.ylabel('$\\mathregular{MAE\\ (log\\ kg\\ ha^{-1})}$', fontsize=font_sml)\n",
    "    #ax1.get_legend().remove()\n",
    "    ax1.set_ylim(bottom=0.0)\n",
    "    ax1.xaxis.set_major_formatter(mtick.NullFormatter())\n",
    "    ax1.tick_params(axis='both', which='major', labelsize=font_sml)\n",
    "    ax1.text(x=0.875, y=0.9, s='(a)', transform=ax1.transAxes, size=font_med)\n",
    "\n",
    "    ax2 = fig.add_subplot(2, 2, 2)\n",
    "    pplot2 = sns.pointplot(x='numb_vars', y='min_MAE_pct', \n",
    "                           data=df_results_best[df_results_best.model == mod_type],\n",
    "                           alpha=0.5, scale=0.5, ax=ax2)\n",
    "    plt.setp(pplot2.lines, zorder=10, linewidth=1.0)\n",
    "    plt.setp(pplot2.collections, zorder=10, label=\"\")\n",
    "    vplot2 = sns.swarmplot(x='numb_vars', y='mae_pct_mean', \n",
    "                           data=df_results_tops[df_results_tops.model == mod_type],\n",
    "                           dodge=False, alpha=0.45, size=2.5, ax=ax2)\n",
    "    plt.xlabel('')\n",
    "    plt.ylabel('Relative MAE (%)', fontsize=font_sml)\n",
    "    ax2.yaxis.set_major_formatter(mtick.PercentFormatter(1.0, 0))\n",
    "    ax2.xaxis.set_major_formatter(mtick.NullFormatter())\n",
    "    ax2.tick_params(axis='both', which='major', labelsize=font_sml)\n",
    "    ax2.text(x=0.875, y=0.9, s='(b)', transform=ax2.transAxes, size=font_med)\n",
    "    #ax2.get_legend().remove()\n",
    "\n",
    "    ax3 = fig.add_subplot(2, 2, 3)\n",
    "    pplot3 = sns.pointplot(x='numb_vars', y='max_R2',  data=df_results_best[df_results_best.model == mod_type],\n",
    "                           alpha=0.5, scale=0.5, ax=ax3)\n",
    "    plt.setp(pplot3.lines, zorder=10, linewidth=1.0)\n",
    "    plt.setp(pplot3.collections, zorder=10, label=\"\")\n",
    "    vplot3 = sns.swarmplot(x='numb_vars', y='R2_adj', \n",
    "                           data=df_results_tops[df_results_tops.model == mod_type],\n",
    "                           dodge=False, alpha=0.45, size=2.5, ax=ax3)\n",
    "    plt.xlabel('# of input features', fontsize=font_sml)\n",
    "    plt.ylabel('$\\mathregular{Adjusted~R^{2}}$', fontsize=font_sml)\n",
    "    #ax3.get_legend().remove()\n",
    "    ax3.set_ylim(bottom=0.0, top=1.0)\n",
    "    ax3.tick_params(axis='both', which='major', labelsize=font_sml)\n",
    "    ax3.text(x=0.875, y=0.9, s='(c)', transform=ax3.transAxes, size=font_med)\n",
    "\n",
    "    ax4 = fig.add_subplot(2, 2, 4)\n",
    "    pplot4 = sns.pointplot(x='numb_vars', y='min_AIC',  data=df_results_best[df_results_best.model == mod_type],\n",
    "                           alpha=0.5, scale=0.5, ax=ax4)\n",
    "    plt.setp(pplot4.lines, zorder=10, linewidth=1.0)\n",
    "    plt.setp(pplot4.collections, zorder=10, label=\"\")\n",
    "    vplot4 = sns.swarmplot(x='numb_vars', y='AIC', \n",
    "                           data=df_results_tops[df_results_tops.model == mod_type],\n",
    "                           dodge=False, alpha=0.45, size=2.5, ax=ax4)\n",
    "    plt.xlabel('# of input features', fontsize=font_sml)\n",
    "    plt.ylabel('AIC', fontsize=font_sml)\n",
    "    #ax4.get_legend().remove()\n",
    "    ax4.tick_params(axis='both', which='major', labelsize=font_sml)\n",
    "    ax4.text(x=0.875, y=0.9, s='(d)', transform=ax4.transAxes, size=font_med)\n",
    "\n",
    "    leg2handles, leg2labels = ax2.get_legend_handles_labels()\n",
    "    for handle in leg2handles:\n",
    "        handle.set_sizes([10])\n",
    "    ax2.set_ylim(bottom=0)\n",
    "    plt.figlegend(leg2handles, [xv_dict[l] for l in leg2labels], fontsize=font_med,\n",
    "                  loc='upper center', ncol=3, columnspacing=0.5, handletextpad=0.1, frameon=False)\n",
    "\n",
    "    plt.tight_layout(h_pad=0.5, rect=(0, 0, 1, 0.925))\n",
    "    if save_all_figs:\n",
    "        plt.savefig(outDIR_figs_R1 + 'Supp/' + mod_type +\n",
    "                    '_model_performance_top' + str(ntop) + '.png', dpi=600, bbox_inches='tight', pad_inches=0.2)\n",
    "    #plt.close()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 46,
   "id": "1c3a5a2b-acec-487d-9c90-3e05e48ef53f",
   "metadata": {
    "tags": []
   },
   "outputs": [],
   "source": [
    "\"\"\"\n",
    "Calculated weighted relative variable importance for each variable across model types\n",
    "\"\"\"\n",
    "# set max vars based on where cross-validated metrics stop improving\n",
    "max_vars = 3\n",
    "\n",
    "# Calculate decomposed R^2, relative variable importance (RVI) and AIC weights\n",
    "df_results_feats_ptn = pd.DataFrame(columns=['model', 'kfold', 'features', 'formula', 'RVI', 'weight'])\n",
    "for m in df_results_full['model'].unique():\n",
    "    for k in df_results_full['kfold'].unique():\n",
    "        df_results_sub_tmp = df_results_full[(df_results_full['model'] == m) &\n",
    "                                             (df_results_full['kfold'] == k) &\n",
    "                                             (df_results_full['numb_vars'] <= max_vars)].copy()\n",
    "        aicw_tmp = df_results_sub_tmp['AIC'] \\\n",
    "            .transform(lambda x: np.exp(-0.5 * (x - x.min())) / np.sum(np.exp(-0.5 * (x - x.min())))).values\n",
    "        df_results_sub_tmp.loc[:, 'AICw'] = aicw_tmp\n",
    "        for f in var_names:\n",
    "            df_results_sub_f = df_results_sub_tmp[[f in n for n in df_results_sub_tmp['formula']]]\n",
    "            if len(df_results_sub_f) == 0:\n",
    "                continue\n",
    "            else:\n",
    "                for f_form in df_results_sub_f['formula']:\n",
    "                    f_form_omitted = re.sub(f + \":[^\\+]* \\+ \", \"\",\n",
    "                                            re.sub(\" \\+ \" + f + \":[^\\+]*\", \"\",\n",
    "                                                   re.sub(\" \\+ \" + f + \":.+?(?= |$)\", \"\",\n",
    "                                                          re.sub(\" \\+(?<=\\+)[^\\+]*:\" + f, \"\",\n",
    "                                                                 re.sub(\"\\+ \" + f + \" (?=\\+)\", \"\",\n",
    "                                                                        re.sub(\"(?<=~ )\" + f + \" \\+ \", \"\",\n",
    "                                                                               f_form))))))\n",
    "\n",
    "                    # print(f_form)\n",
    "                    # print(f_form_omitted)\n",
    "                    RVI_tmp = df_results['R2'][(df_results['formula'] == f_form) &\n",
    "                                               (df_results['model'] == m)].values - \\\n",
    "                              df_results['R2'][(df_results['formula'] == f_form_omitted) & \\\n",
    "                                               (df_results['model'] == m)].values\n",
    "                    OVI_tmp = df_results['mae_orig_mean'][(df_results['formula'] == f_form) & \\\n",
    "                                                          (df_results['model'] == m)].values - \\\n",
    "                              df_results['mae_orig_mean'][(df_results['formula'] == f_form_omitted) & \\\n",
    "                                                          (df_results['model'] == m)].values\n",
    "                    weight_tmp = df_results_sub_tmp['AICw'][(df_results_sub_tmp['formula'] == f_form) & \\\n",
    "                                                            (df_results_sub_tmp['model'] == m)].values\n",
    "\n",
    "                    df_results_feats_tmp = pd.DataFrame(dict(model=m,\n",
    "                                                             kfold=k,\n",
    "                                                             features=in_var_dict[f],\n",
    "                                                             formula=f_form,\n",
    "                                                             RVI=np.mean(RVI_tmp),\n",
    "                                                             weight=np.mean(weight_tmp)),\n",
    "                                                        index=[0])\n",
    "                    df_results_feats_ptn = pd.concat([df_results_feats_ptn, \n",
    "                                                      df_results_feats_tmp])\n",
    "\n",
    "# calculate weighted relative variable importance (RVIw)\n",
    "df_results_feats_ptn['RVIw'] = df_results_feats_ptn['RVI'] * df_results_feats_ptn['weight']\n",
    "\n",
    "# add input variable group field\n",
    "df_results_feats_ptn['group'] = [input_group_dict[f] for f in df_results_feats_ptn['features']]\n",
    "\n",
    "# calculate sorted mean RVIw for each feature after summing across each kfold\n",
    "df_results_feats_grp = df_results_feats_ptn.groupby(['model',\n",
    "                                                     'kfold',\n",
    "                                                     'features',\n",
    "                                                     'group'])['RVIw'].sum().reset_index()\n",
    "df_results_feats_grp['RVIw_mean'] = df_results_feats_grp.groupby(['model', 'features'])['RVIw'].transform('mean')\n",
    "df_results_feats_grp = df_results_feats_grp.groupby(['model']) \\\n",
    "    .apply(lambda x: x.sort_values(['RVIw_mean'], ascending=False)).reset_index(drop=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 47,
   "id": "262dc897-6ef5-4292-898a-bc58596f9500",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 396.85x396.85 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "\"\"\"\n",
    "Figure 2: Plot variable importance calculated from all model subsets\n",
    "\"\"\"\n",
    "for mod_type in df_results_feats_grp['model'].unique():\n",
    "    fig = plt.figure(figsize=(half_pg, half_pg))\n",
    "    ax1 = fig.add_subplot(1, 1, 1)\n",
    "    pplot1 = sns.pointplot(y='features', x='RVIw', hue='group',\n",
    "                           hue_order=['Bands', 'Veg. Indices', 'BAIs'],\n",
    "                           orient='h', dodge=False, ax=ax1, linestyles='None', scale=0.75,\n",
    "                           palette=sns.color_palette()[6:],\n",
    "                           errwidth=1.0,\n",
    "                           data=df_results_feats_grp[(df_results_feats_grp['model'] == mod_type)])\n",
    "    ax1.set_ylabel('')\n",
    "    ax1.set_xlabel('Weighted relative importance', fontsize=font_sml)\n",
    "    ax1.get_legend().remove()\n",
    "    ax1.set_xlim((-0.10, 0.50))\n",
    "    ax1.set_yticks(np.arange(0.5, len(df_results_feats_grp['features'].unique()) - 1, 1), minor=True)\n",
    "    ax1.grid(which='minor', axis='y')\n",
    "    for tic in ax1.yaxis.get_major_ticks():\n",
    "        tic.tick1line.set_visible(False)\n",
    "        tic.tick2line.set_visible(False)\n",
    "    for tic in ax1.yaxis.get_minor_ticks():\n",
    "        tic.tick1line.set_visible(False)\n",
    "        tic.tick2line.set_visible(False)\n",
    "    ax1.axvline(x=0, color='grey', alpha=0.75, linestyle='dotted')\n",
    "    ax1.tick_params(axis='x', which='major', labelsize=font_sml)\n",
    "    ax1.tick_params(axis='y', which='major', labelsize=font_sml)\n",
    "    ax1.text(x=-0.1, y=1.025, s='(a)', transform=ax1.transAxes, size=font_med)\n",
    "\n",
    "    handles, labels = ax2.get_legend_handles_labels()\n",
    "    fig.legend(handles, labels, loc='upper center', ncol=len(labels), fontsize=font_med,\n",
    "               columnspacing=0.6, handletextpad=0.1, frameon=False)\n",
    "    plt.tight_layout(w_pad=3.0, rect=(0, 0, 1, 0.95))\n",
    "    if save_all_figs:\n",
    "        plt.savefig(outDIR_figs + mod_dir_dict[mod_type] + mod_type + '_variable_importance.png', dpi=600,\n",
    "                    bbox_inches='tight', pad_inches=0.2)\n",
    "    #plt.close()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 48,
   "id": "0b4658ae-c834-4657-8357-e3fcacf44595",
   "metadata": {},
   "outputs": [],
   "source": [
    "\"\"\"\n",
    "Create and save tables for top candidate models (at least one AICw > 0.50), sorted by MAE\n",
    "\"\"\"\n",
    "# calculate AIC weights (AICw) for all models with fewer than max_vars\n",
    "df_results_full_sub = df_results_full[df_results_full['numb_vars'] <= max_vars].copy()\n",
    "df_results_full_sub['AICw'] = df_results_full_sub.groupby(['model', 'kfold'])['AIC'] \\\n",
    "    .transform(lambda x: np.exp(-0.5 * (x - x.min())) / np.sum(np.exp(-0.5 * (x - x.min()))))\n",
    "\n",
    "# Calculate AIC rank within model group and K-fold, then tally number of times a model is best based on AIC\n",
    "df_results_full_sub['AIC_rank'] = df_results_full_sub.groupby(['model', 'kfold'])['AIC'].rank()\n",
    "df_results_full_sub['AIC_tally'] = df_results_full_sub.groupby(['model', 'formula'])['AIC_rank'] \\\n",
    "    .transform(lambda x: np.sum(x[x == 1]))\n",
    "\n",
    "# Get stats by model\n",
    "df_results_full_grp = df_results_full_sub.groupby(['model', 'formula']).mean().reset_index()\n",
    "for col in ['mae_orig', 'mae_kg', 'mae_pct', 'AIC', 'AICw', 'R2', 'pe']:\n",
    "    df_results_full_grp[col + '_sd'] = df_results_full_sub.groupby(['model', 'formula']).std().reset_index()[col]\n",
    "for col in ['AICw', 'mae_kg']:\n",
    "    df_results_full_grp[col + '_max'] = df_results_full_sub.groupby(['model', 'formula']).max().reset_index()[col]\n",
    "\n",
    "# define fields to include in saved tables\n",
    "out_features = ['formula',\n",
    "                'mae_orig', 'mae_orig_sd',\n",
    "                'mae_kg', 'mae_kg_sd', 'mae_pct', 'mae_pct_sd', 'pe', 'pe_sd',\n",
    "                'AICw', 'AICw_sd', 'AICw_max', 'R2', 'R2_sd']\n",
    "\n",
    "# save tables as .csv's, sorted by log-transformed MAE\n",
    "for mod_type in df_results_full_grp['model'].unique():\n",
    "    table_tmp = df_results_full_grp.loc[(df_results_full_grp['AICw_max'] >= 0.50) &\n",
    "                                        (df_results_full_grp['model'] == mod_type),\n",
    "                                        out_features].sort_values('mae_orig', ascending=True)\n",
    "    table_tmp['formula'] = table_tmp['formula'].map(\n",
    "        lambda x: 'Y ~ ' + ' + '.join([in_var_dict[v] for v in re.sub('np\\.log\\(bm\\) ~ ', '', x)\n",
    "                                      .split(' + ') if v in in_var_dict.keys()]))\n",
    "    table_tmp['mae_orig'] = table_tmp['mae_orig'].map(lambda x: '{:.4f}'.format(x))\n",
    "    table_tmp['mae_orig_sd'] = table_tmp['mae_orig_sd'].map(lambda x: '(' + '{:.4f}'.format(x) + ')')\n",
    "    table_tmp['mae_kg'] = table_tmp['mae_kg'].map(lambda x: '{:.2f}'.format(x))\n",
    "    table_tmp['mae_kg_sd'] = table_tmp['mae_kg_sd'].map(lambda x: '(' + '{:.2f}'.format(x) + ')')\n",
    "    table_tmp['mae_pct'] = table_tmp['mae_pct'].map(lambda x: '{:.1%}'.format(x))\n",
    "    table_tmp['mae_pct_sd'] = table_tmp['mae_pct_sd'].map(lambda x: '(' + '{:.1%}'.format(x) + ')')\n",
    "    table_tmp['pe'] = table_tmp['pe'].map(lambda x: '{:.1%}'.format(x))\n",
    "    table_tmp['pe_sd'] = table_tmp['pe_sd'].map(lambda x: '(' + '{:.1%}'.format(x) + ')')\n",
    "    table_tmp['AICw'] = table_tmp['AICw'].map(lambda x: '{:.2f}'.format(x))\n",
    "    table_tmp['AICw_sd'] = table_tmp['AICw_sd'].map(lambda x: '(' + '{:.2f}'.format(x) + ')')\n",
    "    table_tmp['AICw_max'] = table_tmp['AICw_max'].map(lambda x: '{:.2f}'.format(x))\n",
    "    table_tmp['R2'] = table_tmp['R2'].map(lambda x: '{:.2f}'.format(x))\n",
    "    table_tmp['R2_sd'] = table_tmp['R2_sd'].map(lambda x: '(' + '{:.2f}'.format(x) + ')')\n",
    "    if save_all_figs:\n",
    "        table_tmp.to_excel(outDIR_tables + '/Main/candidate_models_' + xv + '_table_' + mod_type + '_v3.xlsx',\n",
    "                           index=False)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 49,
   "id": "b9e22629-8dc1-412a-b35b-d39efcf148e2",
   "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>mae_orig</th>\n",
       "      <th>mae_orig_sd</th>\n",
       "      <th>mae_kg</th>\n",
       "      <th>mae_kg_sd</th>\n",
       "      <th>mae_pct</th>\n",
       "      <th>mae_pct_sd</th>\n",
       "      <th>pe</th>\n",
       "      <th>pe_sd</th>\n",
       "      <th>AICw</th>\n",
       "      <th>AICw_sd</th>\n",
       "      <th>AICw_max</th>\n",
       "      <th>R2</th>\n",
       "      <th>R2_sd</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "Empty DataFrame\n",
       "Columns: [formula, mae_orig, mae_orig_sd, mae_kg, mae_kg_sd, mae_pct, mae_pct_sd, pe, pe_sd, AICw, AICw_sd, AICw_max, R2, R2_sd]\n",
       "Index: []"
      ]
     },
     "execution_count": 49,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "table_tmp"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "id": "0cdd4202-8bc3-4362-bb18-71fffe24d8a1",
   "metadata": {},
   "outputs": [],
   "source": [
    "# define final models\n",
    "lr_form_fnl = df_results_full_grp['formula'][97]\n",
    "\n",
    "# define final model parameters\n",
    "wtd = False\n",
    "\n",
    "# save models as objects\n",
    "if wtd:\n",
    "    lreg_all = smf.wls(formula=lr_form_fnl,\n",
    "                            data=df_out_all,\n",
    "                            weights=1 / df_out_all['bm']).fit()\n",
    "else:\n",
    "    lreg_all = smf.ols(formula=lr_form_fnl,\n",
    "                            data=df_out_all).fit()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "id": "7d5a1c42-7f90-43ac-9e32-da6f0ffd346e",
   "metadata": {},
   "outputs": [],
   "source": [
    "#pca_outliers"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "id": "0143973b-88f9-4b15-b3c9-0799d5240119",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "if log_xfrm:\n",
    "    df_out_all['bm_pred_tb'] = np.exp(lreg_all.predict(df_out_all))\n",
    "else:\n",
    "    df_out_all['bm_pred_tb'] = lreg_all.predict(df_out_all)\n",
    "df_out_all['bm_resid'] = df_out_all['bm_pred_tb'] - df_out_all['bm']\n",
    "df_out_all.loc[df_out_all['bm_pred_tb'] < 0, 'bm_pred_tb'] = 0\n",
    "df_out_all= df_out_all[df_out_all['bm_pred_tb'].notnull()]\n",
    "xy_bm = np.vstack([df_out_all['bm'],\n",
    "                   df_out_all['bm_pred_tb']])\n",
    "df_out_all['z_bm_tb'] = gaussian_kde(xy_bm)(xy_bm)\n",
    "ax = plt.subplot()\n",
    "sns.scatterplot(x='bm', y='bm_pred_tb', data=df_out_all, hue='z_bm_tb', \n",
    "                hue_norm=(1.1e-10, 1.1e-6), palette='magma', edgecolor='none', ax=ax)\n",
    "sns.regplot(x='bm', y='bm_pred_tb', data=df_out_all, ci=None, marker='', order=5,\n",
    "           line_kws={'linestyle': 'dashed', 'color': 'black', 'linewidth': 2}, ax=ax)\n",
    "ax.plot([0, 6000], [0, 6000], linewidth=2, c='black', linestyle='solid')\n",
    "plt.legend().remove()\n",
    "#plt.xlim(0, 3000)\n",
    "#plt.ylim(0, 3000)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "id": "2b35b7ef-765c-4c05-b2ed-785bbff2d3b6",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[<matplotlib.lines.Line2D at 0x7fb131832e30>]"
      ]
     },
     "execution_count": 16,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAZQAAAEKCAYAAAA1qaOTAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjUuMiwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy8qNh9FAAAACXBIWXMAAAsTAAALEwEAmpwYAADyN0lEQVR4nOyddZwcRd7Gv9XdY+ub3SQbD3GiWAgS3N0huB1w2PEihx12d3DHweHu7u7BQtAAQQIJgbj7Zn20u6veP6rHZ5MACcnBPHyWzFRXV1f3zNRTPxdKKYoooogiiiji18JY3xMooogiiiji94EioRRRRBFFFLFWUCSUIooooogi1gqKhFJEEUUUUcRaQZFQiiiiiCKKWCsoEkoRRRRRRBFrBeudUIQQDwohlgshpmS0XSWEWCSEmOT97Z1x7BIhxEwhxDQhxB4Z7ZsLISZ7x24VQojf+l6KKKKIIv7IWO+EAjwM7Fmg/Sal1Cbe35sAQojBwBhgiHfOnUII0+t/F3Aq0N/7KzRmEUUUUUQR6wjW+p6AUuojIUTvNex+APC0UioOzBFCzAS2FELMBSqUUhMAhBCPAgcCb61qsNraWtW795peuogiiiiiCICvv/66XinVMbd9vRPKKnCWEOI44CvgfKVUI9AN+Dyjz0KvzfZe57avEr179+arr75aezMuoogiivgDQAgxr1D7hqDyKoS7gL7AJsAS4AavvZBdRK2iPQ9CiFOFEF8JIb5asWLFWphqEUUUUUQRsIESilJqmVLKVUpJ4D5gS+/QQqBHRtfuwGKvvXuB9kJj36uU2kIptUXHjnkSWxFFFFFEEb8QGyShCCG6ZLw9CEh6gL0KjBFCBIQQG6GN718qpZYArUKIrTzvruOAV37TSRdRRBFF/MGx3m0oQoingB2BWiHEQuBKYEchxCZotdVc4DQApdQPQohngamAA5yplHK9oU5He4yF0Mb4VRrkiyiiiCKKWLsQf+T09VtssYUqGuWLKKKIIn4ehBBfK6W2yG1f7xJKEUUU8fvHggULuPPOe1m4cCG77bYrRx89BtM0V39iEf9TKBJKEUUUsU4xb948Ro4cTX19PQBPPPE048Z9wMMP35/q09raysyZs+jfvx9lZWXra6pF/EpskEb5Iooo4veDW265I0UmSTz22JPMnDkLgNtuu5Nu3fqw+eZb061bH+655/5CwxTxP4AioRRRRBHrFHPmzMlrU0oxe/Ycvvvue84553za2toALamcccZf+PHHn37raRaxFlAklCKKKGKdYscdd8hrKy0tZdSokbz+er4zplKK119/87eYWhFrGUVCKaKIItYp/vznUxg1amTqfSDg5+67b6OyspKOHWsLntO5c+ffanpFrEUUCaWIIopYp3jssSf54ouJqfe27VBZWQnAmDGH0b17dtq93r17ccghB/6WUyxiLaFIKEUUUcQ6g1KKq676Z1ablJJ//OMaACoqKvjkk3GcfvppbL31KM4663Q+/vh9SktL18d0i/iVKAY2FgMbiyhinSEWi1FSUp3X3qFDB+rrF62HGRWxNtBeYGNRQimiiCLWGYLBIFtttWVe+8475xvqi/jfR5FQiiiiiHWKu+66jS5d6lLv+/fvx3XX/Xs9zqiIdYVipHwRRRSxTjFixHDmzJnGuHHj8fl87Ljj9us07Uo4HObFF1+hubmZ/fffh549e66zaxWRjaINpWhDKaKI3w1mz57DDjvsyqJFuhySz+fjiSce5tBDD17PM/t9oWhDKaKIItYYy5Yt4403xjJr1uz1PZWfhb///eoUmQDYts0555yP4zjrcVZ/HBQJpYgiisjCzTffTq9eg9h//0MZOHAEZ5117nqdz6RJ33P//Q8zceLXq+1bqM+SJUtZtKjoUfZboEgoRRRRRAqzZs3m/PMvwbYdQKCU4q677uONN8aul/mceeb/sfnm23DaaWex1VY7cPzxp7AqNf3w4UPz2mpra+nSpUuB3kWsbRQJpYgiikjhv/+9xXslvD8DELz//ge/+Vw+++xz7r47O/Pw448/xTvvvN/uOVdc8Tc6dOiQei+E4Nprr8bv96+zeRaRRtHLq4giikhh/PiP0USSje7du/7mc5kw4YuC7Z99NoE99ti14LHBgzdm6tRveeKJp2lubuHggw9g+PBh63KaRWSgSChFFFFECtForECrWC+5tQYOHFCwfeONB63yvE6dOnHuuX9ZF1MqYjUoqryKKKKIFPbee4+8tpEjN6NXr17rZS477bR9VttWW23JwQcf8JvPpYg1Q1FCKaKIIlK45pormD59Bh988BEAgwYN4NFH71svczEMgzfffJmnn36er7/+huHDh3H00Ues0h6yZMlSbrzxViZN+o5NN92E88//SzEV/m+I9R7YKIR4ENgXWK6UGuq1dQCeAXoDc4HDlVKN3rFLgJMBF/iLUuptr31z4GEgBLwJnKNWc3PFwMYiiiiMadNmEIlE2GST4QihbSpKKRYtXEJtxxqCwcB6nmE+2tra2GSTrZgzZ26qrV+/vnz77QRKSkrW38R+h9iQAxsfBvbMabsYeF8p1R9433uPEGIwMAYY4p1zpxAimcPhLuBUoL/3lztmEUUUsYYYOLA/m246IkUm4z/4jKGDd2RA/23ZqNdIbrl5/Ugtq8Kzz76YRSYAM2fO4oUXXlk/E/oDYr2rvJRSHwkheuc0HwDs6L1+BBgPXOS1P62UigNzhBAzgS2FEHOBCqXUBAAhxKPAgUB+fdEiiijiZ6GlpZUxR/yZlpZWAJqbW7nk4n/RpUtnpkyZxoIFi9lpp204+piD283RtXDBYkzLokuXTutsnkuXLv1Z7UWsfWwIEkohdFZKLQHw/k1+C7sBCzL6LfTaunmvc9vzIIQ4VQjxlRDiqxUrVqz1iRdRxP8iXNdtN2Dw/fc+TpFJJk495a/ceMO9PPfs65xx+qWcesqFeX2WLFnOnrsfxcAB29G/7zYcesgpNDfnj7U2sOeeuxds32uvwu1FrH1sqITSHvId5EGtoj2/Ual7lVJbKKW26Nix41qdXBFF/Bz88MNPfPDBx8RihVx105j07WSOPuo0tt5qDy666O80NjattTk0NjZzwvH/R22HYXTrshmXX3YdCxcuYc4cvW+TUvL8c28WPNe2JQILgQkInn36dcaOHZ/V55y/XM7HH+t4EqUUb705jr9duuap6+PxOPX1DWvUd7PNNuG6664hGAwCEAqFuPHG/zB06JA1vl4Rvw7r3SgP4Km8Xs8wyk8DdlRKLRFCdAHGK6UGegZ5lFL/9vq9DVyFNtx/oJQa5LUf6Z1/2qquWzTKF7E+EIlEOPzwExk79j0AamtrePLJ+9l55+3z+k6fPoutRu1OOBxJtW2xxSZ88umbKfvGr8GYI87g9dfeTb0XmAih95mbbT6MffbdiX9cdQuujJO7RzONgNdXZM3lnHNP4vAj9uH7737itFP/ipRu1nm1tR2Yt2Aiq8O//30LN914Dy0trWy22TDuve8Ghg7deLXnNTQ0MG3aDAYNGkB1dX61yCJ+Pdozym+ohHI9sFIpda0Q4mKgg1LqQiHEEOBJYEugK9pg318p5QohJgJnA1+gvbxuU0oV3lp5KBJKEesDV1/9X666KnuX3qVLZ+bM+R7LyjZrXnTR37n5prvzxhj3wctsu+2oXzWP5uZWenTbAiklAAKDtI+LRlVVBU1NLSilkMpGKYlAYBi+FPEk/20PUtpIlc7227//Rkz6/r1VnvP8869xzNFnZLX17NWdH3/85FfVUnnlldd56qlnCQQCnHLKiYwevc0vHuuPjA3Wy0sI8RQwARgohFgohDgZuBbYTQgxA9jNe49S6gfgWWAqMBY4UymV3P6cDtwPzARmUTTIF7GB4t138/NiLVmyjMmTp+a1N6xsLDjGynbaCyEcjvD5hG9ZsmR5VrthCAwjcwnIXw6amloAnRPLNPxYZhDTDGSQyOqlJMPwZfU748wTVnvOc8++mtc2f95Cvvjim9We2x5uuuk2Dj54DM899yKPP/4UO+20J6+88vovHq+IfKx3QlFKHamU6qKU8imluiulHlBKrVRK7aKU6u/925DR/xqlVF+l1ECl1FsZ7V8ppYZ6x85aXQxKEUWsL/Toke0vIjDwmSUcdeSZnHj8/zFn9vzUsf32z/d+Ly8vY6edRq/RtV564W0GD9iNvXY7nmGD9uCiC65NGd/Ly8s4+JC9Vnl+oXiTmtpMNdKa/cy61NWx7bYjue+B/3Lqacestn8gUDh4saQktEbXy4Xrulx77X+z2qSU/Otf1/2i8YoojPVOKEUU8UfDeeedSSiUXBgFllmKEBbz5i7kmWdeYffdxhCJRAHYf/89ueSS/0sZmrt168KTT91LeXnZaq/TsLKJM067jBbPq8p1Xe69+0lefTltM7nt9qs59bRjqKmtpqamMm+MU087imeev5PR241kwMA+nPN/JzHx69c4+y8n0bGmDp9RaIEXiJylZfTorXjnvac56qiD1uAJQWtrJK+tR49ubLJJfnr6NUE0GqW+fmVe+4IFCwv0LuKXYoOwoawvFG0oRawvTJ48lTvvvJ8Jn33D9Gnz8o4//MgtHH7E/qn3TU3NLFm8lP4D+ubZWVqa23j1lfeJhKPss9+OdOtex/SfZnP7rY/zxGMvI5VLpiRx9LEHcvtd/yg4r1defocH7nuaSDTGIYfuxWl/PjpHLaYTSG4+7EBWrGhEKYkrEygUSYdL4am3DGEBAqlsNtl0CB98/OQaPZt4PE5dp02Ix2OpuQsMNtqoF1N/+miNxiiEbbfdmc8//zKr7ZhjjuSRRza8IM0NHe3ZUNZ7YGMRRfwRMWzYYO6660auuvK/XPefO/KOt7a2Zb2vqqqkqipfgpg+bQ4H7P1nVqzQWuHLL70ZQxi4riYQ0whgKIWrEijPMF5X1767/AEH7s4BB646bmPce5+zYoW24Qhh6L9M1ZcSngeYJhYTH/PmLFnlmJlQSrsYC2FgZhj8V2f8Xx3uvfd29t//MObO1QS+2Wab8J//XP2rxiwiG0WVVxFFrEccdPDeee6/gYCfffbdbY3O/9c/70qRCWi1luNku+kKITCFtklUd6jkhJMP+1VztqxsLytD+BCYdO7ckcOO2Jvy8sq8e2prja1xQKNSit12z3ehPvqYg3/5pIEhQwYzffr3jB//Np9//iETJ35CXV0xceTaRJFQiijiV2DWzHl89unXxOOJVNv8+Ys55eSLGDZkDw456M9M/PK7ds8fMWIwd951LZ061QDQs2c3nnjqzoJSxLdfT+WsU6/mmMMv5LGHX0VKybffZHuGZcaRZLULwYEH7cF7HzxB9+51v+heI+EYj9z/KuPe/praDrVZY/v9AZ576Q7uuf9f+KzCio/XXxm/2ms88tALDB6wG++N/YKArxS/FaK8rJzzzj+NCy86Y7Xnrw6mabLddtsycuTmv3qsIvJRVHkVUcQvQDye4KQT/srrr+lytDU11Tz0yH/ZepvN2Hfvk5g3Vxt7589bxGeffs1nX7zIRhv1KDjW8ScczlFHH8SKFQ3U1XXMs1kAfPn5ZA7d7xyv1ju8/87nTJ0ykyFD+7No4TIATBHAMgI4Mo4iLaUITIJBHzffdgWVVeW/6H5jsQSH7nM+U76flWrr0rE7LdEG+vbtwcWXncbQYbogVrfudbRMnZU3Rlm5zvgrpSx4jz/+OJPzzrnaU3cJUNq4n4iBzwww7r3P6N6zK4MH9/tF91DEukdRQimiiF+Au+96PEUmoONC/nTShbz15gcpMkkiEonyxGMvr3I8n89H166dCy60SxbVc/tNT6TIJIknHnmdM84+morKckzhxxLaxdcUPkC7IweMCoJmBdghzjr5P7S0hNudQ2NDC3fe9Dx/PfMWnnns3azrvfbi+CwyAWhqCPPIo9fz/kePstvu26bar7z6rLyxy8tL2WyLIZx+0j/pU7cXQ/ocyH///XBW/rCxb3zYbj6xm298iDGH/YXRow7lhGMvwLbt7Ge0ZDknnXgevXqMZJut9+fll8a2e59FrDsUCaWIIn4B3nv307y25ctX8tOP+TtzICt1yurw0w9zOPaQy9hq2PEM6XUoWw07jo/fn5yygyRh2w6L56+k0t+ZErMavxHCwo8QBpYI4TfKMDIi3z/64Bv22f4vvPTM+FRbPKZVdU2NrRy061/57z8f44WnxnHJObdz+nHXpvrNmlnYvfbriT/mte28y1bcfvfl1HbU8SpDh/fn2Zdv4dLzbuHVl8Zj2w5Nja3cdN1j3H/3i6nzqjvkOx0Uwqsvv8eTj2cHPh56yJ945ulXqK9v4LtJP3DM0WelcogV8duhSChFFPEL0K1bvjHXMAz23X8Xysryizltu82WxKLx1Y77/JPvsed2Z/LRB9+wZNEK2rx4DCkVpvBjZGipe/bqwg1XP0FTY9rYbQgTAwshRBaZJDF//hIuOP0Wzj7pv+y25dkM7XYU+4w+j+v/8Rjz52aneR/39kS+nTgNgC22HFxwvlO+nV2w/fAj9+aHGW8wd8kHjPv4UXr37sYH7+fn73r+6XdSrw8+ZA+6dGnfA02g3ZABPhyfJouvv/qeSd/+kNVXKcVDDz7T7lhFrBsUCaWIIlYD23Z4/+0veO3FD1MqozPOOo5QKJjVb8yR+zFkyACeeOpW+vbTNdhrKzuzUccRnHPirYwadBIP3pmfUiSJSDjGFRfevcrYcx3bAV26duSCi05m+bL8TLyG97MurD7SC/LYVz5n7iztyjv9x/m88MR4DCx8BLSU4/WbO3sxrz7zMQ/c+BYmvpzrWMyatrjduQohKCnRz0gYomAyy8y8XBWV5bz5zsMcevjeec/WEH5Mw4cpAoCgZ8+uqWOJRIJCSMQLtxex7lA0yhfxh4Druti2SzCYVhsppXjrtU8Z//5XdOlay5HH7UVdl5qs8xYvXMGRB17C3Nl64QwEfNx41wXse+B2vPfBE9x91xMsW7qCPfbcgRNOPBSAHXfaiq+/fZ2Z0+Zx0C4X09qspYy21gjXXPYQQzfpy5bb5KdUnzNrUSpCHsDCDwgcEiQDE/fcZzSn/eVgNtlsEA31LZimgevKrHGGbdKfI0/YnYkTfuSlZ7Pzhhleqvn85yMx8WFi6VQwBLGJcd/1bzJ/9nIUihAVuDg4xFFesGGf/vllh5RSvPXS5zx291iam8LsvNdmnHnRIeyx9za89fonWX3HHJOd+qVX727c+8C/UErx0gvvcNrJl4MyUmQkhCAUKGHEJkNYuGAp3XvUseWoTenTpyezM1LWAIw58oC8uRWxblEklCJ+97j5uid44O6XaGkOs/1Om3HtzefQrXsnrrjoTh65/7VUv0cfeI1X37uFHj3TbrXXX/NIikwA4nGbs07+D1WVFYzeaQSXXfp/TPjgBzrWVaUM6j9+N49XnvyUGT8tIBHNjgkBGPvqhIKE0qN3HaFQgGg0QQlVmN7PUyFxcbDwM+/7FqZOXMSnY39i0bx6Ro3clM8+/zo1RjAU4Nqb/8LQEf046rh92Grb4dx6/VMsXliPwExJL4VgYmHiS0knJhYLZq/wYt8FKhkBj0mCKKGSAH+58AiklEz8eBpNDa1M/mY2zz70AZFIDIVC4fLgbW8y6YuZ3PbU+QSDQd5/dwIVFWWceOqBHHfSfllzeOX58Tzx4JvU1zfTo3dHXWslh/9cR3HaSVdgGAbHnXgg/7nhAp5/8X7OPvNvfPrpRDp2quGvfz2dfffbjR9/nME7b39IXV1H9j9gjzzJp4i1i2LqlWLqld81nn5sLBeeczOgvZ4UkuGb9ufeR65gm02OT6VutwjgI0iv3nVccNXRvPfaN7z/+lc02ctxZNrbyVAmPgKUV5QxdNAgvvtmJtKTEDYe0YuTztmLv536QFZwYZQ2HNL2k7P/ejj/d8mRBed7z60vcOOVz+IX7eXIypcuBg7vTlU3HzW1lRx/6n5sPGSjrOOtLWFOO+ZavvhkCgB1XWpoaYoQi2WohJQgSFnB8RVpCcjFQQgYc+pOHHnCHtTUVnLaQTcyfcoCr6/KOE9ToSYjhc9vkEg4BII+Djt+Zy665qgsr7b7bn+Rf13xYPp8pYiqJpTKlsBy8cAj17DfgTsDOi1MIODHMAzuvPMRLrwgHQnfv/9GvP3uU6mYnyJ+OTboeijrC0VC+d+DlJIXnxrPQ3e/xrw5S+jRs46evesYvlk/jjxhdzrUVjBnxhKmfDuHAYO7c8Wlt/PFJz8QoAwTC4XCJsa/7zid88+8AQA/JQQoTV1DYKQW1jBNuOiFN6TKKaNDaoFMEM/bPQf8fuxEtnuvoQQuNjYJfGXwxqc3071n4drqi+bWs9+WF2M7Tv5BJTAwUCL/N/v8x39nwJDCcS5JTJs6j3BblBGb9+e7r2dww9VPMu2H+QwZvhGTv5yDEy0svWQSisShorwMp8WgtDxIz0Ed+f6rwp5t+lxNKpJ8Se3S/xzH0aekMwKMHHgM9Suasvq4KoE0YyRy3IQzcfSx+3HT7ZdmtTU2NjOg32ii0exqmCeedDhDhw2iS5fO7LXXTvh82XahItYMRUIpgCKh/G+guSnMsw+/z6cfTOaH72fT2hJBKb1UZS7o3Xt2ZNfdt+Kp+9LxIVWdg6xY1uTZDtI45bz9uPfeZwi3RSmjRmfHVeBD1+5whV4AHWwiNGIokxq6p4hGIrFFAaOvAiMjUj0og/hJ2202HtmD7fYYzvIFjWy6Qz92OHB4apfesLyVS49/gAmfT0YKN2/cECXY2Km5ZWKLLQZx+a3HsdGgXxYF//rzn3HpqfcXPJYkFG03EQRVSYpUbWwoQHCZcHCyAi1Tc95mII+8fhmgNwr9Ox2AlPljhUJBdtt3JJuM7M9F51+fd7yqupJT/3w4/3fBcanEmRMmfM1uu4zJvg8ls6SdYcMG8dbbT9GhQ9Uq519EPoqEUgBFQtnwEQnHOHyXy5k1fZFWp+RIBLk7X1P5sDK8kTTt5O/2R20/hH2P3JKL/nIrAbsSn/JRoSpTNoYECVpFKwhwSKCQlKsOWeMWklAMZaTaTGVSqkrJhbZk6OvsfuTmXHjn4dxx8au8+sAEXEcSJ0GzaMoa268C+PFrOSeXyBSUqFIMDHY6YASX3n4UpeU/31aw38hLmDdrWfb9CIGrXKpryomsSOBHe1kpFK7QcpcSMjWPQvW2XJyCEsoOu2/CnU+fTzyW4JP3J3PvHS/w5eff5/RKp8Kv7VRFdReL776ZjkAgcTwVmz5+wEE7c9vdlxEMBli5spGB/bcjFtOqRqUU6Vp8aVx8ydlcceV5P+cxFcEGXLGxiCIysXBOPWOfmciP32qPnTdfmMDs6Z5RvMBilavzz9ThrwrlFSEOOWJXJkx+hO4bdaRcVWQZrP34CaHtGBZ+jj15n7zrmjlSj6EMaqnSpIImlELInPE7T33Nc7d/xEv3fIrr6IU5gJ8OqgOW8mEpH34ZSF3LwsoeV0FABVNz/+CV77jtspeYN20Z4ZZsdc/q8OjYv1FeVoJQAkMZhGSIjoFazj73MLp16EyAEIZHhyYmljIxMbAsE6EEPpVfFGtVn8eRf9qN+256jS27n8b/HXsbUz9fTIVVm/OE0q9XLGtg7vTl+IwglhHAJ0owCeiHgOSVl95jxKD9+eD9L6ipqebSv5292nv+auKkNX4+RaweRS+vItY7lFKMe3ESj9/8HtMmpyOyN9m6L1Gp4z78yk9crD4wMNeLSSAYNLQXP01J1xwRQjDmpF0B6Nipmkv+fhxXHfMYBoKgYQGKmHQJGAHK64Icfdpu7HvA1nw/bg6zZ6eD/0wsggE/PmmCLSgTIUxMulJLSY8ApaEQs37KDhYE8GNiCZOEchFC8MT17+f1sbAIqRCu0DXchbf7FwiCBJFKIpVEL+lpgjEQvP3oN7z9yDcES3wcde7OHPvXXVf73AAe+OdbmC1+KjI8veIRm8dveB/DyncIEBhYwuKa+0/iX2c+RbQ1ASqpBsMjGR9KKBxsXFxPNrMoLQ0QCvq59Z8vZI0pHdhx9NaU1pq8+fJnWcdcbCKRjBxlQgBK57v30NjYwhmnXMWkH1/hvPNPY8cdt+Htt8dTUhLiisv/k5e+ZuPBA9bo2RSxZigSShHrHTdf8CIvP/Apdo5aZNKEWZoeBFSqalaqelyRsSAosgzUBxy+PTO+XsysmWk33zJ/iBvuOYsXnh7Ph29MorylBNmseOjPb7L89EYOPGd7ho3sQ8C0qBT+VLxDqaHot10P/v3SqcyfuowLd7wT1WrTw6oiICwMUzBgdC8Ov2BnLtwru0CTgaAkbBFbHsbvEUcSpue86zMMAl4ke6S1cABewLAo6RCksT6sd/oqrfIzMLStRumGAD4sj1ikUkgUsYjNg9e8zaDNezBy54Gr/Rzee0HXay/o6VXA0UogOOuqAxk0rCfR1gQCQUAF8RNAeicIhGeb8uPiJKdLPOxy7pi79fxFtmfYN19Op7w6iN8IYctEyv6S6RyQhCygxlq5solJ3/7IlqOGM3zExmy2+TAAmptbuPbft6X6devWhb+c86fVPpci1hxFQilivWL5wkZee2gCsh3ViF/4KFeVmFjUyI60imYSIoGFxcZ9+nDLa2fy3dcz6dOvKz16duKkIf+hE9XESBDAR0WilA8en8QF1xxF4+dtzJq/CIDWhghPX/M+oTI/W+03lLqKcmJt6YXdEILGGU1IV/LyrR8TbYtjGgKfl3gRCdM/msc1Hz2UNV8BlFomdmsCA+jkCxF2bdpcJ0UmlmFgCoEldFmqoGGSyAlONBGUixJohv2P25ofv51Px65V9Ohbi2kZbLPHEGb9sJg7r3oNERFYGWlWTAQomXqmn7w+ZY0IRRUwiGfeVyE8e9PH3H/F2Jy+haQZgYmFo5zUYLGw/hwd0h5cLg6OLYkvj2eQkoFCUVPdgZWNK/PGzZ21EIIXnnuLIw49g7bWCHvtvSM33nIZV1x5HjvvvC1vvz2eLl06ceRRB1NdvWb5w4pYMxQJpYjfBG8/8AUv3PAhkZYotd2qOO2WA9h4q94snd+IlKrgrhggZFhUiQ7ElI0rBWWiI7V+P0HTxN9mUT+9kd333RKAOVOWEG1JUCnKyFwmZny9kMUz65n17aJUW4kpCJqCl696h5eufJtYPH+n27isjaevHMvS2XoRKzRDBQRNg5hHCD5DYGSkGBFCUGb5UUrgKKXVaqaJT6RTkfiEXoAj0kEqhSXS0ouSMGrbgVx865i8a2+2bX92P3RzDu93Nbm+NQYiRSjhpmxbSiJm8/ZjX+EPWOw8ZlN8fr0MjNxpIB+9PjnvOpbPpM/gLvz03YKcaxi0Nka8Z9Me5aRRqI9lmTiuJhTluVAASKWDOTPPbWtOUNuxA/VeQTGFwhAGuTw4dHg/7rsnXW749dfeJxyO8OKr9zB6u1GM3m7UaudaxC/DBm2UF0LMFUJMFkJMEkJ85bV1EEK8K4SY4f1bndH/EiHETCHENCHEHutv5kUAxNoSTP14NlfseQ8PXvwGzSvasOMuS2av5O8HPETD0hb6j+hGqCzgRX7kVC40THqEgnQL+RhYVsrgilI2KgkS9PI/JVoT3HHiU6xc2ARA557VBEry4wp6btwZwzQyxhWELAMy0nmYBdZDU8AnT33LgC1WHd9R6TMptQxMAZZZ+CcVNEwsDCp9viwySV4/aJiUGD7KTD9Bw8o63rlXdaEh9RxNs2COLEOIFCktmZne1U98dxqH9vg7d1/4Gree8xKH9fwH075ZyNypS9njkM3p1S87OWPn7tUMHNwdSxqUBP0YnpTlF9p6A3qxT6rbVmWEL3TszL8dyKaj+lNSFqRz12q0hkwCElNYmMJK2YiUUhxy2O784z9n0ndANxQuwjDYfvstOeLIvdhtj2349/Xn4cq4R04KLacpxo37jGXL6rPnoxRLly7HKRTzU8Qvwv+ChLKTUirzm3Ax8L5S6lohxMXe+4uEEIOBMcAQoCvwnhBigCrkK1jEOoOUinfv+pQ3b/uESKs2oueqc0CrVx4+/1X88QTDQiW0Cov5kSiOEvgMCJkmtf4APmHgM/Ti7jMEpT4DVyqaEwpHgRN3mfDMJAZu3pUJD3zOsI5B6hskyyMuZX6D0lI/2+7RH2IJevSrYcHMlfgLsEeJz6Q1kVmUCgKmgR1zUCta6FpiEnYUYSd7UbSEJoRyn0W5D6wSHy0t+c4DFX6TUmUiJQUsAWAUIDQFjNixL4O36t3u8y6rDNFrQCfm/JTt7usXBj5h4pcmjUvS2YivPfGplDcZgB13uHjfe3Fi+t4tv8mAPl1ZMrcBwzBoWxShbZEnhShBZWkp8UhaonC8uzEw8CFSC3gyTiV9L/lxQ7setBnHnLUbJ/zfnqm24w/4JxM+mZwie/Dq1ntxR4sX1rP1DkNYUd+Quk7vPt257qbzU/E8Tz7xYt61AdyM7AXvvjOec865jNmz59GpUy1//8dFnHhivhRYxM/D/wKh5OIAYEfv9SPAeOAir/1ppVQcmCOEmAlsCUxYD3P8w6FleRufPjaRDx/8gkhrnFgGjRfasxoCFn8+h6S+otSy2LiinJjjgEh/LU2h+2aSgGkIqgJQH1MYwHf3fcwP9+lxDKBz0KA2YOAqELi8es7zCKBMKjr49bllFjTb0OKp732GoCpgknAVQROq/AZhR9Ghdwemvz+dkGUQNBWmkLTZXvSDR3KZGDSqJ+V15Yx/alJq4fYJrQYTgGNQMHjPMgTlhkHEO8dvmRz5zz3Z47g8V3+ibXEiTTFqulfyzTvTaJjRgF8YJJTURCgMAsJAAqYQbDyqJwD1i1uIhfMdABJRJ6WmS8QdlszRKiXXdbNkRlMI7Eja3iGEwFIGEu0EkHSl1gu9xMHNMqQLoaUbBfz5sn058fw9ycUxp+yuCSUHSVuJY7uceco/U1mflVI8+egbbLHlUMYcrRNNDh7Sn6++zh/js8++5tDD9mbFipUcccSpqUScy5fXc8bpFzJ8+MZsvvmIvPOKWHNs6ISigHeEEAq4Ryl1L9BZKbUEQCm1RAiRzGHRDfg849yFXlsWhBCnAqcC9OzZc13O/Q+DxkVN3HrQg7Qsb8MQCp8BmTkRLSGI59BKiSXIU34DlX6L5owsGwIKqqMMIfAbigofWBljJ7sGTHCl3p0GPXt1xIWeZekkh2UWtNlpicEQgjKfoNKnF7+qMh8VIYv6DNVYdcCkOgC2VMRdRdRVVPgEJRYYhqBx4lzqXcWArqWU9+vEtx/MxjKS5+NF0mffujbka9IxhSDmKkoMQbdulfgC6Z+oUorn/vku4x6eiB1z6DqwI5XdqzCEoNS0yAuhVHppX/z9Mm4++RmMYOG4mMzxV4fcj0IIgcg5L1t1me3FJRBYlsE+R25VcPwhm/TFMEQe6SaljJIyf8Gqk++/MyFFKD175WdABpg/X3v/vfnme1lZnUHf+4svvFEklF+JDZ1QtlVKLfZI410hxE+r6NuezTS7QZPSvaAj5dfONP9YmPHhDMb+cyzNS5oRUqGkwlRQ7QeJoM3OfqymoQ3gcTetiOi3aTdWTi0Qo2EqyhW0OskUJ+0jaCrKfYWNwUqBaUDITBOSo8DOGNAyoC6kpZSYC34DgiZpY7kBbku0wOjJuQpqOgQhkplkUWEIiDdGcSYt4JLnj+e7D2by4cNfYsccLAEhn8AyIOkHEDRBKoGr9Gu/KfAZgnH3fMaIPQam7D+fPjOJt+9OC9yLp61g5aLmVTwh/aNoXtrKl69NxTUFJgI352dheP3W5McgCvRsvy35TNKTEYbgwhuOoFPXqryxE3Gbb774icHDN2LKpHThrmSanY6dqzn06F149rnX887t1Dmd8HGHHUdx9T9vy+uzw47aGF9Wlp+9AKC0QGG0In4eNmijvFJqsffvcuAltAprmRCiC4D373Kv+0Ig03raHWi/+k8RPxtKKp466TGeP+0p4osaCEqXEiGp8EG5D0p9UOlT9ClTDKmUDKqQ9C+XVPsVftOgR5WPA/8ymvumXcyYi3cqdAW6BBV9y126h1yChiJkSC9tRvaC5ReKoZWaBFY554zTCvW1DKj2a0mnxMr20LJsG6OpNe8cvcBBsDxAyJ+/60+O4NqSD657l8Mv341LX/8TW+w1kLKgwG9qaShk6T9dXVGfYwpBiSnwG7D4u0XcsPe9tNbrHfn4x/LTBMXbElhWO0kdVY5KzpYETYOQ52VmCUHA0DExPsPAb2jztyVEuwtDIcN6QZVmys0imTpF/3v5Hcew/3Hb5PUPt8UYs9flnHfqrSkyqelUSVVNKRXVQbbefihPvXY1o3fYlF12y5ZuSstCnHjKQan3W47ahPMu+FPKpmKaJn+98DQ29+JR9tlnV3r0zJZiyspKOfroQ9u56yLWFBtsLi8hRClgKKVavdfvAv8AdgFWZhjlOyilLhRCDAGeRJNOV+B9oP+qjPLFXF6rR3hZC/7yIL4SPz+8MYXXzn+RkKnwG6AFkWwJIWRIAmb6O2UKrVtfGhVUbtKbk586AYCP//YiHz83mSZbIEnaPiQ9SlzKLRdDeHF8AlwFy2MmS2IWcSkImYpav8LyVGvNdm50vJZOQFFqpu27toSEhLjMnrNfSMosyYq45R1TBA2o8mtpozkhaHVAIfAJRYVPEgj52PGvu/HmTR8RbcpXn7gqKeUoug+ro35OA4lwgribrQ5Mn6PnZ6TmnkZZTQndBtfx3YT5xCL5WXePv2F/nrthPMvnNabaqruU07SkNYsgQeFm/Nx1AGQ+XKWQSuEUiGZ0CtCHmxHzkr6SwvbcfhUKwxJYIYMOncvZ9aDN6Ni1klnTFtFvUDf2OXxrnnt8HFdf8lDe2AkVTpFYp87VvDD2Bjp1rub+u1/go/Ff0b1HHaedcRgDN94o79z58xfzw5Tp9OjZhXlzF9KjRxeGj9CljOfMmc9VV13PF59/w8Yb9+eyy88tqrt+Bv7nkkMKIfqgpRLQqrknlVLXCCFqgGeBnsB84DClVIN3zt+AkwAH+D+l1FurukaRUNrHyunL+OBvL7Fy2jLMgEXv0X1pXNLM3ElLKfPpBdBBUGK6VPm08bbNMQi7BmVWps5cUWpIotJgxEFD6L77cCpHDWLsyQ+x7Ov5rEwIwq5e9HqXOHQOFnbhVEqrvwKGZEXcT1ymV93SUmh2LFraJFIqupckKLUkfqF9i+LSZGXCR9QVSCVISL14KyBgKDr6JSU+PefmhCDsmp6aTOETmhST9+tLXVYhFVRuN5gf3pmeNVe/kEiliavUIkWYCmhzNKnlmo9MoSUo6c0r7q3lmXwQthWJnDVeANd99X906FpJPGrz3oNf8s3rP1A/t4HWphgJN5M0sgnFVYWdfF1PIkwSTrJXIOgjGrM1eXhsr6U1hRU0icfSn52LTP2XlUAyNZN0ipbBm/Si56AaXn7mo7y52CqWldzz0CN35YY71jyZ4zNPv8KZZ/wtlSRyn3124bEnbsXvz887VsSa43+OUH4LFAklGzNf/IopD3yEm3BoCbsk2rI9gsothxbbQGGgFARNSZegnbXoNSZMHCWo9NuUWdqp1FAGy2N+upbECJiKmt02I9yxK5PuGp86TyqoDLiUGu17eQcMh4AlaYj7iLkGdaE4QVP3T7gmcWlgiGwjvlJawnEltDom9QkfManVVAaKUlNhCUXQI8GEC422Pl7hk1lqsoSEhMwUHxSlPWqwBvRg8mtTMJBU+rQrMWjSSi6jLqCUIO5qUpEqwxkAbesxDKEXcSVwpHaLTiXw9fprZwD92gQ6bdSBf3yikyA2LGrmih1uJxF1UnKjQhG20ylbXJntUJs5jyTcjDVBE4s+3bAM7vziHD56eTKv3D+B+sXafhMs8XPFo0fT3Bhl6fwGZv+0hHee01Uk0+n289cZBwfpEc1eh43kpec/yOuTKaEADBjUi3c/uyuvXyE0NbUwoN/oPAP8DTdeyWl/PmaNxiiiMNojlA3dKF/Eb4SPznuSheOnAXoBTrj5WnRHCfqXR5neWooQUO13ssgEoDaQoGMoht9TeylvUQxZTspQu/Ldbxh4yyiWfdOHJV/MwmdIREmITjWK8LJ8D54kApZLdShO0LIxEFmxG6ZwiCf8eR5hQkDQcPH7JNWBBN1LYrTYFktiQaSXFdgy0guWZUCJ6aKUyLO5+A1wpEJ69yGA8MIGNjl4CxaN/R5TpCWKpFNA0sJiKUgohc8QhEwtgSTJwmeQUk0lp28K7UQg0F5hrucdFjQFQdPLr6sgvLCRh059hmNvO4RnLn8LN+ZkPReBoMwHbV78jKFTa3nX8xwflCLhiUzVdeXUL2nJfIKp8SpqSlg2v4mFs+pTZAIQiyT4z5+f5bkZl6ccGvY6aiRvPf0lH783mcaGfDtUcm5J1NZWpwp/JeGoRJ7NZtCQ3gXHKoSJX07KIxOADz+csFpCaWtr46mnXmD27LnstNP27L57IZtfEbkoEkoRNE5ZQMsnU6j2K2KuSdgp7F5aYjp0LY0Sdg0WRUJYBQor1QQTKTKBtKtsud8h6pipRbz57sfZakQ3YrINmppQPj8tTTCHqpRbbzYUHcsilAQcjKifuJ0dEW8a4DOSyqVsGELhM12S1Wb9loshFPXxAFIJzIwVuMxyqPVLhABbClodC0cZGWMl1VV6ln4T2ubXI0S2eip31y+EJivH9Ty5BMRV2qMss58PhSugRKTtLUbO+II0WU95Zxov//Ntprw7XZNUBjklPwm/oXCkvkDu3Awh6NKziu2O2IT9ztiGL978kdvOehHXzZZUVi5r5bKDHkQa+Z97c32YR655hxMu0wkqKjqU8P4r3xCLt19pMZMstth2IOf+/TDeff0LLj//HhoaG/NqqFiWydnnH9HueLno1bt7wfbevVed+aChoZHtttuTadNmAnD99bdy5pl/4pZb/gPo+JympiY6dOhQMEvBHxkbtJdXEWsXS9+YyEc7X8a4bS5m3OhL+OGKp2ibsZBZ595O15IYtYEEnYJxagIJzByy6F/VwpZd6ykviTOiUyNb1DZgFCCUgNm+yso0JFVVUWpqwvgb5pH46FNEiw6iE3acqvIYfTs1UhaIYxkOPtPFMlx61jYzovcyKivirCq5h9nOkYCVJhNQmIaiY0mcjatb6FPRhoWkyh+nLhghZMrUwu0zFJU+m0x1TSaZCKHwKZclr07E8huYQhE0JD6hCqQsTP/Y0lJJ/lyTRnm/oQmoUJ9UX5GWaKa8m7bjCNKuwJ5vAnhJKY12FsC+I7pwxEU7U1IeZPh2fdhm/6H4Q1ZqwExXY+EW9tF/6/GJqdfP3D2eRNxJpWrJPSNpXwHY86At2WHPEViWyV4HbsMzb11Drz51nl+YiYGFwMB1JAvnr2j/geRgwIA+HDFm/6y2mtpq/nz6sas87557HkqRSRJ33vkAM2bM4qmnnmOjjYbQqdNGDBq0KWPHvrvG8/kjoCih/AGw+NF3WfTw27iOAtcEDJCKZe99R/Mn31Eu4mCCD/BJSXkgTr+aRmY3l7MiGqBraZQB1Wm1hWkoulRGqC6NM6u+OiVR+IykrrzwotWtWwtl5V7aDgV22MBNGGBIzKDEjZlUhBJUhJK2G4U/5GJkCEyWT5CwLRI5EopUUFYax3FNwjE/oA3GflPmkWPmmlric+lT3UY8YRBz8n8OpgCfUNhKYCApMSHimgih8AulJROgXNqIpPEERUDp+JZMaUviRf17acSkC7GM2BwDctRV4LcgXqi8vNch2T/eFEmNkfv0c8csBNcL0Fk2r5GLd7+blpW6zLIpwM7x9hJCYCojlXYlidbGKPGoTSDko2F5Wm2WdCFWKNzUOYLRuw7jtAv3Z8TIvlnj9B3QnSHD+zJ39tKUWkxgopAsmJcfu7Qq3Hvfdey449aMG/cpPXt245RTj6Z79y6rPOeHH/LD3ZRSvPXWu5x//sVIqe9h5szZHHLI0cyaNZm6us4/a16/VxQJ5XeK1s++Y9ldLxBZ2oJt67JThgE+wyHimNieYToWU5QF04usaSgqQtojZlB1K4OqWzFErpIEUIIOVVGq6+KYZZCIGTTPCYApsO30spayoZQkUmQCnmqnROLaUFYnSbQZuDkFBk2fyiITAMuvCIVsXGkQTfhQSuDzOVRXhwkFtXQUtw1aWwJYQhKO+ZHuqiPEDVVYVZaEz5CEhINUOp2LpkyRZS/J3fibQgdeRj0PNtNv0Xe3jZn5xuTUoh5MVgsO+lGQldYk+YwqKkPYCUk8kkAqhe0WDvZ0bZeQX2TFWKbHSWfWEkJgFHAXblupCemNeyfQ4r1OqnNkgQsaQuTZ2V1Xsk/Xy9hy94GM2KovX4xLL8xpYvCeByZ7H7x1iky+/ORHnrj3HVqaw+y6zxbMn7ssL1koCIYO71fg7tuHaZoce9yhHHvcmseYjBy5GU8/nV34yzRN5s6dmyKTJGKxGC+//Dp//vPJP2tev1cUCeV3BBmOEPviGxofeRm3sQVTGjh2KK9fwHRThFJwHJVt8G5P5jBLoaS7Xk1LHYfyWolVKoi3GjQvtnBsA59fIhIOJaH8lU4YYJUqDB/4qySJsAFejIivEnwBhYzkX1cYilAgQYeOYQxTIYRKnQcQ8LkEOkW04wAxYlGL+uVlKNXe/lzDZ0gcN22sBk0WlT4bqQRhx9CxKAZZ7rf6Zf7YAQOUkoiQnzGv/4VvH/osx5EAggLwCaKtiZQTgOt5pkkF4aYoyWXYFALDVMTcdoIJvSj9XMdNlaMmtAwDV2oLRXI6/pCW+JZ6ebwykWmLyURNbRkr69tS13C85ItfvP0T0dY4ex4xkref/SrlWuzm0FjjCi31fvbBZE477PpUupUvP/6RitpggXkI/nvNo4w5bg8OOHTHVPvk76bx3+seZOb0eYzcchgX/e0UunTtlHf+muJPfzqWZ599iS++SHuAXnHFhfh87dgWS/J/Y39UFAnld4Dwjf/BmTodJ+4Sa7UAgc+vCFgujeFgqqpfEgZgmhLDVBhu/u7aZ7lIaaQWYKX0/jazn+mTlPa3EAFLpxyPOfg8T6JghSRYoQlEeIEb8ZUmqilbd6MUmKV+II4ZEpR2l0SXGhg+QelGCrdNoBywOuivqdPgIBMKqxIChk2wWiH8ApkQ2A0KM6Aw/GC3krWND4YcOnVpoXFlCbZt5j0P6b03DUUQl4R37z7DRSojFVzpei62mnJUSp0lhECYBsrJ38oHTRh5we6Udixn5jtT844LdJndTKKxhK6D4qa8fVWKtHSqfVUoDZomkgLtHbpV0GVoN75648f0RcmOhu+3eTeWzl5Jp55VeeeHgj4isRzpCWhbGcX0nkSuVu77z+bw9P1/48+X7cfn46Zy9TlP5I275U6DAHj4jrfycne1NeRnbFYoPv9kMp9/MpmF85dz5nmH89brH3PayVcQi2nxdsb0eUz4bBKffPkUPt8vW95KS0v56KM3eeut95g7dx477bQ9Q4YMYtGixfzrX/+ltTWt/q2r68zBB++/itH+WCjGofyPxqHYM6bgPnwTbksCp1UvcYmYQSJmUtbJIVDh6oXQESycXU5rc0CfKBRdBkSo7RrHMCEeESz4rgw7ZgKKis42wUQcKcF2rBSpBDtIlK2QCTBNl6rNLHxV6R+sbIlnb9uT8FKLUBbAaXZxFkWSIfb4+pVhdQzgzG4B10F4q6oyBUaZD4IWQqp0u6uQtsQIaBuQijkk0xorpZARiUqA3ZI/DQDD1Lv++iWlxGN67pbpYrsGMdvnEYvej/tNHdKXcLWp33Wh2Q7gKAOpBCU9a+h1xLYYfouaYT1Y+tVcvrr2jdS1kqo+M+hj2Ok7M3fsZBZMXYbMeUauhEQBycl2yfJxSgZ2gsB2FQmVfR3lxRq6Ga9T1goBV395Ls0r2lg+t4Fugzrx1l0T+Pqtn7ATDpG29OItlcLOiFUxTIOhozfiq3HTM6wfmkSVV2o4+fztnHj5ngM7cfJlezJ6v6HcefWrPHrzO7iuxLQMTr1oH066QCdzPGSHy/hp8ry8ZzBqlwF8OO5rL+2O/lyS5YDLK0oYPnIjxr33Rer6UjmeZ5hizNH78u//nEdZeeG8Xb8UX331DVdeeQ2TJ/9Az57d2Xbbrdl//73Zdtut1+p1NnQUAxsL4H+NUKQdQ712F+70H6FZ78gSzeBG9IIUj5oYAUV55+z9ousKpk2qQUpBlz5hOvXJ3v1JF9pWWgSrFYEyiCyC6BKQjlZrVQyx8HcOpPorwwDbgYwduWxNZL1PwSMU0aEEUVOmF7+WBMRtjFK9qCupcGY3ITyViagJYlaHUOFEwYzEqXkkoxYFYEucZXGUo7CbC/dP2mOUgqZl2ZKb4wpsxyBu6+JWwjPkZ9pJHNfADZXQ/87zCNVVMfUfz7Lio6lIr96L2aWGxmZJZKWOpRGGoKRbB1rm6wJXroKEFGQmhXck2HkSE3kR8cl2qbwo/1SgY+4zSRORfjyaHLY/cRT9t+zJiD0GYnmqmzfu/JQn//5OgesoL94mG46UWW1SZSvTXCXzUrMIAbseuRkfvz6ZSDjG4K16cf4Nh/HdZzN57fHPKasMUdU9xAtPjM86r+/Abrw64VpuvOYp7roxbc9QSuESx1U2DtnfY6UUjkob4oaPGMjY9x/8xZJKe0gkEuy778G8/356zqeeehJ33XXrWr3OhowioRTA/wqhqCU/ocbdB176CBV2cJdHwVXEG0DG9YLkOgKrTBEoz1+Nmhb5cKImNRvbWAWSqipPl5NMcZ5UnwgTTQZJnYwhwDJR4XjWYq9sF9WWE3MgQHiJC0XvGoQ/44dtO5BIE58K+HHnNUJzFHOjKk+nUsDCnIuMOci4i7skhhMFmaMxMQKQqZeJRUyizdnpNxzPgG4YCsfV5JJpG5EKXNek57//zOQrnyHRku1FoJQO/mxzBFg+yjfqxLKflmaNoRREXIGr0o7FCZntDeaqfKJIXt+R6fNcqVIpWnL7JZvjbrZ6rP+onvzfU8fiC1hcuvNdzPsh32tKeYGOuSuDUgpbeS7TgERlR9ULSMhst/Fkwa1MlHcKUr+iJatPhLCWLgR06VbDDQ+exUYDurDtkFOIRbO/B1I5xFVrXpwKgCPjZFLhA4/8i/0P3CX/If0KPP740xx//J/y2idO/ITNNttkrV5rQ0V7hFKMQ9lAoZREffUw6s3L4ZMHEfGEt3PW6iCzi2YFM8N2aVrtKNGB0l4WVdtWpKSCrGulBkgvakJoozmmkSaT5AHQ/qwZED4TUe6HEh8ELB3xZwrwW4guldlkAmBlGziFZWANrsMc3hVheGqyVQVhFIARMMESmEEwQyAsEH7w1Qj8NQZWtUh94wMlLv5QMsZEIYSLz3LwWS6moQj4XEoC2QuZ8G5/6l8fIFEgrb12IVaEDIVyHFbmkEmyj99Im/EFOsjRSM7DEPTbeUD2M/cgVfanu7pSpK7Mt7XM+GI+E1+ZAkCoLFDgrOQTKQwzWSxMCExhYIn0ErLFLgOy+iaVVFn3gMwiE63EcgkQJECIgArSsDDCSftcz/kn3pFHJumRC383cq+3YP6Sdu7kl+Orr75up/2btX6t/zUUjfIbGOSScdA8C6SLiDRqJXss/0dlhCzckIV/VE/UvAhGRRCjpgS5uBk1dzEio+yuKPNRstNGWu+9IoCcmp3VX3QsR5QHob4FQn5oynCtMoxsLy+pdD6RuipY2gRJY60QiLpKjKAfSoNaHSUVGALRVqCmSObvXgjoWK37lpfAnJg+N2BBNC31KKWyI5MLSNfCJ8BRmlTK9PvkOYZfYFWC06iQtiAQlASCCaQLTkIb35UC1zFwnGReMBdXJdOv64m7sn3PJ/CEOCDRzqKX5wwrNKnoR6bY+uRtqenfmU/u+jjVp7S2jIZlSY8qL1NwOxNQJNU/hY8vmLIUDoc9T92Kn77Itl0k7S+F7q+gZ5kQGEo/40nvzyRkWEhPNSYFCOUiCsahJMdMj2p4bO/i4tgOn4//sSBvKCQGvpS9JAmpnLxZbrfDyMIP4Vdg+PBhBdtHjCjc/kdCUeW1Aai8XLsJIhOgqR4Ri6V/zHEHVVUJ4Rjip9mIxmzjgKzpgOjfBSwry1VLNbTivvoNSoLwC8zRffVCnzxveTNqUSM4ElFThujVAWEYKGEg/D5ULIFa0ggGiE4ViJYoZOZjKg1CTbm+VjShSS/k15X7DCNP+iCeyFJv6RMVROIQ8EGXmvQ9aGOFDhe3HdSyBmgKp5NQZRBM3ooqwK33yKtQsIa39bYbJXZr+nnZsexwQMcxiEV9qTYpIe4aCAxsx6DF9pFwTWSOgC8VuFLgoB9Jm+MrKEXYLkRlunJk0oMseRO9t+3DwP1HYAUsmhY2UdWjmr6j+3LDzrfTtqItRRbKu2bmT1gmvdGUSmVUzsWJNx/I1oeNAGDCi5N59t/v0bwiTEVNCRJYMb8JV6m8uctVZCfORMpYryCOk/VsXVziIpH1PtMyk5mFGMAhgS3SqkVhQNyNZFzHoaw8xOX/PIUH7nuOyZN1PjrDMDj/wpO48JJTCsz41yEWi7HrrvswYcIXqbajjjqCxx57YK1fa0NF0YZSAOubUFz5ESTLtbguNDZiNDUBIKuqoLIyszPi/c8Q9Y2At5McPRwRDEIiX4JRUsDU2dDQgjGga87BdpQahgG+jAh0x9Hz8q5PS0T3qSrVNpBcCAMqy6ClDfw+CHn6uHAUWsJ6ZRZAwA9V5V64uNQElFwZ84IoFDS2QnMbJAtt+X0QjunUwHEXr5IU+ExUQ1THPYSd7FvMuOVEg8QJCwwDXFvgJIysy0XCgbyYFVdCwjFJuCaONGjzvMKSiSKVZ99w0UShFLQ6VpaRHDQvVm3am4aFLTQtasL1bCK5n4bnvEa/Hftz+F1jEIZg6nvTeOqs50gksr2pkrF2mQSi7R35nFvRsZRrvzwXK6cw2Ms3jue5/3yQ8Ry0YV6R5mZVICCyUBto8gGwlZtXITJOXEsqQqvA3Iwn5OJ62YkzxsJl6103ZtpPc1i0cHmWMd4wDG66+3xuuPYhZs9cgFQSiYNpmDz/6i1ss92mBWb365FIJHjxxVf48cef2Hrrrdhjj13/UHm9itmGNxC4cg6QoWZIfgktCzp2RLkutLVBRUX2iaaBGtIP8eFEVMiPGtwf+vZFNTQgChAKlWWwzQhEOIxavCxV91sJoKwU0dq2+skahl68/X49z+pKiEQ16chommySfasq9L+V5WkyAf3a79OSCmhCCQTSBCMV7VoEhIAOFRD0QyyOsB3dFrf1v5k1123PS0wICJmoWIbvrUr+o5CuNrxrr4EcHb8UBQMg0yYNgWUoSi2bhDRxpUCWl9PaGEsdT19OeWlKMh5Fx3JGnbs7E+78iKZFze2WOE5mG54xfgZP/ekJjnrwGBZMWoSdyC1lpR+5zJFGhBD40O7Fme1tjRGmTZjLxtv14ZNnJvHt2J8oqQzy8fPfZ40phC5TbIV8jNx3MF+9M422xigiQ0rJdE1uDwb5JYd9+DBwcZULwtCfSYGRkkpGA5P+/Xsy/v2vMISJTwVxPQ+LHXceyQ3/fpi5s5dgCMtL3mkilcOjD72yzgjF7/czZsxh62Ts/2UUCeU3gCvDwMTV9gNQVVW6YlSh3U5FGWrIRqgunaFrF72ShELQnK0KU6BJwDD0jr6qAhqb9c+zYy2UBFGOg4jGss4RZo6qyjT0wp+EYUBpie5tWymPL8rL9b9K6a18ITfN0pAmIMvS41omqORKWGBZSqq/kggFtNosuXstDWopJWkrMg0vyFKgWuOIqEQETZSt42eSkC0KKwBOQksRwtCR90qmPbwokBtAobMWS6kDIRUitVJHG6Mp/X/y7Kij1WjJW7A94Su+tJVXj74P1+8vZAJKP+oMI8bsT2dz3TY30rwy0r7dxhKkglNSjzCZcCUN6ShevOY9em7ajfGPpo3LyXQsuaS0/xnbsvufRtGyIszk8TNBGGy6a39++nYhDUvbCttaMm6sUCVH0KlX9HEHExMDTSwmJjH09zIzXctP3y/ImJeBhVbhTp40jWUrlmU/C2EilUs4XMB2lzHHa/99O3ff9SitrW0ceNBe3HDjlVRXV7Z7ThGrR1HltQ5VXnnSyJogtQIVSPttGHpBzu1fXw8tLekfd00NmF7t27IyiMVg4UItyVR6ko+U0NwCkZhevUxT2y7wDLJKIXxWtgoM0lt119USR1nZmnljKQVR71o+n76elB6hKK1CS742jOzzkn/ROLTl5GJJOBBJ6HMNgXJcHb/SlvDiVLwbMsBtdFFxhZLJ4Me0usp1BNIVOLZBNOLPUl8k/QtijkVbIuCV6xUkpEmSNFwlUmV/o67wKkqKVDGsPIJS0OZkx6Rk3ZbMjhNNuGlSKiQVDNtvKBNfmpJzjcJuxa4sPIZCeeWL0yQxfOd++II+vng9O8pfegVfbNtFolIqLj2OhuUz2fvUUXw/YQ5L5jXQvDKcxT4KRQL9PU+5Fwvo1KeCeXOySUIIgVFh09KSLVnbKoKr8n8rrrK55e5LOGzMHgXuFO64/SEu/OvVWW1777MLzz1/b8H+RWSjqPL6jeBKB/iU9n2AVoNMFZjjZLfnShDJ9o4doaoKlUhoiUIIiETSZBAMQnU1yrYRrjemYUB1FVSTXtCVAilRjpuWHPI36xqmCSWhNXftFUL3z5Q8kvcqhJZqXDctceQ+Dx26n30saddJPmoXHSkY1eF1KprjUeTt4N0oWTclBFg+RVwKIhE/UhraEpK8tEcm4YQPhebdhDJTYwgBltCqNFcJEjJtzC9c2yU93UJIxmtmvk8GNRb6VkkF34/Nz5CbmRQyNVa7s0mfY5L2EOvUs5q3H56YOpaEAUhX4TMMXCWxyZZMBDph5Zhzd+K0q/dlwcwVnLDF9dnXQmBherXnhf7PEKmcYJlQSjF46EZ8/tnknDFMIJ9QTjzloIJk8tqr7/Gff9/J95N/yDv21pvjWLmykZqa6sIPp4jVokgoawmu/SUYGdUGf62BLmkgT/5IDWPV/X2+dP94XBNLklCU0iow1y28iskMqcAwPNuIIlUdK/dekp5YQuRLTMnjWdWmPDLwWdmk6Lh6rkp56rN2vo5KwbKVmRZnsCUqEss1gSAsA+UzUOH8RUYEBTKikAWEP73hliiVdB9Oqqv0vxHbh/LUWqYBplQpiSQJQyhsmY6RyFUhZcL2xs6ag/e/HM1VyqMrX7WUJhsnmu8k4Xltp8gp9fjamVMSyYzEQggWTV+O5TNxEm5en+R30/BkY0OIrP1HoMRHsESrppbOy0866Y2UTXiuYvm85rwIudrOlQwe3juPUEz8DN2sN5O+SRPqmeeO4eIr8rP/fj7hG449+v+QUuLK/B9CMs6riF+OIqH8Csjl36Fav4EOHbQRfW1/GXPLAK7pOcGcTK1JUjJNkBa4GYtPUs2UhOsiMo3t7V0/uZq5Uq+wme3xhCaP5E4zOX7C9nK2C63Si2VEmkvvPMPIvqZS2kssc9UJx3WREE/tUhCFzDImxBsFSmr1FoBpyVRKFjcjzX16vdQLa7pSoydQGRI3r0yy8GJDlbffVt7/8xEvQOwCz2lCpbMNa6nES5WfaZvIMMILoUODCqVSM4TAV+oj5mUdMLwLZaq19Izz7S8A0yfMp0OPSpbPayo8X6+vXxjeV0LbYiSKfU7eioCXxXjjLXoSLPERy0nR7xbY4ZhYuMpN1Zuv6VRBl17VPHbXu/p5irR9paZjBfc//nd+nDqLubMWMWrbEWw8ZKP8BwE89uiLqfTzQpgolU3C++67Kx06VBU8t4g1w+8qUl4IsacQYpoQYqYQ4uJ1dR316qXIcVfDvHEYzS16Ad+QdzaGoSUUnw9lmmD5wNQurXl+pYVsN6tCPJ62fyTtKlJqe0k0mq+misVh6QpoLlBnXAh9nlLatrNgGcxeBMsbdfAGaAJLVpwqsIIqpVLeXrmQcXBtAztm4tr6LxG1cB2BkhAJZ6diSX6krhQ4MvvzzaUJqTRJxGW6ioeuuJg/x7hLStrJmjtJ92PvCiodq5J9j/lSRmb1xlxsstsADEtnF04mIEj3z01un72/AKjpXF5gDjmqtBTZCkwh2PO4LTjhqrTKqawyxHm3HJoiGEi6DBf27vIpP93qOvHo25dw7Nm78c2X0zySNhAZd9pQ38K1VzzKjruM5IRTD2yXTADsjO+iwEAIvYGwLJPy8lJWNjTywQeftnt+EavH74ZQhP523AHsBQwGjhRCDF6b11CzvkS9eLGOCWluQyxahmqL5huu1yfac7IQIiURKL8fFQgUtsmsiZNGIRKKxdNkAvp9JAYtrZokkqhvhNZwvq0EtCvw8iaYvwyW1GuJJnVN6VmTM85TCuXIlO5eSYVqTehdrCFwwgoZ944pRaJJ4Dq5NQ0Fdsykob4Ux85+Hkl1UlvCR+5y7UqB6xnO466g2baISgtbGZ7tAqyAxU7XH8bgY7dCCYEjIezofF/CM4BnwRDEhJfnS+WThCnScyqELMFO3xpCwKzP51HZoTTLZGUKTXim0FaI5KmCfLNYVacyjrhkZwIlGWTgzcEftPICG/WtGHnqo02370v/Yd10KWIEhlfLMRdJ4/yQzXozfGRfJn76Y6q9UGjlR+9NSr1ua40w9o1PmfDpd3nP94gx+6Vea6nTwu/34zgura1hPv1kIgfufxJff5XtQl3EmmO1Ki8hxHmrOq6UunHtTedXYUtgplJqNoAQ4mngACC/CMUvxXcv5P1qRUMzqrV13ai8fi6U0ou7z5c/FzuBiGo3SiUEuC5GOIIK+NP9lfK8vQoELSYXcqmyF/V255ExRjSWtuEkXTlbw6iAXqBE0n7T6tmg3MIShnYWyFhkTaFJxk2ripSUOPUuiaUZc7QU0oECmTm86QraYlo6sUyJ3/RS/0tBzDExvBrxqch2CSsTfmxlZoyRP27CH2D+t4uY9PhET8WWnLki4WUZTkoNUmn7gRRGymNKoD+OofsNo2V5Gx371NCwtLWgAR5vHIN8433z8jbcHBYSIltllbmzTHp6JY+PPnwEw3boR+PiZr54ZSqBEouhO/ajQ7dK3n3kK6LRfKm2plt2HNXkT2dzycEPkEikY4VMBCVGkKib8OJQlCd9GJRWBzjt0n0B6Nqj1nsehX9fHTtrI/onH37Dqcf+ndZW7Qk4bER/nnjxWqo76Lnssuu23Hzrlfzn33exZMlyhg0fwKRJ2TYZx3F48MGn2HyL4QWvVcSqsVq3YSHEld7LgcBI4FXv/X7AR0qp/LSb6wFCiEOBPZPzEUIcC4xSSp3V3jk/121YvniR/vEd/J9fPd8iiiiiiPWNXxo28ovdhpVSf/cGeAfYTCnV6r2/CnjuF81m3aDQ9iXvaQkhTgVOBejZs+e6nlMRRRRRxB8GP8fLqyeQmeMjAfReq7P5dVgI9Mh43x1YnNtJKXUvcC9oCeXnXEBFHQhZOPedzaCLH2HmyuzSgF07VzFgozqOPnhbTjzjEAiFaGuLsnx5I1271hIM+tsZuR0kVT+FbB2rwuLFiPaM61LC8vpUbixhmbrCVp5dxNU2jVBAq8UcF4I+nURSKZi3AhX2Ipo9pbwCKAlg1FVluzk7DmppM6JS195WDeH8ZJGAbIgiOpUjKoM6xYfUzrIqbCMXt6Ei2eckGlWeZk/JdH2YTMQiFgmvSqPtQNTWdpGEbZJrSmy1LWJu9k/DVRB1TU/DZqRqtztKx56YQhFzdW6vhEsqx1fqcUpS8Smpuar8qoxxzyCfROdBnUgkXHxBH1sdswWhqhIePu2ZvHFA755y/RSSgZDtxbAkOwUqQ1R2LqemRxWJmMPUj2ZnFBfTcSmCHG9wpXC8IMlCO13tYCBJKB22qP+yoYtyuXm2EZ/fZOcjR/DCYx/ljevg4GIjBGy721C+mfgT5ZWlLFy0kGginNV3r/1Gc++jV6beP/bIS9x3z9O0toY54MBd2WOv0fzlL5cz9YfpVFSU8X/nnsJFF7er1ChiNfg5hPIY8KUQ4iX0d/Mg4NF1MqtfholAfyHERsAiYAxw1Fq9Qu1A5IyvUfUxbtxjJN8tbeD7ZY38uKKJGU1tLF7WxOJlTey0zWBoWAlV1Xw4/jsOOOBvAFRVlVFX14HajlXU1lRSU1PB9f89nYoKXab0g3Hf0NLcRmlZiBInQYkJJUE//m5dqexQQXW19raRUuI4LpZlYuTGpyQSiHg8LwWK9n5ycGfNR8YTuJ4dpDToT60Ui+pbcFyJKxWO62K3RHGbo9gJl+5VpXSsKoXuNSyetZQfpi7AcXRfnS1XYcdspIJD998MU2rf1NcnTGNZOIJyJTLkQ/lNVCSBtF0Gdq5it8F6D7CyLcajn/6gVyfPDUlFXXAVwhQcPKgX3cp1DZhPFyxnyopGVFTr1U1D1+gwhaDS72PPLj1IVmN8a8ECpAt2LIDpmYLDsRAGBt2ClXQKlGIIaHXiLItFkNJHQloYwsQSBqYwMIWgxAx4C6iBCxgK4ioz0l2kKi0WWrgLFbNMhvmANuC3RbIz8wIs+nFZyqvr+YtfI1QZpPvQOhZMzi+MVejCQkAgYFHRpYKlswvHghhCYLgul711CqbP4LpDH8RRce26q1xMYWIS0t7eMkGjswxXubjKwfH+taWDi0tXf3+Chv4+L7GnsyKxEFu5SOV4mYVdpJIEjDL6hrb34uMVU8IvoZSbYXhXfHnH4ygknYLDqfJr761mez5LY9+C5/E194VxekOzQCCESXVoENJLz+OIldR2kzzxxBNUVVUx8csp3HzDIwhhITC57ZZHWbq0nolfvcWSJcuprq4kGCxcI6aINcMaE4pS6hohxFvAdl7TiUqpb9fNtH4+lFKOEOIs4G20W/6DSqn8cNhfAbHFvqjvdMrqPft3Y8/+3VLHZL865nepYtbiBnp374iQEhpWYi9ZSo/uHVm6rJGmpjaamtrgp/mp82686czU639efj8fff5jwWuPGbMzjz9xOQDTpi1g2NAT0vMSAsMrSiWA9/55JNv174KSivMeHc+tb3xdcAc5uGctU+48jWQ0Wt+T7iBRIEoZ4OZDt+HsHYeh5i7nvc+nceLTHxfsB7Bvn26EvCDF6976igkLVxTsd/SoASlCWbqyjQveaD/f2WZ1NSlCeeGnedzxVWHD9MCqSnbcqzc+n8RxTI7/4GPiBYLYAM7ruwNHdd8UqeDd5XP55/S32r3+y1ucT8zVwY1Xz3iYmZGFOgOV0H/J15tWDOaQLnsDsCLRwL3zn/RcXb0/4Tm+CoMDOu1Dp0AdAF8nJvHV4q9Jm57TRFVhVbJ7rfZQijbHeGL6jTjK1U4IGZ5PSimGl4+ib8nGAMyOTOPLlvEoJHKuwrAg4ThIqWWFI+vOwRQmCnhmzj3cWno5bgGHiP6hTdi28gAU0Oo28FbDg+0+p50rT0wRytLELGbGCtsoK/2dufT+u7j1wpdpXNlKkzO/gPyiYctI1uuIs7zd62+7xYH0G9KZH76fxWdfT+Tf1/6jYD+/VU1ZqDcvvfAO555/PHfceRu9e/emd+/e9O3bl379+lFaunbr0f8RsCZeXhVKqRYhRAdgrveXPNZBKdVeCOxvDqXUm8Cb62p8o6YrcuNR8HW+r7qc2UjNp430+es2iIxgu4P2GsmBB++ArKqiobGNpUsbqK9vZmV9EyvnLKQkFoXlKyASZrvB3agyFG1xh2gsQdh2icVt4gmHTuXBlHeVam7BMg0cz/VWKYWbqeuwXbD13lk5bopMdAyCwDQMTFMQ8Jmp8wWCHh0riNsOPtPEFDoa3GeaWKaguiTgjSHoUl7CTn27YBlaOjANAxMwlUdsGbvsffp3Z1BNpY6DCGgZQXhSx8henZG2xG12qXLh7K03Tt+DApXwlksFdaFgSrWzddeORAe7SKl0kkcUUkhicZNqs5zlyyu8+4Lta3rj4KBwcZQi4QpsqXCVpKO/lKRkIVSQbsEOSCVxM//Q/9rSwjIEUqmsY7lSQcyrPawUxF2bpfH6dr9Pcak1yJ037kxk0UTmRGcX7Ffj65j1fk5kZsGAQICNQgMzxo+yPJGh9c2r0CwxUkkaVYpMDAxMYWFgYggDvxFIRdz7RIBaX3ed0NHrYwrTG8ckYIRS49f5+uMXpUgEBpZ2Evb6HvGX3dj5kE147Np3aV0ZZkjJgXpOQnvD6c9Uf/6WUaq/wwLKfN3oVbYbmko9RZqXst7FpqqqgkTcZsa0eQStTlhGGVI51HQsp6m5gUikDaUchNBLn+M4TJ8xg9tuuy3vWXbr1o0hQ4Zwxx130K9fv4LPu4hsrImX1+tKqX2FEHPI/vno2F2l+qzLCa5L/JLkkEpKnFvOh+Y0jybiBst/9NOhOoZTUkZgZFfMziUgFIGNO+qAwmBQJ1I0TYjHUctWYsybh5q+GJJlTstDCB/gKlTMBgXG1ulQmlRktFKwcBk4OixMmbqqorR1bRAr4WB4wQRS6h+lUArDNHVFxpwCWHJpCyRsRFUJyjR0OpOEjWqIpGrCZ/UP59ciVK5CRSVuVIEBRk5ojrIMEm0WseU6OaO0Qblg+SSBEgcjCP5aAyNgooQP2RTRpWK09gO7Vcdm6DLHnn4+poMUAVra/ETj2eoKgcJ2BQnXRAnwV5cim7J17KDrmDQl8lUdSilcBAlpkJDp55BwFbZUxJTMIBitrpHSwjJ0oSpbOqy0G3BchZ1U7ijlLYCKTv7OBM0g/qoSSjcv45OXJ6QIVJGu1+4TfnqGeqeuPzsywwssTKcKEQhQgiqrhhKrEoCoG6bZbgLhxXwIA5RBWU0p0YY4ZWYFwivh6ygbKbXLbm78iEo+C0/1VigORimlSw6nMgxkfA6mYOjO/Vi5tIXyDiWM+evODB7Vi4UzV3DiFv/1rpGu5ljXo5q5C7NVei4uVV1KWbG0iYSK5bngSCSusLn0X8dz5aW3piLi0/eg2G3fLXj1lXdS8xVCsN0OI7ntzst49tlnmTt3LnPnzmXGjBnMnj0b27NDLlu2jE6dOgFw4okn8uOPP7LNNtuwzTbbMHr0aOrq6vIfyO8cxQJbBfBLsw23fTeNlTfeTKjCIRo2qV8Qwo4LendqTunuk6g4qCfBkV21PcN2oDWCPbcJY+FijBILFY4jmx1k1EUYAqPGn0oFAiBG9keUhbIn4LioBUv1b8pLIS8dFzmnCdUWB6kwqoIYHUOpxUFJXa8cga5H4hGFaoygljSnh25ycZokoszC38lABK3szLu2i4o6CDP7PmVMgqOwmxUyDv7a7CmHl1lE682C8Rqmz6W0Sv94Q3+7Gmo7seTvdxL+fg5SahLRAYkQCDhUVkZBilTKeaVgeUMZuauMVNCa8Ge1K8DAxW+kDfpNcR+2NPMM/HFXEJX5DhHSM8TbMtv4rg334KhsElYqmbsre4zMY7l5wbTBP02euXMTlomTkREgmQlHJV8LMHwGToH6KRtt0YPZExeQC+mRRu48JGmDfNJmplLHkxaPdH9XpUs1K6UwfQYPTr+EYE4N+6b6No4YeA0yJ8h1y90H0W1IDY/d/G6qrceAjsyerqWtBHFcke/UcfBx23HhP45mWJ+Dcd18Qtlki4EM3aQXjz/6MrFYnJ132Zpb77yS2tpqpJT4/WmnGcdxmDt3Lj/99BP77rtvqn3AgAHMmDEja+zBgwezyy67cNhhh7HddtvxR8CvzjYshNgWmKSUCgshjgE2A25WSs1fzam/O8x/ZypfftcptWstMR0GVrXlkQlAdNwSjJiDUWngLA4T/b6Bko4OrmEQ6CxxVyZSWXEVCndJDGp8GCFPHTVzCQzrnVKjKdtFLm7CsF3w6UqHzvwW3NmNGL709eXSMEiFWaf1wCqcQLlSe1DFvNTuLTFUcyxrvsoFJKgWh2jMwAq5+Ot8Xmp4CXEXu9HF8AnM8uSctLuTdBROmz4/3gp2wiLSbGnDerz9uE/XNrETLo1OVyKn3AwIhHRQmICCjLUjHvexYoWpF2ABpYGEnm+BRTdqW+SRjISo9BNG4QuYlA7pTXziHFzAR3oMqSDqGgWd0VNpRvByYikv9xa6Vkpef9ILc9JelVyBnQJkgjcPkZQI0E4AmeP5/AZ+n0EiYqfStWRxgaIgmQB51RrbQ7KccMoDK+O+UwRC9nV1XjGRZQ1xbckjV72Nbbs0LW9j+PZ92OukUVTVlrH7kZsx9vH0ps4wBIecuR2b7dCPPQ8fyZSJc+g1oI5/nfd4qo+JhVQuKietzQ/fzKO8opRd9hjFO29OyDqmUMSiCa69/kKu+uc5JBI2JSVBLrv0vzzy8HPE4zb77rczN996FR06VGFZFv369ctTdX355ZdMnDiRTz/9lE8//ZTPPvuMqVOnMnXqVKqqqlKE0tTUhG3bdOyYra78vePneHndBYwQQowALgQeQHt+7bAuJrYhY8q4uVkqkIhrsTAcokeH/PxUolNn7C9msXRZGc0tIaSqomZFlKDfoVMHlZdiHUCG3RSh0NiG+vwnHLMUpMKd0wBxB6t3CWbvKlSrjfNTI0YobxhkfRSjcwlEbNTKCEiFaomD30S22BgWCCtj9+4o3FY9H+lCdKlBqNxFRuJYFYauKdKqcNsUhqVwGiTKADMgkA64YfRC5giWTC3HSKpjhKKiJJ4/wQw0ix5EliZrXSgUhtbZG9mLhiMFUSdAcqUPxwMEDBuFwGfm9FX56jqtCdSruh2XNE9fgkSnQwGFofRCGnZNrVLKIaqkJ5eL8HbzAidTAslcbdF1TDKtHZlp6DXJtJ9dIUVC5Oe6jLfZKNPAzcntuSZIRBIIQ3hu2clrqRTXSdIkWRAe2bk5+bxSh3OegSMlbz30Zer9V+9M4/uPZnPAWaNZPL2ekqAfYYK/xM/gUb0o99zL+2zchT4bd6G5MUz94hYEBhJXJ6MkiELhYCO9J/zTlHk8/eC7dKnrjN8K4jiOly1MHz/g4J0ACAYDBIMB/v2vO7jzjrSj6ssvvYPjuDz5dL49JYmqqip22203dtttN/0sEwm+/PJL3nvvPQ455JBUvwcffJALL7yQnXbaiaOOOopDDz2U8vL8nGi/N6yxyksI8Y1SajMhxBXAIqXUA8m2dTvFdYdfovKKNUd5dLvr8tp9hmSvvgvJUq8oCJx2EtEnn2LhXK16MYSkc3kYIaBqsIvPzl9ohV9gdUqrB2RC4izMLvMrQgKzcwBlC9wlUYwS8nXfCtqWm5R2UVglGfOSCtnipcAI6ey+Mo6WPLw+dsygpSFAZU0sP3O+odIZehOgpMhadBcvL0Pl5MUqC8YxTVVw8RNVlaxcmFMD3pu/acissSO2hZtLFEInUA76HHyGvkar7UMqIy8vlVQQy8gq7CqIuQa6DrwiIfXrvDmSJBpBZmJHR+XHjkhPaknkFLJKxq1kf0d07XedgiVzDJ2eJZ10Mf/ZJGuWtPcTTkoPSilNZN6VtzxgKAunL2PxjysyksGk5y69/smEM4UgDEGikC80OUQoFYlCFTkBW7qeTUZllQkWQvDXuw5ntzF6abns5Id498WvdfX5nHrzCkWctBeYWerQFg5nHTf8iqOP35srrzkdK8N+uOmIvZg1c17WeIZhsGDxF5SX/zoPr7/+9a/cfPPNOF4ao5KSEg499FBOPfVUttlmm//5NPntqbx+TnLIViHEJcCxwBteMsYNKCvibwNfyIevrJABF96YWcf85hAJxyCSsJjbUME7133K8sZykouIlbFANv1YePFSgYwcUa7CXZEfpKikgrBNZLFDU0MAp0AwnxMXrFxYSvNsgd3g4kYlMiaRrZ7LqYTEMklsniSxVKLiGYuTUMTifpobQ1kLluMIIq0+om0WzSsCtDYHcF2BlNAS9rNgcSWRSP7zicR9uG422QIkHIM5sxQtcR/xvHTw4MjstkIqJZQOOlwZC7EsEmJZNETE8WFLkbfY2jnBhVKlDdt2O2QCYJUHGXLKDlT165wzHuTqxfwhXx6ZeNPM65vMo+V4C7lSCkemyaS9miprsg8UgOkzcGRyfE2gP3w6h229xTpTZSWEwPWCGPV1279Irt0j+z71fUipUnagVSGvTLBS3H/lm0gpqV/azLiXv9U2HAxEzuevo1GSnmpuFpkkx9p2203zyATAXyCpq2kamO2VRPgZuP7661m2bBn3338/2223HZFIhEcffZTRo0fz5z//+VePv6Hi56i8jkAHCp6klFoqhOgJXL+ac353MP0Ww4/bmq/vHJ9qU8CMVoOo6+enllosA6p9klJLYYo2Gipsyrwn7Ugjre9Xgsa5ASq6xbF8emddvzLEjM+rqOrk0q2ulQpfrODOPdpg0rSslqYF+oe9dJGkz5BmgmU6OK612cfsHysxFIgml1BJG6JZgQDDD64raFweoCUSpLEthCEUNRURykIxAiFJIi5oS/iI2BaJuEUwpL2AohFL1wMx9XWjtkVTcwlR1ySasIh7RuxKBWX+tPFDKoOFjaUIIQiYLpahF7hlkRAuOlW7wEeFP0GFd55UOvIcV0squsxufjZcVxjIVE33TCO5QdxVWIZeGhNegsZMO4mtMvuLgnaeikFd2fXu4zEDPr5+5ttU4SpIk0YyC7ACZMRGZRZQWUOkCmFlmyxQ3rHMZU6itFOA8ojDcwlPIhnd7sTy3YtbVoSJtsQ57sYDGHv7x7TWR+g2qCOjDtuEB857Nauv8CS3PK+qAremUpJN+vp6jHxSzEx/X+gpNSxrpbk+zJn73aodStDJJA18JFQiaz6pOJyMkZRSOMRRKMaN+5xtNj2Gh5+6msFD+6b6HH/iYVx84b+zrnv4EftSUlJAf/wL0KFDB04++WROPvlkZs6cyYMPPsh9993H7rvvnuqzZMkSQqEQVVVVa+Wa6xs/y8tLCNEL6K+Uek8IUQKYydxe/4v4NTXlZ74xmVlv/4AtFW+8OoOEk95/Vvgh4HlBmUKxUanDkMpoarEq9SeoDKVVXU3RgN5RuwZxJ83xQih6dWuirjaM592J60DzihBiwECWTliUN68OdW0sX1aG66R3Y4aQdK4MU1YZxzQl0aiPluYAzbEgMSd7l2YJ7bzZavtSi3PQdOhaGsEAlkSCNHoutiHToWtJFAk0xoKEnWwjeE0wSnVQq+oaYz6WRUOp4yWmQ8BwCbu+HKlA0SUUIVWeynNDjbkmbbYPpRTlfgfLs61oTy4z5f7aHpSChDKQSuIV+M0yoktP9ZVLVkpBTAp2vvEIPr/3U5ZPWZzhvg1S6DEyI+QzVT65Y+WqvACirl6Ik5mHV5UuxfPVJ1ogDMXyzpekC3S19+ve9ohNOOHGA7LaPnj0Kx656PW8voHyAJHmWJpEvftIen4l4UhPMiHtcQaexJIzZlLdBeCofOeBup7V/Onfe3HxMffnzcfBSam+XBxs4qnrxGgFFI5KpGwnSYzYdCBvjrsz9V4pxR23PcL99z9NJBzlkEP35oqrziEUyilQtxYRj8cxTRPLC/w94YQTePHFFzn33HM577zzqKysXGfXXpv41W7DQohT0EkVOyil+goh+gN3K6V2WbtT/e3wawgliYfPf5WPn/o2q80AaoJpnXiFBf3LY3QJJbAMnRWl0TbxWZKoY+IHKnyShBTIAobkXrVNVIbi2LZBNOqVoh29Dc3vfp3XV6uN8rfZ5f44AUviSIErDZoSJq7Mr/MhUIQdM89YXOWPI5WgPp79Y/MJlxLLJeaYKSkhc6wqf4yo66PVtnCUrvWhd9SKUsvJOyc5ZpnPJeEIEAJH6eeSuVBrY73SaioFfnPV32NXaSO9lCp1zYhrpBZ44cV/WCJ7kYy7EJUG/uoywg2RrDFtrwywXkSzn1chg3bKG0zp74by7CSZO3lDQOeBnVj80/KCxKQA2813701fQ3nj6xFTqfBzRK/uI7qyw9Gbsf2YTTE8Fc+UD2fx3zGP5Y9pmdjxbDfdpKos6T2ejFNxZPuKMoV2J97qgCGMf/G7rDk7GWcZpsF1L/+J+fOW86+zn8wbJ0koUkkSIpp1zMUhocIkiFGITqfMfonq6oq89vUBKSX77LMPY8eOBaC6upqLLrqIv/zlL4RCa0dKWldYG4QyCV1z5Aul1KZe22Sl1LC1OdHfEr+EUJob2rjx4hf48M3vqKgqZSNfJW3L8oPlOgTAMvSiETSgZ5UJto3P0NHaTnJRcyAmwRKKjgGHbiE3b5dcEYjSIRgnGdXdXNGVBXMi9CyN4SpoTliYQlHhcz3vs0KEksBnujjSoCXhY1ZrKd1DiSz1T1wK/IYk6uZrQk0kUZnfrhdh6c05/7pJVU1UijxjerllFzROChQhU3o74XTkvZQQkwa2V4DKFIqAqU3kliGznlvm1zrpxaWUIOIIfAZEXUHSBUEvhPrKAknAIFXyN6EMEi7YnhY/EymPr1WQR0bp9ZTUEXfTJJKL3pv3YOrEBSCT5KY8SUCTQ1LNVQg64DL93vB0TULosKPks9Y2Gj3I6MNGcPrtBzPvh6V88doPvH7zR7o2fDIGBk1gBYMdgY7dK2lY1IxS0HNoHdO/y8vHCt78a7pV0HNoZ1wpWTB9BYtnr9R2I9J2FNMU3DH+L/Qd1pWlCxs4ZMRVyJwbtr3EkDEV8+JR0iovW8WJ0+LVV8lGSWmQQ8bsQKfOtRx17P7U1dXm9Vkf+Pjjj7nsssv46COdCLNXr15cf/31HHrooRus8X5tEMoXSqlRQohvlVKbCp274Bul1P9sJZpfQihnH3w7X36QziPVw1dOpZUvItcGIeIomhJ6x1hRFaSviFEd0IuhiyDiQKuT/YWpC9j0LU/vBnVFQAOf4XrlZQWNcQtXCbqUxJgXCaXiGAKGS/+yKL68Tb+ixKcrGcZdkxktpbgY1PhtSi1JQ9xkacyH9HbqVT6XSn/2D9IKWkQi+T/SpDdTqZVvzk4GAJoowgWIzsSl1Cfz2g00OcRcg5APz4FA0eyYOaSU/u4GDIWBxG/oCly2FKmKhxHX9KQ/kQrKs4S3WMrMmIm0z1MQRQJNMWE36f5cmFAKqZa0io0UDSXVXUnHqNxAR/CKWJb4U/Xf02OplJeZnkf+uUopovmOclkQQhWca8/hXZj17WLvzvNtHgZgGvmEgiG478cL9b0lHBZOr+fKgx7Ku+4mO/fj8At34tKDHiAWTt9bWU2IxvpwlkJMIDjj2v056M/b8tB/x3L3Na97z1B7nJVXhTj89B3Z7aDNmTltIa89+ymtLWE+/ugbL/Gkg0OMQtUdHRnDVfr6HTpU8tZ7D9K334ZRwkIpxbvvvssFF1zA5MmTEUIwZcoUBg9eq0Vn1xrWhpfXh0KIS4GQEGI3dC2U19bWBP8XsGxhYxaZAKxwInleKiWW9v5piKvUItDSFOPbRqh3DToM60ZJbRmRAnrwpXGLloRnMM5IeT6tJcDctiAr4z6i0qDFMZgdTpMJQFyazA0HWRD2EfMWwbgrsF2D5niIZreUBeFQame+KOpjadRksUcmoI3ajbaZOh8AAcMuOQizNN97y1U6ZiPpRZSEjgAXeHGSBeEoAyfDEysZoBeTJhHXwl8Rov9fD6LzsTtjl5Xluwsn56wg7BpEpUGb6yPmmkhl0GKbLI/7aHVMbGl496j/khJZZlvmuDG0e3CbK7CTzgEF7h1y6TB9LOLokr9tDoRdiLn6My0s0SiiriLSEqfQJk+h4z7CjsTJSCuSjEzXdepXDbcAmbhSMevbxVneXnnkCClJIvPPcSRTJ8yjvEMJ1XUV9NusGyXl+d+RLXYfyNhHJ2aRiVKKxvqwl5AmPaZEIV3JU3eM48Hrx6b6J/u1NEUZPqoPUkp23XcLbnn0HB58+VJ23GNTQHuC6X8zcsopcFQ0RSYADQ3N3HrzhpMsXQjB7rvvzrfffss999zDBRdckEUmTqEqqhsgfo6X10XAn4DJwGnoJIz5FrPfMQr90GPKZXasiU5mGYYQRF0XJybpUlMGBYIWv1vuEm6Acx84njv2vDPvuFKCj1YEGVDmUBtUtNqCWW0WUVfvuKv8BnEpqLQUPjN/KWtxLJY6PuZGtCpIIuhYabD31ftSNrAHL+x/q46bUBCR2jMtfxjBzDYfplBstmUdW155EOV9OkNJkAnnP0EyQFkBza6JVIIVMYuIBH/KBqGo8SsMAY4XBJh7nYQUROMWthL4DaUJSEGt36VD345UjBrIu1e+AUBASAoFeLtKG8cRybokWr0VdQ0vWFEjGW2eDLrTx1atTpBKpcZIlsRNnuGqDEIR2gU77pIqrVvatZLWBc3ZT1WkVWHZ19D5PFPfFlVYCkl49omoC4YrU6WD2yuNC9CxZxXL5jV6tg2VOge0V5ieU1IuKYwkmSmlMEXyeehcY6/e8Qlb7q2TeobKApx9x8HcesaLRNu0oXyr/Qaz+wkj+WzM1Lwx27O2zJm1hOcf+iiP4CSShIjxp4N0xdTNthrAzY+dTXVNOXc8ciEvPvUBX372A1OnzuCHKTNSpOIqB1flu97PnD633XteXzBNk1NPPTWrbfz48XTo0IHhwzd8ZdAaEYrQLjTfK6WGAvet2yltuKjr0YHNtxvA1x9Pz2oXmCxwGomiDYRBAnQJVJOX3tXDjK8W8O+TnyMScekUyl4ltU5e8F2TRZVfEMtZaCKurvsRV4pCZrs056VVOcuaJPOWJOg/yGBO3EdUR+LhM6DGrwoQiia2Ntdg6hKXXfro2Itu2w9i9O0n8MPDnxCrbyNeVcnCj+ZiOxK/IQiZkBlKEo0qgobClpowyiyFZSSlF+09Jb25xjLiQ1Y4fpbNCiN/+JykokwZIsvw7khoc4RnYFf4BIS841KJLDJJQgLJp70mit7k3FLXTNpClI6AD5T4sKM2jtLSB+joc2EZbLHXYJbfm53+w5WKsO2RG+C3IF5ATaUglbxQeQt3IscQL73xUAqfUTiVii9gsdcZ23DPha+njPOZF3OVwu83kW6+3S7vWagksWTL41Mnzmf61wsZsHl3AEbtM5j7f+jLtIkLqO1aQfeBOqniZjv155tx6RxYhVyJk3j3mW8oxHE2iSzbyDefT+fGK5/ln7efjN/vY8zxuzPm+N1pamxhWL+DkK6mLUnh3f2WW41Y9U1vIGhra2PHHXdc39NYI6yRykspJYHvvNiTPzSuefBEdj1oMyyfSYdO5Zx55QGUdfcRFmGkkEghiYgo8+PL8Aez+VolA9eEYOZ3i1kSkbQkZCqxX9yFNo+DElKyIuZi50hFyQ8s5oq8ok1KJaO203CVIuYqZn86m4dOeSZFJqDVL812tgFbn5MeZ8XsBh488wU+ffIbXNulblRfdrnreEZcfhBfvDebcFyScLV6pzmRPZZCpIzothKEXYMWW9DqaAlC75TzVzLlShJRG8dzx41LaLW1zSmJsJuphtLjazsJhJ1Vr46r3o97n1NVOV323ISOg/IzySarFLa02bQ5wlMPpu0sriOxozZBT/1jS0VrQtFqZ6dSiWWQifbO0t8PwzKQhkHCVUQdRcRReUZ8W0psqaPsbZm9yAsBfTfvzl8ePoLnb/sE25WpsXORSLhrRK7+kHYvz+0rXcVtZ7+AUoo3HvicC/e6l6uPfpzWxkiKTBqWtdK0MkxNlzXzroqF7YIfkFFgufronUl5bYmEjePaSJxUqhZT+LJsYEOH9efsc45do/msb2Qmp9zQ8XOM8uOAkcCXQMqtSSm1/7qZ2rrH2nAbXrmihZ0GnVPwx3rhpWP48MFvaakPE5U2DW6EaqMEP+mspn5D0ClkZe3YlFLUx9xU8cKagJkKWiu1tPeY15MSUxEwtRqkPgaVPkHI47GGuGJlXI/r85sEpEvQyv6lCqBzSKUSEToKIk56sVZK0eCpnofvPpBT7z8CgJv2uYeFPyzLGssUUOUnSzWVNiYrQmZa7SUVtDra3pQrIdky6ZWFt7MWqfF9hsISisLpUZRnJBeU+fIlr6Rx3FbgryqhsmMZK2fVE/Us5H5TS4hhR6RsHabfpPeQTiz4bgm21It1Ml7Elsmsw4qEm3ajtQxwTYOY7cVluKtwpVV685D0zhKAaWjS8nufc+6mQilFvICr1yY79uXw83ek59A6omGb83a6g8ZlbVl9kktyroOBIDsw0jDT9qkt99mYnoM78/S14/LshUnsf/o2vHRndp2gY/+2G3ucMJIzt7+N+sVp9Z92s1apao2ZKjudn8vFxkGK7N2RTQJbZDssDBjSgxc+/mf+fPY8k4mfT85qO+3Mwxk0uDe1HavZaZetMH9uae0iUvjV2YaBv6/F+fxuEAnHsshE/yDiuLg88ezbLF8Sxsx4zCtFhIBrE8fGQFCqAvjiISr9mjSkUrTYacFeu5kqQh4RtNmKqhLTcwkSOMEASxrjKXVIfRyqFAilqM9IE2YnXGwUcVcbfxVQYgnKLFgaM5ASAiYpHXnqvIzf9PfvTGP+94vpObwry+c05D2LZIxFct6u1MboCh+AIOKC4enxXaWljrir6BAgL2gu+Tpzqyo8+0hCQtDItzMkpCDiORPYCaj0KXxGMnMwnoRgYAnwr4yx83WH8fxfX6FlaWtqfMP7DBNJb6yYy4/fLMmSvAwgZOqZOVIbxDOfgesqEhmp5ZN2m0JLsS2zXX2V98wVEJPaDrbd4SOY+OaPxD2jdnvkFG6NM3Dr3jQub+WKwx5i+dJmTRQYKbJI5tWyEJgZwaBJO0lyY3PYRTvz2as/sGhWPd98NAszaDL64GF89ML3BaWHsY/mV9t89Op3GPfCpCwy0c9DG5MMBK5HLMnrJqnXxMxSbxUKjgQ46Zy9Cz6LO+69nDNP+QcTv5iCZZkccvjuXHrlafj9f7hsUb8pfk4J4A9XdVwIMUEptfWvn9L/Fp5/7m1cZWMKHzpRXTgVoTtj5lwEBpXUpvMNuZK4Z1uRKFpVDGEL+o8cQJf+tbz3+Dd510gWWrKlol5KFq2MUun5BnffuA4jsgI3pvVBCmhMFK5hrhd1mSKfVlsRc6CsRNAS0aFlXUoMgqbwPHu0h1Im6uc10nN416xMtZloTIDPScefKKFJrsTUu902R6vqdKoQbRNqcxRlPkHQVFgFrNHJDL1CQVxp7yx8WuJJIum+nH5mgqUxvXBZBlkSjaNXLj554PMUmSTHcMn3wsoVPiVaPYnQtpR8JJ1cM1pE/jh6nvmNAki42sJvCkFVlwrunHwBP02Yh2O79N+8B+fucDvNK7LjnwZv3RuAK494hNlTlui5g1dZUpNKcuF2lAIlMTMyf0rPXoMJj/zzndRcYiLBuGcm0bVvDV5S5pwMzEqrqQpg3o/LvHnkPA9vHgaCBE5KSkn2EoBQ+imqZOVGLI4+Yxdmz1iE5TM59Pgd2X63wnaQHj3rePXtO1m6pJ5QKEBl1e8/0++GgF+fBS2NdZevYANFS3Mbt9/0FK3UI5Wb0tlmQglJ3DPW54r3SQQ6+7ni9T+xx6lbF/TuabZdVsRslkZtYjGbNjfBgmiUpbEE3386lxH75Puqt2cjyF2/Yq5iWYtN1NW2ljmtLj822UxtspnRYhPPSAJo+gz6bqnNaFaBBJmQlCLSbqZJtDmaWGJuOkNvTOrdeNQVrIhBU0LooD+pHeSE0G639XFoSMDyOJ4dRdBsC1ptrZqq6NWBiEueIV54EyqkHovaMHdi4VI+qp3XmQg7irCtChJC6lzPmB51FHFHpQ3jHmp6VLX7QSUlhoSU1PWpIVgaYJNdB7DFXhtT2amM0288IGXXAOgzvCsHnbMdYx+byMxJ+Sl5JApHSS/Zu/Qkad2m825JbK+SpMxg1KSDAMCimfX4/JZ3b/pPSoUv6KNjj6r2nkLK5TcLhsgimUJxIwYGMuNs0zI46OjRVHQI8emH33HVuffz0O1vtHNdjboutf/TZDJhwuece+4F/O1vl+cV9toQ8XNUXqvDmhljfkdYvGgFsajWKzWxGJ8KYYp8kTpJMirLzyjzuH50PTfuzInX7suT/3iHWDjh/dD1WVFXESjz0dTcknrQjpI4ymXwHgMZPHojvnljKvG2BHX9agjVlfHGTZ/gZmyhZSrPkvL09tJLa66XXEuY+AwjNb6rYGXcpVMQ/AGLQ6/ak8rO+sdZ1rWKluVtJE0yirQOXilFc0IRLPcRSjgpe8Pq0GZDwknX0/KJ/MU9ISEotAdU2NXG+a3HjGLB1WPzxtOLIQUXbanALSDGqQy1D7TvjeQofU8G2q6TGYXueBkb426GChBwXDz3aK3m6t+/I8dcvx//GfN4VvVF1/uckmhckZ8ub+ReG3Pv9xfw/fhZVNSUMnS7jXjl3s+486JXsQrkNXOVRGZ8VhKvfJnSAZzam241btQoorFE1vguCicaJzovjkE6c3My+WO7H7sCJ2PzlbnZSlaLV6isz+6My/bnP1c8zifvfw9ANBzn+iuepKKylEOO3bHdeX/x+Xc88dirxKIx+vTtwaitRrD9jltu8DaU++57gNNOPT31/uabbuWdd99i2223WY+zWjXWJqH84dCnX3dqO1ZRv6IJAJsYhrLyDJ76WAIXG9P7LxO9N06nRN/jpC3Z/vARfPv+DO658DWa67VaI1QWoKZ/Jcu+askZV1JWW8KQvTZm1KFa/P/sralcfswjmFJR5fMTNExAG48FihbHJi51ksSkzUQBCaV9lDWppJfRUL9arnruBCo6lqXaRh+1GY9+vTD1Xi+kEtNIx1REGuOU+ASlVjJ4cfXeV04GARTKLJxsz/QteOm6D6jrWUNk/sqsPsmNtswZJ1ld0XVlyj0Xksbi7Ij0lP0jQ82TmeIkmYjRQBF20ilNknebu0hH3TRZfP32NHY5bgu6j+jCzIkLvDT2irh0sxbnlpwcYkmUV5ew7UE681E8avPIv95NSRRZZZtRyALPUSdlVBgYWfaUXGhpKR2CmLS11PWoZtmCxlQ/mXxQJJPYrCoBfvYV3IxE/Yp8pWHHukr2PHwk1/79kbyzX3xifLuE8vqrH3Dy8Zfiug4qg8D69uvFS6/dTbdunQuet77hOA6XX3ZlVls0GuUff7+at995cz3NavVYmyqv1Xiy/4yBhLhKCLFICDHJ+9s749glQoiZQohpQog9Mto3F0JM9o7dKn6DJDh+v4//3HQuwWDSa0vRsWsZJaVa+2dZJltvP4y4iNAqVhIRrcSJ4eCkRHxb2Jz19wOzxg2VBdjmgKHc/fV5nHv3oZx928HcO+l8OnStLDiPtqbsBeeBq8ciXYmtFCsScRbFosS8HWNCSuJepHWhD99Wbt4yMOP7xUSj2TryYbsPYK9zt6eqrhxf0KJuYCetynKzd6WtccnisMvK2OpFlDX2OMycr1SEm+PM/Gkl9TFoSUBTPGlU10t6cl6O1BJOq61ojCuWtjq0+f1EXakD9dCLWbDMT6chXbA9lZz2OoOYo91/Izl2JQlU9qrOIhNIFqrKbhPo+JG46xJ3Xd568At+/Ho+McehzXGIuA6uUthK4noG9JF7DMp7Bq1NUZYvakq9b1jWQluzVq06pK+bjGNpD0kbS3vPXqE8NVmmdUOft3RRU0HCqO5URt1GHeg+sP3yt1Jl003y96DL+qZlFNBG/DOvPKDdOebm+srEDdc9iJQyi0wAZs2cx9V/v73d89Y3mpqaWL58eV77Tz9NWw+zWXP8bAlFCFGReZ5SKunus7adum9SSv0359qDgTHAEKAr8J4QYoBSykWXKD4V+Bwdxb8n8NZanlMe9txnNBOnPM349yfSoaaS7XbcjEg4zk8/zKW0PMhVl91Gm7scgUnAKCMqWpFIry62Q8Jq47yzr8M0TXbbaytO/vOBlJbpkMVQWYAdDtskda0tdxnIR69+n3X9YKmfoaM2ymqbP2MZcc/TzMLCh48V8QhX3n0Un775A+NfzXanzEXmzzNpqF08eyWde1YTbolx3RnP8tmbPyClYuQuA/jXvacx+YNZ3HnGC+2OlZCKNtul1DJSAXtAlnQQsSXVpSaGZWHHHF1hMMcAnIw0V1JlBBQmr6E94iTa1TZkaskkISFmK0r8AkcqWu20GBRuiWP5DAwUlt9i8C4DOPz6A3ATLq9e+x5T3vkJ15E0N0Tz67Z7OPyfezHz+8XMnbEyq10IfT1fxn0mpMwini/fnqbVUQVGdpXiwNO3ZohnbAdobYpw5fGPMumTmbiuZOAmPbj8/mPoulENtV0rqF+sJVjXkySSlgmdJSB7j6VQuLgIBDbgw8xSOylvnNT95OwZk44ZuTu3Ln1rufShozh2xLV50hJoMnFSauDklTLUW96/XbpVM3LXQex9+Cg23UbXdt96h6FM+HBK1njTps3l+Sfe59CjsxOfK6WYOX1eQWcIgE8/yc/WvaYIh8Nce+1NjB37Dl26dOH8889mhx1G/+LxclFbW8ugQQPzCGT06A1X3QU/g1CEEKcB/wCiZIRMAH0AlFJT2jl1beIA4GmlVByYI4SYCWwphJgLVCilJnhzfRQ4kN+AUAA61FRy8OG7pt6XV5SwxVYbs8t2xzNlsjakKSRR2YgwBI6wkcrBJY6wLX74fjYA3387g48/+Ibn3ihct2zPo0cy6ZOZvPfctwCUlAe48PYjKK1I+0Mk4jZNqpm4kfbXD6gAJaqELoNq2avLVilCKWRe6DOsCwt+WKZVFt5OOVDio/+m3Vi2sJFrT3+a7z+ek+o/8f3p3Hzei1x675GMve9zZn+bNgjnGuYjjt6ZBw2BYYBpGBhCq0kijiaCUx87jmDQ4pGzX6BhYTOOSqdMkV6/hFQYQhAwRUZMjoYO9lPEXUHEUOm69oagzdbn557j2JL9/r+9s46P4mgD8DN3MSIkuAV3p7hbcdciLe5OobR80AKlLVBaWqy4uxV3h+JWnOJOkCQkxC93N98fe5q7QALB2n345cfd7O7s7N7dvDuvjmlAuRbFcDEH0SRzpdXYBrQa24CQx2EMLz8JoTM4BoEaJUVq5uHC4ds4w+w5pkUSY1odxp2YNSb1kDMy5bE+5V89e59+9aYSHamsFiWSK2fv0bvuZAZPaEHfX5owpssydBaPP+v9NyLxSZ6MyBfRpnFJu0k9FgNIRQVqmt5fC4kSw3J480XrOGyEiq0wMZ/bgNGpfsPbNxnDJ35u1/brnL6MGDibPVtOmQReLKFhkXzTbyo+fp5cunAdnS6Wxs2qcfz4WWJiYk3qM0dy5Mz8mlcJnTv3pmLFsnz//TA0GkF0dCRnz57F3d25s8rrsHTZIp4+fYY0u3q7uJAuXTquXLmSZOd4FR4eHvj7++PqpLqlMxKzQvkKKCilDHytkSWevkKI9sApYLCU8jmQCWUFYuaBqS3W9DpuuwNCiO4oKxmyZHl7gf+nTl60CBNbdDISd+GFkVgEGoentxNHL7Fz6xGq1ijFrm3HCA4KpWqNkmTOkh4XFy3fzv6CDt/U5smD5xQokQVPH3vnum1rTxBjsA/+ihExuGvdSJ8lJcm83HDx0aAPU9QYSCzpyrMUTkut7qWY3G8trqavhkRS5NMcbF99kt+GrsFD7+4wIR7ecgmNi4bv1nXk0OpzXDhyh/N/3+X2tSdoEXhqXHE3pQfRSyOBOkVjrhUCbxctGpP+XqMVpM+ZCr90Pow8PICHVx5z7dBtQp6Ec3DRKV68iLEJ6FNiapK7aXBz02KINRKlN9ptjzUqgZMa00rBw8cDQuzrZ5hZP/EgldqWcLrNL70PxRoW5Miqc5bgSIRia3H3csUvrTe5S2fh+MZLdp+nwSiJNhowmmwOCOdmbyUuw+mpyZjTmmJ94tfr7ISJeV0T9OwFw9rPI3PONMw7OZizB29xfOc/HNxwAXPHtdqUYMCEppzac43RnZcQG+uYjkTp07pidLaisb0CjVbg5ZuMsOBIuxVnqRp5Ld5glnshjRgs/mVmm4m1X2foYhzHmCKVD9Xqf8LWLX85jK1X+zHEGhWBOWPyKjJmS2EZk5AuSJsULG5urgz5pjs7tu9nxvTFhIWF07hJbXr36fBKY/3Vq9epWLEspUuXQKu1FpXz9fUje/ZsLz02sej1ekJDX6DVakiePDkaTVJaKV6OlJKgoCAePHhA9uzZX30AiRMoNwHn1sHXQAixG3DMawHDUdRXP6D8Gn4AJgCdcW6nicePx/m3VEo5C5gFSqR8ogeeQMzeX3HJlCUNoU90GGNjkc7qowOdPv8Ob08voqIUwaDVahg/+UtatqnJlUu3cHd3o2TVPE6P/ef8Xaftn1TPhW9KL3auP0VA+DNchSsuaNGjR0cMRmFEFxTNgikhPCccxYwv0BHLrl2n2b39jGLExtH1OZmXOxqNQBerZ/eucxzbetViSDEgCTPq0Ag3ilbKyamD1y0fjEFKQmP1+Lq6Kok1dQZmf7uVIbNboYuOJVP+9GTKr3xFSjYrwtDq0+3Oq3iTGZAxBnw8XR2ixxWbkcRdA5GxkhQpPXnxPMrRaC0l9289Z8bA9fSc2MTp/Ws8qAqnt18lMjTa0rmU8Em9/BzaeJFPauZhxU+7iYlUKqcYpCTCYH1CN3ucuQiNQ6xNXDuLUSorBo2bhgunbpOrWCY8PF25dOKO3bXF/fLev/mM73suYdqW/tT+vCQPbwXyz+n7ZMuXjpyFMwJQsWEh3Dxd0IU6xo0o8SoCJfuVASE1aBFohdZyTkz3zigVL75e4xswvs8qoqKV76q3dzK2rTlJxqyp8PL1ICI0GiPSEnsFOKlVYgrSifNzKFw6h5NPAoJMTjD2PQiMNp4HBoORh/etNgiN0CKlQGKgXYfG9Ojdlps379C6VS+LMDx27Ay3bt3j94mjnJ732rWbDBo0jH37DrF69Tw7YQKg1zsNSnojXFxcSJUqZZL3mxCEEKRKlYpnz54l+JjECJT/AUeEEMcBy2wppeyfiD4sSClrvHovEELMBsx1SR8AtutUf+CRqd3fSft7o2z5YmTImIaAR/Yfxo8/96NylVL8MmYBs/9Yi4hjGlfcVoVFmIDy4xg5dDozJq/m+lUldqJcxSLMXPgdfil80OlimThuKRvWHEDqnD9d9Riq5AN69jgEhFKkKJZY9FhLpQY8fsbjx4FoccFW1sVGx+JOMrRoiUWPu03qGIAm3cpz49IjutX6jdhIA244Lo+LN8qPf5ZUnDzguGqL0usRQoNeSg6uPc/V0/cJuB1M1vzp6DGuAUUr51QmUKPV0BxrtHetDY3Q4ap14i5rhAhTll6Ni+ISHWsEF43VPVhnyhywf+VZ2n1fGy9fx7SbabOmZPzhfmyceJArh+/g4ePGnduBbFpykk1LTuLh5UbnEXXYPf8kD64HEm2wcQO2eQbXSYOSXsW0KpPS+rRufh1tfpLWGZg5agsbFhwhS/60uHppiYmINeW0cv4sdPbIDQIfh5I6vS+ZcqQmU47UGI1GNi07wpHdl0id3hef1MksBnxbjEjT98FqQQHw9klG1qxpuXHhkbJFSgxIIp/HMLLLIrsHjPCwKG5ciOLGhUd4JHOjcIXsXDx9Byxy2N4QbxYuIs5jio+fJ52G1MYZufI6V1XFfdCJjZPYTggNNWtV5Pcp3wEwcOAIB0P/wgWr+H70YJInt49diY2NpX79z7hzR/n9aTRKDgLbz8HPz7nTzMdMYn2bErN+mgnsRVE5nbb5S3KEEBls3jYFzPaZjUBrIYS7ECI7kBs4IaUMAMKEEGVN3l3tgQ1vY2xxefAggK+H/EDZ0vX4pGgNWn/Wgz17DuHq6sKSlb9StJjioZMiRXJG/diP2nUrkczTg75ftiF33sxIabBWx5MSAzoHIQMQHhZpESYARw+dZ9zoeQCM/t8spk9cw6MHzwh4+hg99iqvpp9XpFBxZclavnpBqyEc6RCIacRoqoJngwC96QkzVuiJNhn8/dJ60XtMQzoMq8mkb9cSE+k8WhrAM7kHru7On18ijHrCDDpijQaMRkmAKa3L3StP+L71IoKfhOGfNw2pMik/WHNsji2x8SQ/tE0Hk69cVtJk8bMIFZ1RESYxJv9iQ6yRJ3efc2bfda47CQ5MntqLL36sy0/7euGXIyUBD0Is26IjdKyecpDfDvflx61dLROneXViS4w0EGGMIdKoI1rqTY4PStChLs6VGTFy/85TDm+/SGRkNAZhwIApTN8JUkrC4nj9/ThgCd/3WcSudadZPn0v9wKe2AUMmid2W+9DW8JfRHHpwl1L+6vSSZr3i47SkbtkJgpVyOawjxGj4kov9BiE3lQZXocBA3XblGLpkWH4Z3fuJVauUmFcXBx/I1IY7e6Lb/LkGE2/L+XPSN4CWWjcoAuFC9Tk4oVrDn3odLGEhzlWYN2//5BFmJgxz7UajSBt2jSkTp0qvlvynyExKxS9lHLQWxuJPeOFEMVQxP8dlPorSCkvCSFWAZdR4t/6mDy8AHoBC4BkKMb4t26QDwp6TpXKTQl4ZE2SePXqTTZu3MX8BRNp1boRuw7MJ+T5C7y8PXF1td7ulKl82fHXTHZvP8qNa/eZM+NPngUqXkLKGuXVsn7PzhPodLGsWb7Hrj2KEDKmz0Cnbk0oWionpSrm49CBvzmw5xTpMqSm77dNmPXLZqKirWo5IwZiiUGZqgVuJMMVd5vt1olOLwykzOjDzD0DSZ1emeQvnrptGrtjwj8hBLVaFSdNBl/WTv2LGBsX5FhpQGcSagaMaKUGDQIXk30pJiqWIxsvUq9LGaq0L86m3/+y5LSyRalGqSTKtPUc0xmVGh7JPN0oXDMPJRoWIOThCxZ/t53QoEi76dszZTK+aTKHsOfK03vh8tkZtby9ndODmUvH7zi0PXsQwtP7IeQrnYWSNfNyctfVeKdeiRLYp0WDRIMBs7E+jvrLicHaxUPDZ92qsmbuX3ZCXCJJnyUF2fJaNckB94LYtNQ+jX5MlI7SVfIRHaQnNDiCZClcuXxFuR6tUfEKdBxvwqNKbHkRHMG1i1bzpjm+xIDe/rqEchad0HFg71n8C6RkyewdBAe9oHqdEgwb057Uaf0A8EjmTst2NVg2f4dNvIrRZLy2CvKQsBDMZn+lzchvE2bZPXgI7G0rxYsXJmMmR0280Rj3EUYhZcoUZMnij4vL64f0eXt7Ex4e/uodE0hISAjLli2jd+/egFJP5ddff2Xz5s2vOPLNScxd2GcyaG/CXuXlmCXwDZFSxuuCLKX8CfjJSfspoFBSj+VlLFu61k6Y2PLDD7+zetUWLpy/QrFPCjJy1JcUKKjYPW7euMvDh08oUbIQdRtWAiCZlxsj/qf4xRsxIKTWbrnp7uFGTLROSfEiY5EYCQwysGv7EfTODKzaWCKNIXz//VRCgsO5d/uxZVuGjKlZsmcMwc8i+KrbFAKfhaAjGlvnPR2RSsAb5rTl1hQyWXOnZeG2ofilsgY6ZsuTnn/O3De5oMbiggsCQTIvdwb91owi5RV9+A9rOrFk7G6unLpHeFS04l1kwoBNhDTgIRUd9dKJe5k1ahtRETEIwFPjYjHkW65XmuIl9AIXjQYPbzeeh1qf1KNijAxvMR+AlOl8GDq3FYuGbefOpcfESiPCTUNYdAwRL6It9/3CkdusmniATiMcVS/+udJw7x/7OAGv5B6kSKuoSgZMac6wJnMsuaziEovBIqT1GIgROoTU4GqyXlmuy8kkrovR8/fZ64RGvcAVN4u6SApJ0Qo57b43Tx4+d7py07pqmHfoK+7dfELT0iOs5xPx+EabxpnMzR2DLv6VKJgzLygBk1nzpmPt4kNoTS7JSsbgWIfgXrCqrJ4+DWb8yKWW9q3rjvL08XOWbB5pafvupy48eviEfTuVbOFSSmJlFOa1q8SI0Wi/+jbIWId7odwrDVIaKVgoL3Pm2UUqWKhevTKZM2fi/n37lWvatKnfSJi8DUJCQpg2bZpFoLxLEqPyagsMBQ6jeF6Z//6zBAQ4nywA7t5+xI7t+3n06Albt+ylXt32PHsWROeOQyherD4N63cmf94abNu6H4AOXZrgnkyLURoxSgM6Yzg6YySp0/uwdttv9BnYyhSkFmP50URHx9Cry/eUreQoRz28tPz84zz+PnWFu7cC7Mf9KJCVy3ZQpnJ+fp3TB6/k7jibRexUZwJihY4IEcrlm9dZtXQXPwybw7aNRzAajfT6riFaNyX3khFJLLEUrJSVNf98R81WVu+p8Ogo/nl4j8dRwXbCxIztBBqNnliMPHsYSlREjGk7RBsNDhODHiMx0ki0NBJj0BMcGmkJ6tNLo8WFFZT6HD91Wk66wmnQpHQj0qgnIlpH2IsoB/fdM/sdbT7hoVHkKJERjYcGoykA0SglTfpWZPOSYyz4dSchweFMPzKQgiZ1j+116TGgN3k8ufm4EC10yjO2MBIjYtHbhRE6qrYMGDh1+B+Mwkis0KEXscSKWMXBIs5kn69oFnx8PR36KF1FUcUe2nXR7l4aUbyxFFdiHTqTehMgeWpPVl38ljzFrOZKcxii+Z9ih9GDFjp+XYuiFXKAAIMwEC2i0Gl0TtVqtvfI4KQw3amj/3D3lvWhyN3DjXkrRrHt0FQmz/mKGBmGUQnptPw+3N0S5sI79Y+f+PvcTo6f2EyePM4dAVxdXdm8eSXVqlVCo9Hg6upCtmxZ8Pb2dro/wIwZMyhWrBjFihUje/bsVKtWLd59Bw8eTPHixfn0008tRvCqVavy5ZdfUrlyZfLnz8/Jkydp1qwZuXPn5ttvv423r6FDh3Lz5k2KFSvGkCFDAHjx4gVNmzalQIEC9OzZM94V15uSGNFaAOgNVET5Hv0FzHgbg/pYqFWrKhN/dyxg6UxdFRT4nKFfj+PPNVZNXGjIC7p1+YazF7bh6uJKWLg5jbo56ZIevYyiZJkClCxTgEePHrN4gb1pKDZWT+4CmUie3IedW47i4upC/cYVWLXKnNtKODWsXftH8QYrU6kgv8zqQ5fWPzh4P1Wt8wnXzj3kaUCIaaKJVSYNaWDcqAUAzJ+xiQZNKzJ57hCWHxnGtlUn0ccaqNWsJLkL23tu37j0kIGt/7DJWxWfg56C8qRrxCXO19SA4o6b0T8FgY9C0ZtWJ+ZjdBIE0mQRsB6jlcKysgkNjGDb0pNonXxWtvchXeYU6GL0rJtziBP7riKE4Ozhm0RHKsJWaMBgNKLRCub8ts2izpv90xZa9a7CqGUdWDhmF3vX/s2jZ4GmK7aaoCPCHb0BDcKAURpQwi3tVYiK7cHeY8q2v/I1Ctr15eHpxshp7RnRYz6RpnOVr1GQz7opk1uqtHGKXgnQEYOrl4boSNPYJKTxS8nIme34a895dFol24MWrUV9pRdKtTCjyY5RvkZ+Og+rA0DuQpm4fvGhRTApLtdGFOWmrT3PZKDXaEA6Tnh3bj8iaw57dVS+AtnInjMjyTzdiYq0v5flKhTjeWgwZ05fQqvVUrFyOfbus3c39kuRnOYt6uLt7eVwvrgUKJCXnTvXAnDlypVXel/17NmTnj17EhsbS/Xq1Rk0yLnFICIiguLFizNhwgRGjx7N999/z9SpirbCzc2NgwcPMmnSJBo3bszp06dJmTIlOXPm5MsvvyRVKke7zbhx47h48SJnz54FFJXXiRMnuHz5MlmzZqVOnTqsXbuWFi1avPKaE0tiBMpC4AUw2fS+jants6Qe1MdC1Wrl+fqbPvw2YSZ6vfkJWJAxUzoeBziG61y65Pi0GxYWQYG8NcifPxdGGWOazLRoTBXmqlYvY9m3TPnCDgIFwMPDnWkL/kd0VAwarYZbNx6wYoU53490GlPwScl8TP59MXNn/UlwYKipZrgRjXTBVXjionWlz+AWXDx3m1FDrELTVk9tZvO6Q3Tt24Qin+Smx7D4q8ttWn4Ufazy9GsQBvTS4CAsLOexPKE7x4CkYLUcbF9iXSQbpTK5mUcnbP6ZjxHSWlo3PpuAWcy5ebjQckBlRnVZxMHNzrMLSKPJZdUg7WxDUsLyP/ZzZPdleo1syMql+9ALg+UEWrTKhKrBeQZFAZlzpSFPiYxsXX8MfYwBjWkCd7A9mKLgK9YuRL5i/g5dVa1XjG2Xf+bssRukSe9Hxqwp+f2nFRzac47U6VKQLlMKnjx8btnfzVNLZGS03TmEp5HZUzZx4uAVtFKLi3C1OGoo99aAUWO9kEMHz3PnxmOy5UrPhBU9+X3Yn+zYdMyqVBUSgzSgEQIXDw1IgSEG0vunJEqG8+SRdTwABqmnc5uRTF8wjFr17KtkLJy/jvCIcDTY5NET8L+R3Sn2ST4e3H9MMk8PUqXyY+aMpYwd8wfBQSHkzZeDyVO+T5AweRMGDBhA9erVadiwodPtGo2GVq2UwnVffPEFzZo1s2xr1EipX1i4cGEKFixIhgyKv1KOHDm4f/++U4HijNKlS5Mjh7L6atOmDYcOHXrvAiWvlLKozft9QohzST2gj41R339Fr94duHH9NjpdLH4pfPHx8aZU8fo2QkYJpCpWrABXLt9w6CMmRsfZs5ctk5+UiiPpJ8UK8+1Iqx60dt2KpEjpy/PgULt+m7eqBSjGSoC8+bORPWcmbt9U9L0SpR6G+cdWqEhO3D21fP/dH4B9aVUjeoSbjj/mDqdoiTwULZGH8LBIFs/aRlhYJGnSJefWzYcmjzTFlqNBy7Ur9yjySe6X3ivzysRsaI4iCk/padGn22ecVTBgxAXH2Jci5bNRo3Vxi0BRIr/tbUnO1EZmYaG32DAcdfnFq+bCP2dq6ncui8ZVxCtM4p4rbuVBgNtXHzNm4DKLys40IGUyRUOBklk5d+KmXV8ajSB/icwEhz9nw+pDikrKFPsYtwyuAQOR4gUI2LLzAFt3/kXpCgWYsnAQfims6hgvHw8q1FRUox2b/Mixvy4BcOPqQ7QaDY1aVCEw4AW5C/qzf9cZIm9H253naUAITwNCTOc0oiVOduA41RVjYmL5Y8w6fpnXi3SZUjBuYVdSDPVk+ay9dvchhhjCY6JBQqVPi9G8XVV6dfwJLa4mpxCBET06okAvGff9fAeBcmDfCRR/RcUOBcqDT3iYYuj2z2xd1fTo+Tmdu3xGyPMXpEn79r2yFixYwN27dy0rjoRgZzs1Rd5rNBq7KHyNRmM3vySmT2fvk4rE2FD+FkKUNb8RQpRBsaf850mXLg0VKpamWvUKfPJJIXLlysb8hRPInFkJJsuWzZ9FSybx1ZDuJE8ev87V9onZxVXDrv0LSGvzpff28WTlugmUr/gJ7u5uFCqSmwXLx5Irt33EvxCCGfNHkDO34q+v0UKdRuX5YXxv5i4bxcbdk9mwVvEMc6ajj46JJmNm63l7DGjKoUuzOHdvCb0Ht0BKIzoiMaAzOZvGsG3rgVfepzotSimrA7MnjpBEiAjCRTjhIhzh5SSdPIoB23JvBLQaUIXW/6uGTsZSqXEhy37OsCThJJZoYtChI1JGE0E00UJHGJHE2giieh1KM259V/pOaELmvGkIDHgRT89W4rrgxgodMSKaGBGNXsQSHOikDwHtvqzBxD97Ub9NGbQmN9i8RTNTvWURjp+6wPV/HsapBmrvaWXEqEy0Nh+hxMiJw5eZ8P0yu9MZDAYO7TvLwplbLcLEss1oxKCJZfamr/h6XGvyFXbMIJHM08YeIZT7ab5ux0BFhXOn7AVl32+bkjqTt81nEoOOSMuq69C+c0ydsBxQ3OejCSOaF+iIxPwJ37z+wCGAMJO/NWOwrQ0lYzyZhF1dXd+JMDl9+jS//vorS5YseWmEu9FoZM2aNQAsW7aMihXfLCeYj48PYWH2JQ9OnDjB7du3MRqNrFy58o3PER+vXKEIIcz5G1yB9kKIe6b3WVHcd1WckDNXVuYt/JVcObORMlUKyxdq7/7l/DxuBqtXvbwwkIzHaFa4aB7+3DwJgJ3bDzJn1jKm/bGQlp/V47PW9S37FSqciwPH53Pz+n38UviQOk0K+/5fkdnXxcV5gGTD5pX5ffxibt2298nfvuUw16/eJXferA7HXPvnHqEhYRQrkZdR0zowst889ObCXcLqkixdJTp0uFhWDcqKzYDR5FYL+UtlZsvmo8yYvBEA/+xp6DOhITfOBrBp0VGHcxswEEmUJVBTRyxILLYUKSRRMgYDku7f1qPdVzW4/Pcdxn29nIunb5MpSyo8PF0tKU/iYk6waE4Drxd6DMLWc82AXli9izRoccGFPIUy03ZAdTy9PRgxrR0DxzQnOlJH8hSelM/Z09J7XIwYqFK7KHt2nFKM104fNCX7d/5teff4URDtm47k5vWHCAQuwtFYfeSvCzwPfkGKlMnp9XUTjh+8wosQJR5DCEG7XrWYNWGT9QxCiSNp3KoiTwOec/TgZYeHE98U9s4ARw5c4O6jB5iTr1jsSSaPRimlXayVMwoVzeXw3ezRqzV/rt7Bi1Cr622Lz+qQI2dmdDodv/82iw0bdpDCz5c+/TpRr96ncbt9K0ydOpXg4GCLMb5kyZLMmTPHYT8vLy8uXbpEiRIl8PX1ZeXKlW903lSpUlGhQgUKFSpE3bp1qV+/PuXKlWPo0KFcuHCBypUr07Rp0zc6R3yIV00sQgjHGcIGKaXzXB8fASVLlpSnTiWto1pgYDAtWnTh2FGl3xw5srJ27Xzy5bdXB7Vo1otdO+0NhLb6folk7YaZxMTEkDJVCsqUKWa37+qVW+jRdbhd24hR/Rk4uHOCxjln5hqGfzMRcFSjfFIiP1v3zIz32F6df2Djuv0O7fOX/kCtetZsqBHhUfTs8BMH9ylljdOmT8nMhd/y7GEY33S2719iJGVGLx4/CkaLFneZDFfpZnE3BZPgSWaw89iSSNJnS4EUktDACAwvwMMSPyOJEFE4m3Vt7zUogZt5P/Fn2C9f0L/VZEKCrcFtLmhJ4e1HxItouydySz4tofyGOgyoxbzJ1loVir+Us3QcQrkaYaBa3eL8NLUHvn6KHv9FaATlsve02VNjN06f5MmYv+V/tKw5nOgY5+l9QFCoSA7W7hsLwOBek1i/cr/N9bg7qDz0UkeVmp8wd6XiPXTt0j3+GL8Wvd5Il34NKF42Dz8PW8aSGTstx9RvWY6xM7oTq9NTJW8/wl/E2H1WY6Z3o1GrCpb9J41byZTxqzDiqKoxoMfHxxOhgdCQcKcqGW/vZCxYNZpSZQs6bLt58x5zZ63m0YOnVK9RlrbtGuLi4kL3rl+xZIk1C7YQglVrZlG/foISdcTLlStXyJ8//xv18bHg7FqFEKellCXj7vtKgfJv5m0IlF69vmb+vOV2bWXLlmD/gfV2bWFhEYz5cSrbtu0nJkbHo4dP7CYOozTi4eFGtCn4sHSZYtSsVZnFC/8kLDwCrXAh5PkLG+OyxMXFhZHfD6Bj55b4+Lzc0CilJFfWWoSFRtpNrl7eHhz7eyVp0sbvwbJg9nqGfz3Frs3NzZWTF5fbrYTG/7iQP36zf9rKki09B0/PZeGU7UwbuwFdjOKLpfHWExFhTQcipAZP6Y1GanHBVZmcNQaEjCsI9A76+2zpMyB1gsdBQRhNK4e4OBMoBmEgVdrkBD8Nc3jabt2tGqtnH7TU8dDGWdxLJFkLpubm5QDL+RRvLOe/L6Mp6gagQcsKjJ9ptZV1ajyGE39ZM8oKBKnS+FKmckG6fdmA3AX8mTVpA7+OXurQr7K/ht9m96ds5YJotYKaZfoSFBhquWolnM8VITSmomJK+WohBCeuLWDD6gOM+W4eBoPiseXq6oKnpzvFS+endbtahIdEk7uAP4WKW11s/z5+jaG9ZvLgzjPc3V1o16s2X45oZTeu7ZuO0qfDeBzzHIBexmBfs0SDOXlqgyaVqFy9BLXrl8MvRcLL+QYGBpM9a2kMBnuhXq1aBbZsW5LgfpyhChRVoDjwNgRKzpylefggwKH9ydNL+Pomd2gPDg5h146/6N7tf5Y25TNx/FyETTCfxqQWips6BaBwoXzs2reEZMkcI7xtWTB3LUMG/WzXNmX6CFq3rR/PEQoxMTo6fz6C/XtOAkryytHj+tKxa2O7/epX68/Fc45OCPtOzCJHLn8MBiNXzt3l7OmrjB46z2G/LJkyEvzAPo2Im4urxbhvto3EXYAU/CQbLwKjCLgfhJDCcq9sibsa1BFjWWlo4qwKANr3q8WlM3c4ffiaYoA39atUJjSaXGcVV2I3lPsei2NEvxmjyQHYgB408PWoL8icPS3rVu4n4kU0wQGR3Ln+GK1WQ63Gpflhcld7OwYwsMvvbF1vr+bL5J+GL79rzYql2zh84KzD9Qq0CDRotRqTwLCi0WiYMn8wvTvYfyds71rqNH7sOz3dUrPHFiklD+4+I2UqH7x8HLd3bDmCg3sc/XgUhwrH3GJZsmak75dtaNOhjmW/sBcRJPeN3w5py53b9ymQv7JDe7FPCnHk6CYnRySc1xEoZcqUISbOqnLx4sUULlz4tcYQFBTEp586qu/27NmTYO+vhJAYgfJhhXj+C0ibJhX3799H+QkrT1h+fr54etr/wC5fvkaP7kM4feo8Pj7eZPJPx8MH8QdKxp01X5YG4+LFq6z7cwdtv2gc7z4AHbs0I0PGNKxcvhUhBG3bNeTTGuVeegyAu7sbS9eM4+TxSzy495iyFYqQIaNj3qUMGVM7CBQ3NxdSpVbStWi1GgoVz86ZU/84PY9wN5AitQ/PAxUDo0cyV9p0/5RFk3ZZbUBObAjRkTryF81CwP0ghzgOUFY1fsm9iYnUE6vXoyOaKBFueXJ21bqBQZl+XXDD3dWdhq3L06hNefq0mMTjR8EYhRGjNNqMweRlJiRIlBUAIt7PSUnDaPKkMsLYEQssK00z/b9qTYceDUiRyvlT+YjxnTl96jIB958DEndPLWOn9WT6pNV2wkQ5nzSNR8kDVrthObaut/epqd2gDPNnvGyilQQ+C2H3thM0blnFbktEeBQLZm3i5NFLZMuZkS69GpM5q9W76vTxKxzYcwaj09Wdc3thhkypLMJk88a9jBg+iXt3H5EzVxbGjv+K6q/4rmbLnpnCRfJz4bx9/ZDGjZ0nnXzbHD9+PEn7S5UqlSXW5ENBFShJyJEjJ7h85RJSmgscgQY3vhzUw65AjdFopGXzbtw2GbbDwsIJCwunW7f2eHp6kidvdr7+agwxMfE/4ZpzbsXHzRsJM23VrluJ2nUrJWjfuJQqU5BSZRz12QAXzl0l1hhlV50R4ItO9fH1s58gc+f3dxorc/PmA+av+pbQpzHoYmKpWq8YqdP5Ur5GITYvO4qUksuX7vDPBXtDbvUGxanfvCxnjl4nJCgco7SuaAwYkMJIyixeDB3zOT2aTiDKGGZZnQDEGnWWW6sjms79W5Ezv+Kxt/Hvnzh58CqbVh1m2+rjYBIaOqIwCj0atOhFLHpiEVLg4+VFVIS9QV9JxOho5I8rgGZPW0erjjUB5wJl5pQ/uWeTCiQ2Er7s+QsBj4Kc7m9m3OQ+fPZFTYoWz83iOVsJD4ukbuPyDP2+Iw2rfvnSY0ESq3O0gXRqNYqTRxXvsQN7TrN57UG2HJhMugzKk/K9u48tAtOIwWIbepmXmE9yRW174/pdunYcZvHuunnjHu3bDuHkuXVkyBB/mWGAxUum0L5df86fu4yLiwut2zThy0HdX3GNKq/Lu6vW8h/gyy+HERlpv3QvUDAX33zTz67t1KlzFmFiy/r1W/jjjzn8/tt0qn9qX+rTNrGkGWteTEcqVHJYjb4UKSWLFv7JZy16073rUE6fenXsRXwcO3KWujW6sW3rfqL0oeiNMWTNkZ5fpgxkxBjHH3OBwtkRGmu2YCmlKUusns6ff4dPGhdadKpC6nTKyqZEhTyM/KMDo6Z1JFuhlOixpvPQEU1AYAA58mVk/ckfGPbb56TL4keECCFEPCFMBKIjktpNSlGqUj4G/NDMTpg4Y9N6q0u0q6sL5T8tiHSLJYJQoggjklD0Qufg1qsnlmcRAUQQTLQIpefQBvT/vgm+/hqMmlfHEERGRlO3Sm/u3X3sdPuhOKsQgEcPA02p1Z3jkcydOo2U71bXvo05cHYmp28u5sffeuHtk4xK1Yu9dEwuLlry5M/Cj8NnM3vqWvR6PcePXLQIEzNBgaGsWLTD8r50+YJoTZ6O5tzGBvQYZCxaV6PTVcpfB49To3JHfv15joOrcHR0DFs37XvpWAHy5MnJseNbuHzlILfvnmDW7F9wc3N75XEqr4e6QkkiDAYDZ8446ofv3nUUHHHVX2aePVOeLG/cuMPNm3cZOfIrbty4R8qUfnTq3JLVq7Yye/ZyIsIjiY2NtQgUW9uKQNCxYwuqVX+16sqW/33zM9OnWQ2Vf67ZzsbNc6hQ8dWCKSIiisUL1nLm9CXyFcjJkb/OWXJKSQzEykjuPbxNvcYVnXrv+Pp5U6VmMfbuOI3i3ytNevUYpM7IwF7jOXFpGe7ujhPB9s1HiSYSS8g4sOnPg4z5vQ/J/bxo0bEK127d5PJUa4XqaMLRCUXw5y306jKwd289Rq832Lmr3r/7xOTyHFeoK86wymooxhQAGk2slPw8bjbmpY9Ai0ucujLmtDaKgFSe4oMCY5gzfS2jxzkm+suUOa2DStHdww2j0ehgHzGr/HoNbEHy5FaHjY1/7mfCuMXcvR1A2YpF+Hp4B65cuM3pE87VkNlzZaBxDesq5rcxSxj8rfNcro8Dgnj04Bm7dxzD18+boaM6MXbUfAyWpI0S4WIkWqeo/jRoEUKDq6uWqJgIYsINnD93lXNnnZe8dfdIeLndbNlfv9yvSsJRBUoSodVqyZMnJ9eu2QdzFSiQz2HfQoXyUb58SY4cid8hQErJ7Tt3mTXHaiD9dkQ/vh3Rj5MnzlK9aivTfgZTOhRBzpxZWLZiKgUK5iY8PIJVKzdz7+5DqlYrR9Vq8QuY589DmTvH3htLr9czaeK8VwoUg8FA88a9OHnivKXNw93RwywyMprgoFC8vR0TFQJMnjeEQllaYNArT7CxMgojigopKCiES+dvULxUAYfjXF3Nk7xNUGic7K9/rtztcNyiuZvpO6gVxcvkxcs7GRHhzksDAxT5xDH2oWzFQpw4bB+G5ZPck5af10CjEcyYugqTOLWMTUprDQ1p9qwyxWB4JHPD28eDB0+sqd7NQXrX/3Guvuw9oCX7d58iJtqqGu3cozErlmwjOOgFVuElcHXVMnf5KCpXL27Zd/P6g/TpOsb0TnDk4Fl633zIX2fmc+3KPc6dvsbv45YR9CwUc+rHf/65befQEB2tY82y3bi5uaCLowrz9PagYvGOltVF5izpWL/rN47+dZ5rV+/il8Kb6X8stdwfI3olBYtOF0dQOz6EpErlR6PG7yae5ENm+/btDBgwAIPBQNeuXRk6dKjd9qVLl/Lzz8oc4u3tzfTp0ylaVEl4ki1bNnx8fNBqtbi4uJAUDkqqyisJGTt2pN1k5uHhwY8/Dne676o1s+na7XOyZMlE/vwvT1kSlyJFC5AqtW2golIPolXrhhQomJugoOdUqtCMAf1GMOHXmTSs35Hmzbrz8IFz1UlwcKhDllqAgIBXl/7cs+uInTCRUhITE+2wX46cmdFqBcuWbmT/vuNER8fYZTxNlswdn5TuRMsXRMtQDOhM2nZlXM6M/gCtOzgaWFt3qGX3PuyFY8Ekc5ubmyuL1o1EqzWn7FD+zPgk9+S7cUpsz53bD+nZeRSlCn/G4aPHyV8km2U/Ty8Pxv/Rl//92J6vv/+CHLkzOeQ9i7s6MxBLLFHEyDBcvHQ0bevokWQegzOKlcjLpt0TadelPg2bVmbqnG8YOrITbTvUM+1hzgVspH23BnbC5NTxS/TtNsayj7meyKOHzzh84Cx6fSxzZ63j6bNADMRaMvmCNfuAmTs3HzFu8gCL55dGo+GLzvVYu2q3narq/r0nLF24lR79mzPhj0Ggccwa7dy7UXF1NlO5amnWbZ6eYG+vfysGg4E+ffqwbds2Ll++zPLly7l82f4hJ3v27Bw4cIDz58/z3Xff0b27vcp53759nD17NkmECagrlCSlUaO6nDlzgJUr16LVamnbtgU5c2Z3um/KlH5MnvIjoKwGihSqbrKrKHp4jdDSrp3z5G3u7m5Mmz6GLp0GEx6uuNVWrFSafgM6ATB75lJuXL9jd8zO7QcomL8aP475hr79Otpty5EjMzlzZXUw5CdL5k71qq3JnDkDA77sQvHijmny79yxGoXN7s5GGas8w5pqkadOk4IGjatQvEgju/xDWo0LWq2WUmWK0KlLc3LlzkzAE0ehlzFLKjJkci5QvvxfW1xctKxZtgedLoa0/t7cvHuDyb8v5sihvzl98jLJvLyJibEXmPUaWVNPFCmeizrNSrFu1X7ME5pGavH28WT3qSmkTOVLVFQMzRv059FDRcg+uP8YL+9kTFv8PUJqKFupoMWILIRg4syv6Nr2ex4FWK/HmeOBeaXy7Nlzrl+74/QaS5eLv8xP3gLZ+PGXPnZtg4a2QyMEK5ftREpJi9Y1GPy/9nb7TBi7EINDDXRF1RgREUXfrmN58SLSaWqeOEfgm8Kbpp9Vo2bdMlw8f5Os2TIQFRXNwnkbHfY/d+aq5fWJY+cdtoPyvQt/SfXPn8YNwiOZEqPlEUftJaUkOjrmlS7z75q9q/9mweidPHsYQppMfnQcUYvqLT95oz5PnDhBrly5LEkfW7duzYYNGyhQwLqSL1/eaostW7YsDx48cOgnKVFXKElM/vx5GDVqKN99NyReYRIXFxcXlq+chk9yD6V4ltRjlDouXIw/s029+tW5euMvVqyaxr4Dq9m2Y7Ela+rly45ZjUGpTf/tsPHcv//Irl0Iwew548iY0Zr7KEUKX44dPcOpk+dZt3YHdWq249JF+5KpN27cITQ01Oo+a/N0aZA69MZoatYpy4GjS5g1Y5lDMjuD0UBsrJ4jh87QtdMwTp897dTRII2pUp8zXFy0fPm/tsxZOYygiABOnjrH+j938+OoaezdfYwXL8J5EvgYKQwIoaTyr1GnjINzwMCvPydNOut5hFby/S89SJlKcQTYsfWQRZiYiQiP4vLla3incOWbr8bTqd1QNq5XcqQVLpaLIxcWULK0rRecfZliKY12UeN58md3UK15JHOncfP462jEd0++Gt6Bk5eWcuryMoaO6Ozg1HH9mvMUJ5n80xDy/AUvnKzq4mI0JcUJePyMT/K04uC+0/hnTotPck8yZkrjdAWRN7818UbmLBkctgshqPJpCXLnyQZYHTTM3630GVPRpFEXihSshX+GUhQtVJOh34zlyeNnzJm9jJw5ypIqRX6qVW3OxYvO7UDvmr2r/2bSgHU8fRCClPD0QQiTBqxj7+q/X33wS3j48CGZM1ttQ/7+/jx86Fi62szcuXOpW7eu5b0Qglq1alGiRAlmzZr1RmMxowqUD4Qd2/cSGmrNIiyl5OshowgOfh7vMcmTe1O/waeULFUEgH17D9Og3hfs2X3QVNnRUaVgMBg4cthxeVuiZGEuXN7Bzt2LWbJsosN5o6NjmDPbmgHg53HTKVGsPj+MnoTBqENviHGqvrh29SbXrt62i4K33W5GIHjxIhw0eod+qlYvHe89MDNtyjLCwyLjtFon8BhDBB171uP01WXMXTaSg/tP0PazL/m81SC2bNpHthwZ2Xt8FuN+70+vgS1Yt/NXmn1W3dJTpNPxw4XzV2lSvzerV25n88Z9dOkwjMm/LwIUu9qi1T/RsVtjMmZKQ8HCOfjsixpUqlYMA7EYbAIf3dxcadexATPmjyBrNmWizZMvKwuW/0iatCkJDHxOYGD834XEUrK0oz3K09ODpWvHvjSRISjC0sfX007tFfgshB4dRlOu6Bd8kvczfv95MUOG26+KUqbypd/gtpb3vfu3IZmndSUhpSTWGM3mTfu4du0W2XNm4rM2tS12sjx5sxHw+BGPHj0BjMTExHDr1j3+mLKAihWa0b/fcB4HKFU0jx87Q6OGHfj77wtERMT9XrxbFozeaVfaACAmKpYFo3fGc0TCcBaUHl8W4X379jF37lyLPQXg8OHDnDlzhm3btvHHH39w8ODBNxoPqALlg+HI0ZMObdHR0Zw9e9HJ3o6cOP43jRt1ZO/ewzwPCcWqPzcb7a3kzJnNaR8uLi6UKftJvIZz84R269Y9xvw4NU4mXGmzUrHinyUD2XNkRqt1lmzSSTVCg4HkNjaDylVL0rt/W4f9bLl86TpbNu1/6T4ATwMDadWiL6l9i9Ox3Vfs2P4XO7b9RYfPv2bBvLU8fvyUWXMW8cuvU6ldoz2DB/5kWVXVqlvBUh7AMnohuHHjjkP1uykTFxNrKsvsk9yLH8b3ZfqC4dy+f5OFC1exY9ce3Dwk7h5KbFIm/7TMmD+SjJnSUrdBRQ6fWcz1B5vZd3QehYvk4vNWgymQqy4FctXl81aDCXn+6gzIEeGRDBn4Czky1iRbuppUKPk5Xw/8leWLtxAcHMo3I7rY2aW8vJMxb/looqJiKFg4Jz4+nsSXNqZN+zoYDDb51KRigzETHRXDtEkr8fJOxua9U+g3uDXf/dCN3UdmkiuP9Ym6UJHc7Ng/my49mpM7bxY7AQtw6+Z9DEbJwqXj6dO/LWUrFDHdV8dxPQ54im1hO6M08PDhA8qWqUO2rMWZOXPhK+/Z2+LZw5BEtScUf39/UxC1woMHD8iYMaPDfufPn6dr165s2LDBLoLevG/atGlp2rQpJ06ceKPxgCpQPhjy58vj0KbRaMid23lJ0rjMmb3UNPk5Vmi09fGvW6+aZUUTH2XLFSdlKj+H9voNlCf248fOxpOt2H5idXFx4cvBXciYMS09erVxsr9j/RAXFxf2H1nMqvW/s/ev+axePxEvL+du1gA6XSwtm/XlxYuXT7IajWDThp38feaS1egujSYvOcmUiQvp0uEbzp9T1CSxsXoWzPuTmdOVVVnqNCmYv/QnsmVXqlCmTZeSiX8M5fnzUIdzhYS8cFjRDBrwE0FBISYVjp7QsGBCw58RrX9B8TK5qVPPatMRQuBpuuZh30xg53alJoqUkp3bDzF86G8vvVaAoYN/Z8mCTURF6tDpYrl98yFLFm5mcP9fKFmwJQGPnnLg1Hymz/+W36d/zeI1Y/nfoEnUrdKL+tX7kDtvZooUy+XQb7YcGWncvKqDcHXGlg0HKVIsN0OGd6Rbn+akTuPnsE/uPFn58ecB+GdN67SPDet206pFHyZPnM/8OaudptGJi3KvrKrTFy/CGNB/OBcvOHc/ftukyeSXqPaEUqpUKa5fv87t27fR6XSsWLHCUpDLzL1792jWrBmLFy8mTx7rHBMREWFJcR8REcHOnTspVCh+W11CUQXKB0K//t3w97d/uujbryuZM2eK5wh7XpiKCTlb8gohaNGyPj17t+POnTuk8M1LvTqfcyGeH1iyZB4sXjKRTJmU1Bmuri707tOeVq2VinN58sZnGxJIacTXz4fmLeuwY89CKldR1FU/jBnE8lWTaNuuMYUK58HT0xNXk0ecxPqE2+bzBjx+/JSChXJSsPCrvd8O7j/B44BnDisxrYuW7Nn9EUKQPYc/bdvVdQg6RZhdc/U8fPiEy5cc845t2rDH8rpq9dIcObOMC9c3cPrSGj5rW5dq1cs6HFOiZEG7bABhL8K5eOG6KZFn3CdsI6tXbnVaeE05v2Pw3uaNLw/oex78grWrdxJf+efoaB19u/9EsmTuNGhShRata/LdkCmWgmwAp09epkyFwly+9yfDvu9CvUYV+fKbz1m3YwLePp507tHkpWMA7OJdAgNDOH3yEhHhjuqnyMho0qZN6fQhJTaO96EQAmcltoUQloBOZ0GSUko2b34zFdPr0nFELdyTudq1uSdzpeOIWvEckTBcXFyYOnUqtWvXJn/+/Hz22WcULFiQGTNmMGOGUp199OjRBAUF0bt3b4oVK0bJkkoYwJMnT6hYsSJFixaldOnS1K9fnzp16rzReOA9enkJIVoCo4D8QGkp5Smbbf8DugAGoL+UcoepvQSwAEgGbAUGSCmlEMIdWASUAIKAVlLKO+/sYpKAjBnTc+LkLhYtWsWDBw+pWbMqdeok3M++adO6bNyww6knkdFoINYQw/Rp8y3qmf37j9Cwfjsu//OX00DLSpVLc/HKLq7+c4t06dOQymbFUqJEYZo0rcX6ddYfaO482Zk2/Ue8vL0oVMhxtQVQq04latWxT/Ny+tRF5s1ZTVDgc1Kn8WP16g3MnbsUNzdXvhzUjW9HDHjpdWttjNhGYk2BkYI2beoxaep36HSximvwgrUwP/5+PJK5YjDoHdRXcQuiCSHsMip/930f/rlyizOnlUjxLFkyMHHqt3bHeHl7kjZdKp48js8NW3Ljxl3yF3BcEfgk97JknLa2vdxdduhXEzAazbUpnfM4IJCwF+H4JPfmcUAgly7cdNhn787jfPdDD3r0a+6wrc+XrXjxIpz5MzeYPOjsU+y4uGhp30V5Wh7/01z+mLQMnS4WHx8vfhw/gM/aKJPXhnW7GTRgDKEhYU7Ha3TipJE3Xy6ePw8mMDAYg8GAt7cnP/w4hAyZ0jLmp8lcvXqdsDDHFWvKVCkc2t4FZm+upPbyAqhXrx716tWza+vZ01r+YM6cOU5rsOTIkYNz55K+4O57yzYshMiP8lg6E/jKLFCEEAWA5UBpICOwG8gjpTQIIU4AA4BjKAJlspRymxCiN1BEStlTCNEaaCqlbOV4VnveRrbh98n3oyYw8fdZxMTEWoSK8lRsQKPR4MTEwYCBXXn+PIR06dJQr34NHj9+Sr58ucib13Fys8VgMLB+3U4OHz5FrlzZ+KJd05dWozRz+9Y95s9fSUhIKI0b1+bTGoqAefTwCQXzV7fYLMx5nwoUzEOZMp8waHB3smd3rCSo1+spU7wZd+5Y3SG1Wi3bdy+geImClnuwY/tBunT8hqjIGLt7Y7taaNDwU7Zstj79CyFYvmoSNWu/urrdubP/EB0dQ8lShZzaixbOX8ug/j8hnWSHFkLDxas7nOal+v3XBYz5YbpdW4fOTfn196EO+wK8CA2nQI6GppTtzlcoZi7d2kiKFMmJCI+iWJ6WREfZC65yFYuyatOvlvc6XSw7thzhzq2HlK1QhFJlFRXJg/tP0Gg0zJ2xlv27T5E+Yyp6D2hFhcqfcHD/KVo1GeRw7t4D29CnXxuK5G/gIDBTpPQjXbpU1G1QhV9/nu5w7ISJ39K562eEhoZx88YdcufJjo+P9bsXHh7BJ8Wqce+edcWVMWN6zp7b5zTj9+ugpq//QNPXCyH2Yy9Q/gcgpRxrer8DZSVzB9gnpcxnam8DVJVS9jDvI6U8KoRwAR4DaeQrLu7fJlAAHj9+Rr685W3iLqxR2liy7gpTKV7nSfkAOnduw9Q/xiZp7emzZy9Rq0YrwsOtLqmjf/iar4b0YtHCNfTpNdw0Ysf0/WnSpuL4iS12JZHN3Ln9gO+G/85fB0+SPbs/X/+vB3XrKdlwg4NCaNa0B3+fUVYQLlotUmpM3VvP4e3jxYUr25k+dQlbt+wnRUpfevX5nDp1qzic73U5sP8EbT/rT1SUfeBnu/ZNmfTHSKfHBAQ8pUThxpYodInE2ycZJ/9e7/RePH/+goI5GtqsFpwLFX//dJy4YM2O8MN3M5k1dY3lvRCC2YtHUru+UiArIjyKlg0Hc+Gs1SW9VNlCLFr1E94+VieKpQs3M3PaKkKeh1GnXkU0Wlg4d4PD+Y0Yad+pEfPnrnbY1rBxdXLny4zRKLl5/S7r11pXwiVLFWbDljnxpi8yc//+Q37+eQp/n7lAkSIF+GZof7JlS7r0K6pA+XjS12dCWYGYeWBqizW9jttuPuY+gJRSL4QIBVIBgXE7F0J0B7oDZMni+MT7sZM+fRoGDerJ2LGTLW1SGi32BWWaMdXFeImsmDdvOfXq13jjyna2jP95qp0wUdr+oGev9qRL9/Kssc+eBrF0yZ9OM8Vmy+7P4mUTHNqllHTuNMQiTAD0BgNp0vg61Hnv3fcLfH19aNaiNn4pfciUKT2f1igft8tXcv3aba5fv02JEoVJl97+mqpULc22XQvo13skF85fxdMzGT17t+Xbkf3i6Q3W/7mTGJ39E3x4WAQb1u2iW4/WDvs/Dw6lYuXiHDxgflCSSAleXp5ERSr9JE/uxbS5I+yO+3Z0dwwGI/NnrcdoMCKlZPTwGRQsnAv/LOlYsWSbnTABOHnsIs3rfcm2A9PRaDSsW7OHIQN/NcXXSBYuWIdvcm9L6vy4nDpxyaENYOOGnej+VCpturq68O2o/hhiDeTKnY0GjT7Fzc3V6XG2ZM6cialTx71yP5Wk5a0KFCHEbiC9k03DpZSOjy2mw5y0xacQtj6Gxb/NvlHKWcAsUFYo8Yzho2bEyMEUKVqADeu3I4Rg6dK4T4HS7oZZVT/2T7P79x9JUoFy88Ydh7bw8AgeP35KjZoVKVw4H+fPX0ZKI8JJLMSzZ8GJOt+Qr35i754jDk/oz54FM2vOOLZvPUBUZDTNWtThs9b1mTp5AcP+N96yX5Ei+di8bQF+fq9Wk0RERNK541ds27ofUBwZfvhpCL372CdOLFI0HwcOryQ4KASf5F52ZQ0A9u49wpxZK3jxIpwmTWs5JHk0Y4zTfu3qHbp1+pYrl26iEVpc3Vzt0unUa1iJZi1qEhUVQ9XqpSxeZABRkdFs3nCQtSt32/V77+5jfh2zgIkzvrEkoTTnF1O+LRouX7zFwX2nqfppKZYt2myqAGntI/RFuBLXYjNcs0efX4rklC1XjGNHz1o3CtDFRmF2e4+N1bFy+UZOnH6zglgq74a3KlCklK8zGz0AbNem/sAjU7u/k3bbYx6YVF6+QOJmn38ZTZrUpUmTuqxaucGJQIGSpYpy69ZdgoKC7FwspdSg0Shfixw5knYFV75CKS5csI9ezpQpA9mzZ0Gr1fJJifycOavUn9dKx7rndeomPGL86ZNA5s1Z5XRbMk8P6jeoTqvWDSxtgc+CGTXyd7v9zp//h1kzl/H1Nz3jdmHHyhUb6d93BFFR1jiK2Fg9w4b+TP0G1cma1dFTz5lb9rat+2nTqp9FXfXXwROULFkErVZrV8Y2WTIPGjWtaXds987fceWSYlg3SqNDbrZVK7bTrWdLihTLa9f+6OEzmtUbyP17j52WSj55TImDKlQ0F6tX7LBL2miUeiSCo4fPUqV6SXS6WKdqVKPRiLeXpyW4VQm61ZM7jz9D/teNhfPWcvjQaXx8vFi/fgtxnwX/+ce5B5zKh8eH6Da8EWgthHAXQmQHcgMnpJQBQJgQoqxQZpr2wAabYzqYXrcA9r7KfvJfoWDBvE7bm7eoz1dDejlJdaKox3LmzMbnnzt697wJ/xvW3248Xl6e/DF9LFqtlmPHTjN3rrVGupKCRvkIPTzc+W7EQCpXLpPgcwUEPDUZ+KWDO2pUZCSLF62xa7tw4R+nCTJfVRcm4NETevUYRmSkY0JMo9HIoYPOg8UCHj2xRHWbmTJpgcNYT506jy7WmoWgcJG8LF89yc6Af+P6PS5ffPWke/GCY0qen0bO5N7dR4pa1Mmi3ss7GZGR0bT+oi6u7lZnA3MtegN6Jv62kMpl21Gleql4z+3qriVH7kzojTHEGsIxGKOZPWspn1ZpS6u29Vm3aToVKhVzKpDSpU/9ymtT+TB4bwJFCNFUCPEAKAdsMRnWkVJeAlYBl4HtQB9pnfV6AXOAG8BNYJupfS6QSghxAxgEOHeB+Q9SsFA+One2jzQvWFBpi+sia6Zq1XIc/GsDyZP7EBsby9atu9i0aTvR0Y6TZmJImzY1x05sYeOmhSxaPIXrN49Sq5Zi9D56xD5TgJRGDMYY2nVoxvWbh/lmaB9nXcZLgYK5SZcutakvgyXNv9Gox2g0MGzoz3YTet58OZ16ZxVw4sorpeTnsX+QM1sFCuavTmxs/IkMs8RZnQQGBtO4YWfy5KpM7pyVaN60uyU40lwPxxElUM/FRfDn+mlUqmw/cY/9cXo8gab2FI2zOvlt/ELWrtmpxPBgwCBjiZUx6GQUeqnDKA1cuHiVutW7E6vXkyKlVfUXt8ritat3OHH8HI2bVScuUkqCg4K4cuUyBqP9d+j+vUf8PEbx5CpQ0HncUceOLV95bf9Vtm/fTt68ecmVKxfjxjnajPbv34+vry/FihWjWLFijB49OsHHvg7vTaBIKddJKf2llO5SynRSyto2236SUuaUUuaVUm6zaT8lpSxk2tbXvAqRUkZLKVtKKXNJKUtLKW+9j2t6H+h0Or79djS5cxehSJEy/PHHTId9pk0fz8ZNS/jqqz7MmPkrhw5vJnlyH7Jm8XfSI7T9vDkpU/px/fpN8uUrRaNGrWna9HNy5SrOuXMJSwUTH1qtlho1K9OiZQM720SePDmd7l+lSllSpPBN9HlcXV2ZOXscvr4+KBOy0bQaUyZevV7P8ePW5HwZM6ZjwMBOdn1ky+ZPr96OxaOmT1vEjz9M5unTQEuKFWcroSpVylCpsn0esq8G/cDePdZa7jt3HGDoN2MBqB2PR5nZoK3TxXLrln1Sx4BHT9m0YS/2aiJpF+QppZEaNctSoJBVON66eZ9fx85TBLdUKicapM60MlTa9KYSAtf+ucP82WupVqO03TnicmDfKWbMHUWevFlsqm8a0RnCiTFExlvq9/RJZRVYqXIZGjS015KXKFGYr75+ucrxv0pC0tcDVKpUibNnz3L27FlGjBiRqGMTy4eo8lJJBAMGfM3Ysb9y69YdLl26Qv/+Q5g2zT5zqBCC2rWr8dOYYXTq1MbictmocR2KFLGvCZ83by5atFAi4r/66jvu3rXmCnr8+An9+n39Vq6jTt3qVKpsH3VesmRRmjatF88Rr6b6p+Xp279DvNtz57aP+B81ehBbty9k8Ffd+G3iCA4dXUvadPbqFqPRyJJFa+PpUVEXurm5MmLkAFavneGwx8YNuxzaNm1Q3GKH/q+3g5uyrXeUj48X+fLbr5iePXuuBGTKWAxGvSU7r/Jeh8EYg1HGsmPHASqVbc2d24qj5KnjFzEY9SabiLOaJPaT/6ULNxg2sjslnBQ5M5M+Q2qEEKz483fy5suK0agnxvDCLvbGmVrNNqBzybLJLF0+lS8Hd2P23PHs2O28UufHxu2t51hX7zeWFB/Junq/cXvrmwcV2qavd3Nzs6Svf9vHvowP0W1YJYFERUWxcOFSh/YZM+bSu7eje21c3Nzc2L1nLTNnLODMmfMULVqQHj074umpxBUcPHjY4ZgjR46j1+sdqiK+KVqtls2bl7B8+TpOnzpHkSIF+PyLFri7J7zMq5SSY0dPExERSaXKZXB3d2fblr3KtjixLTVrVqFAQceI/oqVSlGxkqMtIDw8giGDf2DVqk3ExuptdP3WCb9smeIULJSHAV92jdehwdfPh8A43mrmYDtvb09Wrv6Du3cf8vDBYwYP/JF/riiGdo1Gw0/jhuDjY01nYjAYmPT7XIzSrHLTI3FBIxyrWAL8c+UWgwaMYe3GaYSEvsAo9Xblo+NiTvpoRJI7b1ZSpfJj085pXLxwnS0bD/DbLwvs9h84WMkunCq1Hz36NGfx4nX2HlxYBYpZUPr5JWfI0B6W7VqtlkaNa9Kosb3TwcfM7a3nOPbDJgzRyucUERDKsR8Ur7Xs9Yq+dr/O0tcfP37cYb+jR49StGhRMmbMyK+//krBggUTfGxiUQXKR4zBYHCqv3fIWfUSfH2T8/U3/Z1uy5UrB3//bV8EKVu2LEkuTMy4u7vTsWNrOnZ0jK94FU+fBtKoYQfOn1OW7WnTpmLm7F8JCQ21qn9snMYioxKX0nzI4B9YvPhPJ1tMLrB+vqzdMJvkyX2c7GOld58OjB5l703Wu6/9Kipr1kxkzZqJv46uZuf2v3j8+Bk1alZwsMesXrmFtWu2249G6kmdJhWBgSGm99Ii/AQaDu4/SUjIC3792VrfPn4ksUblPk2ZPJdixfNQu25lChXOTaHCuSlTrgjLlmzBaDDwWZu6aLRGWjbvzuHDp5RSBAi0GseYkRw5s5I9uz/lyhWnXcdmFlvXv5WzU/dYhIkZQ3QsZ6fueSOBkpD09cWLF+fu3bt4e3uzdetWmjRpwvXr1xOV+j4xqALlI8bb25v69euwadNWu/bWrZ1XekwsI0cOpXnzdhaXVSEEo0cPS5K+k5rvR02wCBOAp0+DaNGsm8nxwDFa/PChk9y8eZecObPyKgwGAytXbnSaJ81MSEgoU6csYNjw+IMUAYZ83ZMUfslZtmwDQgjatW9Ox07Ojc4uLi7UaxC/q/TBA849yFq2rsv0qcuR0ohe6rBdqaTyS82li9cJDg7FHOAaHwajNaAyMjKKAX2/5/yV7ZbAwqrVS1tq1WzetJs2zXvHmagUW47tKsjXLznrNs506kr9byXisWNG6pe1J5SEpK9Pntxqp6xXrx69e/cmMDAwwanvE4sqUD5y5sz5gy5derN16w5cXV1p374tI0YkjZNbgwa1OXp0F4sWLcdgMNC27WeUKxe/a+j7ZP++I3FaRLxebGYi4kTtA6xfv53Zs5YRFRlF8xb1qF+/Bh07DiQmRplclQnTeTqTw4cda9qAImy8vDwtQYxdu7ela/eX13hJCPFNys1a1CZP3hx8NehH9DH2T6KePm5kyGBOFW823luvx5z7zWCMdcik8PRpENev3aagk+SfkybNdfrUa5R6hNSSM2dWKlUpQ//+Hf9TwgTAK70vEQGOwsMrfeKdTWyxTV+fKVMmVqxYwbJly+z2efz4MenSpUMIwYkTJzAajaRKlQo/P79XHvs6qALlIyd16lRs2LCSsLAwXFxcSJbs5TmOEkvx4kUpXvz1l+WvYv/+QyxevAIhBO3bt6Fy5cSnOwHImtWf27edl7V1lmghY6b0FCiYx27VsXLFBrp0/sqyz4kTZ5k0cS4PHwa8sj+AXLmy2b2/fPkaPXoM4dTJs/j5JWfQoJ4M+Tpx7s8vo1OXz1g4/08CbNyf69SrwifFC5Lc1wtXNy1x8i5y/94j/DOnp1GTT9m4fo9FHebh4U7BIjk4elQJLBVoHNLEu7u7WUoaxCUw8GVxxEZmzB5D2bLFE3+R/wKK9f3UzoYCoPVwpVjfhGcTd4Zt+nqDwUDnzp0t6etByTq8Zs0apk+fbpkbVqxQfmvxHfumvPfkkO+Tf2NyyI+JJUtW0rFjb8t7IQSLF8+kdevEB1Tu33eExo062tiU4q4ihOUcXl6elCxZhL/+Ok7y5N706t2BYcP7U6VSc86csQ9kjJuRGBTDcalSxezcjlOmSsH+A6vJkUNRoRkMBgoVrMKdO/ftjl2ydBrNm9dP9PXFx+PHz5g7ayW3bt2jYqVSlCpTmC6dvuTChX9w0SZzMLpnzpyBi1d2odPFMnfWavbuPkqGjGno0bsNo0b+xs4df1n21aC1u4eDh3Rl+Ii+Tscx7H/jmDxprkO7QEPnLq2YNGW0k6M+XhKbHPL21nOcnbqHiMeheKX3pVjfT9/IfvIu+aiyDb9PVIHyfsmTpwS3bt2xa8ubNxeXLr2et8nff19k/rwVRIRHUK5CScb8NMVSg8RWMKRJm4pnT4PtJsvffh/FlMnz7NLgm4lbQrlEiSIcPLSOrVv2sGPHAdKnT0OHDi3JaPP0fvToKapXcxSMjRrXZuXKWQ7ttgQGBvPll9+xccMOUqTwpV+/rgz+qpfDftHR0Tx9GoS/fwZLHfgypetx4bxSOE0ILVqNNYWNEILpM3+kTdvGTs+7Yf0u2n/xpV1bhgzpqFWrMvXqV6VOvfgzL4eHR9Cx/UC2b98PKAGdX3zejLr1qvNJ8TevBPihoWYbVgWKA6pAeb+4u6e1y1GltLkTEfEoniMSR1hYOL/+Mp1fxk9z2GayhFgm21KlilGufEmmTJ5nt1+GDGl59OixXdu8+b/TqrV9qdW4nD17kXJlrSsRs0ArW7YE6zcsxM8vfv15vbpt2bv3L7u2adPHkzlzRiZPnk1QYDCpUqXg+PHThIaGkS17FqZPH0/OHNnJn69SnN40pEuXjiZN6tCqdQPKlnu52mnO7BVM+n0eT54EUrtOFX6ZMIz06V+eCdqWu3cfEB4eSUEnLtn/JlSB8vGkr1f5j1C9emV27drn0JZU+Ph4O0TZ28ajSEBKpaSsi4uWb78bwN27D9i0cRdSSkqWLML8hb+zft12/vxzC95eXnTv8QXNmr862LJYsUKUKv0JJ0/8bapxrqjijh49Tq6cJflz7UKqVHG0Fz18GOAgTACmTJ7NP/9cd+pocOf2PVq26MKp07txcXGxFClTMFKwYA5+nzTC4ThndO3Wmq7dEu+2bSZrVufZF1T+G6iR8iqvZP/+g0yaNJXDh+N6Ur0Zkyf/TI4c2Szvc+XKwcSJY5P0HPnyx80P5RgVLjHSodNneHl5smz5H1y/cYiLl/ey/+CfZM+ehS8HdefQ4Q1s37ksQcLEzJ9/zqVjx1Z4eNjHYoSFhdO/X+I88R4/fvpSr7Xw8AhOnzpL28+b2bULIejTr3OizqWi8rqoK5SPCCklUVFRlkj2t43RaKR16y9Ys2adpa1Dhy+YP392kvSfO3dOrlw5wcGDSs2SSpXKodVqCQ19wY4de/Dx8aZmzWqvHUgZExPD3DlLLLEQztJ+ALi7u9KunTV2J73FrfbNSJMmFdNnjGfnzj08eGAfSHnlyjV+/nki+/b9Rbp0aejXrwclS35CpkwZ+PTTSuzZY79KSZHCl+Dg5y89X3Lf5EyZ+iO5c2dn/fptpPDzpXffTtSpk/C0/yoqb4SyHP9v/pUoUUJ+LKxYsUpmz55bCuEmP/mklDx8+MhbP+fmzVulEB4Of/v3H3xr59y796BMkSKbdHFJLV1cUssCBcrIBw8evVZfv/02Xbq5ZpRurhmlu6u/dHfLLD3cszj8ubtllqtWrk/iK7FSu3ZL6eaa3u7Px9tfarUpLX/JkqWXx4+fklJKGRgYLDt06Cf9fHPJHNlLyt9/myF//XWa5Vqc/RUpXEXq9fq3dg0q9ly+fPl9D+Gd4exagVPSyZyqqrw+As6ePcfnn7fnzp27lvf16zcmNPTNIm1fxYkTzh0WTpxwHsD3pkgp6dVrEGFh4Za2a9duMmpUwlJrG41GJkyYRskSn1K2TC3mz7cGakmMlhT2juc1snCR82JcScHIkUNIlszDri0q2j49jk6nY+JEJY17qlQpWLBgMs9DrnPz1kkGftmD/v270rFja0uK/WzZslClSjny5MlJl66fs33HSqfp91X+3bwqBf0vv/xiSV1fqFAhtFotwcFKzFC2bNkoXLgwxYoVo2RJB/v6a6GqvD4Cli9f6aA/Dw0NZdOmLXzxxZtHXMdHoULOM8sWKvTmAVDOePToMTdv3nFoP3ToKOHh4URERJIuXfzqqJEjxjF+/JQ4rQL7IESJ0WiwxGcoAib+lCpJQblypTh9Zh+LF60kMiqKIkUK0LGjoxtwQMCTePtwdXVl5qwJ/DRmOM+fh5A7d463Nl6VjwNzCvpdu3bh7+9PqVKlaNSoEQUKWH+3Q4YMYciQIQBs2rSJ33//nZQpU1q279u3j9Spky6XmrpC+Qhwc3Oevju+9qSiSZNGVKli74Zat25tatd+O5lgU6dOSYoUfg7tOl0s6dPnw9+/IJUq1XOIXQFldTNjxgKH9rhy4vPPm1OmzCeWoltmI32H9p+9+QW8hJw5szHq+28YP34Ubdo0x9/fMW9SvXqvvq+pU6dUhclHSOzxw0QOG0hEz3ZEDhtI7HHHTN6JJbEp6JcvX06bNm3e+LwvQxUoHwHt23+Oh4e9yiRDhgw0bJh0EdfOcHV1ZefOLSxduoCvvx7MypVL2LBhjSWILqlxd3dnxAj7eiuurq7cu/fAkkvr2LGTtG3bzeFYg8HgNMuyr29yunb7gvr1azJl6lhmzprAqtVzad68AR4eHmTJkomJk36iRcuXx5UkJVqtlqVL55AhQzpLW/Pmjejfv8dLjlL5WIk9fhjd0nnIYKUipwwOQrd03hsLFWcp6B8+fOh038jISLZv307z5tZgWyEEtWrVokSJEsya9fJg24Siqrw+AnLnzs3WrRv57rtRXLnyD+XLl+Xnn8e8Vt6u8PBwbt26Ta5cORPkLebq6kqbNq1o06bV6ww90fTt242iRQuyZs1G3N3dWLNmI/fv2/9ITp8+y+3bd8meXUlzEhMTw5o1G8mVKxv//GNfW/2zVk2YMsVet5w+fVqWLXesbPkuqVChDLdunePMmXOkSZPazn3aGTExMYwa+SOrVq3B09OTnj270qevo9pM5cMjdsNq0OnsG3U6YjesxrVMhdfuVyYiBf2mTZuoUKGCnbrr8OHDZMyYkadPn1KzZk3y5ctH5cpvFgemCpSPhKpVq/DXX/teveNLmDp1GsOHjyAsLAw/Pz8mTBhPp07xVzR8X1SqVB5fX19q1WpKYGCww49Eo9Hg5aUIw4iICKpXb8yZM0oFPIEGITRotVqaNq3P2LHfvfPxJxRXV1fKlEmYMbRvn4HMm7fQ8r5//8FoNBp69VZXNR865pVJQtsTSmJS0K9YscJB3WXeN23atDRt2pQTJ068sUBRVV7/Ef7++yz9+39JWFgYACEhIXTr1pMbN2684sj3w9dfjyAwMAhntctbtGhE2rRKOpCFC1dYhAko3lxGqWfHztUsXTYTb28vh+M/NiIiIliyZLlD+8yZjskYVT48RMpUiWpPKLbp63U6HStWrKBRI0fVbWhoKAcOHKBxY2sOt4iICMtcEBERwc6dOylU6M1zrr03gSKEaCmEuCSEMAohStq0ZxNCRAkhzpr+ZthsKyGEuCCEuCGEmCxMj65CCHchxEpT+3EhRLb3cEkfNBs3bnZoMxqNbNq05Y36lVKii7ucTwKOH7e6LEtptPi5f/VVX+bMmWzZdv78JafH371732n7x0h8lTmjohJemVPl/eHauCXEdaBxc1Pa3wDbFPT58+fns88+s6SvN6ewB1i3bh21atXCy8v6cPXkyRMqVqxI0aJFKV26NPXr16dOnTpvNB7g/QU2AvmBvMB+oKRNezbgYjzHnADKofiBbgPqmtp7AzNMr1sDKxMyho8psPFNmTp1mhTCzeFv8eKlr93ntD9myIwZskmNSCarVa0lr1z5J8nGW758LbvAP602pcyfv5SUUspbt27Le/fuSymlnD17kSUI0vbv8uWrSTaWD4GGDZtJjUhm9zdyxOj3Paz/LIkNbNQdOyQj/jdAhvf4Qkb8b4DUHTv0lkaW9CQmsPG9R6snVKAAGYB/bN63AWaaXu8AypleuwCBmDIpv+zvvyRQgoODZcaMWe2ESfbsuWVERMRr9bdlyzaHCS5XzgKvFa2t0+nkhQuX5PPnzy1te/celJ6eGSzCxN09rZw7d5GsUKGm1Gp9pVbrK+vXbyEDAh7LqlUb2AmTQYOGv9Y1fcg8ffpUNm7cQrpovaRnshSyT+8BUqfTve9h/WdRI+WdC5QP1SifXQjxN/AC+FZK+ReQCbAtVvHA1Ibp//sAUkq9ECIUSIUiWFSAFClScOjQPsaOHc/58xcoUaI4w4Z989p5wZYuXeHQduvWbY4cOUalSgn3XNm8eRvdu/fjyZOneHh4MGTIAEaNGk61apU4d+4QS5euxmAw0Lp1M775ZgTHjlnrqG/fvotRo8awe/d6tm7dxfXrN6lQoSxlypR4rWv6kEmTJg3r168mPDwcFxcXBzdyFZUPgbcqUIQQuwFnNUOHSynji8AJALJIKYOEECWA9UKIgjiruWq12L5sW9wxdQe6A2TJkuVlw//XkT17dmbNmp4kfcWXsNFcNz0hPH/+nDZtOhEZqSROjI6O5ocffqZ06ZLUq1ebXLlyMHLkN4CSmmT79l0OfWzYsIUZMybRsGES6H8/Ary9vd/3EFRU4uWtGuWllDWklIWc/MUbzimljJFSBplenwZuAnlQViS2xRb8AXMlpgdAZgAhhAvgCzgtci2lnCWlLCmlLJkmTcILB6nY07lzBwd33sKFC1GmTKkE97Fr1z6LMLFlwwZHRwEXFxd8fZM7tKdO/WaeMioqKknHB+c2LIRII4TQml7nAHIDt6SUAUCYEKKsyburPWAWTBsBc0BFC2CvSc+n8paoUqUSK1YuoWjRwqRIkYKWnzVny9Z1icqJFZ8wSJPGMbeQRqOhX7+eDu0DBvR2aFNRUXk/vDcbihCiKTAFSANsEUKclVLWBioDo4UQesAA9JRSmlcbvYAFQDIUL69tpva5wGIhxA2Ulcnrl5xTSTAtWjSlRYumr3181aqVKF68GGfOnLW0JU+enC5d2jvdf8SIoWTIkJ6lS1fh6upCt24dadXKsW67iorKe8KZpf6/8vdf8vL6UAkKCpJDhgyXZcpUkZ9/3lmeP3/xfQ9JReWVfCheXp06dZJp0qSRBQsWdLrdaDTKfv36yZw5c8rChQvL06dPW7Zt27ZN5smTR+bMmVOOHTs23nP8G7y8VP4jpEyZkvHjf0z0cTdu3OTWrTuULl0CPz+/1z5/aGgoc+cs4NKly5QuU4oOHb5QPahUPho6duxI3759ad/e+ap+27ZtXL9+nevXr3P8+HF69erF8ePHE5T6/nVQBYrKR4XRaKRr194sWrQUKSWenp5Mnz6ZL75IvJYzIiKCShU/5dKlywAsWLCYVSvXsHvPtrdaH0Xlv4cMuQZPjkFsOLh6Q7qyCL88b9xv5cqVuXPnTrzbN2zYQPv27RFCULZsWUJCQggICODOnTuW1PeAJfX9mwqUD84or6LyMpYtW8XChUvMwa1ERkbSvXsfAgPtQ46OHj1O3bpNyJWrEB06dOP+/QdO+lppESZm9u8/yM6du9/eBaj855Ah1+DhfkWYgPL/w/1K+1smvhT3iUl9nxhUgfIvYsuWrfTo0Ztvvx3J3bt33/dw3gq7du1xaIuJieHgQWttievXb1CzZgN27tzN7dt3WLJkOTVq1HfIh3Xj+k2n57hx/cNMmKnykfLkGEi9fZvUK+1vGfODly1CiHjb3xRVoPxL+O67UTRs2JTZs+cyZsw4ihUr5fD0/W8gc+ZMTtuzZLE+bS1YsMQhceKNGzfZscN+5VGpsvOI/spxqlSqqLwR5pVJQtuTkPhS3Ccm9X1iUAXKv4Dnz5/z66+/2bWFhobyyy8T3tOI3h49e3ZziFOpU6cWJUsWt7wPD3f+Q43bXr9+XZo0cUz3vXzZqiQYqYqKCdd4shvE156ENGrUiEWLFiGl5NixY/j6+pIhQ4YEp75PLKpR/l/AvXv3LSVybbl27fp7GM3bxd8/E8ePH2Tq1BncuHGLatUq0b17F7t9WrZsxtSpM+zafHx8qFu3ll2bEAJvH8cf9cSJUxj81QBSpVKj8FWSgHRlFRuKrdpLuCjtb0ibNm3Yv38/gYGB+Pv78/3331tUuz179qRevXps3bqVXLly4enpyfz58wH71PcGg4HOnTtTsGDBNx6PKlD+BeTLl5c0adLw7Nkzu/ZKlSq+pxG9XbJmzcIvv4yJd3vFiuWZOvV3Ro36kcDAIHLnzsX06ZPw9fV12PfmDUc7SkxMDPfvP1AFikqSIPzyKIkF34KX1/LljoXX7M4tBH/88YfTbfXq1aNevXpvPAZbVIHyL8Dd3Z1Zs6bRtm17i+2gRIniDB065D2P7PU4cOAvli5ZjlarpUPHdpQtWzpRx8fExJAxY3qmTJlAqZIlyZY9a7wGx8qVK3H06HG7trRp01KgQP7XHr+KSlyEXx5IAgHyweMs2vG/8vdvi5QPCgqSK1asknv27JVGo/F9D+e1mD9/kV2NFRetl/zzz3UJPv7q1WsyS+ZcluPTpsksT506He/+ISEhslzZypb9fbxTy40bNyfBlaj8m/lQIuXfBYmJlFeN8h85Fy9eokmT5mTJkpNOnbpSoEB+qlev9s4D88LDwzl8+AgBAQFv1M/3o+yj5o1GIz+MHpvg4wcP/oYHD6z+9IGBgfTr+2W8+/v6+nL4yH727d/JqtXLuHf/Og0b1k/8wFVUVFSV18dMcHAw1arVJCgoCFBc/w4fPsrVqxffqf5/yZLl9OkzgLCwMFxcXOjfvw+//jou0f0YDAbu3XOsBX/r1u0E93H40FGHtuPHT6LT6XCLW9fbhBCCypX/nfYmFZV3ibpC+YhZuXK1RZiYCQ4OZsUKR7fXR48ecfToMYf4jDfl8ePHdOnSg7CwMAD0ej2//TaJTZsca5q8Cq1W63Rir1a9SoL7yJMnl0NbtmxZ4xUmKioqSYcqUD5inBWnitsupaRv34FkzZqHChWq4e+fk/XrNybZGPbs2ecQgQ6wdeuO1+pv6h8T8fe3Bi/mzJmD334bn+DjR33/nV3VSCEEP/w46rXGoqKikjhUgfIR06xZE4eSuy4uLjRvbq1RsnLlaqZNm4nBYACUIMh27ToTGhqaJGPIkCGD0/aMGZ23v4qCBQtw4+Zltm3fyM5dW7jyzzly5Mie4OPr1KnFyVOHGTRoAP369ebosQO0bdvqtcaiovKh07lzZ9KmTUuhQoWcbl+6dClFihShSJEilC9fnnPnzlm2ZcuWjcKFC1OsWDFKliyZNANyZqn/r/z9G7y81q5dL7Nlyy2FcJNZs+Zy8Ihq166TFMLD4W/9+o1Jcn6j0SjLlq1k13fatJnlo0ePkqR/FZUPkQ/Fy+vAgQPy9OnT8dZDOXz4sAwODpZSSrl161ZZunRpy7asWbPKZ8+evfIcaj2U/xBNmzamceOGBAcHkzJlSjQa+0VnunTpnB6XIUP6JDm/EIIdOzYzceJUDh06TJ48uRk8eGC8KxcVlf8iRuMTJLeAGMAdQQ40Gue/zcTwqvT15cuXt7wuW7YsDx44Zt1OSlSB8i9Ao9GQOrVjHXaAXr26M2fOfDsVV7VqVShdulSSnT958uSMGDEs3u337t3j2LET5M+fj8KFnS/NVVT+rSjC5CpgNLXEILmK0UiSCJWEMnfuXOrWrWt5L4SgVq1aCCHo0aMH3bt3f+NzqALlX06OHNk5evQAEyZM4vbt21SrVpWBA/u+s/OPG/cL33030mLD+eKLtixYMNdhJaWi8m9FWZkY47QaTe3vRqDs27ePuXPncujQIUvb4cOHyZgxI0+fPqVmzZrky5ePypUrv9F5VIHyHyBfvrzMnj3tnZ/3xo0bDB/+HYrKVWHJkmU0btzIznFAReXfjWPi1pe3Jy3nz5+na9eubNu2zS4+zZyuPm3atDRt2pQTJ068sUBRHxNVHDh8+Ajly1fGw8OHcuUq8ddfh159kBMOHjxkJ0zM7N9/4E2HqKLyEeGeyPak4969ezRr1ozFixeTJ481l1hERIQldiwiIoKdO3fG6ymWGN7bCkUI8QvQENABN4FOUsoQ07b/AV0AA9BfSrnD1F4CWAAkA7YCA6SUUgjhDiwCSgBBQCsp5Z13eT3/Fh4/fkzdug0ttUOOHz9BvXqNuHr1YqIL8OTMmcNpe65cOd94nCoqHwuCHHFsKAAaBM5/H4nhVenrR48eTVBQEL179waUsIJTp07x5MkTmjZVtAR6vZ62bdtSp06dNx7Pe3PZBWoBLqbXPwM/m14XAM6hiO/sKMJGa9p2AigHCGAbUNfU3huYYXrdGliZkDH8G9yGk5rJk6dKIdwc/n77beJr9Ve7dj27fnLnzi9DQkKSeNQqKu+WxLoNGwyPpd5wROoN+6TecEQaDI/f0siSno/CbVhKudPm7TGghel1Y2CFlDIGuC2EuAGUFkLcAZJLKY8CCCEWAU1QBEtjYJTp+DXAVCGEMF24SiKI75a97q3csGEtixYt4ciRo+TNm5fu3bs4rUuiovJvRvHmenceXe+LD8WG0hlFMABkAmwzBD4wtWUyvY7bbneMlFIPhAJOsyMKIboLIU4JIU7FLUilAi1bNsfT09OuLVmyZHz2WYt4jng57u7udOvWhfnz5zB06BBSpkyZFMNUUVH5AHmrAkUIsVsIcdHJX2ObfYYDemCpuclJV/Il7S87xrFRyllSypJSypJp0qRJ+MX8R8iQIQNbtmygZMkSaDQaihf/hM2b1+Pv7/++h6aiovKB81ZVXlLKGi/bLoToADQAPrVRTz0AMtvs5g88MrX7O2m3PeaBEMIF8AWC3/gC/qNUqVKZEyeOvO9hqKh80Egp33ndoXdNYlXd703lJYSoA3wDNJJS2qbN3Qi0FkK4CyGyA7mBE1LKACBMCFFWKJ9ie2CDzTEdTK9bAHtV+4mKisrbwsPDg6CgoNe2LX4MSCkJCgrCw8Mjwce8z8DGqSieXLtMUv6YlLKnlPKSEGIVcBlFFdZHSmkwHdMLq9vwNqx2l7nAYpMBPxjF00tFRUXlreDv78+DBw/4t9thPTw8EqXuFv9mCfsqSpYsKU+dOvW+h6GioqLyUSGEOC2ldMh5/6F4eamoqKiofOSoAkVFRUVFJUlQBYqKioqKSpLwn7ahCCGeAXdf8/DUQGASDiepUMeVONRxJQ51XInnQx3bm4wrq5TSIZDvPy1Q3gQhxClnRqn3jTquxKGOK3Go40o8H+rY3sa4VJWXioqKikqSoAoUFRUVFZUkQRUor8+s9z2AeFDHlTjUcSUOdVyJ50MdW5KPS7WhqKioqKgkCeoKRUVFRUUlSVAFymsghKgjhLgqhLghhBj6Ds+bWQixTwhxRQhxSQgxwNQ+SgjxUAhx1vRXz+aY/5nGeVUIUfstj++OEOKCaQynTG0phRC7hBDXTf+neJdjE0LktbkvZ4UQL4QQA9/HPRNCzBNCPBVCXLRpS/T9EUKUMN3nG0KIyeINU97GM65fhBD/CCHOCyHWCSH8TO3ZhBBRNvdtxjseV6I/t3c0rpU2Y7ojhDhran+X9yu++eHdfceclXFU/15auliLUpY4B+CGUq64wDs6dwaguOm1D3ANpWTyKOArJ/vHW075LY3vDpA6Ttt4YKjp9VASUOr5LX92j4Gs7+OeAZWB4sDFN7k/xFMKO4nHFV+J7my2+8Xp512MK9Gf27sYV5ztE4AR7+F+xTc/vLPvmLpCSTylgRtSyltSSh2wAqUE8VtHShkgpTxjeh0GXMFatdIZlnLKUsrbwA2U8b9LGgMLTa8XopRtfl9j+xS4KaV8WTDrWxuXlPIgjnV6EnV/hBAZMJXClsovf5HNMUk2LinlTqlUPwWlRPdLU86+q3G9hPd6v8yYnuQ/A5a/rI+3NK745od39h1TBUriia9E8TtFCJEN+AQ4bmrqa1JPzLNZ0r7rsUpgpxDitBCiu6ktnVRq2WD6P+17GhsoZQ1sf+gfwj1L7P15WSnst4VtiW6A7EKIv4UQB4QQlUxt73Jcifnc3vX9qgQ8kVJet2l75/crzvzwzr5jqkBJPAkuN/zWBiCEN/AnMFBK+QKYDuQEigEBKEtuePdjrSClLA7UBfoIISq/ZN93OjYhhBvQCFhtavpQ7ll8vE4p7KQfhGOJ7gAgi5TyE2AQsEwIkfwdjiuxn9u7/jzbYP/Q8s7vl5P5Id5d4xnDa49NFSiJJ74Sxe8EIYQrypdlqZRyLYCU8omU0iClNAKzsapo3ulYpZSPTP8/BdaZxvHEtIQ2L/Ofvo+xoQi5M1LKJ6YxfhD3jMTfn5eVwk5ShLVE9+cm1Qcm9UiQ6fVpFL17nnc1rtf43N7l/XIBmgErbcb7Tu+Xs/mBd/gdUwVK4jkJ5BZCZDc99bZGKUH81jHpZ+cCV6SUv9m0Z7DZrSlg9j5xWk75LY3NSwjhY36NYtS9iH155g7Yl21+J2MzYffk+CHcM5vzJfj+yJeXwk4yRDwluoUQaYQQWtPrHKZx3XqH40rU5/auxmWiBvCPlNKiLnqX9yu++YF3+R17E6+C/+ofUA/Fg+ImMPwdnrciytLzPHDW9FcPWAxcMLVvBDLYHDPcNM6rvKEXySvGlgPFY+QccMl8X4BUwB7guun/lO9hbJ5AEOBr0/bO7xmKQAsAYlGeAru8zv0BSqJMpDdRSmmLtzCuGyj6dfP3bIZp3+amz/cccAZo+I7HlejP7V2My9S+AOgZZ993eb/imx/e2XdMjZRXUVFRUUkSVJWXioqKikqSoAoUFRUVFZUkQRUoKioqKipJgipQVFRUVFSSBFWgqKioqKgkCapAUVFRUVFJElSBoqKSAExpyC++es/3gxCioxBi6ku2NxFCFLB5v18IUfLdjE7lv4IqUFRUPmDMUdZJQBOUdOUqKm8NVaCoqCQcFyHEQlOm2zVCCE+hFFMaI4Q4KoQ4JYQoLoTYIYS4KYToGV9HQoiqQoiDQiledVkIMUMIoTFtCxdCjBZCHAfKCSG+EEKcEEqBppk2qTw6CSGuCSEOABVecq7yKIkxfzH1kdO06QshxBEhxEUhxLsua6DyL0QVKCoqCScvMEtKWQR4AfQ2td+XUpYD/kJJv9ECKAuMfkV/pYHBQGGUDLrNTO1eKEWZyqCkjGmFksm5GGAAPjfltPoeRZDU5CWrDynlEZQ0JUOklMWklDfN55FSljddx7yE3AAVlZehChQVlYRzX0p52PR6CUruJLAmB70AHJdShkkpnwHRwlQ6Nx5OSKVQmwElP5S5PwNKxlhQioKVAE4Kpazspyh508oA+6WUz6RS6G0liWc5WApGJX/FWFVUXonL+x6AispHRNzEd+b3Mab/jTavze9f9huLr79ok5ABpTbFQinl/2x3FEI0cXJ8Yonv/Coqr4W6QlFRSThZhBDlTK/bAIfesL/SpjIIGhS1lrP+9gAthBBpAYQQKYUQWVEq8VUVQqQy1cBo+YpzhaHUGbellanPikColDL0Da5FRUUVKCoqieAK0EEIcR5IiVI98E04CoxDSRN+G6UomR1SysvAtyillc8Du1BStgcAo0x97EZJjf4yVgBDhFKK1myUfy6EOALMQEkNr6LyRqjp61VU3gNCiKrAV1LKBu95KCoqSYa6QlFRUVFRSRLUFYqKyltECFEYpcqgLTEml+C3cb7hONpTVkspf3ob51NRsUUVKCoqKioqSYKq8lJRUVFRSRJUgaKioqKikiSoAkVFRUVFJUlQBYqKioqKSpKgChQVFRUVlSTh/681Erc1dZDEAAAAAElFTkSuQmCC\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "xy_bm = np.vstack([df_out_all['bm_pred_tb'],\n",
    "                   df_out_all['bm_resid']])\n",
    "df_out_all['z_bm_tb'] = gaussian_kde(xy_bm)(xy_bm)\n",
    "ax = plt.subplot()\n",
    "sns.scatterplot(x='bm_pred_tb', y='bm_resid', data=df_out_all, hue='z_bm_tb', \n",
    "                hue_norm=(1.1e-10, 1.1e-6), palette='magma', edgecolor='none', ax=ax)\n",
    "sns.regplot(x='bm_pred_tb', y='bm_resid', data=df_out_all, ci=None, marker='', order=5,\n",
    "           line_kws={'linestyle': 'dashed', 'color': 'black', 'linewidth': 2}, ax=ax)\n",
    "ax.plot([0, 2000], [0, 0], linewidth=2, c='black', linestyle='solid')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "id": "2fabb6d1-270d-492f-9ae1-b74205f8eced",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.03840631730078966"
      ]
     },
     "execution_count": 19,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_out_all[df_out_all['bm'] > 2500]['bm'].count() / df_out_all['bm'].count()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "f0a3b978-26f7-4778-b687-5b4608101ec2",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[<matplotlib.lines.Line2D at 0x7f3cf7d36a40>]"
      ]
     },
     "execution_count": 16,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 427.125x360 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "sns.lmplot(x='bm', y='bm_pred_tb', hue='Project', data=df_out_all, ci=None, order=1,\n",
    "           line_kws={'linewidth': 2, 'linestyle': 'dashed'},\n",
    "           scatter_kws={'alpha': 0.2})\n",
    "plt.plot([0, 6000], [0, 6000], linewidth=2, c='black', linestyle='solid')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "453d7dc0-0f36-40e4-a0a5-289d52072394",
   "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>index</th>\n",
       "      <th>ID</th>\n",
       "      <th>Date</th>\n",
       "      <th>Project</th>\n",
       "      <th>gps_E</th>\n",
       "      <th>gps_N</th>\n",
       "      <th>Method</th>\n",
       "      <th>bm_gm2</th>\n",
       "      <th>dfi</th>\n",
       "      <th>ndvi</th>\n",
       "      <th>...</th>\n",
       "      <th>swir1</th>\n",
       "      <th>swir2</th>\n",
       "      <th>bai_126</th>\n",
       "      <th>bai_136</th>\n",
       "      <th>bai_236</th>\n",
       "      <th>bai_246</th>\n",
       "      <th>bm</th>\n",
       "      <th>Year</th>\n",
       "      <th>bm_pred_tb</th>\n",
       "      <th>z_bm_tb</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>0 rows × 24 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "Empty DataFrame\n",
       "Columns: [index, ID, Date, Project, gps_E, gps_N, Method, bm_gm2, dfi, ndvi, ndti, satvi, ndii7, nir, swir1, swir2, bai_126, bai_136, bai_236, bai_246, bm, Year, bm_pred_tb, z_bm_tb]\n",
       "Index: []\n",
       "\n",
       "[0 rows x 24 columns]"
      ]
     },
     "execution_count": 15,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_out_all[df_out_all['bm_pred_tb'] > 10000]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 35,
   "id": "2bda63fc-2a87-4b7a-a786-e3b6f30cef33",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<Figure size 432x288 with 0 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 720x720 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.figure()\n",
    "plt.figure(figsize=(10,10))\n",
    "plt.xticks(fontsize=12)\n",
    "plt.yticks(fontsize=14)\n",
    "x_is = 'PC1'\n",
    "y_is = 'PC2'\n",
    "plt.xlabel(x_is,fontsize=20)\n",
    "plt.ylabel(y_is,fontsize=20)\n",
    "plt.title(\"Principal Component Analysis of Vegetation Indices\",fontsize=20)\n",
    "plt.scatter(pca_df[x_is],\n",
    "            pca_df[y_is], s = 15)\n",
    "for i, label in enumerate(df_out_all['ID']):\n",
    "    if label == 'Section_8_Pt7_No':\n",
    "        plt.text(pca_df[x_is].iloc[i],\n",
    "                 pca_df[y_is].iloc[i],\n",
    "                 label)\n",
    "        plt.scatter(pca_df[x_is].iloc[i],\n",
    "                    pca_df[y_is].iloc[i],\n",
    "                    c='black', s=15)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "id": "45625694-0488-4c7f-baa1-7f7f20938301",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<AxesSubplot:xlabel='n_years', ylabel='rmae'>"
      ]
     },
     "execution_count": 17,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "n_years = [7, 4, 5, 3, 1]\n",
    "rmae = [0.30, 0.37, 0.52, 0.47, 0.34]\n",
    "df_error = pd.DataFrame(data={'n_years': n_years,\n",
    "                              'rmae': rmae})\n",
    "sns.regplot(x='n_years', y='rmae', data=df_error, ci=None)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "id": "fffdcd6a-556d-4457-ac72-58cdff00ad98",
   "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>index</th>\n",
       "      <th>ID</th>\n",
       "      <th>Date</th>\n",
       "      <th>gps_E</th>\n",
       "      <th>gps_N</th>\n",
       "      <th>bm_gm2</th>\n",
       "      <th>geometry</th>\n",
       "      <th>ndvi</th>\n",
       "      <th>dfi</th>\n",
       "      <th>ndti</th>\n",
       "      <th>...</th>\n",
       "      <th>bai_236</th>\n",
       "      <th>bai_246</th>\n",
       "      <th>bai_346</th>\n",
       "      <th>nir</th>\n",
       "      <th>swir1</th>\n",
       "      <th>swir2</th>\n",
       "      <th>bm</th>\n",
       "      <th>Year</th>\n",
       "      <th>bm_pred_tb</th>\n",
       "      <th>z_bm_tb</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Project</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>BIRD</th>\n",
       "      <td>352</td>\n",
       "      <td>352</td>\n",
       "      <td>352</td>\n",
       "      <td>352</td>\n",
       "      <td>352</td>\n",
       "      <td>352</td>\n",
       "      <td>352</td>\n",
       "      <td>352</td>\n",
       "      <td>352</td>\n",
       "      <td>352</td>\n",
       "      <td>...</td>\n",
       "      <td>352</td>\n",
       "      <td>352</td>\n",
       "      <td>352</td>\n",
       "      <td>352</td>\n",
       "      <td>352</td>\n",
       "      <td>352</td>\n",
       "      <td>352</td>\n",
       "      <td>352</td>\n",
       "      <td>352</td>\n",
       "      <td>352</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>FIRE</th>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>...</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>NEX</th>\n",
       "      <td>4</td>\n",
       "      <td>4</td>\n",
       "      <td>4</td>\n",
       "      <td>4</td>\n",
       "      <td>4</td>\n",
       "      <td>4</td>\n",
       "      <td>4</td>\n",
       "      <td>4</td>\n",
       "      <td>4</td>\n",
       "      <td>4</td>\n",
       "      <td>...</td>\n",
       "      <td>4</td>\n",
       "      <td>4</td>\n",
       "      <td>4</td>\n",
       "      <td>4</td>\n",
       "      <td>4</td>\n",
       "      <td>4</td>\n",
       "      <td>4</td>\n",
       "      <td>4</td>\n",
       "      <td>4</td>\n",
       "      <td>4</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>PDOG</th>\n",
       "      <td>42</td>\n",
       "      <td>42</td>\n",
       "      <td>42</td>\n",
       "      <td>42</td>\n",
       "      <td>42</td>\n",
       "      <td>42</td>\n",
       "      <td>42</td>\n",
       "      <td>42</td>\n",
       "      <td>42</td>\n",
       "      <td>42</td>\n",
       "      <td>...</td>\n",
       "      <td>42</td>\n",
       "      <td>42</td>\n",
       "      <td>42</td>\n",
       "      <td>42</td>\n",
       "      <td>42</td>\n",
       "      <td>42</td>\n",
       "      <td>42</td>\n",
       "      <td>42</td>\n",
       "      <td>42</td>\n",
       "      <td>42</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>WTGN</th>\n",
       "      <td>15</td>\n",
       "      <td>15</td>\n",
       "      <td>15</td>\n",
       "      <td>15</td>\n",
       "      <td>15</td>\n",
       "      <td>15</td>\n",
       "      <td>15</td>\n",
       "      <td>15</td>\n",
       "      <td>15</td>\n",
       "      <td>15</td>\n",
       "      <td>...</td>\n",
       "      <td>15</td>\n",
       "      <td>15</td>\n",
       "      <td>15</td>\n",
       "      <td>15</td>\n",
       "      <td>15</td>\n",
       "      <td>15</td>\n",
       "      <td>15</td>\n",
       "      <td>15</td>\n",
       "      <td>15</td>\n",
       "      <td>15</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>5 rows × 25 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "         index   ID  Date  gps_E  gps_N  bm_gm2  geometry  ndvi  dfi  ndti  \\\n",
       "Project                                                                      \n",
       "BIRD       352  352   352    352    352     352       352   352  352   352   \n",
       "FIRE         1    1     1      1      1       1         1     1    1     1   \n",
       "NEX          4    4     4      4      4       4         4     4    4     4   \n",
       "PDOG        42   42    42     42     42      42        42    42   42    42   \n",
       "WTGN        15   15    15     15     15      15        15    15   15    15   \n",
       "\n",
       "         ...  bai_236  bai_246  bai_346  nir  swir1  swir2   bm  Year  \\\n",
       "Project  ...                                                            \n",
       "BIRD     ...      352      352      352  352    352    352  352   352   \n",
       "FIRE     ...        1        1        1    1      1      1    1     1   \n",
       "NEX      ...        4        4        4    4      4      4    4     4   \n",
       "PDOG     ...       42       42       42   42     42     42   42    42   \n",
       "WTGN     ...       15       15       15   15     15     15   15    15   \n",
       "\n",
       "         bm_pred_tb  z_bm_tb  \n",
       "Project                       \n",
       "BIRD            352      352  \n",
       "FIRE              1        1  \n",
       "NEX               4        4  \n",
       "PDOG             42       42  \n",
       "WTGN             15       15  \n",
       "\n",
       "[5 rows x 25 columns]"
      ]
     },
     "execution_count": 25,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_out_all[df_out_all['bm'] < 100].groupby('Project').count()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "id": "ea8efade-1d0d-4590-882a-fa7432c1d2f3",
   "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>index</th>\n",
       "      <th>ID</th>\n",
       "      <th>Date</th>\n",
       "      <th>gps_E</th>\n",
       "      <th>gps_N</th>\n",
       "      <th>bm_gm2</th>\n",
       "      <th>geometry</th>\n",
       "      <th>ndvi</th>\n",
       "      <th>dfi</th>\n",
       "      <th>ndti</th>\n",
       "      <th>...</th>\n",
       "      <th>bai_236</th>\n",
       "      <th>bai_246</th>\n",
       "      <th>bai_346</th>\n",
       "      <th>nir</th>\n",
       "      <th>swir1</th>\n",
       "      <th>swir2</th>\n",
       "      <th>bm</th>\n",
       "      <th>Year</th>\n",
       "      <th>bm_pred_tb</th>\n",
       "      <th>z_bm_tb</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Project</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>BIRD</th>\n",
       "      <td>1743</td>\n",
       "      <td>1743</td>\n",
       "      <td>1743</td>\n",
       "      <td>1743</td>\n",
       "      <td>1743</td>\n",
       "      <td>1743</td>\n",
       "      <td>1743</td>\n",
       "      <td>1743</td>\n",
       "      <td>1743</td>\n",
       "      <td>1743</td>\n",
       "      <td>...</td>\n",
       "      <td>1743</td>\n",
       "      <td>1743</td>\n",
       "      <td>1743</td>\n",
       "      <td>1743</td>\n",
       "      <td>1743</td>\n",
       "      <td>1743</td>\n",
       "      <td>1743</td>\n",
       "      <td>1743</td>\n",
       "      <td>1743</td>\n",
       "      <td>1743</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>FIRE</th>\n",
       "      <td>90</td>\n",
       "      <td>90</td>\n",
       "      <td>90</td>\n",
       "      <td>90</td>\n",
       "      <td>90</td>\n",
       "      <td>90</td>\n",
       "      <td>90</td>\n",
       "      <td>90</td>\n",
       "      <td>90</td>\n",
       "      <td>90</td>\n",
       "      <td>...</td>\n",
       "      <td>90</td>\n",
       "      <td>90</td>\n",
       "      <td>90</td>\n",
       "      <td>90</td>\n",
       "      <td>90</td>\n",
       "      <td>90</td>\n",
       "      <td>90</td>\n",
       "      <td>90</td>\n",
       "      <td>90</td>\n",
       "      <td>90</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>NEX</th>\n",
       "      <td>351</td>\n",
       "      <td>351</td>\n",
       "      <td>351</td>\n",
       "      <td>351</td>\n",
       "      <td>351</td>\n",
       "      <td>351</td>\n",
       "      <td>351</td>\n",
       "      <td>351</td>\n",
       "      <td>351</td>\n",
       "      <td>351</td>\n",
       "      <td>...</td>\n",
       "      <td>351</td>\n",
       "      <td>351</td>\n",
       "      <td>351</td>\n",
       "      <td>351</td>\n",
       "      <td>351</td>\n",
       "      <td>351</td>\n",
       "      <td>351</td>\n",
       "      <td>351</td>\n",
       "      <td>351</td>\n",
       "      <td>351</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>PDOG</th>\n",
       "      <td>259</td>\n",
       "      <td>259</td>\n",
       "      <td>259</td>\n",
       "      <td>259</td>\n",
       "      <td>259</td>\n",
       "      <td>259</td>\n",
       "      <td>259</td>\n",
       "      <td>259</td>\n",
       "      <td>259</td>\n",
       "      <td>259</td>\n",
       "      <td>...</td>\n",
       "      <td>259</td>\n",
       "      <td>259</td>\n",
       "      <td>259</td>\n",
       "      <td>259</td>\n",
       "      <td>259</td>\n",
       "      <td>259</td>\n",
       "      <td>259</td>\n",
       "      <td>259</td>\n",
       "      <td>259</td>\n",
       "      <td>259</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>WTGN</th>\n",
       "      <td>343</td>\n",
       "      <td>343</td>\n",
       "      <td>343</td>\n",
       "      <td>343</td>\n",
       "      <td>343</td>\n",
       "      <td>343</td>\n",
       "      <td>343</td>\n",
       "      <td>343</td>\n",
       "      <td>343</td>\n",
       "      <td>343</td>\n",
       "      <td>...</td>\n",
       "      <td>343</td>\n",
       "      <td>343</td>\n",
       "      <td>343</td>\n",
       "      <td>343</td>\n",
       "      <td>343</td>\n",
       "      <td>343</td>\n",
       "      <td>343</td>\n",
       "      <td>343</td>\n",
       "      <td>343</td>\n",
       "      <td>343</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>5 rows × 25 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "         index    ID  Date  gps_E  gps_N  bm_gm2  geometry  ndvi   dfi  ndti  \\\n",
       "Project                                                                        \n",
       "BIRD      1743  1743  1743   1743   1743    1743      1743  1743  1743  1743   \n",
       "FIRE        90    90    90     90     90      90        90    90    90    90   \n",
       "NEX        351   351   351    351    351     351       351   351   351   351   \n",
       "PDOG       259   259   259    259    259     259       259   259   259   259   \n",
       "WTGN       343   343   343    343    343     343       343   343   343   343   \n",
       "\n",
       "         ...  bai_236  bai_246  bai_346   nir  swir1  swir2    bm  Year  \\\n",
       "Project  ...                                                              \n",
       "BIRD     ...     1743     1743     1743  1743   1743   1743  1743  1743   \n",
       "FIRE     ...       90       90       90    90     90     90    90    90   \n",
       "NEX      ...      351      351      351   351    351    351   351   351   \n",
       "PDOG     ...      259      259      259   259    259    259   259   259   \n",
       "WTGN     ...      343      343      343   343    343    343   343   343   \n",
       "\n",
       "         bm_pred_tb  z_bm_tb  \n",
       "Project                       \n",
       "BIRD           1743     1743  \n",
       "FIRE             90       90  \n",
       "NEX             351      351  \n",
       "PDOG            259      259  \n",
       "WTGN            343      343  \n",
       "\n",
       "[5 rows x 25 columns]"
      ]
     },
     "execution_count": 26,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_out_all.groupby('Project').count()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "a989b877-19f6-4676-9e32-44f3512bd6fd",
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3 (ipykernel)",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.10.5"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}