comparison_plots.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {},
   "outputs": [],
   "source": [
    "import os\n",
    "import pandas as pd\n",
    "import matplotlib.pyplot as plt\n",
    "import seaborn as sns\n",
    "from matplotlib.colors import LogNorm\n",
    "import numpy as np\n",
    "from scipy.stats import linregress\n",
    "from scipy.optimize import curve_fit\n",
    "\n",
    "datadir = f\"./data/comparison\"\n",
    "plotdir = f\"./plots/comparison\"\n",
    "\n",
    "if not os.path.exists(plotdir):\n",
    "    os.makedirs(plotdir)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {},
   "outputs": [],
   "source": [
    "def prepare_performance(perf_df, algo_join_cols,last_task=10):\n",
    "    # get stability plasticity and capacity\n",
    "    stability_df = perf_df[(perf_df['phase'] == last_task) & (perf_df['task'] == 0)]\n",
    "    plasticity_df = perf_df[(perf_df['phase'] == last_task) & (perf_df['task'] == last_task)]\n",
    "\n",
    "    # drop columns manager, ridge result, run, index\n",
    "    stability_df = stability_df.drop(columns=['manager', 'ridge_result', 'run', 'index'])\n",
    "    plasticity_df = plasticity_df.drop(columns=['manager', 'ridge_result', 'run', 'index'])\n",
    "\n",
    "    # rename cl_performance to plasticity and stability\n",
    "    stability_df.rename(columns={'cl_result': 'stability'}, inplace=True)\n",
    "    plasticity_df.rename(columns={'cl_result': 'plasticity'}, inplace=True)\n",
    "\n",
    "    plasticity_df = plasticity_df.drop(columns=['task', 'phase'])\n",
    "    stability_df = stability_df.drop(columns=['task', 'phase'])\n",
    "\n",
    "    # merge both dataframes\n",
    "    stability_plasticity_df = pd.merge(stability_df, plasticity_df, on=['tstamp', 'algorithm', 'dense_size'].extend(algo_join_cols))\n",
    "    stability_plasticity_df['capacity'] = stability_plasticity_df['stability'] + stability_plasticity_df['plasticity']\n",
    "    return stability_plasticity_df\n",
    "\n",
    "def prepare_trajectories(filter_traj_df, last_task=10):\n",
    "    run_names = filter_traj_df['run']\n",
    "    run_names = run_names.str.split('_', expand=True)\n",
    "    run_names['tstamp'] = run_names[0]\n",
    "    filter_traj_df = filter_traj_df.drop(columns=['run'])\n",
    "    filter_traj_df['tstamp'] = run_names['tstamp']\n",
    "    filter_traj_df = filter_traj_df.reset_index()\n",
    "    filter_traj_df_last = filter_traj_df[filter_traj_df['phase'] == last_task]\n",
    "    distances_to_first = filter_traj_df_last[['tstamp', 'distance_to_first_null_space', 'distance_to_first_in_plane']]\n",
    "    filter_traj_df = filter_traj_df.drop(columns=['distance_to_first_null_space', 'distance_to_first_in_plane', 'phase', 'index'])\n",
    "\n",
    "    # group by tstamp\n",
    "    filter_traj_df_grp = filter_traj_df.groupby(['tstamp'])\n",
    "    distances_cumulative = filter_traj_df_grp.sum(numeric_only=True).reset_index()\n",
    "\n",
    "    # join on tstamp\n",
    "    distance_df = pd.merge(distances_cumulative, distances_to_first, on=['tstamp'])\n",
    "    return distance_df\n",
    "\n",
    "def join_frames(distance_df, perf_df):\n",
    "    # tstamps are floats\n",
    "    distance_df['tstamp'] = distance_df['tstamp'].astype(float)\n",
    "    perf_df['tstamp'] = perf_df['tstamp'].astype(float)\n",
    "\n",
    "    # sort both frames by ascending tstamp\n",
    "    distance_df = distance_df.sort_values(by=['tstamp'])\n",
    "    perf_df = perf_df.sort_values(by=['tstamp'])\n",
    "\n",
    "    # rename tstamps to tstamp_d and tstamp_s\n",
    "    distance_df.rename(columns={'tstamp': 'tstamp_d'}, inplace=True)\n",
    "    perf_df.rename(columns={'tstamp': 'tstamp_s'}, inplace=True)\n",
    "\n",
    "    # reset indicees and drop index\n",
    "    distance_df = distance_df.reset_index()\n",
    "    distance_df = distance_df.drop(columns=['index'])\n",
    "\n",
    "    perf_df = perf_df.reset_index()\n",
    "    perf_df = perf_df.drop(columns=['index'])\n",
    "\n",
    "    # concatenate both dataframes along the columns\n",
    "    combined_df = pd.concat([distance_df, perf_df], axis=1)\n",
    "    return combined_df"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {},
   "outputs": [],
   "source": [
    "############################### EWC\n",
    "ewc_performance_path = f\"{datadir}/ewc_cifar110/performance.csv\"\n",
    "ewc_trajectories_filter_path = f\"{datadir}/ewc_cifar110/filtered_trajectories_ewc_plasticity_cifar110_nodrop_task_0_relu_fc1.pkl\"\n",
    "\n",
    "ewc_perf_df = pd.read_csv(ewc_performance_path)\n",
    "ewc_trajectories_filter = pd.read_pickle(ewc_trajectories_filter_path)\n",
    "\n",
    "############################### SI\n",
    "si_performance_path = f\"{datadir}/si_cifar110/performance.csv\"\n",
    "si_trajectories_filter_path = f\"{datadir}/si_cifar110/filtered_trajectories_si_plasticity_cifar110_nodrop_task_0_relu_fc1.pkl\"\n",
    "\n",
    "si_perf_df = pd.read_csv(si_performance_path)\n",
    "si_trajectories_filter = pd.read_pickle(si_trajectories_filter_path)\n",
    "\n",
    "############################### Replay fixed readout\n",
    "replay_perf_path = f\"{datadir}/replay_fixed_readout/performance.csv\"\n",
    "replay_trajectories_filter_path = f\"{datadir}/si_cifar110/filtered_trajectories_si_plasticity_cifar110_nodrop_task_0_relu_fc1.pkl\"\n",
    "\n",
    "replay_perf_df = pd.read_csv(replay_perf_path)\n",
    "replay_trajectories_filter = pd.read_pickle(replay_trajectories_filter_path)\n",
    "\n",
    "replay_perf_path = f\"{datadir}/replay_fixed_readout/performance_high.csv\"\n",
    "replay_trajectories_filter_path = f\"{datadir}/replay_fixed_readout/filtered_trajectories_replay_plasticity_cifar110_frozen_readouts_high_task_0_relu_fc1.pkl\"\n",
    "\n",
    "replay_perf_df2 = pd.read_csv(replay_perf_path)\n",
    "replay_trajectories_filter2 = pd.read_pickle(replay_trajectories_filter_path)\n",
    "\n",
    "replay_perf_df = pd.concat([replay_perf_df, replay_perf_df2])\n",
    "replay_trajectories_filter = pd.concat([replay_trajectories_filter, replay_trajectories_filter2])\n",
    "\n",
    "############################### GEM\n",
    "gem_performance_path = f\"{datadir}/gem_buf_cifar110/performance.csv\"\n",
    "gem_trajectories_filter_path = f\"{datadir}/gem_buf_cifar110/filtered_trajectories_gem_sweep_buffersize_task_0_relu_fc1.pkl\"\n",
    "\n",
    "gem_perf_df = pd.read_csv(gem_performance_path)\n",
    "gem_trajectories_filter = pd.read_pickle(gem_trajectories_filter_path)\n",
    "\n",
    "gem_performance_path = f\"{datadir}/gem_cifar110/performance.csv\"\n",
    "gem_trajectories_filter_path = f\"{datadir}/gem_cifar110/filtered_trajectories_cifar110_nodrop_task_0_relu_fc1.pkl\"\n",
    "\n",
    "gem_perf_df2 = pd.read_csv(gem_performance_path)\n",
    "gem_trajectories_filter2 = pd.read_pickle(gem_trajectories_filter_path)\n",
    "\n",
    "gem_performance_path = f\"{datadir}/gem_high_cifar110/performance.csv\"\n",
    "gem_trajectories_filter_path = f\"{datadir}/gem_high_cifar110/filtered_trajectories_cifar110_gem_capacity_exp_nodrop_augment_high_buf_task_0_relu_fc1.pkl\"\n",
    "\n",
    "gem_perf_df3 = pd.read_csv(gem_performance_path)\n",
    "gem_trajectories_filter3 = pd.read_pickle(gem_trajectories_filter_path)\n",
    "\n",
    "\n",
    "gem_perf_df = pd.concat([gem_perf_df, gem_perf_df2, gem_perf_df3])\n",
    "gem_trajectories_filter = pd.concat([gem_trajectories_filter, gem_trajectories_filter2, gem_trajectories_filter3])\n",
    "\n",
    "############################### LwF\n",
    "\n",
    "lwf_performance_path = f\"{datadir}/lwf_cifar110/performance.csv\"\n",
    "lwf_trajectories_filter = f\"{datadir}/lwf_cifar110/filtered_trajectories_lwf_plasticity_cifar110_frozen_readouts_task_0_relu_fc1.pkl\"\n",
    "\n",
    "lwf_perf_df = pd.read_csv(lwf_performance_path)\n",
    "lwf_trajectories_filter = pd.read_pickle(lwf_trajectories_filter)\n",
    "\n",
    "lwf_performance_path = f\"{datadir}/lwf_cifar110/performance_temp.csv\"\n",
    "lwf_trajectories_filter_path = f\"{datadir}/lwf_cifar110/filtered_trajectories_temp_lwf_task_0_relu_fc1.pkl\"\n",
    "\n",
    "lwf_perf_df2 = pd.read_csv(lwf_performance_path)\n",
    "lwf_trajectories_filter2 = pd.read_pickle(lwf_trajectories_filter_path)\n",
    "\n",
    "lwf_perf_df = pd.concat([lwf_perf_df, lwf_perf_df2])\n",
    "lwf_trajectories_filter = pd.concat([lwf_trajectories_filter, lwf_trajectories_filter2])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {},
   "outputs": [],
   "source": [
    "distance_df_ewc = prepare_trajectories(ewc_trajectories_filter)\n",
    "perf_df_ewc = prepare_performance(ewc_perf_df, ['ewc_lambda'])\n",
    "combined_df_ewc = join_frames(distance_df_ewc, perf_df_ewc)\n",
    "\n",
    "distance_df_si = prepare_trajectories(si_trajectories_filter)\n",
    "perf_df_si = prepare_performance(si_perf_df, ['si_lambda', 'si_eps'])\n",
    "combined_df_si = join_frames(distance_df_si, perf_df_si)\n",
    "\n",
    "distance_df_replay = prepare_trajectories(replay_trajectories_filter)\n",
    "perf_df_replay = prepare_performance(replay_perf_df, ['replay_mem_size'])\n",
    "combined_df_replay = join_frames(distance_df_replay, perf_df_replay)\n",
    "combined_df_replay = combined_df_replay[combined_df_replay['replay_mem_size'] <= 60000]  # remove unreasonably large replay mem sizes\n",
    "\n",
    "distance_df_gem = prepare_trajectories(gem_trajectories_filter)\n",
    "perf_df_gem = prepare_performance(gem_perf_df, ['gem_memory_strength', 'gem_patterns_per_experience'])\n",
    "combined_df_gem = join_frames(distance_df_gem, perf_df_gem)\n",
    "\n",
    "distance_df_lwf = prepare_trajectories(lwf_trajectories_filter)\n",
    "perf_df_lwf = prepare_performance(lwf_perf_df, ['lwf_alpha', 'lwf_temperature'])\n",
    "combined_df_lwf = join_frames(distance_df_lwf, perf_df_lwf)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/Users/daniel/miniconda3/envs/pt/lib/python3.10/site-packages/scipy/optimize/_minpack_py.py:1010: OptimizeWarning: Covariance of the parameters could not be estimated\n",
      "  warnings.warn('Covariance of the parameters could not be estimated',\n",
      "/Users/daniel/miniconda3/envs/pt/lib/python3.10/site-packages/scipy/optimize/_minpack_py.py:1010: OptimizeWarning: Covariance of the parameters could not be estimated\n",
      "  warnings.warn('Covariance of the parameters could not be estimated',\n",
      "/Users/daniel/miniconda3/envs/pt/lib/python3.10/site-packages/scipy/optimize/_minpack_py.py:1010: OptimizeWarning: Covariance of the parameters could not be estimated\n",
      "  warnings.warn('Covariance of the parameters could not be estimated',\n",
      "/Users/daniel/miniconda3/envs/pt/lib/python3.10/site-packages/scipy/optimize/_minpack_py.py:1010: OptimizeWarning: Covariance of the parameters could not be estimated\n",
      "  warnings.warn('Covariance of the parameters could not be estimated',\n",
      "/Users/daniel/miniconda3/envs/pt/lib/python3.10/site-packages/scipy/optimize/_minpack_py.py:1010: OptimizeWarning: Covariance of the parameters could not be estimated\n",
      "  warnings.warn('Covariance of the parameters could not be estimated',\n"
     ]
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 1000x1000 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "DENSE_SIZE = 128\n",
    "SI_EPS = 1.\n",
    "LWF_TEMP = 1.\n",
    "MARKER_SIZE = 100\n",
    "REFERENCE_STABILITIES = [.375, .6, .735]\n",
    "\n",
    "def func(x, a, b, c, d):\n",
    "    # 3rd order polynomial\n",
    "    return a + b * np.exp(c * x)\n",
    "\n",
    "plt.figure(figsize=(10,10))\n",
    "# vertical line at ref\n",
    "for REFERENCE_STABILITY in REFERENCE_STABILITIES:\n",
    "    plt.axhline(y=REFERENCE_STABILITY, linestyle='dotted', color='k', alpha=.3)\n",
    "\n",
    "########## EWC\n",
    "    \n",
    "df = combined_df_ewc[combined_df_ewc['dense_size'] == DENSE_SIZE]\n",
    "df = df.groupby(['ewc_lambda']).mean(numeric_only=True).reset_index()\n",
    "log_norm_ewc = LogNorm(vmin=df['ewc_lambda'].min(), vmax=df['ewc_lambda'].max())\n",
    "\n",
    "popt, pcov = curve_fit(func, df['stability'], df['plasticity'], p0=(1, 1, 10, 0))\n",
    "x = np.linspace(df['stability'].min(), df['stability'].max(), 100)\n",
    "plt.plot(func(x, *popt), x, 'g--')\n",
    "\n",
    "plt.scatter(df['plasticity'],\n",
    "            df['stability'],\n",
    "            c=df['ewc_lambda'],\n",
    "            cmap='Greens',\n",
    "            norm=log_norm_ewc,\n",
    "            s=MARKER_SIZE,\n",
    "            marker='s',\n",
    "            edgecolors='k',\n",
    "            linewidths=1,\n",
    "            label='EWC'\n",
    "            )\n",
    "\n",
    "########## SI\n",
    "df = combined_df_si[combined_df_si['dense_size'] == DENSE_SIZE]\n",
    "df = df[df['si_eps'] == SI_EPS]\n",
    "df = df.groupby(['si_lambda']).mean(numeric_only=True).reset_index()\n",
    "\n",
    "x = np.linspace(df['stability'].min(), df['stability'].max(), 100)\n",
    "popt, pcov = curve_fit(func, df['stability'], df['plasticity'], p0=(1, 1, 10, 0))\n",
    "plt.plot(func(x, *popt), x, 'r--')\n",
    "\n",
    "norm = LogNorm(vmin=df['si_lambda'].min(), vmax=df['si_lambda'].max())\n",
    "plt.scatter(df['plasticity'],\n",
    "            df['stability'],\n",
    "            c=df['si_lambda'],\n",
    "            cmap='Reds',\n",
    "            norm=norm,\n",
    "            s=MARKER_SIZE,\n",
    "            marker='o',\n",
    "            edgecolors='k',\n",
    "            linewidths=1,\n",
    "            label='SI'\n",
    "            )\n",
    "\n",
    "########## Replay\n",
    "df = combined_df_replay[combined_df_replay['dense_size'] == DENSE_SIZE]\n",
    "df = df.groupby(['replay_mem_size']).mean(numeric_only=True).reset_index()\n",
    "\n",
    "# fit poly\n",
    "x = np.linspace(df['stability'].min(), df['stability'].max(), 100)\n",
    "popt, pcov = curve_fit(func, df['stability'], df['plasticity'], p0=(1,0,0,0))\n",
    "plt.plot(func(x, *popt), x, 'purple', linestyle='--')\n",
    "\n",
    "\n",
    "plt.scatter(df['plasticity'],\n",
    "            df['stability'],\n",
    "            c=df['replay_mem_size'],\n",
    "            cmap='Purples',\n",
    "            norm=LogNorm(),\n",
    "            s=MARKER_SIZE,\n",
    "            edgecolors='k',\n",
    "            linewidths=1,\n",
    "            label='Replay',\n",
    "            marker='^'\n",
    "            )\n",
    "\n",
    "########## gem\n",
    "df = combined_df_gem[combined_df_gem['dense_size'] == DENSE_SIZE]\n",
    "# df = df[df['gem_patterns_per_experience'] == GEM_PATTERNS]\n",
    "df = df[df['gem_memory_strength'] >= 0.]\n",
    "df = df.groupby(['gem_patterns_per_experience', 'gem_memory_strength']).mean(numeric_only=True).reset_index()\n",
    "\n",
    "# fit poly\n",
    "x = np.linspace(df['stability'].min(), df['stability'].max(), 100)\n",
    "popt, pcov = curve_fit(func, df['stability'], df['plasticity'], p0=(1,0,0,0))\n",
    "plt.plot(func(x, *popt), x, 'gray', linestyle='--')\n",
    "\n",
    "\n",
    "plt.scatter(df['plasticity'],\n",
    "            df['stability'],\n",
    "            c=df['gem_memory_strength'],\n",
    "            cmap='Greys',\n",
    "            s=MARKER_SIZE + 10,  # stars looks smaller\n",
    "            edgecolors='k',\n",
    "            linewidths=1,\n",
    "            label='GEM',\n",
    "            marker='*'\n",
    ")\n",
    "\n",
    "########## lwf\n",
    "df = combined_df_lwf[combined_df_lwf['dense_size'] == DENSE_SIZE]\n",
    "df = df[df['lwf_temperature'] == LWF_TEMP]\n",
    "df = df.groupby(['lwf_alpha']).mean(numeric_only=True).reset_index()\n",
    "\n",
    "# fit poly\n",
    "x = np.linspace(df['stability'].min(), df['stability'].max(), 100)\n",
    "popt, pcov = curve_fit(func, df['stability'], df['plasticity'], p0=(1,0,1,10))\n",
    "plt.plot(func(x, *popt), x, 'blue', linestyle='--')\n",
    "\n",
    "plt.scatter(df['plasticity'],\n",
    "            df['stability'],\n",
    "            c=df['lwf_alpha'],\n",
    "            cmap='Blues',\n",
    "            norm=LogNorm(),\n",
    "            s=MARKER_SIZE,\n",
    "            edgecolors='k',\n",
    "            linewidths=1,\n",
    "            label='LwF',\n",
    "            marker='D'\n",
    "            )\n",
    "\n",
    "plt.legend()\n",
    "plt.xlabel('Plasticity')\n",
    "plt.ylabel('Stability')\n",
    "sns.despine()\n",
    "plt.xticks(fontsize=18)\n",
    "plt.yticks(fontsize=18)\n",
    "plt.savefig(f'{plotdir}/plasticity_vs_stability_128_units_new.png', bbox_inches='tight')\n",
    "\n",
    "\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 800x400 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 800x400 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 800x400 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# for each algorithm, get closest mean stability to reference stability and parameters\n",
    "\n",
    "def get_closest_stability(avg_df, reference_stability):\n",
    "    # get closest stability to reference stability\n",
    "    avg_df['distance_to_reference'] = abs(avg_df['stability'] - reference_stability)\n",
    "    avg_df = avg_df.sort_values(by=['distance_to_reference'])\n",
    "    closest_stability = avg_df.iloc[0]\n",
    "    return closest_stability\n",
    "\n",
    "for REFERENCE_STABILITY in REFERENCE_STABILITIES:\n",
    "    # EWC\n",
    "    df = combined_df_ewc[combined_df_ewc['dense_size'] == DENSE_SIZE]\n",
    "    df = df.groupby(['ewc_lambda']).mean(numeric_only=True).reset_index()\n",
    "    closest_stability_ewc = get_closest_stability(df, REFERENCE_STABILITY)\n",
    "    # collect runs\n",
    "    ewc_ref_runs = combined_df_ewc[(combined_df_ewc['dense_size'] == DENSE_SIZE) & (combined_df_ewc['ewc_lambda'] == closest_stability_ewc['ewc_lambda'])]\n",
    "    # SI\n",
    "    df = combined_df_si[combined_df_si['dense_size'] == DENSE_SIZE]\n",
    "    df = df[df['si_eps'] == SI_EPS]\n",
    "    df = df.groupby(['si_lambda']).mean(numeric_only=True).reset_index()\n",
    "    closest_stability_si = get_closest_stability(df, REFERENCE_STABILITY)\n",
    "    # collect runs\n",
    "    si_ref_runs = combined_df_si[(combined_df_si['dense_size'] == DENSE_SIZE) & (combined_df_si['si_eps'] == SI_EPS) & (combined_df_si['si_lambda'] == closest_stability_si['si_lambda'])]\n",
    "    # Replay\n",
    "    df = combined_df_replay[combined_df_replay['dense_size'] == DENSE_SIZE]\n",
    "    df = df.groupby(['replay_mem_size']).mean(numeric_only=True).reset_index()\n",
    "    closest_stability_replay = get_closest_stability(df, REFERENCE_STABILITY)\n",
    "    # collect runs\n",
    "    replay_ref_runs = combined_df_replay[(combined_df_replay['dense_size'] == DENSE_SIZE) & (combined_df_replay['replay_mem_size'] == closest_stability_replay['replay_mem_size'])]\n",
    "    # gem\n",
    "    df = combined_df_gem[combined_df_gem['dense_size'] == DENSE_SIZE]\n",
    "    df = df[df['gem_memory_strength'] >= 0]\n",
    "    df = df.groupby(['gem_patterns_per_experience', 'gem_memory_strength']).mean(numeric_only=True).reset_index()\n",
    "    closest_stability_gem = get_closest_stability(df, REFERENCE_STABILITY)\n",
    "    # collect runs\n",
    "    gem_ref_runs = combined_df_gem[(combined_df_gem['dense_size'] == DENSE_SIZE) & (combined_df_gem['gem_memory_strength'] == closest_stability_gem['gem_memory_strength']) & (combined_df_gem['gem_patterns_per_experience'] == closest_stability_gem['gem_patterns_per_experience'])]\n",
    "    # lwf\n",
    "    df = combined_df_lwf[combined_df_lwf['dense_size'] == DENSE_SIZE]\n",
    "    df = df.groupby(['lwf_alpha', 'lwf_temperature']).mean(numeric_only=True).reset_index()\n",
    "    closest_stability_lwf = get_closest_stability(df, REFERENCE_STABILITY)\n",
    "    # collect runs\n",
    "    lwf_ref_runs = combined_df_lwf[(combined_df_lwf['dense_size'] == DENSE_SIZE) & (combined_df_lwf['lwf_alpha'] == closest_stability_lwf['lwf_alpha']) & (combined_df_lwf['lwf_temperature'] == closest_stability_lwf['lwf_temperature'])]\n",
    "\n",
    "    # force a copy of all dfs\n",
    "    ewc_ref_runs = ewc_ref_runs.copy()\n",
    "    si_ref_runs = si_ref_runs.copy()\n",
    "    replay_ref_runs = replay_ref_runs.copy()\n",
    "    gem_ref_runs = gem_ref_runs.copy()\n",
    "    lwf_ref_runs = lwf_ref_runs.copy()\n",
    "\n",
    "    # concatenate all runs after adding algorithm column\n",
    "    ewc_ref_runs['algorithm'] = 'EWC'\n",
    "    si_ref_runs['algorithm'] = 'SI'\n",
    "    replay_ref_runs['algorithm'] = 'Replay'\n",
    "    gem_ref_runs['algorithm'] = 'GEM'\n",
    "    lwf_ref_runs['algorithm'] = 'LwF'\n",
    "\n",
    "    # concatenate all runs\n",
    "    ref_runs = pd.concat([ewc_ref_runs, si_ref_runs, replay_ref_runs, gem_ref_runs, lwf_ref_runs])\n",
    "    # select algorithm and distances\n",
    "    ref_runs = ref_runs[['algorithm', 'distance_to_first_null_space', 'distance_to_first_in_plane']]\n",
    "    # avg over algorithm\n",
    "    ref_runs = ref_runs.groupby(['algorithm']).mean(numeric_only=True).reset_index()\n",
    "    plt.figure(figsize=(8,4))\n",
    "\n",
    "    # ewc\n",
    "    plt.scatter(\n",
    "        ref_runs[ref_runs['algorithm'] == 'EWC']['distance_to_first_in_plane'],\n",
    "        ref_runs[ref_runs['algorithm'] == 'EWC']['distance_to_first_null_space'],\n",
    "        c='g',\n",
    "        s=MARKER_SIZE,\n",
    "        marker='s',\n",
    "        edgecolors='k',\n",
    "        linewidths=1,\n",
    "        label='EWC'\n",
    "    )\n",
    "\n",
    "    # SI\n",
    "    plt.scatter(\n",
    "        ref_runs[ref_runs['algorithm'] == 'SI']['distance_to_first_in_plane'],\n",
    "        ref_runs[ref_runs['algorithm'] == 'SI']['distance_to_first_null_space'],\n",
    "        c='r',\n",
    "        s=MARKER_SIZE,\n",
    "        marker='o',\n",
    "        edgecolors='k',\n",
    "        linewidths=1,\n",
    "        label='SI'\n",
    "    )\n",
    "\n",
    "    # Replay\n",
    "    plt.scatter(\n",
    "        ref_runs[ref_runs['algorithm'] == 'Replay']['distance_to_first_in_plane'],\n",
    "        ref_runs[ref_runs['algorithm'] == 'Replay']['distance_to_first_null_space'],\n",
    "        c='purple',\n",
    "        s=MARKER_SIZE,\n",
    "        marker='^',\n",
    "        edgecolors='k',\n",
    "        linewidths=1,\n",
    "        label='Replay'\n",
    "    )\n",
    "\n",
    "    # GEM\n",
    "    plt.scatter(\n",
    "        ref_runs[ref_runs['algorithm'] == 'GEM']['distance_to_first_in_plane'],\n",
    "        ref_runs[ref_runs['algorithm'] == 'GEM']['distance_to_first_null_space'],\n",
    "        c='grey',\n",
    "        s=MARKER_SIZE + 10,\n",
    "        marker='*',\n",
    "        edgecolors='k',\n",
    "        linewidths=1,\n",
    "        label='GEM'\n",
    "    )\n",
    "\n",
    "    # LwF\n",
    "    plt.scatter(\n",
    "        ref_runs[ref_runs['algorithm'] == 'LwF']['distance_to_first_in_plane'],\n",
    "        ref_runs[ref_runs['algorithm'] == 'LwF']['distance_to_first_null_space'],\n",
    "        c='blue',\n",
    "        s=MARKER_SIZE,\n",
    "        marker='D',\n",
    "        edgecolors='k',\n",
    "        linewidths=1,\n",
    "        label='LwF'\n",
    "    )\n",
    "\n",
    "    # add a dotted line through the origin with slope sqrt(10/118)\n",
    "    # x = np.linspace(0, 250, 100)\n",
    "    # y = np.sqrt(10/118) * x\n",
    "    # plt.plot(x, y, 'k--', alpha=.5)\n",
    "\n",
    "    # set text sizes\n",
    "    plt.xticks(fontsize=14)\n",
    "    plt.yticks(fontsize=14)\n",
    "    # plt.legend(fontsize=14)\n",
    "\n",
    "    plt.title(f'Distance to first in plane vs null space - Stability = {REFERENCE_STABILITY}')\n",
    "    plt.xlabel('range')\n",
    "    plt.ylabel('null')\n",
    "    sns.despine()\n",
    "    # equal axes\n",
    "    # plt.gca().set_aspect('equal', adjustable='box')\n",
    "    plt.xlim([0, 120])\n",
    "    plt.ylim([0, 280])\n",
    "    # plt.savefig(f'{plotdir}/distance_to_first_in_plane_vs_null_space_{REFERENCE_STABILITY}.png', bbox_inches='tight')\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAA10AAANcCAYAAACkJ8X6AAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjcuMSwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy/bCgiHAAAACXBIWXMAAA9hAAAPYQGoP6dpAAEAAElEQVR4nOzdd3RU1dfG8e9MOgmEbugl9N6lBUF67yAgVUBAFF8UG0UQFewiikgRkCYICoj0moTee6/Sa0gIkDb3/WPM/AiZNEiYBJ7PWlk657Z9k0yYfc85+5gMwzAQERERERGRFGF2dAAiIiIiIiLPMiVdIiIiIiIiKUhJl4iIiIiISApS0iUiIiIiIpKClHSJiIiIiIikICVdIiIiIiIiKUhJl4iIiIiISApS0iUiIiIiIpKClHSJiIiIiIikICVdImncyJEjMZlMjg7jmbVixQrKlSuHu7s7JpOJoKAgevToQf78+R0WkyOvX7t2bWrXru2Qaz+rzp49i8lkYvr06bY2va9Tp+ify40bNxLcN3/+/PTo0cP2esOGDZhMJjZs2GBrc/TfEhF5epR0iaQi06dPx2Qy2b7c3d3JmTMnDRs25IcffiAkJCRZrnPp0iVGjhzJ3r17k+V8jrZ582ZGjhxJUFBQsp735s2bdOjQAQ8PD3766SdmzpyJp6dnsl4DUi5+kedFYGAgjRs3JleuXLi7u5M3b16aN2/OnDlzbPvcu3ePkSNHxkh6Upu0ECOAxWLhyy+/pECBAri7u1OmTBnmzp2bqGNr164d49+5h79cXFxi7Pt///d/VKhQgcyZM5MuXTqKFy/OyJEjuXv3boz9ohNae19bt25NtvsWeRLOjg5ARGL75JNPKFCgABEREVy5coUNGzbw9ttv8+2337JkyRLKlClj23fYsGF88MEHSTr/pUuXGDVqFPnz56dcuXLJHP3Tt3nzZkaNGkWPHj3ImDFjsp13x44dhISEMHr0aOrVq2drnzx5MhaLJdmuk9T4k/v6ImnZH3/8QceOHSlXrhyDBg0iU6ZMnDlzBn9/fyZPnkznzp0Ba0IzatQogKfSW3vs2DHM5vifbT/6Xn7aMT6uoUOHMnbsWPr06UPlypVZvHgxnTt3xmQy8corryR4bO/evWO0hYaG0q9fPxo0aBCjfceOHfj5+dGzZ0/c3d3Zs2cPY8eOZc2aNfj7+8f6/r711ltUrlw5RluhQoWe4E5Fko+SLpFUqHHjxlSqVMn2+sMPP2TdunU0a9aMFi1acOTIETw8PABwdnbG2Vlv5ZRw7do1gFiJ0KNPY+2JjIzEYrHg6uqa7HEl5voiz4uRI0dSokQJtm7dGuv9Fv0edgQ3N7cE90mL7+WLFy/yzTff8MYbb/Djjz8C0Lt3b1566SWGDBlC+/btcXJyivP4+vXrx2qbNWsWAF26dInRHhgYGGtfX19f3n33XbZv307VqlVjbPPz86Ndu3ZJvieRp0HDC0XSiJdffpnhw4dz7tw52z9QYH/ux+rVq6lZsyYZM2bEy8uLokWL8tFHHwHWYRjRTwJ79uxpG4IRPZ8kICCA9u3bkzdvXtzc3MiTJw//93//x/3792Nco0ePHnh5eXHx4kVatWqFl5cX2bJl49133yUqKirGvhaLhXHjxlG6dGnc3d3Jli0bjRo1YufOnTH2mzVrFhUrVsTDw4PMmTPzyiuv8O+//8b7fRk5ciRDhgwBoECBArb7OXv2LGBNfkaPHo2vry9ubm7kz5+fjz76iLCwsHjPW7t2bbp37w5A5cqVMZlMtvkZj87DiJ6T8/XXX/P999/brnX48GEAxo8fT8mSJUmXLh2ZMmWiUqVKtmFPCcVvT3zXnzRpku36lStXZseOHfHeJ/xvWKu/vz+vv/46WbJkIUOGDHTr1o3bt2/He2x4eDgjRoygYsWKeHt74+npiZ+fH+vXr4+xX1JjPHr0KO3atSNz5sy4u7tTqVIllixZEm8sERERZM6cmZ49e8baFhwcjLu7O++++66tLb6fS1yihzHNnz+fzz77jNy5c+Pu7k7dunU5efJkjH0fndMTLTnnxZ04cYK2bdvi4+ODu7s7uXPn5pVXXuHOnTu2fUwmEwMHDmT27NkULVoUd3d3KlasiL+/f4xznTt3jgEDBlC0aFE8PDzIkiUL7du3t/u7GBQUxP/93/+RP39+3NzcyJ07N926dYsx1yksLIyPP/6YQoUK2f6WvPfeewm+95Lq1KlTVK5c2e4DjuzZswPW379s2bIBMGrUKNv7bOTIkQDs37+fHj16ULBgQdzd3fHx8aFXr17cvHnT7jVv3LhBhw4dyJAhA1myZGHQoEE8ePAgxj5x/fwf9vB7Ob4Yp02bhslkYs+ePbHO8fnnn+Pk5MTFixfjvVZyWbx4MREREQwYMMDWZjKZ6N+/PxcuXGDLli1JPuecOXPw9PSkZcuWCe4b/f2Kazh2SEgIkZGRSY5BJKXp8bhIGtK1a1c++ugjVq1aRZ8+fezuc+jQIZo1a0aZMmX45JNPcHNz4+TJk2zatAmA4sWL88knnzBixAj69u2Ln58fANWrVwesQ3Xu3btH//79yZIlC9u3b2f8+PFcuHCBP/74I8a1oqKiaNiwIS+++CJff/01a9as4ZtvvsHX15f+/fvb9nvttdeYPn06jRs3pnfv3kRGRhIQEMDWrVttPXqfffYZw4cPp0OHDvTu3Zvr168zfvx4atWqxZ49e+IcdtemTRuOHz/O3Llz+e6778iaNSuA7cNL7969mTFjBu3ateOdd95h27ZtjBkzhiNHjvDXX3/F+b0eOnQoRYsWZdKkSbbhnr6+vvH+fKZNm8aDBw/o27cvbm5uZM6cmcmTJ/PWW2/Rrl072wez/fv3s23bNjp37pxg/EkxZ84cQkJCeP311zGZTHz55Ze0adOG06dPJ+qJ+sCBA8mYMSMjR47k2LFj/Pzzz5w7d86WaNgTHBzMlClT6NSpE3369CEkJISpU6fSsGFDtm/fHmv4amJiPHToEDVq1CBXrlx88MEHeHp6Mn/+fFq1asXChQtp3bq13VhcXFxo3bo1f/75J7/88kuMD+GLFi0iLCzMNvQpoZ9LQsaOHYvZbObdd9/lzp07fPnll3Tp0oVt27YleGxyCQ8Pp2HDhoSFhfHmm2/i4+PDxYsXWbp0KUFBQXh7e9v23bhxI/PmzeOtt97Czc2NCRMm0KhRI7Zv306pUqUA61CuzZs388orr5A7d27Onj3Lzz//TO3atTl8+DDp0qUD4O7du/j5+XHkyBF69epFhQoVuHHjBkuWLOHChQtkzZoVi8VCixYtCAwMpG/fvhQvXpwDBw7w3Xffcfz4cRYtWpRs34d8+fKxdu1aLly4QO7cue3uky1bNn7++Wf69+9P69atadOmDYBtqPbq1as5ffo0PXv2xMfHh0OHDjFp0iQOHTrE1q1bY/3+d+jQgfz58zNmzBi2bt3KDz/8wO3bt/ntt98e+z7ii7FAgQK88cYbzJ49m/Lly8c4bvbs2dSuXZtcuXLFee6IiIgYiXh8MmfOHO+wyD179uDp6Unx4sVjtFepUsW2vWbNmom6FsD169dZvXo1HTt2tDtnNjIykqCgIMLDwzl48CDDhg0jffr0tus9rGfPnty9excnJyf8/Pz46quvYowaEXEoQ0RSjWnTphmAsWPHjjj38fb2NsqXL297/fHHHxsPv5W/++47AzCuX78e5zl27NhhAMa0adNibbt3716stjFjxhgmk8k4d+6cra179+4GYHzyyScx9i1fvrxRsWJF2+t169YZgPHWW2/FOq/FYjEMwzDOnj1rODk5GZ999lmM7QcOHDCcnZ1jtT/qq6++MgDjzJkzMdr37t1rAEbv3r1jtL/77rsGYKxbty7e88b18+jevbuRL18+2+szZ84YgJEhQwbj2rVrMfZt2bKlUbJkyceKPy5xXT9LlizGrVu3bO2LFy82AOPvv/+O93zR91mxYkUjPDzc1v7ll18agLF48WJb20svvWS89NJLtteRkZFGWFhYjPPdvn3beOGFF4xevXo9Vox169Y1SpcubTx48MDWZrFYjOrVqxuFCxeO915Wrlxp956bNGliFCxY0PY6MT8Xe9avX28ARvHixWPc97hx4wzAOHDggK0tX758Rvfu3WOd49HvYfT35uH346Pva3v27NljAMYff/wR736AARg7d+60tZ07d85wd3c3WrdubWuz997fsmWLARi//fabrW3EiBEGYPz555+x9o9+T8+cOdMwm81GQEBAjO0TJ040AGPTpk3xxpwUU6dONQDD1dXVqFOnjjF8+HAjICDAiIqKirHf9evXDcD4+OOPY53D3r3PnTvXAAx/f39bW/TPpUWLFjH2HTBggAEY+/bts7U9+vOP/t1Zv369re3R93J8MXbq1MnImTNnjPvavXt3nH/LHxZ97cR8JfR3qGnTpjHeS9FCQ0MNwPjggw/iPf5R48ePNwBj2bJldrdH/w5GfxUtWjTG99AwDGPTpk1G27ZtjalTpxqLFy82xowZY2TJksVwd3c3du/enaR4RFKKhheKpDFeXl7xVjGM7hFavHjxYxVbiJ4rBtbJzTdu3KB69eoYhmF3aEu/fv1ivPbz8+P06dO21wsXLsRkMvHxxx/HOjb66fGff/6JxWKhQ4cO3Lhxw/bl4+ND4cKFYw1VS6xly5YBMHjw4Bjt77zzDgD//PPPY503Lm3bto3VQ5UxY0YuXLiQqGF+T6pjx45kypTJ9jq6F/Phn0d8+vbtG6NHrH///jg7O9u+j/Y4OTnZepQsFgu3bt0iMjKSSpUqsXv37iTHeOvWLdatW0eHDh0ICQmx/S7cvHmThg0bcuLEiXiHUb388stkzZqVefPm2dpu375te5Ie7Ul/Lj179ozRk5bU73VyiO7JWrlyJffu3Yt332rVqlGxYkXb67x589KyZUtWrlxpGw788Hs/IiKCmzdvUqhQITJmzBjjZ7lw4ULKli1rt8cx+j39xx9/ULx4cYoVKxbjPf3yyy8DPPZ72p5evXqxYsUKateuTWBgIKNHj8bPz4/ChQuzefPmRJ3j4Xt/8OABN27csM0Xsvd7/MYbb8R4/eabbwLE+155Ut26dePSpUsxvnezZ8/Gw8ODtm3bxnts2bJlWb16daK+fHx84j3X/fv37c5Xc3d3t21Pijlz5pAtWza7c70ASpQowerVq1m0aBHvvfcenp6esaoXVq9enQULFtCrVy9atGjBBx98YOuh/PDDD5MUj0hK0fBCkTTm7t27tnkK9nTs2JEpU6bQu3dvPvjgA+rWrUubNm1o165dgpW0AM6fP8+IESNYsmRJrPk8jw5PiZ6f9bBMmTLFOO7UqVPkzJmTzJkzx3nNEydOYBgGhQsXtrv9cSebnzt3DrPZHKt6lY+PDxkzZuTcuXOPdd64FChQIFbb+++/z5o1a6hSpQqFChWiQYMGdO7cmRo1aiTrtcH6Qfph0clNQvOyoj36/ffy8iJHjhzxzi8DmDFjBt988w1Hjx4lIiLC1m7v+5FQjCdPnsQwDIYPH87w4cPtXu/atWtxDqVydnambdu2zJkzh7CwMNzc3Pjzzz+JiIiIkXQ96c/lSb/XyaFAgQIMHjyYb7/9ltmzZ+Pn50eLFi149dVXYwwthNg/W4AiRYpw7949rl+/jo+PD/fv32fMmDFMmzaNixcvYhiGbd+H3/unTp1K8EP+iRMnOHLkSJzDZOMrcHH37t0YH6qdnJwSHG7bsGFDGjZsyL1799i1axfz5s1j4sSJNGvWjKNHj8b7NxOsyf6oUaP4/fffY8Vmb1jeo99PX19fzGZzgu+VJ1G/fn1y5MjB7NmzqVu3LhaLhblz59KyZUvSp08f77GZMmWKUYH1SXh4eNidlxc9p+3hBDYhp0+fZsuWLQwcODDOglAZMmSwxd6yZUvmzJlDy5Yt2b17N2XLlo3z3IUKFaJly5b8+eefREVFxVvcQ+RpUNIlkoZcuHCBO3fuxFsC18PDA39/f9avX88///zDihUrmDdvHi+//DKrVq2K9x+eqKgo6tevz61bt3j//fcpVqwYnp6eXLx4kR49esTqOUuuf8QsFgsmk4nly5fbPaeXl9cTnf9pLTJr78NG8eLFOXbsGEuXLmXFihUsXLiQCRMmMGLECFtp6OQS18/j4Q/PyW3WrFn06NGDVq1aMWTIELJnz46TkxNjxozh1KlTSY4x+nfs3XffpWHDhnb3TagE9CuvvMIvv/zC8uXLadWqFfPnz6dYsWIxPqA96c8lMd/ruH7vkvMD4DfffEOPHj1YvHgxq1at4q233rLNM4prflNc3nzzTaZNm8bbb79NtWrV8Pb2tpUAT2qvucVioXTp0nz77bd2t+fJkyfOY7/++usYP4N8+fIlOplJly4dfn5++Pn5kTVrVkaNGsXy5cttRXHi0qFDBzZv3syQIUMoV64cXl5eWCwWGjVqlKh7fxp/Y5ycnOjcuTOTJ09mwoQJbNq0iUuXLvHqq68meGx4eDi3bt1K1HWyZcsW7+9njhw5WL9+PYZhxLjvy5cvA5AzZ85EXQewFa55tGphfNq0aUPXrl35/fff4026wPp7Fh4eTmhoKBkyZEj0NURSgpIukTRk5syZAHF+GI1mNpupW7cudevW5dtvv+Xzzz9n6NChrF+/nnr16sX5AeHAgQMcP36cGTNm0K1bN1v76tWrHztmX19fVq5cya1bt+Ls7fL19cUwDAoUKECRIkWSfI247idfvnxYLBZOnDgRY9L31atXCQoKIl++fEm+1uPw9PSkY8eOdOzYkfDwcNq0acNnn33Ghx9+iLu7+1NLChNy4sQJ6tSpY3t99+5dLl++TJMmTeI8ZsGCBRQsWJA///wzxn3YG06aGAULFgSsvZuP+2S+Vq1a5MiRg3nz5lGzZk3WrVvH0KFDY+2X0M/lSWXKlMluhbVz587Z7jM5lC5dmtKlSzNs2DA2b95MjRo1mDhxIp9++qltnxMnTsQ67vjx46RLl87Wi7RgwQK6d+/ON998Y9vnwYMHse7B19eXgwcPxhuTr68v+/bto27dukn+/e7WrVuMQgxJ6Tl5WHQBhehkIK44bt++zdq1axk1ahQjRoywtdv7nj287eGe3JMnT2KxWGJUFX0cCX2vunXrxjfffMPff//N8uXLyZYtW4L/HoB1LcCH39vxOXPmTLz3Ua5cOaZMmcKRI0coUaKErT26iExS1n6cM2cOvr6+sUq/xycsLAyLxZKowiCnT5/G3d39iR/ciSQHzekSSSPWrVvH6NGjKVCgQLxPBe09zYz+RzB6SEh0hahHP0xFP918+Gm9YRiMGzfuseNu27YthmHY7T2Ivk6bNm1wcnJi1KhRsXplDMOIs2xztLjuJzpZ+P7772O0Rz99b9q0aaLv43E9GrurqyslSpTAMAzbULy44n/aJk2aFGN44M8//0xkZCSNGzeO8xh7vzPbtm17rLLRYC3xXbt2bX755Rfbh+WHXb9+PcFzmM1m2rVrx99//83MmTOJjIyMMbQQEvdzeVK+vr5s3bqV8PBwW9vSpUsTXAYhsYKDg2OVxi5dujRmsznW8K8tW7bEmJv077//snjxYho0aGD7GTo5OcV6/40fPz7WEhBt27Zl3759dqt/Rh/foUMHLl68yOTJk2Ptc//+fUJDQ+O8r4IFC1KvXj3bV0JDPteuXWu3PXp+VdGiRQFs1RcT83cPYv/deNhPP/0U4/X48eMB4n2vJEZcMUYrU6YMZcqUYcqUKSxcuJBXXnklUes0JuecrpYtW+Li4sKECRNsbYZhMHHiRHLlymWrhAvWhPfRYcfR9uzZw5EjR+KsFhoUFGT3uClTpgDEqEpo7+/Cvn37WLJkCQ0aNEjU0HqRlKaeLpFUaPny5Rw9epTIyEiuXr3KunXrWL16Nfny5WPJkiXxPoX/5JNP8Pf3p2nTpuTLl49r164xYcIEcufObXt67OvrS8aMGZk4cSLp06fH09OTF198kWLFitkWnrx48SIZMmRg4cKFTzRPpU6dOnTt2pUffviBEydO2IbrBAQEUKdOHQYOHIivry+ffvopH374IWfPnqVVq1akT5+eM2fO8Ndff9G3b98Y6ys9KrpAwNChQ3nllVdwcXGhefPmlC1blu7duzNp0iSCgoJ46aWX2L59OzNmzKBVq1aJfvL7JBo0aICPjw81atTghRde4MiRI/z44480bdrUNg8jrvjtlU9OSeHh4dStW5cOHTpw7NgxJkyYQM2aNWnRokWcxzRr1ow///yT1q1b07RpU86cOcPEiRMpUaJErMnuifXTTz9Rs2ZNSpcuTZ8+fShYsCBXr15ly5YtXLhwgX379iV4jo4dOzJ+/Hg+/vhjSpcuHau8dWJ+Lk+qd+/eLFiwgEaNGtGhQwdOnTrFrFmzElx6ILHWrVvHwIEDad++PUWKFCEyMpKZM2fi5OQUa85VqVKlaNiwYYyS8UCMhyHNmjVj5syZeHt7U6JECbZs2cKaNWvIkiVLjHMNGTKEBQsW0L59e3r16kXFihW5desWS5YsYeLEiZQtW5auXbsyf/58+vXrx/r166lRowZRUVEcPXqU+fPns3LlymQr5d2yZUsKFChA8+bN8fX1JTQ0lDVr1vD3339TuXJlmjdvDlh7zEqUKMG8efMoUqQImTNnplSpUpQqVYpatWrx5ZdfEhERQa5cuVi1ahVnzpyJ85pnzpyhRYsWNGrUiC1btjBr1iw6d+6c4HC3hMQXY7Ru3brZ/h4mZmghJO+crty5c/P222/z1VdfERERQeXKlVm0aBEBAQHMnj07xtDEDz/8kBkzZtjtPZs9ezYQ99DCDRs22JZ1KFy4MOHh4QQEBPDnn39SqVKlGPfesWNHPDw8qF69OtmzZ+fw4cNMmjSJdOnSMXbs2GS5b5En9jRLJYpI/KJLd0d/ubq6Gj4+Pkb9+vWNcePGGcHBwbGOebS09Nq1a42WLVsaOXPmNFxdXY2cOXManTp1Mo4fPx7juMWLFxslSpQwnJ2dY5QcPnz4sFGvXj3Dy8vLyJo1q9GnTx9j3759scoSd+/e3fD09EwwHsOwlhX/6quvjGLFihmurq5GtmzZjMaNGxu7du2Ksd/ChQuNmjVrGp6enoanp6dRrFgx44033jCOHTuW4Pdu9OjRRq5cuQyz2Ryj7HFERIQxatQoo0CBAoaLi4uRJ08e48MPP4xRjjwuSS0Z/9VXX8U6xy+//GLUqlXLyJIli+Hm5mb4+voaQ4YMMe7cuZOo+O1JyvWJo/y0vfvcuHGj0bdvXyNTpkyGl5eX0aVLF+PmzZsx9n203LnFYjE+//xzI1++fIabm5tRvnx5Y+nSpU8c46lTp4xu3boZPj4+houLi5ErVy6jWbNmxoIFC+K9l4fjypMnjwEYn376aaztif25PCq69PajZdrtlX03DMP45ptvjFy5chlubm5GjRo1jJ07dyZbyfjTp08bvXr1Mnx9fQ13d3cjc+bMRp06dYw1a9bE2A8w3njjDWPWrFlG4cKFbT+nR8tu37592+jZs6eRNWtWw8vLy2jYsKFx9OhRu6Xvb968aQwcONDIlSuX4erqauTOndvo3r27cePGDds+4eHhxhdffGGULFnScHNzMzJlymRUrFjRGDVqVILf56SYO3eu8corrxi+vr6Gh4eH4e7ubpQoUcIYOnRorL+ZmzdvNipWrGi4urrG+L27cOGC0bp1ayNjxoyGt7e30b59e+PSpUuxfjejfy6HDx822rVrZ6RPn97IlCmTMXDgQOP+/fsxrvU4JePjizHa5cuXDScnJ6NIkSKP+y17YlFRUbb3vaurq1GyZElj1qxZsfaLXlrk0b9nUVFRRq5cuYwKFSrEeY2TJ08a3bp1MwoWLGj7uZYsWdL4+OOPjbt378bYd9y4cUaVKlWMzJkzG87OzkaOHDmMV1991Thx4kSy3K9IcjAZRgrOsBYRkVRv+vTp9OzZkx07dmgh0WeQyWTijTfe4Mcff3R0KJIMbty4QY4cORgxYkScFT5FJPXRIFcRERGRNGL69OlERUXRtWtXR4ciIkmgOV0iIiIiqdy6des4fPgwn332Ga1atXriSoki8nQp6RIRERFJ5T755BPbkgDR1RJFJO3QnC4REREREZEUpDldIiIiIiIiKUhJl4iIiIiISApS0pVEhmEQHBwca+V6ERERERERe5R0JVFISAje3t6EhIQ4OhQREREREUkDlHSJiIiIiIikICVdIiIiIiIiKUhJl4iIiIiISApS0iUiIiIiIpKClHSJiIiIiIikICVdIiIiIiIiKUhJl4iIiIiISApS0iUiIiIiIpKClHSJiIiIiIikICVdIiIiIiIiKUhJl4iIiIiISApS0iUiIiIiIpKClHSJiIiIiIikICVdIiIiIiIiKUhJl4iIiIiISApS0iUiIiIiIpKClHSJiIiIiIikICVdIiIiIiIiKUhJl4iIiIiISApS0iUiIiIiIpKClHSJiIiIiIikICVdIiIiIiIiKUhJl4iIiIiISApS0iUiIiIiIpKClHSJiIiIiIikICVdIiIiIiIiKUhJl4iIiIiISApS0iUiIiIiIpKClHSJiIiIiIikICVdIiIiIiIiKUhJl4iIiIiISApS0iUiIiIiIpKClHSJiIiIiIikICVdIiIiIiLy2B48eMDSpUsdHUaqpqRLREREREQeS0REBB06dKB58+b8+OOPjg4n1XJ2dAAiIiIiIpK2hITA1q1RvPded/bu/Ruz2Z2ffy7FggWQOTOUKwcVK0KVKpAtm6OjdTyTYRiGo4NIS4KDg/H29ubOnTtkyJDB0eGIiIiIiDwVhgFbt8KECTBvnkFERD9gEtZ+nEVA01jHmEzQuDEMGACNGoGT09ONObXQ8EIREREREYnX3r1Qvbr1a9Ysg4iI97EmXCZgFvYSLrAmasuWQbNmUKwYLF/+9GJOTZR0iYiIiIiIXeHhMHIkVK5s7eWy+hz46r//nwR0TNS5Tp6EJk2gVy8ICkruSFM3JV0iIiIiIhJLUBDUqwejRkFk5MNbwv/77zdA7ySfd9o0qFQJTp9+8hjTCs3pSiLN6RIRERGRZ11QENStC7t3x7XHZqA6AM7OUKgQeHlBWBicOAEPHiR8jZw5wd8ffH2TKehUTD1dIiIiIiJiExEBLVs+mnBtBEJtrzJlqs7gwdYhhyEhcOQI7NgB+/dbX+/fD59+CnnyxH2dS5egfn24cSOl7iT1UNIlIiIiIiI2Y8ZYe6D+ZxVQH2iI2RzCsGFw4QJ88w28+CK4u8c83tkZSpeGoUOtQwgnTrT2gtlz5oy1suGzTsMLk0jDC0VERETkWbVvn3W+1f/mcG3CmnDdx8OjHRs2/E6VKkmv+37uHLRrBzt32t8+fz60b/+YQacB6ukSEREREREMw9rr9L+Eaw/QBLiPm1sj9u6dHW/CZbFY4tyWLx+sXWtdLNmeN9+Ee/ceN/LUT0mXiIiIiIiwYwds3hz96ijQAAgG/FizZiFFirjGeezFixfJmDEjAQEBce6TIQMsWQKZM8fedvUqzJv3BMGnckq6RERERESECROi/+8sUA+4AVTknXeWUrNmuniPXbBgASEhIcycOTPe/V54AX780f62n36y9rY9izSnK4k0p0tEREREnjWhoZA1a3Sp9z1Ye7mykz79Ri5fzoqnZ/zH16xZk02bNpElSxauXLmCs7NznPsaBlSsCHv2xN62bx+UKRP3dQzD4Pr162TPnj0Rd5V6qKdLREREROQ5t3Pnw2trlQf8gVW89lrCCdeVK1fYvHkzRYoU4ebNm/EOMQQwmeKuWBgYGP+1vvjiC/LmycvpNLayctwpqIiIiIiIPBc2bQoBTgAV/mspDsSuKHj16lX27dsXo23t2rWYTCYqVKjAlStX+Pnnn4mIiIixj6+vL74PrYLcrh306RM7jl274o4xODiYsWO+ICw8jM8++4ypU6cm8u4cT8MLk0jDC0VERETkWXL//n3y52/MtWs7gSXAywCYzdaFjtM9NJ3r9ddfZ9KkSbHOkSdPHurWrcvOnTs5ePBgrO3Vq1dn06ZNMdoKF4aTJ2PuV778o4sy/89nn33GxyNGUtC7JqeDAzl+/BgFCxZMyq06jHq6RERERESeU+Hh4bRv355r1zYC6QFv2zZf35gJF1iH9wUFBTF//nxy5cpF5cqVcXZ2xsPDA4CKFStSrFgxDMPgwIEDHD9+nKpVqzJnzpxY1y5VKnbSdeWK/TiDg4P56suvyZ+hKiWzNubivT1pqrdLc7pERERERJ5DUVFRdOvWjX/++Qez2R1YClS0bbc3lytjxoz8/vvvTJ06lVu3brFx40YiIiJwcrKu32UymTCbzWzZsoUTJ04wfPhwAgICyJcvX6xzeXnFPn9YmP1Yx48fz927dymWuS7OZlcKZajNjBm/pZm5XUq6RERERESeM4Zh0K9fP+bNm4eLiwtlyvwJ1Iqxz/8Ka8RkMpno1asXu3fvxtnZmd2PjAc8evQoly9fZt26dXzyySdxVjK0d35XO0uBPdzL5eGSEYCCGavj5uzJZ599ltCtpgpKukREREREniOGYTBkyBCmTJmC2Wxm9uzZFC3aONZ+p09DeHjc58mRIwdBQUHkyJGDyMhITpw4wYMHD/Dx8cEwDNKnTx9vHEeOxG7LkiV228O9XNHSWm+Xki4RERERkedIVFQUJ06cAGDy5Mm0b9+esmVj7xceDnZqYtgsXbqUiIgIvL29Wb58OZs2bWLp0qUYhoGHhwcLFy6M89jQUPtJV7lyMV/b6+WKlpZ6u5R0iYiIiIg8R5ydnVmwYAHLli2jV69egHWxYnsWL477PAsWLMBsNrNhwwayZcvG6tWrqVSpEmvWrCEyMpJ58+YRV6H0ZcvAYond/mgc9nq5bPeRhnq7lHSJiIiIiDwHduzYYUuCXFxcaNz4f0MKK1eG/2phxDBpEjyy5BYAoaGhrFixAovFQs+ePdm1axf16tVjzZo1fP7551gsFk6fPm23fDzAhAn2Y6xW7X//H18vV7S00tulpEtERERE5Bm3YMECqlatSr9+/YiKioq1PVMmaNEi9nFXrthPkB48eEDZsmVZsGABv/zyC+n+qy3v5OTEBx98QGBgINWqVbN7rbVrYcOG2OcsUgSqVPnf6/h6uaKlld4uJV0iIiIiIs+wFStW0LlzZywWCxaLBbPZfgowYID94z/8MPZ6WlmyZGHr1q20bdvW7jFVq1Zl8+bNlHtkklZICLz2mv3r9O9vXZAZEtfLFS0t9HYp6RIREREReUYFBATQpk0bIiIi6NixIxMnTsRkMtnd9+WXoUSJ2O3370PbtnD79pPFEhEBXbvCuXOxt6VPD927/+91Ynq5oqWF3i6TEdfsNrErODgYb29v7ty5Q4YMGRwdjoiIiIiIXbt27eLll18mODiYJk2a8Ndff+FqbyGsh6xbB3XjyHPKlYO//4bcuZMeS3CwNeFassT+9vHjYeDA6H2DyZsnH07h3hTOVMv+AY+INCLYfXU+PXv2YOrUqUkPMIUp6UoiJV0iIiIiktodPnyYWrVqcfPmTV566SWWL1+Oh4dHoo4dMAB+/tn+Nm9v+O476NED4ugwi2X1aujdG86ft7+9dm3rPK/ooYX79u2jUqVKREZGJu4CD6lQviK7du9M8nEpTUlXEinpEhEREZHUbsGCBbzyyitUqFCBNWvWJOlz69271oIW9tbRilaihDU569ABsmWLvT0kBJYutRbhCAyM+zyZMsGuXVCgQMz2+/fv2y3CkRB3d3ecnZ2TfFxKU9KVREq6RERERCQtWLVqFRUrViRLlixJPvb8efDzi7t36mH58kHp0uDpCWFhcPQoHDsGCWUZnp7WHq4XX0xyeGmOkq4kUtIlIiIiIqnRjRs3CA8PJ2fOnMlyvvPnoX59OH48WU4XQ+bM1gWSn4eEC1S9UEREREQkzQsODqZRo0b4+flx5syZZDln3rywdau1AEZyql0bdu58fhIuUNIlIiIiIpKm3bt3j+bNm7Nr1y6Cg4N58OBBsp07Uyb47TdYvBhy5Xqyc2XIAD/9ZB1S+Ogcrmedki4RERERkTQqPDycdu3a4e/vT4YMGVi5ciXFixdP9uu0aAGnTsGsWVC9etKOLVYMxo2zrs81YMD/qhQ+TzSnK4k0p0tEREREUoOoqCg6derEH3/8gYeHB6tWraJmzZpP5dpHj1qrEu7aBbt3w5Ur1iIarq7W+VrlykHFilC1KlSqlPjy8s8qJV1JpKRLRERERBzNMAz69OnD1KlTcXFx4e+//6Zhw4aODkvi8Bx27omIiIiIpG23b99m06ZNmM1m5s6dq4QrlUt9K4eJiIiIiEi8MmfOjL+/P5s2baJVq1aODkcSoOGFSaThhSIiIiLiKCdOnKBw4cKODkOSSMMLRURERETSgKlTp1KsWDEmT57s6FCeOsMwOLvhLLObzObulbuODifJNLxQRERERCSVmz9/Pn369MEwDE6cOOHocJ4awzA4vfo0/qP9OR94HoDNX2+mwdcNHBxZ0ijpEhERERFJxZYtW0aXLl0wDIO+ffvyxRdfODqkp8ISZWFGnRmcD7AmW06uTpTvXZ4qb1ZxcGRJp6RLRERERCSV2rhxI23btiUyMpLOnTszYcIETM/woleGYdjuz+xkJkvRLFzaeYmKr1ekxpAapM+Z3sERPh4V0kgiFdIQERERkadh586dvPzyy4SEhNC8eXMWLlyIi4uLo8NKEZYoC4fmHSLg8wDazG6DT1kfAO5euYvJbMIzu6eDI3wy6ukSEREREUmFli5dSkhICHXq1GH+/PnPZMIVFRHFgdkHCPg8gFsnbgGw5ZsttP6tNQBePl6ODC/ZKOkSEREREUmFPv74Y3Lnzk3Hjh1xd3d3dDjJKjIskr3T97Jp7CaCzgYB4JHZg6qDq1JlYNqbs5UQDS9MIg0vFBEREZGUcvXqVTJmzIibm5ujQ0kxhmEw5cUpXNpxCQDP7J5Ue7calftXxtXL1cHRpQz1dImIiIiIpALXr1+nTp065M6dm7/++gtPz7Q9j+lh4aHhOLs7Y3YyYzKZKNmxJCGXQqjxXg0q9K6AS7pnb+jkw9TTlUTq6RIRERGR5Hbnzh1efvlldu/eTe7cudm8eTN58uRxdFhPLCw4jO0/bmfLt1to+nNTSrYvCUDkg0gwgbPb89EH9HzcpYiIiIhIKnXv3j2aNWvG7t27yZYtG2vWrEnzCdf9W/fZOm4r23/YzoOgBwAcnHPQlnQ5uz9facjzdbciIiIiIqlIWFgYbdq0ITAwEG9vb1atWkXRokUdHdZjC70WypbvtrDjpx2Eh4QDkLV4VvyG+lGqYykHR+c4SrpERERERBwgMjKSLl26sHLlStKlS8eyZcsoV66co8N6IgteWcDZ9WcBeKHMC/gN86NE2xKYzM/ugs6JoaRLRERERMQBTp48yerVq3F1dWXx4sVUr17d0SEl2Z1/7+CWwQ13b2tJ+2qDqxEeEk6t4bUo0qzIc59sRVMhjSRSIQ0RERERSS579uzhwoULNG/e3NGhJMnt07cJGBPAvhn7qDW8Fi8NfwmwloMHMJmUbD1MPV0iIiIiIk/RlStX8PHxAaB8+fKUL1/ewREl3o1jNwj8PJD9s/djRFkTrKv7rtq2K9myz+zoAEREREREnhfffPMNxYoVY/PmzY4OJUmuHbzGwk4L+an4T+z7bR9GlIFvQ196BvSkw4IOjg4v1VNPl4iIiIjIUzB58mTeffddADZt2pSm5nBt/mozB38/CEDRFkXxG+pHriq5HBxV2qGkS0REREQSJdISScC5ALZd3Mauy7vYf3U/t+7fIiIqAndnd3JlyEUFnwpUzFmRl/K9RPFsxR0dcqrx+++/8/rrrwPw/vvvM2TIEAdHFL8L2y6QLms6MvtmBqDmRzWJuB+B30d++JTzcXB0aY8KaSSRCmmIiIjI8+bK3StM2T2FX3b9woXgC4k+rnqe6gyoNIB2Jdrh5uyWghGmbkuXLqV169ZERkbSv39/fvrpp1Q79+mc/zn8R/tzes1pSncpTZtZbRwd0jNBPV0iIs+B69dh1y44fRru3QMnJ/D2hlKloEwZcHd3dIQikhpFREUwNnAsnwV8RlhUWJKP3/zvZjb/u5kP137IL81+oXHhxikQZeq2fv162rVrR2RkJK+++io//vhjqku4DMPgzNoz+I/255z/OQDMzmZc0rlgGEaqizctcmghjYsXL/L999/ToEED8ubNi6urKz4+PrRt25Zt27bF2n/kyJGYTKY4v86ePWv3OitXruSll14iffr0ZMiQgTp16rB27doUvjsREcc6eBAGDoR8+SB7dmjcGN54A4YMgcGD4bXX4MUXIX16qFwZxo2DoCBHRy0iqcWBqwd4ccqLjNgw4rESrof9G/wvTeY0odfiXgSHBSdThGnD+PHjCQsLo2XLlkybNg2zOXXVsTu95jRTq01lZv2ZnPM/h5OrExVfr8ibJ96k+aTmSriSiUN7usaPH88XX3yBr68vDRo0IFu2bJw4cYJFixaxaNEi5syZQ8eOHWMd1717d/Lnzx+rPWPGjLHaZs2aRdeuXcmWLRs9evQAYN68edSvX5/58+fTrl27ZL4rERHH2rwZPvoINm5M3P6RkbBzp/Xro4+ge3f45BPImjVl4xSR1GvFyRW0mdeG+5H3k/W80/ZOY/vF7azquoqc6XMm67lTqzlz5vDVV18xZMgQnJ1T3yCzi9svcnHbRZzdnanQtwI1htQgQ25NoUluDp3T9eeff5IlSxZeeumlGO0BAQHUrVsXLy8vLl++jJubdQzwyJEjGTVqFOvXr6d27doJnv/27dsULFgQZ2dn9uzZQ+7cuQG4cOGCbT2E06dPkz59+kTHrDldIpJa3bsHQ4dae6ye9C979uwwcSK0bp08sYlI2rHi5ApazG1BhCUi3v1czC6UfqE0RbIUwdXJlZCwEPZd3cfp26cTvEahzIUI6BmAj9ezWZDh1q1bZMqUKdX1ElmiLBz+4zCeL3hSoE4BAMKCwwgcG8iLb72Il4+XgyN8djm0f7NNmzaxEi4APz8/6tSpw+3btzlw4MBjn/+PP/4gKCiIN99805ZwAeTOnZuBAwdy48YN/vrrr8c+v4hIanH2LFSsCN9//+QJF8C1a9CmDfTrB1FRT34+EUkb9l3ZR5t5beJMuEyYaFq4Kf90/oeQD0PY1XcXc9vOZUarGfzZ8U9OvXWKG0Nu8FOTnyiZrWSc1zl56ySNZzfmQeSDlLoVh/n333+pUKECQ4YMIbXUq7NEWtj32z4mlJzAwk4LWfPeGltsbhncqPt5XSVcKSx1DSp9iIuLC4Ddblh/f3+++OILvvrqKxYtWsTdu3ftnmPDhg0ANGjQINa2hg0bArAxseNvRERSqZMnoWZNOHo0+c/9yy/QubN1CKKIPNvCo8LptqhbnEMKS2Qrwbbe21jaeSlNCjeJsxphlnRZGFB5AAf6H+CXZr/g5Wr/w/zeK3sZuWFkcoWfKly7do369etz7tw5/v77b+7cuePQeKLCo9g9ZTc/Fv2RRd0XcfPYTdwzuVOkeREskRaHxva8SX0DS4Hz58+zZs0acuTIQenSpWNt//jjj2O8zpgxI+PGjaNbt24x2k+cOAFA4cKFY50jui16n7iEhYURFva/yaPBwc/X5E8RSd2uXYP69eHixbj38fCADh3Azw8qVIBs2ay9V6dOWSsaLl0K/v5xHz9/PmTMaE3AROTZ9an/p+y/ut/utp7levJz05+TVPbdZDLRt2JfGvg2oOmcphy+fjjWPl9t/orWxVrzYu4XHzvu1CIoKIiGDRty7Ngx8uTJw+rVq+3WG3haDi88zMr/W0nwv9bPrumypaPaO9Wo3L8ybhme3/L9jpLqeroiIiLo2rUrYWFhfPHFFzg5Odm2lS1bll9//ZXTp09z//59zpw5w/jx4zGZTPTo0YMlS5bEOFf00wVvb+9Y14mej5XQE4gxY8bg7e1t+8qTJ8+T3qKISLIwDOjb1zq00B43N/j0U7h0CaZPt1YrLF8ecue2VjR8+WVrJcONG+HAAWjePO5rTZoECxakxF2ISGpw/s55xgaOtbvttfKvMbXF1DgTLovFwrlz5+I8d/6M+dnQfQPFshaLfaxhYdCKQY8XdCoSGhpK06ZN2bt3L9mzZ2fNmjXkzZs33mMMw2D58uVEpuBQguB/g/HK4UWDbxsw6Mwgar5fUwmXg6SqxZEtFgtdu3Zlzpw59OnTh0mTJiXquLVr11K/fn1KlSrF/v3/e0JTpEgRTpw4QURERKxhihEREbi6ulKmTBn27dsX57nt9XTlyZNHhTRExOHmzIEuXexvK1cO5s6FYrE/48TJMGDWLHj9dbhvZ3RRtmxw6JD1vyLybBm6diifB34eq73MC2XY0WcHrk6ucR47bdo0Xn/9dc6dO0eOHDni3O/A1QNUnFTR7nyx7b23UzlX5ccL3sHCwsJo3ry5rWdr48aNlClTJsHjNm/eTI0aNZgxY0as0VpJjiEkjB0TduCe0Z1Kr1cCwLAY7J+1n5IdSuLsnioHtz1XUk1Pl8VioVevXsyZM4dXX32ViRMnJvrYunXr4uvry4EDB2IM/4vu4bLXmxW9n71esIe5ubmRIUOGGF8iIo4WGgpvv21/24svWnuv4kq4tm7dyvXr12O1m0zQtSusWmUdkvio69dh2LDHj1lEUqewyDCm7JkSq93J5MT0ltPjTbjAWhI9IiIiweJkpV8ozfBaw+1um7BzQuIDTmU2bNjAmjVr8PT0ZPny5YlKuAAW/Dd84I8//njsa9+/fZ+Nn2zk+3zfs/aDtWwYsYGIe9ak1mQ2UbZbWSVcqUSqSLosFgs9e/ZkxowZdOrUienTpyd54bis/y0oc+/ePVtbfPO24pvvJSKS2s2da02CHpUtGyxZAnE9H7p48SIvvfQSXbt2jfPcNWvC5Mn2t/32G9y69RgBi0iqtfr0aq6FXovV3rp4a8rnKB/vsbdu3WL9+vUAzJ8/P8FrDa42GG+32A+85x2cl2YrGTZs2JDZs2ezePFiqlatmqhjDMOwfb9WrlxJSEhIkq5578Y91g5dy7j849jw8QYe3H5AlqJZqP9VfZxcnRI+gTx1Dk+6ohOu3377jY4dOzJz5swY87gSIzQ0lEOHDuHp6WlLvgBbOfpVq1bFOmblypUx9hERSSsMA376yf62CROsa2zFZezYsURGRrFy5Uq2bt0a536dO0OrVrHbHzyAadOSFq+IpG5bL9j/W9C/Uv8Ej12yZAlR/60rERAQwI0bN+Ld39PVkx7lesRqvx95nwNXH3+ZoKfNMIwYiVKnTp2oW7duoo/ftWsXF/+rgBQREcE///yT6GMPzDnA9/m/J/DzQMKCw8heKjttf2/LgEMDKNutLGZnh3+8FzscOqcrekjhjBkzaN++PXPmzIlzpe6QkBAuX75MkSJFYrTfv3+f1157jblz59KzZ09+/fVX27bbt29ToEABXFxctDiyiDwzDh2CUqVit1eqBDt2xH3cxYsXKViwIA3rd2PXnrWULl2EFStWxLn/0aNQvHjs9rJlYe/epMctIqlT49mNWXEy5t+C9K7pCfogCLPpfx/gly9fzvHjx2PsN2vWLPbs2UNUVBQmk4mePXvGGl7XqFEjihYtanu99vRa6s2sFyuOn5v+TL9K/ZLjllLc0KFD+euvv1i9ejW5cuWKd9+lS5fGKvZ28OBBtm/fTlRUFM7OzhQpUoQaNWrE2Kdq1ar06tULsCZ50QstXz1wlYllJuJT3odaw2tRrGUxTObUtQizxObQQZ6ffPIJM2bMwMvLiyJFivDpp5/G2qdVq1aUK1eOmzdvUqxYMSpXrkzx4sXx8fHh6tWrrFmzhgsXLlC6dGm++uqrGMdmypSJH3/8ka5du1KhQgU6duwIwLx587h58ybz5s1LUsIlIpIabNliv71/Ag+lx44di6urB3Ve6kD2bHmYOv1jtm7dGudwmGLFrBUO162L2X7ggHVOmafnYwQvIqnOviuxC4qVz1E+RsIFMHz4cHbt2gVgG5VkGAYWi3W9J7PZzIwZM2K1jx49mmEPTQitkKOC3Tj2Xtn7ZDfylHz55Zd8/rm16Mi6deviHa4NcODAASb/N2bbycnJljxF9xBGRkZy/PhxW0IbXc0wMjKS1nVaEzg2ELOTmaYTmgLwQukX6L29Nzkr5bSdS1I/hyZdZ/+rc3z37l0+++wzu/vkz5+fcuXKkTlzZgYMGMD27dtZtmwZt2/fxsPDg+LFi/PWW28xcOBAPOzM/H711VfJmjUrn3/+OdOmTcNkMlGxYkWGDRtGvXqxn7KIiKR2/33miaVFi7iPuXjxIpMmTaJRgx54eHhSoXwdlq+azsiRI+Pt7WrePHbSZbFYe7oeeSgrImnUzfs3Y7UVyxK7Es/q1at57bXX+Ouvv2wJw8MebnNycsLZ2Zkvv/ySQYNiloTP5JGJFzxf4Gro1Rjttx/cftxbeGomTpzI+++/D8AXX3yRYMIF8OGHH1KwYEFee+01wsLC7JaIj26L/r6NfW8svhd9GV94PEaUgdnZTK3htUifw9pZkKty/L1rkvo4NOmaPn0606dPT9S+GTJk4Mcff3ys6zRq1IhGjRo91rEiIqnNATvTHvLlg4emtMYS3ctVu1Y7AMxmJxo36JFgb1elSvbPt3+/ki6RZ0VEVOwS7vbW5MqUKRMLFy5kypQpvPnmm0RFRdlNIMxmM3ny5GHBggVUrFjR7jXtnT88Kvwxon965syZw4ABAwD46KOPeO+99xJ9bMeOHalSpQodO3Zk586d2JvdYzabKe1TmjdKv8GlsZfYZ7H2QBasX5Baw/6XcEnapBqSIiJpTFBQ7DZf37j3f7SXK1piersKFbJ/zgTWlRd7QkJg7VrrxLtdu+DkSes4TZPJWm6yRAmoWNFa8/+ll8DFxdERy3PCzdktVuXAu+F37e5rMpno06cP5cuXp3Ll2Otqmc1mfH192bVrV7xTOELDQ2PH4ZR6F+1dsmQJ3bp1wzAMBg4caHdKTEIKFCjApk2byJMnD1evXo21vYRRglaXWtkKbBRpVgS/oX7krpr7ieMXx1PSJSKSxtgZ1UMcNYiA2L1c0RLT2xXXee083Ja4HDxoLSs5cybctf9BlsuX4dgxiF7nyMcH+vSBvn0htz5wScry8fLhbNDZGG0Hrx2M95hz587ZbbdYLLbpI3G5FHLJ7pBGHy+feI9zlMjISD744AOioqLo1q0b48aNe+y5VKdPn46RcLngQgTWnsZTximc0jlRpFER/Ib5kaN83AtNS9qjmpIiImlMunSx2+Kq0hzdy/Vy7Y4xermiVSj/MjlzFmDkyJF2j4/rvPZikEcEBUHPnlC6NPz8c9wJlz1XrsDo0dYuzFGjICL28C+R5FLeJ/ZaXPuu7ot3uN+CBQtsFadNJhMmk8n2OiIigmXLlsV57K5L9iem2osjNXB2dmb16tUMHjyYqVOnJnkt2YctXLjQOvySPLzKq/R07mnbFu4UztmWZ+mwsIMSrmeQki4RkTQmf/7YbQcPQridz0dx9XJFM5vNNGnYM851u3bvth9DgQJJCPh5tHIllCwJiZy3HKfwcBg5EqpUsf6QRVJApZyxJ2+GR4Wz+Ohiu/uHhYWxZMkS23yu9957j1OnTlGtWjXbPuPHj4/zevMOzUt0HI50//592//nypWLb775Js6ljRLDMAzWTllLV0tXXuM1ClGIXEYuvhv6Hc7OzkRFRfHHP38QoYcszyQlXSIiaYy9eenh4bBzZ8y2hHq5opUvVyfO3q7AwMTHIP+ZPBkaN4ZLl5LvnHv3QvXqsGFD8p1T5D9+ef3stv+0w/4q7GvXruXevXtkzZqV1atXM3bsWAoUKMD69etp3749AFu2bOHBgwexjr0Weo0/Dv8Rqz2LRxaKZY1dMdFRjh8/TuHChZk7d+4Tn8swDE6uOMkvVX6h1plaFKAAJmcTFfpU4M3jb/L2p2+zc+dOChUqRHBwcLwL10vapaRLRCSNiaui4JQpMV8n1MsVLa7ertBQmDMn9v7ZskGePEmN+jnx66/WeVh2KpPFkD491KoFHTpA+/ZQuTK4JVBEICQEmjSBTZuSL14RoEbeGhTOXDhW+8ZzG1l6fGms9jx58tCvXz8OHz4cY/kdJycnW9l4k8lEWFhYrGOHrh1qd9hiz3I9cTI7PcltJJvz589Tr149Ll68yNdff223QmNSnFh2gtmNZ3N151VwhlK9SjHo1CCaT2pOpoKZAChbtix79+7lu+++o7i9VeklzTMZ9mpWSpyCg4Px9vbmzp07ZMiQwdHhiMhz6N49yJEDgoNjtru7w4kT1roLFy9epGDBgjRq0IPGDbsneE6LxcJnX3SjdOkitkqGP/wAjyyvA1inKf36a3LcyTPG3x/q1LEuZGaPhwe8+qo1KatQAR6dFxIRYe3J+vlnWLw47vNkymSt2a8CG5KMvtvyHYNXDY7VnsMrB4cGHCKTR6YEz3Hv3j2yZMli6+H6888/ad26tW37ypMraTTb/hI+J988iW/meMqwPiVXr17Fz8+PEydOULRoUfz9/cmePXuSzmGJshB0NojMvpltrydXmkz+l/NT/d3qKv3+nFLSlURKukQkNRg0yJoUPapRI1i2DN5+exA//PADNWu0xN09cVUvzpw5xKnT+9m2bRvZs1ehdGn7tR+2b7d2zMhDQkOhTBk4fdr+9oYNYdIkyJs3cefbudOa3cY1j6txY/jnH2u5eZFkcPv+bQr+UJCgB0GxttXKV4vlXZaTzuV/f0uCg4PZt29fjP22bdvGkCFDAGvxiUaNGtnWsjp19xSDdg8iOPKRp0VAi6ItWPyK/fljT9Pt27epXbs2+/fvJ1++fAQEBJAnCd36lkgLB38/SMBnAYSFhPHWqbdwdrPOATMsBiaz3q/PMyVdSaSkS0RSg6NHIa4RKF9/DbduDWX+/NjzJhLi5OTEhAnTGDasKlu2xN5eqZJ1mSl5xNtvw7hx9reNHg1DhyY9QQoLg1697I/xBGuRju4J92KKJNave37ltSWv2d1WPU915rWbR+4M1h7WDz/8kLFjx8baz9nZOfZwvAJAe8DO8x8vVy8O9D9A/oz5nyz4J3T37l3q16/P1q1b8fHxISAggEJxLVT4iKiIKPbP3E/A5wHcPnUbAPeM7nRd3ZWclXKmZNiShijpSiIlXSKSWvToATNm2N/29dcweHDSP+cHBUGbNrB+vf3tq1ZB/fpJO+cz79IlyJfP/uJlI0ZYS77HYeDAgZQoUYIBAwbY3yEqCtq1g0WLYm/Lk8fas/YE1dREHmYYBk3nNGX5yeV2t2dwy8C3Db6lR7kehN4NpX///syJ66EAgDtQF4inZ/znpj/Tr1K/J4o7Ofzwww8MGjSITJkysXHjRkqXLp3gMZFhkeydtpfAsYHcOWddMd4jiwfVBlej8huVcfd2T+mwJQ1R0pVESrpEJLW4fdtalfzyZfvbmzSxjmjLlStx51uxAnr3hosX7W/v3dtamE8eMWqUtaz7o2rUgI0bwcl+cYAtW7ZQvXp1PD09OXv2LFmzZrV//lu3rD/oK1dib1u8GFq0ePzYRR5xKeQSlSdX5lJI3NU383rnpW+FvjQp3IS9q/fyRr83iIiIsPZwuQE5gVJAGcAl7mu1LtaahR0WPvZCw8nJMAyGDx9OixYtqFKlSqKOubz7MpMqTgLA8wVPqg+pTqXXK+Hq5ZqSoUoapaQriZR0iUhqsmwZNG0a9/Z06aBzZ+jTx1q74dFOkZAQWL4cJkyw5gdxyZ/fWrXc2zs5on6GREVZ52k9Wh7ebIbDh6Fo0TgPbdSoEWfOnOHixYu8+eabjBkzJu7rLFhgrXIY+yTWH6BIMjp8/TC1p9fm+r3rCe7r6uRKLs9cXDhzgQhzBCRcbwOAlwu8zD+d/8Hd2XG9QRaLBYvFkui1t8LvhvPvln/xrf+/gh9L+izhhTIvUKF3BVw84skw5bmnpCuJlHSJSGozbpx1SlFCPDygbFlryfeoKDh1Co4fT7i6edasEBAAxVLPEjqpx+7d9hcta9nS/pDA/0T3cs2dM5c9e/fy008/xt/bFRUFhQrB2bMx252drZmzu4YxSfI6duMYDWY14Pyd88l+7pZFWzK37Vw8XDyS/dyJZRgGb7zxBpcuXeL333/HPZ730IM7D9j+43a2freV8JBw3jr9Fhly6TOgJI3W6RIRSeMGDYLvv094v/v3YetW+Ptvaw/ZsWMJJ1w5cljndynhisOjK1JHe81+MYJoo0aNokSJErRt247B/2ct0/3NN9/EfYCTk7WoxqMiI+HAgcRGK5JoRbMWZe/re+lapmuynTOdSzp+aPQDf3b806EJF8BHH33Ezz//zJIlSwiMYxX4ezfvsX7Eer7P9z3rh63n/s37eOfzJvhC7AqMIglR0iUi8gwYNMiaTPn4JN8569WDbdugVKnkO+czZ9cu++3Vq8d5yJYtW1i5ciXDhw3HbDaTNWtW3nhjIOPHj+fGjRtxXyuuc8YVg8gTyuSRid9a/8aSV5ZQMFPBJzpXg4IN2N9vP2+++CZmk2M/fo4dO9ZWeXHixIkxFngGuH/7PqvfX824/OPwH+1P2J0wspXIRps5bXjjyBvkflFr5EnSKekSEXlGNGsGhw5Z1999EhkywMSJ1kqFSVii5vn06HA/sM7xypIlzkMe7uWKlqjervLlEx+DSDJqXrQ5xwceZ2mnpTQp3AQTiSt84YYbpp0m+AmG5R+WKhY/njBhAh9++CEAX331FX379o21jyXSwvbx2wm/G45POR/aL2hP/wP9Kd2pNGYnfXSWx6M6syIiz5DMmWHmTPjgA2txjN9+s7/AsT1FisCAAdalnzJmTNEwnx0PHsRuy5w5zt2je7nmzpmL2fy/D28P93a988479ud2ZYqjQoG9GESSmZPZiaZFmtK0SFNu3rvJzks72XV5F7sv7WaV/ypy5MpBwbwFyZk+JxVzVqRijoqU8ynHwX0HGTBgQOy1uxxg1qxZvPHGGwAMGzaMd999F4Cgc0Ec+fMI1f6vGgCe2Typ/1V9vPN6U6RZkVRRXVHSPhXSSCIV0hCRtOTuXWtVwp07raPQTp+Ge/esU4QyZIDSpa11IKpWtS58rM8WSVS3LqxbF7OtZEk4eNDu7o0aNeLff/9l7559MZIugBs3buBbqGDclQwjI8HFTnW0d96xLswmInG6desWBQsW5M6dO7z55puMGzeO26dvEzgmkH0z9mGJtNAzsCd5a+R1dKjyjFJPl4jIM8zLy1pSPr6y8vIE7HUJnjljrTb4yPpccfVyRUuwt+vECfsxqI6/SIIyZ87M8uXLmTt3LsP6DWNR90UcmHMAI8ra91Dg5QIq+S4pSgNTRUREHlfp0rHb7t2Do0djNduby/WoeOd2xVUwo0yZRIUq8iQGDx7MX3/95egwkuzhYY1li5Sl1tVa/FzqZ/bP3I8RZVCocSF6bepFt7XdyFEhhwMjlWedki4REZHHZW+NLoDFi2O8fLRiYVzirWT4yDkTjEEkmVy+fJnvv/8+/gW8U6G9e/dSokQJ9uzZA4BbBjcubr8IBhRrVYw+O/rQZVkX8lRXxSBJeZrTlUSa0yUiIjbXrlnr9D/6T2mePNYJdM7WUfzxzeV6lN25XZcuWasiRkXF3Dl3bjh/XpPxJEVNmDDBVoDiwoUL5MqVy8ERJezYsWO0q9oO3yBfohpH8feyvwE4veY0ntk9eaHMCw6OUJ436ukSERF5XNmzQ+PGsdv//Rd+/BFIfC9XNLu9XR9+GDvhAmupSSVcksLmz5+PyWTCZDKliSGG2+Zt45uy39AuqB3lKc+HTT+0bStYr6ASLnEI9XQlkXq6REQkhn/+sS6S9igPD9i7l15jxzJt2rTHOvUPP/zAmwUKQPPmsTeazdaiHXlVbU2Sj8ViifH6xo0b5MiRA4vFgtlspnr16mzcuDHGPtEJmSMZhsHp1adZM2INV7ZdAcCChZJdSlJvVD0y+8a9lIPI06CkK4mUdImISAxRUdYy8ceOxd5WqBCnpk5lnb1tCTCZTLTNm5dM7dpBSEjsHdq3h/nzHyNgEfsWL15M69atSepHwwwZMnDx4kW8vLxSKLL4hd8N57e6v1nnawGRRHLC6wQfr/6Y4lWLOyQmkUepZLyIiMiTcHKCiROhTp3Y206exLd7d3xnz4bq1RN/TsOAefMgroQrQwawV+FQ5An4+flRp04d1j269lw8TCYT77zzjsMSLgBXL1dc07tiOBlsi9rGiWwnWLF5BYUKFXJYTCKP0pwuERGRJ1W7NvxXaCCWs2ehZk0YNAguXEj4XAcOQNu20KmT/YQL4NtvrcU6RJJR5syZWb16NV9++SVOTk44PbLW3MOcnJzw8fFh/fr1jBgx4qnFaImycGDuASZXmczdq3dt7U0nNGXA8QF4d/Rm0bpFSrgk1dHwwiTS8EIREbErNBRq1YLdu+Pex8nJulL1Sy9ZS737+IDFAufOWdfhWrECAgPjv06nTjB7tgpoSIravn07bdu25fLly0TZKeLSrFkzpk+fTpYsWZ5KPFERURyYfYCAzwO4deIWADU/qkmd0XUSVaBGxNE0vFBERCQ5eHpak6aXX4aDB+3vExUFS5ZYvx5H8+YwfboSLklxVapUoXz58ly+fNnu9nbt2j2VhCsyLJJ9M/YRODaQoDNBAHhk9qDq/1WlYv+KdO3alSJFijBixAiHF/MQiY96upJIPV0iIhKvmzehVauEe6ySqnt3mDwZXFyS97widoSEhJAlSxYiIiIAMJvNmM1mIiMjcXJyolGjRixdujRFY4iKiGJCiQncOmnt2fLM7km1d6pRqX8lXL1c6devH5MmTcLZ2Zl9+/ZRokSJFI1H5EmoP1ZERCQ5ZckC69fDF1+Am9uTny9rVmuVwunTlXDJU7Ns2TIiIiJsvUfdu3enZMmSmEwmoqKiWLlyJcHBwcl+3ciwSNv/O7k4UbBBQdLnTE/D7xsy6MwgarxXA1cvV95//30mTZqEyWRi1qxZSrgk1VPSJSIiktycneG992DvXmjTxjqXK6nSpYN+/eDQIWt5eJGnaMGCBQCkT5+ehQsX8uuvv7J9+3YGDx4MQGRkJP/880+yXS8sOIyAMQF8l+c7Lu64aGuv+1ld3jr1FlUHVcUlnfWhw+eff85XX30FwKRJk+jYsWOyxSGSUjS8MIk0vFBERJLswgXr0MC//rImUY8sQGvj5gbly1uLZXTvDt7eTzdOkf8UK1aMLFmyMHfuXPI+sgD3ihUr6Nq1K3369OHzzz9/ouvcv32fbeO2sW3cNh4EPQCgQt8KNP/FzoLgwPjx43nrrbcA+Oabb2xJoEhqp6QriZR0iYjIE7l3D/btg+PHrf9vNoOXl3WB5ZIlNYRQUoW7d++SLl26OCsDPnjwALPZjKur62OdP/R6KFu+3cKOn3YQHhIOQNZiWfEb6kepV0phdo593cOHD1OqVCkMw2DEiBGMGjXqsa4t4ghKupJISZeIiIjI4zMMg5+K/8TNYzcByF46O7WG1aJ42+KYneKf+TJlyhSOHDnC119/rWqFkqYo6UoiJV0iIiIiSRN8IRgvHy9bD9a28dvYN2MftYbXomjzopjMcSdQhmEowZI0T4U0RERERCRF3D59m7/7/s24guM4OO9/69dV7l+ZPjv6UKxlsXgTrsDAQGrVqsX169efRrgiKUZJl4iIiIgkqxvHbrCoxyLGFxnP7sm7sURYOLvhrG272dmcYO/V7t27adq0KYGBgYwcOTJlAxZJYc6ODkBEREREng3XDl4j4LMADs0/hGGxzmDxbeCL3zA/8vnlS/R5jhw5QsOGDQkODsbPz89WIl4krVLSJSIiIiLJYtkbyzjnfw6AIs2LUGtYLXJVyZWkc5w9e5b69etz48YNKlasyN9//026dOlSIlyRp0ZJl4iIiIg8lovbL5K5UGY8MnsA4DfMj3S/pKPWsFr4lPNJ8vkuX75MvXr1uHjxIiVKlGDFihV4a706eQZoTpeIiIiIJMm5gHPMbDCTKS9OYev3W23tvvV96bCgw2MlXAC9evXi1KlTFChQgNWrV5M1a9Z49580aRLNm9tfSFkkNVFPl4iIiIgkyDAMzqw9g/9of9sQQpOTibDgsGS7xoQJE+jevTvTp08nZ86cidp/3759nDhxgsKFCydbHCLJTet0JZHW6RIREZHnzckVJ9k4aiMXtl4AwOxiplzPctT8oCaZCmRK1msldl2uU6dOUahQIQDGjBnDBx98kKxxiCQnDS8UERERkXgdXnCYC1sv4OzuTJU3qzDo9CCa/9L8iROu8PBw2rVrx+LFi21tiV0IeeHChXh4eNCkSRMWLlz4RHGIpDT1dCWRerpERETkWWaJsnB4wWFeKP0C2UpkA+DWqVvs/Hkn1d+tjpePV7JcJyoqii5dujBv3jzSp0/P2bNnyZw5s919L1++zJ07d2K0vfrqq+TLl4/27dvTqVMn/P39yZYtm2272WymcOHCiU7iRFKSkq4kUtIlIiIizyJLpIUDcw4Q8HkAN4/dpGSHkrSb1y5FrmUYBn379mXKlCm4uLiwePFiGjdubHffe/fu4e3tTWRkZKxtv//+O02aNMHHx4d79+7F2j5r1iy6dOmS7PGLJJUKaYiIiIg8x6LCo9j32z4CxwRy+/RtANwzupOtVLZEz69KCsMwGDJkCFOmTMFsNjN79uw4Ey6AdOnS8dtvv/H6668THh7OmDFjqFKlCq6urlSsWBGz2cz+/fu5cuUKt2/f5v/+7/84efIk/fr1o02bNskau8jjUk9XEqmnS0RERJ4V+2ftZ93Qddw5bx26ly5rOqq9U43KAyrjlsEtRa45evRoRowYAcCvv/5Kz549E3Xc6dOn6dSpE7t37+bTTz9lyJAhmM3/K08QGBhI586dCQkJYerUqUq4JFVRT5eIiIjIcyrkcgh3zt/By8eL6kOqU/H1irh6uqbY9VasWGFLuL7//vtEJ1wABQsWJDAwkHfeeYcPPviASpUqUbduXdv2Fi1akC9fPgIDA8mbN2+yxy7yJJR0iYiIiDwHwkLC2DFhB9lLZadI0yIAVO5fGVcvV8r3LI+ze8p/LGzQoAH9+vUjR44cDBo0KMnHu7i4UKxYMZydnSlfvjyXLl3iwoULVKlShSpVqhAWFqaES1IlDS9MIg0vFBERkbTkQdADto3fxrbvt3H/1n1eKPsCr+953WFV/aI/ej7u9evWrYuTkxP9+/fntddes83j8vX15a233uLKlSsxqhiKpAbq6RIRERF5Bt27cY+t329l+/jthAWHAZClSBaqDa6GYTEwOT2dpGvFihX8/vvvTJ48GRcXlydK9m7cuMHGjRvJmTMnbdq0oXXr1lStWpVhw4aRI0cOLBYLixYtok+fPsl4ByJPTkmXiIiIyDNm58SdrHp3FRGhEQBkK5mNWsNqUaJ9CcxO5gSOTj4BAQG0adOG+/fvU6ZMGQYPHvxE51u8eDFRUVFcv36dn3/+mddft/bY1a9fn1deeQWwLpqspEtSGyVdIiIiIs+Y9LnSExEagU95H2oNr0WxlsUwmZ/ucMJdu3bRtGlT7t+/T5MmTRg4cOATn/P27dtUrFiR6dOnU6pUKVt7+fLl2bVrF2+//TZHjx594uuIJDfN6UoizekSERGR1CTobBCBXwSS2Tcz1d+tDljnTZ1df5b8dfI7ZO7W4cOHqVWrFjdv3uSll15i+fLleHh4PPU4RFIL9XSJiIiIpEG3Tt4i4PMA9s/cjyXSgkcWDyoPqIxLOuu8qQIvF3BIXGfOnKF+/frcvHmTypUrs2TJEiVc8txT0iUiIiKShlw/cp2AzwI4OPcghsU6YKlgvYLUGl4Ll3QuDo0tKiqKZs2acenSJUqWLMny5cs1MkgEJV0iIiIiacb2n7az/M3l8N/kkMJNC+M31I881fI4NrD/ODk58cMPP/DOO++wfPlysmTJ4uiQRFIFJV0iIiIiqVhURBROLk4AFKxbEJPZRNEWRak1rBY5KuRwcHSx1a1bl927d2M2P70qiSKpnd4NIiIiIqnQv5v/ZXbj2Sx5bYmtLWuxrLx99m06/tkx1SRc9+7do1OnThw+fNjWpoRLJCb1dImIiIikEoZhcHbDWfxH+3N2/VkAnNycaPhtQ9JlTQdAhtypZ45UeHg47dq1Y/ny5ezYsYOjR4/i7KyPlyKP0rtCRERExMEMw+DUqlP4j/bn303/AmB2MVOuRzlqflDTlnClJlFRUbz66qu2cvDTp09XwiUSB70zRERERBxs95TdLO27FLD2bFXoXYEa79XAO6+3gyOzz2Kx0LdvX/744w9cXFz466+/qFmzpqPDEkm1lHSJiIiIPGWGxSD0WihePl4AlOxQko0jN1KyY0mqv1ud9DnTOzjCuBmGwTvvvMOvv/6K2Wxm7ty5NGzY0NFhiaRqJsMwDEcHkZYEBwfj7e3NnTt3tO6EiIiIJIkl0sLBeQcJ+CwAVy9Xem/rjclkAiAqPAonVycHR5iwX375hX79+gEwffp0unfv7uCIRFI/9XSJiIiIpLCoiCj2z9pP4OeB3Dp5CwA3bzeCzgaRqUAmgDSRcAG88sorzJo1iw4dOijhEkkk9XQlkXq6REREJLEiwyLZO30vm8ZuIuhsEAAeWTyoNrgald+ojLu3u2MDfEyRkZEqmiGSBHq3iIiIiKSQk8tP8k+/fwDwfMGT6u9Wp1K/Srh6uTo4sqT5448/+Pfffxk8eDCAEi6RJNI7RkRERCSZhN8N5/qR6+SqnAuAoi2K4tvAl8LNClOhdwVcPFwcHGHSLVu2jC5duhAREUHhwoVp3ry5o0MSSXM0vDCJNLxQREREHvXgzgO2/7idrd9txexsZtCZQWkywXrUxo0badSoEQ8ePKBTp07MnDkTJ6e0MfdMJDVRT5eIiIjIY7p/6z5bx21l+w/beRD0AIDMhTITdCaIbCWyOTi6J7Nz506aN2/OgwcPaN68OTNmzFDCJfKYlHSJiIiIJNG9G/fY/PVmdvy0g/C74QBkLZ6VWsNqUbJDSczOZgdH+GQOHTpEw4YNCQkJoU6dOsyfPx8Xl7TfcyfiKEq6RERERJIo5HIIm77YBMALZV+g1rBaFG9THJPZ5ODInlxQUBD169fn1q1bvPjiiyxevBh397RZZVEktVDSJSIiIpKAO+fvcC7gHGW6lAHghdIvUPPDmuSumpsizYvYFjh+FmTMmJH33nuPadOmsWzZMtKnT+/okETSPBXSSCIV0hAREXl+3D59m4AxAeybsQ8MePPkm2TMl9HRYT0VYWFhuLm5OToMkWdC2h5wLCIiIpICbhy7waLuixhfZDx7puzBEmEhr19eIkIjHB1airhz5w6vv/46QUFBtjYlXCLJR8MLRURERP4TfCGYVe+u4tD8Q/DfWKBCjQrhN8yPvDXyOja4FHLv3j2aNWtGYGAgZ8+eZeXKlY4OSeSZo6RLRERE5D8u6Vw48c8JMKBoy6L4DfWzLXT8LAoLC6NNmzYEBgbi7e3NF1984eiQRJ5JSrpERETkuXVh6wWOLjpK3TF1MZlMeGT2oNmkZmQrkQ2fsj6ODi9FRUZG0qVLF1auXEm6dOlYtmwZ5cqVc3RYIs8kJV0iIiLy3Dm78SwBnwZwes1pwDqEMH/t/ACU7lTagZE9HRaLhT59+rBw4UJcXV1ZtGgR1atXd3RYIs8sJV0iIiLyXDAMg9NrTuM/2p/zAecBMDubKdOtDN75vB0c3dP18ccfM336dJycnPj999+pX7++o0MSeaapZHwSqWS8iIhI2hNyKYR5beZxcdtFAJxcnSjXqxw1369JxvwZHRucAxw9epQGDRrw6aef0q1bN0eHI/LMU0+XiIiIPPM8s3ty/+Z9nN2dqdC3AjWG1CBD7uf34WmxYsU4cuQInp6ejg5F5Lmgnq4kUk+XiIhI6maJsnD4j8PsnbaXV5a8grOb9Rnzxe0X8c7rjZePl4MjdIxff/2VPHnyaCihiAOop0tERESeCVERURyYc4DAzwO5efwmAHun76XS65UAyFXl2S39npDff/+d3r174+Liwq5duyhVqpSjQxJ5rijpEhERkTQtKjyKvTP2EjgmkKAzQQC4Z3Kn6v9VpVRHJRdLly6la9euGIbBa6+9RsmSJR0dkshzR0mXiIiIpFn3b99nYtmJBP8bDEC6bOmo9k41Kg+ojFt6NwdH53jr16+nXbt2tjW5fvzxR0wmk6PDEnnuKOkSERGRNMUSZcHsZAbAI5MH2UtmxxJpocZ7NajYtyIu6VwcHGHqsG3bNlq0aEFYWBgtW7Zk2rRpmM1mR4cl8lxSIY0kUiENERERxwgLDmPHhB3smLCD3lt7kz5nesBaDt4jswfO7nqWHO3UqVNUrlyZ27dvU7duXZYuXYq7u7ujwxJ5bumvk4iIiKRq92/fZ9sP29g2bhsPbj8AYNekXdQeWRvAlnzJ/+TNm5emTZty6tQpFi1apIRLxMHU05VE6ukSERF5OkKvh7L1u61s/3E74SHhAGQpmgW/oX6U7lQas7OGysXHYrEQGhpK+vRKSkUcTT1dIiIikupEPohkQskJ3Lt+D4DspbLjN8yPEu1K2OZzSUzXrl1jwoQJDB8+HCcnJ8xmsxIukVRCSZeIiIikCqHXQ/HM5gmAs7szpbuU5rz/eWoNr0XRFkUxmVV1Ly5BQUE0bNiQvXv3cufOHb777jtHhyQiD9HwwiTS8EIREZHkdfvMbQLHBrJ32l56bOxBnmp5AGtvl5Obk0qcJyA0NJQGDRqwefNmsmfPTkBAAEWKFHF0WCLyEPV0iYiIiEPcPH6TwDGB7Ju5DyPK+gz42JJjtqRL1QgTFhYWRuvWrdm8eTMZM2Zk9erVSrhEUiH9NRMREZGn6tqhawR8FsCheYcwLNZkq2D9gtQaXot8fvkcHF3aERkZSadOnVi9ejWenp4sX76cMmXKODosEbFDSZeIiIg8NYbFYG7zuQSdCQKgSLMi+A3zI/eLuR0bWBr0+uuv89dff+Hq6srixYupWrWqo0MSkTio/I+IiIikqEs7LxEVEQWAyWyi5gc1Kd6mOH1396XT352UcD2m1q1b4+XlxR9//EHdunUdHY6IxEOFNJJIhTREREQS53zgefxH+3Nq1SlaTmtJuR7lHB3SM+fGjRtkzZrV0WGISAI0vFBERESSjWEYnFl3Bv/R/pzbeA4Ak5OJ26dvOziyZ8OkSZOoV68eBQsWBFDCJZJGKOkSERGRZHFi+Qn8R/tzYcsFAMwuZsr1LEfND2qSqUAmB0eX9k2cOJH+/fuTI0cO9u/fr4RLJA1R0iUiIiLJYvNXm7mw5QJObk5U6FOBGu/VwDuPt6PDeibMmTOHAQMGANCzZ08lXCJpjJIuERERSTJLlIUjC4+Qv3Z+PLN7AlB7ZG2Olj9K9Xerkz5HeofF9tdff7FhwwbGjRvnsBiS05IlS+jWrRuGYTBw4EA+/fRTR4ckIkmkQhpJpEIaIiLyPLNEWjgw9wCBnwdy4+gNarxfg3pj6zk6rBiqVavG1q1buXDhArly5XJ0OE9k3bp1NGnShLCwMLp168a0adMwm1V8WiSt0btWREREEhQVHsXuqbv5sdiPLOq2iBtHb+Ce0Z10WdM5OrQYLl68yNatWwFrj1datnPnTlq0aEFYWBitW7dm6tSpSrhE0igNLxQREZF47Z66G/9P/Llz/g4A6bKmo9o71ag8oDJuGdwcHF1Mf/31F2azmaxZs/LHH38wcOBAR4f02PLnz0+xYsXInDkzc+fOxdlZH9tE0iq9e0VERCRel3dd5s75O3j5eFF9SHUqvl4RV09XR4dFVFQUFoslRtsff/xBzpw5yZs3L4GBgVy6dIls2bLZtptMpjSTvGTNmpW1a9fi7OyMm1vqSm5FJGnURy0iIiI2YSFhbPpqE5f3XLa11Xi/Bo3HN+at029RbXC1VJFwhYaGkjlzZlxdXWN8+fv7kzdvXvLmzQtArly5Ymx3cXFJ1cMOz58/z8yZM22vvb298fT0dGBEIpIc0sajHhEREUlRD+48YPv47Wz9biv3b92nWKtidPyrIwAZ82WkysAqDo4wJk9PT9577z1GjBiByWSiTJkyeHp6YjabyZcvH05OTtSvX5/Q0FAsFgsHDhwgJCSEevXq8dJLLzk6fLuuXr1K/fr1OX78OOHh4bz22muODklEkol6ukRERJ5j927eY93wdXyf73vWD1/P/Vv3yVIkC8VaF3N0aAkaOnQoGzduxMfHhyNHjuDs7EzBggVxcnICIEeOHGTPnp2TJ09y7949vvrqK1auXEnmzJkdHHlst2/fpkGDBhw/fpx8+fLRsGFDR4ckIslIPV0iIiLPqc1fb2bjqI2E3w0HIFvJbPgN9aNkh5KYndLGc9maNWty4MABevXqxaJFi2jcuDEvvPCCbfvKlSvJmjUrW7ZsoXLlyg6MNG53796lSZMm7N+/Hx8fH9asWUPu3LkdHZaIJCMlXSIiIs8pJ1cnwu+G41POh1rDa1GsVTFMZpOjw0qyTJky0apVKxYvXkz69OmJjIwkPDycdOnSkT59ekqXLp1qE64HDx7QqlUrtm7dSqZMmVi1ahWFChVydFgiksyUdImIiDwHgs4FsemLTeSvnZ+SHUoCUKF3BTL5ZqJwk8KYTGkv2XrYggULeOGFF7h37x6rVq3i7t27VKpUiTx58rBq1SpCQkJInz69o8OMISoqildeeYW1a9fi5eXFihUrKF26tKPDEpEUkDbGDoiIiMhjuXXyFotfW8z4QuPZ+fNONo7aiGExAHBJ50KRpkXSfMIVHBzMqlWrePDgAcuXL6dgwYL07NmTrVu3cubMGcLDw/nnn38cHWYsZrOZSpUq4e7uzt9//02VKqmrWImIJB8lXSIiIs+g60eu81fXv/ix6I/s/XUvlkgLBeoWoMmEJmlyCGF8/vnnH8LDwwkKCmLw4MFs2bKFSZMmsWjRIiIjIwHr+l2pjclkYtiwYRw5coTatWs7OhwRSUEaXigiIvKM2Th6Ixs+3gDWDi0KNymM3zA/8lTL49C4Usrhw4fJnj07M2fOpEGDBrb2li1bcuDAATp37syhQ4ccGGFMv/76Kx06dMDLywuA/PnzOzYgEUlxJsMwDEcHkZYEBwfj7e3NnTt3yJAhg6PDERERAcCwGLYerJMrTzK70WyKtSqG3zA/clbM6eDoUlZERAQWiwU3Nze72w3D4N69e6likeExY8bw0UcfUa1aNTZu3IiLi4ujQxKRp0A9XSIiImnYv1v+xX+0P7mq5KL2yNoA+Dbw5Y2jb5C1aFbHBveUJJS4mEymVJFw/fTTT3z00UcAtGnTRgmXyHPEoXO6Ll68yPfff0+DBg3Imzcvrq6u+Pj40LZtW7Zt22b3mODgYAYPHky+fPlwc3Mjf/78DBkyhLt379rd32KxMH78eEqXLo2HhwfZsmWjU6dOnD59OiVvTUREJMUYhsHZDWf5re5v/Fr9V04uP8n2H7cTGWadv2QymZ6bhCutmDlzJgMHDgRg2LBhvPvuuw6OSESeJocOL/zggw/44osv8PX1pXbt2mTLlo0TJ06waNEiDMNgzpw5dOzY0bZ/aGgoNWvWZO/evTRo0IDy5cuzZ88eVq1aReXKlfH398fd3T3GNfr06cOUKVMoWbIkTZs25dKlS8yfPx8vLy+2bt1K4cKFkxSzhheKiIijGIbBqVWnCPg0gPOB5wEwO5sp26MsNT+oSWbfzA6OUOz566+/aN++PVFRUbz55puMGzcuzVeMFJGkcWjS9eeff5IlSxZeeumlGO0BAQHUrVsXLy8vLl++bBuj/fHHH/PJJ5/w/vvvM3bsWNv+0cnb559/zocffmhrX79+PS+//DK1atVi9erVuLq6ArB8+XKaNGlCgwYNWLlyZZJiVtIlIiKOsnH0RjaM2ABYFzYu37s8Nd+viXdeb8cGJnFas2YNTZs2JTw8nB49ejB16lTMZhWPFnnepNpCGg0bNmTVqlXs2LGDSpUqYRgGuXPnJjg4mCtXrsQYmx0aGoqPjw/Zs2fn1KlTtvbOnTszd+5cNm7cSK1atWKcv06dOmzYsIFz586RN2/eRMelpEtERJ4Ww2IQFhyGe0brKI5bJ2/xS4VfKP9aeWoMqUH6nKlrsV+JLXp0jp+fH/PmzcPZWdPpRZ5HqfZRS/Tk0ug/TidOnODSpUvUqFEj1mRYT09PatSowenTp/n3339t7Rs2bLBte1TDhg0B2LhxY7xxhIWFERwcHONLREQkJVmiLByYc4CfS//M0n5Lbe2ZC2XmnUvv0Oi7Rkq40ohy5cqxbds25syZo4RL5DmWKpOu8+fPs2bNGnLkyEHp0qUBa9IFxDkHK7o9er/Q0FAuX75MgQIFcHJySnD/uIwZMwZvb2/bV548z+YaJyIi4nhREVHsnb6Xn4r/xJ9d/uT64eucXn2aB3ce2PZx9XJ1YISSGMeOHWPz5s221wUKFIiznL2IPB9S3SOXiIgIunbtSlhYGF988YUtYbpz5w4A3t72x61HD/WL3i+p+8flww8/ZPDgwbbXwcHBSrxERCRZRYZFsnf6XjaN3UTQ2SAAPLJ4UPX/qlJlYBXcvd3jP4GkGufOnaNevXrcunWL5cuXx5reICLPp1SVdFksFnr06IG/vz99+vSha9eujg4JNzc3PZ0SEZEUteuXXawYtAIAz+yeVHu3GpX7V1avVhpz5coV6tWrx4ULFyhWrBjFixd3dEgikkqkmqTLYrHQq1cv5syZw6uvvsrEiRNjbI/usYqrZyp6rlX0fkndX0RE5GkJDw0n5GIIWYpkAaB8r/LsmbqHcr3KUbFPRVzSadHctObWrVs0aNCAkydPki9fPlavXk22bNkcHZaIpBKpIumyWCz07NmT3377jU6dOjF9+vRY5VQTmoP16JwvT09PcuTIwZkzZ4iKioo1ryuhOWIiIiLJLSw4jO0/bmfLt1vwzuNN3919MZlMuHq58vre17V2UxoVEhJCkyZNOHDgAD4+PqxZs4bcuXM7OiwRSUUcXkjj4YSrY8eOzJw5M87CFzlz5mTTpk2EhobG2BYaGsqmTZsoUKBAjPlWL730km3bo6LX59JYaxERSWn3b91nw8gNfJ/ve9YNXcf9m/cJCwkj5FKIbR8lXGnTgwcPaNWqFdu2bSNz5sysXr2aQoUKOTosEUllHJp0RQ8p/O2332jfvj2zZs2ym3CB9R+j3r17c/fuXUaPHh1j2+jRo7l79y59+vSJ0d63b18Ahg8fTnh4uK19+fLlbNiwgQYNGpAvX75kvisRERGr0OuhrPlwDd/n/56NozbyIOgBWYtlpfXM1gw8OpAMubTeY1pnMpnw9vYmffr0rFixglKlSjk6JBFJhRy6OPLIkSMZNWoUXl5eDBo0yO76Fa1ataJcuXKAtUerRo0a7Nu3jwYNGlChQgV2797NqlWrqFy5Mhs3bsTDwyPG8X369GHKlCmULFmSpk2bcvnyZebNm4eXlxdbtmyhSJEiSYpZiyOLiEhiHVtyjN9b/g7AC2VewG+YH8XbFMfs5PCBJpKMIiMjOXbsGCVLlnR0KCKSSjk06erRowczZsyId59p06bRo0cP2+s7d+4wcuRIFi5cyJUrV8iRIwft27fn448/Jn362AtFWiwWfvzxRyZNmsTJkyfx8vKiXr16fPbZZ/j6+iY5ZiVdIiISlzv/3uHGkRv4NrD++2IYBot7LKZYm2IUbV4Uk1lDCJ8FhmGwcOFC2rZtq2GhIpIoDk260iIlXSIi8qjbp28TMCaAfTP24ZbBjbfPvq1y788owzB4//33+eqrr+jbty+//PKLo0MSkTQgVVQvFBERSYtuHLtB4OeB7J+9HyPK+gzzhTIvcO/GPSVdz6jPP/+cr776CoDKlSs7OBoRSSuUdImIiCTR7dO3WTd0HQfnHYT/xov4NvSl1vBa5K2R17HBSYoZP348w4YNA+Cbb76hd+/eDo5IRNIKJV0iIiJJFBkWaUu4irYoit9QP3JVyeXosCQFTZ8+nbfeeguAESNGMHjwYAdHJCJpieZ0JZHmdImIPH8ubL3Aha0XqPp2VVvb1u+3kr92fnzK+TgwMnkaFi5cSIcOHbBYLAwaNIjvvvtOBTREJEnU0yUiIhKHc/7n8B/tz+k1pzE5mSjSvAiZfTMDxEjA5NkWFRWF2Wyme/fufPvtt0q4RCTJlHSJiIg8xDAMTq85jf9of84HnAfA7GymTNcyOLk6OTg6cYQOHTpQoEABKlSogNmsNdZEJOmUdImIiPzn5vGb/NXtLy5uuwiA2cVM+V7lqflBTTLmz+jY4OSp2rdvH1myZCF37tyAKhWKyJNR0iUiIvIfz+ye3DhyA2d3Zyr0qUCN92qQIbfm7z5vjhw5Qr169fD09GT9+vUUKFDA0SGJSBqnpEtERJ5LligLh/84zMnlJ2k5vSUmkwn3jO60m98On7I+ePl4OTpEcYCzZ89Sv359bty4Qb58+ciSJYujQxKRZ4CSLhERea5YIi0cmHOAgM8DuHnsJgClu5TGt4EvAIUaFnJkeOJAly9fpl69ely8eJESJUqwYsUKVSoWkWShpEtERJ4LUeFR7PttH4FjArl9+jYA7pncqfp/VbXGlnDz5k3q16/PqVOnKFCgAKtXryZr1qyODktEnhFKukRE5Jl3+/RtpteeTvC/wQCky5aOau9Uo/KAyrild3NwdOJoISEhNG7cmEOHDpEzZ07WrFlDzpw5HR2WiDxDlHSJiMgzyTAM23pKGfNnxNXLFa8cXlQfUp2KfSvi6unq4AgltQgNDeXevXtkyZKF1atXU7BgQUeHJCLPGJNhGIajg0hLgoOD8fb25s6dOxrnLSKSCoWFhLHjpx0cmHOA3tt64+LhAljLwXvn9cbZXc8bJbabN29y4cIFypYt6+hQROQZpH95RETkmfAg6AHbftjG1u+38uD2AwAOzD5Ahd4VAMhSRFXo5H+ioqLYtGkTtWrVAiBLliyqVCgiKUbLqouISJp278Y91g5dy/f5vmfDxxt4cPsBWYpmodWMVpTrUc7R4UkqZBgG/fr1o3bt2kycONHR4YjIc0A9XSIikmaFXg/lh4I/EH43HIDspbLjN9SPEu1LYHbSc0WJzTAM3n33XaZMmYLZbFbvlog8FUq6REQkTQkLCbNVHPTM5kmBlwsQfCGYWsNrUbRFUUxmk4MjlNRs9OjRfPvttwBMnjyZ9u3bOzgiEXkeqJBGEqmQhoiIYwSdDSJwbCAHZh9gwKEBeOf1BiAsOAzX9K62SoUicRk3bhxvv/02AN9//z2DBg1ybEAi8txQT5eIiKRqt07eIuDzAPbP3I8l0gLA4YWHqfZ/1QBwy6B1tiRh06ZNsyVco0aNUsIlIk+Vki4REUmVrh+5TsBnARycexDDYh2UUbB+QWoNr0U+v3wOjk7SmuPHjwMwePBghg8f7uBoROR5o+GFSaThhSIiKS88NJxvc35LWHAYAEWaFcFvmB+5X8zt4MgkrTIMg2XLltGkSRMNRRWRp05JVxIp6RIRSRnXDl0je8nsttdrPljDrRO38BvmR47yORwYmaRV+/bto0iRInh4eDg6FBF5zinpSiIlXSIiyev8pvP4j/bn1MpTdN/Qnfwv5QesPRPqkZDHtWvXLurUqUOlSpVYsmQJXl5ejg5JRJ5jmtMlIiJPnWEYnF1/Fv9P/Tm7/iwAJicTl3ddtiVdSrjkcR0+fJiGDRsSEhKCxWLBycnJ0SGJyHNOSZeIiDw1hmFwauUp/Ef78+/mfwEwu5gp16McNT+oSaaCmRwcoaR1Z86coX79+ty8eZPKlSuzZMmStDW88P592LcPdu2CAwfgzh2IjAR3d8ibFypUgIoVIV8+0IMJkTRDSZeIiDw1RpTB8jeXc+vkLZzcnKjQpwI1htSwrbkl8iQuXbpEvXr1uHTpEiVLlmT58uVpYyqAYcCGDTBhAixZAuHhCR9ToAD06QOvvQbZsye8v4g4lOZ0JZHmdImIJJ5hMTi6+CiFmxTG2c36nG//7P1c3n2Z6u9WJ32O9A6OUJ4VN27c4KWXXuLw4cP4+voSEBBAjhxpoADLihXw7rtw6NDjHe/qCj16wNixkEk9xSKplZKuJFLSJSKSMEukhYPzDhLwWQA3jtyg2S/NqNi3oqPDkmfY7t27qVu3Lp6engQGBpI/f35HhxS/O3dg8GD49dfkOV+OHDBpEjRrljznE5FkpeGFIiKSbKIiotg/az+Bnwdy6+QtANwzumOJtDg4MnnWVahQgY0bN+Li4pL6E66jR6FxYzh7NvnOefkyNG8O77wDX34JZnPynVtEnph6upJIPV0iIrEZhsGuX3YRODaQO+fuAJAuazqqDq5KlTeq4JbBzcERyrMoPDycU6dOUbx4cUeHkniHDkGdOnD9euL2z5YNXFysPWOhoYk7pk8f+OUXFdoQSUX0GERERJ6YyWTi6KKj3Dl3B88XPKn/dX0GnR2E34d+SrgkRURFRfHqq69StWpVAgICHB1O4pw/D/Xrx59wZc8OH34I69dDUBBcuwYXL0JwMBw/DrNmQaNG8V9n8mR4//1kDV1Enox6upJIPV0iIhB+N5wdP++gTJcypM9pLYZxcftFLmy7QIXeFXDxcHFwhPIsMwyDPn36MHXqVFxcXPj7779p2LCho8OKn8UCdetaqxTa4+VlHRb42mvW4hgJOXkS3nzTWogjLitWQGr/vog8J5R0JZGSLhF5nj2484Dt47ez9but3L91n6qDq9LwG32ok6fHMAzeeecdvvvuO8xmM/Pnz6dt27aODithEybAG2/Y31atGsyda117KykMA6ZOhQEDICIi9vbcueHgQfDWkgwijqZCGiIikqD7t+6z9futbPthG2F3wgDIXDgzOSvmdHBk8rz55JNP+O677wD49ddf00bCdfUqvPee/W21a8M//0C6dEk/r8kEvXtDrlzQsmXsxOvCBRgxAsaNS/q5RSRZaU6XiIjEa8OoDXyf73v8R/sTdieMbCWy0WZOG9448galO5d2dHjyHPn+++8ZOXIkAD/88APdu3d3bECJ9csv9otg5M0LixbFm3AtXLiQKlWqEBUVFff5GzeOO7GaMsU6N0xEHEpJl4iIxOvB7QeE3w3Hp5wP7Re0p/+B/pTuVBqzk/4JkafHYrGwcuVKAD799FPefPNNB0eUSBER1qTLnilTEhz6N2nSJHbs2EFgYGD813n9dXj55djt9+7Bb78lMlgRSSma05VEmtMlIs+yoHNBbPpyE2W7liV31dwAhFwK4dKuSxRpVgSTSlCnWf7+/nh5eVGhQgVHh/LYwsLC+OOPP+jSpUva+V1cutS6ftajmjWDv/+O99Dbt2+TPXt2IiMjefPNN/nhhx/iv9aBA1CmTOz2UqWs20TEYfSYUkREuHXqFkt6L2F8ofHsnLAT/9H+tm3pc6anaPOiaedDrsRy584dmjdrQfNmLQgPD3d0OEly/Phxop8Pu7m58eqrr6at38W4qhXGVVTjIUuWLCEqKop27drx559/YrEksMh46dLg5xe7/eBBuHkz4VhFJMUo6RIReY7dOHqDv7r9xY9Ff2TP1D1YIi0UqFuA6u9Vd3Rokox++OEHQkNDuXz5EtOmTXN0OIm2ceNGypYty9tvv51wwpFa7doVuy1zZmjQIEbT1atXOXToUIyvOXPmUL16dd58800uXrzIn3/+GWufyMjImOfu3Nl+HLt3J9MNicjjUPVCEZHn1Mp3VrL1u63w3yDzQo0LUWt4LfJUy+PYwCRZ3blzh6+/+oZcTpUJd7rLJ6NG07NnT1wTsxaUA+3cuZPmzZvz4MEDzpw5Q1RUFGZzGntWbBj2k51KleChezEMg5IlS3LTTm/UuHHjqFGjBjlz5qR9+/axtn/11Ve8++67/2uoXNl+LLt2WRdmFhGHSGN/vURE5Ek8PI03c6HMYEDRlkXps6MPXZZ1UcL1DPrhhx+4F3qPAq5+FHSpnSZ6uw4dOkTDhg0JCQmhTp06zJ8/HxeXNLjgdmgoBAfHbi9bNsZLk8nEnDlzyJ49O2azmZEjR7Jlyxa2b9/OgAEDcHJyYvv27WzZsoU1a9ZQrlw5AF555RX69u0b89ylS1tLyT/q0qVkuikReRxKukREngP/bvmX2U1ms++3fba28r3K029fP15Z9Ao5K2m9rWdRdC9XTqdKuJnT4+WUnRdcSvHJqNGpdm7XqVOnqF+/Prdu3eLFF19k8eLFuLu7OzqsxxMWZr89ffpYTQ0aNGD//v3Ur1+fkSNHsnDhQsqWLYuzs3VQUq5cuXB3d+eNN97gxIkTTJs2jTlz5sQu6uXqCva+Xw8ePOndiMgTUNIlIvKMMgyDsxvO8lvd3/i1+q+cXH6STV9ssvV2Obs580KZFxwcpaSkh3u5oqXm3q4LFy5Qr149Ll++TOnSpVm2bBnp7SQoaYZzHLM4Hl3E+D8vvPACy5Yt47PPPuPrr79m3rx5MbZ37tyZqKgodu3aRY8ePewXFDEM++dPiz2FIs8QJV0iIs8YwzA4teoU02tNZ0adGZxZdwazs5lyvcrRaUmntFX5TR7bo71c0VJzb9eWLVs4f/48hQoVYtWqVWTOnNnRIT0ZLy/7ideJE3EeYjabKVasGABVq1blxo0bbPivAuKLL76Ik5MTRYoUifuaZ8/Co8U1ADJmTHzcIpLsHruQRmRkJBs2bODUqVN07tyZ9OnTc+nSJTJkyICXl1dyxigiIkmwYtAKto/fDoCTqxPlXytPjfdrkDFfRscGJk9VdC9XeY/YJcQLutRmy+UfmTZtGq+//roDorOvffv2uLi4UL58eXx8fBwdjl2GYbB582ZOnjxJREQE2bJlo169enh6esbe2ckJSpSA/ftjtu/cGe81Fi5cSOnSpTl//jy1atXiypUrvPrqqzRp0oTp06dz+PBhSpYsaf/guM5tb/0uEXlqHmtx5HPnztGoUSPOnz9PWFgYx48fp2DBggwaNIiwsDAmTpyYErGmClocWURSG8NiEBkWiYuHdfjQmfVnmNN0DhVfr0iNITVInzMND8+Sx3Lnzh3y5slHxgclKObexO4++x/Mx5z1FmfOnnZoJcN79+4RGhpKtmzZHBZDYty/f5/Jkyfz888/c/To0RjbvL296d69O4MGDaJgwYIxD3ztNfj119gnPHYM7PRYhYWFkS1bNtKnT8/ly5epW7cubdu2ZciQIWTMmJELFy4wcuRIPv74Y/uB9uoF9oaOnjwJvr6JvV0RSWaPNbxw0KBBVKpUidu3b+Ph4WFrb926NWvXrk224EREJG6WKAsH5h7g5zI/s/GTjbb2/LXz83///h+NvmukhOs5ZW8u16NSw9yusLAw2rRpg5+fH//++6/D4kjI9evXqV27Nu+88w5lShZn/fK/eXDrKpHBNzmxfzf9e/dizpzZVKhQgfXr18c8+MUX7Z80jgfUq1evJiQkhGvXrjF27FhWrlxJv3792LNnDzly5ACsPWF23boFc+fGbs+SBR5NBkXkqXqspCsgIIBhw4bFejKWP39+Ll68mCyBiYiIfVERUeydvpefiv/En53/5Pqh6xyYdYCoiCjAWn46XZZ0Do5SHCWuuVyPcvTcrsjISLp06cLKlSv5999/U+3nh9DQUJo0acLZs2fYumEN836bRu1afri5ueHk5EQh34KM+eRjTh3YQ5WKFWjWrBk7Hx7i16YNuLnFPvHUqWDnnoODgylZsiSbNm3ivffes61NVqhQIQIDA3nvvfeIjIzE7kClb76xX6Wwc2f7ZeRF5Kl5rKTLYrEQFRUVq/3ChQtpu8qQiEgqFhkWyc5fdvJjkR9Z3HMxt07cwiOzB3VG16H/gf44uTg5OkRJBRLTyxXNUb1dFouFPn36sHDhQlxdXVm8eDFVq1Z9qjEk1vjx4zlw4AAr/lpAxfLl4twvQ4YMLJo3m+JFi9C/f///bciaFewsakxwMLz+urXa4EM6d+7MwYMHqVKlSqxDXF1d+eKLLzh8+HDsgji7dsEXX9gPrl+/OOMWkafjseZ0dezYEW9vbyZNmkT69OnZv38/2bJlo2XLluTNmzdVlqFNLprTJSKOsvr91Wz+cjMAntk9qfZONSr1r4RbejtP0eW5lJi5XI962nO7DMPg7bff5ocffsDJyYkFCxbQqlWrFL/u44iKisLX15c6fjWY9suERB2zdPkKmrd7he3bt1O5cmVr47ZtEFdS+cUX8N57TxbotWtQs6b9qogvvwya+iHicI+VdF24cIGGDRtiGAYnTpygUqVKnDhxgqxZs+Lv70/27NlTItZUQUmXiDwt4aHhhN0Js83LCjoXxPSXplP17apU7FsRl3Rad0diGjNmDB999BFZnHxxMdmppmfHA+MOQVHnmDJlCq+99loKRwgjRoxg9OjRAPz222907do1xa/5uFasWEHjxo3Z7r+OyhUrJOqYqKgofEuVp36DBkyePPl/G159FWbPtn/Q55/DBx883hDAf/+FRo3g8OHY25ydYccOKFcu6ecVkWT1WCXjc+fOzb59+5g3bx779u3j7t27vPbaa3Tp0iVGYQ0REUm6sOAwtv+4nS3fbiFfrXx0/LMjABnzZWTQ6UGYzJqbIfYVLFiQWn4vAUl5npoRyE/u3LlTJqiHBAcHM2fOHAB++umnVJ1wARw/fhx3d/dEJ1wATk5O+FWvyvHjx2NuGDcO1qyBq1djH/TRR7BhA0yeDHnzJu5ChgHTp8P//R/cuWN/n6FDlXCJpBKPvU6Xs7MzXbp0oUuXLskZj4jIc+v+7ftsG7eNbeO28SDIOhn+2oFrhN8Nx9XLOuxLCVf8IiIiWLRoES1btnRoGXRH6dixIx07dnR0GHHKkCEDAQEB/PPPP/Tu3dvR4SQoLCwMN3tFMBLg5uZGWFhYzMYsWWDKFGjRItY8LgBWrbKu6fXqq9C/v3VdLXs9X/fuwfz58NNP8a/3VamSNZkTkVThsZKuMWPG8MILL9CrV68Y7b/++ivXr1/n/fffT5bgRESeB6HXQ9ny7RZ2/LSD8BBrFbmsxbLiN9SPUq+U+n/27jqsyvMN4Pj30AiI2N0Ys3O2s7vb2V2bzk3dZutv1qZOZ81hd3d3N4qtiIgFNt2c9/fHKyhyUDgcPMT9uS6ucd543vu4TbjP/Tz3g4mZXj2PUqUlS5YwcOBAZs2axU8//WTscMR7np6eUe3Os2XLliwSLoAMGTLg6+uLt7c36dKli/N9Ho+fkD59+pgnmjaFefNg8GDdNwYEwL//ql+ZM0O5clCwoDpN0Nsbrl2DmzchLOzzATg6wu7dkAo/eBAiqdJrTVfevHlZu3YtVapUiXb8woULdOzYEXd3d4MFmNTImi4hhKGdn3OeA8MOAJClZBaqj6lO0dZFMTGVZCs+QkJCcHR05MWLFzg4OPDw4UPSpJHW+ca2e/du2rVrx+LFi5P8dMJPeXl5kStXLv6aMpmhgwd++QbgkYcH+YuV5t9//6Vv3766L1q4UE284v8r2JcVK6ZWzbJnN/zYQgi96fUT3cvLK+oTq49lypQJT0/PBAclhBApmc8TH55d+rA/T7m+5SjYsCAdtneg/9X+FGtXTBIuPSxbtoynT5+ydetWXr9+zb///mvskFK9Y8eO0bZtW4KDgzl06JDuvaWSsKxZs9KmTRsW/LeEsC9Vl95bsHgJdnZ2dO7cOfaLBg6EvXsNnxh16QKnT0vCJUQSpFely9HRkfHjx9OlS5dox1etWsX48eN5+PChwQJMaqTSJYTQ17uH7zg97TTXll8jQ6EMDLw+UOcaLUVRcPd258rzK9x5fQe/ED8UFGzMbXDM4Ej57OUplKEQJhpJzCJFVrmqVavG2rVr6d27N3v27JFqlxFduHCBunXr4u/vT4sWLdi0aRPm5smv4+bly5epUqUKndu3ZcnCeZiaxr4f3tYdO2nXpQdjxoxh4sSJXx7c21tthLF8ecKCzJYNFi1S14sJIZIkvdZ09e3bl2HDhhEWFkbt2rUBOHLkCCNHjuTnn382aIBCCJHcvb73mtNTT3N99XWUCPVzLptMNgS+CcQm04e23rde3mLh5YVsuLWB14GvPztmWsu0tCjcgkEVBvFtjm9jbpSaykRWucaOHQvA6NGjWbFiBf/++6+s7TKCGzdu0KhRI/z9/alTpw7r169PlgkXQPny5Vm+fDldu3bl5avXTBzzW4xuhp6eXsxf7MTUv2bRrl07xo8fH7fB06WDZctg+HBYsABWrVLXdcVV6dIwaBB07gw2cdsiQAhhHHpVuhRF4ddff2Xu3LmEhqqLvq2srBg1ahTjxo0zeJBJiVS6hBBx9frea05MOMHNDTejOngXaFCAGmNqkLvah7bQ119cZ9j+YRx7dEyv55TNVpaZ9WfyXd7vDBB18vNplSuSVLuM48GDB1SrVo0XL15QqVIlDh06hK2trbHD0tupU6d48OABWbJkYdCgQXh4eFCuTGmqVKqIuZk5Dx89Yve+A5ibmzNs2DAmT5782WrYZ/n4qNMOL16EK1fgxg31mKKoTTFy5lSba5Qrp256XL68fnt7CSG+Or2Srkj+/v7cuXMHa2trHB0d9WqrmtxI0iWEiKsH+x+wppG6GWqhZoWoMaYGOSrmiDofFhHG1NNTmXxyMuHa8AQ/b1D5QUyvNx1bi+T7C64+Fi1axKBBg7h16xZFixaNOv7w4UMKFSrEn3/+KdWur2jChAlMnDiRkiVLcvz4cRwcHIwdUoLUrFmTy5cv8+rVKywtLdm3bx9OTk64ubkRGhpK5syZadeuHd26dYtXh8M4UxT1y0SmEwuRnCUo6UqNJOkSQsTm6YWneLt7U7xjcUCdFXD418OU6FSCrKWzRrv2XdA7mq1rxpknZwwaQ+EMhTnQ5QB50uUx6LhJVWxVrkhS7fr6FEVh1qxZdOnShSxZshg7nATx8vIie/bsKIrCli1baN26tbFDEkIkU3onXZcvX2bjxo08fvw4aophpK1btxokuKRIki4hxKc8TnlwcvJJHh56iFU6K4Z5DMMybeyV/7dBb6m9ojYuL1wSJZ6caXNyssdJ8jnkS5Txk5LYqlyRpNoVP8ePH2fs2LEcOXIkXptL+/r6YmVlleI2pF60aBFDhgwhT548VKpUiTVr1hg7JCFEMqVXrXr9+vVUqVKFO3fusG3bNsLCwrh16xZHjx7F3t7e0DEKIUSSoygKDw8/ZHnN5SyvsZyHhx6iMdVQpGURwoJiby0dGhFKs3XNPptwWZtZ06t0Lza03cCDHx4QOiaUsLFhPB72mG0dtvFjxR9JZ5Uu1vuf+j6l3qp6vA16m5C3mOSFhIQwZcoUOnbsqDPhAsifPz/du3dn+vTpBAYGfuUIk59FixZx+vRpjh2L+/rCgIAAGjVqROvWrQkKCkrE6L6+LVu28N1339GtWzd2795NSEiIsUMSQiRTelW6SpYsSf/+/Rk8eDB2dna4uLiQL18++vfvT7Zs2eLWJjWZkkqXEMLrmhd7Bu7h6fmnAJiYm1CmVxmqjqqKQ77Pr18Zd2wck09O1nnOVGPKiCojGFl1JA7Wnx8nMCyQfy//y+ijowkK1/2LbpeSXVjValUc3lHy9KUqVySpdsVNcHAwmTJlwt/fn759+7J48eIv3hMSEkKzZs04dOgQ6dKl49y5cxQpUuQrRGtYPj4+NGjQAB8fn2jH79+/z/z586levTrFixcnX7580davm5mZsXjxYipXrvy1QxZCJDN6JV02NjbcunWLvHnzkiFDBo4fP06JEiW4c+cOtWvXTtEbJEvSJYR4++At8wrPw9TClLJ9y1J1ZFXS5vzy3wfOns5U/K8iEUpEjHN57POwuf1mymcvH69YHrx9QPtN7bnqdVXn+R0dd9C8cMrbu+dLa7k+JWu7vmznzp20aNGCZs2acf78eTw9PT/bhS88PJz27duzbds2bGxsOHz4MJUqVfqKERtOeHg4rVu3ZteuXaRJk4aePXtiaWmJlZUVv/76K7a2tsyZM4cnT56gKApbt27Fw8ODIkWKcPDgQXLlymXstyCESOL0ml7o4OCAn58fADly5ODmzZsAeHt7y/QNIUSKoo3QcnPDTY6OPRp1LH3B9LRa1Yqh7kNpNLdRnBIugKH7h8aacJ3qeSrWhMvFxYXjx4/rPFcwfUGOdT9GuWzldJ7/Yd8PhEXEPt0xufp0X64vGT16NK9fv+bff/9N5MiSBz8/P548eRLta926dXzzzTeMGTOGV69esWPHjhjXaLVaALRaLb169WLbtm1YWFiwY8eOZJtwgVqx2rFjB3PnziU8PJzTp0/Tt29f/vjjD+zs7NBoNAwbNoyxY8fy9OlTPDw86Nu3L5cvX5aESwgRJ3pVujp37kz58uUZPnw4kydP5p9//qFFixYcOnSIsmXLSiMNIUSypw3XcmPdDU5POc3ru69BA4PvDCZj4Yx6jefs6Uy5xTETI1ONKRf7XqRstrI67oKwsDAKORbi9Zs3eHg8In369Dqve+b7jGILiuET4hPj3Jb2W2hdNOV0XYuIiCBfvnwEBgbSpk2bON+3Y8cOADw8PFLFFiefkz9/ftzd3WMcHz9+POPHjydfvnx4eHjEOD9x4kTGjh3LDz/8wPz58zE1NWXr1q00b55yqqnXrl2jY8eOPHnyhKtXr1KoUCFATTQLFSrE69ev+e+//2jXrp2RIxVCJCdm+tw0b948goODAfXTQ3Nzc86ePUubNm0YM2aMQQMUQoivKSI0ApeVLpyeepp3D98BYOVgRaVhlbDNov/+VwsvLdR5/Ldqv8WacAGsXr2aRx6PMDM15++//2bSpEk6r8uRNgd/N/ybnjt6xji34NKCFJV0KYpCyZIlefHiBc7OznG+L1euXGTMmBHZKQXmz59Pt27dePPmDSNGjKBOnTqYmJhQtWpVNBoNx44dw9XVlbCwMCZPnsyFCxeoU6cO/fv35+HDh6xYsQKNRsPKlStTVMIFULp0aUaPHk23bt2wtrYG1IRLo9Gg1Wpp3769JFxCiHiTfbriSSpdQqRcnlc92dByAz6P1WpRmoxpqPxzZSoMqvDZFvBfEqGNIP2M9PiG+EY7bmthy/Phz7GztNN5X2SVi6AM2Fpl5NHbs5+tdmkVLYX+KYTbO7cY57x+9iKLbfLeM0kYlqenJ127duXo0aOMGjWKSZMmYW5uHnX+4cOHdOrUCWdnZ/73v/8xYsQITN5v0HvhwgVu3bpFr169jBV+omrdujXPnz9ny5YtdOvWjUePHrF69Wq2bt3KypUref78+WfXuwkhxKf03t48IiKCzZs3M3nyZCZPnsyWLVsIDw83ZGxCCPFVpS+YnlD/UGyz2lJ/Zn2GPhpKtV+rJSjhArj35l6MhAuga8musSZc8KHKVSpPU0rkbkBIcCh///13rNebaEwYUH6AznOXnl+Kd9wiZcuWLRsHDx7kf//7H9OmTWPp0qXRzn///fc8efKE06dPM2rUKPz9/aPOffvttyk24fL392ffvn2kSZOGkiVLcu/ePTJmzEj16tV58OABL1++5PTp08YOUwiRzOg1vfDWrVs0b94cLy8vChcuDMD06dPJlCkTu3btonjx4gYNUgghDC3EL4RLCy7hcdyDzns7o9FosLSzpMuBLmQunhkzK73+etTp8vPLOo83LNgw1nvCwsKYNHESeTOXI4OdulC/cPbvmD37b4YNGxZrtathwYaMODQixvErz6/QtFBTPaJPfOPHj4/aT0t8XSYmJlHtzsuUKUNAQAC3bt2iYsWKlC5dGk9PTypWrMjatWsZNmwY+/bto1w53U1bUop9+/YRHBzMsWPHaNGiBUuWLMHe3p6JEyfyxx9/AOr+XTVr1gTA1dU1qrlYfNjZ2eHo6GjQ2IUQSZdev1X06dOHYsWKcfnyZRwc1L1k3r17R48ePejXrx9nz541aJBCCGEowd7BXJh7gfN/nyf4nbo21e2gGwUbFAQge/nsBn/m7Ve3dR7/XHv4yCpXiwofEpESuRtw7/nxz67tKpKxCNZm1jH27rr16pYekSe+69evM2nSJGxt7GjevHnUzxTx9WzevJlcuXJhZmZG2bJluX//Pj179qRFixYsWrSIWbNmMWrUKCIiItiwYUOKT7pOnDiBpaUls2bNYuDAgWg0GgAmT55M7dq16dKlS1Q3UVdX16hGG/q4f/++JF5CpBJ6remytrbm8uXLFCtWLNrxmzdvUqFChRS3I/3HZE2XEMlT4JtAzs8+z8V/LhLiGwJAhkIZqD66OsU7FcfUPPHWZwzaM4iFl6M30jA3MSd0bKjO6z9ey1W7+MBo5y4+2PTFtV2F5xXm/pv70Y41LNiQfd/vS8C7SBxtWrdh/66jhGoD+X3Mr0ycONHYIaUqWq2WnDlzYmZmhpeXF8WLF6dLly6MGzeOLFmy8PDhQ0xNTYmIiKBbt24sW7Ysal1XSuXl5UVwcDB58+bVef7du3e8fPmSwoUL4+zsTLly5fjrv+kULFQgzs94cN+NX/qO4sqVK5QtG3sjHSFEyqFXpatQoUK8ePEiRtL18uVLChYsaJDAhBDCUF7decV/Ff4jLEDdrypTsUzUGFODb9p9g4lp4v8CqeuzrchPz3XRVeWKFJdql4km5nvSKtp4RPx1XL9+na3btlLEtDkBvGLWzNkMGzZMql1f0blz5/D09ARg+PDhTJkyBUtLS5o2bUqzZs0AdQ13q1atWLJkSYpPuACyZs362fMODg4x/hstWKgAxUp/k5hhCSGSOb3+9pw6dSo//vgjmzdv5unTpzx9+pTNmzczbNgwpk+fjq+vb9SXEEIYW8bCGUlfID1Zy2Sl/db2DLw+kOIdi3+VhAsgjXmaGMdCI0LxDvaOcVzXWq6PWVukjVrb9fbt2xjnFUXhhf+LOMVgbBMnTMTGLANZTUqR27QKwUEhn20UIgzP09OTbNmysWfPHmbOnBm1f1lwcDAvX74EwNzcnJUrV2JmZrh1jkIIkdro9Tdo06bqYuz27dtHfVob+Ulu5CdjiqKg0WiIiIgwRJxCCKE3jYmGLge7YJPZ5rMVpsRSML3uGQDOns7Uzlc72rHPVbkifa7a9cj7Ee+C38W4xzF90lo38nGVy8TEFEvsyEY5qXZ9ZW3btqVNmzYx/r+YPHky3t7eVKlShQMHDmBrq/8edUIIIfRMuo4dO2boOIQQIlElZGPjhIqtYcbxR8ejJV1fqnJF+rja9WknwxMeJ3TeUy5b0mp+EFXl0pSKOpbbtAqeQVf4+++/ZW3XV6Trg4gVK1aQM2dOxo8fLwmXEEIYgF5JV2SbVCGEEF9WIksJzE3MCdOGRTu+5OoSxtYYi7mpuiFtXKpcUWPGUu1afGWxzus/1ynxa/u0yhXJUiPVLmMKDAwkTRp1GmqaNGmYPXu2kSNKwh4/hilT0PToYexIhBDJhF4LGvbv3x9tY8D58+dTunRpOnfuzLt3Mae1CCFEamZlZkUjx0Yxjj/3e87q66uBuFe5Iula23XS4yTnnp6LcW3xzMXJ75A/ge/CcD5ey/UpWdtlHO/evaNy5cqMGTNGZ+MX8ZEnT6BCBfj3X7LPm2fsaIQQyYReSdeIESOimmTcuHGD4cOH07hxY9zd3Rk+fLhBAxRCiJRgYPmBOo//fPBnPP08Wb9+PY88HhEU4suJW//F6eud/zP8/f2YN28egWGB9N7ZW+czBpUfZJS1bLpEVrlyKVUx0cRs0/9xtUs+xPs6/P39ady4MdevX2fJkiW8evXK2CElbTlzQr16UKoULzt1MnY0QohkQq/phe7u7nzzjdoadcuWLTRr1owpU6bg7OxM48aNDRqgEEKkBPUL1Kdg+oI8ePsg2vF3we/otKUTI7OPpEaNmija+FQZLMmvqUHefHnpu6tvjLEB7Czs6FKySwKjNxxda7k+JWu7vp7g4GBatmzJ+fPncXBw4ODBg2TOnNnYYSU94eEQGgpp0oBGA//9B1otYffuGTsyIUQyoVfSZWFhQWBgIACHDx+mW7duAKRPn17axAshhA4mGhNm1J1B642tY5w74XECUxNTtu3fSnpr3RsexyYoLIi+u/qy9sZanefH1xyPnaWdXjEbWmxruT4la7u+jrCwMDp27MiRI0ewtbVl//79lChRwthhJT3v3kGHDmBjA1u2gIkJWFvH6VatotW5b54QIvXR62+CatWqMXz4cCZPnszFixdp0qQJAPfv3ydnzpwGDVAIIVKKVkVb0bF4R53njrofpdiCYuy4uyPOa2rOPD5D6X9Ls+bGGp3nK+eszLBKw/QN1+A+t5brU7K2K3FptVp69uzJjh07sLS0ZNeuXVSsWNHYYSU9d+/Ct9/CoUNw8CDcvh2n2x57P8Fx9jdUXFiFN4FvEjlIIURyoFfSNW/ePMzMzNi8eTMLFy4kR44cAOzbt4+GDRsaNEAhhEhJ/mn0Dznscug85+XvRcsNLSm3uBz/XfkPD2+PGAmYl78X62+u57vl31FtWTXuv7mvcyxbC1uWtViG6WcqSl/Tl9ZyfUrWdiWuEydOsGbNmqif5d99952xQ0p69u1TEy5XV8idG86eheLFv3jbCfeT1FnWAACfEJn9I4RQ6TW9MHfu3OzevTvG8U/by06bNo0BAwaQLl06vYITQghDCAwMxM/PjyxZshg7FDKmyciBLgeosbwGb4Pe6rzmqtdV+u3uF3V9rrS5MNGY4OnvyXO/5198hqWpJTs77qRwxsIGjT0hFixYAMBDDvFIeyRO90RowgkLCGH16tX88MMPiRleqlOrVi2WLl2KlZUVTZs2NXY4SYuiwMyZMHKk+n21auq0ws+sdXtw3w2Am+9uMeba+KjjPxX9Ea/7L/Dihc7rhRCph0ZJxN6wadOm5dq1a+TPn3RaFSeUr68v9vb2+Pj4kDZtWmOHI4SIgwYNGnLnzm1cXV2xtLQ0djgA3Hhxg/qr6+Pl72XQcW0tbNnRcUe0TZeTgqtXr3Lo0KF436fRaOjQoQO5c+dOhKhSn9DQUCwsLIwdRtL2yy9q0gXQpw/Mnw+x/Jm5urpSqFChDweaA2Xff78QPsm1Yrh//z6Ojo4JjVgIkQwkatJlZ2eHi4uLJF1CCKM5e/YsVatWBdQ9BQcNGmTkiD545vuMfrv7sdd1r0HGq5ijIstbLKdopqIGGU+kLAsWLGDx4sUcOHAgSVR9dXF1dcXPzy/e99nZ2RkueblwAerUgalTYcgQtVvhZ9y7f48NdzbQOGdjwrXhrHBbQdcCXbEytfp6MQshkjxJuuJJki4hkpf69Rvw6NFjChf+hqtXL+Dm5pZkql0AiqKwwmUFvxz8hTdB+i24tzG3YWyNsfxc5WfMTPSaNS5SuFWrVkV1Gp49W+0KmdTEqBrFU4KqRv7+YGv74fWrV5Ap0xdvexXwisx/qdMOR1QZwYx6M/R7vhAixZOfzkKIFOvs2bMcOnSQ2bP/o3Dhb2jatDpLlixJUtUujUZDj9I9aF+sPetvrmfBpQVc8bwSp3sLZSjEoPKD6FaqGw7W0lZd6LZt2zZ69uwJwA8//MDQoUONHJFukRWupSuXUKRI3Ncj3r17j17deutVIQNg/Xq1onXoEJQpox6LQ8J18dlFvnX6Nup1qSxf7sophEi9JOkSQqRYEyZMxNGxCPXrN8XExIQmTVozZcoUevfunaSqXQBpzNPQq0wvepXphdtbNy4+u8gVzyvcfnUb/1B/tIoWGwsbCqUvRLns5SifvTzFMhVD84WpTyJ1O3ToEB07diQiIoIePXpQvHhxrl69StmyZb98s5EUKVKYMmXLJP6DtFoYOxamTFFfL1wIixfH6VYnZyf67uob9frPen/yfcnvEyNKIUQKIUmXECJF+rjKZWKi7o4xcOBPNG26NclVuz5VIH0BCqQvQKcSnYwdikjGzp49S8uWLQkNDaVNmzbMmDGD7Nmz07hxY3bs2GHs8IzLzw+6dIGdO9XXI0d+SL6+oPv27qx0WRn1+ki3I0mucY0QIulJ1G3Sq1evjnUcd20XQghD+rjKFSlfvoJR1a6QkBAjRidE4tJqtQwYMIDAwEDq16/PmjVr2Lt3L+Hh4Rw4cED/qXgpwcOHULmymnBZWsKqVTB9Oph+fv84RVH45eAv0RIuj2EeknAJIeIkzkmXr69vnL8i7d27l2zZsiVK4EIIEZtz585x6NBBBg36OarKFWngwJ/w9PRkyZIlRopOiMRnYmLCrl276NatG1u3bsXS0pItW7ZQoEABQkJC2LvXMB0zjUFRFE6dOEWXTl0plK8wjeo1xtzMjLFjx3Du3LkYG4pHc/8+VKwIt25Btmxw8qRa8YqDh+8eMv/S/KjXQaODyG0vWxkIIeImzt0LTUxMvrh2QFEUNBoNERERBgkuKZLuhUIkfQ0aNMTd3YPt24/FSLoARo4cjLPzuSTXyVAkX5E//4wtPDwcM7OYKwf8/PzIlCkTU6dOZc2aNeTPn5+NGzcaIcLYOTs7U65cOc5ePB3rmq57d+/RtVM3bty4SeH8eWlZ7zvs7Wx56+3DlgNHcX/yjIoVKrB+wwby5csXc4CICGjaFF6/hu3bIUeOL8b1JvANXv5eFMtcjE23NvEi4AVDKg5J4LsVQqQ2cV7TdezYscSMQwghDOLcuXMcPHgg2lquTyWXtV0ieVizZg1Tp07j/Plz2H7cdvwr8/Lyom7durRv355t27ah1WqjzgUGBhISEkLr1q0JDg5m/PjxlCoVvdtehgwZ2Lt3L1ZWn99fylhu3bxFvVr1yZYxPYdXLqRWpfLREt1pI37gwKlz/Dj5LypXrsTp02coWLAghIWBoqgbHJuawoYNYG4OcVj+sOveLpqvb052u+xc7HORdsXaJeZbFEKkYIm6T1dKJJUuIZK2L1W5Ikm1SxhCSEgI+fLmx9PrOdOnT2fkyJFGiePt27fUrFmTmzdvkjt3bhRF4cmTJzg6OtKgQQMAHB0d+fHHH3n16hXTpk0jNDSUoKAgVq1aFdVsY+PGjZ/9/yYxfa7SFRQURKlvSpPeNg3HVv9LurR2sY7z8s1banTqi8bCkpvHT2DWsSM4OqqdCeNRjRx7dCz/O/U/AMxMzHAZ4MI3mb7R780JIVK9OFe6rl+/HudBS5YsqVcwQgiREHGpckWSapcwhKVLl+L1wgsHCjBt6nQGDRr01atdfn5+NGrUiJs3b5I1a1aOHDlChgwZ6NevH5s3b6ZWrVrMnj2bNGnSAJApUyZmzpzJrVu36NixIyYmJixYsIABAwYkiSmSumzeuJmnT59x+OCWzyZcAJkzpGft7D/o0bILIaVKYfbyJVy5Ar//DrqmHOpQeUllzj89H/XauZ+zJFxCiASJ95quL10ua7qEEMYS1ypXJKl2iYSIrHJpX2Qgr1KLS5r5TJ025atWu4KDg2ncuDHHjh0jffr0nDhxguLFiwPqOjMnJyeGDh1KsWLFuHTpUtR9Li4uVKpUiQIFCrB+/fqoe4zpc5WuapWqkymNFXvHj4RnnhAcAmamYGcHjvnBJk30wQ4fJ2jwSNJotZA/v9qpsFixL8YQFhGGxf8soh17N+od6azSJfTtCSFSuThXutzd3RMzDiGESJDIKlfjxi3ZunVdnO7JkiUrz549w8nJicGDBydyhCKliaxylVdaYIU9WZXSX7XaFRYWRvv27Tl27Bi2trbs378/WvKk0Wjo27cvR44c4c6dO8CHhh+mpqYEBwczZcqUJJFwxerNG4IW/MvUy87UsLZGU6dljEsUjQby54UyJaBFY3B2QTN7IWmAI0CVEyewzpnzi4/yD/XHbuqHKpoGDeHjwjHRGGe6pRAiZYlz0pUnT57EjEMIIRLk4cOHAOzdu529e7fH6175UEnEV0hICJMn/Y/MFCMNGQHIRVUu+VxjwYIFX6XatWzZMnbt2oWVlRW7du2iQoUKOuPcu3cvv/zyC6dOnaJbt26ULl0aJycnChUqxPbt22nevHmixxpfmgdumE6dgemGzVgFB1MPIChI97WKAm7u6tfmnVHHH35XjYbHT/PEzIy47Bg6/tj4qO+bFWrGzk47P3O1EELEj16NNFauXPnZ8926ddM7oKROphcKIYRYuHAhgwcPobzSPyrpAnBlL0Hp3Hn8xCPRq11arZbhw4dTr149mjRpovOa3bt306xZM9q3b8/mzZupUKECDx48wMrKigIFCnDz5k28vLwwNzdP1FjjwtnZmQrlynG2cwcqbNmGSUhogsZ7nS8PVd09uOjtjb29fazXub9zJ2+6vPiH+tNmYxvG1xxP1dxVE/RsIYT4lF5Jl4ODQ7TXYWFhBAYGYmFhQZo0aXj79q3BAkxqJOkSQojU7eO1XEWUltHOBeOTqGu7FEVBUZQ4dxjs2bMny5cvx8TEhHHjxjF69GhevHhBly5dOH78OACHDh2ibt26Bo81vh5euMCTSpWoacAxgwHLJUvQ9Oql8/zvR37n7/N/M6nWJH6p8osBnyyEENHpNVH53bt30b78/f25d+8e1apVY926uK2lEEIIIZKjyLVcuZVqMc59vLYrICDA4M+eMmUKnTp1IjQ0blWgY8eOkStXLk6cOMH48eMxMzMjR44cHD58mP/973+YmpomjX04PT3J36OHQRMuACtA07s3zJwZ7biiKKSblo6pp6cSFB7E8UfH0Spa3YMIIYQBGHSfrsuXL9OlSxfu3r1rqCGTHKl0CSFE6vW5KlekyGrXtOlTGTFihMGe/c8///Djjz8CsHnzZtq0afPFe1xcXMiTJw/p0qXTef7OnTukT5+eLFmyGCzOePP2hurV4ebNz152GggvXYKaPTqBrQ1s2Ibm0HEAFEtLNCEhn3/O4sXQty9+IX6knRb953fY2DDMTOK8zF0IIeLNoC15zMzMeP78uSGHFEIIIZKMz1W5IkVWu6ZOmWawateKFSuiEq5x48bFKeECKFWqVKwJF0DRokWNm3ABDB78+YSrb1+4c4dzM2ZQ69oNpt+5ByvWf0i4hvSFS0dQ/pqMkvszXQoHD+bemZ3REq5imYqhjFck4RJCJDq9/pbZuTN6Rx9FUfD09GTevHlUrSqLT4UQQqQ8ujoWxubjToYJrXZt3bqVXu/XJA0dOpQJEyYkaLwkZft2WLtW97k8eWDFCqipTjr8pXBh7J4/p9bff6MBwszMePP7T5g3a8jbl6/Y5OnFyrAwBgM/6BjuevowSh1uEfV6eKXhzGwwU8eVQghheHpNL/x0Aa9GoyFTpkzUrl2bmTNnki1bNoMFmNTI9EIhhEidYutYGJvIToZPnj7GxsZGr2ceOHCAZs2aERYWRq9evfjvv//i3EQjyQsIgIIFwcsr5rmiReHoUcia9cMxFxeUmjXR+Pjw0tKSphERXAoPjzptZWVFx44d+eGHHyh75gy8rwwC+FpCgR/h9ft/DRvbbqRdsXaJ9c6EECIGvSpdWq0sNhVCCJF6RFa5HJT8mGBGMN5fvCcTxXDxdta72uXn50fnzp0JCwujXbt2LF68OOUkXKBWuHQlXOnSwYED0RKuN2/esGLvXmoGBxMMtA4J4SXqh8AVK1SgVevW9O7dmwwZMqg3lC0LT57An3/iawlpQ2DSMVhdElafzUK+0a2+xjsUQogoBm2kkRpIpUsIIVKfo0ePUqdOHb3uLVmyFC4u1/S698SJEyxYsIBVq1ZhYWGh1xhJkqKoidG1azHPrVgBkft9hoRw6epVmjZvjrf3O3pVLEGb2pXImTUjIWHhnLjlyqL9Z7jzxJNevXqxaNGiqD3HwgL96T4kB1dsfDm3BNIHQbgJmGmBrVuhlSReQoivR6+kKyIiguXLl3PkyBFevnwZo/J19OhRgwWY1EjSJYQQqY9Wq+XMmTOEfKlDng4FCxYkb968cb5eURQ0Gk28n5OsXL2qJl2fKlcOLl0CjQZevCCgYUMW3brF5gK52DyyN1kdYv7cVRSFFUcvMGDher7v0oVly5bh5e9F9lnZo65ZtRW6XP/opqZNYdeuRHhjQgihm17TC4cOHcry5ctp0qQJxYsXT/k/HIQQQqRqJiYmVK9ePdGf8+jRI9q3b8/SpUspXrx4oj/PaE6f1n188GA14XJ2RmnRApunT+lroqHP0C7Y6ki4QF1X3qNOJSzMTOn29woK1SnE6Iejo873uW9Ll+v+0W86c0attsnvL0KIr0SvSlfGjBlZuXIljRs3ToyYkjSpdAkhhEgMnp6eVK9eHTc3N6pXr86JEydS7oeaPXqo0wg/ptGoe3bt36+eDwriLvByyPdUrVMpTsM6Lp2Ie7nXUa/nNJzDj1ufwYwZMS92c4P8+fV9B0IIES96VbosLCwoWLCgoWMRQgghUqU3b95Qr1493NzcyJcvH+vXr0+5CRfoXsvl6Ah//QWTJwPgkj0bvZRwzteqGKchBz5ZHy3hOtnjJNXzVAePjbHHIEmXEOIr0asN0s8//8ycOXOQHhxCCCFEwpw8eZKKFSty69YtsmfPzuHDh8mePfuXb0zO3ryJeSwwMCrh4uef6ZrWniplv4lTx8bVby/y39szUa8XFlyoJlygtp+PawxCCJFI9Kp0nT59mmPHjrFv3z6KFSsW1Sko0tatWw0SnBBCCJGSBQUF0bJlS969e0f69Ok5dOgQ+VND9SUsLOaxXLng1Sv491/o3h2/TZtIY/n5jo0RihZTjQn17YqS1yI9Za1ysXWEC5b/WX64yNJS9826YhBCiESiV9KVLl06WkmrVSGEECJBfv31V969ewfAyJEj+eabb4wc0VeiKxFKlw4ePICcOd+/TMcrH79Yh3gc+pauj1fQK31luqevxCXHUQT6hLI1woV06dJ9uNDXN+4xCCFEItEr6Vq2bFmcrjtz5gzly5fHUv5iE0IIIWKITLKyZs3KNV3rnFKqnDnh8ePox1xcohIugPoNG/LfwvnMCQmNUfHa5n2Ndh5LALgf8pI29mVwMEuD08mzWFpaRO80Gduf60fPEkKIxJaoW9s3atSIZ8+effaa1atX079//6jkTKPRsHz5cp3XTpgwAY1GE+vXo0ePdN534MABatasiZ2dHWnTpqVWrVocOXIkge9OCCGESJi9e/dSpUoVBg8ezO7duwkODjZ2SIkvLAz8dFSwnj+HJ0+iXvbv3x9v/0DWn7oS7bJRz7dHJVwA+/MPxtbUkogILYsPnqVDh45kzJjxww0XLuiOo1y5BL0NIYSID70qXXEVl0YbY8aMwcPDg4wZM5ItWzY8PDy+eE/37t11bjQZbTrBe6tXr6Zr165kypSJHj16ALBhwwbq1avHxo0badu27RefJ4QQQiREQEAALVq04N27dzx79gwzMzOyZs2Ki4sL06dPp1GjRowdO5ayZcuSJk2aqPtMTU2ZM2cOlSrFrWV6kvfmDbRvDzdu6D6/YgWMGQNA/vz5adWyJaNW7qBS4bx8kysb5e5NwyX4w4e5NwqPpqhVVhRFYZjTZh6/esuPP/74YbyAANi0KeZzcueGjxMzIYRIZImadMWFk5MTjo6O5MmTh2nTpvHbb7998Z4ePXrw3XffffG6d+/e8cMPP5AxY0acnZ3J+X4qwahRoyhTpgwDBw6kQYMG2NnZJfRtCCGEELGysLDA3NwcZ2fnqGPffvstVatWpXv37mTIkIGpU6dGzdg4ePAg7u7u5M6dmwwZMhgpagO7dQuaN4eHD8HGRk2IPrVoEfz8M1hbA7Bk6VJqVK/Gd2Pn8van6Bscvy0+g7Sm1jx+9ZYxa3az9sQlFi9eTLmPK1grVoCPT8zn1KtnyHcmhBBflKjTC+Oibt265MmTJ1HG3rRpE97e3vzwww9RCRdAzpw5GTJkCK9fv2bbtm2J8mwhhBAikrm5OQ0aNIh6nStXLmbMmMGcOXOikqpff/2VWbNmodVqcXd3p127dri4uODo6GissA3n3DmoXFlNuPLlU183bRrzumfPYPz4qJfp0qXj6PHjmHa3ijpmrjVhnmd7lu09T6upiyk4YCK7rtxhzZo19O3b98NYz5/D6NG64+nf31DvTAgh4sToSZc+Tp48yfTp0/nzzz/Zvn07/v7+Oq87fvw4APXr149xLvKH34kTJxItTiGEEALUBlQ//fQTAP369cPCwoIyZcpw9+7daNdVqVKF1atX4+TkxIYNG3ROm0+WihVTp/R99x1cvAglSsCgQbqvnTkTPlp3nTFDRoa0/gFLE0uK+RUj3QIHfvxvM2PW7eVpqAkLFizg2fPndO7c+cMYYWHQowd4e8ccv3x5qFDBkO9OCCG+KFGnF2o0mkQZd/xHn4KB+knYnDlz6NatW7Tjrq6uADo/JYw8FnlNbEJCQggJCYl67Rtb61khhBBCh82bN9OnTx8Ahg8fzl9//cXmzZtp3749pqam0a4NCAigQ4cO9O7d2xihGlZICFhYgEYDadPCoUPqOqrIvT0bNICKFdUk7GNaLbRsyYXV05kcsJeN7TYytsZYOhXvhGMGR/hLXTMe6+8YoaHw/ffq83QZN85w71EIIeIoUStdcWmkER+lSpVi6dKlPHz4kKCgINzd3fnnn3/QaDT06NGDnTt3Rrve5/08bnt7+xhjpU2bNto1sZk6dSr29vZRX7ly5TLQuxFCCJHSubu707lzZ7RaLX369OGvv/5Co9Gwfft2SpQogYODA61bt+abb77h8uXLtG7dmt27dxMeHm7s0BPm2TOoVk2tWkXKlu1DwgVgYgJLl6qJ2Sc25PGn5uXB7HHdw4QjY9FoNGrC9V6sCde1a2oit3mz7vOdO0OzZnq8ISGESBi9kq7atWvjraNk7+vrS+3ataNe+/n5kT9/fr2D+1SrVq3o2bMn+fLlw8rKirx58zJkyBA2ve9MNOZ9xyND+u233/Dx8Yn6evJRO1shhBDic/Lly8fs2bPp2LEjixYtQqPREBISwu7du8mcOTOlSpXi5MmTWFpaUqVKFR4/fszr1685deqUsUPX34UL6hS+y5dhxgzdjSwiFSsGkyZFvVSASTWhYzsIMYOm92DsL7tg/vzYNzkGuHQJevZUpw26uOi+JmtWmDtXv/ckhBAJpNf0wuPHjxMaGhrjeHBwsFF+UNSpU4cCBQpw48YNfH19o6pYkRUuHx+fGN2fIqcJ6qqCfczS0lI2dxZCCKG3wYMHM2jQoKjqzOHDh/H19eXIkSPUqlWLVatWkSlTJkaPHs1ff/0FwJYtW6hVq5Yxw9bPqlXQt686tbB4cdixA77wc5YRI+DGDYI3rKFPc1hTUj380zn48yCYKq4wZAj89BOULAllykD69BAeDq6uanLn6fn5Z9jaws6dkFI6QQohkp14JV3Xr1+P+v727dt4eXlFvY6IiGD//v3kyJHDcNHFQ8aMGXnw4AGBgYFRSZejoyOXL1/G1dU1RtL1ufVeQgghhL5u377NyJEjWblyJenTpweiT4c7evQopqamTJo0iVGjRkWt6/rzzz+pW7cu3bp148hHjSSShYgI+O03+PNP9XWLFmoCFpctWUxMeDlvOq1yHuCs9WtMtTB/D/S/8sl1YWFw5Yr6FR9p08KePdI8QwhhVPFKukqXLo1Go0Gj0USbRhjJ2tqaf/75x2DBxVVAQAC3bt3CxsYm2i70NWvWZN26dRw8eDDGxpIHDhyIukYIIYQwBHd3d+rVq8fz588ZPnw4y5cvj3HNTz/9RK9evShWrFiMcw0aNODmzZvJayq7okDr1molCdQ27ZMmqWu24iiQMFwzaLAPsGTz2jDqPtAaJrbixWHtWrVbohBCGFG81nS5u7vj5uaGoihcvHgRd3f3qK9nz57h6+tLr169EiVQPz8/7t+/H+N4UFAQffv2xc/Pj/bt22Nm9iGPbN++Pfb29vzzzz88ffo06vjTp0+ZN28eGTNmpFWrVokSrxBCiJTh3r17zJo164vXPX/+nLp16/L8+XOKFSvGzI+bSHwkZ86cOhOuSJkyZaJs2bJ6x/vVaTTqZsNWVrB+Pfzvf/FKuADypsvL7s67OTf4KnU3XEp4kmRqqiZ/ly9LwiWESBI0iqFbDMaTk5MTp0+fBuDGjRs4OztTtWpVChYsCEC1atXo06cPjx49In/+/FSoUIGiRYuSNWtWXrx4weHDh3n69CklSpTg2LFjMaYRrl69mq5du5IpUyY6dOgAwIYNG3j9+jUbNmygXbt28YrX19cXe3t7fHx8oqYxCiGESLl69+7N0qVLcXNzi7U51OvXr6lZsya3b98G4OrVq5QuXforRmkEoaEfOg8qCjx+DHnyxPn2BZcWkDNtTpoXbh7zZFgYbNwICxbA2bNxjyltWnV/riFDQJYPCCGSEkUPy5cvV3bv3h31esSIEYq9vb1SuXJl5dGjR/Eaq3v37gpqwyKdX927d1cURVF8fHyUwYMHKxUqVFAyZcqkmJmZKXZ2dkrFihWVGTNmKIGBgbE+Y9++fUr16tUVGxsbxdbWVqlZs6Zy6NAhfd664uPjowCKj4+PXvcLIYRIPsLCwpQMGTIogDJjxgyd1/j4+CjlypVTAMXCwkIBlIULF37lSL8irVZR5sxRlG++UZS3b+N9e1hEmPLD3h8UJqDY/GGjuL9z//wNLi6KMnWqorRtqyh58yqKRqMoapqnKDY2ilK5sqL88IOirFmjKP7++r0nIYRIZHpVugoXLszChQupXbs2586do06dOvz999/s3r0bMzMztm7darCkMKmRSpcQQqQeR44coW7dujg6OpI+fXrOnz8f45r27duzadMmMmTIwNu3bzE3N6dGjRocim1z3uQsNBQGDYIlS9TXs2apXQUBXryAly/VaywtIXt2tcvgR3xDfOm4uSP7HuwDYGqdqYyqOir2fbd0URT1GaamYKZXE2YhhPjq9Prb6smTJ1HT/7Zv307btm3p168fVatW5bvvvjNkfEIIIYTRbNmyhTx58jBu3Di6du3K06dPyZkzZ7Rr/ve//3H37l2aN2/O1KlTGT16NJMmTeLNmzcxprwnay9fQps2cPq0umZr7Fh1GmCzZuraqY86GkfJkwfKlYM6dXjUtBrN9nzPzZc3sTazZlWrVbT5pk3849Bo1KROCCGSEb0qXZkzZ+bAgQOUKVOGMmXKMHz4cLp27YqbmxulSpXC398/MWJNEqTSJYQQKY+iKDRt2pTHjx9HO+7m5sbAgQMZN24cmTJlIlu2bKRNmxZfX9+o/Srr169PsWLFWLduHQ4ODqxZs4YcOXLQvXv3GA0zmjZtSpEiRb7a+zKYa9fUNvCPH6t7XpUrp661CguL0+3nckLLTvDSBrJaZWJX172Uz14+cWMWQogkRK9KV7169ejTpw9lypTh/v37NG7cGIBbt26RN29eQ8YnhBBCfBV58+Zl7969mJiY0K1bN+zt7alfvz5Dhw7F3t6e//77j6tXr6IoCvPnzyciIgKArVu3smPHDkxMTBg1ahTZsmWjffv2bNu2jW3bthEeHk5AQAAAWq02+SVdR4+q1azAQEiXDry94cSJeA2xqZiacJX2hJ2bvcmlPQQjSsv0QCFEqqFXpcvb25uxY8fy+PFjBg4cSMOGDQEYP348FhYWjB492uCBJhVS6RJCiJRrw4YN9OvXj/Tp07Nu3boYezyGhIRQpUoVnJ2do47169eP2bNnkyZNmhjj3b59m44dO3L//n1mzZrFwIED47d+KSl49QpKl1aTrcBAvYYIN4Fp1WDYebANfX+wQgXYvBly5zZUpEIIkWTFO+kKDw9nypQp9OrVK8a89tRAki4hhEjZ3N3d6dy5M5cuXeLgwYPUrl076lzBggVxc3MDYOnSpURERPDjjz+SN29edu7cGbXeGWDNmjX07duXfPnysX79ekokp/2iwsLA3Fz9/vx5aNAAfH3jfHuwGfxdCX4+C+af2+c4Vy44ckTauwshUrx41/XNzMyYMWMG3bp1S4x4hBBCCKPKly8fs2fPpnLlylFVKUVR+Pvvv6MSrrlz59KzZ08ASpYsybfffsvBgwejJV3//vsv+fLl49KlSzqrYEmWh4e6fuunn6BMGWjY8PMJl0YDlSqplatixXhhHkrL57M4H+7OkwIZmb/idez3PnkCdeqo68NS4Qe5QojUI35bxr9Xp04dTsRzPrcQQgiRXGzbto1MmTJRunRpunbtSrp06Rg+fDgAGo2Gzp07R1378OFDAJo0acKZM2f4+eefef36NU2bNsXd3R09ZvEbz+nTavLk4gJjxkCrVuDjo/taS0sYMQLc3NSkac4cbraswrc+f3E+3J10VuloM34D3LgB/fqpHQ91efIEOnaE92vkhBAiJdJrTdeiRYuYOHEi33//PeXKlcPGxiba+ebNdewun0LI9EIhhEjZFEWhYMGCpE+fnrdv3+Lp6UlQUBCg7lN57949lixZQq9evQBo27YtDx8+pGXLlkycOBGArFmzMmnSJPr06cPmzZtp00aP1uhfm5OTugdXWJi6hqtMGVi2TPe1334Ly5fDR01B9rnuo8PmDviF+lEwfUH2dN5DoQyFPtxz/jz06AH37ukec/ZsGDbMQG9GCCGSFr2SLpPYPq1C/QQwIgV/WiVJlxBCpGzXrl2jTJkyAFSsWJGVK1cydepUXF1duXTpEmFhYTRu3Jg9e/YQGBhIpkyZCAkJQVEUxo4dS69evejevXvUjJAOHTqwbt06Y76lzwsPh59/hrlz1dft2sHgwVCrlroR8adatoT166PtlfXPhX8YdmAYWkVLzTw12dJ+CxnS6NijzNsbGjeGc+dinrO2hjt31L29hBAihdEr6UrNJOkSQoiU7e+//2b48OGMGjWKSZMmYW5ujlarRVEUbty4QadOnfDw8CAgIIAdO3bQqlUrcubMyZo1a6hRowYAERERTJ8+nXHjxmFtbc2bN2+wsLAw8jvTISICGjWCQ4fU15Mnw+jRauK1ZUvM67/9Fk6cwPXxY/z8/AB4FfyKNsfaEBAeQPNczfm95O+Ym5jrfJydnR2OmTJB2bLg7h7zgp9/hr/+MtS7E0KIJEOSrniSpEsIIVI2Hx8fPDw8KFmypM7zAQEB3Lt3j7Jly3L9+nWcnJyYMGEC6dOnj3HtuXPnWLVqFf/88w+mpqaJHbp+JkxQE51Vq9Q1XM+eqdWmT2etWFnBtWu4mphQqFCh6OcKAlmAM19+3P3793F89kytpH3KwUF9vrW1nm9GCCGSpjgnXXPnzqVfv35YWVkxN3IKQix+/PFHgwSXFEnSJYQQItn7uCW8VguPHkH+/OrrKVPUatenfvsNpkzB2dmZcnXK0f73NlQoUi7Oj3z64Blzhs/nypUrlC1bFjp3Bl3TLlevhu+/j/97EkKIJCzOLeNnz57N999/j5WVFbNnz471Oo1Gk6KTLiGEECLZUhT48091U+LjxyFNGrWrYGTCBaCrO7GJCQwYAIDLWxfoC7tD9vFdnupktcuiXyxDhuhOuk6elKRLCJHixDnpunbtGvb29oC6caQQQgghkpHgYOjbV60kgdoM430HxiiKAleuxLy3WjXInZu1N9Yy4NwAsAEHi3SYmcR7u88PKldWpzF6eEQ/ruv5QgiRzMV5n6706dPz8uVLAGrXro23t3dixSSEEEIIQ3r+HGrWVBMuU1OYPx/eb+4czdOn8OZNjMNKhfKMPzae77d+T6g2FO7AoEL9yGijo0NhXGk0UL58zOPXr6sdFYUQIgWJc9Jla2vLm/d/ER8/fpywsLBEC0oIIYQQBnLpkrrh8cWLkD49HDyo7sel0cS89v2Hqx8LMoPOWc4w6eQkALoX6A4bwcLUAN0YixePeSwsLPYNmYUQIpmK87yAunXrUqtWLYoWLQpAq1atYm1/e/ToUcNEJ4QQQgj97d4NbdtCSAh88w3s3AkFCsR+fWhojEMzqsL6wAuYmZixqMkiylCGFcoKw8SXJo3u4yEhhhlfCCGSiDgnXatXr2bFihW4ublx4sQJihUrRprY/rIUQgghhPGVKAFp06r7a61Zo37/OTo+TB15Bi42KsGIDnP5Lu93ODs7Gy6+gADdxz/aeFkIIVKCOCdd1tbWDHjfuejy5ctMnz6ddOnSJVZcQgghhNBHeDiYvf/xnicPnDsHefOqa7m+JFs2AC5lh3KeYKKAdTjs8aoDeb8zfKzXr8c8ZmkJ8vuFECKFifOaro8dO3YsTglX2rRpefjwoT6PEEIIIUR8ublBmTLqNMJIBQrELeEClKxZmVPXlkp9YPx3H524cMGgYaoPU9T1Zp8qVSrO8QohRHKhV9IVV3Hcd1kIIYQQCXX0qNow4+ZNGDUq3h0AwyLCGLxvCMOq+aM1AS9biPopfu6cmtAZ0vHj8OxZzOPl4r7hshBCJBeJmnQJIYQQIpEpitoCvn59ePcOKlZUEzCzuO+h5R3sTZO1TVh4eSEaNPx1ABbvgmj9DefNM2zc//yj+3jt2oZ9jhBCJAEJ2NVQCCGEEEYVGgo//ACLF6uvu3SB//4DK6s4D/Hw3UOarm3Kndd3SGOehrV1FtBiSj/gk06G8+ZBjx5RL58+0FGl+oxo1+/dC9u2xbwoc2Zo3jxe4wohRHIgSZcQQgiRHAUHQ4MGcPKkuufW9Onwyy+699+KRVBYEDWW1eCZ3zNy2OVgV6ddlMlWBtodUrsdfiw8HLp1I+3KlQDMGT5fr7DThYRAv366T/btq7ODohBCJHeJmnRp4vEXvxBCCCHiwcpKbQl/7RqsWweNG8d7CGtza6bVncacC3PY0XEH2e2yqydGjoT16yEiIvoN169TcPhwXK9exVerjffz0oWEkL9/f91ruezsYPDgeI8phBDJgUZJxG4XdnZ2uLi4kD9//sR6xFfn6+uLvb09Pj4+pP3SfidCCCGEoUVEfOjuFxYGjx9/fsPjT2gVLZ5+nuRImyPqWLg2HDOTTz6HHT0apkzRPUjRorBihdq4I64OHIA+feDpU93nFy9WK11CCJECJbiRhqIosXYp3LdvHzly5NB5TgghhBDxoCjwv/+pUwrDwtRj5ubxSrgCwwLpsLkDlZdUxsvfK+p4jIQLYNw4KF5c90B37kClSmoSde3a52M+dQratYOGDWNPuOrXV8cSQogUSu9K18qVK/nzzz9xdXUFoFChQowYMYKuXbsaNMCkRipdQgghvrqAAOjZEzZtUl9v2gRt28ZrCC9/L5qva86l55cwNzFnc/vNNC/8haYVrq5QrRq8fPn56woXhvLloVgxsLYGf3914+NLl+DRo8/fW6gQnD4NmTLF6/0IIURyotearlmzZjF27FiGDBlC1apVATh9+jQDBgzg9evX/PTTTwYNUgghhEi1Hj+Gli3h6lW1sjV/frwTLhcvF5qta8YT3yekt07Ptg7bqJGnxpdvdHSEw4ehXj148SL26+7dU7/iq1AhdXxJuIQQKZxela58+fIxceJEunXrFu34ihUrmDBhAu7u7gYLMKmRSpcQQoiv5uxZaNVKrTRlygRbtkD16vEaYvf93XTa0gn/UH8KZSjEns57KJi+YPzicHNTpwhevRq/+z6nbl21Q2LmzIYbUwghkii91nR5enpSpUqVGMerVKmCp6dngoMSQgghUr1Nm+C779SEq2RJdapePBOubXe20WJ9C/xD/amdrzbne5+Pf8IF6rqxCxdg0iS12pYQtrawaBEcPCgJlxAi1dAr6SpYsCAbN26McXzDhg04OjomOCghhBAi1fvmG7C0hNat4cwZyJMn3kPUyleLwhkK07dsX/Z/vx8Hawf94zE3h7Fj4eZNGDRITZ7iI0MGGDUK7t6F/v3jtZ+YEEIkd3pNL9yyZQsdOnSgbt26UWu6zpw5w5EjR9i4cSOtWrUyeKBJhUwvFEIIkWg+bgcPapfAwoXBJO6fkQaGBWJtZh21V6Z3sDf2lvaG3zvTzw82blQ3Z75yRY314727zM3V7oflykHt2uo0SSsrw8YghBDJhN7dC69cucLs2bO5c+cOAEWLFuXnn3+mTJkyBg0wqZGkSwghRKK4d09tkLFgQbynEUZye+tG03VNGVh+ID9++6OBA/yCoCB49w5CQ8HCAjJmVP8phBAicTdHTokk6RJCCGFw+/dDx47g46NWhi5divf0u1Mep2i1oRVvgt6Qxz4PtwffJo15mkQKWAghRHzotabL1NSUlzr27Hjz5g2mH0+LEEIIIUTsFAVmzoQmTdSEq2pV2LMn3gnXSpeV1FlZhzdBbyifvTznep+ThEsIIZIQvZKu2IpjISEhWMhUAiGEEOLLgoPVDY9/+UVdC9W7Nxw5AlmyxHkIraJl9JHRdN/enTBtGG2/acuJHifIZpctEQMXQggRX/HaHHnu3LkAaDQanJycsP2oc1FERAQnT56kSJEiho1QCCGESGn8/KB+fTh/Xm2SMXs2/PBDvCpciqLQeUtnNtzaAMDo6qOZVGsSJhq9Pk8VQgiRiOKVdM2ePRtQ/6JftGhRtKmEFhYW5M2bl0WLFhk2QiGEECKlsbWFvHnV9ukbN0K9evEeQqPRUClnJbbe2YpTcye6lepm+DiFEEIYhF6NNGrVqsXWrVtxcEjAfh/JlDTSEEIIoTet9kP798BA8PRUNx6OB0VRotq/K4rCg7cPcMwge2QKIURSptcchGPHjsUp4UqbNi0PHz7U5xFCCCFEyqHVwujR0L79h72s0qSJd8K1895OKi2phE+wD6BWuyThEkKIpC9RJ35LN3ohhBCpnp+fujHwlCmwZYvaLCOeFEVh5tmZtFzfkovPLvLX2b8SIVAhhBCJJV5ruoQQQggRD+7u0Lw53LwJlpbg5BTv9VthEWEM3juY/5z/A6B/uf6MqzkuMaIVQgiRSCTpEkIIIRLD8ePQti28eQPZssG2bfDtt/Ea4l3QO9puastR96No0DCrwSyGfjs0ak2XEEKI5EGSLiGEEMLQVq5U990KD4fy5WH7dsiRI15DPHz3kEZrGnH/zX1sLWxZ12YdTQs1TZx4hRBCJKpETbrkkzghhBCpkqOj2qWwc2d1SqG1dbyHsDazJigsiFxpc7G7825KZimZCIEKIYT4GhI16ZJGGkII8XW9fAmZMxs7ilTq43bwlSvD5ctQvHi8Njz+WDa7bOz7fh8Z0mQgq21WAwYqhBDia0tQ98LQ0FDu3btHeHi4zvP79u0jRzynUwghhNDPo0dQsiQMHw4REcaOJpW5eRNKlwYXlw/HSpSIV8KlVbT8dvg31lxfE3WsWOZiknAJIUQKoFfSFRgYSO/evUmTJg3FihXj8ePHAPzwww9MmzYt6rpq1aphaWlpmEiFEELEytsbGjeGFy/g6FEICjJ2RKnIjh1qZevGDfjpJ72GCAgNoN2mdkw7M43eO3vz1PepgYMUQghhTHolXb/99hsuLi4cP34cKyurqON169Zlw4YNBgtOCCFE3Bw7Bvfuqb0adu8GW1tjR5QKKIq691arVuDvD7VqwaZN8R7mud9zai6vydY7W7EwtcCpuRM50+ZMhICFEEIYi15rurZv386GDRuoVKlStGYZxYoVw83NzWDBCSGEiJtWrdQGeblzQ075fT3xBQaq3QnXr1dfDx4Ms2eDuXm8hnH2dKb5uuY883tGxjQZ2d5hO1VzV02EgIUQQhiTXknXq1evyKxjpXZAQIB0LBRCiK8oJETdcxegWTPjxpJqvHkDDRrAlStgZgbz5kH//vEeZsfdHXTe2pnAsEC+yfQNuzrtIr9D/kQIWAghhLHpNb2wfPny7NmzJ+p1ZKLl5ORE5cqVDROZEEKIz1qxAsqUAXd3Y0eSytjbQ8aMkCEDHD6sV8IFcOn5JQLDAqlfoD5ne52VhEsIIVIwvSpdU6ZMoVGjRty+fZvw8HDmzJnD7du3OXv2LCdOnDB0jEIIIT5x9Cj06aPuvbtmDYwZY+yIUoHIlvBmZuq0wnfvIF8+vYebVGsS+dLlo3vp7piZJOoOLkIIIYxMr0pXtWrVuHbtGuHh4ZQoUYKDBw+SOXNmzp07R7ly5QwdoxBCiI/cvg2tW6sJV4cO8Pvv0c/LHokGFhEBI0bAwIFq8wyAdOninXC9DXrLT/t/IihMbS1pojGhd9neknAJIUQqoFHkp3O8+Pr6Ym9vj4+PD2nTpjV2OEKIVMbLCypVAg8PqFpVnd32URNZVqxYwahRo1ixYgUNGjQwXqAphY8PdOwI+/err8+fh2+/jfcw99/cp+napri+daV3md44NXcycKBCCCGSMr0qXXv37uXAgQMxjh84cIB9+/YlOCghhBAxBQSozTI8PMDRUe1W+GnC1bNnT0zNzGjRogUHDx40Wqwpwv37aoK1fz9YW8OGDXolXMfcj1HJqRKub13JbZ+bod8OTYRghRBCJGV6JV2//vorERERMY4risKvv/6a4KCEEELE9MsvcPmy2r9h7161l0OkyISrZ68+3Lr7gDp169G8eXNJvPR18KCaYN27p/bgP30a2reP9zBLnJdQf3V93gW/49sc33Kxz0VKZCmRCAELIYRIyvRKulxdXfnmm29iHC9SpAgPHjxIcFBCCCFiGj8eqlWDnTuhYMEPxz9OuP5ZsAgrKyvWbtgsiZe+Fi2CRo3A2xsqV4ZLl6Bs2XgNoVW0jDw0kj67+hCuDadj8Y4c636MLLZZEidmIYQQSZpeSZe9vT0PHz6McfzBgwfY2NgkOCghhBAxZc0KJ09ClSofjn2acJmYqH+tW1paSuKlrzx51H/26AHHjql/8PH03O85S68uBWB8zfGsbb0Wa3NrAwYphBAiOdGrkUb//v05d+4c27Zto0CBAoCacLVp04YKFSrg5JRyFwhLIw0hxNe0fTv4+0OXLjHPxZZwfSwkJITOHdpy5PAhdu7cSf369RM/6ORIUeD9npOAuvFx2bLRj8XTKY9TPPF9QucSnQ0QoBBCiORMr6TLx8eHhg0bcvnyZXLmzAnA06dPqV69Olu3biVdunSGjjPJkKRLCPG1XLwI330HQUGwZw80bvzhXFwSrkiSeH3B1avQqxds3gzvP0jUh7OnM28C31CvQD0DBieEECIl0LtlvKIoHDp0CBcXF6ytrSlZsiQ1atQwdHxJjiRdQoivwd1dbQ3/8qW6vGjnTnVPXohfwhVJEq9YbNoE3burmW2rVrB1q17DbLuzjS7bumCqMeV8n/N8kynmumchhBCpl+zTFU+SdAkhEtu7d+q6rbt3oXRpdR2XnZ16Tp+EK5IkXh/RamHCBJg8WX3dsCGsW6duehwPiqIw48wMfj2idu5tWLAhG9puIK2l/HwQQgjxgd5J15EjRzhy5AgvX75Eq9VGO7d06VKDBJcUSdIlhEhMISHq7//Hj6udys+fhxw51HMJSbg+jC+JF/7+0K0bbNumvv75Z5g+HUxN4zVMaEQoA3YPYNm1ZQAMqTCE2Q1nY2ZiZuiIhRBCJHN6JV0TJ05k0qRJlC9fnmzZsqH5ZKHxtsgfZCmQJF1CiMSiKGousHq1Wtk6fRpKllTPhYeHkzZtWnLlzo2zyy1M45kgfMzPz4+ijvlwcHDA1dXVQNEnE56e0KAB3LgBFhaweLE6vTCe3gS+ofXG1pz0OImJxoQ5DecwpOKQRAhYCCFESqDXx3GLFi1i+fLldO3a1dDxCCFEqpY3r7p2a/PmDwkXgJmZGfPnz6d3794M+2Ewc+Yt0KvSFRoaSs9u3+Pv78/q1asNF3hy4eAAadJAlixqpatyZb2GmXNhDic9TpLWMi0b2m6gYcGGBg5UCCFESqJX0hUaGkqVjzeKEUIIkWAajbrEqFs3cHSMeb5nz54A9O7dGyDeiVdoaCidO7Tl8KGDbN++nYYNU1GiENkS3spKTbbCwyFXLr2HG1tjLE98n/BL5V8olrmYAQMVQgiREum1OXKfPn1Yu3atoWMRQohU6epVtXleJF0JV6SePXuyZMkSljgtZuiQQTHW1MYm1SZcYWHwww8wZsyHY9my6ZVw7bm/h3BtOADmpuYsa7FMEi4hhBBxolelKzg4mMWLF3P48GFKliyJubl5tPOzZs0ySHBCCJHS3boFtWpB0aKwezdkyPDle+Jb8Uq1CdebN9C+PRw9qla5vv8evol/K/cIbQSjDo9i5rmZ/FDxB+Y2mpsIwQohhEjJ9Eq6rl+/TunSpQG4efNmtHOfNtUQQgihm5eXuuGxjw+Ym4OtbdzvjWvilWoTrlu3oEULcHMDGxtYs0avhMs/1J/vt37Pzns7AciYJiOKosjPOiGEEPGiV9J17NgxQ8chhBCpSkAANG0Kjx9DoULqMiNLy/iN8aXEK9UmXLt3Q+fO4OendibZuRNKlIj3ME98ntBsXTNcXrhgaWrJ8pbL6Vi8o+HjFUIIkeLJZiJCCPGVRURAp05w5QpkzAh798ZtWqEusSVeqTbhmjULfvlFbZxRs6baBjJjxngPc+nZJZqvb46XvxeZbTKzo+MOKuWslAgBCyGESA30TrouX77Mxo0befz4MaGhodHObd26NcGBCSFESqQoMGwY7NqlVrZ27oQCBRI25qeJ15+z/qZLp/apL+ECtRW8osCAATB3rjpvM54CQgNosrYJrwJfUTxzcXZ32k2edHkSIVghhBCphV5J1/r16+nWrRsNGjTg4MGD1K9fn/v37/PixQtatWpl6BiFECLF8PKC9evV71ev1nubqBg+Trz279/Lq5cvU0/CFdkOHtRmGQUKQCX9q1I2FjY4NXfiP+f/WNN6DWkt0xooUCGEEKmVRlEUJb43lSxZkv79+zN48GDs7OxwcXEhX7589O/fn2zZsjFx4sTEiDVJ8PX1xd7eHh8fH9KmlR/EQoj4e/AATpyA94Upg1q2bBmjRo1i5cqVqSPhunhRbQm/Ywdkzar3MCHhIbh7u1MkY5GoY9IwQwghhKHolXTZ2Nhw69Yt8ubNS4YMGTh+/DglSpTgzp071K5dG09Pz8SINUmQpEsIoY+wML1muukl1SQLq1dDnz4QEgI9esCyZXoN8zrwNa03tObem3tc6HOBvOnyGjRMIYQQQq/NkR0cHPDz8wMgR44cUW3jvb29CQwMNFx0QgiRAjx8qO7DtX//13leik+4IiJg1Cjo2lVNuJo3V9dv6eHu67tUcqrEqcenCA4PxsPbw8DBCiGEEHomXTVq1ODQoUMAtGvXjqFDh9K3b186depEnTp1DBqgEEIkZ2/fqntxubnBuHGg1Ro7omTO11fdf2vGDPX177+r/fbt7OI91OGHh6nkVAm3d27kS5ePc73PUTNvTQMHLIQQQug5vfDt27cEBweTPXt2tFotM2bM4OzZszg6OjJmzBgcHBwSI9YkQaYXCiHiKiQEGjRQ12/lzAkXLkD27MaOKhnz8IBGjeDOHbCygqVL1d77evj38r8M3juYCCWCqrmqsq3DNjLZZDJwwEIIIYRKr6QrNZOkSwgRF4qizn5bs0Ytwpw5o9f+vOJjvr5qV0JfX9i+HcqX12uYlS4r6b69OwBdSnbBqZkTlmbx3JlaCCGEiAe9ki5TU1M8PT3JnDlztONv3rwhc+bMREREGCzApEaSLiFEXIwbB5Mng5mZuvlxvXrGjiiZivwRFblO7eFDsLaGbNn0HjIgNIDvVnxHy8It+b367yl/DZwQQgij02ufrtjytJCQECwsLBIUkBBCJHf79qkJF8CiRZJw6S00FAYNgmLF4Kef1GP58+s11MuAl2RKkwmNRoONhQ1nep3BwlR+XgkhhPg64pV0zX3fHUqj0eDk5IStrW3UuYiICE6ePEmRIkViu10IIVKFunXVTuZZsiTOXlypwsuX0KYNnD6t9tpv0wZy59ZrqIvPLtJ8XXMGVxjM2JpjASThEkII8VXFa3phvnz5APDw8CBnzpyYmppGnbOwsCBv3rxMmjSJb7/91vCRJhEyvVAIERefzooT8XDtmtqh8PFjSJsW1q1TW0DqYeOtjXTf3p3g8GBKZSnF+T7nsTKzMmy8QgghxBfEq9Ll7u4OQK1atdi6dWuK7lIohBDx4ekJ8+fDhAnqOi5JtvS0ZQt06waBgeDoCDt3gh4zKBRF4Y9TfzD2mFrZalqoKWtbr5WESwghhFHotabr2LFj0V5HRERw48YN8uTJI4mYECLV8feHpk3B2Rm8vWHePGNHlEz98QeMGaN+X68ebNgAevxMCQkPoc+uPqy+vhqAnyr9xJ/1/sTUxPQLdwohhBCJQ6/NkYcNG8aSJUsANeGqUaMGZcuWJVeuXBw/ftyQ8QkhRJIWHg4dO6oJV6ZMMHy4sSNKxiKnbP/0k9ryUY+ES6toabSmEauvr8ZUY8qiJouY1WCWJFxCCCGMSq+ka9OmTZQqVQqAXbt28ejRI+7evctPP/3E6NGjDRqgEEIkVYoCQ4fCnj3qXr07d+rdXC/1+nhZ8ZAhcOoUzJqlztHUg4nGhK4lu5LOKh37u+ynf/n+BgpUCCGE0J9e+3RZWVnx4MEDcubMSb9+/UiTJg1///037u7ulCpVCl9f38SINUmQRhpCiEizZsHPP6vrtzZtUhvsiXg4fRp+/13NVtOlS9BQIeEh0TY4fh34moxpMiYwQCGEEMIw9Kp0ZcmShdu3bxMREcH+/fup934TmsDAwGgdDYUQIqXasgV++UX9/q+/JOGKNycnqF1brWxNnJigoRZeWkiJhSV4FfAq6pgkXEIIIZISvZKunj170r59e4oXL45Go6Fu3boAXLhwQfbpEkKkClZWkCYNDB78Yd9eEQfh4eqczL59ISwM2rWD//1Pr6EitBEM3TeUQXsH4frWlSVXlxg4WCGEEMIw9Jo0P2HCBIoXL86TJ09o164dlpbqlA5TU1N+/fVXgwYohBBJUZMmcOUKFCgg7eHj7O1b6NABDh9WX0+apHYr1OMP0DfEl05bOrHXdS8AU2pPYVTVUYaMVgghhDAYvdZ0pWaypkuI1OvtW/Dxgff7xIv4cHVVNzh+8EAtEa5aBa1bx7gsIABu3YI3b9SimJUV5MqlbtkVOXvdw9uDpuuacvPlTazNrFnZaiVtv2n7ld+QEEIIEXdxrnTNnTuXfv36YWVlxdy5cz977Y8//pjgwIQQIikJCYGWLeHuXdi9GypWNHZEyUzatOofYp48sGMHvO+Aqyhw/jwsXw5nzsCdO6DVxrzd1hbKlIFS9a+zwaI+r4JekNU2Kzs77qRCjgpf970IIYQQ8RTnSle+fPm4fPkyGTJkIN9nPubVaDQ8fPjQYAEmNVLpEiL10WqhSxdYt07NHc6cgeLFjR1VMnT7NmTMCJkzoyiwZg3MnAnXrsVjDOu30Kcy9mms2dhyF/Ur5UqsaIUQQgiDkemF8SRJlxCpz5gx8Mcf6tZR+/bB+95B4nOCg6F/f6hTB7p1i3bq0SPo0weOHInrYArw0bovew8IyoCZ1pbff4fRo8HCwkBxCyGEEIkgzknX8OHD4zagRsPMmTMTFFRSJkmXEKnLkiVqggCwdCn07GnceJIFT09o1QouXAAbGzXLyqi2cN+8Wf0z9PeP41hmwdC8NzyuBpcH6rykVCl1q6/cuQ0TvhBCCGFocV7TdfXq1WivnZ2dCQ8Pp3DhwgDcv38fU1NTypUrZ9gIhRDCSA4eVIs1AGPHSsIVJ5cvQ4sW8Pw5ODiou0a/T7iWLlUT2DjPr7B5CR1bQq5zUHQr3G0J/tliXObiAtWqwdGjULCgwd6JEEIIYTBxTrqOHTsW9f2sWbOws7NjxYoVODg4APDu3Tt69uxJ9erVDR+lEEJ8ZYoCf/4JERHqeq4E7t+bOqxbB716qVMLixZVy0/vs6BNm76ccKVLpzY4rFABLHPdZJJrM7xCHmGlpOObu5u5GpCN2G5/8kSd9nnuHGSLmZcJIYQQRqXXmq4cOXJw8OBBihUrFu34zZs3qV+/Ps+fPzdYgEmNTC8UIvUIDIQZM+C33+D9doRCF0VRF75NmaK+btIE1q5Vu44A7u5QooTaDl6XnDnVpLZjR7Wb/P4H+2m/qT1+oX4UcCjA7s67KZKxCI8ewT//wJw5ajKsS716cOCA7J0mhBAiaTHR5yZfX19evXoV4/irV6/w8/NLcFBCCGEsH/8ynyYNTJggCdcXaTQfNtEaNUptCf8+4dJq1eJXbAlXz55w86Z6TZo0MP/ifJqsbYJfqB818tTgfJ/zFMlYBIC8edVuhxcvwjff6B7v0CFwcjLw+xNCCCESSK9KV7du3Th16hQzZ86k4vvNai5cuMCIESOoXr06K1asMHigSYVUuoRIucLD1b24KlSAceOkWhIvWi2cOAG1akU7vHx57Gvh1OqWK/7+Hz6sW/FgBXPvzKVZrmaMLjkacxNznfcqij2DBxfgwoWY5+zswM0NMmXS980IIYQQhqVX0hUYGMgvv/zC0qVLCQsLA8DMzIzevXvz559/YmNjY/BAkwpJuoRImRQFBg2CRYvA2lptzuDoaOyokrBjx9S5l1u3qn9gOiiKOq3w1q2Y53r1glGjXClcuFDMk46A65dDuHjxAa1aFeDZs5jn/vgDfv/9y2MIIYQQX0OC9ukKCAjAzc0NgAIFCqToZCuSJF1CpEx//QUjRqjVrS1b1I7nIhYLFsCPP6pzMSdMgPHjdV528iTUrBnzeO7ccOMGPHjgTLk65cg2NBs1cn4Xa1XrU2893rB30h6uXLnCixdladw45jW5cqlrySJnPQohhBDGFOfuhbrY2NhQsmRJQ8UihBBGsWmTmnABzJolCVesQkNh6FC1HAjw/fcwcmSsly9bpvv4//6nLvlyeesCfcFT48lN7U0aFm0U75AaNYKGDWH//ujHnzxRN1+uXz/eQwohhBAGp1cjDSGESCnOnoWuXdXvf/hBzSmEDq9eqRnMokVqOXD6dFi1KtaphQCnT8c8ljEjtGsHa2+sZcC5AWADDubpqVqgmt6hDR6s+/iZM3oPKYQQQhiUJF1CiFTL21vdxzckBJo1g9mzpXmGTrduQcWKaqMMOzt1/62RIz/7h+XtDQ8exDzetJnCtPMT+H7r94RqQ+Eu1M/cADsrO73Da9QIdM1uv3xZ7yGFEEIIg5KkSwiRaqVLp26AXLmyuq+vrP+JRZo04OcHBQrA+fPQtOkXb7l+XcdBsyBuFOrMxBPqTtPdC3SHDcR5LVdsTE2hTJmYx69dS9CwQgghhMFI0iWESNV69FCnwaWCPkD6y5dPXTR14ULsG2R94s0bHQet3/FIOYmZiRlOzZz48ZsfQe9WTtEVLRrz2Nu3hhlbCCGESChJuoQQqYpWC5MmwYsXH46ZyN+E0QUGQufOsHv3h2Ply0OGDHEe4v1uItH5ZWd66Z0c7HKQ3mV7JzzOj+jawDo01KCPEEIIIfQmv2oIIVKVMWPUDuc1asgv5To9eQLVqqnzLXv0UKcV6sHK6v03BfdB0S1Rx7NTjlr5aum+KQF8fD4TgxBCCGFkknQJIVINJyeYOlX9/vffwcLCuPEkOWfPqhWtq1fVNoPbtqmNM/SQPbsC386Fzk2hdRfIoi7ycnExZMAf6Fq/lTNn4jxLCCGEiC9JuoQQqcLBgzBggPr9+PHQvbtx40lyli2DWrXg5UsoWVJt/Ve9ul5DhWvDcfIcAo2GgokWbnSG10UAtQ+Hofn6wu3bMY+XK2f4ZwkhhBD6kKRLCJHiXb8ObdtCRIS6J9f48caOKAnRamH4cOjVS51v2bq1usFVnjx6DecT7EOTtU3413kBKBo4+CfsdIIItay4d2/09XSGsHq1+u/2U5J0CSGESCok6RJCpGjPnkGTJurSpO++U6cYyl5cHzExgeBg9fvx42HTJrC11Wuoh+8eUmVpFQ66HSSNeRpah22Fs78AH/7Aw8Jg4UIDxP1eeDjMm6f7XL16hnuOEEIIkRBmxg5ACCESU2ious1UkSKwdaus49Jpzhxo0wbq1EnQMCtdVnL71W1y2OVgV6ddpA0sw7apoHzSFn7aNOjY8cPrtx66+svH7uPr//oL7tyJeU3lyuosSSGEECIp0CjKpz8Oxef4+vpib2+Pj48PadOmNXY4Qog4ePtWrXTpOWMu5TlwQC35rVsHZob77C1CG8GYo2MYUnEIOdLmAKBxY9i3L+a15cvDkiUPKFXKUe/nbd7sQefOuXV2oVy5Up1KKoQQQiQFRk+6Vq9ezalTp7hy5Qo3btwgNDSUZcuW0aNHD53X+/r6MmHCBLZs2YKXlxfZsmWjXbt2jB8/HlsdU2K0Wi3z589n8eLFPHjwAFtbW+rWrcsff/xB/vz54x2vJF1CJH2KAjduSKUjBkWB2bNhxAh1Ldfs2TBsmN7DaRUtS5yX0K1UNyzNdGyUBZw7B1Wrxqx2AdSuDTNnPkCr9Y33s93dM9KvX26dGyAXLqyu45OqphBCiKTC6ElX3rx58fDwIGPGjNjY2ODh4RFr0hUQEEC1atW4du0a9evXp0yZMly9epWDBw9SoUIFTp48idUnG7P07dsXJycnihUrRpMmTXj+/DkbN27E1taW8+fP4+gYv09ZJekSIun780/49Vd1rc/AgcaOJokICVHbNy5frr7u1QsWLNC9q3AcBIUF0WNHDzbe2kjXkl1Z0XIFmlgWy/30E/z9t+5xHB1h6VJ1a7C4iIiAWbNg7Fj1LX3KxETtA1KpUtzGE0IIIb4GozfScHJy4tGjR7x69YoBkf2cYzFjxgyuXbvGqFGjOHDgANOmTePAgQOMGjWKS5cuMXv27GjXHzt2DCcnJ2rUqIGzszPTp09n1apVbN++nbdv3zJkyJDEfGtCCCPYtAlGjlQLObL58XteXmo7+OXL1axkzhx1eqGeCZeXvxffrfiOjbc2Ym5iTq28tWJNuAD++ENNrnRxdVU3qm7fHk6c0F0RAwgIUEMuU0b996sr4QL4+WdJuIQQQiQ9Rq90fWzatGn89ttvOitdiqKQM2dOfH198fLywsbGJupcQEAAWbNmJXPmzLi5uUUd79y5M+vWrePEiRPUqFEj2ni1atXi+PHjeHh4kDt37jjHKJUuIZKus2fVKWshIfDjj2pukepdvQrNm8PTp5AuHWzcmKC2ftdfXKfp2qY88X1Ceuv0bG2/lZp5a37xvnv31G2/Xr36/HUZMqit3osUUacHenurGx9fv/7lJLppU7VZirl5nN+OEEII8VUYvdIVV66urjx//pyqVatGS7gAbGxsqFq1Kg8fPuTJkydRx48fPx517lMNGjQA4MSJE4kbuBDiq3jwQM0tQkLUf86aZeyIkggzM3j3Ts1iLl5MUMK15/4eqi6tyhPfJxTKUIjzvc/HKeECdZ3VkSOQJcvnr3vzRt3Ieu5ctTOhk5O6T3NcEq6NGyXhEkIIkTQlq6QLiHUNVuTxyOsCAgLw9PQkX758mJqafvH62ISEhODr6xvtSwiRtLx+DY0aqb+wly8Pa9eCjv/tU6cSJWD/fjh/PvY5fnHgF+JH9+3d8Q/1p3a+2pzvfR7HDPEbr0QJtRppyOl/Go26t/PWrWBtbbhxhRBCCENKNkmXj48PAPb29jrPR071i7wuvtfHZurUqdjb20d95cqVK/7BCyES1erVaqUrTx7YtQs+KYanLv7+0KGD2k0iUrVqEMvfhXFlZ2nHxnYbGVBuAPu/34+DtYNe4+TPD6dPq81O9FxSFsXREU6dgpkzpcIlhBAiaUs2SZex/Pbbb/j4+ER9fTx9UQiRNAwdqk4n3LsXsmY1djRG5O4OVaqo8+w6dYq920QceQd7c/HZxajXtfPVZmHThZibJizDMTWFX35Rm2j89htkyhS/+8uVgyVL1G0BdMweF0IIIZIcw+2KmcgiK1axVaYip/1FXhff62NjaWmJZUI/jhVCJAqtVm3Gp9GobclTtRMnoE0bdY5l1qxq4pWAv7vc3rrRdF1TPP08Od/nPEUyFjFgsKpcuWDKFBg/Xk2Yz5yBK1fAxUVdhgbqv99cudREq1w5aNBA/acQQgiRnCSbpOtLa7A+XfNlY2NDtmzZcHd3JyIiIsa6ri+tERNCJG2LF6vreDZuhFTfSPTff2HIEAgPVzOS7dshZ069hzv9+DQt17fkTdAbcqbNSUh4wipmX2JpCa1aqV+RtFp1Ty4zMzWpFkIIIZKzZDO90NHRkezZs3PmzBkCAgKinQsICODMmTPky5cv2pqrmjVrRp371IEDBwBitJIXQiR9+/fDoEFw4IC6nivVCg+HwYPVTY/Dw9UphadOJSjhWumykjor6/Am6A3ls5fnYp+LlMpayoBBx42JibpOSxIuIYQQKUGySbo0Gg19+vTB39+fyZMnRzs3efJk/P396du3b7Tj/fr1A2Ds2LGEftRveN++fRw/fpz69euTJ0+exA9eCGEwLi7Qrp1aBenWDQYONHZERmRqqm58pdGo8/TWrNG7hZ9W0TLm6Bi6b+9OaEQobYq24USPE2Szy2bgoIUQQojUx+ibIzs5OXH69GkAbty4gbOzM1WrVqVgwYIAVKtWjT59+gBqRatq1aq4uLhQv359ypYti7OzMwcPHqRChQqcOHEC609+4ejbty9OTk4UK1aMJk2a4OnpyYYNG7C1teXcuXMUKlQoXvHK5shCGM/Tp2q78WfPoFYtteJlYWHsqIwsIADOnYO6dRM0zKLLixi4R81gf6/2O5NrT8ZEk2w+lxNCCCGSNKMnXT169GDFihWxnu/evTvLly+Peu3j48OECRPYsmULXl5eZMuWjXbt2jF+/Hjs7Oxi3K/Vapk3bx6LFy/mwYMH2NraUrduXf744w8KFCgQ73gl6RLCOHx9oXp1uH4dvvlGbbqQLp2xozKCXbtg2za1fZ8B596FRoTSdG1TupTsQrdS3Qw2rhBCCCGSQNKV3EjSJYRxtGsHmzdDlizqPr958xo7oq9MUWDaNBg9Wv1+2TLo0SNBQ7q+cSW/Q35MTUzfP0JBI4uohBBCCIOTuSNCiGTh99/VzXB3706FCVdQEHz/vfqHoChq84zvv0/QkLvu7aLMv2UYeWhk1DFJuIQQQojEkWxaxgshUrcyZeD2bbWFeKry9Cm0bKluYGVmBvPmQf/+eg+nKAqzz8/ml4O/oKDg8sKF0IhQLExT++I4IYQQIvFIpUsIkWRt3ar2iIiU6hKuCxegQgU14cqQAQ4fTlDCFRYRxoDdA/j54M8oKPQv15993++ThEsIIYRIZKntVxghRDJx+jR07qz2ijh7Vq10pTrh4fDmDRQvDjt3Qr58eg/1LugdbTe15aj7UTRomFl/JsMqDZMphUIIIcRXIEmXECLJcXWFFi0gJESdWVeypLEjMpKqVWHvXvj2W9DRnTWutIqWOivrcNXrKjbmNqxrs45mhZsZMFAhhBBCfI5MLxRCJCmvXkGjRvD2rTqzbs0adQ/gVMHHB9q3hxs3PhyrWzdBCReAicaEMTXGkMc+D2d6nZGESwghhPjKpGV8PEnLeCEST1AQ1KmjruPKm1dtDZ8li7Gj+kpcXaFZM7h3T92I7Pr1BGebb4Pekt46fdTroLAgrM2tP3OHEEIIIRKDVLqEEEmCVgvdu6sJV7p06qy6VJNwHToEFSuqCVfOnLBqVYISLq2i5bfDv1F8QXGe+DyJOi4JlxBCCGEcknQJIZKEsDCIiABzc9i2DYoWNXZEX4GiwJw50LAheHtD5cpw6RKULav3kIFhgbTb1I5pZ6bh6e/JHtc9hotXCCGEEHqR6YXxJNMLhUg8ERHg7Kyu5UrxQkNh0CBYskR93aMHLFoElpZ6D/nc7znN1zXniucVLEwtWNJ8CV1KdjFMvEIIIYTQm1S6hBBGde+eWvABdUZdqki4AExMwMND/eesWbB0aYISrqueV6n4X0WueF4hY5qMHOl2RBIuIYQQIomQlvFCCKO5ehWqV1fbwi9ZkqCcI/kxM4ONG9XSXp06CRrq7JOz1FtVj8CwQIpmLMruzrvJ75DfQIEKIYQQIqEk6RJCGMWTJ9C0KQQEgJeXuglyird5s9opZOZM9bWDQ4ITLoCSWUrimN6RzDaZ2dhuI+ms0iV4TCGEEEIYjqzpiidZ0yVEwvn6QrVq6nZUxYrB6dNqx8IUS6uFiRNh0iT19bZtankvAcIiwjAzMUPzPlt9FfAKB2sHzEzkszQhhBAiqZE1XUKIryosDNq1UxOurFlhz54UnnD5+6tvODLhGj5cLfElwNugtzRY3YDpZ6ZHHctkk0kSLiGEECKJkp/QQoivRlHUhn0HD0KaNLB7N+TJY+yoEtGjR9CihbrRsYWF2p2wZ88EDen6xpWm65py/819rnheoXeZ3mSyyWSYeIUQQgiRKCTpEkJ8NbduwYoVasO+9euhXDljR5SITp2C1q3h9Wt1l+etW6FKlQQNefzRcVpvaM274Hfkts/N7k67JeESQgghkgFJuoRIDYKC1F/+Q0LUikvGjGqp6SsrXhwOHIAHD6BZs6/++K/L1xfevIEyZWDHDsiVK0HDLb26lP67+xOuDefbHN+yo+MOsthmMVCwQgghhEhM0kgjnqSRhkgWAgLUTnlHj8KVK3DnjtrMIZJGA0WKqKWm776D9u3Bzi7RwlGUVNKd8FM7d0LduglOcEcfGc2U01MA6FCsA8taLMPa3NoQEQohhBDiK5BGGkKkJI8ewbBhkCMH9OgBK1eqc/o+TrhAzYLu3IHVq6FPH/X6wYPVEpSB3b+vbnh8+7bBh05a3r6FDh3A3f3DsebNDVJRzJNOXfg2rsY41rZZKwmXEEIIkcxIpSuepNIlkqSICJg7F37/HYKD9R/HwgImTIARI9TNexPo1SuoXBnc3KBBA9i/P8FDJk23b6sJlpsbVKoEZ88avLTn7OlM2WxlDTqmEEIIIb4OqXQJkdw9ewY1a6qtyBOScAGEhqqJW+XKatUsAYKC1MZ9bm6QL59adEuRdu9WEy03N7UV47//JjjhcvZ05rvl3/Em8E3UMUm4hBBCiORLki4hkjM3N3WX4TNnDDvu5ctQtao6BVEPWi106wbnzoGDA+zdC5kzGzZEo1MUmD5drXD5+UGNGnDpEpQsmaBht93ZRvVl1TnhcYJfD/9qoGCFEEIIYUzSvVCI5OrZM6hTBzw8Pn+doyNUrKi2DrS1hcBAuHlTTRDu3o39vufP1fHPnFFLVfHw669qHw8LC9i+Xe3ZkaIEB0PfvuqaOID+/dXpnRYWeg+pKAozzszg1yNqotWgQAP+qv+XIaIVQgghhJFJpUuI5CgiAjp2/HzC9f33cOGC2sli9Wr2lixJzmnT2FKggDrX784dtbNhr17qxlm6eHpC27YQFhbn0Favhj//VL9fulQtAKU4igL37oGpKSxYoG56nICEKzQilN47e0clXEMqDGF3593YW9kbKmIhhBBCGJE00ognaaQhkoS//4afftJ9Ln9+NdupWTPq0N69e2nVqhV29g54v33Nhg0baNOmzYd7zp2Dnj3VREKXyZNhzJg4hfbunboncJ06cb4leXr2DFxd1Zb7CfA26C2tN7TmhMcJTDQmzGk4hyEVh3zxvnfv3vHy5UtMTU3Jli0bNjY2CYpDCCGEEIlHKl1CJDceHmqzC10qVFCnDepIuCrXqMu249eo27gVHTp0YMuWLR/uq1wZLl6E6tV1jztp0uenIn7EwQEOHoTRo+P6hpKJ1avhf//78DpHjgQnXABaRcsT3yfYWdixp/OezyZcERER7Ny5k4YN6pM+fXqKFCmCo6MjGdKnp1u3rly4cAH5HE0IIYRIeqTSFU9S6RJG99NPaqXrU/nzqwlX+vRRhz5OuKb8sxxzCwvCw8OZ+MsADu/dFrPi5eurJmC6NtXq0wf++09nSE+ewIED6iUpTkSEmuTOmKG+Pn48WlJrCHdf3yVcG07xzMVjvebly5c0b9qUC5cuUSFXFvqUKYhjhrREKAqXn71msbMr7m986NqlC05LlmCRgOmOQgghhDAsSbriSZIuYVQBAWqFxccn5rlPkgFdCVekzyZely6pLdA/3VDZ2lqdUufgEO2wj4/aQPHmTZg5U+1cn2L4+kLnzrBnj/r699/VqZaxrYGLIydnJ0w0JvQq0ytO1797945qVarw9vkT1rWtQfU8WWNcE6HVstrFjYG7z9GseQs2btqESQLjFEIIIYRhyE9kIZKTzZt1J1zffx/nhAvAzMyM8X8t0j3VsEIF6Ncv5jOCgmDNmmiHwsKgXTs14cqaVe25kWI8eKAmn3v2gJUVrF0Lf/yRoIQrQhvBiIMj6LurL/139+fGixtxuu+nn4bh+diDw93q60y4AExNTOhexpEN7b5j2/Zt/Pvvv3rHKYQQQgjDkqRLiOTk2DHdx3/8MerbLyVckT6beA2JZV3R8eNR3yoKDBwIhw6BjY2am+TOHd83lEQdOaK22b9zR60snjoFnTolaEj/UH9ab2zNX+fUNvBja4z97HTCSC9fvmTd2nX8Vq04RTKl++L1zYrkpnXRvPwz529Z3yWEEEIkEZJ0CZGcXLkS81jBgmp1irgnXJFiTbyKFYMyZT77/KlTYckStfCzfj2ULav3u0p6nj9X2zB++6063bJ8+QQN99T3KdWXVWfnvZ1Ymlqyrs06xtUch0aj+eK9y5YtwwSF7mUKxvl5AyoU5s69+5w4cSIhYQshhBDCQCTpEiK5CA7W3eCiYkXQaOKdcEWKNfF6n8hF8+gRvHvH2rUfuhP+8w80barfW0qyunaFDRvUyl62bAka6vLzy1T8ryLXvK6R2SYzx3scp2PxjnG+//z581TPk4UMaazifE/NvFlJl8aK8+fP6xOyEEIIIQxMki4hkotXr2I2twAoUULvhCuSzsSrRAndF3t58fKl+u3PP8OgQfF8H0nRq1dqw4wXLz4ca99eXcuVQIfcDuHp70nxzMW52OcilXJWitf9/n6+2Fuax+sejUZDWitL/Pz84nWfEEIIIRKHmbEDEELEUWio7uO2tvTr1w87ewcmz1kS74QrkpmZGeP/XMiNqxfp06cPrWfOROfkt9BQhg1TC2GVK+v1qKTFxQVatFD3P/P2hr17DTr8r9V+xdrcml5lepHWMv4dT+3S2vPWPZZ/97HQahW8g4Klw6oQQgiRREilS4jkIrZkKjCQOXPm4P32NX/8+gPh4eF6Da/Vapk2bjiez54wd+5cNEFB0c6/IiM+pI2Ko2rVBHdON76tW9U34uGhro2bOTPBQ4aEhzDx+ET8Q/0Bteo0rNIwvRIugBo1anDK4wVefoFxvueg2zN8g0KoUaOGXs8UQgghhGEl91+ZhEg9MmYEXY0Xbt6kTZs2bNiwgcN7tzHxlwHxTry0Wi1TRg9l16bVrFixgq5du6p94N8Lwopm7KI6p3gSprtlebKiKDBpErRpo+59VrcuXLwIRYsmaNjXga+pt6oeE05MoMf2HgYJtXv37piZm7PE+X6c71l0+R6lS5agUqX4TWUUQgghROKQpEuI5MLaGooUiXn84kUAvRMvnQkXwOXL6nk0dGE1F6jEU5PcBFo6fGa0ZCAgADp0gPHj1dc//gj79sXY9Dm+7r6+SyWnSpx6fIq0lmnpW7avAYIFBwcHunXvzowzt3B+/vqL1692ecDuu48Z+tPwOHVHFEIIIUTik6RLiOREV+vye/fA2RmIf+IVa8J1715U0jWSGWylDRaEsL3qXxQubNB39PWFh8P162BuDv/9B3PmgFnClrceeXiEyksq4/bOjbzp8nK211kaFGxgoIBh5syZfFO8BPVWHmLnXQ+02pj7bwWHhTP77E16bT9Nz5496d69u8GeL4QQQoiEkUYaQiQnNWvCqlUxj8+bB0uXAh8Srw4dOgAw/q9FmOlIKmJNuAAWLABgPoOYyS8ALKMnNdp8a+A3ZAT29rBrl9qpsFq1BA+3+MpiBu0ZRIQSQZVcVdjeYTuZbDIZINAPbGxsOHTkCO3btaX1usMUzJiOXqULUCijPeFaLZefvWa5y0Ne+wcybNgw/vrrL6lyCSGEEEmIRlGUmB+Zilj5+vpib2+Pj4+PdAYTX5+fH+TIof7zYyYmcPp0tHaCW7ZsoUOHDtRt3CpG4vXZhMvFBcqXZ3d4A1qwAy2m/MHv/G45C549gwwZEvtdGt7SpRAUBIMHG3RY72BviswrwouAF3Qp2YX/mv2HlVnC28zHRlEUzp49y8IFC9i0eROhoWEApLNPS89evRkwYACFChVKtOcLIYQQQj+SdMWTJF3C6IYMgfnzYx4vXFhd3/XRf5e6Eq/PJlyBgVClCuEuNynOTe5RhN448R990XTvDsuXJ/77M6TwcBgxAv7+W01ML1+GMmUM+oiLzy5y+OFhfqv221etLoWHh+Pt7Y2pqSn29vaYJPtWkkIIIUTKJUlXPEnSJYzuwQMoVkz3vl3Vq8Pu3bEnXn8uZNq44bEnXK1awcGDADwlBzMYyUx+xtxUgatXY98wOSl69w46dox6P0yYAGPHJrjP/ROfJ9x9fZd6BeolPEYhhBBCpAqSdMWTJF0iSZg2DX77Tfe5okXVilTFilGHIhOvLNly4PnsScyE69o16NEDxcVF94bIo0fD//5nwDeQyO7ehebNwdUV0qSBlSvV9vAJdPHZRVqsb4FviC+nep6ibLayBghWCCGEECmdzEcRIjn65ReoUEH3uTt31LVdAwbArVvAh+Yagf6+MbsU/vgjVKhAqMttmrOTjbSLPl7x4mqFKLnYtw8qVVITrty54cwZgyRcm25toubymnj5e1HAoQAZ02Q0QLBCCCGESA2k0hVPUukSScajR1C1Kjx//vnrypRRE7QSJVDSpEETFKRufHz5clRbeAXozRKW0Qs7fHlIfjLyRt2Q+fRpklWf+NmzYfhwtTPhli2QOXOChlMUhSmnpjDm2BgAmjg2YV2bddhZ2hkiWiGEEEKkApJ0xZMkXSJJuXMH6tb9cuL1Bf9jNGP5HyZEsItmNGYfZMqkrocqXdowsX4tiqJOJ+zUCSwsEjRUSHgIfXf1ZdV1tU3/T5V+4s96f2JqYmqISIUQQgiRSsj0QiGSs6JF1elzZfVfW7Sa7xmLul5rPoPVhKtYMTh1KnkkXJ6e0K0b+PiorzUa6N49wQkXwL9X/mXV9VWYakxZ2GQhsxrMkoRLCCGEEPEmSZcQyV3evHD+PEyeDObm8br1BDXohbqp8ghmMMDUSW2aceVK8phSePmyOnVy1SoYNMjgww+uMJjOJTqzv8t+BpQfYPDxhRBCCJE6yPTCeJLphSJJu3sXZs6ENWvUzYA/4zG5KIUL3jjQ1mQLG7rsxuSX4cmnLfy6ddCrFwQHqxW/nTuhYMEED3v+6XnKZiuLhWnCK2VCCCGEECCVLiFSliJF4L//1DVe8+apXfvy5tV5ac5cJvR1PE7lvM9Z6fEdJiuWJY+ES6tVq3GdO6sJV5MmaqXPAAnXwksLqba0GgN3D0Q+jxJCCCGEoZgZOwAhRCJIlw4GD1a/QN0o2MtL3VDZwgIyZ8YkQwZmoOYtVlbGDDYe/Pzg++9h1y719ciRMGUKmCZsnVWENoKfD/7MnAtzAAjThhGhRGCmkb8ihRBCCJFw8huFEKmBgwM4OBARAfPnQ//+YPn+VLJJuAACAuDqVbC0BCcn6NIlwUP6hvjSaUsn9rruBeCP2n/wW7Xf0Gh0bhMthBBCCBFvknQJkYqMRvWToQAAW21JREFUHAmzZsHu3XDggNroL1nJmhV27IDwcKhYMcHDeXh70HRdU26+vImVmRWrWq2i7TdtDRBo3Li6uuLn5xfv++zs7HB0dEyEiIQQQgiRGCTpEiKVmDdPTbhA7T+RbBKuhQvB3l5dwwUJao//sXBtOPVX1+f+m/tktc3Kzo47qZCjgkHGjgtXV1cKFSqk9/3379+XxEsIIYRIJiTpEiIV2LULhg5Vv58yBTp2NG48cRIWBj/+CIsWqdMJK1QAAyYZZiZm/NPoH34/8jvbOmwjl30ug40dF5EVrgL9KmKdPe6dUIOe++K2+KJeFTIhhBBCGIckXUKkcFeuqEmWVgt9+sCvvxo7ojh4/RratoUTJ9SS3MSJBulOqCgKj7wfkc8hHwD1C9Snbv66mGiM18jVOntabPI6GO35QgghhEh80jJeiBTMwwOaNoXAQKhfHxYsSAbTCm/cUKtaJ06AnZ26/9aoUQkOPDg8mK7bulL+v/K4vnGNOm7MhEsIIYQQqYP8tiFECvb8OYSEqNtvbdoE5ubGjugLtm+HypXh0SPInx/OnVOzxgR6GfCSOivrsObGGnxDfLnieSXBYwohhBBCxJVMLxQiBatcGc6eBRsbSBv3ZUPGc+mS2ha+dm3YuBEyZEjwkLde3qLpuqY88n5EOqt0bG63mTr56xggWCGEEEKIuJGkS4gURlHg6VPI9b4vRJEixo0nXiZPhrx5oUcPg5TlDjw4QPvN7fEN8aWAQwF2d95NkYzJ6Q9ECCGEECmBTC8UIoX53/+geHE4csTYkcTBkyfQty8EB6uvTUzU1wZIuA65HaLJ2ib4hvhSI08NLvS5IAmXEEIIIYxCKl1CpCCrV8O4cer3rq5QJynPojt3Dlq1ghcv1Jbw8+YZdPjqearzbc5vKZShEP82/RcLUwuDji+EEEIIEVeSdAmRQhw/rm56DDByJAwYYNRwPm/FCujXD0JDoWRJ+OUXgwzrH+pPGvM0mGhMsDKz4kCXA9iY26BJ8i0bhRBCCJGSyfRCIVKAO3fUolFYGLRrB1OnGjuiWEREqAlWjx5qwtWqFZw5o67jSqBH3o+o5FSJ34/8HnXM1sJWEi4hhBBCGJ0kXUIkcy9eQOPG4O0NVaqoRSSTpPh/tre32v595kz19bhxsHkz2NomeOhzT85R8b+K3Hp1i5UuK3kb9DbBYwohhBBCGIpMLxQimZs2Td3WqmBB2LEDrK2NHVEs3r6FixfVAFesUEtyBrD2xlp67ehFSEQIpbOWZlenXaS3Tm+QsYUQQgghDEGSLiGSuenT1Vl7P/wAGTMaO5rPyJ8ftm0DOzsoUybBwymKwsQTE5l4YiIAzQs3Z03rNdhaJLxy9jUFPfdN1OuFEEIIYXySdAmRzFlYwNy5xo5CB0WBv/+GwoXV+Y8ANWoYbPjeO3uz7NoyAEZUGcHUOlMxNTE12PiJzc7ODgC3xRcTdL8QQgghkj5JuoRIhv75B9zc1OVRpkkxzwgJUdsnLl8OadOqnT6yZzfoI+oXqM/q66tZ2GQhvcv2NujYX4OjoyP379/Hz88v3vfa2dnh6OiYCFEJIYQQIjFoFEVRjB1EcuLr64u9vT0+Pj6kTZvW2OGIVGjHDrXpn6LApk3Qtq2xI/qElxe0bq3uw2ViArNmwY8/ggG6CEZoI6JVsx77PCa3fe4EjyuEEEIIkZiSYo8zIUQsLl2CTp3UhKtfP2jTxtgRfcLZGSpUUBOudOlg3z4YOtQgCdc+132UWFiC537Po45JwiWEEEKI5ECSLiGSiUePoFkzCAqChg1h/nyD5DKGs3EjVKsGT5+q67guXID69Q0y9D8X/qHpuqbceX2HqaeS6iZkQgghhBC6yZouIZIBb2+1F8WLF1CqlJrfmCW1/3uPHlUzwkaNYN06sLdP8JDh2nCG7R/G/EvzAehVuhczG8xM8LhCCCGEEF9TUvu1TQjxCUVRt7S6cwdy5IDdu9Wu60nO3LlQsiT072+Q7h4+wT502NyBA24H0KBhet3p/FLlFzRJqrwnhBBCCPFlMr1QiCROo4FBgyBzZtizB3LmNHZE7z16pDbICA9XX1tYqIEaIOF64vOEKkurcMDtAGnM07C1w1ZGVB0hCZcQQgghkiVJuoTQk7+/P926dWPdunWJ/qxWrcDdXZ1amCScOKE2zPjnH5g0yeDD21vZo0FDdrvsnOp5ipZFWhr8GUIIIYQQX4u0jI8naRkvQE24Gjduwpkzp1EUheXLl9OtWzeDPmP7dihTBvLkMeiwCbd4MQwerFa4ypVTA02E8puHtwdmJmbkSJvD4GMLIYQQQnxNUukSIp4iE64rzs5MWrqN2q0606NHD1auXGmwZxw9qq7jqlQJHj822LAJExYGQ4aoa7bCw6FjRzh50iAJl1bRMu7YOGadmxV1LE+6PJJwCSGEECJFkEYaQsTDxwnX7wvWUrhUeRxLlgOgR48eAAmueN2+re4tHB4ONWsmkTVcb95A+/ZqNgjwxx/w228G6VkfFBZEjx092HhrIxo0NCrYiKKZiiZ4XCGEEEKIpEKSLiHiSFfCBWBiYkK/MdOBhCdeXl5qa3gfH6haFZYvB5OkUI9+/lzdd8vWFlavhhYtDDKsl78XLda34OKzi5ibmPNv038l4RJCCCFEiiNJlxBxEFvCFckQiVdAgLr5sYcHODqqS6WsrAwRvQGUKKFuDpYrl/q9AVx/cZ1m65rx2Ocx6a3Ts7X9VmrmrWmQsYUQQgghkhJJuoT4gi8lXJESknhFREDnznD5MmTIAHv3QsaMBglfP4oCf/4J1apBlSrqscaNDTb8nvt76LilI/6h/hTKUIjdnXbjmMHRYOMLIYQQQiQlknQJ8RlxTbgi6Zt4+fjAkydgaQk7d0LBggkOXX9BQdCnD6xdC1myqLsyOzgY9BGPvB/hH+pPrby12Nx+M+mt0xt0fCGEEEKIpESSLiFiEd+EK5I+iVf69GojwCtXPhSWjOLZM2jZUi25mZnB+PEGT7gABlccTIY0GWhTtA3mpuYGH18IIYQQIilJCkv0hUhy9E24IkUmXl9qJ+/p+eF7W1u1W6HRXLigbngcOcfx0CEYONAgQ3sHezNg9wDeBb2LOtaxeEdJuIQQQgiRKkjSJYQOP//8M6dOnWTknBXxTrgiRSZeFWo1pEePHty+fTva+YsX1WmEU6eqS6iMavVqNePz9ITixdXgvvvOIEM/fPeQKkuq8O+Vf+mxo4dBxhRCCCGESE4k6RJCh3bt2mFpacnuVYsICw3Re5wHN69y88IpqlSpSr58+aKOu7urnQoDA+HUKdBqDRG1nhQFduyAkBBo3hzOnoX8+Q0y9OnHp6n4X0XuvL5DDrscTKg5wSDjCiGEEEIkJ5J0CaFD3bp12b59OzfOn2TWiH56JV73r1/hj4GdKFO6NPv378Pa2hqAd+/URoAvX0Lp0rBhA5iaGvgNxIdGo24INmcObNsGdnYGGXaVyyrq/L+9+46OqnjYOP7dTdn0hA6hBZBepIMiTRSwU0R6k6aAyI8qFkAsqJTXrihKERBFilKkSZUmHRQ0ofdOEhJSNrv3/WMlsGQTkpiQkDyfcziwc+fOnRtGzJOZO3dmMy7HXKZ2cG3+6PMHNYrUyJC2RURERO4lCl0iyWjZsmW6g9ftgcvPzw+A+Hho0wb+/huKFYMlSzIs46RNWBiMGnVzXaOvLwwalCFvYrYbdl5f8zrdFnUj3hZP24ptWd9jPcH+wf+5bREREZF7kUKXSArSE7ySC1yG4diJfd06R9BauhSKFs3kG3Bl9WqoVw/ee88xu5XBwmPDmbVvFgCjHhrFj+1+xMfDJ8OvIyIiInKvMBlGlj/Cf0+JjIwkMDCQiIgIAgICsro7cpcsX76cVq1aUbV+I4ZM+AoPT4vLeskFLnBknUcfdSwlXLYMmje/W73/l2HAJ5/AkCGOtzHXr+9YTli4cIZf6s8Lf7Lr7C663Z+6l0OLiIiI5GQKXWmk0JV73Sl4pRS4bvjqK0fo6tXrbvX6X3FxMGAAfPON43O3bjBlCnh5ZUjze8/tJfRyKO0qt8uQ9kRERERyEoWuNFLoyt2SC16pCVxZ5sIFx4NkmzY5ntmaMAH+9z/HBhoZYPE/i+k4vyPxtnjW91jPA8UfyJB2RURERHIKPdMlkgaunvFKKXD99Rc8+SRcupSFnQ4Lc7x3KyDAsXPHkCEZErgMw2Dylsk8M/cZoq3RNCrZiAr5K2RAh0VERERyFs10pZFmugRuznjdV7UWRw/ucxm4zp51PDZ14oRjNd+MGVnY4R9+gPvvhwoZE4qsNisDlw3kq11fAdCvVj8+eewTPNw8MqR9ERERkZxEoSuNFLrkhhvBq3btOkkCV3Q0NG4MO3dCuXKO9w3ny3eXOma3w7vvQqtWUKVKhjd/NeYq7ea147ejv2HCxOQWk3m53suYMmi5ooiIiEhO457VHRC5V7Vs2ZKwsDAKFiyIxXJzUw2bDTp2dASu/PkdOxXetcAVFQXdu8OCBfDtt/Dnn+CTsdu1z9w7k9+O/oavhy9zn53Lk+WezND2RURERHIahS6R/6B48eJOnw0DBg+GxYvBYoFffoEyZe5SZ44fh6efhn37wNMT3ngjwwMXwKB6gzgWfowe1Xtwf+H7M7x9ERERkZxGywvTSMsLJSWffQYDBzr+PG8ePPvsf2vv779hyxbHrNn+/RAe7phJ8/GBkBCoVQtq14aHjI1YOrWFixehYEHH+7cefPC/3k6iRX8vonmZ5nrJsYiIiEg6aKZLJAM9+qhjZuuFF9IfuK5fh7lz4fPPHWErOdu3O4JdL6bSkP6AldhKNfD6dRGUKJG+i9/Gbth57bfXeG/Tezxb6Vl+ePYHzCZteioiIiKSFgpdIhmoXDnYvRvS85ouw4Aff4RBgxyv1koNE3Y68j2eWPmRdvQ8MI32Y32ZPBmCgtLeh1tdt16n68KuLDi4AIAK+bQdvIiIiEh6aHlhGml5odzuyBE4fNgxy5VeFy86ZscWLEj7uXm5TGdm8wkvAY4dBIODYepUeOyx9PXnzLUzPP390+w8uxNPN0+mPjWVrvd3TV9jIiIiIrmc1gmJ/AdXrsDjjzvCTXoCEzhCW/36qT+/Agd5lXdu9oF8fMIgbgQugDNnHP365JO092fPuT3Um1qPnWd3kt8nP791+02BS0REROQ/0PJCyXHsdggLc2xCER3tWLbn5wflyzuW/5kz6EcNcXHQpg388w8UK+YITml14gQ0aQInT965btGi0CXPUt442BFf2zXOmosxzd49xXMGDXL8/tJLqeuP1WalzQ9tOBV5igr5K7C001JK5ymdupNFRERExCUtL0wjLS/Mnq5dg1mzHBtL7Njh+OyKn59jx7927aBrV0jvX6FhOM6fPRv8/WHTJqhaNW1txMRAnTrw11/J16leHfr3h6efMig0cwK88orj4o0aETvrJ7YfK8DUqfDDD44QmJwlS+CJJ1LXr80nNzP+9/F81/o7gryC0nJLIiIiIuKCQlcaKXRlLxcuwFtvwfTpjvcCp4WvL3TrBqNHQ+HCaTt39GjHdd3dHS8/Ts/zXMOGwaRJro/lyweffgrt24MpLhb69HGkSoC+fR3rBj09E+ufOOGosnKl6/YKF3aEu7x5kx6Lt8Xz14W/qFGkRtpvQkRERETuSM90yT3JMByzO5UqOcJJWgMXOJYefvGFo43Zsx1tpsa0aY7ABfDll+kLXFu3wuTJro/Vru0ISB06gOnsGWjc2BG43NwcN/vll06BCxw7xC9fDh984LrNc+ccL22+3ZWYK7Sc1ZKG0xqy99zetN+IiIiIiNyRQpfcc+LiHDNUHTrA5cv/vb2rV6FLF+jYEWJj71x/61bH76+9Br16pe+ar77qOuRVrQqrVkGhQv8W7NmDsX071oAAPnjkER6aNYu6tWvxWMuWfPXVV0RHRyeeazLB8OEwYYLra373nfNSxrDLYTzwzQOsPbYWk8nE+ejz6bsZEREREUmRlhemkZYXZq3YWGjd2jGrkxmaNYPFi8HbO/k6hgELFzr6YTIlXy85Bw5A5cpJyz09He/4qlTpZtm2bdtY/uyzzDx1Cru/Dw8G+eJpNnEiJp61F8Px9/Xjf0OHMnr0aMz/7hBiGI7nt379Nek1+veHzz6D9cfW0+bHNlyJuUKJwBIs6biEqoXS+FCaiIiIiKSKQlcaKXRlHbsdnn3WEXhSUq8ePPKIY8OMoCDHusOICD927oTffoMtW1I+/8knYdEix2q+G65cgcBA57L0GjTI9Vbu48bBG6/a4J13oFs3VoWF8fRTT1HFz8JrZYNpVjAI8y0p78T1OKYcPcvHh8/RsUN7vps1OzF4nTrl2KkxJsb5Gn5+8MGqaby8qh9Wu5V6Revxc4efKeRXCBERERHJHApdaaTQlXU++sj1c0k3tGrl2OCixr/7QRw7dozGTZpis9lYv24tZcqUAWDvXnj7bfjpp+Tb+uADx1I9cDwv1rixY8v27793bMDxX5QqBceOOZd5e8Ppg5HkGdgZliwhrkIFipw4QZ0AL+bULouXW/IrgReeuUy3HWG89tprvHXjYTMcG2tMnXpb5fI/Q8dWALSv3J5pz0zD2yOFaT0RERER+c8UutJIoStrhIXB/fcnnbkBx7bvX30Fzz13c7nfjcB13WrHZDbjiXPwAkfo6tvX8UzX7SwW2LMH7rvPEeaWLoUCBWDbNkdoSq9Llxzt3G5oq8NMDH3asfbQYmF6w4aM3Pw7+5pWw8/9ztNrYw4c58tTVzlz7lziuPzjD8esnxNzAmVGP0HnRvUZ02QMZpMe6xQRERHJbPqOS+4JL73kOnDlyQPr1v27tbqLwNVz4iyenzibeNxo3KQphw8fTjz32Wdh/XrH9uy3i4uDF16Al192BC4vL/jll/8WuAB27kxa1pQ1vL2qriNwBQcTvXw5L23eTM9i+VIVuAD6lSpMbFws3333XWJZzZrg4wP4nQOz1VFod6f8jqW82fRNBS4RERGRu0TfdUm29+efsGKF62M//XRzOSEkDVxBBYMJyF8o2eBVtarjGTFXG2KsXw+ff+44NmsW1K//3+/lyJFbPxn05zNW0hyv6CtQty5s387yy5eJun6d7iVT/5xVsLeF5gWD+OnHHxPL3N2hzEO7oG8teGwQ4JjUPnbE/b/fiIiIiIikmkKXZHtffum6fMAAePjhm59dBa4bUgpeDRum/KzYxInQtu1/vIl/Xb9+888eWOnJNNyxEd26iyPlBQdz8eJFzCYTxb09k2/IhRAfCxfOnUv8vOjvRRyo1xACzkDJDWCJTNIHEREREcl8Cl2SrSUkOGaZbuftffMFxZBy4LohpeD15puOnf1u5+7u2GY9o9y6+6EVT1qxiEF8xMVJMx1rGAEPDw/shoEtjU9bWu0Gnp6eGIbBB5s+oM0PbbC5XYdDzeGbzRAXmKQPIiIiIpL5FLokWztwACIikpZ36uR4ngtSF7huSC54+ftD165J6yckOJY3ZpSQyH28xMeJn09TjE8YxJGjN9c33nfffQBsu3ot1e0ahsG2iOuUKn8fvX7pxcjVIzEwyBPWH+YsTQxc4Nh4RERERETuHoUuydZ27HBd3qaN4/e0BK4bkgteN9q8navNL9Jl4UKeeu9BPuZlnuIXp0O33mfDhg0pW7o0U49dSHXT265Gsf/qNc40Osu0PdMwm8x80PRjwud8BnbnZ7iqVPlPdyEiIiIiaXRPhq6QkBBMJpPLX02aNElSPy4ujnHjxlG2bFm8vLwIDg6mb9++XLiQ+m9qJWvs2+e6vFat9AWuG24PXgcOHObTT13X3bs3HR2/lWE41kK2aYNbTDSrTY/wOw85VVm69OafzWYzLwwYwKKzl9kbHn3H5hPsBu/8c4oyISG81uI1gryCWNxxMSHnX8LVCyFq1/6P9yMiIiIiaXLPbmMWGBjIYBe7H4SEhDh9ttvtPPPMM6xYsYL69evTtm1bwsLCmDp1Kr/99htbt26lgKsXJ0m2cOVK0rKgIIiPP5nuwHXDjeD1zdAu1K69n5iYMpjNYLc713P1Hq9Ui46Gnj1h3jzH50GDeGPzJK7ucP5Pb8MG+OsvqFzZ8fmFF17g+9mzaLP9IAvrlqNaoOs3Msfb7fTed5gNlyNZ9t0PNC/fnKMvHyXIK4iH+7nuUpJ3d4mIiIhIprpnQ1dQUBBjx469Y70ZM2awYsUKOnbsyOzZszH9uzf4l19+yYsvvsjrr7/OlClTMrm3kl4JCUnLPDzgt99+48TxY7ToPTxdgeuGgPyFCCz0DZdON8VstpMnj5nLl53rWK3pbPzECceblXfvdnT688+hd286fARbXSybHDUKfv7ZsUW9j48PS39dzmMtmtP09z9pF5yXPiGFqZXHsdvHlXgrs09cZHLgWS4+G8/H1T+mefPmAAR5BbFyJaxdm/QaZctCnTrpvB8RERERSZd7cnlhWnz99dcAjB8/PjFwAfTr14/SpUsze/ZsYly9dVeyhX839HMSGQmdO3ejX79+rPxmIjt+nZfu9hdOOsThXU0BmDLF8VLk1PQhVTZudASuAgXgt9+gd28Aund37L54u8WLYfbsm58LFizIht83Mfatt9lot9B4w37yLfmD4OU7KbF8B68VO8nF5vHgD2eLnE08LzIy8VJJvPgimHP8f/UiIiIi2cs9O9MVFxfH9OnTOXPmDAEBAdSpU4d6t62bio2NZdu2bZQvX56SJUs6HTOZTDz66KNMmTKFHTt20LBhw2SvE3fLd+KRkZEZfzOSrBIlkpbFxUFoqJnPP/8cgK8mvwpA7cfapant5V/vZsevrQB4/XWDxo3NREWlrg+p0rkzXL4MzzwDt4y/oCDo0QO++CLpKf36QUgIPPTvI1++vr688sorDB8+nNWrVxMaGsq1uGv8ZP+J3TG7ARjbeCyjG48GICbG8U6xkyeTtu3v77iuiIiIiNxd9+zPvM+dO0fPnj157bXXeOmll6hfvz5169Z1evfS4cOHsdvtlC1b1mUbN8rDwsKSvc748eMJDAxM/FW8ePGMvRFJUa1arsu3bHFsOPH555/Tt29fFk1+NU0zXhvmbmDjD80Bdzp3Nhg3zsSWLWnrQxIJCTBuHNy6QcugQU6B64Zx46BgwaRNXL8OLVrA99/jtAmGm5sbLVq0oHX31szzn8fumN1Y3CzMaTOHMU3GYDKZOHECmjeH1atdd+/9929usy8iIiIid889Gbp69uzJb7/9xvnz54mOjmb37t107dqV7du306xZM65dc7zfKOLfFzwFBga6bCfg3xcWRbh6EdS/Ro0aRUREROKvk66mECTTJBd4vv3W8Xt6gtee1YtYMXUQxYsfoXFjg2++MWEy3WwztX1wcvUqPPEEjBkDzz6bdDeO2+TP73qmCxzBq1MnRzOhoTfL/7n0D3W/rsuec3so4FOAtd3X0rFqR65fh88+c2wF//vvrtt8+GHHLJqIiIiI3H335PLCMWPGOH2uXr06M2fOBOC7777j66+/ZsiQIRlyLYvFgsViyZC2JO2Cg6FqVdi/37l8yxbYuhXq178ZvODOSw33rF7ET+8Pp2fPHkyZUo2YGBMWC+zaBevXJ61foYLLiSpn//wDTz/tSEg+PvDyy3d8cCosLIyQkGu0bl2ChQvzu6yzYIHjV716kTRqFEGZ8rHkcStIYN68fNZgCcc3hTBnk+M5sJR2WCxUCKZN07NcIiIiIlnlngxdyenXrx/fffcdmzZtYsiQIYkzXMnNZN14Piu5mTDJHl54AQYMSFrep4/jxcWenqkLXrtW/sz8D3bSo0cPpk6ditlsxsPDsTthchtPvPCCYzfBZC1fDh06QESE4+Gvn3+G6tVTvJ+wsDDKlSv37yc3YA7wXDK1DbZt82fbtgCgOPiugAQvmsWlbszmywcrV/6H59JERERE5D/LUT/7zp/fMWMQHe14oWzp0qUxm83JPrN1ozy5Z74ke+jSBfz8kpb/+SeMGHHz2aeUlhruXrWI+R/4AHPw8fkG8y3TPq++6thk8Hbe3o6dBl0yDJg0ybGkMCICGjSA7dvvGLiAxOWvlidL4dW9HJaub+FW1sWySLd4aNUTmt4ysxtdCFIZuEqUcLz/q1q1VFUXERERkUySo2a6tm3bBtx8QbK3tzd169Zl69atHD9+3GkHQ8MwWLVqFb6+vtSuXTsruiupFBAAw4c7Hpe63UcfOcLRO+84ls+5mvFy9/Dgp/fPARMwmQweecQxdWUYMHYsTJzo+rpDhjh2GnTp+nX46ivHs1u9ejneweXpmab7MuXzwq2w46XH7m0+IOHgLuJWjYCYPOBzCZ5rCyEbwO4Ge7vCldT/cKBnT5g8OYX+i4iIiMhdc8/NdP39999cv37dZfnIkSMB6NSpU2J53759AceGGMYt28FNmTKFI0eO0LlzZ7xdvTRJspVXXoH773d97L334NFH4dgxx+fbZ7zmvbcVmADApEkmWrd2vLe4RQvHLoKuVKoEb7yRQod8feGXX+DTT+Hrr9McuFxxr7gan17tcXvwM+hdzxG4YgNg9tJUB64aNeDXXx2bgihwiYiIiGQPJuPWJHIPGDt2LJMnT6ZRo0aULFkSX19fQkNDWbZsGVarlVGjRvHuu+8m1rfb7Tz++OOsWLGC+vXr07hxYw4dOsSCBQsICQlh27ZtFChQINXXj4yMJDAwkIiIiMTdD+Xu2LMH6tUziI93/ZCVjw907ep4AXC1amAYdrp3/5K5c/uQkODBgAHQt6/jJcgzZsC/q1CTcHeHzZuhTp3bDuzc6djR4z++7GrXrl3UqlULr+4VE2e6brCZI4n1PAwmG8TkhYWzILQFKf18JG9eePJJ6N8f6ta9wzNoIiIiInLX3XOha/369Xz++efs3r2b8+fPc/36dfLnz0+9evXo378/zZs3T3JOXFwc7733Ht999x0nT54kb968PPnkk7z99tsUKlQoTddX6Mo6VquVhg0/Ydu2wbgOIQYEnIaC+/EvEE6RAt4cXfkY1lgLBQraMexmLl1K+RomE3z3neO9xk7mznWs2bNaYc0aaNQo3feRXOiyul0k3uMEmAzMNl+84u/DhAdGnA/2C+VJOFychG0x9Os3kDJlgild2rGdfcmSCloiIiIi2dk9F7qymkJX1rBarXTs2IlFP/9M1ceWsHfpoxh2E5itUP4XuP87KLYF/G55MfGf7WD+HCi8F7o1g6v3QegTsKsPRBZLcg2z2bEsz2nzDLvdsc7wxuzp44/DnDnwH3a8TD50XSLe8xhuCXmxWEMw3RYsbeeiiZ1xkJ07d1KzZs10X19ERERE7q4ctZGG5Ey3Bq6Ww/6P0vVKULxWGCu2/Eh85S8g4IzrE6vMA5/LUOAAeEeA904I3gmN3oF/noa1b8KFqoDjfWDTpoHTROm1a471ij//7Pg8fDiMHw9ubplynx62/JjjPDHb/TGhqSsRERGRnEKhS7K1pIGrGeeu7+N3y6vEP3A46Ql2E8QFOEIWQOk1SeuYbVBxIZRbDBveoFupUXz0fx7OG08cPep44fGff4LF4tgso2vXDL03w81KrOchLPElMeEBgJtds6ciIiIiOc09t3uh5B63B65SdR9m54Wv+SmsI1fjXAQugLVvwVc74VIqdvtzS4CmY/jrgQeIcTvrfGzxYkfgKlwY1q/P8MBFMMQFH8XmFk6c5/GMbVtEREREshWFLsmWXAWuTWcnsPnsZAzsrk/a2Qs2vgZXy8Dpeqm+1s6zO3lo2kOciDhxs/Cll+Dtt2HHDqiX+rZSY/WZ1dATcLdhsnvjaS2eoe2LiIiISPai0CXZzo3AtXD5IiqNbMeRImv49kAjdl+clvxJhx6FpV8C4FtpCuadc3iv9Hss77yc8c3G82S5J3EzJf8s1olLR1jYuSZXzh1zFJhM8NprULRoht2XYRi8u/FdRu4cCR5gvu6Ld1wFzIYlw64hIiIiItmPQpdkK1arlSd6PMEC6wJMw8zs95jDwasLuJ6Qwl7v56rCjz+B3Z2C9XfTaUwNyoY8zms9X+Panmu88tArLO64mGODjzH0gaFJwle+aFj5Hby89DKHn2oAmbChZ1xCHN0Xdee1Na85CraCx/nimMicTTlEREREJPtQ6JJsIzYhlofefIhV963CqGnHZoq/80mRwTBnKcQHQMhaLjxSj7mHWlH++acJqd2EDh06sH//fgCKBRRjYvOJbO29lXL5ygFQ+Tz88TU0PQbXPOGt+84w78BPGX5vMQkxbD+zHTeTG69UfQWWA5fjsJ2LTvUv43JshvdLRERERDKf3tOVRnpPV+bYd34f7X9qz9+X/k7biT/+CAfaQf6D0OtB8A5PPOS2x5MiewqxY+v2JC/BvnT9Em8PrcNb3x7DPx4O54GnO8KBglDQtyCHBx3Gz9MvA+7spsNXDnPk6hFC7CGUK1cu3e2EhoZStmwqNgoRERERkWxBoSuNFLoy3u8nfueJOU8QGReZ9pOj88Hir6DFUMhzLOnxQ1BgVV769n6RF198kaJFizqWD44fj/H665gMgzUh0O45uOJz87QpT06hb62+6b0lAFYdXsXR8KMu2wkLC+PatWtpbtPf31+BS0REROQeo9CVRgpdGWv76e08PPNhouKjUn2On0dhygQ253JsKBeu/0W8PeXwUuSYG5FzwNPLl0WLF9OoShWoVg1Onya0Y3Mq37eShNserapeuDq7+u7CZErfS4q/3PElA5cNxMBgXfd1NCzZMF3tiIiIiMi9T890SZa5GnOVVj+0SjZwmXAdeEr9uYS8f71H6zLTeL7CegrsrIDpVPLXORti4+mnfShns/F4y5bsPHoUFi2CKVMoO3s51YvXTnLOnnN72Hd+X5rvyWa38b/l/+PFpS9iM2x0qtqJukXrprkdEREREck5FLokywxeMZgz1864PFazcE0s7km3UvcLHcT+72ux9osinDnozaGNy7m0+G8+W1KQ17blw9vqOqgdC7jGO4XdKQX07N4No1Yt6NsXk8lE/9r9XZ6z+eTmNN3PtbhrPDP3GT7c9iEAbzV9i5mtZrq8DxERERHJPRS6JEssP7ScmXtnujz2bKVnmfr0VGITbtut78QDRP84EYD7n7hCcMUY9i/5jgd9fGjs40unfwKYtbwIeWOch3W3PbB2BjQ4FsFoP2/2/3WAzZtvBqrWFVu77MeOMztSfT/Hw4/T4NsGLA1bipe7Fz88+wOvN3o93csTRURERCTnUOiSLPH2hrddljcr1Yw5beawZv8a5wOXy8D3v2AkeBBc6TTlGq/l7/WLOX/0IJ38/ROrVbhqYcpvhfG0mTDbYcIKmLEILDZYVhZOVjQR4uXFl198kXhOkFcQ9+W9L0lf9p7fm+r7WRq2lP0X9lPItxDre6znucrPpfpcEREREcnZ3LO6A5L77D23l00nNyUpD7AEML3VdI4dOcawN4bB4/8eiM4Hs3+FmPzgtp0zB5owb8T1xPMe8PZyaqfSFQvDtgfw0KEIHjvkKBvXCMY2geqXoqi3x5P9e/Y4nVM+X3kOXTnkVHY19mqq7+nF2i8SERtB52qdKRFYItXniYiIiEjOp9Aliex2O6tXr2bRokVcvnwZi8VChQoV6NmzJ0WKFMmw63y962uX5W82eZNiAcXYdWjXzTnYBA+YuwiulIXAY1CpG9h8AV+4EIPpWBSety3h84y38sZv0XhZ4bo79GgF86o4ju0uGEfx/O7ExMQ4n+PmmaQ/Vps12XswDIMvdnxB56qdCfQKxGQyMarhqNR9AUREREQkV1HoEgzD4Ntvv+Wdt97i6PHj5HX3xMduYDeZ+AEbo0ePpnXr1kyYMIGQkJD/fL3VR1YnKfP18OX5Gs/fLEj493c3K5RdBhcrQefH4fRliP43IMUkYACXbDYKuN8cyoFRUXhZE4jwMtO0m53dwc7XCitqJU9sPqcyV7Na3h7eLvsflxBH78W9mbVvFotDF7O001LMJq3UFRERERHXFLpyOcMwGDZsGJMnT6YMHjyDD4USzInbtcdhEIaVlYt+Zv3adaxZt5YqVaqk+3qRcZGEXg5NUt6qQisCLLe89yzi399NQKPxUOsr8L0M4UE3Q1ceCyYT/BIVRa+goMRTL+YJAgxO5PNjT5HTSa71T34r42o94/Q12HNuT5J6xQOKJym7GH2R1j+0ZtPJTbiZ3GhVvpUCl4iIiIikSN8t5nL/93//x+TJk2mAhUfwojBuTu/HsmCiCp60SvCE8AhaPPooFy5cSPf19p7bi0HS93HXK1rPueBkS4jzvfnZ97Ljd//4m2WebhiFvJkXEUmeq1cx3XjPt8nExbx58TY8KRXhkeRa9iIGvXr1Svz896W/CY8NT1KvdrDz+7sOXDxAvan12HRyE4GWQJZ3WU6/2v1SvmERERERyfUUunKx6OhoxoweTWU8qELSZ5pu5Y2ZFjZPLl24yGeffZbuayb3Xi7O3/zj5s0BELsYvt0AsQHO9QrEgNme+LFgsA/f22wUuRpO4YuXkjRb/mrS+7Lks1C4cOHEz9/u/tZll2oVqZX45xWHVvDANw9wNPwopfOUZmvvrTxS+hHX9yIiIiIicguFrlzs+++/Jzo6mvvvELhu8MXMfXYzU774Aqs1+U0mUhJni3NZPnzwCDZu3MjevTByZCnAHfwPgCXSuaK7AYUcOxdWj4xn+1/hNADCgfHWeA7FxztV90lI+p4sNy+3xD+Hx4bz7Z6kocvD7EHTUk0dfU6I44WlLxAZF0nDEg3Z1nsbFfJXSPU9i4iIiEjupme6crHvZs6kuNkDf3vqs3dFPJh/8SJr166lefPmab6mq10CAXyKhdC8eS/8/fdz/boFWAMlh4CrdwuXjOTZP2HG3kh87Ab/+LjzVHFfjhy+xoRTp6nj40MdT088TSZ2WGOTnH5rH4auGMqVmCtJ6rSr3I78PvkBsLhbWNh+IV/u+JKPWn6Exd2S5vsWERERkdxLM1252OmTpwiy37nerfL8O2TOnElmmeAdJEQmuCwv/vhzJBhLuXjRQpEiV4E2cA245vxMlskOYzcazNsdgY/dYHk+C/XqFCCsuB+26nkBMCpUYKGnJ7MMg3OFkg7xG2Fq3l/zXM5yAXSr1o01R2++oLl64ep8+eSXClwiIiIikmYKXbmYyeRqGillN7bASM+5x44dY0SPV5IesLkTNr0VCXFlMbtd5vLl+iRuX3jC+ZmuIlHw0jbHnyfVcOPJGnmI8Ph3GJsdfZry9decOX+ekxfPYiuQNFXWKFyD+Qfm03lBZ5f9bFiiIcNWDePx2Y+z9dTWNN+niIiIiMitFLpysRIhJbliTlt4uowjxJQoUSJN5x07doyGjRsTHWnG21TQ+WBECWJOF8HN00bjEafwK3rLqterXnDeJ/HjmQBo9xz0eAaGPWPDVvMS+Lt+TmzBwQUk2JPOrJ2MOMmz857Fak/6XJqPuw8HLx7kzwt/ksc7D+5mrcAVERERkf9GoSsX6/n885yyxxNO6tcYHsRK8aJFadSoUarPuRG4IuMMGo38ioI+zluxk/cI9K5H2T5zKVDezv2dhjkdfminN4/8fXOorikNM2r8+8EnAapdggqXIV984jNghmHw2XbXuyxuPZ387FW8PZ5LMZeoXrg6f/T+I8m28SIiIiIiaaXQlYs9++yz5A3Kw27iXb4763YR2DlittF/4EDc3NzuWB+SBi6fvIUo5dfKcTDW/2bFoBMcLzyUePs13D1uPjfV+3Q0a7ZfZt5PBmWSez2YCcgfC/UiYBT02tSLJtObsPnk5lT18VYJ9gSeKvcUG3tupHhg0pcji4iIiIiklUJXLubl5cWESRMJxcqOOwSva9hZ4R5PyZAQXnjhhVS17ypwARTwqoXP6Xbw4TE4+Exi/RjbefZcnoiBgTvw8aFIvj4YjocBywM8OfN3IMTfYTmkJ+y5socNJzakqo+3G/rAUBa2X4ifp1+6zhcRERERuZ0eWMnlnn/+eS5cuMCoUaO4YDaobHejBO6Y/12nF4Wdg1j5281OweBgVqxaRVBQ0B3bPX36tMvABRB13ou42dMh1gf2dYUKPycuCzwWvZi8cQbLgWZXHM9qvQa8eyUO1sTBHqANUCxj7t/Hw4f3H3mfTSc28XCph+lTq0/GNCwiIiIi8i+FLuGVV16hYsWKvPPWW6zYuRMfdw/8TGZswNWEeLy8vOjeoxdjx46lYMGCd2wP4O+//+bUiROUafos3nlunhMb6c7GyWWwXbfgXuwvElp3dXoXV6UL8MP3S7gPiPGw8H6b/iyMjoAl3zJ48GC6du1Kgj2BWUdmMeWfKcTb45NePJUalmjItGemUSZvGQbUGZCuHRlFRERERO7EZBjGnR/mkUSRkZEEBgYSERFBQEDAnU+4x+zcuZOff/6ZK1euYLFYKF++PB06dEjXvX7yyScMGjSI8i27Uq39YGxWM+s/uI/Lh/zwzR/HA6M2sC6qFQnG9cRzPloGg/6Ao0HwbOcALnk8xIkvl9Grcy+++uorzOabK2JPR55mys4pfLHjCy5dv5SqPplNZuoWrcu+c/t4psIzzG4zW2FLRERERDKVZrrESa1atahVq1aq6kbFR7H77G52nt3J35f+Jio+CgMDXw9fyucrT62navG+8T4jXx6JYZi4fuVDLh/yw8MngYZDDhOQLz/1vN5i88XhGP/uoDisOSSYYXxDuOQbCSzDfYQ758uf560Nb1E+f3k83Ty5GnOV3ed2s+XUljsGLi93LzpU7sCDxR/kXNQ5xq4fi92wc/raaaKt0Xp+S0REREQylWa60iinz3Tdid2ws+rwKj7f8TnLwpa5fA/WrUyYCLGFcPTrhnBuBiY3O42HHaZgxSgAPOJjqbvsTYZUW4XNnPqt61OrYv6K/NbtNwr4FmDw8sGJ28j3qN6DKU9OwdPNM8OvKSIiIiJyK810SaqtOLSCQcsHEXo5NNXnGBgcdTsKfY/DLzUJ9IqiQIXHABN5rl5g+MdDKHP0AAWimtK90Q6s9msZ1t8mIU2Y124eHmYPnpzzJCsOrwDgvWbvMaLBCC0rFBEREZG7QjNdaZQbZ7oiYiMYsmII3+75NkPa879Ukr4Bw3n187HkDb9EpF8QHw54nx1lQ9gY9hLh3qkPda7c2JGwf53+mDDRZEYTNhzfgLe7N7PazKJNxTYZch8iIiIiIqmh93RJio6HH6fO13XSF7jO3g9LPoMED6fip08f5+33B5I3/BInit3Ha6Nn8lfFOpzfto3wD0JpdqkZDUs0TPPlfD186VerH3/1/4uBdQdiNpkxmUyMazKOkKAQNvbcqMAlIiIiInedZrrSKDfNdB0PP07DaQ05GXnyjnVNmMjnk4/o6GhijBi4VhSmbnP8/tC78MhrmO0wfjWM2Ow455eyZqa9NAPPgMoc27SE7VPH8PzzzyfuUrj//H5m7J3B5pOb2XNuDzEJMUmuW9S/KLWCa9GyTEs6V+tMgMXxd3I+6jyF/G6+GyzeFq/nt0REREQkSyh0pVFuCV2RcZHU/qo2YVfCkq0T7B9Mn5p9eLT0o9xf6H5e6PUCs2fPpvQzwzi7/RVizuTDVPBvjJ71wTuCspdg9xTwtcLbDWF0U3Cz+lD1ykD2fDXBKXDdLsGewKErh7gacxWr3Yq3uzclAks4BSsAwzAYu24sH237iE3Pb6JywcoZ/rUREREREUkLbaQhLg1bOSzZwOXn6ccHj3xA75q98XDzwDAM2rVrx/z58ynXegiX/36NmDNBeAbEU2dAFOG+g/gn+kPC8l+jW2vwsMEPVR1tJVius/vKB/R6Pul7uG7lbnanQv4KKfY5xhpDz5978sNfPwCwLGyZQpeIiIiIZDmFLkli5eGVfL3ra5fHahapyYLnFlAyqCSAc+BqM5jocyO4fDAIs6eNGi/8w4OXthB7rST5SsxhX+QbLKi0J2mjleGRNo8kG7hS41zUOVrNbcW209twN7vz5RNf0qtmr3S3JyIiIiKSUbSRhjgxDIPBywe7PFazSE3WdFvjMnCVbzsYu+1/nN5cCEwG1XqE0vXIdMZ/9j/GfDWS4Eg7NQP/jyD36i7bHrZqGFabNV193n9+P/Wm1mPb6W3k8crDqq6rFLhEREREJNtQ6BIna46u4eClg0nKfT18mf/cfAK9AoGkgatE0/YElIjC3SuBKm1CeefPVxjw02Tc7DZ2VahLpG8gbiYvqgW8g4cpMEn7p6+d5ud/fk5zf/ec28OD3z7IiYgTlM1blm29t9EkpEma2xERERERySwKXeLk8x2fuyx//5H3CQkKSfzcu3dvp8AFkK98JI8PWcucPe15fPMv2ExmprR+iQld38DqYQHAYs5Leb8hLq8xaeOkNPe3UoFK1AmuQ9OQpmztvZWy+cqmuQ0RERERkcykZ7okUYw1hiWhS5KUF/ItRJ9afRI/L1u2jOnTZwAQfdGH6HNe+BaOpfSpUN6cMpJCV88R7eXLuz3fYnvlB5K0V9jzEY6Yp3Ld7rwV/dZzW7kYfZECvgVS7KfVZsVkMuFudsfTzZOF7Rfi7eGtLeFFREREJFvSTJck2nd+H/G2+CTlvWr0Sgw0y5Yto1Wr1uSp9ABFmw7m1IaebH2/LBHHfGn32xwKXT3HqQLFGTRsqsvABWAymSnq1crlsR1ndqTYx/DYcJ6Y8wTDVg5LLAv0ClTgEhEREZFsS6FLEiUXeB4p/QhwM3AFlK9LmefeIvLIm8B92K3nObdrGh91GMGiRs8yaPhUThYOSfFaAQmut3/feXZniufFJcQRejmUr3d9zbHwY3e6JRERERGRLKflhZLon8v/uCyvWaSmU+Aq2/FNwn6ojO24Gy95fMT8epc4seb/cLdE8vlzQ+94nfiocMK++D/oTpIR6GoTj1sV8ivEkk5LiLfFOz1jJiIiIiKSXSl0SaKo+KgkZQGWAA7sPpAYuMp1fpMTy8visy+GVTSnhnUPFQP68f5jfTny61cAlHmid7LXiI8KZ8eH/XGLuUpwQDBnrp9xOn7dev2O/axSsEoa70xEREREJOsodEkiA8Nluc1mw8DAZHbn3LailFx/goW0piAXueKXh30h1ShWqipAisErPiqcHR8N4PqFE3z26Se8H/d+0j4YrvsgIiIiInKv0jNdksjPwy9JWWRcJNXqVGPejz9yef/vNFs2h7U0pSAXCQ0uS59BX7H/38BV7OEulHysL0d+/YbDS6c6tZMYuM4f552336JPvz6cjz6f5Hq+nr6Zc3MiIiIiIllEM12SqHz+8i7Ld53Zxapff2WykcDg+HEArKnSiHc7vkasp7dT3WIPdwGcZ7xuD1yjRo1i3/l9xCbEJrlWhXyuN9gQEREREblXKXRJotrBtV2WD/7wNcxfbeZDkxkMO2NMZj4Dyrl5uJwqvTV42ayxXPprC9Hnj1M0uAgDBgwAYM3RNS6vVSu4VkbcioiIiIhItqHQJYnuL3Q/HmYPrHbrzcLYAPbO+ZKAchd5v/Jyonz8WWA2c2XOWEJnv0m5zmMwuyUdRsUe7oJhGBxf/jWY3ShY90ku7vuNR5u3YNPvG/lyx5cu+5Bc8BMRERERuVfpmS5J5O3hzRPlnkj8/NjfZirOmgiXqhJ1qgYLK7ZiXZVG5K30EPd1GsuVA5sInf0mdltCkrYMu53Yy46dCYu36EPRpl1x9w3i1KlT/LD/B5fb0zcq2YiCvgUz7wZFRERERLKAQpc4ebH2i2DAsN9hyVyDX069Tx73E9jbN+Pa1VWJ9VIKXobdzqGfJnBhxzJKt3uV/NUfIfSbIQR6GCxcMZ//rf6fy2v3r90/0+9PRERERORuU+gSJ48EP8TCZQFMWA1mDH7jYaLavAAldxNme5Pw438k1nUVvJwC17OjCLyvJqHfDCHA3cbqtSsZtXMUF69fTHLdIn5FaF2x9d28VRERERGRu8Jk6MVIaRIZGUlgYCAREREEBARkdXcy1tmz0KoV/PEHCbgxmA/57LG/oN4tz1+dMVHWbSx5SzROLLpy4HcOzRlL3koNcPPycxm4Fq1cwCs7XmHtsbUuLz27zWw6Ve2UyTcoIiIiInL3KXSlUY4NXdu3OwLXmTNcIQ/tmMeaB3ZDi+FJ68ZCkdiOFC/YG5PJDbgZvAy7zSlw+bsn8MrMEby9620uXb/k8tKtKrRiwXMLMJlMmXiDIiIiIiJZQ6ErjXJS6DIMg40bN7Js2TI6z5lD1ZMnuRoczNrBq/h+533srF+JoxGHkz3fIyEPhSxtCDTXxsdUmqijB7Fb4/AqUoJ/fvkfHsWvk/fRPBy/djzZNor4FWFXv10U9iucGbcoIiIiIpLlFLrSKCeELsMwmDlzJh+MH8+Bf/4hn6cnpTAxKCGBgXYb+PnRs1cveg7pyeM/Pc6Za2fu2KYJN9wJwjDsJNivgtud+5HXOy/re6ynSsEqGXBXIiIiIiLZkzbSyGUMw2Dw4MH06NED36PHGe0bxIdehQjgG677VeZd/3w8bLUz9bPP6NS8PfOfmE/pPKXv3C42rFwmwZS6wFXEr4gCl4iIiIjkCgpducy7777Lxx9/TC9vP4b7BFDJ3cInMZ+zJr4z46PmkN/sTidvP97yDuD8kSO8+Fw/NnbdSNdqXTOsD60rtGZ3v90KXCIiIiKSKyh05SKXL1/m7XHjeNriQwuLDwCzYkfzh/VJ3Imjp8+rmE2O1abF3NwZafFj7759LJu/jJmtZ/JLh19SNeuVnKL+Rfm+7ffMf24+hfwKZcg9iYiIiIhkdwpducj06dOxJSTw1L+Ba3lcT5bEOV5I3N9nMJXctzrVD3H3oKanF59+/DGGYfBU+acIHRjK4o6Leey+xzCb7jx8TJh4uNTD/NTuJ46+fJQOVTpol0IRERERyVW0kUYa3csbaVSrXJmAQ0d42SeAndZH+SB6BgZudPB6lzZeH7k8Z5c1jveiI9i3bx9Vq1Z1OhYeG86us7vYeWYnBy8dJNoajd2w4+vhS4X8FahVpBY1i9Qkn0++u3F7IiIiIiLZkntWd0DunpMnT/G42Y0jCVX5MPorDNx42HM2rS2uAxdASTfHEDl16lSS0BXkFcTDpR7m4VIPZ2q/RURERETuZQpduYiBgQkIMF+moPk4eczn6e09gpRW+904pAlREREREZH0UejKRYKLFOHksRPkN5/hLf+nAHA3JaR4zkmb43hwcHCm909EREREJCfSRhq5hNUKDRqNZmuClSi7HR/TNXxM1+543mprHJXKl+f++++/C70UEREREcl5FLpyoF27dvHGG28QFxcHgGFAv37wzTedsDGQZXHXU9XOKVsCO6xxDBg0SDsOioiIiIikk5YX5jDbt2/n4WaPEHUtkp27drFwwQImTLAwbRqYzSbaPluDeT9ep7CbG408vZNt54LNxvuxUZQrW45u3brdxTsQEREREclZFLpykBuBK84zPx412rJixSzq1vuYfXuHA/DZZ9C3b1d8fdfz6bRp7Euw0tLTm/vcPRLbiLDbWRMfwzJbPPmCg/l15Qr8/Pyy6pZERERERO55ek9XGmXX93TdGrjMNfth8vDCetiG7Z/+gIUhQxKYNMmRsQ3D4NNPP2Xi++9z4vRpilksBGEiHjhijcfd3Z2OnTvz/vvvU6BAgSy9LxERERGRe51CVxplx9DlKnDZo/IQv6UdWL2A+bR8bBqLFs7HYrEknmez2fj1119ZtmwZ4eHheHt7U7VqVbp160bevHmz7oZERERERHIQha40ym6hy1XgAkg4XIuEfxpgCjqLW5n3sO/5ghYtHmXhggVOwUtERERERDKXQlcaZafQlVzguiHhZEXcCh7DZInBdvFvbLumKniJiIiIiNxl2jL+HuUqcBkGGPabf6XuxQ9issQA4FagAm41e7NixSpat2mTuJ28iIiIiIhkLoWue1CySwr/fgjrH60wrK5nsRS8RERERETuPoWue0yygetYNWxHa2K/Ugz7laLJnq/gJSIiIiJydyl03UMuXLjAw80eIdY9j1Pgsp0vRcKBRgC4l9uMW6EjKbZzI3gt/3U5/fv3z/R+i4iIiIjkZgpd9xB/f39KliwJMVcwYq8AYI8ogHVPS8CMW/E/cSuzI1Vt2a8exTDsVKtWLRN7LCIiIiIiCl33EG9vb+Z+PweT3Ur81k+wXbxO/I6nweaBOf9x3Cuvw2S6czvW0F+xHVrOu+++y8svv5z5HRcRERERycXcs7oDknqnT5/m6WdaYZjcweSBdUcjMHwx+V/Co8avmMz2O7Zxa+AaNWrUXei1iIiIiEjuptB1jzh9+jQNGzXm1JmLmEu3wuJmIS5sOMRPxL3iOkwevndsQ4FLREREROTu0/LCe8CtgctU6mnMlkBM7l5YylYEy9NYd7+D/dqZFNtQ4BIRERERyRoKXdmcq8B1g8ndC8t9z2By9yZ+66fJBi8FLhERERGRrKPQlY2lFLhuMLl74Vkm+eClwCUiIiIikrUUurKp1ASuG5ILXgpcIiIiIiJZT6ErG0pL4Lrh9uBl/XMetkPLAYiPj8/sLouIiIiISDIUurKhkSNHcvTIYUwlWqYqcN1wa/Cynfgd9yL1MQeVZezYsezbty8TeywiIiIiIslR6MqGhg4dSkBAIMbZDRgJcWk61xG8WuFR5hlMlkCMiMN07NiRypUrZ1JvRUREREQkJQpd2VCNGjVYt24t3qYY7MeXpCN4WcAWS8LxVXTo0J6ZM2fi5uaWSb0VEREREZGUKHRlU/8leNnCDzkFLnd3vQNbRERERCSrKHRlY+kJXgpcIiIiIiLZi0JXNpeW4KXAJSIiIiKS/Sh03QNSE7wUuEREREREsieFrntESsFLgUtEREREJPtS6LqHuApeClwiIiIiItmbyTAMI6s7cS+JjIwkMDCQiIgIAgICsqQPu3fvpkmTpkRb3bDHhitwiYiIiIhkY5rpugfdmPEK8DbRqVNHBS4RERERkWxMM11plB1mum6Ij4/H09MzS/sgIiIiIiIp00zXPUyBS0REREQk+1PoEhERERERyUQKXSIiIiIiIplIoUtERERERCQTKXSJiIiIiIhkIoUuERERERGRTKTQJSIiIiIikokUukRERERERDKRQpeIiIiIiEgmUugSERERERHJRApdIiIiIiIimUihS0REREREJBMpdImIiIiIiGQihS4REREREZFMpNAlIiIiIiKSiRS6REREREREMpFCl4iIiIiISCbKNaFr+/btPP744wQFBeHr60v9+vX58ccfs7pbIiIiIiKSw7lndQfuhrVr19KiRQu8vLzo0KED/v7+zJ8/n/bt23Py5EmGDh2a1V0UEREREZEcymQYhpHVnchMCQkJVKhQgVOnTrF161aqV68OQEREBHXr1uXYsWOEhoZSsmTJVLUXGRlJYGAgERERBAQEZGLPRUREREQkJ8jxywvXrFnD4cOH6dSpU2LgAggMDOTVV18lPj6eGTNmZF0HRUREREQkR8vxoWvdunUANG/ePMmxFi1aALB+/fq72SUREREREclFcvwzXWFhYQCULVs2ybHChQvj5+eXWMeVuLg44uLiEj9HRkZmfCdFRERERCTHyvGhKyIiAnAsJ3QlICAgsY4r48eP580330xSrvAlIiIiIiIA/v7+mEymZI/n+ND1X40aNYohQ4Ykfj59+jSVKlWiePHiWdgrERERERHJLu60yV6OD103ZriSm82KjIwkT548yZ5vsViwWCyJn/38/Dhw4ACVKlXi5MmT2sFQUhQZGUnx4sU1VuSONFYkNTROJLU0ViS1NFYyhr+/f4rHc3zouvEsV1hYGLVq1XI6du7cOaKioqhbt26q2zObzRQtWhRwLE3U4JTU0FiR1NJYkdTQOJHU0liR1NJYyVw5fvfCxo0bA7By5cokx1asWOFUR0REREREJKPl+NDVrFkzSpcuzZw5c9izZ09ieUREBO+++y6enp5069Yt6zooIiIiIiI5Wo5fXuju7s7UqVNp0aIFjRo1okOHDvj7+zN//nyOHz/OxIkTCQkJSVObFouFMWPGOD3rJeKKxoqklsaKpIbGiaSWxoqklsbK3WEyDMPI6k7cDX/88Qdjxoxh8+bNWK1WqlatypAhQ2jfvn1Wd01ERERERHKwXBO6REREREREskKOf6ZLREREREQkKyl0iYiIiIiIZCKFLhERERERkUyk0JUG27dv5/HHHycoKAhfX1/q16/Pjz/+mNXdkixw+vRpPvzwQ5o3b06JEiXw9PSkcOHCtG3blm3btrk8JzIykiFDhlCyZEksFgshISEMHz6cqKiou9x7yWrvv/8+JpMJk8nE1q1bkxzXWMndFi5cyKOPPkq+fPnw8vKiVKlSdOzYkZMnTzrV0zjJvQzDYMGCBTRt2pQiRYrg4+ND+fLl6devH0eOHElSX2Ml55s1axb9+vWjdu3aWCwWTCYT06dPT7Z+WseE3W7nk08+oWrVqnh7e1OgQAE6duzocrxJMgxJlTVr1hgeHh6Gv7+/0adPH2PIkCFGyZIlDcCYOHFiVndP7rKRI0cagFGmTBmjV69exiuvvGK0bdvWcHNzM8xmszF37lyn+lFRUUb16tUNwGjevLkxcuRIo3nz5gZg1KlTx4iJicmiO5G7bf/+/YbFYjF8fX0NwNiyZYvTcY2V3Mtutxt9+/ZN/Lelf//+xsiRI42uXbsaJUqUMDZu3JhYV+MkdxsyZIgBGEWKFDFeeOEFY8SIEUaLFi0Mk8lk+Pv7G/v370+sq7GSO9z4njR//vyJf542bZrLuukZE7179zYAo3LlysaIESOMLl26GJ6enkbevHmN0NDQTL67nEGhKxWsVqtRpkwZw2KxGLt3704sDw8PN8qVK2d4enoax44dy7oOyl03f/58Y926dUnKN2zYYHh4eBh58uQxYmNjE8tHjx5tAMbIkSOd6t8Ib++++26m91myXnx8vFGzZk2jXr16RpcuXVyGLo2V3OvDDz80AKN///5GQkJCkuNWqzXxzxonudfZs2cNs9lslCxZ0ggPD3c6NnnyZAMwevbsmVimsZI7rFq1KvF70fHjx6cYutI6JtasWWMARqNGjYy4uLjE8mXLliUGN7kzha5UWLFiRZJ/xG6YPn26ARhvvvlmFvRMsqMbPy3avn27YRiOn14HBwcbfn5+RlRUlFPdqKgow8/PzyhdunRWdFXusjFjxhgWi8X466+/jO7duycJXRorudf169eNPHnyGKVLl3YKV65onORuW7ZsMQCjU6dOSY6FhoYagPHkk08ahqGxklulFLrSMyY6duxoAMb69euTtNekSRMDMI4fP56h95AT6ZmuVFi3bh0AzZs3T3KsRYsWAKxfv/5udkmyMQ8PDwDc3d0BCAsL48yZMzRo0ABfX1+nur6+vjRo0IAjR44keV5DcpZdu3bxzjvvMGbMGCpVquSyjsZK7rVy5UquXr1Kq1atsNlsLFiwgPfee48vv/ySQ4cOOdXVOMndypYti6enJ5s2bSIyMtLp2JIlSwBo1qwZoLEiSaVnTKxbty7x2O30fXDqKXSlQlhYGOD4h+52hQsXxs/PL7GO5G4nTpxg9erVFClShKpVqwIpj59byzWGcq64uDi6detG9erVGTFiRLL1NFZyr507dwLg5uZGtWrVaNu2LaNGjeLFF1+kfPnyDBs2LLGuxknuli9fPt577z1OnDhBhQoVePHFFxk5ciQtW7Zk5MiR9O/fn4EDBwIaK5JUWsdEdHQ0Z8+epVSpUri5ud2xviTPPas7cC+IiIgAIDAw0OXxgICAxDqSe1mtVrp27UpcXBzvv/9+4j9OqRk/t9aTnGf06NGEhYWxc+dOl//TukFjJfe6cOECAJMnT6ZmzZr88ccfVKxYkd27d9O3b18mTZpEmTJlePHFFzVOhP/9738ULVqU3r178+WXXyaWP/TQQ3Tq1ClxpYXGitwurWNCYyjjaKZLJAPY7XZ69OjBhg0b6NOnD127ds3qLkk2sWXLFiZOnMjrr79OlSpVsro7kk3Z7XYAPD09WbRoEXXq1MHPz4+GDRsyb948zGYzkyZNyuJeSnYxbtw4unTpwquvvsrJkye5du0aGzduJDY2liZNmvDLL79kdRdF5DYKXalwI90nl+IjIyOT/QmA5Hx2u53nn3+eOXPm0KVLF6efOkLqxs+t9STnSEhIoHv37lSrVo1XXnnljvU1VnKvG3+ntWvXJjg42OlYlSpVKF26NIcPHyY8PFzjJJdbvXo1Y8aMYeDAgbzyyisUK1YMPz8/HnroIRYvXoyHhwdDhw4F9G+KJJXWMaExlHG0vDAVbl2vWqtWLadj586dIyoqirp162ZF1ySL2e12evbsycyZM+nYsSPTp0/HbHb+Wcad1jvfaX213LuioqIS/349PT1d1nnggQcAxwtxb2ywobGS+5QvXx6AoKAgl8dvlMfExOjflFzu119/BaBp06ZJjhUuXJgKFSqwe/duoqKiNFYkibSOCV9fX4oUKcLRo0ex2WxJlshrDKWeQlcqNG7cmPHjx7Ny5Uo6dOjgdGzFihWJdSR3uTVwtW/fnu+++y7Zh0yDg4PZtGkT0dHRTrsFRUdHs2nTJkqVKkXx4sXvZvflLrBYLPTq1cvlsQ0bNhAWFsbTTz9NgQIFCAkJ0VjJxW58A33w4MEkx6xWK4cOHcLX15cCBQpQuHBhjZNcLD4+HoCLFy+6PH7x4kXMZjMeHh76N0WSSM+YaNy4MXPnzmXTpk00atTIqb0b3wffXi4uZPWe9fcCq9VqlC5dOsWXIx89ejTL+id3n81mS3zPUrt27e74Xh29nFJu5+o9XYahsZKb3XjH39dff+1UPm7cOAMwunTpklimcZJ7ff/99wZgVK5cOcnLkb/44gsDMBo0aJBYprGS++jlyNmTyTAM4y7nvHvS2rVradGiBV5eXnTo0AF/f3/mz5/P8ePHmThxYuL6ackdxo4dy5tvvomfnx8vv/xy4k5Rt2rVqhXVq1cHHD89atCgAXv37qV58+bUrFmTXbt2sXLlSurUqcP69evx9va+y3chWalHjx7MmDGDLVu2UL9+/cRyjZXc6/Dhwzz44INcuHCBJ554InGZ2Jo1ayhZsiRbt26lcOHCgMZJbmaz2Xj44YfZsGEDBQsW5OmnnyYoKIhdu3axZs0avL29WbduXeJjDxorucPUqVP5/fffAdi/fz+7du2iQYMG3HfffYBjZ8vevXsD6RsTffr0YerUqVSuXJknnniCs2fP8sMPP+Dn58eWLVsoV67c3b3he1FWp757ybZt24yWLVsaAQEBhre3t1G3bl1j7ty5Wd0tyQI3ZilS+nX7T5jCw8ONwYMHG8WLFzc8PDyMEiVKGEOHDjUiIyOz5iYkSyU302UYGiu52YkTJ4wePXoYhQsXNjw8PIzixYsbAwYMMM6fP5+krsZJ7hUbG2uMHz/eqFGjhuHj42O4u7sbRYsWNbp06WIcOHAgSX2NlZzvTt+XdO/e3al+WseEzWYzPvroI6Ny5cqGxWIx8uXLZ7Rv3944dOjQXbi7nEEzXSIiIiIiIplIW8aLiIiIiIhkIoUuERERERGRTKTQJSIiIiIikokUukRERERERDKRQpeIiIiIiEgmUugSERERERHJRApdIiIiIiIimUihS0REREREJBMpdImIiIiIiGQihS4RkVykSZMmDB48GICQkBA+/PDDLO3Pvej69eu0bduWgIAATCYT4eHhWfK17NGjB61atbpr1zt27Bgmk4k9e/bctWuKiOQUCl0iIrnU9u3b6du3b6rq3qsBLTP6PWPGDDZu3MjmzZs5e/YsgYGBafpaupKeQPPRRx8xffr0dF9TRETuHves7oCIiGSNAgUKZHUX7kmHDx+mYsWKVKlSJbHsTl9Lq9WKh4dHhvYjMDAwQ9sTEZHMo5kuEZEcKjo6mm7duuHn50eRIkWYNGmS0/FbZ4EMw2Ds2LGUKFECi8VCcHAwgwYNAhxLEo8fP87//vc/TCYTJpMJgMuXL9OxY0eKFi2Kj48PVatW5fvvv3e6RpMmTRg0aBAjRowgb968FC5cmLFjxzrVCQ8Pp1+/fhQqVAgvLy+qVKnCkiVLEo///vvvNGzYEG9vb4oXL86gQYOIjo6+4/0n12+A+fPnU7lyZSwWCyEhIUm+Nim1OWnSJDZs2IDJZKJJkyZJvpYAJpOJL774gqeffhpfX1/eeecdrl69SufOnSlQoADe3t6ULVuWadOmAVCqVCkAatSo4dRuSm5fXpiar3VKbvT5sccew9vbm9KlS/PTTz8lW99ms9GrVy9KlSqFt7c35cuX56OPPnLZx4kTJ1KkSBHy5cvHgAEDsFqtiXXi4uIYNmwYRYsWxdfXl3r16rFu3bpU91tE5F6g0CUikkMNHz6c9evX8/PPP7Ny5UrWrVvHrl27XNadP38+//d//8eUKVMICwtj0aJFVK1aFYAFCxZQrFgxxo0bx9mzZzl79iwAsbGx1KpVi6VLl/Lnn3/St29funbtyh9//OHU9owZM/D19WXbtm188MEHjBs3jlWrVgFgt9t57LHH2LRpE7NmzeLAgQO89957uLm5AY5ZpZYtW9K2bVv27dvHDz/8wO+//87AgQPveP/J9Xvnzp0899xzdOjQgf379zN27FjeeOONVC3VW7BgAX369OGBBx7g7NmzLFiwINm6Y8eOpXXr1uzfv5/nn3+eN954gwMHDvDrr79y8OBBvvjiC/Lnzw+Q+DVbvXr1HdtNSUpf69R44403aNu2LXv37qVz58506NCBgwcPuqxrt9spVqwY8+bN48CBA4wePZpXX32VH3/80ane2rVrOXz4MGvXrmXGjBlMnz7d6Ws9cOBAtmzZwty5c9m3bx/t2rWjZcuWhIWFpetrICKSLRkiIpLjXLt2zfD09DR+/PHHxLLLly8b3t7exssvv2wYhmGULFnS+L//+z/DMAxj0qRJRrly5Yz4+HiX7d1aNyVPPPGEMXTo0MTPjRs3Nh566CGnOnXq1DFGjhxpGIZhrFixwjCbzcY///zjsr1evXoZffv2dSrbuHGjYTabjZiYmDv2x1W/O3XqZDz66KNOZcOHDzcqVap0x/YMwzBefvllo3HjxileBzAGDx7sVOepp54yevbs6bLNo0ePGoCxe/fuVPXBMAyje/fuxjPPPJP4+U5f6zsBjBdeeMGprF69esaLL76Y6j4OGDDAaNu2rVMfS5YsaSQkJCSWtWvXzmjfvr1hGIZx/Phxw83NzTh9+rRTO82aNTNGjRqVqn6LiNwLNNMlIpIDHT58mPj4eOrVq5dYljdvXsqXL++yfrt27YiJiaF06dL06dOHhQsXkpCQkOI1bDYbb731FlWrViVv3rz4+fmxYsUKTpw44VSvWrVqTp+LFCnChQsXANizZw/FihWjXLlyLq+xd+9epk+fjp+fX+KvFi1aYLfbOXr06B2/Dq4cPHiQBg0aOJU1aNCAsLAwbDZbutp0pXbt2k6fX3zxRebOnUv16tUZMWIEmzdvzrBr3ZDS1zo1HnjggSSfk5vpAvjss8+oVasWBQoUwM/Pj6+++irJ33/lypUTZy5v79P+/fux2WyUK1fO6e94/fr1HD58ONX9FhHJ7rSRhoiIULx4cf755x9Wr17NqlWr6N+/PxMmTGD9+vXJbgAxYcIEPvroIz788EOqVq2Kr68vgwcPJj4+3qne7eebTCbsdjsA3t7eKfYrKiqKfv36JT5fdqsSJUqk5RbvOl9fX6fPjz32GMePH2fZsmWsWrWKZs2aMWDAACZOnJhh10zpa53R5s6dy7Bhw5g0aRIPPPAA/v7+TJgwgW3btqW6T1FRUbi5ubFz506nYAbg5+eXKf0WEckKCl0iIjlQmTJl8PDwYNu2bYnh5OrVq4SGhtK4cWOX53h7e/PUU0/x1FNPMWDAACpUqMD+/fupWbMmnp6eSWaBNm3axDPPPEOXLl0AxzM+oaGhVKpUKdX9rFatGqdOnSI0NNTlbFfNmjU5cOAA9913X6rbvJWrflesWJFNmzY5lW3atIly5col+cY/oxUoUIDu3bvTvXt3GjZsyPDhw5k4cSKenp4AGTrTlh5bt26lW7duTp9r1Kjhsu6mTZt48MEH6d+/f2JZWmenatSogc1m48KFCzRs2DB9nRYRuQdoeaGISA7k5+dHr169GD58OGvWrOHPP/+kR48emM2u/9mfPn0633zzDX/++SdHjhxh1qxZeHt7U7JkScCxO9+GDRs4ffo0ly5dAqBs2bKsWrWKzZs3c/DgQfr168f58+fT1M/GjRvTqFEj2rZty6pVqzh69Ci//vory5cvB2DkyJFs3ryZgQMHsmfPHsLCwvj5559TtZFGcv0eOnQov/32G2+99RahoaHMmDGDTz/9lGHDhqWp72k1evRofv75Zw4dOsRff/3FkiVLqFixIgAFCxbE29ub5cuXc/78eSIiIjK1L8mZN28e3377LaGhoYwZM4Y//vgj2a912bJl2bFjBytWrCA0NJQ33niD7du3p+l65cqVo3PnznTr1o0FCxZw9OhR/vjjD8aPH8/SpUsz4pZERLIFhS4RkRxqwoQJNGzYkKeeeopHHnmEhx56iFq1armsGxQUxNdff02DBg2oVq0aq1evZvHixeTLlw+AcePGcezYMcqUKZP4TqrXX3+dmjVr0qJFC5o0aULhwoWdtjBPrfnz51OnTh06duxIpUqVGDFiROKMT7Vq1Vi/fj2hoaE0bNiQGjVqMHr0aIKDg1PVtqt+16xZkx9//JG5c+dSpUoVRo8ezbhx4+jRo0ea+54Wnp6ejBo1imrVqtGoUSPc3NyYO3cuAO7u7nz88cdMmTKF4OBgnnnmmUztS3LefPNN5s6dS7Vq1Zg5cybff/99sjOX/fr1o02bNrRv35569epx+fJlp1mv1Jo2bRrdunVj6NChlC9fnlatWrF9+/Zsv3xURCQtTIZhGFndCREREclaJpOJhQsXpis4i4hIyjTTJSIiIiIikokUukRE5J60ceNGp23Gb/+VXdpMr5T6sXHjxjS1NXv27GTbqly5cibdgYiI3KDlhSIick+KiYnh9OnTyR5Pz46HmdFmeh06dCjZY0WLFr3jdvu3unbtWrKbnHh4eCRumCIiIplDoUtERERERCQTaXmhiIiIiIhIJlLoEhERERERyUQKXSIiIiIiIplIoUtERERERCQTKXSJiIiIiIhkIoUuERERERGRTKTQJSIiIiIikon+H4lpe1hvLNwMAAAAAElFTkSuQmCC",
      "text/plain": [
       "<Figure size 1000x1000 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "def get_closest_stability(avg_df, reference_stability):\n",
    "    # get closest stability to reference stability\n",
    "    avg_df['distance_to_reference'] = abs(avg_df['stability'] - reference_stability)\n",
    "    avg_df = avg_df.sort_values(by=['distance_to_reference'])\n",
    "    closest_stability = avg_df.iloc[0]\n",
    "    return closest_stability\n",
    "\n",
    "\n",
    "# plot distance in null vs distance in range\n",
    "X = 'distance_to_first_in_plane'\n",
    "Y = 'distance_to_first_null_space'\n",
    "\n",
    "plt.figure(figsize=(10,10))\n",
    "\n",
    "# lwf\n",
    "df = combined_df_lwf[combined_df_lwf['dense_size'] == DENSE_SIZE]\n",
    "df = df[df['lwf_temperature'] == LWF_TEMP]\n",
    "df = df.groupby(['lwf_alpha', 'lwf_temperature']).mean(numeric_only=True).reset_index()\n",
    "df = df.sort_values(by=[X])\n",
    "\n",
    "# regression line\n",
    "slope, intercept, r_value, p_value, std_err = linregress(df[X], df[Y])\n",
    "plt.plot(df[X], slope * df[X] + intercept, '--', c='blue')\n",
    "\n",
    "\n",
    "plt.scatter(df[X],\n",
    "            df[Y],\n",
    "            c=df['lwf_alpha'],\n",
    "            cmap='Blues',\n",
    "            norm=LogNorm(),\n",
    "            s=MARKER_SIZE,\n",
    "            edgecolors='k',\n",
    "            linewidths=1,\n",
    "            label='LwF',\n",
    "            marker='D'\n",
    "            )\n",
    "\n",
    "# ewc\n",
    "df = combined_df_ewc[combined_df_ewc['dense_size'] == DENSE_SIZE]\n",
    "df = df.groupby(['ewc_lambda']).mean(numeric_only=True).reset_index()\n",
    "slope, intercept, r_value, p_value, std_err = linregress(df[X], df[Y])\n",
    "plt.plot(df[X], slope * df[X] + intercept, '--', c='green')\n",
    "log_norm_ewc = LogNorm(vmin=df['ewc_lambda'].min(), vmax=df['ewc_lambda'].max())\n",
    "plt.scatter(x=df[X],\n",
    "            y=df[Y],\n",
    "            c=df['ewc_lambda'],\n",
    "            cmap='Greens',\n",
    "            norm=log_norm_ewc,\n",
    "            s=MARKER_SIZE,\n",
    "            edgecolors='k',\n",
    "            linewidths=1,\n",
    "            marker='s',\n",
    "            label='EWC',\n",
    "            )\n",
    "\n",
    "# SI\n",
    "df = combined_df_si[combined_df_si['dense_size'] == DENSE_SIZE]\n",
    "df = df[df['si_eps'] == SI_EPS]\n",
    "df = df.groupby(['si_lambda']).mean(numeric_only=True).reset_index()\n",
    "df = df.sort_values(by=[X])\n",
    "\n",
    "# regression line\n",
    "slope, intercept, r_value, p_value, std_err = linregress(df[X], df[Y])\n",
    "plt.plot(df[X], slope * df[X] + intercept, '--', c='red')\n",
    "\n",
    "norm = LogNorm(vmin=df['si_lambda'].min(), vmax=df['si_lambda'].max())\n",
    "plt.scatter(df[X],\n",
    "            df[Y],\n",
    "            c=df['si_lambda'],\n",
    "            cmap='Reds',\n",
    "            norm=norm,\n",
    "            s=MARKER_SIZE,\n",
    "            edgecolors='k',\n",
    "            linewidths=1,\n",
    "            label='SI',\n",
    "            )\n",
    "\n",
    "# gem\n",
    "df = combined_df_gem[combined_df_gem['dense_size'] == DENSE_SIZE]\n",
    "df = df[df['gem_memory_strength'] >= 0]\n",
    "df = df.groupby(['gem_patterns_per_experience', 'gem_memory_strength']).mean(numeric_only=True).reset_index()\n",
    "df = df.sort_values(by=[X])\n",
    "\n",
    "# regression line\n",
    "slope, intercept, r_value, p_value, std_err = linregress(df[X], df[Y])\n",
    "plt.plot(df[X], slope * df[X] + intercept, '--', c='k')\n",
    "\n",
    "\n",
    "plt.scatter(df[X],\n",
    "            df[Y],\n",
    "            c=df['gem_memory_strength'],\n",
    "            cmap='Greys',\n",
    "            norm=LogNorm(),\n",
    "            s=MARKER_SIZE,\n",
    "            edgecolors='k',\n",
    "            linewidths=1,\n",
    "            label='GEM',\n",
    "            marker='*'\n",
    "            )\n",
    "\n",
    "# replay\n",
    "df = combined_df_replay[combined_df_replay['dense_size'] == DENSE_SIZE]\n",
    "df = df.groupby(['replay_mem_size']).mean(numeric_only=True).reset_index()\n",
    "df = df.sort_values(by=[X])\n",
    "# regression line\n",
    "slope, intercept, r_value, p_value, std_err = linregress(df[X], df[Y])\n",
    "plt.plot(df[X], slope * df[X] + intercept, '--', c='purple')\n",
    "\n",
    "plt.scatter(df[X],\n",
    "            df[Y],\n",
    "            c=df['replay_mem_size'],\n",
    "            cmap='Purples',\n",
    "            norm=LogNorm(),\n",
    "            s=MARKER_SIZE,\n",
    "            edgecolors='k',\n",
    "            linewidths=1,\n",
    "            label='Replay',\n",
    "            marker='^'\n",
    "            )\n",
    "\n",
    "\n",
    "# plt.legend()\n",
    "\n",
    "# axis labels\n",
    "plt.xlabel(X)\n",
    "plt.ylabel(Y)\n",
    "# equal axes\n",
    "# plt.gca().set_aspect('equal', adjustable='box')\n",
    "# despine\n",
    "sns.despine()\n",
    "\n",
    "# increase size of tick font\n",
    "plt.xticks(fontsize=18)\n",
    "plt.yticks(fontsize=18)\n",
    "\n",
    "# x = np.linspace(0, 270, 100)\n",
    "# y = np.sqrt(10/118) * x\n",
    "# plt.plot(x, y, linestyle='dotted', c='k', alpha=.5)\n",
    "\n",
    "# plt.savefig(f'{plotdir}/distance_to_first_in_plane_vs_null_space.png', bbox_inches='tight')\n",
    "\n",
    "REFCOLORS = ['red', 'blue', 'green', 'purple']\n",
    "for REFERENCE_STABILITY, REF_COLOR in zip(REFERENCE_STABILITIES, REFCOLORS):\n",
    "\n",
    "    # EWC\n",
    "    df = combined_df_ewc[combined_df_ewc['dense_size'] == DENSE_SIZE]\n",
    "    df = df.groupby(['ewc_lambda']).mean(numeric_only=True).reset_index()\n",
    "    closest_stability_ewc = get_closest_stability(df, REFERENCE_STABILITY)\n",
    "    ewc_ref_runs = combined_df_ewc[(combined_df_ewc['dense_size'] == DENSE_SIZE) & (combined_df_ewc['ewc_lambda'] == closest_stability_ewc['ewc_lambda'])]\n",
    "\n",
    "    # SI\n",
    "    df = combined_df_si[combined_df_si['dense_size'] == DENSE_SIZE]\n",
    "    df = df[df['si_eps'] == SI_EPS]\n",
    "    df = df.groupby(['si_lambda']).mean(numeric_only=True).reset_index()\n",
    "    closest_stability_si = get_closest_stability(df, REFERENCE_STABILITY)\n",
    "    si_ref_runs = combined_df_si[(combined_df_si['dense_size'] == DENSE_SIZE) & (combined_df_si['si_eps'] == SI_EPS) & (combined_df_si['si_lambda'] == closest_stability_si['si_lambda'])]\n",
    "\n",
    "    # Replay\n",
    "    df = combined_df_replay[combined_df_replay['dense_size'] == DENSE_SIZE]\n",
    "    df = df.groupby(['replay_mem_size']).mean(numeric_only=True).reset_index()\n",
    "    closest_stability_replay = get_closest_stability(df, REFERENCE_STABILITY)\n",
    "    replay_ref_runs = combined_df_replay[(combined_df_replay['dense_size'] == DENSE_SIZE) & (combined_df_replay['replay_mem_size'] == closest_stability_replay['replay_mem_size'])]\n",
    "\n",
    "    # gem\n",
    "    df = combined_df_gem[combined_df_gem['dense_size'] == DENSE_SIZE]\n",
    "    df = df[df['gem_memory_strength'] >= 0]\n",
    "    df = df.groupby(['gem_patterns_per_experience', 'gem_memory_strength']).mean(numeric_only=True).reset_index()\n",
    "    closest_stability_gem = get_closest_stability(df, REFERENCE_STABILITY)\n",
    "    gem_ref_runs = combined_df_gem[(combined_df_gem['dense_size'] == DENSE_SIZE) & (combined_df_gem['gem_memory_strength'] == closest_stability_gem['gem_memory_strength']) & (combined_df_gem['gem_patterns_per_experience'] == closest_stability_gem['gem_patterns_per_experience'])]\n",
    "\n",
    "    # lwf\n",
    "    df = combined_df_lwf[combined_df_lwf['dense_size'] == DENSE_SIZE]\n",
    "    df = df[df['lwf_temperature'] == LWF_TEMP]\n",
    "    df = df.groupby(['lwf_alpha', 'lwf_temperature']).mean(numeric_only=True).reset_index()\n",
    "    closest_stability_lwf = get_closest_stability(df, REFERENCE_STABILITY)\n",
    "    lwf_ref_runs = combined_df_lwf[(combined_df_lwf['dense_size'] == DENSE_SIZE) & (combined_df_lwf['lwf_alpha'] == closest_stability_lwf['lwf_alpha']) & (combined_df_lwf['lwf_temperature'] == closest_stability_lwf['lwf_temperature'])]\n",
    "\n",
    "    # force a copy of all dfs\n",
    "    ewc_ref_runs = ewc_ref_runs.copy()\n",
    "    si_ref_runs = si_ref_runs.copy()\n",
    "    replay_ref_runs = replay_ref_runs.copy()\n",
    "    gem_ref_runs = gem_ref_runs.copy()\n",
    "    lwf_ref_runs = lwf_ref_runs.copy()\n",
    "\n",
    "    # concatenate all runs after adding algorithm column\n",
    "    ewc_ref_runs['algorithm'] = 'EWC'\n",
    "    si_ref_runs['algorithm'] = 'SI'\n",
    "    replay_ref_runs['algorithm'] = 'Replay'\n",
    "    gem_ref_runs['algorithm'] = 'GEM'\n",
    "    lwf_ref_runs['algorithm'] = 'LwF'\n",
    "\n",
    "    # concatenate all runs\n",
    "    ref_runs = pd.concat([ewc_ref_runs, si_ref_runs, replay_ref_runs, gem_ref_runs, lwf_ref_runs])\n",
    "    # select algorithm and distances\n",
    "    ref_runs = ref_runs[['algorithm', 'distance_to_first_null_space', 'distance_to_first_in_plane']]\n",
    "    # avg over algorithm\n",
    "    ref_runs = ref_runs.groupby(['algorithm']).mean(numeric_only=True).reset_index()\n",
    "    \n",
    "    # ewc\n",
    "    plt.scatter(\n",
    "        ref_runs[ref_runs['algorithm'] == 'EWC'][X],\n",
    "        ref_runs[ref_runs['algorithm'] == 'EWC'][Y],\n",
    "        s=MARKER_SIZE * 4,\n",
    "        marker='o',\n",
    "        edgecolors=REF_COLOR,\n",
    "        linewidths=4,\n",
    "        facecolors='none',\n",
    "        label='EWC'\n",
    "    )\n",
    "\n",
    "    # SI\n",
    "    plt.scatter(\n",
    "        ref_runs[ref_runs['algorithm'] == 'SI'][X],\n",
    "        ref_runs[ref_runs['algorithm'] == 'SI'][Y],\n",
    "        s=MARKER_SIZE * 4,\n",
    "        marker='o',\n",
    "        edgecolors=REF_COLOR,\n",
    "        linewidths=4,\n",
    "        facecolors='none',\n",
    "        label='SI'\n",
    "    )\n",
    "\n",
    "    # Replay\n",
    "    plt.scatter(\n",
    "        ref_runs[ref_runs['algorithm'] == 'Replay'][X],\n",
    "        ref_runs[ref_runs['algorithm'] == 'Replay'][Y],\n",
    "        s=MARKER_SIZE * 4,\n",
    "        marker='o',\n",
    "        edgecolors=REF_COLOR,\n",
    "        linewidths=4,\n",
    "        facecolors='none',\n",
    "        label='Replay'\n",
    "    )\n",
    "    \n",
    "    # GEM\n",
    "    plt.scatter(\n",
    "        ref_runs[ref_runs['algorithm'] == 'GEM'][X],\n",
    "        ref_runs[ref_runs['algorithm'] == 'GEM'][Y],\n",
    "        s=(MARKER_SIZE + 10) * 4,\n",
    "        marker='o',\n",
    "        edgecolors=REF_COLOR,\n",
    "        linewidths=4,\n",
    "        facecolors='none',\n",
    "        label='GEM'\n",
    "    )\n",
    "\n",
    "    # LwF\n",
    "    plt.scatter(\n",
    "        ref_runs[ref_runs['algorithm'] == 'LwF'][X],\n",
    "        ref_runs[ref_runs['algorithm'] == 'LwF'][Y],\n",
    "        s=MARKER_SIZE * 4,\n",
    "        marker='o',\n",
    "        edgecolors=REF_COLOR,\n",
    "        linewidths=4,\n",
    "        facecolors='none',\n",
    "        label='LwF'\n",
    "    )\n",
    "\n",
    "    # add a dotted line through the origin with slope sqrt(10/118)\n",
    "    # x = np.linspace(0, 250, 100)\n",
    "    # y = np.sqrt(10/118) * x\n",
    "    # plt.plot(x, y, 'k--', alpha=.5)\n",
    "\n",
    "    # set text sizes\n",
    "    plt.xticks(fontsize=14)\n",
    "    plt.yticks(fontsize=14)\n",
    "    # plt.legend(fontsize=14)\n",
    "\n",
    "    plt.title(f'Distance to first in plane vs null space - Stability = {REFERENCE_STABILITY}')\n",
    "    plt.xlabel(X)\n",
    "    plt.ylabel(Y)\n",
    "    sns.despine()\n",
    "    # equal axes\n",
    "    # plt.gca().set_aspect('equal', adjustable='box')\n",
    "    # plt.xlim([0, 120])\n",
    "    # plt.ylim([0, 280])\n",
    "plt.savefig(f'{plotdir}/distance_to_first_in_plane_vs_null_space_references.png', bbox_inches='tight')\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAA2gAAANLCAYAAAA0Lr6lAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjcuMSwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy/bCgiHAAAACXBIWXMAAA9hAAAPYQGoP6dpAADGpElEQVR4nOzdZ3RUZf/18T3pgZCQ0Iv03qSqVEUEBMsNAgICIl0RpIXeEQi9SVcQkCKIoKD0LkjvvYbeAyQkIXXmecGT+TMSIBmSzCT5ftbKWsl1zpnsuO4b2Pmdc43BZDKZBAAAAACwOQdbBwAAAAAAPEVBAwAAAAA7QUEDAAAAADtBQQMAAAAAO0FBAwAAAAA7QUEDAAAAADtBQQMAAAAAO0FBAwAAAAA7QUFLJCaTSUFBQeJ9wAEAAADEFQUtkTx+/FheXl56/PixraMAAAAASCYoaAAAAABgJyhoAAAAAGAnKGgAAAAAYCcoaAAAAABgJyhoAAAAAGAnKGgAAAAAYCcoaAAAAABgJyhoAAAAAGAnKGgAAAAAYCcoaAAAAABgJyhoAAAAAGAnKGgAAAAAYCcoaAAAAABgJyhoAAAAAGAnKGgAAAAAYCcoaAAAAABgJyhoAAAAAGAnKGgAAAAAYCcoaAAAAABgJyhoAAAAAGAnKGgAAAAAYCcoaAAAAABgJyhoAAAAAGAnKGgAAAAAYCcoaAAAAABgJyhoAAAAAGAnKGgAAAAAkkzDhg3l6uqqqKgoW0exSxQ0AAAAAElmzZo1ioiI0Ny5c20dxS5R0AAAAAAkiW3btunJkyeSpNmzZyfq9woODlZERESifo/EQEEDAAAAkCTGjRsnBwcH5cuXT0ePHpXRaEyU72M0GpUta3blz18gUV4/MTnZOgAAAACAlOf333/X5cuXLda2bNmiHDlyqECBArp06ZLat2+vokWLmo8bDAZ99dVX8vHxea3vPWDAAAWHPFZwyGMtWbJETZs2fa3XS0oGk8lksnWIlCgoKEheXl4KDAyUp6enreMAAAAASSY0NFRp06aN9di7776r3Llza9myZQoLC3vueLNmzbRw4UKrv7fRaJRH2nRKY8iisKggefg46/btW1a/XlJjggYAAAAgQaVJk0bTp0/Xd999p+joaJUrV065c+eWg4OD0qRJI4PBoM8++0zh4eEKDQ3Vv//+q8DAQFWqVOm1n00bMGCAnoSFqsIbdfUk8qH2316crKZoTNASCRM0AAAApHb+/v6qUqWKbt68qUKFCumtt96Sk9P/zYiuX7+uHTt2KDo6Wn5+furZs+drfb+Y6VlaQ1a9l6uzjKZorff3S1ZTNDYJAQAAAJAo8ubNq2vXrunjjz/WuXPndOPGDYvju3btkpOTk44fP/7a5Uz6v+lZsYx1JEkOBkcVy1Bbd+7c1pIlS1779ZMCBQ0AAABAonFwcJCXl5ckKUuWLAoMDNTly5dlMpmULVs2GY1Gi41CrGU0GjVp4mRldM+nzGn+b/fGNzzLKq1zBnXr1v21v0dSoKABAAAASFRr165VlixZdO3aNa1atUrbtm3Tli1blDVrVoWFhWnjxo2v/T3+Oz2LkdymaDyDlkh4Bg0AAACQLl68qAIFCsjFxUUREREqWLCgPvroI02ZMkVOTk6KiIhQ7dq1tW7dOqu/x3+fPXvueDJ6Fo0JGgAAAIBE4+fnJ0mKjIxU165dde7cOU2cOFF79uyRu7u7JOmff/55re/xoulZjOQ0RaOgAQAAAEg0zs7Oypgxo3bs2KGJEyea1ytUqKC7d+/qgw8+eOF7psXFi549+6/k8iwaBQ0AAABAopkxY4bu3bunKlWqPHfMxcVFGzdu1N27d61+/VdNz2Iklykaz6AlEp5BAwAAABLXq549e+78ZPAsmtOrTwEAAAAA+xMzPcuXsbDuhl6I0zXZ0hbThTv/aMmSJWratGkiJ4w/JmiJhAkaAAAAkLhKlSql48ePW3Vt3bp19ffffydwotfHBA0AAABAsrRnzx7t2bPHqmurVauWwGkSBhO0RMIEDQAAAEB8sYsjAAAAANgJChoAAAAA2AkKGgAAAADYCQoaAAAAANgJChoAAAAA2AkKGgAAAADYCQoaAAAAANgJChoAAAAA2AkKGgAAAADYCQoaAAAAANgJChoAAAAA2AkKGgAAAADYCQoaAAAAANgJChoAAAAA2AkKGgAAAADYCQoaAAAAANgJChoAAAAA2AkKGgAAAADYCQoaAAAAANgJChoAAAAA2AkKGgAAAADYCQoaAAAAANgJChoAAAAA2AkKGgAAAADYCQoaAAAAANgJChoAAAAA2AkKGgAAAADYCQoaAAAAANgJChoAAAAA2Am7KGgBAQH6+eef1bx5cxUrVkxp06aVq6urcubMqXr16mnlypUvvHbevHkyGAyv/Ni0adNLM1y8eFEdOnRQ3rx55ebmpkyZMql27dr6/fffE/rHBQAAAIBYOdk6gCRlzZpVUVFR5q/d3Nzk7OysGzdu6MaNG/rzzz9Vp04dLV++XGnSpIn1NRwcHJQpU6YXfg9XV9cXHluzZo0aNWqk0NBQSZKnp6cePHigDRs2aMOGDWrVqpXmzJkjg8Fg5U8IAAAAAK9mFxO0qKgovfXWW5o+fbouXryoJ0+eKDg4WP7+/mrTpo0kae3aterQocMLX+ONN97Q7du3X/hRtWrVWK/z9/fX559/rtDQUFWuXFlnz55VYGCgAgMDNWjQIEnSzz//rLFjxyb8Dw4AAAAAzzCYTCaTrUNs3bpV1atXf+Hxr7/+WrNmzZIkXb16VW+88Yb52Lx589SqVSvlzp1bly9fjvf3btGihRYuXKisWbPq9OnTSp8+vcXxDh06aPbs2fL09NTly5fl7e0dp9cNCgqSl5eXAgMD5enpGe9cAAAAAFIfu5igvaycSTJP0STpwIEDCfZ9Q0JCzM+YffPNN8+VM0nq27evpKeF648//kiw7w0AAAAA/2UXBe1V3NzczJ9HR0cn2Ovu3LlTT548kSTVqVMn1nPy5MmjokWLSpI2bNiQYN8bAAAAAP4rWRS0bdu2mT8vWbJkrOfcu3dP5cqVk4eHh9zd3ZUvXz41b97c4tr/OnHihPnzEiVKvPC8mGMnT56MX3AAAAAAiAe7L2iPHj2Sn5+fJKlq1aoqXLhwrOeFhobq0KFDcnFxkdFolL+/vxYtWqTq1aurdevWFrtExrh586YkydvbW+7u7i/MkCNHDovzYxMeHq6goCCLDwAAAACID7suaEajUS1atNCtW7fk5uamqVOnPndO9uzZNXjwYB09elRhYWF68OCBQkNDtWvXLn3wwQeSnu7C2K1bt+euffz4sSS9cOv+GDHHY86PjZ+fn7y8vMwfz25kAgAAAABxYdcFrUuXLvrrr78kSdOmTVOpUqWeO6dWrVoaMmSISpUqZX6vM0dHR1WqVEnr16/X//73P0nS9OnTdf78+UTL2rdvX/P2/IGBgbp27VqifS8AAIDkqGfPnho6dKitYyQ7Q4cOfeFjPkh57Lag+fr6midmEydOVOvWreP9Gg4ODho3bpykp9O41atXWxxPly6dJJnfoPpFYo7HnB8bV1dXeXp6WnwAAADgqaioKI0fP17Dhw+X0Wi0dZxkZeLEiTpx4oR27Nhh6yhIAnZZ0Hr16qXx48dLksaNG6euXbta/VoFChRQxowZJUmXLl2yOJY9e3ZJ0sOHD827Ocbmxo0bFucDQGqwa9cu5c+fXxcvXrR1FAApwI8//iiTyaSoqCj9+eefto6TbJw9e1aBgYGSpLFjx9o4DZKC3RW0nj17mv/HN2bMGPXo0SPRvtezOzc+u6Pjf8UcK168eKJlAQB707JlS126dEktW7a0dRQAKcBPP/1k/nzKlCk2TJK8xGyWJ0lbt261YRIkFbsqaL6+vuZbEseMGaOePXu+9mtevHhR9+/flyTlzZvX4liVKlXMuzeuW7cu1uuvXLmi06dPS3r6vBsApAabN2/WxYuXlCljDu3evSdRn+EFkPIZjUYdOXLE/PW///5ruzB2zGg0KioqyuIjZj8GSQoJCdHevXufOwcpi90UNF9fX4vbGuNSzkwm0yuPx7yOg4ODPv74Y4vjadOmVYMGDSRJM2bMMI+PnzV69GhJT58/q1ev3iszAUBK0KFDB6XzSK+unafIwcFRrVq1snUkAMnEmjVr5ObmJmdnZ/OHq6urxXNnERERFsednZ3l7u6u3bt32zC5bQUFBcnJyem5/y4BAQEW573zzjvPnfPdd9/ZKDUSg10UtGefOZswYUKcb2u8cuWK3nrrLc2aNUuXLl0yFzaj0ag9e/aoTp06WrlypaSn/9iI7T3Uhg0bprRp0+rWrVv65JNPzL8lDgkJ0bBhwzRz5kxJ0oABA+Tt7f3aPysA2LuY6VmdD1vKxyerqr/bkCkagDgrXLiwHBwcXjnl+e9xZ2fn5+52Sk08PT1Vu3bteF+XLl06dejQIRESwVYMpleNoRLZ1atXlTt3bklPp1yZMmV66fm+vr7y9fWVJF2+fNni/8iurq5Kly6dHj9+rPDwcPN6q1atNHv2bDk5OcX6mmvWrFGjRo3MuzV6eXkpODhY0dHR5uvnzJkjg8EQ558rKChIXl5eCgwMZEdHAMlKgQIFdPfOA40Y9rucnV0V9Pih+g/6TG+/XUE7d+60dTwAyUBwcLDef/997d+/P07nv/fee1q7dq3c3NwSOZn9mz17tjp27Gj+d+jL1KhRQ2vWrJGLi0sSJENSsfkE7dlxt9Fo1J07d176ERwcbD4/S5Ys+uGHH/TFF1+oWLFi8vT01KNHj+Ts7KwiRYqodevW2rlzp+bOnfvCciZJdevW1bFjx9SuXTvlyZNHYWFh8vb2Vs2aNbV8+XLNnTs3XuUMAJKrZ6dnzs5P31vSM503UzQA8eLh4aF9+/Zp5MiRL/03lIODg3744Qdt3bqVcvb/tW/fXpcuXVLOnDlfeI6Dg4OmT5+uTZs2Uc5SIJtP0FIqJmgAkqP/Ts9iMEUDYK2qVau+8M+Njz76yGITDPyfbt26adKkSbEe8/b21oMHD5I2EJKMzSdoAAD7ENv0LAZTNADW2rt37wuPbdu2LemCJDPLly9/4bGHDx/yHpUpGAUNACDp/3ZurFLp01iP16zxBTs6AoiXNWvWKDIy0vy1s7OzxWMnISEhOnDggC2i2bUHDx7o+vXr5q9dXFxUtGhRi3OefX80pCwUNADAS6dnMdKl89b77zViigYgziZOnGj+vHjx4rp9+7auXLmifPnymdeHDRtmi2h2bezYsebPS5YsqTt37ujUqVMaOHCg+Zm+VatW2SoeEhkFDQDwyulZjA/eb8oUDUCcHTp0SJLUp08fnThxQj4+PsqePbsuXrwoZ2dnSdLWrVttGdEu7d+/XwaDQf369dOxY8eUPn16SU/L7IEDB+Tl5cUzaCnYi7c2BACkCjHTs88bdnnh9CxGzBRt05Zfdf78eRUsWDCJUgJIjn7//XdlzJhRJUqUsFg/ceKE+dbHtGnT2iKaXfvrr78UGhoqHx+f546VLVtWDx480N27d22QDEmBCRoApHJxnZ7FYIoGIK7ee++958qZZPn81J07dxQUFJSUseyem5tbrOUshoODg7JmzZqEiZCU2GY/kbDNPoDk4OLFiypQoICcnJzl7uYR5+tCQoNkMpkUFRUpBwd+1wfgxX766SedPn3aYm3WrFkKCQkxf129enWVKVPG4py2bds+tzEGkBpQ0BIJBQ1AchAaGqovv/xSAQEB8b42V65cmj9/fiKkApBSGI1GOTs7y2g0xvvaOnXqaM2aNYmQCrBvFLREQkEDACSUkydPqm7duvr1119VsWJFW8cB4mX9+vWqV6+ewsLC4nxN1apVtWHDBrm5uSViMsA+cV8KAAB2rmXLlrp69aq++uorW0cB4q127dq6d++eypUr98pzHRwcNHHiRO3YsYNyhlSLggYAgB07duyYDh8+rEKFCun8+fPatm2brSMB8ebh4aEDBw6oe/fuLzzH1dVVZ86cUdeuXZMuGGCHKGgAANix1q1by93dXZs2bpaPj486dOhg60iA1V52m2NERIRy586dhGkA+0RBAwDATsVMz7p1667s2bOrb59+TNGQrK1cufKFx0wmk2bNmpWEaQD7xCYhiYRNQgAAr6t8+fI6c+aMLvtfkbe3t0JDQ5U3Xx5lyJBBZ8+etXU8IF7u3r2rLFmymL8uWrSoOnbsqG7duikqKkqS9Oabb+rIkSM2SgjYByZoAADYoWenZ97e3pKkNGnSMEVDsjV27Fjz5z179tSpU6fUqVMnXblyRXnz5pUkHT9+3Kot+YGUhIIGAIAdinn2rMt3XSzW27dvz7NoSJYyZ86sbNmy6d9//9WYMWPM69mzZ9elS5f09ddfK1OmTDZMCNgHbnFMJNziCACw1rFjx1SmTBn169dfQwYPee74pEmT1LOXr7Zs2aL33nsvyfMBABIPBS2RUNAAANb677Nn/8WzaACQcnGLIwAAdiS2Z8/+i2fRYK/CwsIsNv0AEH9M0BIJEzQAgDVeNT2LwRQN9ui7777TDz/8oO+//14DBgywdRwgWWKCBgCAnYjL9CwGUzTYo99//12StHDhQhsnAZIvJmiJhAkaACC+4jo9i8EUDfbk/v37FrswPnnyRG5ubjZMBCRPTrYOAAAAnj67c+jQITk4OKhQ4YJxvi44OFgBAQG6fv26cubMmYgJgZd7dut8SZo+fbq6d+9uozRA8sUELZEwQQMAxIfRaFTbtm1148aNeF+bKVMmzZ07Vy4uLomQDHjel19+qR07dlis3bp1SxEREeav3dzclCVLFotzPvroI02bNi1JMgLJFRM0AADsgIODg+bOnWvrGECcnDp1SleuXHnpOWFhYc+dc+7cucSMBaQIbBICAACAeNm3b5/atWsXr2t8fX21cePGREoEpBzc4phIuMURAACkdOvXr1e9evUUFhb2wnM8PDy0YcMGVaxYMQmTAckXBS2RUNAAAEBqcPfu3eeeNXtWSEiI0qRJk4SJgOSNWxwBAABgtTlz5rz0+G+//ZZESYCUgYIGAAAAqy1YsMDia0dHR4uvZ8yYkZRxgGSPggYAAACrREREWLxJetOmTRUWFqaPPvrIvHbgwAEZjUZbxAOSJQoaAAAArLJo0SKZTCY5OztrxYoVWrx4sZycnPTXX39p3rx5cnR0VHR0tDZv3mzrqECyQUEDAACAVWrWrKlmzZrp9u3bql+/vsWxli1b6urVq/riiy9Urlw5GyUEkh92cUwk7OIIAAAAIL6YoAEAAACAnaCgAQAAAICdoKABAAAAgJ2goAEAAACAnaCgAQAAAICdoKABAAAAgJ2goAEAAACAnaCgAQAAIEUzGo3y8vJSq1atbB0FeCUKGgAAAFK0JUuWKCgoSCtXrrR1FOCVKGgAAABI0aZNmyZJCgwM1P79+22cBng5ChoAAABSLKPRqCNHjuizzz6Tk5OTRo0aZetIwEs52ToAAAAAkBDCwsK0atUqmUwm89qJEyf05MkTtW7dWiEhIdq0aZOWLl1qcV2uXLlUsWLFpI4LxIqCBgAAgBShffv2+uWXX55b9/Ly0gcffKCAgACtX79eTZo0ee6ckJAQpUmTJiliAi/FLY4AAABIESZNmqSiRYtKkgoXLqz169dr//79OnnypFxdXdW8eXMdOXJE+/fv1+jRo+Xs7CxnZ2cNHjyYcga7QUEDAABAiuDj46NTp07J19dXFy9eVKtWrRQWFqYcOXJIkhwcHFS8eHGtWLFCffr0kaenp/bs2aMhQ4bYNjjwDIPp2Zt0kWCCgoLk5eWlwMBAeXp62joOAABAqrJ7927VqFFDXl5eunXrlnn9zz//VL169VSlShVt3rxZLi4uNkwJPI8JGgAAAFKcihUrysnJSVWqVJEkbdu2TTdu3NDbb78tg8GgggULUs5glyhoAAAASHF27typx48fq06dOmrdurWqV6+uYsWKaffu3apYsaLWrFlj64hArLjFMZFwiyMAAIDt/O9//9OqVauUNWtW3bt3T40aNdI///yjGzduKHv27Lp9+7b8/f2VK1cuW0cFLDBBAwAAQIqzfft2SVJwcLBWr16tJUuW6OrVq2ratKlu374to9HIm1bDLlHQAAAAkOKYTCaVKVNGN27cUJ06dSQ93cVx8eLFWr58udKlS6egoCAbpwSexy2OiYRbHAEAAADEFxM0AAAAALATFDQAAAAAsBMUNAAAAACwExQ0AAAAALATFDQAAAAAsBMUNAAAAACwExQ0AAAAALATFDQAAAAAsBMUNAAAAACwExQ0AAAAALATFDQAAAAAsBMUNAAAAACwExQ0AAAAALATFDQAAAAAsBMUNAAAAACwExQ0AAAAALATFDQAAAAAsBMUNAAAAACwExQ0AAAAALATFDQAAAAAsBMUNAAAAACwExQ0AAAAALATFDQAAAAAsBMUNAAAAACwExQ0AAAAALATFDQAAAAAsBMUNAAAAACwExQ0AAAAALATFDQAAAAAsBMUNAAAAACwExQ0AAAAALATFDQAAAAAsBMUNAAAAACwExQ0AAAAALATFDQAAAAAsBMUNAAAAACwExQ0AAAAALATFDQAAAAAsBMUNAAAAACwExQ0AAAAALATFDQAAAAAsBMUNAAAAACwExQ0AAAAALATFDQAAAAAsBMUNAAAAACwExQ0AAAAALATFDQAAAAAsBMUNAAAAACwExQ0AACQagUFBSkqKsrWMQDAjIIGAABSpYiICGXIkEH16tWzdRQAMKOgAQCAVGnWrFmKiorS5s2bbR0FAMwoaAAAIFWaM2eOJCksLIySBsBuUNAAAECqExUVpZMnT6pgwYJydHTUuHHjbB0JACRJTrYOAAAAkJiuX7+u6dOny2QymdfOnz+vqKgo5c+fX+Hh4dq2bZv69u1rcV2ePHnUoUOHpI4LIJUzmJ790woJJigoSF5eXgoMDJSnp6et4wAAkGp99dVXmj9//nPradKkUcOGDXXt2jVt3bo11mtDQkKUJk2axI4IAGYUtERCQQMAwD5ERUXp448/1vr16+Xp6alq1arJw8NDzs7OcnR0lPR0R0ej0agLFy7o4MGDcnZ21oIFC9S4cWMbpweQ2vAMGgAASNGcnJy0bt06/fjjj3ry5InWrVun27dvm8uZJBkMBu3Zs0cHDhxQgQIFdO3aNcoZAJugoAEAgFShbdu2unDhglxdXbV3716LYzdv3tTly5fVuHFjnTt3TpkzZ7ZRSgCpHQUNAACkGjlz5lRkZKSyZ88uo9Go8+fP68GDB8qaNasMBoOMRqOtIwJI5djFEQAApBqrV69WRESEMmfOrA0bNuj27dsyGAwqX768smbNqo0bN9o6IoBUjgkaAABINSZPnixJ2r9/v+7du6fvv/9eZcqU0f79+3X//n09evRIp0+ftnFKAKkZBQ0AAKQau3fvliT5+Pjo5MmTGjBggA4ePKghQ4YoOjpakjRq1ChbRgSQytlFQQsICNDPP/+s5s2bq1ixYkqbNq1cXV2VM2dO1atXTytXrnzlazx+/FhDhgxRyZIl5eHhIS8vL1WoUEHjx49XRETEK6+/c+eOevToocKFC8vd3V0+Pj6qWrWqfvrpJ/FOBAAApAyZMmVSw4YNdevWLRUuXNi8PnjwYB08eFA5c+ZUxowZbZgQQGpnF++D5uzsrKioKPPXbm5ucnR0VEhIiHmtTp06Wr58eaxvFnnlyhW99957unz5sqSnbzwZHR2t8PBwSVKZMmW0efNmeXt7x/r9Dx48qNq1aysgIECS5OHhobCwMHOm2rVra9WqVXJxcYnzz8T7oAEAAACIL7uYoEVFRemtt97S9OnTdfHiRT158kTBwcHy9/dXmzZtJElr165Vhw4dYr32k08+0eXLl5UtWzZt3LhRISEhCg0N1a+//qp06dLp8OHDat68eazfOzAwUB9//LECAgJUpEgR7d+/X48fP1ZISIimTp0qZ2dnrV+/Xl27dk3M/wQAAAAAYB8TtK1bt6p69eovPP71119r1qxZkqSrV6/qjTfeMB+bM2eO2rZtK0n6999/VbFiRYtrlyxZoi+++EKStGnTJtWoUcPi+MCBAzV8+HC5u7vr5MmTyps3r8VxPz8/9evXT46Ojjp16pQKFSoUp5+JCRoAAACA+LKLCdrLypkk8xRNkg4cOGBxbP78+ebX+G85k6QmTZqYS9eCBQueOx6z9ux5z+rcubM8PDwUHR2tRYsWveInAQAAAADr2UVBexU3Nzfz5zE7LElSaGiodu3aJenpM2qxMRgM+vDDDyVJGzZssDh29uxZXb169aXXe3h4qGrVqrFeDwAAAAAJKVkUtG3btpk/L1mypPnz06dPy2g0SpJKlCjxwutjjt2+fVsPHjwwr584ceK5c152/alTp+IXHAAAAADiwcnWAV7l0aNH8vPzkyRVrVrVYkvcmzdvmj/PkSPHC1/j2WM3b96Uj4+PVdcHBQUpODhYHh4ez50THh5u3jUy5lwAAAAAiA+7nqAZjUa1aNFCt27dkpubm6ZOnWpx/PHjx+bPY9t+P7Zjz17zutc/y8/PT15eXuaPZzcyAQAAAIC4sOuC1qVLF/3111+SpGnTpqlUqVI2TvRiffv2VWBgoPnj2rVrto4EAAAAIJmx21scfX19zROziRMnqnXr1s+dky5dOvPnoaGhL3ytZ489e81/r3/Rdvgvuv5Zrq6ucnV1fWEGAAAAAHgVu5yg9erVS+PHj5ckjRs37oVvEp09e3bz5zdu3Hjh6z177Nlr4nu9p6dnrM+fAQAAAEBCsLuC1rNnT40dO1aSNGbMGPXo0eOF5xYtWlQODk9/hGd3ZPyvmGNZs2Y1bxAiWe7cGJfrixUrFoefAAAAAACsY1cFzdfXV+PGjZP0tJz17NnzpeenSZNGlStXliStW7cu1nNMJpPWr18vSapVq5bFsUKFCilXrlwvvT4kJET//PNPrNcDAAAAQEKym4Lm6+trcVvjq8pZjJYtW0qStm7dqr179z53/LffftOlS5ckSV9++aXFMYPBYF779ddfdfny5eeunzZtmoKDg+Xo6KhmzZrF+ecBAAAAgPiyi4L27DNnEyZMeOltjf/VsmVLlSxZUiaTSQ0aNNDmzZslPd2i/7ffflO7du0kSXXq1FGNGjWeu97X11dZs2ZVaGioPvroIx08eFCSFBERoRkzZmjgwIGSpPbt26tQoUKv9XMCAAAAwMsYTCaTyZYBrl69qty5c0uSHBwclClTppee7+vrK19fX4u1y5cvq3r16uYJWJo0aWQ0GhUWFiZJKlOmjDZv3ixvb+9YX/PgwYOqXbu2AgICJD3dqTEsLEyRkZGSnt7auGrVqnjt0hgUFCQvLy8FBga+cHdIAAAAAHiWzbfZNxqNFp/fuXPnpecHBwc/t5YnTx4dO3ZM48aN04oVK+Tv7y9nZ2cVL15cTZs2VefOneXi4vLC1yxXrpxOnjyp0aNH66+//tK1a9eUNm1alShRQi1btlTr1q3Nm5EAAAAAQGKx+QQtpWKCBgAAACC+GAsBAAAAgJ2goAEAAACAnaCgAQAAAICdoKABAAAAgJ2goAEAAACAnaCgAQAAAICdoKABAAAAgJ2goAEAAACAnaCgAQAAAICdoKABAAAAgJ2goAEAAACAnaCgAQAAAICdoKABAAAAgJ2goAEAAACAnaCgAQAAAICdoKABAAAAgJ2goAEAAACAnaCgAQAAAICdoKABAAAAgJ2goAEAAACAnaCgAQAAAICdoKABAAAAgJ2goAEAAACAnaCgAQAAAICdoKABAAAAgJ2goAEAAACAnaCgAQAAAICdoKABAAAAgJ2goAEAAACAnaCgAQAAAICdoKABAAAAgJ2goAEAAACAnaCgAQAAAICdoKABAAAAgJ2goAEAAACAnaCgAQAApADz58/Xzp07bR0DwGsymEwmk61DpERBQUHy8vJSYGCgPD09bR0HAACkYBEREXJzc1PmzJl1+/ZtW8cB8BqYoAEAACRzM2bMkMlk0p07d3Tz5k1bxwHwGihoAADA7MiRI/JM56lFixbZOgriYe7cuTIYDJKkUaNG2TgNgNfBLY6JhFscAQDJUZkyZXTkyBH5+GRQQMB9W8dBHERFRcnV1VUlS5bU+fPn5e3trevXr9s6FgArMUEDAACSnk7Pjhw5Ki+HN/TgQYBmzZpl60j4j23btqlQoULKnz+/+SN37twyGo1q06aNqlatqhs3blgcz58/v0qVKqVHjx7ZOj6AOGCClkiYoAEAkpsyZcro+JFTqubhq72hs+WW3sQUzc4sWbJEX3zxhSTJwcFBHh4ekqT06dPr/PnzOnbsmGrXrq2oqChJT/89IklOTk7y9/dXzpw5bRMcQJwxQQMAAObpWR6XKnI2uKuASw2maHaoadOm2rx5s9KmTSuj0ajKlSsrICBAV65ckYuLi8qXL6+AgACdPXtWmTJlkiTlzp2bcgYkI0zQEgkTNABAcvLs9MzZ4C6TyahdoT8wRbNTwcHB+uCDD7R3715lypRJN2/elJOTkyTpxIkTKlu2rCIjI9WuXTvNnDlTDg78Th5ILvh/KwAAqdx/p2eSZDA4MEWzYx4eHtqzZ4+KFi2q+/fvy2g0mo9dvnzZXM5mz55NOQOSGf4fCwBAKteqVSs5ylm5XN6xWM/iVExpHDKqX7/+NkqGlzEajbpw4YLy58+v8+fPK1euXHrrrbdUrVo1OTs7a9u2bbaOCMAKFDQAAFKx2KZnMZiivZ4///xTNWvWVNmyZfX222/rq6++0u3btxP09SMjI+Xh4aFSpUrp+vXr2r9/v7JmzarMmTPrwoULCg0NTbDvByBpUNAAAEjFXjQ9i8EULf5GjBghHx8f1atXT3t279bjoEDdvnVTCxYsUM6cOVWqVCmdPn36tb/PlClTJD0t2d7e3jpy5IgWLlyoqKgo3bhxQyaTyXwOgOSDggYAQCr1sulZDKZo8dOgQQMNGDBAhQvk16/z5yrgur/OHzukK2dO6PLpY+rTo5uuXrmismXKvPYtiPv375ckffTRR7p9+7ZKlSqlZs2a6erVqypYsKAkaenSpa/7IwFIYuzimEjYxREAYO/+u3Pji7CjY9x07txZU6dOVY/vOmnMiGEv3Jzj8pUrer/Op7p1545OnTqlvHnzWvX9evfubS5lsRk0aJBy586tNm3aWPX6AGyDgpZIKGgAAHt25MgRlSlTVvldqiu/a/VXnn878oSOhS3VzJkz1aFDhyRImLyEhobK29tbn9T9UL8tnC+DwfDS8y/5X1bxcm+rcpUq2rRpUxKlBJAccIsjAACpUKtWreQgJ+V0Lq9oU+QrPzI6FVIaQwaeRXuBwYMHKyIiQgN793xlOZOkfHnzqFmTz/XPP/8oLCwsCRICSC6YoCUSJmgAAHtmMDhIsu6fAPv27VOFChUSNlAylzVrVuV+I6f2bt8c52sOHT6iclXe0/Dhw9W/P8UXwFNOtg4AAACS3pQpk3XmzJl4X5c5c2aVKVMmERIlbyEhIapQrmy8rnmzVElJ0rlz5xIjEoBkioIGAEiVjh07psOHD6tly5a2jmITnTt3tnWEFMYkB4dX39r4LIPBIAcHg6KiohIpE4DkiIIGAEiVqlWrpsDAQJUvX17Fixe3dRwkc66ubjp1+my8rjl/4aKMRpPy5MmTOKEAJEtsEgIASHUmTZqkwMBASU83ywBe16effqot23fo4iX/OF8ze+48OTs5qWfPnomYDEBywyYhiYRNQgDAfqVPn14REREqU6aM/v33X504cYIpGl7LgwcPlCVLFrVr1VLTJ41/5fn37t1XgZJlVKRoUe3duzcJEgJILqyeoEVFRWnTpk2aNWuWHj9+LEm6efOmgoODEywcAAAJLWZ61qNHD82bN08SUzS8Ph8fH3388cea8eMc/TBj1kvPffjwkerUb6gnYWGaNm1aEiUEkFxYNUG7cuWKPvzwQ129elXh4eE6d+6c8uXLpy5duig8PFwzZ85MjKzJChM0ALBPMdOz4OBgOTg4qEqVKtq1axdTNLw2o9Got956SwcPHlTjhp+p67ff6O0K5c3vixYSEqLFy5bLb+wEXbtxQz///LOaN29u49QA7I1VE7QuXbqofPnyevjwodzd3c3r9evX1+bNcX//DwAAktKz0zMHh6d/Bf7888+SmKLh9U2aNElLly7Vl19+qT9X/62K1WuqcOnyqlOvoap/+LGy5i2k9p26KDQsTH/99RflDECsrJqgZciQQf/++68KFy6sdOnS6ejRo8qXL58uX76sYsWKKTQ0NDGyJitM0ADA/vx3ehaDKRpe19mzZ1WkSBGVK1dOBw4cUFRUlEaOHKnFixcrKChITk5OypMnj0aOHKkqVarYOi4AO2bVBM1oNCo6Ovq59evXrytdunSvHQoAgIQW2/QsBlM0vC4/Pz9J0pEjR2Q0GuXk5KRBgwbpzJkzunnzpq5evaodO3ZQzgC8klUFrVatWpo0aZL5a4PBoODgYA0ePFh169ZNqGwAACSYIUOGyN3dXUOHDn3uWMGCBVW5cmXt379fJ0+etEG65KdHjx4aP/7VuxWmFmvWrJHBYFB0dLQWL15s6zgAkjGrCtr48eO1a9cuFStWTGFhYfriiy+UJ08e3bhxQ6NHj07ojAAAvJaXTc9iMEWLu7CwME2ZMkVDhgyxdRS7cOXKFd27d0+ffPKJDAaDpk+fbutIAJIxq98HLSoqSkuXLtXRo0cVHByssmXLqlmzZhabhqRmPIMGAPbjRc+e/RfPosXN+PHj5evrK0naunWr3nvvPdsGSkLz58/XiBEj9Ow/nx4+fKiAgADt27dPzZo108WLF5UvXz6L67Jnz67t27cndVwAyRBvVJ1IKGgAYB8mTZqkbt26acCAAfr+++9feu758+dVqFAhVahQQfv27UuihMlPyZIldf/+fT148EA1a9bUX3/9ZetISaZ///4aOXKkJMnJycn8i+n8+fPr8OHDWrhwob799luZTCaZTCbz+8N6eHjo4cOHcnJysll2AMmDVQXNz89PWbJkUevWrS3W586dq3v37ql3794JFjC5oqABsIXg4GB5eHjYOoZdiev0LAZTtJeLiopS2rRp9e233+rixYvavn27Hj16ZOtYSWrOnDn6+uuvFRUVpU6dOumHH3547pyTJ0+qevXqunfvnsqVK6ctW7bw7wEAcWJVQcuTJ48WL16sSpUqWazv3btXTZo0kb+/f4IFTK4oaACS2okTJ/RmqTfVrHkzLViwwNZx7MLkyZPVtWtXpU+fXgUKFIjTNYGBgTp//rzeeust7d27N5ET2reVK1eqZcuWFrfzGY1GhYaGaufOnfL391eLFi2e+6WAg4OD/vnnH5UqVSqpIyeZq1evqlq1arpy5YrefPNNHTlyxHxs5cqVatiwoUwmk4YNG6YBAwbYLiiAZMeqOfvt27eVLVu259YzZcqkW7duvXYoAED8ffXVVzKajFq8aImmTJmi9OnT2zqSzUVHR8vFxUUhISE6evRonK9zdnZmEqmnu1tGR0crNDRU2bJlU8OGDWUwGJQpUyZVrFhRZcuW1alTpxQSEqKoqCgtXLhQQUFB8vb2Vs6cOW0dP1HlypVLly5dUsaMGXX+/HmLY4cOHZLRaNT06dP1zTff2CghgOTKql0c33jjDe3ateu59V27dil79uyvHQoAED8nTpzQoYOHlD/L24o2Rqtt27a2jmQXunfvrvDwcEVERMT7Y/PmzbaOb3MlSpTQtWvXVKJECd26dUtXrlzRoEGDNGDAADk4OMjd3V0jR45U9+7ddfjwYT1+/Fjvv/++bt++LR8fH1vHT3SPHj3Sw4cPVblyZW3ZskVZsmRR/fr11aVLF0nSH3/8YduAAJIlqyZo7dq1U9euXRUZGan3339fkrR582b16tVLPXr0SNCAAIBX++qrr+To4KJ3CjWVDA76Y+WfevToEVM0vDYfHx8dP35cffv21bhx4/Tmm2/q/Pnz5s0xbty4oZIlSyoiIkLTpk1LVROjsWPHSpKePHmiGjVqyMHBQX/88Yd27dolLy8v/fPPPzZOCCA5smqC1rNnT7Vp00YdO3ZUvnz5lC9fPnXu3Fnfffed+vbtm9AZAQAvETM9K5GrplydPVQ6z8cpdoq2e/duW0dItfz8/FS6dGmFhobK2dnZvEuhm5ubQkJC9Omnn6aqciZJS5culSTt3LlT+fLl07Vr1zR06FDdv39fgYGBevLkiXbs2GHjlACSm9faZj84OFinT5+Wu7u7ChYsKFdX14TMlqyxSQiApFK+fHkdPXxCjSuPlqvz0+emtp+ao8t39+t+wL0UM0UbNWqU+vbtq44dO2ratGm2jpPqGI1GpU2bVs2bN1fnzp3VpEkTeXt7a/HixWrVqpWOHz+ue/fu2TpmknJwcJDJZFKHDh00c+ZM8/qxY8dUvXp1PXjwQI0aNdKyZcskSRs3btTdu3fj/X0yZ86smjVrJlhuAPaN90FLJBQ0AEnhxIkTKlWylN7M85HK5qtnXg8MvaPf9wxQgwafafny5bYLmECMRqO8PNMrOOSxnJ1cFBzyWC4uLraOlaosW7ZMjRs3Vr169fT333/LxcVFUVFRcnZ2VuXKlbVhwwadPXtWBQsWtHXUJPPVV1+pRYsWqlGjxnPHoqKi9M0336hhw4aqXbu2Nm7cqFq1aln9vTZs2EBJA1IJq98t8cCBA1q2bJmuXr2qiIgIi2MrVqx47WAAgFeLefas+BsfWKx7pcmi/FnfSTHPoo0ePVrBIY+V17G6/KO2qmvXrpo+fbqtY6UqU6dOlfR044uSJUtqx44devjwod59912tX79e0tPbIOfOnWvLmElq3rx5Lzzm5OSkH3/80fx1zORs3I+jVaBQ/jh/jwvnLsq3XW+rJm8AkiernkH79ddfValSJZ0+fVorV65UZGSkTp48qS1btsjLyyuhMwIAYvHfZ8/+K6U8i2Y0GjVyhJ/SGbIrj0M1ZTAU1E8/znnul4NIXMeOHZOTk5P69eunY8eOKX369MqbN68uX76sL7/8Ug4ODtq+fbutY9q9AoXyq3jpYnH+iE+ZA5AyWFXQRo4cqYkTJ2r16tVycXHR5MmTdebMGX3++efKlStXQmcEAMTiRdOzGP+doiVXMdOzfI7VZTAYlNfxPUVGRahr1662jpaqLFmyRIcPH9aIESMs1h0cHDR//nzt2LGDN0gHgARgVUG7ePGiPvroI0kyvwGowWBQt27dNHv27AQNCAB43qumZzGS+xTt2emZj6GAJMnTIQdTNBuoU6eOSpQo8cLjlStXVuXKlZMwEQCkTFYVNG9vbz1+/FiSlCNHDp04cULS0zdsDA0NTbh0AIBYvWp6FiO5T9H+Oz2LwRQNAJBSWVXQqlWrpo0bN0qSGjVqpC5duqhdu3Zq2rRprDsZAQASTlynZzGS6xQttulZDKZoAICUyqpdHKdOnaqwsDBJUv/+/eXs7Kx///1XDRo00IABAxI0IADA0ldffSVJyuSZT3cDL8bpmixeBZLdjo4x07M3nT61mJ7FyOv4ng5E/ciOjgCAFMWqgubj42P+3MHBQX369EmwQACAlztx/KRMMmnjsSnxvvb3339XmzZtEiFVwnrZ9CzGs1O0SZMm8b5oAIAUwer3QYuOjtbKlSt1+vRpSVKxYsX0v//9T05OVr8kACAOTp85pePHj8f7uvTp06tatWqJkCjhvWp6FoMpGgAgpbGqTZ08eVKffvqpbt++rcKFC0t6+pdppkyZtHr16pfu8gQAeD158+ZV3rx5bR0j0cRlehaDKRoAIKWxapOQtm3bqnjx4rp+/boOHTqkQ4cO6dq1aypVqpTat2+f0BkBAKnIi3ZufBF2dAQApCRWTdCOHDmiAwcOyNvb27zm7e2tESNGqEKFCgkWDgCQusRMz5zkpnDTY92MPhyn61zlxRQNAJAiWFXQChUqpDt37qh48eIW63fv3lWBAi+/HQUA7NnatWv17rvvKk2aNLaOkioZjUaFhobIKKPORK+K17WGKINu3LiRom//RPJ34Vzcdl619nwAyZ9VBc3Pz0/fffedhgwZonfeeUeStGfPHg0bNkyjR49WUFCQ+VxPT8+ESQoAiWzz5s2qW7euSpcurcOH4za5QcJycnJSeES4jEZjvK91cHBgoyrYrcyZM0uSfNv1fq3rAaR8BpPJZIrvRQ4O//foWszzATEv8+zXBoNB0dHRCZEz2QkKCpKXl5cCAwMpqUAyUbBgQV24cEEODg46ePCgSpcubetIAGKxceNG3b17N97XZc6cWTVr1kyERHGTXHMDSFpW/apx69atCZ0DAGxq8+bNunjxojp16qWff56uVq1aMUUD7NDGjRtVq1Ytq6/fsGGDzcoOJQtAXFhV0N59992EzgEANvX1118rfXoftWnTUQaDQdOmjdWRI0eYogF2JmYCNXfBHBUpUjjO1505c1atv2xj1QQLAJKSVQVt3bp18vDwUJUqVSRJ06ZN048//qhixYpp2rRpFrs7AoC9i5me9ekzTG5u7mrRoq3mzp3GFA2p1v3799WxY0ctWLBAbm5uto4TqyJFCqtM2TK2jgEACc6q90Hr2bOneSOQ48ePq3v37qpbt678/f3VvXv3BA0IAIktZnr2+ectJEnp0nmqdetvdezYMR05csS24QAb6N69u3777TeNHj3a1lEAINWxqqD5+/urWLFikqTff/9dn3zyiUaOHKlp06Zp7dq1CRoQABJTzPTs66+7ys3N3bzeokVbubunUatWrWyYDrCNNWvWSJIWLVpk4yQAkPpYVdBcXFwUGhoqSdq0aZP5YV0fHx+LLfYBwN79d3oWgykaUit/f38FBATI0dFRFy5cUFhYmK0jWc1kMmnnjp1q36aDOn3TWY4ODvruu+/Uq1evZP1zAUjZrCpoVapUUffu3fX9999r3759+uijjyRJ586dU86cORM0IAAkli1btsQ6PYvBFA2pkZ+fnySpX79+MplM+uGHH2ycyDpbNm1R+TcrqOb7tbV3xz8qnC2zmn5SW3mzZ9HYsWPl5empli1b2jomADzHqoI2depUOTk5afny5ZoxY4Zy5MghSVq7dq0+/PDDBA0IAImlQ4fYp2cxmKIhob377rvq3LmzrWOY1apVSz4+PhYfc+fOVbp06TRkyBA5OTmpX79+z53TrFkzW0d/qd+WLdf/Pq6vzOnSaOP86Tq9frkWTRyhBeO+1/6Vv+jQn4v0ed0PtGDBAlWtWtXWcQHAglUFLVeuXPrrr7909OhRtWnTxrw+ceJETZkyxfz1qFGj9OjRo9cOCQAJ7en07MILp2cxmKIhoSxdulQ7duzQtKnTdf36dVvHkSRFRUXp4cOHevjwoSTJw8ND2bNn14ABA+Tg4KDOnTsrW7Zs8vDwsDjXYDDYOPmL7du7X21atlXjj2pqw7xpqlHprefyli5WWAvGfa/Rvb7Tzp07+f83ALtiVUGLq5EjR+rBgweJ+S0AwCqvmp7FYIqGhNKtaze5KJ0k2U0h2LJliwYMGCCDwaDAwEB17dpVV69eVa9evSRJEyZM0NWrV9WkSRMFBwfL0dFR06ZN08KFC22c/MXGjh6nQnlyaa7fYDk5vfzdhHq2+1KNP6qlJUsWKyIiIokSAsDLJWpBM5lMifnyAGCVuE7PYjBFw+taunSpbt2+pbyqruwqp82bttjNFC3meXJPT0/16NFD7dq1szhep04djR07VtmzZ9f58+fVsWNHGyV9tatXr2nNX2vUqcXncnaO21u9dmv1hcLDIzRy5MhETgcAcWPVG1UDQHLWocPXkqTw8DAtWDA7TtcULFhEx44d0qFDh1S2bNnEjIcUqFvXbnKVl7KopLyVVzd1UK1atdLGjRttHU2SVL58eZ0+fVrZsmXTtWvXLI7dvHlTBoNBly9ffuVEytZW/bFKLs7OavZpnThfU6FUcRUtkFe//PKLhgwZknjhACCO7PtPWgBIBCEhITIYHDRp0qh4XWcwGHT27FkKGuIlZnpWWJ/KIAe5ytNiimYvux9PmDBBktSjRw81adJEK1as0LfffqtWrVqpW7dumj9/vsVz5/bo3t17ypzRRx5p08T5GoPBoCJ582jnkROJmAwA4o6CBiDVuXnzhq0jIBV5dnoW4w1Vsrsp2rJly+To6Kgvv/xSt2/flru7uyZNmqT8+fNLkmbNmmVXBe3MmbPPrd2/f19RUdHxfq2IqEg5OCTqUx8AEGcUNAAAEsl/p2cx7G2KFhQUpCtXrkiS7ty5o759+2r48OFq06aN5s2bJ0k6fPiwjEajzYtM5syZJUmtv4y9LBoMBl25cUu5c2SL0+tFRkZp39GTypErd4JlBIDXkagFrWrVqnJ3f/UD+AAApESxTc9ixEzRWrdurQ0bNtgg3f+ZP3++JMnLy0sbN25UhQoVJEk///yzGjVqpAYNGigsLEzbtm3T+++/b8uoqlmzpjZs2KC7d+8+dywoKEidO3fW7KUrNKL7t3F6vT83bdO9Bw81fVbcnkcFgMRmMMVxq8WgoKA4v6inp6fVgVKKoKAgeXl5KTAwkP8eAJAKLV26VE2aNFFhfaqsejPWcy5onW7qoK5eu2LTKdr9+/c1fvx4DR06VC4uLs8df/Tokfz8/DR48GClSRP357tsoXTp0rpy6aKO/b1UObNleem5T8LCVLFRK129fVcPHz5KmoAA8ApxLmgODg6vfGNKk8kkg8Gg6Oj43/+d0lDQACB1y54tux7cDtXb6mRxe+OzwhWkvfpBH9SsYfMpWkpx7NgxvVWhgnJnz6qNC6brjWxZYz0vOCRUjTr31sZdezV37ly1bNkyiZMCQOzifIvj1q1bEzMHAAApxouePfuvmGfRNm3cbBfPoqUEpUqV0so//lD9evVUos7natu4vr5u2kAF8+SSJN0NeKA5v/2paQuX6c79AI0YMYJyBsCuxHmChvhhggYAqVdcpmcxmKIljpMnT6pVq1Y6fPiwoqKi5JHGXQ4ODnocEiqDwaB8+fJp4sSJ+vjjj20dFQAsxHmCduzYsTi/aKlSpawKAwBAchfX6VkMpmgJ7+zZs+rVq5fu3r0rHx8fhYWFycvLSwUKFFCePHnUv39/89sHAIC9ifczaK86nWfQnmKCBgCpU5EiRXT27Fk5ylUGvfzZ7RhGRcuoSDVt2lSLFy9O5IQp19mzZ/VR3bry9/eXo4OD3iqUW+nc3XT7YZCO+F+Xk6OjKleponXr1snNzc3WcQEgVnGeoPn7+ydmDgAAUoS+fftq0aJF8b7OYDDo22/jtjU8nrd//35Vq1pFbk6OGtniU31V4x1l9PQwHz97445mrv1H09bsUN48eXT+wgV5eHi85BUBwDZ4Bi2RMEEDACBpPHr0SDlzZJeXm4u2jeiqfFkzvvDcdYdOqd7IWcqfv4DOnD2bhCkBIG6seqPqBQsWvPT4l19+aVUYAACA+OrZs6dCQp9ox/DvXlrOJOnDssU0oXUDfffjb9qyZYvN33gbAP7Lqgmat7e3xdeRkZEKDQ2Vi4uL0qRJowcPHiRYwOSKCRoAAEnDyzOd3s6fU2sHP3+LaMx7tD7rSXiEcrTqr0LFS2j//v1JFRMA4uTV20vF4uHDhxYfwcHBOnv2rKpUqaIlS5YkdEYAAIBYbdmyRUGPg9WhdpXnjq3ed1wNRv2o+0HBFuvuri76qsY7Onb0SBKlBIC4s6qgxaZgwYIaNWqUunTpklAvCQAA/j+j0ai8efNqxIgRto5iV86cOSNJKpQjs8X6jYBHajt1kVbtO66y3Ubp+v2HFscL58iiiMgoGY3GJMsKAHGRYAVNkpycnHTz5s14XxcaGqq1a9dq+PDh+uyzz5Q7d24ZDAYZDAYNGTLkpdcOGTLEfO7LPi5cuPDS1zl06JCaN2+unDlzytXVVdmyZVP9+vW1ZcuWeP88AAAktD///FOXL1/W5MmTbR3Frjg6OkqSnn1gIzraqK8m/6KAxyGSpAoFcytHhvQW1xlNRhni9i4IAJCkrNokZNWqVRZfm0wm3bp1S1OnTlXlypXj/Xr79u1T3bp1rYli5uzsLB8fnxced3J68Y/6008/6ZtvvlFUVJQkycvLS3fu3NEff/yhP/74Q4MHD35lUQQAIDHFFLN79+7pypUryp07t40T2YeyZctKkg5fuqbiubJJksb9sUlbj5+TJOXIkF6zO37x3HNoRy5dl6uzixwcEvR31QDw2qwqaPXq1bP42mAwKFOmTHr//fc1fvx4q4J4e3urbNmy5o9u3brp9u3bcb6+UqVK2rZtW7y/7+7du/X1118rOjpa9erV0w8//KCcOXMqICBA/fv316xZszR06FAVK1ZMn3/+ebxfHwCAhLBnzx6lT59ejx490qhRozRjxgxbR7ILFSpUUAYfH01f+4+av/eW9p27rMFL/pb09N8n87q0UAbPtBbXPAoJ1aLt+1WpajVbRAaAl7KqoCX0/dpVq1Z9bufHPn36JOj3eJFevXopOjpaJUuW1LJly+Ts7CxJypAhg2bOnKnLly9r/fr16t27txo0aGC+lQIAgMRy//59i79rd+zYofDwcHXp0kXTpk3TH3/8oaFDh1pc4+HhoTRp0iR1VLvQuk0bjR07Vqv3HVePuSsUFf30v13vz2qqeslCz50/adVWhUdFWf1LZQBITHYx17dV6bl06ZJ27twpSfL19TWXs2f17dtXknT58mXt2LEjSfMBAFKfli1bKlOmTMqSJYv5o1GjRpKevt/Xe++9p9u3b1scz5Ilizw8PBQREWHj9LYxbNgwZfDxVqMxP+nSnfuSpLcK5dHgJs8/PvHThl0avmydqlSpqtKlSydxUgB4NasmaNHR0Zo3b542b96su3fvPjdRSy4ba2zcuNH8+YcffhjrOVWqVFG6dOn0+PFjbdiwQdWrV0+qeACAVKhnz55atWqVHj16JFdXV3311VdydnZW+fLllTFjRi1YsEDDhw9XZGSkjh49qn/++UeS1KRJE7m4uNg4vW24ublpwMBB6tatmyTJ2dFRAxp9KGenp78ANhqN2nj0rKb/vV1/HzypEiVKWPVYBAAkBasKWpcuXTRv3jx99NFHKlGixHMP3trCyZNP/8C9dOmSHBwclCNHDlWrVk0dO3ZUmTJlYr3mxIkTkqTMmTMrc+bMsZ7j6OioIkWKaP/+/Tp58mSi5QcAQJJKlCihO3fu6KOPPtKmTZu0ePFirVmzRlWqPH2fLx8fH40bN05fffWV/vnnH7m4uGjRokVq2LChjZPbzqVLlzRo0CDz10aTSZ+OmKms3p7ycHPV/aBgPQp5orRp3NW2bVvNmjWLzUEA2C2rCtqvv/6qZcuWvfbOiwnp/v37evDggdKnT6+goCCdO3dO586d05w5c9SvXz8NHz78uWti3hIgR44cL33tHDlyaP/+/S99C4Hw8HCFh4ebvw4KCrLyJwEApHYuLi7auHGjZs2apW+//VY1atSw+Dtm4MCB+uWXX1SoUCHt2rVLGTNmtGFa24qMjFTTpk31+PFjSdKXX36pGTNmaPDgwdq9e7dCQ0OVs7CXGjVqpK+//ppiBsDuWfWnlIuLiwoUKJDQWaxSsGBBjRkzRmfPnlVYWJgCAgIUEhKi9evXq1y5cjKZTBoxYkSsDwLH/GH+qoeqY47HnB8bPz8/eXl5mT/eeOON1/ipAACQOnToIHd3d/PbyOzYsUMPHjwwT9MaNGiQqsuZJA0ePFj79u2TJBUoUEBTp05VmjRpNHbsWO3cuVOHDh3S1q1b1bFjR8oZgGTBqj+pevToocmTJ8v07LtC2kizZs3Us2dPFSpUyLzJh4uLi2rVqqWdO3eqQoUKkp6+oXVgYGCi5ejbt68CAwPNH9euXUu07wUASB0OHDig4OBgffTRR6pRo4beffddZcuWTY8fP5arq6uWLl1q64g2tWXLFo0aNUrS0/c7Xbx4sdKlS2fjVADweqy6xXHnzp3aunWr1q5dq+LFiz+3++GKFSsSJNzrcnNz08iRI1WzZk0FBwdr8+bN+uyzz8zHY/4QDw0NfenrxBx/2R/6rq6ucnV1TYDUAAA8NXr0aEnSokWLFBYWpvLly+vEiRNq3LixXFxcdOnSJQUHB8vDw8PGSZPe/fv31aJFC/Mvi0eMGGH+pSwAJGdWTdDSp0+v+vXr691331XGjBktbu3z8vJK6IyvpWLFiubPL126ZHEse/bskqQbN2689DVijsecDwBAUojZbTgyMlKTJk3S/v37devWLb355pvmLfUnTJhgy4g2YTKZ1KZNG/Oz4TVq1JCvr6+NUwFAwrBqgvbzzz/H6bxdu3apfPnydjtZKlGihCTp7t27unfvnjJlyvTcOdHR0Tpz5owkqXjx4kmaDwCQuj1+/FjZsmXT9u3bVbBgQUlPf0l65MgRDRw4UCNGjNDhw4dtnDLpzZgxQ6tWrZIkZciQQQsWLOD5MgApRqL+aVanTp1XTqcS2549e8yf582b1+JYzZo1zZ+vW7cu1ut37dpl3hykVq1aiZAQAIDYPXz4UDdv3jSXs2d9//33CgoK0m+//WaDZLZz4sQJ9ejRw/z1zz//zB0uAFKURC1oib2JyKtePzw8XP3795ckpU2bVjVq1LA4ni9fPvNOWOPHj1dkZORzrxHz8HHu3LlVrVq1hIgNAEihZs2apS1btsTpXKPRKKPR+NJzPD09X3rcw8NDTk5W3QyTLD158kRNmjRRWFiYJKlTp0765JNPbJwKABKW3dwP8PDhQ92/f9/8EfOXVmhoqMV6cHCw+ZodO3bogw8+0C+//KLr16+b1yMjI7V582ZVrVpVe/fulSQNGjRI6dOnf+77jh49Wo6Ojjp69KiaNGlinvg9ePBAHTt21Nq1ayVJY8aMkaOjY2L9+ACAZC4sLEzffPONmjZtGqfzc+TIwa3z8dSzZ0+dPHlSklSyZEmNHTvWxokAIOEZTIk45kqXLp2OHj2qfPnyvfLcPHny6MqVK688r2XLlpo3b54kadu2bapevbr5mLu7u9KmTavAwEDzNMzBwUF9+vTRiBEjXviaP/30k7755htFRUVJenp/f2BgoHlCN3jwYA0ZMuSV2Z4VFBQkLy8vBQYGvvI3oACA5G/ChAnmW++uXbumnDlzvvDc06dPq1ixYpLE3xNxtGrVKv3vf/+T9HSX5oMHD5r/GwJASmI3EzRrlCxZUuPGjVODBg1UqFAhubu769GjR3J3d9ebb76pTp066ciRIy8tZ5LUtm1b7d27V1988YVy5Mih0NBQZc6cWfXq1dPmzZvjXc4AAKnPzz//LIPBIEny8/N76bkxt89L0rhx4xI1V0pw48YNtW7d2vz1xIkTKWcAUqxEnaB5enrqyJEjcZqgpTRM0AAg9YiKipKrq6vefPNNnTt3Tl5eXi/dJCtTpkyKjIxUaGiocufOrfPnzydh2uQlOjpatWrVMj/bV69ePa1YscJchgEgpUnUJ4sTe5MQAACS2saNG9W+fXtFR0eb18LDw2U0GtW2bVutXr1a69atU65cuWQymRQcHCyTySQXFxd9/fXXcnR01P3799WwYUPdvHlTu3fvlp+fn0Xh8PLyUrt27VLVBiAvMnbsWHM5y5Ejh3766SfKGYAUzaoJ2vvvv68VK1Y8t+lGUFCQ6tWrF+cdrFIyJmgAkDItW7ZMjRs3lvT0OWcvLy9Jko+Pj86cOaMTJ06odu3aioyMVFhYmJ48eRLr6xw8eFD+/v5q2LBhrMc3bdr03O7Dqc2+fftUuXJlRUVFyWAwaMuWLXrvvfdsHQsAEpVVz6Bt27ZNERERz62HhYXpn3/+ee1QAADYq88//1w7d+5UunTpZDQaVbZsWd2+fVsXLlyQk5OTSpcurTt37ujIkSPKmDGj+TqDwaCBAwfq33//1YULF1S2bFk1aNBAZ86c0b///qtOnTqZz/P19U315SwoKEhNmzY1b+DVr18/yhmAVCFe904cO3bM/PmpU6d0+/Zt89fR0dFat26dcuTIkXDpAACwQ5UrV9bdu3dVq1Ytbd68WdmyZdOdO3fMtyQeOXJEFSpUUFRUlDp37qxatWqpUaNG+v7773Xu3Dn9+uuv5tcqWLCgatasqS1btsjT01Pr1q1TxYoVbfWj2Y1vv/1Wly5dkiS9/fbbGjx4sI0TAUDSiNctjg4ODub7vmO7zN3dXT/88IPFTkupFbc4AkDqUKxYMZ05c0YRERHmgrZ27VrVrVtXHTp00MyZMyVJjx49UokSJXTjxg1dvnxZuXPnliStX79eH374oYoWLapDhw7Jzc3NZj+LvVi4cKFatGgh6elb9qTWDccApE7xmqD5+/vLZDIpX7582rdvnzJlymQ+5uLiosyZM/NmzgCAVMNoNOrChQsqVKiQjh07po8//lhZs2bVli1b5OLiok2bNpnPTZ8+vRwcHOTu7q7cuXNr1KhR8vb2Vps2beTg4CAXFxfKmaSLFy/qm2++MX89c+ZMyhmAVCVeBS3mt31GozFRwgAAkJysWLFCkZGRcnFxUfny5WUwGHTr1i1ly5ZNGTNm1KVLlxQaGqo0adLowYMHunbtmqpUqaKSJUvqxIkTkp5Oi4oWLaoTJ04oKioqVe/cGBkZqS+++ELBwcGSpC+//FJffPGFjVMBQNKyapOQ+fPn6++//zZ/3atXL6VPn16VKlXSlStXEiwcAAD27IcffpAkHT9+XJkyZdKpU6e0fPlymUwm3bx5UyaTSZMnT5b0dLt4Sdq1a5dOnDihxo0bq2bNmtq5c6dOnTql6OhozZs3z1Y/il0YPHiw9u3bJ0kqUKCApk6dauNEAJD0rNpmv3DhwpoxY4bef/997d69WzVq1NCkSZP0119/ycnJSStWrEiMrMkKz6ABQMqXNm1ahYaGqkGDBlq2bJkcHJ7+3vP+/fuqVq2aTp8+rdKlS+vw4cPKly+f/P395erqqiVLlqh+/fqSpNmzZ6tjx46Kjo5W+fLltX//flv+SAlu48aNunv37ivPO3nypEaNGiWTySRHR0dNmjTJvLMlAKQmVhW0NGnS6MyZM8qVK5d69+6tW7duacGCBTp58qTee+893bt3LzGyJisUNABI+QYNGqQ333xTDRo0iPW4n5+f3njjDTVv3lylS5eWyWTS5s2bLbbfl6QrV67ogw8+UK5cubR58+akiJ4kNm7cqFq1all9/YYNG1SzZs0ETAQA9s+qG909PDwUEBCgXLlyacOGDerevbskyc3N7YVvyAkAQEozbNiwlx7v27ev+fMjR4688LzcuXPr/PnzCRXLbsRMzrpM+FY5C8T+Njwmk0nzRizUqX2nJUkF3yygOi1qaYrv9DhN3gAgpbGqoNWsWVNt27ZVmTJldO7cOdWtW1fS09sT8uTJk5D5AABAMpezQA7lL5E31mNrf9lgLmeePunUe1Z3Pbz7KAnTAYB9sWqTkGnTpqlSpUq6d++efv/9d2XIkEGSdPDgQTVt2jRBAwIAgJTpytlrmj9yofnrTqO/lk9mbxsmAgDbi/cELSoqSlOmTFHv3r2VM2dOi2NDhw5NsGAAACDlCg+L0IQuUxQRHilJqvNlLZWvUdbGqQDA9uI9QXNyctKYMWMUFRWVGHkAAEAK9/S5s1907dx1SVLuwm+oZd9mNk4FAPbBqmfQatSooe3bt/O8GQAAiJeLxy9pvt8indhzSpLk4uqs7lO+k4uri42TAYB9sKqg1alTR3369NHx48dVrlw5pU2b1uL4p59+miDhAABAynD3+j0tHr9UO/7cZbHeelBLvVEw5wuuAoDUx6qC1rFjR0nShAkTnjtmMBgUHR39eqkAAECK8dfPa7Xr792Kivi/xyOy5s6iL/s00zu1K9gwGQDYH6sKmtFoTOgcAAAgBQkPD9fatWslSdtX/mNeT+ftoc87N1CtLz6Qs4tV/wwBgBSNPxkBAECCMZlMWrZsmfr27St/f3/zurOLsz5uVUefffM/pfVMY8OEAGDf4lzQpkyZovbt28vNzU1Tpkx56bnffffdawcDAADJyz///CNfX1/t27fPYr1s9TLqMKy1MuXIaKNkAJB8xLmgTZw4Uc2aNZObm5smTpz4wvMMBgMFDQCAVOTs2bPq06eP/vjjD4v14sWL6+TJk2rarRHlDADiKM4F7ciRI/Ly8pIki1sWAABA6nT37l0NGTJEs2fPttggrESJEhozZowCAgLUokULXb9wI16vG9/zASAliXNB8/Hx0a1bt5Q5c2a9//77WrFihdKnT5+I0QAAgD0KDQ3VhAkTNHr0aAUHB5vXs2XLpu+//15fffWVHB0dtXHjRknS5O7TrPo+mTNnTpC8AJCcxLmgeXh4KCAgQJkzZ9a2bdsUGRmZmLkAAICdiY6O1vz58zVw4EDdvHnTvO7h4aFevXqpe/fuFu+NWrNmTW3YsEF3796N9/fKnDmzatasmSC5ASA5iXNB++CDD1S9enUVLVpUklS/fn25uLjEeu6WLVsSJh0AALA5k8mk9evXq1evXjp+/Lh53dHRUe3atdOQIUOUJUuWWK+lZAFA/MS5oC1cuFDz58/XxYsXtX37dhUvXlxp0rBNLgAAKdmRI0fUs2dPbdq0yWL9008/1ejRo1WkSBEbJQOAlMlgMplM8b2oevXqWrlyJc+gvURQUJC8vLwUGBgoT09PW8cBACBerl27pgEDBuiXX37Rs/9UqFChgsaNG6dq1arZMB0ApFwO1ly0devWOJUzT09PXbp0yZpvAQAAbCAwMFB9+/ZVoUKFtGDBAnM5y5s3r5YsWaI9e/ZQzgAgEcX5FkdrWDGcAwAANhAREaFZs2Zp2LBhun//vnnd29tbAwYM0LfffitXV1cbJgSA1CFRCxoAALBvJpNJK1asUJ8+fXThwgXzuouLizp37qz+/fvL29vbhgkBIHWhoAEAkErt3r1bvr6++vfffy3WmzZtqhEjRihv3rw2SgYAqRcFDQCAVObChQvq06ePfv/9d4v1d999V+PGjVP58uVtlAwAYNUmIXFlMBgS8+UBAEA83L9/X999952KFi1qUc6KFi2q1atXa+vWrZQzALAxNgkBACCFe/LkiSZPniw/Pz8FBQWZ17NkyaKhQ4eqTZs2cnLiphoAsAev/adxTAmLbVq2du1a5ciR43W/BQAAsILRaNTChQs1YMAAXbt2zbyeJk0a+fr6ytfXV+nSpbNhQgDAf1l9i+OCBQtUsmRJubu7y93dXaVKldIvv/xicU6VKlXYkhcAABvYtGmTypUrp5YtW5rLmYODg9q2bavz589r6NChlDMAsENWTdAmTJiggQMHqlOnTqpcubIkaefOnfr66691//59devWLUFDAgCAuDl+/Lh69eqldevWWazXrVtXY8aMUfHixW2UDAAQFwaTFQ+K5c2bV0OHDtWXX35psT5//nwNGTJE/v7+CRYwuQoKCpKXl5cCAwPl6elp6zgAgBTuxo0bGjRokObNmyej0WheL1u2rMaOHav333/fhukAAHFl1QTt1q1bqlSp0nPrlSpV0q1bt147FAAAiJvHjx9rzJgxGj9+vJ48eWJez5Url0aOHKmmTZvKwSFRN20GACQgq/7ELlCggJYtW/bc+tKlS1WwYMHXDgUAAF4uMjJSM2bMUIECBTR8+HBzOfPy8tLo0aN19uxZNWvWjHIGAMmMVRO0oUOHqnHjxtqxY4f5GbRdu3Zp8+bNsRY3AACQMEwmk1atWqXevXvr7Nmz5nVnZ2d17NhRAwcOVIYMGWyYEADwOqx6Bk2SDh48qIkTJ+r06dOSnr7JZY8ePVSmTJkEDZhc8QwaACCh7du3Tz179tSOHTss1hs1aiQ/Pz/lz5/fRskAAAnF6oKGl6OgAQASyqVLl9SvXz8tXbrUYr1KlSoaN26c3n77bRslAwAkNKtuTHd0dNTdu3efWw8ICJCjo+NrhwIAANKDBw/UvXt3FSlSxKKcFSxYUCtXrtSOHTsoZwCQwlj1DNqLhm7h4eFycXF5rUAAAKR2YWFhmjp1qkaMGKFHjx6Z1zNlyqTBgwerffv2cnZ2tl1AAECiiVdBmzJliiTJYDDop59+koeHh/lYdHS0duzYoSJFiiRsQgAAUgmj0ahff/1V/fr105UrV8zrbm5u6t69u3r37s1t8wCQwsWroE2cOFHS0wnazJkzLW5ndHFxUZ48eTRz5syETQgAQCqwbds29ezZUwcOHDCvGQwGtWzZUt9//71y5sxpw3QAgKQSr4Lm7+8vSapevbpWrFghb2/vRAkFAEBqcerUKfXu3Vt//fWXxXrt2rU1ZswYlSpVykbJAAC2YNUmIVu3bo1TOfP09NSlS5es+RYAAKRot2/fVocOHVSyZEmLclaqVCmtX79e69ato5wBQCpk1SYhccUO/gAAWAoODtb48eM1duxYhYSEmNdz5Mih4cOHq0WLFuyIDACpWKIWNAAA8FRUVJR+/vlnDRo0SLdv3zavp0uXTn369FHXrl2VJk0aGyYEANgDChoAAInIZDJpzZo16tWrl06dOmVed3JyUocOHTRo0CBlzpzZhgkBAPaEggYAQCI5dOiQfH19tXXrVov1zz77TH5+fipUqJCNkgEA7JVVm4TElcFgSMyXBwDALl25ckXNmzdXuXLlLMrZO++8o507d+r333+nnAEAYpWoBY1NQgDAfoWFhSkoKMjWMVKUR48eqVevXipcuLAWLVpkXs+fP7+WLVumf//9V5UrV7ZhQgCAvXutghYREaGzZ88qKioq1uNr165Vjhw5XudbAAASwZUrV5QlSxZlyZJFu3fvtnWcZC8iIkKTJk1S/vz5NXbsWIWHh0uSfHx8NGnSJJ06dUqNGjXizhIAwCtZVdBCQ0PVpk0bpUmTRsWLF9fVq1clSZ07d9aoUaPM51WpUkWurq4JkxQAkCCuXLmikiVLSpIyZMio6tWrU9KsZDKZtGzZMhUtWlTdunXTgwcPJEmurq7q1auXLl68qC5dusjFxcXGSQEAyYVVBa1v3746evSotm3bJjc3N/P6Bx98oKVLlyZYOABAwoopZwaDQes3bdPm7TuVKXNmSpoVdu7cqYoVK6px48a6dOmSeb1FixY6d+6cRo8erfTp09suIAAgWbKqoP3xxx+aOnWqqlSpYnG7RvHixXXx4sUECwcASDj/LWely5RR7ty5tWnrP5S0eDh79qzq16+vqlWrau/eveb1999/XwcPHtSCBQuUK1cuGyYEACRnVhW0e/fuxfqeLSEhIdxfDwB2KLZyFoOSFjd3797Vt99+q+LFi+uPP/4wrxcvXlx///23Nm3apLJly9ouIAAgRbCqoJUvX15///23+euYUvbTTz+pYsWKCZMMAJAgXlbOYlDSXiw0NFQjRoxQgQIFNH36dEVHR0uSsmXLph9//FFHjhxR3bp1+QUlACBBWPVG1SNHjlSdOnV06tQpRUVFafLkyTp16pT+/fdfbd++PaEzAgCsFJdyFiOmpH1QvaqqV6+urVu3pupfukVHR2vBggUaOHCgbty4YV5Pmzatevfure7duytt2rQ2TAgASImsmqBVqVJFR44cUVRUlEqWLKkNGzYoc+bM2r17t8qVK5fQGQEAVohPOYvBJO2p9evXq2zZsmrdurW5nDk6Ourrr7/WxYsXNXDgQMoZACBRGEy8m3SiCAoKkpeXlwIDA+Xp6WnrOABSGWvK2X+v/6B6Vd27ezdVTdKOHj2qnj17auPGjRbrn3zyiUaPHq2iRYvaKBkAILWwaoK2Zs0arV+//rn19evXa+3ata8dCgBgvatXr75WOZOen6Q9u1thSnT9+nV99dVXKlOmjEU5K1++vLZt26ZVq1ZRzgAAScKqgtanTx/zQ9LPMplM6tOnz2uHAgBYb/z48Xr8+LG6dOthVTmLkTt3bg37fqTCw8M1ePDgBExoPwIDA9WvXz8VLFhQ8+fPV8xNJXny5NHixYu1d+9evfvuuzZOCQBITay6xdHd3V2nT59Wnjx5LNYvX76s4sWLKyQkJKHyJVvc4gjAViIiIszvSzl3/kI1afqFVa+zY/s2ffrRh0qXLp3Onj0rHx+fBE5qO5GRkZo1a5aGDh2q+/fvm9e9vb01YMAAffvtt3J1dbVhQgBAamXVBM3Ly0uXLl16bv3ChQs8NA0ANubi4qKTJ08qf/78at2yuX5dsjjer5FSy5nJZNKKFStUvHhxde7c2VzOXFxc1KNHD124cEHdu3ennAEAbMaqgva///1PXbt21cWLF81rFy5cUI8ePfTpp58mWDgAgHVep6Sl1HK2e/duVa1aVQ0aNND58+fN602aNNGZM2c0bty4FPOzAgCSL6tucQwMDNSHH36oAwcOKGfOnJKePmBdtWpVrVixQunTp0/onMkOtzgCsAfxvd0xJZazCxcuqG/fvlq+fLnFerVq1TRu3DhVqFDBRskAAHie1dvsm0wmbdy4UUePHpW7u7tKlSqlatWqJXS+ZIuCBsBexLWkpbRydv/+fX3//feaMWOGIiMjzetFihTR6NGj9cknn8hgMNgwIQAAz+N90BIJBQ2APXlVSUtJ5ezJkyeaMmWKRo4cqaCgIPN6lixZNHToULVp00ZOTk42TAgAwItZ/TfU5s2btXnzZt29e1dGo9Hi2Ny5c187GAAg4cQ8k1a8eHG1btlckswlLaWUM6PRqEWLFql///66du2aeT1NmjTy9fWVr6+v0qVLZ8OEAAC8mlUFbejQoRo2bJjKly+vbNmycYsIACQDsZW07Nmzp4hytnnzZvXs2VOHDx82rzk4OKh169YaOnSosmfPbsN0AADEnVW3OGbLlk1jxoxRixYtEiNTisAtjgDs1bO3Ozo7O8vT0zPZlrMTJ06oV69eWrt2rcV6nTp1NGbMGJUoUcJGyQAAsI5V2+xHRESoUqVKCZ0FAJAEYiZpJUqUUJYsWZJlObt586batm2rN99806KclSlTRps2bdKaNWsoZwCAZMmqgta2bVstXhz/Nz4FANgHFxcXHTt2TFevXk1W5ezx48caNGiQChYsqDlz5pifgc6VK5d++eUXHThwQDVq1LBxSgAArGfVM2hhYWGaPXu2Nm3apFKlSsnZ2dni+IQJExIkHAAAkhQVFaWffvpJgwcP1t27d83rXl5e6tevn7777ju5ubnZMCEAAAnDqoJ27NgxlS5dWtLT+/+fxYYhAICEYjKZtHr1avXu3Vtnzpwxrzs5Oenbb7/VgAEDlDFjRhsmBAAgYfE+aImETUIA4PXs379fvr6+2rFjh8V6w4YN5efnpwIFCtgoGQAAiYd36gQA2BV/f3/169dPv/76q8V6pUqVNG7cOFWsWNFGyQAASHxWF7QDBw5o2bJlunr1qiIiIiyOrVix4rWDAQBSlwcPHmjEiBGaOnWqxd8rBQsW1OjRo1WvXj1uowcApHhW7eL466+/qlKlSjp9+rRWrlypyMhInTx5Ulu2bJGXl1dCZwQApGDh4eEaP3688ufPrwkTJpjLWcaMGTV16lSdPHlS9evXp5wBAFIFqwrayJEjNXHiRK1evVouLi6aPHmyzpw5o88//1y5cuVK6IwAgBTIaDRqyZIlKlKkiHx9ffXo0SNJkpubm/r166eLFy/q22+/fW6nYAAAUjKrCtrFixf10UcfSXr6XjohISEyGAzq1q2bZs+enaABAQApz/bt2/X222/riy++0OXLlyU93QW4ZcuWOnfunEaMGMEGSwCAVMmqgubt7a3Hjx9LknLkyGHeav/Ro0cKDQ1NuHQAgBTl9OnT+vTTT/Xee+/pwIED5vWaNWvq8OHDmjdvnt544w0bJgQAwLas2iSkWrVq2rhxo0qWLKlGjRqpS5cu2rJlizZu3KgaNWokdEYAQDJ3+/ZtDRkyRD/99JOio6PN66VKldLYsWNVq1YtG6YDAMB+WPU+aA8ePFBYWJiyZ88uo9GoMWPG6N9//1XBggU1YMAAeXt7J0bWZIX3QQMAKSQkROPHj9eYMWMUEhJiXs+RI4eGDx+uFi1ayNHR0YYJAQCwL7xRdSKhoAFIzaKjo/Xzzz9r0KBBunXrlnk9Xbp06tOnj7p27ao0adLYMCEAAPbJqmfQHB0ddffu3efWAwIC+E0oAKRiJpNJa9as0Ztvvql27dqZy5mjo6O+/fZbXbhwQf369aOcAQDwAlY9g/aioVt4eLhcXFxeKxAAIHk6dOiQevbsqS1btlis169fX35+fipcuLCNkgEAkHzEq6BNmTJF0tOtkH/66Sd5eHiYj0VHR2vHjh0qUqRIwiYEANi1K1euaMCAAVq4cKHF+ttvv61x48apSpUqNkoGAEDyE6+CNnHiRElPJ2gzZ860uJ3RxcVFefLk0cyZMxM2IQDALj169Eh+fn6aPHmywsPDzev58uXTqFGj1LBhQxkMBhsmBAAg+YlXQfP395ckVa9eXStWrGC3RgBIhSIiIjRjxgx9//33CggIMK/7+Pho0KBB+uabb7jdHQAAKyXILo7R0dE6fvy4cufOTWn7/9jFEUBKYzKZtHz5cvXt21cXL140r7u6uqpLly7q27ev0qdPb7uAAACkAFbt4ti1a1fNmTNH0tNyVq1aNZUtW1ZvvPGGtm3blpD5AAB2YNeuXapUqZI+//xzi3LWvHlznT17VqNHj6acAQCQAKwqaL/99pvefPNNSdLq1at1+fJlnTlzRt26dVP//v0TNCAAwHbOnTunzz77TFWqVNGePXvM6++//74OHjyoX375Rblz57ZhQgAAUharClpAQICyZs0qSVqzZo0aNWqkQoUKqXXr1jp+/HiCBgQAJL179+6pU6dOKl68uFauXGleL168uP7++29t2rRJZcuWtWFCAABSJqsKWpYsWXTq1ClFR0dr3bp1qlmzpiQpNDSUN6oGgGQsNDRUI0eOVP78+TVt2jRFRUVJkrJly6Yff/xRR44cUd26ddmdEQCARGLVG1W3atVKn3/+ubJlyyaDwaAPPvhAkrR3717eBw0AkqHo6Gj98ssvGjBggG7cuGFeT5s2rXr16qUePXoobdq0NkwIAEDqYFVBGzJkiEqUKKFr166pUaNGcnV1lSQ5OjqqT58+CRoQAJC4NmzYoJ49e+rYsWPmNQcHB7Vr105Dhgwx39IOAAASX4Jss4/nsc0+AHt39OhR9erVSxs2bLBY/+STTzR69GgVLVrURskAAEi94jxBmzJlitq3by83NzdNmTLlped+9913rx0MAJA4rl+/rgEDBmjBggV69nd05cuX19ixY/Xee+/ZLhwAAKlcnCdoefPm1YEDB5QhQwblzZv3xS9oMOjSpUsJFjC5YoIGwN4EBQVp9OjRmjBhgsLCwszrefLk0ciRI9W4cWM5OFi1dxQAAEgg3OKYSChoAOxFZGSkZs+eraFDh+revXvm9fTp02vAgAHq1KmT+Vlie7Zx40bdvXs33tdlzpzZvNswAAD2Ls63OHbv3j1O5xkMBo0fP97qQACAhGEymfTHH3+oT58+OnfunHndxcVFnTp1Uv/+/eXj42PDhHG3ceNG1apVy+rrN2zYQEkDACQLcS5ohw8ftvj60KFDioqKUuHChSVJ586dk6Ojo8qVK5ewCQEA8bZnzx75+vpq165dFutNmjTRyJEjX3qruj2KmZzVHfSRfHJniPN1D64EaM2wv62avAEAYAtxLmhbt241fz5hwgSlS5dO8+fPl7e3tyTp4cOHatWqlapWrZrwKQEAcXLhwgX17dtXy5cvt1ivVq2axo4dq7feestGyRKGT+4MylKYbf8BACmXVU+Djx8/Xn5+fuZyJkne3t4aPnw4tzcCgA0EBASoa9euKlasmEU5K1KkiP78809t27Yt2ZczAABSA6veqDooKMjiQfMY9+7d0+PHj187FAAgbsLCwjRlyhSNHDlSgYGB5vXMmTNr6NChatu2rZycrPqjPtkIDw5X8P1gZcgT91sfAQCwV1b9rV2/fn21atVK48ePN/9Gdu/everZs6c+++yzBA0IAHie0WjU4sWL1b9/f129etW87u7uLl9fX/Xs2VPp0qWzYcKkcff8Ha0e+KeMUUY1n9tS7p7uto4EAMBrsaqgzZw5U76+vvriiy8UGRn59IWcnNSmTRuNHTs2QQMCACxt2bJFPXv21KFDh8xrDg4OatWqlYYNG6bs2bPbMF3SMJlMOr76mLZM2qToiGhJ0tZJm1V30Mc2TgYAwOuxqqClSZNG06dP19ixY3Xx4kVJUv78+ZU2bdoEDQcA+D8nTpxQr169tHbtWov1OnXqaPTo0SpZsqSNkiWtyCcR2jRuo06tP2ley1Ikqyq3rWLDVAAAJIzXejAhbdq0KlWqVEJlAQDE4tatWxo0aJDmzp0ro9FoXi9durTGjRunGjVq2DBd0gq8+Uhrh69RgP9989qb9cvovc7V5eSSsp+1AwCkDvxtBgB2Kjg4WGPHjtW4ceMUGhpqXn/jjTc0YsQINWvWTA4OVm3Gm2ytHb5GUeFRkiRnd2fV6v2hinxQ1MapAABIOBQ0ALAzUVFRmjNnjgYPHqw7d+6Y1z09PdWvXz999913cndPPZthhIWF6eeff5YkcznLmC+jPhn+P/nkYudGAEDKQkEDADthMpn0119/qXfv3jp9+rR53cnJSR07dtTAgQOVMWNGGyZMepcuXVKjRo0sNkQpXqeEavSoKWc3ZxsmAwAgcdjFvTGhoaFau3athg8frs8++0y5c+eWwWCQwWDQkCFD4vQad+7cUY8ePVS4cGG5u7vLx8dHVatW1U8//SSTyfTK6y9evKgOHToob968cnNzU6ZMmVS7dm39/vvvr/nTAcCr7d+/X9WrV9enn35qUc4aNmyo06dPa/LkyamunP3xxx8qW7asRTkr+UkplWlYVg+uBOjO2duv/HhwJcCGPwEAAPFnFxO0ffv2qW7dulZff/DgQdWuXVsBAU//Ivbw8NDjx4+1c+dO7dy5U8uXL9eqVavk4uIS6/Vr1qxRo0aNzM94eHp66sGDB9qwYYM2bNigVq1aac6cOTIYDFZnBIDYXL58Wf369dOSJUss1itVqqRx48apYsWKNkpmO5GRkerbt6/Gjx9vXsuRI4du3Lih46uP6fjqY/F+zcyZMydkRAAAEo1dFDRJ8vb2VtmyZc0f3bp10+3bt195XWBgoD7++GMFBASoSJEi+uWXX1S+fHlFREToxx9/VLdu3bR+/Xp17dpV06dPf+56f39/ff755woNDVXlypU1d+5cFSpUyPxw/rBhw/Tzzz+rSJEi6tWrV2L86ABSoYcPH2rEiBH64YcfFBERYV4vUKCARo8erfr166fKXwpdv35dTZo00a5du8xrDRs21Jw5c7R3717dvXs33q+ZOXNm1axZMyFjAgCQeEx2ICoq6rm13LlzmySZBg8e/NJrBwwYYJJkcnd3N126dOm54yNHjjRJMjk6OprOnj373PHmzZubJJmyZs1qevjw4XPH27dvb5Jk8vT0ND148CDOP1NgYKBJkikwMDDO1wBI+cLCwkzjx483eXt7mySZPzJmzGj64YcfTBEREbaOaDPr1683ZcyY0fzfxNnZ2TRlyhST0Wi0dTQAAJKMXTyD5ujoaPW1CxYskCQ1adJEefPmfe54586d5eHhoejoaC1atMjiWEhIiPkZs2+++Ubp06d/7vq+fftKkoKCgvTHH39YnRNA6mY0GrVkyRIVKVJEPXr00MOHDyVJbm5u6tu3ry5cuKBOnTrJ2Tn1bXwRHR2twYMH68MPP9T9+0/f3yxXrlz6559/1Llz51Q5SQQApF52UdCsdfbsWV29elWSVKdOnVjP8fDwUNWqVSVJGzZssDi2c+dOPXny5KXX58mTR0WLFo31egCIi+3bt+udd97RF198ocuXL0uSDAaDWrZsqXPnzmnkyJHy8vKybUgbuXv3rj788EMNGzbMvKFT3bp1dejQIb399ts2TgcAQNJL1gXtxIkT5s9LlCjxwvNijp06deq1rj958qRVOQGkTqdPn9ann36q9957T/v37zev16xZU4cOHdK8efP0xhtv2DChbf3zzz8qXbq0Nm3aJElycHCQn5+fVq9erQwZeH8zAEDqlKwL2s2bN82f58iR44XnxRwLCgpScHDwc9d7e3u/9E1fY65/9vsBwIvcuXNH33zzjUqWLKnVq1eb10uWLKl169Zpw4YNKl26tO0C2pjRaNSYMWNUvXp13bp1S5KUNWtWbdmyRX369JGDQ7L+qwkAgNdiN7s4WuPx48fmz9OkSfPC85499vjxY3l4eFhc/7Jrnz3+7Pf7r/DwcIWHh5u/DgoKeulrAkh5QkJCNGHCBI0ZM8bil0HZs2fX8OHD9eWXX77WM7cpwYMHD/TVV19ZFNf3339fixcvVpYsWWyYDAAA+8CvKROIn5+fvLy8zB+p+bYlILWJjo7WTz/9pIIFC2rQoEHmcubh4aHhw4fr/PnzatWqVaovZ/v371fZsmUtytmAAQO0YcMGyhkAAP9fsi5o6dKlM38e8ybTsXn22LPXxHz+smufPf7stf/Vt29fBQYGmj+uXbv28vAAkj2TyaQ1a9bozTffVLt27cy36zk6Oqpjx466ePGi+vfv/8opfUpnMpk0bdo0ValSRVeuXJEkZciQQWvXrtX333+f6osrAADPSta3OGbPnt38+Y0bN+Tp6RnreTdu3JAkeXp6mm9vfPb6hw8f6smTJy98Di3m+me/33+5urrK1dU1fj8AgGTr0KFD6tmzp7Zs2WKxXq9ePY0aNUqFCxe2UTL78vjxY7Vr105Lly41r1WsWFFLly7lTgMAAGKRrCdoz+68+OyOjP8Vc6xYsWKvdX3x4sWtygkg5bh69apatGihcuXKWZSzt956Szt27NDKlSspZ//f8ePHVb58eYty1q1bN23bto1yBgDACyTrglaoUCHlypVLkrRu3bpYzwkJCdE///wjSapVq5bFsSpVqpinZi+6/sqVKzp9+nSs1wNIPQIDA9WnTx8VKlRICxcuNK/ny5dPS5cu1Z49e8zvuQhp3rx5evvtt3Xu3DlJT+9g+P333zVhwgS5uLjYOB0AAPYrWRc0g8GgL7/8UpL066+/mt8A9lnTpk1TcHCwHB0d1axZM4tjadOmVYMGDSRJM2bMUGBg4HPXjx49WtLT58/q1auXsD8AALsXERGhyZMnK3/+/Bo9erR5t1YfHx9NnDhRp06d0ueffy6DwWDjpPYhNDRUbdq0UatWrfTkyRNJUpkyZXTo0CF99tlnNk4HAID9s5uC9vDhQ92/f9/8YTQaJT39y/7Z9We3rpYkX19fZc2aVaGhofroo4908OBBSU//UTVjxgwNHDhQktS+fXsVKlToue87bNgwpU2bVrdu3dInn3yi8+fPS3o6eRs2bJhmzpwp6elOY97e3on28wOwLyaTSb/99puKFSumrl27KiAgQNLT50179uypCxcuqGvXrjx7+oxz587pnXfe0dy5c81r7du317///qv8+fPbMBkAAMmHwWQymWwdQpLy5Mlj3t3rZVq2bKl58+ZZrB08eFC1a9c2/wMqXbp0CgsLU2RkpKSntyauWrXqhf+QWrNmjRo1amTerdHLy0vBwcGKjo6WJLVq1Upz5syJ12/Ig4KC5OXlpcDAwBduXgLAPu3atUu+vr7as2ePxXqzZs00YsQI5c6d20bJ7Ndvv/2mNm3aWLy/5KxZs9S8eXMbJwMAIHmxmwna6yhXrpxOnjypbt26qWDBgoqMjFTatGlVpUoV/fjjj1q7du1Lf8tdt25dHTt2TO3atVOePHkUFhYmb29v1axZU8uXL9fcuXO5fQlIBc6dO6fPPvtMVapUsShn1atX14EDB7Rw4ULK2X9ERETou+++0+eff24uZ0WLFtX+/fspZwAAWMFuJmgpDRM0IPm4d++e+ZbmqKgo83qxYsU0ZswY1a1bl1/SxOLKlSv6/PPPtW/fPvNas2bNNHPmTIu3NAEAAHGXrN8HDQBeR2hoqCZNmqRRo0aZpz+SlDVrVg0bNkytWrWSkxN/TMbm77//VosWLfTw4UNJkouLi6ZMmaL27dtTZgEAeA38ywNAqhMdHa1ffvlFAwcO1PXr183radOmVc+ePdWjRw8mQC8QFRWlgQMHatSoUea1fPny6bffflPZsmVtmAwAgJSBggYgVdmwYYN69eqlo0ePmtccHBzUtm1bDRkyRNmyZbNhOvt28+ZNNW3aVDt27DCv1a9fX3PnzlX69OltFwwAgBQkRWwSAgCvcuzYMdWuXVu1a9e2KGcff/yxjh8/rlmzZlHOXmLLli0qU6aMuZw5OTlp/Pjx+v333ylnAAAkIAoagBTtxo0bat26tUqXLq0NGzaY18uVK6ctW7Zo9erVKlasmA0T2jej0ajvv/9eNWvW1N27dyVJOXPm1Pbt29W9e3eeNwMAIIFxiyOAFCkoKEijR4/WxIkT9eTJE/N67ty5NXLkSDVp0kQODvyO6mXu37+v5s2ba/369ea12rVra+HChcqYMaMNkwEAkHJR0ACkKJGRkZo9e7aGDh2qe/fumdfTp0+v/v37q1OnTnJzc7NhwuTh33//VePGjc2bqDg4OGjo0KHq168fxRYAgEREQQOQIphMJv3xxx/q06ePzp07Z153dnZWp06d1L9/f2XIkMGGCZMHk8mkiRMnqnfv3ub3hMucObOWLFmi999/38bpAABI+ShoAJK9vXv3ytfXVzt37rRYb9y4sUaOHKl8+fLZKFny8ujRI7Vu3VorV640r1WrVk1LlixR9uzZbZgMAIDUg/tUACRbFy9eVOPGjfXOO+9YlLOqVatqz549+vXXXylncXT48GGVK1fOopz17t1bmzdvppwBAJCEmKABSHYCAgL0/fffa/r06YqMjDSvFy5cWKNHj9ann37K7oJxZDKZNHv2bHXp0kXh4eGSJG9vby1YsEAff/yxjdMBAJD6UNAAJBthYWGaMmWKRo4cqcDAQPN65syZNWTIELVt21bOzs42TJi8BAcH6+uvv9aiRYvMaxUqVNCyZcuUJ08e2wUDACAVo6ABsHtGo1GLFy9W//79dfXqVfO6u7u7evTooV69eildunQ2TJj8nDp1Sg0bNtTp06fNa507d9bYsWPl6upqw2QAAKRuFDQAdm3Lli3q2bOnDh06ZF4zGAxq1aqVhg0bphw5ctgwXfK0cOFCdejQQaGhoZIkDw8PzZkzR59//rmNkwEAAAoaALt08uRJ9erVS2vWrLFY//DDDzVmzBiVLFnSRsmSr7CwMHXp0kWzZ882r5UsWVLLly9XoUKFbJgMAADEYBdHAHbl1q1bateunUqVKmVRzkqXLq2NGzdq7dq1lDMrXLx4URUrVrQoZ61atdKePXsoZwAA2BEmaADsQnBwsMaOHatx48aZb72TpJw5c2rEiBFq3ry5HBz4nZI1VqxYoVatWikoKEjS02f3pk2bplatWtk4GQAA+C8KGgCbioqK0pw5czR48GDduXPHvO7p6am+ffuqS5cucnd3t2HC5CsiIkJ9+vTRxIkTzWuFChXS8uXLmUICAGCnKGgAbMJkMumvv/5S7969LXYSdHJy0jfffKOBAwcqU6ZMNkyYvF27dk2NGzfW7t27zWuNGzfWjz/+yI6XAADYMQoagCR34MAB+fr6avv27RbrDRo0kJ+fnwoWLGijZCnDunXr1Lx5cwUEBEiSnJ2dNXHiRHXs2JE38AYAwM7xQAeAJHP58mV98cUXqlChgkU5q1ixonbt2qXly5dTzl5DdHS0Bg4cqLp165rLWe7cubVr1y59++23lDMAAJIBJmgAEt3Dhw81YsQI/fDDD4qIiDCvFyhQQKNGjdJnn31GeXhNt2/f1hdffKGtW7ea1z7++GPNnz9fPj4+NkwGAADig4IGINGEh4dr2rRpGj58uB4+fGhez5AhgwYPHqwOHTrIxcXFhglThu3bt6tJkya6ffu2JMnR0VEjR46Ur68vO18CAJDMUNAAPOf27dvy9/dXxYoVrbreZDJp6dKl6tevn/z9/c3rbm5u6tq1q/r06SMvL6+EiptqGY1GjRkzRv3795fRaJQkZcuWTb/++quqVatm43QAAMAa/GoVgIXTp08rX/78qlKliiZPnhzv63fs2KG3335bTZs2NZczg8GgL7/8UufOnZOfnx/lLAE8ePBAn376qfr27WsuZzVq1NCRI0coZwAAJGNM0ACYnT59WuXKl5eLq5sKFSul7t27S5K6dOnyymvPnDmj3r17a9WqVRbrH3zwgcaMGaMyZcokSubUaN++fWrUqJGuXr0q6WkBHjhwoAYNGiRHR0eLc3///XctX75cgYGB8vT0VP369dW4cWNbxAYAAHFgMJlMJluHSImCgoLk5eVl/kcRYO+eLWezf12rbDneUPd2TXRo705NmDDhhSXtzp07GjJkiH788UdFR0eb10uUKKGxY8eqdu3abACSQEwmk6ZOnaoePXooMjJSkpQxY0YtWrRItWrVMp9nNBrVs2dPzf3pJz0KCpKDwaA0zo56EhmtaJNJXh4eavHVV5o8eTLPqAEAYGcoaImEgobk5L/lLG+BwpKksCehLyxpISEhmjBhgsaMGaPg4GDzevbs2fX999+rZcuWz01zYL2goCC1bdtWv/32m3mtUqVKWrp0qXLmzGlei4qKUpkyZXTixAlVzZ1Fnd4upk+L5JKzo4Oioo36+9w1Tdt3Wlsu3VLhQoV07PhxNmoBAMCOUNASCQUNycWLylmM/5a0Tp06ad68eRo0aJBu3rxpPs/Dw0O9e/dWt27dlDZt2qT+MVK0o0ePqlGjRjp//rx5rUePHvLz85Ozs7PFuRUrVtTePXs07eOKal+hyAtfc8GR82qzcqfeLP2mDh8+kljRAQBAPFHQEgkFDcnBq8pZjLAnoerWtrEO7d2prFmz6tatW+Zjjo6Oat++vQYPHqwsWbIkVfRUY+7cufr2228VFhYmSfLy8tK8efNUr169587duHGjatWqpREflFPvqqVe+dpTdp9U93X7tHz5cjVo0CChowMAACvw8AGQSsW1nEnSlUvnJZlkMpksytn//vc/nThxQtOnT6ecJbDQ0FC1atVKbdq0MZezsmXL6tChQ7GWM0nq27evvNxc1OWdYnH6Ht+8VVSZ0rpp8ODBCRUbAAC8JnZxBFKhuJaz2zevacb44Vr35zL9d9jeuXNnTZkyJSnipjpnz55Vw4YNdeLECfPa119/rYkTJ8rNzS3WayIiInT08GF9904xuTnH7Y92Z0cHdShfRH47jiooKIhpPwAAdoAJGpDKxKWcBT8O1NQxQ9SwRnmt/WOpuZxlfyO3Bo+doXLvVNW0adOsep80vNzSpUtVvnx5czlLmzatFi1apBkzZrywnEnSxYsXFWU0qnKuzPH6fu+8kUnRJpNOnz79WrkBAEDCYIIGpCL+/v6vLGfrVy3XuGG9FPjwgXnN0yu9WnfqqYbN2srF1VU16vxP3ds1Uffu3eXi4qJvvvkmKX+MFCk8PFzdu3fX9OnTzWvFihXT8uXLVbRo0ThdL0mO8dw239nx6fkhISHxug4AACQOJmhAKrJhwwY9CQ1V1fc/VJ78hWI9xz1NGnM5c3ZxUfO2nbVi6xF90fpbubi6SpLc3NOoYfM2MhqNWrRoUZLlT6n8/f1VpUoVi3LWokUL7du3L07lTJLy5Mnz9LUePo7X97788OlbJBQoUCBe1wEAgMTBLo6JhF0cYa+aNGmipUuXqkmrb9St/8jn3kTaZDL9v/buOzqqam/j+DPpk0qAhNCb9HIJRa8FAREEBPEVQZAmICCCKFK8eKXaBQSxUgTEAlL0ehVRgtKLFCkCkRpAEQiB9J7Mef/gZkxIhyQzmXw/a81i5py9z/xmsleYJ/ucPXq6fw9VrFRZo8a/pCrVamY7xvZNP2riU/0VGBioE8ePy9vbu6TKdzjffvutBg0apKioKEmSu7u73n33XT355JOF/oLvKpUryz89SYdGP1zgvncs+K/OJ0vhERGFLR0AABQDTnEEypiVK1de/3fph5KULaSZTCbNX/aVXHP58mLCWdFITU3VSy+9pLfeesu6rW7dulqzZo1atGhxU8d8ZuxYvfjii9oUdlH31amSb/tdf4Rr/19XNXny5Jt6PgAAUPSYQSsmzKDB3uU3k5YTwlnRuHDhgvr27avt27dbtz3yyCNasmSJ/Pz8bvq4aWlpqljeXy7pqdo27EHVr5j7scIiY3XP4nWKt5h05erVPBcgAQAAJYdr0IAyauXKlXrssce0cumHmvvqi9mW0b8R4axobNy4UcHBwdZw5uLionnz5mnNmjW3FM4yjrVj124lpEt3LvxWH/wSqpiklCxt4pJTtXDv77pjwbeKTknTlm3bCGcAANgRZtCKCTNoKC0KMpNGOLt16enpeuWVVzRjxgxrGK5evbpWrVqlf/7zn0X6XGFhYWrfrp3++OMPubs4q2u9aipvdldUUorWn/xTialpqlKlikI2bizwIiQAAKBkcA0aUMbld00a4ezWXblyRQMGDNCGDRus27p27arly5erYsWKRf58tWvX1rnz57Vv3z5NmDBBWw4fUlpqqlxdXRV8+x2aNWuW7rzzziJ/XgAAcOsIaAByDWmEs1u3Y8cOPfbYY7pw4YIkycnJSTNnztTkyZPlVMjvLCus1q1ba/PmzcX6HAAAoGgR0ABIyh7Sbr+7PeHsFhiGobffflsvvPCC0tPTJUmVKlXSihUr1KFDBxtXBwAA7BXXoBUTrkFDaZVxTZrJZFJQ5cqEs5sQGRmpIUOG6JtvvrFua9eunVasWKHKlSvbsDIAAGDvmEEDkMXKlSvl6empvXv3ateuXYSzQtq/f7969+6tsLAw67YXX3xRM2bMkIsLv3IBAEDemEErJsygAWWLYRj66KOP9Nxzzykl5frS9uXLl9enn36qbt262bg6AABQWvDnXAC4RXFxcRoxYoRWrFhh3XbHHXdo1apVqlGjhg0rAwAApQ1fVA0At+DIkSNq06ZNlnD27LPPauvWrYQzAABQaMygAcBNWr58uZ566iklJiZKknx8fLRkyRI9+uijNq4MAACUVgQ0ACikxMREjR07VosXL7Zu+8c//qHVq1erXr16NqwMAACUdpziCACFcPLkSd15551ZwtmTTz6pXbt2Ec4AAMAtI6ABQAGtXbtWrVq10qFDhyRJZrNZy5Yt06JFi2Q2m21cHQAAcASc4ggA+UhJSdGkSZP0zjvvWLc1aNBAa9asUdOmTW1YGQAAcDQENADIw/nz59WnTx/98ssv1m19+/bVwoUL5ePjU2J1hISEKDw8vFB9AgMD1alTp2KqCAAAFAcCGgDkYv369RowYICuXbsmSXJzc9O8efP01FNPyWQylVgdISEh6ty580313bBhAyENAIBShIAGADdIS0vTtGnT9Nprr1m31apVS6tXr1br1q1LvJ6MmbO6I26XuYpvgfok/hWj0wv3FHrWDQAA2BYBDQAyuXjxoh5//HFt3rzZuq1nz55aunSp/P39bVeYJHMVX3nVsm0NAACgeLGKIwD8z+bNmxUcHGwNZ87Ozpo1a5a+/vprm4czAABQNhDQAJR5FotFr732mjp27KjLly9LkqpWraotW7ZowoQJJXq9GQAAKNs4xRFAmXb16lUNHDhQ69evt27r1KmTPv/8cwUEBNiwMgAAUBYxgwagzNq9e7eCg4Ot4cxkMmn69Olav3494QwAANgEM2gAyhzDMDR//nxNmDBBaWlpkqSAgAB9/vnnLEkPAABsioAGoEyJjo7WsGHDtHbtWuu2e+65RytXrlTVqlVtWBkAAACnOAIoQw4ePKjWrVtnCWcTJ07Uzz//TDgDAAB2gRk0AA7PMAx9/PHHGjNmjJKTkyVJ5cqV0yeffKKHHnrIxtUBAAD8jYAGwKHFx8dr1KhR+vTTT63bWrVqpdWrV6t27do2rAwAACA7TnEE4LBCQ0N1xx13ZAlnTz/9tHbs2EE4AwAAdokZNAAOacWKFRo+fLji4+MlSV5eXlq8eLH69u1r48puXuJfMcXSFgAA2A8CGgCHkpSUpOeff14ffvihdVvTpk21evVqNWzY0IaV3bzAwEBJ0umFe266LwAAKB1MhmEYti7CEcXExMjPz0/R0dHy9fW1dTlAmXDmzBn17t1bv/76q3Xb4MGD9cEHH8jT09OGld26kJAQhYeHF6pPYGAg3+sGAEApQ0ArJgQ0oGR98803Gjx4sKKjoyVJHh4eev/99zV06FAbVwYAAFBwLBICoFRLTU3VhAkT9PDDD1vDWb169bR7927CGQAAKHW4Bg1AqfXnn3+qb9++2rFjh3Xbo48+qo8//piZawAAUCoxgwagVNqwYYOCg4Ot4czV1VXz58/XqlWrCGcAAKDUIqABKFXS09M1bdo0denSRREREZKkGjVqaNu2bXrmmWdkMplsXCEAAMDN4xRHAKVGeHi4+vfvr40bN1q3devWTcuXL1eFChVsWBkAAEDRYAYNQKmwbds2tWjRwhrOnJyc9Prrr+vbb78lnAEAAIdBQANKwI4dO9ShQwddunTJ1qWUOhaLRW+99ZY6dOigixcvSpKCgoL0888/61//+pecnPg1BgAAHAenOALFbMeOHbrvvo5KSUlWo0aNFBoaqqCgIFuXVSpERkZq8ODB+vbbb63bOnTooC+++IL3EAAAOCT+9AwUo4xw5luhop5/a4ESk66HNGbS8rd37161bNkySzh76aWXFBISQjgDAAAOi4AGFJPM4Wzmkq91Z+cemvLRSkJaPgzD0Pvvv6977rlHZ8+elSRVqFBB69ev18svvyxnZ2fbFggAAFCMCGhAMbgxnAVUriZJatCiDSEtD7GxsXr88cc1ZswYpaSkSJL++c9/6sCBA+rSpYuNqwMAACh+BDSgiOUWzjIQ0nL222+/qXXr1lq5cqV127hx47RlyxZVr17dhpUBAACUHAIaUITyC2cZCGlZLVu2THfccYdOnDghSfL19dXatWv19ttvy83NzcbVAQAAlBwCGlBEChrOMhDSpISEBA0bNkxDhgxRYmKiJCk4OFi//vqrHnnkERtXBwAAUPIIaEARKGw4y1CWQ9qJEyf0z3/+U0uWLLFuGzFihHbu3Km6devasDIAAADbIaABt+hmw1mGshjSVq9erdatW+u3336TJHl6eurTTz/VggUL5OHhYePqAAAAbMdkGIZh6yIcUUxMjPz8/BQdHS1fX19bl4NiEhcXpwoVK8qnXAW9vOw/hQ5nmR0/uFczRz6m8v7ldPHixSKs0n6kpKRowoQJevfdd63bGjVqpDVr1qhx48Y2rAwAAMA+MIMG3AIPDw95mj0VFxOlyCuXb+lYly+cV0pKsgIDA4uoOvty7tw5tW3bNks469+/v/bs2UM4AwAA+B8CGnALXFxcFBp6TB5ubpo58jGdOLz/po6zdd1avfvSWNWvV0/799/cMezZunXrFBwcrD179kiS3Nzc9NFHH+nTTz+Vt7e3jasDAACwHwQ04BYFBQXp999DbzqkZQ5nR48elYuLSzFVWvLS0tI0efJkde/eXZGRkZKkOnXqaNeuXRo5cqRMJpONKwQAALAvBDSgCNxsSHPkcHbx4kV17NhRb7zxhnXbww8/rP3796tly5Y2rAwAAMB+EdCAIlLYkObI4eznn39WixYttHXrVknXTwWdM2eOvvrqK5UrV862xQEAANgxAhpQhAoa0hw1nFksFr3yyivq1KmTwsPDJUlVq1bVli1b9Pzzz3NKIwAAQD4IaEARyy+kOWo4i4iIULdu3TRlyhRZLBZJ0gMPPKADBw7orrvusnF1AAAApQMBDSgGuYU0Rw1nu3btUnBwsH788UdJkslk0syZM/X9998rICDAxtUBAACUHnxRdTHhi6ohSZcuXVLDho2UlJysBweM0Fcfz3eocGYYhubNm6dJkyYpLS1NkhQYGKgvvvhCHTt2tHF1AAAApQ8BrZgQ0JAhI6RFR0epfv36DhPOoqKiNHToUH399dfWbffee69WrFihKlWq2LAyAACA0otTHIFilnG646RJkxwmnB04cECtWrXKEs5eeOEF/fTTT4QzAACAW8AMWjFhBg2OyDAMLVy4UM8++6ySk5MlSf7+/lq+fLm6d+9u4+oAAABKv9L/p3wAJSIuLk5PPfWUPv/8c+u2Nm3aaNWqVapVq5btCgMAAHAgnOIIIF/Hjh3T7bffniWcPfPMM9q2bRvhDAAAoAgR0ADk6bPPPlObNm0UGhoqSfL29taXX36p+fPny93d3cbVAQAAOBZOcQSQo6SkJD377LNauHChdVuzZs20Zs0a1a9f34aVAQAAOC4CGoBsTp8+rd69e+vAgQPWbUOGDNF7770nT09PG1YGAADg2DjFEUAWX3/9tVq1amUNZ2azWUuWLNGSJUsIZwAAAMWMGTQAkqTU1FS98MILmjt3rnVb/fr1tWbNGjVr1syGlQEAAJQdBDQA+uOPP/TYY49p165d1m19+vTRokWL+B4/AACAEsQpjkAZ98MPPyg4ONgazlxdXfXee+9p5cqVhDMAAIASRkADyqj09HRNmTJF3bp109WrVyVJNWvW1I4dOzR69GiZTCYbVwgAAFD2ENAABxIREaGhQ4eqUmCg/Ly8VM7HR3Vq19Y777wji8VibXf58mV17txZr7zyigzDkCR1795dv/76q9q0aWOr8gEAAMo8k5Hx6QxFKiYmRn5+foqOjuY0MRQ7i8Wibt266acNG5RuGGpb0Vd1vDyUZjG081qMzsQny9PdXa+9+aaCg4PVt29fXbx4UZLk7Oys1157TRMmTJCTE3+zAQAAsCUCWjEhoJU+KSkpcnNzs3UZhWaxWNS4USMdP3FCo+oEaVSdyqrj5WHdbxiGtkTE6JXQ89odGSeTyWSdNatcubJWrlype++911blAwAAIBP+XA5ImjVrljzMZnXs2NHWpRTagw8+qOMnTmhh8G2a1ax2lnAmSSaTSc39POXren3R1oxw1rFjRx08eJBwBgAAYEeYQSsmzKCVHrNmzdIL//qX/IOq69pf59S+fXtt2rTJ1mUVyLVr11SpYkUNq1VJc5rXzrHNvshYDdx7Qn8kpli3lStXThEREXJ2di6pUgEAAFAAzKChTMsIZ3WD79TYxd+r09DntXnzZnXo0MHWpRXICy+8oDTD0NN1grLtMwxDH565qE7bjlrDWQU3Fw2uEaCYqCidOnWqpMsFAABAPviiapRZmcPZgJcXytXNXe0ff1qSFLLkbXXo0MHuZ9K+/e9/dXcFH9Xy8tBficn6KylFFxKv37ZERGvdpUhr23+W99EnrevJ3clJn5y/olmzZmnx4sU2rB4AAAA3IqChTMopnGWwt5CWlJSkCxcu6MKFC/rzzz/1559/Wu9HREQo1iSV/3a30vM4WXls3cqa0biGXP+3SqOns5N1FUcAAADYDwIaypy8wlmGkghphmEoJiYmS+DKKYRlfIl0bhLy2Ofn4qyPWt6mHpXLZ9mebhhyd8/+ugEAAGBbDhHQli1bpiFDhuTbLiQkRPfff3+O+06fPq233npLGzZs0MWLF+Xj46OWLVtqxIgR6tWrV1GXDBspSDjLcCshzWKx6MqVK7mGrox/4+Pjb/k1OUlq7uelqmY3VfZwU1Wzm6p4XL+18ve2rt6Y4VBUvJIthpo3b37Lzw0AAICi5RABLYOTk5MCAgJy3Z/bjMH333+v3r17KyHh+lyEr6+vrl27pg0bNmjDhg0aMmSIPv74Y5lMpmKpGyWjMOEsQ04hLSUlRRcvXswzfP31119KTU29pXpdXFxUpUoVVatWTdWqVVPVqlWz/fvtt99q7NixmtG4hjoGlivQcRedvSQ3Z2f961//uqX6AAAAUPQcKqBVr15dZ8+eLVSfsLAw9enTRwkJCbr77ru1ZMkS1a9fX3FxcZo1a5ZmzpyppUuXqmHDhpo0aVLxFI5iV9BwlpwYr5iIy4q5ckkxEZcVHXFRMRGXFVCjrjZv3iw3N7dbDl6S5OnpmS143Ri+AgMD5eSU90Kro0eP1uRJk/TK73/ongq+cnfOu31oTIK++OOK7m7XXh4eHnm2BQAAQMlzqIB2M6ZOnar4+HgFBQXpu+++U7ly5SRJ3t7emjFjhi5duqSFCxfq1Vdf1fDhw+Xv72/bglFoc+bMyTecffLiMJ0/+quS4mPzPFZBwlmFChVynO3KHML8/PyKZEbWyclJb82Zo2dGj9agfSe0rHU9mXP5brPQmAT12HVMrm5uWr58+S0/NwAAAIpemQ5o8fHxWrt2rSRp1KhR1nCW2eTJk7Vw4ULFxMToP//5T4GudYN9mT5jhpxdXPXoC7NznTlLSUrIN5yZnJzk5OSs9LRUdenSRfXq1csWvqpUqSKz2VwcLyNXTz/9tC5duqRXX3lZzTce1MjaQRpUM1CB7q4yDEOHoxO0MOySVvx5RS6ubtqyfbuqVatWojUCAACgYMp0QNu+fbsSExMlSV27ds2xTa1atdSoUSOFhoZar0dD6bJwwQINGDhQy//9pIbO+lRmb99sbcpXrqmYK5fkGxAkv4qV5VuxknwDguRboZL8AoLkUyFQW1cs0C/ffq7+/fvrs88+s8Eryd3MmTPVqlUrTZgwQTNDT2l66HmZnZ2UZjGUahhyc3ZW2w73admyZYQzAAAAO+ZQAe3KlStq1aqVjh8/rvT0dFWuXFl33XWXnnzySbVv3z5b+yNHjljvN23aNNfjNm3aVKGhoTp69GhxlI1i1q9fP0nSgIEDtWTiwBxDWq+Jb+Ta3zAMfffeTLsNZxl69uypnj176vTp05o1a5YuXrwoV1dXNWvWTC+88ALXnAEAAJQCDhXQEhIS9Ouvv8rf31/x8fEKCwtTWFiYPv/8cw0ZMkQLFy6Ui8vfL/mvv/6SJPn7++d5WlrVqlWztM9JcnKykpOTrY9jYmJu9eWgCBUkpOUkI5zt/uZTuw5nmdWtW1cfffSRrcsAAADATch7ybdSokqVKpo2bZoOHTqkpKQkXbt2TQkJCdqxY4f1e8+WLl2qcePGZekXG3v9miNPT888j5+xP6N9Tl5//XX5+flZb9WrV7+Vl4Ri0K9fP3326ae6dOZ3LZk4UIlxeYfo0hjOAAAAULo5REDr3Lmzpk+frubNm1u/68zZ2Vl33XWXfvzxR/Xs2VOS9MEHH+jkyZPFUsPkyZMVHR1tvf3xxx/F8jy4NQUNaYQzAAAA2IJDneKYEycnJ82ePVvffPONLBaLvv32Wz3//POSJB8fH0myfkF1bjL2Z7TPibu7e65fhA37kt/pjvYWzkJCQhQeHl7ofoGBgerUqVMxVAQAAIDi4vABTZJuu+02VaxYURERETpz5ox1e5UqVSRJkZGRSkxMzPU6tAsXLmRpj9Ivt5Bmj+Gsc+fON91/w4YNhDQAAIBSpEwEtNxkXrnxyJEjatOmTY7tMlZ7bNKkSYnUhZKRLaS9tVwbl82zm3AmyTpz5t69tkwVCr4Ko3E1Scnfhd3UzBsAAABsp0wEtNOnTysiIkKSVLt2bev2e+65R2azWYmJifrhhx9yDGjnzp1TaGioJN3STAZsx2KxyMkp58stM4e0twd3VEJMlN2Es8xMFTzkHORV4PbpxVgLAAAAik+pXyTEMIx890+cOFHS9evRunfvbt3n5eWlXr16SZI+/PBDRUdHZ+v/5ptvSrp+/dnDDz9cRFWjpAwdOlSubm569dVXc22TsXBIWnKiBg4caHfhDAAAAGVHqQ9o586d0+23364FCxbozJkz1sBmsVi0e/dude3aVV9//bUkaeTIkWrQoEGW/jNnzpSXl5cuXryoHj16WFd5jI+P18yZM63fJ/XSSy/J39+/BF8ZbtXQoUO1dOlSmf0qasqUqfmGtOSkJC1fvrwEKwQAAACycohTHPfu3au9e/dKur6aoo+Pj2JjY7N8cfSQIUM0f/78bH1r166tVatWqXfv3tq2bZvq168vPz8/xcXFKT093do3YxYOpUNGOGvSuY/uHfaivn9rrKZMmSpJ+ve//23j6gAAAICclfoZtEqVKundd9/V448/rsaNG8vX11dRUVFydXVVw4YNNXToUG3fvl1LliyRi0vOebRbt246fPiwhg8frlq1aikpKUn+/v7q1KmT1qxZoyVLlshkMpXwK8PNyhzOOoycJhc3dz34wruq0fKefGfSAAAAAFsq9TNoZrNZY8aM0ZgxY27pOHXr1tXChQuLqCrYyo3hzPS/xUGcXd304Avvat2bzzCTBgAAALtV6mfQgAy5hbMMGSGNmTQAAADYKwIaHEJ+4SwDIQ0AAAD2jICGUq+g4SwDIQ0AAAD2ioCGUq2w4SwDIQ0AAAD2iICGUutmw1kGQhoAAADsDQENpVJSUpKWLlsm/6q11X7E1EKHswzOrm7qMv5tuXl5a8bMl4u4yqJjXE1S+qX4At+Mq0m2LhkAAAA3odQvs4+yycPDQ0OeeEJLly7V5oUzb2oGLTU5Ued+3aZfVr6n5LgY9ejRQxaLRU43GfaKQ2BgoCQp+buwW+oPAACA0sFkGIZh6yIcUUxMjPz8/BQdHS1fX19bl+OwbuY0x5jwCzr03XL9vnGtkhLjs+zzcHVVtx49tGDBAlWsWLG4yi6UkJAQhYeHF7pfYGCgOnXqVAwVAQAAoLgQ0IoJAa3kFCakXTiyR9+/Nkruqal61MtTvXx9VM3FRemSDiUlaUVMjELiE+Tl6alde/aoSZMmJfdCAAAAUOZxiiNKvSVLlkiSli5dKkm5hrTw00f13csj1MLFWfOrVZFPpjbOktqYzWpjNutYcrKGX7ykf7Zpo9Nnz3KaIAAAAEqM/VxsA9yCJUuWaMiQITq6YZU2LZghw2LJ1mbrh9NV28mkDwIDsoSzGzV2d9fiykFKTkzUgAEDirNsAAAAIAtm0OAw8ppJCz99VBdPH9GLlQJlLsB1ao3c3fWgt7e+//lnpaSkyM3NrfgKBwAAAP6HgIZSL/MiGh07dtTp06e1dcMqJcZEqnWvETKZTNq7+iMFOLvoXk/PbP2PJ6coLDVFXby9s2zv5+er/8TF6Y033tDUqVNL5LUAAACgbCOgoVQLCQlR586dc9x3ZneIzuwOkXT9XN4uXl5yMZms+y+mpem9a5H6Ji5OniaT2pjNquDsbN3f1N1dbiaTDhw4UKyvAQAAAMhAQMNNSUlJ0RtvvKFt27YpPj5efn5+GjBggPr371+idViXn2/iL3nlPpxNRyPl9r9wFmuxaHFUlD6NjlHy/xYxjTcMfRYdrWfLl8/Sz1XXvxQbAAAAKAkENBRKXFyc+vbtqw0//KDU9HS5SXKRSSky9MMPP+ipESP05IgRmjt3bskW5uUi+eZ+nVi62UV/xKXq0+hofRQZpahMi4j4OjlpZLly6ufrk6VPrMWiBMNgFUcAAACUGAIaCuzSpUtq0qiRIqOidJtc1ESeCpSTTDLJIkPnlaYjCcmaN2+etm3bpj179sipAAtyFDvDkNydtD8iQfuTkq2bXSX19/PV8HLlVC7TqY0Z/hsbK0kaN25cSVUKAACAMo6AhgJJS0tT82bNFR0Vpe4yq8oNQ8dJJtWSq2rJVUeUoh3796tz587auHGjjSr+n2vJ0qloKSY1y+bu3l4a6++vqq6uOXZLNQx9Fh2jalWrqkWLFiVQKAAAAMD3oKGAZs6cqSsRV9Qph3B2o6ZyU7Dc9NNPP+nw4cNF8vzff/+93Nw9dP/99xesQ6pFOnhV+jUiWzh73t9fbwYG5hrO0g1D069E6I+0NM14+eVbLR0AAAAoMAIaCuSjDz9UBTmphrKfCpiTf8hNTpLGjx9/y8/9/fffq0fPnnJy99RPP/2kDh065N0h/X/hLCLT4h7eLlKL8lKgh+ZGRuqda9d0JS0tW9ejyckafemy/hMXpxEjR2rIkCG3XD8AAABQUJziiHwdPHhQVyIi1E7uMsmUfwdJ7jKpnly0ZdOmW3rujHDmFVBdHSYvUtjW/2jzmvfUoUMHbcrp2BZD+i1Sik65/tjVSarnK1X2lEwmqYKHjNOxWnw+RoujotXO01M1XF2UZkgHkpJ0LCVFbs7OmjJlimbOnHlLtQMAAACFRUBDvvbv3y9JCijg7FmGQDnreHqyLBbLTS0WcmM48/Atr0bdh0qSNaQ9+eSTf3cwDOl41N8zZ84mKbhC1tUdTSbpNl9ZyrtJv17VHotFuxMS5WQyyce/nF4aPkJTpkyRm1vuK0ICAAAAxYWAhnylpFyfjSrY3NnfMtqnpaUVOvDkFM4yZA5ply9f/rtTWKx0IeHvJ/9H+dyX3ne5HhgXLF5c4t/dBgAAAOSGgIZ8NWjQQJIULYvKF2IWLUoWOZtMRRrOMmSEtN/WvHd9Q3iidDbu7wZN/KXyHoV6XgAAAMDWWCQE+brvvvvkaTbrmFLzb/w/6TJ0XKlq2rx5oZ6rIOEsQ6PuQ3Vbxz7XH2QOZ/V8pSDPQj0vAAAAYA8IaCiQPo89pj+VrmtKL1D7k0pVsqRXXnmlwM9RmHCWwb9W46wbanhLNX0K/JwAAACAPSGgoUDmzJkjd1dXrVeiYmXJs+1fStM2JatmjRrq3r17gY5/M+Es5q8wHfxi9t8byrtLlTykmJT8b/HZl9gHAAAAbI1r0FAg5cuX10+bNqlDu3b6Kj1BreSmenKVe6alQ2JlUahSdVApKleunA4eOlSgY99MOEuMvKKtc8YoNTHTqY3Xkq/fCiEwMLBQ7QEAAIDiZDIMw7B1EY4oJiZGfn5+io6Olq+vr63LKTJHjhxR1y5d9OeFC3KWVFnOcpNJCbLokiwySWrVurU2bdokb2/vAh0zKChIl8OvqOvrX8knqEa+7VMSYrX59eGK+uOEJMmncm21HvyiXDw8deCL2Yo4cUD//ve/1ahRozyPExgYqE6dOhWoRgAAAKAkENCKiaMGtAy//vqrJk6cqNDQUCUnJcns6al27dppzpw5CgoKKtSxVqxYoQEDB8q3Wj21n7RAbl65X0OWnpqibW+PUXjoPkmSV8Uquu+lpTKXC9DJjV/qwGdvqlWrVtqzZ89NffcaAAAAYEsEtGLi6AGtqBUkpBkWi3Z/9KL+2LNBkuTmXU73/XuJfCvXIpwBAADAIfApFnahX79++uzTTxXz50ltfmukUuJjs+w3DEMHV75tDWfObh5qO+4dwhkAAAAcCp9kYRfmzZunp58erXHPPZdjSDv+w6c6ueELSZLJyVl3Pv2GKtRtRjgDAACAQ+HTLGxu3rx5Gvf8eEXHxmruO+9kC2nndn6vw1/Os7ZvNfjfqtLiXsIZAAAAHA7XoBUTrkErmIxwVr5eSzUZ+JJ+ff85JUb8qXHPPqu58+bJXKGKEiL+kmG5/gXZTf5vlJr0HE44AwAAgEPiUy1sJnM4a/HULHn4V1LrZz+QuWI1zX3nHT3er5/iw/+whrO6HR5V44eeJJwBAADAYTGDVkyYQcvbjeHM2c3Dui8lNlJ75gxXYsQF6zZXs7e6vfVfnf/lR8IZAAAAHJaLrQtA2fPyyy9r6rTp8s8hnFll+rtBlSpVdPHSJYVMf1wJVy8RzgAAAOCw+ISLEpU5nAXnEM7SkxN14KPxSrz6lyTJ5Oyqi5cv6/lx45QUFaE2bdoQzgAAAOCw+JSLEpNfOLOkp+nwkpcUcy5UkuReLkC3T1gkz4DqmvvOO1qzehXhDAAAAA6NT7ooEfmFM8MwFPrFG4o4ulOS5GL2VvCot+VbvYF14ZBH+/TRf//7X1uUDwAAAJQIAhqKXX7hTJJOf7dQf/2yTpJkcnHVP0a8KZ+qt0mS3Hz8rSHt/3r1IqQBAADAYbGKYzFhFcfrzp49q9p16sgzsIb++cKyHMPZH1vX6vdVs68/MJnUfOgrqhR8X7Z2KbGR2jGzj5yNdCUlJhR36QAAAECJYwYNxSY8PFz/aBEsmZyUcPmcLuz8Nlubywc36/fVc6yPG/Qal2M4MwxDZ35cprTEOD3a65FirRsAAACwFZbZR7EIDw9XvfoNFJeQqKZPvaPIozt0fM3bkqQa7XtLkiJPHdSRZdOsS+rX6jTQui8zwzB0fO08/bF5lfr166fPPvus5F4IAAAAUIIIaChymcNZkxFvy7dWM/nWaiZJ1pBWvkErHVwwSZa0FElS5du76raHRmU71o3h7Isvvii5FwIAAACUMAIailRO4UySTCaTaj74lKTrIc3F7K20xDhJUoVGd6hx/xdlMpmyHItwBgAAgLKGa9BQpFq2aqXYuLgs4SyDyWRS1fv6y8XTzxrOfGs0UvMnX5OTc9a/FWQOZw899BDhDAAAAGUCAQ1FqlnTpjLS0xRz5lCO+51d3eVbt4X1cUDzdnJx98zSJnM4c3Fx1ZdfflmcJQMAAAB2g1McUaTWr1+v4OBgHVy/UJJU7b4BWfY7ubqr4YDpOvfjx0qLi9Lp7z6Si4endXGQG8PZuXNn5eGRfWl+AAAAwBER0FCkdu3apSNHj8nJ3VPncglpJidn1eo6QoZhyMXsY104pHq7R7OFsypVqpT4awAAAABshYCGIrNr1y7d2669nDzLqenI+bqyf32uIU3KvnBI+MFNijx1QJJJd911J+EMAAAAZQ4BDUUiLCzMGs4aj5wvd/8gVbt/iCQVKKSlxEboyq8hci9fReWbtdfWLV+od+/eWr16dYm+DgAAAMCWCGgoElevXlV6errcvf3lbPaxbi9ISIv7I1TXju2UV7WGavTk24oM3SlJOv/HHyVQOQAAAGA/WMURRaJ169ZatnSJEi6e0u8fj1daUrx1X7X7h6hqx8E6t36h/vz5syz9Ys8f09FF42WuVNsazk6velV1b7tN27ZuLemXAQAAANgUM2goMoMGDZIkPTFkqH7/eLwaDpsjFw8vSTnPpGUOZw2HvPV3OKtbV8eOHpWbm5ttXggAAABgIwQ0FKmChrTkqHBdORBCOAMAAAAy4RRHFLlBgwZdP93xr5O5nu54add/CGcAAADADUyGYRi2LsIRxcTEyM/PT9HR0fL19bV1OUXi5MmT2rJli2JiYlS1alX93//9X55havny5XpiyFB5VqmXZSZNkmLOHJRX1fqEMwAAACATAloxcaSAtmDBAr3x+us6d+6cMg8WVydntbuvg+bNm6cmTZrk2DevkBZxcCPhDAAAAMiEgFZMHCGgpaWl6fbbb9eBAwcU4OSkrm6eaubqJjdJkYZFW5KTtC01SRZJ8+bP1zPPPJPjcXIKaYQzAAAAIDsCWjEp7QHNYrGoZcuWOnzokJ4we+sBN7OcTKZs7WItFs2Jj1ZoeqoWLV6sYcOG5Xi8zCEtsM2DCvvP24QzAAAA4AYEtGJS2gParFmzNGnSJA0ze+sBd88826YYhqbHReq8DMUnJcnFJefFQTNCmmFJV93bbiOcAQAAADdgFUfkaO7bb6uak7M6u5nzbetmMmmI2Ucp6el67bXXcm03aNAgrfjicz300EOEMwAAACAHzKAVk9I8g7Zv3z61adNGT5p91Nk9/4AmSYZhaFLsNSVVKK9L4eHFXCEAAADgmJhBQzY//PCDJKmla8FnuEwmk1q5uuva1avFVRYAAADg8AhoyCYmJkaS5K7si4LkxcNkksXChCwAAABwswhoyCYgIECSFG1YCtUv2rDI2ZkhBQAAANwsPk0jm4EDB8pJ0vaUpAL3STMMbU9JUp169YqvMAAAAMDBEdCQTVBQkJo0a6YNKYlKLeAaMvtSkxVtGJoxY0YxVwcAAAA4LgJaGRYeHq5KQZXVvHlzpaWlZdn35ptvKt4w9FFCjCz5hLTw9HQtSoxV+XLl1KdPn+IsGQAAAHBoBLQyKjw8XPXqN1D4lQj99tsRNWrUKEtI69q1q8Y884y2pSbr7fhoXU5Pz3YMi2Ho19RkvRh3TSnOztq6fXtJvgQAAADA4bjYugCUvIxwFhMbL7fbn5aReE2nDn2uRo0aKTQ0VC4u14fF/Pnz5e3trbdef0N7Yq+qhYub/uHiJjeTSZGWdG1OSVKEYZGfj4/27dypJk2a2PiVAQAAAKUbX1RdTOz1i6qzhrNRcipfV5KUfmGvUg99rttuq5slpElSRESEJk6cqDVffqmExERZJDmbTKpZq5Ymv/iihg4dKicnJmMBAACAW0VAKyb2GNByC2cZ8gppmVksFgIZAAAAUAz4lF1G5BfOJMm5ahu5/qO/Tp06ne2atMwIZwAAAEDx4JN2GVCQcJahoCENAAAAQNEjoDm4woSzDIQ0AAAAwDYIaA7sZsJZBkIaAAAAUPIIaA7qVsJZBkIaAAAAULIIaA6qX79+iomOkmvzx28qnGVwrtpGzrd11qlTpzR58uQirBAAAADAjQhoDuqVV16Rs4urUn//r4zEazd9HEvUOaWHbZaXt4/GjRtXhBUCAAAAuBEBzUHdeeederjnQ1JytJJ3zb+pkGaJOqeUX96Xl9lNv4ceU5UqVYqhUgAAAAAZCGgOauzYsVq7dq1MPjUlS7pSdr9XqJB2YzirVq1aMVYLAAAAQCKgOaSxY8fq3XfflZN/fbnV7iL323pK6akFDmmEMwAAAMA2CGgOJnM4c63RUSaTk0xuvnIrYEgjnAEAAAC2Q0BzIDmFswwFCWmEMwAAAMC2CGgOIq9wliGvkEY4AwAAAGyPgOYAChLOMuQU0jLCmburE+EMAAAAsCECWilXmHCWIUtI2zVfKXs+kAyLkpOStG3bthKoGgAAAEBOTIZhGLYuwhHFxMTIz89P0dHR8vX1LZbniIiIUEBAgEzminKr37tA4SwzIyVGKae+kVy95Fqjo1KOfykPdxclJiQUS70AAAAA8sYMWilWsWJFdejQQUZihNLDfy10f5Obr9waPi7Xuj2U9tdOyZKq58eNK4ZKAQAAABQEAa2U+/nnn9W+fXulXfxFaZf3Ff4AJint3EZZos9o8uTJevXVV4u+SAAAAAAFQkBzAJs2bbqpkGYY6Uo9u8Eazl577bVirBIAAABAfghoDqKwIY1wBgAAANgfApoDKWhII5wBAAAA9omA5mDyC2mEMwAAAMB+EdAcUG4hjXAGAAAA2DcCmoO6MaQRzgAAAAD7xxdVF5OS+KLqgujQoYM2b94sk0d5GUnXCGcAAACAHWMGzcFt2rRJ9913H+EMAAAAKAWYQSsm9jKDliEuLk7e3t62LgMAAABAHphBKyMIZwAAAID9I6ABAAAAgJ0goAEAAACAnSCgAQAAAICdIKABAAAAgJ0goAEAAACAnSCgAQAAAICdIKABAAAAgJ0goAEAAACAnSCgAQAAAICdIKABAAAAgJ0goAEAAACAnSCgAQAAAICdIKABAAAAgJ0goAEAAACAnSCgAQAAAICdIKABAAAAgJ0goAEAAACAnSCgAQAAAICdIKABAAAAgJ0goAEAAACAnSCgAQAAAICdIKABAAAAgJ1wsXUBjsowDElSTEyMjSsBAAAAYA98fHxkMpnybENAKyaxsbGSpOrVq9u4EgAAAAD2IDo6Wr6+vnm2MRkZUz0oUhaLRX/99ZcMw1CNGjX0xx9/5PvDQNkVExOj6tWrM06QL8YKCoqxgoJirKCgGCu3jhk0G3JyclK1atWspzj6+voykJEvxgkKirGCgmKsoKAYKygoxkrxYpEQAAAAALATBDQAAAAAsBMEtGLm7u6uadOmyd3d3dalwI4xTlBQjBUUFGMFBcVYQUExVkoGi4QAAAAAgJ1gBg0AAAAA7AQBDQAAAADsBAENAAAAAOwEAQ0AAAAA7AQBrRjExsZq+vTpatasmby9veXn56c2bdpozpw5SklJsXV5KAFXr17V0qVLNWDAADVu3FheXl5yd3dXtWrV9PDDD+vrr7/O9xiMo7LrjTfekMlkst7ywjgpm2JiYvTmm2/qrrvuUkBAgPX3S4cOHTR9+nRFRUXl2I/xUraEhISoT58+qlmzpjw8PGQ2m1WnTh31799fW7ZsybMvY8UxJCQkaP369XrllVf0yCOPqGbNmtb/W6ZPn16gY1y+fFnjx49XgwYNZDabVb58ebVt21aLFy9WQdYaPH36tEaOHKnatWvLw8NDAQEBeuCBB7R27dpbfHUOzECROnv2rFGrVi1DkiHJ8PT0NNzd3a2Pg4ODjWvXrtm6TBQzFxcX689ckuHh4WF4eXll2da1a1cjPj4+x/6Mo7Lr999/Nzw8PLKMldwwTsqmn3/+2ahUqZL15+zm5maUK1cuy5g5cOBAtn6Ml7LDYrEYI0eOzDImzGazYTabs2wbN25cjv0ZK45j06ZNWX7mmW/Tpk3Lt/++ffuMChUqWPt4e3tn+YzzwAMPGMnJybn2X7duneHp6Wlt7+vrazg5OVkfDxkyxLBYLEX4ih0DAa0IpaamGs2aNTMkGZUrVzZCQkIMwzCM9PR0Y+XKlYaPj48hyejWrZuNK0Vxk2TcfvvtxgcffGCcPn3auj0sLMwYNmyY9RfTgAEDsvVlHJVd6enpxl133WVIMu688848AxrjpGzavn279UP2I488Yuzdu9f64SY+Pt7Ys2eP8e9//9s4c+ZMln6Ml7JlyZIl1t8fjz76qHHixAnrvt9//93o2bOndf9XX32VpS9jxbFs2rTJ8Pf3Nzp27GhMnDjRWLFihREUFFSggBYVFWVt27BhQ2Pv3r2GYRhGcnKy8d577xmurq6GJGPUqFE59j9z5oz1j9N33323cfz4ccMwDCM2NtaYOnWqdQy++eabRfqaHQEBrQgtXrzYOth27tyZbf8XX3xh3b9x40YbVIiS8vPPP+e5P/NfNs+fP59lH+Oo7Jo3b54hyejfv78xbdq0PAMa46TsiY+PN+rUqWNIMp555plC9WW8lC3t27c3JBm33XabkZqamm1/SkqKdSz17ds3yz7GimNJS0vLtq1mzZoFCmgvvfSSdfb1xj/6GIZhvPbaa4Ykw9nZ2Rq+MhswYIAhyQgKCjIiIyOz7R8xYoR1Vo0Z2awIaEWobdu2hiSjQ4cOOe63WCxG7dq1DUnGoEGDSrg62JM9e/bk+tdLxlHZlPGXxgoVKhjh4eH5BjTGSdnz0UcfWT/sJCYmFqov46VsadCggSHJ6NWrV65tHnnkEUOS0b179yzbGSuOr6ABrUaNGtbTEHMSGxtreHt7G5KMqVOnZtkXFxdnne2fMWNGjv3DwsKs/88tWbLkpl6Lo2KRkCKSkJCgHTt2SJK6du2aYxuTyaQuXbpIkjZs2FBitcH+eHh4WO+np6db7zOOyq7hw4crPj5eb7/9tgICAvJsyzgpm5YvXy5J6t27d5bfIflhvJQ9derUkSQdOnRIaWlp2fanpqbq4MGDkqTWrVtbtzNWkOH48eM6f/68pNzHgre3t9q2bSsp+1jYvn27EhMT8+xfq1YtNWrUKMf+ZR0BrYiEhobKYrFIkpo2bZpru4x9ly5d0rVr10qkNtifzZs3W+83a9bMep9xVDYtWrRIP/30k+6//34NGjQo3/aMk7InOTlZ+/btkyS1atVK58+f14gRI1S9enW5ubmpUqVK6tGjh9atW5etL+Ol7Bk1apQk6dSpU+rXr59OnTpl3Xf8+HH16dNHZ86cUd26dTVu3DjrPsYKMhw5csR6vyBj4dixY7fU/+jRozdVp6MioBWRv/76y3q/atWqubbLvC9zH5QdUVFRev311yVJbdu2VYMGDaz7GEdlz4ULFzRx4kSZzWYtWLCgQH0YJ2XP2bNnrUubnzlzRk2bNtWiRYsUHh4uLy8vhYeH67vvvlP37t01fPjwLEtfM17Knh49emju3Llyc3PTmjVrVK9ePXl6esrT01MNGzbU5s2bNWrUKO3Zs0e+vr7WfowVZCjsWIiJiVFcXFy2/v7+/jKbzfn2ZxxlRUArIrGxsdb7np6eubbLvC9zH5QNFotFAwcO1MWLF+Xh4aH33nsvy37GUdkzcuRIRUdHa/r06dbTkvLDOCl7IiMjrfdfeeUVubq6avXq1YqLi1NkZKTOnTun3r17S5IWL16suXPnWtszXsqm5557Tl999ZUCAwMlSYmJidZTzlJSUhQXF6fo6OgsfRgryHCrYyHjfl59M+9nHGVFQANK0LPPPqvvvvtOkvT++++refPmNq4ItvTZZ59p3bp1atGihZ5//nlblwM7lnHaWcb9jz/+WI8++qhcXV0lSTVq1NDKlSv1j3/8Q5L02muv5XjtEcqGhIQEPfbYY+revbtq1KihDRs26MqVK7py5Yo2bNigxo0b69NPP9Xtt9+uw4cP27pcADcgoBURHx8f6/2EhIRc22Xel7kPHN+ECROsM2Zz587V0KFDs7VhHJUdly9f1nPPPSdnZ2ctWrRILi4uBe7LOCl7Mv/86tWrp4cffjhbGycnJ02YMEGSdPXqVe3fvz9bX8ZL2TBx4kStWrVKDRo00LZt29SpUydVrFhRFStWVKdOnbR161bVr19fERERGj16tLUfYwUZbnUsZNzPq2/m/YyjrAhoRaRKlSrW+xcuXMi1XeZ9mfvAsU2aNElz5syRJM2ePVvPPfdcju0YR2XHv/71L129elUjRoxQw4YNFRcXl+WWcb2RpGzbGCdlT+ZrQBo2bJhru8aNG1vvnzt3ThLjpayJjY3VwoULJUmjR4/OccVPs9msMWPGSLq+2l54eLgkxgr+Vtix4OvrK29v72z9IyMjrafW5tWfcZQVAa2INGrUSE5O19/OzCvX3ChjX1BQkMqXL18itcG2Jk6cqFmzZkmS3nrrLY0fPz7XtoyjsiMsLEyS9OGHH8rHxyfbLWMhGUnWbZMmTZLEOCmLypcvn+eF+hkyLw5iMpkkMV7KmhMnTlhPb61bt26u7erVq2e9n/H7iLGCDJlXXizIWMj8x6Gb6d+kSZObqtNREdCKiKenp+6++25J0g8//JBjG8Mw9OOPP0qSOnfuXGK1wXYmTJig2bNnS7oeziZOnJhne8YRCoJxUjZl/BxDQ0NzbZN5qevatWtLYryUNRkBS/p7FjUnly9ftt7POL2MsYIM9evXV40aNSTlPhbi4+O1bds2SdnHwj333GNdvTG3/ufOnbP+PmMs3cCGX5LtcBYvXmxIMkwmk7F79+5s+7/88kvrN6Zv3LjRBhWiJI0fP9768549e3aB+zGOYBiGMW3aNOvPOSeMk7Jn69at1p/p119/nW1/enq60bx5c0OSUbVqVSM9Pd26j/FSdiQkJBhms9mQZLRs2dJITU3N1iYtLc246667DEmGv7+/kZaWZt3HWHF8NWvWNCQZ06ZNy7PdSy+9ZEgyPD09jbCwsGz733zzTUOS4ezsbBw/fjzb/gEDBhiSjMqVKxtRUVHZ9o8aNcqQZPj4+BjXrl272ZfjkAhoRSg1NdVo1qyZ9T/HjF9c6enpxqpVqwxfX19DktG1a1cbV4riNnHiROt/YG+//Xah+jKOYBj5BzTGSdn06KOPGpKMChUqGGvWrLF++D537pzRp08f65hZtmxZln6Ml7LlmWeesY6FLl26GIcPHzbS09ON9PR049ChQ0bnzp2t+2fMmJGlL2PF8Vy7ds24cuWK9Va9enVDkjFx4sQs22NjY7P0i4qKMoKCggxJRuPGjY19+/YZhmEYycnJxgcffGC4ubkZkoxRo0bl+LxnzpwxvLy8DElG27ZtjRMnThiGYRhxcXHGjBkzDJPJZEgy3nzzzeJ9A0ohAloRCwsLM2rVqmX9xefp6Wl4eHhYHwcHB/NXAgd37tw568/bycnJqFSpUp63WbNmZTsG4wj5BTTDYJyURXFxcca9995r/Rm7u7sb/v7+1sd5/VWc8VJ2JCQkGF26dMkyLtzd3Q13d/cs2/r165dl9iwDY8WxZMyY5XcbPHhwtr779u0zKlSoYG3j4+NjuLq6Wh937tzZSEpKyvW5161bZ3h6elrb+/n5Gc7OztbHQ4YMMSwWSzG++tKJa9CKWK1atXT48GFNnTpVTZs2lclkkqurq1q1aqXZs2dr9+7d8vf3t3WZKEY3fl/R5cuX87zFxcVlOwbjCAXBOCl7vLy8tGnTJi1atEj33nuvvLy8FBcXp6pVq6pv377asWOHpk+fnmNfxkvZYTab9f3332v16tXq2bOnqlWrZl1Apnr16urVq5e+++47ffHFF3J2ds7Wn7GCDK1atdLRo0c1btw41atXT6mpqfLy8tI999yjRYsWaf369XJ3d8+1f7du3XT48GENHz5ctWrVUlJSkvz9/dWpUyetWbNGS5YssS5ohL+ZDCPTkk8AAAAAAJthBg0AAAAA7AQBDQAAAADsBAENAAAAAOwEAQ0AAAAA7AQBDQAAAADsBAENAAAAAOwEAQ0AAAAA7AQBDQAAAADsBAENAAAAAOwEAQ0AkKP27dvrueeekyTVqlVL8+bNs2k9pVFCQoJ69eolX19fmUwmRUVF2eS9fOKJJ/Twww+X2POdPXtWJpNJBw8eLLHnBABHQUADAORr7969GjFiRIHaltYwVxx1f/LJJ9q2bZt27typixcvys/Pr1DvZU5uJvy88847WrZs2U0/JwCg5LjYugAAgP0LCAiwdQml0unTp9WoUSM1bdrUui2/9zI1NVWurq5FWoefn1+RHg8AUHyYQQMAKD4+XoMGDZK3t7cqV66sOXPmZNmfeXbJMAxNnz5dNWrUkLu7u6pUqaKxY8dKun5a5Llz5zRu3DiZTCaZTCZJ0tWrV9WvXz9VrVpVnp6eatasmVasWJHlOdq3b6+xY8dq0qRJKl++vIKCgjR9+vQsbaKiojRy5EhVqlRJHh4eatq0qb777jvr/u3bt6tt27Yym82qXr26xo4dq/j4+Hxff251S9LatWvVpEkTubu7q1atWtnem7yOOWfOHG3dulUmk0nt27fP9l5Kkslk0ocffqiHHnpIXl5eevXVVxUZGan+/fsrICBAZrNZ9erV09KlSyVJtWvXliQFBwdnOW5ebjzFsSDvdV4yau7atavMZrPq1KmjNWvW5No+PT1dw4YNU+3atWU2m9WgQQO98847OdY4e/ZsVa5cWRUqVNDo0aOVmppqbZOcnKwJEyaoatWq8vLy0h133KHNmzcXuG4AKA0IaAAATZw4UVu2bNE333yjDRs2aPPmzfr1119zbLt27VrNnTtXCxYs0MmTJ/Wf//xHzZo1kyR99dVXqlatmmbOnKmLFy/q4sWLkqSkpCS1atVK69at05EjRzRixAgNHDhQe/bsyXLsTz75RF5eXvrll1/01ltvaebMmQoJCZEkWSwWde3aVTt27NBnn32mY8eO6Y033pCzs7Ok67NVXbp0Ua9evXT48GF9+eWX2r59u8aMGZPv68+t7v3796tPnz7q27evfvvtN02fPl1Tpkwp0OmCX331lYYPH64777xTFy9e1FdffZVr2+nTp+v//u//9Ntvv2no0KGaMmWKjh07pvXr1ys0NFQffvihKlasKEnW92zjxo35Hjcveb3XBTFlyhT16tVLhw4dUv/+/dW3b1+Fhobm2NZisahatWpavXq1jh07pqlTp+rFF1/UqlWrsrTbtGmTTp8+rU2bNumTTz7RsmXLsrzXY8aM0a5du7Ry5UodPnxYvXv3VpcuXXTy5Mmbeg8AwC4ZAIAyLTY21nBzczNWrVpl3Xb16lXDbDYbzz77rGEYhlGzZk1j7ty5hmEYxpw5c4z69esbKSkpOR4vc9u8PPjgg8b48eOtj9u1a2fcc889Wdq0adPGeOGFFwzDMIwff/zRcHJyMo4fP57j8YYNG2aMGDEiy7Zt27YZTk5ORmJiYr715FT3448/bnTq1CnLtokTJxqNGzfO93iGYRjPPvus0a5duzyfR5Lx3HPPZWnTo0cPY8iQITkeMywszJBkHDhwoEA1GIZhDB482OjZs6f1cX7vdX4kGU899VSWbXfccYcxatSoAtc4evRoo1evXllqrFmzppGWlmbd1rt3b+Oxxx4zDMMwzp07Zzg7OxsXLlzIcpyOHTsakydPLlDdAFAaMIMGAGXc6dOnlZKSojvuuMO6rXz58mrQoEGO7Xv37q3ExETVqVNHw4cP19dff620tLQ8nyM9PV0vv/yymjVrpvLly8vb21s//vijzp8/n6Vd8+bNszyuXLmywsPDJUkHDx5UtWrVVL9+/Ryf49ChQ1q2bJm8vb2ttwceeEAWi0VhYWH5vg85CQ0N1d13351l2913362TJ08qPT39po6Zk9atW2d5PGrUKK1cuVItWrTQpEmTtHPnziJ7rgx5vdcFceedd2Z7nNsMmiS9//77atWqlQICAuTt7a2FCxdm+/k3adLEOiN6Y02//fab0tPTVb9+/Sw/4y1btuj06dMFrhsA7B2LhAAACqV69eo6fvy4Nm7cqJCQED399NOaNWuWtmzZkuviFrNmzdI777yjefPmqVmzZvLy8tJzzz2nlJSULO1u7G8ymWSxWCRJZrM5z7ri4uI0cuRI6/VwmdWoUaMwL7HEeXl5ZXnctWtXnTt3Tt9//71CQkLUsWNHjR49WrNnzy6y58zrvS5qK1eu1IQJEzRnzhzdeeed8vHx0axZs/TLL78UuKa4uDg5Oztr//79WUKcJHl7exdL3QBgCwQ0ACjj6tatK1dXV/3yyy/WIBMZGakTJ06oXbt2OfYxm83q0aOHevToodGjR6thw4b67bff1LJlS7m5uWWbXdqxY4d69uypAQMGSLp+TdKJEyfUuHHjAtfZvHlz/fnnnzpx4kSOs2gtW7bUsWPHdNtttxX4mJnlVHejRo20Y8eOLNt27Nih+vXrZwsJRS0gIECDBw/W4MGD1bZtW02cOFGzZ8+Wm5ubJBXpDN7N2L17twYNGpTlcXBwcI5td+zYobvuuktPP/20dVthZ72Cg4OVnp6u8PBwtW3b9uaKBoBSgFMcAaCM8/b21rBhwzRx4kT9/PPPOnLkiJ544gk5OeX8X8SyZcv08ccf68iRIzpz5ow+++wzmc1m1axZU9L1VQq3bt2qCxcuKCIiQpJUr149hYSEaOfOnQoNDdXIkSN1+fLlQtXZrl073XvvverVq5dCQkIUFham9evX64cffpAkvfDCC9q5c6fGjBmjgwcP6uTJk/rmm28KtEhIbnWPHz9eP/30k15++WWdOHFCn3zyid577z1NmDChULUX1tSpU/XNN9/o1KlTOnr0qL777js1atRIkhQYGCiz2awffvhBly9fVnR0dLHWkpvVq1dryZIlOnHihKZNm6Y9e/bk+l7Xq1dP+/bt048//qgTJ05oypQp2rt3b6Ger379+urfv78GDRqkr776SmFhYdqzZ49ef/11rVu3riheEgDYBQIaAECzZs1S27Zt1aNHD91///2655571KpVqxzblitXTosWLdLdd9+t5s2ba+PGjfr2229VoUIFSdLMmTN19uxZ1a1b1/qdXy+99JJatmypBx54QO3bt1dQUFCWZd8Lau3atWrTpo369eunxo0ba9KkSdaZpObNm2vLli06ceKE2rZtq+DgYE2dOlVVqlQp0LFzqrtly5ZatWqVVq5cqaZNm2rq1KmaOXOmnnjiiULXXhhubm6aPHmymjdvrnvvvVfOzs5auXKlJMnFxUXz58/XggULVKVKFfXs2bNYa8nNjBkztHLlSjVv3lzLly/XihUrcp0RHTlypB555BE99thjuuOOO3T16tUss2kFtXTpUg0aNEjjx49XgwYN9PDDD2vv3r12fworABSGyTAMw9ZFAACA0sNkMunrr7++qZANAMgbM2gAAAAAYCcIaAAAh7dt27YsS7PfeLOXY96svOrYtm1boY71+eef53qsJk2aFNMrAABk4BRHAIDDS0xM1IULF3LdfzMrPxbHMW/WqVOnct1XtWrVfL+iILPY2NhcF3BxdXW1LgYDACgeBDQAAAAAsBOc4ggAAAAAdoKABgAAAAB2goAGAAAAAHaCgAYAAAAAdoKABgAAAAB2goAGAAAAAHaCgAYAAAAAdoKABgAAAAB24v8BvifJt4PsZmYAAAAASUVORK5CYII=",
      "text/plain": [
       "<Figure size 1000x1000 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# plot distance in null vs distance in range\n",
    "X = 'distance_to_first_in_plane'\n",
    "Y = 'distance_to_first_null_space'\n",
    "\n",
    "plt.figure(figsize=(10,10))\n",
    "\n",
    "# lwf\n",
    "df = combined_df_lwf[combined_df_lwf['dense_size'] == DENSE_SIZE]\n",
    "df = df[df['lwf_temperature'] == LWF_TEMP]\n",
    "df = df.groupby(['lwf_alpha', 'lwf_temperature']).mean(numeric_only=True).reset_index()\n",
    "df = df.sort_values(by=[X])\n",
    "\n",
    "# regression line\n",
    "# slope, intercept, r_value, p_value, std_err = linregress(df[X], df[Y])\n",
    "# plt.plot(df[X], slope * df[X] + intercept, '--', c='blue')\n",
    "\n",
    "plt.scatter(df[X],\n",
    "            df[Y],\n",
    "            c=df['lwf_alpha'],\n",
    "            cmap='Blues',\n",
    "            norm=LogNorm(),\n",
    "            s=MARKER_SIZE,\n",
    "            edgecolors='k',\n",
    "            linewidths=1,\n",
    "            label='LwF',\n",
    "            marker='D'\n",
    "            )\n",
    "\n",
    "# ewc\n",
    "df = combined_df_ewc[combined_df_ewc['dense_size'] == DENSE_SIZE]\n",
    "df = df.groupby(['ewc_lambda']).mean(numeric_only=True).reset_index()\n",
    "# slope, intercept, r_value, p_value, std_err = linregress(df[X], df[Y])\n",
    "# plt.plot(df[X], slope * df[X] + intercept, '--', c='green')\n",
    "\n",
    "log_norm_ewc = LogNorm(vmin=df['ewc_lambda'].min(), vmax=df['ewc_lambda'].max())\n",
    "plt.scatter(x=df[X],\n",
    "            y=df[Y],\n",
    "            c=df['ewc_lambda'],\n",
    "            cmap='Greens',\n",
    "            norm=log_norm_ewc,\n",
    "            s=MARKER_SIZE,\n",
    "            edgecolors='k',\n",
    "            linewidths=1,\n",
    "            marker='s',\n",
    "            label='EWC',\n",
    "            )\n",
    "\n",
    "# SI\n",
    "df = combined_df_si[combined_df_si['dense_size'] == DENSE_SIZE]\n",
    "df = df[df['si_eps'] == SI_EPS]\n",
    "df = df.groupby(['si_lambda']).mean(numeric_only=True).reset_index()\n",
    "df = df.sort_values(by=[X])\n",
    "\n",
    "# regression line\n",
    "# slope, intercept, r_value, p_value, std_err = linregress(df[X], df[Y])\n",
    "# plt.plot(df[X], slope * df[X] + intercept, '--', c='red')\n",
    "\n",
    "norm = LogNorm(vmin=df['si_lambda'].min(), vmax=df['si_lambda'].max())\n",
    "plt.scatter(df[X],\n",
    "            df[Y],\n",
    "            c=df['si_lambda'],\n",
    "            cmap='Reds',\n",
    "            norm=norm,\n",
    "            s=MARKER_SIZE,\n",
    "            edgecolors='k',\n",
    "            linewidths=1,\n",
    "            label='SI',\n",
    "            )\n",
    "\n",
    "# gem\n",
    "df = combined_df_gem[combined_df_gem['dense_size'] == DENSE_SIZE]\n",
    "df = df[df['gem_memory_strength'] >= 0]\n",
    "df = df.groupby(['gem_patterns_per_experience', 'gem_memory_strength']).mean(numeric_only=True).reset_index()\n",
    "df = df.sort_values(by=[X])\n",
    "\n",
    "# regression line\n",
    "# slope, intercept, r_value, p_value, std_err = linregress(df[X], df[Y])\n",
    "# plt.plot(df[X], slope * df[X] + intercept, '--', c='k')\n",
    "\n",
    "plt.scatter(df[X],\n",
    "            df[Y],\n",
    "            c=df['gem_memory_strength'],\n",
    "            cmap='Greys',\n",
    "            norm=LogNorm(),\n",
    "            s=MARKER_SIZE,\n",
    "            edgecolors='k',\n",
    "            linewidths=1,\n",
    "            label='GEM',\n",
    "            marker='*'\n",
    "            )\n",
    "\n",
    "# replay\n",
    "df = combined_df_replay[combined_df_replay['dense_size'] == DENSE_SIZE]\n",
    "df = df.groupby(['replay_mem_size']).mean(numeric_only=True).reset_index()\n",
    "df = df.sort_values(by=[X])\n",
    "# regression line\n",
    "# slope, intercept, r_value, p_value, std_err = linregress(df[X], df[Y])\n",
    "# plt.plot(df[X], slope * df[X] + intercept, '--', c='purple')\n",
    "\n",
    "plt.scatter(df[X],\n",
    "            df[Y],\n",
    "            c=df['replay_mem_size'],\n",
    "            cmap='Purples',\n",
    "            norm=LogNorm(),\n",
    "            s=MARKER_SIZE,\n",
    "            edgecolors='k',\n",
    "            linewidths=1,\n",
    "            label='Replay',\n",
    "            marker='^'\n",
    "            )\n",
    "\n",
    "\n",
    "# plt.legend()\n",
    "\n",
    "# axis labels\n",
    "plt.xlabel(X)\n",
    "plt.ylabel(Y)\n",
    "# equal axes\n",
    "# plt.gca().set_aspect('equal', adjustable='box')\n",
    "# despine\n",
    "sns.despine()\n",
    "\n",
    "# increase size of tick font\n",
    "plt.xticks(fontsize=18)\n",
    "plt.yticks(fontsize=18)\n",
    "\n",
    "# x = np.linspace(0, 270, 100)\n",
    "# y = np.sqrt(10/118) * x\n",
    "# plt.plot(x, y, linestyle='dotted', c='k', alpha=.5)\n",
    "\n",
    "# plt.savefig(f'{plotdir}/distance_to_first_in_plane_vs_null_space.png', bbox_inches='tight')\n",
    "\n",
    "REFCOLORS = ['red', 'blue', 'green', 'purple']\n",
    "for REFERENCE_STABILITY, REF_COLOR in zip(REFERENCE_STABILITIES, REFCOLORS):\n",
    "\n",
    "    # EWC\n",
    "    df = combined_df_ewc[combined_df_ewc['dense_size'] == DENSE_SIZE]\n",
    "    df = df.groupby(['ewc_lambda']).mean(numeric_only=True).reset_index()\n",
    "    closest_stability_ewc = get_closest_stability(df, REFERENCE_STABILITY)\n",
    "    ewc_ref_runs = combined_df_ewc[(combined_df_ewc['dense_size'] == DENSE_SIZE) & (combined_df_ewc['ewc_lambda'] == closest_stability_ewc['ewc_lambda'])]\n",
    "\n",
    "    # SI\n",
    "    df = combined_df_si[combined_df_si['dense_size'] == DENSE_SIZE]\n",
    "    df = df[df['si_eps'] == SI_EPS]\n",
    "    df = df.groupby(['si_lambda']).mean(numeric_only=True).reset_index()\n",
    "    closest_stability_si = get_closest_stability(df, REFERENCE_STABILITY)\n",
    "    si_ref_runs = combined_df_si[(combined_df_si['dense_size'] == DENSE_SIZE) & (combined_df_si['si_eps'] == SI_EPS) & (combined_df_si['si_lambda'] == closest_stability_si['si_lambda'])]\n",
    "\n",
    "    # Replay\n",
    "    df = combined_df_replay[combined_df_replay['dense_size'] == DENSE_SIZE]\n",
    "    df = df.groupby(['replay_mem_size']).mean(numeric_only=True).reset_index()\n",
    "    closest_stability_replay = get_closest_stability(df, REFERENCE_STABILITY)\n",
    "    replay_ref_runs = combined_df_replay[(combined_df_replay['dense_size'] == DENSE_SIZE) & (combined_df_replay['replay_mem_size'] == closest_stability_replay['replay_mem_size'])]\n",
    "\n",
    "    # gem\n",
    "    df = combined_df_gem[combined_df_gem['dense_size'] == DENSE_SIZE]\n",
    "    df = df[df['gem_memory_strength'] >= 0]\n",
    "    df = df.groupby(['gem_patterns_per_experience', 'gem_memory_strength']).mean(numeric_only=True).reset_index()\n",
    "    closest_stability_gem = get_closest_stability(df, REFERENCE_STABILITY)\n",
    "    gem_ref_runs = combined_df_gem[(combined_df_gem['dense_size'] == DENSE_SIZE) & (combined_df_gem['gem_memory_strength'] == closest_stability_gem['gem_memory_strength']) & (combined_df_gem['gem_patterns_per_experience'] == closest_stability_gem['gem_patterns_per_experience'])]\n",
    "\n",
    "    # lwf\n",
    "    df = combined_df_lwf[combined_df_lwf['dense_size'] == DENSE_SIZE]\n",
    "    df = df[df['lwf_temperature'] == LWF_TEMP]\n",
    "    df = df.groupby(['lwf_alpha', 'lwf_temperature']).mean(numeric_only=True).reset_index()\n",
    "    closest_stability_lwf = get_closest_stability(df, REFERENCE_STABILITY)\n",
    "    lwf_ref_runs = combined_df_lwf[(combined_df_lwf['dense_size'] == DENSE_SIZE) & (combined_df_lwf['lwf_alpha'] == closest_stability_lwf['lwf_alpha']) & (combined_df_lwf['lwf_temperature'] == closest_stability_lwf['lwf_temperature'])]\n",
    "\n",
    "    # force a copy of all dfs\n",
    "    ewc_ref_runs = ewc_ref_runs.copy()\n",
    "    si_ref_runs = si_ref_runs.copy()\n",
    "    replay_ref_runs = replay_ref_runs.copy()\n",
    "    gem_ref_runs = gem_ref_runs.copy()\n",
    "    lwf_ref_runs = lwf_ref_runs.copy()\n",
    "\n",
    "    # concatenate all runs after adding algorithm column\n",
    "    ewc_ref_runs['algorithm'] = 'EWC'\n",
    "    si_ref_runs['algorithm'] = 'SI'\n",
    "    replay_ref_runs['algorithm'] = 'Replay'\n",
    "    gem_ref_runs['algorithm'] = 'GEM'\n",
    "    lwf_ref_runs['algorithm'] = 'LwF'\n",
    "\n",
    "    # concatenate all runs\n",
    "    ref_runs = pd.concat([ewc_ref_runs, si_ref_runs, replay_ref_runs, gem_ref_runs, lwf_ref_runs])\n",
    "    # select algorithm and distances\n",
    "    ref_runs = ref_runs[['algorithm', 'distance_to_first_null_space', 'distance_to_first_in_plane']]\n",
    "    # avg over algorithm\n",
    "    ref_runs = ref_runs.groupby(['algorithm']).mean(numeric_only=True).reset_index()\n",
    "\n",
    "    ref_runs_select = ref_runs[\n",
    "                        (ref_runs['algorithm'] =='LwF') |\n",
    "                        (ref_runs['algorithm'] =='EWC') |\n",
    "                        (ref_runs['algorithm'] =='SI')\n",
    "                ]\n",
    "    # sort by Y\n",
    "    ref_runs_select = ref_runs_select.sort_values(by=[Y])\n",
    "    \n",
    "    plt.plot(\n",
    "        ref_runs_select[X],\n",
    "        ref_runs_select[Y],\n",
    "        c = 'k',\n",
    "        linewidth=2,\n",
    "    )\n",
    "# lwf\n",
    "df = combined_df_lwf[combined_df_lwf['dense_size'] == DENSE_SIZE]\n",
    "df = df[df['lwf_temperature'] == LWF_TEMP]\n",
    "df = df.groupby(['lwf_alpha', 'lwf_temperature']).mean(numeric_only=True).reset_index()\n",
    "df = df.sort_values(by=[X])\n",
    "\n",
    "# regression line\n",
    "# slope, intercept, r_value, p_value, std_err = linregress(df[X], df[Y])\n",
    "# plt.plot(df[X], slope * df[X] + intercept, '--', c='blue')\n",
    "\n",
    "plt.scatter(df[X],\n",
    "            df[Y],\n",
    "            c=df['lwf_alpha'],\n",
    "            cmap='Blues',\n",
    "            norm=LogNorm(),\n",
    "            s=MARKER_SIZE,\n",
    "            edgecolors='k',\n",
    "            linewidths=1,\n",
    "            label='LwF',\n",
    "            marker='D'\n",
    "            )\n",
    "\n",
    "# ewc\n",
    "df = combined_df_ewc[combined_df_ewc['dense_size'] == DENSE_SIZE]\n",
    "df = df.groupby(['ewc_lambda']).mean(numeric_only=True).reset_index()\n",
    "# slope, intercept, r_value, p_value, std_err = linregress(df[X], df[Y])\n",
    "# plt.plot(df[X], slope * df[X] + intercept, '--', c='green')\n",
    "\n",
    "log_norm_ewc = LogNorm(vmin=df['ewc_lambda'].min(), vmax=df['ewc_lambda'].max())\n",
    "plt.scatter(x=df[X],\n",
    "            y=df[Y],\n",
    "            c=df['ewc_lambda'],\n",
    "            cmap='Greens',\n",
    "            norm=log_norm_ewc,\n",
    "            s=MARKER_SIZE,\n",
    "            edgecolors='k',\n",
    "            linewidths=1,\n",
    "            marker='s',\n",
    "            label='EWC',\n",
    "            )\n",
    "\n",
    "# SI\n",
    "df = combined_df_si[combined_df_si['dense_size'] == DENSE_SIZE]\n",
    "df = df[df['si_eps'] == SI_EPS]\n",
    "df = df.groupby(['si_lambda']).mean(numeric_only=True).reset_index()\n",
    "df = df.sort_values(by=[X])\n",
    "\n",
    "# regression line\n",
    "# slope, intercept, r_value, p_value, std_err = linregress(df[X], df[Y])\n",
    "# plt.plot(df[X], slope * df[X] + intercept, '--', c='red')\n",
    "\n",
    "norm = LogNorm(vmin=df['si_lambda'].min(), vmax=df['si_lambda'].max())\n",
    "plt.scatter(df[X],\n",
    "            df[Y],\n",
    "            c=df['si_lambda'],\n",
    "            cmap='Reds',\n",
    "            norm=norm,\n",
    "            s=MARKER_SIZE,\n",
    "            edgecolors='k',\n",
    "            linewidths=1,\n",
    "            label='SI',\n",
    "            )\n",
    "\n",
    "# gem\n",
    "df = combined_df_gem[combined_df_gem['dense_size'] == DENSE_SIZE]\n",
    "df = df[df['gem_memory_strength'] >= 0]\n",
    "df = df.groupby(['gem_patterns_per_experience', 'gem_memory_strength']).mean(numeric_only=True).reset_index()\n",
    "df = df.sort_values(by=[X])\n",
    "\n",
    "# regression line\n",
    "# slope, intercept, r_value, p_value, std_err = linregress(df[X], df[Y])\n",
    "# plt.plot(df[X], slope * df[X] + intercept, '--', c='k')\n",
    "\n",
    "plt.scatter(df[X],\n",
    "            df[Y],\n",
    "            c=df['gem_memory_strength'],\n",
    "            cmap='Greys',\n",
    "            norm=LogNorm(),\n",
    "            s=MARKER_SIZE,\n",
    "            edgecolors='k',\n",
    "            linewidths=1,\n",
    "            label='GEM',\n",
    "            marker='*'\n",
    "            )\n",
    "\n",
    "# replay\n",
    "df = combined_df_replay[combined_df_replay['dense_size'] == DENSE_SIZE]\n",
    "df = df.groupby(['replay_mem_size']).mean(numeric_only=True).reset_index()\n",
    "df = df.sort_values(by=[X])\n",
    "# regression line\n",
    "# slope, intercept, r_value, p_value, std_err = linregress(df[X], df[Y])\n",
    "# plt.plot(df[X], slope * df[X] + intercept, '--', c='purple')\n",
    "\n",
    "plt.scatter(df[X],\n",
    "            df[Y],\n",
    "            c=df['replay_mem_size'],\n",
    "            cmap='Purples',\n",
    "            norm=LogNorm(),\n",
    "            s=MARKER_SIZE,\n",
    "            edgecolors='k',\n",
    "            linewidths=1,\n",
    "            label='Replay',\n",
    "            marker='^'\n",
    "            )\n",
    "\n",
    "# axis labels\n",
    "plt.xlabel(X)\n",
    "plt.ylabel(Y)\n",
    "sns.despine()\n",
    "\n",
    "# increase size of tick font\n",
    "plt.xticks(fontsize=18)\n",
    "plt.yticks(fontsize=18)\n",
    "\n",
    "# add a dotted line through the origin with slope sqrt(10/118)\n",
    "# x = np.linspace(0, 250, 100)\n",
    "# y = np.sqrt(10/118) * x\n",
    "# plt.plot(x, y, 'k--', alpha=.5)\n",
    "\n",
    "plt.savefig(f'{plotdir}/distance_to_first_in_plane_vs_null_space_matched_stability.png', bbox_inches='tight')\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 1000x1000 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "def get_closest_stability(avg_df, reference_stability):\n",
    "    # get closest stability to reference stability\n",
    "    avg_df['distance_to_reference'] = abs(avg_df['stability'] - reference_stability)\n",
    "    avg_df = avg_df.sort_values(by=['distance_to_reference'])\n",
    "    closest_stability = avg_df.iloc[0]\n",
    "    return closest_stability\n",
    "\n",
    "\n",
    "# plot distance in null vs distance in range\n",
    "X = 'distance_to_first_in_plane'\n",
    "Y = 'distance_to_first_null_space'\n",
    "\n",
    "plt.figure(figsize=(10,10))\n",
    "\n",
    "# lwf\n",
    "df = combined_df_lwf[combined_df_lwf['dense_size'] == DENSE_SIZE]\n",
    "df = df[df['lwf_temperature'] == LWF_TEMP]\n",
    "df = df.groupby(['lwf_alpha', 'lwf_temperature']).mean(numeric_only=True).reset_index()\n",
    "df = df.sort_values(by=[X])\n",
    "\n",
    "# regression line\n",
    "slope, intercept, r_value, p_value, std_err = linregress(df[X], df[Y])\n",
    "plt.plot(df[X], slope * df[X] + intercept, '--', c='blue')\n",
    "\n",
    "\n",
    "plt.scatter(df[X],\n",
    "            df[Y],\n",
    "            c=df['lwf_alpha'],\n",
    "            cmap='Blues',\n",
    "            norm=LogNorm(),\n",
    "            s=MARKER_SIZE,\n",
    "            edgecolors='k',\n",
    "            linewidths=1,\n",
    "            label='LwF',\n",
    "            marker='D'\n",
    "            )\n",
    "\n",
    "# ewc\n",
    "df = combined_df_ewc[combined_df_ewc['dense_size'] == DENSE_SIZE]\n",
    "df = df.groupby(['ewc_lambda']).mean(numeric_only=True).reset_index()\n",
    "slope, intercept, r_value, p_value, std_err = linregress(df[X], df[Y])\n",
    "plt.plot(df[X], slope * df[X] + intercept, '--', c='green')\n",
    "log_norm_ewc = LogNorm(vmin=df['ewc_lambda'].min(), vmax=df['ewc_lambda'].max())\n",
    "plt.scatter(x=df[X],\n",
    "            y=df[Y],\n",
    "            c=df['ewc_lambda'],\n",
    "            cmap='Greens',\n",
    "            norm=log_norm_ewc,\n",
    "            s=MARKER_SIZE,\n",
    "            edgecolors='k',\n",
    "            linewidths=1,\n",
    "            marker='s',\n",
    "            label='EWC',\n",
    "            )\n",
    "\n",
    "# SI\n",
    "df = combined_df_si[combined_df_si['dense_size'] == DENSE_SIZE]\n",
    "df = df[df['si_eps'] == SI_EPS]\n",
    "df = df.groupby(['si_lambda']).mean(numeric_only=True).reset_index()\n",
    "df = df.sort_values(by=[X])\n",
    "\n",
    "# regression line\n",
    "slope, intercept, r_value, p_value, std_err = linregress(df[X], df[Y])\n",
    "plt.plot(df[X], slope * df[X] + intercept, '--', c='red')\n",
    "\n",
    "norm = LogNorm(vmin=df['si_lambda'].min(), vmax=df['si_lambda'].max())\n",
    "plt.scatter(df[X],\n",
    "            df[Y],\n",
    "            c=df['si_lambda'],\n",
    "            cmap='Reds',\n",
    "            norm=norm,\n",
    "            s=MARKER_SIZE,\n",
    "            edgecolors='k',\n",
    "            linewidths=1,\n",
    "            label='SI',\n",
    "            )\n",
    "\n",
    "# gem\n",
    "df = combined_df_gem[combined_df_gem['dense_size'] == DENSE_SIZE]\n",
    "df = df[df['gem_memory_strength'] >= 0]\n",
    "df = df.groupby(['gem_patterns_per_experience', 'gem_memory_strength']).mean(numeric_only=True).reset_index()\n",
    "df = df.sort_values(by=[X])\n",
    "\n",
    "# regression line\n",
    "slope, intercept, r_value, p_value, std_err = linregress(df[X], df[Y])\n",
    "plt.plot(df[X], slope * df[X] + intercept, '--', c='k')\n",
    "\n",
    "\n",
    "plt.scatter(df[X],\n",
    "            df[Y],\n",
    "            c=df['gem_memory_strength'],\n",
    "            cmap='Greys',\n",
    "            norm=LogNorm(),\n",
    "            s=MARKER_SIZE,\n",
    "            edgecolors='k',\n",
    "            linewidths=1,\n",
    "            label='GEM',\n",
    "            marker='*'\n",
    "            )\n",
    "\n",
    "# replay\n",
    "df = combined_df_replay[combined_df_replay['dense_size'] == DENSE_SIZE]\n",
    "df = df.groupby(['replay_mem_size']).mean(numeric_only=True).reset_index()\n",
    "df = df.sort_values(by=[X])\n",
    "# regression line\n",
    "slope, intercept, r_value, p_value, std_err = linregress(df[X], df[Y])\n",
    "plt.plot(df[X], slope * df[X] + intercept, '--', c='purple')\n",
    "\n",
    "plt.scatter(df[X],\n",
    "            df[Y],\n",
    "            c=df['replay_mem_size'],\n",
    "            cmap='Purples',\n",
    "            norm=LogNorm(),\n",
    "            s=MARKER_SIZE,\n",
    "            edgecolors='k',\n",
    "            linewidths=1,\n",
    "            label='Replay',\n",
    "            marker='^'\n",
    "            )\n",
    "\n",
    "\n",
    "# plt.legend()\n",
    "\n",
    "# axis labels\n",
    "plt.xlabel(X)\n",
    "plt.ylabel(Y)\n",
    "# equal axes\n",
    "# plt.gca().set_aspect('equal', adjustable='box')\n",
    "# despine\n",
    "sns.despine()\n",
    "\n",
    "# increase size of tick font\n",
    "plt.xticks(fontsize=18)\n",
    "plt.yticks(fontsize=18)\n",
    "\n",
    "# x = np.linspace(0, 100, 100)\n",
    "# y = np.sqrt(118/10) * x\n",
    "# plt.plot(x, y, linestyle='dotted', c='k', alpha=.5)\n",
    "\n",
    "# plt.savefig(f'{plotdir}/distance_to_first_in_plane_vs_null_space.png', bbox_inches='tight')\n",
    "\n",
    "plt.savefig(f'{plotdir}/distance_to_first_in_plane_vs_null_space.png', bbox_inches='tight')\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/var/folders/7n/wpwh80rn01z3xkb5bl7v9wyh0000gn/T/ipykernel_44233/2084229118.py:13: FutureWarning: The default value of numeric_only in DataFrameGroupBy.mean is deprecated. In a future version, numeric_only will default to False. Either specify numeric_only or select only columns which should be valid for the function.\n",
      "  df = df.groupby(['ewc_lambda']).mean().reset_index()\n",
      "/var/folders/7n/wpwh80rn01z3xkb5bl7v9wyh0000gn/T/ipykernel_44233/2084229118.py:20: FutureWarning: The default value of numeric_only in DataFrameGroupBy.mean is deprecated. In a future version, numeric_only will default to False. Either specify numeric_only or select only columns which should be valid for the function.\n",
      "  df = df.groupby(['si_lambda']).mean().reset_index()\n",
      "/var/folders/7n/wpwh80rn01z3xkb5bl7v9wyh0000gn/T/ipykernel_44233/2084229118.py:26: FutureWarning: The default value of numeric_only in DataFrameGroupBy.mean is deprecated. In a future version, numeric_only will default to False. Either specify numeric_only or select only columns which should be valid for the function.\n",
      "  df = df.groupby(['replay_mem_size']).mean().reset_index()\n",
      "/var/folders/7n/wpwh80rn01z3xkb5bl7v9wyh0000gn/T/ipykernel_44233/2084229118.py:33: FutureWarning: The default value of numeric_only in DataFrameGroupBy.mean is deprecated. In a future version, numeric_only will default to False. Either specify numeric_only or select only columns which should be valid for the function.\n",
      "  df = df.groupby(['gem_patterns_per_experience', 'gem_memory_strength']).mean().reset_index()\n",
      "/var/folders/7n/wpwh80rn01z3xkb5bl7v9wyh0000gn/T/ipykernel_44233/2084229118.py:39: FutureWarning: The default value of numeric_only in DataFrameGroupBy.mean is deprecated. In a future version, numeric_only will default to False. Either specify numeric_only or select only columns which should be valid for the function.\n",
      "  df = df.groupby(['lwf_alpha', 'lwf_temperature']).mean().reset_index()\n",
      "/var/folders/7n/wpwh80rn01z3xkb5bl7v9wyh0000gn/T/ipykernel_44233/2084229118.py:13: FutureWarning: The default value of numeric_only in DataFrameGroupBy.mean is deprecated. In a future version, numeric_only will default to False. Either specify numeric_only or select only columns which should be valid for the function.\n",
      "  df = df.groupby(['ewc_lambda']).mean().reset_index()\n",
      "/var/folders/7n/wpwh80rn01z3xkb5bl7v9wyh0000gn/T/ipykernel_44233/2084229118.py:20: FutureWarning: The default value of numeric_only in DataFrameGroupBy.mean is deprecated. In a future version, numeric_only will default to False. Either specify numeric_only or select only columns which should be valid for the function.\n",
      "  df = df.groupby(['si_lambda']).mean().reset_index()\n",
      "/var/folders/7n/wpwh80rn01z3xkb5bl7v9wyh0000gn/T/ipykernel_44233/2084229118.py:26: FutureWarning: The default value of numeric_only in DataFrameGroupBy.mean is deprecated. In a future version, numeric_only will default to False. Either specify numeric_only or select only columns which should be valid for the function.\n",
      "  df = df.groupby(['replay_mem_size']).mean().reset_index()\n",
      "/var/folders/7n/wpwh80rn01z3xkb5bl7v9wyh0000gn/T/ipykernel_44233/2084229118.py:33: FutureWarning: The default value of numeric_only in DataFrameGroupBy.mean is deprecated. In a future version, numeric_only will default to False. Either specify numeric_only or select only columns which should be valid for the function.\n",
      "  df = df.groupby(['gem_patterns_per_experience', 'gem_memory_strength']).mean().reset_index()\n",
      "/var/folders/7n/wpwh80rn01z3xkb5bl7v9wyh0000gn/T/ipykernel_44233/2084229118.py:39: FutureWarning: The default value of numeric_only in DataFrameGroupBy.mean is deprecated. In a future version, numeric_only will default to False. Either specify numeric_only or select only columns which should be valid for the function.\n",
      "  df = df.groupby(['lwf_alpha', 'lwf_temperature']).mean().reset_index()\n",
      "/var/folders/7n/wpwh80rn01z3xkb5bl7v9wyh0000gn/T/ipykernel_44233/2084229118.py:13: FutureWarning: The default value of numeric_only in DataFrameGroupBy.mean is deprecated. In a future version, numeric_only will default to False. Either specify numeric_only or select only columns which should be valid for the function.\n",
      "  df = df.groupby(['ewc_lambda']).mean().reset_index()\n",
      "/var/folders/7n/wpwh80rn01z3xkb5bl7v9wyh0000gn/T/ipykernel_44233/2084229118.py:20: FutureWarning: The default value of numeric_only in DataFrameGroupBy.mean is deprecated. In a future version, numeric_only will default to False. Either specify numeric_only or select only columns which should be valid for the function.\n",
      "  df = df.groupby(['si_lambda']).mean().reset_index()\n",
      "/var/folders/7n/wpwh80rn01z3xkb5bl7v9wyh0000gn/T/ipykernel_44233/2084229118.py:26: FutureWarning: The default value of numeric_only in DataFrameGroupBy.mean is deprecated. In a future version, numeric_only will default to False. Either specify numeric_only or select only columns which should be valid for the function.\n",
      "  df = df.groupby(['replay_mem_size']).mean().reset_index()\n",
      "/var/folders/7n/wpwh80rn01z3xkb5bl7v9wyh0000gn/T/ipykernel_44233/2084229118.py:33: FutureWarning: The default value of numeric_only in DataFrameGroupBy.mean is deprecated. In a future version, numeric_only will default to False. Either specify numeric_only or select only columns which should be valid for the function.\n",
      "  df = df.groupby(['gem_patterns_per_experience', 'gem_memory_strength']).mean().reset_index()\n",
      "/var/folders/7n/wpwh80rn01z3xkb5bl7v9wyh0000gn/T/ipykernel_44233/2084229118.py:39: FutureWarning: The default value of numeric_only in DataFrameGroupBy.mean is deprecated. In a future version, numeric_only will default to False. Either specify numeric_only or select only columns which should be valid for the function.\n",
      "  df = df.groupby(['lwf_alpha', 'lwf_temperature']).mean().reset_index()\n"
     ]
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 800x400 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 800x400 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 800x400 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# for each algorithm, get closest mean stability to reference stability and parameters\n",
    "\n",
    "def get_closest_stability(avg_df, reference_stability):\n",
    "    # get closest stability to reference stability\n",
    "    avg_df['distance_to_reference'] = abs(avg_df['stability'] - reference_stability)\n",
    "    avg_df = avg_df.sort_values(by=['distance_to_reference'])\n",
    "    closest_stability = avg_df.iloc[0]\n",
    "    return closest_stability\n",
    "\n",
    "for REFERENCE_STABILITY in REFERENCE_STABILITIES:\n",
    "    # EWC\n",
    "    df = combined_df_ewc[combined_df_ewc['dense_size'] == DENSE_SIZE]\n",
    "    df = df.groupby(['ewc_lambda']).mean().reset_index()\n",
    "    closest_stability_ewc = get_closest_stability(df, REFERENCE_STABILITY)\n",
    "    # collect runs\n",
    "    ewc_ref_runs = combined_df_ewc[(combined_df_ewc['dense_size'] == DENSE_SIZE) & (combined_df_ewc['ewc_lambda'] == closest_stability_ewc['ewc_lambda'])]\n",
    "    # SI\n",
    "    df = combined_df_si[combined_df_si['dense_size'] == DENSE_SIZE]\n",
    "    df = df[df['si_eps'] == SI_EPS]\n",
    "    df = df.groupby(['si_lambda']).mean().reset_index()\n",
    "    closest_stability_si = get_closest_stability(df, REFERENCE_STABILITY)\n",
    "    # collect runs\n",
    "    si_ref_runs = combined_df_si[(combined_df_si['dense_size'] == DENSE_SIZE) & (combined_df_si['si_eps'] == SI_EPS) & (combined_df_si['si_lambda'] == closest_stability_si['si_lambda'])]\n",
    "    # Replay\n",
    "    df = combined_df_replay[combined_df_replay['dense_size'] == DENSE_SIZE]\n",
    "    df = df.groupby(['replay_mem_size']).mean().reset_index()\n",
    "    closest_stability_replay = get_closest_stability(df, REFERENCE_STABILITY)\n",
    "    # collect runs\n",
    "    replay_ref_runs = combined_df_replay[(combined_df_replay['dense_size'] == DENSE_SIZE) & (combined_df_replay['replay_mem_size'] == closest_stability_replay['replay_mem_size'])]\n",
    "    # gem\n",
    "    df = combined_df_gem[combined_df_gem['dense_size'] == DENSE_SIZE]\n",
    "    df = df[df['gem_memory_strength'] >= 0]\n",
    "    df = df.groupby(['gem_patterns_per_experience', 'gem_memory_strength']).mean().reset_index()\n",
    "    closest_stability_gem = get_closest_stability(df, REFERENCE_STABILITY)\n",
    "    # collect runs\n",
    "    gem_ref_runs = combined_df_gem[(combined_df_gem['dense_size'] == DENSE_SIZE) & (combined_df_gem['gem_memory_strength'] == closest_stability_gem['gem_memory_strength']) & (combined_df_gem['gem_patterns_per_experience'] == closest_stability_gem['gem_patterns_per_experience'])]\n",
    "    # lwf\n",
    "    df = combined_df_lwf[combined_df_lwf['dense_size'] == DENSE_SIZE]\n",
    "    df = df.groupby(['lwf_alpha', 'lwf_temperature']).mean().reset_index()\n",
    "    closest_stability_lwf = get_closest_stability(df, REFERENCE_STABILITY)\n",
    "    # collect runs\n",
    "    lwf_ref_runs = combined_df_lwf[(combined_df_lwf['dense_size'] == DENSE_SIZE) & (combined_df_lwf['lwf_alpha'] == closest_stability_lwf['lwf_alpha']) & (combined_df_lwf['lwf_temperature'] == closest_stability_lwf['lwf_temperature'])]\n",
    "\n",
    "    # force a copy of all dfs\n",
    "    ewc_ref_runs = ewc_ref_runs.copy()\n",
    "    si_ref_runs = si_ref_runs.copy()\n",
    "    replay_ref_runs = replay_ref_runs.copy()\n",
    "    gem_ref_runs = gem_ref_runs.copy()\n",
    "    lwf_ref_runs = lwf_ref_runs.copy()\n",
    "\n",
    "    # concatenate all runs after adding algorithm column\n",
    "    ewc_ref_runs['algorithm'] = 'EWC'\n",
    "    si_ref_runs['algorithm'] = 'SI'\n",
    "    replay_ref_runs['algorithm'] = 'Replay'\n",
    "    gem_ref_runs['algorithm'] = 'GEM'\n",
    "    lwf_ref_runs['algorithm'] = 'LwF'\n",
    "\n",
    "    # concatenate all runs\n",
    "    ref_runs = pd.concat([ewc_ref_runs, si_ref_runs, replay_ref_runs, gem_ref_runs, lwf_ref_runs])\n",
    "    # select algorithm and distances\n",
    "    ref_runs = ref_runs[['algorithm', 'distance_to_first_null_space', 'distance_to_first_in_plane']]\n",
    "    # avg over algorithm\n",
    "    ref_runs = ref_runs.groupby(['algorithm']).mean().reset_index()\n",
    "    plt.figure(figsize=(8,4))\n",
    "\n",
    "    # ewc\n",
    "    plt.scatter(\n",
    "        ref_runs[ref_runs['algorithm'] == 'EWC'][X],\n",
    "        ref_runs[ref_runs['algorithm'] == 'EWC'][Y],\n",
    "        c='g',\n",
    "        s=MARKER_SIZE,\n",
    "        marker='s',\n",
    "        edgecolors='k',\n",
    "        linewidths=1,\n",
    "        label='EWC'\n",
    "    )\n",
    "\n",
    "    # SI\n",
    "    plt.scatter(\n",
    "        ref_runs[ref_runs['algorithm'] == 'SI'][X],\n",
    "        ref_runs[ref_runs['algorithm'] == 'SI'][Y],\n",
    "        c='r',\n",
    "        s=MARKER_SIZE,\n",
    "        marker='o',\n",
    "        edgecolors='k',\n",
    "        linewidths=1,\n",
    "        label='SI'\n",
    "    )\n",
    "\n",
    "    # Replay\n",
    "    plt.scatter(\n",
    "        ref_runs[ref_runs['algorithm'] == 'Replay'][X],\n",
    "        ref_runs[ref_runs['algorithm'] == 'Replay'][Y],\n",
    "        c='purple',\n",
    "        s=MARKER_SIZE,\n",
    "        marker='^',\n",
    "        edgecolors='k',\n",
    "        linewidths=1,\n",
    "        label='Replay'\n",
    "    )\n",
    "\n",
    "    # GEM\n",
    "    plt.scatter(\n",
    "        ref_runs[ref_runs['algorithm'] == 'GEM'][X],\n",
    "        ref_runs[ref_runs['algorithm'] == 'GEM'][Y],\n",
    "        c='grey',\n",
    "        s=MARKER_SIZE + 10,\n",
    "        marker='*',\n",
    "        edgecolors='k',\n",
    "        linewidths=1,\n",
    "        label='GEM'\n",
    "    )\n",
    "\n",
    "    # LwF\n",
    "    plt.scatter(\n",
    "        ref_runs[ref_runs['algorithm'] == 'LwF'][X],\n",
    "        ref_runs[ref_runs['algorithm'] == 'LwF'][Y],\n",
    "        c='blue',\n",
    "        s=MARKER_SIZE,\n",
    "        marker='D',\n",
    "        edgecolors='k',\n",
    "        linewidths=1,\n",
    "        label='LwF'\n",
    "    )\n",
    "\n",
    "    # set text sizes\n",
    "    plt.xticks(fontsize=14)\n",
    "    plt.yticks(fontsize=14)\n",
    "    # plt.legend(fontsize=14)\n",
    "\n",
    "    plt.title(f'Distance to first in plane vs null space - Stability = {REFERENCE_STABILITY}')\n",
    "    plt.xlabel(X)\n",
    "    plt.ylabel(Y)\n",
    "    sns.despine()\n",
    "    # equal axes\n",
    "    # plt.gca().set_aspect('equal', adjustable='box')\n",
    "    plt.xlim([0, 110])\n",
    "    plt.ylim([0, 300])\n",
    "    plt.savefig(f'{plotdir}/distance_to_first_in_plane_vs_null_space_stability_{REFERENCE_STABILITY}.png', bbox_inches='tight')"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "pt",
   "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": "3.10.6"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}