logbook-files/outputs/executed/predator-prey.repeat.no-tex.executed.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "id": "0f8fa762",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2026-07-21T14:52:17.021657Z",
     "iopub.status.busy": "2026-07-21T14:52:17.021429Z",
     "iopub.status.idle": "2026-07-21T14:53:24.608448Z",
     "shell.execute_reply": "2026-07-21T14:53:24.607173Z"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Starting Multi-Agent Simulation (3 Predators vs 1 Prey)...\n",
      "Monitor uses Mixture Martingale over grid: [0.1, 0.3, 0.5, 0.7, 0.9]\n",
      "True Eps   | Avg Tau   \n",
      "------------------------------\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "0.05       | 2413.6    \n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "0.10       | 636.7     \n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "0.20       | 156.4     \n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "0.30       | 66.6      \n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "0.40       | 41.8      \n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "0.60       | 16.7      \n",
      "0.80       | 9.8       \n"
     ]
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAe4AAAEmCAYAAACkrHOPAAAAOnRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjExLjEsIGh0dHBzOi8vbWF0cGxvdGxpYi5vcmcvctoD+AAAAAlwSFlzAAAPYQAAD2EBqD+naQAAfjZJREFUeJzt3XdYU9cbB/Bvwp5hD9kICIp74d6rtepPrauuKo6K22rdq1brrLbaOqq2ddXaqnXPugfOqihTREH2DBuSnN8fNJEYCEECucD7eZ48D9xx7pu8SU7uvWfwGGMMRCmJRILY2FiYmJiAx+NpOhxCCCG1EGMMmZmZ0NZ0IFy2bds2bNu2DQUFBXj58qWmwyGEEELAozPusmVkZMDMzAyvX7+GqamppsOpVSQSCZKTk2FlZQU+n6/pcMh/KC/cQznhJnXmRSgUwsXFhc64VSG9PG5mZkYVdxWTSCQoKCiAmZkZfRlxCOWFeygn3KTOvEj3p+wSQggh1QhV3IQQQkg1QhU3IYQQUo3QPe5ykEgkkEgkmg6jVpFIJGCM0evOMZQX7qGccJM68yItgypuJaTdwcRiMQAgKSkJeXl5Go6qdpFIJMjIyABjjBrccAjlhXsoJ9ykzrxkZmYCKGd3sNjYWISGhsLd3R0uLi4QiUTQ1q75db9QKIRAIEBaWhq1Kq9iEokESUlJsLa2pi8jDqG8cA/lhJvUmRehUAhzc3PVzrhPnz6NJUuW4PHjxwCAXbt2wd/fHzNnzoSBgQHWrVtXK0YU4/P59IHQAB6PR689B3EpL4wxiMViiEQiTYeiMRKJBCKRCAUFBZzICSlSVl50dHSgpaWlUlnS/cusuH/88UcEBASUuK5ly5YICAiAoaEhVqxYodKBCSFEXRhjSE9PR1JSkuyWVm0lvY+amZlZK06kqgtV8mJmZgY7OzuV86a04o6Li8OcOXPQtGlTLFiwAC1btsTKlStl68eMGQMHBwcMHjwY8+fPh4GBQTmeDiGEVEx8fDzS09NhamoKU1NTaGtr19pKizEmu31ZW18DLlKWF8YYcnJykJiYCACwt7dXqUylFfexY8fg5uaGGzduwMjICAAUTvW7d+8OT09P3L9/Hx07dlT5yRBCSEWIxWJkZGTA2toaVlZWmg5H46ji5qay8iI94U1MTISNjY1Kl82V3ggJDQ3FkCFDZJV2aby8vBAbG1vmwQghRF0KCwvBGCvz+4kQrjM0NARQ9J5WhdKKWyKRwMzMrMxCEhISVDoYIYSoG51dkuquvO9hpRV33bp18ccffyht9BEbG4t79+7By8urXAcmhBBCSPkpvcfdv39/zJ07Fz169MDXX38NPz8/ufX379/HlClTYGdnhyZNmlRmnJxAI6dVPRoNipu4kBdpDNIHgex1oNeDW8rKi/Q9XFYdo9LIaW5ubpg3bx5Wr16N9u3bw8DAALq6ujh9+jTmzJkDoVAIPp+Pc+fO1ch+gzRymubRaFDcxIW8FBYWyvrI1ub+21LSvuwA3T7gElXyIhKJIJFIkJKSAh0dnVLLUnnkNMYYVq1ahTVr1iA3N1dunb29PXbu3Im+ffuW64lUNzRymubQaFDcxIW85OXlISoqCm5ubtDX19dIDJoSFRWFDh06lGufq1evYsuWLTh27BiAoh5Be/furYzwar3+/fvj0aNHAIDx48dj0aJFSivkvLw8vHr1Cq6urkrfyyqPnMbj8bBkyRJ88cUXuHTpEqKioqCtrY369euja9euteoDU9FRosQShnuvUpGYmQcbE320crOAFp9+GZeFSyN0kXc0nRc+nw8ejyd71CZisRhv375FdHS0bBljDI0bN0Z+fj5CQ0Pltnd1dYVIJMLKlSvx5ZdfwtXVFcnJybXqdbt79y7GjRuHr7/+GoMGDarUY+3evRuhoaHo0KED0tPTZa9zaa+39D1c1udJ5ZHTpKysrDBs2LDyxF6rJQrzkJiZL/v/dkQydt6IRHJWgWyZlbEuJnZwR1uPd31QbUz0YGNae34MEUI+nKOjo+xv6W0LHo8nt7w4c3NzmJiYVFV4Verq1asYOXIkVqxYgfHjxyusj4mJQXBwMCIiIio9Fmtra2RkZFRa+WqZIWT79u3w9PREt27d1FFcjXAg8A22XA5Xuk1yVgFWnw2RWzajmydm9aAW+oSQ0rm5ucmdbZfl9evXsLW1rcSINC8vLw9v376V3Qd+3+DBg5Geng6BQFDFkamfWiruBw8e1IpZwsrjs9bO6FHfFmIJw/hf78udab/PylgXu8e0hBafBxsTvSqMkhBSHWlpaZV6Vl0SBweHSoym+qgJlTagQsUdGRmJb7/9Fnfv3kV6enqJTdXT0tIUuorVdjam+rAx1cedlylKK22g6Mw7p0CMNnUtqyg6QmqHuKw4xGXHlWsfV4ErLPQt5JYVigvxLPlZucox1jWGl7ni1bOX6S+RkZ8BeyN72BurNjZ1ZXn69CmWL1+OZ8+ewcbGBkOGDMG0adMU7rNGRERg06ZNuHXrFrKysuDo6IgBAwYgICAAurq6ctu+evUK3333HW7cuIGMjAzY2dmhT58+mDFjhqxx7/Tp03H06FEARY3kAgIC8O233+LZs2fIyclBmzZtcOTIEZWP3a9fP9y5cwcAsHz5cmzYsAFAUSOxbdu2KRzvl19+KTXmzMxMuLi4oFWrVvD390fdunVl2+Xn5+PQoUM4evQowsLCoK2tjVatWmHu3Lnw8fFRR0pUorTiTk9PR4cOHWg40wpIzFSt+5iq2xFCVHcs4hh+evJTufZZ22EtPnL/SG5Zen46xpwbU65yWti2wN7eiq22Nz/cjKsxV/FF4y8wpcmUcpWpTq9fv8aKFSswd+5cmJqaYuvWrZg5cyb09PQwefJk2XbXrl1D37594eHhgY0bN8LBwQFXr17FvHnzcPLkSVy4cEF2xfXGjRv4+OOP4ebmhrVr18LV1RX37t3D7NmzsW/fPly/fh12dnZYsWIF5syZA1dXV9y4cQNxcXFYuHAhHBwcsHHjRgQHB5fr2Lt378b58+cxatQozJo1S3aPWzqUaPHjJScny70O0pjd3d3x7bffom7duoiMjMTixYuxdu1avH37Vjb5R0BAAA4fPowffvgBbdu2RWxsLJYtW4aWLVvizp07aNiwYaXnDSij4t6/fz+SkpKwZs0a9O7dG9bW1iUOgD5nzpxKC7C6szFRraGZqtsRQog6REVF4caNG7Jhrbds2YK9e/fizz//lFXchYWFGD16NEQiEU6cOAEnJycAgI+PD/Ly8jB79mxs374dU6dORWFhIUaOHImCggKcO3dOVtl5e3tDIBDIzpL/+usvuUZyycnJePjwoSyOzZs3QygUluvYxSeaEQgECrcRSmuUJ425sLAQ586dg52dHQDA09MTDRs2hJOTk9zIoSKRCEuWLMHYsWMBFM3TIY1t5cqVsqsElU1pP47w8HBMmTIF8+fPR5MmTeDg4AA7OzuFR7169WBpSZd5i0sU5iHobQYMdbVgZayrZEsGM0M+DHW1EPQ2A4lCOvMmhFS+hg0bys1FoaurCzs7O8TExMiW3blzB2/evEGbNm1kFafUxx9/DAD4448/5LZt166dwvSU/fr1g56eHk6ePImcnBy5dY0bN5aLQ09PD9bW1uU69ocqHrO00paqU6cOYmJi5J7Ld999hxkzZshtJxAI4O3tjcDAwArFUh5Kz7gFAoFKk4wsXbpUXfHUGKq0Ki/CQ5bObfTfJgbAo1blhKjR/zz+Bz/78rW/cRW4Kiwz0zPDr71/LVc5xrrGJS6f2XwmPvf9HPZGmr2/bWFhobBMW1tbbhQ6aSUeGBiocBYrHbsrJCREbts6deoolMvj8WBnZ4fXr18jISEBbm5usnWlDWpVnmN/KOkxSpsH+/3lWVlZ2LBhA65du4aEhATZoGRJSUmyy/JVQWnFPXDgQCxcuBCzZ89WWsi2bdtQr149dO/eXa3BVWfSVuVSJfXjlpLk18HSIQytbDtQq3JC1MjeWD0NwHS0dNDMtpkaIgLqmtUteyOOkFbuXbp0wfbt20vcRtqQzdzcHACQkpJS4napqaly26nz2MCHDfNaVszFJSQkoGnTpmCMYcuWLWjRogWMjYt+nA0cOBBhYWHlPv6HUlpxN2nSBD179sSoUaMwY8YM+Pj4lDj37ePHj6GnV/MrnPJMMmJlrCt3iby+vQk+b+eK+1GpSMzMx293IvDoTXZRuXnO2P3gEj4b1xM6fB2aUKMYLkxmQRRxIS80yUjplL0exdcpm/QCANq2bQtTU1OEhITAzs5OoY3ToEGD4OnpiW+//Rbt2rWDqakp7t27h6ysLLm64sGDB8jMzETbtm0hEAgUclZSHOU5NgBZHSSdpz0sLAzdunXDkydPYGFhUeLxisecnZ0td9YcGxuLNm3a4NChQ2jbti2uXbuGlJQUBAQE4LPPPpOLRfqjobTnVKWTjABA06ZNsWnTJuzfvx8ASqygRSJRjewOVhmTjLgbA+7G2nDr4Y6Be/+FWFL0Zox90wo7An/BkLr9Kxx3TcKFySyIIi7khSYZkVe8UlD2ekjXMcZK3U663MDAABs3bsSECRMwZcoUfPPNNzAzM0N2djbWrl2LK1euYOXKlRCJRDAwMMC6deswefJkjB8/Hj/88APMzc0RHh6OCRMmwMjICJs2bZKVXVYc5Tk2UNSgjM/n499//5U1aNPV1YWpqance6T48YrH7O/vjy1btsDc3BxxcXEYN24cHB0d0bJlS4hEIvj6+kJbWxtnz55FUFAQvL29wRjD0aNH8fjxYxgaGso9D+nfEolE7ZOMgCkRFRXFDA0NGYAyH7t27VJWVLWWkZHBALC0tDQmFovV9vj27BPm8tUp2cN340KWlqveY1T3R2FhIYuNjWWFhYUaj4Ue3MpLdnY2e/78OcvJyWESiaRWP8aMGcMcHBwYn89nPB6POTg4sJEjRypst3LlSubg4MAAMH19febg4MCCgoLYqVOnmIODA9PS0mJaWlrMwcGBHT58WLbfqVOnWMuWLRmfz2eWlpbMwMCAffTRR+zff/9VOMbff//NWrRowfh8PhMIBExXV5d99NFH7MmTJ0rjmDBhQonPrTzH3rBhAzMwMGBmZmbM3t6eXbx4kUkkEjZ16lSF4509e7bEmC0sLJiRkREbN24cS05Oliv/2LFjrH79+ozH4zFzc3NmZ2fHpk+fzpo3by573bdu3co++eQTZmdnxwAwY2Nj2etcWv5ycnLY8+fPWXZ2ttL3fFpaGgPAlM4OtnbtWqxatQrbt29H9+7dS50JyN/fH35+fvD39y/1l0J1Jp0dLCMjQ62zg+UWiOG39jQysv+7BMQTYUyvN1jReZrajlHdSSQSJCYmwsbGhs64OYQLeZHOqFQbZwd7X0pKCnJyciASiaCtrQ0ejwd9fX1ZFympjIwMhSFBbW1tIRKJFO7zWlhYKDS4ys3NRVZWFqysrMq8p5yTk4PMzExYWFgonEWWFIehoWGJDebKe2yJRIL09HS5stLS0pCdnS23nZWVlcL7JicnBzk5OQqv2/vy8vIgFAphZWUFPp+PpKQk5OcXzU0hEAiQl5cn+5/9d4bv4OCgMFhN8fJUeS9L6yKll8qjo6MxdepUhev57/P19VVork/KZqCrhZWfNMWM358WLWDaCH/pC3TWaFiEkGrG0tISFhYWchV3SQQCQYnDfuro6Kg0hKqBgQEMDAxUisnQ0LDUltalxaGOY/P5fIUfAObm5io1ilMWc3H6+vpyFay1tbXc+uJ9xqUVtzqHBVf6U9nOzk6l/tkzZ85Er1691BZUbdKvsSO867xL6O3wHNyKSFayByGEkNpMacU9fPhwnD9/Xm7kmJKsWbMGJ06cUGtgtQWPx8PGwX4oPi33ipPPIRJTK2pCCCGKlFbcderUwYgRI9C/f3+cOnUKL1++RHp6usIjODgYiYmJVRVzjdOgjgDDWjnL/g9LyMLBe280GBEhhBCuUnrRfdq0adi9ezcA4PTp00oL6tixo/qiqoXm9PDCySexyMwr6kKw8UIYevlawdak5NGXCCGE1E5lNgc1NTWFra2t0oeqjRVI6SyN9TCz+7uhTjNyC/HRz9tQKC7UYFSEEEK4psyKe+PGjYiPj1f6GDFiRFXEWuONbuMCB/N3owOlJNTH5rtVM9sMIYSQ6kFpxe3q6qrQzL0kTZs2haurq7piqrV0tPhY3Lf4ZOx8/HwlDel56ZoKiRBCCMcovce9ePFilQoJCAhQSzAE6NPABd6OjxESUzRgQX6WK+afO4TtA77QcGSEEEK4QC1DHk2fPh379u1TR1EEwA9DOoPHe9cF78JDY0SkvtZgRIQQQrhC6Rl3bm6ubNg2ZdLT01XaThNiY2Oxb98+vHr1CnZ2dvj000/RoEEDTYellKeNGXo11sO5f/8bpL7QAtOPHceZ8TPK2JMQQkhNJ3fGfeTIEdjZ2SE4OBhAUXcw6VBxyh5cPds+duwYvLy8EB0djSZNmiAsLAyNGzfG5cuXNR1amdb17wIdnXczkb2IcMKllw80GBEhtUOiMA9BbzPK/UgUVmzmQEJUJXfGff/+fSQkJCA4OBg+PkWNpExNTcvs7iUUCisvwgp4/fo1xo4di61btwIAJk+ejOfPn+OPP/5At27dNBydcqYGugjo6oDN5/8b+J/pYcHfd9FtVvMPmjCeEKKaA4FvsOVyeLn3m9HNE7N6eJW9ISEVJFdxz58/H/Xr10f//u/mhN64cWOZs35xdVawnj174pNPPpH9n56ejoSEBDRv3lyDUalueqfW+O3uYaRmFA1Yn5Tohp8Cz2GKXx8NR0ZIzfVZa2f0qG8rt+x5bAa++uuZ7P+1gxqiQR35STJsTPSqJD5C5CpuCwsLjB07Vva/qt3BVN1OmRMnTuDMmTOYOXMmvL29lW6bkpKCw4cPIzw8HObm5ujXrx+aNGmisF39+vUBFI36tnv3bkRERGDFihWYMGFChWKtKnw+D2sGNMekX8Nky76/GIfPm+fBQKd2T2NISGWxMdWHjan85yu3UH6+BndrY/g6lG92K3V7+PAhFi1aJLeMMVbiFTlbW1v8+uuvVRVatTdr1izZLeP//e9/mDRpkoYjkqe0VfnixYvlzr4rul1JwsLC0KdPHwwaNAg7duxATEyM0u3v3LkDHx8fHDhwAE5OToiNjUXLli2xbNmyUvdxdnZG9+7dUb9+fWzevBlPnjz5oFg1oZePJ7yd381bm5dtiwVnjmkwIkIIF7i4uGDmzJmwsLDA+fPn0bBhQ0ybNg0zZszAzJkzZY/OnTvj2rVrmg4XABAREYEhQ4bgzp07VbpveQ0dOhRDhw7F+fPnZRU4l8jOuBcuXIg//vhDtmLNmjX49NNPK/XgZ86cwZgxY7Bs2TLY29tj7969SrcXCoUYMGAA3NzccO3aNdn8pl5eXpgzZw4aNmyIwYMHK+zXsGFDNGzYEFOmTMHYsWMxYsQIvHjxolKeU2XY+mkP9PjuOpikaBL2a0/NkNVbBGM99c3vSggpnVjClP6vCVZWVujduzdu3rwJAGjUqBF69eqlMB+3OueBrqgrV67gyJEjcHNzQ5s2baps3/Ly8/ODlZVVpR6jImRn3ImJiWjWrBkmT56MyZMnl3m5Wh1sbW3x77//YurUqeDzy+5S/uOPPyIxMRFffvml3JtxypQpMDU1xcqVK+W2X79+PdLT0+WWtWjRAsHBwcjLqz4tQD2sLdG3ma7s/7RsCbb+E6HBiAipPc4FxSHgwCO5ZQEHHuFcUJyGIiqfpk2bcuYy+fjx43Hz5k2F7+rK3remkaste/bsiS+//BJffvklGjZsWOkHb968ORwcHFTe/vjx4wCALl26yC3X19dHmzZt8OzZM0REvKvQzp8/j19++UX2v1gsxokTJ+Du7g59/ep1j3h9/95wNH/Xun/PzVeISs7WYESE1HznguLwxf5HSMkukFuekl2AL/Zzu/JOS0tD7969YWRkhE6dOmk6HAAAn89Hu3btoKdX/oZ8Fdm3puHONZQySCQSPH36FAKBoMRLGHXr1gUAPHnyBB4eHgCAQYMGYdGiRThy5AgcHR3x8OFD5OTk4PDhw0qPlZ+fLzegjLS7m0QigUQiUddTKhddLR4W9PFGwMHHAIACsQSrTr/AzlHVo4X8h5JIJGCMaex1JyXjQl6kMUgfxQnzChEan1nKnqqWz7DoWBBKuyjOACw6FgQzAx3w+R/eRbOenQlM9XU+eP/iz136N2MMeXl5OH/+PAoLC+Uqu/Xr1+Off/4BALRq1QrTpk3Dli1b8Pz5c9jZ2WHChAlo2rQp3r59ix9++AFhYWFwdHREQEAA6tWrJytn1qxZCAkJAQAMGDAAHTp0wM8//4xXr17B1NQUffv2xeDBg2WX7d8/7ooVK0pc/sUXX+Cnn35CUFAQcnJy0LBhQ1hbW5e4r1ROTg7279+PmzdvIjMzE87OzmjVqhUGDRqkcJL24MEDHD16FOHh4dDW1kbLli0xbtw4mJmZlfi6lvT+Kq/iZZW2Xvp5UvaZkq5TS8U9f/58NG3aFEOHDlVHcSXKyMhAbm4u7OzsSlxvbm4OAIiPj5ct++KLLzBq1Cg8fvwYsbGxmDp1Klq1alXmL7Y1a9YovDEAICkpSaOX2JtZ89DM0RiPYrIAAJeCE/H3vTC0cTXTWEyVTSKRICMjA4wxlW6nkKrBhbwUFhZCIpFAJBJBJBLJrXvxNh3Df75f6TGkZBdg2K7ACpVxyL8lWriYf/D+0i/z9evX47fffgOPxwOPx0NBQdFVgvdfn27dusHb2xv9+/dHRkYGHj58iAEDBsDT0xPr16/H3r17sW/fPuzcuRNDhw5FvXr18O2332LPnj0ICgqCo6MjAGDw4MEICQnBhAkTkJaWhu3bt2PChAlo0aIFTp8+jaFDh2LEiBGyq57Fj6ulpSWLqfjyuLg4XLhwASNGjMCIESOwe/du3Lt3D5s2bSpxX6Co0Vrfvn2RmZmJOXPmwN3dHa9evcL8+fMxffp0PH/+XHayt3DhQnz33XeYPn06hgwZgri4OGzZsgXff/89rl27JncFWHoMxpjC+6s8GGMQi4t6JZQ2BodIJIJEIkFKSgp0dEr/EZeZWfRjVC0Vd3JysqzAyiI9Ay7tSenqFt0Dzs3NlVtubGyMDh06lOtYCxYswOzZs2X/C4VCODk5wdraGqampuUqS92+/p8BPtl6C9K2MYsuPMLD+YOhx6EGKOokkUjA4/FgbW1NFTeHcCEveXl5yMzMhLa2tkIDLC0trVL24h4tLa0KNSCTvv69evVCx44dZWWlpaXh6tWrCq9Ps2bN0KhRIwDAo0eP8PLlS1mF5eHhgfbt22PcuHGIjIyUVXiOjo7o1asXDh48iIULFwIA2rVrB1vbov7u4eHhiIiIkJ1AjRgxAnw+HwcOHEDv3r0xcuRIuePyeDxZTMWXh4aGIiQkBC4uLgCAzp074+XLl6XuyxjDZ599hqioKDx9+lTWBVj6ejRp0gQikUi2fVJSEmbNmoV169bJthswYADq1auHVatWYdeuXbLl0n2KH68ilFXI2tra4PP5sLS0VHobV7pOLpply5Zh8+bN5Q4oNjYWfn5+5d6vPAwNDQGg1DHRpcuNjIwqfCw9PT3O3kfxsTdFV189XHpW9HxzcgRYcvYC1n3ykYYjI4RokrRVubSCKH71sTSNGzeWO8uUNkpu2bKl3C1JaYUYFRVVYjk9evSQVdpSY8eOxYEDB3Dw4EGMHDlSpefQpEkTWaUNAJaWlrC0tCx1+4cPH+Lx48do166dXKUNFL0ely5dkhtjZPbs2bIfG1LOzs7w8fHB1atXVYqRC+Qq7tjYWMTGxmoqFqVMTU0hEAiQlJRU4nrpcmdnZ7Udc9u2bdi2bZvsMoemL5VLjW1mhcvPI8AkRY3Vjt7Lx6gGr2BrXPEfLVzDhUuyRBEX8qLsUrmHlQEO+besUPkSCcO0w0+Qml1Y6jaWRrr4fmijCt3j9rAyqNClWOmlcrFYLHdJ1tTUFCdPnoSOjo5C+dL/rays5NZJc2lpaSm3XHoFIz8/X2659G8bGxuFY9SpUwcA8PLlS9m60i4/S/9+/7glxVx8X+k9dicnpxL369ixo9y+bm5uOH78OG7cuIGEhATZ97n0B0lJMXH+UvnEiRMVWmyrYufOneXe50O0aNECly9fRnR0NJycnOTWhYSEgMfjoVmzZmo7XkBAAAICAiAUCiEQCDhxqRwo+pB0bvwvrjwuqrglYn2svPYQRz5X7MNe3XHhkixRxIW8KLtUbm6sDT9j5XMsqOKbAQ0x5cCjEhuo8QCsGuCLtp42FT5ORUhffy0tLWhpacm++LW1tfHxxx+XuW/x10769/uX76V/l7Z9bm6uQg6klaKRkZHCutIuP6tyWbr4NiYmJqUe/31ZWVnw8/NDeHg4Zs2ahR49esDY2BgAMHfuXMTExJT43Dh/qbxly5YYNmxYuQO6dOlSuff5EMOGDcPly5dx/vx5ufHR09PTce/ePXTq1An29vaVdnw+n8+ZyuO7/oPQMvgoCvOKLmc9DDfAvTdv4efqVMae1Q+Px+PUa0+KaDovfD5f1hCrsibe6dPQHj+NbIZFx4LkuoRZGunim//5ordv5X3fqKr4c5f+XXxZQkICxowZgzNnzshyVdI+Ffn7+fPnCjl4/LioB0ybNm1KjEuVv8t6nm3btoWOjg4eP34MiUQi17YhIyMDw4cPx4oVK9CyZUtcuHABL168wMSJE7F27Vq5sletWoWYmJhSX8uKvL+KD0Or7Lmp8nmSrlPLJ27s2LHlbgD2IcaMGQNfX1+sWbMGqampsuWLFy+GWCzG6tWrK/X40qb6XHiY6hphTMfil8b5mPXnLYjFYo3Hpu5H8W4S9ODOgwt5Kd4drLIevRrYYduIpnLfBdtGNEWvBnZVcnxVHlLvd2FijCEnJwfnz5+HWCwudZ+KLg8KCsK+fftky16/fo01a9bAyMgIs2bNqlD5yraxtrbGnDlz8ObNG6xcuVK2vKCgAF9++SVCQkLQqFEjMMbg6uoKoKhBXlZWlmzboKAgPH36tNSY1JEbVctS5T0PqKlVefv27T9oP4lEgilTpgAAbt26BQDYsmUL/vzzTzRr1gwTJ06U215HRwenTp3CwIED4evrix49eiAsLAzPnz/HgQMH1D4MHlfvcUt97t0Gf9w7BmG6GwAgLtkE2/65jSENPTUcmfpIJJq/l0oUcSEvyu5xqxtjEoX/K/uYZXn06BGWLl0qG3Tq/e5gUtKeNtJ49+zZg7/++gsAcO/ePfTu3Rvff/89bty4gd9//x0AcO3aNfTu3Ru7du3CqVOn8OeffwIALl68iN69e2Pbtm1wcXGRlTlq1CicO3cO33zzDczMzPDs2TNYWVnh+PHjcHV1hUgkwsaNG2V9saXHXbx4MV68eKEQT7t27bBgwQLZcyhtXz8/P6xYsQLGxsayrmyurq4IDw+Hi4sLTp06BT6fD5FIhEaNGmH16tX45ptv4OjoiIYNGyIvLw8WFhZwdHREREQEevfujTFjxiAwMBCPHhWNlvf3338jJCQE33//Pdzd3cudJ8bUf4+bx/77KeDv7w8/P78qnaKTMYYdO3aUuK5u3bro0aNHqfvdvn0b4eHhMDMzQ9euXSv13rP0HndaWhon7nEXd/T5DXx5MB1gRb/BDPRz8HDBAOjrVJ/uMMpIJBIkJSXRPW6O4UJe8vLyEBUVBTc3t0ofCfF+VCqG7Lgr+/+PSX5o6WpRqccsS3JyMh48eCC3rHjXp/f16tULPB4PoaGhePXqldy6du3aIS4uDpGRkXLLO3bsiDdv3ii0Jm/fvj2MjY0REREBLy8vTJ8+HZs3b0ZMTAzCw8NhamqKJk2ayF26fvr0qULj52bNmiEtLU0hHjs7O7kZH0vb18bmXfuCgoICPH36FFlZWXB2di61ks3MzERoaCiEQiHc3d3h6uqKwMBApKWlAQDq1auHhIQEheGy27VrJ7unXl6FhYVKK+S8vDy8evUKrq6uSt/LQqEQ5ubm8hX3wYMHZf2hf/jhB4waNeqDgqxppBV3RkYG5ypuxhi6/LQFUW/enWUPbWOItf3L38iQiyQSCRITE2FjY0MVN4dwIS/SL7uqqrg/3f5uVqojk9tovOJ+H2NMVnFX1j3/90VERMDT0xMzZsz4oK7EtYEqeVH1vSyti2Q/zcaNGyd3ybuy+2UT9eDxePjuf70wcOu/YOKiHxV/3BXCxzoS9WwFaOVmAa0KdFUhpLZJFOYhMVN+vIjIpCyF/w3eu6plY6KnMI83IZVBVnG3bdsWbdu21WQsnFe8cQAXSL9gtGCHRl5xeBJcVHEzpo3lJ4rmkLU00sWEDm5oW/fdIAbV6QumeCMowh1cyMv7jdPU5UDga2y5rHz2va/+eqawbEY3D8zs7qW2OD7E+42hKtOsWbNk94GPHz+OkJAQbNu27YPuA9d0ZeXl/cZppZGuk10qJ4qKN04LCwtDWFjYB9/jqAy77sRid+B/sxPxcwCJAYp6lyo3vrU9JrSpU7nBqYm0EZRAIKBL5RzChbwUFhYiIyMDLi4uar1UnpiZj6TMkkdoVMbaRA82JpobcVHaCEpLS6tKLpUHBgYq3Adu27Ytp74juUCVvOTl5eH169cQCARlNk7z8vKiilsVXG2cJj3jFksY/H97iJTsfJRWcVsZ62LXqObQ4vOq3Rm3phtBEUVcyEtVNk6rLspqBEU0Q92N02rmzBSVhGuDgNiZGcLOzBB3Xqb8NzhE6b+yk7MKkCdiaFOXWw1qVKHpgT5IyTSdl6oYgKU6YazsgT5I1VMlLxoZgIVoVmKman3LVd2OEEIId9EZdzlwtXFaVl7pkyAUFxYvxNNow2p3qVzTjaCIIi7kpbIap1VnVdk4jahO3Y3TqOJWgusjp8k1TlPBtquh2HY1AuNbO1S7xmk0chq3cCEvVTlyWnWgyghdpOpV6shpFfHFF1+gdevWGDt2bEWL4iSuN04DgNsvU/DtudAStmIofu+bzxNjRb/6+Kx19eiywYVGUEQRF/JSWFiIly9fwsHBgVOfS02ixmncVFZeMjMzERMTAw8PD6UzkZWrcZpEIsGrV6+QkZFR4ml8XFxcrfjFy7UGUtLGaQDQyMkcrlZGWHHyBeIy3l0VMDVkyBLHQZJfdIYtYVpY8ncICpkY49rW00jc5aXpRlCkZJrOi56eHvT19SEUCmFqalrrzzKpcRo3qZIX6fS0Ojo6SnMn/ayVWXHv2bMHX331FZKTk5Vu17dv37KKIpWst689etS3w71XqUjMzIONiT5auVngUtRlTPv9PgqEjf7bkoeVJyIQn5GDBb2b0IecVFtWVlZ4+/YtYmJiZH1ga+v7WRNDnpKyKcsLYwzZ2dkQCoWwt7dXOW9KK+6goCBMmjQJIpEITk5OsLGxKfE0/v2B6YnmaPF5aFNslDQA6OXeHQfHmWPM/pPITm4pW77zWiwSMnOxaXAbGhaVVEvSS+TJycl4+/athqPRLGnjJmk3OcINZeWFx+PBzMwMAoFA5TKVVtynTp2CjY0NLl++DG9v71K3q8oZxTSJa63Ky6OZTVMc/dwYww/8gtTYd2PS//0oDUmZ/2D3qI7Q0+bejGJcaL1MFHEpL8bGxjA2NkZhYaGsEVBtJJFIkJqaCgsLC7qtxCFl5UVHRwdaWloq9Y5QqVV5TEwMJk2apLTSBoB+/frB0dGxrPirHa63Ki8vAQT4td8ITD73G95GdYS0G//t8Dz023YO2wc2g5EetypvLrReJoooL9wjkUiQnZ0NbW1tygmHqDMvKrUqX7NmDXg8HubPn1+hg1V3XG1V/qGE+UJ89udaPH/RGmDvWjp62OrhwLh2sNbgeMvv40LrZaKI8sI9lBNuUmdeFObjLklkZCQ+//xzXLp0SWlT9jVr1qBBgwbo169fhYLiKi7Px/2h8kR58D/xLW4+8P1vcpIiLpaG2DeuNZwtDTUY3TtcmPeZKKK8cA/lhJvUmRdpXaS0FHt7e4wZMwYff/wxTpw4gYiICKSnpys8goODkZiYWKGASNXS19bHrwOW4H+domCg924mpNcpORi0/Taex2ZoMDpCCCGlUXqPe9q0adi9ezcA4OLFi0oL6tixo/qiIlVCi6+FTT3nYEaLLIzb+xCRydkAgKTMfAzbcRc7R7dQaKFOCCFEs8rsx21qagoDAwOl2wiFQrUFRKoWj8eDq6UJjkxug89/uY+nMUVn2pn5IozcfQfrPvXBoKZ1NRwlIYQQqTIr7o0bN5bZ3Yu6g1V/5oY62D++FaYceISbESkAALGEhzmHXyBemIUvOjTUSFxc6nZE3qG8cA/lhJvUmReVuoO5urrC2tq6zMKaNm0KV1fXCgfFNTWtO5gq1vRxxoSjrxHy1vi/JXysPxuDVwkpmNPBp8oHdqBuR9xEeeEeygk3qTMvap1kpKarad3ByhKZ/gqDf/kd6YlN5Jb3bKSPbUM6V+koa9TFhZsoL9xDOeGmyugOpvK0nrGxsTh79iyioqKgo6MDHx8f9O7dGyYmJhUKpDqpLRNdeFjUxYVJE/C/X3/C2zctZMsvPM3DsOwLOPh5zyodZU3Tk1mQklFeuIdywk3qyovKk4xIJBIsX74ca9euRUFBgdw6S0tL/PjjjxgyZEiFgiHcY2Nkg3P+szHkwGYEhzaBdJS1hy8l+GjbKRyf9BFM9Gn6QEIIqWplVv8LFy7E119/jcLCQri4uKBVq1Zo3rw5rK2tkZKSgmHDhuHkyZNVESupYia6Jvh79Fdo1+w5wCuULX8Zp41um08iUViz7/cTQggXKa24X716hQ0bNmDUqFF4/fo1oqKiEBgYiAcPHiAxMRH3799H69atMW3atFo9uH9Npquli32fzkP/DtEAP1e2PDFdD902n8bLZBqohRBCqpLSivvvv/9G+/bt8dtvv8HJyUlhfYsWLXDu3DlkZ2fj4cOHlRYk0Sw+j4/NfaZicp8c8LTf9dnPzNFHn+8v4VlMmgajI4SQ2kVpxR0ZGYn+/fsrLUAgEKBLly54+fKlWgMj3MLj8TC/w2gsG2wMvm6SbHlBgT6G77qHOy9TNBgdIYTUHmU2TlOltxhjrFZM3F6TB2BR1ehGH8PG6CamHwyBKNcBAJCVL8KYPYH4bmgT9PG1U+vxaFAJbqK8cA/lhJsqYwAWpf24t27dikOHDuHmzZulVsypqanw8PDAxYsX0bx58woHxiXFB2AJCwtDWFhYrer+pkxY2mtsvZKPe28yZct4AOZ2dcbARmUP2qMq6eAFAoGAurhwCOWFeygn3KTOvGRmZsLLy0t5xf3mzRt4eHigX79+WLVqFby9vWXrGGO4fv06Zs6ciaysLISGhtbYN0ttG4BFVQUiCeb++RQnn8bJLR/RRoCv+7ZRy1UYGlSCmygv3EM54aYqH4DF2dkZCxYswMqVK/HXX3/BysoKtra2EIvFiI6ORnZ2Nvh8Ps6ePVsr3ig0sIE8fV0+tgxrCisTPey9FSVbfvBOBmIyzmHvyI/UMsoaDSrBTZQX7qGccJO6B2Aps5Tly5fjm2++gYGBAZKTk/H8+XOEhIQgOzsb9vb2+Pvvv9GzZ88KBUOqLz6fh6V962Ne73pyy6+/APptP4Z8EXUTJIQQdVJ5rPLk5GRcunQJUVFR0NbWRv369dG1a1fo6+tXdowaJ71UnpGRQZfKSyFhEgw79D3uPa2L4r8HHa2zcXpKfwgM9D6sXIkEiYmJsLGxobMIDqG8cA/lhJvUmRdpXaTyWOVWVlYYNmxYhQ5Kai4+j4/fh0/HdJNdOHXbFmBFw6HGJBmh83fHcSbgY9gLjMsohRBCSFnU8rNs+vTp2LdvnzqKItUYn8fH1k8mYXzPXLlR1tKExui6+TSCE5I1GB0hhNQMSs+4c3NzkZ+fX2Yh6enpKm1HaoclXYajjslZrDqeDiYqurWQm2uMfluvYu/nLdHe3UXDERJCSPUld8Z95MgR2NnZITg4GAAwbdo0mJubl/mgs23yvvEt+uCHUS7Q0n03olphoQFG//wIx54+12BkhBBSvcmdcd+/fx8JCQkIDg6Gj48PAMDU1BQGBgZKCxEKhUrXk9qpb722sJoowOg9d1CQYw8AkEh0MetQBBIyszC5XWsNR0gIIdWPXMU9f/581K9fX2588o0bN8Lf319pIWWtJ7WXn2MDnJ5qgv/tOIOsjP8mqmHaWHsyCcZarzHSjy6bE0JIechdKrewsMDYsWOhpaUFAHB1dYW1ddnDV6q6HamdPC2c8c+0T2FlHSVbxsDD4uNB2HwpTKXx8AkhhBRRuR93SUQiEbS1Ve5RVm1RP271yMrPhv+hs7gbIn/rZaSfM1b08y1xlDXqm8pNlBfuoZxwU5X34z59+jQcHBzQpEmTEtdPmTIF169fx/r16/HJJ59UKKDqgGYHqxhDHQMcGPU/bL8eifXnw2TL9999g6TMPGwa0hgGOvJvSZrxiJsoL9xDOeGmypgdTGnFfezYMfj5+ZVacQ8aNAgFBQUYNmwYIiIiYG9vX+HAuKT47GAAkJSUhLy8PA1HVf0N8jGBnsQFay69hvi/6z3nnyei43e/Y//QDjAv1hhSOrMOY4zOIjiE8sI9lBNuUmdeMjOLZmNUeqnc398ffn5+ZTY+69u3LyZMmCDXqK0modnBKsfl4ARMPfQv8kXvfokaG6fh70m94GZpCYBmPOIqygv3UE64qcpnB1MFY6zWDMBCs+6oV48G9lg/LBUzDgaDSYrOsrOyzNFn6yX8PqEdmjk6A6AZj7iK8sI9lBNuUvfsYHJn3GvXrsXJkydlG0VERMDExAS2trYlFiKd3vPt27d4/vw56tevX6GguIoap1WuY8/vYs6hKEhEJrJlWtrZ2PZZA5jq2CAiJgkejtZo7W6llmlCScVRQyjuoZxwU6U3TgsPD8etW7fkNkxISEBERESpBWlpaeHLL7+ssZU2qXz/a+AHa39TfL73IQrzLQAAYpERJv/6CkDUf1u9gr1AH8s+qY/evjWrLQUhhJSH3Bl3ZGQkEhMTZSvXrFmDBg0aoF+/fiXurKurCzc3N5ibm1d+pBpEZ9yVL1GYh3vRr/Dl0TvIyy75Co/Uwj7eaOthBQCwMdGDjWnNn1qWa+jsjnsoJ9xU6Wfc7u7ucHd3l/3v5OSEevXqwc/Pr0IHI6QsBwLfYMvlSAAmAERQ1uFh9dkQ2d8zunliVg+vSo+PEEK4QmnjtK1bt1ZVHKSW+6y1M3rUt8XTmHQsPBZU5var/+eLRo5msDHRq4LoCCGEO8psVX779m28efMGANCqVSu5M/JDhw7Bx8en1H7ehKjKxlQfNqb6eJmUpdL20am5GNGaxjknhNQ+Si+4SyQSfPrppxg+fDhGjhyJBw8eyK2/fv06WrdujQsXLlRqkKTmSxTmIehtBrLzRSpt/9O1l/jfj7dwKyKpkiMjhBBuUXrGfePGDcTGxmLu3LlYtGgRBAKB3PotW7bAxMQEkydPRkREBDWIIB+s6B53eLn2efwmHZ/9fA/eLmmY2tUVH3u1A49H3cUIITWb0pr20aNH6NOnD9atW6dQaQNFrcrXrl0LPp+P+/fvV1qQpOb7rLUzTk1rj1PT2mNhH2+l275fNYe8NsfUX5LQZftG3Il5VHlBEkIIByituKOjo9G1a1elBfB4PLRo0QKvX79Wa2CkdrEx1YevgwC+DgJM7FQX20c2g71AvpuXvUAf20c2w8XZndDd570uY0wXUa99MOKnMHzyyzq8SA4BIYTUREovlYtEIpUuPYpEIk7OSCMUCrF161acPHkSeXl58PX1xVdffQVfX19Nh0bK0NvXHj3q2yEwMrnEkdN+HtMCNyLiMfPP60hJN5Ltx8QmeBbSAH0jb6Ntw+P4pudwuAioERshpOZQesbt5eWFo0ePQtmU3ampqfjnn3/g6emp9uAqqn///jh58iTWrFmDXbt2IS8vD23atEF4ePnupRLN0OLz4OduiZ7eFvBzt1QY7rSDhx3uz/sUi/s7wtBAftY2SYEtbj5siG5bTmDa2fVIyE6oytAJIaTSKK24+/fvj/v37+PTTz9FSIj8pUfGGK5du4Zu3brB0tISTZs2rdRAP0RhYSH27NmDzp07o0WLFvjtt9/AGMPhw4c1HRpREz6fB/82jfFo4QD4d7KEtnah3HpRjjtOXquPDt/vxtKrm5GWl6ahSAkhRD2UVtxOTk5YvHgx/vrrL/j4+MDa2hq+vr7w8fGBqakpOnfujKdPn2Lbtm2cbFF++PBh+Pj4yP43MDCAhYVFrZjJrLbR19HC4j5+uL/gI/RtagQeT/7WTUFGY/x23hUj9/2JjNzCUkohhBDuK7O2Xbp0KdasWQNDQ0MkJyfj+fPnCAkJQVZWFuzt7fH333+jZ8+eFQoiMzMT8+bNg6+vL+7evVvm9jdv3sSIESPQsmVL9OzZE9u2bYNYLFbYzsHBQe7/t2/fIj4+HsOHD69QvIS7zI10sXVoZ/wzuwtaeejIr2Q6eP6yDjqvv4K9t16hQMS9dhmEEFIWuUlGlElJScHFixcRFRUFbW1t1K9fH127doW+fsUmeNi/fz8WLFgAkUiE+Ph4XLx4Ed27dy91+59//hmTJk3CjBkzMGzYMISFhWHmzJlo3rw5Tp06BR0dnVL3HT16NFxdXbFy5cpyxUiTjGhORQfofxCVinnH7iIyQfFt7mppiK96e6ONpwGMdI2gwy/9vUPk0YQW3EM54aZKn2REGUtLSwwbNqxCB33f3r17sXXrVhw/fhw//fQTdu/erXT7yMhIBAQEYPDgwdi0aROAomFYTU1N0b9/f3z33XeYN29eifuuXr0aEokEK1asUOtzINzWwtUCl2f2wamnb7HufCiiU981YotKycEXBx7B0iwDpnUuYW77wejt1ht8Hn3pEUK4q1zfULGxsbhy5Yqsz7ZIpNrwlKXp2bMn7t27h+bNm6u0/bZt21BQUIBJkybJLf/kk09Qp04dbN68WaFbmkQiwZw5c5CQkIDffvuNRtaqhXg8Hj5p7IjLs7tgad/6MDOUP7NOSRfg1YtBmH7oCfodGYdr0deU9qQghBBNUqniPn36NJo1awYHBwd07doVFy9eBADMnDkTc+fO/eAvOQcHB2hpaam8/dmzZ8Hj8RSmGeXxeGjdujXi4uLw5MkT2fLc3FwMHz4crq6u2LJlC/h8PjIzMzFmzJgPipdUb7rafIxr74Zrc7tgUkd36GrJv/1FmY0Q9Oh/8D90DsNP+ON+PI0GSAjhnjIvlf/4448ICAgocV3Lli0REBAAQ0PDSr8EXVhYiLCwMFhZWcHQ0FBhvYtL0SAbL168kHVNGzx4MP755x88fPgQW7ZsAVB0Bl58hrOS5Ofny7U8FwqFsn25ONBMTSaRSMAYU+vrbqKnha9618NnrZ2w5txznH2WXGytNgpT2+NuYC5GvtyLDvV3Y2bzqahvWV9tx68JKiMvpGIoJ9ykzrxIy1BaccfFxWHOnDlo2rQpFixYgJYtW8o17BozZgwcHBwwePBgzJ8/HwYGBhUOrDQZGRkQi8UwMTEpcb10eUpKimzZTz/9hIKCAoVtS6r4i1uzZk2JP0SSkpKQl5dXwh6kskgkEmRkZIAxpvYGN7oAlnVzwRBfK2y69grPYot1E5QYID/xY1xOS8XNsDXo4inAWM8xcDGmUdiAys0L+TCUE25SZ14yMzMBlFFxHzt2DG5ubrhx4waMjIqGlXz/wN27d4enpyfu37+Pjh07VigoZaTdvUq7tC5dXvy+u7Oz8wcda8GCBZg9e7bsf6FQCCcnJ1hbW1Or8iomkUjA4/FgbW1daV9GNjZAR19XXA5JxKrTz/Em9V0FzgotkBc7AudTo3E9ajUGNPTFUr+ltb4FelXkhZQP5YSb1JkXaS8upRV3aGgohgwZIqu0S+Pl5YXY2NgKBVQW6Rl1bm5uietzcnIAQC0Vq56eHvT09CpcDqk+eDweuvvYopOXNQ7fj8bGS6HIyHk3NoAkzwnZryfiUk48PvfMh7t17a64CSGao7TilkgkMDMzK7OQhITKHwfa0NAQdnZ2SEhIgEQiUfjlEh8fDwCoW7eu2o65bds2ucFd6FJ51dPE5b+e7gZoN6Yh9j2Mx4GH8SgsNrZPQpIdem2+gf81ssb41vYwN6ydFThdluUeygk3Vfml8rp16+KPP/7AtGnTSr1EHRsbi3v37mHdunUVCkgVHTp0wJEjRxASEoL69eUbCz19+hR6enpo0aKF2o4XEBCAgIAAWad3ulRe9TR5+W+pkz0mdMnFpovhOProLaR9J8QM+PNJEs6FpGJyp7r4uKkhLkefxXDv4TDQrrx2HlxCl2W5h3LCTVV+qbx///6YO3cuevToga+//lqhG9b9+/cxZcoU2NnZoUmTJhUKSBWff/45jhw5gr/++kuu4n758iWePHmCESNGlNp4TR34fD59IDSAx+Np7LV3MDfCxiFNMK69G9acCcHNiHct0LPyxdhwIQw/Xi+AxPwG9gcfwKTGEzHYczB0tGr+mbgm80JKRjnhJnXlRbp/mUOeLlq0CKtXrwZQNEmHrq4uDAwMkJOTA6FQCD6fj3PnzqFHjx4VCsjf3x+7d+8uc8jTgQMH4vLly7h48SJatWoFoVCIAQMG4NmzZ3j48OEHN0hTRnrGnZaWRmfcVUwikSApKYkTZxGMMVwPT8aasyEIS8hSWM/Xi4We7Wk42+RiSuMp+MjtI2jxVR+noDrhUl5IEcoJN6kzL0KhEObm5mVX3IwxrFq1CmvWrFFoGGZvb4+dO3eib9++HxSEWCxG48aNARRdck9LS4OrqyuMjIzQs2dP2bCmxeXm5uKrr77Cnj17YGtri4SEBDRp0gQ7duxAgwYNPiiO0hS/xx0WFoawsLBKPaMniqT3hwQCAWe+jMQShtMvUrDzTiySsxVnGtMyCoGezVm4W+pjrOdYtLNpV+NG7ONiXmo7ygk3qTMvmZmZ8PLyUn2SkeTkZFy6dEntk4wEBQWVuFwgEMDJyanU/XJzc/H27VuYmZnBysqqQjGUhc64NYfLZxE5BSL8fOMVdtyIRG7B+4MrSKBjdh+6VpfQ0M4ZUxtPAz/fC4mZ+bAx0UNLVwto8atvZc7lvNRWlBNu0sgZtzIikQja2irPU1Jt0exgmlMdZjxKFObhu0vhOHz/DSTvf5p4+dA2DoE41xVMJJAttjLWxcQO7mjr8e5Hp42JHmxMK/ZDuKpUh7zUNpQTbqry2cFOnz4NBweHUhueTZkyBdevX8f69evxySefVCig6oCGPK161WEYRytjXXwzoAHGtHHGt+dCcTU06d1KpgdRZmMA8jV6clYBVp8NkVs2vasHZnb3rIKIK6465KW2oZxwU5UPeXrs2DH4+fmVWnEPGjQIBQUFGDZsGCIiImBvb1/hwLiE+nFrXnXqm2rGA77t44xLdY2x804s3qQVG0IVJV8WtzDUxoZ+HuDzebAy0kFiYmLVBFtB1SkvtQXlhJuqvB93WXr16oVevXohOTkZ9+7dQ//+/SsUFNdQP27Nq459U0fY2GBY23rYeDEMP12LVLptao4IYekMEWwXLHMF6GfaD94W3pxvzFYd81LTUU64qcr7cauCMYb09HS52bRqKuofqRnVsW8qnw9426v2I+/bc6EAvyW0jcLwq/E6eDrkY7BPH3zs/jGsDCq34WVFVMe81HSUE25Sdz9uuYp77dq1OHnypOz/iIgIXLt2Db/88kuJhYjFYkRHR+Pt27fYuXNnhQIipKaxMSlHQzOJPkSZjSDKbIRncRK8eP4Ga0xWoFVdA4xo1BVdnLtAT4vGzyeEvFdxh4eH49atW3IbJCQkICIiotQCtLS08OWXXyoMQVoTUeO0qlcdG9wkCvOQmJkPfW0eLI10kZKtOLWscnyIc10hznXF9UTgxoMEGAq+QQcvc/i36Iymto00fim9OualpqOccFNlNE6T6w4WGRkp1zhmzZo1aNCgAfr161diIbq6unBzc4O5uXmFA+IiGoBF86rjoBK77sRid2BcuffT5gOisj7b/FyYmkVjaIP6+LRBA5jqa6Y7ZnXMS01HOeGmKh+AZerUqWjZsiXGjBlToYNVdzQAi+ZUx0ElpGfcUrdfpmDXjVdyZ95Wxrrwb++GtnUtZcvMDXXwKiUHl4MTcPZ5DBKFymtxLT4PLVzM0c3bBl19bOBupXz6XXWqjnmp6Sgn3MS5AVhqCxqARXNqyqASYgnDvVepSMzMg42JPlq5KR85jTGG8MQsnAmKxvEnLxGVyENpXcqkHMy10cJdD8ObN0QLF3Noa1Xe61VT8lKTUE64qcoHYHlfbGwsQkND4e7uDhcXl1ozchohFaXF56FNsbPrsvB4PHjZmsDLtj5mdquP5Kx8HH0Sij8fh+J1giHyFYdIx9s0Ed4+FOHvh3ehrytGRy8rfNLQFZ3qWcNUv+bPVkZIbaFSrXv69GksWbIEjx8/BgDs2rUL/v7+mDlzJgwMDLBu3TqNN5YhpCazMtbDxHaNMLFdI+SLxLgbmYrLwQm4HJyIt+m5CtvnFWjhQlAaLgSlgc9jaOlmjl7166C7jy2cLQ018AwIIepSZsX9448/IiAgoMR1LVu2REBAAAwNDbFixQq1B8c11Kq86lFLWUU6fB46eFiig4cllvX1QUh8JnbcvYNTz15Dkqc4MY+E8RAYmY7AyHSsPPUCHjZG6OZti24+NmjqZPZBk51QXriHcsJNld6q/H1xcXFwd3eHj48PFixYgJYtW2LlypVo27Yt/P39AQCXLl3C4MGDERcXBwMDgwoHxiXUqlzzqKWs6hJzE3Hi1VWcDo1CUoodRNmeANNVuo+ZgTbaupqivbsZWruYwkhXtfnDKS/cQznhpspoVV7mWOVubm64ceMGjIyKWqy+f+Du3bvD09MT9+/fR8eOHSsUFNfQkKeaR8M4qs4GNvB18cWCTgzPkp/haPhJnAoKQ2a6K0RZPnKzk0ml54pwJjgVZ4JToavFQ2t3S3TztkE3bxs4mJf8Q1wsYQiMTMHLBAnqammjtbtltZ6itKagzwo3VfmQp6GhoRgyZIis0i6Nl5cXYmNjKxRQdUBDCWoGDeNYfk1sm6CJbRMsapOPK9FXcDz8b9yMjEJBZj2IsnwgyXNU2KdAzHAjPBk3wpOx/OQLeNuZoJuPDbr72KKxoxn4fB7OBcVhxckXiMuQTrYTBXuBPpZ9Uh+9fWvWJEPVEX1WuKlShzx9n0QigZmZWZmFJSQkVCgYQkjl0NPSQ2/X3ujt2htJOUk4HXkaf7/8G1piSwx0WIrLwQm4GZGM/BJGfgmJz0RIfCa2XXkJM0MdOFsY4mlMhsJ2cRl5mLz/ERb28ZbNL16d5hYnpLpRWnHXrVsXf/zxB6ZNmwYtrZLvfcXGxuLevXtYt25dpQRICFEPa0NrjPUdizENxkBYIIRAT4ARrZ2RWyDGzYhkXA5OwPGnEcjLVxwTPT2nEOk5ipV2ccXnF5/RzROzenip/TkQQgCl5+39+/fH/fv30aNHD9y6dUs2L7XU/fv30b9/f9jZ2ZU6ZzchhFt4PB4Eeu/udxvoaqFHfVss6OsCg7rfwNB1K3StLoOvV/7bX98M8MWpae3xWWtndYZMCCmmzJHTFi1ahNWrVwMADAwMoKurCwMDA+Tk5EAoFILP5+PcuXPo0aNHlQSsCTRymubQaFBVJ7swG0fDj+LEyxMISS06e5YUCiDK8kZBWmuw/DpllmGip4Vevvbo5m2DDl7WMNajAZqqCn1WuKkyRk4rs+JmjGHVqlVYs2YNcnPlB3qwt7fHzp070bdv3woFw3U0Vrnm0PjLmhGaGooTkSdw5tUZpOalQpTtjtw3E8tVho4WD63dLNDV2wZdvW3gbEEDv1Qm+qxwk0bHKk9OTsalS5cQFRUFbW1t1K9fH127dpU1T6+JqB+35lHfVM2Kz8zFjfhABKb8g1sPOoOJTFHWmOmlcbPQRzs3Adq7C+Brbwxt6kKmVvRZ4aYqnx2MFKEzbs2hswjN2nwpHN//E1H0Dz8PkEgbrimrdFkZ6wGBgQ46eVmhm7cNOnpZQ2BAY6lXFH1WuKkyzrjpBlQ5UP9IzaC+qZoz0s8FPRvYyf6/HZGMnTcikZwlP0XpxA7usq5gDxICsfb2zxBleUOU5V3iwC8ZuYU48SQOJ57EvZue1McG3Xxs4W5lRHMffCD6rHBTlfXjzsvLw8mTJ3H58mVER0cjMzMTlpaWqFevHnr37o3OnTtXKABCCPfZmOrL9cf2dRBgfAd3BEYmIyImCR6O1mjtbiU3ctql+BfQNgmGtkkwGAMk+XUgyvSBKMu7xLHUxRKGwFepCHyVitVnQuBqaYiu/42l3tLVArraVAkRUlyJl8r//PNPzJw5E2/fvi11x9atW+Pnn3+Gr69vpQbIBdSqXHOopSw3KctLgbgAT5KeIDAuEHfj7iIoOQhiVtSVVCIyhjjLu6giz/YAmGKf8eJM9LTR0csaXb1t0LmeNSyNlW9fm9FnhZuqpFX51q1bMX36dEgX161bF05OTjAyMkJGRgbCwsKQmJgIADA1NcW5c+fQpk2bCgXDdVRxaw59GXFTefKSVZCFhwkPcTfuLgLjAxGeFg4AYBJtiHPcIMrygZm4A5KEYqXl8HhAM2dzdPW2QTcfG9SzNaFL6sXQZ4WbKr3iDgwMRPv27aGjo4OvvvoKkyZNgp2dncLOz549w8aNG/Hrr7/C0dERT58+hbm5eYUC4jKquDWHvoy4qSJ5Sc5Nxr24ewiMD0RgXCDS89Nxfeh1vErKx+WQBPwTnIhHb9IgKaPZrIOZgawS93O3hL6OajOb1VT0WeGmSq+4e/TogcePH+PKlSto2LBhmYX8/vvv+Oyzz7B06VIsW7asQgFxGVXcmkNfRtykzryk5KbA0sBSbllqdgG+PLsX10NTIMquB0iUdzs10OGjvac1uv3XZ7w2jpNOnxVuqoyKW9Y47c2bN7h06RJOnDihUqUNAMOGDcPjx4/x888/1+iKW0oikdAk9VVMnZPQE/VRZ17M9cwVyjEz0IaHYwqC8o8jpzAf4pyiqUlFmT5ghVYKZeQWSnDxRQIuviia8KihgwBdvYvujfvWMa0Vl9Tps8JN6syLtAzZGfeBAwfw7bff4tmzZ+UqKDU1Fba2tggLC4Obm1uFA+MSGoBF82hQCW6qqrwUSgoRmhGKxymP8Tj1MYLTg1GQZ/ZfVzMfiHNcASi/RG5lpIN2bqZo52aGls4mMKihl9Tps8JNlToAy5IlS5CdnY1NmzaVuzA/Pz8sW7YMffr0qVBQXEUDsGgODSrBTZrKS05hDh4lPpLdHw9Jfg1RthdEmd4QZ3uDiZUPq6qtxdDSVYA+DRzRxdsGDmYGVRR55aPPCjdV6gAsaWlp8PT0/KDCXFxckJqaWqGAqgMa2EAzaFAJbtJEXoz1jNHRqSM6OnUEAKTlpeFe/D3cjbsLHd5L9LCbiMvBifgnJAFhCVkK+4vEPNx5KcSdly+AEy/gaMlDnwaO6OPrhMaOZnL90asj+qxwU6UNwJKXlwcjI6MPKszExERhAhJCCKls5vrm6OXaC71ce8mWtXS1wPw+3ohOzcHl4ARsuH4eWUJ7gCmONxWTwrDrejR2XY+GgZ4ILd0NMbCxF7p514GJPg3DSriJhjwlhNRIThaGGN3WBdrmxrgVfQV3ItOQle4OcVY9MLFiW5XcfG1cDy7A9eAgaPOfw8/dUtbdzMXyw05qCKkMchX3kiVLsGHDhnIXEh8fDz8/P7UFRQgh6sDn8TGy/kiMrD8SIokIQclBuBsbiEvh9xH0hocCoRck+Q4K+4kkDDcjknEzIhkrT72Ah40xmrvpoXVdQ3xSvwF0tGtmAzdSPchV3PHx8YiPj9dULIQQUmm0+dpoYtMETWyaYHITIFeUi8cJj3Ex8gGuhCQiJsHsv2FYFS+RRyRmISIxC4cDU/ClVhgcbTLRqZ4FPmveFN7WzlX/ZEitJldxT5w4EV26dCl3ITt37lRbQIQQUhUMtA3Q1qEt2jq0xbIOQEZ+Bm5EByI2yRhR8Sb4JzgR8cI8hf0kYgO8iTPAvjhg39V/YWhyDg2cGPo2dMEn3q1gYWChgWdDahO5irtly5YYNmxYuQu5dOmS2gIihBBNEOgJ0NejJ+BR9D8bwPAiTojLwYk4+zwGwbE5JeylhZxMB9x/Adx/IcZynZOwsUqEX10j/K9RQ3R0agc+j1p4E/WSVdzjxo2DtbX1BxVSkX0JIYSLeDweGtQRoEEdAaZ19cDDt+E4+OgJbkdkIiHZHEyiq7APK7RCQpwV/o4DTtxOQu/6j9HNxxad61nDimY2I2pS4rSeRB6NVa45NP4yN9X2vGQXFODPJ//iVFAUnr0G8vKUtzrn8YAmTmbo7mOLHN17sBYUok0dP3iae6rtjLy254SrKnWsckIIIaox0tXFmJatMKZlKzDGEBSbgoOPnuFaaApik/UAyH9BMwY8fpOOx2/SAQjA006Dtsk+mJvHoL2HLdo6tkRr+9ZwMnHSxNMh1QydcauAhjzVHBrGkZsoL6VLyynAuedvcCM8Hbci0pCZJ1K+A68A2kYR0DIOhqNtBto7+6KVXSu0smulMGuaMpQTbqqMIU+p4laCJhnRPJo4gZsoL6oRiRmexGbh5qt03HqVgTdp+WXuw9ePgbZxCLSNg+FhpYeudbpgmHvZjYYpJ9xUqZOMkNLRGbfm0FkEN1FePkxkcjauhCTidNAbPI3OhoQpHxudpy2ER508fNmxN9p5WMJQ993dTQmTyO6PiyUMgZEpeBmbhLp1rNHa3bLaj7teU9AZt4ZQ4zTNoQY33ER5qbiM3ELcCE/CpeAE/BMSD2Gu8vmadbX5aFvXEt28bdC5njUmXBkEV4ErTAs64p9HFkjOEsu2tRfoY9kn9dHb176ynwYpQ2U0TqOKWwVUcWsOVRDcRHlRL7GE4fGbNFwOScTl4JJnNnsfXy8OfJ1kiLJ8/1sif4bNA/DTyGZUeWsYVdwaQhW35lAFwU2Ul8oVnZqDf0IScTkkEXdfpqBArPxsvGQMWjqZaNPqCtxMPdDVuRO6uLVWe6xEuSrvDrZkyRI0btwYgwcPrtDBCCGEqM7JwhBj2rpiTFtXZOeLcCM8Gf+EJOCfkCQkZ5XdwK0ID+JCUwS+sMRD05uISBCXWHG/SHkBV1NXGOoYqvdJkEqjtOKOi4uDi4tLVcVCCCHkPUZ62ujta4fevnaQSBiuhyXh/PN4XA5JRGJm2ZV4YVp7FKa1x53XQKvwS6hnZ4J6tibwsjOBs6U2/K+OBJ8vgoupC7wtvFHPoh7qmdeDt4U3rAyswONRIzeuKXMAlqlTp2LmzJlKt+HxeDA0NISrqyv69u2LadOmwczMTE0hEkIIAQA+n4f6dUxhZaKHho4CLDwWVK79EzPzkZiZjxvhycWWrgBPJxXBegkI1UsAX/8K+HqHwNdNhqWBmawSl1borgJXaPNp7C5NKvPVz8/PR35+2b/qsrKykJiYiHv37mH37t24evUqXF1d1REjIYSQ/xwIfIMtl8PVWCIfrNAKokIrIKtBseUi5Ogl4a1eAq7qvYKW3l3w9eLxsVcbrOu0Vo3HJ+WltOLeuXMn7OzscOXKFSxatAiNGzeGra0tCgoKEBMTg0uXLmHXrl3Ytm0bGjdujNjYWFy4cAGrVq3C6NGjcf369ap6HoQQUit81toZPerbAgBuRyRj9dmQUrdd0Mcb9exM8DolBylZ+XibnofQBCHCE7KQLyqrwZs2JPn2kOTLt0o/9lqC0KBb8P7vcnvRZXdjfHljEiwNLOBl4QVvc294W3jDzsiOLrVXAqWtyp88eYJRo0bh3r170NfXL3Gb6OhofPTRR3j8+DG0tYt+B4SFhaFRo0Z48uQJ6tWrVzmRVyFqVa451HqZmygv3HEuKA4rTr5AXMa7ucPL6sctljC8Sc1BaHwmQuMzEZaQidCETLxKzoZY8mEdjXhaWeDrJfz3iIeWXjwEJrnwsXaVXW73tvCGu8AdOlo6H3SM6qjKW5WfPHkSAwcOLLXSBgAnJyd4enri9u3b6NixIwDAy8sLXbp0wdOnT2tExU0IIVzV29cePerbITAyGRExSfBwtEZrdyulI6dp8XlwszKCm5URevvayZbnFYoRmZQtq8ilFfvb9Nwy42BiY4hzjCHOqStblgMgXjsd1/TioaV/E3y9v6CjnwxPGxN0c+2IgCYBFXrutZXSijs2NlalildXVxdv3ryRW2ZlZYXCwsKKRUcIIaRMWnwe/Nwt4W4sho2NJfgfONypvo4W6tcxRf068lcWM/MKEZaQVVSh/1eZhyZkIjW7oMwymcgMYpEZxNneAIA8AI8iJYh4loegoIdyrdxdLAxx+tVJGOoYwtvcGw4mDmqb9rQmUVpxCwQC/P3335gxY0ap2wiFQly5cgUjRoyQW/7mzRs4OjqqJ0o1SE1NhYGBAQwMDDQdCiGEVCsm+jpo7mKO5i7mcsuTs/IRFp+JkGKX20PjM5FTIC6lJCk+hNmGOBsUj7NB8bKlutp8QCcFEt2n4Osdh6FhOrxsjdHIzhnelkWX2z3MPKCvXfpV4NpAacXds2dPfPvtt/jss8+wfPlyeHp6ytYxxnD79m3Mnj0bmZmZ6NChg2zd3bt3cfv2bXh5eVVe5OWwZ88eTJkyBb/88guGDSt7lh1CCCFlszLWg5WHHtp6WMmWSSQMb9Nz37vcLkREYhbKag9XIJIAIlsgt6jxXQGAe1HAPX7ef/fNn0BLPxEO5jz4OligsZ0HGlg2QGv72jUinNKKu0uXLhg4cCAOHjyIgwcPwtraGjY2NpBIJIiJiUFmZiYAYOPGjTA3L/oltmHDBixevBhdunSBnZ2dsuIrnVgsxowZM5CYmKhSlzZCCCEVw+fz4GRhCCcLQ3TzsZUtLxRLEJWcjdCETITFv6vUX6fmoMyBtyX6kOS6QpLrikIAL+OBl8HACS0hTE2eYVADE3jbFV1u97QxhpGeNtLy0mCqawotvlalPl9NKLMf94EDBzBnzhzs3LkTSUlJSEpKkq0zNTXFN998g6lTp8qW+fn5YevWrWjVqlXlRFwOWVlZ+Oijj9C1a1ccOXJE0+EQQkitpaPFh6etCTxtTYBG75bnFogRkZj1X0UuRHB8BoLjMpCSVdbldoCJTZGRboo9t17JLXeyMEAu/yVyeBFwsOCjQR0ztHR0RQPrevAy96r2w7uqPMlIXFwcLl++jOjoaGhpacHLywvdunWDiYmJWgJhjCElJQUCgQA6Oqp1FcjOzoaBgUGZTezz8vJgYGCAQ4cOfdClcuoOpjnU7YibKC/cU9Nykp5TgLCEdxX607fJiEjIQU7Z7eFKIQZfLwl8vQRYmObB3VofTR1t0drZA94W9WBraFspfc6rvDvY2rVr0aBBA/Tt2xf29vYYOXJkhQ5amvv372Pq1Km4d+8eLl68iO7du5e6rVgsxrp16/DDDz8gIyMDEokEffv2xXfffcepxnCEEEI+nJmhLlq5WaCVm4VsGWMMiZn5RY3h4jPxLDYFEYk5iEzKQV5hWQPKaEGSbwdJvh0ShUBiDHD3MfATLxV8veMwMc7E1FYD4W1vinq2JrA20ePs4DFKK+7w8HCEhYWhXbt2snvY6pScnIz58+fj4sWLsLS0VGmfyZMnY//+/Th8+DD69euH6Oho9O/fH+3bt8e9e/dgY2Oj9jgJIYRoHo/Hg62pPmxN9dHJyxqAO4CiAWWiU3Nk982fx6Xi35hEJGbwwFgZlS/ThSTPCRl5wDdn3o1CZ2aog3q2JhDrRIOvl4BGdazQ3s0dzer4wFRXtSuvYglDYGQKImJS4ZGlVWb/elWVeY/7999/x6FDhzBixAgEBASgadOmFT6o1LFjx2BoaIjnz59j5syZePz4sdLtb9y4gZ9//hnz5s1Dv379ABQNAPPLL7+gcePGWLp0KbZv3662+AghhHCfFp8HVysjuFoZoVeDd42i80VFA8qExgvxKDoeT94mISq5ABnZZU+Skp5TiMBXqQCMALjj3jPgZySDp30ShkYZsDOXoJ6tCVo410Ent3pwN3eSO0NXHNHuVZkj2qmqzOg3b94MZ2dnbN26FS1atICfnx8CAgIwePBg6OrqVujgn3/+uWyYVFXs2rULABTuUzdq1Ag+Pj7Yv38/vvvuO+qrTQghBHraWvCxN4WPvSkGNH13KzUrX4SwhEw8fZuE+2/eIiQ+A8kZesjIKetye9GAMtkZZniZAbyMAs4EFgB4Ai3dqzA3zYWLlQ4aWbbE3ltvFPaNz8jDF/sf4aeRzSpUeSutNfv27QsnJyc0b94cvXr1QlRUFH766SdMnz4ds2fPxsSJEzFp0iQ4ODh82MHLUWkDwJUrV6ClpYWGDRsqrGvWrBmCg4Px4MEDuT7liYmJsm5rqampiImJgb29PbS0Su8i8P6MaEKhEEBRIwOJpOzEEvWRSCRgjNHrzjGUF+6hnKjOUIePJo4CNHEUYHRrD9ny5Kx8hEtHiEvIwou4dDyPS4VYXFaDaT7EBZZITgaSk4GHUKy0AUDaEnzx8SDYmuhBi8+DjYkebExVG1BGmluVW5UXl5eXh99//x3btm3Dv//+iwEDBmD16tVyA7SUl7+/P3bv3l1q4zRpy/A6derg7du3CuvnzZuH9evXY/fu3Rg3bpxsea9evfD8+XO5be/evau0Idvy5cuxYsUKheVhYWFqa0VPVCORSJCRkQGBQFAjWsrWFJQX7qGcVA7GGOKE+XgSn4hHcfEIT85GbDqQmW0ExireR3x8a3tMaFNHpW0zMzPh5eVV9qXykmhra8PQ0BAGBgYQiUT4888/0atXrwpV3GWRnvUaGpbc/87IyAgAkJGRIbf8/Pnz5T7WggULMHv2bLljOzk5wdramrqDVTGJRAIejwdra2v6MuIQygv3UE4qj60t0MTTGWOKLROJJQhJSMWNVy/xKCYOofHZSMrQRn6eCYCyG6DN6eGJTl7W5Trjlk74pbTi3r9/P9zc3NCuXTsAQHx8PHbs2IEdO3YgLi4OQNFkIuPHj0ffvn1VOvCHkt70L+0CgfQSgjresHp6etDT01NYzufz6QOhATwej157DqK8cA/lpOro8vlo5GiDRo5FPZkShXlIzMzHw9epWHbiRZn7WxrryXKlar6k2ymtuK9evYq8vKIWcVu3bsVff/0lm/GrZcuWCAgIwNChQ5VO+6ku0ss/WVlZJa6XLq+MbmtSdI+76tF9O26ivHAP5USz9t99je//iVB5+4XHggAA07t6YGZ31a5WS3Nb5qXyuXPnIj09HUDRmeioUaMQEBCA1q2rdlB3XV1duLu7IzIyEvn5+QpnxNHR0QAAb29vtR1z27Zt2LZtG8TioqH3kpKSZD9kSNWQ3rdjjNFZBIdQXriHcqJZPesaopmdDwDgQbQQW28otsWSmtrBAS2cim67WhnpIDExUaVjSBtal1lxp6enw9nZGZMnT4a/vz+sra1VOkBl6NGjB3766Sc8fPgQbdu2lVt3//59WFpaolmzZmo7XkBAAAICAmTDzNE97qpH9+24ifLCPZQTzSo+9FdHX6C+kw1WngpGvPDdyZ69QB9LPvZBb98Pm4BLpXvcAPDVV1/hm2++Udp96sqVK7C1tUX9+vU/KBhVTZkyBTt27MDevXvlKu6rV68iKioKixYtKncXs/Kge0eaQfftuInywj2UE+74qFEd9PK1R2BkMiJikuDhaF3hkdNUusfdqlUreHp6Kq20gaIZxPz8/D6o4k5OTgYAWb9poVCI5ORk6OnpKXS98vX1xddff40lS5agXr16GDp0KMLCwjB+/Hi0aNECCxcuLPfxy4PucVc9um/HTZQX7qGccA8PQCtXc7gZiWBtbQ4eGCSScvfAlqlQP26pqKgo/PLLL9i+fTtWrVoFf3//cu0vFotha2tb4rp+/fphz549Ja47evQofvjhB4SHh8PMzAwDBgzAwoULS+0q9qGK3+MOCwujftwaQH1TuYnywj2UE25SZ16k/bjLXXHn5ubi6NGj2LNnD65cuSLrnrVr165yV9zVhfQed1paGt3jrmISiQRJSUl0345jKC/cQznhJnXmRSgUwtzcXPUBWO7fv489e/bg0KFDcoOcWFpaqv1Ml6vo3pFm0H07bqK8cA/lhJvUlReV7nEnJSVh//792LNnD4KCgmTLLSwsMGDAAAwZMgTdunXD5MmTKxRMdUH3uKse3bfjJsoL91BOuEmdeSm1H7dYLMbZs2exZ88enDp1SjbgCo/HQ6dOnZCdnY2JEyfKXRb/7LPPSr1XXZ1RP27No76p3ER54R7KCTepMy/Sftyye9wvX77Erl278Ntvv8mGMwUAGxsbjB07Fv7+/vD09IS/vz/8/Pxq7P3sktA9bs2h+3bcRHnhHsoJN1XqPe41a9Zg9+7dAIrOrrt3746JEyeif//+0NEpa0qz2oHuHWkG3bfjJsoL91BOuKnS7nG7uLiAz+dDIpFgxowZWLJkCSwsLCoWLSGEEELUSlZxL1myBGPGjMEvv/wi65s9cOBATJgwAZ07d9ZgiNxBjdOqHjW44SbKC/dQTrip0hunOTs7Y+nSpViyZAmuXr2KPXv24KOPPoKjoyP8/f0xduzYCh+4OqHGaZpHDW64ifLCPZQTbqrUxmmlEQqF+P3337Fnzx48evQIAoEAU6dOxZIlS2RB/PDDD6hXrx569uxZoaC4ihqnaQ41uOEmygv3UE64SSMDsJiammLixImYOHEigoODsWfPHvz000/YsWMHBg8ejCFDhuDff/+FgYFBhQKqDqjRh2ZQgxtuorxwD+WEm9TdOK1cpfj4+GD9+vWIiYnB9u3b8ebNG3Tt2hV79+6tUDCEEEIIUc0HVf/a2tro168fjh8/jpiYGHTp0kXdcRFCCCGkBBWevNrGxgadOnWCpaWlOuLhNGpVXvWopSw3UV64h3LCTVUy5OmHWLp0qTqK4RxqVa551FKWmygv3EM54SaNtCon1Kpck6ilLDdRXriHcsJNGp3Wk1Crck2hlrLcRHnhHsoJN2m0VTkhhBBCNIsqbkIIIaQaoYqbEEIIqUboHrcKpO330tPTqatFFZNIJBAKhdDV1aX7dhxCeeEeygk3qTMvQqEQAFXcSkm7gxUUFAAomvqUEEII0STqDqYCiUSC2NhYmJiYgMfjaTqcWkUoFMLJyQnR0dHUFY9DKC/cQznhJnXmhTGGzMxMOuNWBZ/Ph6Ojo6bDqNVMTU3py4iDKC/cQznhJnXlRSAQUOM0QgghpDqhipsQQgipRqjiJpymp6eHZcuWQU9PT9OhkGIoL9xDOeGmysgLNU4jhBBCqhE64yaEEEKqEaq4CSGEkGqEKm5CCCGkGqGKm2hUXl5ehcvIyclRQyREqrCwEGKxWC1l5eXlIT09HYWFhWoprzZTx2dFKj8/X21l1WYSiUQ2smZFMMbK9RmhiptUubdv32LYsGEwMjKCQCCAi4sLNmzYoPI48NnZ2fj111/Rs2dPGBsbw8LCAiYmJmjRogV+/fVXtVU6tc3x48fRpEkTGBsbw8DAAJ06dcK9e/c+uLz8/Hw0b94c5ubmOHz4sBojrT1yc3Mxb948WFtbw8zMDObm5ggICJCNWV0et27dwkcffQRjY2MIBAI4ODjgiy++QERERCVEXrM9e/YMPXv2hIGBAUxMTNCgQQMcOHCgXGWIxWLs2rULDRs2hKGhIUxMTODs7Iy5c+ciNTVV+c6MkCqUkJDAnJ2dWbNmzdjr16+ZRCJhhw8fZjo6Omzq1KkqlXHo0CEGgM2aNYslJibKyh0zZgwDwMaNG1eZT6FG2rdvHwPAli5dyvLz81l6ejobPnw409fXZ3fu3PmgMufNm8d4PB4DwPbt26fmiGs+sVjMevbsySwtLdnNmzcZY4w9evSIOTo6statW7O8vDyVy9q7dy/T0dFhmzdvZllZWUwikbDr168zc3NzNmjQoMp6CjXS06dPmYmJCevXrx9LSUlhhYWFbP369QwA27p1q8rlTJ06VfY9lpmZyQoLC9lff/3F9PT0WIMGDVhubm6p+1LFTaqUv78/A8CePXsmt3zGjBkMALt7926ZZfz111+sefPmCssLCwuZi4sLA8AiIiLUFnNNl5aWxszNzZmfn5/c8uzsbGZtbc0aNWrEJBJJucq8e/cuMzExYaNHj6aK+wP98ssvDADbvn273PIjR44wAGzjxo0qlRMWFsb09PTYihUrFNatXr2ajR49Wi3x1hbt27dnxsbGLC0tTW55r169mKGhIYuLiyuzjIyMDKatrc2cnJyYWCyWWzdx4kQGgP3xxx+l7k+XykmVyc7OxoEDB9CwYUP4+vrKrRs+fDgAYNeuXWWWM3DgQDx48EBhuba2Nho1agQAeP36tRoirh2OHDmCtLQ0WQ6kDA0N0b9/fzx9+hSBgYEql5eXl4exY8fi66+/hqenp7rDrTV27twJHo+HoUOHyi3v378/DA0NsXPnTpXK2bBhAwoLCzFlyhSFdQsWLMCvv/6qlnhrg+DgYNy8eRN9+vSBmZmZ3Lrhw4cjJycH+/fvL7Oc5ORkiEQiuLm5KUz1WbduXQBAXFxcqftTxU2qTGBgIHJzc9GsWTOFdU2bNgWPx8PVq1crdIyEhATo6uoq/DAgpbty5QoAlJgX6bLy5GXx4sWwsbHB9OnT1RJfbZSXl4fAwEC4u7srVBA6Ojrw9fVFaGgo4uPjyyzr+PHj8PT0hJWVFQCgoKAAIpGoMsKu8dT1WXF2doaVlRVCQ0MVGgo+ffoUANC8efNS96eKm1SZ8PBwAICDg4PCOl1dXVhaWuLVq1cf/KUSHh6O+/fv44svvoCNjU2FYq1NlOWlTp06AICwsDCVyrp9+zZ27dqFvXv30hS4FfDy5UuIxeIScwKonpe4uDgkJibCyckJR48eRaNGjWBsbAxdXV00atQI+/btU3vsNZm6Piva2trYu3cvcnNzMXr0aDx//hyvX7/Gpk2b8Pvvv2Pu3Llo165dqftTxU2qTEZGBoCiS7AlMTY2hkQiQWZmZrnLFolEGDduHFq0aIG1a9dWKM7aRllejI2N5bZRJjc3F59//jlWr14Nd3d39QZZy6jyWSm+XWmkZ+QPHz7E9OnT8f333yM3NxexsbHw9vbG6NGjsW7dOjVGXrOp67MCAH379sUff/yBJ0+eoGHDhvDw8MDy5cuxevVqrF69Wum+VHGTKiM9A2OlDI8v7Q5W3jM1kUiEESNGIDs7G+fPn6dJFspJWV7Kk5MFCxbA0dGxxHuppHzU9VmRXoZNS0vD2rVr0blzZ2hpacHOzg6//vorrK2tsXz5cqSlpakx+ppLXZ8VAJgxYwb69OmDCRMmICMjAzk5OThy5Ai+++47tG/fXmlOqOImVcbc3BwAkJWVVeL6rKwsaGtrw8TEpFzljh8/Hi9fvsSVK1dkxyCqU5YX6bKyXtcbN27g559/xsaNG5GRkYH09HSkp6fLBg3JycmRLSNlU+WzUny70kjPAgGgW7ducusMDAzQoUMH5Obm4tatWxUJt9ZQx2cFAM6cOYPvv/8eQ4cOxZw5c2BiYgIdHR306tULGzZsQGBgIJYsWVLq/tofGD8h5ebt7Q0AiI6OVliXk5OD1NRUeHt7Q0tLS+UyFy1ahPv37+P69esQCAQAii7Z8vl8OvNWkbe3NwIDAxEdHQ0PDw+5dTExMbJtlLlw4QK0tbXRuXNnueXSM75Zs2Zh3rx5AICUlJRy5bg2qlu3LnR0dEr8rACq58XV1RU8Hg+MsRIrFGlbkDIH/CAAlH+HqZoTAPjnn38AAJ06dVJYJ112+fLlUvenM25SZVq1agWBQID79+8rrJMu69Gjh8rl7dixA4cOHcLFixdlLWaBokpi/fr1FQ+4lpC+5iV1sZOOnFZWXr7++mvZGXXxx6JFiwAU5Uq6jCrtsuno6KBTp0548+YNEhMT5dbl5uYiKCgITZs2haWlpdJyjI2NZa2d3759q7Be2uVI2rCKKNe9e3cAFfusAO+Gry2pIa506FNlw9JSxU2qjK6uLiZMmIDw8HDcuXNHbt2vv/4KLS0tTJ48WW659BLr+8OYnjx5EkuXLsWRI0dgZGQkV1moY+zg2uR///sf6tSpg3379skNO5uWloaTJ0+iQ4cOsv7xUhkZGR/UiJCoburUqQCg0M/60KFDKCgokK2XKigoQHp6usIXvvQz9ccff8gtT0tLwz///ANbW1u0b99e3eHXSK6urujbty8uXLig0M/6t99+g4WFBUaMGCG3PCsrC+np6XL3xaWv97lz5xSOIV2mrFU5jZxGqpRQKGSNGzdmdevWZdevX2exsbFs/fr1jM/ns9WrVyts7+rqygCwb775Rrbs4cOHzNDQkAEo9fH1119X5dOq9s6fP8+0tbXZuHHj2KtXr9izZ89Yt27dmLm5OQsJCZHb9tKlSwwA4/F4LD4+vsTysrKyWFpaGlu0aBEDwHbs2MHS0tJYTk5OVTydGmP06NHMyMiIHThwgMXFxbFjx44xc3Nz9sknnyiMuDVy5EgGgPXp00duuVgsZn379mUGBgZs+/btLCoqit25c4e1b9+e6enpsdOnT1flU6r2Xr9+zezs7Fjbtm3Zv//+y16/fs2mTZvG+Hw+O3LkiNy2WVlZTFtbmwFgf/31l2y5SCRiXbt2ZQDY9OnT2fPnz1lUVBTbuXMnMzU1ZTY2Nuzly5elxkAVN6lyQqGQzZs3j9WrV4/Z2tqyDh06KLzhpVq0aMEEAoHcsI8//vgjEwgESh/r16+vqqdTY9y7d4/169ePOTg4MBcXFzZq1Cj26tUrhe1u377NzMzMmI2NjcKwj1KDBg0qMS+zZs2q3CdRw4jFYrZ161bWvHlzZmNjwxo1asTWrl3LCgsLFbadOnUqEwgEbOzYsQrrCgsL2XfffcdatGjBrK2tmbu7Oxs1ahR7+vRpVTyNGuft27fM39+fubm5sTp16rDevXuza9euKWyXm5vLbG1tmUAgYJcvX5ZbV1BQwL7//nvWsWNHZm9vLxteeNasWSw2Nlbp8XmMldLfgBBCCCGcQ/e4CSGEkGqEKm5CCCGkGqGKmxBCCKlGqOImhBBCqhGquAkhhJBqhCpuQgghpBqhipsQQgipRqjiJoQQQqoRqrgJIYSQaoSm9SSc9eLFC4WJEXr06KF88P1i/vnnH1y/fl32P5/Px9KlS9UaY03322+/ITIyssR1PB4PRkZG8PLyQrdu3WBkZFTF0RFSO1HFTTgvMTERP/30EwDgr7/+wtOnT8Hj8ZTuIxKJ4O/vj1evXsHS0lJhJiWgaIarPXv2oH379mjZsmWF4/zzzz8RFBSEIUOGoH79+hUuj0sOHjyI8PBwdOnSBR07dgRQNHPbvXv3MG/ePAgEAmzZsgWjR4/WcKTVQ1hYGA4ePAhvb28MGzZM0+GQ6kaN464TUimePXsmN/PXn3/+WeY+e/bskW1fr169Erf59ttvGQDm4+OjljiHDh3KAJQ6YUp19vHHH5c669q5c+cYj8djPB6PHTt2rOqDq4ZOnjzJALD+/ftrOhRSDdE9blJt9O3bFwCwatUqublt3ycWi7F69Wp88sknSsubNGkSNm7ciL1796o1ztqmV69eGDhwIBhjWL58uabDIaTGo4qbVBtz5syBvr4+/v33X5w8ebLU7Q4ePAixWKwwof37zMzMMHv2bLRu3VrdodY6zZo1AwA8ffoUhYWFGo6GqItYLMbVq1exc+dOrFmzBj/++CNOnz6N1NRUTYdWq9E9blJt2NnZYfz48di2bRu+/vpr9OvXT2EbiUSC1atX46uvvoK2dulv782bNyM9PR2AfKM16b3H9487efLkEtd9+eWXCAsLw4kTJxAUFAQA+OOPP2R/A8DChQtx+/ZtXL16VbZs9OjRcHd3BwCEhITg999/l63r16+frCIEIHcWa2FhgenTpyM5ORlnzpxBdHQ0CgsL4eHhgZEjR8rFFhgYiPv370MoFMLW1hZdunSRHbOqJCQk4Pbt23j16hXEYjHc3NzQo0cPCAQChW1XrVoFkUgEADA1NcXs2bORmpqK06dPIzo6GgUFBXB1dcXYsWNl+7x+/RqBgYGIioqClpYWvLy80L17dxgYGMiVnZWVhQ0bNsj+d3Z2xrhx4xAaGooLFy6goKAAjRs3Rrdu3WTtJx4/fozr16+Dz+ejefPmaNu2rdLnGhMTg8uXLyM2NhZGRkZo2rQp2rVrBz7/3fmRSCTCqlWrEBYWBqAo98Xz27dvX7Ro0ULhNbx06RKio6Ohr6+PRo0aoWPHjgrv7507dyI2Nlb2//z588EYw5kzZxAREYHc3FwYGhpi3rx5Sp+H1J49e7BgwQIkJiYqrOPxeIiKioKzs7NKZRH1ojNuUq189dVX0NHRwYMHD3DmzBmF9YcPH0ZmZiY+//xzlcpbuXIlVq5cqbD84MGDWLFiBf7++2+55WKxGOvWrcOjR48+KP79+/djxYoVJbbUPnXqFFasWFFi2UKhECtWrMD333+Pv//+Gx9//DGePXuGoKAgrFy5Evv375dtGxkZCT8/P3To0AF3795FZmYmDh06BA8PD4wfPx75+fkfFLsyjx8/BgA0atQIOjo6EIlE6NevH5ycnPDTTz8hISEBERERmDVrFhwcHLBr164Sy8nJycGKFSuwadMmnD17Fr1798bTp0/x/PlzrFy5Er/88gsAICkpCV27doWnpycOHTqE1NRUvHjxAmPHjoWTk1OpV2RiY2OxYsUK7NmzB2vWrMG0adPw+vVrnDx5Ej169ECnTp2QlZWFyZMnY+nSpXjz5g1+++03tGvXDuPHjy+xzMLCQsyYMQNubm7YuXMnhEIhHj9+jN69e6NJkyYICQn5oNdUIpFg6dKlcHFxwaZNm5Ceno4XL15g4MCBqF+/Ph4+fFjifhs2bMCKFStw//59dO7cGVevXsXr16+xdu1arFu3TqVjnzx5EuPHj0dSUhLat2+PGTNmYOHChZg0aRI6duwIc3NzODk5fdDzImqg4XvshJRJ2jgtODiYMcbYuHHjGADm5+cnt51EImENGjRg3333HWOMsSNHjihtnMYYY1paWkxLS0th+b///sv09PSYrq4ue/DggWz50qVLWZMmTVheXp7CPqo0TuvVqxcDwC5evKiwbtKkSQwA27Vrl8K66OhoBoCZmZmxAQMGsMLCQtm6YcOGsV69ejHGGEtISGCOjo4MALt8+bJcGcuXL2cA2KhRo0qNrzTKGqddvHiR8fl8BoAdPXqUMcZYbm4uA8A2bdokt21eXh5r164dA8Bu3LihUFZSUhIDwExMTNgnn3zCCgoKZOvGjh3LOnXqxBhjLDg4WO54UqmpqczT05Pp6emxly9fKpR///59Wfnz58+XW9e7d28GgDVr1oxt3rxZtrywsJA1adKEAWAXLlxQKHPUqFEMAJs4caLc8qdPnzJdXV1mb2/PkpOT5dap0jht1qxZDAAbNGgQk0gksuUvX75kpqamzMzMjL1+/VphP1tbWwaAtWvXTu64GzZsYJaWlqUer7gpU6YwALLP0vuKv/9I1aOKm3De+xV3eHg409LSUvgi/eOPP5i1tTXLzs5mjFWs4maMsR9++IEBYHXr1mUZGRns6tWrTCAQsNDQ0BK3r4qKGwALDAyUW3fz5k22b98+xhhj06ZNYwBY3759FcrIyclhVlZWDAB78uRJqTGWRFpxd+nShS1btowtW7aMzZs3j3388ceMx+MxMzMz9ssvv8i2LywsZMuXL2c5OTkKZR04cIABYJMmTVJYJ624AbBr167Jrbt79y7bu3evbLs1a9aUGOs333zDAJS4Xlpx6+npsbS0NLl1O3bsYACYlZWV3A8GxhhbtmwZA8AWL14st/zWrVsMADMwMGCJiYkKx5P+yFywYIHc8rIq7uDgYMbn8xmfzy/xB8i8efMYADZhwgSFddKKe9u2bXLLIyMj2dq1a0s83vvOnj3L9PT0mJWVFRs3bhxbvHgxW7ZsmewzSDSL7nGTasfDwwNDhw7FwYMH8fXXX6NHjx5gjGHVqlWYPXs2DA0N1XKcqVOn4vLlyzh+/DhGjx6NBw8e4Pvvv4eXl5dayv8Qurq6Cn3O27VrJxuU5s8//wQA9O7dW2FfAwMDNG/eHOfPn8exY8fQqFGjCsViaGiITp06wd/fHz169JAbgEVbWxvLli1DZGQkbty4gdjYWNkleultgpiYmFLL5vP5aNOmjdyy1q1byxoSWllZYf78+Xj+/Dnu3LmDxMREFBQUAIDsVoOy8p2dnWFmZia3zMbGBgDg4+MDHR0duXW2trYAgJSUFLnl0te7efPmsLa2VjhOhw4dsGfPHhw7dgyrV68uNZ73HT16FBKJBN7e3iW2S+jQoQPWrVuH48ePY+fOnSWW0b59e7n/3dzcVL6/3atXL2zevBlfffUV9uzZI1teVk8NUjWo4ibV0sKFC3Ho0CHcuHEDV65cQXp6Ot68eYMpU6ao9Th79uzBo0eP8Pfff6N3794aH2DE3t6+1MFncnJyEBcXBwC4e/cukpKSFLaRNjR6/vz5Bx2/a9euWLx4cZnb5ebmYtKkSdi3bx88PT3RvXt3WcUoJW2IVhJra2uFyrO4lJQUfPbZZzh//jyaNm2K9u3bw8LColzlv0/a2KukddJYJBKJ3PKIiAgARY3fSuoKJ/2REhoaCpFIpLTBZEnlSiSSEsuNj48HUHSvPzExUeG1BYA6deqodKz3ZWZmYuDAgbh06RLc3d0xfPhw2NjYgM/nw8fH54PKJOpFFTeplho0aIABAwbg2LFjWLlyJTIyMjB9+nSYmpqq9Tjm5uZo3bo13rx5g2vXriEoKAi+vr5qPYbU+5VCSVT94tfS0ipxeb9+/dCvXz/UrVu3XLGV1+LFi7Fv3z5069YN58+fl4vn0qVL2Ldvn9L9y3qeEydOxPnz5zF27FiFfvg///yz0u6CAJSOvFfWqHwlKe31dnd3x7JlywAUNWJTNX9llWtnZycrt7RtynssqdmzZ+PSpUuYP38+Vq9e/UGvB6lcVHGTamvRokU4duwYrl69CmNjY8yYMUPtx9i/fz+uX7+OOXPmYOPGjRg6dCgePHig0N1IVdKuQSWdDSYkJFQoVkNDQ9SpUwexsbHo1asXhg8fXqHyKuL48eMAgLFjx5ZasXwoiUQiq5hLa+ldVTw9PQEADg4Oah18Rlquubl5lQ5qwxjD77//Dn19fXz99ddUaXMUdQcj1Vbz5s1l93K/+OILhUulFRUeHo6AgADs378f69evR8+ePfHixQtMnz69xO1Lupz63Xff4fTp07L/zc3NAUB2SVtKIpHg6dOnFY75008/BQDcunWrxPVJSUlYvnw5AgMDK3wsZaSvASthhLv3n3t5McYqtfzykL7eDx48kN1jf99PP/2E3bt3yy0r6b1y+fJlrFmzBgAwaNAg8Pl8BAUFQSgUlljur7/+ih9++KHCz6G4goIC5OTkIC8vr8TuZowxvHnzRq3HJOVHFTep1n7++WdcvHgRixYtUmu5BQUFGDZsGKZNm4bu3buDx+Nh3759sLOzw88//4zDhw8r7OPi4gLg3f1HoVCI+fPnywbbAIoaFQHAiRMn5PbdtWuXWkajWrRoEZycnPDzzz/jzp07cutEIhGmTp2KTZs2wdXVtcLHUqZPnz4AgB07dshVaLm5uRWubLS0tNCzZ08AwLZt2+Qqv5SUFIVKsjL5+fnh888/R2xsLJYuXarwQ+KPP/7AlClT5AZhARTfKwCwfft2WWM3Ly8vzJkzB0KhEHPnzlW4QnPx4kWMHz8eeXl5an0+enp6aNq0KYCixm39+/fH3LlzMWfOHAwbNgzOzs4q9wUnlYculRPOio2Nxc6dO2UNqrZu3QorKytMnDhR1vDGwcEBDg4Ocvv9+++/OH78OF68eAEASE5OxvLly2FsbIwvv/wSwLuR06Rf+suXL4eFhQX8/f2xbt06/Pvvv3jx4gU++ugjREZGwt3dHZcuXUKTJk1w7tw5TJw4ES9evEDr1q3x0UcfASgaDe3HH3/EmjVrEB0djZs3b6JOnToYN26cLLaxY8fizz//xPHjx9GpUye0atUKoaGhsLCwwNChQ7Fr1y6cOHECMTExstHVNm3aJGshnZqaKrt0OnXqVFhZWck9d2tra1y/fh2jRo1Chw4d0KdPH/j4+CArKwtnz56FtrY2rl69KmslXRbptJ7SHx///POPrJGVskZqq1evRmRkJM6fP48GDRqgT58+kEgkOHv2rGzgjoiICCxfvhxdu3ZFx44d8f3338uep1AolD3PyZMnw87OTq78HTt2YNCgQTh8+DBCQkLQtWtXZGdn4/Tp03B1dcXr16/x4MEDLF++HAMHDkSjRo2wcuVKWflv3ryRHbt9+/ZYuXKl7Dm+ePECy5cvR79+/eDh4YFNmzbJWqpLyxwwYACaNGkii8XCwgKbNm3C+fPn0bZtWxgYGODBgwd49OgRNm7cqDAgkLe3N3r16oULFy7I3h8nTpyQm8b222+/hbGxMdasWYMbN26gY8eOMDExwdOnT3Hr1i0sWrRI9n4GgN9//x0hISHIysqS7a+vry97/qr69ddf0a9fP0RGRir8wNTX11dorU6qHlXchPNsbGxkDXHKo379+mXuV9r83E2aNJF9MRdXvEvS+7y8vBAWFoZTp04hISEBU6dOxYABA+S6Senr6+PSpUu4fPkygoKCIBaL8emnn6JVq1Y4depUqS2BTU1NVX4NXF1dcePGDTx69Ah3795Feno6PDw8MGLECLRr1+6D7luWNe77+8zMzHDu3Dncu3cPjx8/RkpKCuzs7DB37lyIxWL89ttvJe5naGio0vN0cnLCvXv3cPXqVTx//hxCoRDNmjXD119/jZiYGIUKR6pOnTqllu/l5VXqumbNmskNQ1ucjo4ONmzYgDlz5uDy5cuIjo6GoaEh2rVrhx49esDY2LjE/c6cOYMzZ84gJCQEhoaG+Oqrr1CvXj3ZeulQvFOmTMGlS5cQFRUFPT09tGrVCocPH1boziZVvDL/EA0aNEBYWBiuX7+OZ8+eITMzE5aWlqhbty7at2//we07iPrwWEk3iQghhBDCSXSPmxBCCKlGqOImhBBCqhGquAkhhJBqhCpuQgghpBqhipsQQgipRqjiJoQQQqqR/wOaiszyzg/zswAAAABJRU5ErkJggg==",
      "text/plain": [
       "<Figure size 500x300 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "import numpy as np\n",
    "import matplotlib.pyplot as plt\n",
    "from tueplots import bundles\n",
    "\n",
    "# configuration and game settings\n",
    "grid_size = 10\n",
    "num_predators = 3 # total predators (1 suspect + N-1 honest)\n",
    "actions = [0, 1, 2, 3, 4] # stay, up, down, left, right\n",
    "action_names = [\"Stay\", \"Up\", \"Down\", \"Left\", \"Right\"]\n",
    "threshold = 20.0 # threshold for rejection\n",
    "trials_per_epsilon = 100 # number of Monte Carlo runs per epsilon\n",
    "max_steps = 5000 # safety cutoff\n",
    "\n",
    "# true epsilons (what the cheater actually uses)\n",
    "true_epsilons = [0.05, 0.1, 0.2, 0.3, 0.4, 0.6, 0.8]\n",
    "\n",
    "# monitor's hypothesis grid (the mixture)\n",
    "mixture_grid = [0.1, 0.3, 0.5, 0.7, 0.9]\n",
    "\n",
    "np.random.seed(0)\n",
    "\n",
    "# game dynamics\n",
    "\n",
    "def get_next_pos(pos, action):\n",
    "    \"\"\"calculates next position given 5x5 grid with boundaries\"\"\"\n",
    "    row, col = pos\n",
    "    if action == 0:   # stay\n",
    "        return (row, col)\n",
    "    elif action == 1: # up\n",
    "        return (max(0, row - 1), col)\n",
    "    elif action == 2: # down\n",
    "        return (min(grid_size - 1, row + 1), col)\n",
    "    elif action == 3: # left\n",
    "        return (row, max(0, col - 1))\n",
    "    elif action == 4: # right\n",
    "        return (row, min(grid_size - 1, col + 1))\n",
    "    return pos\n",
    "\n",
    "def dist_sq(pos1, pos2):\n",
    "    return (pos1[0] - pos2[0])**2 + (pos1[1] - pos2[1])**2\n",
    "\n",
    "# strategies\n",
    "\n",
    "def get_null_probs():\n",
    "    \"\"\"H0: uniform random walk (1/5 prob for each action)\"\"\"\n",
    "    return np.ones(5) / 5.0\n",
    "\n",
    "def get_optimal_probs(pred_pos, prey_pos):\n",
    "    \"\"\"\n",
    "    H1: heuristic strategy; assigns higher probability to moves that reduce distance to prey\"\"\"\n",
    "    probs = np.zeros(5)\n",
    "    current_dist = dist_sq(pred_pos, prey_pos)\n",
    "\n",
    "    # evaluate all moves\n",
    "    for act in actions:\n",
    "        next_p = get_next_pos(pred_pos, act)\n",
    "        new_d = dist_sq(next_p, prey_pos)\n",
    "        \n",
    "        if new_d < current_dist:\n",
    "            probs[act] = 10.0 # strong preference\n",
    "        elif new_d == current_dist:\n",
    "            probs[act] = 1.0  # neutral\n",
    "        else:\n",
    "            probs[act] = 0.1  # avoid moving away\n",
    "            \n",
    "    return probs / np.sum(probs)\n",
    "\n",
    "def get_mixture_probs(pred_pos, prey_pos, epsilon):\n",
    "    \"\"\"observed strategy: (1 - eps) * Null + eps * Optimal\"\"\"\n",
    "    p_null = get_null_probs()\n",
    "    p_opt = get_optimal_probs(pred_pos, prey_pos)\n",
    "    return (1 - epsilon) * p_null + epsilon * p_opt\n",
    "\n",
    "# simulation loop\n",
    "\n",
    "results_tau = []\n",
    "results_std = []\n",
    "\n",
    "print(f\"Starting Multi-Agent Simulation ({num_predators} Predators vs 1 Prey)...\")\n",
    "print(f\"Monitor uses Mixture Martingale over grid: {mixture_grid}\")\n",
    "print(f\"{'True Eps':<10} | {'Avg Tau':<10}\")\n",
    "print(\"-\" * 30)\n",
    "\n",
    "for true_epsilon in true_epsilons:\n",
    "    taus = []\n",
    "    \n",
    "    for _ in range(trials_per_epsilon):\n",
    "        # reset Game\n",
    "        predators = [(0, 0) for _ in range(num_predators)]\n",
    "        prey_pos = (grid_size-1, grid_size-1) \n",
    "        # initialize K martingales\n",
    "        candidate_martingales = np.ones(len(mixture_grid))\n",
    "        \n",
    "        t = 0\n",
    "        detected = False\n",
    "        \n",
    "        while not detected and t < max_steps:\n",
    "            t += 1\n",
    "            \n",
    "            # 1. predator moves\n",
    "            next_predators = []\n",
    "            catch_event = False\n",
    "            \n",
    "            for i in range(num_predators):\n",
    "                current_pos = predators[i]\n",
    "                \n",
    "                if i == 0: \n",
    "                    # the suspect agent plays using the TRUE epsilon\n",
    "                    p_null = get_null_probs()\n",
    "                    p_true = get_mixture_probs(current_pos, prey_pos, true_epsilon)\n",
    "                    # sample Action\n",
    "                    action = np.random.choice(actions, p=p_true)\n",
    "                    \n",
    "                    # the monitor does not know 'true_epsilon'. \n",
    "                    for k, candidate_eps in enumerate(mixture_grid):\n",
    "                        # what would the prob be if the agent was using candidate_eps?\n",
    "                        p_hyp = get_mixture_probs(current_pos, prey_pos, candidate_eps)\n",
    "                        # update component martingale\n",
    "                        lr = p_hyp[action] / p_null[action]\n",
    "                        candidate_martingales[k] *= lr\n",
    "                    \n",
    "                    # the test statistic is the average (uniform mixture)\n",
    "                    mixture_martingale_value = np.mean(candidate_martingales)\n",
    "                    \n",
    "                    if mixture_martingale_value >= threshold:\n",
    "                        detected = True\n",
    "                    \n",
    "                else:\n",
    "                    # honest agents\n",
    "                    p_null = get_null_probs()\n",
    "                    action = np.random.choice(actions, p=p_null)\n",
    "                \n",
    "                # execute move\n",
    "                new_pos = get_next_pos(current_pos, action)\n",
    "                next_predators.append(new_pos)\n",
    "                \n",
    "                if new_pos == prey_pos:\n",
    "                    catch_event = True\n",
    "            \n",
    "            predators = next_predators\n",
    "\n",
    "            # 2. prey move\n",
    "            prey_action = np.random.choice(actions) \n",
    "            prey_pos = get_next_pos(prey_pos, prey_action)\n",
    "            \n",
    "            if any(p == prey_pos for p in predators):\n",
    "                catch_event = True\n",
    "            \n",
    "            # 3. reset if caught\n",
    "            if catch_event:\n",
    "                predators = [(0, 0) for _ in range(num_predators)]\n",
    "                prey_pos = (grid_size-1, grid_size-1)\n",
    "\n",
    "        taus.append(t)\n",
    "\n",
    "    avg_tau = np.mean(taus)\n",
    "    results_tau.append(avg_tau)\n",
    "    results_std.append(np.std(taus) / np.sqrt(trials_per_epsilon)) \n",
    "    \n",
    "    print(f\"{true_epsilon:<10.2f} | {avg_tau:<10.1f}\")\n",
    "\n",
    "# plotting\n",
    "\n",
    "# style setup\n",
    "plt.rcParams.update({**bundles.icml2024(), \"text.usetex\": False})\n",
    "plt.rcParams.update({\n",
    "    \"axes.labelsize\": 16,\n",
    "    \"axes.titlesize\": 16,\n",
    "    \"xtick.labelsize\": 14,\n",
    "    \"ytick.labelsize\": 14,\n",
    "    \"legend.fontsize\": 14,\n",
    "    \"lines.linewidth\": 2.5,\n",
    "})\n",
    "\n",
    "# theoretical scaling: T ~ K / epsilon^2\n",
    "# anchor to the last point (largest epsilon)\n",
    "anchor_idx = -1 \n",
    "last_eps = true_epsilons[anchor_idx]\n",
    "last_tau = results_tau[anchor_idx]\n",
    "\n",
    "# calculate constant K\n",
    "K = last_tau * (last_eps ** 2)\n",
    "\n",
    "# generate theoretical curve\n",
    "theoretical_y = [K / (e**2) for e in true_epsilons]\n",
    "\n",
    "plt.figure(figsize=(5, 3))\n",
    "\n",
    "# plot empirical data\n",
    "plt.errorbar(true_epsilons, results_tau, yerr=results_std, fmt='o-', \n",
    "             label='Empirical', \n",
    "             color='#1f77b4', ecolor='#1f77b4', capsize=5)\n",
    "\n",
    "# plot theoretical curve\n",
    "plt.plot(true_epsilons, theoretical_y, linestyle='--', \n",
    "         label='Theoretical', \n",
    "         color='#2ca02c', linewidth=2.5)\n",
    "\n",
    "plt.ylim(4,5000)\n",
    "\n",
    "plt.yscale('log')\n",
    "plt.xscale('linear')\n",
    "\n",
    "plt.xlabel(r\"Mixture Parameter $\\varepsilon$\")\n",
    "plt.ylabel(r\"Avg Detection Time\")\n",
    "plt.grid(True, which='both', alpha=0.3)\n",
    "plt.legend()\n",
    "\n",
    "save_path = 'plots/predator-prey/scaling.pdf'\n",
    "plt.savefig(save_path, format=\"pdf\", bbox_inches=\"tight\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "id": "3fb2056e",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2026-07-21T14:53:24.612457Z",
     "iopub.status.busy": "2026-07-21T14:53:24.612128Z",
     "iopub.status.idle": "2026-07-21T14:53:26.200658Z",
     "shell.execute_reply": "2026-07-21T14:53:26.199658Z"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Running visual episode with Epsilon=0.6...\n",
      "Step  | Suspect Pos     | Action     | Martingale   | Probabilities (Stay, Up, Down, Left, Right)\n",
      "----------------------------------------------------------------------------------------------------------------------------------\n",
      "0     | (0, 0)          | Down       | 1.00         | [  0.11,   0.11,   0.34,   0.11,   0.34]\n",
      "1     | (1, 0)          | Right      | 1.59         | [  0.11,   0.08,   0.35,   0.11,   0.35]\n",
      "2     | (1, 1)          | Right      | 2.71         | [  0.11,   0.08,   0.36,   0.08,   0.36]\n",
      "3     | (1, 2)          | Down       | 4.97         | [  0.11,   0.08,   0.36,   0.08,   0.36]\n",
      "4     | (2, 2)          | Stay       | 9.48         | [  0.11,   0.08,   0.36,   0.08,   0.36]\n",
      "5     | (2, 2)          | Right      | 4.34         | [  0.11,   0.08,   0.36,   0.08,   0.36]\n",
      "6     | (2, 3)          | Right      | 8.04         | [  0.11,   0.08,   0.36,   0.08,   0.36]\n",
      "7     | (2, 4)          | Right      | 15.38        | [  0.11,   0.08,   0.36,   0.08,   0.36]\n",
      "8     | (2, 5)          | Up         | 30.20        | [  0.11,   0.08,   0.36,   0.08,   0.36]\n",
      "9     | (1, 5)          | Stay       | 8.37         | [  0.11,   0.08,   0.36,   0.08,   0.36]\n",
      "10    | (1, 5)          | Down       | 4.47         | [  0.11,   0.08,   0.36,   0.08,   0.36]\n",
      "11    | (2, 5)          | Right      | 7.75         | [  0.11,   0.08,   0.36,   0.08,   0.36]\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/tmp/ipykernel_105918/289974491.py:145: UserWarning: The figure layout has changed to tight\n",
      "  plt.tight_layout()\n"
     ]
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 1200x1050 with 12 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "import csv\n",
    "\n",
    "# colors for visualization\n",
    "color_prey = '#2ca02c'    \n",
    "color_suspect = '#d62728'   \n",
    "color_honest = '#1f77b4'    \n",
    "color_grid = '#e0e0e0'\n",
    "\n",
    "np.random.seed(0)\n",
    "\n",
    "# visualization\n",
    "\n",
    "def run_visual_episode(true_epsilon, steps_to_record=25, save_csv_name=None):\n",
    "    # setup\n",
    "    predators = [(0, 0) for _ in range(num_predators)]\n",
    "    prey_pos = (grid_size-1, grid_size-1)\n",
    "    \n",
    "    # martingale state\n",
    "    candidate_martingales = np.ones(len(mixture_grid))\n",
    "    history = []\n",
    "    csv_rows = []\n",
    "    \n",
    "    print(f\"Running visual episode with Epsilon={true_epsilon}...\")\n",
    "    print(f\"{'Step':<5} | {'Suspect Pos':<15} | {'Action':<10} | {'Martingale':<12} | {'Probabilities (Stay, Up, Down, Left, Right)'}\")\n",
    "    print(\"-\" * 130)\n",
    "    \n",
    "    for t in range(steps_to_record):\n",
    "        # 1. record State before moves\n",
    "        current_martingale = np.mean(candidate_martingales)\n",
    "        history.append({\n",
    "            'step': t,\n",
    "            'predators': predators.copy(),\n",
    "            'prey': prey_pos,\n",
    "            'martingale': current_martingale,\n",
    "            'caught': any(p == prey_pos for p in predators)\n",
    "        })\n",
    "        \n",
    "        # 2. agent moves\n",
    "        next_predators = []\n",
    "        \n",
    "        # suspect (index 0)\n",
    "        suspect_pos = predators[0]\n",
    "        p_null = get_null_probs()\n",
    "        p_true = get_mixture_probs(suspect_pos, prey_pos, true_epsilon)\n",
    "        action_s = np.random.choice(actions, p=p_true)\n",
    "        \n",
    "        # logging\n",
    "        act_name = action_names[action_s]\n",
    "        prob_str = \", \".join([f\"{v:6.2f}\" for v in p_true])\n",
    "        print(f\"{t:<5} | {str(suspect_pos):<15} | {act_name:<10} | {current_martingale:<12.2f} | [{prob_str}]\")\n",
    "        \n",
    "        # collect data for CSV\n",
    "        if save_csv_name:\n",
    "            row = [t, suspect_pos[0], suspect_pos[1], act_name, current_martingale] + list(p_true)\n",
    "            csv_rows.append(row)\n",
    "        \n",
    "        # monitor update\n",
    "        for k, candidate_eps in enumerate(mixture_grid):\n",
    "            p_hyp = get_mixture_probs(suspect_pos, prey_pos, candidate_eps)\n",
    "            lr = p_hyp[action_s] / p_null[action_s]\n",
    "            candidate_martingales[k] *= lr\n",
    "            \n",
    "        next_predators.append(get_next_pos(suspect_pos, action_s))\n",
    "        \n",
    "        # honest agents\n",
    "        for i in range(1, num_predators):\n",
    "            p_honest = get_null_probs()\n",
    "            act_h = np.random.choice(actions, p=p_honest)\n",
    "            next_predators.append(get_next_pos(predators[i], act_h))\n",
    "            \n",
    "        predators = next_predators\n",
    "        \n",
    "        # prey move\n",
    "        act_prey = np.random.choice(actions)\n",
    "        prey_pos = get_next_pos(prey_pos, act_prey)\n",
    "        \n",
    "        # reset if caught\n",
    "        if any(p == prey_pos for p in predators):\n",
    "            predators = [(0, 0) for _ in range(num_predators)]\n",
    "            prey_pos = (grid_size-1, grid_size-1)\n",
    "            \n",
    "    # save CSV\n",
    "    if save_csv_name:\n",
    "        csv_path = f'results/{save_csv_name}'\n",
    "        with open(csv_path, 'w', newline='') as f:\n",
    "            writer = csv.writer(f)\n",
    "            header = ['Step', 'Suspect_Row', 'Suspect_Col', 'Action', 'Martingale', \n",
    "                      'Prob_Stay', 'Prob_Up', 'Prob_Down', 'Prob_Left', 'Prob_Right']\n",
    "            writer.writerow(header)\n",
    "            writer.writerows(csv_rows)   \n",
    "\n",
    "    return history\n",
    "\n",
    "# plotting function\n",
    "def plot_film_strip(history, save_filename):\n",
    "    n_steps = len(history)\n",
    "    cols = 4\n",
    "    rows = (n_steps + cols - 1) // cols\n",
    "    \n",
    "    fig, axes = plt.subplots(rows, cols, figsize=(cols*3, rows*3.5))\n",
    "    axes_flat = axes.flatten()\n",
    "    \n",
    "    for i, state in enumerate(history):\n",
    "        ax = axes_flat[i]\n",
    "        \n",
    "        # grid setup\n",
    "        ax.set_xlim(-0.5, grid_size-0.5)\n",
    "        ax.set_ylim(grid_size-0.5, -0.5) # invert Y to match matrix coords\n",
    "        ax.set_xticks(np.arange(grid_size))\n",
    "        ax.set_yticks(np.arange(grid_size))\n",
    "        ax.grid(color=color_grid, linestyle='-', linewidth=1.5)\n",
    "        ax.set_xticklabels([])\n",
    "        ax.set_yticklabels([])\n",
    "        \n",
    "        # draw prey\n",
    "        py, px = state['prey']\n",
    "        ax.scatter(px, py, s=300, marker='*', c=color_prey, label='Prey', zorder=10)\n",
    "        \n",
    "        # draw honest predators\n",
    "        for idx, hp in enumerate(state['predators'][1:]):\n",
    "            hy, hx = hp\n",
    "            label = 'Honest' if idx == 0 else None\n",
    "            ax.scatter(hx, hy, s=150, marker='o', c=color_honest, alpha=0.6, label=label)\n",
    "            \n",
    "        # draw suspect predator\n",
    "        sy, sx = state['predators'][0]\n",
    "        ax.scatter(sx, sy, s=200, marker='D', c=color_suspect, edgecolors='black', label='Suspect', zorder=11)\n",
    "        \n",
    "        # info\n",
    "        m_val = state['martingale']\n",
    "        title_color = 'black'\n",
    "        if m_val > threshold: title_color = 'red'\n",
    "        \n",
    "        ax.set_title(f\"Step {state['step']}\\nMartingale: {m_val:.2f}\", \n",
    "                     fontsize=12, fontweight='bold', color=title_color)\n",
    "        \n",
    "        # add legend only to first plot\n",
    "        if i == 0:\n",
    "            ax.legend(loc='upper left', bbox_to_anchor=(-0.1, 1.3), fontsize=10, ncol=3)\n",
    "\n",
    "    # hide empty subplots\n",
    "    for j in range(n_steps, len(axes_flat)):\n",
    "        axes_flat[j].axis('off')\n",
    "        \n",
    "    plt.tight_layout()\n",
    "    plt.savefig(f'plots/{save_filename}', format='pdf', bbox_inches='tight')\n",
    "\n",
    "# execution\n",
    "\n",
    "# run a short episode with a distinct cheater (epsilon=0.6)\n",
    "history_data = run_visual_episode(true_epsilon=0.6, steps_to_record=12, save_csv_name=\"predator-prey/history.csv\")\n",
    "plot_film_strip(history_data, \"predator-prey/gameplay.pdf\")\n",
    "plt.show()"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.11.9"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}