roofline.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 5,
   "id": "0b3b8876-b3fd-4480-a4fc-ff953d35f9e2",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAkgAAAHHCAYAAABEEKc/AAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjguMywgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy/H5lhTAAAACXBIWXMAAA9hAAAPYQGoP6dpAACq+0lEQVR4nOzdd3gU1dfA8e/upncSSEFCR3oHISAdQiAiSC8KKGJDpPhDRBEBURAULICiohQpAkoVkFBCkdA7CArSBJJASG+72Z33jzH7GgiQsptNOZ/n2Yed2Tt3zlwWcnLnzr0aRVEUhBBCCCGEmdbWAQghhBBCFDaSIAkhhBBC3EMSJCGEEEKIe0iCJIQQQghxD0mQhBBCCCHuIQmSEEIIIcQ9JEESQgghhLiHJEhCCCGEEPeQBEkIIYQQ4h6SIAlRQP766y+Cg4Px9PREo9Gwbt06Fi1ahEaj4cqVK+Zybdu2pW3btjaL0xbyc80VK1Zk6NChFoslIyODt956i8DAQLRaLT169LBY3UKIokMSJFFiZCYjmS87Ozsee+wxhg4dyo0bN6x+/iFDhnD69Gk+/PBDli5dSpMmTax+zty4cuWKuW2mTZuWbZlBgwah0Whwc3Mr4OgKzvfff8+sWbPo3bs3ixcvZsyYMQ8s27ZtWzQaDd26dbvvs8z2/OSTTwB444030Gg0XLx48YH1vfvuu2g0Gk6dOgWoyd9TTz2Vpcy932Fvb28aN27MqFGjOHfu3CPjuNfmzZvRaDSULVsWk8n0wNgeZO/evfTt25fHHnsMBwcHPD09adasGVOnTiUqKipL2cz2ynx5e3vTtGlTvv/++yznbtu2LXXq1Mn2fHfu3EGj0TB58uRcxypEbtjZOgAhCtrUqVOpVKkSaWlpHDhwgEWLFrFv3z7OnDmDk5OTVc6ZmppKREQE7777Lq+//vpDy27bts0qMeSUk5MTK1asYOLEiVn2Jycns379equ1UWGxc+dOHnvsMebMmZPjYzZt2sTRo0dp3LjxA8sMGjSIL7/8kuXLlzNp0qRsy6xYsYK6detSr169h56vU6dODB48GEVRiI+P5+TJkyxevJj58+fz8ccfM3bs2BzHvmzZMipWrMiVK1fYuXMnHTt2zPGxkyZN4oMPPqBy5coMHTqUypUrk5aWxtGjR/n0009ZvHgxly5dynJMuXLlmD59OgC3b99myZIlDBs2jD///JMZM2bk+NxCWJskSKLE6dKli7n35sUXX6R06dJ8/PHHbNiwgb59+1rlnLdv3wbAy8vrkWUdHBysEkNOde3alV9++YWTJ09Sv3598/7169ej1+sJCQlh586dNozQuqKjo3P095SpfPnyJCYmMmXKFDZs2PDAcs2aNaNq1aqsWLEi2wQpIiKCy5cv5yhJePzxx3n22Wez7JsxYwbdunXjzTffpEaNGnTt2vWR9WQmvdOnT+eHH35g2bJlOU6QfvrpJz744AP69u3L0qVL7/vezpkzJ9sk09PTM0vsL7/8MtWrV2fu3Ll88MEH2Nvb5+j8Qlib3GITJV6rVq0A7vtNd+fOnbRq1QpXV1e8vLzo3r07f/zxx33HHz9+nC5duuDh4YGbmxsdOnTgwIED5s8nT55MhQoVABg3bhwajYaKFSs+MJ57x+OEh4ej0WhYtWoVH374IeXKlcPJyYkOHTpke7vm4MGDhISE4OnpiYuLC23atOH333/PcXsEBQVRqVIlli9fnmX/smXLCAkJwdvbO9vj5s+fT+3atXF0dKRs2bKMGDGCuLi4+8p98803VKlSBWdnZ5544gn27t2bbX3p6em8//77VK1aFUdHRwIDA3nrrbdIT0/P8bX8V3JyMm+++SaBgYE4OjpSvXp1PvnkExRFAf7/VtSuXbs4e/as+TZQeHj4Q+t1d3dnzJgxbNy4kWPHjj207KBBgzh//ny25ZYvX45Go2HAgAF5uj4fHx9WrlyJnZ0dH374YY6OWbt2LampqfTp04f+/fvzyy+/kJaWlqNjJ02aROnSpVm4cGG2Sb2np2eOboO5uLjQvHlzkpOTzb9I5FZiYiKjR4+mYsWKODo64uvrS6dOnR759yHEw0iCJEq8zAHSpUqVMu/bvn07nTt3Jjo6msmTJzN27Fj2799Py5YtswyoPnv2LK1ateLkyZO89dZbvPfee1y+fJm2bdty8OBBAHr27Gn+TXrAgAEsXbqUzz77LNdxzpgxg7Vr1/K///2PCRMmcODAAQYNGpSlzM6dO2ndujUJCQm8//77fPTRR8TFxdG+fXsOHTqU43MNGDCAlStXmpOHO3fusG3bNgYOHJht+cmTJzNixAjKli3Lp59+Sq9evViwYAHBwcEYDAZzuYULF/Lyyy/j7+/PzJkzadmyJU8//TTXr1/PUp/JZOLpp5/mk08+oVu3bnz55Zf06NGDOXPm0K9fvxxfRyZFUXj66aeZM2cOISEhzJ49m+rVqzNu3Djz7agyZcqwdOlSatSoQbly5Vi6dClLly6lZs2aj6x/1KhRlCpV6pEJQebf173Jp9FoZNWqVbRq1Yry5cvn+voylS9fnjZt2nDgwAESEhIeWX7ZsmW0a9cOf39/+vfvT2JiIhs3bnzkcX/++Sd//vknPXr0sMh4tL///hudTpernrv/euWVV/jqq6/o1asX8+fP53//+x/Ozs7Z/kIjRI4pQpQQP/zwgwIo27dvV27fvq1cv35dWbNmjVKmTBnF0dFRuX79urlsgwYNFF9fXyUmJsa87+TJk4pWq1UGDx5s3tejRw/FwcFBuXTpknnfzZs3FXd3d6V169bmfZcvX1YAZdasWdnGdPnyZfO+Nm3aKG3atDFv79q1SwGUmjVrKunp6eb9n3/+uQIop0+fVhRFUUwmk1KtWjWlc+fOislkMpdLSUlRKlWqpHTq1Omh7fPfGM+cOaMAyt69exVFUZR58+Ypbm5uSnJysjJkyBDF1dXVfFx0dLTi4OCgBAcHK0aj0bx/7ty5CqB8//33iqIoil6vV3x9fZUGDRpkuY5vvvlGAbJc89KlSxWtVms+f6avv/5aAZTff//dvK9ChQrKkCFDHnpt69atUwBl2rRpWfb37t1b0Wg0ysWLF8372rRpo9SuXfuh9WVXdsqUKQqgHD16VFGUB/+dN23aVClXrlyWttq6dasCKAsWLMhStkKFCkpoaGiWfYAyYsSIB8Y0atQoBVBOnjz50DiioqIUOzs75dtvvzXva9GihdK9e/dHXvf69esVQPnss8+y7DeZTMrt27ezvAwGg/nzNm3aKDVq1DB/9scffyhvvPGGAijdunXLUu5Bfwe3b99WAOX999837/P09HxomwiRF9KDJEqcjh07UqZMGQIDA+nduzeurq5s2LCBcuXKAXDr1i1OnDjB0KFDs9xOqlevHp06dWLz5s2A+lv/tm3b6NGjB5UrVzaXCwgIYODAgezbty9Hv8Xn1PPPP5/lVkbmrcG///4bgBMnTvDXX38xcOBAYmJiuHPnDnfu3CE5OZkOHTqwZ8+eHD+lVLt2berVq8eKFSsAtceje/fuuLi43Fd2+/bt6PV6Ro8ejVb7//+lDB8+HA8PD3799VcAjhw5QnR0NK+88kqW6xg6dCienp5Z6ly9ejU1a9akRo0a5uu4c+cO7du3B2DXrl05uo5MmzdvRqfT8cYbb2TZ/+abb6IoClu2bMlVfdnJ7EWaMmXKQ8s9++yz/PPPP+zZs8e8b/ny5Tg4ONCnT598x5HZo5OYmPjQcitXrkSr1dKrVy/zvgEDBrBlyxZiY2Mfemzm9/re3qP4+HjKlCmT5XXixIksZc6fP2/+rGbNmnz55ZeEhoby/fff5/QS7+Pl5cXBgwe5efNmnusQ4l6SIIkSZ968eYSFhbFmzRq6du3KnTt3cHR0NH9+9epVAKpXr37fsTVr1jQnHbdv3yYlJeWB5Uwm0323jvLj3lsvmbcEM3+Y/fXXX4A6ncC9P6S+++470tPTiY+Pz/H5Bg4cyOrVq7l48SL79+9/4O21B7WXg4MDlStXNn+e+We1atWylLO3t8+SYGZey9mzZ++7jscffxxQB1LnxtWrVylbtizu7u5Z9mfePsuMLT88PT0ZPXo0GzZs4Pjx4w8s179/f3Q6nfk2W1paGmvXrqVLly5ZbvPmVVJSEsB913qvH3/8kSeeeIKYmBguXrzIxYsXadiwIXq9ntWrVz/02My6M8+Vyc3NjbCwMMLCwhg3bly2x1asWJGwsDC2b9/Ovn37iIyMZNOmTZQuXTqnlwio0x1kmjlzJmfOnCEwMJAnnniCyZMnm39xECKv5Ck2UeI88cQT5qfYevTowZNPPsnAgQO5cOFCoZ7fR6fTZbtf+XecUGbv0KxZs2jQoEG2ZXNzfQMGDGDChAkMHz4cHx8fgoODcxdwPphMJurWrcvs2bOz/TwwMLDAYsmNUaNGMWfOHKZMmfLAcWaZA4h//vln5s2bx8aNG0lMTLxvPFlenTlzBp1OR6VKlR5Y5q+//uLw4cPA/QkrqGOTXnrppQceX6NGDfO5/svOzs78FNw///yT7bGurq6PfFLOycmJ1NTUbD9LSUkxl8nUt29fWrVqxdq1a9m2bRuzZs3i448/5pdffqFLly4PPZcQDyIJkijRdDod06dPp127dsydO5e3337b/MTZhQsX7it//vx5SpcujaurK05OTri4uDywnFarLdAf5FWqVAHAw8MjV3PZPEj58uVp2bIl4eHhvPrqq9jZZf/fxX/b6789QXq9nsuXL5tjySz3119/mW+VARgMBi5fvpxlSoEqVapw8uRJOnTokKWnIK8qVKjA9u3bSUxMzNKzcv78+Syx5VdmL9LkyZMZMmTIA8sNGjSIrVu3smXLFpYvX46Hh0e2k03m1rVr19i9ezdBQUEP7UFatmwZ9vb2LF269L7Ee9++fXzxxRdcu3btgQPGq1evTrVq1Vi3bh2fffYZrq6u+Y79vypUqMDOnTtJTU3F2dk5y2eZ/97u/TsLCAjgtdde47XXXiM6OppGjRrx4YcfSoIk8kxusYkSr23btjzxxBN89tlnpKWlERAQQIMGDVi8eHGWx9TPnDnDtm3bzPPL6HQ6goODWb9+fZYn26Kioli+fDlPPvkkHh4eBXYdjRs3pkqVKnzyySf33foA8vQI9bRp03j//fcZOXLkA8t07NgRBwcHvvjiC3NvFqhPrMXHxxMaGgpAkyZNKFOmDF9//TV6vd5cbtGiRfdNB9C3b19u3LjBt99+e9/5UlNTSU5OztV1dO3aFaPRyNy5c7PsnzNnDhqNxqI/REePHo2XlxdTp059YJkePXrg4uLC/Pnz2bJlCz179sz3BJx3795lwIABGI1G3n333YeWXbZsGa1ataJfv3707t07yyvz1ljm+LMHmTx5Mnfu3GH48OFZnlTM9N/vQm517doVg8HAggULsuw3mUx89dVXODg40KFDB0AdC3jvrWNfX1/Kli2b5ykhhADpQRICUOcn6tOnD4sWLeKVV15h1qxZdOnShaCgIIYNG0ZqaipffvnlfXO7TJs2jbCwMJ588klee+017OzsWLBgAenp6cycObNAr0Gr1fLdd9/RpUsXateuzfPPP89jjz3GjRs32LVrFx4eHjl6hPu/2rRpQ5s2bR5apkyZMkyYMIEpU6YQEhLC008/zYULF5g/fz5NmzY1Twpob2/PtGnTePnll2nfvj39+vXj8uXL/PDDD/eNQXruuedYtWoVr7zyCrt27aJly5YYjUbOnz/PqlWr+O2333K1VEu3bt1o164d7777LleuXKF+/fps27aN9evXM3r0aHPvmyV4enoyatSohw7WdnNzo0ePHuZxSLm9vfbnn3/y448/oigKCQkJnDx5ktWrV5OUlMTs2bMJCQl54LEHDx7k4sWLD5zR/bHHHqNRo0YsW7aM8ePHP7CegQMHcubMGaZPn86hQ4fo378/lSpVIjk5mTNnzrBixQrc3d3zNK6qW7duBAcHM2bMGA4dOkSLFi1ISUlhw4YN/P7770ybNo0yZcoA6mD0cuXK0bt3b+rXr4+bmxvbt2/n8OHDfPrpp7k+txBmNn2GTogClPlI/eHDh+/7zGg0KlWqVFGqVKmiZGRkKIqiKNu3b1datmypODs7Kx4eHkq3bt2Uc+fO3XfssWPHlM6dOytubm6Ki4uL0q5dO2X//v1ZyljiMf/Vq1dnW+cPP/yQZf/x48eVnj17Kj4+Poqjo6NSoUIFpW/fvsqOHTse2j4PivFe9z7mn2nu3LlKjRo1FHt7e8XPz0959dVXldjY2PvKzZ8/X6lUqZLi6OioNGnSRNmzZ89916wo6rQAH3/8sVK7dm3F0dFRKVWqlNK4cWNlypQpSnx8vLlcTh7zVxRFSUxMVMaMGaOULVtWsbe3V6pVq6bMmjUry5QIipL3x/z/KzY2VvH09Hxoe/76668KoAQEBGR55P+/HvSYf+ZLq9UqXl5eSsOGDZVRo0YpZ8+eva+Oe/9eR44cqQBZpqa41+TJk7NMFfAw4eHhSu/evZWAgADF3t5e8fDwUJo0aaK8//77yq1bt7KUzU3bpqWlKZMnT1Zq1KihODo6Kq6urkrz5s2VH3/8MUu59PR0Zdy4cUr9+vUVd3d3xdXVValfv74yf/78HJ1HiAfRKEo++kGFEEIIIYohGYMkhBBCCHEPSZCEEEIIIe4hCZIQQgghxD0KTYI0Y8YMNBoNo0ePNu9LS0tjxIgR+Pj44ObmRq9evYiKispy3LVr1wgNDcXFxQVfX1/GjRtHRkZGAUcvhBBCiOKkUCRIhw8fZsGCBdSrVy/L/jFjxrBx40ZWr17N7t27uXnzJj179jR/bjQaCQ0NRa/Xs3//fhYvXsyiRYuYNGlSQV+CEEIIIYoRmz/FlpSURKNGjZg/fz7Tpk2jQYMGfPbZZ+ZFD5cvX07v3r0BddbbmjVrEhERQfPmzdmyZQtPPfUUN2/exM/PD4Cvv/6a8ePHc/v27SwLYgohhBBC5JTNJ4ocMWIEoaGhdOzYkWnTppn3Hz16FIPBkGXJhBo1alC+fHlzghQREUHdunXNyRFA586defXVVzl79iwNGzbM9pzp6elZZlg1mUzcvXsXHx8fiyxrIIQQQgjrUxSFxMREypYti1Zr2ZtiNk2QVq5cybFjx8yLJv5XZGQkDg4OeHl5Zdnv5+dHZGSkucx/k6PMzzM/e5Dp06c/dJZbIYQQQhQd169fp1y5chat02YJ0vXr1xk1ahRhYWH5XoMotyZMmMDYsWPN2/Hx8ZQvX54///wTb2/vAo2lODEYDOzatYt27dphb29v63CKNGlLy5G2tAxpR8uRtrScu3fv8vjjjz90cea8slmCdPToUfOKy5mMRiN79uxh7ty5/Pbbb+j1euLi4rL0IkVFReHv7w+Av78/hw4dylJv5lNumWWy4+joiKOj4337vb298fHxyc9llWgGgwEXFxd8fHzkH30+SVtajrSlZUg7Wo60peVZY3iMzZ5i69ChA6dPn+bEiRPmV5MmTRg0aJD5vb29PTt27DAfc+HCBa5du0ZQUBAAQUFBnD59mujoaHOZsLAwPDw8qFWrVoFfkxBCCCGKB5v1ILm7u1OnTp0s+1xdXfHx8THvHzZsGGPHjsXb2xsPDw9GjhxJUFAQzZs3ByA4OJhatWrx3HPPMXPmTCIjI5k4cSIjRozItodICCGEECInbP4U28PMmTMHrVZLr169SE9Pp3PnzsyfP9/8uU6nY9OmTbz66qsEBQXh6urKkCFDmDp1qg2jFkIIIURRV6gSpPDw8CzbTk5OzJs3j3nz5j3wmAoVKrB582YrR6ZOBaDX661+nqLMYDBgZ2dHWloaRqPR1uEUGHt7e3Q6na3DEEIIYUGFKkEqrPR6PZcvX8ZkMtk6lEJNURT8/f25fv16iZtPysvLC39//xJ33UIIUVxJgvQIiqJw69YtdDodgYGBFp+IqjgxmUwkJSXh5uZWYtpJURRSUlLMDwoEBATYOCIhhBCWIAnSI2RkZJCSkkLZsmVxcXGxdTiFWuZtSCcnpxKTIAE4OzsDEB0dja+vr9xuE0KIYqDk/BTLo8yxNLKum3iYzOTZYDDYOBIhhBCWIAlSDsnYEvEw8v0QQojiRRIkIYQQQoh7SIIkhBBCCHEPSZCKuYiICHQ6HaGhofd99sYbb9C4cWMcHR1p0KBBtsefOnWKVq1a4eTkRGBgIDNnzryvzOrVq6lRowYuLi60aNGiQOalEkIIIaxJEqRibuHChYwcOZI9e/Zw8+bN+z5/4YUX6NevX7bHJiQkEBwcTIUKFTh69CizZs1i8uTJfPPNN+Yy+/fvZ8CAAQwbNoyjR48SGhpKz549OXPmjNWuSQghhLA2ecy/GEtKSuKnn37iyJEjREZGsmjRIt555x3z51988QUAt2/f5tSpU/cdv2zZMvR6Pd9//z0ODg7Url2bEydOMHv2bF566SUAPv/8c0JCQhg3bhwmk4l3332XvXv3MnfuXL7++uuCuVAhhBDCwqQHKZcURUGfrLfJS1GUXMW6atUqatSoQfXq1Xn22Wf5/vvvc1VHREQErVu3zjLFQefOnblw4QKxsbHmMh07dsxyXHBwMBEREbmKVQghhChMpAcplwwpBqa7TbfJuSckTcDBNefzMS1cuJBnn30WgJCQEOLj49m9ezdt27bN0fGRkZFUqlQpyz4/Pz/zZ6VKlSIyMtK8779lIiMjcxynEEIIUdhID1IxdeHCBQ4dOsSAAQMAsLOzo1+/fixcuNDGkQkhhBCFn/Qg5ZK9iz0TkibY7Nw5tXDhQjIyMihbtqx5n6IoODo6MnfuXDw9PR9Zh7+/P1FRUVn2ZW77+/s/tEzm50IIIURRJAlSLmk0mlzd5rKFjIwMlixZwqeffkpwcHCWz3r06MGKFSt45ZVXHllPUFAQ7777LgaDAXt7NTkLCwujevXqlCpVylxmx44djB492nzc9u3bCQoKstwFCSGEEAVMbrEVQ5s2bSI2NpZhw4ZRp06dLK9evXqZb7NdvHiREydOEBkZSWpqKidOnODEiRPo9XoABg4ciIODA8OGDePs2bP89NNPfP7554wdO9Z8rlGjRrF161Y+/fRTzp8/z4wZMzhy5Aivv/66Ta5dCCGEsATpQSqGFi5cSMeOHbO9jdarVy9mzpzJqVOneOONN9i9e7f5s4YNGwJw+fJlKlasiKenJ9u2bWPEiBE0btyY0qVLM2nSJPMj/gAtWrRg+fLlTJw4kXfeeYfKlSvzyy+/UKdOHetfqBBCCGElkiAVQxs3bnzgZ0888YT5Uf/w8PBH1lWvXj327t370DJ9+vShT58+mEwmEhIS8PDwyFW8QgghRGEjt9iEEEIIIe4hCZIQQgghxD0kQRJCCCGEuIckSEIIIYQQ95AESQghhBDiHpIgCSGEEELcQxIkIYQQQoh7SIIkhBBCCHEPSZCEEEIIIe4hM2kLi7hw4QJt2rTh8OHDxWYmbb1ez+OPP86aNWto0qSJrcMRouhKT4f//Q/27bN1JIWCnaLQJiEBu/ffB43G1uEUaXYZGdar22o1C5saOnQocXFxrFu3Lsv+8PBw2rVrR2xsLCdOnGDOnDkcOnSIhIQEqlWrxrhx4xg0aFCWY1avXs17773HlStXqFatGh9//DFdu3bNUmbChAm8/vrruLu7W/vSCoyDgwP/+9//GD9+PDt27LB1OEIUTSkp0LMn/PabrSMpNDSAl62DKCasmV5KglSC7d+/n3r16jF+/Hj8/PzYtGkTgwcPxtPTk6eeespcZsCAAUyfPp2nnnqK5cuX06NHD44dO2ZekPbatWts2rSJzz///IHnMhqNaDQatNrCc1dXr9fj4ODw0DKDBg3izTff5OzZs9SuXbuAIhOimEhKgm7dIDwcXFxg3jwICLB1VDaXkZHBocOHeaJpU+zs5MdwfmQkJEDfvtapXBFKfHy8Aih37ty577PU1FTl3LlzSmpqqg0iy7shQ4Yo3bt3v2//rl27FECJjY3N9riuXbsqzz//vHm7b9++SmhoaJYyzZo1U15++WXz9qxZs5QmTZooRqNRiY2NVYxGo/LDDz8onp6eyvr165WaNWsqOp1OuXz5stKmTRtl1KhRWerr3r27MmTIEPN2hQoVlA8++EB57rnnFFdXV6V8+fLK+vXrlejoaOXpp59WXF1dlbp16yqHDx/OUs+aNWuUWrVqKQ4ODkqFChWUTz75JMvnFSpUUKZOnao899xziru7uzJkyBAlPT1dGTFihOLv7684Ojoq5cuXVz766KMsx7Vr106ZOHFitu2VydLfE71er6xbt07R6/UWqa8kk7a0jFy3Y1ycorRooSigKO7uirJ3r3UDLELkO2k5d+7cUQAlPj7e4nUXnl/niwhFgeRk27wUxfrXFx8fj7e3t3k7IiKCjh07ZinTuXNnIiIizNt79+7NdoxOSkoKH3/8Md999x1nz57F19c3x3HMmTOHli1bcvz4cUJDQ3nuuecYPHgwzz77LMeOHaNKlSoMHjwY5d9GOXr0KH379qV///6cPn2ayZMn895777Fo0aIs9X7yySfUr1+f48eP89577/HFF1+wYcMGVq1axYULF1i2bBkVK1bMcswTTzzB3r17cxy7ECXe3bvQsSPs3w9eXrB9Ozz5pK2jEiJXpG8vl1JSwM3NNudOSgJX15yX37RpE273BGs0Gh9YftWqVRw+fJgFCxaY90VGRuLn55elnJ+fH5GRkebtq1evZpsgGQwG5s+fT/369XMe9L+6du3Kyy+/DMCkSZP46quvaNq0KX369AFg/PjxBAUFERUVhb+/P7Nnz6ZDhw689957ADz++OOcO3eOWbNmMXToUHO97du358033zRvX7t2jWrVqvHkk0+i0WioUKHCfbGULVuWq1ev5voahCiRoqOhUyc4dQp8fCAsDBo2tHVUQuSa9CAVY+3atePEiRNZXt999122ZXft2sXzzz/Pt99+m+uxNqmpqTg5Od2338HBgXr16uUp9v8el5mg1a1b97590dHRAPzxxx+0bNkySx0tW7bkr7/+ypIU3pvIDR06lBMnTlC9enXeeOMNtm3bdl8szs7OpKSk5Ok6hChRbt6Etm3V5MjPD3bvluRIFFnSg5RLLi5qT46tzp0brq6uVK1aNcu+f/75575yu3fvplu3bsyZM4fBgwdn+czf35+oqKgs+zJ7bTKVLl2a2NjY++p1dnZGc88jrFqt1nxbLJPBYLjvWHt7e/P7zDqy22cyme479mFc7+mCa9SoEZcvX2bLli1s376dvn370rFjR9asWWMuc/fuXcqUKZOr8whR4ly7Bh06wMWLUK4c7NgBjz9u66iEyDOb9iB99dVX1KtXDw8PDzw8PAgKCmLLli3mz9u2bYtGo8nyeuWVV7LUce3aNUJDQ3FxccHX15dx48aRYcV5ETQa9TaXLV7WmC4jPDyc0NBQPv74Y1566aX7Pg8KCrrvEfewsDCCgoLM2w0bNuTcuXM5Ol+ZMmW4deuWedtoNHLmzJk8Rv//atasye+//55l3++//87jjz+OTqd76LEeHh7069ePb7/9lp9++omff/6Zu3fvmj8/c+YMDeW3YCEe7O+/oXVrNTmqWBH27JHkSBR5Nu1BKleuHDNmzKBatWooisLixYvp3r07x48fN9/mGT58OFOnTjUf4/KfbhSj0UhoaCj+/v7s37+fW7duMXjwYOzt7fnoo48K/HqKml27dvHUU08xatQoevXqZR5X5ODgYB6oPWrUKNq0acOnn35KaGgoK1eu5MiRI3zzzTfmejp37syLL7740PFNmdq3b8/YsWP59ddfqVKlCrNnzyYuLi7f1/Lmm2/StGlTPvjgA/r160dERARz585l/vz5Dz1u9uzZBAQE0LBhQ7RaLatXr8bf3x8vLy9zmb179/LBBx/kO0YhiqULF9Seoxs3oFo1tecoMNDWUQmRbzbtQerWrRtdu3alWrVqPP7443z44Ye4ublx4MABcxkXFxf8/f3Nr//O0rxt2zbOnTvHjz/+SIMGDejSpQsffPAB8+bNQ6/X2+KSipTFixeTkpLC9OnTCQgIML969uxpLtOiRQuWL1/ON998Q/369VmzZg3r1q0zz4EE0KVLF+zs7Ni+ffsjz/nCCy8wZMgQBg8eTJs2bahcuTLt2rXL97U0atSIVatWsXLlSurUqcOkSZOYOnVqlgHa2XF3d2fmzJk0adKEpk2bcuXKFTZv3myerykiIoL4+Hh69+6d7xiFKHbOnIE2bdTkqFYtdcyRJEeimNAo9w4IsRGj0cjq1asZMmQIx48fp1atWrRt25azZ8+iKAr+/v5069aN9957z9yLNGnSJDZs2MCJEyfM9Vy+fJnKlStz7NixB94WSU9PJz093bydkJBAYGAgt27dwsfHJ0vZtLQ0rl+/TsWKFbMdiCxU8+fPNz8u7+7uft/Yo6Kqf//+1K9fnwkTJjy0XFpaGleuXCEwMNAi3xODwUBYWBidOnXKMvZK5J60pWXc147Hj2PXtSuamBiUevXI2LIFZKxejsh30nJiYmIICAggPj7e4stc2XyQ9unTpwkKCiItLQ03NzfWrl1LrVq1ABg4cCAVKlSgbNmynDp1ivHjx3PhwgV++eUX4MGPoGd+9iDTp09nypQp9+3ftWtXllt4AHZ2dvj7+5OUlCS9Ug/Rv39/IiMjSUxMtHUoFpO5FtsLL7xAQkLCI8umpqayZ88ei46BCwsLs1hdJZ20pWWEhYVR6s8/aT5lCprkZGKrVSNi3DgMhw/bOrQiR76T+WfNJ4xt3oOk1+u5du0a8fHxrFmzhu+++47du3ebk6T/2rlzJx06dODixYtUqVKFl156iatXr/Lbf9b4SUlJwdXVlc2bN9OlS5dszyk9SNahKAqJiYnFqgcpp6QHqfCStrSMzHbs7OqKY8+eaBITMQUFYdywATw9bR1ekSLfScsp1j1IDg4O5kfRGzduzOHDh/n888+zTFaYqVmzZgDmBMnf359Dhw5lKZP5SPp/H0O/l6OjI46Ojvftt7e3v+/L+t81xArTOmKFUeYj94VtzbWCoNVq0Wg02X6H8sPS9ZVk0pb5V/rkSRw//hhNSgq0a4d2wwa0tpo5txiQ72T+WbP9Ct1PMZPJlKV3578yxxoF/LvYYVBQEKdPnzZPFghql6WHh0e2PVBCCCHyRrNlC82nTVOTo86d4ddfbbesgBAFwKY9SBMmTKBLly6UL1+exMREli9fTnh4OL/99huXLl1i+fLldO3aFR8fH06dOsWYMWNo3bq1eZbl4OBgatWqxXPPPcfMmTOJjIxk4sSJjBgxItseIiGEEHmwdi26fv3QGAyYnnoK7Zo1IP/HimLOpglSdHQ0gwcP5tatW3h6elKvXj1+++03OnXqxPXr19m+fTufffYZycnJBAYG0qtXLyZOnGg+XqfTsWnTJl599VWCgoJwdXVlyJAhWeZNEkIIkQ8rV8Kzz6IxGrnRogW+P/2EVpIjUQLYNEFauHDhAz8LDAxk9+7dj6yjQoUKbN682ZJhCSGEAFi8GF54AUwmTIMGcbRnT7rImBlRQhS6MUhCCCEKgQULYOhQMJlg+HCMCxeiPGLZHiGKE0mQhBBCZPX555C57uXIkWqyVMKeTBVCvvFCCCH+34wZMHq0+v6tt9RkqYTNayYESIJU7EVERKDT6QgNDb3vs8OHD9OhQwe8vLwoVaoUnTt35uTJk1nKnDp1ilatWuHk5ERgYCAzZ868r57Vq1dTo0YNXFxcaNGihYwJE6IoUhSYPBkyl9V5/301WZLkSJRQkiAVcwsXLmTkyJHs2bOHmzdvmvcnJSUREhJC+fLlOXjwIPv27cPd3Z3OnTtjMBgAdYbx4OBgKlSowNGjR5k1axaTJ0/mm2++Mdezf/9+BgwYwLBhwzh69CihoaH07NmTM2fOFPi1CiHySFHg7bchcwmm6dPVZEmSI1GCSYJUjCUlJfHTTz/x6quvEhoayqJFi8yfnT9/nrt37zJ16lSqV69O7dq1ef/994mKiuLq1asALFu2DL1ez/fff0/t2rXp378/b7zxBrNnzzbX8/nnnxMSEsK4ceOoWbMm7777Lo0aNWLu3LkFfblCiLxQFPWWWmbv8Jw5arIkRAknCVKuKUCyjV65WzZv1apV1KhRg+rVq/Pss8/y/fffk7n0XvXq1fHx8WHhwoXmhVYXLlxIzZo1qVixIqDenmvdujUODg7mOjt37syFCxeIjY01l+nYsWOW8wYHBxMREZGrWIUQNmAywcsvwxdfqNtfffX/44+EKOEkQcq1FMDNRq/crVq8cOFCnn32WQBCQkKIj483zy3l7u5OeHg4P/74I87Ozri5ubF161a2bNmCnZ06PVZkZCR+fn5Z6szcjoyMfGiZzM+FEIVURob6GP+336pPqP3ww/8/uSaEkASpuLpw4QKHDh1iwIABANjZ2dGvXz/z5JypqakMGzaMli1bcuDAAX7//Xfq1KlDaGgoqamptgxdCGFtBgMMHAhLl4JOB8uWqcmSEMLMpjNpF00uQJINz50zCxcuJCMjg7Jly5r3KYqCo6Mjc+fOZc2aNVy5coWIiAi0/85vsnz5ckqVKsX69evp378//v7+REVFZak3c9vf39/8Z3ZlMj8XQhQy6enQrx+sXw/29vDTT/DMM7aOSohCRxKkXNMArrYO4qEyMjJYsmQJn376KcHBwVk+69GjBytWrMBgMKDVatH85ymVzG2TyQRAUFAQ7777LgaDAft/lxcICwujevXqlCpVylxmx44djP7PuIXt27cTFBRk5asUohCI/wPS79g6ipxLT4eJ78L5Q1DHHj6cBs1LQ/TeRx6qycjA23gWzW0PsJMfHfkhbWk5mth4q9UtfzPF0KZNm4iNjWXYsGF4enpm+axXr14sXLiQpUuXMm7cOEaMGMHIkSMxmUzMmDEDOzs72rVrB8DAgQOZMmUKw4YNY/z48Zw5c4bPP/+cOXPmmOsbNWoUbdq04dNPP6VLly4sWbKEI0eOZJkKQIhix5AEe3tB5DZbR5J7nf99YYCk8bA9Z4fZAa0Awq0UVwkibWk5drkbmpu7uq1XtbCVhQsX0rFjx/uSI1ATpJkzZ6LX69m4cSNTpkwhKCgIrVZLw4YN2bp1KwEBAQB4enqybds2RowYQePGjSldujSTJk3ipZdeMtfXokULli9fzsSJE3nnnXeoXLkyv/zyC3Xq1Cmw6xWiQJkMsK+vmhxpdOBWGTSFfDin0QT//AOpqeqA7HLlwMU5V1UoikJSUjJubq5Zep5F7klbWo6iywAuWaVuSZCKoY0bNz7wsyeeeML8qD9Ap06dHlpXvXr12Lv34d3vffr0oU+fPphMJhISEvDw8MhdwEIUFYoCh16BW1tA5wwddkLp5raO6uHu3oWQEDj8F3h5wdat0KxZrqvJMBjYuXkzXUO6mm+5i7yRtrScjJgYoLRV6pYESQghcur0ZPj7e7XH6MlVhT85un0bOnWCkyfBxwfCwqBhQ1tHJUSRIAmSEELkxMVv4cxU9X3Tr+Gxp2wbz6PcugUdO8K5c+DnB9u3g9z6FiLHJEESQohHufErHH5VfV/nPag63LbxPMr169ChA/z1Fzz2GOzYAdWr2zoqIYoUSZCEEOJhYg6rg7IVI1R+HupOsXVED3f5MrRvD1euQIUKsHMnVK5s66iEKHIK+aMXQghhQ4kXITwUjCkQEAJPLCjcK9z/+Se0aqUmR1Wrwp49khwJkUeSIAkhRHbSomFXCKTfBu/G8ORq0BbiJ47OnoXWreHGDahZE3bvhvLlbR2VEEWWJEhCCHGvjGQIfwqSLoFrJWjzK9i72TqqBztxAtq2hagoqFcPwsPhP8sMCSFyTxIkIYT4L1MG7OsHdw+Dow+02wrOfraO6sEOHYJ27eDOHWjSBHbtAl9fW0clRJEnCZIQQmRSFDj8Gtz8FXRO0HojeDxu66gebN8+9VH+uDho0UJ9lN/b29ZRCVEsSIIkhBCZzkyDS9+qE0G2XAllCvGiyzt3QufOkJio3l777TfIZnkhIUTeSIJUTA0dOpQePXo8tIyiKHTp0gWNRsO6deuyfHbt2jVCQ0NxcXHB19eXcePGkZGRYf7caDQyY8YMatSogbOzM97e3gQFBbFkyRIrXI0QBeDS93B6kvq+yTwo19228TzM1q0QGgopKRAcDL/+Cm6FeIyUEEWQzINUgn322WfZLpRoNBoJDQ3F39+f/fv3c+vWLQYPHoy9vT0fffQRAFOmTGHBggXMnTuXJk2akJCQwKFDh4iMjCzoyxAi/25ugUP/LsJc+x2o9opt43mY9euhTx8wGKBbN1i9GhwdbR2VEMWOJEgl1IkTJ/j00085cuQIAQEBWT7btm0b586dY/v27fj5+dGgQQM++OADxo8fz+TJk3FwcGDDhg289tpr9OnTx3xc3bp1SUhIKOhLESJ/Yo7Avj7qRJCVBkO9abaO6MFWrYJBgyAjA3r3hmXLwMHB1lEJUSxJgpRbiqJ2a9uCi4tFJqlLSUlh4MCBzJs3D39///s+j4iIoG7duvj5/f+TO507d+bVV1/l7NmzNGzYEH9/f3bu3Mlrr71GmTJl8h2TLaTFp6EYo9Dap+e7Lr3eSEZaFDeOfICdcjvf9SkKVHVI4kbE3EI9L2FR8LC2tMtIwT/qCDqTnlRHb6IN/8C+YNsE+giuW2/iM+McGhMkBfsT89pdOBhaYOeX76TlSFtaTkJSxqML5ZEkSLmVkmK7e/1JSeDqmu9qxowZQ4sWLejePfsxFpGRkVmSI8C8nXkLbfbs2fTu3Rt/f39q165NixYt6NatGy1btsx3fAXBlGEiNeYmpSrftUh9ShrYOaXxWM39ODldtUidwsrSgDDABJQC5453qWC/08ZBPcA3wEf/vn8R3L6OxE0nt7OFsOZNC0mQSpgNGzawc+dOjh8/nq96atWqxZkzZzh69Ci///47e/bsoXv37gwcOJBFixZZJlgrSrmTjPtj8QAYUt0xGf4dw5HH3+bUHqQ0bhzqi84Ul+/4TCYTd+7coXTp0mi18ixFfmTXlhpTBv4xv+JoiCZD58Yth+4Yf3excaTZc//5DD5zIwBIeKYWdwe0gL0F3+0g30nLkba0nPjENGCpVeqWBCm3XFzUnhxbnTufdu7cyaVLl/Dy8sqyv1evXrRq1Yrw8HD8/f05dOhQls+joqIAstyS02q1NG3alKZNmzJ69GiWLFnCkCFDeP/996lSpUq+Y7UWxaSQkX4HO0cjJpMd9s7VwDl//0kpaWnYOel5rOZUnJyc8h2jwWDg9ObNNGzVFXv7Qry8RRFwX1uaMmBvLzBEg4M3dp1+J9Czhq3DzN7HH8Pcb9X348bh8fHHeNjonox8Jy1H2tJyYmJikASpsNBoLHKby1befvttXnzxxSz76taty5w5c+jWrRsAQUFBfPjhh0RHR+P774y8YWFheHh4UKtWrQfWnflZcnKylaK3jLS4VFx81H5ZDWWQ2S5KEEWBIyPhxgZ1Isg2G6AwJkeKAlOnwuTJ6vakSep7GbAiRIGRBKkYi4+P58SJE1n2+fj4UKdOnfvKli9fnkqVKgEQHBxMrVq1eO6555g5cyaRkZFMnDiRESNG4Pjv48S9e/emZcuWtGjRAn9/fy5fvsyECROoWrUqNWoUwh84/5GecBevigYURYNGK0sylCjnpsPFrwENtFgOZQrhmDlFgQkT1N4jgI8+UreFEAVKEqRiLDw8nIYNG2bZN2zYML777ruHHqfT6di0aROvvvoqQUFBuLq6MmTIEKZOnWou07lzZ1asWMH06dOJj4/H39+fdu3asWDBAuzsCu/XSp+kx9EjDgBF8Uajke7tkkJzZQmcfFfdaPwFBD5j24CyoygwejR88YW6PWeOui2EKHCF9yeZyJdFixbleLC0oij37atQoQKbN29+4DHDhw9n+PDhWfaZTKZCPw9S6t04PALTANBqC/ECpMKiymQcR3fkQ3Wj1nio/rptA8qOyQSvvQYLFqjb8+fDq6/aNiYhSjCbDr746quvqFevHh4eHnh4eBAUFMSWLVvMn6elpTFixAh8fHxwc3OjV69e5sHCmR61JIYQmTL0GegcY9BowGR0AwrnU0vCwmKP80T6x2iUDKg4COp/9OhjCprRCC+8oCZHGg18/70kR0LYmE0TpHLlyjFjxgyOHj3KkSNHaN++Pd27d+fs2bOAOl/Pxo0bWb16Nbt37+bmzZv07NnTfHzmkhh6vZ79+/ezePFiFi1axKRJk2x1SaIQS7mdhIuPOsmnVnf/BJmiGEq6jN2+7tiRhsm3PTT7Xl2ItjAxGNTZsRcvBp1OnR37+edtHZUQJZ5Nb7FlPjWV6cMPP+Srr77iwIEDlCtXjoULF7J8+XLat28PwA8//EDNmjU5cOAAzZs3z9GSGEIAmIwmUG6jtVMwGR3R6mTV82IvPQbCu6BJiyReWxGXFj+h1RWy/xPS06F/f1i3DuztYeVK+M8vgUII2yk0v0oZjUZWrlxJcnIyQUFBHD16FIPBQMeOHc1latSoQfny5YmIUCdNe9CSGAkJCeZeKCEAUmNScCmtzl+l0fqR5xkhRdGQkQq7u0HCBRSX8hxwnAT2hSwpTk2FHj3U5MjREdauleRIiELE5oO0T58+TVBQEGlpabi5ubF27Vpq1arFiRMncHBwuG9CQz8/P/NyFzlZEiM76enppKf///pbmQOLDQYDBoMhS1mDwYCiKJhMJkwmU56vsyTIHOyd2V6FiSH1Dq6+RkwmLeCNolg2PpPJhKIoGAwGdDpdvuvL/B7e+30UOaAY0e3vj/ZOBIq9F2nNfyHt4D+Fqy2Tk9H17Il21y4UZ2eMv/yC0qGDerutkJLvpOVIW1qONdvQ5glS9erVOXHiBPHx8axZs4YhQ4awe/duq55z+vTpTJky5b79u3btwuWe2art7Ozw9/cnKSkJvV5v1biKi8TERFuHkIUx2Yibt5oE69M9SEu3/Ezoer2e1NRU9uzZY9GHBMLCwixWV4mgKNTVf0vljM0YsWe/bhx3D/4DFJ62tEtJofkHH+Dzxx9kODkRMXEid9PT4SFPjRYmhaUdiwNpy/xLseLi8TZPkBwcHKhatSoAjRs35vDhw3z++ef069cPvV5PXFxcll6kqKgo83IXOV0S414TJkxg7Nix5u2EhAQCAwNp164dPj4+WcqmpaVx/fp13NzcLLKERHGmKAqJiYm4u7ujKUQz/ibER+LorkdRwMExEAdHy899lJaWhrOzM61bt7bYUiNhYWF06tRJliLIBe35WehOb0ZBgxK0hOblehWutoyNRffUU2j/+APF0xM2baJ5s2a2jSmHClU7FnHSlpajLjViHTZPkO5lMplIT0+ncePG2Nvbs2PHDnr16gXAhQsXuHbtGkFBQUDel8RwdHQ0zwj9X/b29vd9WY1GIxqNBq1WK4sKPkLmbbXM9ioMDKkGHFxjAVBMpdDq7v97twStVotGo8n2O5Qflq6vWLu8FE6rE0FqGs3BrlL/LB/bvC1v34bOneHECfDxQbNtG3aNGtkunjyyeTsWI9KW+WfN9rNpgjRhwgS6dOlC+fLlSUxMZPny5YSHh/Pbb7/h6enJsGHDGDt2LN7e3nh4eDBy5EiCgoJo3rw5kLMlMUTJlnInHo/HUgF5tL9Yi9wOB15Q39f8H9QYZdt47hUZCR06wLlz4OsLO3ZANkv+CCEKD5smSNHR0QwePJhbt27h6elJvXr1+O233+jUqRMAc+bMQavV0qtXL9LT0+ncuTPz5883H5+TJTFEyWU0GNHaxaDRginDBa1d0V1kWDxE7AnY0xOUDKjQHxp8bOuIsvrnHzU5+vNPKFtWTY4K+XqFQggbP+a/cOFCrly5Qnp6OtHR0Wzfvt2cHAE4OTkxb9487t69S3JyMr/88st9Y4syl8RISUnh9u3bfPLJJ4V6LbCCMnToUHr06PHQMoqi0KVLFzQaDevWrcvy2aNmKDcajcyYMYMaNWrg7OyMt7c3QUFBLFmyxApXkzcpd5JxKZ0MgEZ6j4qn5KsQ3hUyEsGvHTRfVLgmgrxyBVq3VpOj8uVhzx5JjoQoIiSTKME+++yzbAdTZ85Q7u/vz/79+7l16xaDBw/G3t6ejz5Sl2mYMmUKCxYsYO7cuTRp0oSEhAQOHTr00OkVCpJiUjBl3EZnb8JktEerK2XrkISlpd+FXV0g9RZ41oFWv4CVxpjlyV9/Qfv2ag9SlSqwc6eaJAkhigRJkEqoEydO8Omnn3LkyBECAgKyfJaTGco3bNjAa6+9Rp8+fczH1a1bt9AsVpsam4pLaXW6AZkYshgypsGe7pDwB7iUg3ZbwMHL1lH9v3Pn1NtqkZFqj9GOHertNSFEkVGI+qKLBkVR0Ov1NnllTsSYXykpKQwcOJB58+ZlOx1CTmYo9/f3Z+fOndy+fdsiMVmSoigYku9g75yBYtKi0ZS2dUjCkkxG2P8s3N6nzo7ddquaJBUWJ05AmzZqclS3LuzeLcmREEWQ9CDlksFgYPr06TY594QJEyyyvtyYMWNo0aIF3bt3z/bznMxQPnv2bHr37o2/vz+1a9emRYsWdOvWjZYtW+Y7vvzSJ+lx9IwHQMEHjXzNiw9FgWNj4frPoHWA1uvBq7ato/p/hw9DcDDExUHjxvDbb3DP3GpCiKJBepBKmA0bNrBz504+++yzfNVTq1Ytzpw5w4EDB3jhhReIjo6me/fuvPHGG5YJNB/SYmNx8kxHUUCrlcHZxcr5T+HPL9T3QUvAr41t4/mv339Xb6vFxUFQkHpbTZIjIYos+dU6l+zt7ZkwYYLNzp1fO3fu5NKlS/etcderVy9atWpFeHh4jmco12q1NG3alKZNmzJ69GiWLFnCkCFDeP/996lSpUq+Y82LjPQM7JzvAqCYPNEUpkG7In+uLIfj49T3DT+FCv1sG89/7doF3bpBcrJ6e23jRnB3t3VUQoh8kAQplzQajUVuc9nK22+/zYsvvphlX926dZkzZw7dunUD8j5DeeZnycnJVor+0VLuJOAWIBNDFjuRO+HAUPV99TFQc+xDixeorVvhmWcgLU29vbZ2LdyzpqMQouiRBKkYi4+P58SJE1n2+fj4UCebGXzLly9PpUqVgJzNUN67d29atmxJixYt8Pf35/Lly0yYMIGqVatSw0bzvJiMJjSaO2i1CqYMZ7R2bjaJQ1hY7CnY+wyYDFC+LzT6xNYR/b/166FvX9Dr1R6kVatA1mwUoliQBKkYCw8Pp2HDhln2DRs2jO++++6hx+VkhvLOnTuzYsUKpk+fTnx8PP7+/rRr144FCxbYbKLO1JhkXEonAZkTQ8qj/UVe8nV1IkhDAvi2hqDFhWciyNWrYeBAyMiAXr1g+XIowr3LQoisJEEqphYtWsSiRYtyVDa76QMyZyh/kOHDhzN8+PAs+0wmk83mQVIUhYz02+gcTJiMdjIxZHGgj4XwLpB6AzxrQ+t1oCskvTNLl8LQoWAywaBBsGgRyAz+QhQrheRXMSHyJy0+DWfvfyeG1PgiX+0izpgGe3pA/FlwLgttt4BDIUl6v/0WhgxRk6Nhw2DxYkmOhCiG5KeIKBb0CTE4uBpQTBo0Wl9bhyPyQzFBxBCI3gP2Hmpy5Bpo66hUc+fCSy+p8zG99hp88w3odLaOSghhBZIgiSLPkGLAwT0OAEXxQe4cF3HH/gfXVoHWXr2tVqqerSNSzZoFI0eq7998U02WtPJfqBDFlfzrFkVeSkwsTl5pAGh1fo8oLQq183Pgwhz1ffPF4NfOtvGA2ls0dSq89Za6PXGimixls9CzEKL4kF+1RZFmNBixc4xBowFThjtaO2dbhyTy6upP6jIiAA1mQsUBto0H1OTonXdgxgx1e9o0ePdd28YkhCgQkiCJIi3lTiKuvikAaO1kYsgiKyocIgar7x8fCTX/Z9NwADU5GjMGPv9c3Z49W90WQpQIkiCJIksxKSim22h1CqYMR7R2HrYOSeRF3Bn1iTWTHgJ7QaM5tr99ZTKpg7AXLFC358+HV1+1bUxCiAIlCZIoslLvpsjEkEVdyj/qXEeGeCjzJLT4EbQ2firMaIQXX1TnNtJo4Lvv4IUXbBuTEKLASYIkiiRFUTCk3saltBGTUYdWJ6umFzn6ONjVRU2SPGpC6/W2nwjSYIDBg2HlSvXx/SVL1NmyhRAljiRIokjSJ+pxKvXvrN2aMsgDmUWMMR32PAPxZ8A5ANptAUdv28aUng4DBqiLzdrbw4oV6hIiQogSSX6qlBAzZsxAo9EwevRo8760tDRGjBiBj48Pbm5u9OrVi6ioqCzH7dixgxYtWuDu7o6/vz/jx48nIyMjS5lTp07RqlUrXFxcqF27NrNmzbL69aTFx+DopkdRNGhlYsiiRTHBgaEQHQ527tB2M7hWsG1MaWnQs6eaHDk4wC+/SHIkRAknCVIJcPjwYRYsWEC9elkn3BszZgwbN25k9erV7N69m5s3b9KzZ0/z5ydPnqRr166EhIRw/PhxfvrpJzZs2MDbb79tLpOQkEBwcDAVKlTg8OHDTJ06lSlTpvDNN99Y7Xoy0jKwd4kFQDF6AbJAaJFyYjxcXQkaO2j9C5RqYNt4kpPhqadg82ZwdoZNm9RtIUSJJrfYcktRwJhim3PrXHL9dE9SUhKDBg3i22+/Zdq0aeb98fHxLFy4kOXLl9O+fXsAfvjhB2rWrMmBAwdo3rw5P/30E/Xq1WPSpEkAVK1alZkzZ9K3b1/ef/993N3dWbZsGXq9nu+//x47OzsCAwP5888/mT17Ni+99JLlrv0/UmLicC+bCoDWLsAq5xBWcuEL+OMT9X3zH8C/o23jSUhQk6G9e8HNTU2O2rSxbUxCiEJBEqTcMqbAKjfbnLtvEti55uqQESNGEBoaSseOHbMkSEePHsVgMNCx4///gKpRowbly5cnIiKC5s2bk56ejpNT1kGzzs7OpKWlcfToUdq2bUtERAStW7fGwcEBk8kEQHBwMDNnziQ2NpZSpSy7wKgpw4RWd+ffiSFd0dq5WLR+YUXX1sDR0er7+tOh0rM2DYfYWAgJgUOHwMMDtm6FoCDbxiSEKDTkFlsxtnLlSo4dO8b06dPv+ywyMhIHBwe8vLyy7Pfz8yMyMhKAzp07s3//flasWIHRaOTGjRtMnToVgFu3bpnr8fPzu6+OzM8sLSUmCZfSyQBodNJ7VGRE74X9zwIKVHsNao23bTx37kD79mpy5O0NO3dKciSEyEJ6kHJL56L25Njq3Dl0/fp1Ro0aRVhY2H29QDkVHBzMrFmzeOWVV3juuedwdHTkvffeY+/evWhtsEinoiiYMm6jtVMwGR3Q6jwLPAaRB/HnYPfTYEqHcj2g8Re2nQgyMhI6doSzZ8HXF7Zvh7p1bRePEKJQkh6k3NJo1Ntctnjl4ofK0aNHiY6OplGjRtjZ2WFnZ8fu3bv54osvsLOzw8/PD71eT1xcXJbjoqKi8Pf//yU7xo4dS1xcHNeuXePOnTt0794dgMqVKwPg7+9/35Nvmdv/rccS0mLTcPFJBECj8UMmhiwCUm6qcx0Z4qB0C2ix3LYTQf7zjzrG6OxZKFsWdu+W5EgIkS1JkIqpDh06cPr0aU6cOGF+NWnShEGDBpnf29vbs2PHDvMxFy5c4Nq1awTdc6tBo9FQtmxZnJ2dWbFiBYGBgTRq1AiAoKAg9uzZg8FgMJffvn071atXt/j4I0PKbeycMjCZtGi0pS1at7ACfbw6S3bKNfCoDm02gC0XE75yBVq3hj//hPLlYc8eqFHDdvEIIQo1ucVWTLm7u1OnTp0s+1xdXfHx8THvHzZsGGPHjsXb2xsPDw9GjhxJUFAQzZs3Nx8za9YsQkJC0Gq1/PLLL8yYMYNVq1ah06m9AAMHDmTKlCkMGzaMcePGcfjwYb744gvmzJlj0evRJ+tx9IhXN5TSgI2XoxAPZ9TD3l4Qdwqc/KDtFnC03WznrjdvYjdyJFy/DlWqwI4dUMHGcy8JIQo1SZBKsDlz5qDVaunVqxfp6el07tyZ+fPnZymzZcsWPvzwQ9LT06lfvz7r16+nS5cu5s89PT3Ztm0bI0aMoGnTpvj4+PDee+9Z/BH/tNi7eJRLR1FAq/N79AHCdhQTHHwBonaAnZs6EaRbJdvFc+4cT777LprYWLXHaPt2eOwx28UjhCgSJEEqQcLDw7NsOzk5MW/ePObNm/fAY3bu3PnIeuvVq8fevXsxmUwkJCTg4eGR31CzMOqN2DndBUAxeqKxc7Ro/cLCTr4DV5apE0G2+hm8G9kwlpPYdeqEfWwsSp06aLZvBz9JsIUQjyZjkEShlxITj7O3OjmnTAxZyP05D859rL5v9h0EBNsuliNHoF07NLdvE1e5MhlhYZIcCSFyTBIkUaiZjCY03EGjBVOGM2CjSTrFo11fC0dGqu/rTYPKQ2wXy/790KEDxMZiataM36dOBR/bjYESQhQ9kiCJQi31bgrOpdV5pzQ6y04bICzo9u+wfyCgQNWXofY7toslPByCg9VlRFq3xrh5MxluklgLIXJHEiRRaCmKglEfjc7ehMloj0bjbeuQRHbiz8PubmBMg8eehiZzbTcR5G+/QZcu6gK0nTrBli3g7m6bWIQQRZpNE6Tp06fTtGlT3N3d8fX1pUePHly4cCFLmbZt26LRaLK8XnnllSxlrl27RmhoKC4uLvj6+jJu3DgyMjIsGquiKBatTzxaekI6zqUSANBofCnME0OW2O9H6i0IDwF9LPg0h5YrQGujZz82bICnn4a0NHUB2g0bwEXW6hNC5I1NE6Tdu3czYsQIDhw4QFhYGAaDgeDgYJKTk7OUGz58OLdu3TK/Zs6caf7MaDQSGhqKXq9n//79LF68mEWLFplXoM+vzPl+9Hq9ReoTOadPvIO9SwaKSYNGW8bW4TxUSoo6iNze3t7GkRQgQyKEh0LyVXCvBm02gq0WD169Gnr1Ar1e/fPnnyGPS+wIIQTY+DH/rVu3ZtletGgRvr6+HD16lNatW5v3u7i4PHDZim3btnHu3Dm2b9+On58fDRo04IMPPmD8+PFMnjwZBweHfMVoZ2eHi4sLt2/fxt7e3iZrkBUVJpMJvV5PWlpavtvJkGZAcbhLWhqYMrzQ2mUAlu0VtARFUUhJSSE6OhovLy9zQl3sGfWwtzfEHgcnX2i3FZxsNLv5jz/CkCFgMsHAgbB4MdjJDCZCiPwpVP+LxMerMyV7e2cda7Js2TJ+/PFH/P396datG++99x4u/3adR0REULdu3Swrynfu3JlXX32Vs2fP0rBhw3zFpNFoCAgI4PLly1y9ejVfdRV3iqKQmpqKs7MzmnyOQUmLS8TJ6y7cATQOwGWLxGgtXl5eFl97rtBSFDg0HCK3qWsEtvkV3CrbJpbvvoOXXlJjeuEF+OYbKClJqhDCqgpNgmQymRg9ejQtW7bMskTGwIEDqVChAmXLluXUqVOMHz+eCxcu8MsvvwAQGRmZJTkCzNuRkZHZnis9PZ309HTzdkKCOs7FYDBkWVMsk0ajoWLFihgMhpI71iQHMjIy2L9/Py1atMAuH7/Bp8Wlce3QeCr1O0XK3ebYu39rwSgtS6PRYGdnh06ns+i4t8zvYXbfR1vTnn4P3eUlKBodxuYrUDzqgw3i1H71FbpRowAwvvwyps8/V3uRTKYs5QpzWxYl0o6WI21pOdZsw0KTII0YMYIzZ86wb9++LPv/u2RF3bp1CQgIoEOHDly6dIkqVark6VzTp09nypQp9+3ftWuXuWdK5N2ePXvydXzclr8Z8OlWHJwMHPl7GDExj57Nu7gKCwuzdQhZVDRspb7+awBO2L/GteMmOL65wOOosm4ddRYtAuDi009zNiQE7rllf6/C1pZFlbSj5Uhb5l/m+E9rKBQJ0uuvv86mTZvYs2cP5cqVe2jZZs2aAXDx4kWqVKmCv78/hw4dylImKioK4IG3PCZMmMDYsWPN2wkJCQQGBtKuXTt8ZDK5PDMYDISFhdGpU6c8D1Y2GowcOtMLB1cDKTGP06zZeArz02vWYom2tDTNjQ3o9n8DgLH2JOrUmkidRxxjDdqPPkL3b3JkfPttKkyZQoWH3NItjG1ZFEk7Wo60peXExMRYrW6bJkiKojBy5EjWrl1LeHg4lSo9ekHLEydOABAQoC45ERQUxIcffkh0dDS+vr6AmpV7eHhQq1atbOtwdHTE0fH+9bzs7e3ly2oB+WnH8z+fpsGQvQA4eryFzj5/g+yLukLznbwdAQefBUxQZTi6epPRFfRcR4oCEyfCRx+p2x98gG7iRHI64qjQtGURJ+1oOdKW+WfN9rNpgjRixAiWL1/O+vXrcXd3N48Z8vT0xNnZmUuXLrF8+XK6du2Kj48Pp06dYsyYMbRu3Zp69eoBEBwcTK1atXjuueeYOXMmkZGRTJw4kREjRmSbBInCS1EUbp/5irr9E9EnlcLB7VlbhyQAEv6EPf9OBFk2FJrOL/iJIBUF3nwT5sxRtz/5RN0WQggrsekz61999RXx8fG0bduWgIAA8+unn34CwMHBge3btxMcHEyNGjV488036dWrFxs3bjTXodPp2LRpEzqdjqCgIJ599lkGDx7M1KlTbXVZIo/+ibhO9aczx7O8BkiCa3OpkbArBNJjwLspPPlTwU8EaTLBa6/9f3I0d64kR0IIq7P5LbaHCQwMZPfu3Y+sp0KFCmzeXPADRYVl/bVlMe0/uInRYI+D2xu2DkcYkmD3U5B8GdyqQNtN6mP9BclohBdfhEWL1F6rb7+FYcMKNgYhRIkksx6KQiHuahz+DVYCoE/sA/jaNqCSzmSAfX3g7lFwLPPvRJAF/HdiMMBzz6nJkU4HS5dKciSEKDCSIIlC4dSyDdTo8QcAzt4TbBxNCacocOgluLUVdC7QZhO4Vy3YGPR66NcPVqxQZ8X+6ScYNKhgYxBClGiSIAmb0yfpcfL4Fq1OIfl2C7DJw+PC7PT78Pci0GjVMUelnyjY86elQc+esHYtODjAL7+o66sJIUQBKhTzIImS7dTy36n33GEAXHzesXE0JdzFb+DMB+r7pl/DY08V7PmTk6FHD9i+HZydYd06CA4u2BiEEALpQRI2ppgUUqLn4uSZTmpsBTTaLrYOqeS6sQkOv6q+rzMJqg4v2PMnJkKXLmpy5OoKW7ZIciSEsBnpQRI29dfm89QdqC4lYu/8PyRnt5E7B2FfX1BMUPkFqDu5YM8fF6cmRwcOgIeHmhy1aFGwMQghxH/ITyNhUzcOfU2pynEYUtyxc3rB1uGUTAl/qY/zG1MhIASe+LpgJ4KMiYEOHdTkqFQp2LFDkiMhhM1JD5KwmahTUVTqsB4Ao+FF7JGFggtcWjSEd4H0O+DdGJ5cDdoCXPogKgo6doQzZ6BMGfX22r+z5AshhC1JD5KwmT/WLqNim6uYMnQ4ecrMyAUuIxnCn4KkS+BWGdr8CvZuBXf+GzegTRs1OQoIgN27JTkSQhQakiAJm0iOTsan2hIAUmO7AY/ZNqCSxpShjjm6exgcfaDtFnD2K7jzX70KrVvDhQtQvjzs2QM1axbc+YUQ4hEkQRI2cerHrdTqcxoAl9LyaH+BUhQ4/Arc3Aw6Z3UiSI/HC+78Fy9Cq1bw999QubKaHFUt4IkohRDiESRBEgUuIz0DjXY+OnsTybcbodE0tXVIJcuZqXBpoToRZMuVULp5wZ37jz/UnqPr16F6dTU5qlCh4M4vhBA5JAmSKHDn1hyh3rMRADiVkmVFCtSlhXB6svq+yXwo93TBnfvUKXXM0a1bUKeOOuboMbm1KoQonCRBEgVKURTi/p6LS+lU0uIC0Nk9Y+uQSo4bm+HQy+r72u9CtZcL7txHj0K7dnD7NjRsCLt2gV8BjnkSQohckgRJFKire65Qs9dvAGjtRwM6m8ZTYsQchn19QDFCpSFQ74OCO3dEBLRvD3fvQrNmsHMnlC5dcOcXQog8kARJFKiru7+hTK07GFKdcXB9xdbhlAyJlyA8FIwp4B8Mzb4tuIkgd++GTp0gIUEdmB0WBl5eBXNuIYTIB0mQRIG5e+kujzVbDYAhZTDgYduASoK027ArBNJvQ6mG0GpNwU0EuW2bunxIcrI6GeSWLeDuXjDnFkKIfJIESRSYc6tXU7XzJUxGDS4+420dTvGXkQK7u0HSRXCtCG03g30BJSibNkG3bpCaCl27wsaN6gK0QghRREiCJApEekI6rn4LAUi5EwxUsm1AxZ0pA37vDzEHwcEb2m0FZ/+COfeaNfDMM6DXq3+uXQtOTgVzbiGEsBBJkESBOLVsJ3X6HwfA1VcmhrQqRYEjI+DGRtA5QZuN4FG9YM69bBn06wcZGTBgAPz0Ezg4FMy5hRDCgvKdIBmNRk6cOEFsbKwl4hHFkMlowpAyF3vnDJJv10SjaWXrkIq3sx/CxW8ADbRYDmVaFMx5Fy6E554Dkwmefx6WLgX7Alz4VgghLCjXCdLo0aNZuFC9VWI0GmnTpg2NGjUiMDCQ8PBwS8cnioE/N52m7sA9ADh6jgcK6AmqkujvRXDqPfV9ky8hsIDmmZo3D158Ue29evVV+O470MkUDkKIoivXCdKaNWuoX78+ABs3buTy5cucP3+eMWPG8O6771o8QFH0RZ/5EveAJNISfLBzGGDrcIqvm7/BweHq+1pvw+MjCua8n34Kr7+uvh8zRk2WtHL3XghRtOX6f7E7d+7g768O9ty8eTN9+vTh8ccf54UXXuD06dMWD1AUbTeP3qBal1/VDWUEIONRrOLuUdjXC5QMqPgs1P+oYM47bRr873/q+3feUZOlgppjSQghrCjXCZKfnx/nzp3DaDSydetWOnXqBEBKSgo66VIX97i49QcCGkWSke6Ak+cbtg6neEq6rE4EmZEM/h2h2ULrJymKAhMnwnv/3s774AP48ENJjoQQxYZdbg94/vnn6du3LwEBAWg0Gjp27AjAwYMHqVGjhsUDFEVX4q1E/OouAyA9vi92vj42jqgYSrujTgSZFgVe9aHVz6Czci+doqi9RrNnq9uzZv1/L5IQQhQTuU6QJk+eTJ06dbh+/Tp9+vTB0dERAJ1Ox9tvv23xAEXRdWblOpqPOg+Aq6+MT7O4jBTY8zQk/gku5f+dCNLKs5ObTDByJMyfr25/+eX/jz8SQohiJNcJEkDv3r3v2zdkyJB8ByOKD0OqAQe3BWi0kBjZCnd/6V20KJMR9g+COxHgUEqdCNKlrHXPaTTCSy/B99+rt9K++UZ9ck0IIYqhPD1qsmPHDp566imqVKlClSpVeOqpp9i+fbulYxNF2LnV+6k74BAArmWk98iiFAWOvgH/rAOtI7TeAJ41rXvOjAwYPFhNjrRaWLJEkiMhRLGW6wRp/vz5hISE4O7uzqhRoxg1ahQeHh507dqVefPmWSNGUcQoikJS1Oc4uBlIuVMJrS7Y1iEVL+dmwF/zUSeCXAa+T1r3fHo99O8Py5eDnR2sXAnPPmvdcwohhI3l+hbbRx99xJw5c3j9P+MO3njjDVq2bMlHH33EiBEFNPeKKLSu7LpI7b47AbB3GYdMDGlBl5fCyX+Xamn8GZTvZd3zpaVB797w66/qkiFr1qiL0AohRDGX6x6kuLg4QkJC7tsfHBxMfHy8RYISRVvk0a/wqhCPPskDe5ehtg6n+LgVBgdeUN/XHAfVrTxtQkoKPP20mhw5OcGGDZIcCSFKjFwnSE8//TRr1669b//69et56qmnLBKUKLrSbqRRqe06AIz6lwBnm8ZTbNw9Dnt7qhNBVhgADWZY93yJidClC4SFgasrbNkCnTtb95xCCFGI5PoWW61atfjwww8JDw8nKCgIgAMHDvD777/z5ptv8sUXX5jLvvGGTAxY0uj+OkLga/9gNNjh7P2mrcMpHpKuQHhXyEgCv3bQ/AfQWHEpj7g4NTk6cAA8PNTkqEUBLXgrhBCFRK4TpIULF1KqVCnOnTvHuXPnzPu9vLzMi9gCaDQaSZBKmNS7qdQM2gpAyp3uuAf42ziiYkB/F8K7QFokeNWFVmtB52i988XEQHAwHDsGpUrBtm3QpIn1zieEEIVUrhOky5cvWyMOUQz8sXobTV85C4Cbvzzan19aJR3dvp6QcB5cyqkTQTp4Wu+EUVHQqROcPg1lyqi31/5dmFoIIUqafPXTK4qCoih5Pn769Ok0bdoUd3d3fH196dGjBxcuXMhSJi0tjREjRuDj44Obmxu9evUiKioqS5lr164RGhqKi4sLvr6+jBs3joyMjDzHJXLPaDCi1c5Ha6eQeLMxGk1DW4dUtClGGqfPQRuzH+y9oO1WNUmylhs3oG1bNTkKCIDwcEmOhBAlWp4SpCVLllC3bl2cnZ1xdnamXr16LF26NNf17N69mxEjRnDgwAHCwsIwGAwEBweTnJxsLjNmzBg2btzI6tWr2b17Nzdv3qRnz57mz41GI6Ghoej1evbv38/ixYtZtGgRkyZNysuliTy6sOEodQZEAODgNd7G0RRxioL2xJuUNR5A0TpA63XgVdt657t6Fdq0gfPnITAQdu+GWrWsdz4hhCgKlFz69NNPFRcXF+Wtt95S1q9fr6xfv14ZN26c4uLiosyePTu31WURHR2tAMru3bsVRVGUuLg4xd7eXlm9erW5zB9//KEASkREhKIoirJ582ZFq9UqkZGR5jJfffWV4uHhoaSnp+fovPHx8Qqg3LlzJ1/xl2S/zxqoKApK3M0yil6fZutwirazHyvKMhRlGYrh0jLrnuviRUUpX15RQFEqVVKUy5etez4b0Ov1yrp16xS9Xm/rUIo0aUfLkba0nDt37iiAEh8fb/G6cz0G6csvv+Srr75i8ODB5n1PP/00tWvXZvLkyYwZMybPyVrmPEre3t4AHD16FIPBQMeOHc1latSoQfny5YmIiKB58+ZERERQt25d/Pz8zGU6d+7Mq6++ytmzZ2nY8P5bPenp6aSnp5u3ExISADAYDBgMhjzHX1LdOHid6k+rg7P/vtaN6p5GQNoxLzTXVmB3Qu2BO+3wAlX8e6BY6zt5/jx2ISFobt5EqVaNjN9+g8ceg2L2byDz37T8284faUfLkba0HGu2Ya4TpFu3btEim0d+W7Rowa1bt/IciMlkYvTo0bRs2ZI6deoAEBkZiYODA15eXlnK+vn5ERkZaS7z3+Qo8/PMz7Izffp0pkyZct/+Xbt24eLikudrKKk0+zfx9Md3SUt05sbtTlwLC7N1SEVSaeNJgtI+AOCi3dP8bf80f1upLd2vXKHF++9jHx9PQvny7H/nHdJPnYJTp6xyvsIgTL6XFiHtaDnSlvmXkpJitbpznSBVrVqVVatW8c4772TZ/9NPP1GtWrU8BzJixAjOnDnDvn378lxHTk2YMIGxY8eatxMSEggMDKRdu3b4+PhY/fzFScL1BBLc1V7D9MTnMBqd6dSpE/b29jaOrIiJO4XdrsFoyMAU2Ieyjb7n7PYd1mnLY8ewe+EFNPHxKA0a4Lx5Mx1Kl7bsOQoRg8FAWFiYfC/zSdrRcqQtLScmJsZqdec6QZoyZQr9+vVjz549tGzZEoDff/+dHTt2sGrVqjwF8frrr7Np0yb27NlDuXL//6SOv78/er2euLi4LL1IUVFR+Pv7m8scOnQoS32ZT7lllrmXo6Mjjo73zyVjb28vX9Zc+mvjGpqPuozJqMWlzFvAGWnH3Eq+BvuehowE8G2DtsVS7E3q8xMWb8uICHUSyPh4aNYMzZYt2JcqZbn6CzH5XlqGtKPlSFvmnzXbL9dPsfXq1YuDBw9SunRp1q1bx7p16yhdujSHDh3imWeeyVVdiqLw+uuvs3btWnbu3EmlSpWyfN64cWPs7e3ZsWOHed+FCxe4du2aeRbvoKAgTp8+TXR0tLlMWFgYHh4e1JIncaxKn6zHtcy3ACRFdgbK2zagokgfC7tCIPUmeNZWn1iz1kSQe/aok0DGx8OTT6qTQJaQ5EgIIXIr1z1IoCYuP/74Y75PPmLECJYvX8769etxd3c3jxny9PTE2dkZT09Phg0bxtixY/H29sbDw4ORI0cSFBRE8+bNAXWR3Fq1avHcc88xc+ZMIiMjmThxIiNGjMi2l0hYztnVO6g74AQAbv4TMZpsG0+RY0yD3d0h4Q9wfgzabgEHL+ucKywMuneH1FTo0AHWr1fXWBNCCJGtHCVImU955YSHh0eOy3711VcAtG3bNsv+H374gaFDhwIwZ84ctFotvXr1Ij09nc6dOzN//nxzWZ1Ox6ZNm3j11VcJCgrC1dWVIUOGMHXq1BzHIXJPMSkYkr/AztFI4q06uAe0wGiSJzJyTDFBxGC4vRfsPaDdFnANtM65Nm2C3r0hPR26doU1a8BZFhEWQoiHyVGC5OXlhUajeWgZRVHQaDQYjcYcn1zJwSzcTk5OzJs3j3nz5j2wTIUKFdi8eXOOzyvy71LYWWr32QOAk9fbNo6miFEUODYWrq0G80SQda1zrl9+gf791Uf3e/SAlStBelaFEOKRcpQg7dq1y9pxiCIm+sxnVO2cQmpsGZxL9bN1OEXL+dlw4XP1ffPF4NfOOudZvhwGDwajUU2SliwBGRAqhBA5kqME6fPPP2fRokV4eHiwZMkS+vXrJ+N7SrDos1FU6bRR3VBeJ49D2UqmKyvh+P/U9w1nQcX+1jnPDz/AsGFqb9WQIbBwIeh01jmXEEIUQzl6im3Tpk3m9dGef/5584zXomS6tO0b/OpFY0h1xNn7DVuHU3REhcOBIer7x9+AGm9a5zzz58MLL6jJ0csvw/ffS3IkhBC5lKNf/WvUqMGECRNo164diqKwatWqBw7G/u8SJKL4SbmTQpka6hOMabH9sXf2sm1ARUXcadjTA0x6COwNjWbDI8b15cns2fDmv4nXqFEwZ451ziOEEMVcjhKkr7/+mrFjx/Lrr7+i0WiYOHFitoO2NRqNJEjF3Nk1P9P0lT9RTBrcAt61dThFQ/J12NUFDPFQphW0WApaK/TofPghTJyovp8wQd2W5EgIIfIkRwlSixYtOHDgAABarZY///wTX19fqwYmCh+j3oi9kzo1Q8KNNngG5n1pmRJDHwfhXSD1BnjU/HciSCfLnkNRYNIkmDZN3Z46VU2UJDkSQog8y/FM2lOnTiUlJYXLly9TpkwZa8YkCqnz6/ZTq88RANz8pffokYzp6m21+LPgHADttoKjt2XPoSgwbtz/J0czZ8J770lyJIQQ+ZTjBGnKlCkkJSVRoUKFR86JJIofRVFIipqDg6uBpKiq6Ow72Dqkwk0xQcQQiN4Ndu7qLNmuFl6KxWSCkSPh00/V7S++UJMlIYQQ+Zbj57NzMqmjKL6u7/+bms9sB8DeZRwgSfJDHX8Lrv0EWntovRZK1bds/Uaj+oTawoVqb9GCBTB8uGXPIYQQJViuFquVnqOS6+ahz/Eol0havBeO7kNsHU7hdv4zOP9vr06zH8Dfwr1tGRn/P7eRVguLF0tyJIQQFparGf4ef/zxRyZJd+/ezVdAovCJvXyXwCd/BsCY/jIgk4Q+0LXV6jIiAA1mQKVBlq1fr4dBg9T11OzsYNky6NvXsucQQgiRuwRpypQpeHp6WisWUUj9uWkRzUbexKi3x9XXSpMbFgfRe2D/s4AC1UZAzbcsW39ampoMbdwIDg6wahV0727ZcwghhABymSD1799fHu8vYdIT0/Es/wMASVHd8QyUJxizFXcWdndXJ4Is9ww0/tyyT5KlpMAzz8C2beDkBGvXQkiI5eoXQgiRRY7HIMn4o5Lpj5838/hTZwHweGySjaMppFJuqHMdGeKgTEtoscyyE0EmJkJoqJocubrCr79KciSEEFaW4wRJnmIreUxGE4rpC7Q6hfjrzdBo69o6pMJHH68mRynXwaM6tF4Pds6Wqz8+Hjp3hvBwcHeH336D9u0tV78QQohs5fgWm8lksmYcohC6uPUYtXrvB8ClzDs2jqYQMuphb091nTUnf2i7FRx9LFa9fUICus6d4dgxKFVKTY6aNrVY/UIIIR4sV4/5i5Il7vIcHD30JEcHYu/0lK3DKVwUExx4HqJ2gp0btN0MbhUtV390NC3few/tsWNQujTs2iXJkRBCFKBcDdIWJUfkiRtU67oZAK39WCSXvseJCXB1OWjsoNXP4N3QcnXfvIldhw54Xr2K4u+PZscOqFXLcvULIYR4JPmpJ7J1dc+XlKocR3qiG86lXrJ1OIXLhS/hj5nq+2YLISDYcnVfuwatW6O5cIFUHx8yJDkSQgibkARJ3CcpMomAhisB0CcOBVxsGk+hcu1nODpKfV//Q6g82HJ1X7oErVvDpUsolSqx76OPoFo1y9UvhBAixyRBEvc5v34Z5VtdxZShw73s27YOp/CI3gf7BwEKVH0Fak2wXN0XLqjJ0dWrUK0aGTt2kOLnZ7n6hRBC5IokSCKLjLQMXHwWAJBwIwR4zLYBFRbxf8Cep8GUDo89DU3mWm4iyDNnoE0buHlTvZ22ezeUK2eZuoUQQuSJJEgiiz/W7aR695MAeDz2vo2jKSRSbsKuENDHgk9zaLnCchNBHjsGbdtCVBQ0aKDOdxQQYJm6hRBC5JkkSMJMURQMiXPQ2ZuIv14frZ08Vo4hAcK7Qso1cK8GbTaCnYXGZB04oE76GBMDTzwBO3dCGVnKRQghCgNJkITZ1d1/UKPnbgCcSo23cTSFgFEPe3tB3Elw8oV2W8GptGXq3rMHOnVSZ8p+8kkIC1MngxRCCFEoSIIkzKLPzsbFJ5WUO344uvW1dTi2pShw8EWI3A52rv9OBFnZMnVv366upZaUpPYgbd0KHh6WqVsIIYRFSIIkAIj56zaV2q8HQDGNBCy42GpRdPJduLIUNDp4cg14N7ZMvb/+Ck89Bamp0KULbNqkLkArhBCiUJEESQDw9/avKFPzDoYUJ1x9R9o6HNv6cz6cm66+b/YdlA2xTL1r18Izz0B6OnTvrm47W3BhWyGEEBYjCZIgLS4Nn2pLAUi+PRAowbd7rq+DI6+r7+tOhcpDLVPvihXQpw8YDNCvH6xeDY6OlqlbCCGExUmCJPhj3Roqd7yIyajBs/y7tg7Hdm7vh/0DAAWqDIc6Ey1T76JFMGgQGI0weDAsWwb29papWwghhFVIglTCmTJM2DnOAyD+Wjs0GgsNRC5qEi7A7m5gTIOyT0HT+ZaZCPLrr+H559VB3y+9BD/8ALoSPr5LCCGKAEmQSri/Nu+n5jOHAXAv+56No7GR1Mh/J4K8Cz5PwJMrQWuX/3o/+wxefVV9/8YbarKklX9yQghRFMj/1iVcUtSn2DkZSbhRHTvHNrYOp+AZEiE8FJKvgFtVaLNJfaw/v6ZPhzFj1Pfjx6vJkqWWJhFCCGF1kiCVYDeO/M3jT20HwN5lHFDCfoCbDLCvD8QeA8cy/04Emc+ZrBUF3n8f3nlH3Z48WU2WJDkSQogiRRKkEuzW4c9wD0gi9a43zqUG2zqcgqUocHA43PoNdC7Q9ldwr5L/OsePh6lT1e0ZM9RkSZIjIYQocmyaIO3Zs4du3bpRtmxZNBoN69aty/L50KFD0Wg0WV4hIVnnpLl79y6DBg3Cw8MDLy8vhg0bRlJSUgFeRdGUcCOeckGrATCkvgyUsKeqTk2Cy4v/nQhyFfjkc905k0kdZzRrlrr9+edqsiSEEKJIsmmClJycTP369Zk3b94Dy4SEhHDr1i3za8WKFVk+HzRoEGfPniUsLIxNmzaxZ88eXnrpJWuHXuT9tfk7/BtEkpHmgMdj/7N1OAXrrwVwdpr6vunX8Fho/uozGuHll2HuXLW3aMECNVkSQghRZFngUZ2869KlC126dHloGUdHR/z9/bP97I8//mDr1q0cPnyYJk2aAPDll1/StWtXPvnkE8qWLWvxmIsDQ4oBj3LfA5B48xlKVfa2cUQF6J8NcOQ19X2d96Hqi/mrLyNDfYz/xx/VJ9R++EGd60gIIUSRZtMEKSfCw8Px9fWlVKlStG/fnmnTpuHj4wNAREQEXl5e5uQIoGPHjmi1Wg4ePMgzzzyTbZ3p6emkp6ebtxMSEgAwGAwYDAYrXk3hcG7tBuoNOAeA62PvWuyaM+sprG2oiTmI7vf+aBQTpkrPY6zxjjqzdV4ZDOgGD0b7888oOh3GxYtR+vbNX53mqgt3WxYl0paWIe1oOdKWlmPNNizUCVJISAg9e/akUqVKXLp0iXfeeYcuXboQERGBTqcjMjISX1/fLMfY2dnh7e1NZGTkA+udPn06U6ZMuW//rl27cHFxsfh1FCaKolAu/VM0Wrh2qg7HL/8N/G3Rc4SFhVm0PktwNd2kVerb2JFKlK4RB6NCUbZsyXN9WoOBJrNmEXDoECY7Ow7/739EurnB5s0WjLpwtmVRJW1pGdKOliNtmX8pKSlWq7tQJ0j9+/c3v69bty716tWjSpUqhIeH06FDhzzXO2HCBMaOHWveTkhIIDAwkHbt2pl7p4qrKztPUKHvEQDKVPyIrjUttBAraiYfFhZGp06dsC9MS2mkRWG3cywaEjCVaoR32+10sXPLe30pKej69kV76BCKkxOmVatoFGK5doRC3JZFkLSlZUg7Wo60peXExMRYre5CnSDdq3LlypQuXZqLFy/SoUMH/P39iY6OzlImIyODu3fvPnDcEqjjmhyzWSjU3t6+2H9ZE6/PwcHNQOLNiriXfQprzH1UqNrRkAS/PwPJf4NbZbRtN6N1LpX3+pKSoEcPCA8HFxc0Gzdi1769paK9T6FqyyJO2tIypB0tR9oy/6zZfkVqHqR//vmHmJgYAgICAAgKCiIuLo6jR4+ay+zcuROTyUSzZs1sFWahdfuPW1TupN4C0tqPodhPDGkywL6+cPcIOJaGtlvB2S/v9cXHQ+fOanLk7g6//QZWTI6EEELYjk17kJKSkrh48aJ5+/Lly5w4cQJvb2+8vb2ZMmUKvXr1wt/fn0uXLvHWW29RtWpVOnfuDEDNmjUJCQlh+PDhfP311xgMBl5//XX69+9f6J9gS0iAX3+F/4wVt76rSxn6fjyJd0uxbutLGI2Wrd5o1HDyZCB37mgKwXqsCi10r/C4bgsZijNbkzZx55dqea7NIekuwZ92pvSVI6S7eBE2+jfuXHwCLj762LwoXG1ZtElbWoa0o+VIW1pOUpIVf9FXbGjXrl0KcN9ryJAhSkpKihIcHKyUKVNGsbe3VypUqKAMHz5ciYyMzFJHTEyMMmDAAMXNzU3x8PBQnn/+eSUxMTFXccTHxyuAcufOHUte3kO99pqiqFMvF9xr//7miqKgTJ48qcDPXdCv93u+ryjLUDKWapWnGm7IV11liFJOUE9RQImmtFKf4za/PnnJS17ykpeigPrzOz4+3uI/pzWKoijWS7+KhoSEBDw9Pblz506BDdKuXh3+/BNatAAvL+ufL8DtN777KQR9ugMvDLtCbGyAxc+hKCaio6Px9fVFo7Hd3dvgKt8xqvlwAL48+DVbL76c57q8027y4aGOlE/6g7uO/rz7xHauude2VKgPVFjasjiQtrQMaUfLkba0HL0+ju3bSxEfH4+Hh4dF65YEiYJPkGJioHRp9f2dO2DtUxoNRv7a3IQa3U9w92I3vKtusMp5DAYDmzdvpmvXrrYbeHjjV9jTHRQj1J4I9T/Ie13XrkGHDnDxIpQrBzt2wOOPWy7WhygUbVlMSFtahrSj5UhbWk5MTAylS5e2SoJUpJ5iKy4OHFD/rF7d+skRwF+bd/B46EkAPMq/b/0T2krMYXVQtmKEykOh3tS81/X33+oA7KtXoWJF2LkTKlWyVKRCCCEKOenbs4GICPXPoCDrn0tRFNITPkVrpxB7uSF2Do2tf1JbSLwI4aFgTIGAzvDEN+q6aHlx4QK0bq0mR9WqwZ49khwJIUQJIwmSDezfr/5ZEAnSjYMXqN5tNwDOpSZY/4S2kHYbdnWB9NtQqhE8uRq0eey2PnMG2rSBGzegVi3YvRsCAy0brxBCiEJPEqQClpEBhw6p71u0sP75os/OwskrnaTIsjh59bL+CQtaRjLsfgqSLoJrRWj7K9i7562u48ehbVuIioL69dX5jgIsP5hdCCFE4ScJUgE7cwaSk8HDQ+2gsKa4q3ep0Go9ACbj6xS7v25TBuzrBzGHwNEH2m0F5wfPoP5QBw+qY45iYqBpU3XMUZkylo1XCCFEkVHMfmIWfpnjj5o1A62VW//yjrn4PB5DeqILHo+NtO7JCpqiwOHX4OavoHOC1hvBo3re6tq3Dzp1grg4tVsvLAy8vS0arhBCiKJFEqQCVlDjj/RJeryrLgYgOXoAkI/FWQujM9Pg0reg0UKLFVAmjw26Y4e6fEhiIrRrpy4f4ulp2ViFEEIUOZIgFbDMHiRrjz/689dVVGj9N6YMLaUqTbLuyQrapR/g9L/X1PhLCOyRt3q2bIHQUEhJgZAQde0Xt2KWSAohhMgTSZAKUHQ0XLqkvrfmWrqKSUFn/yUAsZfbo9GWt97JCtrNLXBInSWbWhPg8dfyVs/atdC9u7oYXvfusG4dODtbLEwhhBBFmyRIBShzgshatay7vMjfO/ZTLfQoAB6PFaOJIWOOwL4+6kSQFZ+D+h/mrZ6VK6FPHzAYoG9fWL0aHB0tG6sQQogiTRKkAlRQ44+Sbs3CztFI3JVa2Ls8ad2TFZSkv2F3qPpYv39HaPZd3iaCXLwYBg0CoxGeew6WLQOZ6l8IIcQ9JEEqQAUx/ijq9FWqhoQB4OA2znonKkhpd2BXCKRFQ6kG0Opn0Dnkvp4FC2DoUDCZYPhwWLQI7GS1HSGEEPeTBKmAGAxw+LD63po9SDePzMLVN4Xk26VxKf2s9U5UUDJSYHc3SPwLXCtAm1/BPg8LEn7+Obzyivp+5Eg1WbL2PAtCCCGKLPkJUUBOnoTUVHXsUfU8TtfzKMnRSTz2xGoADMkvU+TXIjZlwO8DIOYAOJSCtlvApWzu65kxA0aPVt+/9ZaaLOV1nTYhhBAlgiRIBSTz9lrz5tbruLi0bQG+taMxpDjiWeFN65ykoCgKHBkJNzaA1hHabATPmrmvY/JkmPDvGnTvv68mS5IcCSGEeARJkAqItccfZaRn4Ob/HQAJ//RCoyllnRMVlHPT4eLXgAZaLocyLXN3vKLA22/DlCnq9vTparIkyZEQQogcKOL3YIqOzATJWuOPLm7ZRI0e51FMGryK+sSQfy+Bk++q7xt/DoE9c3e8oqi31L74Qt2eM+f/b7EJIYQQOSA9SAXg1i24ckXtvHjiCcvXrygKRsMcAO5eaonO3kqDnArCrW1wcJj6vuZbUD2Xa8iZTOpg7Mzk6KuvJDkSQgiRa9KDVAAye4/q1AGPPDyA9SjX95/g8VB1kiU3v4mWP0FBuXsM9vYCJQMqDoIG03N3fEYGvPACLF2qDvT6/nsYMsQ6sQohhCjWJEEqANYefxT390zKt8wg/nplPAODrXMSa0u6AuGhkJEEfu2h2ffqQrQ5ZExL49CYMfxtMmEcMgRq11a77JYssV7MVmQymYiJiWHZsmVoZTqCfJG2tAxpR8uRtrSc5ORkq9UtCVIBsOb4o7uXoqjY/lcAtLqxQBEchJweA+EhkBYJXvWg1S+5mggyIzmZNRMncsHfH/z91Z0pKXD5spUCLjhJSUm2DqHYkLa0DGlHy5G2zL+0tDSr1S0JkpXp9XDkiPreGgnS9d9nU39wIql3PXEv+6LlT2BtGamw+2lIuAAugdB2Mzh45vhwQ0ICP73/Ppe8vLAzGGhTqRKe1hjoVcCMRiMnTpygQYMG6HQ6W4dTpElbWoa0o+VIW1pOfHw8M2bMsErdkiBZ2fHj6oLxPj5QrZpl605PSMO3zjIAUu8+j7N3EVtw1WSE/YPgzn6w9/p3IsjHcny4/u5dVnzwAVe8vLDX6xnQqBGVevWyXrwFyGAwcPXqVWrXro29rBWXL9KWliHtaDnSlpYTExNjtbrl5qeV/ff2mqWn4PlryyICGt0gI92OUlUmWLZya1MUODoK/lkLWgdosx68auf48LToaJb+mxw5pKfzbPPmxSY5EkIIYXuSIFmZtcYfmYwmnDy+AiD28lNoNL6WPYG1/TET/poHaKDFj+DbOseHpt68ydLp0/nHywuntDQGt2lD+W7drBerEEKIEkdusVnZfvXpe4snSH9v30Hl4NMAeFV837KVW9vlZXDibfV9o9lQvk+OD02+epWln39OlJcXzqmpPBcSQkDbttaJUwghRIklPUhW9M8/6kung6ZNLVt3WtwnaHUKMX81xt6pgWUrt6bI7XDwefV9jbFQY3SOD028eJFFn39OlKcnrikpDH36aUmOhBBCWIX0IFlR5u21evXAzc1y9d46foFqXcIBcCr1juUqtrbYE7CnJ5gMUL4fNJyV40Pj//iDJd99x11PT9yTkhjSty8+ls46hRBCiH9JD5IVWWv80e1zH+PooSfhxmO4lu5h2cqtJfkqhHeFjETwbQtBi3M8EWTsqVP88P333PXwwCsxkeeffVaSIyGEEFYlPUhWZI3xR4m34ggMWgeAyTCSIpHjpt+FXV0g9RZ41oHWa0GXsykJ7hw5wpJVq0h0c8M7IYHBL76IZ82aVg5YCCFESVcEfroWTWlpcOyY+t6SS4xcCf+cUpVjSYt3xatiLhdytQVjGuzpAQl/gPNj/04E6ZWjQ6P372fR6tUkurpSJj6eoa+9JsmREEKIAiE9SFZy7BgYDODrC5UqWaZOQ6qBUpUWAZB4axBOni6WqdhaTEbY/xzc3gv2ntBuC7gG5ujQW+HhLN26lVQXF/zi4nhu9GhcK1SwcsBCCCGESnqQrMQaE0Re+u0nyjW/glGvw6fqe5ap1FoUBY6Nhetr1IkgW68Dr7o5OvSfbdtY8ttvpDo7UzYujiHjxklyJIQQokBJgmQllh5/pCgKWrsvAIi52BGtXTnLVGwt5z+FP9V4ab4Y/Nrm6LBrmzaxNDycNCcnAmNjeW7CBJzLlrVenEIIIUQ2bJog7dmzh27dulG2bFk0Gg3r1q3L8rmiKEyaNImAgACcnZ3p2LEjf/31V5Yyd+/eZdCgQXh4eODl5cWwYcNsvkKyovx/D5Klxh9d2xdBlc5HAfAMnGSZSq1Ec20lHB+nbjT8BCr2z9Fxf//8Mz9GRKB3dKRibCzPTpqEk28RmyFcCCFEsWDTBCk5OZn69eszb968bD+fOXMmX3zxBV9//TUHDx7E1dWVzp07k5aWZi4zaNAgzp49S1hYGJs2bWLPnj289NJLBXUJ2bp2DW7dAjs7aNLEMnUm3fwYnb2Juxdr4+huwVHfFlbaeArdoWHqRvVR6mSQOfDXypUsP3ECg4MDVWJjGThlCg7e3laMVAghhHgwmw7S7tKlC126dMn2M0VR+Oyzz5g4cSLdu3cHYMmSJfj5+bFu3Tr69+/PH3/8wdatWzl8+DBN/s1EvvzyS7p27conn3xCWRvdmsnsPWrQAJyd81/fnT+vUaljGAD2rm/lv0JriTvFE2kz0GBQlw9pNDtHA7D+WLyYNZcuYbKzo3pcHL0//BA7V9cCCFgIIYTIXqF9iu3y5ctERkbSsWNH8z5PT0+aNWtGREQE/fv3JyIiAi8vL3NyBNCxY0e0Wi0HDx7kmWeeybbu9PR00tPTzdsJCQkAGAwGDAZDvmPft08L6GjWzIjBYMp3fTcPf0y9Qakk3vLFqXRfi8RocSnX0e19Gi0pGH2exNRkIWQYAeNDDzv3/fesv3kTRaejZlwcT0+ZguLgUDivsQBlXn9JbwdLkLa0DGlHy5G2tBxrtmGhTZAiIyMB8PPzy7Lfz8/P/FlkZCS+94xRsbOzw9vb21wmO9OnT2fKlCn37d+1axcuLvl/dP6331oDpXByOs7mzTfyVVdGop42TVYBcP5UMDfTfst3fJZmpyTRKvUdPJSbJGgC2Zf6Cobfdj76wN9/56SzM4pWS5XISBw7duS3XbusH3AREhYWZusQig1pS8uQdrQcacv8S0lJsVrdhTZBsqYJEyYwduz/j41JSEggMDCQdu3a4ePjk6+6U1PhyhW1WV95pT4VKtTPV33nf5lD6Rp3SE90on67z2ig8chXfRZnTEe3NxRtyjVMTgEc0EyibfAz2NvbP/Sw419+yZZ/b6M1TEoiZM4cNDpdQURcJBgMBsLCwujUqdMj21I8nLSlZUg7Wo60peXExMRYre5CmyD5+/sDEBUVRUBAgHl/VFQUDRo0MJeJjo7OclxGRgZ37941H58dR0dHHB3vX+rC3t4+31/WAwcgIwMCAqBKFft8zYFkNBjxCFgIQPy1XvjWzl/yZnGKCQ4Og9t7wM4dY6sNpO6/8ch2PDBzJr+lpgLwRGoqITNmSHL0AJb4TgqVtKVlSDtajrRl/lmz/QrtPEiVKlXC39+fHTt2mPclJCRw8OBBgv6dXCgoKIi4uDiOHj1qLrNz505MJhPNmjUr8Jgh6/xH+Z0g8u8dG6jQ+k9MRg3eVSfnOzaLO/Y/uLYKtPbq+mpej+4t2/fhh+bkqIVeT8hHH0lyJIQQotCxaQ9SUlISFy9eNG9fvnyZEydO4O3tTfny5Rk9ejTTpk2jWrVqVKpUiffee4+yZcvSo0cPAGrWrElISAjDhw/n66+/xmAw8Prrr9O/f3+bP8FmifmPTPrZAMRcaEWZWlXzX6ElnZ8DF+ao75svAv8O6toqD6CYTOz+4AN2/7vdRlFo88EHaLSFNkcXQghRgtk0QTpy5Ajt2rUzb2eOCxoyZAiLFi3irbfeIjk5mZdeeom4uDiefPJJtm7dipOTk/mYZcuW8frrr9OhQwe0Wi29evXiiy++KPBrgawTROZ3Bu2bR05SJVitzNV/Yj4js7Crq9RlRAAafAwVBz60uGIysX3SJPb/2xXaQafjyYmF7JqEEEKI/7BpgtS2bVsURXng5xqNhqlTpzJ16tQHlvH29mb58uXWCC/XLl+G6Giwt4dGjfJXV+zf0ynbxEjs39UoVbnjow8oKFG7IeI59f3jr0PNcQ8trphMbH33XQ79m9R2dnKi+fjx1o5SCCGEyJdCO0i7KMocf9SoEfynkyvX4q9HU6H1rwBodKMBC612m19xZ2BPdzDpodwz0Oizhw60UoxGNk2YwLF/n1YLdXenydiczawthBBC2JIMALEgS40/+ifiE9z8k0i+XQqvCsPzH5glpPwD4V3AEA9lWkKLZaB98OBqk8HA+vHjOebqisZkoru3tyRHQgghigxJkCzIEuOP9MnplK75IwApt58HCsEjoPp4CO+qJkkeNaD1BrB78BoqxrQ0fnn7bU66u6MxmehZtiwNRo4swICFEEKI/JEEyUKSkuDUKfV9fhKkv7d/j1/dWxhSHPCpPsEyweWHMR32PgNxp8HJH9puAceHLCKbns7aSZM46+GB1mikT8WK1Hn55YKLVwghhLAASZAs5PBhMBqhXDn1lReKScHB7SsA7l58Cq2utAUjzFNAcOB5iNoFdm7Qbgu4VXxgcUNiIjHh4fzp5YUuI4P+NWpQ8/nnCy5eIYQQwkJkkLaFWGL80dU9O6jU7jQApSpPzndM+XZiPFxdARo7aPULlGrwwKL62FhWT53KdT8/7AwGBjRoQOXevQsuViGEEMKCpAfJQiwx/igt7mM0Wrh9rikObnUtE1heXfgC/vhEfd/8ewjo9MCiabdv8+OUKVwpVQqH9HQGNGkiyZEQQogiTXqQLMASE0Te/uMvKndU55l29n7HQpHl0bWf4eho9X39j6DScw8smnrrFj/OnMnNUqVwSkujlqcngaGhBROnEEIIYSXSg2QBf/0FMTHg6AgNG+atjttnP8LBzUDc1fK4+Xe3bIC5Eb0X9g8CFKj2KtR6+4FFk69dY/GsWdz08sI5NZWB7dtjqlmz4GIVQgghrEQSJAvI7D1q0gQcHHJ/fMqdBB5rthYAo/51bDYxZPy5fyeCTIdy3aHxlw+cCDLx0iUWf/YZUZ6euKakMLRbN/zbti3YeIUQQggrkQTJAvJ7e+3Kntl4BsaTEuOOd9XXLRdYbqTchF1dQB8LpYOgxfIHTgQZf/48i+bP57anJ+5JSQzt1Qvfli0LOGAhhBDCemQMkgXkJ0Ey6o14lV8EQOKNQbj4PHgCRqsxJKizZKdcA/fHoc1GsHPJtmjs6dMsWbyYOA8PPBMTGfLss5Rq0KBg4xVCCCGsTBKkfEpIgNPqk/l5SpAu71xB1ZCrZKTrKF3jPcsGlxNGPezpCXGnwMkP2m0FR59si8YcPcqSlStJcHfHOyGBwcOG4VmrVgEHLIQQQlif3GLLp0OH1KfYKlaEgIDcHasoChrtZwDcOR+MzqGsxeN7RABwcBhE7QA7V2j7K7hVyrbo7QMHWPTTTyS4uVE6Pp6hr7wiyZEQQohiS3qQ8ik/t9duHjlApfbHAPAsP8mCUeXQyXfgyo+g0cGTa8C7cbbFInfvZumWLaS4uuIXF8dzo0bhWrFiwcYqhBBCFCDpQcqn/CRIiTemo7VTuP1HPZxLNbdsYI/y5zw4N0N93+w7KBuSbbEbYWEs3rqVFGdnAuLiGDJunCRHQgghij3pQcoHkynvS4zEXrlOxbZhADi4vmXhyB7h+lo4MlJ9X+8DqDw022LXfv2VZb//jt7JiXKxsQx65x2cfH0LLk4hhBDCRqQHKR8uXIC4OHB2hnr1cnds5LEZOHmlkfCPP57lB1glvmzd3g/7BwIKVH0Jar+bbbHLv/zCj/v3o3d0pEJsLM++954kR0IIIUoMSZDyIbP3qGlTsLfP+XHpian41f8JgLT4lyiwv4b487C7GxjT4LFu0GRethNBXly5kuXHjmFwcKBKbCyDpkzB0Sf7J9uEEEKI4kgSpHzI6/ijK+Fz8a4SQ1q8M2Vq/s/ygWUn9RaEh4D+Lvg0g5YrQHv/HdbzS5aw4tw5MuzteTwujv7TpmHv6VkwMQohhBCFhCRI+bB/v/pnbsYfmYwmXMt8A0Ds333QaN2tENk9DIkQHgrJV8Gt6r8TQbreV+zsd9+x+uJFTDodteLj6fvRR9i5uVk/PiGEEKKQkUHaeRQXB+fOqe+b5+IBtKt71lOp3UVMGVp8qhfAo/0mA+ztDbHHwclXnQjSqcx9xU7On8/6qCgUnY56CQl0nzEDbV4WlhNCiP9r787joirbx49/hm3YQUA2ZVFzX1EUSS1LSs0szcofaanZYl/JynqetEXFTGzVxzKr50Ftc8lKUzPLzCWNJRdSNBEMw5BNkE3WYe7fH+TkABrgsMn1fr3m1cy573Ofa66ouV73Oec+QlwHZAapnmJiKv/ZqRPU5drlitK3AMg8fhNWtp0aILLLKAUxj0D692BuCzdvA4fqxzy0bBmbMzNRZmYEFBZKcSSEEKLVkxmkeqrP9UeZ8UfxvyUaAAevRnisyNGXIfnjvxaC3AiuA6t1iXnjDXYUFYFGw8CiIkYvWYLGvOaH1AohhBCthcwg1VN9rj/KOb0IC20F2ae6YOd+a8MEdkni+3D81cr3gz6AdndU63Jg8eLK4ggILitjdESEFEdCCCEEMoNUL3r936fYajuDVJiehc+N31R+0MxumMAu+fNrODiz8n3vBdBpulGz0uvZ98or7Pnr8016PcNfeQWNmdTLQgghBMgMUr2cOAH5+WBnB7161W6f1NjXsWtbREGaK66dp//zDvV1PhoOhILSQ6dHoJfxheBKr2fX/PmG4uhWMzNuCQ+X4kgIIYS4jPwq1sOl648GDQKLWszB6UrKcen8CQAXM6bRYBN3+adg751QUQzed8DAlUYLQSq9nu9efJEDfwV9u1bLsJcb4VooIYQQooWRAqke6nr90Zm9/6Nt9wzKCrW493qhYYIqTofdo6A0G1wCYcgGo4UgVUUF38yZQ4y1NQB32NsTPGdOw8QihBBCtHBSINVDXe5gU0phZbsCgPMJYzGzaGP6gMoLK2eOLiaDfScY/g1Y/r3Ao768nK/nzOGQnR0oxV1t2jDw2WdNH4cQQghxnZCLtOsoO7vyIbVQuwUiU2N+wHfYcZReg8sNC0wfkL4c9t8HOYdA6/bXQpB/L8xUUVLCphdf5LijIxq9nvHe3vR+/HHTxyGEEEJcR2QGqY6iK5cxoksXqM3zW4svvAZAxrEgrJ16mjYYpSD2cUjbAeY2fy0EeYOhWXfxIl+88ALHHR0xq6jgPj8/KY6EEEKIWpAZpDq6dHqtNtcf5ZxOxP/mvQDYuMw1fTDHFsDvq0FjBkM/B7cgQ1N5QQGfz5tHkrMz5jod93frRpdJk0wfgxBCCHEdkhmkOqrL9UdZx1/F0lZHTpI/Tj5jTRtI0ocQv7Dy/cCV0O5OQ1PZhQus+6s4sigv54E+faQ4EkIIIepAZpDqQKeD2NjK9/9UIJXk5uMd+FXlfiVhgObqO9RF6jb45YnK971ehhseMzSVZmWxdvFiUpydsSor44FBg/C7+27THVsIIYRoBZr1DNKCBQvQaDRGr27duhnaS0pKmDlzJq6urtjb2zNhwgQyMjIaLJ74eCgsBEdH6NHj6n1TDryFg3cBF7McadszzHRBnI+F/RMrF4LsOA16hxuaitPT+eSv4khbUsKDw4ZJcSSEEELUQ7MukAB69uxJWlqa4bV//35D2zPPPMPWrVvZuHEje/fu5dy5c9xzzz0NFsul02tBQXC1R5bpdRU4+awCIC9lEhqN1jQBFCTB3jFQUQReoyqfsfbXQpBFZ8/y8WuvkersjE1xMVNuu432o0aZ5rhCCCFEK9PsT7FZWFjg6elZbXteXh6RkZGsXbuWW2+tfPDr6tWr6d69O9HR0QyuzT34dVTb649S9q/Ff/iflBdb0LaHiVaqLsn8ayHI89CmPwzdCGaWABT+/jsfv/suWc7O2BUV8eDYsXgMHWqa4wohhBCtULOfQUpMTMTb25uOHTsyadIkUlJSADh06BDl5eWEhIQY+nbr1g1fX1+iLlUyJlb7C7SXAZB1fCSWNl7XfmDdRdhzJxSeBrsORgtB5icksGbFCrKcnHC4eJGpEyZIcSSEEEJco2Y9gxQUFMSaNWvo2rUraWlphIeHM2zYMOLj40lPT8fKygpnZ2ejfTw8PEhPT7/quKWlpZSWlho+5+fnA1BeXk55eXmN+2RmQlJS5YzNgAHlXKEbmcdi8B12BAA77xevOF6t6XWYH7gPs5xfUFau6IZtBQtXKC8nNz6etZ99Rq6jI06FhTwwcSJOffte+zHr6dJxm+r41xPJpelILk1D8mg6kkvTacgcNusCafTo0Yb3ffr0ISgoCD8/Pz7//HNsbGzqPW5ERATh4eHVtu/evRtbW9sa94mN9QSC8PHJ5+efd19xbN/SpbTrr/gjthtxaZnA9nrHiVL0LXsPf91OKrDigNm/uPBTEpCEWUoKiX/8QYGDA855eXTs3Jmo1FRITa3/8Uxk586dTR3CdUNyaTqSS9OQPJqO5PLaFRUVNdjYzbpAqsrZ2ZkuXbqQlJTEbbfdRllZGbm5uUazSBkZGTVes3S5uXPnMnv2bMPn/Px8fHx8uOWWW3C9wvLYBw5Uno0MCbHnjjvuqLFPwblU7NpULrVt6/QCdwTU3K+2zE4swvz4ThRmqBvXEtzuLgCyoqNZd+AAhQ4OuOXlEfrIIzh07nxNxzKF8vJydu7cyW233YalpWVTh9OiSS5NR3JpGpJH05Fcmk52dnaDjd2iCqTCwkJOnz7Ngw8+yIABA7C0tGTXrl1MmDABgISEBFJSUgj+h4uEtFotWm31O8ssLS2v+Md6af2jIUPMsLSs+dKt8/Gv4TKmlNwz3rTtOplrWvvodCQcr1wIUjPwXSz8K79j+r59fLZ9O0V2drjn5fHQrFnY+fvX/zgN4Gp5FHUjuTQdyaVpSB5NR3J57Royf826QHruuecYO3Ysfn5+nDt3jvnz52Nubk5oaChOTk5Mnz6d2bNn4+LigqOjI08++STBwcEmv4OtvPzvAulKjxgpLyrBvdcGAIovPIaz/zUUR6nbK5+xBtDzBehcuShk6g8/8OmuXZTY2OCVm8vkZ5/Ftn37+h9HCCGEEDVq1gXSn3/+SWhoKNnZ2bRt25ahQ4cSHR1N27ZtAVi6dClmZmZMmDCB0tJSRo4cyXvvvWfyOI4eheJicHaGrl1r7pOy/x063X6B4hw7PPo8V/+DZR+E/feBqoAOD0GfRZXjb9/O2v37KbW2pv2FC0yaOxdrD4/6H0cIIYQQV9SsC6T169dftd3a2poVK1awYsWKBo3j0u39gweDWQ1n15RS2Lh+AEB24n20D7Kr34EKf/97IUjP22DQf0Gj4czmzaz95RfKtVr8Llwg9OWX0V7hWikhhBBCXLtmXSA1F/+0/tGf0ZvxCT5NRZk5bXvMq99BSrIqF4IsyYQ2/WDYl2BuRdKGDWw4dgydlRUdL1zg/4WHY+nkVL9jCCGEEKJWmv1Ckc3Bzz9X/vNK1x/pSt4EIP3ozWgdOtT9ALoi2DsWChLBzg+GbwdLBxI++YT18fHoLC3pkptL6KJFUhwJIYQQjUBmkP5BejqcOVP5yLNBg6q3ZycexXdo5a39Dp71eKyIXgcH/h9kx4CVCwzfATZeHI+M5Ks//kBvYUH3vDwmLF6M+TWs/SSEEEKI2pMZpH9w6fRar17g6Fi9PSdpEeaWejKPd8ex/fC6Da4UHAyD1K1gbg03bwGnbhxduZIvU1LQm5vTOz+fe5cskeJICCGEaEQyg/QPrnb9UVH2edoN2vbXp6frPvjxxZD0AaCBG9dC2yEc/s9/2HrhApiZ0a+wkLFLlmAm62QIIYQQjUpmkP7B1a4/Sv1lCbauxeSddaNtj4frNvDva+DoS5XvBywHn/HEvvkmW3NzQaMhsKiIu6Q4EkIIIZqEFEhXUVYGBw9Wvq86g1RRXo7rDR8DUJg2DY2mDpNx576DmEcr3/d4HrqG8XNEBN9evAjA4NJS7oiIQGNufq1fQQghhBD1IAXSVcTFQWkpuLpC1UednT3wX1xuyKI03xqPvnNrP2jOYdg/AZQO/Ceh+rzK3oUL2VlWBsCwigpuX7QITU0LLgkhhBCiUciv8FVcvkCk5rInhyilsLB5F4DM43dhoW1TuwELk2HPHaC7CB4jUAP/x48LwtmjFAC3aDTcunChFEdCCCFEE5OLtK/iStcfZfy6i/ZBv6Gv0ODaeX7tBis5/9dCkBng3Ac1ZCPfzV9IzF8Pzb3dyorguXWYiRJCCCFEg5Gpiqu40h1sRdkRAKQfCcbWrcc/D6Qrgn13QcEpsPVB3bSNb+YvMRRHd9jbS3EkhBBCNCMyg3QFqalw9mzls9cGDvx7e97ZRHyH7APAuk0tihp9Bfw8Cc5HgaUz+mFb2Rq+nDh7e1CKu1xcCJg1q4G+hRBCCCHqQ2aQruDS7FGfPmBv//f2rBOvYGGt4/zJDrh0GnP1QZSCQ7Pgz81gpqUieCObFn9MnL09Gr2e8Z6eUhwJIYQQzZDMIF1BTdcflRUW4NlvEwDlxWGApvqOlzvxGiS+B2ioCFzFF299w0lnZ8wqKpjg70+Ph+u4dpIQQgghGoXMIF1BTdcf/RnzBvYehRSkOePZN+zqAyR/Ar9WnoKr6LWEDSt+4aSzM+Y6Hfd37izFkRBCCNGMSYFUg9JSOHy48v2lAknp9Th4rwIgN3kSGjOrKw+QthOiKwugig4zWbvqHInOzliUlxPauzddH3ywIcMXQgghxDWSAqkGhw9XrqLt7g4dO1Zu+zPmU9p2T6XsoiUefV6+8s4X4uCnyoUgKzzH8/EGc35v0wbLsjImDRxIp/vvb5TvIIQQQoj6kwKpBpeuPwoOvmyBSLUUgIxfR2Fl71Hzjhf/gN2jQVdARZshfLytHSltXNCWlvLgkCH43313wwcvhBBCiGsmF2nXoOr1R+dPRtF+cBwAzv7zat6pNOevhSDT0dt25eNdvUlxcsOmuJjJt92G94gRDR+4EEIIIUxCZpCqUKp6gZSfugiNGaQd6o+Dd2D1nXTFlQtB5p9Eb+XFp1HDSHHwxLaoiCljxkhxJIQQQrQwMoNUxdmzcO4cWFhAYCBczPqT9oN3AmBu/Vz1HfQVEDUZsg6gzB1Yf+Q2km3bY3/xIg9NmEDboKBG/gZCCCGEuFYyg1TFpeuP+vUDW1tIj3sVK7tyshPb07bHROPOSsHhZ+DsVyiNJV/9NppEq444FRYyLTRUiiMhhBCihZIZpCouP72mKy2hbff1ABTnPIZGU6We/O1NOPUOANt/H0m8pgdtCgp4aNo0nHv2bMywhRBCCGFCMoNUxeUFUmrMchzb53Ixyx6v/rONO55ZC3H/BmBP6nAO6gJxzc9n6mOPSXEkhBBCtHBSIF2muBiOHKl8HxyssHZ+H4DshPswt7T7u2P6LoieCsDBrAHsLRqOe14eU8PCcOzSpZGjFkIIIYSpySm2y8TFadDpwMsLrHI24dE/GV2pOW17XnZr/4Wj8NM9oC/nZG43vskdg2duLg/Ono2tj0/TBS+EEEIIk5EZpMv88kvlqpDBwaAregOAc4duwaaNf2WHiymwZzSU55Ny0Ycvzk+gXW4+U+bMkeJICCGEuI7IDNJlLhVIwwYm0H5wDAAOXn89VqTsQmVxVHyOrBI31qU/QLucQh54+WW0rq5NFbIQQgghGoDMIF3m4MHKAmloj4WYWSjSjvSkTYeboKIE9o2DvBMUlNvzadqDeGeXMGn+fCmOhBBCiOuQzCBdJitLg7NzAb1u+qpyg3oKlB6iHoLMfZRUaPn03GQ8zivuX7QIC3v7pg1YCCGEEA1CCqQq/jXrfaydS7iQ7I5nwMNw+FlI2UiFMmND2kRcs2yYsPhVzG1smjpUIYQQQjQQKZAuo9FUMGXKcgAKUqfRpuQ/kLAMgM3p47HPcGP8kgjMrKyaMEohhBBCNDQpkC4zcuR3tOv4J8UXbPD26Ay/PALAzqzbsEjvwNglizGztGziKIUQQgjR0KRAuszMmSsAyPtlMO5Zj4MGYnIHUXauH3ctiUBjbt7EEQohhBCiMchdbJe56ab96M9rcMv4GTNNBScKu5P75y3cEbFEiiMhhBCiFbluCqQVK1bg7++PtbU1QUFBxMbG1n2QIijbpsXCvJQ/in3JSL2b2xctRmN23aRJCCGEELVwXfzyb9iwgdmzZzN//nwOHz5M3759GTlyJJmZmXUap/BLW6ytSsgqdSM1LZRbwiOkOBJCCCFaoevi1//tt9/m0UcfZdq0afTo0YP3338fW1tbVq1aVadx7LVFFOjsScmaxo0vLWmgaIUQQgjR3LX4AqmsrIxDhw4REhJi2GZmZkZISAhRUVF1GqtUb8kfOY8w4N+vmzpMIYQQQrQgLf4utvPnz1NRUYGHh4fRdg8PD06ePFnjPqWlpZSWlho+5+XlARCfPoE+M18iOzu74QK+jpWXl1NUVER2djaWshzCNZFcmo7k0jQkj6YjuTSdnJwcAJRSJh+7xRdI9REREUF4eHi17bfOXw/z1zdBREIIIYSor+zsbJycnEw6ZosvkNzc3DA3NycjI8Noe0ZGBp6enjXuM3fuXGbPnm34nJubi5+fHykpKSZPcGuSn5+Pj48PZ8+exdHRsanDadEkl6YjuTQNyaPpSC5NJy8vD19fX1xcXEw+dosvkKysrBgwYAC7du1i3LhxAOj1enbt2kVYWFiN+2i1WrRabbXtTk5O8sdqAo6OjpJHE5Fcmo7k0jQkj6YjuTQdswa447zFF0gAs2fPZsqUKQQGBjJo0CCWLVvGxYsXmTZtWlOHJoQQQogW6LookCZOnEhWVhbz5s0jPT2dfv36sWPHjmoXbgshhBBC1MZ1USABhIWFXfGU2j/RarXMnz+/xtNuovYkj6YjuTQdyaVpSB5NR3JpOg2ZS41qiHvjhBBCCCFasBa/UKQQQgghhKlJgSSEEEIIUYUUSEIIIYQQVUiBJIQQQghRRasvkFasWIG/vz/W1tYEBQURGxvb1CE1O/v27WPs2LF4e3uj0WjYvHmzUbtSinnz5uHl5YWNjQ0hISEkJiYa9cnJyWHSpEk4Ojri7OzM9OnTKSwsbMRv0fQiIiIYOHAgDg4OuLu7M27cOBISEoz6lJSUMHPmTFxdXbG3t2fChAnVVolPSUlhzJgx2Nra4u7uzr/+9S90Ol1jfpUmt3LlSvr06WNYaC84OJhvv/3W0C55rJ8lS5ag0Wh4+umnDdskl7WzYMECNBqN0atbt26Gdslj7aWmpjJ58mRcXV2xsbGhd+/eHDx40NDeaL85qhVbv369srKyUqtWrVLHjx9Xjz76qHJ2dlYZGRlNHVqzsn37dvXiiy+qr776SgFq06ZNRu1LlixRTk5OavPmzerXX39Vd911l+rQoYMqLi429Bk1apTq27evio6OVj/99JO64YYbVGhoaCN/k6Y1cuRItXr1ahUfH6/i4uLUHXfcoXx9fVVhYaGhz4wZM5SPj4/atWuXOnjwoBo8eLC68cYbDe06nU716tVLhYSEqCNHjqjt27crNzc3NXfu3Kb4Sk1my5Yt6ptvvlGnTp1SCQkJ6oUXXlCWlpYqPj5eKSV5rI/Y2Fjl7++v+vTpo5566inDdsll7cyfP1/17NlTpaWlGV5ZWVmGdslj7eTk5Cg/Pz81depUFRMTo37//Xf13XffqaSkJEOfxvrNadUF0qBBg9TMmTMNnysqKpS3t7eKiIhowqiat6oFkl6vV56enuqNN94wbMvNzVVarVatW7dOKaXUiRMnFKB++eUXQ59vv/1WaTQalZqa2mixNzeZmZkKUHv37lVKVebN0tJSbdy40dDnt99+U4CKiopSSlUWq2ZmZio9Pd3QZ+XKlcrR0VGVlpY27hdoZtq0aaP+97//SR7roaCgQHXu3Fnt3LlT3XzzzYYCSXJZe/Pnz1d9+/atsU3yWHvPP/+8Gjp06BXbG/M3p9WeYisrK+PQoUOEhIQYtpmZmRESEkJUVFQTRtayJCcnk56ebpRHJycngoKCDHmMiorC2dmZwMBAQ5+QkBDMzMyIiYlp9Jibi7y8PADDQxYPHTpEeXm5US67deuGr6+vUS579+5ttEr8yJEjyc/P5/jx440YffNRUVHB+vXruXjxIsHBwZLHepg5cyZjxowxyhnI32RdJSYm4u3tTceOHZk0aRIpKSmA5LEutmzZQmBgIPfddx/u7u4EBATw3//+19DemL85rbZAOn/+PBUVFdUeR+Lh4UF6enoTRdXyXMrV1fKYnp6Ou7u7UbuFhQUuLi6tNtd6vZ6nn36aIUOG0KtXL6AyT1ZWVjg7Oxv1rZrLmnJ9qa01OXbsGPb29mi1WmbMmMGmTZvo0aOH5LGO1q9fz+HDh4mIiKjWJrmsvaCgINasWcOOHTtYuXIlycnJDBs2jIKCAsljHfz++++sXLmSzp0789133/HEE08wa9YsPvroI6Bxf3Oum0eNCNGSzJw5k/j4ePbv39/UobRYXbt2JS4ujry8PL744gumTJnC3r17mzqsFuXs2bM89dRT7Ny5E2tr66YOp0UbPXq04X2fPn0ICgrCz8+Pzz//HBsbmyaMrGXR6/UEBgayePFiAAICAoiPj+f9999nypQpjRpLq51BcnNzw9zcvNpdBBkZGXh6ejZRVC3PpVxdLY+enp5kZmYatet0OnJyclplrsPCwti2bRu7d++mffv2hu2enp6UlZWRm5tr1L9qLmvK9aW21sTKyoobbriBAQMGEBERQd++ffnPf/4jeayDQ4cOkZmZSf/+/bGwsMDCwoK9e/eyfPlyLCws8PDwkFzWk7OzM126dCEpKUn+JuvAy8uLHj16GG3r3r274XRlY/7mtNoCycrKigEDBrBr1y7DNr1ez65duwgODm7CyFqWDh064OnpaZTH/Px8YmJiDHkMDg4mNzeXQ4cOGfr8+OOP6PV6goKCGj3mpqKUIiwsjE2bNvHjjz/SoUMHo/YBAwZgaWlplMuEhARSUlKMcnns2DGj//h37tyJo6Njtf+ptDZ6vZ7S0lLJYx2MGDGCY8eOERcXZ3gFBgYyadIkw3vJZf0UFhZy+vRpvLy85G+yDoYMGVJt+ZNTp07h5+cHNPJvTt2vMb9+rF+/Xmm1WrVmzRp14sQJ9dhjjylnZ2ejuwhE5R0uR44cUUeOHFGAevvtt9WRI0fUH3/8oZSqvOXS2dlZff311+ro0aPq7rvvrvGWy4CAABUTE6P279+vOnfu3Opu83/iiSeUk5OT2rNnj9GtwEVFRYY+M2bMUL6+vurHH39UBw8eVMHBwSo4ONjQfulW4Ntvv13FxcWpHTt2qLZt27a6W4HnzJmj9u7dq5KTk9XRo0fVnDlzlEajUd9//71SSvJ4LS6/i00pyWVtPfvss2rPnj0qOTlZHThwQIWEhCg3NzeVmZmplJI81lZsbKyysLBQr776qkpMTFSfffaZsrW1VZ9++qmhT2P95rTqAkkppd555x3l6+urrKys1KBBg1R0dHRTh9Ts7N69WwHVXlOmTFFKVd52+fLLLysPDw+l1WrViBEjVEJCgtEY2dnZKjQ0VNnb2ytHR0c1bdo0VVBQ0ATfpunUlENArV692tCnuLhY/d///Z9q06aNsrW1VePHj1dpaWlG45w5c0aNHj1a2djYKDc3N/Xss8+q8vLyRv42Tevhhx9Wfn5+ysrKSrVt21aNGDHCUBwpJXm8FlULJMll7UycOFF5eXkpKysr1a5dOzVx4kSjtXskj7W3detW1atXL6XValW3bt3Uhx9+aNTeWL85GqWUquMMmBBCCCHEda3VXoMkhBBCCHElUiAJIYQQQlQhBZIQQgghRBVSIAkhhBBCVCEFkhBCCCFEFVIgCSGEEEJUIQWSEEIIIUQVUiAJ0UokJCTg6elJQUHBNY915swZNBoNcXFxV+03fPhwnn766Ws+nqksWLCAfv36NXUYjUKj0bB58+ZrGmPHjh3069cPvV5vmqCEaEGkQBKiBYiKisLc3JwxY8ZUa6ttsTJ37lyefPJJHBwcqrV169YNrVZLenp6reLx8fEhLS2NXr16AbBnzx40Gk21h3E2pZoKhOeee87oGU51VZ8Cy9/fn2XLltX7mPWVlpZmeMJ8bf9Gqho1ahSWlpZ89tlnDRChEM2bFEhCtACRkZE8+eST7Nu3j3PnztV5/5SUFLZt28bUqVOrte3fv5/i4mLuvfdePvroo38cq6ysDHNzczw9PbGwsKhzLE3J3t4eV1fXpg6jUXh6eqLVaq95nKlTp7J8+XITRCREyyIFkhDNXGFhIRs2bOCJJ55gzJgxrFmzps5jfP755/Tt25d27dpVa4uMjOSBBx7gwQcfZNWqVdXa/f39eeWVV3jooYdwdHTkscceM5qROHPmDLfccgsAbdq0QaPRGBVier2ef//737i4uODp6cmCBQuMxtdoNHzwwQfceeed2Nra0r17d6KiokhKSmL48OHY2dlx4403cvr0aaP9vv76a/r374+1tTUdO3YkPDwcnU5niBlg/PjxaDQaw+eaZoBWrVpFz5490Wq1eHl5ERYWVuu8Tp06lXHjxvHmm2/i5eWFq6srM2fOpLy8HKg8xfjHH3/wzDPPoNFo0Gg0hn3379/PsGHDsLGxwcfHh1mzZnHx4kWjvC9evJiHH34YBwcHfH19+fDDDw3tZWVlhIWF4eXlhbW1NX5+fkRERBjl9dIMWocOHQAICAhAo9EwfPhw9u3bh6WlZbVZw6effpphw4YZPo8dO5aDBw9Wy78Q171rfKacEKKBRUZGqsDAQKVU5UMcO3XqpPR6vaE9OTlZAerIkSNXHOOuu+5SM2bMqLY9Pz9f2dnZqfj4eKXT6ZSHh4fat2+fUR8/Pz/l6Oio3nzzTZWUlKSSkpKMjqnT6dSXX36pAJWQkKDS0tJUbm6uUqrywaeOjo5qwYIF6tSpU+qjjz5SGo3G6MGygGrXrp3asGGDSkhIUOPGjVP+/v7q1ltvVTt27FAnTpxQgwcPVqNGjTLss2/fPuXo6KjWrFmjTp8+rb7//nvl7++vFixYoJRSKjMz0/Ag4LS0NMMT1efPn6/69u1rGOe9995T1tbWatmyZSohIUHFxsaqpUuXXjGPVfefMmWKcnR0VDNmzFC//fab2rp1q7K1tTU8XDM7O1u1b99eLVy4UKWlpRkeTpqUlKTs7OzU0qVL1alTp9SBAwdUQECAmjp1qlHeXVxc1IoVK1RiYqKKiIhQZmZm6uTJk0oppd544w3l4+Oj9u3bp86cOaN++ukntXbtWqO8btq0SSlV+YR0QP3www8qLS1NZWdnK6WU6tKli3r99dcN+5SVlSk3Nze1atUqo+/t4eFh9FBlIVoDKZCEaOZuvPFGtWzZMqWUUuXl5crNzU3t3r3b0F6bAqlv375q4cKF1bZ/+OGHql+/fobPTz31lJoyZYpRHz8/PzVu3DijbVWPuXv3bgWoCxcuGPW7+eab1dChQ422DRw4UD3//POGz4B66aWXDJ+joqIUoCIjIw3b1q1bp6ytrQ2fR4wYoRYvXmw07ieffKK8vLyMxr1UIFxStcDx9vZWL774oqqtmgokPz8/pdPpDNvuu+8+NXHiRMNnPz+/akXX9OnT1WOPPWa07aefflJmZmaquLjYsN/kyZMN7Xq9Xrm7u6uVK1cqpZR68skn1a233mpULF/u8u9/pb+R1157TXXv3t3w+csvv1T29vaqsLDQqF9AQICh+BSitZBTbEI0YwkJCcTGxhIaGgqAhYUFEydOJDIysk7jFBcXY21tXW37qlWrmDx5suHz5MmT2bhxY7U73QIDA+sRfaU+ffoYffby8iIzM/OKfTw8PADo3bu30baSkhLy8/MB+PXXX1m4cCH29vaG16OPPkpaWhpFRUW1iiszM5Nz584xYsSIen2vS3r27Im5ubnhc03fr6pff/2VNWvWGMU/cuRI9Ho9ycnJhn6X50Wj0eDp6WkYe+rUqcTFxdG1a1dmzZrF999/X+fYp06dSlJSEtHR0QCsWbOG+++/Hzs7O6N+NjY2tc6rENeLlnWFpRCtTGRkJDqdDm9vb8M2pRRarZZ3330XJyenWo3j5ubGhQsXjLadOHGC6OhoYmNjef755w3bKyoqWL9+PY8++qhhW9UfzLqwtLQ0+qzRaKrdNn55n0vX6dS07dJ+hYWFhIeHc88991Q7Xk2FYE1sbGxq1e+f1Ob7VVVYWMjjjz/OrFmzqrX5+vrWauz+/fuTnJzMt99+yw8//MD9999PSEgIX3zxRa1jd3d3Z+zYsaxevZoOHTrw7bffsmfPnmr9cnJyaNu2ba3HFeJ6IAWSEM2UTqfj448/5q233uL22283ahs3bhzr1q1jxowZtRorICCAEydOGG2LjIzkpptuYsWKFUbbV69eTWRkpFGB9E+srKyAyuKqMfTv35+EhARuuOGGK/axtLS8ajwODg74+/uza9cuw0XmDcHKyqpaHP379+fEiRNXjb82HB0dmThxIhMnTuTee+9l1KhR5OTk4OLiUi0GqPnfzyOPPEJoaCjt27enU6dODBkyxKi9pKSE06dPExAQcE2xCtHSyCk2IZqpbdu2ceHCBaZPn06vXr2MXhMmTKjTabaRI0cSFRVl+IEsLy/nk08+ITQ0tNrYjzzyCDExMRw/frzW4/v5+aHRaNi2bRtZWVkUFhbW+fvWxbx58/j4448JDw/n+PHj/Pbbb6xfv56XXnrJ0OdS8ZOenl5t9uySBQsW8NZbb7F8+XISExM5fPgw77zzjklj9ff3Z9++faSmpnL+/HkAnn/+eX7++WfCwsKIi4sjMTGRr7/+uk530L399tusW7eOkydPcurUKTZu3IinpyfOzs7V+rq7u2NjY8OOHTvIyMggLy/P0DZy5EgcHR1ZtGgR06ZNq7ZvdHQ0Wq2W4ODgun95IVowKZCEaKYiIyMJCQmp8TTahAkTOHjwIEePHq3VWKNHj8bCwoIffvgBgC1btpCdnc348eOr9e3evTvdu3evUwHWrl07wsPDmTNnDh4eHnX6oa+PkSNHsm3bNr7//nsGDhzI4MGDWbp0KX5+foY+b731Fjt37sTHx+eKsx9Tpkxh2bJlvPfee/Ts2ZM777yTxMREk8a6cOFCzpw5Q6dOnQynqfr06cPevXs5deoUw4YNIyAggHnz5hmdSv0nDg4OvP766wQGBjJw4EDOnDnD9u3bMTOr/r91CwsLli9fzgcffIC3tzd33323oc3MzIypU6dSUVHBQw89VG3fdevWMWnSJGxtbevx7YVouTRKKdXUQQghGt6KFSvYsmUL3333XVOHIpqZ6dOnk5WVxZYtW4y2nz9/nq5du3Lw4EHDWkpCtBZyDZIQrcTjjz9Obm4uBQUFNT5uRLQ+eXl5HDt2jLVr11YrjqDyESXvvfeeFEeiVZIZJCGEaKWGDx9ObGwsjz/+OEuXLm3qcIRoVqRAEkIIIYSoQi7SFkIIIYSoQgokIYQQQogqpEASQgghhKhCCiQhhBBCiCqkQBJCCCGEqEIKJCGEEEKIKqRAEkIIIYSoQgokIYQQQogqpEASQgghhKji/wOaTORDPWHg8gAAAABJRU5ErkJggg==",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "import matplotlib.pyplot as plt\n",
    "\n",
    "def plot_roofline(peak_performance, memory_bandwidth, color, label):\n",
    "    # peak perf TFlops, mem bandwidth TB/s\n",
    "    coef = peak_performance / memory_bandwidth\n",
    "    ai_intevals = range(1000)\n",
    "    # ai_values = coef * ai_intevals\n",
    "    tflops_values = [min(i * memory_bandwidth, peak_performance) for i in ai_intevals]\n",
    "\n",
    "    plt.plot(ai_intevals, tflops_values, color=color, label = label)\n",
    "\n",
    "# A100\n",
    "peak_performance = 312\n",
    "memory_bandwidth = 2.039\n",
    "\n",
    "plot_roofline(peak_performance, memory_bandwidth, 'purple', \"A100\")\n",
    "\n",
    "\n",
    "# H20\n",
    "peak_performance = 144  \n",
    "memory_bandwidth = 4.0\n",
    "\n",
    "# 绘制 Roofline 曲线\n",
    "plot_roofline(peak_performance, memory_bandwidth, 'blue', \"H20(rumors)\")\n",
    "\n",
    "# H20\n",
    "peak_performance = 312  # Tflops\n",
    "memory_bandwidth = 2.03   # TB/s\n",
    "\n",
    "plot_roofline(peak_performance, memory_bandwidth, 'yellow', \"A800\")\n",
    "\n",
    "# L40\n",
    "peak_performance = 362 \n",
    "memory_bandwidth = 0.864  \n",
    "plot_roofline(peak_performance, memory_bandwidth, 'red', \"L40S\")\n",
    "\n",
    "# L20\n",
    "peak_performance = 119.5 \n",
    "memory_bandwidth = 0.864  \n",
    "plot_roofline(peak_performance, memory_bandwidth, 'gray', \"L20\")\n",
    "\n",
    "# 4090\n",
    "peak_performance = 330  \n",
    "memory_bandwidth = 1.008 \n",
    "\n",
    "plot_roofline(peak_performance, memory_bandwidth, 'orange', \"4090\")\n",
    "\n",
    "plt.xlim(0, 600)\n",
    "plt.ylim(0, 400) \n",
    "\n",
    "plt.xlabel('AI (Arithmetic Intensity)')\n",
    "plt.ylabel('Tflops')\n",
    "plt.title('Roofline Model of NVIDIA GPUs')\n",
    "\n",
    "plt.grid(True)\n",
    "plt.legend()\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "id": "af094454",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Defaulting to user installation because normal site-packages is not writeable\n",
      "Collecting matplotlib\n",
      "  Downloading matplotlib-3.8.3-cp39-cp39-macosx_10_12_x86_64.whl.metadata (5.8 kB)\n",
      "Collecting contourpy>=1.0.1 (from matplotlib)\n",
      "  Downloading contourpy-1.2.0-cp39-cp39-macosx_10_9_x86_64.whl.metadata (5.8 kB)\n",
      "Collecting cycler>=0.10 (from matplotlib)\n",
      "  Downloading cycler-0.12.1-py3-none-any.whl.metadata (3.8 kB)\n",
      "Collecting fonttools>=4.22.0 (from matplotlib)\n",
      "  Downloading fonttools-4.49.0-cp39-cp39-macosx_10_9_x86_64.whl.metadata (159 kB)\n",
      "\u001b[2K     \u001b[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[32m159.1/159.1 kB\u001b[0m \u001b[31m1.6 MB/s\u001b[0m eta \u001b[36m0:00:00\u001b[0ma \u001b[36m0:00:01\u001b[0m\n",
      "\u001b[?25hCollecting kiwisolver>=1.3.1 (from matplotlib)\n",
      "  Downloading kiwisolver-1.4.5-cp39-cp39-macosx_10_9_x86_64.whl.metadata (6.4 kB)\n",
      "Collecting numpy<2,>=1.21 (from matplotlib)\n",
      "  Downloading numpy-1.26.4-cp39-cp39-macosx_10_9_x86_64.whl.metadata (61 kB)\n",
      "\u001b[2K     \u001b[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[32m61.1/61.1 kB\u001b[0m \u001b[31m961.8 kB/s\u001b[0m eta \u001b[36m0:00:00\u001b[0m \u001b[36m0:00:01\u001b[0m\n",
      "\u001b[?25hRequirement already satisfied: packaging>=20.0 in /Users/jiaruifang/Library/Python/3.9/lib/python/site-packages (from matplotlib) (24.0)\n",
      "Collecting pillow>=8 (from matplotlib)\n",
      "  Downloading pillow-10.2.0-cp39-cp39-macosx_10_10_x86_64.whl.metadata (9.7 kB)\n",
      "Collecting pyparsing>=2.3.1 (from matplotlib)\n",
      "  Downloading pyparsing-3.1.2-py3-none-any.whl.metadata (5.1 kB)\n",
      "Requirement already satisfied: python-dateutil>=2.7 in /Users/jiaruifang/Library/Python/3.9/lib/python/site-packages (from matplotlib) (2.9.0.post0)\n",
      "Collecting importlib-resources>=3.2.0 (from matplotlib)\n",
      "  Downloading importlib_resources-6.1.3-py3-none-any.whl.metadata (3.9 kB)\n",
      "Requirement already satisfied: zipp>=3.1.0 in /Users/jiaruifang/Library/Python/3.9/lib/python/site-packages (from importlib-resources>=3.2.0->matplotlib) (3.17.0)\n",
      "Requirement already satisfied: six>=1.5 in /Library/Developer/CommandLineTools/Library/Frameworks/Python3.framework/Versions/3.9/lib/python3.9/site-packages (from python-dateutil>=2.7->matplotlib) (1.15.0)\n",
      "Downloading matplotlib-3.8.3-cp39-cp39-macosx_10_12_x86_64.whl (7.6 MB)\n",
      "\u001b[2K   \u001b[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[32m7.6/7.6 MB\u001b[0m \u001b[31m5.2 MB/s\u001b[0m eta \u001b[36m0:00:00\u001b[0m00:01\u001b[0m00:01\u001b[0m\n",
      "\u001b[?25hDownloading contourpy-1.2.0-cp39-cp39-macosx_10_9_x86_64.whl (257 kB)\n",
      "\u001b[2K   \u001b[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[32m257.1/257.1 kB\u001b[0m \u001b[31m7.1 MB/s\u001b[0m eta \u001b[36m0:00:00\u001b[0m\n",
      "\u001b[?25hDownloading cycler-0.12.1-py3-none-any.whl (8.3 kB)\n",
      "Downloading fonttools-4.49.0-cp39-cp39-macosx_10_9_x86_64.whl (2.3 MB)\n",
      "\u001b[2K   \u001b[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[32m2.3/2.3 MB\u001b[0m \u001b[31m2.8 MB/s\u001b[0m eta \u001b[36m0:00:00\u001b[0m00:01\u001b[0m00:01\u001b[0m0m\n",
      "\u001b[?25hDownloading importlib_resources-6.1.3-py3-none-any.whl (34 kB)\n",
      "Downloading kiwisolver-1.4.5-cp39-cp39-macosx_10_9_x86_64.whl (68 kB)\n",
      "\u001b[2K   \u001b[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[32m68.1/68.1 kB\u001b[0m \u001b[31m2.5 MB/s\u001b[0m eta \u001b[36m0:00:00\u001b[0m\n",
      "\u001b[?25hDownloading numpy-1.26.4-cp39-cp39-macosx_10_9_x86_64.whl (20.6 MB)\n",
      "\u001b[2K   \u001b[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[32m20.6/20.6 MB\u001b[0m \u001b[31m4.7 MB/s\u001b[0m eta \u001b[36m0:00:00\u001b[0m00:01\u001b[0m00:01\u001b[0m\n",
      "\u001b[?25hDownloading pillow-10.2.0-cp39-cp39-macosx_10_10_x86_64.whl (3.5 MB)\n",
      "\u001b[2K   \u001b[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[32m3.5/3.5 MB\u001b[0m \u001b[31m3.7 MB/s\u001b[0m eta \u001b[36m0:00:00\u001b[0m00:01\u001b[0m00:01\u001b[0mm\n",
      "\u001b[?25hDownloading pyparsing-3.1.2-py3-none-any.whl (103 kB)\n",
      "\u001b[2K   \u001b[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[32m103.2/103.2 kB\u001b[0m \u001b[31m3.7 MB/s\u001b[0m eta \u001b[36m0:00:00\u001b[0m\n",
      "\u001b[?25hInstalling collected packages: pyparsing, pillow, numpy, kiwisolver, importlib-resources, fonttools, cycler, contourpy, matplotlib\n",
      "Successfully installed contourpy-1.2.0 cycler-0.12.1 fonttools-4.49.0 importlib-resources-6.1.3 kiwisolver-1.4.5 matplotlib-3.8.3 numpy-1.26.4 pillow-10.2.0 pyparsing-3.1.2\n",
      "Note: you may need to restart the kernel to use updated packages.\n"
     ]
    }
   ],
   "source": [
    "pip install matplotlib"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "3502c6cb",
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3 (ipykernel)",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.9.6"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}