mlirl_gridworld_5x7.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Python imports\n",
    "import sys, numpy as np\n",
    "# Torch\n",
    "import torch, torchvision, torch.nn as nn, torch.optim as optim\n",
    "# Navigation MDP\n",
    "import navigation_mdp as NvMDP\n",
    "from navigation_mdp.plotting import *\n",
    "from navigation_vis.Raster import Raster as NavGridView\n",
    "\n",
    "# RL utils\n",
    "import rl.planning as Plan\n",
    "import rl.policy as Policy\n",
    "from rl.model import RewardLinear\n",
    "# Utils\n",
    "from utils.utils import *\n",
    "# Matplotlib\n",
    "from matplotlib import cm as cm, pyplot as plt, colors as mplotcolors\n",
    "%load_ext autoreload\n",
    "%autoreload 2"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Gridworld 5x7"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {
    "scrolled": false
   },
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAABQgAAANYCAYAAACIPgCjAAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADh0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uMy4xLjIsIGh0dHA6Ly9tYXRwbG90bGliLm9yZy8li6FKAAAgAElEQVR4nOzdd5hU5fnG8fuZ2V16X0CadAlFRUXF3hVrrInGWKImiiXGFkv8RWKiRo0lKnaNLbGLHbFiAxFEQDpLk96Xtizb3t8fOyzgDDuoM/OemfP9XNdc7sycWZ897rr3PPs+5zXnnAAAAAAAAACEU8R3AQAAAAAAAAD8oUEIAAAAAAAAhBgNQgAAAAAAACDEaBACAAAAAAAAIUaDEAAAAAAAAAgxGoQAAAAAAABAiNEgBAAAAICAMLNBZvac7zoAAOFCgxAAPDOz/c1shJmtNrOVZvalme1pZuea2Rc/4vN0MjNnZnnprBcAAPx8ZvYbMxtjZuvMbJGZDTWz/X3XBQAIJ95EAoBHZtZY0tuSBkp6SVKBpAMkbfRZFwAASB8zu1LSdZIukjRMUpmkAZJ+KWm9x9IAACHFCkIA8GsnSXLOPe+cq3TObXDOvS+pXNLDkvaJrSwoliQzO9bMvjWzNWY2z8wGbfG5Pov9szj2mn1irznPzKaY2SozG2ZmHWOPm5ndY2ZLY6sXJ5hZn0x94QAAhJGZNZF0s6RLnHOvOefWO+fKnXNvOeeuSXD8y2a2OPa7+jMz673Fc8eY2WQzW2tmC8zs6tjjhWb2tpkVx6YTPjezSOy5tmb2qpktM7PZZvbHLT7fXrFVjWvMbImZ3Z3+MwIACAIahADg13RJlWb2tJkdbWbNJMk5N0XVqwpGOucaOueaxo5fL+lsSU0lHStpoJmdGHvuwNg/m8ZeMzL23A2STpbUUtLnkp6PHXdk7DU7xT7fryWtSOPXCgAApH0k1ZU0ZDuPHyqpu6RWksZK+u8Wzz0h6ULnXCNJfSR9HHv8KknzVf27v7Wqs4CLNQnfkjReUjtJh0n6k5kdFXvdvyX92znXWFJXVU83AABCgAYhAHjknFsjaX9JTtJjkpaZ2Ztm1nobxw93zn3nnKtyzk1QdbPvoFr+FRdKus05N8U5VyHpVkl9Y6sIyyU1kvQLSRY7ZlHqvjoAAJBAC0nLY7+Xk3LOPemcW+uc2yhpkKRdY6sQperf5b3MrLFzbpVzbuwWj7eR1DG2OvFz55yTtKekls65m51zZc65WarOH6dv8bpuZlbonFvnnPsqJV8xACDwaBACgGexxty5zrn2qv7rf1tJ9yY61sz2NrNPYmNBq1W9yrCwlk/fUdK/YyNGxZJWSjJJ7ZxzH0t6QNJgSUvM7NHYNREBAED6rJBUuD2biplZ1Mz+aWYzzWyNpDmxpzb97j9F0jGS5prZp5suLyLpTklFkt43s1lmdl3s8Y6S2m7KBbFscIOqVxlK0vmqniyYamajzey4n/m1AgCyBA1CAAgQ59xUSU+pulHoEhzyP0lvSurgnGui6usU2qaXJzh+nqpHj5pucavnnBsR+/fd55zbQ1JvVb8hiLv2EQAASKmRkkolnZjsQEm/UfXGJYdLaiKpU+xxkyTn3Gjn3C9VPX78umIjwbEVh1c557pIOl7SlWZ2mKpzwewf5IJGzrljYq+b4Zw7I/b5bpf0ipk1SMlXDQAINBqEAOCRmf3CzK4ys/ax+x0knSHpK0lLJLU3s4ItXtJI0krnXKmZ7aXqNw6bLJNUJanLFo89LOn6TRc0N7MmZnZa7OM9YysS81V9bcNSSZVp+UIBAIAkyTm3WtJfJQ02sxPNrL6Z5ceuRXzHDw5vJGmjqlcd1lf1pUIkSWZWYGZnmlkT51y5pDWK/R43s+PMrJuZ2RaPV0r6WtIaM7vWzOrFVij2MbM9Y6/7rZm1dM5VSSqO/avIBgAQAjQIAcCvtZL2ljTKzNarujE4UdUXF/9Y0iRJi81seez4iyXdbGZrVf3moubi4c65Ekm3SPoyNjbU3zk3RNUrAF6IjSZNlHR07CWNVX3doVWS5qr6zce/0vnFAgAAyTl3t6QrJd2o6j/wzZN0qapXAW7pGVX/jl4gabKqc8KWzpI0J/Y7/iJJv4093l3Sh5LWqXrF4oOx6xhXqnpFYV9JsyUtl/S4qlcnStIASZPMbJ2qNyw53TlXmoqvGQAQbFZ9rVoAAAAAAAAAYcQKQgAAAAAAACDEaBDmADMbYGbTzKxoix3KAABACJADAAAA8HMxYpzlzCwqabqkIyTNlzRa0hnOucleCwMAAGlHDgAAAEAqsIIw++0lqcg5N8s5VybpBUm/9FwTAADIDHIAAAAAfra82p486pAGbsXK7N7V/psJG4c55wb4riON2ql617NN5qt6R9TNzAZJuinRiwsldUpPXQAA4CeYkFdfZeXrbTsPT54DJLIAAABZ5EdmgdDJ9l5VUPtUtTYIV6ys1NfDdsxULWkRbTOj0HcNaZbofxpbz407N0jSoEQv7mTmmtmpKS8qm33gXtFRDc72XUbgDFv/jAa0Gui7jMB5b+lDOrrTFb7LCJShc+7RgF7X+y4jcN6bfJuO3D1hfya03h/7Nx2+/z98lxE4dcc9+GMOT54DJLLAj0QWSIwsEI8ckBhZIB45IDGyQGI/MguETrb3qoLap6q1QegkVakqQ6XgJ5ovqcMW99tLWuipFgAAkFnkAAAAECr0qtKDaxBmv9GSuptZZzMrkHS6pDc91wQAADKDHAAAAICfrdYVhJJTpaMrG2TOuQozu1TSMElRSU865yZ5LgsAAGQAOQAAAIQPvap02I4R4/jL2CBYnHPvSnrXdx0AACDzyAEAACBM6FWlByPGAAAAAAAAQIglGTHmwo8AAAAAAAAIDnpVqZdkxNip0rFsEwAAAAAAAP7Rq0oPRowBAAAAAACAENuOEWO6sgAAAAAAAAgGelWpl3QX40pOOgAAAAAAAAKAXlV6MGIMAAAAAAAAhBgjxgAAAAAAAMga9KpSL/mIMTvDAAAAAAAAIADoVaUHI8YAAAAAAABAiG3HiDEAAAAAAAAQDPSqUi/JiLFjZxgAAAAAAAAEAr2q9GDEGAAAAAAAAAix2keMnVRJUxYAAAAAAABBQK8qLZLuYsxcNwAAAAAAAIKAXlV6MGIMAAAAAAAAhFiSXYxNlbLMVAIAAAAAAADUil5VOiQfMWauGwAAAAAAAAFAryo9GDEGAAAAAAAAQizJiLFYtgkAAAAAAIDAoFeVeklHjDnpAAAAAAAACAJ6VenBiDEAAAAAAAAQYklHjKscXVkAAAAAAAAEA72q1GPEGAAAAAAAAFmBXlV6MGIMAAAAAAAAhFiSFYSmSnqIAAAAAAAACAB6VenBNQgBAAAAAACQNehVpR7XIAQAAAAAAEBWoFeVHklXEALp1qhZQ135+EDtceQuWrN8rZ644X/65PkvfJfl1QkXHq4jztxfnXq31/CXv9JdFz3uuyTv8gvydOntp6vvAT3UqFkDLZy9TE/d+obGfDzZd2leXXPPmeq7b3fVrVeglcvX6JVHPtGwF0f5Lisw2u7YQg+//kd98f5E3XHdy77L8e7OR85Vz53bq7KySpK0fOkanX/KA56r8u/gw3rprHMPUKvWjbVq5Xrdcetbmjhhnu+yQoMcEI8ckBhZIDGyQO3IApuRA7aNLAAkbRCaKh1z3Uivyx44XxVlFfrVDr9X176ddMvb12vW+DmaO3m+79K8WbFolf53x5vqd/jOKqib77ucQIjkRbRswSr9+aR7tHT+Ku15eG/d8NgFGnjwP7Rk3krf5Xnz4oMf6t5rX1B5WaXad2ml21+4WDMnLVDRxPD+/GzpkhtP0PSJC3yXESgP3PGu3nt9rO8yAmP3fp31+4sO1T9uek1TpyxU8xaNfJcUOuSAeOSAxMgCiZEFakcW2Bo5IB5ZIBvRq0qHWs+ok1SlSFbfEGx169fR/qf011N/fUGl60s16cupGvnmGB1+1kG+S/Pqyze/0ci3x2rNynW+SwmMjSVleu5f72jJvJVyzunrDyZqyfcr1G2XHX2X5tX3M5aovKxSkuTkJCe16djCc1XBcNDRu2j92lKN+2qm71IQYOecd6CefepzTZm8UM5JK5av1Yrla32XFRrkgMTIAYmRBRIjC2wbWQDbgyyQfbK9VxVUwa0ModBupzaqqqzSghmLah6bOWGOOvZq77EqZIOmLRupXZdWmjttUfKDc9wlN5+iIZP/qcc/ul4rl67R6E+m+C7Ju/oN6uisSw/Xo3e+67uUwDnv0sP08kd/1j1PnKdd9ujkuxyvIhHTTr9oo6ZN6+vp5wfq+Vcv06V/OkoFBVyBJVPIAfg5yAKbkQXikQUSIwdsjSwAbJb0u54LPyKd6jWsq/WrS7Z6bP3qEtVvVM9TRcgG0byIrn3wd/rwpa80v2iJ73K8G/zXV/XQoNfUc/dO2rl/V5WXVfguybuzLztCw14bo+WLV/suJVCeuP8DzZ21TBXllTr4qD66+Z4zNPA3D2vR/FW+S/OiWbMGys+P6oCDe+qKS59VRUWlbr7tNJ15zv76z2PDfZcXCuQA/FRkga2RBeKRBeKRA+KRBbIXvarUq33E2FXPdWfzDcG2YV2p6jfe+k1Ag8b1VLJ2g6eKEHRmpmsGn6vysgoNvv5F3+UERlWV06Qxs1W4Q1Md+9v9fJfjVZdftNFu+3TVkGe+9F1K4EyduEAbSspUXl6pD94er0nj52mv/br7LsubjRur30C//uporVyxTmtWb9CrL47S3v27eq4sPMgB+CnIAomRBTYjCyRGDohHFshO2d6rCirWzcKrBdMXKZoXVbtuO2hB0WJJUpddOoX6wuSo3RX3/FbNWjbW//1msCorqnyXEzjRvIja7Bju6w7tsmdntW7bTM98+GdJUr36BYpEInqgaytdetpgz9UFjHMyC+9fX9etK9XSJWuqL2QDL8gB+CnIArUjC5AFtlvIc4BEFgC2lLR1WSXL6huCrbRko754bZTO+duvVbd+HfXet4f2/eWe+vDZT32X5lUkGlF+nXxFIpHNH0eD+5eGTLnsjjPUYacddNNvH1JZabnvcrxr0qKhDjqur+rWL1AkYtr9wB46+PjdNH7kDN+leTX05dH63dH/0iWn3K9LTrlf77z4tb7+bJr+8oenfJfmVYOGdbXHPl2VX5CnSDSiQ4/eWTvv3lFjRhb5Ls2rYUPH68RT+qlp0/pq2LCuTj5tL301ItznJJPIAYmRA7aNLLA1skBiZIF45IBtIwtkJ9+9plzsU9W6gtBJqszxfUzMrK6kzyTVUfX5eMU5d5OZdZb0gqTmksZKOss5V+av0tx1/yWP66onBuqlJY9r7Yp1+vfFj4V+5cBvrj1BZ91wUs39w8/YT8/eOkTP3fq6x6r8atW+uY495wCVlZbr+Ym31Tx+3zXP65NXR3uszCPndOxv99Olt5ymiJmWLFilR/7+hr76YJLvyrzaWFqujVu8aSwtKVP5xnKtXrXeY1X+5eVFdO7AQ9WhU6GqqpzmzVmuQVe9oPlzV/guzavnnvpCTZrU11P/G6iysgp9+skU/ffZL3yXFSrkgHjkgMTIAgmQBRIiC8QjB2wbWSD75HqvylefihFjaaOkQ51z68wsX9IXZjZU0pWS7nHOvWBmD0s6X9JDPgvNVWtXrdOgk+/0XUagPHfr66F/E/BDS+ev1IDWF/suI1BWr1yvP5/OmEwyzz34ke8SAmF1cYkuO/sx32UETmVlle67+z3dd/d7vksJLXJAPHJAYmSBeGSB7UMWIAfUhiyAAPLSp0rSILRAX0AxFZxzTtK62N382M1JOlTSb2KPPy1pkGgQAgAAAAAAeJTbvSpffaqkI8ZV2b9ss9DMxmxx/1Hn3KNbHmBmUUnfSOomabCkmZKKnXMVsUPmS2qXiWIBAAAAAACQWA70qgLZpwrDiPFy51y/2g5wzlVK6mtmTSUNkdQz0WHpKA4AAAAAAAChEcg+VdIGYaUL7g4rqeacKzaz4ZL6S2pqZnmx7mx7SQu9FgcAAAAAAIDQ9Koy2adKMmJsOb0zjCSZWUtJ5bGTXk/S4ZJul/SJpFNVvUPMOZLe8FclAAAAAAAAcr1X5atPFYYR42TaSHo6Nt8dkfSSc+5tM5ss6QUz+4ekbyU94bNIAAAAAAAA5DwvfaqkDcKqHN4ZRpKccxMk7Zbg8VmS9sp8RQAAAAAAANiWXO5V+epTJd3FOJeXbQIAAAAAACB70KtKD84oAAAAAAAAEGLJNykJyc4wAAAAAAAACDZ6VemR/BqELDIEAAAAAABAQNCrSj3OKAAAAAAAABBitY8YO6kyh3eGAQAAAAAAQPagV5UeSUaMTVVirhsAAAAAAABBQK8qHWi5AgAAAAAAACGWZBdjlm0CAAAAAAAgGOhVpUfSXYwrWWQIAAAAAACAgKBXlXqcUQAAAAAAACDEkowYm6ocF34EAAAAAACAf/Sq0oMRYwAAAAAAAGQNelWpxxkFAAAAAAAAQizpLsZV7AwDAAAAAACAAKBXlR5JRoxNlWKuGwAAAAAAAEFAryodaLkCAAAAAAAAIcaIMQAAAAAAALICvar02I5djFm2CQAAAAAAgGCgV5V6tFwBAAAAAACAEKt9xNgZyzYBAAAAAAAQCPSq0iP5iDEnHQAAAAAAAAFBryr1km9Swlx3oJnZk5KOk7TUOdfHdz0AACCzyAIAACBM6FWlBy3X7PeUpAG+iwAAAN48JbIAAAAAfoYkI8bGss2Ac859ZmadfNcBAAD8IAsAAIBwoVeVDuac2+aTbXo3c+c9f0gGy0m9W3cd8o1zrp/vOtIp9qbg7W2OFZkNknRToqf6SRqTproAAMCPV691e21YPO9Hzc2QBQAAyB0/JQuESbb3qoLap0q6SQlygHODJA1K+JyZO+Tw2zJZTeB98uH1OuD4O3yXETifv/Vn7Xvanb7LCJwRL1+jvc76l+8yAuXrZ6/W7hfc5buMwBn7+FXa9dK7fZcRKOMfuFJ9ruKcxHkuDeeELPCjkAUSIwvEIwckRhaIRw5IjCywDenIAkASyXcx5jKFAAAAAAAACAh6VamXZBdjU5VjVSsAAAAAAAD8o1eVHrRcs5yZPS9ppKQeZjbfzM73XRMAAMgcsgAAAAB+rqQjxlX0EAPNOXeG7xoAAIA/ZAEAABA29KpSr/YRYydVsmwTAAAAAAAAAUCvKj1ouQIAAAAAAAAhlnzEmK4sAAAAAAAAAoJeVeptxy7GLDIEAAAAAACAf/Sq0oMzCgAAAAAAAIRY0hHjSrFsEwAAAAAAAMFAryr1kowYM9cNAAAAAACAYKBXlR6MGAMAAAAAAAAhlmTEmAs/AgAAAAAAICjoVaVD0msQVjHXDQAAAAAAgICgV5V6tFwBAAAAAACAEKt9kxInVXLhRwAAAAAAAAQAvar0SD5izFw3AAAAAAAAAoJeVepxRgEAAAAAAIAQq33EWKYqlm0CAAAAAAAgAOhVpQe7GAMAAAAAACBr0KtKvdCPGJtZBzP7xMymmNkkM7s89vggM1tgZuNit2N81woAAAAAAIDc5atPlWTEWGFYtlkh6Srn3FgzayTpGzP7IPbcPc65f3msDQAAAAAAADEh6FV56VOFfhdj59wiSYtiH681symS2vmtCgAAAAAAAInkcq/KV58qd8/oZoVmNmaL2x+2daCZdZK0m6RRsYcuNbMJZvakmTXLQK0AAAAAAADIXYHsU9W+gtDlxM4wy51z/ZIdZGYNJb0q6U/OuTVm9pCkv6t69erfJd0l6by0VgoAAAAAAIBty/5eVSD7VMmvQRiCnWHMLF/VJ/2/zrnXJMk5t2SL5x+T9Lan8gAAAAAAAKBw9Kp89KnCMGJcKzMzSU9ImuKcu3uLx9tscdhJkiZmujYAAAAAAACEh68+1XZsUpLbXVlJ+0k6S9J3ZjYu9tgNks4ws76qbk7PkXShn/IAAAAAAACwSY73qrz0qZKPGOf2SZdz7gsp4drUdzNdCwAAAAAAALYt13tVvvpUoR8xBgAAAAAAAMKMEWMAAAAAAABkDXpVqZdkxDjrt45GwL375pVb3S8oyNMbb32r+wd/4KmiYNihVWNdedER6vOLtiorr9TwEdN0/2Mfq7LK+S7Nq47tmuvq8w9Xjy6tVbymRA88+6k+G13ku6yMOu3wvjrugN7q2r5Q7381VTc/NqzmuT177ahrzj5MO7RopIkzF+nmx97T4hVrPVabOb8+pK+O37eXurUr1Hujp2nQf6rPS140olt/f4x6dWyttoVN9Ps7X9I30+d7rjYzTj9wV52wd291b9NCQ8dO01+fe1+StHOnHXTJsfuqV4fWqqyq0pii+br9leFavma954oz44z9dtWJ/arPy7vfTtONL1afly6tm+u20weofYsmkqTJ85fqtjc+0awlK32WGwpkgcTIAvHIAdXIAvHIAYmRBRIjC2Q/elXpwYgxvDrmhLtrbif/6n6VlVXo08+m+i7LuysvOkLFq0t04jkP6rzLn1bf3h104jG7+S7Lq2jEdPs1J+rLb2ZqwO8e0O2PvK+bLjtWHdo0811aRi0rXqcn3xiltz7besOqJg3r6fY/nqBHXv1Sh188WFPmLNEtlxznqcrMW1a8To+/M0pvfDkp7rlxMxboxieGalnxOg+V+bNs9Xo9NmyUXv9q63PSuH5dvfrldzp60BM6+qYnVFJappvPPNJTlZm3bPV6PfLhKA35elLc41c887b2++tDOuCmhzV88kz968xjPFUZLmSBxMgCWyMHbEYWiEcOSIwskBhZAEgsaYOwSpbVN2SPgw7ooVXFJZrw3TzfpXjXpnUTffzFVJWVV2pl8Xp9PXa2Ou9Y6Lssrzq2a6HC5g31wjvfqMo5fTNpnr6btkADDuzlu7SMGj6mSJ+OLdLqdaVbPX5Iv26atWCFPho9XWXllXrstRHqvmNLdWzT3FOlmfXxt0UaPm6mVq/fsNXjFZVV+t9H32pc0UJVuXCtuvlofJE+mTBTxeu3/l75cvIcfTBuhtaXlqm0vELPfzZefbu09VRl5n04sUgfT5qp4pKtz8va0o1auGqNJMlMqqxy6lDY1EeJoUYW2IwssDVywGZkgXjkgMTIAomRBXKD715TLvapar8GoWOuG5lz5BE76/0PJiY/MAReeesbHXZAT3373Tw1alhXe+/RWY//90vfZQWPSV06hPfN0pa6tCvUjO+X1twvLavQgqWr1aVdC81dxFgEtm2Pbu00c9EK32UExoi/D1T9ggJFzPTAsBG+ywkdssBmZIHtQA7YClkAPxVZYGtkgSxAryotkm5SAmRCq5aNtesuHXTn3WndtTtrjJs4T8cfuYvee/Fy5UUjGvrRRH3+1QzfZXk1d+FKrVpdojNP2FMvvPON9ujdQbv16qCxE7/3XVog1K+br1Vrt/6r+bqSjapft8BTRcgG3dsW6sIB/XX5Y2/6LiUw9v2/h1SvIE8n9OulRaty/7pdQUIW2BpZYGvkgOTIAvgpyALxyAIIq1pHjJ2qu7LZfEN2OPKIPpo4ab4WL17tuxTvzKS7/naaPh05Q0eedq+OPfN+NWxYRwPPPch3aV5VVlbpujtf1767d9Hbjw7UGcf108cjp2npyvBdTyaRktJyNai39RuABvUKVFJa5qkiBF2HwiZ6cOBJuuPV4fp25gLf5QTKhrIKvTRygm49/Sg1b1jPdzmhQRbYjCwQjxyQHFkAPxZZYNvIAsGW7b2qoEp+DcIAnLxcPPHY2pFH9NGw9xkpkqTGDeupdcvGeu2dsSqvqNSataUa+uFE9d+js+/SvJv5/XJdMuhFHX3+YF1x66tq26qJJhct8l1WIMxasFzdO7SsuV+3IE/tWzXVrAWMiyBem2aN9Milp+jR90bp7dFTfJcTSBEz1S3IV6vGDX2XEhpkgc3IAomRA2pHFsCPQRZIjiwQbL57TbnYp0qygtD/icvVE4/Nevdqp8IWDTWcHQslSavXbtDCxcU68ei+ikZMDRvU0YBD+6ho9jLfpXnXdcdCFeRHVacgT2cc308tmjXUu8Pjd6vLZdGIqSA/qkjEFIlEVJAfVTRiGv5Nkbq2L9Qh/bqrID+qC07cRzPmLQvNNYeiEVNBXlQRiyhq1R9HI9W/A/LzoirIi8Z9nOs2nZNoxBS1SM3HrZo00GOXnaoXPx+vl7+c4LvMjKs5L2aKRjafl32676hftG2piJka1CnQn48/SGtKSjVraTh+hnwjC2yNLJAYOaAaWSAeOSAxskBiZIHsl+29qqDiGoTw7qgj+ujzL6drwwbGHzb5y21v6I8XHKIzT9lblVVV+va7ebr/iU98l+XdgAN76/hDd1ZeXkTjpyzQ5X9/WeUVlb7Lyqjzftlfvz9p35r7x+zXS48NGaHHhozUtfe/qWvOOkx/u+hoTZq5WDc++I7HSjPrgmP768IT9qm5f+w+vfTImyP1yFsjNeTv56ptYRNJ0oNXnFL9/HWPa9GKNT5KzZjfH7W3Bh6z+Zwct1dPPfTuSDlJHVo21UVH99dFR/eveX6fqwd7qDLzLjx8b1185ObzcvwePfXg+yNVtHiFrj/pEO3QpKFKyys0cd4SXfT4EJWF7P8xvpAF4pEF4pEDqpEF4pEDEiMLJEYWABJL2iB0Ae5uIjfc/e9hvksInKLZS/XHv7zou4zAGfzcpxr83Ke+y/DqsSEj9diQkQmfGz3pe/3quv9kuKJgeOSt6jcBiRx3/RMZriYYHh76lR4e+lXC5x7ZxuNh8OD7X+nB9xN//e9PCO8GEL6RBeKRBeKRA6qRBeKRAxIjCyRGFsgN9KpSL2mDsEqcdAAAAAAAAAQDvarUS7pJCQAAAAAAAIDcVesKQucU6AsoAgAAAAAAIDzoVaUH1yAEAAAAAABA1qBXlXqMGAMAAAAAAAAhlmQFobFsEwAAAAAAAAFBryodGDEGAAAAAABA1qBXlXqMGAMAAAAAAAAhVvsuxmJnGAAAAAAAAAQDvar0qH3E2FVvHw0AAAAAAAB4R68qLRgxBgAAAAAAAEIs6SYlVWLZJgAAAAAAAIKBXlXqJb0GITvDAAAAAAAAINNGFoUAACAASURBVAjoVaUHI8YAAAAAAABAiCUZMTZ2hgEAAAAAAEBA0KtKh6TXIGRnGAAAAAAAAAQFvarUY8QYAAAAAAAACLHtWEHIsk0AAAAAAAAEA72q1Kt9F2PHSQcAAAAAAEAw0KtKD0aMAQAAAAAAgBBLOmLMzjAAAAAAAAAICnpVqccuxgAAAAAAAMga9KpSjxFjAAAAAAAAIMTYxRgAAAAAAABZg15V6tW+i7GMkw4AAAAAAIBAoFeVHowYAwAAAAAAACGWfMQ4E1UAAAAAAAAA24FeVerV3iB0zHUDAAAAAAAgIOhVpQUjxgAAAAAAAECIJR0xZt0mAAAAAAAAAoNeVcolvwYhyzYDzcw6SHpG0g6SqiQ96pz7t9+qAABAJpADAABAGNGrSr3kKwgRdBWSrnLOjTWzRpK+MbMPnHOTfRcGAADSjhwAAACAn207VhBmogz8VM65RZIWxT5ea2ZTJLWTxBsDAAByHDkAAACEEb2q1DNXy1mt07Wda3/rxRksJ/VmnX7jN865fr7ryAQz6yTpM0l9nHNrtnhikKSbEr2mn6Qx6S8NAABspzqd22rjrAU/em5mmzmg+slBIgsAAJAVfmoWCIts71UFtU8V+hHjbV27x8yaS3pRUidJcyT9yjm3yledyZhZQ0mvSvpT3JsC5wZJGrSNF7qeN9yd5uqyy5Rbr1SPv3FOfmjaTVeq2+13+S4jcIquvUqd7/2X7zICZfafrlbHh+7wXUbgzB34Z3V88p++ywiUueddp87/vcV3GcFz4+Af/ZJac4BEFviRyAKJkQXikQMSIwvEIwckRhbYhp+QBZA7fPWpIrU+6yQ5y+5bcpuu3dNTUn9Jl5hZL0nXSfrIOddd0kex+4FkZvmqflPwX+fca77rAQAAmUMOAAAAoZLtvarkvPSpQn8Nwlqu3fNLSQfHDnta0nBJ13oosVZmZpKekDTFOcefugEACBFyAAAACKNc7lX56lPVvoJQinVms/j2I8Su3bObpFGSWsf+o2z6j9Pqx322jNlP0lmSDjWzcbHbMb6LAgAAGUEOAAAA4eO715SDfaowXIOw0My2vPb2o865R3940A+v3VP9B/ngc859ISk7igUAAClFDgAAAMg6gexTJWkQmtz2zUcH2fJku8Ns49o9S8ysjXNukZm1kbQ03YUCAAAAAACgNlnfqwpknyr0I8a1XLvnTUnnxD4+R9IbyT8bAAAAAAAA0sp3rykH+1RhGDFOZtO1e74zs3Gxx26Q9E9JL5nZ+ZK+l3Sap/oAAAAAAAAQDl76VLU3CJ2yfdlmUkmu3XNYJmsBAAAAAABALXK8V+WrT5V8BeGP3GEFAAAAAAAASBt6VSmX/BqEAAAAAAAAAHLWdlyDMHeXbQIAAAAAACDb0KtKNUaMAQAAAAAAkD3oVaUcI8YAAAAAAABAiLGCEAAAAAAAANmDXlXK1d4gdJJyeOtoAAAAAAAAZBF6VWnBiDEAAAAAAAAQYklHjB3LNgEAAAAAABAQ9KpSj2sQAgAAAAAAIHvQq0o5RowBAAAAAACAENuOFYRc+BEAAAAAAAABQa8q5ZI2CI1lmwAAAAAAAAgIelWpx4gxAAAAAAAAEGK1ryB04sKPAAAAAAAACAZ6VWmRZMTYmOsGAAAAAABAQNCrSgdGjAEAAAAAAIAQ245djDNQBQAAAAAAALA96FWlHA1CAAAAAAAAZA96VSnHiDEAAAAAAAAQYqwgBAAAAAAAQPagV5VytTcIndgZBgAAAAAAAMFAryotGDEGAAAAAAAAQizpiLGxbBMp9Jv+u+qk3Xtrp9Yt9M6Eabrh1ffjjrn40P667LB9dN6Tr2rkzO89VJl5Z+65q07etbd2atVCb0+cpuvfrD4v7Zo01seXn6/1ZWU1xz7+5Rg9+PkoX6Vm1Fm79dXJO/dSj8JCvTVlmq4dOqzmubp5ebr+kAN1dI8eyo9GNGXpMv3m+Zc8VpsZZ+/SV6f06q0eLQr11vSpuuaD6nPyyx6/0C2HHlFzXMRM9fLzdfzzz2ri0qW+ys2Yc3rvplN79FGPFoV6s2iqrv5kaM1zx3btoSv77acdGjTSonVrdMfXn+v9OUUeq82Mc3rurlO79VGPZi315qwpuvqLd2ueO737Lhq4S3+1rNdAo5fM1zVfDNXSDes8VpsZBZGobt7zKO2/Q2c1KairuWtX6c7xw/XpolmSpH1bd9LN/Y5S2waNNW7FQl0z8i0tKFnjuercQxZIjCwQjxyQGFkgHjkgMbJAPLJA7qBXlXpcgxAZtWzNej38ySjt372j6uTHf/t1aN5ER/XurqVrcv+X05aWrl2vBz8fpQO6dlSdvPjzsuftD6rShe+Hccm6dRo8YpQO6NxJdX9wXm456ghFI6YBTzyl4tJS9WzV0k+RGbZk/To98PUoHdix41bn5I1pU/XGtKk190/p2VuX7dU/598QbLKkZJ3uHztSB3bovNV5ad2goe499Fj9/r0hGj5vtg7dsYsePOIE7fffR7WitMRjxem3pGSd7h8/Uge266y60c3nZO/WHXTNHgfq9Pde0Jw1K3XT3ofr/oOP16+HPu+x2syIWkSLStbq1x8+p4XrV+uQtt30wP4nacC7j6ukokwPHXCyrh/1rj5cMENX7XKQ7t//JJ38/tO+y845ZIHEyALxyAGJkQXikQMSIwvEIwvkkHD9SswIRoyRUR9MLtJHU2aquKQ04fM3Hn+I7hr2ucorKzNcmV8fTC3SR9NmqnhD4vMSVu/PKNKHRTNVvGHDVo93bt5Mh3XrohuHfaiVGzaoyjlNWpL74VeShs0s0gezilRcWvv3yik9e+m1qZMzVJV/782eoffnFKm4dOvvlTYNGmlNWamGz5stSfr4+1kqqShXxyZNfZSZUe/Nna73v5+h4o1bn5PDd+ymd+dM04zi5SqvqtJ940ao/w47asdGuX9ONlSW69/ffa4F61fLSfp4YZHmr1utnZvvoAHte2jG6uV6d95UlVVV6t7vPlfPpq3UpXEL32XnHLJAYmSBeOSAxMgC8cgBiZEF4pEFgG2jQYjAOKpPd5VXVumz6XN8lxI4n1x+gT790wW69YQj1axeXd/leLdrmx20YM1aXb7/Pvr60oF653dn66iduvsuKzDaNWqkvdq112tTJvkuxbsJyxaraNVKHd6xqyJmOrJTN5VVVmrKimW+S/PGYrea+7E7PZoV+ijHq8K6DdS5cXNNX71c3Zu21JTizQ2GDZXlmruuWDs1Cd958YkssG1kgc3IAcmRBaqRAxIjC2xGFgA24xqECIT6Bfm64oj9dP5Tr/kuJVBWlWzQKY/9T1MWL1XT+vX016MP1Z0nH60L/jvEd2letWnUSD1aFmrY9Bna78FHtFu7tnrslBNVtHyFZq5c6bs8707u2VujFy7Q/DVcL6XKOb02fZLuO/w41YnmqbyyUhd/8KY2VJT7Ls2bT+bP0gMHn6Dnpo3T7DWrdHnf/VTlnOpF832XllF5FtE9+56gV2d9p1lrVqh+Xr5W/mDcbG15qRrkFXiqMHzIAomRBeKRA5IjC1QjByRGFqhGFshu9KpSjxWECIRLD9tHb46bogWrwh1ifqikvFwTFy1RpXNasb5Efx/6sQ7o2kkNCsL9S6q0okJllZUaPOIrlVdV6et58zXq+3nav3NH36UFwkk9e+nVkK8Y2GS/dh11ff+D9Os3X1C3R+/Sr958QbcfdJR6tWjluzRvvlw0V/d8+4UePvREjTjtIs1fu1rrysu0qGSt79IyxiTdve8JKq+q1E1jqi/uX1JRrob5dbY6rmF+Ha2vKEvwGZAOZIHEyALxyAHJkQWqkQMSIwuQBYBEtmOTEkt6CPBz9e/aQTs0bqTT995VktS8QT3dffqxeuKz0Xr88zGeqwuOTX8ksZD/WE5dGu6xkNrs0aatWjdoqKEzpvsuJRB6F7bSqEXz9d2yJZKqR42+XbpI+7fvqMkrwnO9qh96Zuq3embqt5Kkzo2b6bJd99G0VeH5ubp972NVWLeBfjf8RVW4KknSjOJlOrnLzjXH1Ivmq2PDZpq+ermvMkOHLLB9yALkgGTIApuRA7aNLEAWyHr0qlKu9hWELgduCJRoxFSQF1UkYopaRAV5UUUjpvOeeFUn3PeMTn7gOZ38wHNauma9Br3xof43arzvkjMiaqaCaFQRM0UjERVEo4qaaZd2O6hzi2YySU3r1dWNAw7WqDnztG5jOP6Ktem8RCOR6u+d2HkZPX+BFq1Zq4v676WomXZv11Z77dhBn8+e47vktNvqe8U2f69sckrP3nqvaLrWl4drdCZqpjo156X646iZxi9dpL12aFezUqB3i1baq037UFx7aFvnpE40qp2aVl9Lp22DRrpt3wF6cvI3WlO20XPFmfGPPQeoW5NCXfDpS9pYWVHz+LD507VTk5Ya0KGHCiJR/XHn/TW1eKlmrVnhsdrcRBZIjCwQjxyQGFkgHjkgMbJAYmSBHOC7z5SjfarkKwiBFLro4L116WH71Nw/YbeeeuCjkRr88VdbHVflqrRmw0aVlIUj2Aw8cG9ddtDm8/LLXXrq/k9HavbyVbry0P3UvEF9rdtYphGz5urKV9/1WGlmXbJvf/1xv83n5cTevXTflyN135cjddFrb+jWAUfowr330sI1a3TNO+9p1spVHqvNjEv36q8/9d+35v5JPXvp3q9G6N+jRqogGtWxO+2kge+85bFCPy7bYx9d0W+/mvsn79Rb94z5UveOGaF7x4zQQ0eeoMJ6DbSytESDx36lz+fP8Vdshly26766Yrf9a+6f3K2P7vn2Cz05eYzuO+h4dWzUVOvKy/Ry0Xe669vPPVaaOe3qN9aZ3XfXxsoKfX3S5TWP/2X0UL0xZ5Iu/vw1/a3fUbpnnxM0bsVCXfbl6x6rzV1kgcTIAvHIAYmRBeKRAxIjC8QjCwDbZs5tu31Zp0MH1+7KKzJYTurNvvKqb5xz/XzXEVT9zNz6G+72XUagTLn1SvX4G+fkh6bddKW63X6X7zICp+jaq9T53n/5LiNQZv/panV86A7fZQTO3IF/Vscn/+m7jECZe9516vzfW3yXETgLbxysjbMWZGxuhiwQjyyQGFkgHjkgMbJAPHJAYmSBxDKdBbJNtveqgtqnYhdjAAAAAAAAZA16Vam3HZuUZKAKAAAAAAAAYHvQq0q52jcpAQAAAAAAAJDTWEEIAAAAAACA7EGvKuVqbRCaY64bAAAAAAAAwUCvKj0YMQYAAAAAAABCbDtGjNlZGwAAAAAAAAFBryrluAYhAAAAAAAAsge9qpRjxBgAAAAAAAAIsaQrCLnwIwAAAAAAAIKCXlXqMWIMAAAAAACA7EGvKuUYMQYAAAAAAABCrPYVhI5lmwAAAAAAAAgIelVpwYgxAAAAAAAAsge9qpQL/YixmT1pZkvNbOIWjw0yswVmNi52O8ZnjQAAAAAAAMh9vvpUyRuELstvyT0laUCCx+9xzvWN3d7drs8EAAAAAACA9PLda8rBPlXSEeNcn+t2zn1mZp181wEAAAAAAIDkcrlX5atPFYYR40IzG7PF7Q/b+bpLzWxCbGlns7RWCAAAAAAAgDAIZJ8qDA3C5c65flvcHt2O1zwkqaukvpIWSborrRUCAAAAAAAgDALZp2IX4wScc0s2fWxmj0l622M5AAAAAAAA2CRkvapM9KnCsILwRzOzNlvcPUnSxG0dCwAAAAAAAKRLJvpUta8gdLl94UdJMrPnJR2s6hnw+ZJuknSwmfVVdU96jqQLvRUIAAAAAACAajneq/LVpwr9iLFz7owEDz+R8UIAAAAAAACQXA73qnz1qRgxBgAAAAAAAEIs9CsIAQAAAAAAkEXoVaVcrQ1CU27PdQMAAAAAACB70KtKD0aMAQAAAAAAgBBjxBgAAAAAAADZg15VytXeIMzxraMBAAAAAACQRehVpQUjxgAAAAAAAECIMWKc5cysrqTPJNVR9X/PV5xzN/mtCgAAZApZAAAAhA69qpSjQZj9Nko61Dm3zszyJX1hZkOdc1/5LgwAAGQEWQAAAIQLvaqUS94gRKA555ykdbG7+bEbPyoAAIQEWQAAAAA/l1VnysTqtengupx7ZQbLSb3J/7zyG+dcP991pJOZRSV9I6mbpMHOuWt/cMAgSQlHjfpJGpPm+gAAwPZr06upFk1aZT/mNWQBAAByx0/JAmGS7b2qoPapGDHOAc65Skl9zayppCFm1sc5N3GLAwZJGpTwxWauxyt/y0CV2WPaqTdp1zdv9F1G4Iw/4R/ad9iffZcROCOOukNHfnK57zIC5f1D/q1fjfiD7zIC56V9H9XvR5/lu4xAeWzPZ3XduJN9lxE8v/n4R7+ELJBaZIHEyALxyAGJkQXikQMSIwtsw0/IAqFDryrl2MU4hzjniiUNlzTAcykAAMADsgAAAAB+itobhC4HbjnOzFrGVgvIzOpJOlzSVL9VAQCATCELAACAUPHdZ8rRPlXSEWMLcPGQJLWR9HTs2kMRSS855972XBMAAMgcsgAAAAgVelWpxy7GWc45N0HSbr7rAAAAfpAFAAAA8HOxSQkAAAAAAACyB72qlGPEGAAAAAAAAFmDXlXqsYIQAAAAAAAA2YNeVcrVvosxAAAAAAAAgJxW+wrCgG/BDAAAAAAAgBChV5UWtTYILXYDAAAAAAAAfKNXlR6MGAMAAAAAAAAhxiYlAAAAAAAAyB70qlIuaYOQraMBAAAAAAAQFPSqUo8RYwAAAAAAACDEGDEGAAAAAABA9qBXlXI0CAEAAAAAAJA96FWlHCPGAAAAAAAAQIjVvoLQceFHAAAAAAAABAS9qrRgxBgAAAAAAADZg15VyjFiDAAAAAAAAIRY0hWELNsEAAAAAABAUNCrSj1GjAEAAAAAAJA96FWlHCPGAAAAAAAAQIgxYgwAAAAAAICsQa8q9WpvEDqxbBMAAAAAAADBQK8qLRgxBgAAAAAAAEKMTUoAAAAAAACQPehVpVytDUITc90AAAAAAAAIBnpV6cGIMQAAAAAAABBijBgDAAAAAAAge9CrSrmkDUJznHUAAAAAAAAEA72q1Eu+ghBIofxIVIP6HqN9WnVW04J6mrt+pe6Z+Ik+W1KkfIvoX3udrD7N2qp9g6Y669On9fXyub5Lzoj8SFR/2fl47V3YVU0K6mne+pW6b+oH+nLpDO3ctL0u+cVh6tWkrSqd05gVs3X7xHe0fOM632WnVb5FdXXPk7Rni25qlF9fC0qW6+GiYfpq+TR1atBK/9fn12pXv7kkadqaBbpn6puas36p56rTL9+iumyn07Rbs53UKK++Fm5Yrv/MfkejV07Z6rjfdjxKZ3c+WteOf1DfrpruqdrMyLM8XdDlDO3cpKca5jXQ4tKlev771zWueJJa1mmhwbvfqtLK0prj31gwTK8ueNdjxZmRZ3k6c8dz1LNxbzXIa6ClG5dqyPyXNXHNBElSQaRAp7Y/Q/2a7aWoRTV/w/e6c9qtnqtOr6jl6cR2f1C3RruofrShVpQt0nuL/qfpa79V36YH6KT2F9YcaxZRQaSO7p9+jRZsmOWx6txDFohHDkiMLBCPHJAYWSAxskA8sgCwbbU3CJ1YtomUyrOIFm1Yo7M+e1oLS1broB266969T9HxHz6spRvWauyKeXqmaJTu3ftU36VmVJ5FtHjDap0/4gkt2rBaB7Tqrjv3+LVOHf6AGhfU06tzx+iqZUWqdFW6vs+xurnvybp41DO+y06raCSipRuLdfHoR7SktFj7FPbQ33c5U2eNuEfLN67RX8Y/p8WlqxSR6ZQd99HNu/xGZ4+813fZaRe1qJZtLNbV4x7Q0tJV2qtFT/2l1zm6cMwdWlK6UpLUpm4LHdByV63YuNpztZkRtYhWbFylQZPu0vKNK7Vb0z66Yqc/6OrxN9ccc+7XV6hKVR6rzLyIRbSyfKXunHarVpat0M5NdtWFXS/RoEl/0Yqy5Tqr4+8Usaj+Ouk6ra9Ypw71O/ouOe0iFtXq8uV6tOj/VFy+XD0a7a4zO16le6ddoXHFn2tc8ec1x+7R7BAd2vpU3hCkAVkgHjkgMbJAPHJAYmSBxMgC8cgCOYJeVVpsx4hxJspAWGyoLNcDUz6tuT988QzNLylW76ZttKBktZ4uGiVJqgrZcuENleV6ePonNfc/WzpdC0pWqWfTtvpo0eStjn1+zig9ue/5mS4x40ory/XEzA9r7o9YPlWLNqzULxq30/ClE7WuovqvwGamSufUvn4LX6VmVGlVmZ6d817N/VErJmtx6Up1b9i+5o3Bpd1P1eOz3tJlO53mq8yM2lhVppfnv11zf2zxd1paulxdGuyoWeu/91iZX2VVZXpr4ZCa+xNWj9PyjcvUsX4n5Vmedm26u/48/nKVVlX/LH1fMsdTpZlTXrVRHy55qeb+1LXfaGXZUrWr31WrVi/b6tjdmx+ssauGZ7jCcCALxCMHJEYWiEcOSIwskBhZIB5ZIHfQq0o9RozhVYs6DdSpYQsVrV2W/OAQaV7QQB0btNDMtfFjMns075Tw8VzXrKChOtQv1Ox1S2oeG3bIINWLFihipseLPvBYnT9N8xuqff2WmluyWJJ0QMtdVe4q4kaNwqRJfiO1qdda80oW1jz24B63yjlpwuopem7uK1pbsd5jhX40ymus1nV30MLSBerSsKtWbFyuE9qdrP7N99Pq8mK9tXCIxhaP8V1mRjXMa6LCOm20pHTeVo83zW+pzg166pV5gz1VFi5kgXjkgMTIAvHIAYmRBRIjC8QjCwCbhX4XYzN7UtJxkpY65/rEHmsu6UVJnSTNkfQr59wqXzXmqjyL6F97naQhc8dr1toVvssJjDyL6LbdT9Nb88dpzrrlWz3XvVFrXbjTwbp89P88VedH1CIatPPpGrpwrOaWbH4DedQng1Q3mq9j2u6hxRuKPVboR9Qiuq7XWfpg8WjNK1mqutECndf5OF0/4SHfpXkTtYgu63a+Pl02UgtLl6hOpI6um3Cr5qyfp0b5DXR+5zN0WffzdeuU+3yXmlFRi+qCLgM1YsUXWly6SLs37af29TtobPEYXTPhj+raoJsu636VFk5ZqMWlC5N/whwQUVS/3vFPGrtquJZtXLDVc7s3O0hz1k/RqrJwNmEyiSwQjxyQGFkgHjkgMbJAYmSBeGSBLJfDvSpffapI0sJcdt+2w1OSBvzgseskfeSc6y7po9h9pJBJumPPE1VeVam/jxvqu5zAMJlu2e1UlVdV6rbv3t7quQ71m+vBvc/WHZPe1bcrc/+C7ZuYTDf1OV3lVZW6a+rrcc+XVpZryLxR+r8+v1KzggYeKvTDZLq2529VUVWpB2a8Ikk6u9PR+nDJGC2OjRiFjcl0abfzVOEq9OTs5yVJG6s2atb6uapSlVaXr9UTs19Q36a9VS9a13O1mWMyndf5QlW6Cj3//bOSpDJXpoqqCr2z8A1VukpNXzdN09ZOUe/GfTxXmxkm0693/KMqXYXemP943PO7Nz9Y36wcnvnCQoYsEI8ckBhZIB45IDGyQGJkgXhkgeznu9eUi32q0I8YO+c+M7NOP3j4l5IOjn38tKThkq7NWFEhcMseJ6iwTkP9/sv/qcKF62LBtfnbrieqRZ0GumTUs1udlzb1muiRfc7VozOG6+354z1WmHk39D5Vzeo01FVjn1TlNr5XImaqGy1QyzpNtKosHOMiV/Y4XU3zG+nG7x6tOS+7NdtJhXWa6Ph2+0mSmuQ31F96naOXvv9YL837yGe5GXFR17PVJL+xbpt6/za/V3L6T43bcE6n89U4r7Hum3GXKl2lJGl+ybwkr8ptp3S4WA3zm+o/s25RlSq3eq5j/R5qnNdM360e6am68CALxCMHJEYWiEcOSIwskBhZIB5ZAEHmq08VhhHjQrP/Z+++4+Sqy8WPf54tyaaHkE4CoQsqzQiKKHhVQATk/mygCPZyhasXrqhYALkq9+pVuXYERERAREVUFFGKSpUmvRcT0nvPtu/vj3MSNpnZnSWZ3TO783m/XvvKzDlnZp59dmGefeZboutCCuellM6r8JhJKaW5ACmluRExse/Cqz9n7XsEO48az3v/+hPWd7Zvcq65oZHocntIQyOtnR2lTzIIfe6lR7HjqAl86NaLNsnLxJZR/PCV7+Nnz9zOz5/9e4ER9r9P7vGv7DBiIh+/64e0dsnJy8ftyvK21Tyxci4tjUP48C6HsaJtLc+sro8pAP++29vYfsQkPvWP79La2bbx+Gn3foemhsaN97+93yl8/8mr6mIdog/u+E62GzaZsx/6Jm1dcrLLyBmsbl/LvHULGNE0nPfOOJYHlj/K2o51BUbbf47f/j1MaZnK1x/7b9rS83l5fNWjLGldzBunHMXv5/6GHUfuzO6jXsSVsy8vMNr+ccx2H2Li0Gmc/9RZtKfWkvP7jXstDyy/jdbO+vgdKYq1QCnrgPKsBUpZB5RnLVCetUApa4FBYmD3qmqyT9Vzg7D3wx9r2aKU0syig1Bm6vAxHLvTTNZ3tPO3I0/dePyMu3/Lb2Y9wB8O/RjTRowF4MJXHw/Av/z+XJ5bs7yQePvLlGFjeNuM/Vnf0cb1h5628fjZ913N9BHbMn3EOD6y22v5yG6v3Xjulb//ryJC7TeTW8byr9NfwfqONn5z8Oc2Hv+fh35JW+rglD2OZsLQMazvbOPh5bM55e4LNvnDYbCaOHQbjpz6Klo72/jZgWdvPH7uo1dw/YK7Nrm2g8Sq9rWs6ygtfAaT8UPG8YbJB9Pa2cYPZ/7PxuPnPfVTUkoct+sxjG4exdqOddy3/CHOfbx0GslgNG7Ithw88V9o62zla3t/a+PxS579EbcvuZXvPPFNTpjxft44+UgWty7iwqfPY966uQVG3PfGNk/gFeMPo62zlc/u+fzvwa9m/4B7l/2Vpmhmr7EHcskzXy0wysHPWqCUzLQZ/QAAIABJREFUdUB51gKlrAPKsxYoz1qglLXAIDHwe1U12aeq+ynG3ZgfEVPyruwUYPB/FNlP5qxZzu6/+GK351/3h/paLHiDuWuXs/dvPt/t+R88dkM/RlMb5q1bxoF/7H7E9A3z7+/HaGrHgvVLOfTGT/Tq2hNu6/6/tcFkUesS3n7rh7s9f/Pi+htxA7CkdTEfvPOEbs/PWfcc5zxSH78jGyxrW8in//GWbs+3pzbOeqD7nKk6rAVKWQeUZy1QyjqgPGuB8qwFSlkLaADr8z5VxU1KSAP8a8tcDZyY3z4R+PUWP5MkSZIkSZKqp+he0yDsU/U4gjAY8MM2K4qIy8gWehwfEbOBM4BzgCsi4v3AP4G3FRehJEmSJEmSYPD3qorqU9X9FOOU0nHdnHpdvwYiSZIkSZKkulZUn6oXuxgP4rasJEmSJEmSBhZ7VVVXsUE4mIdtSpIkSZIkaWCxV1V9PTcIt24BRUmSJEmSJKl67FX1icq7GEuSJEmSJEkatCpPMe7sjzAkSZIkSZKkyuxVVV8vNinphygkSZIkSZKk3rBXVXVOMZYkSZIkSZLqmLsYS5IkSZIkacCwV1V9vdjF2KxLkiRJkiSpBtir6hNOMZYkSZIkSZLqmFOMJUmSJEmSNGDYq6o+dzGWJEmSJEnSwGGvquqcYixJkiRJkiTVsR5HEAYO25QkSZIkSVJtsFfVNyrsYpzcGUaSJEmSJEm1wV5Vn3CKsSRJkiRJklTH3MVYkiRJkiRJA4a9qupzF2NJkiRJkiQNHPaqqs4pxpIkSZIkSVIdc4qxJEmSJEmSBgx7VdVXYRdjoNOsS5IkSZIkqQbYq+oTTjGWJEmSJEmS6piblEiSJEmSJGngsFdVda5BKEmSJEmSpAHDXlX1OcVYkiRJkiRJqmO9mGJsW1aSJEmSJEk1wl5V1TnFWJIkSZIkSQOGvarqc4rxIBERjRFxT0T8tuhYJElS/7IOkCRJ0tboeQRhwp1hBo6PAw8Do4sORJIk9TvrAEmSVB/sVfWJHhuEAYTzumteREwD3gR8CTil4HAkSVI/sg6QJEn1xF5V34jUQ1JHj56WZh5wUj+GU303/Okzd6WUZhYdR1+KiCuBrwCjgP9MKR252QVnAmeUe+xM4M4+jk+SJPXezL2Gcuc/1kVvr69YB2QXnYm1gCRJA8ILrQXqzUDvVdVqn6ryLsad/RCFtlhEHAksSCndFRGHlL0opTOBM7t5gvTfDx7WR9ENTJ968bVc+OiBRYdRc963+y1c9cReRYdRc47Z5T5ueXrHosOoKQfu+DSP/HNK0WHUnBdtP5fFs6cWHUZN2XbaHDrn7lJ0GLXnsFm9vrRXdQBYC7xA1gLlWQuUsg4oz1qglHVAedYC3XgBtUDdsldVdb3YxdhhmzXuVcDREXEE0AKMjohLUkrHFxyXJEnqe9YBkiSp7tirqj53MR7gUkqfSSlNSynNAI4FrvePAkmS6oN1gCRJkqrBXYwlSZIkSZI0MNir6hMVphgncNjmgJFSuhG4seAwJElSAawDJElSfbBX1RecYixJkiRJkiTVsV5sUtIfYUiSJEmSJEmV2auqvooNQodtSpIkSZIkqWbYq6o6pxhLkiRJkiRJdaziLsbR2U+RSJIkSZIkST2xV9UnnGIsSZIkSZKkgcNeVdX1okHYD1FIkiRJkiRJvWGvqupcg1CSJEmSJEmqYxVHEIbDNiVJkiRJklQj7FVVn2sQSpIkSZIkaeCwV1V1TjGWJEmSJEmS6ljPIwgTUAdbR0fEM8BKoANoTynNLDYiSZIkSZIklaiDXlURfaoeG4RBqqd53a9NKS0qOghJkiRJkiSVV0e9qn7tUznFWJIkSZIkSapj9bBJyfiIuLPL/fNSSudtdk0C/hgRCfhBmfOSJEmSJEmqBQO7V1WTfap6aBAu6sVc7VellOZExETguoh4JKX0l/4ITpIkSZIkSS/AwO5V1WSfyinGQEppTv7vAuBXwP7FRiRJkiRJkqR6VESfqu53MY6IEUBDSmllfvtQ4IsFhyVJkiRJkqTNDfJeVVF9qopTjOtgZ5hJwK8iArJ8XJpS+kOxIUmSJEmSJKmcQd6rKqRPVXkNwkEupfQUsHfRcUiSJEmSJKm+FdWnqodNSiRJkiRJkjRY2KuqugoNwmTSJUmSJEmSVCPsVfUFdzGWJEmSJEmS6ljlXYztykqSJEmSJKkW2KvqE5XXIBzEW0dLkiRJkiRpgLFXVXVOMZYkSZIkSZLqWMURhOGwTUmSJEmSJNUIe1XVV3mKsUlXFTVGM4dOPokdRuxDS+MolrXO4S8LL+KpVXcC8KLRr+agCe9mZNN4VrYv5C8LLuLxlbcWHHXfa4hmDpx4KlOHv5yhjaNZ0Tqbuxb9gNlrbgNgt9FHsde44xnWNI75a+/jr/O+wtqORQVH3fdeNvHLTBi2P40Nw1jfvpjHl/2IZ1f+CoDxw/Zn7/GnM6xpMkvXP8DdCz7P2va5BUfcP3Yd/3XGtBxIQ8Mw2joW8dzyH7Bg1RUEzew24ZuMGPpSWpqm8cC841ix7vaiw+0XU8Z9ixEtBxExnI6OBSxe+T2Wr76UliH7MWHMabQ0v5REJ2vW38L8pZ+no3NB0SH3i5Hb/B/NQw+CGE7qXMjald9l/ZrLaGzalZHbnEtD0w4AtLfez5rln6ej/fGCI+57MeZrMOSVEMOhcyFp9Q9h7c83vWjESTSM+jidS06E1luKCXQQsxYoZR3QPWuBUtYB5VkLlGctUMpaYJCwV1V1lRuEUhU10MCK9oVc+uxprGhbwM4jX87R253OhU99lM7UzpHbncYvZ53FU6vuZKeR+3PMtNP5/uMnsqZjedGh96kGGlndvoBrZn2MVe3zmT7ilbx26tn86pl3M7J5Mi8b/2F+P/tkVrTO4oCJn+CQKWfy+9knFR12n3ts6QXcs+AMOmljZPMMDpp6AcvWP8La9rkcMOnr3LPwLOatuYk9xn2Ml0/6H/7y3LuLDrlfzF7+PZ5Y9GkSrQxr3okXT76M1a0Psab1UVasv5M5K37E7hO+XXSY/Wrxim8xb8mpJFoZ0rQL20+8knWt99PYMIZlqy5h9bobSXQwaeyXmDLuG8xe9K6iQ+4Xa1d+m1VL/xNopaFpZ8aMv5L2tgfobH+WlUs+RGfHbKCBlhHvYeS477J8wRuKDrnPpdXfh+WnA63QuBMx7hJS20PQ/mB2QeP2RMvhpI75hcY5mFkLlLIO6J61QCnrgPKsBcqzFihlLSCV1/MahAnoTAP7SzWlLa3n5oWXsKJtPpB4ctUdLG+bz+SWXRjVPJ51Has3jiB4atUdtHWuY+yQqcUG3Q/a0zruWXwhq9rnAYlZq29hZdsctm15EdNHvIpnVl3Pstan6aSdexdfxJTh+zKqebuiw+5zK9uepJO2/F4CEiOapzNlxOtY0fYkc1ZfR2dq5ZEl32fMkN0Y2TyjwGj7z9q2x0m0ApBSgpRoadqeRBtzV/yIlevvJNFRcJT9q7X9sY052fC7MqRpBqvX3cDKtb+lM60ipbUsXfUjhg19eZGh9quO9sdgY14AEo1NM0hpRf4HAUCQ6KCxcccCIixA+xOw2e8KTdtvPB2jv0Ba+VXY+P8eVZu1QCnrgO5ZC5SyDijPWqA8a4EyrAUGvoHeq6pRFUYQJodtqk8NbxzLuCHbsWj9syxtncPi9f9kl5Gv4MlVd7DLqAPoSG0sXPdU0WH2u5bGbRjdPJ1l659iUstLgNh4LvKb2wzZkZVtzxUTYD/aa/zpbD/qaJoahrFs/cPMX/NX9hx3MivWP7bxmo60ltXtsxk9ZGdWtT1TXLD9aKdxX2TCyLfQ2DCMVesfYOnaG4sOqXCTtvkyY4a/g4aGYaxrvZ9V6/5ccs3woQfQ2vZoAdEVZ8SYLzN0+NuJhmG0t95Pa5e8bDPlISJGAA2sXfm14oLsZzH6TBj2/4gYRmp7ENbflJ0YejikNmi9qdD46o21QCnrgE1ZC5SyDijPWqA8a4FS1gIDnb2qvuAUYxWmgUaO2u5TPLD8TyxpzT69enD5nzlq2qdoiiF0pDaumv1l2tL6giPtX0Ejh0w+gydW/J7lbf9k1upbee3UL/LIsqtY0TaLfca9l5Q6aWpoKTrUfnHfoi9z36JzGNeyN+NbZtKZ2mhqGM76jqWbXNfWsZKmhhEFRdn/nlryBZ5aciajhu7HmJYDSKm18oMGuflLT2f+0s8xbMhMhre8siQnQ5v3YPzo/2D2ovcWFGExVi8/ndXLP0fTkJfRPPRA6JKXpXP3hBjG0OFvp7N9dg/PMrikFWfCii+SmveFIQdkOYnhxKhTSUvr6/ejaNYCpawDSlkLlLIOKM9aoDxrgVLWAlKpnqcYQ9aVHchfqlHBkdt9ko7UxnVzvwPADiP25ZCJ7+eyZ07jqw8fyaXPnMYbp3yCiUN3KjjW/hQcPPkLdNDOrQu+DsDctXdxz6IL+JepX+LtO/6SVW3zaOtcw+r2hQXH2p86WbLuHoY1TWLHMW+nvXNNyR8ATQ0jae9cXVB8Relk5fo7GdI0hUmj6mMdnco6Wdt6B02NU9hm5IkbjzY3zWDa+EuYv+wLrG29o8D4itJJe+vfaWicQsuIEzY9ldayfvXFjBx3LtGwbTHhFaIT2u4iGifD8HcSI/+dtPbX0FE/fxwVz1qglHVA96wFSlkHlGctUJ61QClrgQGt6F7TIOxTuYuxCnHE1P9geNM2XPnPz9OZr5EyqWUnZq25n3nrsp2z5q17jDlrH2HGyH1ZsL4+phYdNOkzDGsaxx+fO3WTtWMeXv5LHl7+SwBGN09n721PZGmd5KSriEZGNE1jReuTbD/qqI3HG2MYI5qz4/UoaKSlefvKF9aRoInmfFe+psbt2H7Cz1i84pusWPOLgiMrWuPG3Qo31UBECw2Nk+noXNzvURWrkWjcHobsTzROguHvzA43jCPGnpvtbLj6vGJDHKSsBUpZB1RmLVDKOqA8a4HuWAuUshYYkOxVVV3lEYRSlR06+WS2HTKdX/zzDNq7DG+fu/Yxpg1/ycZRAhNbdmb68JewYN3TRYXarw6c+EnGDpnBdc+dRkeXvDTGEMYOyRYMHtE0iVdNOo2Hlv6c1s6VRYXaL4Y0jmO7kYfTGMOABiYOO5BpI9/IwrV3MHf19YwesgtTR7yOhhjC7tt8mBWtj9fFmkPNDduy7YgjaYjhQANjW17N+BFHsXztrQAEQ4gYkt9u3nh7MGts2JZRw95M5DkZ0XIwo4cfw5p1N9PUOJntJ/6cpasuYtnqnxQdar+Khm0ZMuxoyPPSPPRghg47hrb1N9M89NU0Nr+Y7I+BkQwfcwapczkdbU8UHXbfahgHLW/amBOGHAQtR5JabyUtOYG06E2kxUeTFh8NnQtIyz8Pay4pOupByVqglHVAKWuBUtYB5VkLlGctUIa1gNStnkcQJmp6hxUNPKObJ7LvuDfR3tnKSbtftvH4tXP+j4dW3MDNCy/hmOmfY3jjWNZ2LOfWRZfzzOq7C4y4f4xomsSLxh5De+d6jtv56o3Hb57/VWavvoVDppzJqObtaOtcw+MrruHuxT8sMNp+khI7jn4b+4z/LEQDa9rmcv+i/2HemhsBuGP+qew1/jO8bOKXWbr+fu6c/6li4+0nicTkUe9i523/CwjWt8/h6SVns3TtnwDYd9qfaGmaBsCLJ18MwF2zX8369sG8kH1im5EnMHncOUADbe2zmb/sC6xady3bjj6FIU0zGD/6FMaPPmXjIx57btfiwu03iZYRJzBibJaXzo7ZrF5+Bm3r/siQliMZtc1/0dA4hZTW0d52LysWHQ8M8nXeEsTwd8LoLwIN0PEcaeWXYP2fN55/XgekFZDWFBDo4GYtUMo6oBvWAiWsA7pjLVCetUAJa4HBwV5Vn4jUw7DMMUMnpQOnDuz1LP7wzDfuSinNLDqOWjUzIr3twcOKDqOmfOrF13LhowcWHUbNed/ut3DVE3sVHUbNOWaX+7jl6R2LDqOmHLjj0zzyzylFh1FzXrT9XBbPnlp0GDVl22lz6Jy7S9Fh1Jz9D5vFnf9YF5WvrA5rgVLWAuVZC5SyDijPWqCUdUB51gLl9XctMNAM9F5VrfapnGIsSZIkSZIk1TE3KZEkSZIkSdLAYa+q6lyDUJIkSZIkSQODvao+4RRjSZIkSZIkqY45xViSJEmSJEkDh72qqrNBKEmSJEmSpIHDXlXVOcVYkiRJkiRJqmMVRhAmu7KSJEmSJEmqEfaq+kIvdjHu7J9IJEmSJEmSpJ7Yq+oTrkEoSZIkSZKkgcNeVdW5BqEkSZIkSZJUxxxBKEmSJEmSpIHDXlXVVd6kpNOkS5IkSZIkqRbYq+oLTjGWJEmSJEmS6ljFXYxTcmcYSZIkSZIk1QB7VX2i8hqEDtuUJEmSJElSrbBXVXVOMZYkSZIkSZLqmLsYS5IkSZIkaeCwV1V1FdYgTNDpvG5JkiRJkiTVAHtVfcIpxpIkSZIkSVIdc4qxJEmSJEmSBg57VVVXsUGYHLYpSZIkSZKkGmGvqvqcYixJkiRJkiTVsQojCJPDNiVJkiRJklQj7FX1hQq7GAOdJl2SJEmSJEk1wF5Vn3CKsSRJkiRJklTHerGLsQs/SpIkSZIkqUbYq6q6HhuECUgO25QkSZIkSVINsFfVN5xiLEmSJEmSJNWxCpuUJIdtSpIkSZIkqTbYq+oTFdcgrIdhmxFxOHAu0Aicn1I6p+CQJEmSJEmSVMZg71UV0aeq+ynGEdEIfAd4I7AncFxE7FlsVJIkSZIkSao3RfWp3MUY9geeSCk9BRARlwNvBh4qNCpJkiRJkiSVGty9qkL6VJFS98MyI+IPwPi+DKAftADrutw/L6V03oY7EfFW4PCU0gfy++8GDkgpndS/YRZjRMTqNfBw0XHUkmkwdTbMKTqOWmNeyjMvpcxJeeallDnp1g4ppQn99WLWAqX83SzPvJQyJ+WZl1LmpDzz0q1+rQUGmkHQq6rJPlWPIwhTSof35YvXiChzbHBPZu9iNQwnpZlFx1FTIhIpTS06jJpjXsozL6XMSXnmpZQ5qQnWAmX4u1meeSllTsozL6XMSXnmRVugDnpVhfSp6n4NQmA2ML3L/Wn4CYYkSZIkSZL6XyF9KhuE8Hdg14jYMSKGAMcCVxcckyRJkiRJkupPIX2qypuUDHIppfaIOAm4lmz76AtTSg8WHJYkSZIkSZLqTFF9qh43KVEdyNZ8KDe/vX6Zk/LMS3nmpZQ5Kc+8lDIntcGfQylzUp55KWVOyjMvpcxJeeZFqhlOMZYkSZIkSZLqmA1CnVV0ADXInJRnXsozL6XMSXnmpZQ5qQ3+HEqZk/LMSylzUp55KWVOyjMvUo1wirEkSZIkSZJUxxxBKEmSJEmSJNUxG4SSJEmSJElSHbNBWKci4vCIeDQinoiITxcdTy2IiAsjYkFEPFB0LLUkIqZHxA0R8XBEPBgRHy86pqJFREtE3BER/8hz4topXUREY0TcExG/LTqWWhARz0TE/RFxb0TcWXQ8tSIixkbElRHxSP7/l1cWHVO9sRYoZS1QyjqgPGuB7lkHlLIWKM9aQKotrkFYhyKiEXgMeAMwG/g7cFxK6aFCAytYRLwGWAVcnFJ6SdHx1IqImAJMSSndHRGjgLuAY+r59yUiAhiRUloVEc3A34CPp5RuKzi0mhARpwAzgdEppSOLjqdoEfEMMDOltKjoWGpJRPwY+GtK6fyIGAIMTyktKzquemEtUJ61QCnrgPKsBbpnHVDKWqA8awGptjiCsD7tDzyRUnoqpdQKXA68ueCYCpdS+guwpOg4ak1KaW5K6e789krgYWC7YqMqVsqsyu82519+2gJExDTgTcD5Rcei2hURo4HXABcApJRa/YOg31kLlGEtUMo6oDxrgfKsA9Rb1gJS7bFBWJ+2A2Z1uT8bCz31QkTMAPYFbi82kuLl02fuBRYA16WU6j4nuW8CpwGdRQdSQxLwx4i4KyI+VHQwNWInYCHwo3wa2vkRMaLooOqMtYBeMOuATVkLlGUdUJ61QClrAanG2CCsT1HmWN1/4qmeRcRI4BfAJ1JKK4qOp2gppY6U0j7ANGD/iKj7qWgRcSSwIKV0V9Gx1JhXpZT2A94IfCyfwljvmoD9gO+llPYFVgOugde/rAX0glgHlLIW2JR1QI+sBUpZC0g1xgZhfZoNTO9yfxowp6BYNADka+v8AvhpSumXRcdTS/KpEDcChxccSi14FXB0vs7O5cC/RMQlxYZUvJTSnPzfBcCvyKZ21rvZwOwuo22uJPsjQf3HWkC9Zh3QM2uBjawDumEtUJa1gFRjbBDWp78Du0bEjvlisMcCVxcck2pUvgj3BcDDKaWvFx1PLYiICRExNr89DHg98EixURUvpfSZlNK0lNIMsv+vXJ9SOr7gsAoVESPyRf3Jp80cCtT97qgppXnArIjYPT/0OqCuNzwogLWAesU6oDxrgVLWAeVZC5RnLSDVnqaiA1D/Sym1R8RJwLVAI3BhSunBgsMqXERcBhwCjI+I2cAZKaULio2qJrwKeDdwf77ODsDpKaVrCoypaFOAH+e7gDYAV6SUfltwTKpNk4BfZX9f0wRcmlL6Q7Eh1YyTgZ/mzamngPcWHE9dsRYoz1qgLOuA8qwF1FvWAt2zFpBqSKTkcjOSJEmSJElSvXKKsSRJkiRJklTHbBBKkiRJkiRJdcwGoSRJkiRJklTHbBBKkiRJkiRJdcwGoSRJkiRJklTHbBBKkiRJkiRJdcwGoSRJkiRJklTHbBBKkiRJkiRJdcwGoSRJkiRJklTHbBBKkiRJkiRJdcwGoSRJkiRJklTHbBBKkiRJkiRJdcwGoST1s4i4OSL2LTqO7kTE9yPi81v42PdExN+63F8VETtVL7otExFfj4iPFB2HJKk+RcRFEfFfvbz2KxHxib6OqZZERIqIXbbwsZMi4uGIGFrtuCSpntgglFTTIuKZiFibN5o2fE3diuc7JCJmVzPGF/j6RwErU0r35PfPjIhLioqnnJTSR1JKZ1fpuUamlJ7amud4IX9U9eCrwGcjYshWPo8kqcZsVivMy983RhYd15aIiAnACcAP8vuF1i0DQUppPnAD8KGiY5GkgcwGoaSB4Ki80bTha05RgURE01Y+xUeAn1QjFvVORDSmlOYCjwBHFx2PJKlPHJVSGgnsA+wLfKaIIKpQJ7wHuCaltLYK4Qx6XfL9U+DDRcYiSQOdDUJJA1JEvCIibomIZRHxj4g4pMu59+ZTTVZGxFMR8eH8+Ajg98DUrqMRNx+htvmn9fnIhE9FxH3A6ohoyh/3i4hYGBFPR8S/d7l+/4i4MyJWRMT8iPh6fnwI8C/ATfn9w4HTgXfksfwjIt4WEXdt9r2eGhFX5bcvyqcAX5d/fzdFxA5drn1Rfm5JRDwaEW/fgtxuzMeGXOQxLIiIuRHx3i7XbhsRV+ff6x3Azps918YpQxExLCL+NyKejYjlEfG3iBiWn/t5PupjeUT8JSJenB//EPAu4LQ8R7/Jj+8RETfmP/8HI+LoLq95UUR8LyKuiYjVwGvzUzcCb3qh+ZAkDRwppXnAtWSNQiJiaER8LSL+mb8nf7/Le89NEfGW/PZB+XvWEfn910fEvfntnSPi+ohYHBGLIuKnETF2w2t2UyfsGxF35+/VPwNaulw/PiJ+m7+HLYmIv0bEhr/L3sjzdUJ3dcuaiNi2y/O9LK9HmiNb6uPmiPhW/p76SES8rsu1YyLigvz9/LmI+K+IaHwhOY6IXfLcLc/z8bMu517cpQ6ZHxGn58f3j4hb8+95bkR8O7oZ1R8Rb4qIe/LaYlZEnNnl3Iz85/T+iPgncH1+6nZgp641kSTphbFBKGnAiYjtgN8B/wWMA/4T+EVk03IAFgBHAqOB9wLfiIj9UkqryQrvOVswGvE4subSWKAT+A3wD2A74HXAJyLisPzac4FzU0qjyRpmV+THdwU6U0qzAVJKfwC+DPwsj2Vv4Gpgx4jYo8trH8+mow7fBZwNjAfuJfvUfMMfEtcBlwIT85i/26XZ9um8MC/71cP3PhkYk3+v7we+ExHb5Oe+A6wDpgDvy7+68zXgZcCBZD+30/JcQvYH0K553Hdv+J5SSuflt/8nz9FREdFMlv8/5tefDPw0Inbv8lrvBL4EjAI2rIn4MLB3D/FJkga4iJhG9l7/RH7ov4HdyBqGu5C9l30hP3cTcEh++zXAU8DBXe7ftOFpga8AU4E9gOnAmZu9dNc6oQG4iuy9exzwc+AtXa49FZgNTAAmkX1YmPJzLwUeBeihbrkR6PoB4PHA5Smltvz+Afn3Mh44A/hlRIzLz/0YaM9zsS9wKPCBPHcH9VQnRMRB+XOcTfYevA0wDfhW/vhRwJ+AP+S52gX4c/6YDuA/8pheSVY7/RvlrSabZj02z+lHI+KYza45mOxncVieq3ayn7nv85K0hWwQShoIrupSnF5FVghfk1K6JqXUmVK6DrgTOAIgpfS7lNKTKXMTWRH76q2M4f9SSrPyKT8vByaklL6YUmrN19j7IXBsfm0bsEtEjE8prUop3ZYfHwus7OlFUkrrgZ/l3yN5c28G8Nsul/0upfSX/NrPAq+MiOlkTdFnUko/Sim1p5TuBn4BvDV/7nNSSmO7++ohrDbgiymltpTSNcAqYPd8xMFbgC+klFanlB4g+8OjRD4y4n3Ax1NKz6WUOlJKt+TfAymlC1NKK/P7ZwJ7R8SYbuJ5BTASOCfP//V5fo7rcs2vU0o3578f6/JjK8l+BpKkweeqiFgJzCL7oPCMiAjgg8B/pJSWpJRWkn0wt+H9+iY2bQh+pcv9g/PzpJSeSCldl1Jan1JaCHy9y3UbdK0TXgE0A9/M3zuvBP7e5do2sg/WdsjP/zWltKFBWLFWIHuv3VAnNJK9/3X9IHFBl9f+GVnD8U0RMYn76Fh/AAAgAElEQVSs4fiJ/H17AfCNDflIKf2tpzohpbThA7c2YAdgakppXZfjRwLzUkr/mx9fmVK6PX/uu1JKt+X1yTNkayxunkPya29MKd2fv4ffB1xW5toz8++h61Rs3+claSvYIJQ0EBzTpTg9hqwofdtmo98OIiu2iYg3RsRt+fSWZWSNw/FbGcOsLrd3IJvu0/X1TycbBQDZKLvdgEci4u8RcWR+fCnZiLZKfgy8M//D5t3AFRsaaZvHklJaBSwh+6R+B+CAzeJ6F9kIwK2xOP9kfoM1ZA26CUATm+bm2W6eYzzZ9KonNz8REY0RcU5EPBkRK4BnujymnKnArJRSZ5djz5KNCtlgFqVGAT2NlJQkDVzHpJRGkY0IfBHZe8gEYDhwV5f3xT/kxwFuBXbLG2f7ABcD0yNiPLA/8BeAiJgYEZfnU3JXAJdQ+h7V9X1nKvBcl6YfbPr++FWy0W5/jGwplE93OdebWuHXwJ4RsRPwBmB5SumOLufLvfaGOqEZmNslHz8gG43/QpxGNqryjsiW+dgwe2A6Zd7nASJit8imVc/Lc/hlunmfj4gDIuKGyKZNLydbv7mnfG/g+7wkbQUbhJIGolnATzb7VHtESumciBhKNmrua8CkfGTcNWSFLDw/haer1WR/QGxQrqHW9XGzgKc3e/1RKaUNIxgfTykdR1Zw/zdwZT7993Eg8inS5Z6X/PG3Aa1kox7fSemmJtM33Ihsl8ZxwJw8rps2i2tkSumj+bWnx6a7QW/yVeZ7rmQh2TSl6V2Obd/NtYvIpiLvXObcO4E3A68nm8o8Y8O3l/+7eY7mkP0B1/U9bHvguS73y/2c9yCbFi5JGqTymQMXkdUBi4C1wIu7vC+OSdlmJqSU1gB3AR8HHkgptQK3AKcAT6aUFuVP+xWy95W9UrZ8yPE8/x618aW73J4LbJd/0LfBxvfHfGTdqSmlnYCjgFPi+XUC7yP7kLHc8254/Dqy5UveRfZB4uZ1QrnX3lAnrAfGd8nH6JTShqVIXt1TnRARr85ff15K6YMppalkG4N8N7L1hmdR/n0e4Htkm4XtmufwdEpzuMGlZEuuTE8pjQG+X+baTfIS2WYlu+D7vCRtMRuEkgaiS4CjIuKwfPRZS2SbaUwDhgBDyZtXEfFGsvV1NpgPbLvZ9NV7gSMiYlxETAY+UeH17wBWRLYg+bA8hpdExMsBIuL4iJiQj3Db8El2R8rWBvoTm06TmQ/M2KzZBdkohm8D7V2m7mxwRL5O0BCydYBuTynNIptmu1tEvDuyhcqbI+Llka9nmFL6ctp0N+hNvip8zyVSSh3AL4EzI2J4ROwJnNjNtZ3AhcDXI1tgvTEiXpk3dEeR/cGymKxR++XNHj4f2KnL/dvJmrqn5d/jIWR/YF1eIeSDydY6lCQNbt8kG1m3F9kSIN+IiImQrWMcz68ZDNk04pN4fr3BGze7D9n71CpgWf4h3ycrvP6tZB+g/XtkG5b8P7IRieQxHBnZRh8BrCBbn68jP30NpXXC5nULZHXCe4Cjyeqiribmr90cEW8j+4DsmpTSXLJlV/43IkZHRENkG7AcDJBPde62Tkgp/TWP/215zQXZiMeUx/9bYHJEfCKyzWFGRcQBXXK4AlgVES8CPtpD/kYBS1JK6yJif7IPEivZn2yZle5mMkiSKrBBKGnAyZthbyb79Hkh2SfWnwQa8vWF/p3sk/WlZEXl1V0e+wjZWjZP5dNrppJ98v4PsqmtfyRbA7Cn1+8ga0jtAzxNNkLhfLLRbwCHAw/mo/LOBY5Nz6+D9wOyT/s3+Hn+7+KIuLvL8Z8AL6F0VABkn6yfQTa1+GVkIwjIv/dDydYSmgPMIxvBOLSn72crnUQ23Xge2YiNH/Vw7X8C95Otw7Qkj62B7I+cZ8lGAD4E3LbZ4y4gm0q1LCKuykd4HE22jtIi4LvACfnPtqyImALsSbZovCRpEMvXCbwY+DzwKbLpvLflU1v/BHTd1OomsobUX7q5D3AWsB+wnGyTtF9WeP1W4P+RNfCWAu/Y7DG75nGsImsmfjeldGN+7mKyDwKH5c9Vrm4hpXQz2UZfd+dr+nV1e/4ai8g27HprSmlxfu4Esg9TH8pju5J8iZYX4OXA7XmdczXZ+sJP53XIG8hqpHlkMydemz/mP8lqspVkTdueaq1/A74Y2ZqSX+D5zd568i6ykYaSpC0Umy5PIUnqaxHxN+DklNI9PVwzjGyR8f1SSo93OX4RMDul9Lk+D3QQiYj/JZsu9t2iY5EkqScR8WVgQUrpmxWuux64NKV0fpdj7wE+kFI6qNsHDjL56NCbgH27fCArSXqBmooOQJLqTS+L9o8Cf+/aHNSWSymdWnQMkiT1Rkrp9ErX5Mua7Ec2o6KupWw35j2KjkOSBrq6n2IcERdGxIKIeKCb8xER/xcRT0TEfRGxX3/HWElEHB4Rj+YxfrryIyTVsoh4hmzBdJtakiqyDpDqS0T8mGyK8ifyab2SpEGmUn2Xr/X6s/z87RExY6tfs96nGEfEa8jW/7g4pfSSMuePAE4GjgAOAM5NKR2w+XVFiYhG4DGy9T5mk63tdVxK6aFCA5MkSX3OOkCSJGlw6U19FxH/BuyVUvpIRBwL/GtK6R1b87p1P4IwpfQXssXyu/NmsuZhSindBozNF7uvFfsDT6SUnsoXRL4cpxpIklQvrAMkSZIGl97Ud28GfpzfvhJ4XUTE1rxoj2sQHvbaEWnxko6tef7C3XXf+geBrovVnpdSOu8FPMV2ZDukbjA7Pza3CuFVQ7n4Nh3hGHEm2Y6nJcYDM/omLkmStAWe2aaBRUs6elvgVa4DoMdaYEQ0srZl3AuMUpIk9ZXUupLUvnarmj2D2aGHHZ4WL1pUdBhb7O6776rUp+pNfbfxmpRSe0QsB7Yl28F+i/TYIFy8pIM7rt1+S5+7JjROeXxdSmnmVjxFuf8oa2leduX4UjoTOLPcg2dEpDvm7lLxRQ6bus8Lj2yAui5dyRvirb269to59/bqusGQvxeSl94YLLmrdl56q5bzV1ROXogi8jcQ8tIb1czdYMnJC9Gb/O1/2KyK13TRuzqlh1pg7fCJaejub38hrylJkvrQ+kevKDqEmrZ40SJuvv3OosPYYsOao1Kfqjf1XdV7VT02CBPQSefWPP9gMBuY3uX+NGBOQbGUU+vxSZKkvmMdIEmS6koCBvl2Gr2p7zZcMzsimoAx9Lx8XkU9Nggh0ZHqvkF4NXBSRFxONqRzeUqpVqYXQ7ZY5a4RsSPwHHAs8M5iQ5IkSf3EOkCSJNWdVFMTO6uuN/Xd1cCJwK3AW4Hr01buQlyhQTj4RcRlwCHA+IiYTbY+TzNASun7wDVkOxg/AawB3ltMpOXlc81PAq4FGoELU0oPFhyWJEnqB9YBkiRJg0t39V1EfBG4M6V0NXAB8JOIeIJs5OCxW/u6vZhiPKi7sqSUjqtwPgEf66dwtkhK6RqyRqYkSaoz1gGSJKnuDO5WVdn6LqX0hS631wFvq+ZrVhxB6BqEkiRJkiRJqhWDvD9YiIaiA5AkSZIkSZJUnApTjBMdg3xrGEmSJEmSJA0ctqqqrxdTjM26JEmSJEmSasMg38W4EE4xliRJkiRJkupYxV2MO+zKSpIkSZIkqVbYqqo6pxhLkiRJkiRpQEjYH+wLTjGWJEmSJEmS6ljlKcZuDTPoHTZ1n6JDGLDM3ZYzd1vH/G0d87flzN3W6U3+HkuL+yESSZKkASq5i3Ff6MUUY0mSJEmSJKk2uItx9TnFWJIkSZIkSapjFaYYJ3cxliRJkiRJUs1winH19TzFOEGHSZckSZIkSZIGLacYS5IkSZIkSXWs4i7GblIiSZIkSZKkWpBwinFfqLCLcdBB9E8kkiRJkiRJUo+Suxj3AacYS5IkSZIkSXWs8hRjm7KSJEmSJEmqEU4xrr4KU4xxirEkSZIkSZJqhv3B6nOKsSRJkiRJklTHKk4xdgShJEmSJEmSaoZDCKuu4hTjzmSDUJIkSZIkSbXBXYyrzynGkiRJkiRJUh1zirEkSZIkSZIGhJTcxbgvVGgQBh0OMpQkSZIkSVKNsD9YfXb/JEmSJEmSpDrmJiWSJEmSJEkaOBxCWHWuQajCjdpmJKec/1FeduherFi0kgtOv5QbLvtb0WEV6s0fO5xDTzyEGS/dnhsvu5mvvu87RYdUuOYhTZz83Q+y3+teyqhxI5nzxDwu/Oyl/P0P9xYdWqE+dfHJ7Pu6l9IyYihL5y3jiq/+mt9fcH3RYdWM7XaZzHn3/S9/ufI2/vuEbxUdTuG+dv2Z7PGKXelo7wRg0XNLeN8eHy84quId8o4DefcX3saE7cezdN4yvvre7/DA3x4pOqxBqXPtYtrn3EznmoXQsY6WfT7W8/VrFtI26wbSuqVEyzY0T38tDcMn9FO0/ce8lGdeSpmT8sxLKXNSnnkZPNzFuPoqjiCU+trJ334/7a3tvH3yB9l5nxl86bef4al/PMOzD80uOrTCLJ6zhJ9+6RfMPGwfhrYMKTqcmtDY1MjCWYs49ZAzWPDPRex/xL587men8KG9TmX+swuLDq8wl5/zK77+ge/R1trO9N2n8rUbzuKJe57h8bufKjq0mnDytz/Ao39/sugwasq3T77AJnIX+71+Lz5wzvF86dhv8MgdTzBuyjZFhzS4RQMNY3ehcfxLaHv69z1emjo7aH36Gpom7E3j+JfSsfgBWp++hqF7HE80NPZTwP3EvJRnXkqZk/LMSylzUp55kbpVYQ3CoCM1DOgv1baW4UM56C2v4KIvXM661et48OZHuPXqO3n9uw8uOrRC/e1Xd3DLr//OisUriw6lZqxbs56fnPVz5j+7kJQSt//ubuY9vYBdX7ZT0aEV6tmHZtPW2g5s2M0rMWXnSQVHVRsOeceBrFq2mnuuv7/oUFTDTjjz7Vxy9pU8fPvjpJRYPGcJi+csKTqsQauhZRuatt2TaBlX8drOVc8BicYJexMNjTRN2Ds/Pvg+QDQv5ZmXUuakPPNSypyUZ14Gjw07GQ/Er1rVYwctAZ00DOgv1bbtdptCZ0cnzz0+d+OxJ+97hh32nFZgVBoIxk4cw7TdpvDsg7OKDqVwJ3/nA/xm1SX86JFzWTJ3KXdcc0/RIRVu+KhhnHjWO/jBf15cdCg1531ffhdXLriAb/71bPY6eM+iwylUQ0MDu83cmTETRnPRY9/i0n9+n5O+9X6GOHK7JqR1S2ho2ZaI55e7aRi2LWldfTdwzUt55qWUOSnPvJQyJ+WZl9qWBvBXrbKDpkING9nC6uVrNjm2evkaho8aVlBEGggamxr5zCX/zh8vvolZj84pOpzCfetj5/Pm0SfwiVd/nr/96nba1rcVHVLh3nP2sfzhwutZOHtx0aHUlPM/fQkn7Pwxjpv2YX73wz9x9tWfZspO9TvidJtJY2ge0sRr3vIKTnnN5/nIvp9kl31m8K7PvaXo0ATQ2QaNmzVrG4eQOur8/3HmpTzzUsqclGdeSpmT8syL6kzFNQjdpER9ae2qdQwfvWkzcMToYaxZubagiFTrIoJPXXwy7a3tfPukC4oOp2Z0dnby4M2P8PrjX81RHz2Uq77V85oqg9nOe89g39e9lI/ud1rRodScR+54YuPt6y6+idceexD7H7Evv/72HwqMqjjr17YCcNW3f8+SecsAuPIbv+Vdn30LP/rcZUWGNmh0LHmUttk3AtAwYipDdj6q9w9uaIaO1s2esI1obK5egAUxL+WZl1LmpDzzUsqclGdeBqdEbU/VHah63sU4hev4qU8999hcGpsa2W6XyTz3xDwAdtprRl1vUKKenXr+R9lm0hg++6av0NHeUXQ4NaexqZGpO08uOoxC7XXInkyaMYGfPvs9IBup3NDYwA57TuPfZn6q4OhqS0ppk2kz9WbVstUsmLWIZIXZZxrH7U7juN236LHRMo7Ohfdu8nvauXYRjeNfUs0QC2FeyjMvpcxJeeallDkpz7wMZtZv1Wb3T4Vat2Y9f/vl7Zx41jtoGT6UFx+4Owe++eX86Sc3FR1aoRoaG2ge2kxDY8Mmt+vdx7/3QbbfYzs+f/R/07qutfIDBrmxE0ZzyDsOpGVECw0NDcw8dG8OOfZVdb8pxzXn/YkTdzmZj+z7ST6y7yf57Q/+yO2/u5vPHP6lokMr1Igxw5l56N4b/3/yL+88iJe+Zg/uvPYfRYdWqD9edCPHnPRGxk4YzcixI/h/H38Tt/3urqLDGrRSSqTOdkid2f3OdlJn+Q97GkZuBwQdi+4jdXbQvvC+/PjgW6fYvJRnXkqZk/LMSylzUp55kbpXcYpxp1OM1ce+9bHzOfWCj3LF/PNZuXgV5/7bD+t+BOG7PvcWTjjj7Rvvv/7dr+His67gJ2f9vMCoijVx+/Ec+eFDaV3XyhVzf7jx+Dc/8gOuv/RvBUZWnJTgqI8cxse/9yGiIVjw7CK+9x8XcevVdxYdWqHWr23dOHUUsqUMWte1snzRigKjKl5TcxPvOftYpr9oOzo7Opn1yHOc+a9fZfZj9b2O5yVnX8no8aP40aP/R+u6Nm76+S1c+qVfFh3WoJVaV9L68E823l9/3w+geRQtLz4BgNYnf0PDyCk0TZpJNDQyZMcjaJt1A+1zbiVatmHIjkcQDY1Fhd9nzEt55qWUOSnPvJQyJ+WZl0GixncDHqiip2k1u750WDr36p37MZzqe9NOD96VUppZdBy1amZE2ibeWnQYNeW6dCVvMCclzEt55qWUOSnPvJQyJ+Xdnv7MirSk3z6hbRg+MQ3d/e2VL5QkSf1i/aNX0LlmgaO1urHXPi9Lv73+lqLD2GI7bNtSk32qCiMIXYNQkiRJkiRJtcMRhNVn90+SJEmSJEmqYz3vYgx02kOUJEmSJElSjUjuYlx1FTcp6UhOe5ckSZIkSVKNsD9YdQ4PlCRJkiRJkupYhSnGQYc9REmSJEmSJNWAhAMI+0LFKcad7mIsSZIkSZKkGuEuxtVn90+SJEmSJEmqYxV3MXaKsSRJkiRJkmpDchfjPlB5DUJ3MZYkSZIkSVKtsD9YdQ4PlCRJkiRJkupY5U1K7CFKkiRJkiSpRjiAsPp6nmKcoMNdjCVJkiRJklQLkrsY9wW7f5IkSZIkSVIdqzDFOOjETUokSZIkSZJUvATuYtwHKuxi7BRjSZIkSZIk1RD7g1Vn90+SJEmSJEmqYxV3Me6whyhJkiRJkqQa4QDC6qswxTjoTK5BKEmSJEmSpNrgLsbV5/BASZIkSZIkqY45xViSJEmSJEkDhrsYV1/FXYw73cVYkiRJkiRJtcL+YNXZ/ZMkSZIkSZLqWIUpxkEHblIiSZIkSZKk2uAAwupzirEkSZIkSZIGhIS7GPcFu3+SJEmSJElSHevFLsZOMZYkSZIkSVJtcBfj6ut5inEKpxhLkiRJkiSpNiRchLAP2P2TJEmSJEmS6ljlKcaOIJQkSZIkSVKNcABh9fXY/UtAJzGgv3ojIg6PiEcj4omI+HSZ89tHxA0RcU9E3BcRR2xZuqsvIi6MiAUR8UDRsUiSpP5nLSBJkupNSgP3a2tExLiIuC4iHs//3abMNftExK0R8WDew3pHb5677ocHRkQj8B3gjcCewHERsedml30OuCKltC9wLPDd/o2yRxcBhxcdhCRJKsxFWAtIkiTVg08Df04p7Qr8Ob+/uTXACSmlF5PViN+MiLGVnrjCFOOohynG+wNPpJSeAoiIy4E3Aw91uSYBo/PbY4A5/RphD1JKf4mIGUXHIUmSimEtIEmS6k0d72L8ZuCQ/PaPgRuBT3W9IKX0WJfbcyJiATABWNbTE/e8izHQmXo3TbeGjY+IO7vcPy+ldF6X+9sBs7rcnw0csNlznAn8MSJOBkYAr++LQPtMxJnAGd2dvi5d2X+xDBDmpDzzUp55KWVOyjMvpcxJqdGUzBTZej3UAlOahrOk+q8oSZLUdwZ2f7BSn6onk1JKcwFSSnMjYmJPF0fE/sAQ4MlKT1xxk5JBYFFKaWYP58t1QDf/VTsOuCil9L8R8UrgJxHxkpRSZ9Wi7EspnUnW5CwVkd4Qb+3PaGredelKzEkp81KeeSllTsozL6XMSTfSn/vgObuvBeYOn5iGVv8VJUmSVF6PfaqI+BMwucypz76QF4mIKcBPgBN707+qvIvx4F+mcDYwvcv9aZROIX4/+do+KaVbI6IFGA8s6JcIJUmSJEmSRGKgDyDsWUqp21mrETE/Iqbkowen0E1fKiJGA78DPpdSuq03r1thinEMhinGlfwd2DUidgSeI9uE5J2bXfNP4HXARRGxB9ACLOzXKCVJkiRJkrTVuwEPYFcDJwLn5P/+evMLImII8Cvg4pTSz3v7xIN+eGAlKaV24CTgWuBhst2KH4yIL0bE0fllpwIfjIh/AJcB70mpNn4dI+Iy4FZg94iYHRHvLzomSZLUf6wFJEmS6sY5wBsi4nHgDfl9ImJmRJyfX/N24DXAeyLi3vxrn0pPXHGKcWcd9BBTStcA12x27Atdbj8EvKq/4+qNlNJxRccgSZKKYy0gSZLqS6rbXYxTSovJZrhufvxO4AP57UuAS17oc/c8xThBx+CfYixJkiRJkqSBoj77g32q8ghCG4SSJEmSJEmqEfYHq2/wzx+WJEmSJEmS1K1e7GJsD1GSJEmSJEk1INX1LsZ9puIU4w6cYixJkiRJkqTiJajbTUr6ksMDJUmSJEmSpDpWYYqxm5RIkiRJkiSphjiAsOoqTDF2DUJJkiRJkiTVDvuD1VdxDUINftfOubfiNYdN3acfIhl4epM7MH/lmLutY/62jvnbcuZu6/Qmf/sftqYfIpEkSZKeV7FB2OkmJZIkSZIkSaoR7mJcfT2vQZigwzUIJUmSJEmSVCPcxbj6XGBQkiRJkiRJqmOVpxi7SYkkSZIkSZJqhQMIq67nKcYEnU4xliRJkiRJUo2wP1h9Dg+UJEn6/+zdd5yU1fn+8evMbGNdQHbpZaUIKIhAJJaYABrR6DciGqyhqGBsKCoq8kMpRhNLTCygYgUBo4jdEHE1KJEoIoorSFOK6II0qcvWOb8/dlmBeWZmhZnnmfJ55zUvd2aeHe5cDOt4c+5zAAAAgBTGKcYAAAAAAABICFZSgGOMoy7CiLEYMQYAAAAAAEDcoD0YfYwYAwAAAAAAACmMU4wBAAAAAACQGKzEhHH0hW8QWk4xBgAAAAAAQLywsgwZRx3LAwEAAAAAAIAUFvmQEk4xBgAAAAAAQByoOlDX6yqSTy32IKRBCAAAAAAAgPjAiHH0RWwQIvmd0byb1yUkLLI7eGR3aMjv0JDfwSO7Q1Ob/FbYLS5UAgAAAPwk8ogxKwgBAAAAAAAQJzjFOPoYMQYAAAAAAEDCYMQ4+jjFGAAAAAAAAEhhEUaMDSsIEVNv7Ji63/2MOhl687HZmnj9Mx5VFB+aHNFI108cqqNP6qDy0gr99+WP9egNzypQGfC6NE/lH9VCwyYMVYfj2mrbph168tapmvfaJ16X5apzrv2dTh/cW6275Ov9f87T/ZdPrHmu+6nHaNiEoWqc31DL5q/U/ZdN1MZvN3tYrXtC5ZKWnqZR04erQ4+2atq6sUacMlaFH3zlcbXuCJXJ0Se01+A7L1L749oqUBlQ4ftLNHH4M9q6YZvHFbsjVC75R7fUyCnD1KxdE0nSyoWrNHH4s/p26Xdelpt0Anu2qKJongLFm6TKEmV1uzb89cWbVL5ujmzJjzJZDZTe6hT5shu5VK17yMUZuQQjE2fkEoxMnJFLcuAU49iIuIIwIJPQN8S3vvUG1twuaDpUZXvKNPelj7wuy3PXTxyqbRt36MLmf9JV3W/RsT07qe81Z3hdlqd8fp/Gv3ar5v9roc7Lu0wPXjlJI6depxbtm3ldmqu2FG3V9Ltf1uxn5+z3eL28uhr78i2aMuYFnZd3mVYsXKXbX7jRoyrdFyoXSVo8b6nuGfiItqz/0YPKvBMqk5wGh2nWkwUa2OYaDWh9tYp37dHNz4T/cJxMQuWypWir7jz/AZ2Xd5n6Nxqij978VKP/eYNHVSYx45Pv8COVnn9KxEttoFJlq2fJ36CDMrsMlT+3o8pWz5INVLpQqMvIxRm5BCMTZ+QSjEyckUvSsAn8v3jFiDHixm/6n6htG7fry/8u9boUzzVt01gfvPQ/lZeW68cftmnB7EU6onMrr8vyVP5RLZTXPFcv/+MtBQIBLZqzWF/NW67TBvb0ujRXffjqJ/rf6wu0Y8vO/R7/9XknaM2SdZo782OVl5Zr6rgZatu1tVp1bO5Rpe4KlUtFeYVefWiWlsxblnIrcENlsuDtRZo782MV79yj0j1len3C2+p8ckePqnRfqFx2by/WD2s3Vd0xUqAyoOZHNvWgwuTmy2qgtLxOMlm5Ea8N7PpekpW/UVcZn19pjbpWP558qzrJxRm5BCMTZ+QSjEyckQsQWvhDSiyHlMA9fQb1VsHUD7wuIy68+vAs9b7wZH3x/hLlNMjRL3/XTVPGvOh1Wd4yDj+LjFHrzvnu1xKHWndupVWFa2vulxSXquibDTqicyutW17kYWWId116Hq21S9Z5XUbceHXrZNXJyZLxGU0Zm+I/dz1mS7bKl5Uns8/Pf1+dPNmSrVK9IzyszFvk4oxcgpGJM3IJRibOyCW+cYpx9EXYg5AGIdzRqFVDHdurk/4+9DGvS4kLhR98pTOHnqbXtz8nf5pf70x+P+X22jvQumXfa9vG7brglr56+R//UrdTOuvYXp30xZzFXpcWF7JysrR90479HiveXqzsunU8qgiJoE2XfA2443yN7Xev16XEjXNzL1VWdqb6DO6lH9amxh6ecStQLvkz9n/MnyFbWe5NPfGCXJyRSzAycUYuwcjEGbnEL0uDMBbCryAEXNJnUE8t+XCZNqzZ6HUpnjPG6K9v365/PVGgG04eraycLN389DUaeu8APYLnq6EAACAASURBVDVymtfleaayolJjz71Pwx4eogtv7acVn36jD2b8T+VlFV6XFhdKdpUou97+zcDsetkq3rnHo4oQ75q3a6q/zBqtx254Vos/XOZ1OXGlpLhUbz1eoJkbn9aQTjdo2wHNd9Re5dblKv/ufUmS77Dmymh3du2/2ZcuVZYd8ILlMv706BXoEXJxRi7ByMQZuQQjE2fkAtRexAYhKwjhhj4De+mFe1/zuoy4UDc3R43zG+q1CW+rvKxC5Vt3afbkObr0zxeldINQklZ/+a1GnDK25v6DH96lgufe966gOLJmyTqdPqhXzf2s7Ew1a9eE0VE4apzfUPcW3KHpd83Uu9Pmel1OXDI+o8zsTOW1yKVBeAj8uR3lzz24PS5NVq4CmxbJWlsz3hXYs1n+hsdEs0RPkIszcglGJs7IJRiZOCOX5GQlBeL4sI9EFfaQEiujgE3sG+Jfp5M6KK9FLqcXV9uxZafWr/pBZ199unx+nw6rn60+g3pr1RdrI39zkmvTJV/pmenKrJOh/iPOVm6zBnpn8vtel+Uqn9+n9Mx0+fy+/b6e9+onan1Mvn593glKz0zXgDH9tbpwbcrsPxgqF0lKz0hTemZ60NfJLlQmec1zdf97Y/XGo7P11qQCr8t0XahcfnHasWrXrbV8Pp+y69bRVX8frF0/7tK3S7/3uuSkYq2VDVRIturQIBuoCHkapC+nhSSjys2FsoFKVWwqrH68pVvluoZcnJFLMDJxRi7ByMQZuSQPaxP3Fq8YMYbn+gzurXmvfKI9u0q8LiVujP/D33T1Py7Vhbeeo0BlQF/MWaLHb5rsdVmeO21gL5055FSlpafpy/8u1W2n/znlRoz/ePsfNGjsBTX3TxvYU8+Nn6Gp41/S+P5/07BHhui2qddr2fyVuvviBz2s1F3hcnlm2UNq2rqxJOme2XdIkga0ueanE2uTVKhMrLVq3q6pBo45XwPHnF/zfN96A70o03Whclm7ZJ2uffhyNWqZp9I9ZVqx4GuNOvNulZeyz1A02bKdKls6teZ+aeEkKb2usjoPkiSVffOmfDnNlNakh4zPr4w2Z6l83RxVFH0kk9VAGW3OkvH5vSo/ZsjFGbkEIxNn5BKMTJyRCxCasWHal3U7NrXHPTrAxXKi74PTHlhore3hdR3xqocxtoHp73UZcaXAzlQfMglCLs7IJRiZOCOXYGTibL59TzvsVtfGIHzZjW1mxwsiXwgAAFxRunyGAsUbGYkMoX3nrvYfL77jdRkH7ewuTeOyTxV5D0LxngQAAAAAAEB8iOdR3UQVdg9CAAAAAAAAAMkt7ApCaznFGAAAAAAAAPGBU4xjI+KIsaVBCAAAAAAAgDjBiHH0RWgQGlYQAgAAAAAAIG7QH4w+9iAEAAAAAAAAUhgjxgAAAAAAAEgYlhnjqAt/SIk4pAQAAAAAAADxoeqQEkQbI8YAAAAAAABACgs/Ymw5GQYAAAAAAABxwjJiHAsR9yAMiBFjAAAAAAAAxAfag9HHiDEAAAAAAACQwiIeUsIpxgAAAAAAAIgXjBhHX4QRY8MpxgAAAAAAAIgLnGIcG4wYAwAAAAAAACks4iElrNoEAAAAAABAfLCMGMdALRqEjBgDAAAAAAAgPtAfjD5GjAEAAAAAAIAUFv4UY8sKQgAAAAAAAMQPFhBGX8QRY04xBgAAAAAAQDywkgLMGEcdI8YAAAAAAABACuMUYwAAAAAAACQMWlXRxynGAAAAAAAASAxWsqxmizpGjAEAAAAAAIAUFv4UYxlWEAIAAAAAACAuWEkBr4tIQpFHjN2oAgAAAAAAAKgFJoyjjxFjAAAAAAAAIIWFX0FoOaQEAAAAAAAA8SPAEsKoizhizIwxAAAAAAAA4oEVI8axwIgxAAAAAAAAkMIiNgitNQl9qw1jzO+MMcuNMV8bY24Lcc0FxpivjDFLjDHP/8ycY8YY08oYM8cYs7S6tuFe1wQAANzB5wAAAJCKAtYm7O1QGGNyjTEFxpiV1f9sEObaesaY740xE2rz2pFPMU7yZZvGGL+kiZL6SPpO0gJjzBvW2q/2uaa9pFGSTrbW/miMaexNtY4qJI2w1n5mjKkraaExpmDf+gEAQNLicwAAAEgtVgokea8qjNskvWetvad6gdttkkaGuPbPkj6o7QtH3oMw+R0v6Wtr7SpJMsa8IOkcSft+sL5C0kRr7Y+SZK3d6HqVIVhr10taX/31TmPMUkkttH/9AAAgCfE5AACQ6H5cUKvFTSnl5BM+9roExK9zJPWu/nqKpPfl0CA0xhwnqYmktyX1qM0Lh20QVm38mPCnGDc0xny6z/0nrLVP7HO/haR1+9z/TtIJB7xGB0kyxsyT5Jc0zlr7diyKPRTGmNaSukuaf8AT4ySNDfV9BXZmDKtKTGTijFyckUswMnFGLsHIJFg9hZwUCSvk54CqJ8cpxGeBZmnZ2npQvyIAAID7kuCQkkh9qnCaVP8Fsay1650mXI0xPkkPSBoo6be1LSr8CkIrKfEbhJutteG6pU7/Bw98q6VJaq+qLm1LSf81xhxjrd0WnRIPnTEmR9LLkm6w1u7Y70lrx0kaF+IbbR/TP8bVJZYCO1NkEoxcnJFLMDJxRi7ByCQE+97P/pawnwOksJ8F1mc3tpk/+1cEAADwilUgqG2TUML2qYwx70pq6vDU6Fq+/jWSZllr1xlT+54eI8ZVKwZb7XO/paQih2s+ttaWS1ptjFmuqobhAndKDM8Yk66q/yiYbq19xet6AACAe/gcAAAAkDystaeFes4Y84Mxpln16sFmkpy2wDtJ0m+MMddIypGUYYzZZa11PJR3r5Q/pERVTb72xpg2kr6XdJGkSw645jVJF0uabIxpqKqR41WuVhmCqWoHPy1pqbX2717XAwAA3MPnAAAAkIpSoFcVyhuSBku6p/qfrx94gbX2j3u/NsZcKqlHpOagJPki/tI2wW+R/u9ZWyFpmKTZkpZKmmGtXWKMudMY07f6stmSthhjvpI0R9It1totkV/dFSeraq78VGPMourbWV4XBQAAXMHnAAAAkFKsqk4xTtTbIbpHUh9jzEpJfarvyxjTwxjz1KG8MCPGkqy1syTNOuCxMft8bSXdVH2LK9baD+W8jyIAAEhyfA4AAABIHdWL1YIOHrHWfippqMPjkyVNrs1rR2gQmmQ4xRgAAAAAAABJIpDCM8axEnkFIZkDAAAAAAAgHtiU3oMwZiLvQQgAAAAAAAAgaYVfQWjFiDEAAAAAAADiwt5DShBdjBgDAAAAAAAgYVhmjKOuFqcYs4IQAAAAAAAA8YEVhNHHHoQAAAAAAABACmPEGAAAAAAAAAmBPQhjgwYhAAAAAAAAEoalWRV1jBgDAAAAAAAAKSz8CkIryXJICQAAAAAAAOKAZcQ4FiKOGHNyNAAAAAAAAOIFvaroY8QYAAAAAAAASGEcUgIAAAAAAICEUHWKMc2qaKtFg5A9CAEAAAAAABAf2IMw+hgxBgAAAAAAAFJYxBWEhq4sAAAAAAAA4gQTxtEXvkFoxR6EAAAAAAAAiAvWWvYgjAFGjAEAAAAAAIAUFmHE2HBICQAAAAAAAOIGCwijrxanGLtQBQAAAAAAAFALAa8LSEKMGAMAAAAAAAApjBWEAAAAAAAASAhW4pCSGKBBCM0uWhTxmjOad3OhksRTm+wk8nNCdoeG/A4N+R08sjs0tcnv+DOKXagEAAAgcdEfjD5GjAEAAAAAAIAUFn4FoRWnGAMAAAAAACBuBFhBGHURR4wNoQMAAAAAACAOWCtZZoyjLvIehEAUnXPt73T64N5q3SVf7/9znu6/fGLQNQPG9NfgcRfq1j536vP3vvSgSveFyqXJEY00bfWj2rOrpObaF+97TdPvetmrUt2VPUATPxng+H7JrJOhP/1tkHqdf5LS0tP0zRdrNKL3WA+Ldcfe94pp0koqeUt2+8iqJ7L6ytS786cLjU/G1FH7X4zUys9WeVOsi8L9bOl5/kkaPO4CNWyZp03rtuiZ0c9Lqt0+eoksXCZnDjlVF448V7lND9fiD5fpgSGPasv6Hz2s1h3pGWm67tEr9IvfdlHd3BwVfb1Bz4x+Xgverno/dD/1GA2bMFSN8xtq2fyVuv+yidr47WaPq04ugT1bVFE0T4HiTVJlibK6XRv++uJNKl83R7bkR5msBkpvdYp82Y1cqtY95OKMXIKRiTNyCUYmwbZu3aqr/jRE7xW8o7yGDXXnXX/VRRdfEnSdtVa3/7/bNPmZpyRJgy8borv/eq+MYboSyS3yHoQ2wW+IK1uKtmr63S9r9rNzHJ9v1raJev7hJG0p2upyZd6KlEu/BoPVt95A9a03MHWag5JUuTFkLjdMulJ1G+RoSKcbdV7eZXr8pikeFOi+ve8V7Zm5/xMlb8hu7PbTbcc42YpvU6I5KIX+M5TXPFe3Tb1ej4+YonPqD9ITt07VqOnDJV+uR5W6J1Qmx/bspMvuvkRj+92r8/Iu04Y1G/X/nr/Boyrd5U/za9O6zRrRe6z6HT5Yk8e8oNtfvElNjmikenl1NfblWzRlzAs6L+8yrVi4Sre/cKPXJScf45Pv8COVnn9KxEttoFJlq2fJ36CDMrsMlT+3o8pWz5INVLpQqMvIxRm5BCMTZ+QSjEyC3HD9tcrIyNDa73/Qs1Oma/iwq/XVkiVB1z395BN6843XNH/hF/rks0L9e9ZbeuqJSR5UjHACNnFv8YpDSuCqD1/9RP97fYF2bNnp+PywR4boqdumqbyswuXKvBUpl5RV+o5jLi07NNdJfXvowSsnafvmHQoEAinTCNv7XlFgW9jrTNa5sntedakq74X6M9SoZa52bdtds0Lsk1mfqWR3qeTP96JMV4XK5MSzj9N/Z36ktV99p4ryCk3/80wd26uTmrVt4lGl7ikpLtXU8S/ph7WbZK3V/H99pg2rN6r9cW316/NO0Jol6zR35scqLy3X1HEz1LZra7Xq2NzrspOKL6uB0vI6yWRFbtIHdn0vycrfqKuMz6+0Rl2rH/8uxlW6j1yckUswMnFGLsHIZH+7d+/Wa6+8rLHj/qycnByd/Otf6/9+31fPT58adO20qVM0/IYRatmypVq0aKHhN4zQ1Ocmu180wvK6yUeDEIihnv1PVEVZhT759+delxJ3pq95TM9/+7hufvoa1cur63U5njvqhCO1ce1mDRp/gWZufFpPfPGAfn3eCV6XFT98zaWMX0olr3ldiedWfLpK65Z+r5PO7iGfz6dfnfNLlZeWSxXLvS7NM8YYad8RmeqvWx/TyqOKvHN44/pq2aGZ1i5Zp9adW2lV4dqa50qKS1X0zQYd0Tn1cokXtmSrfFl5+410+erkyZak1pTBgcjFGbkEIxNn5BIsFTJZuWKF/H6/2nfoUPNYl65dtfSr4BWES79aoi7Hdv3pumOdrwOSTcQGobGJfUNiyDosS5fffYkeu3Gy16XEle2bd+raX47UH1tfrWt6jFSdunU0atr1XpfluUYt89SmS752by/WRS2u1ITrntatk4cp/6gWXpcWH+qcK5V/KlUmz9/6HqxAIKCCqR9o1PThmlXyvEZNH64Hr3pCsnu8Ls0zn8z6TL3OP0ltuuQrIytDA8b0VyAQUFZ2ptelucqf5teoadfrnec+0LrlRcrKydLu7cX7XVO8vVjZdet4VCEUKJf8Gfs/5s+QrSz3pp54QS7OyCUYmTgjl2ApkMmu3btUv379/R6rX6++du4MnuDatWv/a+vXr69du3ZxKEYcsaraKzJRb/Eq8iEllo04EXuDx1+gd6fN1YY1G70uJa6U7C7RioVVo7PbNm7XhOue1oz1Tyq7bh0V70zdBkfpnjKVl1Vo+l0vK1AZUOHcr7RozmIdd3pXfbvse6/L85yp00929+NelxEXuv+2i664d4BuPmWsVn62Wu2Pa6s7Xx8ppX0kVSz1ujxPfP6fxXpu3AyNnXmzsutn65UH39KenSXa9N0Wr0tzjTFGI5+7ThVlFZow7GlJUsmuEmXX278ZmF0vO6V/1kZD5dblKv/ufUmS77Dmymh3du2/2ZcuVZYd8ILlMv706BXoEXJxRi7ByMQZuQQjk/ByDsvRjh079ntsx84dqls3eDorJ2f/a3fs2KGcnBwOKYkzAa8LSEKcYoy40P3ULmrYMldnX32GJKl+o3q6/cWbNOO+1/Tifa97XF38qPnbhhT/l9PqfcYAcYD0X0i+xlLJ215XEhfadWutwrlLaxrtKz79Rsvmr9RJp/4qZRuEkvTGo7P1xqOzJUkt2jfTJaP/oDWL13lclXtGPHW1GjSpr9H/91dVVlRtwL5myTqdPqhXzTVZ2Zlq1q6J1i5JnVxiwZ/bUf7cjgf1vSYrV4FNi2StrfmPssCezfI3PCaaJXqCXJyRSzAycUYuwcgkvPYdOqiiokJfr1ypI9u3lyR9+cUXOrpT56Brj+7UWV8WfqFfHn981XWFztcBySb8iLFNghviis/vU3pmunx+335f33LaeF3RZYSu6n6Lrup+i7YUbdVDV03S6xNne12yK0LlctTxR6plh+Yyxqhubo6ufehyLZqzWMU7iiO/aFLwO+ZSOHepNn67WRePOlc+v0+df9VRXXt31qezF3ldcMztzaHqx7dPUoYkf83zps65Uulsye72qEJvhPoztGLBN+rym6PUrmtrSVUNwy6/OTol9iAMlUl6ZrpaV++r16hVQ9046Uq99vAs7dqWGu+Z4Y9dofyjW+iOvveqrOSn1RLzXv1ErY/J16/PO0HpmekaMKa/Vheu1brlRR5Wm3ystbKBCslW/b2/DVSEPCXTl9NCklHl5kLZQKUqNhVWP97SrXJdQy7OyCUYmTgjl2Bksr/DDjtM55x7nu4cP0a7d+/W/+bN01tvvq5L/jgw6No/Dhikhx/6u77//nsVFRXpoQcf0MBBl7pfNMLyekw4RUeMXagCKeOPt/9Bg8ZeUHP/tIE99dz4GZo6/qX9rgtUBrTzx90q2V3idomeCJXLd8uLdNndl+jwxvVUvGOPPiso1F8uecjDSt1lcq7RrD0/7bm47/tlbL97ddOTV+vCkf20ce1m3Td4Qkr8h/yB7xVTp5/srodldz0iKUPKOkt22zDvCvRIuJ8tU8e/pDteGqEGTepr+6Yd+udfX9EVN3/oYbXuCJXJKw/+S6OmD1ezdk20Z2eJZk+eo8l3vOhhpe5pnN9Qv7/ydJWVlGnG+idrHn/wqkn6z/Mfanz/v2nYI0N029TrtWz+St198YMeVpucbNlOlS396cTI0sJJUnpdZXUeJEkq++ZN+XKaKa1JDxmfXxltzlL5ujmqKPpIJquBMtqcJePzh3r5hEUuzsglGJk4I5dgZBLsoUce1ZVXXK785o2Vm5enhyY8pk6dO+vDD/+rfr8/U5u37ZIkDf3TlVq9epV+2b2LJOnSy4dq6J+u9LJ0HMhKcdxnS1gmXPcys1Ur2+KmG10sJ/pW3zRiobW2h9d1xKsexthP1h8Z8bozmndzoZr4UGBnqo/pX6trZxfVbsVaMuT3c3KpjWTJLtq51FY85+dVJj+HF/klQi61Ec3skiWTn6M2+R1/xjp9+kWJa3tJ+LIb28yOF0S+EACAGPhxwQSvS4g7J5/QQwsXfpra+0qF0bjdMfYP983wuoyD9nj/znHZp4q4gpCTgAEAAAAAABAPrKwCLCGMOkaMAQAAAAAAkDDoD0Yfpxgj7kc44xnZHTyyOzTkd2jI7+CR3aGpTX4r7BYXKgEAAAB+wgpCAAAAAAAAJIx4Pg04UYVtEBrLHoQAAAAAAACIH/QHo8/ndQEAAAAAAAAAvFOLEWNO1gYAAAAAAID3rBWnGMcAexACAAAAAAAgYdCqir6IDUL2IAQAAAAAAEC84JCS6GMPQgAAAAAAACCFMWIMAAAAAACAhGAlBehVRV34BqFlxBgAAAAAAADxgxHj6GPEGAAAAAAAAEhhjBgDAAAAAAAgYbCAMPpoEAIAAAAAACAxWMuIcQwwYgwAAAAAAACksIgrCDmkBAAAAAAAAPGAU4xjI/KIMQAAAAAAABAnGDGOPkaMAQAAAAAAgBTGISUAAAAAAABIGLSqoi98g9CyByEAAAAAAADiR4AR46hjxBgAAAAAAABIYYwYAwAAAAAAICFYW3VDdNEgBAAAAAAAQMLgFOPoY8QYAAAAAAAASGFhVxAacUgJAAAAAAAA4gcLCKOPEWMAAAAAAAAkBCtOMY4FRowBAAAAAACAFBZ+BaFlxBgAAAAAAADxgwWE0Rd5BaFN8FstGGN+Z4xZboz52hhzW5jr+htjrDGmR+1eOfaMMVnGmE+MMV8YY5YYY8Z7XRMAAHAPnwUAAEBKsVY2gW+HwhiTa4wpMMasrP5ngxDX5Rtj3jHGLDXGfGWMaR3ptVN+xNgY45c0UdKZkjpJutgY08nhurqSrpc0390KIyqVdKq1tqukbpJ+Z4w50eOaAACAe/gsAAAAkBpuk/Setba9pPeq7zt5TtL91tqjJR0vaWOkF2YFYVVQX1trV1lryyS9IOkch+v+LOk+SSW1elWX2Cq7qu+mV99YbAsAQIrgswAAAEg1AZu4t0N0jqQp1V9PkdTvwAuqF72lWWsLJMlau8taWxzphSOeYpwCexC2kLRun/vfSTph3wuMMd0ltbLWvmWMudnN4mqjehXkQklHSpporY23VY4AACCG+CwAAEhUDX45zOsS4k7p8m+9LiGuVa0HS/5mVQhNrLXrJclau94Y09jhmg6SthljXpHURtK7km6z1laGe+GIDcIk0NAY8+k+95+w1j6xz33j8D017zRjjE/SPyRdGpvyDl31b3I3Y8zhkl41xhxjrV1cc4Ex4ySNDfX9BXZmzGtMNGTijFyckUswMnFGLsHIJFg9OW4lE9ahfBZolpatrQdXKgAAAH6+sH0qY8y7kpo6fN/oWr5+mqTfSOou6VtJL6qqp/V0pG8KL/GbsputteEOFflOUqt97reUVLTP/bqSjpH0vjFGqvpNesMY09dau+9vqOestduMMe9L+p2kxfs8MU7SOMdvMsb2Mf1jX1wCKbAzRSbByMUZuQQjE2fkEoxMQrDvHfy3HsRngfXZjW3mQf+KAAAA7kvwU4zD9qmstaeFes4Y84Mxpln16sFmct5b8DtJn1trV1V/z2uSTlSEBmH4PQi93j/QnT0IF0hqb4xpY4zJkHSRpDdqIrB2u7W2obW2tbW2taSPJcVNc9AY06h6tYCMMXUknSZpmbdVAQAAt/BZAAAApBqvTyL26hRjVfWrBld/PVjS6w7XLJDUwBjTqPr+qZK+ivTCKX+KsbW2QtIwSbMlLZU0w1q7xBhzpzGmr7fV1UozSXOMMYWqehMUWGvf8rgmAADgHj4LAAAApIZ7JPUxxqyU1Kf6vowxPYwxT0k1W8/cLOk9Y8yXqtpa78lIL8whJZKstbMkzTrgsTEhru3tRk21Za0tVNVcOQAASEF8FgAAAKkmCqcBJyRr7RZJv3V4/FNJQ/e5XyDp2J/z2qmwByEAAAAAAACSgLWKxqguDpDyI8YAAAAAAABAKmPEGAAAAAAAAAmDBYTRx4gxAAAAAAAAEkaADmHUMWIMAAAAAAAApLDwKwitWEEIAAAAAACAuMECwugL2yA01TcAAAAAAADAa1aWU4xjgBFjAAAAAAAAIIVxSAkAAAAAAAASg2XEOBYiNggNoQMAAAAAACBOMGIcfawgBAAAAAAAQMKgPxh97EEIAAAAAAAApDBWEAIAAAAAACBhMGIcfeEbhJY9CAEAAAAAABAfrGgQxgIjxgAAAAAAAEAKY8QYAAAAAAAAicFySEksRGwQMmIMAAAAAACAeMGIcfQxYgwAAAAAAACkMEaMAQAAAAAAkDBYQBh9jBgDAAAAAAAgQVhGjGOAEWMAAAAAAAAghYVfQWjFiDEAAAAAAADighUjxrHAHoQAAAAAAABIDJZTjGOBEWMAAAAAAAAghYVdQWjEISUAAAAAAACIHywgjD5GjAEAAAAAAJAwGDGOPkaMAQAAAAAAgBQWcQWhoSsLAAAAAACAOEGrKvrCNwitGDFGVKVnpOm6R6/QL37bRXVzc1T09QY9M/p5LXh7kdLS0zRq+nB16NFWTVs31ohTxqrwg6+8LtkV4XI5+oT2GnznRWp/XFsFKgMqfH+JJg5/Rls3bPO67JgKl0n+0S01csowNWvXRJK0cuEqTRz+rL5d+p3HVcdeuFz2NWBMfw0ed6Fu7XOnPn/vS4+qdUe4TJoc0UjTVj+qPbtKaq5/8b7XNP2ulz2s2B2R3iuZdTL0p78NUq/zT1Jaepq++WKNRvQe63HVsRUuk1Mv+bVuePzKmmuNzygrO1PX9BiplZ+t8rDq5BLYs0UVRfMUKN4kVZYoq9u14a8v3qTydXNkS36UyWqg9FanyJfdyKVq3UMuzsglGJk4I5dgZOKMXJKDFSPGscCIMVzlT/Nr07rNGtF7rPodPliTx7yg21+8SU2OqPohu3jeUt0z8BFtWf+jx5W6K1wuOQ0O06wnCzSwzTUa0PpqFe/ao5ufCf8vsmQQLpMtRVt15/kP6Ly8y9S/0RB99OanGv3PG7wu2RWR/gxJUrO2TdTzDydpS9FWDyt1T20y6ddgsPrWG6i+9QamRHNQipzLDZOuVN0GORrS6Uadl3eZHr9piscVx164TP7z/Ic175G+9QbqkWufUtE3G2gORpvxyXf4kUrPPyXipTZQqbLVs+Rv0EGZXYbKn9tRZatnyQYqXSjUZeTijFyCkYkzcglGJs7IBQgpYoPQ2MS+Ib6UFJdq6viX9MPaTbLWav6/PtOG1RvV/ri2qiiv0KsPzdKSecsUqAx4XaqrwuWy4O1FmjvzYxXv3KPSPWV6fcLb6nxyR69LjrlwmezeXqwf1m6qutBIgcqAmh/Z1NuCXRIul72GPTJET902TeVlm9fE8QAAH8ZJREFUFR5W6p7aZJKKwuXSskNzndS3hx68cpK2b96hQCCQEo2wn/Ne6TOolwqmfuBBlcnNl9VAaXmdZLJyI14b2PW9JCt/o64yPr/SGnWtfjz5VouTizNyCUYmzsglGJk4I5ckYatGjBP1Fq84xRieOrxxfbXs0Exrl6zzupS4Ei6XLj2PTsm8nDJ5detk1cnJkvEZTRn7oofVeefAXHr2P1EVZRX65N+fe1yZd5zeK9PXPCZrrT4rKNQTt07Vji07PazQG/vmctQJR2rj2s0aNP4CnTagp7au36bnxs/Qh6/M97pMV4X6Wds4v6G69OykB4Y85lFlkCRbslW+rDwZY2oe89XJky3ZKtU7wsPKvEUuzsglGJk4I5dgZOKMXOIbI8bRx4gxPONP82vUtOv1znMfaN3yIq/LiRvhcmnTJV8D7jhfT9w61aPqvBEqk3NzL1W/wwdrwnVP6+vP13hXoEcOzCXrsCxdfvcleuzGyV6X5pkDM9m+eaeu/eVI/bH11bqmx0jVqVtHo6Zd73WZrjswl0Yt89SmS752by/WRS2u1ITrntatk4cp/6gWXpfqmnA/a/sM6qXF/12qDWs2elQdJEmBcsmfsf9j/gzZynJv6okX5OKMXIKRiTNyCUYmzsgFKaYWpxi7UQZSjTFGI5+7ThVlFZow7Gmvy4kb4XJp3q6p/jJrtB674Vkt/nCZRxW6L9J7paS4VG89XqCZG5/WkE43aNumHR5U6T6nXAaPv0DvTpubsk0Np0xKdpdoxcKq0dltG7drwnVPa8b6J5Vdt46Kd+7xslzXOOVSuqdM5WUVmn7Xy1WHH839SovmLNZxp3fVt8u+97ji2Iv0c6XPwF76519f8aCy5FO5dbnKv3tfkuQ7rLky2p1d+2/2pUuVZQe8YLmMPz16BXqEXJyRSzAycUYuwcjEGbkkLxYQRh8jxvDEiKeuVoMm9TX6//6qygo2ed0rVC6N8xvq3oI7NP2umXp32lwPK3Rfbd4rxmeUmZ2pvBa5KdMgdMql+6ld1LBlrs6++gxJUv1G9XT7izdpxn2v6cX7XveyXFfU5r1SM4qwz6hIsnPKZXXhWo+r8la490rnX3VUbvMGmjvzY4+qSy7+3I7y5x7cvrkmK1eBTYtkra0Z7wrs2Sx/w2OiWaInyMUZuQQjE2fkEoxMnJFLsrKMGMcAI8Zw3fDHrlD+0S10R997VVay/9/IpGekKT0zPejrVBAql7zmubr/vbF649HZemtSgYcVui9UJr847Vi169ZaPp9P2XXr6Kq/D9auH3fp26XJv/JJCp3LLaeN1xVdRuiq7rfoqu63aEvRVj101SS9PnG2h9W6I1QmRx1/pFp2aC5jjOrm5ujahy7XojmLVbyj2MNq3RMql8K5S7Xx2826eNS58vl96vyrjurau7M+nb3Iw2rdEe7fQZLUZ3BvffjyfO3ZVeJBdcnPWisbqJBs1WFkNlAR8jRIX04LSUaVmwtlA5Wq2FRY/XhLt8p1Dbk4I5dgZOKMXIKRiTNyAUILv4KQk4ARZY3zG+r3V56uspIyzVj/ZM3jD141Sf95/kM9s+whNW3dWJJ0z+w7JEkD2lzz04m1SSpcLs2PbKrm7Zpq4JjzNXDM+TXP9a030ItSXRMuk4qyCl378OVq1DJPpXvKtGLB1xp15t0qL03+/UAi/RnaV6AyoJ0/7lbJ7uRudITLxAasLrv7Eh3euJ6Kd+zRZwWF+sslD3lYrXsivVfG9rtXNz15tS4c2U8b127WfYMnJP1+sJEySc9MV6/zT9Kd/f/mYZXJzZbtVNnSn/bRLS2cJKXXVVbnQZKksm/elC+nmdKa9JDx+ZXR5iyVr5ujiqKPZLIaKKPNWTI+v1flxwy5OCOXYGTijFyCkYkzckkOVhxSEgsmXKiH5bWyx5x1o4vlRN8n00YstNb28LqOeNXDGNvA9Pe6jLhSYGeqD5kEIRdn5BKMTJyRSzAycTbfvqcddqtrc/C+7MY2s+MFbv1yAAAggtLlMxQo3pg6e+L8THWadbDthkz0uoyDtuTu0+OyT8WIMQAAAAAAAJDCwo4YGzFiDAAAAAAAgPjBiHH01eIUY0JPdrOLIm9Kf0bzbi5Uknhqk51Efk7I7tCQ36Ehv4NHdoemNvkdf0ZqHKIDAABwsGhVRR8jxgAAAAAAAEAKi7iCkBFjAAAAAAAAxAMrKRCgWRVt4RuEtvoGAAAAAAAAeM0yYhwLjBgDAAAAAAAAKSzyiHHAjTIAAAAAAACAyDjFOPpqcYqxC1UAAAAAAAAAtUB/MPoYMQYAAAAAAABSGKcYAwAAAAAAIEFYRoxjoBanGBM6AAAAAAAAvEerKjZYQQgAAAAAAIDEYDmkJBbYgxAAAAAAAABIYZxiDAAAAAAAgITBCsLoC9sgNGLEGAAAAAAAAHGEXlXUMWIMAAAAAAAApLAIpxhbjoZJAWc07+Z1CQmL7A4e2R0a8js05HfwyO7Q1Ca/FXaLC5UAAAAkLkaMo49TjAEAAAAAAJAwaBBGHyPGAAAAAAAAQArjFGMAAAAAAAAkBGstKwhjgBFjAAAAAAAAJAwahNHHiDEAAAAAAACQwiKcYiwpQFcWAAAAAAAAcYJWVdSxByEAAAAAAAASBiPG0ceIMQAAAAAAAJDCOKQEAAAAAAAACYMVhNFXixFjQgcAAAAAAEAcsDQIY4ERYwAAAAAAACCFRWwQGpvYt9owxvzOGLPcGPO1MeY2h+dvMsZ8ZYwpNMa8Z4w54udHHVvGGL8x5nNjzFte1wIAANzF5wAAAJBSbALfDoExJtcYU2CMWVn9zwYhrrvPGLPEGLPUGPOwMcZEeu3wDUKvQ3MheGOMX9JESWdK6iTpYmNMpwMu+1xSD2vtsZJmSrov8iu7brikpV4XAQAAPMHnAAAAkBKsrKxN3Nshuk3Se9ba9pLeq76/H2PMrySdLOlYScdI+qWkXpFemBFj6XhJX1trV1lryyS9IOmcfS+w1s6x1hZX3/1YUkuXawzLGNNS0v9JesrrWgAAgLv4HAAAAJAyzpE0pfrrKZL6OVxjJWVJypCUKSld0g+RXjjsISVGkkn8jR8bGmM+3ef+E9baJ/a530LSun3ufyfphDCvN0TSv6NYXzQ8KOlWSXUdnzVmnKSxob65wM6MSVGJjEyckYszcglGJs7IJRiZBKsnx0mRcMJ/DpDCfhZolpatrT/3VwQAAPBQgh9SEqlPFU4Ta+16SbLWrjfGND7wAmvtR8aYOZLWq6q1N8FaG3HSJPIpxoFalhi/Nltre4R53mkO2/GdZowZIKmHarE00y3GmN9L2mitXWiM6e14kbXjJI0L8QK2j+kfo+oSU4GdKTIJRi7OyCUYmTgjl2BkEoJ9r9aX1upzgBT2s8D67MY282cVCAAA4K0EbxCG7VMZY96V1NThqdG1eXFjzJGSjtZP068Fxpie1tq54b4vcoMw+X0nqdU+91tKKjrwImPMaar6zehlrS11qbbaOFlSX2PMWapaQlrPGDPNWjvA47oAAEDs8TkAAAAgiVhrTwv1nDHmB2NMs+rVg80kbXS47FxJH1trd1V/z78lnSgpbIOwFqcY24S+1cICSe2NMW2MMRmSLpL0xgG/Ad0lTZLU11rrFL5nrLWjrLUtrbWtVVX7f/iPAgAAUgOfAwAAQMqx8vygEQ8PKXlD0uDqrwdLet3hmm8l9TLGpBlj0lU1BRtxxDjlTzG21lZIGiZptqoCm2GtXWKMudMY07f6svsl5Uh6yRizyBjzRoiXAwAAAAAAQCx53WuKYZ8qgnsk9THGrJTUp/q+jDE9jDF7D6ybKekbSV9K+kLSF9baNyO9MCPGkqy1syTNOuCxMft8HXJ5Zzyx1r4v6X2PywAAAB7gcwAAAEBys9ZukfRbh8c/lTS0+utKSVf+3NeO0CC0UmJv/AgAAAAAAIAkkuCHlMSliCsIDZkDAAAAAAAgDlhFZS8/HCDiISUAAAAAAAAAklfkPQjpygIAAAAAACBOsIIw+sI3CK1kAi5VAgAAAAAAAIQTndOAcQBGjAEAAAAAAIAUxogxAAAAAAAAEgYjxtFXiwahC1UAAAAAAAAAtUCDMPoYMQYAAAAAAABSWMQVhIauLAAAAAAAAOIEKwijjz0IAQAAAAAAkBCsLA3CGGDEGAAAAAAAAEhh4VcQWkkBdwoBAAAAAAAAwrLiQN0YCNsgNLLsQQgAAAAAAIC4wYhx9LEHIQAAAAAAABIGDcLoYw9CAAAAAAAAIIWxghAAAAAAAAAJgxWE0cchJQAAAAAAAEgQlgZhDDBiDAAAAAAAAKSwiCPGnGIMAAAAAACAuGCrb4gq9iAEAAAAAABAQrBiD8JYYMQYAAAAAAAASGERVhBaVhACAAAAAAAgbrCCMPoin2JM6AAAAAAAAIgX9KqijhFjAAAAAAAAIIVFPqQk4EIVAAAAAAAAQCTWSpZmVbRFbBAalm0CAAAAAAAgXtCrijpGjAEAAAAAAIAUFnnEmK4soig9I03XPXqFfvHbLqqbm6OirzfomdHPa8HbiyRJPc8/SYPHXaCGLfO0ad0WPTP6ef3v9QUeVx17kXI5c8ipunDkucptergWf7hMDwx5VFvW/+hx1bE38rnr1P23XZR1WKZ+3LBNM+5/Xf9++j+SpO6nHqNhE4aqcX5DLZu/UvdfNlEbv93sccXuCJVLWnqaRk0frg492qpp68YaccpYFX7wldfluiJUJkef0F6D77xI7Y9rq0BlQIXvL9HE4c9o64ZtXpfsilC55B/dUiOnDFOzdk0kSSsXrtLE4c/q26XfeVxx7IX7ubLXgDH9NXjchbq1z536/L0vPao0OQX2bFFF0TwFijdJlSXK6nZt+OuLN6l83RzZkh9lshoovdUp8mU3cqla95CLM3IJRibOyCUYmTgjlyTCiHHUhV9BaCUFbGLfEFf8aX5tWrdZI3qPVb/DB2vymBd0+4s3qckRjZTXPFe3Tb1ej4+YonPqD9ITt07VqOnDdXijel6XHXPhcjm2ZydddvclGtvvXp2Xd5k2rNmo//f8DV6X7IoX7nlVA9tco36HD9aYc+7VpX++WO1/0Vb18upq7Mu3aMqYF3Re3mVasXCVbn/hRq/LdU2oXCRp8bylumfgIynRQN5XqExyGhymWU8WaGCbazSg9dUq3rVHNz8T/oNgMgmVy5airbrz/Ad0Xt5l6t9oiD5681ON/mdq/1zZq1nbJur5h5O0pWirh1UmMeOT7/AjlZ5/SsRLbaBSZatnyd+ggzK7DJU/t6PKVs+SDVS6UKjLyMUZuQQjE2fkEoxMnJFL8rA2cW9xihFjuKqkuFRTx7+kH9ZukrVW8//1mTas3qj2x7VVo5a52rVtd82quU9mfaaS3aVq1q6px1XHXrhcTjz7OP135kda+9V3qiiv0PQ/z9SxvTqpWdsmXpcdc2u/+k7lZRWS9v4stWrWrol+fd4JWrNknebO/FjlpeWaOm6G2nZtrVYdm3tcsTtC5VJRXqFXH5qlJfOWKVCZWn+jFiqTBW8v0tyZH6t45x6V7inT6xPeVueTO3pcrXtC5bJ7e7F+WLup6iIjBSoDan5k8v+slUJnstewR4boqdum1VyD6PJlNVBaXieZrNyI1wZ2fS/Jyt+oq4zPr7RGXasfT76VruTijFyCkYkzcglGJs7IBQgtwohxfHc3kfgOb1xfLTs009ol6/T9yg1at/R7nXR2D83/12c68ezjVF5artWFa70u03X75tL5Vx0lY356svrr1se00vpVP3hUoXuumzhUpw/urazsTK38bJU+mfW5Lr/7Yq3a531RUlyqom826IjOrbRueZGH1brHKZdUV5tMuvQ8WmuXrPOgOu+Ey+XVrZNVJydLxmc0ZeyLHlbprlCZ9Ox/oirKKvTJv/nzFA9syVb5svJk9vl3oK9OnmzJVqneER5W5i1ycUYuwcjEGbkEIxNn5BLPOMU4FtiDEJ7xp/k1atr1eue5D2qaOgVTP9Co6cOVkZWu8rIK/fmCv6ukuNTjSt11YC6fzPpMo1+4UW89/o6+X7lBA8b0VyAQUFZ2pteluuKRa5/SxOue0dEndVDX3p1UXlqurJwsbd+0Y7/rircXK7tuHY+qdJ9TLqkuUiZtuuRrwB3na2y/ez2q0Bvhcjk391JlZWeqz+Be+mFtauzhKYX4uXJYli6/+xLddsZdXpeHvQLlkj9j/8f8GbKVKf7zjlyckUswMnFGLsHIxBm5xC8rGoQxwIgxPGGM0cjnrlNFWYUmDHtaktT9t110xb0DdPMpY3Vm5sUa0XusbnryKrXr2trbYl3klMvn/1ms58bN0NiZN2vamkf1w5qN2rOzRJu+2+Jxte4JBAJaMm+ZGrXM09lXn66SXSXKrrd/MzC7XraKd+7xqEJvHJgLQmfSvF1T/WXWaD12w7Na/OEyDyv0Rrj3Sklxqd56vEAjpwxLiT1f9zowk8HjL9C70+Zqw5qNXpeWVCq3LldJ4SSVFE5S2Tdv/rxv9qVLlWUHvGC5jD89egV6hFyckUswMnFGLsHIxBm5ALXHCkJ4YsRTV6tBk/oa/X9/VWVF1Sav7bq1VuHcpVqxcJUkacWn32jZ/JXqfloXffPFGg+rdY9TLpL0xqOz9cajsyVJLdo30yWj/6A1i1NrTFKqWl3ZvF1TrVmyTqcP6lXzeFZ2ppq1a5Jyo6N77c0FP9k3k8b5DXVvwR2aftdMvTttrseVeSvUe8X4jDKzM5XXIlfbDlidm+z2ZnJsz05q2DJXZ199hiSpfqN6uv3FmzTjvtf04n2ve1xl4vLndpQ/9+D2/TRZuQpsWiRrbc14V2DPZvkbHhPNEj1BLs7IJRiZOCOXYGTijFySGL2qqOMUY7hu+GNXKP/oFrqj770qK/npb2RWLPhGXX5zVM2KwXbdWqvLb45OmT0IQ+WSnpmu1p1bSZIatWqoGyddqdcenqVd23Z7VaorDm9UT70v/JWyDsuSz+dTj9O7qvdFJ+vz/3ypea9+otbH5OvX552g9Mx0DRjTX6sL16bE/oPhcpGk9Iw0pWemB32dzMJlktc8V/e/N1ZvPDpbb00q8LpUV4XL5RenHat23VrL5/Mpu24dXfX3wdr14y59u/R7r8uOqXCZ3HLaeF3RZYSu6n6Lrup+i7YUbdVDV03S6xNne112UrHWygYqasaCbKAi5GmQvpwWkowqNxfKBipVsamw+vGWbpXrGnJxRi7ByMQZuQQjE2fkkiyq9yBM1FuciryCEIiixvkN9fsrT1dZSZlmrH+y5vEHr5qk/zz/oaaOf0l3vDRCDZrU1/ZNO/TPv76ihQWFHlbsjnC5zP/XZxo1fbiatWuiPTtLNHvyHE2+I/kPE7BWOvuqMzT8sT/J+Iw2rt2sx26crI/e+FSSNL7/3zTskSG6ber1WjZ/pe6++EGPK3ZHpFyeWfaQmrZuLEm6Z/YdkqQBba756cTaJBQukwFj+qt5u6YaOOZ8DRxzfs339K030MOK3REul579T9S1D1+uRi3zVLqnTCsWfK1RZ96d9HtZRvrzs69AZUA7f9ytkt0lHlSavGzZTpUtnVpzv7RwkpReV1mdB0mSyr55U76cZkpr0kPG51dGm7NUvm6OKoo+kslqoIw2Z8n4/F6VHzPk4oxcgpGJM3IJRibOyAUIzdgwyzLrZzaxv2r+RxfLib631/xjobW2h9d1xKsextgGpr/XZcSVAjtTfcgkCLk4I5dgZOKMXIKRibP59j3tsFtN5Cujw5fd2GZ2vMCtXw4AAERQunyGAsUbXfsskGh89VvZzJNu9LqMg1Yye0Rc9qnYgxAAAAAAAACJI45HdRMVpxgDAAAAAAAAKSz8CsK9h5QAAAAAAAAAXrOWadcYYMQYAAAAAAAAiYMR46hjxBgAAAAAAABIYawgBAAAAAAAQOKgVxV1ERqEzHUDAAAAAAAgXlhGjGOAEWMAAAAAAAAghdXiFGO6sgAAAAAAAIgTTLtGHXsQAgAAAAAAIDFYMWIcA4wYAwAAAAAAACmMFYQAAAAAAABIHAF6VdEW+RRjQgcAAAAAAEBc4BTjWGDEGAAAAAAAAEhhEU8xtnRlAQAAAAAAEC/oVUVd5D0IGTEGAAAAAABAPLDivIwYYMQYAAAAAAAASGGcYgzNLloU8ZozmndzoZLEU5vsJPJzQnaHhvwODfkdPLI7NLXJ7/gzil2oBAAAIFFxSEksRNiD0EoBQgcAAAAAAECcYDFb1LGCEAAAAAAAAImDFYRRxx6EAAAAAAAAQAqLuILQMmIMAAAAAACAuGCZdo2BCA1CQgcAAADw/9u7v1C/6zqO489Xm2auTEipzY1mJN10kTIWIkiUxiJxFUYLkoiimxaGRFhd9Oeum+iiiGQzjKwlM2mYuAqT8iKbM0PnNNYQPKxYYlFbhNjeXZyvcfT7Pef8Vtv5fLfv8wFjv+/v9+Xsvff5cXj93uf7+XwlSRqJwiXGp4FLjCVJkiRJkqQJW+YuxsAJryCUJEmSJEnSSLja9ZSb4S7GXrYpSZIkSZKkMShnVaeBS4wlSZIkSZKkCVvyCsL5fR+9bFOSJEmSJEkj4RLjU26ZPQi9bFOSJEmSJEkj4qzqlHOJMZBkS5KnkhxKcsvA669M8qPu9YeSbFz5KiVJkiRJkjRVST6Y5ECSE0k2LXHeknOuIcvepORsX2KcZBXwLeBaYA7Yl2RPVT2x4LSPA3+tqjcn2QZ8DfjQylcrSZIkSZI0YcWUlxg/DnwA+M5iJ8w45+rxLsawGThUVYcBkuwCtgILG7cV+HL3eDfwzSSpmu47UpIkSZIkaeVNdzu8qjoIkGSp02aZc/VkqRlXkvuAi06y3rE5D/jXguNbq+rWFw+S3ABsqapPdMc3Am+vqu0Lznm8O2euO/5jd86zK/EfOJ3WJMf/CQdb1zEm62HdHBxpXcfY2Jdh9qXPngyzL332ZFFvrKqLV+ofMwv0+d4cZl/67Mkw+9JnT4bZl0WtaBY405wFs6ol51SzSPIA8NmqenjgtWXnXEOWvotx1ZaTKfAMNTR2ffnUdJZzzkjH4XyqFl23PklJUbWudRmjY1+G2Zc+ezLMvvTZk1EwCwzwvTnMvvTZk2H2pc+eDLMv+h+c7bOqJL8A3jDw0her6iezfImB55adYS2/xPjsNwdsWHC8nv5vMF48Zy7JauC1wHMrU54kSZIkSZKmoKqu+T+/xCxzrh7vYgz7gMuSXJrkXGAbsOdl5+wBPto9vgG43/0HJUmSJEmSNDKzzLl6Jj8grKoXgO3AXub337mzqg4k+WqS67vTdgKvS3IIuBmY6RbRkiRJkiRJ0qmQ5P1J5oArgZ8m2ds9vy7JvbD4nGvZr+2FcBM3v+fDkre/mRx7Msy+DLMvffZkmH3psyfj4Pehz54Msy999mSYfemzJ8PsizQak7+CUJIkSZIkSZoyB4T6SusCRsieDLMvw+xLnz0ZZl/67Mk4+H3osyfD7EufPRlmX/rsyTD7Io2ES4wlSZIkSZKkCfMKQkmSJEmSJGnCHBBKkiRJkiRJE+aAUJIkSZIkSZowB4QTlWRLkqeSHEpyS+t6xiDJbUmOJnm8dS1jkmRDkl8mOZjkQJKbWtfUWpLzkvw2ye+7nri58gJJViX5XZJ7WtcyBkmeTvJYkkeTPNy6nrFIcmGS3Ume7H6+XNm6pqkxC/SZBfrMAcPMAoszB/SZBYaZBaRx8SYlE5RkFfAH4FpgDtgHfLiqnmhaWGNJrgaOAd+rqre2rmcskqwF1lbVI0leA+wH3jfl90uSAGuq6liSc4AHgZuq6jeNSxuFJDcDm4ALquq61vW0luRpYFNVPdu6ljFJcjvw66rakeRc4Pyq+lvruqbCLDDMLNBnDhhmFlicOaDPLDDMLCCNi1cQTtNm4FBVHa6q54FdwNbGNTVXVb8Cnmtdx9hU1Z+q6pHu8T+Ag8Albatqq+Yd6w7P6f742xYgyXrgvcCO1rVovJJcAFwN7ASoquf9QLDizAIDzAJ95oBhZoFh5gDNyiwgjY8Dwmm6BHhmwfEcBj3NIMlG4HLgobaVtNctn3kUOAr8vKom35PON4DPASdaFzIiBfwsyf4kn2xdzEi8CfgL8N1uGdqOJGtaFzUxZgGdNHPAS5kFBpkDhpkF+swC0sg4IJymDDw3+d94amlJXg3cBXymqv7eup7WqurfVfU2YD2wOcnkl6IluQ44WlX7W9cyMldV1RXAe4BPdUsYp241cAXw7aq6HDgOuAfeyjIL6KSYA/rMAi9lDliSWaDPLCCNjAPCaZoDNiw4Xg8caVSLzgDd3jp3AXdU1Y9b1zMm3VKIB4AtjUsZg6uA67t9dnYB70zy/bYltVdVR7q/jwJ3M7+0c+rmgLkFV9vsZv5DglaOWUAzMwcszSzwX+aARZgFBpkFpJFxQDhN+4DLklzabQa7DdjTuCaNVLcJ907gYFV9vXU9Y5Dk4iQXdo9fBVwDPNm2qvaq6vNVtb6qNjL/c+X+qvpI47KaSrKm29SfbtnMu4HJ3x21qv4MPJPkLd1T7wImfcODBswCmok5YJhZoM8cMMwsMMwsII3P6tYFaOVV1QtJtgN7gVXAbVV1oHFZzSX5IfAO4KIkc8CXqmpn26pG4SrgRuCxbp8dgC9U1b0Na2ptLXB7dxfQVwB3VtU9jWvSOL0euHv+8zWrgR9U1X1tSxqNTwN3dMOpw8DHGtczKWaBYWaBQeaAYWYBzcossDizgDQiqXK7GUmSJEmSJGmqXGIsSZIkSZIkTZgDQkmSJEmSJGnCHBBKkiRJkiRJE+aAUJIkSZIkSZowB4SSJEmSJEnShDkglCRJkiRJkibMAaEkSZIkSZI0Yf8BlBCTYILxb+kAAAAASUVORK5CYII=\n",
      "text/plain": [
       "<Figure size 1296x864 with 8 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "np.random.seed(5)\n",
    "# Specify state space\n",
    "S = NvMDP.state.DiscreteStateSpace(5, 7)\n",
    "goal = (2,6)\n",
    "# Attach classes\n",
    "class_ids = NvMDP.class_.XYClassDistribution(np.arange(5 * 7).reshape(5, 7))().flatten()\n",
    "S.attach_classes(class_ids)\n",
    "# Attach features\n",
    "S.attach_feature_spec(NvMDP.features.FeatureClassIndicatorOneHot(key=\"indicator\"))\n",
    "# Attach rewards\n",
    "S.attach_reward_spec(NvMDP.reward.RewardStateScalar(loc_to_reward_dict={goal: 0.}, \n",
    "                                                    class_id_to_reward_dict={}, \n",
    "                                                    default=-1, \n",
    "                                                    key=\"scalar\"))\n",
    "R = S.rewards(key=\"scalar\")\n",
    "# Specify dynamics\n",
    "T = NvMDP.dynamics.XYDynamics(S, slip_prob=0.)\n",
    "# Set goal\n",
    "# S.set_terminal_status_by_loc([goal])\n",
    "plot_irl_world(S, [], r_key=\"scalar\", phi_key=\"indicator\", v_range=[None, None, None, (0,-1)])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "RewardLinear(\n",
      "  (fc_reward): Linear(in_features=35, out_features=1, bias=False)\n",
      "  (sigmoid): Sigmoid()\n",
      "  (reward_activation): Sigmoid()\n",
      ")\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1296x864 with 8 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "PHI = torch.FloatTensor(S.features(key=\"indicator\"))\n",
    "\n",
    "r_model_linear = RewardLinear(PHI[0].numpy().shape[0], lr=0.2, weight_decay=0.)\n",
    "r_model_linear.init_const(0)\n",
    "print(r_model_linear)\n",
    "preprocess_fn = lambda x: torch.FloatTensor(x)\n",
    "postprocess_fn = lambda x: x.squeeze()\n",
    "\n",
    "S.attach_reward_spec(\n",
    "    NvMDP.reward.RewardStateFeatureModel(\n",
    "        r_model = r_model_linear,\n",
    "        preprocess_fn=preprocess_fn,\n",
    "        key=\"linear\", feature_key=\"indicator\")\n",
    ")\n",
    "\n",
    "plot_irl_world(S, [], r_key=\"linear\", phi_key=\"indicator\", v_range=[None, None, None, (0,-1)])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Setting goal as terminal state!\n",
      "Learning values [  0 ] Converged @ 10.\n",
      "Reasoning [  10 19 ] Done (10 iters).\n",
      "\n",
      ">>> Iter: 0000 (0.307s): loss = 37.110191, likelihood = 0.0000\n",
      "\n",
      "\n",
      "Learning values [  0 ] Converged @ 11.\n",
      "Reasoning [  11 20 ] Done (10 iters).\n",
      "\n",
      ">>> Iter: 0001 (0.321s): loss = 16.945639, likelihood = 0.0000\n",
      "\n",
      "\n",
      "Learning values [  0 ] Converged @ 11.\n",
      "Reasoning [  11 20 ] Done (10 iters).\n",
      "\n",
      ">>> Iter: 0002 (0.302s): loss = 06.913630, likelihood = 0.0010\n",
      "\n",
      "\n",
      "Learning values [  0 ] Converged @ 11.\n",
      "Reasoning [  11 20 ] Done (10 iters).\n",
      "\n",
      ">>> Iter: 0003 (0.302s): loss = 03.699116, likelihood = 0.0247\n",
      "\n",
      "\n",
      "Learning values [  0 ] Converged @ 11.\n",
      "Reasoning [  11 20 ] Done (10 iters).\n",
      "\n",
      ">>> Iter: 0004 (0.299s): loss = 02.138701, likelihood = 0.1178\n",
      "\n",
      "\n",
      "Learning values [  0 ] Converged @ 12.\n",
      "Reasoning [  12 21 ] Done (10 iters).\n",
      "\n",
      ">>> Iter: 0005 (0.315s): loss = 01.609090, likelihood = 0.2001\n",
      "\n",
      "\n",
      "Learning values [  0 ] Converged @ 12.\n",
      "Reasoning [  12 21 ] Done (10 iters).\n",
      "\n",
      ">>> Iter: 0006 (0.314s): loss = 01.460700, likelihood = 0.2321\n",
      "\n",
      "\n",
      "Learning values [  0 ] Converged @ 12.\n",
      "Reasoning [  12 21 ] Done (10 iters).\n",
      "\n",
      ">>> Iter: 0007 (0.313s): loss = 01.417490, likelihood = 0.2423\n",
      "\n",
      "\n",
      "Learning values [  0 ] Converged @ 13.\n",
      "Reasoning [  13 22 ] Done (10 iters).\n",
      "\n",
      ">>> Iter: 0008 (0.330s): loss = 01.404566, likelihood = 0.2455\n",
      "\n",
      "\n",
      "Learning values [  0 ] Converged @ 13.\n",
      "Reasoning [  13 22 ] Done (10 iters).\n",
      "\n",
      ">>> Iter: 0009 (0.337s): loss = 01.401793, likelihood = 0.2462\n",
      "\n",
      "\n",
      "Learning values [  0 ] Converged @ 14.\n",
      "Reasoning [  14 23 ] Done (10 iters).\n",
      "\n",
      ">>> Iter: 0010 (0.349s): loss = 01.403154, likelihood = 0.2458\n",
      "\n",
      "\n",
      "Learning values [  0 ] Converged @ 14.\n",
      "Reasoning [  14 23 ] Done (10 iters).\n",
      "\n",
      ">>> Iter: 0011 (0.355s): loss = 01.406593, likelihood = 0.2450\n",
      "\n",
      "\n",
      "Learning values [  0 ] Converged @ 13.\n",
      "Reasoning [  13 22 ] Done (10 iters).\n",
      "\n",
      ">>> Iter: 0012 (0.340s): loss = 01.411286, likelihood = 0.2438\n",
      "\n",
      "\n",
      "Learning values [  0 ] Converged @ 15.\n",
      "Reasoning [  15 24 ] Done (10 iters).\n",
      "\n",
      ">>> Iter: 0013 (0.357s): loss = 01.416824, likelihood = 0.2425\n",
      "\n",
      "\n",
      "Learning values [  0 ] Converged @ 15.\n",
      "Reasoning [  15 24 ] Done (10 iters).\n",
      "\n",
      ">>> Iter: 0014 (0.351s): loss = 01.422952, likelihood = 0.2410\n",
      "\n",
      "\n",
      "Learning values [  0 ] Converged @ 15.\n",
      "Reasoning [  15 24 ] Done (10 iters).\n",
      "\n",
      ">>> Iter: 0015 (0.351s): loss = 01.429476, likelihood = 0.2394\n",
      "\n",
      "\n",
      "Learning values [  0 ] Converged @ 16.\n",
      "Reasoning [  16 25 ] Done (10 iters).\n",
      "\n",
      ">>> Iter: 0016 (0.358s): loss = 01.436228, likelihood = 0.2378\n",
      "\n",
      "\n",
      "Learning values [  0 ] Converged @ 16.\n",
      "Reasoning [  16 25 ] Done (10 iters).\n",
      "\n",
      ">>> Iter: 0017 (0.363s): loss = 01.443060, likelihood = 0.2362\n",
      "\n",
      "\n",
      "Learning values [  0 ] Converged @ 16.\n",
      "Reasoning [  16 25 ] Done (10 iters).\n",
      "\n",
      ">>> Iter: 0018 (0.361s): loss = 01.449839, likelihood = 0.2346\n",
      "\n",
      "\n",
      "Learning values [  0 ] Converged @ 17.\n",
      "Reasoning [  17 26 ] Done (10 iters).\n",
      "\n",
      ">>> Iter: 0019 (0.373s): loss = 01.456447, likelihood = 0.2331\n",
      "\n",
      "\n",
      "Learning values [  0 ] Converged @ 17.\n",
      "Reasoning [  17 26 ] Done (10 iters).\n",
      "\n",
      ">>> Iter: 0020 (0.385s): loss = 01.462781, likelihood = 0.2316\n",
      "\n",
      "\n",
      "Learning values [  0 ] Converged @ 16.\n",
      "Reasoning [  16 25 ] Done (10 iters).\n",
      "\n",
      ">>> Iter: 0021 (0.355s): loss = 01.468754, likelihood = 0.2302\n",
      "\n",
      "\n",
      "Learning values [  0 ] Converged @ 17.\n",
      "Reasoning [  17 26 ] Done (10 iters).\n",
      "\n",
      ">>> Iter: 0022 (0.396s): loss = 01.474296, likelihood = 0.2289\n",
      "\n",
      "\n",
      "Learning values [  0 ] Converged @ 18.\n",
      "Reasoning [  18 27 ] Done (10 iters).\n",
      "\n",
      ">>> Iter: 0023 (0.388s): loss = 01.479352, likelihood = 0.2278\n",
      "\n",
      "\n",
      "Learning values [  0 ] Converged @ 18.\n",
      "Reasoning [  18 27 ] Done (10 iters).\n",
      "\n",
      ">>> Iter: 0024 (0.409s): loss = 01.483883, likelihood = 0.2268\n",
      "\n",
      "\n",
      "Learning values [  0 ] Converged @ 17.\n",
      "Reasoning [  17 26 ] Done (10 iters).\n",
      "\n",
      ">>> Iter: 0025 (0.367s): loss = 01.487862, likelihood = 0.2259\n",
      "\n",
      "\n",
      "Learning values [  0 ] Converged @ 18.\n",
      "Reasoning [  18 27 ] Done (10 iters).\n",
      "\n",
      ">>> Iter: 0026 (0.389s): loss = 01.491277, likelihood = 0.2251\n",
      "\n",
      "\n",
      "Learning values [  0 ] Converged @ 18.\n",
      "Reasoning [  18 27 ] Done (10 iters).\n",
      "\n",
      ">>> Iter: 0027 (0.373s): loss = 01.494124, likelihood = 0.2244\n",
      "\n",
      "\n",
      "Learning values [  0 ] Converged @ 19.\n",
      "Reasoning [  19 28 ] Done (10 iters).\n",
      "\n",
      ">>> Iter: 0028 (0.382s): loss = 01.496413, likelihood = 0.2239\n",
      "\n",
      "\n",
      "Learning values [  0 ] Converged @ 19.\n",
      "Reasoning [  19 28 ] Done (10 iters).\n",
      "\n",
      ">>> Iter: 0029 (0.383s): loss = 01.498159, likelihood = 0.2235\n",
      "\n",
      "\n",
      "Learning values [  0 ] Converged @ 18.\n",
      "Reasoning [  18 27 ] Done (10 iters).\n",
      "\n",
      ">>> Iter: 0030 (0.372s): loss = 01.499384, likelihood = 0.2233\n",
      "\n",
      "\n",
      "Learning values [  0 ] Converged @ 19.\n",
      "Reasoning [  19 28 ] Done (10 iters).\n",
      "\n",
      ">>> Iter: 0031 (0.386s): loss = 01.500116, likelihood = 0.2231\n",
      "\n",
      "\n",
      "Learning values [  0 ] Converged @ 19.\n",
      "Reasoning [  19 28 ] Done (10 iters).\n",
      "\n",
      ">>> Iter: 0032 (0.386s): loss = 01.500388, likelihood = 0.2230\n",
      "\n",
      "\n",
      "Learning values [  0 ] Converged @ 19.\n",
      "Reasoning [  19 28 ] Done (10 iters).\n",
      "\n",
      ">>> Iter: 0033 (0.385s): loss = 01.500235, likelihood = 0.2231\n",
      "\n",
      "\n",
      "Learning values [  0 ] Converged @ 19.\n",
      "Reasoning [  19 28 ] Done (10 iters).\n",
      "\n",
      ">>> Iter: 0034 (0.388s): loss = 01.499694, likelihood = 0.2232\n",
      "\n",
      "\n",
      "Learning values [  0 ] Converged @ 19.\n",
      "Reasoning [  19 28 ] Done (10 iters).\n",
      "\n",
      ">>> Iter: 0035 (0.385s): loss = 01.498801, likelihood = 0.2234\n",
      "\n",
      "\n",
      "Learning values [  0 ] Converged @ 19.\n",
      "Reasoning [  19 28 ] Done (10 iters).\n",
      "\n",
      ">>> Iter: 0036 (0.388s): loss = 01.497597, likelihood = 0.2237\n",
      "\n",
      "\n",
      "Learning values [  0 ] Converged @ 19.\n",
      "Reasoning [  19 28 ] Done (10 iters).\n",
      "\n",
      ">>> Iter: 0037 (0.382s): loss = 01.496118, likelihood = 0.2240\n",
      "\n",
      "\n",
      "Learning values [  0 ] Converged @ 20.\n",
      "Reasoning [  20 29 ] Done (10 iters).\n",
      "\n",
      ">>> Iter: 0038 (0.394s): loss = 01.494400, likelihood = 0.2244\n",
      "\n",
      "\n",
      "Learning values [  0 ] Converged @ 20.\n",
      "Reasoning [  20 29 ] Done (10 iters).\n",
      "\n",
      ">>> Iter: 0039 (0.393s): loss = 01.492480, likelihood = 0.2248\n",
      "\n",
      "\n",
      "Learning values [  0 ] Converged @ 20.\n",
      "Reasoning [  20 29 ] Done (10 iters).\n",
      "\n",
      ">>> Iter: 0040 (0.395s): loss = 01.490387, likelihood = 0.2253\n",
      "\n",
      "\n",
      "Learning values [  0 ] Converged @ 20.\n",
      "Reasoning [  20 29 ] Done (10 iters).\n",
      "\n",
      ">>> Iter: 0041 (0.392s): loss = 01.488155, likelihood = 0.2258\n",
      "\n",
      "\n",
      "Learning values [  0 ] Converged @ 20.\n",
      "Reasoning [  20 29 ] Done (10 iters).\n",
      "\n",
      ">>> Iter: 0042 (0.396s): loss = 01.485811, likelihood = 0.2263\n",
      "\n",
      "\n",
      "Learning values [  0 ] Converged @ 20.\n",
      "Reasoning [  20 29 ] Done (10 iters).\n",
      "\n",
      ">>> Iter: 0043 (0.399s): loss = 01.483382, likelihood = 0.2269\n",
      "\n",
      "\n",
      "Learning values [  0 ] Converged @ 19.\n",
      "Reasoning [  19 28 ] Done (10 iters).\n",
      "\n",
      ">>> Iter: 0044 (0.384s): loss = 01.480891, likelihood = 0.2274\n",
      "\n",
      "\n",
      "Learning values [  0 ] Converged @ 20.\n",
      "Reasoning [  20 29 ] Done (10 iters).\n",
      "\n",
      ">>> Iter: 0045 (0.396s): loss = 01.478360, likelihood = 0.2280\n",
      "\n",
      "\n",
      "Learning values [  0 ] Converged @ 20.\n",
      "Reasoning [  20 29 ] Done (10 iters).\n",
      "\n",
      ">>> Iter: 0046 (0.401s): loss = 01.475807, likelihood = 0.2286\n",
      "\n",
      "\n",
      "Learning values [  0 ] Converged @ 21.\n",
      "Reasoning [  21 30 ] Done (10 iters).\n",
      "\n",
      ">>> Iter: 0047 (0.405s): loss = 01.473248, likelihood = 0.2292\n",
      "\n",
      "\n",
      "Learning values [  0 ] Converged @ 21.\n",
      "Reasoning [  21 30 ] Done (10 iters).\n",
      "\n",
      ">>> Iter: 0048 (0.403s): loss = 01.470699, likelihood = 0.2298\n",
      "\n",
      "\n",
      "Learning values [  0 ] Converged @ 21.\n",
      "Reasoning [  21 30 ] Done (10 iters).\n",
      "\n",
      ">>> Iter: 0049 (0.407s): loss = 01.468171, likelihood = 0.2303\n",
      "\n",
      "\n",
      "Learning values [  0 ] Converged @ 21.\n",
      "Reasoning [  21 30 ] Done (10 iters).\n",
      "\n",
      ">>> Iter: 0050 (0.412s): loss = 01.465675, likelihood = 0.2309\n",
      "\n",
      "\n",
      "Learning values [  0 ] Converged @ 21.\n",
      "Reasoning [  21 30 ] Done (10 iters).\n",
      "\n",
      ">>> Iter: 0051 (0.406s): loss = 01.463222, likelihood = 0.2315\n",
      "\n",
      "\n",
      "Learning values [  0 ] Converged @ 21.\n",
      "Reasoning [  21 30 ] Done (10 iters).\n",
      "\n",
      ">>> Iter: 0052 (0.406s): loss = 01.460816, likelihood = 0.2320\n",
      "\n",
      "\n",
      "Learning values [  0 ] Converged @ 21.\n",
      "Reasoning [  21 30 ] Done (10 iters).\n",
      "\n",
      ">>> Iter: 0053 (0.405s): loss = 01.458464, likelihood = 0.2326\n",
      "\n",
      "\n",
      "Learning values [  0 ] Converged @ 21.\n",
      "Reasoning [  21 30 ] Done (10 iters).\n",
      "\n",
      ">>> Iter: 0054 (0.404s): loss = 01.456173, likelihood = 0.2331\n",
      "\n",
      "\n",
      "Learning values [  0 ] Converged @ 21.\n",
      "Reasoning [  21 30 ] Done (10 iters).\n",
      "\n",
      ">>> Iter: 0055 (0.407s): loss = 01.453944, likelihood = 0.2336\n",
      "\n",
      "\n",
      "Learning values [  0 ] Converged @ 21.\n",
      "Reasoning [  21 30 ] Done (10 iters).\n",
      "\n",
      ">>> Iter: 0056 (0.410s): loss = 01.451780, likelihood = 0.2342\n",
      "\n",
      "\n",
      "Learning values [  0 ] Converged @ 22.\n",
      "Reasoning [  22 31 ] Done (10 iters).\n",
      "\n",
      ">>> Iter: 0057 (0.431s): loss = 01.449683, likelihood = 0.2346\n",
      "\n",
      "\n",
      "Learning values [  0 ] Converged @ 22.\n",
      "Reasoning [  22 31 ] Done (10 iters).\n",
      "\n",
      ">>> Iter: 0058 (0.442s): loss = 01.447654, likelihood = 0.2351\n",
      "\n",
      "\n",
      "Learning values [  0 ] Converged @ 22.\n",
      "Reasoning [  22 31 ] Done (10 iters).\n",
      "\n",
      ">>> Iter: 0059 (0.442s): loss = 01.445695, likelihood = 0.2356\n",
      "\n",
      "\n",
      "Learning values [  0 ] Converged @ 22.\n",
      "Reasoning [  22 31 ] Done (10 iters).\n",
      "\n",
      ">>> Iter: 0060 (0.461s): loss = 01.443804, likelihood = 0.2360\n",
      "\n",
      "\n",
      "Learning values [  0 ] Converged @ 22.\n",
      "Reasoning [  22 31 ] Done (10 iters).\n",
      "\n",
      ">>> Iter: 0061 (0.439s): loss = 01.441980, likelihood = 0.2365\n",
      "\n",
      "\n",
      "Learning values [  0 ] Converged @ 22.\n",
      "Reasoning [  22 31 ] Done (10 iters).\n",
      "\n",
      ">>> Iter: 0062 (0.427s): loss = 01.440225, likelihood = 0.2369\n",
      "\n",
      "\n",
      "Learning values [  0 ] Converged @ 22.\n",
      "Reasoning [  22 31 ] Done (10 iters).\n",
      "\n",
      ">>> Iter: 0063 (0.430s): loss = 01.438535, likelihood = 0.2373\n",
      "\n",
      "\n",
      "Learning values [  0 ] Converged @ 21.\n",
      "Reasoning [  21 30 ] Done (10 iters).\n",
      "\n",
      ">>> Iter: 0064 (0.424s): loss = 01.436910, likelihood = 0.2377\n",
      "\n",
      "\n",
      "Learning values [  0 ] Converged @ 22.\n",
      "Reasoning [  22 31 ] Done (10 iters).\n",
      "\n",
      ">>> Iter: 0065 (0.423s): loss = 01.435348, likelihood = 0.2380\n",
      "\n",
      "\n",
      "Learning values [  0 ] Converged @ 22.\n",
      "Reasoning [  22 31 ] Done (10 iters).\n",
      "\n",
      ">>> Iter: 0066 (0.433s): loss = 01.433847, likelihood = 0.2384\n",
      "\n",
      "\n",
      "Learning values [  0 ] Converged @ 22.\n",
      "Reasoning [  22 31 ] Done (10 iters).\n",
      "\n",
      ">>> Iter: 0067 (0.432s): loss = 01.432406, likelihood = 0.2387\n",
      "\n",
      "\n",
      "Learning values [  0 ] Converged @ 22.\n",
      "Reasoning [  22 31 ] Done (10 iters).\n",
      "\n",
      ">>> Iter: 0068 (0.430s): loss = 01.431022, likelihood = 0.2391\n",
      "\n",
      "\n",
      "Learning values [  0 ] Converged @ 22.\n",
      "Reasoning [  22 31 ] Done (10 iters).\n",
      "\n",
      ">>> Iter: 0069 (0.421s): loss = 01.429693, likelihood = 0.2394\n",
      "\n",
      "\n",
      "Learning values [  0 ] Converged @ 22.\n",
      "Reasoning [  22 31 ] Done (10 iters).\n",
      "\n",
      ">>> Iter: 0070 (0.429s): loss = 01.428418, likelihood = 0.2397\n",
      "\n",
      "\n",
      "Learning values [  0 ] Converged @ 23.\n",
      "Reasoning [  23 32 ] Done (10 iters).\n",
      "\n",
      ">>> Iter: 0071 (0.442s): loss = 01.427194, likelihood = 0.2400\n",
      "\n",
      "\n",
      "Learning values [  0 ] Converged @ 23.\n",
      "Reasoning [  23 32 ] Done (10 iters).\n",
      "\n",
      ">>> Iter: 0072 (0.437s): loss = 01.426019, likelihood = 0.2403\n",
      "\n",
      "\n",
      "Learning values [  0 ] Converged @ 23.\n",
      "Reasoning [  23 32 ] Done (10 iters).\n",
      "\n",
      ">>> Iter: 0073 (0.439s): loss = 01.424892, likelihood = 0.2405\n",
      "\n",
      "\n",
      "Learning values [  0 ] Converged @ 23.\n",
      "Reasoning [  23 32 ] Done (10 iters).\n",
      "\n",
      ">>> Iter: 0074 (0.437s): loss = 01.423811, likelihood = 0.2408\n",
      "\n",
      "\n",
      "Learning values [  0 ] Converged @ 22.\n",
      "Reasoning [  22 31 ] Done (10 iters).\n",
      "\n",
      ">>> Iter: 0075 (0.436s): loss = 01.422772, likelihood = 0.2410\n",
      "\n",
      "\n",
      "Learning values [  0 ] Converged @ 22.\n",
      "Reasoning [  22 31 ] Done (10 iters).\n",
      "\n",
      ">>> Iter: 0076 (0.439s): loss = 01.421775, likelihood = 0.2413\n",
      "\n",
      "\n",
      "Learning values [  0 ] Converged @ 22.\n",
      "Reasoning [  22 31 ] Done (10 iters).\n",
      "\n",
      ">>> Iter: 0077 (0.429s): loss = 01.420817, likelihood = 0.2415\n",
      "\n",
      "\n",
      "Learning values [  0 ] Converged @ 23.\n",
      "Reasoning [  23 32 ] Done (10 iters).\n",
      "\n",
      ">>> Iter: 0078 (0.439s): loss = 01.419899, likelihood = 0.2417\n",
      "\n",
      "\n",
      "Learning values [  0 ] Converged @ 23.\n",
      "Reasoning [  23 32 ] Done (10 iters).\n",
      "\n",
      ">>> Iter: 0079 (0.442s): loss = 01.419016, likelihood = 0.2420\n",
      "\n",
      "\n",
      "Learning values [  0 ] Converged @ 23.\n",
      "Reasoning [  23 32 ] Done (10 iters).\n",
      "\n",
      ">>> Iter: 0080 (0.446s): loss = 01.418167, likelihood = 0.2422\n",
      "\n",
      "\n",
      "Learning values [  0 ] Converged @ 23.\n",
      "Reasoning [  23 32 ] Done (10 iters).\n",
      "\n",
      ">>> Iter: 0081 (0.437s): loss = 01.417352, likelihood = 0.2424\n",
      "\n",
      "\n",
      "Learning values [  0 ] Converged @ 23.\n",
      "Reasoning [  23 32 ] Done (10 iters).\n",
      "\n",
      ">>> Iter: 0082 (0.433s): loss = 01.416569, likelihood = 0.2425\n",
      "\n",
      "\n",
      "Learning values [  0 ] Converged @ 23.\n",
      "Reasoning [  23 32 ] Done (10 iters).\n",
      "\n",
      ">>> Iter: 0083 (0.455s): loss = 01.415816, likelihood = 0.2427\n",
      "\n",
      "\n",
      "Learning values [  0 ] Converged @ 23.\n",
      "Reasoning [  23 32 ] Done (10 iters).\n",
      "\n",
      ">>> Iter: 0084 (0.446s): loss = 01.415091, likelihood = 0.2429\n",
      "\n",
      "\n",
      "Learning values [  0 ] Converged @ 23.\n",
      "Reasoning [  23 32 ] Done (10 iters).\n",
      "\n",
      ">>> Iter: 0085 (0.439s): loss = 01.414394, likelihood = 0.2431\n",
      "\n",
      "\n",
      "Learning values [  0 ] Converged @ 23.\n",
      "Reasoning [  23 32 ] Done (10 iters).\n",
      "\n",
      ">>> Iter: 0086 (0.439s): loss = 01.413723, likelihood = 0.2432\n",
      "\n",
      "\n",
      "Learning values [  0 ] Converged @ 23.\n",
      "Reasoning [  23 32 ] Done (10 iters).\n",
      "\n",
      ">>> Iter: 0087 (0.452s): loss = 01.413077, likelihood = 0.2434\n",
      "\n",
      "\n",
      "Learning values [  0 ] Converged @ 23.\n",
      "Reasoning [  23 32 ] Done (10 iters).\n",
      "\n",
      ">>> Iter: 0088 (0.458s): loss = 01.412455, likelihood = 0.2435\n",
      "\n",
      "\n",
      "Learning values [  0 ] Converged @ 23.\n",
      "Reasoning [  23 32 ] Done (10 iters).\n",
      "\n",
      ">>> Iter: 0089 (0.436s): loss = 01.411856, likelihood = 0.2437\n",
      "\n",
      "\n",
      "Learning values [  0 ] Converged @ 23.\n",
      "Reasoning [  23 32 ] Done (10 iters).\n",
      "\n",
      ">>> Iter: 0090 (0.446s): loss = 01.411278, likelihood = 0.2438\n",
      "\n",
      "\n",
      "Learning values [  0 ] Converged @ 23.\n",
      "Reasoning [  23 32 ] Done (10 iters).\n",
      "\n",
      ">>> Iter: 0091 (0.444s): loss = 01.410721, likelihood = 0.2440\n",
      "\n",
      "\n",
      "Learning values [  0 ] Converged @ 24.\n",
      "Reasoning [  24 33 ] Done (10 iters).\n",
      "\n",
      ">>> Iter: 0092 (0.447s): loss = 01.410184, likelihood = 0.2441\n",
      "\n",
      "\n",
      "Learning values [  0 ] Converged @ 24.\n",
      "Reasoning [  24 33 ] Done (10 iters).\n",
      "\n",
      ">>> Iter: 0093 (0.458s): loss = 01.409666, likelihood = 0.2442\n",
      "\n",
      "\n",
      "Learning values [  0 ] Converged @ 24.\n",
      "Reasoning [  24 33 ] Done (10 iters).\n",
      "\n",
      ">>> Iter: 0094 (0.449s): loss = 01.409166, likelihood = 0.2443\n",
      "\n",
      "\n",
      "Learning values [  0 ] Converged @ 24.\n",
      "Reasoning [  24 33 ] Done (10 iters).\n",
      "\n",
      ">>> Iter: 0095 (0.458s): loss = 01.408682, likelihood = 0.2445\n",
      "\n",
      "\n",
      "Learning values [  0 ] Converged @ 24.\n",
      "Reasoning [  24 33 ] Done (10 iters).\n",
      "\n",
      ">>> Iter: 0096 (0.443s): loss = 01.408216, likelihood = 0.2446\n",
      "\n",
      "\n",
      "Learning values [  0 ] Converged @ 24.\n",
      "Reasoning [  24 33 ] Done (10 iters).\n",
      "\n",
      ">>> Iter: 0097 (0.452s): loss = 01.407765, likelihood = 0.2447\n",
      "\n",
      "\n",
      "Learning values [  0 ] Converged @ 24.\n",
      "Reasoning [  24 33 ] Done (10 iters).\n",
      "\n",
      ">>> Iter: 0098 (0.449s): loss = 01.407329, likelihood = 0.2448\n",
      "\n",
      "\n",
      "Learning values [  0 ] Converged @ 24.\n",
      "Reasoning [  24 33 ] Done (10 iters).\n",
      "\n",
      ">>> Iter: 0099 (0.453s): loss = 01.406908, likelihood = 0.2449\n",
      "\n",
      "\n"
     ]
    }
   ],
   "source": [
    "# # T.ACTIONS: ['U', 'D', 'L', 'R']\n",
    "# Sk = S.at_loc((2,2))\n",
    "# Ak = \"U\"\n",
    "boltzmann_temp = 0.05\n",
    "goal = S.at_loc((2,6))\n",
    "log_likelihoods_history = []\n",
    "s_lst_lst = [[(2,0), (2,1), (1,1), (1,2), (1,3), (1,4), (1,5), (2,5), (2,6)],\n",
    "            [(2,0), (2,1), (3,1), (3,2), (3,3), (3,4), (3,5), (2,5), (2,6)]]\n",
    "a_lst_lst = [T.loc_lst_to_a_lst(s_lst) for s_lst in s_lst_lst]\n",
    "\n",
    "for _iter in range(100):\n",
    "    _iter_start_time = time.time()\n",
    "    r_model_linear.zero_grad()\n",
    "    \n",
    "    VI = Plan.ValueIteration(S, r_model_linear(PHI), T, verbose=True, log_pi=False, gamma=0.95, goal=goal)\n",
    "    Pi, V, Q, iterno, v_delta_max, _ = VI.run(100, lambda q: Policy.Boltzmann(q, boltzmann_temp=boltzmann_temp), \n",
    "                                           verbose=True, debug=False, eps=1e-4, ret_vals=True)\n",
    "    values = V.detach().numpy().round(2)\n",
    "    loss = 0\n",
    "    for k in range(len(s_lst_lst)):\n",
    "        for i in range(len(s_lst_lst[k])-1):\n",
    "            loss += -torch.log(VI.Pi[VI.get_tbl_idxs(S.at_loc(s_lst_lst[k][i]), a_lst_lst[k][i])])\n",
    "#     for i in range(6):\n",
    "#         loss += -torch.log(Pi[VI.get_tbl_idxs(S.at_loc((2,0+i)), \"R\")])\n",
    "#         print(Pi[VI.get_tbl_idxs(S.at_loc((2,1+i)), \"R\")], loss)\n",
    "\n",
    "    ll = np.exp(-loss.detach().item())\n",
    "    log_likelihoods_history.append(ll)\n",
    "    \n",
    "    print(\"\\n>>> Iter: {:04d} ({:03.3f}s): loss = {:09.6f}, likelihood = {:02.4f}\\n\\n\".format(\n",
    "                    _iter, time.time()-_iter_start_time, loss, ll))\n",
    "    loss.backward()\n",
    "    r_model_linear.step()\n",
    "    # print(torch.sum(r_model_linear.fc1.weight.grad))\n",
    "#     print(torch.sum(r_model_linear.w.weight.grad))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1728x1728 with 9 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "plot_irl_results(S, [], VI.V.detach().numpy().round(2),\n",
    "                 log_likelihoods_history, r_key=\"linear\", phi_key=\"indicator\")\n",
    "plt.savefig(\"./figures/mlirl_gw_5x7_one_hot__traj_2_turns\")"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python [conda env:irl]",
   "language": "python",
   "name": "conda-env-irl-py"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.7.4"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}