Baseline.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {
    "collapsed": true,
    "ExecuteTime": {
     "end_time": "2023-05-04T00:26:11.818306704Z",
     "start_time": "2023-05-04T00:26:10.724608836Z"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[Dataset] Generating Dataset 2d\n"
     ]
    }
   ],
   "source": [
    "import numpy as np\n",
    "import scipy\n",
    "import matplotlib.pyplot as plt\n",
    "import pickle\n",
    "from utils.dynamics import import_dynamics\n",
    "from utils.dataset_generator import generate_dataset\n",
    "\n",
    "generate_dataset(system_name='2d', baseline=True)"
   ]
  },
  {
   "cell_type": "markdown",
   "source": [
    "# Plotting and Comparison Trajectories"
   ],
   "metadata": {
    "collapsed": false
   }
  },
  {
   "cell_type": "code",
   "execution_count": 82,
   "outputs": [],
   "source": [
    "# baseline['rollout_len']\n",
    "# baseline['rollout_len'], baseline['i_rollout'], baseline['dist_to_target'], baseline['x0']\n",
    "f = open('datasets/dataset_2d_baseline_stats.pkl', 'rb')\n",
    "b_stats = pickle.load(f)\n",
    "f.close()\n",
    "\n",
    "f = open('datasets/dataset_2d_baseline_data.pkl', 'rb')\n",
    "b_data = pickle.load(f)\n",
    "f.close()\n",
    "\n",
    "f = open(f'inference/2d_arch256_e100_d100_edim256_ks5_par3_70e06_date04_26_02_22_20/data.pkl', 'rb')\n",
    "data = pickle.load(f) # REWARD, TRAJECTORIES\n",
    "f.close()\n",
    "\n",
    "f = open(f'inference/2d_arch256_e100_d100_edim256_ks5_par3_70e06_date04_26_02_22_20/trajectories.pkl', 'rb')\n",
    "traj = pickle.load(f) # REWARD, TRAJECTORIES\n",
    "f.close()\n",
    "\n",
    "finish = b_data['finish_token'].astype(np.int8)\n",
    "episode_ends = np.array([idx for idx, a in enumerate(finish) if a == 1])\n",
    "episode_ends = np.concatenate(([0], episode_ends))\n"
   ],
   "metadata": {
    "collapsed": false,
    "ExecuteTime": {
     "end_time": "2023-05-04T04:46:16.432714634Z",
     "start_time": "2023-05-04T04:46:16.220962814Z"
    }
   }
  },
  {
   "cell_type": "code",
   "execution_count": 99,
   "outputs": [
    {
     "data": {
      "text/plain": "<Figure size 2000x400 with 5 Axes>",
      "image/png": "iVBORw0KGgoAAAANSUhEUgAABlMAAAFzCAYAAACq3P/BAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjcuMSwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy/bCgiHAAAACXBIWXMAAA9hAAAPYQGoP6dpAACvGElEQVR4nOzdd3hU1drG4d+ekt5Ig0ASQu+9gxRFEBEVOFZs2HvvDXvv5dgV9CB2UEFUEJUiSO+9hZCQEFJIL5OZ+f6IRvnopOzMzHNfFxdksmfv500mi8m8s9Yy3G63GxERERERERERERERETkki9kBRERERERERERERERE6jM1U0RERERERERERERERI5AzRQREREREREREREREZEjUDNFRERERERERERERETkCNRMEREREREREREREREROQI1U0RERERERERERERERI5AzRQREREREREREREREZEjUDNFRERERERERERERETkCGxmB6hLLpeLPXv2EBoaimEYZscRkXrK7XZTUFBA48aNsVi8o+es8U9EjpXGQBHxVd44/oHGQBE5NhoDRcSXHesY6FPNlD179pCQkGB2DBHxELt37yY+Pt7sGDVC45+IHC+NgSLiq7xp/AONgSJyfDQGiogvO9oY6FPNlNDQUKDyixIWFlYr13A4HMyaNYvhw4djt9tr5Rr1ga/UCb5Tq6/UCUevNT8/n4SEhKoxwxvUxfgHvvM48pU6wXdq9ZU6QWOgngNWn6/U6it1gu/U6ovjH2gMrEm+UieoVm+kMVBjYHX5Sp3gO7X6Sp1Qc2OgTzVT/p7OFxYWVqsDaFBQEGFhYV79IPSVOsF3avWVOuHYa/WmKcB1Mf6B7zyOfKVO8J1afaVO0Bio54DV5yu1+kqd4Du1+uL4BxoDa5Kv1Amq1RtpDNQYWF2+Uif4Tq2+UifU3BjoPYsgioiIiIiIiIiIiIiI1AI1U0RERERERERERERERI5AzRQREREREREREREREZEjUDNFRERERERERERERETkCNRMEREREREREREREREROQI1U0RERERERERERERERI5AzRQREREREREREREREZEjUDNFRERERERERERERETkCDymmfLMM8/Qq1cvQkNDiY2NZfTo0WzevNnsWCIiIiIiIiIiIiIi4uU8ppkyd+5cbrzxRv78809mz56Nw+Fg+PDhFBUVmR1NRERERKTWuFxudmUX8ce2LGatz2Dh9iw2ZxSQV+IwO5qIiIiIiIjPsJkd4Fj99NNPB3w8adIkYmNjWb58OYMGDTIplYiIiIhI7UjOKuL9+Tv4cV0GOUXlhzwmMTKIzvHh9GkWySntGtIkIrCOU4qIiIiIiPgGj2mm/H95eXkAREZGHvaYsrIyysrKqj7Oz88HwOFw4HDUzjv5/j5vbZ2/vvCVOsF3avWVOuHotXrD18CM8e/v8//7b2/lK3WC79TqK3WCxkBPeA7odrv54I9kXvllGw6nGwB/m4X4BoGE+NsoKK0gt7ic3GIHKTnFpOQUM2NNOg9/t562DUMY1TmOsd0aExPqX+0sh+IrPy++Uif4Tq2+MP6B54+B9Zmv1Amq1RtpDNQYWF2+Uif4Tq2+UifU3BhouN1ud42lqiMul4uzzjqL/fv3s2DBgsMe9+ijj/LYY48ddPuUKVMICgqqzYgi4sGKi4sZN24ceXl5hIWFmR3nhGj8E5ETpTHQfN8mW/gtvXI13rbhLk5p7KZFmBvb/1ugt7gCdhca7CqETfst7CgANwYAFtx0aOBmSGMXLULBMOq6ChHP4w3jH3j+GCgi5tAYKCK+7FjHQI9splx//fX8+OOPLFiwgPj4+MMed6hudEJCAllZWbX2H4PD4WD27NkMGzYMu91eK9eoD3ylTvCdWn2lTjh6rfn5+URHR3v0k0gzxj/wnceRr9QJvlOrr9QJGgPr+3PAH9ZmcNuXawCYcEZbLu6TgHGMnZDc4nJ+2biPr5ansnJ3XtXt3RMjuG5QM4a0jj7mcx2Jr/y8+Eqd4Du1+sL4B549BtZ3vlInqFZvpDFQY2B1+Uqd4Du1+kqdUHNjoMct83XTTTcxY8YM5s2bd8RGCoC/vz/+/gcvb2C322v9AVIX16gPfKVO8J1afaVOOHyt3lC/meNfXV7HbL5SJ/hOrb5SJ2gMrC3VuUapw8kzP20G4KaTW3LFwBbHdf/YcDvj+gYzrm8SW/YWMGlhMl8vS2VFyn6umbySnk0b8MAZ7eie2OCE8v1/vvLz4it1gu/U6s3jH3juGOhJfKVOUK3eSGNg7fH1x5A38pVafaVOqP4YaDn6IfWD2+3mpptuYtq0afz66680a9bM7EgiIiIiIjVmxpp09uaX0Tg8gJtOaVmtc7VuGMrTYzqx4N6TuWZQcwLsFpbtymXsWwu58dMV7MouqqHUIiIiIiIivsFjmik33ngjkydPZsqUKYSGhpKRkUFGRgYlJSVmRxMRERERqbbvVqUBcFHfpgTYrTVyztiwAB4Y2Y7f7zqZ83rGYxjww9p0Tn15Ls/M3EhxeUWNXEdERERERMTbeUwz5e233yYvL48hQ4YQFxdX9eeLL74wO5qIiIiISLWUOpws2ZkDwGkdGtb4+RuFB/D8OV2YectABrWOweF08+68HQx/ZR6/bc6s8euJiIiIiIh4G4/ZM8XtdpsdQURERESkVmzLLKSswkVEkJ0WMSG1dp12cWF8ckVv5mzcy4Tv1pOaW8LlE5cyqnMcE85sT2xoQK1dW0RERERExJN5zMwUERERERFvtTOrcg+TFjEhGIZR69cb2q4hs24fxJUnNcNiVO7XMuzlecxYs6fWry0iIiIiIuKJ1EwRERERETFZdmEZAI3C6m5mSLC/jYdHtef7m06iY5Mw8koc3DRlJbd/sYr8Uked5RAREREREfEEaqaIiIiIiJissKxyI/jQgLpfhbdjk3Cm3TCAm09picWAaSvTOP3V+SzekV3nWUREREREROorNVNEREREREzmcFbuD2iz1v4SX4dit1q4c3gbvrquH4mRQaTtL+GC9//k1V+24HRp70IRERERERE1U0RERERETGb/q4lS4TS3cdGjaSQzbx3IuT3icbvh1V+2ctlHS8j6axkyERERERERX6VmioiIiIiIyfxtVgBKHE6Tk0CIv40Xzu3Cy+d1IdBuZcG2LM54fT5LduaYHU1ERERERMQ0aqaIiIiIiJgsNswfgL35pSYn+cfY7vF8f9MAWsaGsDe/jAvf/5MP5u/A7dayXyIiIiIi4nvUTBERERERMVmjsAAAMvLqTzMFoFXDUL6/aQBjujXB6XLz5A8buXfaehwus5OJiIiIiIjULTVTRERERERMFhceCEBGfmm9m/kR5Gfj5fO68MiZ7bFaDKat3MMb6631ahaNiIiIiIhIbVMzRURERETEZH8v81XqcJFX4jA5zcEMw+DyAc345IrehAfa2FVoMPadxazavd/saCIiIiIiInVCzRQREREREZMF2K1EBfsBsGd//Z3xMaBlNN9c15dGgW4yC8q44L1FzFqfYXYsERERERGRWqdmioiIiIhIPRAfGQRAcnaRyUmOrGlkELd3cjK4dTSlDhfXTl7OxwuTzY4lIiIiIiJSq9RMERERERGpB1rHhgCwZW+ByUmOLsAK74zryoW9E3G74ZHv1/PUDxtwuerXfi8iIiIiIiI1Rc0UEREREZF6oHXDUMAzmikANquFp8d05J4RbQB4f/5Obv5sJaUOp8nJREREREREap6aKSIiIiIi9UC7uDAA1qTmmZzk2BmGwQ1DWvLaBV2xWw1+WJvOxR8sJq/YYXY0ERERERGRGqVmioiIiIhIPdAlIRzDgNTcEjLz6+8m9IdydtcmfHJFH8ICbCzblcv57y0is8CzahARERERETkSNVNEREREROqB0AA7bf5a6mtFSq7JaY5fvxZRfHFtP6JD/NmUUcC57yxid06x2bFERERERERqhJopIiIiIiL1RLfEBgCsSNlvbpAT1C4ujG+u70dCZCC7sos5552FHrMHjIiIiIiIyJGomSIiIiIiUk/0aPpXM2WX581M+VvTqGC+vq4/rRuGsDe/jPPeXcRKD5xpIyIiIiIi8m9qpoiIiIiI1BPdEyMAWJOWR3mFy9ww1dAwLIAvr+1H14QI9hc7uOiDxSzYmmV2LBERERERkROmZoqIiIiISD3RLDqYBkF2yitcrNuTZ3acaokI8uPTq/pwUstoisudXDFpKbM37DU7loiIiIiIyAlRM0VEREREpJ4wDIPezSIBvGImR7C/jQ/H9+T0jo0od7q4fvJyflybbnYsERERERGR46ZmioiIiIhIPTKkTSwAv2/ONDlJzfC3WXnjwm6c3bUxFS43N322kumr95gdS0RERERE5LjYzA4gIiIiIuKrih3FfLvtW7JLs0kMTaRvXF+GtIkBYOXu/eQWldMg2M/klNVns1p4+byuWC0GU1ekcevnK6lwuRjTLd7saCIiIiIiIsdEzRQREREREZPcN/8+ftv9W9XHBgYDmgwgMb4dKanNmLd1H2d3bWJiwppjtRi8cE4X7BYLXyzbzR1frqbC6ebcnglmRxMRERERETkqLfMlIiIiImKSRXsWATCs6TA6R3fGjZsFaQvIDX2f4OYv89mGqVS4KkxOWXOsFoNnxnbioj6JuN1w99dr+GxJitmxREREREREjkrNFBERERERkwTYAgC4tP2lfHrGp8wYM4MrOl5BsC0Mi38W6x3vMWraKL7e8jUOl8PktDXDYjF4cnRHxvdPAuD+qWv536JkUzOJiIiIiIgcjZopIiIiIiIm6dWoFwDzUucB0DSsKbf3uJ2f/vMT7uwzcFUEk1aYxmOLHmPMd2OYlTwLt9ttZuQaYRgGj5zZnqsHNgPg4e/W8+GCnSanEhEREREROTw1U0RERERETDI0cSgA323/7oCZJxEBoZwUcy5F2+6lT9jlRAZEsit/F3fOvZOLZl7E0oylZkWuMYZh8MDIdlw/pAUAT8zYoIaKiIiIiIjUWx7VTJk3bx5nnnkmjRs3xjAMvv32W7MjiYiIiIicsGFNhxEVEEVmcSa/7PrlgM8NbRcLbj9Sknswc+xMru9yPYG2QNZmreWKn6/g+l+uZ3POZpOS1wzDMLjntDbcckpLoLKhMvEPNVRERERERKT+8ahmSlFREV26dOG///2v2VFERERERKrNz+rH+W3OB+CDtR/gcruqPje8QyP8rBa27C0kNdvFDV1vYObYmVzQ5gJsho0FaQs4d/q5PDD/AfYU7jGrhGozDIPbh7XmxpMrZ6g8Nn0DHy9MNjeUiIiIiIjI/+NRzZTTTz+dJ598kjFjxpgdRURERESkRoxrN44QewhbcrcwK3lW1e3hgXYGt4kBYPrqymZJdGA0D/Z9kO9Gf8eIpBG4cTN9x3RGTRvFS8teIq8sz5QaqsswDO4a3qZqya9Hvl+vTelFRERERKResZkdoDaVlZVRVlZW9XF+fj4ADocDh8NxuLtVy9/nra3z1xe+Uif4Tq2+UiccvVZv+BqYMf79ff5//+2tfKVO8J1afaVO0BhYH58DBlmCuKTtJby99m3eWPkGgxsPxmapfJp+eodYZm/Yy/er07jl5GYYhgFAXGAcT/d/movbXMyrq15l2d5lTFo/ialbp3JVh6s4r/V5+Fn9aq7Af6nNn5fbT2mOo8LJBwuSefi79bjdLi7slVDj1zkWGhe8jy+Mf+B5Y6An8ZU6QbV6I42BGgOry1fqBN+p1VfqhJobAw232+2usVR1yDAMpk2bxujRow97zKOPPspjjz120O1TpkwhKCioFtOJiCcrLi5m3Lhx5OXlERYWZnacE6LxT0ROlMZAc5S5y3gp/yWK3cWMDhxNT/+elbc74cFlVhwug7s6VZAQcvB93W43Wyq28HPJz2S6MgGIsEQwLGAYneydsBgeNRkdtxu+22Xht/TK3Oc3d9K/oUf+yiIexhvGP/DMMVBEzKcxUER82bGOgV7dTDlUNzohIYGsrKxa+4/B4XAwe/Zshg0bht1ur5Vr1Ae+Uif4Tq2+Uiccvdb8/Hyio6M9+kmkGeMf+M7jyFfqBN+p1VfqBI2B9fk54ORNk3l5xcs0CmrEt2d+WzWz5NYvVjNz3V6uHNCU+0a0Oez9nS4n03dO5+01b7OvZB8AbRu05bZut9G7Ue8TynQodfHz4na7efrHzUxalALA06Pbc26P+Fq51uFoXPA+vjD+geeOgZ7AV+oE1eqNNAZqDKwuX6kTfKdWX6kTam4M9Oplvvz9/fH39z/odrvdXusPkLq4Rn3gK3WC79TqK3XC4Wv1hvrNHP/q8jpm85U6wXdq9ZU6QWNgbanONca1H8fkTZPJKM7gq21fMb7jeADO6tqEmev28uO6vTx4RgcsFuPQ18bOuW3PZVTLUUzeMJkP133IptxNXPfrdZzU5CRu73E7rRu0PtHSDr5eLX89HzmrIxgWJi1M5sHvNhAc4MfZXZvU2vUOR+OC9/Hm8Q88dwz0JL5SJ6hWb6QxsPb4+mPIG/lKrb5SJ1R/DPSsOf8iIiIiIl7K3+rPTV1vAuDdNe+SXZINwJA2sYT429iTV8qKlNyjnifQFsjVna9m5tiZjGs7DpthY0HaAs75/hwe/uNhMooyarWOmmIYBo+c2Z6L+iTidsOdX67m1017zY4lIiIiIiI+yqOaKYWFhaxatYpVq1YBsHPnTlatWkVKSoq5wUREREREasDZLc+mXWQ7Ch2F/HfVfwEIsFsZ3r4hANNX7znmc0UGRHJ/n/v5bvR3DG86HDduvt32LWdOO5PXV7xOQXlBrdRQkwzD4ImzO3JWl8ZUuNxcP3kFi3dkmx1LRERERER8kEc1U5YtW0a3bt3o1q0bAHfccQfdunVjwoQJJicTEREREak+i2Hh3t73AvDN1m/YnLMZgDO7Ngbg+9V7KKtwHtc5E8MSeWnIS0weOZnusd0pdZby/tr3OWPqGXy68VMcTkfNFlHDLBaDl87rwiltYymrcHHVx8tYl5ZndiwREREREfExHtVMGTJkCG63+6A/kyZNMjuaiIiIiEiN6NGwB8ObDsfldvH80udxu90MbBlNwzB/cosdzNmYeULn7RLThUkjJvHaya+RFJZEblkuzy55lrO/O5ufk3/G7XbXcCU1x2618NZF3endLJKCsgou/WgJO7OKzI4lIiIiIiI+xKs3oPd2DqeLjLxS9uwvIbuonJyicnKLyskpLqe4zElphZNSh5NShws3YDHAahgYhoHVAlaLQZCfjbAAO2GBNkID7IQF2IgI8iMuPIC48AAig/0wjENvcioiIiIiteOOnnfw++7fWZKxhF93/8rQxKH8p3s8b/2+na+W7WZkp7gTOq9hGJySeAqD4gcxdetU3lr1FrsLdnPX3LvoFN2JO3veSY+GPWq2mBoSYLfy4WU9Gff+Ytam5XHlpKVMu2EA4UG+sVmmiIiIiIiYS80UD7CvoIwtewuq/mzLLCQ1t4S9+aW4avkNhP42CwmRQbSMCaFlbAitGobQIiaExAj/2r2wiIiIiA9rEtKEyzpcxvtr3+elZS8xsMlAzulR2UyZu2UfGXmlNAoPOOHz2yw2zmtzHqOaj+Lj9R8zcf1E1matZfxP4xmSMITbu99O84jmNVhRzQgNsPPh+J6MfvMPdmQVceOUFUy8vBd2q0dNuBcREREREQ+kZko943a72bK3kKXJOSxLzmHZrlxSc0sOe7yf1UJcRACxof40CPIjMtiPiCA/QgNs+NssBNitBNitWAxwuty43eB0u3G53ThdbgrLKsgvqaCg1EF+aQX5JQ5yi8vZs7+UrMIyyipcbMssZFtmIaz/57qGAbEBVuYUraVzQgQdGofTvnEY4YF6Z6CIiIhITbiq01V8u+1bdhfs5tONn3J5x8vpldSApcm5fLMilRtPblntawTZg7i+6/Wc2+Zc3l71Nt9s/Ybfd//OvNR5jG01lhu63EBMUEz1i6lBsaEBvH9ZT859ZxELtmXx+PQNPDG6o9mxRERERETEy6mZUg84nC6W7Mxh1voMZm3YS3pe6QGftxjQNCqYVrEhtG4YSquGITSNCqZxRADRwf5YLLWzDFdZhZO9eWUkZxex9a+GyvbMQrZmFpBb7GBvicH3a9L5fk161X1axYbQq1kkvZMi6dUskiYRgbWSTURERMTbBdmDuLX7rTz0x0O8u+ZdzmxxJuf2TGBpci5fLN3N9YNb1NjzwOjAaB7u9zAXtb+I15a/xq+7f+XrLV/zw44fuKzDZYzvMJ5ge3CNXKsmdGgczqvnd+Xaycv535+7aN84jAt7J5odS0REREREvJiaKSZal5bHZ0tSmLEmnbwSR9XtgXYr3ZtG0LNpJD2TGtAtsQEh/nX/rfK3WUmMCiIxKohBrQ98R+KenEI+/v5Xgpq0YUNGAevS8knbX8LWzEK2ZhYyZXEKAE0iAumV1IC+zaMY1DqGxmquiIiIiByzM1ucyWebPmN99nreXPkm9/R8iCdmbCAlp5gF27IOeo5WXc3Dm/PaKa+xYu8KXlr+Emv2reGd1e/w5eYvubrT1Zzb5lws1I8ltYZ3aMRdw9vwws+beeS79bRtFEq3xAZmxxIRERERES+lZkodKy6vYOqKND5fmsK6tPyq26OC/Ti1XUOGd2jIgJbRBNitJqY8uphQf9o3cDNySHPs9sqlvbIKy1iWnFu1RNm6PZUNlrRVJXy7ag8ALWKCGdQ6hkGtY+jbLIpAv/pdp4iIiIiZLIaFe3vfy6U/XsrUrVO5oO0F/Kd7PJMWJvPp4l013kz5W/eG3Zl8+mR+SfmFV5e/SkpBCs8tfY6PN3zM1R2uxuauH79G3DCkBWtT8/hpfQbXT17B9JtPIiZUe/uJiIiIiEjNqx+/BfmAglIHnyzaxYcLdpJTVA5U7ndyWsdGXNArgb7No7DW0nJddSU6xJ8RHRsxomMjAIrKKli1ez+Ld+awYOs+Vu3ez/Z9RWzfV8TEP5Lxs1ro1awBg1pVNlfaNgrFMDz7ayAiIiJS07rFdmNE0gh+Sv6J55Y8x329X2PSwmR+2ZhZ7Y3oj8QwDIY1HcaQhCF8t+073l79NhlFGTyx5AmiLFH4JfsxsuVILIZ5M1UMw+DF87qw7b+VS9Le+OkKPr26jzakFxERERGRGqdmSi0rLKvg/Xk7mPjHTvJLKwBIjAzi0n5NGds9nshgP5MT1p5gfxsDWkYzoGU0dwxrTV6Jg4Xbspi3dR/ztmSRtr+EP7Zl88e2bJ75cRPxDQIZ3r4Rwzs0pFdSpMc3l0RERERqyh097uC33b+xbO8ykksX0TspkiXJOXyxdDe3ntqqVq9tt9g5p/U5nNniTL7c/CXvrXmP7LJs7l94PxM3TuTmbjczOH6waW+KCfG38e4lPRj95h8sSc7hqR828uhZHUzJIiIiIiIi3kvNlFridrv5dmUaT8/cSGZBGVC5xNVNp7TkzM6Nsfngu+XCA+2c3imO0zvF4Xa72b6viPlb9zFvyz4W7cgmNbeEj/7YyUd/7CQy2I+hbWM5rUMjBraOxt+m5cBERETEd8WFxHFFxyt4e/XbvLTsJa7p/Q5LknP4bEkKN5zcok5mYvhb/bmk/SWcmXQmj05/lMXOxWzJ3cLNv95M5+jO3NL9FvrE9an1HIfSIiaEl8/vytWfLGPSwmS6JIQzplu8KVlERERERMQ7qZlSC/LK4dpPV/Lb5iwAmkYFcfdpbTi9Y5xmW/zFMAxaxobQMjaEywc0o6Tcybyt+5i1fi9zNu0lp6icr5an8tXyVCKC7JzZuTH/6RFPl/hwLQUmIiIiPunyjpczbds00ovS2Wv8THRISzLyS/l5fQajOjeusxzB9mBODjiZh4Y+xOTNk/l046esyVrDVbOuok+jPtzc/Wa6xHSpszx/G9a+Ibec0pLXf93Gfd+spVVsKB2bhNd5DhERERER8U6+Nz2ili3Yls1zq638tjkLP6uFO4e15ufbBjGqc2M1Uo4g0M/KaR0a8dJ5XVj24Kl8dnVfxvdPomGYP/uLHfzvz12M/u8fnPryXN6bt528YofZkUVERETqVKAtkDt73AnApPUfcVbP4Mp//5FsSp5w/3Bu63EbP/7nR8a1HYfNYmNxxmIunnkxN8+5mQ3ZG+o8022ntubkNjGUVbi4bvJycv/aq1BERERERKS61EypIW63m7d+38YVnyynqMKgXaNQpt98EjcPbUWAXUtUHQ+b1UK/FlE8elYHFt43lP9d2ZvRXRsTYLewfV8RT8/cRN9n5vDAtLVs3VtgdlwRERGROnNa0ml0j+1OqbOUbPtU7FaDZbtyWZuaZ1qm6MBo7u9zPz+M+YExLcdgMSz8nvo75884n5t/rdumisVi8Or53WgaFURqbgm3fL4Sp8tdZ9cXERERERHvpWZKDahwunhg2jqe/2kzbjf0i3Xx1TW9adMo1OxoHs9qMRjYKoZXL+jGsoeG8ezYTrRtFEqJw8mUxSkMe2UeV3+yjC1qqoiIiIgPMAyD+3rfh4HBr6k/M6BDPgATF+40ORk0DmnM4wMe59uzv2VU81GVTZXddd9UCQ+y894lPQm0W5m/NYsXft5cJ9cVERERERHvpmZKNTldbu78ajWfLUnBMGDCGW25oIULf81GqXEh/jYu6J3Ij7cO5PNr+nJah4ZYDJi9YS8jXp3HXV+tJjW32OyYIiIiIrWqXVQ7xrYaC0B2wFeAixmr09lXUGZusL80C2/GMwOfYdrZ0zij+RmmNFXaNArlhXM7A/DO3O3MXJte69cUERERERHvpmZKNbjdbu79Zg3frdqDzWLw1rjuXNI30exYXs8wDPo2j+LdS3oy6/bBjOjQCJcbvl6eyikvzuWZHzdSWFZhdkwRERGRWnNL91sItYeSXLCF5s02Uu50MfnPXWbHOkDz8OY8O/BZ05oqozo35ppBzQG466vVmsksIiIiIiLVomZKNbw+ZxtfL0/FajF448JunN4pzuxIPqdlbAjvXNKDb28cQL/mUZQ7Xbw7dwenvPg7U1ek4tIa2SIiIuKFIgMiua7LdQCUhEwHSymfLEqmpNxpcrKDmdlUuee0NgxoGUVxuZNr/7ecvBJHrV1LRERERES8m5opJ+indem88ssWAJ44u6MaKSbrmhDBlKv78MGlPWkaFURmQRl3fLmac95ZyKaMfLPjiYiIiNS4C9tdSLPwZhRW7Cc6fi65xQ6+XpFqdqzDOmJTZc7NrN63usavabNaeOPC7jSJCGRnVhF3fLFKb7YREREREZETombKCUjNLeaer9cAcOVJzRjXR0t71QeGYXBq+4bMun0Q94xoQ5CflRUp+znzjQW8+etWKpwusyOKiIiI1Bi7xc49ve4BwBE8D8NvHx/O34GznjcLDtlUSf2di2dezFU/X8Wf6X/idtdcDZHBfrxzcQ/8bBbmbMrk9V+31ti5RURERETEd6iZcpxcLjd3frma/NIKuiZEcN/pbc2OJP+Pv83KDUNa8uudQzi1XUMcTjcvztrC2LcXaq1sERER8SonNTmJQfGDcOEkJG4mydnFzN6w1+xYx+Tvpsq3Z3/L6JajsRk2Fmcs5upZV3PRzIuYkzIHl7tm3gzTKT6cp8d0AuDVX7YyZ6NnfI1ERERERKT+UDPlOH25bDeLd+YQaLfy+gXdsFv1JayvGoUH8P6lPXjl/C6EBdhYk5rHqNcX8O7c7VreQURERLzG3T3vxmaxQdBGrMGbeG/edrMjHZdm4c14YsATzBw7k3Ftx+Fv9Wdt1lpu++02xn43lunbp1Phqqj2dc7pEc+l/ZoCcNsXq9iZVVTtc4qIiIiIiO9QJ+A47C8u55kfNwFw5/DWJEYFmZxIjsYwDMZ0i2f2HYMZ2jaWcqeLZ37cxGUTl5CZX2p2PBEREZFqSwpP4uJ2FwMQ0GgGK1KyWLIzx+RUxy8uJI77+9zPT//5ias6XUWIPYTtedt5YMEDjJo2ii83f0mZs6xa13jojPb0bNqAgtIKrvlkGUVl1W/SiIiIiIiIb1Az5Ti89ft28koctG0Uyvj+SWbHkePQMCyADy7ryTNjOxFgtzB/axYjXpvPr5u0xIOIeJ/conJWpOSycFsW69LyyCkqNzuSiNSyazpfQ2RAJBa/LOyRi3jr921mRzph0YHR3Nr9Vn4+52du6XYLDfwbkFaYxhN/PsGIb0Ywad0kCssLT+jcfjYLb13cnYZh/mzNLOTur1fX6P4sIiIiIiLivWxmB/AUe/NLmbQwGYB7T2+LTct7eRzDMLiwdyK9kiK55bOVbEjP54pJyxjfP4kHRrbDz6bvqYh4rrIKJ9NWpDFlSQpr0/L4/68NNokI5LQOjRjTrQmd4sPNCSkitSbUL5Tbut/GhIUT8I/+hbnburEurQ0dm3juz3uYXxhXd76ai9tfzNStU5m4biJ7i/fy0vKXeHfNu5zT+hwuancRjYIbHdd5Y0MDeOuiHlzw3iJmrs3g3Xk7uG5wi1qqQkS8jdMJc+ca7NsHcXEwcCBYrWanEhERkbqgV4+P0cQ/kimvcNGzaQOGtI4xO45UQ8vYEKbd2J8rBjQDYNLCZC75cLHeuS0iHmv5rhxOf3U+901dy5rUykZK4/AAWsaGEBPqj2FA2v4SPvpjJ2e+uYAxb/3BjDV7qHDWzMbOIlI/nN3ybNpHtcewluEX8zNv/+5Ze6ccTqAtkIvaXcSPY3/k8f6P0yy8GYWOQiatn8Tp35zOffPvY2P2xuM6Z4+mDXj0rA4APP/TJuZv3Vcb0UXEy0ybZnDNNcMZNszGuHFw8smQlARTp5qdTEREROqCminHoKisgk8X7wLg2sEtMAzD5ERSXf42KxPObM+Hl/UkxN/G4p05jP7vH2zLLDA7mojIcfly2W7Oe/dPdmQVERPqz0NntGPJg0NZeP9QfrljMEsfPJU1jwzng0t7cmaXxvhZLaxM2c9NU1Yy/JV5/LQuQ0vciHgJi2Hh/t73A2CPWMZPW5exfd+JLYdVH9mtdsa0GsO3Z3/Lf4f+l16NelHhruCHHT9w3ozzuOrnq5ifOv+Yx7RxvRM5v2cCLjfc/NlKducU13IFIuLJpk6FCy6wkp0dcMDtaWlwzjlqqIiIiPgCNVOOwQ9r0ykorSApKoihbWPNjiM1aGi7hky7oT+JkUGk5BRz3rt/smFPvtmxRESOyeQ/d3HP12twutyM6hzHL7cP5qqBzYkNPfCX/NAAO6e2b8gbF3bjj/tO4dahrYgM9mNHVhHXTV7O+e/+yc6sIpOqEJGa1DW2KyObjcQw3Pg1/J63f/PcvVMOx2JYGBQ/iI9O+4jPR33O6c1Ox2pYWZyxmBvm3MCY78YwdevUo25WbxgGj53dgS7x4ewvdnDd5OWUOpx1VIWIeBKnE269ueKvZVQPfHNl5W1ubrulAqeGEBEREa+mZsox+Hp5KgDn9kzAYtGsFG/TqmEo3944gM7x4eQUlXPh+3+yfk9e5TPm33/H+PxzotauRc+MRaQ+mbNxLxO+WwfA1QOb8caF3QgPsh/1fjGh/tw+rDXz7jmZm09pSYDdwpLkHEa+Np8vl+2u7dgiUgdu73E7fpYAbEG7mL59Jqm53jvjokNUB54f9Dw/jv2RS9tfSrA9mO1523lk4SMM/3o4b6x8g4yijMPeP8Bu5e2LexAV7Mf6Pfk8MG1tHaYXEU8xfz6k7rHx/xspf3O7DXan2Zg/v25ziYiISN1SM+Uo0vNKWLIzB8OAMd2amB1HaklksB+Tr+pDt8QI8kocfHL3yzgTm8LJJ2O79FJOevhhbC1bau62iNQLqbnF3Pb5KlxuOL9nAg+MbHfcS1CG+Nu4c3gb5tw5hH7NoyhxOLnn6zU88+NGXI4K+P13+OwzjLlz1UwW8TCNghtxTeerALDFzOSN3zaYnKj2xYXEcXevu5l9zmzu7HEnDYMaklOaw3tr3mPENyO44/c7WJqx9JBLgDWOCOS/F3XHajGYuiKNact26Q01InKA9PSaPU5EREQ8k5opR/HrpkwAuiVE0Dgi0OQ0UpvCAux8fEVvLs9YwTNTHseyJ+3AA/bs0WK4ImI6l8vNHV+spqCsgm6JETw5pmO19vJqEhHIp1f14dahrQBIfm8y+XHxlTuqjhuHbdgwhl9zDca0aTVVgojUgcs6XEa0fyMs9jy+2/mpV89O+bdQv1DGdxzPj//5kZcGv0TPhj1xup3M3jWbK36+grHfj+XLzV9S7Djw69G3eRS3Dm3FaZsX0m9oT72hRkQOEBdXs8eJiIiIZ/K4Zsp///tfkpKSCAgIoE+fPixZsqRWr/fbX82UU7RXik8Is1t4cPa7wMETuI2/38l42216h6KImGbayjSWJOcQ7Gfl1fO7YrdW/79yi8Xg9mGtmRKZxtvfPk1Y9t4DPh+QnY31ggv0YqKIBwmwBXBf37sAsEXO5aVfl5qcqG7ZLXaGJw1n4oiJfHPWN5zb+lwCbYFs27+NJ/58glO/OpXnljzHttx/9pS5MXcN73z7NLH5WQeeTG+oEfF5AwdCfDwYxsGz2yq5iIkrZ+DAOo1VqwpK8miwdg3G559XzljW78AiIiKe1Uz54osvuOOOO3jkkUdYsWIFXbp04bTTTiMzM7NWrudyuVm8MweAwa3VTPEJ8+dj25N2+B8Mtxt270aL4YqIGUodTl6atRmAm05pRdOo4Jo7udNJ/zeewODgJwdVzWU1k0U8yvCmw2kZ1hHD4uCntIk+Mzvl/2vdoDUT+k3gl3N/4Z5e95AYmkiBo4DJGycz5vsxjPthHF9u/ILi+24DDjEG6g01Ij7PaoXXXqv8t4Hr/3228uOAMRP4ZcfKug1WS5Z8+RInfzOEX3a+i+3SSytnLCclqaksIr7B6cSYO5cm8+Zp2Ws5iEc1U15++WWuvvpqLr/8ctq3b88777xDUFAQH330Ua1cb0dWEQWlFQTYLbSLC62Va0g9c4hFblNi/XD9/2kqWgxXREzw1bLd7MkrpXF4AJcPSKrZk8+fD6mpB83Km90zjNwQa+WLiWomi3gUwzB47KT7AbCGreDpX34xOZG5wvzCuKT9JUwfM523hr7F0MSh2Awba7PW8sSSJznlnjAevLoJS9oG4/xrMExu+NfzQI2BIj5v7Fj4/HMnkVGlB9we18RFsxueIbz3D9w1/3p+2rLCpIQ1ZOpU1n38LC6LwbRBDcgOtVbenpamWXoi4v2mToWkJNwjhtH6nVexDRumZrIcwGZ2gGNVXl7O8uXLuf/++6tus1gsnHrqqSxatOiQ9ykrK6OsrKzq4/z8fAAcDgcOh+Oo11yeXDnFv2PjMNwuJw7X0TuRf5/3WM7vyby1TiMm5oAfit0xdi54pAX91xXy5AepBDgq35lYEROD28tq99bv6aEcrVZv+BpUd/w7Ub7yODKjTrfbzf8W7QLgigFNseLC4fj/74w8ccbu3Qc9KZgyNJJnLmlM7w2FvPdCMlY3VOze7XXjH/jOYxc0BtZWffX1MdQuoh29o4eyJGsOv2d9yM7MwcQ3CKrWOetrrcejb8O+9G3Yl+ySbGYmz+TblZ+w0z+b6QMaMH1AAyLzKui7oZCZ/SJISi/joU/20GdjkcZAD+cL4x9oDKxNo0Y5sNlmExR0Gvv22YiLg5NOcpNeeCXnfLeQclsKdy+4nqKylzmrbW+z4x4/pxPbLbfQNfCfmYwr2gQzbFk+uN24DQNuvZWKkSMrp+t4CV95/GoM1BhYXd5epzFtGtYLLuB/wyJ55dEOjJ6fyyMf78H9VzPZ+fnnuMeMMTtmjfL27+m/1dQY6DHNlKysLJxOJw0bNjzg9oYNG7Jp06ZD3ueZZ57hscceO+j2WbNmERR09F8iZ+2yABYCS3OYOXPmceWdPXv2cR3vqbyuTqeT4VFRBGRnYwBvjG1IQbCV9Cg7BuAGSqKjmZ2fD8f5mPAUXvc9PYLD1Vpc7PnLoFR3/KsuX3kc1WWdO/JhS6YNP4ub4H3rmTlzfY2eP2rXLk76f7f12VhEUImTJe1D+GBUDNdO38efu3aR7aXjH/jOYxc0BtaW+vgYGuTqxhL3XKxBO7jpi3e4JrFtjZy3PtZ6IqKI4u7dQwmf+AxTBzVgVq9wcsJtzOwXAUBynD9X3duMj57ZQZnGQK/gzeMfaAysbVYrlJX9TFgYFBXBzz9X3n5j+CW8nvMJTv/dPLL0VpasvIT+4UmmZj1eUWvXclJaGp3/tYbJilZBlc0U/lr2MDWVxS++SHanTialrD2+8PgFjYG1ydcfQx7N6WT4DTdgdbuJyq+gwmawo7E/UDn2uYHyG29kts3mVc3kv3nl9/QwqjsGGm63+3A7qNUre/bsoUmTJixcuJB+/fpV3X7PPfcwd+5cFi9efNB9DtWNTkhIICsri7CwsKNe88bPVjFrQyYPjmzD+H5Njymnw+Fg9uzZDBs2DLvdfkz38UTeXOffnWiAAW+2pSDYyv+e2E6XHZXTub2xEw3e/T39/45Wa35+PtHR0eTl5R3TWFEfVXf8O1G+8jgyo85Hpm9gypJU/tO9Mc+O6VjzF3A6sbVsCXv2/LM/ADC9fwQPXBOP1enmow8L6TRnk1c+efSVxy5oDKytmur7Y+iheS8xM/VTXOXRfDnyS1rHRpzwuep7rSfkX2NghcXN0rYh/NIzjG9PisBhr3xV8e1JhfT5caPGQA/mC+MfaAysTUerM7MwjzHTrqHEuhW3y85N7Z/kyu7DTEh6YozPP6/cIwWYNCKKly6I44yF+3n2vdQDjqv45BPcf/3O7A30+K2kMfDE6THk+Yy5cyuX9AI2NA3g/MdaEl5YwfybNh2wFHbF7Nm4Bw82J2Qt8Obv6f9XU2Ogx8xMiY6Oxmq1snfv3gNu37t3L40aNTrkffz9/fH39z/odrvdfkwPkN25lS+et4gNPe4H1LFew9N5ZZ3nnQc2G447bqUguPKX5WbpZdCkCcZrr2EbO9bkgLXLK7+nh3G4Wr2h/uqOf9XlK4+juqrT7XYzd0s2AGd0blw717Tb4fXXK9fCNozK/QGAMxfuZ2GHYGYMaMB918bzlVFGuD285q9fT/jKYxc0BtaW+voYmjDoZmZ9Op0Kvywe/PVDvr3kvmqfs77WekL+NQbaXQb91xfSf30hD3+8h4xIOxU2g4S3P4OAALOT1iqv+p4egTePf6AxsC4crs4mDaL5+YJPOOPLqymwrOPNDQ9S6izntv6j6z7kiUhIqPpnw9wKADIiD67TlpBQOW56GV9//HpL7RoDa59X1rlvX9U/W6WWEVDmIi/Exo7G/rTY809zzrZvn8Y/D1fdMdBjNqD38/OjR48ezJkzp+o2l8vFnDlzDpipUpNyiip/WGJCvPuXJjmEsWMp2rCq6sNV9z9CxdatlbsOiojUsW2ZhaTtL8HfZqFf8+jau9DYsfD119CkyQE33zGznAQjknR3Ho8sfAQPmdQqIv8SbA/mivbXA7DNMZU/k3ebnKgeOsQYaACNghpWNlL0PFBEjkGDoBB+vvAjGtAdw1LBB1se4cnfPzU71rEZOBDi48EwiMmtXDs+O/xf78E1jMqGy8CBJgUUEaklcXFV/7Q73XTZXrnk0/I2QYc9TnyTxzRTAO644w7ef/99Pv74YzZu3Mj1119PUVERl19+eY1fy+12k1tU+eQhIsg3OnNyoCJX5cwkf6s/+zt18colHUTEM8zdUvkumb7Nowj0q+WxaOxYSE6G336DKVOomD2bha+/wzOnvY7NYmNOyhymbZtWuxlEpFbc0PNCgo0EDGsJD/7+stlx6qd/jYEVn3zCgiee0BtqROS4hfoH8tO492ho6YdhuPhi17Pc9uPrZsc6OqsVXnsNAP+KyjfPlNkrF7hxUbmHKK++qt+NRcT7/KuZDNBjcxEAy9oEV35ezWT5i0c1U84//3xefPFFJkyYQNeuXVm1ahU//fTTQZvS14SyChflThegZoqvKnJUDpxBttrfrFtE5EhWp+YB0LtZZN1c0GqFIUPgwgsr14O1Wmkf2Z5but0CwHNLniOtMK1usohIjbFarNzf+24A9vIr361bZW6g+uqvMdB9wQWVGyzrRUMROQFBdn9+HPc2zfxOA2BO5vtcMvURXC6XycmO4q9ZevYGMQA4bJUvLGaERvPKNU9TcfZoE8OJiNSSfzWTMQx6bK6cmbKsTTCuvxosaiYLeFgzBeCmm25i165dlJWVsXjxYvr06VMr1yl1OKv+HWDXD4ovKnZUDpxqpoiI2dalVTZTOjUxd6+SS9tfSvfY7hRXFPPQgodwuev5iwEicpCz255MQ1s3DMPFM4tf1LJ9IiK1yG618u35z9M99EIAVhVM5ewvbqO8wmFysqMYOxbnjBkAOCPCSZv6A8NvmcTrEZ145sdNJocTEakl/1rytfP2YgLKXOxrYGdlx8aVt2umsuCBzZS6Ul5R+QKRYYDNYpicRsxQ1Uyxq5kiIuYpdTjZmVU5U6594zBTs1gtVp4c8CSBtkCW7V3Gpxs9ZP1vETnA04Pvx+22UGRbzTuLZ5kdR0TEq1ksFj4e+wDDYm7C7TZILv+NYVMuJ7ekwOxoR1T817LXwSGRNBkzkufO6w7Ahwt28vHCZBOTiYjUor+WfLXNnEVCUeVKSLffcKkaKVJFzZTDqHBVvkvPZjEwDDVTfFFRReWLl8H2YJOTiIgvS80tASDE30ZUsJ/JaSAhLIG7et4FwGsrXmNH3g6TE4nI8eod3442QcMAeG/9q5Q56vk7pEVEvMDLI6/l4mYTcLvs5LhXc9oX57Mzt/4um1pYXghAiD0EgDM6x3H3aW0AeGz6en5cm25aNhGRWmW14h48mMSGfQHIYg25ReUmh5L6Qs2Uw7BZKxsofzdVxPdozxQRqQ9251TOkotvEFhvmvvntj6X/o37U+Ys48H5D1LhqjA7kogcp5eG3QOuACpsqUz49ROz44iI+IT7Bp/D3Z1fxV0RQomxmzHfns+ClFVmxzqkjOIMAKIDo6tuu2FICy7olYDLDTdMWaEZKiLi1boFVjaQLQEpfLJkvclppL5QM+Uw7JbKL43bDU41VHzS380UzUwRETPtza9cYqFxRKDJSf5hGAaP9X+MUHso67LX8eHaD82OJCLHKalBLINjK9fw/yFlEtN+LOKzz+D338HpPPJ9RUTkxF3WYxCvDpyIUd4IpyWP6+dcycervzU71kF2F+4GICE0oeo2wzB4cnRHLuydiNsNj3y/nqd+2IBLr5mIiBeKsEQQ49cUw3AzZe1svT4sgJoph/XvTedLHPqN0hdpA3oRqQ/ySiqX34kItJuc5ECNghtxf5/7AXhn9TtszN5ociIROV7PDbuBgiWj2XzvV4wdGcy4cXDyyZCUBFOnmp1ORMR7ndq6LZ+fORl7eVuwlPPiqoe599dncbrqz2sP2/dvB6BpaNMDbrdZLTw9pmPVkl/vz9/JTZ+toFSvm4iIFzqt2RAA8ljD3C2ZpmaR+kHNlMMIsFvws1V+ebQunm8qcFRuCPj3GrEiImb4u5kSVs+aKQCjmo/i1MRTqXBX8MCCByh36v9LEU/y84wAdr31OBW5sQfcnpYG55yjhoqISG1q36ghP13wMRHllXtYzdz9Ked9ezW5pbkmJwOX28WarDUAdIrudNDnDcPgxpNb8ur5XfGzWpi5NoML3vuTrMKyuo4qIlKrTkkcDIAtZDOTFm0zOY3UB2qmHIZhGEQGVW70u79Ym3L6ooLyymZKqF+oyUlExJcVlFbuRxIWYDM5ycEMw+Dhfg8TGRDJtv3beHPVm2ZHEpFj5HTCrTf/vd/Rgb8SuN0Abm67pUJLfomI1KLY0CBmj3+BVsY1uF02thQsZcRXo1mU9qepuTZkb6DQUYgffrSKaHXY40Z3a8L/ruxNeKCdVbv3M/athaTnldRhUhGR2tU5ujOR/tEY1lIWpv3JruwisyOJydRMOYLo0MpmSmZBqclJxAyF5YWAmikiYi5n5aua2Kz187/syIBIHun3CACT1k1iZeZKkxOJyLGYPx9S99gA45Cfd7sNdqfZmD+/bnOJiPiaALuVry6+iRGRT+Esi6HYlcM1v1zDS0tfxeEy542ds5JnAdDG3gab5chv6OnTPIqpN/QnMTKIlJxiLv1wCcXlFUe8j4iIp7AYFoYnnQqALXQtk//cZXIiMVv9fGWmnkhoULlXRkpOsclJxAxVM1PsaqaIiHncfzVTLId+vbNeOCXxFM5qcRZu3Dy04CFKKvSORJH6Lj29Zo8TEZETZ7UYvHj2SB7o8i4V+3sDbiZt+JBxMy5ld8HuOs3icDr4YecPAHS0dzym+7SICeGza/oSG+rP1sxCHp++oTYjiojUqeFJwwGwhW7g86XJ2iPKx6mZcgSJkWqm+LL88nwAQvy0Z4qImMflqvzbMOpxNwW4r/d9xAbFklKQwpsrtdyXSH0XF1ezx4mISPVd3KcVn5z9AtZ9l+F2BrApdx1jvzuHLzZ9UWeb08/cOZPM4kyiA6NpY29zzPdrEhHIqxd0xTDg86W7mblW3XgR8Q7dY7sTGRCJYS2h2LqZ3zfvMzuSmEjNlCNIig4GYOveQpOTiBkyizMBiAmMMTmJiPgyy1//U7tcbnODHEWoX2jVcl//2/A/VmWuMjeQiBzRwIEQHw+GceixxTDcJCRUHiciInWnV1Ik06+4gYYFD1BR3JRSZzFPLn6SS368hA3ZtTvjo9hRzOsrXwdgXJtx2Izj27Ovf4torh/cAoCHv11HblF5jWcUEalrVouVUxP/XuprHdPX7DE5kZhJzZQj6NA4DIB1e/KqllkR31DhqiCrJAuA2KBYk9OIiC/zt1kBKK2o/1OJB8UPqlru6+E/Hqa0QnuOidRXViu89lrlvw1cB3zu749ffbXyOBERqVsJkUF8d92Z9A+aQGnGmbid/qzNWsuFP1zI04ufrlpFoaY9v/R5MosziQ+J58I2F57QOW47tTWtG4aQXVTOEz9ouS8R8Q7DkoYBYAtdz5yNeygq095QvkrNlCNo0ygUu9Vgf7GD1Fyt/+5LskqycLqd2Awbkf6RZscRER8W6Ff5SmZJuesoR9YP9/S6h+jAaJLzk3lr9VtmxxGRIxg7Fr7+2qBJ/IHLCMbHG3z9tcHYsSYFExERQvxtvH9JL67qcilFO+7EkdcFl9vFZ5s+Y8Q3I3h79ds12lSZvGEy32z9BgODCf0m4G/1P6Hz+NksPPufzhgGTF2RxrwtWg5HRDxfz4Y9ifCPwGIrwmHfzi8b95odSUyiZsoR+NustI+rnJ2yNDnH5DRSl9KLKtd3jQ2KxWrRWzJFxDwBf81MKfGQTe7C/cN5uO/DAHy8/mPW7ltrciIROZKxYyE52eC332DKFPjtN9iZrEaKiEh9YLEY3DuiLS+NHYhr70UU77oSW0UjCsoLeGvVW4z4egRvrXqraonqE+Fyu3h79ds8t/Q5AG7pfgv9GverVu7uiQ24rF8SAI9+v57yCs94U5CIyOHYLDaGJg6t/HfYGqav1r5QvkrNlKMY0DIagPlbs0xOInVp+/7tADQLb2ZyEhHxdcH+lc2UglKHyUmO3SmJp3B6s9NxuV08/MfDlDu1XrZIfWa1wpAhcOGFlX9raS8RkfplbPd4Pr+2L1HWjuRuvQVXxsXEBjSlwFHA26vfZvjXw7lpzk3M2TWHYkfxMZ93fdZ6rvz5St5aVTmb+NrO13JlxytrJPOdw1sTHeLHjqwiPlmUXCPnFBEx08hmIwGwh61h7tY95BV7zu/oUnOObzcxH3RSq2je+n0787dm4XK5sViMo99JPN7fzZQWES1MTiIivi4qxA+A3GLPakjc3/t+FqcvZnvedt5Z/Q63dL/F7EgiIiIiHqt7YgN+uGUgt36+koXbO7I9tz0nd8/AETKP1ftWMTd1LnNT52K32Oke252ejXrSIqIFSWFJhPqF4mf1o6C8gPSidNZlreO3lN9Yk7UGgEBbIPf2upf/tP5PjeUNDbBz1/A23Dd1La/P2cr5vRIIDbDX2PlFROpaz0Y9iQ2KJbM4E3fgRn5e343zeiWYHUvqmJopR9GjaQNCA2xkFZaxJDmHvs2jzI4kdWDb/m2AmikiYr4GQZXNlJwiz3rXS4OABjzU9yHu+P0OPlr3Eac2PZX2Ue3NjiUiIiLisWJC/fnflX147ZctvPHbNn5b0ZiOTa7hnbOiWLzvJ35K/on0onQWZyxmccbio57Palg5vdnp3Nj1RuJD42s877k9E3h//g627yvi08UpXDdYv1+LiOeyGBbOaHYGE9dPxBa2iu9XD1EzxQdpma+j8LdZGdkxDoDvVqWZnEbqgsvtYn32egDaNGhjchoR8XVRwZWbf+YWedbMFIBhTYcxvOlwnG4nD//xMA6nZzWEREREROobq8XgjuFtmDi+Fw2C7KxLy+e6ibvpGDiOn//zM9NHT+e+3vdxVouz6BjVkXD/cGyWyvfRBtuDSQxNZHjT4dzf+35+OfcXnhn4TK00Uv7Oev2QlgB8MH8npR6yB6CIyOGc0fwMAGwhG1m4czeZ+aUmJ5K6ppkpx+Dsro35YtluZqxJ56Ez2hPsry+bN9uZt5OC8gICbYG0jmwNer4nIiaK/GuZr+yiMo9cbvKBPg+wNGMpW3K38MHaD7i+6/VmRxIRERHxeEPaxPLDLQO5acoKVqTs55r/LeeaQc25+7Q2XNQu6aDjnS4nVkvdb4p1dtfGvDJ7C2n7S/huVRrn90qs8wwiIjWldYPWtIxoybb927CGrGPGmh5ccZL2W/YlmplyDPo0jyIpKoiC0gq+Xp5qdhypZSszVwLQMbojdovWdBURc8WG+mMxwOF0k1VYZnac4xYVGMX9fe4H4L0177E5Z7PJiURERES8Q+OIQL64th9X/vVC3nvzdjDmrT/YsrfgoGPNaKQA2K0WLupb2UD5ZrlW+xARz2YYxj+zU8JX8t3qPSYnkrqmZsoxsFqMqi7jhwt24nS5TU4ktWnhnoUA9GjYw+QkIiKVv4A2CgsAIHV/iclpTsyIpBGcknAKFe6KyuW+XFruS0RERKQm2K0WHh7Vnncu7kHEX8t+jXpjAe/P21FvXrsY060JhgFLknNIyS42O46ISLWMbDYSAGvQTtakJ5OcVWRyIqlLaqYco3N6xNMgyE5KTjFTlqSYHUdqSbmznD/S/gBgSPwQc8OIiPwlvkEQAKm5ntlMMQyDh/s9TJhfGBtzNjJx3USzI4mIiIh4lREdGzHrtkGc3CaG8goXT83cyIXv/Vkvmhdx4YH0bRYFwG+bM01OIyJSPY1DGtOjYQ8Mw409bDXfa3aKT1Ez5RgF+dm4fVhrAF78eTM5HrgRsBzd0oylFFcUExsYS7uodmbHEREBIL5BIACpueb/MnyiogOjua/3fQC8s/odtuVuMzmRiIiIiHeJDQvgo/G9eGZsJ4L9rCxJzmHEa/OYsjgFt9vcWSoDW0cDsGBblqk5RERqwj9Lfa1g2qpU08dYqTtqphyHcb0TaRcXRl6Jg0e+X68fFC/0w44fABiSMASLoR8PEakf/m6m7M7x3GYKwKjmoxgcPxiHy8GEhRNwupxmRxIRERHxKoZhcGHvRH66bRC9m0VSXO7kgWlrGT9xqanPJfs0iwRgbWqeaRlERGrKaUmn4Wf1xxqwl10Fm1m/J9/sSFJH9GrxcbBZLTw1piNWi8H01Xv435+7zI4kNaigvIDZu2YDcHbLs01OIyLyjxaxIQBsyyw0OUn1GIbBw30fJsQewtqstUzZNMXsSCIiIiJeKSEyiM+v7stDZ7TDz2Zh7pZ9nPryXF77ZSuljrp/Q0vz6Mrnsxn5paZcX0SkJoX5hTGs6akA2COWaamvf3E64fff4bPPKv92etmQr2bKceqe2ID7T28LwBMzNrBoe7bJiaSmfL/9e0qdpbQIb0Gn6E5mxxERqdLyr2bKlr2FHj8rsmFwQ+7oeQcAb6x8g9SCVJMTiYiIiHgni8XgqoHNmXnLSfRvEUVZhYtXftnC8Ffm8eumvXWaJSzQXvVvNVNExBuMaTkGAHvYKqat2kmF02VyIvNNnQpJSXDyyTBuXOXfSUmVt3sLNVNOwJUnNWNkp0Y4nG6umLSUxTvUUPF0DqeDSesnAXBh2wsxDMPcQCIi/9IiJgTDgLwSB1mFnr9n139a/YeeDXtSUlHC44se9/gGkYiIiEh91jI2lE+v6sMbF3ajYZg/KTnFXDFpGVd9vKzONqgv+VcDxWrR79si4vl6NepF4+DGGNYyclnB3C37zI5kqqlT4ZxzIPX/vV8yLa3ydm9pqKiZcgIMw+Dl87oysFU0JQ4n4ycu9fkfGE83fcd0MooyiA6MZnSr0WbHERE5QIDdSmJkEOD5S30BWAwLj/Z/FH+rP4vSF/H99u/NjiQiIiLi1QzD4MwujZlz5xCuHdQcm8Xgl417Gfry79w/dS1p+0tq9fqb0iv3E4gO8Sc0wH6Uo0VE6j+LYWF0y9FA5VJfXy7bbW4gEzmdcOvNFYd8o2TlTW5uu6XCK5b88phmylNPPUX//v0JCgoiIiLC7DgE2K28f2nPfzVUlvDy7C04XXp3racpLC/kjZVvADC+w3j8rf4mJxIROVirqqW+CkxOUjOahjXl+i7XA/D80ufJKskyOZGIiIiI9wvxt3H/yHb8dNtABraKxuF089mSFIa88BsPTlvLnlpqqny7Kg2APs0ja+X8IiJmOLvl2RgY2IK38+vWTewrKDM7kinmz4fUPTbg0DMP3W6D3Wk25s+v21y1wWOaKeXl5Zx77rlcf/31Zkep8ndD5cLeibjd8PqcrVw2aRkZxzlL1ts35qnv3l3zLlklWTQNa8qFbS80O46IyCG1jwsDYP2ePJOT1JzLOlxGu8h25Jfn8+ySZ82OIyIiIuIzWsaG8r8r+/DVdf0Y0DIKh9PNp4tTGPLC79z91WpW7d5PTa3EuiIll8+XVL5j+6I+iTVzUhGReqBxSGP6xPUBwAhbxrSVvrknaHp6zR5Xn3lMM+Wxxx7j9ttvp1On+rUxeIDdyjNjO/Hq+V0J8rOyeGcuz6228uj0jWQXHr0b6Qsb89RnK/au4JMNnwBwT6978LP6mZxIROTQOjYJB2BNqvc0U2wWG4/2fxSrYeXn5J/5LeU3syOJiIiI+JReSZF8elVfvrimL/2aR1HudPHV8lTOfW8JL6yx8umS3eQVO074/Au2ZjH+oyVUuNyc0TmO/i2iazC9iIj5qjaij1jO50t3+eSeoHFxNXtcfeYxzZT6bnS3Jvx460CGtYvFhcGnS3Zz0nO/ceeXq1m8I/uQP0i+sjFPfZVXlse98+/F5XYxqvkoBsUPMjuSiMhhdY6PAGBrZiEl5d4zhbF9VHsu7XApAE/++SQF5d6xjJmIiIiIJ+nTPIrPrunLN9f3Y2y3JvjZLKQVGzw6fSPdnpjFmLf+4OXZW1i8I5uC0iM3V5wuN0t25nDjpyu4+MPF5JdW0KNpA54dW7/eHCsiUhNOSTyFUHsoFvt+dhWvYWlyrtmR6tzAgRAfD4Zx6EaSYbhJSKg8ztPZzA5Qm8rKyigr+2d2SH5+5YZnDocDh+PE31lxOI3D/Hjt3A689XU6v+aGsyG9kG9WpPLNilQSGgRyUssousSH0zk+nKQGwdxyswW328r/X0/O7a58kN16i5ORI91YrTUetdr+/vrVxtexLjhcDu6Zew8ZRRkkhCRwb497D1uLp9d6rHylTjh6rd7wNajr8e9vvvI4MqPOyEALMSF+7CssZ+3uHLolRtTJdeui1qvbX80vyb+wu3A3Ly99mQd6P1Br1zocX3nsgsbA2qpPjyHv4yt1gu/U6gvjH2gMrE2+UGfnxqF0HtuBu05txgtfzWV9cRhbMotYmbKflSn7eX3OVgDiGwTSIiaYyGA/GgTacbrdFJU52Z1bzOaMQvaX/PM1Gtc7nntPa02AtX5+7Xzh+woaAzUGVp+v1AnHV6sVKyOSRvDV1q+wRyxh0h8n0S0+tLYj1oia/J6+9JLBBRdYARf/nr9h4AIMXnyxApfLjctV7UudkJoaAw23iXOP7rvvPp577rkjHrNx40batm1b9fGkSZO47bbb2L9//1HP/+ijj/LYY48ddPuUKVMICgo67rzHw+2G5EJYnGlhRZZBmcvAsOUT1PQdLH45FG7sSfJzE496nieeWECnTtm1mtXXuNwuppVMY2X5SuzYuTrkahrbGpsdS+qR4uJixo0bR15eHmFhYWbHOSFmjn9Se97bZGF9roX/JDkZFOddU4d3OHbwUdFHAFwZciXNbM1MTuS7NAaKiK/yhvEPNAZKzdtfBpvyDDbtN9hZYLC//NAbDP9boNVNp0g3J8e5aBxcByGl2jQGipy4DGcGbxa8idttoWTbfUzoHESEv9mp6t6iRXG8/G5zHPv/WdIxOqqYK69aR79+9XvDlGMdA01tpuzbt4/s7CM3Cpo3b46f3z/7WBxPM+VQ3eiEhASysrJq7T8Gh8PB7NmzGTZsGHa7HYDi8grmbc1mfvImfsy7HYD9f55O6jvPH/V8n3xSwQUX1L8XzA5VpyeocFXw+OLHmbFzBhbDwsuDXmZQkyMv7+WptR4vX6kTjl5rfn4+0dHRHv0k0ozxD3zncWRWnW/8up3Xf9vO2V3iePGculkmoS5rfWLxE0zbPo2moU35fOTn+Fvr7tmnrzx2QWNgXT4H9Fa+Uquv1Am+U6svjH+gMbA2+UqdcORac4vL2ZxRSEpOMbnFDvaXOLBZDALtVhqF+9MqNoS2jUKxWz1jZXlf+b5qDNQYWF2+UiecWK2Xz7qc1VmrKcs8jas7XcltQ1vWcsrqq+nvaYXTRc/Jgyne1o6xUfdzerfmnHRS/Vh1qabGQFOX+YqJiSEmJqbWzu/v74+//8EvxNjt9lr/of/3NcLtds7sGs+ZXeO5bv93fLvtO6anFpN6lHMAvJX8MMsXlNAiogUtI1rSIqIFzcKaYbfWj0GrLr6WNSW/PJ975t/DH2l/YDWsPH3S0wxNGnrM9/ekWqvDV+qEw9fqDfWbOf7V5XXMVtd19moeBb9tZ3nK/jr/+tZFrXf1vosFexawq2AXH274kFu731qr1zsUX3nsgsbA2qLHkPfxlTrBd2r15vEPNAbWBV+pEw5da2y4ndhw75tu4ivfV42BtcfXH0Pe6HhqPb/t+axesBp7g8V8vmwYt5zaGn9bPegiHIOa+p7+sXs9FnsRwW1W8eTFTQm0178dRqo7Bta/ig4jJSWFnJwcUlJScDqdrFq1CoCWLVsSEhJibrjj0DyiOXf0vJ1bu0HSO5CW5sbtPtQUWRf2yL0UxM9kToqLOSlzqj5jM2wkhiUe0GBpGdGSxLBE7BbfGMyO17qsddw3/z525e8iwBrAs4OeZWjisTdSRETqg+6JDbBaDFJzS0jbX0KTiECzI9WoML8wHuz7ILf9dhsT103ktKTTaBvZ9uh3FBERERERETHRsKbDeG7Jc+Sxn/2sY+baDozpFm92rDr1y/ZlAASRSKDdO9c585hmyoQJE/j444+rPu7WrRsAv/32G0OGDDEp1YmzWuG11+Cccyo34nH//415DIN33wyi7eB32L5/O9v2b2Pb/m1s37+dQkchO/J2sCNvB7N3za66n81iIyks6YAGS4uIFiSEJmCzeMy3ukYVO4r5YO0HfLTuI5xuJ42CG/Haya/RPqq92dFERI5bsL+Njo3DWJ2ax9KdOTTp1sTsSDVuaOJQhjUdxuxds5nwxwSmnDHFZ/8PExEREREREc8QYAvg7JZn88mGT/CL+JMP5vdmdNcmGMbR95nyFqsy1wCQGOy9b4r0mFcnJk2axKRJk8yOUaPGjoWvvza49VZI/deaX/HxBq++ZjB2bDjQj36N+1V9zu12s7d4b1WD5d9/F1cUVzVd/s3P4kdSeBItIlrQIrwFzSOa0yK8BQlhCV47k8XhcjBj+wzeXPkmmSWZAJze7HQe7PMg4f7hJqcTETlxvZtFsjo1jyXJOYz2wmYKwAN9HmBx+mI25mzkfxv+x+UdLzc7koiIiIiIiMgRndP6HD7Z8AnWkM1s2LaLBdvaMrBV7W1xUd+klWwBO/Ru3MXsKLXGY5op3mrsWDj7bIP58yE9HeLiYOBA47Ab8xiGQaPgRjQKbsSAJgOqbne73aQXpR/QYNm2fxs783ZSUlHCltwtbMndcsC5/r1cWPPw5jQPb06LiBYkhSfV6aa/NSm3NJfvtn3H5I2T2Vu8F4AmIU24u+fdDG2qZb1ExPP1Sork/fk7WbIzx+wotSY6MJq7et7FhIUT+O+q/zI0cSiJYYlmxxIRERERERE5rGbhzejTqA+LMxZjb7CUd+a29JlmSur+fBzW3RjAGa37mB2n1qiZUg9YrVDdlcoMw6BxSGMahzRmUPygqttdbhdphWls37+dHXk7Kv/eX7lEWHFFcdVyYf9mMSw0CWlCi/AWJIYlkhiaSEJoAglhCcQFx9W75VZyS3NZuGchP+38iQVpC6hwVwCVL8aN7zCeC9teiJ/Vz+SUIiI1o1dSJADbMgvJKiwjOsQzm99HM7rlaGbunMmf6X/y2KLH+GD4Bz41PVpEREREREQ8z7ltzmVxxmL8Ipbyx9ahrE3No1O896+S88OmFRgWJ4YrmPYxzc2OU2vq16viUuMshqWyERKawJCEIVW3/3u5sL8bLX83W/LL89ldsJvdBbsPOp/NsNE4pDHxIfE4i53kbMqhcWhjYoNiiQ2KJSYwplYbFy63i90Fu9mQvYEN2RtYmrGUDdkbcOOuOqZdZDsuaHsBo5qPUhNFRLxOg2A/2jYKZVNGAX9sy+Lsrt651JdhGEzoN4Gx341lScYSvt32LWNajTE7loiIiIiIiMhhnZJwClEBUWSXZmML3cA7c+P570XdzY5V6xakrAAg1q+lV78RUs0UH3Wk5cKyS7PZsX8H2/O2k5KfQmpBKikFlX+Xu8pJKUghpSAFgMUrFh907gb+DYgJiiE2KJbowGjC/cIJ9QslzD+MUL9Qgm3B+Fn9sFvs+Fn9sFlsON1OKlwVVLgqcLgcFJYXsr9sP/vL9pNbmktaYRqpBamkFaZR6iw96JqtG7RmcPxgzmh+Bi0iWtTeF05EpB4Y3CaGTRkFzN2yz2ubKQAJoQnc0PUGXl7+Mi8tf4nBCYOJDIg0O5aIiIiIiIjIIdmtdsa2Gsv7a9/H3mARM9d1YltmAS1jQ82OVqu25m0AP2gf1dHsKLVKzRQ5gGEYRAdGEx0YTe+43gd8zuV2kVmcye6C3ezM3cnc1XPxb+hPVmkWmcWZ7CveR7mrnNyyXHLLcg/ao6Wm+Fv9adOgDe2i2tE5pjP94voRE+Qb6w+KiAAMbh3Du3N3MG9LFi6XG4vFe9/1cXH7i/lhxw9szt3MC0tf4JmBz5gdSUREREREROSwzmtzHh+t+wiCd2D4pfPqL1t5c5z3zk4pKHVQ4N6BBRjarKfZcWqVmilyzCyGpWo2S9eorvhv8WfkSSOx2+1A5ayWvLI89hbvZV/JPjKLM8kqyaKgvICC8gLyy/PJL8+n2FGMw+XA4XRQ7irH4XJgNazYLXZsFhs2i40QewgR/hGE+4cT4R9RubRYaDzxIfHEhcRht9hN/mqIiJinZ9NIgvysZBWWsSE9n45NvHf9VbvFzqP9H2XcD+OYsWMGZ7Y4k/6N+5sdS0REREREROSQGgU3YmjiUGbtmoU98g9+WBvHzRkFtGnknbNTft+SgsV/HwAnJXYzOU3tUjNFaoxhGEQERBAREEEb2pgdR0TEa/nZLPRvEcUvGzOZt3WfVzdTADpGd2Rcu3F8uvFTnlj0BFPPnkqgLdDsWCIiIiIiIiKHdHH7i5m1axYBEaspzzyd1+Zs4a2Lepgdq1b8tG0ZAIFGNFGBUSanqV0WswOIiIjI8RvcunJ5w7mb95mcpG7c3O1mGgY1JLUwlXdXv2t2HBEREREREZHD6hrTlXaR7XDhwB6xhJlrM9iwJ9/sWLVi5d61ALQIb2dyktqnZoqIiIgHGtw6FoDlu3LJL3WYnKb2BduDeaDPAwB8vP7jWtuXS0RERERERKS6DMPg4vYXAxAauwRw8tKszeaGqgX7CsrIqdgOQP/4LianqX1qpoiIiHigxKggmscEU+Fy89umTLPj1IlTEk9haOJQKtwVPLboMVxul9mRRERERERERA5pRNIIIgMiKScX//D1zNmUycJtWWbHqlELt2dhDUgFoHfjruaGqQPH3Uy57LLLmDdvXm1kERERkeMwokMjAGat32tykrpzf+/7CbYHs2bfGr7c/KXZcUREREREREQOyc/qx3ltzgMgLqFyX5Enf9iI0+U2M1aN+m3rTix++wFoH9Xe3DB14LibKXl5eZx66qm0atWKp59+mrS0tNrIJSIiIkdx2l/NlN82Z1LqcJqcpm40DG7Ird1vBeC1Fa+xt8h3GkkiIiIiIiLiWc5vcz42i41s5xZCw9LZkJ7PNytSzY5VI9xuN3/sXglAbEACoX6hJieqfcfdTPn2229JS0vj+uuv54svviApKYnTTz+dr7/+GofD+9dsFxERqS86x4cTFx5AcbmT+Vu9a6rwkZzX+jw6R3em0FHIs0ueNTuOiIiIiIiIyCFFB0YzImkEAG3brAbgxZ83U1RWYWasGrEpo4A81w4AejTqbHKaunFCe6bExMRwxx13sHr1ahYvXkzLli255JJLaNy4Mbfffjtbt26t6ZwiIiLy/xiGUTU75ef1GSanqTtWi5UJ/SZgM2z8kvILv6X8ZnYkERERERERkUO6qN1FAGwrXkB8tIPMgjLenbfD5FTV9+umzKr9UrrEdjI5Td2o1gb06enpzJ49m9mzZ2O1Whk5ciRr166lffv2vPLKKzWVUURERA7j72bKLxv3UuH0nQ3Z20S24dIOlwLw1OKnKHIUmZxIRERERERE5GAdozvSLbYbFa4KundcD8A7c7eTnOXZv8fO2bQXS2BlM6VDVAeT09SN426mOBwOvvnmG0aNGkXTpk356quvuO2229izZw8ff/wxv/zyC19++SWPP/54beQVERGRf+mV1IDIYD/2FztYvDPH7Dh16rou19EkpAl7i/fy5so3zY4jIiIiIiIickjjO4wHYFnOD/RrGUJ5hYsJ36/H7fbMzehzi8pZvScZi60Qi2GlTWQbsyPVieNupsTFxXH11VfTtGlTlixZwrJly7juuusICwurOubkk08mIiKiJnOKiIjIIdisFoa1awjAjDXpJqepW4G2QCb0nQDAlE1TWJe1zuREIiIiIiIiIgcbkjCEpLAkChwF9O60BT+rhXlb9jFzrWcu2f37lkyMv5b4ahXRkkBboMmJ6sZxN1NeeeUV9uzZw3//+1+6du16yGMiIiLYuXNndbOJiIjIMTira2MAZq5Np7zCd5b6AujfpD8jm43E5Xbx2KLHqHB5/iZ+IiIiIiIi4l0shoXLO14OwA8pX3DN4KYAPD5jPQWlDjOjnZAf12Zg+auZ0jG6o8lp6s5xN1MuueQSAgICaiOLiIiInIC+zaOIDfUnr8TB75szzY5T5+7pdQ9hfmFsytnE5A2TzY4jIiIiIiIicpBRzUcRHRjN3uK9NE/aStOoIPbml/H0zI1mRzsuRWUVzN2yr2rz+Q7RvrFfClRzA3oRERExn9VicFaXytkp363aY3KauhcVGMVdPe8C4K3Vb5FWmGZyIhEREREREZED+Vn9uKjdRQBM3vQxz4zpBMBnS3Yzd8s+M6Mdl982Z1JW4cQeVPm7t69sPg9qpoiIiHiF0d2aAPDLxr0eOUW4uka3HE3Phj0pqSjhyT+f9NhN/ERERERERMR7ndfmPIJsQWzN3YorcBPj+ycBcO/Xa8gr8Yzf5X9cm4Fhz8ZtKcHP4kerBq3MjlRn1EwRERHxAh0ah9EiJpiyChc/r99rdpw6ZxgGE/pNwG6xsyBtAT8n/2x2JBEREREREZEDhPmFcW7rcwGYuG4i945oS7PoYDLyS3l8+gaT0x1dUVkFv27KxBpYucRX28i22C12k1PVHTVTREREvIBhGIzuWjk75btVvrnMVbPwZlzd6WoAnl/6PAXlBSYnEhERERERETnQxe0vxmbYWJKxhB35m3jx3M5YDPhmRSrTV9fvpbt/WpdBicNJgwaVb+L0pf1SQM0UERERr3H2X82UP7ZlkZlfanIac1zR6QqahjVlX8k+3lz5ptlxRERERERERA7QKLgRI5uPBODDdR/So2kkNwxpCcD9U9eyM6vIzHhHNG1l5Zs3wyMyAOgY3dHMOHVOzRQREREvkRgVRPfECFzuf57g+Bp/qz8P9nkQgM83f8767PUmJxIRERERERE50BUdr8DAYPau2WzN3cptp7aid7NICssquOHTFZQ6nGZHPEhGXil/bM8CnOx37gR8a/N5UDNFRETEq5zfKwGAL5bu9tlN2Ps17sfpzU7H5XbxxKIncLrq35NQERERERER8V0tIlpwatNTAXh/zfvYrBbeuLAbUcF+bEzP57F6uH/K1JWpuN3QqVkZZc5SgmxBJIUlmR2rTqmZIiIi4kVGdW5MsJ+VHVlFLNmZY3Yc09zT6x5C7aGsz17Pl1u+NDuOiIiIiIiIyAGu7XwtAD8l/8SOvB00DAvg1Qu6Yhjw2ZIUvliaYnLCfzhdbj79szJPp+b5ALSPao/VYjUzVp1TM0VERMSLBPvbOKtrYwA+X7rb5DTmiQ6M5ubuNwPw+orXySrJMjmRiIiIiIiIyD/aRLbh5ISTcePmgzUfADCwVQy3DW0NwEPfruPPHdlmRqzy66ZM0vaXEBFkxy+4cllxX9svBdRMERER8Trn90oEYObadPKKHSanMc95rc+jQ1QHCh2FPL/0ebPjiIiIiIiIiBzg2i6Vs1Nm7pxJSn7lzI+bT2nJGZ3jcDjdXDd5Ocn1YEP6TxYlA5VLi2/KqVyCrEO0b+2XAh7STElOTubKK6+kWbNmBAYG0qJFCx555BHKy8vNjiYiIlLvdIkPp22jUMoqXHy7yjc3ogewWqw83O9hLIaFH3f+yKI9i8yOJCIiIiIiIlKlQ1QHBjYZiNPt5IO1lbNTLBaDl87tQpf4cPYXO7j0oyVk5pealnFjej7zt2ZhMeD8nnFszt1cld3XeEQzZdOmTbhcLt59913Wr1/PK6+8wjvvvMMDDzxgdjQREZF6xzAMLuxdOTvlsyUpPrsRPVQ+ubugzQUAPLX4KcqcZSYnEhEREREREfnH37NTpm+fTlph5RsiA+xW3r+sJ4mRQaTkFHPJh0vYX2zOxII3f90GwBmdG1NipFHhqiDCP4L4kHhT8pjJI5opI0aMYOLEiQwfPpzmzZtz1llncddddzF16lSzo4mIiNRLo7s2wd9mYVNGAWtS88yOY6qbut1ETGAMu/J38dG6j8yOIyIiIiIiIlKlS0wX+sX1o8JdUTU7BSA2NIDJV/YhNtSfzXsLuGzi0jpfyntbZgEz16UDcNPJLVmXtQ6ofOOiYRh1mqU+sJkd4ETl5eURGRl5xGPKysooK/vnHaj5+fkAOBwOHI7aeeD9fd7aOn994St1gu/U6it1wtFr9YavgRnj39/n//ff3soT6gyyw4gODfludTqT/0ymfaMTm37rCbUeTYARwB3d7+D+P+7ngzUfMDxhOImhiQcc4w11HiuNgXoOWF2+Uquv1Am+U6svjH+gMbA2+UqdoFq9kcZAjYHV5St1gjm1XtXhKhalL+Lbrd9yadtLq2Z9xIXZ+ejS7lz00VJW797PBe8tYuL4HkQF+1X7msdS54s/b8bthmHtYmkeFcAnW9cA0K5BO496LNTUGGi4PXDtj23bttGjRw9efPFFrr766sMe9+ijj/LYY48ddPuUKVMICgqqzYgi4sGKi4sZN24ceXl5hIWFmR3nhGj8E4Dt+fD6eht2i5vHujsJtpudyDxut5uPiz5mW8U2WtpaclnwZT75LppjoTFQRHyVN4x/oDFQRE6MxkAR831c+DFbK7bS1d6Vc4LPOeBzaUXw1kYrhQ6DRoFurm/nJMK/dvNsy4M3NtgwcHNPZyeNg+GN/DfY69rLRcEX0c7ernYD1KFjHQNNbabcd999PPfcc0c8ZuPGjbRt27bq47S0NAYPHsyQIUP44IMPjnDPQ3ejExISyMrKqrX/GBwOB7Nnz2bYsGHY7d77qpWv1Am+U6uv1AlHrzU/P5/o6GiPfhJpxvgHvvM48pQ63W43o9/+kw3pBdw1rBXXDmp23OfwlFqPRUpBCuf9cB7lrnKeGfAMpzU9repz3lTn0WgM1HPA6vKVWn2lTvCdWn1h/AONgbXJV+oE1eqNNAZqDKwuX6kTzKt1Q84GLv7pYgwMvhz5JS0iWhzw+R37irh00jL25pcRG+rP2+O60jk+/ISvd6Q6K5wu/vPuYjakF3Bhr3geP6s9hY5Chnw9BJfbxc9jfiYmMOaEr13XamoMNHWZrzvvvJPx48cf8ZjmzZtX/XvPnj2cfPLJ9O/fn/fee++o5/f398ff/+AWnd1ur/UfhLq4Rn3gK3WC79TqK3XC4Wv1hvrNHP/q8jpm84Q6Lx/QjLu/XsOUJbu5bkhLbNYT2y7NE2o9mhaRLbiq81W8teotXl7xMoMTBxPqF3rAMd5Q57HSGFg79BjyPr5SJ/hOrd48/oHGwLrgK3WCavVGGgNrj68/hrxRXdfapWEXhjUdxuxds3ln3Tu8evKrB3y+TeMIvrm+P1dOWsbmvQWM+3ApL57bhTO7NK7WdQ9V51tzt7IhvYDQABt3ndYWu93OxsyNuNwumoQ0oXFY9a5pluqOgaZuQB8TE0Pbtm2P+MfPr3L9t7S0NIYMGUKPHj2YOHEiFoup0UVERDzCmV0aExXsx568UmZv2Gt2HNNd2fFKksKS2FeyjzdXvml2HBEREREREZEqN3a9EYthYU7KnKrN3v8tvkEQX1/fj1PaxlJW4eLmz1Zy/9S1FJdX1FiG5btyef3XrQA8fnYHokIqG5Qr960EoFtstxq7lqfxiI7E342UxMREXnzxRfbt20dGRgYZGRlmRxMREanXAuxWxvWp3Gx94h/J5oapB/ysfjzY90EAPt/8Oeuz15ucSERERERERKRSi4gWjGo+CoDXV7x+yGNCA+y8f2lPrhvcAsOAz5akMOr1Bazavb/a10/JLuba/y3H6XIzumtjxnSLr/rcykw1UzyimTJ79my2bdvGnDlziI+PJy4uruqPiIiIHNlFfZpisxgsSc5hXVqe2XFM1zeuLyObjcTldvHEoidwupxmRxIREREREREB4Pou12Oz2FiUvojF6YsPeYzVYnDf6W359Mo+NAoLYEdWEWPe+oO7vlrN3vzSE7pu2v4SLv1oMVmFZbSLC+OJ0R2rPlfhqmDNvjUAdI3tekLn9wYe0UwZP348brf7kH9ERETkyBqFB3B6p8o3IHy8MNncMPXE3b3uJtQeyvrs9Xy55Uuz44iIiIiIiIgAEB8azzmtzgHgxWUvHvENgP1bRvPTbQMZ260Jbjd8vTyVIS/8zos/b2ZfQdkxX3PxjmzOemMBydnFJEQG8vHlvQgN+GcfkS25WyipKCHUHkrLiJYnXpyH84hmioiIiFTP5QOSAPhu9R6yC4/9CZW3ig6M5pbutwCVU6ezS7JNTiQiIiIiIiJS6fqu1xNqD2VTziam75h+xGMjgvx4+fyuTLuhP90TIyhxOHnzt20MePZXrp+8nJlr08krdhx0P7fbTVoR3PX1Ws5/70+yi8ppHxfG59f0IzYs4IBj/17iq3NsZyyG77YUbGYHEBERkdrXLSGCLvHhrE7NY8riFG4e2srsSKY7t/W5fLvtW9Znr+e1Va/Rl75mRxIREREREREhMiCSazpfw0vLX+L1Fa8zvOlwguxBR7xPt8QGfHN9f35en8G783awMmU/P67L4Md1GVgMaB4TQsuYEIL8rRSWVrA5o4BdOTYgHYDzesbz6FkdCPI7uGVQtV9KjO/ulwKamSIiIuITDMPgipOaATBxYTIl5donxGqx8lDfhzAwmLFzBskVyWZHEhEREREREQFgXLtxxIfEs69kHxPXTzym+xiGwYiOcUy7YQA/3HIS1wxqTouYYFxu2JZZyE/rM5i6Io1ZG/ayK6cYq+FmWLtYvr9pAM+f0+WQjRS3283Kvdp8HjQzRURExGec0SmOF2dtZndOCV8sTWH8gGZmRzJdx+iO/Kf1f/h6y9dML57ONa5rsGM/+h1FREREREREapGf1Y87et7BHb/fwaR1k/hPq//QKLjRMd+/Q+NwOjQO54GR7dibX8rG9Hx2ZRdTVuHE32YloYE/ezcs4ZyzumK3H/734J15O8ksycTP4kfnmM41UZrH0swUERERH2GzWrhmUAsA3p+/E4fTZXKi+uHWbrcS7hfOXtdebUYvIiIiIiIi9capiafSPbY7pc5SXl/x+gmfp2FYAEPaxHJZ/ySuGdSCy/onMbBlNEHHMNViUfoiALo17EaALeAoR3s3NVNERER8yLk94okO8Sdtfwnfrdpjdpx6ISIggpu73gzAO2vfYV/xPpMTiYiIiIiIiFQu23VPr3sAmL5jOqv3ra7zDH/u+ROAfnH96vza9Y2aKSIiIj4kwG7lyr/2Tnln7nZcLrfJieqH0S1G08TahEJHIS8vf9nsOCIiIiIiIiIAdIjuwOiWowF4YtETVLgq6uzaDpeDpXuXAtCvsZopaqaIiIj4mIv7JhIaYGNbZiGzNuw1O069YDEsnBl4ZuVm9DtmsCxjmdmRRERERERERAC4vcfthPmFsTl3M1M2Tqmz667dt5YiRxER/hG0jWxbZ9etr9RMERER8TGhAXYu7dcUgLd/34bbrdkpAPG2eMa2HAvAU4ufwuFymJxIREREREREBCIDIrm9x+0A/HfVf8koyqiT6/6ZXrnEV5+4PlgMtRL0FRAREfFBlw9ohr/NwurUPBZuzzY7Tr1xY5cbifCPYNv+bXy+6XOz44iIiIiIiIgAMLbVWLrEdKG4opjnlz5fJ9dckLYAgL5xfevkevWdmikiIiI+KDrEnwt6JQDw1u/bTE5Tf0T4R3Bb99uAynf7aDN6ERERERERqQ8shoWH+z6M1bAye9ds5qTMqdXrpRemszZrLQYGg+MH1+q1PIWaKSIiIj7q6kHNsVkM/tiWzfJdOWbHqTfGtBpDp+hOFDmKeGn5S2bHEREREREREQGgTWQbxncYD8Djix4np7T2fpf/JeUXALrFdiMmKKbWruNJ1EwRERHxUfENgjinRzwAL/y8WXun/MViWHiw74MYGPyw4weWZiw1O5KIiIiIiIgIADd0vYGWES3JKc3hyT+frLXf5Wfvmg3A8KThtXJ+T6RmioiIiA+7eWgr/KwW/tyRwx/btHfK3zpEdeDc1ucC8PTip7UZvYiIiIiIiNQLflY/njrpKWyGjdm7ZvNT8k81fo29RXtZmbkSgKGJQ2v8/J5KzRQREREf1iQikHF9EgF4YZZmp/zbLd1vqdqMfsrGKWbHEREREREREQGgfVR7rul8DQBPLX6KjKKMGj3/30t8dY3pSqPgRjV6bk+mZoqIiIiPu/HklgTarazevZ9fNmaaHafeCPcP5/YetwPw1qq3yCzW10ZERERERETqh6s6X0X7qPbkleVxz7x7amxFBbfbzbSt0wA4Lem0Gjmnt1AzRURExMfFhPozfkASAC/N2ozLpdkpfxvdcjSdoztTXFHMi8teNDuOiIiIiIiICAB2i50XB71IiD2ElZkreWPlGzVy3vXZ69mcuxk/ix+jmo+qkXN6CzVTREREhGsHNSfU38amjAJmrE03O0698e/N6H/c+SNL0peYHUlEREREREQEgISwBB4f8DgAE9dNrJH9UyZvnAzAsKRhRAREVPt83kTNFBERESEiyI+rBzUH4JXZW6hwukxOVH+0j2rPeW3OAyrXotVm9CIiIiIiIlJfDGs6jEvaXwLAQwseYl3WuhM+157CPfy0s7Ihc2n7S2sknzdRM0VEREQAuOKkZkQG+7Ezq4hvVqSaHadeubnbzTTwb8COvB18tvEzs+OIiIiIiIiIVLmzx50MbDKQMmcZN/96M7sLdp/Qed5a9RZOt5O+cX1pH9W+hlN6PjVTREREBIAQfxs3DGkBwGu/bKWswmlyovoj3D+cW7vfCsDbq98muyTb5EQiIiIiIiIilawWK88Pep5WDVqRVZLFVT9fxZ7CPcd1jg3ZG/h++/dA5RsK5WBqpoiIiEiVi/s2pVFYAHvySvnfol1mx6lXRrccTbvIdhQ6Cnl95etmxxERERERERGpEuIXwrunvktSWBJ7ivZw5c9Xsiv/2H6vL3eW88jCR3Dj5vSk0+kc07mW03omNVNERESkSoDdyh3DWgPw+pyt5BaXm5yo/rBarNzf534Apm2dVq11aEVERERERERqWkxQDB8M/4CE0ARSC1MZ98M4/kz/84j3cbvdPL34aTblbCLcP5x7et9TR2k9j5opIiIicoD/9IinXVwY+aUVvPnbDrPj1CvdYrsxqvko3Lh5ZskzuNyu47q/0wm//w6ffVb5t1MrqYmIiIiIiEgNahjckE9O/4TO0Z3JL8/nutnX8dyS5ygoLzjo2Ap3Bc8ue5Zvtn6DgcEzJz1DdGC0Cak9g5opIiIicgCrxeChM9oBMGXJbjJLTA5Uz9ze43YCbYGs2beGGTtmHPP9pk6FpCQ4+WQYN67y76SkyttFREREREREakp0YDQfjfiIs1qchdPtZPLGyYyaNorXVrzGwj0LWb1vNd9s+4a3Ct7iq61fAfBQ34cYGD/Q5OT1m5opIiIicpABLaMZ2jaWCpeb73bp6cK/xQbFcm3nawF4ZfkrFJYXHvU+U6fCOedAauqBt6elVd6uhoqIiIiIiIjUJH+rP0+d9FTVPio5pTl8sPYDrp19LRfPvJinljxFpiuTML8wXh3yKue1Oc/syPWeXh0RERGRQ7p/ZDusFoN1uRb+2J5tdpx65ZL2l5AYmkhWSRbvrXnviMc6nXDrzRW43e6DPld5k5vbbqnQkl8iIiIiIiJS4/o36c/Us6bywuAXGJE0gmbhzYgLjqNbTDeGBQzj+7O+Z2jToWbH9AhqpoiIiMghtYwN4aLeCQA8PmMT5RXHtz+IN/Oz+nFv73sB+N/G/7Ezb+dhj50/H1L32ADjkJ93uw12p9mYP782koqIiIiIiIivs1vtjEgawQuDX+D70d8z65xZfDjsQwYHDCbML8zseB5DzRQRERE5rFtPaUGIzc2OrCI+Xphsdpx6ZVD8IAY2GUiFq4Lnlz5/2OPS04/tfMd6nIiIiIiIiIjUPY9pppx11lkkJiYSEBBAXFwcl1xyCXv27DE7loiIiFcLC7RzZtPKGSmvzdlKZn6pyYnql3t63YPNYmNB2gLmpc475DFxccd2rmM9TkRERERERETqnsc0U04++WS+/PJLNm/ezDfffMP27ds555xzzI4lIiLi9XrHuOkcH0ZhWQXP/rTJ7Dj1SlJ4Epe0vwSA55Y8R7mz/KBjBg6E+HiAg/dMATAMNwkJlceJiIiIiIiISP1kMzvAsbr99tur/t20aVPuu+8+Ro8ejcPhwG631+i1nE4nDofjhO7rcDiw2WyUlpbi9OKdZL21TrvdjtVqNTuGiEi9YjFgwhntOOfdxUxdkcZ5PRPo2zzK7Fj1xrWdr2X69umkFKTwvw3/48pOVx7weasVXn7FyXnnWgAX/34vi4ELMHj11crjRERERERERKR+8phmyr/l5OTw6aef0r9//yM2UsrKyigrK6v6OD8/H6hsBByqWeJ2u8nMzKw67kS43W4aNWpESkoKhnHojWa9gTfXGRYWRmxsbFVdfz9WTrTB5il8pU44eq3e8DU43vGvpvjK48hX6oR/amzfMIgLesXz+dJUHpi6lu9v7Ie/zWMmuB5Vdb6nfvhxS9dbmLBoAu+ueZfTE08nJijmgGPWhE8m4abVpH96PxW5DatubxIPL73k5Mwz3dTVw0ljYO3U54vjgrfX6it1gu/U6gvjH2gMrE2+UieoVm+kMVBjYHX5Sp3gO7X6Sp1Qc2Og4Xa7D73mRD1077338uabb1JcXEzfvn2ZMWMGUVGHf2fso48+ymOPPXbQ7VOmTCEoKOig20NDQ2nQoAHR0dH4+fl5XZNAjsztdlNeXk5WVha5ubkUFBSYHUlMUlxczLhx48jLyyMsLMzsOCfkeMc/kWNRXAFPr7JS4DAYmeDktHiPeQpR61xuF+8Xvs9u52662LtwbvC5VZ8rqijnmf2vgi2fpNLT6LNvNLm5ATRoUEr79tn1bkaKxkAR8VXeMP6BxkAROTEaA0XElx3rGGhqM+W+++7jueeeO+IxGzdupG3btgBkZWWRk5PDrl27eOyxxwgPD2fGjBmHbXocqhudkJBAVlbWQV8Up9PJjh07iImJOWKD5mjcbjcFBQWEhoZ6dTPGm+vMzs5m3759NG/eHKvVisPhYPbs2QwbNqzGl5SrT3ylTjh6rfn5+URHR3v0k8jjGf9qkq88jnylTji41hlr0rn9q7X42SzMuLEfzaKDzY5YI2rie7o+ez2X/nwpbtx8NOwjusZ0BeDK759lZeGXGBUN+O286YQFmPtLnMbA2qnJl8cFb+UrdYLv1OoL4x9oDKxNvlInqFZvpDFQY2B1+Uqd4Du1+kqdUHNjoKnLfN15552MHz/+iMc0b9686t/R0dFER0fTunVr2rVrR0JCAn/++Sf9+vU75H39/f3x9/c/6Ha73X7QF83pdGIYBiEhIVgsJ75sicvlAsAwjGqdp77z5jpDQkLIysoCOOBxcqjHjTfylTrh8LV6Q/3HM/7VBl95HPlKnfBPraO7JzBtdQbztuzjkembmHJ1H69qqlfne9q1UVfGtBrD1K1TeWH5C3x2xmds2pfGivxpGBa4oOV1RIWG13DiE6cxsHb44rjg7XylTvCdWr15/AONgXXBV+oE1eqNNAbWHl9/DHkjX6nVV+qE6o+BpjZTYmJiiImJOfqBh/D3i/n/7jbXBG96QUhOjB4DIiKHZxgGT57dkWGvzGXRjmy+WLqbC3onmh2r3ril2y3MSp7FxpyNfL/9e95ZPBvD4iDA2Zz7Bl5gdjwREREREREROUEeMaVg8eLFvPnmm6xatYpdu3bx66+/cuGFF9KiRYvDzkoRERGR2pEYFcRdw9sA8OQPG0nbX2JyovojKjCKaztfC8CEhRPY45wPwL297vG6mZwiIiIiIiIivsQjfqsPCgpi6tSpDB06lDZt2nDllVfSuXNn5s6de8ipewJDhgzhtttuAyApKYlXX3216nMZGRkMGzaM4OBgIiIiDntbdSQnJ2MYBqtWrar2uUREpP654qRm9GjagMKyCu77Zg0mbsFW74xrN474kISqj6MsXTin0wATE4mIiIiIiIhIdZm6zNex6tSpE7/++qvZMTzW0qVLCQ7+Z4PgV155hfT0dFatWkV4ePhhb6uOhIQE0tPTiY6Orva5RESk/rFaDF44pzOnvzaf+Vuz+GzJbsb10XJfAH5WP1rbLyCVFwA4pXkHkxOJiIiIiIiISHV5xMwUqZ6YmBiCgoKqPt6+fTs9evSgVatWxMbGHva26rBarTRq1AibzSP6dSIicgKax4Rw92mVy3099cMGducUm5yofli/J48f/oys+nhJ5h8mphERERERERGRmqBmihcoKiri0ksvJSQkhLi4OF566aUDPv/vZb6SkpL45ptv+OSTTzAMg/Hjxx/ytkMt07V//34Mw+D333+v+vjiiy8mJiaGwMBAWrVqxcSJE4FDL/M1d+5cevfujb+/P3Fxcdx3331UVFRUfX7IkCHccsst3HPPPURGRtKoUSMeffTR2viSiYhIDbl8QDN6Nm1AUbmTO75chdPl28t9OZwu7v5qDS57etVtu/J3sSpzlXmhRERERERERKTaNG3gMNxuNyUO53Hfz+VyUVLuxFZeccIbzQbarRiGcczH33333cydO5fvvvuO2NhYHnjgAVasWEHXrl0POnbp0qVceumlhIWF8dprrxEYGEh5eflBt+Xm5h71uk899RQbN27kxx9/JDo6mm3btlFScuhNiNPS0hg5ciTjx4/nk08+YdOmTVx99dUEBAQc0DD5+OOPueOOO1i8eDGLFi1i/PjxDBgwgGHDhh3z10NEROqO1WLw8nldGfn6fJYm5/Lf37Zxy9BWZscyzdu/b2dDej7hDVNw/ev255Y8x6dnfIrF0PtYRERERERERDyRmimHUeJw0n7Cz6Zce8PjpxHkd2zfmsLCQj788EMmT57M0KFDgcqGRHx8/CGPj4mJwd/fn8DAQBo1alR1+/+/7ViaKampqXTt2pWePXsClbNeDuett94iISGBN998E8MwaNu2LXv27OHee+9lwoQJVY2nzp0788gjjwDQqlUr3nzzTebMmaNmiohIPZYYFcQToztw+xereW3OVga0jKJH08ij39HLbM4o4I1ftwLQrlkW6/PgonYXMW3rNNZlr+OHHT9wZoszTU4pIiIiIiIiIidCb4/0cNu3b6e8vJw+ffpU3RYZGUmbNm1q/dpXXHEFX3zxBV27duWee+5h4cKFhz1248aN9OvX74AZNwMGDKCwsJDU1NSq2zp37nzA/eLi4sjMzKz58CIiUqPGdItndNfGOF1ubv18FfmlDrMj1akKp4u7v16Nw+lmaLsY0ss2ADC86XCu7nw1AK+ueJVih/aVEREREREREfFEmplyGIF2KxseP+247+dyuSjILyA0LLRay3yZ7e/sbvc/a987HAe+MDZs2DB27tzJTz/9xOzZsxk6dCg33ngjL7744glf1263H/CxYRi4XK7DHC0iIvXJ46M7smxXLv/X3p2HRVnv/x9/zQy7DioKogmCprhLikuZqSfN8nw7me3qN0l+djJL1JNlZmnZUetYafbNVnMtO5rlyWyxBTUxzX1DDRPBFc0FFIGBmd8fHClCWWcYmfv5uC6uS+65l/ebppdevOe+P4fPXNQzn+7S6/dHl+uxldXZGz8ka8fhcwr089Lw3oF66NvT8jH7qE29Nmpdr7WW7l+qI+ePaO7uuXo0+lF3lwsAAAAAAMqJO1OuwGQyKcDHq0Jf/j6WCh8b4ONVrl88NW3aVN7e3tqwYUPhtjNnzmj//v2V6j84OFiSdOzY7wvo/nEx+T/uN2TIEC1cuFAzZszQO++8c9nztWzZUuvXry8ynFm3bp2sVusVH0kGAKheAv28NfP+62Qxm/T59qOal5ji7pKqxOr9JzXzu4LHez1/R2ulXNglSWob3FY+Fh/5Wnw1uuNoSdIHuz7Q8QvH3VYrAAAAAACoGIYp1VzNmjUVFxensWPH6vvvv9euXbsUGxtb4btiLvH391fXrl01bdo0JSUlafXq1ZowYUKRfaZMmaLly5crOTlZu3fv1ooVK9SyZcvLnu/RRx9VWlqaHn/8ce3du1fLly/XxIkTNWbMmErXCgC4enRsXEdP39ZCkvTiF0nalHLazRW51pGzFzVq8VY5HNIDncN153WNdODsAUlS67qtC/e7pfEt6hDSQdn52Zq5Zaa7ygUAAAAAABXEb7E9wL/+9S91795dt99+u3r37q0bb7xRHTt2rPR558yZo7y8PHXs2FGjRo3Siy++WOR1Hx8fPfPMM2rXrp1uuukmWSwWLV68+LLnuuaaa7Ry5Upt3LhR7du31yOPPKK4uLhiAxoAQPUXd2Ok/qddA+XZHXp00RalZ2a7uySXyM2za8SiLTqTZVObawI18fZWkqR8R74kydfiW7ivyWTSk52flEkmrfh1hXae3OmWmgEAAAAAQMWwZooHqFmzphYsWKAFCxYUbhs7dmzhn1NSUors/9lnnxU7x+W2tWzZstii8pce02W32/XEE0/ohRdeuOydJREREUUe6SVJPXr00MaNG6/YR0JCQpnqAgBc3Uwmk166q532n8jU/hPn9diirVo0rIu8LZ71GY5/frFH29LOqpa/t2YP6ii//655ZncUrPVlNhXtt3Xd1rq96e36z4H/6KWfX9KC2xYYZk0ZAAAAAACqO8/6rQYAALgq1PD10tv/GyOrr5c2ppzWc8t3FxuyV2fLtx3RvPWHJEmv3ddeYUEBxfb58zBFkuI7xMvfy1/bT27XqkOrXF4nAAAAAABwDoYpAADAJSLr1dCM+6NlNkkfbUzV7NUH3F2SU+w9nqGnlxU8puuxXtfqLy3qF3n90p0pl7vrJCQgRENaD5EkzdwyU7Z8m4urBQAAAAAAzsAwBQAAuMzNLetr4u0FC7G//NU+Ld92xM0VVU7a6Sw9+P5GZeXmq9u1dTW6T/Ni+xQ+5usK/8x6qPVDqutXV6mZqfr3/n+7tF4AAAAAAOAcDFMAAIBLDbkhQv/vxkhJ0tglO/TTr7+5uaKKSc/I1qD3Nig9M0ctQq16c2BHWczF7z5xqOBxZpd7zJckBXgH6NHoRyVJb21/S5m5ma4rGgAAAAAAOAXDFAAA4HLj+7XUbW1ClZtv18PzN2nXkXPuLqlczmXZ9OCcjUo9naXwoADNH9pZtQK8L7tvSY/5umRAswGKrBWpszlnNWfXHJfUDAAAAAAAnIdhCgAAcDmz2aTX7otWTOM6ysjO0+D3N2j30eoxUMnKzdNDczdq7/FMhVh9tTCui0IC/a64v8NRcGeKSVcepniZvTS6w2hJ0oI9C3T8wnHnFg0AAAAAAJyKYQoAAKgSft4WffBQJ10XXltns2wa9N4G7Tma4e6ySpRty9cjC7doS+pZBfp5aX5cZ4XXDSjxGD+vgkHLedv5EvfrGdZTHUI6KCc/R7O2znJazQAAAAAAwPkYpgAAgCpj9fPWvKGd1T7s0kDlJyUduzoHKmezcvXg+xu1Zv9J+Xtb9MFDndUiNLDU4yJrFawPc+DsgRL3M5lMeiLmCUnS5wc+177T+ypfNAAAAAAAcAmGKQAAoEoF+nlr/tDOat+ols5k2XT/Oz9ddYvSHzl7UXe/tV4bU07L6uelObGd1LFxnTIde23tayVJyWeTS923bXBb3Rpxqxxy6NXNr1aqZgAAAAAA4DoMU5wtP19eP/4offSRlJAg5ee7u6IKmTRpkqKjo91dBgDAQ9Xy99b8uC7qEF5b5y7a9L/vb9CnWw+7uyxJ0p6jGbrz/9YpOf28QgP9tOSR63V907plPr5p7aaSpMOZh3Ux72Kp+4/sMFJeZi8lHk1U4pHECtcNAAAAAABch2GKMy1bJlOTJqp5++0yDx4s9eolRURIy5a5uzIAAK46tfy99eGwrurXNlS2fIdGf7xdr3/3S+EC7u6wNvmU7n17vdIzc9S8fk0te/SGMj3a64/q+tVVbd/acsihg+cOlrp/mDVM90fdL0l6dfOryrdXzw9iAAAAAADgyRimOMuyZdLdd0uH//Sp2iNHCrYzUAEAoBg/b4veeKCD/n5TE0nSq6v267GPtupclq1K68jJs2v5IbPi5m/R+Zw8dYkM0pJHblDD2v7lPpfJZCp81Fdp66Zc8vd2f5fV26p9Z/Zpxa8ryn1NAAAAAADgWgxTnCE/X4qPlxwOmf782qVP144a5bJHftntdk2dOlWRkZHy9/dX+/bttXTpUklSQkKCTCaTvvvuO8XExCggIEA33HCD9u0rusjttGnTVL9+fVmtVsXFxSk7O9sltQIA8Gdms0lP92upF/u3kcVs0hc7jum2mWu0/kDVrKPyy4lM3fP2Bn1/1CyHQ7q/U5jmx3VWLX/vCp/z0qO+fjn7S5n2r+1XW/+v3f+TJM3aOkvZefw9DAAAAADA1YRhijOsXVv8jpQ/cjiktLSC/Vxg6tSpmj9/vt566y3t3r1bo0eP1uDBg7V69erCfZ555hm98sor2rRpk7y8vDR06NDC1/79739r0qRJmjJlijZt2qQGDRrozTffdEmtAABcyeCujfXJ8BsUUTdAR89la+B7P2nql0nKyXPNhxHy8u36YN1B/c+sH5V0PFM1vBx684FoTburnXy9LJU6d3nvTJGkQS0HqUGNBjqRdUILkxZW6voAAAAAAMC5GKY4w7Fjzt2vHHJycjRlyhTNmTNHffv2VZMmTRQbG6vBgwfr7bffLtzvn//8p3r06KFWrVpp3LhxSkxMLLz7ZMaMGYqLi1NcXJyioqL04osvqlWrVk6vFQCA0kSH1dYXI7vr/k5hcjikt1f/qr9MX61PNh9Wvt05a6nk2x1avu2IbnltjZ7/fI9y8uzqfm1dPdU+X31ahTjlGpeGKUm/JZV5DRhfi68ev+5xSdKcnXN0LuecU2oBAAAAAACVxzDFGRo0cO5+5ZCcnKysrCz16dNHNWvWLPyaP3++Dhz4/dOw7dq1+0MZBXWkp6dLkpKSktSlS5ci573++uudXisAAGVRw9dL0+5qp7cGd1T9QF8dOXtR/1iyXf1mrtWqPSdkr+BQJS/frs+3H1XfGWsUv3ibfj11QXUCvPXCHa31/oMdVMvHeT20DW4rfy9/nbx4UntO7ynzcX9t8lc1q9NMmbZMzdk1x3kFAQAAAACASvFydwEeoXt3qVGjgsXmL/fpU5Op4PXu3Z1+6fPnz0uSvvjiC11zzTVFXvP19S0cqHh7//7cd5OpYGUXu93u9HoAAHCWW9uEqmdUsOYmpujNH5K170Smhs3fpNBAP/Vr20B/bddA14XVltlcbMWyQtm2fP34yyl9tfu4vk06obP/Xdi+lr+3Hr6piYbcEKGavl6y2Zy74L2vxVfdGnbTt6nf6ofUH9S6busyHWc2mTXyupF6/PvH9WHShxrccrCCA4KdWhsAAAAAACg/hinOYLFIM2dKd98th8kk0x8HKv8dXGjGjIL9nKxVq1by9fVVamqqevToUez1P96dciUtW7bUhg0b9OCDDxZu++mnn5xaJwAAFeHnbdEjPZrqgU7hmr36gBb9dEjHM7I1Z91BzVl3UCFWX10bUlON6virUZ0A1anho2NnLyrtzEWlnc7S/hOZysr9fc2VoBo+ir0hQrHdIhToV/EF5suiV3gvfZv6rRLSEvTYdY+V+bgejXqofXB7bT+5XW/veFsTuk5wXZEAAAAAAKBMGKY4y4AB0tKlUnx80cXoGzUqGKQMGOCSy1qtVj3xxBMaPXq07Ha7brzxRp07d07r1q1TYGCgGjduXOo54uPjFRsbq5iYGHXr1k2LFi3S7t271aRJE5fUDABAedUK8Na421poVO9mWvvLKX2x46hW7Tmh9MwcpWfmlHhsg1p+6ts6VH1bh6pTRB15WarmKafdr+kus8msfWf26cj5I7qm5jWlH6SCO0jjO8Rr6NdD9cn+TzSk9RCFWcNcXC0AAAAAAChJtRum5OTkqEuXLtq+fbu2bt2q6Ohod5f0uwED5Lj9dl34+msFnDsn8zXXFDzaywV3pPzR5MmTFRwcrKlTp+rXX39V7dq11aFDB40fP75Mj/K67777dODAAT355JPKzs7WXXfdpeHDh+vrr792ad0AAJSXn7dFfVrVV59W9ZVty9fOI+eUdjpLh/97J8qZLJsa1vZTozr+CqsToIh6NdQi1Fr4iMuqVMevjq4LuU6bT2xWQlqCBrUcVOZjO4V20g0Nb1Di0US9ue1NTe0+1XWFAgAAAACAUlW7YcqTTz6phg0bavv27e4u5fIsFuXdeKMUGCiZq+aTryaTSfHx8YqPj7/s644/reMSHR1dbNv48eM1fvz4Itteeukl5xYKAIAT+Xlb1CkiSJ0igtxdyhX1CutVoWGKJI3sMFKJRxP1xa9f6KE2D6l5neauKRIAAAAAAJSqan7b7yRffvmlvvnmG02fPt3dpQAAAJSqZ1hPSdKm45uUkZtRrmNb122tPo37yCGHZm2d5YLqAAAAAABAWVWbO1NOnDihYcOG6bPPPlNAQECZjsnJyVFOzu/PUc/IKPglhs1mk81mK7KvzWaTw+GQ3W4v06OxruTSHR+XzuWpPLlPu90uh8Mhm80mi8VS+F7583vG0xilT6n0Xj3hZ1Ce/HMmo7yPjNKnZJxeXdVnQ/+GalqrqQ6cO6AVySt0T7N7ynX8I20e0Xep3ykhLUGbj21Wu3rtKl0TGeia/ozy/4pknF6N0qdknF6NkH8SGehKRulToldPRAaSgZVllD4l4/RqlD4l52WgyfHn5z1dhRwOh/r166du3bppwoQJSklJUWRkZKlrpkyaNEnPP/98se0ffvhhsYGMl5eXQkNDFRYWJh8fH2e3gGokNzdXaWlpOn78uPLy8txdDtwgKytLAwcO1Llz5xQYGOjuciqkPPkHwLUScxK18uJKhZhD9Lj18XKv37Isa5m25G5RpFekhtYY6vL1X8hAAEblCfknkYEAKoYMBGBkZc1Atw5Txo0bV+q6HElJSfrmm2/073//W6tXr5bFYinzMOVy0+iwsDCdOnWq2A8lOztbaWlpioiIkJ+fX4V7cjgcyszMlNXqnsVuq4on95mdna2UlBSFhYXJz89PNptNq1atUp8+feTt7e3u8lzGKH1KpfeakZGhevXqVet/RJYn/5zJKO8jo/QpGadXV/aZmZupvp/2VXZ+tt69+V11rN+xXMcfu3BM/T/vL5vdptl/ma0uoV0qVQ8Z6JqejPL/imScXo3Sp2ScXo2QfxIZ6EpG6VOiV09EBpKBlWWUPiXj9GqUPiXnZaBbH/P1j3/8Q7GxsSXu06RJE33//fdav369fH19i7wWExOjQYMGad68eZc91tfXt9gxkuTt7V3sh5afny+TySSz2SxzJRaOv/TIq0vn8lSe3KfZbJbJZCr2Prnc+8YTGaVP6cq9ekL/5ck/VzDK+8gofUrG6dUVfQZ5B+l/mv6Plu5fqiXJS9S1UddyHR9eO1z3Rt2rRUmL9M6ud9StUTenfJCBDHQNo/y/IhmnV6P0KRmnV0/OP4kMrApG6VOiV09EBrqO0d9DnsgovRqlT6nyGejWYUpwcLCCg4NL3e/111/Xiy++WPj90aNH1bdvX3388cfq0qVyn84EAABwtfuj7tfS/Uv1fer3Ss9KV0hASLmOH9pmqJbsW6Kt6Vu14fgGdW1QvoEMAAAAAAConGpxS0F4eLjatGlT+NW8eXNJUtOmTdWoUSM3VwcAAFCyqKAoXRdynfIcefpk/yflPj4kIET3RBUsXj9722xVgyXvAAAAAADwKNVimAIAAFDd3R91vyRpyf4lstlt5T5+aJuh8jH7aEv6Fm04vsHZ5QEAAAAAgBJUy2FKRESEHA5HiYvPAwAAXE36NO6jev71dPLiSS1PXl7u40MCQnR387slcXcKAAAAAABVrVoOU/C7nj17atSoUW6tITY2Vv379y/8vipqSkhIkMlk0tmzZ116HQAAnMXb4q2hbYZKkt7e8bZy83PLfY4/3p2y8fhGZ5cIAAAAAACugGEKnG7ZsmWaPHmyu8sAAOCqc2/UvQrxD9HxC8e17Jdl5T6+fo36uqv5XZKk2du5OwUAAAAAgKrCMAVOFxQUJKvV6u4ySpWbW/5PBAMAUBm+Fl8NazdMkvTujneVnZdd7nPEtYmTt9lbm09s1s/Hf3Z2iQAAAAAA4DIYplyBw+FQli2rQl8X8y5W+NgsW1a5P2Wal5enxx57TLVq1VK9evX07LPPFjnHggULFBMTI6vVqtDQUA0cOFDp6emFr585c0aDBg1ScHCw/P391axZM33wwQeFr6elpenee+9V7dq1FRQUpDvuuEMpKSlXrOfPj/mKiIjQlClTNHToUFmtVoWHh+udd94pckx5r3E5c+fOVXh4uAICAnTnnXfqlVdeUe3atQtfnzRpkqKjo/Xee+8pMjJSfn5+5To/AADOMKDZADWo0UDpF9O1ZP+Sch9fv0b9wrVT3tz+prPLAwAAAAAAl+Hl7gKuVhfzLqrLh13ccu0NAzcowDugzPvPmzdPcXFx2rhxozZt2qSHH35Y4eHhGjas4JOvNptNkydPVlRUlNLT0zVmzBjFxsZq5cqVkqRnn31We/bs0Zdffql69eopOTlZFy9eLDy2b9++uv7667V27Vp5eXnpxRdfVL9+/bRmzZoy1/jKK69o8uTJGj9+vJYuXarhw4erR48eioqKuuI1br31Vu3YsUM+Pj6l/8w2bFBcXJymTp2q/v3766uvvtLEiROL7ZecnKxPPvlEy5Ytk8ViKXP9AAA4i4/FRw+3e1jPr39e7+18T3c1u6tcf+9LBWunLN2/tPDulE6hnVxULQAAAAAAkBimeISwsDC99tprMplMioqK0s6dO/Xaa68VDlOGDh1auG+TJk30+uuvq1OnTjp//rxq1qyp1NRUXXfddYqJiZFUcCfJJR9//LHsdrvee+89mUwmSdIHH3yg2rVr68cffyyy8HxJ+vXrp0cffVSS9NRTT+m1117TDz/8oKioqBKvkZCQoFtuuaXU88+cOVO33nqrnnzySUlS8+bNlZiYqK+++qrIfrm5uZo/f76Cg4PLVDcAAK5wx7V36P2d7+vw+cOat3uehkcPL9fxoTVCNaDZAH2872O9v/N9hikAAAAAALgYw5Qr8Pfy14aBG8p9nN1uV2ZmpqxWq8zmij1Fzd/Lv1z7d+3atXAIIUnXX3+9XnnlFeXn58tisWjz5s2aNGmStm/frjNnzshut0uSUlNT1apVKw0fPlx33XWXtmzZoltuuUX9+/fXDTfcIEnavn27kpOTi62Bkp2drYMHD5a5xnbt2hX+2WQyKTQ0tPBRYyVd48CBA2U6f1JSku68884i266//vpiw5TGjRszSAEAuJ232VvxHeM1dvVYvb/rff3t2r/pmprXlOscsa1jtXT/Uq07uk57T+9Vi6AWLqoWAAAAAAAwTLkCk8lU7kduSAXDlDyvPAV4B1R4mOJMFy5cUN++fdW3b18tWrRIwcHBSk1NVd++fQsXYL/tttt06NAhrVy5UqtWrdLNN9+sESNGaPr06Tp//rw6duyoRYsWFTmv3W6Xr69vmevw9vYu8r3JZCoc6lzpGpKcPvioUaOGU88HAEBF9W3cV0tCl2jj8Y3618//0oxeM8p1fCNrI90ScYu+PPil5uyco5d7vOyaQgEAAAAAAAvQe4ING4reQfPTTz+pWbNmslgs2rt3r3777TdNmzZN3bt3V4sWLYosPn9JcHCwhgwZooULF2rGjBmFC8R36NBBv/zyi0JCQnTttdcW+apVq5ZT6nfGNVq2bHnZnwMAAFcrk8mkcZ3HyWKy6LvU75R4JLHc5xjapuBRnl8f+lppmWnOLhEAAAAAAPwXwxQPkJqaqjFjxmjfvn366KOPNGvWLMXHx0uSwsPD5ePjo1mzZunXX3/Vf/7zH02ePLnI8c8995yWL1+u5ORk7d69WytWrFDLli0lSYMGDVK9evV0xx13aO3atTp48KASEhIUHx+vI0eOOKX+K11j5MiROnz4cJnOMXLkSH311VeaPn26fvnlF73xxhvFHvEFAMDVplmdZnqgxQOSpKkbp8qWbyvX8S2CWqjbNd1kd9g1b/c8V5QIAAAAAADEMMUjPPjgg7p48aI6d+6sESNGKD4+Xg8//LCkgjtO5s6dqyVLlqhVq1aaNm2apk+fXuR4Hx8fPf3002rXrp1uuukmWSwWLV68WJIUEBCgNWvWKDw8XAMGDFDLli0VFxen7OzsYmucVFRJ1wgMDCzTObp27ap3331XM2fOVPv27fXNN99owoQJTqkPAABXGh49XEF+QUrJSNH8PfPLfXxcmzhJ0mfJn+nUxVPOLg8AAAAAAIg1U6q9hISEwj/Pnj37svs88MADeuCBB4psczgchX+eMGFCiYOH0NBQzZtX9NOudrtdGRkZkqS5c+desSZJSklJKXbObdu2lXqNkvTs2bNID5I0dOhQDR06tPD7P9c1adIkTZo0qczXAACgKgT6BGp0x9F6dt2zmr19tv4S/hdF1oos8/Ex9WPUtl5b7Ty1U0v2LdHw6OEurBYAAAAAAGPizhQAAAA3u6PpHbqh4Q3Kyc/Rs+ueVb49v8zHmkwm/W+r/5Ukfbzv43I/KgwAAAAAAJSOYQoAAICbmUwmTbp+kmp419D2k9u1MGlhuY7v3bi3QvxD9Fv2b/r60NcuqhIAAAAAAONimAKPFRsbq7Nnz7q7DAAAyqRBzQYaGzNWkjRr6ywdPHewzMd6m711X4v7JEmL9iwq9ihMAAAAAABQOQxTAAAArhIDmg0ofNzXhB8nyGYv+yO77m5+t3zMPtr12y7tOLXDhVUCAAAAAGA8DFMAAACuEpce92X1tmrHqR16Y+sbZT42yC9I/Zr0kyQtSlrkqhIBAAAAADAkhikAAABXkQY1G2jSDZMkSXN2zdG6I+vKfOwDLR6QJH176FudyznnivIAAAAAADAkhikAAABXmVsibtF9UQVroIz/cbxOZp0s03Gt6rZSVJ0o2ew2rTy40pUlAgAAAABgKAxTAAAArkJjO41V8zrNdTr7tMatHac8e16Zjruz2Z2SpE9/+dSV5QEAAAAAYCgMUwAAAK5CvhZf/avHv+Tv5a+Nxzfq1c2vlum4fpH95GX2UtLpJO07vc/FVQIAAAAAYAwMU5wsP1/68UcvffSRlJBQ8D0AAEBFNKnVRP+88Z+SpAV7Fmh58vJSj6njV0e9wnpJkj5L/syV5QEAAAAAYBgMU5xo2TKpSROTbr+9pgYPNqtXLykiomC7O8TGxqp///5OO1/Pnj01atQop50PAACUrk/jPnqk/SOSpOfXP6/tJ7eXeswdTe+QJH1z6Bs5HA6X1gcAAAAAgBEwTHGSZcuku++WDh8uuv3IkYLt7hqoOENubq67SwAAwNCGtx+uv4T9RTa7TSO/H6m0jLQS9+/asKv8vfyVnpWupNNJVVQlAAAAAACei2GKE+TnS/HxUsEHP01FXrv0YdBRo1z3yK+lS5eqbdu28vf3V926ddW7d2+NHTtW8+bN0/Lly2UymWQymZSQkCBJeuqpp9S8eXMFBASoSZMmevbZZ2Wz2QrPN2nSJEVHR+u9995TZGSk/Pz8FBsbq9WrV2vmzJkymUyyWCxKTU11TUMAAKAIs8msKd2nqEVQC53OPq2/f/t3/Xbxtyvu72vx1fUNrpckrU5bXVVlAgAAAADgsbzcXYAnWLu2+B0pf+RwSGlpBfv17Oncax87dkwPPPCAXn75Zd15553KzMzU2rVr9eCDDyo1NVUZGRn64IMPJElBQUGSJKvVqrlz56phw4bauXOnhg0bJqvVqieffLLwvMnJyfrkk0+0bNkyWSwWNW7cWPv371ebNm30wgsvyG63y9fX17nNAACAK6rhXUOze8/W4JWDlZaZpke/e1Rz+s5RDe8al92/Z1hPfZ/2vX5I+0HDo4dXcbUAAAAAAHgWhilOcOyYc/cr37WPKS8vTwMGDFDjxo0lSW3btpUk+fv7KycnR6GhoUWOmTBhQuGfIyIi9MQTT2jx4sVFhim5ubmaP3++goODC7f5+PgoICBAoaGhstvtysjIcH5DAADgiur519Nbvd/Sg18+qD2/7dHwb4drdu/Zlx2odG/UXZKUdDpJp7NPK8gvqKrLBQAAAADAY/CYLydo0MC5+5VH+/btdfPNN6tt27a655579O677+rMmTMlHvPxxx+rW7duCg0NVc2aNTVhwoRij+xq3LhxkUEKAAC4OkTUitDsPrNl9bFqa/pWPbLqEZ3PPV9sv3r+9RQRGCFJ2n1qdxVXCQAAAACAZ2GY4gTdu0uNGkkm0+VfN5mksLCC/ZzNYrFo1apV+vLLL9WqVSvNmjVLUVFROnjw4GX3X79+vQYNGqR+/fppxYoV2rp1q5555plii8zXqHH5R4YAAAD3a123td695V0F+gRq28ltGvr1UJ3MOllsv5ZBLSVJv5z9papLBAAAAADAo1SbYUpEREThQuqXvqZNm+busiRJFos0c2bBn00mR5HXLg1YZswo2M8VTCaTunXrpueff15bt26Vj4+PPv30U/n4+Cj/T6veJyYmqnHjxnrmmWcUExOjZs2a6dChQ2W6zuXOBwAA3OPSQCXIL0hJp5M0cOVA7Tu9r8g+DWoW3BZ74sIJd5QIAAAAAIDHqDbDFEl64YUXdOzYscKvxx9/3N0lFRowQFq6VLrmmqLbGzUq2D5ggGuuu2HDBk2ZMkWbNm1Samqqli1bppMnT6ply5aKiIjQjh07tG/fPp06dUo2m03NmjVTamqqFi9erAMHDuj111/Xp59+WqZrRUREaMOGDUpJSdGpU6dkt9td0xQAACiTVnVbaWG/hYoIjNDxC8c18IuBWrhnofLtBR9+8DIXLI/nkKOk0wAAAAAAgFJUq2GK1WpVaGho4dfV9iiqAQOkX3916PPPz2vhQrt++EE6eNB1gxRJCgwM1Jo1a9SvXz81b95cEyZM0CuvvKLbbrtNw4YNU1RUlGJiYhQcHKx169bpb3/7m0aPHq3HHntM0dHRSkxM1LPPPlumaz3xxBOyWCxq1aqV6tevr8OHD7uuMQAAUCZh1jAt7LdQNzW6Sbn2XL3080u66z93ad7ueVp1aJUkKbRGqJurBAAAAACgevNydwHlMW3aNE2ePFnh4eEaOHCgRo8eLS+vK7eQk5OjnJycwu8zMjIkSTabTTabrci+NptNDodDdru9UndcmM0O3XhjnqxWh0ymgvO48gaOqKgorVy5sth2u92uunXr6quvviq2fdq0acUekTZy5MjCvp977jk999xzxX4O1157rdatWydJcjgcyszMLPyZeRK73S6HwyGbzSaLxVL4Xvnze8bTGKVPqfRePeFnUJ78cyajvI+M0qdknF6re58B5gC91v01LU1eqje2vaED5w5o+qbpkiQ/i59ubnRzsR7JQOeq7u+h8jBKr0bpUzJOr0bIP4kMdCWj9CnRqyciA8nAyjJKn5JxejVKn5LzMtDkcDiqxXMfXn31VXXo0EFBQUFKTEzU008/rYceekivvvrqFY+ZNGmSnn/++WLbP/zwQwUEBBTZ5uXlpdDQUIWFhcnHx8fp9aP6yM3NVVpamo4fP668vDx3lwM3yMrK0sCBA3Xu3DkFBga6u5wKKU/+AfAsF+0XtTl3s1LyUuRt8lY3325q5NWozMeTgQCMyhPyTyIDAVQMGQjAyMqagW4dpowbN04vvfRSifskJSWpRYsWxbbPmTNHf//733X+/Hn5+vpe9tjLTaPDwsJ06tSpYj+U7OxspaWlKSIiQn5+fhXopsClOzasVqtMl1af90Ce3Gd2drZSUlIUFhYmPz8/2Ww2rVq1Sn369JG3t7e7y3MZo/Qpld5rRkaG6tWrV63/EVme/HMmo7yPjNKnZJxejdKnRAa6qifeQ57HKH1KxunVCPknkYGuZJQ+JXr1RGQgGVhZRulTMk6vRulTcl4GuvUxX//4xz8UGxtb4j5NmjS57PYuXbooLy9PKSkpioqKuuw+vr6+lx20eHt7F/uh5efny2QyyWw2y2yu+FIylx55delcnsqT+zSbzTKZTMXeJ5d733gio/QpXblXT+i/PPnnCkZ5HxmlT8k4vRqlT4kMdBXeQ57HKH1KxunVk/NPIgOrglH6lOjVE5GBrmP095AnMkqvRulTqnwGunWYEhwcrODg4Aodu23bNpnNZoWEhDi5KgAAAAAAAAAAgN9ViwXo169frw0bNqhXr16yWq1av369Ro8ercGDB6tOnTpOvVY1WUIGLsR7AAAAAAAAAADwR9VimOLr66vFixdr0qRJysnJUWRkpEaPHq0xY8Y47RqXbuXJysqSv7+/086L6icrK0uS59ziCgAAAAAAAAConGoxTOnQoYN++uknl17DYrGodu3aSk9PlyQFBARUaGF1u92u3NxcZWdne9xaIn/kiX06HA5lZWUpPT1dtWvXlsVicXdJAAAAAAAAAICrQLUYplSV0NBQSSocqFSEw+HQxYsX5e/vX6FhTHXhyX3Wrl278L0AAAAAAAAAAADDlD8wmUxq0KCBQkJCZLPZKnQOm82mNWvW6KabbvLox0R5ap/e3t7ckQIAAAAAAAAAKIJhymVYLJYK/0LdYrEoLy9Pfn5+HjVk+DOj9AkAAAAAAAAAgGcsdgEAAAAAAAAAAOAiDFMAAAAAAAAAAABKwDAFAAAAAAAAAACgBIZaM8XhcEiSMjIyXHYNm82mrKwsZWRkePRaIkbpUzJOr0bpUyq910sZcSkzPEFV5J9knPeRUfqUjNOrUfqUyEBX4T3keYzSp2ScXo2YfxIZ6ExG6VOiV09EBpKBlWWUPiXj9GqUPiXnZaChhimZmZmSpLCwMDdXAqA6yMzMVK1atdxdhlOQfwDKiwwEYFSelH8SGQigfMhAAEZWWgaaHJ42ci6B3W7X0aNHZbVaZTKZXHKNjIwMhYWFKS0tTYGBgS65xtXAKH1KxunVKH1KpffqcDiUmZmphg0bymz2jKchVkX+ScZ5HxmlT8k4vRqlT4kM5N+AlWeUXo3Sp2ScXo2YfxIZ6ExG6VOiV09EBpKBlWWUPiXj9GqUPiXnZaCh7kwxm81q1KhRlVwrMDDQ49+EknH6lIzTq1H6lEru1ZM+iSNVbf5JxnkfGaVPyTi9GqVPiQx0Fd5DnscofUrG6dVI+SeRga5glD4levVEZKDr8B7yPEbp1Sh9SpXPQM8ZNQMAAAAAAAAAALgAwxQAAAAAAAAAAIASMExxMl9fX02cOFG+vr7uLsWljNKnZJxejdKnZKxeq5pRfrZG6VMyTq9G6VMyVq9VyUg/V6P0apQ+JeP0apQ+3cEoP1uj9CnRqycySp/uYJSfrVH6lIzTq1H6lJzXq6EWoAcAAAAAAAAAACgv7kwBAAAAAAAAAAAoAcMUAAAAAAAAAACAEjBMAQAAAAAAAAAAKAHDFAAAAAAAAAAAgBIwTKkCOTk5io6Olslk0rZt29xdjtOlpKQoLi5OkZGR8vf3V9OmTTVx4kTl5ua6u7RK+7//+z9FRETIz89PXbp00caNG91dktNNnTpVnTp1ktVqVUhIiPr37699+/a5uyyXmzZtmkwmk0aNGuXuUjyeJ2egJ+efRAZ6MjKw6pCB1ZenZ6BR808iA6uKJ+ef5NkZ6On5Jxk3A8m/quPJGejJ+SeRgZ7MGRnIMKUKPPnkk2rYsKG7y3CZvXv3ym636+2339bu3bv12muv6a233tL48ePdXVqlfPzxxxozZowmTpyoLVu2qH379urbt6/S09PdXZpTrV69WiNGjNBPP/2kVatWyWaz6ZZbbtGFCxfcXZrL/Pzzz3r77bfVrl07d5diCJ6cgZ6afxIZSAbCWcjA6skIGWjE/JPIwKrkyfkneW4GGiH/JGNmIPlXtTw5Az01/yQykAwsAwdcauXKlY4WLVo4du/e7ZDk2Lp1q7tLqhIvv/yyIzIy0t1lVErnzp0dI0aMKPw+Pz/f0bBhQ8fUqVPdWJXrpaenOyQ5Vq9e7e5SXCIzM9PRrFkzx6pVqxw9evRwxMfHu7skj2bEDPSE/HM4yEAyEM5ABlZfRsxAT88/h4MMrEpGzD+HwzMy0Ij553B4fgaSf1XLiBnoCfnncJCBZGDpuDPFhU6cOKFhw4ZpwYIFCggIcHc5VercuXMKCgpydxkVlpubq82bN6t3796F28xms3r37q3169e7sTLXO3funCRV6/9+JRkxYoT++te/FvlvC9cwagZW9/yTyECJDETlkYHVl1Ez0NPzTyIDq4pR80+q/hlo1PyTPD8Dyb+qY9QMrO75J5GBEhlYFl5OqAeX4XA4FBsbq0ceeUQxMTFKSUlxd0lVJjk5WbNmzdL06dPdXUqFnTp1Svn5+apfv36R7fXr19fevXvdVJXr2e12jRo1St26dVObNm3cXY7TLV68WFu2bNHPP//s7lI8nlEz0BPyTyIDyUBUFhlIBlY3np5/EhlYVYyaf5JnZKAR80/y/Awk/6qOUTPQE/JPIgPJwLLhzpRyGjdunEwmU4lfe/fu1axZs5SZmamnn37a3SVXWFl7/aMjR47o1ltv1T333KNhw4a5qXJU1IgRI7Rr1y4tXrzY3aU4XVpamuLj47Vo0SL5+fm5u5xqyygZSP4ZExmI0pCBZKCn8uT8k8hAZzBK/klkoBF5cgaSf85hlAwk/4yJDCwfk8PhcDjlTAZx8uRJ/fbbbyXu06RJE9177736/PPPZTKZCrfn5+fLYrFo0KBBmjdvnqtLrbSy9urj4yNJOnr0qHr27KmuXbtq7ty5Mpur76wuNzdXAQEBWrp0qfr371+4fciQITp79qyWL1/uvuJc5LHHHtPy5cu1Zs0aRUZGurscp/vss8905513ymKxFG7Lz8+XyWSS2WxWTk5OkddweUbJQCPnn0QGkoG4EjLwd2Sg5/D0/JPIQGcwSv5Jxs5Ao+Wf5PkZSP45h1Ey0Mj5J5GBZGDZMExxkdTUVGVkZBR+f/ToUfXt21dLly5Vly5d1KhRIzdW53xHjhxRr1691LFjRy1cuNAj/jLu0qWLOnfurFmzZkkquO0tPDxcjz32mMaNG+fm6pzH4XDo8ccf16effqqEhAQ1a9bM3SW5RGZmpg4dOlRk20MPPaQWLVroqaee8shbGd3JSBnoifknkYGehgysWmRg9WeEDDRK/klkYFUyUv5JnpmBRsg/yTgZSP5VLSNloCfmn0QGehpXZCBrprhIeHh4ke9r1qwpSWratKlHhadUEKA9e/ZU48aNNX36dJ08ebLwtdDQUDdWVjljxozRkCFDFBMTo86dO2vGjBm6cOGCHnroIXeX5lQjRozQhx9+qOXLl8tqter48eOSpFq1asnf39/N1TmP1WotFpI1atRQ3bp1+QekCxglAz01/yQykAxEZZCBZGB1YJT8k8jAqmSU/JM8NwONkH+ScTKQ/KtaRslAT80/iQwkA0vHMAWVtmrVKiUnJys5ObnYXw7V+can++67TydPntRzzz2n48ePKzo6Wl999VWxhaiqu9mzZ0uSevbsWWT7Bx98oNjY2KovCKhGPDX/JDKQDARKRwZWb+QfUDmemoFGyD+JDAQqw1PzTyIDycDS8ZgvAAAAAAAAAACAElTvlYEAAAAAAAAAAABcjGEKAAAAAAAAAABACRimAAAAAAAAAAAAlIBhCgAAAAAAAAAAQAkYpgAAAAAAAAAAAJSAYQoAAAAAAAAAAEAJGKYAAAAAAAAAAACUgGEKAAAAAAAAAABACRimAAAAAAAAAAAAlIBhCgAAAAAAAAAAQAkYpsCQTp48qdDQUE2ZMqVwW2Jionx8fPTdd9+5sTIAcC3yD4CRkYEAjIwMBGBU5B+cxeRwOBzuLgJwh5UrV6p///5KTExUVFSUoqOjdccdd+jVV191d2kA4FLkHwAjIwMBGBkZCMCoyD84A8MUGNqIESP07bffKiYmRjt37tTPP/8sX19fd5cFAC5H/gEwMjIQgJGRgQCMivxDZTFMgaFdvHhRbdq0UVpamjZv3qy2bdu6uyQAqBLkHwAjIwMBGBkZCMCoyD9UFmumwNAOHDigo0ePym63KyUlxd3lAECVIf8AGBkZCMDIyEAARkX+obK4MwWGlZubq86dOys6OlpRUVGaMWOGdu7cqZCQEHeXBgAuRf4BMDIyEICRkYEAjIr8gzMwTIFhjR07VkuXLtX27dtVs2ZN9ejRQ7Vq1dKKFSvcXRoAuBT5B8DIyEAARkYGAjAq8g/OwGO+YEgJCQmaMWOGFixYoMDAQJnNZi1YsEBr167V7Nmz3V0eALgM+QfAyMhAAEZGBgIwKvIPzsKdKQAAAAAAAAAAACXgzhQAAAAAAAAAAIASMEwBAAAAAAAAAAAoAcMUAAAAAAAAAACAEjBMAQAAAAAAAAAAKAHDFAAAAAAAAAAAgBIwTAEAAAAAAAAAACgBwxQAAAAAAAAAAIASMEwBAAAAAAAAAAAoAcMUAAAAAAAAAACAEjBMAQAAAAAAAAAAKAHDFAAAAAAAAAAAgBIwTAEAAAAAAAAAACjB/wf0XT85YUwmhQAAAABJRU5ErkJggg=="
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "fig, axs = plt.subplots(1, 5, figsize = (20,4), sharex=True, sharey=True)\n",
    "for i, ax in enumerate(axs):\n",
    "    # data\n",
    "    idx = i + 20\n",
    "    tmp = np.array(traj[idx][0])\n",
    "    ax.plot(tmp[:,0], tmp[:, 2], label = 'diffusion')\n",
    "\n",
    "    ax.plot(tmp[0,0], tmp[0,2], '*')\n",
    "    ax.plot(tmp[-1,0], tmp[-1,2], 'ro', label='end') # end\n",
    "\n",
    "    x, _, y, _ = b_data['obs'][episode_ends[idx]+1:episode_ends[idx+1]].T\n",
    "    ax.plot(x,y, label = 'baseline lqr')\n",
    "\n",
    "    assert np.mean(b_stats['x0'][idx] - tmp[0]) < 1e-6 # same initial condition\n",
    "\n",
    "    ax.plot(b_stats['x0'][idx,0] ,b_stats['x0'][idx,2], 'bo', label='start') #start\n",
    "    ax.set_xlabel('x')\n",
    "    ax.grid()\n",
    "\n",
    "axs[0].set_ylabel('y')\n",
    "axs[0].legend()\n",
    "\n",
    "plt.savefig('2D_baseline_trajectories.pdf', bbox_inches = 'tight', pad_inches = 0)"
   ],
   "metadata": {
    "collapsed": false,
    "ExecuteTime": {
     "end_time": "2023-05-04T04:52:05.894139423Z",
     "start_time": "2023-05-04T04:52:05.342584343Z"
    }
   }
  },
  {
   "cell_type": "markdown",
   "source": [
    "# Baseline 500 average on 500\n"
   ],
   "metadata": {
    "collapsed": false
   }
  },
  {
   "cell_type": "code",
   "execution_count": 59,
   "outputs": [],
   "source": [
    "def one_rollout(A, B, x0_rollout, x_target_rollout, n_state, n_input, qr_seed):\n",
    "    np.random.seed(qr_seed)\n",
    "    Q = np.random.uniform(low=0.0, high=1.0, size=(n_state, n_state))\n",
    "    Q = Q @ Q.T\n",
    "\n",
    "    R = np.random.uniform(low=0.0, high=1.0, size=(n_input, n_input))\n",
    "    R = R @ R.T\n",
    "\n",
    "    # steady state lqr\n",
    "    X = scipy.linalg.solve_discrete_are(A, B, Q, R)\n",
    "    F = np.linalg.inv(R + B.T @ X @ B) @ B.T @ X @ A\n",
    "\n",
    "    x_rollout = np.zeros((rollout_len, n_state))\n",
    "    x_rollout[0] = x0_rollout\n",
    "    u_rollout = np.zeros((rollout_len, n_input))\n",
    "    for i_rollout in range(rollout_len):\n",
    "        dist_to_target = np.linalg.norm(x_rollout[i_rollout] - x_target_rollout)\n",
    "\n",
    "        if dist_to_target < finish_threshold or i_rollout == (rollout_len - 1):\n",
    "            finish_token_rollout = i_rollout\n",
    "            break\n",
    "        u_rollout[i_rollout] = -F @ x_rollout[i_rollout]\n",
    "        x_rollout[i_rollout + 1] = A @ x_rollout[i_rollout] + B @ u_rollout[i_rollout]\n",
    "\n",
    "    # x_rollout,\n",
    "    return finish_token_rollout"
   ],
   "metadata": {
    "collapsed": false,
    "ExecuteTime": {
     "end_time": "2023-05-04T04:11:57.877409646Z",
     "start_time": "2023-05-04T04:11:57.834898179Z"
    }
   }
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "outputs": [],
   "source": [
    "A, B, x_target, n_state, n_input, name = import_dynamics(system_name='2d')\n",
    "\n",
    "n_rand_controllers = 500  # how many random controllers we want to generate\n",
    "rollout_len = 250  # how many time steps a rollout is\n",
    "finish_threshold = 0.02  # stopping condition\n",
    "\n",
    "load_results = True\n",
    "if load_results:\n",
    "    f = open(f'datasets/baseline_2d_500average_on500.pkl', 'rb')\n",
    "    finish_step_avg = pickle.load(f) # REWARD, TRAJECTORIES\n",
    "    f.close()\n",
    "else:\n",
    "    finish_step_avg = []\n",
    "    for i_contr in range(n_rand_controllers):\n",
    "        print(i_contr, end = ' ')\n",
    "        np.random.seed(i_contr + 10000)\n",
    "\n",
    "\n",
    "        x0 = np.random.uniform(low=-5.0, high=5.0, size=(n_state))\n",
    "\n",
    "        finish_step = [one_rollout(A=A, B=B, x0_rollout=x0.T, x_target_rollout=x_target.T, n_state=n_state, n_input=n_input, qr_seed=jj+ int(3e4)) for jj in range(n_rand_controllers)]\n",
    "\n",
    "        finish_step_avg.append(np.mean(finish_step))\n",
    "\n",
    "    ## SAVE ##\n",
    "    # with open(f'datasets/baseline_2d_500average_on500.pkl', 'wb') as handle:\n",
    "    #     pickle.dump(finish_step_avg, handle, protocol=pickle.HIGHEST_PROTOCOL)\n",
    "    # handle.close()"
   ],
   "metadata": {
    "collapsed": false
   }
  },
  {
   "cell_type": "code",
   "execution_count": 81,
   "outputs": [
    {
     "data": {
      "text/plain": "<Figure size 800x700 with 2 Axes>",
      "image/png": "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"
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "fig, (ax1, ax2) = plt.subplots(2,1, figsize=(8,7), sharex=True)\n",
    "ckpt_path = '2d_arch256_e100_d100_edim256_ks5_par3_70e06_date04_26_02_22_20'\n",
    "f = open(f'inference/{ckpt_path}/data.pkl', 'rb')\n",
    "data_loaded = pickle.load(f) # REWARD, TRAJECTORIES\n",
    "f.close()\n",
    "freq, thres, _x_ = ax1.hist(data_loaded[:,-1], bins = 60); #, orientation='horizontal');\n",
    "ax1.set_title('Diffusion Policy')\n",
    "ax1.grid()\n",
    "\n",
    "ax2.hist(finish_step_avg, bins = 60);\n",
    "ax2.set_xlabel('Len trajectories')\n",
    "ax2.set_title('Mean over 500 LQR')\n",
    "ax2.grid()\n",
    "plt.savefig('2D_diffusion_vs_lqr.pdf', bbox_inches = 'tight', pad_inches = 0)\n"
   ],
   "metadata": {
    "collapsed": false,
    "ExecuteTime": {
     "end_time": "2023-05-04T04:40:44.786181375Z",
     "start_time": "2023-05-04T04:40:44.314324108Z"
    }
   }
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 2
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython2",
   "version": "2.7.6"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 0
}