visualize-defaults.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Use this notebook to visualize results obtained from using the default evolutionary operators"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Note: this notebook assumes a certain structure of the saved population files. Your custom implemented operators may not conform."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**You may need to manually set ffmpeg path**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "# import os\n",
    "# os.environ[\"IMAGEIO_FFMPEG_EXE\"] = \"...\""
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**Imports**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "import pickle, numpy as np, os, seaborn as sns, sys\n",
    "from matplotlib import pyplot as plt\n",
    "from tqdm import tqdm\n",
    "import moviepy.video.io.ImageSequenceClip\n",
    "from IPython.display import Video\n",
    "sys.path.append(os.path.dirname(os.path.abspath('')))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**Fill in data directory**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [],
   "source": [
    "outdir = \"./eldir-outputs\""
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**Compute number of generations**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [],
   "source": [
    "subdirs = os.listdir(outdir)\n",
    "generations = [s for s in subdirs if s.isdigit()]\n",
    "generations = sorted(generations, key=int)[1:]"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**Collect fitness statistics across generations**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 40/40 [00:00<00:00, 3405.85it/s]\n"
     ]
    }
   ],
   "source": [
    "generational_means = []\n",
    "generational_stds = []\n",
    "generational_best = []\n",
    "for g in tqdm(generations):\n",
    "    fitness = np.load(os.path.join(outdir, g, \"robots-fitness.npy\"))\n",
    "    fitness = np.max(fitness, axis=1)\n",
    "    generational_means.append(np.mean(fitness))\n",
    "    generational_stds.append(np.std(fitness))\n",
    "    generational_best.append(np.max(fitness))\n",
    "generational_means = np.array(generational_means)\n",
    "generational_stds = np.array(generational_stds)\n",
    "generational_best = np.array(generational_best)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**Plot fitness stats**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.legend.Legend at 0x7f57e0e6a530>"
      ]
     },
     "execution_count": 12,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAA04AAAHWCAYAAABACtmGAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjguMywgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy/H5lhTAAAACXBIWXMAAA9hAAAPYQGoP6dpAACUSElEQVR4nOzdd3xT5f4H8M/JXk26B22hZe89REVAUVwoPxUQB0NcF8VR8SrXBXqvCCqiwhVFxYmiiOM6AEFQUZQlUzYdrO6RtmnmOb8/0oaGFtq0aZO0n/fr1Vfbk+Scb9qI+fR5nu8jSJIkgYiIiIiIiM5JFugCiIiIiIiIgh2DExERERERUR0YnIiIiIiIiOrA4ERERERERFQHBiciIiIiIqI6MDgRERERERHVgcGJiIiIiIioDgxOREREREREdWBwIiIiIiIiqgODExGFrBEjRmDEiBF+Pefs2bMhCIJfz1lfZWVluPPOOxEfHw9BEPDQQw8hIyMDgiDgvffeC0hNLdmUKVOQkpLit/M1xeuRWgZ/v9aIKDAYnIioVu+99x4EQfB8aDQadO7cGffffz9ycnICXV6jWCwWzJ49Gxs3bgx0KV6ef/55vPfee/jHP/6BDz/8ELfffnut9/v+++8xe/bs5i0uQP773/82KjSeOnUKs2fPxs6dO/1Sz99//43Zs2cjIyPDL+ejlsPfrzUiCj6CJElSoIsgouDz3nvvYerUqXj22WeRmpoKq9WKTZs24cMPP0S7du2wd+9e6HS6gNZY9dd9XwNQfn4+YmJi8Mwzz9QIIE6nE06nExqNxj9F+uCCCy6AQqHApk2bPMckSYLNZoNSqYRcLgcA3H///Vi8eDFawz/fPXv2RHR0dIND7rZt2zBo0CAsW7YMU6ZM8brN4XBAFEWo1ep6n2/lypUYN24cNmzYUGN0yW63AwBUKlWDaqXQ5u/XGhEFH0WgCyCi4HbVVVdh4MCBAIA777wTUVFRWLBgAb7++mtMnDgxwNX5n0KhgEIRmH8ac3Nz0b17d69jVaN95H9KpdKv52utgUmSJFitVmi12kCX4ldWqxUqlQoyWeMn5/j7tUZEgcGpekTkk0svvRQAkJ6eDsA9QvPcc8+hQ4cOUKvVSElJwb/+9S/YbDavx6WkpODaa6/F2rVr0bdvX2g0GnTv3h2rVq3yut+51hhVTR083xQpu92Op59+GgMGDIDJZIJer8ewYcOwYcMGz30yMjIQExMDAJgzZ45nKmLVyFNt1/f1OW7atAmDBw+GRqNB+/bt8cEHH5znJ+oeMRMEAenp6fjuu+88NWVkZNRY4zRlyhQsXrwYALymUlY9N0EQ8NJLL+Gtt97y1Dto0CBs3bq1xnUPHDiAm266CZGRkdBoNBg4cCC++eYbr/s4HA7MmTMHnTp1gkajQVRUFC6++GL8+OOPnvtkZ2dj6tSpSEpKglqtRkJCAq6//vo6p7PV9biUlBTs27cPP//8s+d5Vo3yFBYWYubMmejVqxcMBgOMRiOuuuoq7Nq1y+vnOmjQIADA1KlTPeeo/rM8e93Jp59+igEDBiAsLAxGoxG9evXCq6++CsD9Ghw3bhwAYOTIkZ7zVY2G1bbGyWq1Yvbs2ejcuTM0Gg0SEhJwww034OjRo/W65vmUl5fjkUceQXJyMtRqNbp06YKXXnrJaySyZ8+eGDlyZI3HiqKIxMRE3HTTTV7HFi5ciB49ekCj0SAuLg733HMPioqKvB5b9Tpfs2YNBg4cCK1WizfffPO8tS5evBjt27eHVqvF4MGD8euvv9b687LZbHjmmWfQsWNHqNVqJCcn45///GeN/9YEQcD999+Pr776Cj179oRarUaPHj2wevXqGtc+efIk7rjjDsTFxXnu9+6773rdp+q/wU8//RRPPvkkEhMTodPpYDabm+y1Vp/fny/PtbS0FA899BBSUlKgVqsRGxuLyy+/HDt27Djv74aI6o8jTkTkk6o3fFFRUQDco1Dvv/8+brrpJjzyyCP4888/MXfuXOzfvx9ffvml12MPHz6MCRMm4N5778XkyZOxbNkyjBs3DqtXr8bll1/e6NrMZjPefvttTJw4EXfddRdKS0vxzjvvYPTo0diyZQv69u2LmJgYvPHGG/jHP/6B//u//8MNN9wAAOjdu/c5z+vLczxy5AhuuukmTJs2DZMnT8a7776LKVOmYMCAAejRo0et5+/WrRs+/PBDPPzww0hKSsIjjzwCAIiJiUFeXp7Xfe+55x6cOnUKP/74Iz788MNaz7d8+XKUlpbinnvugSAImD9/Pm644QYcO3bM85fvffv24aKLLkJiYiIef/xx6PV6fPbZZxg7diy++OIL/N///R8Ad5CcO3cu7rzzTgwePBhmsxnbtm3Djh07PL+zG2+8Efv27cOMGTOQkpKC3Nxc/Pjjj8jKyjrvgvi6Hrdw4ULMmDEDBoMBTzzxBAAgLi4OAHDs2DF89dVXGDduHFJTU5GTk4M333wTw4cPx99//402bdqgW7duePbZZ/H000/j7rvvxrBhwwAAF154Ya31/Pjjj5g4cSIuu+wyzJs3DwCwf/9+/Pbbb3jwwQdxySWX4IEHHsBrr72Gf/3rX+jWrZvn91cbl8uFa6+9FuvXr8fNN9+MBx98EKWlpfjxxx+xd+9edOjQoc5rnoskSbjuuuuwYcMGTJs2DX379sWaNWvw6KOP4uTJk3jllVcAABMmTMDs2bORnZ2N+Ph4z+M3bdqEU6dO4eabb/Ycu+eeezxTdB944AGkp6dj0aJF+Ouvv/Dbb795jZocPHgQEydOxD333IO77roLXbp0OWetb7zxBu6//34MGzYMDz/8MDIyMjB27FhEREQgKSnJcz9RFHHddddh06ZNuPvuu9GtWzfs2bMHr7zyCg4dOoSvvvrK67ybNm3CqlWrMH36dISFheG1117DjTfeiKysLM+/Tzk5Objgggs84SMmJgY//PADpk2bBrPZjIceesjrnM899xxUKhVmzpwJm80GlUqFv//+2++vtfr+/nx5rvfeey9WrlyJ+++/H927d0dBQQE2bdqE/fv3o3///uf8/RCRDyQiolosW7ZMAiCtW7dOysvLk44fPy59+umnUlRUlKTVaqUTJ05IO3fulABId955p9djZ86cKQGQfvrpJ8+xdu3aSQCkL774wnOspKRESkhIkPr16+c59swzz0i1/dNUVU96errn2PDhw6Xhw4d7vnc6nZLNZvN6XFFRkRQXFyfdcccdnmN5eXkSAOmZZ56pcZ2zr9+Q5/jLL794juXm5kpqtVp65JFHalzrbO3atZOuueYar2Pp6ekSAGnZsmWeY/fdd1+tP6Oq+0ZFRUmFhYWe419//bUEQPrf//7nOXbZZZdJvXr1kqxWq+eYKIrShRdeKHXq1MlzrE+fPjVqqq6oqEgCIL344ot1Pr+GPK5Hjx5ev+MqVqtVcrlcXsfS09MltVotPfvss55jW7durfHzqzJ58mSpXbt2nu8ffPBByWg0Sk6n85z1fP755xIAacOGDTVuO/v1+O6770oApAULFtS4ryiK9b5mbb766isJgPTvf//b6/hNN90kCYIgHTlyRJIkSTp48KAEQHr99de97jd9+nTJYDBIFotFkiRJ+vXXXyUA0scff+x1v9WrV9c4XvU6X716dZ112mw2KSoqSho0aJDkcDg8x9977z0JgNfP68MPP5RkMpn066+/ep1jyZIlEgDpt99+8xwDIKlUKs/zlCRJ2rVrV43nOm3aNCkhIUHKz8/3OufNN98smUwmz/PfsGGDBEBq376951iVpnit1ff358tzNZlM0n333Vfj2kTkP5yqR0TnNWrUKMTExCA5ORk333wzDAYDvvzySyQmJuL7778HAKSlpXk9pmrE5LvvvvM63qZNG89IBgAYjUZMmjQJf/31F7Kzsxtdq1wu96wzEUURhYWFcDqdGDhwYIOnq/j6HLt37+75azPgHjXq0qULjh071qDrN8SECRMQERHh+b6qnqoaCgsL8dNPP2H8+PEoLS1Ffn4+8vPzUVBQgNGjR+Pw4cM4efIkACA8PBz79u3D4cOHa72WVquFSqXCxo0ba0zpOp+GPq6KWq32rD1xuVwoKCiAwWBAly5dGvy7Dg8PR3l5udc0xMb44osvEB0djRkzZtS4rWp6ZUOv+f3330Mul+OBBx7wOv7II49AkiT88MMPAIDOnTujb9++WLFihec+LpcLK1euxJgxYzzrkj7//HOYTCZcfvnlntdDfn4+BgwYAIPB4DXdFQBSU1MxevToOuvctm0bCgoKcNddd3mtHbz11lu9XqNVNXTr1g1du3b1qqFqevDZNYwaNQodOnTwfN+7d28YjUbP61ySJHzxxRcYM2YMJEnyOufo0aNRUlJS47UyefLkGmu1muK1Vt/fX32fK+B+Lf355584depUg2oioroxOBHReS1evBg//vgjNmzYgL///hvHjh3zvGHKzMyETCZDx44dvR4THx+P8PBwZGZmeh3v2LFjjfVDnTt3BgC/tXd+//330bt3b896nJiYGHz33XcoKSlp0Pl8fY5t27atcY6IiIgGhYOGOruGqjeoVTUcOXIEkiThqaeeQkxMjNfHM888A8DdqAIAnn32WRQXF6Nz587o1asXHn30UezevdtzbrVajXnz5uGHH35AXFwcLrnkEsyfP7/OINzQx1URRRGvvPIKOnXqBLVajejoaMTExGD37t0N/l1Pnz4dnTt3xlVXXYWkpCTccccdta6Zqa+jR4+iS5cu52020tBrZmZmok2bNggLC/M6XjVtsPrrcsKECfjtt988YXjjxo3Izc3FhAkTPPc5fPgwSkpKEBsbW+M1UVZW5nk9VElNTa37B1CtjrP/+1EoFDWmcR4+fBj79u2rcf2qfyPOrqGu/9by8vJQXFyMt956q8Y5p06dWus5a3teTfFa8+X3V5/nCgDz58/H3r17kZycjMGDB2P27NnN+gcbotaAa5yI6LwGDx7s6ap3Lv7cMPZc53K5XHU+9qOPPsKUKVMwduxYPProo4iNjYVcLsfcuXO9FuP7s66zVbUMP5vUjK3D66pBFEUAwMyZM885alD1RveSSy7B0aNH8fXXX2Pt2rV4++238corr2DJkiW48847AQAPPfQQxowZg6+++gpr1qzBU089hblz5+Knn35Cv379zllnQx8HuPe8euqpp3DHHXfgueeeQ2RkJGQyGR566CHP8/NVbGwsdu7ciTVr1uCHH37ADz/8gGXLlmHSpEl4//33G3TOYLjmhAkTMGvWLHz++ed46KGH8Nlnn8FkMuHKK6/03EcURcTGxuLjjz+u9RxVDVWqNEUHPVEU0atXLyxYsKDW25OTk72+r+/r/LbbbsPkyZNrve/Zaxtre15N8VrzVX3+XRk/fjyGDRuGL7/8EmvXrsWLL76IefPmYdWqVbjqqquapU6ilo7BiYgarF27dhBFEYcPH/ZaIJ+Tk4Pi4mK0a9fO6/5VIx3VQ8ihQ4cAwPPX56rRkeLiYoSHh3vud/ZfYGuzcuVKtG/fHqtWrfK6RtUoShVfgp6vz7E5NDaotm/fHoC7RfKoUaPqvH9kZCSmTp2KqVOnoqysDJdccglmz57tCU4A0KFDBzzyyCN45JFHcPjwYfTt2xcvv/wyPvroo/Oeu67Hneu5rly5EiNHjsQ777zjdby4uBjR0dGe7339WalUKowZMwZjxoyBKIqYPn063nzzTTz11FO1jpjW9dz+/PNPOByO87ajruuatWnXrh3WrVuH0tJSr1GLAwcOeG6vkpqaisGDB2PFihW4//77sWrVKowdO9ZrT6EOHTpg3bp1uOiii/waiqrqOHLkiFd3P6fTiYyMDK/g0qFDB+zatQuXXXaZX/4YExMTg7CwMLhcrnq9zs+lKV5rvvz+fJGQkIDp06dj+vTpyM3NRf/+/fGf//yHwYnITzhVj4ga7OqrrwYALFy40Ot41V+Mr7nmGq/jp06d8upCZzab8cEHH6Bv376ejl9V8/h/+eUXz/3Ky8vr9df3qr/KVv8r7J9//onNmzd73a9q497i4uI6z+nrc2wOer0eQP3qr01sbCxGjBiBN998E6dPn65xe/VOfgUFBV63GQwGdOzY0dMe2mKxwGq1et2nQ4cOCAsLq9FCurr6Pk6v19f6POVyeY1RvM8//9wzHa3644H6/azOfq4ymczzxr6qJl/Od+ONNyI/Px+LFi2qcVtV7fW5Zm2uvvpquFyuGud+5ZVXIAhCjTfKEyZMwB9//IF3330X+fn5XtP0APdohcvlwnPPPVfjWk6ns8GvtYEDByIqKgpLly6F0+n0HP/4449rTF8dP348Tp48iaVLl9Y4T0VFBcrLy326tlwux4033ogvvvgCe/furXH72R0rz3cef7/WfP391cXlctWYNhgbG4s2bdqc93VERL7hiBMRNVifPn0wefJkvPXWWyguLsbw4cOxZcsWvP/++xg7dmyN/WM6d+6MadOmYevWrYiLi8O7776LnJwcLFu2zHOfK664Am3btsW0adPw6KOPQi6X491330VMTAyysrLOW8+1116LVatW4f/+7/9wzTXXID09HUuWLEH37t1RVlbmuZ9Wq0X37t2xYsUKdO7cGZGRkejZsyd69uzZ6OfYHAYMGAAAeOCBBzB69GjI5XKvttL1sXjxYlx88cXo1asX7rrrLrRv3x45OTnYvHkzTpw44dmjpnv37hgxYgQGDBiAyMhIbNu2zdPyGHCPGF522WUYP348unfvDoVCgS+//BI5OTnnram+jxswYADeeOMN/Pvf/0bHjh0RGxuLSy+9FNdeey2effZZTJ06FRdeeCH27NmDjz/+2DOaVqVDhw4IDw/HkiVLEBYWBr1ejyFDhtS6luXOO+9EYWEhLr30UiQlJSEzMxOvv/46+vbt6xlt7Nu3L+RyOebNm4eSkhKo1WpceumliI2NrXG+SZMm4YMPPkBaWhq2bNmCYcOGoby8HOvWrcP06dNx/fXX1+uatRkzZgxGjhyJJ554AhkZGejTpw/Wrl2Lr7/+Gg899JBXIwHAHUpmzpyJmTNnIjIyssYIzPDhw3HPPfdg7ty52LlzJ6644goolUocPnwYn3/+OV599VWvPZ/qS6VSYfbs2ZgxYwYuvfRSjB8/HhkZGXjvvffQoUMHr1Ga22+/HZ999hnuvfdebNiwARdddBFcLhcOHDiAzz77zLNvlC9eeOEFbNiwAUOGDMFdd92F7t27o7CwEDt27MC6detQWFhY5zma4rXm6++vLqWlpUhKSsJNN92EPn36wGAwYN26ddi6dStefvlln85FROcRiFZ+RBT8qtp/b9269bz3czgc0pw5c6TU1FRJqVRKycnJ0qxZs7zaXEvSmVbba9askXr37i2p1Wqpa9eu0ueff17jnNu3b5eGDBkiqVQqqW3bttKCBQvq1Y5cFEXp+eefl9q1ayep1WqpX79+0rffflujFbAkSdLvv/8uDRgwQFKpVF6tyWtrh+7rczzb2XWeS33bkTudTmnGjBlSTEyMJAiCp96q+9bW4rv6c6xy9OhRadKkSVJ8fLykVCqlxMRE6dprr5VWrlzpuc+///1vafDgwVJ4eLik1Wqlrl27Sv/5z38ku90uSZIk5efnS/fdd5/UtWtXSa/XSyaTSRoyZIj02Wefnfe51vdx2dnZ0jXXXCOFhYV5ta+2Wq3SI488IiUkJEharVa66KKLpM2bN9f6s/7666+l7t27SwqFwutnefbrYuXKldIVV1whxcbGel5799xzj3T69Gmv8y1dulRq3769JJfLvVqT13Zti8UiPfHEE57XTnx8vHTTTTdJR48e9ematSktLZUefvhhqU2bNpJSqZQ6deokvfjii55W52e76KKLam2tX91bb70lDRgwQNJqtVJYWJjUq1cv6Z///Kd06tQpz33O9To/n9dee83z3+XgwYOl3377TRowYIB05ZVXet3PbrdL8+bNk3r06CGp1WopIiJCGjBggDRnzhyppKTEcz8AtbbebteunTR58mSvYzk5OdJ9990nJScne34Hl112mfTWW2957lPVjry2f4+a4rUmSfX//dXnudpsNunRRx+V+vTpI4WFhUl6vV7q06eP9N///rfG44io4QRJasYVy0TUaqWkpKBnz5749ttvA10KEQWYKIqIiYnBDTfcUOvUPCKiYMQ1TkRERNRkrFZrjTVCH3zwAQoLCzFixIjAFEVE1ABc40RERERN5o8//sDDDz+McePGISoqCjt27MA777yDnj17Yty4cYEuj4io3hiciIiIqMmkpKQgOTkZr732GgoLCxEZGYlJkybhhRdegEqlCnR5RET1xjVOREREREREdeAaJyIiIiIiojowOBEREREREdWh1a1xEkURp06dQlhYmNfGe0RERERE1LpIkoTS0lK0adMGMtn5x5RaXXA6deoUkpOTA10GEREREREFiePHjyMpKem892l1wSksLAyA+4djNBoDXA0REREREQWK2WxGcnKyJyOcT6sLTlXT84xGI4MTERERERHVawkPm0MQERERERHVgcGJiIiIiIioDgxOREREREREdWh1a5zqQ5IkOJ1OuFyuQJfSYsjlcigUCraAJyIiIqKQxOB0FrvdjtOnT8NisQS6lBZHp9MhISEBKpUq0KUQEREREfmEwakaURSRnp4OuVyONm3aQKVScYTEDyRJgt1uR15eHtLT09GpU6c6NxgjIiIiIgomDE7V2O12iKKI5ORk6HS6QJfTomi1WiiVSmRmZsJut0Oj0QS6JCIiIiKieuOf/WvB0ZCmwZ8rEREREYUqvpMlIiIiIiKqA4MTERERERFRHRiciIiIiIiI6sDg1EJMmTIFgiB4PqKionDllVdi9+7dfjn/7Nmz0bdvX7+ci4iIiIgo1DA4tSBXXnklTp8+jdOnT2P9+vVQKBS49tprA10WEREREVHIYzvyOkiShAqHKyDX1irlPu0jpVarER8fDwCIj4/H448/jmHDhiEvLw8xMTE4fvw4HnnkEaxduxYymQzDhg3Dq6++ipSUFADAxo0b8c9//hP79u2DUqlEjx49sHz5cmzYsAFz5swBAE89y5Ytw5QpU/z6fImIiILBJ1uy8MHmTEiSFOhSiFq05XddgEi9KtBl1BuDUx0qHC50f3pNQK7997OjoVM17FdUVlaGjz76CB07dkRUVBQcDgdGjx6NoUOH4tdff4VCocC///1vz3Q+mUyGsWPH4q677sInn3wCu92OLVu2QBAETJgwAXv37sXq1auxbt06AIDJZPLnUyUiIgoKxwsteOabfbA7xUCXQtTiOcXQ+u+MwakF+fbbb2EwGAAA5eXlSEhIwLfffguZTIbly5dDFEW8/fbbXqNG4eHh2LhxIwYOHIiSkhJce+216NChAwCgW7dunnMbDAYoFArPiBYREVFL9MLqA7A7RQxOjcQDl3YKdDlELZpJqwx0CT5hcKqDVinH38+ODti1fTFy5Ei88cYbAICioiL897//xVVXXYUtW7Zg165dOHLkCMLCwrweY7VacfToUVxxxRWYMmUKRo8ejcsvvxyjRo3C+PHjkZCQ4LfnQ0REFMy2ZhTiu92nIROAOdf1QLcEY6BLIqIgwuBUB0EQGjxdrrnp9Xp07NjR8/3bb78Nk8mEpUuXoqysDAMGDMDHH39c43ExMTEA3CNQDzzwAFavXo0VK1bgySefxI8//ogLLrig2Z4DERFRIIiihGf/9zcA4ObBbRmaiKiG0EgE1CCCIEAmk6GiogL9+/fHihUrEBsbC6Px3P8z6NevH/r164dZs2Zh6NChWL58OS644AKoVCq4XIFpkkFERNTUVv11EntOliBMrUDa5Z0DXQ4RBSG2I29BbDYbsrOzkZ2djf3792PGjBkoKyvDmDFjcOuttyI6OhrXX389fv31V6Snp2Pjxo144IEHcOLECaSnp2PWrFnYvHkzMjMzsXbtWhw+fNizziklJQXp6enYuXMn8vPzYbPZAvxsiYiI/KPc5sT81QcAADMu64hogzrAFRFRMOKIUwuyevVqz5qksLAwdO3aFZ9//jlGjBgBAPjll1/w2GOP4YYbbkBpaSkSExNx2WWXwWg0oqKiAgcOHMD777+PgoICJCQk4L777sM999wDALjxxhuxatUqjBw5EsXFxWxHTkRELcaSn48it9SGdlE6TL4wJdDlEFGQEqRWtkmB2WyGyWRCSUlJjSlrVqsV6enpSE1NhUajCVCFLRd/vkREFGxOFFlw2cs/w+YU8ebtAzC6B7vHErUm58sGZ+NUPSIiImq15q0+CJtTxND2Ubiie1ygyyGiIMbgRERERK3S9sxC/G/XKQgC8OS13Tz7HBIR1YbBiYiIiFqd6u3HJwxMRo82pgBXRETBjsGJiIiIWp2vdp7ErhMlMKgVeOSKLoEuh4hCAIMTERERtSoWuxPzKtuP3zeyI2LC2H6ciOrG4EREREStypKfjyHHbENypBZTL0oJdDlEFCIYnIiIiKjVOFlcgTd/PgoA+NdV3aBRygNcERGFCgYnIiIiajXmrz4Am1PE4NRIXNmTezYRUf0xOBEREVGrsD2zCF/vdLcff/ra7mw/TkQ+YXAiIiKiFk8UJTz3rbv9+LgBSeiZyPbjROQbBqcWYsqUKRAEAffee2+N2+677z4IgoApU6Y0f2FERERB4Jtdp7DzeDH0Kjlmsv04ETUAg1MLkpycjE8//RQVFRWeY1arFcuXL0fbtm0DWBkREVHgWOxOvPCDu/349JEdEWvUBLgiIgpFDE51kSTAXh6YD0nyqdT+/fsjOTkZq1at8hxbtWoV2rZti379+nmOiaKIuXPnIjU1FVqtFn369MHKlSs9t7tcLkybNs1ze5cuXfDqq696XWvKlCkYO3YsXnrpJSQkJCAqKgr33XcfHA5HA3/QRERETeOtX44h22xFYrgW0y5ODXQ5RBSiFIEuIOg5LMDzbQJz7X+dAlR6nx5yxx13YNmyZbj11lsBAO+++y6mTp2KjRs3eu4zd+5cfPTRR1iyZAk6deqEX375BbfddhtiYmIwfPhwiKKIpKQkfP7554iKisLvv/+Ou+++GwkJCRg/frznPBs2bEBCQgI2bNiAI0eOYMKECejbty/uuusuvzx9IiKixjpdUoElVe3Hr2b7cSJqOAanFua2227DrFmzkJmZCQD47bff8Omnn3qCk81mw/PPP49169Zh6NChAID27dtj06ZNePPNNzF8+HAolUrMmTPHc87U1FRs3rwZn332mVdwioiIwKJFiyCXy9G1a1dcc801WL9+PYMTEREFjfmrD8LqEDEoJQJX92L7cSJqOAanuih17pGfQF3bRzExMbjmmmvw3nvvQZIkXHPNNYiOjvbcfuTIEVgsFlx++eVej7Pb7V7T+RYvXox3330XWVlZqKiogN1uR9++fb0e06NHD8jlZ/5yl5CQgD179vhcMxERUVP4K6sIX/51srL9eA+2HyeiRmFwqosg+DxdLtDuuOMO3H///QDcAai6srIyAMB3332HxMREr9vUajUA4NNPP8XMmTPx8ssvY+jQoQgLC8OLL76IP//80+v+SqXS63tBECCKol+fCxERUUNIkoRnK9uP39g/Cb2S2H6ciBqHwakFuvLKK2G32yEIAkaPHu11W/fu3aFWq5GVlYXhw4fX+vjffvsNF154IaZPn+45dvTo0SatmYiIyJ++2XUKf2UVQ6eS49HRbD9ORI3H4NQCyeVy7N+/3/N1dWFhYZg5cyYefvhhiKKIiy++GCUlJfjtt99gNBoxefJkdOrUCR988AHWrFmD1NRUfPjhh9i6dStSU9mJiIiIgl+F3YV5Ve3HR3RAHNuPE5EfMDi1UEaj8Zy3Pffcc4iJicHcuXNx7NgxhIeHo3///vjXv/4FALjnnnvw119/YcKECRAEARMnTsT06dPxww8/NFf5REREDbb012M4VeJuP37nsPaBLoeIWghBknzcLCjEmc1mmEwmlJSU1AgXVqsV6enpSE1NhUbDv075G3++RETU1LJLrBj50kZUOFx4fWI/jOkToC1FiCgknC8bnI0b4BIREVGLMX/NAVQ4XBjQLgLX9k4IdDlE1IJwqh4REREFlEuUcKLI0ujzpOeXY9WOkwCAp6/tzvbjRORXAQ9Oixcvxosvvojs7Gz06dMHr7/+OgYPHnzO+y9cuBBvvPEGsrKyEB0djZtuuglz587l1C8iIqIQdDC7FPd+tB3p+eV+O+cN/RPRJzncb+cjIgICHJxWrFiBtLQ0LFmyBEOGDMHChQsxevRoHDx4ELGxsTXuv3z5cjz++ON49913ceGFF+LQoUOYMmUKBEHAggULAvAMiIiIqKG+3X0Kj36+GxUOF5RyASp541cQJIRr8diVXf1QHRGRt4AGpwULFuCuu+7C1KlTAQBLlizBd999h3fffRePP/54jfv//vvvuOiii3DLLbcAAFJSUjBx4sQaG7M2Vivrl9Fs+HMlIiIAcLpEvLj2IN78+RgA4OKO0XhtYj9E6lUBroyI6NwC1hzCbrdj+/btGDVq1JliZDKMGjUKmzdvrvUxF154IbZv344tW7YAAI4dO4bvv/8eV1999TmvY7PZYDabvT7ORalUAgAslsbPs6aaqn6uVT9nIiJqfYrK7ZiybKsnNN0zvD3emzqIoYmIgl7ARpzy8/PhcrkQFxfndTwuLg4HDhyo9TG33HIL8vPzcfHFF0OSJDidTtx7772e/YdqM3fuXMyZM6deNcnlcoSHhyM3NxcAoNPpuLDUDyRJgsViQW5uLsLDw2tsyktERK3D3pMluPej7ThRVAGtUo4Xx/XGtb3ZLpyIQkPAm0P4YuPGjXj++efx3//+F0OGDMGRI0fw4IMP4rnnnsNTTz1V62NmzZqFtLQ0z/dmsxnJycnnvEZ8fDwAeMIT+U94eLjn50tERK3Ll3+dwONf7IHNKaJdlA5v3j4AXePPv2cKEVEwCVhwio6OhlwuR05OjtfxnJycc765fuqpp3D77bfjzjvvBAD06tUL5eXluPvuu/HEE09AJqs581CtVkOtVte7LkEQkJCQgNjYWDgcDh+eEZ2PUqnkSBMRUSvkcIl4/vv9WPZbBgBgRJcYvDqhH0w6TtsmotASsOCkUqkwYMAArF+/HmPHjgUAiKKI9evX4/7776/1MRaLpUY4qnoz7u/GA3K5nG/0iYiIGiG/zIb7Pt6BP9MLAQAzLu2Ih0Z1hlzGafBEFHoCOlUvLS0NkydPxsCBAzF48GAsXLgQ5eXlni57kyZNQmJiIubOnQsAGDNmDBYsWIB+/fp5puo99dRTGDNmDEMOERFRENl1vBj3frQdp0usMKgVeHl8H4zuwenaRBS6AhqcJkyYgLy8PDz99NPIzs5G3759sXr1ak/DiKysLK8RpieffBKCIODJJ5/EyZMnERMTgzFjxuA///lPoJ4CERERneWzrcfx5Nd7YXeKaB+jx1u3D0DH2LBAl0VE1CiC1Mo21zGbzTCZTCgpKYHRyEWpRERE/mJ3inj223346I8sAMDl3eOwYHwfhGm4nomIgpMv2SCkuuoRERFRcMo1W/GPj3dge2YRBAF4eFRn3D+yI2Rcz0RELQSDExERETXK9swi/OOj7cgttSFMo8CrN/fFpV3j6n4gEVEIYXAiIiKiBpEkCcu3ZGH2N/vgcEnoHGfAm7cPRGq0PtClERH5HYMTEVEdfj+aj0+3HIfDJQa6FGpmkgRIkCBJgCgB8HwtQaq8XaxcKixK7tu8b688FsDn0JSsDhf2nTIDAK7uFY8Xb+oDvZpvLYioZeK/bkRE5/HlXyfw6Oe74RRb6ltfosaRCcA/r+yKey5pD0HgeiYiarkYnIiIzuGdTel47tu/Abj/mj60Q3SAK6JAEADIBAGCcOZrCN7Hqz7j7GNA5fEzt7c0HWIM6BhrCHQZRERNjsGJiOgskiThpbUHsXjDUQDA1ItS8NQ13dkdjIiIqBVjcCIiqsYlSnjyqz34ZMtxAMCjo7tg+ogOnIJERETUyjE4ERFVsjpceOjTnVi9LxsyAfjP//XCxMFtA10WERERBQEGJyIiAKVWB+76YBv+OFYIlVyG1yb2xZU9EwJdFhERUYshihKcogRRkqBRygNdjs8YnIio1csrtWHKsi3Yd8oMg1qBtyYNwIVsBEFERFQrSZLgqgxBLlGCS5LgctX2vej53ulyb89QpXsbI+QhtnaYwYmIWrXjhRbc/s6fyCiwIEqvwvt3DEbPRFOgyyIiImoyVcHHJUkQRcApihBFuANP5YiQS/T+2v35zH1bIwYnImq1DmSbMemdLcgttSEpQosPpw1BarQ+0GURERHVqXqw8Yz0nHVMrDYqVD0MSdyasEEYnIioVdqaUYhp722F2epE1/gwvH/HYMQZNYEui4iIWiFJkuConNrmcFVOg3OJntBTWzBi+Gl+DE5E1Oqs+zsH9y3fAZtTxMB2EXhn8iCYdMpAl0VERC1I9TDkFN1rfKrCkNMlwVG5/sfhar1T30INgxMRtSort5/AY1/shkuUcFnXWCy6pT+0qtDr7ENERE1LrGpyUH2dT+Uan6q1Qe7P1dcLVXsMw1CLw+BERK3G0l+O4T/f7wcA3NA/EfNu7A2lXBbgqoiIqClVX+dT1digqttb9SlwTk6FozowOBFRiydJEl5YfQBv/nwMAHDXsFTMuqobZCHWBpWIiM5aD+SsOeXt7FDEABREJAkypwVyawFw0g4k9QeE0Pl/MYMTEbVoTpeIf325B59tOwEAmHVVV9wzvEOAqyIiorOJojsEudcCnfna4VkXJMLO9UDBRXRCbiuGwloAhbUIcmsBFNZCKKyFkJ/js8xlO/P4xzIBbXjAyvcVgxMRtVhWhwv3L/8L6/bnQCYAL9zYG+MHJge6LCKiVsOrM1y1DVEd4pkNUp1skBB4kgSZoxxyW3HlRxEUnq9Lqh0vrgxBRVBYCyC3lUCA70N6olwNQR8DwVbK4EREFGi5pVbc++F27Mgqhkohw6KJ/XBFj/hAl0VEFNLObpHtdInVvpY4TS7QRBfk9lLI7WcHHvfXispQJLfWPC5IzgZf1qkOh1MTBZcm4qzPkXBqIr2OOTWRkJQ6dG9jhDzEpswzOBFRi7PnRAnu/nAbTpdYYdQosHTSQAxpHxXosoiIgsrZTRPOF4CclXsLUTORRHegqch3Bx1bCeS2knOOAsntVd+bGzQCVEWUq+FSh8OlDoez8vOZD5P7uCYCLk0UnFXhSB0OyFpHpGgdz5KIWo2vd57EP1fuhs0pokOMHm9PHoTUaH2gyyIiajZVjRIcrjNrhByiBIdT9NpTiKNBzahyKpzCmu9e61ORD0VFgXttUEW+57O86pi1EILkavDlXEo9XCpTzdBTIwxVhiSN+7Ok0PrxSbc8DE5E1CK4RAkvrjmIJT8fBQBc2jUWC2/uC6OGG9sSUctRFYbsLhFOlwiHq1pIEiXYnSIDUVMTXZDbzTVHfuxVI0KVx6oCUmUo8mqKUE8ulRFOTaR3yPEEoWrB6KyQJMlVTfDEicGJiEKe2erAQ5/uxE8HcgEA/xjRATOv6BJyc6eJiETRHYpsThE2pwt2p+j+qAxMDEX+JTgsUFgrR38q8t2NDzxT485Mi1N4psOVQG43N/h6olwDpzYGTm2Ue72PNhpObVTl1Lcz31etD2IACi4MTkQU0tLzy3Hn+1txNK8caoUM82/qjev7Jga6LCKic/KEI4cIm+tMOLI53eGIGk5w2c8a6XFPfZNXG/lxfy6EwpoPmbOiwddyKQ1nRn1U1UeA3F87a4ShaEhKnR+fLTU3BiciClm/HMrD/ct3wGx1IsGkwVu3D0SvJFOgyyKiVkySpBpNFarCka0yIDEcNYDohKIiD8rybPeHJRuKys/K8hz39xX5DRoNEuVqd7jRRHmaHngHonC4NO6vq0+Tg4xTwVsbBiciCjmSJOGdTel4/vv9ECVgQLsIvHFbf8SGaQJdGhG1MGd3nHPvPVSzEx07zzWczFEOZfnpakEoG0pLjtf3ioo8CFL9NnqSBLm7BXZVGNJGe33v1EZ5BSVRqQcETu1ubqH4E2dwIqKQYnW48MSXe/HFjhMAgPEDk/Dc2J5QK+QBroyIQpEoSp6RIJvTVbm2yN1sgfsQNYIkQWYvgdKSC4Ul1/25ItfzvaIiz3Nc7iit3ykFORy6WDh18XDoKz908XBWfdbFVLbHNgGCrImfIJ1NEACFXIBCJkAhk0EuEyq/l7mPVX1deR8hBMMqgxMRhYwcsxX3fLgdO48XQy4T8OQ13TDlwpSQ/MeXiJqPJEmeQFTVcIHT5hpIdHgaKSgrcqGwVAagispAVC0k+dJFzqU0VIaguMpAlOAJR06d+5hTEw3I+Eey5iIIcIcfmVD5+UzocYci70DUGhoyMTgRUUjYebwY93y4DTlmG0xaJf57a39c1DE60GURUZBwVrXjdolewcjmdMHhZDg6H8FpdY8AVRRUfs6vbKKQf+brqg9bkU/ndqmMcGhj4dS5Pxy6WDi1lZ89x+IhqgxN9OyoSvURIXm1USB59RGiVhaEfMXgRERBb9WOE3h81R7YnSI6xRrw9uSBaBfFTW2JWjpJkuCoXDfkEN2jQ87KzVyr9jByimzTfU6is7KZwkmoyk5BWXYSyrJTlSEo78ymq44yn04rCTI41ZE1wpDn62rHJAXXnvqbIAAyoXroEby+lwuVwUde7TaZABmDUKMxOBFR0HKJEuatPoC3fjkGABjVLRavTOiLMG5qSxTyXKJ741a7S4TDeWYjV2e1UMRGC+cns5dCWXYKqvKTUJaerBaQTkBZdgpKS3a9GyqIMtWZttnamGqNFCq/rvbhUkdwypyf1DYdTl59OpzMe0So6oMCg8GJiIJSSYUDD3zyF34+lAcAuH9kR6Rd3pl/MSMKEQ6Xu8GCw+mePueonEJXFZbE+r2fb9UEpxUqcwbUJelQmTOgKj8FZekJKMtPQVV2sl6tt0WZEg59AhyGJDgMbeDQt4FDVy0MadyfRZWRneX8QCaD17Q3eS3T4ZRnfU+hg8GJKIQdyS3FX1nFgS7D70RJwps/H8Ox/HJolDK8NK4Pru3dJtBlEVGlqil0Z0aLznSiqxo54tS5+hFcdihLj0NtTncHpJJ0qM3HoCpxB6W6ONUmdyjSt4HdkAiHIREOw5mvndoYdphroNrWBNU1OsRmRS0bgxNRiFq/Pwf3frQdjhbcEaqNSYO3Jg1Ez0RuakvU3LxGiJyip+kCg1EDiC4oy09CXZIBdckxqMzpUJdkQGU+BlXpCQiS65wPdamMsJlSYTemwl45auQJSPo2bKrgI0EAlHJ3dzhlVZe46l+HeLtsaloMTkQhaOPBXPzjox1wuCT0aGNEbJg60CX5XbxJi7TLOyOmBT43omDg9KwvkmBzudwjSNWCEoNR/Qgum7sld0UulOU57o1bLblQWNxfK8tPQWXOgky0n/McokILm6k9bMYU2D2fU2EztXevJ+Ib+DpVTYGrapGtlMvOeYyooRiciELMr4fzcPeH22F3ibiqZzxen9gPCjmnYRCRt6rmC7bqo0bVptUxGJ2f4LJXbtTqDkSez5Yc955FVSGpnu25Rbka9rC2ntEjmynVE5KcujiGo/OQyQCVXAaFXAalXIBKLoNSLoNS4Q5EKrmM61+pWfgcnGw2G/78809kZmbCYrEgJiYG/fr1Q2pqalPUR0TV/H40H3e+vw12p4jLu8fhNYYmolZLrNqzqHKdkd3TiMEFu5Md6eokSZBbC6AqzYLanAmVOROq0izPZ6Ulp96nEmWqyhbccTU+O/TxsBtT4NC3YSe6WggCoFKcGRFSKSpDkVyo/MxRIgoe9Q5Ov/32G1599VX873//g8PhgMlkglarRWFhIWw2G9q3b4+7774b9957L8LCwpqyZqJWaUt6Iaa9tw02p4hLu8Zi0S39oGRoImrRaltjVPXZ2YLXN/qN6ICq7CRU5qyzgpH7a7mj/PwPlynhrB6CvIJRXOWeRXFwqcM5YlSL2kKRQiZAqZC5R5BkAv/4RyGlXsHpuuuuw44dO3DLLbdg7dq1GDhwILRaref2Y8eO4ddff8Unn3yCBQsW4IMPPsDll1/eZEUTtTbbMwsxddkWVDhcGNYpGv+9tT/UCv7lkijUVY0a2Zy1N2LgdLo6iE73OqLKjV3d+xkdh8qcBXVpJpRlp87beEGC4B4RCmsHu7Ed7GFtz3wOS4ZLE8lAdA4MRdQa1Ss4XXPNNfjiiy+gVNa+6WT79u3Rvn17TJ48GX///TdOnz7t1yKJWrOdx4sx+d2tKLe7cFHHKCydNBAaJUMTUahwVgtGVYGoat0RR43OT2Yvq9zU9aQnHCnLTkFV5t7LSFmefd5gBJxZW+QVjKq+NiRBUmia6dmEBpmssutcZSCq6jTHJgtEgCBJrevvWWazGSaTCSUlJTAajYEuh+i89p4swcSlf6DU6sSQ1Ei8N3UwtCqGJqJg5A5ELtgqg5HV4YLNIXKt0blIEuS2EqhKM92jRKXH3Ru8Vo0elZ2o1wavkqCAw5BQrVV3UmVAagt7WDs4dbHcxwhnRojODkVnt+RmkwVqbXzJBj43h2jfvj22bt2KqKgor+PFxcXo378/jh075uspiagWf58y49a3/0Sp1YlBKRF4d8oghiaiAJMkyROMbI6qkOSC1cFpdbURXHYoy05Wri3Kcn8uzXKHpNKsegUjp9oEh969qavDkOjew0h/5munNoZNFyop5AJUlVPlqn+umkpHRI3jc3DKyMiAy1VzWNxms+HkyZN+KYqotTuYXYrb3vkTJRUO9GsbjmVTB0Ov5u4BRM3J6nBVfpwZSeK6o7NU60znDkbHq4WjLCjLT0OQxPOewqGNgT2sLRxhyZUbu3oHJFHFhlNVZDJArZBBJZdDqRC8gpFaIeOGrURNrN7vxL755hvP12vWrIHJZPJ873K5sH79eqSkpPi1OKLW6EhuKW59+w8UltvRO8mE9+8YDANDE1GTqRpFqrC7UOGo/LC7GJCqSBIUlhyozJlQmzOgMmdAbU53d6czZ9TdmU6uqZw252644P5cOZXOkAxJqWumJxIaBMEdjtQKOdRKdyBSVX7PdUVEgVXvd2Njx44FAAiCgMmTJ3vdplQqkZKSgpdfftmvxRG1NkfzyjBx6Z/IL7OjRxsjPrxjCIya2puyEJHvRFGC1enyhKSqEaVWH5IkEQpLDtQlNYOR2pwBmbPi3A/1dKZzByJHZSCqCktObQw7051FEOAZJXKHIhnUSrlnBImIglO9g5MouofaU1NTsXXrVkRHRzdZUUStUUZ+OW5Z+gfySm3oGh+Gj6YNgUnH0ETUUKIoeY0gWSvXJLXakOQJR+lQmdOrhaQMqMyZkLms536oIHM3XTCmwG5Mgc2Y4u5MZ0plZ7rzqFpzVH3USK3gtDqiUOXz/J/09PSmqIOoVTteaMEtS/9AjtmGznEGfHznEEToVYEuiyhkOF2iJyTZHKLnc6sjSVBU5LuDkTkDqpJjlUEpA+qS9DrCkdw9lc4TjFJgN7m/dhiSIMn5b1J1VaNGSvmZfYxUchmUCvcxlZzhiKiladDCifXr12P9+vXIzc31jERVeffdd/1SGFFrcbK4AhOX/oFTJVZ0iNHj4zsvQJRBHeiyiIJWVRc7q+PMlLvWth+S3FpcOWqUftbnDMgdped8nDsctYXNlFq5n1GqOySZUmA3JAIyjnJXqd6+uyoInfnefYzBiKh18Tk4zZkzB88++ywGDhyIhIQE/qNB1AinSyow8a0/cKKoAqnRenxy1wWICWNoIgK8mzZUX5cktpaBJNEFVelxqEuOQF1c9XEU6pKjUNiKz/kwCQIchiTYTCnuYGRKhd2UCpsxFfawJIajStWDkapylEhd2a2uKhwREVXnc3BasmQJ3nvvPdx+++1NUQ9Rq5FrtuKWpX8iq9CCtpE6LL9rCGKNXCdArVNrbtogOK1Qlxx1h6Liw2cCkjkdMpftnI9z6OIrR45SYDO193xtD2vLNUc4E4yqps+p5FVfu0ePFAxGROQjn4OT3W7HhRde2BS1ELUaeaU2TFz6B9Lzy5EYrsXyu4YgwaQNdFlEzaK1rkeSW4u8gpGmchRJWXYCAmpPiKJcDZupA2zhHWELr/xs6gCbMYVtvOG94ataKYNaLveMHjEYEZG/+Ryc7rzzTixfvhxPPfVUU9RD1OKU2Zw4nFOKw7llns97TpSgoNyONiYNPr37AiRF8A0QtUw2pwtWu+g1mtTS1yMJDgs0RYegKToATeGBys8HobAWnPMxTnV4ZTjqCJupI2wRHWEN7wiHPhGQyZux+uAiCO5wpFbIPQHJ06VOLoOM+xoRUTPyOThZrVa89dZbWLduHXr37g2l0nuu9IIFC/xWHFEoKbU6cDi3DEdyynCoWlA6VVJ7F6s2Jg2W33UBkiMZmij01baJrLWlr0cSnVCbM6AuPABN0UFPSFKZs845gmQ3JFYLSB1gDe8EW3hHuDSRrXavo7P3NPJ8yNm2m4iCi8/Baffu3ejbty8AYO/evV638R83ag3MVgcO55ThSG4pDuWUeQLS6XMEJACICVOjc5wBnWLD0Knyc69EE7Sq1vuXZApdTpcIa1XTBkcr2B9JkqCw5HiFI03hQaiLD59zDZJDEw1bZBdYI7vCGtHV/Tm8U6udXlcVjqrvaVR95IjvH4goFPgcnDZs2NAUdVArJkkSCsrtyCwoR2aBBZkFFmQVWpBZUI6swgrYnK5Al3iGBJTanOe8OTZMjc5xYegYa0DnuKqQZEC4jvufUGg6e6qd1emCw9lSExIASYKy7AS0ebugy9sFbf4eaAr/PmcXO1GhhTWic2U46uIJSS5t69wkXqnwnlanVnLkiIhajgbt4wQAR44cwdGjR3HJJZdAq9VCkiT+o0jn5BIlnCqucAejwnJkVQakzEILsgrKUW4PonBUD3HGswJSrHsUyaRjm18KTdW72lUfTWqxo0iVFJZcaPN2QZu/uzIo7YbCWljjfpIgg82Y6h5FqhpBiugKu7EtILSuJgRymeAZLVIrZVAr5O6vGY6IqIXzOTgVFBRg/Pjx2LBhAwRBwOHDh9G+fXtMmzYNERERePnll5uiTjoPq8OFL3acQPl5RkKam0t071FUNXp0osgCx3kWhAsC0MakRbsoHdpF6dA2Ul/5WQe9usH5vklE6lUwaRmQKHSJooQKhwsW+5lRpNbQ1U5mK4Yubze0+buhzdsNbf4uqMpP17ifKFPCGtkVFdG9URHTBxVRPWAL79SqWnxXX3d0djhitzoiaq18fkf68MMPQ6lUIisrC926dfMcnzBhAtLS0hicAuCdTel4cc3BQJdRJ5VChuQILdpFuUNRu0gd2kXp0TZKh6QILdQKrvch8reqpg0WuwsWuxMV9ha+HqmS4LBAW7AX2rzd0OXvgjZvF9TmjBr3kyDAFt4RFTF9YInpg4roPrBGdm01IUkmQ2Ugkp/5rOToERFRbXwOTmvXrsWaNWuQlJTkdbxTp07IzMz0W2FUf2v/zgEADG0fhYTw4PifvQABsUa1Jxy1i9Ih3qhh61iiJmavnGZncTg9I0otPSRBdEBTdAjavJ3Q5e6CNn8XNEUHIUg1R9FsYW3do0jR7qBkje4JUakPQNHNSy4ToFHKoFa6A5Km8rOSo0dERPXmc3AqLy+HTlezK1BhYSHUarVfiqL6yy+zYfeJYgDAwpv7Is4YHMGJiJqeS5Q8o0iWVrJHEiQJKnOmOyTl7XSvTyrYW2t3O4cuzjOKVBHTGxXRveHSRASg6OajkAueUFQ9IHF6HRFR4/kcnIYNG4YPPvgAzz33HAB3C3JRFDF//nyMHDnS7wXS+f18MA+SBPRoY2RoImrhnC4R5TYXyuxOlNucrWJdkrt5w0732qS8ndDm74LCVlLjfi6VEZbo3qiI6euZdufUxweg4uZRvTmDxjO9Tg45R/WJiJqMz8Fp/vz5uOyyy7Bt2zbY7Xb885//xL59+1BYWIjffvutKWqk8/jpYC4A4NKusQGuhIj8zekSUW53odzmRFkrCEoyuxna/D1nWoHn7YKq/FSN+4lyNaxRPTyjSZbYvrAbU1pkdzuZDNAo5dAq5Z7PaoWM056JiALA5+DUs2dPHDp0CIsWLUJYWBjKyspwww034L777kNCQkJT1Ejn4HCJ+OVQHgBgRBcGJ6JQ5xIllNnco0nlNiesLTgoyeyl0Bbsq+xutxva/D1Qm9Nr3E+CAFtEZ3dIiukDS0xf2CK6QJK3vL3RVAoZNEqZOxxVhiSVouWFQSKiUNWgPs8mkwlPPPGEv2shH+3ILEKp1YlIvQp9k8MDXQ4R+UgUJc+0u6qg1BIbOcgc5dAU7IU2b497RCl/N9QlxyCg5pO1G5JhiTkz5a4iqidElSEAVTcdQXCPIlWFJE3lB6fZEREFtwYFJ6vVit27dyM3Nxei6P0X0euuu84vhVHdqqbpDe8cw//hEoUAUZRQbne61ynZnC1yg1l3G/B9nlEkbf5uqIuPniMkJbr3Soru5fnc0po3VDVr0FZNs1O6GzYQEVHo8Tk4rV69GpMmTUJ+fn6N2wRBgMvl8kthVLcNB9zBaSTXNxEFJUmSvNYotbjW4KITmqJD0OVuhy73r8qQdKTWNuB2fUK1gFQZkrRRASi66VSfaqdRuYMS230TEbUcPgenGTNmYNy4cXj66acRFxfXFDVRPZwosuBQThlkAnBJp+hAl0NEcAclS7WgZGlhQUleUQBd7o7Kj7+gzdsJudNS434OXaxXQKqI7gWnruX8gUcQ3CHJ07ChMiRx5J+IqGXzOTjl5OQgLS3Nb6Fp8eLFePHFF5GdnY0+ffrg9ddfx+DBg895/+LiYjzxxBNYtWoVCgsL0a5dOyxcuBBXX321X+oJFRsOuptCDGgXgXBdy1skTRQKJElChcNV2dDBHZhaTFASndAU7ocu96/KEaUdUJtrbnLuUhpgiemLitj+lV3uerWoNuDu9UiyM9PtVHJoFHJ2tSMiaoV8Dk433XQTNm7ciA4dOjT64itWrEBaWhqWLFmCIUOGYOHChRg9ejQOHjyI2Niaf5202+24/PLLERsbi5UrVyIxMRGZmZkIDw9vdC2hhtP0iAKjwu460/nO7oTYQhrfKSx50Ob9BV2Oe9qdLn8XZM6KGvezhneEJXYALLH9YIkdAFt4R0DWctbsqBQy6FTugKSrHEkSBIYkIiICBEny7e+jFosF48aNQ0xMDHr16gWlUul1+wMPPFDvcw0ZMgSDBg3CokWLAACiKCI5ORkzZszA448/XuP+S5YswYsvvogDBw7UuG59mc1mmEwmlJSUwGg0NugcgWZ1uND32bWwOkT88OAwdEsIzedBFAocLhGlVidKrQ6U2VpIUJJEqIsOQZ+9FfqcrdDlboeq9HiNu7lURlhi+sIS2x+WuAGwxPSBqA5v/nqbiEwGaJVy6FQK6NRy6JRyKLgmiYioVfElG/g84vTJJ59g7dq10Gg02Lhxo9df4gRBqHdwstvt2L59O2bNmuU5JpPJMGrUKGzevLnWx3zzzTcYOnQo7rvvPnz99deIiYnBLbfcgsceewxyee1/8bTZbLDZbJ7vzWZzveoLZpuPFcDqEJFg0qBrfFigyyFqcawOF8wVDpit7oYOoU5w2aDN2w19zhbosrdCn7MNcrv3v4Xu/ZI6uUNS5YctvGOL2lRWo5RVjiQpoFPJ2d2OiIh84nNweuKJJzBnzhw8/vjjkMka/j/U/Px8uFyuGmul4uLicODAgVofc+zYMfz000+49dZb8f333+PIkSOYPn06HA4HnnnmmVofM3fuXMyZM6fBdQaj6tP0OIWEqPGqut+ZKxwotTphd4b2sJLMboYuZzv02Vugz9kKbd4uyFw2r/u4FDp3QIofVDn1ri9EVcsZvZbLBOiqpttVhiU2byAiosbwOTjZ7XZMmDChUaGpoURRRGxsLN566y3I5XIMGDAAJ0+exIsvvnjO4DRr1iykpaV5vjebzUhOTm6ukv1OkiT8VBWcunB9E1FDuUQJpVZ3UDJbHSE9BU9Rnu0JSbrsrdAU7q+xb5JTE4Xy+MEojxsES/wgVET1AGQN2sovKKmV7rVJepXC3cCBo0lERORnPv9fc/LkyVixYgX+9a9/NerC0dHRkMvlyMnJ8Tqek5OD+PjaOzIlJCRAqVR6Tcvr1q0bsrOzYbfboVLV7C6nVquhVqsbVWswOZJbhhNFFVApZLioY8vaA4WoqdmdIsxWB8wVjpBuFa4yZ0B/6ndPWKptfZLNmFIZkgajPH4Q7MZUd4u4FkAQUDma5F6bpOdoEhERNQOfg5PL5cL8+fOxZs0a9O7du0aThgULFtTrPCqVCgMGDMD69esxduxYAO4RpfXr1+P++++v9TEXXXQRli9fDlEUPSNehw4dQkJCQq2hqSXacNA92nRB+yjoVC3nr8VETaXC7vKEJasjNIeVZLZiGE79DsPJXxF28pcaQUkSZLBGdkd5/CCUxw+GJW4gnLqWs8+eUiFApzwTkjRKGacpExFRs/P5nfeePXvQr18/AMDevXu9bvP1f2RpaWmYPHkyBg4ciMGDB2PhwoUoLy/H1KlTAQCTJk1CYmIi5s6dCwD4xz/+gUWLFuHBBx/EjBkzcPjwYTz//PM+dfILdVXT9C7tEhPgSoiCkyRJKLM5YbY6Ya5wwOkKvWElwWWHLncHDCd/heHkr9Dm74YgnQl9kqCAJa4/yuMvcAel2H4QVS2jUUzVvklVDRx0KgVUipbToIKIiEKXz8Fpw4YNfrv4hAkTkJeXh6effhrZ2dno27cvVq9e7WkYkZWV5bWWKjk5GWvWrMHDDz+M3r17IzExEQ8++CAee+wxv9UUzMxWB7ZlFAHg/k1E1VWtVzJXOFFqC8H1SpIEdfGhyqC0CfrTf0DutHjdxRreEWWJw1CWOAzl8RdAVBkCVKx/VY0mVd83iZvLEhFRMPJ5H6cqR44cwdGjR3HJJZdAq9VCkqSQmDoRyvs4fb/nNKZ/vAPtY/T46ZERgS6HKKAcLtHTMrzc5gy59UoKSx4MpzZ5RpWUFu/1ng5NNMoTL0Zp4jCUJV4Mpz4hQJX6T/V9k6qCkpL7JhERUQA16T5OBQUFGD9+PDZs2ABBEHD48GG0b98e06ZNQ0REBF5++eUGF07nd2aaHkebqHU6s7+SAxX20BpWElw26LK3IOzEz+7pd4X7vW4X5WqUxw/xjCpZI7uG9B5KVVPutCoFdEo5O90REVHI8zk4Pfzww1AqlcjKykK3bt08xydMmIC0tDQGpyYiihI2HjyzfxNRa1Fuc1Y2dwi9/ZWU5kyEndiIsOMbYTj9O2TOCq/bK6J6oqxyVMkSNwiSQhOYQv2AU+6IiKil8zk4rV27FmvWrEFSUpLX8U6dOiEzM9NvhZG3PSdLkF9mh0GtwKCUyECXQ9RkJElCqc2JEot7jyWXGDpz8ASnFfrTfyDsxAaEnfgZ6pJjXrc7dLEoTRyOsqThKGtzEVza0N1SQCEXYFAroFcroFfLoVZwNImIiFo2n4NTeXk5dDpdjeOFhYUtar+kYFPVhvzijtHsMEUtjii6w1LVNLyQae4gSVCZ0xF23B2U9Kc3Q+aynblZUKA8biDKkoejNGkkrJHdQnYvJblMgF4t94QlTrsjIqLWxufgNGzYMHzwwQd47rnnALhbkIuiiPnz52PkyJF+L5DcNlStb+I0PWohRFFCqdWJksqwFCrNHQSHBYbTv7un353YCHVpltftdn0CypJGoDR5JMraXAhRFVpNaKrIZIBe5Q5JBrV7Ch4REVFr5nNwmj9/Pi677DJs27YNdrsd//znP7Fv3z4UFhbit99+a4oaW728Uht2nSgBAIzg/k0UwqrahpdUuKfhhUpYUpWkw5i1DobjG6DP3gKZaPfcJsqUsMQPQmnSSJQmj4AtvHNIjioJAqBTuUeUDBoFtEp5SHRKJSIiai4+B6eePXvi0KFDWLRoEcLCwlBWVoYbbrgB9913HxISQr9dbjD6+VAeAKBnohGxxtBdPE6tk0uUYK5wh6WyUGkbLknQFOyFKWMNjJmroSk65HWz3ZCM0uQRKE0ajvI2F0FU6gNUaMMJAqBVnZl6p1cxKBEREZ2Pz8EJAEwmE5544gl/10LnsIFtyCnEOF0izJXT8EJmjyXRBX3OVhgzVsOYuRaqshOemyRBgbKEC1Da9lKUJY2AzdQhJEeVNEoZDJrK6XcqBbveERER+cDn4LRs2TIYDAaMGzfO6/jnn38Oi8WCyZMn+604cm/y+UvliNMIrm+iICaKEkoqHCgOobAkOK0wnNoEY8YaGLN+hMJa6LlNVGhRmjQC5najYW57KUR1eOAKbSCVwh2UDCp35zsFN5slIiJqMJ+D09y5c/Hmm2/WOB4bG4u7776bwcnPtmUUodTmRKRehT5J4YEuh6iGUqsDxRb3VLxQCEsyuxlhxzfAmLEaYSc2Qu4o99zmVIejtO0olKRcibLEYZAU2gBW6ruqFuFV0+/YgZOIiMh/fA5OWVlZSE1NrXG8Xbt2yMrKquUR1BhVm96O6BwDOafVUJCwOlwotjhQXGGHwxn8aUlhyUNY1lqYMtZAf+o3yESH5za7PsE9qpQyGuXxQwBZg2YwB4RMBq+gxBbhRERETcfndwixsbHYvXs3UlJSvI7v2rULUVGhu5ljsPqpcn3TSE7TowBzukQUVzhQbLGjwh78Gy0pLDkwHfsWpvTvoMvZDgFnAp7V1BHmlNEwp1yJiujeIbNeqaqhQ5hagTCNki3CiYiImpHPwWnixIl44IEHEBYWhksuuQQA8PPPP+PBBx/EzTff7PcCW7PjhRYczi2DXCbgkk5sQ07NT5IkmK1OFFvsIdE+XG4thjHje4Qf/Qb605u9wpIlpk/lyNKVsIV3DGCVvqmafmfUKGHQKDjyTEREFCA+B6fnnnsOGRkZuOyyy6BQuB8uiiImTZqE559/3u8FtmZV0/QGtI2ASacMcDXUmljsThRZHCixOOASgzstyRzlMGauhenoNzCc/MVrGl557ACUtB8Dc8qVcBjaBLDK+uOoEhERUXDyOTipVCqsWLECzz33HHbt2gWtVotevXqhXbt2TVFfq8ZpetSc7E4RxRV2FFscsDmCeyqe4LQi7MTPMB39GsasdZC5rJ7bKiK7o7jDdShpPwaOsOQAVll/HFUiIiIKfg1eBd25c2d07tzZn7VQNRV2F34/WgAAGNmV0/SoaThdIkoqN6ctt7kCXc75iU4YTv0O09GvYcpcA7nd7LnJZkxBcYfrUdL+OtgiOgWwyPrxjCppFAhTc1SJiIgoFDQoOJ04cQLffPMNsrKyYLfbvW5bsGCBXwpr7TYfy4fNKaKNSYMucWGBLodaEFflfkshsTmtJEKXsw3hx/4H47HvoLTme26y6xNQ0n4MijtcB2tUr6Bv8CCTAUaNkqNKREREIcrn4LR+/Xpcd911aN++PQ4cOICePXsiIyMDkiShf//+TVFjq7ThgHvT25FdYyEE+RtCCn4uUfLst1QW7GEJgKZgP8KPrILp2P+gKj/lOe7URKIk9RoUd7gOlrhBgBDc+xQp5AKMWiWMGnfLcP63TEREFLp8Dk6zZs3CzJkzMWfOHISFheGLL75AbGwsbr31Vlx55ZVNUWOrI0mSZ33TpVzfRA0kihJKrU4UV4RGRzyZrRjhR79BxKEV0OXv8Rx3KQ0wp4xGcfvrUZZ4ESAL7kYpaqXMPbKkVUCnCp09oYiIiOj8fP6/+v79+/HJJ5+4H6xQoKKiAgaDAc8++yyuv/56/OMf//B7ka3N4dwynCyugEohw9AO3BuL6q+qfbi5cipesIclSCL0p35D5KHPYMxYDZnLBgAQZUqUth2F4g5jUZo8EpJCE+BCz0+rksOodTd34Ca0RERELZPPwUmv13vWNSUkJODo0aPo0aMHACA/P/98D6V62lA52jS0fRT/Yk11kiQJZTYnii0OmK0OiMHdEA8AoCw9johDnyPi8Eqoyk54jldEdEVRlwko7vh/cGkiA1jh+QkCoFcrYNQoYNQqoZQH95RBIiIiajyf35VfcMEF2LRpE7p164arr74ajzzyCPbs2YNVq1bhggsuaIoaWx1O06P6cLpEFJTbUVBmD/q9lgB3C3FjxmpEHloBw6nfPMddKiOKO4xFYZfxQd3kQSYDwtTuKXhhGiWbOxAREbUyPgenBQsWoKysDAAwZ84clJWVYcWKFejUqRM76vlBSYUD2zKLAAAjuzA4UU1Ol4j8MjsKym3BP7okSdDm70bEoc8QfvRrrxbipW0uRlGXCTC3Gx20U/GqOuGZdEqEsbkDERFRq+ZzcGrfvr3na71ejyVLlvi1oNbu18N5cIkSOsTo0TZKF+hyKIhUBab8MlvQr12SVxQg/OiXiDj4GbRFBzzH7YYkFHUeh6JONwXt5rRymQCjVgGTVslOeEREROTBBTRBpqoNOafpUZWQCUySCMOJXxB58BOEZa2DTHQAAES5GuaUq1DYeTzK21wYlC3Eq9qGm7RK6FVyhiUiIiKqoV7BKSIiot5vJAoLCxtVUGsmihJ+PuRe3zSSwanVC5XAJLcWIuLQZ4jc/xHUpVme45bo3ijqPB7FHa6DqA4PXIHnoJALMGmVMFaOLBERERGdT73eLSxcuLCJyyAA2H2yBPlldhjUCgxsF7wdxahpOV0i8spsKCizB29gkiTocncgcv+HMKV/52kj7lIZUdTxBhR1uRnWqO4BLrImpaIyLGmU0DMsERERkQ/q9c5h8uTJTV0H4Uwb8mGdoqFSBN90JmpaoRCYZI5ymI5+haj9H0FbsM9zvCKqJwq6T0Jx++sgKYNrbZ5KIfOsWWJ7fyIiImqoBr2LcLlc+PLLL7F//34AQPfu3XH99ddDoeCbksbYcJDT9Fojh0tEfpAHJnXRIUTu/wgRh7+A3FEKwL12qaT9GBR0m4SKmD5B1UZcEACTVolIvYojS0REROQXPr+j2LdvH6677jpkZ2ejS5cuAIB58+YhJiYG//vf/9CzZ0+/F9ka5JZasftECQBgRJeYAFdDzSHYA5PgssOYuQaRf38IQ/YfnuM2YwoKu92Ook7j4NKEB67AWujUckToVAjXKiHjPktERETkRz4HpzvvvBM9evTAtm3bEBERAQAoKirClClTcPfdd+P333/3e5GtwcaD7m56vRJNiA0Lzj1tyD+sDhcKyu0oKg/OwKQsO4XIA8sRcfATKCvcr0tJkMHc9nIUdrsdZYkXB1VnPIVccIclnRIapTzQ5RAREVEL5XNw2rlzp1doAtxd9/7zn/9g0KBBfi2uNdnIaXotXqnVgYIyO0qtzkCXUpMkwnDyV0Tu/xDGrHUQJPfOug5tDAq73oKiLhPhMLQJcJFnCAIQplEgQq/ixrRERETULHwOTp07d0ZOTg569OjhdTw3NxcdO3b0W2GticMl4tdD+QC4f1NLI4oSiiscKCizweoQA11ODTJ7GSIOf46ofe9BbU73HC9LuBAF3W6DOWU0IFMGsEJvaqUMEToVInRKKOTBM+pFRERELZ/PwWnu3Ll44IEHMHv2bFxwwQUAgD/++APPPvss5s2bB7PZ7Lmv0Wj0X6Ut2NaMQpTanIjSq9A70RTocsgPHC4RheV2FJTZ4RKDbz6eypyBqL/fR8TBzzzNHlzKMBR1vgmFXW+DLaJTgCs8QyY70+iBXfGIiIgoUHx+F3LttdcCAMaPH++ZHiNVLtQYM2aM53tBEOByufxVZ4tWtb5peJcYLmgPcRV2F/LLbCipcATf+iVJgv7074je+w7CstZDgLtAq6kDCnpMRXGnGyEq9QEu8gx9ZaMHExs9EBERURDwOTht2LChKepo1X6q3L+J0/RCV0nldLxyW/D9sUBwWhF+ZBWi9y2Dpuig53hp0gjk97gDZUmXBE2zB0EAwnVKRBvUbPRAREREQcXn4DR8+PCmqKPVOl5owZHcMshlAoZ1YhvyUCKKEgot7ul4dmfwrV9SlJ9G1N/vI/LAcihsxQAAl0KH4k43oaDHFNjCg2dNokIuIEqvQqRexbVLREREFJR8Dk6rV6+GwWDAxRdfDABYvHgxli5diu7du2Px4sVe3faoblWjTQPaRcCkDZ5F+HRudqeIgnIbCsvtEIMtL0kSdLnbEbX3XZgyfoAguUfA7IZkFHSfjMIuEyCqg2cdnVYlQ7RBDZNWyc54REREFNR8/tPuo48+6mkAsWfPHqSlpeHqq69Geno60tLS/F5gS7fhIKfphQqHS8TxQgsO5ZQivzS4QpPgsiP88Cp0+HoMOvzvBoSnfwtBcqEs/gJkjnoLB8f/gvzedwdFaBIEwKhVoH2MHh1jwxCuUzE0ERERUdDzecQpPT0d3bt3BwB88cUXGDNmDJ5//nns2LEDV199td8LbMkq7C5sPloAgMEpmImihLwyG/JKbUHX8EFekY+o/R8hcv+Hns1qRbkaxR2uR0GPqbBG9ajjDM1HJgMi9SpE6dVQKTgdj4iIiEKLz8FJpVLBYrEAANatW4dJkyYBACIjI71akVPdfj+aD5tTRGK4Fp1iDYEuh84iSRKKLA7kmK1wuoIrMcnsZsTsWoLofe9A5qwAADh0sSjoNgmFXW+FSxsV4ArPUClkiDKoEKFTQc7ueERERBSifA5OF198MdLS0nDRRRdhy5YtWLFiBQDg0KFDSEpK8nuBLVnVNL2RXWM4VSnImK0O5JRYg27TWsFpRdTf7yNm12JPwwdLdC/k97wL5tSrIclVgS2wGr1ajqjK9UtEREREoc7n4LRo0SJMnz4dK1euxBtvvIHExEQAwA8//IArr7zS7wW2ZNf3TYRSLsPoHvGBLoUqWR0unC6xoszqDHQp3kQnIg59jrgdr0BpyQYAWMM7ImfgYzC3u8K9cCgICIJ7s9qYMLYTJyIiopZFkKRgW7XRtMxmM0wmE0pKSmA0GgNdDgUJh0tEdokVxRZHoEvxJkkwZnyP+G0vQl1yDABg17dBzoBHUNzxBkAWHOFEJgOi9GpEGVRQsp04ERERhQhfskG9RpzMZrPnRHWtY2IYoVASzI0f9Cd/RfzWedDl7wYAODWRyO17Pwq73gZJoQlwdW5KhYBogxqROhVkXL9ERERELVi9glNERAROnz6N2NhYhIeH17oeR5IkCIIAl8vl9yKJmkJhuT0oGz9o83YhfusLMJz6DQDgUuqR3/Mu5Pe6C6IqLMDVuXH/JSIiImpt6hWcfvrpJ0RGRgIANmzY0KQFETW1UqsD2UHY+EFdfARx216EKeMHAIAoU6Kw2+3I7Xs/XNroAFfnFqZRIDpMDYPa5+WRRERERCGNa5yo1QjWxg+K8tOI2/4KIg5/BkESIUFAcacbkdP/YTjCkgNdHhs+EBERUYvl9zVOu3fvrvfFe/fuXe/7EjUHUZSQbbaisNweVOuY5NYixOz6L6L+fg8ylw0AUNLuCuQMeBS2yC4Bro4NH4iIiIiqq1dw6tu3LwRB8KxjOh+ucaJgUmZz4mRRBezO4JmWJzgsiN77DmJ2L4HcUQoAKI8fjOxBj8MSNzDA1bHhAxEREVFt6hWc0tPTPV//9ddfmDlzJh599FEMHToUALB582a8/PLLmD9/ftNUSeQjlyjhdEkFisqDqL246ETEoRWVezG5Nz+uiOyO7EGPoSxpRMD3YmLDByIiIqJzq1dwateunefrcePG4bXXXsPVV1/tOda7d28kJyfjqaeewtixY/1eJJEvzFYHThVXwOEMknl5kgRj5lrEbZ0HTckRAIA9LBnZAx5FSYfrACGw0+A0ShniTBoYNcqA1kFEREQUzHxujbVnzx6kpqbWOJ6amoq///7bL0URNYRLlHCquCKoNrHV5WxD/Jbnoc/ZBgBwqiOQ2+8BFHa7DZJcHdDaVAoZ4oxqhOtUAa2DiIiIKBT4HJy6deuGuXPn4u2334ZK5X7DZbfbMXfuXHTr1s3vBRLVR0mFe5QpWPZkUhcfQdzWeTBlrgEAiHIN8nveibw+90JUBbabo0IuIDZMjUi9ilPyiIiIiOrJ5+C0ZMkSjBkzBklJSZ4Oert374YgCPjf//7n9wKJzsfpEnGq2IqSiuAYZVJYchC74xVEHlwBQXJBEmQo6jweOf3T4NTHB7Q2uUxATJgaUXo2fSAiIiLyVYP2cSovL8fHH3+MAwcOAHCPQt1yyy3Q6/V+L9DfuI9Ty1FsseNUsRUuMfCjTDJ7KWJ2v4novUshc1YAAMxtL0f2oMdgi+gc2NpkQLRBjWiDGnIGJiIiIiIPv+/jdDa9Xo+77767QcURNZbDJeJUcQXMFYHfyFZw2RF54GPE/vUqFNZCAEB57ABkD54FS/zgwNYmAFEGFWIMaii4DxMRERFRozQoOBEFSlG5HadKKiAGelsmSYTp2LeI2/4i1OZMAIDN1B7Zgx6Hud3ogLYWFwQgXKdEbJgGKgUDExEREZE/MDhRSLA7RZwsrkCZNfCjTPpTvyN+y/PQ5e8GADi0Mcjt/xAKu9wMyALb0tukVSLWqIZGKQ9oHUREREQtDYMTBb2CMhuyzdaAjzIpyrORuOlxGI//BABwKfXI730v8nveCVEZ2PV9YRoF4owaaFUMTERERERNgcGJgpYoSjheZAmKtUzG9O+QuOlxKGwlkAQFCrrdhtx+D8CljQ5oXVqVHAkmDfRq/qdMRERE1JR8frc1efJkTJs2DZdccklT1EMEwD01L7OgHFZHYIeZZPZSJGyejcjDnwMALNG9cGLEq7CFdwxoXUqFgHijhpvXEhERETUTn4NTSUkJRo0ahXbt2mHq1KmYPHkyEhMTm6I2aqXKbU5kFlgC3mZcl7MNyRsfhKr0OCRBhrw+9yGn/0MBXcckkwExYWpE69Xci4mIiIioGfnccuurr77CyZMn8Y9//AMrVqxASkoKrrrqKqxcuRIOR3BsQkqhq7DcjvT88sCGJtGB2O0vo/23N0FVehx2QxKOXbMCOQMfDWhoitAr0TkuDLFhGoYmIiIiombWoA1wq9uxYweWLVuGt99+GwaDAbfddhumT5+OTp06+atGv+IGuMFJkiRkm63IL7UHtA5VSTqSNz4IXd5OAEBRxxtw6sJnIaoC91rRq+VoE65lpzwiIiIiP/MlGzRqk5fTp0/jxx9/xI8//gi5XI6rr74ae/bsQffu3fHKK6805tTUirhECZkFlsCGJklCxIHl6PTlldDl7YRLZUTWyEU4MWJhwEKTWilD2ygd2scYGJqIiIiIAsznNU4OhwPffPMNli1bhrVr16J379546KGHcMstt3hS2pdffok77rgDDz/8sN8LppbF5nQhs8ACWwCbQMgrCpC46TGYMtcCAMoSLsSJ4QvgMLQJSD0yGRBn1CBKr4IQwI10iYiIiOgMn4NTQkICRFHExIkTsWXLFvTt27fGfUaOHInw8HA/lEctWZnNiawAN4EwHN+IpF8egbIiD6JMhZyB/0R+rzsBoVGDsQ0iCECkXoXYMDUU8ua/PhERERGdm8/B6ZVXXsG4ceOg0WjOeZ/w8HCkp6c3qjBq2QrKbDhdYkXjVtg1nOC0In7L84j++z0AgDW8E46PfB3WqO4BqSdMo0C8ScMpeURERERByufgdPvttzdFHdRKSJKEUyVWFJYFbj2TJn8vkjc+AE3xEQBAfo+pyB40C5Li3H8MaLJalDIkhGth4Aa2REREREGtXu/WbrjhhnqfcNWqVT4XsXjxYrz44ovIzs5Gnz598Prrr2Pw4MF1Pu7TTz/FxIkTcf311+Orr77y+brUvFyihKxCC8qszsAUILoQvectxG1/CTLRAYc2BieGL0BZ0vBmL0UuExBv0iBSzw1siYiIiEJBvRZSmEwmz4fRaMT69euxbds2z+3bt2/H+vXrYTKZfC5gxYoVSEtLwzPPPIMdO3agT58+GD16NHJzc8/7uIyMDMycORPDhg3z+ZrU/KwOF47klgUsNCnLTiL1h1uQsHUuZKIDJe1G4/CNPwYkNIXrlOgcZ2BoIiIiIgohPu/j9Nhjj6GwsBBLliyBXO5ej+FyuTB9+nQYjUa8+OKLPhUwZMgQDBo0CIsWLQIAiKKI5ORkzJgxA48//nitj3G5XLjkkktwxx134Ndff0VxcXG9R5y4j1PzK7U6kFVogRigxnmG4xvQdsMMyO1muBQ6nB46G0WdJ7i7MTQjpUJAm3AtjJrAbaJLRERERGc06T5O7777LmbOnOkJTQAgl8uRlpaGd99916dz2e12bN++HaNGjTpTkEyGUaNGYfPmzed83LPPPovY2FhMmzatzmvYbDaYzWavD2o++WU2ZBYEKDRJEmJ2LkLKmimQ282wxPTFkf9bjaIuNzd7aIo0qNApNoyhiYiIiChE+bwi3el04sCBA+jSpYvX8QMHDkD08d1xfn4+XC4X4uLivI7HxcXhwIEDtT5m06ZNeOedd7Bz5856XWPu3LmYM2eOT3VR40mShJPFFSgqdwTk+jJ7GZJ+eQSmjB8AAAVdb8PpobMhyZt3epxaKUNiuBZ6Nn8gIiIiCmk+v5ubOnUqpk2bhqNHj3oaOPz555944YUXMHXqVL8XWF1paSluv/12LF26FNHR0fV6zKxZs5CWlub53mw2Izk5ualKJLhDU2aBBaUBWs+kKklHux/vhKb4MESZCqcufA5FXSc2aw2CAEQb1IgzqrmJLREREVEL4HNweumllxAfH4+XX34Zp0+fBuDeFPfRRx/FI4884tO5oqOjIZfLkZOT43U8JycH8fHxNe5/9OhRZGRkYMyYMZ5jVaNcCoUCBw8eRIcOHbweo1aroVarfaqLGi7QoSns+E9I3vAA5HYzHLpYZI56CxWx/Zu1Bq1KhqQIHfdkIiIiImpBfG4OUV3VeqHGNFkYMmQIBg8ejNdffx2AOwi1bdsW999/f43mEFarFUeOHPE69uSTT6K0tBSvvvoqOnfuDJXq/FOx2Byi6QQ0NFWuZ4rb/hIESCiPHYCsUUvg1MXV/Vg/EQQgzqhBtEHFUSYiIiKiEOBLNmjwwou8vDwcPHgQANC1a9d6T507W1paGiZPnoyBAwdi8ODBWLhwIcrLyz3T/iZNmoTExETMnTsXGo0GPXv29Hp8eHg4ANQ4Ts1LkiRkFARmj6ZgWM+kV8uRGKGFWsFRJiIiIqKWyOfgVF5ejhkzZuCDDz7wTJOTy+WYNGkSXn/9deh0Op/ON2HCBOTl5eHpp59GdnY2+vbti9WrV3saRmRlZUEm87n5HzUjUZSQGaCNbQO9nkkmAxJMWu7JRERERNTC+TxV75577sG6deuwaNEiXHTRRQDcne4eeOABXH755XjjjTeapFB/4VQ9/wpkaAr0eiajVoE24Voo5Qz2RERERKHIl2zgc3CKjo7GypUrMWLECK/jGzZswPjx45GXl+dzwc2Jwcl/AhaaAryeSSEX0MakhUnHPZmIiIiIQlmTrnGyWCw19l0CgNjYWFgsFl9PRyEqUKGpxnqmbrfj9AXPNNt6pjCNAkkRWig4ykRERETUqvj87m/o0KF45plnYLVaPccqKiowZ84cDB061K/FUXAKVGhSlaSjwzfXw5TxA0SZCieGzcepi/7TbKEpyqBCSrSeoYmIiIioFfJ5xOnVV1/F6NGjkZSUhD59+gAAdu3aBY1GgzVr1vi9QAougQpN3uuZ4pA56s1mW88kCECCSYMoA/cDIyIiImqtGrSPk8Viwccff4wDBw4AALp164Zbb70VWq3W7wX6G9c4NVxAQtPZ65niBiLrsiVw6mKb5fIyGdA2UocwDdczEREREbU0Tb6Pk06nw1133dWg4ig0BSI0Cc4KJG98KGDrmZQKASlRemiU3JuJiIiIqLVrUHA6ePAgXn/9dezfvx+Ae8Tp/vvvR9euXf1aHAWHgIQmhwUpa++A4fTv7v2ZLvo3irrc3GzX16rkaBelY6txIiIiIgLQgOYQX3zxBXr27Int27ejT58+6NOnD3bs2IFevXrhiy++aIoaKYACEZpk9jKkrpkEw+nf4VIakH71x80amkxaJdpH6xmaiIiIiMjD5zVOHTp0wK233opnn33W6/gzzzyDjz76CEePHvVrgf7GNU71F5jQVIqU1ZOgz90OlzIM6Vd92Kyb2saEqRFv0jTb9YiIiIgocHzJBj7/Sf306dOYNGlSjeO33XYbTp8+7evpKEgFJDTZSpD6w23u0KQyIv3q5c3aOS8xQsvQRERERES18jk4jRgxAr/++muN45s2bcKwYcP8UhQFVmBCUzFSf7gVury/4FSH49jVn6Aipk/zXFsGpETrEalvnqYTRERERBR66tUc4ptvvvF8fd111+Gxxx7D9u3bccEFFwAA/vjjD3z++eeYM2dO01RJzSYQoUluLULqD7dCW7AXTk0k0q9aDmtU92a5tkohQ7soHTvnEREREdF51WuNk0xWv4EpQRDgcrkaXVRT4hqn88sqsKCkwtFs15NXFCD1h1ugLdwPhyYa6Vd/Altkl2a5tk4tR7tIHRRsAkFERETUKvl9HydRFP1SGAW3/DJbs4YmhSUPqT9MhKboEBzaGKRf/SlsEZ2a5drhOiWSIrQQBKFZrkdEREREoa1B+zhRy1NhdyG7xNps11NYcpD6/URoio/AoYvDsas/hT28Q7NcO9aoRpyRTSCIiIiIqP7qNUfp008/rfcJjx8/jt9++63BBVHzc4kSsgot8K0xfcMpyrPR/rsJ0BQfgV2fgGPXfN4soUkQgORILUMTEREREfmsXsHpjTfeQLdu3TB//nzs37+/xu0lJSX4/vvvccstt6B///4oKCjwe6HUdE4UWWB3Ns90TGXZKbT/bhzUJcdgNyTi2DWfwW5KafLrymUCUqP1CNexcx4RERER+a5eU/V+/vlnfPPNN3j99dcxa9Ys6PV6xMXFQaPRoKioCNnZ2YiOjsaUKVOwd+9exMXFNXXd5Cf5ZTaYK5qng56y9ATafz8BqtLjsIcl49jVn8IRltzk11XI3aGJnfOIiIiIqKHq1VWvuvz8fGzatAmZmZmoqKhAdHQ0+vXrh379+tW7+14gsaveGRa7E8fyyptlip7SnIn230+EquwEbMZ2SL96BRyGNk1+XZVChpRoHdQKhiYiIiIi8ub3rnrVRUdHY+zYsQ2tjYJEc65rUpVkIPX7m6EqPwWbqT2OXf0pnPr4Jr+uWilDarQeSrYbJyIiIqJGYle9VupEkQUOZ9OnJlXJMbT/bgKUlhxYTR2Rfs0ncOqafiqnViVDSpSeezQRERERkV8wOLVCeaXNs65JXXQYqd/fDGVFHqwRnZF+1Sdw6mKa/Lo6tRwpUXrIZdyjiYiIiIj8g8GplbHYncgxN/1+Teqig0j9biKU1nxURHZD+lXL4dJGNfl1DRoF2kXqIGNoIiIiIiI/YnBqRZprXZO6+AjafzcBCmshKqJ6uEOTJqJpLwrApFUiOVILQWBoIiIiIiL/avQCEJfLhZ07d6KoqMgf9VATOl7Y9Oua5BUFSFkzGQprISzRvZB+1SfNEprCdQxNRERERNR0fA5ODz30EN555x0A7tA0fPhw9O/fH8nJydi4caO/6yM/ySu1odTatOuaBKcV7X68E6rS47CFtUXG6A/g0oQ36TUBIMqgQnKkjqGJiIiIiJqMz8Fp5cqV6NOnDwDgf//7H9LT03HgwAE8/PDDeOKJJ/xeIDVes6xrkkQk/ZwGfe52ONUmZIx+v1nWNMUa1WgTrm3y6xARERFR6+ZzcMrPz0d8vHsPnu+//x7jxo1D586dcccdd2DPnj1+L5Aax+kSm2VdU9y2FxGe/i1EmRJZl70Fe3iHpr0ggHiTBnFGTZNfh4iIiIjI5+AUFxeHv//+Gy6XC6tXr8bll18OALBYLJDL5X4vkBrnRFFFk69riji4ArG7FgMATg6bh/I2Q5v0egCQGKFFTJi6ya9DRERERAQ0oKve1KlTMX78eCQkJEAQBIwaNQoA8Oeff6Jr165+L5AaLrfU2uTrmvQnNyFx0ywAQE6/B1Hc6aYmvZ4gAEkRWoTrVE16HSIiIiKi6nwOTrNnz0bPnj1x/PhxjBs3Dmq1+6/+crkcjz/+uN8LpIYptzmRa7Y16TXURYfQbv29ECQnijuMRW7/tCa9niAAbaN0MGqUTXodIiIiIqKzCZLU+NUvxcXFCA8P90M5Tc9sNsNkMqGkpARGozHQ5TQJp0vEkbyyJp2ip7DkocM3Y6EqO47yuEFIv+pjSIqmW28kCEBKtB4GNbceIyIiIiL/8CUb+LzGad68eVixYoXn+/HjxyMqKgpJSUnYvXu379WS3zX1uibBWYF2P06Dquw4bMYUZF6+tElDk1wmoH0MQxMRERERBY7PwWnJkiVITk4GAPz444/48ccf8cMPP+DKK6/EzJkz/V4g+abJ1zVJIpI3PgRd3k441eHIGP0eXJrIJrucQu4OTToVQxMRERERBY7P70azs7M9wenbb7/F+PHjccUVVyAlJQVDhgzxe4FUf82xrilu6zyYMn6AKFMic9RS2E3tm+xaggCkROmhUbJbIxEREREFls8jThERETh+/DgAYPXq1Z6uepIkweVy+bc6qrfm2K8p4sByxO5+AwBwctiLsCQ0XVAWBKBdlA5aFUMTEREREQWezyNON9xwA2655RZ06tQJBQUFuOqqqwAAf/31Fzp27Oj3Aql+ss1WOF1Nl5oMJ35B4m9PAABy+j+M4k43NNm1AKBNuBZh7J5HREREREHC5+D0yiuvICUlBcePH8f8+fNhMBgAAKdPn8b06dP9XiDVzeESUWxxNNn51YUH0Xb9PyBILhR1/D/k9nuoya4FALFGNSL13KeJiIiIiIKHX9qRh5KW2I48x2xtsrVNCksuOnxzPVRlJ1EeP9jddlyubpJrAUC4TonkSF2TnZ+IiIiIqEqTtiMHgA8//BAXX3wx2rRpg8zMTADAwoUL8fXXXzfkdNQIoiihoMzeJOc+03b8JGzGVGSOWtqkocmgUSApQttk5yciIiIiaiifg9Mbb7yBtLQ0XHXVVSguLvY0hAgPD8fChQv9XR/Vochih0tsgkFDSUTyhgehy9sFpzoCGaPfh0sT4f/rVNKqZGgbqYMgCE12DSIiIiKihvI5OL3++utYunQpnnjiCcjlZzqeDRw4EHv27PFrcVS3/CYabYrfMhemzNUQZSpkXr4UdlNKk1wHAJQKAe2i9JDLGJqIiIiIKDj5HJzS09PRr1+/GsfVajXKy8v9UhTVT0mFA3an6PfzRu7/EDF73gQAnLjkJVjiB/v9GlVkMvdeTUp5g2aNEhERERE1C5/fraampmLnzp01jq9evRrdunXzR01UT/ll/m8IYTjxM9r8/jQAIGfAIyjpONbv16jCDW6JiIiIKFT43I48LS0N9913H6xWKyRJwpYtW/DJJ59g7ty5ePvtt5uiRqqFxe6ExebfDYfVRYfPtB3vdCNy+z7g1/OfLTlCB73a55cgEREREVGz8/ld65133gmtVosnn3wSFosFt9xyC9q0aYNXX30VN998c1PUSLXIL/Xv2ibBZUfyxgcgd5ShLP4CnLx4nntIqInEmzQw6bjBLRERERGFhkbt42SxWFBWVobY2Fh/1tSkWsI+TjanC4eyy/x6zrit8xC7azGc6ggcvnEtnLo4v56/uiiDCm3C2XaciIiIiALLl2zQqHlSOp0OOh03K21u/t63SZe9BTG73wAAnLz4hSYNTUatgqGJiIiIiEKOz80hcnJycPvtt6NNmzZQKBSQy+VeH9S0XKKEwnL/BSeZvRTJGx+GIIko7DQO5tSr/Hbus2lVciRHMGgTERERUejxecRpypQpyMrKwlNPPYWEhARuWNrMCsptaPjkypoS/ngWqrLjsBuScHrobP+d+CxqpQwpUTrIuFcTEREREYUgn4PTpk2b8Ouvv6Jv375NUA6djyRJfp2mZ8xYg8hDKyBBwPERr0BUhfnt3NXJZQLaRemg4F5NRERERBSifH4nm5ycjEb0k6BGKLY44HT552evsOQicdNjAIC83vfCEj/EL+c9myAAKdE6qBWcxklEREREocvn4LRw4UI8/vjjyMjIaIJy6Hz8tuGtJCHx139CYS1ERWR35A5I8895zyIIQNsoHXQq7tVERERERKHN53e0EyZMgMViQYcOHaDT6aBUeu/FU1hY6Lfi6IxSqwNWh+iXc0Ue+BjG4z9BlKtxfMSrkORqv5z3bAkmDYwa7tVERERERKHP5+D0yiuvsCFEAOT7aW2TqiQdCX8+BwDIHvgYbJFd/HLes0WHqRBlaJpARkRERETU3BrUVY+al9XhQpnV2fgTiU4kb3wIMmcFyhIuREHPOxp/zlpoVXLEGzVNcm4iIiIiokDweY2TXC5Hbm5ujeMFBQXcx6mJ5JX6Z21T7M5F0OX9BZfKiBPDFwCC/7vcyWUC2kbqOCpJRERERC2Kz++cz9VRz2azQaVSNbog8uZwiSipcDT6PNq8XYj961UAwMkL/w2HoU2jz1mbpEgtVAq2HSciIiKilqXeU/Vee+01AIAgCHj77bdhMBg8t7lcLvzyyy/o2rWr/yts5fLLGr/hreCsQPLGByFILhS3H4OSDtf7p7izRIep2AyCiIiIiFqkegenV155BYB7xGnJkiVe0/JUKhVSUlKwZMkS/1fYirlECYXljW8KkfDnf6AuOQaHLh6nLvyPu0+4n3FdExERERG1ZPUOTunp6QCAkSNHYtWqVYiIiGiyositsNwOsZEdyA3HNyBq/wcAgBOXvASXJrzxhZ2F65qIiIiIqKXzuavehg0bmqIOOoskSSgob1xTCLm1CEm/zAQA5PeYirKkS/xRWg1c10RERERELV29glNaWhqee+456PV6pKWlnfe+CxYs8EthrV1JhQMOZyMWN0kSEjc9DmVFHqzhHZE9aJb/iquG65qIiIiIqDWoV3D666+/4HA4PF+fC6dq+U9+WeNGm8KPfAFTxg+QBAWOj3gVksL/6490aq5rIiIiIqLWoV7BacOGDTh27BhMJhOn6jWDMpsTFfaGL25Slh5Hm9+fBgDkDEiDNbqXv0rzkMsEJEdwXRMRERERtQ71XpjSqVMn5OXleb6fMGECcnJymqSo1i6/MRveii4k/ZwGuaMM5bEDkNf7H/4rrBquayIiIiKi1qTe73zP3vj2+++/R3l5ud8Lau2sDhdKrc4GPz5671IYsv+ES6nH8RELAZm8zsf4fA2uayIiIiKiVoZDBkGmMWubNAV/I27biwCA0xc8A4exnb/K8uC6JiIiIiJqjeodnARBqLGehetb/MvpElFscTTosYLTiuSND0ImOlDS7goUdZ7g5+q4romIiIiIWq967+MkSRKmTJkCtVoNALBarbj33nuh1+u97rdq1Sr/VtiKFJTbITWwA3nc9pegKToIhyYaJy9+AWiCcMN1TURERETUWtX7XfDkyZMRGxsLk8kEk8mE2267DW3atPF8X/XREIsXL0ZKSgo0Gg2GDBmCLVu2nPO+S5cuxbBhwxAREYGIiAiMGjXqvPcPFaIooaDM3qDH6rL/RPSepQCAk5fMh0sb7c/SAAAxYWquayIiIiKiVqveI07Lli1rkgJWrFiBtLQ0LFmyBEOGDMHChQsxevRoHDx4ELGxsTXuv3HjRkycOBEXXnghNBoN5s2bhyuuuAL79u1DYmJik9TYHIosdrjEhg03xf71GgRIKOw8AaVtR/m5Mve6pjij2u/nJSIiIiIKFYJ0dru8ZjZkyBAMGjQIixYtAgCIoojk5GTMmDEDjz/+eJ2Pd7lciIiIwKJFizBp0qQ67282m2EymVBSUgKj0djo+v3lUE4pbA7f925SFx5A51VXQBJkODj+VzjCkv1al1wmoGOsgVP0iIiIiKjF8SUbBPTdsN1ux/bt2zFq1JlREplMhlGjRmHz5s31OofFYoHD4UBkZGStt9tsNpjNZq+PYFNS4WhQaALgmaJXknKV30MTwHVNRERERERAgINTfn4+XC4X4uLivI7HxcUhOzu7Xud47LHH0KZNG6/wVd3cuXO91mAlJ/s/XDRWQ1uQKyw5CD/6lfscve7yY0VuXNdEREREROQW0kMJL7zwAj799FN8+eWX0Ghq31to1qxZKCkp8XwcP368mas8P4vdCYvN1aDHRv39IWSiA+WxA1AR29+vdXFdExERERHRGfVuDtEUoqOjIZfLkZOT43U8JycH8fHx533sSy+9hBdeeAHr1q1D7969z3k/tVrtaaEejMqszgY9TnBWIHL/hwCA/F53+rMkyGUC2kZyvyYiIiIioioBHXFSqVQYMGAA1q9f7zkmiiLWr1+PoUOHnvNx8+fPx3PPPYfVq1dj4MCBzVFq0Ik4/AUUtiLYw5JhbnelX8+dHKmFUh7Sg5FERERERH4V0BEnAEhLS8PkyZMxcOBADB48GAsXLkR5eTmmTp0KAJg0aRISExMxd+5cAMC8efPw9NNPY/ny5UhJSfGshTIYDDAYDAF7Hs1KEhG19x0AQH6POwCZ3G+njg5TIYzrmoiIiIiIvAQ8OE2YMAF5eXl4+umnkZ2djb59+2L16tWehhFZWVmQyc6Mfrzxxhuw2+246aabvM7zzDPPYPbs2c1ZesCEHd8ATclRuJRhKOo8wW/nVSlkiAurfa0YEREREVFrFvB9nJpbsO3jlGu2IsfsW1e91O9vhuHU78jrdQ+yhzzht1pSY/QwqAOepYmIiIiImkXI7ONEvtMU7IPh1O+QBDkKekzx23kjDSqGJiIiIiKic2BwCjGeDW9Tr4HDkOiXcyoVAuKNnKJHRERERHQuDE4hRFGejfCj3wDw74a3bcK1kMvYepyIiIiI6FwYnEJI1N/vQ5CcKI8fjIqYPn45Z7hOCSO76BERERERnReDU4gQHBZEHvgIAJDf0z+jTXKZgAQTp+gREREREdWFwSlERBxeCYWtBDZjO5jbjvLLOduEa6DgRrdERERERHXiu+ZQIImI3vs2AKCgxzS/bHgbplEgXKdq9HmIiIiIiFoDBqcQEJa1DmpzBlwqI4o6j2v0+WQyd0MIIiIiIiKqHwanEBC9p3K0qeutEJX6Rp8v3qiBSsFfPRERERFRffHdc5DT5O+GIfsPSILCLxve6tVyRBnUjS+MiIiIiKgVYXAKcjGVo03F7cfAqU9o1LkEAUiM4BQ9IiIiIiJfMTgFMWXZKZiOfQsAyO91Z6PPF2tUQ61ofGMJIiIiIqLWhsEpiEX9/R4EyYmyhKGwRvdq1Lm0KhliOEWPiIiIiKhBGJyClMxRjsgDywEA+T0bN9okCEBShA6CIPijNCIiIiKiVofBKUhFHPoMcrsZNmMqStte1qhzxYSpoVFyih4RERERUUMxOAUj0YWove8AqBxtEhr+a1IrZYgN4xQ9IiIiIqLGYHAKQsasH6EuzYJTHY6iTjc26lxJEVpO0SMiIiIiaiQGpyAUvWcpAKCw622QlLoGnyfKoIJOpfBXWURERERErRaDU5DR5u6EPmcrRJkSBd0nNfg8SoWAeKPGj5UREREREbVeDE5BJnqve7SppMP1cOrjG3yexHAtZDJO0SMiIiIi8gcGpyCiLD0BU/r3ABrXgjxcp0SYRumvsoiIiIiIWj0GpyDi3vDWhbI2F8Ea1b1B51DIBbQJ1/q5MiIiIiKi1o3BKUjI7KWIPPAJACC/110NPk8bkxZyTtEjIiIiIvIrBqcgEXFoBeSOUlhNHVGaNKJB5zBqFTDpOEWPiIiIiMjfGJyCgehE9N5lAID8XtMatOGtTAZO0SMiIiIiaiIMTkHAmLkGqrLjcKojUNyxYRvexoZpoJTz10lERERE1BT4TjsIRO95GwBQ0H0SJIXvey/JZECkXuXvsoiIiIiIqBKDU4ApTm2FPnc7RJkKhd0atuFthE7FhhBERERERE2IwSnAdNvfBAAUd/w/OHUxDTpHlIGjTURERERETYnBKZCKMqA+8h2Ahm94a9QqoFbI/VkVERERERGdhcEpkP58E4IkojTxEtgiuzToFNEGtZ+LIiIiIiKiszE4BVLSQDiiuzd4w1utSga9WuHnooiIiIiI6Gx81x1IPW9EUfLVKDPbGvRwjjYRERERETUPjjgFmiC4P3ykkAswaZVNUBAREREREZ2NwSlERRlUEBoQuIiIiIiIyHcMTiFIEIBIHVuQExERERE1FwanEBShV0Eh56+OiIiIiKi58N13CIrSc7SJiIiIiKg5MTiFmDCNAholN7wlIiIiImpODE4hJsrA0SYiIiIioubG4BRCNEoZwjRsQU5ERERE1NwYnEJIFDe8JSIiIiIKCAanECGXCQjnhrdERERERAHB4BQiogwqyGTc8JaIiIiIKBAYnEKAIACRbEFORERERBQwDE4hwKRVQskNb4mIiIiIAobvxkNANJtCEBEREREFFINTkNOr5dCquOEtEREREVEgMTgFObYgJyIiIiIKPAanIKZSyGBiC3IiIiIiooBjcApiUQZ20iMiIiIiCgYMTkFKJgMidQxORERERETBgMEpSEXqueEtEREREVGwYHAKQoIAROnZFIKIiIiIKFgwOAUho0YJlYK/GiIiIiKiYMF350EoOoxrm4iIiIiIggmDU5DRquTQqRSBLoOIiIiIiKphcAoy0WxBTkREREQUdBicgohSIXDDWyIiIiKiIMTgFEQi9SoIAluQExEREREFGwanIMEW5EREREREwYvBKUhE6FWQc8NbIiIiIqKgxOAUJKL0bApBRERERBSsGJyCQJhGAY1SHugyiIiIiIjoHBicgkB0GNc2EREREREFMwanAFMr5TCoueEtEREREVEwY3AKMO7bREREREQU/BiciIiIiIiI6sDgREREREREVAcGJyIiIiIiojoERXBavHgxUlJSoNFoMGTIEGzZsuW89//888/RtWtXaDQa9OrVC99//30zVUpERERERK1RwIPTihUrkJaWhmeeeQY7duxAnz59MHr0aOTm5tZ6/99//x0TJ07EtGnT8Ndff2Hs2LEYO3Ys9u7d28yVExERERFRayFIkiQFsoAhQ4Zg0KBBWLRoEQBAFEUkJydjxowZePzxx2vcf8KECSgvL8e3337rOXbBBRegb9++WLJkSZ3XM5vNMJlMKCkpgdFo9N8TISIiIiKikOJLNgjoiJPdbsf27dsxatQozzGZTIZRo0Zh8+bNtT5m8+bNXvcHgNGjR5/z/jabDWaz2euDiIiIiIjIFwENTvn5+XC5XIiLi/M6HhcXh+zs7Fofk52d7dP9586dC5PJ5PlITk72T/FERERERNRqBHyNU1ObNWsWSkpKPB/Hjx8PdElERERERBRiFIG8eHR0NORyOXJycryO5+TkID4+vtbHxMfH+3R/tVoNtVrtn4KJiIiIiKhVCuiIk0qlwoABA7B+/XrPMVEUsX79egwdOrTWxwwdOtTr/gDw448/nvP+REREREREjRXQEScASEtLw+TJkzFw4EAMHjwYCxcuRHl5OaZOnQoAmDRpEhITEzF37lwAwIMPPojhw4fj5ZdfxjXXXINPP/0U27Ztw1tvvRXIp0FERERERC1YwIPThAkTkJeXh6effhrZ2dno27cvVq9e7WkAkZWVBZnszMDYhRdeiOXLl+PJJ5/Ev/71L3Tq1AlfffUVevbsGainQERERERELVzA93FqbtzHiYiIiIiIgBDax4mIiIiIiCgUMDgRERERERHVIeBrnJpb1cxEs9kc4EqIiIiIiCiQqjJBfVYvtbrgVFpaCgBITk4OcCVERERERBQMSktLYTKZznufVtccQhRFnDp1CmFhYRAEIdDlwGw2Izk5GcePH2ezCvIZXz/UWHwNUWPw9UONxddQ6xRMv3dJklBaWoo2bdp4dfKuTasbcZLJZEhKSgp0GTUYjcaAv3AodPH1Q43F1xA1Bl8/1Fh8DbVOwfJ7r2ukqQqbQxAREREREdWBwYmIiIjo/9u796Coyj4O4N9Ty64LLhCIiJJcFkNIQQHHkAYZuUziEJBKKs2YeGtEBynQ7DKSOsIolXTBbEoqJtLRAU2JUTJclXREEkFHaMAKLyBOJbJcvMDz/uG8O+954W31VTm79f3M7Mye53nO2e9hzxz2t+eyRERmsHBSmEajwdq1a6HRaJSOQlaI2w89KG5D9CC4/dCD4jb0z2St7/s/7uYQRERERERE94tHnIiIiIiIiMxg4URERERERGQGCyciIiIiIiIzWDgRERERERGZwcJJIUeOHEFcXBxGjhwJSZKwZ88epSORFeno6MDKlSvh4eEBrVaLKVOmoKqqSulYZKHM7W8kSRrwsXnzZmUCk0XJzs7GpEmToNPpMHz4cCQkJKChoUE2JiIiot/288orryiUmCzN1q1bERAQYPqx09DQUJSVlZn6ly5dCr1eD61WCxcXF8THx6O+vl7BxPQo5OTkQJIkrFy50tTW09OD1NRUODs7Y+jQoZg5cyauXr2qXEgzWDgppLOzE4GBgfj444+VjkJWaNGiRSgvL0dhYSHq6uoQExODqKgoXL58WeloZIHM7W9aWlpkj+3bt0OSJMycOXOQk5IlMhgMSE1NxYkTJ1BeXo7bt28jJiYGnZ2dsnGLFy+WbUebNm1SKDFZGnd3d+Tk5KC6uhqnTp3CtGnTEB8fj3PnzgEAgoODUVBQgPPnz+PAgQMQQiAmJga9vb0KJ6eHpaqqCtu2bUNAQICsPT09Hfv27cOuXbtgMBhw5coVvPDCCwqlNI+3I7cAkiShpKQECQkJSkchK9Dd3Q2dToe9e/dixowZpvbg4GBMnz4dGzZsUDAdWbp72d8kJCSgo6MDhw4dGrxgZDWuXbuG4cOHw2AwIDw8HMDdI04TJkzAli1blA1HVsPJyQmbN2/GwoUL+/XV1tYiMDAQjY2N0Ov1CqSjh8loNCIoKAj5+fnYsGGDaV/R3t4OFxcXFBUVYdasWQCA+vp6+Pn54fjx43jmmWcUTt4fjzgRWZk7d+6gt7cXQ4YMkbVrtVocO3ZMoVT0d3H16lWUlpYO+GGGCADa29sB3P3g+5++/vprDBs2DOPGjcOaNWvQ1dWlRDyycL29vdixYwc6OzsRGhrar7+zsxMFBQXw8vLCk08+qUBCethSU1MxY8YMREVFydqrq6tx+/ZtWfvYsWMxevRoHD9+fLBj3hOV0gGI6P7odDqEhoZi/fr18PPzg6urK7755hscP34cPj4+SscjK/fll19Cp9NZ9KkSpJy+vj6sXLkSYWFhGDdunKl93rx58PDwwMiRI1FbW4vVq1ejoaEBxcXFCqYlS1JXV4fQ0FD09PRg6NChKCkpgb+/v6k/Pz8fq1atQmdnJ3x9fVFeXg61Wq1gYnoYduzYgZ9++mnA67BbW1uhVqvh6Ogoa3d1dUVra+sgJbw/LJyIrFBhYSFSUlIwatQoPP744wgKCsLcuXNRXV2tdDSyctu3b0dycnK/I5pEwN1vjs+ePdvv6PaSJUtMz8ePHw83NzdERkaiqamJp1oRAMDX1xc1NTVob2/H7t27MX/+fBgMBlPxlJycjOjoaLS0tCA3NxdJSUmorKzkvsiKXbx4EWlpaSgvL//bvI88VY/ICun1ehgMBhiNRly8eBEnT57E7du34e3trXQ0smJHjx5FQ0MDFi1apHQUskDLly/H/v37UVFRAXd3978cO3nyZABAY2PjYEQjK6BWq+Hj44Pg4GBkZ2cjMDAQeXl5pn4HBweMGTMG4eHh2L17N+rr61FSUqJgYnpQ1dXVaGtrQ1BQEFQqFVQqFQwGAz744AOoVCq4urri1q1buH79umy+q1evYsSIEcqENoOFE5EVs7Ozg5ubG/78808cOHAA8fHxSkciK/b5558jODgYgYGBSkchCyKEwPLly1FSUoIffvgBXl5eZuepqakBALi5uT3idGSt+vr6cPPmzQH7hBAQQvzPfrIOkZGRqKurQ01NjekREhKC5ORk03MbGxvZjYgaGhrQ3Nw84PVvloCn6inEaDTKvon75ZdfUFNTAycnJ4wePVrBZGQN/n27Vl9fXzQ2NiIzMxNjx47FggULlI5GFuhe9jc3btzArl278O677yoVkyxUamoqioqKsHfvXuh0OtO1Bw4ODtBqtWhqakJRURFiY2Ph7OyM2tpapKenIzw8vN+th+mfac2aNZg+fTpGjx6Njo4OFBUV4fDhwzhw4AAuXLiAnTt3IiYmBi4uLrh06RJycnKg1WoRGxurdHR6ADqdTnYtJHD3C19nZ2dT+8KFC/Hqq6/CyckJ9vb2WLFiBUJDQy3yjnoAAEGKqKioEAD6PebPn690NLICO3fuFN7e3kKtVosRI0aI1NRUcf36daVjkYW6l/3Ntm3bhFar5XZE/Qy07QAQBQUFQgghmpubRXh4uHBychIajUb4+PiIzMxM0d7ermxwshgpKSnCw8NDqNVq4eLiIiIjI8XBgweFEEJcvnxZTJ8+XQwfPlzY2NgId3d3MW/ePFFfX69wanoUpk6dKtLS0kzT3d3dYtmyZeKJJ54Qtra2IjExUbS0tCgX0Az+jhMREREREZEZvMaJiIiIiIjIDBZOREREREREZrBwIiIiIiIiMoOFExERERERkRksnIiIiIiIiMxg4URERERERGQGCyciIiIiIiIzWDgRERERERGZwcKJiIjoAX3xxRdwdHRUOgYRET1CLJyIiGjQtLa2Ii0tDT4+PhgyZAhcXV0RFhaGrVu3oqurS+l498TT0xNbtmyRtb344ov4+eeflQlERESDQqV0ACIi+me4cOECwsLC4OjoiI0bN2L8+PHQaDSoq6vDp59+ilGjRuH5559XJJsQAr29vVCp/r9/i1qtFlqt9iGnIiIiS8IjTkRENCiWLVsGlUqFU6dOISkpCX5+fvD29kZ8fDxKS0sRFxcHALh+/ToWLVoEFxcX2NvbY9q0aThz5oxpOVlZWZgwYQIKCwvh6ekJBwcHzJkzBx0dHaYxfX19yM7OhpeXF7RaLQIDA7F7925T/+HDhyFJEsrKyhAcHAyNRoNjx46hqakJ8fHxcHV1xdChQzFp0iR8//33pvkiIiLw22+/IT09HZIkQZIkAAOfqrd161bo9Xqo1Wr4+vqisLBQ1i9JEj777DMkJibC1tYWY8aMwbfffvvQ/t5ERPRwsXAiIqJH7vfff8fBgweRmpoKOzu7Acf8uwiZPXs22traUFZWhurqagQFBSEyMhJ//PGHaWxTUxP27NmD/fv3Y//+/TAYDMjJyTH1Z2dn46uvvsInn3yCc+fOIT09HS+99BIMBoPsNV9//XXk5OTg/PnzCAgIgNFoRGxsLA4dOoTTp0/jueeeQ1xcHJqbmwEAxcXFcHd3x7p169DS0oKWlpYB16WkpARpaWl47bXXcPbsWSxduhQLFixARUWFbNw777yDpKQk1NbWIjY2FsnJybL1JCIiCyKIiIgesRMnTggAori4WNbu7Ows7OzshJ2dnVi1apU4evSosLe3Fz09PbJxer1ebNu2TQghxNq1a4Wtra24ceOGqT8zM1NMnjxZCCFET0+PsLW1FT/++KNsGQsXLhRz584VQghRUVEhAIg9e/aYzf7000+LDz/80DTt4eEh3n//fdmYgoIC4eDgYJqeMmWKWLx4sWzM7NmzRWxsrGkagHjrrbdM00ajUQAQZWVlZjMREdHg4zVORESkmJMnT6Kvrw/Jycm4efMmzpw5A6PRCGdnZ9m47u5uNDU1maY9PT2h0+lM025ubmhrawMANDY2oqurC9HR0bJl3Lp1CxMnTpS1hYSEyKaNRiOysrJQWlqKlpYW3LlzB93d3aYjTvfq/PnzWLJkiawtLCwMeXl5sraAgADTczs7O9jb25vWg4iILAsLJyIieuR8fHwgSRIaGhpk7d7e3gBgurGC0WiEm5sbDh8+3G8Z/3kNkY2NjaxPkiT09fWZlgEApaWlGDVqlGycRqORTf/3aYMZGRkoLy9Hbm4ufHx8oNVqMWvWLNy6dese1/T+/NV6EBGRZWHhREREj5yzszOio6Px0UcfYcWKFf/zOqegoCC0trZCpVLB09Pz/3otf39/aDQaNDc3Y+rUqfc1b2VlJV5++WUkJiYCuFuE/frrr7IxarUavb29f7kcPz8/VFZWYv78+bJl+/v731ceIiKyHCyciIhoUOTn5yMsLAwhISHIyspCQEAAHnvsMVRVVaG+vh7BwcGIiopCaGgoEhISsGnTJjz11FO4cuUKSktLkZiY2O/UuoHodDpkZGQgPT0dfX19ePbZZ9He3o7KykrY29vLipn/NmbMGBQXFyMuLg6SJOHtt9/udwTI09MTR44cwZw5c6DRaDBs2LB+y8nMzERSUhImTpyIqKgo7Nu3D8XFxbI79BERkXVh4URERINCr9fj9OnT2LhxI9asWYNLly5Bo9HA398fGRkZWLZsGSRJwnfffYc333wTCxYswLVr1zBixAiEh4fD1dX1nl9r/fr1cHFxQXZ2Ni5cuABHR0cEBQXhjTfe+Mv53nvvPaSkpGDKlCkYNmwYVq9ejRs3bsjGrFu3DkuXLoVer8fNmzchhOi3nISEBOTl5SE3NxdpaWnw8vJCQUEBIiIi7nkdiIjIskhioD0+ERERERERmfB3nIiIiIiIiMxg4URERERERGQGCyciIiIiIiIzWDgRERERERGZwcKJiIiIiIjIDBZOREREREREZrBwIiIiIiIiMoOFExERERERkRksnIiIiIiIiMxg4URERERERGQGCyciIiIiIiIz/gWD3ypfLO0NXwAAAABJRU5ErkJggg==",
      "text/plain": [
       "<Figure size 1000x500 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.figure(figsize=(10, 5))\n",
    "sns.lineplot(x=generations, y=generational_best, label=\"Best\")\n",
    "sns.lineplot(x=generations, y=generational_means, label=\"Mean\")\n",
    "plt.fill_between(generations, generational_means - generational_stds, generational_means + generational_stds, alpha=0.2)\n",
    "gen_labels = [int(g) for g in generations]\n",
    "max_gen = max(gen_labels)\n",
    "while len(gen_labels) > 10:\n",
    "    gen_labels = gen_labels[::2]\n",
    "    if 1 not in gen_labels:\n",
    "        gen_labels = [1] + gen_labels\n",
    "    if max_gen not in gen_labels:\n",
    "        gen_labels = gen_labels + [max_gen]\n",
    "gen_labels = [str(g) for g in gen_labels]\n",
    "plt.xticks(gen_labels)\n",
    "plt.xlabel(\"Generation\")\n",
    "plt.ylabel(\"Fitness (body displacement)\")\n",
    "plt.title(\"Population fitness statistics over generations\")\n",
    "plt.legend()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**Select a generation from which to plot the best robot**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [],
   "source": [
    "generation = int(generations[-1])"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**Where to save robot and simulation results for dynamic visualization**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Visualization output directory:  ./tmp-viz/gen-40\n"
     ]
    }
   ],
   "source": [
    "viz_outdir = \"./tmp-viz/gen-{}\".format(generation)\n",
    "print(\"Visualization output directory: \", viz_outdir)\n",
    "if not os.path.exists(viz_outdir):\n",
    "    os.makedirs(viz_outdir)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**Load the robots**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {},
   "outputs": [],
   "source": [
    "with open(os.path.join(outdir, str(generation), \"robots.pkl\"), \"rb\") as f:\n",
    "    robots = pickle.load(f)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**Select the best**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {},
   "outputs": [],
   "source": [
    "i = 0\n",
    "best_robot_points = robots['points'][i]\n",
    "best_robot_springs = robots['springs'][i]\n",
    "best_robot_id = robots['id'][i]\n",
    "best_robot = {\n",
    "    \"points\": [best_robot_points],\n",
    "    \"springs\": [best_robot_springs],\n",
    "    \"weights\": []\n",
    "}"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**Load trained weights**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {},
   "outputs": [],
   "source": [
    "gen, idx = best_robot_id.split(\"-\")\n",
    "with open(os.path.join(outdir, gen, \"weights\", idx, \"best.pkl\"), \"rb\") as f:\n",
    "    best_robot[\"weights\"].append(pickle.load(f))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**Save the phenotype and weights together for simulation**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Saving best robot to ./tmp-viz/gen-40/best_robot.pkl\n"
     ]
    }
   ],
   "source": [
    "robot_save_file = os.path.join(viz_outdir, \"best_robot.pkl\")\n",
    "print(f\"Saving best robot to {robot_save_file}\")\n",
    "with open(robot_save_file, \"wb\") as f:\n",
    "    pickle.dump(best_robot, f)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**(optional) View the static robot phenotype**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(0.0125, 0.28750000000000003, 0.08700961894323342, 0.3727980021920982)"
      ]
     },
     "execution_count": 19,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "fig = plt.figure()\n",
    "\n",
    "green = plt.colormaps['Greens'](0.75)\n",
    "purple = plt.colormaps['Purples'](0.7)\n",
    "gray = plt.colormaps['Greys'](0.4)\n",
    "\n",
    "for j in range(len(best_robot_springs)):\n",
    "    if best_robot_springs[j][-1] == 0:\n",
    "        c = gray\n",
    "    else:\n",
    "        c = green\n",
    "    plt.plot([best_robot_points[best_robot_springs[j][0]][0], best_robot_points[best_robot_springs[j][1]][0]],\n",
    "                [best_robot_points[best_robot_springs[j][0]][1], best_robot_points[best_robot_springs[j][1]][1]], color=c)\n",
    "\n",
    "for i in range(len(best_robot_points)):\n",
    "    plt.plot(best_robot_points[i][0], best_robot_points[i][1], 'o', color=purple)\n",
    "\n",
    "plt.gca().set_aspect('equal', adjustable='box')\n",
    "plt.axis('off')\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**Command to generate state for dynamic visualization (don't forget to run it!!)**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "python visualize.py ./tmp-viz/gen-40/best_robot.pkl ./tmp-viz/gen-40\n"
     ]
    }
   ],
   "source": [
    "print(f\"python visualize.py {robot_save_file} {viz_outdir}\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**RUN THIS ^^^ !!!!**"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**Load state data for dynamic visualization**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {},
   "outputs": [],
   "source": [
    "state_dir = os.path.join(viz_outdir, \"state\")\n",
    "x = np.load(os.path.join(state_dir, \"x.npy\"))[0]\n",
    "v = np.load(os.path.join(state_dir, \"v.npy\"))[0]\n",
    "act = np.load(os.path.join(state_dir, \"act.npy\"))[0]\n",
    "spring_actuation = np.load(os.path.join(state_dir, \"spring_actuation.npy\"))[0]\n",
    "spring_anchor_a = np.load(os.path.join(state_dir, \"spring_anchor_a.npy\"))[0]\n",
    "spring_anchor_b = np.load(os.path.join(state_dir, \"spring_anchor_b.npy\"))[0]\n",
    "center = np.load(os.path.join(state_dir, \"center.npy\"))[0]\n",
    "loss = np.load(os.path.join(state_dir, \"loss.npy\"))[0]"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**Make directory to save visualization frames**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {},
   "outputs": [],
   "source": [
    "os.makedirs(os.path.join(viz_outdir, \"frames\"), exist_ok=True)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**Generate frames**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "  0%|          | 1/500 [00:00<00:51,  9.76it/s]"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 500/500 [00:55<00:00,  9.02it/s]\n"
     ]
    }
   ],
   "source": [
    "x_min, x_max = np.min(x[:, :, 0]), np.max(x[:, :, 0])\n",
    "y_min, y_max = np.min(x[:, :, 1]), np.max(x[:, :, 1])\n",
    "\n",
    "gray = plt.colormaps['Greys'](0.4)\n",
    "greens = plt.colormaps['Greens']\n",
    "purples = plt.colormaps['Purples']\n",
    "\n",
    "steps = x.shape[0]\n",
    "\n",
    "offx = []\n",
    "offy = []\n",
    "vx = []\n",
    "vy = []\n",
    "\n",
    "for t in range(steps):\n",
    "    offx.append(x[t, :, 0] - center[t, 0])\n",
    "    offy.append(x[t, :, 1] - center[t, 1])\n",
    "    vx.append(v[t, :, 0])\n",
    "    vy.append(v[t, :, 1])\n",
    "\n",
    "offx = np.array(offx)\n",
    "offy = np.array(offy)\n",
    "vx = np.array(vx)\n",
    "vy = np.array(vy)\n",
    "\n",
    "offx = 0.0 + (offx - np.min(offx)) * (1.0 / (np.max(offx) - np.min(offx)))\n",
    "offy = 0.0 + (offy - np.min(offy)) * (1.0 / (np.max(offy) - np.min(offy)))\n",
    "vx = 0.25 + (vx - np.min(vx)) * (1.0 / (np.max(vx) - np.min(vx)))\n",
    "vy = 0.25 + (vy - np.min(vy)) * (1.0 / (np.max(vy) - np.min(vy)))\n",
    "\n",
    "for t in tqdm(range(0, steps, 2)):\n",
    "    fig = plt.figure(figsize=(10, 5))\n",
    "\n",
    "    for j in range(act.shape[1]):\n",
    "        if spring_actuation[j] == 0:\n",
    "            c = gray\n",
    "        else:\n",
    "            a = act[t, j]\n",
    "            c = greens((a + 2) / 3)\n",
    "        plt.plot([x[t, spring_anchor_a[j], 0], x[t, spring_anchor_b[j], 0]],\n",
    "                 [x[t, spring_anchor_a[j], 1], x[t, spring_anchor_b[j], 1]], color=c)\n",
    "\n",
    "    for i in range(x.shape[1]):\n",
    "        c0 = purples(offx[t, i])\n",
    "        c1 = purples(offy[t, i])\n",
    "        c2 = purples(vx[t, i])\n",
    "        c3 = purples(vy[t, i])\n",
    "        plt.plot(x[t, i, 0], x[t, i, 1], 'o', color=c2, markersize=8)\n",
    "        plt.plot(x[t, i, 0], x[t, i, 1], 'o', color=c3, markersize=6)\n",
    "        plt.plot(x[t, i, 0], x[t, i, 1], 'o', color=c0, markersize=4)\n",
    "        plt.plot(x[t, i, 0], x[t, i, 1], 'o', color=c1, markersize=2)\n",
    "\n",
    "    plt.hlines(0.0915, x_min, x_max, color='black')\n",
    "\n",
    "    plt.gca().set_aspect('equal', adjustable='box')\n",
    "    plt.axis('off')\n",
    "    plt.xlim(x_min - 0.05, x_max + 0.05)\n",
    "    plt.ylim(y_min - 0.05, y_max + 0.05)\n",
    "    plt.text(x_min - 0.015, 0.06, \"0.0\")\n",
    "    plt.text(x_max - 0.025, 0.06, f\"{np.round(x_max - 0.025, 2)}\")\n",
    "    plt.tight_layout()\n",
    "    plt.savefig(os.path.join(viz_outdir, \"frames\", f\"{t}.png\"))\n",
    "    plt.close()\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**Create and save video from image frames**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Moviepy - Building video ./tmp-viz/gen-40/sim.mp4.\n",
      "Moviepy - Writing video ./tmp-viz/gen-40/sim.mp4\n",
      "\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "                                                              "
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Moviepy - Done !\n",
      "Moviepy - video ready ./tmp-viz/gen-40/sim.mp4\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\r"
     ]
    }
   ],
   "source": [
    "frames = os.listdir(os.path.join(viz_outdir, \"frames\"))\n",
    "frames.sort(key=lambda x: int(x.split('/')[-1].split('.')[0]))\n",
    "frames = [os.path.join(viz_outdir, \"frames\", f) for f in frames]\n",
    "clip = moviepy.video.io.ImageSequenceClip.ImageSequenceClip(frames, fps=100)\n",
    "clip.write_videofile(os.path.join(viz_outdir, \"sim.mp4\"))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**Load the video**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<video src=\"./tmp-viz/gen-40/sim.mp4\" controls  width=\"800\"  height=\"400\">\n",
       "      Your browser does not support the <code>video</code> element.\n",
       "    </video>"
      ],
      "text/plain": [
       "<IPython.core.display.Video object>"
      ]
     },
     "execution_count": 25,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "Video(os.path.join(viz_outdir, \"sim.mp4\"), width=800, height=400)"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "rss-2024",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.10.13"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}