NNagent_IRC_01262020(170602).ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {
    "toc": true
   },
   "source": [
    "<h1>Table of Contents<span class=\"tocSkip\"></span></h1>\n",
    "<div class=\"toc\"><ul class=\"toc-item\"><li><span><a href=\"#import-packages\" data-toc-modified-id=\"import-packages-1\"><span class=\"toc-item-num\">1&nbsp;&nbsp;</span>import packages</a></span></li><li><span><a href=\"#check-teacher-data\" data-toc-modified-id=\"check-teacher-data-2\"><span class=\"toc-item-num\">2&nbsp;&nbsp;</span>check teacher data</a></span></li><li><span><a href=\"#load-data-for-IRC\" data-toc-modified-id=\"load-data-for-IRC-3\"><span class=\"toc-item-num\">3&nbsp;&nbsp;</span>load data for IRC</a></span></li><li><span><a href=\"#compare-policies-of-NN-and-POMDP\" data-toc-modified-id=\"compare-policies-of-NN-and-POMDP-4\"><span class=\"toc-item-num\">4&nbsp;&nbsp;</span>compare policies of NN and POMDP</a></span></li><li><span><a href=\"#IRC\" data-toc-modified-id=\"IRC-5\"><span class=\"toc-item-num\">5&nbsp;&nbsp;</span>IRC</a></span><ul class=\"toc-item\"><li><span><a href=\"#IRC-with-line-search-based-method\" data-toc-modified-id=\"IRC-with-line-search-based-method-5.1\"><span class=\"toc-item-num\">5.1&nbsp;&nbsp;</span>IRC with line-search based method</a></span></li><li><span><a href=\"#IRC-with-gradient-descent\" data-toc-modified-id=\"IRC-with-gradient-descent-5.2\"><span class=\"toc-item-num\">5.2&nbsp;&nbsp;</span>IRC with gradient descent</a></span></li></ul></li><li><span><a href=\"#Check-points\" data-toc-modified-id=\"Check-points-6\"><span class=\"toc-item-num\">6&nbsp;&nbsp;</span>Check points</a></span><ul class=\"toc-item\"><li><span><a href=\"#Contour\" data-toc-modified-id=\"Contour-6.1\"><span class=\"toc-item-num\">6.1&nbsp;&nbsp;</span>Contour</a></span></li><li><span><a href=\"#Hessian-of-final-points\" data-toc-modified-id=\"Hessian-of-final-points-6.2\"><span class=\"toc-item-num\">6.2&nbsp;&nbsp;</span>Hessian of final points</a></span><ul class=\"toc-item\"><li><span><a href=\"#Std,-Covariance,-Correlation\" data-toc-modified-id=\"Std,-Covariance,-Correlation-6.2.1\"><span class=\"toc-item-num\">6.2.1&nbsp;&nbsp;</span>Std, Covariance, Correlation</a></span></li></ul></li></ul></li><li><span><a href=\"#Check-IRC-result\" data-toc-modified-id=\"Check-IRC-result-7\"><span class=\"toc-item-num\">7&nbsp;&nbsp;</span>Check IRC result</a></span><ul class=\"toc-item\"><li><span><a href=\"#Average-Posterior\" data-toc-modified-id=\"Average-Posterior-7.1\"><span class=\"toc-item-num\">7.1&nbsp;&nbsp;</span>Average Posterior</a></span></li></ul></li><li><span><a href=\"#save-simulation-results\" data-toc-modified-id=\"save-simulation-results-8\"><span class=\"toc-item-num\">8&nbsp;&nbsp;</span>save simulation results</a></span></li></ul></div>"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# import packages"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "from twoboxCol import *\n",
    "import pickle\n",
    "from datetime import datetime\n",
    "import matplotlib.pyplot as plt\n",
    "from sklearn.decomposition import PCA\n",
    "from sklearn import random_projection\n",
    "import matplotlib.gridspec as gridspec\n",
    "\n",
    "import matplotlib as mpl\n",
    "import matplotlib.pyplot as plt\n",
    "import os\n",
    "path = os.getcwd()\n",
    "\n",
    "\n",
    "from matplotlib.ticker import FormatStrFormatter"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [],
   "source": [
    "\"\"\"\n",
    "color code for color cues\n",
    "\"\"\"\n",
    "rgb = [[255, 0, 0],\n",
    "       [255/4*3, 0, 255/4*1],\n",
    "       [255/4*2, 0, 255/4*2],\n",
    "       [255/4*1, 0, 255/4*3],\n",
    "       [0, 0, 255]\n",
    "      ] \n",
    "rgb=np.array(rgb)/255.\n",
    "\n",
    "cmap_col5 = mpl.colors.ListedColormap(rgb,\"\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [],
   "source": [
    "# date of the files\n",
    "datestring_data = '01262020(150900)'  #POMDP teacher data\n",
    "datestring_train = '01262020(151953)' #train neural neworks\n",
    "datestring_NNagent='01262020(170602)' #NN agent behavioral data, closed loop"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# check teacher data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [],
   "source": [
    "dataN_pkl_file1 = open(path + '/Results/'+ datestring_data + '_dataN_twoboxCol.pkl', 'rb')\n",
    "dataN_pkl1 = pickle.load(dataN_pkl_file1)\n",
    "dataN_pkl_file1.close()\n",
    "\n",
    "idx_Sample = 20\n",
    "act = dataN_pkl1['observations'][idx_Sample, :, 0]\n",
    "rew = dataN_pkl1['observations'][idx_Sample, :, 1]\n",
    "loc = dataN_pkl1['observations'][idx_Sample, :, 2]\n",
    "col1 = dataN_pkl1['observations'][idx_Sample, :, 3]\n",
    "col2 = dataN_pkl1['observations'][idx_Sample, :, 4]\n",
    "\n",
    "lat = dataN_pkl1['beliefs'][idx_Sample, :, :]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [],
   "source": [
    "\"\"\"\n",
    "import parameters of the NN agent in the test case\n",
    "\"\"\"\n",
    "\n",
    "para_pkl_file = open(path + '/Results/'+ datestring_data+'_para_twoboxCol.pkl', 'rb')\n",
    "para_pkl = pickle.load(para_pkl_file)\n",
    "para_pkl_file.close()\n",
    "nq = para_pkl['nq']  # how many belief states\n",
    "na = para_pkl['na']  # number of actions\n",
    "nr = para_pkl['nr']  # number os rewards, 2\n",
    "nl = para_pkl['nl']  # number of locations, 3\n",
    "Numcol = para_pkl['ColorNumber'] # number of colors\n",
    "discount = para_pkl['discount']\n",
    "\n",
    "# nq, na, nr, nl, Numcol, discount, parametersAgent, parametersExp = para_pkl['NNtest_params']\n",
    "# Numcol = parametersAgent[7]  # number of colors\n",
    "Ncol = Numcol - 1  # number value: 0 top Numcol-1\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAABDUAAAKsCAYAAAD8yYZQAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjMuMiwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy8vihELAAAACXBIWXMAAAsTAAALEwEAmpwYAADsUElEQVR4nOzdd3gU173/8fdZdQkJRJUAgSQ6wg33jgHllxun2Y4dO9iOE5cU0pPrJL6JYzvXcXLTnIJTcNxx7CS2EztdgMFNLiA3RFcBdZBQ77t7fn/sriKEykra1e5Kn9fz8Ajtnp35zs6Z0cx3TjHWWkREREREREREIo0j1AGIiIiIiIiIiIyEkhoiIiIiIiIiEpGU1BARERERERGRiKSkhoiIiIiIiIhEJCU1RERERERERCQiKakhIiIiIiIiIhEpOtQBiPSVaBLtFKYMWqaK2UMuJ53KgMSTTPOol1HpR7z+CEQs4N/3N8mPdfkTz2yqBn2/kvQhlxGo/e1PvP7sK3+WM1bfsT+xNJM8ZJmx5E88s/3Yn/4sZ6gygdqXi9k3ZBl/BOpcEU7fnz8CVScCdfyGk0B9N4HiTzyBOteO1XYFKpZAnWsDVUeTaRmyzFB/gwP1NyZQ9Xg/S4YsE6jrv6GuXwB2cvqg75/OzoDE0swkP8qMXf0bq+sKf/bBPhYHaF2BOcb9idmfa19/DHV94k/98yeWQNSJBhpos21myAWNkJIaEnamMIVbuGXQMndyx5DLucWPMv5YxbZRL+OOMIoF/Pv+TvdjXf7Ecwd3DvH+4PsaAre//Ys3MMsZq+/Yn1i2sWrIMmPJn3j82Q/+LGeoMoHal78J0HccqHNFOH1//ghUnQjU8RtOAvXdBIo/8QTqXDtW2xWoWAJ1rg1UHV3F9iHLDPU3OFB/YwJVjy/xI55AXf8Ndf0CYNgx6Ps7CMw93LYhkieeMqsCsq5wuq7wZx+s4jcBWtcdQ5bxrx4PHbM/177+GOr6xJ/6508sgagTv+W3Qy5jNNT9REREREREREQikpIaIiIiIiIiIhKRlNQQERERERERkYikpIaIiIiIiIiIRCQlNUREREREREQkIimpISIiIiIiIiIRSUkNEREREREREYlISmqIiIiIiIiISERSUkNEREREREREIpKSGiIiIiIiIiISkZTUEBEREREREZGIpKSGiIiIiIiIiEQkJTVEREREREREJCIpqSEiIiIiIiIiEUlJDRERERERERGJSEpqiIiIiIiIiEhEUlJDRERERERERCKSkhoiIiIiIiIiEpGU1BARERERERGRiKSkhoiIiIiIiIhEJCU1RERERERERCQiKakhIiIiIiIiIhFJSQ0RERERERERiUjRgVjI+nUFFmDDppUmEMsbS+vXFawE1gJnAGcCmd63ztywaeWOUMU1lPXrCm4AHgQe3rBp5Q2hjUZERERERERk7AUkqRHhbgc+FOogRERERERERGR4lNSAfOBdYCewA3gJmB/SiERERERERERkSBM+qbFh08of9P59/bqCUIUiIiIiIiIiIsMQ8KTG+nUFtwCfAZYA7cB24PYNm1buGqD8fODrwHuBOUAb8BawccOmlY/3KmeAv3vLbdywaeUtfZbjAPKA1cCGDZtWfi6wWzY869cVTAO+CHwAWABEAZXAy8BvN2xa+Uqf8n59D8NY/3nAV4HzganAMTytUH60YdPKV/spvw24GLgEcHtjOdv72cs3bFr55+HGICIiIiIiIhJMAZ39ZP26gp8CvwIagb8AtcBlwGvr1xVc0E/5s/HcuH/G+9IzeLqAnA9sWr+u4BFvMoMNm1Za4DqgArh5/bqCa/os7nY8CY038dzMh8z6dQWn4enS8m1gHrAN+BuexMI1QN+EjN/fg5/r/wzwInA5cBj4k/fnFcDL69cV3DzIx68EnvfGnQdsAbr9XbeIiIiIiIjIWAl0S41bgEs2bFr5AvS0rvge8A3g8fXrChZv2LSyw/tePPBHYApwL/C1DZtWurzvrcBzM30dnpYNvwHYsGll7fp1BVfjSRL8Zv26gh0bNq08sH5dwSV4EgjNwFUbNq3sDPB2+W39uoJk4FkgHfg18JUNm1a293p/Bp5WLL7fh/09DLH+U4Cfe3+9asOmlX/s9d7VwCZgw/p1BfkDtJ75LPCpDZtW/tbPTRYREREREREJiYC21AB+5UtoQE/rim8BxUAGnpYCPld6XzsE3Oq7kfd+bhdwh/fXr/VewYZNK1/C0yojGfjD+nUF84DHvdty84ZNKw8GeJuG60ZgLvAq8NneCQ2ADZtWHvVug8+IvodBfAFPsuqJ3gkN7/KewNNqIwZP15j+5CmhISIiIiIiIpEg0EmNx/q+4L1J/73311W93rrY+3PThk0r++ve8CBggYXr1xXM6fPePcC/gFOBd4A04DcbNq18csSRB857vT9/503qDGU038Ngy3togPcf8P5cNcD7T/uxDhEREREREZGQC3RSo2SA10u9P+f2es13g97vZ7zdVCr7lPW95xtfowWYDOwGvjTsaIPDNx3sXj/Lj/h7GMnygKIhlnXIj3WIiIiIiIiIhFygkxpD6d1ywfTzWl+DDY75YWCS9/9z8e+GPxyN9nsYyEDLG2pZ7UO8LyIiIiIiIhIWAp3UyBzi9cper5V7f2b39wHvAJrp3l8r+ry3AvgZ0IVn4MsU4Mn16wpihx1x4PlaOiwZtNR/jPh7GICvTL/LA7KGsSwRERERERGRsBXopMa6vi+sX1cQBXzU++u2Xm9t9/68Zv26gv5mYfk4nlYFBzdsWtlzA75+XUES8AcgAfg6cD2eKUhPB344yvgD4V/en5/0cxrWEX0Pfizv+gHe/4T35zY/liUiIiIiIiIStgKd1Pjs+nUFF/h+8d7U3wksxNMy4KleZf8IlOFpOXDP+nUFjl6fW+79HMCP+qxjA7AMeHbDppX3bti00o0nmXIE+ML6dQUfDugWDd/9eFqknAf8wtvSosf6dQUzen9HjPx7GMjPASeeJMllfdZ9JXAV0M1/pn0VERERERERiUj9tQwYjY3A9vXrCl4AqoCVeLphtAPrek9vumHTyo716wquAv6BZ7rSy9avK3gDmIpnZo5Y4FGgZ3rR9esKrsfTcqGM/7Q4YMOmlVXr1xVcB/wTeGD9uoI3N2xa6deAl+vXFVwKfLvXS76uHg+uX1fQ6v1/1YZNKy/DDxs2rWxev67gQ8DfgPXA1evXFbwMdOAZRPQ0PLPBvDTS72GI9b+9fl3BF4FfAk+vX1fwGp7BQRcCZwFu4HMbNq1815/liYiIiIiIiISrQLfU+ArweTw35B8GZgJ/Bs7esGnl9r6FN2xa+SqeaVl/DUQBlwNnA68C1wIf902Lun5dwVLgPrytEDZsWnmsz7L+DfwASAWeWL+uIMbPmGd41+n75xuXY0Wv107zc1m+WHYAJwHfB6qBXOBSPN/L497t7V3e7+/Bz/XfB1wIPIOnBchVeMY1eRq4YMOmlX4lSERERERERETCWUBaamzYtLL32BH3ef/5+9lDwGf8KLeX/8x2MlCZbwLf9Hfd3s88BDw0nM/4udwj3lj8isff76FX+YcYJO4Nm1a+DLw8jOWt8resiIiIiIiISDgY6yldRUREREREREQCQkkNEREREREREYlIxlq/h2oQGRPGmKOAXwO9ioiIiIiISFibb62dEayFK6khIiIiIiIiIhFJ3U9EREREREREJCIpqSEiIiIiIiIiEUlJDRERERERERGJSEpqiIiIiIiIiEhEUlJDRERERERERCKSkhoiIiIiIiIiEpGU1BARERERERGRiKSkhoiIiIiIiIhEJCU1RERERERERCQiKakhIiIiIiIiIhFJSQ0RERERERERiUhKaoiIiIiIiIhIRFJSQ0REREREREQikpIaIiIiIiIiIhKRlNQQERERERERkYikpIaIiIiIiIiIRCQlNUREREREREQkIimpISIiIiIiIiIRSUkNEREREREREYlISmqIiIiIiIiISERSUkNEREREREREIpKSGiIiIiIiIiISkZTUEBEREREREZGIpKSGiIiIiIiIiEQkJTVEREREREREJCIpqSEiIiIiIiIiEUlJDRERERERERGJSEpqiIiIiIiIiEhEUlJDRERERERERCKSkhoiIiIiIiIiEpGU1BARERERERGRiKSkhoiIiIiIiIhEJCU1RERERERERCQiKakhIiIiIiIiIhFJSQ0RERERERERiUhKaoiIiIiIiIhIRFJSQ0REREREREQikpIaIiIiIiIiIhKRlNQQERERERERkYikpIaIiIiIiIiIRCQlNUREREREREQkIimpISIiIiIiIiIRSUkNEREREREREYlISmqIiIiIiIiISERSUkNEREREREREIpKSGiIiIiIiIiISkZTUEBEREREREZGIpKSGiIiIiIiIiEQkJTVEREREREREJCIpqSEiIiIiIiIiEUlJDRERERERERGJSEpqiIiIiIiIiEhEUlJDRERERERERCKSkhoiIiIiIiIiEpGU1BARERERERGRiKSkhoiIiIiIiIhEJCU1RERERERERCQiKakhAWeM+aYx5o/GmGJjjDXGlIY6JhERERERERl/jLU21DHIOGOMscAxoAA4HWiy1maGNCgREREREREZd6JDHYCMSwustcUAxphdwKQQxyMiIiIiIiLjkLqfSMD5EhoiIiIiIiIiwaSkhoiIiIiIiIhEJCU1RERERERERCQiaUwNCQvGmFuAWwB+85vfnH7LLbeEOCIREREREREJABPUhWv2Ewkm30Chw5z9RJVSRERERERkfAhqUkPdT0REREREREQkIimpISIiIiIiIiIRSUkNEREREREREYlIGihUAs4Ycx0w3/vrDCDWGPMt7++HrLWPhiYyERERERERGU80UKgEnDFmG3DxAG9vt9auGmIRqpQiIiIiIiLjg2Y/kQlHlVJERERERGR80OwnIiIiIiIiIiJ9KakhIiIiMoE53W5uKyrinJ07ua2oCKfbHbQykRTLRKbvL/jCqR6H23Gn+ifDpe4nEo5UKUVERMbIbUVF/Li8nC5riQJykpI4PTn5uDI7mprY3daGCwYts6e9Hae1JDgcfHnuXO7Ozh52LPdWVNDudo86Fl+ZkcYykd1WVMRPysvpHMW+lMEN97gL5n4It+Ou93ej+jduBLX7iWY/EREREZnA/nD0KF3eh1wuYG9bG/VO53Flarq6cHn/P1gZp3c57W43W+rruXuYsWyur6fd+1R2tLH4yow0lonsr8eO0TnKfSmDe6q2dljHXTD3w5aGhrA67p7p9d2o/ok/lNQQERERmaDeaWmhtKMDB+Bm4KervZ/kDlbG93TfAaxOTR12PJ29mpmPNpYflZfTbS3xDgdrRhDLRNXodFLe0XHcazlJSSGKZnza39ZGaXs7Bk/z5MHq8U/Ly+mwFgNcOGVKUOJx92q5P9rjzlcGIDM+ftixHO7o4FBHR893A3BGnxYhIn0pqSEiIiIyATV0d3P5rl3MiInhqpkzea2piTWpqdyZmXlC2buysjDGsKW+fsgyj9XUcLizkwTH8IZu21hZyTttbZyXkoLL2lHHcqC9nT/V1nL1jBn9lpETua3l43v20Ohyce3Mmexua+NAezubjx2jrrubaTExoQ4x4rU4nVy+axeToqK4ZtYsdjQ3D1nXnzp6lH3t7ZR3dGCtxZjAteT/fU0NO1paOGPSJKKMCcg54N/HjnGos5N/HTtGSXs7WQkJfsXS4XLxkcJCoozhlrQ08pua2NvWxqtNTXS4XMRHRY1ya2W80pgaEo5UKUVERILIbS0ffPdd/l1fz7ZTT+W8yZMDtmxrLdfu2cPvjxzhnyefzHumTh3yM683NXHhm29yyZQp/O3kk4kKwE3bm83NrNy5k6dycrh8xoxRL28i+N6hQ/xPSQn3LlzIF+fOBf6zb1ZNmcLfA7RvJiprLdfs3s0fjx7lXyefzFo/jg2fuw8d4lt99s1ovdvSwjkFBaxMTmbrKacQM8xE5GCK2ts5Y+dOMuPjefm000j0IyFxy759bKyq4umcHC7zHrN/PnqUywoLuSk9nY1LlgQsPhlzmtJVRERERALnu4cO8bdjx/jpwoUBTWgAGGP47ZIlrEhK4prduyltbx+0/JGuLq4oLGR2XBybli8P2E1zRlwcAGWdnQFZ3nj3r2PH+FZJCR+bOZMvzJnT8/pZKSn8ctEi/l1fz3dKSkIYYeT7aXk5Tx49yt1ZWcNKaAB8c948PjhtGl8rKuLFhoZRx9LQ3c1lu3aREh3NH5YvD2hCA2BBQgKPLVvGWy0tfGb/foZ6kH5/ZSUbq6r45rx5PQkNgA/PmMFt8+Zxf1UVGysrAxqjjB9KaoiIiIhMIH+vq+PO0lKunzWLz86eHZR1JEVF8XRODi5rubywkHaXq99yTrebq3fvpra7m6dycgLavWFaTAzxDgdlfcaHkBOVtrfzsd27WZGUxG+XLDmhe8PNs2dzY1oadx8+zF9qa0MUZWTbVl/PrUVFXDZ9Ol+fN2/Yn3cYwyPLlpEVH8+VhYVUjiJZ57aW6/bu5VBnJ3/KySHdmwAMtEunTeOOzEweqanhV4MkJN5oamL9gQPkpqby3aysE96/KyuL96Sm8rkDB3i9qSkosUpkU1JDREREZIIoam9n3Z49nDJpEr9evDigffP7WpiYyGPLlvFmSwufPXCg3ye1t5WU8HxDA79evJiVAR4M0BhDRlwc5WqpMah2l4vLCwtxWcvTOTkkDdBN4JeLFnFGcjLX79nD/ra2MY4yspV3dPDR3btZlJjIQ0uXjvi4mxwdzTMrVtDiHXuiq9fAusNx96FD/LWujp8uWMD5AW6p1de358/n0qlT+dLBg+Q3Np7w/lFvS6302FgeX7as35ZaUcbw+PLlpMfG8pHCQo52dQU1Zok8SmqIiIiITABtLheX79qFAZ7OySFhDAbde//06dw+fz4PVVfzmz5Pav945Ag/LCvjM7Nn8/G0tKCsf25cnLqfDMJay2f27+fNlhYeW7aMhYmJA5aNj4riqZwcYh0OLtu1i5Y+03lK/zrdbq7cvZs2t5unc3JIiR7dPA05SUk8sHQp+U1NfOXgwWF//h91dXyntJRrZ81ifa9uRsHiMIZHly0jIy6OjxQWUt3reHS63VyzezdHurp4esUKpsfGDricaTExPL1iBUe7u7l6926cI0zoyPikpIaIiIjIOGet5ZZ9+3i3tZXHly3zezaCQPhOZibvmzqVL/R6Uru7tZVP7N3LOSkp3LtwYdDWnaGkxqB+XVnJwzU13D5/Pu+fPn3I8vPi43li+XL2trVx4759Q46TIPClgwd5tamJh5YuZVmApsa9auZMvjp3LhsqK3mkutrvzxW3t/OxPXs4OSmJ3wS5pVZvqTExPLNiBfVOJ1ft3k23NyHxPyUlbGlo4Fd+ttRamZzMrxYtYmtDA7dpfBfpRbOfSNg5Z+dOu3rKFO7KyiI6wIMWBZrT7eb2khK2NjQQKTGLyNjSeUJCyVf/Hj9yhEOdndwxfz7f6afPerDVd3dzxs6dtLtcfHD6dB6ursYAe84+m/nx8UFb77eKi/n+4cN0XnzxhJi1w5/zja/MX+rq2NPWxv9LTeVvJ5+MYxjfzw8OH+YbxcWsnjKFVpdr3JzbAnW+9i3nyaNHKe7o4Gtz5/LDACfvnG43ue+8Q35jIx/zTr872D7/ZnExGyorcVvLO2eeyeJBWuUEy+M1Nazbs4czJk2i0eXiQHs7t6Sn85thzmry2f37+VVlJR+eNo2qrq6A7KvR7HP9nfdLUE/Ao2v/JBIErzU3805rK8YY7s7ODnU4g7q9pISfVlTQ4XZHTMwiMrZuLynh3ooK2nWekBC4vaSEn5SX02ktDqAzRA+zUr1Nx0/fsYPfVFUBEGcMv62sDOrxkBEfjwuo6uxkbhCTJ+Gi9/nmzZYW9ra3s27WrOPKbKqu5m/HjtFlLQZYlpQ0rIQGwK0ZGTxYVcVW7ywc4+XcFqjzde/rQwcQHYSEWrTDwZPLl7Pg1Vd5qKYGC4Pu82fr6nDhOe4erq4Oyb762KxZ/KSsjB0tLYDnLnfKCLrj3LtwIX+preXPdXXA6OpfIK7lby8p4WcVFbTp73zIKIUkYand7WZLfX2owxjSH48epcPbhC5SYhaRsZVXX0+7zhMSIv+sr+9JZLiBrSGsf6dMmkRmr8RCp7VBPx5807pOlMFCtzY09JxvuqzlmdpaPlJYeNy/Z+rq6PLWCQu80s/gjUMxxpDca0yW8XJu6/39jWabtjY09FwfuoHnAzAFa39mxsYyPz4eX6pysH3um39oLI67wfRuMWWB7SP4bmIdDmb1Gn9jNPvqyQBcy29taKBNf+dDSkkNCUvRwJrU1FCHMajC1lZKe00TF2NM2McsImPLWktLr6ksDXBekEeaF/Hpcrs50muWgASHI+R/p66aMYMEb7PssYhnrjepMVHG1bhkypSeNt7xxnBTWhrvnHHGcf9uTEsj3ntjOZp9kJuaSqx3OXHj5Bro4ilTev4/mmvRk3qNnRHsev7BadN6jqlg7/NAWDNlSkDOAe9NTSXOu00OYPUIlvNmczOHAnAtP7XPVNSLx3DMIvFQ9xMJOylRUVjg9szMUIcyoEank8t27WJaTAxXzZzJ4zU11DmdXKCbFRHp5afl5extb2fVlCkc7epiT1sbu1pacFk7Ifr3S2h9raiIiq4uPjx9OlWdnaxJTeXOEP9tvSsrC2MMW+rrxySejAmW1Fidmsr3y8pYEB/PR2fO5M7MzBP69v968WJmxcaOeh/clZVFN/CTsjIWJSSEvG4FwhnewSpjjCHGGL41f/6IlpMYFYUDOHXSJN47dWpQv5v+jqlg7fNACNQ5wLecx2pqONzZSdIwx7A41t3NFYWFpMXG8pEZM3ikpoYGp5PVvRJb/nijqYktx46RGR/PtOhoijs6+Hd9PVWdnaR7zz8SfBooVMLOH48csVft3s3fTzqJ/5o2LdThnMBtLZft2sXfjx1j6ymncKF3kKxzCwqo6Oxk5+mnk6kMrciEt62+nrVvv82Hp0/njzk5GO/4AZ/av59vzZ/Pd0MwWKNMHI9VV3Pd3r18ee5cfhLE2UXCnbWW5Bdf5ObZs/npBPgertm9m38dO0bluecSPwZT9gJ8Zv9+Hq6upvLcc5nS54l1pPmvd95hV2srv1i4kMsKC3k6J4fLZswY1jI6XC7m5OezJjWVP+TkBClS8bHW8rE9e/jDkSP88+STyZ06dcjPuKzl0nfe4fmGBl447TTOTkmhyenk7IIC6rq7KTj9dL/G4Dna1cXpO3fiAHacfjrTY2PZ1dLC2QUFnDZpEltPPZVYDRjqE9QnOfqWJex8aPp0ZsTEsNE7kFi4uefwYZ6tq+PHCxZwoTebmxQVxdM5Obis5fLCQtp7NTcXkYmnvKODq3bvZlFiIg8uXdozbd7N6el8Mi2N/z10iGdra0McpYxXb7e0cMv+/Vw8eTI/mOCD1RljyIiPnxBjatR2dfH00aNcN2vWmCU0AG5KT6fd7WbTkSNjts5gONTRwb+OHeOTaWm8f9o0ZsfGcv8IrkX/XFvLMaeTm9PTgxCl9GWM4f4lS1ielMQ1u3cf151kIHeUlvKv+np+sWgRZ6ekAJASHc0zOTm0u918pLCQTu8YGQNxut1cs3s3R7q6eHrFCqZ7x/hYMWkSDyxdystNTXytqGj0Gyh+UVJDwk6sw8HH09J4rq6O6jC7CPnXsWN8u6SEj82cyefnzDnuvYWJiTy2bBlvtrTwmf37NXe7yATV6b0gane7eSYnh+ReI7sbY9iwaBGnT5rEdXv2cKCtLYSRynh0rLuby3btYmp0NE/m5BCjp4TMjYujzI8bnUj3WE0NXdZy0xjfTJ+enMxpkyaxsbIyoq99HvQmMD6Znk60w8En09P557Fjw647G6uqyIyPHxdjjEQK38PFbmu5fNeuQR8uPltby/8eOsQn09JOSDwtTUrioaVLea25mS8dPDjoOr9VUsKWhgZ+tXgxK73dlnw+OnMmX547l19UVPBYdfXIN0z8pr90EpZuTEvDaS0P19SEOpQeJe3tXLN7NyclJfHbJUt6nrz29v7p0/nO/Pk8XFPDrysrQxCliITaFw8c4LXmZh5eupSlvQaL84mPiuKpFSuIMYbLdu2ixekMQZQyHrmt5do9eyjv7ORPOTnHzQ4wkWXExY37MTWstWysquLs5GROmjRpzNd/c3o6b7e2srO5eczXHQgua3mgupr3pKYy39vt4JNpabiBB4ZxU1rU3s7WhgZuTEsb9jS5MjqLvA8XC1paWH/gQL8JtgNtbVy3Zw+nT5rEhkWL+r2Wv2LGDL6ekcGvKyt7El19PXX0KD8oK+PTs2fziQGSiD/IzubiyZO5Zf9+3vZOYSvBo6SGhKWlSUlcOHky91dVhUXWv93l4vLCQizw9IoVJA3SrPP2zEzeN3UqXzx4kPwRTJMmIpHrwaoqflNVxdczMrh8kH7Y8+PjeWL5cva0tXHTvn1hcZ6TyHdnaSn/OHaMny9cyDkauLpHRlwcVV1ddA/RnDyS5Tc1sbutjZtnzw7J+j82axYJDkfYdh0eyr+OHaOss/O47y8rIYHc1FQeqKrC5ec5+ndVVThgwBtdCa4PTJ/Ot+fP58Hqan7bpy62OJ1cXlhIjDE8tWLFoF20/jcrizVTpvCZ/ftPSNTtaW3lhr17OTs5mXsHGacnxuHgyZwcpkZHc/muXdR3d49u42RQSmpI2Lo5PZ2D7e1sC9Lc3v6y1vLp/ft5q6WFx5YtY8EQg4A6jOGxZcvIiIvjI4WF1PSaTk9Exq+dzc18Zv9+1qam8r9+DAK6dupU7s7K4smjR7m3vHwMIpTx7K+1tdx16BA3pKXxqRDd2IarjLg4LFA1jv8e319VxaSoKD46zEEtA2VydDRXzZjB40eORGTrs/urqpgRE8MH+gxQf1N6Ooc7O8k7dmzIZXS73TxYXc2l06YxR7NehMx3MjN579SpfP7AAV71Ply01nLTvn3sbm3l98uX97TGGUi0w8Hvly9nZmwsl+/aRa333NHknf0w0eHgTzk5xA3RvW9WbCx/ysmhrLOTdXv24NYDjKBRUkPC1hUzZjA5KmpEgzQF0q8qK3mkpoY7MjO51M/ZWFJjYnh6xQrqnU6uKiwc10+HRMQzQN/lu3YxKzaW3y9bdsJ0egP5+rx5XDZ9Ov9dVMS2+vogRynj1cG2Nq7ds4eVkyZx3wBNqieyueN8Wtcmp5MnjxzhmpkzmdRrDJ+xdvPs2bS4XPzh6NGQxTAS1Z2dPFdXxw1paSfMVPGh6dOZ7ufg9X+rq6O6q2vMxzSR40UZw6Zly5jb6+HiveXlPHn0KHdnZfk1OwrAjNhYns7Joaari2v27MHpdvOJvXs52N7Ok8uX+zU7CsA5kyfz84UL+cexY9xZWjqKLZPBKKkhYSsxKoprZ83iqaNHORaCJltOt5vrd+9m/YEDLIiP55vz5g3r86dMmsTGJUt4obGR8wsKOGfnTm4rKsI5ggSH0+3mtqIiztq5k1sPHqTF6aTD5TruX4vTya0HD3L2KNYTSE63m28UFXHWjh2Dxvz1oqJRfTcSHnx1dLT70ul2800/6s2tBw+OqkygjhdfPV/w2muUd3by5PLlPSOg+8MYw0NLl7IwIYH3vfMOK994Y0y+Px1344PT7eZrBw9y8o4ddLrdPLl8OQljOOtFpMjw3nyM18FCH6+poc3tDvlsG+elpLAsMXHYXVB8561QnZMeqq7GOcAAq3EOBx+fNYtn6+qGbHl7f1UVs2NjeZ+fN80SPFNjYng6J4fa7m5Oev11vlJUxOKEBL6akTGs5ZyRksKGxYvZXF/P3Px8nq6t5eIpU7hgmN37PjV7NtfPmsVdhw6x5LXX9Lc3CEKXzh0hY0wpML/PyxZoBvYCTwIbrLUBSccbY24AHgQettbe0Ov1TKAEOGStzQzQulKB7wGXAul49s9frLUfDsTyI9HNs2ezobKSR2tq+OLcuWO67q8cPMhj3unJKru6uLO0lLuHOTXeulmz+GlZGW94Bwh6p7UVY8ywl3N7SQk/Ki+n21reaG7mh0M0VX93hOsJpG8UF/OT8nIs8EZLy5Axj/S7kfDw7ZISflhWhovR7cv/LiriZxUVftebQJR5u6VlxPHeXlLCT7zHZqwxPFdXN+yxDFKio7lo8mQ2trfzZmsre9vbRxzPVw4e5JeVlTruJojbS0r4WUUFTmuJM4YHq6u1L/uRMc5batxfVcXJSUmc0WcGhrFmjOGm9HS+WlTErpYWVvg5YOmXDx5kg/e8NdbnJGst91dVcdHkySxOTOy3zI3p6fy4vJyHq6u5dYAHXOUdHfzj2DG+OW+e3y31JLhOTU7mPampPOftOlTW2ckdI7iWvzE9nV+Wl/NWayvgGb/mO8NcjjGGmTExGGB/ezv3VlTob2+ARVxSo5d/Ab7hiKOBDOA84CzgSmPMJdbaSEvJbwSuwJMs+RPQCRSENKIQO2XSJM5ITub+qiq+MGfOmDWp7Xa7ebCmBl/Pt3a3my319dw9gmVF9Yp5pMt5praW7l798DJiY/lsnyll76uooMz7FGE08QbKM7W19O45GAkxy8g9VlODbwK1ke5Ll7U8WF09rHoz0jK93++0dsR1b2tDQ8+x2TWK5bzjvViCkX9/Xd7+3DruJo68+nqc3vo3mno83qVER5MSFTUukxpvNjezs6WFXyxcGBbdjq6fNYtvFBdzf1UV9y5aNGT5vuetsT4nbWtooKijgzsyMwcssywpiQu8g9f/d0ZGv9/zA9XVuPFMByvh40ivlt6jqVu9p8Ye6XJebGwMWT2fCCI5qfF9a+223i8YYxYDLwPnAJ8CfhbE9VcAy4CA9IswxsQAHwI6gFOttU2BWO54cHN6Op/av5/XmprGbDT3W4uLaXG5iDWGLmtJcDhGPN/4milTeKe1lQ63GwOcP8xtKOvo4FBHBwZPk6QEh4Pr0tL4xvzjGyw1OZ3cW1FBu9uNA0I+P7pve4eK+QdlZbi9ZUIds4zMP+rqKO/qIgp6EhvJI+jXfWdpKY0uFzHG0O097oaq6yMt0/t9YNhNSX0WxMfzmndk9NHU4dXe84QvnukxMcNexteKimhxu487bw313QCcGeKnuzJyvads1Tl0cBlxcZSPw6TGxqoq4h0O1s2aFepQAJgeG8tl06fzaE0N38/OHnSGCfC00mh1u3v+fsQZM6b1eGNVFVOio7liiAFWb05P5+N79/JCYyMXT5ly3Hsua/ldVRVrU1PJHmIweRlbvf+2juYcuXbKFHaNcjmrp0zh7ZYWOrwtO3W+DqxITmqcwFq73xjzG+B/gFUEMalhre3G090lUHzdTSqU0DjeNTNn8pWDB9lYVTUmSY3f19Rwb3k5n589m+ToaLbU17MmNZU7B8niD+aurCyMMfy5tpY9bW3sbWvDba1f85d3uFxcUVhIlDHckpbGWy0tA8biW8/D1dVUdHVxYwifFhS2tlLZ1cWaKVNocbkGjfmNlhY219fzybS0EX/HEjrF7e2s27OHkxMT+X/TpvFCQwP1Tidb6+t5oaGBi/pc/A3k2dpavnvoEDfMmkV6XBxbBznufHV9sGNzqDK+9/9SW0thWxuLRngh6sZzEX5yUhK5U6eO+jyxub6emq4u8o4do6C5mZV+Jhweq67mFxUVfGnOHBKjovz6bv5aV8fetjZea2qi0+0echR3CT/13d1MjY5mUULCqP5OTQRz4+LGXUuNVpeLTTU1fGTGDFJHkAgNlpvT0/nD0aM8U1vLNYMkWx6urua+ykq+OncuUcbwk7IyMuPjx6we13V389TRo9wye/aQY9F8ZMYMvnDgABsrK09Iamyur+dwZyc/XLAgiNHKSPhzvTBWy7krKwsn8MOyMs5OSdH5OtCstRH1DyjF8/B31QDvf877/l8HeD8JuBV4A2gC2oFC4A5gUj/lb/Au76E+r2d6Xy8d7Xq8yxnoX2aov/MQ/DvBJ/fssYnbt9vG7u7+3g6Yd5qbbeL27fbCggLb5XIFfPm/KCuzPP+8vaukxK/yN+/da3n+efvMkSN+r6Okrc2a55+33ykuHmGUo/elAwdszLZt9khn55Bl321utjz/vP1dZeUYRCaB1Op02lPfeMNOefFFe7Ctref1xu5uu/jVV+2sl16yFR0dQy5nf2urTXnhBXv6G2/YdqczmCGfwO1225zXXrNn7dgx7M/WdXXZuG3b7Of27w9oTEc6O23GK6/Y+a+8Ymu7uoYs/1Zzs03Yvt1eXFBgu4d53nrqyBHL88/bT+3dO9JwJUQKW1oszz9vf3joUKhDiQg37d1rZ730UqjDCKiHqqoszz9vt9fXhzqU47jcbpuVn28vefPNAcvsbGqy8du320vefLPnvPXVAwds9LZtttqPa4dAuNd7TfZ2c7Nf5T+7b5+N27bN1vU5L1/x7rt22osv2o4gXDfK+JOZn2+vKSwMdRihENT7x/H4WOYs7889fd8wxswFXgd+gGew0Xzg30Aq8B3gZe9gnaMygvU8DDzl/X+r93ffv5bRxjMe3JyeTpvbzRPegTuDoaG7m8t27WJydDR/WL78uP5zgbJ+zhyunTWL75SW8o+6ukHL3l9ZycaqKm6bN48PD2Pe+cyEBHJTU3mguhqXHfv5sDtcLh6pruay6dOZ4ccsEDlJScyKiWGLprOMKNZaPr1/P2+3tLBp2TIW9GrpkBIdzTMrVtDicvGRwkK6Bhnhu8Xp5PLCQmKM4akVK4Zsqhxoxhhunj2b15ubeadleKfbR2tq6LT9j5g/GjNiY3kqJ4eqri6u2b170OP4mPe8NTU6midzcoY9QN3lM2bwjXnz+E1VFQ+EePpsGZ77q6qIMYbr09JCHUpEyIiLo6a7m85xNOPAxspKFickcOEYdc31l8M7YOjzDQ0cbGs74f267m4u37WL6TExPLF8ec9568b0dJzW8nB19QmfCTRrLRsrKzkrOZmT/RzQ9Ob0dDqtZVNNTc9rR7q6+EtdHR9PS1NrN/HL0sRE9vZzXMjojIujzxgTbYzJNMbcDlwLNAD39SljgD8Ay4Ff4mkB8f+stR8CFgCPAScDPx1lLMNej/XMqvI176+11tobev2rHU0848XZKSnkJCZyf5Auut3Wct3evRzu7ORPOTmkeUdKDzRjDL9ZvJiTk5L42J49FLe391vujaYm1h84wHtSU7krK2vY67k5PZ2yzk7+7R3xeSw9U1vLMafT76nljLdf4Zb6el/LJYkA93lnJbojM5P3TZt2wvvLk5J4cOlS8pua+PLBg/0uw1rLTfv2sbu1lSeWL2e+n3O+B9p1s2YRZ8ywzi/WekbMPzM5mVP8vCAejjNTUrhv8WLy6uu5vaSk3zJua7l2zx7KveetWcOYSra3/83KYm1qKp/dv58dTer9GAk63W4eqa7mQ9OnM3OE+32i8c2AUjlOuqDsbm3l5aYmbkpPD4sBQvu6IS2NKOB3fRIULmv52O7dVHV18VROznH1t/eAnMG+HnitqYnCtrZhJaVPTU7m9EmT2NgrvocHmQ5WpD/LEhPZ5+2KLoETyUmN540x1hhj8QzWWQLciWdWlLOttX2vAt8LnAu8CnzRWtuTIrPWtgOfBo4A60bZWmOs1jOh+J6mvtHczNvDfJrqj/89dIi/1tXx04ULOS/ITzwSo6J4esUKDHD5rl20uVzHvX+0q4srCgtJj43l8eXLj5s9xV8fnD6dGTExw54rPhDur6oiKz6e1cMYAGlNaio13d0U9poBQsLXy42NfOngQd4/bRrf6jMIZW9XzpzJ1zIyuK+yst8nb/eWl/Pk0aPcnZXF2qlTgxnyoKbGxHD5jBk8WlNDe5/jcSCvNTWxq7XV7+TdSNyYns7N6el87/Bh/nz06Anv31layj+OHePnCxeOaryhKGP4/bJlzIqN5YrCQmp7zRoj4enPtbXUDSN5LJ4xNWD8TOv6O29LnY+HaUud2XFxXDptGg9WVdHdq3XM7SUl/Lu+nl8uWsRZKSknfO7m9HQOtLfzQmNjUOPbWFVFksPB1TNnDutzN8+ezbutrbze3NyT3D4/JYVlSUlBilTGm6WJibS53eNy4OJQiuSkxr84vpvG3/HMSPJe4BfGmL4jE73P+/Mpa+0JbQ+tta3ADjyDdZ45irjGaj0Tju9p6sbKyoAu9+91ddxRWsp1s2bx2dmzA7rsgWQnJLBp2TLeaW3lU/v392T8nW43V+/ezdHubp5esYJpIxz4K9bh4ONpaTxXV0f1GJ40i9rb2drQwI3p6X4NhOrjGwF6S0NDkCKTQKnq7OTKwkLmx8Xx6NKlQ+7ne7KyuGTKFD69fz9vemcJAdhWX89/FxVx2fTpfH3evGCHPaSb09NpcDp5qp/kQX9GekE8XL9YtIgzk5O5fu9e9vVqrvpcbS13HTrEJ9LS+FQAzlvTvV1earq6uGbPnpB0XRP/baysZH5cHGs1er7fMsZRUqPT7ebhCGipc3N6OjXd3fzV2932L7W1fO/wYW5MS+PmAc5bH5kxg8lRUQG/1uutyenkiSNHuGbWrGHP1HXNzJkkOhzcX1XFi42N7G9vH3BbRPqzNDERgD3qghJQkZzU+H6fbhqXAlnA/wHvAf5ljOndOTvb+/OHvhYeff/xn4SE/wMYnGis1jPh+J6mPjaMp6lDKfLO3HDKpEn8evHiMW3C+V/TpnFnZiaP1dSwoaICgP8pKWFrQwO/WrTI71kPBnKTr29qr76fwXZ/VRUO4BPDfHI0Pz6eBfHxGlcjzHW73Vy1ezeNTidPr1jBFD+SbtEOB08sX870mBguLyykrrub8o4OPrp7N4sSE3lo6dKwaDq9asoUFiYk+NUFxXdBfPXMmSOaunY44hwOnsrJId7h4PJdu2hxOjnY1sZ1e/awctIkNixaFLDv7wxvl5fN9fV8a4AuLxJ6Re3tbBlB8niiy/B2byvr6AhxJKPna6kT7l0e3jt1KrNjY7m/qop93vPWGcnJ/HLRogE/kxgVxbpZs/jT0aPUd3cHJa4njhyhze0e0feXEh3NR2fO5Pc1Nfy0vJyUqCg+Moxxz0R8SQ2NqxFY421K125jzDeBTwKn4Gm18Tfv274Ex3Y8M6gM5tAowhir9UxIn0xL4/dHjnDSG29w1YwZ3JWVdcLAeE63m9u9yYHVU6YMWOabxcVsqKzEbS1/WL6cxDEeoBDgf+bP57WmJr5w8CDfO3yYqq4ubklP54YAXKgsSUzkQm/f1FszMoJ+49jtdvNQdTWXTpvG7BGMSbImNZUnjhzB6XYPe7DDofhTJyLRUNvl77Hgb5kHqqup6e7m0aVL/R5YDWCmtxXABQUFnLFjB0e7u+lyu9l8yimkBDkp4C9jDDempfHNkhL2t7Wx2HvR0R/fBfFYPZ3LiI/nieXLWfv225y+cydlnZ1Ya3ly+fIhpyEcrk+mp/NaUxPfP3yYfd6pmYNZt/wxlsdvILYr2N/NAyNMHk90SVFRpEZHR3STb1+92VBZSYrDwSV+TpkdKtEOBzekpfG9w4fZ8sYbOIAnly8fckDom9LTua+yksdqavj83Ll+r8/fY/PnFRVMj45m5QjHQ/pEWhoPVlfz59paTktKIk7JRRmGGTExTI2Ojoikxlj+rR+t8LiaDCBrrdsYUwpMB5bxn6RGmffnH621G4IYwlitZ0LaUl+PAYo6OvhBWRmPHzly3KwLAAfb2ynv7MQNvNHcPGCZss5OLBBnDA9VV3N3djZjzWEMixMS+DtQ1dWFAaYE8Cbv5vR0rt+7l+0NDawKcjPlv9XVUd3VNeI+3mtSU/ltVRU7mptHNT5Af24vKeEn5eV0Wss7ra0YY0KyvwPtv4uK+EVFBS76r+v+HgvDKRPNyJpMnpWSwprUVP7pbY0TawxPHDnCSUEYZHOkbkhL49ulpdxfVcX/LVgwYLmNVVWsSErirFG2phqO1amprJo8mee9/czjjOHBIJ23fr5oEc/V1vKMt8l4IOrWaI6720tK+HF5OV1jcPx+s7iYn5aXD3hMwdDbHszvxul282B1Ne+bNo25IRpYN5LNjYuL6O4nt5eU8NOKCjrcbqKN4c7S0rD/W+ZrbdFpLXHG8LuqqiFjPq3XgJyfmzPH74cyXz54kF9VVvr1N7HLWu4Y4ff397o6DGCB3W1tfCcC9oOED2MMSxMT2RMB48gN929iKK+xI/9RZR/GGAeQ6f2194iS//D+vDLIIYzVeiak5xsa8PX0dgN1Tidd1h7375jTiduPMr7ldFob0m4PrzQ19cRige0BHFeip2/qGAwYurGqitmxsfzXCAd89D1xCsa4Gn8/doxO7xgB7W73uOnm8nRtLb6OWP3VdX+PheGUccKIv796p7Pn/10hPu76kxYXxwemTePh6uoBp6F9q7mZHc3N3ByCGQfaesUUzPNWnMNBWq9++oGoW6M57p6praVrDI5f651KcrBjyp9tH8l3s9nPbfrbsWNUdXWFfbeDcJUR4UmNvPp6OrznAWcYnkP7U9BrcPfhnLd8A3K+0WsspsF0ehN+/v5N7B7F99f7WjTU15ASmSJhWldr7ZDHVCD/1o/WuGqpYYyJBu7G00qjG/hnr7f/DOwELjbG/Bq4zVp7rM/ns4H/GmULi7Faz4S0esoU3mltpd3tJsHh4Atz5pyQDbytqIh7KyqGVWZNCAdb67tNgYwlISqKa2fN4v6qKn7R3c3UEQ48OpSyjg7+eewY35w3b8RNzmbExnJKUhJb6uv5n0Fm1BiuVpeLij4XscH6HsaStZZWl6vnaVF/dX0kx0Iwj5dg1vVAuSk9nWdqa3muro4r+uknfX9VFXHGcO2svmNRB99Yfn/vnTqVve3tA9aL4dYbYETjj5S2t1Pa0dFTz8HTtS4YfltVxVGnk2g8ybuRHjMj+W7i/Dxv3l9VRXpsLJeGcLagSJYRF8frft4khxtr7XGzpYXrObSvkZ63rpk5k68cPMjGqqp+Z0np68sHD9LqdhNrDF3W+nVsjue/ZRLeliYm8kB1NfXd3aSG6TXpfZWV1DmdRBuDc4BjCsLnniqSkxrfMMbc0Ov36cCpwBw8yaQvWmtLfW96u6V8GM8sKZ8CPmaMeRso9352HrAYqAFGnGwYq/VMVHdlZWGMYUt9PWtSU7kzMzNoZcZKsGO5efZsNnj7pn5hGH1Th+PB6mrceKagHI21qan8oqKCNpcrIGOcWGu5ed8+ap1Orp4xg+KODo50d7P52DHeaGriTD8ulMLVK01N1DmdvG/qVOq6u/utO+F2vITTcTeQ/zd1KhlxcWysrDwhqdHmcvFYTQ1XzJgRksTYWH5/Q61ruPWm3ulka309LzQ0cJGf4wC0u1xcXlhIrDHckJbGG83NFLW38+9jx6ju7CRtBGP3DOTVxkY+f+AA70lNZWVyMs+P4ngY7nfT6nLxYmMjf6+r433Tpg0YY3lHB3+vq+Mbo0geT3QZ8fHUdnfT7nIFfDyaYPtFRQW729u5aPJkOt3usD2H9jXS81bvATl/smDBoEnRh6qq+FVlJV+bO5dYh2NUx2Ywt0nEZ5k3Ob+vrS3gXa4D4eXGRr508CCXTp3KyZMmsTUC7qmMbyrJSOEdL6O/x7ideKZ0fQn4ubV25wCfjwduBK4CTgImAXXezz4PPGOtfaVX+RuAB4GHrbU39Ho9EygBDllrMwOwnkGXN8FEVqWMAGft3Em72807Z5wR8CbzLmvJfvVVliQm8u9TThnVsv5RV8f73n2Xf598MrkBeBL5s/JyvnTwIHdnZXGbt/VHbVcXZ+zciRvYefrpzAjj6fAGc8OePTxdW0vluecyKUwG2xwv7igp4a5Dhyg55xzm9xq34NHqaq7fu5fnTzkl6GPUjDdNTidn7txJo9NJwRlnDDmYsLWWT+7bx0PV1Ty3YgXvnz4dgHdaWjinoIAzkpPZcsopxATg5r6mq4vTd+wg1uFgx+mnj3nCqs3l4vw336S0o4Mdp59+Qn9ln++WlnJ7aSkHzz57wDIyuEeqq/n43r0cOOssFgapxU8wvNjQwOq33+bSqVN5esWKCTPrTX5jI+e9+SYbFy/mpgEGZi5obua8ggLOnzyZf518shJ+EhEOtrWx6PXXeWDJEj4RZt0Jqzs7WblzJ0lRUbyxcqVfM935Kagnrog78q21mdZa08+/eGvtAmvtxwdKaHg/32Gt3WCtvdhaO9VaG2utTbfWnmGt/e/eiQZv+Ye8y7+hz+ul3tczA7SeQZcnMho3paezq7WV15qaAr7szfX1HO7sDEgf7wsnTybam+0drRcaGvhaUREfmjaNb8yb1/P69NhYnl6xgiNdXVyzezfOAcZOCGeNTid/OHqUa2bOVEIjCHwXGA/0GYtmY1UVCxMSuDjMZxwIRynR0TyzYgUtLhcfKSwccMwSn99UVvJQdTW3z5/fk9AAOHnSJO5fsoQXGxu5tbh41HE53W6u3r2bOqeTp3NyQtICJzEqiqdzcjDAFbt2HdfFwMdtLb+rqmLNlClKaIzCXG8yLZLG1ajs7OTKwkKy4+N5eNmyCZPQADgnJYXliYkDjgtW193N5bt2MTM2lieWL1dCQyJGZnw8scaE3bga3W43V+3eTaP3b2IAExpBp6NfZAK4ZuZMkhwO7g/CgKEbKyuZHhPDh3rdeIzUpOhozklJGfVgoZWdnVw1yEXgyuRkfrV4MVsaGvhWScmo1hUKj9fU0O52j3imGRnc/Ph4/t/UqTxQXY3L25pxb2srLzY2clMIBggdL5YnJfHg0qXkNzXx5YMHByyX39jIFw4e5H1Tp/KdfpqxfmzWLL44Zw73lpfz+5qaUcX0jeJitjU0sHHxYk4dw9ls+spKSODxZct4p7WVT+3fT99WtJvr6znU2Tlm0wiPVxkRltTocru5srCQFpeLp1esYPIES2IbY7g5PZ3Xm5t5p6XluPdc1nLN7t1UdXXxVE5OxLa6lIkp2uFgcRgOFvrfRUW82NjI75YsCavZ6fyhpIbIBJAcHc3VM2fyxJEjNPeagWK0arq6+EtdHR+fNcvvge6GsmbKFHY2N/dMAzdcXW43H/FeBD4zyEXgJ9LT+fTs2fygrIynjh4dTchjbmNVFadOmsTpIbwJG+9uTk+nvLOTfx3zjPP8u+pqoo3h4yEYIHQ8uXLmTL6WkcF9lZU8XF19wvs1XV18pLCQjLg4HhvkqfQPFyzgwsmTuWnfPt7tc7PjryePHOHH5eV8bs4crk1LG9EyAum906ZxV2Ymj9XUsKGi4rj3NlZVMS06mg8HIHk8kUVaS42vHDzIK01NPLB0KTlJSaEOJySuS0sj1pgTHsrcXlJCXn09GxYtiujxsWTiWpqYyJ4wSmo8XlPDzyoq+NLcuVwdgdc6SmqITBA3pafT6nbz+yNHArbMh6urcVo76gFCe1uTmooFto2wtcaXDx4kv6mJB5cuZfkQF4H3LlzI2cnJ3LB3b0TMFw6e/sNvtrSoxUCQvX/aNGbGxLCxqoout5uHq6v5wLRpAR2ccqK6JyuLS6ZM4dP79/Nmr5kout1uriospN7p5OkVKwYdET7G4eAPy5czOTqay3btomGYSdDC1lZu3LuX81JS+PGCBSPelkC7bf58PjBtGl8uKuIl7znwSFcXf6mt5fq0tIAljyeqhKgopsfEUB4BSY1Hq6vZUFnJV+fO5aqZM0MdTshMi4nh8hkzeLSmhnZv16w/Hz3K9w4f5ub09AHH2hAJd0sTEylub6czDLpBv9PSwk379nHR5Mn8X5/ZTSKF/jqKTBBnp6SwIikpYF1QrLXcX1XFBZMnsyyAT5DOTkkh0Tt6+XA9Ul3NfZWVfC0jgyv9uAiMczj4U04OiQ4HlxcW0hTAVizBsrGqigSHg3UT+CJ3LMQ6HNyQlsZztbX8prKSo93d6u4TINEOB08sX870mBguLyykzpuQ+HpxMS80NvLbxYs5xY9mr2lxcfwpJ4fDnZ1ct3cvbj8HPm90Orls1y6So6P5Y04OsWGUKHAYw6PLlpEVH8+Vu3dT1dnJI9XVdFur+hcgc+PiKOvoCHUYg3qzuZlb9u9n1ZQpfD9CbzAC6eb0dBqcTp6urWVfWxvX793LmcnJ/GLRolCHJjJiSxMTcQFF7e0hjaPeOzZNanQ0f8jJCcgA3KEQmVGLyLAZY/ikd1rEk994g9uKikY8SKbT7ebaPXs40N7OlKiogA62GetwcNGUKX6Pq+F0u7mtqIiT3niDT+7dy8WTJ3NPVpbf65sbH8+Ty5ezv62NM3fu5JydOwf8bnzrGqxMMLW6XGyqqeHKGTMiavCmSHVDWhou4AsHD5LscLBaM54EzMzYWJ7KyaGio4Mzd+xg4auv8tPyctbPnj2sriDnTZ7MTxcu5K91dVz85ptDHr/fLCoiMz+fg+3t/H7ZsiFnYQmFydHRPJ2TQ2N3N2fu3Mm3S0qYExvLIg0QGhAZcXEB737iz98Gf8t8+cABzikoIBrYtGyZBr8EVk2ZQnZcHLcWFXHqG2/gdLt5cvlytVySiOab1jUY42r4e775ZlER2a++SnFHB08sX86sCB6bZmKNOCQywR32Pp16t7WVg+3tGGO4ewRPgW4vKeFJbzeWLQ0NfKe0dETLGciaKVP47+JiKjo7mTPETcftJSXcW1FBu/eEfeqkScO+CFyVmsqqKVPY6k2kFLS08HpzM5dOm3Zcub/V1fFCYyPd1vJOa+uIv7+R+uORIzS7XAGZaUaG9mh1NQ7ADbS73dwV4Ho+0Z2VksKa1FT+6W2V5QAmRUUNezmfnT2bX1dU8JJ3dqfBjt9tDQ24gBhjyKuvD9upeVdMmsR7pk7lL3V1ABzt7g74eXaiyoiL46XGxoAu8/aSEn5SXk6ntX79/RiszPMNDbjxzH24oaJC+xxPC6aM+Hi2e/dbnHeMDX03EskWexPVwRhXw99zUu+/if88dowLI3h2t5AkNYwxdwDfAe601t4RihiCzRhzA/Ag8HDf6WDDnTEmCrgcOAM4EzgdSAEKrbUrQhmbjE5+ryld291uttTXc/cIlvNMbS2+SQdHs5yBrPHeaGypr+f6IZ7abmlo6EloALw6wmlrW3tNo9htLVsaGgZtLRKM7R7KxqoqliYmcsHkyWO41olrq/fmAsAJY76/J4L6Xl2+3IxsLB1jDAm9Epn+HL/d1ob9/qzu6ur5f1cExBspMuLiqHc6aXW5SBpBEq0/T9fW0unt/uR3/RuiTKf2+XGae/2N1ncj48Gk6Ggy4uKC0lJja0PD8M9JEX5MqaXGIIwxpcB8IMtaWxraaMZUMvCHUAchgbd6yhTeaW3tSQKcM4IRww91dFDa0YEBLJDgcPQkIQLllEmTmBYd7VdSI7rXYJmjiaX3d5PgcPCZ2bO5vc90kneWlrKhooIua4k1JuDbPZjC1lZeaWriRwsWaIDQMdK3Tozl/p4oAvUdr01NZVdb25DH768rKyNmf6r+BYdvBpTyzk6WeJt/j0Zxe/sJfxP9qX/joY6Opf+XmsqeXse4vhsZD5YGaVrXUyZN4jXvQNwT5XyjpIb0pxt4DNgJ7AAmA38NaUQSEHdlZWGM4bm6Ova0tvJmczPdbrffgwK1u1xcvmsXMcZwQ1oab7a0sCY1lTv7nChHy2EMq1NT2VJfj7V2wJv4v9TW8kpTE6ckJRHvPSGPNBbfd7Olvr5nOX27sfxfdjaxxvDjsjLmxccHfLsHc39VFTHGcH0ETrMVqfqrExJYgfqO/T1+E7yDEEfC/lT9C46M+HgAyjo6Rp3UaPP+TYx3OPjkrFkUNDcPq/5Feh0dSzoeZDxalpjIA9XVg17rjsQK7wD+pyQlcem0aRPifGOsn6OFB3SlEdL9ZDQtNSK5+0lfxphVwPOMXfeTsa+UE9Cmmhqu3bOHL8+dy08WLhyyvLWWG/ft48Hqap5dsYIPTJ8e1Ph+U1nJp/fvZ+9ZZ/V74ekb2HNxYiIvnnoq8QFqRuyPbxYX83+HD3P43HOHHPMjEDrdbma/8gprUlP5Q05O0NcnIjJeFbe3s+C113hgyRI+MYrxiay1fHzvXh6rqeFvJ53Ef/Xpry4iMpRfVVTw2QMHKDvnHOZ6E66BcMWuXexobqb0nHPCqXVvUAMZ9bDBxpizjTE/NMbsMMbUGGO6jDGVxpg/GWPO8ePzWcaYx7yf7TDGFBpjvmqM6bcVifG4zhizzRhT7/1MkTFmgzEmo5/ymcYY601QDBSDNcbYXr/f4P19vvelEl8Z77/Mobarz/KnG2N+ZYwp7xXv/xpjBnxEYIy51BjzD2NMrfc7LTPGPGyMWdan3GJjTJMxxmmMuaif5Sw3xrQaY7qNMecOJ24Z39bNmsXn58zhp+XlPFFTM2T531ZV8WB1Nd+aPz/oCQ3wDBYK9Du1a4vT2dNi5E85OWOa0AC4MS0NN/BggKbHHcozR49yzOnUlI4iIqPkS0SPdgaU+yorebSmhjsyM5XQEJERWRqEGVBc1vJ8QwNrUlPDKaERdIGYC+lu4MtADPA68CxQB1wBvGSMuXKQz2bh6d5wCbANT2uABcCPgD8aY46Lz3j2zGPAI8B5wBvAn/Fkfj4LvGWMOTMA23QQeBho9f7+lPd337+WYSwrFXgNuMr781/ADOB/gC39JTaMMffg6e7xHqAQ+BPQCFwPFBhjLvWVtdbuBz4FRAGPG2Om91pOIvBHIBG4zVqbP4y4ZQL40YIFXDB5Mjfu28e7LQNX69eamvj8gQO8d+pU7hij5mkLEhKYFxd3QlLD12JkT1sbTyxfzvwAZrb9tTAxkUumTOF31dW4x6C128aqKjLj4yO+v6OISKjFORzMjImhfBRJjZcbG/nSwYO8f9o0vjV//tAfEBHpRzCSGm+1tFDvdE64a8ZAJDV+BMyx1p5irf2AtfYj1tqTgA/iGcz814O0SLgebyLDWvtRa+1/AScBFcCHgU/3Kf8Z4GNADXCatfY91tqrgUXAL4CpeJIho2oPbq19ydtlpNb70testTf0+lc7yMf7+iBQhWcbr7DWfsgb77vAOcAdvQsbY94HfANPQuUSa+3F1tqPebt93ArEA5uMMTN7xft74LfAHOAR85+03AZgOfB3PPtJ5DixDgd/WL6clOhoLi8spKG7+4QyNV1dXLFrF3Pj4ti0bBlRY5T1Nd6BOJ9vaMDVK3Hw0/Jy/nD0KN/Lzmbt1KljEkt/bk5Pp7Sjo9+WJIFU1N7O1oYGbkxLwzGBMu4iIsGSERc34pYaVZ2dXFlYyPy4OB5dulTnZREZsbTYWCZHRQV0WlffdenqCJ6edSRGndSw1v7TWntC23Vr7XN4WglMxdMSoz/twGettR29PncA+Lb31y/3Kf9V789vW2sLe33GBXwNKMPTZeQjI9iUYLHAZ6y1DT0veL6vL3p//bQxpvejZt82/sxa+8JxC7L2h3hae0wGbu6zni8C7wD/Bfy3MeZ64AagHLjehmLwFIkI6XFx/Cknh9KODq7fu/e4lgdOt5urd++mzunkqZwcpsbEjGlsa1JTqXc6edM7gvO2+npuLSri8unTuTXjhN5mY+qy6dOZGh3NxiB3QfldVRUOGFXfbxER+Y+M+HjKOjqGLthHt9vNVbt30+h08syKFUwZ47+JIjK+GGMCPgPKlvp6licmkj4GY76Fk0C01PCNGXGDMeZHxpj7jTEPGWMeAnyDSi4e4KP/ttYe6ef1TXhaeSw0xszxrmMukO19/dG+H7DWdnk/B7BqxBsTeO9Ya9/t+6K19nk8LVKSgdMBvOOInO8t8tAAy3vQ+3NVn+V1AFfi6RpzN/ArwAVcY62tG9UWyLh3/uTJ/HTBAp6rq+PuQ4d6Xv9GcTHbGhr47eLFnJacPOZx+bLMWxoaKO/o4Krdu1mUmMiDS5eGvJ9gfFQU182axZ9razna1RWUdXS73TxYXc37pk0bkwFJRUQmgpG21PhaUREvNTZy/5IlnDRpUhAiE5GJJpBJjU63mxcbGydc1xMIwJSuxphPAT/BM27DQFIGeL2kvxettV3GmCo83Snm4rn5n+N9u6p3y44+irw/5wzwfij0u41epfxnGwGmAXF4EjeHBvjMgNtord1vjLkVuA/Pvv2OtfalEcQsE9D6OXN4rbmZ75SWckZyMs0uFz8uL2f97Nlcl5YWkpjS4+JYnpjIP+rqeOboUdrdbp7JySElOjxmo7559mx+VlHBIzU1fDUILUf+fuwY1V1dGiBURCSAMuLiaHK5aHI6/f57sqmmhp9XVPCluXO5RlNri0iALEtK4uGammGdjwaS39hIu9vN2gmY1BhVSw1jzBl4WgTEAP8NLAUmAQ5rrQHu8RUdxWp8beFNn9/7DWm4C+87GGmI9N3G3q/1NeA2GmOigKt7vXSmCfXjbIkYxhh+s3gxJyUmctmuXVyzezdzYmP5vwULQhaT0+0mweFge2MjrzU388CSJSz1zr0dDnKSkjg3JYWNlZUEuoeX0+3mqwcPEmMMrzQ24nS7A7p8EZGJaq635Zs/g4U63W5u3LuX6/bsISM2lu9lZQU7PBGZQAI5WOiWhgYcwMUTbDwNGH33k4/gucn+ubX2R9bafdba1l7jNywc4vOZ/b1ojIkFfI8mK70/y70/Zw8yEKjvL01Fr9d87cIHaicY7GGrM/14z7eNtUAnnv0y0Of620afO4CL8MxCUwi8H/iKv4GKJEZFcd7kyXRaixuoczqP644y1m4vKWFXq2cSomg8IzqHm5vS09nX3s7LjY0BXe4n9+6lqKODbmv5eUUF3yktDejyRUQmqoxhTOv67ZISHqyuxgJHnU7+N4R/E0Vk/AloUqO+njOTk5kcJi2ax9Jokxq+qQfK+r5hjJkB5A7x+fd4y/V1jTe2ImttOYD3Z7H39Wv7WV8MnplRwDM9rM9RPImNaQOs632DxOdLiIymZpxijFnR90VjzMV4upC0ADsBrLVO4GVvkesHWN4N3p/b+ixvDXAb0AB81PuvDbjHGHPWKOKXCebNXomDDrc76LN7DGZrQwOd3hypE0Iay0A+OnMmyVFRAR0wtLqzkyeOHu35vT3E+0FEZDzJ8E4F7s9goc/V1fU0nQ3130QRGX+y4+OJNmbUSY0mp5PXm5om5HgaMPqkxl7vz+uNMT0tIYwxycADwJQhPp8I/LJ3ywtjzALgu95ff9an/E+8P79rjFna6zNRwP/haXVxCPiT7z1rbTfwovfXu3p3xzDGXADcNUh8vtYQy4bYjsEY4D5jzORe653Bf7btt9ba9l7lfdv4JWPM+b1exxjzFeBcoBG4v9frs/AMkuoAbrTWlnpnh/kCnq5BTxpjpoxiG2QCWT1lCgkOz6khweEI6ckxnGIZSFJUFB+bOZM/Hj3a75S4w9XtdnPl7t1Ya4n3nq7CddtFRCLR7NhYDP51P4nu1YtX52IRCbQYh4NFCQmjntb1hYYGXDBhz1GjbZvyIPAlYCVQbIx5Cc9N/EV4Wjk8AHxykM8/ClwKFBljXsbTRWQ1EA88B2zoU/4+PLODXAO8bYx5HqgHzsIzM0o9cKW1tu9fqduBC4FPAxcbYwrxJEBOB74HfGuA+J7BM8vIJmPMv/G0ggD4+jBmFHkWzywwRcaYbXi+80vwDJ76hje2HtbavxljfgB8HXjBGPMinu4pJ3mX0wFc65tG1zsmyCZgFrDBWvt0r2X9zhizGk8Llt8BV/gZM8aY+/DsV/jPQK/ZxphXexW731p7PzKu3JWVhTGGLfX1rElN5c7MTMUyhJvS0/lNVRWPHznCZ+eMbpxi3+j6jy5dyp62trDfdhGRSBPjcJAWGztk95MWp5OD7e2clJREojehoXOxiATa0sTEUSc1tjQ0EO9wcF7KQPNzjG+jSmpYa+u9g4V+F09Xk0uBI8DTeG7WPzXEIoqBM/EkFlYDk72vPQDca609bmQ8a601xqwD/gHcDJwDJOC56f8VcI+19oSuMNbaV7zdM+4AzsaT0CgErrfWbjLGDJTU+CWeG/p1eMan8LUo+V/A36RGvTfO7+Hp6jIdTwuQXwLfs9a29hPvN7wJos/h+X7Ow/O9Pgp831q7u1fx/wHWAG8BX+1n/Z/yLuNyY8znrLW/9DPu5Xi+q94S+rz2Tz+XJREk2uHg7uxs7g51IIRXLIM5PTmZUydNYmNVFZ+ZPXvE080+3mt0/Wu9M86E+7aLiEQif6Z1ffLoUVrdbn69eDHnTZ48aFkRkZFampjIc3V1dLvdxDhG1pFiS30956ekEB8VFeDoIoMJ9Ij9IgGgSikR576KCtYfOMAbK1dyxgiy5O+0tHBOQQFnJiez+ZRTRvxHTUREhvaRXbsobGtjz1kDDzt2zs6dNLtc7DrzzBEnq0VEhvJIdTUf37uXvWedxRLvwKHDcaSri1mvvML3srL45vxgz4ExYkE9ieqqWUQkAD42cyYJDgf3j2DA0Prubi7btYvU6GieXL5cCQ0RkSDLiI+nrKNjwOm4321p4bXmZm5KT1dCQ0SCapk3kbGn9YQG/H7Z6h3AeO0EHU8DlNQQEQmIKTExXDljBo8fOUKL0+n359zWcu2ePZR1dvKnnBzS4gaasVpERAJlblwcrW43jQOcr++vqiLWGK6bNWuMIxORiWbJKKd13Vxfz5ToaFYmJwcyrIiipIaISIDcnJ5Os8vFH3pNxzqU7x46xN+PHePehQs5V322RUTGRIY3gdzfuBodLheP1tRw+YwZTI+NHevQRGSCSYmOZnZs7IiTGlsaGlg1ZQpRE7hVmZIaIiIBcv7kySxNTPS7C8rf6uq4o7SU62fN4jOzZwc5OhER8RksqfFUbS31Tic3paePdVgiMkGNdAaU4vZ2Sjs6WDNlSuCDiiBKaoiIBIgxhk+kpZHf1MTJb7zBbUVFON3uE8o53W4+s28fH3r3XWbGxPDLRYvUZ1tEZAwNltS4v6qK7Ph4LpngNwkiMnaWJSayt61twHF+BrLFO57Gmgk8ngaMckpXERE5XoX3Avnd1lZ2tbby66oqpsXEHFemtquLBpcLgCank+8fPszd2dljHquIyESVHhdHFFDeJ6lxoK2NbQ0NfC8rC4eSzSIyRpYmJtLkclHd1UX6MMZX21JfT3psLEtHMGvKeKKWGiIiAfRaU1PP/y2ek+xZycnH/evd57HD2p4su4iIjI0oY0iPi6Oso+O41++vqiIKuCEtLTSBiciEtHQEg4W6rWVrQwNrUlMnfItftdQQEQmg1VOm8E5rK+1uNwkOB5+aPfuEVhi3FRVxb0VFT5mJ3mRQRCQUMuLijut+0u1281B1Ne+fNm1YT0pFREZrWVISAHva2rjEz+vCXa2tHO3unvDjaYCSGiIiAXVXVhbGGLbU17MmNZU7MzNHVEZERIIrIy6ON1taen5/rq6OI93d3KyBm0VkjM2OjWVSVNSwWmpoPI3/UFJDRCSAoh0O7s7O5u5RlhERkeDKiIvjubo6rLUYY9hYVcWc2FjeO3VqqEMTkQnGGMNS72Ch/tpSX8/ihAQy4uODGFlk0JgaIiIiIjLhzI2Lo93t5pjTyeGODv517BifTE8/btwjEZGxMpykRrfbzfbGRrXS8FJSQ0REREQmHN/TzbKODh6oqgLgxvT0UIYkIhPYssREyjo7aXE6hyz7enMzLS6XkhpeSmqIiIiIyIST4R0MtLSjgweqq3lPairz1YxbRELENwPKvvb2Ictuqa/HAJdokFBASQ0RERERmYB8SY3fVVdT1tnJTWqlISIhNJxpXbfU13PapElMjYkJdlgRQUkNEREREZlwZsbGEm0Mf62rY0ZMDB+cPj3UIYnIBLYwIYEoYE9r66DlWl0u8pua1PWkFyU1RERERGTCsdaS6B0UNDs+XhfFIhJSDiAlKor7Kiu5ragIp9t9Qhmn282Ne/fSbS2HOzr6LTMRGWttqGMQ6UuVUkRERILqtqIiflBWhhuIN4avZGRwd3Z2qMMSkQnqtqIi/q+sDBcQDZw6aRLnT558XJmXGxspaGnBDSQ4HHx57txIOW8FdVqp6GAuXEREREQkHG1taMD3jLPDWrbU13N3SCMSkYlsa0MDLu//ncCbLS0c6DNoaIvL1XPeane7dd7yUlJDRERERCac1VOm8E5rK+1uNwkOh/qni0hI9T0n9dcK47aiIu6tqNB5qw8lNURERERkwrkrKwtjDFvq61mTmsqdmZmhDklEJjB/zkk6b/VPY2pIOFKlFBERERERGR+COqaGBnoWERERERERkYikpIaIiIiIiIiIRCQlNSTgjDEOY8yXjTF7jTEdxpgyY8yPjTFJoY5NRERERERExg8lNSQYfgr8BNgNfB74I/AF4DljjOqciIiIiIiIBIRmP5GAMsbk4ElkPG2tvaLX6yXAz4GrgcdDFJ6IiIiIiIiMI3pqLoF2DZ7Rbe/t8/pGoA24dqwDEhERERERkfFJSQ0JtDMBN/B67xettR3AW973RUREREREREZN3U8k0GYDtdbazn7eqwDOM8bEWmu7er9hjLkFuMX762PW2uuCHKdIUBljbrHW/jbUcYiMhuqxRDrVYRkPVI8l0gW7DqulhgRaItBfQgOgo1eZ41hrf2utPcNaewawLFjBiYyhW4YuIhL2VI8l0qkOy3igeiyRLqh1WEkNCbQ2IG6A9+J7lREREREREREZFSU1JNAqgenGmP4SG3PwdE3p6uc9ERERERERkWFRUkMC7Q089eqs3i8aY+KBU4EdfixDfQZlPFA9lvFA9VgineqwjAeqxxLpglqHjbU2mMuXCcYYcxLwNvCMtfaKXq9/Hvg5cJ219rFQxSciIiIiIiLjh5IaEnDGmF8AnwOeAf6OZ+DPLwAvA6utte4QhiciIiIiIiLjhJIaEnDGmCjgS3hGuc0EaoEngduttS2hi0xERERERETGEyU1RERERERERCQiaaBQCQvGGIcx5svGmL3GmA5jTJkx5sfGmKRQxybSmzFmsTHmLmPMq8aYo8aYZmPMW8aY/+mvvhpjlhhj/myMqTfGtBpjXjTGrA5F7CIDMcYkGmNKjDHWGPPLft5XPZawZIyZaoz5kTHmoPf64agx5nljzIV9yqkOS1gyxkwyxtxmjHnXe01Ra4x5xRhzgzHG9CmreiwhY4z5pjHmj8aYYu/1QukQ5f2ur6O9F4wewfaIBMNP8Yy78QzwY/4zDsdpxpi1GodDwsgngfXAs8AmoBu4BPhf4CpjzDnW2nYAY8wC4BXACfwf0AjcDPzLGPNf1trNIYhfpD93AdP7e0P1WMKVMWY+sA2YBPwO2A9MBk7GM428r5zqsIQlY4wD+AdwHvAw8AsgEbgGeBDP9fDXvWVVjyXUvgccAwqAKYMVHEF9HdW9oLqfSMgZY3KAdxl4xpR11trHQxWfSG/GmDOAA9baxj6v/y/wP8DnrbW/9L72B+AK4HRr7Vve1yYBhUAHsNTqJCwhZoxZCbwO3IrnQmKDtfZzvd5XPZawZIx5Ec/YXWdZa6sGKac6LGHJGHMunhu/e621X+71eiywF5hqrZ3ifU31WELKGJNtrS32/n8XMMlamzlAWb/rayDuBdX9RMLBNYAB7u3z+kagDbh2rAMSGYi1dkffhIbXk96fKwC8zeU+CGzzncy9n28B7gcWA2cGN1qRwXkHdt4I/BN4up/3VY8lLBljLgIuAP7PWltljIkxxiT2U051WMJZivdnZe8XrbVdeAbabwXVYwkPvoTGUEZQX0d9L6ikhoSDMwE3nieFPay1HcBb6CQtkWGu92eN9+fJQByQ30/ZV70/Vbcl1L4MLMUzDXd/VI8lXL3P+/OwMeY5oB1oNcbsN8b0vgBWHZZw9jrQANxqjLnSGDPPOw7BPcDpwB3ecqrHEkmGW19HfS+opIaEg9lArbW2s5/3KoDp3mZ4ImHJ+7T7djz9Bn3N42Z7f1b08xHfa3P6eU9kTBhjsoA7gbustaUDFFM9lnC1xPtzIzAV+DhwI9AFPGqM+YT3fdVhCVvW2no8T7SPAX8ADuHpdrIeuMJau9FbVPVYIslw6+uo7wU1UKiEg0Sgv0oMnj5XvjJdYxOOyLDdC5wD3Gat3ed9zdcMur+63dGnjEgo/AooAX4ySBnVYwlXyd6fzcAl3ub6GGOeAYqB7xljHkZ1WMJfC7ALzwDkr+BJ0q0HHjfGfMham4fqsUSW4dbXUd8LKqkh4aANmDnAe/G9yoiEHWPMd/E03f+ttfaeXm/56mxcPx9TvZaQ8jbPfw9wkbW2e5CiqscSrtq9P3/vS2iA58m3MeZZ4Ho8rTlUhyVsGWNOwpPI+LK19te9Xv89nkTHRu8sEqrHEkmGW19HfS+o7icSDirxNCvqr+LPwdMcSa00JOwYY+4AvoVn2rVP93nbN+hXf81Bfa/11yxPJKi859qfAH8Hqo0xC40xC4H53iKTva9NQfVYwle592d1P+/5ZkJJRXVYwtuX8dy0/bH3i9baNuBveM7LmageS2QZbn0d9b2gkhoSDt7AUxfP6v2iMSYeOBXYEYKYRAZljPkO8B3gEeCmfqZRexdPU7pz+/n4Od6fqtsSCgnADOBS4ECvf9u871/r/f0mVI8lfPkGlJvbz3u+146gOizhzXeDF9XPe9G9fqoeSyQZbn0d9b2gkhoSDp4ELPClPq/fjKf/1KaxDkhkMMaY2/GMSP4o8AlrrbtvGe+0Vc8Bq4wxp/T67CQ8N4sH6DPKs8gYaQWu7OffZ73v/9P7+7OqxxLG/oxnPI1rvfURAGNMOvBh4IC19qDqsIS53d6fN/R+0dtS7kNAPVCkeiyRZAT1ddT3gubEh4siY88Y8ws84xI8g6dJ9DLgC8DLwOr+bhpFQsEYsx74JXAY+DaeKah6q/EO6oW3Sf/rQDfwU6AJzwn6JOBSa+2/xipukaEYYzLxDBy6wVr7uV6vqx5LWDLG3AL8BigEHgBigc8A6cD7rbX/9pZTHZawZIyZDxTg6Sq1Cc9171Q89TMTWG+tvc9bVvVYQsoYcx3/6ar6eTzn3B97fz9krX20V9lh1dfR3gsqqSFhwTsl5peAW/CcxGvxZO1u92b7RMKCMeYhPFMHDmS7tXZVr/LLgO8DF+M5+RcAd1hrNwcxTJFhGyip4X1P9VjCkjHmcuBWPBfKbiAfuNNa+3KfcqrDEpa8A4HeDqwBZuEZBPct4F5r7dN9yqoeS8gYY7bhqXv9Oe7611ve7/o62ntBJTVEREREREREJCJpTA0RERERERERiUhKaoiIiIiIiIhIRFJSQ0REREREREQikpIaIiIiIiIiIhKRlNQQERERERERkYikpIaIiIiIiIiIRCQlNUREREREREQkIimpISIiIiIiIiIRSUkNEREREREREYlISmqIiIiIiIiISERSUkNEREREREREIpKSGiIiIiIiIiISkZTUEBEREREREZGIFB3qAET6YUMdgIiIiESY006Dt94autypp8KbbwY7GhGZ4HRKOo4J5sLVUkNEREREIt+550Js7OBlYmPhvPPGJh4RmdB0Sho7xlo9FJewo0opIiIiw1NVBdnZ0NExcJmEBCguhrS0sYtLRCYknZKOo5YaIiIiIiKDSk+HT3xi4EejsbGe9yfA3YOIhJ5OSWNHLTUkHKlSioiIyPAN9mh0Aj0SFZHwoFNSD7XUEBEREREZ0kCPRvVIVERCQKeksaGWGhKOVClFRERkZPp7NDrBHomKSPjQKQlQSw0RERERET/1fTSqR6IiEkI6JQWfWmpIOFKlFBERkZHr/Wh0Aj4SFZHwolOSWmqIiIiIiPjP92jU4dAjUREJOZ2SgkstNSQcqVKKiIjI6FRVwdVXw5NP6g5CREJugp+SgtpSQ0kNCUeqlCIiIiIiIuODup+IiIiIiIiIiPSlpIaIiIiIiIiIRCQlNUREREREREQkIimpISIiIiIiIiIRSUkNERERkQDKL8vnnhfvIb8sP9ShTGj+7AftKxEZKzonBU90qAMQERERGS/yy/JZ88gaOpwdxEfHs+X6LZybcW6ow5pw8svyueThS+h0dRJlolidtZrpidOPK1PbVsvWkq24rVv7SkSCKr8sn9WPrKbD2aFzUhAoqSEiIiISINtKt9Hh7MBi6XR2sq10my5KQ2Bb6TY6XZ0AuKyLgqoCpiZMPa7MsfZjuKwLgC5Xl/aViATNttJtdDp1TgoWJTVEREREAmRV5iocxoHLujDGsCpzVahDmpAumHcBAAZDfHQ8z13z3Ak3B/ll+ax6eBVdri6iHdHaVyISNL3PLwnRCQOek3pamDmidE4aBo2pISIiIhIgZ805i8SYRABS4lI4Z+45IY5oYvvYSR8bsAn3uRnn8peP/gWAa0+6Vk9ERSRoZk2ahcXy/kXvH/SclHd9HnFRcbx3wXt1ThoGJTVEREREAqSgqoDmrmbes+A91HfU8+6Rd0Md0oS0uXgzDuPgF//1i0FvDN676L2cMfsM9tbtHcPoRGSi2Vy8GYAfvueHg56TLpx3Ie9d+F7eOfIO1tqxCi/iKakhIiIiEiB5xXkA3LPmHs/vRXmhDGfCyivO44zZZ5CakDpk2bVZa3m1/FWaOpvGIDIRmYjyivOYkzyHJdOWDFl2bfZaShtKKaovGoPIxgclNUREREQCZHPxZk6ZdQor01eydPpSNpdsDnVIE05jRyOvV7xObnauX+VzF+Tisi62l24PcmQiMhG53C62FG8hd0Euxpghy/vOXb7WHTI0JTVEREREAqCtu42Xy17uuSDNzc5le+n2nhHvZWw8X/o8LuvyO6lxfsb5JEQn9LSyEREJpIKqAuo76v0+Jy2etpiMlAydk4ZBSQ0RERGRAHjh0At0ubrIXfCfpEa7s51Xyl4JcWQTS15RHkkxSX4PshcXHcdF8y/SDYSIBIXv3LI2e61f5Y0x5GbnsrVkKy63K5ihjRtKaoiIiIgEQF6RZ9T6C+ddCHim8It2ROtmeYzlFedxcebFxEbF+v2Z3Oxc9tbupbypPIiRichElFecxymzTmFm0ky/P5O7IJeGjgZ2VO4IYmTjh5IaIiIiIgGQV5zHBfMuICEmAYDkuGTOmXuOkhpj6FDDIQ4cO+B3M28fX+saDewqIoHU2tXKy4dfHvY5aU3WGgD9/fCTkhoiIiIio1TdUs27R9494cI1NzuXnZU7qWurC1FkE4vvBmC4NxAnzTyJWUmzdAMhIgH1wqEX6HZ39yRO/TUjaQanpZ2mc5KflNQQERERGSXfKPV9L1xzs3OxWLaWbA1FWBNOXnEes5Nns3zG8mF9zhjD2uy1bC7ejNu6gxSdiEw0ecXHd0scjtzsXPLL8mnpaglCZOOLkhoiIiIio5RXnMe0hGmcmnbqca+fOedMUuJS9LRtDLitmy3FW1ibvdavaRP7ys3O5WjbUd6peScI0YnIRLS5ePNx3RKHI3dBLt3ubk037QclNURERERGwVpLXlEea7LX4DDHX1pFO6K5JPMS8orzsNaGKMKJ4a3qt6hrrxt21xMf38wEGldDRAJhoG6J/rpg3gXER8crKe4HJTVERERERmH30d1UtVQNeOGam51LaUMpRfVFYxzZxOJLRvg7bWJfc1LmsHzGct1AiEhADNQt0V/x0fFcOO9CnZP8oKSGiIiIyCgMNTilZtYYG3nFeZw08yTSJqWNeBm52bm8ePhFOpwdAYxMRCaigbolDkdudi67j+6moqkicIGNQ0pqiIiIiIxCXnEei6YuYv6U+f2+v2jqIuZNnqenbUHU3t3OS4dfGnErDZ+12WvpcHbw8uGXAxSZiExEg3VLHA7fOc3X6kP6p6SGiIiIyAh1ubrYXrp90D7Txhhys3PZWrIVl9s1htFNHC8efpFOV+eI+677XDz/YqId0UpAicioDNUt0V+npJ3CjMQZbC5RUmMwSmqIiIiIjNCr5a/S2t06ZJ/p3OxcGjsb2VG5Y4wim1jyivKIjYrlovkXjWo5yXHJnDv3XCU1RGRUhuqW6C+HcbAmew2bizdrsOlBKKkhIiIiMkJ5RXlEmSguybxk0HJrstdgMLpZDpK84jzOyziPpNikUS8rNzuXN6vepLatNgCRichENFS3xOHIzc6luqWaXUd2BSCy8UlJDREREZERyivO46w5ZzE5fvKg5aYnTue09NOU1AiCmpYa3q55e9RPRH1yF+RisWwp3hKQ5YnIxOJPt8Th8C1Hfz8GpqSGiIiIyAjUt9fzRuUbfg9OuTZrLfll+bR0tQQ5solla8lWYPTNvH3OmH0Gk+Mm6wZCREbE326J/sqYnMGSaUt0ThqEkhoiIiIiI/B86fO4rdvvm+ncBbl0u7vZXro9yJFNLHnFeaTGp7IyfWVAlhftiGZ11mryivPUh11Ehs3fbonDkZudy/bS7XQ6OwO2zPFESQ0RERGREcgrymNS7CTOmXuOX+UvmHcB8dHxetoWQNZa8oo90yZGOaICttzc7FwONx7mwLEDAVumiEwM/nZLHI7cBbm0O9t5peyVgC1zPFFSQ0RERGQE8orzWJW5ipioGL/Kx0fHc+G8C5XUCKB9dfsobyoPWNcTH1+z8bwi7SsR8Z+vW2Kgz0mrMlcRZaL092MASmqIiIiIDFNJfQlF9UXDvnDNzc5l99HdVDRVBCmyicWXdAj0DcSC1AVkTsnUDYSIDEtPt8QAjafhkxKXwjlzz9E5aQBKaoiIiIgMk+/CcthJDe+F7ubizQGPaSLKK85jQeoCslKzArpcYwy52bk8X/o8TrczoMsWkfErryiP5Nhkzp5zdsCXnZudy87KndS11QV82ZFOSQ0RERGRYcorzmNO8hyWTl86rM+dPOtkZiTO0NO2AOh2dbOtdFvAW2n45Gbn0tTZxBsVbwRl+SIy/gy3W+Jw+Kab9s34JP+hpIaIiIjIMLjcLrYUbyF3QS7GmGF91mEcrM1ey+bizZpZY5Req3iN5q5mv6fUHa7VWasxGCWgRMQvvm6JwTonnTn7TJJjk3VO6oeSGiIiIiLD8Gb1m9R31I+4hUBudi41rTXsOrIrwJFNLHlFeTiMg9VZq4Oy/GmJ01iZvlI3ECLil5F2S/RXTFQMl2Rdou6L/VBSQ0RERGQYfINTrslaM6LP+57i6WZ5dDaXbOaM2WeQmpAatHXkZufyavmrNHc2B20dIjI+bC7ePKJuicORm51LSUMJRceKgraOSKSkhoSde168h/yy/AHfzy/LH5dlAiHctikQZcIplrEuEwjhtk2B2u5Iq1tjScdm8Mv8afefSJuURnF98YBlBpMxOYN5KfP49Y5fh802hVMZf5aRV5RHflk+y6YvG7BMIOQuyMXpdrL+7+vH5BjXcRf6ujVRy4RTLJFY5qXDL/HX/X/llLRTht0tcTh8rUC++u+vhsV2+1sm2Iz6c0q4MXcaG2WiWJ21mumJ0497r7atlq0lW3FZF5FWBiA2KpYt12/h3IxzR/Ud9Se/LJ9LHr6ETldnWG33aMqEUyzBKGOxxEXFBbVOrH5kNZ3Ozp4m2mGx3dYSFz3y7X7l8Ctc/PDFON3OsKlbbusmPjo+aPvSX/ll+Vz00EWj+m78KROux9RYlckrzsNgRrzP88vyuejBi3Da4O6nSCzj7zK2lGzBbd3ERcXx/MefD9pxt710O6seXjWq/T2cda19ZO2A9SKc9lMklgmnWMKtTDjFEollep+TYqNi2fbxbUE7T7xy+BUuePACLDbk2927jB/XtMHL9ADRwVy4yEi5rIuCqgKmJkw97vVj7cdwWVdElalrr+sp0+XqYltpcE5020q30enqHJNtGqsy4RRLsMoEvU44O7HYcbXdf9z9x54pFsOpbgVzX/rr2X3Pjvq78adMOB9TY1EGwGJHvM+3lW7DjTustilcyvi7DLf1fH9OtzOox90rZa8Ao9vf/nrwrQdx2oGP33DaT5FYJpxiCbcy4RRLJJbpfU5yuV1BPU9sP7S95/+h3u6+ZUJ5HaSkhoSdKBNFbFQsz13z3AkHRX5ZPmseWUOXqyuiypz/wPlYLLFRsazKXDXKb6h/F8y/AKDnaVI4bPdoy4RTLIEuc+GDF+KyrqDWiVWZqzDGYK0lITohLLb74ocuptvdTUxUzKi3ezTnikBu0+pHVtPh7MBhHEHbl/6Kdnj+rAf7PBqOx1Soyoxkn6/KXEVcVFzYblOknfeDedytylxFjCMmYOetwbR3twMDH7/htJ8isUw4xRJuZcIplkgsM9bnpPjo+LDY7v7KhOo6SN1PJOzc8+I9dlXmqgGzfPll+Wwr3UYklbnyD1fy531/ZvN1m7k48+J+y4zWq+Wvcu7vzuXqFVfzhbO+EBbbHYgy4RRLIMt8d/t3uX3b7Wx8/0ZuOv2mfsuMVoezg8nfn8zp6afz4/f8OCy2e9M7m7j2mWv5xvnf4J6194xouz7w+w/wZtWbrD9zfVjUrVcOv8Klj1/KyWkns/2G7f2WGSvXPn0tfz/wd7523te4JPMSHZshLDOUcIs3nMqM1T7w16NvP8r1f76e2y64jbvX3B209ay4bwUJMQlcvvRyHXdBKhNOsYRbmXCKJRLLjOU5KZy2298yBLn7iZIaEo7GXaV8dt+zfOiJD7H1+q1cknVJUNbx3e3f5TvbvsOR/z5yQn83CT9HW48y80cz+e4l3+VbF30rKOvYWrKVNY+s4dmrn+UDSz4QlHUMl7WWOT+Zw4XzL+TJjzw57M93u7qZ+n9Tufaka/nV+38VhAhH5qZnb+JPu/9E7a21Pa0lxpq1lvQfp7M6azWPX/F4SGIQGY98x9aa7DVsunxTUNZR1VzF7J/M5gdrf8Ct598alHWIiIRQUJMamv1EZAysylxFlIkK6vR9ecV5nJZ+mhIaEWJG0gxOSzstuHWiKI9oR3TIu0T0ZoxhbfZathRv6el/OhyvVbxGS1dLz5SY4SI3O5fGzkZ2VO4IWQzvHnmXmtaanpHRRSQwfOetzcWbR3Te8sfm4s0AYXduExGJBEpqiIyBlLgUzpl7Ts9FS6C1dLWQX56vm5kIk5udS35ZPi1dLUFZ/uaSzZwz9xyS45KDsvyRys3Opa69jreq3xr2Z/OK8npmcgkna7LXYDBBO8b94Vt37gKdB0QCLTc7lyOtR3i35t2gLD+vOI/pidM5Ne3UoCxfRGQ8U1JDZIzkZueyo3JHz8j5gbS9dDtOt1NJjQiTuyCXbnc3Lxx6IeDLrmurY2flzrCsE74nkXlFw2+lklecxxmzzyA1ITXQYY3K9MTpnJYe3JY3Q8krzmPp9KXMTZkbshhExque81YQjnFrLZuLN7Mmaw0Oo0tzEZHh0plTZIzkLsjFYtlasjXgy84rziM+Op7z550f8GVL8Fww7wLio+NHdHM/FN+c4eGY1EhPTmfFzBXDvjlo7Gjk9YrXw3KbIPgtbwbT6exke+n2sP1uRCLdnJQ5LJu+LChJjcKjhVS1VOn4FREZISU1RMbIWXPOIiUuJSg3sHnFeVw0/yLio+MDvmwJnvjoeC6cd2FQLpLzivOYHDeZM+ecGfBlB0Judi4vHX6pZwpDfzxf+jwu6wrbC//cbE/Lm+2lYz8Dyitlr9DubA/b70ZkPMjNzuWFQy/Q4ewI6HJ91wXqOiYiMjJKaoiMkWhHNJdkXhLwG9iKpgp2H92tm5kIlZudS+HRQiqbKwO2TGstecV5XJJ1Schm4hhKbnYuna5OXjz8ot+fySvKIykmKehTpY3U+fPO97S8CUEXlLzi8BsUVmS8yV2QS4ezg5cPvxzQ5eYV57F42mLmTZ4X0OWKiEwUSmqIjKHc7FxKGkooOlYUsGX2DA6opEZE8j2ZC+QAk0X1RZQ2lIZ1nbho/kXERsUOq+VSXnEeF2deTGxUbBAjG7n46Hgumn9RyJIa4TgorMh4cvH8i4l2RAf0GO90drL9kLqOiYiMhpIaImPIdwMbyAuivOI8ZibN5KRZJwVsmTJ2Tp51MjMSZwS2TviaMofxRXJSbBLnZZzn93YfajjEgWMHwnqbwPOd7z66m4qmijFbZzgPCisyniTHJXPu3HMDer7OL8+nrbtNx6+IyCgoqSEyhhZNXURGSkbALog0YnrkcxgHa7LXsLl4M9bagCwzrziPeZPnsXDqwoAsL1hys3N5u+ZtalpqhizrO2bC/cLfN0PCWE7t6hsU1rduEQme3Oxc3qx6k9q22oAsb3PxZqJMlLqOiYiMgu6CRMaQMYbc7Fy2lmzF5XaNennvHnmXmtaasL/Rk8HlZudS3VLNriO7Rr0sp9vJ1pKt5GbnYowJQHTB46u3/swItLl4M7OTZ7N8xvJghzUqwWh5M5S84jxS4lI4a85ZY7ZOkYkq0DOZ5RXncfbcs5kcPzkgyxMRmYiU1BAZY7kLcmnoaGBH5Y5RL0sjpo8Pvpv7QNwI76jcQWNnY0QkulamryQ1PnXI7XZbN1tKtrA2e23YJ2ocxsHa7LUBbXkzmJ5BYTPDd1BYkfHkjNlnMDluckBmMqtvr2dH5Y6IOF+LiIQzJTVExtiarDVAYG5g84rzWDp9KXNT5o56WRI6GZMzWDJtSWDqRFEeBsOa7DUBiCy4ohxRrMleQ15x3qAJgLeq36K2rTZiLvxzs3Opaa3h3SPvBn1dkTAorMh4Eu2IZnXW6iHPW/7YWrIVt3Xr+BURGSUlNUTG2IykGZyWdtqob2A7nB28cOgFXQyNE7nZuWwv3U6ns3NUy8krzuO09NOYnjg9QJEF19qstZQ3lbOvbt+AZXxPRH0JwXDXMyBwAJ7kDkWttUTGXm52LocaD3Hw2MFRLSevOI/k2GR1HRMRGSUlNURCIDc7l/yyfFq6Wka8jPyyfNqd7UpqjBO5C3Jpd7aTX54/4mW0dLWQX54fUXXCnwRAXnEeK2auID05fazCGpW5KXNZOn0pm0uCP1jo5pLNzJs8j0VTFwV9XSLi4RuUd7QPJ/KK81iVuYqYqJhAhCUiMmEpqSESArkLcul2d/PCoRdGvIy84jyiHdEaMX2cWJW5iigTNaqn+9tLt+N0OyMqqZGdmk12avaANwft3e28dPiliNomCFzLm8G43K6IGRRWZDxZOHUh8yfPH1VSo7i+mOL64og7t4mIhCMlNURC4IJ5FxAfHT+qG9i84jzOmXsOyXHJAYxMQiUlLoVz5p4zqovkvOI84qPjOX/e+QGMLPhys3PZVrqNblf3Ce+9ePhFOl2dEXfhn5vtaXnzStkrQVvHjsodNHQ0RNx3IxLpes9k5nQ7R7QMdR0TEQkcJTVEQiA+Op4L51044hvYurY6dlbu1M3MOJObncuOyh0caz82os/nFedx0fyLiI+OD3BkwZWbnUtzVzOvVbx2wnt5RXnERsVy0fyLQhDZyPW0vAni1K6+ZUfCoLAi403uglyaOpt4o+KNEX0+rziPuSlzWTJtSYAjExGZeJTUEAmRtdlrKTxaSGVz5bA/u7VkKxbb069Xxoe12WuxWLaWbB32ZyuaKth9dDdrsyKvTqzOWo3DOPptuZRXnMd5GeeRFJsUgshGLjkuedQtb4aSV5zHaWmRMyisyHiyJmsNBsPm4uGPnaOuYyIigaWkhkiI+FpZjOSCKK84j5S4FI2YPs6cNecskmOTR9QtyVePIrEpc2pCKmfMPuOEgTWPtB7h7Zq3I7ZFUm52Ljsrd1LXVhfwZbd0tZBfFlmDwoqMJ9MSp7EyfeWIEpcFVQXUd9Tr+BURCRAlNURC5JS0U5iROGPYF0TWWvKK87gk8xKiHdFBik5CISYqhkuyLhnRRXJecR4zEmdw8qyTgxBZ8OVm5/Ja+Ws0djT2vLaleEvPe5Eod0HuiFveDGV76Xa63d0RmcQSGS9ys3PJL8+nubN5WJ9T1zERkcBSUkMkRBzGwZrsNWwu3oy11u/PFdUXUdpQGrE3ejK43OxcShpKKDpW5PdnrLVsLt7M2uy1OExkntZzs3NxWRfbSrf1vJZXnEdqfCor01eGLrBROGvOWaTEpQSlC4pvUNgL5l0Q8GWLiH9yF+TidDvZfmj7sD6XV5zHqWmnMjNpZpAiExGZWCLz6ldknMjNzqW6pZpdR3b5/RmNmD6++ZJVw7kRfvfIu9S01kR0ouvcjHNJiknq2W5fi6Q12WuIckSFOLqRiXZEc0mmp+XNcBKX/sgrzuPCeRdG3KCwIuPJ+RnnkxCdMKwug61drbx8+OWIPl+LiIQbJTVEQmgkN7B5xXnMmzyPRVMXBSssCaHF0xaTkZIxvDoxDhJdsVGxXJx5cc9276vbR3lTecRf+Odm51LaUEpRvf8tb4biGxQ20r8bkUgXFx3HRfMvGtb5+oVDL3i6jun4FREJGCU1REIoY3IGS6Yt8fuCyOl2asT0cc4YQ252LltLtuJyu/z6TF5xHkunL2VuytwgRxdcudm57K/bz+HGw/9J1ET4hb8v0TSSwV8HEsmDwoqMN7nZueyp3UN5U7lf5fOK84iLilPXMRGRAFJSQyTEcrNz2V66nU5n55Bld1TuoLGzMeJv9GRwuQtyaehoYEfljiHLdjg7eOHQC+OiTvimKM4ryiOvOI/s1GyyUrNCHNXoLJq6iHmT5wV0XI1IHxRWZDzxnbf8ncksrziPC+ZdQEJMQjDDEhGZUJTUEAmx3AW5tDvbyS/PH7Ks76JJI6aPb2uyPPvXn4vk/LJ82p3t4yKpkTMjh/RJ6fz94N/ZVrptXGzTSFreDGY8DAorMp6cNOskZibN9CtxWdVcxa4ju8bFuU1EJJzoikgkxC6efzFRJsqv5ul5xXmclnYa0xOnj0FkEiozkmZwatqpfl0k5xXnEWWiuDjz4jGILLiMMazNXsuf9/6Z5q7mcXPhvzZ7LY2djX61vBnKriO7qGmt6Xk6LCKh5TAO1mav9Wsmsy0l3mmq1XVMRCSglNQQCbHJ8ZNZPmM5D7/9MPllA7fW2FK8hZcOv8TyGcvHMDoJldzsXF46/BJ3bLtj0HrxzN5nmJsyl8IjhWMYXfDkZufitm4Mhkmxk0IdTkD4Wt586/lvDbov88vyuefFewYts7FgIwBT46cGNkgRGbHc7FyOtB7hS//80qDH7+PvPk5iTCLt3e1jGJ2IyPhnAj3NnEgATKhKmV+Wz0UPXYTT7STKRLE6a/UJLTFq22rZUrIFt3UTFxXH8x9/nnMzzg1RxDIWfvbqz/jSv74EMGC9ONp2lM3FmzEY4qPj2XL9loivF8/te44PPvFBABKiE8bFNuWX5XP+A+djsYMe41tLtuKyrkHLbC7ejMWOm+9GZDx4du+zfOjJDwEDn69r22p7Wt/p+BWRCSioMxxEB3PhIjK0baXbcFs3AC7roqCqgKkJxz+FPdZ+rKeM0+1kW+k2XQyNc02dTT3/H6xeAFgsXa6ucVEvdh3ZhcGMq23aVrqt5/+D7UuXdQ1ZxnpzvuPluxEZDwqP/qel3FDna9DxKyISaEpqiITYqsxVxEXF0eXqIjYqlueuee6EC538snzWPLKmp8yqzFWhCVbGzNrstdzz0j0Trl6sylxFfHT8uN4mf/blRNnfIuPBqsxVJEQn6PgVEQkRdT+RcDThKmV+WT7bSrexKnPVgE9u/Ckj48tErRcTdZsm6v4WGQ90/IqIDCqo3U+U1JBwpEopIiIiIiIyPgQ1qaHZT0REREREREQkIimpISIiIiIiIiIRSUkNEREREREREYlISmqIiIiIiIiISERSUkMkDFRVwcUXQ3V1qCORsKKKMX74sS+1u0Uil45fEZHQUVJDJAx897vw0kuenyI9VDHGDz/2pXa3SOTS8SsiEjqa0lXC0YSqlFVVkJ0NHR2QkADFxZCWFuqoJORUMcYPP/aldrdI5NLxKyIyJE3pKjKeffe74HZ7/u9y6SmPeKlijB9+7EvtbpHIpeNXRCS01FJDwtGEqZS9n+746CmPqGKMI37sS+1ukcil41dExC9qqSEyXvV+uuOjpzyiijGO+LEvtbtFIpeOXxGR0FNLDQlHE6JS9vd0x0dPeSYwVYzxw499WWXTtLtFIpRO1yIiflNLDZHxqL+nOz56yjOBqWKMH37sS+1ukcil41dEJDyopYaEo3FfKQd7uuOjpzwTkCrG+OHHvqyKzyKbIjo6Bn54od0tEp50uhYRGRa11BAZbwZ7uuOjpzwTkCrG+OHHvvxu1624u5yDltHuFglPOl2LiIQPtdSQsGOMucVa+9tQxxFMxvAmcKofRd+yltOCHI4EwYjqsTF+1wusVb0IZ37sy9Mo4C3/Du+QnQcmwvlYxrdg1WH9HZexpHOxRLpg12ElNSTsGGN2WGvPCHUcIqOheizjgeqxRDrVYRkPVI8l0gW7Dqv7iYiIiIiIiIhEJCU1RERERERERCQiKakh4Uh9BmU8UD2W8UD1WCKd6rCMB6rHEumCWoc1poaIiIiIiIiIRCS11BARERERERGRiKSkhoQVY8x7jTH7jDEHjTHfCHU8IkMxxmQYY543xuwxxhQaY77ofX2qMSbPGHPA+zM11LGKDMUYE2WMedMY81fv76rHElGMMVOMMX8yxuz1npfPVT2WSGKM+bL3emKXMeb3xph41WEJd8aYB4wxR4wxu3q9NmC9NcZ803u/t88Y8/9Gu34lNSRsGGOigA3AfwHLgWuMMctDG5XIkJzAV621y4BzgPXeevsNYIu1dhGwxfu7SLj7IrCn1++qxxJpfgb801q7FDgFT31WPZaIYIyZA3wBOMNauwKIAq5GdVjC30PAe/u81m+99V4nXw3keD9zn/c+cMSU1JBwchZw0FpbbK3tAp4APhTimEQGZa2tstYWeP/fjOcCeg6euvuwt9jDwIdDEqCIn4wxc4FLgft7vax6LBHDGJMCXAT8DsBa22WtbUD1WCJLNJBgjIkGEoFKVIclzFlrXwCO9Xl5oHr7IeAJa22ntbYEOIjnPnDElNSQcDIHKOv1e7n3NZGIYIzJBE4DXgNmWWurwJP4AGaGMDQRf9wL3Aq4e72meiyRJBs4Cjzo7UZ1vzEmCdVjiRDW2grgR8BhoApotNb+G9VhiUwD1duA3/MpqSHhxPTzmqbnkYhgjJkEPAV8yVrbFOp4RIbDGPN+4Ii1dmeoYxEZhWhgJfAra+1pQCtqpi8RxDvmwIeALGA2kGSMuTa0UYkEXMDv+ZTUkHBSDmT0+n0uniZ3ImHNGBODJ6GxyVr7tPflGmNMuvf9dOBIqOIT8cP5wAeNMaV4uv6tNsY8huqxRJZyoNxa+5r39z/hSXKoHkukWAuUWGuPWmu7gaeB81Adlsg0UL0N+D2fkhoSTt4AFhljsowxsXgGkHk2xDGJDMoYY/D0395jrf1Jr7eeBT7u/f/Hgb+MdWwi/rLWftNaO9dam4nn3LvVWnstqscSQay11UCZMWaJ96U1wG5UjyVyHAbOMcYkeq8v1uAZq0t1WCLRQPX2WeBqY0ycMSYLWAS8PpoVGWvVul/ChzHmfXj6dUcBD1hr7w5tRCKDM8ZcALwIvMt/xiK4Dc+4Gn8A5uG5SLnSWtt3ACWRsGOMWQV8zVr7fmPMNFSPJYIYY07FM9htLFAMfALPQzzVY4kIxpg7gY/imV3tTeAmYBKqwxLGjDG/B1YB04Ea4DvAnxmg3hpj/gf4JJ56/iVr7T9GtX4lNUREREREREQkEqn7iYiIiIiIiIhEJCU1RERERERERCQiKakhIiIiIiIiIhFJSQ0RERERERERiUhKaoiIiIiIiIhIRFJSQ0REREREREQikpIaIiIiIiIiIhKRlNQQERERERERkYikpIaIiIiIiIiIRCQlNUREREREREQkIimpISIiIiIiIiIRSUkNEREREREREYlISmqIiIiIiIiISERSUkNEREREREREIpKSGiIiIiIiIiISkZTUEBEREREREZGIpKSGiIiIiIiIiEQkJTVEREREREREJCIpqSEiIiIiIiIiEUlJDRERERERERGJSEpqiIiIiIiIiEhEUlJDRERERERERCKSkhoiIiIiIiIiEpGU1BARERERERGRiKSkhoiIiIiIiIhEJCU1RERERERERCQiKakhIiIiIiIiIhFJSQ0RERERERERiUhKaoiIiIiIiIhIRFJSQ0REREREREQikpIaIiIiIiIiIhKRlNQQERERERER+f/t3Xl8XFX9//HXmWzTJE2XdG9pQunepm1CtQUVwQ0LqAiiYkVxQ6QKbl/9uePydd9YCiLKIlZERPkitFRF1kKBki7pRkv3JE3bTNbJnpnz+2Nu6jSdJJNktpu8n49HH9PcOffec2fOPXPv555FXElBDRERERERERFxJQU1RERERERERMSVFNQQEREREREREVdSUENEREREREREXElBDRERERERERFxJQU1RERERERERMSVFNQQEREREREREVdSUENEREREREREXElBDRERERERERFxJQU1RERERERERMSVFNQQEREREREREVdSUENEREREREREXElBDRERERERERFxJQU1RERERERERMSVFNQQEREREREREVdSUENEREREREREXCk92RkQ6S7DjLJeJvWaZgqVfW5nJP4+01Qyuc80jYwcdF6iUcmUPtNEs69otjOHPVFsp+/PZiSNg85PIj+/aPKbSIn6/PoqwwBHo/j8Jsfou4pGNPnJjcH3Gc250EjuoPcTrWi+q2jSRCOa8tdXmmjyEotyHkv+KPIcTVmfwtE+07zC2YPeT6ykWv0XjViU9Wi2Eat6NBqxOh8Sledo8ruHOX2miaasx6o+idV3Ho2+9pVK9XW0YlUfR5Ofvn7ro/mdT2SZSGQZjcV2YnXc0dxT9XWtVEUr9bbD9LmhAVJQQ1KOl0mczR29prmRG/vczvk83WeaG7mmzzRPcf6g8xKNaLYTqzRPcUEU2+n7szmfpwadn0R+ftHkN5ES9fn1VYYBvhvFdq6J0XcVjWjyc3YMvs9ozoWn+rgxjaVovqto0kQjmvLXV5po8hKLch5LT0eR52jK+o18t880hk2D3k+spFr9F41YlPVothGrejQasTofEpXnaPJ7QRRpoinrsapPYvWdR6OvfaVSfR2tWH020eSnr9/6aH7nE1kmEllGY7GdWB13NPdUfV0rfZpX+tzGYKj7iYiIiIiIiIi4koIaIiIiIiIiIuJKCmqIiIiIiIiIiCspqCEiIiIiIiIirqSghoiIiIiIiIi4koIaIiIiIiIiIuJKCmqIiIiIiIiIiCspqCEiIiIiIiIirqSghoiIiIiIiIi4koIaIiIiIiIiIuJKCmqIiIiIiIiIiCspqCEiIiIiIiIirqSghoiIiIiIiIi4koIaIiIiIiIiIuJKCmqIiIiIiIiIiCspqCEiIiIiIiIirqSghoiIiIiIiIi4koIaIiIiIiIiIuJKCmqIiIiIiIiIiCspqCEiIiIiIiIirqSghoiIiIiIiIi4koIaIiIiIiIiIuJKCmqIiIiIiIiIiCulx2Ijq1aWWoDVa0pMLLaXKKtWlmYDbwVWAK8HCoA84CjwDPCr1WtKNicvhz1btbL0auBu4N7Va0quTm5uRERERERERBJvuLfU+BDwCPAZYBywAfgH0AFcBby8amXpp5KXPRERERERERHpyXAPanQAdwElq9eUFK5eU3Lp6jUl7wNmA18C0oDVq1aWzk5mJkVERERERETkdDHpfuJWq9eU3AvcG2G5BX65amXpRYS6p3wA+H6CsyciIiIiIiIivYh5UGPVytJrCHXnmAO0AE8D3169pmR7D+kLgK8C7wSmAs3AFuDO1WtK/hSWzgBrnXR3rl5Tck237XiAfwFvAVavXlPy2RgczmZCQY1p/V1x1crSfOAG4F3AWYRafVQS6uLy29VrSp7vlj6qz6Ef+z+XUGuTNwBjgRrgOeDnq9eUbIyQ/ingzcAFQNDJyzJn3ctWryl5uL95EBEREREREYmnmHY/WbWy9FfA7UA98H9ANfBe4MVVK0vfGCH9MkI37p9xFv0d2EToRnzNqpWlf3CCGV2tJ64CKoBPrVpZemW3zX2bUEBjM6Gb+ViY5bwe7c9Kq1aWFgNlwLeA6cBTwGOEAgtXAt0DMlF/DlHu/zPAs8BlwGHgr87r5cCGPsYJuQJ40sn3v4AnCHXTEREREREREUkpsW6pcQ1wweo1Jc/AydYVPwT+H/CnVStLZ69eU9LqvOcFHgRGA78Gvrx6TUnAeW8hoZvpqwi1bLgDYPWakupVK0s/SChIcMeqlaWbVq8p2btqZekFhAIIjcD7V68paRvsgaxaWVoEXAxYQkGGaNcbSWjw0cnAb4Avrl5T0hL2/nhCrVi6/u7359DH/hcDNzt/vn/1mpIHw977ILCG0DghL/TQeuY64NOr15T8NspDFhEREREREUmKWA8UentXQANOtq74JrAfOINQS4EuVzjLDgFf6bqRd9bbDtzo/Pnl8B2sXlPyHKFWGSOBv6xaWTod+JNzLJ9avabktcEexKqVpbnONtOBu1evKdnaj9U/Qai7ykbguvCAhpP/E84xdBnQ59CL6518/zk8oOFs78+EWm1kEOoaE8m/FNAQERERERERN4h1UOOP3Rc4N+n3O3+eH/bWm53XNavXlETq3nA3oVYSM1etLJ3a7b0fAeuBJcA2YBJwx+o1JQ8MOOeOVStLMwi1nFhIqEvI9f3cxDud1987QZ2+DOZz6G179/Tw/l3O6/k9vP+3KPYhIiIiIiIiknSxDmoc6GH5Qec1fMDNrhv0iOs43VQqu6Xteq9rfA0/MArYCXy+37ntZtXK0nTgz4QCE7uAC1evKWnq52YKnNfdUaYf8OcwkO0B+/rY1qEo9iEiIiIiIiKSdLEOavQlvOWCibCsu94Gx7wUyHX+P43obvh7tGplaRqh8SYuA14D3rZ6TcnxwWwzSoP9HHrS0/b62lZLH++LiIiIiIiIpIRYBzUK+1heGbas3HmdEWkFZwDNyc6fFd3eWwjcBLQTCkTkAQ+sWlma2e8cczKgcR/wfkItHN6yek1JZe9r9airpcOcXlP914A/hx50pYm4PeDMfmxLREREREREJGXFOqixsvsCJ2DwAefPp8Leetp5vdLp9tHdRwm1Knht9ZqSkzfgq1aW5gB/AUYAXwU+QmgK0rOBn/U3w6tWlnqAPxCaavUQodlbjvR3O2HWO68fj3Ia1gF9DlFs7yM9vP8x5/WpKLYlIiIiIiIikrJiHdS4btXK0jd2/eHc1H8XmEmoZcBDYWkfBI4QajnwIye40LXefGc9gJ9328dqYB7wyOo1Jb9evaYkSCiYchy4ftXK0kujzayzz7uBDwGHCQU0BjumxO8ItUg5F7jFaWkRvs/x4Z8RA/8cenIz0EkoSPLebvu+glBrlA7+O+2riIiIiIiIiCtFahkwGHcCT69aWfoMcBQoIdQNowVYGT696eo1Ja2rVpa+H1hHaLrS965aWfoyMJbQzByZhLqEnJxedNXK0o8QarlwhP+2OGD1mpKjq1aWXgU8Dty1amXp5iiDE5/lvy0a9gPfWbWyNFK651avKfldFNtj9ZqSxlUrS98DPAasAj64amXpBqCV0CCixYRmg3luoJ9DH/vfumpl6Q3ArcDfVq0sfZHQ4KAzgdcDQeCzq9eUlEWzPREREREREZFUFeuWGl8EPkfohvxSYALwMLBs9ZqSp7snXr2mZCOhaVl/A6QRGqRzGbAR+DDw0a5pUVetLJ0L3IbTCmH1mpKabtv6J/ATYAzwZ2dq1r6MDfv/+YQCJpH+vfG0NXuxek3JJqAI+DFQBbwduNjZ35+c4w1PH/XnEOX+bwPeBPydUAuQ9xMa1+RvwBtXrymJKkAiIiIiIiIikspi0lJj9ZqS8LEjbnP+RbvuIeAzUaTbzX9nO+kpzdeAr/Vj3zcCN0abvj+cmVOizk+0n0NY+nuAe3p5fwOwoR/bOz/atCIiIiIiIiKpINFTuoqIiIiIiIiIxISCGiIiIiIiIiLiSgpqiIiIiIiIiIgrGWujHn9SJCGMMSeAwU6tKyIiIiIiIslXYK0dH6+NK6ghIiIiIiIiIq6k7iciIiIiIiIi4koKaoiIiIiIiIiIKymoISIiIiIiIiKupKCGiIiIiIiIiLiSghoiIiIiIiIi4koKaoiIiIiIiIiIKymoISIiIiIiIiKupKCGiIiIiIiIiLiSghoiIiIiIiIi4koKaoiIiIiIiIiIKymoISIiIiIiIiKupKCGiIiIiIiIiLiSghoiIiIiIiIi4koKaoiIiIiIiIiIKymoISIiIiIiIiKupKCGiIiIiIiIiLiSghoiIiIiIiIi4koKaoiIiIiIiIiIKymoISIiIiIiIiKupKCGiIiIiIiIiLiSghoiIiIiIiIi4koKaoiIiIiIiIiIKymoISIiIiIiIiKupKCGiIiIiIiIiLiSghoiIiIiIiIi4koKaoiIiIiIiIiIKymoISIiIiIiIiKupKCGiIiIiIiIiLiSghoiIiIiIiIi4koKaoiIiIiIiIiIKymoISIiIiIiIiKupKCGiIiIiIiIiLiSghoiIiIiIiIi4koKaoiIiIiIiIiIKymoISIiIiIiIiKupKCGiIiIiIiIiLiSghoiIiIiIiIi4koKaoiIiIiIiIiIKymoISIiIiIiIiKupKCGiIiIiIiIiLiSghoiIiIiIiIi4koKaoiIiIiIiIiIKymoISIiIiIiIiKupKCGiIiIiIiIiLiSghoiIiIiIiIi4koKaoiIiIiIiIiIKymoISIiIiIiIiKupKCGiIiIiIiIiLiSghoiIiIiIiIi4koKaoiIiIiIiIiIKymoISIiIiIiIiKupKCGiIiIiIiIiLiSghoiIiIiIiIi4koKaoiIiIiIiIiIKymoISIiIiIiIiKupKCGiIiIiIiIiLiSghoiIiIiIiIi4koKaoiIiIiIiIiIKymoISIiIiIiIiKupKCGiIiIiIiIiLiSghoiIiIiIiIi4koKakjMGWO+Zox50Biz3xhjjTEHk50nERERERERGXqMtTbZeZAhxhhjgRqgFDgbaLDWFka7/rhx42xhYdTJRUREREREJEW98sor1dba8fHafnq8NizD2lnW2v0AxpjtQG5/Vi4sLGTTpk1xyZiIiIiIiIgkjjHmUDy3r6CGxFxXQCNZAtayzudjs99PcW4uK/LzSTMmpdOIiIiIiIhI/ymoIUNKwFou3LqVFxsaaAoGyfF4WJaXx/rFi08GElItjYiIiIiIiAyMxtSQuOrqftKfMTWWLl1qB9r95NHqaq7cuRN/MHjK8nQ4JdDQGWHdRKXJ9Xi4f/58Lhk3LsqjEhERERERcSdjzCvW2qXx2r5aakhKMMZcA1wDMH369AFvp9TvPy2gAXBOXh7njhoFwIb6ep5raEhamqZgkC1+v4IaIiIiIiIig6SghqQEa+1vgd9CqKXGQLdzuLX1tGW5Hg9fmT79ZBDh0epqtnRrzZHINCM8Hpbk9mvsVBEREREREYnAk+wMiMTKQydO8PuqKiZkZJDr8WAIBRmW5eWxIj//ZLoV+fksy8tLSpouy/Ly4vY5iIiIiIiIDBcaU0PiKlFjamysr+eCrVspzs3ln4sW8VRdHVv8fpb0MSNJotOkG8O3Dxzg3FGjWL94MVkexRVFRERERGToiveYGgpqSFwlIqixv6WF5aWljExLY2NJCeMzMweQ08RZc+wYH961iw9PnMgf5s7FaBYUEREREREZojRQqLiOMeYqoMD5czyQaYz5pvP3IWvtfbHaV01HBxdt20antaxdtCjlAxoAKydO5EBLC986eJAZXi/fPfPMZGdJRERERETElRTUkHj4BPDmbsu+77w+DcQkqNEWDHLZ9u0caG3lX4sXMyc7OxabTYhvFBSwr7WV7x06xIwRI/jopEnJzpKIiIiIiIjrKKghMWetPT9e2+4an6LU7+fJujqerq9nzbx5nDd6dLx2GRfGGO6YPZsjra18YvduytvaCFpLcQ/jcsjQ0lWON/v9PX7n0aQRd9B3KUNBrOqtvtIksn5M1DGJiLiRm+o/BTXENQLWcuHWrbzY0HByitQZXi8fmDAhyTkbmEyPhwfmz6dg40a+eeAABshxZlBZv3ixLoiGqPBy3BQMRvzOo0kj7qDvUoaCWNVbfaVJZP2YqGMSEXEjt9V/GihUUk5PA4U+Wl3NlTt3ngxoQGga1fvnz+eSceMSmcWYebS6mg/s3EnzEDom6V2kcpwGFOfmMtEZE+ZYezub/X4CYeupXLjTUKy3ZPgZaL3V3zQD3cZAzql4HpPOcRFxu0erq/ngzp00xeh6Nd4DhWo+SXGNzX7/KScWQFMwyBa/P0k5GrzNfj8tQ+yYpHelfv8pF9EAAaCyvZ0q519le/spPxCgcuFWQ7HekuFnoPVWf9MMdBsDOacinZuxOiad4yLidqUDrCOTVf+p+4m4Rn5GBt3bFeV4PCzJzU1KfmKhODeXHI/nlItFtx+T9O5AS8tpy3I9Hu6YPftkVDvSE0SVC3cam54+5OotGX72DbDe6m+agW4j3Zh+n1MjPJ7Tzs1YHZPOcRFxu82NjactS+X6Ty01xBWq2tr4yaFDZBpDtseDIXRiLcvLY0V+frKzN2Ar8vNZlpdHjid0KmYa4/pjkp49ePw49xw7xoSMDHJ7Kcdd5aKrgh4KZX04OtrWxo8PHybTGDKdvqUj9F2Ky9x/7Bh/OHaMiVHWW4NJM5BtZBhDh7URAy892d/Swk8OH8ZrDDkxOqaup4Sqr0XE7W6vqOBhn49pmZmDrvcTRWNqSMrpPqZGUyDA+Vu2sLOpiSeXLOF4eztb/H6WDJERxrtGDV65axdLR47knxpcbEh6vr6et2zZwtKRI1m/aBFP1tX1Wo4D1nLB5s3saG7m3rlzh0RZH078nZ28ecsWXm1u5qklS9hQX8/n9+3ja9On8/0zz9R3Ka7wXF0db926lWV5eTy+aBH/qa3ts95a5/MNKk1/t1GUm8vvKyt5rKaGhxcu5F19jGNR09HBuaWlnOjo4LniYva1tMTkmK7atYv7jx/nbwsW8O5x43SOi4grrfX5eFdZGRfl5/PQggX8s6Zm0HUkxH9MDQU1JOWEBzUC1nL59u084vPx8MKFvHsID7r1xtJS0o3hqeLiZGdFYmxfSwvLS0sZnZ7OC8XFjHMGWOrLV/ft49fl5bScdx4eXSC7RsBa3rt9O4/5fPyjqIiL8vOp6eggf8MGfnHWWXzxjDOSnUWRPu1tbmZ5aSnjMjJ4oaSEsRkZyc5Sj8IffjxTXMzZI0dGTNcWDPKOrVvZ2NDAvxcv5k0xnA7+nqNH+dirr7Ln9a9nVnZ2zLYrIpIomxsbedPmzczJzubpJUvITY/dSBUaKFSGtS/v28f/+XzcNHPmkA5oABR6vRxqa0t2NiTGfB0dXLRtG9Za1hYVRR3QgFCZaLeWqvb2OOZQYslay+dfe41/+HzcMmsWFzlNMMekpzMyLY1Dra1JzqFI36rb27morAyPMaxdtCilAxoAOWlp/GPhQsZnZHBJWRmHI5xn1lo++eqrPFNfz91z58Y0oAGh+hrQOS4irnSktZVLysoYm5HBo0VFMQ1oJIKCGpKybikv59fl5dwwdSqfmzYt2dmJu0KvlyOtrXR2G2lY3KstGOS927dzqLWV/ysq6vfTu66L5IO6SHaNm8rLubWigi9Nm8Z1U6eeXG6ModDr1XcpKa81EOA927dzpLWVRxYu5KwRI5KdpahMysrisUWLaAkEuGjbNuo7O095/8aDB/njsWP84Mwz+dDEiTHfv+prEXGrhs5OLi4rwx8I8FhREZOzspKdpX5zVwhGhoWj7e18a/9+/vfwYd6Tn88vZs5MdpYSosDrPTlV0nTn4kjcp6tvYanfzxO1tTxbX8+f58/nDaNG9XtbBWFP/s4dwPoSf13f92a/n9ZgkB8ePszl48bx07POOi1tgderp7iSksLrrX/X1PB8QwN/mT+fc1xW7yzIyeGhhQt557ZtXL59O5+dOpWypiZ8HR3cVFHBxydN4uvTp8dl31OzsvCAWly6XHidXtzHGAK9pUk1bszzcJTI8te1nU2NjTxSXc3OpibWLVpEkUtnblJQQ1JOZVsbPzh8mJFpafxh3rxhU+mGP+VRUMOdAtZy4datvNjQcHJ6q7O8Xt43fvyAtlegJ38pLfz7bgoGsUBeWhr3zJ0bcQyUQq+XZ+vqEp5Pkd5Eqrdmer1cNsB6K9neOmYMv5k1i0/u2cPTdXUEAEuoC9jqWbMwcbqmyPB4mJaVpfraxbrX6TnOTA7rwwZwjyZNqnFjnoejRJa/SPX+vOxs3jJmTFyOLRHU/URSVsBanhlGNwC6gXW/dT7fKT8QAMfa21nn8w1oezlpaYzPyFCZSFHh33fXkNsBa3mqh3qr0OulPhCgrqMjYXkU6UukeqtqEPVWKpiYmUmGMXTCyXOzPRjk37W1cd2vupi5W/c63R8M8mx9Pde++iq/PnKEXx85wrWvvsqz9fWnpHmxoSGlz5dIx5XqeR6OEln+ItX7R1pbXV0mFNSQlNUSDLLF7092NhJmutN/Tc3T3Wuz309TtzFRmgZZjtVlIXVF+r6be/m+C7rOcTVPlxQSj3or2Tb7/XR2m92vt3MzVlRfu1ukc6HdWn5XVcUX9u3jC/v28buqKtq7la1UP182+/2n3LxC6ud5OEpk+RuK9b6CGpKycjwelri0X9dAeNPSmJyZqac8LjY2PZ3uk2QPthzryV/qWpiTQ/fGnr193xpIUFLR6DjUW8lWnJtLjufUS9xEHFOh10tFWxvtGvDblYpzc8mOUG7unzeP2je8gdo3vIH75807rWx5U/x8KY8QSHf7OT4Uzc3OjnhN0Vf5GzGA7zLL4xly9b6CGpKScp0+Yiuc6RCHiwLdwLrW0bY2fnz4MJnGkO3xYIhNOe6a6tfa7j8/kkzWWv5+4gRBIMuYqL5vdTGTVFPZ1sZP4lBvJduK/HyW5eWRm+BjKvR6CRL5JlJS34r8fCY50653lZvleXlcMWECozMyGJ2RwRUTJrA8rGwZoCMYZGaKzhL0SHU1dx49yrj09FNuhidkZrr6HB9qAtZyb1UVQUJBsmjLH0DQWpaOHBn1vl5rbuYnhw4xwuMhZwjV+xooVFLOlKws7pg/f1iOzFzo9fJyQ0OysyH95O/s5JKyMmo7O9lQXExVeztb/H6WxGCE8YKsLFqDQY53dDDRudiS5Pv+oUPcd/w43ykoYOnIkVF93+MyMsj2eNQ8XVJCozOFX30gwAslJVS2tcWs3kq2NGNYv3gx63y+hB5T+IxVM1L0Jld6Zq2lLRhkUU4OV4wfH7HcdC9bEzMz+dr+/bxr+3ZeKC5mXAr9Tm9qaODKnTtZOnIk/168mGfq6tjszMz2dH09fzl+nCvjML2x9I+1ls+/9hqP1dRw68yZFHi9PdZb3cvfiLQ0vrF/P+/dsYP/LF7MiLS0XvdV3d7OirIyPB4Pm5csYW9Ly5Cp9xXUkJQzOTOTS8aNS3Y2kqLQ6+WhEycIWOvqimU4CVjLlbt2scXv55GiIpbm5QHErAyHd1lQUCM13FdVxXcOHuSjEyfyncJCjDFRfd/GGHUnkpTQGQzywZ07KfP7+UdRESUjR1IycuSQ+u1Nc87LRB6Tupi528PV1ZS3t7N69mze3Uu56V62FubkcMGWLbxn+3aeWLwYbx83lolwqLWVS8rKmJCZyT+KishLTz+Z569Mn847tm7l6t27mZaVxZtGj052doe1m8rLubWigi9Nm8aqadOA3q8hu5e/M71e3rdjBx/ZvZsH5s+POPsaQGsgwKXbt3OktZX/LFnCnJwc5uTkDJl6X91PRFJIQVYWHdZyVE1XXaEruv6oz8cts2ZxcRya7ekiObU8VVvLJ159lbeMHs1v58zp9/SQ6mImyWat5frXXmNtTQ2rZ892dXPjVHNGVhYG1ddudVNFBWd6vf3+LT9n1CjumzeP5xsauHr3boJJ7i5a19HBxdu20RoM8lhR0WkPRLI8Hv6+cCFner1cun07e5qbk5RT+fuJE3xx3z4uHzeOn5511oC2cdn48fz8rLP464kT/L/9+yOmCVrLx159lQ0NDdw3bx7njho1mGynJAU1RFJI1w2sZkdwh/Do+nVTp8ZlH+HNmSW5djU18d4dO5g1YgQPLVhApqf/P6GFmh1BkuyX5eXcXlnJV844g09PmZLs7AwpmR4PUzIzdY67UGljI8/V1/O5qVMH1FL2igkT+OmMGTxw4gTfOHAgDjmMTnswyPt27GBPSwt/W7iQ+Tk5EdONzchg7aJFpBnDRdu2caK9PcE5lZcaGli5axevHzmS++bN67GFRTS+MG0a102Zws+OHOE3FRWnvf/NAwf48/Hj/GTGDK6YMGEw2U5ZCmqIpBA9lXePruj6ZYOIrkdjZHo6Y9PTVSaS7Fh7OxeVlZFlDI8VFTE6I2NA2yn0eqnp7KShszPGORTp20MnTvA/+/Zxxfjx/GjGjGRnZ0hSFzN3uqm8nByPh49PnjzgbXz5jDP49OTJ/PjwYe6srIxh7qJjreXaPXt4oq6OO+fM4S1jxvSafsaIETyycCEV7e28Z/t2WgKBBOVUDrS08K6yMiZlZvJIUVGfY2H0xRjDTTNncvHYsazau5d1Pt/J935XWcmPDh/mmsmT+Z8zzhhs1lOW68bUMMYcBAq6LbZAI7AbeABYba2NyaNuY8zVwN3Avdbaq8OWFwIHgEPW2sIY7WsM8EPgYmAyoe/n/6y1l8Zi+5L6puup/CkC1rLO52Oz309xnAcximZfXWn+4fNxT1UVr8vNHXR0PRrqspAcXd/3S42N/Pn4cara2nimuJjCQQwAGN7ypsjFU6cNZ4msl2KhK7+POPXWspEjuXfu3LjXW8NVgdfL8xrw21WOtbfz5+PH+dTkyYxKH/itkTGGW2fN4lBbG9fu2cPR9nYMJOz65Vfl5fynro5vTZ/ORydNimrd5aNG8cd583jfjh28fetW3jF2LCV9XANFc53klvoxVvrz2Tzf0MB9x47RHgzy9JIlTIjReGnpHg9/nj+f87Zs4YodO/jRjBlsamzkj8eOceGYMayeNavfXWbdxHVBjTDrgSrn/+nAGcC5wOuBK4wxF1hr3XYXcCdwOaFgyV+BNqA0qTmShMpOS2NCRoZuYAlV/hdu3cqLDQ00BYPkONNNrV+8OOY/kN33NcLjoSgnhztmzz65r4C1fHrPHrY1NdESDGKArLQ0sgbQBaG/Cr1eXlWf14QKLxP+YBCARTk5lPRj2rRIChXUcLWucrGxoYHmONdLsdCV3xec/Bogw+MZUNcpiU6h18sDx4/TGQySrs/ZFe6orKTdWj7nDNI4GOkeD/fPm8cZL7zAdw4exEBCrl+eq6+nzVrSgQ0NDf0acP7SceOY6fWyoaGBDQ0NZPdyDVTmXAP1dp3UlSbV68dY6c81ZFlTE83ONUVJbi6zsrNjmpfc9HT+b+FCZr34Ije89hqWULeM1mBwSAc0wN1BjR9ba58KX2CMmQ1sAJYDnwZuiuP+K4B5QEcsNmaMyQDeA7QCS6y1CvMPU2q6GrLO5zvlhtIfDPJiQwPrfL6Yj9S8zudjo/NjBNAcDPJiYyNLXnmlx3UssLmxMS756a7Q62V9TQ3W2iH/o5Qqupc/gP0tLYP+vtXFzN3W+XwnAwQQ33opFrrqtq78JrLeGq4KvV4CQEV7+8mWWZK62oNBbq+sZMXYscyJ0Q3mc/X1dP1yWOJ//fJ8QwNtzuCknYTGaujPvtb5fFSFjakRzTVQNGlSvX6Mle7XC9F8NgB7mpvj8tls9ftJMwbrlIkg8MowqPeHVAjZWrsHuMP58/w476vDWrvbWrsvRpvs6m5yTAGN4a1AAwkCsNnvPxlk6NIUDLLF74/5vkoj7MsAHxw/ngfnz+fB+fP5wPjxdA8nxCs/3RV4vTQHg/g6YhJDlShs9vtPCWhAbL7vCRkZeD0eBTVcarPffzJA0CVR9cBAvNzYmLB6VEI0uLO7/OX4cara27khBq00umz2+2lJ0Hn3r9raQe8r0vVWNNdAqXSdlEwD/fzi9dkksvylEje31OhJV5eUiKO4GWNygFXAFcAcJ91+4EHg59baqL7xvsbU6M9+jDHhcz8VdPv7TGvtwWjyJENDodfLI9XVBK0d1n2eR6al0X1SNAvkxmH+951NTacty/F4WDlx4smottfj4TGf75Qb3RyPhyUJ6EIQ/nR/XIz6XkrvOrpdEEBsvm9jDAVZWZrhyKWKc3NJA8KH00tUPdBf1lqeqa8/bXmq5neoCK+vz0tyXqR31lpuqqhgzogRvL2PQTX7ozg3lxyP55TrBQ9Q1MNMJAN1or2dB44fP215f8/xSPmN9hqorzTZw6C+Kc7NxevxnBJISOY1ZE/f51D/HoZUSw3H653XXd3fMMZMA14CfkJosNEXgH8CY4DvABucwToHZQD7uRd4yPl/k/N317+hHVaT0xR4vbRZy7FhPL3WodZWfnz4MFnGkO3xYAj9MGYZw48PH+ZwDJ+A3X30KA+cOMHkzExynX3lOv1AV4TNVb8iP59leXm9pokXdVlIrNLGRn5x5Agj09LIicP3rS5m7vWmUaMIAhlOwDkNElYP9NeNBw/yVF0dM7zepNRbw9X0rCxA9bUbbGxoYFNjI9dPmxbTh0jdrxeyjCEAPO50I42FlkCAd2/fTl1nJ0tHjhzUOR7N9c1A0kBohpWhXt+syM8/+cAtFa4hk3m9mkxDoqWGMSYdmAZ8BPgwUAfc1i2NAf4CzAduBb5qrW123hsB/NZZ91fA1YPIS7/3Y6292mn5cTlQHT7Ligw/4QMJTnYujoaT+s5OLt62jdZgkJfPPptDra1s8ftZkpvLdK+X8zZv5qJt29hQUjKoUcoBnqit5Zo9e3j7mDE8snAh/66tPbmv7iNXpxnD+sWLWefz9ZgmXgqccqCn+/F3pLWVS8rKGJeZyYbiYrb4/TH/vgu8Xkqrq2OQW0m0p+vrscCNhYWsr6nhmfp6fhM2GFyquLeqiu8dOsTHJ03ijtmzebymJuH11nDlTUtjUmamup+4wE3l5YxKS+MjEyfGdLuRrheerqvj5+XlzMrO5ouDnFYzaC0f2b2bFxsa+OuCBbxn3LhBXZtEc33T3zSlfj+3lpczOi1tyNc3FW1t+Do6uGzcOIpzc5N+DZnM69VkMrGKGCZKD1O6hnscuMEZXyN8vRXAWmAj8AZrbbDb+zmEuoeMBSZYa2ud5VfTjyldB7GfiNsbjpYuXWo3bdqU7GwkzY6mJha+/DL3z5vHB2P8Q5vq2oNBLtq2jafr61m/aFHEOdafqK3lndu2cf7o0awtKiJjgKPL72hq4tzSUqZ7vTxXXDzoAEm8jX72Wa6aNIlbZs1KdlaGrPrOTt64eTOHW1t5vqSEBTFuKtzlR4cO8fUDB/C/6U3kxKE7lcTPta++yp+OH6f6DW/gREcHBS+8wA3TpvGLmTOTnbWT/lNby4UxqCNl4M4pLSXb4+GJJUuSnRXpQXlrK4UbNybs/A1aywd37uSvJ07w4IIFXD5+/IC39ZV9+/jZkSP84qyzBh0giaefHz7M/+zfz+azz2bJIGcOS2Vf3bePXxw5wv7ly5muwYF7ZIx5xVq7NF7bd/Mv3XpO7aaxltCMJO8EbjHGdL8bvMh5fah7oAHAWtsEbCLUeuV1g8hXovYjQ1TBMG26aq3l2j17eKKujt/NmRMxoAHw1jFjuHP2bP5dW8u1e/YMqClnVVsbF23bRnZaGo8VFaV8QAPUZSHeOoJBrtixg93NzTy0YEHcAhqggQTdylrL2poa3j5mDJkeD1Ozsnjf+PH8/uhR/J2dyc4eEBof6LLt25kzYgR/XbBAAY0kUX2d+m6vrMQCn506NSH78xjDvXPnsjwvjw/v2sWLDQObE+A3FRX87MgRrpsyhS/EcHDTePjE5MlkezzcXFGR7KzETVMgwJ1Hj/Le8eMV0EgyN//a/dhae3XYv4uBM4GfAu8A1htjwh+BzXBef2aMsZH+8d+AxMDDp4nbjwxRuenp5KenD7uuBj88fJi7q6r4dkEBH500qde0V0+ezLcKCrirqoofHT7cr/00BQJcUlaGr6ODx4qKXPMjpFlx4sday3V79/Kv2lp+O3s2bxs7Nq77K1RQw5V2NDVxpK2Ni8L6Jd8wbRr1gQB/OHYsiTkL6QrWjkhL47FFi1wRrB2qCrKyONLWRsBlraGHi5ZAgDsqK3n3uHGcOWJEwvY7Ii2N/1u4kCmZmbyrrIz9LS39Wn+dz8eqvXu5eOxYbpo5M+WneB+TkcFHJ03iT8eOcWKIjhO35tgxajs7uSFBwTHp2ZD6xbPWdhhjvgZ8HFhMqNXGY87bXQGOp4GDfWzq0CCykaj9yBCW6k95Atayzudjs99P8SD66nVt596qKv5aXc3KCRO4sbAwqnW/W1jIgdZWvnHgALUdHeSlp/eYl679vOL3s9bp6/nIwoWUuKg5ZKHXy5N1dVhrU/5Cxg3Cy/BrLS384dgxvllQwMcmT477vjXwqzutrakB4J1hQa/leXm8buRIbi4v59opU2I62GA09WxXmpcaG/nz8eMcb2/n2ZKSk62BJDkKvV46rOVoWxvTXPpdxOp3PpV0HdPvjh7F19nJqilTEp6H8ZmZrF20iHNKS7lo2zZuLCxkb0tLn+f4upoa7qqqYlFODn+eP590l7TC+tzUqdxeWckdlZV8M8rrO7ew1nJzeTnFubm8YdSoZGdn2BtSQQ0Aa23QGXdjHDCP/wY1jjivD1prV8cxC4najwxhBV4vu5qbk52NiALWcuHWrbzY0EBTMEiOM6ry+sWL+3XB07Wd5xsaaAkG8QCV7e0E+W9ksDfGGO6YPZtHq6v5eXk5BiLmJTy/XdNbzXbhaNwFXi+NgQC1nZ2MzYg4Y7VEqXsZtsDEjAy+U9DbcE2xMykzk0xjFNRwmXU1NSzOyWFq2ADOxhhumDaND+/axT9ranhnjOqVaOrZSHXbopycIT9tnxuc7GLm0qBGrH7nU0n388UD/OjwYS4YMybhxzQnO5uHFizgrVu3snLXLiy9X7+80NBAczAYmskiLY0RLhqLaV5ODu8YM4bbKiv5yvTpZLokGBONJ2pr2dHczD1z5+phUwoYOiXLYYzxAIXOn+HToa5zXq+IcxYStR8ZwrpaaqTiQL7rfD42OhcFFvAHg7zY0MA6n6/f23nBCWgABIGX+7md/9TW0uF8Rl15eaKujnEbNjD5+eeZ/PzzjNuwgSfq6k6Zr7uyra3f+U02dVmInXU+38kL264zzB8I8LjzJD7ePMYwXd2JXKW+s5Pn6utP6XrS5Yrx45mUmclNMew33r2MRlu37W9pcV3dNhS5vTVWpPI3kN/5VBJ+TBC65ngpicfUFAiQaQxB+r5+aXbybIEtfr/rvocbpk3jaHs7D504keysxNRNFRVMyMjggxMmJDsrwhALajhTu/6IUCuNDkIzoXR5GHgFeLMx5jfGmNM6TRtjZhhjVg0yG4najwxhhV4vLcEgJzo6kp2V07zi99MUPHUM3KZgkC1+fw9rRPZcff3JH+qBbmez33/aNgBmer28Oz+fd+fnMzPCU7KB5DfZ3H6RnEo2RyjDzQkuE6nexUxO9e/aWjqt5aII461kejx8ZsoUHq+p4dUYtbCLVEZhaNZtQ1GBy+vrSOXP7WUr1Y5ps99Pe4QHV0PxHH/n2LHMGjGCm8rLk52VmHmtuZnHfD6unTKFrCHU+sTN3Nz95P850612GQcsAaYSCsDeYK092PWm0y3lUkKzpHwa+JAxZitQ7qw7HZgNHAMG3G0kUfuRoS18doQJmZlJzs1/WWt5OcKI3VkeT7+aPLcEAjxcXX3a8px+bqc4N5ccj+eUJ5W5Hg/fKSzkknHjAHi0upord+48JU1/95MK3H6RnEoilZtEl4mCrCweddmF6XC21udjdHo6y/PyIr7/6SlT+N9Dh7ilvJxbZ88e9P6y09LofrszVOu2oSg7LY3xGRmubY1VnJtLpjG0hd10pxvj6rJVnJvLCI/nlAchyTxfhtP1i8cYPjd1Kte/9hovNjSwrId61E1uragg3RiuTcK4LBKZm0NLFwIfDfv3NqAN+APwemvt7d1XsNaWA68HPgtsBhYAlwMLgUbg58Blg81YovYjQ1eqPpW/qbycx2pqOCMri1yPB0OoEmkPBpkUZfAlaC0f2b2bV1taKMrJObmdXKc/aX/GuliRn8+yvLxetxFNGjcYm55Oblqaay+SU8mK/PxTBolNRpko9Ho51tFBSyCQsH3KwHRN5fqOMWN6HJxvYmYmV06YwD1VVdQNsoXdgZYWfnzoEF5jyBkGddtQ5ebWWO8cO5Y0Y/AABsgwhg5rXXs8EDpfxjnjUaXC+TKcrl8Arp40iby0NG4eAq01Gjo7uauqivePH8/ksDGWJLlc11LDWls4yPVbCbWQiKqVhLX2HuCeCMsPEqoXY7WfXrcnw0sqPpV/+MQJvrhvH5eNG8ef589nfU0NW/x+pnu9fGP/ft6zfTsbS0o4o49B0f7f/v389cQJfn7WWXx+2jTW+Xxs8ftZMoDR1dOMYf3ixb1uI5o0bmCMcfVFcipJM4ZPT57MM/X1XDN5Mu/Kz094megKXB5ua2NOdnbC9iv9t8Xvp6q9PeJ4GuGunzaNe48d466qKr54xhkD2ldtRwcXlZURADadfTYHWluHfN02VBV6va7rJtDlSWcchy9Mm8bY9HSKcnP5XWUlN7z2GmeOGMHFLrypbgkEqOvo4E15ebxj7Nikny/D6foFYGR6Oh+fPJlbKyr4WVsbU1wcDLinqorGQIAbpk1LdlYkjEnFgQhleFu6dKndtGlTsrORdGOee46VEybEpCnzYL3U0MD5W7ZQlJPDk0uWkN1t5O0yv583bt5MgdfLc8XF5KVHjpf+pqKCz+zdy3VTpnDrrFkaLbqfLtm2jYr2djYvXZrsrLjeR3ft4jGfj2NveENSLhCfravjvC1beHzRIi6MME6DpI4fHjrENw4coOrcc5nYR4u0N23eTEVbG3uXLet3uWoPBrlw2zY21Nfzr8WLefPo0YPItSTb/+zbxy3l5TSfd15Mp/pNhEu2bWNTYyOHzjnn5HgB/s5O3rxlC682N/NMcbGrpkQHuK2iglV797KhuJhzNf1mUuxraWHWiy/yjYICvn/mmcnOzoAErWXOSy8xPiOD50tKkp0dVzHGvGKtjdsFrJu7n4gMaanyVP5ASwvvKitjUmYmjxQVnRbQACjKzeWvCxawq7mZ9+/YQUeEAe7W+Xys2ruXi8aO5aaZMxXQGIBUKRNuF7SWdTU1J5tYJ4Nms3GPtT4fS0eO7DOgAXDD1KkcaG3l0X7OTmCt5ZOvvspTdXXcPXeuAhpDQKHXS5u1HGtvT3ZW+mVvczOP1dScNgBibno6jxYVMTYjg0vKyjjiororaC03l5ezdORIzhkC4zm41VkjRnBJfj53VFbS6tKul+tqanitpYXrp05NdlakGwU1RFJUQVZW0m9gazs6uLisjHZreayoqNeL+rePHctvZs9mfW0tq/buPWU62q1+P+/fuZNFubk8MH9+j/3SpXeFXi91nZ3Ud3YmOyuu9kpjIyc6OvrsThBPU7KySDcm6ee49K6mo4MXGhoiznoSyaXjxjE9K6vfo/x/9+BB7jt2jO8VFrJy4sSBZFVSTKqOjdWXWyoqyOhhAMTJWVmsLSqiKRDg4rIyGlzyW/Sv2lpebWnhhqlT9UAlyW6YNo0THR38+fjxZGdlQG4qL2dqZiaXjx+f7KxIN64bU0NkuCj0enmirg5rbUJ/hAPWss7nY1NjI3+rrmZvczP/WryYeTk5fa77icmT2d/Swg8PH+ZMr5einJzQk8eqKkalpfFoURG5PXRNkb6Fz4qzyGWjn8N/y9Zmv5/iJPYNXltTgwHeMWZMwvfdJc0YzsjKGtItNaL5vlOlTPTknzU1BIEVUQY10p3pXb924ACr9uxhxdixfR53TWcnvy4v5+pJk/hmQUEcjkKSocAZM+BQayvnuKS7Q0NnJ3dXVfGBCROY1MOYBwtzc3lowQJWlJXxvu3buW7qVMqamuJ+jg9mOzeVlzMpM5P3T5jQ7/1KbL1l9GgWZGfzvUOHONzWRkkK1PvR/lbdUVnJv2pruWriRNd1KRsOdHchkqIKvV78gQA1nZ3kOyN2x1vAWi7cupUXGxpOTiE2PzubN/WjKfT3zzyT11pa+PqBA2SFTQn3+pEjo54hRSILf/LntqBGeNlqCgbJcUZxX794ccIvZtb6fCzLy2NcksvjUO5OFM33nUploidra2rIT0/ndVE2WQ84XZsAbqus5A9VVX0etwXGpKVxm8YZGlJSccDvvtxdVYU/EOCGPprWv23sWG6fNYtP7dnDk/X1BKyN6zk+mO282tzMupoabiwsJFOtRJMuSOj7PNDayo0HDya93u/Pb9XT9fUA/O3ECSrb2lLqt0oU1BBJWeEXRIkKaqzz+U4JaAAcbm1lnc93ct70vniM4QPjx/PXEydOmeN+Z1NTv7Yjp3Nrc2Y4vWz5g0FebGhIeJk40d7Oy42NfLewMGH77EmB18s/nRvgoSbS9/10XR1v27qVac4T4PK2Np6rq6OrAXuyykRPgtbyeD/HXlnn81Ha2Hjy72iOG6DdWp6orU2J45bYGJmeTn56umvq64C13FJezrl5eSyNIog3KTPz5FSvEN9zfJ3Px0bnprO/27mlvJzMHrrTSOKt8/k40tYGgCX59f5AfquaUuy3SkIUshRJUckYSHCz33/yoqFLUzDY72npdjQ3031epYFsR041LiODER6PK7ssxKpsDdb6mhosJHU8jS6FXi9H29tpizCwrttF+r47CY2v83x9Pc/X17PV76d7j/xUqicGMvbKQI+7OYWOW2KnwOvlkHMDl+rW+nzsa22NeprKzX4/nd1mUIzXOV46wN+P+s5O7qmq4oMTJkQ10K/E32a/n+YUuBYIz4/bf6skJClBDWPMjcYYa4y5MRn7TwRjzNXOMd6T7Lz0lzFmujHmWmPM340xu40xzcaYRmNMqTHm28YYDR2dAMl4Kp9uzGnBiByPhyX97OpQnJtLTrdmngPZjpzKGOPaLgtpEcqWNwllYm1NDRMzMihOgbJY6PViwVWzCERrdnb2actyPR7+MHcu+5YvZ9/y5fxh7lxyu9UTmcakTD3RNfZKf6bcjVT3RXPcqh+HJjfV1zdXVDA1M5P3RvnkeaBlfcQAyvrWCDeP2VFs566jR2kKBrk+ykCNxF+kcpOVxHq/ODeX7u3wVGe7k1pq9MIYc9AJTBQmOy8J9ifgduASoAF4BHgBOAv4LrDVGOPOCaZdZHR6OiPT0hL2VH5XUxM/OXyYbI+HHI8HQ6hiX5aXx4p+PtVekZ/Psrw8cge5HTldgYsukrvsbGriJ4cOnVK2DKHmzvOjGIA2VgJOd4IV+fkpMchX10CCbvs++9IRDHJnZSWWUOCqpzqgez2RBrRZmzJ9lNf6fCzPy+tX979o6j7Vj8NHgdfLodbWU2YDS0U7mpr4d20tq6ZOJSPKcScGUtYh1K3r7JEjo87bbyoq+Ft1NVMzM0/Zjtfj4e29DPYcsJZbKip4Q15ev/Yn8dW9THgI1ftjkjSI/PiMDIKEAuqqs91NY2pIJBXAF4D7rLW+roXGmPHAX4DzgXuANycjc8NFIp/KH29v5+KyMrI8Hl4pLmZXczNb/H6WDHBU6jRjWL94Met8vkFtR05X6PXyckNDsrMRtWNO2cpOT2dDcTE7m5rY4vczNiODr+3bx7vLythQUsKoBFzQvNjQQG1nZ9QzWcTbyS5mLmmeHg1rLdfu2cMTdXX8fvZsJmRm9lgHdK8nZmdn8/2DB/nAzp1sKC6mKIlPwQY69ko0dZ/qx+Gj0OulJRjkREcHE1K4+8PN5eV4PR6u6ce4EwMp69lpaXxj/37eu307Ty5Zwoi0tF73sc7nY9XevVw8diwPLVjAv2pr2eL3c7StjduOHuVL+/ZxSw8D7D7q83GgtZWfzJgR/Qchcde9TJzp9fKtAwd4744dbCwpYcaIEQnNz62VleR4PPx+7lz2NjerznYxBTXkNNbaD/Sw/IQx5irgCHCeMeYMa+2RxOZueElEUKM5EODdZWVUtbfz1JIlnJWdzVnZ2YMe/CjNGC4ZN06DKMVYodeLr7OTxs5ORqb49LjNgQDvKivjeHs7Ty9ZwowRI5gxYsTJMjE3O5sLt23j8u3bWbtoUdxHpl/r85EGvT7dS6RpWVmkMbRaavzo8GHuqqrimwUFfNy5QeqtDuheT5yTl8fy0lIuKivjxZISpvQwrWS8DWbslWjqPtWPw0N4N9JUDWrUdHRw37FjfHjixH4PSj6Qsn6m18vlO3Zw1a5d/GXBgh5bzW1pbOT9O3eyODeXP8+fT1Za2inbyU5P5+dHjnDWiBF84YwzTlv/5vJyzsjKiro7jSRO9zKxNC+Pc0pLuWjbNp4vKWFsggbHr2pr44Hjx7l2yhQ+0Md0v6qzU9+gryCNMcuMMT8zxmwyxhwzxrQbYyqNMX81xiyPYv0zjTF/dNZtNcbsMMZ8yRgT8WrdhFxljHnKGFPrrLPPGLPaGHNarWaMKXS6kBzsJQ/WGGPD/r7a+btrwvgDXWkG0h3FGDPOGHO7MaY8LL8/MMac3un4v+tcbIxZZ4ypdj7TI8aYe40x87qlm22MaTDGdBpjzouwnfnGmCZjTIcx5pz+5DsSa205UO38qU6KcdbVdDVegtZy1a5dvNTYyJ/mzeP1UU5bKMnT1WUh1QcLDVjLyl272NTYyJ/mz484mv5bxozhd3Pm8ERdHdfu2RP3Jtpra2o4d9QoxiTogqkv6R4PU7OyhkxQ4/5jx/jGgQN8aMIEvjfA2WXO8Hp5tKiI2o4OLikrw9/ZfXi2xEilsVfEvQqSMOB3f9159CgtwSDX9zGNa6y8d/x4fnHWWTxUXc1X9++PmKa8tZWLy8oYnZ7Oo0VF5EYI4P9kxgwuHzeOL+3bx99PnDjlvTK/n//U1bFq6lTSNY1rypuTnc3DCxdyoLWVy7ZvT9jg2b+prKTDWj6XoLIv8RWLM/1/CXVVyABeIjT+gg+4HHjOGHNFL+ueCWwCLgCeAp4kNG7Dz4EHjTGn5M+E2pf9EfgDcC7wMvAwoe7Z1wFbjDGvi8ExvQbcCzQ5fz/k/N31rz/D3Y4BXgTe77yuB8YD3wCeiBTYMMb8CHgUeAewA/grUA98BCg1xlzcldZauwf4NKHuyH8yxowL20428CCQDXzdWvtCP/IdkbP9rsecRwe7PeldoddLfSBAXUdHXLb/lX37+Ft1Nb846ywuHT8+LvuQ2HJLl4X/2bePh6ur+fXMmbynlycbH500iW8XFHB3VRU/PHw4bvk52tbGZr+fi1Kk60mXwjgHLhPl2bo6rt69m/NGjeKuuXMjNgePVvHIkfxlwQK2+v18cOdOOhM8O0yqjb0i7pXq4+Z0BoOsrqjggtGjE9rd6/PTprFqyhR+fuQIt1dUnPJeQ2cnF5eV0RgIsLaoqMfWWh5juG/ePJbl5bFy1y5eDOuWeXNFBSM8Hj45eXJcj0Ni57zRo7l77lyerq/nk6++GveHHG3BILdXVnLR2LHMijCwtbhPLNou/xxYaa09Fr7QGPMuQsGA3xhjHrPWNkdY9yNOmg9ba1ud9WYRCm5cClwL3BaW/jPAh4BjwFuttTucddKAXwGfIxQMmWOtHfAVv7X2OUIBmfOBHODL1tqDA9zcu4ENwNnW2jonvxOBfwHLgRuBr3QlNsZcBPw/QgGVi6y1z4S99z/AT4E1xpjZ1trjTn7vd/J6DfAHY8zFNlQbrAbmA2sJfU+x8GVCAZTSQXwmEqXwpqtLYvB0OWAt63w+Nvv9VLa18ZujR/ns1Kl8XiODu0YyZsWJRnjZqmhr446jR7l+6tSoRp2/sbCQA62tfPPAAQqyshidns5mv5/iGPZbfbymBkiNqVzDFXq9PFlXl+xs9Fv49z0hI4Ov7d/PmV4vf1+4kKwYPBm9KD+fW2fN4rq9e/nc3r1clJ/Pll7KRHh+BlNuAtbyyyNHqO3sZGJGBoEUGrhU3Gd0Rgaj09NjVl9HU877k+aPx45xpK2NX8+cGZP8RcsYw69nzuRQWxuf3buXaVlZGELTKD/i87G9qYl1ixb1GWgZkZbG/y1cyDmlpbyrrIzni4t5qbGRe6qqeMvo0YxO8S6acqoPTZx48lqgMCuLZXl5gy7rPXng+HGOd3REPYWxpL5Bn+3W2sd7WP4PY8yDhIIQFwCPRUjWAlzXFdBw1ttrjPkWcBehFiDhQY0vOa/f6gpoOOsEjDFfJhQIKQDeB6wZ8EHFlgU+0xXQALDWHjPG3AD8B7jWGPPtsM+g6xhvCg9oOOv9zBhzObAM+BShVjJdbiAUJFkB/I8xpgq4GigHPmJjEPI0xryNUFAjGJZPiaOTXQ3a2lgyyNG7A9Zy4datvNjQQFMwiAXy09P5xVlnDeqpqiTWhMxMsoxJqaf7kcrWuPR0fnbWWVGtb4zhzjlzONTSwkd278br8dAaDJLjjDC+fvHiQd9Yrq2pYWpmJkUJnG0lGgVeLxVtbXQEg1HPOpBs3b9vCE0HvbGoKKZ9oT8zdSqvtbTwy/Jy7qqqosPaiGWie34GWm66tvNMfT0At1ZUsKmxMSblT4avgqysmNTX0ZTz/qbxB4MYYHVFBe8ZNy6h5Tzd4+H+efN40+bNXLp9O1keDy1OfTJ3xAjeGuXYRxMyM1m7aBHLX3mFxZs20WEtndbyXH09F27dqvPXZb4+fTqvNTfzg8OHyTKG9jjU+9Zabi4vZ152dsqMsSWDF5MrKGfMiKuNMT83xvzOGHOPMeYeYKGTZHYPq/6zq7VBN2sI3TjPNMZMdfYxDZjhLL+v+wrW2nb+G8g4f8AHE3vbrLVl3Rdaa58kNMvISOBsAGcckTc4Se7pYXt3O6/nd9teK3AFoa4x/0toStYAcGX4DCYDZYwpItSVJQ34trX2qcFuU/oWy6fy63y+kxcxXRGu1mCQfzpPsMUdPMak3LSukcpWSz/LVpbHw2ecfq0tznb8wSAvNjSwzje4KqzDycuK/PyUC+AVer0EgfIU704Urvv3bQkFNfY0R2qQOThvHj2aNKDd2pNl4rn6er68bx/3VlVxb1UVX963j+fq60/mZ6Dlpuu4OpxnAE0xKn8yvMVqwO/u512050JvaSB0/r6UpHKem57O5884AwsnAxoQqg/7k5852dl8dfp0moPBk+dvs85fVzLGhAJshKZ6jUe9/3xDA6/4/Vw/dWrKXRPIwA26pYYx5tPALwmN29CTnkYfPBBpobW23RhzFJhKaDDKCuf/AEfDW3Z0s895TaURXyIeo+Mg/z1GgHwgi1Dg5lAP6/R4jNbaPcaYrxBq3ZIOfMfpSjMoxpi5wL+B0cAvrLX/2/saEiv5GRlkezwxuSDa7PeffKrapTkYZIvfr9GcXSZRU/1GK1Zla29Ly2nLmmJQRp9vaKAhEEi58TTg1MDlmQmeym6gIn3frXGqS7b6/XQfUaPNWn5dXt7regMpN5GOKxblT4a3Qq+Xf9fWYq0d1A1UpPIZzbkQr/MlVg5H+C0bSH46rMUA4c2Sdf66U1lTU1zr/ZvKyxmdns5VkyYNMqeSSgbVUsMYs5RQi4AM4H+AuUAu4LHWGuBHXUkHsZuu+sl0+ztilvq78e6DkSZJ92MMX9Zdj8fojC3ywbBFrzODDEEaY2YT6iYzAbjNWvvlwWxP+scYE7OBBItzc/F2a96e4/GwRKP7u068Z8Xpr9kRBtkaSNkqzs0lp1sZzY5BGV3r85FhTNTNmRPJDbMjdFecm0t6t5+WeNUlPZWJO2fPZv+yZexftow7Z88mOwblxmPMaT+8qiNlsAq8XpqCQWoGOZPPYM6FvtIks5xHOq5Y/X7o/HWneNb7R1pb+duJE3xy8mRy0tJinndJnsHe0L+P0E32zdban1trX7XWNoWN39DXyEOFkRYaYzKBriGLK53XrvDcFGNMTxPXn+m8hg+l3O689lTKC3pYHiuFUbzXdYzVQBuh76Wn9SIdY5cbgfMIzUKzA7gE+GK0Ge0ubNDWycCdwGcHui0ZuFg9lV+Rn0+eU4EbINfpf7gixQZOlL4Ver0c7+igORBIdlboCAa5s7ISC3g9nkGVrRX5+SzLyyPX2Q5AdloaFw6yhcW6mhreNGoUeSk4aNwZzgB5qdTypi/lbW10WEu6MXGvS7qXiVyPh3Py8vjY5MmcOWIEZ44YwccmT+acbuUm3Rje0o8g1o6mJn566BA5Hg85gyzHIuFi1Y10RX7+yVkaoj0Xok2TzHIe6RyPxe9Hso9LBm6g9b7X4+EdfVwv3OZcr3xW07gOOYO9wusqOUe6v2GMGQ+8vY/132GMGW+tPdFt+ZWEbuz3WWvLAay15caY/YTG1fgw8Ptu+8sgNCgphKaH7XKCUGAjv4d9XdRL/roCIoP5nBYbYxZaa7d3y++bCXUh8QOvAFhrO40xG4C3EJoZ5lsRtne18/pUt+29Ffg6UAd8gNCsLS8BPzLGPGutfak/mTbGnEUooDGF0Dgen47FYKPSfwVeLxvDpiobqGPt7VR3dPDu/HxeN3IkS2I4s4QkVvjT/XlJHPjSWstn9uzhibo6fjd7NhMzM9ni9w+4bKUZw/rFi1nn87HF7+dgayu/r6rimwcO8JMoBx3t7khrK2VNTfxsxowBrR9vmR4PUzIzXRPUeNzn47N79/LOMWO4dsoUypqa4lqXdC8TkfbVPU1DIMDPjhzhk6++yh/nzetzWtaqtjYu2raN7PR0ni8uZkdT06DKsUi48Pr67EEM+J1mDGdkZrIvLY0vTJvG0pEj+zwXojlfkl3OY5WfVDsuGbiBlOPKtjZuP3qUL772GrfMmhWxq1dzIMBvKyu5dNy4k+elDB2DDWrsdl4/Yoz5vbXWD2CMGUlo9pLRfayfDdxqjPlI1xSszs309533b+qW/pfArcD3jTEbrLW7nXXSCE11WkBoLIq/dq1gre0wxjwLvBX4njHmuq6bc2PMG4Hv9ZK/CmAWMA94rY9j6YkBbjPGvMtaW+/sd3zYsf3WWhvekfyXhIIanzfGPG6t3XByQ8Z8ETgHqAd+F7Z8IqFBUj3AJ7qmWjXGXO+ke8AYUxw+A0uvGTbmTEIBjanAvcAnFdBInkKvl5rOTho6Owf1pPn2ykqCwC9nzuQsl/Tdl8gKUySo8ePDh0NBh4ICPjFlCsCg+y6nGcMl48ZxybhxWGvJ8nj46ZEjzBgxgk87++iPdSk6lWu4Qq+XQy4YKHSr38/7d+6kKDeXvyxYwMj0dN4zfnzc9xteJqJNMzY9na8dOMAMr5cf9BLQagoEeNf27VR3dPBMcfHJp4Dqgy+xEquWGvtbWvhHTQ1fmz6dG888s8d0Azlfki1W+Um145KBG0g5zklP5+dHjnDWiBF84YwzTku/5tgxajo7o5pqXtxnsEGNu4HPAyXAfmPMc4Ru4s8j1MrhLuDjvax/H3AxsM9poZBL6IbeC/wDWN0t/W2EZge5EthqjHkSqAVeT6gFRy1wRVeAJMy3gTcB1wJvNsbsIBQAORv4IfDNHvL3d0KzjKwxxvyTUCsIgK/2Y0aRRwjNArPPGPMUoc/8AkKDp77s5O0ka+1jxpifAF8FnnECMpVAkbOdVuDD1tpjcHJMkDXARGC1tfZvYdv6vTHmLYRasPweuDzKPD8EnMF/u8Lc1cPQHD/uCixJ/ITfwPY1Z3tPWgMB7qis5JL8fAU0hoBYzoozUPcfO8bXDxzgQxMm8L3CwrjswxjDTTNncqi1lVV79jA9K6vfTYnX+nwUZGUxL8K4H6mi0OtlQwxaY8VTRVsbF2/bRl5aGo8WFTEyBbvyhPvq9Onsa23lfw8fZsaIEXx88uTT0gSs5UM7d1La2MjDCxcO6im6SE/GpKczMi1t0PX1rRUVpBnDdWo2LxLRT2bM4EBLC1/at49Cr5f3hgXdrbXcXFHB4pwczhs1Kom5lHgZ1Jga1tpaYCnwW0LdKC52/v4boUDHad1SutkPvA54jtCN/tsIzQjyFeBya+0pg986rQVWEuqa8SKwHLjMOY7bgcXW2pcj5PN5Qi01niB0s97V5eQj1tpIXTy63EqoC0gFofEpPuH868+VT62Tz78TamWxAvARCqZcYK1tipDf/we8C/gXoWDG+4AxhIJAZ1trHw1L/g3n2LYAX4qw/08De4HLjDHRjonR1a0oC7gK+GgP/zRscALEYiDB+48f50RHBzcoOj0kTM7MJMOYpAU1nqur4+rduzlv1Cjumjs3rlOipXs8/Hn+fBbl5vL+nTvZ6vdHvW5bMMi/a2u5KAWncg1X4PVypLWVzmD38d5TQ2NnJ5eUlVEfCPDYokVMzeppWKvUYYzhtlmzeMeYMXx6zx7+FWF64S+99hqP+HzcNHMm79KTXYkT40zDPZjf8MbOTn5/9CjvGz/eFeefSDJ4jOG+efNYlpfHyl27eCnsYcGTdXVsb2rihmnTUvp6QAbOqFeBpJqlS5faTZs2JTsbKeNYezuTnn+eW2bO5LMDCEpYayl55RU6rWXb0qWqzIeImRs38rq8PO6fPz+h+93T3Mw5paWMz8jg+ZISxmZkJGS/lW1tLCstxVrLxpISpkXRH/bfNTW8fds2/rFwYUo3R76zspJr9uzh0PLlTE+xfr6dwSDv3r6df9bU8GhREe9M4W48kTR0dvLGzZs51NrKhuJiFjqt3W4uL+eG117jC9Om8cuZfY1pLjI47yor43BrK1tf97oBrb+6ooLP7t3LC8XFLNdTZpFeHW9vZ3lpKU2BABtLSjhzxAjeU1bG8w0NHFm+HK9mPUkKY8wr1tql8dp+KkxnKiK9mJCRgdfjGfBT+Wfr69ni93P91KkKaAwh0cyKE7CWR6ur+f7BgzxaXU1ggEHsru18bd8+zt+yBQM8tmhRwgIaAFOyslhbVERDIMDF27bxl+PHez2ugLWsrqggDWgNBgd87ImQCt2JugtYyz+qqzl382bW1dRw66xZrgtoAOSlp/NYURE5aWms2LaNe6uq+NDOndzw2mtcmp/PzwY4AK1IfwxmFrOgtdxcXs7rRo5kWV5ejHMmMvRMyMxkbVERHdayYts2fn3kCI/4fLxl9GgyPLr1HapSu1OsiISarmZlDXggwZvKyxmbns7KiRNjnDNJpgKvl7URmtR3CVjLhVu38mJDA03BIDnO9HbrFy/u12jwXdvZ6GwH4Ozc3JM34olUlJvLA/Pnc1FZGR/auZMgoXnpi3Nz+dP8+SePq2ushA0NDVjgY7t385vKyn4fe6IUhAU1zktyXuC/3/mz9fW0W0uGMfz1xAk+NWVKSn5+fTnD6+X/Fi5keWkpH9u9G0voiU5dZ2eysybDREFWFg2BAHUdHYzuZzB4fU0Ne1pa+OO8eXowIRKluTk5/HXBAt62dStf2LcPgMd8Pi7cujVlrwVkcBSuEnGBgT7lOdjSwsPV1Xxq8mSy1dxuSCn0eqlqb6c1EIj4/jqfjxcbGvAHg1jAHwzyYkMD63zRjnF86naawsZ7eLW5ud/biZWAtWQZQwCwQFMwyHMNDUzfuJGpL7zA1BdeYPrGjTznBDRg4MeeKNOdPvKD6XMfS+t8Pp5vaKDdad3SYW1Kf37RONbeTqbHc7JMBIFNjY2uPiZxj8G0xrqpvJzJmZlckYCZhkSGkuZAgMyw4EVTil8LyOAoqCHiAgMNatxWWYkBVmm09CGn6yL5cA8teDb7/acEIiD0g76lHwNtdm3HH4PtxMpmv//kzXa4d+fnc8fs2dwxezbvjtBNIpl57os3LY3JmZkp0/1ks99PSwp957Gw2e+nbYgdk7jHQIMau5uaWF9by2emTCFTzeZF+iXS9YLq/aFL3U9EXKDA66W6o4OmQICcKFtcNAUC3Hn0KJeNH88ZKTb4oAxe+Kw4syNMV5qblkb3W38LTMrM7Nd+jrW3n7Ysx+NhyQCnFx6s4txccjyeUwItuR4Pn5o8+eRgoFMyM/lPbe0paZKZ52gUDKLPfawVRKgvUv3z60ukcuP2YxL3OFlf97Mb6S0VFWQawzVTpsQjWyJDmur94UVhXxEXKBzAtK73VVVR19mpaVyHqN6e/B1qbeXHhw7hNYZsjwdDaOyJNOAXR45Q29ER1T4e9/m4vbKS/PR0cpzt5Dpjc6xI0qCRK/LzWZaXR24v+YkmTaopHOSUj7G0t7kZ4GTZccPn1xc3lgkZOsZlZJDdzwG/6zo6uLeqiisnTGBiP4PRIqJ6f7hRSw0RFwgPaszPyekzvbWWmysqODs3l3M1WvqQNCUzk3RjTrtIruvo4KJt22gHXj77bA62trLF72dJbi4jPB5WlJVx+Y4dPL5oUa/Nmbf6/bx/504W5eby5OLFPOfMorMkN5cV+flJG2QrzRjWL17MOp+vx/xEkybVFHq9PHTiBEFr8SQxn23BIL89epSLx47l2ilTXPP59cWNZUKGDmNMvwOXv6+qoikY1IMJkQFSvT+8KKgh4gIF/eyP++/aWnY1N3Pv3LkaLX2ISvd4mJaVdcpFcnswyOU7drC3pYX1ixaxMDeXhbm5J7tlANw1Zw5X7d7Np159lXt6KB8VbW1cvG0beWlpPFpUxOiMDC4ZN+6U7SRTmjF95ieaNKmkICuLDms52t7OVGfg0GR44Phxjnd08Plp03jb2LGu+fyi4bYyIUNLf7qYBazl1ooK3jRqFMUjR8Y5ZyJDl+r94UPdT0RcYFJmJpkRnsr35KbyciZmZPCBCRPinDNJpvABZK21fHrPHv5TV8fv5szhgjFjIq7z4UmT+G5hIX84dozvHzp02vuNnZ1cUlZGfSDAY4sWJfUGezgZzOwIsWKt5abycuZnZ/PWHsqPiAxMfwb8/kd1NQdbW7leg3yLiERFQQ0RF/AYw/Qom67ubW7msZoarp0yhSyNlj6kFWRlnbxI/t9Dh7inqorvFBTwkUmTel3vWwUFfHTiRL5z8CD3VVWdXN4ZDPKBnTsp8/t5cP58FmswrYTpb2useNhQX0+p38/106aphZdIjBV4vdR2dtLQ2dln2psqKpielcWlerosIhIVdT8RcYGAteR4PGyor+fR6uqIfQID1rLO5+MHhw6RBnxq8uTkZFYSZnpWFhXt7by3rIyHfT5WTpjAdwoL+1zPGMNv58zhSFsbH9+9m/K2NjqsZWNDA+tqarhj9mzeqYG0Emqa0yLm7qNHGZWWlpR+vzdVVDAmPZ2rJk5M6H5FhoPpzjn+9f37eefYsT3+jt9WUcFTdXVcPXGigosiIlEy1naf9E8kuZYuXWo3bdqU7GykjIC1XLh1K0/X19Np7cnRm9cvXnzygqgrzcaGBpqCQdKBN48efUoaGVoC1lL08svscmaq8BD6zv/Vj+/c197O9I0baQ6b7qwgK4t9y5er3CRQ1/n7RF0dQMRzPN4Ot7YyY+NGvnTGGfzkrLMSsk+R4SJgLctLS9nU2IghNK1kT7/jT9fV0emkWZ7gekBEJF6MMa9Ya5fGa/tqqSGS4tb5fLzY0ECnE4D0B4P8p66Owo0bGZmWBkBjIEBFWxtdIcpO4MWGBtb5fBocaYha5/Od0lUhCLzcz+/8hYaG05b5OjpUbhKs6xzv4g8GE37+3lZRgQVWqQ+/SMyt8/nY1dQEgCW63/GmJNQDIiJupQ73Iilus99PU9iTdAhdFI1KS2NhTg4Lc3IYlZZG9zZXTcEgW/z+hOVTEmuz309rt3LR3+98s99PyyC3IYMX6RxP5PfQHAjw26NHee+4cUx3xvYQkdjZ7Pef0iIO9DsuIhJLaqkhkuKKc3PJ8Xjwh10Q5Xo8/HjGjJNPbx6trubKnTtPSZPj8bBEAz0OWZHKRX+/81hsQwYv0veQaUzCvoc/HjtGbWcnN0yblpD9iQw3+h0XEYkvtdQQSXEr8vNZlpdHrseD4b/97VeEDeQYTRoZWmLxnavcpIbu30Ma0GZtQvrRW2u5ubyc4txc3jhqVNz3JzIc6XdcRCS+NFCopBwNFHq6rplNtvj9LMnN7XX2k97SyNASi+9c5SY1hH8Ps7Oz+f7Bgxxqa2NDcTFFcXxS+0RtLW/bupV75s7lo31MBSwiA6ffcREZzuI9UKiCGhJzxhgPcAPwaaAQOAH8Bfi2tbapr/UV1BCR4e5IayvLS0tJM4aNJSVMcaaDjLV3l5WxsaGBw8uX43UGLBQRERGJpXgHNdT9ROLhV8AvgZ3A54AHgeuBfzgBDxER6cUZXi+PFhVR09HBJWVl+Ds7Y76PfS0tPOrzce2UKQpoiIiIiGvpBlNiyhizgFAg42/W2sustXdaa78IfBG4APhgUjMoIuISxSNH8pcFC9jq93Plrl0EYtyy8taKCtKM4TNTpsR0uyIiIiKJpKCGxNqVgAF+3W35nUAz8OFEZ0hExK0uys/nllmzeNTn43N79vCP6mq+f/Agj1ZXRwxyBKzl0SjS/OX4cW6vqOANeXlMyMxMxKGIiIiIxIWmdJVYex0QBF4KX2itbTXGbHHeFxGRKF03dSqvNTfzq4oKfl9VRYe15DgzI6xfvPjkQIIBa7lw61ZebGigKRjsNc2z9fW0W8tLjY1cuHXrKWlERERE3ERBDYm1KUC1tbYtwnsVwLnGmExrbXuC8yUi4lrnjxnDzRUVtDstL/zBIM/U13NpWRlnjhgBwIGWFp6pr6ejH2lagkFebGhgnc/HJePGJeHIRERERAZHQQ2JtWwgUkADoDUszSlBDWPMNcA1ANOnT49b5kRE3Gir30+w27IOa/l3XR0jGhqAUICio1t3k2jSNAWDbPH7FdQQERERV9KYGhJrzUBPcw96w9Kcwlr7W2vtUmvt0vHjx8ctcyIiblScm0uO59Sf7FyPhwfnz6fmjW+k5o1v5MH588kdQJocj4clublxPwYRERGReFBQQ2KtEhhnjIkU2JhKqGuKup6IiPTDivx8luXlkevxYAgFK5bl5bEiPz/maURERETcRN1PJNZeBt4BvB54tmuhMcYLLAGeSU62RETcK80Y1i9ezDqfjy1+P0tyc1mRn3/K4J6xSiMiIiLiJsbGeN57Gd6MMUXAVuDv1trLw5Z/DrgZuMpa+8fetrF06VK7adOm+GZURERERERE4s4Y84q1dmnctq+ghsSaMeYW4LPA34G1wDzgemAD8BZrbffx7rqv3wi8Gu98isTZOKA62ZkQGSSVY3E7lWEZClSOxe3mWGtHxmvjCmpIzBlj0oDPE5rNpJBQJfwA8G1rrT+K9TfFM5InkggqxzIUqByL26kMy1CgcixuF+8yrDE1JOastQHgF84/ERERERERkbjQ7CciIiIiIiIi4koKakgq+m2yMyASAyrHMhSoHIvbqQzLUKByLG4X1zKsMTVERERERERExJXUUkNEREREREREXElBDRERERERERFxJQU1JCUYYzzGmC8YY3YbY1qNMUeMMb8wxuQkO28i4Ywxs40x3zPGbDTGnDDGNBpjthhjvhGpvBpj5hhjHjbG1Bpjmowxzxpj3pKMvIv0xBiTbYw5YIyxxphbI7yvciwpyRgz1hjzc2PMa871wwljzJPGmDd1S6cyLCnJGJNrjPm6MabMuaaoNsY8b4y52hhjuqVVOZakMcZ8zRjzoDFmv3O9cLCP9FGX18HeC2pKV0kVvwKuB/5OaCrYec7fxcaYt1lrg8nMnEiYjwOrgEeANUAHcAHwA+D9xpjl1toWAGPMWcDzQCfwU6Ae+BSw3hizwlr77yTkXySS7wHjIr2hciypyhhTADwF5AK/B/YAo4BFwNSwdCrDkpKMMR5gHXAucC9wC5ANXAncTeh6+KtOWpVjSbYfAjVAKTC6t4QDKK+DuhfUQKGSdMaYBUAZ8Hdr7eVhyz8H3AystNb+KVn5EwlnjFkK7LXW1ndb/gPgG8DnrLW3Osv+AlwOnG2t3eIsywV2AK3AXKtKWJLMGFMCvAR8hdCFxGpr7WfD3lc5lpRkjHkWKAReb6092ks6lWFJScaYcwjd+P3aWvuFsOWZwG5grLV2tLNM5ViSyhgzw1q73/n/diDXWlvYQ9qoy2ss7gXV/URSwZWAAX7dbfmdQDPw4URnSKQn1tpN3QMajgec14UATnO5dwNPdVXmzvp+4HfAbOB18c2tSO+MMWmE6trHgb9FeF/lWFKSMeY84I3AT621R40xGcaY7AjpVIYlleU5r5XhC6217UA10AQqx5IaugIafRlAeR30vaCCGpIKXgcECT0pPMla2wpsQZW0uMM05/WY87oIyAJeiJB2o/Oqsi3J9gVgLvDZHt5XOZZUdZHzetgY8w+gBWgyxuwxxoRfAKsMSyp7CagDvmKMucIYM90Zh+BHwNnAjU46lWNxk/6W10HfCyqoIalgClBtrW2L8F4FMM5phieSkpyn3d8m1G+wq3ncFOe1IsIqXcumRnhPJCGMMWcC3wW+Z6092EMylWNJVXOc1zuBscBHgU8A7cB9xpiPOe+rDEvKstbWEnqiXQP8BThEqNvJKuBya+2dTlKVY3GT/pbXQd8LaqBQSQXZQKRCDKE+V11p2hOTHZF++zWwHPi6tfZVZ1lXM+hIZbu1WxqRZLgdOAD8spc0KseSqkY6r43ABU5zfYwxfwf2Az80xtyLyrCkPj+wndAA5M8TCtKtAv5kjHmPtfZfqByLu/S3vA76XlBBDUkFzcCEHt7zhqURSTnGmO8Tarr/W2vtj8Le6iqzWRFWU7mWpHKa578DOM9a29FLUpVjSVUtzuv9XQENCD35NsY8AnyEUGsOlWFJWcaYIkKBjC9Ya38Ttvx+QoGOO51ZJFSOxU36W14HfS+o7ieSCioJNSuKVPCnEmqOpFYaknKMMTcC3yQ07dq13d7uGvQrUnPQrmWRmuWJxJVT1/4SWAtUGWNmGmNmAgVOklHOstGoHEvqKndeqyK81zUTyhhUhiW1fYHQTduD4Quttc3AY4Tq5UJUjsVd+lteB30vqKCGpIKXCZXF14cvNMZ4gSXApiTkSaRXxpjvAN8B/gB8MsI0amWEmtKdE2H15c6ryrYkwwhgPHAxsDfs31PO+x92/v4kKseSuroGlJsW4b2uZcdRGZbU1nWDlxbhvfSwV5VjcZP+ltdB3wsqqCGp4AHAAp/vtvxThPpPrUl0hkR6Y4z5NqERye8DPmatDXZP40xb9Q/gfGPM4rB1cwndLO6l2yjPIgnSBFwR4d91zvuPO38/onIsKexhQuNpfNgpjwAYYyYDlwJ7rbWvqQxLitvpvF4dvtBpKfceoBbYp3IsbjKA8jroe0Fz+sNFkcQzxtxCaFyCvxNqEj0PuB7YALwl0k2jSDIYY1YBtwKHgW8RmoIq3DFnUC+cJv0vAR3Ar4AGQhV0EXCxtXZ9ovIt0hdjTCGhgUNXW2s/G7Zc5VhSkjHmGuAOYAdwF5AJfAaYDFxirf2nk05lWFKSMaYAKCXUVWoNoevesYTKZyGwylp7m5NW5ViSyhhzFf/tqvo5QnXuL5y/D1lr7wtL26/yOth7QQU1JCU4U2J+HriGUCVeTShq920n2ieSEowx9xCaOrAnT1trzw9LPw/4MfBmQpV/KXCjtfbfccymSL/1FNRw3lM5lpRkjLkM+AqhC+Ug8ALwXWvthm7pVIYlJTkDgX4beCswkdAguFuAX1tr/9YtrcqxJI0x5ilCZS+SU65/nfRRl9fB3gsqqCEiIiIiIiIirqQxNURERERERETElRTUEBERERERERFXUlBDRERERERERFxJQQ0RERERERERcSUFNURERERERETElRTUEBERERERERFXUlBDRERERERERFxJQQ0RERGRQTLGnG+MscaYq5OdFxERkeFEQQ0RERGRKBljlhhjbjTGFCY7LyIiIgLpyc6AiIiIiIssAb4DPAUcDFv+DDAC6Eh4jkRERIYxBTVEREREBslaGwRak50PERGR4UbdT0RERESiYIy5Ebjb+fNJZwwNa4y5J9KYGuHLjDHXGWNeNca0GmPKjDEXO2mKjDGPG2MajDE+Y8zNxpiMCPueZYy5zxhz1BjTbow5aIz5mTEmJxHHLiIikqrUUkNEREQkOn8DJgPXAD8EdjnL9wFZvay3ChgD/I5Qa47rgYeNMVcAdwL3Aw8D7wA+BxwHftC1sjHmbOA/QB1wB1ABLHa28wZjzJutter2IiIiw5Kx1iY7DyIiIiKu4LTEuBu4wFr7VNjy84EngY9Za+/ptqwSmG+trXeWLwK2AhZ4n7X2b2HbeQWYYq2dHLZsK6GgyeustY1hy99LKNBycp8iIiLDjbqfiIiIiMTXPV0BDQBr7TagAagMD2g4ngMmGWNyIdQ9BVgE/AnIMsaM6/rnpG0i1MJDRERkWFJQQ0RERCS+9kdYVgsc6GE5QL7zOs95/S5wotu/40AOMDFmORUREXEZjakhIiIiEl+Bfi4HMN1efwE83kPa2h6Wi4iIDHkKaoiIiIhEL9GDke11XgPW2n8neN8iIiIpT91PRERERKLnd17HJmh/m4HtwLXGmBnd3zTGpBtjEpUXERGRlKOWGiIiIiLRexkIAt8wxowhNFBnpLExYsJaa40xVxGa0nWbMeYuYAeQDcwELgO+BtwTrzyIiIikMrXUEBEREYmStfYw8HFgBHA7cD/wmTjvcwtQDPwReDdwC/BNYDmhYMYT8dy/iIhIKjPWJrprqIiIiIiIiIjI4KmlhoiIiIiIiIi4koIaIiIiIiIiIuJKCmqIiIiIiIiIiCspqCEiIiIiIiIirqSghoiIiIiIiIi4koIaIiIiIiIiIuJKCmqIiIiIiIiIiCspqCEiIiIiIiIirqSghoiIiIiIiIi4koIaIiIiIiIiIuJK/x9NFH3Ymtik/AAAAABJRU5ErkJggg==\n",
      "text/plain": [
       "<Figure size 1080x720 with 5 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "#fig_posterior = plt.figure(figsize= (15, 10))\n",
    "\n",
    "showlen = 100\n",
    "startT = 100 \n",
    "\n",
    "endT = startT + showlen\n",
    "showT = range(startT,endT)\n",
    "\n",
    "fig, [ax3, ax1, ax_loc, ax4, ax2] = plt.subplots(5, 1, figsize= (15, 10))\n",
    "\n",
    "ax3.imshow(np.tile(col1[showT], (5,1)), cmap = cmap_col5,vmin=0, vmax=Ncol)\n",
    "ax3.yaxis.set_label_coords(-0.1,0)\n",
    "ax3.set_ylabel('box 1 color', color = 'slateblue', rotation= 360, fontsize = 22)\n",
    "ax3.set_xticks([])\n",
    "ax3.set_yticks([])\n",
    "\n",
    "\n",
    "ax1.plot(lat[showT, 0], color = 'c',marker ='.', markersize = 7)\n",
    "ax1.yaxis.set_label_coords(-0.1,0.25)\n",
    "ax1.set_ylabel('Belief \\n about box 1', rotation= 360, fontsize = 22)\n",
    "ax1.set_yticks([0, nq-1])\n",
    "ax1.set_yticklabels(['0','1'])\n",
    "ax1.tick_params(axis='both', which='major', labelsize=18)\n",
    "ax1.spines['top'].set_visible(False)\n",
    "ax1.spines['right'].set_visible(False)\n",
    "ax1.spines['bottom'].set_visible(False)\n",
    "ax1.spines['left'].set_visible(False)\n",
    "ax1.set_xlim([0,showlen])\n",
    "\n",
    "ax_loc.plot((np.remainder(loc[showT]+1, 3) - 1 ) * 10, 'g-o', markersize = 3)\n",
    "box1_r = act[showT] // 4 * 7 * np.remainder(loc[showT]+1, 3) * np.insert(rew[showT][1:], -1, 0) * 1.0\n",
    "box2_r = act[showT] // 4 * 7 * (np.remainder(loc[showT]+1, 3) - 2) * np.insert(rew[showT][1:], -1, 0) * 1.0\n",
    "box1_n = act[showT] // 4 * 7 * np.remainder(loc[showT]+1, 3) * (1-np.insert(rew[showT][1:], -1, 0))* 1.0\n",
    "box2_n = act[showT] // 4 * 7 * ((np.remainder(loc[showT]+1, 3) - 2) * (1-np.insert(rew[showT][1:], -1, 0)))* 1.0\n",
    "box1_r[ box1_r==0 ] = np.nan\n",
    "box2_n[ box2_n==0 ] = np.nan\n",
    "box2_r[ box2_r==0 ] = np.nan\n",
    "box1_n[ box1_n==0 ] = np.nan\n",
    "ax_loc.plot(box2_r, '^', c = 'red', markersize = 10)\n",
    "ax_loc.plot(box1_n, 'v', c = 'blue', markersize = 10)\n",
    "ax_loc.plot(box2_n, '^', c = 'blue', markersize = 10)\n",
    "ax_loc.plot(box1_r, 'v', c = 'red', markersize = 10)\n",
    "\n",
    "ax_loc.set_xlim([0,showlen])\n",
    "ax_loc.spines['top'].set_visible(False)\n",
    "ax_loc.spines['right'].set_visible(False)\n",
    "ax_loc.spines['bottom'].set_visible(False)\n",
    "ax_loc.spines['left'].set_visible(False)\n",
    "ax_loc.set_yticks([])\n",
    "\n",
    "\n",
    "ax2.plot(lat[showT, 1], color = 'c',marker ='.', markersize = 10)\n",
    "ax2.set_xlabel('time', fontsize = 18)\n",
    "ax2.yaxis.set_label_coords(-0.1,0.25)\n",
    "ax2.set_ylabel('Belief \\n about box 2', rotation= 360, fontsize = 22)\n",
    "ax2.tick_params(axis='both', which='major', labelsize=18)\n",
    "ax2.set_yticks([0, nq-1])\n",
    "ax2.set_yticklabels(['0','1'])\n",
    "ax2.set_xlim([0,showlen])\n",
    "\n",
    "\n",
    "ax4.imshow(np.tile(col2[showT], (5,1)), cmap = cmap_col5,vmin=0, vmax=Ncol)\n",
    "ax4.yaxis.set_label_coords(-0.1,0)\n",
    "ax4.set_ylabel('box 2 color', color = 'slateblue', rotation= 360, fontsize = 22)\n",
    "ax4.set_xticks([])\n",
    "ax4.set_yticks([])\n",
    "\n",
    "\n",
    "plt.tight_layout()\n",
    "plt.show()\n",
    "\n",
    "\n",
    "#fig_posterior.savefig('posterior.pdf', format='pdf', bbox_inches = 'tight')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    " "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# load data for IRC"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [],
   "source": [
    "dataN_pkl_file1 = open(path + '/Results/'+ datestring_train + '_data' + datestring_data + \n",
    "                       '_agentNNdriven' + datestring_NNagent + '_twoboxCol.pkl', 'rb')\n",
    "dataN_pkl1 = pickle.load(dataN_pkl_file1)\n",
    "dataN_pkl_file1.close()\n",
    "\n",
    "idx = 0\n",
    "\n",
    "obs_IRC = dataN_pkl1['observations'][idx,:5000, :5].astype(int)   #NN agent behavior\n",
    "lat_IRC = dataN_pkl1['POMDP_agent'][idx, :5000, 1:]  #POMDP agent beliefs, for comparison\n",
    "\n",
    "obs = dataN_pkl1['observations'][idx,:, :5].astype(int)   #NN agent behavior\n",
    "lat = dataN_pkl1['POMDP_agent'][idx, :, 1:]  #POMDP agent beliefs, for comparison\n",
    "\n",
    "act = obs[:, 0]\n",
    "rew = obs[:, 1]\n",
    "loc = obs[:, 2]\n",
    "col1 = obs[:, 3]\n",
    "col2 = obs[:, 4]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "dict_keys(['observations', 'beliefs', 'trueStates', 'neural_response', 'allData', 'POMDP_agent', 'POMDP_agent_dist'])"
      ]
     },
     "execution_count": 8,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "dataN_pkl1.keys()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [],
   "source": [
    "dataN_pkl_file = open(path + '/Results/'+ datestring_train +'_data'+ datestring_data+\n",
    "                      '_agent' + datestring_NNagent + '_mainPara_twoboxCol.pkl', 'rb')\n",
    "dataN_pkl = pickle.load(dataN_pkl_file)\n",
    "dataN_pkl_file.close()\n",
    "nq, na, nr, nl, Numcol, discount, parametersAgent, parametersExp = dataN_pkl['NNtest_params']\n",
    "Numcol = parametersAgent[7]  # number of colors\n",
    "Ncol = Numcol - 1  # number value: 0 top Numcol-1\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[0.17, 0.1, 0.1, 0.03, 0.2, 0.15, 0.3, 5, 0.45, 0.55, 0.1]"
      ]
     },
     "execution_count": 10,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "parametersAgent"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# compare policies of NN and POMDP"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Text(0,0.5,'policy $\\\\pi(a_t|b_t)$')"
      ]
     },
     "execution_count": 11,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 720x288 with 3 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "fig_NNtraining, ax = plt.subplots(2,1,figsize=(10,4))\n",
    "\n",
    "ax[0].imshow(dataN_pkl1['POMDP_agent_dist'][idx,-100:, :na].T, vmin = 0, vmax = 1)\n",
    "ax[0].set_ylabel('action',  fontsize = 20, color = 'green')\n",
    "ax[0].set_title ( 'target POMDP', fontsize = 20, color = 'green')\n",
    "ax[0].set_xticks([])\n",
    "ax[0].set_yticks([])\n",
    "\n",
    "im = ax[1].imshow(dataN_pkl1['observations'][idx, -100:, -na:].T, vmin = 0, vmax = 1)\n",
    "ax[1].set_xlabel('time', fontsize = 16) \n",
    "ax[1].set_ylabel('action',  fontsize = 20, color = 'green')\n",
    "ax[1].set_title ( 'neural network', fontsize = 20, color = 'green')\n",
    "#ax[1].set_xticks([0, 40])\n",
    "ax[1].set_yticks([])\n",
    "\n",
    "cbar_ax = fig_NNtraining.add_axes([0.93, 0.2, 0.02, 0.56])\n",
    "cb = plt.colorbar(im, cax=cbar_ax)\n",
    "for j, lab in enumerate(['0','1']):\n",
    "    cb.ax.text(2.4, j , lab, ha='center', va='center', fontsize = 16)\n",
    "cb.ax.get_yaxis().set_ticks([])\n",
    "cb.ax.get_yaxis().labelpad = 10\n",
    "cb.ax.set_ylabel('policy $\\pi(a_t|b_t)$', rotation=90, fontsize = 16, color = 'green')\n",
    "\n",
    "\n",
    "#plt.outline.set_visible(False)\n",
    "\n",
    "#fig_NNtraining.savefig('NNtraining.pdf', format='pdf', bbox_inches = 'tight')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [],
   "source": [
    "POMDP_act_flatten = dataN_pkl1['POMDP_agent'][idx:idx+1, :, 0].reshape(-1,1)\n",
    "NN_act_flatten = dataN_pkl1['observations'][idx:idx+1, :, 0].reshape(-1,1)\n",
    "L = len(NN_act_flatten)\n",
    "\n",
    "POMDP_act_hist = np.zeros(5)\n",
    "NN_act_hist = np.zeros(5)\n",
    "\n",
    "for i in range(na):\n",
    "    POMDP_act_hist[i] = len(np.where(POMDP_act_flatten == i)[0])/L\n",
    "    NN_act_hist[i] = len(np.where(NN_act_flatten == i)[0])/L"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(0, 1)"
      ]
     },
     "execution_count": 13,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 720x360 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "N = 5\n",
    "ind = np.arange(N)  # the x locations for the groups\n",
    "width = 0.27       # the width of the bars\n",
    "\n",
    "\n",
    "fig, ax = plt.subplots(nrows=1, ncols=1, figsize = (10,5))\n",
    "label = ['POMDP', 'NN']\n",
    "\n",
    "ax.bar(ind, POMDP_act_hist, width, color='r')\n",
    "ax.bar(ind+width, NN_act_hist, width, color='b')\n",
    "ax.set_title('frequency of actions')\n",
    "ax.set_ylim([0,1])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "heading_collapsed": true
   },
   "source": [
    "# IRC"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "heading_collapsed": true,
    "hidden": true
   },
   "source": [
    "## IRC with line-search based method"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {
    "hidden": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      " [ 262.7643762  1137.44111331  165.9406098  -625.78595953 -493.16745481\n",
      " -776.68959875 -412.49988863    0.           59.72847248  528.00104095\n",
      " 3015.06647793]\n",
      "[0.17, 0.1, 0.1, 0.03, 0.2, 0.15, 0.3, 5, 0.45, 0.55, 0.1]\n"
     ]
    }
   ],
   "source": [
    "\"\"\"\n",
    "check gradient of the true parameter\n",
    "\"\"\"\n",
    "twoboxd = twoboxColMDP_der(discount, nq, nr, na, nl, parametersAgent)\n",
    "twoboxd1st = twoboxd.dloglikelihhod_dpara_sim(obs_IRC)\n",
    "print('\\n', twoboxd1st)\n",
    "print(parametersAgent)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {
    "hidden": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[0.2  0.12 0.05 0.07 0.2  0.15 0.3  5.   0.4  0.6  0.1 ]\n"
     ]
    }
   ],
   "source": [
    "pointIni = np.array(parametersAgent)\n",
    "pointIni[0:4] = np.array(parametersExp)[0:4]\n",
    "pointIni[8:10] = np.array(parametersExp)[4:]\n",
    "print(pointIni)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {
    "hidden": true
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[array([0.20607935, 0.11446118, 0.04468221, 0.06962941, 0.19649229,\n",
       "        0.15983267, 0.29715115, 5.        , 0.40875096, 0.60294378,\n",
       "        0.09095523]),\n",
       " array([0.19758934, 0.12189848, 0.05024994, 0.07594882, 0.20981327,\n",
       "        0.15018192, 0.3078951 , 5.        , 0.40181438, 0.60410095,\n",
       "        0.10760163])]"
      ]
     },
     "execution_count": 17,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "point = [np.array(pointIni) + (np.random.rand(len(pointIni)) * 2 - 1) * 0.01 , \n",
    "         np.array(pointIni) + (np.random.rand(len(pointIni)) * 2 - 1) * 0.01]\n",
    "point[0][7] = 5\n",
    "point[1][7] = 5\n",
    "point"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {
    "hidden": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[[-2.39807343e-01  4.46579404e-01  2.18286766e+00 -2.31854336e-01\n",
      "  -2.76345168e-01  2.94513572e-01  6.94352346e-03  5.76152419e-01\n",
      "  -3.53893914e-01  1.01185866e+00 -2.47960923e-01]\n",
      " [-5.17335109e-01 -4.52174111e-01  3.64270847e-01  5.02264425e-01\n",
      "   1.03827753e-01 -1.33053215e+00 -8.76084839e-01  6.36789421e-01\n",
      "   1.31607144e-03  1.73273092e-01  6.69807062e-01]]\n",
      "[0.1927932  0.13083006 0.0939073  0.07131173 0.20428636 0.15607219\n",
      " 0.30803397 5.         0.3947365  0.62433813 0.10264241]\n",
      "-5119.474376460369\n"
     ]
    }
   ],
   "source": [
    "randProjNum = 10\n",
    "for l in range(randProjNum):\n",
    "    transformer = random_projection.GaussianRandomProjection(n_components = 2)\n",
    "    transformer.fit_transform(point)\n",
    "    projectionMatRand = transformer.components_\n",
    "    print(projectionMatRand)\n",
    "    \n",
    "    # Contour of the likelihood\n",
    "    step1 = 0.02   # for u (1st principle component)\n",
    "    step2 = 0.02  # for v (2nd principle component)\n",
    "    N1 = 8\n",
    "    N2 = 8\n",
    "    uOffset = - step1 * N1 / 2 \n",
    "    vOffset = - step2 * N2 / 2\n",
    "\n",
    "    uValue = np.zeros(N1)\n",
    "    vValue = np.zeros(N2)\n",
    "    Qaux1 = np.zeros((N2, N1))    # Likelihood with ground truth latent\n",
    "    Qaux2 = np.zeros((N2, N1))    # Expected complete data likelihood\n",
    "    Qaux3 = np.zeros((N2, N1))    # Entropy of latent posterior\n",
    "    para_slice = []\n",
    "    LL_slice = []\n",
    "\n",
    "    for i in range(N1):\n",
    "        uValue[i] = step1 * (i) + uOffset\n",
    "        for j in range(N2):\n",
    "            vValue[j] = step2 * (j) + vOffset\n",
    "\n",
    "            para_slicePoints = point[-1] + uValue[i] * projectionMatRand[0] + vValue[j] * projectionMatRand[1]\n",
    "            para_slicePoints[7] = parametersAgent[7]\n",
    "            para_slice.append(para_slicePoints)\n",
    "            para = np.copy(para_slicePoints)       \n",
    "            #print(para)\n",
    "\n",
    "            twoboxCol = twoboxColMDP(discount, nq, nr, na, nl, para)\n",
    "            twoboxCol.setupMDP()\n",
    "            twoboxCol.solveMDP_sfm()\n",
    "            ThA = twoboxCol.ThA\n",
    "            policy = twoboxCol.softpolicy\n",
    "            pi = np.ones(nq * nq)/ nq /nq  # initialize the estimation of the belief state\n",
    "            Trans_hybrid_obs12 = twoboxCol.Trans_hybrid_obs12\n",
    "            Obs_emis_trans1 = twoboxCol.Obs_emis_trans1\n",
    "            Obs_emis_trans2 = twoboxCol.Obs_emis_trans2\n",
    "            twoboxColHMM = HMMtwoboxCol(ThA, policy, Trans_hybrid_obs12, Obs_emis_trans1, Obs_emis_trans2, pi, Ncol)\n",
    "\n",
    "            #Qaux1[j, i] = twoboxHMM.likelihood(lat, obs, ThA, policy)  #given latent state\n",
    "            Qaux2[j, i] = twoboxColHMM.computeQaux(obs_IRC, ThA, policy, Trans_hybrid_obs12, Obs_emis_trans1, Obs_emis_trans2) \n",
    "            Qaux3[j, i] = twoboxColHMM.latent_entr(obs_IRC) \n",
    "            \n",
    "            LL_slice.append(Qaux2[j, i] + Qaux3[j, i])\n",
    "\n",
    "    Loglikelihood = Qaux2 + Qaux3\n",
    "    Loglikelihood = np.nan_to_num(Loglikelihood, nan = -100000000000)\n",
    "    max_point = np.where(Loglikelihood == np.max(Loglikelihood))\n",
    "    new_point = para_slice[max_point[1][0] * N2 + max_point[0][0]]\n",
    "    point.append(new_point)\n",
    "    print(new_point)\n",
    "    print(Loglikelihood[max_point[0][0], max_point[1][0]])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {
    "hidden": true
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[array([0.20607935, 0.11446118, 0.04468221, 0.06962941, 0.19649229,\n",
       "        0.15983267, 0.29715115, 5.        , 0.40875096, 0.60294378,\n",
       "        0.09095523]),\n",
       " array([0.19758934, 0.12189848, 0.05024994, 0.07594882, 0.20981327,\n",
       "        0.15018192, 0.3078951 , 5.        , 0.40181438, 0.60410095,\n",
       "        0.10760163]),\n",
       " array([0.1927932 , 0.13083006, 0.0939073 , 0.07131173, 0.20428636,\n",
       "        0.15607219, 0.30803397, 5.        , 0.3947365 , 0.62433813,\n",
       "        0.10264241])]"
      ]
     },
     "execution_count": 19,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "point_all = point\n",
    "point_all"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "heading_collapsed": true,
    "hidden": true
   },
   "source": [
    "## IRC with gradient descent"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {
    "hidden": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[0.1927932  0.13083006 0.0939073  0.07131173 0.20428636 0.15607219\n",
      " 0.30803397 5.         0.3947365  0.62433813 0.10264241]\n",
      "\n",
      " [  -39.8885004    597.06151842  1999.64561943 -1748.22522513\n",
      "   821.10385756   299.30147048   290.37759668     0.\n",
      "   877.46587178  -272.96400913  2092.77682825]\n",
      "-5107.204829656552\n",
      "[0.19275331 0.13142713 0.09590694 0.0695635  0.20510747 0.15637149\n",
      " 0.30832435 5.         0.39561397 0.62406516 0.10473519]\n",
      "\n",
      " [  -31.73365258   506.7463062   1780.28061873 -1553.19257556\n",
      "   842.22824353   450.3483533    346.24784348     0.\n",
      "   863.10211645  -318.81224641  1317.92086268]\n",
      "-5098.336134453799\n",
      "[0.19272158 0.13193387 0.09768722 0.06801031 0.2059497  0.15682184\n",
      " 0.3086706  5.         0.39647707 0.62374635 0.10605311]\n",
      "\n",
      " [  -11.11611073   448.55647411  1558.1912694  -1428.96565558\n",
      "   780.5281748    465.60721239   340.71902246     0.\n",
      "   861.09812219  -335.60136762   936.68726549]\n",
      "-5091.379149949262\n",
      "[0.19271046 0.13238243 0.09924542 0.06658135 0.20673022 0.15728745\n",
      " 0.30901132 5.         0.39733817 0.62341075 0.1069898 ]\n",
      "\n",
      " [   12.28728252   404.59869069  1354.02290744 -1330.74386758\n",
      "   698.58563256   435.35271925   316.23120412     0.\n",
      "   859.63035417  -341.80284456   722.66339794]\n",
      "-5085.735612046613\n",
      "[0.19272275 0.13278703 0.10059944 0.0652506  0.20742881 0.1577228\n",
      " 0.30932755 5.         0.3981978  0.62306894 0.10771246]\n",
      "\n",
      " [   34.6992274    368.13456973  1173.2957737  -1243.2613139\n",
      "   617.87345021   392.18634538   287.8946616      0.\n",
      "   854.98802218  -343.92593352   588.7047746 ]\n",
      "-5081.076956618263\n",
      "[0.19275745 0.13315516 0.10177273 0.06400734 0.20804668 0.15811499\n",
      " 0.30961544 5.         0.39905279 0.62272502 0.10830116]\n",
      "\n",
      " [   55.00832776   336.35187538  1015.93038289 -1160.43623166\n",
      "   545.42366433   348.37990916   261.16102617     0.\n",
      "   846.22686427  -344.16969538   496.83315319]\n",
      "-5077.182549427592\n",
      "[0.19281245 0.13349151 0.10278866 0.06284691 0.20859211 0.15846337\n",
      " 0.3098766  5.         0.39989901 0.62238085 0.108798  ]\n",
      "\n",
      " [   72.92797454   307.94091072   879.8582985  -1080.40493433\n",
      "   482.84652985   308.40984982   237.77786191     0.\n",
      "   833.79322405  -343.49701582   428.81902118]\n",
      "-5073.897038366037\n",
      "[0.19288538 0.13379945 0.10366852 0.0617665  0.20907495 0.15877178\n",
      " 0.31011438 5.         0.40073281 0.62203735 0.10922682]\n",
      "\n",
      " [   88.59278387   282.32792647   762.48252844 -1003.27302152\n",
      "   429.490643     273.31956153   217.89130278     0.\n",
      "   818.6113082   -342.31330301   375.6574215 ]\n",
      "-5071.095657734641\n",
      "[0.19297397 0.13408178 0.10443101 0.06076323 0.20950444 0.15904509\n",
      " 0.31033227 5.         0.40155142 0.62169504 0.10960247]\n",
      "\n",
      " [ 102.12209054  258.96156876  660.93710757 -929.41122966  384.34959879\n",
      "  243.2311494   201.28531014    0.          801.78785538 -341.08069667\n",
      "  332.16997053]\n",
      "-5068.68781693407\n",
      "[0.1930761  0.13434074 0.10509194 0.05983382 0.20988879 0.15928833\n",
      " 0.31053356 5.         0.4023532  0.62135396 0.10993464]\n",
      "\n",
      " [ 113.8250409   237.69155723  572.87274285 -864.6473616   345.97530794\n",
      "  217.41036926  187.38290782    0.          787.19516114 -341.85589266\n",
      "  295.82447951]\n",
      "-5066.580574333394\n",
      "[0.19318992 0.13457844 0.10566482 0.05896917 0.21023477 0.15950574\n",
      " 0.31072094 5.         0.4031404  0.6210121  0.11023047]\n",
      "\n",
      " [ 123.95373597  218.13154126  496.15541047 -811.08716313  313.50180416\n",
      "  195.42477912  175.88548963    0.          776.3918502  -345.54711601\n",
      "  264.69704335]\n",
      "-5064.7066172327395\n",
      "[0.19331388 0.13479657 0.10616097 0.05815808 0.21054827 0.15970116\n",
      " 0.31089683 5.         0.40391679 0.62066655 0.11049517]\n",
      "\n",
      " [ 132.75244146  199.90512283  428.94875423 -760.95101213  286.07972945\n",
      "  176.7347685   166.45632422    0.          764.98974704 -349.25704495\n",
      "  237.69349536]\n",
      "-5063.025280357639\n",
      "[0.19344663 0.13499647 0.10658992 0.05739713 0.21083435 0.1598779\n",
      " 0.31106328 5.         0.40468178 0.6203173  0.11073286]\n",
      "\n",
      " [ 140.39393227  182.89050968  369.75332023 -714.02275898  262.73301137\n",
      "  160.67225624  158.6540875     0.          753.08937267 -352.84247133\n",
      "  214.25253635]\n",
      "-5061.504799721862\n",
      "[0.19358702 0.13517936 0.10695967 0.05668311 0.21109708 0.16003857\n",
      " 0.31122194 5.         0.40543487 0.61996446 0.11094711]\n",
      "\n",
      " [ 147.04081059  166.97484534  317.35240827 -670.04775156  242.69909409\n",
      "  146.73004443  152.13987172    0.          740.73023097 -356.15198249\n",
      "  193.87392695]\n",
      "-5060.120192623236\n",
      "[0.19373406 0.13534634 0.10727703 0.05601306 0.21133978 0.1601853\n",
      " 0.31137408 5.         0.4061756  0.6196083  0.11114099]\n",
      "\n",
      " [ 152.82845561  152.05700129  270.7523945  -628.80210953  225.37615769\n",
      "  134.5185492   146.65033814    0.          727.94297648 -359.074892\n",
      "  176.1202011 ]\n",
      "-5058.851626141281\n",
      "[0.19388689 0.13549839 0.10754778 0.05538426 0.21156516 0.16031982\n",
      " 0.31152073 5.         0.40690354 0.61924923 0.11131711]\n",
      "\n",
      " [ 157.87147277  138.04682021  229.1347655  -590.06030187  210.28590254\n",
      "  123.73522713  141.97957807    0.          714.75238201 -361.53093879\n",
      "  160.61284714]\n",
      "-5057.683217773292\n",
      "[0.19404476 0.13563644 0.10777691 0.0547942  0.21177544 0.16044355\n",
      " 0.31166271 5.         0.4076183  0.6188877  0.11147772]\n",
      "\n",
      " [ 162.26758908  124.79067482  191.8200287  -553.6018316   197.0438725\n",
      "  114.14192431  137.96472103    0.          701.24844387 -363.56324381\n",
      "  147.02867611]\n",
      "-5056.602004862194\n",
      "[0.19420703 0.13576123 0.10796873 0.0542406  0.21197249 0.16055769\n",
      " 0.31180067 5.         0.40831954 0.61852413 0.11162475]\n",
      "\n",
      " [ 166.10120565  112.22621924  158.23889953 -519.21976046  185.3329677\n",
      "  105.54643541  134.47241872    0.          687.45122735 -365.12285624\n",
      "  135.09948712]\n",
      "-5055.5975290069355\n",
      "[0.19437313 0.13587346 0.10812697 0.05372138 0.21215782 0.16066324\n",
      " 0.31193514 5.         0.409007   0.61815901 0.11175985]\n",
      "\n",
      " [ 169.44512663  100.37333232  127.91543031 -486.72205958  174.89899255\n",
      "   97.79712855  131.39840394    0.          673.32765593 -366.08242044\n",
      "  124.58874062]\n",
      "-5054.661296952777\n",
      "[0.19454258 0.13597383 0.10825489 0.05323465 0.21233272 0.16076104\n",
      " 0.31206654 5.         0.40968032 0.61779293 0.11188444]\n",
      "\n",
      " [ 172.36201438   89.18341791  100.45073395 -455.92978859  165.53751266\n",
      "   90.77472987  128.66079305    0.          658.92079147 -366.43795647\n",
      "  115.29013682]\n",
      "-5053.78627095142\n",
      "[0.19471494 0.13606302 0.10835534 0.05277872 0.21249826 0.16085181\n",
      " 0.3121952  5.         0.41033924 0.61742649 0.11199973]\n",
      "\n",
      " [ 174.90592563   78.61633185   75.5080855  -426.6869837   157.0804834\n",
      "   84.38170607  126.19303288    0.          644.28352562 -366.19939056\n",
      "  107.02896998]\n",
      "-5052.966538631307\n",
      "[0.19488984 0.13614163 0.10843085 0.05235204 0.21265534 0.16093619\n",
      " 0.31232139 5.         0.41098353 0.61706029 0.11210675]\n",
      "\n",
      " [ 177.12410499   68.64001807   52.80255664 -398.8636754   149.3890577\n",
      "   78.53810894  123.94136502    0.          629.4794739  -365.38892164\n",
      "   99.65642312]\n",
      "-5052.197051291443\n",
      "[0.19506697 0.13621027 0.10848365 0.05195317 0.21280473 0.16101473\n",
      " 0.31244534 5.         0.41161301 0.6166949  0.11220641]\n",
      "\n",
      " [ 179.05654749   59.22905439   32.09333499 -372.35698801  142.34874743\n",
      "   73.17837117  121.86260028    0.          614.58160963 -364.03988452\n",
      "   93.04504601]\n",
      "-5051.473347812623\n",
      "[0.19524603 0.1362695  0.10851574 0.05158082 0.21294708 0.16108791\n",
      " 0.3125672  5.         0.41222759 0.61633086 0.11229946]\n",
      "\n",
      " [ 180.74026957   50.36358198   12.94354934 -347.09182546  135.87383273\n",
      "   68.25531636  119.92689178    0.          599.9274581  -362.3690327\n",
      "   87.07903271]\n",
      "-5050.79107427873\n",
      "[0.19542677 0.13631986 0.10852868 0.05123372 0.21308295 0.16115617\n",
      " 0.31268713 5.         0.41282752 0.61596849 0.11238653]\n",
      "\n",
      " [ 182.19367212   42.02842683   -4.68935468 -323.01967985  129.92352185\n",
      "   63.75390421  118.12826506    0.          585.5045838  -360.37322207\n",
      "   81.63600887]\n",
      "-5050.146522646753\n",
      "[0.19560896 0.13636189 0.108524   0.0509107  0.21321287 0.16121992\n",
      " 0.31280525 5.         0.41341302 0.61560812 0.11246817]\n",
      "\n",
      " [ 183.4389459    34.21264773  -20.83937306 -300.11196213  124.43110791\n",
      "   59.63027434  116.44309689    0.          571.28122262 -358.02698767\n",
      "   76.65190151]\n",
      "-5049.536783419468\n",
      "[0.1957924  0.13639611 0.10850316 0.05061059 0.2133373  0.16127955\n",
      " 0.3129217  5.         0.4139843  0.61525009 0.11254482]\n",
      "\n",
      " [ 184.49085746   26.92393355  -35.51993177 -278.38117785  119.40890181\n",
      "   55.90510136  114.89301869    0.          557.35063779 -355.40119552\n",
      "   72.0385533 ]\n",
      "-5048.958048531271\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[0.19597689 0.13642303 0.10846764 0.05033221 0.21345671 0.16133545\n",
      " 0.31303659 5.         0.41454165 0.61489469 0.11261686]\n",
      "\n",
      " [ 185.38557379   20.1164703   -49.09573818 -257.75897939  114.64796262\n",
      "   52.40533301  113.36655916    0.          543.7694017  -352.5127886\n",
      "   67.85306141]\n",
      "-5048.407962090765\n",
      "[0.19616227 0.13644315 0.10841854 0.05007445 0.21357136 0.16138786\n",
      " 0.31314996 5.         0.41508542 0.61454218 0.11268471]\n",
      "\n",
      " [ 186.12605254   13.79821333  -61.51895741 -238.27786936  110.20427337\n",
      "   49.19152252  111.9082217     0.          530.61593644 -349.43858827\n",
      "   63.96354092]\n",
      "-5047.883934941857\n",
      "[0.1963484  0.13645694 0.10835702 0.04983617 0.21368157 0.16143705\n",
      " 0.31326186 5.         0.41561604 0.61419274 0.11274868]\n",
      "\n",
      " [ 186.725204      7.95846063  -72.87449716 -219.93769951  106.0432769\n",
      "   46.23688255  110.50725243    0.          517.94885894 -346.22869225\n",
      "   60.34008948]\n",
      "-5047.383569762127\n",
      "[0.19653513 0.1364649  0.10828415 0.04961624 0.21378761 0.16148329\n",
      " 0.31337237 5.         0.41613399 0.61384651 0.11280902]\n",
      "\n",
      " [ 187.19370564    2.58541877  -83.23640941 -202.73483005  102.13830319\n",
      "   43.5195434   109.15647817    0.          505.81452251 -342.92909913\n",
      "   56.95464279]\n",
      "-5046.904654921058\n",
      "[0.19672232 0.13646749 0.10820091 0.0494135  0.21388975 0.16152681\n",
      " 0.31348153 5.         0.4166398  0.61350358 0.11286597]\n",
      "\n",
      " [ 187.54018038   -2.33479022  -92.66995312 -186.66040523   98.46738885\n",
      "   41.02076764  107.85077666    0.          494.24661302 -339.5806234\n",
      "   53.78261994]\n",
      "-5046.445161275605\n",
      "[0.19690986 0.13646515 0.10810824 0.04922684 0.21398821 0.16156783\n",
      " 0.31358938 5.         0.41713405 0.613164   0.11291976]\n",
      "\n",
      " [ 187.77192963   -6.81727306 -101.23376978 -171.69889179   95.01293971\n",
      "   38.72417074  106.58690826    0.          483.26634442 -336.21808507\n",
      "   50.80315077]\n",
      "-5046.003239319331\n",
      "[0.19709763 0.13645834 0.10800701 0.04905514 0.21408323 0.16160655\n",
      " 0.31369597 5.         0.41761731 0.61282778 0.11297056]\n",
      "\n",
      " [ 187.89487694  -10.87854798 -108.9807447  -157.82770515   91.76029198\n",
      "   36.61514074  105.36255832    0.          472.88269343 -332.86966936\n",
      "   47.99916405]\n",
      "-5045.577216020585\n",
      "[0.19728553 0.13644746 0.10789803 0.04889732 0.21417499 0.16164317\n",
      " 0.31380133 5.         0.4180902  0.61249491 0.11301856]\n",
      "\n",
      " [ 187.91429466  -14.53623736 -115.95903379 -145.01722671   88.69707926\n",
      "   34.67996794  104.17619615    0.          463.09378376 -329.55747974\n",
      "   45.3568091 ]\n",
      "-5045.165589850479\n",
      "[0.19747344 0.13643292 0.10778207 0.0487523  0.21426368 0.16167785\n",
      " 0.3139055  5.         0.41855329 0.61216536 0.11306391]\n",
      "\n",
      " [ 187.83449195  -17.80877119 -122.21313864 -133.23210078   85.81242855\n",
      "   32.90606674  103.02696955    0.          453.88772378 -326.29776888\n",
      "   42.86466719]\n",
      "-5044.767024334856\n",
      "[0.19766128 0.13641511 0.10765985 0.04861907 0.2143495  0.16171075\n",
      " 0.31400853 5.         0.41900718 0.61183906 0.11310678]\n",
      "\n",
      " [ 187.65907043  -20.71536801 -127.78402743 -122.43150923   83.09679833\n",
      "   31.28179378  101.91390811    0.          445.24435361 -323.10101233\n",
      "   40.51366705]\n",
      "-5044.380547811584\n",
      "[0.19784893 0.1363944  0.10753207 0.04849663 0.21443259 0.16174204\n",
      " 0.31411045 5.         0.41945242 0.61151596 0.11314729]\n",
      "\n",
      " [ 187.38170638  -23.30757616 -132.62353423 -112.45690621   80.67921135\n",
      "   29.90468602  100.9154621     0.          437.12704337 -320.01834734\n",
      "   38.24284886]\n",
      "-5044.004702725367\n",
      "[0.19803632 0.13637109 0.10739945 0.04838418 0.21451327 0.16177194\n",
      " 0.31421136 5.         0.41988955 0.61119594 0.11318554]\n",
      "\n",
      " [ 187.03058231  -25.53207287 -136.96564656 -103.50398497   78.241962\n",
      "   28.5118067    99.85384168    0.          429.52727381 -316.94751306\n",
      "   36.21220367]\n",
      "-5043.6388122058115\n",
      "[0.19822335 0.13634556 0.10726248 0.04828067 0.21459151 0.16180045\n",
      " 0.31431121 5.         0.42031908 0.61087899 0.11322175]\n",
      "\n",
      " [ 186.59054967  -27.45134498 -140.72100566  -95.38879749   75.96326213\n",
      "   27.25416925   98.83550592    0.          422.39404502 -313.94860707\n",
      "   34.27177671]\n",
      "-5043.2821089355375\n",
      "[0.19840994 0.13631811 0.10712176 0.04818528 0.21466748 0.16182771\n",
      " 0.31441005 5.         0.42074147 0.61056504 0.11325602]\n",
      "\n",
      " [ 186.06544836  -29.08298393 -143.92534194  -88.06224469   73.82962394\n",
      "   26.11472337   97.85594648    0.          415.68934193 -311.01475815\n",
      "   32.43327319]\n",
      "-5042.933940790604\n",
      "[0.198596   0.13628902 0.10697783 0.04809722 0.21474131 0.16185382\n",
      " 0.31450791 5.         0.42115716 0.61025403 0.11328845]\n",
      "\n",
      " [ 185.45822877  -30.44481127 -146.61061687  -81.47448898   71.83057915\n",
      "   25.08070247   96.91221931    0.          409.37300309 -308.1375171\n",
      "   30.7006203 ]\n",
      "-5042.5937599516665\n",
      "[0.19878146 0.13625858 0.10683122 0.04801575 0.21481314 0.1618789\n",
      " 0.31460482 5.         0.42156653 0.60994589 0.11331915]\n",
      "\n",
      " [ 184.77212507  -31.5539919  -148.80590516  -75.57541358   69.95774766\n",
      "   24.14221944   96.00272369    0.          403.40493433 -305.30738059\n",
      "   29.07416456]\n",
      "-5042.261111435752\n",
      "[0.19896623 0.13622702 0.10668242 0.04794017 0.2148831  0.16190304\n",
      " 0.31470082 5.         0.42196994 0.60964058 0.11334823]\n",
      "\n",
      " [ 184.00978479  -32.42781622 -150.53868538  -70.31616133   68.20316412\n",
      "   23.29041672   95.12565521    0.          397.74505967 -302.51425142\n",
      "   27.5522525 ]\n",
      "-5041.935622540898\n",
      "[0.19915024 0.1361946  0.10653188 0.04786986 0.2149513  0.16192633\n",
      " 0.31479595 5.         0.42236768 0.60933807 0.11337578]\n",
      "\n",
      " [ 183.17430749  -33.08246596 -151.83472442  -65.64894011   66.5598327\n",
      "   22.51764636   94.27970508    0.          392.35506392 -299.747594\n",
      "   26.13247125]\n",
      "-5041.616992195481\n",
      "[0.19933342 0.13616151 0.10638004 0.04780421 0.21501786 0.16194885\n",
      " 0.31489023 5.         0.42276004 0.60903832 0.11340191]\n",
      "\n",
      " [ 182.26863267  -33.53383181 -152.71910888  -61.52781862   65.02049564\n",
      "   21.81681339   93.46326806    0.          387.19795521 -296.9969064\n",
      "   24.81154843]\n",
      "-5041.30497771031\n",
      "[0.19951569 0.13612798 0.10622733 0.04774268 0.21508288 0.16197067\n",
      " 0.31498369 5.         0.42314724 0.60874133 0.11342672]\n",
      "\n",
      " [ 181.29804539  -33.79695499 -153.26113953  -57.90870819   63.57904749\n",
      "   21.18176053   92.67530822    0.          382.2697654  -294.26552192\n",
      "   23.58580878]\n",
      "-5040.999317978839\n",
      "[0.19969698 0.13609418 0.10607406 0.04768477 0.21514646 0.16199185\n",
      " 0.31507637 5.         0.42352951 0.60844706 0.11345031]\n",
      "\n",
      " [ 180.26607904  -33.88634013 -153.52764331  -54.75011039   62.23981291\n",
      "   20.61583291   91.92051948    0.          377.57450173 -291.56263372\n",
      "   22.44223651]\n",
      "-5040.699804433321\n",
      "[0.19987725 0.1360603  0.10592054 0.04763002 0.2152087  0.16201247\n",
      " 0.31516829 5.         0.42390708 0.6081555  0.11347275]\n",
      "\n",
      " [ 179.17080822  -33.81580063 -153.43687119  -52.01281147   61.00119936\n",
      "   20.11636479   91.20008872    0.          373.0195167  -288.85295887\n",
      "   21.37582123]\n",
      "-5040.406330855497\n",
      "[0.20005642 0.13602648 0.1057671  0.04757801 0.2152697  0.16203258\n",
      " 0.31525949 5.         0.4242801  0.60786664 0.11349413]\n",
      "\n",
      " [ 178.01593458  -33.59795664 -153.01429539  -49.66035067   59.85215921\n",
      "   19.67316166   90.50942026    0.          368.57879968 -286.12893129\n",
      "   20.39151968]\n",
      "-5040.1188204444425\n",
      "[0.20023444 0.13599288 0.10561409 0.04752835 0.21532955 0.16205226\n",
      " 0.31535    5.         0.42464868 0.60758052 0.11351452]\n",
      "\n",
      " [ 176.81047393  -33.24490626 -152.2841203   -47.65753965   58.7834038\n",
      "   19.27834091   89.84513897    0.          364.22897301 -283.37942604\n",
      "   19.48982208]\n",
      "-5039.837358483626\n",
      "[0.20041125 0.13595964 0.1054618  0.04748069 0.21538834 0.16207153\n",
      " 0.31543984 5.         0.42501291 0.60729714 0.11353401]\n",
      "\n",
      " [ 175.54661918  -32.82606758 -151.20189801  -45.79665529   57.9426287\n",
      "   19.04197688   89.29208434    0.          359.92858557 -280.66944924\n",
      "   18.61524925]\n",
      "-5039.561608023346\n",
      "[0.20058679 0.13592681 0.1053106  0.04743489 0.21544628 0.16209058\n",
      " 0.31552913 5.         0.42537284 0.60701647 0.11355262]\n",
      "\n",
      " [ 174.25423903  -32.2261867  -149.95348556  -44.42426871   56.96944936\n",
      "   18.68465552   88.64828851    0.          355.71180888 -277.84559188\n",
      "   17.93499905]\n",
      "-5039.291702285467\n",
      "[0.20076105 0.13589459 0.10516065 0.04739047 0.21550325 0.16210926\n",
      " 0.31561778 5.         0.42572855 0.60673862 0.11357056]\n",
      "\n",
      " [ 172.9205469   -31.53032139 -148.45345049  -43.30192041   56.069739\n",
      "   18.37749551   88.03252058    0.          351.53548743 -274.99353655\n",
      "   17.29223914]\n",
      "-5039.027614241778\n",
      "[0.20093397 0.13586306 0.10501219 0.04734717 0.21555932 0.16212764\n",
      " 0.31570581 5.         0.42608008 0.60646363 0.11358785]\n",
      "\n",
      " [ 171.55203113  -30.74636425 -146.72962516  -42.61612776   55.22927495\n",
      "   18.10683261   87.43748003    0.          347.4921532  -272.18216292\n",
      "   16.700792  ]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "-5038.769041912968\n",
      "[0.20110552 0.13583231 0.10486546 0.04730455 0.21561455 0.16214575\n",
      " 0.31579325 5.         0.42642758 0.60619145 0.11360455]\n",
      "\n",
      " [ 170.15544017  -29.8874711  -144.81022481  -42.99998091   54.44685212\n",
      "   17.86913708   86.86362889    0.          343.89207394 -269.64576682\n",
      "   16.16264126]\n",
      "-5038.5156030915305\n",
      "[0.20127568 0.13580242 0.10472065 0.04726155 0.21566899 0.16216361\n",
      " 0.31588011 5.         0.42677147 0.6059218  0.11362072]\n",
      "\n",
      " [ 168.73818822  -28.99920219 -142.73077159  -43.43581077   53.73627937\n",
      "   17.67007052   86.31965284    0.          340.2653374  -267.09323265\n",
      "   15.67262734]\n",
      "-5038.267302685813\n",
      "[0.20144441 0.13577342 0.10457792 0.04721812 0.21572273 0.16218128\n",
      " 0.31596643 5.         0.42711173 0.60565471 0.11363639]\n",
      "\n",
      " [ 167.30605785  -28.08457521 -140.51489234  -43.91653783   53.0815114\n",
      "   17.49718831   85.797461      0.          336.61292855 -264.5217146\n",
      "   15.23924129]\n",
      "-5038.024138421917\n",
      "[0.20161172 0.13574534 0.10443741 0.0471742  0.21577581 0.16219878\n",
      " 0.31605223 5.         0.42744835 0.60539018 0.11365163]\n",
      "\n",
      " [ 165.86432594  -27.1473491  -138.18327534  -44.4345651    52.47052195\n",
      "   17.34247689   85.29124352    0.          332.93686556 -261.93074609\n",
      "   14.86287329]\n",
      "-5037.786099546379\n",
      "[0.20177758 0.13571819 0.10429922 0.04712976 0.21582828 0.16221612\n",
      " 0.31613752 5.         0.42778128 0.60512825 0.11366649]\n",
      "\n",
      " [ 164.41784555  -26.19182033 -135.75503544  -44.98183716   51.89398416\n",
      "   17.20053842   84.79651569    0.          329.24044899 -259.32137305\n",
      "   14.54062294]\n",
      "-5037.5531657991\n",
      "[0.201942   0.135692   0.10416347 0.04708478 0.21588018 0.16223333\n",
      " 0.31622232 5.         0.42811052 0.60486893 0.11368103]\n",
      "\n",
      " [ 162.97111324  -25.22242903 -133.24757492  -45.5507161    51.34420371\n",
      "   17.06735839   84.30990965    0.          325.52759512 -256.69611932\n",
      "   14.26795097]\n",
      "-5037.325306963324\n",
      "[0.20210497 0.13566678 0.10403022 0.04703923 0.21593152 0.16225039\n",
      " 0.31630663 5.         0.42843605 0.60461224 0.1136953 ]\n",
      "\n",
      " [ 161.52837179  -24.24339255 -130.67749478  -46.13264719   50.81464678\n",
      "   16.93975901   83.82862416    0.          321.80293874 -254.0570531\n",
      "   14.03954866]\n",
      "-5037.1024830620845\n",
      "[0.2022665  0.13564253 0.10389954 0.0469931  0.21598233 0.16226733\n",
      " 0.31639046 5.         0.42875785 0.60435818 0.11370934]\n",
      "\n",
      " [ 160.09336832  -23.25939295 -128.06042923  -46.72121759   50.29968997\n",
      "   16.81528556   83.35004986    0.          318.07125833 -251.40858179\n",
      "   13.84976349]\n",
      "-5036.884644173514\n",
      "[0.2024266  0.13561927 0.10377148 0.04694638 0.21603263 0.16228415\n",
      " 0.31647381 5.         0.42907593 0.60410677 0.11372319]\n",
      "\n",
      " [ 158.66974172  -22.27465848 -125.41102296  -47.30933506   49.79445566\n",
      "   16.69188168   82.87212859    0.          314.33798995 -248.75437612\n",
      "   13.69360234]\n",
      "-5036.671731053559\n",
      "[0.20258527 0.135597   0.10364607 0.04689907 0.21608243 0.16230084\n",
      " 0.31655668 5.         0.42939026 0.60385802 0.11373688]\n",
      "\n",
      " [ 157.26068705  -21.29347558 -122.7429475   -47.89055492   49.29507577\n",
      "   16.56787572   82.39291947    0.          310.6082886  -246.09851334\n",
      "   13.56580015]\n",
      "-5036.463675883277\n",
      "[0.20274253 0.13557571 0.10352333 0.04685118 0.21613172 0.16231741\n",
      " 0.31663907 5.         0.42970087 0.60361192 0.11375045]\n",
      "\n",
      " [ 155.8691738   -20.3196505  -120.0684228   -48.45930198   48.79811058\n",
      "   16.44203621   81.91121924    0.          306.8875094  -243.44509893\n",
      "   13.46166755]\n",
      "-5036.260377120611\n",
      "[0.2028984  0.13555539 0.10340326 0.04680272 0.21618052 0.16233385\n",
      " 0.31672098 5.         0.43000776 0.60336847 0.11376391]\n",
      "\n",
      " [ 154.49149396  -19.42971721 -117.34402597  -48.81026143   48.46062438\n",
      "   16.42911229   81.51488782    0.          303.15608637 -240.8855089\n",
      "   13.32287866]\n",
      "-5036.061776136058\n",
      "[0.20305289 0.13553596 0.10328592 0.04675391 0.21622898 0.16235028\n",
      " 0.3168025  5.         0.43031092 0.60312759 0.11377723]\n",
      "\n",
      " [ 153.14929442  -18.46672149 -114.72023925  -49.37113135   47.91761057\n",
      "   16.25501864   80.999901      0.          299.47493931 -238.23216452\n",
      "   13.31846215]\n",
      "-5035.867787030104\n",
      "[0.20320604 0.13551749 0.1031712  0.04670454 0.2162769  0.16236653\n",
      " 0.3168835  5.         0.43061039 0.60288936 0.11379055]\n",
      "\n",
      " [ 151.82798597  -17.52559001 -112.10750017  -49.89814624   47.38743701\n",
      "   16.09632909   80.48997825    0.          295.81476338 -235.59833789\n",
      "   13.28871167]\n",
      "-5035.678316878812\n",
      "[0.20335786 0.13549996 0.10305909 0.04665464 0.21632429 0.16238263\n",
      " 0.31696399 5.         0.43090621 0.60265376 0.11380384]\n",
      "\n",
      " [ 150.53084644  -16.60744783 -109.52030789  -50.39048574   46.86066313\n",
      "   15.94322293   79.97946705    0.          292.17998326 -232.98459655\n",
      "   13.24936875]\n",
      "-5035.493269124749\n",
      "[0.20350839 0.13548336 0.10294957 0.04660425 0.21637115 0.16239857\n",
      " 0.31704397 5.         0.43119839 0.60242077 0.11381709]\n",
      "\n",
      " [ 149.25975847  -15.71388839 -106.96895333  -50.84650911   46.3324468\n",
      "   15.790627     79.46552432    0.          288.57431334 -230.39271673\n",
      "   13.20705269]\n",
      "-5035.3125445059195\n",
      "[0.20365765 0.13546764 0.1028426  0.0465534  0.21641748 0.16241436\n",
      " 0.31712343 5.         0.43148696 0.60219038 0.1138303 ]\n",
      "\n",
      " [ 148.01614816  -14.84653512 -104.46113538  -51.26490032   45.79992219\n",
      "   15.63613387   78.94652572    0.          285.00056942 -227.82422275\n",
      "   13.16374073]\n",
      "-5035.136034842073\n",
      "[0.20380567 0.1354528  0.10273814 0.04650214 0.21646328 0.16243\n",
      " 0.31720238 5.         0.43177196 0.60196256 0.11384346]\n",
      "\n",
      " [ 146.80089316  -14.00391284 -102.00264842  -51.76352352   45.26194698\n",
      "   15.47849388   78.42189279    0.          281.50133494 -225.29990656\n",
      "   13.11979577]\n",
      "-5034.9635935431115\n",
      "[0.20395247 0.13543879 0.10263614 0.04645037 0.21650854 0.16244548\n",
      " 0.3172808  5.         0.43205346 0.60173726 0.11385658]\n",
      "\n",
      " [ 145.61467015  -13.19329294  -99.59927866  -52.26726444   44.72198816\n",
      "   15.31987164   77.89356096    0.          278.05419722 -222.81430211\n",
      "   13.07292223]\n",
      "-5034.7951085471295\n",
      "[0.20409809 0.1354256  0.10253654 0.04639811 0.21655326 0.1624608\n",
      " 0.3173587  5.         0.43233152 0.60151444 0.11386965]\n",
      "\n",
      " [ 144.45780016  -12.41859445  -97.25526616  -52.71078589   44.18054596\n",
      "   15.16041357   77.36187217    0.          274.63909737 -220.35926213\n",
      "   13.02277815]\n",
      "-5034.630481743713\n",
      "[0.20424254 0.13541318 0.10243928 0.0463454  0.21659744 0.16247596\n",
      " 0.31743606 5.         0.43260615 0.60129408 0.11388267]\n",
      "\n",
      " [ 143.33048512  -11.67985192  -94.97348674  -53.0960051    43.63640801\n",
      "   14.9989055    76.82621526    0.          271.25749966 -217.93463657\n",
      "   12.96967457]\n",
      "-5034.469474178082\n",
      "[0.20438587 0.1354015  0.10234431 0.0462923  0.21664108 0.16249096\n",
      " 0.31751288 5.         0.43287741 0.60107615 0.11389564]\n",
      "\n",
      " [ 142.22823756  -11.06181298  -92.71440467  -53.21063782   43.24850761\n",
      "   14.94907974   76.3744938     0.          267.88682499 -215.63793598\n",
      "   12.85438975]\n",
      "-5034.312246605526\n",
      "[0.2045281  0.13539044 0.10225159 0.04623909 0.21668433 0.16250591\n",
      " 0.31758926 5.         0.4331453  0.60086051 0.1139085 ]\n",
      "\n",
      " [ 141.16587772  -10.37920265  -90.59167223  -53.51901927   42.65529878\n",
      "   14.74125747   75.80558486    0.          264.58077355 -213.25512898\n",
      "   12.85925464]\n",
      "-5034.158594072158\n",
      "[0.20466927 0.13538006 0.102161   0.04618557 0.21672698 0.16252065\n",
      " 0.31766506 5.         0.43340988 0.60064725 0.11392136]\n",
      "\n",
      " [ 140.12850352   -9.7363727   -88.51988696  -53.76755416   42.07682923\n",
      "   14.55135862   75.24361718    0.          261.3070028  -210.90612245\n",
      "   12.82171252]\n",
      "-5034.008427145931\n",
      "[0.2048094  0.13537032 0.10207248 0.0461318  0.21676906 0.1625352\n",
      " 0.31774031 5.         0.43367119 0.60043635 0.11393418]\n",
      "\n",
      " [ 139.11702808   -9.13111307  -86.50544169  -53.96087363   41.50545374\n",
      "   14.36990988   74.68365675    0.          258.06530175 -208.58746848\n",
      "   12.76047715]\n",
      "-5033.861659499475\n",
      "[0.20494851 0.13536119 0.10198598 0.04607784 0.21681057 0.16254957\n",
      " 0.31781499 5.         0.43392925 0.60022776 0.11394694]\n",
      "\n",
      " [ 138.13149781   -8.56204497  -84.55127863  -54.10266476   40.93742518\n",
      "   14.19225282   74.12326522    0.          254.85510741 -206.29640494\n",
      "   12.68461529]\n",
      "-5033.718208101185\n",
      "[0.20508665 0.13535263 0.10190143 0.04602374 0.2168515  0.16256376\n",
      " 0.31788911 5.         0.43418411 0.60002146 0.11395962]\n",
      "\n",
      " [ 137.17160436   -8.02773138  -82.65780798  -54.19580611   40.37144299\n",
      "   14.01655754   73.56188416    0.          251.67533295 -204.03067992\n",
      "   12.59854889]\n",
      "-5033.577993298774\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[0.20522382 0.1353446  0.10181877 0.04596954 0.21689188 0.16257778\n",
      " 0.31796268 5.         0.43443578 0.59981743 0.11397222]\n",
      "\n",
      " [ 136.23646282   -7.5276821   -80.82471231  -54.24315532   39.80703423\n",
      "   13.84159805   72.99885783    0.          248.52449133 -201.78823979\n",
      "   12.50440891]\n",
      "-5033.440938980909\n",
      "[0.20536005 0.13533707 0.10173794 0.0459153  0.21693168 0.16259162\n",
      " 0.31803568 5.         0.43468431 0.59961565 0.11398473]\n",
      "\n",
      " [ 135.32534831   -7.06039555  -79.05065104  -54.24698065   39.24451448\n",
      "   13.66745983   72.43450364    0.          245.40112008 -199.56681263\n",
      "   12.40337406]\n",
      "-5033.306972508114\n",
      "[0.20549538 0.13533001 0.10165889 0.04586105 0.21697093 0.16260529\n",
      " 0.31810811 5.         0.43492971 0.59941608 0.11399713]\n",
      "\n",
      " [ 134.43731677   -6.62524417  -77.33414895  -54.21005972   38.6841084\n",
      "   13.49400281   71.86907897    0.          242.30342842 -197.36470676\n",
      "   12.29590362]\n",
      "-5033.176024618857\n",
      "[0.20562982 0.13532339 0.10158156 0.04580684 0.21700961 0.16261878\n",
      " 0.31817998 5.         0.43517201 0.59921871 0.11400943]\n",
      "\n",
      " [ 133.57148237   -6.22090192  -75.67326611  -54.13440522   38.12675641\n",
      "   13.32152078   71.30294853    0.          239.22953642 -195.17961482\n",
      "   12.18268517]\n",
      "-5033.048029473025\n",
      "[0.20576339 0.13531717 0.10150588 0.04575271 0.21704774 0.1626321\n",
      " 0.31825128 5.         0.43541124 0.59902353 0.11402161]\n",
      "\n",
      " [ 132.72678916   -5.8465248   -74.06610985  -54.02314036   37.5728232\n",
      "   13.15014015   70.73642973    0.          236.17738043 -193.00954682\n",
      "   12.06383604]\n",
      "-5032.922680976651\n",
      "[0.20589611 0.13531132 0.10143182 0.04569869 0.21708531 0.16264525\n",
      " 0.31832202 5.         0.43564742 0.59883052 0.11403367]\n",
      "\n",
      " [ 131.89969832   -5.54776307  -72.48047768  -53.65500965   37.18076368\n",
      "   13.09307663   70.2568168     0.          233.0813486  -190.89457055\n",
      "   11.87561429]\n",
      "-5032.800454743476\n",
      "[0.20602801 0.13530577 0.10135934 0.04564503 0.21712249 0.16265835\n",
      " 0.31839228 5.         0.4358805  0.59863963 0.11404555]\n",
      "\n",
      " [ 131.10130203   -5.08067933  -71.00349376  -53.51172422   36.59448885\n",
      "   12.88449403   69.66675937    0.          229.94896517 -188.55679627\n",
      "   11.81360039]\n",
      "-5032.6811273143085\n",
      "[0.20615912 0.13530069 0.10128833 0.04559152 0.21715909 0.16267123\n",
      " 0.31846194 5.         0.43611045 0.59845107 0.11405736]\n",
      "\n",
      " [ 130.31751405   -4.6420173   -69.55670415  -53.32330511   36.03660298\n",
      "   12.70202847   69.09136118    0.          226.82536473 -186.23097094\n",
      "   11.70698215]\n",
      "-5032.564645516326\n",
      "[0.20628943 0.13529605 0.10121878 0.0455382  0.21719512 0.16268393\n",
      " 0.31853103 5.         0.43633727 0.59826484 0.11406907]\n",
      "\n",
      " [ 129.54941758   -4.22872199  -68.14624703  -53.09574954   35.49798748\n",
      "   12.53451046   68.5247096     0.          223.710108   -183.913\n",
      "   11.5782068 ]\n",
      "-5032.450958132126\n",
      "[0.20641898 0.13529182 0.10115063 0.0454851  0.21723062 0.16269647\n",
      " 0.31859956 5.         0.43656098 0.59808093 0.11408065]\n",
      "\n",
      " [ 128.79730184   -3.83900624  -66.77403508  -52.83293558   34.97384296\n",
      "   12.37637753   67.96420712    0.          220.60249921 -181.60005857\n",
      "   11.43844293]\n",
      "-5032.340015756053\n",
      "[0.20654778 0.13528798 0.10108386 0.04543227 0.2172656  0.16270884\n",
      " 0.31866752 5.         0.43678159 0.59789933 0.11409209]\n",
      "\n",
      " [ 128.06071845   -3.47108471  -65.44043299  -52.538614     34.46209939\n",
      "   12.22489482   67.40843764    0.          217.50196356 -179.29047772\n",
      "   11.29337579]\n",
      "-5032.231770289501\n",
      "[0.20667584 0.13528451 0.10101842 0.04537973 0.21730006 0.16272107\n",
      " 0.31873493 5.         0.43699909 0.59772004 0.11410338]\n",
      "\n",
      " [ 127.33939639   -3.12394513  -64.15311145  -52.21549509   33.96180273\n",
      "   12.07887635   66.85705011    0.          214.41660283 -176.98890042\n",
      "   11.14604038]\n",
      "-5032.126150988627\n",
      "[0.20680318 0.13528139 0.10095426 0.04532751 0.21733402 0.16273315\n",
      " 0.31880179 5.         0.43721351 0.59754305 0.11411453]\n",
      "\n",
      " [ 126.63362395   -2.79635151  -62.97692016  -51.86657108   33.47604888\n",
      "   11.94048491   66.31186443    0.          211.41336492 -174.74483229\n",
      "   10.99510519]\n",
      "-5032.0230682966385\n",
      "[0.20692981 0.13527859 0.10089129 0.04527565 0.21736749 0.16274509\n",
      " 0.3188681  5.         0.43742492 0.59736831 0.11412552]\n",
      "\n",
      " [ 125.93921747   -2.4870651   -61.82290417  -51.4956082    33.01036486\n",
      "   11.81395055   65.77622753    0.          208.41516834 -172.5025113\n",
      "   10.83694588]\n",
      "-5031.9224801654855\n",
      "[0.20705575 0.1352761  0.10082946 0.04522415 0.21740051 0.1627569\n",
      " 0.31893388 5.         0.43763333 0.5971958  0.11413636]\n",
      "\n",
      " [ 125.25632792   -2.1944943   -60.69244046  -51.10534767   32.56159107\n",
      "   11.69571624   65.24833771    0.          205.42161747 -170.26012938\n",
      "   10.67813628]\n",
      "-5031.824345523998\n",
      "[0.20718101 0.13527391 0.10076877 0.04517305 0.21743307 0.1627686\n",
      " 0.31899912 5.         0.43783876 0.59702554 0.11414704]\n",
      "\n",
      " [ 124.58509536   -1.91764866  -59.58614747  -50.69799045   32.12767297\n",
      "   11.58396208   64.7271745     0.          202.4321393  -168.01632046\n",
      "   10.521494  ]\n",
      "-5031.728625355199\n",
      "[0.20730559 0.13527199 0.10070919 0.04512235 0.21746519 0.16278018\n",
      " 0.31906385 5.         0.43804119 0.59685753 0.11415756]\n",
      "\n",
      " [ 123.92960252   -1.6557351   -58.5032958   -50.27533462   31.70794298\n",
      "   11.47795956   64.21255694    0.          199.44113137 -165.76222333\n",
      "   10.36862886]\n",
      "-5031.635282806634\n",
      "[0.20742952 0.13527034 0.10065068 0.04507207 0.2174969  0.16279166\n",
      " 0.31912806 5.         0.43824063 0.59669176 0.11416793]\n",
      "\n",
      " [ 123.28511457   -1.40779775  -57.44289501  -49.83929466   31.30176719\n",
      "   11.37733489   63.70409574    0.          196.45445809 -163.50597616\n",
      "   10.21999196]\n",
      "-5031.544278430288\n",
      "[0.20755281 0.13526893 0.10059324 0.04502223 0.2175282  0.16280304\n",
      " 0.31919177 5.         0.43843708 0.59652826 0.11417815]\n",
      "\n",
      " [ 122.65154874   -1.17312084  -56.40451127  -49.3915004    30.90847804\n",
      "   11.28144959   63.20159355    0.          193.47220132 -161.24719696\n",
      "   10.07594801]\n",
      "-5031.455228752408\n",
      "[0.20767546 0.13526776 0.10053683 0.04497284 0.21755911 0.16281432\n",
      " 0.31925497 5.         0.43863056 0.59636701 0.11418822]\n",
      "\n",
      " [ 122.02686048   -1.05383515  -55.36606477  -48.70349403   30.68501337\n",
      "   11.30374039   62.79184345    0.          190.47894421 -159.09268222\n",
      "    9.86590112]\n",
      "-5031.368758892514\n",
      "[0.20779749 0.1352667  0.10048147 0.04492414 0.2175898  0.16282562\n",
      " 0.31931776 5.         0.43882103 0.59620792 0.11419809]\n",
      "\n",
      " [ 121.42122159   -0.82711995  -54.40015502  -48.27236262   30.27251296\n",
      "   11.17411193   62.27567428    0.          187.51383595 -156.81126843\n",
      "    9.80040295]\n",
      "-5031.284510318516\n",
      "[0.20791891 0.13526587 0.10042707 0.04487587 0.21762007 0.1628368\n",
      " 0.31938004 5.         0.43900855 0.59605111 0.11420789]\n",
      "\n",
      " [ 120.82211833   -0.61625178  -53.4387982   -47.82559517   29.88997494\n",
      "   11.06925538   61.77656996    0.          184.55286863 -154.53375454\n",
      "    9.70100155]\n",
      "-5031.202443933288\n",
      "[0.20803973 0.13526526 0.10037363 0.04482804 0.21764996 0.16284786\n",
      " 0.31944181 5.         0.4391931  0.59589657 0.11421759]\n",
      "\n",
      " [ 120.23118688   -0.41873955  -52.48873458  -47.36710707   29.5281443\n",
      "   10.97885252   61.28908717    0.          181.59761839 -152.25794596\n",
      "    9.58786313]\n",
      "-5031.122520541819\n",
      "[0.20815996 0.13526484 0.10032114 0.04478067 0.21767949 0.16285884\n",
      " 0.3195031  5.         0.4393747  0.59574432 0.11422718]\n",
      "\n",
      " [ 119.64933674   -0.23274697  -51.55331019  -46.89964135   29.18234168\n",
      "   10.89750822   60.81018546    0.          178.64934671 -149.98334882\n",
      "    9.47087119]\n",
      "-5031.044700716174\n",
      "[0.20827961 0.13526461 0.10026959 0.04473377 0.21770867 0.16286974\n",
      " 0.31956391 5.         0.43955335 0.59559433 0.11423665]\n",
      "\n",
      " [ 1.19076556e+02 -5.73472789e-02 -5.06339575e+01 -4.64251689e+01\n",
      "  2.88499323e+01  1.08225513e+01  6.03385324e+01  0.00000000e+00\n",
      "  1.75709082e+02 -1.47709914e+02  9.35478329e+00]\n",
      "-5030.968944864271\n",
      "[0.20839869 0.13526455 0.10021895 0.04468735 0.21773752 0.16288056\n",
      " 0.31962425 5.         0.43972906 0.59544662 0.114246  ]\n",
      "\n",
      " [ 1.18513043e+02  1.08389941e-01 -4.97309547e+01 -4.59453768e+01\n",
      "  2.85292390e+01  1.07523929e+01  5.98731613e+01  0.00000000e+00\n",
      "  1.72778025e+02 -1.45438125e+02  9.24177675e+00]\n",
      "-5030.895213134346\n",
      "[0.2085172  0.13526466 0.10016922 0.0446414  0.21776605 0.16289132\n",
      " 0.31968412 5.         0.43990184 0.59530118 0.11425524]\n",
      "\n",
      " [ 117.95870341    0.2654142   -48.84431655  -45.4615174    28.21944054\n",
      "   10.68653819   59.41389099    0.          169.85767525 -143.16869147\n",
      "    9.13301756]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "-5030.823465337504\n",
      "[0.20863516 0.13526492 0.10012038 0.04459594 0.21779427 0.162902\n",
      " 0.31974354 5.         0.44007169 0.59515801 0.11426438]\n",
      "\n",
      " [ 117.41338858    0.41414751  -47.97379916  -44.97469035   27.91974475\n",
      "   10.62437696   58.96026596    0.          166.94893384 -140.90244076\n",
      "    9.02884221]\n",
      "-5030.753661101114\n",
      "[0.20875257 0.13526534 0.1000724  0.04455097 0.21782219 0.16291263\n",
      " 0.3198025  5.         0.44023864 0.59501711 0.11427341]\n",
      "\n",
      " [ 116.8772269     0.55514101  -47.11920519  -44.48587697   27.62973418\n",
      "   10.5655763    58.51207516    0.          164.05332735 -138.64045104\n",
      "    8.92934167]\n",
      "-5030.685759701237\n",
      "[0.20886945 0.13526589 0.10002529 0.04450648 0.21784982 0.16292319\n",
      " 0.31986101 5.         0.4404027  0.59487847 0.11428234]\n",
      "\n",
      " [ 116.34990187    0.68951795  -46.28000625  -43.99612226   27.34887767\n",
      "   10.50999981   58.06923491    0.          161.17207724 -136.38323435\n",
      "    8.83454049]\n",
      "-5030.61972027784\n",
      "[0.2089858  0.13526658 0.09997901 0.04446248 0.21787717 0.1629337\n",
      " 0.31991908 5.         0.44056387 0.59474209 0.11429117]\n",
      "\n",
      " [ 115.83128435    0.81709732  -45.45597312  -43.50612198   27.07671229\n",
      "   10.4573819    57.63173704    0.          158.30608663 -134.13273155\n",
      "    8.74426587]\n",
      "-5030.55550171541\n",
      "[0.20910163 0.1352674  0.09993355 0.04441898 0.21790424 0.16294416\n",
      " 0.31997671 5.         0.44072217 0.59460796 0.11429991]\n",
      "\n",
      " [ 115.3213525     0.93869585  -44.64675112  -43.01675381   26.81282422\n",
      "   10.40773805   57.19918681    0.          155.45722999 -131.88954563\n",
      "    8.65829134]\n",
      "-5030.493062697266\n",
      "[0.20921695 0.13526834 0.0998889  0.04437596 0.21793106 0.16295457\n",
      " 0.32003391 5.         0.44087763 0.59447607 0.11430857]\n",
      "\n",
      " [ 114.8199608     1.05469462  -43.85203738  -42.52785311   26.5569588\n",
      "   10.36065714   56.77169156    0.          152.62662964 -129.65446604\n",
      "    8.57681061]\n",
      "-5030.432361878935\n",
      "[0.20933177 0.13526939 0.09984505 0.04433343 0.21795761 0.16296493\n",
      " 0.32009068 5.         0.44103026 0.59434641 0.11431715]\n",
      "\n",
      " [ 114.32712745    1.16570118  -43.07130894  -42.04203105   26.3087004\n",
      "   10.31633747   56.34917034    0.          149.8155907  -127.42991021\n",
      "    8.4994108 ]\n",
      "-5030.373357579408\n",
      "[0.2094461  0.13527056 0.09980198 0.04429139 0.21798392 0.16297524\n",
      " 0.32014703 5.         0.44118007 0.59421898 0.11432565]\n",
      "\n",
      " [ 113.8425896     1.27187832  -42.30465129  -41.55780607   26.0676934\n",
      "   10.27442158   55.93144942    0.          147.0253992  -125.2158836\n",
      "    8.42591817]\n",
      "-5030.316008334682\n",
      "[0.20955994 0.13527183 0.09975967 0.04424983 0.21800999 0.16298552\n",
      " 0.32020296 5.         0.4413271  0.59409377 0.11433407]\n",
      "\n",
      " [ 113.366189      1.37361803  -41.55172792  -41.07645418   25.83364494\n",
      "   10.23481491   55.51834238    0.          144.25699192 -123.01404149\n",
      "    8.35620267]\n",
      "-5030.260272660317\n",
      "[0.20967331 0.1352732  0.09971812 0.04420876 0.21803582 0.16299575\n",
      " 0.32025848 5.         0.44147136 0.59397075 0.11434243]\n",
      "\n",
      " [ 112.89808117    1.47139235  -40.81207499  -40.59823277   25.60623489\n",
      "   10.19739557   55.10990923    0.          141.51190862 -120.82541343\n",
      "    8.29015517]\n",
      "-5030.206109098846\n",
      "[0.20978621 0.13527468 0.09967731 0.04416816 0.21806143 0.16300595\n",
      " 0.32031359 5.         0.44161287 0.59384993 0.11435072]\n",
      "\n",
      " [ 112.43790232    1.56540864  -40.08557789  -40.12360841   25.38513399\n",
      "   10.16222177   54.70599262    0.          138.79120252 -118.65116267\n",
      "    8.22762104]\n",
      "-5030.153476368496\n",
      "[0.20989864 0.13527624 0.09963723 0.04412804 0.21808682 0.16301611\n",
      " 0.3203683  5.         0.44175166 0.59373128 0.11435895]\n",
      "\n",
      " [ 111.98555967    1.65597157  -39.37217662  -39.65293672   25.17004214\n",
      "   10.12888879   54.30639521    0.          136.09581674 -116.4924679\n",
      "    8.16830925]\n",
      "-5030.102333401056\n",
      "[0.21001063 0.1352779  0.09959785 0.04408838 0.21811199 0.16302624\n",
      " 0.3204226  5.         0.44188775 0.59361478 0.11436712]\n",
      "\n",
      " [ 111.54109143    1.74330762  -38.67151918  -39.18639868   24.96056459\n",
      "   10.09745301   53.91116792    0.          133.42700458 -114.35044325\n",
      "    8.11212431]\n",
      "-5030.052191325555\n",
      "[0.21012217 0.13527964 0.09955918 0.0440492  0.21813695 0.16303634\n",
      " 0.32047652 5.         0.44202118 0.59350043 0.11437523]\n",
      "\n",
      " [ 111.10304331    1.71407555  -37.96871624  -38.48218694   24.91851319\n",
      "   10.18615694   53.60963496    0.          130.77349558 -112.3359807\n",
      "    7.97904704]\n",
      "-5030.003885952224\n",
      "GD finish\n"
     ]
    }
   ],
   "source": [
    "alpha_rate = 20\n",
    "learnrate = 10 ** -6\n",
    "\n",
    "point_add_gra = [point_all[-1]]\n",
    "max_ll_last = np.max(Loglikelihood)\n",
    "LL_add_gra = [max_ll_last]\n",
    "\n",
    "while True:\n",
    "    alpha = 1\n",
    "    \n",
    "    p_last = point_add_gra[-1]\n",
    "    twoboxd = twoboxColMDP_der(discount, nq, nr, na, nl, p_last)\n",
    "    twoboxd1st = twoboxd.dloglikelihhod_dpara_sim(obs_IRC)\n",
    "    print(p_last)\n",
    "    print('\\n', twoboxd1st)\n",
    "    \n",
    "    para_temp = p_last + alpha * learnrate * np.array(twoboxd1st)\n",
    "    \n",
    "    twoboxCol_test = twoboxColMDP(discount, nq, nr, na, nl, para_temp)\n",
    "    twoboxCol_test.setupMDP()\n",
    "    twoboxCol_test.solveMDP_sfm()\n",
    "    ThA = twoboxCol_test.ThA\n",
    "    policy = twoboxCol_test.softpolicy\n",
    "    pi = np.ones(nq * nq)/ nq /nq  # initialize the estimation of the belief state\n",
    "    Trans_hybrid_obs12 = twoboxCol_test.Trans_hybrid_obs12\n",
    "    Obs_emis_trans1 = twoboxCol_test.Obs_emis_trans1\n",
    "    Obs_emis_trans2 = twoboxCol_test.Obs_emis_trans2\n",
    "    twoboxColHMM_test = HMMtwoboxCol(ThA, policy, Trans_hybrid_obs12, Obs_emis_trans1, Obs_emis_trans2, pi, Ncol)\n",
    "    max_ll_new_temp = twoboxColHMM_test.computeQaux(obs_IRC, ThA, policy, Trans_hybrid_obs12, \n",
    "                                               Obs_emis_trans1, Obs_emis_trans2) + twoboxColHMM_test.latent_entr(obs_IRC)\n",
    "    max_ll_temp = max_ll_new_temp\n",
    "    print(max_ll_temp)\n",
    "    \n",
    "    while max_ll_new_temp < max_ll_last + 0.2 * alpha * learnrate * np.array(twoboxd1st).dot(np.array(twoboxd1st)):\n",
    "        alpha /= alpha_rate\n",
    "        para_temp = p_last + alpha * learnrate * np.array(twoboxd1st)\n",
    "        print(para_temp)\n",
    "\n",
    "        ## Check the ECDLL (old posterior, new parameters)\n",
    "        twoboxCol_new = twoboxColMDP(discount, nq, nr, na, nl, para_temp)\n",
    "        twoboxCol_new.setupMDP()\n",
    "        twoboxCol_new.solveMDP_sfm()\n",
    "        ThA_new = twoboxCol_new.ThA\n",
    "        softpolicy_new = twoboxCol_new.softpolicy\n",
    "        Trans_hybrid_obs12_new = twoboxCol_new.Trans_hybrid_obs12\n",
    "        Obs_emis_trans1_new = twoboxCol_new.Obs_emis_trans1\n",
    "        Obs_emis_trans2_new = twoboxCol_new.Obs_emis_trans2\n",
    "        \n",
    "        twoboxColHMM_test = HMMtwoboxCol(ThA, policy, Trans_hybrid_obs12, Obs_emis_trans1, Obs_emis_trans2, pi, Ncol)\n",
    "        max_ll_new_temp = twoboxColHMM_test.computeQaux(obs_IRC, ThA, policy, Trans_hybrid_obs12, \n",
    "                                               Obs_emis_trans1, Obs_emis_trans2) + twoboxColHMM_test.latent_entr(obs_IRC)\n",
    "        \n",
    "        print('max_ll_new_temp: ', max_ll_new_temp, '  (alpha = ', alpha, ')')\n",
    "        if alpha < 10** -6 or max_ll_new_temp < max_ll_temp:\n",
    "            break\n",
    "        \n",
    "        max_ll_temp = max_ll_new_temp\n",
    "            \n",
    "    point_add_gra.append(para_temp)\n",
    "    LL_add_gra.append(max_ll_temp)\n",
    "\n",
    "    if len(LL_add_gra) >= 2 and np.abs(LL_add_gra[-1] - LL_add_gra[-2]) < 5 * 10** -2:\n",
    "        print(\"GD finish\")\n",
    "        break\n",
    "    \n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {
    "hidden": true
   },
   "outputs": [],
   "source": [
    "point_all = point_all + point_add_gra"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {
    "hidden": true
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([0.21023327, 0.13528135, 0.09952121, 0.04401071, 0.21816186,\n",
       "       0.16304653, 0.32053012, 5.        , 0.44215195, 0.5933881 ,\n",
       "       0.11438321])"
      ]
     },
     "execution_count": 22,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "point_final = point_all[-1]\n",
    "point_final"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Check points"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "heading_collapsed": true
   },
   "source": [
    "## Contour"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "metadata": {
    "hidden": true
   },
   "outputs": [],
   "source": [
    "uniques = rmv_dup_arrary(point_all)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "metadata": {
    "hidden": true
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([[-0.28479654, -0.03616195,  0.01501443,  0.30765107, -0.16624557,\n",
       "        -0.07873555, -0.18756126,  0.        , -0.71605748,  0.47941111,\n",
       "        -0.10544762],\n",
       "       [ 0.15208068, -0.24724468, -0.83729448,  0.29624163, -0.13295645,\n",
       "        -0.07835239, -0.10235004,  0.        , -0.00381576, -0.23848546,\n",
       "        -0.18904621]])"
      ]
     },
     "execution_count": 24,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "pca = PCA(n_components = 2)\n",
    "pca.fit(uniques - uniques[-1])\n",
    "projectionMat = pca.components_\n",
    "projectionMat"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "metadata": {
    "hidden": true
   },
   "outputs": [],
   "source": [
    "# Contour of the likelihood\n",
    "step1 = 0.02   # for u (1st principle component)\n",
    "step2 = 0.02  # for v (2nd principle component)\n",
    "N1 = 15\n",
    "N2 = 15\n",
    "uOffset = - step1 * N1 / 2\n",
    "vOffset = - step2 * N2 / 2\n",
    "\n",
    "uValue = np.zeros(N1)\n",
    "vValue = np.zeros(N2)\n",
    "Qaux1 = np.zeros((N2, N1))    # Likelihood with ground truth latent\n",
    "Qaux2 = np.zeros((N2, N1))    # Expected complete data likelihood\n",
    "Qaux3 = np.zeros((N2, N1))    # Entropy of latent posterior\n",
    "para_slice = []\n",
    "\n",
    "for i in range(N1):\n",
    "    uValue[i] = step1 * (i) + uOffset\n",
    "    for j in range(N2):\n",
    "        vValue[j] = step2 * (j) + vOffset\n",
    "\n",
    "        para_slicePoints = point_all[-1] + uValue[i] * projectionMat[0] + vValue[j] * projectionMat[1]\n",
    "        para_slice.append(para_slicePoints)\n",
    "        para = np.copy(para_slicePoints)\n",
    "        #print(para)\n",
    "        \n",
    "        twoboxCol = twoboxColMDP(discount, nq, nr, na, nl, para)\n",
    "        twoboxCol.setupMDP()\n",
    "        if np.any(twoboxCol.ThA<0) == True:\n",
    "            Qaux2[j, i] = np.nan\n",
    "            Qaux3[j, i] = np.nan\n",
    "        else:\n",
    "            twoboxCol.solveMDP_sfm()\n",
    "            ThA = twoboxCol.ThA\n",
    "            policy = twoboxCol.softpolicy\n",
    "            pi = np.ones(nq * nq)/ nq /nq  # initialize the estimation of the belief state\n",
    "            Trans_hybrid_obs12 = twoboxCol.Trans_hybrid_obs12\n",
    "            Obs_emis_trans1 = twoboxCol.Obs_emis_trans1\n",
    "            Obs_emis_trans2 = twoboxCol.Obs_emis_trans2\n",
    "            twoboxColHMM = HMMtwoboxCol(ThA, policy, Trans_hybrid_obs12, Obs_emis_trans1, Obs_emis_trans2, pi, Ncol)\n",
    "\n",
    "            #Qaux1[j, i] = twoboxHMM.likelihood(lat, obs, ThA, policy)  #given latent state\n",
    "            Qaux2[j, i] = twoboxColHMM.computeQaux(obs_IRC, ThA, policy, Trans_hybrid_obs12, Obs_emis_trans1, Obs_emis_trans2) \n",
    "            Qaux3[j, i] = twoboxColHMM.latent_entr(obs_IRC) "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "metadata": {
    "hidden": true
   },
   "outputs": [],
   "source": [
    "Loglikelihood_original = Qaux2+Qaux3"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "metadata": {
    "hidden": true
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([[-5554.6061138 , -5543.82151594, -5527.06319728, -5506.75583891,\n",
       "        -5485.10511412, -5457.43374335, -5423.66505436, -5387.26111913,\n",
       "        -5356.44479799, -5336.20928668, -5317.85322821, -5306.76942748,\n",
       "        -5303.58865251, -5308.5940297 , -5315.38111307],\n",
       "       [-5541.50630707, -5524.92179055, -5503.00253512, -5477.9246969 ,\n",
       "        -5449.71735812, -5414.99072393, -5374.38329495, -5335.7903743 ,\n",
       "        -5305.69669282, -5286.04713692, -5270.81507501, -5263.83537719,\n",
       "        -5262.19015665, -5268.28795316, -5280.96238352],\n",
       "       [-5531.40180791, -5503.39568364, -5474.82578828, -5444.09136259,\n",
       "        -5407.80165464, -5366.86898991, -5320.84547316, -5281.33927083,\n",
       "        -5254.46404398, -5235.45702683, -5223.20325111, -5220.12543083,\n",
       "        -5222.53716362, -5228.89327731, -5244.31486968],\n",
       "       [-5515.4230668 , -5476.11985767, -5442.73688066, -5404.30553856,\n",
       "        -5358.37251461, -5312.75593333, -5264.04611485, -5225.73032846,\n",
       "        -5202.27817979, -5186.24854537, -5176.00087583, -5176.32596975,\n",
       "        -5183.81354026, -5193.76081724, -5209.40446295],\n",
       "       [-5490.27196003, -5442.34666182, -5403.82110502, -5356.81265709,\n",
       "        -5301.67750275, -5252.71175676, -5205.25547745, -5170.96397748,\n",
       "        -5150.22982141, -5138.58290654, -5132.61313136, -5136.5230354 ,\n",
       "        -5147.8724687 , -5164.27618193, -5183.72423535],\n",
       "       [-5464.53550649, -5408.78804895, -5356.9041632 , -5300.8108727 ,\n",
       "        -5240.44156174, -5188.90522606, -5146.43196208, -5118.64946774,\n",
       "        -5102.3993576 , -5094.03782911, -5095.44405623, -5104.555541  ,\n",
       "        -5120.55797276, -5142.86756564, -5169.60915492],\n",
       "       [-5439.15043107, -5371.64474645, -5302.86116035, -5240.73083816,\n",
       "        -5179.32748729, -5127.68693009, -5090.65632716, -5071.78913099,\n",
       "        -5062.53152263, -5059.46093162, -5067.61296811, -5083.98163135,\n",
       "        -5106.88262773, -5136.01414356, -5169.21002365],\n",
       "       [-5409.77197503, -5322.97866684, -5247.02862565, -5183.52643615,\n",
       "        -5123.41275511, -5076.95999759, -5046.8440748 , -5036.88939261,\n",
       "        -5036.22848693, -5043.41238926, -5059.7660125 , -5083.25654978,\n",
       "        -5113.2707122 , -5150.64978064, -5190.27087467],\n",
       "       [-5384.11844794, -5287.99350557, -5206.02490373, -5139.9158234 ,\n",
       "        -5084.85701811, -5048.89423524, -5033.06638759, -5032.77916058,\n",
       "        -5041.37754404, -5059.02547542, -5085.39834611, -5116.67641912,\n",
       "        -5152.64457271, -5194.97330799, -5240.5860115 ],\n",
       "       [-5378.62095482, -5283.38052091, -5199.50693555, -5137.98581839,\n",
       "        -5092.07324297, -5069.40359101, -5067.92133381, -5083.35057792,\n",
       "        -5102.86266303, -5127.65936123, -5155.26109824, -5191.82353925,\n",
       "        -5235.32772083, -5279.01028922, -5325.43311547],\n",
       "       [-5418.53936768, -5333.72487065, -5258.29934934, -5204.48165759,\n",
       "        -5167.37449377, -5160.15920226, -5165.76323673, -5187.25481242,\n",
       "        -5218.81401231, -5250.98728257, -5275.15371499, -5307.99348274,\n",
       "        -5356.30968962, -5403.08921307, -5449.53914638],\n",
       "       [-5517.62048041, -5434.74646918, -5366.0584131 , -5319.57689626,\n",
       "        -5301.9979704 , -5312.48810011, -5327.66435473, -5340.38070753,\n",
       "        -5372.65236013, -5409.69190323, -5437.32444505, -5464.01411564,\n",
       "        -5511.48542342, -5558.37063665, -5605.97932497],\n",
       "       [-5679.06180088, -5592.37090721, -5523.25681498, -5483.89065719,\n",
       "        -5488.890563  , -5513.45878972, -5536.82950594, -5543.21174748,\n",
       "        -5567.81024745, -5600.42834021, -5629.30645908, -5657.67292942,\n",
       "        -5701.87743541, -5746.20196201, -5795.61091009],\n",
       "       [-5870.29960816, -5794.04535985, -5735.31950842, -5702.66371394,\n",
       "        -5720.83740857, -5752.13846281, -5777.49767603, -5784.89198135,\n",
       "        -5808.87460176, -5833.75293425, -5858.89879605, -5891.26121158,\n",
       "        -5931.78576754, -5970.29184269, -6019.91382825],\n",
       "       [-6052.39580607, -5991.89718925, -5951.04785075, -5943.15327724,\n",
       "        -5973.84705077, -6005.76138073, -6033.4271591 , -6047.35257386,\n",
       "        -6081.68025455, -6108.65895254, -6130.01291956, -6165.66140714,\n",
       "        -6206.57603114, -6240.19158895, -6288.313863  ]])"
      ]
     },
     "execution_count": 28,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "Loglikelihood = np.nan_to_num(Loglikelihood_original, nan = np.nanmean(Loglikelihood_original))\n",
    "Loglikelihood"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "metadata": {
    "hidden": true
   },
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 720x720 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# project the trajectories onto the plane\n",
    "point_2d = projectionMat.dot((point_all - point_all[-1]).T).T   \n",
    "\n",
    "# true parameters projected onto the plane\n",
    "true_2d = projectionMat.dot(parametersAgent - point_all[-1])\n",
    "\n",
    "fig, ax = plt.subplots(figsize = (10, 10))\n",
    "uValuemesh, vValuemesh = np.meshgrid(uValue[:], vValue[:])\n",
    "cs3 = plt.contourf(uValuemesh, vValuemesh, Loglikelihood[:, :], \n",
    "                  np.arange(np.min(Loglikelihood[:, :]), np.max(Loglikelihood[:, :]), 20), cmap='jet')\n",
    "#plt.xticks(np.arange(0, 1, 0.1))\n",
    "#plt.yticks(np.arange(0, 1, 0.1))\n",
    "plt.plot(point_2d[:, 0], point_2d[:, 1], marker='.', color = 'b')   # projected trajectories\n",
    "plt.plot(point_2d[-1, 0], point_2d[-1, 1], marker='*', color = 'g', markersize = 10)        # final point\n",
    "plt.plot(true_2d[0], true_2d[1], marker='o', color = 'g')           # true\n",
    "\n",
    "\n",
    "ax.grid()\n",
    "ax.set_title('Likelihood of observed data')\n",
    "plt.xlabel(r'$u \\mathbf{\\theta}$', fontsize = 10)\n",
    "plt.ylabel(r'$v \\mathbf{\\theta}$', fontsize = 10)\n",
    "#plt.clabel(cs3, inline=1, fontsize=10)\n",
    "#plt.colorbar()\n",
    "plt.show()\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "hidden": true
   },
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Hessian of final points"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "metadata": {},
   "outputs": [],
   "source": [
    "L =len(parametersAgent)\n",
    "hes_final = np.zeros((L,L))\n",
    "\n",
    "paraHessian = point_final\n",
    "#paraHessian = parametersAgent\n",
    "twoboxd = twoboxColMDP_der(discount, nq, nr, na, nl, paraHessian)\n",
    "twoboxd1st = twoboxd.dloglikelihhod_dpara_sim(obs_IRC)\n",
    "\n",
    "perturb = 10** -6\n",
    "for i in range(L):\n",
    "    if i != 7:\n",
    "        para_perb = np.copy(paraHessian)\n",
    "        para_perb[i] += perturb\n",
    "\n",
    "        twoboxd_perb = twoboxColMDP_der(discount, nq, nr, na, nl, para_perb)       \n",
    "        twoboxd1st_pert = twoboxd_perb.dloglikelihhod_dpara_sim(obs_IRC)\n",
    "\n",
    "        hes_final[i, :] = (twoboxd1st_pert - twoboxd1st) / perturb"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(array([-5.07817131e+05, -1.40464925e+05, -1.17227460e+05, -3.43136854e+04,\n",
       "        -1.83645366e+04, -2.38137960e+02, -1.43144662e+03, -4.91850877e+03,\n",
       "        -7.51482057e+03, -9.25024302e+03,  0.00000000e+00]),\n",
       " array([[-0.06835068,  0.05614177, -0.11024528,  0.05492367,  0.06974349,\n",
       "         -0.52735272,  0.81474913, -0.16769625, -0.04552426, -0.00547496,\n",
       "          0.        ],\n",
       "        [-0.06580259,  0.18747366,  0.26697153,  0.06896951, -0.03622803,\n",
       "         -0.34094243, -0.17887315,  0.33717735, -0.78727484,  0.03811827,\n",
       "          0.        ],\n",
       "        [ 0.29637767, -0.36491994,  0.62442954, -0.11412993, -0.12846543,\n",
       "          0.10256072,  0.30180107,  0.34812075,  0.14849887,  0.33868849,\n",
       "          0.        ],\n",
       "        [ 0.09201297, -0.53031212, -0.65548474,  0.00198159, -0.13942414,\n",
       "          0.1668255 ,  0.12461281,  0.31773425, -0.30684339,  0.1507726 ,\n",
       "          0.        ],\n",
       "        [ 0.32547198, -0.40567953,  0.13868478, -0.11532042,  0.05872056,\n",
       "         -0.14419815, -0.03987292, -0.06374   , -0.08470771, -0.81233651,\n",
       "          0.        ],\n",
       "        [ 0.3871717 , -0.10091111,  0.0580649 ,  0.83596232, -0.05933245,\n",
       "          0.02137619, -0.0612954 , -0.32554046, -0.08936283,  0.12654554,\n",
       "          0.        ],\n",
       "        [ 0.20703958, -0.16768527,  0.0470054 , -0.47285292,  0.0669797 ,\n",
       "          0.01266418, -0.08666667, -0.67668736, -0.3450561 ,  0.33777153,\n",
       "          0.        ],\n",
       "        [ 0.        ,  0.        ,  0.        ,  0.        ,  0.        ,\n",
       "          0.        ,  0.        ,  0.        ,  0.        ,  0.        ,\n",
       "          1.        ],\n",
       "        [-0.02732827,  0.08837667,  0.08441519,  0.08188226,  0.72968738,\n",
       "          0.54011497,  0.28040121,  0.05122039, -0.25432437, -0.08668906,\n",
       "          0.        ],\n",
       "        [-0.08134814,  0.24997129,  0.0930954 , -0.02375158, -0.64325809,\n",
       "          0.50389519,  0.32310392, -0.15929294, -0.24662866, -0.25175092,\n",
       "          0.        ],\n",
       "        [-0.76722123, -0.52357507,  0.23747102,  0.19028005, -0.01419666,\n",
       "          0.01621575, -0.04119388, -0.20025826, -0.04679416, -0.01368397,\n",
       "          0.        ]]))"
      ]
     },
     "execution_count": 31,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Check eigenvalues of the Hessian\n",
    "w, v = np.linalg.eig(hes_final)\n",
    "w, v"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Std, Covariance, Correlation"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([0.04047267, 0.02484563, 0.01238955, 0.01303729, 0.01279313,\n",
       "       0.0070596 , 0.01158708, 0.0363209 , 0.03438527, 0.00392591])"
      ]
     },
     "execution_count": 32,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "selec_idx = list(range(7)) + list(range(8,11))\n",
    "var_para_woCol = np.linalg.inv(-hes_final[np.ix_(selec_idx, selec_idx)])\n",
    "std = np.sqrt(np.diagonal(var_para_woCol))\n",
    "std"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 33,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[Text(0,0,'Appearance rate (box 1)'),\n",
       " Text(0,0,'Appearance rate (box 2)'),\n",
       " Text(0,0,'Disappearance rate (box 1)'),\n",
       " Text(0,0,'Disappearance rate (box 2)'),\n",
       " Text(0,0,'Grooming reward'),\n",
       " Text(0,0,'Traveling cost'),\n",
       " Text(0,0,'Pushing button cost'),\n",
       " Text(0,0,'Color parameter 1'),\n",
       " Text(0,0,'Color parameter 2'),\n",
       " Text(0,0,'policy temperature')]"
      ]
     },
     "execution_count": 33,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x432 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "fig_correlation, ax = plt.subplots(figsize = (6, 6))\n",
    "ax.imshow(np.linalg.inv(np.diag(std)).dot(var_para_woCol).dot(np.linalg.inv(np.diag(std))))\n",
    "ax.set_title('Correlation of the parameters', fontsize = 30)\n",
    "ax.set_xticks(list(range(10)))\n",
    "ax.set_xticklabels( ('Appearance rate (box 1)', 'Appearance rate (box 2)', 'Disappearance rate (box 1)', \n",
    "                    'Disappearance rate (box 2)','Grooming reward', 'Traveling cost', 'Pushing button cost',\n",
    "                    'Color parameter 1','Color parameter 2', 'policy temperature'), rotation = 60, ha = 'right', fontsize = 16)\n",
    "ax.set_yticks(list(range(10)))\n",
    "ax.set_yticklabels( ('Appearance rate (box 1)', 'Appearance rate (box 2)', 'Disappearance rate (box 1)', \n",
    "                    'Disappearance rate (box 2)','Grooming reward', 'Traveling cost', 'Pushing button cost',\n",
    "                    'Color parameter 1','Color parameter 2', 'policy temperature'), ha = 'right', fontsize = 16)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 34,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[Text(0,0,'Appearance rate (box 1)'),\n",
       " Text(0,0,'Appearance rate (box 2)'),\n",
       " Text(0,0,'Disappearance rate (box 1)'),\n",
       " Text(0,0,'Disappearance rate (box 2)'),\n",
       " Text(0,0,'Grooming reward'),\n",
       " Text(0,0,'Traveling cost'),\n",
       " Text(0,0,'Pushing button cost'),\n",
       " Text(0,0,'Color parameter 1'),\n",
       " Text(0,0,'Color parameter 2'),\n",
       " Text(0,0,'policy temperature')]"
      ]
     },
     "execution_count": 34,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x432 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "fig_correlation, ax = plt.subplots(figsize = (6, 6))\n",
    "ax.imshow(var_para_woCol)\n",
    "ax.set_title('Covariance of the parameters', fontsize = 30)\n",
    "ax.set_xticks(list(range(10)))\n",
    "ax.set_xticklabels( ('Appearance rate (box 1)', 'Appearance rate (box 2)', 'Disappearance rate (box 1)', \n",
    "                    'Disappearance rate (box 2)','Grooming reward', 'Traveling cost', 'Pushing button cost',\n",
    "                    'Color parameter 1','Color parameter 2', 'policy temperature'), rotation = 90, ha = 'center', fontsize = 16)\n",
    "ax.set_yticks(list(range(10)))\n",
    "ax.set_yticklabels( ('Appearance rate (box 1)', 'Appearance rate (box 2)', 'Disappearance rate (box 1)', \n",
    "                    'Disappearance rate (box 2)','Grooming reward', 'Traveling cost', 'Pushing button cost',\n",
    "                    'Color parameter 1','Color parameter 2', 'policy temperature'), ha = 'right', fontsize = 16)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Check IRC result "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 35,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 720x360 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "\"\"\"\n",
    "Comparison of parameters\n",
    "\"\"\"\n",
    "\n",
    "N = 10\n",
    "ind = np.arange(N)  # the x locations for the groups\n",
    "width = 0.27       # the width of the bars\n",
    "\n",
    "fig_parameters = plt.figure(figsize = (10, 5))\n",
    "ax = fig_parameters.add_subplot(111)\n",
    "\n",
    "zvals = parametersAgent[:7] + parametersAgent[8:]\n",
    "yvals = point_final.tolist()[:7] + point_final.tolist()[8:]\n",
    "rects1 = ax.bar(ind, zvals, width, color='r')\n",
    "rects2 = ax.bar(ind+width, yvals, yerr=2*std, width = width, color = 'b', \n",
    "                align='center', alpha=1, ecolor='black', capsize=8)\n",
    "\n",
    "\n",
    "\n",
    "\n",
    "\n",
    "ax.set_title('Comparison of parameters', fontsize = 20)\n",
    "ax.set_ylabel('Parameters', fontsize = 16)\n",
    "ax.set_yticks([0, 0.5])\n",
    "ax.set_xticks(ind+width)\n",
    "ax.set_xticklabels( ('Appearance \\n rate (box 1)', 'Appearance \\n rate (box 2)', 'Disappearance \\n rate (box 1)', \n",
    "                    'Disappearance \\n rate (box 2)','Grooming \\n reward', 'Traveling \\n cost', 'Pushing button \\n cost',\n",
    "                    'Color \\n parameter 1','Color \\n parameter 2', 'policy \\n temperature'), rotation=60, ha = 'right', fontsize = 12)\n",
    "leg = ax.legend( (rects1[0], rects2[0]), ('teacher\\'s parameters', 'estimated parameters' ), fontsize = 16 , frameon=False)\n",
    "leg_c = ['r', 'b']\n",
    "for i, text in enumerate(leg.get_texts()):\n",
    "    plt.setp(text, color = leg_c[i])\n",
    "\n",
    "ax.xaxis.set_tick_params(length = 0)\n",
    "ax.spines['right'].set_visible(False)\n",
    "ax.spines['top'].set_visible(False)\n",
    "ax.spines['bottom'].set_visible(False)\n",
    "plt.show()\n",
    "\n",
    "fig_parameters.savefig('parameters.pdf', format='pdf', bbox_inches = 'tight')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 36,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([0.05, 0.15, 0.25, 0.35, 0.45, 0.55, 0.65, 0.75, 0.85, 0.95])"
      ]
     },
     "execution_count": 36,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "belief_array = (np.arange(nq) + 1/2)/nq\n",
    "belief_array"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 37,
   "metadata": {},
   "outputs": [],
   "source": [
    "twoboxCol = twoboxColMDP(discount, nq, nr, na, nl, point_final)\n",
    "#twoboxCol = twoboxColMDP(discount, nq, nr, na, nl, parametersAgent)\n",
    "twoboxCol.setupMDP()\n",
    "twoboxCol.solveMDP_sfm()\n",
    "ThA = twoboxCol.ThA\n",
    "policy = twoboxCol.softpolicy\n",
    "pi = np.ones(nq * nq)/ nq /nq  # initialize the estimation of the belief state\n",
    "Trans_hybrid_obs12 = twoboxCol.Trans_hybrid_obs12\n",
    "Obs_emis_trans1 = twoboxCol.Obs_emis_trans1\n",
    "Obs_emis_trans2 = twoboxCol.Obs_emis_trans2\n",
    "twoboxColHMM = HMMtwoboxCol(ThA, policy, Trans_hybrid_obs12, Obs_emis_trans1, Obs_emis_trans2, pi, Ncol)\n",
    "\n",
    "alpha_est, scale_est = twoboxColHMM.forward_scale(obs)\n",
    "beta_est = twoboxColHMM.backward_scale(obs, scale_est)\n",
    "gamma_est = twoboxColHMM.compute_gamma(alpha_est, beta_est)\n",
    "xi_est =  twoboxColHMM.compute_xi(alpha_est, beta_est, obs)\n",
    "\n",
    "belief1_est = np.sum(np.reshape(gamma_est.T, (gamma_est.T.shape[0], nq, nq)), axis = 2)\n",
    "belief2_est = np.sum(np.reshape(gamma_est.T, (gamma_est.T.shape[0], nq, nq)), axis = 1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 38,
   "metadata": {},
   "outputs": [],
   "source": [
    "quan_true_bel1 = (lat[:, 0] + 1/2)/nq\n",
    "quan_true_bel2 = (lat[:, 1] + 1/2)/nq"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 39,
   "metadata": {},
   "outputs": [],
   "source": [
    "belief1_est_MAP = (np.argmax(belief1_est, axis = 1) + 0.5)/nq\n",
    "belief2_est_MAP = (np.argmax(belief2_est, axis = 1) + 0.5)/nq\n",
    "\n",
    "belief1_est_EXP = belief1_est.dot(belief_array)\n",
    "belief2_est_EXP = belief2_est.dot(belief_array)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 40,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1440x288 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "showT_start = 0\n",
    "showT_end = showT_start+150\n",
    "#fig = plt.figure(figsize= (20, 4))\n",
    "fig, ax = plt.subplots(2,1, figsize= (20, 4))\n",
    "#ax[0] = fig.add_subplot(211)\n",
    "ax[0].imshow(np.flipud(belief1_est[showT_start:showT_end].T), interpolation='Nearest', cmap='gray')\n",
    "ax[0].plot(nq - 1 - (quan_true_bel1[showT_start:showT_end] * nq - .5), color = 'b',marker ='.', markersize = 10)\n",
    "ax[0].set(title = 'belief of box 1 based on estimated parameters', ylabel = 'belief 1')\n",
    "ax[0].set_yticks([0,1,2,3,4,5,6,7,8,9])\n",
    "#ax[0].set_ylim(ax[0].get_ylim()[::-1])\n",
    "labels = [item.get_text() for item in ax[0].get_yticklabels()]\n",
    "labels[0] = '1'\n",
    "labels[-1] = '0'\n",
    "ax[0].set_yticklabels(labels)\n",
    "\n",
    "\n",
    "ax[1].imshow(np.flipud(belief2_est[showT_start:showT_end].T), interpolation='Nearest', cmap='gray')\n",
    "pl1 = ax[1].plot(nq - 1 - (quan_true_bel2[showT_start:showT_end] * nq - .5), color = 'b',marker ='.', markersize = 10)\n",
    "#pl2 = ax[1].plot(nq - 1 - (lat[showT_start:showT_end, 1] * nq - 0.5), color = 'c',marker ='.', markersize = 10)\n",
    "ax[1].set(title = 'belief of box 2 based on estimated parameters', ylabel = 'belief 2')\n",
    "#ax[1].set_ylim(ax[1].get_ylim()[::-1])\n",
    "ax[1].set_yticks([0,1,2,3,4,5,6,7,8,9])\n",
    "#ax[0].set_ylim(ax[0].get_ylim()[::-1])\n",
    "labels = [item.get_text() for item in ax[1].get_yticklabels()]\n",
    "labels[0] = '1'\n",
    "labels[-1] = '0'\n",
    "ax[1].set_yticklabels(labels)\n",
    "plt.show()\n",
    "\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 42,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1152x576 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "showT_start = 150\n",
    "showT_end = 300\n",
    "\n",
    "fig, ax = plt.subplots(2,1, figsize= (16, 8))\n",
    "ax[0].plot(quan_true_bel1[showT_start:showT_end], color = 'b',marker ='.', \n",
    "           markersize = 10, label = 'POMDP belief1')\n",
    "#ax[0].plot(belief2_est[showT_start:showT_end].dot(belief_array), color = 'r',marker ='.', markersize = 10)\n",
    "ax[0].plot(belief1_est_MAP[showT_start:showT_end], \n",
    "           color = 'r',marker ='.', markersize = 10, label = 'belief1 from IRC (MAP)')\n",
    "ax[0].plot(belief1_est_EXP[showT_start:showT_end], \n",
    "           color = 'g',marker ='.', markersize = 10, label = 'belief1 from IRC (EXP)')\n",
    "ax[0].set(title = 'Comparison of beliefs of NN and from IRC', ylabel = 'belief 1')\n",
    "ax[0].legend()\n",
    "\n",
    "\n",
    "ax[1].plot(quan_true_bel2[showT_start:showT_end], color = 'b',marker ='.', \n",
    "           markersize = 10, label = 'POMDP belief2')\n",
    "ax[1].plot(belief2_est_MAP[showT_start:showT_end], \n",
    "           color = 'r',marker ='.', markersize = 10, label = 'belief2 from IRC (MAP)')\n",
    "ax[1].plot(belief2_est_EXP[showT_start:showT_end], \n",
    "           color = 'g',marker ='.', markersize = 10, label = 'belief2 from IRC (EXP)')\n",
    "\n",
    "ax[1].set(ylabel = 'belief 2')\n",
    "\n",
    "\n",
    "ax[1].legend()\n",
    "\n",
    "\n",
    "plt.show()\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 43,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.colorbar.Colorbar at 0x11b042908>"
      ]
     },
     "execution_count": 43,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 576x576 with 3 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "mutual1 = np.zeros((nq, nq))\n",
    "for i in range(nq):\n",
    "    index = np.where(lat[:, 0]==i)\n",
    "    #mutual1[i] = np.sum(belief1_allest.T[:, index[0].tolist()], axis = 1)/np.size(index)\n",
    "    mutual1[:, i] = np.sum(belief1_est.T[:, index[0].tolist()], axis = 1) /np.size(index)\n",
    "    mutual1[:, i] = mutual1[:, i] / np.sum(mutual1[:, i] )\n",
    "\n",
    "mutual2 = np.zeros((nq,nq))\n",
    "for i in range(nq):\n",
    "    index = np.where(lat[:, 1]==i)\n",
    "    mutual2[:, i] = np.sum(belief2_est.T[:, index[0].tolist()], axis = 1) /np.size(index)\n",
    "    mutual2[:, i] = mutual2[:, i] / np.sum(mutual2[:, i] )\n",
    "    \n",
    "\n",
    "fig_beliefdist, ax = plt.subplots(1, 2, figsize=(8,8), sharex=True)\n",
    "fig_beliefdist.text(0.12, 0.8, 'Average distribution of the posterior given the true belief', fontsize = 16)\n",
    "\n",
    "ax[0].imshow(mutual1, origin='lower', extent=[0,1,0,1], vmin=0, vmax=1)\n",
    "ax[0].set_ylabel('average distribution \\n of the belief posterior', fontsize = 14)\n",
    "#ax[0].set_xlabel('true belief', fontsize = 12)\n",
    "for index, label in enumerate(ax[0].xaxis.get_ticklabels()):\n",
    "    if index != 0 and index != nq:\n",
    "        label.set_visible(False)\n",
    "for index, label in enumerate(ax[0].yaxis.get_ticklabels()):\n",
    "    if index != 0 and index != nq:\n",
    "        label.set_visible(False)\n",
    "ax[0].xaxis.set_major_formatter(FormatStrFormatter('%.0f'))        \n",
    "ax[0].yaxis.set_major_formatter(FormatStrFormatter('%.0f'))\n",
    "ax[0].set_title('box 1', fontsize = 14)\n",
    "\n",
    "\n",
    "im = ax[1].imshow(mutual2, origin='lower', extent=[0,1,0,1], vmin=0, vmax=1)\n",
    "#ax[1].set_ylabel('average distribution \\n of the belief posterior', fontsize = 12)\n",
    "#ax[1].set_xlabel('true belief', fontsize = 12)\n",
    "for index, label in enumerate(ax[1].xaxis.get_ticklabels()):\n",
    "    if index != 0 and index != nq:\n",
    "        label.set_visible(False)\n",
    "for index, label in enumerate(ax[1].yaxis.get_ticklabels()):\n",
    "    if index != 0 and index != nq:\n",
    "        label.set_visible(False)\n",
    "ax[1].xaxis.set_major_formatter(FormatStrFormatter('%.0f'))        \n",
    "ax[1].yaxis.set_major_formatter(FormatStrFormatter('%.0f'))\n",
    "ax[1].set_title('box 2', fontsize = 14)\n",
    "\n",
    "fig_beliefdist.colorbar(im, ax=ax.ravel().tolist(), shrink=0.5)\n",
    "# fig_beliefdist.subplots_adjust(right=0.8)\n",
    "# cbar_ax = fig.add_axes([0.85, 0.3, 0.03, 0.4])\n",
    "# fig_beliefdist.colorbar(im, cax=cbar_ax)\n",
    "# fig_beliefdist.text(0.4, 0.2, 'true belief', fontsize = 14)\n",
    "#plt.tight_layout()\n",
    "\n",
    "\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Average Posterior"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "selec_idx = list(range(7)) + list(range(8,11))\n",
    "N_para_test = 10\n",
    "\n",
    "mean_para = point_all[-1]\n",
    "mean_para_woCol = np.concatenate((mean_para[0:7], mean_para[8:]))\n",
    "var_para_woCol = np.linalg.inv(-hes_final[np.ix_(selec_idx, selec_idx)])\n",
    "para_sample_woCol = np.random.multivariate_normal(mean_para_woCol, var_para_woCol, N_para_test)\n",
    "\n",
    "para_sample = []\n",
    "for i, para_temp_woCol in enumerate(para_sample_woCol):\n",
    "    para_temp = np.insert(para_temp_woCol, 7, 5)\n",
    "    para_sample.append(para_temp)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "belief1_test= np.zeros((N_para_test, 20000, nq))\n",
    "belief2_test= np.zeros((N_para_test, 20000, nq))\n",
    "\n",
    "\n",
    "for i, para_test in enumerate(para_sample):\n",
    "    twoboxCol_test = twoboxColMDP(discount, nq, nr, na, nl, para_test)\n",
    "    twoboxCol_test.setupMDP()\n",
    "    twoboxCol_test.solveMDP_sfm()\n",
    "    ThA = twoboxCol_test.ThA\n",
    "    policy = twoboxCol_test.softpolicy\n",
    "    pi = np.ones(nq * nq)/ nq /nq  # initialize the estimation of the belief state\n",
    "    Trans_hybrid_obs12 = twoboxCol_test.Trans_hybrid_obs12\n",
    "    Obs_emis_trans1 = twoboxCol_test.Obs_emis_trans1\n",
    "    Obs_emis_trans2 = twoboxCol_test.Obs_emis_trans2\n",
    "    twoboxColHMM_test = HMMtwoboxCol(ThA, policy, Trans_hybrid_obs12, Obs_emis_trans1, Obs_emis_trans2, pi, Ncol)\n",
    "\n",
    "    alpha_test, scale_test = twoboxColHMM_test.forward_scale(obs)\n",
    "    beta_test = twoboxColHMM_test.backward_scale(obs, scale_test)\n",
    "    gamma_test = twoboxColHMM_test.compute_gamma(alpha_test, beta_test)\n",
    "    xi_test = twoboxColHMM_test.compute_xi(alpha_test, beta_test, obs)\n",
    "\n",
    "    belief1_test[i] = np.sum(np.reshape(gamma_test[:, :].T, (-1, nq, nq)), axis = 2)\n",
    "    belief2_test[i] = np.sum(np.reshape(gamma_test[:, :].T, (-1, nq, nq)), axis = 1)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "fig, ax = plt.subplots(1,1, figsize= (16, 8))\n",
    "plt.imshow(np.flipud(np.mean(belief1_test, axis = 0).T[:, :100]), vmin = 0, vmax = 1)\n",
    "plt.plot(nq - 1 - (quan_true_bel1[:100] * nq - .5), color = 'b',marker ='.', markersize = 1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "para_sample = []\n",
    "for i in range(N_para_test):\n",
    "    para_temp = mean_para + (np.random.rand(len(mean_para)) * 2 - 1) * 0.05\n",
    "    para_temp[7] = Numcol\n",
    "    para_sample.append(para_temp)\n",
    "\n",
    "belief1_test1= np.zeros((N_para_test, 20000, nq))\n",
    "belief2_test1= np.zeros((N_para_test, 20000, nq))\n",
    "\n",
    "\n",
    "for i, para_test in enumerate(para_sample):\n",
    "    twoboxCol_test = twoboxColMDP(discount, nq, nr, na, nl, para_test)\n",
    "    twoboxCol_test.setupMDP()\n",
    "    twoboxCol_test.solveMDP_sfm()\n",
    "    ThA = twoboxCol_test.ThA\n",
    "    policy = twoboxCol_test.softpolicy\n",
    "    pi = np.ones(nq * nq)/ nq /nq  # initialize the estimation of the belief state\n",
    "    Trans_hybrid_obs12 = twoboxCol_test.Trans_hybrid_obs12\n",
    "    Obs_emis_trans1 = twoboxCol_test.Obs_emis_trans1\n",
    "    Obs_emis_trans2 = twoboxCol_test.Obs_emis_trans2\n",
    "    twoboxColHMM_test = HMMtwoboxCol(ThA, policy, Trans_hybrid_obs12, Obs_emis_trans1, Obs_emis_trans2, pi, Ncol)\n",
    "\n",
    "    alpha_test, scale_test = twoboxColHMM_test.forward_scale(obs)\n",
    "    beta_test = twoboxColHMM_test.backward_scale(obs, scale_test)\n",
    "    gamma_test = twoboxColHMM_test.compute_gamma(alpha_test, beta_test)\n",
    "    xi_test = twoboxColHMM_test.compute_xi(alpha_test, beta_test, obs)\n",
    "\n",
    "    belief1_test1[i] = np.sum(np.reshape(gamma_test[:, :].T, (-1, nq, nq)), axis = 2)\n",
    "    belief2_test1[i] = np.sum(np.reshape(gamma_test[:, :].T, (-1, nq, nq)), axis = 1)\n",
    "\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "fig, ax = plt.subplots(1,1, figsize= (16, 8))\n",
    "plt.imshow(np.flipud(np.mean(belief1_test1, axis = 0).T[:, :100]), vmin = 0, vmax = 1)\n",
    "plt.plot(nq - 1 - (quan_true_bel1[:100] * nq - .5), color = 'b',marker ='.', markersize = 1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "fig, ax = plt.subplots(1,1, figsize= (16, 8))\n",
    "plt.imshow(np.flipud(belief1_est.T[:, :100]), vmin = 0, vmax = 1)\n",
    "plt.plot(nq - 1 - (quan_true_bel1[:100] * nq - .5), color = 'b',marker ='.', markersize = 1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# save simulation results"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 44,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "FormatStrFormatter\t HMMtwoboxCol\t L\t LL_add_gra\t LL_slice\t Loglikelihood\t Loglikelihood_original\t MDP\t N\t \n",
      "N1\t N2\t NN_act_flatten\t NN_act_hist\t Ncol\t Numcol\t Obs_emis_trans1\t Obs_emis_trans2\t PCA\t \n",
      "POMDP_act_flatten\t POMDP_act_hist\t Qaux1\t Qaux2\t Qaux3\t QfromV\t QfromV_pi\t ThA\t Trans_hybrid_obs12\t \n",
      "ValueIteration_opZW\t ValueIteration_sfmZW\t a0\t act\t alpha\t alpha_est\t alpha_rate\t ax\t belief1_est\t \n",
      "belief1_est_EXP\t belief1_est_MAP\t belief2_est\t belief2_est_EXP\t belief2_est_MAP\t beliefTransitionMatrix\t beliefTransitionMatrixGaussian\t beliefTransitionMatrixGaussianCol\t beliefTransitionMatrixGaussianDerivative\t \n",
      "belief_array\t beta_est\t binom\t block_diag\t cb\t cbar_ax\t cmap_col5\t col1\t col2\t \n",
      "col2im\t cs3\t dataN_pkl\t dataN_pkl1\t dataN_pkl_file\t dataN_pkl_file1\t datestring_NNagent\t datestring_data\t datestring_train\t \n",
      "datetime\t discount\t division\t expm\t fig\t fig_NNtraining\t fig_beliefdist\t fig_correlation\t fig_parameters\t \n",
      "find_closest\t find_closest_array\t g0\t g1\t g2\t gamma_est\t gridspec\t hes_final\t i\t \n",
      "idx\t im\t im2col\t ind\t index\t j\t kronn\t l\t lab\t \n",
      "label\t labels\t lat\t lat_IRC\t learnrate\t leg\t leg_c\t loc\t max_ll_last\t \n",
      "max_ll_new_temp\t max_ll_temp\t max_point\t mpl\t mutual1\t mutual2\t na\t new_point\t nl\t \n",
      "norm\t np\t nq\t nr\t obs\t obs_IRC\t optimize\t os\t p_last\t \n",
      "para\t paraHessian\t para_perb\t para_slice\t para_slicePoints\t para_temp\t parametersAgent\t parametersExp\t path\t \n",
      "pb\t pca\t perturb\t pi\t pickle\t pl1\t plt\t point\t pointIni\t \n",
      "point_2d\t point_add_gra\t point_all\t point_final\t policy\t projectionMat\t projectionMatRand\t quad\t quan_true_bel1\t \n",
      "quan_true_bel2\t randProjNum\t random_projection\t rects1\t rects2\t reversekron\t rew\t rgb\t rmv_dup_arrary\t \n",
      "scale_est\t selec_idx\t showT_end\t showT_start\t softmax\t sqrt\t std\t step1\t step2\t \n",
      "tensorsum\t tensorsumm\t text\t toeplitz\t transformer\t true_2d\t twoboxCol\t twoboxColHMM\t twoboxColHMM_test\t \n",
      "twoboxColMDP\t twoboxColMDP_der\t twoboxColMDPdata\t twoboxCol_test\t twoboxd\t twoboxd1st\t twoboxd1st_pert\t twoboxd_perb\t uOffset\t \n",
      "uValue\t uValuemesh\t uniques\t v\t vOffset\t vValue\t vValuemesh\t var_para_woCol\t w\t \n",
      "width\t xi_est\t yvals\t zvals\t \n"
     ]
    }
   ],
   "source": [
    "%who\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "datestring_IRC = datetime.strftime(datetime.now(), '%m%d%Y')  # current time used to set file name\n",
    "datestring_IRC"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 45,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "'/Users/zhengwei/Dropbox (Personal)/BCM/_Code/pomdpwithnn/Results/01262020(151953)_data01262020(150900)_agentNN01262020(170602)_IRC2_twoboxCol'"
      ]
     },
     "execution_count": 45,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "path + '/Results/' + datestring_train + '_data' + datestring_data +  '_agentNN' + datestring_NNagent + '_IRC'+ str(idx) +'_twoboxCol'"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 46,
   "metadata": {},
   "outputs": [],
   "source": [
    "IRC_dict = {'Loglikelihood': Loglikelihood,\n",
    "            'Qaux1': Qaux1,\n",
    "            'Qaux2': Qaux2,\n",
    "            'Qaux3': Qaux3,\n",
    "            'Loglikelihood': Loglikelihood,\n",
    "            'Loglikelihood_original': Loglikelihood_original,\n",
    "            'uValue': uValue,\n",
    "            'uValuemesh': uValuemesh,\n",
    "            'vValue': vValue,\n",
    "            'vValuemesh': vValuemesh,\n",
    "            'point': point,\n",
    "            'point_add_gra': point_add_gra,\n",
    "            'LL_add_gra': LL_add_gra,\n",
    "            'point_all': point_all,\n",
    "            'point_final' : point_final,\n",
    "            'uniques': uniques,\n",
    "            'projectionMat': projectionMat,\n",
    "            'belief1_est': belief1_est,\n",
    "            'belief2_est': belief2_est,\n",
    "            'belief1_est_MAP': belief1_est_MAP,\n",
    "            'belief2_est_MAP': belief2_est_MAP,\n",
    "            'belief1_est_EXP': belief1_est_EXP,\n",
    "            'belief2_est_EXP': belief2_est_EXP,\n",
    "            'quan_true_bel1_POMDP': quan_true_bel1,\n",
    "            'quan_true_bel2_POMDP': quan_true_bel2,\n",
    "            'hes_final': hes_final,\n",
    "            'var_para_woCol': var_para_woCol,\n",
    "            'std_para_woCol': std,\n",
    "            'LL_add_gra': LL_add_gra\n",
    "            }\n",
    "\n",
    "para_output = open(path + '/Results/' + datestring_train + '_data' + datestring_data +  \n",
    "                   '_agentNN' + datestring_NNagent + '_' + datestring_IRC+ 'IRC'+ str(idx) +'_twoboxCol' + '.pkl', 'wb')\n",
    "pickle.dump(IRC_dict, para_output)\n",
    "para_output.close()"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.8.5"
  },
  "toc": {
   "base_numbering": 1,
   "nav_menu": {},
   "number_sections": true,
   "sideBar": true,
   "skip_h1_title": false,
   "title_cell": "Table of Contents",
   "title_sidebar": "Contents",
   "toc_cell": true,
   "toc_position": {},
   "toc_section_display": true,
   "toc_window_display": false
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}