jupyter-notebooks/QoIAwareProjection-MSE-loss-convergence.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "<a id=top-page></a>\n",
    "# Inspect convergence of the MSE loss\n",
    "\n",
    "### Table of contents\n",
    "\n",
    "- [**Load the training data**](#load-data)\n",
    "- [**Upload the MSE losses**](#upload-MSE-loss)\n",
    "- [**Visualize the MSE loss**](#visualize-MSE-loss)\n",
    "\n",
    "***"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "%run -i imports.py\n",
    "%matplotlib inline"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [],
   "source": [
    "plot_save_location = '../figures/'\n",
    "MSE_loss_directory = '../results-MSE-loss/'"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [],
   "source": [
    "n_components = 2"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "<a id='load-data'></a>\n",
    "***\n",
    "\n",
    "## Load the training data\n",
    "\n",
    "[Go to the top ↑](#top-page)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Select one dataset:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [],
   "source": [
    "data_tag = 'H2-air'\n",
    "# data_tag = 'CH4-air'\n",
    "# data_tag = 'C2H4-air'"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      "The data set has 58101 observations.\n",
      "\n",
      "The data set has 9 variables.\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "array(['T', 'H2', 'OH', 'H2O', 'O2', 'H2O2'], dtype=object)"
      ]
     },
     "execution_count": 5,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "########################################################################\n",
    "## Load data - Hydrogen/air flamelet\n",
    "########################################################################\n",
    "\n",
    "if data_tag == 'H2-air':\n",
    "    \n",
    "    state_space = pd.read_csv('../data/' + data_tag + '-state-space.csv', sep = ',', header=None).to_numpy()[:,0:-2]\n",
    "    state_space_sources = pd.read_csv('../data/' + data_tag + '-state-space-sources.csv', sep = ',', header=None).to_numpy()[:,0:-2]\n",
    "    state_space_names = pd.read_csv('../data/' + data_tag + '-state-space-names.csv', sep = ',', header=None).to_numpy().ravel()[0:-2]\n",
    "\n",
    "    selected_state_variables = [0, 2, 4, 5, 6, 8]\n",
    "\n",
    "########################################################################\n",
    "## Load data - Syngas/air flamelet\n",
    "########################################################################\n",
    "\n",
    "if data_tag == 'CO-H2-air':\n",
    "    \n",
    "    state_space = pd.read_csv('../data/' + data_tag + '-state-space.csv', sep = ',', header=None).to_numpy()[:,0:-1]\n",
    "    state_space_sources = pd.read_csv('../data/' + data_tag + '-state-space-sources.csv', sep = ',', header=None).to_numpy()[:,0:-1]\n",
    "    state_space_names = pd.read_csv('../data/' + data_tag + '-state-space-names.csv', sep = ',', header=None).to_numpy().ravel()[0:-1]\n",
    "\n",
    "    (n_observations, n_variables) = np.shape(state_space)\n",
    "\n",
    "    print('\\nThe data set has ' + str(n_observations) + ' observations.')\n",
    "    print('\\nThe data set has ' + str(n_variables) + ' variables.')\n",
    "\n",
    "    selected_state_variables = [0, 1, 2, 4, 5, 8, 9]\n",
    "\n",
    "########################################################################\n",
    "## Load data - Methane/air flamelet\n",
    "########################################################################\n",
    "\n",
    "if data_tag == 'CH4-air':\n",
    "   \n",
    "    state_space = pd.read_csv('../data/' + data_tag + '-state-space.csv', sep = ',', header=None).to_numpy()[:,0:-1]\n",
    "    state_space_sources = pd.read_csv('../data/' + data_tag + '-state-space-sources.csv', sep = ',', header=None).to_numpy()[:,0:-1]\n",
    "    state_space_names = pd.read_csv('../data/' + data_tag + '-state-space-names.csv', sep = ',', header=None).to_numpy().ravel()[0:-1]\n",
    "\n",
    "    species_to_remove = 'N2'\n",
    "    (species_index, ) = np.where(state_space_names==species_to_remove)\n",
    "    state_space = np.delete(state_space, np.s_[species_index], axis=1)\n",
    "    state_space_sources = np.delete(state_space_sources, np.s_[species_index], axis=1)\n",
    "    state_space_names = np.delete(state_space_names, np.s_[species_index])\n",
    "\n",
    "    selected_state_variables = [0, 4, 5, 6, 14, 16]\n",
    "\n",
    "########################################################################\n",
    "## Load data - Ethylene/air flamelet\n",
    "########################################################################\n",
    "\n",
    "if data_tag == 'C2H4-air':\n",
    "   \n",
    "    state_space = pd.read_csv('../data/' + data_tag + '-state-space.csv', sep = ',', header=None).to_numpy()[:,0:-1]\n",
    "    state_space_sources = pd.read_csv('../data/' + data_tag + '-state-space-sources.csv', sep = ',', header=None).to_numpy()[:,0:-1]\n",
    "    state_space_names = pd.read_csv('../data/' + data_tag + '-state-space-names.csv', sep = ',', header=None).to_numpy().ravel()[0:-1]\n",
    "\n",
    "    selected_state_variables = [0, 4, 5, 6, 15, 22]\n",
    "\n",
    "(n_observations, n_variables) = np.shape(state_space)\n",
    "\n",
    "print('\\nThe data set has ' + str(n_observations) + ' observations.')\n",
    "print('\\nThe data set has ' + str(n_variables) + ' variables.')\n",
    "\n",
    "state_space_names[selected_state_variables]"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "<a id='upload-MSE-loss'></a>\n",
    "***\n",
    "\n",
    "## Upload the MSE losses\n",
    "\n",
    "[Go to the top ↑](#top-page)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "2D-LDM-lr-0.001-bs-58101-n-epochs-20000-architecture\n"
     ]
    }
   ],
   "source": [
    "random_seeds_list = [i for i in range(20,40)]\n",
    "if n_components == 2: decoder_architecture = (6,9,10)\n",
    "if n_components == 3: decoder_architecture = (7,10,11)\n",
    "learning_rate = 0.001\n",
    "batch_size = n_observations\n",
    "n_epochs = 20000\n",
    "case_name = str(n_components) + 'D-LDM-lr-' + str(learning_rate) + '-bs-' + str(batch_size) + '-n-epochs-' + str(n_epochs) + '-architecture'\n",
    "print(case_name)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [],
   "source": [
    "architecture_without = str(n_variables) + '-' + str(n_components) + '-' + '-'.join([str(i) for i in decoder_architecture]) + '-' + str(len(selected_state_variables))\n",
    "architecture_with = str(n_variables) + '-' + str(n_components) + '-' + '-'.join([str(i) for i in decoder_architecture]) + '-' + str(len(selected_state_variables) + n_components*2)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "'9-2-6-9-10-10'"
      ]
     },
     "execution_count": 8,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "architecture_with"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [],
   "source": [
    "MSE_losses_without = np.zeros((n_epochs, len(random_seeds_list)))\n",
    "MSE_losses_with = np.zeros((n_epochs, len(random_seeds_list)))\n",
    "\n",
    "for i, random_seed in enumerate(random_seeds_list):\n",
    "\n",
    "    # Not including projection-dependent QoIs: - - - - - - - - - - - - - - - - - - - - - - - - - -\n",
    "\n",
    "    filename = MSE_loss_directory + 'QoIAwareProjection-MSE-training-losses-without-Seta-' + data_tag + '-SLF-' + case_name + '-' + architecture_without + '-random-seed-' + str(random_seed) + '.csv'\n",
    "    \n",
    "    MSE_loss_without = pd.read_csv(filename, sep = ',', header=None).to_numpy().ravel()\n",
    "\n",
    "    MSE_losses_without[:,i] = MSE_loss_without\n",
    "\n",
    "    # Including projection-dependent QoIs: - - - - - - - - - - - - - - - - - - - - - - - - - - - -\n",
    "    \n",
    "    filename = MSE_loss_directory + 'QoIAwareProjection-MSE-training-losses-with-Seta-' + data_tag + '-SLF-' + case_name + '-' + architecture_with + '-random-seed-' + str(random_seed) + '.csv'\n",
    "    \n",
    "    MSE_loss_with = pd.read_csv(filename, sep = ',', header=None).to_numpy().ravel()\n",
    "\n",
    "    MSE_losses_with[:,i] = MSE_loss_with"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "<a id='visualize-MSE-loss'></a>\n",
    "***\n",
    "\n",
    "## Visualize the MSE losses\n",
    "\n",
    "[Go to the top ↑](#top-page)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [],
   "source": [
    "n_epochs_list = [i for i in range(0,n_epochs)]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAABNwAAADdCAYAAABg32+mAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjcuMiwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy8pXeV/AAAACXBIWXMAAA9hAAAPYQGoP6dpAAC/jUlEQVR4nOzdd3wc9Z3/8dfMbF/13mVJrnLFBQwGg+mE5JIQIAmXQCoh53CX/sslubtwKZcLpENIuCROgCQkBEij2oANrrg32ZZt9d61K23fmd8fs1pJlm1kWcXl83w8FGlnd2e+Kwnnq/d+vt+PYhiGgRBCCCGEEEIIIYQQYlyoUz0AIYQQQgghhBBCCCEuJBK4CSGEEEIIIYQQQggxjiRwE0IIIYQQQgghhBBiHEngJoQQQgghhBBCCCHEOJLATQghhBBCCCGEEEKIcSSBmxBCCCGEEEIIIYQQ40gCNyGEEEIIIYQQQgghxpEEbkIIIYQQQgghhBBCjCMJ3E7DMAw8Hg+GYUz1UIQQQgghxDiSeZ4QQgghJpIEbqfh9XpJTk7G6/VO9VCEEEIIIcQ4knmeEEIIISaSBG5CCCGEEEIIIYQQQowjCdyEEEIIIcRF45FHHqG8vJxly5ZN9VCEEEIIcQFTDNm44pQ8Hg/Jycn09vaSlJQ01cMRQgghhBDjROZ5QgghhJhIUuEmhBBCCCEuGlLhJoQQQojJIBVupzHR73wahoFhGKiq5J5CCCGEEJNpMuZ5GAaKzPOEEEKIi5LMAKZIj7efvJs+xfMbd031UIQQQgghxDgKh0OEnvwGoRceneqhCCGEEGKKSOA2RawWjVA4wqHqxqkeihBCCCHERWMylpQqimJ+0d02YdcQQgghxLlNArcp0tnbh24YvLb9wFQPRQghhBDiorF69WoqKirYvn37hF1DNmwRQgghhARuUyQYCgNQUdUwxSMRQgghhBDjydD1qR6CEEIIIaaYBG5TxNPvA8DrC0zxSIQQQgghxHjSNG2qhyCEEEKIKSaB2xTxB0JTPQQhhBBCiIvOpOzhpirxr/VoZMKuI4QQQohzlwRuU6Q/IJVtQgghhBCTbTL2cFOUwSl2tK12wq4jhBBCiHOXBG5TZFpeVvxrQ3bWFUIIIYS4YMS7lAL6sd1TOBIhhBBCTBUJ3KZIWlJC/Ov2bs8UjkQIIYQQQkyY1pqpHoEQQgghpoAEblNEUwc30/37hp1TOBIhhBBCCDFh/N6pHoEQQgghpoAEblMkLXmwwu2pVzZN4UiEEEIIIS4ek9E0QQghhBBCArdzwOGaxqkeghBCCCHERWEymiYAMGQ1gxBCCCEuPhK4nQP6/MGpHoIQQgghhBhPienxLw1Dn8KBCCGEEGIqSOAmhBBCCCHEeCsuj3+pe3umbhxCCCGEmBISuE0hizr47TcMYwpHIoQQQgghxpNasjD+tdFaPYUjEUIIIcRUkMBtCmWkJMa/bu7ombqBCCGEEEKIcWVJTI1/HW2WwE0IIYS42EjgNoUWzCyOf11RVT+FIxFCCCGEEONJUZTBG+0yzxNCCCEuNhK4TaErF82Kf/3S5r1TOBIhhBBCiIvDI488Qnl5OcuWLZu8i/o9k3ctIYQQQpwTJHCbQlddMif+9YZdFVM4EiGEEEKIi8Pq1aupqKhg+/btk3dR6VIqhBBCXHQkcJtCM4vz4l/Xt3RM4UiEEEIIIcREkgZZQgghxMVFArcp5HLY418Hw5EpHIkQQgghxPntL3/5Cx/96EenehjDDd3HLeibunEIIYQQYtJJ4HYO6fXKREwIIYQQ4kx96Utf4itf+cq5V0XmSIp/qfe0TeFAhBBCCDHZJHCbYlZt8EfwyjZpnCCEEEIIcaYuvfRSHn300akexkgF0+NfRmplv14hhBDiYiKB2xS7enF5/Ovv/OrZKRyJEEIIIcS567HHHmPevHnDPpqamgC44447UIYu3zxHqDOGdEKt2jNl4xBCCCHE5LNM9QAudp+/+12s234AgJrmDjp7PaQnJ73Ns4QQQgghLi733nsv995771QP44xYUrMJDdyIhIjWH0YrnD2VQxJCCCHEJLmoKtzOxc10Ly2fPuz2e77w0BSNRAghhBBCjCdFHT7Vjrz1AoauT9FohBBCCDGZLprA7VzdTFfTVBw2a/z2gWP1bNpzeApHJIQQQggx8YLBIPPmzWP9+vXxY4FAgI9//OOkpKSQm5vL97///akb4ETw9WJ01E/1KIQQQggxCS6awO2c3UwXWLFw1rDb3/3NX6ZmIEIIIYQQkyAQCPDBD36QgwcPDjv+pS99iR07dvDaa6/xs5/9jAceeIA///nPozrnNddcw29+85sJGO1ZsjmG3Yw2VE7RQIQQQggxmS6owO183EwX4KPvXjXs9sY9R2hs65qi0QghhBBCTJyKigqWL1/O8ePHhx3v7+/nl7/8JT/+8Y9ZvHgx733ve/nyl7/Mww8/PC7XDQaDeDyeYR+TIrt02E295sDkXFcIIYQQU+qCCtzuvfdeDhw4MOwjLy9vqof1topzM0cce+bVrVMwEiGEEEKIibVhwwZWrVrFli1bhh3fu3cv4XCYK664In7syiuvZNu2bejjsO/Z//zP/5CcnBz/KCwsPOtzjoZaOn/4gf4e9HDo5A8WQgghxAXjggrczlc56ckjjv3h5U1TMBIhhBBCiIn16U9/mh/+8Ie4XK5hx5ubm8nIyMBms8WPZWdnEwgE6OzsPOvr/vu//zu9vb3xj/r6ydlLzZJbNuKYXrl9Uq4thBBCiKlzVoFbc3Nz/Ovq6mq+/OUv87WvfY3q6uqzHhhcPJvppiUlcOJi18raZvp8gSkZjxBCCCHERM/zTuTz+bDb7cOODdwOBoNnfX673U5SUhJPPPEEy5cv57rrrjvrc46GYrWPOBY98takXFsIIYQQU8cylid1d3dz6623Eg6H2b59Oz09PVx++eW0t7djGAY///nP2bp1KzNmzBjzwAKBAHfddddpN9Otra3lnnvuobi4mNtvv/1tz3nNNddwzTXXjHlME0VVVbLSk2nt7I0f0w2D//vLq3zurluncGRCCCGEuNhMxjzvZBwOx4hgbeD2idVwZ2P16tWsXr0aj8dDcvLIVQYTwmKDyJBlpP09GNEIijamqbgQQgghzgNjqnB74IEH2LZtG1dffTUAjz/+OG1tbXzzm9/k5Zdfxmq18s1vfnPMg7roNtMFCrPSRxxb89fXMQxj0sYghBBCCDHR87xTyc/Pp6Ojg0gkEj/W0tKC0+kkJSVl3K7zyCOPUF5ezrJly8btnG8rJXvEIb3+8ORdXwghhBCTbkyB29///nc+8YlP8NBDDwHw/PPPk5KSwle+8hVuuOEGPv3pT/Pqq6+OeVAX22a6ALmZqdht1mHH6ls72br/6KSNQQghhBBioud5p7Jo0SKsVitbtw42jtq4cSPLli1DVcdv2+HVq1dTUVHB9u2Tt4+aWjhrxLHIvg2Tdn0hhBBCTL4xzV4aGxu57LLLAAiFQmzcuJGrr746PhkqLCykq6trzIO62DbTBSjKycBhHbms4IkX3pi0MQghhBBCTPQ871RcLhf33HMP9913H9u3b+cvf/kLDz30EP/2b/82rteZigo3Lf8ky29729A9Zz9/FUIIIcS5aUyBW0ZGBr295n5jb775Jn6/n+uvvz5+f1VVFVlZWeMzwiEmazPdoR+TpSQ/iz7/yNfwp1c2EwyFJ20cQgghhLi4TdU8D+AHP/gBS5YsYdWqVaxevZoHHniA2267bVyvMRUVbkpKFqjaiOORDU9hGGe/SkMIIYQQ554xBW6XXHIJv/rVr9ixYwf/+7//i6qqvOtd7wJg+/bt/OIXv2D58uXjOlCYvM10p0JZfjbRkyyLjeoGP/z981MwIiGEEEJcjCZznmcYxrCGVi6Xi9/+9rf09fXR2NjIZz/72XG5zlRTFBWlYOSyUqOnjciWv03BiIQQQggx0cYUuH3zm9+kqamJyy67jHXr1vHRj36UoqIiXnvtNS677DLC4TBf+9rXxnusk7aZ7lQoLTA303U7R7aO/9/f/JWm9vFfuiGEEEIIcaKpmudNlilpmgBYltx00uP68d2Et/wVQ49O6niEEEIIMbHGFLgtWrSInTt38r3vfY+nnnqKX/ziFwBMnz6dT3ziE2zZsoUFCxaM60AHrjsZm+lOhYKsNOxWC4XZI7uVAsy/84v8bcMO6VoqhBBCiAk1VfO8yTIVS0oB1IQUcJ18uxL92C4iO16e1PEIIYQQYmIpxjme4CiKwuuvvx5fbnDfffexceNG1qxZQ2NjI/fccw9r1qwZ9/09ADweD8nJyfT29k7Kfm6Xf+TrZKUl8cauQ6d8jM1qYVl5Gbdfv5xVS+dSnJs54eMSQgghhLjQTPY8DyBSc4Dom0+f4l4F222fRXGnTMpYhBBCCDGxxlwW1tbWxosvvhi//de//pVLL72UFStW8Nxzz43L4E5mMjbTnSrzphfSf5LGCUNFo1E27T3C577/WxZ98MvMv/ML/MfPnqK+VbpcCSGEEGJ8TNU8bzJM1ZJSAK24/DT3GoQ3PStLS4UQQogLxJgq3CoqKrjyyivJz89n//79HDt2jPLycgzDwGq1EgqFeOmll4Z1tDofTfY7nz956kW+95u/4nDY6Ozxjrjf7bQTjepgGCyaVcLuI9VYNA1NUwlHonz/cx/mgzdfOeHjFEIIIcSFS+Z5Eyv48q+grW7kHZoFohGU9Hwsi29AzSmZtDEJIYQQYvyNqcLtv//7v9F1nf/6r/8C4PHHHycajbJ27VpaWlqYN28eDz744LgO9GIwf3oR/YEgq5aa735qJ+xL5wsECYTCLC0vY+uBo9x0+SLys9PISEnk3VcvZfX//poXNu6aiqELIYQQ4gIh87yJZZlz+cnviJpNwQxPB+G1vyG07nEMrzTNEkIIIc5XYwrcNmzYwL/+679y++23A/DSSy9RWFjINddcQ1JSEh/5yEcmfSPaC8G8skIAlpaXARDV9WH3GwYU52TgD4X52Vc+zqtv7ScciVLT3M7yBTO5+YqFfP4Hj9PnC0z62IUQQghxYbjQ53lTuaQUQM2dfuo7FcWsdAOM5uOE/vYwkcPbJmlkQgghhBhPYwrcuru7KS0tBaCnp4ddu3axatWq+P0JCQmEQqHxGeFFJDM1ibyMVBrburFYNACcNtuwxzR3dLPzUBXTC3NZ++h/0NXbR2qimz+v28p3P/PPtPd4+NPaLVMxfCGEEEJcAC70ed5UdSkdoFhtKNPmn/xO1YJid2N99/2o864CRSG6/QVCG/4ke7sJIYQQ55kxBW4FBQXU1tYC5ruehmFw4403xu/fvn07+fn54zPCi8xl86ezdX8lS2ebE12H3TLs/lAkit1m5Q8vb2ROST6Pff1eOnv72Lz3CGnJCdxyxSU88fwbUzF0IYQQQlwAZJ438SylC09+RzSM0dsGuo71kuuxvuM+AIz6CqI7X5nEEQohhBDibI0pcLvmmmv46U9/ykMPPcR//Md/4HQ6ecc73kFvby8PPvggv/nNb7j11lvHe6wXheXzZ7LnSA3XXjoPgG6vj0Uzpw17TDAU5ncvbsTnD3Lj8oVcf9kCdMPgzd2HePc1S9lTWUNjm+z5IYQQQogzJ/O8iafkloLFfvI7VQ299qD5ZUomasFscCYSPbJN9nQTQgghziNjCty++93vUlxczJe//GVqamr4/ve/T3JyMnv37uX//b//x5w5c/j3f//38R7rReHyBTMJR6KkJLjix0rys1BVZdjjQuEI7/nigwRDYR78t38G4LFn13HjZQuxaBovbt4zmcMWQgghxAVC5nkTT1E1tOmLTn6n3RkP3ADU4nLwecBiI3p0x+QMUAghhBBnbUyBW0ZGBtu3b2f79u3U1dXxqU99CoB58+bx5JNPsmXLFjIzM8d1oBeL8pICktxO2ns8uOzm/m1/f2Mnn3jPtSMeu7Oiin99cA3T8rIozE5n094jWCwaly+YwStb90720IUQQghxAbjQ53lT3TRhgFqy4OR3BPwYve3oPW3m4/JnAKCk5xGtOYhhGJM1RCGEEEKchTEFbgCqqrJkyRJycnJob2/H4/GQlpbGXXfdhdPpHM8xXlQ0TWXV0rm8+tYBrlo8B6fdRiQaZUZRLqoyvMrNMAz+tHYLr2zdy+3XLScciZpVbssX8uauQ/gCwSl6FUIIIYQ4n13I87ypbpowQEnPh4TUkXcYUVAt6LUHzMfZXSjJmSiqBfp7MGJBnBBCCCHObWMO3Do7O/n4xz9OamoqOTk5pKamkpqayic/+Uk6OjrGc4wXnZuvWMSuw9VcMruEUCQCwA+efJ4bLhve0coA8jJS+e6av3Dj5ea7pD/6/fOsWDSLQCjMm7sPT/bQhRBCCHEBkHnexFMUBW3WpSe/0+5Cr62IV7MpWUUY/d1gsaE3yPxOCCGEOB+MKXDr6enhiiuuYM2aNaSlpfGOd7yDm266iZSUFH71q1+xfPlyPB7PeI/1onHDZQtQVQVNVYlGdWxWC80d3bz/xhUjHtve3cvuIzX4gyFsVgv9/iDrtu1nWl6mLCsVQgghxBmTed7k0U7VrdTvwehtj1ezqZlFGL0dKNnT0BuPTuIIhRBCCDFWYwrcvvWtb3Hs2DEeffRRqqqq+Pvf/84LL7xAdXU1P//5z6muruZ//ud/xnusF430lERWLJzFhp0VzC0rpKwgG4BH//wKBdlpwx4bjupkpCTyh5c2say8jNyMVH761ItcsWAWr2zdJ/t8CCGEEOKMyDxv8igON0rhnFPcOditVMksND8npmG0N2AE+idriEIIIYQYozEFbs899xwf/vCH45voDnXvvffyoQ99iGefffasB3cx++dbrmLjnsNcfckc6po70FSV7RXHue+2G0Y8tt8f4G8bdvCulUtobO/CYbdSUdVAQ2sne47UTP7ghRBCCHHeknne5LLMPFXzBp1ozX4Mw0BJTAO7CxQFMNCbjk3mEIUQQggxBmMK3BoaGrj88stPef8VV1xBXV3dmAcl4J+uXkqS20mvz09/IMilc8sA2Lz3CJo6/MfmD4aJRHWC4TB5mankZqSyp7KGRJeDXzy7biqGL4QQQojzlMzzJpeSWwKu5JF3GAZ4uzA6GlEUBTWjAKOnDSUtF72xcvIHKoQQQogzMqbALTU1ldra2lPeX11dTVJS0pgHJcBpt/Gxd6/iL6+9xZI5pSiqiqLAC5v3cOm8svjjBhqXaqrKmr+t50df/AgVVY3MKcnH6wvwp7Vb2Fd56p+VEEIIIcRQF/o875FHHqG8vJxly05VWTa5FEXFsuDqU94fOb7TfFx2MUZbHUpuGXrTMQw9OllDFEIIIcQYjClwW7VqFT/72c+orBz57trhw4d59NFHueaaa852bBe9z9x5MwDJCS427z3CtcvMLqWefj8AqqIwsEVbKBKhpqmdYDDMb/97NU3t3TjtVgzD4KbPfJv1Ow5OyWsQQgghxPnlQp/nrV69moqKCrZv3z7VQ4lTSxeCxX7S+4zjezGiYbTCORANo1jtEApgtDdM8iiFEEIIcSYUYwy76h8+fJglS5ZgGAZ33XUX5eXlABw8eJDf//73qKrKtm3bmDdv3rgPeDJ5PB6Sk5Pp7e2dsndyf/LUi3zjF38iweXk9usu47f/2ICuGzjtVvzBMGBWt0V1HYCinAx2//5/ae3q5cHH/84Tz28gEjXvy8tI5VO3X8+H37GS1KSEKXk9QgghhDi3yTxvakQqNhPd+fJJ71MvvRXrrEsJvfAYWO0YPa1oJQuwLL15kkcphBBCiNEaU+AG8Oabb/Kxj32M48ePDzteWlrKL3/5y/P6nc8B58JELBKJcvP936GyrplgKMxdt1zJb/62HlUB3QBz69zhPvW+G/ju/XcB0OPt5+E/vsTDf3yJYDgSf0yy28niOaXcsuISVl4yh7KCbCwWbfJemBBCCCHOWTLPm3yGHiX0p+9BODDyTs2K7YNfRa+tIPLm06jF89BbqrC97wsommXyByuEEEKItzXmwA3AMAx2797N8ePHMQyDsrIyLrnkElR1TCtVzznnykSsqb2baz/1AO09Ht537WWs3baPHq/vtM/5+Vc/wftvXBG/res6T6/dysN/eomDVQ2c+GNXgJRENzOKc7ls7nRmFudSkpfFtLwscjNSLpifqRBCCCFG53yY5wUCAe6++25aW1sJBoP86Ec/Yvny5aN67rkyzxsqUnOQ6Jt/Oul9yvTFWC97F+F/PAoWC0ZnE5YV70MrXTDJoxRCCCHEaJxV4HahO5cmYodrGrlp9bfx9Pu597breeyE7qMWTSMSHb557vf+9UN88rbrRpwrEAyz71gtb+yqYO3W/VTWNdHb5+NUvwmaqpCdlkxhTgal+dkU5qSTn5lGftbgR6LLOW6vVQghhBBiNH72s5/R0tLCf//3f3PkyBE+8pGPsGXLllE991ya5w0VfOlX0H6KLrA5ZVjKFhLZ9Cyk5qBEw1jftRpFlVUKQgghxLlmVIHb3XfffeYnVhR++9vfjmlQ54pzbSJW09zOio9+HV8gRFl+FlVN7cMq1awWjXBkeOj2/huv4IdfuAen3Xbac+u6Tk1TO+ve2s+WfZUcPF5PfVsngdg+cUMN3TNugNtppzA7g8KcdPIyU8nLSCUvM5XcDPMjPzONpAQnykBbVSGEEEKcE87neZ7X60VRFBISEjh48CD/8i//woYNG0b13HNtnjfACPoIPfMDiI6cg8VZ7WDoEAmjXXIDlnlXTt4AhRBCCDEqowrcxrJ0QFEUotHzu135uTgRO3i8nlWfeoDIkGDtVD/AgWCsrCCbH37hHq66ZM4ZX6/H28+2A0d5fftBth04yv5j9Vg0lRlFufgCIepbO4aFfHabFYumEonqBEPDJ4pOu5W8zLRhQVxe5mAwl5eZSlZqMpp27ixVEUIIIS5058M877HHHuMnP/nJsGOvvPIKeXl5ALS3t3PTTTfx4IMPct11I6v7T+ZcnOcN0DsaCb/42Kgfr+TNQJs2DzWrCBJS5Q1OIYQQ4hwwqsCttrZ2TCcvLi4e0/POFefqROyFjbv48H88TEqimy5P32kfa7dasFg0+v1B3nnVYh741J2UFmSP+dpN7d184xd/4ul1W7nvfTfwzU/fSUtXL7VN7VQ3tVHT1E5Nc3vsdjudvd74c20WC4luBxZNQzcMgqEw/f4AUX3wV1BVFHLSU8wQbkgQl5sxWDWXl5mGw24d82sQQgghxKDzfZ5XWVnJ7bffzre//W3e9a53jfp55+o8b0C0/giR9b8/8yc6E1Gzi1GzilGyilFSMlEUeTNTCCGEmGyyh9tpnMsTsb+s384nv/kL8jJTqWvpOO1jNVVh0cxpNHf20NLZw/uuvYzPf+idzJ6WP+br/+qvr/GlHz3J+2+8nJ9+6WOn7HDq9fmpbe6gZkgY19zeTXNHNy2dPbR29aIPCdw0VcXltGO3WlAUhUgkii8QHNZhVVFgZlEeC2cWmx8zilkws1j2kRNCCCEuMvX19dxwww08/vjjXHrppWf03HN5njdAbz5OeN0TnHo9w0nYnKBZwN9nPs/mRM0sRBkI4dJypbOpEEIIMQkkcDuNc30itmnPYT7+3z+nPxCkz3eSFvIn+MwdN1GYm8FPnnqRxrYubly+kI+9exXXXzp/TMs4n3l1G/d95/+4+YqF/PI/7sNuO/Oqs0gkSlu3h5bObprbe2iKBXGDoVwvLZ099Hj7hz1PU1U0TSUSiaLHfoXTkxMpK8hmXlkBy+ZN58qFs8nPSpNlFUIIIcQ5JhgMsmTJEh5++GGuueYawOw4unr1ap555hmcTidf/OIX+cIXvnDa89x777288MILTJ8+HYDMzEyefvrpUY3hXJ/nDTC8XYReXgN+zxk8SwEMUDVwJ5uH+ntBj4JmRUnPQ80oQMksMD+7zt3XL4QQQpyvJHA7jfNhItba2ctXH/kDz762bVSPv/OG5fzkSx/j6XVb+L/nXmXf0TqKcjK4511X84EbV5CXmXpG139p8x4+8l+PsGLRLB7/7/txO+1jeRlvKxAM09plVui1xkK4ls4emju6qWpoo6Gtk87evhH7ximKGcTNLMrl8gUzueWKS1gwswirRd7ZFUIIIaZCIBDgrrvu4rnnnuP111+PB273338/b7zxBmvWrKG2tpZ77rmHX//619x+++3jct1gMEgwGIzf9ng8FBYWntPzvAGGrhM+vBVj5yucUbUbgNUBkTAYZtiGOwlF1TD8fRD0mY9xJZnBW0YsgEvLQbFOzJxOCCGEuFhI4HYa50PgNuCNXRV88UdPcrSu+W0fa7Na+OmXPsb7rruU3Udq+PVfX+e5198iFIlwzZJyPnDTCm69cjEux+gmWm/uPsRdX/0JuRkp/PI/72PBjKnb0yUUjtDQ1snW/UfZUXGcA8fqOVbfQk9fP0N/0xOcDrLSkinOzWBGUS7zygpZPLuE2dPy0LSTL48VQgghxNmpqKjgrrvuwjAM9u3bFw/c+vv7ycjI4MUXX4wHcN/61rdYt24d69evH5drf+Mb3+CBBx4Ycfx8mOcNMHSd8J5XMQ5u4oyDNzCr3awO8HaZXVA1K0pSOlhtGOEQ9HaAbm7joSSmoaTmmOFbai5qWg44E2XlgBBCCDFKEridxvkUuAEYhsGTL7zBF370BOHw23cOczlsfONTd/Chd6wkFI7w1/Xb+cPLm9i6/yiJLgfvvmYZH7hpBZfPn/G2Hcwqa5v55Ld+zqHqRv7ljpv44offRYLLMV4v7azpus6uQ9X8Y+Mutu6vpK6lgy5P/4iKOACrRSPB5SA9OZG8zFRK87OYV1bEghlFFOVkkpWWJJNNIYQQYgweffRRKisr+fa3v43b7Y4Hbps3b2blypX4fD5sNhsA69ev55ZbbqG/v39MnVRPdD5XuJ3IMAwi1fvQt/zVXCZ6plQ11lAhC6O3E6OjHsJB0CwoqdkoziSw2jH6ezC6WyAU27rE7jJDuOQM1ORMlKQMlORMcCbI3EgIIYQ4gQRup3G+BW4DfIEgd331J2zYVTGqx1s0jXtvu47P/fOtZKQkUd3YxlOvbOKplzdT19JBcW4mH7jxCj5061UUZKWf8jzBUJgf/+EFfvi750lLTuCB++7ktlWXjsskeaIEgmEa2jrYe7Q2Xg1X19JBW1cvPV4fgZMEcqqikJTgpDA7gxULZ3HV4jlcPn8GqUkJU/AKhBBCiPOToijxwO2ZZ55h9erVtLS0xO8/dOgQ5eXltLW1kZmZOW7XfeSRR3jkkUeIRqNUVlaed/O8oQzDIHJoC/pYlpoOsNhQSheiFc3B6GrBaK1Bb6uFcNDc661kAWpmEYbfi9HVjNHTitHbgeHtGgz7rA6U5IxYAGeGcEpSBkpCMoomneWFEEJcnCYkcOvt7aW2tpYFCxaM96kn1fkauA146uVNrP7fXw3rAno6CnDj5Qv55qffz4yiXHRdZ8v+ozz18ib+8vp2fMEg11+6gHvedTU3XrbglJ1Ja5vb+dojT/H8xl0smFHEf37ydq5dNu+8fOczFI5Q09TGvmN17DlSw+GaRmqb22np7B3WqEIBSguyedfKJXzkXddQnDt+fxgIIYQQ55LxmucNDdyeeOIJvv71r1NbWxu/v6qqirKyMurr6ykoKDjbYY9wvs/zhjLCIcJb/opRe+DsTuRwo85fiVZ2CUbTcfSqvehNR0HXUTILUYtmoxXOQUlMw9CjGN5uDE8HRm+7GcL1tmN4OsxquQHOBBR3CkpCyrDPJKSguJNRLLazG7MQQghxjhpV4LZ48WK+853vcPPNN8ePhcNhXnzxRVasWEF6+vCqp9/97nfcfffdRKNjKHE/h1wIE7GGtk6u+eQ36OztQ1WVUYdv0wty+NEX7+GKhbNQFAWvz8+zr73Fb/++nt1HasjNSOHDt67kE++5jszUk39vtuyr5IHH/sy2A0e5ctFs/vOTt7Nsbtk4vrqp1ePtZ9Oew7y4aQ+v7TxIc3t3/L6URBdLy8t47zXLuPbS+WSnJZ+XgaMQQogL31TN84YGbk8//TT333//SSvcOjs7SUtLO6trDXUhVbidSG+rI7zlr+DpIN6pdKzsLpTcUtT0fAyfB6OzCaO9AYwoSmoOatEc1MI5KClZw+Y4hmGA34vh6TSXpPb1DH7u6wGfBwx98Do2J4oryeyU6koc8nVS7OtE8zEyjxJCCHGeGVXgpqoqTz75JHfddVf8WGdnJ1lZWaxdu5Zrr7122OMlcDu3hCMRbvyXb7OnsgYAq6YSjuqnf1JMosvBg5/9EHfecEV8orPvaC2//ccG/vjyZqK6zgdvvpLP3HkTpQXZI55vGAYvb9nLf//fnzlU3ch1y+bx5Y+8m0vnTh+313eu6PH2s2FXBX96ZQtb9x+ly9MXv89ltzFrWh6XzpvOopnTWDhzGjOLctG0c3e5rRBCiIvDVM3zhgZuA3u4BQIBLLFO4q+//jq33norfX19E7I9xYUyzzuRoevox3cR2f2qufeazTHYjXQ8qJr5EQ2DYYAzCXVaOVrxfJSMPBTl9D8rQ4+agVwsgDN8Hgy/F3we82ufF/x9DAsLNasZvLmSUJyJ5tfOWDg39LZFlq8KIYQ4d1jO5smy/dv5wWqx8Nov/pOPfuNn/HXDDjiDdwi9vgD3feeX3P+9NXz41pX89313smBGMd//3N38xyfex6/++jq/eGYtv/n7et599VL+30fezexp+fHnK4rCzVcs4obLFvDXDdt58PG/c9Pqb3PNknK+fM+7uXzBzAl4xVMjJdHNu69exruvXgaAp9/HUy9v4plX32L3kWp2H6lhT2VNvFuq3WphTmkBy8rLuGTWNBbPKWVmUa68gyuEEOKcMJnzvEWLFmG1Wtm6dStXXnklABs3bmTZsmXjHrYNrXC7ECmqijZjKWrxPKIHNxI9vA00K6RkQXdrvAvp2BlgsYLdBYE+8HvQD21FP7TVDMayitBKF6EUl6NqI//UUFQN3LFlpSPfqzWvoEfB3xcL4MwPYmGcuZdckxnMRULDn2hzmA0fXInxIE5JzjSr8JIyJJATQggxqc6qwi0zM5N169ZJhdt55As/fJxf//V1El0O7DYrHT1eNE0lOsqKN4AFM4r52sfey7XL5mGxaASCYf7w8kZ++LvnaWjr4o7rl/OVj7yHkvysEc/VdZ2/vbGT7/32rxyqbmTl4jl8+Z53s2LhrPF8meecUDjCvqO1bNlXyZu7D7P1wFG8/X6AYUt9nQ4b5SUFrFg4ixuWL2DpnDIcdpkcCiGEmDhTNc8bWuEGcN9997Fx40bWrFlDY2Mj99xzD2vWrOG22247q+ucyoU4zzsZw99nBm9HtptBWdEcaDgCgf4zP5nNab5xG/LD0D8hrHaz6i0cGh7oORNRMgrQCuegFs1GsdrP/gXFGIYB4WCsOi4WxPk8ZvXcwO1+D/h6zScoCkpimtllNTUHJS0HNTXHHKO82SmEEGICnFWFmzj/fP9zd5OTlsJ31jxHJKrzmfffxK//+jq+aAiXw4YvEBrxHIfNQiA0OHnad7SW9//7j3DYrXzghiv4wM0r+Mi7ruGum6/kiRfe5PtP/J1nXt3Gh95xFV/48LsozB7c+0VVVd5zzTL+aeUSnt+4i+/99m+889++y4qFs/j8h97JqqVzL8hJj81qYWl5GUvLy7j/A7dgGAZ1LR0cOFbPgeN17KmsZf/ROpo7u9l5qIqdh6r4yVMvApCc4GJabiZzywpZPn86s6cVkJ+VRnZasixJFUIIccH4wQ9+wKc//WlWrVpFcnIyDzzwwISFbRcTxZmAZenNaOVXENn/BvrRnWYlWPFcjIZKc2noaIX8g1+n5aNoqllp5vOAERz+WFWDQB9G/SEi9Ydgi2KGb/NWoubPOOv5nqIo5uuwOSD51M2qjHAQo6fd7K7a3Yre3YLetHGwsYPdZe5Jl5qNkpqNkpKNkpIp3VWFEEKcNalwO40L+Z3Px59/g88+uAZFUXjg03ey+1A1z77+FgBWi0o4MrLiLSXBSU+ff8RxgPzMNO688XLuvOFyinMz+dVfXuNHv38eT7+fD9+6ks/9860UZKWPeJ6u67y0eQ/fe/xv7K2spby0gNV33MT7rrsMu+3im+hEIlEa2jqprG1m874j7DhUxbG6Fjp6vSOqEFVVISs1meLcDPKz0snLTCU/M5W8zDTyMlPJzUglOy35lN1khRBCCLj45nkXctOE0TD6uolUbEY/tgtQwJkAfd3gSga/Z3jl2plwp6JYbRiBfnOpqaKaFXGRIERPWMaqaqhFc9EWrDSXek7ym62GYUB/D3pXC0Z3C0Z3qxnIebsBI1YNl252Zs0tRc0pQXEmTuoYhRBCnP8kcDuNCzlwA1i3bR93fe0nhCNRrlgwk0+97wYeeOzPVDW2nvZ5SW4Hnv7AiOMWTSUS1Vl5yRw+fceNXL5gBr/+63p++scX6fcHzeDtrlvJzxrZacwwDDbuOczDf3yZV7buJTstmU+89zo+/I6VZKcnj9trPl8NVMS9vuMAG3YdYtehaupaOgCzes7ttKOg0OcPEAoPTmpVVSEnLcUM4DJTyRsSyM0qymNmce5FGWwKIYQYJPO8C3Oe93YMfx/Rw1uJHn4LorEVDrEGCHpNxeBSzNFQVEjKAM0Cve1m1dzAMtOgz6xEyy3DsDnM5az+wcZSWO2oBbNR5yxHS88b3xd5hgar4VrQu1sxWmswetoAUNLzUYvmoBWVoySNfBNZCCGEOJEsKb2IXX/ZAtY/9g3e84UH2XrgKDsPVfEvd9xEUoKTh574O/5gKL63mKYoRGPZ7MnCNoBIrAJr24GjvLH7EMW5GfzLHTex7bff4YkX3uDhP77EE8+/wV03X8mn3nf9iOYKV10yh6sumcOR2iYeffoVvv/E3/numr9w8xWLuPudK7lu2fyLdgmloigU52bykXet4iPvWgWAp9/PjorjbDtwlG37j7Gj4jihcASrRWNmcS4ledlkpyXhsNvo7fPR1N7Fhp0VNLZ3x/ePs2jmY+eVFTKvrJC5sc9ZaRJyCiGEEBcyxZmA5ZLr0eZeSfToTrO5gq8XvWILav5MlGs+iHFkG3r1/rdvtGDo0NsWO7EKziRzvzifJ37MqD8MioJWfgVK2SVED7yJUbUXwkH06r3o1XuJuJOxzFuJWrZwSpZ0KlY7SmYBZBYwsD7A8PehN1eh1x8ium8D0d3rUFKzzQq9kvkoiSPfSBZCCCHgDCrcVq5cSWlpafxYMBjkqaee4sYbbyQ3N3fY46uqqnjzzTflnc/zRHu3hw9+9cfsOlyNpqm4HXbuvnUl4UiUx5/fQDAUIaqbYZrNYiEcjYxYbaAoJ1+BoABJbief/edb+cCNK/jdSxt57Jm1tHV7uGZJOZ+87XpuvGzBSZc99nj7eXrdVh7/xwYOHK8nLzOVD9y0gtuuvZS5pYUT8J04v0UiUQ4crzcDuFgI19TRDUBeZipzywqZW1rA3LJCinMyCUXCHK5p4sCxeg4er6eiqoH+gLmfSVZqEvOnF3HV4jnccNkC5pTkX5B76wkhhLj45nkX+5LSUzEMHb2hksjOl8HbBYCSmos69woMVxL6m08Pr0wbC1UzJ4wOF5ZLrkcpnodeuYPovvWxPdWM+OOUvOloxfNQC2eNa7OFs2FEQuhNx9BrK9AbjkAkhJJZiFayALV4LorDPdVDFEIIcQ4ZdeB2xidWlPN2IjbgYgncwOyi+e1fP8tP/vAi+ZlpdHv7CIWj3HDZfBLdTl59az+dvYOTLLfDblbAnfDroyqgDznkdtjjIY7dZuGDN13JVz/2Hl7fUcH/PbuOHYeqyEpN4n3XLeeO65ezaNa0EcGOYRjsOVLDb5/fwF9e305vn4/Z0/K47drLuO3aSykryJm4b8x5zDAM6ls72X7wOAeO13GwqoGDx+tpajdDOJvVQml+FmUFOZQVZlOan43LYScQDFHX0sHeo7Vs2nMYXyBEotvJzKJc86M4l5nFecwuzqM4N/OirToUQogLhczzLvx53pnSW2oIv/UPc3komAFY/kywOTCq9sRysTHu9TaEklmI5er3o1hsRA9tIVqxZTB4szogHADNilo4G7VkAWpeGYp6buxNa4RD6A1H0Kv3oTcdA0DNm446bZ7ZFMLumuIRCiGEmGqjCtw2bNgwppNfffXVY3reueJinIit33GQ+x9cQ1tnD1ctLqe+pYPKumbSkhKYP6OI1s5eDtc0xh+f6LLTHxhcegonr3bLSk2ip89HKBxBURQWzSzms3fdSn5mKs++/hZ/XreVtm4P0/IyuXH5Qm5cvoAVC2fjsA9fThAKR3ht+wGefW0bL27aQ58/wJySfG5cvpCbLl/IsvIyaRLwNnq8/fHw7Vh9C8cbWjhW30p9aycD/xwkuZ2UFWRTlGMGauFIhH5/kLbuXmqbO+JLUu1WCzOKzAAuLyOF7PQUstOTyUlPIT05kbTkBNKSErBZZfW6EEKcq2Sed/HM886U3tVMdPer6M3Hhk/uFNX80CPgcENiGnQ1j2yOcKKhzxsqKQOtYBYkZWB4O9GP7oBQAKx2lJwSjN4O8HSA3YVaPBetdCFKRsE5U31vBPrRaw8Qrd6P0V4PioqSVYxaOButcBZKQupUD1EIIcQUGFXgdrG6WCdivkCQHzz5D376x5dwO+zcfv1yVFXlH2/upLGtiwSng/ysVGqaOwiGzFbyNotGRNdPCN4UTvz1SnI7cdittHWZe3rYrRZuXnEJ77l6KVaLhVe3H+CVrXtpbOvC5bCxfP4MLp8/k+XzZ7KkvBSn3RY/lz8YYu3Wfby0eQ9rt+2jo8dLSqKb6y6dx9WLy1mxcBYl+VnnzGTsXBcIhqluaosFcC0cb2ilurGNmuZ2mtq74z9Li6aSn5VGWlIiNpuFaCSK1x+g3x+kq7cPX6yicahEl4O05ETSktykJSWYQVxyAunJiaQmJQw7npJofu122uVnJ4QQYsJcrPO8sTAC/USr9qIf24UxUPUWpwAGpOejzb4MPJ1Ej+85s6YLJ7LYweYYPIfFhpJTioKB3tkEfi8kpKKVLjAr35Iyxn6tcWb4POgNR4jWH8ZoqQY9ipKYhpJbippThpozTarfhBDiIjEugVs4HKaiogKXy8WMGTPGY1znhIt9IlbX0sGP//ACT77wJhZN5b2rLuXyBTNpaO3ila172XW4GkUBq2YhFHmbdzRPIiXRhaffHw/pLBaNKxbM5LpL51OWn83R+mY276tk2/6jePr9WC0a88oKWTCjmHnTi1gwo4i5pYW4nXZ0XWfX4Wpe2bqPtVv3se9YLbpukJuRwhULZ3HFglksLS9lTkk+VotUW52pYChMfWsn1U1t1DS1U9vcHg/japva48uGAdKSE8jPSCU9NYnUBBdulwOb1YKiKITDEbo8fXR5+unq7aPb00dnbx+RkyxLslktpCUlkDo0pDvhdmqi2wzsBu5LdEuFoxBCjLMLbZ4ne7idHcPbRbT+ENGaA9DZNPIBdhdK0RzUtDyMvh706n1nHr7ZHGaX01AgtsSUwSUUigqJqaBo0NdlVtWl5aGVLkSbNg/FmXD2L3KcGKEAeks1evNxjJYqDE8noKCk5aLmlKDklKBmFZ0ze9QJIYQYX6MO3Px+Pz/+8Y958803ef755+PH165dy0c+8hFaWloAWLhwIU8++STl5eUTM+JJdLEHbgOaO7p58oU3eeL5N6hv7aQgO513rLiEKxbMoNvrY9PeI7z21gG6PIN7vJ2suu10VFXBMCAjJQFPn59gOEJBdjo3XLaAa5fNIys1ib1Ha9l9uJp9x+o4UtNEJBpFURRK87OYXphjLm0symV6YQ656SkcqWtm094jbN57hD1HaojqOnarhfnTi1g0axqLZ5ewYGYxMwpzZcnjWTAMg/ZuDzXN7ScN4wYaN4BZ0ViUk8G0/Cym5WYyLS+T4pwM0lMSSXA4CEejdHv7zVAuFsh1eczb3b19g2Gdpy++rPVESW7nkHDOrJ4b9jl2X1pSAsW5GaQmnTsTcyGEmCoyz7t453lnQw/0E93+AnrNATMIM/TYPbGqNwBnImQUgB6FjgYI+sZ+waH7lpzqegmpqLll5r5vmYUoNsfYrzfOjP5e9JYqs+tpS7VZqaeoKOl5qDklZgiXWYhisb39yYQQQpzzRhW4hcNhVq1axebNm1FVFb/fj9Vqpbq6mrlz5xIIBFi1ahULFizgmWeeIRgMcvDgQTIyzp3y7rGQidhwuq6zcc9h/vHmLp7fuIum9m6cdhuXzp3OFQtnUpCdzs6Dx3lq7WZ8gdBJz6HE/uftfuuSEpzMKMiho9dLbXMHFk1j8ZwSVl4yh6sXlzN/RhE1TW3sO1rH4ZpGKuuaOVrXQl1LRzzoS01yU5qfTVFOBnmZqWiaSp8vQHNHD0frmjlWb/7xYNE0ygqymVOSz5ySfGZPMz+X5GVJtdQ48MeaMJwYxg3c9gcHf1cS3U4Ks9MpzE6nKCfD/Hrgc3Y6malJ8WWm4UiEHq/vhHBuePXcyW6HI4PVdIqisGR2CR/5p2u484bLpfpRCHFRknmezPPOlt7eQHjLc+DtQknNNZedRmL//65qYLGZDRAGJoA2p1m9diaNF04M2BKSzeq3oH/wnIZhXjf+OECzoiSmoqTloSRnoGaXoGYWnO1LPmuGYWB4OjFaq80quJYaCPabDSoyClBzpqHmlJp71WkyPxFCiPPRqAK3n/3sZ3zmM5/h85//PF/5ylfiE6xPfOIT/PrXv+ad73wnf/vb3wBoaWlh3rx5fOITn+C73/3uxI5+gslE7NQMw2Df0Vre3H2YjXsOs2VfJZ5+P6qqMKs4j+z0FPZW1tDt6T/teYa8H3lKiS4HxbmZpCUlsP9YHd3efhw2K8vnz2Dl4nJWLp7DopnT0DSVQDDM8cYWjtW1cLSu2ayyau6I7UPWFV++qqoKuekppCUn4rBb0XUDb7+fls4ePLHKKZvVQkmeWT1XVpjN9IIcpheaHxkpibK/2DgwDIPWrl7qWjqob+mkvrWD+tZO86Olk4bWTvr8gfjjHTYrBUMCueLcTIpzM2KVcpmkJSec9udiGIa515zHDOAOHK/n72/s5OUte1k6p5THv/kZcjNkY2MhxMVF5nkyzxsPRjRMdN8Gogc3gSsRtbgcvfGY2elUVWNt7A0zfLO7zKWggb63Pe+pxWaRmhUlPQ8sFoyedvCZ+wRjsZvXjYTM6jqIV8gpyZloS25Cyz93lkgbhoHR247RUoXeUoPeWgMhP2gWlMxCMyjMKUHJyD9nOrUKIYQ4vVEFbldffTUWi4VXX3112PGsrCw6OztZv349V111Vfz4Zz/7WdatW8eBAwfGf8STSCZioxeN6lRUNbD7SDW7j9Sw50gNB6vqh1UTjRe3w05+VhqKolDX3IE/FCLJ7eTyBTPjH4tmThuxTDQUjtDY1hUL4YZ+dFDb3E5n7+Ckz6qpJCe6cdisGAb0BwL0en3xcDDJ7YyHb9MLcygryGZGYS6lBdm4nbIPx3gxDIMeb388gBsayNXGquQ8Q5aWJrocFOWay1RL87OYU1LAnJJ8Zk3Lw+U49c9lR8Vx7vnPR0hOdPHiT79KcoJsZiyEuHjIPE/meeNJ93QQ3bUWvf4wSnIWasl8dF8vRtU+M/yyu0HTwN9nVqJpFnAmmZ1LA32g629/kVOxOlDyylAzCiHgRW9vwOhsNMM9RQHNan49UAFnc6AUz0PLn2HupXYONTMwDB2juxW9pRqjpRq9rdas6NOssQCuGDV7mhnAadapHq4QQoiTGFXglpmZyZe//GW+9KUvxY/t37+fhQsXkpycTFdX17Cqkl/+8pd87nOfw+v1TsyoJ4lMxM5OMBTmUHUjh6ob2Lr/KK/tOEBDa9eEXMtpt+Kw2+jrDxCORrFbrSwtL401TJjJ0vIyElyn38PD6/NT39JJbUsHdc3t1LV0xMO42paOYXuG2SwWnA4biqIQCIYIxLq1AhRkpTFrWj6zp+UxsyiXmcV5zCrOlb3CJkiPtz8enMbD1KZ2jjW0UttsdlJTFIVpuZmUlxbElw7PKSlgemF2fBnp4ZpGblr9bS6bP4M/fPvf0DR1Kl+WEEJMmottnidNEyaH3lZH9MCb6I2V4E5BK7sE3EkYzVXoDUdi4ZsLxZ0CgNHfPbg81O6CSBii4VNf4O0oKqRkoZRegpaSDn096F0tGN0tGF0tZsB3IkcCSkYeau501KI5qK5z5/fC0KMYXS3orTUYbbVmABcKDC5BzSpGzS5GSc87p4JDIYS4mI1qQwCPxzNin46NGzcCsGLFihFLuKIn6TgoLj52m5VFs6axaNY0PnjzlQB4+v386ZXN/OYfGzh4vH7cruUPhvEHBydlwXCYzfuOsGV/JQ/qBoqiML0gm1XL5rLyknKWz59BekrisHMkupyUlxZQXjpyXw/DMOjt8w2riKtr6YiFc+btgdCtoa2L1q5e3tx9iFB4cDKXnpwYD3sWzixm4cxpzC7Ok33izlJKopuURDcLZ04bcV+fL0BlXVMs+DXD3ydfeJOWzh4ArBaNsoIcZhWbwehH372Kn/7hRf7j0af4zmfumtwXIoQQU+Rim+etXr2a1atXx99YFRNDzSpCvfaf0btbiFZsIXrwTdB11MLZWFbcBqpmLp9sOo7R22Y+KSENxZVk7vfr84B34I3a0WxCcgJDh+4WjJ0vEgGzwi0hDTV/Jtryf0K1WIm2VJtLYL2dZvVboA+joZJoQyXR7S+AakFJSEZJz0fNm26GWYlpU7KkU1E1lIx81Ix8mLsCQ9cxetow2mrQW2uJHt1J9MAb5oMT01DT8sxmDOl5KGm551TzCCGEuFiMKnDLyMiId6ca8Nprr6EoCtdcc82Ix+/fv5+srKxxGaC4sCS5nXzivdfxifdeR21zO0+v28rzb+5kT2UtYO6bNjSkOhuGQbyBgmEYHK1v4Wh9C489ay6ZyUlPZuXici6bN4Nlc8soLyk4ZVWToiinDXYGOnXWNndQ22KGcXXNHRyta2b/sXq8Pj+dvV62HTjKrsPV/N9z5hgcNivzygpZNGsaS+aUcsWCWRTmpMv+cOMkweVg8exSFs8uHXa8q7ePQ9UNHKxqoLKumcraJh7/xwZau3oBePTPa3n8+TeYV1bIkjmlzCrOk0pFIcQFS+Z5YiKpqTmoK96LsfRm9Op9RI/uJLLhKbA5UYvKsSy7BRJTMVpr0dvqMNrrzKYLAM5EFHeyGc4F/eZ+cGcavA0wDPB2oh/egn54i3lMUcBqB1ey2TFUs8K0+SgOF3S3YnjaMfp6MDyd6NX7YidSwJ2EkpaLmpyFkpyJkpyBkpSBYp287qKKqqKk5UBaDtrs5eYecN5OjM4mjM4m9M5G9L1HiA5UCbpTzHEmpKI4XKiZRSgpWeBMQFGkql8IISbCqJaU3n777VRVVbFr1y4Auru7KSwsxO/3U1FRwaxZs+KP7enpYcaMGdxyyy08/vjjEzfyUQqHw9x99900NDTgdrt58sknR91VS5aUTp6m9m5e2ryHl7fsZdOew/QHgtgsFpISnITCEbz9/rFOr0ZNAQqy01k8u4Srl8zl5isWjssG+oZh0NjexYFj9ew/Vse2/UfZsq8SXzCE3WYlKy2JSCRKc0cPAHmZqfG96K5YMItZxbmoqkyEJkOPt58jtU386i+v8fc3dxIIhlEVBX3IP5Muh538rFRK87KZXpTDtLwspuVmUpCdRlZqMimJLvl5CSHOK+fzPO9syDxvahiGgdHTil5zgGjNAejrBkcCauEs1ILZqLklEAmjt9VjtNeidzRhdDWbXU4B7G4UhwsjHASflzEHcBNBs4DNATYXisOF4kgAVyI4E1CtDrNhhMWKYrGB1QY2pxnu2RwTEnoZum42YuhuwehuNbui9ndjdLcOPkhRwZWIYrWjZBWjJKWjuFNQnAkQCaFkl6DIvEYIIcZkVIHb66+/znXXXceNN97ILbfcwu9+9zt27NjBLbfcwvPPPx9/3KFDh7j33nvZvHkzr7/+OitXrpzQwY/Gk08+yZ49e3jooYf4zW9+w8GDB3nwwQdH9VyZiE2NSCTKnsoa3th9iI27D7PtwFF8gRCqolCUm0F6ciK6rtPZ66Wjx4svEJrQ8TjtNjJTk5hRlMOlc6dzyawS5pYVkJOeMuZgJRyJsLeyllff2s8/3tzFgeP12K0WFs0qISstiaa2bvYerSUSjZKa5ObSudNZPn8Gl82bwSWzSnDYZXPciRaORNi05wg7D1ez50g1R2qaqG/tHLZf38koikKC005KopuM1ETyM9LIz04jMzWZtCQ3aUkJpCa5SU1KIC0pgbTkBJz2yXtHXAghTnQ+z/POhszzpp5hGGY1Vu0BovWHzSWkmhU1b7oZwOXPRHG4MQwdvN3oXc0YXU3onc1mCBca2F9XiXUgPUnDBc16dnvBndLAaoQhf0opqnnc0DmzIFABuxPF7jT3tRuo7HMno7iSUdxJ5l53Dve4rIIw9CiGtxvD0wE+D0Z/L9GDG1ES0zF8vWZjiaFcSea+cDYHRn8vSkIKak6peczuQrHZQVExPB2oJQvAYpPVGkIIwSgDN4CHHnqIr33ta4TD5v9hzZ07l7Vr15KTkwPAF7/4RX74wx9iGAZf/OIX+d73vjdxoz5DkUgEi8XCt771LVRV5atf/eqonicTsXNDJBLlUE0jOyqOs6Oiih0Vx6msawYgLSmBxXNKmZGfhQ50efqoamyjvrWTju5eswP9BHLabaQlJ5CXkUppQRblpYUsnVPK0vJSbNbRh2I1TW387Y2dPPXyJg5VN5KTnsJ7r11GeUkhDa2dbDtwlB0VVfT5A9isFhbNmhYP4JbMLiU7XfagmQyGYdDV20ddSweN7V20dPbS0tFNVWMbDa2dtHV76PL0DWuwMRoWTSXR5SQ50UVqYgIZKYlkpSWRkZJIenIiackJpCaaIV1GSiK5Gam4nXaZzAohxs35PM8bK5nnnVsMw8DobUdvOILecASjvcHM0TIKzAAut8zcQy22f5phGOD3mp08e1oxetrQu1uhu+XkF1BUcLhB1czlo/o47EWoKOZS1dM/CFTV/FBUswtrNMKwQM5iM5e2Wqzm+AwgEgR/PxhDxmlzoqRmo6Zmm/vK5ZSiuBJPvOBZMQwdgn4MvxejtRYUBSPQhxHwQThoNsEIBczGFiH/yV+/osZekw18HrPCL7PIvG2xQb8HveEwlivfBxa7uRR34PX396LmTR/X1ySEEFNl1IEbQFNTE1u3biUpKYlVq1ahaYMbhj788MP84x//4JOf/CTve9/7JmSwp/PYY4/xk5/8ZNixV155hby8PABuvfVWtm/fztq1a1m4cOGozikTsXNXj7efnYeq2H7wuBnEHaqit8+HoijMnpbHsvIyFs8uIScjBW+/n/3H6tl1uIqKqka6PH2TMkab1UJGSiIzi3K5PNYpdXpBDvlZaafcK84wDPYdreX3L23i6XVb6Pb0s6y8jA+94yredfUS6po72HbgGFv3V7J1/9H4MtSc9BQWzChmwYwiFs4sZsH0YtkLbgpFIlHaezy0dvbS1t1Ll6efrlhFZktHD61dPbR3e+n29NHb56ffHxi2bHWoU20TrSoKTocNt8NOottJsttFarJZMZeZkkhmalJ838GinHRSExNIdDtIcruwWUe1facQ4iJzLs/zxpN0KT0/GH4vekMleuNR9JZqc0mpzYGaU4qaW4aaV4aSMHLrj4HqrWhHPfrxPdBef+pwzWIzg7BQEDDMr81NgEc3SNUCFgvohhmQnWhUgdxoDIR2sf8m9ejga7K7zGWgqdlmcwS72wywLHaw2lCsdjPMstrGfdmqYehm+BYOYkSj6I2V5lLUcBAjHIJIkOi+DSjp+eZS20gIwiGMnta3P7kryWz0oFkwOpvMTrdFc8zzqBp6zQGUlCy0mcvMJbkWm1mx505GTc4cPs6gTzq3CiGmxBkFbue7Y8eOceutt3LkyJFRPV4Ct/OHruscq29le8Uxtscq4Q5VN6DrBoluJ0vnlLJsbhmXzp3OzKJc6lo7OVTdyOGaRiqqGjhwvH5YVZKiKEzkfxopCS5yMlIpyk1nRkEu5WUFLJlTyozCnPgy1WAozEub9/DEC2/y2vYDOO1W3nPNpXzoHVexfP4MAOpbOtlTWcO+o7XsPVrLvspa2ro95jUS3cwtLWDe9ELmlRUyr6yI2dPyZTnqOUjXdTz9fjp6vHR5+ujx9tPj9dHj7ae3z0dnr5f2Lg9t3R46e714+v30+QL4AkHCkTN/h15TVRx2Ky67HbfLTpLLSVKii9QEN2kpCaQkuEhKcJGc4MLtdJDodJDgcuBy2HE77SS4HCS5nSQ4HdJlVwhx3pJ53vnD0KPm0tOmY+jNxzE6GsAwzI6h2dNQs4pRs4ohIeWkbzbq/R4i+zdgVO8zQ5/TGRqSDQRUJ1uq+nYsVtBssfNFIRx6+6o6RRlclophVsON1x51ihLfWw67a3A56JDbijPR3M9tIOyaQIZhmD+LSCgWzoUgHCR6cBNKTgmEAhghP0Sj6Md2gqKiJKaZe/dFI0OWE5+E3QVWs3LO8HRBNIySkm1WzsWq7BSLlUjFZtT0fCyLrgWb07xPUTF621FTpDGMEOLsXfCB22OPPUYoFOIzn/kMLS0trFq1ikOHDo3quTIRO795fX52H67mrYEquIrjdPb2oakql8yaxopFs7ly0Wwumz+dBKeD1q5eDlU3cqi6gcPVjew7Wsuh6iZCkfHpmjpaqqKQ4HKQnpxIYU46pfnZ5KQnU9PUxoZdh2nu6GZ6QTYfunUlH7hxxYjlpC2dPeytrGX/sToOHq/nwPF6jje0YhgGmqoyoyiHuWWDIdzAfnRSDXd+CoUj9Pb56O3zDQZ1febntq4emtq7ae7oiYd3fb4A/f4gvmAQ/TRrrkcbOlstGm6HHZfTQYLLTnKCm0SXHbvNisNmxWGzYbdZcdpt2G2W2OfYfXYbDpsVu82KzWrBZrFgs2pYLZb4bVVT0KMGRbkZ2K1WbFZNmlIIIcaFzPPOX0bIj95SY4ZvbbUYPW3mHa6kWPhWZDYASMkcUdWl97QRObQVo/YAhIdWpSlmtZrBafZ8O1Xd+UlY7WaIoyhmmBTwDT5XUcHhMqvQFMUMkCIhs9LOOE0op1ljQZ7FPIdhmCFeJASR8OnHpsaWeaqa+dyBYDEaMZ+rnzDftdhQXEmDAZzL3EdOSUhFSUw195dTp+5NN8PQwd9nBnChAIRD6J0NGD6vOe5wECMSAn8fel2F2aVVVc1jkbAZgJ7u+2V3mRWDFiuGvw+jowG1eC5qblm8+UVky1+xLLkRNW9G/Bj9PRhBH2p6/vDx+r1moCmEuKiMKnAby6a4iqKwYcOGMQ0KIBgMsmTJEh5++OF4S/pAIMDq1at55plncDqdfPGLX+QLX/jCac/T09PD3XffTW9vL7qu853vfIerrrpqVGOQidiFxTAMjta1sGnvYTbuOcKmPYdp7eodEcAtnz+DBJf5rp6u6zS0dnG4tpEdB4+zad8RDh6vp7dv+LtqZzD9GnfJCU4WzChm+fwZlOZnU5iTQWF2OnkZqfHqo35/kEPVZiXfgeP1HDhWT0VVPV6f2fErPTmBeWVF8Wq4uWWFzCrOk6WHFzDDMAiEwvEgrsc7GNoNDfC6PH109vbR4+mnp68fT78fb79ZXXcyCuZyalVVURQFTVPNawVD2G02dF0nEo0SjeqnXEb7djRVxWrRsFot2CwaNqsFq8ViHouFdjaLBcuQ+4Yds1hwOmw47TYsmoY19jiLpqEoZidau82KRVOxaFr8MRaLhkVTsVosOGxWEt3O+GNsVgt9vgDpyYlYNBVNU1FVFTX2PdCGfD0QbhuGMSLoPtkxISbaVMzzzgUyz7twGEEfelsdRlut+bmzyaxKs9pR0vNQ0/PNveAy8s0wJkbvaSNydCdG1R4ztBnKagd37E1Nn2fk/WNhtZtVeA73YDOHvh6Mvp7BKjpVM5dTOtxmiGPoZgVYOABBvxkSnqriTrOZoaFmgWgsiDsxPFS1weBNj46sFFM1c9mmxYqiaub8Nho1l8wGAwyGhgq4kuMBnJIw5CMp7ZxfwmkYhllF19MKhoER9JlBXCRE5K3nUUsXmrejYQyfZ3hn11Ma/ItAScowAziLDaOv2/wdciWZy2JjgWe0YjMAliveGwtCFaKV2zFaa7Decm8sWFXMys7ag1ivvzu+5NnoaYVoJL7fnaHrGD2tqGm5g6+xrwclIWXwtr9vWOMNI+g3q/9c8u+fEBNlVIHbwB9OwKiX2SmKQjQ6ts1IA4EAd911F8899xyvv/56PHC7//77eeONN1izZg21tbXcc889/PrXv+b2228f03VOFAwGCQYH/4j0eDwUFhbKROwCZRgGxxta2bjnMBv3HGbTniO0dPZg0TSWzS1j1dK5rFo6l0tmlYzYc03XdTbvq+Tnf36FTXsr6fH2n/I6iqKQkuAkNSmR5AQnFk2js9dLc0cP/uDEdli1WzWSE9xkpSVTlJtBeUkBi2ZNY+mcUjJTk6hr6YgHcAeO13Owqp6apnbArFyaWZwXq4SLfUwvJCNF/lsQEI3qeH3+YSHd8YZWmtq7sVkt8WWvnn4/oXCYHRVV5KSnoBs6R+ta8AdDOGxWVFU9ZXg3YGi1XUZKIoqi4A+E6PMHKCvIBqDPF6C1q5fyknzCkSi9/T4ActJTY0vOWyjISsdq1QhHonT2eHE77UQNg0Cs07FuGOi6jj8YIhIdw/KhUVIUBUUhXmGYnpyA3Walqb0bgEtmTcNqtaCpKo1tXdS1dHDNknJsViu6odPnC7B1/1HeffVSnA5bPMzbvK+Sq5eUY7VoRKM6v/zLa+Skp3D3O1ei6waRaJTnXt/O7dddRlpyAuFIlKqGVry+gPk8TUPVVDRVQY1VhCgK8SDxKw//nm99+v0kuBwEgmF2H6lmSXkZ1lggWdfSQWlBNhkpiQSCIYLhCFmpSaiqGTgOBI+aqpoFH1Edp92G1aJhGBAIhej3B1FVhZz0FCLRKFaLhc4eL+kpUhUwkSZ7njeewuEwd999Nw0NDbjdbp588kkyMjJG9VwJ3C5cRjiE0dGA3tGA0dmI3tFoNkoAcCaawVt6Pmp6HkpqDoozAb2njWhdBXrVHvB2jzyp1Q4JqWYQFg5g9HacUCE3SorCYCdTICULJasINSULxe7C8PeDtwvd2wnerhFhnOJOgYRkFLsbNA0iYYxw0GxqEOg3A7STVbsNrWo7MbCzu1ESU1CSMs1AMBILmfxe8HkxfJ6Rwd1Atd1Ad9jwCVVyVgckpaMmpZt7zCUOfE6b8OWqE8nQdfN7EQmZv1eKEmtyYe5Zp7fUoNfsR5tzOUY0bP4sgj70xqOD+9INhGaejrMfkDMBFA18vfFDSlYRRludecORgJozDUJB9KajAGizl5vh3qEt5u3FN6BYzOXPenMVel0F2vyrUTILUBSNyME3MVqqsd56X7yyMvz679BmL0ebvhhUFb25iujBjdhu+WR8HJFDW1CsdvMxMXpPG4rDbf53BOhdZpMTNS0n/phozQHU7GnmfoDxc23FaK/DuvLOEd8C3dNpdva12uLHjP5e0Czx64BZFRnd8TLa3BUnDRkNn2fEcSMaxuhpR03PO8UPYGwMXUeRFRsXhVEFbiUlJdTW1pKdnc0dd9zBBz/4QS6//PIJGVBFRQV33XWXuXn8vn3xwK2/v5+MjAxefPHFeAD3rW99i3Xr1rF+/fpxufY3vvENHnjggRHHZSJ2cTAMg2P1LWzYdYj1Ow7yxu5DePv9JCe4WLl4DtcsMQO4kvyRezr0+QI89uxa/rh2C0drm+NTHKumYbdZCIYihIf8YWK3WcnLTGV2cR7zpheSmZJEV28f2w4eZeehajz9fjTVrIYJR6Ljs9/uSWiqit1mITnBRU56CmWFOcyelodF0+ju7aOxo5uqhlYO1zTii4USOekpzJ9RxJLZJSyeU8qS2aWkJSe8zZWEOLVgKExrV298X7qorlPd2IbdZsUfCBEMh+nt87Fp7xHmlRUSieq0d3v44yubeedVi+OPP1zTxMKZxdhtVt46cAwAl8NGMBQhqp9ZgOZ22OmPBYFlBdkcb2glNdFNQXY6nb1estNTUBWFsoJsHDYrmqbx0ubd3LB8ITMKs0l2u3C7HOiGQTSqYxgGUV2PV/ZFdZ2Xt+zl8vkzMTAIBMNs3nuETXuPcNfNV8Yf39bVy/qdFVy3bB42mxnCefsDbNhVwVWXzCYUjqLrOlHdDOIG3hxQVZWDx+sByMtIRdVULJoaD9TdDjtWq+W0bxZMpYGQVVWVeDCZ5HaixSoKB0JGYsGlt99PXmYqumFgGAZWiwVNU2nr6sVq0Wju6GHx7BIcdptZgaiqLJlTytc+ftsUv9Jzw2TO88bbk08+yZ49e3jooYf4zW9+w8GDB3nwwQdH9VwJ3C4uhs+D3tFoBnGdjWYV3EBg5kyMdf/MMZcRJqQQ9Xaj1x6E1ppT7BemgCMB3Elm+OT3QH/v8H3aBsKYM1kHYXOgJGeZe9PlTUdJSjOXmvb3YHi7hnx0Y/R1xTqexsbjSjQDrcQ0cx+zSBDD12cGZgGv2fV0NN1ZYx1dleRMlIx8lIQ0FGeCWfFmRCHgGwzj/B4Mn9e87ffG9p0beq6TfA9UC9gdZjVcVrFZgZhkjlux2LjYGNGI+X0zzA/D32f+nKKR+LLfaN1B1OwS8wmKgtHVQvTwVrS5K+J7/UX3v4GaPxPsTtCjZmOJpAxwJprPaakyn56cOTzwUy1ALIgdjz86tNgKmdjvppKYbobDihKvFFRySgEDo6XaHELJgvhz9LoK8zRzYv8/ZBhED281jy1chWJ3DvnmQWT7CwBYrn5/fN/FyPo/AGC98WPxBxq9HUS2/d08/k/3mwWJMXptBdE9r2K56g7UvLJ4KBre9CxG0zFs7//3YUGx4e0i9JcfY333v6ImpceGaRB582m0eVfFKw2NcJDIm3/GcuX74s/XPR2E//pTrNffg5pbOuxbF37zz6hF5WjF5YPX0qNEKzajzVmOollj19KJHt6GNn2x2RglJlq9H8WVhJpdPPh8fx+GpwM1e9rgMZ8Hve4Q2uzLhl1fb61Bb6nGsnDVsONGKEC0YhPagmuGLSU3m9R0jWhUEnplDdqcK9AKZ3Eio697RMMbvafNDORPPHdP27CqTYhVZlos8e/FuW7Ue7ht27aNP/7xjzz99NM0NTVRVFTEBz7wAT7wgQ+MuuvnaDz66KNUVlby7W9/G7fbHQ/cNm/ezMqVK/H5fNhs5j/E69ev55ZbbqG/v39c9vSRCjcxVCQSZefhKtbvOMjrOyrYUXGcqK5TnJvJNUvLWbVkLisXzyE1aXjY1Nnj5el1W/nNP9ZzpKYJVVHiS+dy0pNxOex4fQG6er1ET9hDK8HpYEZRDnmZqXh9Aaqb2qhv6UQBls6dzrLyUqblZdHW1cvGPYc5Vt9Cb59vTBvnnwmrRYttoG/HarEQDIXp6PHGK/TyMlNZOKOY5fNnsnzBDOZPL8Jpv/gmTOLcpes6Xl+AcCSCqqh0efrMcC+q0x2r0LNoKqFwlNe2H2BuWQFH61qobW6nvNT8OhqNEtV1guEI9a2dBENhXA47nn4foVAkHtANKMxOp761k+QEF1/40DvZe7SWu26+Mr6fXTgSpd8fZNXSuVO+jFTXdXQ9FgrGvjYw0HUDq0WjzxcgEtXRY1URFs2sorNaNEKRCJFINFbNGCES1bFoKpGoTjAUjp8zGtXj4aBZcWeeb+DfQafNnDhVVDeQk55iVj/6g2zae4Sl5aVoqnlO81zRWLgGvX0+mtq7mJabCYqCoihEIlGC4TA9Xh9Wi0YgGCIl0R0LdA0MQ2fe9CI+/8/vnLLv+blmsuZ5EyESiWCxWPjWt76Fqqp89atfHdXzJHC7uBmGDn096F0tGN3NGF0t6N0t5tI/AM1qdv5MzUFxJWJEIhi97RidQ6rlTsbmMvdnM3RzCeiJYZ2ijq0Jg9UOrmTUhBRITDVDk8R0VIsVIxKG/l4Mb2c8iDP6PWbF27CgS4MhS1UZ2P9sNCHcsNfoNPcjc7jMCqJ4Ewanufebopjf30jErAYLBzEC/bHgr99cvhnrbjqCMxElKcP84zspbbAyLiEVRZOtTiaDoUcHfy/0qBmU+fvNgErXzdvebgj2gzMp/rjI7rVY5q0EzP9/1msPYrTXoZVfAdEohqGjV24HQC2ea1bU1RwAQMmKBUSKgtFaY36ZlDF4rLd9cICqxmBaZpz57+9YnapKdGD5dNA3eCwh1XxM/5DKw5RsM+jsbRs8llk48BVgYLSbb5YqsaXCGAZGcxUD/x0PfN+Mnrb43pVq2aLYfo46es1+89jMZebzdd1sOgKo0+ajuBJB1+MBppKSjVo0J74fZPTgxvjzFYcrHjxGD7xpns9iw7L4erPSU48S3bUWAO2S61EcCeZrCIeI7njRfPiK28zgVY9i6FH0Y7vMEK1kgflaDAPCQSKbnwPAet2HGfjZRnavxehqRrvketTMwvj3PbzucfOx77jX/LYYOuHXfgchP9Z3/kus+YyColmHLameKmfcNMEwDN58802eeuopnn32Wdrb25k5cyYf/OAHef/738+sWSNTzDEPTlHigdszzzzD6tWraWlpid9/6NAhysvLaWtrIzMz8zRnGhuZiImhPP1+Nu45HAvgDnKsvgVVVVg8q4QbL1/IDZctYMGMomHhb3VjG39+dSt/eGkT1U1tOO02VFWh3x/EoqnMKyskOz0Fry/A8foWWrvMf5TVWFv6gaBOVRScdhv+UAhdN0hJdLN0Tik3X7GId1+9lIxU8/ezs8fLa9sP8MKmXWyvqKK1s2dClsYNZAOn+9fD7bCTkZpEUU4Gs6flMrukgJy0ZDJTk8lKSyIzNQmXw37qEwhxnolEorR29dLW1UtNczs7Ko7zs6dfedvnqapiNp4Y0gXW7XTgdtrjxx2xpZeJLgcOmxWbzRrbv86K3WbuU2e3WrBazc+22IfdasVq1cyGExYLNttAcwoLdtvAvnWyZ5wYNJnzvDPx2GOP8ZOf/GTYsVdeeYW8PHOZz6233sr27dtZu3btqANCmeeJkzEC/RjdreixEM7obsXwdA4ul1QtZtMAh9v8wzfQD/0ec4+1U1LMDpgQ/+PzlDTLkKo1Rh/QaZZ4EKYkpKIkZ0ByRqz6RcGIhlH8ZrWbEegzq6j8fWZlWmCM1c6qxVxWqqqDe8adOFZVA2eCOS5XohmouZLMwM5qQ/d2Y3Q0mCFL0Aco5h5yimIGcgPfK0UxGx8MLEtNHFyuOtUNHMTUM2L775k3Ykt2Q/7B8AXzs95SDYZuBtYniFZsRi2eF6vGU+N/8ET3vIY2Y7H5u2zo5n2qRvTgJnOvvwEWK9EdL6Gk5ph76w3s1bd/A0pOKUpyhtnEZWA5ryvJrHCL/z1loFftBUAtnBPLncwgTq87ZC4pzyo2X5+mYTSZqzmUzMLYtVSMVrNiUEmNLdFVNfNNAjCXNFusoGjDq2OdibHXpEJfbDm9xTb43yGK+UbE0H0mFdX8Pp1sb8uT/ZsV+56duiHN+FPScrHdet+kXe+U4zibLqW6rvPqq6/yxz/+kb/85S90d3ezcOHC+KSsqKjo7AY3JHB74okn+PrXv05tbW38/qqqKsrKyqivr6egoOCsrnUyMhETp1Pf2sn6HQd59a39vLbjIN5+P9lpyVx/2XxuuGwB1yydS3KC+Y6HYRjsPlLNn9Zu4a/rt9PS2YvDZiU1yU23p59AKExygovL5k2nMCcdm8VCbXM7u4/U0NzRA5gVJW6nnXAkOmLPK6fdRkF2GotmTmPFwtnMmpZHaX4WmalJRKJRDtc0se3AUV7YuJs9lTV0eyZmGZnTbiUl0Y3daiUQCuHp98eXop5MWlICJflZTMvLpCQvi9L8LKblZVGSn0V2WrIEAeKCEgiGaevuxR8MEQpF8Pj8/G3DDmZPy8MXCNHvD9AfCOLtD9DvN5fX+gJB+v0h/MEg4UgUry9AKBQmGI4QjkTGtFz2REMDOvMjFubZrGaIZ7HEQz2vz89bB45x723X09vn44+vbOZT77ue6YW57K2sie8F19vno6K6gVuuuIQjtU0kOB3MLSugs7ePvMxUvvXLZ/n4e65lXlkhigI3f+Y7pCUl8PIjXxu2l9jhmibmlOSfdNy1ze0U5WTEH//6joNcvXiOdLEdRxM9zxtvx44d49Zbb+XIkSOjerzM88RoGbpZqWJ4OtA9HRieTgxPh1nl4+sd/g6kNfZHajQ8PDgbKyX2Rz6GGapZ7WYQNXBum3OwAcLprqkoZiDnSopVj8X2VcsoMPdtC/rMpWcDIVxvO3pvu7mHXKDvDJtGKOYf7M4Ecx8uqz22NNLcz8zweUdW/mlWcCeZS0qjEXMMA9dMSEVxJ6NY7Wb1VX+v+b2Ph6BarHlDmvm6kjPMZbkpWef1fnFCnC+MWMgJxPemM/SoeUxVh3WLjldQxu6Lh5sDx4aeNxKKhaaxvWYVBSMSxvB0mtVrsfvQdQxvp7kHn2EABlhsI5ajToWzCtyGikQivPLKKzz66KO88MILsSUdZ/d/MkMDt6effpr777//pBVunZ2dpKWlne1LGEEmYmK0wpEI2/Yf45Vt+1i7dR+HaxqxaBrL58/ghuULuOGyBcyeloeiKOi6zr6jdby8ZQ8vb9nL7iM1qIpCXmYqiqLQ2NaFbhgU5WRwzZJylpaXkZLgorq5nb2Vtew7Wsux+sH/DlwOGwYQDIZHdH10OWyU5mdTmp/NtPxMSvOzKcnLpCA7ne7efl7asodNe45wpLaJLk/fhHxvNFUlLyOF4txMAqEwh2ub6It1R010OXDYbYTCESyaSmfv4BjcDjtzSvKZO72QuaVm04by0oJ4iCmEMEWjOqFIhFAoQigSIRgOEw5Hh30OhiOEw5FhQV04Yt4OhsKEIwOPjwx77MB5B563bf9R2ro9zCnJJxSOcLxhNF3bzoymqvHlrAPMJaaK2QFXUfAFQrR09gCwdE4pR2PL6wFuWG7uA+Pp87PtwFGm5WWyeHYp+4/WsmBmMQkuBxZNo7ykgI+9e9WI64uTm4h53nh47LHHCIVCfOYzn6GlpYVVq1Zx6NChUT1X5nliPMQDoL4ecylnXw9GX3f88+irx8bQ816zmoFbNDwYYNmckJge63JqgUhosAovMHSuN+R6seWzavY0lLzpZlOJk+ylpvu8GB316G116J1N0Ns+fBndaMdsd5qVbrHKN6x2cyzhYCzM7BxctqtoYHfEOqqa80clORMls8gM1BxuCPnN53nN5xp93YN/uLuSUFKyUJMzzccnZ6GkZA7b90oIISbKuARura2tPPPMM/z5z3/mjTfeQNd1SkpKOH78+NkNbkjgNrCHWyAQwGIx1++//vrr3HrrrfT19U3IO9oyERNjVdfcwdq3zPDtjV2H8AdDFGanc8PyBVx/2QKuWjSbBJf5jltLZw9rt+7j5S17Wb/jIP2BIKlJbrJSk/H2+2nq6EZRFBbMKOLqJebecfNnFNHQ2smh6kYqqhs5VN3AoepGGtu6RozFoqk4HXYUzOYOA6GczWqhOCeDkvxsSvIzmZabiT8Y4khNM4eqG6huasfrG3z3UVUUVFVBU1XC0eiwP4bHQlNVbDYLkWiUcNhcLqAoCoXZ6eRnpaGqCt5+P4FQhKqGlvjS2LKCbJbMKWVpeRnLysuYW1aA1SJ7eggxlYxYswJ/MEwkEjX/u45GCQRD+GIdZe1WC1FdJxCKcOBYHaX52YQiEQzdYMv+Sl7fcZBP3XY9kageX7b+0ON/5+53rjTDnahOOBKJN6J45E8vs2jmNOZNL0RVFB5//g0A3rHiElTV3Mvt72/sZFZxHpmpSfT2+XDYrUQiUcKRKJfOm873P3f3FH7Xzh8TMc8LBoMsWbKEhx9+ON4MKxAIsHr1ap555hmcTidf/OIX+cIXvnDa8/T09HD33XfT29uLrut85zvf4aqrrhrVGGSeJyaDEQ7FArhus9lBX7e5F9xAqHTaP8XOMIRTNHAlmkvBQv7B/aMU1WykkJhm7sukamZFiN9r7o3V7xl5HZvTrBZLz0crXYCSnn/SlQdm4OjB6OtG72xCb6/D6G6BgY3/z/Q1WWxmIOdIRLE7zMcHfeYG/+GgGdi5Es1Kvth+e0pWMdq0eajFc1EcbrOzZG8nRm9bfK8rvbc91n02dm13CmpKprlHX3qe2YXSJasrhBDja8yBW0tLC8888wxPP/00mzZtIhqNUlhYyB133MH73/9+li1bdvaDGxK4+Xw+0tPTWbt2LVdeeSUA3/zmN1m3bh0bNmw462udjEzExHgIBMNs2nuYV7aaAVx1Uxs2q4XL58/k+svmc/1l85lVbFa/DTz2te0HeG37AQ7Hmi4U52XitNtobu+i2+vDYbOyfP4MM4BbOpf5082943r7fByuaeRQdSP7jtbx1sFjVDe2jljWmeR2kpGSSHKCC6vFQre3n9rmdkJhs1pBVRXyM9OYlpeJzWKht99HW7eHjm7vsOWsA3tP2WxWDF3H2+8nPA57xg3sRzrAbrWQk55COBqlqb2b0oIs6po7iER1HDYrC2dOY9ncMi6bN51L504nKy35rMcghBAXs4mc5wUCAe666y6ee+65+DwP4P777+eNN95gzZo11NbWcs899/DrX/+a22+/fVxekzTHEucaQ9fB7zXDqt4O9I4G6GzC6O85eUOBEUYRyGlWs4mDKwlQY91UewYnWlY7JKSZ+6tpFrPKzO8B38Ay0uHLZRVXkhlQFc9DTcs2K9VOEVLpgT6Mhkr05iqzIq6va/C6I/Z5UmKb4Y9mE/zY61ZU87WhmAGfoqBkFaKVLEApno9qG17FZkRCGL0dZggXC+P0rpbBajq7Kx6+KWl5qOm5EsIJIc7KGQVuJ5t85eXlcccdd3DnnXeOewv5oYEbwH333cfGjRtZs2YNjY2N3HPPPaxZs4bbbrttXK87QAI3MRGqGlpZ99Z+1m7bx8bdhwmEwhRkp3PdpfO44dIFrFwyh0SX2e66oa2T1946wKtvHWD9zoN4+v0kuZ0U52YS1aNUNbYRCIZJS0rg6iXl3HzFIm64bP6IzqkdPR72Ha3jte0H2HbgGMcbWujx+hj6n3+iy0FORgqF2enkZaZhtWh09nipbmqnpqkNr29w746UBBdWq4VwJII/ECYYHr4BpsNmJdHlwG63EQqF6fb2D+ukatE0dEM/6yo5AIuqoqoqoSFLm/KyUlk+bwblpQW4HXY+/p5rpQpOCCHexmTM8yoqKrjrrrswDIN9+/bF53n9/f1kZGTw4osvxud93/rWt1i3bh3r168/6+sCfOMb3+CBBx4YcVzmeeJcZMQ6HOpt9bHAqsEMyiLjsOm4agFnAiSmgjsFDAOlr8vcF21oB1arHRyJoBjmdcPBkUGgqpkVaf5+tKU3mUs3BxoZKMNXIBnRMEZnE3pbHUZHI3pHvRmUgVlRZ3eBqpob4Pv7BvdoGya2P5zVFltm6j91laAaq/hzJaM6E8DhRol9MOQzho7h6cLoasLobDbDwYHltxYbREJYLnunuYw1OTO+R5UQQrydUQVuDz/8cHzypes62dnZ3H777bz//e+PV5tNyOBOCNx8Ph+f/vSneeaZZ0hOTuZLX/oSn/3sZyfs+hK4iYnmD4bYvPcI67btZ91b+zlW34LVonHp3Olcdclsrlw0myVzyuLLoHYcquK17Qd49a397D5Sg2EYFOdmkJaUSLenj5rmdjRVZfn8Gdyy4hJuvmIhZQU5J722ruvsPVrLy1v2su3AUSprm2nt6iU6pEJNU1Wy0pKYWZTL/OnF5GQko6oqPd5+apraOVrXzL6jdfFN2xVA01RzX9wTNnJXFYUEl4PkBBdpyQlYNQuBUIij9S0EQxPfscZqMZtOuBx2mtq7+ci7ruGT772OtOQEEl1Onn1tG9npydy4fHh3u0gkiqap8u6mEOKCNZnzvEcffZTKykq+/e1v43a7R2wd4vP5sNnMvaPWr1/PLbfcQn9//7hsHSIVbuJCYOg6hqfdDOFaajC6msxwajSdTE/LbKpAUjqk5ZpNCgwd+nvMpbDeruFLTwe6Gur6ya8du0+dNh81fwZKara5RFWzDr4WwwCfB72jAaOjAb29wXw90Yi50XpylrkU1uHGUFRzb7aOhpPvGzd0eYTVYXZPDflj57KA1Wq+Rj0K4RAjKgOHdFPF7gKLFcUwMII+s4PqwPktNpSMfBR3CmredNSicgnghBCnNKrATVXNPzanT5/OnXfeydVXXz2qic+11147LoOcKhK4iclW09TGurcO8Pr2A2zeV0mPtx+71cLS8jJWLJrFlYtms7S8DKfdRmePl/U7K3h9xwE27T1CTVM7ANnpyVgtFlo7eghHo8wsyuXmFYt499VLuWRWyWmDI8MwqG1u57XtB9mwq4IDx+qob+0cVp0G5hLPopwMFs6cxvzpRaQmu7FqGoFQmLauXo7Vt3DweAP1bR14+0/f1UpTVRJdDtKTE8hOTyErLYm0pAR++483cNisBMPh+P5tkyE/M5XUpARmFOWQlZrML55dB8Dvv/NvJDodJLgc9Hj7+davnuWln34Vi2WwDX0kEuUPL2/irpuvRNNk8iWEOD9M1Txv6BurzzzzDKtXrz5pc6y2tjYyMzPP6lpDPfLIIzzyyCNEo1EqKytlnicuCEbQj95Wi954FKO9zgzIxqNLKpjhmsVq7g1nscWWfmKePxKGcGCUS2Bj3ClYyi83q8WSMszGBgMdqvUoRnerGcB1NKC314M3tkex3W3utZaajepKQjcMcy+8tlqMnrdp4qNqse6xJ3xPVNWsrnO4zeq6aNRcsGroEAqaod3plrgqqhnUDXRJTUg1m084E8zlttaRzSeEEBePUQdu8SecQZVHNPp26+/PbRK4iamk6zoHqxrYvPcIG/ccZtPeI3R7+rFoGnPLCszGAXNKWTKnjOmF2TR39LBp7xE27z3Cpr1H4p1M3U474UiUUDhCdloy7732Uu64fvnbhm8DBkK4XYdr2Lj7EDsOHedoXQuBk1SluRx2Zk/LY9GsaWZn0emFTMvNpLfPx7YDRwmGI9Q0tXOkupGjDS20d3vwBUKc6p8hm9VCgtNBRkoiNqtGZV0LuRkp+PxB2nu8J33OZCvKTsdht2GxaFRUNQDmv5Pfvf8ubBYLmalJuJ12guEIn33oN7z8yNcoysmIP/+ZV7dx+YKZ5GWmxo8ZhkFDWxeF2emT/nqEEBefqZrnDQ3cnnjiCb7+9a9TW1sbv7+qqoqysjLq6+spKCg4q2udjMzzxIXOCPrM0KqpCr21Cjyd4xfCgRk2adpgAKfrgxVvZ1xxp0B2EZYZS1GyS8wOpgMhnN+L3tGE0dmI0dU0fN81qx0lNQc1NQdSs1HsLvRQwAwdj+0a3aVVzdy/TtXM4PBkAZvVbu6Hp0dRXEkYsf3uCPrNwPFUoZxmRXElxarnYhV0sc9mKJeA4kgAh2vEElwhxPlvVIHbyfa7GI3/+q//GtPzzhUyERPnEl3Xqahu5K0Dx9h5qIqdh6o4UtsEmE0Qlswp5ZLZJcwrK2T+9CJcDhvbYo/dfvA4e47UDNvnLMHlYOUlc/joP13DdZfOP6M/sgzDoL6lkz2VNew6XM22A5UcONZAn9+sZrNoKtGoHi/WL8xOZ+HM4ngIN7e0kOLcDFRVxTAMPH1+Onu9HDhWx+9f3sSR2ibauzxnVN1m0VQiUT3++VRObMgwFVRFweGwYbNo9HjNZRFL5pQyoygHu83KGzsPUd3UxsKZxXzqtutJcDlIcDm57YsPAbD79/+L2+nAZrHQ7e3jyo/9J9se/w75WWnxa+w7Wku3p5+rl5THj/V4++n29FOSnxU/1u3poz8QpCBreLjX6/WRnOiayG+DEOIcMVXzvKGB29NPP839999/0gq3zs5O0tLSTnOmMyMVbuJiZvg86O0N6K3V6C3V4OkY/4mR1WHu4ZaYBs4E1ORMjFAQw9OO0VYP0ZBZPXbSPdpGUssWo5QuQHMnm+dVNQx/H0Z3C3pXM0ZXC0Z3M4ani4FmCkpiGkpKJiRnotjdZmOK7lboaR2+JFUxK9wI+WKfT9gTzpVkVr9Zzeo+o7vF7J4ajWKE/BDof/sXMNAgwmIFzQbRMERCIx8TC9/iQVw8lEuMhXWx40OW5Qohzm1j7lJ6MZDATZzrevt87D5czY5DVeyoOM6+o7U0d/QA4HbYKS8tYN70IuaVFTKnJB/DMNh/rI4XNu5m64Gj8a6kFk1lemEON12xiPffcDmzp5289fvpDFRl7a2sYc8RM4jbfbianj5zUmONLb0cWJ7qctiZW1YwLISbP70It3N4Rymvz8++ylp2Ha5m6/6j7DpcRVu3J95wwaKpRHV9XOeKqqKgDznh24V454qMlEQsmoZFU2loM5df3LLiEhw2K5qm8ud1WwH4lztvJCs1GYum8fWfPQXAbx9YjcNuTuA27j7MT//4Ev/2wXfw0X+6BoumYbVo3PH/fsC+o3V0r18z7Lqrv/srfIEga77xL8OOP/r0KyyaNY3LF8yMHwuFI2zYVcENly0Y9tj9R+uYN71w2O9dR48Hl8Pcd2+AYRh09HjJTB3+b7Ku6/R4faQlD28YIoQ4dw0N3Ab2cAsEAlhiTW5ef/11br31Vvr6+sZlD7cTyTxPCDCiEYyuZvT2eoz2evS22sEQSdXM8Oms94cbKtZh1OZEm7EEEtMgHCTq7TbDP0+HGYidrhJPs4AzEZIyIOhDm7kUrXgeYJjLUWNdSPWedozetsHGDIpqNnRISIVIGMPvBV/vyEYUimqePxI0O5dabBiBvsHzgLkMNSHF7GLqcKPY7GZgZ3VAyIcR9JvdZvt6MbydIwO2oeJLdVUziASzoi4SMivoTtxvzuYwK+Vi4ZySkoWaU4KSnoeiaiNOL4SYOhK4nYZMxMT5qKPHw4Fj9ew/VseB4+bnytrmeBOD/Kw0ZhXnMrMoj7TkBI7Xt7L7SDVVja3xUMlutTB/ehE3r1jEioWzWTy7BJv1zLt8GoZBU3u3GcJV1rL7iBnCdfb2xa9jtVjwBYPouoGqKswpKWDpnFIWzy5hSXkps4vzR+yHFo5EOFrXEn+Nb+yq4ODxhvhrdNptBMNhlpaX4enz4w8G6ejx0u8f3F9EUcBhs6Eo4AuE4sculn8RVUXBwBjz601yO3HabVitFhpaOwFYPn8GVosZzimqyqvb9gPw8fdci81iQVHgZ0+/AsDsafncfetKALZXHOe5198C4Gdf+ThOhx2n3cYH/v1HAOx48n/QVBVVUXnsuXU88qeX2fvUgxRmp8cDum//6lkeeuLvtK79vxG/q6nXfJTPf+id/Mcn3jfs+Eub95CWnMClc6fHj9W1dFDX3MGVl8we2zdGCDFqQwM3n89Heno6a9eujTdq+OY3v8m6devYsGHDuF5XKtyEOD2jvxc9Fr4ZLTVmaAWxQGkgCBpSJWaxxRsdjMuSVUesqsudhBH0m00O+rrNcxunW8qumFVkDhckpKLaHJCSg5pRAOEARqAfw9OBMRDEDQSLigJ2t/l5oNHCSangTkKxOcGimbejYYygL9ZVdcjYElJRElJQXMkoriQUdxK6ZgVfL4anE6O3HXraBq9lc5qdV4O+U3eidbjje+gpioIRDkEw9hqiEXPJa2YhWk6JuY9ceh6KdubzdyHE+JHA7TQkcBMXikAwzJHaRg5VN3KktomjdS1U1jZR1dgWD6lSEl1kpSWb+6N1ewiGBycbdpuVKxbM5LpL57Ny8RzmlhacVbVBc0c3eypr2Xukhj2VNew8VEVHbE+25AQXqqLQ0+fDMAzcDjuLZk1jcXzPutJhSyeHvsaK6noqa5t5Y9ch/vDyJiyaRiQaRVUVSvKyON7QyoffsZI5pfk4rFZau3rp9vZzvKGVV9/aT5LbiaffP+Lc2WnJ2GwWDAPzexMKY9E0NNWshDuxqcT5JvZe89ieO7QpWKyKMarr8QpEAJvFgqoqI/b9O7HqcSxsVgt2mxXvkJ9baV5WvJlFKBKJNxS5+YpFZnCnKmiqyl/Wbwfgk++9Dofd3NT4p0+9CMDnP/ROUhJc6IZBY1sXb+4+zNc/fhs/+N0/6PcH+d6//TNb9lXy3d/8lZcf+RrpyQlU1rXwf8+u41f/eR+GAZ/45s9Zv7OCrtd/HR/bG7squPP//ZDmVx6Lh4WhcITLP/I1Hvv6p1gypzT+2OMNLfxtw04+98+3xo8ZhsGfX93Ku69eNixY7PH2k5Lojt8+XNOILxBk8ezB8wE0tnWRmZo07LmhcARdN+IVjkJMlhO70d93331s3LiRNWvW0NjYyD333MOaNWu47bbbJuT6Ms8TYnSMQD96ay1Gaw16Ww1Gd6xBgTPRrPLSrBjhgLlH3EDzBJvTXP7pcJshWCRiVpQF+sxQaar+BFU1s4ItJQvcyaiKiqFHzPGEAhj+PvB2nvr5Fpu5b13s8cMMBJIKEOhHcadgRMOx5gsBhs22bA5ISEOx2TH8fShWG0bAB/29wyoKlawi87zhoNlgQo9A0G+Ghx0Np3+tigruFJTUbNS0XAC0uSskhBNiEkngdhoyERMXulA4QlVjK5W1zVTWNlFZ10xNczvVjW3xAOxknHYbC2YUsWLRbG6+fCGLZ5eeVVdOwzCoaWpn28FjbNt/lLcOHos3INBUFV3XSU500+M138XLzUhhyZxSFs8uZU5JPjddvvCkS2AHQrgDx+o5XNPEo39+JX6fw2ZlXlkhgVCYGUU5fOczd5GTnkKfL0BTezddnj7+tmEHXp8fp91GR4+Hjh4vHT1eOnu8tHV7hl0rJcFFJKrT5w9w+fwZJLqd9PkDHKltpt8XoCgnw9yvrt9Pa1ev+c7kBfbPb3xz4yGva6Je59BlvokuJzarRpenLz5/L8pJR1M1NE2NNxApyslAVRQURaG6qS1+rqKcdKK6Qb8/GP8dA7PhiKqqw8K8sbBaNCyahmEYw0LHwux0Et1Ouj198aXgNyxfgKaqaKrK8xvNzZ6TE1xcsXAmFs1szHG8wfxD5+ufeB/hcITKumaee/0tEpwO/uOT70NVVb70oycAeObBL5iViBaNcCTKO/71fwDY8H/fIMntxG6zUn775wFoW/d/WDQzqPzHmzu5+z8foenlX+CMhZGGYZC26mN8+vYb+c5nPhh/He3dHr72yB/4yZc+Fg/tdF2ns7cvvvT3L+u3k5mSyIpFg5WDvkAQRVHi52/r6kU3DHLSU+KPqW5sIyXRRWrS4FLhzh4vUV0nKy35jH8WhmHQ3u0Z03PF+DsxcPP5fHz605/mmWeeITk5mS996Ut89rOfnbDryzxPiLExgr7BAK61BqM7tvdiQgpqWh44EwAFo63W7JY6sJzSakdJyjC7k6bnotgTwGLFCPnNZaDtdWbX0VM1LpgsFhu4EkC1QkIKCgr4vRg+j1kVd7IltgONFzQLoJjBotURCyBPnAcpZmgXr25zmN8j/YTzKqpZNahHBwNKRTGDN4cbNbvYXNJqd6LYXWY1YNCH3tuO3nDErAwM+kaOV7OYS2XdKSjJsZ9HYpq5f5zdaQakzsQz3l5GCDGSBG6nIRMxcTHr8wWobWmntqmd7RVVbN1fydG65vhy0BM57TZy0lPIz0pj4cxiZhTlkpuRQm5GKnkZqaQlJ5zR/3H3en3sPFzFK1v38otn1sWDm5z0FPzBEEluJw1tXRiGQVZqEkvLy1hSXsqSWAh3qj+ovT4/+4/Vx6vr/rR2S/y+3IwUyksLsWgqeZlp/Ncnbz9l44Ber4/KumY6er10xsK4wzVN/PGVzcwrK6Srt4/ePh/9geBJn5/gdMSbTHziPddisWi0dHTT7fGxZX8liS4HiS4n3n4/Hp+fSDR60Sx3PZcNLJmN6gbBWHiWnZ4MhvnfQE2zWU2Xk55iLtluaI0/bsGMIhJcDjbvrWReWSEHjteTn5lGbmYqOyqOAzCzKBe7zYqBwfH6VvzBENPysshMSaS2pYO2rt74WFISXditVjAMWmMBsC02tuiJk/az4LTbCEeiRIZ0pFw0sxglttfh3srBrpLvXbUMUHh+4y5C4QjvWHEJiW4nf3xlMwBfvuefMAyIRKP88HfPA/CDz99NKBzhKz/9PQBPfPMzRKM6oUiEe7/1GABrf/Z1vvGLp7nrlitZ/d1fAXDgT9/HatFQVYWnXt7Mfzz6R/b/6fvkxzr+fu2Rp3j0z68M23PwsWfX8f9+8juO/uXHZKTI/69frGRJqRDj65QBnDsFNavIDIisNoxIGKOtFvp7zSBuaKhmtaMWzkZJSDWbLdgc5tLKoA+9uxW9u9lcgjmw3FKzmmX2A6HVuO4zdwqKYgZrug5pOWbIpajmsk6f1wzXwief95nPV80wS1XN8RuG+XpiTRiIRs3mDac+AW+7JkHVUNJyzUAtPRc6miA1B5xu6GxC7+sGbxcEfG97LiW3NLYkNhHD50VJSEXNnmbuIWdzYPR1o2QVSzAnxGlI4HYaErgJMVI4EmFHxXFe3LyHV7cd4Eht04g/7k9sOgDmfm05GankZabGQrgUcjJSSU9JJC3JTaLL3BNs0axpJ71uV28fb+4+xIZdh1jzt9cBc96Tl5FKcqKLqoY2NFWNB1xlBdn809VLuXTudJaVl5GeknjK19Tc0c32g8fZe7SWg8freXnL3vh9eZmpzCkp4K2Dx7hiwUx+/tVPDlu6N5rvV2VdM97+AH2+AH3+AN5+P16fn0f++DJNHd2UlxYQCIYIBMP4gqFhVVYnKsnLwmG3ousGrV099Hh9lBVkY7FoBIIhgqEI7d2etw1cVFUBgxE/JzHxLJqKy2GPL1/WVHVcA7Kh7FZLfHl4QXZ6fL+93IwU3E4HkUiUmuZ2FAVK87Pj1XMDCrLTyUwZfJd71+FqAHLSk8nLNJd2N7R10tblIcFpJxCKEIlGyctMpam9m5yMFNq7Bn8fUxJcePr9pCS56RoS3ls1jXB04qoZslKTzOq2IZW7973vBv7n/rsm7Jri3CfzPCEmxrAArq3ODOAM3QyDUnPMvcXScs2N/xXQ649A0Ifh95pB3NDOn3Y3ijsJxZWEoagYgT7UzGIIeDH6esyKuEDs/080K7iTzQq7gSWfQT+E/afeF20iOBPMLCscQEnKMPd40/VYd9LwGMNBxdzjjVi4Zeinb8QwGq7Yv3v+PvP7ZbGecE5lcPns6SoOE1LNzqmahtHVDK4k1MwiM1hUFPT6I6iFs8yGFaoFAn3orbVoZYvMJbjRCNG6CrQZS80KO0U1u8q21mKZd6V5HlUzPyJhM/Cz2mPfhqi53NY++Aa5ETWXCCuW4VtlmLGHgaKMfxMeIU5FArfTkImYEG8vFI6w/1gdOyqO88auQ2zdf5Quz+Af0ooCSW4XWWlJZKYkk5RgVm21dPbQ0tEzogIsye0kOy2ZrLRkEt1OXtq8h8/ceROXzZ9BTnoKORkppCS42X+sjiO1jeytrGN7xXEOHq+PP19VFdwOO00d3fGqsNQkN92efr73bx9i6ZxSZpfkx5eyncgwDA5WNXCouoHDNU0cqm7gxU174vfnZaRSVphNbkYqf1q7hd8+sJqrF5fHq+GCoTA2q2XM7/j1ePupbmzD6/PHQrogTe1dPPDYn7n71pWEIhFC4ShtXb1s3HMYu9WCpqn4g+FRLd9MSXRjiS0BHlg6PPD9sVksRHQdfYICIHHuyM1IxTB0WjoHq+YKstJoaOvCZrXEuxg77Tbys9KwaCqHa5rij01NdBOJRvH6AigKWDRtXPczdDnsOO3WEVW10wtzOFbfwqqlc4nqOm/sOsTVS8rZsLMCMMO1d65cgqqq/PK5V5lTkk9TezeaqvDhd6zkx0+9yBULZ/Huq5dy723Xj9t4xflH5nlCTA4jGsboasXobETvbMToaDQ7dxpGvHOokpyJkpKJkmzurUY0gtHZZDYl8HnA50FvOn7qsMqdjKJZzf3Ygv7BSjNFNe9LTBsMc9zJGP1eaKsBv3dql68OBGgY5hJUDDOsM3SzQcS5Ph8b6GRrtZvhnL/fDAZtztiy2Sj095i3wfxeD4R6ijr2ysSBbqwDP7uEVLNDq6bF9xhUckpiDSZUUFT0hsPmU2ddau5jN7AMGIge2gKhAJbL3xPb508xPwC9sRI1qxglowDz52OArqM3Hzf3xDuhM6wR9IOmoVhO/neGuLhI4HYaMhETYmw6e7wcrDIrxbbsq+R4Q+uIZgRpyQn/v707D5OrLBP+/z2n9qqu3vdOd1ZCNkKILBFCCILKCDhmiCCIojIuKIs6g4joAAJmfJGRlxdHf8gMMvITERFm5DVsspOFPZKd7J3e19rXc573j6e6ujtLZ+ukk8r9ua6+uvucOqfOqVMn/eSu+7lvTmisZdqEeqY21VNaHGD1pmZqKkro7A3R0RtiW2sX76zbMuJzfXzebGrKS/B7PfSGovzxrys4Z+4MWrp687W7QNdsG1o/yzQMGmsr2d7WxQ//8R/49IJTmdRQM2Itug93tLHqw+2s29rC1pYONmxvy9eaA/0f/cbayvwxP3z7t5hQX0V9VRnlxUVsb+umqbbykOrdjUTlGjgkUml2tHUTjidIJNOkMxmS6QxbW7v4/bNv8Kn5pxBPpMhaNqs37eDtdVuYd9IJWJadnzq4OhfAPBSmmavfpg6+KcN+Pc/AgCj35yzg9eB0OEhm0nhcLiLxBA7D4PSTTsCyFCtXfwhAsd+H3+ehrrIUw4C1W1pIpjNUlgbp7o8Ma6IxNAC1pwxOMfqcDgeWPfJU6qG1/A6Ew2Fy8glN/PVXtx7CEYpjlUwpFWLsqXQS1deBCnVi93eiQl2o/q7BbDXD1EGy4gr9vahM11MrrtRBkmQMlYxCPIK16V2MkkrdTCCdhGwG1ds68gHoJxmskQYQrBgMCB3ROnK7HAe5YJHDOdhswsrox8X69QMcTh2MOxJTaQ830wG5bEcMB8TDUN2EgQKnW9fQMx1gmjr4alt62rFSqFg/KhbGrG4EW4GysXtaIBbCbJyOUrnXyLZRHdv0tmU1YFk6E862AKWz/A6Ww5U/Pgwz/x42aiaAw6kDcoaB3awDfo6ZZ+ltnLlp0bEQ1vqVOM+5TD9+IEDYshEcThwnnq6vd2455IKA42fKdN5jgATcRiABNyFGTzqT4fX3NvCXN97lP/77JXweN4nU7qnwQb+XCfXVTJ/YQCZrURzw8blPnInP6yZr2fSGo/RHYvyf3z+DaRrUV5XR2RumszdEc8fwrlJej4tivy/f4GDyuN2nzA3ldDjwed1EYglu/vJnuPTjH9WF9kfoyJrOZFm/rYWNO9rYtKOdzS0d/PGFFbs9bmjzgBsu/xTj6yppqq3E63bx1trN3HD5p476P5o9/RF6QlESqRTxZJq+cJQPm9tprKkgmdLBzLfWbqa5o4czZk1BKYVtK2yluPu//ocFc6dTHPCRtWzauvtYtXE7JzTVkc1a2ErXHRuY8uh2OclkLQyOnmmvLqdTl10xTZLpTL4Ta1VpkHTWwjQM3G4n3X26qP/kcTWYpoFtK7a3dZO1LAI+Dw7TZNqEepLpDOu2tgzLCjtpSiOWZbN5ZyepzGCAuLI0SDSeJJnOUFUapDsUoTSosxIHj8+BZdm4Xc7dOsKKPRta300cf2ScJ8TRR6XiqP4uVEgH4exwj55mGgsNBqRMJ0ZRie7AWVSqi/8XleV/xhfAMEwdbMlmIJNCZVKoeAR7xzqdeRXrRyWiuitpLogzjC+os98GMqFsO9e8YAwDXE4PeHwYLg8q1AXKxjH7XF2rzeXV2V3b10JxOWblOH3M2TQqEcUOd6PC+nVUPTv3HKzz+HUAKJ3c/fUYCBAppV8/jMEpsgOUPXbdZ/fHQIdaZYM3oOv/2RYEy/V7yuFEOZw6cBYN6YxLlxuVSWOYTjCN/NRWe+d6DJcXI1imp8kOvJZKYbdtBsvCqBk/pOGFQnXu0I/xF+vnHqg/eChTgwem8zp1sw7V3wlWBrNxeq5eoD5ne9M7GMUVmFPm6mC2aWL3daC6mnGeeoF+bC5wqDq3genAnDBbX2vD0PeTbUM6oTsPD6FScd0QZZdMP6FJwG0EMhAT4vAKReKs3tLMu+u3sGzVBp5ZtmrfG+V86dMLaaypoLa8NNcZdCozJ49jU3MHPaEIrV19dPWF6OqP8Ogzr9PW3U9tRSkdvaE9Trv0edykM9kR69FVlhbz2zuu5dTpk3A6R/6jopSiJxRhe1s3bd19tHb1cdN9/z9ul5PailJaOnuHPVdpMMCkhmrqKsv4YNMOdrR388sffJUTGmupryqjuqyEp197h/PPmE3A59nv16lQKKUIRXUhYYfDxOlw0NbVRziWQCmFYRgYBvSGY/SGoricDhKpNIZh8O+PP0tNeSmZbJZwLEE8mWRHew+xRAq/141pmGAoEqkMlmVjGEf3ePFoNlI5572t23X5fpSEHjUScDu+yThPiGOHUjbEI9j9nRDp0fXbYv36e7Qf0rt0FA+UYlaNGxKM0wE5PfXUOXzftq2nuMZCqEgPdssmVLRP14cbCKT4izFKqzFLq3WAxukBZaESMT1FNtShmxEcLQMIb0DXNXO6wOHWQRTAceoFunae26tnIPS1o8LduptprhFCPjC0K8PUAZihQTaXFwLFGN4i8HgxG2dg5qbv2g4Hhm3nshHjqHRCX6dUUv+cimP35majZFM6AJVJoQv7jXFwc5gjOTLZw3ObDrBzwblc7Toyaf2zUjrw6XIP1tqzsrrTbnZguT14zQwzHwA8aG7vYNMS0AFrwKioH6y1ZzowTAd2qBMSUZ2x6fRguL1gZbBbN2HWTkLZFmZFvc5WTMUhGceoqEMl45jVTToLETCsDHZXM2bNBAxfUDdg6W7GbDhBZ0VmUqhEBKO0OndcBobThREY+670EnAbgQzEhDjy+sJRNmxvY/22FjZsa+VXTzy/18c6HY5hnRMdpklDVTnj6ytZt7WFifXVfP2Sj+uAVXkx6YxFXWUpkXiS1q5eXnxrNS+s/IA5J06gpbOXjp4QbT19dAypabU3bqeDdC4z6fILzuTyT85n/pxpGIZBOpPF7XKOuH02a9Ha1cfW1k5efmcNPo+bnR09tHb389c3Pxhx28Xnz6OmvISKkiC9oQj3/+FZ3n5kCRPqqvPTVdOZLC2dvUxsqN7nuYg9iydTRONJOnpDtPf0YxoGsYRe9q2f/geXf/IsAj4PkViCWCrF06++C+g6aIlcA4uMlSWVzu71OUzTwGGamIaRb24wMP25KBdYVUAsoWvR+DwuXQc6k9XN2eyD/xMe8HpwOR3EU2m8bifhWBK/10M8maKmooRILIHf6yFrWfRH4nhcTkqCfvwefVy2sonEk1iWTTiWwEA3WQjHEvngKEBVWTHJVBqP20XWsumPxHC7nCil8OeOYaCW4EBdRcMwiA+p7+h06AB3dpQbK0jA7fgm4zwhCodKJ/MBOLv1Q8ikUfGwDsbFQ4OBMMMAX7HOhisanhlnFJXqwFouU0dZWVS4B9XfiervQPXr6a9E+waf2BPArGnCCFZg+INQPQGViulurL1t2KFuHYjbE8PMTRU09tzd1DD0Y2wrN+XU5rB9Kuh06TprTheE9WwDo3F6LmChUF3NqEw614givv/BMIdTB+Vcbp2J5Q1g+IrA68dweXVwyOXBKK3RQU/DxCyvy22sA0hKKZ3tuPEtzCkfQWWS+vpG+3S31EQE1d2ia//ZFmSzunOs6dDHaQ1k8x3lYY98hpihA2wDWYWgX3OAQGku68yhzzEZg+LKwe63ybhuKKFsVDqF4QvoYFYyjuHxArrGnOH0gMPU1zTap4PI6YQOJNuW/nI4czX5csdg66m5ujZfdvB9mH8vjGVgcg+KyvAs+vZYH4UE3EYiAzEhjh69oSibd3awpaWD1Zubuf+xZzjlxAl09oXp6OkfVsdp10Dcrk6a0kR1WTGpTIaOHp0F99//diN1lWWUBQMk02neWbeVUCxOT3+Els5eWjp7+e9X3t7jNNihBjLifB43550+i6baSu761uVs2N6Kz+2mqa4SgHAsQXHAt9f9KKXoDUVp7eqjpauX1q4+/unn/8VHZ08FoKsvTHd/ZFhHU9M0qCgJUlESZP22FgB+8JVFjKsupzrXiMK2bf724Q6+cOGCYc+XzVr7zNoTB0cpxc7OXvojsfy00OdX/I2O3hCnzZhEKpMlncny2nvreO299Vy84COkc8uiiSThaIIN23UtGp/Hna+zty9Dx+ROh4nb5dKlQnLBu3HV5XqqM+Sn8jodDqrK9N87W9kkU5lhwbOD4XY5sW2F1+OiLBjQwTnDoKosSFt3PyeOr6e5vZuK0iCTGmpwOR04c9NjvW4XLqcDh8NBJJagqqyYjTvaOKGxloqSIhwORy7j0WTj9jYmjavGthWJVJrxtVWEYwnauvuYPqEBWyncLiebmtupKA0yb9YUTjph/CGdmzg2SQ03IY4vyrZ09lq0X2dwDXyP9aMifYN140AHuPzFGIGSwaBcYMh01UCxDgKFe1A9Laj+Ll2DLtI7WF/N6YZsGqNyHI4pp+hGEG6fzp7r2gmhTuzunTrIZu3lQznTkavxZYLHlwsiJcDawzjU4dTPaZo68ymbHp1Mpn0ZmGo78Bz+ksE6ZpFevdzp0XGYgcyrUTEksDO08cJAA4f9CAYaZbXgcutsPrdPN0Qwcpl16Gm1KpPAKK5AZbPQvlVPBXU4h2dTWlmdYabU4QuGFjwTGHLNzFwQ2srooCPG4Pve6R58nbNDlg0UjLaGZF+6vXguu/nInMIIJOA2Agm4CXFsUErRF46xdNl7dPSGcDoc9IWj9IVjtHT28sI+MsZGctHZc6kpL8Xv9fD8ylWcfMIEFsydzkemT6KzL8TGba088eKbLP9gIz6Pi2QqM+LQxuNy8oWLFvDgky9y8YKP8PDt3yKeTBHwefOPSaTSOB0mLufIWXIAmWyW55b/DdNh0t7dR09/hO5QhFffXce6rS0E/V4i8eRu21WWBpnSWEt5cRH90TjLVm3gorPncuWnFlBbUULQ7+Ovb63mE/NmM76uatjz7c9xicNLKYVl2bR09RJNJHU9OQV/fHEFs6eMJ53JkEhlyFoW//V/X2XxeWdgmia2bfPCyg94Y9UGPnXWKdi5RhutXb2s29qCYRiMqy4nmdYdbwcy0g7G0HFnkc9DNBfoqygpwrJs+nOBvOryEmJxnV0XisUpDfqxbUV3f4TK0iAup4NYIjWs8UqR36uH2wqiicH3t8M08XpcpDPZfXZMHV9XxfuP/q+DOjdRGGScJ4QA9PS0WEjXdYv2oQaCc7mMORKRwQfvmiE3pI4cLjd2PAzRPqx1K3KF+NXgH8NACWZJNUZxBfiKUIaJo24yKhnTz9XXjgp1D07xHClA5fYNZsd5/ZC1ctlQCsPK6tp0u06zzZ+DOZgxN2bTNnNdQIc+/0hdS51uqGyAWBjKanDWTMCobNDLnS4MX5HOoMtRtgXJOBiGvqaJqO58292s65RlM5BNY29ZhVHRoKdnZlKQSaLSqeFB2F35i3OPTWHUTc53PLW7WzDHnYBZUo1yeTAGgo9Ol+5cqmxw+yHahx3txfQGIZvSDUPiIYziKlS0Tx9fqEu/B0qqdbApndTnmpu+icM1rD5cwRv63hjofqtXDJ9yO7AMBcVVeP7+2iN8oLuTgNsIZCAmRGGJxpN094cJxxKEowkeWfoatRWlxBJ62mBvKMqqjduH/Qd+bwzAsUsm3byTTqA44MfvdfPUy2/llw8NNuxNSZGPBXNnMLG+mvt+vxSAW76yiG8s/gRFfh2MG6h5d9bJJx7QeSdTGTr7QnT1hdnR3s0dv36CC+efQld/hN6wzqJbM0JX0qlNdZQWB4glUqzZ3MyMSeO46qJz8lMfp+SyjcpLgjTWVOQbWHxk+qQDOk5x9Epnsjq7U4FCoZQiEk8SisTpi0RRCsKxOG3d/ZSXFOF0OAhH4/SGo2xqbmdcdQWReIKnXnqLC8+ei9Ph4Jd/fA4gH/izbcVzK1YxeVwNyXSGls5easpLcDodJFNpekK5rl9DGpAcis9+fB4P3PL1Q96POHbJOE8IsT+UlQvIRUOoWF++btweA3IY4A8OBuH8QQzTqWtRpZOoeAii/Tqolt/E1MG4shodjMuNLY3G6XpqYCYFkV7sjq2QiKFCnTrgkk3vvZPqQIDC489/N4orddCpfQtGVZOuxWVldPApk9IZdLa190Dd0W4gABko09mG2TRG7UTMpplgmhjldZheP3Z3C0ZZDWZxZX5TletkOrS+n4r0opLxXNONpJ626vaCndWZjX3tmGW1OuPRtlBdzeSbSey1w63BwCeGRrBCH7Pbi+ENYO9YixGswJw4K9+Iwcgdo0pEMOsmY7jcWDvW4zjxtGFNCpRtYbdvxSip0oHaXEBxoEOr3bpJ1zezsigrt87KgunA2vQeZtU4lGWhmtdhVDTo94SVGZw+bToOrmuvw6njgUMDYgMdV4c1ixhoHDeKoSlPAM+l3xu9/R0kCbiNQAZiQhzfEqk0vaEokXiCWCLF7Q88TlNtJc0dPZiGQSiWYMO2FuJJ/QfDYZq7NV0Yyf4GDvxeNzXlpWxt7QR0J8tZU5r4h3NPZ0J9NW6nk5Mvv5H/+vG1XLzgIwd3sujstYGOr9FEkj+9+CbpTJaSIj/9kRg9oSjPrdCNLTwuZ77m2FBDu8+eM3cGDdXl+L1uXntvPZd94kymNtVRUuQn4PMwrqYCy7apKS8hlkgRS+jaYUKMRCmV74AbS6YwTYNILEk2a+FwmLicDjJZi7ZuPUjcvLODTDZLRWkxlmVhmiYO0+SMWVMoDQb28WyikMk4TwgxGpSV1VNWhzRxyP8c64d4hMFAQi4gFyzXUxlzXTFVOqmLxsfCuubcUA4nRmk1hq8Y/EUY3iKd0VVUrj8ES0TB5UGFu1Ctm3WGVCoOvW0HfjKGqTt4DtSPAwy3BwxTP4+Zm6pqZQfrih0WR6AemOkccj65qYj1J0Ayilk3CSNQimrdjN3bhmPexZhFZRimSXbTuxhFZThPGD7mVomIDqT5i3VQK5PWwbpUHFIJ/d3KolJxrNWv4Zh8Sq6RRFLXC2zbDN4ifUz78/oOTCV2ewdr6wFm0wz9Hoj1Y7h9uWYGJtam9yDah/Psz+p1xZUYDhd253Yck+fk68WpRAyzvFb/H8W28kFI27YwDFO/VukUKqunL6toP/lMzmwGMkmyq17GMe2MId1bwVr9KkZVo84mtG1UT4sO9BaV5aYcW/qyJw9uZsVuSqpwX/xNfcxjSAJuI5CBmBDiYGSzFj2hCL975g227OzA6TTpj8QJReP0R2K8t2HbYXtun9vF+LoqxtVWUFoU4PkVqwjFEjxwy1eZNXk8ZcUBnl2xiinjamnr6WPxefN4Z90WTmiqG1ZTLpFK43W7MAY6EAGpdAalwOvRf5BtWxfN39TcTiSW4O11Wwj6vazZ3Mxv//IaF5w5h57+CPFUesQMuqHBu08vOJV0Nsuzy1fx/S/9PZFYgvv/8Cx/+Ol3OKGxFr/Xw9vrtnDBR0/Ol8swzbH9QyqEODbJOE8IcSQoKwu5Bg75GnLh7tz01bAufj+U0w2+It1UYKBovpUBy9JBjnQql4U25L/xpkMHMlweDJcb1deBOenkfIYUHr/uDtm5AwUYKheoS8X20NTB0N0tBxo5BEr1c8VCej+ZJPlOovtr11pupnOXaYCj4QgX7R9oeqFsfc1yAS+jYaoOWjpcGF4/qnk9qq8d58IrMFxu/do6XNhb3scoqcaceFI+W83asRazqgnDV6QDXpkUZDO65h+Aw4G9bbXu7Am66UEmqQNaa98Ap1sHtWwb1bFVb+ML6unDu77P9vc0gxWA0nUKi8owiit0oK55nT6kGWfpB9pZjNIasn97GcekkzHrJuuGDZkkqqdVvx/9xYN1/qysbvCA/jB16P85lG3r94mda86QSen7aGAarWGg4pHce8rW+zNA9bbr979SGIFSHE3TD+qcR5ME3EYgAzEhxOGWyWbpC8dIpjP84fnl1FWV0dsfob2nn9Wbmnn1vXX4vW6ylkU6M7odGkFnzw1k6M2dNhGnw+TNNZvz62+5+h9o7+mnLxzjTy+uBOCVX99Gkc+L3+ch4PMQjib4+DfvYPlDd1ES9A/b/2vvrWPutEkEfB4y2SyhaJy+cIxIPEFrVx+rNm7H63Gzubmdx55fxsKPzKCjN8yazc1UlBTlpxGO5BPzTs5n3t397S/gdTv5w/PL+dLFC5l30lS6+8P8872/5el7v7/P7rFCiMInTROEEEcTlc2g4mGdJZeI6kypRER330xE9c/xyC5T8NDBHo9fByCcLh3MyKbBW4Rq2aiDPra95ymiHj+Gx6+7hnpy3UlNPUZSsTCYBircC9FeKK7Q003Tid2bAri84PboYBxgNkxF6ZAepOKo3jbyXTeH8gX1dys78hTW/W6EsIeacMcyfwmYhg6uBsuhRzdDMybMwvAFc/X9FPbGt3HMnI9R3QSZFNnXn8B51j9gVo4DhwNl2WRf/yOucy7DCJToQFY2jbKyZN9eimPibB3sMwwd0AI9tTVQCoDq78QoKgMU1tplGI3T9AydbAbVukkfa1HZ8M69B8JbpN+zuem/hunUQWXbwqxswA51o9o24zzjIv1+x0Cl4qjunThOOkcH1+IRnTGa695quNyH9NKPNgm4jUACbkKIo00qnWFLSwcffLiDVR9uZ/22Fl58a81YH1ZeY00FyXSG8bWVWLadz+b7xuKPU+Tz4nE5+d+PLiWaSPLE3f9E0O/D7XLw+Asr2NrSySN3Xpf/hOu199bRF45x6ozJdPWFCEXjRONJHntuORMaqnjxzdVYts3E+mr+8sZ7wL471J558okUB3zs7OjBMAwWnz+PB/70Atde+kmqykr4xzt+xcO3f4tPn3MqlmXjcJjSKEKIAiXjPCHEsURlUsMCcAOBObuvA9Wvy44Mryc3yCirBacb1deGUdmIivTqenG2lZsemNJTINOJPWeueYtyQb1ctpTHn+t0rnTmXTwMoS4dKMok9b72mW2Wy0hzeTFnzscIlIBh6KynTEo3OPAEIJvGWvO6DuZlUvk6eIfE4cp1HE0OHqfbrwOD2TRHPFvuSDIcgD0YPPUEdIZYMqJfk0AZhLty6/w64OcJ6Gy7zm3gC+KYeRb4ghiGiYr1Y73zLI65H8eobNJ1B9MJjGwaOx7C8AYhEUalkhjBUp25mUli97bra25b2NvXgj+oszQjfbp5hK9oSPblAVwPXxAcTszyOlznXDa6r91BkIDbCGQgJoQ4lqTSGTp6QrT39tPR088X/+UXzJrcSEVpkOb2bvoiMfrCo1QXYT+4nQ7SQzpFup1ObGWTtfb+6aNhGBT5PKSzFqm0rqdx/hknsbOjh+ryEk6eOoHf/PllIrEE93znC5QGA1z941/xzc9+gkULTydr23T3hensCzNtYgOxRJJYIsmXb/sli887AwyDaDzJs8tXoZTK1/va1dBadKDr0fWGo3ywaQffu+rTVJYWU19ZhsvloCxYRCKV4iPTJ/P7597ggo/OoaG6nPfWb+WVd9fy7SsuHLbvWCJFwOc5xFdXCHGoZJwnhCg0+emryTiqr03XButp1dP4kjHslg8HO3LuKle/zfD4dOaaaergm+nQdeY6mzHKanQTh2yuRlk2tecAnS+ot8+kMUoqdYArndBNCFwePUU2Ht77tFSnS2dkhbr1VMSgDg5aW1bhOv+LGN4AyunGsDJ6v6mErpOWTUMyhrVlFWZ5HSoVx976Nx28cbr3MH12Pw3UtgP92nn8OrgY7dXn6S/RgcienUPOwa2DdwMdYcX+8fhzjR9y43CnW78HnS7980DDEZcHSqv1NU3G9LTl0ioMZev3fLAcz2duGLvzyJGA2whkICaEKDTpTJb121ro6gvT3tNPW1cfd/3nkwB8ZNpE2npDdPb07xYUM00D2z6wPxdul5P0HhorjMQwdp+tMNANFaDI7yUa33cXWYAin5eMNRi4A/jqovPwedwEfB6Cfh8/+MWjGAY8fNu1lJcEeOntNdzzyNNcOP8UOntDnDHrBO7/w7PMnTaR9zdsw84F6SzbHvH1OGlKIx9s0nXrPrPwNMbVVFAaDLCjrYv/+r+vcur0Sdz6tc/icbu45Hv3EIkl6Hj+1wDs7OjhseeX8f0vfQbDMEhnstz2/z3OjV+8mLLiovxzZLJZDAycTscej0EIMTIZ5wkhjmc6OBVHhbuxu5p1ECsZ1UGxVByVTuoupuFeSO6lxMeQgZtRXKlzkHIBEaOsFtXXrh/ncA02A9hTkM3h1F+5KY97eCJ2y3Ay9FRI7KxuFOANYG98C6O0GsfsczHcXt0B1O3DCJbtfu657D67r113Go2FUPEI9oYV4MpNszUMPV3S5ck1BdjTse2F6RgeKMpNtTWC5bojarAM1bkDo7xOH/vWv0GgBIordQApEYVwT/4cAfAV52roMeRYDkOHzwLhvvK2YbXhxoIE3EYgAzEhxPFIKUV/JEZ3f4Rnl6+itrKUrt4w4VicSDzJL/7wLADjqsvp7o/kg2G7OpiAm8M0cTjMA95uV6ZpgAJ7lz9xQb+XSC5gtz9dZWsrSnE5HTR39OSXXfmps7Esm22tXZQGAyxd9h4Xnf0RXnxrNZWlRexo76E44GNSQw3vb9zGtAn1hGO6Zt2BqK0opamuktUf7iCey7a77BNnsrW1k0kN1fz+2WUA/Oa2b5K1bJ5Z/j5vvL+ep+/9Pj6vmxmLv8v3v/T33PSlzwCQTGXo6g/TWFNBd38Yy1L5rrCReIKg37fH4xCiUMk4Twgh9p/OnouglIXq79LBs0yS7Jo3MFwejIoG3Zhh698AMEqq9HahrhH2agxm0g0E1QYaQGRSuibb/oQrDDMXmNpLxpzh0DW+bAujYSqGP4jhC2J4A7qjZ38HjqmnDdYUM01d424v7KzO0FPhbkjkgpTphA5YZtJ6SmQ2k69PttdA46EoKtNTMh3OfD01c8pcDKcLZVvYG9+G6gmY9ZMgk8Ze8zoEyzAbZ4CVxd72gX5tHU793R/Ux2xbuSm7x36YyPOF28f6ECTgNhIZiAkhxL6l0hnCsQThWJz/9fD/MO+kE0hnskRiCeKpNDs7enj8hRUATKivoj8SJ5lK7zVQZxgGDtMYcerpvuxPMG1/GYbBaP+pPGPWFFxOJ6FonA827Ri27vN/N58/vfgmc6aOx+fx8OLbq/PrTps5mbfWbD6gyiKmYQwLPE6or2Jbqx781leXMa6qgjfXbKI0GODC+afQ0tlLWXGA+qpyfvt/X+Vn3/kCazY3M7WpjjfXbmbuiRMZX1fF3b/9H378jcswDYOO3hAfP+MkHn76FQC+/OlzAWjp7KXI76WkyM/SN95lwSkzCfg9KKVYt7WFGZPGHfyLuAeWZROOxYdlAgqxNzLOE0KIwy/fbTOT1N0zsxns1g91k4d0Uk97bd+KUVKJ3bxeTyk0TFSuUQBur874OtLcPh0MxADTgVE/CdW2FWL9OD72BczicsimsVs3Y06cheHMdYndS0bVQMMCkjHdoTarp8Na3S0Y3oDu+tnfiTHuxFxmYY9uRrGn6R8j2bUDrGHoc9hXQ4lc8wSswW2N8jodkDQMCPfmO52ajdN0oC6bwd65QS+beBJ2+zaM4gpUMparj4euK2iY+rX0BnLnZB5YgwvTgf4kPbeN2zu4f48vlxmZGb7PYAWez1y//89xmEjAbQQyEBNCiMNHKUU4mmD99hYMw2BzczubdnZQV1lGLJHk3//wLJ19YRafP49YIkkylSEUi/PuOt3m3ONykjrETLh92TVYdTQaGIdVlRXT1RcGwON24TANigM+LFvllzsdJjXlJSjIZ9wNNJpwOkwaqsvZ3tZ9yMfk87hwOp1EYrrz2MknjGfVh9sBOP/02bzwpv70+6QpjUxsqOGddVvwuFxc+vF5/OLx5/jGJR/H43LywJ9e4OavLOLpV99h6vh6pk9s4LHnlnHzVz7Dkv98ilu/9llS6Qztvf2cd9osbvzfj/DEX1fS+9J/ks1apLMWAZ+H//P7pQR8Xr7y9+fS0ROiurx4zKcYiLEn4zwhhDh2qIFgS6wfMFCxflAKlU5id20Hl1d3Z3W4dJAo1KWDPW6vDiJZh3fMOMzQINnQ4JA3gFFShervxKyfgkonUD2tOGacqaf1+oKYjdNRoS5U+xbMGWfqOnilVTqolIigHE6wsqhov24wkc1gbV+NESzHUDa2baO2r8EIlGL4irDTSYxsWtfeO6BuokM+3nW6dcBs4DwcTp0J53Ad2DTbPXE49b6V0kGzAcFyHag1nblmGm5UpA9c7sEAbK6Dar6JRqBEb5NOYlQ04P7U1w7t2EaBBNxGIAMxIYQ4ulmWTTqbxbYVqzfvwOt2kUhlcsGWLM8se5/O3hAnTx1PNJ4kkkjS2RPiv195mymNtfkMLKfDpK27H9CBvEzWwlbqkHpUmQbsWuZtb00aYDDwdaz1xRqaTTg0COr3uslaNulMdrfAZTDgywfj9rav0VJbUUp7Tz8A4+uq2N6ms/uuWfwJfnLt5aP6XOLY8Itf/IJf/OIXWJbFxo0bZZwnhBDHAZ1pl9RZdekkdusmDLcPlYphtW7W00uzWVTrRr1BrpvmUc0wdLbYwHTVXDAOt08HpmIhvdxfrM8PdEdbhwujbhJqx1q9m7pJevCpbF3frqcVo/4EiIdQVlZPr7VtVH8HRkklKhbWHW0dztzz2zqrzsgFzg51JLtrBtxA44RsCkyX/j5wvoYxPFA3sItgBW7JcDu6ScBNCCHEQE27vnAMy7bp6g/T2Rviv19+m5IiP26XzuT6sLmdd9ZtAaCkyI/X40LZilQmSygaB3SG2aFMlT1QBzoLYU8O5ZgPptnGkVJVGmTjU/eN9WGIMSTjPCGEEPuilNJ12tIJHawL9+Rry6lsGtXTigp1YpTX62wv29bTQ0FnYFlZQO/j6Gbkktpy47aBmQBHOly0p+mmpmOwY26+Ft6Qj6j3tI3Hj+fSmw730e6Tc6wPQAghhDiaGYZBWXFRvi7YCU11AHxm4ekHvc9kKkMilSIST7Kzs5dVG7Yxa0ojSilsW7Fy9SYMw+CeR/7Mwo/MJJ5M8fr765kxaRyWZbFhexugg2GGYew1a25/x0hup4P0kH0MZNsBhxQgHI1g29AMtbqKUtpyPw8YX1tJXWUp3aEo21o788d79inTWPHBh5w0pYl31+tpyNVlxXT2hfF53Jw1Z9ohH5sQQgghCpthGOAPYviDekF100HvSw00UlBKZ5Olk7r5grLz0yLsLe/r4FFVU67mW1R3U23dhFFajept1c0drAxk0npaZ662GoYJqIMMkqnhiWljlZc1NHA28MmxbQ3P5Bt8APq4c9sMDbyN0PTiSJIMtxHIJ59CCCGOBbZtk0hl6A1HSabSpDJZslmLrv4w8USKRCrNm2s201RbwfK/fUhbdx9zp0/k4T/rJgfj66oo8ntZs7kZgMnjati8s2Ofz2sYYBrDp4HWlJeQymSoKi3hw+a2/LKO3lBum+FNKKrLS+jMrduT0a6j5zBNTNNgyrhalv3mzlHbrxhbGzdu5NRTTyUcDu/3NjLOE0IIUYiUUjogl07qKaK2hbV+5WAWXiaFSkRRPS2o3jaM0prB6aTZDKTi+34Sp0s/1uPPTffU+yVQBpZuDkFFPfR1QEml/l5Wq5cnInofwXIdJAvn6gd7i3SAcWD/plNPXYVcMBH9XNaeG68N4/Lg+dwPDuyFOwwk4DYCGYgJIYQ4niml6A1FyViWDlQZBq++t47Nze001VURSyTJWjapdIYf/fIxPj5vNlPG1RKKxukNR3lm2fsANNZU0NzRc0DPXVLkxzQNTMOgJzQ4DePC+XMpDepPLbv6Ijy3YhUAF5w1h2fe0M830M0V4OSp4wn4vCxbpbtoVZQUUV9VztTxdTz4o28cyssjjhLxeJzPfe5zrFixgs7Ozv3eTsZ5QgghxJ4pZee6y+YaLmTSkGu+YH34NmZ1kw64ZTMoK4vdthmUwvD4UL36A9d9diM1jFx3VDPffMGoasJwebBbPwTAMf2j+jHKxlq7DHPyKZgNU3Xj1UgfRnkthunA2rkRx/R5OkgXD6GyGRyHkI04WiTgNgIZiAkhhBCjK5PNkkhlSCTTJNNpbFvR3tNPLJGiuz/MNUse5M5vfg7btrGVIplKs2F7K0++9BYAZ558IplMlnRWN2dYtVF3P62vLKO1e/+6bxmGwQVnnszv7rrh8JykOKK++tWv8vWvf53Fixezbdu2/d5OxnlCCCHE4aVUboprtC9fFy67+nXMmvG5mncWyrZRHduxd67HnHQyZNL5OnhGcSXKtnQtvIHMuP1glNbguugbGAOZcWNEAm4jkIGYEEIIcWyJxBPYlh7avL1uM16PO9+Fds3mZkLRODMnN1JTXsIFZ84Z24MV++2BBx7gvvuGN7l47rnnePrpp0mn01x77bVMmDBBAm5CCCFEAVNWNp8Nh5VF5ZpR2K2bMIordGDOymLWT8EIlIzhkWoScBuBDMSEEEIIIY5eCxYswDT1p9crVqzg3HPPZenSpfu1rYzzhBBCCHE4SZdSIYQQQghxTHr11VfzP0+YMGG/g21CCCGEEIfb2E5oFUIIIYQQx51UKsWsWbN4+eWX88uSySRXX301paWl1NXVcc8994zdAQohhBBCHCLJcBNCCCGEEEdMMpnkiiuuYM2aNcOW33jjjbz99tu8+OKLbN++nauuuorx48ezePHi/drvvuq3pVIpUqlU/vdwOHzAxy6EEEIIsb8k4CaEEEIIIY6ItWvXcsUVV7BrCeFYLMaDDz7I0qVLmTt3LnPnzmXNmjXcf//9+x1w25clS5Zw++23j8q+hBBCCCH2RQJuIxgYDMonoEIIIYQ4UMFgEMMwxvowjiqvvPIK5557LnfddReBQCC/fNWqVWQyGc4888z8svnz53PXXXdh23a+McKhuPnmm/nud7+b/z0UCtHU1CTjPCGEEEIcsP0Z50nAbQSRSASAxsbGMT4SIYQQQhxrpPvl7q655po9Lm9ra6OyshK3251fVlNTQzKZpKenh6qqqkN+bo/Hg8fjyf8+EGiTcZ4QQgghDtT+jPMk4DaC+vp6mpubD9sn1OFwmMbGRpqbm2VAXoDk+hY2ub6FTa5vYTtS1zcYDB62fReaeDw+LBgG5H8fWndtNMk4TxwKub6FTa5vYZPrW9iOpnGeBNxGYJom48aNO+zPU1xcLDd6AZPrW9jk+hY2ub6FTa7v0cPr9e4WWBv43e/3H5bnlHGeGA1yfQubXN/CJte3sB0N1/fQC2IIIYQQQghxCBoaGuju7iabzeaXtbe34/P5KC0tHbsDE0IIIYQ4SBJwE0IIIYQQY2rOnDm4XC5WrFiRX/b6669z2mmnjUrDBCGEEEKII02mlI4hj8fDrbfeulvNElEY5PoWNrm+hU2ub2GT63v08fv9XHXVVXzjG9/goYceoqWlhZ/97Gc89NBDY31oB03eZ4VNrm9hk+tb2OT6Fraj6foaSik11gchhBBCCCGOL4Zh8NJLL7Fw4UJAN0645ppreOKJJygpKeHGG2/k29/+9pgeoxBCCCHEwZKAmxBCCCGEEEIIIYQQo0iKYgghhBBCCCGEEEIIMYok4CaEEEIIIYQQQgghxCiSgNsYSqVSzJo1i5dffnmsD0Xsw5NPPolhGMO+Fi9eDMB7773HGWecgd/v57TTTuOdd94Ztu2jjz7K5MmT8fv9LFq0iO7u7vw6pRTf//73qaqqory8nO9973vYtn1Ez+14tqd7cOvWrZx//vkEAgFmzJjBc889N2ybF154gVmzZuH3+/nYxz7Gli1bhq2/9957aWhoIBgMcvXVVxOPx/PrkskkV199NaWlpdTV1XHPPfcc1vM73u3p+t5www273cv3339/fv2h3K89PT1ccsklBINBJk6cyCOPPHJEzvN409LSwuLFiykvL6ehoYHvfve7JJNJQO5fcXSRcd6xQ8Z5hUnGeYVNxnmFqeDGeUqMiUQioRYtWqQA9dJLL4314Yh9uPPOO9XFF1+s2tra8l99fX0qGo2q2tpa9U//9E9q7dq16vrrr1c1NTUqGo0qpZRauXKl8vl86uGHH1arVq1S55xzjrrwwgvz+/3Zz36mGhsb1WuvvaZefPFFVV9fr+6+++6xOs3jyp7uQdu21ezZs9XnP/95tXbtWvWTn/xE+f1+tX37dqWUUtu3b1eBQED97Gc/U6tXr1aXXnqpOumkk5Rt20oppf74xz+qkpIS9ec//1m9+eabasaMGepb3/pW/jmvvfZaNXv2bPXOO++oP/3pTyoYDKrHH3/8iJ/78WBv/8aef/75asmSJcPu5VgsppQ69Pv1oosuUuedd5764IMP1IMPPqg8Ho9auXLlETvn44Ft22revHnq7/7u79Tq1avVq6++qqZMmaL++Z//We5fcVSRcd6xRcZ5hUfGeYVNxnmFqRDHeRJwGwNr1qxRJ598spo9e7YMxI4Rn//859XNN9+82/L/+I//UBMnTszfyLZtqylTpqiHHnpIKaXUF77wBXXVVVflH79jxw5lGIbasmWLUkqpxsbG/GOVUuq3v/2tGj9+/OE6DZGzt3vwr3/9qwoEAvmBtFJKnXfeeerWW29VSin1ox/9SJ1zzjn5dbFYTAWDwfz2Z599dv6xSin12muvKZ/Pp2KxmIpGo8rr9Q673++4445h+xOjY6R/YxsaGtSzzz67x+0O5X7dtGmTAtTWrVvz66+++uph+xOHbt26dQpQ7e3t+WW/+93vVH19vdy/4qgh47xjj4zzCouM8wqbjPMKVyGO82RK6Rh45ZVXOPfcc1m+fPlYH4rYT2vXrmXq1Km7LV+xYgXz58/HMAwADMPgrLPOyl/bFStWsGDBgvzjGxsbaWpqYsWKFbS2ttLc3Dxs/fz589m+fTttbW2H+YyOb3u7B1esWMHcuXMJBAL5ZfPnz9/r9fT7/cydO5fly5djWRZvvfXWsPXz5s0jnU6zatUqVq1aRSaT4cwzzxy275UrV8r0klG2t+sbDodpaWnZ470Mh3a/rly5ksbGRiZMmDBsvfw7P7pqa2t55plnqKmpGbY8FArJ/SuOGjLOO/bIOK+wyDivsMk4r3AV4jjPeUhbi4NyzTXXjPUhiAOglGLDhg08++yz/OQnP8GyLD772c/y4x//mLa2NmbOnDns8TU1NaxevRqAtrY26uvrd1u/c+fO/GBr6PqBf1x27txJXV3d4Tyt49re7sGRrte+1vf395NMJoetdzqdVFRUsHPnTkzTpLKyErfbPWzbZDJJT08PVVVVo3V6x729Xd9169ZhGAZ33XUXS5cupaKigu9+97tcddVVwKHdr/t674jRUVpayic/+cn877Ztc//993PeeefJ/SuOGjLOO7bIOK/wyDivsMk4r3AV4jhPAm5C7MOOHTuIx+N4PB7+8Ic/sHXrVq6//noSiUR++VAej4dUKgUw4vqBIo1D1w/8PLC9OLJG+3oOXa+U2uM6kOt9pKxfvx7DMJg2bRrXXXcdr7zyCl/72tcoLi5m0aJFh3S/7uu9Iw6P733ve7z77ru89dZb/PznP5f7VwhxwGScd/yQcV5hk3Fe4SmEcZ4E3ITYh/Hjx9PT00NZWRmGYTBnzhxs2+bKK69k4cKFu92EqVQKv98PgNfr3et6r9eb/33oz0B+e3Fkeb1eenp6hi3bn+tZWlq62zXcdXvLsva4DuR6Hylf/OIXufjiiykvLwdg9uzZbNy4kV/+8pcsWrTokO7XkbYVh8dNN93Evffey2OPPcasWbPk/hVCHBQZ5x0/5O9EYZNxXmEplHGe1HATYj+Ul5fn63cATJ8+nWQySW1tLe3t7cMe297enp8m0NDQsNf1DQ0N+d+HrgNkmsEYGel67Wt9RUUFXq932PpsNktPT0/+end3d5PNZodt6/P5KC0tPXwnJfIMw8gPwgZMnz6dlpYW4NDu1329d8Touu6667jnnnt45JFHuOSSSwC5f4UQB0/GeccH+TtR2GScVzgKaZwnATch9uHZZ5+loqIin4oK8P7771NRUcHZZ5/NsmXLUEoBug7IG2+8wbx58wBdjPH111/Pb9fc3ExzczPz5s2jvr6epqamYetff/11mpqa5B/vMTJv3jzeffddEolEftnrr7++1+sZj8d57733mDdvHqZpctpppw1bv3z5clwuFyeffDJz5szB5XKxYsWKYfs+7bTTME35p/hI+Jd/+RfOP//8Ycvef/99pk2bBhza/Tpv3jy2b98+rJbH0PeOGD233347v/rVr/j973/P5z73ufxyuX+FEAdDxnnHD/k7UdhknFcYCm6cd8h9TsUhQdrFH/XC4bBqaGhQl19+uVq/fr36y1/+ourr69VPf/pTFQqFVFVVlbr++uvVmjVr1PXXX69qa2vz7YqXLVum3G63evDBB9WqVavUwoUL1cUXX5zf95IlS1R9fb166aWX1EsvvaTq6+vVPffcM1anelwaeg9ms1k1Y8YMddlll6nVq1erJUuWqKKiIrV9+3allFJbt25VXq9XLVmyRK1evVpdeumlavbs2cq2baWUUo8++qgqLi5WTz75pHrzzTfVzJkz1XXXXZd/rq9//etq5syZ6s0331RPPvmkKi4uVk888cQRP+fjydDr++abbyqn06nuvvtutWnTJvXv//7vyuPxqGXLlimlDv1+/eQnP6kWLlyoVq1apR588EHl9XrVypUrj+j5Frq1a9cqh8OhfvjDH6q2trZhX3L/iqORjPOOfjLOK2wyzitsMs4rLIU4zpOA2xiTgdixYfXq1er8889XRUVFqq6uTt122235m3flypXqlFNOUV6vV51++unq3XffHbbtQw89pBobG1UgEFCLFi1S3d3d+XXZbFZ95zvfUaWlpaqyslLddNNN+f2KI2PXe/DDDz9UCxYsUB6PR82cOVM9//zzwx7/l7/8RU2dOlX5fD513nnnqS1btgxbv2TJElVdXa1KSkrUV77yFZVIJPLrYrGY+uIXv6gCgYCqr69XP//5zw/nqQm1+/V96qmn1OzZs5XX61XTpk3b7Q/podyvHR0d6uKLL1Zer1dNnDhR/e53vzvs53e8WbJkiQL2+KWU3L/i6CPjvGODjPMKl4zzCpuM8wpLIY7zDKVyOdJCCCGEEEIIIYQQQohDJhPKhRBCCCGEEEIIIYQYRRJwE0IIIYQQQgghhBBiFEnATQghhBBCCCGEEEKIUSQBNyGEEEIIIYQQQgghRpEE3IQQQgghhBBCCCGEGEUScBNCCCGEEEIIIYQQYhRJwE0IIYQQQgghhBBCiFEkATchRMG57bbbMAxjn1/f/va3x/pQ+c1vfoNhGDz44INjfShCCCGEEEc9GecJIY4VzrE+ACGEOFy+9rWvcfbZZ+91/fTp04/g0QghhBBCiNEi4zwhxNFOAm5CiIL10Y9+lCuvvHKsD0MIIYQQQowyGecJIY52MqVUCCGEEEIIIYQQQohRJAE3IcRxb+HChUyZMoV3332XBQsW4Pf7qa+v5+tf/zpdXV27PX7p0qWcd955FBcX4/P5mDNnDvfddx+2be/22D/+8Y+cc845lJSUUF5ezrnnnssLL7yw2+MSiQQ33XQTTU1NeDwepk2bxn333YdS6rCcsxBCCCHE8UDGeUKIsSIBNyFEwYpGo3R3d+/1a+ggp6enh4997GMUFxdz9913c+GFF/LrX/+aM888k3A4nH/cPffcw6c+9Sl27NjBTTfdxJIlSygpKeGGG27gkksuGTYYu+OOO/jsZz9Lf38/t9xyC7fddhsdHR1ccMEFPP3008OO9eabb2bp0qV85zvf4ac//Sm2bXPDDTfwq1/96vC/UEIIIYQQxxgZ5wkhjnpKCCEKzK233qqAfX719fUppZQ655xzFKC++tWvDtvPvffeqwD1wx/+UCml1ObNm5XT6VTTp09X0Wg0/zjbttXll1+uAPWb3/xGKaXUli1blMPhUAsWLFCpVCr/2J6eHlVRUaHmzJmjlFLqoYceUoCaPn26SiaT+cdt3rxZAWr+/PmH5TUSQgghhDgWyThPCHGskKYJQoiCdeONN/KJT3xir+uLioqG/f7jH/942O/f/OY3ufXWW3niiSe44447ePLJJ8lms9x0000EAoH84wzD4Cc/+QmPPvoojz32GFdddRVPPfUUlmVx/fXX43a7848tLy/n1Vdf3e25L7vsMjweT/73SZMmUV1dTWtr60GduxBCCCFEIZNxnhDiaCcBNyFEwZoxYwbnn3/+fj22srKS2traYctcLheTJk1i7dq1AGzevBmAmTNn7rb9hAkTKCoqYuvWrQBs2bIFgGnTpu3xuHa163MD+Hw+UqnUfh2/EEIIIcTxRMZ5QoijndRwE0IIGPap41DZbBanU382ofZR2NayrPx+MpkMoD8V3R8Oh2N/D1UIIYQQQhwAGecJIcaCBNyEEAJoa2sjGo0OW5ZKpdi6dSsnnngiAJMnTwZgzZo1u22/ZcsWEokETU1NAEycOBGADRs27PbY++67j3/8x38kEomM6jkIIYQQQojdyThPCDEWJOAmhBCAbdvcfffdw5bde++9RKNRrrjiCgAWLVqEw+Hgpz/9KbFYLP84pRQ/+tGPAFi8eDEAn/nMZzAMg/vvv59sNpt/bF9fH//6r//KsmXLCAaDh/u0hBBCCCGOezLOE0KMBanhJoQoWMuXL89PE9gTt9vNpZdemv/97rvvZsuWLZx55pmsXLmShx9+mDPOOIPrrrsO0J983nnnndx8883MmTOHL3/5ywQCAZ566ilefvllLrroIq688koATjzxRG655RbuvPNOPvrRj3LFFVeglOLXv/413d3d/O53vzu8Jy+EEEIIUcBknCeEONpJwE0IUbAeeOABHnjggb2uLykpGTYQe+WVV7juuut4/PHHqaur4wc/+AE//OEPh3Wf+v73v8/06dP5t3/7N5YsWYJSimnTpnH//fdzzTXXYJqDicN33HEH06ZN47777uOWW24hEAhw6qmn8vDDD3P66acfnpMWQgghhDgOyDhPCHG0M9S+qkMKIUSBW7hwIa+88gqZTGbET0qFEEIIIcSxRcZ5QoixIjXchBBCCCGEEEIIIYQYRRJwE0IIIYQQQgghhBBiFEnATQghhBBCCCGEEEKIUSQ13IQQQgghhBBCCCGEGEWS4SaEEEIIIYQQQgghxCiSgJsQQgghhBBCCCGEEKNIAm5CCCGEEEIIIYQQQowiCbgJIYQQQgghhBBCCDGKJOAmhBBCCCGEEEIIIcQokoCbEEIIIYQQQgghhBCjSAJuQgghhBBCCCGEEEKMIgm4CSGEEEIIIYQQQggxiiTgJoQQQgghhBBCCCHEKPp/pBjlAtqN/ysAAAAASUVORK5CYII=",
      "text/plain": [
       "<Figure size 1500x200 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "fig = plt.figure(figsize=(15,2))\n",
    "spec = fig.add_gridspec(ncols=2, nrows=1, width_ratios=[1,1], height_ratios=[1])\n",
    "\n",
    "lw = 1\n",
    "fontsize = 14\n",
    "\n",
    "color_map_colors = cmc.batlow\n",
    "cluster_colors = color_map_colors(np.linspace(0, 1, 10))\n",
    "\n",
    "figure = fig.add_subplot(spec[0,0])\n",
    "for i, random_seed in enumerate(random_seeds_list):\n",
    "    plt.plot(n_epochs_list, MSE_losses_without[:,i], lw=lw, c=cluster_colors[1])\n",
    "plt.gca().spines[\"top\"].set_visible(False)\n",
    "plt.gca().spines[\"right\"].set_visible(False)\n",
    "plt.xticks([1,5000,10000,15000,20000], **csfont)\n",
    "plt.yscale('log')\n",
    "plt.yticks(**csfont)\n",
    "plt.ylabel('MSE loss', fontsize=fontsize, **csfont)\n",
    "plt.xlabel('Epoch', fontsize=fontsize, **csfont)\n",
    "\n",
    "figure = fig.add_subplot(spec[0,1])\n",
    "for i, random_seed in enumerate(random_seeds_list):\n",
    "    plt.plot(n_epochs_list, MSE_losses_with[:,i], lw=lw, c=cluster_colors[7])\n",
    "plt.gca().spines[\"top\"].set_visible(False)\n",
    "plt.gca().spines[\"right\"].set_visible(False)\n",
    "plt.xticks([1,5000,10000,15000,20000], **csfont)\n",
    "plt.yscale('log')\n",
    "plt.yticks(**csfont)\n",
    "plt.ylabel('MSE loss', fontsize=fontsize, **csfont)\n",
    "plt.xlabel('Epoch', fontsize=fontsize, **csfont)\n",
    "\n",
    "plt.savefig(plot_save_location + data_tag + '-MSE-losses-' + str(n_components) + 'D-LDM.svg', dpi=200, bbox_inches='tight')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "***"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3 (ipykernel)",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.10.12"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 4
}