MetalArtifact_Detection_Pytorch.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 12,
   "id": "a6be3f19",
   "metadata": {},
   "outputs": [],
   "source": [
    "import numpy as np \n",
    "import pandas as pd\n",
    "import matplotlib.pyplot as plt\n",
    "\n",
    "from sklearn.model_selection import train_test_split\n",
    "\n",
    "import torch\n",
    "import torch.nn as nn\n",
    "from torch.utils.data import Dataset, DataLoader\n",
    "from torchvision import transforms as T\n",
    "import torchvision\n",
    "import torch.nn.functional as F\n",
    "from torch.autograd import Variable\n",
    "\n",
    "from PIL import Image\n",
    "import cv2\n",
    "import albumentations as A\n",
    "import time\n",
    "import os\n",
    "from tqdm.notebook import tqdm\n",
    "import os\n",
    "\n",
    "!pip install -q segmentation-models-pytorch\n",
    "!pip install -q torchsummary\n",
    "\n",
    "from torchsummary import summary\n",
    "import segmentation_models_pytorch as smp\n",
    "\n",
    "device = torch.device(\"cuda\" if torch.cuda.is_available() else \"cpu\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "id": "6fdc020f",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "\"\\npath = 'C:/Users/hrips/OneDrive/Documents/Task/Metal/images'\\nfiles = os.listdir(path)\\nfor index, file in enumerate(files):\\n    os.rename(os.path.join(path, file), os.path.join(path, ''.join([str(index), '.png'])))\\n\""
      ]
     },
     "execution_count": 13,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "\"\"\"\n",
    "path = 'C:/Users/hrips/OneDrive/Documents/Task/Metal/images'\n",
    "files = os.listdir(path)\n",
    "for index, file in enumerate(files):\n",
    "    os.rename(os.path.join(path, file), os.path.join(path, ''.join([str(index), '.png'])))\n",
    "\"\"\""
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "id": "a151aa55",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "\"\\npath = 'C:/Users/hrips/OneDrive/Documents/Task/Metal/masks'\\nfiles = os.listdir(path)\\nfor index, file in enumerate(files):\\n    os.rename(os.path.join(path, file), os.path.join(path, ''.join([str(index), '.png'])))    \\n\""
      ]
     },
     "execution_count": 14,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "\"\"\"\n",
    "path = 'C:/Users/hrips/OneDrive/Documents/Task/Metal/masks'\n",
    "files = os.listdir(path)\n",
    "for index, file in enumerate(files):\n",
    "    os.rename(os.path.join(path, file), os.path.join(path, ''.join([str(index), '.png'])))    \n",
    "\"\"\"    "
   ]
  },
  {
   "cell_type": "markdown",
   "id": "16ceb13c",
   "metadata": {},
   "source": [
    "### Preprocessing"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "id": "c2e1d0cb",
   "metadata": {},
   "outputs": [],
   "source": [
    "IMAGE_PATH = os.path.join(\"C:/Users/hrips/OneDrive/Documents/Task/Metal/images/\")  #4083\n",
    "MASK_PATH = os.path.join(\"C:/Users/hrips/OneDrive/Documents/Task/Metal/masks/\")    #4083"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "id": "578b1158",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "width:   512\n",
      "height:  512\n",
      "channel: 3\n"
     ]
    }
   ],
   "source": [
    "# Check the shape of image\n",
    "\n",
    "img = cv2.imread('C:/Users/hrips/OneDrive/Documents/Task/Metal/images/200.png')\n",
    "h, w, c = img.shape\n",
    "print('width:  ', w)\n",
    "print('height: ', h)\n",
    "print('channel:', c)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "id": "9ca7f157",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.image.AxesImage at 0x1a2552d1b08>"
      ]
     },
     "execution_count": 24,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAQYAAAD8CAYAAACVSwr3AAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADh0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uMy4xLjEsIGh0dHA6Ly9tYXRwbG90bGliLm9yZy8QZhcZAAAgAElEQVR4nOy9XaxtS3Ye9I2ac6619t/5vefcvn+OTbuJgpCAyLe7I96wLNxtW22QopAgYiFL/YAjgZQXizcQD+GBBIWHICMiHAlBkAk4Eh1QsJAQInZsRSFgTNIdy3bfvrfvz/ndf2utOWcNHkaNqlE1a669z7nndO9z7y5p77XWnDXrb9YY9Y1RY4wiZsZ1uk7X6TrZ5H7YDbhO1+k6Xb10zRiu03W6TpN0zRiu03W6TpN0zRiu03W6TpN0zRiu03W6TpN0zRiu03W6TpP0UhgDEf00Ef1jIvoOEf3yy6jjOl2n6/TyEr1oOwYiagD8EwA/BeA9AL8N4M8y8//7Qiu6TtfpOr209DIQw5cBfIeZf5+ZtwD+WwDfeAn1XKfrdJ1eUmpfQplvAfiu+f0egK/semBBS17h4CU05TpdNo13DkCXAY9lHrrEIwQ0D0+fp1nX6QWmYzz6hJnvXSbvy2AMtakymXJE9E0A3wSAFfbxFfrJl9CU61Sm4z/zVZBnkAdAAHmAvLweJkr3NLHkAwPcyCW9r7+zt0upHM3b7zs0PYNGjve5AXxDOPqbv/nyOnudsvS/8q/94WXzvgzG8B6Ad8zvtwG8X2Zi5l8B8CsAcIPuXDtsvIR0/Ge+CjcmQmcC2rXPCR+IhE3EAAPEhlEwgyGfGCkxAwA0hk+fXh8TgcCRmeh913NWjwehHTzWP/fl9KwDuCHs/63fekEjcJ2eN70MxvDbAL5ERD8G4HsA/g0Af+4l1HOdTDr7178iq7oD3ChE6AZhCsQMJoIbGW7YwYMZCSEgMAiGEDqEwKviRvEME2XX3FAwj3AjMgvFmCxI4+xf+0qsm53c3P8frpnFDzK9cMbAzAMR/QUA/wuABsBfZ+bffdH1XKfADBAIVgmPSQiTEoMAhLgjU8iI0RRIyFb6mHjmu81rmYNnIWgGuE0IZMJ0ijoInIsho9wgD5z//JejCHLwa9dM4mWnl4EYwMzfAvCtl1H25z2tf+7LUd5XAsyJiadET4HISgK3+Vyef6oVqjyrn6VoEhgM+SCG2LqVAYASQqmkiHRCI2kE4BjcEJ7+ua+iO/MAA3u//vd3NPQ6PW96KYzhOr3YpMyAOCEDGuQekzAJhdzkA/z3uXIwIQrkRG3TRcwAyDe4VdHokk5Cy6bxArVRjUn5vE1MgamUjIqTfmP9s1+WdrhrJvEi0zVjuKKpxgxiMj+JRe5nEtgedxACAiBviLQmIlhRAMhRROV7JP4ib6aoLNtr6yuZkv4O1yxKqPVXGWB5n7yMGTspZ+9/vGYSnyZdM4YrltY/G7T0ujuwSx9gVljyDNqKVj/qFfR5u9qXsF9TRVcQkwsMwVavZdu2IigqDQOgUZgUeUwZU5kKhhMVmbaZBLhSLKr04fwbX77e4fgU6ZoxXIG0+fq7YEcJHdSg/oxCkLwQSwbh7coM5KuyPlvC8zJ/wRxURIm/A0rJMxV1AKCB4VcOvgXcNiEXGhCZWyyrEBcUhaRLAa14zq+TYYZBxOBGynjyb34V45KwvUH4wn/6f+I6XS5de1f+kNPmZ94FUGjuS3hfEHK0QyDDFFA886xJ61ADJF2pGYIyIgop6iPELcyJfYRmUSVpmx6kkeHbgrOoODEis6WwdcU22PJ3+fsQcPY64exNxj/9T746n+86ZekaMfwQ0ubr705ka0uYk2v2nqaaiFHbPizrKJ+38nsTLB8risMMIRTf4/NFecQAmOG2ov/gJtQR7BrIA35BCUkYkSUyBBKkkCMEW2e6p6KHNcpikrXPdwx2wLf/s6/EHZgv/YVrMWMuXTOGH2BSdLCTWO33CjQHCgg/x0hslgrs12vx3pwe4xJ6AZ6pgwmguN0YGEFHwiSYoln2Tt1HrcpC+Ro/DbPU3QxiRndC2NwhoGU05w7USxl/+B/+KQwHjC/+xWuz7DJdixI/oLT+2S8nhmDFBWBK/HOye00HUGEcVFyzBJuJAoaY2VHa6qwwA2UeE73CLhRf3vOiW1CiVVHCt5TPxBI9Idcj1NCKVGgUlpRQyeKY0Wwk/9HvA7f+CXDzO8DB94D2lPCdv3wtYpTpGjG8xLT5+ruySwAkwyNLmLpiVghhdrU2sn4kkAJZKBIADCqolBF/NpQpFyfEP1N3VQFpswfiRCBqYgBBTKFRRAvfim/FuHCRkDkwCVUklm1Qq8pyPFWMUFGCidBspMzVA4ezDlg98mjPpQ2+E7hy/McIf/Af/SkcvAfc+8//3nyHPkfpGjG8pKR6hGxPfxcsn7t3GTg/s8JHNKCMgvL7QGAK5bboZdq7K2WiTvphGRYgzLLZ+kDsQvC+peDxGfIY3QKNQZHZmDp2MCZA/EV8CywfMVwP9AcuIgs3MJZPGMvHhHHF2NwmnPzpr2SOXZ/XdM0YXkLafD3XJdAoprwTVIAd3821XKYOl0vCmNHYx2JKJmGTle8rhFbmt2WV1+bqV8aUiQHhz/XJDZwrlpU2uWAXwRnDoUwJqfdk+1cYTnvOaM8I6zsUtzLJM5oNY/lQnutvMPp9B99Ssif5nKZrUeIFpfXPfVnQwdyW3cBJrCCzQpeiQ/iM9gl6O8B2NRaqytkXoYuKTD65X9FvZFVcIG7U2hyvc37NijjN1otisiGMCxLPUCtK6PPjVH4ptyuz3yyogRtg9Qnj+EeBzccNVo9GEAtiWT0Clg8arO95nN9zcAOhOxO9kJa1/J9+u97hz2i6RgwvIG2+JgiBiXJnpCKJZj54HurI11AE6qux/JhpRFh9a7sP9jsVjKTq5TjJk7dlsitiPzU50R/4BcEvxYaBWxEVuBUGENFB+HQ9x7K4oVxkMG1wasRktjat3UVUVIZn3ACMHaE7Y3ALbG9S1t5mw1g9ZDRrwvY2Y3tE8LpkhnIiCvycpGvE8ClTJjbUPiuig1o3qrXjxOagSDUoP1l5Z1bq8rn8d+7XEFFJTak5aVTRx9AfWfWFKQwrwrCS624IsRyCpSONQHfOQYwILuEsruKqa3AwOhrTJjcwfEeinwDyMVRmF8cmBInZJ7gBaNbA5iYwLgjtOj23fMzY3nA4e8Njc9uhO3No12NW9+br72L5rc8HcrhmDM+ZyhWEmJODT7FLMEmqewAnxID0KV6F6UFlBCU038UAJkRetEvFmol9RA29KAoyZfmFEO+wJAz7hLEj+A4Y9yS/DwZNSV6gbDcDAGggNFvAbYDlE0Z3xmg2PjJPUUQmkSI6ho1C2M0WGUPQ75Hhhfa6ntFsxcBp9Qnh9E3G+o7D4fcSY+lOPQ4+ANZ3CZs7DNcTlk+R+3hQeu+fdQZxzRieI21+5t20utuVPmjWq96FhYwdZfX4z5TVYGL0Y5mD/q4pAXduIZp67Co7q7OwKYgG49LBN8D2BmFcEsYVMC6CQjAyAlX+hcaZfpBhhH7BYia9Dwz7hPaMsHhKWD32po8EuBSeTrczVQGZiUIGJZRbmc2G0R84dCdS1uYW4fB9ZEy6O2Msnjis73n0NyBbqINHdAwz7/uzjh6uGcMzpvNvfDkPj2bnpY15oNcYmaa9JObS1BgICjaHyBxKGX9XNOdy1Z/YM9j7hSJxwlRI9vpVLOgP5NMvgWGFwAjloci0DCMgJbpAzNZGAQCIkp7BL4HtQhgEsUOzYdmBaAzC8NO2NRsh2owhGwSmzMMN0rR2zWhPHbY3BOFZl3TXM1YPGMMhYdhnDCsSw6jRMCAz9p9l5nDNGC6ZNj/zboCoqOsCdmjp53YYypQxAP3uAPbTPFWCjw9P2xA/lQkUXpl6n1vREwxLQn9I2Nwk+EUuGpDXJR9wYyGSFCgkI2rOiTu6UCMhjnGPcfwjhMUxYfWA0a6lbqemzyGyE42UxAnkeoVMpAjPuYHRrkWc2P+I8eSLYtOweJq2Psgz9h54+K7ByTvA6ZsENzgsjv1UCRrq2Hxd0OPy73y2GMT1rsQlkjIFAHUxATDERRMmMTFJriRFFrV9fuxAHKmAep0ZwgiiTokmVEYf9h3Wtxocv9Xg5G2Hsy8Q+iNgXHEkdhoEEYgHJAQdGJg/myKxzoxF0B3QSOCWsb0JsTkgEWHEOtPkZcB3O+qqpGYreov2nNGuCZsblAd+YcmzfOyxeELY3mJsblCwkNxRD6Wdqc9KumYMFyQ1dKmtSLOpJLzyttUxVFb8TJ/AKmdPyywRgE21fHEXBEZHQbJyru80OHmrwelbhM0dYDiQlVoiQAVrxDH8WdRE03qqhFkqQu0zoa+xDi8MYXsTGJfykG/y53QXI25V6jjFgs0np2fYyfZldyzbllxiZhZxY/WA4TvG+jVCv08oDa20T7b8zxJzuBYlZlJ0jTbpQvNmD1EcapY5QtDyCoZgZXGbB+pn4UK15YEwttwSURhmFuVpAsaVw7AinN13GPZExs/6EaqcoAAKbVS0UOgk7BkW9pMsGtd8Fs0YEQQDgEbqOXmLcPA+0GwFIWgfyDPaDTCuCO15sW2puxeVYC7RRPoh48k/C2yOgrGTxtJkRrMF9h54nD9qsL3tcbp2WJx4YCBkEap0fEvkQK/+rsU1YphLVsMNpNXJTUUFyVf/bq+V+oHyvq6ek23GUH9kCKqYVGZhjKVi+wyasGKJ7wjr2w1OviDiwvaWMAV2HMUC9VUovTQjEjKBYuxqH5uqbUN+L8ZrUARTMJX4NXhhAsC4kh0EQLYofUuxXjeIgnAS8KUoT3dhiEXPMHaEdsMBlVA+ThCm40ZBDTQQNncZ/Z7bjRIr4/Qqp2vGUEkxbgKQQeZ4lFuFObBhJLWdgzlLxqq9wUw+QKC2MggNHa8MImuTMhLlaw2hP2pwfqfB6ZsOm7tAr8eFekSFopXjywke4b4RJcjmY8OkZpjfznzmPrFAfgDobwDDnogX3EytHKNNBipjYPIBokPQ9jcbwvYoeFkWogcNwOIpY/GY4DuW7dnFDpGiSK+6peS1KGFStE8A8glrYCN5jpORBp6IA/GINiSir1kp2usX7VJUEcYYbhQyfXY8HMmkH/Yczu86bG8AfoG00gciZ/M72+q0Iehj/+xYpO/RCKpos+odJuHY5hy3LHMYACKJvnT2OuHGH3BCI6Gd7YbR7xPc1liR2jIr4mB35jGsHFYPgLM3GetbDgdrXzAmxuLE4/B7BG4dTt5h0OCw/2AEbSvtxfQ9vsrGUNeIYS6VcBwwqyTnXoCVVEMIljmUhk56vfZdn51YDhYQPttlcBCGcKfByZsO67tBj0CICsR49iQjd1gKTKW6BVoRceb8M6ooqRQjChhfJi1r3GOMy7BLocyKxPaASXwy0kNAyeCtvkGtILsTYSTbG1R91zQC3anH8qHUv7lLSaQox2FuvF7RdI0YkHvRZRZzcy/XEAmjRAT5Q5fxW5gzWiqZgP2sPRcZkBOUcHZvihImIoIl+CB+WK9HmyezySgZRaUdE+Mu+8Wgd9pRlhAogRvG5iZh/2PZcnQhNoPGVej3gjESirKKfqjfhhsZGIHmXMYo8xvR9nsxjFo+9jgZxFR68ZiwOCnyVsQuW6fuVrxKtg6fe8Sw+Zl3U8h1TZXVIPseVqrJDkAxmpZAymv2erl1KT/MQTIzfxnTCfL35naDpz/S4vEXHdavpd0GPSYuI1ZL4D7dz5K1YrTtNm2z4zLJW+gjMg/PcvyAKYGpyDPKNur5XVmx7falGC4Zm4QaAwz1aei37kxurh4Aw4GgESuiaHIDY3Hqsf8+wbeMk3dke7eq5C1TMZ6v0nbm554xAKiKC1liTF5+eSLSnBgwp1gsEwf9RXzeKLkmBGWe0fb7BWFz0+HkTYfN7YASiv5QQfw1I6lae6MIUxNb9Lupp6ZXmdRpmAVgdBRlk9iIPgz0R8CwVLftpAxstoxxQVMmXqYgCrqe4RtCeyZOWsMeTVGitnEAVo/Ej2LcE52GbwtF5Mxicpl3fxXT51aUiCbOZTyCHdy/3LMu4bN6Re4SDcrfGWoowqDp9TnFpIZDW99uxBDnAMnGQMstg52UiKUij8d8hU2CMo6aWFOil+jjEcqXQ3hTPbEZjDw2RYncLKJpxApzfY/QfI+BhcZ0FNuD7ZELuw6c96lkFpyuN1sh+LP7hOUTnysw9RHPWBx7HHzPwXeEk3fEVHr5NCgiK8gkorPi3b0qCsnPLWKwVoA1CFmVGRk5cRWweO7sA02lyDBH9Jp3YgBVrLDjUiwWT98i9IeIDkfWqCj2z2xnxjJ1F4JSHg6GReI3AXMeBOKOQIkqSiOtuS3IWYXjHIKx46HKUpYYD7oTYm0mwMCwsg0znxVRUaNVd8dAfyh2EtmzZk64nrE4YSwfEfpDxvoOYVy4ydjO9qdASlc9fS4Zg25LZqctaWJceMbB9BzHdH2X/cI0vyHkYP8A5vg9M43OCgD6wwZn9xxO3ySMS2m33WXQfPqGJ2jFELzv5LdvEY2nIoNoODGG2l/JJOfkbu1LyTgIExFnFrXp9mrLAv2L1PSMYeXyuBgzOgAxdvLwjegbxv2iTGUKlPI3Wx/jQ27uMLaHyRx7NlmGYIq/6vqGz5Uosf3pd+sWhfYTmJ+YetsctppD61yMqH0vZW+tTxAMojgxKSPMv3HlsLkhDCGGPitFAu2SDbDihOdIXboMYgJJRJcQgqo4jn4LqZ2pYaztLHQPNRuFeJw9Kp+xwelaDCFf6jJGefD8NcD1iLoHiTgtIdz6fYfFyZjKtFWYIC5Nz+gRFMlMOL/nsHpUeS6s8q5nLJ967H3Y4PwNj5O3HZZPZDfEBoapiS/kp/zjKu9WfK4QQ7IneAbdQrUcPb+ggPuGmK3tQmnHULYnNz9mqM91+Ry3hLN7DmdvUPQsVAMiVmgdVvJoKq3XKKz+lDeKInWbdrkgZ3NgCmTaUzzLikq0LtsWl77PDyZMO1KfqgzGMPFxj9EfUIgWRdFrtNlyLk4UqMGauNOgjlUSJKY/xBQBFESuMRuaM0J/w2N76MQRS9tn2mh3ZGbFpGeYdz/I9LlhDBG61XpcwMbseu07MHnZOjFruxM1caAqIgCCGIodCUD0CWf3GpzfI4wLU3bpPwEhbF3Z02rP008lfKbpxLXPTayqOKILrSP+qYiijML8rjHQNIDpLzKcAnFpPiE4CvYHyHQNGkRnsnU5k5pe2tedyLZlJHKbDCJTi8jVQ8m0viu6hmz7Uh/bQfRpIbia5tMXMgYi+utE9BER/T/m2h0i+rtE9O3weTtcJyL6q0T0HSL6R0T0J19m4y+bIlMwq0Zm0FKKE8ooKp6FADIz4TiBfSLEqg4B02s1JDFBCQ1w/lqLx19scPI2ZQpGbgLhGYYwKcg2JvwlEabgTqZRESlUGkme8vKJc8cubZsyBAdwl7c1Nm1GrCgZQvTGjK7ZCJGkQrzJNjHm7sxj2HMoY2PYg3JVWdxs5CUvnkiF2wNXb5MyGC92E/vf99j7yOHsDcbZfZfsIHYsBqlzxethEXOvUroMYvivAPx0ce2XAfwGM38JwG+E3wDwNQBfCn/fBPDXXkwznz9NlDzh5WbxFcqkE7Y2MUyqoQCgzhyy/PEg1pnqzcTs9x1O3yIM+4iEoffsBIswf66BphGqR7D3MpGiZohgGmdRiK2XLdzP+sFpZS+YQ+xLyZSByXhnjwRVQH8jnGpdMJx+3zAFs9rHcg3icoMcPNOcSdSq+Ur1GUZ37rF8wIBjrO8GhlKZN9V3XIhN+nmVFJIXMgZm/t8BPCwufwPAr4bvvwrg5831v8GSfhPALSJ640U19lOlktCBnRPvmWS/YiLM7u+zEmCKmFwuyPF5kp2Hk7ca2XUI7U16BK48w1MGYVf6kmh3Yd1d2ymVS6WYMFeWtt+3SQyoKk1jBfYGDHHK7+1R2KK07yAwz+gNeYGYqOdbdqdAf2hc68u+GIal25fNGaE/8tjcDv4aFYKf60vNHuSqpOfVMbzOzB8AQPi8H66/BeC7Jt974dokEdE3ieh3iOh3emxqWT51yjiwVQrZ1cneZ0zzzyQxwQ3fYwQhMyEc5FQlPWxFw6ohzydKNp4Q7cmbLR7/uNjxx7ZYRWKF9slT4hKB81gkUDKDKqOozdJwPUMZu1IQQzKRxX4a/UPKP61yljgVAYwAtyx6l46ikxVYnJ+2R26iM4jihZkPTS/nVCyeMDZ3JJr0pD9lO4PIcvhdgAbC2RuMzVGTueRPmMIMA7T5rgpqeNHKx9rMqZIYM/8KM/8EM/9Eh2Uty6dKm6+9m8/xCCdnFsoaopj7bVcth7hDoSlqyQP0L5lCQbtAYRfRHzqc36doxUiMqPEHcoSgF9KKTTs4R8qfxJoKoyhgzwQVTDiK+WlsG7I2mc9s58Kgh2qaW3GB6EPhF7rSJ2JLqCGf4plRW/hNg7ShXUt0p35/Ggsyflo0OMqhuKuHIU5liNlQa2upmM7KLfp7FZjD8zKGD1VECJ8fhevvAXjH5HsbwPvP37znS+nIuCk62IWeJ2nHi1SbBXu9rItGDn/mGk9p134Oew6nX2iyI9IisdVW/oLTlau0VRfEFd/qBIrfUbFY0y8g5Z9lsDOPZDcyWQtxe1X7CyDb+tR81SIDA9jeFKbgu4TkxH8CmARXqeiLwBIYhry4YZdxKGKzi6LajegamjVhfVccrKoRpfAMc+8KMIfnZQx/G8AvhO+/AODXzfU/H3YnvgrgiYocP6w0Kyr7YsJcRs6zjIDz32kFpHjkWgqRlgov0bpdnPtDh8dfbLC5Y1ZpO0HLzjBFY6RSZ0GeUuDWUA+73PYiAwY1RWWoYyJmkDHkqjw3qwitdV6rCTss2RgjtX/yfkJf1Wt0XAbnKtJ3ALTn4ihlA75GUaKopztn+AbY+z5hewvVlb+mR6EB2H8w4uC7wPaWx8nbJKLI3Lwz32sMdGIZ+kNKl9mu/G8A/D0Af5yI3iOiXwTwlwD8FBF9G8BPhd8A8C0Avw/gOwD+CwD/zktp9Y60k9OWkBhIst8lubkcm2aehXnBWo6v0EURdn6yU+iAs9cb2X3Qa+X+v33OFMCOS7qVZ4xvg+Uakwm+Y/mf3a2YYyT2txVHPGXllOJJZKwGOdTQ3mS7M2OsktErg6GgPzAreOYkFd6XmDvLaeTdiUSH7vddLD89bL5HxiWiyPIJo9kStjcZ69vTrVLbV7ubVCY7tD9M1HChSTQz/9mZWz9ZycsAfunTNup50/an363OWwByaItDfrKTmrE+g3hBoyHyMHEivZSTFkqPM8fWMQNEADP8Iiga9VaDzEhJlJR1EcCaLUdiQX0cbAOTEVAagHKLEcW2ZtZ8kgNnJroEm4GMWXWtHiQ0w43c0FiPSrisi7wlVDLfg21DfwNYPE2BXskHcaIjNOvCaS5rJoWQfYgiTH9E4I9NHSi+m/YRh+3Lhy3O78sOxfBRiPlQYywFUyxVNtaUffO1d38oJtOfGcvH0g9iogCb4d7lkemzSkglcJ4yExsk1ho6RdEirFKZwhGAKh39wuHpj7QSHj1ASSWU4YDR3/bYvDZic9ujP/QYVwy/EBfkYY8xrBh+yRgX8ucXDN9yRBIlQQuBUW4QFfLoPV3V1cpQP20eJWj7vO1grKcUO2o4WhGNK94ZI4fXNTjPoh/gBji/T/BNiiZNHuGMzdzhyX6376c/IHTHhPUd5O/5gsWDBuDgex6LRw7b2x7nrzmMy0KkKBBO2fWJziog0x8GcvhMOVFVdVs6TzkQLvFUhLOrTykWID07yRcnq5FHvakzrCY2rkFEEg4xTNrmpsP2KE0EOUAW8MuwinqAQOClx3gQMjWBkBuWVd1yHBLmxVsHDA60JbQnLkVxUr3EmGZtZCAld1UG6vLL+ZcZlGacsDLEUC6VDHBjtkMdgrNUPlaT1de+I8g4jXvAsCdH3/EQmEbPGJYSu0Hd7e3Bt8LQUz2LY+DsCyHvkNqYJcqvidFTOvtyc0uO2XN9KIMqZcyk6S7Q5Z57kekzwxhmlPRZmgThICMW1FBC8ayvHJNm69bnFAr6lsRzD7p6cmIioZ1+KYe+2GAl7MRm3y/kjEYaKBALgc5dZECl9aO6JIMB7hhoGLQ3oL05AvcZ/aYFTjq0T1zQwE8hllUjpL4nIvdmC1WZiexkhPrdlOgLKSK7r5p+MswtMsPyUBqbKrCePOCDS/biRHQLzVb0B/2BAzeI78N6WQKI+hzyQBvCvvmO0AxmfsyJFeG72zJWjxmb2w7bm4zNsZPj8IoyLtqdKBHTDyN9JhiDQi2rX5hMcBUFiKcrTbhflV+L5FuC206hOSuTMUhCdybAQZ6vnElwer/BcGDa20hYNr9guE2A5F5gv4vbjdofqdyNZBgVxYVeVml5xUwAdQx/OAL/zLmY8j5dYvFhBxooI/QoJmRMx4gHZmAjqAhf0lhURBXzUqr6D90BIQZaAgZlqKa/dtzLQ24CCtreBvYepOvkhSFsD1yM0kTM+Zmgxukqnoh96LC39vH6zsSyeLTnHofvER7eBM7eYLitw9Ha5343pXhhb9k5a0SKH7Su4ZXXMajCEZjCWa69ACvj2U9f/C4TI4VJM4jYEk80airLqcRvBAA4YHM7GTJpGpccFGqpcBqp2i4rDlgjLm0vhb7RCLTnhO6TFvyHB9g83EOzP2D4Y2sMhx6+SYXHPlHlWvgSFaIFE86kBSOKaN9L6q7VJUylGENbBooxDmgh6mdIti6BsIvE6UDb9O6KmI1KkKGsxVMTVr4mRpjn7H0NHtudEMYli/el1TUQ8jZzPidKpGb7u/1XfwI/qPRKM4bN15LCsdRn2QFPfzZD+FT0QMhfnp0QBkmU0Yayl+dTAJeaQZXVL3BL2B42MfqSyLoChbm1RJLq54YxLkW5OK68fO578Zvo5J46KbHpgx5GCy9KumYDLB40cH+wB3y8xHh7gP/iuZxqbfvENBlXzRAVjwY52EGJiksYBGHyzJVvFZrZgbNm7MiMV6kcdAFlrO+JElIZjBsBN3IwgFL0w1oXc0kAACAASURBVNlnfIcOWD7x2NwChlVpOQlUFw5tIwPNxuPge4zuxGF7m3F+18lpWUVb7cJSEy/K++zoB+aF+UqLEiVSqBKhXRSoUD4qU6gley+uJGmLrxRZIlNw6SXqAaxl+/yCQhDXfEXKVkknJzBRIACJb8DglRfC0eccYzwgkGPAMYaRgN4BA8GtHdxAiaFB9JRugOgtWkZzTmjWHYbDBvTmGuOjBbonTsQT0/Yc4k4HbYfeEiAGG2qw42YfSIrixFztIb8WlSjjKdGWor9xKX+q4NWdi3FBEz2D7ibZOdNsWHZ/2tSW2PVdjCHkXR579A8bDHvA+g5h8TRElbJxNkuGODcX82H6gaRXmjFMJqIZ3JoMW4sAnBeIKUyMD4dLDaJ2P6t7oq209aYVnBuZmP2BE4McXTG1/RyIlhCsARm6fUlMcCcNaARcT1n9kp8lfmPngZbhbwzwgxOksHVRZ8EMseHgoL13QHvq0DcdmrtbjJsV6Jwjsdp+ZONuBpmtEqBQZEZxqBA3srLs0Fk0Uua1CK9EM0rDXnQ0/SGhPQ8u2R6gkTHsObTnyMTCFJ9DwsVxQ2h6D3biN7E4xmQuZGjIEnVoV7NhrB56bG849EcsuxQnwNxCNK8w333/ZaVXljGUugX9rCkg9Z49VxLAlHNb8QLmnrmufg/sUknELMQ2YUSGEEi03K6XFWp7SLmVo0LhUfbR/VImNJ1REicYUw9HMyuJKbTXZbCdCeBOEIjvGMOCQT2hOXXRP4A2hOUnDfzTBv39HuNIWHzUxm1DW+9kjHVL0hUemJWXQAz4hiMiKcfJogYmAG0i4qnuxoxD+c6Y0B8B3XHQMQTXat+K3qEZ08vSU8xFWUzRv6I9I6xfA/Y/Dugvlp2/s8x1XJm8l0hPB98jPP4TwNkbhIMPg+J6F+ooUomCmWTuL/7nl6uIfCV1DNaYqdQx2Hk4FS/sMo/sJUsG86f3LLPQ72H10T/oNp016S0SNxR3KfyC0B9IfiqVnkGJ5jYyMZuNbFeKci1nCnE1zpZ1Mx7BIYo8SXknDt0Th+5xA3jCcHtAf8QY9oRpMEn9zcMWtBixvTdkrsHV1b5QRGaNMy9JGcesorJ4NGPopQ+FfW8Vph7nQgMMB4YB+RQTUgPpqk5ID5AZOzGI2h6K3cewj3QuZoESY701ImfIidmnDNcDw4pTuPnKIlL2fYp2c8b5so2eXknGAOTED+QrTpmPCSEisl40GewEK7h+dj98J8/B1bqgR1ZZNeWNk8eF5zyCm3FAC0HezE5wDjBVZeO48lsOqFXsgJdp4nL2GHkxD148ISw+bkEMjDcH9HcHcCv53JbQfLgELUeMe7yzHqsstCLDxHwbii4QDaxEV2JepCIcx1l3qzsUu6B1QHZMYuwEICr/VM+Qi3rCLM5eFxdu8iyH9zSyQ5SdN5EP65TAbbmjbF82a2FE/T7l+ebB37RLBSqeKIRfcHrlGIN1qS6ZQTlY+apEU7ip3+210vy2Ik5o2VmdPk1gVUb6JUVFozKBYeWwuRHOgrArsIWo8XCXBIMoTHYajSY/av/rs8kq8jJCC/Df9YTFY4fFhy0Wn7ToXxuwfb0PzAFoP1hivDFie29MOyXENR6VXciQRWgseQpGTAXKC0wivShksz/pT8w7mEvmXapn6bgnjNiHMXW9jNmwdJEhM8l9ELD/scfh+yP2P/Foz6SM9a1pxbORr+0CE/IsHkt7zr5Asy7ZOm6T8UO5AKX7LxM1vHKMAZhy1RKS1tBDNVx8+Y5q1yrihD1TIoOUgYmozOm2ckaimtX6ToigP6Apc7KWjxYlWIJRgjJyv11d4xgYOFpdWSzBkeg14IHFRy3ccYvt/QE+yPbN0wZYjtjeGauF1VDaJIPhBLV3l6khYqCVhD5qYsxkpebK94BKhpX89I0UUIaYJ2aB/ANw+oUGw8rh9H6DvU8kzkJ/mBCGKopndQQFoyfPWDyVsvvD4LU5Z+w0k+zOyyxTfsHplWMM5YQvv5dQS19kaV9QNWaqEtHl26LbX9k5EVoGIW51qtIxTjKTbwKvgWQ7UfPVN9Si8RbiDoozYoSB6XoxQ1tedBndsex8DPd6DPuipGw/XgBLj3GVOp41URWFJXwz+co+WRuIBMtNXwzaqY1ThuQqDN42pT+kGLWaWMSJ6LAGeW/tucfBBx7NRnZ2ulMWC8jz8L4SeJtnCkj5rC6hO/NoTwm+BTa3aKqs1D7ZvtnuFG78pS7tZaRXijGo0rG26hS6rnwgi2ApVV0CKtcsLES6bl2otS5BAwzVLcSVO+gLfCerzrhQOTZviyrn4o9YpzAasT0wk6JHhgz0uWx15QJRlMsvFR6Y4X731GHxQQdeMIZ7PQCg+6jDcHvA5v4oodi0rcbGYNJ2U20ZPSp/yIyzikd21SeAG56eETmHIgwTpYHQ3wjh2sz7a7aYwPr23OPg+yNWj0bsPRgFTfVyuA1XRIAaYdaYR7tm7H0iDnHr1wjDnkvm85WFau5U8dhNM5QvS5x4ZRiDtfjK5tN0kdoBoTFdeWJBO+5n9eUQ1+aloXjetDH244AyZqCTPZOzSzHCtkefbdLkz2T2fGGefM+uFRDfys1uILRPRIs+3NuCATSPRTvZ3+8ncRozvwjT6Vo3MlwcEYR9tvIM01Sut6t41hbkY+U47k7o+RPNliV4bBRdOI+1AaBde7itjI3sWlS7mHWrpjwkz+hOpLHDPmN7WAk2a/pQioaZuFpJL8Ma8pVhDLXBmmMA1clYYxQlGqiJFfaFGUebDMrp5BpzMSJq1KMoIQFAasRb0ytE4jXGNHo9iQBFvw2BljA/7hRUTqoiXamM0s8NQPuRnIU3vrkRz8PHLbB16O+MGFe5+fakU0hl1gg99Vfhtw6syWYsH+eQ3ByktuPUHyQlpIgTjH4v1TU9qDiFlAcS4ttV35QBpq/tmuEGkuhQR/b91vtUr+CC+y8wvTKMoYYKJko35HkypSQLYWcx/2qIIRaC9LIIyUEqrChxTjdSpmq7a0oyDkqvYUWyDVasDJEpFB1mJ0orjc9gT5i2Id/imGgwFUUAVJQX6on2DRbi1waSCa4HFu8vwKctxgMvzlhPJRClf3ONzTtb+GUak5Jj50ggryv2X/vCM0FmYz6R/2vvrxQdYwpj4rvksGYjJI2B4KPHJQyT4KAfWBM2Nw0xV1IVoRlidwNj8UjqOr/PMTboRKSYKfPCul5weiUYg4VKNRF1F4PIf3MGFTNjmRIxAPnks4wCiSnY043m6taDV4c9ykdcibeE1uHhzBZgBgExIW2xKoqxHMM2xn5a5WMN/5p+wAPdowZ0c4vhgMUuoyfw0wXoaYf+fo/tXY9xkcqNooRLdWbvyyomMY/+Ju2kktmkvk8YvVlEQGLIRYzoXEWjROWOj0QxkeOnGxnNGtgeJeerHcMV72d9YcTdCUBsK8alu1BEKFFxOd/tQvSidQ2vBGPQNCfjzU4qFPd19S8jBceMtvB0jYmSlaMmFyaQ5faVF6xogSkZuEQts51ABXfT9mZEUBppEaZh102fYznxQtnfgnGE76WykrzEYfTnLXh/BBhozhzapw7dCaF50IH3RvRvbGXnwjC7ybuhZHthGUQuSxfnXhi0k+lCLEOoMXWkfEyQ4DeNoAc3Au2GMXZTJWSZ2lPZmaiFlLfDaFNk5to2D9m23MoZFHEulItTZY6XCDOr4yWlK88YakpHq2Mo05wCSB7UPHapQnoZ2QRGNQ9xEh/UFJobYFw5kUOLEVXDnHEhx6yX5WdEGC5ksmfISwYVZG3LgtsWZWn5pd7Ctk/vG+rM5f3UrvZRCzAw3Biz510PdB92aD9ciJn1Oxv0RzxVFiKf5CrSZKip0KdMHgKSAZi5HZtpRYWYISCFLoginKJ9u1FMoKMVahGngUYh6OGAJy7Y9h1OkGyBMAFgceqxfCAZ13cdfOfCOBQoFtP3tEt80N8vEjVcecZgk4VWtQleEymm+eTF26PE5AaqULTcb5a9cI5ONda/37fJsk1FDZCsUsNeOt8gWdwVHTANpgFoz4DuGNj/gLH6hLF4isk2p7bJrk7WOCgSeVGNhfo18SMj1vDpBgDBXdsvOCIeZWTkw87FSQe+3WM45ECoHI2xyrgMWn8GqwukU/bBoj+YucDKQAvmaQPZ9ro70QA0cjJ2IkREWG5tS2h50RGV+qNayhSmsRGIYeZpIGxuJ4O3SZ9QzI3i9w8COVx578pd4sLcPfu9hizUzXanJti+iCCPpgvFvRA6LELNMGG5EaZAHKIBBUgZR73ULShs3wBH3/XoToPlZIhN6FvC+Z0G63sk1nx2VdIJFr0/EQ16ZuEVckhvoVh2DoTe9kBz7uAPtuDOwY8AgeCM96MbAHfs4Ncd+P4G270W3cNWYiBYNFToHRIj0HENTARmV8UOl5P2WD4X4zZkwW0QmQcjmaIrU3MD8rGspGYrFKsnjs/tiMTXiIJYIwNjLI7l/In+kCUi9ak+zEBw/d5F6OWUyb6X1r2fIl1pxKCu1RMZNKSLdAua7LNRL1DC8rmXESZceZYhsXjd6THqutq4QaIE+QXFE5FoZPSH+WpmFXTaSDcAe99n3Pn/RqwejGjWXsoNDMX1jIMPB9z+x4PY8Y/ITKlBiMzJDeLvQNaSuTbjFKHYmI7F2AGJUKkn8LaJDmHcJp0EK9F4iRfh3l+B1g361wZs3hiwvTtiXHDOiJDEiowo2BgkcV4+BaabhZqHESNs95ShB9Qx7ukp2Hqfw1F2Dro7UW5dunDM4OZWeKcFM7CibRWpGtTZnnssHhG4Y5y97jAuXGKOIXrU3A5FqXws9Q8vMsLTlWYMNlkG8TzP7WQgNXhoP/W7BhzVF6RbmOqrH/Krj4TI35zCehUQV/InPL73EePwgxHNxlcnh7bFDYyj9zwaG3DEtDk72NULhK1FhLY6BTXFzhiWWaUVQbgewEAhkpQhRgswlDg80KwJ7rgBrR34YMDw5hbb10b0hyKOiMgVLDCt49hcUjTmOFMG1qB7HAuDUnzHghrM0DcbiSxtdVCZroGBZkMYV1NF5VxTuTIe2s7lY7nRHwX9hi1nR3mXqfdFpSvPGCxHBj7lgBjira0MszsMFcYRVy8TgjwrSgN/sEQnjtcJVSvB5pxw8P1RAnmwthEZQ7GTpzvx2PuYc8ZgGI/19Iy3S9hu2jTRnhUEan0b3MYBLQMth6jWckq0j6JLaP8oqMVtJR4EPe3AZw2w9ODXthjfXmP7Ro/tnTFs4XFV828VtHFXIioti8wuvRsA0eDMjtOwF5BK2DEiH869bArFo7FtcJvQzy4RcrlQ1RAOcc60JICLvOP+QCJ52aS+NrVUKt13idefNl1ZxqCQ6DIrfk3MKO8p1JJCw0fN4xLIUQKK75qcrNw2v4oc3FCK1kSE9V3K5N6ysW4LHP2Rj7EItd9xghdIQIiOsXo8YvG00tZAPBKGXhyG5saodKpKN2jykL6H5ozguhHceflbiAUkd+F0aaMM5eCs1ITAM+1xg/bjDu7DJfDxErRpRJl5fwP/9hr+x86x/ZEttndGiZZt6tW2ZLC62KHIztswaMEyh+FAxImxM6ihZ2yPXDYnLGpYPJG8auhkX+OuOToZ9yBONBsRJ9Z3KXp9Tp6/zJyuIIkXsTtx5ZWPmnZBqZ3IM5P7OEcFJQMwK00VBYTQcJHJsIoTHNGDQk3dghr2TFAWF/60USQa6r2PGcunPrNvKI2uNPCpved6xuIJi0XfDNKpMsxc8slXPtO2yacMgqykDljeWmNzugCvG4BDyLatnHzFpv2uJ+hhMGhMGzzJLgccmCRojObhpcfwWg886AR1jCksHKxLtu4AUNEPVESb8Pp9C4wroN0E1DBqlKXECLJFY5TYkXCFM5bJsmvXwI67iJ7CXMcVy3mlqkiNCwejCEKY1TdX/vOI2nPpyjMGI4JfKp9+t0knkBo2ZYfSAjmV2N/hWlXsQH6dGMAIuHD0mTKOzS0XYxtM2jUQDr7HOPhwBI0mAIoyCDvxir15zdOuPdg1OcPTFbMk9vBdmVmMpBREiTgVa6p2JHGCRmB82qG5M2J1uMV6XIJJggb4lQd3oqRMdhXIGB0TYqxJezI1MdCEQLg4b8DUYDz0GJYj3HGL9pSqMDvuPNQguB1H061hn7B8zGlXaWRADd/iPEoIsztnuD7YopB5R2V1M8wiXZR5s3oAbG8B/Q2PceHQeC8BenekKtMz118UUwCuqCihYkSmwAlpDhKXCCKDmw5RAXihcVNZdpgoE7EDaeL4lqJsLJNdLCX9grANDjN6PxJhcKXefzDmFpWYn1C19jWqk8CUMCJ8ri1nTJkYQbqCzzYgv94+adB/tIfN6ULO0VwEhakHuPXwCy8OVhSax4m5qdzNAhbkPWsfA9FFu4hjh+ZxC7/04v4ct4QrTMw21Zm5Ur5jBIYE0YtMUFmsghPqGcRqcVwlHUDUASHNt3J+VsUDDlaQfTCO09Bx2QE4PNc1uV28qnLx/LTixJVkDJpqyp1yAEp5qzZgAIKbMmOCz6w4sQMC2u1KVVolRR9Ft1wL6X1H0T5f2wAgxmhYPuTJcXdl/9MP0157Wbcjy607QxSk8oLec8WkY5pMaOS3J8kNwOJhg+67C7QfL+CetNH4CQDQsOgeuqB/sDYehLDViSBqID/lyxAqcXABP24wHnj0N+WgHW1z8hidjtkc0qwtHEr8yR8iIEF1mguKVN8lpiMZp2NUSmGl4hgs5tjNRu7HiF4XjHvJC0t62MmQnjFdacYATF907X5NS2uJ0XeUE1SNGcyt0vaFKiNQL8cg8zfbsIQUe9Bnr7kIc/NGAfvvM258d0ha86LdKv7Ea0WebNIUcSplMir8N2MVDYCMTsLAsgk6cyloiq3PL4DtnRGb+wPGfRYrzVMJP7/4uEVz0oCC3oEdYzzwGPc9xpWXk7Q6jvEkXI/cB8Qgi1inR/LqZKC/FVy+7e5OJVZDKUJkCswBEZUkvVBCEqWnJTGL6NEE+5XKXMyYQaqqKna05x7LhyKWrV+jYEGLiBqIg/1EERTI1mOvl+V/WrHiyjGGXRCoqlXfMQD2xbiBM+KZbEdy8Wdvl8xBxYsYKCWdF6H1MVH0u0dY0TU1W2D12AvsLl5q9tKLlSFODOsnEAx9sjqKWaviS2KCRcF6zTJhP5382WMjwZ07sQvYkwAkw57ca84lRFxz5kC9kzFS0SIk38oWp9oUWMYX4bk92CeIGG7t4M6dHM3XFuNnkMYcoy91qdnCM0rwFmGu0wLcNmck+vwuvYJFDdkixMDiiTDe7VHaBs2aewFxX6Rs/DTixIWMgYjeIaL/jYh+j4h+l4j+3XD9DhH9XSL6dvi8Ha4TEf1VIvoOEf0jIvqTz9KgUk67CB5d/FIoETsVf8AUPZQvMMDeqAcgRBuFGEXJIAn7fVwmiGxhb3umZrZ52y3h1VaACYMCwt57ZQzKZYWscrNemdoH2LxRJxIneEAXvRCQnoilVpBq0wASzXt3TOgeN2iOG9AoW3SKtrhhiMgFEYWsqKVWneZ9CEMKdVI4/JeMd2ahV8jO8IwDlOZYFAsMFYydHleXK5xpFAUk+dxyMhvzNDxTeF8gGGIWcWIthlPDMhcn5ub9rjmi9z+tGAFcDjEMAP4iM/8JAF8F8EtE9M8B+GUAv8HMXwLwG+E3AHwNwJfC3zcB/LVnadCcHsF+1lL5IpjSREsFFX9AnVEAuQl0VDalcw5le47SBHQUVzg9Co2LiQ3IpF8+4onCsezfHFTN9xmBs3vhFZYBUahilwCkGI8hT3USsZpyT8PWA+KlKAwzwG01dgrEzR1j2GOxoWjFMIg8hcNuHJpzCuWkg27k9Cfz3qyhEpvvIbmNuC9r/ElipINpNK/lf8ZCk8noNXSuGYbuO9m2VAMnFSuaLaPZBIWyZVYzi1NNxE0ZgfacsXhC8AvG+q6L7cTE8K4u0s2lsk3Pky5kDMz8ATP/g/D9GMDvAXgLwDcA/GrI9qsAfj58/waAv8GSfhPALSJ647INKju/iwPWEMIU/hbBWaoFIU06JfJA/KWmPzIby/1HxMmjEzMeLlJMUDDQnftsktf6MPty7TVKzAnQVTNliMjAFFRTRMaHFQ2MiXCoL/tL5qBcEqvGLYE24usBADSEPEi/wcIk9MCb7tihPRdmKigimEQX+oVoo2DGVt+XnlMxGZtCJIk7IYzk1t2aa7E8RIZfGjppW9w2WU7qeNZ2BnYpwrVNbmC0J1Kuhp6Lfa+IMpdlCpfNuys9k46BiH4UwL8E4LcAvM7MHwDCPADcD9neAvBd89h74dqlUk2BUx1YpHyXUrzMEZkVMQyDUKVghmA4TZIoBxvPRwm7JqtNfxhWWDftg1pM7mJ4u8SmxDwoxhfQG7OMdAc6mGRV2T60H0DmJxLPteiDwiwcXkODmD67IeWJnpeBwP2SxYiJBEZ3J07sE0YCnDAIRRncIjuejsz7iUpEZdy+yKfXLLMwyCCOm7mm7VXEkIaOYxnNWoyjxE4lf4/le55VFpvx7M4k03BY1zOUYsRF4vRFeS+bLs0YiOgQwH8P4N9j5qe7slauTZpIRN8kot8hot/psQGQH1S7C4bVEMWEU5Y9o8p3ml6PaAGVl6kvXSf9wHCDKKw2NxpsD52ce3jDYXtkeh2sHXV1Tsewl6Ni+uEKBhFEFitP+zL4i5bhikE016uH4paVu1RXtNgEzGpWMLqwZer6VFQ0ZnIhjoN5j74T/Uusbht0EccO7YlDsxYTagoncfuOM1FAxhBojyn3FQFy8dHOgTBm+k7GDjHmIo3B9oTl3aiOJGMOQe/QnTL6A7FRQcY8poxgl5ghAXYYi2OxeB32GeubLkN/oDzk/ZxbdQ056/XnVUBeijEQUQdhCv81M/+tcPlDFRHC50fh+nsA3jGPvw3g/bJMZv4VZv4JZv6JDsvy9u72mI7r52R135WKVUIKRbaroPVkqKR4TncfaERQqAm0HRcmKItFNEyTcxi0jkm7bR+p+B3S9sAlBR0XlVWwbNymnIEkmXJTk84QjXDkOOuDD0aXygSikRAj29KNiKMXonY9REfQMRTG06D3BHk0GxImcU4hWlZgrGOwnNQ+w7wXMmNWBG3JCLdBImyTx43h/Tkh3BjYJ7znZgM5OVx3JjKRY/cilg+2jIcbxSweJAfjzImYpXhRF5tT3k+rZ7jMrgQB+C8B/B4z/2Vz628D+IXw/RcA/Lq5/ufD7sRXATxRkeOi9DxyUckU8ps518/SjGktMVfPEMjaGBhEzWJxWBK2N4zprvWNANJKatpt/0rEBBhGZdtMEvUYMMQ3p7Ekg1BmcGc8tcoGOSHEqNRap8ZNQJtEgkhYLhUvq25QClrXZ+sIFnQP0aKxgOE0Sp5mTWjX4oHarCk8l15QZASGgcaxj5mQUEt4fxEVmdcTHeNiW0z06OA4x0B0wCp1AbX3OCcKA4jMBsAE/dnyiRNqqInQNYT9aZjDZXwl/mUA/xaA/5uI/mG49u8D+EsA/jsi+kUAfwTgT4d73wLwdQDfAXAG4N9+vqZdnEr5yybVKl9ciHz4lkKUIYAQwnsP+bHtqS5O5ZO+MCmLGOIYEwjCRkmWt5m3t5w4tUllxYsIhZcO/WFapZVoMzGC5bcb6wZNEWF5AqISEpHIMliuaEGLaRgIQUv8IjCVENEJJCu6eJlyZAAEAKO2VcppNrI6+zYxGN0CjQwqWkhSUk46yPZnk4eLk2uIkawmySM6clkT7XKm+IaMyJeQZLOFOFMdElaPkD04p3gsEa7tG3nZmegPWeJkNmRCzJnsFyyac3PoedOFjIGZ/w/UhxgAfrKSnwH80vM05qLOlEQ6xxnl947CIpGED+OXr88DhugwM+C6CpkVXaF0ihyUN5p0tapw+xoTygcg5R9WchainZjRDsEUGsur4EyF4ipecOAPKgbF+sK/ZDHJQOuBkYAhiVLllqbf4+BYFp71gf8YOwpFYGpQ5RtBItoGt5H+uKBvUHTghtCNIpakjn02jsrsijlSRntKptGlEVNYCDwHZzeKhlCTV1RBsDuJmoH2VOwjVOGqfYtlEGVIIXuvJtWuPQ8KB66Q5WOpJCkhNjCFS7OyFAUuP8cgzKqokFIh5E6FUZhcMfw8wqRpRFbMXyaSGBEaHWXbsjlU/z6nL2GHdIqS+Yv5L4Ie2VglBmIDzWZWh+Vn6BvvD8GOQBSIanTU3xyBN9fg5QheeiF6JdYmIQ9iRFd0NWpyI0V9AyC7GGP4m7TTzUx8axmqDMC2PY4jRcSX6ajGFOE66hcicxDCjQf81ocz1W3mbk1MBIDuTJgqSAys5uZDYnRTRGHvl6/7eRSQV8bteg4t7EIRu5hDZAjW9dcmm9+sJnGCjDzRxuuORYSqoXwagMWxh+8I/b6EBHdjDBuQqlQinulHmXZx+/6QshgNNhKzfTBem+F2ep9BOdpRuG8hsY7lCGAkLG9usPEEnLRAH8ycbwhFuz/cw+JEHh5XEradoAxHXphvivMjYCY0Q2wolAeTaO6bTWJCdvCsmOMbhtMHI7NE8c4lqIwGY2UC4AjsGW4MEZ2MOGp9Jlwv50z4luA8V4d2Tt0TP8277c482vMG26Wca9mdVuxcAmpIhXG6junrvUi3cVG6MoihRAfA5aAYmReTy3bmOLpJZZXfNSbDgJ4boUeKSZ3IyqVRvCTbMy/elJXYC/aCTLgaMzOT+ILkF6nd0qbc23BSLaeTrUDpmDq9l7T8CDI/0q6CFZeCFSQ84BxjebQBHw4Y7m/Bd7dAw6CzRg5WCc5mzUaOutMtR1mVNWJ0WiHVkjIzhda6GeKlGVdyZSBTKog/7VjasY7iTqrHBnvJguLEd8PxWQpnUMrZFFOqswtWbT6X7XEDozmXa8P+dM5O5lLcTZlOlNoi+TzixJVhDFETXazcitcfsgAAIABJREFU5XegGPDaFo4+py/NcujJoKUdBCuiKJG4gQUVGPdbe/ahPqf1qFXcVLzhOGkFlub9sf2aOwbd9k332okRDXlmVwmzGttj7xI6KsYkODvFYK9221GJdevgPWFv2WN1Y4PFfg/eNIAPxkrEWX6ngVuUmQWizsQfbXghusQV1ss2sCXaNPapw0xIXq1WhChQgwbT0fb4qFvJD8uxId5UAclN0ENMDsPFJFkmUdNRgcV/RhiDtKtK3DE4LcfPUqy29Wl7XmnEUEs1xBDlwGIw6gWY73ZCUrpf6iDs4GZEGYOHiKLIulPrxLXxF2T7jfMCQzt0ks21faeMyQih6c1qnjFDQ2PhaLnyFOvJMqZGRMFHofSstMxBv7u1w+ZkiZEJq0UP9oTmYAC1PhF9YHB+Ie1rzwh+GUSLNm+PPZUqWkmGFV2ZgZwaZbY2KWew+iONPyZbllFHwemMiagwDQ5plphqod7aM3n1GmBljii1LXPvVjLI4rM4ltOwtzd5erBNlp8m322by4XkQhqZSVeaMWiaU9zUuKHqAvICbIb8exaAZccAx/wNZWghruALwrB0uUa/pgCsMKO5PsfylfEsCOOew9m9Jl8Nbb9KPpShpekMsXoJu7IlRMZZW8QjMkSn3jicnS3Rjw2a1qNbiH7Br3x0jkoVyR8NIdpyCBrLRs8gRCuVR69N3WUIZtbcpbDzZX+ifiXoHGpH5KUForjPhqmGvxQPw3hasoSbBxDdzCksGCVTiV2vzV+LYBho16K7EMcyyhewrP2p8MnJ6wVSeB6GoOlKKB/5xv78vRJ+XZCIZeyIObOfr2dGRlDVeIIlZA1le9INdQBOTGmHm6S0H1e0ufJse+N1MxnGVqJNjwtCv5eOuBuXwZxYEUsJj217Q9sunCxszpbQ72HcOVPjI94gBtzGYdw02DoPImAcwpbLcsS458R/YsyHQcOZgcRq0pX6mECwjOBXsfJozhzchsQQKrh2ux71gDOU+pDanMY1IStlPJSLnNrOIf22uxKAOFKp2Bjfn5ZTKAPL8Y7vwgZvCajB9Q7jymNYEZYzIgmBInMQ5hiYA2jne9587V3gW782LXQmXQnGAMzDr0zm3/HchMDsZLDPzjGKmfKr9dqyESbFyNkWVrai2TflgP6gQXcqgr13QvzjkrC5KZF8xqWJJOQloEe7ZtkmWwJn9ymJEECm85i0V2mlRFGmnXGMjf6h+mkUf+JpSOCTFj0TqPXgPtj6DwR/OKJnYPHUxa1HHbNmI+KFbolqXAatgx0wHnq4NWH5UStm0w7gVoyl/JJBx3E5n3bK6hvUFqWGLNucOK0eoOkZviE0A5tihTk0WzGkGvdQiHMVRFYw7SrDYFlYumNgexPY3iQcfIiqErtkDrVk32mtHZdJV4YxAPONv4gp1G+az2K1SM8rl5d7jPzFpVWHJ4RlGZAaFo1LYwpdIgbzpo7fcVg9EIu/7Q1hBDqBm43EglycMJpgiRlDxwUC922D83uFHKrf1enJQlZO9U+2IHk6YbNJZTLEXQ1PYJiDURjgwYnB0xj0BYMgg+1NL6dR9QCNFE+c9p14Vyp6yHSOami0kRW5Pddguw7DoVQ6rFisEA1SmBBh0c9008Bv5S/WhyWIFRnzN1uX6jQ2LvKyyvpjeahfT2hHmEB7xgGJBEbueULgsT8V5lCqtJ5H6ajpSjEGmy7TsSr3VUIoVvXpw8hkfY3KVJat97JnAlNQOChWayQTpVhdpw0MK+IC6E7lqDkakc6/vKjPI+PovRHru23Szmt/ARPezVwLaCDzrgzdqE4kJQxjSUkDpT65sIKHvBgDV/US7o3s+If65OAbj/HAy1ZlLyu/Jk+QGA6B0NxGXMpdj2j/4DYEFwyAxiOPcaB4EAxxamtkEoSddiysW7KKZLYpnDyNogRsNtN34gZGs3YYVwnSl4x1jkirczrMq+5E2jUcSN3dKVenw6SMGWX2p2EOV4Yx1BQ2Nl0aFll0wMVnkdiJPwQY5iAUnmwXZQxoB3rhcjRrkIaBG3/o0Z55TM63QD6hZuHgwBIXYC88WopM5RjoT8MU7C6GdefOiLr4rrsxyWgKGfGr9Z4qDJmCiXVQMvqO4c4ceGnhQUJL0SJyENQRw8+rbD8A7YnDcODl4JmGwU0TtpNzIyfV9VQRoKl3XBDadQjm05DRFYQYG9HE2pwhMqb+lbqpGZBYhfe2LSBEhDgupF2L4+n7rz/Pqe6KU9fzpFdiVwJIXHiXSDkZhBkCkczmmoWMlT1kLTvuSJhrWlaMF8AFWjBLBbEorrrTMaGDykQq+1Prn3rkTZgCI/ccJWDCFO0zOmFdvvsQtfQ+5YuM0fbRmH0rIWseLVfOfJTOuUGiPWl746E3wRgqlglhNOMKk1mqoeHQiFgSU4UKaoQRV/TgDGUZvu2jxITU1TgwBRUntuGdN/NbizXxYrra601BLMJ805alRRq2HhGD6hXbHYrnRQxXijHMaXBzRQ1X8wJ2EHl+FS0HSqEjc3UgdYAnzIXT8xoU1k6qokHyiGMxZCngbY3wIxRGhVkQ4mnalghLws7+bF9sG/V5o+G3thBR4aioopEoS9H2gdJYu7WLz4sBEAJc58zBKuoWIPdUxtY4k3qStaKXYT8FkBVfBcLi4wYYCcOtAdu7I4ZVvmpEQioZo32HJGbRyiTIIzpHZWNZJGJGe46EjhgAJ1+ZaX7DjCoiTcw3cgyJpxHGS8QxKWNHfEi70Dwrg7hSjKFMVblsxja8TLNxGLLyNRAH4qqYlxE+G9k6lNOr83zqiDO/75wuqsFRujdtf01sySZ5eE4iCBXPmPgCtV2ZOEEskirrVkQQJr3NY3cufIgKHQOwbCnFUwyzKnpA9pTiPna64lKyNDTjYT1EGQm1yFaqMo8wlmsH2joQE4abo+h4TMp2I0qiDdd9F1yshzTOOvY0hpOqiveqOiHpq76E6bzMyrNjjJxZWLEubYVikmxZ+UJw8Vx/VpHiSjEGSxRz3HfXc/IwpgonnvnUPEaLrwQxLgl+SeFwEY4edxNUE+wJMsKN9RIyFs/IFJy1VJtIXPaDMNldqKKrcpuuwogyMSGUzW0akygymP4JEuB0TsQoBajowC2nMYllhmuBmSiDTCjFzHiX73jIsXepfk1uTeDVKK7fA0XUpswls0sph50lj0aylmtK7FKXG/JoTtLe8DvYOVzkLVtbrS3DyPRWLGgKCNaimF/payi6/P1pdiiujPKxpKFn6ghzzjUJyTvSElUFxonFGoEUElPwmjPnPmi7dJW2+/K+Ec11PM2oBhfjMoE8LmJR/mWuk9AXFk8Zm5tmu8/nE6AMsxbHgCvDUGMqkCGN5ztw8J5cAOOeB3eM9rgR0UKfd4hhyrg1h+Y2ZjfEySrcrIUIfFBEZtGqwVH8QCDCaCXpKSKj9pzgPuzCQqATJ3wUxmoqLkT9iVmt+0PC4tTDjSTu803wshxYfF9OkCHQGLDFi5t0W46bMXS6lJ4B2qYgorBEtdJTsLX9c++L4goxTzSvLGIoIdauNEETM1CKC2aRbhSfhKBEEq2063M5zSbXh0CgSDCS2+khJFmndqRnZgrh2vLJiKZP8F9XnskzMWCMGeOgQ0iRn6RwKvLEsXHi2zDuMYajERzOiRhXHjHku9YRHaWSaJHthnjEyNEaTdoGek1IIT2TmB9l/QALI+ImxEcgZOKDRSxZn8w13zCG/TwqtPpAuNHUbY8HAKJYqXlzHZBhIgUKnt1ZCP1xW8nj1UkLFYRhu5GhX8o+dynsL0pXhjE8q+gw29lMxishFnLkoKtHtDlP+UsNsOwoBMu3jUT2jWcTqOKsJqoUhc31saZPqSlCdZUbl07kfF0FlbAMkwDMfVs3F39mfDL4HxSN40IIbzz0oJAh6g3iGOaEl8WrGPM6vDprBREACO33BUFR+vMqtoR5Mq4Y67d6DPuM/u6A8WDMno0u3GbsJoro8M6GwFTsjguZeaELhh071XXURIlsrJHe22UIVP0w5JyNhDq0X7vmR6qQJ+jhlVU+Pi9nAwpoZaBihhjKSYHp7+iGW2lb/AzlqzVicy6f6kQzqaPAjs4SCfIJpJN+l9JKRZrHP+7gl6a/9ig8zr9HfYT+FS7UcYWHyR/KGJfBZ2HPgzYOGMSISYmKzBF1Uh4lRDIaQh+TeMGdMBt2nJ1hwQ3HPHZ8OHxhQjwKDyEg7fjaNgzsdFDjzgbSu8uUi6HffjlloNEickA678G8XzfKMfbWqK3cCbDvrlQczqXulNMRfIt8spYMYnYhLd2znyNdGcawK+1CE1WZ7RnQRyrIfJ2xZYj5CAKJe47XMmLMIKvFkvmlrNiMoaV8k0lAiCdFWQSUmfb6HZOvuJ5FoM5gaSJCblkUjECM3By71Rji1r5xiskg7QlE3XLcXfCLdPan7t1X35tph4oebktotgQ6aeVvcGgfq8dpPiHieZblTC/qsbDdjlWKLckTdEUDwO10ManpA8rf1YWQg7l10EOVjMGWMacXmrhlPydzuHKMoYRgF8GwCbTSVbCGDCyUhvnOmIodM23Jkq6MlROJJgUF5abvMF05KrBzV700MhZPOPYzyfWmn2UfzTV7kGu2xensfVmV/ZKja3SM3hQ7isiUrK4jl3sR7R5i1CiWzo/7GvyUsrZYfQkTgLDTEedDCKdnPSDjNnAoQJgBZ6u8FcXSe5frwyowBp+YiZQrBwqV29+KrDSmw1yq6c5miRrybtXlvNx+rT2zW2fBuMxWZi1dOcYwgcHPwPBk4iTYd6nw8fFhs6oYwi2ZE/mpQ1VUuumzBjWIeJPaYVeByCAKDfquiQMGxpWLZ0pU8xlizfQKhkgsysjyFOW0x052EvbGGAshruLGAYpGMuHaYMyUTdljOjmKRopxGyKDG5H0DLGt6UzL6GXZiD/BuO/BLaN94mLdOt76vnzDORoyzFjRFSDObBa1aUwE8pjaW4Rymo26wF+8sttxriFc/XQjR6vW4YCqTMcym9ldjk8pTlw5xqBpDm7vEinyldf+KAuvVaj1clbXRWglvmzrABU/KZOXY75a23ckW/9w4HD2mkvxCoFsW9YyG0vwMQJVsWpNNNfl/RHoHjeiX9Bym1Rm3FVQolJFLBAC5QYzaFVOmt2H5tSF06WlUo3fEI+001O2PYCBMBwxxgVj2GcMB14MnAZjr1AZ79inGoqMTILFo7FJp3uPiyQiWHNtu9i4PmcatdV8VmzEdMy1nc1GPtUk276jsrydiAFAFCeekUFcGTuGi1INYpfoghUl1MQIoD45YO6VjxXll/niyu4oKRVt+ZNtBROdZ+YF1160/vYLwsmbjZyLacuYMMV0j0yfrU0ChecskUtfTJlBvKAe6I4d+sOgHAz6BsuEyJvtPdt9a0sR/lxPCcY3wHA0ygG5wQjKG9NnjNKYZgSGGyP8ClHh6baEGIuVTJTsIpGiQYvQ7FxgscL0jTgykacI47UPaWyDyMFi5yBWsQAq9inygMB5ywzmlYZIJ41zLqbMKaJLHdRUb6ET4dlEileGMdQ4ZZk0YEq6gCmEhvluid0QeRn5d0LfIR5Jgq6AHo6atPPIOAoFGN0fav46zKxOGgL6wwYnbzj0R0B7Dhy879GGra31TYdhTwKH9EdmwqhSkJD1MQa0DboEu8pao6jsRCctLxw9bwOgRHPnJjY3Pq/bldwhihm+E5+AsWU0GwnGku+6cPJqHAEE/UF71kJEKdE3jIceiwdNeCch1hRT7hCmr0H7Remd6btliKVhv09oHzHcyBgRDKm8nORVW1Bcj2DgJQNUEr0wZcp/a50VNMoQhtOeM+AJwx5jXAQPYPPcpZgMTN3PIU5cGcZQk5Vm5aciVV9IYZBSVcgpsRR5mAgUMtc4ddz+08kcJrEbgTEzIZ4yKW6AYeXQnlXiyFWQBDeE9Z0GZ69LQJe9jxn7H3nZLg3tOFh7CeCyIDz8441sY4a2U1F+nJA+dN8QesY89Ni6ok2pr3KfSQg9Qn5TTGQ6GnhlCKPaiF1CsyG0pwmyq77CLxMTSRVTLHjc43hCVek2HttemziGKcSQ9ObZcY+ARwkBxTG0+iODfFSssjseFynJL0PUGsC3LHuu/NouSEK4z84UgCukY9jV4Rq0rq2wlB34WT5UqxTVPBYx1GQ79dfP6vYShzCurlbuHUIEo3Dt/DWXKaySYjNvE7eE9e0GJ28LU1g8AY7eG9CsfWbRKKs5o1l77H+f0+QtfTtsu2AnD3IGqTetOS6l3QMgrJTBUEm3NDNbAHWGIkTz5kiUob/Nxg5gagdrW8fs5Yqb9JEP24SBOZTvcBdDMO+x1EmpngGQsZQj8VTnIGHeMtENCU3Z4+zi3KzI9SVTqDIRRrS81RD15Y6Gnfvl9YkeggjPs215ZRADMO0oUOkoZga01LADUHsDa7J7YRuanHjKRAywF3lfrdQ0vzKG5IKdjH1sOnud0B+0uPVPBzSbPJryuHTwLXD6RoNhTyDu4jFw+P6Idu0TSqmtNgysHntsbzZJpCh3PFQEsCgpPBvvm4mbMz8ISoA9QCb4RAyCILxDJn4oY4xoK4g3viX4NpwYpczNrvr27M8w7r5hNGfCDPobHs3aGQWrmTxmwOPXkkFGhhqqcOnYOWHSDN8KCtSzLN3A2Xi6cI6lb/P5Kl+mL6ims5ogBwp6jp7g93SrdFLUZGGcU3Luas+udGUQQ02ZeKEMNSmkMkCXeT5CUENIF4yMDdSpeo1ozmpXfSeaa28s5EDAcAA8/ZFWlJEq7zeE7ZHDky822NwWrXR7Ctz4owHtuY8a/119aLYeR98dJe5DAX3LsWDTnuo4Bfge4Wy5s6HOUYzoTMUuWEp2yl2Qr9iDbGvK+Y+yy5AxbH1kzF3UVW/gdFtUT95iTFeTCcSc9n8yr5iyYDtAPpciQ7VNVWtTPRS5mK9zSmSrK5i2A1AFrc6duTls0WyNTj6NOHFlEEMG1c2KVdO41jhuDOway7Bv9YLK40RHmLyIk3lWx2HnIMvk0P3nKB+bBnPDGB1lsuv2FvDJP99mZapc2Z0ABx94LB+Pu1GCEr/pS7v2WD506A+oiqQyBSlMPw1KUMJzGjglIIlm7WJAV7KiRZd2KyIzaMQb020JLsjM45IjwbvgAOUXLOdRavtPKWuXb5KVpO1rMqcuBqZCKWSQfXwtzrxf4gjd1aI1nSsRDps1zCHFiQDGLq+n9j2rdwYZW/2H6yEoZkmTZzXN7VRM6n4OI6crwRjo6dn02g7xoXpv4oNuIkADE+Kpr5AwK1xSQM6228p3jcZaKFbZ4u2xSycmL54Cex97mfwtcHZPdh0AUTIun4y4LFMrr7VraUs5ebIAskU/yuZmfQ2ETQPAK07igScxegpiBZN8p5EktmOUwxHPsYwEZsW8IPZRaL+20wdnKGvZaE/PLkUQ/VLutmRjZd+P9i08O3YmCAslA63SyCk6WjGyWJ9zRF/+nruHIMaoLYduWe5aHMudiqyd4fry7/w2niVdCcZgU22VLifu7CpepnI1RfG7YBBl/IKL6smUPZwMcyZMoVjBaAD2P2Dsf2JiPwJYPPU4u9/g9M1Lcviyf0Y0UHm4ZARqyRcPetF7FUgbL2nbPYOCUpE7Fl6sodmCbQE6gDkwCACEcJxexxLlKbRPw7iNyyDLq3mzaUvy/wirpuouyLSdwvUGiCdRkUbaCm0wY6+7KjoW8TPksQzAdwAHnxSrWE2FzTBfOy+Q39PhtOJEmSJiAMLZFwRXeP7WFJglw3kmWinSldExaCqZQA0lzj9soSHmmcIc3el9u999QVutsoxGnuoAKrNj9ZCx//EICsSrWdzA2P9oxPIxcPI24fR+mxF7tb3ZShOumZXRfqoMrRrvdAZk0UR1lbb9C+10PaE7Dqtyx9EjkhtRmDUnTpiEOkUFE+rm1MENhOac0KxFR0BezrNstsD2lkd/wKldBsZHInPKvDl1X5FH3FKcihF55+ZvgeTQIPWyFWTHZheqEE9CvdzSpNxduga9ZplHNk04hHhDEFMoPXOR3mJXnc+Srgxj2KU82aWVnb0+Jy5kFZT1aYXzj9jyrYxK3jgaMTKCNxUA0H3qJPpYnRl5xv6HI9gB63uE/rCJ9dXaHuspJw0Vk9WssBkiCn/aDi0LZsKlXQaC2iN0jx26xw3apw3c2oFXHuMtwfvN2okJNUG2ajcUHZ9iX3Sog05l8djJwTJRJEEUfXxUbKquQcys1fPRDG2OzObMo2fmGpNxsca07Mzvw6SagrC2ktvyymvZvEUQZ4KYksyx5+vYyWieI13IGIhoRUR/n4j+LyL6XSL6D8L1HyOi3yKibxPR3ySiRbi+DL+/E+7/6GUaMteRGiSr5mdM3KXtQaSTCWGIIi9Yypm46RbtkRiQNHmhWVh3Lh6URu3UNIOB7tSj2cpqcfp6bvNQlj1ZRcL3scsZ1ywCCuNix1l3F8r6EvGJUrLZEJo10D1x6B60aB+2crZky2jOhHF0j5ycRLWlGGeAzLj7pXhYRoMha9Jcez8IQVsCgwLBbEcHkaO2oBT9q8niEUkZUcYiQui4BFEoPlsxQpow/B2iRfVe1M2keVbb1bio3OdFDpdBDBsA/woz/wsA/kUAP01EXwXwHwP4K8z8JQCPAPxiyP+LAB4x848D+Csh3wtJdounus0DO3iWZaNCqOavKCfGCbiI65pnBSWwnB5kmUVFFtrepIlzjC0HDNz69oj2XM4yfPqjLbY3mlRfrV3K+FhsLPqDAPdrE8NCU+PbkK2QljnYcZpZbZs1oT0ndE8dFk+ciAjnFHc1xj1xfIp1koSFF6tJIYBxwVERKzsfokB0xtDJdwLr3dYQawhVb4PVWEFcRSirl4jGVMVq7CcMVep2vcT3rI1ldraFGRP7vTZvq7oGkrnbBJsYDj4ctXlS0ylouVU08v+39/WxkiVXfb9z7+3u9958z+zOetldWK9tESCJwFobI6KIAElgQzB/mIgIKSRy4ogoEpH/AFuJIiVRJMgfwUkUBaw4ElFCMCRYNsas42D7rwh7FzD+iDFelvXu7PfszJuZN++97r63Tv44dapO1a3br2f27fvY6SO1uvt+1D1Vt+rU+T63AHsSBhbY8n9H/sMAvh+Als/9FQA/5n+/0/+HP/8DRHvbS8aPPrawA5Y65oOU75jECO7C8aLsvz1ufufusEOghMAWHKnmHNJ/A+jrG3yjbgTMTlUxiUg2WQCg3nU4+2SLtVcY07PA5lsqXP/mptwHgz/XBNeQ1CVYwJVoH8KhqlwnMUPd36hchZ2N8RwAdBuM3fvn2H1DCzeRhdVsx9Do+UkhCM1WJfdaxyoXFbkKYRfsEMSJJNGtf7aKcYowmffek8NVZ2GOabxHn55znzMwwVjKneaKSHss58qKCkK/MWhiGPYcrAW78EucT/78W7VIAEvqGIioJqIvAHgJwKcA/CmATWZWqfESgPv87/sAPAMA/vw1ABcKbb6HiB4nosfnmCYdKeLAQMlRIyEYuatyziHY75J4YdstjExv0WgIcrhAqgklCkHLNXhkufZu0QOVhIIycip1KnVnnJ4ndGvV8Fvzu+Hu2SqtS2AJle1zIh5Qtmi41+chkc4mROGKMT/lML/QyrMah27C8hkLl9BucHDWKj3HEi4tSOPUrbqKO6j2STMe5ay/bhDKKSTvyooDhF7JgcQ7lCNBKkESuTkwRsX7hogwpe+qtIEkYl9GeCw3cbu6hqUIAzN3zPydAO4H8HYA31a6THFecM62+UFmfpiZHx5hUn5u1lJpIQ29CPnYLTG/CJGlzDC1RGHPhaGhzJCX2ezGClFBls7eEhMwPSduz0nbhRdJLeOuL7dYuyw71ivfVuPKt47Qbnjdg+LnPSdv3F9j+96YVj5MZiUIGXsZuYD4YNm1Ux99Oz72QuUamGTxdm/ewfjBLTQn5kDNqK43aLa9jqGDZF6ae3OpT7oSfBEMC6+yftWK6zQ5Qrcups2RD7wKosMI3gnLf2pT8wLx/cB6tpr+WA4jWWSGmyCOSXai45P4GyyqLTG0s4dnlLg07RvFMco55bzNknj9ahSQt+THwMybRPRZAO8AcJaIGs8V3A/gOX/ZJQAPALhERA2AMwCu3A5y+SDcMuUdGpgSJxGeKbEANrV6T/Nt8fPVrAgMMMcAmJIG2zTEFWHnbsLJ5+IER2mSQIjDyec6NNs1th4gtCeAa2+qMd5kjG8w6rmE5+6erzA9mz7bKh8tt9AjREzlSZpDicsiWbhuBEzWZPvu2gr1lZF4PzYIpr0Q8u0A6HGzY4eQ78orFx1AJE5Uat1Q0YMbJQixj6rYlX5IMFQ1A1DF7FEp7vE7iB1+rILpNPdPge7iffGh1/zAHBxctIFjiQ5adkPM52IuruQb2WumfCSiu4norP+9DuAHAXwVwGcAvMtf9lMAPup/f8z/hz//aebbT1c7JEeVWKhcAamWCrmo1Ll4T//BSBbR4Is012kuB3WpDvJuriDx4MYSNFVUpiqKvv1qzti43GJyRZ7ZTYCdi4Rrb6pw5VtrXHuowu55pK7D2UQv2drtjrYXAbZ9Lup5GOi6Cm1boXppjHoXMfrS5nhQXHTiJgQK3nNSx0jSyVVzWeBB6ejiGIoSU2Iv3ERqX7h1SfvWTYxvhLYP9MY8Br6ZTzjp7yk4Oalb95AVa3AIs4Wc9B9mUVepbiPnMoZECdv27cAy3bkXwGeI6IsAHgPwKWb+OICfA/BeInoCokP4kL/+QwAu+OPvBfC+ZZEZP9pXkiwWFeKxoHEeuGevNksEYqj+ZU9haJ1xHDC5GhdCosgsvKmdc1VRhND27GSlDjjz1Bxnv67JFPyzrQmNzbEq3luyaAwSAX9PMnkL+gko0fMXVnNgd3MNbSsWlG7NWyu2SCos+bFKxss/KzHh+udowpZmm0KItfPp7OdnHGZJPgQXAAAgAElEQVSnnX8uJTjRrIrihG9bfSB6c0T7WsFzPjo48uVqQjuJIgSrfwgB09MV6p30+p74u4jgL7rOEIA8fZzdLHNCkBOJ21E8AkuIEsz8RQDfVTj+JETfkB/fBfDjt4XNErBIXst0fP449zgDziagXKjnNFQ62o/tbmoHPugSMrl1dBPYzRZg0DUYBBmE2RkCXliOuuvzxlsdTv8ZcPPeyicjRdJHu/D6bH/s9xBbascDfvh6XIeKIg5o1zm48FZbNbDRwq0x6psUnaOchC07uxVZfHMLgz/ebFNQ8rqRmD3DcY+XGyNksgb87s8A+1ySUuaNYh+yKEq9B+SdpirAhXcvpsPw/muA5hJs1W7o2KdZnHIY4nqLm5viYwh6zjGUiM8QcbhdODKej8vCkNw0SKnzwdmTm+BwXZJFmZDsOIpLaQRHOy5J2gLXv0bfZrsOzE/0tVcL5UMG1q50OP+1FuuXBd/gd5HvfP765HfeNscJFfpKcXKyn5zJJK5jW6iA9iQDFVDvELobI8knMOaw47lRTNcmCEe8kkltfA60dgY3krVpfs7BjRAcpbj2KdYpEn/qKHzTjEIJu4C7jYXJE+P4YrpuFJW3zoRUk0/eoveMbvos0ZWZNxkMKRiHjqVWs8ipBByzDcr+LykkbxeONGFYJHvn18VJnDkOsWH/gHQHNWy6/g8RdX4iuDGJb8CY4Eby6dYqSaCh9QWzl1HNGRsvcIxN6DAwC2TS37i/En97c2oh1deJ3TJOXWrTQrmMvitw3mezW1klZNCJAIm5LIQY58TW6w/qHUnkOj8jOSPWnm/Q3BARIugKRsrSI8Y3+OeoziCkiR9LbQVykiZ+ds6hW2c0N8Rtuh+PgqDLkWSqFHNT1j7TE1Py3FznopaYbiy4dhM5ltQBcfCsiLgsj2+4MH4SXxFfVklRncxn7uel5Hw++r7l2b6G2re/l9IVLYAjTRiGOrdoUAaBgZwIhO9s4YTF4d2e7flqzqF2pfrV58ljq5Yxue76C3Bgtc9PAvMT8VVwNmFtP3sTw4npz5Z6C30ssK9WF1PknvJxMW3ZXTeIV96sOdqsUN+s4MbA7hta2W19ABApwtZSY1jzEDRViQKx3pE223Wg2/AJarTitdr0TUAYdT5LlC3g60O6ecwxg7clBhmHlGfJDmnszS4sfhc5a2re0R5+fKmoRsPngBB7Y3EF+ptlrmjk7NrbhSNHGEoKyJKYsGynZTHpm0X8ztlrPZVniHZRkWg/1DKqmZgnk8nhTXH1dCDVUvYGdSLuXKgCvkP9yG/XvtS7ZoLnu80AYdHzi1AM1xTazPGoOnGJruZAc6PG7GIb5WRPFINpUe+1ixzCyle74gLdTVhCtecUEsAEkycLV5Hgy4qH5XD8wcwUasdDTsRvriUpS5LNCwicQsJVef1VaedfNEcXKijJfArn9+IE7Dy5XcUjcAQJQwlKHMKwtSLNoWgXUDw48CAdVKO8q6em6rD/tmy3yvbEIlYES4YDJq+YtpTdz7d8Em357l1ebKHFLz9nGcFICqsOSCzJeIRPxb2JnO88iYmOpR+aNKWHI4v5rtkmjF9uJM17rUlaInFWvQOrz0EF8IhR70pg1vyUQ3tS4ipCAlpDREOORU8QyBkGx7o5E0DTKuZzsGJEMkgI3IRUuYocmKsBZfutRUVQonRTQMohDs/R/rFEXMmHNdMxLNoU90PxCBxhwnCrbNAibay+wAA5O114VmpA8DuD4yTzcrKQ1GRo6lpsXHZh4veCsnorWOIn8gWea5t7u01NYYH1/Bb22lmstmrhNoRU8cqIyU7McxLHKqNkrGemDb/A7H3C9ZDPEiW+CDKWCN6MInZwEGNEj4Cwi4ecDGzwqIXYxMFC+Xc2MDaCMhEfehxYxrVg74WbQ+k9B/wKnNoiZWbp9+3CkSQM40cfS5Q1Q6JETgysFt0CeTawaM8fECusllcLnwStdGjXvNA6LXYjkZYdRlvxmdRi4azZubtKQ7nt5QXOhytgekaySvcsH0ogVImYPTpOwAFKhPRa4Yi8aVA9FvOdzyxOLbCjtSmrKYUckeruHCwQFYfkuvPTTiprd1qZSnIvcCVKRHV0StDMA+Z0YU+roJgtKefCcStSeAWmaxCtBDoXsjFW03Y+TotgSEcQDxT+K5Hi9P4hXQJTWSS/FThyqd0USoO8F5sdZE3/3ZPdgPJOkXEP1h0aAKhluLHRsI/FJm/LzQdtO3vOgmRHO/V0hyt/rpZdXXcgb/u2CDIB8xPAK9/RYO0VSfsGSHGadk3qSjiNjfCLSuotxv+2Dz0OAv1n9sfB8rR9/MLErBiwcRqIHIJVblZTiW/oJojWBzUBmySwzTah3WB0Jx1oTqhmnnhoODOLGTRYCCxHRAh6HTcCtAhOvUOo5lWiR+h132wUIRtT5iwGiDhRB/dkO1z+AiN6lhbrXrv5IEHh/jVB6bsPXMEiOLKEYRlrROm8Thw7kfsXDjWQ/rb3VnMJsZZwWAltrmZRn1HNGW4kx1TjTAw0Ow6TzRrTs36y5UE35i1zTWAH7NxN2L3QxEngq1wBCFF+Qii0DaS8n7Lr2f/ebF2kBePYBxteXRpjKzYo4VMzbT0VRSJM8VlyYn0ARMzgGujOt8C0ApyYgas5ofNtMMmQBndv5RA8gXA1JIWc+j/M0rgI3e3z7pV0JBIp6n0fZj5eQRd7wSWalSvL6GmukNxLYZiIENqW+T9EXEqvdD+IxpElDONHH8Psh94W/tsOL9N5yyoml9qd1EKye2g2Z2My0sXlAAIHHwfZCRlBrq4AVk6CZWKdutSGhb5Qtg9EBjj9tBOzaIdoKTHckNrO5ycqXHuTUAEbD9BnUU1QDgFlitkHvTZMOnWDVnBmwWa7MxhS7NWXnJN08gweS5bp0dUK7QlG9YZd0NWJmF7bSFSaHXWFjlyHui53Jx0w6UDbDeqtCrUnBoJjXKVsXnTOEOVzQR2hErdtW8uD02/1j7GWihIss2B7dNpuTAsC+RTs+VcrRgBHmDAoLKNlzddZ79K95Lj8GJtjdt12DG5ikVHlErgiX/kY0foAIGb/YdRThChDrWxUgmoOnHrGYXSjC89PFFMc+0gdY3y9w9kngGsPVehyEWIZ8I3HSD4zi8231XUUd1o9zunurIrC9mQXdvlqu0Jzk9BtMOieKbpXJiHLtHJImsnaTUQZ2E18dqcTHeoTc9DLaxhdrpOQaX1evsgG2fR8l9ZxDroPS0hjcZ29xpYKQ3jbu7gXlXLlZDLfF3HHtwlHUvloIR/UEuXta+4LZb1ucdB62YHMc2z+BS1O4hoKJq4efg7YeNEE8ShH0VOiAOPrjPFWl2r7s8mQT7bRjQ6nn3JRN2LuDX1nM8kLvKiaLofEi8GJrZdX3kuwjseV2FRTYHylRnOjxvjlGuNrMoDVfdvothpRVDKk+tQsyiia3k3yMQB8bgZ0hPrJdYw3Jet0gp8flDBeWRXvBN9SVwxHoCbZZL7ZDcgqPAtjky/YXIwpLWL7/IQ7ceXrcnXQfuodjjRh0HRvi6h/fk4+nLJiyw7YEpwFtT64yqd0IwdxdJqxKVTSx23j5RZrV2TxEAPUphNX2gbWrriEdbT9LOkFtc/j6x1OPMvpJDKTMhm3vbaWwkzrKe1UdAKgxWYABKViUFKyOCxVM3GRrjrC7jfPcOatl9FeXhc/gxaA96BEpU5KFPQNfH4Ot9Fh/I0JJi/VqNosdwTruyisuLCKzCfvU8YZguAtSIhihknIGutdcE/vMLTgcx1BTpuTeV6Yt0PExALT/ogRwBEnDMDebFhpgHN2T/PnxQuHGsu+hy5rObDFFscSmxfu6YATL3TR/t8hUZCBfYWqUjKRDCzhsX1ev9LFmpoGh9KupSeDzqP0gOzaeF6PIRLgEs+caG/lMzvrcOLcDl5+5pwQF7/rV51wFtVUEsO6EcOtObjTLeqXxhi/1PQ6YR+XEIoCJ2bxDffncRM6FuyJU6GcgOoTbKKWYBJOqq0PD8WiY8kcDJxP2q2SeP2qxJUCHHnCYKFHWVHmGPo7v7zskp/7Iujbrc1E7DgEValM6sYUvBdL0Gw7SdGmu47LuAazGy3Ei8vf1Yxx8tly6O9eSquhiyPrTOWx3aPBnux7YYqb19YktmEuKeXrXYQkum4Sx4BPthi9MJLYCfVezMTGQUguNIeHRArbN13Y+fuv4reG5gcdDTOsNSrHcYhI9Ppgr1PTKfffdQnuKB2DZY2iqFAe8EUQXpYGRWVchYWgTMxeYMqOy0JUCwbI/1fLRGETJgecutQFZVIQKdqoUW/XqI/XEJ7UX3gbL7a48JUOo5sRT722OCC2sQUgYgHSXdfeQhyIR6nQC1eiK3CzGvXlMcjXmSAWYtCNfQamhuE2HNy6w/jSOCRhSXQyhXeezAe/IsM70++ca6T46eWbrNKFmVsflGOIIkyfEOQ7fI6vnc/988oWInGc07ZL8+vVxEbkcOQJwxAs2gGtsii5R81+ZkIk1+WspnmxQ4o/icJjU4GKU212dj91jPGmz8qjE83gMz1T9fDOF0LAwQZ8mQlZ7zqcerpLd8EhGJiZvUMJa5YdynQR9niigGsg4dA+70E1h5gofWVnrsXqAAbGl2vvQ8ChHcATqEU7bgntbLcu6kvsWIV3bZLDDsRZlNj5kk6sxNmW5ke8KMMxP10gDvsJx4IwKNewkAXzkGh291hgvdh8nSSOw6RYxDWU2EPxr2e4CRWvIwZOPt9FD0ggZgRugNlpYL5xi6/FcEC6EEbbLuZp6E26jNc125zuuLoTpjtx2te8neSdFGZuN2ZgHll73aX1MrfuUG20QhQ6StoOMKRcLD+yh7uFXNnXq76VbQj2cZZzyBe11TMUF312nf1OcFMxZcBHwl6/n9wCcEwIAzC8KAcJhOufDzt6XtcRyHYVH5iUubr21lMV2c7k5Xsz5uxMnZY889+jLYczTzqj3Zbnc8VwE+Dqt9bo1gucAxC5C6PoAuBzGfiU8r7d9ctczIzcT+QSuYYwIc0ubTkwy57nisAEeivJfyqJmFQRQb0327MtaFph9GdrwepQEnd6i38vRSlxYqkJXbVEgbPbMiIY3p+PtCyBnW+5OFES+XJUi+c9bmo6VdiLVu4HHBvCsIjqAgODU2DBhW3sJ2BJ2vRWDDeK1yTPIaQVjgmJmzMxxHe/Tnd/28bkWofJVTOJfJYnZbmvvqVOy9Lpc2spdKvms6oT1nq+UWH7rgqzUxXatQrzE3XANSrVzOLvjRWV/yeLTM/1Tw3qfDIRpNqtIoemYdQjgOYVxptV4oKcgMF9L65t0aqxokVO8CzX10tYa/oNIOgfuEYs1muuL+FoOYkB9NL29X7XJzgWltWz3QocG8KQ22cXcJTp+QIFT6oODezKak7sJhVcQ+HDta87qXoFTUtGlL5MBkbbjG4iloocZ+qAU892aVYjg3s3Aa5/S5VWuybg5sUaW/dqsRny11bYuq/G7gXCzl0Vti9WuHlPhel5ShVoPQVJZsHIBtUuiETWpuy4aceywLbZMLl95uYQdFUD/IZdjK5Uw0TFbqcDosOixWgViDkkzRZ2eu2nnFCuM3J5OlfsGC3awfO2LbEoLXCNUrX9S/Q2tH++CxaODWEA+haKvV6ANVHmL80qCeNN/tuLHNQx6lnkQ/WYaontCw3Vr82I1jOJdWjXq6Is2Wy7qIikNJyYa0m/vvnmCrMzIlps391g+5sI85PxGVwRNh+qMT8pIdHTc8DuXYTZWSEuSQp5i7T52x+3PiHTcQlNeCXgUH3O4iqtJUOTZmluNxzcxSnqS2tRn+EJd0JE1VTq207eZUnsMf+17y6v/eCP97gB4oRjCJerlYLj/5SQcO/95kNiYaFY7DlD/V91hQb9ffutW1A48rESiyCnoL3zWsuyjubHcI/f+SR5rB4ceo7sclwTrPmo93wlNprok8VRphvZtlJ8Tz3bYX6qjoFInYgScrEQiGtvrHx8BVKzGgFuQiG3ovTZ//bOUs60VZZjsxnqL8gXR68UW1iJpfFKdzUwSZ2HkQPdbOKYnZuhfnYN1dxfW2nwWrZgPcEgpGMnnEhGKOwL4bjbyiZR5izDmJlzVZsNjYkMTbKH67lMXCiNSe+5OUEIJ5GKMlnOh9Lv/YZjxTEAQEn5MyhjeccmrVScr4FgH+4tFvu8+KGOpS2vX8hZwFJdxMq7UMvk7BOUetfhxHOm2pBJFxcUnI3PWqyZmihmr56dqFDPgHpHcj9KlmRJbT65ymi2lWilOOqCDb/zGc06KWmQTQ7Mh9muqXQt+VU9duH57ZkO1XNrkmHJNNpj+fV0QarI8Yk6A08UskVldQq57igcZzLvLbtm4D0nuSjyNg3OOa75f9v3XDGdz/uhzXC/4NhxDONPPn5L4djEkF3YWxBschUgchUJKDHIJmjYubzCjysqOJ+w+Q3UOw6TzhMmE/ZsJ9vGiy2YGty8zxOyFiFJic2GrNdzDVz/5hprV8TB6sTzLiwe6oB6LnEbriZsX6zQrVES+CUcRVy92naSy5ERismqyJCtuzi5dSG6dMXa99KeYFTXRiAHzC62qG7UaHYocAKCG/Xe4VI7ZFh5+oq0b0i4gNC2JRD2W8dhjsCh5aZV3WT0PiYMZomyOJc4CasHSUQfkoS0APtcHAzVJ9lrXwvdgsLx4xg8lBQ2iygoqYKw7tvmi+bLAaKQTyJuqC+7ZrtDPfXh2dkk112CGNh4OYYlCzeTdSqw2nLv7CywfY8oRMc3HCbXHNZe6TC51qHZdpIwBpD6F0bPoHJ8UKIZHLToLBjinenIPDsSgRLrXtq59ZhrALfRoZoRZne1QEcY3agSlrl3c+FYiegPvfOEACwLhnCoyBBiT4CU6zJEJWf1hziDpXZ4PweC0rgtJJ1Z0O/9gmNJGDQn5KCMlkF4MSoK1JQQljxlfLzRfCtRyLgNVWImOJRkTHViKrxYJsFt/SVOF69PikJ2ARllW7sBbN1H0EpIdjFwRZidkhJ2g30riAcAgheiVTJCOQMUJqb5YxeBHm7PONC8krYqYPJSnSaTteOQiS6ltpNumOsSIqOWg5KokY+DEk3PIdQzBny28dpUL6+UaPr7uKLgs1Iy4drn5gSjREACPkDi/EaZCAq8ttwCcEwJA9C3UCwDcVf0okBNaam1vbiG/DlhEXCoFqSciWUTRRmZWkFK8uWpS1JyThWRSfYgA2HnrkTkuPwXGmw+NML2xQbzUzV2LjS49lCDrfskV2SPZc76E0yJtfnYSWy18QMTOleqKriGgZZQ36zQne4web4JJebysbA6gNBeSfzKxs/+icQeGYHJBjFfqH4l1FP5hMu8J6sqsHvKx8oHgNmmC2KD7V8u/vaJBoVEPlWLhGMgfu2JAnAMdQwJmIg2oDwxgfRYnHB6P2Iux4XPWnDMsH9WuRcnuuwsdrSLLKaT0OzZmUZqRTCivsF0QoiNXzAVg0eE6Tlget4XZ6HCxBwQhfQcefGh5NhnlWI5p8Al9kgfQcD8lKRwb8+2qLYk/qGUOzG/T58VlJu5YsMMYFF+tybPAYIf+uW/mYQoSF7NuDCZRBzSit2JxSbjXIcIZf6+c31Bghv5xLaMUL/zoOHYcgyAKCKB5WS4oV2OLAdB0YGp3MhQ40j8GIh5z5HNcdaJUk8Z578qJd40DbpwOTJzU02/mVnqiedNmuHegYmnv1Vs0X5YXYK2b3UJi/h5S+AAYHaX5Idoz3SodmrRKwBiuszfByMkeynqjfQZRmnaw4E46mesuKPtU/wNmMVtmqp3GLWKBgTYYjO1z9YVOAbP7lc+P0dJB1KCnL7l16kPi3IjVqF9ENwCcMwJg8IyosRCxaSKGP4FhBRsQ6JFfi5hSZX17S+g3DMyHM93u5YxeSW2HSpe5fkbtCGKFaXCxLfeeSUo4N8zw2U5EHq7n7E8JKAZndgTtJrR3CwT22RXNwMxJJf3uuHyG5AGx+V9hTlnCYfS1xYhwC2Jf6j645PoggbwS3DlPj0vKdFFlIAvdcj9miEHAMeeMGj6t0XQ06Kb44myrIP3bIS3ItDek8qwosSciCQl+dGe6+0UBpdTz7ZSX0GJkH9mvoPmjSVsuuop2vSDEsGwky/jFIb6kKNhj83OMOrtCm7i0FxtoqPRHtAjkhx/LLyfKSGg/fvlY7kEey6vjqUWibCjE3ph7lyRNGs4lEVjlfdvaHNQ07omDtbrDopbAF4HhAHY2y003/FK58J/L1qouc81VA644sK3yqpV9Ka0rKJ1XMonUc5S1lPGha+0UsnKt681FUSPUO6MTmKuERyi7E4VTJSeOOhiCKnXzULMn5HLxSX5mUnSt0lSV4BPdKh3kRAaDakeoG3l46XVZjwWo1jYvyw2tngxavp66iQC1+bwJGOR0LELeoA2Nlp6LTkxGNKFhetrkgpe8HU36GCJAnALhIGIaiL6QyL6uP//RiL6HBF9nYg+TERjf3zi/z/hzz/42qC+N5Q4BAuLJmIgEHPDAQzpHij7Zi9KZK6zKr+XWPMifh3j9NNd5BrsPV15a0p2LELQOSzD5pYaWlaZq7e0J4TyUAe051rQVqrf1r702s2UqyURwrLdPbrliWbOCZV0GeF/dn3i4JUQSQoWomCh8cRXjxeVyVnXltGFAYZjaCmZfwcJt8Ix/AyAr5r/vwDgF5n5LQCuAni3P/5uAFeZ+c0AftFfdyiQ7w5DL6So7Ao3RhbSpnHruU3rt2VPqe/Jp7j0NO/mnMWluekwuomQ98E+o1f8hdM8BkGUUVdqeywL6w04VxlC2hzx4OQPE7+RgjKjGz5SUnULBQKTtGO4p0XcXX7C6lWS64eIYPJu+qc1ZkOVsNbVORAN4xvhGvj0dIsXb/6u83ec496NKBS/sdzIQcJShIGI7gfwNwD8Z/+fAHw/gP/pL/kVAD/mf7/T/4c//wP++tcUlgnLLsl/e+2kQGThtTpVonsoTMhQRHeIwVhAoJJzXiN+7mstxptm0ZRk6AX9YZKU7jbYKql1yYgK14KqXBWN1pchcQwiKUbbnvBEoZP8jfXVJiYZycUOozMI8r0haHrxEGtO2e7OhMQxS04uIBieuHMVCwGNthj1lPtu6ACaXdUwKgcmiYBVh5OMVQYlgjqkf2IidGN5Rj2V7NmvVQTlIliWY/gAgJ9FVFFdALDJzDoklwDc53/fB+AZAPDnr/nrEyCi9xDR40T0+BzT/PRtwSKnp9KLWXSdnrM7WIirICNW5JyDJQ5AHOGBBbvwt1ayakWkIF++LW8ryOAFvYDtpAZkWTde3QGTyMn0tjIho4jL7EIHrhn1tiDGNeDOtCFAijhbaAOr3UZKDuESX5y9N/bHFryxXBEbfBOFrrat4qMNxw9RtUbsMETHNRj0TBxUoi4YAm23m8i7pDmw8ZufG7jwtYU9CQMR/QiAl5j59+3hwqW8xLl4gPmDzPwwMz88wpDf7q2DzQ+5SLk3uBMt4tzYiwdDi724+OPBsAgz1nLRhFKopw4nL8UdNXhFmknd63DeKHGirwink13W3lfe1WShM7hm7F6U2Id6OyZamZ0TdX6eQDUUpbFEsMq4gkxkyUHfYzifWyL8Qs5Do0NeCsT7LPfUaHSqi+Ht8kCpMq6ZsgAE5yk38iJGYV4Nza9cgdu/SPJqgIAH/vX/HbjotYdlPB+/F8CPEtEjANYAnIZwEGeJqPFcwf0AnvPXXwLwAIBLRNQAOAPgyr5jvgcso/0F0he1F7EgloQfurPIwaGGEXYYmagUjg8p7+yx3rMdsH65xfTsCNNz2TOAtL5hgaUQj0NDWDxurolsPBuqVdQh+MXSTRjze+agnRrNVp14581PO4zOTYFvbARuIYpdKV5WqRieX7rAnAtyv1/oIeFtzgkQwuJns6B7Slw//s1NRCViHXNCcEXB+pDEMZAs4MnVFN2hOZSMY8aVJoSyAdr1BYTjgGBPjoGZ38/M9zPzgwB+AsCnmfknAXwGwLv8ZT8F4KP+98f8f/jzn2beQzuzz6BcwyLttkJO4RexgEMviynK33Kg/1uUWeVhsLvgXlzN6Wfavi+/JRD2vpKeoNCoPZeMA6Xj0q0xpvfOMb97jmpTqkyrr0UIaz8zx3xzksjdFpfejmlfSnZNuI61tiUCF2PFB138YRxKnJwltobTcr7AlcRBiH7B1UjMk7mYFTkG0UnYLg7pGPL3WpxvJISIG+DN7/29fkMHCK/Gj+HnALyXiJ6A6BA+5I9/CMAFf/y9AN736lC8PbDEQWFPjTcGXlj2Pw8VDrqH3gL1wVOmeIhtaxllaP6/mjJOP+WislDbZQQNUFwkGTtEsShMmiqdy/32t7brjOnFFu2FOartGs2VkSjGPKegC9Q1QDNp0Vyre45GOYUuvoOMW0lwcSkhIH2uEgnf/6iLyDivLt4b+x267/Me+CrklKVT46wvAGIlc5SJco4/ekPQB8+RPvCvDk+EULilICpm/iyAz/rfTwJ4e+GaXQA/vg+4vWoYP/pYktTlVmCIeKSTfeC3HvJp4zQ6L28732Hs/1yXaNf4+EaHE88Sbn4TJZOb7SKBP+XI40xhdySXrg/R6HP47tYY3UkHWuvAHYG2GlS7FaprdbhPWXj2opIbM/jiFPyNE6hnSDoWdmmPVOBQ8gGx+OTjUaf/w0LPCEEoH2feUzVHSri1Tc/lVDOgngtHp/krml0Oopb0BwmRd42aKwcmCjMIqbl6kcJR8Tr5G4ejbMzheEdXHiAsVBgtAt3FK2+b7tLEpSV9gp5TyIkFQ6Iw240G0/MIyVNJ80UGUSIuFEDPM0Cib2AC5qclFyMIQO05H80DsTlCPaOwE2tOCUWGa0HGjRjd2RbjcQfM0SMKcjMlfdEOJpaIPZRCg0QhDJoZcyUASiz8/8BAmQjJeiZcX9UBrYqF/r5uTEFcUKKg5js5ccAAABl5SURBVEyu+jUfIi4p92P70FOglsSfQ4bXhUv0IlgmluK2YZl2/USoWvZuyn6Xp1SUGOIi8nPhWMc49UyHesew1TbQxxMKXQCqM+BaRINuDZifdeCJkxTlswr1jRr1tRr19QbNtRrVlCJ3YSYxtRSzSY2AbkMqWLfPbaS7tdVRaFBVwmVR+p0pWhICaohbLrLlcRBFAsv9sdbxUmuEtlXP4+IVjkFEQvu8ztccGQrXXzTnenh4/CafOHh/hSF43RMGIDqI5DLeMgSjp7AsiQ9FeRnJZAp1KDRzdUXgRmpOSL0Kz5KH2hURRytG2E213nW48JU2KP/CszRJSRV/C1FgzM84tBfmaM/PZRFsVah3KtS7kstBqkRFogL7CXK9OEu5MWN+psPGPTex++QpNFsxH0RpIEuu2UsRbe6LB3kbPc/QjMuy4wLAu4ozqhmwdtWFWBFX+wxOQLAkVT52QnVGAEICHM3wlEOJU+jhbI4dhhPTIrgjCAMQ08Ep5JrioQma7NZ+Z0p2rCHO13vHWSeaYJXwcRjUSiyGyq6uEW86N6JEJ6H45TgD0s6JZ3UiZ7sRpwQjLBxHoB3PEbQUfQF88ZziwvY5HlwjZjo3Ybh1xuTCDnaeP4lmOzVNWkKWD25QjCYHzDGjhExELcXTiEr6rUFHyN6p5QQsKF6N5xaqTvJyAt4N2SsClUjKPbERN6aYI3MJWKRnOGpEAbiDCAOweHeyE3lp0WNYHB4+Z2RdXbxaMVs/Q/iVjpFjnHyhlXyR+gjr0msWD3WEeqcSwtAJUUgWGaHvCAQjjnglYzdhuDWHycVtOEcYX6mKQV2hjxlFHlIl6PU9c54liEqYkxvlwhDavYgrUTQ9N1XNI8FMUraThN43ao5MuE2KJswFc2CISygR+KMGd5TyUSnz9IfflouzAMovqycPWhHBTlI7IRmDJFc86/yDXbYA/ITsJjRIIHoaex/5d/obLZqdBjfeGLkUbT8pHDOV4269A7iW/ItKFH0AFZmFQgy4OvWYdCc7nL14A5vPncbkZalMLe7Gcn+PsJoD6mjFMOnls5WcWFmd8S7stYvgtZkXiAmXlN6DTyRT7xDG1zmOZyVihKLbTYDJdZdxISIqtWvKZeUyA6OUbtB+6/GjyCko3FEcg4K+kGWodo+628mZ7cj5NSEjVKYnyAvdBAJFwqJ2I58MdIi4MPqLD8DG5VYWPhCyEOU2dnKE0aasJm7ErZkbjlGVlURIci0fN5IP14Bbc3AnO4xPT3Ht0pm0XL3vyEJui7i4iwbk+j/THbkgQgTdg+tJItmD4jkVO5odfR/xHlE0+mcYnYTVF3El4lRipQnIGyI4ML+OOlEA7lDCsAwMTa7UMWivNvqyctpWOSzbtr2IePXOOeDMky6N9vP2f1sToZoT6hu1JwyecDXeF0GVnl6nIJYHBk8cqpNzjE5NMducYPJyjaotiw65OJAoDYeoGqcu2z1uwYyJ5V4SPQLMtSVO0PcTLAFbzQ6j6iTHo2bsqoLLNXkzZh9VVxNcw4HbKlWJyrsbx+KIyxAe7ihRwoJ6Rg45QC0tA+ruku9mA+1YFhmaNLYzC2HOaHad1xMQ4Mp5EIZk9PH1Dnd9xWHzTQ3mJyJuuaWi2QHqWS0hviOfM3LkxM+hYTAxiACq1f+XQC9NUG1VWLM7peJlg5IssTM+DAv9FLLjoncBco5BPT45F8MsDiWCTQiWC2KJjZhcY4BFv9OumVICJIrH8U3XX8jk3ahHQHNDOT3tn/R1SKnNBEwefbzc/yMGdyxhKEFp3uayYQ8KIoTcKP/zNvv/4wEhDoyq8xaKBqgYQOKem9anyOVWYnWb7vDKnzdO/oZAqEaeWqBpJUqQG6BDBZoBNKfAZYS8g476FgtdmBUSwmD7WkI0sVhkAxOUc7a4i8ddTbrqs4E2PjPRFTlzTLk8TxgAgFpgsumJQhe9GS0+3YQw3rI5GMRcyfDmZZJ6oQlkaUdyM/NxgjtelNgrh0O+Q5NP2BIvQvo7YWHZ5++jfhGXsAC8u7SeMwqtbkxo18XfIeCQF7MxuFsiMbrpcO6PnQQ5hapSSEQKXUD1jNBsEyZXK4w3K4y2CM1NQrNDkqGozYgCpZ/cEpDqCCLVsmbKXh/I18mELNyE0FgCZAKp2G5rhhCzIQIWZ9UnjK8Bzba4qtfe9ZkrCm7QXAmxLLH9xOIcBojbdHinSxRbPup6BQsrjgF9saJniUD8zzATFOizrdmOSr6oDQAhw3Y301s8cVCCULWq9CPAZ/SpAaCNYkWJiOnzdcGPb3Q49TRw41sqSUeOuOOG2oglbsaKRqV+6bW+j3lbwedAz3E2LEoUDXGweoV+wJL/sq7OYnUVcSMPpgJifQ0g6EtAjNGWWCKqzsRD1CIejG56gjyiJIJV+hYzd3UTeZhWoJKxGGYJjhNBULjjOYYSLLIzD+Zh0B20dJ8q/xhJUV19VmhXF0AnWaKrlsNO70bCOZTMXjke4vZcoZtUGG85nHo6jZpK7PVZv3ObP4DiLEn8Ckr32vtLayYfJ/VD6MrX8NDYZjgk9+p/TcLbEkY3JYGMDZd2DQWiwn7h2/M5dGMZu2bK6VwpcA3HTYRQWHEMBkqcQ0nzr+JEMb/CoongRQdbCi7JcqSysFdCjjpGN6nQjQnO8891mzoI5Qo4damu5sIqd+MKo23GqacYWw9IrL8NisordVOWnarnWKTyfgGHMD6l7cbfG/0vsmHzRCG1YESOI+BAKc5cIVhhLNcUxA/1c5gDoy0pQdfsilLRaaVyAkY7wg5yIxza2tXUfyEoViugW5f2pJitKBwjQuz1EQef8n0/YcUxFEADr4Y0/xJGXbgx0zEMuk17wtKrku2APBVcPXOSRMSw7Mkjsx1eRRLqWCoZ7coEX7vW4dTTLIvIyOFJijgj6izc6TgjaBYsAUFfVCm1Feo5ZO7LPeeknJgYYpOndwvKRq9XqHdFr1BPOZSaA4QTc41P006RiCQWjtAX9qKEmCqDE1qg8iL6HXeiAKwIw1KQKw2DWJCnki/shGGRZMo66rz7cxYR2Mvd4MWK3PNv0M9CdQxmQdYzBybC2tUOJ5+JolBQSCo7bAhZiJuwxG1RP7UPhXM5hxCepQvaiiEWL8RjJR8G2+ccD1u1u9kGxjdYCtZmIpCrfcYmc7w2Ze1jH+JDuPIcQztcwfy4w0qUGADrPm1BJ1BeiIZgFu9A4E64PlngBCJOFyVn4kLHaLYZbkJo1ys0Oy5huRMW2rehSkglZGIhAdZf6UCuxtYDFNLJB/k91y8oTgPcAVfmkZbDyrgmgiEaQIxpMIuR7QK1HyUQRmEKF4/bHJD1LsRRqwFcI1aOehdYf5m92ZUlctLjotxCT2ma6zkCB+OrklUilgSCahzZjjunoLDiGPaAye88lkwchWC2DAtZZ49egN5Cijfbdhg9ApL5BKiCq/IKsW5SJbbxfBcsPVtL7nENrF9pcepplyRSzU2DQzhHYmUsAiYyk+Yi81dzyYwkJd+iyJKIDL7dxEFJzxmxKZgfSXdqw9EQgnIxcAFNxHGyyRhvOYx2nD9mlY5yj42PCAQ/J7Z+bJx3kqqUwIRz/LohCsCKMCwFQ8ll1U8hEReAuDCzSdVnT+FZ4bjjBH+EXP/godkRC4ObUJ9l10rbVQFXh6Qe59rVDqf/zKW7c4YfDeFsWP18kfcsOkZMSfIh+DZioZuso6onUE6EkCaYDeOEkDDGjSLXUbWE8Q1gdDOOTT2LKe21DIDoFDj0SUvaF4EQ3nezo3gL4uNPHg+PxmVhJUosCUUXagcQsVmQUqmq5yaN7H/hfF5mXRVdjHShkQPqHQduRKxIRAYgCd3u+WF4a4fmfBhvOZz/f4zrD9ZoNyJe2l7QP+hi0l1cCVpu0dDALXMsAYY4ZHBKvIocj7YZdC/+kNXXOEng2k2A6RvmeNt3PIl/eO9n8c/+5Mdw8//cg7XLwi2AgGrqok7AixHdOJovQbGgLRheIRm5QWKGqwntRI6PtkWce70RBIUVx3CLYAvaBG14F3di1yzQQC3YiYYsF8llZiFRy2i2ndc3RBm6mnGINCy64nLKOYy2Hc482WF8HTEXg9V3aC6GnAjkXIY+z3hZhqxKZjHrwrZmy5K1BTmxsFYCRiJuccWotms89qU34R88+vfx3Re/Aa518XLshybD8e7mmqBFcamnLumztUaEJLBjoJpJyrvXk+iQw4ow3AaUskHJd8aGluRU/W8X1BJvIXgG5s91ontodhxq43CTiCXZ88kTh6oVP4d6yjjzZAzZ7qfHN88zHESvGpav4ziYhs0u8qzt0Ecb3WmOpWbD2F4o/jojVNMKzfldfODex8UrkTLi6n92IxHTmh0Oz7C6D6soTa0RhHZNOIaNjxyNbM6vFaxEidsEK1qIV51u59mFGRFIjgMIZs+Kk4jBHPJgHFFaluszFAkIMgLB6qAjOSDqqcNdX2qxfVEsFkFcMBGHrHj73TrRD1gWH55AUFQEWndn6zdRzFXpxZHgep7pbJgQFY6VHCBmTC5XaJ45gbc/+tM4/8IM85O1pIf3CkXN+gyIrqCaOnBFmG94/QOLP0M3BkbbpZcA3PMfDr/mw0HAimN4laDcQ7JAS8TBEIIAQX7mxGwWbltUwCv3tjOLZhFxyc9pfsNuXEn+yBdbbDwvBVUXWiu4f9zu9qWMzcGC0CD4GRTbRHpP4BgyTsHO3tE1wsbzjBPPO0lYM+dgeow6EhljN6IQADVfF1fs0U2Hblxh95wgpZ6jIv4I4TmsArOHASvCsA8QPCWdsqYDyoR84qu83fX1CaV2rHgQrwGGIi4XOdxY//566l2EJ8IWnHq2xemnHBrdNTUISW8p7PTBN8H3M1n4xokp0XsYxWLY/cm0abgEyzXkBGW0BZx4jrG26UQ8YIAbCv0KpedG5MvPRQWtGwGT6x2qljHa6jC55jC5ZnUTkqx38tuvX31CCVaixD6BjaDb/ZG3J9GSCRSUbMIZLFjFC8DqEJYlCr02fFAR1+IjQY6xttlhbbPDzoUGN+8lCTVWYmDRtZyDtzoE0UWtGn5RB7dug3vuXl10Nde2KoBrE57dARsvMMbX2cc/mHb8M6tWuLFu7C1GDWG0bQcqdVYLQXJ8tOo8HDSsOIbXANY+/vnhkyWuAYvFhtIi32vhL4oQTeMr4h9xvxa5uxtXcCPxdzj39Q4bL3D0CMxmTcnywNkOb/0eEu6hMvcXZmPCSVBEnlpgcgWYbDqMbrp+QBgQxtaNvAjhrTAyPkIw6ll2vSdWdzJRAFaE4TWDtd/6fD+hy5BJrqQchEzevZSR+ruoYCxAroOIbD0DrCHJzrDaIl6sXXE4cYnRbCE1HSoxKOgMerhn+ofecSM2hPbyMekAmgNrlxnrl11SJUqVl+ogBhLdiXIl3Yhi0R+IInK0Y0LSPUG404kCsBIlXlOwnMP0kbcNcgtA9L7Lg4L2UiYuc2zR+Wj6M5yD84tnDgCipKtJEpusXRO2fOeuCvOTPi2cX+xJdCQBbKM2VYxQkSPro1V05nEXWv9h/UXG+KYQsHrOJtw6WlE0zoRrkoQqBNS7jOnZCpMbXcBPiXa9yysOoQArwnBAMPnEYxKQVeIaSrt8QW8QbllALOy5RdfdCkRnJkbdMupdIWTtmtj1p+dFB9GzHmQRkgnx4Hi+Z141iWAnm4xmR6wGo20XLATBG7EhwylweK7ztSXrKaNbozSHJAu3MN4SJeOKKPRhRRgOEELE5iPGrTrI5RRY3JJT0hDoogoekXaBJabRmEAkv/dWniN/4L0uJVPUeEtyWs5OEqbnJGqzl3ORPUJats/nfpC4BuEIGk9wmm3hWOoZo9nxuSWUUAZTho4ZEk4seCiOKPhpdN48Ga6pCfN1wvlf+73lOn8HwoowHALYHSqIGAN+CCWLQ4kjsMShdM+QM1SSgSjPRmRvL4kgLDu5xl40O4wTL8qFXAs30Y3FgUi4A59ZqhViMLnhUKmOwKVtB8JoTaWh3xT+x6I+EW9xyvIZrCZCFJTorv2WiHcbxV6uQGFFGA4ZJp94DNNH3iYT12RQUo9IAg/u8uTZ+3wxL+ICcuKR3DvkfzFwf8CjZdTeSUsOyI7f7IheYnzD31vJf7JBXsrJuD5Hk4hYligk+gnqWXS4pqCA7CYkVgsXicIK9oYVYTgCkHMQYZpn6zQXL5gAbmQ7DYE+6piTWSlKDlB7chAFGGqH1RfCKBE1JNn6NXRrFepWd3kgZFfOaztkXIKKDomS0efD0P6LY5MqQwntGnDmv63EhduBpQgDET0F4AaADkDLzA8T0XkAHwbwIICnAPwtZr5KRATg3wF4BMA2gL/LzH+w/6i/PkGJRBAxKsioe7CLTHfaaA3gaPPvIOx0QbO5UH9BC8rmLQHEUZ2QKEIrcZ5yDaE7XUvkY8tlQmSJQq6cteKDxlIY86nqHE7++oogvBq4FY7hrzDzZfP/fQB+l5l/noje5///HIAfBvAW//luAP/Jf6/gFsByERqopRBDlzlRvOnOzTUBVfQQ1HuWWuwD1Zr3aic/lltHurUqcWhyIwqcQ9qQNoBIFDj2l73yUvNiqq8HY2Vy3E94NaLEOwF8n//9KwA+CyEM7wTwX5mZAfweEZ0lonuZ+flXg+idDDbuv5coxhV0DFrdyjhIFUUITwQC+8/C2i9SdOZKzpwzyO8BgOnZWmIRth1mTS3ORmOKsRiIxK7oaq3+Hb5MnDV93mkxDAcFyxIGBvC/iYgB/DIzfxDAPbrYmfl5Irror70PwDPm3kv+WEIYiOg9AN4DAGsrHfHSkBCJv/7wsOJxQVXlmC5NS+eZNrpUL7AISkSnBz5TU7PLSTBViCPRZwerjBePjAiSOi/RYpfzFewLLEsYvpeZn/OL/1NE9McLri1JsL2p44nLBwHgNJ2/TYn2zgabVqxYtds4G5VMm3q+MsFeXMOnjY/WgtzUmXARBY7Fghv5Ohc+hJk6jnUiLZ5m8av3Z3wOYe23V8TgIGEpwsDMz/nvl4joIwDeDuBFFRGI6F4AL/nLLwF4wNx+P4Dn9hHnFRSglGZs92++PdjzrRJSXYYTxaU/HlPfR2tBTgASZ6oCUbAJXNUz0TXk9QQxB0RRX8HHs9bj6w32JAxEdAJAxcw3/O+/BuBfAvgYgJ8C8PP++6P+lo8B+MdE9GsQpeO1lX7hcKBkt5/90Nu8nT8lCiVIRIWcAKiSMveotJYQw7EA4tUouoF48PWcN/E4wzIcwz0APiJWSDQAfpWZHyWixwD8OhG9G8DTAH7cX/8JiKnyCYi58u/tO9YruG0YWohBFPFOVqXdPM3oPMAplCRJAtY/uhIFjhMQL0ofdlBIEN0A8LXDxmNJuAvA5T2vOnw4LngCxwfX44InUMb1W5j57mVuPiqej19j5ocPG4llgIgePw64Hhc8geOD63HBE3j1uK4StaxgBSvowYowrGAFK+jBUSEMHzxsBG4BjguuxwVP4PjgelzwBF4lrkdC+biCFazgaMFR4RhWsIIVHCE4dMJARD9ERF8joid8lOZh4vJfiOglIvqyOXaeiD5FRF/33+f8cSKif+/x/iIRvfWAcX2AiD5DRF8loq8Q0c8cRXyJaI2IPk9Ef+Tx/Bf++BuJ6HMezw8T0dgfn/j/T/jzDx4Engbfmoj+kIg+fsTxfIqIvkREXyCix/2x/Xv3zHxoH0hR9D8F8BCAMYA/AvDth4jPXwbwVgBfNsf+DYD3+d/vA/AL/vcjAH4H4t/3DgCfO2Bc7wXwVv/7FIA/AfDtRw1f/7yT/vcIwOf8838dwE/4478E4Kf9738E4Jf8758A8OEDHtf3AvhVAB/3/48qnk8BuCs7tm/v/sA6MtC57wHwSfP//QDef8g4PZgRhq8BuNf/vhficwEAvwzgb5euOyS8Pwrgrx5lfCGpFv8A4ip/GUCTzwMAnwTwPf5346+jA8LvfgC/C+D7AXzcL6Qjh6d/Zokw7Nu7P2xRYihE+yhBEl4OYK/w8gMHz8Z+F2Q3PnL4evb8C5BAu09BuMRNZm4LuAQ8/flrAC4cBJ4APgDgZxFDyS4cUTyBmArh930KA2Af3/1hez4uFaJ9ROFI4E5EJwH8LwD/hJmv03AI9KHhy8wdgO8korMAPgLg2xbgcih4EtGPAHiJmX+fiL5vCVwO+/3veyoEC4fNMRyHEO0XfVg5jlp4ORGNIEThvzPzb/rDRxZfZt6EZPp6B4CzRKQbk8Ul4OnPnwFw5QDQ+14AP0qS3/TXIOLEB44gngDSVAgQYhtSIXicXtW7P2zC8BiAt3jN7xiixPnYIeOUg4aXA/3w8r/jNb7vwAGHl5OwBh8C8FVm/rdHFV8iuttzCiCidQA/COCrAD4D4F0DeCr+7wLwafaC8WsJzPx+Zr6fmR+EzMNPM/NPHjU8AUmFQESn9DckFcKXsZ/v/iCVTwNKlEcgGvU/BfBPDxmX/wFJQTeHUNl3Q+TG3wXwdf993l9LAP6jx/tLAB4+YFz/EoQd/CKAL/jPI0cNXwB/EcAfejy/DOCf++MPAfg8JDz/NwBM/PE1//8Jf/6hQ5gH34dolThyeHqc/sh/vqLrZj/f/crzcQUrWEEPDluUWMEKVnAEYUUYVrCCFfRgRRhWsIIV9GBFGFawghX0YEUYVrCCFfRgRRhWsIIV9GBFGFawghX0YEUYVrCCFfTg/wP0+e0QCHBGMQAAAABJRU5ErkJggg==\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "img_array = np.array(Image.open(\"C:/Users/hrips/OneDrive/Documents/Task/Metal/images/200.png\"))\n",
    "plt.imshow(img_array)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "id": "a5be9ed8",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.image.AxesImage at 0x1a255334ec8>"
      ]
     },
     "execution_count": 25,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "img_array = np.array(Image.open(\"C:/Users/hrips/OneDrive/Documents/Task/Metal/masks/200.png\"))\n",
    "plt.imshow(img_array)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "id": "f79b9c8f",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Total Images:  4083\n"
     ]
    }
   ],
   "source": [
    "def create_df():\n",
    "    name = []\n",
    "    for dirname, _, filenames in os.walk(IMAGE_PATH):\n",
    "        for filename in filenames:\n",
    "             name.append(filename.split('.')[0])\n",
    "    \n",
    "    return pd.DataFrame({'id': name}, index = np.arange(0, len(name)))\n",
    "\n",
    "df = create_df()\n",
    "print('Total Images: ', len(df))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "id": "ae858131",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Train Size   :  2939\n",
      "Val Size     :  735\n",
      "Test Size    :  409\n"
     ]
    }
   ],
   "source": [
    "#split data\n",
    "X_trainval, X_test = train_test_split(df['id'].values, test_size=0.1, random_state=19)\n",
    "X_train, X_val = train_test_split(X_trainval, test_size=0.2, random_state=19)\n",
    "\n",
    "print('Train Size   : ', len(X_train))\n",
    "print('Val Size     : ', len(X_val))\n",
    "print('Test Size    : ', len(X_test))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "id": "5be77495",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Image Size (512, 512)\n",
      "Mask Size (512, 512)\n"
     ]
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAQYAAAEICAYAAAC9P1pMAAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADh0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uMy4xLjEsIGh0dHA6Ly9tYXRwbG90bGliLm9yZy8QZhcZAAAgAElEQVR4nO29e9RlV1Un+pv7fJVUVapCCJAAqUhIiK9mKGJEHDRJBLEl0sJoUbG5Envgzb2tUoWGINjebnXovS0hQJU61CBegw0CojzkYisNpIL38n4KRKBS0KaSkBDIoypVldR39rx/rMeea+251l57n3O+75yq/Rv11dl77fWYe++1fmvOuR6bmBkjRowYIVFttgAjRoxYPozEMGLEiBZGYhgxYkQLIzGMGDGihZEYRowY0cJIDCNGjGhhJIYIRPR3RHTlZssxFET0QiL6h8z1y4no0EbKlAMR/TkR/c5my9EFKScRPZ2IvjQwn6V6/imcksRARF8jomNEdISI7iSi/5uIdgAAMz+bmW8ozONHFi9tPzDzm5j5R905ETERPWFofkR0o83je6Pwd9rwy2cQd6hMf05E60T02I0uGwCY+UPM/B2bUfZG4ZQkBot/y8w7ADwZwA8A+I2NLJyI1jayvBnxZQAvcidE9AgATwXwjY0WhIjOAPCTAO4D8MKNLv9UwalMDAAAZr4NwN8BeCLge8hfcNeJ6H8lopuJ6DARfZGInkxEfwHg2wD8rdU6Xq6piFKrIKLfJKK3E9F/I6L7Afw8EVVE9AoiuoWIvklEbyOiszU5iWg/Ef2kPf7Xtre+wp7/CBF9xh7/PBH9oz2+ySb/rJXzZ0R+VxPRXUR0BxH9h47H9CYAP0NEE3v+swDeAeAhkd9TiOjDRHSvzfMPiOg0e42I6LW2vPuI6HNE9ETlHncS0QeJaB8RUUKWnwRwL4DfBhCYfOIZv9W+r09JTce+j1fa93iP1RS32muXE9EhIvp1IrrbxlWJJ37XRPRYIvprIvoGEX2ViHaLa9ushnMPEX0RphNaepzyxEBE5wO4AsCnlWs/BeA3YXrLMwH8BIBvMvPPAfgXWK2DmV9VWNxzAbwdwFkwjW03gOcBuAzAYwHcA+APE2n3A7jcHl8K4KBN5873xwmY+VJ7+L1Wzrfa80cDeBiA8wC8GMAfEtHDM3LfDuCLAJyJ8iIAb4ziTAH8CoBHAvghAM8E8Iv22o9aGb8d5t5/BsA3ZWKrhbwfwP/LzLs5PVf/SgB/CeAtAL6TiJ4cXX8ugL8CcDaANwN4JxFtEddfCODfALjIyiM1xUdb+c+z5VxPRFmTgYgqAH8L4LM23TMBvJSI/o2N8l9sWRfZclfCf3UqE8M7ieheAP8I06j+TyXOLwB4FTN/nA0OMPP/nKHMDzPzO5m5ZuZjAP43AP+JmQ8x84MwJPT8hJmxHyER/F/i/DIoxJDBCQC/zcwnmPm9AI4A6LKZ3wjgRbahnMXMH5YXmfmTzPwRZl5n5q8B+BMh3wkAOwF8JwBi5puZ+Q6R/LFW/r9i5qRJR0TfBuCHAbyZme+EIZK4oX2Smd/OzCcAvAbAVhizx+EPmPlWZv4WgN+F0X4k/g9mfpCZ9wP4fwD8dO6hwGgAj2Lm32bmh5j5IIDXA3iBvf7TAH6Xmb/FzLcC2NeR31JglezceeN5zPw/OuKcD+CWOZZ5a3T+OADvIKJahE0BnAvgtijuhwF8OxGdC+BJMNrLbxHRIwE8BcBNKMc3mXldnB8FsKMjzd8AuA6mp/+L+CIRfTtMQ7wEwHaYuvVJAGDmDxDRH8BoQ99GRO8A8DJmvt8m/3EYcvrjDhl+DsDNzPwZe/4mANcR0cssEQDiGTNzbVV+6aSU7+B/RtfuYeYHMtc1PA7AY20n4zAB8CF7/FilzKXHqawxlOBWGBVQQ6zqPgDTIAAA1h5/VEeaWwE8m5nPEn9brd8jTMh8FKah7QHweWZ+CMD/B+BXAdzCzHeX3tQQ2PL/DsB/hEIMAP4IwD8DuJiZzwTw6wBIpN/HzN8P4F/BqPDXiLSvB/DfAbzXOhdTeBGAC4no60T0dRgieiSAZ4s457sDq+bvgjGFWtdh/ETy2sOj8uPrGm4F8NXoHe5k5ivs9TuUMpceIzHk8acAXkZE328daE8gosfZa3cCuFDE/TKArUT049am/Q0Ap3fk/8cAftflSUSPIqLnZuLvB/DLaMyGG6NzDbGcs+DXAVxmTYUYOwHcD+AIEX0nDIEAAIjoB4joB+1zeQDAcRjNSOKXAXwJwHuIaFucORH9EAxJPwVGY3oSjMP4zQjNie8non9nzbGXAngQwEfE9V8iol3WyfvrAN6KEL9FRKcR0dMBPAfGX5HDxwDcT0S/Zh2NEyJ6IhE5J+PbALySiB5ORLsAvKQjv6XASAwZMPNfwdihbwZwGMA7YZxagLHxf8N64V/GzPfBONv+FMYMeABA10SWvQDeDeAfiOgwTAX+wUz8/TAN8KbEuYbfBHCDlbPLXs6CmW9n5n9MXH4ZgH8P85xej7DBnWnD7oFRpb8J4NVR3gzgKpge+F1utEDgSgDvYuZ/Yuavuz+YZ/gcakZz3gXj3LwHxvT4d8LMAMy7/AcY5+1BAHJy1ddtutthzJT/nZn/OfNIwMxTAP8Whqi+CuBumDrwMBvlt+w9f9WWq2lbSwcaN2oZcbKAiH4TwBOY+X9JXP8agF/QfEtkJmr9N2betUgZVwWjxjBixIgWFkIMRPRjRPQlIjpARK9YRBkjRoxYHOZuSlhv/JcBPAvGxv44gJ9l5i/OtaARI0YsDIvQGJ4C4AAzH7RDam+BmY02YsSIFcEiJjidh3BCxyHkPe04jU7nrcgNX49YOCqCmXYQa5AyzB3LZQzxeSKsrjFic3EY99zNzPHcGhWLIAZt8UvLXiGiq2CGp7AV2/GDk2ctQJQRMWirm1rhXpN8NV3EIMNSyBBHKysGH3+wS+QRc8L/mL6teNblIojhEMKZXvHMMwAAM18P4HoAOJPOHsdMF4CGBHqlQhk5AGXaQ3Q5yJJAW7eGHEUAHzveU+YR88YiiOHjAC4mosfDTPR5AczElxELRKgJlPBsRyP2ebkfAgY5qikSSZolCK0TF7R1K7A2seaNTXX/kQFljxiKuRMDM68T0S8D+HuYxSR/xsxfmHc5I4ZqBIWoKoCo+QOMn6CubdvWSEIhmi5lg5UwAKhZEAOBztzp8+P7Dve8mRF9sZDVlXYp73sXkfepjm4y6NurKy2XKmAyac6dplBV5m99miEHLesEC7CiNfjLNcCV6tf0JAGA7x9JYhE4lZddrwxoq1w2ME93TMJHMBGj2CfWw/CqMj351IZ1+SH1kwbagAZgyKiuQ4LSkp+50+cxahLzw0gMS4pQMyjxB7RywGAScaYDx3p+HIfbl9ViEyaGkm0AZhTdu82DdpwBTGsAPDowZ8RIDEuGhfoN+oIaGx+A0RaApsGmfAciSRCe0g5c/NhBWRcSm0tbEWB4AbTNaFkjQQzDSAxLgDIyiFtcrtH00RZi+9+q8I4Etogqwux7ZGNGJMghJV7gl4j8C4Evwh1ymbLky7KO0pqb7Lc1ZthIEuUYV1duMhajIczoh6jrxoyQjdKTgi0jW4y4GCgeFIbnnJiM7iHSmDSIktwpSWJEHqPGsAmYjQz6+Bv6zGmQpwysr8P3wISmgTILE6OkbAq4pEUOKRn8sKXQXgpER0VmNKOu9RGNbVvHiVQFGIlhAzFf7aCk0WvDhH3yY0MEHHXB0YSkdPYKAWRFieI7jcGRUe5W/DUyRCLXZsTyuiTbtwE1g4+PBBFjNCU2CPMlhZSHr1XqgDwLs2LxpxaT0AAah0K+zHjYsguBI5PMXIwUKDwIh4NHAKPGsFAsdoShw6RwzjvfVQ51Vmo9dak5UzDXoUSbqRmoEiZMSvSKzHazLTNGT+/IYdQeDEaNYUHYuGHHvvMbSvO0f6qDkMJoOVPCr7NweRTKG5gqwoSJEYe5NBXlfRM+cSgPbd06ahAYiWEhWBwpLIIEovydLS8bJslrFDXaSDyfRsQvMh8Ew2hRpn33c3AzNXNaVdqReqqTw2hKzAkboyEozkFTevuaHxQoNCdkA9HmG2jFy7IgSECSBtvyK8ka7npi6DOeVQlYP0PVziNGkIzMlOp6PZSJogQVqZOpaOtWo3VMKvDhU2t150gMc8BSzVaMUaK9+yFEbhqyNBc6IdJKTcHPKUjkJQmodiMgbBtpYkSjr9IUaC0KmK2jUrkuyqJtW0+p4c2RGGbExvoSNIM6ozV4RC0qOBVdKQOtlpdzOsqobuWlHFYsJiQAE2rKciMRtTBBGN2LqnKOyOz06tzoiL1WVaAztoMfOJrJ5+TBSAwDsfFaQsrzJlth1DKkeS+1AhkhTg6gPdtQOhucaWBU7GCmYYnIGlpikdmoxZkibh8IN9kpNcEqNW3DrRZ1RBOTFifMCafRrK/7c9q2FVhbO+lNi9H5OADLZToUzAlQrys9OyvRYlBlGu0Ws8MSx420JI9IjCa+SOgOK+snmExM5GmtEBfa6eJCqmhvh9Zjyzkpo/jMwMPOTMc/CTASQ08sDymkKnKB/q7NIixZk7A2MYuqbA/MIJDrgfsQQqyABFAyIpgyJ9YXMGSLORLkoI2ouDglsIu06Izt3XFXFKMpUYjlIQQHTW/uGIFw5oRz8nmtgaP0IoFT65VhP8qaNx2ig2GmWkemTTwjSWbnhh/XazPaUtqQHSoCpmLkJNAaaqMNlZAOM4jZ+B1O0olRo8ZQgOUjBYdcI0yNKghnXjxMKNOSbYRbFFII1OpEuUWIE0tiEy1XlkNkCKIWKz2Li0z4Q3wYF+bFjV/Cnp9s8x5GYujA8pJCCooDjWAaVNDA3H+xM5IatX1tLe/oE8n0BpXrfSUBlKZxSW2aWpJcIUgpU46GlLJMzeDIaXEykcNoSiSweELIDS/2QSIP1/Bd3XWNyI3rMwNcA9Wkie/s8Eqo8DEByN9OMUq7cs0kUtLLKJOJ0Bp6aCmTCYCp2elJbjQjCTSpSQnUbFqPF90cnCymxagx9EJPmzabZh6kkCnLawfU2M2+97f2OVUNSQCm0VRRHjkENvos96OxTwGqqnz7tzhdXExgThT4SFiSR44xVxMjMSgItQVtjKsPFkEAubIoqqccHbs4Yg7CljVrNkiROUyb0vg5Puk7Xpm4h6AQpWAfZcAohdOMZD4+++gZZuQkP2uyLeuqmxWjKSGgmw9dFXXJoDVg51+APbZzA7gi8JYK9FAN4tqMElCUUZIXNXU/xR7y+VE7qCVvwpSQBORknUyA6TQ/8UlDPEIRFMGhJpWCnxily2rIYTW/zzlqDJ1wRuQKkIJDSzOnhhCYG1I4fYLqwXXT87n7jAcGkpARCOkZkInnFwRRGD1bpsjApa8KhxnjvJIrLwv9FnUdmmdt725ZPkuIUWNAl6NxhQgBEA3EVkjXA1NlnI1rE+9kowenoofuZIIBdby04XGiLXY1UOcPoLwjMuXn9T4KDuP6vBPpnGhOG2sllp5bWkmH5CmvMazecGQKFFVSbiouVcBaBZy2xa8JoPUpSGyz3gUGRcNzTTG5VKEK0pEg0CDinlhei9IQzPBqrhFrcFqUel/Sf5DKl8EVNeTUGlkJ810lv8MpTwzDkLOrNwOk927en5Cyv0WPm4EkhIAckg3OZSkbTB/NS4sfhbE4cAQoJz2VQhuhKEY82iMhX8iKaZ04hU2J2bdw146zJaLd882pwhDaNrabqOS/HiUixx+KkR5+AGBlcRQANwXar5GQaeO8vFzBgRJJC4q1i0QjizvlqjKLrCYFG7r4oijjaGR0LtlmKHGkfyEUclXMilFjmBu6tAjNQ98338R1VVMQpCClkBpCa3RRVOCEfCzTyt+MeIkTUbjy7EiLH0WT/oCWvT8HFAxdkiOPZKS+73s5cEoSw+K//pRzmrVacY98FVAcx/oU/IYmovXLjkw2qpYobW0gRRShpLl7cQ2eEn7ODidjCQjDJj1VPTQMCYZdaRmTh1Z+SBDL7m84pYiBtp6+IFLIaQhD4qXiF1TetYn5a+VNlhvMHH8maWKw/Uc+ntaZ5xp+YF5ot1FEBqLxtByRkW+jVRQ3an1zY2VwZpec7+Gz7XBCModaWOdISmhWLCtBnFLEsDiktIC4NSTU5lnKCdp+lXCmJRyG/nP34oImHllSsLdDTvvwyeepwudsFFIPW2mrzAhFKks3utDKt2PkpjXdXMixwjhlnI8bMyyZ8sDHzrNSx2OHNhFkQ80WZrKhK7W60RrITG6ytjSRCSc783D3dbeB7V6MJMSmusbeq3e1NISs1tB5n5r2kHFYpm7PZeWHYjv8FPL5uU/bxeLkbsld8zMgNWHi40iMJdxotpMYiOjPADwHwF3M/EQbdjaAtwK4AMDXAPw0M99DRARgL4ArABwF8PPM/KnFiF6OhhRkTShtnH2gjTqkzIdcQ4grU1zB4qRk1jq4VZFBnkr2DE8CvvEIR9vua28DQGDb85JfMGTy46rCS153u5k3tT7Fvmt26Y+hyG2gEUBihKBIM2GY6dJUnsQXbdPF5TNnxHJrJpwKlZLd/Sr1ggHavh18dHk2mi0xJf4cwI9FYa8A8H5mvhjA++05ADwbwMX27yoAfzQfMYcj1BRiB2Fn6o7zHGQl0EwI2XpSJKKV5/V50VuJpdJxeaptb7QEYxK4eBGpMINO1KimNWidQTWD6tr/VVOraVRV25Qofkzas8ll0JBXZ75+CDLznlVrRfhYSmViRIQi36f77TIhGXTmzrSsG4xOYmDmmwB8Kwp+LoAb7PENAJ4nwt/IBh8BcBYRPWZewm48Wp4olDqW9PRa5ZCNI2VmZLQbgnA2RnknRXXOAaE1yEZUVXY7d7bqMTei1jbM9ZTWBi8ih67GRjJiqkEXankE01hzIxSajOpirAI/Q9ZsyWkR4nTIXpYLwlDn47nMfAcA2N9zbPh5AG4V8Q7ZsBaI6Coi+gQRfeIEFrP6bPHDkn2uOcR+CK1LT2kKURYEY0IEUTKkQO0T4tqMUEQaA7sFV24FYTAJiINb2HPtIZFQ+YO8nDHePe92qQSU0ISivCrq39jcbNHWK8kU5kcmZOHuN6URxmTPxr9x1nLsPj3vUYmifgIAmPl6Zr6EmS/Zgvk34GGk0MdU6IuYDID2o9HCMxXbbayi1EO3tqE1IUnrDF2PV9egms0erWjIgIlsWJTQaRQar+WIaeaOsSQDof2k4qey0Ugg2x90yVOoMbjid5zRkd/iMZQY7nQmgv29y4YfAnC+iLcLwO3DxRuG4ZpCX2/V0DLi8x6+Dyee++CLTOY5oH0PTIIk/OXGpPB5TKdNw5iQXVqMUNHRtAGKfuNwf9qhlvs0XSpBwXtytn9fIqoI6tyFVLcXkG5KKyhgT//VKwJt29ZH4rljKDG8G8CV9vhKAO8S4S8ig6cCuM+ZHKuDRdt5KYeU5ozMYMukSSayikkhGD4kSxBEZj+Gyo48eIdb7b+6xBUwPX0N9RmGZKmGWYcwndohPammAHtedQitBVZFJjeph2X83OXU47b/pAik7NXAaXMiGJnQSL7L99TkA6Ax7TYRncRARH8J4MMAvoOIDhHRiwH8VwDPIqKvAHiWPQeA9wI4COAAgNcD+MWFSJ2Td1OWUfd5ibJBaRUnoz24etdaGBUeN0YEB2GqT68CeELgSQXevhW8Yzv2vXwXeK0CrwHrT3oQh59IWN8+8Tm1P1ir2PybW68bDNUaWvlYpsuSnXwIGhm4v/TD8WYdbVZdNuicx8DMP5u49EwlLgP4pVmFGorNe5Az17pMPpYN5LDiZJIlhVbWFF5jNhORGGGDmZ5e4fDjTeTr/vO34erf/hccvX8bvv+8r+KTD12As45MEjI25sHgSU4pn4T6UZoU3M3GhEpG00l9EFf6PrxCp2huSUXONniKw7RCOiC/GA6zyctmrMQ8aWY+njwbrgDtVi8ciKclXlmq3ikOP7LLrp29zyCgAqp1xul3V3jGkz6HtfMZv/sH34szbmOc9V/WcRbWwVvMAiWSPgmXvfPm1wyOPJXqcu2s0CnkvJepVisamlf39WjBr7Yc22sFSjmeTDpkUUcoxOWajR4vTJnNIIdxrcTSQ6iv2jBaj2zkr/c1iPxpvcbp99X44Ke/B3976/egetBoFHtfdh7o+AnQ1IxawP0J7Zi0dmCR3P0pli0boYv5mtL0aAlfhkuiZa/6GRJi5EhHlS/vZ9hsW2wkhqWHqECxCeFQWoc0B6XshGtG9dAU2+58EA//5BGc+aUHMHnQzHTce82uhgy8WJYVnHNvEXXa59VlnmSZKez9Sy0d7UveqZvLEoPmW8g8JIa6Uc5G4qQghpPLjIhhK4hbC5G43IXWvIbmAojNdGcf5D5PV9egE3b4clJh39XnNTUm9pfKGYbsorD/7e13aEUn8RPfdMrhp0RJnauES1Fj9zfWhmv3pGXWwzPrp6ojKHujl2evPDGc3KQAU0FaayGGZMPqOfkeH6FpXJH5mrNr8BUBdW3IQfo8BPa8qpn0GuSvouM+tHYU/CqNrUUakU8i1hpKSFWbHq3BTS/vdI525MUI34cUZQPJYaWJ4aQnBVfRvAkRDDH0Utml4y8khTgzbnYlWpv4OQt+bL1m7HuZJYfYx9YxA7ApL7rFvODd5pMygKA6AAlmLobrkUv8oPFmublb9L18/HLkTVB0rpCF/M5oXMQGkcNKE8PJC1EhqrgFzJpzyjmHwJno9msAEHrqbePfd40gh97y9dR6sn5HhWyyGXHoK4mUinaShKeyZWHEvXzKnCjwNaibv2wsRmJYKsS6I4XTngeaESbnfFp1daRvNHaIUi6mYkEOXrQS+Th7mkyS1Oi18pUGKOMr/pAkNM2sd4PVtISEytdXo1oQVpYYVteMyNkAouI4VZ5kpZIOuA0QiRGQgRnadL2lIIfkp94WCOlvoDhQsxFkj872GxSJPIPz2AGpFCHDgrkM7rfUoeHyEUSsYCPMiZUkhtUlBSBtNEeQaxj6OhRKio/FSJm8olwKemQbmWGGMheh9mrypZA0aWK/jNMwEo4/NW/l3hRrINj4RnU0av4HBb4fSMddNDmsJDEsL+bROGxlUL4JMdciNLBo/Gqny0oDNBH3/tqudvy+yDkZu3wBrYziHhuNBqTlk9KYOiPloigO0ICBMzcjN9HZBKwcMSy3tjDcB2BgK4L/HuI88hRw7aUwy5aGIIfSRIZkd3Xa+/Lz/ZBlax+I+LgppC1jKq5s7yV5qQIopkHXM/EflIl9BLHMLEytFGKmS8jsF1OlH9AitYaVI4bVQh+6FxXKawszkoJmNmR6yuyEJFfhgxEKcdlvGBvm2dnTl9yi1pZSj9b5BXIZqzMaSwQQ8RN+B24JG2su8mXETskc4niLVSVGYlgoSl6k7Y18Xab8EOWQOlQIdT2DopG3HHnuFpix9+Xn+w412BwmJbzWNjSCGSR/T6ffULRMrZwJEfsNFD+C1MwCZ+bG2RUrRQzLbUZ0IaH7ykoF2I1dM57srroRbEqSYgjZsHtA1k3Xa3rz3dyL30naNZKqYwGVJsfA+l+8cjP+KE9JefHU6NiPmCxa83cUsLbUzEgGhqSzKHNiZYhhtUkBSNY+9v+JHYp7VBw1LEEsJA56Wjlt9ZnDa4h67FhjTkFrJ3Nw1XRmQpGsRXkm4NtrzpbKaRGpTKGQQjuvRZDDyhBDP8ypC+pVXmdXrieTrWfSMRKR6uV849IcCf3uvfhzc65MS2SNX8Kk3/3qQ+Z63w/MOsztlc3ReetB7UedLCaOUKiqBCSvqSajj2EA5t0FlZSX8eqpoPDlt76CJNTOosoXk4HCHGJxjlxp6X47V0FKrUHMgGzfPgE1Y891Zlt5vxCrRIuemymt2fH2mQa9cGl2yvOUan786ny8Lr9H3ivLckJZmHlBHsOxEsSwmmZE7mUpqnxKW5ARB9UJnSQ1AlC1BTV/an5acxvCtHuuO9RW7Z2TdbPgRHELqnojZYZoDkMXrpFBh3PUX441hrYPad7mxEoQw+qhtNLbN++2RQtqWYFHPTAf0ORVWHxqKbZeTlhMzuYFgH0vb74i0Np0pMQHUIpUNr5BJSLPukFsTG4trS5W/WONjpU4CagaiNQgCv1SPbD0xHByaAtKw5BR/EhEIn6yiKhyzNAJB7tI58TXTGY56SmIa/Lcc+0h7H71bYp8M1bm4uQJ/46fuFSKjobsn4nWUDU/gTvWZSBlMlk6L5qr1rD0xGCwiWrnXKBVJBvWZyRChVTrtWz6PjsOflrHPltnr6fUatg5Dex9E21yiO57yGvOpWldk2YZQZ2klMqn8HuW4Qd1XGNOEQUnrsn0rmz5ghlh3lH8OWBFiGH+zpXNg1Nh7QttraCL7VP9tNXTxD7LIE1aGwkckLH6LeteXP80ywft+f17r7EmRc1KDy0yWhj3a7Z5T8xU/WQjjl+S5otQyszuJSmP5/cQl5oYVtOMSEFR94ny31ZMwRKLbrvH8aLyO/PNRI/P3UdjxYYuJEY+GrPDNvz1Grtfcxt2v0p8BFdU6OKh0ljWosen9Kzia93ZJCUWh2zbrYasvZiUFtEWMzyh6C8knnm1maUmhtVESeV2FTL1+KMKk6u4KTIgJbCrgkfaQee27zGETSyHQAMfxJTVEZi4nOKyu0ihpdYHgcg+EM2fkirb+3y0yCnVP84kgViTazmDVDVxJozEMFcU2IvSa90aiRDXtPcbvXuK9xTQnNYyQQbaiETXnIZg81Nvh7PgIXOwz62fcBpGzcbfAJsmsF4oSNshdPjbGbEH+iZpPfccQRQWVPyNiRTpDMfSEsNqmhGJRu5+5XubVOk9F3JZybqldUTJeqc4FX2UdCP0n7Jr5WevUXR/vucMy9p3jd2vgbkhh+tuR7C3pJe0gBz6dLxaj9tqwCX5KY1PaYvceiEplpcvNOMDaWl/seMnzGceoxNLSwyrj6ilBo2G8qQQ27aa5tjVESUzlcnKvjMpPzEXOxdJKzxn/wejGIw9192GPdce8nkWf4Mi5cfrRM6O70hHvRJEhcSNWcqTuKbuuq1pmXHFGCJjiJEYFt1oYQUAACAASURBVIbMy5lUiUrmjftuh1hpWRnNtXP6cw6B30tUxsDMkWHcdszZ7egBYPdrbsOe6w6Vf4FJcxN0ag3RSbzHxCzwWpzmzJQCxqyZY1EXmtI4ZPxCv0khlpIYVtOM6IBUsasK2ZfXRQqtTjq+0KGaliDrk2trDqHJZH0Jsckjd5p2S5/JhNN6DbAhiGJy6GGuh4mihL3aUayq9Sk3l0+p2tOlRZjfWc2JpSSG1UFpjYrYPKeSlviZOi+k1NYBENnGGkbLP+BcC74n5uZe7a7MXBEwoXbNY5gvaa/X2PPq27D7uoYgkqMTJR1pV8I+/Nn1pWz1eqySlKg3qSip9yn9FYV5d2Akht5QeoyuRkjioEtbyKUvet/U/MR/ST9bufbSfAGz6S09QTgyAMyXq5yNXHMTPiHUWybgqkr7TWqjPex59W3Yfe0htBygBbffq230eh05Qk8xlYyk3XCm/kj3hKp1aC/3JDQllt+M0CpGqtuRL9yZEZkWmusQ4uxIXhQRchyV7HgH9DBs0wV+NWNC+PwqSw7TKcB22NISCDt+rGw6zSkonG+9ZNS424crD2EePgYA+S9ep5wi0lkD5TgFmU7XSGZpS53EQETnE9EHiehmIvoCEe2x4WcT0fuI6Cv29+E2nIhoHxEdIKLPEdGTB0u38oheWJ/5+TGSyeZgLgyBV0waXwK7eQpTs70bn75mNIcT6z5NdaL206N5YtnB80LEenYPSaej7L5Wzpi0SLWdFoEmEnV8bzOdsZYl6x+y6Wz8MWmQOOvyKaRkm82ULNEY1gFczczfBeCpAH6JiL4bwCsAvJ+ZLwbwfnsOAM8GcLH9uwrAHw2WLolNagxDIN/hpMoQQ8c9yc6B4gtzRM/s3JAnMZsNWSaVaRwV4cEz13DkgtNxz1MehgcffQZ4UmHvy3fh96/eBZqyMSvWhBniycHc6N5rzveTuPZceytQEV7ymttDv0bOdEjeizS3evgv4sVunIlbJkhHganrGpm48Pk4nzuJgZnvYOZP2ePDAG4GcB6A5wK4wUa7AcDz7PFzAbyRDT4C4CwiesxgCXWpZki7kaQSVeChpBAj+c6TToTZ+SOh4Tr73y/GWpvggcftwD1PPA3TbQSeMB542kW490cvwv3PvBi8VoEJ2PeyXaATta3j7fv3i69gSQEwzsmasefVh9pOz5Tfr+ueel+b1RTRevWSFxRf58w16XsYhl4+BiK6AMD3AfgogHOZ+Q7AkAeAc2y08wDcKpIdsmFLgjn3sKWgqt97Sg0yJFXkRv1udxoD7fM43NXhwLy12oKd0Xj8kVtw3nfciR947FcxPQ2oJ8DkGKFaB6anV7j/GU/A4cufAADYd/V5qNY11btRoVtDl7bcPa8+hN3XumnV5bc3VwwqN9egcwQRF5Zr+LOPThQTAxHtAPDXAF7KzPfnoiphLQmJ6Coi+gQRfeIEHjRhS+947AMSZgQrXzNSkDQXENaZpN2s9UBzJkJJTFYLYiJwReC1CaoTwL/ceg4+fvvjwWedALZPgUc+hBO7TuD42QSaMqppjSNPvxBHLr0Q+152Hn7/Vx4T5l03y7fdl62C+2QGahNfDmsGjyInf9b/sEiUFBhpQkloFUL2HqY+DG1TRcRARFtgSOFNzPw3NvhOZyLY37ts+CEA54vkuwDcHufJzNcz8yXMfMkWnEyEkEDJZixd2l+RLZvrkVxovxahxfeDls4/UBly2HJ0ijNuqzE5RrjorLux9fQTqG4/DfTNLf7+6MQUWDctm+0Wa/t+9TzsvXqXkXlSYc+1txpSSD6yhiBapsVQE6IzXZSw2MfgIqV8A6XQGC3ON+5ZhrFfyagEAXgDgJuZ+TXi0rsBXGmPrwTwLhH+Ijs68VQA9zmT45SDr0fUz7ekaQvF6FIvU57uXI4F8dm2+wenqB6a4rT7gQO3PRrrX9+G0w7XOOOOdZxxxxRMAFcViM3oxJGnX2jrNoO4NoutpD+mk08FOQzx1ciDocpVUbp4FEJqdn3kjh9ITAqxk3LYTZVoDE8D8HMAnkFEn7F/VwD4rwCeRURfAfAsew4A7wVwEMABAK8H8IuDJFt1EOBfTJVj7w7S6FYAMgnaKnbvxoMwTXafBAZoWoPWa2z9+jE87GN3Y8ctD+Csv/8y6KEpqqPrmDxUg9eM2eHq+OHLLmqIoGbsu/o8MUKh3FbL6c5+u/rkdOrssyt8Jl0b7Ra3wZRmV6BVBnFjEogx3FZa64rAzP+YKeGZSnwG8EuDJTopED2uYM+++WQZdEDqV5Y1dVP0/j06qnjmYXIptlO17S+ddpqZs+C+K0EApgxy6yWoqeRcEaiGCWNg79W7sMd9tCbuXOWjlPdRm41n916zq71XhexUh6LrFfYyKbSEfbSHVI/RVwPRsTQzHzfX8ThnL1Tw7gmDJjXFPWLLl8hmvkBdm+8jTGvx1ScpQEJb6clTranQQZZNbx/Y4WsToCKQneJsxOYmnpjwdeSyi8xohT0nMPZes6std9AGBMlxc3HPtRnNIQ7eEAek7N1jja6PSVGqFUiVapgDcmmIYXMx1BtVkF3rC1OFRfhw25CmU/Nr7XFDBlPzy5YkptMoA73XbH1jUjtWwIIewmJMQyc368812MkEqCocufSiJn+GiVezaMw2fkU4/IyLcfhyE79Zuh1pQ57zKCQipz3BkkPJ++ujQQzWFlK2fsqk6IMUWQzRQhqMxJDFLHqnxdAp0BIVNcuUGVY7qNvitcTVy5YLoAqi6+miDpB8Q3eRqfnlKD3b+AgbNZP5AxEOWzLxOz/lhHO7Qom8QHZiVOczSgaWY9ArTvibVI0MaDfwVHrtpfa/v5EY5o6IqbV31YqrQKjGxk6vQ1U99a771oEelTrVAwebsgABKRy+7KKoOA58AtLiIQZo2vgkHDkYk4IT9V5qPxw+t+Kvcs1I3imFIJu/NPVKe/eUdqGZKtI50x8jMSwaMy2aItNI1q3JsD4F1u2CJO/AcwlsBZjqMwkD5Cpy4lqwDZzkALkhSwRPCoTAGegmL4Vfv3LmUZhZQw7nN9OkY2enBmfVdD3/mNQ60RFXLa/LVE3n2Sbj3Dkrv8Pq30gMi8aspoQ2qaZmkW/0y6kPtXIzQpDTYpI+Ow5+A/+WrIB+WbVStaRJ4fwNkZN054duwc6bblHcI8a7se+aXeYLV4RmGNj5cVpfpDa+hrj8MOOsShfJr/hsQvHzZKVnOmN8+SJI+RumMXQOV24IWl9jWlUonv9ZoVY0BqhqaN1tiuKdk0AzFGjP67pRtbdM8oSVtXCaUYNAbRcVkKvKTFwSaYgSmoWszyKrnTfdkpSHarPyEiSnTHf3vt5ROq+hCKmt+/vok3csa0+nTzK/NiHQ1tOBB8pzWg5iWAkMYN+uSpK67IrRHIzuerwpCFGz6cl0ajLnRPoTUzOUKE0Rpb0480GSgToL0o0OEHD4hy+GXMvQNn3Z7NEQmRdgYMeHDuafhxDSmTCNc9LIu+faW4WPA2IIF+1nkd1YRZMjozEMQpyB8hKCKH0L1MyMMozEUIycnZiw6bKVrsDJ1LJ/EzI4ApHDdxnb3++q5L6yLXs+KT7CiUxtP4M8NpOUcmLGqrfPm4Ad0nxI1X9tqbX0dxB7oij7NsWCNNXiNtjlr9Cixi8q1XPEx/3udSSGmVDA4L01y6iHC3rLKFO/G7OtKK6H9CpzBs6puSaqgG/XbU3Bhbdhyj182UVGxCmr9yxJhbg2WsPEFSjmQGjf40wQl0YOTbKMpjDUKdfKRxTuXT0FJk1nIaFszb3kOokkmw6SZXQ+BujhiALQWcFm6ZFikoidjH7XIyuz/8Asmt8ta+IbFjapPHekIu1kpHtZ3YxoRg5aj4MQBKQmVu3cf0tTvvgoLtv5DG4FZvbVdD1qigotTdcVV2uzgzIquZZD0uYclNuoMQQY8hBTqlrPF5xTv6my/gJhO0vSqUjUAWFCMBsimFTGr+AmSrnJRQy7DdskJAeh5qtk4ElLfF9C68DkjXkyEnLXjB0fOmi0CdFwGYQjl12Yr9ctAsrAyxepHX19DNm8ZVBXPeqymdzllGlWqDapZZZhJIbZPUgWMUEMqXQiD4Kd7WiPUYVzFFpZi8bu8lmfNnG3bDHHtRjOdBqEkme+cltSIMKRyy5qmy2RBhKkJMKODx00y65tgVxRsE4inqjUWhA1BJp51XcEocucd+W0C07nmU4YxYvNiNi3lUP/Oj6aEjOTQlfXlXnhwXtVNI9grgI1H8KN5wi0ehZnEnETPp2aA++ohNFEIvn83o0QeUo13383wvbq8vnF5nXKJaE19GCEpCnH38pcNOwokz5Dl9LMSWSnl1lavxLPxF+KC+tkpAEyNBiJYSbETMzi/RQ6oYreWaRJyDKyW9KHarshBEE2FbXqU2siky9fNiLCkUsvDBOWqPwpsLEwGl5zGpPuk5gdjnR6sI1GZi1ekYRcmndsFiQF6Ehfqn2UYTQlZkLK/p4xH2Yz9Vn27PEUaIbxOwCCHHL2qs23sj4Hl7cb548nmak9rTGPOGn7irRSHJH1jv0HxTX2l8l9a4IAAoEnppERJ6Z4l7a9VrspJOxWeTmHR1Re3Sf/jnfmy24FdqefASukMczOgotHTxlT0YP9C9DsuRBMCOJwUVXn7kJRXEcEqZWaqWxAwazGYth2deTSC0Mzxf7uvPGA2cTFiehUd+cIjXlvcHXg4Gc4xD2osvRhLXdcelNdws9OFitEDIthxvlCsbdzfodeZgSHNqfrwYIlxx0Vq2V2JMgkKVci/9jq6MjvyKUXmnkMMoIlhx03HvCkRwygcr6MLtk6rklQ9FuClinB7UNHYsXCaOaalE04Y9U0BSbkQKwQMawqEhUk51SrqpBYKjKbngRtWskgu/5B8SkQzOzHtUmiLiWcbTmzuLDjO3LphZH5Ys2KmrHjRrtOwq643HHTQbTmQGiafVF7iB9ACVghhoTpRkh8pq4/Ok22BXaWS0wMQ17gEiCrKXQmgndgTibGF+AarmxEhMZ0kEk7F6P16iLbx1pnlcsydc026sNPvwi2m4V0rBLX2PnBA0aL2H8LqLZrNCjOBD3aRtTD9xkCdfLlMLiKUvRXmiaHk9qUkMbbMpgRhS9tiB2sag3UDE8m0wnTQNMIUnLl4J2CrqJG6m2X9lpgSqjpgnim3J03fgWHL3tCM/MxsJj6tsQ5dy5xdoy8xpZEyoko/SoZMyYp0GxYYmIAloMQHHrK0vs9iQQ1NxuznFgHHlpvNmtxcb3WEBHC2lpjivSQQV8HEfkuGNbxqDVU5TgRlt5UFhE5ADv3HwAIfh/IhPC6aQF7XzFZz2PGYw6DJmTp8siviXcUOqDMNJacGFYZOXs/k4wRbeVmZyo6sogrueyp3QjDpDIE4cwRZ2JkbF99oY4rJNHq5KWcvR9d8xvKemddiznCU2bs2H8Qhy+7qLG742eYIML2fAwMbLgq6+hWQO/slQfo32t8bWPM6pEYlgWBeqz1EraCTKONEmPUtZkD4aZDV9ZPIUc1UmDtROvZEyTRSq404kj09tRree+hZrLjpoMtTcN9N9MttmrtYh1kT/A7aveBSiTUyOr+BmshGTVLrQfu+uJIYiSGRWHoOwv2UVC6YT/nQOsebWVxzjW3R2TNZq1EccXtcoYp4apvITMiA9PQQxs6YV4Q7Be1I42JrVnC7PNKbibj8nTmWJ1bdzIAzrRTfQKlUO4/ua+GZyMReX5EscTEsDEq06ZgVnPQzV1gpSJ534MLc2QSe/VLoVS6eKfq2KSJogfh4lq7ASc0pkCdFr20yDcmhM75BPGmLzlohBo8bxncmZlyHjfwRJlqPsrDmAOWmBiWyfHYAyWNr6iB5nvbYA9Hf430P0kO2eJyfoZCdCkZth20N3y1Fbz0tXeIlF4ubhPPa59R2S5b7hhN61JMpzg8ML0y5LlAc2JcKzFvzJ+8LWwFmEzaGoNcnl3ZjWGdPe2TCzU37o16yLrjQwebKdFdTlTXUARCQuAwLswaiSOXP8GE1zV27D8o9rUVmcn8fXpFhjiBnw4+hwbVarMU7jGpRioBNe9Lfo0L2nFhfj3lGIlh3uj/DsK0roFz0G3Y68IhJXt4OSVa2rrBZrGtgwiydUU3IO9pHoTn82syZqrM3g5Ss5Bf3/KBCZGzhYnfrqHKovdHahyGW/Q15GFFMpHdCi96Tv3zHRJ/JIb5wzfYgS2IyH7zUZCB0w6qSmy0IvKva9sTVg0hlHaIqgmR1igIjJ033dLs8eiia0ndLbDdPzJyKDpZD192EXitapGXcThqzzHPCMGmtTK+J97Mwyl+bbmIXeyiXRf3ZFewUh2PQM3Cyv00pOXwMbTUr1UHDeeGwBvNwnHYVJpWnXd/tUhTSxs07jXj8rrs4DZ23BRt9R5r54HW35TLkRZz+NKLvL1PNUBTBk1ru2bigCKrOEjU9Xh360CmeUxsSrVRQqi9qRG8IFE46adzahp8/Hiv+MtBDCclBr5RRvhpe/e3PjV/lbCRYxvak4KyTDuueIGYqV454RADWt+ECH5l/KgNyGXXhy+9KJyA5cmQAUsOjdqBRusoaDQUMOaMZFC0gCom3dQzjY9z2svmdZhLSgyL8bRuKHLvNHWtpdZzeO7G8d3iqnjIzY1SuOO6znZMvdpMS0ths7uzy1PTjMWx/B7FkcsuwuEfvrgxH2o2H7P1QrovUYWZJqdRZ+WNHCRDqlZXA5XPIBhajTS1vKBNVLc+po7rgiysBMOdrEtKDBvJlEtIQsnbj3r9SdWtGnd1TCWPutVLc/gr12so+WXH9m2d37n/QLN7dQk0LcUHJW6WjTS9EM8XyYBaDTinSrljzyhNmI+ikUwfG3WoPVtADES0lYg+RkSfJaIvENFv2fDHE9FHiegrRPRWIjrNhp9uzw/Y6xcMkmzDsGT+Dd/rAKrOLCupW4GZQmpdgcuy2KMfJ2ywc7/8xmTCXIFiBriNWPzQq8lr5/5bok/VJR0JHSbFHEwIQN/dSrUkpGkXR9DUtJyPITZLZr2P/ulLNIYHATyDmb8XwJMA/BgRPRXA7wF4LTNfDOAeAC+28V8M4B5mfgKA19p4pxBmIJpA81Rqn7ZxhyOHPu++JWKs8oqKmspX9mrODJCzKwMXhGLyuIVigU1uSMJ9lzItfk4jkBpKgR3fBW0vzaQpKDSpVpkxkyV6c7veo+376UN0cbz+dbKTGNjgiD3dYv8YwDMAvN2G3wDgefb4ufYc9vozibpdwXz8wR5iLxJzNC2GOo9alQKNuh40KCtrVdkdngpkb7UVzSGWAyHu4XbedIuRy9nGiWzc8KN3XNr7TGsIklBYafhdosdqyxy1w1anT5n8Uyq98r5svuEchkTcpGChgEPaVpGPgYgmRPQZAHcBeB+AWwDcy8zrNsohAOfZ4/MA3AoA9vp9AB6h5HkVEX2CiD5xAstCCsBcTYvSvRg7QWLDFo7MDZt90qxQjf5EHMVZVfQ4GDv32/0apY3MpoJrC5tIlNlsLqtoRLEwmlrfhT6Pv9XZKgXK71H4dhgTg3YTcZhCGPKjO0GaEn+B9Fe4vwU6H5l5ysxPArALwFMAfJcWTUiXuibzvJ6ZL2HmS7bg9FJ5VweE5lsOWoVIvWPfNgl+0pJs8MEeC1EmFZnvVbo0FDV0V08ymqwqj4q2DU0l81GUKEZj6KrAwlGXe3aJMoo3Z+HoJPX1Km/u2d9eKysTWhEg9s6Im1NJAy8hozL0GpVg5nsB3AjgqQDOIiI3c3IXgNvt8SEA5wOAvf4wAN8aJN3KQtjcKeTes/vaVNx7+IVRmbREwFqlj1i0uCSjKRQjrLg7brolKLflD9C0layWHTNYRs4uspuHlSj4Sd46q6aERgCKHyfIO9YYHIZqsgvSGIjoUUR0lj3eBuBHANwM4IMAnm+jXQngXfb43fYc9voHmDdxpsZmwWuAiVuPNUUP0e3FmmmwoEao3d4WzfREQXCpupCQuSSOxjWJDi0Y5pPpgl5ZICa4rro/tPa1HhM17yGegUqtA6Vgqa6liiQlSpdJmnsAw26+ZK3EYwDcQEQTGCJ5GzO/h4i+COAtRPQ7AD4N4A02/hsA/AURHYDRFF5QKgwffxC09WQxK2zlmdYAJbZnT5mA0rvtt3m0YUQNnauVR5z7oA4tQyKnjsdlsZVLJCJxuYlm1y7EMiSNzqYhcEU49pJzbZtibPv9b9hHG6v4scya47HHc/BpEjKymDym9vKxQKlrQiCrDVJriXwXoXAUT1wZ6NTvJAZm/hyA71PCD8L4G+Lw4wB+apA0JxW6WliCFVK9ofcPROnilZUpMWSxOUQqsi5bptL7HlWJFuevNhATgYlwbM85NqghyGO//CjfgLbtuyvI15MPI8gredOZS2kIzcbno2kJHWUnBXJJ+qaTxFHE7lmMqysXCWlO5HYB0sJZRhC9bc1GYyjp+Zwq3qd+5nxdQcdEAEWEYO9xx4cONh+9JfjVlcEOS3Z+wM798YYthKMvtRpCjUSvbbShY7vPabWBbXvvCk2TWG7tflKIN2LV2n+rDZY87IRAPi/u2aY19p0NSzoleh6Yh6dplvxFzyK3fVeihEGE1g5MzhlZZaZA+29ZSvlsJS0aRlMaUhylpQUoLU2xteMVlUyEHTcdNN+rFLIxWVIAwu9yKrIQ2zi1NLfQaBkxhqyqbBVPikwEVGTnZri/WM0TcVsFiHrils3X8fvqw+rzqfcnscYwO2vOnr9QJ1XLIQxsbHHXXUbmgfwYrQwD0j1rzdBHMgbp0U1SJ37KXo7zJsCPUBAQz2s4uuccc2+thtEPgXlR0khyjyG1OrWVB4ll7nFDToVHedpOgFprM3IC5vwYHTJ3YOk0hmHOkllZcoHaBSdPFCmk/UrtduZ6Ufmla+0L1q4YTwoJ/4MoOemjkFGS4RT0yNmvYVckdocWeVSV4cKYFBKdIMdxWhcK32myLSnqudZGHeG1vv8QZ5iz0Ux4M+TZpeFlX4b/7bsHg8RJojHMqh10pZ/dmeM/BKMV3dI8bSP3U4c5vBar2TJNJTJzx6q2kAnSNN648SUbLDXyKPb5Trn5Ss506GjXBOPmkBOOtr3O+Re6EmauJ30FmhlhwwloOYUDJ6TLL3FThGZ1amv9SFZIkUEcPltnd5IQw6JRQgiZGud6+5TTUOsg/JZuLowR7gMJQRwwRndNQDUR8uRuQZFVNXeia1wQF1Fcko3H+BKO7T4HmNh7rG2Gcf3O1P/gUVjLK9C4Wh2+1Z6KuT264dTrlZvMJL2cAFqNNtaMqB2tmMUU7WZGLCUxtOczxHbsov0HQ5Bjdytz6bRcsNiBibt7EKnOdzXYVD6lj9Q9/pLbiPwhO/YfxNE95yrTfjvkCbIRJo/93bb3rsTekDJZzx60lZ1lHy1fBtIO3rjuuusRs6m+oj71PGwXsy5KXEpiaGPow1oW2IqQm+wk4zEAjlVK0SAqapY6E8z6CUBU/tKuPApOiZSD0oB37D9odny2lXznTbeAQTj20nPAVWUn8ETpY41EEzDQ1Bnb9n2jWaKde6aekGdTrzWRGoehDMzVV0XYykzk8pqhmq6kc5xfp7kixJBDycPYDC0jKtOp+y2tQTR6px+3ejd7naomvSOH2PSI00uuCCqehKisSmeWbXgtV4Ilg/0H4EZZPClMKrPQSqv72r4HJA6cP4G5matQ9EopLbsiv4fmTFRIsNnNOY7Ug4Ts+2p9gq9VeK4eDyg3g6UlhvLp0R22WyuOxLwIQzVq2+cM41zyfoAoeqBpRs4uN4fB7cfg9nycTpvt4108Ld+AFBR5Xby4Xsn0OXKI0wAgYjM34Vcfbc6njMAvIgsNFopF+dpL2/beGZoMmk+iJSOHcROX2+FRvXKzOeOyCGJL/5KMo4dJSOzxmEOOJHgue5ssLTEMQ99GvihSyEAdo9dqdKxxuAZle5Xp1H7FetJcd2TinX0p8XrIq6WPG6TW6Ag4uvtc8Bazu5QnBY4iuYwCQrD3URtzwcQc+q7IPLNePgZGewRC01A0v0B0zaX1D7L9rjmYv5EjlPg493JmwwoSQ5+GmLPL5qUtDMijZuXbiXEvQk09Ajeee5nHtLb+BZuWo+yCJDmbtQM5hShBCgwyy78dKeQmLUlSYGD7675eJlcKnDxRZVXvraSqkBK3VW6shsXqlzMjNDJK5Rnnrb342bDUxKCbE31uPPfA5vMA+8OaAuS2YuPgUuBgrKU5EdsasM5ManrDoJGWVjLtSvwlp6horTOMcGz3OWY5xXodkpqPH/kNSuYgpBCr+K4Q3wY1G0nGj8MKmMHKbvwCGRNNIYFAFj9Co5kjKUFjRmrIZpZJTRJLTQzDkesN56UpzADXO2SdYrbBE8M4JeOKIPKSY+AEdN971s6IZEVbzpLHRxCORo5EJzOng4Ftr7tzBjMhIV/r+qxmBBIaAynrGlJCBAmbH79PZsZUab0rjSTmY0I4LN2U6PmglHk3CQzbQ3SgtQtT9PLdwqpeBctf/bL8OEwrekH982siHDEEmq8lhRrYvndGUmjJo+j1ffcI4jhN5ob97NQ4fuqdKdeC9RE5rTZBLv44adMMwtJrDMu1eUuJtpHrjUUPwLA+gkpP6gLWJsCJqcjKXqwqxU/RV9aEuZA6V5KkfAw+rt+bwaj23mQo7eRm6Qx7fCzGI94kJbn7M4GpAvE0LhSh7Z+oC9aPxBWB1qeKDyb1MmLTsmHdeZkRwEmrMSwKJZUsp6ILVtd6prgReLWbEDjo3LWkiixt2fmqmGoxSTgVxETcLs2GBYrl85dmRGl56itm/dkTehCPUiekGdFphihmZEBC88VKEMPyfHOiFCVaBYRd2dG7TyZGc3BzGSoKTQjm8GO2stcj/9/sKNSwj73knLCzhDEbjKLQU5au6L5NxSYE2v6XEpRs4MvRsSqUSxBrDiKz5JL5tUK0qwAAGRdJREFUlCki8wnUsrlqC8AKmBKrg1KnZmRO1HUzpdkGq+Yjkd06PtI43BJsmafsqeVmL0CiocXOS4R1O2MRxfmxvGhNne177/TpinwKfcyMrot9HI+tHZvEzUuZrNOwvT4jdhDm/RPsNngJzIiUbZbzN8wfK6ExhNgIHXRImZqqV1BOYNPmXrjiVPNfpWKhfbCeJLC3YxljYokvDa+I2/bepV/IZZkSrzgjKvOFxIgdwnKFqyzKmRHBtx9STsaEehHMdtQeugzXylmsubgyxNCYE4tmy5i9uzSB+EWl4ooXGGuH044RCm8OyEbPSA9xybIo/Ou6h1aURM+X6hDJXiTCtn2ZuQm9Gr9yPSAsJXEdzXbszF8SZ4TWfdrnq/oGpOkgE0fvP5i/kMg/ci62ycLkMW8zAlghYgA2w9eQcwrFPUIXecRqKoSNaX0DUqVXixM1VNsT0aq4wd6QLWIgvQFonZ6HkD3TOTHI+Bf8oqDEM+EwTe66KlvKWceRgKlOXLuH1leto+fkGrP9Ahi35jDkBA4EBEB28VWXGaGRS3h9EaQArKSPodSWXzT62HxxXKXyBjs8FaiGzv51WVey8SPcYTnOrtTuDh41ZYgjCiczApHN1x8mSEpDl1lDImI827HrdcUbrZALEwQrFnqZIVc1I0Uol4/NS53tmHsnsYNj8VgpjcFgGUjBoeQlFb7IoBPMqbQ2v0kFUNX0ZN7JmDMZWCeLvmp9Am4796RfoRTJ8gsFc3td5tD5WkRDdt8KJWqOPUnFHZWmLUQdgdsNuniehdROoRzPHytHDIs3J+bNyAkvmK8YIsr6tImTfe+2R1urgC1bgLU1M2JR2b3jXAWexEOacUXssKm9at79TNhNYNp7V7jnQin8Y0mYU0Xtx/bq0zotstbGcr2/M/ncjlo1m+fqTY+e5gRVdrv5WvgXNGaWWkKsVZjzRZkRwAoSw+LRp0YPYW3ZaCLfg6vUubx9faNEuAxzKiuHoxc++4JeNbDt27fi4LQF8jMd81l3umR687NQz3PaQqn2D/GcmBuirQjNrtBR2V0ZW+3O+yfUZxCZmsHDiAljcVhJYticCU8lBnYpOMpCqKNy+LKr8aRsaNcDBn/2Ury2IlenYx9fIu7R3ee2ZejiTMmJgY8v5VQsEMQ587JTxRVo8xGkWSe1rWltZAyGKrmJGwitmBSSrFNxgjxk3o4k5rMZSw4rSQzALOQwtFF3vfSe5fosoorAsCYFNeedWXIT1+UZ+B0qM3NyEi31zpFCkHe6/Ga7+GZ2YzbvUmR9LkpDJpjNa0iEJeNHeSU3zwmL8Os+vEkW32xH51ER2PsXaugyyjylttBoDRvRMa4sMQzHrE4b1UgdVq5vgJE96VX+gtYVRJG2MoVDl0EdS+Q7oFEf220+Cecdjr0JQZBaLAeQsP21eO6clPvraVrkECyqiutCR2b2PaT3dnTvvKg3WChWmhg2xqSIe4F5vTRhLmi7O69Pm16po8PTNyKJ1FC5tFfmq3XGWrhS7tHd5uOz2YlMXZnI42D2ZoEQ8rE5R15f/0Q8MhDPdpThgP1ilrb5qys8IbNzCDMSE9oaM0G/AfM8FulwlFhpYtgYaC86Vftm0CICFZiNqlnH6mYGgVlSauxzR5S01nJ0T/tL08ULpGTvnpKpNB/nD3DmgBuZKc5DMUtyUJdhd5gQwulIdT3w25yDPLKDsfLEsHGOyNjJVPJyS16k6Cn9uxce9i5HpBSr5bMIi+kSo2WWyN+Wdt4EOG0hpTWkCWNWs85lY59hdg5HJl1KLk1r8M87afu0s7PaArlVsEV7O7bjbJS2AJwExAAsghzmoQ3EaVIVVqiQwcdTbE+4vg7VDk+JS9BNi672EmjBabkZZL4mZaP4lZO9MQdScKI51b7XblYIzSufoeLvcOdVFfkXJJsnHDQEYDIxTsf1afe6mJZmYPLdSFIAehADEU2I6NNE9B57/ngi+igRfYWI3kpEp9nw0+35AXv9gsWI3gd9VLCuBjwUqYbgap4kCFGUV5OVSptDTntIipfQhMhFIRzbc44PCz85nxNF5DknJaHJT/T6fV+PptKr8jH8CE9wvcOEsOHsJpplTYiUebJx5oNEH4rdA+Bmcf57AF7LzBcDuAfAi234iwHcw8xPAPBaG28DkHuAM9j+Qdi8a3Uq/4gcps4R2eX8jJDSHpqLUflanAbH9rgNWLp8E12Y13MUjsK+RNiSw5oJKdnYLZxKOR4ThGrnVNC0y4Rw77bUP7RYFBEDEe0C8OMA/tSeE4BnAHi7jXIDgOfZ4+fac9jrz7TxF4q2qjWkh+/Z8HrnnethlFEFyQUn1htyGOS7yj2P2HQgyCFOBpkRCL+7NTWfictA9y2U2EM94Hrh3h+tjf0L4tm2XpMhIO8jCC5mTIiqAq+tGYfjNN4bUoNeRzbajADKV1e+DsDLAey0548AcC8zr9vzQwDOs8fnAbgVAJh5nYjus/HvlhkS0VUArgKArdg+VP4A4cax8/AJzEoSscNShsdxEmXL4PUpsGXNBDKVt6VAK6WQcGJ5ojyP7j4HvmdWrueL7fP8+j5raWsxQD13y27tg5AzIdl8VlDd/yIjX2UXua2ntIWYhZQYm0AKQIHGQETPAXAXM39SBitRE9QZXGsCmK9n5kuY+ZItmN8u0PNzRC5Sc4jMBvU6mni+DbheLuFvyLkx4iLcr3ZsIUmBosSD5i30SlLCQOJ5zPqZ+/jjP7GJH5gRaeesh5zlmFwXsZykAJRpDE8D8BNEdAWArQDOhNEgziKiNas17AJwu41/CMD5AA4R0RqAhwH41twlXxrInisO72oJcZxEGjkLcn3drqT0HsHZ/VQinXcwWmdka7OV2C1RWqZ3bpaiIC5bgfoOU9YMMyU5zk9oad4/YN8JoXz+AZH96DDM1vDB1n2KybhJDsYcOjUGZn4lM+9i5gsAvADAB5j5hQA+COD5NtqVAN5lj99tz2Gvf4C57+b+syGvNcz7JaSckpqxGjf8VJev+SJEpZ1GE59S/jIXp4Cj3DCkIQXW1Wa54/KGvNGceo+mwQ0ZpmwHNllLLcGNSARDm5k6ZAmlmcwktYzyureZ2gIw2w5OvwbgLUT0OwA+DeANNvwNAP6CiA7AaAovmE3EYUh/qGYjOWqoh7nDJmCYClfZ77xpTjNE6n6iTjLIL5kORhzQoiQfsG3fXf34tbe2UIi67u9jVic1JRy6nqcpWk2ZAzUbsQQjER0+DHF9s0kB6EkMzHwjgBvt8UEAT1HiHAfwU3OQbWaE5JDqNufhZOxCiclQGgYTtj41W8n7r11Dd0gmOqpgTkK0YUxsLcjAYesi5vR83b0QmkY3maAXM2gzD4mMaRGYZVZLcGscTjg/e6bnt3F5UjVflyq69YYcloEUgJNk5mMZtEa3CFIo8b1mu6eOeOKaHwYT91Kg5nstAQjt6lTpJK70MSMiDSQPrevX/Bs2jttBqWTvheCmevgJbNpmYxZHCok8Kmp2zSqa+pwScvNx0hND3t+wCFKYNc8e6Z1JkVNX5SCIjBYNfshmyXH6OOvCAYN+6ME4AeEUCMPiQC2Cg2fhQUCz8WsrszaqyvgWOiczJcRcEm0BOAWIAUiRQ1wDkDnvCtfydPFL85bXerj62ZoVfgINN+EJeG1B8V9q7SPQFgrF6qdaaNCegRuZsffa2+moZan4Z/zUZ6sBJEcjhJpVWROiZmU9RPezWCZSAE4RYgBicuhQWec2hBRXiC6NYqDq6bYe82ZFonhYtdjxVc9b3Pb7BWsj5uZoTORR1/BDjX1fUet7FPYhxN+PEEOU4TTo+MGJX7klfGsjlvzDXjZSAE4hYmjQ1ThVhToT1lXWkGsl2oyMww05BDs/RRWe0KjFwYdo2jk2bYCCep10PPo2N2/zTAjEHE4t7jWpKaVFKWHeF0iR2aIwLcEMT04mxoRQvyyVwXK5FjxOKWLg4w92sPM8nZFaXpy5lkKJ49JdYjMBan29UYWVZNv33glMTQCv2f0g1ypgYns+9yu/rp3bDs73xIsgBVc22/UiA4vRHKFy0xXpaXXcOqmijVU0H07lJzP5kZJSV8nx4+Bjy6ctACv5JarZkZ7j0Bel2kVMBF01pxm+6tZuZG8Weey3REN5khz23QkmwtFfeTS4snlxuvtqfSuixBk3M6y2wmzWG7Qu9+huc85A+bgJcPdkJinFi58oPJxUzWSm5F6OijhLaD5InJLEEEIdoCtESfy4cXed98k/dZ0ap6TfGbqdlJhB67X/JiO7+VIyJ7slu19NuSGEEMqJloo+0AdU8jjl/AVGxzRoux4CKNiAxSdZCZyyxOCckbR1K0IVHwh761TD1cJLTIRcWXGcAWXJIKcGVxRtHd/Uzu377vTpOCIQ708IihnqZ+npWCUY2eWmuEFvTT0aGUN1POZ8DhWBptNEHHhtAYTyyUyEpTUdYpyyxNAgZfdz9AsRLw53afsQRR/TosRp1m70Pl4NAFO7nXykIQnzof116pi0tJZY0vBLSUHk70ZZ1PkKPYdU/KtJmF6aeFWV3kOBYHwL3gdRMG/Bj3asBk55YjD+hq19UhSG5cIdclqDFrer8WUqnqvA7gM0bq8ANR/tPB6ii8sbqkHZOPILTdNp2dyB3ojJPAG5CEqDXT3JRKATHVqFSMNHj/UVeNNwyhMD0DiCyh2SuUZQ6jj0pUe/Wlnx9VT5JcVxY7PbXm9Yb1bimyklCqvWu29sap9vC5L2lLVlRiREdNlWVeRPieJUFbgi65DtJoVVMR8kRmIIoKnNJQ1QU+MLe8recbpIocRnYSs9W086Vb7Cp9tciVmRguZwlVlbsvJe/Q1ybLZEs3JVlNFYKFwPockaaQqriJEYBBrNIXZIphrZkF421hD6jkqUmieFTlNm0yBhVXg36ck5AKNJUHk/Q4dc3hVje9pajvtLYo00BXc5yG4ODS7mc6cNwDoetfgTsR5CjkTE4qyY6RBjJAYFfPy4IIecil96LXde4lfoa5505ZuJG38305kZjiQAQR6IGoTi+HQ/8Ze3+ywymrcCwQDI3WPEDhXZb0tG/gXpcHQaTpCfiEhYaVIARmIohGYipNT5VC/fx8zoKqsrTc4pWKKpiHiuN2cRP2oEUaCSR5SXavpopELzschSkd18BVcOA0yVHaZU0jqnZDCE6kSn5ndt9ScUj8SQQGhWxIhJYIi9rWkBfbSQWBYnT6wfz7O7dQ1c5N1q7HHDj2WL8gquKVpR67Zj/0Sx8AgbsgtiwU3W6RiPRjgTYm1i4ga+BUEIWybg+w73EWhpsfrUtmCUTV3V/AZQjmUc7byEaJRrrSBvzGfy6EsYOblzadxfywiPfrVrhcUU83JHROd01FZbWocjuenmTsuxw7+8Ze2kIQVg1BiKEGoPKW2hT+Uv0RZSjVdroH0dcX1IIen965AjvpdU3C5zpuPehnCcmg+lJzWRMSG8tiDNp7UJ+P6ThxAcRmLoAZ0gghjI11RW4sW9e0w0Jc5JeS7TpsrvArVPuUSWrnI07UimTRFrBiVc2iUXoZmbcILDV+FWTyLSFk5SQnAYTYkB4OPHE3U25ejTGn4cJw7LORBTYfKaVlxXQ7OqsWYJJZ2MqcauxZFORvmrmT19taBM8SWoyE4JZyGmMSGCyUwEYzacxKQAjMQwGHzseORMlxU99rJr56keUyOTnjU+FkGVU0mkaffSTVBUcC6s9D6KC+wgsbjo9L0zKVu4VfZL1WQzsL4E3Hd/mWwrjNGUmAFyqms4ehH7DeIGn3NOynTxcVB6mD5V51VFI84fCikMtf279PmY7HK+mR7ZaWHBbRLacydsOZUlRbmFm9UM3JRxvufkJwOJUWOYE0LzItYQEJ2nHHU50kgdZ1oNKyc+TBBCQAoptb5Utfdjf4oQSTVmftAeDWnPTtx/VYl1D9T4HLZMwGsE3H3v/OVccowawxzhNAjatrXdAFVPfE4z6PIxiGtap6tGFxU/Vq05mShMm3VspsggJj4tXg+U+CizmoYkRbPZil9Jabex49O3gL5xzzD5TgKMxLAA5AkiPpbnKRLpMj2UJK2LiSITl9OFaOcl5lAKObIZCM1lo/GuG6KcEBh2pem91mQ4Mj9xVhGjKbFA8LHjmREMh7g154zm1LkIVkkhIh5JVr077ZLuujStROae1PBZnZNWM5hU4Amh3rrWkMKIUWPYCOgaRIxS00GLnyx5QPqUSVNKAHH6Lq1CS6NESfFlSdcWP1rrc+C1Cfi0Ceiuexbh7VhpjMSwgQhGMZIkETcWTR/ODUEUmB0+rkwTayypsL6Yh6zl0Rjmk3IsIhJgTIa6tvMSGpNhJAQdIzFsEgItAuggifjYJciNdKTysvGSvo+USTNPUhiaXxqSCPzXtty5W0EpfQgjshiJYZPR0iIAhSRKetYu7UAZput0UvYhnlJZZoDi3uA4MHhUZD6ecwoON86KkRiWCCFJbAPAhT6JLiTiBuQQj4LkJiBtAlqkIIY9SYTZ72PQN0YymAUjMSwp+FizA1A3SfREsp1rBWycOaBnmxn2JBjnY1WBCSCrGSwBja08RmJYAUiSABxRAO3ZjAXIzl+ITJaSwZEhiEca1CkbQhY/c9HKZD85L/0FIxnMF0XEQERfA3AYwBTAOjNfQkRnA3grgAsAfA3ATzPzPUREAPYCuALAUQA/z8yfmr/opy46iSKIPKAA2TipQ1PJDScGeRVcIxEozQP7x4dP8VlHG4g+E5x+mJmfxMyX2PNXAHg/M18M4P32HACeDeBi+3cVgD+al7AjdPCxY/bvOMxW8IRwp2fx1xwYZOcrcTsuxXGidBTFzUJkKAgAk8p8Um9tAn7gKPjIAyMpbDBmMSWeC+Bye3wDgBsB/JoNfyMzM4CPENFZRPQYZr5jFkFHlIGPHk1eozO2w68yZGcqcNiI1YEJ5aKq8reDWv4CAoLdpoGx0S8hSomBAfwDETGAP2Hm6wGc6xo7M99BROfYuOcBuFWkPWTDAmIgoqtgNApsxfbhdzCiGPxAhjR2nCEiRuwQk0B2MmPsIBwb/yqilBiexsy328b/PiL650xcrbq0R8wNuVwPAGfS2YtwcY3oAT7ywGaLMGKJUORjYObb7e9dAN4B4CkA7iSixwCA/b3LRj8E4HyRfBeA2+cl8IgRIxaPTmIgojOIaKc7BvCjAD4P4N0ArrTRrgTwLnv8bgAvIoOnArhv9C+MGLFaKDElzgXwDjMKiTUAb2bm/05EHwfwNiJ6MYB/AfBTNv57YYYqD8AMV/6HuUs9YsSIhYK4zzcEFyUE0WEAX9psOQrxSAB3b7YQBVgVOYHVkXVV5AR0WR/HzI8qSbwsMx+/JOZHLDWI6BOrIOuqyAmsjqyrIicwu6zjDk4jRoxoYSSGESNGtLAsxHD9ZgvQA6si66rICayOrKsiJzCjrEvhfBwxYsRyYVk0hhEjRiwRRmIYMWJEC5tODET0Y0T0JSI6QESv6E6xUFn+jIjuIqLPi7Czieh9RPQV+/twG05EtM/K/TkievIGy3o+EX2QiG4moi8Q0Z5llJeIthLRx4jos1bO37Lhjyeij1o530pEp9nw0+35AXv9go2QU8g7IaJPE9F7llzOrxHRPxHRZ4joEzZsfu+emTftD8AEwC0ALgRwGoDPAvjuTZTnUgBPBvB5EfYqAK+wx68A8Hv2+AoAfwezaOypAD66wbI+BsCT7fFOAF8G8N3LJq8tb4c93gLgo7b8twF4gQ3/YwD/0R7/IoA/tscvAPDWDX6uvwrgzQDeY8+XVc6vAXhkFDa3d79hN5K4uR8C8Pfi/JUAXrnJMl0QEcOXADzGHj8GZjIWAPwJgJ/V4m2S3O8C8KxllhfAdgCfAvCDMLPy1uJ6AODvAfyQPV6z8WiD5NsFs+nQMwC8xzakpZPTlqkRw9ze/WabEqm9G5YJwb4TALr2ndhwWDX2+2B646WT16rnn4FZgfs+GC3xXmZeV2Txctrr9wF4xEbICeB1AF4OwH7hFo9YUjmBZo+UT9q9TYA5vvvNnhJdtHfDkmIpZCeiHQD+GsBLmfl+ijdKEVGVsA2Rl5mnAJ5ERGfBLNv/rowsmyInET0HwF3M/EkiurxAls1+/3PfI0ViszWGVdi7YWn3nSCiLTCk8CZm/hsbvLTyMvO9MFsAPhXAWUTkOiYpi5fTXn8YgG9tgHhPA/ATZDY+fguMOfG6JZQTwOL3SNlsYvg4gIut5/c0GCfOuzdZphhLue8EGdXgDQBuZubXLKu8RPQoqymAiLYB+BEANwP4IIDnJ+R08j8fwAfYGsaLBDO/kpl3MfMFMPXwA8z8wmWTE9igPVI20vmUcKJcAeNRvwXAf9pkWf4SZm/KEzAs+2IYu/H9AL5if8+2cQnAH1q5/wnAJRss67+GUQc/B+Az9u+KZZMXwPcA+LSV8/MA/rMNvxDAx2D27fgrAKfb8K32/IC9fuEm1IPL0YxKLJ2cVqbP2r8vuHYzz3c/TokeMWJEC5ttSowYMWIJMRLDiBEjWhiJYcSIES2MxDBixIgWRmIYMWJECyMxjBgxooWRGEaMGNHC/w902u3Q2uLqXgAAAABJRU5ErkJggg==\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "img = Image.open(IMAGE_PATH + df['id'][100] + '.png')\n",
    "mask = Image.open(MASK_PATH  + df['id'][100] + '.png')\n",
    "print('Image Size', np.asarray(img).shape)\n",
    "print('Mask Size', np.asarray(mask).shape)\n",
    "\n",
    "\n",
    "plt.imshow(img)\n",
    "plt.imshow(mask, alpha=0.6)\n",
    "plt.title('Picture with Mask Appplied')\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "0ff8f8df",
   "metadata": {},
   "source": [
    "### Dataset"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 36,
   "id": "6ae717c4",
   "metadata": {},
   "outputs": [],
   "source": [
    "class MetalDataset(Dataset):\n",
    "    \n",
    "    def __init__(self, img_path, mask_path, X, mean, std, transform=None, patch=False):\n",
    "        self.img_path = img_path\n",
    "        self.mask_path = mask_path\n",
    "        self.X = X\n",
    "        self.transform = transform\n",
    "        self.patches = patch\n",
    "        self.mean = mean\n",
    "        self.std = std\n",
    "        \n",
    "        \n",
    "    def __len__(self):\n",
    "        return len(self.X)\n",
    "    \n",
    "    def __getitem__(self, idx):\n",
    "        img = cv2.imread(self.img_path + self.X[idx] + '.png')\n",
    "        img = cv2.cvtColor(img, cv2.COLOR_BGR2RGB)\n",
    "        mask = cv2.imread(self.mask_path + self.X[idx] + '.png', cv2.IMREAD_GRAYSCALE)\n",
    "        \n",
    "        if self.transform is not None:\n",
    "            aug = self.transform(image=img, mask=mask)\n",
    "            img = Image.fromarray(aug['image'])\n",
    "            mask = aug['mask']\n",
    "        \n",
    "        if self.transform is None:\n",
    "            img = Image.fromarray(img)\n",
    "        \n",
    "        t = T.Compose([T.ToTensor(), T.Normalize(self.mean, self.std)])\n",
    "        img = t(img)\n",
    "        mask = torch.from_numpy(mask).long()\n",
    "        \n",
    "        if self.patches:\n",
    "            img, mask = self.tiles(img, mask)\n",
    "            \n",
    "        return img, mask   \n",
    "    \n",
    "    def tiles(self, img, mask):\n",
    "\n",
    "        img_patches = img.unfold(1, 480, 480).unfold(2, 500, 500) \n",
    "        img_patches  = img_patches.contiguous().view(3,-1, 480, 500) \n",
    "        img_patches = img_patches.permute(1,0,2,3)\n",
    "        \n",
    "        mask_patches = mask.unfold(0, 480, 480).unfold(1, 500, 500)\n",
    "        mask_patches = mask_patches.contiguous().view(-1, 480, 500)\n",
    "        \n",
    "        return img_patches, mask_patches"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "99a4f778",
   "metadata": {},
   "outputs": [],
   "source": [
    "mean=[0.485, 0.456, 0.406]\n",
    "std=[0.229, 0.224, 0.225]\n",
    "\n",
    "t_train = A.Compose([A.Resize(480, 480, interpolation=cv2.INTER_NEAREST), A.HorizontalFlip(), A.VerticalFlip(), \n",
    "                     A.GridDistortion(p=0.2), A.RandomBrightnessContrast((0,0.5),(0,0.5)),\n",
    "                     A.GaussNoise()])\n",
    "\n",
    "t_val = A.Compose([A.Resize(480, 480, interpolation=cv2.INTER_NEAREST), A.HorizontalFlip(),\n",
    "                   A.GridDistortion(p=0.2)])\n",
    "\n",
    "#datasets\n",
    "train_set = MetalDataset(IMAGE_PATH, MASK_PATH, X_train, mean, std, t_train, patch=False)\n",
    "val_set = MetalDataset(IMAGE_PATH, MASK_PATH, X_val, mean, std, t_val, patch=False)\n",
    "\n",
    "#dataloader\n",
    "batch_size= 3 \n",
    "\n",
    "train_loader = DataLoader(train_set, batch_size=batch_size, shuffle=True)\n",
    "val_loader = DataLoader(val_set, batch_size=batch_size, shuffle=True)   "
   ]
  },
  {
   "cell_type": "markdown",
   "id": "87a07294",
   "metadata": {},
   "source": [
    "### Unet Model"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 38,
   "id": "8ad27aad",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Unet(\n",
       "  (encoder): MobileNetV2Encoder(\n",
       "    (features): Sequential(\n",
       "      (0): ConvBNActivation(\n",
       "        (0): Conv2d(3, 32, kernel_size=(3, 3), stride=(2, 2), padding=(1, 1), bias=False)\n",
       "        (1): BatchNorm2d(32, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
       "        (2): ReLU6(inplace=True)\n",
       "      )\n",
       "      (1): InvertedResidual(\n",
       "        (conv): Sequential(\n",
       "          (0): ConvBNActivation(\n",
       "            (0): Conv2d(32, 32, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), groups=32, bias=False)\n",
       "            (1): BatchNorm2d(32, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
       "            (2): ReLU6(inplace=True)\n",
       "          )\n",
       "          (1): Conv2d(32, 16, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
       "          (2): BatchNorm2d(16, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
       "        )\n",
       "      )\n",
       "      (2): InvertedResidual(\n",
       "        (conv): Sequential(\n",
       "          (0): ConvBNActivation(\n",
       "            (0): Conv2d(16, 96, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
       "            (1): BatchNorm2d(96, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
       "            (2): ReLU6(inplace=True)\n",
       "          )\n",
       "          (1): ConvBNActivation(\n",
       "            (0): Conv2d(96, 96, kernel_size=(3, 3), stride=(2, 2), padding=(1, 1), groups=96, bias=False)\n",
       "            (1): BatchNorm2d(96, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
       "            (2): ReLU6(inplace=True)\n",
       "          )\n",
       "          (2): Conv2d(96, 24, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
       "          (3): BatchNorm2d(24, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
       "        )\n",
       "      )\n",
       "      (3): InvertedResidual(\n",
       "        (conv): Sequential(\n",
       "          (0): ConvBNActivation(\n",
       "            (0): Conv2d(24, 144, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
       "            (1): BatchNorm2d(144, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
       "            (2): ReLU6(inplace=True)\n",
       "          )\n",
       "          (1): ConvBNActivation(\n",
       "            (0): Conv2d(144, 144, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), groups=144, bias=False)\n",
       "            (1): BatchNorm2d(144, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
       "            (2): ReLU6(inplace=True)\n",
       "          )\n",
       "          (2): Conv2d(144, 24, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
       "          (3): BatchNorm2d(24, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
       "        )\n",
       "      )\n",
       "      (4): InvertedResidual(\n",
       "        (conv): Sequential(\n",
       "          (0): ConvBNActivation(\n",
       "            (0): Conv2d(24, 144, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
       "            (1): BatchNorm2d(144, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
       "            (2): ReLU6(inplace=True)\n",
       "          )\n",
       "          (1): ConvBNActivation(\n",
       "            (0): Conv2d(144, 144, kernel_size=(3, 3), stride=(2, 2), padding=(1, 1), groups=144, bias=False)\n",
       "            (1): BatchNorm2d(144, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
       "            (2): ReLU6(inplace=True)\n",
       "          )\n",
       "          (2): Conv2d(144, 32, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
       "          (3): BatchNorm2d(32, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
       "        )\n",
       "      )\n",
       "      (5): InvertedResidual(\n",
       "        (conv): Sequential(\n",
       "          (0): ConvBNActivation(\n",
       "            (0): Conv2d(32, 192, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
       "            (1): BatchNorm2d(192, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
       "            (2): ReLU6(inplace=True)\n",
       "          )\n",
       "          (1): ConvBNActivation(\n",
       "            (0): Conv2d(192, 192, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), groups=192, bias=False)\n",
       "            (1): BatchNorm2d(192, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
       "            (2): ReLU6(inplace=True)\n",
       "          )\n",
       "          (2): Conv2d(192, 32, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
       "          (3): BatchNorm2d(32, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
       "        )\n",
       "      )\n",
       "      (6): InvertedResidual(\n",
       "        (conv): Sequential(\n",
       "          (0): ConvBNActivation(\n",
       "            (0): Conv2d(32, 192, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
       "            (1): BatchNorm2d(192, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
       "            (2): ReLU6(inplace=True)\n",
       "          )\n",
       "          (1): ConvBNActivation(\n",
       "            (0): Conv2d(192, 192, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), groups=192, bias=False)\n",
       "            (1): BatchNorm2d(192, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
       "            (2): ReLU6(inplace=True)\n",
       "          )\n",
       "          (2): Conv2d(192, 32, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
       "          (3): BatchNorm2d(32, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
       "        )\n",
       "      )\n",
       "      (7): InvertedResidual(\n",
       "        (conv): Sequential(\n",
       "          (0): ConvBNActivation(\n",
       "            (0): Conv2d(32, 192, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
       "            (1): BatchNorm2d(192, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
       "            (2): ReLU6(inplace=True)\n",
       "          )\n",
       "          (1): ConvBNActivation(\n",
       "            (0): Conv2d(192, 192, kernel_size=(3, 3), stride=(2, 2), padding=(1, 1), groups=192, bias=False)\n",
       "            (1): BatchNorm2d(192, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
       "            (2): ReLU6(inplace=True)\n",
       "          )\n",
       "          (2): Conv2d(192, 64, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
       "          (3): BatchNorm2d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
       "        )\n",
       "      )\n",
       "      (8): InvertedResidual(\n",
       "        (conv): Sequential(\n",
       "          (0): ConvBNActivation(\n",
       "            (0): Conv2d(64, 384, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
       "            (1): BatchNorm2d(384, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
       "            (2): ReLU6(inplace=True)\n",
       "          )\n",
       "          (1): ConvBNActivation(\n",
       "            (0): Conv2d(384, 384, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), groups=384, bias=False)\n",
       "            (1): BatchNorm2d(384, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
       "            (2): ReLU6(inplace=True)\n",
       "          )\n",
       "          (2): Conv2d(384, 64, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
       "          (3): BatchNorm2d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
       "        )\n",
       "      )\n",
       "      (9): InvertedResidual(\n",
       "        (conv): Sequential(\n",
       "          (0): ConvBNActivation(\n",
       "            (0): Conv2d(64, 384, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
       "            (1): BatchNorm2d(384, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
       "            (2): ReLU6(inplace=True)\n",
       "          )\n",
       "          (1): ConvBNActivation(\n",
       "            (0): Conv2d(384, 384, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), groups=384, bias=False)\n",
       "            (1): BatchNorm2d(384, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
       "            (2): ReLU6(inplace=True)\n",
       "          )\n",
       "          (2): Conv2d(384, 64, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
       "          (3): BatchNorm2d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
       "        )\n",
       "      )\n",
       "      (10): InvertedResidual(\n",
       "        (conv): Sequential(\n",
       "          (0): ConvBNActivation(\n",
       "            (0): Conv2d(64, 384, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
       "            (1): BatchNorm2d(384, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
       "            (2): ReLU6(inplace=True)\n",
       "          )\n",
       "          (1): ConvBNActivation(\n",
       "            (0): Conv2d(384, 384, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), groups=384, bias=False)\n",
       "            (1): BatchNorm2d(384, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
       "            (2): ReLU6(inplace=True)\n",
       "          )\n",
       "          (2): Conv2d(384, 64, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
       "          (3): BatchNorm2d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
       "        )\n",
       "      )\n",
       "      (11): InvertedResidual(\n",
       "        (conv): Sequential(\n",
       "          (0): ConvBNActivation(\n",
       "            (0): Conv2d(64, 384, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
       "            (1): BatchNorm2d(384, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
       "            (2): ReLU6(inplace=True)\n",
       "          )\n",
       "          (1): ConvBNActivation(\n",
       "            (0): Conv2d(384, 384, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), groups=384, bias=False)\n",
       "            (1): BatchNorm2d(384, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
       "            (2): ReLU6(inplace=True)\n",
       "          )\n",
       "          (2): Conv2d(384, 96, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
       "          (3): BatchNorm2d(96, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
       "        )\n",
       "      )\n",
       "      (12): InvertedResidual(\n",
       "        (conv): Sequential(\n",
       "          (0): ConvBNActivation(\n",
       "            (0): Conv2d(96, 576, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
       "            (1): BatchNorm2d(576, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
       "            (2): ReLU6(inplace=True)\n",
       "          )\n",
       "          (1): ConvBNActivation(\n",
       "            (0): Conv2d(576, 576, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), groups=576, bias=False)\n",
       "            (1): BatchNorm2d(576, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
       "            (2): ReLU6(inplace=True)\n",
       "          )\n",
       "          (2): Conv2d(576, 96, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
       "          (3): BatchNorm2d(96, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
       "        )\n",
       "      )\n",
       "      (13): InvertedResidual(\n",
       "        (conv): Sequential(\n",
       "          (0): ConvBNActivation(\n",
       "            (0): Conv2d(96, 576, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
       "            (1): BatchNorm2d(576, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
       "            (2): ReLU6(inplace=True)\n",
       "          )\n",
       "          (1): ConvBNActivation(\n",
       "            (0): Conv2d(576, 576, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), groups=576, bias=False)\n",
       "            (1): BatchNorm2d(576, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
       "            (2): ReLU6(inplace=True)\n",
       "          )\n",
       "          (2): Conv2d(576, 96, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
       "          (3): BatchNorm2d(96, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
       "        )\n",
       "      )\n",
       "      (14): InvertedResidual(\n",
       "        (conv): Sequential(\n",
       "          (0): ConvBNActivation(\n",
       "            (0): Conv2d(96, 576, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
       "            (1): BatchNorm2d(576, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
       "            (2): ReLU6(inplace=True)\n",
       "          )\n",
       "          (1): ConvBNActivation(\n",
       "            (0): Conv2d(576, 576, kernel_size=(3, 3), stride=(2, 2), padding=(1, 1), groups=576, bias=False)\n",
       "            (1): BatchNorm2d(576, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
       "            (2): ReLU6(inplace=True)\n",
       "          )\n",
       "          (2): Conv2d(576, 160, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
       "          (3): BatchNorm2d(160, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
       "        )\n",
       "      )\n",
       "      (15): InvertedResidual(\n",
       "        (conv): Sequential(\n",
       "          (0): ConvBNActivation(\n",
       "            (0): Conv2d(160, 960, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
       "            (1): BatchNorm2d(960, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
       "            (2): ReLU6(inplace=True)\n",
       "          )\n",
       "          (1): ConvBNActivation(\n",
       "            (0): Conv2d(960, 960, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), groups=960, bias=False)\n",
       "            (1): BatchNorm2d(960, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
       "            (2): ReLU6(inplace=True)\n",
       "          )\n",
       "          (2): Conv2d(960, 160, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
       "          (3): BatchNorm2d(160, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
       "        )\n",
       "      )\n",
       "      (16): InvertedResidual(\n",
       "        (conv): Sequential(\n",
       "          (0): ConvBNActivation(\n",
       "            (0): Conv2d(160, 960, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
       "            (1): BatchNorm2d(960, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
       "            (2): ReLU6(inplace=True)\n",
       "          )\n",
       "          (1): ConvBNActivation(\n",
       "            (0): Conv2d(960, 960, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), groups=960, bias=False)\n",
       "            (1): BatchNorm2d(960, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
       "            (2): ReLU6(inplace=True)\n",
       "          )\n",
       "          (2): Conv2d(960, 160, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
       "          (3): BatchNorm2d(160, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
       "        )\n",
       "      )\n",
       "      (17): InvertedResidual(\n",
       "        (conv): Sequential(\n",
       "          (0): ConvBNActivation(\n",
       "            (0): Conv2d(160, 960, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
       "            (1): BatchNorm2d(960, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
       "            (2): ReLU6(inplace=True)\n",
       "          )\n",
       "          (1): ConvBNActivation(\n",
       "            (0): Conv2d(960, 960, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), groups=960, bias=False)\n",
       "            (1): BatchNorm2d(960, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
       "            (2): ReLU6(inplace=True)\n",
       "          )\n",
       "          (2): Conv2d(960, 320, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
       "          (3): BatchNorm2d(320, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
       "        )\n",
       "      )\n",
       "      (18): ConvBNActivation(\n",
       "        (0): Conv2d(320, 1280, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
       "        (1): BatchNorm2d(1280, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
       "        (2): ReLU6(inplace=True)\n",
       "      )\n",
       "    )\n",
       "  )\n",
       "  (decoder): UnetDecoder(\n",
       "    (center): Identity()\n",
       "    (blocks): ModuleList(\n",
       "      (0): DecoderBlock(\n",
       "        (conv1): Conv2dReLU(\n",
       "          (0): Conv2d(1376, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n",
       "          (1): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
       "          (2): ReLU(inplace=True)\n",
       "        )\n",
       "        (attention1): Attention(\n",
       "          (attention): Identity()\n",
       "        )\n",
       "        (conv2): Conv2dReLU(\n",
       "          (0): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n",
       "          (1): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
       "          (2): ReLU(inplace=True)\n",
       "        )\n",
       "        (attention2): Attention(\n",
       "          (attention): Identity()\n",
       "        )\n",
       "      )\n",
       "      (1): DecoderBlock(\n",
       "        (conv1): Conv2dReLU(\n",
       "          (0): Conv2d(288, 128, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n",
       "          (1): BatchNorm2d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
       "          (2): ReLU(inplace=True)\n",
       "        )\n",
       "        (attention1): Attention(\n",
       "          (attention): Identity()\n",
       "        )\n",
       "        (conv2): Conv2dReLU(\n",
       "          (0): Conv2d(128, 128, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n",
       "          (1): BatchNorm2d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
       "          (2): ReLU(inplace=True)\n",
       "        )\n",
       "        (attention2): Attention(\n",
       "          (attention): Identity()\n",
       "        )\n",
       "      )\n",
       "      (2): DecoderBlock(\n",
       "        (conv1): Conv2dReLU(\n",
       "          (0): Conv2d(152, 64, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n",
       "          (1): BatchNorm2d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
       "          (2): ReLU(inplace=True)\n",
       "        )\n",
       "        (attention1): Attention(\n",
       "          (attention): Identity()\n",
       "        )\n",
       "        (conv2): Conv2dReLU(\n",
       "          (0): Conv2d(64, 64, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n",
       "          (1): BatchNorm2d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
       "          (2): ReLU(inplace=True)\n",
       "        )\n",
       "        (attention2): Attention(\n",
       "          (attention): Identity()\n",
       "        )\n",
       "      )\n",
       "      (3): DecoderBlock(\n",
       "        (conv1): Conv2dReLU(\n",
       "          (0): Conv2d(80, 32, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n",
       "          (1): BatchNorm2d(32, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
       "          (2): ReLU(inplace=True)\n",
       "        )\n",
       "        (attention1): Attention(\n",
       "          (attention): Identity()\n",
       "        )\n",
       "        (conv2): Conv2dReLU(\n",
       "          (0): Conv2d(32, 32, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n",
       "          (1): BatchNorm2d(32, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
       "          (2): ReLU(inplace=True)\n",
       "        )\n",
       "        (attention2): Attention(\n",
       "          (attention): Identity()\n",
       "        )\n",
       "      )\n",
       "      (4): DecoderBlock(\n",
       "        (conv1): Conv2dReLU(\n",
       "          (0): Conv2d(32, 16, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n",
       "          (1): BatchNorm2d(16, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
       "          (2): ReLU(inplace=True)\n",
       "        )\n",
       "        (attention1): Attention(\n",
       "          (attention): Identity()\n",
       "        )\n",
       "        (conv2): Conv2dReLU(\n",
       "          (0): Conv2d(16, 16, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n",
       "          (1): BatchNorm2d(16, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)\n",
       "          (2): ReLU(inplace=True)\n",
       "        )\n",
       "        (attention2): Attention(\n",
       "          (attention): Identity()\n",
       "        )\n",
       "      )\n",
       "    )\n",
       "  )\n",
       "  (segmentation_head): SegmentationHead(\n",
       "    (0): Conv2d(16, 23, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))\n",
       "    (1): Identity()\n",
       "    (2): Activation(\n",
       "      (activation): Identity()\n",
       "    )\n",
       "  )\n",
       ")"
      ]
     },
     "execution_count": 38,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "model = smp.Unet('mobilenet_v2', encoder_weights='imagenet', classes=23, activation=None, \n",
    "                 encoder_depth=5, decoder_channels=[256, 128, 64, 32, 16])\n",
    "model"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "25e94346",
   "metadata": {},
   "source": [
    "### Training"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 39,
   "id": "72969241",
   "metadata": {},
   "outputs": [],
   "source": [
    "def pixel_accuracy(output, mask):\n",
    "    with torch.no_grad():\n",
    "        output = torch.argmax(F.softmax(output, dim=1), dim=1)\n",
    "        correct = torch.eq(output, mask).int()\n",
    "        accuracy = float(correct.sum()) / float(correct.numel())\n",
    "    return accuracy"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 40,
   "id": "2891767c",
   "metadata": {},
   "outputs": [],
   "source": [
    "def mIoU(pred_mask, mask, smooth=1e-10, n_classes=23):\n",
    "    with torch.no_grad():\n",
    "        pred_mask = F.softmax(pred_mask, dim=1)\n",
    "        pred_mask = torch.argmax(pred_mask, dim=1)\n",
    "        pred_mask = pred_mask.contiguous().view(-1)\n",
    "        mask = mask.contiguous().view(-1)\n",
    "\n",
    "        iou_per_class = []\n",
    "        for clas in range(0, n_classes): #loop per pixel class\n",
    "            true_class = pred_mask == clas\n",
    "            true_label = mask == clas\n",
    "\n",
    "            if true_label.long().sum().item() == 0: #no exist label in this loop\n",
    "                iou_per_class.append(np.nan)\n",
    "            else:\n",
    "                intersect = torch.logical_and(true_class, true_label).sum().float().item()\n",
    "                union = torch.logical_or(true_class, true_label).sum().float().item()\n",
    "\n",
    "                iou = (intersect + smooth) / (union +smooth)\n",
    "                iou_per_class.append(iou)\n",
    "        return np.nanmean(iou_per_class)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 41,
   "id": "ec98e1e3",
   "metadata": {},
   "outputs": [],
   "source": [
    "def get_lr(optimizer):\n",
    "    for param_group in optimizer.param_groups:\n",
    "        return param_group['lr']\n",
    "    \n",
    "def fit(epochs, model, train_loader, val_loader, criterion, optimizer, scheduler, patch=False):\n",
    "    torch.cuda.empty_cache()\n",
    "    train_losses = []\n",
    "    test_losses = []\n",
    "    val_iou = []; val_acc = []\n",
    "    train_iou = []; train_acc = []\n",
    "    lrs = []\n",
    "    min_loss = np.inf\n",
    "    decrease = 1 ; not_improve=0  \n",
    "    \n",
    "    \n",
    "    model.to(device)\n",
    "    fit_time = time.time()\n",
    "    for e in range(epochs):\n",
    "        since = time.time()\n",
    "        running_loss = 0\n",
    "        iou_score = 0\n",
    "        accuracy = 0\n",
    "        #training loop\n",
    "        model.train()\n",
    "        for i, data in enumerate(tqdm(train_loader)):\n",
    "            #training phase\n",
    "            image_tiles, mask_tiles = data\n",
    "            if patch:\n",
    "                bs, n_tiles, c, h, w = image_tiles.size()\n",
    "\n",
    "                image_tiles = image_tiles.view(-1,c, h, w)\n",
    "                mask_tiles = mask_tiles.view(-1, h, w)\n",
    "                \n",
    "            image = image_tiles.to(device); mask = mask_tiles.to(device);\n",
    "            #forward\n",
    "            output = model(image)\n",
    "            loss = criterion(output, mask)\n",
    "            #evaluation metrics\n",
    "            iou_score += mIoU(output, mask)\n",
    "            accuracy += pixel_accuracy(output, mask)\n",
    "            #backward\n",
    "            loss.backward()\n",
    "            optimizer.step() #update weight          \n",
    "            optimizer.zero_grad() #reset gradient\n",
    "            \n",
    "            #step the learning rate\n",
    "            lrs.append(get_lr(optimizer))\n",
    "            scheduler.step() \n",
    "            \n",
    "            running_loss += loss.item()\n",
    "            \n",
    "    else:\n",
    "            model.eval()\n",
    "            test_loss = 0\n",
    "            test_accuracy = 0\n",
    "            val_iou_score = 0\n",
    "            #validation loop\n",
    "            with torch.no_grad():\n",
    "                for i, data in enumerate(tqdm(val_loader)):\n",
    "                    #reshape to 9 patches from single image, delete batch size\n",
    "                    image_tiles, mask_tiles = data\n",
    "\n",
    "                    if patch:\n",
    "                        bs, n_tiles, c, h, w = image_tiles.size()\n",
    "\n",
    "                        image_tiles = image_tiles.view(-1,c, h, w)\n",
    "                        mask_tiles = mask_tiles.view(-1, h, w)\n",
    "                        \n",
    "                         \n",
    "                    image = image_tiles.to(device); mask = mask_tiles.to(device);\n",
    "                    output = model(image)\n",
    "                    #evaluation metrics\n",
    "                    val_iou_score +=  mIoU(output, mask)\n",
    "                    test_accuracy += pixel_accuracy(output, mask)\n",
    "                    #loss\n",
    "                    loss = criterion(output, mask)                                  \n",
    "                    test_loss += loss.item()\n",
    "                    \n",
    "            #calculatio mean for each batch\n",
    "            train_losses.append(running_loss/len(train_loader))\n",
    "            test_losses.append(test_loss/len(val_loader))\n",
    "\n",
    "\n",
    "            if min_loss > (test_loss/len(val_loader)):\n",
    "                print('Loss Decreasing.. {:.3f} >> {:.3f} '.format(min_loss, (test_loss/len(val_loader))))\n",
    "                min_loss = (test_loss/len(val_loader))\n",
    "                decrease += 1\n",
    "                if decrease % 5 == 0:\n",
    "                    print('saving model...')\n",
    "                    torch.save(model, 'Unet-Mobilenet_v2_mIoU-{:.3f}.pt'.format(val_iou_score/len(val_loader)))\n",
    "                    \n",
    "                    \n",
    "            if (test_loss/len(val_loader)) > min_loss:\n",
    "                not_improve += 1\n",
    "                min_loss = (test_loss/len(val_loader))\n",
    "                print(f'Loss Not Decrease for {not_improve} time')\n",
    "                if not_improve == 7:\n",
    "                    print('Loss not decrease for 7 times, Stop Training')\n",
    "                    \n",
    "                #break\n",
    "                    \n",
    "                    #iou\n",
    "            val_iou.append(val_iou_score/len(val_loader))\n",
    "            train_iou.append(iou_score/len(train_loader))\n",
    "            train_acc.append(accuracy/len(train_loader))\n",
    "            val_acc.append(test_accuracy/ len(val_loader))\n",
    "            print(\"Epoch:{}/{}..\".format(e+1, epochs),\n",
    "                  \"Train Loss: {:.3f}..\".format(running_loss/len(train_loader)),\n",
    "                  \"Val Loss: {:.3f}..\".format(test_loss/len(val_loader)),\n",
    "                  \"Train mIoU:{:.3f}..\".format(iou_score/len(train_loader)),\n",
    "                  \"Val mIoU: {:.3f}..\".format(val_iou_score/len(val_loader)),\n",
    "                  \"Train Acc:{:.3f}..\".format(accuracy/len(train_loader)),\n",
    "                  \"Val Acc:{:.3f}..\".format(test_accuracy/len(val_loader)),\n",
    "                  \"Time: {:.2f}m\".format((time.time()-since)/60))\n",
    "        \n",
    "    history = {'train_loss' : train_losses, 'val_loss': test_losses,\n",
    "               'train_miou' :train_iou, 'val_miou':val_iou,\n",
    "               'train_acc' :train_acc, 'val_acc':val_acc,\n",
    "               'lrs': lrs}\n",
    "    print('Total time: {:.2f} m' .format((time.time()- fit_time)/60))\n",
    "    return history\n",
    "    \n",
    "    "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 42,
   "id": "5fc1994f",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "a545f27c90dc408ea36c186dccbf31ea",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/980 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "114f0c095e9d4089b5b4c9ccadcbc195",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/980 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "0fc4c519bbc841a49adc30ed269ebe64",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/980 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "2c92183f50804085869c7cb716a885ee",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/980 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "aab71c0f01d34a1fb5a3a26b4dad059b",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/980 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "58e7204619e448f3a95aa08f4bca2a9e",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/980 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "015d977d391745c09a402f7bb3272b30",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/980 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "6dc829a975224b38843055b9bca1215e",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/980 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "c375e559a3924159afbece26dfa52cd1",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/980 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "415843cd890b4e289616615ce97c6f14",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/980 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "cac9f947a14e4bbdb6eb7f065d350365",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/980 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "0c217b63d0e6409dbba5f4635026dd86",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/980 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "c9669bb9e5b94945a7e060765d160803",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/980 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "4d3a9f02c1bd4695a317bef2dc5cec6e",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/980 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "645fcef7f3cb4661a95645090349c44b",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/980 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "ddcf8ed9cf014659a9e2a181f14cf461",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/245 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Loss Decreasing.. inf >> 0.009 \n",
      "Epoch:15/15.. Train Loss: 0.014.. Val Loss: 0.009.. Train mIoU:0.815.. Val mIoU: 0.878.. Train Acc:0.995.. Val Acc:0.997.. Time: 43.41m\n",
      "Total time: 592.46 m\n"
     ]
    }
   ],
   "source": [
    "max_lr = 1e-3\n",
    "epoch = 15\n",
    "weight_decay = 1e-4\n",
    "\n",
    "criterion = nn.CrossEntropyLoss()\n",
    "optimizer = torch.optim.AdamW(model.parameters(), lr=max_lr, weight_decay=weight_decay)\n",
    "sched = torch.optim.lr_scheduler.OneCycleLR(optimizer, max_lr, epochs=epoch, steps_per_epoch=len(train_loader))\n",
    "history = fit(epoch, model, train_loader, val_loader, criterion, optimizer, sched)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 43,
   "id": "67981cca",
   "metadata": {},
   "outputs": [],
   "source": [
    "torch.save(model, 'Unet-Mobilenet.pt')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 44,
   "id": "deddce1d",
   "metadata": {},
   "outputs": [],
   "source": [
    "def plot_loss(history):\n",
    "    plt.plot(history['val_loss'], label='val', marker='o')\n",
    "    plt.plot( history['train_loss'], label='train', marker='o')\n",
    "    plt.title('Loss per epoch'); plt.ylabel('loss');\n",
    "    plt.xlabel('epoch')\n",
    "    plt.legend(), plt.grid()\n",
    "    plt.show()\n",
    "    \n",
    "def plot_score(history):\n",
    "    plt.plot(history['train_miou'], label='train_mIoU', marker='*')\n",
    "    plt.plot(history['val_miou'], label='val_mIoU',  marker='*')\n",
    "    plt.title('Score per epoch'); plt.ylabel('mean IoU')\n",
    "    plt.xlabel('epoch')\n",
    "    plt.legend(), plt.grid()\n",
    "    plt.show()\n",
    "    \n",
    "def plot_acc(history):\n",
    "    plt.plot(history['train_acc'], label='train_accuracy', marker='*')\n",
    "    plt.plot(history['val_acc'], label='val_accuracy',  marker='*')\n",
    "    plt.title('Accuracy per epoch'); plt.ylabel('Accuracy')\n",
    "    plt.xlabel('epoch')\n",
    "    plt.legend(), plt.grid()\n",
    "    plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 61,
   "id": "f817fdef",
   "metadata": {},
   "outputs": [],
   "source": [
    "##plot_loss(history)\n",
    "##plot_score(history)\n",
    "##plot_acc(history)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "1eecd4f1",
   "metadata": {},
   "source": [
    "### Evaluation of Model"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 47,
   "id": "ecab96df",
   "metadata": {},
   "outputs": [],
   "source": [
    "class MetalTestDataset(Dataset):\n",
    "    \n",
    "    def __init__(self, img_path, mask_path, X, transform=None):\n",
    "        self.img_path = img_path\n",
    "        self.mask_path = mask_path\n",
    "        self.X = X\n",
    "        self.transform = transform\n",
    "      \n",
    "    def __len__(self):\n",
    "        return len(self.X)\n",
    "    \n",
    "    def __getitem__(self, idx):\n",
    "        img = cv2.imread(self.img_path + self.X[idx] + '.png')\n",
    "        img = cv2.cvtColor(img, cv2.COLOR_BGR2RGB)\n",
    "        mask = cv2.imread(self.mask_path + self.X[idx] + '.png', cv2.IMREAD_GRAYSCALE)\n",
    "        \n",
    "        if self.transform is not None:\n",
    "            aug = self.transform(image=img, mask=mask)\n",
    "            img = Image.fromarray(aug['image'])\n",
    "            mask = aug['mask']\n",
    "            \n",
    "        if self.transform is None:\n",
    "            img = Image.fromarray(img)\n",
    "        \n",
    "        mask = torch.from_numpy(mask).long()\n",
    "        \n",
    "        return img, mask\n",
    "\n",
    "\n",
    "t_test = A.Resize(480, 480, interpolation=cv2.INTER_NEAREST)\n",
    "test_set = MetalTestDataset(IMAGE_PATH, MASK_PATH, X_test, transform=t_test)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "1641865a",
   "metadata": {},
   "source": [
    "### Results"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 48,
   "id": "fe0e3485",
   "metadata": {},
   "outputs": [],
   "source": [
    "def predict_image_mask_miou(model, image, mask, mean=[0.485, 0.456, 0.406], std=[0.229, 0.224, 0.225]):\n",
    "    model.eval()\n",
    "    t = T.Compose([T.ToTensor(), T.Normalize(mean, std)])\n",
    "    image = t(image)\n",
    "    model.to(device); image=image.to(device)\n",
    "    mask = mask.to(device)\n",
    "    with torch.no_grad():\n",
    "        \n",
    "        image = image.unsqueeze(0)\n",
    "        mask = mask.unsqueeze(0)\n",
    "        \n",
    "        output = model(image)\n",
    "        score = mIoU(output, mask)\n",
    "        masked = torch.argmax(output, dim=1)\n",
    "        masked = masked.cpu().squeeze(0)\n",
    "    return masked, score"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 49,
   "id": "bec57ed2",
   "metadata": {},
   "outputs": [],
   "source": [
    "def predict_image_mask_pixel(model, image, mask, mean=[0.485, 0.456, 0.406], std=[0.229, 0.224, 0.225]):\n",
    "    model.eval()\n",
    "    t = T.Compose([T.ToTensor(), T.Normalize(mean, std)])\n",
    "    image = t(image)\n",
    "    model.to(device); image=image.to(device)\n",
    "    mask = mask.to(device)\n",
    "    with torch.no_grad():\n",
    "        \n",
    "        image = image.unsqueeze(0)\n",
    "        mask = mask.unsqueeze(0)\n",
    "        \n",
    "        output = model(image)\n",
    "        acc = pixel_accuracy(output, mask)\n",
    "        masked = torch.argmax(output, dim=1)\n",
    "        masked = masked.cpu().squeeze(0)\n",
    "    return masked, acc"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 50,
   "id": "8b38563c",
   "metadata": {},
   "outputs": [],
   "source": [
    "image, mask = test_set[3]\n",
    "pred_mask, score = predict_image_mask_miou(model, image, mask)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 51,
   "id": "7762332a",
   "metadata": {},
   "outputs": [],
   "source": [
    "def miou_score(model, test_set):\n",
    "    score_iou = []\n",
    "    for i in tqdm(range(len(test_set))):\n",
    "        img, mask = test_set[i]\n",
    "        pred_mask, score = predict_image_mask_miou(model, img, mask)\n",
    "        score_iou.append(score)\n",
    "    return score_iou"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 52,
   "id": "1a6520de",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "f9dcc1978a334474a43735c137ce623e",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/409 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "mob_miou = miou_score(model, test_set)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 53,
   "id": "11e138cf",
   "metadata": {},
   "outputs": [],
   "source": [
    "def pixel_acc(model, test_set):\n",
    "    accuracy = []\n",
    "    for i in tqdm(range(len(test_set))):\n",
    "        img, mask = test_set[i]\n",
    "        pred_mask, acc = predict_image_mask_pixel(model, img, mask)\n",
    "        accuracy.append(acc)\n",
    "    return accuracy"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 54,
   "id": "353655d8",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "b6c3473ce15645d68ab409d3084206ac",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/409 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "mob_acc = pixel_acc(model, test_set)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 55,
   "id": "de1b20d6",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1440x720 with 3 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "fig, (ax1, ax2, ax3) = plt.subplots(1,3, figsize=(20,10))\n",
    "ax1.imshow(image)\n",
    "ax1.set_title('Picture');\n",
    "\n",
    "ax2.imshow(mask)\n",
    "ax2.set_title('Metal Artifact truth')\n",
    "ax2.set_axis_off()\n",
    "\n",
    "ax3.imshow(pred_mask)\n",
    "ax3.set_title('UNet-MobileNet | mIoU {:.3f}'.format(score))\n",
    "ax3.set_axis_off()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 56,
   "id": "6ce9db4a",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAABIQAAAF3CAYAAADU7JAmAAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADh0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uMy4xLjEsIGh0dHA6Ly9tYXRwbG90bGliLm9yZy8QZhcZAAAgAElEQVR4nOy9fZhc2V3f+T31XtVV1S/qbrWkmZE0GmvsmeDxOsSELLvAYgIevczYY2N7eMBeICSb3YfNC5slu5CQBDYhm0AIyYaEQAwkkICX4DcggfCQALGz4MRJMINnNLI0oxlNS+pu9VtVV9WtOvtH1e/Or359zq3WjKSR1N/P89RTVfeee97ureq+3/r+fsd570EIIYQQQgghhBBC9g+5N7oDhBBCCCGEEEIIIeT2QkGIEEIIIYQQQgghZJ9BQYgQQgghhBBCCCFkn0FBiBBCCCGEEEIIIWSfQUGIEEIIIYQQQgghZJ9BQYgQQgghhBBCCCFkn0FBiNyzOOd+2Tn3oTe6H4QQQm4PzjnvnHvoFrcx9rfFOfd9zrlrzrlXbmW7twrn3Iedc7/1RveDEEIIIbcfCkLkrsc5d8E513bObTnnlp1z/8Q5V/fev8t7/5N7PP6dt6OvhBCyHxh9r3adc/Nm++dGos2xPdTxVc65S7egbx9xziXOucN7KPu9zrl/qrfpvy3OufsB/HkAj3jvl15Hn46N5qVwI325Fe0QQsjdyiSBWwv6d5oYPurPR97ofpD9BwUhcq9wxntfB/B2AH8EwHffrob5jzUhhAT5IoAPyhvn3JcAqL5x3QGcc1MAngKwDuAbJ5Tdy3f7UQAr3vsrN6F7rws3hP/XEUJuiJCzUgvQI3HeO+f+vinzW865D7/WNgJlXvcPCZPY64/Fgb59ZNSHd6htDznn/B6Pv63i042I/7EfG7LOmXNuzjn3L51z2865i865pzPqn3HO/aRz7sro8b2Rcl85avP7zPY/65x7xTm37pz7CedcedKYyI3BfxzIPYX3/iUAvwzgDznnfsM5922yzzn3J5xzzzjnNp1zv++ce7tz7qcBPADgEyOH0V8I/SqtXUSjL86POuf+qXNuA8CHnXM559x3Oeeed86tOOd+zjk3dxuHTgghdxo/DeCb1fsPAfgpXcA5V3bO/S3n3Asjh+ePOueqI+HmlwEcHn03bznnDjvn3uGc+7Rz7rpz7rJz7u8550o30KenAFwH8FdH/dF9sd/tfwrA/wHg/aP2//Oo3G84575t9DfhV1UfPzLa//Pqn9d/55x7VLVRdc797dE/0OujG6oqgH83KnJ9VNeXm759fUZfvt8599sAWgAedMb1av7Zj7YzOg9rzrkvOufedQNzSgi599kG8M03Q5SZwB33Q4JiFcD3TSy1P/j7ALoADmL448o/0H/rDD8EoAbgGIB3APgm59z/qAs454oAfhjAfzDbvw7AdwH4mtHxDwL4KzdrEGQIBSFyT+GG9v3HAfwns/19AL4Xw5uTJoCzGP6q+00AXsDIYeS9/5t7bOoJAB8FMAPgnwH4DgBPAvhKAIcBrGH4ZUkIIfuVzwBoOufe4pzLA3g/APsr5A8AOAngbQAeAnAEwF/y3m8DeBeAl0ffzXXv/csA+gD+LIB5AF+O4T+Jf/oG+vQhAD8L4J8DeLNz7u1mv/5u/3EA/xeAfzFq/zFd0Hv/a6aPHx7t+mUAbwKwCOA/Yvg3QvhbAP4wgD8GYA7AXwAwAPDfj/bPjOr6tGnrVzL68k0Avh1AA8DFCeOPtfNlAL6A4bz+TQA/7pxzE+oihOwfrgP4CIC/HCvgnPuW0Q+va865f+WcOzraLkL0fx4J0e/PaGcvPyRMO+d+yjl3dSSuf7cbd0c659yPjET3P3DOfY3aMfZjsan3zc65X3XOrTrnvuCc+wZT5CcBvNU595WR46edcz8++rHiJTfML5d3zr0FwI8C+PLR+K9njD/KSOz/35xz/8UNnTk/7pw76IZhcJvOuV9zzs1m9C1rzm6kH+K0/R7v/Zb3/rcAfBzDv0UhzgD4m977lvf+AoZ/W7/FlPnzAP41gD8w2z8E4Me995/33q8B+GsAPvxa+k3iUBAi9wq/OPqC/S0A/xbDf5w134bhl9Hv+CHnvPeT/nHO4tPe+1/03g+8920AfxLA/+m9v+S972AoPr3XMZyMELK/kX/uvxbDf/Rekh0jweFPAPiz3vtV7/0mht/dH4hV5r3/rPf+M977ZPSP5T/EUIifiHPuAQBfDeBnvPfLAP4NjEsIu7/bbxjv/U947zfV34LHRv+M5zD8J/h/9d6/5L3ve+///ajc6+Ejo3+WE+997zXWcdF7/2Pe+z6GNz2HMPzllxBChO8H8JRz7mG7wzn3JIYuxvcAWADwmxiK7/DeixD92EiI/hcZbezlh4QfATCNoVvkKzH8G6MdJ18G4DyGAvdfBvALboJrfyRy/CqAn8FQzP8ggP/HuF5aGP6N+v5INT8JIMHwx43/BsAfB/Bt3vtnMHScfno0/pmsvkzgKQz/np7EUGj5ZQznfR7D+/rviBw3ac5uhJMA+t77Z9W2/wwg5hACAGde/6H0zVA4/BYMnbuWR0d163YOOucO3GinSRwKQuRe4Unv/Yz3/qj3/k8H/pG/H8DzN7G9F837owD+pRuGMVwH8AyGv2TzH2pCyH7mpwE8jeEvej9l9i1gaCP/rPru/JXR9iDOuZPOuU+OQrI2MPznfD5W3vBNAJ7x3n9u9P6fAXh6ZFUX7Hf7DTH6NfhvuGH48AaAC6Nd86NHBTf3bxHwOvs8Il0hzXvfGr2s34R6CSH3CN77VzB0uoRu3P8kgL/uvX/Ge59g+N38NnEJ3SBZPySISPQXR8L7BQB/G+PulCsA/o73vjcSn74A4NSENk8DuOC9/ycjcf0/Avh/AbzXlPuHAB6wYbXOuYMYOkb/jPd+e5RX7oeQ8QPHa+RHvPfLoxQZvwngP3jv/9Poh4V/iaEQNcYe5+xGqGOYh0+zjqFLNcSvAPgu51zDDXMSfQuGf/uFv4uR22gPbcnrWFvkNUBBiOwXXgRwIrLPJoTbhvqiGn2R2hsUe8yLAN41EqXkURl9YRNCyL5k5MT8IoahvL9gdl8D0AbwqPrenB4tEADs/p4FgH+A4Q3Cm7z3TQx/Gd1raNM3Y5hj5xU3XCL+BzEUafQ/9rbNPSUMVTyNYdjZOzH8NfbYaLvDcLw7CP8t2ks7sTKZf8MALGWUJYSQPoCi2VYEEHIc/gCAr3POPWa2HwXww0rcX8Xwe+9IqMFRmJPkh7MJ/rN+SJgHUMJ4eOxF085L3ntv9k9aVfIogC+T/o/G8I0Y//7ESHj5a6OHM8cXAVxWx/9DDN1GN5Nl9bodeB8S8ifNWQJz/tUPJaFrYAvD9BuaJoDNSJ+/Y9S35wB8DEPn2KVRO2cANDJcY7YteR1ri7wGKAiR/cI/BvCdzrk/7IY8pH61WMbQQik8C6DinDs1+kL8bgCTMtr/KIDvV/HSC865J27yGAgh5G7kWwH8D6O8QCne+wGAHwPwQ865RQBwzh1xwySSwPC7+YBzblod1gCwAWDLOfdmAP/TXjrghsmTT2CY0PJto8cfwjA8wIaNaZYBHLuBXAsNAB0AKxiKMmn48mi8PwHgB90wQXbeOfflbrhiylUMcwk9GKjzRvvyOQAfcM4VnXNfivFfuPfSDiFkf/ECXhWvheMI5CTz3q8A+DsYCiKaFwH8SfPDaNV7/+9DDY5W+5L8cP/M7Jv0Q0IPQwFGeADKRQTgyCgkWe9/OdQP0/9/a/pf996H/sb8EwwF/3eb4zsA5tXxTe+9hFG9kWL8pDmLnf8+xudVeBZAwTn3JrXtMQCfDzU+Cgn/Ru/90mg+cgD+v9HurwHwpeqHmvcD+DPOuY+N9n9+VLduZ3l0HZKbBAUhsi/w3v88hjG/P4OhqvyLGCb0BIC/DuC7R4r+d3rv1zFMUvqPMfwi3MZIyc7ghzFMqPavnXObGMZAf9lNHwghhNxleO+f997/bmT3/w7gHIDPjEKsfg3Aw6Pj/gDDXxLPj76fDwP4Tgx/Od7EUEzKykWh+RCAj3nv/6v3/hV5YPjdfTojv8TPj55XnHP/cQ/t/BSGN1EvAfh9DP8WaL4TwH8F8DsY/oL+AwByozCt7wfw26Ox/tHX0ZfvwVD8WsNwNZafkR17bIcQsr/4Fxj+H3yfG66a+04M89N8NFL+BzFMjP8Wte1HAfxFybkzypv2PrXf/vg6idgPCX0AP4fhj7CN0Q+xfw7jeYYWAXzHSBR/36ifvzShvU8COOmc+6bRcUXn3B9xw4TQY4xC4r4Xw79fsu0yhkmR/7ZzrjmaxxPu1QTUywDucze2KuZNYQ9z9isAHlZjn8Pwx4yPjsZq69vGUKj7q865Kefcf4uhM/anQ+2P5uHA6EeQd2G4CIKs1vY9eHVhibdheC/1Y3g1v9FPAfhW59wjbpgw+7sxTG5Obibeez744IMPPvjggw8++OCDDz722QPDZd3/bwxznq1juDriWbX/qwBcMsf8BQxdLx9W274JQ8F7A0PHzE+ofX8KwGUMVyv7hkg/LgB4Z2B7YdTWsdH7WQzFjKujdv4ShsI6MAwz+20Af280lmcB/HFV129gmOhZyv6W2vcwgE+N6l0B8OsA3jba9xEA36fK5gD83vBWOt02jWFY86VR2/8JwAdG+0qjulcBXIuM/8MYLhIQO09j8zOag+9V778NwK+NXh8bzVlh0pyN9v8xDBfmWcPQTfXjAGYz+jKH4Y/r2xg6jJ5W+/47AFvq/TeM6mxh6GD9uox6x+Z5tO3PYSiobWDoziq/0Z+Ze+3hRhNNCCGEEEIIIYSQ24xz7sMAvsp7/+E3uCtkn8GQMUIIIYQQQgghhJB9xi0RhJxzX++c+4Jz7pxz7rtuRRuEEEIIIYQQQsg9wOcwDMMi5LZy00PGRkt0PwvgazGMofwdAB/03v/+TW2IEEIIIYQQQgghhLwmboVD6B0Aznnvz3vvuwD+OYaZxwkhhBBCCCGEEELIHUDhFtR5BMPs5cIlBJbfds59O4bLzgHAH74F/SCEkHsC7717o/vwRvO1ufdxBQRCCInwq4Of598J/p0ghJAosb8Tt0IQCjW06wvae/+PAPwjAHDO8QucEEIIIYQQQggh5DZxK0LGLgG4X72/D8DLt6AdQgghhBBCCCGEEPIauBWC0O8AeJNz7rhzrgTgAwA+fgvaIYQQQgghhBBCCCGvgZseMua9T5xz/wuAfwUgD+AnvPefv9ntEEIIIYQQQgghhJDXxq3IIQTv/S8B+KVbUTchhBBCCCGEEEIIeX3cipAxQgghhBBCCCGEEHIHQ0GIEEIIIYQQQgghZJ9BQYgQQgghhBBCCCFkn0FBiBBCCCGEEEIIIWSfQUGIEEIIIYQQQgghZJ9BQYgQQgghhBBCCCFkn0FBiBBCCCGEEEIIIWSfQUGIEEIIIYQQQgghZJ9BQYgQQgghhBBCCCFkn0FBiBBCCCGEEEIIIWSfUXijO0DI7eIbvuEbAACDwSDd5pyD9x4A0udcLgfvPZxz6Pf7aRnn3Nhxwkc/+tHb0X1CCCGEEEIIIeSmQUGI3DN8+MMfRqfTwWAwwGAwgHMOxWIRzjl0u130ej0AYUHIOYfBYADvPfL5PLz3yOVy6Pf76X4RioBXRaNcLoennnoqbU/K6eef/dmffUPmgxBCCCGEEEIIiUFBiNwVvP/970/dOuLO0a6dwWCAVqsFAGMizmAwQL/fR5IkqZgD7HYG6feCCEu6Lr3PPuu+5XLDaMx8Po+nnnoqbd85lwpOg8EA+XweAPALv/ALN3fCCCGEEEIIIYSQDCgIkTuW973vfalw0uv1kCQJgPFwrVwuh0JheBkPBoPUlSP7+v0++v3+rrq997vCwGR7CCtCWTFJ16kRwUf2i8Bk+/Dkk08il8uhXC7TUUQIIYQQQggh5JZDQYjcdj74wQ+mooh27ojQIqJJt9tNnTdabNGCTCikK5/Po9frpSKSFl5CAo4Ve0JtWrFH75d25Xide6jb7SKXy40JVboucQnJvp2dHZw9ezYdR6FQgHMOSZIgSZLUeeS9x8c//vFbd5IIIYQQQgghhNzTUBAit40PfOADqdtHhBwJuRJBB3jVpZPP58fEFl1WnnUYlxwrIWJ2nxyjRaCQQ8hus8eH6rOJqeW1fojwI4KTriMkFomwZOdA5uk973kPcrkck1oTQgghhBBCCLlhKAiRW8LTTz+Nfr+PXq+X5vHZ3t7eJaAAGEvgrPcXi8Ux14120mgBJZ/Pj7mKtAhjhZeQcKPf63on7bPCUshtJK/7/f6ukDbtiLJ9ERFI5zDSwhAA9Ho95HI5PPnkk2Pi2Mc+9rEbOleEEEIIIYQQQvYfFITITePpp59OxYtut5u6dEICkH0dKqPDwYDduYOAV5M5S2iYztFjsfVnuYNC7caWpxcBJ1avxrqDQqKVdgzpcrlcLk10La+lrIzdOYczZ86kwtMv/uIvRvtCCCGEEEIIIWT/QkGIvC6+8Ru/EQDQarXQarWiSZPlvWCFEJ1TR8QVcQzJdl1e9onzJhYeZtu1+2JhZzHRxq5GluVCsvVr95PNaSRjtk4j2S77xEllBSLrfpL5OX36dJrDaDAYMO8QIYQQQgghhBAAFITIa+Tpp59OQ7W63W5UvIm5ZbJW84q5iWJ1hBI+7+V43VYonCy0FL0OYbPbQ0JRVl32eHEcZaHFI2BcQIutmCaClPceZ86cgXOOwhAhhBBCCCGE7HMoCJE989RTT6WhSe12OxUiJE8QkO3Kke0x4WSv+X409hhbp2y3DiPbPvCqG0cLLqG69XHy0PWHHEe2Dnu8hHjpcDAA6UpjAFAqlcZcQQDSsLwkSdK2retJv5d2T58+nbbLnEOEEEIIIYQQsv+gIEQyee9735uGIPV6vTHxRkQGnfgY2C3ATHLLWJFlUh4eS6h8zCUUSgitt8dWEYu1a+vfC6GxhfoifSwWiyiVSqjX62libV2m1Wphc3MTnU5nbO5jya/1tsFgkC5z/4lPfOKGxkEIIYQQQggh5O6FghAJ8oEPfABJkqDT6aTbtLCgcwXFcgR5P1wBTLB5ePYSHhVL2hxz/2SFm+n3ewnPsmMOhaXFQshC+61zR+ZHi2BWgCqXy1hcXMSRI0cwPT2NXC6HQqGQ5hIqFApYX1/H8vIynn32WXS73bE2xFFkxbiQq+ns2bPpfoaUEUIIIYQQQsi9DQUhMsb73vc+5HI59Ho99Hq9dLteEj6Wq0aLDlmhY1rkuRF3TZbAEyobEp4m5fKxx04qo/sRcz7ZectyQdm8P7lcDpVKBfV6HbVaDc45FAoF5PN5dLtd5PN5NJtNdLtdlMtlJEkysf7QPOoHAJw5c4aOIUIIIYQQQgi5h6EgRPDe974X3W4XANL8NEmSjCWKjq26FQq9CokpWXmCQvVZx0xWfbHQLSvOZCVfjglAoRxFoXxEep5i49dIziC9Xx79fj/dJi6tTqeDZrOZ1i3nJkkSVKtVHDp0CMvLy3jllVdw/fr1TAdXaO5kXHo+Tp8+PXYeKBARQgghhBBCyL0DBaF9yrvf/W4Aw6TEOzs7Y2LHYDDY5QgKPYeICT8hx1BoqXhNLEQr1m5WKFpIwImJSSExyjqadPhXLFeSrleXl3pEWLL91tt2dnawurqKSqWC+fl5eO/x0ksvYWtrC51OB/l8HsePH0e9XscDDzwA5xy2t7fR6/VS4Sg0pzERLyT0yTGSiJrCECGEEEIIIYTc/VAQ2oe85z3vSQUDLQyIUJEVKpWV/HhSSJklFDYWE5tC9YbcS6H+hsQe239bnxZwdD9Dzp/QnITqD5Wx20IOo52dHWxubqZ9v3z5Mq5fv45+v49cLoeZmRkUi0VMTU1henoaxWIR/X5/TLAKiWux97GxyDycOXMGACgMEUIIIYQQQshdDAWhfcK73/1u9Pt9eO/HlokXJLmxlAEmr6IVctjEQr9sPaHQsZgwEQrbivUjVCYkMoWEHUnUHHJEhfppRbC9hNWF3EG2X94Pk02LU0tcQhsbG2g0GlhfX8fq6moqCD333HPodrt48MEHsbS0hMuXL+Pq1avodDrpcvRZQptO2J01z4KEo505cyYdG8UhQgghhBBCCLm7yL3RHSC3nne/+92pCBFyfQg2RErKhLD5aEJCSUwg2Qv6+JgDZ5LTJSbm2MekerPa2Ev50FxowUk7kWz/5bnb7WJrawv9fh/VajUVlvr9PjY3N7G2tgbvPSqVCmZmZjA1NZWuRrZX9FzEhLzQmL33eOKJJ/bcDiGEEEIIIYSQNx46hO5RtAgkq4VpgUFu+AeDQeoOArBrWfRYWJZ2nOhjQyJQSDSyglPIpRJqLyS+WIeOuF4kH1LomNB2mwNoLyKRrSeWQ2iSA8q2KeKdnKNut4uXXnoJpVIJR44cwdbWFlZWVuC9x/b2Nl588UXcd999mJ+fxwMPPIBCoYBLly6h1+shSZKx86rb06FwVtCTfuiwvpCQBQxzUZ06dSot+8lPfjI4VkIIIYQQQgghdwZ0CN2DnD17Fv1+PxUCrBMoK0dQaH8ohMzmvAmFP2kmuYNCOXSs0BIKHQu1LSt42edJfZxUb6z8XlxDsfnPcjlZx8729jY2NjYwNTWFRqMxtq/b7WJlZQXtdhvlchmzs7NoNBrI5/NBYW/SOY3lS5o0F3IMHUOEEEIIIYQQcmdDh9A9xBNPPLErNCzkygmJJwB2rUilj4k5YWLOF+t6iTl7bH9C+Xh0X2V/oVCA9x7FYhHFYhHlchm5XA7FYjF1G+Xz+TTPjnPDJdol2XK320W32wUAtNvtdCyx8K29CkmhY2VbyFFlVy6TvsscSA4g7z02Nzdx+fJl3H///Th69CguXLiQznOSJPjiF7+Ifr+PRx99FLOzs8jlclhdXU3Hbc9DzAEWc0pl5WGyjrFer4dTp07hU5/6VOa8EUIIIYQQQgh5Y6AgdI/w5JNPTkwgLGSFB9ljJ9Vl6wuJPCFi+WmywqpyuRzy+TwKhQLK5TIKhQKq1SpqtRpKpVJQENJClogiSZKg1+ulSZc3NzeRJAl2dnbQ7/d3hdhJ+1osynIPaZeTHVdsPuUYyfsjYxDBSBKBt9tt9Pt91Gq1XXW0Wi2srKyg1+uhUChgamoKU1NTuH79+tg50quP6fGEXEF2FThBi4EhN5Ts41L1hBBCCCGEEHJnQkHoLkdcQXZ1sFBeHS1qCIVCIbg/64ZfY90kIaHH5qsRsvISAUCxWIRzDo1GA6VSCbOzsyiVSigWi6jVaigUCtGQsZC4FerDYDBAkiRpSFa328WVK1fQarXQ7/fRbrfHRCVx9cRyBcXyCem8O6F5kPNQrVbR6/XQ6/VSEaxUKqFaraLT6aDVauHKlSs4dOgQCoVCKl4Bwzw+y8vLuHTpEg4fPoxSqYTjx49jbW0Nm5ubu4Qg6ZN1KcUEKxlnzOmky+Tz+bF9FIYIIYQQQggh5M6CgtBdyBNPPJGGhWkhSBCBJCSK2Jt/ER0kAXMs38xeBJAQk8QkKSOv8/k8qtUqSqUSDh8+jGq1mq6YJaKJ7pOEuemQKJvwWLcRCnMT4alUKsE5hwMHDqSizOrqKjqdDq5fv45Op4N2u73LTRNayUv3LSSS6WTO4nyqVCrY3NxEv98f61+320WpVEKtVkOn08Hy8nLq/pHyeg7OnTuHJElw3333YWFhAYuLi2i1WhNDv/R51+JRKOQvJAICCB4n1+FgMGDiaUIIIYQQQgi5Q6AgdBdx9uxZDAaDsVw/e01WbMvIzbqEJ9n8QTHRQIsAWnAKhUnpvthwJBFCpA+FQgFzc3Oo1WqYn59PXUAaEUp0uzaMS7ejBQktwNhxSd/F1VIsFlEqleC9R6PRgPevLvsu7qHNzc00jMuOVddr59oKbtJmpVIZ66sdY6fTQaPRQLlcxvXr13H58mUcPnwYFy9exPb29tiYr1+/jvPnzyNJErz5zW/GiRMnsLy8jJ2dnTRvkp2rkOATO58xgTEk8Onx6pXkBoMBTp8+DQAUhgghhBBCCCHkDYCrjN0lnDlzZuwmWyeOtmTl5ZmU/Fi22zCs0DYbemSdJNYZY4UEACiXy5iensb8/DwOHjyIgwcPol6vo1qtAsCYc0mP2ybOntRmSJCy70Vs0/WLaFOr1TAzM4NDhw5haWkJzWYTtVptl4MmNqd2fjUiytncRbZ/SZIgn88jSRJsbW2h2WyiXC7vEnMGgwFarRZWV1fR7/dRr9dRLpfH5ig2Z/Y8ZrmDQnNo59eKQ7acFoYIIYQQQgghhNw+6BC6Czh79mzU3RIjFOYV2i9kJUgOtZfVdlY4kYhKxWIRhUIBBw4cQL1eTwUXWT3MOoBi7h8rZljHSiicKxb6ZlcY064eYOgcmp6eRrVahfev5hySZN527KE5DM2zHJMkSVSsAYBOp5O23W630Ww2x5JG6zH1ej2sr6+j1Wqh2WyiXq9ja2srKGCFRB87D3rObXnrENPH2XrtOSSEEEIIIYQQ8sZAQegO58yZM8FwrpjwEArV0ccJOnwndAMfuvkPuT1CTiLbTxFVJD9QpVLBwYMHUavVMDs7i3w+PyYAyXhjYo4WIKRunfhZP+swLZvTR46R0DWd68aKJtIXyWV03333wXuParWK7e1tbGxsYGNjY6IQp4UTcQaVSiW02230er2gs0bY2dnB5uYmKpUK2u02Go0GDh8+jMuXL4+FmMlxW1tbeO655/D2t78dDz/8MAqFAr74xS/uErBiWIFNXw+h51AZe03o8DE9xwwfI4QQQgghhJDbCwWhO5QzZ84AwK4cNULMpSGvNVnikRYQbJ6dUBv6+Bi2PS3gzM7OotFopHmCRGjRQlDMQaLFJxFxrAgU6psVkayglDUm6xrSIhIALCwsoF6vo1KppG4hWbUsSxiSuorFIvL5/Fhy8Njc9vt9JEmCqamp9Phmszk2BuuuWl1dxc7ODprNJg4cOICLFy/uEo7sHNtrYC9jySIWNhZyKp05c4YrkRFCCCGEEELIbYCC0B2GhIdpIcjeOMdu6EPuHX2MdtEAYQeODbkKOT90ezoBskbe5/BGeqcAACAASURBVPN55HI5HDhwAFNTUzh69GiaSNl7P+Z+smKL3qa3S+JlPVYbWqbFLe3GiYlVFmlTu4aAoSgjdedyOUxNTaFer2N+fh7NZhPr6+t4+eWX0e1206XjteCm+wwMHUcyh7YvIcGv1+shn8+nyaXn5ubQaDSwurq66zwMBgOsra3h4sWLePTRR3H8+HG88MILWFlZQafTCY45JPrtJWQsdI3avsdEQn2d9fv9sZxCdAwRQgghhBBCyK2BSaXvIGQVsSwnTigHjBATgrJcPlow0dtjAo/UoUWjkNAhx1SrVczMzKTLn4sYJMmibX0aK35p0Ucnmg7NR8hlFHqt6wmN186XdVbpOZ+bm8PBgwcxNzeHqampdKwhrPAF7A6jsu3LanDiLNrc3IT3HrVaLdPpdPXqVfR6PRSLRczNzaFcLgfPd5brS5eTfsXmJhYuZgm50fZ6LCGEEEIIIYSQ1wcdQncA4oiQZdWBcOiObBdioVFZAo0VRaSszrejy1uBKhTmY/snq3LV63UcPnwY9XodpVIJzrk0f03I6WRFGD2OkFvFtmnDrmxZLVjYlctk/FYYkW2CXQ5eHEO5XA75fB6NRgMPP/wwdnZ2cOnSpTS/kOQHsvXrfEmDwQD5fD7T+eW9R6vVQqPRwPLyMhqNBg4dOoStra00ubQ+vt/v48qVK7h06RLuv/9+nDx5EoVCAb/3e7+3S1DLEuUmbZ/kbMp6BjAmFGpOnz4N7z0+9alP7eoDIYQQQgghhJDXDh1CdwjWdRJyX1hRxDoqQiE7sRt6TUgAsQKJFTJC20WYqVQqWFhYwMGDBzE9PY1SqZQKKHpc1p0TcuqEBJ7YPMTK2KTVsj3Wrn6vBaGYMKWPEfeOCDVLS0uo1+uZjq4sV5BuS/qeJAn6/X6aZHpqagq1Wi14rPfD1ctWVlZw/fp11Go1LC4upsvVh4Sd0Lm2cxO7pmLiUey4kGgXqkNyahFCCCGEEEIIuTm4OyEswzn3xnfiDUCcQbEQmVB4mBYyYk4gjRVURFSQYyXHj9QbC1sS940N88rlcigUCpiZmcGRI0cwNTWFSqWSLh8vjiDbbqjP4o4JhQ1Z8UfnEdKCyaSQKykXW31M5/PRziErVoQcTroO29/19XVcuHABW1tbaLfbY30rl8uoVqvY2tpKQ8JkbkN5n7z3qNfr6Pf7qNVq+Iqv+Aqsr6/jN3/zN9Htdsf6JmMuFouo1+v4qq/6KkxNTeH8+fM4f/48rl69Ouasil0/MbLEyBAhd1Ds+railOzfj3mFvPfZqu4+4Gtz79uXfycIIWQv/Org5/l3gn8nCCEkSuzvBB1CbwCnTp1KQ2FiLpGQqBEKJ9vLDXjs5lu7WkJOFy2wxFb1yuVyqetkdnYWtVoNhUJhrE7tOgmJXtYVovMLxVxKdmWtkLtIz50WdGw/QkKPhIHpOmzeIiuO6Xnr9/vpAwCazSYOHz6MxcXFdHU1LWTlcjkUi8WxMYeQvkioWq/XS1cek+ND5QeDAVqtVhpWduDAAczOzgbn0b6WOQrNV+j86XbtNWv3h9rSx4XEUp10mhBCCCGEEELIa4M5hG4z2hVkBZOYGyMmYMTEJCkDYJdoYY/Xq2fpm3sRNiTcK1RHs9lEs9nEAw88gKmpqfQ4cZyEcgBZ0UHnzJE+iEAS6qceGzBcpUvas6KSJRROpZ1G+hgdZiZOJ+mbdlqJw6rf7+/KgyPH93o9FAoFLC4uYn5+HtVqFSsrK9jY2ECn00GSJBgMBqhWq+h0OrvGY8VB5xw6nQ7m5ubQarWwvLyM+++/H81mEzs7O+k86rmWOp977jn0+30cP34cjzzyCJaXl7G+vh5Nrh2au5i4qLeFQuT0/Omx6P3aJRQS6uS1fI72o1uIEEIIIYQQQm4GdAjdRqwrSAi5I2JuiizniBVurKgQutm3gpEOWYqJA7lcDuVyGQcOHMDCwkKav0a7irJWEbPigT5OhKRYLhopHxLF9LNdwUuLXLbOrFAzaSeUS8iKGdZFpedBO6bm5+dx8OBBNBqNtOzOzs5Y3boPtk/yWsS07e1t9Ho91Ov11CUUO3ZjYwNXr16F98PVyRqNBorFYqbLLFTfjWyzDq+Q0KXLhlxBWfUTQgghhBBCCLlx6BC6TZw6dWrXDX3MNaPLWLKEFb0t5LAIOYnsMSIKaaeQdfpUq1U0Gg3Mz8+nCZOt0KJdHratkPNDO25Cbidx4ejxWuFErwAWwubjseVCq6xJ/7JCq3Rf9HEiCIlwI2OsVqvI5/MYDAZYX19HkiTodrtjiZ713Oh2Ja+QPANAq9XCzs4Oms0mSqXSLnFJjvPeo9Pp4Nq1a9jZ2UGlUsGBAwewtbWV7ptETDjTYw+91315LcQEK0k2/YlPfOI11UsIIYQQQggh+xU6hG4DIgbpBxBeQcs6WyxWeBBigpJ1C2WVsyE7VigqFApoNpt46KGHcPLkSTSbTeRyuTTkKTQuLRRp949+SP4c61SxrptCoYB8Pj8msMiKW7Iv5MiRcnqfhLZZd1BIrLJ9lj7o+RTBRec+kqXopQ+6vXK5jMOHD+PEiRNoNBqp00dC9KwDyZ4f2TY9PY2NjQ1cvnwZS0tLOHDgQNqmLT8YDNDr9bC2toZz585hc3MTJ0+exMmTJzEzMzN2TYUcUDIuu20v12TIcRYaoz0XoXr0a5lPrkJGCCGEEEIIITcGBaFbzOOPPx51Bsl7/QzEQ2n06yyXhq03duOt643d3OvtzWYTCwsLaDabKJfLYwLIpD7IswglWeFDersOJQuFslnBKfQ65JSybYbCxrLCzLRQpB1UIeFD97/f7yNJkrTM/Pw8Dhw4kIo35XIZhcKrxr2sUK5erwfnhqukbW9vo1wuo9lsBstaIUWWoS+VSpifn8fs7OzEsLFYnfZ16JxmiT32OrCil36OOdy89zh79uwN9Z8QQgghhBBC9jMMGbtFhPIFZeWGie0TUcI6fLRjIxa2FKvHhoBpR5B1CeXzeZTLZUxNTeGBBx5ArVaD9z4N3xJCzqbQTb92CWl3jZTXdeu+WtePHb+Euem2rehkxSOpSxJCy6pgUqc8hwS90Hh1fVlldNLoWq2GY8eOod1uY2VlBYVCAY1GA2tra2N16rkQtre3AQwTa0ti6KWlJTzzzDO7hDorsF2+fBlJkmBpaQkzMzN4+OGH8fzzzyNJkrHj7Ng09roJXcNWCNJzGBKUQk6oSaFm+hydOXMGzjl8/OMfj5YnhBBCCCGEEEKH0C1BxCAgvuy2JuYskX3W4WP3x7aF3D+TQtLsjXilUkGz2cSBAwdQq9V2CQuakBNHtylzYEO3tNNIjo05g3T9Orm0oB1IeqwSRiV9t33W50AeItxY0UP3KVanfW3nSsQnEd0OHDiAer2OJElQKBSiIVt6biUUrlAowPvhSmjVajWYJFpfXzK2zc1NbG5uwvthgulisTh2ruwcheoMCTt7FXJCdWYdG5vLUHmGkBFCCCGEEEJINhSEbjI2XxAQXzZeXsdEmpA7xTopNFYgseE2sTpD5HI5lEolPPDAAzh+/DiWlpZ2CSF6fDq0So8pawUxCXcSR5AIJLJPjreuIS00ybLvOr+Q5PfR20Uo0nMlDiAbOqbnUEK8rIilE1jbHErSfyta6bHJQ8a/tLSEBx98EO12G91uN125LTTH+vzu7OykY1xeXkY+n8fMzExQ2NF96/f7aLVa+MIXvoCrV68CAA4ePIhqtbqrbNY1Y4UyvT8kZtqV50KfEyv+xcIS7bUiZXVIHiGEEEIIIYSQMBSEbhEhN4N1z9jXsTomhYLp1yFXxV5vjnWf8vk8arUams1mmtMmy8WUVZ8+di9hRfrZ9l+LTuKkkTokR491Fgmh5ext+JoNV9L9yRLorINHi2C2jK5D+pvL5VCv1+G9R6/XQ7lcTpeQz3LGSDv5fB6tVgu9Xg9TU1NjglwMWeXs6tWr6HQ6WFxcHHOBaUKOIU3MtWPLxNxuWfWHBKXYuLRr6YknngiWIYQQQgghhBDCHEI3jXe9611jYoIVAcThYUWPkNvG7g+FRoVEDClvkxzbkKosp4UIQUePHsXMzAyKxWK6Opbtr7Qbct4ASEOXYjmLZEl2Oc46p/RYRbgRkcSuFhZaFt46jvL5PAqFwtiqY3pM8j6UR0j3RcqIa8nOf0z00Mj1oM9ZsVjE4uIiVlZW0mXk19bW0nZkdTXb5sbGBqanp3H16lXk83mcOHECvV4Ply9fRrvdDvZRxnvt2jVsbGzg6tWr+Oqv/mosLS3hN37jN9BqtdDtdjPDs2Jio3b66OtYXwdZ8yJlpFxWfipBz7u03+v1cPr0aeRyOeYUIoQQQgghhBADHUI3gVOnTgVvuOV1SJixzpGQeyWEdZ1klbF9Cb0XpA/FYjF1BslKYnJcyJ0UEnCsCCPlJJxLL8mub+L1POhwMP1e6rerf01yWenjxF2kxxASuvbqcLFing2Li7nE9DYRr+r1OgqFApIkQZIkqaiW5RLq9XpIkiRdbaxUKqXOriynGvBqiNX6+jq895iamhpzhYWuydi1ZcuHws1C5Wy9oeThISeWbLfilHaRSRmuQEYIIYQQQggh41AQep3oZeWzQrs0WoSYFG5lxQUrYITCrCyhJdVtP0WwOXz4MI4ePYqpqanMm3L70C4X/RDxxwpeMnbZp8vbcUiOHJ2AWsQl2x8pWywWU0FJyomrR1xMpVIpbU8LT7o+KS/9sXl8dPuCHGOFE32+dUiZvO/1elhaWsLS0hKcc9jZ2cHU1BQqlUrw2pDXSZKg1WqhWCxia2sLuVwOi4uLmJ6eHmvX9kM7olqtFlZWVpDP5/Hoo4/i0KFDY4JdiCwB1G7X+0LuNivIZV0zds4ntSkuMopChBBCCCGEEPIqFIReB+IMyiLkzNACTSi8JnRs6L1ss8JSzA2iy2vkBrxcLmN6enpXQuPQsSGRSi/THnO06H0hUSUmQGjXh7QlZCWX1g9pX7uLJoke1uFjRZWQayrk9spyyui5KhaLmJmZSUPKvPeoVCqpY8fOm9Dr9cbC4Wq1GhqNRmabMjaZj2vXrqHdbqPZbGJ+fn5MMLNC5M0idq1Kv0LXXNZ4ss7lYDDg6mOEEEIIIYQQMmKiIOSc+wnn3BXn3O+pbXPOuV91zj03ep4dbXfOub/rnDvnnPsvzrm338rOv5GIMyh2gyyCg3acxMSUEFnuEl1H6Bgd8hQST+wy8M45NJtNHDx4ELOzs6mwoIUKW5fti+S3kdcxF4i4brQzx+aI0WVt34FxEadUKqFUKqFYLKJQKIzNt+2fhFBpEUTa0Pl89LF2VSxdZ2wlr72IhFo08t6P5SSam5vDkSNHUKvV0Ol0UCgUUK/XdzmY9LXR7XbTObp27RoKhQIWFxfHwsasG0ePdzAY4Ny5c3j22WdRqVTw4IMPYnZ2dmwJezs38lqfo0ljD82lTeyt59f23Yqoep+tPyaSUhQihBBCCCGEkL05hD4C4OvNtu8C8G+8928C8G9G7wHgXQDeNHp8O4B/cHO6eWfx+OOPZ4o6VjgJOXJi+7LqtNtDIkXoJjmrfRFkGo0GZmdng6JBFqHQnSxRQNdtw7Ns/wQbdibzK2JHSPCwbQBInUJaENIig5STfD460XXI1RQbWygEKtS/UF1aFJqdnUWSJKn7x7Zhz1W/30ehUMD29jb6/T7q9fousS10ruT4VquF1dVVJEmCfD6Pubk5lMvl4PUYc0fdyHUTmjvLJGF0r9eqXXXt9OnTe+onIYQQQgghhNyrTBSEvPf/DsCq2fwEgJ8cvf5JAE+q7T/lh3wGwIxz7tDN6uydgNxIhkKHsoSeUBhV6KY49GyJiUV2X0icsH1xbrhyV7PZTJc9D91w2wTOIQeH3WbLW8EqJj5pEcbWa4UhjQ4F02PU+XwktEy29/v9oNvHjiurj8Du1dtCQpXdbrdp99LMzEwaOtbr9dI2tFNG91NCxXK5HNrtNrrdbnDpet2Onb9er4eNjQ1sbW3BOYeFhQU0Go1d5a2QZAmdGyE0j6G5ic1jlgiVNccyR3obnUKEEEIIIYSQ/cxrzSF00Ht/GQBGz4uj7UcAvKjKXRpt24Vz7tudc7/rnPvd19iH246EiYVudic5FWIiURYxAUXvj9WT5UzRVCoVzM3NYWFhAYVCIRUkQmKFYEUZ/bCrf2mRQwtJug09Bi0e6VXGdF/sGG34kJ0vK57oZNfiktI5enQuJD0OcSRJziIJUZMQNu1asucx1P/Q2Pv9PjqdDpxzmJubw8zMDJIkQbvdRqPR2BViJ2MUIch7j83NTVy9ehW5XA4zMzO7nEtS3gqYg8EArVYLzz77LK5fv4777rsPjzzyCKrV6q5cQvbassKMvgZCc2HPYWguQtds6Fq3YXQ635Qem90/GAyYaJoQQgghhBCyb7nZSaVDtoCg6uG9/0fe+y/13n/pTe7DLeHxxx9PX4dcDiH09liZSYRulu37mMsm5mrR70ulEur1+tgqXrrOkOvCLgdvV4SyYoAcK6FYVpiIjTc07pgwFhKudDtWXLPtAgj2RerT4o9OUK3FiFASa1suNKaYWFgoFNBsNlEqldJwMHH82LkSManb7cJ7j62tLXQ6HVSr1ehcWneb7FtdXcWVK1dSQWlqamqs3Vj7sXNi27V1xMRK2+eYyGTr1zmhQteEPGe5nAghhBBCCCHkXue1CkLLbhQKNnq+Mtp+CcD9qtx9AF5+7d27Mzh16lT6OuQQsu4UWzYmYNjXmtANtb25FmEjy5Fi29AiQKlUwtLSEhYXF3eF04TELBE99HYtCGnXjOS+kWTJshS7zt0Teg3sTnwtz1nCUJY4pvcL0l/vh8u2e+9RKpVQLpfTlcoKhQIKhcLYNjlGBBjdR11Gi0chN46dXzuefr+PJElw6NAhHDx4ME30LSFc9hwDSJefd87h2rVruHLlSrqEvS4bE/qkjrW1NZw7dw5bW1uoVCp46KGHMDc3N5bMO+RUi30u9PkLCV/6c6Kx168W7KywZUUfcXlZB1NonplPiBBCCCGEELIfea2C0McBfGj0+kMAPqa2f7Mb8kcBrPtRaNndyunTp6NhLUKWC0K22ZvZvQgYdtlt+6yPj90gx0JvCoUCarUa5ufnUSwWd4XZaGxYlBaCrFgk+0QMSZIkfWgBCMDYe+99METMjjd0HrKECD0HUk6LWOJ80eMvFAoolUqpICQikBa2pB451uYl0sva23C00KpqdgzSVrlcxuHDh3Hw4EG0220MBoPUzaXD8KRtSUDd6XTw4osvpomhQ44oOz/S536/j83NTZw7dw7b29s4evQo3vKWt6TL0MdcZJrQubO5nyaJovaa1sfYELZY+dhnVRxr8v7s2bPMKUQIIYQQQgjZV+xl2fmfBfBpAA875y45574VwN8A8LXOuecAfO3oPQD8EoDzAM4B+DEAf/qW9Po2ITmDgPANpnXvhG5qZX9IpNB1hIjdxIecFHbpbisyWCdRvV7H0tISKpXKLoeOxa7GpZd3t4KTdnOISKRXXJN9up86h88ksUCLLbLfCjKhpdm1Y0RC16RcsVhMnU9W7JKHXnkstOqZ3afdNPJaj9+ueGbHqgWfarWKQ4eGudm73S6mpqbSc2DPf5Ik6Tna3NyE9x7NZnNsviaJheJOOn/+PH7/938fuVwOhw8fTq+X2LUYOnfSjj7/9rMUEiFDxMTRmCil+xUqb0U8AMwpRAghhBBCCNk3FCYV8N5/MLLrawJlPYD/+fV26k5AcgbZm9iYU8jun+SasOXsTXHMFRMqG2pD0PlUdLuVSgW1Wi0awhQKrwmt7iWCTGg8WWKB3IBr90xMzNKuECug6POT5aYKuXBkm4hWVhALHRPqT6ysFeRkmxbntEvFHgO86lwqlUooFotpXqBcLjd2rBV3isUiWq0Wut1uKuLoPoauIVtXt9vF2toaer0eKpUKFhYWsLW1he3t7bG5DQk/sXnQ15v9bO2VSdd/lvgaE6V02SeeeAIf+9jHQAghhBBCCCH3Mjc7qfQ9wenTp6OhKPp9TDAAwqtzWRdNTACS428kJCqUyDkUZiNtLywsYHp6eswpoYUVOw4pp0UQLczYstoNE3LE6PmxIont/17cWSGRKCRc6XHE5lY7SOxqaLrPMg4Zm27D1m1dRXKMdifJnOo+iCOqWCxicXExXcWsXq8H50ZcQiJyXb16FYuLi4hhr3N9HSVJgpWVFZw/fx7tdhvHjh3DyZMndwlwsevUMul6zjo+JsTG2ojt09eVvX51vis6hQghhBBCCCH3OhSEAkxyqmTdnIZcKlmCTijcJSZoxAgJKVmhNLlcDpVKJSpKhcZnRSXdVpYzxgpjuh4tMGUtQ6+f9X69T7+2uZcA7BKMQvXYnDyhObCheXqcWvAJ5ViyyaUlxEy7pOwcSjhTv99HvV5HuVxO8wTp+dXhhRI2lsvlsLW1NZaI+kbcZdLHq1evYm1tDeVyGQcOHIiKaaF6QoKRfZ91bdtzs5d+Z9WZ9bkAXnV93ahriRBCCCGEEELuNigIKR5//HGcOnUqKuro1/omPCYQTQoFCrmIdFvyHLsZtkKLLRPKQZTP51EqlTA1NTVRNJJ6tbihy8fyFlnHkBZJZCzaMWRvwkPCQcwlZEUFXa9s18mqtfNGn8NQH7Qoo5HEzjpPkB2vbtu6oWy/JIm1XsHMOnX6/T5mZ2extLSETqczJl7ZBNM7OzsYDAaoVCrY2NhAvV5HqVTadV3JfNrxaZGu3+9jeXkZzzzzDPr9PhqNBur1+q5xxcTP0NjtdRkTqmKi6KTrPnTObF9sPaF54epjhBBCCCGEkHsZCkIGKwiEnAaTQpn0TXZMcJkUXhO6aZ10nA7VCgkbgogjsfCm0HH6Jt4KQ5PmSdcfcvVkCQt7cVPZJNPWMWMFvpgIFxLtYljHTwzpnz2fk+Y31O98Po9arYbBYIAkSdLtNgG2rOgmq6QVi0WUSqVd82fFtNDcAkCv18P6+jq2trbgnMPc3Fyaxyg2d3ZbSGQNtRXbFjqfWuCyx4S2xa7ZUN9kG1ceI4QQQgghhNyrUBAaIUmkhZDIY4kJNTEx40YQ18deBKBQv2IrlOVyOZRKpbF+hsZnb5Bt+JMdbyhkK5TXKCai2G2TwnZCc6tFLrvEvRU/Qm3p8xYTcXR/QyKXbSskwOhV12z9tk3tAgKAarWaij569TIrZEh5SRxeLpfHyltCfZY+DAYDdDodrK6uotfrYX5+HrVaLRjul3XeYuLNXj4bWYKRPc/2vGRdbyERVdfjnKMoRAghhBBCCLknoSAE4NSpU7u26Rvd2M1oluATuwmNCSm2nlgYTegm197Mh27SgaEQUS6XMTs7u8vBFHOLaPeJDRGLiQjyHBJ/9PZQIumQ+GLdILH9+r2eA3tsaK70kvG2fi0Oha4JqcvmENJjklXVpKy4eDSSMNrOoe5TrVbD1NQUer0eyuXyrmtAkDxClUoF6+vrWFhYQLFYHKvXnq8sV02SJPiDP/gDXLhwAcePH8eb3vQmLCwsBI+JvbfCYchZFDs+9NkJ7dPimRVFY59Ze93pa0fOGyGEEEIIIYTca1AQQni1sEk3q7bsXvaFbkhjjhW7P9ZGKGQqVF4Ei3w+P7YMeai/WXVk3eQL1gFj+2DDf0LjDgku9sZdb9dOGi2+ZI0rJC6F8tvYPunxxVwoMbeJJeQI0iKEFa0ApO6cYrEYbMf2qdVqYWpqald5KwaG+qVptVpYWVlBsVjE3Nwc5ubmxvIz2fMSY69Cata1H7pGZHvssxLqn35vz6e+HukSIoQQQgghhNxr7GtB6MyZMzh16lT0xnOS0GPdNCFiTojYTbN1b1iRJMthEXqvb3Cr1SoWFhYwPz+/q4y9EbaCi+6DHps9zvav3+8HxRzBikd2vDb0S2/v9XppfTqvjl3eXY4VoSVJktRBY8dsz53eL04h7SCxK4fJMvG6Dt0PeR9KSC7l7LHSnvT7vvvuw+LiIsrlMmq12q55E7FI5v/ixYs4dOgQ7rvvvmiuIttne4699+h0Orh48SI2NjYwPT2Nt771rThy5EiasHqSqGpFGFvWth3qQ0gUjNVtr6lQG6F69dzr+Tlz5gyXoyeEEEIIIYTcM+xrQciG/8ScQvr1jQhB8qzdHbG6bb9suZjLZFJ/tChULBZRrVZ3rY5l67E38brtmLtHbws5g7S4EQvbidUzyV2jRRgrgmm3jeyzLqWQqyUkrNl+2/MU6q8+NhaOFhL5QudHrqNisYhKpQLvfZoPKjSPIlD1ej1471Gv16MiiT4mJpzIPK+traU5jBYWFlCpVNL+xeoNbdNjtJ/FkJBkP6tZ444JO7ZOfazdFroGsxKNE0IIIYQQQsjdxL4VhGR5eSCe2yTmiAmJHhYtCIQEDXu8Lh/KfaLL2P6EhBx7XC6XQ71eR7PZ3LUse0igiLl/rKPC1iN16Vw69kZebqp1HiON7kO/30+FCC04iNDR7Xbh/avL2Mv2TqeDbreLXq+HnZ0ddDod9Ho9JEky1jfdv5gwI2W06JEkSVDA0A6mSWKEbSOfz6NQKIz1ydbT7/dRLpcxNzeHTqeDYrE45oiy9fb7fSRJgk6ng5mZmTEHkhC6BkKOITnfzz33HF544QW0220cP34cJ06cSPtgRS8rsuiHnT/rthKsm0qLf7aP9jk2RnusFixD50uuRQBcjp4QQgghhBByT7AvBSFJIh1z1uxF8LHE3A8hp0SWi0YLHzfSh72Ew8hKU1bciQlfkmg5JDqFRCHbvi4nx+r2ZKzykPCyUEJsYLdYJWVFnBHxSOqxdTQaDSwsLODIkSNp2NXMzAzK5TIKhcLYXMZEPBmDFaiEUCiYnbfQGPS5ylrRTcoXi8WoW0WLblJvu91GqVQaO/+2P3b8sffb29u4du0atre3USqVMDc3F3Q/hfpl58QKRHquYsJsP8x2swAAIABJREFUlrhjsfNu+xHbbo+31/0TTzwRbZMQQgghhBBC7gYKk4vcW7zrXe8CkH3zF3LkxMSZ0A2rrcOKJjGxIItQ+7bOLAEpn8+j0WigUCiMCQk6h0/sRtw6KGLiU2i+tICit4vbQq/gFepHLpfDYDBIhQztQCqVSqkY1Ov10mPEZVMoFFAqlTA7O4vDhw/jscceQ6FQgPevho31+31sbm5iZWUFn/3sZ7GxsYF+vz82RikbmjdpU8Zi0fVYZ4udT5kLOUexc+q9R7FYRJIk6Pf7KBQK6fg14pjK5XJYWVnB9PQ0arUaer3erlC60Dmzr4V2u41XXnkFhUIBc3NzWFhYQLPZxOrqalpW+m+PzzrfdpsWavT8Z4mQ9voMCUch8UmXCdUn4ihXHyOEEEIIIYTcK+w7h1DoRl7IcsqEsO6BECGxaFK7sXZCr2OCk3aayOtSqRR1aYTe67qzxA4tbIT6qm+itYPHjl/K6xxHEkoV6peMSZ6LxSKKxWL6ulAo4NixY/iSL/kSPPLII2muG93ffD6Pubk5HD9+HI8++igOHz6c3vhrUSaUO8k6hvZC6NyE5lPqD+WT0q8ll4/eLvMtY8jn82i1Wuj1eqjVammYWahfenwhgU/G3Ov1sLW1hSRJUCgU0Gg0oqvLZbl5JrmKpExMwNGExCW7P9S/UJnYcXqOuPIYIYQQQggh5G5mXzmEHn/88eBNYZYTKHSjHCqrRYasG045LlQu1o9QuZjrItTXYrGIcrmMcrmcCgX5fH4sV4wWSkQQEcEkdtNuw450P2SbFoC0GGLnQm+3LhLbP+DVHEXiltHzksvlMDs7iwcffBCPPfYYvPfY3t7Gr//6r6PVaqFaraYrdJXLZTz88MPI5XJ45JFHcOLECfzcz/0cdnZ2xkKzRJjS7hAJc5O+S//1OGLiyiTxQrtS5BgJ30uSBKVSKXUGadePbdv7YX6l9fV1XL9+HfPz81hbWxubL+0GC10/tp8y/tXVVWxubgIAHnroIWxvb2N5eTn6mYk5gULEPltZhK4Z+37SZzOrnyHn0pkzZ/CJT3xiYp2EEEIIIYQQcqexbwQhm0RaY2/07E2fLSeEnDmxG86QIJBFVrhQiFCojBWF7HL2sRtt7YgJOUm0eCDHilihy9k8NtrRJNvz+fyuxNOhubBjj4X2yJLrb37zm3H//fcDAK5du4bl5WW89NJLaVLmfD6f5tSZn59Hs9lEqVRCqVTC4uIiLl++nNYtoWXybJ1Peg5D502PTebFClx2/NKWrls7lmTukiRJxy3Hy5xK+VwuhyRJ0G63UavVUCqV0Gq1dvXBXj/6tT3fg8EgTdhdr9fThOWXL1/eJfrZczTJqRM6t68Fe22HxKlJn3nbr1AbhBBCCCGEEHI3sm8EoawQJbsfCN/YW9EjJmJkCU+xtux2e7NuuRFhKpfLoVqtjok32oWjx6VFFTsWLTDE+h0SR3R5Xa/UGRKltItJ9zdUTo/zxIkTOHHiBB544AEkSYJPf/rTeOmll9JVxvr9PtbX19O6xPUzNzeHt73tbSiXy3jHO96B8+fP4/nnn8fm5mZwPHqlNJk/GzIVC5XSY7dzq8uGrgHd3tTUFLrdLkqlEqanp9MQLnEPyZxMT08jl8uh0+ng8OHDmJ2d3ZUrSRNzxNn2+/0+XnnlFdTrdUxNTeHYsWN4/vnn03oniaUhASrkxNkrIcE2q+yksU6qR9qjS4gQQgghhBByN7IvBKHHH39817bYr/0x54k+LivkJybkhMQgKwBktRvaNkmMkrKFQgHlcnlXHSGXUOiG2goSIjboeRBhJ+QokuNCfY3No/fhvDzWaaQTMTebTbzlLW/B/Pw8kiTBK6+8ki6PniTJrqXi5XljYwPOOWxtbcE5h2q1iiNHjmB7extbW1sAxkU0K5JJX733KBQK6XabsFuO0+FusetLnz97rNQpCbLz+TwqlQq63S76/T7y+TyazWbal2KxmIbJlUolVKvViW6r0JyHHD6bm5up82hqagqFQmFsnu01pJ/ttaeP2YsYFHI37QU7hqxjQ5+30LazZ8/i4x//+J77QAghhBBCCCFvNPd8UunTp0/vuuEOOU1C22S7ELqJjh0XQ3LN2Hpj4SyTHA9ybNYNbj6f3+X4CQlUMbeOlNNLxGsRxN7Q20eov9Ke7YMWPiT0yYYghcSCZrOJBx98EAsLCygUClheXsYXv/jFNBdQt9tNhQppW85dt9tN899cv34dvV4PlUoF999/P2q12q7zZVcLs3Nk58OOXV+DMVeQ5G7S58mOW0Q+7eoqlUool8toNBqo1+vY3NzEtWvX0G630+ugXq8HPweh11nXt/cerVYrFc3K5TKq1Wo6R6GwMTkuJkBmtRUTWS2xvoeu+dD70HmdJDzdqJuJEEIIIYQQQt5o7mmH0KlTp8ZcJpPCVrIIORdibhrtEspy9di2Q0JQqL9ZN6gh90OlUkG9Xk/nQkJ65FgRqfTKV1lClHbKaJeOdWyIc8c5NxZGpPuXz+fTOu0KcHacEqqk3UH5fB5HjhzBO9/5TuRyOVy+fBnPPfcczp07h263i06nk9Ypy7DLsu7S71arhSRJ8Pzzz+PixYs4dOhQ2peHHnoIly5dwtWrV3clyLbnRyd0tqFsIWFI9tncQDJe6a84r2y7zWYT1WoVGxsbqNfrmJ2dxbVr19BqtbC9vY3p6WkUi0WsrKyg0WhgdXUVU1NTOHjwYDrvWWKIdfCExrGxsYFnn30WR48eRT6fx8mTJ/Hcc89hZWUleoyt3xISJWNCjJ03Pb8xR0/IeRWryyZdt/3UdZ4+fRq5XI5OIUIIIYQQQshdwT3vEIqJMXZbzDkUqie0TScYtnXq8npb7EY2q/3QzXWWeJPL5VAoFFIBIIT3PhUesvqkV9Ky43mtbgzrttmL48qOd2lpKRWIrl+/jrW1tTHhRvfbilbaMSRJkie5UULnUZ//WL6j0LVgw+KsiyjkTpL94vjp9XpjOZK63S7W1tawubmZuoh2dnawsrKCra2t1H2lxxUS6yaJN7Kt3W6j0+kgn8+jXq+P5auy82Tfh4g5i0LXRUzQil1HoWs0dJ60CBb6vFqnlxAKcySEEEIIIYSQO5F71iGkl5jXxISVmFshxCS3kBWEYuJIFnJjafPQ2D7o9kOuCOeGOXEkZMyW1cKA98O8NHaFKisUyYpbkpBZjrchYBL21O/3UzeP7kPoZjokHoRWIZMb9ampKTz88MPw3qPT6eCFF17A9evXkSTJWDu6X9ohJduSJEG/30exWMTi4iKKxSKuX7+Oz372s9jZ2UmXWpf8STohs64jlJDbJoy2ooJ2COnznc/n0ev1gtdZv99HrVZDo9HA2toakiQZm+/19XVsbm6i0Wig3W6ny8OvrKxgfn4ejUYjdfGEsNe3dYXpa6jX62F5eRmVSgXz8/NYWlrCK6+8ks6HPg+xz2RI6Axd86Ftug773vY5hL6+Y32x5WP7CCGEEEIIIeRu4Z53CFmsA0Nv3yt7cX7otrLqjzmTQkKGdqLoMpNuWguFQhq6FXIThVwWoTFpt0Qof5DNMRRzmFjXi3Zj2G1St30vDqADBw6gUqmkfZQVtkQcERFKjykmCkq/qtUqtre3cfHiRWxvb+8S56yDJJ/Pp0JabB4nXV+xayV2nUqfKpVK+l6EOjkXvV4PW1tbqFarafl2u41ut4tqtTomPmWJp3abFYoGgwE2NjbQ6XRQKpXQaDRSV5Z1zNzInExy+Nj6dL4pe81Masdek7p/9rqxfckSmwghhBBCCCHkTuWeFIRkVbFYSIjsmxTOtFessLDXm2tdfpIoI9jkyllCE/DqCmNaKNA3y9KOzqmjx2RdFyFRSMppIUi7gUIJku2cWJErdnOuBb1isYiHHnooXWWrUqmgVCphamoKxWJxLCeSrUNyFsnS9sViEbVaDUePHsXy8jI+97nP4YUXXkj7LYKajMuGz0lduv82rCjkSrNjtqu1aeeLrsO5YX6mmZmZdJztdhulUil9L6LQYDDA9PQ08vk8Njc3sba2hpmZmdTtFDoXMkeh8xASQl555RUsLy+jUCjg0KFDqFQqqSAk7ip7rdt50ufotWAdQaGx2fK2T7Z/QDj5uW1P7zt9+jTOnDnzmsdBCCGEEEIIIbeDe1IQEkI3ma/31/yYe+ZGHEcht0qMkNMhJnDZ8YqooN1BMfEqtN+OK+QU0jfU1h2U5UjSiZdt2b0IXVoUknGWSiXMz8+jWCyiWCyiXC6PCVPSrp4beV0oFFJR6cqVK9jc3NwVJhXKnyT12fFImSxxxZ63LNEsJHR471Eul9Pl57vd7pgYJ4LMzs5OOr7BYIBOp4NarYZyuRw85/p1yLlkxRPvPXZ2dtKwtFKplLq2Ym4wPYbQeGNkXRshV1oWsc9ulgtpL/sJIYQQQggh5G7gnhOETp06lXlztlexYdJNp3V0yEM7R/Zy4xorox07VuyJhcNYIQJAGs4UO86KOHbsIUeRlEmSBEmSoNPpjCWHFqeM9Knf7yNJEnS73V1t9Xo99Hq9tK5QXySPjg5JE/fL9vZ2Gj4GAMeOHcNb3/pWTE1NpW4ZEXzkocO8CoUCZmdnMTc3h4WFBczOzuLChQtot9vpHIRyH+n34oIRJ4ydO2C3s8ueeztmLVzpUCh9zUmI2MLCApwb5g7a2tpCrVZL+yb92NraQqlUgvcea2trmJubw/T09C7nkb0eQ9eZ7oPs73Q6ePnll9Hv95HP53H48OG0PXvtaWExRuwzrIXLSYJmzNkj20Kfd+sqs3Xp+iY5jwghhBBCCCHkTuaeE4SA3a6LWJkYe3EpxMraG8sbJRSOY290ddnYDbvugw0Xk36G6tVl7PhCzh9dT0yk0jfTkmDaLs1uHUb6OYTse/nll9OVwQCgXC5jfn4ex44dw/T0dBoyp3MKFYtFlEol1Ot1NJtNFAoFdLtdNBoNOOfSMCvpq8ylOG5CjiHpk31t58nut8uyyzkI1SlYIUNC5KTvhUJhV2Lrfr+PVquFUqmUrggmDqGQ88u6zUKfJ339iPOo0+kAAKanp9MVzkJCY9Y4s65p/T40n3q7FndCx4fEr9j3Qki8iomoso9hY4QQQgghhJA7mXtqlbFTp06N3aBqJ0HohtCyl3AT7boAsMsNlCWOxOoKCQn6fVZISugGVvLjiBNG+miT/GaJOOICsY4M6Xcul0sdMeJO0SuU6fNgxR+9ype9CZfjbWJg2Wb7/dxzz6HX6+Ftb3tbuuT51NQUHnvsMZw8eRIXLlzAxsYGNjc30Wq1ACANM2u326lIVKvVUCwWceXKlXS1MRmbdu3Ie+lLsVhEkiS7rjs7VttvOd7mDCoUCml90jctlkmfgOGqZs45zM/Po91u48UXX0zPzdTUFLrd7piQ0+v10Gw2U8fU/Pw8zp07N3a+pS8xB02W26bb7WJ1dRXlchlLS0tYXV3F2traWH2xa86KjKE2bHt6LvT20OdpL46drLL2sxDqvxXPBoMBTp8+jU9+8pMT2yaEEEIIIYSQ2809IwidOnUKQFiQibkaLFniiD02dvMac7SEsCKS7UtMZAj12Yoqel/MxRRzgYRubEOilJ0bu7KUdvpYccEKIlK/tKVFrUmCxMsvv4xSqYSFhQUcO3YMpVIpdQWdOHEC7XYbW1tbuHz5Mrwf5trpdDppu5VKBdVqFb1eD9evX981Di2WhJwpWcJeTAQZDAap2KPrsSKYiEIhsUbK5/P5dGUvGV+xWEydT9rF0u12UavV0O/30zw/1omVNdexa8D7YQjh5uYmDh48iHK5jEajMVZPSFSZJKDasrEwSsteBKBJ7YXq3KvYq88RRSFCCCGEEELIncg9IQhpZ5Bgb+xCN756XwzrbrBMCqnRZfQxViCJiTbSh6w29fh0P+R1oVDY5crR8xATBWLOpFCYTC6XQ5IkYzlvBMnBY50/Whiy8yfbdb9lu2yTHDqdTgef//znkc/n8du//dsAhsuxl8tlHDp0CIuLizhw4ACq1SquXbuGjY2NtC2Zm0ajgQsXLuDixYtjcynjFPHGCldZfZZt2uGjx6sFkqxQNC2i6KXl5bh8Po8DBw5genoaGxsb2N7eRrlcxszMDK5evTp2zWxubqLf76PT6WB2dhaVSgXtdjvoupnklNHXjnD+/HmUSiU89NBDePDBB/GZz3xml7srtnrZpM9BqI/WfTVJYAr1WdcRE191aF+WaGTrvBGBmBBCCCGEEEJuJ/dEDiErTgjWySHb5BHK8xMSdLQwEBJDJvVrL21MQvc7JtTE+q4TXeub2azwnawxhlwyQHgZet2OTdgbmhf9iLmv7Bhk22AwQJIk6Pf7aLfbWF9fx4ULF/DMM8/g/PnzqFQqAJAmud7Z2UGn00nr29zcRJIku/pixybtageUfh86B6Hx2rFnbdfzZ8+NCGuVSiXdbseh56nX66XJn+3S87ru0LmyYpF1NIkby7nhCnASshhyocU+F3v9vNi+6n7FCLVh65nkOIoJV7HrFhguRU8IIYQQQgghdxJ3vSCkQ8U0oTCj2E2c3RaqQ463YkhMnAm1M+mmf69kCSkaLchkhQHpOrTIMSnBtEbyCdljbY4lO6d2JaiQ0GLbC928h86jCBQr/z977xIjSZal5x1zdzP38HhkRGVFZnZldU93o6cx1GymAUIYQEBDIAUQkzMAtRBFbqSRQGA23AjSQoJ2ArSQNhK4ojDALIaCAJKQBJBQZYMQKHHBhSgMNCMNpJ6ZLk2rqqururIqMzKe/jBzNy0ij9Xvv59zzTzy5TF1PiAQ7mbX7j33mnlE3D/+e+7Tp/LRRx81goruSlZVlUwmEymKQkRErq6u3J2pLGGFd2jD1+wQQacQ98Fqr23cLJFyf39fiqJY2fWMhUCNWV1Mmlha20gJPti2N0ZVVcnZ2VnjCsId2qx7h5+NlDjJ11t0+XxZwh0ebxOD8JgljFr3S7+HKBQEQRAEQRAEwTZxqwWhR48edZogch4anMjxBK7rhLDLshGrjk1cIll2nT9GvzQvjH7x9u4qwOhEHAUZjAMnsyy86ARexQSMScvqFvE6scY4VXCxxlfLqCsF4+MdxtDZhI4ujR0n9doexqvnVRh5/vy5/PznP5d79+41iZXzPJfxeCwHBwcymUxkOp2uCD8auy5T0/rqupayLNfGAGOzng0ez7qum/osAUj7pM+ClexZRJpd0Y6Pj+XBgwfNPbq6upLxeLz23C+XSzk5OZG6rmVvb2/FNeOJfp4gx46bqqrk2bNnTY6m3d3d5n5bAqXl9kq1awm0VlnE+/zivU6507zPrHefu8YQBEEQBEEQBEHwNrnVglAbOMFEYcSayKVcClgGsbai7nqt1bbnRsK4cCt5dGCgYKDCi8hXAo81ibfianPliKwmgmaxDevU+AaDQavogIIO7orGbXvx8heKLyrmnJycyMnJyVo9mlNHxRmGBUR2ArFAY7lPuI9W3d4Obdof/M51qyNHk0trv9T9xELXZDJpEkzjM9JFuLBEVLwOk3aPx+M10QjrwOOpzwiKc3jME3C79gMFNmRT117XtoIgCIIgCIIgCLaFv9CCEHKTyRiLMV3qaHMTcB18nM9j3hhrCZc6XfI8X3HqYHl2cqDjhkUnzgPkuaowDqxH3Uvo7tEY8Zi6iVA00nwz+l3rRtFLv/R6FmpQROG+oWslz/Nmd7HLy0t59uxZIwhp/Vgvil/evdbdwESuXTup5wHHDevDe41CkZUDip1T/X5fjo6OZH9/v0mWrePEz9n5+blcXl7KO++8I7u7u2t943iwL6lcUOpO+uKLL6SqKrl3754URbHyHGM/2KHGnzlLALTGH5+DLuLumwTjiGVjQRAEQRAEQRBsC7dWEPqN3/gN99ymDoG2855rxMMSX7BOrx7LIaSTZkXdLlabLAZ16Vev11sRYzwnj9dPFG+4v9YSIewz70jmOWcsF4cliuhxXdqFiZVVCCqKQqqqWhmv2WzWJJe2XEnaF+yPilgqfnlOIU/YSB33HFwq2KXEkl6vJ+PxuDmmfeX6NKn2cDiUPM/d+rq853tb17VcXl5KVVWyt7fXJPPG/nniKMNiEYLjYJXZRAS2PqubCEqpe4LPb13XTd6zIAiCIAiCIAiCt8mtFIQePXq0MglENhFhuooe1murXj1mLanx6kDQCYI5eUSk2T1L67N2t9IlRyqGsPsCY0anjQpCurQry67zAKHLRY9pbiF1/WicRVGs5AbCib/l5EExSYUYdOKg+FHXdZMnCCfXlkCgdep44TioMKGx61jobmNaB07iLWeOiKzkclJRRRNrY9/1GPYdhTC8P3oO+4V5i1DE43ZEpCn34MGDJk/SdDqVnZ2dNSFmMpnIs2fPZHd3tylriVp8//C5RrAPi8VCPv/8c3n+/Lk8ePBAjo+PTeeP5erxPjPW56arYOvFj9elPiv8OhWXJ0KlhK0gCIIgCIIgCIK3wa0UhNBh0mVC1nXSZpEqexMHQaoNa3KLjiAUf/gLXUHsDmLhANtkdwU6frzlPCLr7p6U6IROJ6suHqMu99ASFqwx1q3ol8ulzOdzqaqq6Z+KWLwjV2ryj0IdupNQPMJxsEQGbc8SCHjJHgtUXJbvfV1f5+3R3ECLxaIRv7AeHZfBYCCj0UgGg8FabiirH1Yc1rjPZjM5Pz+Xoihkd3fXdPPgWHdxAbFglPrcWPcB69zUDdQmHluiL7ZzU+dSEARBEARBEATB6+JWCkLsjulCakJnTQ49cUHp0jZPWlOxIZg3B5cqYZ3qxEARSHe+wmt0yZDlWrIm+IPBYMU9g9uUs0sF27dEOq1XYyrLcsXhY7kyLNeLik94rV6Hu6GhOIPuo7q+3hXs8vJSnj9/LrPZrDk/nU7l4uJipX+ct0nB5WE4Fnqt1ol5kDAXEsaPz4S6ryyBg8cDk3izy6YsSynLUoqikPv378udO3caQQjvj47r2dmZDAYDOTo6alxEqeeSRRnsB/fr4uJCPv/8c8nzXB48eLDidvOENx5vLIN91nHlHFjWdV3oKvTwc4qCnNUe/4zCdiKXUBAEQRAEQRAEb5tbJwjpcrEuAg6SWoZiJV+2SAlK1vsu9XlLYURkZQkS98EShnCCqmV06ZQVJ4oIKCJhXHq9ChZaPy7h0lj1CyfLKgJp3ZxcGPuL71nswDIsBKAow0vlVBwaDAZycXEhJycnMp1OpSxLOT09bd6zcOe5nFDsYdcTCyOc1whjwn5pvdqeioEiXz2b7AbiHcg0Rh3nw8NDOT4+bu6N9ZxNp1NZLpdyeHgoR0dHSbGwzU3D92ixWMjz588lyzI5PDyU8XhsOsr4WbCwng98pqzYrGsYr2yb60yv8RxAKGql4olcQkEQBEEQBEEQvE0GbzuATXj06JGIbDaZZHdMapt0zFmDx3miz3SZfHrXeUtp8FhKQGlrqyzLZrcpbA/FCtxdS+Q6f5HmxFGxCZcQ6RhijiEdW50MqyME8wpxvCyMYH95go3nOC+PiKwkkMZrsB7dDn0ymYiIrI0L3w/sm3U/eFy8OjBOdArxvcNn07o/KBrxNTy2IiJ3796V4XAo8/lchsNhIwBpWU0sPR6P5ejoaK1uLYeuII47JZhcXV3JcrmUPM9lf3+/eW/1FfMnWaIRx8BjzuPc5nayxguPczttPwNSbbGrqms9QRAEQRAEQRAEr5Nb5xASWc9l42EJP97kmetHUm1gPF2dQRxjW91ema5OBl5yZrWL7g2ctCu4uxUvb8IJu16HeXXYScEuGR63lGND36PjhnMsWdfjEjfeMYzHg3MdefffcgZZAgU6Wfgci2CWGOKNAef54fjzPG9yA2nCcI5/Pp9Lr9eT0Wi0JlzgmGIsHI837nidbm3PsfL4cZtWXRhDmxjbVZxlPEGpS30sZrLIedOYgiAIgiAIgiAIXiW3ThDa5L/rXVwBntvBcndYy0P0PE5i28SmTePy+tIWMztZrEm8XqfJkq32sR6tV0UVrRPdK7w0CMtYQhTmCML22ZnD90FkNQ8RlsP8PSy6oJBkiYso4nC+Kv1Sl5IljPFxbzz5uPWej+E4cH4pq/xoNFpzF+HYlWUpWZbJcDg06/CeOy93Do+Tush2d3ebXEaWMGLVZbXP99Aq/zrdOF3qs36mhEMoCIIgCIIgCIJt49YJQjyp2vS/7eyo2WRy1uaE8I6xY8BLnIssl8uV3ay8NlBYYPFFHTGTycR0kNR13QhBnIxZr2UHDn+vqqpJZm05XjD3kOeUQaEJv/Pyviy7TlCNibPZcaRJsTXXj4pXmlwbhSeMw3LfoPjFiabRbaRjoXVgQml2XmF/OEE01pFywrAwifHiWGdZ1uQGGg6HMhqN1u7f6empLBYLGY/HZt2bwNcuFgu5urqSuq7l+PhYDg4OVspYycKZNnG1LYZN8D7fm1yfwhJkI49QEARBEARBEARvi1snCCltk6+ukzPPrdNF9NHjm4hLlpDlOY7wXGqSa12vx9Wl0TbpxmVSKASxi8OLCYUSTUSdEpI4fnaWWOPFY62iAm8Dr8IWCiRW2+qa4WTc1viiM4hFIEu0Y+GH+4MikeXsUqw8RVxOxTOOo67rxiGUZZkURbF2fjabyWKxkKIoVmLwxkG/8zI67xnUvEHj8VjG4/FKOxaW44nHx3qW8V57pD5D+GynPseWMGfF6V3btZ0gCIIgCIIgCILXza0RhPA/6amJlOWesEi5AazJmrd0hSerVjs8Se0yGVTnjbpdUs4FFCZQ1NFjp6enjSgksp5QerFYyGw2a77KslxZcmXFr+d1K/Y8z6UoChkMBitLz9Rhw7GzoKCiRr/fX9nJzHLzsKtJ69M8OegCQqeTVRcLRHg9XqcJqdmhxGISPnfctiUG6vjiPeRx4pxMOK5ahu+Rjt0777wjOzs7jeiDzqssy+Ti4kImk0kxtge4AAAgAElEQVQjFrFAhfcRx84STPkzVde1fPHFFzKdTmVvb0/u37/f5CriutuEHi2XEj+5Lo6zbYndJoIux8ZlUuItXxMuoSAIgiAIgiAI3ga3QhB69OiR61ThL56kea4Ni64OHK6ri0BlHe/iclKHCIofVgyW60Yn99PpVObzebIdFkC8+DVunIDjEind4t1yTnRxQuE5vq+eUKciBx9POWtQZNNjKIbodxXMOCG1JYhw3Aw/q9bYpJ5tFdr0O8bIzwIKkcPhsFkyx8JJWZYym83Wxolht5dXlgWTy8tLmc1mkmVZ4xBKubGsNvgZ9wQi7Buew3ucui9W+6ljqfgtcWvTeoIgCIIgCIIgCF4nt0IQEum2zMgTgvCa1CQ25cC5Cam2eIt7r935fC5Zlpkiiycs4NdyuZSLiws5OTlp6mRBh90nVsx8vq6/2sWKhRcUrzBm3jadBTx9jcvA9LiKIOiAUtGD28RlVux08oRFnsBjO5ofabFYrC3Nwv5rHCjgcR4oFJ1Y5NMYrGei3+83whSKPZ7ooGV7vZ7cv39f5vO56fSaTCZyfn6+Eo8nDHoiGB/DmM7OzuTs7EyyLJPDw0O5e/fu2nPLCc8Z6750cfNouVTdqfasZxexfibxea9dHt/f+q3f6hxfEARBEARBEATBq+DWCELWZKutvOe+sCa6lnvDwlsq06UsHusyCUaBQ0TWRJLURBnfV1Ulk8lkRazAsizMaJuW6KATeHapYJmUWNHVtdG2exY6mrA+fu25RkTse8DPhieOcF+wDY7REums5wbHBO+vJxxxvPge7/FoNGratHaSU/eY50JDUp8NK6bZbCZXV1cicp3we2dnx3zurNcWbaLvqyTVVpe2rc8Gn7fE7CAIgiAIgiAIgjfB1gtCulzsJhMmXuZiTdBYRGhriyf9KVHHEib4nCcq4LWat6au6yZPjxUDxo/1LpdL+fLLL5stwPE6zqvC26yjCIVLwjAGnuBb48A5f1iE0zw43Afc3UtfLxaL5jqNBx0x6KbRuDG/EMep7Vm5gNA9pWPBsfF90HHl9jwhSfvOdbEIxTmdrOfackTt7e1JURQynU5lZ2dn5brZbCZPnz6V09PTZuv5VB9R4OJ+W6Lk1dWV/OxnP5OqqqQoCvnmN78pg8Fg5Vp8HrBf3mcxJRhZ13V1CG3iOrLEM66j7ecJj1fkEgqCIAiCIAiC4E2y1YLQo0ePRKR9Quc5e7pOBLvADhRs05vwWeKO5zRg5wzXU9d1I3DoMiWrfiyPfa+qaiVXjOV4sWL0nDGeQ4XjFrGdJ1xG8ZxBnqtGz/M4WuIS3iscv9R90gTWujyORSEWkjCelNBkxc73DM+jEGOJl97zr+OZ57ksl8uV+PVLxbM8z1eu9caERUQeY+zLcrlslqvVdS3D4XBN0LTGxWMTAZbHCa+1RKyuonOX57DtZwK+x/IhCgVBEARBEARB8KbYakGoKzphbltm1FZH6jhPIC0hwMOaAHaBxQ2R6+U9ZVmuiEIYpzXB1bF5+vSpVFUlIqs5UbiPuAuVts05eSyHCLbLAg7nTEJ3km5Xj44kFsh4wq55gnCHLn2vwhmKP5YQhPWhk0ev1+M61uoS0h25UBji1+giybJrJ1NKmMC4sd86PjjenLPJEpE03uVyKYPBQHZ3d11BZjqdytnZmYzH45WlgOzUYQGNsYS5uq5lOp3KdDqVLLtevjYej9fK4z3Ee+dhiWEcG8ep48l9SrV3U0HZqtcSqazxCoIgCIIgCIIgeBNsrSCk7iCEJ3w82fKSImMZxXPhpK7hNtqcCpZjoKszAiev3C/MB6PLb1L16PeLiwuZz+drYhBPxrU+HmecrLMQYiUs5vfWmFtleFJv9UdFFFxuxSIS99/aicwSDyxhUfvvOXIsIW6Tyb33fOD5lKuJY2YHT57nzT3G+6hC1Gw2k9FotBKDJViwA6iLYLJYLJo8RVmWyc7OjjmW/JlBLCHKirONTQUe/hljjUOqbe8epX6OBUEQBEEQBEEQvAm2VhDyxAJrkqrgf/wt2BHS1qZ1fdvEkyfsmNwX+2DFbDkdvGUoZVk2bp+iKNwlZBjHycmJPHnypJmco/OE41JnDDqRMJ+QN17YH17mpCKO1ycFXTFYF99fy/mDk++2+43xes4NLKf1YFJtHAtr7Nllg/WhyKP1qksI7xt+DQaDNVeN1R66rrIsa1xNItIIP0pVVXJ5eSl7e3try8as8cKx98YKY1wsFnJyctLksDo6OpKiKFbasBw+fC898WsTXkZw8QReT7TENtvEM6w7lo0FQRAEQRAEQfAm2FpBSMR39KQmoeyM8Nh0aUbbhM5b8mHFzf2yRIlU3Pq1WCwalwxO0L3rqqqSi4sLmUwmaxNayymjbXA/UXSw+u65STAW3uI8Nb54XkUSr00dExxXFhSsmHh8tR0UVTgWq32+BymHE97zlOMoJRimnCYs3ulY4PI1/czM53MZjUZru5B1ETPwmbD6Wte1zGaz5j6Mx+PGseQ989bzaPUt9XmzSAk5XtlN4Wch9Tzgca9PQRAEQRAEQRAEr4OtFoRE2ieEfCzlYOnqDmhzPVjHvQktk4rJqt9a5oRlMfeOB8Z3eXkpFxcXTV1W3CoSaN3cLr7mZTAi/pKrlPDDIpI1OWZXitWOlUya31vOHbwO8yR5k3lP0LFEPxTO2EXEY9AmgnoiG48RU1VVc0/V4YTPVlmWzQ52Vv3sjOFxsZZS4vhdXV01yaXH43HjEErB7VtLrrzPB5dj2oRXq65NsX4eWSLbTX9OBUEQBEEQBEEQvAxbKQjpVvOYKPqmk6QuDgcmlayZX9/UqcB4E3wWPbRezANTlmXjFLKEFYxlNpvJZ599ZvaR28Ft2z2nCraFfWCXxGKxkKqqVkQjdAq1TZRZSGHxBuO1ci9ZW7Wr4DWfzxvBROvUZ4/bxCWAVv2cnwnvH28J7y2fs0RGbpfvHQpB2C4n7S7LshEP8X6VZSm7u7syHo/dZ53vCceLx1kQOzk5kdPTU1kul3Lnzh05PDx068XnYFMh1iu3KZ4LK9VOqi4rJu9nS13XsWwsCIIgCIIgCILXzlYKQkxqUujBE9eujgDLNcMT3K4uhDa8eDCGlJtGsXai4thxDHTHJ6zLczml3Au8nTq+tpw4eE5FC8sVkxpLz0GE8afuPYsli8XCFJ6Wy2WTo8mqI+UeYpcOXosCJ7tb2AFkCWAYI45d6vNhuZV4t62qqmQwGLTmieI+t32uNMayLGU2m0ld11IUhQyHQ1eU5HG2PpOvm5Ro0+Uz6dXpXWPd3yAIgiAIgiAIgtfJ1glCurtYl//Mp8pYE7kupESRNoHEOmY5XrylIhiDVSYlbOn24tYEWsUHvVYdMZ6wgEum9EvzzmAceo7zNrEAVdd1s0QJr8U2rKVOZVkmRTNux3LvoAtHhTPtv8hq8myNUV/rd0VjtNxX/Br7o3WoC4ldPtgnHC99v1gsmtxIWZY1ib55eRyOA2/fvrOzI6PRaCUnFLa/WCxkNBrJ/v5+8jPF51Jlcdxns5k8f/5c5vO5DIdDeffdd1vvK7+3hJIu4szLiiueS69r+xapMi+TNDsIgiAIgiAIgqArWycIKTzpSk28LEePV2fK0dAmIlnX8mScr+drUs4SPIZ98xwbVrm2nCrK5eXl2vhZ48nikBWnHuedzrBOzzVktY1uGm9MrT6qGGIJheqCUUeQ53ThaweDgeR5vuae4fIIixns3sL+sajF9fHYo6CE9aWeXc3bs7u7mxQa+/3+ym5kXF/KjYRYolFVVTKZTGQ2m0mWXW89bz0n3ufLI+WM4rjbYu8iHKU+613iVXgXPP4cBUEQBEEQBEEQvG4GbQWyLPumiPx9EXkgIksR+d26rv9ulmXviMg/FJFvi8j/JyL/dl3XJ9n1rObvisgjEbkSkX+vruv/o2tAnhCCE63Ujlp8nVV31xg2jfkmdJnQbrK0qt/vr4geWk5zDOm1P/7xj+Xhw4fyrW99q3HLYHvqNrLaYccGuzfYRYPlME8P9k2TGStVVa0tX8KxYWcO1i8izQ5hOG6aUBm3etelYbr7FiZcLopC7t69K1mWyXQ6bXZow53H2M2D482x9/v9FRGI3VFZljUxaxw6Vvod+8euKO/ZWC6XzVbvX375pcxmMxkMBs028OpCyrJM7t69a372vGeBnwksj5/lqqrk6dOn8sUXX8jR0ZEcHR3JcDiUyWSSrA/H2BNhUp8/S5j0sJ41S9zSc95noyt4PY9dXV/nEfrggw861xcEwdvhn376Ryvv/9p7v/aWIgmCIAi2Efw9Eb8jgm2jy7+iKxH5j+q6/ksi8usi8neyLPtXROQ/EZF/Vtf1L4vIP3vxXkTkN0Tkl198/Y6I/L1NAmKhh10k7CroOiHr+p/7l4Vj80StlKDFWHVZZbl+vIavnc1m8uzZM1dUwDp4Is7tWjtyWfVhP7hPlgPIupc8OfccJixAWSKFCjuWgKVuGd0RK89zyfN8bXywLe9+WoKKNZbo/EkJDtYuYdye1X+sm4VXjQP7aN0H6/PI/eByymKxkOl02rTJbiSmi7jSVsb63HStoy026/Pi/ezqAt6LcAkFwe2AxaAgCIIgQOL3RLDttDqE6rr+TEQ+e/H6PMuyH4vIQxH56yLyr78o9vsi8s9F5D9+cfzv19czof8ty7LDLMu+8aKeVvg/8EY8nZeuWPXq65uIQ12u6xobTijRTcJxoqPEwhNtUnGoE+XZs2fy6aefyje+8Y0Vlwo6ilB8sHKboNDBeWs8RwnnEsK6FHUM8fV6Lectso6ha0fkWuxAEQhzLqEb6eDgQPb29mR/f78Zh8lkIpPJpHFTcf4e7i8/v3iPMLE0xqkOJr4WE0Djc8I5jXjsFc0blOe5DIfDxgE2n89X6i3LUkajUeMe8u4fu6E8OK6yLOXs7ExErt1S4/FYLi4umjH1BJbUc3LTzzK3Y8XetV4WG604rfqtOlIxBUGwPXh/4Md/foMgCAKREIKC28NG/4bOsuzbIvIDEfmXInJfRZ4X3++9KPZQRH4Gl33y4lhn2sSPLmUp7pWJ9JuebKXaazvX5pLhOrzX+p7FhqdPn8rTp09Nxwm7TNAJ1FY3lvMm9Oj44mTHXSbXnoMGBap+v9/0DfvHMYhIk0h6f39f9vb2JM/z5tqrqyuZz+dNYmrdoUzkK3HIEuH4GIt3bQIev2fBiUUw7JMlFOKyNW4HlyayuMfwZ8q7D9i+JvTW8zs7OyvuNHxeWPziMbPGeBNBhrHGfRNxt42UU8kSdTeJPQiCIAiCINh+4p8GwTbSWRDKsmxPRP4HEfkP6ro+SxU1jq3NqLIs+50sy/4gy7I/0GOPHj1yJ8AidrJVT6BYaXzDpRtWGyLriWA3vd6KSySdRNYTQvScnlcHD28njnXg96IoZH9/X3Z3d2WxWEhVVSuiCQoNOsaab8dzpKTEA5zkq7ikoPigu2lh/Xje2uqe+6e7euGOYSrs5Hne9FPr0/oPDw/l4cOHsr+/L1VVydXVlZyensonn3wi5+fnTVLqxWIhZVk2rzUHEYoq2r7Gqu4fhe8VfsfrUJxBgQJfo+sJxxzHSMdlOBxKXV9v/a7jpGXLspSiKNbcYtimJQJZnwt+9nQMdBxFpMlrhOX5Guu9Jw5xHKmfJYx1Dp1cVrkuog2LqtZ11nOsZXXXxSAItoN/+ukfhTsoCIIgSGL9nojfEcG20rpkTEQky7JcrsWg/66u6//xxeHPsxdLwbIs+4aIPHlx/BMR+SZc/r6IfMp11nX9uyLyuy/qd1UWa1LXdaJnTcC61NcWA4sU3jVtE8auZSxwoq5OGEsM4jY07jt37sju7q689957jTjAzgwWbiwnA4sz6DLx+sYTYK4PkzSn7k9KBGE3kOVUwq3bUdxQUaqua3n+/LlcXV2tjSe36z1XlkNI2+a4WGSxtqZHVLixciFZbiGRr8QyHD+9Z2VZNs+S5S6yxj11n714y7KUPM+b5WkM3yeMhcfaEgQ9NhVzU2IsvvY+x54AlHIjtbUfBMF2En/oB0EQBCIhBgW3j1aHUHY9K/k9EflxXdf/FZz6JyLy2y9e/7aI/GM4/u9m1/y6iJzWHfMHtWFN6q0yEHtrctZNJok4uba2jeY62+pmUSklXFntaP8Gg8Fa7hnLbZBl17lydnZ25P79+3L//n3Z2dlpRAIVByxhKRUD9yElBHHf0BXj9ZtFE6sdjVnHwoqV28LzKr7obmN5nst0OpXz8/NmmRM6l9CFpNdpfVZ+obbnkPvD7hsW3HCntC7JpbUuzM3EDiBdAlcUhSu0pJw8KTSOsiyb3c2Gw6HkeZ78nFjiW5vA4tXTNU5uu63+1PNuiYV8X7zx7dJ2EARvn/hDPwiCIAiC20oXh9C/JiL/joj8cZZlKnn+pyLyX4jIP8qy7G+LyMci8jdenHss11vOfyjX287/+680YklPALuIKlw+Vc4SoLxjVp1WWZ5o4uQcv1JiDH7XpUvozrEm9MfHx3L37l05PDyU8XjcCAwcv8i6+wS3Qed6rb6ycMZihzpg9Du6gyzxhyfHuNQLx4yTX2NdPAHXc7qjWJ7nMhgM5KOPPpLJZNKIFyyq4Jbv2HddTodiDYo5OpZajzpxMDZM5m3dc1yaZrmSsN8i0oynLlnT2DRfUlVVTYxXV1eyXC5lOBw2zxWOL94LT6ixxpifp6urK9nf35f9/X05ODiQL7/80kysbh27qUDS9Toew5d16bQJqql2wiEUBNtNCEFBEASBR/yOCG4LXXYZ+xdi5wUSEfmrRvlaRP7OpoH85m/+pl6/6aVrsCCSwprccl1Yp75OCTZtEzyeSA8Gg0YMwZi8XaTY1cEuEm5bBY/j42M5ODiQ8XhsOpws14KKKyx+eOPliQX6HR0ueMwaIxSOrHHrItZ5jhYUkLTcfD6Xqqrk+fPnK2IEO33wOit3EF/njZuKLuzuSt1zzi+Eybi9McD2eMkY3quyLGW5XLrbwVtCHffLu/d4TJNzqyiF5druq/ecWs9/m9jbdp7b3vQcxuddi3Hg2AZBsN3EH/pBEASBR/yOCG4TnXIIvQleRgjyXAl6jJdheJNNb1JrlbHi1Ymq5YaxwAS+6jBReGvyVL+xLUv0uHv3rty7d0/u37/fxMMJc1F40ffqghG5TgjsOUW89tlFg3Wzewav4Qkyn7cEE8zJo/R6PSmKYmV3MIwRhZzFYiHPnj1bcVphsmhNSq05hvhZQ/eOluf4cTz0OC4549g8YYuPWfeBy+kxFXtQHNKYdAv48Xi8lsjau+d8n6xnHfu1XC6bdkajkRwdHa2MBwqBXpt4nD/7eo/axB7rOfJIjad3vRefVa/XvyAIto/4Iz8IgiBIEb8ngtvGRtvObyvef9atCaolmmhZvM6iTbSy4kC3D5ZTIaCqKpnNZo1oocu/MH6OTcUNnNTyBFTbG4/H8u1vf1vefffdpv6UmIX9QLFCz+MSM08488QcbgcdNOyw4eVg+Jon4yg66dgcHBzI+++/vyIsYEwqiOCSKhWDqqpqlorptbqzGO40hu4sbVv7pfdVl2SJXOc38p4/vQ6PWWJCXdcynU5lMpmsLPfi/rFYpHHqrmsaD8Yxn89lPp/LeDyWPM9X4vCefStWS0TC+C8uLmQ2m0m/35f9/f2V5N5WO5ZYwl9aNz5HqXpexo3T5Tr83HfJH8V9x89GEASvD2/HsCAIgiAQid8TwV98boUg1DYxSk0APaGIX7/Mf+S9a612ULBRccEST9ghYU24LQEA49nb25P33ntP9vf3V5xBlhMKd7RS5wruPIb5ZNjFo9ewcGG5R9Q5w8vEUn2y+oZlUDjLskz29/fl+PhYzs7OmgTG1hhjHZZQxpNyHDur75aIVVWVzOfzlbpYaNHxVmGLl9VpmdlsJtPpdGWZmOci4nHS2PM8lzzPV1w0+lzMZjMpy1LG47EMBgPTVcWwuGfFxM/excWFTCYTEbl+RjnRNbrTLFHXGn9u18J75l4FKZHVi5+fiTbhKAiCV0v8kR8EQRCkiN8TwdeBrVkyZtG23MISDLrgOVe8c6n4UuW9uPXLc+vgtdbEF897Dqcsy+Tw8FAODg6aY5aQ49XDy7pErpeNab4j7Is3FtiWJ1x594GPcXxeXYeHh3J8fCwnJycr28V7fWNRSfspsrrzGbuLePKO/VcxQ+tjEQ5J3Vvur4oq2DYLcTgW1mdEHUIaIwt2i8WiWR6HgqAl2HGs1rPAsdV13TiR6ro2XVPYB2vMUu16Y2KNKz+/L0MXUdi6hp9nfs5eNq4gCGz0j/yw9gdBEAQMCkHxeyL4i85WCUI8keMJHDsmUqQmetYkt2u9CCda1rq6TOQ4ObAl6PDE0JtgM1l2naj6/v37cufOHSnLcm3pCk+evXFCVChRRwfWge4Y7kubg8MbB8/90u/3V8QWEZHRaCTHx8dyeHgoX375pZyenq4IP1xe8zWxE0sdWzpemkdIhTBPfOA+Y54hketEyrp7F/fdcpFwnCjg4HXW88Zih/ZP+5TnuUyn07V65vO5TCYT2d/fl9FoJKenp6bI44lQ3C/8rud7vZ7MZjM5PT2Vb37zm5LnuRRFIbPZbKVcF1EkJZamRCmOkfF+dqTq8mgrb53nZ+vOnTsbtRkEgU/8kR8EQRCkiN8TwdeNrRKEUk6gNpGFBZ5ULhGvrTbhgtvqIl6wC4An8FafvK3T28CJ7GAwkOFwKCKy4vTgOr1+pCbmnhPIGiM9x8ml+X5ZTh5PIGRnR6/Xk4cPH8qdO3fkiy++kNlstpbriPPUoGDD4pse5yViCO9ClhoDBcUmLKPOHKxX60LBykuWzAKGFxeOhYisiHjaVlmW0u/3Jc/zlXbwmeT6vTisstrXsiybWFhs4zH0SP1M2ES44ba9Oq3d7m6Kdc9SgngQBC9PWP+DIAiCFPF7Ivg6sjWCUOq/8m1l9FzKwZCCJ2VtwtBN6ubrPDHEEmCsnZNSMfT7fRmNRs2k3sphgrtyYZ9x0mtNgDG/C295bjmurHuG4gyPgZ7DJV3o8uE8PQcHB/Kd73xHTk5O5OOPP14RVywRR7d5V5cRO24UdiThsX6/3wgn3HfuM47xfD5fGWPMz5Napqb185ipoML3wXLz6PhpIm19jfFrvqOiKGQ4HCbFCuszmDrGwth0Om3q3N3dlZOTk87ir44Fj7tXto024Tgl3lr1tJXje+i979puEATtxB/5QRAEQYr4PRF8Xdm6LKY4CUY8YcUrkyrXpf2b/nceJ3FWPJ4oZMXrOaa8L6Xf7zfbi1vXY+4YEWkEEs4pwyKRggmcu0zePQeQ10ctYzmFsO3BYCDvvfee1HUt5+fnInK9nK0oChmNRiuCB16Ly+d4KZnXBxRXUJzjsWeBDfvojZceVyeX1oOJmHm3tFS9eB95fHlMWcRQwQx3IGO6CBXeMy1yPdYqjul9bKuH27eeUzyfEo3bjnFdVhye+OU95/w8eE4gL+4gCG5O/JEfBEEQpPB+T8Tvj+DrwNY4hLos8drU/ZOaXHWZzLYtzdHXljhgTTx5GVMqZnYQWa4Tvk7bV1EkNfll0QAn2JZIgu2jsMLnrf7pNSh6aHtWG9wOOnS0/N7enhwcHMi7774reZ7L3t5eE09VVfLTn/60yf+DrqGiKJr2+/2+VFXV5OlhQczaeY1FGIRFJe2f50bCZ9rK72QJPVgWz+E988QGjJlFLXQx9ft92dnZWVu6yHDScQSPW+M0mUyasdnd3V0bl7bPqSdIWp8VHleupwvsSNr0Z1Dq88HlQwAKgldH/DEfBEEQpEj9nogcQsHXga0RhLxJEE/4WCDZFE8k8v6b3+Va/I5leVJrCTRc1iqXqsO7XndvYlcLXsOJphWc6KMbhJN663nctQpj1TKecGDVzW16ApW+n8/njXBRlqU8efJEptOpPH36VESuHS96nW55703OLdHRyufk3QccP2scFEuk4XYt0Y+vtQQoBgU1PMbPp35hbh8sb8XfVXyxnnvcYU9zXbHzqO0zbolCqc9hmziUgvvfdn2bo8j7Geb9jAiCIAiCIAjeHCEGBV8XtkoQshw51n//N50k3URASgkYljOBJ/VdBR6vz10nhqmJOC5h8cYN3SfWe02EzCINikyek4qTFmO8KZGFHTU8NnVdS1VVcnV1JZ988okMBoPGcXJ6eipVVa0l0rZy82AOH8uphImwOY+TXmMJJ3xvsX9YxhIMPfHT+s5iB4439wfHEsUY7l9VVdLr9RqRhvvDsVvncQyt57mu60aoq+ta8jxPtpMScfnZSQk1eL+7lPfoco3l1EqJZF472McgCDbnr733a2v//Y0/8oMgCAKFf0/E74jg68ZWCEIHBwfmpIcniDf9b3nKqWNNKvk1HrMmaJ4QlGKxWMhoNGp2XNJ6UGzyBASrLygI4C5O6sawEkSjSwiFAXTs6FIrbluX+1iCSWpccdkcu5NUJECXCoozWH6xWMjFxYWcnp6u1KfjpMKG1oFJqnUZmSZYrqpqJR59z0mcsR1LDON7h/dUy1sJsnG8rB23eKw5n5CW1/vM/dXjKDBYQtlyuZTZbCZZlsl4PHbFSewviixt4LM8m81kMplInudyeHjoioApYdD7PPDyLrwOy1rnvPNWP9qOWfVt8jPMcu8FQbA58cd9EARBkCJ+TwRfZ7ZmxpGaKHVZWqGkJnPWxJUdA10mbJbjJTVpTsXHwoHl/Ej1hZ0emITYcnFgHCiQLBaLxjnCriLL6aATb0587OUEskQ4LcsJnlP91GPocsEy+sXuJGzPuucoJnm5k3isu8D3kUU/Fhg9RwgKVHyPtK+6NAyFLDzn9QWfORXlUIDiOPB+tIlG/CxjDCo26vJGvIbHj+tksG4WOVP3y/pctGHFxOJNSlx+1fEEQRAEQRAEQRDchK1wCImsO1esZR3WhNez7WgAACAASURBVJa5qYsodS0LD5u0520Xn2WZlGUpg8FgZfJqtcmOEzxuXaMum7aJZ1VVTZJldjnhkh5Fl4+x0KQTe3V51PVXeWssl1OWZc0uU1xORCTP87UJNucv0teei4oFAX6m1EGkbaOYgm4aro/dNV4yZ6yH+2c9EypioViGbWMZHW8sx8+mJRhl2bVzSp8NHBPO04TPheVUwfitflrPrL7W8rr1vG5x3yYuWVhOIeszY/VjE1cQ99mKI1UXC4JtbaTaCoIgCIIgCIIgeBVsjSBkiSZd/1vO1+FxS8BJTdisMuzE2WTC6jl90FGDk3wrHguvzxor57vB2NEpo8dYjEBHDbs8rDhQwMDlUJ4gpGIKjjNvB6/14S5jnjDm3S8LHmvuV7/fbxxIKhax28QTB7uIEyyoWeOLx1XMQIGI61dhjstonbgcTpfEaR3oiFosFo1YpuNgLdvyBBBL+LREMRYSbyrkes8lC1YYn/d8tH22uwo0qZ9fXfq5qegcBEEQBEEQBEFwE7ZGELIcMV3/a4/XW+esdrzjnuhgTdi7ik0sViCLxaJxxHAuIeuaVH/1vE7qRdbFDf7O4ojlWuHzPNlWYaeqqhWHkYoKWB8KHbi0R4UJjBVdKlaM+J1dVp4IgW4RFKdEvhKhBoPBWv24Uxkm2sa+Yt+5bUvwtO4nPmcYl8aA9wN3D0MXETqGWMz69NNP5fz8fOVavDfo8EIHVErMwGfBu1fW5+fq6qpxyeV5LvP5vFX84M8YPwvcFoqjXA9jxd/W/iZ0EZs2+ZkVBEEQBEEQBEHwMmyNIKRYEzyrjDeRZtoEFK98qk5vcpwSaTyhiJ0g3K7VV6sMt6dii+U+svqWErqwjNcu9gHdSSjo4PUscGDSZU4gzTHgMqmU+wbzG1n94qVhGJ+3qxiKLCwoeQm4LbcMnveEN2tbebyeE1Cji0jLo6gzm83k2bNnK0mmeQz1XqBg5Imt/DxuIlxkWSaz2UwWi0WT3LutDSuWNpFlE3fNJj8jXrVI01VcDoIgCIIgCIIgeFVsnSBk0VX80XPechYLa9KIk31vMqzXeDsapfrB7ahLoiiKldw93uS3zeWU57kcHBw0IgOLLuj6UFSEGQwGjVMnlccHBR92ULFIg4IKtstiCQsDfB95py28Pyr+aCw6hur2UfFJXUsoGKEbhkUr7IO+VhFDxTaNWb+rUwoFIk6gjfcUnzWNm5Nia/0q8uh4sQDE9wbHaTgcyne/+1357LPP5OLioomTyy0WCymKotmBjePge45j1dUFo3mutF9FUZifj9Tnsw1PtPJi7SLCdBVp2Bm2iTjFn/1NRK0gCIIgCIIgCIKubM0uYyJpR04Xp8BN/qtuTTa91ywMWLtipVxCKVcRb5OemmCz2MJ1FUUh4/HYdOZ4deKSJxU72vrGzhx096BrBcUNFWT0C4UgTbDtxWjFzzGgwJSKncUwPm6JZp4AheezLJM8z1fEG87ng2PBTiMeR/3K81wGg8Ha2GmMmI8Jd3tjge7o6Ej29vZW+sjjiG4tHhfPscbjk0Kf2/l83ohC6nbCceTx8j4/qTYtodITs1LtbNImtmuJZ13h3dKCIAiCIAiCIAheJVvlENp0yQlOQDedNHn/gfcmbilhBuvs4pDg6/X4fD5vxIT5fN55sovve72eHB0dyTvvvLMmrqAAgTtVqWPDcudYDhRM8qxuFp5oY1l0SbCrhe8fTqBR1FA3jtUOtsdjwcu4tF4Uvzg3D4p/eh6vVRaLReOoQsFB69PcUHqNJnNGwYzHQx1N+l3bRwHJE2UsIYiFr9FoJMfHx/KLX/xClsul5Hkus9lsJQZ1BWkMngvJEi75OzqW+HmdzWbNmAyHw+Z8Kom15VSy4HOeoGUJyVbfujp2vHKbOIQ8wS0IgiAIgiAIguBVslWCUArLfcBiwib/hffEnE0mbm11WvWnrq3r68TM6gTRyTJOUNlFYR07PDyU0Wi0IqLgeZHVpW4sBGXZdWJlS4DBa0XWHUAYFzpeUsuMrPHxJuvehBvz+mg5dnSh+0rfe/ViXfgdBR7ur/eM8v3BLe7VFWW5cnjsUQxiJ5O+VgEMnSXYVxWqWEzFNrAO634gKUEGxxKP49JDbYeTXHNd1ti2fZ6smFgM4tfe+654z+tNf6YEQRAEQRAEQRC8LrZKEPIm1NZxvs67fpO29fsm/6HvKiBZLgusQ7/rhLwoCimKQsqyXFkOZIkDKLjkeS53796VXq/X5IjB8iLXE3DN2cKiB+Yb0mU87DRiQcJyVHBuJT2m7hnL2aX1eMudsA2sAwUVFYYwcbK1U5bldOF7hPFpXSqAsSsIXT/oLqrrWvI8N+8B9lWdQPi8q0CI9xr7on2zxAx2gOmSxOFwKDs7OzIcDqWqKhmPx1KW5cp9wG3neWys+4bjxXgOIhGR6XQq5+fn8u6778p4PF67R1y3JdJ03RlMv7NIiJ8fq183xYqZz3tt8bkQk4IgCIIgCIIgeB1slSDkiQtdeJlJkyW24LnUZJev6RKHJZbod60HtzJviwWP4/ItPIfuE0v8shwjLFRhndaOVliWtyzHclifda/ZcaPfedy8OLktFVLa7g27hjBWFN0sUs4ZTTyNcVrLzERk5bu27bmOrOeDz6fi80S5qqoaQVDP873gOtrga7TPvDyNr2kbb+t+W9dbzxtf31Xc9eL0xGzvmlR9QRAEQRAEQRAEr5utEoRuKuq8jDMo1W7bBC/l+ul6HV/rCSss1nDd7H7x+uk5NnjXrtSkGctbZbq8t2JjLBcKizPs8sD3eA2KASzOeKJbSoxjcaFNTLBEK76vItIs01ssFq1CDCeFthww1vhgsmLrOq3P2hGNx6XtM5v6bKkgVNdfuaj4frWREqdYNLKewU36onWkHEyv0tlj3ZcgCIIgCIIgCIJXxVYJQoonBHjHXteEyXKsbFKej3tYZXUiqMmeeftvb6mMuju0DnYL6SSfkz2nxhNFI+4PO1y4/5gbh0Ukdt1YookluGC8/Br7zMIK7+CmsfJ28xib5f7gslb8Wi8KPSr2lGW5lp+Hl7cxWBfuxua5fFjEY6FrPB7LdDpdW7pW13WzhAxdO5bw2AaLj8xsNpOrqyup61qKolhph+vxBEMee+szaD03XWP0BFSPm4rTXh9fVlQKgiAIgiAIgiDw2Kpt5y28CZbnYEmVu0nb3uQav7Tsy7apgoDnkPDaYCEF6/CuZxHFEjxYnPLgJMseWG+bIwjPWzmA9L2VhJnFC3XEsBvGE/CsGDy6Pgu8hTiLXpYQxO3zbm7WffPq4PjUkaPCII6j5h8aDAYvJUhwnNg+jzkKlFY9XZ0yXd1AqZg5/tTPIC+mtp9ZXWMLd1AQBEEQBEEQBK+LrXEIoQjSRexBV4PlWMFyLxOT1a51jGO+SbvcDxQKdNcvnvBabgqMwRobXE6Frh1rgovCiiUoiay6lTwXCTs5RNaXqXG9DLeBLh6Fj2EsmtAa+6JjwI4ddjaxE4nzI2FsKm7g2GodKPxY9Xhjj8dw97e2MbcEkrquZWdnx9zavq6vE1kvFgtTEOoiXKTuKYtx2n9Nco7OJ+tzzG6ftrg4xjaRlZ+xVP3WM81CqhUji5FcB8YTBEEQBEEQBEHwutgaQSj1n3iR9eUhlmjU5lC5CZ7wYpVRUu6XrvFZW6LzxJ8n156zKBWX5xqyElBjGY0RRQ/+jjF2cUfoLmR8jtu2xjIlJFriEL7n8fNEMT2uceIzy9/5Wl3yxwKQ17Y1hniMhTwWSL1nUMUv3Hreqqeu6+RW8Dye3jPJ5Vj4VXGLl+hZ9wvHbVM855DVh5vSVTx7VX0KgiAIgiAIgiB4GbZGEBLxc+nge35tvff+y38TPAFhk3o9x5DlIPDECJFV90vKUYLCAO72lRLUUBywYrXq0PdeQmqti4UHFk9QfPIEFYyp1+utJV1GAU37rFuna/zL5bLZvY1jx35aDhZuD+PFNvn8crmU+XzexIhuJBxvb6mgJWBoe/qek2y3fWb6/b4URdGIQoPBQGazWfPsaCy8ixmTEhj5ObEEMBEx8xV5dW8qqlrCkiVWcb2WuNh1HNrK4jX8OfRExSAIgiAIgiAIgtfB1ghCllujDW9ShXW+CiwnBeNNePlYSthihwS2q1uXc0xWHJoDxnKcsOBgvUaR4VWP7SYuDE6cbLWNopOeYyEKr9E+sVCWul9WOV0WhjFg0moVn/Qr1fcuQge2paINj48naHjCDScmR/GMx8F6n6o79Xzg5xbdSNbSQ2sMupISTTe5rq3dTV1Gqc+u1X4QBEEQBEEQBMHrYGsEIXa/eLtoIdZElR0fbdenyqIgsOlkPuWeSGE5dFiY0PfsDNFtvDVeXtaEMXhjzG1Z/bXEBK07Nam1cj5pWXZkeHASacu1g2Mh8pW7J8/zNbHLckehYKF9xTY9F5O6XXBHOKsNFoiwDd6KPpUniYUwHV9vnEWud6HLsqwRYERkJedRr9eTsiwbAZJzLvG4W3jPFB/TMcMYvDqYNuE4Jfx0aaeLMG09s6kx2UTstmINgiAIgiAIgiB4lWyNIGRNsG96bVfaynb5rz+XeRX/3dfJPW8Nr+14ghHH1dXZgKKFJXi0OSCstlh4SiVPxmvaxASME+vGcrzduyUacRte3zARNZbHNtWxo2KLVy/Hbo2x50hKxWnlXcLrWRzUc1YOKBajOHZun4WONiGUhTt97y1P6+LW8QSnTeA+vGrR5ibCThfnWBAEQRAEQRAEwU3ZCkHo7Oyseb3pJOhNT5o8saSLi0iv52u9ZTIqCqnjBJfWqBMIr9MJ/+Xlpezu7jZ5Wax42fWC5azJv+Uowjqs/mGbupMVl+EdvLAeyy2G4pWW8xxNWXadH0fjVHGoqqq15UlYN9eB1/EuYdgPbwkXjxWKIFxW3TvoImInlN4/TcjM48f3RXcM07FWt1RRFDKZTNZ2RdN+FkXRLCOzRD2OraswhOfVVaW7jFn1W3W2faastvnZQdqcQClXkSfIYlku00VAfpUicxAEQRAEQRAEAbMVghCy6X/n9RoPa8L9MhOsTZwzqUmrFxOLJZ5LiOvFMhcXF3J4eLiSqDflivHGxpuQWg6XLljjxcve9Jj232o/VX+/33djwzY9N5LI9bIqnsDrVuyYx0lFIJz0azssPnAZz33jvbf6amHdM2sHLxWVPPEFHUKWIMVlLKcRP6epYyhSWmNhiUXemOA1lujK49Um8LSJzVbf2pxN3mcqCIIgCIIgCILgTbF1gpBI90S1PMFNiQB4zaugrR5PDLLee8dEvtpevG1iixP+s7MzmUwmsrOzs1KWBQpLPGAXCjtirO3OPRcE9827Dif83LcuYyeyKnpYfcH3KmRwvbyjmHWOhRBsU497S7g8Zw1+t1w4Xp+wXyxGcBmOqdfrmUvGMA5PsEJ4KRr3GWNiLCENx7pLvXyOn0evbYzBE0K9+L3PjVUvx+d9VlICX5sgFQRBEARBEARBcBO2UhCysASMLhMqhSd97CJoK2+d9+JMxdY2oWVRQ11CmCgaXSrW9c+fP5cPP/xQ3n//fdnf35f9/f0VEUSXGonYy7603bIsmzbrum6WHelxrUOXOLErBSf2vORInTw4JixM4Fgtl0sZDAbN+PCYsvDBbiMtp2NZlmUTk47HYrEwl55hnLg0TMebHU3WPWbBCttBp4637IzdOJYDKSXy8Xd0n/E1nmsI+8z9tAQQFrxwvDReTILuCXjWeHi0fWbbnDtWHZbQZIlwLGTi57mL6GTFyc96EARBEARBEATBq+LWCEIi6Ql324Sxi1vAKt9Wf5tzJxUT4zlLvLZS9VxcXMhHH30k+/v78su//MsyHA7NXcoU3vlKhRLMQ6TJk3GSivlurL6g0wTbwLpEroUQ3u0qJZJY7hoeL42BlyINBoNmWRiW0zrYoWI9Cyyy4GutgwUVFkewLY9NBYEuooonjOIx7luXdvE19ovPpZ5jSxT1YtlUMEq5cKzzKYePV0eXetoIASgIgiAIgiAIgjfB1ghCXZw6nhCUcvpY/7nn8x6WU8KKD7+zS2MT8cpyalj1q2CCggpep2LE2dmZnJ+fy/n5udy/f1++853vNOU40TIKOJ5LRp1BWk6TW/MYcawsAqCYpOV5CRO7gzRWrhuTVXO86PrRPusXikIqVOk1GhfGiqIUiy4sPFmCALuMUBTCNrUeLoN95v5aAg+6lix3SpZlUhRF0x6WUQcVbjtvPZPemFtl8N6jQIfuMxx3ry0WBS1BC9vk8bFghxL3h9tu4yaC9ab1BUEQBEEQBEEQvCxbIwh5k63U5FrfW69Tx7rE0uW6NhHKmxS31c2TUKudlFuBkwFPp1M5Oztb2UKdc8SgcKCTdm/C7Tkg+L5YS2y0fUyWreXRNWRtQ85igB7D45ikmmE3CTqSuE+W6OKJflyGE2VzrHotClS80xeLa3q9lvfEU47Fcgq1iZvs0OF+e+PbJoRabVv31CvrlUvhfT6xvjZXUUqIbiuH7dyUcAwFQRAEQRAEQfA62BpBqCtvYnLEk/GU8ONda5ESvTyhiyfgOllXcaBLAuPFYiFXV1ciIk3uHlwmxS4SdouguIQOD46bHTMi18uz0AXDrhdcRob5ZLBdzHejcN4efc25iXTJG+YHUtFGHUe6nG4+nzeiGbqf+v1+416xlsdZ95KPYVxFUaztMsb3DkUlHAMtpy4efc/iE9aLaL9xBzrOI4XClpWsm49xfzlZNWKJTuzCsuJn4beL48YTm1ICT+qepK5vE3w2cQhtKnoFQRAEQRAEQRDcBDuj8C3jppOnrv+1x8lc12Uo3I41Yec2PLeCnkvlUMLj3EZd1zKfzxsBRMUZvYbFFoyz3+83y3n0K89zyfPcXSKF9YhIs0QLnUscs8aGS6esmCy3jeXGYVTcEPlqWRy6ltSlg6IEbmGPwpkl+LDbivuAy8rYucOJr602WAzT/lqCnSdYeEIDx5bnuYh8JeZZgiTGZAlR3JaXKykl+nhuIu+8R+qz6rnOrPN8bFORp61PXptnZ2et9QdBEARBEARBEGzKrXMIKThJvelyDEs4SdWHE8FN2kzVa9XliTMpMcqatKNoMJ/P5ec//7k8fPhQhsNhkzuIxRQWplCEUvEBl30tFguZz+drooAKO0VRrIkRPOHWNtS1goIN5g2yxkDb4ZxKLBChy4hdUVVVrQgzWo8es8YWY0X3j5ZHl4we07HgccfxVdSdhE4iFGd6vZ4URSFlWa4JlikRA99bu9UVRSEPHjyQPM9lNput9JvHtE2U6fK8s8jqfba6CLDe58zqgxUjj7PCbic+v4k4zGU5tlTZIAiCIAiCIAiCV8nWCUJdl0tsKgJ5DgprcmjFYE0m8biH157Vljfp5C3X8bzl4NBjmGumrmv58z//c3ny5Im899578vDhw5X6tW4UNPQ6FWjyPF8RVPBaBBM3e5NcFFp4FzAUm3CXM/7OW9l744wOH5HVZNY8lizo8PInXNKGeOKO1ssuLCxnPQN5nq+JdNiOjnGWZTKdTldit54LHiPsD97rur5OHo7JwvneKpyHKtWm51pitxaO3SaCryWaskCG5fja1GfbE41S7W+KNz5BEARBEARBEASvi61bMtZlaYn3H/4u9VqTcp70ec4hLM/te3F47aXqxDbb+pqa7OKEVtu8vLyUp0+fmruKYV94ko5LvhaLRSMYIP1+X4qiWCmLS7V4jHGpEuePUREF60FhB4UpFLa4H1gX18PjpjHw0jgVtnRZGcaI9eBSLh53vE77xcKbBYtmeE1VVSsOKnQ24fVYh/XsYvtlWcqzZ89kPp+v9QXvGcduiUHYhvUcax3etdYYe3ifyy5OHnxO+MsTkNra74rnGNJz4RIKgiAIgiAIguB1sXWCUBdhBSdfqQlVl+NWHVad3vXsPknBk+JNJ3xtk2NrXPhYXV/nE8L3KKywCwmFGpy8q3OGEzVbApkVcyofUmqHMQ8U0LQveE9SeXqwv7isDIUiFg28mNpcOW07hPF9sIQyS0zkcnzeusYb47IsTRFP32OMVh+wrrZz1q5q/Lx2EVpSDhuvvymw3dcpynAbIQAFQRAEQRAEQfCm2JolY48fP5ZHjx4179vEDxYu8PimeK4Gq82uZVPxpJbRpGL02vecQezA0WtVEMK6cDt6dbSgaKPndIlRXdcrLiFMHI1xWMKA7tRlbRHP77F/3pixiwXLW/WLXLt09vb2VhxAZVnKfD6X8/PzRlDSMdBdyjTBtsK7jqEIoW2r44iXrPE1OGZ4P/A49jXP86Z9XuKFQgOODb9mBxLXZTmYLIHJunfaD/6c8rX4nFmfacQTiFJOnlTsKVKupTasMWoTc4Mg2E5OH39v7didRx++hUiCIAiCbYR/T8TviOA2sXUOIeUmy0ReZdspF4cXi15rnWMXBLaVatea2HOsbXGiawbbSU2+rWt0W3grbnQPWVhuDW7Xqzvl8EBnEn7xjmEsKmm9w+FQ7t69K++++67cuXNHjo6OZGdnp9lFjceKhRmsm7eSt9w7PA6p+8dOGTyGZS1HjScSpurGcm3CS0oo4bHp6vKxsJxNm17X9Tjf6y7Xpdrne9Ol/ZdtNwiCV4slBgVBEASBYv2eiN8dwW1iawUhkZfLoZGa3LVxE6GpTTRItWXVlaq7SyzW9Sry4A5W1nVeTFiHOlPYjcN9s0QKdsiwIGFdb23ljuW9ODj3DseR53nj3pnP57JYLOTw8FD29/dlOBxKv9+XwWCw0n5d141jCLerZ5cSx8TPlR5nIQxjxJ3PWPTCceXt7LU+SzDSe+cJk3rNYrFochS1iZncplUWnwUeBys5Ofav7bPcReS04upyDo9v8vMkJfBa9d/UjRQEwZvn9PH34g/+IAiCwCV+TwS3ha1ZMmbhuWG6Ltm4KZu0k2rXcpJ0cRltctwTYhA9r8JDVVUynU5lZ2fHdYpwvDg518k7JkbGOPSYyFeChm5zj8vEUk4SFA5wmRrHYrlbdBt4Fk2wXFVVcnl5Kf1+Xx4+fNj04fLyUrIsk93dXTk+PpaLiwuZTqfS6/VkPp+7IovVH3QqWWW8Xbp0nBaLhZkIGq9DIUj7jWUtYYrvrfVMobMplZCdxz4Vq1WHFUtXEWXTMryjmzUGnuDT5tZq+xy2CUnW9UEQvH3uPPow+Qc9noslAkEQBF8/4vdEcNvZKofQ48ePk+e9/6K/jJNok3a4TNsxdoyk3EAsWFjijPc+Jep4sWkeIQ+vHktcsJwpVnkWPPhaayw2wRoTPY5jo4JWWZZydnYmJycnMp/PZTgcynA4lMvLSzk9PZWrqyvZ2dmRo6MjuXfvnhwcHLg5daz7zOKK54RKCRNWgm3vPe4yxo4XjslLQu31DceR4f5zPV6/ORZ2EFljssmzYQlUfL5rvW2fSc8t1VZ328+ztp+JQRC8frr+AR//CQ6CIPh6Er8ngtvMVglCTFeHThcBR+kySbOu8Y5hXSkxhieflrDilRf5Kqmx5aDxhBCrPpFr4eDp06cymUzW2uHJOMbKCZStybsnAuDEX2NmcYUdNHgt90fHgvMH4XVVVUlZllJV1coX7kI2m83kww8/lJ/85CcyGo3k4cOHsrOzI6PRSGazmTx//lyeP38ug8FAHj58KN/61rfkzp07TSJq3ZUMx0qP53m+cn8st49eo3HhGHA/+VmzhEbtFz/jLBLx/ead1CxHFj9v1rPoiaL83OB5kVUxyxOWUsKjRdfPOY+NJSLxsU1+5vD1GLuVN6tNsAuC4M2zyR/78Qd/EATB14/4PRHcVrZ61vEyrh9rUpmauFrXKN5kLrVEJFWfRZtgxMIPO0HaXAh8fjabSVVVaxNhq28sFHgTYU+c4frahDPsE+buSYHj5n1Z7WTZ9e5i5+fncnJyIovFQkaj0cpSr+VyKRcXF3J+fi57e3vy3nvvNcvtLFEKxSDPqcMikSUsWsIEO5D4fKo83wPOi8TPnJWIPHWe7we/t8QcLMd5hKxxsPr4qrDEKI7lprT9rLDKv+r+BUHwcmxi9Y8/9oMgCL5+xO+J4DaydTmEuggcm9TD7z1hha+xJmSbuHLweJvDwXKNeOJSnucrDhevPF/L7T158kSqqpJf/dVfXZkIa8JnzfmD56ydxFQAQWcMjpU3EUZHCPdbc+cobduSo7iC28Rj23hOwb7O53P54z/+YxkMBvLDH/5QqqqS09NT+fLLLxtn1GQykaurKymKQu7duydFUchPf/rT5l5o/SoG8dirwIRJnVlk6/V6a7ujWQ4dHScUMLSP1rNujaHWzzmCtB1NQM5jbblXrPtsfbb4OeVnynpmLFJiUepnSOrz79WTEje9GNrELO9ajOeDDz5IXhsEwZtH/9jv8oe8lomcEUEQBF8f4vdEcNvYOocQLnfZZGLY9Zw3mfPcJFyWxZ2u/8Xv4nKw2vYEK+zLJhNaFGQmk0nSgWOJGla9lnPIuwaFIGs7dhwHFiisvvN4cW4ivr/6WndKQ+GoruvGLdTr9WQ8HsvBwUHjGNJlYsvlUsqylNlsJkVRNDuNoSBmxcbijnesq8iBywh1/Nt2juPrORYu2yb08FinPhOe28gbn9RnA8WxLuPMMXB/uvy88cbDou1ng9deV0EsCIIgCIIgCILgZdk6QSglAig8oewituh1qXatSS1PFlMTtq4TUxY8OF5sm9vDHbe89lgssSbfi8VCLi8v5bPPPpOyLBtRQ/PsWJNtS1RAASQVt1cHxsQ7kzE8ZiqI4HkrJ4tVR5ZlUlXV2hbsdV3LT37yE/n0009luVzKzs6O7O/vy3g8lt3dXdnZ2ZE8z6WqKpnNZvL+++/L8fFxIxZ5bYqsbx2Pzh6NQcccl0+xY0jrKopC8jxfS8qMr1Pj7Tlb0IFkjSWPc0rcSI0/xsaOMStWTyhL9cOLI4UlkqXasmgTdTb5uRUEwfax3aPSfAAAIABJREFU6bKAWBoQBEHw9SJ+TwS3ha0UhLpMiCw3Cr63HD+bTOi4Hcstk3ILpJwZlkDEZTx3Uyr+TR1Ly+VSnj9/LpeXl2vnvJg4sbSW57I8bpwfRt0sLHZ1cZFYbaaEgS7HMK6rqyt5+vSpnJ+fN8u/8jyX4XAoRVFIXV8vm8rzXESuRTHNKWSJWd4z1FWc4bhRkOHrLREwJdbweXU56fNhCT5WHim+lylSTh/rc9vFveP1z3u/SXypuq3PiZbpEnNXcTkIgu0jLP5BEARBivg9EdwGtk4Q+uCDDzpNyFITfZ1kaj4Wzkdzk8klT+LblrSk4sRYsU5rFypsy3LtsAukTTTC+uq6li+++EI+/PBDKctSRKRxuvAYWnViMmXulyUAYdtaBs/haz3HAgs7ijBnEY4BHuP2dfkX7hCWZZkMBgMZDAayWCzk+fPn8sknn0hd13J0dCTj8bhJND0ajWRnZ0f29vZkuVzKfD6Xsiyb+nRscQxxTPE+YlmM03L44LPM+YI0j5OOnfecYt2eANnr9WQwGKy0a13H59C5lhJxsE+cvJvPe33oihdLytXkfc44JjzXJhx5hEsoCG438cd+EARBkCJ+TwTbztYJQiLpJRubwhM7y5GQus6amHeZxPEEP1XWmqx6YpMKLbxlu9eHVN/09dXVVbPDlte/1OTeu19em54zxhLAvHJ4nifw+Npzs+C1nNhYr59Op/Ls2TOp61p2d3elKIpGNLpz547s7+9LWZZSlmWTjydF6lm2RD6+x9YY6LPQ9nymnjduQ7/QheTdEx6z1C50Vkzec9PlmepKSlCyPntW+9bn+VXG2BZXEAS3m5gQBEEQBCni90TwtthKQchjk6UcXN4qm5oosqMj1W7b8bYlL1qmS4JnkXXxwEvO3HVivlwu5enTp3J2drbWb0+EsOLA197YertdcV/ZfWSJPRyXVQe+57LqMGIXlLaxWCzkyy+/lCdPnshwOJThcCh5nje5hAaDgcznc6mqam18vGeRxUWOCYUsa+xwbHkMvdxL3ljxs9PVkWMJrHxcz3lt4z1g8cmKua1PKbxnpq0uzw1kvbbed43vVV0TBMHbIf6ID4IgCFLE74lgm9lKQeiDDz5YyV9i0SYOdV0iYl3TtsQkNSm06vQm2CwSsONFsUQlS1DxXEksdPB1y+VSPv/8c/nTP/1TOT09XYuHRQirD9iW1R4eVzeNlRwbJ+FVVTXCjCbT5uV/lksFx4dFGlxy5bk+1PGjO4792Z/9mfz85z+X8Xgsx8fHUhSFPHnyRH7xi1/IxcVF0x9LQOB7gGOE99V6jnjMUTTEcdVreRkcjyffA/2MYbLruq6bOjTptvXs4T3Ac1ZyaL43+h53Z1ssFo2wtqk75qZumpsITZ7gmhLGuvzsYR4/frxR+SAI3i6pP/ZjIhAEQRDE74lgW2kVhLIsG2VZ9r9nWfZ/Zln2f2dZ9p+9OP6dLMv+ZZZlP8my7B9mWVa8OD588f7DF+e/fZPAPAFCpNt/5j13QptTx3tvuRvwtScmpSaDXF+byGX1x4Jj8UQBLLtcLmU2m8mTJ09WnEJWW95YbhJ/27ioEIRJrLE/npuFxRMsj7t5cZ3YV3QnqUjxySefyEcffSRVVUlZlnJ2dibn5+eNeOQJmF36mRoj75m0RDcUWLz2UcBZLBZrOZiwfl4yZsWHceLYcRnvedDcRyh0vQl3jCVyvUy7bc9+l59jQRDcbuIP+iAIgiCF9XsifncEb5tBhzIzEfkrdV1fZFmWi8i/yLLsRyLyH4rIf13X9T/Isuy/EZG/LSJ/78X3k7quv5dl2d8Skf9SRP7mywTJgskmEzdroovnLCeDJfK0iS9e/V65Lv1hoQKPd3U7YV+svqMgslgs5NNPP5XPP/9cHj58KHfu3JG7d++uOGvUtaPX4BIw6z6x4MJxq0uIBQh0A2GSZESv4aVSvV5PqqpaEXQ8EQzj1GNcX13Xcnl5KT/96U/liy++kH6/L/P5XJbL5Yq7BuPy+optcJ+tcbN2E9P6MVZPXOKEzSJfiTBVVTXJqfF8nucyGAwasYvHguOwzlvx6nl+JjFBtrVbWUoY6yqqpJ5BL65N6mWRsgtvQvgKguDNEn/YB0EQBCni90SwbbQ6hOprLl68zV981SLyV0Tkv39x/PdF5N988fqvv3gvL87/1ewlZz7WxFiPt12X+s//y07c2pxKllDQ1aHkTVpx4pnKs5OK22oTBYiqquSzzz6Tjz/+WJ49eyaz2awRLzwxyjuufU4tJcKyWJ7dLlynFQMKJJZApdd6Y80uF0zsvFwu5eLiQi4vL9cSSVvuJRZ4LLyYvH57z5PldML+oYuIhTq+d7iMC8tYMXikxldRR1Nd1ysuKy5rjV8XUdSLxxKG2sbewrrPXqzWtV78sVwsCIIgCIIgCII3QaccQlmW9bMs+yMReSIi/7OI/L8i8ryu6+pFkU9E5OGL1w9F5GciIi/On4rIXaPO38my7A+yLPsDq02eFHVZhmEJMh6pCTqWSU3cePKdEp64zE3EKi2PdaFwYbXD9VliEYsnunzs6upKjo+PZTgcNvlkEBWPLKEHxaN+v7+2/Ihz4Vh1aJ/6/f7Ktuoi0iSDtspjkuiiKJrdwbQ9FjhwYq+uG8zHw/FqfiFdPlaWpcxms8Zpg44bdb8o2n5VVY2LCdvm541f8zPJzwKKaNbYovAyn8+b/uDY53neuKzwnlluH4+UgIOC5mg0kizLZDqdJoWUru2mylhjp89Lqt62zzV/fnj8U9cGQRAEQRAEQRC8LToJQnVdL+q6/jUReV9E/lUR+UtWsRffrVnO2qyoruvfrev6L9d1/Zfb2u/q0OkywbImpl45nDBuOln0YvWcGdbEPwUud9L3KfdLm1Bk9Wt/f1+Wy2UjIlhlRL5KEp0Su6xlWFpHyp2RErmwLus6FUd4FzF0oljjbzliUBzCtlDEsZZX8b22vnipXFvfrXFA94+3jIwdN9ouC1G4nI3FrJSA54k2/CxjfHmei4g0Y8fxdhGCLVKfAeuz5bl7vPJdaHtugyAIgiAIgiAI3iYb7TJW1/VzEfnnIvLrInKYZZnmIHpfRD598foTEfmmiMiL83dE5NnLBOk5B1ITvLZ62v5Dz5N+ntSnYuvi5LBeW+6FNry4FJzcW2UsYSTLMnnnnXfk29/+9spEnZd9aX3qNEFBgceobUt0rR/FGn3tCRw8VhiX5ZjC5VIsqHEcPC7qGMIx4utZhOB29Jg6b3h5G7uVOC6+19Z4sGuFBSIdA3U24U5vWD7LspV7bwlLHKOVhBpzQbFg2O/3pSgKybKsyclk3YuXhePFZ9H7nG5Cl+tTP6NCMAqCIAiCIAiC4G3QZZex4yzLDl+83hGRf0NEfiwi/6uI/Fsviv22iPzjF6//yYv38uL8/1Lf8F/sjx8/NifZXV0uFiwcbELKdcDHuB1PiPFis+BrtHxq0uzFlCLLMjk4OJDRaNQpdm/i3iZu8XmrfFv/tJ42h5UllrTVwwKG5d7RMuhG8sbEEgc3uXccp9VnTFhtiUtdnn1djte2yxi2y3Xra28MtLyKbFZybuuaFJ4oatE2nla5trq8662fWQheF/mDgiAIgiAIgiB4U3TZZewbIvL7WZb15VpA+kd1Xf9PWZb9PyLyD7Is+89F5A9F5PdelP89Eflvsyz7UK6dQX/rZQK0JrN8rk2o8a7vKiR5LpC2+vm45yzp6lxKuVFScVk7VbXFfnx8LHmeJ90P2JeyLGU4HLqxeG4SdqhY/WGxSMuoE4iXy+kxFIHqupayLJs68BqOgXcaw53QsB129uj3fr/fLKNTlw2X0zFgtwrGYu0QZt0vSwji7dxZtNA8SKPRaGWpW13XUhRFM14oKqWeGz5vOZe4TJ7njeg4mUxWrkOHER7b1GnDny1P9PPw2rOe0bZr8Lq2z2wQBEEQBEEQBMHrplUQquv6/xKRHxjH/1yu8wnx8amI/I1XEl0LXYQNkXbBpAssCKTKcDzWZPmm8WAc1mS5beLeJebBYCBFUTTnWazhfnltp9oU6bZtOh/nOFJiFcbkiYZYj4pIg8HA3H5er+F7qu+trd0tIcsqY42DJ2a0uXUwLq4P49D+snimXyn3kndfrNeWOCSyustYVVXS6/XWElxz/zbFG0+O8yb1Yp3WeGzycyIIgiAIgiAIguBNs1EOobeBt4Rik8nhTSeYOKHXCbu3E1TKIcGuIE+w8RxQVpyDwWDF9bKpc0LhXDu9Xk/G43EjiqTcFKPRqHERLRYLmc1m5th4O4xhziBcMmSNF+ehYXcO53nCa/j+qNjjuV/atkC37pUVD/ab36t7iEUbHAfsB4s1/IUuJmzTEnfqum52RcOx0Rj42cKE3Hq99TxY8BjiNcPhUIbDofT7fZlMJmtJufH6NteNF48nMnIsmwo0/Dzwa+s9HrPiiuViQRAEQRAEQRC8SbZeEBLp7gQSWU/q7F3XRVDyJsE8qbMcKJ445LXDZdjpgWVxSZR1TUpc8t7r1u5ZljWJfhnLnYTbseuOZBwzgkuYtB9Yh373RLeUQOW5htr6gn3ytiD3nilLAGDnFN4rdhN5z0ZKWGgTGfAZt+LVJW16npeVYeJpq05+tm/q3FGHkD4DnljS5qh5GcG3a9lNQGEydX9fV/tBEARBEARBEARduBWCUBd3gMKTfJ6YpQQFi9R/9L3rrQl7Kg7vmhSWK6JNhGKRSb90iVie5zIYDJpcQG3tV1W1In6IiFRV1eyiZYGCEYpKKlBYfbdECTyH+XLQ8cKuIXTa6HvMu8NOHE+ESj0TeH/ZWYPXoCsHY25zuaDglGofy3uxoysJy3pixaaihSdc6XEVIEVkRaDi9ru6d9o+U1a5LmItxswiWhdBL1W38qMf/ai1fBAEQRAEQRAEwaukS1Lpt443QeWJmufo4TraHAdM205QOElPtd+GNQn2Yk3tymQlEtbXfC7Pc8myTKbTqYiIHBwcyN7e3lqiZowREzlbbVdV1ThR8DqcSGN/tR9Ypye88H30xoVFoSy7XqKlO2ixAIUixXK5XNlinpde8fI1jguPodiE13jii9aPy+pExMxnxDmksIyKLXi/9T0u1WMhRgVCdqCxUMMuKOs55ecOHUgi10vGNIH1dDp1xc1XCX8mugqvXlxtAh4ee119CoIgCIIgCIIguAm3wiH0wQcfNK+9iZXnRsDrvLIeKecGT4K5rdQ17A7x+uL1tW0yi0ufODZ2PGRZJlVVNcJBURSyu7vr9ktE1kQGrFtFBv2yHDEqeuR53nwNBgMZDAYrfed2UQDR91y3vsf2UVTRa1QU0i9L9GMRxxJDrPHEsUAHDJ/H6zgHEMZqtY/3AJ8jHCPNBaTl0EGl+Z7Y6dLv95uE4uoAs+K1hCBPCLGe4yzLZDweS1EUslgsGhHRwxNW8TPF7afcP13Epy6CkQV//jjWIAiCIAiCIAiCbeBWCEIi6wlXUxO1TSaGqevbJqiWc6mLiMOiANfZJT6uiyfGnmiD6GQcxYqiKFZ2GLMmztwmxyYiUpZlIzpwWcxXxEmt+Tv2g91FLMjpRJyXhqlAVFWVlGW5stRNnUAqUO3t7cnh4WEjpqijiNvDnblwrFjUYcGnLXk29tkbcwTvDwpImJeHXUT9fl/KspQs+ypPELYzGAxWxCC+f/iaBSxP4OQ+i1wnJc+ya+fWfD4382Jh+17/LZEqVQ+f9z6zvPtaVzgnEsaKMejrSCYdBEEQBEEQBMHb4FYsGesKuxm6CDvWtUibW4dJuQ02xXJDtMXTJiBwvTwJV7HGmnizO0dj0N2yeNKugoPWi/Xo8i0PS2CxYrKWtrGTh3MJafsqoGifHzx4IEdHRzKdTuX09FQuLi7Wdk6zBCDutyfY8fizYGPVj+OBS/XY0cSiDn4XkZX+1vXq7nRaN9dn7frlkfqsefdwMBhIr9eTsiw3asuCRSHvPMeQOpb6DHl0+Vxa9yoIgiAIgiAIguBNc2scQoo34eQJacpt4NWbmqR5E0p2AHiuCivel3ExtLmNUtR1LYPBtRaoE/Fer9ckl0ZxgCfz7EDCcbfyxGjC6Nls1rhA1JVkCQGWwMLtWIIKjgc6VaqqWtlG3nJ8FEUhR0dHsre3J7PZTLIsk3v37skv/dIvyXg8boQLRNsty7LJm8SuLD2PS9JwnFBQ4txOuPsai0Q49pZwhNfguKDwM5/Pm3xJKOqpI2o+n5vihuXISWG5ZLSdnZ0dybLr/FWpnFhtWEIkx8Cf2ZcRd622rbasnwU8puEOCoIgCIIgCILgbXFrHUKppSltbh7rPE4mrUm2tfSD2+5Sjq/htlP1YyybuJ+8tq1rdSmXN9HnONom157LgkU8z1Hj3Us8js4ZFogwBn6tDAYD6ff7cn5+3ohlKlqoGGQlk8ZJPsfLopnlIhJZTxptiQbaZkp8ST3PeD06jDB5tV6jYhrvFMfPncbOuY9SMeHxLMuaccbE1t51lmOM20Exjeu5qdCEop1VT6oNjNcTN4MgCIIgCIIgCN4Wt1YQ8vKVWMKDSHrilnLiWOWta612ujqJrHhSk2kRWdkJq+0arx6crGoOmNFoZMaJ+VQwB5AlHGA8OjnX8sPhcEWUQOGEkyDrd00AzWIPChjcvpLnefNa8+JwTOyYUrdPnueyu7sry+VSzs/PV8QBjYudP3pM3UG6/E7HSseP8wRpviFeRsbOGc8RxOPCThWtX9uuqkp2dnaa+6fnVRzTsfKcbJ7gxs+OJ+70ej0ZDodS17VMJpM1J1qb6MP18Vh69WAd1s8KFtA2EW+8cfLabHNXBUEQBEEQBEEQvE5u3ZIxpW0ytclEDidtPNnusiTGet0mVvF1qcmwVU/KfdB1omlNTnEXLitOjHcwGEie5ytJnNnpoxP0fr/fbHGv5XWZmraFy9fa3CXeOeu19s9yP9V1LfP5XKbTqdy/f1+Ojo5Wkk/rDmg8sUd3Er7X5WG8fE7bZ1cR33t08XBfrfG17g3WhaKQfqEIhOdEvsr1hEJfaswtvGeT+6D3eTabNXFgH1J4nyWv33yt9ZlBEeh1u3jCHRQEQRAEQRAEwdvmVglCjx8/7iR2sKiyiShkHedJvFXGmqB3FXq8fDGpiau1LKatLyl0cq6iDW/FrhN43MWrLMuV3aEs0aQoijVRRfuC9Q4GgyavEIsl6HDxXDAquqCYxK4MzV+k16lAJCJycXEhWZY1gtDFxYVcXl5Kr9eTnZ2dRriyBBy+P7zDl94vHU9dIoVjyTmvWDTBfliihbbFuZ1Y6NAkzjpmnJxaXVq8+5hHSvi0RBn9PhwOm2ft8vKyiY3H0oPHpEucljil31OisNd2G6k2N6knCIIgCIIgCILgdXCrBCGR1zOJ6uLmsQSeri4lrcM6n1rWknKCeHHfdLLKuWAsh40e123CebkYllNXEC4vsxwjVuwqmFiTcx4znmBbAgHWq2hMKtz0ej0Zj8cru17psjEUjlhA4TZ53FDowfZ5m3ZvvD2s8cH2LBFG29Gla/yl90xEVpaMYR2W2Gq1w3A7KHRpAmtuw3uWLcePdy51bdvns0sdqXNWH1LOpSAIgiAIgiAIgjfJrROE2nbl6eJq4PJeHV2XrqTq8MpxGzdxMXmCiRVLW/ye2MVOFHWhWBNfrEdz0bT1BevWJM6c6JjHzduRKjWG6kDp9/uNa0mP9/t92d/fl/39fZnNZo2jablcymg0kvF4vBZzFyHPcp1oniFP2FPBBEUdFlw8oYHdQDwmmDAa7xlem+e5LJdLmc/na23qd+v+t42/1qX9U3eQyLUgtIlzxirrCYHWlxfbq6BL7Pr1ox/96JW0GQRBEARBEARBcBNubVJpxHOMdBGHrPKemwfLW2IET5JZSPHa7YrVz037aAkROkFXwQRFCas/KqbosiK9XhMW83I2vBbFDu6bLhkaDAaNQ0WFIgXz3jDcNi6fUrEJx6zX68mdO3fkhz/8oWRZJn/4h38o0+l0ZXexyWQie3t7sru7K2dnZ0386CxS8jxf6RvuwqX9w/5gniEcJ+seWfcZ68WxtXbt0tdlWcrp6WmT3BvPDwYD2dvbk+l0KmdnZ65otSnYviZDv3PnThPv2dnZWr88sc0S4/Bes9Clx25CykGU+txZ7fL4hRgUBEEQBEEQBMHb5tY5hJjUcpWuE8EuyzfahCKv/ZSbwYtjk2UvXetIOSnwOj3HdVhLfdgF05a/B5czeaIHxqC7faEA4/UX29S4OEE2u6Hu3r0r3//+92U8HsuzZ89WXDEarx7b29tb2TEsNfbslsL4MAm1N1Z4nvuXciCh6GTdfx3TsixXnD4cn7qI9FjbZyv1vOl5jltFxeVyuZI3Cq/h+j2hxRvrVPxcj3fe+9x3qde6nymXUhAEQRAEQRAEwZvkVjqEHj9+LI8ePRKRV5NTKDXB90QIzymUqocnq5tMXlNLdLo4IvB6TSaMOW3UWYLHLEELHTFYHyYoRkFBk0Xr+8Vi0YgWmuCZRSUVNlAowB26OCZ8j7mBtLyKSbor2ne/+11577335Dvf+Y70ej35kz/5E7m4uGiWhp2enjZLy4qikCzL5Pj4WObzuXz66acrOZEYvoc4LpagxuOGdaooxUvIvC3q8blioVLjLctSZrOZ9Pv9RhjSMtpnzA+VSl6OQox1zruuKArZ398XkWvH0nQ6XRF0rPuLnyGrv/xZsAQ3L0bv3Mt89lPnXsZtFQRBEARBEARB8Kq4tQ4h3HGsi7uAjyOeG4GFG8+947WnWC4hfY0T8i4uJWxnEzGsre8qEFn5eyyHBraNrheNS8WMsixXhB1tBxM363fcxYvLV1W1sozM6pc1JrqDWb/fl3fffVe+//3vyw9+8AP53ve+J71eT6qqktFoJHmey2w2k6urq8a9osmzZ7OZjEYj2d/fb/qWui/arh7XJWLaN8zB00WU8FxBqWs4Pv2azWaNEKTiH8bZ7/elqqo1YbANT8y0juV53ghtZVmujSdfa4lb1vuu8Xbtj3f8ZQWdWC4WBEEQBEEQBME2cCsdQoo3OU4JMKl6+LXlgOA62P2DZbwlLpaAkHIkiHyVH0bBpUGeUMV9xC+dhGvuHxVldEtyvlZj0RwwvMwI41GBSJMX62trORM6UDS5sG59juKSOpg0Vr0G8wthvf1+X3q9nty7d0/u3r0rd+/elV/5lV8RkWuhZzqdyieffCInJyeNY+bBgweyWCzkyZMnTT6lq6srqapKvvWtb0mWZfLxxx/LdDpduS/aNj4LmM9IxwdjY5FRy2qeI8uZY+UcwuPYNl+L/ZnP57KzsyPz+Xzl2uFwKKPRSM7OzmQymayJom3CCz+jjLYzHo9lf39flstlM774POFr7hvfZ3zfVSBtExHbRB/LyeS1s4loGwRBEARBEARB8Ca51YJQ20QQJ+ibOCo2WUriTZI94ceKgye7VhkrUTMf30QUElkVKjQRMW83jrHzluYsfKHTg5dH8WuMWa/XJWRal57XZNXoIGobv36/LwcHB/KDH/xADg8PZTQaNQLD5eWlfPbZZ/LjH/9YqqqSq6srybKs2VFMHU26G5nWPxqNZDgcNu4mfg48wYITHvM1eE9RBOPleZ4zimHRD2NQoQeFo/+/vXeLsSTLzvP+fU7EueTJS1VmVnbd+jJT3SR6wLFleygQEB8Eyg+sGkKjBwqmoAshENSLH2hAgij5xZBgP+jFFAwbBgRRMCUaHgpj0zNkZz0IIjkUHyyqZ0ZqzaBnWt3V03Wvyvu538MPmStqnXX2joisqq6syvw/IJHnROzYe+0dkQXEX/9aW8aM4xhxHKfzs+PaZzhrzho7x0qlks5N123S8w85/vS4RQTbPLJEWD3vvL97X3y+Nnm7JBJCCCGEEELIi+KVFoRu3ryZ1hLyUeR/8rNesn0v7Pbl2NdP6LsVQOy4PkeGrh+j49E7Stkxs9wcut6Pdf4kSYJOp4O1tbW53cB8beW4TokSoiiac6xMp9PUASN1ePTaiGNF6tfIeDoOPTeZR5Ikaa0imde7776Lt99+G6urqzM1d9rtNv7kT/4EOzs7aUyj0Qjj8RitVgtJkmBhYQFxHM84kb7//e/j8uXLuHjxIu7evTvjErJzl1Q0LbzIrmUypl1T+4xY8U6LN3YdbD9arLLPQb/fB4A0BU/f00qlglqthuFwmAqEvj70fbHunSykzdLSEur1errjmbjKrBiUJ4CF4vIJVk9D6P4WuS7PXUUIIYQQQgghJ80rW0PIhxUPgKcr7qpfxH1uhVDNkyysuyDru27vE4NsXFkuB/3digxasJA5iRgg/Wpnj+5LxxKqqWPX0cZs+3+al27tqNHzW11dRb1eT4WnJEmwtbWF27dvp4LJYDBIY9f1kySlq9vtzjhqAGBtbQ3Ly8teF5VvrQTt9rH3wa5bkbWw4qL+7Xsu5PNoNJopZK3vgbihxCGl+wsJnHrs0Nzs/a1Wq2n8IlD5/qZ8/drjReLJ6jfrWB55Lq2sGAghhBBCCCHkpHnlBSGdgvE80kV857OcClqYyHo59322rqFQPD6ng915Kg/pX9w84qwRp4jE0mq10O1202usG8iugQghct4nHNjUJ/vduo10upl8F8eR7lfOSxvnDusPRVGEg4ODmV3MyuUyHj58iHv37mFrawudTiedP3C485W4eNrtNprNZlrf5uDgIO1/Y2MDly5dSuORAswiStm0OuBQYBJnlL7fWujz7d4ma2MFDvluHUcAUjHPPhNSe0nqMIkQo/ut1+solUro9XrpnHSNpNDzaZ8Hn1gmlEolLC0tpfWMxJVl3TRZImRWHFbsDIlH9nxR4cj24/scgulihBBCCCGEkJeJV14QsoReGPOEkyKuAd+Lrx3bpvT4HEZ57hAf1m0TijnUlxVmbB8S53A4nEuHyorJFlXOil0LPL54dTpbkiRpTRueNwOmAAAgAElEQVTZJSxUQ8iKFc45bG9vY2tra+Z+PHjwAO12G+VyGbVabaZGEIA0XU1+i2Bk7+nS0pK3qLcW0HzioO4j5Jiya2K/hwQO+e1zsMl9kh3bJB3O3jPZ+UvPOWss7fIK4ZuX7OLW6/XSGkJ2jqHxs47L/LP+DbBxZf3dHUdgzhLMAIpBhBBCCCGEkJePV7qGkLC5uZnWEgq9mFmXj94RKeSAyXIthJwFtk3ei77vWJZrQVwwsl13yD1iYwdmd8SqVCrpy7i+bjAY4NGjR3jttdfmdrWy4oV2BumxZH313CVWcdFIKpe+Vtf6AZ7suCX9WleUb60l9evevXu4c+cOPv74Y/zkT/4kNjY2cO3aNXz00UdoNpvplvJW8EmSZE58KpfL2NvbQ7vdxmQywcWLF/HWW2/h/v37ODg4AIB0dzYrBpVKpVRo0Wtn781kMplzEem1s9f67ofUQtLXyH1PkgS7u7tpPBKvvvf1eh1JkuDg4GCu4LT97NvxzCeQ6s9RFKU7jDnn8OjRIzSbzbn5hPqzMdgx8jhO2yJ9ZQnGofgJIYQQQggh5GXh1DiENjc35wQei03T0W3zXDdZzow88txJPkJ9i4smLzXH9133KXVyfGKUOIRCYpcVebIELOlTtj3Pcillzd3er6w0NR3To0eP8Nlnn+Hg4AD1eh31ej29zhbWFvFHC1e6/1KplG7bfuXKFWxsbMzE5Fsv31x8aWW+Z7KIoFjEtSbjDQYDbwFwIY7jOZHRjhGKI68GkCA7jJVKJfT7/ZkC4vpehOpShfoNxVUU69jKI+/fCX2e7iBCCCGEEELIy8ipEYSErHSRPJeP/R9+XxvbZ14svs9yrf3J61sECxtXSHDS89XtfEKS/q7TeIoIF3Yc33rqukBFU4OsE8gXg10v+2I/GAxw//593L9/H8vLy3jjjTdQrVbTtr710k4qHYvU4JlMJqjValhZWZmbh73n2onmizW0flniWoistZxMJhgMBnOuKz1mpVKZEWOsAPg8HDZSe2k6nWIwGMz8XfqeJ8tx/w6zBFONCGHP6jh6ni4kQgghhBBCCPk8OVWC0M2bN9PPNnUn9PJmX+B9gkbWtVYwCbk8Qtdnjenrwzk3s6V5njtHr4fgcxjp8cfjMZrNZuE+tZtDjvlq1Iigpev2TCaTtG6PXK+3qdeFmqUPO552ffle+tvtNn7wgx+g0WjgrbfewrvvvosLFy6gVqultXUkNl2culqtYmFhAY1GA9VqFVEU4cqVK1hfX8fCwgLW19dnil37hC6fuGGfG9/994mT9nnR89a7pNnroihCp9NBu91GHMcYDofeZ1ccQjrtzLeuvnhDO6zpdSiXy1hcXEyPt9vtmWfZJ9haUc7+LWSJOPZclgPouO4geU5919AdRAghhBBCCHkVOFWCkMa6LopQJDXG19431tO4Kuz49lpJZ8pr5yMr1cgeS5IEvV4v83yRvkNChjhnfC//PgFAv4BLf1oM8bW38QwGg3RnreXlZSwvL2NpaSkVe0Qkk9Q2EYSkAHOlUkGlUkGtVsN0OkUcx+lubL4t5QGkApONMSQSabGriEvI57YKuY56vR5GoxGccxgOh977XyqVZsQ5K5LYNbZjh1xS8rtcLqdbzksNJz2f0HOV9XeZ5dIJiTxZolAInyAWukdyjGIQIYQQQggh5GXm1AlC9iXMvozbcz4nhpzzCR0hp4S9Puu6rNhC7gvZTl07Knx92c8a6btWq3l3ktJukMePH8+JQlqIsQKGXRcRfLRY4puniC/STrZN14hrRdK17Jg2JvmudyybTCb49re/jdu3b+MLX/gCfvZnfxZXr17FuXPnEMcxqtUqqtVquo19FEWYTCZot9sAgLW1NXz5y1/G5cuXsbS0hB/84Ad4//33Z2oNyTgytq6ZZJ8nKw7JObtjl8zVrqPdal5cPVoMmU6naarewcFBWkhapwTqewEAnU5nRhDS4/pEPJ+wpX/raxuNBtbW1lKBSuKwcwt9zhLJNHptrMMp1Efe30/euMcRfgkhhBBCCCHkZeBU7DKWRV6aCOB/Wfd9Dr1IhlwaRWILxWJjFndQVh8hh4Z1jpTLZa/woun1euh0Omm9nZCoJX2GXEK6cHPIuRFaB4l7PB6nrpLJZJIKZD5RLLT2zjk8ePAAzjlcvXoVcRzjjTfewHQ6RbPZRKlUwmQySXdvAw533ZpOp1hdXcXVq1exsbGR9v/ZZ5+lW9jrdZe4Q0KQLy69XnYuIcFFj2FFKd86i5spK7VM3EM6HivWiNAUume+fuVzrVZDrVYDgLR+kO43JFCG5q8FQDvvvH58522fvr5CZPVPCCGEEEIIIS8jp1IQ0tvQA8X/d98nUoQcEfazxooAxxnfvszLi7+4ZEKiU5Y7CXgiKEVRFEw9k37EwfHgwQOsrq7OtBPXiX6Z19vI6zFlDvq8PaedMRJfFEVzRYflvIhCIhLJlu5W5PC5uYbDIT777DN8+9vfxrvvvourV6/itddew7lz5/Dw4UPs7e2lAoi4ltbW1nDhwgW8+eabAA7rEf3oRz9Cq9WaWQcRrPR85L7Jebt2dt1DaVf2HttnS9ZEb1uv13o0GqHb7aJWq6HdbqdioBZAyuUyRqMRWq3WzBraAtNWvAmJqvZzHMdYXl5Go9HAeDzG3t7ejEvNJwr51sMnlIVikGNF2uT9PeWJStaB9N577wXbEkIIIYQQQsjLwKkUhISsF0TB56YJ9ZH1Umn70y+4Ra4LvRiLGGK3Ks+aly9eebmX/kJzBJ6IPuIqOa44lpd+45uj/SwpVyIaaLFFC2VWNPB9tzuz3bt3DwBw8eJF1Go1bGxsoNvtot/vYzQaYWFhAbVaDQsLC/jiF7+Ier2e9vHZZ5/hk08+AYAZwUf3b1OrrMhnn4vQ+vjujW1nx7DjlMtl9Hq99J5qMUiLU/JZizT2vocEvTwXVJIkiOMYCwsLqdjX7/fn+sgi5JjKEnzs8+K7vijWFZUVJ2sHEUIIIYQQQl4FTrUglOWysIReSEMuIF9/eQ6jEHnXSbFf206LSFn92hhkB7CsOSTJYWHpcrk8s+uX7VdEBrvbln4J1wKJzxUkx61QFcdxKgYBmNuhzPanxRmdQiUxScHk8XiMjz76CJ1OB++88w6+9KUvYWNjY6ZWkTiPpJbQw4cP8cknn+CDDz5I+9POJS1sxHGczsU6gqz4odfKijO+/vXY0+k0XRu7JtI+iiLs7OwgjmNUKpW01pC9n5VKBf1+P91dzhePdQtJHz7Xk4230Wik9Zr29/ext7c39/zqMUPPZxahv9XQ32jWueMKR0WEMUIIIYQQQgh5mTjVglCILGeDzw0RSmcJCROhMY8bH/CkhsuzvqxKH1InR4ScoqkwPueNbmMdJyHHUJ6zSn8Xd4/0YXftyupP5irCkAhCOp47d+6g2WxiZWUFq6urqNfrqXAi7Xu9HnZ3d/H+++/j0aNHM8KVz+HiE8Cs40fHbUUWn7tFz8mO6XO+aCEHQLrdvG98QURHm5YYuuf6WNbzI7FVKpVU4Ov1emk6oP27EaHL9/cZck0Vfb6yrsu6NnR9qA0hhBBCCCGEvAqcWkFI0jZ0LSHB58IJpaKE2of69L1wF3kRDaXOxHGM0WgUFBZCsdu0IWk/GAwwGAzSdjZe6/7Y2trCxsbGjDCknSUyxmAwmBFDRFiQIsJamJH0Nx2vCC1aKHHusPaMbP0eEij0XK37SNxQWlCS9K/pdIrd3V18/etfT2v+6J3BdH/WJaWFKnHqyLV2Djblz+7cpdfGJyj4hJEkSTAcDr2CkvRXLpfR7Xaxu7uL8+fPY39/P+1LnD3SbnFxEbu7u3M1hPRv3/OfJ9jIul+5cgUrKysolUq4d+8e9vf30/nrMewc9H31iV82hpAgY8Xd0Hzy8IlCen1u3rx57D4JIYQQQggh5CQ4ddvOZ3Hc/70v4pzR34/Tv08wCjmQfE6KIn37Xp59IlAW7XZ7TqzRfenf4/EYw+EQw+FwbhczK2T5nEMhV0zomtB89HfnDtOmJI1LC0aSgibulSyHhxVztEhhr/EVYg65eXzz0Md8n/Uxn9NH5ubcYXFw4HCredl2XqPXaDQapQLas4gm+hqJp9FopDF1Op2ZWkZ6PiGB8mmdN741DPWVN9eQaCfHWTuIEEIIIYQQ8ipx6gWhzc3NTHdNlpBjX+h95/R3/aP7t7VxrGAT+tHFkIvGHopVUsTEzRISJnQ/ALC1tYWtra2Z+jw+MUsX3bUOIh2PnBNxRmKz330CmRZl5JyvQLZtKzHLzmE6Ju0K0qKRvfey+5neLU27pfQctCvJurV0n+Ia0v3ZNZOaRtahJe4gvUa29hAAPHr0CAsLC2i1WjP1g/T6xnGMlZWVmTZZ4knoubHinRDHMRqNRurYarVaQVdUlvh5HHHKJ3rav08rNNqd6nz9ST9PK1ARQgghhBBCyMvCqReEAODmzZve1JbQd02eS8N3LuSk8V0TevHW6VWhtJyic/C9wGq3StbL93A4RLPZTNPWQn2FXuJ9u3CF4rax+FwiofYhd0lIuNNYQcqOKfOwRau1+CJikwgx0l6LNlkCo0bECblWp5v5YtFC3GQySeNIkgSdTgdxHM+IQRbnDlPzdCqh7VuTJZzo62V9oihCtVqFcw6j0Qij0WiuQLkdK8tNZWMPHc9bZ58LKu/vwfYv0B1ECCGEEEIIedU4E4KQjyzXkG4DzDsLivZtHQiWkACiz2l3io0jJIj4xrEuCJ97x8YPHDpUWq1Wuk24dbvkzcmKA6HzoZhCAo29LiseADPCjHXb+GK01/v6BDDjGtLz94l9PkHNdy9F2JE6TLqNuGyynl0RpaRelG+nMj1+pVJBtVrFYDBI3WMaG69PSLHrp3+082o4HM45lew4IRHP97xm3XtfXyG0Ky3r7+hp+iaEEEIIIYSQl5EzJwhZkUfwvdxpx0ceeQKJffENvdzKMV0LJkkOC/NWq1VUq1XEcZw6UuwLemiL9yyHi14XPb68IDebTdy/fz/9rscIvdDLj9SlsaKGz3ml1yUkJDk3W3DYrpvvs6yLvic+QdAKbHZ+ekw5Xi6XUS6XU8HO93yNx2Pvs+abg52niFaj0SgVeHQKWej+S0Fw5xy63W56T+36lstl1Ot1NBoN9Hq9GUeafr5CTjaNFbuAQ8Hs/PnzaUziNrPYdQ0Vm/b9nfmeJd2XL86sPkKE2tAdRAghhBBCCHkVOTOCkLy05f2vvs+xYs/7jmV9zxtP/9gXYtmFSteSEVeHFSGEkKtGHwvF4pvnZDJBu92e6zPkhLJOmJAIV2RtdNy++fjmXmSMLKEhr//Qd9+cbZ0fm0InQp0W7qQv61qxhbpDMTh3uOtbp9NBpVJJU8FC19RqNURRNFP7KDR/TeicPi6Ck8yt0+kExTvbhxUCixByL+l+fc4tQgghhBBCCDlrnBlBCJgtMC1kvfz6XhR9L6dWTPAJMllpQvrH50wRMUBcIsPhEIPBAEmSoFqtzqQs2bFke3HrHAphHSfS1qaN2Xno9nZNbaFie52v6HbWmoiQYkUD7WjxCTw+YUr3oc+HRCw7f+t+EkeXLpoMHIpquqi33M/xeJyKffIzHo/TOjv6On1/fWKGFMbu9/t49OgRdnd3g7WBJN44jrG4uIhqtTpzn3zPbJ7Y4mNpaQkXLlyAcw6tVgtbW1szfdv75LsfViy1+OL1PeO2j6eZj4XuIEIIIYQQQsirypkShICweOM7H3KNaPIcJPZ46OU2y43ka6tdQ9Yp5Hs59okaRdxS8jOdTtFqtbxz9h2zL/Z5bgyfuyZ0XVbKj22X5Z4KiUZZjiF9bZbgJ/1osUsLWbofncqli0nb+x+6b9KfCIP7+/s4ODgAgLQOUWjdoihCpVJJd6F73tTrddTrdTjn0O/30/Q1G3sRh1ZIFCr6PGc9N1mE2lAMIoQQQgghhLzKnDlBSF7iQu6fLIpeExJgskQl61YQR4gIPiHBaDQapbWGfM4kADNt9LUhB4juQ4sb9+7dw8HBQWYqj65ho8caj8czLhefA8o6fwRxzYhgYQUU33w0Mge75bwcs4ScJHpNbZxa+CqXy4jjOD0vRZVlbWTMcrmcbiEvu2/Z+kB2DnJO+hIxUO7/ZDLBnTt30Ol0sLCwgGazOXc/dF9SP0ivpR1bX6N/+9bZHl9fX0ej0cBwOMTOzg7a7facyGMFspBop9c/FGeIrGcj1D7090QIIYQQQgghp4EzJwgBh9vQ2y3Cs172Qq6RkOsly4kQElHsddJuOp2mL/tWVNIv9j5hQ6crjUajtDC1CBS++EJzT5IEBwcHuH379oywZGPVayKxl0qlND1Kxx1ydUwmEwyHQ/T7ffR6PQwGg5l10H3o8fU6hMQ7O6akZek6Pnredi2tMKLPyZy1qKPbSPFpaSMFou14PgFC1xOS8yIoicg0mUzw8OFDDAYDLCwspGuZ5TZaXFzE4uLiXC0jn7PLurh8Yo0dY319HZVKBfv7+9je3p6pZ6TjsutsYwiJQ1rE0n1ajuseynKXbW5u0h1ECCGEEEIIeeWJTjqAk+K9997D9evXAYR3NCri6Am5JXxtfcdDL7BSm0a/EPtEC99nH1r0kFoz2mXjSxey8xyPx9jd3cVgMECtVkvTkaz7RrtgtICgY/DNXac36Ri0A0riFYFF2mjnDhBO/dP4xCGf6CbnZI6yVtLe1gayO5pJwWa5RgshvnXS91rcYXo9bLzVahUAcP/+fdy9exdLS0uYTqfY39+fuSc+gWlpaQlxHKPdbs8ITPqeyjzsM+tro+ddKpWwvLwM5xx2dnawu7s7s+W8735odP/22dcx+cS0Iti5EUIIIYQQQshZ4kw6hISbN296XzLty2VIhLEUfam0qTfWtRFyn/goel6PLcKFHlOcPHLM53CRl+9ms5npeLKx2BQve17cKdKfFn98LiwRG3yOqjyseGC/+5xWMge5XmIRV46kw9l45ZzvvF3TUGx2TX3Pppzb2dnBcDiEc4e7jGlHVmhO1WoVpVIJw+FwLoa8a0Mx6muj6FBz7nQ6M8+cbe9zC+l+QjHZv5fjcBzxSKAziBBCCCGEEHJaOLMOISEvVccSSisKpdD4xB1fe3tOBAWdnhXqC5jdqUsLTdZBIYhYIO4bEW3EJWIFI+1cuXXrFpxzWFtbS51MtmaQriUkY0jqmhYKJHVPxhaiKEIURamYIj+hOkAyvnas6Hj0XMRtousP6fXNWjeJX5AUKLtGVuwIuY/ku+xOpr/re2CdYroekbiBdnd3Ua/XcXBwgMFgMDPHUAqUuIlardac+Gafb33M5zrS551zabHq6XSKra0tDIfDmTn6nmcr8NhnWcdi5/I0WGE2NC+AYhAhhBBCCCHkdHGmHULAoUtI8DkUQi+avnZWBCjitAi93IYECd+1WgTJemn2XW9r3mj3jRYDdH/9fh97e3vp2L756fYiSmjBx6Z2+eaphSMfIaeJ/aznYgsn+5w2IWeYXifpy1c3ybf2VsDyOZrsfPQW9jpGaSc1iTqdDra3txHHMeI4xnA4nIvL97w6d5huNplM0O125wSgvGdICym++yf3TuolaReQT9i0ooxdu+NSZA46huP87RBCCCGEEELIq86ZF4SA7Bdm2y6LIqkrIdEhlDLztOkwuk9fyo0VNqQOjhYjtHBhHUf7+/tz8fvG1t/leu34CQk4Gp2elCXaZIlLoVh9fWYJEiKYSdpdnlvF5wqywoedm+5Dp9OFYtzd3cXe3h6q1epc+yzkHo/HY/T7/TkBzSfO+OLQv/U9kHQ0SV+zLjLfeoVEmacRa6yz6Ljo6+gOIoQQQgghhJw2KAhhvpZQ0fSTrHZZLp8sd4h+gRXBwRYpFkeQz/Wi+89yztgYdFFkSVXSu3rpPiaTCVqtFh48eOCdi623Iz+SXlYul2dSpLIEHCmALUWwbVFmHZd1uGh8oo+NVZ+37iHZJc221y4YuwuaFdRsOlKe6KidQT7RCABarRZu376N0WiUCnXSxldgW49dq9VQqVTQarVSx5e00TWafGvnW2s9t0qlgtXVVUynU+zt7aHX63nnGFqHkOvoaZxDRf+ebb/HvY4QQgghhBBCXiUoCB1hHQA+QSHk/CiamiLX6f59YobvGut+sWJOnjhlv4cEE+3cSZIkdZHY1LAkSbCzs4NWq5XpxLDih/StY7bCR2gOWmDxpR9JXyG3VZ4oFnIY2aLXcn3efdfine+cnrP8yDpnuYJKpVJaX2l/fx/T6RTVahW9Xi/oWvKtQRRFmE6n6Pf7aTFqX/Fsu2a+edjzURSh0WhgPB6n9YmsGyvrPhURZXznnsZJlNcv3UGEEEIIIYSQ08iZLyqtCbmEQm4IKyCE2ml8x+1Lsj43mUwQRVGapmRjmk6nGI/Hc7V2fI4U/d2KOFqwkXHFzWO3/pYX9q2tLQwGA/z0T//0TIHmkLAj143H4zlHVBRFXrFMnER6zWz/2oEk4+u10vOw/ej+fPdhNBqlayECioyjXUZa3LCikdw7XQxaizuy3hYp8K3nAQBxHMM5h1u3buHx48dYXFyEcw69Xm/mHsvYWhzR672ysoLJZIKDg4OZe2LvY0hkCd1r5xzOnz+P1157Dfv7+9ja2ppZD9s27+8uS8AsIugVISTIUgwihBBCCCGEnFboEFLol7/jiDwhQm4gnxBjX3rl5VkcL6F0HeCJg8X2lZXyU2QO1sGht4OX851OB51OZ2Zsfa2vH92H1JYRkcWmv2UJCPacFq60SCTbv0vaV6hvvUayriLUiJAjwk5onlbossdCY1lhxFfYWz7Lud3d3TRVrdvtzswpJKTI8SiKUKvVMBqNMBgMMBqN5saxa+SbR2gNFxcXUa/X0W630el05sQgO07obyVrDOsyynMHZTm6fGtFCCGEEEIIIacZOoQMIgpdv369kHtBE2obap91TI6LAyiOY5TL5XTrbnv9eDxGtVoFAAyHw7n+fC4R24dP8NLOkSiKUqeNCDij0Qjvv/8+fuqnfgrnzp3zOkrsHLVTRsQV7ZCp1WozdZN8cVq3kwga2nVjRYPJZDITu0+ME1eOdtIAh9vNy5bzOl7tEtLxaHeO3mVNp4Tp9dF1m6TQs+zKJmPKNu7tdhsPHz5EkiRYXl7G48ePvWsVWn/nHFZWVrCxsYGHDx9ia2srFcq0cymUihjqVxxdlUoF165dw+LiIv70T/8Ue3t7M8JmlhCmj+uxQhQVcqxzSl/rE7roDCKEEEIIIYScdugQCpDnoMk7n/WSmuVgCB2Xl3TrGLHijcRmnSq+l/fjOoXEiVQqlWa2kJ9MJnjw4MFMrSEdR5YAYAUkEcDG43EwTch3TAtoWnzQqWghQUnHMZ1O06LXIuiIi0m7j2x/vvXSP9K/OIxsO31/5bsIMnZdd3Z2sLOzg2q1mhaT1uSJJCIu1Wo19Pt9jEYjrxMt5GzyzVPf6yiKUnGy0+mk8w0V8M5y7hQhyzWm2/jmUuT5J4QQQgghhJDTCB1CATY3N3Hjxg0A4R2qsl4m9Yt96CU0L6VI2ugaQSLEyEu2dnGI2FCtVpEkCQaDwUy/IQeTFUZC7hIRW8StpLePf/jwIRYWFnD58mU0Gg0MBoM5V4aIHL4drPTYkualxRBx3GiHTalUSoshazFMYhJHkPSjdwSzopmMK59FmLJOIHs/bV0i33FdQ8h3rb5eO4okZqmjNJ1Osbu7i08//TR1hOl0LL2uvpj08dXVVVSrVRwcHKT3Stdrkv5selzWswwcusjOnTuHer0OAGg2m3P1pUJutZCQVVS4tHMOXZc1HzqDCCGEEEIIIWcFOoQy2NzcLCwGheqchCgiKtn28pJut4O3/Uj6VGjb+JBbyL4g6znpNuImiaIodeAkSYLHjx/jwYMH6PV63hj12FlOIRlPRB35GY1GGA6HqYNICj7rIs/WbaJdR1bUsc4b61Cy9YysiKSP6T61qOaLR19v11XvZqYFHucc2u02Hj16hNFohCiK0O12vX2H1ltTqVTgnMNwOJwTbDRZqXu+MUulUlrkWt+/0PUhbGrX0xISmI4jMhFCCCGEEELIaaSwQ8g5VwbwPoB7SZL8gnPuCwC+DmAVwHcB/PUkSYbOuSqAfw7gvwKwA+C/SZLkx8898hfEzZs355xCPnwiS0hEst99Yog9NxqN0jQmXdxZO0rku7hRKpUKAKRpRdqlo+O2L8dalPClI0mRZu0UAoB2u41er4fHjx/jy1/+crrtuJ6X/NbFmnU8em18W8iXy+UZcSeU5qT7k7bimNJz0S4YX5qXxOhzCtk1FMFOxybCmb5Gu5CsW0ZEINlZTvrpdDr4/ve/j16vh3q9jnq9PrermG+dfcJhtVpFvV7HdDrFwcFBWpja91xIP3Y7et9z45zD8vIyLl26lApY4uSya2edY/IcWwHrOAJSnnspJI4BdAYRQgghhBBCzh7HcQj9GoAP1fd/BOA3kiR5B8AegF85Ov4rAPaSJHkbwG8ctTsV+F4krQNEjuU5ioqIQb5xtPAgKUUiHthULCkIrQWJLFFLx5s3b+BJipre8l7Ei7t376LVaqUili+VKcup4RMbQvH50IKDvs4KTFZ00v1rB1FWKpIWdqzbRosdWWNpMUhqGIloJEWkxRFUq9XS4uL22Qk5kiSeUqk0U3xcFw33iTGhOds1ke+NRgONRgPT6RTdbte7HlmiT8jRYz/n/S36hMysfgkhhBBCCCHkrFFIEHLOXQXwVQD/9Oi7A/BzAL5x1OS3APylo89fO/qOo/N/wb3iuRl2O3rrbshKodEvwSHxSP8OvbhK+o2ICjqNCjgUL8SxI+NNJpO0PozsUOV7IS8ah32Bl/QqETF0mtCdO3fw3e9+F48ePcJkMkEcxzMFnm1/obW0Y8pa+MQK37VZIkZIJJJ2uk6TPq/Xz7pqbN/OuZnCzxWloGEAACAASURBVPJb0tykjYg1UjMIOBRsPvzwQ3zve9/DrVu3UtdXv99Hq9Wam1dI2NExNRoN1Go1VCoV9Hq9me3ms/qzcw+JZ5cuXcLKykoqCmrHlC8me3995+34vr+jrGc79M+PHKc7iBBCCCGEEHIWKeoQ+scA/i4AeftdA7CfJInkA90FcOXo8xUAdwDg6PzBUfsZnHN/yzn3vnPu/aeM/YUi9YSKuAqy3Ai+l96iiNgi/VihCMBcXRwAaYpVHMczThQbV4isOUsM2iUkx0ejEW7fvo179+7NuJpkvLy18IlnWe6XPGEudA/sZ9/5kMgi/errrIiRNQ7w5J7p1Kp2u42trS1sbW2lTpsoilCv19Hv91MhrohYo8eL4zgtUt7v9+fWxrd2oTnYY+VyGcvLyyiVSuh2u2g2m5nX5/09hYQkn9BV5O+yaDtCCCGEEEIIOQvkCkLOuV8A8DhJku/ow56mSYFzTw4kyT9JkuQrSZJ8pVCkLxkhMSMkcIRSo3zXhdw6Oi1JCxS6sLK4dazoMxqNUkEgL/UnhM8hAyDdzcs3j1arhbt372J7e3tm5zBf36H+Q6JV6OU+JOjYn9C1vsLQWdf65pMlIulx5FpxBpVKJYxGI9y7dw937txJHV5JkqBaraaOI0uWC0eO6QLgpVIJvV5vpn1IULL3ITSuCFalUgmdTmem6HWeW8tH0efSJ75lPRv6M91BhBBCCCGEkLNKkaLSfw7AX3TO3QBQA7CMQ8fQOedcdOQCugrg/lH7uwBeB3DXORcBWAGw+9wjPwFu3rwJAGmRaSC75g8wX5xXjtlrdUqU4HuB1jtRSWqRpCTp9C2dwiUv+IPBIE1LkuNZrhobo00h0jGLyKCLSAOHYlGv18MHH3yASqWCr3zlK6jX66mzyIoQdv4+kShLZMi6XtZLp2vp9Dcdh/4cEhysG0hfp7dw1w4aOS/ijBRzlhg++ugj7O/v4+DgYOb6UqmEWq2GVqs1E5dF33O9RlEUodFooNVqYW1tDYPBALu7u3PPqb239hmwY8n5OI5x7tw5NBoNTCYT3L17N52DvU++tbSpZVn44skTW+39BoD33nuv0HiEEEIIIYQQchrJdQglSfL3kyS5miTJWwB+CcAfJEnyVwH8IYBfPGr2ywC+efT5W0ffcXT+D5JTlqcRchVYMcL30upLfbEvuEWcEVrQ0aKECEZaANHnJ5MJhsPhzC5WNj6fI8QesylG8lJvkf5EGLpz5w52dnbS+OyYVlyxgovPLeQTZHRcds2yCKV5+RxDVkCx/dj7LXWedBFwodvtYmtrCw8ePECz2Zyba6VSwXA4xHA49MYX+q6vl/FqtRoGgwG63e6c2OUTCfXc9Vro9YrjGI1GAwDSQti2vlLIMSUiVtZzn+dOCrnL7Dk5TmcQIYQQQggh5KxTeNt5D78O4OvOuf8RwPcA/ObR8d8E8C+ccx/j0Bn0S88W4suJvFBqt9BxBR6fi6homoy4XMrlcur4kTo+evxKpYLxeJzWHpLzUkdGasqEtlXXYoAVAXQbXThahCcAqfNDrv/000/x4x//GOVyGXEc4+LFi1hZWcHy8jLq9ToAzOzIJf37CklnralOYZNjIdFB2trzkjqm3UM6HrvW0o/8juN4xoE0nU7R7/exs7ODTqeDVquFVquVFue28xDRbGlpCaVSCbu7T4x2vmcmJIrIFvP7+/s4f/48Ll26hPv372Nvb29GtLFromMXl5LeRl7alUolXL58Ge+++y4mkwnu37+fpgiK6Cfx+QQmOx8fvr8n3zGJzyfSOccUMUIIIYQQQggRjiUIJUnyRwD+6OjzLQB/1tOmD+AvP4fYTgWhF2AfWcKRdaRY8Um/EMsxXaxYizTCeDxGpVJBHMdz53RMVhyScXwv3lao8Ald4mQaDAa4f/8+dnd3sby8jKtXr2JhYQFxHHvTz3zOlaw1LCKu5TlsQs4Ue0+l/o8IIDq+yWSCZrOJvb09dLtd7O3tpW4fEX3sOsnays5xg8FgJsZQ2ph9Rpw7rBslBZ6r1SpqtVpaqFr3oecTelZDwszS0lLqPBJ3kIiUoWeiKKF77TufJZQRQgghhBBCCHnCsziECA6dQtevX59zlojzIvRSmveSawm9/IsLQ4szsqtYtVpN20r6mHbaJEmC4XCIUqmEarWK6XQ6V7BYF1gWB9JkMpn5LEKICD1ZYo11+Ygw0mq18ODBA5TLZaytrWFxcRGvv/76jMNE108KpXbZ9bHnteNFfuya2H61+CPzFJeMTssTer0e+v0+tre3UwFoMBjMuXGkP71znJ6ncw4LCwsYDofodDqF5iz9SszLy8vo9XqpgHf+/HlUq1U0m82ZePQ11illP1sBJo5jXL58GeVyGdvb23j48OFc37qf0DxCIldIBA2tQ8gpRXcQIYQQQgghhDyBgtBz4ObNm3OFpotQpJ2vpktWG12cV3b1EqIoSreotw4WKUQtqWhaIBHBSbCuITknYlJI/Mqar4hbk8kEjx8/xt7eHpIkQb1ex7lz59IaONqBA2BOeNBpTpqQq6ZUKqWilhZCpNizFsR885c4RqMROp0OBoMBtre30ev10Gw2Z9Y7Kx1Pz8HGPRwO59rkCW7lchmVSiXdXl7GWlhYgHPzO5VZV5HtT4+r5y5FyhcWFpAkCTqdDtrt9sw8fXGGKOIECsWXdT3FIEIIIYQQQgiZhYLQc8I6hUIv+0VfYu3Lb1Z7Xa/HucMCwiJCSAFp4Emal97tSgQL2Za+UqkgSZI0RUkLBdJWxBZdY0ecST4XR54YZAUcEYZu3bqVni+Xy1hYWECtVkO9Xsfi4iKq1Srq9TpqtdrMWkm6mXUDSVyyBiKEheKWNpPJBKPRCJPJBN1uF5PJBO12G8PhEPv7+5hOp2kbnxhlU8ysW0nX6xERzzmHRqOBTqeT9utLv7P9yz28dOkSzp8/n9YdqtVq6PV6OHfuHIbDIfr9vjf1TT+jul5Q6JmN4xgbGxupc+r+/fs4ODhIhULffc8TwXztrQCXlTJnoRhECCGEEEIIIfNQEHqOZDmFQmkuRWq16OvzUqN0AWDtENIv+cCTYsAiSoioZAsIh5w4wBMH0tPWhNGOHF8/2lUju6OJIyWOY1QqFdRqtbR+jdRDqlQqM84lm/al0+r0Z1kHSbEaj8fo9/sYj8fo9Xrp7yRJMBqN0vY6XS8kVPgICSPiyqpWq+k280XWVMZdXFzEysoKAKSpaiKgST2hUDFpKxLZ9DEtGDnnUK1WsbS0lMZ4cHAwI8I97bNhRSpL6HnRsVIIIoQQQgghhJAwFISeM9oppB0dxxGDfGliIYFBCyZaxBHhRF+vBSARQbSrSBxDSZKkxYAHg8HMMbtbWSgeH1pICDlErIvGnp9Op2ndISmMrF0ssta6xo+ev+1Tizg6tUvHIGPYbdTlet+OY9YJZNdIu2d8ayYiixbdskRC/aytrq4ijmPs7+9jZ2cHpVIJ/X4fFy5cQBRF2NvbS+el+7G/bRqdvU+lUgnr6+u4dOkSnHPY29tLxSbZ+U7fUxu3z3Vk3Ukh0SckEhFCCCGEEEIIKQYFoc8B7RTKE3Ke5iU2K1VGO37kpVpvhy5ihHW1jMfjmVpBVkTJciiFYrMCQlFCAkjWmHb+IkaI8CXnrGATijfLuRXavl6nrPkcOD4BKCSA6Z3F8lw29ryIZvV6HRcvXkSSJDh//jzq9TpGoxG63a43bl9/9pwVNpeWltLi5dodpNc3lG7mGzPLMWbPh0iSBDdv3gyeJ4QQQgghhBBCQehzJSQ+ZIlBWS/B9jqfmKG3a9cpX9odAzzZHUsXPRY3iriGgCeFqOXFXruE8twwebViQo6XLIq28fWvz2kXkBVqfPPyzSdrjKwYrQgUur+6dpAVorKcVu12G71eD7VaDcvLy6kQJulw3W63sAjkE6y0C2tpaQlRFGEwGGBvb6+QaJi1Xj4nka99qD4RAIpBhBBCCCGEEFIACkKfE1K/5MaNG3Mv/qF0sCKChL7OXjOdTlOXjzhURMwBntQN0uKQrSUk7iLnDgs5S00efTyO4xnnjX1B16lrdm4+N4x9ubeijW/+up3tt2h6ns/9EhrPdzzLLeQj5HrRJMnhzmpSuyi0biGnjGwp3+12USqV8Prrr6PZbKJUKuHatWtoNptot9tecTIUm09gc86hXq9jbW0N1WoVH330Ee7du5c75yxRTT8/ofQ7aR9ad9YNIoQQQgghhJBilPKbkGdBv6BqgSPkYsly/1h8ApJN/dJpO7IT1nA4nBGKdF8iAgGHDpV+v58WJo7jOC3oXK1W0y3gtYskJMbkpfqE0sR8jpMsUcWub9Y6h0QiPadQnFn9++6L/PY5kvR1tVoNCwsLGAwG3nSr0BqIUFer1dDv99M2pVIJg8EAS0tLWFlZwe7ublo0W9ro2kl2DF2jSSiXy6jVatjY2Ei3m//ss8/S7eZ9a6b7tGMVdWbpH3s/Nzc3KQYRQgghhBBCyDGgQ+gFIC+qdlt6S5YYpF+KdR+6H9lGXTt+dB+CCEcA5sQjnSomxadlW3q7g5c4h2RLdp9Y4hOJ8lwvImToeH1iii+Nyc4zax2t0KGv0/Hoz74iy3kpTz5RJ5T2JAKLLcgcWhtNFEWIomim7lO5XMZkMsH6+nq6w5jce7mPobFCjqt6vY7Lly/jC1/4ApIkwYMHD7C/v5+5lvp6/Wz6+g9d54MiECGEEEIIIYQ8HXQIvUCyaptkpSpZMSjkerFtstJzBBF9pJaQbSu1hoBDwUkEIh2bOIV8Y4TSrvLi9q1JaN5FCdWj8cWSF/NxyXIj6dhKpdLMTm7H6V/EHysYOuewsLAA5xyGw2E6R58QFFojfXxpaQnr6+tYWlpCr9fD7u5ucBt7ix5T/7bjEUIIIYQQQgj5fKEg9IKxKWTyO+T6CIkj+oXaChj6pTvkztHt5ZrJZILhcJg6SIBDB5EIPkmSYDQazeziJW3EORSqS2OdJ1ZkyEoH8q2HnZN1J4XWTtpqV5RvjFDal0+s0/FbfNdpV5btyzmHXq8XFNWyjpXL5bl7LzWiGo0GWq0WhsOh93r7bGQ9dz/xEz+BN954A5VKBR9//DE+/fTTOYeY7znzjePbmj60pjpmpogRQgghhBBCyLNBQegEsC+yoTSZUK2ZEKH2x0nD0TVlpI3UFdJCiohDugB1FEUzrpTQGMchT6Qoeu2zti8qVuXNV9YwJPKMRqOguGTjtG3EXWTXX+5dp9PJ7LeIM6tUKmF1dRXlchnj8Rg7Ozvo9XqZIpDuO2tMn0jlE/zoICKEEEIIIYSQZ4c1hE6Izc3NdAcy7erRhNwoGl3Txu4eJud8L+q6fo2vPo9seV6pVFIBI4oOHxedOibXViqVtO4QgFQskvSnos6fLOeTj5DIZV0mvvlmxeUTJ3xr5XMm+foVMUW31deIC0sKPvscNSEBT9cBGg6HaSFouXZ5eRmTyQT7+/tzrrRQGqLP/RTHMZaWltBoNNDr9XDv3j1sbW2l91g/b9Jf1vbwIWwsuk9uKU8IIYQQQgghzwc6hE6Qzc3NoIsiK10mzz2iBYWsdiGBRosG4/E4rTGkz5dKpbSIsX5hF+FDBCLthjmOW8gnqGT1YcWjIiKET3B7WpeWL7VKCxt2LvpHRBMt4oXGsOfsXPV3qf+0sLCAfr+PTqczV8MnNAeNxBnHMc6dO4fpdIp2u43t7e253ers/OS7T9zzzcW3XrpfQgghhBBCCCHPBzqEThi9A1kI69bIOy+ikBUh8uri6Do2docySQvTfWpX02QyScUfSR+TreknkwkGg0HunIqkW/na6rSioteHXEJWzMnrz/fZ1843tnMO4/E4/VypVNDpdIJiiE1bs+shO77JsVqtBuccVldXsbW1hWazGRSCbP9W4CuXy3j99ddx7do1PHz4EJ9++ikePHgwJzCFnFK+tQq5u3z3kvWCCCGEEEIIIeT5QofQS4JOhSkqkIQEHv0y73Nb6HNF3DTSRlLFtGtIilGPx2OMRqP0RxxF4iKybiFLnghTJN3MtyZ51+pr9FyPQxEnTF6/UrzbxlNkXOBJiqAVncrlMur1OrrdLkajUe49CB0vlUq4cOEClpeXsb29jb29vTStMI8s55HvvF0vikGEEEIIIYQQ8vyhQ+glQl58b9y4kR6zqUQhp4XPuSMv8iLOWOePzyWTV69G/5b2UrNG+ta1hWq1WppCNp1O023PNcdJBcpyuOg2eXPxiTg+4SLkKtLfQ4KH3snMthmPxzPnq9Xq3Liyrjo264aSdDMp8izPiYhLIgjt7+9jPB4XToHTz1qlUsH6+jrefPNNlEolfPrpp+h2uzP9PQ2++6PnxnpBhBBCCCGEEPL5QYfQS4h1RPhcJ/I5lE5mU3m0kKCPyXUhR00Rt5LPgSROoW63O7cTWRzHM+ln9nfeuD63TWgt7PU2tc6e0+tTRFCybWzqmRbKBBHMdLxxHGM8Hs+NkeUs0ueiKErr+UiK12AwwHA4xGg0Qrvd9har1j/6uPRfKpVw8eJFvPvuu3DOYWtrC+12e0YMEvGqSKpdlgAna7a5uUkxiBBCCCGEEEI+ZygIvaRsbm5ic3PTWx8nzyUTal/UiZMlQmSJI75j0+k0TS+zNWlsTaIi7h977nlQRMjQbUOxFLlPoWucc0HHTZZbS7fRLjIt+PT7/RmXmL4mK8ZSqYRSqYS1tTWcO3cO/X4fOzs7x0qrC9UUCrVlehghhBBCCCGEvBgoCL3k2BdkmwqlBSNbFFgcISLAyDWarBd0PYZuD8xvca/b6vYi+gyHQ/T7fQBP3Et6N7KitWieJd2sCCFHkY5Bnzuu2GbXSJxTUoNJj3GcPqMo8u4gtr6+jq2trTlXjp2HHNPPQxRFWFtbwxe/+EXUajV8/PHH+Pjjj+ccTvI7a4c0H1LPSIQgikGEEEIIIYQQ8uKgIPQKIOkzx3HtAIfCi3WNiEMnjyzhSPqRFDDt9rHj6ELJWgiSgtQA0n7EkZI1J33uWVw9Wa6bUIpe0f7zrrEinKR7heLSx30Cnq4zJAKgrH+tVkO3252bk88VZfutVqs4f/48KpUKRqMRtre30ev1MuPzHc9L/3veoh4hhBBCCCGEkHxYVPoVwW5PHxI57Iu/FCaWH+dcKj5Y8pwpcj6O47n6NlYUkuOSMiZtR6NRKiLpcwBSUUjSy3ScPpEmy81j22QJD756QFkFprMEjOPUHQIO51ypVNBut+fGAZ4UjbZj6Dblcjltp4W4paUlLCwsYH9/3xu39KHH0Ne/9dZbePvttzGZTHDv3j1sbW1hOBwG566dT9aBFKpRRFcQIYQQQgghhJwMdAi9Yty8edMrWGS5UrQwJC/8NnXJJ2TkFV2WXa2kRpD8yLb0sv28Fo6kvXYqifgjPzbGUJqWb54hirh4nrUmUZ7TxXdehLqswtH6+lAfGnFcLS0tIUkSDIfDzNpGNgVR3F8bGxtoNBpotVp4/PjxXCFpXx++tL5QEWmKQYQQQgghhBByctAh9Ari255e8LlL9Fbk+qVf13zxFaXOElF0HRktAsg1IkD5atVMp1OMRqNUuJAYtVgEzO6cpWMIpR6FUuGKpEvl9W+dLaGUOt+62Ws1tVotWJA65KqxVCqVtA9Z00ajgY2NDezu7qZpXkXT36rVKi5evIjLly8DAH74wx/i4cOHM/ciax19c5HjdAURQgghhBBCyMsBBaFXmCLCkP45jmBSlJA4Yvv1XSduIO0WsgWLJbUs1F9IvBKyxB9fzPbaIsfscZ+zKYSen8Rx3Jo6WngTx1G1WkW1WsX29vbMdvMyhg95RpaWlrC+vo5SqYR2u41ms4nBYDBXNFqvvd52PlQ3iFvJE0IIIYQQQsjLAwWhU4AVhqzTR17WdYFpScmyO1PZFDItNuiCxZYsIcDXVtDb0UttIjkm8cZxjOl0mgpIWkDJE0+yXD0aPf+QYBJyBQk6LU9/12uj45GC25LSlec88glt0o/0US6XUalUsLq6isXFRXz44Ydzc5C47PdSqYQ4jvH222/jjTfeQLPZxK1bt3BwcIDxeDwzdqjukm8edAQRQgghhBBCyMsHawidUnyuGRFVtHARctfkFUUOpYMJofpDth/9o8WpUqmUxgs8cb5kkefysbVynoZQsWQ7lj5vazXZdqEi3z6sOCR9TSaTGVGnXq8jiiJ0u91CbiURliqVCtbX1xHHMXZ3d7G9vR0s7m3nrO+vfKYYRAghhBBCCCEvJ3QInSLsy7d1DGlnjRZYrCBhXR/WLTQejxFFEeI4xmg0mrvG9hPCtpW+KpUKAKQ7Wo3H41SskPGl1pCvzk7WXOy5rHpE+lrtrtLXhfoNzVG+S1pXv9+fKdas20nfNlVL91upVNDv9zEYDFCr1RDHMa5cuYL19XVsb2+j2WxiOp3OpKZp15J8B4BLly7hnXfewfLyMra2tvCd73wnjc/ORX+39/69996bW1NCCCGEEEIIIS8XdAidYrRAFEr1kWN5dW+0OKTTt3wpZNYJVBQZ21dIWoQgSYvyuVGehay6R0WFrZBAokU1QbZ2z6qPpPvwxSTpdCKkibCzuLgIAOh0OjNjh1L6ZA0vXLiA8+fPo9/vY2dnB8PhcEZEtHPxCWEUgwghhBBCCCHk1YAOoVOOFoXEMaTTs0IuoTzk+jiOASAVJXw1ZIrUE9LXSFpbHMcz29bLNvdRFM0IVKF0tVDamxXDdEHkvHizUtC0qJbnIpLUrl6vF6y9E4pdfotrqtPppI6vWq2GxcVFNJtNPHz4MB1XO5zseos49fbbb6NWq+GDDz7A7du3U+eSjl1S+fTcmRZGCCGEEEIIIa8edAidITY3N+ecPoA/FcjnJLLFkbUwINudi7igaxTZvosgaVvWgSTjikvIF2dev1kUEa+yzoWEMCtCSfyhukehWkX6s+5D1qlSqcA5h263i263O3PPfPOUNaxWq6jVahiPx9jZ2UG/38+cL9PDCCGEEEIIIeTVhg6hM4a8wH/1q19Nj/lEDOtwEayjZzKZpI4du/35ZDKZ2UXMR54rplwuz6SQJcnhdvXiTJL6NqF55JFVR0hjnT8+oazoOCII+dxRx4k5iqK0P/lcq9UwmUzQ6XTSGkzSxjdWqVRCFEVpmlm73Uar1cJoNMpMY6MriBBCCCGEEEJebSgInVFEGPra176WOnpCIkgI7TLSKUkiDsm26pKO5kujyhJCRqMRyuVymjomjiSd7pYVo+3fFms+LjZlSx+Tz75juk6QiDeVSmVGzApdb7/r33EcpwW3y+UyFhYWsLq6imaziUePHqUOodB6AEC1WsXa2hquXbuGx48f48c//jHa7fZc7SC5nkIQIYQQQgghhJwOmDJ2xvnmN785IxT46t/kIdfoVDL5EbEij5BLR287r2Py7bxVhOMKQU8jHIUQMUoEON8c8op6636ccxiPx2lqXbVaxcLCAjqdDvr9vlfUEXTNobW1NZw/fx47OzvY3d2dSQXUNYRu3rz5PJaBEEIIIYQQQshLAB1CBL/7u7+bfpbC08Ch60eKTvucMfLbt6uWCDmVSgWVSgWDwWBOaND9Cj5xajQaIYoiRFGUOoVETCmXy0GnjcWXNqVT3HxFqoukk/mQdfONF8cxptPp3HbzvvZ6POtwiqII/X4fcRyjVCrhwoULWFlZwUcffYRutxucv8QXRRGuXbuGt956CwsLC/jjP/5jtFqtmdpSrBFECCGEEEIIIacTCkJkBkkJ0sJQngCSdX4ymaRuFLtjFRAuvKxFEdnmPo7j1G2U5WQ6bk2ep6HormQ+4UvcQTaVrEjBay0KiWtKxKdGowHnHHq93pxIpmORviRdTIpJS4rZdDplahghhBBCCCGEnHIoCBEvWhiyu4v5dg4LnROnkBSelu3p5TpLSBgRR42koOmdtnSfgq/Gz9NStA9f+pcVjUTEGQ6Hc7u82XWU1DhffScAqagUxzHiOMby8nJaFNqKTXo3MuBQPLpy5QouXLiA0WiE+/fv4xvf+MZTrhAhhBBCCCGEkFcN1hAimWxubs7Ujimym5Z1Aum0LN1PaMv1rL5lxzFxHfnahL6LMBIaL2snNB/H3RkMeJKaZuv7FHEHZSHbzbfb7Zld2eycxVHUaDSwvr6eXvPrv/7rhedCCCGEEEIIIeTVhw4hUghfCpFOK8sTcmTHMO2CsW18fVmHS5IkqVPIJ/DI9Vnbyeudv2wqW5H0L9tX6Lvtr1wuo1arod/vp84p2ya0o5iNQZxRURSh0WjgzTffxN7eHj755JMZd5Ad4/d+7/eC8yOEEEIIIYQQcnagIESeGhGJrl+/nrudvLh7yuVyWhNI4xOKsoQdK+iEtmzP60vGDO365evnuOixy+UyhsPh3Lki+NZG7y62t7eHwWDgFYJ+//d//9hxE0IIIYQQQgg5vVAQIs+M3Y78+vXrczV89O8oevLYWXeOpIHpwslSM8gWYgbgFXGydgbzFaL2Faa2u5ABmKnlY+cUGlf3V61W4ZxLxTDbhwg8to6QjyiK0rlvbGxgaWkJP/zhD9HpdNI2LAxNCCGEEEIIISQEBSHy3BGB6Pr16+kxLeBI/R8tjgB+QciKRUV25HqWXcZ8Ik2oiLZ8LooIW3nxhebnE6dKpRIWFxcBIE1F41bxhBBCCCGEEELyoCBEPje0c0h2K5tOp6mQoQUgYLYAsk8s0q6hyWTyVFvNF60R5HMPZYlReXV/JF1MtoP3CUO2b33eFomWdSuVSqjX6/jVX/3V3LkTQgghhBBCCCECBSHyQrDpSzdu3EC5XE6/a+FDfqzoI+ljAGZ26fKlgz0rVvzxpZDZtK6QaylJEkRRlBbXtmQ5kHxtoijC7/zO76Tff/u3f/s4UyOEEEIIIYQQQigIkZNBC0Q3btxIHUOhLeuBJ/WC9G5llqcVh/JS0fJS1PJ2INPbzdt5FiksWK1RxAAABcBJREFU7ZxjYWhCCCGEEEIIIc8NCkLkxAkVP9bb2gOzrqAoimbSrzR5qVdFt6ovItL4PvvGjuMYw+EQk8lkznWk46XoQwghhBBCCCHkRUBBiLy0WKFIilTrGkSAP53LkrdrlyXLgXScfqQvu9287vtb3/pW4b4IIYQQQgghhJDnAQUh8spgt7f3cePGDW8haB++os0WKwzluYvExSTtuOMXIYQQQgghhJCXEZeXFvNCgnCuBeBHJx3HS8Y6gO2TDuIlgusxD9dkltO6Hm8mSXLhpIMghBBCCCGEnC5eFofQj5Ik+cpJB/Ey4Zx7n2vyBK7HPFyTWbgehBBCCCGEEFKc0kkHQAghhBBCCCGEEEJeLBSECCGEEEIIIYQQQs4YL4sg9E9OOoCXEK7JLFyPebgms3A9CCGEEEIIIaQgL0VRaUIIIYQQQgghhBDy4nhZHEKEEEIIIYQQQggh5AVx4oKQc+7nnXM/cs597Jz7eycdz4vAOffPnHOPnXPfV8dWnXP/yjn3n45+nz867pxz/8vR+nzgnPsvTy7yzwfn3OvOuT90zn3onPuBc+7Xjo6f5TWpOef+1Dn3H47W5B8cHf+Cc+7fHq3J7zjnKkfHq0ffPz46/9ZJxv954ZwrO+e+55z7/aPvZ3o9CCGEEEIIIeRpOVFByDlXBvC/AbgO4EsA/opz7ksnGdML4v8A8PPm2N8D8K+TJHkHwL8++g4crs07Rz9/C8D//oJifJGMAfztJEneBfAzAP7bo+fgLK/JAMDPJUnynwP4MwB+3jn3MwD+EYDfOFqTPQC/ctT+VwDsJUnyNoDfOGp3Gvk1AB+q72d9PQghhBBCCCHkqThph9CfBfBxkiS3kiQZAvg6gK+dcEyfO0mS/DGAXXP4awB+6+jzbwH4S+r4P08O+f8AnHPOXXoxkb4YkiR5kCTJd48+t3D4wn8FZ3tNkiRJ2kdf46OfBMDPAfjG0XG7JrJW3wDwF5xz7gWF+0Jwzl0F8FUA//Tou8MZXg9CCCGEEEIIeRZOWhC6AuCO+n736NhZ5LUkSR4AhwIJgI2j42dqjY5Se/4LAP8WZ3xNjtKj/j2AxwD+FYBPAOwnSTI+aqLnna7J0fkDAGsvNuLPnX8M4O8CmB59X8PZXg9CCCGEEEIIeWpOWhDy/Y89tz2b5cyskXNuEcD/DeC/S5KkmdXUc+zUrUmSJJMkSf4MgKs4dNO962t29PtUr4lz7hcAPE6S5Dv6sKfpmVgPQgghhBBCCHlWTloQugvgdfX9KoD7JxTLSfNI0p6Ofj8+On4m1sg5F+NQDPo/kyT5f44On+k1EZIk2QfwRzisr3TOORcdndLzTtfk6PwK5tMSX2X+HIC/6Jz7MQ5TS38Oh46hs7oehBBCCCGEEPJMnLQg9O8AvHO0U1AFwC8B+NYJx3RSfAvALx99/mUA31TH/8bRzlo/A+BA0qhOC0e1XX4TwIdJkvzP6tRZXpMLzrlzR5/rAP5rHNZW+kMAv3jUzK6JrNUvAviDJElOjSMmSZK/nyTJ1SRJ3sLhvxN/kCTJX8UZXQ9CCCGEEEIIeVbcSb8jOedu4PB/+ssA/lmSJP/TiQb0AnDO/V8A/jyAdQCPAPwPAP5fAP8SwBsAbgP4y0mS7B6JJf8rDncl6wL4m0mSvH8ScX9eOOd+FsC/AfAf8aQ+zH+PwzpCZ3VN/jMcFkUu41C4/ZdJkvxD59wXceiQWQXwPQB/LUmSgXOuBuBf4LD+0i6AX0qS5NbJRP/54pz78wD+TpIkv8D1IIQQQgghhJCn48QFIUIIIYQQQgghhBDyYjnplDFCCCGEEEIIIYQQ8oKhIEQIIYQQQgghhBByxqAgRAghhBBCCCGEEHLGoCBECCGEEEIIIYQQcsagIEQIIYQQQgghhBByxqAgRAghhBBCCCGEEHLGoCBECCGEEEIIIYQQcsagIEQIIYQQQgghhBByxvj/AVtJnzEkOBaBAAAAAElFTkSuQmCC\n",
      "text/plain": [
       "<Figure size 1440x720 with 3 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "image2, mask2 = test_set[4]\n",
    "pred_mask2, score2 = predict_image_mask_miou(model, image2, mask2)\n",
    "\n",
    "fig, (ax1, ax2, ax3) = plt.subplots(1,3, figsize=(20,10))\n",
    "ax1.imshow(image2)\n",
    "ax1.set_title('Picture');\n",
    "\n",
    "ax2.imshow(mask2)\n",
    "ax2.set_title('Metal Artifact truth')\n",
    "ax2.set_axis_off()\n",
    "\n",
    "ax3.imshow(pred_mask2)\n",
    "ax3.set_title('UNet-MobileNet | mIoU {:.3f}'.format(score2))\n",
    "ax3.set_axis_off()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 57,
   "id": "c803240d",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1440x720 with 3 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "image3, mask3 = test_set[6]\n",
    "pred_mask3, score3 = predict_image_mask_miou(model, image3, mask3)\n",
    "\n",
    "fig, (ax1, ax2, ax3) = plt.subplots(1,3, figsize=(20,10))\n",
    "ax1.imshow(image3)\n",
    "ax1.set_title('Picture');\n",
    "\n",
    "ax2.imshow(mask3)\n",
    "ax2.set_title('Metal Artifact truth')\n",
    "ax2.set_axis_off()\n",
    "\n",
    "ax3.imshow(pred_mask3)\n",
    "ax3.set_title('UNet-MobileNet | mIoU {:.3f}'.format(score3))\n",
    "ax3.set_axis_off()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 58,
   "id": "28c8cdcf",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Test Set mIoU 0.9029659132358209\n"
     ]
    }
   ],
   "source": [
    "print('Test Set mIoU', np.mean(mob_miou))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 59,
   "id": "d269caff",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Test Set Pixel Accuracy 0.9976231938501766\n"
     ]
    }
   ],
   "source": [
    "print('Test Set Pixel Accuracy', np.mean(mob_acc))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "b4583378",
   "metadata": {},
   "outputs": [],
   "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": "3.7.9"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}