2.1) Player Detection using Particle Filter.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Getting Colors of Player Jerseys"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "import numpy as np\n",
    "import cv2\n",
    "import os\n",
    "from random import randint \n",
    "import pandas as pd\n",
    "import matplotlib.pyplot as plt\n",
    "import matplotlib.patches as patches\n",
    "import matplotlib.image as mpimg\n",
    "import extcolors\n",
    "from colormap import rgb2hex\n",
    "from colormap import hex2rgb\n",
    "from PIL import Image\n",
    "from matplotlib.offsetbox import OffsetImage, AnnotationBbox\n",
    "from tensorflow.keras.models import load_model "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Function to get a random hit frame from the match to determine jersey colors"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {},
   "outputs": [],
   "source": [
    "def getHitFrame(file_video,file_timestamps): \n",
    "    video = cv2.VideoCapture(file_video) \n",
    "    fps = video.get(cv2.CAP_PROP_FPS)\n",
    "    print('frames per second =',fps) \n",
    "    with open(file_timestamps,'r') as f: \n",
    "        timestamps = f.read().split(',')  \n",
    "        n = randint(0, len(timestamps)) \n",
    "        frame_id = int(fps*float(timestamps[n]))\n",
    "        video.set(cv2.CAP_PROP_POS_FRAMES, frame_id) \n",
    "        success, frame = video.read() \n",
    "        if success:     \n",
    "            framepath = os.path.join('data','masked','base.jpg')\n",
    "            cv2.imwrite(framepath, frame)\n",
    "            print(f'\\t Saved frame')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "frames per second = 30.0\n",
      "\t Saved frame\n"
     ]
    }
   ],
   "source": [
    "file_video = os.path.join('data','match1.mp4') \n",
    "file_timestamps = os.path.join('timestamps.txt') \n",
    "\n",
    "getHitFrame(file_video, file_timestamps)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Function to Convert RGB into Hex and Create and DataFrame "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Convert colors into hex and create a dataframe\n",
    "\n",
    "def color_to_df(input2):\n",
    "    colors_pre_list = str(input2).replace('([(','').split(', (')[0:-1]\n",
    "    df_rgb = [i.split('), ')[0] + ')' for i in colors_pre_list]\n",
    "    df_percent = [i.split('), ')[1].replace(')','') for i in colors_pre_list]\n",
    "    \n",
    "    #convert RGB to HEX code\n",
    "    df_color_up = [rgb2hex(int(i.split(\", \")[0].replace(\"(\",\"\")),\n",
    "                          int(i.split(\", \")[1]),\n",
    "                          int(i.split(\", \")[2].replace(\")\",\"\"))) for i in df_rgb]\n",
    "    \n",
    "    df = pd.DataFrame(zip(df_color_up, df_percent), columns = ['hex_code','occurence'])\n",
    "    return df"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "1. **Tolerance** : 0 means no grouping and 100 means all colors are grouped\n",
    "2. **Limit** : upper limit to the number of extracted colors presented in the output."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Function to Obtain Court Area using Masking"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [],
   "source": [
    "def masking_image(input_image, BL, BR, TR, TL, saved_name):\n",
    "\n",
    "    # mask defaulting to black for 3-channel and transparent for 4-channel\n",
    "    # (of course replace corners with yours)\n",
    "    mask = np.zeros(input_image.shape, dtype=np.uint8)\n",
    "    roi_corners = np.array([[BL, BR, TR, TL]], dtype=np.int32)\n",
    "\n",
    "    # fill the ROI so it doesn't get wiped out when the mask is applied\n",
    "    channel_count = input_image.shape[2]  # i.e. 3 or 4 depending on your image\n",
    "    ignore_mask_color = (255,)*channel_count\n",
    "    cv2.fillPoly(mask, roi_corners, ignore_mask_color)\n",
    "\n",
    "    # apply the mask\n",
    "    masked_image = cv2.bitwise_and(input_image, mask)\n",
    "\n",
    "    # save the result\n",
    "    cv2.imwrite(saved_name, masked_image)\n",
    "\n",
    "    # Show the resulting image\n",
    "    cv2.imshow(\"Masked Image\", masked_image)\n",
    "    \n",
    "    cv2.waitKey(0)\n",
    "    cv2.destroyAllWindows()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Function to Extract colors of player clothes and Plot a Donut graph"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [],
   "source": [
    "def exact_least_used_colors(input_image, tolerance, limit, zoom, num_rows): \n",
    "    \n",
    "    # input_image is the masked image\n",
    "    img = Image.open(input_image)\n",
    "    \n",
    "    # Extract colors - RGB, counts and create dataframe\n",
    "    img_url = input_image\n",
    "    colors_x = extcolors.extract_from_path(img_url, tolerance = tolerance, limit = limit)\n",
    "    df_color = color_to_df(colors_x)\n",
    "    \n",
    "    # We don't want black color of the background and green color of the court, so filter out first 2 rows\n",
    "    list_color = list(df_color.iloc[2:,0])\n",
    "    list_occurence = [int(i) for i in list(df_color.iloc[2:,1])]\n",
    "    \n",
    "    # annotate text\n",
    "    text_c = [c + ' ' + str(round(p*100/sum(list_occurence),1)) +'%' for c, p in zip(list_color, list_occurence)]\n",
    "    fig, (ax1, ax2) = plt.subplots(1, 2, figsize=(160,120), dpi = 10)\n",
    "    \n",
    "    # donut plot with percentages\n",
    "    wedges, text = ax1.pie(list_occurence,\n",
    "                           labels= text_c,\n",
    "                           labeldistance= 1.05,\n",
    "                           colors = list_color,\n",
    "                           textprops={'fontsize': 150, 'color':'black'})\n",
    "    plt.setp(wedges, width=0.3)\n",
    "\n",
    "    # add image in the center of donut plot\n",
    "    img = mpimg.imread(input_image)\n",
    "    imagebox = OffsetImage(img, zoom=zoom)\n",
    "    ab = AnnotationBbox(imagebox, (0, 0))\n",
    "    ax1.add_artist(ab)\n",
    "    \n",
    "    # Convert number of occurences into percentages\n",
    "    list_percentage = []\n",
    "    for p in list_occurence:\n",
    "        p = round(p*100/sum(list_occurence),1)\n",
    "        list_percentage.append(p)\n",
    "        \n",
    "    # second most occuring color: color of the player's clothes: df_player_color\n",
    "    # df contains all colors except black bg and the green color of the court\n",
    "    df = pd.DataFrame(zip(list_color, list_percentage), columns = ['Hex_Code','Percentage'])\n",
    "    df_player_color = df.iloc[1,:]\n",
    "    \n",
    "    # Print RGB value of least used colors\n",
    "    RGB_Value = [hex2rgb(df_player_color['Hex_Code'])]\n",
    "    print(\"RGB Values: \", RGB_Value)\n",
    "        \n",
    "    # Display color palette\n",
    "    RGB_Value = np.array(RGB_Value)[np.newaxis, :, :]\n",
    "    plt.imshow(RGB_Value)\n",
    "    plt.axis('off')\n",
    "    plt.show()\n",
    "    \n",
    "    return RGB_Value[0][0]"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Implementing to get player colors"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Function to draw pure black lines on court left and right court boundaries (to improve particle filter) \n",
    "\n",
    "def change_court_boundary(frame): \n",
    "    \n",
    "    # Boundary Points\n",
    "    BL = (195,654)\n",
    "    BR = (1068,654)\n",
    "    TR = (848,275)\n",
    "    TL = (398,275) \n",
    "\n",
    "    # Left Boundary Lines\n",
    "    cv2.line(frame, BL,TL, (0,0,0), 7)  # Outer \n",
    "    cv2.line(frame, (BL[0] + 70, BL[1]),(TL[0] + 35, TL[1]), (0,0,0), 7)  # Inner   \n",
    "\n",
    "    # Right Boundary Lines \n",
    "    cv2.line(frame, BR,TR, (0,0,0), 7)  # Inner \n",
    "    cv2.line(frame, (BR[0] - 70, BR[1]),(TR[0] - 35, TR[1]), (0,0,0), 7)  # Outer    \n",
    "    \n",
    "    # Center Line \n",
    "    cv2.line(frame, (BL[0] + int((BR[0] - BL[0])/2), BR[1]),(TR[0] - int((TR[0] - TL[0])/2), TR[1]), (0,0,0), 7) \n",
    "\n",
    "    return frame "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [],
   "source": [
    "# original image\n",
    "# -1 loads as-is so if it will be 3 or 4 channel as the original\n",
    "IMAGE_PATH = os.path.join('data','masked', 'base.jpg')\n",
    "image = cv2.imread(IMAGE_PATH, -1) "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(720, 1280, 3)"
      ]
     },
     "execution_count": 7,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "image.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [],
   "source": [
    "MASKED_IMAGES = os.path.join('data','masked') "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "RGB Values:  [(96, 0, 27)]\n"
     ]
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAABNkAAAI1CAYAAADrfiwkAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjYuMSwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy/av/WaAAAACXBIWXMAAAGKAAABigEzlzBYAADWSUlEQVR4nOzdeZxkVX3//9c551b1PvsMAzNsw74IYiIqKiJqEPe4iyiuUeOuqLgjuCDuv+Rr3DWaGIwhxARi4vL9CsYoLqiIsjMwA8y+dU8vVfeec35/3FvVVdU9+3RXd9f76ePaPbXcOnW7qov77s85HxNjjIiIiIiItNHLzcp2D0FERETkgNh2D0BERERERERERGS2U8gmIiIiIiIiIiJygBSyiYiIiIiIiIiIHCCFbCIiIiIiIiIiIgdIIZuIiIiIiIiIiMgBUsgmIiIiIiIiIiJygBSyiYiIiIiIiIiIHCCFbCIiIiIiIiIiIgdIIZuIiIiIiIiIiMgBUsgmIiIiIiIiIiJygBSyiYiIiIiIiIiIHCCFbCIiIiIiIiIiIgdIIZuIiIiIiIiIiMgBUsgmIiIiIiIiIiJygBSyiYiIiIiIiIiIHCCFbCIiIiIiIiIiIgdIIZuIiIiIiIiIiMgBUsgmIiIiIiIiIiJygBSyiYiIiIiIiIiIHCCFbCIiIiIiIiIiIgdIIZuIiIiIiIiIiMgBUsgmIiIiIiIiIiJygBSyiYiIiIiIiIiIHCCFbCIiIiIiIiIiIgdIIZuIiIiIiIiIiMgBUsgmIiIiIiIiIiJygBSyiYiIiIiIiIiIHCCFbCIiIiIiIiIiIgdIIZuIiIiIiIiIiMgBUsgmIiIiIiIiIiJygBSyiYiIiIiIiIiIHCCFbCIiIiIiIiIiIgdIIZuIiIiIiIiIiMgBUsgmIiIiIiIiIiJygBSyiYiIiIiIiIiIHCCFbCIiIiIiIiIiIgdIIZuIiIiIiIiIiMgBUsgmIiIiIiIiIiJygBSyiYiIiIiIiIiIHCCFbCIiIiIiIiIiIgdIIZuIiIiIiIiIiMgBUsgmIiIiIiIiIiJygBSyiYiIiIiIiIiIHCCFbCIiIiIiIiIiIgdIIZuIiIiIiIiIiMgBUsgmIiIiIiIiIiJygBSyiYiIiIiIiIiIHCCFbCIiIiIiIiIiIgdIIZuIiIiIiIiIiMgBmnUh2yte8QqccwwODgIQQmD+/Pk85znP2eN9f/GLX/CqV72K448/noGBAbq7uznqqKN4/vOfzzXXXEMIoen2v/nNb7jiiit49rOfzcqVKzHGYIzZ7WOcc8459dvtapvO2+/ptsYYzj333D0eu1Y7d+7kQx/6EKeddhr9/f3Mnz+fU089lde//vXs3Llz0vv87Gc/4ylPeQqLFi2iv7+fM888k29+85v7/NgAO3bs4D3veQ+nnHIKvb29dHd3c8IJJ/DWt76VjRs3Trj92NgYr3/961myZAl9fX084xnP4L777tvlvpcvX86LXvSi/RqbiIiIiIiIiHSepN0D2Fe/+tWvOOmkk5g3bx4At956K4ODg5x55pm7vE+aprzuda/jq1/9KgAnnHACT3ziEymXy6xevZqrr76a7373u5x77rn8+Mc/rt/v8ssv53vf+95+jfO8885j+fLlbb/9RRddtMvrrrvuOjZv3sxjH/vYvX5cgNWrV/OEJzyB1atXs2rVKs4//3wqlQq33347n//853n3u99Nf39/032uvvpqXvCCFxBC4Oyzz2bJkiX8+Mc/5qKLLuLmm2/mk5/85F4//ubNmznrrLO48847Wb58OU960pMA+OUvf8lnP/tZvvOd7/Dzn/+cI488sn6fN7/5zXzpS1/iYQ97GEuXLuXaa6/l7rvv5uabb8Y517T/D3zgAwwPD+/TmERERERERESks82qkG14eJhbb721KTi68cYbAXYbsr30pS/lqquu4vjjj+frX/86Z511VtP1Dz74IJdddhk/+MEPmi5/1KMexWmnncbDH/5wHv7wh3PUUUdRqVT2aqyXXHIJ55xzzl4+s6m7/Te+8Y1JL9++fTtXXXUVABdeeOFeP26lUuH8889nzZo1fOELX+A1r3lN0/W33HILixYtarps69atvOIVr8B7z9VXX82zn/1sADZs2MBjHvMYPvWpT/G0pz1tr5//Rz/6Ue68806e8Yxn8J3vfIfu7m4gr1a74IILuOaaa/jABz7A3//93wOwbt06vva1r/GUpzyFa6+9FmMMH/nIR3jf+97HNddcw3Of+9z6vv/whz/w+c9/no985COsWLFir4+LiIiIiIiIiHS2WTVd9KabbsJ7zyMe8Yj6ZTfeeCPWWv7sz/5s0vt897vf5aqrruKQQw7hhhtumBCwARx22GF84Qtf4Fvf+lbT5e9617u47LLLePrTn75PVWazwXe/+10qlQqPfOQjOe644/b6fp/73Oe4/fbbedvb3jYhYAM49dRT6e3tbbrsK1/5CoODgzzzmc+sB2wAhxxyCFdeeSUAn/rUp/Z6DDfccAMA7373u+sBG0B3dzfvf//7gbziseaWW24hyzJe8pKX1KfTvuIVrwDgd7/7XdO+3/jGN3LMMcfw1re+da/HIyIiIiIiIiIyqyrZasFJa8h2wgkn1KePtqpN+bv00ks55JBDdrv/Rz/60QdppDPfP/zDPwDwkpe8ZJ/u9+UvfxnIw6i9dd111wE0VYzVPPWpT6W7u5sf/ehHjI2NNYVmu9LV1bXH2yxevLj+/bZt2wBYuHBh/bLa97XrAP7pn/6J66+/nh/84AeUSqU9PoaIiIiIiIiISM2MrmT7yU9+0rRA/9vf/nYAHvrQh9Yv+/3vf8+tt95a//dRRx1Vv//mzZv55S9/iTGGF77whW16FjPPmjVr+OlPf0qpVOIFL3jBXt9v7dq13HXXXaxcuZLDDz+cn/3sZ7zrXe/ita99LR//+Me56667Jr3f73//ewAe9rCHTbiuXC5z6qmnMjY2xh133LFX4/iLv/gLAK644grGxsbql4+NjXH55ZcD8MpXvrJ++RFHHAHQtP/bb7+96bqdO3dy8cUX8+xnP7u+xpuIiIiIiIiIyN6a0ZVshx56aD0siTHy9a9/naOPPprHP/7xANx1111cf/31PPnJT66vn7VkyZL6/WtTAVetWsWCBQumdewz2T/+4z8SY+T8889vqvjakz/96U9APr329a9/PZ///Oebrn/f+97HFVdcUQ9DAQYHB9mxYwcAK1eunHS/K1eu5Ne//jX33Xcfp5122h7HcfHFF3P99dfzve99j6OPPrpe2XjjjTcyNjbGJz/5SV72spfVb//Qhz6UQw89lM985jM8/vGPZ+nSpbzrXe/CGMP5558PwGWXXcb27dv5zGc+s9fHQ3KRiAc8EIrvA+CJRMDUN1P/3u7i8tp1jt138RURERERERGZaWZ0yHbCCSfwla98BYCbb76Zr33ta7zqVa/i3e9+N5CvmXb99dfz5S9/edIAZ8uWLQAsXbp0+gZdqAWBrb7+9a83BUDTdftG+ztVtDa18qabbuLXv/41l156Ka985StJkoRvfvObvPe97+Xiiy/mxBNP5KlPfSqQV4jVtK7VVtPX1wfA0NDQXo2jr6+P6667jr/6q7/iH/7hH5o6wD7+8Y/nMY95TNPtu7u7+cQnPsFLXvISHvKQh9Qvf93rXsdpp53G7bffzmc/+1k+8IEP1CvbAEZHR+nu7q6v4zbXRSJVoEqkSqRSfK1dlrUEaPlXiFMwFgOUgARDCUMJmr4mTZfl3yuYExERERERkXaa0SFbo//3//4f0Bwu/eQnP2HVqlW7rJBqp/POO2/SZgnHHntsW25fc9NNN/GnP/2JBQsW8PSnP323t20VQgAgyzJe97rX8cEPfrB+3Tvf+U42b97MJz7xCT760Y/WQ7apsGbNGp761Keybt06vvnNb/LkJz8ZgO9///u89a1v5ZxzzuEHP/gBj33sY+v3efGLX8yqVav47ne/y9jYGOeeey7Pec5zgHx9uSOOOIJ3vOMdAFx11VVccskl3HfffcyfP583vOENXHbZZVg7o2dX71FKZITIKIEKtARpeZg2U0Soh3t7G+M5oBuzy00hnIiIiIiIiEylWRWy9ff38+d//udAXiF10003ceGFF+7yPrWpkJs2bZqWMTa65JJLOOecc2bM7WtqVWzPe97z9qqBQKP+/v769y9/+csnXP/yl7+cT3ziE/Vpm93d3U33GRkZmbRBxfDwMAADAwN7NY6LLrqIW265hWuuuYZnPetZ9ctf+tKX0t/fz3Oe8xwuvvhibrzxxqb7PepRj+JRj3pU02VXX301P/zhD7n22mvp6uriN7/5DRdccAHnnXcen/vc57j++uv5yEc+wrJly3jTm960V+NrtwqREQIjRag2Uvw7bffAppgHhokM7yKU6wJ6sPRg6MHQW3ztmdlLU4qIiIiIiMgsMWPPLr/xjW80NT343ve+x86dOymVShhjGBgYIMuyptu1hk6nn346APfccw/bt2+f/icxw3jvueqqqwB2G07uypFHHln/vrHBROtl3nu2bt0KwLx585g/fz4A999//6T7rV3euP9dWbt2LT/5yU/o6uqatBLvmc98JuVymV/96ldNTREmMzo6ytvf/nae/vSn1yvvPvWpT9Hf388///M/88xnPpNPf/rTPO5xj+PKK6/c49imW5XIVjxrybidlN9S4X8Z40Yq/IGUu8lYh2dHBwRse6MCbCewDs89ZNxCyq+o8r+McTNV7iFlE55RQruHKiIiIiIiIrPQjK1kO/bYY7nooosA2LhxI9///vd55CMfyQknnADAz3/+c+644w5e/OIXkyT50zjxxBOb9rF06VLOPPNMfvnLX3LVVVfx2te+dnqfxAzz4x//mHXr1nHkkUc2TaXcWyeeeCLd3d2MjY2xbdu2CWvd1YI1aK56O/3007nhhhu46aabOPnkk5vuk6Ypt9xyC93d3Rx//PF7HEMtkOvr68M5N+F65xx9fX1s27aN7du3TzqltuajH/0oGzZs4LOf/Wz9sttuu40TTzyxqaruzDPP5Prrr2dwcHDSSrzpEIsKrUFCsUXGpmQ1tM6TkYdv24G8Hi7/xdiPZQBDP5Z+VbyJiIiIiIjIHszYkO0xj3lMfQH7L37xi3z/+9/nyiuvrIdDj33sYxkbG6tPf9yViy++mOc///lceumlPPvZz2bZsmW7vO3//u//ctZZZx28JzHD1I7VhRdeuF+L+Xd1dXHeeefxve99j5/85CcTQrHrr78eyLu5NoZRT33qU7nhhhv4l3/5lwkVdNdeey1jY2M87WlPo7u7e49jqIVmW7duZfXq1Rx99NFN1999991s27aNvr6+pk6zre6++24+8YlP8K53vYtVq1Y1XTcyMtL079p01ulsgJC2BGpDBNVXTaM9BW/zsczHap03ERERERERqZsVpRk33HAD3d3dnHnmmQCMjY3xq1/9irPPPnuP933e857HC1/4QjZs2MDZZ5/Nz3/+8wm3Wb9+PW94wxv2awrlbDEyMsI111wD7HtX0UbvfOc7Abj88su544476pevXr2a97///QATKgZf9apXMW/ePL73ve/xr//6r/XLN27cWN/f29/+9gmPdeKJJ3LiiSfywAMP1C87+uijOe200wB4zWtew44dO+rXbd++nde85jUAPOtZz6pXOE7mzW9+M4ceeiiXXHJJ0+WnnHIKf/rTn/jtb38L5B1P/+M//oMjjjhir9eM2x+jBB4k43aq/IoKP6fCH0lZW0z3VMDWfrXgbS2eW0j5Xyr8ngr3kbGDQFRloYiIiIiISEebsZVsjX76059y5pln1hfqv/HGG6lUKnsVsgF885vfpLe3l6997WucddZZnHjiiZx88smUSiXuvfdefv3rX+O950lPelLT/a677jouv/zy+r+r1bz/4iMf+cj6Ze9///untJPmwfJv//Zv7Ny5k4c//OH1Kbf746yzzuIDH/gAl112GWeccQaPfvSjcc7xs5/9jKGhIc4//3ze9ra3Nd1n0aJFfO1rX+P5z38+z33ucznnnHNYvHgxP/rRj9i+fTtve9vbJm3icPvttwP5lNJGX/rSl3jiE5/ID3/4Q4499lge8YhHAPCLX/yCLVu2cNRRR+12DbVrr72W6667jmuuuYaenp6m697xjnfw7W9/m8c//vGce+65/Pa3v2Xt2rV84Qtf2J/DtUuByHYCW4tNUz9nnwjsILKDjPvIu5vOx7IAy0IsfbPjbxgiIiIiIiJykMz4kG316tWsXbu2vj4b5JVtwF6HbKVSia9+9au86lWv4itf+Qo33HAD//Vf/4X3nuXLl/Oc5zyHF7/4xRMW0t+0adOEDpVA02Xt6Fy6Pxqnih6oD33oQ5x++ul89rOf5Re/+AVZlnHCCSdw0UUX8YY3vGHStdKe85zncMMNN/DhD3+YX/ziF1SrVU4++WTe8IY3NP1s98YjHvEIfve73/Hxj3+cH//4x/zoRz/CWsvRRx/Nq1/9at7xjnewaNGiSe9bqVR4y1vewnnnndfUmbTmtNNO49/+7d943/vex7XXXsvy5cu54oor6hVyB2K0CNS2Ediu6rQ5x0M9NAUoQT1wW4CjW1NLRURERERE5jQTY1QJjcgUqFWrbSuCl1FVq3W0PgxLcSzB0qsqNxGRCV5uVrZ7CCIiIiIHZMZXsonMJhmRzXi2FOGaqtWkZpjIMBn3Ar0NgZumlYqIiIiIiMwNCtlEDpAnsoXAJjxbCapXkz0aIXJfsZZbL4YlWJbg6FfgJiIiIiIiMmspZBPZD4HIVgIbi2BNFWuyv0aIrMGzBk9PQ+A2oMBNRERERERkVlHIJrIPdhTB2iY8WbsHI3POKJG1eNbi6cawHMdyHGU1TRAREREREZnxFLKJ7MEogY0ENuAZ02RQmSZjRO4l4z4yFmE5FMdCLEaBm4iIiIiIyIykkE1kErGYDvognm2aDCptFIEtBLYQ6AKWk7AcR5fCNhERERERkRlFIZtIg5TIBjwPqmpNZqAKFA0TxqvbFqm6TUREREREZEZQyCYC7Cyq1jbiVbcms8JWAlsJlIFDcByKo1vNEkRERERERNpGIZt0rEhkE4EHyRhU1ZrMUlWoN0tYjOVwEuYpbBMREREREZl2Ctmk41SJrMOzjoxquwcjchDla7dVmY/hcBIW4do9JBERERERkY6hkE06xiiBNcWUUNWtyVy2g8gOUvrIWEnCMq3bJiIiIiIiMuUUssmcN0rgPjI2ERSuSUcZJnI7KfdiWIljOQ6nsE1ERERERGRKKGSTOWuEwBoyNqqVgXS4CpG7yVhDxmEkHIajpLBNRERERETkoFLIJnPOcBGubVK4JtIkBe4j434yluNYSUKXwjYREREREZGDQiGbzBnDxbTQzQrXRHbLAw/gWYdnBY7DSUgUtomIiIiIiBwQhWwy6+0sKtcUronsmwCsLcK2I4pppFZhm4iIiIiIyH5RyCaz1hiR1aSaFipygDLgHjIewHOUupGKiIiIiIjsF4VsMut4ImuKQEDxmsjBUym6kd6P4WgSFuHaPSQREREREZFZQyGbzBqRyAYC95JSbfdgROawYSK3kLIAz9EkDGDbPSQREREREZEZTyGbzAo7CNxNyk5iu4ci0jG2E/gtVZZiOYqEHoVtIiIiIiIiu6SQTWY0rbsm0n6bCGymykocR5DgtF6biIiIiIjIBArZZEbSumsiM0sk70S6kcBxWq9NRERERERkAoVsMqNo3TWRma1SrNe2BM8xlOhSVZuIiIiIiAigkE1mkFECd5CxQ7VrIjPeZgLbqHAUCYfhMArbRERERESkwylkk7aLRB7Acy+Z4jWRWcQDd5OxAc9xlNSFVEREREREOppCNmmrEQJ3kDKorqEis9ZOIr+lymE4jiIhUVWbiIiIiIh0IIVs0haRyFo895EpXhOZIx7Es7lYq22pGiOIiIiIiEiHUcgm024ngdtJGVa8JjLnVIFbSdlYTCEtq6pNREREREQ6hEI2mTaByH1k3I9XvCYyx20hMEiF4yixRFVtIiIiIiLSARSyybQYLNZeG1G8JtIxUuBPpBxC4Bit1SYiIiIiInOcQjaZUpHIvWSsxbd7KCLSJhvwbMdzAmUWqAOpiIiIiIjMUTrbkSlTIfJ7qgrYRIQKcDNVVpMSVdEqIiIiIiJzkCrZZEpswXM7KVm7ByIiM8paPNsJnEiJHv2dR0RERERE5hCd4chBFYjcTcofFbCJyC4MEbmJKhtU5SoiIiIiInOIKtnkoBklcBspQ5oKJiJ74IHbSdmG51hKaoogIiIiIiKznkI2OSg24bmDVHUpIrJPNhIYosrJlOhTcbWIiIiIiMxiOqORAxKI3EnKrQrYRGQ/jRL5HVU267eIiIiIiIjMYgrZZL+NEPgtVdbpxFhEDpAH/kTKveo+KiIiIiIis5Smi8p+2YLnNlWvichBtgbPTiInap02ERERERGZZVTJJvtsLRl/VMAmIlNka1ElO0Jo91BERERERET2mkI22WuByO2krCZr91BEZI4bJfJbrdMmIiIiIiKziEI22SspkZupskEnvCIyTbROm4iIiIiIzCZak032aJjAH0kZ00muiLSB1mkTEREREZHZQJVssltb8fyOqgI2EWmr2jpto1qnTUREREREZiiFbLJLD5BxixociMgMMUrkd1QZUtAmIiIiIiIzkEI2mSASuZOUu9XgQERmmBS4mSrbFP+LiIiIiMgMo5BNmmRE/kDKOp3AisgM5YFbSNmo31MiIiIiIjKDqPGB1FWJ/IEqw1p/TURmuAjcRkpKZIU+ykREREREZAZQJZsAUCHyewVsIjLL3E3GatJ2D0NERERERER//hcYJXAzKRUFbCIyC63FUwWOJ8Fg2j0cERERERHpUKpk63DDBH5PVQGbiMxqG/D8kRSv32UiIiIiItImCtk62FARsFXbPRARkYNgK4E/UCVV0CYiIiIiIm2gkK1D7SBwM1Wydg9EROQgGizWl1R1roiIiIiITDeFbB1oK54/UMW3eyAiIlNghMjNVKkqaBMRERERkWmkkK3DbCrWLQrtHoiIyBQaVdAmIiIiIiLTTCFbB1mP51atViQiHWKEqDXaRERERERk2ihk6xDrybiDtN3DEBGZVsNFRZuCNhERERERmWpJuwcgU28TnjvU4kAOEgeUMZQBh8EBtv41v8wWt7MYDBCJRJiwhYbrUiAlFhtkxVeRAzVcVLSdRpkE0+7hiIiIiIjIHKWQbY7bguc2RRWylyzQi6EHUwRp+dYF9e+nM6SIRDKoB28VImMtW6UI6UR2Z2cRtD1EQZuIiIiIiEwRhWxz2FY8f9IkKZlEQi1Ms/Rh6sFaN6aoPZsZDIYSUNrNmEIRto0SGSm2nQRGFL5JiyEit1DlVAVtIiIiIiIyBRSyzVE7CArYBMinbQ5gmYdhHpY+LF1zKGCwRUjYCyxuuNwTGS4Ct6Hi67DeER1vsAjaHkIZN4feByIiIiIi0n4K2eagnQRuoUpo90CkLbox9UBtXlGpNpOq06aLazgONbXgbYjA9mLzbRyjtEcetKWcSklBm4iIiIiIHDQK2eaYUQJ/oKrgoIOUgUU4FmKZj6Ws0GCXGoO3FeRrvg0R2U5gG4FBgmrdOkSt2vdUSh0ZQouIiIiIyMGnkG0OqRC5mVRtDuY4AwxgWIRjEZb+hkot2TemIXQ7grzSbbCocNtGYKcitzltG4G7yDiOUruHIiIiIiIic4BCtjkiLTrnVRQKzEklqIdqC7FatH2KOAwLcSzEcTQwRmQznk14hvTempPW4enGcLg+DkVERERE5ADprGIO8ET+SJURhQBzSgIswbEUywKsprS1QTeGlSSsJKHSELgN6r02p6wmowvDMly7hyIiIiIiIrOYQrZZLhK5nVQn/XOEBRZjOQTHAixWwdqM0YVhBQkrisBtSxG47dB7b064nZQuDPM1/VpERERERPaTQrZZ7j4yNquP6Kw3H8shWJbgNBV0FujCcBgJh5EwSmAdng14rYc4i0Xgj1R5KGV6FbSJiIiIiMh+UMg2i23Es0Z9RGetBFiO41AcPTqpn7V6sKzCchQJW4rAbbuC71kpA24h5aGU1aVXRERERET2mUK2WWqIwB2qm5mV+jAchmMZDqcT+TnDYliKYymOUQLri+q2arsHJvtkrFjj8jTKen+KiIiIiMg+Ucg2C1WKk0DVysweBliC5TASrfnUAXqwHF1Ut20msJaMnVq7bdYYInIbKSdTUsMRERERERHZawrZZplaJ1FVx8wOJeBQHIeS0KWT9Y5jGqrbtuFZq6mks8YWAveQcQyldg9FRERERERmCYVss0gkcgepKmJmgRKwkoTDNCVUCgtxLMQxROB+MjYpbJvxHsDTUzS5EBERERER2ROdOcwia/A6MZ/hSsDhJByqcE12YQDLSZQ5isD9xbptelfPXHeTMYBlQNO8RURERERkDxSyzRKb8NxH1u5hyC6UySvXFK7J3urBchyWI0lYQ8Y6vGpUZ6AI/IkqD6OLkt7bIiIiIiKyGwrZZoEhArerk+iMVGa8cs3qBFz2QxnDsZRYgeNeTSOdkSrAbaScqkYIIiIiIiKyGwrZZrhUnURnJAusxHE4iSrX5KDoKaaRriwW3N+hd/2Mso3AGjxH6mNTRERERER2QWcLM9ztpOokOsMsxXI0JboVrskUGMByOmW24llNxrAmkc4Y95ExD8NCXLuHIiIiIiIiM5BCthnsfjK2qpplxhjAcAwl5mkBdJkGi3AsxLKRwGqF7TPGbaScgVXILiIiIiIiEygtmKGGCKxWo4MZoQycQImHUlbAJtPKYDgEx5/TxQpVT80IKXArVYIqDEVEREREpIUSgxkoI3IrqU7h2swAh+N4OF0cgtOC59I2SVFFeQZl+vU6bLshIvfojyAiIiIiItJCIdsMdCcpY4rY2qofwxmUOZqSGhvIjDGA5QzKHEOiurY2exDPRny7hyEiIiIiIjOI1mSbYdaTsUnrsLWNBY4kYaUq12SGMhhWkLAEx92kbNbvi7a5g5R+DL36e5WIiIiIiKBKthllhMBdmoLUNvMwPIwyh5MoYJMZrwvDyZQ5hRJd7R5MhwrAraRan01ERERERABVss0YoViHTTUp088BR5NwqKrXZBZajGM+lrtI2ajfINNumMgaMo6i1O6hiIiIiIhIm6mSbYa4m4xhVUNMuwVY/owuDlP1msxiCYYTKXMSJf3lpA3W4hlSwCkiIiIi0vEUss0Am/Gs0wLa0+4oEh5CiW6FazJHLMXxZ3SxUL/ap1UEbte0URERERGRjqczsTarErmDtN3D6Chl4HTKHKHqNZmDujCcSolVJHp1T6MRIqu1pqaIiIiISEdTyNZmd5HqtGwaLSqmh87XS1/mMINhJQkPpUyPorZp8wCeHZo2KiIiIiLSsZQ0tNFmPJt1QjYtDLCKhFMpU1LoIB1iAMvDKLNEv+qnze2keE0bFRERERHpSDrzapOMyF2aJjotujE8lDIrtSS8dCCH4WTKHKXX/7QYI3KP6pNFRERERDqSQrY2uYeMarsH0QEWFpU8A3qpS4c7goRT1X10WqzDs1XNbEREREREOo6ShzbYhme9TsCm3ApcESpoeqgIwCIcZ1CmV++JKXcHKZmmjYqIiIiIdBSFbNPME7lTU4mmlAGOI+EYSuoeKtKiB8sZWqdtylVBSwKIiIiIiHQYnWVNs3vJGFN1w5RJgIdQ4lBNihPZJYfhJEocqffJlNpIYJuqlkVEREREOoZCtmk0SOABnXBNmR4MZ1BmAa7dQxGZ8QyGI0k4Sf12p9RdZAT9YUVEREREpCMoZJsmgcgdmjo0ZRYUU+B69JIW2SdLcTyEkqLpKTJK1B9XREREREQ6hBKJabIWz4iqGabEIVgeogYHIvttAY7TKVNu90DmqDVkVPT7X0RERERkzlPINg1GCaxRs4MpcRiO49XgQOSA9WM5nTI9ei8ddB64R5XMIiIiIiJznkK2aXAPmWoYpsDhOI5VwCZy0PQUQVu/3lMH3SYC2zVtVERERERkTlPINsW24dlCaPcw5pyjSTiaUruHITLnlDGcTpmF+ng46O4iI+pPLiIiIiIic5bOoqZQJHKPpokedMeScDhJu4chMmc5DKdQYok+Ig6qETVBEBERERGZ03QGNYXW4xlW1cJBdQIlDlPAJjLlLIYTKbFYHxMH1X1kVPW5ICIiIiIyJ+nsaYpkRO5VFdtBY4CTKXEIrt1DEekYFsNJlFikj4qDRk0QRERERETmLp05TZG1ZDqNOohOosQSBWwi085iOJmS1mg7iDYS2KG1OkVERERE5hydNU2BMa27c1CdoIBNpK1ssUbbAn1kHDR3kaoJgoiIiIjIHKMzpimwmlQ1CgfJcSSaIioyA9SCtvn62Dgohols1CeFiIiIiMicorOlg2yQwCadOB0Uq0g4VE0ORGYMh+FUSszDtHsoc8J9ZARVs4mIiIiIzBkK2Q6iSORurcR2UByOY6UCNpEZJw/ayvQpaDtgY0TWaWkBEREREZE5QyHbQbSJwJCqEg7YchxHU2r3MERkF5IiaCu3eyBzwBoyvD43RERERETmBIVsB0kkci9Zu4cx6y3Bcpwq2ERmvK4iaNOKiQcmBe5XNZuIiIiIyJygkO0g2UBgTNUIB6QfwwmUMJqGJjIr9GM5iZLesQfofjJSfX6IiIiIiMx6CtkOgkhkrarYDkgZOIUyTqfrIrPKIhzHqvr0gHjyoE1ERERERGY3hWwHwSYCo6pC2G8WOJkyXQrYRGalQ0k4XBNHD8iDeFWziYiIiIjMcgrZDlAkskYVCAfkeErM00tRZFY7ioSleh/vN1WziYiIiIjMfjojOkCbCYyo+mC/HYFjmSpgRGY9U6ypOKCK1P32AJ6qPk9ERERERGYthWwHQFVsB2YJliO1lpPInGExnERZ7+r9FFA1m4iIiIjIbKaQ7QBsITCsqoP90qdOoiJzUjeGEym1exiz1oOqZhMRERERmbUUsh0AVbHtHwecTEmdREXmqEU4jtA08P0SgHX4dg9DRERERET2g0K2/bQVz05VG+yX4yjRo5eeyJx2JAkL9T7fLw+SEfT5IiIiIiIy6+gMaD/dpyq2/bJcjQ5EOoIppo12tXsgs1AKbFQ1m4iIiIjIrKOQbT9swzOkKoN91ovhGC2JLtIxSkUjBE0M33f3K2QTEREREZl1FLLth7U6+dlnFjhJ67CJdJx5WFYpXN9nI0S26bNGRERERGRWUci2j0YIbCe0exizzjEk9OnlJtKRDsOxQO//faZqNhERERGR2UVnPftIXd/23VIsh6qSRaRjGQzHU9JqjPtoG4ER/VFHRERERGTWUMi2DzyR9QrZ9kkXeTdREels3VqTcb+omk1EREREZPZQyLYPNuJ1urOPjqNEonXYRARYTsJifezsk414UjXaERERERGZFXS2sw8eVMS2T5ZhWaQJYiLSIA/eZW8F9NkjIiIiIjJbKGTbSzsIDKuaYK+VgGM0TVREWpQxmkK+j9aREfT5IyIiIiIy46mgYC89SNbuIUypGCND27YdtP0dR4khRg7a/kRk7nBAiSpbDtKi/gMLF2LM3J2WXiWfNrpcH9kiIiIiIjOa/ot9L1SJbJ7jHd6Gtm3jpG2jHHvsse0eiojIPrs+jrZ7CFNqnUI2EREREZEZT9NF98I6fEdM1Fm4cGG7hyAiIpMYIjIyx//YIyIiIiIy2ylk24NIZN0cnyoqIiIz3wY1QBARERERmdEUsu3BFgLVdg9CREQ63gY8sSPqqkVEREREZieFbHvwoCoHRERkBqgC2zRlVERERERkxlLIthsVItt1QiMiIjOEpoyKiIiIiMxcCtl2Y5NOZkREZAbZQiDTlFERERERkRlJIdtuKGQTEZGZJKDPJhERERGRmUoh2y6MEhhStYCIiMwwmjIqIiIiIjIzKWTbhY1ai01ERGagQSKj+owSEREREZlxFLLtgqbjiIjITLVen1EiIiIiIjOOQrZJ7CQwoqmiIiIyQ23EE/U5JSIiIiIyoyhkm8RGVQiIiMgMVgG2a8qoiIiIiMiMopCtRSRqqqiIiMx4WxSyiYiIiIjMKArZWgwSqbR7ECIiInuwRX8QEhERERGZURSytdBUURERmQ0q5GuIioiIiIjIzKCQrUEkslkhm4iIzBKaMioiIiIiMnMoZGuwnUDa7kGIiIjsJU0ZFRERERGZORSyNdiqigAREZlFdhIZI7Z7GCIiIiIigkK2Jpp2IyIis42q2UREREREZgaFbIURgqoBRERk1tEfiEREREREZgaFbAVNFRURkdloB4FMfyQSEREREWk7hWwFhWwiIjIbRfQZJiIiIiIyEyhkAzIiO3SCIiIis5TWZRMRERERaT+FbMB2gibaiIjIrLWVoE8yEREREZE2U8gGbFMVm4iIzGIeGNRnmYiIiIhIWylkQyGbiIjMftv1WSYiIiIi0lYdH7KNEhjTFBsREZnlduizTERERESkrTo+ZFMVm4iIzAVDWpdNRERERKStOj5k0/QaERGZCwIwpJBNRERERKRtOjpki0SFbCIiMmfs0GeaiIiIiEjbdHTINkoka/cgREREDhKFbCIiIiIi7dPRIdugptWIiMgcMkgg6rNNRERERKQtOjpkG9Jf/EVEZA7xwE6FbCIiIiIibdHRIdugQjYREZljNGVURERERKQ9OjZk80SG9dd+ERGZYxSyiYiIiIi0R8eGbJoqKiIic9EOrcsmIiIiItIWHRuyqemBiIjMRRmoUltEREREpA06NmRTJZuIiMxVWnNURERERGT6dWzIphMQERGZq9RhVERERERk+nVkyDZKIG33IERERKbIsP6QJCIiIiIy7ToyZBvSX/hFRGQOGyGq+YGIiIiIyDTryJBNU0UnKrd7ACIictB4YEwhm4iIiIjItOrIkG2nQrYJFuDaPQQRkf2m32ATqcOoiIiIiMj06siQTSceE83vzJeCiMwR+h02kf6gJCIiIiIyvTrurKRCxLd7EDPQgs57KYjIHKKQbSL9QUlEREREZHp13FnJqP6yP0E/hhKm3cMQEdlvCtkmUsgmIiIiIjK9Ou6sZEQnHRPo5FREZrt+jH6TtRgjkukzT0RERERk2nTcOYlCtokGOu9lICJzjMUwT7/LJlA1m4iIiIjI9Om4MxKFbBPpxFRE5gJV5U40rCUSRERERESmTcedkYzohKNJGejWemwiMgfM1++yCVTJJiIiIiIyfToqZPNEqu0exAyjqaIiMlfMwypma6FKNhERERGR6dNRCYumik6kqaIiMldYDAOK2ZqM6XNPRERERGTadFTCopBtIlWyichc0qffaU2qQNBnn4iIiIjItOiosxGtx9bMgKo+RGRO6dPvtAkqCtlERERERKZFR4VsozrRaNKHwemEVETmkP7O+ljbK5oyKiIiIiIyPTrqbETTRZtpWpWIzDWqZJtIlWwiIiIiItOjo1IWnWg069XJqIjMMQ5Dt363Nam0ewAiIiIiIh2iY0I2T8S3exAzjEI2EZmL+vW7rYmmi4qIiIiITI+OCdmqOsmYQCGbiMxFmgrfTFXcIiIiIiLTo2PORKrtHsAMY0BTqkRkTtK6bM1UySYiIiIiMj06KGTTSUajHgxGJ6IiMgcpZGtWIRL1GSgiIiIiMuUUsnUoTRUVkblKVbrNIqrmFhERERGZDgrZOlSPTkJFZI4yGLraPYgZRuuyiYiIiIhMvQ4K2aRRb+f86EWkA3XpDwlNtC6biIiIiMjU65ikRZVszVTJJiJzmaaMNkv1GSgiIiIiMuUUsnWosk5ARWQOUyVbs6zdAxARERER6QAK2TpUud0DEBGZQgrZmnl9BoqIiIiITLmOCNkCkbTdg5hBEsDqBFRE5jBNF22mSjYRERERkanXESGbArZmmioqInOdQrZmmSrZRERERESmXIeEbDq5aKSQTUTmOk0XbaZKNhERERGRqdcRIVto9wBmGK3HJiJzncPg2j2IGURrsomIiIiITL0OCdl0ctFIFR4i0gkUso1TJZuIiIiIyNTriJDNt3sAM0xJIZuIdIBEv+vqtCabiIiIiMjU64iQTdNFm2lNNhHpBEm7BzCDqJJNRERERGTqKWTrQJpCJSKdQJVs4yJaOkFEREREZKp1xB/6teBzs45IVmcBS37iGwG6im8MkHWDG8tvZBour31tVHtp7y5LmOzlb3Zx/a4eZ1f729XtA+ASSCM4P3671jG07mtXj9V4sCY7Brs6Lq12d7tdHlsDmYHEQPR5SZCS+1lBf1BolqHGNyIiIiIiU6kjQjadDzdzqu6YEeqvSwfP+Nz7GYkZd928hb/7+Pv53Lf/lhgiIQLWYULAmPFEyzmLMZYQAjFGTO1HagzOWqIPWGOw1hKJWGOJMU+NQsjrWYwxGGMIJs+RYoQQAxZTPFZR/WLGY9mkIZWKMWJx9eqYCDgTSD2YrMrmQcNn3/khLv+795O5SMwM3gdC8VwMgXK5jDEG5xJwhizzxBiIEUyM+ZiCxxhDyZbwRHwIJKbI6OL4eIwxxbEw9Q3Ij0HxFNI0o1QqkYV8pUZnHTHkY7I2v5GPkSRx2PzJgy1x6BFHsOPutWyJFW782OcZvn+rfq/MAqpka5YRtVyAiIiIiMgUUsjWgVTJNoMYeO6XL2fUBpwb4JK3vJJ/+N6XCaXirVkLjeL4Ty2EwMjYaD1gqlarGCBxCSEGDIboPaEIj5Ik35e1th4kAUSbn2yH2FL2Vfy7VCrlIVsYD+QSaygnSb5PA95FYgjYaLDGYJKEcmLBZXR199BtSmQW0iyQmATrHNY5Yow4lxANuFJ+OUBX2RFiwzs25kFaHv45bDEGQoaplwEWYVuRQRprsM7WLzfWYG2kWslIyiVckhCjzQPGELDWYYIlxFCMqwgkrcOWEpyx3PzH3/Pxt3yQD3/x0zzuw2/j2le8r6EMUWaqjviA2wdqAiQiIiIiMrU64hxE00WbaQpV+1nyIqkFhy9hi00hBIa2ZZyyain//r9VSEpgIq6IRPPqrzzsctEx0Du/CMdiHghFgy+q2qwxJIkBDDGCa6jqMqYhFzLkFWSYpoqwEPJgz1qLjdQrzwBsrcItRDCmXhNTu38ETDRQ6qKbeWyvbqF33gDGeYilfNZnUTGHhdjw3qxVWLr6r6Va2GbzMRavXGcgxpQQI4krAXn1We351av6MAQfAYu1gXJvfizTNMWF4njESE9Xb17d5jNCjBgSjIn15+aDZcXiZdzw21+wrHsBQ12eC750Bd9+9SX142ijwvyZSFW7zfRJKCIiIiIytToiZNPJbzOrE8+2qgVsGDjzg39NxUDqypx2+Clc+6OrCQ3TQmMtMbJ2fEkya0iLaq/xcCsSiGQ+I0kSbK1e0YA3YO3En3mMeRVa6/sjr3CLmOAxGFzDfbMY6lMyYTyIitQq7iIYRzDgbcCWShhryGIAEzDGQYxELKalriaNzSMx9ao7Q8QUz9/nj2nyALEaPWCJBhpLy2yIxED+eED0HrwnEvIpqcEAAesMWbUCaQYm5PsxKYSYp4E+4C1EZ7n6Zz/kK2/4IBd/9TOk5RLHPu5h3PWTm/KAbbI156TtOuIDTkREREREZoyOmDmokK1ZR/zQZ6hawOYiPPv/vIfURmzIGNs6yl8/4y+5bf3tRNvdWI4FUK8sq1WU1aaC1q6r79/aYp1+T2Y83gYy40nJ6psPGSHkq/fH6ImGpo0Q8408PMvM+BacISYWb8k3k6/zVNu8DaQ2kBkLLlBNM2JmCNHmmRWBjEAwkYBp2iKuaQuhCNeKKaOxqNrLfEaGJ7pIdIFoA9FmROsJNiWYFE+KNxlZTMlilcxmZDbD24zgPLbLQhliko8pkJGGjMxXqYSUNGb4GAjWEPHY4DlkwTz+/eafMT+1+EqVo1/2TChbojX51FWRGU4vUxERERGRqdUReYumizbriB/6DBXIj7/vhaFuQwyO4BIOX7yKm27/H7LuLhLrcEwMbmqhGoyvrzY+DdQUDQSKyq2mddYsxPHNJwm+VCJLElLn8MY0bbWwLRTVcY2bj4EseHwMeQhFJBqwicMmjmDywMoYcC6CiSTGEqOBor7OGYMtjkPjNplacdt4o8/8Odri+NS2IolrCAhtkRgaQoDg84DOe08MRWVbcdtaYGlNvn4b0WNMJOLzAfj8saoEvvsf/8K7X/tmyliiMTz7by8lCbH2UDLD6EfSTJ+EIiIiIiJTqyNm0+jEopnWKWonQwAu+tZnGSYwOFxhdMcYH3ztRbzpyjfRe9gSfDqUBz8xwxfB0OhYSjDjQVsgkqYpkAdutbDIGJMHSdX8+1KpRDCWECJd3V35CKzBWFsP8YzLw6skSfLQLgDGYI3FJYbMpxjyTpz5OmrUGww457DWEnwghgDEPOgLAevG6LKRdVu2MLBkPmllrB54G2MwSan+fQgBG2xTV9Doxp9TCHm9m7NdQCRET5Ik9TXooinjQwDybqXWZ3lX0RiJMSPLTFEcWMIYRzVQBGsJkBJCRuISklJC2UaCD/m6cSVHYiw+83TbEocffTR/2HA3Ixu20H34MkZ7LIeccQIP/Pb2YsqqzCT6kbTSp6GIiIiIyFTqiJBNZOaInPXKZ/Pg4BYihjT1jG2NbN92PwPLF1MJGSHJK7+ssXRZRwiB3j5HFvOKqxhjU6DjvSf68TAKgBCL0AysLWGdJUszAMaqKcbkzQ5Sn5HFgKlVu4VIjK64PhCylGjyfSfGgo9YU0zfzCLEar1qzhhDyURCiHnAFca45LMfxieGnUODgMFHg3E2n/pZSZu6nXa7nrwCLnE454gmwyYOY21e2RegXOqmq6sLl0TStIJzLp8eiyVicNZhrYNIPQAE6l9rAV4a8zAwcQmlssXaiDH5ZT2l7nw/jVWCJEW1neW+7Zv42Hs+yI/vvY3EwGMe82jee+zZat04Iylma6SITURERERkanVEyKaF/ptFIkbHZPoZAyYy/+zTioXyHaOuxGfe9TY++pUPEBb2U/KWzOY/HVNMtTQRfARba3LQMBW0FqzVLmpcn632b1t05+zu6gZgIAawpljnDIKNjO/SEk1eIWdsxMRuDC7vJgrjU0GTpL5GWiMbLSZaYsjDtpAAFZ9XxZWTvBFDCFhnCS2vwaJ+jVisp0aWNzmI5MV1nshoNgKjRVOIoruqMYZQNDgwJh+rI4MImffFsTQNx66YOpoFjDX1dezyHQS8Kda6C6bheOZVcQaLi5Z+18OdN/6W1el2cJ5n/u17+d7rPrLHl4BMr7W/v4nv/9tV7R7GjLH8Ja9h4arj2z0MEREREZE5qyNCNsVJzQLg2j2IDmSi4flf/hDbjMeFPDx68I9rSV40TNJbJvVFKZQJmFA0OIghn9ZpbL36rPa1Ueu/G8O2AMWbIL+NK+U1WT7Lu4eaaDChsWNocV9vCORVZ/l0zSJ0ihBSnwdvLaGeNz5fGw2TP2YW82q0WIzDk1fNZUXP0GLYte6gsdilsZZobDEFtPak8mcTYkPJWO2xvAcMwaT5unOEYuppcbNaDY+BGGIxtTXkD2xrC7sV1W7kwaQxBhsNIf8BYG0JDITg+c61/8oX33IZl/zdldBnGRxw0AuMTPqjlzYZ3bmTB+5f0+5hzBhZtdruIYiIiIiIzGkdsQZ+RzzJfaApQ9PIgHH5K9AtSNjuqkTj826d3vHZd1zOlV/4BJlz2GiwEUo+z31s0SwgmiL8ibG+2eLfhEDIMkLw5CFVvi5atD7vtmkyIp4YMoJPCT4lSyv4rEIIKT5U8VmadxyNHh8y0iyQZgEfIIZA9CGfRhpC3ho1GEKMRJOvW9a4ee/xIcVn+VpyGI+PeV9T7yv4mG9ZrORjiuPjiyGDmH/1WZWQVSFk9S34UKz95vOxxDz8897n+yCDEPJ9xFA/FjB+3CjuB83VbbXuqCFGXHCYYPLn5wIOgwNMzDDRkxBYtLiPf/7Zf3LaiiOxiaE7g7/4uw9A0jA1dTpfZzKpxmYhQtP0bBEREREROfg64r+4dbLbTCHb9LCQH2wfwMF5f/teYmIwRXfOW+94gFWLFjDWa8mzgHoPzUn3N1m1WiyCpnyB//HppI1bHmRlYPLQK4ueLAZ8rWto0V3Te1//CtT/XW+2UHzfuDWOrda9s7aFkN8/yzKyIghsun/MQzbvff151LbW55Bf7xu+bx6H954sKwK7YkuzCpmv5l1CG45f7fFqW+bzcM6EWA/uQhEoto6BkB8zEyP/97f/y9Of8lSSkTyUzAg84/+8twj39D6bCYLXQnmNnFUNs4iIiIjIVOqI6aIdkSTuA9V2TI9QW1DMwIWf+yCbYgVDvrYYWeTSV/813/ivrxNtIDpfzJm0xIYpmK3BWu2yxoCtdlntcqDelbPxNvXrijXfMp9hjKHsXB6sFQ8VgiEL49MtI74+Dhvzk/RoQ7GWWfPjt44zy8bvmy+HZqC25lkxV9REi7OlptBusimxtegqktEaRDrn6s87v6VtGEOGKfadj7n5+BoM0eczRvO6t+K2WEw09ammxhgCERciFRtZMn8+P/7F/7CiayEPpjsYJWOwK7L8IatYd/M9E35uMv1C1G+7RqpkExERERGZWh3xX9yqZGsWVWMzPWrrjSWwdYC8e2dMSHw323akPGzVMdx2/51E20Wsv0p3HwpMVq1GBGdtvRsm0FSZ1np/4wM2RGyIkHnSNM2rx0LAh0Dtf1nwBCIxenzMiu6mGaF4YnmVmsenGT71hCwSAwQf8DHDFxVp9UMRAQwx1Jo1hHpVmfc+n3JajKGxis7XKtZ8yNdTi0yoeKs9h1oTiMbAMQ8LGxsfNMur1vIpuLVfiLa4j20J5GKMpERstGRZxs9+/0te/eKXkQVDOQUbIqe95QL90pkhNF20mVHIJiIiIiIypTriv7jVSbOZIrbp0WXLYOCZX/wABI+JYL0hpcRTznwCP/3fH0NXF8Z4nG94KxZrjjUGab5hfTFixAaPCR4XA85EnInYGLDk66jhTVFKl2uaVpqBrwZMsHllWtGUwOfpFA6PiQETI6XMk1QyEm8pe0s5QFcI9FQCvdVIb+bo8Qn9qaMvs9iYT32NwZPFjCqB1NTWOwOKgI8IlTzCo0IgDQEfIcRiCiu+HvyFmOFjhg0BE3zeajUGiKGYDtscPIYQwRtiBibY/FgwXh03YQqoz4M/Y/JOoyUDiYGE/JjWQrpa6JaYhASLc5bu+T387r7b6E6BYEgTR9VGnvml9zW/GPQrqC2qWui/iVPIJiIiIiIypTRdtAOptmN6VEKVU5/8KLIkkBCpRoONCUPDozz7EY/jPZ+7hGRpPxCL7prjTBGm1dRWUopFl89gEvAZ2ViGTRKSaMEW+VMt0Yl5J01TTHusmVjJaLH5HSkVQVeXhwWlbl59wUv57nVX88Lzn8XirgF6smLdssxjrKHkE8aSyHACa3Zs4ve/+g398+bx33/8FdFaqknA21h0F43YaMgMWA+lEMFabKyFh8WgKZaPAzIb8SZSCuBtfrWLAW8gmogpGjIY5+pVasYYvPdYaydMo53sa1PgxngeVrufmWTKYa0TKwE++e0v8533fo53fuOTuOhxEYZLkfmrlrLjnk357WtVjbt/ychBtnN4qN1DmFE0XVREREREZGp1RMimIpJmCtmmiYUjLzwPTwoejIGKtazoXsaWLffTv2Qhoz4QnQHjMWb8BHiyaY05Q8km3PF/b+LOm/4AY1WcK9GzZCFnPeOJuMV9ZDEdX3cs2j3sD2wMeAIxsbixwCOXHcNzH38+/SHBVyJ3bNnIx//pG5w8sJxXPOnpLKCcV8E5y0i35aY7b+Hq63/ISJfhjGVH8soTHs1zTjyTf7vz1/zgDzeSlgwhCwQLJhrmjQVGeh0mBrqjY4RIX7CQ5c0L6ElIskhS9Yx1WaIPZDYPGrutI40GmwWGkwjWUDIlQqThOYO1Df9umfZZP8a1I9ownXRf1O5z7GGH86fB9SxME3YkhpiA8Z6z3v9avv+qyyE05YcyjYZ2KmRrZJ0aH4iIiIiITKWO+LN2RzzJfZDqbH/KWAxgwMKLv/xBrIkkOIJ1eJcQ0sAHX/E6Pv1PX2DEBnCGYC3YZHxKorX1LUmS/HsgWEtXUmbL3Wu5/1d/gHU7YFtGeQSq923lhn/9L1Kf4Zylq1ymq1Siu/ha28qlUn3x/9rmjKPc1cNLTj2bL7/wTbz27KewPJTpr0bSahUSRzUx3LxzA+/57tf4/E+uZbinxH/deTNv+9b/xzd/+SMGew1pAot755OGSF8s8YLjH8UXn/9mXnLGuXQFh0sS+rF88tVvpT8NLKSLj134BvqGMj5+4Rs5buFyjl+4jIcdfiznHH86rzzvORzqe/jMq97B4q4B3n/haznvIY9kqenjuec9i8MWLsM5RyyZ+nOy1uKcy59/uUwpSSiXSpRKJZxzTbdJkoQkSeqXNx5/aK52a/3Z1C8DfGK48jNX8I5Xv56wswo+4orCvKd98X0kcbySTabXToVsTXp7ets9BBERERGROa0j8idVsjVL2z2AucrkUyIdQDcMducLr8eiS6fNuti2dgch3Up5XgkMRBuLN+H4q3SyqqpoLGDIvOcX3/9vRjdshNExzFiF6vZBwkiFsU1bsM7k0y9bKtfq1Vox1kOm2uUWA9WM0WqVavRE7wnVKmnZMJR4wBLSQJVI1UYS5wghsN15vDV4Z/HOkDooJyUqiaHqDFULziSccszxQKSSpfixCjFL6ekq40eqzIsl+gcG+Jdrv8f2tetY2bOAl/3Fs7jrnrs5+tCV+Bi4+57VvOa5F7KIbtLBEbLE8uOfXk/qPcGANdTDLxifElef7llMCW0NMJ1z9a11GmnjbWr3m/TnArjgWXLUCtaPDbEyK+dr4sW8i2xaijzqwmco1m4ThWzN+vsG2j0EEREREZE5zcTdzSObI9bjuUPRUt0qEla2zBTeunUrAIsXL27HkOYEA0RrMCHyrG9dRpUUj8cYQ4iOmPbwtgveyNe/9mG29yXsaoZiLRRq/D7D4CJA4P998VtU/7AWm1pCdNjoCF2Ox7382SQPPZxImDA9tHFtspSMlivzxgcm0j8UuegvnsHDD11Fz4jnB3f8ln+6/UZiFnnskSdxwSOfRNnna6tFZxgrwTf+77X8fvO9jBJ40SmP5enHPYw0eFbv3Mqn//UbDM8rMeoCzljKMbKYMhtNBRsNy2w32ysVer3l2OOOwSew5q572BSqHDd/OaX+Hoa37WDzzkG6MMxfsIBVq1axZdMm/rD2TirdDhvz4w6Mh2G++eC2/jVhsmPcOG3UGoM1+XWhJaZvWqMt5gFot0/49a1/5N8//rdc/NXP0lUq51NcY4rD8t+v/QiuYvGarD1lJvsoe9f738y69Q+2YTQzT3dXN1/622+1exgiu/Vys7LdQxARERE5IB2xJltHPMl9oOmiUyPPwCLP/thbGTEp+UJhEKIhBsvvfncbCy+ssqVkcOTrsAHY4scRdlNY6mIgAl3W8ZBHP4rf3H4/seIhREJiKPX38+CW9RweV+ZhkZ2k6qoIISy2aaH/rLht2UPa6/jiDf/Jt0cDb33Bhdy16X7OPupknnvWk3AVTzeWJARwlhgi3cHw8rOfwvYuz9d+ch22v5vfVjbxpe99h50ukA0kRDwxC1RdxoiBMROoRk+XNzwYhsksbCtZttx3ax5M2bzj6a2jmzHDeZuGpATbo2fTyFZW37INQsD3lICIN3ko1viMG4MzoL4mWywq+WrX1zquNt4n/1nkoenu1moLYfwnVk0Cp558IqsHt7DIdDESiq6kOKp4nva5d3PtX39MCyJOM63JNq6/X1VsIiIiIiJTrSPyp5ImjDZRTd/BV/SahDIMHtZDkgVSa3ExgQAey99+4BN8/XvfIil7Sj7iTf72q+UutVAoxph3CDW1DULw2GCohMCyE45m/ilHMnjjaiBgEkgrOznlz89gRxE2BUse8kE9SKoHSq45ZE2K9cOCIa+CC5HBLsNH//3bWAxx4xr+565bCCHQXSrjQsPzJRICePKupnc/sIaUQHSR4AyYQIghn6YaoctYTIRSNJDYIoeMlEwkswYTbX48DIQsxZk8EMuIROvJYkawDusg+ECoHT+bNzewxdgcSVN4FhgP2LIsa54eWhxz732xRl0+PTT4rJjua5uCuYgp9pkHlYmF6KsQDJdf+TE+f9lnuPSrf0u0hmgCJsI2l/EXb76QH3zuHyHk/V/j+EGUKRBCYHh4Z7uHMWMoZBMRERERmXodErJJI1WyHXy1I/qCL36IHSalSp5lBeshGm775V2seuk8/mloM8ZBmKTSrLnLZRGwFWzRT8GbAInhkec8lv++aXUeH1crnPaoR5PM6yOhSjABW3S0jMU+vc+r5srlMllMCaG5pKqW9xgDtpy/Y3yxVT2UkgRiZNgGggmUbZKHSFTBOPIZlBYXDAGbh3MmFBnheKAVY8Q4i43ja6jVj2GMEMAaC9aS1EKtGDE2X9cuH3cEk0+3dSYfee3YlazLK9DCxE6tjZVqjcd7sqm1eTVgsbZbQ9fXxjEnxTExBoy1ZFlg0bFHsHloO+n2IezSARIH0Vp6vCM97QgoR9xYflxNVMY2lYZHhnfbVbfTaD02EREREZGp1xGND1TJ1kwh20FkxlsWrHjosQyVPaUkwWFIkxSPw/Z0c+X7PsaNN/6QUAYbE7yd+DNoXJi/plaB5kIkmkDiI10ZHH34ETzh5Idyev8hPO0hj+DcY05jwUgGNhJtPoUyi6FenZUkCaVSKQ+iSiWSJGkJ9eqDoOIzojX09PWChZ75XcSyJyYpJefpKkWcy0hMSgkoE+iyji5jcCVPkniszbDWY4q2miaAjS4Pv7xpCsFqrLXYZLyTZ1J2RBvytNJEjLW4JME6lwdgxuHzGjwcJl+bzQdC2jwNtLGxQa2ZwWTNEWr/rlW1+QgBM/lxavj5eO+LKjlD2Vku+fSH+fg73k/m8im+Doglh8PwrC9eii+eut6FU0tND5oNqJJNRERERGTKdUTIlqAOo400XfQgKqqRjIM/f8dLsRhSG0lCwHoHBNbeO8Rxi7r48e9+ThYhWldfI2xXGkOd2tREGyIrdhpOubfCvO/8hpPuGeXh6XxWrBlm03U/J/zXHzh+XWDZWEKXL9Z9K0Kgxn2maUqWZVhrKZXG6zyttThrKScJMQSGh/NKoCzLGyUEU5TTYYoKOYt1BuNsXmlm8w6oMVIEYaYeHDpXym/f0AG0sWtnbdqmMQZsEZJ56Eq6KbsuSq5MKUno6e7Ou5vGCJh6d1BbdE4ljIditcByfAxuQvVc7bjUbtN4WeOabq0hXO2yxscoZvay4pijWF/dQXnzzvy1UXQajQbGbJWnXvzKzvjF22YK2ZppuqiIiIiIyNTriOmiBkOCwqWaqmpoDrpXfOFytriIDQbnI6nNA7axkHDheU/h9/f8npE+MASMSwixDLE5/KqFPMYYojUk0WBjwEcP0XHE+pSFP76DFdsjbhg2pz1sGt5Ot+nlkN6F2HvHcPf/AWyF5ccvo/rQI1mzJGGolGEshKJPpqW5Wq7cVQaKaZUxr/zy3uNckk8jjYFABm68+ivGvKlDxOGsI4SMfLUyS5Lk66FZHJGIjbYe55tiHTrnHNGmhBDJY0rbdAysA2uLX08xUkocWVrBpxll10WSBLIsJY0eawwJJbr7uqlWq4Si6YBzDqBezdd4nBun0DZWtdVCR5PPTc3fKcbgi8q4erBWW7ctBIxNGo5HpBQjb7rs3fzHx77M2790BaYLukwZQ6AMxNMPJ3SDGbPEYkU+rdF28O0cVsjWSNNFRURERESmXscUVGjK6DgPZDqbPzgMkMCmfkhD2lxF5cps3zDEk0//M6766feJBoxzTWutTTZFFKAUIoG8eUJXKHHmHVVWXv1HBu4Z5q5bV5PMW0bP8Sey/MRTOfHsx9J3yHLOfvwTme9LHDNW5vDfbubwn93P0VsN3eRTQ21tGmaSkCRJ/XFtyDdHgjMJFktXUsaSrwVnHfUqsFpFWG3KaWMwWAuvGrVWktWOT762Wm1/DWNp2J/3nizL8N5TraYkSUK5VC6m0ObhmjMOVzzG2NhYU7hWn2pbjL1xbLUmCI2PV29s0BCmNQZzTVV3MKHSLa/Cy29z6kNPZlNlC8uT7vx6G8BGfLG23HO/+KF6wAY2b3Sht+RBtXXblnYPYUYZGFDIJiIiIiIy1RSydahRndEfHBGe+62PUraGsis3hWzGdvGQQ49ly7b76VnQX4RMFmPGAydrbd7NkubGB9HkDQCChRPuGcFf9VOyu9Zz3lOeyWNedAE9p53KmS98EeGII3naa/6KW+5by9IjjuCeNfexY+sWNm9cz13fv4GBX97L/KwIxpJSvtkyNjoSU8rXSSvGZI3FWYcrgiBbdCq1GJzJmxkATYFYbUuSZHzqZvHvxstql+eBmsur5OrHytVvWzsmtfvXwjznHGnVY22CKdazs8aSuITElprCMigq5WKsV9U1dVdteA6N4WHjPrIsawrbamNoDNpap5gCJNESDSQVz4WXvIF3vvy1RSOIPBTMp7VCmkQGjltcNLQomjnIQbVu/YPtHsKMoko2EREREZGp1zEhW7ndA5hhFLIdHH/53tcSXCAQ8EUIVTuy2fZR3vLil/G3//i3GD8+JbK2kFtjZZe1tghbcsY4ujLD0p2R3l/fR/9wxrbt2zn+MY+kvGwZxz/84Xz3O9fw81/+jjf+1Vvw1vHpT3+W/qULuHfDGh4c3EhiLDt/eQf9PiGxZZzJK9Uaq9kag6zWUCz/d4I1SR5oJQlJUSVWC5vy9dCKcApTBHV5YGeNxdh8vbZIw1TLSTbDeOWYtZZYTCPNl2rLj1G5q0zm80nfJZdgMcW0TVccM1OvUKuNrTZ1dFchWe3nEWOo76PWybS2BF1rFVtrI4TGyzIiJsIInoecdhr3jg5xyFgfxvVi6cmr7zB4a3jSpW8ilFG+NkXWb1DI1khrsomIiIiITL2OCdlUydZMIdv+qb2KXPF/7rTl2JCSWY/FQ4wEC66rjzBsKKc7KC8ewAIuBmLIMHiMybAOjI0EEwgmYCi6YEaHjwmpdSwZDswfHWN+T5nBOMpV/3o1/3jNdxgeGeaE00/lE5/6JI8472ye99pX8LzXv5L7N6yjQmAoDHPkysNYVuph8VDIO5PicWQ4Y3C1YMhGbDQkxhUdQPMQyGEweVPPotItwZBgKOFMGYKD6CBa8lq3EsaUMMZhbQlDPi3WGkNqMowLWGcxzmITl3+NCY4yLlhctBgcNjocRXVaMYb8aySQEsiIJg81g3EQSsRgMCWLLVtctyEmHqzJ1zizBqxpWqfNWkt0kWgDwQZisQWTgcsnU+fTOwOYQAwen6YYItaASQIxCfnTN3l309qWYDHGUTIJSVLm4o9fzvtf9TaoOpyNWOOIzpCYSOY8F3zpQy2vLDlYVMnWTCGbiIiIiMjU64jGB6CQrdUoYc83kgnynpbgDfzlpy9hy5ZtGKAaIjHLqBAIPsNk3Xz5HZ/hc1/9CBvT9WSZJZgSxhZdNIupiLXpi7k88w5AKVqciSRbDUPWs7U7ZautsmN4iJc/+0L+9Ptb2L59Ox+59FJOPvpI/uemP1CKKWNFR8WdfT3clW5nxbGrGNy0ldGBHnaGkHfgLKqwgskfu1QqF1V1EELEhPzy2tiiNcXdDI7YVHVGsQ9ifnn+be21FfAOysbhTSB4T2IsoZi+GQkYAw6bV475hiYAtaYDQAi+qFCrVa1FiA6blAGDMTEPw2KEQD5GYwnRQ77aWX0ZvHKpRKVawcf8eOc5XK2a0ACeWFTApWmxxp61+XMJxbTOam2aKVg8sWipYotmF8Y4DJYs86w48gjujSPMT7sYK1Wx3hCdJQspziWM9kee85E3cvX7/gYTVdR2sFTTKpu3bGr3MGaUAYVsIiIiIiJTroNCNmk0ptP5/RYN9B4yn6EeR3/pEELZUa5kVCoZ0XhKaZmRrVsoMca6tEJ3aSUVk4dD0VDvYFlTmzaaZdV8iqMJOBNYGDOOoETlhEH+NLSOHc6z/va7uPHHN/Lxz36an/3pJt761tdR3byFv/vYFdz0y19RdgnDJvDkt7yELRs3s3zlkfhY5dC+hfi+MtHW1hELpFkFgG07qnnkGsk7eTpLqVQiTasYHOVSid7eXqy1lF1CySUMDAxgjKFU6iLLsvrzqK1jFopAL4mGZzz8Ufz7H/6X0dERurpLhECxRpzFFdNQyy4hJWlqSFA1hoGBAcrlMkmSEOL4dNCu7iJsMwFnHEmtk2le90YgkN/CE/EYbNH9NO8AWqscM1gCJg8OyZtNJIDHs2nLFhYtWkR1bISxsTGyakoIgWpWpVpN8WlGJQsMpxW8zyBaQoDgPYNDQyRJgq9U+cfrvsMHX/E6Pnb1F4nOMlrdSSQhHUuJMVJdvgBKEKvT8OLtEBs3rm8JsEVrsomIiIiITL0OCtlUydZI00X3jyEvBnvG595LcJEqGdZXMImlpy+hYiBW+rn8DRfzj//9T/Qv6CcjUCIlFl0+I1m+AH69mjAFLEk5n3qYOItNLPMGYWjzNjY5w39uuYMzDlnAPXf+iXPOeAwffOMbWLxgAb/5139jcMcWhga3MGBLVGJGSCxbVy7gtg13M7jlXqzroqvazXY3ho8BY13+yEUVWn9vHj3llXVgYwBTwSWAyYihynBlOH/uNg/CHtgRMMYSQ+uMc5uHG8X6cvf/8W5e/IhH8btbf4UxDl8cQ4upvwLzYeSVavlIINRSP5OvjxaNwbokr7bzgZhPaM0DNGOIIV/3jZivBUexz2JE+HpFW16ZFol47wk+kCRFyJYY8BlZMajBwUH6+vuBQIwQg8EljhgTwJKlWb7PeuVcwJE3dQgxDxljlhG9597KdsY2jDDYS72rqI8J+EDA8ozPXMq/v/7Sg/IaFU0VbVUul+nq6mr3MERERERE5ryOWZOtrJCtSQpkCtr2WQRe982PU7WjpH6U4FMwkSRm+JjhvOHe1Q8wbyTjtgfuwscUaz3WBpwNWONxxmPIMIT6Zo0vlg+LuOjpqWYsH8rYNjTIHRvW8LJ3v4Ut2QgLF89jw713sqK3TBjcSs/ICF2jYxyxeCm9fWUqScqG0ijr+iI9x61k/eA2qjbSN+ZxBEo2T9aThoYCJgZMjBACJgYigVj7GvL72WKcxAzwED0xpDjGcFSwYQwbKhhTwdoKljGIY2AdPkAXDmyGtQFjItFEbG2/NmJtvj5dNPk6aMZm4AJYDzZgjMf4lJil2OgxVIEKhgxiijFVrEmxtgJmFGtGsWYMa8bywNBUsaaKoQJhFMIYjiqlxGONx5AS0go+eFz0uOCZ39+Nycaw1YyS93RFT5JVKccKpTBCtxulx47R61J6bIU+l9LjKpQZoceO0e0qlPsyBgYMb7zi7bz3tX+NTQzlMtieQFdvoDy/St+8Kiz1nPuaZxdVhvpddaAUsjXTemwiIiIiItOjY0K2Hp24TqBqtv3QDRu6K3k5W97+YFwMRFPmy2//ID+66T+wfT1F1ZgtmgBErIl5qEYksZbE2rwRAWBrwZuB7lHPH278DfdvWs8Rj3kYm+IYpYUDDJAwTMbgyBaMq7Jxy/3EmLFxyzoGh7ZSqQ5z5nP+gh0mwy6dT9plGNqxhdEHHqA7C7h9fR8YU6+3ixRryMVI3u4g73sQbSQmEJNinTX8eIDmLNXgsREI+f2MieOdO52tfx/wRBPyZgLWFd+H+mU+j/0IRCxxPKBseh0XVXT1rg0BjC++jm+m4atpudwT8SHi04iJjmBLBFvC2wRvEjIyvPEEY8gcVK3HW4+3AW9j/WuwkSQEYvQce/KJbCHj2HmHkHpPORQVfTHfpw0l5j/+oYQkFBNe5UCsU2fRJsuWHNLuIYiIiIiIdISOCdnKeb9DaTCikG3PTPP3L/3GR/Kqq2jzxdkaWBLW3LmdxQR+ccdtEPNOk2CxpoQtum3a+iL7+bRJZywm5l/zhgKQhEhfTy+UE0a6IOspsWrJISREyj1lRqqjbBneRl9/H8uXL+eIQ1ey/JBDWbh4Gb3HHoF1hkqXg8X97Bjbyaa1DxKHR/NwsOXHXq9om3Sj6NRZNGwoDknt+hgtIVpCMIRQW9nMAUn+jiu6e9YaJ9RWQqs9bvNAbL7VJo3WH4zxBgv5HQGDLRoMGNP4a8wUyV/tjjZvkmBd/WdR2/L7WWI09ctiNBDyf1ubFKu85UFebXPRFpuhFE3+uyUaXKTemdVG8s0b8BGL4QXveT0ve+bz6TYlKO6DiUUzCU9iDS/6+kcIJuTHW/bbelWyNTl0+WHtHoKIiIiISEfomJANoFv1IU12qsPonhVdKMHwuJf9JZVSHuJEA8b6Ysl8j6eEN5EXnnse9wytIbqIsWBNILGQ2Dxwi03BXF7lFGNeaWVrC74Ridt30k+JkZ2jVHvLZGnE4dg+uJ1seJAqo/T29DBarTCSjfDgpnWs3vAAv99yH+aoxXgioyVDung+d21Yx7qNm6jsrJCRV1lhwZl82xVjGoKqhmCsFo4ZY7Dk9Xzjtwy4omrPYPDREX3RIbQItVyxtYZsJhhstPWd2ZhggsMEi42lemCHbQ4CgbwasGGrvdXzNdjyHTbevrh2/MdcdDzN7xPzrTHzM2CsKTabb/XMzuYdWK3Jq/oatuAM0UWcCTz8tJMY9RWOX3gYHvKKPQwWB8bifaSSZLzu85cWnVzHddQv6oNAlWzNlh+ikE1EREREZDp01Lmbpow2U8i2FwxABBc57MlnEmNRrRUjMeZTIj0WFwNbNnue+chH8cXvfgvb10sMWVOHw9Zuh9ZajBvfHIaSgZIPVDdtxXvPomNWkMZADyWGt+6AGLDzuzlk5UoGx4aILrB951Z6Sz0MdPez8rRTuT+MQTkBa1i2bAlpAmMm0NfXm09FtHv3PpisO2NjSLXb7o21yjPryDJPKC6aUL22G8bYYnNFoDe+YSaGfrWgbF+6SrZW7tUuq3U4zSsLTfPtGQ8V96Qe3MVIqZRw4XvfxAXnP4tsuALGNgWC3uZ9T9ctKRGTgG/Yj96pe29wcAcjI8PtHsaMoko2EREREZHpoZCtg+0k5h0ZZdeixRF51d9fiTeRiMXHQD32iEUtl/GcvHQVabaTrqUDeO/rQVC+mP3419pUxMapixSVXQ5DVzC4imd0xxDLj1hJsLB01LB4R0ZPqZvK4AiLexeycumhVEdHmd8/QF+5j4Bj1RmnkfZ0Ecgrq0bLhu4lC6ArocsHytHucrroXh2OGCeEUjBeSRZoqAozhhCLCrDifvkDQmiY/tm6LxstDtdydMY3U6wLB7sO1lpDswlB2STPYVeBWwhh8uAuFpHfJPsa3yfkP1vABE4//RT+uONB/vyoEydMCc0L4yLOZ7z4Gx/prF/OB9Hq++5u9xBmHIVsIiIiIiLTo6PO4xSyNfPAmEK23TJA7HcM9hqMTWiuX4p4DM7D9soAr/vL5/Hpb3yJ6PKAC5gQ9DjncM5NCHSiydsFeGOJIbBlZJh7hrcw2B3pdSVu/dr3ODaZz8qVK+kqdXHr3bcREsfyFUeB6cOsWEJYsRi3YB6mEjAYXDBEYznxYaezYdsmfvaznxFj2OvmlbuqCouNgdleHMDEJZRcUl/Of09VWflab7H+fePlk4Vgk217HNYuQrjW/dSq2WrPeVdh3u4et3a/GD3eZHzmK5/nuU96OtXhavPtIthgKVlHZgPHPfVMSvqVtc/uuOu2dg9hRkmShKVLlrV7GCIiIiIiHSFp9wCmU09nZYp7ZYhIT7sHMaMUi+4Xogm88GsfB5/hjQHjyQvSSngP0ZTBe0bXD9ProNKdd53EB7Dka3fVA5zIeA7TMnU01rpmJpiQkMXA8qOPIkscvRuGWVZaxvZlgSzLCPMWY+Y7fnj3nWyLY4wB6QaXN1G4+kGe+eh3srU6gknyfqXzlyzj9EefybaxMVILMcSmoG3C2mh7+DcwIQQbnxaZR49Ek/cBLd5ySbB51052neyPV/oxIdCaLNyabDy7u25XIdiE64OvN3kAKDlLlmUQfH5ZvSIvEDEQTdN4G6v9bHF57d9V41lxxJGs3bGd0xYfwZ0j6yDJMN4QrCm6tYJxltNf8nRu/9EvobKLAyaTuvOu29s9hBll2dJDsFZtf0REREREpkNHhWxqfDDREIFl6rs6zhTdN4us7Y1f+gjbLFRNPrG2FqR4n8dFloxguvmbt72Vr/zHN6mUIYkQIoAp+hjkoVZrxjMhOLKuuCzijGXBEcuxGXz7y99k7I/rSDB5V1MAY/DOUDGG6PLHcs4RhrbzL1d+nudf8gZ2jA5TTQJjZcdhp57AMmOpWNfcqXNvD0tLmLbX93MJARhLq2SlvGFA6z739vFbQ7d9uX9jALanx258ruNVaPnWWNVWu03tssZ91yvbWg5XKRoio7zn/1zOP7znU7z3q5/CDBjKBEJDgwkAksgFX/0wV134Pq3JtpeyLOOee+9q9zBmlEPV9EBEREREZNp0VGlXF0ZxUotBnb43qwVsgElg48K8+igJ4GLEFO0mnSthjCPzgXtvvZcuN8aDOzdRSvJOk5mlYc0u19AsYXxrnWYYQ8TGfMNEepct5KfX/YjhtVuJtkTVJIyYyLCNjFmoEMFHTDVC5vFpSswi229dzZf++hLGKhWwMGICww5GncFP8o5vHVfrNNHdXbcnSZJgXELqfbGW3d7b1+mfjePd38dpnTpaY63FOTehEULj4zVOMd3ldFQ83qUcfdxRrBndxhMf9hiyaAnGkph8amoIgRDyxgehDE9884v3eopvp7tv7Wqq1eqeb9hBlms9NhERERGRadNRIRuomq3VznxSX7uHMbNEsBEu/Mcr8go0wFsIBGLI+0oGD8FHTLT8zaUf4/rf/Q/W5lVoECk5R62SrWGd/uaHaQmubMwXv+8tddHX00+XLdPT3UfEkNpIZgPRgImBEDNizIjRE6MHQl4+V/WQGVYccwKlgfkEStiYl9GFaXjpN61nZqC3XIYsEkIGZvKmCftiX0O+xnE1jm9Pj9F4H6AefmVZVg/RatNbG8e0q8CyHqxGQxITTDS8+RMf5LFnnInfMoSxSX0dvzqfV1UuOPt0KO/zU+5Id9yp9dhaHbp8RbuHICIiIiLSMTouZFPzg2aRPGiTcQbLqeeeyZjNaErJIhifYWOGiR5M4I7b72dBVuHHN15P6iI+BryP+BCIZISY4kMVH9JiNxODF+cjJkJmwflAeGAj1177H5TK3dz+s98ShquYEPK1wqLPe8KGmF9mPAYPpBB9Pk7v2bRtPWQjuJCSEchw+TppLRV0e9N1s+nY7OI+zRVe479WbEgxWYaNFusctb6sAYhFw4f61nJcGm/bWAMXY6x3/NxddV1rddseK/KCJ/oMnxcIEjBgHdHY+uYjhBDqTSzAEEO+EW1TmDeZ6BzGgnWeU09cxaZ0lOed85eEkM/iDQZw+fp11rq88jBUecW3rsjX+KNWESeTufNurcfWStNFRURERESmj0I2YbumjNY58gX7H/LaZ00Iabz3BBPJosWbQOoMH3/ze7n7wTtx8+eTmQM7ji6Aq3pu/sVvqKRV/uf7/5ehBzbTRQIxNJfD2bzEzkQmVkAB1Qe2EjNPNIEJC4MdRK1rtcWW68rlblwpwXiP3cV7b2+r2va1g2hrM4RdNXBorFyrVae1/uwb12arfR9CwDTcr7WabXfyUBL++sOXcMZJD8EPjhB98+snNoxjyGY8+U0vITLxOMu4O9VZdAJNFxURERERmT4K2YRtCtnqvIU3/dOngMmrtiIGX1RZbbh/G8ctXMJ3/u/3MSaZMCV0X0IhA1BMJwxdCW9818X8z7//F0TwIWCjwQYwIe+q4Bx0LZ1HTJi4zqAxuAgbbruHiCFO0nBhV2vDNYZJk2kMpFpvE2MeAeWVZ4aAIYuRis8YrYxhwuSVZHs7/bN17bPmpzxxXbXJKtl2d11NkiSUSqWiUg1KpRIxRkqlEqVSiSxEjEuIwWCMxdhY32pjmWy8rY9preWUU0/i1m3384RHnk3acEyttdjidgEoe8O8s06Crr06VB1pw8b17Bjc0e5hzCj9/QMM9A+0exgiIiIiIh2j40K2vs57yns0SMCrNibnItvKKZm1+bTMliqnSN4BtMt186KnXsD96+4jzuvBxUDJduXhyC62PTIGnxjson5+eeONuDRAhGggEAmW/B3bbXjGJa/mwk+8iyPOeRg+icV8ynw9OANEH/jdNT+iK1hMa8q2nxq7ataOR23aZu361sCqp7eHLGaMDY9M+grbn/XV9nask+27VoHWOO7GyrTa1yzL6rerrcMWQh5GN963Zk/TbydjjKErGD7yxU/xyJNOx4ykmAgmxnqT21LMKxYzC9ZEXvKNK/L7HoyDNMfceZemirbSVFERERERkenVcYlTX31VI6mJqMtoLQh55d9fifcRTAAM0Y5vOEtiA44yD94/xONWncBXvv+PZH4MnxiMzY+hiQE7yfEMwdeDm9bNh4AnMGbhtIc/jBuu+wGegMHgjSUaB8ZA2XD2Rc/FnriCbaWMs17ydJIjFmEBGzwWj4sBfMrOTevzaYkHKcdqrNBq3Lz39QAqD4hq1/l8Dmxi6Onuz9c4a7CnNdUmq7ibbEyN1Xf19dx2cZwb13JrfYyAIZrxULQxhKvdF/IqM4KvjyGEgPe+6fa1TqRJkkwIWWv7g4gJGScdezwbtm/jL055BLECIc2waYoBMhNxRfeNYAxZ4nnzVz9W7GiSKsYOdufdmiraSlNFRURERESmV8eFbBZDr2K2CTp5XbbatLyXfeISRko+rxgrOowSIYZYbIFgIy7p42HHn0G5Okzos5hoiSHLp4s2hUW1bqPFv3YzDdMAhHxq4K9/+StcNeZ9DLzHELBkuMRw7isvYMXZD6NaingTyVzkCS95NqEMIYFgI5nxBBPoWTyfNFabxjCZXU3hbA2sapcDE8Kq+vNoqt6yJMFQKpVJkmSPYV/r/ib7997YU4fPxufV+hjee7IsI8uyenDWWq1Xe/61y2shWq26zzlXDwRrAWRr9Vs9xDMRg+fDX/sbzv2zR+NSg6MMLpnkiQWCiawbiMQe8qnE5MdZ4PY7b233EGYcVbKJiIiIiEyvjjw76+/Mp71bOzo4ZAvWQBmyoxYTbcSYiI0mn2YZAoRA9J7oAzGNbFg/xEVPfiqf++7XSFwZaw0Wn9+WxkXzm0O2Pa3NZiP0VzwvOuc8Xn7Ri5i/fAB6ArGUEnoiy//8RJY84lTGrIcQiSYSnWHRkYdSOnwAylDuSVh13OE89tzH8Ljzn0S0rb05J9pdwNYYSHnv69c33q7xstZqt8Q4ApayLWP8xECucQy+2PbUOGBvpmHurT1VyDVWpdW+1gKy5o6q4/trrPCrhWyTrWUXYyQkkYTAiqMP596dW3namY/DeENoqFGr3x6wIUL0XPiNj+ZlbJZJqyY70YPrHmj3EGacQw9d0e4hiIiIiIh0lEnKJea+fgwb2j2IGWbyFbPmKENzcVeMvPrvr2SnSSHGfFWz6AkGYjTkIZXB+0CSWY5ftJye6NluqxACJsZ8p8YCeRAVDNiWtdCMcfmaW0UoUhtCagLOWDCGAZewdd16rIWHn/lQNoeMZNlijjn9FOjvo+o8MXqcj0QiHkO1bHnmZW+lZ7DCz/75WvoqgUOWLyFNHDELxIapirsKlRqvb6oeozmiq1W07eq+ta8hhCKkCkQMpVKJasvtJmue0Hhcdvc4e4qVavtufYx9CeZql9Uq0RoDRFNcni+eNh64hRBwzjUdpxhjvbrNGUPw+Vp7FoMx4AkkWD70Nx/nb971Yb79n//GvJVL8TbUXyvWWmw0xZTW/FX21ItfxnUf/waB/K8litqk1bGrjmv3EEREREREOkpHlnSpkq3DFZmLA7Bw6HFHMlzKCARCjIQYyGIerLhocDhsNDhn2TZW5Y0vfD7//P+uxnTl1UbRGAyWEDJCyAMmfCi+b15oPxRbjDGvSgr5OmCBiI2eAVdicMMG1m/bzMaRnQz3Oo5+9EPJ5pWpuJRgUmLM21QEYwjW4C1UnSfrs+xMh1izdjX/+Z/XsXDhgiLocrtdr6z18knXSjOmPu7d7atWvQVFYOfy8LFnoAdbSnZZEVeTd3DNH2/Cj61xjA1NCmrVYrXHzrKsaZrn7oLBVruqkKtVstW2mtrPt/FY1aaZ1oK38TXYwFCsIRfGV6gzxoAJHHn8KnaS8bZXvp6+OPHvH6b4n8MQIyz88+OhlL+OFbBJq2VLD2H+vAXtHoaIiIiISEfpyLSpX2uyCeCLIrVzP/IGTNGg00WHw5GYhJItEZ0nC5EsVgFDd9ZDV8i4+d5bSaMhYonYvOoN9tj2MRqIJr9PZvKKNxMDpWrgsFHD0tRCNWPnyDC+v4cz/+JcvHV4W8KbhIAF44rHs8X+DBiDCYHHnPlILrzgAi54yQUkXSVwFs/erVE2+YCbr29cny2EMGFR/9appD5GsujpKXfhi9vXQqrW9dZqYVXjVM3Gpga+ZYzNFXOmPpZ61VjxOLVpnns7jTT6LG9sEDwmhvoWshSfpfUxhhBwRXODydaPa328xgYJjR1NQwh5VWKW8YbL3sMhfQsZ2bGzqTNt6zHBgvcZL/rWR4lW3UZlouOOPbHdQxARERER6TgdGbI5DH06Le1ohnym3+u/9WkwgYCh7B2R5oqkLEuJZEQDwyOOj736Yq756Q+xveW8i+a+vIyKDGa8IixP+frTyEldA5wxbwk7121gaGyEjb7Kmec/kWqSEK3BF3eOxuKtxWCL52AxAZLUkAwH+kOJLMvo7usmJZD6WK/62tWaa7uqYIsx5qvKTRIgNTYFqFWm7eo4Ywz9vX0Ya8mybJeVZbXgqdZ0oLVKrnbsWtc4q42p9bk1Hfo9rPPWqDHUarx/4xhhYlOD1oq8NE3r+6gFfY3TTlt5Cyee8VDuHtnG857yzAnzZhtDNmss0TqiizzkWWdPWvknne34Y09o9xBERERERDpOR4ZsAAOd+9TFAFhiD2zpTSF6DJFgJ67Nla+KZehOuxnZMkLZe35zzx+J1mBiBuTVaXl7gYaAJ8R6NVR9ixETYt61tBiGywIn9y5h6VBg0/0PsH7Dg9yyZT0PecoTGSxBFU9mPJgAZryqqpa/5MFOhJCx88H12LEqWQyMDI9STQyeYl21humUreFV41TPxi6iIQSC9/XGDz6tTAi4am0Vak0Lav+ubZnN99XfO0Caps37btkaK9LGn1tzVVhtjbha2Fa7vLEyrLZmm4+RLIR8XLsJEidcbmyx9pkF65qq9RrHlCQJaZoRiozRWosrpnM6DCXrmgI57z3RGHxLp9Haa8EAaXUHH/3cFZy4bAVDm7fhqp6KDUQCnrxr7Pid8s62J13wFGJXLMagsE1yx6uSTURERERk2nVs0jSgSraOFm3gr77ySUrUGhs0dwKF2gL3ERssQ1g+9tfv4Fdrb8HM68ZiCK7hNWTG75OHPPmC+I1ba6CTxUBP6unfmbFz63YeePABNqejHPvYP6faWyI1Bo8pKt5281xipGcsUN26g51DOxjauZPByii2twtPbOpwuTdb6+3HAyrTtM7ZZBVuE2QBFywL583DmN1Xkk0WvO1xOmuDxuYDjWvDtVbOtYZ8uwrdWoPJxtvUHqc2ZdYYO6HybbIx1yolJ61+s4ZSTDj2qCNZUx3mZU95LiOJxXrIJvk5mlq/DWN46TeuBJuvESfS19vHYYeubPcwREREREQ6TseGbPM696l3PBfheW95FaPdYH3em9EQ8y6hDcFHviC9JdiEjQ8OMs/DP//wGkzweAMlbyeua9aScTRdHzKInhgyYsjor3iWjTk2rV7DPevuZ93OQe4fHSJ2d1PxWV6FZiwxWkI0hGiIeWzWsn/Yun49O7dvp4pnzKeM+ZTUgsmy+nOZNPTZxdTRxsq22nTTxsCpdt/JNAV2IcNhGegfaFrfbW82YMJjNVa6NWoMvJIkmdCgoPF+jeu37W6ttsY14Sa7rbWWrq6u/HZ7eTxap7m2BnM2Qqm3zBVf+gzHrjyOscFRXHBEmsdvjMFGyAwkWWSkFHjR5W/B1UripKMde8wJe70GoYiIiIiIHDwdmzT1YnB7vpnMQcEZeh99AiGmZBaYpFpsfFqgg65eLnn5GxnxO+mbP4D3ERMMWct+a/mGKYriWiujWvVWI3bTIPesuY/169ezYWSQpSccw1gp76LgimmnYTepSS3gGRkexca8KUIaPH0rllMp23pF3WRrlu1NdVgMgeAzfFolxolVbI37bRxT7bpaMNTb07Pbtdt2+dwaF/tv0dpwobZO3O72V7tt61pvexrH+HPO71PrYFqfphrHw8DJquL26vkWs4xHY4XDj1zBfUPbed1TL8C5Ekl0E6bSZhbKHjKXv1Yqxy5h/lGHTAh6pfNoqqiIiIiISHt0bMhmMPR37tPvWAZ43gffQDVm46EJRcWYsUQC0UZ88b80BB64bT0n9C/kiq9+lopJgQxiwEZPEiw2gg1gg6k3I0gieehFQ7hT5ESZNZgQ6QkJOzZtYcvQDjZVRtnuAuXlS6hESE0kjYEQMqzPsCHiin2aMD5VMcSMECP9PQN0lbsJxjJWKsEhCxktFkZrbOYAeffMxq117bimSrZYhHzWUSuRatyXMwYTI7b4vvV6TEqwjoU93fUD0DoltXZZaxjVOhW1NRirhVxAvdKstSKvprZGW/6zbq6Ga7xdYxDYuEZbbTM2yddeK5WJtYoywIQIPhR1hvlU4LCLbLT1ecUYwZO/Fl2+dt9YyPjYFz/LUUsPZ9vazVSrGbWZoEWehyUQbL5qoI2Bbmt56qcvbv6tbjr4l3wHO05ND0RERERE2qKjz7/ma15VxzAYSi6ha1Ev5tRDMTErQhDbNJWyFrfk0zsh+h7++qJXAlXs0r6iAUERXJGHcTEW/yrWHLPQVMkWY34bn+XBlfGBJE3pG66wY2gH2/wYD2Q7OeLhZ7AzBrJi2mljGNO6PljtlRt9IPjAQFcfLilTKZU49LQT2ZkUDSfNxIAq75qa7XHb27XcasdrsjXOkiQhDZ6uri6ctU1VbpOtTdZ6/8nWgGuajtrS4bP2/HY1DbYY7IQKs9p1jVNMW7U+3xghFD+T1sfc1fTcSavy6j/rWmMLcDFw9FEreHBkBx98/TshcbuvvDMQXMSHlFd86bKmfe+5Vk/mEuccq446pt3DEBERERHpSB0dsi3UhNGOETFkPuPFf3c5DkuCweUpWlMo4otKLhssxnbz4MbNnDrvED757S9gkzwcCTGCMdiWQKZp0f4idMn/XVwXIUZLKYVFlRKb1q3nwepO7IrFnPHUJ0BXF0SLC3lllIlgdjGtM+Y7JXqLIaHUN8BJf/ZwjjrlZMZsQurNeLOFLJ/uWdti0TG0Hto0bHGSaZR7Cr1awzUYX/esUqngQ4pzjmqaTTpltDFga9xX45poreHcrqZg7mpqadPPaBdVcbsKsFqf53g3VHZ5LHY1TXSPY7cRY/PXF9HzjisvpT84htdsxPvdx2UmGGyIDM03mJ7x46A/JXSWo45YRbnc1e5hiIiIiIh0pI4O2eZpXbaOYIsOom/6yhVUGcMGx5izeVjG+JTDPBhy2Jjfy9luXvus5zGvt4edSZXMNFYx5Y0ITDDEEIvN1NfryiusIj5AiBZfBGc+BMpdPTzhSeez5OTj6DnqMFaefALeJfiioinE5tAm+uYqNO89MUSCj/mzi4ali1Zy1/qtjJT6SE0Jm0GaZlSr46Fa9D7fGsKg1kqwxpCssRqrNRxqaozQUsFVY4wh2sBYpYK1Fp/mq9hZa3dZsVZf46xlPJNVsTVWre1J0/jjeCODxrHWnteutpY9No25Wq3Wf+77Mp7m29ampwaiCRAiq045jjvHNvGhV1+cH5taKNiw8FosKh8DlhRDV7C88BtX4jBF1aV0Ek0VFRERERFpn44O2QyG+Z19CDpCMBESGFmUEBKIJpAEC8YSMRA8tQXtYwj4aKhauOfOezhj5dF8/KtXEpJIiBkmGBIcNkRM5glYIAESYsyriQwBYyIh5OuzxcxiQkL0AVf1POHh57DmgU1sD4GFR65gLLHEmK8BRz2woylga53iWa16vIdS5hi8az19Pf084YlPplqJZGMQvcGlEVf1hMzXw780Tev7qoVvWQhUg8fHPLppDc0ma0DQGHKlPsPjyZqquCDPj0qMVgM9tkzJlQDIsixfP80GgvGkoUowvilgmzTEs0X41DDGWlOCxgq7yaaWNgVm9Wm+Htj1dNjdvKIweAw+H4+D7r5eop3YMbRmd/usTTsORVVlXvLoSZ3HkfHhv/kE/b3zyDaPYXzEmwxvsgn7tjFiMQTnIHpe+c1PKGHrQGp6ICIiIiLSPh2fMC3UIZjzTLS88isfZbTkMVnzz7sWauSBk8cEi7GRcqnMc574TBaVuvCLSniTB3G7CmTqIUq0xWJo410gIw5Piaya0NdzCCesOJkTTj6dgcXLqRiHJ99CNPgY8DHkC+eH/PvGCjbvPT6LVFJPmmYsWbCQJ531BLLBjNGtVdb8/j7W/2kDz33KCzhsyYp8FmjmmwKryaZ8EmmqzGo8PrsLoAKRaCOBSDC+Kczy3oMJeJNX84WQTxUtlUoTAqfdra9WCxbTNCVNU7IsawoNJ30+exj/3oVpe6923Ha3plvtduzmMWPDZmLAuMApJx/PT+76HZdc9DqSJIFoMHHXv7fyajjYWK6CZg12HIVsIiIiIiLt0/EJ0wIdgjnvYec/hrTPQCUlOlvvHBqjIQJZKKZQxrymKPOB7RtHeMIZj+Hz//T3/3979x0eRb2ocfw7M5tO7yi9CAgWpIhKtYEFRRE9VkRBPRZUFMV2BLyWAzb02BVBsSMqoiiigCDSwUbvikgNNWV3Z+b+MbsLIQECCUw2eT/Pk0uY3Zl5dxNybl5/hZ1ZGTiut8Nn9FzHCeM44RzlV+wabnT9MAjbDjt27mbd2n/YsjGdUxq34e/Vuxn/5c9s3baNbMfGtoOEnDC2Gybshgk7DrZjE3ZsHMfGdr3rZ2dnEwwGyc4OEg4Gydi1i7///pPnh77ChjVbydiWzVntz6aMlcpfv6zm2PLVIehgh+1YMXWg9cLgYKWUs8/nTmTKbe7zbdsmFAqRkZHBxs1bME0DJ1KKZWRk5NjEYd/7Rks+27G9NfJcx/vYZ+2z/a2DlmeRZjhgOJG9P/dM6fTu5ezzcWhyTeXda4pr9Ptl72vnKNL2GiHoGGDv9QE2rhtmZ+Z2xnw9luTEVLau34yJNzU554/vfabfmg7Jhsvl7w1Fc+JLjqpVqlGmTFm/Y4iIiIiIlFglvmFKwyTR7xBS+Iw9f55ybRdChouBgYu9Z+dPvFLMNiFsQNhwCbs2pmVRIakyKbbB8l2bIjuHGoRdiIzZ8j5cG4MQBiFMwuA6eJP/HGwA28UMO5ROTKZCcln+767/UqdUdcLuLi68+HQSywawrSAhgtiEvOIOGwjjuiFcwoTdELYTIhwOAja2HSYYzMYOZZORsZvTTz+DZic2Z1d2iEw7mz83/kly+RRCSQ6fffUFQTtE2A7ihIPY4SCOEwJsDCNSPJkuhmlgWt4aZYZpRKbOes/zypsQWDaO4eAYYVz2fODaYIcxHAdsB8cNRUos71zHDZO+dSsB1wUnhOOGwLCx3SCOE8Z1wziRD8N0MEwHDNu7trHnc9cIE0gwMUwH17Ih4OJaDmaigZFA7LzoR7SYc6N/Rss1bNxIqepGdom1jeiHg2043jeNgbcBAeHI++WVZG5ksqnDnrXToq/XsqxY6eet9ecQPctlzzpqjusStm3CkVIu+rnreFM+Ddf1vlNNSAqDayXQ87rr+Xvbbvr3vgc7aOC4Jq7j4LphbMP2Rgviet/3Bt7oQsPFcELc8tbQo/bPTvzVsL7WYxMRERER8VPA7wBFQTlMNh7GCBYpugwXTMOgeY+z2ZbsFUdZdhjDcDFxCdthDMPEBWzbxXZdHCdMdiiMm20x6KZ+fD1pNE7qbkKuQXYwiOu6BMNhHNcB18XCxHQNApb3zyhsR0oV1wHbJQGwQgl0Ofsc6lVvxvrVSxk/7hO+/OYDKhxbkbL1quO4LkkpydiOg+26BAKJ2LaTa2fNcCiMazvgegUQTha4Js89/QTlQ1Vocl4NJvw8mu3bNlOxTi2mzthApSpp4IQxHRvXgJSUFDANkhOSCJgmiYEETCycgBm5nzeVMzmQiAsEAgEs0yIhMZnExERc16BM6RRcxwUTLCOy+YNhgGvg4pVEALjeKC3bMalTphqmbXHaSa1ILJXqrT0WbUENg8hsVUzXBMPbqMIFQsHd3lpjhnc8MzMT14VgKExWdqY3+su28VolxyuyIruHJiYlEAqFvNLLdnBD0V1FXTAi6+tFpsjiWgSD4diGE7aVieM47N6dgQFkOCECponj2gQsC9txMMzIa7QMwo5DwAhgh8JYpgMGWJZBQmIiDg6BQAIGFpZhYpoWmAYJgURMw8LBwDK9TTgc1yv4rMgx04RsyyDRtPhi0lg2b9/JS1ffyc716aQeW4lQove94H3fWASdLO88w8B2DWzLJtG12FrWIFA2gfAOO7JGm37WFVdNGjX1O4KIiIiISIlmuIW1KFEc24DNEkJ+x/DVjq1bOZ1kKlas6HeUAjNMcDEwkhL5ZttKAokpBEPZe6bTAaYbIBgM4xrR3S4hHMxkzT/rCIRT6NiwAVNXLSJouGAY2HaIrFCYzMwQoWAI23FwgkEyQ1mkpqZ40wXDLru2ZxIOhzFCDmE7iy6dzuSXWT8y/qtxuK6FlRDESHYpXbUSYbcOuzN2k5iYRCgYwjVsXMfAwBsxBy5GZNRdyPQ+90Y5GRBIwLTCuG4mARusxCQSTAjaDiHbxTYMXNeITV90DbAsCyNg4hom4ZBDUmIyWBau64BpYlkWFkZsAX7DgEAgkYRAIo7rYGFimSaGaXqjpsAbwQbejp1A0HAx3b3W209IoF6pGjx2+cXc+Nr/EXJSIGDhOGAlJGAlJHjvpeNgOgBGbLMHwzRi1zFwcJyg99pdE4PIumSGt8SZ4YbYe6dYJ+Ct+xbdQdRxQjiuQzgcKcpsb0qraRg4hLxyz/CeaxpgYhEKOuCYWEYAy4QE05t3GdsAwoBw2MUyAiQlJpCUnERSUgJppUphmiZpaWmUSi5DamoSyckmZcqmkmQESElJImBZJCYESEpJxTQMTCsA2BgmmJaBaUKCG4iNmjQIEHbCVHCTKGclsjRjAyHT9kpUiBSITuS1Ot44QzdIODuIY7tkZ4e5tcEZ2K4T+f4qfq7tfZnfEXxlGAYvPv2GpotKXOtl1PA7goiIiEiBaCQb2vyguHEdbx70v55+kOFjP8IyE7zSxjBwbBfHtCMjpAIQW2A/jOua7Ny1mwcuv4Vl/6xk4oIfSXQScA2XsB3ENQ0MF0J2GByXpMRE7GAYFxfLtLw9Il0bGwfDcdm5NZ07+79NOStERnaILKB9+/YkpqURNCAjtItSDjhOhlfyWSYGXoHkmgFvmqrjFUcmXpliuoBhk2FnETACmFY5TMPAMMOYJgQjRZIdJjJCzNu51HUj0wdNhwCREVRWGNe0vSmqhgGGgQk51iuzzBCBQLb3d9OFsIFjmJFpjWDhlXFeDeaSbHtFj1fkeCMEw24atmWSlGriEgQLQo6NEUjENgxsx8YyA5iugWWY3u6quLhW0Pt6Rr6uJpH1y4gMI9urVDMcb+Ri9Nmmm51zgwYDTMvEIBA7zzASMA0Db8yhGxk56BA2A96sS8fb8TTgGhiG5d0rMpsUIt8/4I27s0NAkCzHIDO4Ccsy2bLNjl4ZCBAwErANN1LqOhiGiYETyW1gGiaGa+AGDEK2jRkpWR3HITU5Fcu1+HPtOl696yE+eWsUGTVTcW0DgySvMHRCWKblrWVnQMCBTCebxHCAUDiJi/77KJ/d/+iB9l2QOFavToMCF2w33HADI0eOJD09nTJlyuA4DuXLl+fss8/m008/zfX8yZMn06lTp/1e79RTT2XGjBkHPSc5OZmyZctSt25dWrduzZVXXkmbNm0OmveXX37h6aefZtKkSWzatIly5crRpEkTevbsSa9evQ54bjAY5OSTT2bRokVYlkU4HM71nBEjRhzwOldccQUffvjhQXO+++67XHfddQA89thjPPzwwwc9Z2+2bfPyyy8zYsQIFi9eTCAQ4KSTTuKuu+7i0ksvPaRr7e3zzz/n1VdfZe7cuezcuZPKlSvTsmVL7rnnHtq2bZvjuR9//DGPPvooK1eupH79+gwePJjLLsu72L7lllv48MMPWbJkCVWrVj3sfCIiIiLxSCUbkIhBKgYZxXJ8RwlheAWUV/GYpFRPwz0mjQQ7A9cxscBb88owsLDAcHHdIG5k5Bami2Enk75+GxUDAYaOeY+U6qVxLW+Eo2VEd6I0SbS8Be1DoQxcy7u3a4SILollOg6u6RIoY5EUsFnz2++kHVOVOqc0Z9OurZCxCwOTkGViuyaBxEQSEk0vlxmd0ueNaDOsyJRREzCM2OtLTEjAdC0s08AwLTAtwCXBdbFMh7AVJBSysSMbMLiGSXTTUNvN9qYZEsbAxIqUSi42puECVuS5Lrg2jh1ZtN8Gw7UiRWBkRFdspJhXSIUtO7YcHoAVNslyDK84TLAJO0FcyySMjYs3Ms0wI183y8QxDAJ4U1cxolNPI8VZJJL37piRUW2Rks3aZwMGx2TvIIbhxKalGoaB4Xoj+BzTwMIkOuXUABIc2zsSfcMMAwhFvsmIvEd7PR4LBwE7slZfZPqt4QYwLRPXCBEmEys6Wi5S2JqRXtA0jMj6bt5XPjESCSDBNbHtLLIIUK1OGWauWshdfXrx+BfvYhlhwuYOApZLZKAdAcv7wW663p4HZlI2Se4unPIpJFVIJXtLJtHqTz/xio/mJ7Uo8DVmz55NkyZNKFOmDACLFi1ix44dtG7d+oDn1a9fP1cpEz2+P1WrVqVLly4AhMNhtm7dyi+//MKMGTN44YUXOPfccxk5ciTVqlXL8/w333yTf//737iuS5s2bWjXrh0bNmxgwYIFvPfeewct2Z544gkWL158wOdEnXTSSZx88sm5jp966qkHPXfz5s3069cvNv3/UNm2Tbdu3Rg3bhylSpWibdu2OI7D9OnT6d69O48++igDBw48pGs6jkOfPn0YPnw4aWlptG3blnLlyrF27Vq+/vprWrRokePr+dNPP/Gvf/2LSpUqcf755zNt2jQuv/xypk2bxumnn57j2nPnzuWNN97g2WefVcEmIiIiJZJKtojymGR4y9VLnDHZa+9GA7AMuj/9IIRsggETM7zXLzaRzQm8zyOLzBsuiZgEk0xee/Rxtm/bTLnqFQhaNs4+a6PZtrdTJ4HI4C93r9EPRrQfc3EdBwuXuqc1xzq2LOFQiA27t1MhKYBlOriuiRG08GoQg8TEFAzLjF1nz5/eSDE3MtIMIiPMIqOfvP/rRioT7++m4Y0IMwJeKZcV9kbbxUZ9RRb2dwHbdUhwANPAcL17GK43qir6ml327EZqRNeEi76de+2k6e3i6eYobgzDYPvund4upI73mGGaGK6DZeXc9tKIvkbDje49sH/Rss0gxy+uxl7vUY6M+2Q2cHEMInNNc56T+76Rd9n1bhx9L3PuUBp5DwzHW2cucjy2Rl2uF7TneLRPy/Hrt7HnPTUiI/hMJ0hqoBSlqlQhISmVimaALUY23tzbnFcworeMTI8OGS6mY9Pj5UG8d0X/PbubEukp1bbFvZNPLFjJtnv3bhYtWkTPnj1jx2bOnAlw0JKtbdu2jBgx4pDu17hx4zzPmTp1Kn379mXChAl06tSJmTNnxkq/qB9++IGbbrqJ+vXrM3bsWJo0aRJ7LBgM8scffxzw3osWLeLJJ5+kT58+vP766wfN2q1bt0MusqLuvvtudu/ezTXXXMO77757yOc///zzjBs3jjp16jBx4sRYcbl48WLOOussBg0aROfOnTnttNPyfc3BgwczfPhwunbtyogRI6hQoULssfT0dDZv3pzj+U888QRJSUnMnj2b2rVrs3LlSpo0acITTzzBuHHjYs9zXZfbb7+dpk2bcttttx3yaxUREREpDjRPMkJTRuOXA7FRZLjQ93+PEUqyvdFk7sFHDxiug+GYrF64lOTdu3nmrWEYhoNluxh2GMMOE8AlwYAEA9xQEMIhiEwvynn96EgmFysxiZTUMtSr14TaDY6nZsOGpJSvQEKZUiSULkWgVCmSy5YiMS0JI2B5u3ta3geWCabplWumiRmwYh+GZXpDoAxwDG/BfDuy4H9kEiVgguFtaGBZkWtHCsNomeWa3kg0x4yUeKaFgUVicjIpaWmR++curPb+8Kba2ntKtr2eBxCybXbu3Ont1eB6+3t6C/x7z7EjO2w6kXXZHMfB2et6+17TcZxcX8+9n7dvvryOG5ESjMj7FXZcwo6L6xqRj9zX2LNraM7XFz2+5zET8Nb4cxyvu3Jch7Djgpv/nzFm5Ll7F3im5ZKZncEHEz7j/e/Hc3OPa3GyIlOBHWI5XDeya67r7YrrGt5mCgE3jGVBpwf7eCVtZCCgqYIt7lWsUIlaNesU6Brz5s3Dtu0co7NmzpyJaZq0aFHwUXL51a5dO3766SdOOOEEFi9enGe5dccdd2AYBmPGjMlRsAEkJibSvHnz/V7fdV1uuukmypUrx1NPPVXY8XP47rvvGDVqFA899BD16tU7rGu88sorADz++OM5RgY2btw49t4MGTIk39f766+/ePLJJ6lVqxYfffRRjoINoHz58jRs2DDHsQULFtChQwdq164NQL169ejYsSMLFizI8bwRI0YwY8YM/ve//xEI6L/hioiISMmkZimiHN6UQoljhsG5l15AdopBgu1gYuMabo6SaO+iKVo2mWYAx0rhpfv+DystmaRqFXANB9sNg2OD402ZtG1vGmBaWhqJiYmxRfVzTBuMlCxgQsAkMTmFMqXKUa50ecqWLk/Z0mUpV7oiZUuXJ7VMGVJKpZGUnIwVCEDAwrC8DytgYSUECCQmYCUEcE0DxyD24U3VjEyrNKMjyJzIKDDXK+tMbySbaRiYphn7AK+QDJhmbJRctGdx8EaCZGRk5CiX9i28Yq92r/Iu55fCiDxusC093buGbYPtYIfCuUqrPedEisKDFKP7lmb7Httf2QaRMipSsEV3FM1dluUWfe/2LthyBzNjH65jQGzzhgOOy8t5n/2Vca5F2MjCYAezVswnM+hS2Urz1t3L4/J7f01MXGzDJeQEOaZ5Y1KPreDNxo1NsZZ4dtKJpxT4GrNnzwbIVbI1atQo10iyIy01NZXnnnsOgNdff52srKzYYz/99BMLFy6kY8eOnHDCCYd87ddee41p06bxzDPPUL58+ULLvK+MjAxuueUWmjRpQv/+/Q/rGtu3b2fFihUAdOzYMdfj0bXtvv32W7Kzs/N1zZEjRxIMBundu7e343Q+pKen53qvypcvT3p6eo6sAwYM4KqrrqJ9+/b5uq6IiIhIcaT/1BhhYlABk036lTMuRafvnXTJWWxNysIyAl7B5LjeemSRhfljpYsR2YHUMHCMACt/X0npbokMHvYkTuUUb20uK4BrO7iOC0YQx8Fb/8ywIFJQGSGbsG3HyiZvcl+kyHJsDMcF0yRgBbzixfJqpHA4TGKs1nUia5Dt9b23T2tiYu7JbTs4pht5TZGCKTKGzXUif7pgOzYuLiFccBxsvMItYFqAA45LwDAwcDGNyEYJhoFDdE0ycF2v1fNWSvMKPvaqjQzTm/7pGi6uA67jsHd37wQyCLlhHDeEG3IIJJkYrotjO0Q6Ky+7YUamYzo40VmzhhWb0mi4YBLZLCDWbbnepg043k6je22RgLv3yLLI5gS4uK7jXde1I9cEjLC3ypzhelNY3ViwSFEViv3dBRzvq+Ft/sCedQAdw4593aKTZh3biWzW4GAalvc1c73323ANDHfPa49MYsU17Nj3NOBtpGEYGI4JtrdeXulyKXwxbRJ9Lu/N/707FJLScK3Imn443ss39ioGjQCOEcJwHVw3RJfH7uGTXg9BZDdY/dSLb81PbHnI5+xv04K81h6Lfi/Wrl2b1atX53p82bJlPPDAA2zZsoVKlSrRtm1bunTpEvv+OxxnnXUWlStXZtOmTcyePZt27doB3lRRgNNPP53MzEw+/PBD5s6di2VZtGjRgh49euy3PFq/fj0DBgzgrLPO4pprrsl3lrlz59K/f3927NhBtWrVOPPMM+nQocMBzxk4cCArV65kypQpJCYm5vtee9u9e3fs87wKwehu4JmZmSxdujRfpePe79/69et57733WL58OWXLlqVTp0507tw51380qVWrFkuXLs1xbMmSJdSqVSv290ceeYTMzEyGDh2a/xcoIiIiUgypZNtLZSyVbHHIG9Vlcf4VF7M10SYcWSHfW6fNW3PN9YZHYTouthlZkMz11tqybZP+/76DYPZuzEqpWKbjlTJGZNqm60bKJgjbjjcV1PDqFicAgYCFbYMTWdzKiKyXBt50TNt1MUwD0zWwMXBtB9zIwv2AY0TKtugC/4bpDYYiuhYcYEfqNCfydycygoxoOebVUXtPF3QdvI0ZHK/EsyBSpnnvSXTkGIS9KaZGZGU3w4ysCUbkTyc2xXLfX7683U8j1ZNh4JgurhuOLXVm2gFCBEnGwjBtbwdNZ981yCKL8JthoiO+YiPdjGhh6R2zAMM1cSybPcvDeV/PPQPLnL2mZnrZbWxM1ytdDcPFjBaBhhkpzgysyMgzb4eHPckMd88Y1+ird12wI3WqGx1Jh4Vh2JHvgUiZGylSDcMbNWcZVmQ9Pe9iNi6GEd2yNDLNN/qqokVrdF226CYPkcfnL/uNSzp0plJiGbYbIVwXLAdMLBwrUiriRL6GZuxMF5uU0olc8ewAPun3pH7ixbnkpGSaNG56yOdVr16dG2+8EfD+bb399tvUrVs3VrwtX76cKVOm0KVLF4499lgAKlWqlOe1pk+fzvTp03McO+GEE/j0009zTT08FCeddBITJ05k0aJFsZJt4cKFgDcitXnz5ixZsiTHOY888gjjxo3Ls3C6/fbbycrK4uWXXz6kHOPGjcux9tjgwYPp0KEDH330UZ6L+y9YsIDnnnuOXr16FWhUV4UKFbAsC9u2WbNmDY0bN87x+KpVq2Kfr1mzJl8lW/T9W7hwId27d2f79u2xx4YMGULHjh357LPPKFeuXOz4+eefz3PPPccbb7zBFVdcwfvvv8+CBQu45557APjtt994+eWXeeqppzjmmGMO+/WKiIiIFAcq2fZSPjJlVNsfxBsXx7VJPb0ZruGt+xWdcuiN0wphGiaOY3uDnPDWIwNwsFi/ciNNUivx349eJpwW3WHTxXW9wiQ6hMmMrFHm4BA0XGzXwTS9X4DMgFeaGXi7VuKC6ZiRRfIBw8CxQpi4hMMhTMMlbASxDG9UlGVYsbXKvBLIjRQzNo7txqYdWqZBOGx7hZhpeTt/ml6xYjtObIdK0wkQtsO4YReXbELhsPc6HC+zabJnaqTjjWYiUvbYjoHteLuchu0wZqQwcmwHy0yILZpvmiZmghUp3myvnCO6oL43bTV7dwaZO7LIBnZt3kK25eJaDrYZxon0mN6aYN6acKbpjfYKBAIkWHvWNAtg4rgQJjJd1jHBBStgYVgBnIA3gtCyTFxsjIAbG81mmKY3otAwCCcYmK6J7VqxssrEAtsmZHjrphmBgFfoGRau4+IYDk7ka43rjXr1SkeXPT1YtJFMxsD0Sk7XxTbDRL8zDMOMFcB7tj2IrKJnmuBER9dFRlg6e8o5AwPLsCKjC71rJFZIYtSP47jtqhsZPOoFjJQAXm3nvTdmtIR1XEzDwbEjIw9dl2wjhFOrPE4KBDItwvqpF7dOPKE5iQmHPlKqUaNGvPnmmwD8+uuvDB8+nN69e/PAAw8AcP/99zNlyhTeeOMNatSokec1ypYtS//+/enevXusTFuwYAEPPfQQM2bM4Nxzz2XBggWULVv2sF5btNTbe1pi9PMhQ4ZQrVo1vv76a8444wzWrl1Lv379+O677+jatSsLFy4kNTU1dt4XX3zBmDFjePTRRznuuOPydf/q1aszcOBALr74YurVq0dmZiazZs3ivvvuY8qUKVx44YXMmDEjxyYutm3Tp08fypYtW+BRXcnJybRq1YoZM2YwYsSIXGvIDR8+PPb5zp0783XN6PvXr18/TjvtNIYNG0aDBg2YNWsWffr0YfLkyfTp04dPPvkkds6AAQP45JNPuOmmm7jpppsAb3TbgAEDAK+8PO6447jzzjtj54RC3gjghISEw3jlIiIiIvFLJdteLE0ZjUuuC3PXLmf4t1+QvWsXaYYNGDhhGzscJtpWhLIyCYZDZGcHvQLEsQmHHG6/4mZSDJPda9azy87CMAxKVyxLAJNSZVKxzERv+mhCIqkpqZRKK0NSIIDhQlJKMkmJgch6Z97aZ4mJiViWRXJSEpZlkpyagmFAIGASCCTguN4vHpYdXcvMG/FmRHZvMCLT99y9Vgk0IoUheOWW47qY4P1y57g4eKPXHG+OI65jEMwOE7YdsB3csI3tOri2Q0ZGFgYuWdlZ3vvguriRaY122GbDP5sIhrz3CNPAdhxvSqi7Z82y6Os0nQBmwMC2Q1iWtWcNN8crh7bv2I1d2+H9D96lfc22uAELw/HWkktMSMS0vNFxgUCA1KTSpKSkYFh71s5LSEggIcEiISGBtMQkEhMSCBgmVkICgSSL5NRUjAQT23XAsDBNA8v0SqvY5FHDwXLxRhPi/TsHYkWsbbo4dhjbdSIllu2VjJF33LJMXCdSNLpOpIh3CYaCOIaF7QYJZgbJyMgiOxjEdsIEg0Gyg9lkZ2ZgOzbB7CC2HSbbdgiGQmRmZZKZmcnWbdsJhULeh+sQDocxHJdQKIRpRN57wyBgWQRMK/YLayAQIDk5iSVL/8Y58TSal67Jsr9XY5lhrIRErMjzot9frhv23h/DKxXDGSEcTK548F4+euTpI/nPU46w1i3yv6vk/kyaNAkgx/TRyZMnU69evf0WbADNmzfPtcnAmWeeybRp0+jUqRNTp07l5ZdfjhV3h2rfHXxhz38cCIfDfPrpp7GdT5s1a8aXX35JgwYNWLNmDe+99x59+vQBvAIqWgQdSpbOnTvTuXPn2N/LlClD165d6dSpEy1atGDOnDl8/PHHXHnllbHnDBs2jDlz5jB8+PDYdM6CGDBgAN26deOZZ56hUqVKXHPNNTiOw9tvv82rr75KIBAgHA7ne2pu9P0rX74848ePJy0tDfCm544dO5YTTzyR0aNHs3Tp0lgZWaVKFebPn88bb7zBqlWrqF+/Pr1796ZixYq8//77/Pjjj0ycOJGEhARWrVrFzTffzPfffx+77htvvBHbNEFERESkuFPJtg9NGY1PLWrWP/iT9uP7x98pxCQiR9e3j7zkdwTxSWJiIiedsP+dNPNr0qRJlCpVipYtvbXddu3axbx58w5p3bK9WZbF/fffz9SpU/n2228Pu2TbvHkzQI4dMEuVKgXA8ccfHyvYopKSkrjqqqsYMmQIU6ZMiZVsDz74IH/99RcTJ04kKSnpsLLsrVSpUvTt25fbb7+db7/9NlayrVmzhv/85z+0b9+e66+/vsD3Abj44ov573//y4MPPkj//v1zbKJw0003MX/+fGbPnp3vTRxKlSpFeno6PXr0iBVsUc2aNaNVq1bMmjWLH3/8MceIv0qVKuX6Ou7cuZP+/ftz2WWXcdZZZ+E4Dpdeeinr1q3jrbfewjCM2EjH2bNn51puQERERKQ4Usm2j/KR1ZpUs4mISFF2QtOTSUpKPuTzRowYQa9evXId33dq34gRIxgxYgQAHTp0YPLkyfm+R3T66Pr16w85H3ij2H755RfAK9SioiOi6tSpk+d50eMbN26MHfvyyy9JTk7mscce47HHHst1jm3bsd07n3/++Tw3f9hXXq9v0qRJ7N69m40bN+baVCK6YcRbb73FxIkTOfnkk3n++ecPeh+A++67j0suuYTRo0ezevVqypYtywUXXECHDh1iIw2bNs3funy1a9cmPT39gO/frFmzcrx/+zN48GC2bdvGs88+C8DEiRNZsGABI0aMoGfPnoD3dezVqxfff/89Z599dr4yioiIiMQzlWz7sDCoqCmjIiJSxLU6pc1hndegQYNYCbJx40bGjx9PmzZtaNSoEQA///wzS5cu5eqrryYQ8P7fhH0X3T+Y6Npf+46Wyq/vv/+ezZs3U7p0aVq0aBE7Hp2euvc6bXvbunUrsGfEW1RWVhZTpkzZ7/2ij23bti1f+Q70+hYvXszixYvzPG/16tV57tB6MA0bNsw1kmzt2rWsW7eOBg0axDanOJjmzZuzYMGCQ37/9rV48WKGDRvGwIEDqVmzZuwYQKtWrWLPi442XLhwoUo2ERERKRFUsuWhkqaMiohIEZYQSKD5SS0O/sQ8tG3blrZt2wLw2muvMX78eIYMGRLbwbNdu3ZkZWUxatSow8736aefAnDKKacc8rkZGRn069cPgFtuuSXHFM/zzz+fQCDAb7/9xtatW3NMJYU9Zdnea8UdqNQyDAPLsgiHw4eUMa/Xd/311+93mujAgQMZNGgQjz32GA8//PAh3Wt/XnzxRYDYZgT5cdFFF/H222/nWThGpwkDudba29cdd9xBnTp1uPfee3M9lpGREft89+7dAPleM05EREQk3un/68lDhciUURERkaKo6fEnkpKSevAnHsSPP/5IcnJybMRRVlYWs2fPpn379gc99/nnn+fPP//Mccx1XV577TWee+45DMPg3//+9yHlmTZtGmeccQa//fYbTZs25ZFHHsnxeKVKlejVqxe7du2ib9++BIPB2GMjR47ku+++Izk5uVDWRHvyySdj68JFhUIhBg0axCeffEJKSkqe024L0+7du1m0aFGu49H3uFGjRvTt2zfX42eddRaNGzdm1qxZOY537dqVJk2aMH36dF5++eXYcdu26devH1u3bqVZs2axEjYvo0ePZuLEiQwbNozExD0720anrL733nuxY9HP8zudVURERCTeaSRbHjRlVEREirLC2FUUYOrUqbRu3To2WmzmzJlkZ2fnu2S79957OeWUU6hbty5ZWVn89ttvrFq1CtM0eeGFF3JM9dzb4sWLY0VYOBwmPT2dX375hXXr1gHQpUsXRowYQenSpXOdO3ToUGbMmMF7773H1KlTadmyJWvXrmXOnDlYlsXrr78em8JYEA8++CCDBg2iZcuW1KxZkx07drBgwQL+/vtvkpOTGTVqVL6naR6uTZs2cfzxx9O0aVMaNmxIQkICc+fOZeXKldSpU4fx48fnuZnDihUrWLNmTY5RZeBtSvH+++/ToUMHbrvtNl5//XUaNGjA/PnzWblyZWzH0P1tUpCRkcE999zDxRdfzHnnnZfjsTPPPJNWrVrx/PPPs2DBAgzDYNKkSbRp0ybXGnUiIiIixZVKtv3QlFERESmKUlNSadXi8NZj29uqVav4888/Y+uzgTeyDchXyXbPPfcwYcIE/vjjDxYuXEgoFKJ69epcc8019O3bN8faXPvasGEDI0eOBLxdQcuWLUu9evXo3r07V155JW3a7P/1lS1blp9//pknnniCjz/+mHHjxlGqVCkuvPBCBgwYwBlnnJHft+CA/vOf//Dzzz+zZMkS5s2bh+u61KhRg5tvvpm77747tobdkVShQgVuueUWfvzxR77//nts26Zu3bo8+uij3HvvvQddOy0vJ598MgsWLGDgwIFMmDCBhQsXUrVqVXr37s3DDz8c21wiL0888QQbN27kueeey/WYYRh8/vnn9O3bl2+//RaA7t27x6a1ioiIiJQEhuu6rt8hiiIblxlkY/sd5CjZsXUrp5NMxYoV/Y4iInLIru19md8RjpqzOnam59W9/Y4hUuh6GTX8jiAiIiJSIFp6bD8sDCpj+R1DREQkh04dzvE7goiIiIiI5EEl2wFUV8kmIiJFSL26DahVY//T+URERERExD8q2Q6gNCalyHvxXxERkaOtY7uz/Y4gIiIiIiL7oZLtIKppNJuIiBQByckpnNa6cBb1FxERERGRwqeS7SCqYKlmExER353Wui1JScl+xxARERERkf1QyXYQAW2AICIiRYA2PBARERERKdpUsuWDNkAQERE/1aldjzq16vodQ0REREREDiDgd4B4EN0AYReu31GOqPT0dLZs2eJ3DBGRfEtPT+fBgf38jnHEdWqvDQ9ERERERIo6lWz5VA2L5YT9jnHElC5fnkUAZBXK9cphcjyJhXItESletqZv4bH/PkxWZuH8vElMSiiU6xRVyUnJnNa6rd8xRERERETkIFSy5VMVLFYRxvY7yBFiGAZlKlQotOs5QJgEqmiqrYjs482RL+G6DknJKuLz49TWZ5CcnOJ3DBEREREROQityZZP2gDh0K0gRHYxn2IrIofmh8kT+H3hr37HiCud2mmqqIiIiIhIPFDJdgi0AcKhCQFLCOGqaBMRYNOmDXw4+l2/Y8SV2rXqUq9uA79jiIiIiIhIPqhkOwSlMSmN4XeMuLINh3XFdpKtiOSX4ziMunkAx5ar6neUuNKx3Vl+RxARERERkXxSyXaIamoZu0O2ijC7cPyOISI++vqpl/j1s+9Y/fiHHBcqR4rWGDuo1JRUTj+1nd8xREREREQkn1SyHaKKmKRqNNshcYHFhLA1bVSkRFo15xc+H/hs7O8rRo4ndcJyalet4WOqou/sM88jJSXV7xgiIiIiIpJPKtkOkYGh0WyHIQOXlYT9jiEiR1n27gxev/oO7FAox/FtS9fy1xOjaZxUjYSEBJ/SFV3JScl0PvsCv2OIiIiIiMghUMl2GKpgkuR3iDi0HpstWp9NpER5u3d//lm6Mu8HXZdlL31GxdmbqV6p2tENVsSd1bEzpUuV9juGiIiIiIgcApVsh8HAoIZGsx2WpYTI1rRRkRJhwrA3mfnhFwd93qY5C9n87Fc0KVsT09T/LCUmJnHeuV39jiEiIiIiIodIv80cpmpYaILToQsBiwjiqGgTKdaWTpvFx/f+X76f7wSDLH3mY2qsCFKxXMUjmKzo69T+bMqUKet3DBEREREROUQq2Q6ThcGxGs12WHbgskLrs4kUW9vWb+DlHrdghw/93/m672eT8fpkGlWufQSSFX0JgQTO73yR3zFEREREROQwqGQrgGOwsPwOEafWY/OP1mcTKXbscJiXL/832//ZeNjXCO7YxfInP6T+1iTKlCpTiOmKvnZtO1G+XAW/Y4iIiIiIyGFQyVYAAQyqq2Y7bHO+nsiqOb/4HUNECtFH9z7GsmmzCuVaq8dMxv14HvWrloxRbZZlcWGXbn7HEBERERGRw6SSrYCOJYDhd4g4tPqJVxnTtQ8vXHwD29Zv8DuOiBSCGR98znfD3irUa2b8vYnVj3/IceHyJCenFOq1i5q2p3WgUsXKfscQEREREZHDpJKtgJIwqKrRbPlm785g7uV3Mumhp3Edh21/b+DFbjcSysryO5qIFMCKmfN4u3f/I3f9EV+TNmE5tarWOGL38JNpmlx43iV+xxARERERkQJQyVYIamJpNFs+BFf9xcTTr+DXT77KcXzlrAW83ec+n1KJSEFtXLGaYV17EczIPKL32bZ0LeueGE3jpGoEAsVrf+fTWrelapVqfscQEREREZECUMlWCFIwtTbbQez4/mc+b3UJf/+6KM/Hfx41hk8fHnKUU4lIQe3cvJVnulzDzk1bjs4NXZdlL31GpTmbqV6peJRShmFw0QWX+h1DREREREQKSCVbIalNQDXbfvz13Ag+63w9u7ekH/B54x5/gYkvDj9KqUSkoIKZmTx/YU82Ll991O+9ac5CNj/7FY3L1sQ04/t/ylq3PI3q1Y71O4aIiIiIiBRQfP9mUoQkYFCTgN8xihQnK5sF193Hd/3+D8e283XO+3cNZOZHY49wMhEpKMdxePXK21k5c75/GYJBlj3zMTVWBKlYrqJvOQrCMAwuvqC73zFERERERKQQqGQrRMdikeh3iCIi+Nd6JrW7kvnvjjmk81zH4c3r7uKPiVOPUDIRKQzv9X2E+V9863cMANZ9P5vMN6fQqHJtv6McshbNW1Pj2Fp+xxARERERkUKgkq0QWRjU0Wg2dk6by5ctL2XtnF8P6/xwMMj/LunN6rmHd76IHFnjh77CDy+N9DtGDtnbdrL8yQ+pvzWJ0qXK+B0nXwKBAFd0v9rvGCIiIiIiUkhUshWyqlikluC9Rte/9hGfnXkNOzZsKtB1snbt5tnzruWfZSsLKZmIFIYfXnmHj+973O8Y+7V6zGT4eB71qxb9UW2dz76AqlWq+x1DREREREQKiUq2QmZgULcEjmZzgkF+v/kRvrnlIexQqFCuuXPTFp4592rS160vlOuJSMFMfn0Uo257yO8YB5Xx9yZWP/4hx4XLkZyc4necPJUvV15rsYmIiIiIFDMq2Y6AiliULUFvbfifTfx45nXMfv2DQr/25tV/8lTHHmz58+9Cv7aI5N+UN9/nnVsewHVdv6Pk24oR40n7bgW1qtTwO0oul3e/psgWgCIiIiIicnhKThN0lNUrIaPZMmb/xpctL2HVT3OO2D02Ll/NfztcxuY1fx2xe4jI/k19+yNG3nR/XBVsUduWrGHdk6NpnFSNQCDB7zgANKzfiDPatPc7hoiIiIiIFDKVbEdIaUwqF/O3d+M7nzOm/b/Ytu6fI36vTavW8lT77mxcueaI30tE9pg28hPe7t0/Lgu2GNdl2UufUWnuZqpXquZrFMMwuPbKG3zNICIiIiIiR0bxboF8VodAsdwCwQ2HWXzX43zV815CWdlH7b5b1q7jqQ6XaTMEkaNk+rujGX7DPbiO43eUQrFp9kK2PP81jcvVxDD8+encoe2Z1Kldz5d7i4iIiIjIkaWS7QhKwaQGlt8xClV4Szo/db6Bn4e97cv90/9az3879mD9khW+3F+kpJg24mPevL5fsSnYouysbJY9/TE1V4WoWK7iUb13amoaPS656qjeU0REREREjh6VbEdYLQIkF5PxbJm/LObrlpey7IfpvubY9vcGnupwGesWLvU1h0hxNe7J//FWr+JXsO1t3cTZZL45hUaVax+1e1560eWULl3mqN1PRERERESOLpVsR5iFQYNisAnClo/HM+b0HmxZ/affUQDYsWETT7XvzrLpR27DBZGSxnEc3uv7CJ8++JTfUY6K7G07Wf7kh9RPT6J0qSNbftU4tiZndex8RO8hIiIiIiL+Usl2FFTAittNEFzHYdkDzzD2ijsIZmT6HSeHXVvSGXrWFcwePc7vKCJxL5SdzatX3sbEF/2ZCu6n1Z9Oho/nUb/qkRvVds2/bsCyitfyASIiIiIiklN8Nj9xqD4JcTeezd6+k5ldb2baU6/4HWW/QlnZvHL5v/nmmdf8jiIStzJ37OTZ865l9sdf+h3FNxl/b2L14x9yXLg8yckphXrtVi3acHzjZoV6TRERERERKXpUsh0liRjUjaOaLWvRCr5p3Z1FX0/yO8pBua7LR/c+xqg7HsYpxmtIiRwJ29Zv4Mn23Vk8yd+1FouKFSO+Ju27FdSqUqNQrpeYmMiVPa4rlGuJiIiIiEjRppLtKKqGRZk42AQhfewPfN7mMjYuXel3lEPy/f9G8OIlvckuYtNaRYqqdQuX8vjp3fjzl4V+RylSti1Zw7qnPqVxcnUCgYQCXeuCzt2oVLFyISUTEREREZGiTCXbUWRg0JCEIluzua7Lqsde4otuN5O5Y6ffcQ7LgrET+G/Hy9ixcbPfUUSKtDljvub/Tu3K5iKymUmR4zgs+98YKs3dTPVK1Q7rElUqV+WCLhcVcjARERERESmqVLIdZWmY1KDoLX7t7NrNnMvuYPJ/nsN1Xb/jFMiq2b8wuNUFrJy9wO8oIkWO4zh8+tB/efmym8natdvvOEXeptkL2fL81zQpVxPDyP9/IjEMgz7X30ZiYtIRTCciIiIiIkWJSjYf1CZAchEaz5a9Yi0T2vTg9zHf+B2l0GxZu44n23Vn0muj/I4iUmRkbN/BsK7XM+6JF+O+TD+a7Kxslj79MTVXhahYrmK+zjn3rPNpdFyTI5xMRERERESKEpVsPjAxaFhENkHYPuEnPm91Cev/WOp3lEIXzs7mnVsG8Ob1dxPM1DptUrKtW7iUwa0u4Nevf/A7StxaN3E2mW9O4bjKtQ/4vGpVq9PjkiuPUioRERERESkqVLL5pDwWVX1++/98+i0+P/8GMtK3+5rjSPtp5Cf8X5uL2LB8ld9RRHwx97Px3r+BZfo3UFDZ23ay4skPqZ+eTOlSZXI9bhgGfXppmqiIiIiISEmkks1H9UnwZdqok5nF/KvvYWL/J3Fs+6jf3w9//rqIQS0vYP7YCX5HETlqwsEgH/V/jJe630TWzl1+xylWVn86CeOT+dSvmnNU23nndqVh/UY+pRIRERERET8Zrhbm8dV2HH4heNTuF1z7N1MuuZW/5v1+1O5ZlBiGwXn3/ZtLHutPICHB7zgiR8zfi5bx2lW3s3bBH35HKfYa9LqANWmZVChfgcceGUqCfraIHJZeRg2/I4iIiIgUiEq2ImA1IdZy5EeU7Zwyi2973MHOTVuO+L2KutqnnECfd4dx7PHH+R1FpNB9/9IIPu7/fwQzs/yOUmJUaFqf28e+Rd16DfyOIhK3VLKJiIhIvNN00SKgNgHKHOFpo+v+N4rPzr5OBVvEmnm/MajFeXzzzGs4juN3HJFCsWPjZp6/sCejbn9YBdtR1uFfl6pgExEREREp4TSSrYjIxGEewUIfz+YEg/xx6yDmvPVRIV+5+GjUoQ29RzxHpTo1/Y4icth+Hf8Db/W6hx0bNvkdpcRpeEYrBkwZjWlZfkcRiWsaySYiIiLxTiVbEbIBmyWECu16ofUbmXbp7ayeMa/QrllcJZcuxZXPPUr7G6/0O4rIIcncuYvRA57kh5dH+h2lREopU5rBv0xQSS9SCFSyiYiISLzTdNEipCoWlQvpS7J75i982fISFWz5lLVzF2/37s/zXa9n+z8b/Y4jki9zxnzNQ006qmDz0TX/e0wFm4iIiIiIACrZipyGJJBUwGv8M/xTxnS4iu1/byiUTCXJL+Mm8kDjjkx8cTiOfeQ3oxA5HFv+/JthF9/AS91vIn3dP37HKbFaX3ERp197md8xRERERESkiNB00SJoOw6/EDzk89xwmEV3PcHMl945AqlKnlrNm3HtS4/T4LQWfkcRAcCxbb57YTif/+dpsnbt9jtOiVatUX3+M2scKWVK+x1FpNjQdFERERGJdyrZiqhVhPjzELZBCG/awvQed7JiyowjmKrkMQyDtr0up8d/H6J0pQp+x5ESbPXcXxlx0/2smfeb31FKvKS0VB6ZNY5jjz/O7ygixYpKNhEREYl3KtmKKBeXXwiyg4N/eTLnL2RCt3+zde26o5CsZEqrUI7uTwygQ5+rME3NspajZ9eWdD4f+AyTXnlXU5iLiFs+fJlTr7jI7xgixY5KNhEREYl3KtmKsGxc5pN9wImjmz4Yx4QbBxDMzDpquUqyuq1O4srnB9Hw9JZ+R5FiLhwM8t0Lwxn3+ItkbNvudxyJOPeu3lz53EC/Y4gUSyrZREREJN6pZCvituPwK8Fc49lcx2HZgGf4aehrvuQq6U7ueg6XPn4fNU9o4ncUKWZc12X2J+MY/cBTbFq5xu84spdG7U+l//cfYQUCfkcRKZZUsomIiEi8U8kWB9YRZgXh2N/t9O3MvLIfS76d4mMqMUyTU6+8mEsG30uVerX9jiPFwMLvp/HJgCdZPecXv6PIPsodU5WB876hbNXKfkcRKbZUsomIiEi8U8kWJxYTZCMOmX8s4/tu/2bT8tV+R5IIKyGB9r2v5KJH7qRc9ap+x5E4tGrOL3z64H/547sf/Y4iebASErh/0sc0PKOV31FEijWVbCIiIhLvVLLFCRuXSRMnM/qSW8jetdvvOJKHxNQUzr6jF+fffytp5cv5HUfiwOLJ0xn3xP9UrhVxVw0bxDl9b/Q7hkixp5JNRERE4p1KtjiyZe06Bre6gB0bN/sdRQ4guVQaHW66ms79+lD+2Op+x5EixnVdfhk3kXFP/o8VP8/1O44cRJurunHze//zO4ZIiaCSTUREROKdSrY4s+yn2Qw58wrCwQPtOSpFQSAxkTZXX8J5/W/hmCYN/Y4jPnNsm5kfjeXrp17ir98W+x1H8qHGCY15eMaXJKWm+B1FpERQySYiIiLxTiVbHPpx+Ie8feO9fseQfDIMg2ZdOtK53000Pbud33HkKMvOyGT6u58yfsgr2i00jpSuXJGHf/6CKvXr+B1FpMRQySYiIiLxTiVbnHr/rkf5bthbfseQQ1TzxCacc1dvWl9xkUbHFHN//b6YSa++y8+jPiNz+w6/48ghSExN4f5JH1OvdXO/o4iUKCrZREREJN6pZItTjm3z7PnX8ceEKX5HkcOQUqY0rf91Ee1v/Jd+kS9GQllZzP5kHJNeHcXy6XP8jiOHwbQsbh/zBs0vOtfvKCIljko2ERERiXcq2eJY5s5dPNXhMtbO/93vKFIAxzZrRPsbr+T0a7tTqmJ5v+PIYVi/ZAWTXxvFTyM/YffWbX7HkQK49qXHOfPWnn7HECmRVLKJiIhIvFPJFue2b9jEE2d0Y+MKrfUU7wKJiTS/+Fza3fgvmp7THtM0/Y4kB7D9n43M/mQcsz4ex7Jps/yOI4Xg/PtvpcdTD/odQ6TEUskmIiIi8U4lWzGwccVqHj/jEnZs2OR3FCkkZatV4ZRunWlx6Xk07nQ6ViDgdyQBdmzawpzRXzH74y9Z8uNMXMfxO5IUkjZXdeOmUS9iGIbfUURKLJVsIiIiEu9UshUTaxf8wVMdLiNzx06/o0ghS6tQjpO7nkOLS8+j2bntSUhO9jtSibJrazpzx4xn1kdfsnjSdBzb9juSFLLGnU7nnm9GEUhM9DuKSImmkk1ERETinUq2YmTRpJ949rzrCGdn+x1FjpDkUmmccP6ZtOx+Pk3PaUda+XJ+Ryp2XNdl9dxf+WPCj/z+7WSW/zwPOxTyO5YcIcc2bcSDP31GatkyfkcRKfFUsomIiEi8U8lWzMwePY5XrrhV09hKAMM0qd28GU3OPJ0mZ57Bce1OJSkt1e9YcSn9738ipdoU/pg4lV2bt/odSY6CcsdU5eEZX1Kx5jF+RxERVLKJiIhI/FPJVgz98Mo7vHurFu8uaayEBOqd2jxWujU4rYWmv+3H1r/+ZuWsBSz/aQ6/T5jCut+X+B1JjrLk0qV4YOoYap10vN9RRCRCJZuIiIjEO5VsxdRnjz7N2MHP+x1DfJSYkkzNk46ndosTqH3KCdRpcSLHNj2uxG2ikLVrN6vn/MKKmfNZGfnY9vcGv2OJj5LSUrn763do1L6N31FEZC8q2URERCTeqWQrxkbeMoDJr43yO4YUIQnJSdQ4oTF1WpwYK9+qHVeP5FJpfkcrMMe22bRqLf8sWcn6xcv5e+FSVs5awN8Ll2n6tMQkpqZw99fv0LjDaX5HEZF9qGQTERGReKeSrRhzHIdX/3Ursz8Z53cUKeLKVK1Mlfq1qdKgTo4/qzaoS6mK5f2OFxMOhdixYRPp6/6JlWn/LFnB+sUr2Lh8NeFg0O+IUoQlpiRz57iRHH/mGX5HEZE8qGQTERGReKeSrZizw2Fev6Yvsz4a63cUiVMpZctQtlplSleqQKlKFShdOfJnpQqUrlyRUpXKU7pyRZJSU7ASEwgkJHh/JiZiJQQIJCZgJSTEpqmGQyHC2UHC2UFC2dlk78oga9dusnftJmvXbjK372T7+o1s/2cj29ZvZPs/m9j+z0a2r9/I7q3b0I8sORwJyUnc+eUImp7dzu8oIrIfKtlEREQk3qlkKwEc2+aNnncx473P/I4iJZhhGAAqyeSoS0hOou8Xw2l2bge/o4jIAahkExERkXhn+h1AjjzTsujzzjDaXn+531GkBHNdVwWbHHWBpCTu+PwtFWwiIiIiInLEqWQrIUzT5Ibhz9Chz1V+RxEROSoCiYncPuZ1Tujc0e8oIiIiIiJSAqhkK0EMw6Dna//lzFt7+h1FROSICiQmctunr3PS+Wf5HUVEREREREoIlWwljGEYXPvS45zd9wa/o4iIHBFWQgK3fvIqJ194tt9RRERERESkBFHJVkJdPWwwne+52e8YIiKFKjElmds/fZ3mF53rdxQRERERESlhAn4HEP/86+lHCCQm8NWT//M7iohIgaWVL8udX46g4Rmt/I4iIiIiIiIlkEq2Eu6yJwYQSEzgi0HP+R1FROSwVah5DP2+GcWxxx/ndxQRERERESmhNF1U6DbwHq579SlMy/I7iojIITu2aSMemv65CjYREREREfGVSjYBoNPN19B37HCSS6X5HUVEJN+Oa3cqD0z9lAo1jvE7ioiIiIiIlHAq2STmpPPP4v4poylXvarfUUREDuqUbp25d8J7pJUv53cUERERERERlWySU51TTuDhGWM5RtOuRKQI63jzNdw2+nUSkpP9jiIiIiIiIgKoZJM8VKx1LA/99BmNO53udxQRkVwuHtiPnlpHUkREREREihiVbJKn1HJlueebUbS5+hK/o4iIAGBaFj1fe4puj/bzO4qIiIiIiEguKtlkvwKJidw86kUufKiv31FEpIRLK1+Wu8aNoONN1/gdRUREREREJE8q2eSguv/ffVz/xhCsQMDvKCJSAh3brBH/mf0VJ3Tp5HcUERERERGR/VLJJvnSofdV3P31O5SqWN7vKCJSgrS87AIemfElVerX8TuKiIiIiIjIAalkk3xrek57Bs77hrqtTvI7iogUc4Zp0v2JAdz2yWskpaX6HUdEREREROSgVLLJIalY61genPYZnW651u8oIlJMpZYry91fjeTCB273O4qIiIiIiEi+Ga7run6HkPj00zujeeeWAQQzs/yOIiLFxLFNG3HH529StUFdv6OIyFHWy6jhdwQRERGRAtFINjlsZ1x3GQ/PGEuVBnX8jiIixUDL7ufz8IyxKthERERERCQuqWSTAql54vE8Oudrml/c2e8oIhKnDNOk++P3c9vo10kuleZ3HBERERERkcOi6aJSKFzX5eshLzPmoSE4tu13HBGJE2WrVaH3yOdodm4Hv6OIiM80XVRERETinUo2KVSLJv3Eq/+6jR0bN/sdRUSKuOYXd6bXm0MpXamC31FEpAhQySYiIiLxTtNFpVA16XQGA+d/w/Fnt/M7iogUUYmpKfR87Sn6fv6WCjYRERERESk2NJJNjgjXdfn+pRF8cv8TBDMy/Y4jIkVEnRYnctN7L1K9UX2/o4hIEaORbCIiIhLvVLLJEfXP0pW8cd2drJw53+8oIuIjwzQ5775/c8ngewkkJPgdR0SKIJVsIiIiEu9UsskR59g24578H2MHP48dCvkdR0SOsgo1j6HPu8No3OE0v6OISBGmkk1ERETinUo2OWrWzP+dN667k3W/L/E7iogcJa2vuIierz5JarmyfkcRkSJOJZuIiIjEO218IEdN7ebNeHTO13S592YMU996IsVZSpnS9B75PP/+8GUVbCIiIiIiUiJoJJv4YunUmbzZ8242rVrrdxQRKWStelzIVcMGUa56Vb+jiEgc0Ug2ERERiXcq2cQ3Wbt28+E9g/nxjffRt6FI/KtYuwbXvvQ4J11wlt9RRCQOqWQTERGReKeSTXy3/Oe5vHvrg6xd8IffUUTkMFiBAOfcdSPdBt1LUmqK33FEJE6pZBMREZF4p5JNigTHtvnh5ZGMeeRpMrfv8DuOiORTvdYn0/P1IdQ66Xi/o4hInFPJJiIiIvFOJZsUKds3bOLj/v/H9Hc/9TuKiBxASpnSXPr4fZx5a09MbWQiIoVAJZuIiIjEO5VsUiQtnTqTd259kHW/L/E7iojso2X387nqhcGUP6aa31FEpBhRySYiIiLxTiWbFFl2OMzEF4bz+cBnydq5y+84IiVepTo1ufrFxzj5wrP9jiIixZBKNhEREYl3KtmkyEv/+x8+7DeYWR+N9TuKSImUUqY0Fzx4O+fe1ZuEpCS/44hIMaWSTUREROKdSjaJGwt/+ImP7n2MtfN/9zuKSIlgBQJ0uOlqLh7YjzKVK/odR0SKOZVsIiIiEu9UsklccV2X2Z+M47NHhvLP0pV+xxEptk664CwuH/owxzRp6HcUESkhVLKJiIhIvFPJJnHJDoeZ9vZHfDHoOdLX/eN3HJFio96pzenx1AM07ni631FEpIRRySYiIiLxTiWbxLVQVhbfvzSSr578H7u2pPsdRyRuVW/cgEsfv4+Wl57vdxQRKaFUsomIiEi8U8kmxULmjp1888xrTHj2DbJ27fY7jkjcKF+jOhc/ejftel2BaVl+xxGREkwlm4iIiMQ7lWxSrOzYtIVxj7/ApFdHEc7O9juOSJFVuW4tuvS/hXa9LichOdnvOCIiKtlEREQk7qlkk2Jpy9p1jH3seaa/8ynhYNDvOCJFxrFNG3H+gFs59V8XYwUCfscREYlRySYiIiLxTiWbFGvp69bz7XNvMuX198jaucvvOCK+qXdqcy544HaaX3QuhmH4HUdEJBeVbCIiIhLvVLJJibA7fRs/vPwO3w17i52btvgdR+SoOf7sdlzwwO0cf+YZfkcRETkglWwiIiIS71SySYkSzMxk2ohPmPDcG2xYtsrvOCJHhGEYNO/WmQsfvIO6LU/yO46ISL6oZBMREZF4p5JNSiTXdfll3ES+ffZ1Fk/+2e84IoUiITmJU/91Mefd92+OadLQ7zgiIodEJZuIiIjEO5VsUuKtmf87E557g5kfjsUOhfyOI3LIqjduQMebr+GMnpeRVr6c33FERA6LSjYRERGJdyrZRCJ2bNzMT++MZupbH7J+8XK/44gcUCAxkRaXnkfHm6+mccfT/Y4jIlJgKtlEREQk3qlkE8nDsp9m8+NbHzL74y/J3p3hdxyRmCr1a9Phpqtp2+sKylSu6HccEZFCo5JNRERE4p1KNpEDyNy5i1kfjWXqWx+yYsY8v+NICWUFApx80Tl0vPkamp7THsMw/I4kIlLoVLKJiIhIvFPJJpJP6/5YwtThHzH93U/ZuWmL33GkBDimSUNOvaob7W/8F+WqV/U7jojIEaWSTUREROKdSjaRQxQOhVgwdgJTh3/EH99N1WYJUqiqNqxL6ysuotXlF1LzhCZ+xxEROWpUsomIiEi8U8kmUgAZ27az4MvvmDtmPL9/O4VgZpbfkSQOVa5bi1aXX0jrKy6idvNmfscREfGFSjYRERGJdyrZRApJ9u4Mfh3/A3PHjOfXr34gc8dOvyNJEVah5jG0urwrra/oSr1WJ/sdR0TEdyrZREREJN6pZBM5AkLZ2SycOI25Y75m/tjv2LV5q9+RpAioWLsGp3TrTOsrLqJ+m1O0gYGIyF5UsomIiEi8U8kmcoQ5ts3iKTOYO+Zrfv3qBzav/tPvSHKUJJdKo1HH02jWuQPNzu1AtePq+R1JRKTIUskmIiIi8U4lm8hRtmnVWhb98BOLfviJxZN+Ztv6DX5HkkJiGAa1mjeLlGrtaXB6SwKJiX7HEhGJCyrZREREJN6pZBPx2bqFS2Ol25LJP7M7fbvfkeQQlKtelabntvc+zmlPmcoV/Y4kIhKXVLKJiIhIvFPJJlKEOI7D2vm/x0q3pVNnkb07w+9YEmGYJscc35B6rU+m3qnNqX9aC2qe0MTvWCIixYJKNhEREYl3KtlEijA7HObvhUtZPfc31sz7jdVzf+PPBX8QzMzyO1qJUK56Veqd6hVq9U5tTp2WJ5FSupTfsUREiiWVbCIiIhLvVLKJxBnHtvl70TLWzPud1XN/Zc3c31i74A+NeCugpLRU6rQ4IVao1Tu1ORVqHON3LBGREkMlm4iIiMQ7lWwixYDjOPyzZAWr5/7K2gUL2bBsFZtWrGHTyjUa9bYXwzAod2w1qjduQLVG9ajeuAHVG9WnWuMGVKhRHcMw/I4oIlJiqWQTERGReKeSTaQYc12XbX//w4blq9m4Yg2bVqxhw/LVsT8zt+/wO+IRkZSWSpUGdXKUadUa1ad6o/okpaX6HU9ERPKgkk1ERETinUo2kRJs15Z0Nq5YzaZVf7Jz42Z2bt7Kzk1b2bV5a+TzLezanM6uzVuxw2FfsxqmSenKFSlbrTLlqlehbLUqlK1ehbLVKlO2WhXKVa9CmWqVKVe9Ksml0nzNKiIih04lm4iIiMQ7lWwiki8Z27azY6/SbdfWbYSzswkHQ9ihMOFgEDsUxg4GCYfC2MEQdigUedz70wnbWIkJBBITCCQlkpCURCAp0fs8OYnkUmkklUojuVRq5M80kkunUbpyRcpUqYRpWX6/DSIicoSoZBMREZF4F/A7gIjEh9RyZUktVxYa+p1EREREREREpOgx/Q4gIpIfN9xwA5ZlsWOHt46c4ziULVuW7t27H/Tczz//nC5dulC5cmWSk5OpWbMml1xyCdOmTcvz+enp6dx5553Url2bpKQkateuzV133cW2bdv2ew/btnnuuec44YQTSElJoXLlylx++eUsWrTogNm+/PJLOnToQJkyZShTpgwdO3bkq6++2u89Pv74Y+69917at29PWloahmFw/fXX7/f6q1evxjCMXB+JiYnUrl2b6667jj/++OOAGfe1YcMG3nrrLS655BJq1KhBYmIi5cqVo0OHDowcOZIDDZBetGgRV199NdWrVycpKYk6depw++23s3nz5kPKsLdNmzZx77330qhRI1JSUqhQoQKnnHIK/fv3z/XcLVu2cPXVV1OmTBnKli3Ltddey9atW/O87qpVq0hJSeGBBx447GwiIiIiIlJyaLqoiMSFE044Add1+f333wH4448/aNasGU899RT3339/nuc4jkOfPn0YPnw4aWlptG3blnLlyrF27Vrmzp3LI488wsMPP5zjnM2bN3PaaaexfPly6tWrR8uWLfnjjz/4448/OO644/j555+pUKFCrvtcdtllfPbZZ5QrV46zzjqLzZs38+OPP5KSksKkSZNo3bp1rnzPP/88d999N4FAgLPPPpukpCQmTJhAZmYmL774IrfffnuO52/bto3y5cvnuk7Pnj0ZMWJEnu/B6tWrqVu3LmlpaVx22WWx41u3bmXevHmsW7eOpKQkvv32Wzp06JDnNfZ1zTXX8N577xEIBGjZsiW1a9dm3bp1TJ8+PfZefPjhh1j7TO/94Ycf6Nq1KxkZGTRu3Jjjjz+e33//naVLl1KjRg1+/vlnatQ4tOlic+fOpXPnzmzZsoWmTZvSrFkzduzYwcKFC/nrr78I77OWYOfOnZkwYQLt27fHdV2mTp1K586d+eabb3Jd++KLL2bevHksXryYtDSt8ydypGm6qIiIiMQ7TRcVkSJv9+7dLFq0iJ49e8aOzZw5EyDP8ipq8ODBDB8+nK5duzJixIgc5Vh6enqeo6fuuusuli9fzqWXXspHH31EIOD9mOzbty8vvvgi/fr1y1VoDR8+nM8++4yGDRsydepUqlatCsCnn37KZZddxtVXX82iRYti1wJYsmQJ9957L0lJSUyaNInTTjsNgKVLl3L66adz991306VLFxo0aBA7JyEhgWuvvZaWLVvSqlUrlixZQq9evfL1HlaqVClX7mAwyA033MB7773HHXfcwa+//pqva1WsWJHHH3+cPn36ULly5djx2bNnc/bZZzN69GjeeustbrrppthjGRkZXHXVVWRkZPCf//yHQYMGAd4OuPfddx9PP/00N954I99++22+MoA3gq1Lly5kZmbyxRdfcNFFF+V4fNasWTn+Pnv2bCZMmMAtt9zCK6+8AkCfPn148803mTNnDi1btow995tvvmHs2LF89NFHKthERERERCRfNF1URIq8efPmYds2p556auzYzJkzMU2TFi1a5HnOX3/9xZNPPkmtWrX46KOPco0+K1++PA0b5lxgbv369XzwwQckJiby8ssv5yjFhg4dSuXKlRk1ahQbN27Mcd6zzz4LwJAhQ2IFG0D37t256KKLWL58OV988UWOc4YNG4Zt29xyyy2xgg3guOOO46GHHiIcDjNs2LAc56SlpfHOO+/Qt29fTjvtNJKTk/f7nuVHYmIigwcPBuC333474HTYfbM/+OCDOQo2gFatWjFgwAAAPvjggxyPjRkzhg0bNtCoUSMeffTR2HHDMHjiiSeoU6cOEyZM4Jdffsl3/kcffZTNmzczdOjQXAUb5C5gFyxYAJCjrL3hhhtyPAZe+di3b1/OPPNMLr/88nznERERERGRkk0lm4gUebNnzwbIVbI1atSIMmXK5HnOyJEjCQaD9O7dm5SUlHzd55tvvsFxHNq1a5ejLANISkqia9eu2LbN119/HTu+atUqFi1aREpKChdccEGua0anaH755Zc5jkfXXdt7CufBzjkSqlSpEvt836mVh+Okk04C4O+//85xfO7cuQC0b98e08z5Pz0JCQmcccYZALnKyP3JzMxk1KhRpKWl5Xs0X3p6OkCOKbfRz6OPATzzzDOsWrWKF198MV/XFRERERERAU0XFZEiaPLkyXTq1CnX8ZNPPjnXMcMwAKhduzarV6+OHf/hhx8AOP3001m/fj3vvfcey5cvp2zZsnTq1InOnTvHzo2KjqI65ZRT8sx1yimnMHz48BzTKqPnNGvWjISEhDzPAXKcs23bNtauXQtA8+bNc51Ts2ZNKlWqxJo1a9ixY8d+i8TCMGfOHAAqV65MpUqVCny9lStXAlCtWrUcx3fv3g2Q55py4E1BBfI9km3OnDns3LmTtm3bkpKSwvjx4/nuu+/IysriuOOO4/LLL+eYY47JcU6tWrUAb0puo0aNAG/a7t6P/fnnnzz++OP07duX448/Pl9ZREREREREQCWbiBRB1atX58YbbwS8Nbvefvtt6tatGyveli9fzpQpU+jSpQvHHnssQK6CaOHChbE/u3fvzvbt22OPDRkyhI4dO8Y2KoiKFl/7W3w/enzNmjWFck758uX3u95XjRo12Lx5M2vWrOGEE07I8zkFsW3bNmbMmMEdd9wBUCg7aIZCIV5++WXA2zRgb9GppXu/D3tbtWrVAR/fV/TrW6VKFbp165ZrBNyDDz7IW2+9xZVXXhk71rFjR1JSUhg0aBAnnngirusyaNAgUlNT6dixIwD33HMPZcqUYeDAgfnKISIiIiIiEqXpoiJS5DRq1Ig333yTN998kzvvvBPXdendu3fsWHTa6BtvvBE79tRTT+W4RnT6X79+/TjxxBOZN28eO3bsYOLEidStW5fJkyfTp0+fHOfs2rULgNTU1DxzRQuxnTt3HtFz9ndeQaxZswbDMGIf5cuX57zzzmPnzp188MEH3H333QW+xyOPPMKiRYuoW7cut9xyS47H2rdvD3jTZPfdcGLdunV89913QP5fb/TrO3bsWL755hteeuklNm7cyOrVq7n33nvJzMykZ8+eOdZaq1atGg8//DBz586lTp061K1bl/nz5/Poo49StWpVfvjhBz755BOGDBlC6dKlY+dlZGQc8nshIiIiIiIlj0o2ESnSJk2aBJBj+ujkyZOpV6/efkePATiOA3ijxcaPH0/z5s0pXbo0Z511FmPHjsUwDEaPHs3SpUuP7AsoItLS0ujZs2fs4/LLL+eUU05hw4YNPPDAA7Fpo4frww8/ZMiQISQnJ/P+++/nKhDPPfdcTjnlFHbt2sV5553HrFmz2LVrFz///DPnnXdebD24fddr25/o1zccDjN48GBuvfVWKleuTO3atRk6dCg9evQgFAoxdOjQHOc9+OCDjB8/nttvv5077riDCRMmcN999xEOh7njjjto164d11xzDeBt8FC1alXS0tKoWrWq1mgTEREREZED0nRRESnSJk2aRKlSpWjZsiXgjQKbN29erAjZn1KlSpGenk6PHj1yTcls1qwZrVq1YtasWfz4448cd9xxsXNg/yOXouuK7T3K6Uics7/zCqJSpUqMGDEi1/Fvv/2WCy+8kHPOOYcVK1bk2oU1P3744Qeuv/56TNPkgw8+oE2bNrmeYxgGY8aM4YILLmDOnDk5NrGoWrUqAwcO5OGHH97vmm37ir6HQJ4bH/Tq1YtPPvmEKVOm5HqsS5cudOnSJcexYcOGsWTJEubNmwd4u6HeddddXHPNNfTo0YPRo0fTt29fateunedOpiIiIiIiIhrJJiJFyogRI3JMa/ziiy/YtWsXCQkJGIZB6dKlCYfDOZ4XXU9rb7Vr1wagTp06ed4nenzjxo2xY9HF7//66688z4kej167oOekp6fHyrT8nHckdO7cmW7durFt2zbeeeedQz5/9uzZXHzxxQSDQd544w26deu23+fWrl2bBQsW8PHHH3PXXXdx880388ILL7Bw4cLYbq5NmzbN132j70tqampsvbe95fX13Z9//vmHQYMGceutt3LiiScC8PTTT1OvXj1GjhzJRRddxIgRI6hbty5DhgzJVz4RERERESl5NJJNRIqUBg0a0LNnT8ArSMaPH0+bNm1iu0H+/PPPLF26lKuvvppAwPsR1rhx41zXad68OQsWLIit3bWvrVu3AjlHRJ100kkAsdFM+4oejxYxe5/z+++/EwqFcu0wmtc55cqVo1atWqxdu5b58+fTtm3bHOf8+eefbN68mdq1ax/RnUWj6tatC8CyZcsO6byFCxdy3nnnsWvXLp577rk8R5TtKxAI0KNHD3r06JHj+PTp0wHyLEzzEt2VNTMzk+zsbJKSknI8ntfXd3/69+9PSkoKgwcPjh1bvHgxZ599dmz6qmmatGzZkokTJ+Yrn4iIiIiIlDwaySYiRUrbtm0ZMWIEI0aMiO1QOWTIkNixKlWqUKtWLUaNGhU7NmDAgFzXiU7py2u6YHTKKewpa8CbRmiaJlOnTs01Aio7O5svv/wSy7I4//zzY8fr1q1LkyZNyMzM5Kuvvsp1r9GjRwPQtWvXHMcvuOCCHI/n55wjZeXKlUD+Cqmo1atXc+6557JlyxYGDhzIXXfdddj3/+effxg9ejQVK1bk0ksvzdc5tWrV4qSTTsJ13Ty/xtFje3998zJt2jRGjRrFU089lWOnWcg9nXf37t35XjNORERERERKHv22ICJF1o8//khycjKtW7cGICsri9mzZ8d2qjyQrl270qRJE6ZPn87LL78cO27bNv369WPr1q00a9Ysxyiy6tWrc+WVVxIMBrn11ltji/ED3HfffWzatIlrrrmGKlWq5LhXv379Ys/Zu5wbM2YMY8eOpUGDBrHCMOrOO+/EsixeffVVZsyYETu+bNkyHn/8cQKBAHfeeWd+3qYC+fbbb/n8888BcpSHB7Jx40bOPfdc1q1bxz333MOjjz6ar/N+//13srKychz766+/uPjii9m5cyfPPPMMKSkpOR7/7LPPaNy4Mdddd12u6913330A3Hvvvaxfvz52fMGCBTzzzDMAuXY53Ztt29x+++20adOG66+/PsdjTZs2ZfLkyaxbtw7wdkCdMmVKvqezioiIiIhIyaPpoiJSZE2dOpXWrVvHpgLOnDmT7OzsfJVslmXx/vvv06FDB2677TZef/11GjRowPz581m5ciUVK1bk/fffxzCMHOc9//zzzJgxg08//ZTGjRvTsmVL/vjjD37//XcaNmzIs88+m+teN9xwA19//XWsEDrrrLPYvHkzU6ZMISUlhVGjRsWmtkY1atSIoUOH0q9fP9q1a8c555xDYmIiEyZMIDMzkxdeeIEGDRrkutett94aG4W3ZcsWAL766qscmw3sXdpFbd68OUeRlJmZyfLly2PXuvXWW+nQocNB31eAm2++mWXLlpGamprrulGVKlXi6aefznHs6aef5rPPPuOUU06hevXqbNy4kWnTppGdnc0jjzwSmya8t+3bt7NkyRKqVauW67GrrrqKCRMmMHLkSI4//nhOP/10MjMzmT59OtnZ2fTp0yfXtNS9vfLKK/z222/MmjUr1/fBgAEDuPDCC2nRogVnnHEGP/30E7t37+aBBx7I13skIiIiIiIlj+G6rut3CBGRfa1atYp69erx8MMP89hjjwHw2GOP8Z///IfFixfH1mjLz3UGDhzIhAkT2LJlC1WrVqVLly48/PDD+91UYOvWrQwcOJDPP/+cDRs2ULVqVS655BIGDRqUa0phlG3bDBs2jOHDh7NixQrS0tLo1KkTgwYN4vjjj99vvi+//JKhQ4cyf/58wJveeN9993HhhRfm+fyOHTvmOT1yb3v/WF+9enVszbW9WZZFxYoVadGiBb179873NM38ZqhduzarV6/Ocezzzz/n1Vdf5ZdffmHLli2UL1+e0047jbvuumu/a7GNGDGCXr160aFDByZPnpzrcdd1efPNN3nttddYtGgRhmFw4okncvPNN+dZ2kVt3ryZ4447jssvv5xXX301z+cMHz6cp556itWrV1O7dm0eeuihPAtFESkcvYwafkcQERERKRCVbCIiIiLiO5VsIiIiEu+0JpuIiIiIiIiIiEgBqWQTEREREREREREpIJVsIiIiIiIiIiIiBaSSTUREREREREREpIBUsomIiIiIiIiIiBSQSjYREREREREREZECUskmIiIiIiIiIiJSQCrZRERERERERERECkglm4iIiIiIiIiISAGpZBMRERERERERESkglWwiIiIiIiIiIiIFpJJNRERERERERESkgFSyiYiIiIiIiIiIFJBKNhERERERERERkQJSySYiIiIiIiIiIlJAKtlEREREREREREQKSCWbiIiIiIiIiIhIAalkExERERERERERKSCVbCIiIiIiIiIiIgWkkk1ERERERERERKSAVLKJiIiIiIiIiIgUkEo2ERERERERERGRAlLJJiIiIiIiIiIiUkAq2URERERERERERApIJZuIiIiIiIiIiEgBqWQTEREREREREREpIJVsIiIiIiIiIiIiBaSSTUREREREREREpIBUsomIiIiIiIiIiBSQSjYREREREREREZECUskmIiIiIiIiIiJSQCrZRERERERERERECkglm4iIiIiIiIiISAGpZBMRERERERERESkglWwiIiIiIiIiIiIFpJJNRERERERERESkgFSyiYiIiIiIiIiIFJBKNhERERERERERkQIyXNd1/Q4hIiIiIiIiIiISz/4fK4fIoKfrgooAAAAASUVORK5CYII=\n",
      "text/plain": [
       "<Figure size 1600x1200 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# For Lower Player\n",
    "masking_image(image, (195,654), (1068,654), (848,325), (398,325), os.path.join(MASKED_IMAGES,'Masked_Lower.jpg'))\n",
    "rgb_lower = exact_least_used_colors(os.path.join(MASKED_IMAGES,'Masked_Lower.jpg'),36, 10, 1.7, 5)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "RGB Values:  [(195, 253, 239)]\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1600x1200 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# For Upper Player\n",
    "masking_image(image, (398,360), (848,360), (848,235), (398,235), os.path.join(MASKED_IMAGES,'Masked_Upper.jpg'))\n",
    "rgb_upper = exact_least_used_colors(os.path.join(MASKED_IMAGES,'Masked_Upper.jpg'),36, 10, 1.7, 5)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Convert RGB To BGR for OpenCv\n",
    "\n",
    "bgr_upper = rgb_upper[::-1]\n",
    "bgr_lower = rgb_lower[::-1]"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Tracking Objects in Video with Particle Filters\n",
    "==============================================="
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Import libraries"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [],
   "source": [
    "import numpy as np\n",
    "import cv2\n",
    "import os\n",
    "\n",
    "# Repeatability\n",
    "np.random.seed(0)\n",
    "\n",
    "\n",
    "VFILENAME = os.path.join('data', 'match2.mp4')\n",
    "HEIGHT = 720\n",
    "WIDTH = 1280"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Load video frames from file"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [],
   "source": [
    "def get_frames(filename):\n",
    "    video = cv2.VideoCapture(filename) \n",
    "    frame_id = -1\n",
    "    while video.isOpened():\n",
    "        ret, frame = video.read() \n",
    "        frame_id += 1\n",
    "        if ret:\n",
    "            yield frame, frame_id\n",
    "        else:\n",
    "            break\n",
    "    video.release()\n",
    "    yield None"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Creating a particle cloud"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {},
   "outputs": [],
   "source": [
    "NUM_PARTICLES = 200\n",
    "# pixel per second\n",
    "VEL_RANGE = 0.5 \n",
    "\n",
    "# Global\n",
    "# BL = (195,654)\n",
    "# BR = (1068,654)\n",
    "# TR = (848,273)\n",
    "# TL = (398,273) \n",
    "\n",
    "def initialize_particles(BL, BR, TL, TR):\n",
    "    # particles given random value between 0 and 1; each particle has 4 attr\n",
    "    particles = np.random.rand(NUM_PARTICLES, 4) \n",
    "    # we dont want values between 0 and 1 so we multiply  \n",
    "    particles = (particles * np.array((TR[0] - TL[0], BR[1] - TR[1], VEL_RANGE, VEL_RANGE))) + (np.array((TL[0],TR[1],0,0)))\n",
    "    \n",
    "    # From above, we got velocity in range of 0 to 0.5  \n",
    "    # Now we want velocity to be centered at 0 and being able to go in either direction. \n",
    "    # Thats why we subtract the middle value (0.25) from all random values above\n",
    "    particles[:,2:4] -= VEL_RANGE/2.0 \n",
    "    print(particles[:20, :])\n",
    "    print('----------------------') \n",
    "    return particles"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Moving particles according to their velocity state"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {},
   "outputs": [],
   "source": [
    "def apply_velocity(particles):\n",
    "    # We are adding Vx to x beacuse this is being done for unit time \n",
    "    particles[:, 0] += particles[:,2]\n",
    "    particles[:, 1] += particles[:,3]\n",
    "    \n",
    "    return particles"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Computing x coordinate for a particular y coordinate on the line"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Left Line:  y + 1.88x - 1020 = 0\n",
    "# Right Line: y - 1.73x + 1196 = 0\n",
    "\n",
    "def get_x_left(y): \n",
    "    return (1020-y)/1.88 \n",
    "\n",
    "def get_x_right(y): \n",
    "    return (y+1196)/1.73"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Prevent particles from falling off the edge of the video frame"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {},
   "outputs": [],
   "source": [
    "def enforce_edges(particles,BL,BR,TL,TR, pid):\n",
    "    for i in range(NUM_PARTICLES): \n",
    "     \n",
    "        # If frame width is 100,and particle x position touches 100, we set it back to 99\n",
    "        # This wont be an issue if the position is any other x position because that position value will be less than width-1\n",
    "        \n",
    "        # y bound\n",
    "        particles[i,1] = max(TL[1], min(BL[1]-1, particles[i,1]))\n",
    "        \n",
    "        # x bound \n",
    "        if pid == 'lower':\n",
    "            particles[i,0] = max(get_x_left(particles[i,1]), min(get_x_right(particles[i,1]), particles[i,0])) \n",
    "        elif pid == 'upper': \n",
    "            particles[i,0] = max(BL[0], min(BR[0] -1, particles[i,0]))\n",
    "         \n",
    "    return particles"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Measure each particle's quality"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {},
   "outputs": [],
   "source": [
    "def compute_errors(particles, frame, TARGET_COLOR):\n",
    "    errors = np.zeros(NUM_PARTICLES)\n",
    "    for i in range(NUM_PARTICLES): \n",
    "        x = int(particles[i,0])\n",
    "        y = int(particles[i,1]) \n",
    "        # Every pixel at x,y in frame will have 3 values denoting RGB \n",
    "        pixel_color = frame[y,x,:] \n",
    "        errors[i] = np.sum((TARGET_COLOR - pixel_color)**2)\n",
    "    \n",
    "    return errors"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Assign weights to the particles based on their quality of match"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {},
   "outputs": [],
   "source": [
    "def compute_weights(errors, particles,BL,BR,TL,TR, pid):\n",
    "    weights = np.max(errors) - errors \n",
    "    \n",
    "    # Setting weights of particles on the boundary to 0\n",
    "    \n",
    "    if pid == 'lower':\n",
    "        weights[\n",
    "            (particles[:,0] == get_x_left(particles[:,1])) | \n",
    "            (particles[:,1] == TL[1]) | \n",
    "            (particles[:,0] == get_x_right(particles[:,1])) | \n",
    "            (particles[:,1] == BL[1]-1)\n",
    "        ] = 0.0 \n",
    "        \n",
    "    elif pid == 'upper': \n",
    "        weights[\n",
    "            (particles[:,0] == BL[0]) | \n",
    "            (particles[:,1] == TL[1]) | \n",
    "            (particles[:,0] == BR[0]-1) | \n",
    "            (particles[:,1] == BL[1]-1)\n",
    "        ] = 0.0\n",
    "    \n",
    "    weights = weights**8\n",
    "    \n",
    "    return weights"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Resample particles according to their weights"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {},
   "outputs": [],
   "source": [
    "def resample(particles, weights):\n",
    "    # Normalize to get valid PDF\n",
    "    probabilities = weights / np.sum(weights)\n",
    "\n",
    "    # Resample\n",
    "    indices = np.random.choice(\n",
    "        NUM_PARTICLES,\n",
    "        size=NUM_PARTICLES,\n",
    "        p=probabilities)\n",
    "    \n",
    "    particles = particles[ indices, : ]\n",
    "\n",
    "    # Take average over all particles, best-guess for location\n",
    "    x = np.mean(particles[:,0])\n",
    "    y = np.mean(particles[:,1])\n",
    "    return particles, (int(x),int(y))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Fuzz the particles"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {},
   "outputs": [],
   "source": [
    "def apply_noise(particles, POS_SIGMA, VEL_SIGMA):\n",
    "    noise = np.concatenate(\n",
    "    (\n",
    "        np.random.normal(0.0, POS_SIGMA, (NUM_PARTICLES,1)), \n",
    "        np.random.normal(0.0, POS_SIGMA, (NUM_PARTICLES,1)), \n",
    "        np.random.normal(0.0, VEL_SIGMA, (NUM_PARTICLES,1)), \n",
    "        np.random.normal(0.0, VEL_SIGMA, (NUM_PARTICLES,1))\n",
    "    ), \n",
    "    axis = 1)\n",
    "    \n",
    "    particles += noise \n",
    "    return particles\n",
    "    "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Predict automatic shot direction based on position "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {},
   "outputs": [],
   "source": [
    "import math \n",
    "\n",
    "def apply_shot_direction(frame, location_lower, location_upper): \n",
    "    THRESHOLD_STRAIGHT = 100.0\n",
    "    THRESHOLD_CROSS = 50.0 \n",
    "    font = cv2.FONT_HERSHEY_PLAIN \n",
    "    \n",
    "    distance = abs(location_upper[0] - location_lower[0])  \n",
    "    \n",
    "    if distance>=THRESHOLD_STRAIGHT: \n",
    "        cv2.putText(frame, 'STRAIGHT', (0, HEIGHT-20), font,2,(255,255,255), 2)\n",
    "    \n",
    "    elif distance<=THRESHOLD_CROSS: \n",
    "        cv2.putText(frame, 'CROSS', (0, HEIGHT-20), font,2,(255,255,255), 2)\n",
    "        \n",
    "    else: \n",
    "        cv2.putText(frame, 'NEUTRAL', (0, HEIGHT-20), font,2,(255,255,255), 2)\n",
    "        \n",
    "    return frame "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Display the video frames"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Reading Non Playing Frames"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "metadata": {},
   "outputs": [],
   "source": [
    "PATH_FRAME_TIMESTAMPS = os.path.join('data','cnn','frame_timestamps.txt') \n",
    "frames_nonplay = [] \n",
    "with open(PATH_FRAME_TIMESTAMPS, 'r') as f: \n",
    "    frames = f.read().split(',') \n",
    "\n",
    "frames_hit = [] \n",
    "PATH_TIMESTAMPS = os.path.join('timestamps.txt') \n",
    "with open(PATH_TIMESTAMPS,'r') as f: \n",
    "    timestamps = f.read().split(',') \n",
    "    for i,second in enumerate(timestamps):\n",
    "        frame_id = int(30.0*float(second))\n",
    "        frames_hit.append(frame_id) \n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "metadata": {},
   "outputs": [],
   "source": [
    "def display(frame, frame_id,  particles_lower, particles_upper, location_lower, location_upper):\n",
    "    if len(particles_lower) > 0:\n",
    "        for i in range(NUM_PARTICLES):\n",
    "            x = int(particles_lower[i,0]) \n",
    "            y = int(particles_lower[i,1])\n",
    "            cv2.circle(frame, (x,y),1, (255,255,255), 1) \n",
    "            \n",
    "    if len(particles_upper) > 0:\n",
    "        for i in range(NUM_PARTICLES):\n",
    "            x = int(particles_upper[i,0]) \n",
    "            y = int(particles_upper[i,1])\n",
    "            cv2.circle(frame, (x,y),1, (252, 36, 3), 1)        \n",
    "    \n",
    "    if len(location_lower) > 0:\n",
    "        cv2.rectangle(frame, (location_lower[0] - 60, location_lower[1] - 100), (location_lower[0] + 70, location_lower[1] + 160), (0,0,255), 2)\n",
    "    \n",
    "    if len(location_upper) > 0:\n",
    "        cv2.rectangle(frame, (location_upper[0] - 50, location_upper[1] - 100), (location_upper[0] + 70, location_upper[1] + 90), (0,255,0), 2)\n",
    "    \n",
    "#     if frame_id in frames_hit: \n",
    "    frame = apply_shot_direction(frame, location_lower, location_upper) \n",
    "    \n",
    "    cv2.imshow('frame',frame) \n",
    "    # 27 is the escape key. The number in the function is waiting time\n",
    "    if cv2.waitKey(30) == 27:\n",
    "        if cv2.waitKey(0) == 27:\n",
    "            return True \n",
    "    \n",
    "    return False"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Function for checking frame type "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Main routine"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[[ 6.44966077e+02  5.60297302e+02  5.13816880e-02  2.24415915e-02]\n",
      " [ 5.88644660e+02  5.37499163e+02 -3.12063944e-02  1.95886500e-01]\n",
      " [ 8.31648242e+02  4.51152260e+02  1.45862519e-01  1.44474599e-02]\n",
      " [ 6.53620052e+02  6.29521294e+02 -2.14481971e-01 -2.06435350e-01]\n",
      " [ 4.07098279e+02  5.98931929e+02  1.39078375e-01  1.85006074e-01]\n",
      " [ 8.38378254e+02  5.87923168e+02 -1.92603189e-02  1.40264588e-01]\n",
      " [ 4.51223492e+02  5.35534016e+02 -1.78323356e-01  2.22334459e-01]\n",
      " [ 6.32831745e+02  4.61423778e+02 -1.17722194e-01  1.37116845e-01]\n",
      " [ 6.03267649e+02  5.12014769e+02 -2.40605100e-01  5.88177485e-02]\n",
      " [ 6.73443075e+02  5.27971285e+02  2.21874039e-01  9.09101496e-02]\n",
      " [ 5.59778555e+02  4.68783513e+02  9.88155980e-02 -2.19887264e-01]\n",
      " [ 6.98045022e+02  5.45639859e+02 -1.44808719e-01 -1.85536851e-01]\n",
      " [ 5.39942758e+02  4.44660844e+02  3.50983852e-02 -3.06992433e-02]\n",
      " [ 8.42768227e+02  3.58572743e+02 -1.45561622e-01 -1.69345241e-01]\n",
      " [ 6.91898746e+02  4.08332937e+02 -1.68446136e-02 -1.27787204e-01]\n",
      " [ 4.69536313e+02  3.61313421e+02  7.81647947e-02 -1.80908524e-01]\n",
      " [ 4.86462063e+02  4.46310581e+02  1.60496615e-01 -2.01449362e-01]\n",
      " [ 7.75075208e+02  3.56616376e+02  2.38229733e-01 -1.56743992e-02]\n",
      " [ 8.37542490e+02  5.23994176e+02  1.19631790e-01 -2.30406104e-01]\n",
      " [ 5.25263133e+02  3.64544669e+02 -1.01929901e-01 -1.90636141e-01]]\n",
      "----------------------\n",
      "[[ 4.15996761e+02  3.09558715e+02 -4.58485458e-02 -6.12967137e-02]\n",
      " [ 7.62214237e+02  3.19264964e+02  2.27166908e-01 -7.40318798e-02]\n",
      " [ 8.01894244e+02  3.27795406e+02 -7.12876742e-02  6.08327182e-02]\n",
      " [ 5.27856481e+02  3.42415988e+02 -1.93786341e-01 -1.43782819e-01]\n",
      " [ 4.80364981e+02  2.76423640e+02  1.22616480e-01  1.34537245e-02]\n",
      " [ 6.17454346e+02  2.20076435e+02 -3.72991373e-02 -2.18223113e-01]\n",
      " [ 4.91713963e+02  3.50535151e+02 -1.42300898e-01  1.79168819e-01]\n",
      " [ 7.59302017e+02  2.42280473e+02  5.28559786e-02 -1.92169064e-01]\n",
      " [ 7.25549671e+02  3.09244719e+02  1.55969281e-01 -1.03077253e-02]\n",
      " [ 8.09688390e+02  2.26908853e+02 -1.03555717e-01  1.07526299e-01]\n",
      " [ 5.86149145e+02  2.44213190e+02 -1.96394627e-01  1.58669556e-01]\n",
      " [ 6.10914340e+02  3.43519714e+02  1.16644567e-01 -4.51368972e-02]\n",
      " [ 5.66079956e+02  2.92189369e+02  1.94529977e-01  1.18639290e-01]\n",
      " [ 4.00318834e+02  3.17182099e+02  2.09753703e-01  1.05227880e-01]\n",
      " [ 4.77652602e+02  2.87692538e+02 -1.79841991e-01 -7.05023608e-02]\n",
      " [ 8.19702669e+02  3.49262743e+02 -1.08581574e-01 -8.01844779e-02]\n",
      " [ 6.68095791e+02  3.54847621e+02 -1.76099333e-01 -1.21541678e-01]\n",
      " [ 7.91100572e+02  2.88864912e+02  1.99480546e-01 -1.57241051e-01]\n",
      " [ 6.37700864e+02  2.65677749e+02 -9.17287201e-02 -2.65615180e-02]\n",
      " [ 5.92884852e+02  2.70028563e+02  2.07485385e-01  1.15872093e-01]]\n",
      "----------------------\n"
     ]
    }
   ],
   "source": [
    "# Global\n",
    "# BL = (195,654)\n",
    "# BR = (1068,654)\n",
    "# TR = (848,273)\n",
    "# TL = (398,273) \n",
    "\n",
    "# Lower \n",
    "BL_LOWER = (195,654) \n",
    "BR_LOWER = (1068,654)\n",
    "TL_LOWER = (398,325)\n",
    "TR_LOWER = (848,325)\n",
    "\n",
    "# Upper \n",
    "BL_UPPER = (398,360) \n",
    "BR_UPPER = (848,360)\n",
    "TL_UPPER = (398,220)\n",
    "TR_UPPER = (848,220)\n",
    "\n",
    "particles_lower = initialize_particles(BL_LOWER,BR_LOWER,TL_LOWER,TR_LOWER) \n",
    "particles_upper = initialize_particles(BL_UPPER,BR_UPPER,TL_UPPER,TR_UPPER) \n",
    "\n",
    "for frame, frame_id in get_frames(VFILENAME):\n",
    "    if frame is None: break \n",
    "        \n",
    "    # Re-initialize particles and dont display them if non-playing frame\n",
    "    if str(frame_id) in frames_nonplay: \n",
    "        particles_lower = initialize_particles(BL_LOWER,BR_LOWER,TL_LOWER,TR_LOWER) \n",
    "        particles_upper = initialize_particles(BL_UPPER,BR_UPPER,TL_UPPER,TR_UPPER)  \n",
    "        cv2.imshow('frame',frame) \n",
    "        if cv2.waitKey(30) == 27:\n",
    "            if cv2.waitKey(0) == 27:\n",
    "                break \n",
    "    \n",
    "    else: \n",
    "        frame = change_court_boundary(frame) \n",
    "        particles_lower = apply_velocity(particles_lower) \n",
    "        particles_upper = apply_velocity(particles_upper) \n",
    "\n",
    "        particles_lower = enforce_edges(particles_lower, BL_LOWER,BR_LOWER,TL_LOWER,TR_LOWER,'lower')\n",
    "        particles_upper = enforce_edges(particles_upper, BL_UPPER,BR_UPPER,TL_UPPER,TR_UPPER,'upper')\n",
    "\n",
    "        errors_lower = compute_errors(particles_lower, frame, bgr_lower) \n",
    "        errors_upper = compute_errors(particles_upper, frame, bgr_upper) \n",
    "\n",
    "        weights_lower = compute_weights(errors_lower, particles_lower,BL_LOWER,BR_LOWER,TL_LOWER,TR_LOWER, 'lower')\n",
    "        weights_upper = compute_weights(errors_upper, particles_upper,BL_UPPER,BR_UPPER,TL_UPPER,TR_UPPER, 'upper')\n",
    "\n",
    "        particles_lower, location_lower = resample(particles_lower, weights_lower) \n",
    "        particles_upper, location_upper = resample(particles_upper, weights_upper) \n",
    "\n",
    "        particles_lower = apply_noise(particles_lower, 7.0,3.0)\n",
    "        particles_upper = apply_noise(particles_upper, 4.0,2.0)\n",
    "\n",
    "        particles_lower = enforce_edges(particles_lower, BL_LOWER,BR_LOWER,TL_LOWER,TR_LOWER,'lower')\n",
    "        particles_upper = enforce_edges(particles_upper, BL_UPPER,BR_UPPER,TL_UPPER,TR_UPPER,'upper')\n",
    "\n",
    "        terminate = display(frame, frame_id, particles_lower, particles_upper, location_lower, location_upper)\n",
    "        if terminate:\n",
    "            break\n",
    "            \n",
    "cv2.destroyAllWindows()\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "badminton",
   "language": "python",
   "name": "badminton"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.9.13"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 4
}