synthtabnet_demo.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "id": "61227146-7500-4016-8ce8-f5b76e8d1808",
   "metadata": {},
   "source": [
    "# SynthTabNet Dataset\n",
    "\n",
    "SynthTabNet is a dataset of 600k `png` images from synthetically generated table layouts with annotations in `jsonl` files.\n",
    "\n",
    "The dataset is open sourced by IBM Research Zurich and is freely available to download on [GitHub](https://github.com/IBM/SynthTabNet)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "94ebf3e7-81c6-4460-aec2-68c6564f6b62",
   "metadata": {},
   "source": [
    "## Download the files\n",
    "\n",
    "We have prepared a small subset of SynthTabNet called `samples`. The following code simply downloads it and unzips it in a local directory."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "id": "a353ea1b-b6e8-4018-9728-11680b2d8eee",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Downloading...\n",
      "Saving...\n",
      "Extracting...\n",
      "Done\n"
     ]
    }
   ],
   "source": [
    "import os\n",
    "import json\n",
    "import jsonlines\n",
    "import zipfile\n",
    "import requests\n",
    "from PIL import Image\n",
    "import matplotlib.pyplot as plt\n",
    "import matplotlib.image as mpimg\n",
    "from IPython.display import display, HTML\n",
    "\n",
    "samples_url = \"https://ds4sd-public-artifacts.s3.eu-de.cloud-object-storage.appdomain.cloud/datasets/synthtabnet_public/samples.zip\"\n",
    "samples_fn = \"samples.zip\"\n",
    "\n",
    "# Download and extract the samples\n",
    "# Path(samples_dir).mkdir(parents=True, exist_ok=True)\n",
    "print(\"Downloading...\")\n",
    "req = requests.get(samples_url)\n",
    "print(\"Saving...\")\n",
    "open(samples_fn, \"wb\").write(req.content)\n",
    "print(\"Extracting...\")\n",
    "zipfile.ZipFile(samples_fn, \"r\").extractall(\".\")\n",
    "print(\"Done\")"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "27889535-d33f-4b41-87cf-aee585a78ac8",
   "metadata": {},
   "source": [
    "## Show some example images\n",
    "\n",
    "The next step is to visualize some random images among all dataset parts:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "id": "07c2902f-5065-4bc7-a27a-29141641da26",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "PubTabNet sample:\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Sparse sample:\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "FinTabNet sample:\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Marketing sample:\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Visualize images from each part of the dataset\n",
    "img_fns = {\n",
    "    \"PubTabNet\": \"samples/pubtabnet/images/train/image_000005_1634629104.274936.png\",\n",
    "    \"Sparse\": \"samples/sparse/images/train/image_000005_1634629370.551275.png\",\n",
    "    \"FinTabNet\": \"samples/fintabnet/images/train/image_000014_1634629328.541362.png\",\n",
    "    \"Marketing\": \"samples/marketing/images/train/image_000024_1634629424.186544.png\"\n",
    "}\n",
    "\n",
    "for part_name, img_fn in img_fns.items():\n",
    "    img = mpimg.imread(img_fn)\n",
    "    imgplot = plt.imshow(img)\n",
    "    print(f\"{part_name} sample:\")\n",
    "    plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "73e49a11-82e6-4def-a16a-14a32363bd7f",
   "metadata": {},
   "source": [
    "## Check the annotations\n",
    "\n",
    "The `synthetic_data.jsonl` file contains the annotations for each part in [json lines format](https://jsonlines.org/)\n",
    "\n",
    "More specifically:\n",
    "\n",
    "- Each line of the file contains a `json` object with annotations for one image.\n",
    "- The data format of each `json` object is described in [GitHub](https://github.com/IBM/SynthTabNet).\n",
    "\n",
    "Let's check the first image of the `PubTabNet` part together with its annotation:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "id": "c68dcd42-b48c-45b9-9d50-05495c38961c",
   "metadata": {
    "scrolled": true,
    "tags": []
   },
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAATgAAAD8CAYAAADjcbh8AAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjUuMSwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy/YYfK9AAAACXBIWXMAAAsTAAALEwEAmpwYAACSeklEQVR4nOydd3gU17n/P7Or3VXvvSIkIYFASPRiejEd4oKNS1ziHtuJHSdOu/c6yfV1S+K4YmODAdvYYHoHIUAUSYAQEqij3suqr7R95/cH2vmBK6IEQebzPHq0O3Nm5j1nzr5zypzvK4iiiIyMjMytiOJGGyAjIyNzvZAdnIyMzC2L7OBkZGRuWWQHJyMjc8siOzgZGZlbFtnBycjI3LJcNwcnCMIcQRCKBEEoEQTh99frOjIyMjI/hHA93oMTBEEJFAOzgBrgFLBMFMX8a34xGRkZmR/gerXgxgAloiiWiaJoAr4GFl+na8nIyMh8Lw7X6bwhQPVF32uAsT+UWBAEURCE62SKjIzMrYQoioiieFkO43o5uJ9EEIQngCcAfHx8+Mc//nGjTJGRkbmJeOWVVy477fVycLVA2EXfQ3u3SYiiuAJYATBq1CjxoYceuk6myMjI3Eq89957l532eo3BnQJiBEGIFARBDdwLbL9O15KRkZH5Xq5LC04URYsgCM8C+wAlsEoUxbzrcS0ZGRmZH+K6jcGJorgb2H29zi8jIyPzU9ywSYYfw2az0d3djdVqRRAEnJycUKlU9GWmVRRFzGYzRqMRFxcXFIobt2jDnh9BEHBxccFoNGI2m3F1de1Tnq6XbT09PVgslp8sa1EUpbSurq4oFIobbv+1wGg0YjQapTx9+7sd+33UaDR9ro+XiyiKmEwmDAYDSqUSJycnlErlFZ3Lbq+joyMKhQKr1YpKpZLuuaOjIyqV6rKOuVnvc79cqtXe3s7zzz/P0qVLuf/++/nTn/6EVqvt83mOHDnCr3/9a1paWq6DlZdPR0cHzz//PM888wytra18/fXX/OpXv0Kv199QuwC6urp48cUXWbp0Kffddx+vvvoqOp2O73sBvKKigieffJKXXnqJ6urq7znbzcmRI0d49NFHqaysxGq1smrVKl544QW6u7svSdfa2spzzz1HSkrKdbOlpqaGl156iYceeojHH3+ckydPXvG5Ghsb+eMf/0hWVhZpaWm8++676PV6Ghsb+d3vfse5c+e+c5+bmpp46aWXOHXqFMePH+edd97pF/X0SumXLTiLxUJeXh4jRoxg6tSpvPjiiwwdOpSf//znAJSXl7N7927uvfde2tra2LNnD4sXL2b37t1otVoGDBjA4sWLaW1t5dy5c+h0OjZs2MC4cePw9/fnq6++4uGHH8ZisbB9+3Y6OjqYPXs2w4cPvy4tPYvFQn5+PkVFRUyfPp26ujrOnTuH1Wq95tfqK1arlfz8fIYOHUpcXBz/+te/WLBgAYMHD2bPnj0UFxeTlJTE6NGj+ec//8nOnTt54IEHUCqVbN++ndzcXIYOHcqsWbM4duwY5eXliKLI8OHDcXNzY9++fWg0GpYsWYKDgwOrVq0iKiqKyspK5s6dy+DBgzlz5gwpKSl4eHhw77330trays6dO7FarcybN4+BAweSnJzMmTNn8PX15Y477sDPz++yWxU2m42UlBRKS0sRBAFBEFi6dCkeHh4IgkBkZCT5+fkcOnSIJUuWsH79ekaPHs3x48fJyspCrVYzb948PD09OXfuHDNnziQ5OZmamhruvfdevvzyS2JjYxk5ciR79uyhqKiIpKQkZs6c2efWz65du0hJSeHVV19FEATc3d0xm80cPnyYU6dOERkZycKFC6mpqWH37t34+fmh1Wq56667cHZ2ZuPGjWi1WuLj4xk7diwmkwmdTsdHH33EqVOn0Gq13HvvvdL25cuXM2XKFEJDQ/nss88YMWIEFosFnU7H6tWrOX78OF1dXYwYMQKtVssDDzzAqVOnqKioYOnSpWg0miutev8W+mULzo7RaKSzsxNBEHBwcECn09HZ2YlSqZQK/5tvviElJQWbzUZDQwM6nY6//OUvrF+/XjqPyWRi1apV5OTkUFtby7vvvktVVRUvvvgi69evJzs7m0cffZSqqqrrlheFQsG4cePYsGEDHR0d/a7JL4oiFosFX19fPD09WbVqFX/5y1+or6/nV7/6FSkpKbi4uODo6EhsbCybN2/mt7/9LXV1dfz2t79l8+bNHDp0iL/85S+Ul5fT0NDA448/zqlTp9i2bRu/+c1vqK2t5a233iI5OZnDhw/z17/+lVOnTvHQQw9RWlqKXq+nsrJSaiUdPXqUZ599ltOnT/Pyyy/T1NRES0sLXV1dfc7b4cOHeeutt+js7OTDDz9k5cqV9PT00NHRgb+/P+PGjWPHjh3k5ORQWVnJvHnzaG5uxmg0sn//fl566SU6Ozulcx48eJB169ah1+v5/PPPOXbsGGvWrOF//ud/qK2t5cUXXyQ5ObnP9yE6Ohqr1cry5cvJyMhAo9Gwf/9+nn32WSoqKnj99ddZuXIlxcXFvPrqq+Tk5PD111+zYsUKtmzZwvvvv4/FYqGqqoqenh4yMzNpbW3F1dUVtVqNv78/FouFkydP0t3dzdGjR/niiy/IyMjgo48+wmKxcOLECVpaWnBxcUGlUuHn54eDgwPvv/8+2dnZfPTRR5SWll5x1/nfSb92cCdOnGDfvn08+eSTjB49ml/96lc89NBDNDU1MWXKFL755ht2797NwoULaWxsJC0tjcrKSvR6PSUlJZecy94UN5lMAHR3d5ORkYHBYEAURcLCwmhra7tueRFFkUmTJuHj48Pu3bux2WzX7VpXwtGjR1m9ejUDBw7Ex8eHtLQ0DAYDra2txMXFYbVaGT16NF5eXixevJizZ8/S09NDe3s7cXFxktMZPnw4f/zjH/Hx8aG8vBydToeHhwfOzs6YzWY0Gg2PPvoos2bNora2lry8PEwmEy+++CLPP/88Dg4OZGdno9frpR+XRqNhypQplJSUkJWV1WcHZycuLo7HHnuMUaNGcebMGVauXMn999/PZ599xuLFi8nOzubTTz9lwIABxMXFkZWVRXZ2Nkajkerqanp6eqRzCYKAKIrYbDasViuiKJKZmYnRaKS9vZ2YmJjvdHEvhwkTJvD555+zePFitm7dyocffsipU6dobW2ls7OTAQMG0NPTgyiK+Pn58dRTTzF27FgqKysZNGgQAwcOJCMjg/LycvR6PaIo4uzszNixYwkODuaRRx7B29sbALVazZ133snhw4dZt24dEyZMIDQ0VDpmxIgRBAYG8sADDzB16lTi4uJ4//33KSgoYN68ebKDu1ruvPNO1q5dyx/+8AfCw8N5+umneemll4iOjmb+/PkcOnSI9vZ2pk2bRl5eHiUlJUydOhVXV1fJoQmCIP1QDh48yJo1a6QbOHz4cPz8/Jg6dSqjRo0iMDDwuubH1dWV++67j5qamu8d47qR3HXXXbzzzjtkZ2ezf/9+kpKScHNzY8KECUycOJHw8PBL0o8YMQI3NzfGjBnD5MmTGTBgwCWTFCEhIQQHBzNw4EBmzJhBUlISarUaAAcHBxQKBaIoSsd99NFHrFq1CovFQlxcHKGhoUyfPp0RI0bg6upKfHw8kydP5ty5c+Tm5l5RHvPz8/nggw9IT08nPj6eOXPm8PLLLzN37lzGjBmDh4cH27dvZ968eVgsFg4ePEhcXBzR0dHfOVdISAjV1dWsWrWK8vJyFAoFw4YNw93dnXHjxjFlyhRCQkL6bOOuXbtYu3YtHR0dADg5OTF8+HC8vb0ZPnw4U6dOZfDgwSgUClQqlVSW9gfmjBkziIqKYv/+/bS2tkrn9fT0pKmpic2bN6PT6aTt48aNk1qpd9555yWTDh4eHrS2trJ582YsFgvLli1j165dhIeHEx8f3+96Id9HvxyDc3Jy4p577mHIkCE4OztLTmr06NHAhdbQiBEjeP755/H29mbAgAEsWrSI9vZ2LBYLjzzyCIMHDyY8PJylS5fi7e3NX/7yF3bv3s3AgQMJCQkhLCyMd955h40bN1JdXU1sbCzu7u7XLT9Lly4lMTGRESNG8Oc//xmbzfadGawbgaOjI3fffTexsbFMnDiRl156Cb1ezxNPPEFAQACFhYUEBAQQERGBu7s7Dz74IG5ubjz00EN4enqSm5uLr68vMTExWK1WoqKicHBwIDIykhUrVrBr1y4aGhqYOHEiAQEBPPbYYwQEBJCYmIjNZmPMmDGsXLmS/fv309nZSVhYGO+//z5btmyhpqaGpKQkvLy8UCqVtLe389xzzzF37twrymt0dDROTk48+uijPPTQQ/j4+DBo0CDgwljkSy+9RG5uLnfccQeBgYG8+uqrZGVlERMTQ2xsLH5+fixbtoy4uDgiIiLQ6/U4OzvzzDPPMHr0aMaMGYOPjw/5+fkEBQURERHRZxtHjRpFfX09NTU1PPXUU9x77714eXmhUCg4ceIEgiAQExODKIrcf//9uLm5MXnyZLq7u/H29qazsxN3d3def/114uLiWLZsGQMGDMDf35/KykoqKiqYNGkS999/P2FhYQQEBPDCCy9QVlbGpEmTMJvNPPDAAwwcOJCAgABqamqk7m5UVBQhISEsWbIEJyenK7oH/26ui1xSXxk1apSYmZl5o82QuUWxWq28//77VFZW8tprr/X7gfH+SEtLC3/6058QRZH//d//xc/P74bZMmrUKDIzM/v3YnsZmX8XCoWCX/ziF1gsFqmbLNM3PD09efXVV9FoNLi4uNxocy4b2cHJ3PIIgoCrq+uNNuOmRqlU4uPjc6PN6DP9epJBRkZG5mqQHZyMjMwti9xFlZG5AYiiSHNzM7W1tT+dWOYSLn4f8aeQHZyMzA1i06ZN/Pd///eNNuOmo729/bLTyg5ORuYGsXjxYoYPH36jzbjpePTRRy87rezgZGRuAIIgEBwcTHBw8I025aajLzPi8iSDjIzMLctVteAEQagAugArYBFFcZQgCN7AemAAUAEsFUXx+q1il5GRkfkBrkULbpooiomiKI7q/f57IEUUxRggpff7FdMbA/F7/24Wbgb7f6yc+6vNMjI/xfUYg1sMTO39vAY4DLx8pScrKyvj6NGj3HXXXbS2tpKcnIyrqyuLFi26aRb86vV6tm/fjlarRalUMmTIECZMmNAvFtuLoohOp+PIkSNUVlYSEhLClClTcHJyYvfu3dTW1kpKGePGjesXNt/K6HQ6Wlpa8PPzw9HRkbq6Ojw8PHBzcwMu3K+2tjYMBgMBAQEoFApaWlrw8PCQ7833cLUtOBHYLwjC6d5AzgABoijW935uAAKu5MTt7e0UFhaSnp7OK6+8Qk1NDW+++Sb//Oc/qauru6laEzqdjn/84x9s376dzMxMfvGLX5CX1z+CjBmNRv72t7/xxz/+kbNnz/LKK6/wu9/9jvb2dt577z2+/PJLUlNTeeSRR/qNzbcyBw8eZN68eXz66ad0dnby0EMPsWvXLml/XV0dL7/8Mv/zP//DgQMHqKmpkYQ8Zb7L1Tq420RRHAHMBX4pCMLki3eKF7zQ93oiQRCeEAQhUxCEzObm5ouPob29nRdeeIEHH3yQNWvWYLPZKC8vZ8+ePbS3t9PS0nJTaFHZEQQBhUJBWFgY8fHxKBQKHBz6xwR2dXU169ev55e//CXvvfceL774Irt37yY/Px9BEBgzZgxPP/00RqPxigQcZfpGT08PlZWVfPbZZ9TU1FBbW3uJfltZWRk6nY5hw4Zx9OhR1qxZw2233SaJWMpcylU5OFEUa3v/NwFbgDFAoyAIQQC9/5t+4NgVoiiOEkVx1LelV86fP8/hw4f5y1/+wh133IHNZmPw4MHMnDmTmTNn8l//9V84Ojpejek3hPT0dHbvvhBJ8YcCu/y7MRgMGI1GgoKCUKlUBAYGSt1WgL179/K73/2OpKQkYmNjb7C1/xmEhYURFhbGV1999Z24HUOGDCEyMpKysjKCgoLo6OjAxcWFgwcPXuIIZS5wxQ5OEAQXQRDc7J+B2UAuFyLYP9Sb7CFgW1/PbVcqNRqNksR473Uu+bvZmDt3Lq+88gpGo5GsrKwbbQ4AAQEBREdHs379ek6cOMHGjRvx8fGRVGzHjh3LsmXLKC4upqCgoF845VsdlUrF448/zp49e2hsbLxkn7e3N3/961/54x//SElJCfHx8Xz88cd8+eWXbNu2Tb4/3+JqWnABwDFBEHKAk8AuURT3Aq8DswRBOA/M7P3eJwYNGsTSpUt5++23OX78OCEhITg4OODl5YW3t/dN59wUCgUBAQHs2bOHF154gaSkJGbMmHGjzQLA19eX1157ja6uLp555hkqKyt56623CAkJwc/Pj0GDBvHII48wevRoNm/efMkDR+b6oFAoiI+PZ8KECRiNxkvqu/3hnpycTExMDAMHDkSpVKJWq/tFlLb+Rr9U9LUHbTYYDDg4OEiBho1GI6Io4uTkdFM5uW8HV1apVP0mD/b7r9frMZvNqFQqHB0dEQSB7u5ulEoljo6O6PV6bDYbLi4u/cLuW5Xi4mJOnz7NggULaG5uZv/+/UyaNIn4+HgpjdlsJi0tjaFDh+Lm5sbhw4fR6XTMnDnzusnu9yf6oujbLx2cjIyMzA/RFwcnL9WSkZG5ZZEdnIyMzC2L7OBkZGRuWWQHJyMjc8vSL16nb2trY+PGjTfaDBkZmZuAtrbLFyfqF7OoGo1GDAoKutFmyMjI3ATU19djNBpvnsDPsbGxlywolpGRkfkh5s+ff9lp+4WDU6vVhIWF3WgzZGRkbgLUavVlp5UnGWRkZG5ZZAcnIyNzyyI7OBkZmVuWfjEG921EUaSzs5OSkhJUKhXR0dH9ZnH6lSCKIl1dXZw/f56AgABCQkIwGo0UFxdjNpuJjo7G3d0dQRAQRZHa2loqKysRBAE3Nzfi4uJobW2lpqaGyMhIvL296enp4fz584SHh+Pp6YlWq8XR0VGStr5cbDYbdXV11NfXExISQlBQEBaLhdLSUqxWK7GxsZI4pyiK1NXVUVFRgSAIuLq6MmTIEJRKJQaDgdLSUgYNGoTVaqWoqIiIiAg8PT1paWlBrVbj5uZ2w+6hKIpotVoqKioICgoiODgYhUIhRZivrKzEzc2N6OholEolLS0tlJeXS2l7enooKSkhKioKV1dXGhsbcXd3x9nZ+ZraabPZqKyspL29nbi4OEn4wJ4Hu63BwcEEBgZSWVmJVqslKCiIkJAQTCYTxcXFhISE4O3tTWtrKw4ODlL9uhqMRiNlZWXAhYlBQRAwm80UFRWh0WiIiopCqVReckx3dzcVFRXExsZKwrUWi4W4uDgUCgU1NTV0dXVJWoN1dXUEBwd/5zxXSr90cO3t7bz++usEBgZKP/qEhARsNhtWq/USNVyLxYJCofiOpIzNZkOhUFySxl6hRVHEZrOhVCql4+xKH/aCvfiYq60YPT09/P3vf8fV1ZWGhgaefPJJnJycOHToEAaDgQ0bNvCnP/1JUoJobm4mLy+P+vp6ysvLee+99zhw4AAbNmzgqaeeYs6cOezcuZOmpiYsFgu/+MUvWL9+PXfeeWefHVxnZyebNm1CFEWKi4v51a9+hZeXF6tWraK2tpYPPvgAT09PKb1Wq5VsKy0t5dNPP0UQBLZu3cqXX37Jxx9/THp6Oo2NjRw9epSf//znfPPNNyxatKjPtl0rRFGkvLycDz/8kJCQEGJiYqR4BgA5OTnk5+dTVVXF2LFjmT17Nn/961+JiYmhoKCAF198kaNHj9LT08OJEydYunQpGzZs4MEHH7ymDk4URU6dOsWXX35JcHAwhw4d4tlnn0WtViOKIqWlpXz00UeEhYURHR3N9OnTSU1Npauri7y8PJ599lnKy8upqKjAYrHwxBNP8M033zB37txrojJSV1fHypUraW5uZsWKFahUKjZu3EhxcTE9PT3MnTuXqVOnXvKbWrduHTt27ODzzz+no6ODtWvXUl5ezscff0xPTw9ffvklarWarq4uVCoVhYWFLF269KpttdNvHVx9fT2/+MUvCA4OxsnJCa1Wy+rVq+no6GDIkCEsXryYjRs3Si2NyZMnIwgCarWaCRMm8MEHH/Dggw+SkZFBRkYGjo6O/PznP6eoqIhjx47h6upKcHAwd955J3v27OH06dP4+vrywAMPcODAAc6dO4enpyePPvroVWvQ1dbW0tzczJ///GfWrVtHcnIyTz/9NA8//DC1tbW8++67ks6aIAgkJiYyfPhwvvjiC0JDQ3FxcWHp0qU0NTVhs9kQRZGWlhYiIyM5ceIE6enphIWFERDQ9/AX7u7uPPXUUzg4OPD2229TVFTEggULeOmll/jrX//6HQHFhIQEEhISWLNmDeHh4ahUKkpKSigtLSUkJASr1UpTUxPR0dGcPHmS9PR0AgMDCQ4OvmGtN5vNxq5du/Dx8SEqKophw4ZJD0lBEJg2bRrDhw/n4MGD1NbWotfrMRgMzJgxg/r6etra2mhtbSU+Pp5Tp06RmppKfHz8JY7/WnHkyBFmzJjBzJkzeeSRR1i6dCmhoaGIosiuXbvw8/Nj4MCBDBs2DEdHR+677z6am5tpamqira0NrVbLgAEDOH36NJmZmXh5eREaGnpNyj4iIoLnn3+eN998E1EU6e7uJi0tjZdffpnGxka2bdvG5MmTUSqViKJIQUEBjY2N+Pn5IYoioaGhPPPMM/zf//0fNpuN7u5uFAoFQUFBlJeX09jYyNKlS6+pnH+/HIMLCQlh7ty5fPDBB/z3f/83eXl57Ny5E71ez7x58zh06BBpaWnk5uby7LPPSi2jlpYW2trapGZ+VVUVGzduZMaMGfj4+LBr1y5aWlpwd3fn4YcfJj09neLiYg4cOMCTTz7Jgw8+SH19PXv27GHBggUYjUZSUlKuOj9ubm5YLBZOnTrFuXPnaG1tRRRF8vPz+eSTT/D19b2kJWDXYjtx4gTTpk1DoVBc0tpUKBTMmTOH5uZmJkyYQF5eHs7OzmzZsoX29vY+2aZQKFCr1RQWFlJbW8vIkSMvaclejF1sUafTkZmZydSpU+nu7mbr1q0sWrQIjUaDIAjMmzePuro6Ro8eTX5+Pm5ubmzevJmOjo6rKscrxd46bWpqoqurizfffPOSIC0mk4ldu3axe/duYmNjcXFxwd3dnc8++4zGxkYCAwOZN28e9fX1jBw5kqqqKiwWC9u3b7/mMuGhoaGcO3eOU6dOUVpaitFoBC446eLiYpqbm2lra+Mf//gHPT09dHV18eWXX1JRUUFgYCCzZ8+mtbWV8ePHk5WVha+vL1u2bKG1tfWqbbPXQzsmkwmTyYSLiws+Pj50d3dLD+qOjg52797N4sWLpdc6vn18aGgow4cPlx7aQUFBHD9+nIyMjGsm3tkvHZyDgwN33XUX//d//8eoUaPYuHEjTU1NtLa2kpeXR2JiojQ+5e3tTVBQkBTIxWq1YrPZMBgMdHd3097eTkFBAYDU74+IiMDFxQUHBwd0Oh3Ozs74+vri7e2N0Wiks7OT3NxcPD09iYiIuOr8BAQE8OSTT5KXlyfdSHtAl1deeYW6ujoaGhqkGw1QUFCAi4vLd94PtHexBwwYwAMPPEBTUxPx8fGcOHECk8lEcnJyn2wTRZHz58/z2WefSS3m70tjt00URfLy8nB3dyc8PJyqqipycnLYtGkTmZmZHDx4kLCwMB588EEaGxuJj48nPT0di8XC4cOHb4iktiAI+Pn5MXXqVJYsWYLRaKSnp0fKk0aj4cEHH+Spp54iJSWFoqIiBEHgv/7rvxgxYgTHjh0jLi6O+++/n5qaGuLi4sjOzqalpYW0tLRraueCBQuIiYmhsLCQyMhInJycsNlsCIJAUFAQU6ZMYdGiRVI99fLy4oUXXmDy5Mmkp6cTEhLCgw8+SFtbG4MHD+b48eMIgsDevXuv2r5v3zu1Wo1KpUKv19Pe3o6TkxMqlQqbzUZZWRnnzp1jw4YNZGVlceDAASwWyyXHq1QqZs+ezeTJk2lqakKv16NQKNi7d+81ccjQTx2cVqvlww8/ZMuWLZw5c4ZBgwYxffp0NBoNbm5u+Pr6EhsbS1dXF2vWrOHgwYMAxMTEkJOTw/79+6muriYsLIyRI0eiUCjw8PDA19dXGleDC47U3s1avXo1GzduxM/Pj8GDB+Pg4ICvry++vr5XnR+bzUZzc7M0Ljhz5kzOnj3LypUrWbt2La6urnh6evL5559TU1OD1WqVuioODg7YbDaysrLIycnh2LFjFBcXIwgCBQUFWCwWkpKSEASBtra2Po9zdXR08Morr6BSqcjLy6OiooLu7m52795NWVkZu3btQq/Xs27dOqqqqhBFkbS0NCZNmoRCoSAqKoo33niD+++/n/j4eKm87fEbkpKSUCgUtLa24urqetVleSXYW7ypqamsWrUKPz8/XF1dWblyJXV1dWzdupW1a9eyc+dOhgwZQmBgIO3t7WzatIm8vDyioqIAOHPmDO7u7gwZMgSLxUJXV9c1z1NXVxc9PT1otVpmzZqFn58f69evp6KigunTp3P48GHWrFmDr68vgiDw2Wef8fnnn5OVlcWgQYMAKCkpQafTMWbMGBwcHGhpabkm459dXV3s2rWLsrIy9u7di1KpZMSIEaxdu5YNGzZw2223UVtby7p16xgyZAivv/469957L4MHD2bkyJHo9Xr2799PWVkZycnJ9PT0YDab2bdvH7Nnz8bHxwetVouDg8M166b+5FpUQRBWAQuAJlEUh/Zu8wbWAwOACmCpKIptwoU+1DvAPKAHeFgUxZ+MrvJtRV+LxcL58+epq6vDy8uLoUOH4uDgIA2gBgQEEBcXR3NzM+fPn6ewsBBXV1eWLl1KXl4eBoMBJycnYmNjMZlM5OTkADB06FAMBgMAfn5+FBcXEx0dTXd3Nzk5Obi4uJCQkIBOp+PcuXNoNBqGDh161TLdNpuN8+fPU1tby+DBg6UfUG5uLgDx8fF4eXlRUFBAWFgYzs7OFBcXEx4ejouLCzabjaKiIiorKwEYPHgwERER1NXV4eTkhIeHBxUVFTQ3N5OYmIhGo7ls2/R6PSdPnpSennZbTpw4gdFoRK1WM3HiRMrKyggJCcHV1ZXi4mJpbNBeLlarleLiYiIjI9FoNDQ0NKDRaPDy8qKiooKmpiaSkpJQqVQ3ZCzObl9TUxMJCQl4enqSm5vLwIEDaWhooKKiAh8fH4YMGYJKpaKmpoaSkhKCg4OJjo6WZvw8PT1xcXHh/Pnz6HQ6hg8ffk3HjLq7u8nKysLR0ZGEhATUajVFRUUEBgbi5uZGYWEhLS0tJCQk4OLiQl5eHi0tLURERBAZGYlSqaShoQGVSoW3tzfV1dXU1taSlJR01ZHouru7pda4i4sLY8eOxWq1kpOTg6OjI/Hx8ej1empra4mLi5NmWYuLixk0aBBms1mqVy4uLowaNQqlUklVVRUDBw7EZDKRnZ1NYGAgERERP1hPrqlkeW+sUx2w9iIH9ybQKori64Ig/B7wEkXxZUEQ5gHPccHBjQXeEUVx7E8ZcbWS5Tk5OVgsFkaMGHHTvkoiIyNzefTFwf3ko0cUxSOCIAz41ubFwNTez2uAw8DLvdvX9gZ8zhAEwVMQhKCLIt1fF4YNG3Y9Ty8jI3OTcqVjcAEXOa0GLoQQBAgBqi9KV9O77Tv8UGT7K+Fava8mIyNza3HVkwy9rbU+T439WGR7GRkZmWvBlY6ONtq7noIgBAFNvdtrgYvfawjt3faT9AfhTRkZmVuLK3Vw24GHuBC1/iFg20XbnxUE4WsuTDJ0XM74W0tLC2vXrr1CU/ofoihel+7y5Zz3Wl372+e5Xnm6nGv39/P292vfarS0tFx22p90cIIgfMWFCQVfQRBqgP/hgmPbIAjCL4BKwL54bDcXZlBLuPCayCOXY4SrqyuTJ0++bKNlZGT+c+nLu4eXM4u67Ad2zfietCLwy8u+ei8ajYbIyMi+HiYjI/MfSF/e8+yXKxlkZGRkrgWyg5ORkbllkR2cjIzMLUu/1IMDJIE/jUZzzfSsbhSiKFJTU0N1dTVxcXF4eXlhs9moqKigoaGBgQMHSuKeAI2NjRQUFCAIAgMGDCA8PByr1UphYSFdXV3Ex8dLC9ojIyPx8fGhoaFBkvnpq2329YI6nY4hQ4bg5uZGW1sbRUVFeHh4EBsbK2l8VVdXU1paClyQgRo+fDgmk4m8vDw8PDyIjo7GYDCQn5/PgAED8PX1pampCY1Gg4eHxw29jxaLhfz8fAICAvD396e9vZ2ioiLc3d0ZNGiQJEkliqIkNhoZGUlQUBAVFRVUVFQA4OHhwZAhQygvL0cQBGJiYjCbzTQ1NREaGiqJOcjcePrtndDr9fzxj3/kX//6FxaLBVEUsVqtGI1GzGYzVqtV+nEajcZLpIZsNpt0zMWfTSbTJWnt3+3nslgsmM1mTCbTJde62nf0LBYLqamp/P3vfyc7Oxu4IIf0z3/+k3PnzvHqq69eIg9z9OhRNm7cSGVlJR0dHdhsNrZs2cJXX33FmTNnaGxsZPv27Zw8eZLVq1fT1dXFhg0b6Onp6bNtoiiyadMmvv76a86ePUtjYyNtbW289dZbZGRkcPbsWcxms5S+o6ODyspKjh8/zjvvvIPJZGLFihWkpqby8ccfc/r0abZv387p06f54osvaGtr45tvvrki264loihSWFjIb3/7W44cOYLJZOLNN9/k5MmTrFix4hLZo6amJl599VWysrJ44403aGpqkvKdmprK8uXLKSsrY8uWLWzdupWSkhJSU1PJy8u7gTmU+T76bQuuuLgYV1dXurq6qKurIyAgQJK30ev1zJgxg6ioKD7//HMsFgtTp07l9ttvRxAEmpubWbNmDc899xxnzpyhtLSUESNG8OWXX2Kz2Zg8eTITJkxg7dq1tLa2EhMTw5133smHH35Id3c30dHRtLe309DQQEhICI8++uhVKUY4ODhwzz33SIq8cEEq3Nvbm8mTJ5OdnS0JBdpRKpU4OzsTGBiIwWBgx44dTJ06FV9fX/z9/ens7CQ8PJzMzEyOHTtGVFSUpJzal1ZSW1sbBw4cYOLEifj7+xMQEMCxY8fo6enB39+fmJgYVCqVlH7o0KHEx8ezatUqIiIi6O7upqCggL///e9kZGRw6NAh3NzciIyM5PTp06SnpxMeHk5gYOAVl9/VIooiOp2OPXv2MHr0aOnBqNPpGD9+PEaj8RIxzsLCQvz9/Xn88cd57bXXOHXqFPPnzychIYHPPvuMqKgozGazJEFUVVVFQUEBjzzyyE3d07gV6ZctOFEUSU5OZsGCBSQlJXH8+HEqKiqorKzk17/+NYGBgbS2trJmzRqioqKYMmUKmzZtkloJXl5e9PT0UFxcTHJyMoMHD+arr74iJiaGqVOnsm3bNnp6eggICMDLy4tt27bR3NyMVqtl4sSJzJw5k7y8PMaMGcOsWbOuOgCGIAg4ODhcUvmDgoJoaGhg1apVuLm54eLiIu0bOXIkixYtQqvV8vbbb9PR0UFBQQGOjo7k5eWxfv165s2bR3t7O7fddhtFRUU4ODiwcePGPgsF6nQ6zp8/j5OTE+fOneObb76hoqKC+vp6vLy8+PDDD6mpqZHyYVf0PXPmDJMnT8ZoNOLo6IharcbLy4uuri7mzZuHVqtl7NixFBcX4+joyDfffEN7e/sNWbEiiiKHDx9mwIABkqCnWq0mJCSEdevWUVhYSGhoqJQ+ODiY6upqTpw4QW5uLo2NjZLKclZWFpMmTSI2NpaIiAjCwsIoKSkhLi6OnTt3cu7cOXlVTj+iXzo4rVbL0aNHOXbsGDk5OSQnJ9Pd3S1pn9mFKzs6OmhsbKS5uZmpU6dKLQ2VSsXMmTP5/PPP0ev1REVF0dHRQXNzM42NjUybNo2cnBxKS0uZMWMGLi4umEwmHB0dCQ0NxcfHh2eeeYb6+nree++9S+Str4aLHdzRo0clRV+7AurFar3Tp09n4cKFNDc3IwgCERERzJgxg2nTplFaWkp4eDj33nsvNTU1DBs2jNOnT6NUKjlw4ECfbHJycmLAgAHMmDGDiRMnUlVVRWBgIImJicyYMYPAwEDq6+sl20RRJDc3V9L612g0GAwGzGYzbW1tuLu7ExYWxrJly6iurmbo0KGcPHkSlUrF4cOHr0k59hWj0cjx48fJzMxk7969JCcnU1BQQGVlJf/93//NggULJNtEUWTgwIE8/PDDlJeX4+npyYABA6R8e3h4EBYWhqOjIwsXLiQsLAwnJyeKiorw9/dnx44dkuagzI2n3zk4URTJyclhxIgRvPjii/zud7/Dw8MDi8WCwWBg5cqVHDhwAEEQWLhwodRqc3V1lQZ3BUFg+PDhVFZWMnLkSDw9PZkzZw5dXV2Iooijo6MkT24XDLy4ldXV1cWJEydQKBRSl/JqOXz4MKdOnWL//v0UFRUxaNAgzpw5w5o1a9Dr9fj7+/PJJ59QWVnJ/v37efvtt/nXv/7F8OHD8fb2ZtasWXzyySds2LBBWvWRn5+PUqkkMTERuBDcxsfHp092eXt7M27cOFasWMG2bdsYN24c48ePp7GxkQ8//JCuri4GDhzIypUrqaioQBRF0tPTmTx5MgqFAk9PT4YOHcr777/Pnj17mDFjBgqFgry8PNRqtaToW1NTg5eX1zUpy76i0Wj47W9/y/PPP8/kyZMZP348ERERODs7s2bNGo4cOSLJkG/cuFEKj9fU1ERISAijRo0CIDU1lVmzZkn1TKfTkZaWxuzZs/H396e0tPSSeihz4/lJwct/BxcLXoqiSGNjIwqFAj8/PwRBoLa2FkdHR4qLiykpKeH06dMsW7aMUaNGUV5eTmtrKyEhIVLkJpvNRkFBAStWrODll18mODgYq9VKWVkZbW1thISEEBAQQElJCWazGScnJ0JDQ2lubsbX1xeVSkVZWRnt7e1ERETg7+9/1WMr58+fl1pjkZGR+Pv7SzEtIyIi8PPzo7y8nICAAHp6eigvL8fJyYlBgwah0WgwGo0UFhai0WiIjo7GwcGB5uZmHB0dcXV1pa6ujtbWVuLi4i4ZM7scDAYDhYWFODk5SbEtGxsbqa6uZsCAAfj4+EhKyk5OTlRVVeHv7y/FqtXr9RQWFuLh4cGAAQMQBAGtVitJzF+NbdeahoYGHBwc8PHxob29nbKyMtzc3Bg4cCBdXV3odDqCg4M5f/483d3dDBo0CDc3Nyn0YFBQEE5OTsCFlmFLS4s0TmpXYfb29r6hebzVuaaKvv8OLkfR1x6nwB5QeMaMGVK0nm9jNpvZvn07AQEBjB8//poFkZWRkbnx3JIOzm6nfZbwx1pUF+dJntWSkbm1uKaS5f0Fu6O6HIclOzUZGRnoh5MMMjIyMteKftOCu1aRrGVkZGTs9AsHp9VqWbVq1Y02Q0ZG5iZAq9Vedtp+4eA8PDxYsGDBjTZDRkbmJuD999+/7LSXI1n+fZHtXwEeB+zx/v4oiuLu3n1/AH4BWIHnRVHc91PXUKlUBAUFXbbRMjIy/7n05V3Ky5lkWA3M+Z7tb4uimNj7Z3duQ4B7gfjeYz4UBEF+CU1GRuaG8JMOThTFI8DlruBeDHwtiqJRFMVyLgSfGXMV9snIyMhcMVfzmsizgiCcFQRhlSAI9kWGNySyvYyMjMz3caWTDMuBv3Ehov3fgH8Aj/blBKIorgBWwIWVDBfvM5vNZGVl0dnZiUqlIjEx8YarwV4NoijS1NREYWEhAQEBxMTEoFQqsdlsFBUVoVAoiI2NldLbbDaqqqooLy8nODiY6Oho6Xt8fDwBAQF0dHSQn59PZGQkAQEB1NTU4ObmhqenZ59ss1qtFBcXU19fj7+/P3FxcZJ6b1FREUqlkpiYmEvK3mq1kp2dTVBQEEFBQTQ3N1NQUICfnx+DBg2iu7tbUvQNDAykrq4OZ2dnPD09b9p7+O+itbWVrKwsRFEkLCyM2NhYampqKCkpITo6+hJ1a4vFQlZWFh0dHTg5OTFixAgsFgu5ublYLBZiY2Px9fWlpKQEnU7H0KFDUSqVlJeXM3DgwD4vYWxvbycvLw+r1crgwYPx9fWlp6eHs2fPotFoGDp06HeWT9qVoUeOHIkgCBQWFtLS0kJUVBRBQUGUlpbS1dXFsGHDgAvaepGRkddseeUVteBEUWwURdEqiqIN+IT/3w294sj2F6PT6fjXv/5FeXk5zc3N6HQ6dDodra2tWCwWOjs7pc+iKNLd3U1raytdXV2YTCa6u7uxWq1YrVZ0Op2k3tvS0iJ97+7upquri5aWFkm1156ms7MTvV6PwWBAFEX0ej1Go/FKigq4IG755ptvkpeXx0cffSRphtXX1/PKK6/wxRdfXJK+qKiId955h/Pnz1NaWorFYuH06dO89957nDp1ClEU2bZtGzk5OXz22We0t7dLKhh9xWKxUFRUREVFBcuXLyc9PR1Asu3rr7++RFFFFEXy8vJ4+eWXOXLkCF1dXbzxxhucO3eOjz/+mKysLLZv305ubi6ff/45LS0tbNy4UZYQukzy8vL4+uuvaWpqoqurC0C6z/v3778krdFo5P3336e8vJyWlhasVistLS0UFhZy+vRpXn/9daqrq1m/fj1Hjx4lIyODjIwMTp8+fUW2abVa6dxvvPEGOp2OL774gsOHD/PNN9+wb9+l84kmk4nVq1fzl7/8he7ubnJycli+fDklJSW88cYbFBcXs2HDBo4ePUp6ejonT568Ytt+iCtqwQmCEHRRxPqfAbm9n7cD6wRB+CcQDMQAJ6/kGq6urowbNw4fHx+ys7PZtWsXI0eOJDIykv379yOKIuPHj+e2227jnXfewcHBgerqah5//HHS0tL42c9+hrOzMytXruSZZ57h66+/prKyEoVCwWOPPSYJZAqCwJAhQ5gzZw6rVq2ivr6e8PBwIiIi0Gq1LF26lNWrVzNmzBhGjhx5JVmhoaEBvV7PY489xldffcXBgweJi4tjx44djB079pIfvyiKpKSk4OTkhNVqJTQ0FJVKxeLFi6mpqZFiBhiNRgIDA2lsbOTIkSPEx8fj4+PTZ0VftVrN3Llzqa+vl9SSzWYzu3fvZvTo0Ze8gG1/MBw4cIDbbrsNQRBoa2ujp6eHRx99lD179nDgwAF8fHwIDg6mubmZY8eOERsbi5+f3xWV3X8aSqVSeojbVUnmzp2LXq+XHN7FqFQq2traCA8PR61WM2DAAH72s5+Rl5dHRUUFZrMZjUaDt7c3jY2N1NbW8sADD1xRS3rgwIH4+PiQm5tLZWUlOp2OnJwc/uu//ovm5ma+/vpr5s6di4ODA6IocubMGSwWiyQm2tPTg6enJyNGjCAnJweDwYBarcbHx4f6+noaGhq47777rmkr/ydbcL2R7dOBWEEQanqj2b8pCMI5QRDOAtOAFwBEUcwDNgD5wF7gl6IoXtESBXurZMeOHbS2thIZGck999zD1q1bSUpKYvbs2ezbt4+jR4/i4+PDiy++iJubGwaDgba2NiluQ1tbG6WlpZw8eZIFCxbg7+/P/v370el0TJ48mfvvv59Tp05x7tw5Kisrefnll3nooYcYNmwYubm5VFdXU1RURFRU1JVkA7iguWaz2UhJSeHEiRPodDqys7Mxm83ExMRIrU240D2tra3FZDIRExPDBx98QGtr6yVNdrsWntlsZtKkSZSXl2MymVi3bh19Hc+0699t2bKF0tJSAgICyM7ORq/XExsbe0msC7igaxccHExISAg2mw1XV1dsNhuHDh3i2LFjtLS0sGjRIgwGA+PGjaO8vBybzca6detoaWmR1W5/gkGDBvHUU08RGBjIm2++SVdXFwqF4nt/9BqNhqeffpopU6awbds2Tp48iSAI5OXlsWnTJry8vAgODua2227D3d2dnp4eIiMj2b17N6mpqVgslj7ZJggCOTk5bNq0CV9fXymOiUajwd3dXRI+BWhpaSE5OZl58+ahUCiwWCwEBQWh1Wr56quvcHV1ZeDAgdx22224ublhsVgIDw9nz549HDp0qM+2/RCXM4u6TBTFIFEUVaIohoqiuFIUxQdFURwmimKCKIqLLmrNIYriq6IoRomiGCuK4p4rNczX15fnnnuOxx57DFdXVwIDA1GpVPT09FBTU0N9fT2zZs1CFEWcnJxwdHTExcUFQRBQqVRS8BiTySTp75eVlUlPEKVSia+vr9RSMhgMaDQanJycUKlUBAQEEBwczPLlyxk2bFifo1VdjI+PD8899xytra24u7sTHh5OdnY2JSUlbN68mePHj1NSUoLFYkEQBIKCgkhKSmLMmDEolUo6OzsvUVMBCAwMZPHixZSVlTF8+HDOnj2Ll5cXKSkpfbJNFEU8PT155plnmDVrFocPH+bkyZOUlZWxadMmjh49KnWTrVYrmZmZnDlzhl27drFv3z6MRiO//vWvaWtrw9XVlUGDBhEQEMCSJUsoLS1l+PDhnDlzBi8vL1JTU6+4DP9T8PLyYvTo0UyfPp3u7m5J0PXi+293LHax01GjRjFkyBDq6uqw2WyMHTuWP//5zzQ0NNDY2MiECRMYOnQoer2elpYWwsLCSE9Pp6WlpU+22Ww2Jk6cyB/+8AdKSkrQ6/Wo1Wp0Oh1arRYXFxdUKhUWi4Xq6mrKy8v54osvyMrKIiUl5RIV666uLiorKxk/fjzDhg2jo6ODjo4OgoKCOH78eJ9t+yH6xUqGb2N3UkqlEgcHBxQKBUqlEpVKxdKlSzl8+DBqtRq1Ws2YMWMk9dtz584xd+5cRo4cyZYtW3BxcUGn0xEVFcXw4cOpqanBxcUFZ2dnSb3XruQbFxfHgQMHePPNNwkJCeHuu+9m6tSpPPfcc/zqV7+6KpVWq9XKuXPnKCkpAeD222/Hzc0NvV7PsWPHyM7OZsCAAXz44YcsXLiQGTNmsGLFCoqLiwkMDCQgIIDU1FSOHj1KQUEBwcHBjBgxgnPnzuHk5MTw4cNJS0sjPz+fsWPH9sm29vZ2Vq9ejV6vp6amhocffpjBgwdLtuXm5hISEsLHH3/M7bffzgsvvIDRaGTDhg24uLjg6enJ8ePHKSsrw2g0Mm/ePOlJ7+rqSmJiIhkZGRQUFDBhwoQrLsP/FI4cOcKxY8doa2sjMTERLy8v0tLS2LNnDwaDgdjYWEJDQ9m2bRv33nsvq1atwmAwUFtby8svv8zJkyfZt28fJpMJf39/AgMDMZlMHDhwgFmzZlFZWcmZM2dQKpWScOflkpqaSlpaGnq9Xgp1OWnSJD788EMMBgN33HEHlZWV7N27l8cff5w33niD9vZ2mpubmThxIvX19axatYqmpiYcHBwICQnBZDJJLb3S0lKys7NRKpU4Ojpek/Lsl3pwVquV5uZm/Pz8UCqVdHV1YbPZ8PDwwGq1SmNaXl5eeHl5odVq0el0fP3110ydOpXRo0dTV1cnOUU/Pz/0ej0NDQ0olUoCAwPR6XS4ubmhVCppbW3F19eX7u5uGhsbcXZ2xsPDg3379pGfn8/vf//7q4qqJYoizc3NdHR04Ofnd8mMcHd3NwaDQRoj8fT0RK1WS5MrAQEBuLi40NTURFtbG4Ig4O/vj5eXF52dnTg4OODk5ERrayvd3d0EBQX16U1vq9VKY2Mj3d3dl8S7+CHb7BWvvb0dhUKBq6srLS0tdHR04OPjI83idnV1fce24ODgqyrH/wS6urpoampCoVAQFBSERqOhpaVF6t77+vpKcWt9fX0v+S34+PjQ09NDQ0MDgiAQEBCAs7MzNpuNtrY2vL29sVqt1NXV4ebmhpeXV5/Gu3Q6naS2bVd3NpvN1NfXo1QqCQoKwmw2097eTkBAAIIgXPJbFgRBqmu+vr54eHhItvn4+GCxWC7LtltS8PKnEEWRkydPEhISckmEpCulubmZ/fv3M3XqVEkKXUZG5sZzSwpe/hSCIPS5e/Zj+Pn5cf/991+z88nIyPz7kQUvZWRkbllkBycjI3PL0i+6qPZpbxkZGZmfoi/zBv3CwTU3N7N8+fIbbYaMjMxNQF9eZu8XDs7X15dHHnnkRpshIyNzE/DZZ59ddtp+4eDs71PJyMjI/BR9eelenmSQkZG5ZZEdnIyMzC1Lv+iifh8/NFPyUysKLj6uv6w+uJxZn4tt7css0bXI47fL7Puu/0P2/VD6H9p/o+7Jt238Mbt+KP+Xe1+uNo/fZ9dPbbtcGy9O01c7L+d6P2RbX7lW9aRfOjhRFKmpqaGgoAC1Wk1CQgIODg5YLJafXD/X0tJCW1sbUVFR/cbB2Ww2Tp48iVarxdHRkQkTJmCz2cjMzMRkMhEfH09ISIhkr8FgICcnB51OR3x8PP7+/hQUFKDT6UhMTEQQBCoqKoiOjr5q5VO78GZubi4BAQEMHTqU7u5uDh8+jEKhICwsjGHDhkm22Ww2ysvLOX/+PKGhocTHx9Pe3k5WVhaenp4kJCRI4oZRUVGEhIRQXV2Ni4uLpG/270YURdra2sjKysLHx4fhw4cDUFBQQGVlJSNGjMDf31/Ko06n4/jx41gsFkmdw8/Pj9LSUkpKSoiIiCAmJoazZ88iCALDhg3DZDJRV1d3TeqdXUvNzc2NmJgY4MK64MzMTEaOHImbmxtw4V7k5OTQ1NREYGAg8fHxdHd3SzpsdvXnwsJCOjs7SUpKQqlUUlJSwqBBg/pcdywWCzk5ObS2thIXF0dYWBhms5mzZ8/S0tLCsGHDCA4OlmwrKiq6pHzr6urIzs5GFEXUajXjx4+nqqqKzs5ORowYgSAIlJWVSYrX14J+6eCam5t57bXXGD9+PIBUuaqqqrjrrrtwdnaWpJA8PDxoaWnB0dERPz8/2tvbqaysJDQ0lJ6eHnp6eq5IyvtaYjKZWLNmDRMnTsTX1xe4sKi6trYWo9HIli1beO211/DyuhDaYu/evWRkZBATE8O+fft49tln2bx5M56enlgsFsxmM62trVLlvxq6urp46623iI+PZ9++fdx99924u7vzzTffcMcdd3xH8LKwsJCVK1eSmJiIQqEgKiqK999/Hz8/P8rKyujq6qKiogKFQkFGRgaPPPIIW7ZsYdmyZVdt69XQ0NBAcnIyJpOJN954AwcHB86ePcu2bdsQBIHbb7/9kvQWiwWDwcCKFSuIjo6msbGRtWvXkpCQIAkOHDhwAEEQUKvVFBcX4+7uflW6gXChjMvKynj11VeZOHEiL774IqIosmPHDpYvX87HH3/M4MGDgQtCCRUVFRiNRjZt2sRjjz0mScR3dnayY8cOfv3rX7Nlyxa8vLzQ6/UoFAq0Wi2DBg3qs23Hjx9n27ZtJCYmsm3bNl599VUOHjxIdnY2cXFxtLS0SA7OYrFw7tw5Nm3ahEKhYPbs2dhsNmlBfUpKCmFhYWzdulWyzS6gei3qtZ1+OQZnMBjo7u4mJiaG22+/nZiYGEpLSzl16hQ5OTmsX7+e119/nYyMDM6cOcPGjRt5//33OXPmDG1tbVRWVkqy2nv27OGdd96hra3thubJwcGB8vJydDqdpLwwZ84cYmJivqNjL4oiPj4+hIWFoVKpsNlsODg4oNFo0Gq15OTkMGPGjGvSQm1ubsZgMLBs2TJGjRrFwYMHEQQBg8FASUnJd2asDh06BEBTUxOenp50dHRQU1PDz3/+c+bPn8+xY8ewWq04OTkhiiKpqakMHTpUcuw3isGDB3P//fdLEkGCIHDPPfd8r4STq6sr8+fPZ8iQIQwdOpSIiAiSk5MRBAGtVourqyuiKKJSqVCpVFRVVVFXV8eECROu+p4YDAZ27tzJ5MmTJeWVwsJC6urqSEhIuCStSqVi/vz5JCQkSA/H8PBwZs+eTVRUFCqVClEUcXBwwNHRkZaWFrKzs5kxY8YVyX/ZFX3s6sF6vZ49e/bg5uZGS0vLJY0ItVrN0qVLGTNmjCRLFhoayqJFi/Dx8WHGjBm4urpKtjU3N5OTk8PUqVOvSprs2/RLBxcSEsITTzzBzp07+f3vf09aWhpDhgxhypQpzJw5E71ez9ixY1myZIkkctne3s6+ffswGAxSbIbg4GAeeughAGpqam5YftRqNc899xxLlizhxIkTHDx4EFEUqaysJCUlBUdHR6ky2yVxysvLOXr0KC4uLgQEBDBv3jxiY2Npbm5m4MCBfPXVVyQnJ1/SwroSvL29cXBwYO/evaSmpqLX6wkLC+PPf/4zI0eO5J///OclUtl1dXU4ODgwfvx4VqxYQWNjIxqNBoVCgbOzMwaDgUWLFuHh4cH48eOl1oQ97Y1Qr7H/wC53LFAQBEmlePz48Tg5OdHY2IiTkxOjR49mxYoV+Pj4MGHCBEaOHElJSQlRUVF89tlnHD9+/JIYFn1BFEWOHTuGo6Mj/v7+mM1muru72bp1K5MnT0ahUGAymS4ZQ7MPJzQ3N0vOu6ysjNTUVBwdHQkMDGTOnDlERkbS1dVFZGQkGzZsYO/evX1ePeTt7U19fb10bvtQUlhYGL6+vnz66afSOX9o7NBoNHLixAluu+02goODmTNnDgMGDMBgMBAREcGGDRvYvXu3pAx8tVyOZHmYIAiHBEHIFwQhTxCEX/Vu9xYEIVkQhPO9/716twuCILwrCEJJb1jBEX01ShRFEhIS+POf/8ySJUs4cOAAoihitVqlJ6e9RbBu3TomTpzIrFmzpCAxdjw9PVGpVKjV6qt2BFdLWFgYgwYNIjo6moaGBsxmM0OGDOHFF19Eq9XS3NxMT08PZrOZY8eOceedd/KrX/2KqqoqWlpaGD58OD4+PiiVSlpaWhg4cCAZGRnfq9PfFzw9PXnuueewWq34+PgQFRWFWq0mNjaWYcOGYbFYMBqN6PV6rFYr4eHhDBo0iKFDh6JSqTCZTFKLu6GhAV9fX7y9vZkyZQr5+fkkJSVJEbaOHj16jUqz79hsNkl+3V6P7Nvs281ms1SHdDod+fn5jB49GkEQCAwMJC4ujiFDhkiOxt468ff3Jz8/nxEjRnDw4MErDrBjtVrJz8+npKSEXbt2kZqaKrXetm7dysmTJ0lNTcVgMNDT04PVasXR0ZFHH32UiRMncvr0acxmMwkJCbzwwgtotVoaGxsZPnw4QUFBWCwWmpubGTJkCFlZWWi12j7Zd/jwYWbMmMHzzz+PVquls7OTgQMHEh8fz7Bhw+js7JTqir18Ly5nURQpLS1FFEWioqJQKpUkJCQQHByMXq+nra2N2NhYTp8+/W9V9LUAvxFFMUsQBDfgtCAIycDDQIooiq8LgvB74PfAy8BcLgSbiQHGciHEYJ90jFpbW1mxYgWCIFBbW8vChQsJCwtj+/btBAUFoVQqUSqVCIJAZGQkaWlpklClUqlErVajVCrRaDQA0vcbRXt7O59++qkky/yb3/yG7Oxs9uzZg8lkYsCAAfj5+fHJJ5+wcOFCRo8ezbZt2zhy5Ajh4eH4+PhgNBpJS0tjzpw5VFVVcfToUVxdXa9a+dRms5GVlcW5c+cQBIE5c+aQmZnJ7t276e7uZsSIEbi6uvLpp58yd+5cZs2axQcffMBrr73GgAEDiI+PZ8KECbz11lsYDAaefPJJFAoFOTk5+Pn5kZCQIA0lTJ8+/RqVaN+prq7miy++IDc3lw0bNrB06VL2799Pamoq586dIzQ0FKPRSFFREffddx8lJSX4+/sTFBQEwIIFC/j44485e/YsMTExBAQE0NnZSU5ODvfeey/Hjx/nyJEj+Pv790lw9GKUSiVPPPEEZrOZ/fv3U1NTQ0JCAq+99homkwmj0cjUqVNpbGxk27ZtLFu2jM8++4yenh5aWlp4+umnycrKYufOnVgsFiIiIggKCsJkMpGamsqcOXOorq4mLS1NkvjvC+PGjeOLL74gPz8fPz8/goKCmD9/PqtXr8ZqtTJt2jTq6+vZs2cPTz75JHv27OH48eOUlpYSHBzMsGHDOHXqFOPGjZNam0ajkUOHDrFgwQLKysrIyMhAo9H02bYfos+Cl4IgbAPe7/2bKopivSAIQcBhURRjBUH4uPfzV73pi+zpfuic3xa8tNlstLe3093djUajwcfHB7gQtkyhUKDRaFCr1Tg6OqLX62ltbcXJyUlyaiaTCY1Gg9FoxM3Njc7OTpycnL4z1vXvwmq10traisFgwN3dHXd3d4xGI62trcCFpr9Go5HiGthVhk0mE97e3jg6OmKz2SQVYpvNJmngu7q6XtW4jyiKdHR0oNPp8PT0xMXFBaPRSEtLC4Ig4OPjg1qtlmxTqVS0t7fT09Mj2WY2m6UZYnvs056eHul+dHV1odfr8fX1vWEPGoPBQHNzM6IootFo8PPzk5SGBUHA29sbpVKJyWTC3d1darHay1cURSnfPj4+aDQarFarNIllNptpaWnBw8MDJyenqx6Ls7fQvn19+/ifTqfDw8OD1tZWjEYjrq6uuLu7S6EvFQoFXl5eaDQaRFGkq6vrquuOXX3XYDDg6ekpqQVrtVppaMVeTz09PWlra6O7uxu4EJfE2dmZzs5O1Gq15ODsoT3d3d2xWq1otVqcnZ1xc3P79yv6CoIwADgCDAWqRFH07N0uAG2iKHoKgrATeF0UxWO9+1KAl0VRzPzWuZ4AngAIDw8fWVlZedl2yMjI/OfSFwd32ZMMgiC4ApuAX4ui2HnxPvGCl+xTU1AUxRWiKI4SRXGUHDNTRkbmenBZDk4QBBUXnNuXoihu7t3c2Ns1pfd/U+/2axLdXkZGRuZquZxZVAFYCRSIovjPi3ZtBx7q/fwQsO2i7T/vnU0dB3T82PibjIyMzPXicmZRJwIPAucEQcju3fZH4HVgQ2+k+0pgae++3cA8oAToAS5L6K0/RPeSkZG5tfhJB9c7WfBDA3ozvie9CPyyL0Y0Njby9ttv9+UQGRmZ/1AaGxsvO22/iYt68uTJG22GjIzMTcCYMWNuvrio13L9mYyMjAz007WoMjIyMtcC2cHJyMjcsvSbLurF/Jhi7PUWsbwcRdRbjZ9S6P2+PF+saPFT9+tKFWSvFVeixHs5ZfJjZXA1XO79+Her5l6NMvUPqYtcqXL35dIvHZzJZOLQoUMMGjSIyMhIGhsbSU9PZ/bs2ddkEe6PSVU3NjZy6tQpRFEkKSkJR0dHkpOTUalU+Pv7S4vPbxVHZ5e8yczMJDQ0lBEjRqDT6Th27Bgmk4kRI0YQHh5+SX4tFgupqalERkYSERFBYWEh+fn5jBkzhvDwcNra2jhx4gRxcXEMGDCAsrIy3N3d8fX1vSHlJooi586do6CgAFEU8fDwYMaMGVitVtLT0+nu7mbixIl4e3sjCAIWi4XMzExqa2uZPn06np6etLS0kJGRgZubG5MnT8ZsNnPy5EkEQWD06NEYDAaqqqoYPHjwVeXRZDJx6tQpmpqaGD58OJGRkZw/f56zZ8+SkJBwiRikwWAgJSUFnU6HxWJh7NixeHh4kJqaitVqZeDAgYwYMYKzZ8/S3t7O+PHjUSqVFBQUEB8fL0l0XS7nz58nKytLWs87Z84ccnNzKSkpQa1WM3XqVEnlx2q1cu7cOc6fP8+ECRMICQlBFEWqqqrIysrC39+f0aNHk5ubS1tbGxMnTkQQBAoLC6/Ith+iX3ZRe3p6ePfdd1mzZg02m43Nmzfz5ptv0tbWhsVikQQG7QqhNTU1VFRUYDKZsFqtNDc3U1tbK6X5NjqdjvT09O/s0+v1vPXWW7S2tmK1Wqmvr6esrIwDBw7g4eFBTk4Or776Kjqd7t9VFNedzs5O/v73v6PX6/nmm29IT0+np6dHkkD629/+dkl+7XLab775JqdOncJqtVJWVsauXbukyr9lyxZ0Oh3r16+nsbGRHTt23NAHgiAIuLi44OvrS2NjI3v37sVqtfLpp5+Sk5MDXKhzdiwWC8XFxXz11VfU1l5YhKPVajlz5gzr1q0DoLi4mIyMDDIyMigoKGDv3r1otdqrzufx48fZtGkTAK+//rok4Hro0CGOHTt2SVr7gnofHx927txJbW0tRUVFpKen4+fnh6urKw0NDezbt4+ysjKOHTvGkSNHJMXlvuLk5ISfnx8Gg4GtW7diMpnYuXMnBoMBHx+fS5ySxWKhvLycbdu2kZeXhyiK1NbW8u6772Iymejp6aG6upr9+/dTUVHB0aNHOXLkCJWVlde0rvRLBwcXRC97enooLCykrKyMwYMHY7VaWbduHZ999hnvv/8+R44cobm5mT179rBx40bWrl1LZ2cnL730Etu2beOjjz6SNODhwtMxJyeHQ4cOsXbtWtLT0ykrK5McncViQafT4eTkxPDhwyWd+LCwMKZPn86TTz6JyWSioKDgRhbNNUWr1WI2m1m8eDFJSUmkpqYSGBjIuHHj8PDwwN3d/ZLuRUdHB6mpqUyZMgX4/6qydt00uKBebJfHPnz4MImJiZIizI1AEASio6OZOnUqRqORuXPn0tnZyfHjxzGbzVRVVUnSWgCOjo7ce++9REZGSsfHxcVx1113SfJUSqUSq9WK1WqlpqaG5uZmxo7tkyrY96JSqXB0dEStVuPm5oZSqWTmzJlMmzbtO2k1Gg0TJ05k6NCh+Pj4MHToUEkv8OTJk/T09KBQKLDZbJhMJtrb28nLy2Pq1Kl9CqJjJywsjGnTpqFSqZgyZYokT5afn09BQcElDQaNRsOSJUtITEyUtqWnp9PR0SE1RhwcHCTb2trayM/PZ8qUKSgUimv24n+/dXAajYakpCQ+//xzwsLCcHNzQ6fTsXfvXkaMGMGgQYNITk7GYrHQ09NDZ2cnO3fuxGg0olKpuPvuu5k4caL0hIb//1QpLS2lubmZ4uLiS1Rm3dzceOGFF6ivr+e1115j3bp1klAfXPjh2mW6bxV8fX1Rq9Vs3ryZAwcOYDQapSAt3+fI9+zZQ3h4OF5eXhiNRqlSXzwetWTJEqKiohg7dizNzc00NDTw7rvvUltbe0NXrLS1tXH+/HnGjBmDTqdDq9Uybtw4FAqF1Gq6mB+zNTo6mtmzZzN58mTOnz9PeHg4H3zwAQcPHrxiRV8AZ2dn2tvbKSgokJSIfwxRFElPTycuLg4vLy+GDh3KH//4R8aPH89HH32ERqPhjjvuICkpiY6ODgYMGMDq1avZunVrnxV94ULwm5MnTzJp0iRUKhWPPPIIjz/+ON3d3axfv/577bNjF9icP38+27Zto6enhzvuuIPExES6u7sJDw9n9erVbN68+d+n6HsjSUxMpKGhgfHjx3+nFeHt7c3ChQtJS0vDzc2N+++/H2dnZ0lDy64Xd7HEs5OTE/Pnz+fee+9lwYIFLFu2jDFjxkjNdavVir+/P08++SRPPvkkaWlpWCwWSe21vr6eqqoqIiIibliZXGvc3d15/vnn8fb2Jjg4mJiYGIxGI5GRkTzyyCNotVra2tokGfiqqipOnz7Nnj17OHDgAE1NTZJCrtVqxWaz4e7uzsiRI8nLy2PEiBGUlJQwePBgjh8/fsPyKYqiNC7o6emJt7c3ERERDBgwgMjISNrb2zGZTHR3d1+i+mvPk304xGazYbVaJTXanp4eQkJCKCoq4rbbbuPYsWNXrOgLkJaWxvjx43niiSdoaWlBq9VK17zYjq6uLkRRxGQykZaWxqRJk1AoFKjVakJDQ4mOjpZaR3FxcVIkrqamJkaNGkV+fn6fFX0BSktLUalUREREIIoiXl5ehIeHExUVJfUG7LbZbbb/Dw0NJSIigoEDB+Lk5IReryc2NlbS02ttbWX48OHk5eVJWolXS7+cZFAoFLi7uxMcHMxbb72Fo6Mjbm5ueHh4cMcdd5Cbm4uLiwteXl5ERUWxY8cOOjs7JaVfd3d3SRjT2dlZco7d3d18+eWXkmT43//+d0aOHMns2bNxcHDAYDDwxRdf0NHRQWtrK9OmTcPZ2ZmCggIpNsHtt98udV1uBewhDc+ePYuTkxNz587l7Nmz7N69G5PJxNChQ/Hy8mL16tXMmTOHX//619hsNjZt2oSTkxO+vr4kJydz7Ngx3N3dCQ8PZ9SoUeTk5BAaGsqwYcM4efIkWVlZzJ49+4bl0x7ebt68edLY1e23387y5csxGo08+OCD0qD4z372MzZs2EBBQQFr167l+eefRxRF1q1bR3FxMV9//TV33303er2eoqIi7rnnHjIyMjh8+LAUKOhKue2221i9ejWlpaWS0vPRo0fZt28fJpOJ6OhoIiMj2bFjB0899RRNTU04ODgQHR0NwJkzZ9i+fbsUt8TPzw+TyUR6ejpz586lpqaGI0eO4ObmJjm9y0UURU6fPs306dNRq9UYDAY+//xz6urqaGtr44knnqCqqoo9e/bwxBNPsHv3bjIzM6moqCAkJISJEyeSmZnJ//3f/+Hn58egQYMkpeqFCxdKcUjc3NxwdXW94jK8mH6zVOvbir52BVJ7f7yzs1PKdGtrqxQjVa1WS95epVLh6uoqKd+azWYsFovk5OyKoRc3f11cXCQVWlEU6e7uprOzE5VKhbe3t3SMKIq4uLhIzvNWwV62PT09eHp6Sgq99ihkXl5eqFQqOjs7cXZ2ln68PT09CIKAo6OjpHQL4OHhgYuLC3q9HgcHB1QqFd3d3ej1enx8fG5Y2dlsNqkOOTg4SC2M1tZWVCoVnp6eGI1GzGYzLi4utLS0YDKZAKSZwZaWFimmp4+PDzabDYPBgIuLC1arVVL01Wg0VzxQbu+h6PV6SZG3s7NTmuhxdXXF2dmZnp4e3N3dpTgSdmdlH89SKBSSSrG9XtvfQLAr+l788L9c27q6uqQxQnvd6e7uxtXVFTc3N2nIyM3Njfb2dvR6PYCkAKzX6+no6JDyZrPZ6OnpkZSKL8e266boe734toOTkZGR+SGui6KvjIyMzM2G7OBkZGRuWWQHJyMjc8vSb2ZRr+bdIRkZGZnv4ycdnCAIYcBaIIALkbNWiKL4jiAIrwCPA829Sf8oiuLu3mP+APwCsALPi6K478euISv6ysjIXC7XVNG3N2JW0MWR7YElXIjBoBNF8e/fSj8E+AoYAwQDB4BBoihaf+gaI0eOFNPT0y/baBkZmf9cxo8fz+nTp6+Nom9vRKz63s9dgiAUACE/cshi4GtRFI1AuSAIJVxwdj/owQRBuGFR52VkZG4u+vLuXp8mGXoj2ycBJ3o3PSsIwllBEFYJguDVuy0EqL7osBp+3CHKyMjIXBeuJrL9ciAKSORCC+8ffbmwIAhPCIKQKQhCZnNz808fICMjI9NHrjiyvSiKjaIoWkVRtAGfcKEbCpcZ2V4UxRWiKI4SRXGUn5/f1eRBRkZG5nu5nFnU741sLwhC0EUR638G5PZ+3g6sEwThn1yYZIgB/qNjAup0Onbs2EFbWxtms5lJkyaRlJSEzWbj0KFDkr4WXFjvl5WVxYkTJ7DZbHh5ebFkyRKSk5Opra3F2dmZxYsXYzAYOHLkCImJiQwaNIjc3Fz8/f0JDAy8wbntn1gsFrKysjh79ixDhw5l9OjRNDQ0kJKSQnBwMJMnT0alUiEIAq2trWzevBmTyYTRaGTJkiWSqIBer2fWrFl4eHhw6NAhBEFg2rRpdHd3U1paSlJSEkql8ortrK2t5ciRI1itVqZNm0ZwcDC1tbWkpKTg5+fH9OnTpbWuoihSWFhIWloa0dHRTJgwAaVSSVFREWlpaYSFhTFlyhROnDhBa2srs2bNQqlUcubMGUaNGnVVogA3C5fTgrNHtp8uCEJ279884E1BEM4JgnAWmAa8ACCKYh6wAcgH9gK//LEZ1O/DYrFI0kTV1dWSdM3NilqtJj4+nuHDh3P27FlJT668vJxPPvmEQ4cOXZI+MDCQkSNH4urqSlpaGlarleTkZCIiIhg6dCiOjo7s2LEDR0dH1q9fT1NTEykpKTg5Od2I7PV7RFEkMzOTDRs2EBISgkqlwmg08s477+Di4kJKSgppaWlSeicnJxITE0lISODIkSNYLBa+/vprSktLsdlsLF++nIKCAgoLCyksLCQvL489e/ZgMpmuWkygoaFBWjj/97//ndbWVt5++22cnJzIyspi586dUlqtVsvbb79NYGAgW7duJSsri7KyMlasWIG/vz/Ozs7U19eTlpZGR0cHR44c4eDBg7S1tV0zSfD+zk/eDVEUj4miKIiimCCKYmLv325RFB8URXFY7/ZFF7XmEEXxVVEUo0RRjBVFcU9fjWptbeU3v/kNe/bsYdWqVdzsr5Co1WoSEhKIiIjA0dGR4cOHo9fr2bFjB9OmTbtkBlkQBEJCQhg9ejRtbW3Mnj0btVqNSqUiLS2NzMxMSSGlsbERtVrNwYMHGTNmDM7Ozjf1g+B6YX9A2Gw2srKy6OzspK6uDoPBwPz585k2bRpZWVlSeicnJ+kBEx0dTUhICFVVVUyaNImZM2dSWFiIXq/HYDCg1+upq6ujs7OT4cOHX3X5JyUlMX/+fMaNG4fJZKKjo4Ouri5mzpzJmDFjLpHab2pqwsXFhWnTphEfH8+JEyfIzMykra2Nc+fO0dzcjIODA1arlfb2drq7uykpKWH8+PH/MS/W98ulWnZpIrsw5dGjR2/6G2Lveg4cOBAPDw+OHz+Oi4sLISEhmM1mrNZLG7mtra3k5+czceJE1Go1v/nNb3jsscdobGxk//79ksT4bbfdRmtrKyUlJbz77ruUl5ffoBz2bxobGwkJCWHBggWsXbuWxsZGST/QxcXlOyKVNpuNgwcPctttt+Hk5MSwYcPYunUr69ato66ujpCQEBYuXMisWbM4f/48ISEhfPjhh+zbt+8797Iv2LvIq1at4u677yYkJISgoCBWr17Njh070Ov1khO1153PP/+cgwcPotPp6OjowNXVlTvuuIP9+/fT0dHB0qVLmThxIq2trURERPDZZ5+xceNGSQ7qVqZfOjgAo9EoabPdCvprJpOJo0ePMn36dARBoKGhgYKCAr7++muOHTtGcXExXV1dkoLw6dOniYyMxMvrwts3Hh4eBAUFERAQQEdHB+7u7gwdOpTs7GzGjBlDXV0dQ4cOvaGquf0VhUJBTEwMQUFBhIaGAhd07gwGAy0tLZSUlBAUFHSJom9HRwclJSWMHDkSgHnz5nHXXXcRERHBsGHD8PHxYfDgwZLTKCoqYurUqZw8efKqFH3b29t55513GDNmDOPHj0etVvPLX/6SsWPHMmDAAOLi4iRdNnd3d15++WUGDx4sDV8EBQURFBRESEgIjo6OGI1GoqOjJW2+pqYmxo8fT2lpKS0tLdekfPsz/bYjbjAYWL16Na2trTzwwAM3vZNrampCo9EwZMgQBEHg3nvv5a677uLEiRPk5OQwaNAgPvjgAxYuXEhYWBj5+fnMnj0bhUKB0Wjk/fffR6fTYTQa+dWvfgVAbm4ukZGRDB48mPT0dLKzs5k1a9YNzmn/QxAEFi5cyCeffEJubi4TJkxg4MCBTJ48mXfffReFQsGTTz4pKfrec889VFRUEBUVhb+/PwAVFRXs3r0bnU7HAw88gLOzsxRA5e6778bV1ZVDhw4RGRl5VS+tp6SkcObMGQRBoKmpifvvv5+MjAxOnDiBg4MDjz/+OPX19ezYsYNHH32UQ4cOUVhYiJeXF5MnT8ZqtZKTk8Prr79OUFAQcXFxmEwmsrKymDNnDrW1tRw+fBhPT0/c3d2vVRH3W/ql4GVjYyNvvPEGv/3tb1Gr1VK8ypsZs9ksqbBenBej0YjJZMLV1ZWOjg5cXFxQKpXfUZ9ta2tDr9fj4eEhKRv39PRcoprb09ODr6/vTf8wuB5crFzs6+uLSqXCYrGg1WpxdnbG3d0dg8GA2WzG1dUVg8GAzWbD2dkZuHD/WlpacHR0lBSgLRbLJYq+Wq1WUkW+Urq6uujs7MRms6FSqfDz80Ov19PZ2YmHhwfOzs5YLBa6u7txd3eX8uTt7S1dt7u7W4pbotFoJEXfb6vmXosYwzeCm17R12g0UlZWRmxsrPxjlZGRuYS+OLh+2UVVq9XExcXd9K02GRmZG0u/dHCyY5ORkbkWyP0/GRmZW5Z+0YKz2Wx0d3ffaDNkZGRuAvryTmy/cHBarZZVq1bdaDNkZGRuArRa7WWn7RcOzs/Pj8cff/xGmyEjI3MTsHr16stO2y8cnD1CuoyMjMxPcd0UfWVkZGRuJmQHJyMjc8siOzgZGZlbln4xBvd92Gw2du7cibu7O1OmTJFf/r1OWCwWUlJSKCwsJCwsjDlz5uDk5IQgCHR2drJv3z4aGhoYMWIEY8eOpaKigqNHjzJu3DgGDRpEZmYmkZGR0qL0/ojNZuP48eOcOnUKi8WCp6cnS5cu5fjx49KSwKlTp0qL5K1WK+np6eTk5BAeHs6sWbOoqqri+PHjjB8/npiYGDIyMoiLi8PHx+ea2lpbW8vu3bsxm81MnjyZIUOGkJmZydGjR5k8eTKjRo265LcgiiJlZWWkpqZyzz330NzczKZNmxAEgWHDhjFlyhQOHTpEa2sr8+bNQ6VScfLkSSZOnNhnRV+dTsfu3btpaWlh1qxZREZGkp6eTlZWFj4+PsybN09Sv7HbVl5eTnJyMj//+c9xcnLCarWyfft2vLy8mDhxIocPH0ar1TJ//nwUCgWZmZlXZNsP0W9bcG1tbXz00Ud88cUXVyU/I/PjiKKIn58f48eP5+TJk6SkpEj77OIASUlJrFy5ktLSUnbt2oWPjw87duygsrKSkydP9vtF24IgEBMTw8yZM/H09KSkpASz2YxKpWLUqFFs27aNzMxMSWetpaWFjz/+mMTERPbu3cu5c+fYsWMHAQEBbNy4kYqKCrKysq5Lvh0cHIiPjyc6Opq3335bkgtrbW3l7Nmz30mv1+v58ssv2bFjBz09PVRVVdHY2MjkyZOJi4ujvr6eM2fOYLFYOHr0KMnJyZhMpitS9N2xYwdFRUVERkby3nvv0d3djYeHB+PGjaOyspKNGzdekr6np4c1a9awa9cujEYjoihSUFDA+vXrOX36NDU1NZw+fRqr1UpqaiopKSmYzearknz/Nj/p4ARBcBQE4aQgCDmCIOQJgvCX3u2RgiCcEAShRBCE9YIgqHu3a3q/l/TuH3AlhmVnZ5OQkICbmxvnz5+XlWqvEw4ODsTExNDe3o7VaiUgIEDaFxAQwOzZs0lMTMTFxQWbzYanpyc1NTWS1PfEiRMB+vX9EQSBwMBABg8eTH19PfPmzcPX15dZs2YxcuRIvL29LxGpVKvVeHl5UVVVhYODA56ennh6elJVVYWbmxsHDx6UpImudb79/Pzw8fGhtraW4OBgNBoNI0eOJCkp6Tu9GHtMj7CwMIKCggBwdHREq9Wybds2iouLJTWRlpYWDAYD1dXVJCUlYTab+2xbRUWF1JKvrKykvb2dmJgYDAYD3d3dBAcHf8e2iIgIybaenh727t3LzJkzUSqVksS+VqvFYDBQVVVFQkICFovlSovvO1xOC84ITBdFcTgXQgTOEQRhHPAG8LYoitFAG/CL3vS/ANp6t7/dm65PWCwW9u3bx6RJkxgyZAiHDh3q1z+gmxl78BKj0YjFYkGn012yz2Qy8cUXXzBs2DCioqK44447GDt2LLGxsQBkZmby8ccfU1NTc6OycNk0NzdTW1tLYmIigiBgs9nYtm0brq6ujBo1SkpnNBolHT64ID909913M3r0aAYPHoxCoSA9PZ0VK1ZQXV39Q5e7YqxWKyaTCb1e/6OOqK6ujuPHjzN69GhMJhMGg4H4+HheffVV7rjjDr788kuUSiXLli1jypQpaLVaQkND+fzzz1m/fn2fe0YTJkyQwgjU1dVJv0mTySTVHfu26upqMjIyGDt2rCQVlpycjLe3N97e3pLE07333svkyZNpa2sjNDSUL7/8knXr1l2zXtvlxGQQRVG013pV758ITAfsbdI1wJLez4t7v9O7f4bQhwE0URSpqamhoKCAo0ePkp+fT1pamryU6zphb4VMnz6dhIQE8vLysFgsdHZ2YjQaWbt2LZ2dndx11104ODjg5ubGwIEDyc3NJTExka6uLiIiIsjJybnRWflRRFEkPT2dYcOG4ebmhtVqZceOHWRnZ/PQQw+hVqsxmUzodDqqq6tRKBQsXryY6Oho8vLycHd3Jyoqiry8PIYOHYrBYCAiIoLs7OxraqfBYCAsLIwlS5ZQX19PV1cXVqsVs9mMxWLBarVK96ejowOLxcLatWvJzs7m6NGjCIKAq6sr3t7ewIWWVEREBBaLBRcXF5qbmxk/fjzV1dW0tbX1ybZx48bx9NNPEx8fT1JSEm5ubphMJsaPH8/kyZMvqTt6vR6j0ciaNWvIyckhNTWV7u5uzp8/z+bNmzl27Bh1dXVERERgs9nQaDS0trYycuRIqquraW9vvybleVkdcUEQlMBpIBr4ACgF2kVRtLclL45eL0W2F0XRIghCB+ADXPb6iqKiIu644w6WLVuGxWLho48+oqKigmHDhl3uKWQuk66uLj755BOpsj/22GNUV1ezc+dOZs6cyY4dO4iJieG9997j4YcfJjw8nLNnzzJixAgGDRpEWloa58+f5+67777BOflxTCYTZWVl0mB2U1MT69evx8fHh+XLl3P33XfT09NDUVERixYtwsvLi3/+85+YzWaWLFmCKIpkZ2czevRoYmNjSUtLo7i4mLvuuuua2nnq1Cn279+PyWRi5syZ+Pr6kpqayv79+7FYLERFRREVFcXWrVt55pln+Nvf/kZXVxfOzs7MmjWLnJwcKeSh/XiTyURubi7z5s2jvr6effv24ePjg4eHR59sa2hoYP369ej1eu677z4cHR1ZvXo1NTU1mM1mli1bRmVlJbt37+aXv/wl//u//0tXVxcajYbbb78dd3d37rzzTk6cOEFJSQnh4eEYDAbOnj3LokWLqKqqIjk5GR8fn2umNtwnwUtBEDyBLcB/Aat7u6EIghAG7BFFcaggCLnAHFEUa3r3lQJjRVHUfutcTwBPAISHh4+srKwELjxp9Xo9CoVCiv/Y09ODUqlEo9FcbX5lvoVd399oNOLs7IyzszNWqxW9Xo9Go6GjowObzYYgCHh6eqJWq9Hr9ahUKpRKJT09PZjNZjw8PPr1TLcoiuh0OpydnVEqlVgsFqkFBBdiXigUCsxmM87OzhiNRslx2FV9DQaDlO/u7m4sFss1z7fJZKKrqwtBEHB3d5euZe/BuLi44OjoiF6vx9XVVepqd3d34+LiIrWgFAoF7u7uODg4YLPZ0Ov1UtS19vZ2nJyc+hxm0l5mSqVSUqbu7u5Gr9fj6OiIq6urVHe+zzaFQoEoilJr1NnZ+Xttc3R0lGbyv4/rqugrCMJ/A3rgZSCwt5U2HnhFFMXbBUHY1/s5XRAEB6AB8BN/5ELfVvSVkZGR+SH64uAuZxbVr7flhiAITsAsoAA4BNjb5w8B23o/b+/9Tu/+gz/m3GRkZGSuF5czBhcErOkdh1MAG0RR3CkIQj7wtSAI/wucAVb2pl8JfC4IQgnQCtx7HeyWkZGR+Ul+0sGJongWSPqe7WXAmO/ZbgD694izjIzMfwT9YqmWPXybjIyMzE/RlxeB+4WDa2tr45tvvrnRZsjIyNwE9OX9vX4RF3XkyJFiRkbGjTZDRkbmJmDcuHGcPn365omLKgjCNVMPkJGRubWRFX1lZGRkkB2cjIzMLUy/6KJ+H983NtiflwL9GD+Ul4u32/N2OWOi3z722/uu1rYfOueV2PZD+fx3crnjzH3J5+Wc50q50jrwY/Xih85xNXZ9X3n9WBn2xb5rVU/6pYOzWq2cOHGCkydPotFomDt3LhERETfarCvGbDazd+9eiouLCQkJYdGiRZK+2ubNm3FycmLBggVS+hMnTnDkyBGsVivu7u48/PDDZGRkcPz4ce644w6GDh1KWVkZhw4dIjExkaSkJE6cOMGAAQMu0eS6XLKysjh+/Dju7u4sWrQIR0dHtm3bRmNjI9OnT2fo0KEoFBca+yUlJezZsweAmTNnMmjQIA4fPkx6ejqLFy9m2LBhVFRUkJKSwvDhwxkxYgSZmZkEBwcTGhp6bQq0j4iiyNGjRzl58iQWiwV3d3ceeeQR8vLySE9Px9PTk0WLFknrSjs7O9m6dSstLS0kJSUxadIkTpw4wfHjx7FYLHh5eXHvvfeSnJyMQqFg7ty5dHd3U1RUxIQJE65KsNFqtZKRkUFGRgYDBw5k/vz5GAwGtm/fjlarZfbs2QwePFhKu2/fPs6cOcNdd91FbGws3d3d7Ny5k8bGRpKSkhg7dizJycm0tbUxf/581Go1GRkZTJkypc/j3ufPn+fAgQNYrVZmzZolSWY1NDTw5Zdf8vOf/1xSdtbr9ezatYvq6mqio6OZPXs2jo6OWK1WNm3ahK+vL5MmTeLAgQO0tLSwcOFCFAoFp06dYvLkybe2om9dXR0rV65k/PjxjBw5EpVKhSiKVFRUcPr0aVpaWhBFkaamJs6cOUNJSQlarRatVktLSws2m43y8nJp4fKZM2coLi6W3rcrKSkhOzub5uZmRFGko6ODM2fOUFBQgNlsprW1ldOnT1NeXt6nKNo/hCAIDBgwgBkzZnD27FkOHz6MKIrk5uayZcsWTp06dUn6qKgoFixYQGhoKOfPn0epVOLp6Ulzc7Nk0759+wgKCmL79u3U1tZy8uRJ3Nzc+mxbV1cXn376KQkJCTQ3N7N7926Sk5MpLCxkxIgRfPDBB3R2dkrp3dzcmDBhAuHh4bz//vvYbDa8vLxobW2lpKQEURTZvXs34eHh7N27l6qqKjIyMvD09LzaYrxiBEEgLi6O+fPn4+vrS01NDUajkU8//ZThw4dTUVHBwYMHpfQWi4Xw8HAmTJjAqlWrqK2tZdCgQSxYsICgoCBKS0spKyujubmZhoYGCgoK2Llzp7Sg/GooLCxk06ZNjBkzhpiYGADWrl2LTqdj4sSJ3ylHHx8fGhoaqKioACAtLY2cnBzGjh3L559/Tn5+Pnl5eSgUCo4fP86+fftQKBRXpOhrsVhITEwkIiKCDz/8UBJa2Lp1K7t376a1tVVKa7Va8fPzY9KkSezbt4/Tp09LdX7btm1kZ2dTU1PDuXPnUCqVHDlyhAMHDiAIwr9X0fdGoFarUSgUFBYWolKp8PX1JTs7m3fffZdTp07xj3/8g6qqKv7+97+TkZHB3/72N/bu3cuBAwc4evQoJpOJ9957j7q6Ov71r39x5MgR1q5dy/Hjx0lJSeHNN9+ksLCQt99+G61Wy1tvvcWRI0c4ffo0dXV1vPXWW5w4cYLly5eTm5t71V0WBwcHBg4cSE1NDT09PYSEhNDV1cW+ffuYPXv2JSopgiDg6+tLbGws1dXVzJs3D41GQ1JSEtHR0SgUCgRBwN/fn4qKCjw9PTlw4ADjx4+/IoVZjUZDYGAg+fn5tLS0EB4eTnV1NYMHDyYxMZGWlhZqa2sl2wICAlCpVJSWlhIZGYmDg8N3bAsMDKS0tBQPDw8OHTrEmDFjLlHMvRH4+/sTExNDdXU18+fPx8XFBX9/fwoKCujs7JRUZwG8vb2ZOnUqcXFxkqqFr6+vJAE+Z84cvL290el09PT0UFNTgyAIhIWFXZFS7sUcPXqU1tZW9u/fz+nTp+no6CAtLY3q6mq2b99OU1OTlNbBwYExY8YQGRkpdekCAgLQ6/Xk5eXh7e2Nr68vcEHs02g00tjYyJAhQyQxz74wePBgxo8fT0REhOQgT506hUqlklpzdlxdXZkyZQrx8fG4ubmhUCjo7u5m//79zJo1C6VSiYuLC4Ig0NTUhMlkor6+nsGDB2Myma60+L5Dv3Rw/v7+/OlPf0IQBFasWMHmzZtJSUmRAkTX1dVx+vRpXF1defzxx5k+fTo2mw1RFKUfuM1mo6amhqysLFxdXXF0dOTMmTMATJkyhYULF0oCfEajkSeeeIL777+f5uZmioqKcHV1Bbgmgob2yufg4ICDgwMtLS0cPHgQR0dHVCoVHR0d9PT0XJK+qamJ8vJyxowZ851xDUEQmD9/PpMnT2bkyJGYzWays7P55JNPyM/P75NtJpMJs9mMk5OTJDo4adIkjh49yooVK6ipqfmOc7LLTXd2dko/6Isd6/z587nttttISEjAbDaTm5vLp59+SlFR0Q1RZhYEAUEQaGhooLGxkYSEBAwGA1arFWdnZxwdHS95edQu0bV69WomT55McHAwgiDQ2NhIVVUVo0aNIiIigrvvvps5c+ZQVlaGh4cHa9asYfv27Vclud3V1UVoaChPP/00qampVFZW0t3dzdy5c5k+fTpfffXVjz4suru7cXR0xNHREbPZjFqt5v7772f69Ok0NTURHBzMunXrWLduHXq9vs/lWFtby5dffsl9992H1Wpl27ZtDBkyBL1eT1tbm9TjsatBb9q0CX9/f4YPH87u3btxc3NDrVbT3t6Ou7s7999/P9OmTUOr1RIUFMRXX33F559/fsnv4Wrolw7OaDSiVCpZtGgRM2fOpKCgAC8vL7y9vRk0aBBPPvkkAwcORKfT0dbWRmNjI6Io4ubmRkdHB52dnTQ2NuLk5ISXlxdhYWFMmzaNJUuWIAiCpDMH4OzsjMFgoLW1lc7OTjQaDZ6engwYMIDFixczc+bMq86P2WzGYDAwbtw4YmJiKC0tRRAEOjo6OHz4MLm5udTW1tLS0oLJZEIURTIyMoiPj8fd3V2SFDcajfT09GAymXByciI8PJysrCzGjx9Pa2srw4cP5/Tp032yraOjg4aGBmbOnMmwYcMoKChg8ODBPPPMMyQkJJCQkEBISIhkW1dXF8HBwUybNo26ujr0ej0GgwGDwSBJbGs0GkntdsyYMTQ1NZGYmHjN1W/7giiKHD9+nKSkJFxdXWltbaW5uZlZs2YxdOhQioqK0Ov1tLe3o9fr+fjjj7HZbMyZM0caHM/IyGDw4MGSdlxERARlZWXSmOi0adPIy8u7otaRncGDB0uadQ4ODri6uhITE4NCoUCpVKJQKDCZTGi1Wkk70V72FouF4uJiwsPDmT17NkajkaamJqnH4O3tjVarZcKECTQ0NPRZNbe+vp4333yTqVOnEhsbi81mw93dncOHD1NUVMTp06fR6XS0tLRgNpvZsGEDxcXFLFu2TMqP/eF+5swZmpubCQ4OloLXtLe3M2rUKBoaGi4ZFrka+uUkg06n44svvpBaCA888ADBwcGsWrWKPXv2EBAQwH333UdMTIzUzQwNDWXkyJF88sknrFu3Dj8/PyIiIvjZz37G4cOHcXR0ZMGCBXh6ekoVJTQ0lAEDBjBp0iTeffdd/P39+fnPf87MmTPZt28fbm5u+Pn5XXV+urq6WLlyJVqtFhcXF5544gkCAwOZP38+mZmZ5ObmEhUVxbvvvsuiRYsIDw+npqaGWbNmSaKB+/fvJzs7m+LiYoKDg5k4cSKFhYUkJiYSHR2Nt7c3mZmZ/OxnP+uTbf7+/kyaNIn33nsPpVLJAw88QEdHB+vWrcNoNPLAAw/g7u7O8uXLmTdvHmVlZSQnJyOKIj/72c9wdXVlx44dZGVlUVhYSHh4OOPHj6egoIDExETi4uLIzMzk1KlT3HnnnVddlleKyWSitraWhQsXIggCISEhjBs3jn/9618oFAoeeughcnNzKSoqYsyYMRw6dIjo6Gj+8Y9/8OyzzxIcHExVVRWzZs2SztnZ2UlnZyczZ87E1dWVvXv3MnToUBwdHa/YzilTppCfn89bb73F6NGjiYyMZNmyZWzYsAGbzcY999xDU1MTW7Zs4emnn2bHjh2cO3eO8vJyIiIimDp1Kp9++in/+Mc/GDx4MNHR0RiNRkpKSliwYAENDQ1s376doKCgS0L8XQ6pqalUV1dz6NAhSkpKeOyxx/jDH/4gCaLOmzePpqYmdu/ezdKlS9m7dy/e3t68++673HfffSxZsoRFixZx8uRJSktLCQsLQ6/XU1xczJIlSyQl6cDAwD6rDf8Q/WKp1rcFLy9W/bQr+8KFQU57WDG1Wi0F57DHkbz33nsxmUzYbDYUCgVqtVpq/QDSd7jQzbKHjrPZbFKgEY1GI32/uLV3NdPWoihiMpmwWq04ODigUqmk81mtVmw2Gw4ODphMJmmf2WzGwcFBUkG1B/aw50OlUmGxWBAEAYVCgcViwWKx4Ojo2CdbRVGU8msv14vLzJ5/uz02mw2TyXRJ2XzbNgcHB6xW61Xbdi2xl6FKpZLK9Nv5ttlsUt2xh7mDC5Gq7Gq/9nsCF4ZB7PfUXmYqleqKBvAvttNez+1l+e37Yf992ONI2Mteo9GgVCqluqZWq6UBe3tdhwvKxPZ62BfswxkACoVCup8Xly1c+J06ODhgMBikMtRoNFJe7OVsnzy022bPp30o54Yo+l4PrlbRt7W1FVEU8fb2vmnflZORkbk8+uLg+mUXta/YIwjJyMjIXEy/nGSQkZGRuRbIDk5GRuaW5Se7qIIgOAJHAE1v+o2iKP6PIAirgSlAR2/Sh0VRzO4N8vwOMA/o6d2e9WPXMJvNNDY2XnkuZGRk/mPoy8vUlzMGZwSmi6KoEwRBBRwTBGFP777fiqK48Vvp5wIxvX9jgeW9/3+Q9vZ2Nm/efNlGy8jI/OfSl/f3LifojAjoer+qev9+bOp1MbC297gMQRA8BUEIEkWx/ocO8PPz46mnnrpso2VkZP5zWbly5U8n6uWyZlF7QwaeBqKBD0RRPCEIwtPAq72BoFOA34uiaARCgOqLDq/p3Vb/rXNeHNlefr1DRkbmmnNZkwyiKFpFUUwEQoExgiAMBf4AxAGjAW8uRLq/bERRXCGK4ihRFEddi9UCMjIyMt+mT7Oooii2cyGi/RxRFOvFCxiBz/j/MVJrgbCLDgvt3SYjIyPzb+UnHZwgCH6CIHj2fnYCZgGFgiAE9W4TgCVAbu8h24GfCxcYB3T82Pjb9yGKorSEya4QYrFYflSJwn7Mt9NcvDTkRnHx8huz2fydfH3bbvt2e57tebOns+fpWqxCsZ/bvjTuYru+r8zt6b+97+J7ZP9sL3N7+d/IVTPfLuvvy/f3pf92nr59/LW+JzabTaon37bLZrN95758+35cXPYX1/3+sGLpRnA5Y3BBwJrecTgFsEEUxZ2CIBwUBMEPEIBswD5LsJsLr4iUcOE1kUf6apRer2f58uWMHz9eWrj9zTff8Pzzz//gAmGTycT27duZO3euJHVkJycnB7PZzJgxY7732OtNV1cX69evR6vV0tXVxZw5c5gwYYKk0hsaGsodd9yBk5MToiiSnZ1NSkoKSqWShQsXEh4ezs6dO+no6GDhwoUolUpOnTrF9OnTr2rdI1xYN7hmzRrq6+vx9PRk2bJlmM1m1q9fj9FoZPz48UycOBGlUokoimi1WjZu3IjBYGD+/PmSKGNpaSlbtmzhscceQ6vVkpKSQkJCAmPHjiU9PZ3w8HDCwsJ+wprrgyiKlJaWsn79egYOHMjSpUsRBIHly5fT1taGj48P9913nyQmKYoi58+fZ+fOnahUKh588EFaWlr45ptv8PX15Re/+AU9PT1s3boVhULBwoUL6erqoqioiEmTJl2VYGN2djb79++XHNbzzz8vLTzPzc1l7969KJVK5s2bR1xcHHV1dWzevBmTycQ999yD2Wxm//79DBkyhAkTJnDy5EkCAwMZMGDANSjJm4+fbMGJonhWFMUkURQTRFEcKoriX3u3TxdFcVjvtgdEUdT1bhdFUfylKIpRvfv7vMjUaDRy5MgRtmzZgslkYseOHaSmptLd3U1HRweHDh0iMzNTetJlZmaSlpbGsWPHMBqNdHV1kZqaysmTJzGbzVRUVFBSUtL30rlGODs7M2fOHBYvXoxWq0WtVlNaWsrmzZuZOnUqI0aMkH4Uoiji6OjI9OnTJeXUyspKzp8/j1qtJisri+TkZJycnK6J8qnVaiUvL4/p06ezYMEC3N3dcXZ2Ztq0aUyaNIlPPvmErq4uKf2GDRtQKpXEx8fz0UcfYTKZ0Ov1fP3112RkZEhCnoMGDSIlJYXy8nKysrLw8fG5aluvBhcXF0JDQzlz5ozU6srNzWXGjBnMmzcPFxcXKW1nZycffvghiYmJTJkyBY1Gg7OzM+Hh4ZIcVVFREV1dXbS1tZGfn8/OnTvx9va+akXf6Oholi5dSkJCAvX19ajVammfn58fc+fOJTY2lo8++gidTsf777/PgAEDmDVrFq6urqSkpBAREcGBAweora3lxIkTN7zsbyT9di1qUFAQGo2GU6dOodVqiYmJwWw28/bbb+Pv7091dTU1NTW4urqSnJxMdHQ0xcXFGAwGPvroI3x9fdFqtdTV1d3orODg4EBYWBhlZWVoNBqGDBnC9u3bqaurY8uWLQQGBvLEE0+gVqsleW240PJzcnLCw8MDq9VKTU0Nzs7OtLa2Mn36dEwmk3TMlSIIAs7Ozqxfvx4vLy+effZZfH19UavVHD58mPDwcEnNReyViZ81axaRkZH/r71zD4rqyvP45/QDEVCBpiUICEiMQR2Rhw+IZsREEkjGUeIrjm/NwzFTs5nN7G5qy62kaqq2tlLJTB7rRieaGGejxmfUsRwNEnXGB1EjKAjKWxrk2TSC0HTL3T/oe7ch+EBIGjr3U0XR95zb3d/fvb97+t5zz/0eNm3aRHV1NVlZWYSHh1NVVaU422ZnZ+Pn50dGRgaJiYkudfQVQhAUFERMTAz5+flK2eDBg9m5cycGg4F169YpZg0lJSXk5eUxePBgdDodr732mvL+c+fOARAQEEB9fT0ajQaTyYSnpyfBwcGKk8bD7pOhQ4fi7e3NwYMHmTVrVifrJaPRSFFRERcuXCAqKoq6ujqysrKUOSdeeeUVgoODyc3NxdfXl+PHjzNlypSf7OUp9ONHtXQ6HVOmTOGzzz5TrKMbGxupq6tjyZIlzJs3j8zMTM6dO8ecOXNYsGAB4eHhNDQ0cPHiReUzsrKyXNr/JuNsYunt7U1rayuhoaG8/vrrlJWVKU688oGRm5vLgQMHWLFiBUajkTVr1jB37lxu3LhBYGAgW7ZsYdu2bb22yNbr9fz+97/n7bffxtfXV5lQxsfHh+DgYMxmM62trYq2pKQk9u7dy5YtW2hoaKCyspKjR48SFhaGxWKhtraWZ555hl/84heMHTsWIQTffvstH3/8MTk5OS7vh5MRQrB+/XreeusttFotGRkZSp1se7Ru3TqMRiPHjh373meFhISwfPlyZs+eTXFxMYMHD2bLli3s3r271425xWIhLy+vk5uzrNnPz4+goCBMJhNWq5U7d+6wevVq4uPj2bdvH0lJSfzyl79k0qRJtLa2kp2dzcaNG5Vj4qdGv23gJEli/PjxjBgxgsTERCRJYtCgQdjtdkwmkzIfgcFgoLa2FovFQkNDg+LIGxISwvTp01m4cGGvLxv6gtu3b3P27FmSkpLQaDRERUUpHnCyH9yNGzdoaWkhPz+fDz/8kLlz5xIYGAh0/HqXlZUxcuRILBYLcXFxmEymXls72+12bt26pTgGa7VaGhoa0Gq1TJw4UbFTLy8vp6Wlhbi4OBYvXkxkZCQTJ07EaDQSHh7OmTNnKC4u5urVq+j1egwGA9nZ2UyaNIm6ujpiY2PJycnpi035UNhsNm7dukVraytNTU1YrVaam5uxWq20tbWh1Wq5desWN2/eJDQ0FKPRqPif6XQ6bDYbjY2NWK1WmpqalLknrl69SnR0NKWlpcyYMYNr1671ytFXkiSys7MJCgrCaDQiSRImk4nm5mZqa2vx9/cnJiaG8vJy/P39CQkJwWaz0dbWhk6nw8PDA6PRyPnz55k2bRpVVVVMmTLFpW7KrqRfXqLqdDoiIyMxGo2sX78eu93OqFGj8PPzY/78+Xz++ed4eHiwYsUKPDw8+Mtf/kJBQQHh4eEYDAZWrlzJsWPH0Gq1PPPMMxgMhl4lXV9QX19PWFgYkZGRAMTExJCTk8OHH37Iz372Mx5//HE2b95MamoqZ8+epbGxkQMHDpCXl8fy5cux2+1UV1eTkpJCTU0Ne/bsYcyYMd+7odJTWlpa2LlzJxUVFQQFBSmuvXv27OHOnTukpaVhMBj49NNPSU5Oxm63s3//frRaLa+88grh4eG8/vrrWK1WPDw8ePrpp9FoNFy7do1JkyYxZswYsrKyuHDhAvPnz++LTflQlJWVsWvXLqqqqti5cydz5sxhx44dVFZWEhISwsyZM7l27RrXr19n3rx5pKWlsXHjRvz8/EhOTqasrIzdu3dTV1fH9u3bWbZsGS0tLcqNmGHDhnH06FHi4uJ65egrzyWSnJysGEgePnyY6dOnU1hYyMmTJ9FqtaxZswZ/f39WrFjBtm3b8PDwYM2aNQghKCwsZOLEiYwePZrIyEhOnz5NWlpaX23KAUW/NLzseinR3S38+/VxyO/peorvKu4W091i6U5/13q5rrduw93puNf3d/fdD5pHrnT07S6me+VST7ZJ1/UfNs6Hyf2u33u397jL00ID3vCy64643/KDfIaruVsMd9N5P/19Fd/ddNzr87ur62/buyt3O8B7EuePsU8eJvf74nhxV1zfOaWioqLyA6E2cCoqKm6L2sCpqKi4Lf2iD85ut1NbW+tqGSoqKgMAeZrEB6FfNHBms5ldu3a5WoaKisoAwGw2P/C6/XKYiIqKisrd6MkwEbUPTkVFxW1RGzgVFRW3RW3gVFRU3JZ+cZPB3XF2f9XpdGg0mk4OuBqNppPFjlwnhECr1X6vXDa5lB8E12g03LlzByFEj40FnLUJIRQdsja9Xv+9x5VkPzVnHXJfrqzHWZvs5uLKx4Vkp1wArVaLTqdTHHK1Wm233nqyG668D2SnXPkZUdk1RH5ve3s7Go2m14/OyTq7bntZr7yf4N7b3vmxrb7Y9l23oRz33XK1vb2dO3fuKOvK+StrFUIozt1yPPL6fZUnagP3I9DW1sb+/fvJz8/HaDSyaNEibt26xQcffMCQIUOYMGECs2fPVhKmoKCAL774AoPBwNq1a5XkyMnJYe/evbz22mtUVFSQnp5OdHQ006ZN4/jx40RFRfXYNddut/Pxxx9TVVWFv78/S5cuxWw289VXX9HW1kZSUhJTp05VGuWbN2+yY8cOrFYrzz//PBEREWzdupWamhrMZjMpKSmEhoZy7NgxRdvJkyeJjIxk5MiRP8TmvS+SJHHmzBm++uorfHx8eOqpp0hISCA9PZ2zZ88SGhrK/Pnz8fLyUg6s9vZ20tPTyc3N5dVXX+X27dvs2LGDhoYGUlNTiYiIYPfu3Wg0GtLS0rBYLOTn55OUlNQrI9Li4mJ2796N1Wpl+vTpnRyCs7OzOXToEBqNhtTUVCIjI/nss8+or6+nrq6OhQsXYjQaOXLkCFFRUSQlJfH3v/+dkSNHEhER0evtmJeXx5YtW/Dx8WHKlCkkJydTWFjI/v37GTRoEIsWLVLcb5qbm9m1axfl5eWMHDmSF154gYqKCiWvnnrqKaKjozlw4ABms5m0tDS0Wi2nT5/m2WefVX5EessD/9wLIbRCiO+EEIccyxFCiHNCiAIhxE4hhIejfJBjucBRH94nSgcwWq2W+Ph4lixZws2bNzl58iRmsxkhBAsXLuTJJ5/sdOY1bNgwYmNjuXz5suKn39zczIEDB7h06RK3b9/mm2++ITo6WnHNvX79OgEBAT3W1t7eTklJCampqcybNw9fX188PT1JSUlh1qxZbNy4kaamJmV92SAyMTGRjRs3IoRg/vz5rFy5Ervdjo+PD19//TVxcXGcPHmSgoICcnNzGT58eJ9sy4elvLyciIgIFi9ezIQJE6ioqODLL78kLS2N0tLSTn5wkiRRUVHB4cOHuXz5Mnfu3OHgwYMAzJgxg61bt3L58mVsNhtWq5WcnBz++te/Ehwc3GtrLn9/f+bOnUtqaiqffvopjY2NSt2IESNYtGgRcXFxfPLJJ+j1ehYuXMjKlSuxWq14e3tz4sQJxo0bx6lTpygrKyM7O1tpdHpLTU0Nvr6+LF68mKlTp2Kz2di6dStxcXEYDAZ27tzZ6YwxOjqaRYsW8d1333H+/Hm8vLxISUnhySefZPPmzRQVFVFeXs4jjzzCqVOnOHLkCMOHD++1Db8zPdkbvwWuOi3/F/BHSZIeBczAakf5asDsKP+jY72fNDqdDoPBQEZGBlVVVYwZMwYvLy/q6ur46KOPOHjwYKfBi8OHD2f8+PGdbMy//vprxowZw4gRIwCIiIjg7NmzPPLII6SnpzNt2jTFW66nDB06lC+//JKPPvqI2tpaQkNDGTduHF5eXgwbNkxJOEmSsFgshIWFER4ejslkor6+XmlY29vbGTt2LI8++ij/+Mc/GD58OBkZGYo2Vw5J8vX1JTs7m3fffZdz587R1NSEXq8nLCwMg8HAlStXlHVtNhsHDhxg5syZeHt7I0kSxcXFxMTEMH78eMVJuba2lrq6OkwmE8OGDSMwMLDTJePDMHToUMxmM4cOHSIqKgovLy+lzs/Pj+LiYo4fP05sbCx6vR6j0Uhrayuenp5ERkYSERFBZmYmAQEBHD9+nISEhPtO2PSgeHt7U1ZWxvvvv8+xY8doamqitraW6Ohopk6dSllZGW1tbUCHTX9MTAxBQUHodDqGDBlCSEiIkldDhgzBYDDQ3t7O1atXEULQ1NTEqFGjaG1t7bNceaAGTggRAjwHfOJYFsBMYLdjla10zKwFHTPbb3W83g08JX7KdgYOPD09iYmJwWAwUFBQQEREBO+//z7r168nPT2dqqqqu75Xds01GAzU19dTUVHBrFmzWL16NePGjUOv15ORkcGGDRv47rvveqTLw8ODN954gz/84Q+EhYVx5MgRAEpLS9m6dSurVq1i8ODBQMev8qxZs9i3bx+ffPIJjY2NyiXd6dOnmTBhAkOHDmXWrFmsWrWKxx57jEGDBnHixAk2bNhAdna2yxq5pKQk3nnnHX7zm9+wfft2AgICCAkJ4b333uPUqVOdLisvXbpEcXExGo2Guro6ampqlPhlAgMDWbduHS+++CKlpaVoNBo2bdrE9u3bezTSvitCCEJCQpg0aRKlpaU0NzcrdVqtlkcffZQJEyaQn5+v9IWePHmS+Ph4fHx8mDFjBi+99BJxcXG0t7dz7tw5NmzYQGZm5kNrkomOjua9997jjTfe4PDhw1gsFkVzd4e41Wrl888/57HHHmP8+PFIkkRpaSnbt29nxYoVBAYG8uqrr7Js2TJMJhMBAQFs2bKFTZs2dYq7NzzoGdyfgH8B5NMDA9AgSZK8J+XZ68FpZntHvcWxfieEEC8LIc4LIc7LCeSutLa2Ul5eTmBgIL6+vlRXVyv9JjU1NUrHakFBAc3NzdhsNiwWC62trcolyrhx47h06RImk4mCggL0ej1arZaLFy+SkJDA7du3iYuLU6zPH5S2tjYqKysxm83U1dUxePBgTCYT7777LjNmzCA0NJT29nYKCwtpbm5m3LhxLF++nMcff5zY2FiMRiMtLS2cOXOGn//850oHvlarJTs7m8mTJ3Pr1i3i4+OV+RBcQVVVFWazmcrKSvR6PZ6enixYsIDk5GTCw8OJjY3FbDZTVlaGp6cnoaGhXLx4EZPJRHl5OaGhoWRlZXHt2jW0Wi3+/v74+Pjw7bffEh8fT3l5OTNnzqSkpEQ5i3lYnXa7nfDwcCwWC1arlaKiIhobGzGZTOh0OsLCwqipqaG9vZ2WlhYuXrxIYmKicvNBr9eTmZnJ9OnTqa+vJzExscd50R3yGat8vPr4+ODn50dubi4XLlwgODgYm81GYWEhbW1tbN68mYaGBlJSUgCoqKjgnXfe4emnnyYkJARJkvDx8SE/P5/g4GAsFgsJCQlYLJZO3SK94b4Xu0KI54FqSZIuCCFm9Mm30jGzPbAJOp5k6KvP7Y9YrVaOHj1KZWUlgYGBzJ49G5PJxL59+7DZbKxevZrhw4crrrk2m409e/Zgt9vZu3cvy5YtY+3atdhsNgYNGkRycjJCCIqKipg2bRqjRo0iIiKCnJwc5s6d2yNtbW1tHDlyBJPJRGhoKCkpKZw/fx6LxcLp06fJy8vj5Zdf5sSJE8ycORO73a5Mp7d27Vr0ej0NDQ2KBpmioiISExMZPXo0o0eP5sqVK7zwwgt9vWkfCEmSyM3NVc7U1q1bh5eXF5mZmeTk5BAbG8sTTzxBXl4eRUVFzJkzh/Hjx1NdXY3RaCQ+Pp6oqCi2bdvGoUOHWLp0Kd7e3lgsFrRaLZMnT8bX15ejR4+SkJCgnPE+DDdu3ODw4cNIksTKlSsxGo188cUXJCQkUFJSwqlTp9Dr9fz617/Gw8MDk8lEdHQ0ISEhymeUlJQwefJkIiMjGTt2LOfPn2fOnDm93o4lJSWKtpdeeomAgACWLl3K3r178fDwYMmSJdTW1nLixAmee+45cnNz8fLy4s9//jMLFiygqqqKpqYmTp06xfXr11m1ahVCCGpra0lNTeXmzZscPHiQMWPG3HV60J5y30e1hBD/CSwF7IAnMBTYBzwDPCJJkl0IkQC8JUnSM0KIvzlenxFC6ICbgFG6xxe5+6NazhMDOw8jkPvL5I5pZ2dWeQiCPPRDvuXv7M7a3t7e6bU85KQnPQLdaZOXZeS7uM665e/tTld3Oh9GW18iD29xHkojxykPS7hXHPL6znF0re+LIQ530+m8X7rLCeebG32RFw+irTtNztvEecJpWV/XvHKOz1n7vW7W9KmjryRJbwJvAjjO4N6QJOlXQohdwDxgB7Ac+MrxlgOO5TOO+uP3atx+CsjJ1nWndR1O4JyA3d1J6trX4fx5D3v3rjtt3Wl1/t7udDvX30unq3AeO+Zc1jXue8XRnVPu/fZZX+nsTm93GqBv8qIn2pzzwVlPd8NlutNzr9zqLb2J/l+B3wkhCujoY9vsKN8MGBzlvwP+rXcSVVRUVB6OHv3kSJL0DfCN43URMLmbdVoB102fpKKiouLA9dcOKioqKj8QagOnoqLitvQLw0shxC3AdYOkfhgCAHfyYXe3eMD9YvqpxBMmSZLxQT6gvzxsny9JUryrRfQlQojz7hSTu8UD7heTGs/3US9RVVRU3Ba1gVNRUXFb+ksDt8nVAn4A3C0md4sH3C8mNZ4u9IubDCoqKio/BP3lDE5FRUWlz3F5AyeEeFYIke9wAB4wj3UJIbYIIaqFEFecyvyFEMeEENcd//0c5UII8YEjxmwhRKzrlHePECJUCJEhhMgVQuQIIX7rKB+QMQkhPIUQmUKILEc8bzvKI8QAdqIWbuasLYQoEUJcFkJcEkKcd5T1Wc65tIETQmiB/wZSgLHAi0KIsa7U1AM+A57tUvZvQLokSaOBdP7/OdwUYLTj72Xgf34kjT3BDvyzJEljganAOse+GKgxWYGZkiRFAxOBZ4UQUxn4TtTu6KydJEnSRKchIX2Xc7K9iSv+gATgb07LbwJvulJTD/WHA1eclvOBIMfrIDrG9wFsBF7sbr3++keHO8wsd4gJ8AIuAlPoGDiqc5Qr+Qf8DUhwvNY51hOu1t4ljhDHAT8TOASIgRyPQ1sJENClrM9yztWXqIr7rwNnZ+CBSKAkSZWO1zcBebaPARWn43ImBjjHAI7JcTl3CagGjgGF9NKJ2sX8iT521u4HSMBRIcQFIcTLjrI+y7n+8iSD2yFJkiSEGHC3qIUQPsAe4J8kSWp09uoaaDFJknQHmCiE8KXDpPVx1yp6eMQP5KzdD5gmSZJJCDEcOCaEyHOu7G3OufoMzgQ4T+QZ4igbqFQJIYIAHP+rHeUDIk4hhJ6Oxu1/JUna6yge0DEBSJLUAGTQcQnnKzqcpqGzZiUeR/0woO7HVXpPngBmCyFK6DCZnQm8z8CNBwBJkkyO/9V0/AhNpg9zztUN3LfAaMedIA9gER2OwAMV2c0Yvu9yvMxxF2gqYHE6Be8XiI5Ttc3AVUmS3nOqGpAxCSGMjjM3hBCD6ehPvEpHQzfPsVp3TtTQD52oJUl6U5KkEEmSwuk4To5LkvQrBmg8AEIIbyHEEPk1kAxcoS9zrh90MqYC1+joH/l3V+vpge7tQCVgo6MvYDUdfRzpwHXga8Dfsa6g425xIXAZiHe1/m7imUZHf0g2cMnxlzpQYwImAN854rkC/IejfBSQCRQAu4BBjnJPx3KBo36Uq2O4R2wzgEMDPR6H9izHX458/PdlzqlPMqioqLgtrr5EVVFRUfnBUBs4FRUVt0Vt4FRUVNwWtYFTUVFxW9QGTkVFxW1RGzgVFRW3RW3gVFRU3Ba1gVNRUXFb/g9pc928CS2BXgAAAABJRU5ErkJggg==\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "jsonl_fn = \"samples/samples/pubtabnet/synthetic_data.jsonl\"\n",
    "\n",
    "# Get the first PubTabnet annotation\n",
    "img_ann = jsonlines.open(jsonl_fn).read()\n",
    "\n",
    "# Show the corresponding image\n",
    "img_filename = img_ann[\"filename\"]\n",
    "split = img_ann[\"split\"]\n",
    "img_fn = os.path.join(\"samples/pubtabnet/images\", split, img_filename)\n",
    "img = mpimg.imread(img_fn)\n",
    "imgplot = plt.imshow(img)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "id": "b85f5f4f-38fb-4e16-a37a-bc082c81eeb4",
   "metadata": {
    "scrolled": true,
    "tags": []
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "{\n",
      "  \"filename\": \"image_000001_1634629104.265115.png\",\n",
      "  \"html\": {\n",
      "    \"cells\": [\n",
      "      {\n",
      "        \"bbox\": [\n",
      "          8.533333333333333,\n",
      "          11.100823045267491,\n",
      "          65.73333333333333,\n",
      "          27.325102880658438,\n",
      "          2\n",
      "        ],\n",
      "        \"cell_id\": 0,\n",
      "        \"is_header\": true,\n",
      "        \"tokens\": [\n",
      "          \"P\",\n",
      "          \"-\",\n",
      "          \"v\",\n",
      "          \"a\",\n",
      "          \"l\",\n",
      "          \"u\",\n",
      "          \"e\"\n",
      "        ]\n",
      "      },\n",
      "      {\n",
      "        \"bbox\": [\n",
      "          107.52000000000001,\n",
      "          11.100823045267491,\n",
      "          118.94666666666667,\n",
      "          27.325102880658438,\n",
      "          2\n",
      "        ],\n",
      "        \"cell_id\": 1,\n",
      "        \"is_header\": true,\n",
      "        \"tokens\": [\n",
      "          \"N\"\n",
      "        ]\n",
      "      },\n",
      "      {\n",
      "        \"bbox\": [\n",
      "          181.76000000000002,\n",
      "          11.100823045267491,\n",
      "          267.92,\n",
      "          27.325102880658438,\n",
      "          2\n",
      "        ],\n",
      "        \"cell_id\": 2,\n",
      "        \"is_header\": true,\n",
      "        \"tokens\": [\n",
      "          \"R\",\n",
      "          \"e\",\n",
      "          \"f\",\n",
      "          \"e\",\n",
      "          \"r\",\n",
      "          \"e\",\n",
      "          \"n\",\n",
      "          \"c\",\n",
      "          \"e\",\n",
      "          \"s\"\n",
      "        ]\n",
      "      },\n",
      "      {\n",
      "        \"bbox\": [\n",
      "          290.9866666666667,\n",
      "          11.100823045267491,\n",
      "          349.12,\n",
      "          27.325102880658438,\n",
      "          2\n",
      "        ],\n",
      "        \"cell_id\": 3,\n",
      "        \"is_header\": true,\n",
      "        \"tokens\": [\n",
      "          \"p\",\n",
      "          \"-\",\n",
      "          \"V\",\n",
      "          \"a\",\n",
      "          \"l\",\n",
      "          \"u\",\n",
      "          \"e\"\n",
      "        ]\n",
      "      },\n",
      "      {\n",
      "        \"bbox\": [\n",
      "          371.20000000000005,\n",
      "          11.100823045267491,\n",
      "          453.0266666666667,\n",
      "          27.325102880658438,\n",
      "          2\n",
      "        ],\n",
      "        \"cell_id\": 4,\n",
      "        \"is_header\": true,\n",
      "        \"span\": {\n",
      "          \"colspan\": 2,\n",
      "          \"rowspan\": 1,\n",
      "          \"span_type\": \"colspan\"\n",
      "        },\n",
      "        \"tokens\": [\n",
      "          \"S\",\n",
      "          \"e\",\n",
      "          \"n\",\n",
      "          \"s\",\n",
      "          \"i\",\n",
      "          \"t\",\n",
      "          \"i\",\n",
      "          \"v\",\n",
      "          \"i\",\n",
      "          \"t\",\n",
      "          \"y\"\n",
      "        ]\n",
      "      },\n",
      "      {\n",
      "        \"bbox\": [\n",
      "          8.533333333333333,\n",
      "          46.111111111111114,\n",
      "          24.240000000000002,\n",
      "          62.33539094650206,\n",
      "          2\n",
      "        ],\n",
      "        \"cell_id\": 5,\n",
      "        \"is_header\": true,\n",
      "        \"tokens\": [\n",
      "          \"d\",\n",
      "          \"f\"\n",
      "        ]\n",
      "      },\n",
      "      {\n",
      "        \"bbox\": [\n",
      "          107.52000000000001,\n",
      "          46.111111111111114,\n",
      "          117.92,\n",
      "          62.33539094650206,\n",
      "          2\n",
      "        ],\n",
      "        \"cell_id\": 6,\n",
      "        \"is_header\": true,\n",
      "        \"tokens\": [\n",
      "          \"B\"\n",
      "        ]\n",
      "      },\n",
      "      {\n",
      "        \"bbox\": [\n",
      "          181.76000000000002,\n",
      "          46.111111111111114,\n",
      "          203.88000000000002,\n",
      "          62.33539094650206,\n",
      "          2\n",
      "        ],\n",
      "        \"cell_id\": 7,\n",
      "        \"is_header\": true,\n",
      "        \"tokens\": [\n",
      "          \"O\",\n",
      "          \"R\"\n",
      "        ]\n",
      "      },\n",
      "      {\n",
      "        \"bbox\": [\n",
      "          282.4533333333334,\n",
      "          36.718106995884774,\n",
      "          362.6666666666667,\n",
      "          72.58230452674897,\n",
      "          1\n",
      "        ],\n",
      "        \"cell_id\": 8,\n",
      "        \"is_header\": true,\n",
      "        \"tokens\": []\n",
      "      },\n",
      "      {\n",
      "        \"bbox\": [\n",
      "          371.20000000000005,\n",
      "          46.96502057613169,\n",
      "          401.05333333333334,\n",
      "          63.18930041152264,\n",
      "          2\n",
      "        ],\n",
      "        \"cell_id\": 9,\n",
      "        \"is_header\": true,\n",
      "        \"span\": {\n",
      "          \"colspan\": 2,\n",
      "          \"rowspan\": 1,\n",
      "          \"span_type\": \"colspan\"\n",
      "        },\n",
      "        \"tokens\": [\n",
      "          \"N\",\n",
      "          \" \",\n",
      "          \"%\"\n",
      "        ]\n",
      "      },\n",
      "      {\n",
      "        \"bbox\": [\n",
      "          8.533333333333333,\n",
      "          82.8292181069959,\n",
      "          75.88000000000001,\n",
      "          99.05349794238684,\n",
      "          2\n",
      "        ],\n",
      "        \"cell_id\": 10,\n",
      "        \"is_header\": false,\n",
      "        \"tokens\": [\n",
      "          \"S\",\n",
      "          \"e\",\n",
      "          \"q\",\n",
      "          \"u\",\n",
      "          \"e\",\n",
      "          \"n\",\n",
      "          \"c\",\n",
      "          \"e\"\n",
      "        ]\n",
      "      },\n",
      "      {\n",
      "        \"bbox\": [\n",
      "          107.52000000000001,\n",
      "          82.8292181069959,\n",
      "          159.57333333333335,\n",
      "          99.05349794238684,\n",
      "          2\n",
      "        ],\n",
      "        \"cell_id\": 11,\n",
      "        \"is_header\": false,\n",
      "        \"tokens\": [\n",
      "          \"9\",\n",
      "          \"3\",\n",
      "          \".\",\n",
      "          \"7\",\n",
      "          \"1\",\n",
      "          \"%\"\n",
      "        ]\n",
      "      },\n",
      "      {\n",
      "        \"bbox\": [\n",
      "          181.76000000000002,\n",
      "          82.8292181069959,\n",
      "          233.81333333333336,\n",
      "          99.05349794238684,\n",
      "          2\n",
      "        ],\n",
      "        \"cell_id\": 12,\n",
      "        \"is_header\": false,\n",
      "        \"tokens\": [\n",
      "          \"2\",\n",
      "          \"1\",\n",
      "          \".\",\n",
      "          \"7\",\n",
      "          \"4\",\n",
      "          \"%\"\n",
      "        ]\n",
      "      },\n",
      "      {\n",
      "        \"bbox\": [\n",
      "          290.9866666666667,\n",
      "          82.8292181069959,\n",
      "          334.36,\n",
      "          99.05349794238684,\n",
      "          2\n",
      "        ],\n",
      "        \"cell_id\": 13,\n",
      "        \"is_header\": false,\n",
      "        \"tokens\": [\n",
      "          \"6\",\n",
      "          \".\",\n",
      "          \"3\",\n",
      "          \"8\",\n",
      "          \"%\"\n",
      "        ]\n",
      "      },\n",
      "      {\n",
      "        \"bbox\": [\n",
      "          371.20000000000005,\n",
      "          82.8292181069959,\n",
      "          414.5733333333334,\n",
      "          99.05349794238684,\n",
      "          2\n",
      "        ],\n",
      "        \"cell_id\": 14,\n",
      "        \"is_header\": false,\n",
      "        \"tokens\": [\n",
      "          \"9\",\n",
      "          \".\",\n",
      "          \"6\",\n",
      "          \"3\",\n",
      "          \"%\"\n",
      "        ]\n",
      "      },\n",
      "      {\n",
      "        \"bbox\": [\n",
      "          445.44000000000005,\n",
      "          82.8292181069959,\n",
      "          497.49333333333334,\n",
      "          99.05349794238684,\n",
      "          2\n",
      "        ],\n",
      "        \"cell_id\": 15,\n",
      "        \"is_header\": false,\n",
      "        \"tokens\": [\n",
      "          \"1\",\n",
      "          \"0\",\n",
      "          \".\",\n",
      "          \"4\",\n",
      "          \"1\",\n",
      "          \"%\"\n",
      "        ]\n",
      "      },\n",
      "      {\n",
      "        \"bbox\": [\n",
      "          8.533333333333333,\n",
      "          116.98559670781894,\n",
      "          76.85333333333334,\n",
      "          133.20987654320987,\n",
      "          2\n",
      "        ],\n",
      "        \"cell_id\": 16,\n",
      "        \"is_header\": false,\n",
      "        \"tokens\": [\n",
      "          \"A\",\n",
      "          \"g\",\n",
      "          \"e\",\n",
      "          \" \",\n",
      "          \"y\",\n",
      "          \"e\",\n",
      "          \"a\",\n",
      "          \"r\",\n",
      "          \"s\"\n",
      "        ]\n",
      "      },\n",
      "      {\n",
      "        \"bbox\": [\n",
      "          107.52000000000001,\n",
      "          116.98559670781894,\n",
      "          159.57333333333335,\n",
      "          133.20987654320987,\n",
      "          2\n",
      "        ],\n",
      "        \"cell_id\": 17,\n",
      "        \"is_header\": false,\n",
      "        \"tokens\": [\n",
      "          \"1\",\n",
      "          \"8\",\n",
      "          \".\",\n",
      "          \"5\",\n",
      "          \"6\",\n",
      "          \"%\"\n",
      "        ]\n",
      "      },\n",
      "      {\n",
      "        \"bbox\": [\n",
      "          181.76000000000002,\n",
      "          116.98559670781894,\n",
      "          233.81333333333336,\n",
      "          133.20987654320987,\n",
      "          2\n",
      "        ],\n",
      "        \"cell_id\": 18,\n",
      "        \"is_header\": false,\n",
      "        \"tokens\": [\n",
      "          \"6\",\n",
      "          \"6\",\n",
      "          \".\",\n",
      "          \"7\",\n",
      "          \"0\",\n",
      "          \"%\"\n",
      "        ]\n",
      "      },\n",
      "      {\n",
      "        \"bbox\": [\n",
      "          290.9866666666667,\n",
      "          116.98559670781894,\n",
      "          343.04,\n",
      "          133.20987654320987,\n",
      "          2\n",
      "        ],\n",
      "        \"cell_id\": 19,\n",
      "        \"is_header\": false,\n",
      "        \"tokens\": [\n",
      "          \"4\",\n",
      "          \"8\",\n",
      "          \".\",\n",
      "          \"9\",\n",
      "          \"7\",\n",
      "          \"%\"\n",
      "        ]\n",
      "      },\n",
      "      {\n",
      "        \"bbox\": [\n",
      "          362.6666666666667,\n",
      "          107.5925925925926,\n",
      "          436.9066666666667,\n",
      "          141.74897119341566,\n",
      "          1\n",
      "        ],\n",
      "        \"cell_id\": 20,\n",
      "        \"is_header\": false,\n",
      "        \"tokens\": []\n",
      "      },\n",
      "      {\n",
      "        \"bbox\": [\n",
      "          436.9066666666667,\n",
      "          107.5925925925926,\n",
      "          512.0,\n",
      "          141.74897119341566,\n",
      "          1\n",
      "        ],\n",
      "        \"cell_id\": 21,\n",
      "        \"is_header\": false,\n",
      "        \"tokens\": []\n",
      "      },\n",
      "      {\n",
      "        \"bbox\": [\n",
      "          8.533333333333333,\n",
      "          151.14197530864197,\n",
      "          79.06666666666668,\n",
      "          167.36625514403292,\n",
      "          2\n",
      "        ],\n",
      "        \"cell_id\": 22,\n",
      "        \"is_header\": false,\n",
      "        \"tokens\": [\n",
      "          \"F\",\n",
      "          \"r\",\n",
      "          \"e\",\n",
      "          \"q\",\n",
      "          \"u\",\n",
      "          \"e\",\n",
      "          \"n\",\n",
      "          \"c\",\n",
      "          \"y\"\n",
      "        ]\n",
      "      },\n",
      "      {\n",
      "        \"bbox\": [\n",
      "          107.52000000000001,\n",
      "          151.14197530864197,\n",
      "          159.57333333333335,\n",
      "          167.36625514403292,\n",
      "          2\n",
      "        ],\n",
      "        \"cell_id\": 23,\n",
      "        \"is_header\": false,\n",
      "        \"tokens\": [\n",
      "          \"9\",\n",
      "          \"4\",\n",
      "          \".\",\n",
      "          \"6\",\n",
      "          \"1\",\n",
      "          \"%\"\n",
      "        ]\n",
      "      },\n",
      "      {\n",
      "        \"bbox\": [\n",
      "          181.76000000000002,\n",
      "          151.14197530864197,\n",
      "          233.81333333333336,\n",
      "          167.36625514403292,\n",
      "          2\n",
      "        ],\n",
      "        \"cell_id\": 24,\n",
      "        \"is_header\": false,\n",
      "        \"tokens\": [\n",
      "          \"3\",\n",
      "          \"4\",\n",
      "          \".\",\n",
      "          \"4\",\n",
      "          \"9\",\n",
      "          \"%\"\n",
      "        ]\n",
      "      },\n",
      "      {\n",
      "        \"bbox\": [\n",
      "          282.4533333333334,\n",
      "          141.74897119341566,\n",
      "          362.6666666666667,\n",
      "          175.9053497942387,\n",
      "          1\n",
      "        ],\n",
      "        \"cell_id\": 25,\n",
      "        \"is_header\": false,\n",
      "        \"tokens\": []\n",
      "      },\n",
      "      {\n",
      "        \"bbox\": [\n",
      "          371.20000000000005,\n",
      "          151.14197530864197,\n",
      "          423.2533333333334,\n",
      "          167.36625514403292,\n",
      "          2\n",
      "        ],\n",
      "        \"cell_id\": 26,\n",
      "        \"is_header\": false,\n",
      "        \"tokens\": [\n",
      "          \"5\",\n",
      "          \"1\",\n",
      "          \".\",\n",
      "          \"5\",\n",
      "          \"3\",\n",
      "          \"%\"\n",
      "        ]\n",
      "      },\n",
      "      {\n",
      "        \"bbox\": [\n",
      "          445.44000000000005,\n",
      "          151.14197530864197,\n",
      "          497.49333333333334,\n",
      "          167.36625514403292,\n",
      "          2\n",
      "        ],\n",
      "        \"cell_id\": 27,\n",
      "        \"is_header\": false,\n",
      "        \"tokens\": [\n",
      "          \"3\",\n",
      "          \"0\",\n",
      "          \".\",\n",
      "          \"4\",\n",
      "          \"8\",\n",
      "          \"%\"\n",
      "        ]\n",
      "      },\n",
      "      {\n",
      "        \"bbox\": [\n",
      "          8.533333333333333,\n",
      "          185.29835390946502,\n",
      "          48.973333333333336,\n",
      "          201.52263374485597,\n",
      "          2\n",
      "        ],\n",
      "        \"cell_id\": 28,\n",
      "        \"is_header\": false,\n",
      "        \"tokens\": [\n",
      "          \"S\",\n",
      "          \"t\",\n",
      "          \"r\",\n",
      "          \"a\",\n",
      "          \"i\",\n",
      "          \"n\"\n",
      "        ]\n",
      "      },\n",
      "      {\n",
      "        \"bbox\": [\n",
      "          107.52000000000001,\n",
      "          185.29835390946502,\n",
      "          150.89333333333335,\n",
      "          201.52263374485597,\n",
      "          2\n",
      "        ],\n",
      "        \"cell_id\": 29,\n",
      "        \"is_header\": false,\n",
      "        \"tokens\": [\n",
      "          \"5\",\n",
      "          \"3\",\n",
      "          \".\",\n",
      "          \"8\",\n",
      "          \"%\"\n",
      "        ]\n",
      "      },\n",
      "      {\n",
      "        \"bbox\": [\n",
      "          181.76000000000002,\n",
      "          185.29835390946502,\n",
      "          233.81333333333336,\n",
      "          201.52263374485597,\n",
      "          2\n",
      "        ],\n",
      "        \"cell_id\": 30,\n",
      "        \"is_header\": false,\n",
      "        \"tokens\": [\n",
      "          \"9\",\n",
      "          \"5\",\n",
      "          \".\",\n",
      "          \"6\",\n",
      "          \"0\",\n",
      "          \"%\"\n",
      "        ]\n",
      "      },\n",
      "      {\n",
      "        \"bbox\": [\n",
      "          290.9866666666667,\n",
      "          185.29835390946502,\n",
      "          343.04,\n",
      "          201.52263374485597,\n",
      "          2\n",
      "        ],\n",
      "        \"cell_id\": 31,\n",
      "        \"is_header\": false,\n",
      "        \"tokens\": [\n",
      "          \"1\",\n",
      "          \"1\",\n",
      "          \".\",\n",
      "          \"7\",\n",
      "          \"6\",\n",
      "          \"%\"\n",
      "        ]\n",
      "      },\n",
      "      {\n",
      "        \"bbox\": [\n",
      "          371.20000000000005,\n",
      "          185.29835390946502,\n",
      "          423.2533333333334,\n",
      "          201.52263374485597,\n",
      "          2\n",
      "        ],\n",
      "        \"cell_id\": 32,\n",
      "        \"is_header\": false,\n",
      "        \"tokens\": [\n",
      "          \"4\",\n",
      "          \"4\",\n",
      "          \".\",\n",
      "          \"3\",\n",
      "          \"3\",\n",
      "          \"%\"\n",
      "        ]\n",
      "      },\n",
      "      {\n",
      "        \"bbox\": [\n",
      "          445.44000000000005,\n",
      "          185.29835390946502,\n",
      "          497.49333333333334,\n",
      "          201.52263374485597,\n",
      "          2\n",
      "        ],\n",
      "        \"cell_id\": 33,\n",
      "        \"is_header\": false,\n",
      "        \"tokens\": [\n",
      "          \"8\",\n",
      "          \"6\",\n",
      "          \".\",\n",
      "          \"1\",\n",
      "          \"7\",\n",
      "          \"%\"\n",
      "        ]\n",
      "      },\n",
      "      {\n",
      "        \"bbox\": [\n",
      "          8.533333333333333,\n",
      "          219.45473251028807,\n",
      "          85.02666666666667,\n",
      "          235.67901234567904,\n",
      "          2\n",
      "        ],\n",
      "        \"cell_id\": 34,\n",
      "        \"is_header\": false,\n",
      "        \"tokens\": [\n",
      "          \"M\",\n",
      "          \"e\",\n",
      "          \"a\",\n",
      "          \"n\",\n",
      "          \" \",\n",
      "          \"\\u00b1\",\n",
      "          \" \",\n",
      "          \"S\",\n",
      "          \"D\"\n",
      "        ]\n",
      "      },\n",
      "      {\n",
      "        \"bbox\": [\n",
      "          98.98666666666668,\n",
      "          210.06172839506175,\n",
      "          173.2266666666667,\n",
      "          244.2181069958848,\n",
      "          1\n",
      "        ],\n",
      "        \"cell_id\": 35,\n",
      "        \"is_header\": false,\n",
      "        \"tokens\": []\n",
      "      },\n",
      "      {\n",
      "        \"bbox\": [\n",
      "          181.76000000000002,\n",
      "          219.45473251028807,\n",
      "          233.81333333333336,\n",
      "          235.67901234567904,\n",
      "          2\n",
      "        ],\n",
      "        \"cell_id\": 36,\n",
      "        \"is_header\": false,\n",
      "        \"tokens\": [\n",
      "          \"9\",\n",
      "          \"3\",\n",
      "          \".\",\n",
      "          \"4\",\n",
      "          \"1\",\n",
      "          \"%\"\n",
      "        ]\n",
      "      },\n",
      "      {\n",
      "        \"bbox\": [\n",
      "          290.9866666666667,\n",
      "          219.45473251028807,\n",
      "          343.04,\n",
      "          235.67901234567904,\n",
      "          2\n",
      "        ],\n",
      "        \"cell_id\": 37,\n",
      "        \"is_header\": false,\n",
      "        \"tokens\": [\n",
      "          \"7\",\n",
      "          \"6\",\n",
      "          \".\",\n",
      "          \"1\",\n",
      "          \"1\",\n",
      "          \"%\"\n",
      "        ]\n",
      "      },\n",
      "      {\n",
      "        \"bbox\": [\n",
      "          371.20000000000005,\n",
      "          219.45473251028807,\n",
      "          423.2533333333334,\n",
      "          235.67901234567904,\n",
      "          2\n",
      "        ],\n",
      "        \"cell_id\": 38,\n",
      "        \"is_header\": false,\n",
      "        \"tokens\": [\n",
      "          \"8\",\n",
      "          \"1\",\n",
      "          \".\",\n",
      "          \"7\",\n",
      "          \"5\",\n",
      "          \"%\"\n",
      "        ]\n",
      "      },\n",
      "      {\n",
      "        \"bbox\": [\n",
      "          445.44000000000005,\n",
      "          219.45473251028807,\n",
      "          497.49333333333334,\n",
      "          235.67901234567904,\n",
      "          2\n",
      "        ],\n",
      "        \"cell_id\": 39,\n",
      "        \"is_header\": false,\n",
      "        \"tokens\": [\n",
      "          \"7\",\n",
      "          \"5\",\n",
      "          \".\",\n",
      "          \"1\",\n",
      "          \"6\",\n",
      "          \"%\"\n",
      "        ]\n",
      "      },\n",
      "      {\n",
      "        \"bbox\": [\n",
      "          8.533333333333333,\n",
      "          253.61111111111111,\n",
      "          17.200000000000003,\n",
      "          269.83539094650206,\n",
      "          2\n",
      "        ],\n",
      "        \"cell_id\": 40,\n",
      "        \"is_header\": false,\n",
      "        \"tokens\": [\n",
      "          \"p\"\n",
      "        ]\n",
      "      },\n",
      "      {\n",
      "        \"bbox\": [\n",
      "          107.52000000000001,\n",
      "          253.61111111111111,\n",
      "          159.57333333333335,\n",
      "          269.83539094650206,\n",
      "          2\n",
      "        ],\n",
      "        \"cell_id\": 41,\n",
      "        \"is_header\": false,\n",
      "        \"tokens\": [\n",
      "          \"7\",\n",
      "          \"4\",\n",
      "          \".\",\n",
      "          \"7\",\n",
      "          \"5\",\n",
      "          \"%\"\n",
      "        ]\n",
      "      },\n",
      "      {\n",
      "        \"bbox\": [\n",
      "          173.2266666666667,\n",
      "          244.2181069958848,\n",
      "          282.4533333333334,\n",
      "          278.3744855967078,\n",
      "          1\n",
      "        ],\n",
      "        \"cell_id\": 42,\n",
      "        \"is_header\": false,\n",
      "        \"tokens\": []\n",
      "      },\n",
      "      {\n",
      "        \"bbox\": [\n",
      "          290.9866666666667,\n",
      "          253.61111111111111,\n",
      "          343.04,\n",
      "          269.83539094650206,\n",
      "          2\n",
      "        ],\n",
      "        \"cell_id\": 43,\n",
      "        \"is_header\": false,\n",
      "        \"tokens\": [\n",
      "          \"6\",\n",
      "          \"0\",\n",
      "          \".\",\n",
      "          \"7\",\n",
      "          \"9\",\n",
      "          \"%\"\n",
      "        ]\n",
      "      },\n",
      "      {\n",
      "        \"bbox\": [\n",
      "          371.20000000000005,\n",
      "          253.61111111111111,\n",
      "          423.2533333333334,\n",
      "          269.83539094650206,\n",
      "          2\n",
      "        ],\n",
      "        \"cell_id\": 44,\n",
      "        \"is_header\": false,\n",
      "        \"tokens\": [\n",
      "          \"2\",\n",
      "          \"9\",\n",
      "          \".\",\n",
      "          \"9\",\n",
      "          \"6\",\n",
      "          \"%\"\n",
      "        ]\n",
      "      },\n",
      "      {\n",
      "        \"bbox\": [\n",
      "          436.9066666666667,\n",
      "          244.2181069958848,\n",
      "          512.0,\n",
      "          278.3744855967078,\n",
      "          1\n",
      "        ],\n",
      "        \"cell_id\": 45,\n",
      "        \"is_header\": false,\n",
      "        \"tokens\": []\n",
      "      },\n",
      "      {\n",
      "        \"bbox\": [\n",
      "          8.533333333333333,\n",
      "          287.76748971193416,\n",
      "          17.866666666666667,\n",
      "          303.99176954732513,\n",
      "          2\n",
      "        ],\n",
      "        \"cell_id\": 46,\n",
      "        \"is_header\": false,\n",
      "        \"tokens\": [\n",
      "          \"A\"\n",
      "        ]\n",
      "      },\n",
      "      {\n",
      "        \"bbox\": [\n",
      "          98.98666666666668,\n",
      "          278.3744855967078,\n",
      "          173.2266666666667,\n",
      "          312.5308641975309,\n",
      "          1\n",
      "        ],\n",
      "        \"cell_id\": 47,\n",
      "        \"is_header\": false,\n",
      "        \"tokens\": []\n",
      "      },\n",
      "      {\n",
      "        \"bbox\": [\n",
      "          181.76000000000002,\n",
      "          287.76748971193416,\n",
      "          216.44000000000003,\n",
      "          303.99176954732513,\n",
      "          2\n",
      "        ],\n",
      "        \"cell_id\": 48,\n",
      "        \"is_header\": false,\n",
      "        \"tokens\": [\n",
      "          \"3\",\n",
      "          \".\",\n",
      "          \"2\",\n",
      "          \"%\"\n",
      "        ]\n",
      "      },\n",
      "      {\n",
      "        \"bbox\": [\n",
      "          290.9866666666667,\n",
      "          287.76748971193416,\n",
      "          334.36,\n",
      "          303.99176954732513,\n",
      "          2\n",
      "        ],\n",
      "        \"cell_id\": 49,\n",
      "        \"is_header\": false,\n",
      "        \"tokens\": [\n",
      "          \"7\",\n",
      "          \"2\",\n",
      "          \".\",\n",
      "          \"8\",\n",
      "          \"%\"\n",
      "        ]\n",
      "      },\n",
      "      {\n",
      "        \"bbox\": [\n",
      "          371.20000000000005,\n",
      "          287.76748971193416,\n",
      "          423.2533333333334,\n",
      "          303.99176954732513,\n",
      "          2\n",
      "        ],\n",
      "        \"cell_id\": 50,\n",
      "        \"is_header\": false,\n",
      "        \"tokens\": [\n",
      "          \"3\",\n",
      "          \"1\",\n",
      "          \".\",\n",
      "          \"4\",\n",
      "          \"5\",\n",
      "          \"%\"\n",
      "        ]\n",
      "      },\n",
      "      {\n",
      "        \"bbox\": [\n",
      "          445.44000000000005,\n",
      "          287.76748971193416,\n",
      "          497.49333333333334,\n",
      "          303.99176954732513,\n",
      "          2\n",
      "        ],\n",
      "        \"cell_id\": 51,\n",
      "        \"is_header\": false,\n",
      "        \"tokens\": [\n",
      "          \"9\",\n",
      "          \"3\",\n",
      "          \".\",\n",
      "          \"4\",\n",
      "          \"4\",\n",
      "          \"%\"\n",
      "        ]\n",
      "      },\n",
      "      {\n",
      "        \"bbox\": [\n",
      "          8.533333333333333,\n",
      "          321.92386831275724,\n",
      "          75.88000000000001,\n",
      "          338.14814814814815,\n",
      "          2\n",
      "        ],\n",
      "        \"cell_id\": 52,\n",
      "        \"is_header\": false,\n",
      "        \"tokens\": [\n",
      "          \"S\",\n",
      "          \"e\",\n",
      "          \"q\",\n",
      "          \"u\",\n",
      "          \"e\",\n",
      "          \"n\",\n",
      "          \"c\",\n",
      "          \"e\"\n",
      "        ]\n",
      "      },\n",
      "      {\n",
      "        \"bbox\": [\n",
      "          107.52000000000001,\n",
      "          321.92386831275724,\n",
      "          159.57333333333335,\n",
      "          338.14814814814815,\n",
      "          2\n",
      "        ],\n",
      "        \"cell_id\": 53,\n",
      "        \"is_header\": false,\n",
      "        \"tokens\": [\n",
      "          \"3\",\n",
      "          \"4\",\n",
      "          \".\",\n",
      "          \"7\",\n",
      "          \"1\",\n",
      "          \"%\"\n",
      "        ]\n",
      "      },\n",
      "      {\n",
      "        \"bbox\": [\n",
      "          181.76000000000002,\n",
      "          321.92386831275724,\n",
      "          233.81333333333336,\n",
      "          338.14814814814815,\n",
      "          2\n",
      "        ],\n",
      "        \"cell_id\": 54,\n",
      "        \"is_header\": false,\n",
      "        \"tokens\": [\n",
      "          \"8\",\n",
      "          \"9\",\n",
      "          \".\",\n",
      "          \"3\",\n",
      "          \"1\",\n",
      "          \"%\"\n",
      "        ]\n",
      "      },\n",
      "      {\n",
      "        \"bbox\": [\n",
      "          290.9866666666667,\n",
      "          321.92386831275724,\n",
      "          343.04,\n",
      "          338.14814814814815,\n",
      "          2\n",
      "        ],\n",
      "        \"cell_id\": 55,\n",
      "        \"is_header\": false,\n",
      "        \"tokens\": [\n",
      "          \"7\",\n",
      "          \"8\",\n",
      "          \".\",\n",
      "          \"2\",\n",
      "          \"7\",\n",
      "          \"%\"\n",
      "        ]\n",
      "      },\n",
      "      {\n",
      "        \"bbox\": [\n",
      "          371.20000000000005,\n",
      "          321.92386831275724,\n",
      "          423.2533333333334,\n",
      "          338.14814814814815,\n",
      "          2\n",
      "        ],\n",
      "        \"cell_id\": 56,\n",
      "        \"is_header\": false,\n",
      "        \"tokens\": [\n",
      "          \"6\",\n",
      "          \"6\",\n",
      "          \".\",\n",
      "          \"1\",\n",
      "          \"8\",\n",
      "          \"%\"\n",
      "        ]\n",
      "      },\n",
      "      {\n",
      "        \"bbox\": [\n",
      "          445.44000000000005,\n",
      "          321.92386831275724,\n",
      "          497.49333333333334,\n",
      "          338.14814814814815,\n",
      "          2\n",
      "        ],\n",
      "        \"cell_id\": 57,\n",
      "        \"is_header\": false,\n",
      "        \"tokens\": [\n",
      "          \"2\",\n",
      "          \"4\",\n",
      "          \".\",\n",
      "          \"2\",\n",
      "          \"4\",\n",
      "          \"%\"\n",
      "        ]\n",
      "      },\n",
      "      {\n",
      "        \"bbox\": [\n",
      "          8.533333333333333,\n",
      "          356.08024691358025,\n",
      "          49.52,\n",
      "          372.3045267489712,\n",
      "          2\n",
      "        ],\n",
      "        \"cell_id\": 58,\n",
      "        \"is_header\": false,\n",
      "        \"tokens\": [\n",
      "          \"M\",\n",
      "          \"o\",\n",
      "          \"d\",\n",
      "          \"e\",\n",
      "          \"l\"\n",
      "        ]\n",
      "      },\n",
      "      {\n",
      "        \"bbox\": [\n",
      "          107.52000000000001,\n",
      "          356.08024691358025,\n",
      "          150.89333333333335,\n",
      "          372.3045267489712,\n",
      "          2\n",
      "        ],\n",
      "        \"cell_id\": 59,\n",
      "        \"is_header\": false,\n",
      "        \"tokens\": [\n",
      "          \"7\",\n",
      "          \"6\",\n",
      "          \".\",\n",
      "          \"2\",\n",
      "          \"%\"\n",
      "        ]\n",
      "      },\n",
      "      {\n",
      "        \"bbox\": [\n",
      "          181.76000000000002,\n",
      "          356.08024691358025,\n",
      "          233.81333333333336,\n",
      "          372.3045267489712,\n",
      "          2\n",
      "        ],\n",
      "        \"cell_id\": 60,\n",
      "        \"is_header\": false,\n",
      "        \"tokens\": [\n",
      "          \"5\",\n",
      "          \"3\",\n",
      "          \".\",\n",
      "          \"9\",\n",
      "          \"4\",\n",
      "          \"%\"\n",
      "        ]\n",
      "      },\n",
      "      {\n",
      "        \"bbox\": [\n",
      "          290.9866666666667,\n",
      "          356.08024691358025,\n",
      "          343.04,\n",
      "          372.3045267489712,\n",
      "          2\n",
      "        ],\n",
      "        \"cell_id\": 61,\n",
      "        \"is_header\": false,\n",
      "        \"tokens\": [\n",
      "          \"1\",\n",
      "          \"5\",\n",
      "          \".\",\n",
      "          \"6\",\n",
      "          \"1\",\n",
      "          \"%\"\n",
      "        ]\n",
      "      },\n",
      "      {\n",
      "        \"bbox\": [\n",
      "          371.20000000000005,\n",
      "          356.08024691358025,\n",
      "          414.5733333333334,\n",
      "          372.3045267489712,\n",
      "          2\n",
      "        ],\n",
      "        \"cell_id\": 62,\n",
      "        \"is_header\": false,\n",
      "        \"tokens\": [\n",
      "          \"7\",\n",
      "          \".\",\n",
      "          \"3\",\n",
      "          \"6\",\n",
      "          \"%\"\n",
      "        ]\n",
      "      },\n",
      "      {\n",
      "        \"bbox\": [\n",
      "          436.9066666666667,\n",
      "          346.6872427983539,\n",
      "          512.0,\n",
      "          380.843621399177,\n",
      "          1\n",
      "        ],\n",
      "        \"cell_id\": 63,\n",
      "        \"is_header\": false,\n",
      "        \"tokens\": []\n",
      "      },\n",
      "      {\n",
      "        \"bbox\": [\n",
      "          0.0,\n",
      "          380.843621399177,\n",
      "          0.0,\n",
      "          380.843621399177,\n",
      "          2\n",
      "        ],\n",
      "        \"cell_id\": 64,\n",
      "        \"is_header\": false,\n",
      "        \"tokens\": [\n",
      "          \"<\",\n",
      "          \"F\",\n",
      "          \"L\",\n",
      "          \"O\",\n",
      "          \"A\",\n",
      "          \"T\",\n",
      "          \">\"\n",
      "        ]\n",
      "      },\n",
      "      {\n",
      "        \"bbox\": [\n",
      "          107.52000000000001,\n",
      "          390.23662551440333,\n",
      "          159.57333333333335,\n",
      "          406.46090534979425,\n",
      "          2\n",
      "        ],\n",
      "        \"cell_id\": 65,\n",
      "        \"is_header\": false,\n",
      "        \"tokens\": [\n",
      "          \"3\",\n",
      "          \"5\",\n",
      "          \".\",\n",
      "          \"1\",\n",
      "          \"4\",\n",
      "          \"%\"\n",
      "        ]\n",
      "      },\n",
      "      {\n",
      "        \"bbox\": [\n",
      "          181.76000000000002,\n",
      "          390.23662551440333,\n",
      "          233.81333333333336,\n",
      "          406.46090534979425,\n",
      "          2\n",
      "        ],\n",
      "        \"cell_id\": 66,\n",
      "        \"is_header\": false,\n",
      "        \"tokens\": [\n",
      "          \"5\",\n",
      "          \"2\",\n",
      "          \".\",\n",
      "          \"9\",\n",
      "          \"7\",\n",
      "          \"%\"\n",
      "        ]\n",
      "      },\n",
      "      {\n",
      "        \"bbox\": [\n",
      "          290.9866666666667,\n",
      "          390.23662551440333,\n",
      "          343.04,\n",
      "          406.46090534979425,\n",
      "          2\n",
      "        ],\n",
      "        \"cell_id\": 67,\n",
      "        \"is_header\": false,\n",
      "        \"tokens\": [\n",
      "          \"5\",\n",
      "          \"9\",\n",
      "          \".\",\n",
      "          \"4\",\n",
      "          \"0\",\n",
      "          \"%\"\n",
      "        ]\n",
      "      },\n",
      "      {\n",
      "        \"bbox\": [\n",
      "          371.20000000000005,\n",
      "          390.23662551440333,\n",
      "          414.5733333333334,\n",
      "          406.46090534979425,\n",
      "          2\n",
      "        ],\n",
      "        \"cell_id\": 68,\n",
      "        \"is_header\": false,\n",
      "        \"tokens\": [\n",
      "          \"3\",\n",
      "          \".\",\n",
      "          \"3\",\n",
      "          \"7\",\n",
      "          \"%\"\n",
      "        ]\n",
      "      },\n",
      "      {\n",
      "        \"bbox\": [\n",
      "          445.44000000000005,\n",
      "          390.23662551440333,\n",
      "          497.49333333333334,\n",
      "          406.46090534979425,\n",
      "          2\n",
      "        ],\n",
      "        \"cell_id\": 69,\n",
      "        \"is_header\": false,\n",
      "        \"tokens\": [\n",
      "          \"5\",\n",
      "          \"0\",\n",
      "          \".\",\n",
      "          \"2\",\n",
      "          \"2\",\n",
      "          \"%\"\n",
      "        ]\n",
      "      }\n",
      "    ],\n",
      "    \"structure\": {\n",
      "      \"tokens\": [\n",
      "        \"<thead>\",\n",
      "        \"<tr>\",\n",
      "        \"<td>\",\n",
      "        \"</td>\",\n",
      "        \"<td>\",\n",
      "        \"</td>\",\n",
      "        \"<td>\",\n",
      "        \"</td>\",\n",
      "        \"<td>\",\n",
      "        \"</td>\",\n",
      "        \"<td\",\n",
      "        \" colspan=\\\"2\\\"\",\n",
      "        \">\",\n",
      "        \"</td>\",\n",
      "        \"</tr>\",\n",
      "        \"<tr>\",\n",
      "        \"<td>\",\n",
      "        \"</td>\",\n",
      "        \"<td>\",\n",
      "        \"</td>\",\n",
      "        \"<td>\",\n",
      "        \"</td>\",\n",
      "        \"<td>\",\n",
      "        \"</td>\",\n",
      "        \"<td\",\n",
      "        \" colspan=\\\"2\\\"\",\n",
      "        \">\",\n",
      "        \"</td>\",\n",
      "        \"</tr>\",\n",
      "        \"</thead>\",\n",
      "        \"<tbody>\",\n",
      "        \"<tr>\",\n",
      "        \"<td>\",\n",
      "        \"</td>\",\n",
      "        \"<td>\",\n",
      "        \"</td>\",\n",
      "        \"<td>\",\n",
      "        \"</td>\",\n",
      "        \"<td>\",\n",
      "        \"</td>\",\n",
      "        \"<td>\",\n",
      "        \"</td>\",\n",
      "        \"<td>\",\n",
      "        \"</td>\",\n",
      "        \"</tr>\",\n",
      "        \"<tr>\",\n",
      "        \"<td>\",\n",
      "        \"</td>\",\n",
      "        \"<td>\",\n",
      "        \"</td>\",\n",
      "        \"<td>\",\n",
      "        \"</td>\",\n",
      "        \"<td>\",\n",
      "        \"</td>\",\n",
      "        \"<td>\",\n",
      "        \"</td>\",\n",
      "        \"<td>\",\n",
      "        \"</td>\",\n",
      "        \"</tr>\",\n",
      "        \"<tr>\",\n",
      "        \"<td>\",\n",
      "        \"</td>\",\n",
      "        \"<td>\",\n",
      "        \"</td>\",\n",
      "        \"<td>\",\n",
      "        \"</td>\",\n",
      "        \"<td>\",\n",
      "        \"</td>\",\n",
      "        \"<td>\",\n",
      "        \"</td>\",\n",
      "        \"<td>\",\n",
      "        \"</td>\",\n",
      "        \"</tr>\",\n",
      "        \"<tr>\",\n",
      "        \"<td>\",\n",
      "        \"</td>\",\n",
      "        \"<td>\",\n",
      "        \"</td>\",\n",
      "        \"<td>\",\n",
      "        \"</td>\",\n",
      "        \"<td>\",\n",
      "        \"</td>\",\n",
      "        \"<td>\",\n",
      "        \"</td>\",\n",
      "        \"<td>\",\n",
      "        \"</td>\",\n",
      "        \"</tr>\",\n",
      "        \"<tr>\",\n",
      "        \"<td>\",\n",
      "        \"</td>\",\n",
      "        \"<td>\",\n",
      "        \"</td>\",\n",
      "        \"<td>\",\n",
      "        \"</td>\",\n",
      "        \"<td>\",\n",
      "        \"</td>\",\n",
      "        \"<td>\",\n",
      "        \"</td>\",\n",
      "        \"<td>\",\n",
      "        \"</td>\",\n",
      "        \"</tr>\",\n",
      "        \"<tr>\",\n",
      "        \"<td>\",\n",
      "        \"</td>\",\n",
      "        \"<td>\",\n",
      "        \"</td>\",\n",
      "        \"<td>\",\n",
      "        \"</td>\",\n",
      "        \"<td>\",\n",
      "        \"</td>\",\n",
      "        \"<td>\",\n",
      "        \"</td>\",\n",
      "        \"<td>\",\n",
      "        \"</td>\",\n",
      "        \"</tr>\",\n",
      "        \"<tr>\",\n",
      "        \"<td>\",\n",
      "        \"</td>\",\n",
      "        \"<td>\",\n",
      "        \"</td>\",\n",
      "        \"<td>\",\n",
      "        \"</td>\",\n",
      "        \"<td>\",\n",
      "        \"</td>\",\n",
      "        \"<td>\",\n",
      "        \"</td>\",\n",
      "        \"<td>\",\n",
      "        \"</td>\",\n",
      "        \"</tr>\",\n",
      "        \"<tr>\",\n",
      "        \"<td>\",\n",
      "        \"</td>\",\n",
      "        \"<td>\",\n",
      "        \"</td>\",\n",
      "        \"<td>\",\n",
      "        \"</td>\",\n",
      "        \"<td>\",\n",
      "        \"</td>\",\n",
      "        \"<td>\",\n",
      "        \"</td>\",\n",
      "        \"<td>\",\n",
      "        \"</td>\",\n",
      "        \"</tr>\",\n",
      "        \"<tr>\",\n",
      "        \"<td>\",\n",
      "        \"</td>\",\n",
      "        \"<td>\",\n",
      "        \"</td>\",\n",
      "        \"<td>\",\n",
      "        \"</td>\",\n",
      "        \"<td>\",\n",
      "        \"</td>\",\n",
      "        \"<td>\",\n",
      "        \"</td>\",\n",
      "        \"<td>\",\n",
      "        \"</td>\",\n",
      "        \"</tr>\",\n",
      "        \"<tr>\",\n",
      "        \"<td>\",\n",
      "        \"</td>\",\n",
      "        \"<td>\",\n",
      "        \"</td>\",\n",
      "        \"<td>\",\n",
      "        \"</td>\",\n",
      "        \"<td>\",\n",
      "        \"</td>\",\n",
      "        \"<td>\",\n",
      "        \"</td>\",\n",
      "        \"<td>\",\n",
      "        \"</td>\",\n",
      "        \"</tr>\",\n",
      "        \"</tbody>\"\n",
      "      ]\n",
      "    }\n",
      "  },\n",
      "  \"imgid\": 1,\n",
      "  \"split\": \"train\"\n",
      "}\n"
     ]
    }
   ],
   "source": [
    "# Print the annotation\n",
    "img_ann_pretty = json.dumps(img_ann, sort_keys=True, indent=\"  \")\n",
    "print(img_ann_pretty)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "8cc8c27a-cff9-4219-bc19-fc6d9fed3075",
   "metadata": {
    "tags": []
   },
   "source": [
    "Now let's use the annotations to reconstruct the table. This includes the following steps:\n",
    "\n",
    "1. Use the elements of the `html.structure.tokens` array to build the table structure.\n",
    "2. Use the `tokens` of each `cell` to populate the cell content.\n",
    "3. (Optional). Use the `bbox` of each `cell` to get the bounding box coordinates and class.\n",
    "\n",
    "The `format_html` function will help us to build html code out of the json annotations:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "id": "a5cd7fb0-bd63-45f9-a0e9-e66b3619c77f",
   "metadata": {},
   "outputs": [],
   "source": [
    "def format_html(ann: dict) -> str:\n",
    "    r\"\"\"\n",
    "    Formats annotations in html code\n",
    "    \"\"\"\n",
    "    cells = ann[\"html\"][\"cells\"]\n",
    "    tokens = ann[\"html\"][\"structure\"][\"tokens\"]\n",
    "    table = []\n",
    "    cell_idx = 0\n",
    "    for token in tokens:\n",
    "        if token.startswith(\"<td\"):\n",
    "            cell = cells[cell_idx]\n",
    "            content = \"\".join(cell[\"tokens\"])\n",
    "            cell_idx += 1\n",
    "        if token == \"</td>\":\n",
    "            table.append(content)\n",
    "        table.append(token)\n",
    "    table_str = \"\".join(table)\n",
    "\n",
    "    html_code = \"\"\"<html>\n",
    "    <head>\n",
    "    <meta charset=\"UTF-8\">\n",
    "    <style>\n",
    "    table, th, td {\n",
    "      border: 1px solid black;\n",
    "      font-size: 10px;\n",
    "    }\n",
    "    </style>\n",
    "    </head>\n",
    "    <body>\n",
    "    <table frame=\"hsides\" rules=\"groups\" width=\"100%%\">\n",
    "    %s\n",
    "    </table>\n",
    "    </body>\n",
    "    </html>\"\"\" % \"\".join(table_str)\n",
    "    return html_code"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "18ab9468-9b2b-4f5b-bc68-625dcc16c397",
   "metadata": {},
   "source": [
    "Finally we can apply the `format_html` on the `img_ann` object and reconstruct the html code:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "id": "65b8cedf-b582-4130-8fde-667db331c0d5",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<html>\n",
       "    <head>\n",
       "    <meta charset=\"UTF-8\">\n",
       "    <style>\n",
       "    table, th, td {\n",
       "      border: 1px solid black;\n",
       "      font-size: 10px;\n",
       "    }\n",
       "    </style>\n",
       "    </head>\n",
       "    <body>\n",
       "    <table frame=\"hsides\" rules=\"groups\" width=\"100%\">\n",
       "    <thead><tr><td>P-value</td><td>N</td><td>References</td><td>p-Value</td><td colspan=\"2\">Sensitivity</td></tr><tr><td>df</td><td>B</td><td>OR</td><td></td><td colspan=\"2\">N %</td></tr></thead><tbody><tr><td>Sequence</td><td>93.71%</td><td>21.74%</td><td>6.38%</td><td>9.63%</td><td>10.41%</td></tr><tr><td>Age years</td><td>18.56%</td><td>66.70%</td><td>48.97%</td><td></td><td></td></tr><tr><td>Frequency</td><td>94.61%</td><td>34.49%</td><td></td><td>51.53%</td><td>30.48%</td></tr><tr><td>Strain</td><td>53.8%</td><td>95.60%</td><td>11.76%</td><td>44.33%</td><td>86.17%</td></tr><tr><td>Mean ± SD</td><td></td><td>93.41%</td><td>76.11%</td><td>81.75%</td><td>75.16%</td></tr><tr><td>p</td><td>74.75%</td><td></td><td>60.79%</td><td>29.96%</td><td></td></tr><tr><td>A</td><td></td><td>3.2%</td><td>72.8%</td><td>31.45%</td><td>93.44%</td></tr><tr><td>Sequence</td><td>34.71%</td><td>89.31%</td><td>78.27%</td><td>66.18%</td><td>24.24%</td></tr><tr><td>Model</td><td>76.2%</td><td>53.94%</td><td>15.61%</td><td>7.36%</td><td></td></tr><tr><td><FLOAT></td><td>35.14%</td><td>52.97%</td><td>59.40%</td><td>3.37%</td><td>50.22%</td></tr></tbody>\n",
       "    </table>\n",
       "    </body>\n",
       "    </html>"
      ],
      "text/plain": [
       "<IPython.core.display.HTML object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "html_code = format_html(img_ann)\n",
    "display(HTML(html_code))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "ac3a1f8b-84d2-4cc7-97af-1b0ffbbddaef",
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3 (ipykernel)",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.9.10"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}