Analysis.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "id": "5e7bb888-5983-4548-addd-19cc2193243f",
   "metadata": {
    "jupyter": {
     "source_hidden": true
    }
   },
   "outputs": [],
   "source": [
    "from cloudscape import (\n",
    "    load_studies,\n",
    "    calculate_2d_cooccurence_df_plus_listofmore,\n",
    "    normalize_dict,\n",
    "    extract_svc_info,\n",
    ")\n",
    "from collections import Counter, OrderedDict\n",
    "import matplotlib.pyplot as plt\n",
    "import numpy as np\n",
    "import pandas as pd\n",
    "\n",
    "architectures = load_studies(\"./data/graphs/\")\n",
    "svc_description_fname = \"./data/services.csv\"\n",
    "\n",
    "svc_info = extract_svc_info(svc_description_fname)\n",
    "\n",
    "\n",
    "def is_ml(svc):\n",
    "    return \"ML\" in svc_info[svc][\"aws_product_categories\"].split(\",\")\n",
    "\n",
    "\n",
    "def is_analytics(svc):\n",
    "    return \"Analytics\" in svc_info[svc][\"aws_product_categories\"].split(\",\")\n",
    "\n",
    "\n",
    "def is_storage(svc):\n",
    "    return (\n",
    "        svc_info[svc][\"capability\"] == \"storage\" and svc != \"EBS\"\n",
    "    )  # since all other svcs are built on top of it.\n",
    "\n",
    "\n",
    "def cleanup(ax=None):\n",
    "    if ax is None:\n",
    "        ax = plt.gca()\n",
    "    ax.spines[\"top\"].set_visible(False)\n",
    "    ax.spines[\"right\"].set_visible(False)\n",
    "\n",
    "\n",
    "CAPABILITY_COLORS = {\n",
    "    \"compute\": \"#FF7F0E\",\n",
    "    \"storage\": \"#1F77B4\",\n",
    "    \"networking\": \"#2CA02C\",\n",
    "    \"integration\": \"#FFBB78\",\n",
    "    \"control\": \"#AEC7E8\",\n",
    "    \"other\": \"#98DF8A\",\n",
    "    \"User\": \"gray\",\n",
    "    \"ThirdParty\": \"lightgray\",\n",
    "    \"OnPremDC\": \"black\",\n",
    "    \"Partner\": \"black\",\n",
    "}\n",
    "\n",
    "schema_colors = OrderedDict()\n",
    "schema_colors[\"object\"] = \"#2CA02C\"\n",
    "schema_colors[\"specialized\"] = \"#BCBD22\"\n",
    "schema_colors[\"nosql\"] = \"#1F77B4\"\n",
    "schema_colors[\"sql\"] = \"#9467BD\"\n",
    "schema_colors[\"fs\"] = \"#D62728\""
   ]
  },
  {
   "cell_type": "markdown",
   "id": "2591dd9b-54a1-4889-a122-e16386ef98e7",
   "metadata": {},
   "source": [
    "# Dataset Analysis"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "id": "c273f64f",
   "metadata": {
    "jupyter": {
     "source_hidden": true
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "There are 396 architectures in the dataset.\n",
      "We will analyze 340 of them.\n"
     ]
    }
   ],
   "source": [
    "print(f\"There are {len(architectures)} architectures in the dataset.\")\n",
    "architectures = [a for a in architectures if a.use_for_analysis()]\n",
    "print(f\"We will analyze {len(architectures)} of them.\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "id": "2d82d214",
   "metadata": {
    "jupyter": {
     "source_hidden": true
    }
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>compute_intensive</th>\n",
       "      <th>control</th>\n",
       "      <th>interactive</th>\n",
       "      <th>data_ingestion</th>\n",
       "      <th>other</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>compute_intensive</th>\n",
       "      <td>10% (35)</td>\n",
       "      <td>0% (1)</td>\n",
       "      <td>1% (6)</td>\n",
       "      <td>2% (9)</td>\n",
       "      <td>0% (0)</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>control</th>\n",
       "      <td>0% (1)</td>\n",
       "      <td>12% (41)</td>\n",
       "      <td>2% (7)</td>\n",
       "      <td>0% (0)</td>\n",
       "      <td>0% (0)</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>interactive</th>\n",
       "      <td>1% (6)</td>\n",
       "      <td>2% (7)</td>\n",
       "      <td>22% (78)</td>\n",
       "      <td>8% (28)</td>\n",
       "      <td>0% (0)</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>data_ingestion</th>\n",
       "      <td>2% (9)</td>\n",
       "      <td>0% (0)</td>\n",
       "      <td>8% (28)</td>\n",
       "      <td>34% (116)</td>\n",
       "      <td>0% (0)</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>other</th>\n",
       "      <td>0% (0)</td>\n",
       "      <td>0% (0)</td>\n",
       "      <td>0% (0)</td>\n",
       "      <td>0% (0)</td>\n",
       "      <td>5% (17)</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Total</th>\n",
       "      <td>15% (51)</td>\n",
       "      <td>14% (49)</td>\n",
       "      <td>35% (119)</td>\n",
       "      <td>45% (153)</td>\n",
       "      <td>5% (17)</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                  compute_intensive   control interactive data_ingestion  \\\n",
       "compute_intensive          10% (35)    0% (1)      1% (6)         2% (9)   \n",
       "control                      0% (1)  12% (41)      2% (7)         0% (0)   \n",
       "interactive                  1% (6)    2% (7)    22% (78)        8% (28)   \n",
       "data_ingestion               2% (9)    0% (0)     8% (28)      34% (116)   \n",
       "other                        0% (0)    0% (0)      0% (0)         0% (0)   \n",
       "Total                      15% (51)  14% (49)   35% (119)      45% (153)   \n",
       "\n",
       "                     other  \n",
       "compute_intensive   0% (0)  \n",
       "control             0% (0)  \n",
       "interactive         0% (0)  \n",
       "data_ingestion      0% (0)  \n",
       "other              5% (17)  \n",
       "Total              5% (17)  "
      ]
     },
     "execution_count": 3,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "all_categories = set()\n",
    "for s in architectures:\n",
    "    all_categories.update(s.get_functional_goals())\n",
    "\n",
    "all_categories = list(all_categories)\n",
    "df, _, _, _ = calculate_2d_cooccurence_df_plus_listofmore(\n",
    "    architectures, all_categories, lambda x: x.get_functional_goals()\n",
    ")\n",
    "df"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "id": "569fa40f",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAaYAAAIACAYAAADJ3MhtAAAAOnRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjEwLjAsIGh0dHBzOi8vbWF0cGxvdGxpYi5vcmcvlHJYcgAAAAlwSFlzAAAPYQAAD2EBqD+naQAAU8xJREFUeJzt3XdUFdfePvBn6E0OgkhREARUELChBrGLFzUascSaKPbXBnYkxoaxBo1BjBobmqjx1aAx3ogiEb0QY4dYEJWAyY0gRgUEpAjn94c/5/WEIgcOMMDzWeus5dnTvgP38mTP7JktyOVyOYiIiCRCraYLICIiehuDiYiIJIXBREREksJgIiIiSWEwERGRpDCYiIhIUhhMREQkKQwmIiKSFAYTERFJCoOJJCk5ORmCICA0NLSmSyGiasZgohoRGhoKQRBK/CxevLhKjrlz50706NEDZmZm0NbWhq2tLSZMmIDk5OQqOR4RVYxGTRdA9VtgYCBsbW0V2pydndGsWTO8fPkSmpqaKjvWjRs3YGtriw8++AANGzZEUlISdu7ciZMnTyIuLg6WlpYqOxYRVRyDiWpU//794ebmVuIyHR0dlR7rq6++Ktbm7e0NNzc37N+/v8p6akSkHF7KI0kq7R7TkSNH4OTkBB0dHTg7O+PYsWPw8fGBjY1NhY7zZrv09HSxLT8/H8uWLUOHDh0gk8mgr6+Pbt264dy5cxU7GSJSCntMVKMyMjLw999/K7Q1atSoxHX//e9/Y+TIkXBxccHatWvx/PlzTJo0CU2aNFHqmE+fPkVhYSH++OMPBAYGAgD69OkjLs/MzMSuXbswevRoTJkyBS9evMDu3bvh5eWFy5cvo23btsqdJBEphcFENcrT07NYW2lThAUEBKBJkyaIiYmBgYEBgNeB0rNnTzRr1qzcx2zSpAny8vIAACYmJggODkbfvn3F5Q0bNkRycjK0tLTEtilTpqBVq1bYsmULdu/eXe5jEZHyGExUo7Zu3YoWLVq8c71Hjx7h5s2b+OSTT8RQAoAePXrAxcUFmZmZ5T7mqVOnkJubi/j4eHz77bfIzs5WWK6urg51dXUAQFFREdLT01FUVAQ3Nzdcv3693MchoophMFGN6tSpU6mDH9728OFDAIC9vX2xZfb29koFRq9evQC8HngxePBgODs7w8DAALNmzRLX2bdvHzZu3Ii7d++ioKBAbP/nCEIiUj0OfqB6zc7ODu3atcOBAwfEtm+//RY+Pj6ws7PD7t27ER4ejoiICPTu3RtFRUU1WC1R/cAeE9UKb+4hPXjwoNiyktqU8fLlS/GeEwAcPXoUzZs3R1hYGARBENuXL19eqeMQUfmwx0S1gqWlJZydnbF//35kZWWJ7efPn8fNmzffuf2rV6/w/PnzYu2XL1/GzZs3FS4nvrm/9PYgjEuXLuHixYuVOQUiKif2mKjWWLNmDQYPHgwPDw9MmDABz58/R0hICJydnRXCqiRZWVmwsrLCyJEj0bp1a+jr6+PmzZvYu3cvZDIZli5dKq47cOBAhIWFYciQIXj//feRlJSE7du3w8nJ6Z3HIaLKY4+Jao1Bgwbh0KFDyM/Px+LFixEWFobQ0FC0bNnynW+J0NPTw+TJk3H16lWsXLkSs2fPxsmTJzF69Ghcu3YNDg4O4ro+Pj5Ys2YN4uLi4Ovri9OnT+Pbb78t1yANIqo8QV7aQyNEtUTbtm1hamqKiIiImi6FiFSAPSaqNQoKCvDq1SuFtqioKMTFxaFnz541UxQRqRx7TFRrJCcnw9PTEx999BEsLS1x9+5dbN++HTKZDLdu3YKJiUlNl0hEKsDBD1RrNGzYEB06dMCuXbvw5MkT6Ovr4/3338e6desYSkR1CHtMREQkKbzHREREksJgIiIiSWEw4fUT/pmZmaVOt0BERNWHwQTgxYsXkMlkePHiRU2XQkRU7zGYiIhIUhhMREQkKQwmIiKSFAYTERFJCt/8QER1QmFhIQoKCmq6DCqBpqamOM9ZeTCYiKhWk8vlSE1NRXp6ek2XQmUwMjKCubm5wqzQpWEwEVGt9iaUGjduDD09vXL94aPqI5fLkZOTg7S0NACAhYXFO7dhMBFRrVVYWCiGEl/kK126uroAgLS0NDRu3Pidl/U4+IGIaq0395T09PRquBJ6lze/o/LcB2QwEVGtx8t30qfM74jBREREksJgIiIiSeHgByKqk8q6dMSZBKSNPSYionfw8fGBIAgQBAFaWlqwt7dHYGAgXr16Val9ent7q67IOoQ9JiKicujXrx/27t2LvLw8/PTTT5g5cyY0NTUREBCg1H4KCwurfbBGfn4+tLS0qvWYlcEeExFROWhra8Pc3BzNmjXD9OnT4enpiRMnTmDTpk1wcXGBvr4+rKysMGPGDGRlZYnbhYaGwsjICCdOnICTkxO0tbUxceJE7Nu3Dz/88IPYE4uKikJycjIEQUBYWBh69eoFPT09tGnTBhcvXlSoJTo6Gt26dYOuri6srKzg6+uL7OxscbmNjQ1WrVqFcePGwdDQEFOnTq22n5MqMJiIiCpAV1cX+fn5UFNTQ3BwMG7fvo19+/bh559/xqJFixTWzcnJwfr167Fr1y7cvn0bwcHBGDFiBPr164eUlBSkpKSgS5cu4vpLlizBggULEBsbixYtWmD06NHiZcPExET069cPw4YNw2+//YbDhw8jOjoas2bNUjhmUFAQ2rRpgxs3bmDp0qVV/wNRJXkNOn/+vHzgwIFyCwsLOQD5sWPHFJYXFRXJly5dKjc3N5fr6OjI+/TpI793757COk+fPpWPGTNG3qBBA7lMJpNPnDhR/uLFC6XqyMjIkAOQZ2RkVPaUiKgavXz5Un7nzh35y5cviy0DUOpHWePHj5cPHjxYLpe//rsUEREh19bWli9YsKDYukeOHJGbmJiI3/fu3SsHII+NjS11n28kJSXJAch37doltt2+fVsOQB4fHy+Xy+XySZMmyadOnaqw3X/+8x+5mpqa+HNo1qyZ3NvbW+nzrEpl/a7+qUZ7TNnZ2WjTpg22bt1a4vINGzYgODgY27dvx6VLl6Cvrw8vLy/k5uaK64wdOxa3b99GREQETp48iQsXLtS6bisRSd/JkydhYGAAHR0d9O/fHyNHjsSKFStw9uxZ9OnTB02aNEGDBg3w8ccf4+nTp8jJyRG31dLSgqura7mP9fa6b94t9+Zdc3FxcQgNDYWBgYH48fLyQlFREZKSksTt3NzcKnvKNaZGBz/0798f/fv3L3GZXC7H5s2b8emnn2Lw4MEAgP3798PMzAzHjx/HqFGjEB8fj/DwcFy5ckX8JWzZsgUDBgxAUFAQLC0tS9x3Xl4e8vLyxO+ZmZkqPjMiqmt69eqFbdu2QUtLC5aWltDQ0EBycjIGDhyI6dOnY/Xq1TA2NkZ0dDQmTZqE/Px88TU8urq6Sg140NTUFP/9ZruioiIAQFZWFqZNmwZfX99i21lbW4v/1tfXr9B5SoFk7zElJSUhNTUVnp6eYptMJkPnzp3FG4EXL16EkZGRwn8ZeHp6Qk1NDZcuXSp132vXroVMJhM/VlZWVXciRFQn6Ovrw97eHtbW1tDQeP3f9NeuXUNRURE2btyI9957Dy1atMCjR4/KtT8tLS0UFhYqXUf79u1x584d2NvbF/vUppF3ZZHscPHU1FQAgJmZmUK7mZmZuCw1NRWNGzdWWK6hoQFjY2NxnZIEBARg3rx54vfMzMxyhxPfyUWkOvJa/qCrvb09CgoKsGXLFgwaNAgxMTHYvn17uba1sbHB6dOnkZCQABMTE8hksnJt5+/vj/feew+zZs3C5MmToa+vjzt37iAiIgIhISGVOR3JkGwwVSVtbW1oa2tXaNvtSRtUXA0RVYXqCL02bdpg06ZNWL9+PQICAtC9e3esXbsW48aNe+e2U6ZMQVRUFNzc3JCVlYVz587Bxsbmndu5urri/PnzWLJkCbp16wa5XA47OzuMHDlSBWckDZINJnNzcwDA48ePFSaWevz4Mdq2bSuu8+aG4BuvXr3Cs2fPxO2JiCorNDS01GVz587F3LlzFdo+/vhj8d8+Pj7w8fEptp2pqSnOnDlTrP2fgWpkZFSsrWPHjiVu+0ZycnKpy2oDyd5jsrW1hbm5OSIjI8W2zMxMXLp0Ce7u7gAAd3d3pKen49q1a+I6P//8M4qKitC5c+dqr5mIiCqvRntMWVlZePDggfg9KSkJsbGxMDY2hrW1NebMmYPPPvsMDg4OsLW1xdKlS2FpaSm+X8rR0RH9+vXDlClTsH37dhQUFGDWrFkYNWpUqSPyiIhI2mo0mK5evYpevXqJ398MSBg/fjxCQ0OxaNEiZGdnY+rUqUhPT0fXrl0RHh4OHR0dcZsDBw5g1qxZ6NOnD9TU1DBs2DAEBwdX+7kQEZFqCPLaPixGBTIzMyGTyZCRkQFDQ8My192R/Hk1VUVU902zWVip7XNzc5GUlARbW1uF/2Al6VHmdyXZe0xERFQ/MZiIiEhSGExERCQpDCYiIpIUBhMREUmKZN/8QERUGWW915KDkaWNPSYionL4888/MXHiRFhaWkJLSwvNmjWDn58fnj59qtR+3kylLggCZDIZPDw88PPPP1dR1WULDQ0Va1FXV0fDhg3RuXNnBAYGIiMjo9j6qampmD17Npo3bw5tbW1YWVlh0KBBCm/oUQUGExHRO/z+++9wc3PD/fv3cejQITx48ADbt29HZGQk3N3d8ezZM6X2t3fvXqSkpCAmJgaNGjXCwIED8fvvv5e4bkFBgSpOoVSGhoZISUnBf//7X/zyyy+YOnUq9u/fj7Zt2ypM4ZGcnIwOHTrg559/xueff46bN28iPDwcvXr1wsyZM1VaE4OJiOgdZs6cCS0tLZw5cwY9evSAtbU1+vfvj7Nnz+Kvv/7CkiVLALyeymLNmjWYOHEiGjRoAGtra3z99dfF9mdkZARzc3M4Oztj27ZtePnyJSIiIgC87lFt27YNH3zwAfT19bF69WoAwA8//ID27dtDR0cHzZs3x8qVK/Hq1Stxn4IgYMeOHRg4cCD09PTg6OiIixcv4sGDB+jZsyf09fXRpUsXJCYmKtQiCALMzc1hYWEBR0dHTJo0Cb/88guysrKwaNEicb0ZM2ZAEARcvnwZw4YNQ4sWLdC6dWvMmzcPv/76q0p/3rzHpKTGqe9+nT0RlZNNTRfwbs+ePcPp06exevVq6OrqKiwzNzfH2LFjcfjwYXz11VcAgI0bN2LVqlX45JNPcPToUUyfPh09evRAy5YtS9z/m33m5+eLbStWrMC6deuwefNmaGho4D//+Q/GjRuH4OBgdOvWDYmJiZg6dSoAYPny5eJ2q1atwqZNm7Bp0yb4+/tjzJgxaN68OQICAmBtbY2JEydi1qxZOHXqVJnn3LhxY4wdOxZ79uxBYWEhMjIyEB4ejtWrV5c4M66RkdG7f5BKYI+JiKgM9+/fh1wuh6OjY4nLHR0d8fz5czx58gQAMGDAAMyYMQP29vbw9/dHo0aNcO7cuRK3zcnJwaeffgp1dXX06NFDbB8zZgwmTJiA5s2bw9raGitXrsTixYsxfvx4NG/eHH379sWqVauwY8cOhf1NmDABI0aMQIsWLeDv74/k5GSMHTsWXl5ecHR0hJ+fH6Kiosp13q1atcKLFy/w9OlTPHjwAHK5HK1atSrXtpXFHhMRUTmUdySfq6ur+O83l8n+OW/c6NGjoa6ujpcvX8LU1BS7d+9W2M7NzU1h/bi4OMTExIiX9QCgsLAQubm5yMnJgZ6eXrFjv5n928XFRaEtNzcXmZmZ73wv6JvzFQSh2kcxMpiIiMpgb28PQRAQHx+PIUOGFFseHx+Phg0bwtTUFACgqampsFwQBBQVFSm0ffHFF/D09IRMJhO3e9s/L5dlZWVh5cqVGDp0aLF1334h6tvHfjNcvqS2f9ZTkvj4eBgaGsLExATq6uoQBAF3795953aqwEt5RERlMDExQd++ffHVV1/h5cuXCstSU1Nx4MABjBw5ssznpv7J3Nwc9vb2JYZSSdq3b4+EhATY29sX+6ipqf7PeFpaGg4ePAhvb2+oqanB2NgYXl5e2Lp1K7Kzs4utn56ertLjM5iIiN4hJCQEeXl58PLywoULF/Dnn38iPDwcffv2RZMmTRQusVWFZcuWYf/+/Vi5ciVu376N+Ph4fPfdd/j0008rvW+5XI7U1FSkpKQgPj4ee/bsQZcuXSCTybBu3Tpxva1bt6KwsBCdOnXC999/j/v37yM+Ph7BwcHirOKqwmAiojpJLpeX+lGWg4MDrl69iubNm2PEiBGws7PD1KlT0atXL1y8eBHGxsZVcAb/x8vLCydPnsSZM2fQsWNHvPfee/jiiy/QrFmzSu87MzMTFhYWaNKkCdzd3bFjxw6MHz8eN27cgIWFhbhe8+bNcf36dfTq1Qvz58+Hs7Mz+vbti8jISGzbtq3SdbyNEwVCuYkCj/36uJqqIqr7hrxnVqntOVFg7cGJAomIqNZiMBERkaQwmIiISFIYTEREJCkMJiIikhQGExERSQqDiYiIJIXBREREksJgIiIiSeHbxYmoTtqR/Hmpy6bZLKzy4/fs2RNt27bF5s2bq/xYdQ17TERElRAVFQVBEFT+hu36jMFERFRLvD39el3GYCIieoe8vDz4+vqicePG0NHRQdeuXXHlyhUkJyejV69eAICGDRtCEAT4+PiI2xUVFWHRokUwNjaGubk5VqxYobDf9PR0TJ48GaampjA0NETv3r0RFxcnLl+xYgXatm2LXbt21asX1TKYiIjeYdGiRfj++++xb98+XL9+Hfb29vDy8kKDBg3w/fffAwASEhKQkpKCL7/8Utxu37590NfXx6VLl7BhwwYEBgYiIiJCXP7hhx8iLS0Np06dwrVr19C+fXv06dMHz549E9d58OABvv/+e4SFhSE2NrbazrkmcfADEVEZsrOzsW3bNoSGhqJ///4AgJ07dyIiIgJ79uxBx44dAQCNGzeGkZGRwraurq5Yvnw5gNdzOoWEhCAyMhJ9+/ZFdHQ0Ll++jLS0NGhrawMAgoKCcPz4cRw9ehRTp04F8Pry3f79+8s9221dwGBSkoPBk5ougagOqdx8TNUhMTERBQUF8PDwENs0NTXRqVMnxMfHi8FUEldXV4XvFhYWSEtLAwDExcUhKysLJiYmCuu8fPkSiYmJ4vdmzZrVq1ACGExERFVGU1NT4bsgCCgqKgIAZGVlwcLCAlFRUcW2e7vnpa+vX5UlShKDiYioDHZ2dtDS0kJMTIw4lXlBQQGuXLmCOXPmQEtLCwBQWFio1H7bt2+P1NRUaGhowMbGRtVl12oc/EBEVAZ9fX1Mnz4dCxcuRHh4OO7cuYMpU6YgJycHkyZNQrNmzSAIAk6ePIknT54gKyurXPv19PSEu7s7vL29cebMGSQnJ+OXX37BkiVLcPXq1So+K2ljj4mI6iRVvt1h3bp1KCoqwscff4wXL17Azc0Np0+fRsOGDdGwYUOsXLkSixcvxoQJEzBu3DiEhoa+c5+CIOCnn37CkiVLMGHCBDx58gTm5ubo3r07zMykf++tKglyuVxe00XUtMzMTMhkMmRkZMDQ0LDMdW/dulVNVRHVfc7OzpXaPjc3F0lJSfXqGZ/aSpnfFS/lERGRpDCYiIhIUhhMREQkKQwmIiKSFAYTERFJCoOJiIgkhcFERESSwmAiIiJJYTAREZGk8JVERFQnHfv1canLhrxXd1/5Exoaijlz5iA9Pb2mS6kw9piIiGqYjY0NNm/eXNNlSAaDiYioFigsLBTncqrrGExERO9QVFSEDRs2wN7eHtra2rC2tsbq1asBADdv3kTv3r2hq6sLExMTTJ06VWHqCx8fH3h7eyMoKAgWFhYwMTHBzJkzUVBQAADo2bMnHj58iLlz50IQBAiCAOD1JTkjIyOcOHECTk5O0NbWxh9//IHnz59j3LhxaNiwIfT09NC/f3/cv3+/+n8oVYjBRET0DgEBAVi3bh2WLl2KO3fu4ODBgzAzM0N2dja8vLzQsGFDXLlyBUeOHMHZs2cxa9Yshe3PnTuHxMREnDt3Dvv27UNoaKg4NUZYWBiaNm2KwMBApKSkICUlRdwuJycH69evx65du3D79m00btwYPj4+uHr1Kk6cOIGLFy9CLpdjwIABYtDVBRz8QERUhhcvXuDLL79ESEgIxo8fD+D1rLZdu3bFzp07kZubi/3794tToIeEhGDQoEFYv369OK9Sw4YNERISAnV1dbRq1Qrvv/8+IiMjMWXKFBgbG0NdXR0NGjSAubm5wrELCgrw1VdfoU2bNgCA+/fv48SJE4iJiUGXLl0AAAcOHICVlRWOHz+ODz/8sLp+LFWKPSYiojLEx8cjLy8Pffr0KXFZmzZtxFACAA8PDxQVFSEhIUFsa926NdTV1cXvFhYWSEtLe+extbS04OrqqnA8DQ0NdO7cWWwzMTFBy5YtER8fr/S5SRWDiYioDLq6upXeh6ampsJ3QRDKNZBBV1dXvOdUnzCYiIjK4ODgAF1dXURGRhZb5ujoiLi4OGRnZ4ttMTExUFNTQ8uWLct9DC0tLRQWFr5zPUdHR7x69QqXLl0S254+fYqEhAQ4OTmV+3hSx2AiIiqDjo4O/P39sWjRIuzfvx+JiYn49ddfsXv3bowdOxY6OjoYP348bt26hXPnzmH27Nn4+OOPxftL5WFjY4MLFy7gr7/+wt9//13qeg4ODhg8eDCmTJmC6OhoxMXF4aOPPkKTJk0wePBgVZyuJHDwAxHVSap8u8PSpUuhoaGBZcuW4dGjR7CwsMD//M//QE9PD6dPn4afnx86duwIPT09DBs2DJs2bVJq/4GBgZg2bRrs7OyQl5cHuVxe6rp79+6Fn58fBg4ciPz8fHTv3h0//fRTscuFtZkgL+snUE9kZmZCJpMhIyMDhoaGZa5769ataqqKqO5zdnau1Pa5ublISkqCra0tdHR0VFQVVQVlfle8lEdERJLCYCIiIklhMBERkaRw8IOSnFMP1HQJRHWH89qaroAkiD0mIiKSFAYTERFJCoOJiIgkhcFERESSwmAiIiJJYTARUZ1069atUj+qFhUVBUEQkJ6eXuo6K1asQNu2bZXed3JyMgRBQGxsbIXrq20YTEREZXgz3XlpnxUrVpRrPwsWLCjxDeXKehNUbz4mJib417/+hRs3bqhkv1IIQAYTEVEZ3kx3npKSgs2bN8PQ0FChbcGCBeXaj4GBAUxMTEpdnp+fr1RdZ8+eRUpKCk6fPo2srCz079+/zB5bWZQ9dlVjMBERlcHc3Fz8yGQyCIKg0GZgYCCue+3aNbi5uUFPTw9dunRRmMX2n5fyfHx84O3tjdWrV8PS0lKcv+ny5cto164ddHR04ObmVmpPyMTEBObm5nBzc0NQUBAeP36MS5cuITExEYMHD4aZmRkMDAzQsWNHnD17VmFbGxsbrFq1CuPGjYOhoSGmTp0KW1tbAEC7du0gCAJ69uyJCxcuQFNTE6mpqQrbz5kzB926davUz7UsDCYiIhVZsmQJNm7ciKtXr0JDQwMTJ04sc/3IyEgkJCQgIiICJ0+eRFZWFgYOHAgnJydcu3YNK1asKFeP7M0su/n5+cjKysKAAQMQGRmJGzduoF+/fhg0aBD++OMPhW2CgoLQpk0b3LhxA0uXLsXly5cB/F9PLCwsDN27d0fz5s3xzTffiNsVFBTgwIED7zy3yuAriYiIVGT16tXo0aMHAGDx4sV4//33kZubW+o0D/r6+ti1axe0tLQAAF9//TWKioqwe/du6OjooHXr1vjvf/+L6dOnl3rM9PR0rFq1CgYGBujUqRPMzMzQpk0bcfmqVatw7NgxnDhxArNmzRLbe/fujfnz54vf1dXVAfxfT+yNSZMmYe/evVi4cCEA4Mcff0Rubi5GjBih7I+n3NhjIiJSEVdXV/HfFhYWAIC0tLRS13dxcRFDCQDi4+Ph6uqqEGTu7u4lbtulSxcYGBigYcOGiIuLw+HDh2FmZoasrCwsWLAAjo6OMDIygoGBAeLj44v1mNzc3Mp1Tj4+Pnjw4AF+/fVXAEBoaChGjBgBfX39cm1fEewxERGpyNuzyAqCAAAoKioqdf3K/HE/fPgwnJycYGJiAiMjI7F9wYIFiIiIQFBQEOzt7aGrq4vhw4cXG+BQ3mM3btwYgwYNwt69e2Fra4tTp04hKiqqwnWXB4OJiEgiHB0d8c033yhc/nvTU/knKysr2NnZFWuPiYmBj48PhgwZAgDIyspCcnLyO4/9pudWWFhYbNnkyZMxevRoNG3aFHZ2dvDw8CjvKVWIpC/lFRYWYunSpbC1tYWuri7s7OywatUqvD0bvFwux7Jly2BhYQFdXV14enri/v37NVg1EVHFjBkzBoIgYMqUKbhz5w5++uknBAUFKbUPBwcHhIWFITY2FnFxcRgzZkyZvbY3GjduDF1dXYSHh+Px48fIyMgQl3l5ecHQ0BCfffYZJkyYoPR5KUvSwbR+/Xps27YNISEhiI+Px/r167FhwwZs2bJFXGfDhg0IDg7G9u3bcenSJejr68PLywu5ubk1WDkR1TRnZ+dSP1JlYGCAH3/8ETdv3kS7du2wZMkSrF+/Xql9bNq0CQ0bNkSXLl0waNAgeHl5oX379u/cTkNDA8HBwdixYwcsLS0xePBgcZmamhp8fHxQWFiIcePGKX1eyhLkb3c/JGbgwIEwMzPD7t27xbZhw4ZBV1cX3377LeRyOSwtLTF//nxxSGVGRgbMzMwQGhqKUaNGlbjfvLw85OXlid8zMzNhZWWFjIwMGBoall3U2YDKnxgRveZZuYkCc3NzkZSUBFtb21JHvpFqTJo0CU+ePMGJEycqtL0yvytJ95i6dOmCyMhI3Lt3DwAQFxeH6Oho9O/fHwCQlJSE1NRUeHp6itvIZDJ07twZFy9eLHW/a9euhUwmEz9WVlZVeyJERLVURkYGoqOjcfDgQcyePbtajinpwQ+LFy9GZmYmWrVqBXV1dRQWFmL16tUYO3YsAIhPI5uZmSlsZ2ZmVuxJ5bcFBARg3rx54vc3PSYiIlI0ePBgXL58Gf/zP/+Dvn37VssxJR1M//u//4sDBw7g4MGDaN26NWJjYzFnzhxYWlpi/PjxFd6vtrY2tLW1VVgpEVHdVNVDw0si6WBauHAhFi9eLN4rcnFxwcOHD7F27VqMHz9efDr58ePH4sNsb75X5PXyRERU8yR9jyknJwdqaoolqquri0MfbW1tYW5urvAq+czMTFy6dKnUp6WJiEjaJN1jGjRoEFavXg1ra2u0bt0aN27cwKZNm8SXBwqCgDlz5uCzzz6Dg4MDbG1tsXTpUlhaWsLb27tmiyciogqRdDBt2bIFS5cuxYwZM5CWlgZLS0tMmzYNy5YtE9dZtGgRsrOzMXXqVKSnp6Nr164IDw/n0FEiolpK0s8xVZfMzEzIZDI+x0RU3fgcU71RZ55jIiKi+kfSl/KIiCqsrKsbSvbUevbsibZt22Lz5s2Vq0liQkNDMWfOnApPyV5V2GMiInqHsLAwrFq1qlzrJicnQxAExMbGVm1RSrKxsSkWrCNHjhTfrCMl7DEREb2DsbFxTZdQIrlcjsLCQmhoVOxPua6urjgtu5Swx0RE9A49e/bEnDlzALzueaxZswYTJ05EgwYNYG1tja+//lpc19bWFgDQrl07CIKAnj17ist27doFR0dH6OjooFWrVvjqq68UjvPLL7+gbdu20NHRgZubG44fP67Q+4qKioIgCDh16hQ6dOgAbW1tREdHIzExEYMHD4aZmRkMDAzQsWNHnD17VqH+hw8fYu7cuRAEQZzEMDQ0VGGSQQDYtm0b7OzsoKWlhZYtW+Kbb75RWC4IAnbt2oUhQ4ZAT08PDg4OFX6xa2kYTEREStq4cSPc3Nxw48YNzJgxA9OnT0dCQgIA4PLlywCAs2fPIiUlBWFhYQCAAwcOYNmyZVi9ejXi4+OxZs0aLF26FPv27QPwenTwoEGD4OLiguvXr2PVqlXw9/cv8fiLFy/GunXrxKnYs7KyMGDAAERGRuLGjRvo168fBg0aJE6nHhYWhqZNmyIwMBApKSlISUkpcb/Hjh2Dn58f5s+fj1u3bmHatGmYMGECzp07p7DeypUrMWLECPz2228YMGAAxo4di2fPnlX+B/v/MZiIiJQ0YMAAzJgxA/b29vD390ejRo3EP96mpqYAABMTE5ibm4uXAZcvX46NGzdi6NChsLW1xdChQzF37lzs2LEDAHDw4EEIgoCdO3fCyckJ/fv3x8KFC0s8fmBgIPr27Qs7OzsYGxujTZs2mDZtGpydneHg4IBVq1bBzs5O7MkYGxtDXV0dDRo0gLm5ufg6t38KCgqCj48PZsyYgRYtWmDevHkYOnRosckKfXx8MHr0aNjb22PNmjXIysoSA1kVGExEREpydXUV/y0IAszNzZGWllbq+tnZ2UhMTMSkSZNgYGAgfj777DMkJiYCABISEuDq6qrwjE+nTp1K3J+bm5vC96ysLCxYsACOjo4wMjKCgYEB4uPjxR5TecXHxxebNt3DwwPx8fEKbW+fv76+PgwNDcs8f2Vx8AMRkZI0NTUVvguCUOb05VlZWQCAnTt3onPnzgrL1NXVlT6+vr6+wvcFCxYgIiICQUFBsLe3h66uLoYPH478/Hyl910eyp6/sthjIiJSIS0tLQBAYWGh2GZmZgZLS0v8/vvvsLe3V/i8GSzRsmVL3Lx5U2F27StXrpTrmDExMfDx8cGQIUPg4uICc3NzJCcnF6vr7ZpK4ujoiJiYmGL7dnJyKlcdqsJgIiJSocaNG0NXVxfh4eF4/PgxMjIyALweMLB27VoEBwfj3r17uHnzJvbu3YtNmzYBAMaMGYOioiJMnToV8fHxOH36tHhv580outI4ODggLCwMsbGxiIuLE/f1NhsbG1y4cAF//fUX/v777xL3s3DhQoSGhmLbtm24f/8+Nm3ahLCwMCxYsKCyPxal8FIeEdVNlXwPX0VpaGggODgYgYGBWLZsGbp164aoqChMnjwZenp6+Pzzz7Fw4ULo6+vDxcVFHIZuaGiIH3/8EdOnT0fbtm3h4uKCZcuWYcyYMe98t9ybWRe6dOmCRo0awd/fH5mZmQrrBAYGYtq0abCzs0NeXh5Kek2qt7c3vvzySwQFBcHPzw+2trbYu3evwpD36sCXuIIvcSWqMXyJa5kOHDiACRMmICMjQ5IPwipDmd8Ve0xERBKxf/9+NG/eHE2aNEFcXBz8/f0xYsSIWh9KymIwERFJRGpqKpYtW4bU1FRYWFjgww8/xOrVq2u6rGrHYCIikohFixZh0aJFNV1GjeOoPCIikhQGExERSQov5SnJ5a+TNV0CUZ1xEzUzpJukjT0mIiKSFAYTERFJCoOJiKgOK2kywLe9mXwwPT292mp6F95jIqI6yWWfS6nLbo6/WY2VACtWrMDx48fFmWilpEuXLkhJSYFMJqvpUkQMJiKiOqqgoOCd62hpaZU6cWBN4aU8IqJ36NmzJ3x9fbFo0SIYGxvD3NwcK1asEJenp6dj8uTJMDU1haGhIXr37o24uDgAry+lrVy5EnFxcRAEAYIgIDQ0FAsWLMDAgQPFfWzevBmCICA8PFxss7e3x65duwAARUVFCAwMRNOmTaGtrY22bdsqrJucnAxBEHD48GH06NEDOjo6OHDgQLFzefLkCdzc3DBkyBDk5eUVu5T35tLf6dOn4ejoCAMDA/Tr109hOvZXr17B19cXRkZGMDExgb+/P8aPHw9vb29V/LgZTERE5bFv3z7o6+vj0qVL2LBhAwIDAxEREQEA+PDDD5GWloZTp07h2rVraN++Pfr06YNnz55h5MiRmD9/Plq3bo2UlBSkpKRg5MiR6NGjB6Kjo8U5ks6fP49GjRohKioKAPDXX38hMTFRfLP3l19+iY0bNyIoKAi//fYbvLy88MEHH+D+/fsKdS5evBh+fn6Ij4+Hl5eXwrI///wT3bp1g7OzM44ePQptbe0SzzUnJwdBQUH45ptvcOHCBfzxxx8KU1+sX78eBw4cwN69exETE4PMzEwcP35cBT/l1xhMRETl4OrqiuXLl8PBwQHjxo2Dm5sbIiMjER0djcuXL+PIkSNwc3ODg4MDgoKCYGRkhKNHj0JXVxcGBgbQ0NCAubk5zM3Noauri27duuHFixe4ceMG5HI5Lly4gPnz54vBFBUVhSZNmsDe3h4AEBQUBH9/f4waNQotW7bE+vXr0bZtW2zevFmhzjlz5mDo0KGwtbWFhYWF2J6QkAAPDw94eXlh7969Zc6cW1BQgO3bt8PNzQ3t27fHrFmzEBkZKS7fsmULAgICMGTIELRq1QohISFlDrBQFu8xERGVg6urq8J3CwsLpKWlIS4uDllZWTAxMVFY/vLlSyQmJpa6PyMjI7Rp0wZRUVHQ0tKClpYWpk6diuXLlyMrKwvnz59Hjx49ALyemufRo0fw8PBQ2IeHh4d4yfANNze3Ysd6+fIlunXrhjFjxhQLspLo6enBzs6u2LkCQEZGBh4/foxOnTqJy9XV1dGhQweVTa/OYCIiKgdNTU2F74IgoKioCFlZWbCwsBB7Om97Vy+iZ8+eiIqKgra2Nnr06AFjY2M4OjoiOjoa58+fx/z585WuU19fv1ibtrY2PD09cfLkSSxcuBBNmjQpcx8lnWt1Tt3HS3lERJXQvn17pKamQkNDA/b29gqfRo0aAXg98u3NvaS3vbnPFBkZKd5L6tmzJw4dOoR79+6JbYaGhrC0tERMTIzC9jExMXBycnpnjWpqavjmm2/QoUMH9OrVC48eParw+cpkMpiZmeHKlStiW2FhIa5fv17hff4Tg4mIqBI8PT3h7u4Ob29vnDlzBsnJyfjll1+wZMkSXL16FQBgY2ODpKQkxMbG4u+//0ZeXh4AoHv37njx4gVOnjypEEwHDhyAhYUFWrRoIR5n4cKFWL9+PQ4fPoyEhAQsXrwYsbGx8PPzK1ed6urqOHDgANq0aYPevXsjNTW1wuc8e/ZsrF27Fj/88AMSEhLg5+eH58+fQxCECu/zbQwmIqJKEAQBP/30E7p3744JEyagRYsWGDVqFB4+fAgzMzMAwLBhw9CvXz/06tULpqamOHToEACgYcOGcHFxgampKVq1agXgdVgVFRWJ95fe8PX1xbx58zB//ny4uLggPDwcJ06cgIODQ7lr1dDQwKFDh9C6dWv07t1bvG+kLH9/f4wePRrjxo2Du7s7DAwM4OXlpbLp7QV5dV44lKjMzEzIZDJkZGTA0NCwzHXLepqciJRT2Tcw5ObmIikpCba2tir7o0jKKyoqgqOjI0aMGIFVq1aVuI4yvysOfiAiIqU8fPgQZ86cQY8ePZCXl4eQkBAkJSVhzJgxKtk/L+UREZFS1NTUEBoaio4dO8LDwwM3b97E2bNn4ejoqJL9s8dERERKsbKyKjZCUJXYYyIiIklhMBERkaQwmIiISFIYTEREJCkMJiIikhQGExERSYpSw8Xj4+Px3Xff4T//+Q8ePnyInJwcmJqaol27dvDy8sKwYcNKnXiKiIioPMoVTNevX8eiRYsQHR0NDw8PdO7cGUOGDIGuri6ePXuGW7duYcmSJZg9ezYWLVqEOXPmMKCIqEatXLmy1GXLly9Xal89e/YscVK+0NBQzJkzR5yWnFSjXME0bNgwLFy4EEePHi1zfpGLFy+K0/9+8sknqqqRiKjeKSgoKDYvUn1RrntM9+7dw4wZM9456ZW7uzu+++47LFy4UBW1ERHVGlFRUejUqRP09fVhZGQEDw8PPHz4UFz+ww8/oH379tDR0UHz5s2xcuVKvHr1SlwuCAK2bduGDz74APr6+li9enVNnIYklKvHpGxq19eUJ6L66dWrV/D29saUKVNw6NAh5Ofn4/Lly+L8RP/5z38wbtw4BAcHo1u3bkhMTMTUqVMBKF5WXLFiBdatW4fNmzdDQ6P+vjGuwqPyUlJSMHz4cJiamsLY2BiDBg3C77//rsraiIhqhczMTGRkZGDgwIGws7ODo6Mjxo8fD2trawCv73ctXrwY48ePR/PmzdG3b1+sWrUKO3bsUNjPmDFjMGHCBDRv3lzctj6qcDBNnDgRzs7OOH/+PH7++WeYmZmp7JXnRES1ibGxMXx8fODl5YVBgwbhyy+/REpKirg8Li4OgYGBMDAwED9TpkxBSkoKcnJyxPXc3NxqonzJKXcw+fn5ITs7W/z+4MED+Pv7w8nJCW3btoWfnx8SEhKqpEgioppkaGiIjIyMYu3p6emQyWQAgL179+LixYvo0qULDh8+jBYtWuDXX38FAGRlZWHlypWIjY0VPzdv3sT9+/cVJs3T19evnhOSuHJfxGzatCk6dOiADRs24IMPPsDIkSPRuXNnDBgwAAUFBQgLC8PYsWOrslYiohrRsmVLnDlzplj79evX0aJFC/F7u3bt0K5dOwQEBMDd3R0HDx7Ee++9h/bt2yMhIQH29vbVWXatVe5gWrhwIYYPH44ZM2YgNDQUW7ZsQefOnREVFYXCwkJs2LABw4cPr8paiYhqxPTp0xESEgJfX19MnjwZ2tra+Pe//41Dhw7hxx9/RFJSEr7++mt88MEHsLS0REJCAu7fv49x48YBAJYtW4aBAwfC2toaw4cPh5qaGuLi4nDr1i189tlnNXx20qPUsA9bW1ucOnUKBw4cQI8ePeDn54egoCBx5AkRUV3UvHlzXLhwAUuWLIGnpyfy8/PRqlUrHDlyBP369cPjx49x9+5d7Nu3D0+fPoWFhQVmzpyJadOmAQC8vLxw8uRJBAYGYv369dDU1ESrVq0wefLkGj4zaRLkcrlcmQ2ePn0KExMTPH/+HPPnz8edO3fw9ddfw9XVtapqrHKZmZmQyWTIyMiAoaFhmeu67HOppqqI6r6b429Wavvc3FwkJSXB1tZW4V4NSY8yv6tyD36IjIyEmZkZTE1N0bRpU9y9exd79uzB2rVrMXr0aCxatAgvX76sdPFERFS/lTuYZs6ciUWLFiEnJwchISGYM2cOAKBXr164fv06NDU10bZt2yoqk4iI6otyB1NKSgref/996OjooF+/fnjy5Im4TFtbG6tXr0ZYWFiVFElERPVHuQc/fPDBBxg+fDg++OADREdHY8CAAcXWad26tUqLIyKi+qfcPabdu3dj2rRpyMjIwEcffVTs9e9ERESqUO4ek5aWFmbPnl2VtRAREZWvx/TmtRrlkZOTg9u3b1e4ICIiqt/KFUwff/wxvLy8cOTIEYX35b3tzp07+OSTT2BnZ4dr166ptEgiIqo/ynUp786dO9i2bRs+/fRTjBkzBi1atIClpSV0dHTw/Plz3L17F1lZWRgyZAjOnDkDF5e6+xDq8GS+domIqCqVe6JAX19f+Pr64urVq4iOjsbDhw/x8uVLtGnTBnPnzkWvXr1gbGxc1fUSEVEdp/QUiW5ubpwzhIgkz2bxv0tdlrzufZUcw8fHB+np6Th+/LhK9kevVXiiQCIiUo38/PyaLkFSGExERO9w9OhRuLi4QFdXFyYmJvD09MTChQuxb98+/PDDDxAEAYIgICoqCgBw8+ZN9O7dW1x/6tSpyMrKEvfn4+MDb29vrF69GpaWlmjZsiUA4JtvvoGbmxsaNGgAc3NzjBkzBmlpaQq1nDhxAg4ODtDR0UGvXr2wb98+CIKA9PR0cZ3o6Gh069YNurq6sLKygq+vb6kD16SIwUREVIaUlBSMHj0aEydORHx8PKKiojB06FAsX74cI0aMQL9+/ZCSkoKUlBR06dIF2dnZ8PLyQsOGDXHlyhUcOXIEZ8+exaxZsxT2GxkZiYSEBERERODkyZMAgIKCAqxatQpxcXE4fvw4kpOT4ePjI26TlJSE4cOHw9vbG3FxcZg2bRqWLFmisN/ExET069cPw4YNw2+//YbDhw8jOjq62PGlTOl7TERE9UlKSgpevXqFoUOHolmzZgAgjjzW1dVFXl4ezM3NxfX37duH3Nxc7N+/X5wqPSQkBIMGDcL69ethZmYG4PU06rt27YKWlpa47cSJE8V/N2/eHMHBwejYsSOysrJgYGCAHTt2oGXLlvj8888BvJ5Z99atW1i9erW43dq1azF27FjxRdsODg4IDg5Gjx49sG3btloxPYhKekxvdyGJiOqSNm3aoE+fPnBxccGHH36InTt34vnz56WuHx8fjzZt2oihBAAeHh4oKipCQkKC2Obi4qIQSgBw7do1DBo0CNbW1mjQoAF69OgBAPjjjz8AAAkJCejYsaPCNp06dVL4HhcXh9DQUBgYGIgfLy8vFBUVISkpqWI/hGqmdDCtX78ehw8fFr+PGDECJiYmaNKkCeLi4lRaHBFRTVNXV0dERAROnToFJycnbNmyBS1btqz0H/m3gwuAeAnQ0NAQBw4cwJUrV3Ds2DEAyg2OyMrKwrRp0xAbGyt+4uLicP/+fdjZ2VWq5uqidDBt374dVlZWAICIiAjxF9a/f38sXLhQ5QUSEdU0QRDg4eGBlStX4saNG9DS0sKxY8egpaWFwsJChXUdHR0RFxenMNggJiYGampq4iCHkty9exdPnz7FunXr0K1bN7Rq1arYwIeWLVvi6tWrCm1XrlxR+N6+fXvcuXMH9vb2xT7/7KFJldLBlJqaKgbTyZMnMWLECPzrX//CokWLiv2AiIhqu0uXLmHNmjW4evUq/vjjD4SFheHJkydwdHSEjY0NfvvtNyQkJODvv/9GQUEBxo4dCx0dHYwfPx63bt3CuXPnMHv2bHz88cfi/aWSWFtbQ0tLC1u2bMHvv/+OEydOYNWqVQrrTJs2DXfv3oW/vz/u3buH//3f/0VoaCiA1+EJAP7+/vjll18wa9YsxMbG4v79+/jhhx9q1eAHpYOpYcOG+PPPPwEA4eHh8PT0BADI5fJi/+VARFTbGRoa4sKFCxgwYABatGiBTz/9FBs3bkT//v0xZcoUtGzZEm5ubjA1NUVMTAz09PRw+vRpPHv2DB07dsTw4cPRp08fhISElHkcU1NThIaG4siRI3BycsK6desQFBSksI6trS2OHj2KsLAwuLq6Ytu2beKoPG1tbQCAq6srzp8/j3v37qFbt25o164dli1bBktLy6r5AVUBQS6Xy5XZYNasWTh58iQcHBxw48YNJCcnw8DAAN999x02bNiA69evV1WtVSYzMxMymQwZGRkwNDQsc92VK1dWU1VEdd/y5csrtX1ubi6SkpJga2tbK0abVYXVq1dj+/btYodBqpT5XSk9XPyLL76AjY0N/vzzT2zYsAEGBgYAXg+pnDFjRsUqJiKicvnqq6/QsWNHmJiYICYmBp9//nmtukxXHkoHk6amJhYsWFCsfe7cuSopiIiISnf//n189tlnePbsGaytrTF//nwEBATUdFkqVaHnmL755ht07doVlpaWePjwIQBg8+bN+OGHH1RaHAD89ddf+Oijj2BiYgJdXV24uLgojEqRy+VYtmwZLCwsoKurC09PT9y/f1/ldRARScEXX3yBR48eITc3F/fu3cPSpUuhoVG33pWgdDBt27YN8+bNQ//+/ZGeni4OeDAyMsLmzZtVWtzz58/h4eEBTU1NnDp1Cnfu3MHGjRvRsGFDcZ0NGzYgODgY27dvx6VLl6Cvrw8vLy/k5uaqtBYiIqoeSsfsli1bsHPnTnh7e2PdunViu5ubW4mX+Cpj/fr1sLKywt69e8U2W1tb8d9yuRybN2/Gp59+isGDBwMA9u/fDzMzMxw/fhyjRo0qcb95eXnIy8sTv2dmZqq0biIiqjile0xJSUlo165dsXZtbW2Vv732xIkTcHNzw4cffojGjRujXbt22Llzp0Itqamp4pB1AJDJZOjcuTMuXrxY6n7Xrl0LmUwmft48l0VERDVP6WCytbVFbGxssfbw8HA4OjqqoibR77//jm3btsHBwQGnT5/G9OnT4evri3379gF4/bAvgGIPrZmZmYnLShIQEICMjAzxI/VhlkRE9YnSl/LmzZuHmTNnIjc3F3K5HJcvX8ahQ4ewdu1a7Nq1S6XFFRUVwc3NDWvWrAEAtGvXDrdu3cL27dsxfvz4Cu9XW1tbfBiNiIikRelgmjx5MnR1dfHpp58iJycHY8aMgaWlJb788stS7+lUlIWFBZycnBTaHB0d8f333wOA+Kr5x48fw8LCQlzn8ePHaNu2rUprISKi6lGh4eJjx47F/fv3kZWVhdTUVPz3v//FpEmTVF0bPDw8FF4TDwD37t0T50SxtbWFubk5IiMjxeWZmZm4dOkS3N3dVV4PERFVvUoNftfT04Oenp6qailm7ty56NKlC9asWYMRI0bg8uXL+Prrr/H1118DeP3Swjlz5uCzzz6Dg4MDbG1tsXTpUlhaWsLb27vK6iKiWmCFrIxlGdVXRw1asWIFjh8/XuK4ACkrVzC1a9dOfHPtu6jyXXkdO3bEsWPHEBAQgMDAQNja2mLz5s0YO3asuM6iRYuQnZ2NqVOnIj09HV27dkV4eHi9fW8WEVFtV65Led7e3hg8eDAGDx4MLy8vJCYmQltbGz179kTPnj2ho6ODxMREeHl5qbzAgQMH4ubNm8jNzUV8fDymTJmisFwQBAQGBiI1NRW5ubk4e/YsWrRoofI6iKj+KioqwoYNG2Bvbw9tbW1YW1uL05nfvHkTvXv3hq6uLkxMTDB16lRkZWWJ2/r4+MDb2xtr1qyBmZkZjIyMEBgYiFevXmHhwoUwNjZG06ZNFZ7XTE5OhiAI+O6779ClSxfo6OjA2dkZ58+fF9cJDQ2FkZGRQp3Hjx8XOxGhoaFYuXIl4uLiIAgCBEEQp8hIT0/H5MmTYWpqCkNDQ/Tu3VtSE72Wq8f09huAJ0+eDF9f32LzhCxfvpzDromoTgoICMDOnTvxxRdfoGvXrkhJScHdu3fFWWfd3d1x5coVpKWlYfLkyZg1a5YYAgDw888/o2nTprhw4QJiYmIwadIk/PLLL+jevTsuXbqEw4cPY9q0aejbty+aNm0qbrdw4UJs3rwZTk5O2LRpEwYNGoSkpCSYmJi8s+aRI0fi1q1bCA8Px9mzZwG8fs4TAD788EPo6uri1KlTkMlk2LFjB/r06YN79+7B2NhYtT+8ClB68MORI0cwbty4Yu0fffSROFqOiKiuePHiBb788kts2LAB48ePh52dHbp27YrJkyfj4MGDyM3Nxf79++Hs7IzevXsjJCQE33zzDR4/fizuw9jYGMHBwWjZsiUmTpyIli1bIicnB5988gkcHBwQEBAALS0tREdHKxx71qxZGDZsGBwdHbFt2zbIZDLs3r27XHXr6urCwMAAGhoaMDc3h7m5OXR1dREdHY3Lly/jyJEjcHNzg4ODA4KCgmBkZISjR4+q9GdXUUoPftDV1UVMTAwcHBwU2mNiYnhfh4jqnPj4eOTl5aFPnz4lLmvTpg309fXFNg8PDxQVFSEhIUF8+L9169ZQU/u/foCZmRmcnZ3F7+rq6jAxMSk2lfrbo4s1NDTg5uaG+Pj4Sp1PXFwcsrKyivW6Xr58icTExErtW1WUDqY5c+Zg+vTpuH79Ojp16gTg9dTDe/bswdKlS1VeIBFRTdLV1a30PjQ1NRW+C4JQYltRUVG596mmpoZ/zvNaUFDwzu2ysrJgYWGBqKioYsv+ec+qpigdTIsXL0bz5s3x5Zdf4ttvvwXw+qHXvXv3YsSIESovkIioJjk4OEBXVxeRkZGYPHmywjJHR0eEhoYiOztb7DXFxMRATU0NLVu2rPSxf/31V3Tv3h0A8OrVK1y7dk2cFNDU1BQvXrxQOPY/h4VraWmJM0C80b59e6SmpkJDQwM2NjaVrrEqVOg5phEjRjCEiKhe0NHRgb+/PxYtWgQtLS14eHjgyZMnuH37NsaOHYvly5dj/PjxWLFiBZ48eYLZs2fj448/LvYOz4rYunUrHBwc4OjoiC+++ALPnz/HxIkTAQCdO3eGnp4ePvnkE/j6+uLSpUsKAy4AwMbGBklJSYiNjUXTpk3RoEEDeHp6wt3dHd7e3tiwYQNatGiBR48e4d///jeGDBkCNze3StddWRV68wMAXLt2Dd9++y2+/fZb3LhxQ5U1ERFJytKlSzF//nwsW7YMjo6OGDlyJNLS0qCnp4fTp0/j2bNn6NixI4YPH44+ffogJCREJcddt24d1q1bhzZt2iA6OhonTpxAo0aNALweUPHtt9/ip59+gouLCw4dOoQVK1YobD9s2DD069cPvXr1gqmpKQ4dOgRBEPDTTz+he/fumDBhAlq0aIFRo0bh4cOHKglTVRDk/7xI+Q5paWkYNWoUoqKixOuR6enp6NWrF7777juYmppWRZ1VKjMzEzKZDBkZGTA0NCxz3ZUrV1ZTVUR139uPolREbm4ukpKSYGtrW6cGXyUnJ8PW1hY3btyoM+/9VOZ3pXSPafbs2Xjx4gVu376NZ8+e4dmzZ7h16xYyMzPh6+tb4aKJiIiACtxjevOw1ttzLzk5OWHr1q3417/+pdLiiIio/lE6mIqKiooNcwReD4dUZqgjERGVzMbGpthQ8PpE6Ut5vXv3hp+fHx49eiS2/fXXX5g7d26JD6AREREpQ+lgCgkJQWZmJmxsbGBnZwc7OzvY2toiMzMTW7ZsqYoaiYjKVJ97F7WFMr8jpS/lWVlZ4fr16zh79izu3r0L4PVDZp6ensruioioUt7cVsjJyVHJGxqo6uTk5AAo/haMklToAVtBENC3b1/07du3IpsTEamEuro6jIyMxHfM6enplXvuOKoecrkcOTk5SEtLg5GREdTV1d+5TYWCKTIyEpGRkUhLSys24GHPnj0V2SURUYWYm5sDQLEXoJK0GBkZib+rd1E6mFauXInAwEC4ubnBwsKC/3VCRDVKEARYWFigcePG5XqJKVU/TU3NcvWU3lA6mLZv347Q0FB8/PHHym5KRFRl1NXVlfrjR9Kl9Ki8/Px8dOnSpSpqISIiUj6Y3szaSEREVBWUvpSXm5uLr7/+GmfPnoWrq2uxoX+bNm1SWXFERFT/KB1Mv/32m/i221u3biks40AIIiKqLKWD6dy5c1VRBxEREYBKTBRIRERUFRhMREQkKQwmIiKSFAYTERFJSrmCqX379nj+/DkAIDAwUHxLLBERkaqVK5ji4+ORnZ0N4PW78rKysqq0KCIiqr/KNVy8bdu2mDBhArp27Qq5XI6goCAYGBiUuO6yZctUWiAREdUv5Qqm0NBQLF++HCdPnoQgCDh16hQ0NIpvKggCg4mIiCpFkCs5J7GamhpSU1PRuHHjqqqp2mVmZkImkyEjIwOGhoZlrmuz+N/VVBVR3Ze87v2aLoEkSOk3P/xzYkAiIiJVqtAMtomJidi8eTPi4+MBAE5OTvDz84OdnZ1KiyMiovpH6eeYTp8+DScnJ1y+fBmurq5wdXXFpUuX0Lp1a0RERFRFjUREVI8o3WNavHgx5s6di3Xr1hVr9/f3R9++fVVWHBER1T9K95ji4+MxadKkYu0TJ07EnTt3VFIUERHVX0oHk6mpKWJjY4u1x8bG1qmRekREVDOUvpQ3ZcoUTJ06Fb///ju6dOkCAIiJicH69esxb948lRdIRET1i9LBtHTpUjRo0AAbN25EQEAAAMDS0hIrVqyAr6+vygskIqL6RekHbN/24sULAECDBg1UVlBN4AO2RDWDD9hSSSr0HNMbtT2QiIhIejgfExERSQqDiYiIJIXBREREkqJUMBUUFKBPnz64f/9+VdVDRET1nFLBpKmpid9++62qaiEiIlL+Ut5HH32E3bt3V0UtREREyg8Xf/XqFfbs2YOzZ8+iQ4cO0NfXV1i+adMmlRVHRET1j9LBdOvWLbRv3x4AcO/ePYVlgiCopioiIqq3lA6mc+fOVUUdREREACoxXPzBgwc4ffo0Xr58CQCoxJuNiIiIREoH09OnT9GnTx+0aNECAwYMQEpKCgBg0qRJmD9/vsoLJCKi+kXpYJo7dy40NTXxxx9/QE9PT2wfOXIkwsPDVVocERHVP0rfYzpz5gxOnz6Npk2bKrQ7ODjg4cOHKiuMiIjqJ6V7TNnZ2Qo9pTeePXsGbW1tlRRFRET1l9LB1K1bN+zfv1/8LggCioqKsGHDBvTq1UulxRERUf2j9KW8DRs2oE+fPrh69Sry8/OxaNEi3L59G8+ePUNMTExV1EhERPWI0j0mZ2dn3Lt3D127dsXgwYORnZ2NoUOH4saNG7Czs6uKGomIqB6p0Ay2MpkMS5YsUXUtREREFQum58+fY/fu3YiPjwcAODk5YcKECTA2NlZpcUREVP8ofSnvwoULsLGxQXBwMJ4/f47nz58jODgYtra2uHDhQlXUSERE9YjSPaaZM2di5MiR2LZtG9TV1QEAhYWFmDFjBmbOnImbN2+qvEgiIqo/lO4xPXjwAPPnzxdDCQDU1dUxb948PHjwQKXFERFR/aN0MLVv3168t/S2+Ph4tGnTRiVFERFR/VWuS3lvT6fu6+sLPz8/PHjwAO+99x4A4Ndff8XWrVuxbt26qqmSiIjqDUFejvkq1NTUIAjCO6e2EAQBhYWFKiuuumRmZkImkyEjIwOGhoZlrmuz+N/VVBVR3Ze87v2aLoEkqFw9pqSkpKqug4iICEA5g6lZs2ZVXQcRERGACj5g++jRI0RHRyMtLQ1FRUUKy3x9fVVSGBER1U9KB1NoaCimTZsGLS0tmJiYQBAEcZkgCAwmIiKqFKWDaenSpVi2bBkCAgKgpqb0aHMiIqIyKZ0sOTk5GDVqFEOJiIiqhNLpMmnSJBw5cqQqaiEiIlL+Ut7atWsxcOBAhIeHw8XFBZqamgrLN23apLLiiIio/qlQMJ0+fRotW7YEgGKDH4iIiCpD6WDauHEj9uzZAx8fnyooh4iI6jul7zFpa2vDw8OjKmp5p3Xr1kEQBMyZM0dsy83NxcyZM2FiYgIDAwMMGzYMjx8/rpH6iIio8pQOJj8/P2zZsqUqainTlStXsGPHDri6uiq0z507Fz/++COOHDmC8+fP49GjRxg6dGi110dERKqh9KW8y5cv4+eff8bJkyfRunXrYoMfwsLCVFbcG1lZWRg7dix27tyJzz77TGzPyMjA7t27cfDgQfTu3RsAsHfvXjg6OuLXX38V337+T3l5ecjLyxO/Z2ZmqrxmIiKqGKV7TEZGRhg6dCh69OiBRo0aQSaTKXyqwsyZM/H+++/D09NTof3atWsoKChQaG/VqhWsra1x8eLFUve3du1ahZqtrKyqpG4iIlKe0j2mvXv3VkUdpfruu+9w/fp1XLlypdiy1NRUaGlpwcjISKHdzMwMqamppe4zICAA8+bNE79nZmYynIiIJKJCL3GtLn/++Sf8/PwQEREBHR0dle1XW1sb2traKtsfERGpjtLBZGtrW+bzSr///nulCnrbtWvXkJaWhvbt24tthYWFuHDhAkJCQnD69Gnk5+cjPT1dodf0+PFjmJubq6wOIiKqPkoH09tDtQGgoKAAN27cQHh4OBYuXKiqugAAffr0wc2bNxXaJkyYgFatWsHf3x9WVlbQ1NREZGQkhg0bBgBISEjAH3/8AXd3d5XWQkRE1UPpYPLz8yuxfevWrbh69WqlC3pbgwYN4OzsrNCmr68PExMTsX3SpEmYN28ejI2NYWhoiNmzZ8Pd3b3UEXlERCRtKntFeP/+/fH999+ranfl9sUXX2DgwIEYNmwYunfvDnNz8yoZsk5ERNVDZYMfjh49CmNjY1XtrlRRUVEK33V0dLB161Zs3bq1yo9NRERVT+lgateuncLgB7lcjtTUVDx58gRfffWVSosjIqL6R+lg8vb2VviupqYGU1NT9OzZE61atVJVXUREVE8pHUzLly+vijqIiIgAqHDwAxERkSqUu8ekpqb2zokABUHAq1evKl0UERHVX+UOpmPHjpW67OLFiwgODkZRUZFKiiIiovqr3ME0ePDgYm0JCQlYvHgxfvzxR4wdOxaBgYEqLY6IiOqfCt1jevToEaZMmQIXFxe8evUKsbGx2LdvH5o1a6bq+oiIqJ5RKpgyMjLg7+8Pe3t73L59G5GRkfjxxx+LvTaIiIioosp9KW/Dhg1Yv349zM3NcejQoRIv7REREVWWIJfL5eVZUU1NDbq6uvD09IS6unqp69XG99RlZmZCJpMhIyMDhoaGZa5rs/jf1VQVUd2XvO79mi6BJKjcPaZx48a9c7g4ERFRZZU7mEJDQ6uwDCIiotf45gciIpIUBhMREUkKg4mIiCSFwURERJLCYCIiIklhMBERkaQwmIiISFIYTEREJCkMJiIikhQGExERSQqDiYiIJIXBREREksJgIiIiSWEwERGRpDCYiIhIUhhMREQkKQwmIiKSFAYTERFJCoOJiIgkhcFERESSolHTBdQ2yTpjaroEojoko6YLIAlij4mIiCSFwURERJLCYCIiIklhMBERkaQwmIiISFIYTEREJCkMJiIikhQGExERSQqDiYiIJIXBREREksJgIiIiSWEwERGRpDCYiIhIUhhMREQkKQwmIiKSFAYTERFJCoOJiIgkhcFERESSwmAiIiJJYTAREZGkMJiIiEhSGExERCQpDCYiIpIUBhMREUkKg4mIiCSFwURERJLCYCIiIklhMBERkaQwmIiISFIYTEREJCkMJiIikhQGExERSQqDiYiIJIXBREREksJgIiIiSWEwERGRpDCYiIhIUhhMREQkKQwmIiKSFAYTERFJCoOJiIgkhcFERESSwmAiIiJJkXQwrV27Fh07dkSDBg3QuHFjeHt7IyEhQWGd3NxczJw5EyYmJjAwMMCwYcPw+PHjGqqYiIgqS9LBdP78ecycORO//vorIiIiUFBQgH/961/Izs4W15k7dy5+/PFHHDlyBOfPn8ejR48wdOjQGqyaiIgqQ5DL5fKaLqK8njx5gsaNG+P8+fPo3r07MjIyYGpqioMHD2L48OEAgLt378LR0REXL17Ee++9V679ZmZmQiaTISMjA4aGhmWvvEJW2dMgojdWZNR0BSRBku4x/VNGxuv/ERsbGwMArl27hoKCAnh6eorrtGrVCtbW1rh48WKp+8nLy0NmZqbCh4iIpKHWBFNRURHmzJkDDw8PODs7AwBSU1OhpaUFIyMjhXXNzMyQmppa6r7Wrl0LmUwmfqysrKqydCIiUkKtCaaZM2fi1q1b+O677yq9r4CAAGRkZIifP//8UwUVEhGRKmjUdAHlMWvWLJw8eRIXLlxA06ZNxXZzc3Pk5+cjPT1dodf0+PFjmJubl7o/bW1taGtrV2XJRERUQZLuMcnlcsyaNQvHjh3Dzz//DFtbW4XlHTp0gKamJiIjI8W2hIQE/PHHH3B3d6/ucomISAUk3WOaOXMmDh48iB9++AENGjQQ7xvJZDLo6upCJpNh0qRJmDdvHoyNjWFoaIjZs2fD3d293CPyiIhIWiQdTNu2bQMA9OzZU6F979698PHxAQB88cUXUFNTw7Bhw5CXlwcvLy989dVX1VwpERGpSq16jqmq8DkmohrC55ioBJK+x0RERPUPg4mIiCSFwURERJLCYCIiIklhMBERkaQwmIiISFIYTEREJCkMJiIikhQGExERSQqDiYiIJIXBREREksJgIiIiSWEwERGRpDCYiIhIUhhMREQkKQwmIiKSFAYTERFJCoOJiIgkhcFERESSwmAiIiJJYTAREZGkMJiIiEhSGExERCQpDCYiIpIUBhMREUkKg4mIiCSFwURERJLCYCIiIklhMBERkaQwmIiISFIYTEREJCkMJiIikhQGExERSQqDiYiIJIXBREREksJgIiIiSWEwERGRpDCYiIhIUhhMREQkKQwmIiKSFAYTERFJCoOJiIgkhcFERESSwmAiIiJJYTAREZGkMJiIiEhSGExERCQpDCYiIpIUBhMREUkKg4mIiCSFwURERJLCYCIiIklhMBERkaQwmIiISFIYTEREJCkMJiIikhQGExERSQqDiYiIJIXBREREksJgIiIiSWEwERGRpDCYiIhIUhhMREQkKQwmIiKSFAYTERFJCoOJiIgkhcFERESSwmAiIiJJYTAREZGkMJiIiEhSGExERCQpDCYiIpIUBhMREUkKg4mIiCSFwURERJLCYCIiIklhMBERkaTUmWDaunUrbGxsoKOjg86dO+Py5cs1XRIREVVAnQimw4cPY968eVi+fDmuX7+ONm3awMvLC2lpaTVdGhERKalOBNOmTZswZcoUTJgwAU5OTti+fTv09PSwZ8+emi6NiIiUpFHTBVRWfn4+rl27hoCAALFNTU0Nnp6euHjxYonb5OXlIS8vT/yekZEBAMjMzHz3AfPklSuYiP5Pef4/9/81aNAAgiBUYTEkFbU+mP7++28UFhbCzMxMod3MzAx3794tcZu1a9di5cqVxdqtrKyqpEYiKsU6WblXzcjIgKGhYRUWQ1JR64OpIgICAjBv3jzxe1FREZ49ewYTExP+FxmRRDVo0KCmS6BqUuuDqVGjRlBXV8fjx48V2h8/fgxzc/MSt9HW1oa2trZCm5GRUVWVSERESqj1gx+0tLTQoUMHREZGim1FRUWIjIyEu7t7DVZGREQVUet7TAAwb948jB8/Hm5ubujUqRM2b96M7OxsTJgwoaZLIyIiJdWJYBo5ciSePHmCZcuWITU1FW3btkV4eHixARFERCR9glwu5/hnIiKSjFp/j4mIiOoWBhMREUkKg4mIiCSFwURERJLCYCIiIklhMBERkaQwmIiISFIYTEREJCkMJiIikhQGExERSQqDiYiIJOX/ARHfML5lcPiCAAAAAElFTkSuQmCC",
      "text/plain": [
       "<Figure size 300x600 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "def fig_3a(architectures):\n",
    "    \"\"\"Proportion of each type of service in the dataset\"\"\"\n",
    "    # query\n",
    "    occurence = Counter()\n",
    "    for arch in architectures:\n",
    "        occurence.update(\n",
    "            svc_info[x][\"capability\"] for x in arch.get_all_nodes_services()\n",
    "        )\n",
    "    occurence = Counter(normalize_dict(occurence, sum(occurence.values())))\n",
    "\n",
    "    # plotting\n",
    "    fig, ax = plt.subplots(figsize=(3, 6))\n",
    "    bottom = 0\n",
    "    for classification, usage in occurence.most_common():\n",
    "        ax.bar(\n",
    "            0,\n",
    "            usage,\n",
    "            bottom=bottom,\n",
    "            label=classification,\n",
    "            color=CAPABILITY_COLORS[classification],\n",
    "        )\n",
    "        bottom += usage\n",
    "\n",
    "    ax.set_xticks([])\n",
    "    ax.set_ylabel(\"Number of nodes (%)\")\n",
    "    ax.legend(handlelength=0.5, loc=\"upper left\", bbox_to_anchor=(1, 1), reverse=True)\n",
    "    plt.title(\"Fig 3a\")\n",
    "    cleanup()\n",
    "    plt.show()\n",
    "\n",
    "\n",
    "fig_3a(architectures)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "id": "fbd06ea9",
   "metadata": {
    "jupyter": {
     "source_hidden": true
    }
   },
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAhMAAAGtCAYAAACsi/a8AAAAOnRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjEwLjAsIGh0dHBzOi8vbWF0cGxvdGxpYi5vcmcvlHJYcgAAAAlwSFlzAAAPYQAAD2EBqD+naQAAYrxJREFUeJzt3Xl4TGf/BvB7siciQRKJELGLfVcpqkjkjZQgtLQqQtE21YqlxE5spRRvqbWWklYptdTaIHa1C7FvIYglkohlsn1/f/jlvKaxZDKTzAz357pytfOck3O+GTNn7nnOc56jEhEBERERUR6ZGboAIiIiMm0ME0RERKQThgkiIiLSCcMEERER6YRhgoiIiHTCMEFEREQ6YZggIiIinTBMEBERkU4YJoiIiEgnDBNEpLOrV69CpVJh8eLFBb7vnTt3QqVSYdWqVQW+byJ6hmGCiF5r8eLFUKlUL/wZMmRIvuxz/vz5aNasGVxdXWFtbY2yZcsiJCQEV69ezZf9EVHeWRi6ACIyHWPHjkXZsmU12qpXrw5PT088efIElpaWetvXsWPHULZsWbRt2xZFixbFlStXMH/+fGzYsAEnTpyAu7u73vZFRLphmCCiXPP390f9+vVfuMzGxkav+5o9e3aOtnbt2qF+/fpYunRpvvWIEJH2eJqDiHT2sjETK1euRNWqVWFjY4Pq1atjzZo16N69O8qUKZOn/WT/XlJSUo5lmZmZGDp0KNzc3FCoUCG0bdsW169fz9N+iEg77JkgolxLTk7GvXv3NNqcnZ1fuO5ff/2Fjz76CDVq1MDEiRPx4MED9OzZEyVLltRqn/fv30dmZibi4uIwduxYAEDLli1zrDd+/HioVCoMHjwYd+7cwfTp0+Hj44Pjx4/D1tZWq30SkXYYJogo13x8fHK0icgL1w0PD0fJkiWxd+9e2NvbA3gWAt5//314enrmep8lS5aEWq0GADg5OWHmzJnw9fXNsV5iYiLOnDmDwoULAwDq1q2LDz/8EPPnz8fXX3+d6/0RkfYYJogo12bNmoVKlSq9dr2bN28iJiYGQ4cOVYIEADRr1gw1atRASkpKrve5adMmPH36FGfOnMGyZcvw6NGjF67XrVs3JUgAQMeOHVGiRAls3LiRYYIonzFMEFGuNWzY8KUDMJ937do1AECFChVyLKtQoQKOHj2a6302b94cwLPBn4GBgahevTrs7e3x1VdfaaxXsWJFjccqlQoVKlTgpaREBYADMInIZJQvXx516tTB8uXLDV0KET2HYYKI9C57TMTFixdzLHtRmzaePHmC5OTkHO0XLlzQeCwiuHjxYp6vHCGi3GOYICK9c3d3R/Xq1bF06VKkpqYq7dHR0YiJiXnt72dkZODBgwc52v/55x/ExMS88FTL0qVL8fDhQ+XxqlWrcOvWLfj7++fxryCi3OKYCSLKFxMmTEBgYCAaN26MkJAQPHjwAD/++COqV6+uETBeJDU1FR4eHvjoo49QrVo1FCpUCDExMVi0aBEcHR0xYsSIHL9TrFgxNGnSBCEhIUhISMD06dNRoUIF9OrVK7/+RCL6fwwTRJQv2rRpg19//RWjR4/GkCFDULFiRSxevBhLlizB6dOnX/m7dnZ2+Oyzz7Bjxw6sWrUKT548gbu7O7p06YLhw4e/8NTF0KFDcfLkSUycOBEPHz5Ey5YtMXv2bNjZ2eXTX0hE2VTysovEiYjyQe3ateHi4oJt27YZuhQi0hOOmSCifJGeno6MjAyNtp07d+LEiRN4//33DVMUEeUL9kwQUb64evUqfHx80LVrV7i7u+Ps2bOYM2cOHB0dcerUKTg5ORm6RCLSE46ZIKJ8UbRoUdSrVw8LFizA3bt3UahQIQQEBGDSpEkMEkRvGPZMEBERkU44ZoKIiIh0wjBBREREOnnjw4SIICUl5aW3SSYiIiLdvPFh4uHDh3B0dNSYZpeIiIj0540PE0RERJS/DBomypQpA5VKleMnNDQUAPD06VOEhobCyckJ9vb2CAoKQkJCgiFLJiIion8xaJg4dOgQbt26pfxkT6/bqVMnAEBYWBjWr1+PlStXIjo6Gjdv3kSHDh0MWTIRERH9i1HNM9GvXz9s2LABFy5cQEpKClxcXBAZGYmOHTsCAM6ePYsqVapg//79aNSoUa62mZKSAkdHRyQnJ8PBwSE/yyciInorGc0MmGlpaVi2bBn69+8PlUqFI0eOID09HT4+Pso6Xl5eKF269CvDhFqthlqtVh6npKQAeHafgPT09Pz9I4iIiN4wlpaWr13HaMLEn3/+iaSkJHTv3h0AcPv2bVhZWaFIkSIa67m6uuL27dsv3c7EiRMxZsyYHO1bt27lrYiJiIi0FBgY+Np1jCZMLFy4EP7+/nB3d9dpO+Hh4ejfv7/yOCUlBR4eHmjVqhVPcxAREeUDowgT165dw99//43Vq1crbW5ubkhLS0NSUpJG70RCQgLc3Nxeui1ra2tYW1vnaLe0tMxVVw0RERFpxyjmmVi0aBGKFy+OgIAApa1evXqwtLREVFSU0nbu3DnExcXB29vbEGUSERHRCxi8ZyIrKwuLFi1CcHAwLCz+V46joyN69uyJ/v37o1ixYnBwcEDfvn3h7e2d6ys5iIiIKP8ZPEz8/fffiIuLQ48ePXIs++GHH2BmZoagoCCo1Wr4+flh9uzZBqiSiIiIXsao5pnID5xngoiIKH8ZxZgJIiIiMl0ME0RERKQTg4+ZMFVlhvxl6BIUVycFvH4lIiKifMKeCSIiItIJwwQRERHphGGCiIiIdMIwQURERDphmCAiIiKdMEwQERGRThgmiIiISCcME0RERKQThgkiIiLSCcMEERER6YRhgoiIiHTCMEFEREQ6YZggIiIinTBMEBERkU4YJoiIiEgnDBNERESkE4YJIiIi0gnDBBEREemEYYKIiIh0wjBBREREOmGYICIiIp0wTBAREZFOGCaIiIhIJwwTREREpBOGCSIiItIJwwQRERHphGGCiIiIdMIwQURERDphmCAiIiKd6BQm1Gq1zgXEx8eja9eucHJygq2tLWrUqIHDhw8ry0UEI0eORIkSJWBrawsfHx9cuHBB5/0SERGRfmgVJjZt2oTg4GCUK1cOlpaWsLOzg4ODA5o1a4bx48fj5s2bWu38wYMHaNy4MSwtLbFp0ybExsZi6tSpKFq0qLLO5MmTMXPmTMyZMwcHDx5EoUKF4Ofnh6dPn2q1LyIiIsofKhGR1620Zs0aDB48GA8fPkTr1q3RsGFDuLu7w9bWFomJiTh16hR2796N/fv3o3v37oiIiICLi8trdz5kyBDs3bsXu3fvfuFyEYG7uzsGDBiAgQMHAgCSk5Ph6uqKxYsXo3Pnzq/dR0pKChwdHZGcnAwHB4fXrp9bZYb8pbdt6erqpABDl0BERG+xXIUJb29vDB8+HP7+/jAze3lnRnx8PP773//C1dUVYWFhr9151apV4efnhxs3biA6OholS5bEl19+iV69egEALl++jPLly+PYsWOoXbu28nvNmjVD7dq1MWPGjBzbVKvVGqdfUlJS4OHhgXv37uk1TFQfvUVv29LVqdF+r15hYqmCKSQ3wm8YugIiItKCpaXla9fJVZjILzY2NgCA/v37o1OnTjh06BC++eYbzJkzB8HBwdi3bx8aN26MmzdvokSJEsrvffjhh1CpVFixYkWObY4ePRpjxozJ0R4ZGQk7O7v8+2OIiIjeQIGBga9dR+cw8ejRI2RmZubpW7+VlRXq16+Pffv2KW1ff/01Dh06hP379+cpTLBn4gXYM0FERHmUm54Ji7xuPDY2Ft26dcPRo0ehUqlQtWpVLFq0CPXr18/1NkqUKIGqVatqtFWpUgV//PEHAMDNzQ0AkJCQoBEmEhISNE57PM/a2hrW1tY52i0tLXP1hOSWOlOlt23p6rV/V5YRDVbV478BEREZhzxfGtqnTx989dVXSE1Nxf3799GhQwcEBwdrtY3GjRvj3LlzGm3nz5+Hp6cnAKBs2bJwc3NDVFSUsjwlJQUHDx6Et7d3XksnIiIiPcp1mAgMDER8fLzy+O7du2jbti3s7OxQpEgRtG7dGgkJCVrtPCwsDAcOHMCECRNw8eJFREZGYt68eQgNDQUAqFQq9OvXD+PGjcO6desQExODbt26wd3dHe3atdNqX0RERJQ/cn2ao2vXrmjRogVCQ0PRt29ffPXVV6hWrRqaNWuG9PR0bN++HQMGDNBq5w0aNMCaNWsQHh6OsWPHomzZspg+fTo++eQTZZ1vv/0Wjx49Qu/evZGUlIQmTZpg8+bNyuBNIiIiMiytBmAmJydj8ODBOHbsGObMmQMLCwvs3LkTmZmZaNy4MRo0aJCfteYJ55kAMNqxYArJjdHJhq6AiIj0TKsBmI6OjpgzZw727NmD4OBg+Pr6IiIigpdcEhERvcW0GoCZmJiII0eOoEaNGjhy5AgcHBxQp04dbNy4Mb/qIyIiIiOX6zARGRmJUqVKISAgAJ6enti0aRNGjRqFtWvXYvLkyfjwww+1HoBJREREpi/XYSI8PBw///wzbt++jaioKIwYMQIA4OXlhZ07d8LX15eXaxIREb2Fch0mUlNTUblyZQBA+fLl8fjxY43lvXr1woEDB/RbHRERERm9XA/ADA4ORkBAAN5//30cPnwYn376aY51ihcvrtfiiIiIyPjlOkxMmzYNzZs3x9mzZ9G9e3e0atUqP+siIiIiE6HVpaFt2rRBmzZt8qsWIiIiMkG5GjPx22+/5XqD169fx969e/NcEBEREZmWXIWJn376CVWqVMHkyZNx5syZHMuTk5OxceNGfPzxx6hbty7u37+v90KJiIjIOOXqNEd0dDTWrVuH//73vwgPD0ehQoXg6uoKGxsbPHjwALdv34azszO6d++OU6dOwdXVNb/rJiIiIiOR6zETbdu2Rdu2bXHv3j3s2bMH165dw5MnT+Ds7Iw6deqgTp06MDPL8x3NiYiIyERpNQATAJydnXn7byIiIlKwK4GIiIh0wjBBREREOmGYICIiIp0wTBAREZFOdA4TmZmZOH78OB48eKCPeoiIiMjEaB0m+vXrh4ULFwJ4FiSaNWuGunXrwsPDAzt37tR3fURERGTktA4Tq1atQq1atQAA69evx5UrV3D27FmEhYVh2LBhei+QiIiIjJvWYeLevXtwc3MDAGzcuBGdOnVCpUqV0KNHD8TExOi9QCIiIjJuWocJV1dXxMbGIjMzE5s3b4avry8A4PHjxzA3N9d7gURERGTctJ4BMyQkBB9++CFKlCgBlUoFHx8fAMDBgwfh5eWl9wKJiIjIuGkdJkaPHo3q1avj+vXr6NSpE6ytrQEA5ubmGDJkiN4LJCIiIuOmdZgAgI4dOwIAnj59qrQFBwfrpyIiIiIyKVqPmcjMzERERARKliwJe3t7XL58GQAwYsQI5ZJRIiIientoHSbGjx+PxYsXY/LkybCyslLaq1evjgULFui1OCIiIjJ+WoeJpUuXYt68efjkk080rt6oVasWzp49q9fiiIiIyPhpHSbi4+NRoUKFHO1ZWVlIT0/XS1FERERkOrQOE1WrVsXu3btztK9atQp16tTRS1FERERkOrS+mmPkyJEIDg5GfHw8srKysHr1apw7dw5Lly7Fhg0b8qNGIiIiMmJa90wEBgZi/fr1+Pvvv1GoUCGMHDkSZ86cwfr165XZMImIiOjtoVWYyMjIwNixY1G2bFls27YNd+7cwePHj7Fnzx60atVK652PHj0aKpVK4+f5WTSfPn2K0NBQODk5wd7eHkFBQUhISNB6P0RERJR/tAoTFhYWmDx5MjIyMvRWQLVq1XDr1i3lZ8+ePcqysLAwrF+/HitXrkR0dDRu3ryJDh066G3fREREpDutx0y0bNkS0dHRKFOmjH4KsLBQ7kL6vOTkZCxcuBCRkZFo0aIFAGDRokWoUqUKDhw4gEaNGull/0RERKQbrcOEv78/hgwZgpiYGNSrVw+FChXSWN62bVuttnfhwgW4u7vDxsYG3t7emDhxIkqXLo0jR44gPT1duZEYAHh5eaF06dLYv3//S8OEWq2GWq1WHqekpAAA0tPT9XrpqrW56G1bunrt32VmUzCF5AYvHyYiMimWlpavXUclIlp9KpqZvfzMiEqlQmZmZq63tWnTJqSmpqJy5cq4desWxowZg/j4eJw6dQrr169HSEiIRjAAgIYNG6J58+b47rvvXrjN0aNHY8yYMTnaIyMjYWdnl+vaiIiI6NmFF6+jdZjIT0lJSfD09MS0adNga2ubpzDxop4JDw8P3Lt3Dw4ODnqrtfroLXrblq5OjfZ79QoTSxVMIbkRfsPQFRARkRZy0zORp7uG5pciRYqgUqVKuHjxInx9fZGWloakpCQUKVJEWSchIeGFYyyyWVtbK7dFf56lpWWunpDcUmeq9LYtXb3278p6+urlBUmP/wZERGQctA4TY8eOfeXykSNH5rmY1NRUXLp0CZ9++inq1asHS0tLREVFISgoCABw7tw5xMXFwdvbO8/7ICIiIv3SOkysWbNG43F6ejquXLkCCwsLlC9fXqswMXDgQLRp0waenp64efMmRo0aBXNzc3Tp0gWOjo7o2bMn+vfvj2LFisHBwQF9+/aFt7c3r+QgIiIyIlqHiWPHjuVoS0lJQffu3dG+fXuttnXjxg106dIF9+/fh4uLC5o0aYIDBw7AxcUFAPDDDz/AzMwMQUFBUKvV8PPzw+zZs7UtmYiIiPKR3gZgxsTEoE2bNrh69ao+Nqc3KSkpcHR0RHJysl4HYJYZ8pfetqWrq5MCXr3CaMeCKSQ3RicbugIiItIzre/N8TLJyclITuYHBRER0dtG69McM2fO1HgsIrh16xZ++eUX+Pv7660wIiIiMg1ah4kffvhB47GZmRlcXFwQHByM8PBwvRVGREREpkHrMHHlypX8qIOIiIhMlNZjJnr06IGHDx/maH/06BF69Oihl6KIiIjIdGgdJpYsWYInT57kaH/y5AmWLl2ql6KIiIjIdOT6NEdKSgpEBCKChw8fwsbmf3eizMzMxMaNG1G8ePF8KZKIiIiMV67DRJEiRaBSqaBSqVCpUqUcy1Uq1Qvv1klERERvtlyHiR07dkBE0KJFC/zxxx8oVqyYsszKygqenp5wd3fPlyKJiIjIeOU6TDRr1gzAs6s5SpcuDZXKeO6aSURERIaj9QDM7du3Y9WqVTnaV65ciSVLluilKCIiIjIdWoeJiRMnwtnZOUd78eLFMWHCBL0URURERKZD6zARFxeHsmXL5mj39PREXFycXooiIiIi06F1mChevDhOnjyZo/3EiRNwcnLSS1FERERkOrQOE126dMHXX3+NHTt2IDMzE5mZmdi+fTu++eYbdO7cOT9qJCIiIiOm9b05IiIicPXqVbRs2RIWFs9+PSsrC926deOYCSIioreQ1mHCysoKK1asQEREBE6cOAFbW1vUqFEDnp6e+VEfERERGTmtw0S2MmXKQERQvnx5pYeCiIiI3j5aj5l4/PgxevbsCTs7O1SrVk25gqNv376YNGmS3gskIiIi46Z1mAgPD8eJEyewc+dOjZt9+fj4YMWKFXotjoiIiIyf1ucn/vzzT6xYsQKNGjXSmFK7WrVquHTpkl6LIyIiIuOndc/E3bt3X3ir8UePHvF+HURERG8hrcNE/fr18ddffymPswPEggUL4O3trb/KiIiIyCRofZpjwoQJ8Pf3R2xsLDIyMjBjxgzExsZi3759iI6Ozo8aiYiIyIhp3TPRpEkTHD9+HBkZGahRowa2bt2K4sWLY//+/ahXr15+1EhERERGLE8TRJQvXx7z58/Xdy1ERERkgrTumTA3N8edO3dytN+/fx/m5uZ6KYqIiIhMh9ZhQkRe2K5Wq2FlZaVzQURERGRacn2aY+bMmQCeXb2xYMEC2NvbK8syMzOxa9cueHl56b9CIiIiMmq5DhM//PADgGc9E3PmzNE4pWFlZYUyZcpgzpw5+q+QiIiIjFquw8SVK1cAAM2bN8fq1atRtGjRfCuKiIiITIfWV3Ps2LEjP+ogIiIiE5WrMNG/f39ERESgUKFC6N+//yvXnTZtWp4KmTRpEsLDw/HNN99g+vTpAICnT59iwIAB+O2336BWq+Hn54fZs2fD1dU1T/sgIiIi/ctVmDh27BjS09OV/3+ZvN6b49ChQ5g7dy5q1qyp0R4WFoa//voLK1euhKOjI7766it06NABe/fuzdN+iIiISP9yFSaeP7Wh79Mcqamp+OSTTzB//nyMGzdOaU9OTsbChQsRGRmJFi1aAAAWLVqEKlWq4MCBA2jUqJFe6yAiIqK8ydMMmPoUGhqKgIAA+Pj4aISJI0eOID09HT4+Pkqbl5cXSpcujf379780TKjVaqjVauVxSkoKACA9PV3pXdEHa/MXz7dhCK/9u8xsCqaQ3NDjvwEREeU/S0vL166jdZh49OgRJk2ahKioKNy5cwdZWVkayy9fvpzrbf322284evQoDh06lGPZ7du3YWVlhSJFimi0u7q64vbt2y/d5sSJEzFmzJgc7Vu3boWdnV2ua3udyQ31timdbdy48dUr1JpXMIXkxutqJSIioxIYGPjadbQOE5999hmio6Px6aefokSJEnkeJ3H9+nV888032LZtG2xs9PfNOTw8XGOQaEpKCjw8PNCqVSs4ODjobT/VR2/R27Z0dWq036tXmFiqYArJjfAbhq6AiIj0TOswsWnTJvz1119o3LixTjs+cuQI7ty5g7p16ypt2TNp/vjjj9iyZQvS0tKQlJSk0TuRkJAANze3l27X2toa1tbWOdotLS1z1VWTW+rMvIWo/PDavyvracEUkht6/DcgIiLjoHWYKFq0KIoVK6bzjlu2bImYmBiNtpCQEHh5eWHw4MHw8PCApaUloqKiEBQUBAA4d+4c4uLi4O3trfP+iYiISD+0DhMREREYOXIklixZotMYhMKFC6N69eoabYUKFYKTk5PS3rNnT/Tv3x/FihWDg4MD+vbtC29vb17JQUREZERyFSbq1KmjMTbi4sWLcHV1RZkyZXJ0sR89elRvxf3www8wMzNDUFCQxqRVREREZDxyFSbatWuXz2U8s3PnTo3HNjY2mDVrFmbNmlUg+yciIiLt5SpMjBo1Kr/rIHpmtKOhK/if0cmGroCIyCSYafsLhw4dwsGDB3O0Hzx4EIcPH9ZLUURERGQ6tA4ToaGhuH79eo72+Ph4hIaG6qUoIiIiMh1ah4nY2FiNuSGy1alTB7GxsXopioiIiEyH1mHC2toaCQkJOdpv3boFCwuD3+qDiIiICpjWYaJVq1YIDw9HcvL/BqclJSVh6NCh8PX11WtxREREZPy07kqYMmUKmjVrBk9PT9SpUwcAcPz4cbi6uuKXX37Re4FERERk3LQOE6VKlcLJkyexfPlynDhxAra2tggJCUGXLl30eu8LIiIiMg1ahYn09HR4eXlhw4YN6N27d37VRERERCZEqzETlpaWePrUiO5ASURERAaXp3kmvvvuO2RkZORHPURERGRitB4zcejQIURFRWHr1q2oUaMGChUqpLF89erVeiuOiIiIjJ/WYaJIkSIICgrKj1qIiIjIBGkdJhYtWpQfdRAREZGJ0nrMBBEREdHz8jT/9apVq/D7778jLi4OaWlpGsuOHj2ql8KIiIjINGjdMzFz5kyEhITA1dUVx44dQ8OGDeHk5ITLly/D398/P2okIiIiI6Z1mJg9ezbmzZuH//73v7CyssK3336Lbdu24euvv9a4XwcRERG9HbQOE3FxcXj33XcBALa2tnj48CEA4NNPP8Wvv/6q3+qIiIjI6GkdJtzc3JCYmAgAKF26NA4cOAAAuHLlCkREv9URERGR0dM6TLRo0QLr1q0DAISEhCAsLAy+vr746KOP0L59e70XSERERMZN66s55s2bh6ysLADPptZ2cnLCvn370LZtW/Tp00fvBRIZqzJD/jJ0CYqrkwIMXQIRvcW0DhNmZmYwM/tfh0bnzp3RuXNnvRZFREREpoOTVhEREZFOGCaIiIhIJwwTREREpJNchYl169YhPT09v2shIiIiE5SrMNG+fXskJSUBAMzNzXHnzp38rImIiIhMSK7ChIuLizI5lYhApVLla1FERERkOnJ1aejnn3+OwMBAqFQqqFQquLm5vXTdzMxMvRVHRERExi9XYWL06NHo3LkzLl68iLZt22LRokUoUqRIPpdGREREpiDXk1Z5eXnBy8sLo0aNQqdOnWBnZ5efdREREZGJ0PrS0FGjRsHOzg53797Fnj17sGfPHty9ezdPO//pp59Qs2ZNODg4wMHBAd7e3ti0aZOy/OnTp8qU3fb29ggKCkJCQkKe9kVERET5Q+sw8fjxY/To0QPu7u5477338N5778Hd3R09e/bE48ePtdpWqVKlMGnSJBw5cgSHDx9GixYtEBgYiNOnTwMAwsLCsH79eqxcuRLR0dG4efMmOnTooG3JRERElI+0DhNhYWGIjo7GunXrkJSUhKSkJKxduxbR0dEYMGCAVttq06YNWrdujYoVK6JSpUoYP3487O3tceDAASQnJ2PhwoWYNm0aWrRogXr16mHRokXYt2+fcmUJERERGZ7WYeKPP/7AwoUL4e/vr5yeaN26NebPn49Vq1bluZDMzEz89ttvePToEby9vXHkyBGkp6fDx8dHWcfLywulS5fG/v3787wfIiIi0i+t7xr6+PFjuLq65mgvXry41qc5ACAmJgbe3t54+vQp7O3tsWbNGlStWhXHjx+HlZVVjqtGXF1dcfv27ZduT61WQ61WK49TUlIAAOnp6XqdxdPaXPS2LV299u8ysymYQnLjDarVpF4DRER5ZGlp+dp1VCKi1RGxZcuWcHJywtKlS2Fj8+zA/+TJEwQHByMxMRF///23VkWmpaUhLi4OycnJWLVqFRYsWIDo6GgcP34cISEhGsEAABo2bIjmzZvju+++e+H2Ro8ejTFjxuRoj4yM5BUoREREWgoMDHztOlqHiVOnTsHPzw9qtRq1atUCAJw4cQI2NjbYsmULqlWrlrdq/5+Pjw/Kly+Pjz76CC1btsSDBw80eic8PT3Rr18/hIWFvfD3X9Qz4eHhgXv37sHBwUGn2p5XffQWvW1LV6dG+716hYmlCqaQ3Ai/8erlJlSrSb0GiIjyKDc9E1qf5qhevTouXLiA5cuX4+zZswCALl264JNPPoGtra32Vf5LVlYW1Go16tWrB0tLS0RFRSEoKAgAcO7cOcTFxcHb2/ulv29tbQ1ra+sc7ZaWlrl6QnJLnWk8U4q/9u/KelowheTGG1SrSb0GiIjykdZhAgDs7OzQq1cvnXceHh4Of39/lC5dGg8fPkRkZCR27tyJLVu2wNHRET179kT//v1RrFgxODg4oG/fvvD29kajRo103jcRERHpR57ChL7cuXMH3bp1w61bt+Do6IiaNWtiy5Yt8PX1BQD88MMPMDMzQ1BQENRqNfz8/DB79mxDlkxERET/YtAwsXDhwlcut7GxwaxZszBr1qwCqoiIiIi0pfU8E0RERETPY5ggIiIineQpTCQlJWHBggUIDw9HYmIiAODo0aOIj4/Xa3FERERk/LQeM3Hy5En4+PjA0dERV69eRa9evVCsWDGsXr0acXFxWLp0aX7USUREREZK656J/v37o3v37rhw4YIyAyYAtG7dGrt27dJrcURERGT8tA4Thw4dQp8+fXK0lyxZ8pX3zCAiIqI3k9ZhwtraWrl51vPOnz8PFxcXvRRFREREpkPrMNG2bVuMHTtWuUuhSqVCXFwcBg8erEx7TURERG8PrQdgTp06FR07dkTx4sXx5MkTNGvWDLdv34a3tzfGjx+fHzUSkR7UWFLD0CUAAGKCYwxdAhHpmdZhwtHREdu2bcOePXtw8uRJpKamom7duvDx8cmP+oiIiMjI5Xk67SZNmqBJkyb6rIWIiIhMkNZhYubMmS9sV6lUsLGxQYUKFfDee+/B3Nxc5+KIiIjI+GkdJn744QfcvXsXjx8/RtGiRQEADx48gJ2dHezt7XHnzh2UK1cOO3bsgIeHh94LJiIiIuOi9dUcEyZMQIMGDXDhwgXcv38f9+/fx/nz5/HOO+9gxowZiIuLg5ubG8LCwvKjXiIiIjIyWvdMDB8+HH/88QfKly+vtFWoUAHff/89goKCcPnyZUyePJmXiRIREb0ltO6ZuHXrFjIyMnK0Z2RkKDNguru74+HDh7pXR0REREZP6zDRvHlz9OnTB8eOHVPajh07hi+++AItWrQAAMTExKBs2bL6q5KIiIiMltZhYuHChShWrBjq1asHa2trWFtbo379+ihWrBgWLlwIALC3t8fUqVP1XiwREREZH63HTLi5uWHbtm04e/Yszp8/DwCoXLkyKleurKzTvHlz/VVIRERERi3Pk1Z5eXnBy8tLn7UQERGRCcpTmLhx4wbWrVuHuLg4pKWlaSybNm2aXgojIiIi06B1mIiKikLbtm1Rrlw5nD17FtWrV8fVq1chIqhbt25+1EhERERGTOsBmOHh4Rg4cCBiYmJgY2ODP/74A9evX0ezZs3QqVOn/KiRiIiIjJjWYeLMmTPo1q0bAMDCwgJPnjyBvb09xo4di++++07vBRIREZFx0zpMFCpUSBknUaJECVy6dElZdu/ePf1VRkRERCZB6zETjRo1wp49e1ClShW0bt0aAwYMQExMDFavXo1GjRrlR41ERERkxLQOE9OmTUNqaioAYMyYMUhNTcWKFStQsWJFXslBRET0FtI6TJQrV075/0KFCmHOnDl6LYiIiIhMi9ZjJsqVK4f79+/naE9KStIIGkRERPR20DpMXL16FZmZmTna1Wo14uPj9VIUERERmY5cn+ZYt26d8v9btmyBo6Oj8jgzMxNRUVEoU6aMXosjIiIi45frMNGuXTsAgEqlQnBwsMYyS0tLlClThncKJSIiegvl+jRHVlYWsrKyULp0ady5c0d5nJWVBbVajXPnzuGDDz7QaucTJ05EgwYNULhwYRQvXhzt2rXDuXPnNNZ5+vQpQkND4eTkBHt7ewQFBSEhIUGr/RAREVH+0XrMxJUrV+Ds7KyXnUdHRyM0NBQHDhzAtm3bkJ6ejlatWuHRo0fKOmFhYVi/fj1WrlyJ6Oho3Lx5Ex06dNDL/omIiEh3ebpraFRUFKKiopQeiuf9/PPPud7O5s2bNR4vXrwYxYsXx5EjR/Dee+8hOTkZCxcuRGRkJFq0aAEAWLRoEapUqYIDBw5wkiwiIiIjoHWYGDNmDMaOHYv69eujRIkSUKlUeismOTkZAFCsWDEAwJEjR5Ceng4fHx9lHS8vL5QuXRr79+9nmCB6Q5UZ8pehSwAAXJ0UYOgSiEyC1mFizpw5WLx4MT799FO9FpKVlYV+/fqhcePGqF69OgDg9u3bsLKyQpEiRTTWdXV1xe3bt1+4HbVaDbVarTxOSUkBAKSnpyM9PV1v9Vqbi962pavX/l1mNgVTSG68QbWa1GsAgDWsC6CS18tVrUby3OrzmEFkqiwtLV+7jkpEtHrXOjk54Z9//kH58uXzXNiLfPHFF9i0aRP27NmDUqVKAQAiIyMREhKiEQ4AoGHDhmjevPkL71I6evRojBkzJkd7ZGQk7Ozs9FozERHRmy4wMPC162gdJgYPHgx7e3uMGDEiz4X921dffYW1a9di165dKFu2rNK+fft2tGzZEg8ePNDonfD09ES/fv0QFhaWY1sv6pnw8PDAvXv34ODgoLeaq4/eordt6erUaL9XrzCxVMEUkhvhN1693IRqNanXAADvSO8CqOT19n+8/7XrGMtzm5vnFTtzfnkxiPdHGboCekPlpmdC69McT58+xbx58/D333+jZs2aOXaizc2+RAR9+/bFmjVrsHPnTo0gAQD16tWDpaUloqKiEBQUBAA4d+4c4uLi4O394gOjtbU1rK1zdudaWlrm6gnJLXWm/saK6Oq1f1fW04IpJDfeoFpN6jUAQA31a9cpCLmq1Uie21wdM1Q5ZwQ2CD0e34i0pXWYOHnyJGrXrg0AOHXqlMYybQdjhoaGIjIyEmvXrkXhwoWVcRCOjo6wtbWFo6Mjevbsif79+6NYsWJwcHBA37594e3tzcGXRERERkLrMLFjxw697fynn34CALz//vsa7YsWLUL37t0BAD/88APMzMwQFBQEtVoNPz8/zJ49W281EBERkW7yNM8EAFy8eBGXLl3Ce++9B1tbW4iI1j0TuRmuYWNjg1mzZmHWrFl5LZWIiIjykdYzYN6/fx8tW7ZEpUqV0Lp1a9y6dQsA0LNnTwwYMEDvBRIREZFx0zpMhIWFwdLSEnFxcRqXWn700Uc5ZrQkIiKiN5/Wpzm2bt2KLVu2KHNBZKtYsSKuXbumt8KIiIjINGjdM/Ho0aMXTv6UmJj4wksyiYiI6M2mdZho2rQpli5dqjxWqVTIysrC5MmT0bx5c70WR0RERMZP69MckydPRsuWLXH48GGkpaXh22+/xenTp5GYmIi9e/fmR41ERERkxLTumahevTrOnz+PJk2aIDAwEI8ePUKHDh1w7Ngxvd+vg4iIiIxfnuaZcHR0xLBhw/RdCxEREZkgrXsmFi1ahJUrV+ZoX7lyJZYsWaKXooiIiMh0aN0zMXHiRMydOzdHe/HixdG7d28EBwfrpTAiItKz0Y6GruCZ0cmGroD0TOueibi4uBx39wSe3RY8Li5OL0URERGR6dA6TBQvXhwnT57M0X7ixAk4OTnppSgiIiIyHVqHiS5duuDrr7/Gjh07kJmZiczMTGzfvh3ffPMNOnfunB81EhERkRHTesxEREQErl69ipYtW8LC4tmvZ2VloVu3bpgwYYLeCyQiIiLjplWYEBHcvn0bixcvxrhx43D8+HHY2tqiRo0a8PT0zK8aiYiIyIhpHSYqVKiA06dPo2LFiqhYsWJ+1UVEREQmQqsxE2ZmZqhYsSLu37+fX/UQERGRidF6AOakSZMwaNAgnDp1Kj/qISIiIhOj9QDMbt264fHjx6hVqxasrKxga2ursTwxMVFvxREREZHx0zpMTJ8+PR/KICIiIlOldZjgdNlERET0PK3HTADApUuXMHz4cHTp0gV37twBAGzatAmnT5/Wa3FERERk/LTumYiOjoa/vz8aN26MXbt2Yfz48ShevDhOnDiBhQsXYtWqVflRJxERvU3+Djd0Bf/jM/HVy43lBmqAwW6ipnXPxJAhQzBu3Dhs27YNVlZWSnuLFi1w4MABvRZHRERExk/rMBETE4P27dvnaC9evDju3bunl6KIiIjIdGgdJooUKYJbt27laD927BhKliypl6KIiIjIdGgdJjp37ozBgwfj9u3bUKlUyMrKwt69ezFw4EB069YtP2okIiIiI6Z1mJgwYQK8vLzg4eGB1NRUVK1aFe+99x7effddDB8+PD9qJCIiIiOm9dUcVlZWmD9/PkaOHImYmBikpqaiTp06vOkXERHRWyrXYSIrKwtTpkzBunXrkJaWhpYtW2LUqFE5ptMmIiKit0uuT3OMHz8eQ4cOhb29PUqWLIkZM2YgNDQ0P2sjIiIiE5DrMLF06VLMnj0bW7ZswZ9//on169dj+fLlyMrKys/6iIiIyMjlOkzExcWhdevWymMfHx+oVCrcvHkzXwojIiIi05DrMJGRkQEbGxuNNktLS6Snp+d557t27UKbNm3g7u4OlUqFP//8U2O5iGDkyJEoUaIEbG1t4ePjgwsXLuR5f0RERKR/uR6AKSLo3r07rK2tlbanT5/i888/R6FChZS21atX53rnjx49Qq1atdCjRw906NAhx/LJkydj5syZWLJkCcqWLYsRI0bAz88PsbGxOYINERERGUauw8SLbj3etWtXnXbu7+8Pf3//Fy4TEUyfPh3Dhw9HYGAggGfjNlxdXfHnn3+ic+fOOu2biIiI9CPXYWLRokX5WUcOV65cwe3bt+Hj46O0OTo64p133sH+/ftfGibUajXUarXyOCUlBQCQnp6u0ymZf7M2F71tS1ev/bvMjKgX5w2q1aReAwCsYf3adQpCrmo1kuc2V8cMMc//QnIjN7Uay/vLlJ5X4I06buWFpaXla9dRiYhRvGtVKhXWrFmDdu3aAQD27duHxo0b4+bNmyhRooSy3ocffgiVSoUVK1a8cDujR4/GmDFjcrRHRkbCzs4uX2onIiJ6U2WfHXgVrWfANHbh4eHo37+/8jglJQUeHh5o1aoVHBwc9Laf6qO36G1bujo12u/VK0wsVTCF5Eb4jVcvN6FaTeo1AMA70rsAKnm9/R/vf+06xvLc5uZ5xc6cX14M4v1Rr1/HWN5frzsOAMbzvAKvf26N5XkFcvfc5gOjDRNubm4AgISEBI2eiYSEBNSuXfulv2dtba0xSDSbpaVlrrpqckudqdLbtnT12r8r62nBFJIbb1CtJvUaAKCG+rXrFIRc1Wokz22ujhmqzPwvJDdyU6uxvL9M6XkF3qjjVn7R+kZfBaVs2bJwc3NDVFSU0paSkoKDBw/C29s4vmERERGRgXsmUlNTcfHiReXxlStXcPz4cRQrVgylS5dGv379MG7cOFSsWFG5NNTd3V0ZV0FERESGZ9AwcfjwYTRv3lx5nD3WITg4GIsXL8a3336LR48eoXfv3khKSkKTJk2wefNmzjFBRERkRAwaJt5//3286mISlUqFsWPHYuzYsQVYFREREWnDaMdMEBERkWlgmCAiIiKdMEwQERGRThgmiIiISCcME0RERKQThgkiIiLSCcMEERER6YRhgoiIiHTCMEFEREQ6YZggIiIinTBMEBERkU4YJoiIiEgnDBNERESkE4YJIiIi0gnDBBEREemEYYKIiIh0wjBBREREOmGYICIiIp0wTBAREZFOGCaIiIhIJwwTREREpBOGCSIiItIJwwQRERHphGGCiIiIdMIwQURERDphmCAiIiKdMEwQERGRThgmiIiISCcME0RERKQThgkiIiLSCcMEERER6YRhgoiIiHRiEmFi1qxZKFOmDGxsbPDOO+/gn3/+MXRJRERE9P+MPkysWLEC/fv3x6hRo3D06FHUqlULfn5+uHPnjqFLIyIiIphAmJg2bRp69eqFkJAQVK1aFXPmzIGdnR1+/vlnQ5dGREREACwMXcCrpKWl4ciRIwgPD1fazMzM4OPjg/3797/wd9RqNdRqtfI4OTkZAJCYmIj09HS91WaR8Uhv29LV/fv3X71CmlXBFJIbb1CtJvUaAGDxxDje7rmq1Uie29zUitSM/C8kN3JTq7G8v0zpeQXeqONWXlhaWqJw4cJQqVQvX0mMWHx8vACQffv2abQPGjRIGjZs+MLfGTVqlADgD3/4wx/+8Ic/evpJTk5+5ee1cXxV0aPw8HD0799feZyVlYXExEQ4OTm9OlUVsJSUFHh4eOD69etwcHAwdDmvZEq1AqZVL2vNH6w1f7DW/GPs9RYuXPiVy406TDg7O8Pc3BwJCQka7QkJCXBzc3vh71hbW8Pa2lqjrUiRIvlVos4cHByM8oXzIqZUK2Ba9bLW/MFa8wdrzT+mVm82ox6AaWVlhXr16iEqKkppy8rKQlRUFLy9vQ1YGREREWUz6p4JAOjfvz+Cg4NRv359NGzYENOnT8ejR48QEhJi6NKIiIgIJhAmPvroI9y9excjR47E7du3Ubt2bWzevBmurq6GLk0n1tbWGDVqVI5TMsbIlGoFTKte1po/WGv+YK35x9Tq/TeViIihiyAiIiLTZdRjJoiIiMj4MUwQERGRThgmiIiISCcME0RERKQThgkiIiLSCcOEkeBFNfQ2y8rKyvd9ZGZm5vs+iPTNVD4bGCYMJD4+Htu2bcOSJUuQkZEBlUplMi8aMh2m8poyM3t2KFq+fDlu3boFQL8B49q1a5g5cyZOnTqlt23mpw0bNrz0zsjGJCPDiO7s+Ro3b97Exo0bER8fb+hSXishIQFHjhwBAKO6p9SrMEwYQExMDFq0aIEhQ4YgNDQU77zzDtLT003mRWMqbt68iQ0bNmD16tU4evSoocspMImJicoB01ReUyKChw8fIiQkBD///DOA/wUMXZ06dQp+fn74559/TOJ1MHr0aAwaNAgzZ87E3bt3DV3OS126dAlDhw7FwYMHjT60nj59Gq1bt8aKFSuMPlDGxsYiKCgIw4YNM4nXq0Ln+4STVs6cOSPOzs4yfPhwuXbtmly+fFmcnZ1lw4YNhi5N77KysrRq16eTJ09KuXLlpGHDhuLs7Cz169eXlStX5vt+DS0yMlLeffddKVmypPj7+8utW7dEpGCec11k1zd+/Hjx9fWVmzdv6mW7sbGx4uTkJN9++63yXBizoUOHSvHixWXHjh0SHx+vtBvbv9+JEyekTJkyEhgYKEuXLjV0Oa906tQpKVKkiAwYMEBOnz5t6HJeKSYmRpycnKR///5y+PBhpT3739/YXgfPY5goQElJSeLv7y/9+vXTaPfz85P58+fLtGnTJDY2Vh4/fmygCvUn+0V/8OBBmTJlikyfPl22bt2aY3l+uHjxopQqVUq+/fZbSUpKksOHD0twcLD06NFDMjIyjPoNqYs5c+aInZ2dTJo0SX799VcpWbKkdO/eXWMdY/nbMzMzX9i+c+dOcXFxkS1btoiIbvU+efJEOnXqJF999VWO9jt37si5c+fyvO38sHXrVilXrpzs3bs3xzJ9hSt9OHfunBQvXlwGDx4sjx49MnQ5r/TgwQN5//33ZdCgQTmWpaamyr179wxQ1YvduXNHatSoId9++22OZf9+vxjL+/h5DBMFJPsff8GCBRoHi4iICLGwsJDmzZtLlSpVxNXVVVasWKHxO6Ymu+4//vhDnJ2dxcfHR1q2bCkVK1aUn376Kcd6+qRWq6V///7y4YcfilqtVtoXLlwoTk5ORnXw0KcFCxaIlZWVrF27Vmn7/vvvJTQ0VI4fPy5xcXFKSH3ZB7khbN68WXbu3KnR1rNnT3n33XflwYMHOm07NTVVatasKf/973+Vti1btkjfvn2lWLFiYm9vL7169TKaHouFCxfKe++9l6Otffv2YmtrK23atJHff//dQNU9k5mZKV988YV88sknGu/flJQUOX/+vBw4cEAuX75swAo13bhxQ6pXry6bNm1S2nbv3i3jxo2TSpUqSc2aNWXRokWGK/A5hw4dkvr168vVq1c12n744QepX7++9OjRQ/7++28DVvhqDBMF4MqVKzJ79mw5deqURvuuXbukfPnysm7dOiXht23bVurXr2+IMnXy72CwZ88eKVGihBIeDh48KA4ODmJpaSnffffdS39PV0+fPpXvv/9e5s2bp7H9M2fOiKen5ws/OEw1tGU7fvy42NraSrdu3TTavb29xcPDQxwcHMTT01N69eolDx8+NFCVOZ08eVJq1qwpTk5O0rt3b9m4caOIiOzbt08aNGgg+/fvF5G8h5/k5GQJCAiQzz77TE6fPi3fffedeHl5yUcffSQzZsyQVatWiYWFhXz//fd6+5t0sWjRInF0dJR//vlHEhMTpU2bNtK0aVMJCgqShQsXSoMGDaR169Zy//59g9bp6+sro0aNUh6vW7dOgoODxd7eXuzs7MTb21v5tzS0Y8eOSZkyZWT58uXy5MkTmT17ttSvX198fX0lNDRUPv/8c1GpVBphw1B27dol9vb2ynM3Z84cadq0qdSrV0+6d++unLZ9/vSXMWGYyGcnT56USpUqSfv27WX9+vUay27cuCGXLl0SEZH09HQREZkyZYq88847kpaWVuC15tWMGTNkw4YNyt+Qnp4uEyZMkG+++UZEROLi4qRMmTLSrVs3GTJkiFhaWmr0UOhDSkqK3LlzRxITEyUpKUlpzw4Kt27dkgoVKkhcXJyy7OjRo3qtwVDOnz8vX375pVSvXl3mz58vIiIdO3aUypUrS3R0tFy6dEm+/PJLcXV1lTVr1hiszheFtkuXLsn27dvF29tb3nnnHWncuLFER0dLqVKlpFevXjrv88cff5SaNWuKq6urFClSRObOnSsXL15Ulrdt21Y6duyo83704fLly9KuXTuxtbWVIkWKiLe3t6xbt06Sk5NFRGT9+vWiUqnk2LFjBV7bhQsX5NChQyIiEhISIh4eHrJr1y4ZNGiQlC5dWoKDg2XNmjWye/duad68ufTq1UujZ7AgXb16VRYvXqw8/vTTT8Xe3l7Kli0rNjY28v333ytf7O7fvy/VqlWT0aNHG6TW7OPWrVu3JDU1VT788EMpVaqU1KtXT6ytrWXUqFHK2Inbt2+LSqWS5cuXG6TW12GYyEdnzpyRokWLypAhQ16aJjMyMjQe9+zZU3r06KF8MJuCRo0aibOzs2zbtk2p+86dO7J79255/PixNGnSRHr06CEizwZuOTg4iEqlkqlTp+pl/2fOnJEOHTpIYGCgDBs2TPkm+/yH19mzZ8XJyUkJEyNGjJCiRYvKvXv3TLJn4t81nzt3TsLCwsTLy0uqVKkiderUkevXryvLExMTxdHRUaZMmVLQpYqIZu/CzZs3NbpyRZ6d2z5x4oR06NBBWrZsKU5OTmJvby8HDx7M9T5SUlIkLi5O/vrrL9m+fbvSfvHiRTlw4IAkJCRorP/kyRMJCAiQMWPG5PGv0s2lS5fk6NGjsm3bNuX024MHD2Tz5s3yxx9/5Fh/48aN8u677+Z47vJbSkqKlC1bVgmqp06dEl9fXylVqpR4eHjI8uXLNUJ6r1695N133zXI6bS0tDQZMGCAlClTRuMLy++//y6RkZFy5coVjfXv378vjRs3lp9//rmAK/3fcatdu3YybNgwERH5559/ZN68efLtt9/KiRMnlHUzMjLk4sWLUrduXdmxY0eB15obDBP5JHvwV2hoqEZ7WlqaXL9+Xc6ePau0ZWZmSlpamgwfPlycnZ3lzJkzBV1unjz/gda6dWtxc3OTLVu2yJMnT5T248ePS+3atSUmJkZEnp3y6dSpk0yfPl0vA+BOnjwpzs7OMmjQINm5c2eOcJbt3Llz4uLiIomJiRIRESG2trYao6VNUVpamsYB++zZsxIWFibOzs4aH5BZWVly/fp1qVu3rixbtqzA63y+xoiICKlVq5aUKVNGGjRoIMeOHdN4vYiIHDhwQObPny/29vYyduzYXO3j7Nmz0r59e6lZs6aYm5uLSqWSJk2ayObNm1/4DTkjI0OGDx8uHh4ecuHCBd3+wDwYO3aseHt7S5EiRcTOzk7KlCkjYWFhkpiYKCI5w2J8fLzUrVtX+vTpU+C1Pn78WDw8PGTbtm1KW3p6uly4cEFjXEtWVpZkZmZKjx495MsvvzRY7+rp06clNDRU6tSpIz/++KPS/qJBjMOHD5cKFSoUeED793ErN8/ViBEjpGbNmkY1GPd5DBP5JD09XZo2baox+Gvz5s3Sr18/cXBwkLJly0rLli1F5NnBs3379lKqVCmT63p//kDt7e0tNWvWlC1btig9FEeOHBFLS0ul23Ho0KHi6+ur8+A6kWcH2CpVqsiAAQM02l/U05CQkCB16tSRjz76SKysrEw6SERHR8uwYcOkatWq0qhRI/n444/l9u3bIvKsqzwsLEwqV66scSANCAiQd95556VhqyCMGDFCSpQoIb/88otcuXJFqlatKvXq1ZPNmze/sCduwYIFUrJkSY0elhc5fvy4uLq6SmhoqKxfv17Onj0rf/zxh1SsWFFKlSolK1as0Pi7169fL3369JHixYsb5P02ePBgcXZ2ltWrV8vu3bvlxIkT0rFjR3FxcRE/Pz+N98a1a9ckMjJSateuLR988IHSXpC9aQkJCVKmTBnlC8HLqNVqGTZsmLi5uRn8C1FsbKx8/vnnUqdOHZkzZ06O5fv375dvvvlGnJycCvw18LLj1vNh5/n/P3r0qAwcOFAcHBzk+PHjBVanthgm8klycrJ4eXlJr1695OzZszJhwgSpXLmyBAUFyYwZM2ThwoVSoUIF6devn1y6dElGjx5tdJeqvU72Ae3XX3+VoKAgCQgIEHNzcylTpoxs3bpV0tPT5fHjx9K3b1+xsbGR6tWri6Ojo97O+a5du1bq168v58+ff22X6qlTp0SlUomtra1Bzjnry+LFi6VChQrSuXNnGThwoHTp0kVKlCghHh4eylUR58+fV055zJ49W4KCgqRSpUrKtx9DBIo9e/ZIvXr1JCoqSkSeXQbp4OAgFStWlBIlSsiWLVuUYJr9ujpx4oTUrFlTY4zDv508eVLs7Oxk+PDhOZY9ePBAqlSpIl5eXnL+/HkReRboO3bsKJ06dZLY2Fh9/5mvtXnzZilXrpwyuDRbRkaGjBo1Stzd3aVv376iVqvl4cOHMmrUKPH19dW4nLwgTh/cvHlT+eC6e/euFCtW7JUfZLNnz5aQkBApUaJEgX84X7x4UTZs2KBxWkDkWQ9Fnz59pFatWhqnPBYvXiy1a9eWli1bvjYg5QdtjluzZ8+WNm3aSKNGjXL8fcaGYSIfRUVFiYWFhXh6ekrhwoVlzpw5SpdqWlqatGrVSpkHwJgu19PGvn37xM7OThYuXChnzpyR06dPy/vvvy8lSpRQ5pVISEiQv/76S3766SdlwKk+DBw4UMqWLfvCZdkfSKmpqXL9+nV5+vSphIaGGvwbky7mzJkjVlZWsnz5cmWQaXp6uhw8eFDeffddKVWqlPLBe/bsWRkwYICoVCqpXLmyEiQKaizOv1/PMTExygH977//FhcXF1mwYIGIiHh5eUndunXlzz//1Ag6U6dOFTMzs5d26964cUNUKpV8/PHHSlv2v3v2di5fviwODg7KYGCRZ2MVsgc1FrRZs2bJ+++/L2q1Wqnx+YHLn3zyiZQrV04Z33Hx4kWNiZYK4jiRlpYm9erVE39/fzl58qQkJSWJs7PzSz/Mrl27JkOHDpXg4GCN07cFIXtQokqlkiJFikhgYKAMGjRIYmNj5cmTJ/LgwQP5/PPPxdvbW2bNmqX83u7du+XOnTsFWmu23B637t+/L+fPn5eNGzca7amN5zFM5LO4uDg5fPiw3L17V6M9MzNTOnbsKMOHD5esrCyTHAQoIvLzzz9LrVq1clxy2KxZM/H09JRt27bp9dzp+fPnlWutIyIilAPvy75tR0REyIgRI0RETHoysKVLl4pKpVK+2f/bsWPHpGLFivLee+8pr6WzZ8/KzJkzNT6sCtqKFSuU1358fLxkZGRImzZtlEmEnjx5Iv/5z3/E2tpa2rRpo/xeWlqaLF++/LW9SDVr1pRq1arJ7t27c7wGsv/ezp07i7+/v1H8+/ft21e8vLxytGfXfvHiRSlcuPALZ8QtiGPEuXPn5PTp07J//36pVKmSfPjhh7Js2TKpUKGCHDlyRK5evSqXL1+WK1euyJUrVyQuLk727t0r58+fzzH2paB07dpVnJyc5KuvvpJPPvlEfHx8pFixYlKlShUZP368jBw5Unr16vXSUx4FQdvj1tixY2XkyJEiYjqXrjNMGIBarZbhw4eLu7u70v1qqmbMmCElSpRQHmcfsA8ePChmZmbi5uamjKzX9U1x7Ngxsbe3V75hbN26VVQqlcydO1dZ5/l9PHr0SEJCQpRR6KYq+5IxlUqlfJv69zdUtVotY8eOFVdXV7lx40aObRR0kMjKypKUlBSxsLCQgQMHKu0PHjyQ2rVrK/+GGRkZEhwcLNeuXcvzLH8NGjSQ8uXLy+7du194JU9gYKD4+/vr+ifl2eHDh5UesQkTJkjhwoWVHsp//82xsbFSpEgR2bNnT4HXeezYMbGzs5OZM2eKyLNv72XLlpUGDRqISqUSCwsLcXFxkWLFiomNjY3Y29tLsWLFpESJEq8d16JvV65ckR9//FE5NdyxY0fx9vaWJUuWiMizHtNZs2ZJo0aNpFatWkrvRdWqVQu8Vyqvx63snjtTwTBRwH755Rf5+uuvxdXV1aQGW76sezUhIUHc3Nzkyy+/1Gg/evSodOnSRQICAvQyFuT48eNiZ2cnQ4YMUdrUarWEhISISqWSyMhIjfWzsrJkxIgRUqVKlRyXg5miK1euiL+/v7i5ueX4IMr+74EDB0SlUsmJEycM8m3mRa+RFStWSNmyZTW+aWfP9jp+/Hh57733pEaNGsrvvm48R1xcnCxYsEDmzZuncflndqDYs2ePxvNy+/ZtCQgIUA7kBf28jBs3TurXry/fffedPH78WOLj48XNzU3atm2rsV72333w4EGpV6+ejBgxQrZt21Zgs0n++/2V/TwdOnRIypcvL7Vq1ZIpU6bI8ePH5fTp07Jnzx45dOiQnDlzRq5du1YgNWZ7fu6eP//8U2kPCgqSypUry/Lly5UvNQ8ePJDExESZN2+eDBo0qMDvzfE2HbcYJgrQ2bNn5f3335f27dsbZPBXXmSfvsg+QB87dkx+++032b59u9KrMnfuXKlYsaJ8/vnnkpaWJnfu3JFhw4ZJly5d9PKN+MSJE2JnZydDhw7VaN++fbtERkZKYGCgmJmZyRdffCHLly+X2bNny6effipFixY16cGW/3bt2jXx8/MTV1dX5bl//kN4+vTp8t577xlssqBsixYtkhMnTkhqaqo8fPhQunbtKr1791auOHn48KG0atVKmjVrJoGBgcppsNeNBzhx4oR4enpKw4YNxcnJScqXL69xMH7nnXdy9FAMHTpUqlatapAD86BBg8TNzU1WrlypjBVSq9Uyc+ZMsbKykg8++EDjG332KH97e3upUqWKfP755y/sZdK3l72/Nm3aJMnJyXLw4EEpV66cdO3a1eDvp9fN3dO5c2epVKmSLFmyRFJTUzWWFfS4tLftuMUwUcASEhI0Zmg0ZjNnzpRBgwYpB7w//vhDbG1tpUqVKuLk5CR16tRRZmNbsGCBuLu7i7Ozs1SqVEmKFSsmR44c0bmGuLg4cXZ2lg8//FCjfcyYMVKmTBm5dOmSnDt3TsaOHSvOzs7i6OgoVapUkU6dOhn9HQJf5WUHvuvXr4uvr6+4ublpnCJLSUmRgICAHDeRK2gHDx4UlUolNWrUkEGDBsmFCxfkn3/+keLFi+eYYvn5cTavC53ZB+YhQ4bIo0ePZNu2bVKyZEkJCAjQeD9l91AcO3ZMRowYIXZ2dgY5MP/6669Srlw5ZdbI5127dk2mTZsm5cuXF0dHR2natKm0adNGqlWrJu3btxeRZ89HQXz4vez9FRERIaVKlVJmity9e7eUK1dOPv744xxXohSUV83dc+XKFWWq/N69e0ulSpVk2bJlBrsR2dt43GKYoJcaNWqUuLq6yujRo+XQoUPSuHFj+emnn+TRo0dy4MABCQsLEzc3N+XbYVJSkixZskRWr16tt+7ZK1euSIMGDaRt27bKeeSJEyeKs7Oz/PXXXxrr3r59W65evSqJiYkGGwymD89/iOzatUs2bdokO3bsUL7B37p1S3x9fcXV1VX5xu3v7y/16tVTPpQLqjv/3/u5c+eO/Oc//5Hq1avLlClTxN3dXaKioiQ4OFjKlCnzwhH0r6s1+8DcqVMnjfYGDRpIpUqVJCkpSSOMNG3aVFQqlRQuXFgvgTYvRo4cKZ06dVJehxkZGbJ27Vrp0aOHtG7dWjp37iznzp2TUaNGSceOHWXgwIEaszAW1LfoV72/su9XkX0KZs+ePVKkSBHp3r27PH36tEDqe97r5u4pVaqUdOjQQUSezcL5/E0TC9rbeNximKBXmjJlipQuXVrCw8OlTZs2Gh8GV69ela+//loaNWqUr93I58+fl//85z/Stm1b6dWr10tvUW2qif5lBg4cKCVKlJAKFSqImZmZtG/fXhkRfvPmTWnVqpWULFlSGjVqpHH5pyHmkdi9e7cyi+Dx48fFxcVFfv31V1m3bp14eXlJx44dRaVSSf/+/bX+IHrRgXnChAmiUqmkYcOG0qZNGwkJCZFp06bJo0ePJCMjQ0JCQgwyJin7yqxPPvlE/Pz85MmTJ/L06VPp0qWLNGnSRN59910JCQmR0qVL57gxW7aC7o5/1fvrecnJyXLx4kWDzBiavf/Xzd3j6empzP4aHBys10vRtfW2HbcYJuiFnj+gTZw4Uezt7cXe3j7HxDU7d+4UBweHfO/6PHfunPj6+oqtra1yl8fnL6kdPny4lCpVSh48eGAyl1K9yvz586V48eJy4MABuXv3rhw5ckQaNWokAQEBynN97do18fX1lRo1ahT4PBLP2717t9SpU0caNWok0dHRIvJsPoXg4GB5+vSpnDp1SsaNGyc2NjbSpk2bPP37PH9g/uyzz8TFxUVWrlwp165dkzVr1si4cePE1dVVihcvLh9//LHBXwP79u0TlUol1atXF3t7e2nQoIH89ttvyr/PxIkTpXz58jnuF2Ior3t/DRs2TEqUKGHwu87mZu6e5+cdMbS36bjFMEG58uOPP4qjo6P07dtXoxfi7t27UqlSpQK55fDFixelVatW4u/vL7t27VLaR4wYITY2NiY9Rfa/ffXVV0q3fnawi4mJkcqVK2vcnyE+Pl5ZbqgJqUSehco+ffqIubm5jBkzRmbMmCFdu3aVdevWicizy90uX76s9Jrk5cCZfWC2sbF54Q3L7t27JytXrjSay62PHz8uU6dOlTlz5khGRobG3zxnzhxp1qyZUY2fet3760XjPwzhVXP3dOrUyejm7nlbjlsME6Qh+w146dIliY2N1TitMWnSJClZsqR88cUXcuDAAbly5YoMHjxYihUrVmDXmWd/Q/Xz85OjR4/Kd99990a9IbM/bIODg5VJnLJvBCfy7NLiIkWK5BjJXlBd48/v59ixY/L333/L6dOnlboXL14sTZo0kcDAQClatKjUqVNHuXlVNl1Owzx/YN69e7fSbqibSuXFzZs3pU6dOhrzbxgLU31/GfvcPab6vGqDYYJyWLVqlZQuXVqKFCkiAQEBsnDhQmXZpEmTpHDhwmJnZydt2rSRli1bFvho+fPnz8sHH3wgxYsXF0tLS5N+Q74sBPz++++iUqlk/fr1Gu0rVqyQevXqGWQ66Oe/6Q0ZMkSqV68u7u7u0rx5c/H19VUGj504cUJmzJghNWrUEJVKJZMmTdJrHc8fmA0xuVNe3bhxQ/bs2SM1a9bUmGfCWL5BZzO195epzN1jas+rthgmSEN8fLzUqlVLFixYIBs2bJAPP/xQ3n33XZk6daqyzqxZs5QPiXv37hmkzrNnz0rbtm2VS9dM0fMfImvXrpXFixfL4cOHlQl3vvzyS7G1tZVff/1Vbt26pVwp0bp1a4N+AE2bNk2cnZ1l7969kpWVJYMHDxaVSqXci0XkWU9BXFyc9O3bN19Ov2QfmBs1amSwSxW18fDhQ/nggw+kQYMG0rt3b6XdWO/JYyrvL1Obu8dUnte8YJh4y/373OK9e/ekc+fOyvXZ165dk969e0ujRo00AsWUKVNeeTfHgmBKXduv8u2334qLi4u4urpKjRo1ZPDgwZKamiqPHj2Sb7/9VqysrMTT01MqVaokdevWzfUkT/qWlZUljx8/li5duihTAW/YsEHs7e2VKctTU1NfeBojPwLFmTNnpGPHjgU+A2NexcbGyrZt25THxhokspnK+8uU5u4RMZ3nVVsqERHQW0tEoFKpsGnTJixcuBC2tra4efMmoqKilHXi4uIwfvx4xMbGws/PD8OHDzdgxaYvKysLZmZmEBHcvHkT3bt3x/fffw8PDw/MnDkTmzdvRoMGDTBhwgQULlwY//zzD27dugVzc3P4+/vD3NwcGRkZsLCwKLBan+fj44PevXujUKFC6Ny5M6ZMmYLPP/8cGRkZWLBgAVxcXBAUFJTvtQFAWloarKysCmRf+pT9viN6U5i9fhV6k6lUKkRHR+ODDz6AlZUVYmNjsXv3bo3AULp0aQwbNgylS5fGrl27kJiYaMCKTdvzH873799Heno67O3tUb58eRQrVgzDhw9Hu3btcOjQIQwZMgRJSUlo2LAhAgMD8cEHH8Dc3ByZmZkFEiSOHz+ufOBFRERg2bJlAAAPDw/MmDEDn376KSZPnozPP/8cAHDnzh2sXbsW9+7dy/fasplikADAIEFvHPZMvOXOnz+P2NhYXL9+HX379kV8fDzmzZuHVatWoVOnThg9erSy7o0bN2BhYQE3NzfDFfyGGDFiBH799VdYW1sDAGJiYpSQkZGRge+//x4bNmyAp6cn5s2bh0KFChVofdeuXUPZsmUxZMgQPHz4EMuWLcO+fftQpUoVXLhwAU2bNoW7uzs2btyIokWLIiUlBcHBwUhOTsauXbtgbm5eoPUSkWHl/9cbMlrXr19H48aNoVar8d133wEASpYsiT59+kClUmHFihUwNzfHiBEjAAClSpUyZLkm7fkeiTVr1mDOnDmYPHkyDh48iI0bN6JDhw5Yvnw5ChUqBAsLCwwcOBAPHz7EvXv3YGtrW+D1enp6YuvWrQgICIC1tTV27NiBKlWqICMjAxUrVsQff/yBgIAAtG7dGk+ePIGzszMeP36MAwcOKL0nDBREbw/2TLzFHjx4gKVLl2Ly5Mlo1aoVFi1apCy7desWFixYgFmzZiEsLAyDBw82YKVvjmXLluHhw4ewtbVF9+7dkZmZiaVLl2Lu3LkoVaoUli5dCjs7OwDPAohKpYJKpXrh2IX8lJWVhaioKPj5+cHMzAxDhgzBiBEjYG1trZzvv3HjBjZv3oy7d++iYsWKaN++fYGO5yAi48Ew8Za7d+8eVqxYgSFDhuDzzz/HlClTlGXx8fFYtmwZOnbsiPLlyxuwyjfD5cuX4evriytXrmDWrFn44osvADwbRLh8+XLMnTsXpUuXxqJFizROaxTUYL0X7ef+/fs4ePAgAgMD0a9fP4wfPx6WlpYvrYc9EkRvJ4aJt0T2B8Xp06cRFxeHrKws+Pj4wNraGvfv38eKFSswYsQI9OjRQyNQ8MNBf9LS0rBt2zaMHDkSIoLDhw8rvQ1paWn49ddfMWbMGHTp0gXjx48v0Nqe7/m4d+8enj59qnFa6/fff0fXrl0xYMAAjBo1CjY2NujVqxf8/PzQsWPHAq2ViIwPw8RbIDtIrFmzBgMGDICFhQUKFSoElUqFbdu2wcnJSQkUY8eORYcOHTB79mxDl23SXnZaQq1WY+fOnejXrx+KFy+O7du3K2EtLS0Nf//9N/z8/Ao0wD1f64QJE7BmzRokJiaiXLly+PHHH1G+fHlYWFhg5cqV6NKlC3x8fJCYmIiUlBScOnWKpzSIiGHibREVFYWgoCBMmTIFPXv2VM6HV65cGdu2bUOpUqWQmJiIRYsWYfbs2di/fz9cXFx4CVsePP/hHBkZidjYWJibm6NNmzaoX78+1Go1duzYgUGDBsHJyQnbt2/PETwM0SM0YsQILFy4EBEREWjUqBECAgLg6emJiIgING7cGObm5tizZw/mzZsHFxcXTJo0CZaWluy9IiKGibfBw4cPER4eDnd3dwwdOhQ3b96Et7c3mjRpggsXLuDBgweIjo6Gu7s7EhMToVKpULRoUUOXbfIGDx6M3377DdWqVYOdnR22bNmCNWvWwMfHB2lpadi+fTsGDx6MjIwMnDp1yqDBbdeuXfj6668xffp0vP/++9i+fTvatWsHBwcH2NjY4Oeff8Y777wDa2trjYmiONiSiACA02m/JdavXy+HDx+WxMREqVu3rnIb6xUrVohKpRJXV9ccd6KkvJs7d654eHgot22OjIwUlUolFhYWsnr1ahF5dqfDNWvWSNeuXXW6k2ZePb/PQ4cOyezZs0VEZNu2beLs7Cw///yzpKenS+nSpaVp06aydetWg9RJRMaPXyneENld60+fPoWNjY3SLv8/XuKDDz4AAPz111+wsbHBkCFDAADOzs4ICAiAubk5Hj16ZJDa3zQPHz7EpUuXMGrUKNSvXx8bNmzA559/jqlTpyI2NhadO3fGunXr4Ofnh4CAALRr1w5AwZ3a2LVrFw4ePIj79+8jNDQUHh4eqF+/Ptzd3ZGWloapU6eid+/eCAkJwePHj1GuXDlER0dj/vz58PX1zff6iMj0cDrtN4SZmRni4+PRrVs37NixQ2n/d9f5tWvXcPz4cWWkflRUFFxcXLBixQpUrFixQGt+U8i/zhQWLlwYXbt2RfPmzXHhwgX0798f48aNQ1hYGAIDA5Geng5/f3/s3LkTlpaWyu8VRJBYvHgxPvvsM9y5cwc1a9aEh4eHsszd3R0pKSmIj49HhQoVAACWlpaoUKECrl69it9++y3f6yMi08SeiTeIWq3GjRs3MHXqVFhZWaFx48Y51mnXrh1mzZqFkiVLombNmti7dy8OHjyoTOtM2nl+sOXjx4+VCadq1KgBAPjzzz/h5OSETz75BABQpEgR9O7dG/Xq1UOTJk0KtNZly5YhNDQUv/zyizKzJfBs4OU777yDDz74AM7OzrCwsMCcOXOQnJyMdevWITExEaVKlYKZmRkHWxLRC7Fn4g1Srlw5LFmyBJmZmYiIiMDevXuVZVlZWQAANzc3rF27Fj4+PmjevDmOHDmifPCRdkRECRLTpk3DRx99hODgYGzatEl5vh89eoSDBw/i6tWruHfvHiZPnoz09HT06tULFhYWyMjIKJBaz507h6lTp2LixIno0KGDEiQ6duyIKVOm4Ntvv8Xq1asBABs3bkRWVhZ+//132Nra4tChQzAzM0NWVhaDBBG9mIHHbFA+OH/+vPznP/8RPz8/2bNnj9KemZkp6enpMmjQIPnkk08kOTnZgFWatqysLOX/v//+e3FwcJAhQ4ZItWrV5N1335XJkydLRkaGZGRkSIcOHUSlUknFihWlevXqkpaWlmMb+W3z5s3i6ekpJ0+eVNqGDRsm9erVk3Xr1kmPHj2kRo0asmrVKmV5UlKSUmN6enqB1UpEpoeXhr6hLly4gK+//hoighEjRqBx48ZIS0vDgAEDMGvWLBw9ehS1a9c2dJkm7+jRo5g/fz6CgoLg4+ODx48fY8CAATh+/Djat2+PgQMHQqVSYcOGDcjIyEDbtm0L9P4V6enpsLS0xPjx4zF37lzExcUpy44fP46SJUvCxcUFp06dQkREBPbt24fo6GiUK1dOWU8KaDpvIjJdPM3xhqpYsSJmzpwJlUqFiIgI7Ny5U5mU6MiRIwwSefR89l61ahVCQkKwY8cOuLu7AwDs7OwwceJE1K5dG2vWrMHUqVORmZmJNm3aKDfCyszMzPcgERYWBgDKAM+aNWvi1q1bWLt2rbJO7dq14eLiAgCoXr066tevj9q1a6NEiRIa22KQIKLXYZh4g2UHCktLS3To0AHTp0/Hnj17UKdOHUOXZrKyP1gfPXqExo0bo1y5crh58yb++usvZZ0iRYpg0qRJqFevHubOnYtff/1VYxv5Pe5g7969uHz5ssZ4jFKlSqFs2bKYO3cuTpw4keN3njx5gt27d6NcuXIGueU5EZk4g55koQJx9uxZadu2rZw6dcrQpbwRpk+fLl988YWIiNy4cUM6dOggjRs3liVLlmisl5iYKFOmTCnwiZ7S09OVsQ7Lli1T2mfPni0qlUoCAwNl586dIiLy9OlTOX/+vPj6+krNmjWVsREFOZ6DiEwfx0y8JbLPnZPu5syZg/79++PYsWOoXLkybty4gb59+yIxMRE9e/ZEt27dcvyOIS6pvHjxIt59911UqVIF0dHRAIAZM2YgIiICZmZmaNKkCe7fv4+MjAyICKKjo3mvDSLKE4YJoleQ5wYfZv9/QkICunbtCj8/P/Tv31+ZMKxv375ISkpCp06d8MUXXxi0VuDZvCNbtmxBeHg4XF1dsX37dgDAtm3bsHfvXuzevRvly5dHgwYN0KNHjwIdGEpEbxaGCaKXeH5CqidPnmiMJejbty/+/vtvnDlzRmmLj49Hly5dULNmTfz3v/8t0IGL/77lefbNuDIyMrB161b069cPHh4eiIqKeuk22CNBRHnFMEH0GosXL8aSJUswduxYVKtWDcWKFcPjx49Rt25dhISEYPDgwcqH+d27d+Hk5AQzMzODXFI5efJkHDhwALdu3UK3bt0QEBCA0qVLY9OmTfjmm2/g6emJbdu2Afhf4DBEnUT0ZuHVHET/cvv2bcTExOCXX37B6dOncf36dVSrVg0ffPABevTogRkzZkBE0Lx5c5w/fx4AlKmmXVxclNkiC+IDOnumTQAYPXo0vvvuO5QuXRrVqlXD8OHDER4ejiNHjsDf3x/Tp0/HzZs3UbduXQBQbiPOIEFEumLPBNFzVq9ejYULF+Lo0aNITU2FtbU1mjRpgu+//x5qtRq//fYbfvzxR/j6+uLhw4fYsmULNm3aBD8/P4PWHRcXh59++gm+vr5o0aIFAGDr1q0YOnQoatSogVmzZsHMzAwbNmzA6tWrsWzZMo3TIkREuuDRhOj/zZ8/H5999hlatGiBZcuW4fr16+jbty/OnDmDgIAApKenIyIiArGxsfDy8oKjoyMAYOnSpUhNTc1x99CCsm7dOpQpUwY///yzRnurVq0wbtw4REZG4p9//oGNjQ3atWuHyMhIpfeEiEgvCvpaVCJjNG/ePLGyspI//vgjx7Lff/9datSoIU2bNpWLFy+KyLP7nGRmZsqUKVOkePHicuXKlQKrNTMzU+O/8fHxEhoaKiqVShYvXiwimvfSqFatmkyZMqXA6iOitw97Juitt3PnTvTp0wfDhg1Dhw4dICIQEWUGyU6dOqFPnz44cuSIMnuk/P8dQwcOHAhPT08sXLiwQGr97bff8Nlnn+H8+fN48uQJAMDd3R3Dhw/Hp59+ii+++ALbt29XLu9MSUnBkydPULhw4QKpj4jeTrygnN56JUuWRJMmTXD06FHs3r0bTZs2BQBYWFgoV2mEhoZi7ty52L59Ozp06KCcJjAzM4OtrW2BzM2QkpKC4cOHIyUlBYcPH0bDhg3RpEkTdO/eHW5ubvjpp5+Qnp6OgIAA9OnTB+7u7ti9ezfs7OzQs2fPfK+PiN5e7Jmgt17FihWxcOFCqNVqjB8/Hnv27FGWZV/pkP0N39XVVWk3MzPDoUOHsHv3brRv3z7f6yxUqBA+/PBDREREYPHixfDy8kJYWBg+/vhjTJo0CZaWlpg5cya++OILzJw5E4cOHcKnn36Kw4cPw8LCQuNeHURE+sQwQQTNu6yOGzcOe/fu1Vh++fJllCpVCo0aNQLwv7uHNmjQADdv3kTNmjXzvUZzc3M0bdoUgwYNgoWFBQYOHIhbt26hQoUKGDp0KLy9vfHzzz/D19cX/fr1w+bNm1GqVClYW1tDrVZzZksiyje8NJToORcuXMDXX38NEcGwYcPQtGlTZGRkIDAwEGZmZli7dq3BL6kMDQ0FAMyaNQsAUK1aNVSqVAnly5fH6dOnsWXLFkycOBGnTp3Chg0bsGbNGrz//vsGrJiI3nQME0T/kh0ozMzMMHToUEybNg1nz57F8ePHYWlpmWPq6oK2cOFCLFq0COvXr0fLli1hZ2eHjRs3wsHBATdu3MC+ffvQoUMHqNVqfPLJJzh8+DAuXLjAW4sTUb5hmCB6gQsXLiAsLAxbt25FuXLlEBMTA0tLS6O5EVbDhg1x+PBhvPfee1i9ejWKFSuWY52MjAwkJydDrVbD3d3dAFUS0duCYYLoJc6ePYvZs2dj2rRpygBGQwcJ+f/7aCxbtgzfffcdFi9ejHr16vH+GkRkUByASfQSXl5emDlzptEECeB/V5c0b94c9+/fV27axSBBRIbEMEGUC8YQJJ5XsmRJhIeH4/vvv0dsbKyhyyGit5xxHSGJKNdat26Nw4cPw8vLy9ClENFbjmMmiExY9liJzMxMmJubG7ocInpLMUwQERGRTjhmgoiIiHTCMEFEREQ6YZggIiIinTBMEBERkU4YJoiIiEgnDBNERESkE4YJIiIi0gnDBBEREemEYYKIiIh0wjBBREREOvk/oYhR0BvN1IEAAAAASUVORK5CYII=",
      "text/plain": [
       "<Figure size 600x400 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "def fig_3b(architectures):\n",
    "    \"\"\"Most popular AWS services used by architectures\"\"\"\n",
    "    # query\n",
    "    occurence = Counter()\n",
    "    for arch in architectures:\n",
    "        occurence.update(\n",
    "            x for x in set(arch.get_all_nodes_services()) if svc_info[x][\"is_aws\"]\n",
    "        )\n",
    "\n",
    "    occurence = Counter(normalize_dict(occurence, len(architectures)))\n",
    "\n",
    "    # plotting\n",
    "    fig, ax = plt.subplots(figsize=(6, 4))\n",
    "    x = np.arange(10)\n",
    "    x_names, y = zip(*occurence.most_common(10))\n",
    "    colors = [CAPABILITY_COLORS[svc_info[x][\"capability\"]] for x in x_names]\n",
    "\n",
    "    ax.bar(x, y, color=colors)\n",
    "    ax.set_xticks(x)\n",
    "    ax.set_xticklabels(x_names, rotation=45)\n",
    "    ax.set_ylabel(\"Percentage of architectures (%)\")\n",
    "    ax.grid(axis=\"y\")\n",
    "    plt.title(\"Fig 3b\")\n",
    "\n",
    "    cleanup()\n",
    "\n",
    "\n",
    "fig_3b(architectures)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "id": "1edde1c1",
   "metadata": {
    "jupyter": {
     "source_hidden": true
    }
   },
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 600x400 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "def fig_3c(architectures):\n",
    "    \"\"\"What percentage of architectures use each capability?\"\"\"\n",
    "    # query\n",
    "    occurence = Counter()\n",
    "    for arch in architectures:\n",
    "        services = (x for x in arch.get_all_nodes_services())\n",
    "        capabilities = set(\n",
    "            svc_info[x][\"capability\"] for x in services\n",
    "        )  # use set to avoid double counting\n",
    "        occurence.update(capabilities)\n",
    "\n",
    "    occurence = Counter(normalize_dict(occurence, len(architectures)))\n",
    "\n",
    "    # plotting\n",
    "    fig, ax = plt.subplots(figsize=(6, 4))\n",
    "\n",
    "    x = np.arange(len(occurence))\n",
    "    x_names, y = zip(*occurence.most_common())\n",
    "    colors = [CAPABILITY_COLORS[x] for x in x_names]\n",
    "\n",
    "    ax.bar(x, y, color=colors)\n",
    "    ax.set_xticks(x)\n",
    "    ax.set_xticklabels(x_names, rotation=45, ha=\"right\")\n",
    "    ax.set_ylabel(\"Percentage of architectures (%)\")\n",
    "    ax.grid(axis=\"y\")\n",
    "\n",
    "    plt.title(\"Fig 3c\")\n",
    "\n",
    "    cleanup()\n",
    "\n",
    "\n",
    "fig_3c(architectures)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "id": "0de8ad53-db9e-41e9-b5d6-13513bef7d35",
   "metadata": {
    "jupyter": {
     "source_hidden": true
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Number of Partner nodes: 4\n",
      "Number of OnPremDC nodes: 5\n"
     ]
    }
   ],
   "source": [
    "def count_number_of_partner_and_onpremdc(architectures):\n",
    "    total_partner_nodes, total_onpremdc_nodes = 0, 0\n",
    "    for arch in architectures:\n",
    "        total_partner_nodes += len(\n",
    "            arch.select_nodes(\n",
    "                lambda x: svc_info[x[\"service\"]][\"capability\"] == \"Partner\"\n",
    "            )\n",
    "        )\n",
    "        total_onpremdc_nodes += len(\n",
    "            arch.select_nodes(\n",
    "                lambda x: svc_info[x[\"service\"]][\"capability\"] == \"OnPremDC\"\n",
    "            )\n",
    "        )\n",
    "    print(\"Number of Partner nodes:\", total_partner_nodes)\n",
    "    print(\"Number of OnPremDC nodes:\", total_onpremdc_nodes)\n",
    "\n",
    "\n",
    "count_number_of_partner_and_onpremdc(architectures)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "7ca7141c-facb-4cfc-8e77-fe18a734dbc8",
   "metadata": {},
   "source": [
    "#### There are fewer than 10 uses of partner or On-Premise-Datacenter nodes across the entire dataset, so we do not include them in our analysis or graphs in the paper."
   ]
  },
  {
   "cell_type": "markdown",
   "id": "11f2e6c2-6916-4438-9d2e-a54584f7519f",
   "metadata": {},
   "source": [
    "# Storage Service Analysis"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "id": "2c378c3a",
   "metadata": {
    "jupyter": {
     "source_hidden": true
    }
   },
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 1200x400 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "def fig_4(architectures):\n",
    "    \"\"\"Usage of different storage services / schemas\"\"\"\n",
    "    # query\n",
    "    service_occurence = Counter()\n",
    "    schema_occurence = Counter()\n",
    "\n",
    "    for arch in architectures:\n",
    "        storage_svcs = set(filter(is_storage, arch.get_all_nodes_services()))\n",
    "        storage_schemas = set(svc_info[x][\"schema\"] for x in storage_svcs)\n",
    "        # sets to avoid double counting\n",
    "\n",
    "        service_occurence.update(storage_svcs)\n",
    "        schema_occurence.update(storage_schemas)\n",
    "\n",
    "    service_occurence = Counter(normalize_dict(service_occurence, len(architectures)))\n",
    "    schema_occurence = Counter(normalize_dict(schema_occurence, len(architectures)))\n",
    "\n",
    "    # plotting\n",
    "    _, (ax1, ax2) = plt.subplots(1, 2, figsize=(12, 4))\n",
    "\n",
    "    y_names, x = zip(*service_occurence.most_common())\n",
    "    _schemas = [svc_info[t][\"schema\"] for t in y_names]\n",
    "    colors = [schema_colors[t] for t in _schemas]\n",
    "\n",
    "    # group by schema\n",
    "    x, y_names, _schemas, colors = zip(\n",
    "        *sorted(\n",
    "            zip(x, y_names, _schemas, colors),\n",
    "            key=lambda t: (list(schema_colors.keys()).index(t[2]), t[0]),\n",
    "        )\n",
    "    )\n",
    "\n",
    "    y = np.arange(len(service_occurence))\n",
    "    ax1.barh(y, x, color=colors)\n",
    "    ax1.set_yticks(y)\n",
    "    ax1.set_yticklabels(y_names)\n",
    "    ax1.set_xlabel(\"Percentage of architectures (%)\")\n",
    "    ax1.set_title(\"Storage schemas\")\n",
    "    ax1.grid(axis=\"x\")\n",
    "\n",
    "    y = np.arange(len(schema_occurence))\n",
    "    y_names = list(schema_colors.keys())\n",
    "    x = [schema_occurence[t] for t in y_names]\n",
    "    colors = [schema_colors[t] for t in y_names]\n",
    "\n",
    "    ax2.barh(y, x, color=colors)\n",
    "    ax2.set_yticks(y)\n",
    "    ax2.set_yticklabels(y_names)\n",
    "    ax2.set_xlabel(\"Percentage of architectures (%)\")\n",
    "    ax2.set_title(\"Storage schemas\")\n",
    "    ax2.grid(axis=\"x\")\n",
    "\n",
    "    plt.suptitle(\"Fig 4\")\n",
    "\n",
    "    cleanup(ax1)\n",
    "    cleanup(ax2)\n",
    "\n",
    "\n",
    "fig_4(architectures)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "id": "0033a791",
   "metadata": {
    "jupyter": {
     "source_hidden": true
    }
   },
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 700x400 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "def fig_5(architectures):\n",
    "    \"\"\"Usage of storage schemas across functional goals\"\"\"\n",
    "    # query\n",
    "    functional_goals_to_consider = (\n",
    "        \"compute_intensive\",\n",
    "        \"interactive\",\n",
    "        \"data_ingestion\",\n",
    "    )\n",
    "    df = pd.DataFrame(\n",
    "        0, index=schema_colors.keys(), columns=functional_goals_to_consider\n",
    "    )\n",
    "\n",
    "    for arch in architectures:\n",
    "        storage_svcs = filter(is_storage, arch.get_all_nodes_services())\n",
    "        storage_schemas = set(svc_info[x][\"schema\"] for x in storage_svcs)\n",
    "\n",
    "        for schema in storage_schemas:\n",
    "            for fg in arch.get_functional_goals():\n",
    "                if fg in functional_goals_to_consider:\n",
    "                    df.loc[schema, fg] += 1\n",
    "\n",
    "    # normalize based on total architectures per functional goal\n",
    "    architectures_per_fg = {t: 0 for t in functional_goals_to_consider}\n",
    "    for arch in architectures:\n",
    "        for fg in arch.get_functional_goals():\n",
    "            if fg in functional_goals_to_consider:\n",
    "                architectures_per_fg[fg] += 1\n",
    "\n",
    "    for fg in functional_goals_to_consider:\n",
    "        df[fg] = df[fg] / architectures_per_fg[fg] * 100\n",
    "\n",
    "    # plotting\n",
    "    fig, ax = plt.subplots(figsize=(7, 4))\n",
    "    width = 0.15\n",
    "    y = np.arange(len(df.columns))\n",
    "\n",
    "    for i, schema in enumerate(df.index):\n",
    "        ax.barh(\n",
    "            y + width * i,\n",
    "            df.loc[schema],\n",
    "            color=schema_colors[schema],\n",
    "            label=schema,\n",
    "            height=width,\n",
    "        )\n",
    "\n",
    "    ax.set_yticks(y + 0.25)\n",
    "    ax.set_yticklabels(df.columns)\n",
    "    ax.set_xlabel(\"Percentage of architectures (%)\")\n",
    "    ax.set_title(\"Storage schemas usage across functional goals\")\n",
    "    ax.legend(loc=\"upper left\", bbox_to_anchor=(1, 1), handlelength=0.5)\n",
    "    ax.grid(axis=\"x\")\n",
    "\n",
    "    plt.suptitle(\"Fig 5\")\n",
    "\n",
    "    cleanup()\n",
    "\n",
    "\n",
    "fig_5(architectures)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "id": "78622878",
   "metadata": {
    "jupyter": {
     "source_hidden": true
    }
   },
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 800x400 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "def fig_6(architectures):\n",
    "    \"\"\"Common readers and writers of storage services\"\"\"\n",
    "    # query\n",
    "    common_readers, common_writers = Counter(), Counter()\n",
    "\n",
    "    for arch in architectures:\n",
    "        storage_svcs = set(filter(is_storage, arch.get_all_nodes_services()))\n",
    "\n",
    "        # For each architecture, we are counting the existence of a service reading from/writing to any storage service.\n",
    "        # We are not counting the __number__ of times a service reads from/writes to a storage service.\n",
    "        this_arch_readers, this_arch_writers = set(), set()\n",
    "        for storage_svc in storage_svcs:\n",
    "            _writer_nodes = arch.get_inbound_nodes_for_svc(storage_svc, data_only=True)\n",
    "            this_arch_writers.update(t[\"service\"] for t in _writer_nodes)\n",
    "\n",
    "            _reader_nodes = arch.get_outbound_nodes_for_svc(storage_svc, data_only=True)\n",
    "            this_arch_readers.update(t[\"service\"] for t in _reader_nodes)\n",
    "\n",
    "        common_readers.update(this_arch_readers)\n",
    "        common_writers.update(this_arch_writers)\n",
    "\n",
    "    del common_readers[\"ThirdParty\"]  # we are concerned with AWS services only\n",
    "    del common_writers[\"ThirdParty\"]\n",
    "    common_readers = Counter(normalize_dict(common_readers, len(architectures)))\n",
    "    common_writers = Counter(normalize_dict(common_writers, len(architectures)))\n",
    "\n",
    "    # plotting\n",
    "    fig, (ax1, ax2) = plt.subplots(1, 2, figsize=(8, 4))\n",
    "\n",
    "    y_writer_names = [t[0] for t in common_writers.most_common(10)]\n",
    "    y_reader_names = [t[0] for t in common_readers.most_common(10)]\n",
    "\n",
    "    sorted_y_names = list(set(y_writer_names + y_reader_names))\n",
    "    sorted_y_names.sort(\n",
    "        key=lambda x: (common_writers[x] if x in y_writer_names else common_readers[x]),\n",
    "        reverse=False,\n",
    "    )\n",
    "    colors = [CAPABILITY_COLORS[svc_info[x][\"capability\"]] for x in sorted_y_names]\n",
    "    y = np.arange(len(sorted_y_names))\n",
    "\n",
    "    x = [common_writers[t] if t in y_writer_names else 0 for t in sorted_y_names]\n",
    "    ax1.barh(y, x, color=colors)\n",
    "    ax1.set_yticks(y)\n",
    "    ax1.set_yticklabels(sorted_y_names)\n",
    "    ax1.set_xlabel(\"Percentage of architectures (%)\")\n",
    "    ax1.set_title(\"Writers\")\n",
    "    ax1.grid(axis=\"x\", linestyle=\"--\")\n",
    "\n",
    "    x = [common_readers[t] if t in y_reader_names else 0 for t in sorted_y_names]\n",
    "    ax2.barh(y, x, color=colors)\n",
    "    ax2.set_yticks(y)\n",
    "    ax2.set_yticklabels([\"\"] * len(sorted_y_names))\n",
    "    ax2.set_xlabel(\"Percentage of architectures (%)\")\n",
    "    ax2.set_title(\"Readers\")\n",
    "    ax2.grid(axis=\"x\", linestyle=\"--\")\n",
    "\n",
    "    # legend\n",
    "    capabilities_shown = set(svc_info[x][\"capability\"] for x in sorted_y_names)\n",
    "    handles = [\n",
    "        plt.Rectangle((0, 0), 1, 1, color=CAPABILITY_COLORS[cap])\n",
    "        for cap in capabilities_shown\n",
    "    ]\n",
    "    ax2.legend(\n",
    "        handles,\n",
    "        capabilities_shown,\n",
    "        loc=\"upper left\",\n",
    "        bbox_to_anchor=(1, 1),\n",
    "        handlelength=0.5,\n",
    "    )\n",
    "\n",
    "    cleanup(ax1)\n",
    "    cleanup(ax2)\n",
    "\n",
    "    plt.suptitle(\"Fig 6\")\n",
    "\n",
    "\n",
    "fig_6(architectures)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "id": "0fdd0cba",
   "metadata": {
    "jupyter": {
     "source_hidden": true
    }
   },
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 600x400 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "def fig_7(architectures):\n",
    "    \"\"\"ratio of upload/download edges for S3\"\"\"\n",
    "    # query\n",
    "    ratios = []\n",
    "    for arch in architectures:\n",
    "        download_edges = len(arch.get_outbound_nodes_for_svc(\"S3\", data_only=True))\n",
    "        upload_edges = len(arch.get_inbound_nodes_for_svc(\"S3\", data_only=True))\n",
    "\n",
    "        # only consider architectures with S3\n",
    "        if download_edges == 0 and upload_edges == 0:\n",
    "            continue\n",
    "\n",
    "        ratio = upload_edges / download_edges if download_edges != 0 else float(\"inf\")\n",
    "        ratios.append(ratio)\n",
    "\n",
    "    replace_inf = max([x for x in ratios if x != float(\"inf\")]) + 1\n",
    "    ratios = [replace_inf if x == float(\"inf\") else x for x in ratios]\n",
    "\n",
    "    # plotting\n",
    "    def normalize_x(t):\n",
    "        if t < 1:\n",
    "            return t / 2  # half the graph is for t < 1. (t is already divided by 4.)\n",
    "        # for the other half of the graph, divide it into 4 subsections (because numerator goes up to 4)\n",
    "        return 0.5 + (t - 1) / 2 / 4\n",
    "\n",
    "    y = np.arange(len(ratios)) / len(ratios)\n",
    "    ratios.sort()\n",
    "    fig, ax = plt.subplots(figsize=(6, 4))\n",
    "\n",
    "    ax.plot([normalize_x(x) for x in ratios], y)\n",
    "\n",
    "    x_axis = [0, 1 / 4, 2 / 4, 3 / 4, 1, 2, 3, 4, 5]\n",
    "    x_axis_str = [\"0\", \"0.25\", \"0.5\", \"0.75\", \"1\", \"2\", \"3\", \"4\", r\"$\\infty$\"]\n",
    "    ax.set_xticks([normalize_x(t) for t in x_axis], x_axis_str)\n",
    "\n",
    "    ax.set_xlabel(\"Upload/Download ratio\")\n",
    "    ax.set_ylabel(\"CDF of arch with S3\")\n",
    "    ax.grid(linestyle=\"--\")\n",
    "\n",
    "    plt.suptitle(\"Fig 7\")\n",
    "\n",
    "    cleanup()\n",
    "\n",
    "\n",
    "fig_7(architectures)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "id": "ca497ecf",
   "metadata": {
    "jupyter": {
     "source_hidden": true
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Percentage of architectures with at least 1 ML service: 18.235294117647058 %\n"
     ]
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAk8AAAGbCAYAAADdgHTKAAAAOnRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjEwLjAsIGh0dHBzOi8vbWF0cGxvdGxpYi5vcmcvlHJYcgAAAAlwSFlzAAAPYQAAD2EBqD+naQAAXRtJREFUeJzt3Xl4TGf/BvD7zEwyGdkUIUEiNNZKJKjaWlvUVkXttLZuv5aiqkVba0vQorVUqRJtbbWrrY31ta+JXagtKEIrG0nM8vz+8M68pgnmGcPkyP25rrnezpkz53xzI/N9n/M8ZxQhhAAREREROUTj7gKIiIiI1ITNExEREZEENk9EREREEtg8EREREUlg80REREQkgc0TERERkQQ2T0REREQS2DwRERERSWDzRERERCSBzRMR5Uvnz5+HoiiIjY11dylEpDJsnojoqRQbGwtFUXJ9DB48+LGd99dff0XNmjVRsGBBFC5cGPXq1cOaNWse2/mI6MnTubsAIqLHadSoUShdurTdtsqVK6NUqVLIzMyEh4eHy841ZcoU9O3bFy1atMDYsWORlZWF2NhYvPLKK1i6dClee+01l52LiNxH4RcDE9HTKDY2Fj179sS+fftQvXr1J3LOcuXKoWDBgtizZw8URQEApKWloUSJEmjYsCFWrlz5ROogoseLl+2IKF+635ynxYsXo1KlSvDy8kLlypWxfPly9OjRA6GhoQ89ZlpaGooWLWprnADAz88PPj4+MBgMtm3//PMPBg4ciPDwcPj4+MDPzw/NmjXDoUOHXPXjEdFjxMt2RPRUS01NxY0bN+y2FSlSJNd916xZg44dOyI8PBwxMTG4efMm3nzzTZQoUcKhc9WvXx9LlizBlClT0LJlS2RlZWHKlClITU1Fv379bPudPXsWK1asQPv27VG6dGlcu3YNM2bMQL169XD8+HEUL17c+R+YiB47Nk9E9FSLjo7Ose1+sxWGDBmCEiVKYMeOHfDx8QEANGrUCPXr10epUqUeeq7Jkyfjxo0b6Nu3L/r27QvgbqO2ceNG1KpVy7ZfeHg4Tp06BY3mf4P/b7zxBipUqIAff/wRQ4cOlfoZiejJYvNERE+1adOmoVy5cg/d76+//sKRI0fw6aef2honAKhXrx7Cw8ORlpb20GMUKFAA5cuXR8mSJfHKK68gPT0dkyZNwmuvvYZt27YhLCwMAKDX623vMZvNSElJgY+PD8qXL4+DBw868VMS0ZPE5omInmo1atRwaML4hQsXAMDW4NwrLCzMoaamffv20Ol0+O2332zbWrVqhbJly+Kzzz7DokWLAAAWiwXffvstvvvuO5w7dw5ms9m2f+HChR96HiJyL04YJyJygbNnz2L9+vV49dVX7bYXKlQIdevWxY4dO2zbxowZgwEDBuCll17CL7/8gt9//x1xcXF47rnnYLFYnnTpRCSJI09ERIBtTtOff/6Z47Xctv3btWvXAMBuFMnKaDTCZDLZni9ZsgQNGjTAjz/+aLdfSkrKfSezE1HewZEnIiIAxYsXR+XKlfHTTz8hIyPDtn3r1q04cuTIQ98fFhYGjUaDRYsW2U1Iv3TpErZt24aoqCjbNq1Wm2PS+uLFi3H58mUX/CRE9Lhx5ImI6L/GjBmDVq1aoU6dOujZsydu3ryJqVOnonLlynYNVW4CAgLQq1cvzJo1C40aNcJrr72G9PR0fPfdd8jMzMSQIUNs+77yyisYNWoUevbsidq1a+PIkSOYN28eypQp87h/RCJyAY48ERH9V8uWLbFgwQLcuXMHgwcPxrJlyxAbG4vy5cvDy8vroe+fPn06pkyZgps3b2LIkCEYM2YMypYti7i4OLz00ku2/T799FN89NFH+P3339GvXz8cPHgQa9asQXBw8OP88YjIRfj1LEREDxEZGYmAgADExcW5uxQiygM48kRE9F//ntgNAFu2bMGhQ4dQv3599xRFRHkOR56IiP7r/PnziI6Oxuuvv47ixYvj5MmT+P777+Hv74+jR4/yHkxEBIATxomIbJ555hlUq1YNs2bNwvXr1+Ht7Y0WLVpg7NixbJyIyIYjT0REeZD1MuGWLVvcWgcR5cQ5T0SU58TGxkJRFCiKgu3bt+d4XQiB4OBgKIqCV155xe41RVHQp0+fJ1UqEeVDvGxHRHmWl5cX5s+fj7p169pt37p1Ky5dumT3BbtPmz/++MPdJRDRfXDkiYjyrObNm2Px4sU5VsDNnz8f1apVQ2BgoJsqk3fr1i2p/T09PeHp6fmYqiGiR8HmiYjyrM6dO+Pvv/+2u7/SnTt3sGTJEnTp0sWl57p69Sp69uyJkiVLQq/XIygoCK1atcL58+ft9lu3bh1efPFFeHt7w9fXFy1atMCxY8fs9unRowd8fHxw5swZNG/eHL6+vujatSv69OkDHx8f3L59O9efNTAw0PbdePXr189xe4SsrCyMGDEC5cqVg5eXF4KCgvDaa6/hzJkztn0sFgu++eYbPPfcc/Dy8kKxYsXw7rvv4ubNm3bH2r9/P5o0aYIiRYrAYDCgdOnS6NWr1yMkSJR/sHkiojwrNDQUtWrVwoIFC2zb1q1bh9TUVHTq1Mml52rbti2WL1+Onj174rvvvkPfvn2Rnp6OpKQk2z4///wzWrRoAR8fH4wbNw5Dhw7F8ePHUbdu3RxNlslkQpMmTVC0aFF8/fXXaNu2LTp27Ihbt25hzZo1dvvevn0bv/32G9q1awetVptrfWazGa+88gpGjhyJatWqYcKECejXrx9SU1Nx9OhR237vvvsuPv74Y9SpUwfffvstevbsiXnz5qFJkyYwGo0AgOTkZLz88ss4f/48Bg8ejClTpqBr167YvXu3i9IkesoJIqI8Zs6cOQKA2Ldvn5g6darw9fUVt2/fFkII0b59e9GgQQMhhBClSpUSLVq0sHsvANG7d2+p8928eVMAEF999dV990lPTxcFCxYUb7/9tt32q1evCn9/f7vt3bt3FwDE4MGD7fa1WCyiRIkSom3btnbbf/31VwFA/Oc//7Ftq1evnqhXr57t+ezZswUAMXHixBy1WSwWIYQQ27ZtEwDEvHnz7F5fv3693fbly5fb8iUieRx5IqI8rUOHDsjMzMTq1auRnp6O1atXu/ySncFggKenJ7Zs2ZLj8pZVXFwcUlJS0LlzZ9y4ccP20Gq1eOGFF7B58+Yc73nvvffsniuKgvbt22Pt2rV2XzS8aNEilChRIsfE+HstXboURYoUwQcffJDjNUVRAACLFy+Gv78/GjdubFdjtWrV4OPjY6uxYMGCAIDVq1fbRqOIyHFsnogoTwsICEB0dDTmz5+PZcuWwWw2o127di49h16vx7hx47Bu3ToUK1YML730EsaPH4+rV6/a9jl9+jQAoGHDhggICLB7/PHHH0hOTrY7pk6nQ8mSJXOcq2PHjsjMzMSqVasAABkZGVi7di3at29va4Jyc+bMGZQvXx463f0XSZ8+fRqpqakoWrRojhozMjJsNdarVw9t27bFyJEjUaRIEbRq1Qpz5sxBdna246ER5WO8VQER5XldunTB22+/jatXr6JZs2a2kRNX6t+/P1q2bIkVK1bg999/x9ChQxETE4NNmzYhKioKFosFwN15T7mt8vt3U6PX66HR5Pz/pzVr1kRoaCh+/fVXdOnSBb/99hsyMzPRsWPHR/4ZLBYLihYtinnz5uX6ekBAAIC7I1VLlizB7t278dtvv+H3339Hr169MGHCBOzevRs+Pj6PXAvR04zNExHleW3atMG7776L3bt3Y9GiRY/tPM8++yw++ugjfPTRRzh9+jQiIyMxYcIE/PLLL3j22WcBAEWLFkV0dPQjnadDhw749ttvkZaWhkWLFiE0NBQ1a9Z8aG179uyB0WiEh4fHfffZsGED6tSpA4PB8NA6atasiZo1a2L06NGYP38+unbtioULF+Ktt95y6uciyi942Y6I8jwfHx9Mnz4dI0aMQMuWLV1+/Nu3byMrK8tu27PPPgtfX1/bpawmTZrAz88PY8aMyXWe0PXr1x0+X8eOHZGdnY25c+di/fr16NChw0Pf07ZtW9y4cQNTp07N8Zr477dsdejQAWazGV988UWOfUwmE1JSUgAAN2/etL3HKjIyEgB46Y7IARx5IiJV6N69u8P77t+/H19++WWO7fXr1891UvapU6fQqFEjdOjQAZUqVYJOp8Py5ctx7do12y0R/Pz8MH36dLzxxhuoWrUqOnXqhICAACQlJWHNmjWoU6dOro1NbqpWrYqwsDB89tlnyM7OduiSXbdu3fDTTz9hwIAB2Lt3L1588UXcunULGzZswPvvv49WrVqhXr16ePfddxETE4OEhAS8/PLL8PDwwOnTp7F48WJ8++23aNeuHebOnYvvvvsObdq0wbPPPov09HT88MMP8PPzQ/PmzR36GYjyMzZPRPTU2bNnD/bs2ZNj+xdffJFr8xQcHIzOnTtj48aN+Pnnn6HT6VChQgX8+uuvaNu2rW2/Ll26oHjx4hg7diy++uorZGdno0SJEnjxxRfRs2dPqRo7duyI0aNHIywsDFWrVn3o/lqtFmvXrrVdYlu6dCkKFy6MunXrIjw83Lbf999/j2rVqmHGjBn49NNPodPpEBoaitdffx116tQBcHfC+N69e7Fw4UJcu3YN/v7+qFGjBubNm4fSpUtL/RxE+ZEi/j12S0RERET3xTlPRERERBLYPBERERFJYPNEREREJIHNExEREZEENk9EREREEtg8EREREUlg86QyQgikpaXluDswERERPRlsnlQmPT0d/v7+tq9ZoIczm804efIkzGazu0tRFeYmj5nJY2bymJlzrLm5ApsnlbJ+wzs9nMViQWJiIjOTxNzkMTN5zEweM3OONTdX4B3GVSYtLQ3+/v64efMmChYs6O5yVMFiseDGjRsoUqQINBr+/wVHMTd5zEweM5PHzJxjza1o0aKPfCw2TypjbZ5SU1Ph5+fn7nKIiIjyHbasKmU0Gt1dgmoYjUasWbOGmUlibvKYmTxmJo+ZOceamyuweaJ8wWQyubsEVWJu8piZPGYmj5k5x1W5sXkiIiIiksDmiYiIiEgCJ4yrjHXCeEpKCvz9/d1djioIIZCeng5fX18oiuLuclSDucljZvKYmTxm5hxrbq5YbMWRJ8oXDAaDu0tQJeYmj5nJY2bymJlzXJUbmyeV4mRBx5lMJqxdu5aZSWJu8piZPGYmj5k5x5qbK7B5IiIiIpLA5omIiIhIApsnIiIiIglcbacy1tV2JMdgMCAzM9PdZagOc5PHzOQxM3n5ITNXtydCCJhMJnh4eDzysTjypFJcnuo4jUbDL9B0AnOTx8zkMTN5zMx5rmo4mbxK6fV6d5egGnq9HlOmTGFmkpibPGYmj5nJY2bOMZlM2Lx5s0uOxeaJiIiISAKbJyIiIiIJbJ4oX7h9+7a7S1Al5iaPmcljZvKYmXN0Op1LjsPVdirD1XZERJQf5OX2hCNPKsXVdo7TaDSIjIzkyhRJzE0eM5PHzOQxM+dYLBYkJye75FhMXqW4ysJxer0eI0aMYGaSmJs8ZiaPmcljZs4xm83YtWuXS47F5omIiIhIApsnIiIiIglsnlQqL0+ky2uEEEhKSmJmkpibPGYmj5nJY2bOURQFvr6+rjkWV9upC1fbERFRfpCX2xOOPKmUVqt1dwmqodPpEB0d7bL7e+QXzE0eM5PHzOQxM+dYLBZcuHDBJcdi86RSrvhW6PzCw8MDffr0YWaSmJs8ZiaPmcljZs4xm81ISEhwybHYPBERERFJYPNEREREJIHNk0pZLBZ3l6AaFosF8fHxzEwSc5PHzOQxM3nMzDmKoiAgIMA1x+JqO+fExsaif//+SElJeaLn5Wo7IiLKD/Jye/LYR56uX7+O9957DyEhIdDr9QgMDESTJk2wY8eOx31qO6GhoVAUBQsXLszx2nPPPQdFURAbG/tEa3oUXG3nOJ1Oh06dOnFliiTmJo+ZyWNm8piZc8xmM06ePOmSYz325qlt27aIj4/H3LlzcerUKaxatQr169fH33///bhPnUNwcDDmzJljt2337t24evUqvL29n3g99xJCwGQyObw/V1k4zsPDA506dWJmkpibPGYmj5nJY2bOsVgsSExMdMmxHmvzlJKSgm3btmHcuHFo0KABSpUqhRo1amDIkCF49dVXAQATJ05EeHg4vL29ERwcjPfffx8ZGRl2x/nhhx8QHByMAgUKoE2bNpg4cSIKFixot8/KlStRtWpVeHl5oUyZMhg5cmSOZqRr167YunUrLl68aNs2e/ZsdO3aNUcH70hd97p+/TqqV6+ONm3aIDs7GxaLBTExMShdujQMBgOqVKmCJUuW2PbfsmULFEXBunXrUK1aNej1emzfvl0qXyIiInryHmvz5OPjAx8fH6xYsQLZ2dm5F6DRYPLkyTh27Bjmzp2LTZs24ZNPPrG9vmPHDvzf//0f+vXrh4SEBDRu3BijR4+2O8a2bdvQrVs39OvXD8ePH8eMGTMQGxubY79ixYqhSZMmmDt3LgDg9u3bWLRoEXr16iVd170uXryIF198EZUrV8aSJUug1+sRExODn376Cd9//z2OHTuGDz/8EK+//jq2bt1q997Bgwdj7NixOHHiBCIiInIcOzs7G2lpaXYPIiIiciPxmC1ZskQ888wzwsvLS9SuXVsMGTJEHDp06L77L168WBQuXNj2vGPHjqJFixZ2+3Tt2lX4+/vbnjdq1EiMGTPGbp+ff/5ZBAUF2Z6XKlVKTJo0SaxYsUI8++yzwmKxiLlz54qoqCghhBD+/v5izpw5Dtc1Z84c4e/vL06ePCmCg4NF3759hcViEUIIkZWVJQoUKCB27txpd4w333xTdO7cWQghxObNmwUAsWLFivueUwghhg8fLgDkeHh4eOS6nY+cD09PT9G7d2/h6enp9lrU9GBuzIyZ5c1HfsnM1Uwmkzh48KBLjvXYmychhMjMzBR//PGHGDVqlKhVq5bQarW2RiUuLk40bNhQFC9eXPj4+AgvLy8BQNy6dUsIIURkZKQYOXKk3fG+/fZbu+apSJEiwsvLS3h7e9se/z6OtXkyGo2iWLFiYsuWLaJevXpiypQpQoiczdPD6pozZ47w9PQUAQEBol+/fnb1HT16VACwq8fb21t4eHiIGjVqCCH+1zxdunTpgdllZWWJ1NRU2+PixYtu/wvNBx988MEHH4/7kZc9kfs8eXl5oXHjxhg6dCh27tyJHj16YPjw4Th//jxeeeUVREREYOnSpThw4ACmTZsGALhz547Dx8/IyMDIkSORkJBgexw5cgSnT5+Gl5eX3b46nQ5vvPEGhg8fjj179qBr1645judoXXq9HtHR0Vi9ejUuX75sVw8ArFmzxq6m48eP2817AvDQiep6vR5+fn52D4ATxmV4enqid+/e8PT0dHcpqsLc5DEzecxMHjNzjtlsRnx8vEuO5ZabZFaqVAm3bt3CgQMHYLFYMGHCBNSsWRPlypXDX3/9Zbdv+fLlsW/fPrtt/35etWpVJCYmIiwsLMdDo8n5I/bq1Qtbt25Fq1at8Mwzz+R43ZG6gLvzon7++WdUq1YNDRo0sO1TqVIl6PV6JCUl5agnODhYOq/c8FYFjtNqtWjcuDEzk8Tc5DEzecxMHjNzjsViQVJSkkuO9VhvEvH333+jffv26NWrFyIiIuDr64v9+/dj/PjxaNWqFcLCwmA0GjFlyhS0bNkSO3bswPfff293jA8++AAvvfQSJk6ciJYtW2LTpk1Yt24dFEWx7TNs2DC88sorCAkJQbt27aDRaHDo0CEcPXoUX375ZY66KlasiBs3bqBAgQK51u1IXVZarRbz5s1D586d0bBhQ2zZsgWBgYEYOHAgPvzwQ1gsFtStWxepqanYsWMH/Pz80L1790dIlYiIiNzqcV4TzMrKEoMHDxZVq1YV/v7+okCBAqJ8+fLi888/F7dv3xZCCDFx4kQRFBQkDAaDaNKkifjpp58EAHHz5k3bcWbOnClKlCghDAaDaN26tfjyyy9FYGCg3bnWr18vateuLQwGg/Dz8xM1atQQM2fOtL1unfN0P/+e8/SwuqwTxq2MRqN47bXXRMWKFcW1a9eExWIR33zzjShfvrzw8PAQAQEBokmTJmLr1q1CiP/Nebr353REamqqAGCbg8XHwx8Gg0GsWLFCGAwGt9eipgdzY2bMLG8+8ktmrnbnzp2HLtJyVN6ekXUfb731lqhbt667y3ALa/Ok1Wrd/hdbLQ+dTic6deokdDqd22tR04O5MTNmljcf+SUzVzOZTOLEiRMuOZYqvtvu66+/RuPGjeHt7Y1169bho48+wnfffYe33nrL3aU9cfxuOyIiyg/ycnvilgnjsvbu3YvGjRsjPDwc33//PSZPnpwvG6d7cZWF4/R6PYYPHw69Xu/uUlSFucljZvKYmTxm5hyTyYSdO3e65Fiq+FbBX3/91d0l5Dm5rSKk3Gk0GkRFRTEzScxNHjOTx8zkMTPnCCFw/fp1lxyLyRMRERFJYPNEREREJIHNk0oZjUZ3l6AaRqMRU6dOZWaSmJs8ZiaPmcljZs7RarWIjIx0ybFUsdqO/oer7YiIKD/Iy+0JR55UiqssHOfl5YXJkyfn+J5DejDmJo+ZyWNm8piZc0wmEzZt2uSSY7F5Uql7v56GHkxRFISEhDAzScxNHjOTx8zkMTPnCCGQnp7ukmOxeSIiIiKSwOaJiIiISAInjKuMdcK4oih5ejJdXqLRaBAREYHDhw/DYrG4uxzVYG7ymJk8ZiYvv2Tm6s84i8WCGzduoGjRoo98LDZPKsPVdkRElB/k5faEl+1UiqssHGcwGDB//nwYDAZ3l6IqzE0eM5PHzOQxM+cYjUasWbPGJcdi80T5QoECBdxdgioxN3nMTB4zk8fMnGMymVxyHDZPRERERBJ07i6AnHPp0iUULlzY3WWogtFoxNq1a5GamgoPDw93l6MazE0eM5PHzOQxM/fjhHGVsU4YT0lJ4cRxB1lvjObr68ubyklgbvKYmTxmJo+ZOceam5+f3yMfi5ftKF/gxErnMDd5zEweM5PHzJzjqtzYPKmUqya95Qcmkwlr165lZpKYmzxmJo+ZyWNmzrHm5gpsnoiIiIgksHkiIiIiksDmiYiIiEgCV9upDFfbyRNCwGQyQafTcWWKBOYmj5nJY2bymJlzrLm54vYOHHmifCEzM9PdJagSc5PHzOQxM3nMzDmuyo3Nk0pxlYXjTCYTNm/ezMwkMTd5zEweM5PHzJxjzc0V2DwRERERSWDzRERERCSB322nUiVLlkRWVpa7y1AFg8GAH3/8Ef7+/pwnkIsHrRnR6fgrQhYzk8fM5DEz57gqN662UxnrajsiV+GvACIiObxsp1Jcnuo4jUaDyMhIaDT86y7DYrEgOTkZFovF3aWoBjOTx8zkMTPnWHNzBX6aqJRer3d3Caqh1+sxYsQIZibJbDZj165dMJvN7i5FNZiZPGYmj5k5x5qbK7B5IiIiIpLA5omIiIhIApsnleIkX8cJIZCUlMTMJCmKAl9fX86vk8DM5DEzeczMOdbcXHIsrrZTF662I1fjrwAiIjkceVIprVbr7hJUQ6fTITo6mvdFkWSxWHDhwgWu6JHAzOQxM3nMzDnW3FyBzZNKueJbofMLDw8P9OnTh5lJMpvNSEhI4IoeCcxMHjOTx8ycY83NFdg8EREREUlg80REREQkgc2TSvFat+MsFgvi4+OZmSRFURAQEMAVPRKYmTxmJo+ZOceam0uOxdV26sLVduRq/BVARCSHI08qxdV2jtPpdOjUqRNX20kym804efIkJ6VKYGbymJk8ZuYca26uwOZJpbhyzHEeHh7o1KkTM5NksViQmJjIy50SmJk8ZiaPmTnHmpsrsHkiIiIiksDmiYiIiEgCmycA58+fh6IoLrt51pM4N691O85sNiMuLo6ZSdJoNAgJCYFGw18TjmJm8piZPGbmHGtuLjmWM2+6evUqPvjgA5QpUwZ6vR7BwcFo2bIlNm7c6JKi6OGMRqO7S1CNO3fuYNq0abhz5467S1EVrVaLqKgoLk6QwMzkMTN5zMw51txcQbp5On/+PKpVq4ZNmzbhq6++wpEjR7B+/Xo0aNAAvXv3dklRrvS0Nhmc/Ow4T09P9O7dG56enu4uRVXMZjPi4+M5YieBmcljZvKYmXOsubmCdPP0/vvvQ1EU7N27F23btkW5cuXw3HPPYcCAAdi9ezcAICkpCa1atYKPjw/8/PzQoUMHXLt2zXaMESNGIDIyErNnz0ZISAh8fHzw/vvvw2w2Y/z48QgMDETRokUxevRou3MrioLp06ejWbNmMBgMKFOmDJYsWWJ73XoJbNGiRahXrx68vLwwb948AMCsWbNQsWJFeHl5oUKFCvjuu+9y/Gxnz55FgwYNUKBAAVSpUgW7du2ye3379u148cUXYTAYEBwcjL59++LWrVu210NDQzFmzBj06tULvr6+CAkJwcyZM+2OsXfvXkRFRcHLywvVq1d3+g+S/4/DcVqtFo0bN2ZmkiwWC5KSkriiRwIzk8fM5DEz51hzcwkh4e+//xaKoogxY8bcdx+z2SwiIyNF3bp1xf79+8Xu3btFtWrVRL169Wz7DB8+XPj4+Ih27dqJY8eOiVWrVglPT0/RpEkT8cEHH4iTJ0+K2bNnCwBi9+7dtvcBEIULFxY//PCDSExMFJ9//rnQarXi+PHjQgghzp07JwCI0NBQsXTpUnH27Fnx119/iV9++UUEBQXZti1dulQUKlRIxMbG2r2vQoUKYvXq1SIxMVG0a9dOlCpVShiNRiGEEH/++afw9vYWkyZNEqdOnRI7duwQUVFRokePHrb6SpUqJQoVKiSmTZsmTp8+LWJiYoRGoxEnT54UQgiRnp4uAgICRJcuXcTRo0fFb7/9JsqUKSMAiPj4+FzzzMrKEqmpqbbHxYsXBQDh5eUlAPDhwMNgMIgVK1YIg8Hg9lry4uN+7ty5I1asWCHu3Lnz4F8MZMPM5DEzeczMOdbcXEGqedqzZ48AIJYtW3bfff744w+h1WpFUlKSbduxY8cEALF3714hxN3mqUCBAiItLc22T5MmTURoaKgwm822beXLlxcxMTH/KxYQ//d//2d3vhdeeEG89957Qoj/NUHffPON3T7PPvusmD9/vt22L774QtSqVcvufbNmzcpR84kTJ4QQQrz55pvinXfesTvGtm3bhEajEZmZmUKIu83T66+/bnvdYrGIokWLiunTpwshhJgxY4YoXLiwbX8hhJg+fboA7t88DR8+PNcPPDZPjj/YPD34cT/8BS2PmcljZvKYmXNc2TxJXbYTDnyNw4kTJxAcHIzg4GDbtkqVKqFgwYI4ceKEbVtoaCh8fX1tz4sVK4ZKlSrZrR4oVqwYkpOT7Y5fq1atHM/vPS4AVK9e3fbft27dwpkzZ/Dmm2/Cx8fH9vjyyy9x5swZu/dFRETY/jsoKAgAbOc/dOgQYmNj7Y7RpEkTWCwWnDt3LtdjKIqCwMBA2zFOnDiBiIgIeHl53ffn+bchQ4YgNTXV9rh48SKAp3cu1+NgNBqxcOFCZiZJo9GgfPnyXNEjgZnJY2bymJlzrLm5gtT3VZQtWxaKorjk9ub/nvCsKEqu25y5puvt7W3774yMDADADz/8gBdeeMFuv3/Pgbn3/NYvXLSePyMjA++++y769u2b43z3Ln101c9gpdfrodfrc2znREHHmUwmLFy40N1lqI5Wq0WFChXcXYaqMDN5zEweM3OOK3OTalsLFSqEJk2aYNq0aXYTpa1SUlJQsWJFXLx40TZCAgDHjx9HSkoKKlWq9MgFWyel3/u8YsWK992/WLFiKF68OM6ePYuwsDC7R+nSpR0+b9WqVXH8+PEcxwgLC3N4FVfFihVx+PBhZGVl3ffncRRXjjlOr9dj+PDhuTahdH8mkwk7d+6EyWRydymqwczkMTN5zMw51txcQXrMb9q0aTCbzahRowaWLl2K06dP48SJE5g8eTJq1aqF6OhohIeHo2vXrjh48CD27t2Lbt26oV69enaX05y1ePFizJ49G6dOncLw4cOxd+9e9OnT54HvGTlyJGJiYjB58mScOnUKR44cwZw5czBx4kSHzzto0CDs3LkTffr0QUJCAk6fPo2VK1c+9Nz36tKlCxRFwdtvv43jx49j7dq1+Prrrx1+/704XOs4jUaDqKgoZiZJCIHr1687dLme7mJm8piZPGbmHGturiD9aVKmTBkcPHgQDRo0wEcffYTKlSujcePG2LhxI6ZPnw5FUbBy5Uo888wzeOmllxAdHY0yZcpg0aJFLil45MiRWLhwISIiIvDTTz9hwYIFDx3ReuuttzBr1izMmTMH4eHhqFevHmJjY6VGniIiIrB161acOnUKL774IqKiojBs2DAUL17c4WP4+Pjgt99+w5EjRxAVFYXPPvsM48aNc/j9RERE5H6KUFHrqigKli9fjtatW7u7FLdJS0uDv78/vLy87C7/0f0ZDAYsWLAAnTt3RmZmprvLyXPu9yvAaDRi7dq1aN68OW/K6iBmJo+ZyWNmzrHm1qpVq0c+Fq9jqBRXjjnOaDRi6tSpzEySVqtFZGQkby4qgZnJY2bymJlzrLm5AkeeVMY68kTkKir6FUBElCeoauRJCJGvG6d7ceWY47y8vDB58mS7+2vRw5lMJmzatIkreiQwM3nMTB4zc441N1dQVfNE/2O9DxU9nKIoCAkJYWaShBBIT0/nyJQEZiaPmcljZs6x5uYKbJ6IiIiIJLB5IiIiIpKgqgnj9L8J44qicMjWQRqNBhERETh8+PAjfVXO0+p+f48sFgtu3LiBIkWK8AajDmJm8piZPGbmHGtuRYsWfeRjsXlSGa62I1fjrwAiIjlsWVWKK8ccZzAYMH/+fBgMBneXoipGoxFr1qzh/bEkMDN5zEweM3OONTdXYPNE+UKBAgXcXYIqcSm0PGYmj5nJY2bOcVVubJ6IiIiIJLB5IiIiIpLACeMqw9V28jQaDUqUKIHLly9ztV0u7vf3yHpDOV9fX95g1EHMTB4zk8fMnGPNzc/P75GPxZEnlWLj5Djr8lQ2TvI4yV4eM5PHzOQxM+e4Kjc2TyrF1XaOMxgMWLBgAX/ZSDKZTFi7di0npkpgZvKYmTxm5hxrbq7A5omIiIhIApsnIiIiIgk6dxdAzrl06RIKFy7s7jJUwWg0Yu3atUhNTYWHh4e7yyEiIpXjajuVsa62S0lJ4de0OEgIAZPJBJ1Ox5UpEpibPGYmj5nJY2bOsebmiv8Tzct2lC9kZma6uwRVYm7ymJk8ZiaPmTnHVbmxeVIprrJwnMlkwubNm5mZJOYmj5nJY2bymJlzrLm5ApsnIiIiIglsnoiIiIgksHmifEGn48JSZzA3ecxMHjOTx8yc46rcuNpOZayr7VJTU13y/TxEREQkhyNPKsXvaXOcxWJBcnIyM5PE3OQxM3nMTB4zc441N1dg86RSZrPZ3SWohtlsxq5du5iZJOYmj5nJY2bymJlzrLm5ApsnIiIiIgmccaZSJUuWRFZWlrvLUAWDwYAFCxbA399ftTeW49REIqK8gyNPKsUPU8cJIZCUlMTMJCmKAl9fX379gwRmJo+ZyWNmzrHm5pJjcbWdulhX21H+wn+mRER5B0eeVEqr1bq7BNXQ6XSIjo7mfVEkWSwWXLhwgSt6JDAzecxMHjNzjjU3V2DzpFKu+Fbo/MLDwwN9+vRhZpLMZjMSEhK4okcCM5PHzOQxM+dYc3MFNk9EREREEtg8EREREUlg86RSvNbtOIvFgvj4eGYmSVEUBAQEcEWPBGYmj5nJY2bOsebmkmNxtZ26cLVd/sR/pkREeQdHnlSKq+0cp9Pp0KlTJ662k2Q2m3Hy5ElOSpXAzOQxM3nMzDnW3FyBzZNKceWY4zw8PNCpUydmJslisSAxMZGXOyUwM3nMTB4zc441N1dg80REREQkgc0TERERkQQ2TyrFa92OM5vNiIuLY2aSNBoNQkJCoNHw14SjmJk8ZiaPmTnHmpsrcLWdynC1Xf7Ef6ZERHkH21aV4uRnx3l6eqJ3797w9PR0dymqYjabER8fzxE7CcxMHjOTx8ycY83NFdg8qRRvVeA4rVaLxo0bMzNJFosFSUlJXNEjgZnJY2bymJlzrLm5ApsnIiIiIgl5qnnasmULFEVBSkqKW+sYMWIEIiMjH7jP+fPnoSiKy76hmYiIiNTBpc1Tjx49oCgKFEWBh4cHSpcujU8++QRZWVmuPM1jN3DgQGzcuNH2vEePHmjdurXdPsHBwbhy5QoqV678hKu7y2g0uuW8amQ0GrFw4UJmJkmj0aB8+fJc0SOBmcljZvKYmXOsubmCy7+vomnTppgzZw6MRiMOHDiA7t27Q1EUjBs3ztWnemx8fHzg4+PzwH20Wi0CAwOfUEU5caKg40wmExYuXOjuMlRHq9WiQoUK7i5DVZiZPGYmj5k5x5W5ubxt1ev1CAwMRHBwMFq3bo3o6GjExcUBuDtZKyYmBqVLl4bBYECVKlWwZMmS+x7r9u3baNasGerUqYOUlBRYLBaMGjUKJUuWhF6vR2RkJNavX2/3np07dyIyMhJeXl6oXr06VqxYYXd5zXppcOPGjahevToKFCiA2rVr292y/d7LdiNGjMDcuXOxcuVK26jali1bcr1st3XrVtSoUQN6vR5BQUEYPHgwTCaT7fX69eujb9+++OSTT1CoUCEEBgZixIgRTuXMlWOO0+v1GD58OPR6vbtLURWTyYSdO3fa/R2mB2Nm8piZPGbmHGturvBYx/yOHj2KnTt32j7oY2Ji8NNPP+H777/HsWPH8OGHH+L111/H1q1bc7w3JSUFjRs3hsViQVxcHAoWLIhvv/0WEyZMwNdff43Dhw+jSZMmePXVV3H69GkAd++B1LJlS4SHh+PgwYP44osvMGjQoFxr++yzzzBhwgTs378fOp0OvXr1ynW/gQMHokOHDmjatCmuXLmCK1euoHbt2jn2u3z5Mpo3b47nn38ehw4dwvTp0/Hjjz/iyy+/tNtv7ty58Pb2xp49ezB+/HiMGjXK1lzK4HCt4zQaDaKiopiZJCEErl+/zntMSWBm8piZPGbmHGturuDyy3arV6+Gj48PTCYTsrOzodFoMHXqVGRnZ2PMmDHYsGEDatWqBQAoU6YMtm/fjhkzZqBevXq2Y1y9ehUdO3ZE2bJlMX/+fFvz9fXXX2PQoEHo1KkTAGDcuHHYvHkzvvnmG0ybNg3z58+Hoij44Ycf4OXlhUqVKuHy5ct4++23c9Q5evRo2zkHDx6MFi1aICsrC15eXnb7+fj4wGAwIDs7+4GX6b777jsEBwdj6tSpUBQFFSpUwF9//YVBgwZh2LBhtg/uiIgIDB8+HABQtmxZTJ06FRs3bkTjxo1zPW52djays7Ntz9PS0h78B0BERESPlcubpwYNGmD69Om4desWJk2aBJ1Oh7Zt2+LYsWO4fft2jibhzp07iIqKstvWuHFj1KhRA4sWLbLdmyctLQ1//fUX6tSpY7dvnTp1cOjQIQBAYmIiIiIi7BqgGjVq5FpnRESE7b+DgoIAAMnJyU7fuv3EiROoVasWFEWxqy0jIwOXLl2yHffe81rPnZycfN/jxsTEYOTIkU7VRERERK7n8ubJ29sbYWFhAIDZs2ejSpUq+PHHH22r0tasWYMSJUrYveffc1FatGiBpUuX4vjx4wgPD3d1iQDs79BtbXiexA3H/n1ncEVRHnjeIUOGYMCAAbbnaWlpCA4O5soxCUajEVOnTmVmkrRaLSIjI3lzUQnMTB4zk8fMnGPNzRVc3jzdS6PR4NNPP8WAAQNw6tQp6PV6JCUl2V2iy83YsWPh4+ODRo0aYcuWLahUqRL8/PxQvHhx7Nixw+79O3bssI0ulS9fHr/88guys7NtDdm+ffse+efw9PR86Oq2ihUrYunSpRBC2JqxHTt2wNfXFyVLlnT63Hq9PteJzlxt5ziTyYQNGza4uwzV0Wg0KFWqlLvLUBVmJo+ZyWNmznFlbo99Bm379u2h1WoxY8YMDBw4EB9++CHmzp2LM2fO4ODBg5gyZQrmzp2b431ff/01unbtioYNG+LkyZMAgI8//hjjxo3DokWLkJiYiMGDByMhIQH9+vUDAHTp0gUWiwXvvPMOTpw4gd9//x1ff/01ANhdTpMVGhqKw4cPIzExETdu3Mh1BOP999/HxYsX8cEHH+DkyZNYuXIlhg8fjgEDBjyWicpcOeY4Ly8vTJ48Ocd8Nnowk8mETZs2cUWPBGYmj5nJY2bOsebmCo915AkAdDod+vTpg/Hjx+PcuXMICAhATEwMzp49i4IFC6Jq1ar49NNPc33vpEmTYDab0bBhQ2zZsgV9+/ZFamoqPvroIyQnJ6NSpUpYtWoVypYtCwDw8/PDb7/9hvfeew+RkZEIDw/HsGHD0KVLl0f64Hz77bexZcsWVK9eHRkZGdi8eTNCQ0Pt9ilRogTWrl2Ljz/+GFWqVEGhQoXw5ptv4vPPP3f6vA/yKM1gfqMoCkJCQpiZJCEE0tPTuaJHAjOTx8zkMTPnWHNzBUU85enPmzcPPXv2RGpqKgwGg7vLeWRpaWnw9/eHl5eX6u7c7i4GgwELFixA586dkZmZ6e5ynOKOf6ZGoxFr165F8+bNc8zVo9wxM3nMTB4zc441t1atWj3ysR77yNOT9tNPP6FMmTIoUaIEDh06hEGDBqFDhw5PReNERERE7vfUjTyNHz8e3333Ha5evYqgoCC0bt0ao0ePRoECBdxdmktYR54UReGQrYM0Gg0iIiJw+PDhJ7Ki8nFwx5+1xWLBjRs3UKRIEd5g1EHMTB4zk8fMnGPNrWjRoo98rKeueXraWZsnyl/4z5SIKO9gy6pSXDnmOIPBgPnz5/PSrSSj0Yg1a9bw/lgSmJk8ZiaPmTnHmpsrsHmifOFpuWz7pHEptDxmJo+ZyWNmznFVbmyeiIiIiCSweSIiIiKSwAnjKsPVdvI0Gg1KlCiBy5cvc7Wd5DnT09Ph6+vLG4w6iJnJY2bymJlzrLn5+fk98rE48qRSbJwcZ12eqtbGyZ04yV4eM5PHzOQxM+e4Kjc2TyrF1XaOs95hnL9s5JhMJqxdu5YTUyUwM3nMTB4zc441N1dg80REREQkgc0TERERkQQ2T0REREQSuNpOZfj1LM4xGAzIzMx0dxlOc9dqO5PJBJ1OxxU9DmJm8piZPGbmHGtuHh4ej3wsjjypFP/BOE6j0fALNJ2k5obTXZiZPGYmj5k5x1W58dNEpfR6vbtLUA29Xo8pU6YwM0kmkwmbN2/mih4JzEweM5PHzJxjzc0V2DwRERERSdC5uwByzqVLl1C4cGF3l6EKRqMRa9euRWpqqkuudRMRUf7GkSfKF3Q6/v8EZzA3ecxMHjOTx8yc46rcuNpOZayr7VJTU13y/TxEREQkhyNPKsXvaXOcxWJBcnIyM5PE3OQxM3nMTB4zc441N1dg86RSZrPZ3SWohtlsxq5du5iZJOYmj5nJY2bymJlzrLm5ApsnIiIiIglsnoiIiIgksHlSKd5h3HGKosDX15eZSWJu8piZPGYmj5k5x5qbS47F1XbqwtV2RERE7sWRJ5XiKgvHWSwWXLhwgZlJYm7ymJk8ZiaPmTnHmpsrsHlSKa6ycJzZbEZCQgIzk8Tc5DEzecxMHjNzjjU3V2DzRERERCSBzZNKlSxZEoqi8OHAw9/fHwDg7+//WM9DRET5A5snleK1bsdZLBbEx8czM0mKoiAgIICNoQRmJo+ZyWNmzrHm5pJjcbWdulhX21Hew39KRET5A0eeVEqr1bq7BNXQ6XTo1KkTv4VcktlsxsmTJzkpVQIzk8fM5DEz51hzcwU2Tyrl4eHh7hJUw8PDA506dWJmkiwWCxITE3m5UwIzk8fM5DEz51hzcwU2T0REREQS2DwRERERSWDzpFK81u04s9mMuLg4ZiZJo9EgJCQEGg1/TTiKmcljZvKYmXOsubkCV9upDFfb5V38p0RElD+wbVUpTn52nKenJ3r37g1PT093l6IqZrMZ8fHxHLGTwMzkMTN5zMw51txcgc2TSvFWBY7TarVo3LgxM5NksViQlJTEFT0SmJk8ZiaPmTnHmpsrsHkiIiIiksDmiYiIiEgCmyeVMhqN7i5BNYxGIxYuXMjMJGk0GpQvX54reiQwM3nMTB4zc441N1fgajuV4Wq7vIv/lIiI8ge2rSrFlWOO0+v1GD58OPR6vbtLURWTyYSdO3fCZDK5uxTVYGbymJk8ZuYca26uwOZJpThc6ziNRoOoqChmJkkIgevXr3NETQIzk8fM5DEz51hzcwV+mhARERFJYPP0CHr06IHWrVvbntevXx/9+/d3Wz1ERET0+D3x5klRlAc+RowY8aRLctq3336L2NhYt5ybK8ccZzQaMXXqVGYmSavVIjIykjcXlcDM5DEzeczMOdbcXEHnkqNIuHLliu2/Fy1ahGHDhiExMdG2zcfHx/bfQgiYzWbodE+8zAcym81QFMWtq954W37HmUwmbNiwwd1lqI5Go0GpUqXcXYaqMDN5zEweM3OOK3N74iNPgYGBtoe/vz8URbE9P3nyJHx9fbFu3TpUq1YNer0e27dvx5kzZ9CqVSsUK1YMPj4+eP7553N8GIaGhmLMmDHo1asXfH19ERISgpkzZ9pev3PnDvr06YOgoCB4eXmhVKlSiImJsb2ekpKCd999F8WKFYOXlxcqV66M1atXAwBiY2NRsGBBrFq1CpUqVYJer0dSUlKOy3bA3Q/qPn36wN/fH0WKFMHQoUPtJvVlZ2dj4MCBKFGiBLy9vfHCCy9gy5Yt0jly5ZjjvLy8MHnyZHh5ebm7FFUxmUzYtGkTV/RIYGbymJk8ZuYca26ukCfnPA0ePBhjx47FiRMnEBERgYyMDDRv3hwbN25EfHw8mjZtipYtW+b4jpoJEyagevXqiI+Px/vvv4/33nvPNqo1efJkrFq1Cr/++isSExMxb948hIaGArj7fTfNmjXDjh078Msvv+D48eMYO3as3ZDo7du3MW7cOMyaNQvHjh1D0aJFc6197ty50Ol02Lt3L7799ltMnDgRs2bNsr3ep08f7Nq1CwsXLsThw4fRvn17NG3aFKdPn871eNnZ2UhLS7N7AHcvf5JjFEVBSEgIM5MkhEB6ejpX9EhgZvKYmTxm5hxrbq6Qt66H/deoUaPQuHFj2/NChQqhSpUqtudffPEFli9fjlWrVqFPnz627c2bN8f7778PABg0aBAmTZqEzZs3o3z58khKSkLZsmVRt25dKIpiN3S3YcMG7N27FydOnEC5cuUAAGXKlLGryWg04rvvvrOrIzfBwcGYNGkSFEVB+fLlceTIEUyaNAlvv/02kpKSMGfOHCQlJaF48eIAgIEDB2L9+vWYM2cOxowZk+N4MTExGDlypKPRERER0WOWJ0eeqlevbvc8IyMDAwcORMWKFVGwYEH4+PjgxIkTOUaeIiIibP9tvRyYnJwM4O7KuISEBJQvXx59+/bFH3/8Yds3ISEBJUuWtDVOufH09LQ7/v3UrFnTboSjVq1aOH36NMxmM44cOQKz2Yxy5crBx8fH9ti6dSvOnDmT6/GGDBmC1NRU2+PixYsPrYGIiIgenzw58uTt7W33fODAgYiLi8PXX3+NsLAwGAwGtGvXDnfu3LHbz8PDw+65oiiwWCwAgKpVq+LcuXNYt24dNmzYgA4dOiA6OhpLliyBwWB4aE0Gg+GRL/tkZGRAq9XiwIEDOVZJ3DtR/l56vT7X+U3Z2dmPVEt+kp2djREjRjAzSVqtFrVq1eKKHgnMTB4zk8fMnGPNzRXyZPP0bzt27ECPHj3Qpk0bAHebkPPnz0sfx8/PDx07dkTHjh3Rrl07NG3aFP/88w8iIiJw6dIlnDp16oGjT47Ys2eP3fPdu3ejbNmy0Gq1iIqKgtlsRnJyMl588cVHOg+vdTvOYrEgISHB3WWojkajue/cPsodM5PHzOQxM+e4Mrc8ednu38qWLYtly5YhISEBhw4dQpcuXWwjSo6aOHEiFixYgJMnT+LUqVNYvHgxAgMDUbBgQdSrVw8vvfQS2rZti7i4ONsI1fr166VrTUpKwoABA5CYmIgFCxZgypQp6NevHwCgXLly6Nq1K7p164Zly5bh3Llz2Lt3L2JiYrBmzRqp83DlmOMMBgPmz5/v0Agj/Y/RaMSaNWt4fywJzEweM5PHzJxjzc0VVDHyNHHiRPTq1Qu1a9dGkSJFMGjQINuqM0f5+vpi/PjxOH36NLRaLZ5//nmsXbvW9n1nS5cuxcCBA9G5c2fcunULYWFhGDt2rHSt3bp1Q2ZmJmrUqAGtVot+/frhnXfesb0+Z84cfPnll/joo49w+fJlFClSBDVr1sQrr7wifS5yXIECBdxdgipxKbQ8ZiaPmcljZs5xVW6K4PUfVUlLS4O/vz+8vLyQlZXl7nJUwWAwYMGCBejcuTMyMzMf23metn9KRqMRa9euRfPmzXPMJ6TcMTN5zEweM3OONbdWrVo98rFUcdmOiIiIKK/gyJPKWEeeFEV56kY6HheNRoMSJUrg8uXL0nPlZDxtfx7WG8r5+vryBqMOYmbymJk8ZuYca25+fn6PfCyOPKnU0/ZB/ThZLBbcuHHjsTZOTytOspfHzOQxM3nMzDmuyo3Nk0pxtZ3jrHOe+MtGjslkwtq1azkxVQIzk8fM5DEz51hzcwU2T0REREQS2DwRERERSWDzRERERCSBq+1UxrrajuQYDIbHeo8n4OmbxC+EgMlkgk6n44oeBzEzecxMHjNzjjU3V9wbiyNPKsV/MI7TaDQoUqSI7W7y5LjH3XA+jZiZPGYmj5k5x1W58dNEpfR6vbtLUA29Xo8pU6YwM0kmkwmbN2/mih4JzEweM5PHzJxjzc0V2DwRERERSWDzRERERCSBzRPlC7dv33Z3Caqk0+ncXYLqMDN5zEweM3OOq3LjajuV4Wq7vIv/lIiI8geOPKkUV9s5TqPRIDIykqvtJFksFiQnJ/M7ASUwM3nMTB4zc441N1fgp4lKceWY4/R6PUaMGMHMJJnNZuzatQtms9ndpagGM5PHzOQxM+dYc3MFXjRVqUuXLqFw4cLuLkMVjEYj1q5di9TUVJfcHI2IiPI3jjwRERERSWDzpFKc8+Q4RVHg6+vLzCQxN3nMTB4zk8fMnGPNzSXH4mo7dbGutktNTYWfn5+7yyEiIsp3OPKkUlxl4TiLxYILFy4wM0nMTR4zk8fM5DEz51hzcwU2TyrFVRaOM5vNSEhIYGaSmJs8ZiaPmcljZs6x5uYKbJ6IiIiIJLB5IiIiIpLA5kmluMrCcYqiICAggJlJYm7ymJk8ZiaPmTnHmptLjsXVdurC1XZERETuxZEnleJEQceZzWacPHmSmUlibvKYmTxmJo+ZOceamyuweVIpLlF1nMViQWJiIjOTxNzkMTN5zEweM3OONTdX4HfbqVTJkiWRlZXl7jLyBF55JiKiJ4kjT0REREQS2DypFK91O06j0SAkJAQaDf+6y2Bu8piZPGYmj5k5x5qbK3C1ncpYV9vR//CvMBERPUlsW1XKw8PD3SWohtlsRnx8PEfrJDE3ecxMHjOTx8ycY83NFdg8qZRWq3V3CaphsViQlJTElSmSmJs8ZiaPmcljZs6x5uYKbJ6IiIiIJLB5IiIiIpLA5kmljEaju0tQDY1Gg/Lly3NliiTmJo+ZyWNm8piZc6y5uQJX26kMV9vlxL/CRET0JLFtVSlPT093l6AaJpMJO3fuhMlkcncpqsLc5DEzecxMHjNzjjU3V2DzpFIcrnWcEALXr1/nCJUk5iaPmcljZvKYmXOsubkCP4GJiIiIJLB5IiIiIpLA5kmluNrOcVqtFpGRkbyxqCTmJo+ZyWNm8piZc6y5uQJX26kMV9vlxL/CRET0JHHkSaX0er27S1ANk8mETZs2cWWKJOYmj5nJY2bymJlzrLm5ApsnlVIUxd0lqIYQAunp6RyhksTc5DEzecxMHjNzjjU3V2DzRERERCSBzZMTevTogdatW7u7DCIiInIDNk8qlZ2d7e4SVEOr1aJWrVpcmSKJucljZvKYmTxm5hxrbq7A5snFjh49imbNmsHHxwfFihXDG2+8gRs3bgAAtmzZAk9PT2zbts22//jx41G0aFFcu3ZN6jy81u04jUaDokWL8q7skpibPGYmj5nJY2bOsebmkmO55CgEAEhJSUHDhg0RFRWF/fv3Y/369bh27Ro6dOgAAKhfvz769++PN954A6mpqYiPj8fQoUMxa9YsFCtWLNdjZmdnIy0tze4BAF5eXk/s51I7o9GINWvW8N5YkpibPGYmj5nJY2bOsebmCjqXHIUAAFOnTkVUVBTGjBlj2zZ79mwEBwfj1KlTKFeuHL788kvExcXhnXfewdGjR9G9e3e8+uqr9z1mTEwMRo4c+STKf6pxSa9zmJs8ZiaPmcljZs5xVW4ceXKhQ4cOYfPmzfDx8bE9KlSoAAA4c+YMAMDT0xPz5s3D0qVLkZWVhUmTJj3wmEOGDEFqaqrtcfHixcf+cxAREdH9ceTJhTIyMtCyZUuMGzcux2tBQUG2/965cycA4J9//sE///wDb2/v+x5Tr9fzhphERER5CL+exQk9evRASkoKVqxYYbf9s88+w9KlS3H06FHodLn3pWfOnEFkZCQmT56MRYsW4c6dO9iwYYPDE/+sX8+iKAonjf/Xw3Kw3hjN19eXNxeVwNzkMTN5zEweM3OONTc/P79HPhYv2zkpNTUVCQkJdo933nkH//zzDzp37ox9+/bhzJkz+P3339GzZ0+YzWaYzWa8/vrraNKkCXr27Ik5c+bg8OHDmDBhgvT52TjJMRgM7i5BlZibPGYmj5nJY2bOcVVubJ6ctGXLFkRFRdk9vvjiC+zYsQNmsxkvv/wywsPD0b9/fxQsWBAajQajR4/GhQsXMGPGDAB3L+XNnDkTn3/+OQ4dOiR1fq62c5zJZMLatWs5wVISc5PHzOQxM3nMzDnW3FyBc56cEBsbi9jY2Pu+vmzZsly3Dxs2DMOGDbPb9tprr/GGl0RERCrCkSciIiIiCWyeiIiIiCRwtZ3KWFfb0f84strOZDJBp9NxZYoE5iaPmcljZvKYmXOsuXl4eDzysTjypFL8ByMnMzPT3SWoEnOTx8zkMTN5zMw5rsqNzZNK8caZjjOZTNi8eTNXpkhibvKYmTxmJo+ZOceamyuweSIiIiKSwOaJiIiISAKbJ8oX7vd1OfRgzE0eM5PHzOQxM+e4KjeutlMZrrbLiX+FiYjoSeLIk0pxtZ3jLBYLkpOTYbFY3F2KqjA3ecxMHjOTx8ycY83NFdg8qRRX2znObDZj165dMJvN7i5FVZibPGYmj5nJY2bOsebmCmyeiIiIiCSweSIiIiKSwOZJpThJ2nGKosDX15fzxCQxN3nMTB4zk8fMnGPNzSXH4mo7deFqu5z4V5iIiJ4kjjyplFardXcJqmGxWHDhwgWuTJHE3OQxM3nMTB4zc441N1dg86RSrvhW6PzCbDYjISGBK1MkMTd5zEweM5PHzJxjzc0VeItSlbp06RIKFy7s7jKIiIjyHY48EREREUlg86RSXGXhOEVREBAQwMwkMTd5zEweM5PHzJxjzc0lx+JqO3WxrrZLTU2Fn5+fu8shIiLKdzjypFKcKOg4s9mMkydPMjNJzE0eM5PHzOQxM+dYc3MFNk8qxSWqjrNYLEhMTGRmkpibPGYmj5nJY2bOsebmCmyeiIiIiCSweSIiIiKSwOZJpTQa/tE5SqPRICQkhJlJYm7ymJk8ZiaPmTnHmpsrcLWdynC1HRERkXuxbVUprrJwnNlsRnx8PDOTxNzkMTN5zEweM3OONTdXYPOkUlxl4TiLxYKkpCRmJom5yWNm8piZPGbmHGtursDmiYiIiEgCvxhYZaxT1NLT0+Hh4eHmatTBaDTi9u3bSEtLY2YSmJs8ZiaPmcljZs65NzdfX99H+nobNk8q8/fffwMASpcu7eZKiIiI1OlRF12xeVKZQoUKAQCSkpLg7+/v5mrUIS0tDcHBwbh48SJXKEpgbvKYmTxmJo+ZOefe3Hx9fR/pWGyeVMZ6Xw9/f3/+o5Hk5+fHzJzA3OQxM3nMTB4zc46fn98jXbIDOGGciIiISAqbJyIiIiIJbJ5URq/XY/jw4dDr9e4uRTWYmXOYmzxmJo+ZyWNmznFlbvx6FiIiIiIJHHkiIiIiksDmiYiIiEgCmyciIiIiCWyeiIiIiCSweVKZadOmITQ0FF5eXnjhhRewd+9ed5eUZ8XExOD555+Hr68vihYtitatWyMxMdHdZanK2LFjoSgK+vfv7+5S8rTLly/j9ddfR+HChWEwGBAeHo79+/e7u6w8zWw2Y+jQoShdujQMBgOeffZZfPHFF+Aapv/5z3/+g5YtW6J48eJQFAUrVqywe10IgWHDhiEoKAgGgwHR0dE4ffq0e4rNIx6UmdFoxKBBgxAeHg5vb28UL14c3bp1w19//SV9HjZPKrJo0SIMGDAAw4cPx8GDB1GlShU0adIEycnJ7i4tT9q6dSt69+6N3bt3Iy4uDkajES+//DJu3brl7tJUYd++fZgxYwYiIiLcXUqedvPmTdSpUwceHh5Yt24djh8/jgkTJuCZZ55xd2l52rhx4zB9+nRMnToVJ06cwLhx4zB+/HhMmTLF3aXlGbdu3UKVKlUwbdq0XF8fP348Jk+ejO+//x579uyBt7c3mjRpgqysrCdcad7xoMxu376NgwcPYujQoTh48CCWLVuGxMREvPrqq/InEqQaNWrUEL1797Y9N5vNonjx4iImJsaNValHcnKyACC2bt3q7lLyvPT0dFG2bFkRFxcn6tWrJ/r16+fukvKsQYMGibp167q7DNVp0aKF6NWrl9221157TXTt2tVNFeVtAMTy5cttzy0WiwgMDBRfffWVbVtKSorQ6/ViwYIFbqgw7/l3ZrnZu3evACAuXLggdWyOPKnEnTt3cODAAURHR9u2aTQaREdHY9euXW6sTD1SU1MB/O/Llen+evfujRYtWtj9faPcrVq1CtWrV0f79u1RtGhRREVF4YcffnB3WXle7dq1sXHjRpw6dQoAcOjQIWzfvh3NmjVzc2XqcO7cOVy9etXu36i/vz9eeOEFfiZISE1NhaIoKFiwoNT7+MXAKnHjxg2YzWYUK1bMbnuxYsVw8uRJN1WlHhaLBf3790edOnVQuXJld5eTpy1cuBAHDx7Evn373F2KKpw9exbTp0/HgAED8Omnn2Lfvn3o27cvPD090b17d3eXl2cNHjwYaWlpqFChArRaLcxmM0aPHo2uXbu6uzRVuHr1KgDk+plgfY0eLCsrC4MGDULnzp2lv2CZzRPlC71798bRo0exfft2d5eSp128eBH9+vVDXFwcvLy83F2OKlgsFlSvXh1jxowBAERFReHo0aP4/vvv2Tw9wK+//op58+Zh/vz5eO6555CQkID+/fujePHizI0eO6PRiA4dOkAIgenTp0u/n5ftVKJIkSLQarW4du2a3fZr164hMDDQTVWpQ58+fbB69Wps3rwZJUuWdHc5edqBAweQnJyMqlWrQqfTQafTYevWrZg8eTJ0Oh3MZrO7S8xzgoKCUKlSJbttFStWRFJSkpsqUoePP/4YgwcPRqdOnRAeHo433ngDH374IWJiYtxdmipYf+/zM0GetXG6cOEC4uLipEedADZPquHp6Ylq1aph48aNtm0WiwUbN25ErVq13FhZ3iWEQJ8+fbB8+XJs2rQJpUuXdndJeV6jRo1w5MgRJCQk2B7Vq1dH165dkZCQAK1W6+4S85w6derkuAXGqVOnUKpUKTdVpA63b9+GRmP/EaTVamGxWNxUkbqULl0agYGBdp8JaWlp2LNnDz8THsDaOJ0+fRobNmxA4cKFnToOL9upyIABA9C9e3dUr14dNWrUwDfffINbt26hZ8+e7i4tT+rduzfmz5+PlStXwtfX1zYPwN/fHwaDwc3V5U2+vr455oR5e3ujcOHCnCt2Hx9++CFq166NMWPGoEOHDti7dy9mzpyJmTNnuru0PK1ly5YYPXo0QkJC8NxzzyE+Ph4TJ05Er1693F1anpGRkYE///zT9vzcuXNISEhAoUKFEBISgv79++PLL79E2bJlUbp0aQwdOhTFixdH69at3Ve0mz0os6CgILRr1w4HDx7E6tWrYTabbZ8LhQoVgqenp+MncnoNILnFlClTREhIiPD09BQ1atQQu3fvdndJeRaAXB9z5sxxd2mqwlsVPNxvv/0mKleuLPR6vahQoYKYOXOmu0vK89LS0kS/fv1ESEiI8PLyEmXKlBGfffaZyM7OdndpecbmzZtz/R3WvXt3IcTd2xUMHTpUFCtWTOj1etGoUSORmJjo3qLd7EGZnTt37r6fC5s3b5Y6jyIEb+dKRERE5CjOeSIiIiKSwOaJiIiISAKbJyIiIiIJbJ6IiIiIJLB5IiIiIpLA5omIiIhIApsnIiIiIglsnoiI7jFixAgUK1YMiqJgxYoVT/S8kZGRD9ynfv366N+//xOpJ6976aWXMH/+/Ec6Rs2aNbF06VIXVUT5CZsnojyuR48eUBQFiqLA09MTYWFhGDVqFEwmk7tLe6gn3YA8qhMnTmDkyJGYMWMGrly5gmbNmrm7JDvLli3DF198YXseGhqKb775xqXn2LJlCxRFQUpKikuP60qrVq3CtWvX0KlTJ9u2AQMGoFChQggODsa8efPs9l+8eDFatmyZ4ziff/45Bg8ezO/TI2lsnohUoGnTprhy5QpOnz6Njz76CCNGjMBXX33l1LHMZjM/LO7jzJkzAIBWrVohMDAQer3eJce9c+eOS45TqFAh+Pr6uuRYT4LRaHwsx508eTJ69uxp+2Lh3377DfPnz8cff/yB8ePH46233sKNGzcAAKmpqfjss88wbdq0HMdp1qwZ0tPTsW7dusdSJz292DwRqYBer0dgYCBKlSqF9957D9HR0Vi1ahUAIDs7GwMHDkSJEiXg7e2NF154AVu2bLG9NzY2FgULFsSqVatQqVIl6PV6JCUlITs7G4MGDUJwcDD0ej3CwsLw448/2t539OhRNGvWDD4+PihWrBjeeOMN2wcScPcSUt++ffHJJ5+gUKFCCAwMxIgRI2yvh4aGAgDatGkDRVFsz8+cOYNWrVqhWLFi8PHxwfPPP48NGzbY/bxXrlxBixYtYDAYULp0acyfPz/HKEtKSgreeustBAQEwM/PDw0bNsShQ4cemOORI0fQsGFDGAwGFC5cGO+88w4yMjIA3L1sZh2d0Gg0UBQl12OYzWa8+eabKF26NAwGA8qXL49vv/3Wbp8ePXqgdevWGD16NIoXL47y5csDAC5duoTOnTujUKFC8Pb2RvXq1bFnzx679/78888IDQ2Fv78/OnXqhPT0dLvMrZft6tevjwsXLuDDDz+0jUxabd++HS+++CIMBgOCg4PRt29f3Lp1y/b6/f7sz58/jwYNGgAAnnnmGSiKgh49egDIfZQrMjLS7s9cURRMnz4dr776Kry9vTF69GgAwMqVK1G1alV4eXmhTJkyGDlypG3kVAiBESNGICQkBHq9HsWLF0ffvn1z/wMEcP36dWzatMluJOnEiROoX78+qlevjs6dO8PPzw/nzp0DAHzyySd47733EBISkuNYWq0WzZs3x8KFC+97PqLcsHkiUiGDwWAbzejTpw927dqFhQsX4vDhw2jfvj2aNm2K06dP2/a/ffs2xo0bh1mzZuHYsWMoWrQounXrhgULFmDy5Mk4ceIEZsyYAR8fHwB3G5OGDRsiKioK+/fvx/r163Ht2jV06NDBro65c+fC29sbe/bswfjx4zFq1CjExcUBAPbt2wcAmDNnDq5cuWJ7npGRgebNm2Pjxo2Ij49H06ZN0bJlSyQlJdmO261bN/z111/YsmULli5dipkzZyI5Odnu3O3bt0dycjLWrVuHAwcOoGrVqmjUqBH++eefXDO7desWmjRpgmeeeQb79u3D4sWLsWHDBvTp0wcAMHDgQMyZMwfA3ebtypUruR7HYrGgZMmSWLx4MY4fP45hw4bh008/xa+//mq338aNG5GYmIi4uDisXr0aGRkZqFevHi5fvoxVq1bh0KFD+OSTT+xGAc+cOYMVK1Zg9erVWL16NbZu3YqxY8fmWseyZctQsmRJjBo1yq7eM2fOoGnTpmjbti0OHz6MRYsWYfv27baf05pvbn/2wcHBtjlAiYmJuHLlSo7G8GFGjBiBNm3a4MiRI+jVqxe2bduGbt26oV+/fjh+/DhmzJiB2NhYW2O1dOlSTJo0CTNmzMDp06exYsUKhIeH3/f427dvR4ECBVCxYkXbtipVqmD//v24efMmDhw4gMzMTISFhWH79u04ePDgA5uxGjVqYNu2bVI/IxEew5caE5ELde/eXbRq1UoIcfdb1OPi4oRerxcDBw4UFy5cEFqtVly+fNnuPY0aNRJDhgwRQggxZ84cAUAkJCTYXk9MTBQARFxcXK7n/OKLL8TLL79st+3ixYsCgO1b2+vVqyfq1q1rt8/zzz8vBg0aZHsOQCxfvvyhP+Nzzz0npkyZIoQQ4sSJEwKA2Ldvn+3106dPCwBi0qRJQgghtm3bJvz8/ERWVpbdcZ599lkxY8aMXM8xc+ZM8cwzz4iMjAzbtjVr1giNRiOuXr0qhBBi+fLlwplfi7179xZt27a1Pe/evbsoVqyYyM7Otm2bMWOG8PX1FX///Xeuxxg+fLgoUKCASEtLs237+OOPxQsvvGB7Xq9ePdGvXz/b81KlStkysXrzzTfFO++8Y7dt27ZtQqPRiMzMzIf+2Vu/lf7mzZt223M7V5UqVcTw4cNtzwGI/v372+3TqFEjMWbMGLttP//8swgKChJCCDFhwgRRrlw5cefOnVzr+bdJkyaJMmXK5Ng+fPhw8eyzz4rKlSuLZcuWiezsbFG5cmWxf/9+MWXKFFGuXDlRu3ZtcfToUbv3rVy5Umg0GmE2mx06P5EQQujc1bQRkeNWr14NHx8fGI1GWCwWdOnSBSNGjMCWLVtgNptRrlw5u/2zs7NRuHBh23NPT09ERETYnickJECr1aJevXq5nu/QoUPYvHmzbSTqXmfOnLGd795jAkBQUFCOEaJ/y8jIwIgRI7BmzRpcuXIFJpMJmZmZtpGnxMRE6HQ6VK1a1faesLAwPPPMM3b1ZWRk2P2MAJCZmWmbt/RvJ06cQJUqVeDt7W3bVqdOHVgsFiQmJqJYsWIPrPte06ZNw+zZs5GUlITMzEzcuXMnx0q58PBweHp62p4nJCQgKioKhQoVuu9xQ0ND7eY0OZLnvx06dAiHDx+2mzQthIDFYsG5c+dw5MiRB/7ZP6rq1avnqGfHjh22kSbg7qXPrKws3L59G+3bt8c333yDMmXKoGnTpmjevDlatmwJnS73j6fMzEx4eXnl2D5ixAi7S4gjR45EdHQ0PDw88OWXX+LIkSNYvXo1unXrhgMHDtj2MxgMsFgsyM7OhsFgeMSfnvILNk9EKtCgQQNMnz4dnp6eKF68uO2DJSMjA1qtFgcOHIBWq7V7z72Nj8FgsJsT87APiYyMDLRs2RLjxo3L8VpQUJDtvz08POxeUxTloZPRBw4ciLi4OHz99dcICwuDwWBAu3btpCZVZ2RkICgoyG5ul1XBggUdPo4zFi5ciIEDB2LChAmoVasWfH198dVXX+WYu3RvkwY8PHPAuTz/LSMjA++++26ul6pCQkLw559/Sh3PSqPRQAhhty23CeH//rkzMjIwcuRIvPbaazn29fLyQnBwMBITE7FhwwbExcXh/fffx1dffYWtW7fmyAMAihQpgps3bz6w1pMnT+KXX35BfHw8Zs+ejZdeegkBAQHo0KEDevXqhfT0dFuT+s8//8Db25uNE0lh80SkAt7e3ggLC8uxPSoqCmazGcnJyXjxxRcdPl54eDgsFgu2bt2K6OjoHK9XrVoVS5cuRWho6H1HABzh4eEBs9lst23Hjh3o0aMH2rRpA+Duh+v58+dtr5cvXx4mkwnx8fGoVq0aAODPP/+0+8CsWrUqrl69Cp1OZ5uI/jAVK1ZEbGwsbt26ZfuA37FjBzQajW1CtyN27NiB2rVr4/3337dtu99o170iIiIwa9Ys/PPPPw8cfZLh6emZI9+qVavi+PHjuf59AR7+Z28dLfv3cQMCAuzmgaWlpdkmZT9I1apVkZiYeN96gLuNZcuWLdGyZUv07t0bFSpUwJEjR+xGH62ioqJw9epV3Lx502400koIgXfffRcTJ06Ej48PzGazrcmz/u+9P9vRo0cRFRX10J+D6F6cME6kYuXKlUPXrl3RrVs3LFu2DOfOncPevXsRExODNWvW3Pd9oaGh6N69O3r16oUVK1bg3Llz2LJli23Sc+/evfHPP/+gc+fO2LdvH86cOYPff/8dPXv2zPGh+iChoaHYuHGj7cMOAMqWLYtly5YhISEBhw4dQpcuXexGVypUqIDo6Gi888472Lt3L+Lj4/HOO+/YjZ5FR0ejVq1aaN26Nf744w+cP38eO3fuxGeffYb9+/fnWkvXrl3h5eWF7t274+jRo9i8eTM++OADvPHGG1KX7MqWLYv9+/fj999/x6lTpzB06FDbZPgH6dy5MwIDA9G6dWvs2LEDZ8+exdKlS7Fr1y6Hz/1voaGh+M9//oPLly/bVkIOGjQIO3fuRJ8+fZCQkIDTp09j5cqVtgnjD/uzL1WqFBRFwerVq3H9+nXbasSGDRvi559/xrZt23DkyBF07949x2hnboYNG4affvoJI0eOxLFjx3DixAksXLgQn3/+OYC7q0F//PFHHD16FGfPnsUvv/wCg8GAUqVK5Xq8qKgoFClSBDt27Mj19VmzZiEgIMC2Gq9OnTrYtGkTdu/ejUmTJqFSpUp2o5Pbtm3Dyy+/7EDaRPdw85wrInqIeyeM5+bOnTti2LBhIjQ0VHh4eIigoCDRpk0bcfjwYSHE3Qnj/v7+Od6XmZkpPvzwQxEUFCQ8PT1FWFiYmD17tu31U6dOiTZt2oiCBQsKg8EgKlSoIPr37y8sFosQIufkZSGEaNWqlejevbvt+apVq0RYWJjQ6XSiVKlSQgghzp07Jxo0aCAMBoMIDg4WU6dOzXGsv/76SzRr1kzo9XpRqlQpMX/+fFG0aFHx/fff2/ZJS0sTH3zwgShevLjw8PAQwcHBomvXriIpKem+WR0+fFg0aNBAeHl5iUKFCom3335bpKen2153ZMJ4VlaW6NGjh/D39xcFCxYU7733nhg8eLCoUqWKbZ/7/ZmdP39etG3bVvj5+YkCBQqI6tWriz179ggh7k54vvcYQtydHG3NTYicme/atUtEREQIvV5vV/fevXtF48aNhY+Pj/D29hYRERFi9OjRttcf9mc/atQoERgYKBRFsf15pqamio4dOwo/Pz8RHBwsYmNjc50wntsCgfXr14vatWsLg8Eg/Pz8RI0aNcTMmTOFEHczf+GFF4Sfn5/w9vYWNWvWFBs2bLhP+nd98sknolOnTjm2X716VZQqVSrHAoqRI0eKQoUKiQoVKtjyFkKIS5cuCQ8PD3Hx4sUHno/o3xQh/nURm4goj7l06RKCg4OxYcMGNGrUyN3lkJtdvXoVzz33HA4ePHjfESpHDBo0CDdv3sTMmTNdWB3lB5zzRER5zqZNm5CRkYHw8HBcuXIFn3zyCUJDQ/HSSy+5uzTKAwIDA/Hjjz8iKSnpkZqnokWLYsCAAS6sjPILjjwRUZ7z+++/46OPPsLZs2fh6+uL2rVr45tvvnmkD0oiIldh80REREQkgavtiIiIiCSweSIiIiKSwOaJiIiISAKbJyIiIiIJbJ6IiIiIJLB5IiIiIpLA5omIiIhIApsnIiIiIglsnoiIiIgk/D/NQdBd7/PH9wAAAABJRU5ErkJggg==",
      "text/plain": [
       "<Figure size 600x400 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "def fig_8a(architectures):\n",
    "    \"\"\"Presence of ML services in architectures\"\"\"\n",
    "    # query\n",
    "    occurences = Counter()\n",
    "    arch_with_atleast_1_ml = 0\n",
    "    for arch in architectures:\n",
    "        svcs = arch.get_all_nodes_services()\n",
    "        ml_svcs = set(filter(is_ml, svcs))\n",
    "\n",
    "        occurences.update(set(ml_svcs))\n",
    "        if ml_svcs:\n",
    "            arch_with_atleast_1_ml += 1\n",
    "\n",
    "    occurences = Counter(normalize_dict(occurences, len(architectures)))\n",
    "    arch_with_atleast_1_ml = arch_with_atleast_1_ml / len(architectures) * 100\n",
    "    print(\n",
    "        \"Percentage of architectures with at least 1 ML service:\",\n",
    "        arch_with_atleast_1_ml,\n",
    "        \"%\",\n",
    "    )\n",
    "\n",
    "    # plotting\n",
    "    fig, ax = plt.subplots(figsize=(6, 4))\n",
    "    y = np.arange(5)\n",
    "    y_names, x = zip(*reversed(occurences.most_common(5)))\n",
    "    ax.barh(y, x, color=\"k\")\n",
    "    ax.set_yticks(y)\n",
    "    ax.set_yticklabels(y_names)\n",
    "    ax.set_xlabel(\"Percentage of architectures (%)\")\n",
    "    ax.set_title(\"ML services\")\n",
    "    ax.grid(axis=\"x\", linestyle=\"--\")\n",
    "\n",
    "    plt.suptitle(\"Fig 8a\")\n",
    "\n",
    "    cleanup()\n",
    "\n",
    "\n",
    "fig_8a(architectures)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "d63e2d7b-3bb6-42a3-ad20-e275c8c4b6e6",
   "metadata": {},
   "source": [
    "The submitted version of the paper mistakenly reported that 26.5% of architectures have at least 1 ML service, instead of 18.2%. We will fix this in the final version.\n",
    "This does not affect any claim made in the paper."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "id": "a1d1b420",
   "metadata": {
    "jupyter": {
     "source_hidden": true
    }
   },
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 600x400 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "def fig_8b(architectures):\n",
    "    all_schemas = (\"fs\", \"specialized\", \"sql\", \"nosql\", \"object\")\n",
    "\n",
    "    # query\n",
    "    df = pd.DataFrame(0, index=[\"reads\", \"writes\"], columns=all_schemas)\n",
    "    n_archs_where_ml_interacts_with_storage = 0\n",
    "\n",
    "    for arch in architectures:\n",
    "        schemas_used_in_this_arch_writer = set()\n",
    "        schemas_used_in_this_arch_reader = set()\n",
    "        ml_svcs = list(filter(is_ml, arch.get_all_nodes_services()))\n",
    "        if not ml_svcs:\n",
    "            continue\n",
    "\n",
    "        for ml_svc in ml_svcs:\n",
    "            writers_svcs = set(\n",
    "                t[\"service\"]\n",
    "                for t in arch.get_outbound_nodes_for_svc(ml_svc, data_only=True)\n",
    "            )\n",
    "            readers_svcs = set(\n",
    "                t[\"service\"]\n",
    "                for t in arch.get_inbound_nodes_for_svc(ml_svc, data_only=True)\n",
    "            )\n",
    "\n",
    "            writers_svcs = filter(is_storage, writers_svcs)\n",
    "            readers_svcs = filter(is_storage, readers_svcs)\n",
    "\n",
    "            for st_svc in writers_svcs:\n",
    "                schemas_used_in_this_arch_writer.add(svc_info[st_svc][\"schema\"])\n",
    "            for st_svc in readers_svcs:\n",
    "                schemas_used_in_this_arch_reader.add(svc_info[st_svc][\"schema\"])\n",
    "\n",
    "        if (\n",
    "            not schemas_used_in_this_arch_writer\n",
    "            and not schemas_used_in_this_arch_reader\n",
    "        ):\n",
    "            continue\n",
    "        n_archs_where_ml_interacts_with_storage += 1\n",
    "\n",
    "        for schema in schemas_used_in_this_arch_writer:\n",
    "            df.loc[\"writes\", schema] += 1\n",
    "        for schema in schemas_used_in_this_arch_reader:\n",
    "            df.loc[\"reads\", schema] += 1\n",
    "\n",
    "    df = df / n_archs_where_ml_interacts_with_storage * 100\n",
    "\n",
    "    # plotting\n",
    "    y = np.arange(len(all_schemas))\n",
    "    fig, ax = plt.subplots(figsize=(6, 4))\n",
    "    width = 0.35\n",
    "\n",
    "    for i, action in enumerate(df.index):\n",
    "        ax.barh(y + width * i, df.loc[action], label=action, height=width)\n",
    "\n",
    "    ax.set_yticks(y + width / 2)\n",
    "    ax.set_yticklabels(all_schemas)\n",
    "    ax.set_xlabel(\"Percentage of architectures (%) with ML-Storage interaction\")\n",
    "    ax.set_title(\"ML services interacting with storage\")\n",
    "    ax.legend(loc=\"upper left\", bbox_to_anchor=(1, 1), handlelength=0.5)\n",
    "    ax.grid(axis=\"x\", linestyle=\"--\")\n",
    "    ax.set_xticks([0, 5, 10, 20, 30, 40, 50, 62])\n",
    "\n",
    "    plt.suptitle(\"Fig 8b\")\n",
    "\n",
    "    cleanup()\n",
    "\n",
    "\n",
    "fig_8b(architectures)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "id": "9173780d",
   "metadata": {
    "jupyter": {
     "source_hidden": true
    }
   },
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 600x400 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "def fig_8c(architectures):\n",
    "    \"\"\"Percentage of ML services reading/writing from S3\"\"\"\n",
    "    all_ml_svcs = [\n",
    "        \"SageMaker\",\n",
    "        \"Transcribe\",\n",
    "        \"Kendra\",\n",
    "        \"Comprehend\",\n",
    "        \"Rekognition\",\n",
    "        \"Translate\",\n",
    "        \"Textract\",\n",
    "        \"Polly\",\n",
    "        \"Lex\",\n",
    "        \"SageMakerGroundTruth\",\n",
    "        \"LookoutForVision\",\n",
    "    ][::-1]\n",
    "    df = pd.DataFrame(0, index=[\"reads\", \"writes\"], columns=all_ml_svcs)\n",
    "\n",
    "    # query\n",
    "    n_archs_where_ml_interacts_with_s3 = 0\n",
    "    for arch in architectures:\n",
    "        svcs_writing_to_s3 = arch.get_inbound_nodes_for_svc(\"S3\", data_only=True)\n",
    "        svcs_reading_from_s3 = arch.get_outbound_nodes_for_svc(\"S3\", data_only=True)\n",
    "\n",
    "        ml_svcs_writing_to_s3 = set(\n",
    "            t[\"service\"] for t in svcs_writing_to_s3 if is_ml(t[\"service\"])\n",
    "        )\n",
    "        ml_svcs_reading_from_s3 = set(\n",
    "            t[\"service\"] for t in svcs_reading_from_s3 if is_ml(t[\"service\"])\n",
    "        )\n",
    "\n",
    "        if not ml_svcs_writing_to_s3 and not ml_svcs_reading_from_s3:\n",
    "            continue\n",
    "        n_archs_where_ml_interacts_with_s3 += 1\n",
    "\n",
    "        for ml_svc in ml_svcs_writing_to_s3:\n",
    "            df.loc[\"writes\", ml_svc] += 1\n",
    "        for ml_svc in ml_svcs_reading_from_s3:\n",
    "            df.loc[\"reads\", ml_svc] += 1\n",
    "\n",
    "    df = df / n_archs_where_ml_interacts_with_s3 * 100\n",
    "\n",
    "    # plotting\n",
    "    fig, ax = plt.subplots(figsize=(6, 4))\n",
    "    y = np.arange(len(all_ml_svcs))\n",
    "    width = 0.35\n",
    "\n",
    "    for i, action in enumerate(df.index):\n",
    "        ax.barh(y + width * i, df.loc[action], label=action, height=width)\n",
    "\n",
    "    ax.set_yticks(y + width / 2)\n",
    "    ax.set_yticklabels(all_ml_svcs)\n",
    "    ax.set_xlabel(\"Percentage of architectures (%) with ML-S3 interaction\")\n",
    "    ax.set_title(\"ML services interacting with S3\")\n",
    "    ax.legend(loc=\"upper left\", bbox_to_anchor=(1, 1), handlelength=0.5)\n",
    "    ax.grid(axis=\"x\", linestyle=\"--\")\n",
    "\n",
    "    plt.suptitle(\"Fig 8c\")\n",
    "\n",
    "    cleanup()\n",
    "\n",
    "\n",
    "fig_8c(architectures)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "id": "257c7879",
   "metadata": {
    "jupyter": {
     "source_hidden": true
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Percentage of architectures with at least 1 analytics service: 42.05882352941177 %\n"
     ]
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 600x400 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "def fig_9a(architectures):\n",
    "    \"\"\"Presence of analytics services in architectures\"\"\"\n",
    "    # query\n",
    "    occurence = Counter()\n",
    "    n_archs_with_analytics = 0\n",
    "    for arch in architectures:\n",
    "        svcs = arch.get_all_nodes_services()\n",
    "        analytics_svcs = set(filter(is_analytics, svcs))\n",
    "\n",
    "        if analytics_svcs:\n",
    "            n_archs_with_analytics += 1\n",
    "            occurence.update(analytics_svcs)\n",
    "\n",
    "    occurence = Counter(normalize_dict(occurence, len(architectures)))\n",
    "    n_archs_with_analytics = n_archs_with_analytics / len(architectures) * 100\n",
    "\n",
    "    print(\n",
    "        \"Percentage of architectures with at least 1 analytics service:\",\n",
    "        n_archs_with_analytics,\n",
    "        \"%\",\n",
    "    )\n",
    "\n",
    "    # plotting\n",
    "    fig, ax = plt.subplots(figsize=(6, 4))\n",
    "    y = np.arange(6)\n",
    "    y_names, x = zip(*reversed(occurence.most_common(6)))\n",
    "    ax.barh(y, x, color=\"k\")\n",
    "\n",
    "    ax.set_yticks(y)\n",
    "    ax.set_yticklabels(y_names)\n",
    "    ax.set_xlabel(\"Percentage of architectures (%)\")\n",
    "    ax.set_title(\"Analytics services\")\n",
    "    ax.grid(axis=\"x\", linestyle=\"--\")\n",
    "\n",
    "    plt.suptitle(\"Fig 9a\")\n",
    "\n",
    "    cleanup()\n",
    "\n",
    "\n",
    "fig_9a(architectures)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "id": "057d5045",
   "metadata": {
    "jupyter": {
     "source_hidden": true
    }
   },
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 600x400 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "batch_svcs = [\"Athena\", \"Glue\", \"RedShift\", \"OpenSearch\", \"QuickSight\"]\n",
    "streaming_svcs = [\n",
    "    \"Kinesis\",\n",
    "    \"Firehose\",\n",
    "    \"EMR\",\n",
    "    \"MSK\",\n",
    "    \"KinesisDataStream\",\n",
    "    \"KinesisAnalytics\",\n",
    "    \"KinesisVideo\",\n",
    "    \"DataPipeline\",\n",
    "]\n",
    "\n",
    "\n",
    "def fig_9b(architectures):\n",
    "    \"\"\"Storage services read by batch analytics svcs\"\"\"\n",
    "    # query\n",
    "    n_archs_with_atleast_1_batch = 0\n",
    "    schemas_used = Counter()\n",
    "\n",
    "    for arch in architectures:\n",
    "        analytics_svcs_in_arch = set(\n",
    "            filter(lambda x: x in batch_svcs, arch.get_all_nodes_services())\n",
    "        )\n",
    "        if not analytics_svcs_in_arch:\n",
    "            continue\n",
    "        n_archs_with_atleast_1_batch += 1\n",
    "\n",
    "        schemas_used_in_this_arch = set()\n",
    "        for analytics_svc in analytics_svcs_in_arch:\n",
    "            consuming_from = arch.get_inbound_nodes_for_svc(\n",
    "                analytics_svc, data_only=True\n",
    "            )\n",
    "            consuming_from = filter(is_storage, (t[\"service\"] for t in consuming_from))\n",
    "            schemas_used_in_this_arch.update(\n",
    "                svc_info[x][\"schema\"] for x in consuming_from\n",
    "            )\n",
    "\n",
    "        schemas_used.update(schemas_used_in_this_arch)\n",
    "\n",
    "    schemas_used = Counter(normalize_dict(schemas_used, n_archs_with_atleast_1_batch))\n",
    "\n",
    "    # plotting\n",
    "    fig, ax = plt.subplots(figsize=(6, 4))\n",
    "    y = np.arange(len(schemas_used))\n",
    "    y_names, x = zip(*reversed(schemas_used.most_common()))\n",
    "    ax.barh(y, x, color=\"k\")\n",
    "    ax.set_yticks(y)\n",
    "    ax.set_yticklabels(y_names)\n",
    "    ax.set_xlabel(\"Percentage of architectures (%) with batch analytics\")\n",
    "    ax.set_title(\"Storage services read by batch analytics\")\n",
    "    ax.grid(axis=\"x\", linestyle=\"--\")\n",
    "    ax.set_xticks([0, 5, 10, 20, 30, 40, 50])\n",
    "\n",
    "    plt.suptitle(\"Fig 9b\")\n",
    "\n",
    "    cleanup()\n",
    "\n",
    "\n",
    "fig_9b(architectures)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "id": "2c543215",
   "metadata": {
    "jupyter": {
     "source_hidden": true
    }
   },
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 600x400 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "def fig_9c(architectures):\n",
    "    \"\"\"Which storage services do streaming analytics write to?\"\"\"\n",
    "    # query\n",
    "    n_archs_with_atleast_1_streaming = 0\n",
    "    schemas_used = Counter()\n",
    "\n",
    "    for arch in architectures:\n",
    "        analytics_svcs_in_arch = set(\n",
    "            filter(lambda x: x in streaming_svcs, arch.get_all_nodes_services())\n",
    "        )\n",
    "        if not analytics_svcs_in_arch:\n",
    "            continue\n",
    "        n_archs_with_atleast_1_streaming += 1\n",
    "\n",
    "        schemas_used_in_this_arch = set()\n",
    "        for analytics_svc in analytics_svcs_in_arch:\n",
    "            writing_to = arch.get_outbound_nodes_for_svc(analytics_svc, data_only=True)\n",
    "            writing_to = filter(is_storage, (t[\"service\"] for t in writing_to))\n",
    "            schemas_used_in_this_arch.update(svc_info[x][\"schema\"] for x in writing_to)\n",
    "\n",
    "        schemas_used.update(schemas_used_in_this_arch)\n",
    "\n",
    "    schemas_used = Counter(\n",
    "        normalize_dict(schemas_used, n_archs_with_atleast_1_streaming)\n",
    "    )\n",
    "\n",
    "    # plotting\n",
    "    fig, ax = plt.subplots(figsize=(6, 4))\n",
    "    y = np.arange(len(schemas_used))\n",
    "    y_names, x = zip(*reversed(schemas_used.most_common()))\n",
    "    ax.barh(y, x, color=\"k\")\n",
    "\n",
    "    ax.set_yticks(y)\n",
    "    ax.set_yticklabels(y_names)\n",
    "    ax.set_xlabel(\"Percentage of architectures (%) with streaming analytics\")\n",
    "    ax.set_title(\"Storage services written to by streaming analytics\")\n",
    "    ax.grid(axis=\"x\", linestyle=\"--\")\n",
    "    ax.set_xticks([0, 5, 10, 20, 30, 40, 52])\n",
    "\n",
    "    plt.suptitle(\"Fig 9c\")\n",
    "\n",
    "    cleanup()\n",
    "\n",
    "\n",
    "fig_9c(architectures)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "74026c39-efd7-47f8-b327-43ec17a13d06",
   "metadata": {},
   "source": [
    "The submitted version of the paper had a bug and reported 66% occurence of blob stores. \n",
    "We will update the graph with this corrected version (52.5%). \n",
    "This does not affect the claims made in the paper."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "id": "6b40f76d",
   "metadata": {
    "jupyter": {
     "source_hidden": true
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Table 3a\n",
      "Cumulative percentage of architectures with <= 0 unique storage services: 10.00%\n",
      "Cumulative percentage of architectures with <= 1 unique storage services: 47.65%\n",
      "Cumulative percentage of architectures with <= 2 unique storage services: 86.76%\n",
      "Cumulative percentage of architectures with <= 3 unique storage services: 98.24%\n",
      "Cumulative percentage of architectures with <= 4 unique storage services: 99.71%\n",
      "Cumulative percentage of architectures with <= 5 unique storage services: 100.00%\n"
     ]
    }
   ],
   "source": [
    "def table_3a(architectures):\n",
    "    \"\"\"Number of unique storage services used in each architecture\"\"\"\n",
    "    archs_with_n_unique_svcs = Counter()\n",
    "\n",
    "    for arch in architectures:\n",
    "        storage_svcs = set(filter(is_storage, arch.get_all_nodes_services()))\n",
    "        archs_with_n_unique_svcs.update([len(storage_svcs)])\n",
    "\n",
    "    archs_with_n_unique_svcs = Counter(\n",
    "        normalize_dict(archs_with_n_unique_svcs, len(architectures))\n",
    "    )\n",
    "\n",
    "    cum = 0\n",
    "    for v in sorted(archs_with_n_unique_svcs.keys()):\n",
    "        cum += archs_with_n_unique_svcs[v]\n",
    "        print(\n",
    "            \"Cumulative percentage of architectures with <=\",\n",
    "            v,\n",
    "            f\"unique storage services: {cum:.2f}%\",\n",
    "        )\n",
    "\n",
    "\n",
    "print(\"Table 3a\")\n",
    "table_3a(architectures)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "id": "ed5a22d9",
   "metadata": {
    "jupyter": {
     "source_hidden": true
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Table 3b\n",
      "             sql      nosql     object\n",
      "sql      5% (19)    6% (21)   13% (45)\n",
      "nosql    6% (21)    8% (30)   20% (70)\n",
      "object  13% (45)   20% (70)   28% (96)\n",
      "Total   25% (85)  35% (121)  62% (211)\n",
      "Number of architectures with more than 2 storage schemas: 20 (5.88%)\n"
     ]
    }
   ],
   "source": [
    "def table_3b(architectures):\n",
    "    \"\"\"Co-occurence of storage schemas\"\"\"\n",
    "    consider_schemas = [\"sql\", \"nosql\", \"object\"]\n",
    "\n",
    "    def extract_storage_schemas(arch):\n",
    "        storage_svcs = filter(is_storage, arch.get_all_nodes_services())\n",
    "        storage_schemas = set(svc_info[x][\"schema\"] for x in storage_svcs)\n",
    "        storage_schemas = [x for x in storage_schemas if x in consider_schemas]\n",
    "        return storage_schemas\n",
    "\n",
    "    df, more_than_2, _, _ = calculate_2d_cooccurence_df_plus_listofmore(\n",
    "        architectures, consider_schemas, extract_storage_schemas\n",
    "    )\n",
    "\n",
    "    print(df)\n",
    "    print(\n",
    "        \"Number of architectures with more than 2 storage schemas:\",\n",
    "        len(more_than_2),\n",
    "        f\"({len(more_than_2)/len(architectures)*100:.2f}%)\",\n",
    "    )\n",
    "\n",
    "\n",
    "print(\"Table 3b\")\n",
    "table_3b(architectures)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "id": "41b25579",
   "metadata": {
    "jupyter": {
     "source_hidden": true
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Table 3c\n",
      "Cumulative percentage of architectures with >= 4 different storage services: 0.88%\n",
      "Cumulative percentage of architectures with >= 3 different storage services: 4.71%\n",
      "Cumulative percentage of architectures with >= 2 different storage services: 35.29%\n",
      "Cumulative percentage of architectures with >= 1 different storage services: 89.12%\n",
      "Cumulative percentage of architectures with >= 0 different storage services: 100.00%\n"
     ]
    }
   ],
   "source": [
    "def table_3c(architectures):\n",
    "    \"\"\"Number of different storage services per workflow\"\"\"\n",
    "\n",
    "    arch_with_n_different_storage = Counter()\n",
    "\n",
    "    for arch in architectures:\n",
    "        different_storage_svcs_seen_per_flow = []\n",
    "        for flow in arch.iterate_flows():\n",
    "            nodes_in_flow = set()\n",
    "            for edge in flow:\n",
    "                nodes_in_flow.add(arch.nx_graph.nodes[edge[0]][\"service\"])\n",
    "                nodes_in_flow.add(arch.nx_graph.nodes[edge[1]][\"service\"])\n",
    "            storage_svcs_in_flow = len(\n",
    "                list(filter(lambda x: is_storage(x), nodes_in_flow))\n",
    "            )\n",
    "            different_storage_svcs_seen_per_flow.append(storage_svcs_in_flow)\n",
    "\n",
    "        max_interacting_svcs_seen_in_this_arch = max(\n",
    "            different_storage_svcs_seen_per_flow\n",
    "        )\n",
    "        arch_with_n_different_storage.update([max_interacting_svcs_seen_in_this_arch])\n",
    "\n",
    "    arch_with_n_different_storage = Counter(\n",
    "        normalize_dict(arch_with_n_different_storage, len(architectures))\n",
    "    )\n",
    "\n",
    "    # plotting\n",
    "    cum = 0\n",
    "    for v in sorted(arch_with_n_different_storage.keys(), reverse=True):\n",
    "        cum += arch_with_n_different_storage[v]\n",
    "        print(\n",
    "            \"Cumulative percentage of architectures with >=\",\n",
    "            v,\n",
    "            f\"different storage services: {cum:.2f}%\",\n",
    "        )\n",
    "\n",
    "\n",
    "print(\"Table 3c\")\n",
    "table_3c(architectures)"
   ]
  }
 ],
 "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": "3.11.7"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}