analysis/dataset_sample_analysis.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "id": "initial_id",
   "metadata": {
    "collapsed": true,
    "ExecuteTime": {
     "end_time": "2024-03-18T23:07:17.347945Z",
     "start_time": "2024-03-18T23:07:17.085142Z"
    }
   },
   "outputs": [],
   "source": [
    "import json\n",
    "import re\n",
    "from pathlib import Path\n",
    "\n",
    "import matplotlib\n",
    "import matplotlib.pyplot as plt\n",
    "import pandas as pd\n",
    "import seaborn as sns\n",
    "from datasets import load_dataset\n",
    "from matplotlib import ticker"
   ]
  },
  {
   "cell_type": "code",
   "outputs": [],
   "source": [
    "from tqdm.auto import tqdm\n",
    "\n",
    "tqdm.pandas()\n",
    "matplotlib.rc_file_defaults()"
   ],
   "metadata": {
    "collapsed": false,
    "ExecuteTime": {
     "end_time": "2024-03-18T23:07:17.351668Z",
     "start_time": "2024-03-18T23:07:17.349011Z"
    }
   },
   "id": "d2f842274f3582a4",
   "execution_count": 2
  },
  {
   "cell_type": "code",
   "outputs": [
    {
     "data": {
      "text/plain": "Resolving data files:   0%|          | 0/17 [00:00<?, ?it/s]",
      "application/vnd.jupyter.widget-view+json": {
       "version_major": 2,
       "version_minor": 0,
       "model_id": "67881a008e4c4a4781f566d18df32976"
      }
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/plain": "Resolving data files:   0%|          | 0/17 [00:00<?, ?it/s]",
      "application/vnd.jupyter.widget-view+json": {
       "version_major": 2,
       "version_minor": 0,
       "model_id": "7ae1ab4c271f4b9682f69189b00b39b3"
      }
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/plain": "  0%|          | 0/200 [00:00<?, ?it/s]",
      "application/vnd.jupyter.widget-view+json": {
       "version_major": 2,
       "version_minor": 0,
       "model_id": "bc638b2433134b9c8136b357f3728f0a"
      }
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/plain": "            repo_name           repo_owner  \\\nid                                           \n0   finance-complaint  Machine-Learning-01   \n1           langchain         langchain-ai   \n2     deep_prediction          sapan-ostic   \n3     cv-ferattn-code           HelenGuohx   \n4      diseno_sci_sfw             leliel12   \n\n                                            file_link  \\\nid                                                      \n0   https://github.com/Machine-Learning-01/finance...   \n1   https://github.com/langchain-ai/langchain/blob...   \n2   https://github.com/sapan-ostic/deep_prediction...   \n3   https://github.com/HelenGuohx/cv-ferattn-code/...   \n4   https://github.com/leliel12/diseno_sci_sfw/blo...   \n\n                                            line_link  \\\nid                                                      \n0   https://github.com/Machine-Learning-01/finance...   \n1   https://github.com/langchain-ai/langchain/blob...   \n2   https://github.com/sapan-ostic/deep_prediction...   \n3   https://github.com/HelenGuohx/cv-ferattn-code/...   \n4   https://github.com/leliel12/diseno_sci_sfw/blo...   \n\n                                                 path  \\\nid                                                      \n0                            notebook/Untitled1.ipynb   \n1   docs/extras/modules/model_io/output_parsers/en...   \n2   scripts/.ipynb_checkpoints/test_argo-checkpoin...   \n3                   fervideo/Facial_recognition.ipynb   \n4     00_antecedentes/02_niveles_de_abstraccion.ipynb   \n\n                                          content_sha  \\\nid                                                      \n0   d12c58483c42f93f58d6943065e34ed0a636d6a5ae1732...   \n1   e515f22c581952d6cb0b36104d398722c5186e06e301b4...   \n2   7736c22796f980a4998a16ec0eb26d703d829be1d0c2ab...   \n3   881e69a1e530676b4a28e425af897c09e8ebcc8037fc46...   \n4   b0c26856e090641929400716e6906670c5fde357f3d560...   \n\n                                              content  \nid                                                     \n0   {\\n \"cells\": [\\n  {\\n   \"cell_type\": \"code\",\\n...  \n1   {\\n \"cells\": [\\n  {\\n   \"cell_type\": \"markdown...  \n2   {\\n \"cells\": [\\n  {\\n   \"cell_type\": \"code\",\\n...  \n3   {\\n \"cells\": [\\n  {\\n   \"cell_type\": \"markdown...  \n4   {\\n \"cells\": [\\n  {\\n   \"attachments\": {\\n    ...  ",
      "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>repo_name</th>\n      <th>repo_owner</th>\n      <th>file_link</th>\n      <th>line_link</th>\n      <th>path</th>\n      <th>content_sha</th>\n      <th>content</th>\n    </tr>\n    <tr>\n      <th>id</th>\n      <th></th>\n      <th></th>\n      <th></th>\n      <th></th>\n      <th></th>\n      <th></th>\n      <th></th>\n    </tr>\n  </thead>\n  <tbody>\n    <tr>\n      <th>0</th>\n      <td>finance-complaint</td>\n      <td>Machine-Learning-01</td>\n      <td>https://github.com/Machine-Learning-01/finance...</td>\n      <td>https://github.com/Machine-Learning-01/finance...</td>\n      <td>notebook/Untitled1.ipynb</td>\n      <td>d12c58483c42f93f58d6943065e34ed0a636d6a5ae1732...</td>\n      <td>{\\n \"cells\": [\\n  {\\n   \"cell_type\": \"code\",\\n...</td>\n    </tr>\n    <tr>\n      <th>1</th>\n      <td>langchain</td>\n      <td>langchain-ai</td>\n      <td>https://github.com/langchain-ai/langchain/blob...</td>\n      <td>https://github.com/langchain-ai/langchain/blob...</td>\n      <td>docs/extras/modules/model_io/output_parsers/en...</td>\n      <td>e515f22c581952d6cb0b36104d398722c5186e06e301b4...</td>\n      <td>{\\n \"cells\": [\\n  {\\n   \"cell_type\": \"markdown...</td>\n    </tr>\n    <tr>\n      <th>2</th>\n      <td>deep_prediction</td>\n      <td>sapan-ostic</td>\n      <td>https://github.com/sapan-ostic/deep_prediction...</td>\n      <td>https://github.com/sapan-ostic/deep_prediction...</td>\n      <td>scripts/.ipynb_checkpoints/test_argo-checkpoin...</td>\n      <td>7736c22796f980a4998a16ec0eb26d703d829be1d0c2ab...</td>\n      <td>{\\n \"cells\": [\\n  {\\n   \"cell_type\": \"code\",\\n...</td>\n    </tr>\n    <tr>\n      <th>3</th>\n      <td>cv-ferattn-code</td>\n      <td>HelenGuohx</td>\n      <td>https://github.com/HelenGuohx/cv-ferattn-code/...</td>\n      <td>https://github.com/HelenGuohx/cv-ferattn-code/...</td>\n      <td>fervideo/Facial_recognition.ipynb</td>\n      <td>881e69a1e530676b4a28e425af897c09e8ebcc8037fc46...</td>\n      <td>{\\n \"cells\": [\\n  {\\n   \"cell_type\": \"markdown...</td>\n    </tr>\n    <tr>\n      <th>4</th>\n      <td>diseno_sci_sfw</td>\n      <td>leliel12</td>\n      <td>https://github.com/leliel12/diseno_sci_sfw/blo...</td>\n      <td>https://github.com/leliel12/diseno_sci_sfw/blo...</td>\n      <td>00_antecedentes/02_niveles_de_abstraccion.ipynb</td>\n      <td>b0c26856e090641929400716e6906670c5fde357f3d560...</td>\n      <td>{\\n \"cells\": [\\n  {\\n   \"attachments\": {\\n    ...</td>\n    </tr>\n  </tbody>\n</table>\n</div>"
     },
     "execution_count": 3,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "dataset = iter(load_dataset(\"JetBrains-Research/jupyter-errors-dataset\", split=\"train\", streaming=True))\n",
    "num_examples: int = 200\n",
    "df = pd.DataFrame([next(dataset) for _ in tqdm(range(num_examples))]).set_index(\"id\")\n",
    "df.head()"
   ],
   "metadata": {
    "collapsed": false,
    "ExecuteTime": {
     "end_time": "2024-03-18T23:07:39.988248Z",
     "start_time": "2024-03-18T23:07:17.352440Z"
    }
   },
   "id": "eedda3bf278d144c",
   "execution_count": 3
  },
  {
   "cell_type": "code",
   "outputs": [],
   "source": [
    "def safe_json_loads(json_str):\n",
    "    try:\n",
    "        return json.loads(json_str)\n",
    "    except json.JSONDecodeError:\n",
    "        return {}\n",
    "\n",
    "\n",
    "def contains_error(output_list):\n",
    "    for output in output_list:\n",
    "        if isinstance(output, dict) and output.get(\"output_type\") == \"error\":\n",
    "            return True\n",
    "    return False"
   ],
   "metadata": {
    "collapsed": false,
    "ExecuteTime": {
     "end_time": "2024-03-18T23:07:39.992606Z",
     "start_time": "2024-03-18T23:07:39.990064Z"
    }
   },
   "id": "8945be5c02ee26ba",
   "execution_count": 4
  },
  {
   "cell_type": "code",
   "outputs": [
    {
     "data": {
      "text/plain": "  0%|          | 0/200 [00:00<?, ?it/s]",
      "application/vnd.jupyter.widget-view+json": {
       "version_major": 2,
       "version_minor": 0,
       "model_id": "e7f7c88d732740239ad5fbde362c021e"
      }
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "df[\"json_column\"] = df[\"content\"].progress_apply(safe_json_loads)\n",
    "df[\"cells\"] = df[\"json_column\"].apply(lambda row: row.get(\"cells\", []))\n",
    "\n",
    "df_cells = df.explode(\"cells\")\n",
    "df_cells[\"outputs\"] = df_cells[\"cells\"].apply(\n",
    "    lambda cell: cell.get(\"outputs\", None) if isinstance(cell, dict) else None\n",
    ")"
   ],
   "metadata": {
    "collapsed": false,
    "ExecuteTime": {
     "end_time": "2024-03-18T23:07:40.333215Z",
     "start_time": "2024-03-18T23:07:39.993376Z"
    }
   },
   "id": "fe2fa7b0cdda59ee",
   "execution_count": 5
  },
  {
   "cell_type": "code",
   "outputs": [],
   "source": [
    "df_outputs = df_cells[df_cells[\"outputs\"].fillna(\"\").apply(len) > 0]\n",
    "df_outputs.loc[:, [\"has_error\"]] = df_outputs[\"outputs\"].apply(contains_error)"
   ],
   "metadata": {
    "collapsed": false,
    "ExecuteTime": {
     "end_time": "2024-03-18T23:07:40.341731Z",
     "start_time": "2024-03-18T23:07:40.333962Z"
    }
   },
   "id": "dae569e7c4c11201",
   "execution_count": 6
  },
  {
   "cell_type": "code",
   "outputs": [],
   "source": [
    "df_errors = df_outputs[df_outputs[\"has_error\"]]\n",
    "df_errors = df_errors.drop(columns=[\"content\", \"json_column\", \"cells\", \"has_error\"])\n",
    "df_errors = df_errors.explode(\"outputs\")\n",
    "\n",
    "df_errors = df_errors[df_errors[\"outputs\"].apply(lambda output: output.get(\"output_type\") == \"error\")]\n",
    "\n",
    "df_errors.loc[:, [\"ename\"]] = df_errors[\"outputs\"].apply(lambda output: output.get(\"ename\"))\n",
    "df_errors.loc[:, [\"traceback\"]] = df_errors[\"outputs\"].apply(lambda output: output.get(\"traceback\"))"
   ],
   "metadata": {
    "collapsed": false,
    "ExecuteTime": {
     "end_time": "2024-03-18T23:07:40.348924Z",
     "start_time": "2024-03-18T23:07:40.342668Z"
    }
   },
   "id": "dd009571e356f7e4",
   "execution_count": 7
  },
  {
   "cell_type": "code",
   "outputs": [],
   "source": [
    "del df\n",
    "del df_cells\n",
    "del df_outputs"
   ],
   "metadata": {
    "collapsed": false,
    "ExecuteTime": {
     "end_time": "2024-03-18T23:07:40.363213Z",
     "start_time": "2024-03-18T23:07:40.350221Z"
    }
   },
   "id": "f48c816ec6121278",
   "execution_count": 8
  },
  {
   "cell_type": "code",
   "outputs": [
    {
     "data": {
      "text/plain": "            repo_name           repo_owner  \\\nid                                           \n0   finance-complaint  Machine-Learning-01   \n1           langchain         langchain-ai   \n2     deep_prediction          sapan-ostic   \n3     cv-ferattn-code           HelenGuohx   \n4      diseno_sci_sfw             leliel12   \n\n                                            file_link  \\\nid                                                      \n0   https://github.com/Machine-Learning-01/finance...   \n1   https://github.com/langchain-ai/langchain/blob...   \n2   https://github.com/sapan-ostic/deep_prediction...   \n3   https://github.com/HelenGuohx/cv-ferattn-code/...   \n4   https://github.com/leliel12/diseno_sci_sfw/blo...   \n\n                                            line_link  \\\nid                                                      \n0   https://github.com/Machine-Learning-01/finance...   \n1   https://github.com/langchain-ai/langchain/blob...   \n2   https://github.com/sapan-ostic/deep_prediction...   \n3   https://github.com/HelenGuohx/cv-ferattn-code/...   \n4   https://github.com/leliel12/diseno_sci_sfw/blo...   \n\n                                                 path  \\\nid                                                      \n0                            notebook/Untitled1.ipynb   \n1   docs/extras/modules/model_io/output_parsers/en...   \n2   scripts/.ipynb_checkpoints/test_argo-checkpoin...   \n3                   fervideo/Facial_recognition.ipynb   \n4     00_antecedentes/02_niveles_de_abstraccion.ipynb   \n\n                                          content_sha  \\\nid                                                      \n0   d12c58483c42f93f58d6943065e34ed0a636d6a5ae1732...   \n1   e515f22c581952d6cb0b36104d398722c5186e06e301b4...   \n2   7736c22796f980a4998a16ec0eb26d703d829be1d0c2ab...   \n3   881e69a1e530676b4a28e425af897c09e8ebcc8037fc46...   \n4   b0c26856e090641929400716e6906670c5fde357f3d560...   \n\n                                              outputs                  ename  \\\nid                                                                             \n0   {'ename': 'TypeError', 'evalue': ''range' obje...              TypeError   \n1   {'ename': 'OutputParserException', 'evalue': '...  OutputParserException   \n2   {'ename': 'ValueError', 'evalue': 'operands co...             ValueError   \n3   {'ename': 'TypeError', 'evalue': ''NoneType' o...              TypeError   \n4   {'ename': 'TypeError', 'evalue': 'unsupported ...              TypeError   \n\n                                            traceback  \nid                                                     \n0   [\u001B[0;31m--------------------------------------...  \n1   [\u001B[0;31m--------------------------------------...  \n2   [\u001B[0;31m--------------------------------------...  \n3   [\u001B[0;31m--------------------------------------...  \n4   [\u001B[0;31m--------------------------------------...  ",
      "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>repo_name</th>\n      <th>repo_owner</th>\n      <th>file_link</th>\n      <th>line_link</th>\n      <th>path</th>\n      <th>content_sha</th>\n      <th>outputs</th>\n      <th>ename</th>\n      <th>traceback</th>\n    </tr>\n    <tr>\n      <th>id</th>\n      <th></th>\n      <th></th>\n      <th></th>\n      <th></th>\n      <th></th>\n      <th></th>\n      <th></th>\n      <th></th>\n      <th></th>\n    </tr>\n  </thead>\n  <tbody>\n    <tr>\n      <th>0</th>\n      <td>finance-complaint</td>\n      <td>Machine-Learning-01</td>\n      <td>https://github.com/Machine-Learning-01/finance...</td>\n      <td>https://github.com/Machine-Learning-01/finance...</td>\n      <td>notebook/Untitled1.ipynb</td>\n      <td>d12c58483c42f93f58d6943065e34ed0a636d6a5ae1732...</td>\n      <td>{'ename': 'TypeError', 'evalue': ''range' obje...</td>\n      <td>TypeError</td>\n      <td>[\u001B[0;31m--------------------------------------...</td>\n    </tr>\n    <tr>\n      <th>1</th>\n      <td>langchain</td>\n      <td>langchain-ai</td>\n      <td>https://github.com/langchain-ai/langchain/blob...</td>\n      <td>https://github.com/langchain-ai/langchain/blob...</td>\n      <td>docs/extras/modules/model_io/output_parsers/en...</td>\n      <td>e515f22c581952d6cb0b36104d398722c5186e06e301b4...</td>\n      <td>{'ename': 'OutputParserException', 'evalue': '...</td>\n      <td>OutputParserException</td>\n      <td>[\u001B[0;31m--------------------------------------...</td>\n    </tr>\n    <tr>\n      <th>2</th>\n      <td>deep_prediction</td>\n      <td>sapan-ostic</td>\n      <td>https://github.com/sapan-ostic/deep_prediction...</td>\n      <td>https://github.com/sapan-ostic/deep_prediction...</td>\n      <td>scripts/.ipynb_checkpoints/test_argo-checkpoin...</td>\n      <td>7736c22796f980a4998a16ec0eb26d703d829be1d0c2ab...</td>\n      <td>{'ename': 'ValueError', 'evalue': 'operands co...</td>\n      <td>ValueError</td>\n      <td>[\u001B[0;31m--------------------------------------...</td>\n    </tr>\n    <tr>\n      <th>3</th>\n      <td>cv-ferattn-code</td>\n      <td>HelenGuohx</td>\n      <td>https://github.com/HelenGuohx/cv-ferattn-code/...</td>\n      <td>https://github.com/HelenGuohx/cv-ferattn-code/...</td>\n      <td>fervideo/Facial_recognition.ipynb</td>\n      <td>881e69a1e530676b4a28e425af897c09e8ebcc8037fc46...</td>\n      <td>{'ename': 'TypeError', 'evalue': ''NoneType' o...</td>\n      <td>TypeError</td>\n      <td>[\u001B[0;31m--------------------------------------...</td>\n    </tr>\n    <tr>\n      <th>4</th>\n      <td>diseno_sci_sfw</td>\n      <td>leliel12</td>\n      <td>https://github.com/leliel12/diseno_sci_sfw/blo...</td>\n      <td>https://github.com/leliel12/diseno_sci_sfw/blo...</td>\n      <td>00_antecedentes/02_niveles_de_abstraccion.ipynb</td>\n      <td>b0c26856e090641929400716e6906670c5fde357f3d560...</td>\n      <td>{'ename': 'TypeError', 'evalue': 'unsupported ...</td>\n      <td>TypeError</td>\n      <td>[\u001B[0;31m--------------------------------------...</td>\n    </tr>\n  </tbody>\n</table>\n</div>"
     },
     "execution_count": 9,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_errors.head()"
   ],
   "metadata": {
    "collapsed": false,
    "ExecuteTime": {
     "end_time": "2024-03-18T23:07:40.371082Z",
     "start_time": "2024-03-18T23:07:40.364129Z"
    }
   },
   "id": "62e3e649942070bf",
   "execution_count": 9
  },
  {
   "cell_type": "code",
   "outputs": [
    {
     "data": {
      "text/plain": "ename\nNameError              0.252525\nTypeError              0.161616\nAttributeError         0.124579\nValueError             0.121212\nKeyError               0.050505\nIndexError             0.033670\nModuleNotFoundError    0.026936\nFileNotFoundError      0.023569\nRuntimeError           0.020202\nImportError            0.016835\nName: proportion, dtype: float64"
     },
     "execution_count": 10,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "error_counts = df_errors.ename.value_counts(normalize=True)\n",
    "error_counts[:10]"
   ],
   "metadata": {
    "collapsed": false,
    "ExecuteTime": {
     "end_time": "2024-03-18T23:07:40.377111Z",
     "start_time": "2024-03-18T23:07:40.373462Z"
    }
   },
   "id": "c0d62ea14bce97ff",
   "execution_count": 10
  },
  {
   "cell_type": "code",
   "outputs": [
    {
     "data": {
      "text/plain": "<Figure size 500x500 with 1 Axes>",
      "image/png": "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"
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "s: pd.Series = error_counts\n",
    "top_num: int = 8\n",
    "\n",
    "top8, other = s[:top_num], pd.Series([s[top_num:].sum()], index=[\"Other\"])\n",
    "data = pd.concat([top8, other])\n",
    "\n",
    "edge_colors = [\"green\"] * len(top8)\n",
    "edge_colors.append(\"black\")\n",
    "\n",
    "fig, ax = plt.subplots(figsize=(5, 5))\n",
    "wedges, text, auto_texts = ax.pie(\n",
    "    data, labels=data.index, startangle=250, colors=sns.color_palette(\"Set3\", 12), autopct=\"%1.0f%%\"\n",
    ")\n",
    "\n",
    "for i, (w, t, at) in enumerate(zip(wedges, text, auto_texts)):\n",
    "    group = data.index[i]\n",
    "    lw, fs = 2, 16\n",
    "\n",
    "    z_order = 1 if group == \"Other\" else 2\n",
    "    edge_color = (0, 0, 0, 0.5) if group == \"Other\" else \"black\"\n",
    "    cell_fontsize = 16 if group == \"Other\" else 0\n",
    "\n",
    "    w.set_edgecolor(edge_color)\n",
    "    w.set_zorder(z_order)\n",
    "    w.set_linewidth(lw)\n",
    "    t.set_fontsize(fs)\n",
    "    at.set_fontsize(cell_fontsize)\n",
    "\n",
    "del auto_texts[1]\n",
    "\n",
    "figures_path = Path(\"figures/\")\n",
    "fig.savefig(\n",
    "    figures_path / \"errors_distribution.pdf\",\n",
    "    bbox_inches=\"tight\",\n",
    "    dpi=300,\n",
    ")\n",
    "\n",
    "plt.show()"
   ],
   "metadata": {
    "collapsed": false,
    "ExecuteTime": {
     "end_time": "2024-03-18T23:07:40.630152Z",
     "start_time": "2024-03-18T23:07:40.377825Z"
    }
   },
   "id": "ad9d1a7fd579b87d",
   "execution_count": 11
  },
  {
   "cell_type": "code",
   "outputs": [],
   "source": [
    "def parse_location(error: list[str]) -> str:\n",
    "    patterns = [\n",
    "        re.compile(r\"\\x1b\\[1;32m(.*?)\\x1b\\[0m\"),\n",
    "        re.compile(r\"\\x1b\\[0;32m(.*?)\\x1b\\[0m\"),\n",
    "    ]\n",
    "\n",
    "    err_locations = []\n",
    "    for line in error:\n",
    "        for pattern in patterns:\n",
    "            if loc := pattern.findall(line):\n",
    "                err_locations.append(loc[0])\n",
    "\n",
    "    loc = None if not err_locations else err_locations[-1]\n",
    "    internal_flags = {\"<ipython\", \"In[\", \"In [\"}\n",
    "    return \"internal\" if loc and any(flag in loc for flag in internal_flags) else \"external\""
   ],
   "metadata": {
    "collapsed": false,
    "ExecuteTime": {
     "end_time": "2024-03-18T23:07:40.635473Z",
     "start_time": "2024-03-18T23:07:40.631374Z"
    }
   },
   "id": "1e50b089296dc10a",
   "execution_count": 12
  },
  {
   "cell_type": "code",
   "outputs": [
    {
     "data": {
      "text/plain": "  0%|          | 0/297 [00:00<?, ?it/s]",
      "application/vnd.jupyter.widget-view+json": {
       "version_major": 2,
       "version_minor": 0,
       "model_id": "4ac5ff18f7b34445ab304e383d544483"
      }
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "df_errors[\"location\"] = df_errors.traceback.progress_apply(\n",
    "    lambda x: parse_location(eval(x) if isinstance(x, str) else x)\n",
    ")"
   ],
   "metadata": {
    "collapsed": false,
    "ExecuteTime": {
     "end_time": "2024-03-18T23:07:40.658361Z",
     "start_time": "2024-03-18T23:07:40.638469Z"
    }
   },
   "id": "72b2bf0a0f610f1d",
   "execution_count": 13
  },
  {
   "cell_type": "code",
   "outputs": [
    {
     "data": {
      "text/plain": "            repo_name           repo_owner  \\\nid                                           \n0   finance-complaint  Machine-Learning-01   \n1           langchain         langchain-ai   \n2     deep_prediction          sapan-ostic   \n3     cv-ferattn-code           HelenGuohx   \n4      diseno_sci_sfw             leliel12   \n\n                                            file_link  \\\nid                                                      \n0   https://github.com/Machine-Learning-01/finance...   \n1   https://github.com/langchain-ai/langchain/blob...   \n2   https://github.com/sapan-ostic/deep_prediction...   \n3   https://github.com/HelenGuohx/cv-ferattn-code/...   \n4   https://github.com/leliel12/diseno_sci_sfw/blo...   \n\n                                            line_link  \\\nid                                                      \n0   https://github.com/Machine-Learning-01/finance...   \n1   https://github.com/langchain-ai/langchain/blob...   \n2   https://github.com/sapan-ostic/deep_prediction...   \n3   https://github.com/HelenGuohx/cv-ferattn-code/...   \n4   https://github.com/leliel12/diseno_sci_sfw/blo...   \n\n                                                 path  \\\nid                                                      \n0                            notebook/Untitled1.ipynb   \n1   docs/extras/modules/model_io/output_parsers/en...   \n2   scripts/.ipynb_checkpoints/test_argo-checkpoin...   \n3                   fervideo/Facial_recognition.ipynb   \n4     00_antecedentes/02_niveles_de_abstraccion.ipynb   \n\n                                          content_sha  \\\nid                                                      \n0   d12c58483c42f93f58d6943065e34ed0a636d6a5ae1732...   \n1   e515f22c581952d6cb0b36104d398722c5186e06e301b4...   \n2   7736c22796f980a4998a16ec0eb26d703d829be1d0c2ab...   \n3   881e69a1e530676b4a28e425af897c09e8ebcc8037fc46...   \n4   b0c26856e090641929400716e6906670c5fde357f3d560...   \n\n                                              outputs                  ename  \\\nid                                                                             \n0   {'ename': 'TypeError', 'evalue': ''range' obje...              TypeError   \n1   {'ename': 'OutputParserException', 'evalue': '...  OutputParserException   \n2   {'ename': 'ValueError', 'evalue': 'operands co...             ValueError   \n3   {'ename': 'TypeError', 'evalue': ''NoneType' o...              TypeError   \n4   {'ename': 'TypeError', 'evalue': 'unsupported ...              TypeError   \n\n                                            traceback  location  \nid                                                               \n0   [\u001B[0;31m--------------------------------------...  internal  \n1   [\u001B[0;31m--------------------------------------...  external  \n2   [\u001B[0;31m--------------------------------------...  internal  \n3   [\u001B[0;31m--------------------------------------...  external  \n4   [\u001B[0;31m--------------------------------------...  internal  ",
      "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>repo_name</th>\n      <th>repo_owner</th>\n      <th>file_link</th>\n      <th>line_link</th>\n      <th>path</th>\n      <th>content_sha</th>\n      <th>outputs</th>\n      <th>ename</th>\n      <th>traceback</th>\n      <th>location</th>\n    </tr>\n    <tr>\n      <th>id</th>\n      <th></th>\n      <th></th>\n      <th></th>\n      <th></th>\n      <th></th>\n      <th></th>\n      <th></th>\n      <th></th>\n      <th></th>\n      <th></th>\n    </tr>\n  </thead>\n  <tbody>\n    <tr>\n      <th>0</th>\n      <td>finance-complaint</td>\n      <td>Machine-Learning-01</td>\n      <td>https://github.com/Machine-Learning-01/finance...</td>\n      <td>https://github.com/Machine-Learning-01/finance...</td>\n      <td>notebook/Untitled1.ipynb</td>\n      <td>d12c58483c42f93f58d6943065e34ed0a636d6a5ae1732...</td>\n      <td>{'ename': 'TypeError', 'evalue': ''range' obje...</td>\n      <td>TypeError</td>\n      <td>[\u001B[0;31m--------------------------------------...</td>\n      <td>internal</td>\n    </tr>\n    <tr>\n      <th>1</th>\n      <td>langchain</td>\n      <td>langchain-ai</td>\n      <td>https://github.com/langchain-ai/langchain/blob...</td>\n      <td>https://github.com/langchain-ai/langchain/blob...</td>\n      <td>docs/extras/modules/model_io/output_parsers/en...</td>\n      <td>e515f22c581952d6cb0b36104d398722c5186e06e301b4...</td>\n      <td>{'ename': 'OutputParserException', 'evalue': '...</td>\n      <td>OutputParserException</td>\n      <td>[\u001B[0;31m--------------------------------------...</td>\n      <td>external</td>\n    </tr>\n    <tr>\n      <th>2</th>\n      <td>deep_prediction</td>\n      <td>sapan-ostic</td>\n      <td>https://github.com/sapan-ostic/deep_prediction...</td>\n      <td>https://github.com/sapan-ostic/deep_prediction...</td>\n      <td>scripts/.ipynb_checkpoints/test_argo-checkpoin...</td>\n      <td>7736c22796f980a4998a16ec0eb26d703d829be1d0c2ab...</td>\n      <td>{'ename': 'ValueError', 'evalue': 'operands co...</td>\n      <td>ValueError</td>\n      <td>[\u001B[0;31m--------------------------------------...</td>\n      <td>internal</td>\n    </tr>\n    <tr>\n      <th>3</th>\n      <td>cv-ferattn-code</td>\n      <td>HelenGuohx</td>\n      <td>https://github.com/HelenGuohx/cv-ferattn-code/...</td>\n      <td>https://github.com/HelenGuohx/cv-ferattn-code/...</td>\n      <td>fervideo/Facial_recognition.ipynb</td>\n      <td>881e69a1e530676b4a28e425af897c09e8ebcc8037fc46...</td>\n      <td>{'ename': 'TypeError', 'evalue': ''NoneType' o...</td>\n      <td>TypeError</td>\n      <td>[\u001B[0;31m--------------------------------------...</td>\n      <td>external</td>\n    </tr>\n    <tr>\n      <th>4</th>\n      <td>diseno_sci_sfw</td>\n      <td>leliel12</td>\n      <td>https://github.com/leliel12/diseno_sci_sfw/blo...</td>\n      <td>https://github.com/leliel12/diseno_sci_sfw/blo...</td>\n      <td>00_antecedentes/02_niveles_de_abstraccion.ipynb</td>\n      <td>b0c26856e090641929400716e6906670c5fde357f3d560...</td>\n      <td>{'ename': 'TypeError', 'evalue': 'unsupported ...</td>\n      <td>TypeError</td>\n      <td>[\u001B[0;31m--------------------------------------...</td>\n      <td>internal</td>\n    </tr>\n  </tbody>\n</table>\n</div>"
     },
     "execution_count": 14,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_errors.head()"
   ],
   "metadata": {
    "collapsed": false,
    "ExecuteTime": {
     "end_time": "2024-03-18T23:07:40.679442Z",
     "start_time": "2024-03-18T23:07:40.666503Z"
    }
   },
   "id": "ed951b7c8653573d",
   "execution_count": 14
  },
  {
   "cell_type": "code",
   "outputs": [
    {
     "data": {
      "text/plain": "<Figure size 600x300 with 1 Axes>",
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAuUAAAF+CAYAAAAlc2TjAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjguMywgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy/H5lhTAAAACXBIWXMAAA9hAAAPYQGoP6dpAACS6klEQVR4nOzdd3xN9/8H8PfJzY5MGbIkBLEFsffeW+2qVbtqV1UVNYqi9t5qVc2i9qZGUZuqnYSIERIRkvv6/ZHfOd97JZRm3CRez8ejj8qZn3Pvuee8z+e8P5+PAgBCREREREQmY2bqAhARERERfewYlBMRERERmRiDciIiIiIiE2NQTkRERERkYgzKiYiIiIhMjEE5EREREZGJMSgnIiIiIjIxc1MXgIiIiCgtvH79Wl6/fm3qYtBHwMLCQiwsLD5oHQblRERElKlFRUVJaGioPH/+3NRFoY+Ivb29eHl5SZYsWd5reYUjehIREVFmFRsbK5cvXxZLS0vx8PAQa2trURTF1MWiTAyAvHz5Uh48eCCvXr2SfPnyiZWV1b+ux6CciIiIMq1//vlHXrx4Ifny5RNzcyYIUNqJi4uTy5cvi52dneTMmfNfl2dDTyIiIsqUAMjz588la9asDMgpzZmbm0vWrFnl2bNn8j514AzKiYiIKFOKjY2V+Pj4987pJUppWbJkkfj4eImNjf3XZRmUExERUaak1+tFRESn05m4JPSxUs899Vx8FwblRERElKmxYSeZyoecewzKiYiIiIhMjEE5EREREZGJsSkyERERfbSCg4Pl/v37pi7Ge8mWLZucOnXK1MX4qCmKIn5+fnLr1q0U3zaDciIiIvpo3b9/X0JCQkxdjExj//79UqVKFfnss89kyZIlpi5OhsKgnIiIiD56ZmZm4unpbOpiJCks7Ml79d5BGRuDciIiIvroeXo6y717S0xdjCT5+HSQkJBHpi4GpTI29CQiIiL6yNy9e1d69+4tAQEBYm1tLS4uLlK/fn05evSotszZs2fFyspKXFxc5N69e4m28fnnn4uiKNKtWzcREenQoYNUqVJFRESWLl0qiqJo/40YMeKD96/av3+/KIoiHTp0kPv370uXLl3Ex8dHzM3N5aeffhKRhFxvf39/iY+Pl/Hjx0uePHnEyspKfH195auvvkpy8J6zZ8/K4MGDpXjx4uLm5iZWVlaSM2dO6dmzp4SGhv7Xj/Y/Y1BORERE9BE5duyYFClSRGbOnCkWFhZSr149KViwoOzYsUMqVqwoa9asERGRoKAgGTNmjDx58kQ+++wzo6HiN27cKAsWLJA8efLI5MmTRUSkfPnyUqtWLRERCQgIkM8++0z7Lygo6IP3/6aHDx9KiRIlZOvWrVKmTBmpU6eO2NraGi3Tpk0bGT16tAQGBkrNmjXl+fPnMmHCBOncuXOi7f3www8yZcoUrex169YVADJ79mwJDg5O88BcgeEnTERERJRJvHjxQi5fviz58uVLFLypfHx8JCQkRLy9s6b79BVvb+8ka6w/xLNnzyRv3rwSHh4uS5culbZt22rzTp06JTVr1pTXr1/LjRs3xM3NTQBI9erVZe/evTJx4kQZOHCghIWFSeHChSUyMlKOHj0qwcHB2jb+raHnh+7fcJsiIk2aNJGVK1eKtbW10XbVQXry5csne/fulWzZsomIyM2bN6VYsWLy9OlTuX79ugQEBGjr7Nu3T/Lnzy8eHh7aNL1eL6NHj5bvvvtOOnbsKIsWLUq0nw/pfeV9zkEVa8qJiIiIPhKLFi2SsLAw6du3r1FALJLQPeS3334rUVFRsmLFChFJCEKXLl0qzs7O8s0338iZM2ekQ4cOEhERId99951RQJ4a+zdkZWUl06dPTxSQG5o2bZoWkIuI5MiRQ9q1ayciIocOHTJatkqVKkYBuUhCg9/hw4eLt7e3bN68+YOOLbnY0JOIiIjoI7Fz504REWnatGmS8ytUqCAiIidOnNCm+fj4yNy5c6VFixZSuXJlefbsmZQvX16GDBmSJvtXFStWTLy9vd+6bQsLC61G3VCePHlERCQsLCzRvEePHsnmzZvlwoUL8vTpU4mPjxcRkdevX8ujR4/k8ePH4uLi8i9HlTIYlBMRERF9JNS0i3Llyr1zuYiICKO/P/nkE2nSpIls2LBBbG1tZfny5aLT6dJs/yIi2bNnf+c62bJlS7JM9vb2IiKJGnuuWrVKunbtKlFRUW/d5vPnzxmUZxR6vV47cWxtbbWcJiIiIno3APLixQsREXF1dRUzM2bVpja1v/PmzZuLnZ3dW5fLmzev0d9hYWFa+seLFy/k0qVL4u/vn2b7F5F3pq2IyAedP7dv35YOHTqIiMhPP/0k9erVE29vb7GxsRERkbJly8qxY8ckLZteMihPpoiIiET5SERERPRhHjx4IO7u7qYuRqbn4+MjV69elSFDhkjx4sXfax0A0rFjR4mIiJDWrVvL2rVrpVOnTnL+/HmtMWZq7j81bNu2TV69eiUDBw6UL7/8MtH8GzdupHmZ+EhKRERE9JGoUaOGiIhs2LDhvdeZPn267NixQ8qVKyfLly+Xb775Rh48eJBkN4OWlpYiIhIXF5di+08NT548EZGEh4Q3HTx4UB48eJDWRWJNeXIZdm/z4MGDd76KSSnR0dFa7Xxa7dMUeJyZz8dyrDzOzIXHmTb7/Lfu4ihldOvWTSZPniwTJkyQ7NmzS5cuXYzSPuLi4mTPnj3i7e0tBQsWlEuXLslXX30l9vb2Wh75t99+K7///rts2bJF5syZI927d9fW9/LyEhGRq1evpsj+U4va+HPFihXSpUsX7XwPCQkxOp60xKA8mQxzyO3s7NL8Ym2KfZoCjzPz+ViOlceZufA4Uw/bZKUNJycn2bRpkzRo0EC6desmo0ePloIFC4qzs7Pcv39fTp8+LU+fPpUNGzZInjx5pG3btvLy5UuZO3eu5MiRQ0REzM3NZcWKFVK0aFEZMGCAVKlSRQIDA0VExN/fXwoXLiynTp2SkiVLSoECBUSn00nDhg2lYcOGH7T/1AzKGzZsKAUKFJBTp05Jrly5pFy5cvLy5UvZt2+fBAUFSdmyZZMcXTQ1MSgnIiKij15Y2BPx8elg6mIkKSzsSYpur3Tp0nL+/HmZMmWKbN26VQ4cOCAiIp6enlKpUiVp0qSJVK9eXYYNGyZnz56VTz75RNq3b2+0jdy5c8vkyZOlW7du0rZtWzl27JhYWFiIiMivv/4qgwYNkkOHDsmff/4per1efHx8pGHDhh+0/9RkaWkphw4dkm+++Ua2b98uv/32m3h7e8sXX3whw4cPl7p166bq/pPCET2TKTo6WrJkySIiIlFRUWn2ui+t92kKPM7M52M5Vh5n5sLjzLj7/JARPTOClBjRk9LWh4zoyZpyIiIi+mgZjv6Y3mWkstKHy5RBeVRUlEycOFGOHz8uJ06ckCdPnsjixYu1/ij/zdOnT2Xw4MGyYcMGefHihZQsWVImTZokxYoVS92CExERUZo6deqUqYtAJCKZtEvEiIgIGTVqlFy+fFmKFCnyQevq9XqpV6+erFy5Unr37i0TJkyQ8PBwqVy5svz999+pVGIiIiIi+phlyppyT09PCQsLk2zZssmpU6ekRIkS773uunXr5OjRo/LLL79I8+bNRUSkRYsWkidPHvnuu+9k5cqVqVXs92ZnZ5emI0yZCo8z8/lYjpXHmbnwOIkoLWTKoNzKyuo/512tW7dOPDw8pGnTpto0Nzc3adGihaxYsUJiY2PFysoqyXWjo6Pfut3M2jCIiIjo37zt/viu+ybRxyZTBuXJcebMGSlWrJhRR/YiIiVLlpR58+bJtWvXpFChQkmuqw6AkJS31T4EBwfL/fv3/3uBk0F9k5AWeJypj8eZ8nicqY/HmfJMeZwiSR+r2sMKEb0dg/I3hIWFScWKFRNN9/T0FBGR0NDQtwbl/8X9+/czTFdMycHjzFx4nJkLjzNz+ViOkyizYVD+hpiYmCTTU6ytrbX5b5OcYYnNzMzExcXtP637oR4/fih6vT5N9vUmHmfK43GmPh5nyuNxpr60PE6Rdx9rVFRUktOjo6Pf+ZaZ6GPCoPwNNjY2Ehsbm2j6y5cvtflvk5xhiV1c3GTtykP/ad0P1aJNBYmIeJAm+3oTjzPl8ThTH48z5fE4U19aHqfIu4+V7aqI/l2m7BIxOdSeW96kTvPy8krrIhERERFRJseg/A1BQUFy+vTpRK/gjh8/Lra2tpInTx4TlYyIiIiIMquPOigPCwuTK1euyOvXr7VpzZs3lwcPHsj69eu1aREREfLLL79IgwYN3todIhERERHRf5Vpc8pnzJghT58+ldDQUBER2bJli9y7d09ERL744gtxdHSUr7/+WpYuXSo3b94Uf39/EUkIykuXLi0dO3aUS5cuiaurq8yaNUvi4+Nl5MiRpjocIiIiIsrEMm1Q/uOPP8rt27e1v9evX6/Vfrdr104cHR2TXE+n08m2bdtk0KBBMm3aNImJiZESJUrIkiVLJDAwME3KTkREREQfl0wblN+6detfl1myZIksWbIk0XRnZ2dZsGCBLFiwIOULRkREROmGqQdb+hApNQiVoiji5+f3XrHSx8zf319u37791gEgU1qmDcqJiIiI/g0HW/rvbt26JTly5JBKlSrJ/v37TV2cDI9BOREREX300nqwpQ+R0oNQXb58WSwsLFJse5QyGJQTERHRRy+tB1v6ECk9CFXevHlTbFuUcj7qLhGJiIiIPjaKomi9zqn2798viqJIhw4d5PHjx9KjRw/x9PQUKysrKViwoCxatMho+REjRkiOHDlEROTAgQOiKIr2X4cOHYyWffz4sXz99deSP39+sbGxEUdHR6latar89ttvicp269YtURRFKleuLM+ePZP+/ftLjhw5xMLCQvr27SsiCbneiqKIiMiCBQukcOHCYmNjI9myZZNu3brJ06dPE233+vXrMmLECClTpoxky5ZNLC0txcfHR9q3by/Xrl37bx9kCmNNORERERGJiMjTp0+lTJkyEhUVJRUqVJCIiAg5ePCgdO7cWfR6vXTp0kVEEgZbbNasmfz666/i4eEhtWvX1rZRvnx57d/Xrl2T6tWry927d8Xf319q1aolz58/lz/++EMaNGggEydOlIEDByYqR0xMjFSqVElu374tlSpVkmLFiomzs7PRMoMHD5apU6dK5cqVJVeuXHLkyBGZN2+eXL58WXtQUC1YsEAmTJggBQsWlBIlSoiVlZVcunRJli9fLps2bZJDhw5J4cKFU/rj/CAMyomIiIhIREQ2bdokrVq1kiVLlmgDJm7cuFGaNGki33//vRaUN27cWIKCguTXX3+VvHnzJtmbXXx8vDRv3lzu3r0rEyZMkAEDBoiZWUKSxvXr16VmzZoyZMgQqV27thQsWNBo3RMnTkiZMmXkxo0b4uTklGRZly9fLufOndO6rI6IiJAyZcrIoUOHZN++fVK1alVt2caNG0u3bt202n3V4sWLpVOnTtK3b1/Zu3fvf/rMUgrTV4iIiIhIREQcHBxkxowZRiOYN27cWAoWLCh37tz5oG4Ut2zZIufPn5dmzZrJoEGDtIBcRCRXrlwyadIkiY+Pl/nz5ye5/rRp094akIuIfP/990ZjyLi6ukr37t1FROTgwYNGy5YuXTpRQC4i0rFjRylXrpzs379fIiMj3/vYUgNryomIiIhIRESKFy8uWbNmTTQ9T548cuHCBQkLC0uUj/42O3fuFBGRpk2bJjm/QoUKIpJQK/4mT09PCQ4Ofuf2a9asmWQ5RUTCwsISzYuKipItW7bI2bNn5fHjx/L69WttWQDyzz//SLFixd65z9TEoJyIiIiIRETEx8cnyen29vYiIhIbG/ve21Jr1du2bStt27Z963IRERGJpmXPnv1ft59UWd9Wzr1790qrVq3k4cOHb93e8+fP/3WfqYlBORERERGJiBilmCSX2rd67dq1xcPD463Lubq6JppmbW39r9t/37JGRUVJixYt5PHjxzJ8+HBp1aqV+Pn5iY2NjSiKIm3atJFVq1al2cidb8OgnIiIiIhSnFqT3aVLF2nWrJnJynHo0CF59OiRNG/eXEaOHJlo/o0bN0xQqsTY0JOIiIiIPpilpaWIiMTFxSU5v0aNGiIismHDhjQrU1KePHkiIkmnu1y/fl1Onz6d1kVKEoNyIiIiIvpgrq6uYmFhIf/884/Ex8cnmt+sWTPJnz+//Pzzz/L9998nyvMGIEeOHJEjR46kajnVxp/r1683yil/+vSpdO7cWWvwaWpMXyEiIqKP3uPHD6VFmwqmLkaSHj9+e+NEU7K0tJTatWvLli1bpEiRIlKsWDGxtLSUcuXKSceOHcXc3Fw2btwotWrVkuHDh8uMGTOkcOHC4u7uLhEREXL27FkJDw+XKVOmSLly5VKtnMHBwVKjRg3ZtWuX5MmTRypXriwiCaOYurq6SqNGjWTTpk2ptv/3xaCciIiIPnp6vV4iIh6YuhgZzoIFC2TgwIGya9cuWblypcTHx0tcXJx07NhRRERy584tZ86ckRkzZsj69evljz/+kLi4OMmWLZsULVpUGjZsKC1atEj1cm7atEnGjBkja9eule3bt4u7u7u0atVKRo8eLQMGDEj1/b8PBuVERET00cqWLZupi/DeUqqsSfUyUrly5Xf2PrJkyZIkR+10d3eXZcuWvXN/jo6O8s0338g333zzr2Xz9/f/115Q3jWA0duOw8bGRkaPHi2jR49ONO9tx/YhAyWlBAblRERE9NE6deqUqYtAJCJs6ElEREREZHIMyomIiIiITIxBORERERGRiTEoJyIiIiIyMQblREREREQmxt5XiIiIMpHw8HARSfvBcNQBbtT9pyf/1sUeUWr5kHOPQTkREVEmog53bqrBcJIabt1UzMwSEgLSU5no46Kee+q5+C4MyomIiDIRnU4ner1eRN4vEEgp6j51Ol2a7fPfWFlZiU6nk6ioKHFwcDB1cegjFBUVJTqdTqysrP51WQblREREmYi7u7uEhISIt3dWuXdvSZrt18eng4SEPBJ3d/c02+e/URRF7O3t5dGjhHKZmzPsobQTFxcnjx49EgcHB1EU5V+X59lJREREmZaPj49cvnxZrl27Jh4eHmJtbf1eARLRfwVAXr58KQ8ePJD4+Hjx9vZ+r/UYlBMREVGmZWVlJbly5ZLQ0FC5deuWqYtDHxF7e3vJnj37e6WuiDAoJyIiokwuS5YskidPHnn9+rW8fv3a1MWhj4CFhYVYWFh80DoMyomIiOij8F8CJaK0wsGDiIiIiIhMjEE5EREREZGJMSgnIiIiIjIxBuVERERERCbGoJyIiIiIyMQYlBMRERERmRiDciIiIiIiE2NQTkRERERkYgzKiYiIiIhMjEE5EREREZGJMSgnIiIiIjIxBuVERERERCbGoJyIiIiIyMQYlBMRERERmRiDciIiIiIiE2NQTkRERERkYgzKiYiIiIhMjEE5EREREZGJMSgnIiIiIjIxBuVERERERCbGoJyIiIiIyMQYlBMRERERmRiDciIiIiIiE2NQTkRERERkYgzKiYiIiIhMjEE5EREREZGJMSgnIiIiIjIxBuVERERERCbGoJyIiIiIyMQYlBMRERERmRiDciIiIiIiE2NQTkRERERkYuamLgBRZhIeHi4iIo8fP5QWbSqkyT4fP35otG8iIiLKeBiUE6Wg+Ph4ERHR6/USEfHAJPsmIiKijIdBOVEK0ul0otfrRUTEzCxtssPU/el0ujTZn8jH80bgYzlOIiIyPQblRCnI3d1dQkJCxNs7q9y7tyRN9unj00FCQh6Ju7t7muxP5ON5I/CxHCcREZkeg3Ii+mAfyxuBj+U4iYjI9BiUE9EH+1jeCHwsx0lERKbHLhGJiIiIiEyMQTkRERERkYkxKCciIiIiMjEG5UREREREJsagnIiIiIjIxFK095Xw8HAJCQkRERFvb2/2HkBERERE9B5SpKZ81qxZEhgYKJ6enhIcHCzBwcHi6ekpefPmldmzZ6fELoiIiIiIMq1k1ZTr9Xpp0aKFbNiwQQCIk5OT+Pn5iaIocvv2bbl27Zr07t1b9uzZI7/88osoipJS5SYiIiIiyjSSVVM+b948Wb9+veTJk0c2b94sjx8/ljNnzsjp06fl0aNHsmXLFgkMDJQNGzbIvHnzUqrMRERERESZSrKC8sWLF4uDg4Ps379f6tevn2h+vXr1ZO/evZIlSxZZtGhRcnZFRERERJRpJSsov3TpklStWlU8PDzeuky2bNmkWrVqcunSpeTsioiIiIgo00p2Q8/3yRNnLjkRERER0dslKygPDAyUvXv3SkRExFuXiYiIkL1790pgYGBydkVERERElGklKyj/7LPPJDIyUqpVqyZ79uxJNH/fvn1So0YNefbsmXTo0CE5uyIiIiIiyrSS1SViz5495ffff5ft27dLzZo1xc3NTfz8/ERE5Pbt2/Lw4UMBIHXr1pWePXumSIGJiIiIiDKbZNWU63Q62bJli0ycOFF8fHwkPDxcTp48KSdPnpTw8HDx9fWViRMnyubNm8XMLEXGKSIiIiIiynSSVVMuImJmZiYDBgyQAQMGyN27dyU0NFRERLy8vMTX1zfZBSQiIiIiyuySHZQb8vX1ZSBORERERPSBmFNCRERERGRiH1RT3qlTJ1EURcaOHSseHh7SqVOn915XURRZuHDhBxeQiIiIiCiz+6CgfMmSJaIoinz11Vfi4eEhS5Ysee91GZQTERERESXtg4Lyffv2iYhI9uzZjf4mIiIiIqL/7oOC8kqVKr3zbyIiIiIi+nDJauh58OBBuXbt2r8u9/fff8vBgweTsysiIiIiokwrWUF55cqVZfz48f+63IQJE6RKlSrJ2RURERERUaaV7C4RAaTIMkREREREH6s06ac8NDRUsmTJkha7IiIiIiLKcD54RM9ly5YZ/X39+vVE01RxcXFy9epV2b17t5QuXfq/lZCIiIiIKJP74KC8Q4cOoiiKiCT0PX7kyBE5cuTIW5cHINbW1jJ8+PD/XsoPFBsbK8OHD5fly5fLkydPpHDhwjJ69GipUaPGO9cbMWKEjBw5MtF0KysrefnyZWoVl4iIiIg+ch8clA8fPlwURREAMmrUKAkKCpJGjRoluaylpaV4eXlJzZo1xdPTM9mFfV8dOnSQdevWSd++fSV37tyyZMkSqVu3ruzbt0/Kly//r+vPnj3bKN1Gp9OlZnGJiIiI6CP3wUH5iBEjtH8vWbJEqlevLt99911KlilZTpw4IatXr5aJEyfKwIEDRUSkffv2UrBgQRk8eLAcPXr0X7fRvHlzcXV1Te2iEhERERGJyH8Iyg3dunUrhYqRctatWyc6nU66du2qTbO2tpbOnTvL0KFD5e7du+Lr6/vObQCQZ8+eib29vZaq8z6io6PfOs/Ozu69t0NERJSZvO3++K77JtHHJllBeXp05swZyZMnjzg4OBhNL1mypIiInD179l+D8pw5c0pUVJTY2dlJ48aNZdKkSeLh4fGv+37XMh97t5Dh4eEiIvL48UNp0aZCmuzz8eOHRvsmoo8br0Omwx7YiP5digTlhw8flk2bNsnff/8tz58/TzIAVRRF9uzZkxK7e6ewsLAk89fVaaGhoW9d19nZWXr37i1lypQRKysrOXTokMycOVNOnDghp06dShTo0/uLj48XERG9Xi8REQ9Msm8i+rjxOkRE6VmygnIA0rlzZ1m6dKkWiKuNQFXq3x+SBpIcMTExYmVllWi6tbW1Nv9tvvzyS6O/mzVrJiVLlpS2bdvKrFmzZMiQIe/c94MHD5im8hY6nU70er2IiJiZpUn3+Nr+2FCXiER4HTKlqKioJKdHR0e/15tooo9BsoLyOXPmyJIlSyQ4OFjGjRsns2fPlg0bNsjVq1flxo0bsmbNGlm+fLn0799fevbsmVJlficbGxuJjY1NNF3t0tDGxuaDttemTRsZMGCA7N69+1+Dcjs7Owblb+Hu7i4hISHi7Z1V7t1bkib79PHpICEhj8Td3T1N9kdE6RuvQ6bDeyPRv0tWUL5kyRKxs7OT7du3S9asWWXFihUiIpI7d27JnTu31KpVS+rWrSstW7aUsmXLip+fX4oU+l08PT0lJCQk0fSwsDAREfHy8vrgbfr6+srjx4+TXTYiIiIioqQk6/3d5cuXpWzZspI1a1YRES1FxTB3rnnz5lK8eHH58ccfk7Or9xYUFCTXrl2TZ8+eGU0/fvy4Nv9DAJBbt26Jm5tbShWRiIiIiMhIsoJyvV6vBeQiIra2tiIi8uTJE6PlcufOLefPn0/Ort5b8+bNJT4+XubNm6dNi42NlcWLF0upUqW0nlfu3LkjV65cMVr34cOHibY3e/ZsefjwodSuXTt1C05EREREH61kpa94e3sb9WaipqecOXPGaEj7a9euibl52vS+WKpUKfnkk0/k66+/lvDwcMmVK5csXbpUbt26JQsXLtSWa9++vRw4cMCoUaqfn5+0bNlSChUqJNbW1nL48GFZvXq1BAUFSbdu3dKk/ERERET08UlWTXmxYsXk0qVLWrpKzZo1BYAMHjxYrly5Is+fP5eJEyfKn3/+KUWLFk2RAr+PZcuWSd++fWX58uXSp08fef36tfz2229SsWLFd67Xtm1bOXHihIwYMUL69u0rJ0+elMGDB8vBgwe1twBERERERCktWdXXDRs2lDVr1sjWrVulYcOGUqRIEWnVqpWsXr1aChQo8L+dmJvLmDFjkl3Y92VtbS0TJ06UiRMnvnWZ/fv3J5o2f/78VCwVEREREVHSkhWUt27dWpo2bWqUmrJ06VIpXLiwbNy4UZ48eSJ58uSRwYMHayNqEhERERGRsWQner85UI+FhYUMGTLkX/v0JiIiIiKiBGkypFl4eDiDdCIiIiKit0jVoPzu3bvyxRdfSI4cOd6Z301ERERE9DH74PQVvV4vq1evlh07dkh4eLi4u7tLnTp1pEWLFmJmlhDj3717V0aOHCnLly+XuLg4ERFp0qRJypaciIiIiCiT+KCgPC4uTurWrSt79uwx6t97xYoV8ssvv8ivv/4qS5culd69e8uLFy8EgDRu3FhGjBghhQsXTvHCExERERFlBh8UlM+cOVN2794t1tbW0qFDBylQoIA8f/5ctm/fLhs3bpTu3bvL/PnzBYDUrFlTfvjhhw8e1p6IiIiI6GPzQUH56tWrRafTyYEDB6REiRLa9CFDhkiPHj1k7ty5oiiKTJw4UQYMGJDihSUiIiIiyow+qKHn5cuXpWzZskYBuWrQoEEiIpI3b14G5EREREREH+CDgvLnz5+Lv79/kvNy5MghIiJFihRJdqGIiIiIiD4mHxSUAxCdTpfkPEVRRCRhiHsiIiIiInp/aTJ4EBERERERvd0H91O+dOlSWbp0aZLzFEV563xFUbQ+y4mIKP0IDw8XEZHHjx9KizYV0mSfjx8/NNo3EdHH7oODcsP+ydNiPSIiSl3x8fEikjA4XETEA5Psm4joY/dBQbler0+tchARkYnodDrt+q6OzJza1P29rZ0SEdHH5oNryomIKHNxd3eXkJAQ8fbOKvfuLUmTffr4dJCQkEfi7u6eJvsjIkrv2NCTiIiIiMjEGJQTEREREZkYg3IiIiIiIhNjUE5EREREZGIMyomIiIiITIxBORERERGRiTEoJyIiIiIyMQblREREREQmxqCciIiIiMjEGJQTEREREZkYg3IiIiIiIhNjUE5EREREZGIMyomIiIiITIxBORERERGRiTEoJyIiIiIyMQblREREREQmxqCciIiIiMjEGJQTEREREZkYg3IiIiIiIhNjUE5EREREZGIMyomIiIiITIxBORERERGRiTEoJyIiIiIyMQblREREREQmxqCciIiIiMjEGJQTEREREZkYg3IiIiIiIhNjUE5EREREZGIMyomIiIiITIxBORERERGRiTEoJyIiIiIyMQblREREREQmxqCciIiIiMjEGJQTEREREZkYg3IiIiIiIhNjUE5EREREZGIMyomIiIiITIxBORERERGRiTEoJyIiIiIyMQblREREREQmxqCciIiIiMjEGJQTEREREZkYg3IiIiIiIhNjUE5EREREZGIMyomIiIiITIxBORERERGRiTEoJyIiIiIyMQblREREREQmxqCciIiIiMjEGJQTEREREZkYg3IiIiIiIhNjUE5EREREZGIMyomIiIiITIxBORERERGRiTEoJyIiIiIyMQblREREREQmxqCciIiIiMjEGJQTEREREZkYg3IiIiIiIhNjUE5EREREZGIMyomIiIiITIxBORERERGRiTEoJyIiIiIyMQblREREREQmZm7qAnzswsPDRUTk8eOH0qJNhTTZ5+PHD432TURERESmxaDcxOLj40VERK/XS0TEA5Psm4iIiIhMi0G5iel0OtHr9SIiYmaWNtlE6v50Ol2a7I+IiIiI3o1BuYm5u7tLSEiIeHtnlXv3lqTJPn18OkhIyCNxd3dPk/0RERER0buxoScRERERkYkxKCciIiIiMjEG5UREREREJsagnIiIiIjIxBiUExERERGZGINyIiIiIiITY1BORERERGRiDMqJiIiIiEyMQTkRERERkYkxKM+AoqNfiqI0EEVpINHRL01dnFTD48x8PpZj5XFmLjxOIkoLDMqJiIiIiEyMQTkRERERkYllyqA8NjZWvvrqK/Hy8hIbGxspVaqU7Nq1673WDQkJkRYtWoiTk5M4ODhIo0aN5MaNG6lcYiIiIiL6mGXKoLxDhw4yefJkadu2rUydOlV0Op3UrVtXDh8+/M71oqKipEqVKnLgwAEZOnSojBw5Us6cOSOVKlWSR48epVHpiYiIiOhjY27qAqS0EydOyOrVq2XixIkycOBAERFp3769FCxYUAYPHixHjx5967qzZs2Sv//+W06cOCElSpQQEZE6depIwYIFZdKkSTJ27Ng0OQYiIiIi+rhkuqB83bp1otPppGvXrto0a2tr6dy5swwdOlTu3r0rvr6+b123RIkSWkAuIpI3b16pVq2arF27NsmgHID27/DwcLGzs0ty22+brtfr////eO/W7obL/ZcW8no9tH1HR0d/8Pr/BY/z7T6W4xRJ3rHyOFMPj/PdeJzvu9+3H+vbjt1wuuH9lOhjpCCT/Qpq1KghISEhcunSJaPpe/bskerVq8vmzZulQYMGidbT6/Via2srnTp1klmzZhnN+/bbb2X06NHy7Nkzsbe3N5oXHh4uHh4eKX8gREREH5EHDx6Iu7u7qYtBZDKZLqc8LCxMPD09E01Xp4WGhia53uPHjyU2NvY/rUtERERElByZLn0lJiZGrKysEk23trbW5r9tPRH54HVdXV3lwYMHEh0dLba2tqIoSpLbf1v6ChERUWb3tvQVAPLixQuxs7MTV1fXNC4VUfqS6YJyGxsbiY2NTTT95cuX2vy3rSciH7yumZkZX7cRERG9AyumiP5dpktf8fT0lLCwsETT1WleXl5Jrufi4iJWVlb/aV0iIiIiouTIdEF5UFCQXLt2TZ49e2Y0/fjx49r8pJiZmUmhQoXk1KlTieYdP35ccubMmaiRJxERERFRSsh0QXnz5s0lPj5e5s2bp02LjY2VxYsXS6lSpbTuEO/cuSNXrlxJtO7JkyeNAvOrV6/K3r175ZNPPkmbAyAiIiKij06m6xJRRKRFixayYcMG6devn+TKlUuWLl0qJ06ckD179kjFihVFRKRy5cpy4MABo35Rnz9/LkWLFpXnz5/LwIEDxcLCQiZPnizx8fFy9uxZcXNzM9UhEREREVEmlukaeoqILFu2TL799ltZvny5PHnyRAoXLiy//fabFpC/jb29vezfv1/69esno0ePFr1eL5UrV5YpU6YwICciIiKiVJMpa8qJkis+Pl50Op2pi5GuAXhrF6DpmV6vFzOzTJe5R0REGRzvTJQs8fHxpi5CqlAD8jNnzpi4JOlTfHy8FpCr50BGeL43DMgzQnn/DQBtSPWPFYAMdQ5mZob3g4xwXr55/+L5Q6bGoJySRQ1eL1y4YOKSJJ/hBfrJkyfStGlTad68uezfv990hUpn4uLiRCThe9fr9dKnTx8ZMWKEiEi6rjVXv1szMzMJDQ2VkiVLyqRJk0xcquRTFEXMzMzk/v378ujRI1MXxyQURRGdTid///23/Pbbb6YuzkdJDcDV+8GcOXNk+/btpizSe1HLu3///gz75o8yFwbl9MEMg9eTJ09K/vz5ZeTIkRISEmLCUiWfGmj+9ddfcvbsWbl27Zp8++23kiNHDlMXLd0wN/9fM5S2bdvK1q1bxczMTKKiokxYqn+n3nzDwsJkwIABEhMTI5aWltpDRkbyZg3k999/L2XKlJENGzaYqESm99tvv0lgYKBMnz5dHjx4YOrifHTUt08vX76Upk2bysCBAyU0NFSeP39u4pIlZvj7efr0qbRs2VLq1Kkj69evN2GpiBJkyoaelLp0Op2EhobKq1evZOPGjeLl5SWdOnWSLFmymLpoyRIVFSVt27aVLVu2SP78+aV9+/bSoUMHUxcrXbl79640bNhQWrZsKRERETJ27FipU6dOuv/uL168KLVr15YWLVrI2bNnZfr06VKhQgWjh4yMQg2Azp49K87OzrJ+/Xpp1aqVlChR4qOr7Vu9erUoiiIrV66Unj17SqdOnThUuwncvXtXxo0bJ9mzZ5eHDx/K8uXLpUqVKulybA8zMzN5/fq1HDx4UKytreXy5csyZ84cKVq0qKmLRiQCog904cIFuLu7I0eOHPD29sbevXtNXaQPotfrERcXp/3b0PTp05ErVy5kzZoVJ06cMEXx0o34+PhE044fP458+fJBURTUqFFDm/7m52gqer0+ybLcvn0bRYoUgaIoaNasmdHyGdF3330HnU6H/Pnzo0mTJnj27Jmpi5TmoqOj4eTkBDc3NxQoUAD//POPqYv0UUjqN7Nq1SooioKAgAC0bt3aBKV6fy9fvkTv3r2hKAoKFSqEb775xtRFItIwfYXeSq/XGzV8UV/7FShQQKpUqSLPnz+XvHnzSpUqVUxVxA+mNlDU6XQSFRUlz549k9evX2vzW7duLcWLF5fHjx9ro8Jm1sasb2OYf/2mEiVKSM+ePcXR0VEsLCzk5cuXRo0+TSkuLk4URRFFURKlpXh5ecl3330nOp1OHj16pOVfp4dyv8ub5576d+nSpSV37twSEhIizZs3F3t7+wzRsO6/SCrFCIDY2trKokWLJCIiQsLCwsTT01NEPr7fa1qJj49P9CZGPedatWolzZo1kxs3boifn5+IiLx69cok5VTBoAGwISsrK6lbt64EBwfLpUuXpFChQtryRKbGoJySFB8fL2ZmZqIoijx48ED0er3Rxbhfv35iY2Mjp06d0i6+GeGipuYWDx8+XIKDgyU4OFjKlCkjv//+uzx79kyyZs0qbdu2FVdXVxk1apTROh8L9Xg3bNgg/fv3l4kTJ8qff/4pMTExoiiK1KhRQypUqCBHjhyR8PDwdPP5qKkoEyZMkA4dOkifPn1k69atEhsbK+bm5lKxYkVp3LixXL16Va5evWri0r7bmw3nIiIijP6uXbu21K5dW549eyYXL14UkYzx+/sQ6vGo3+uuXbvk2LFjEhoaql2LGjRoIA0bNpSXL19qDbLTy/mYmej1etHpdKIoipw4cULmzp0ru3fvNmpHNGzYMLGzs5OtW7fK06dPxdLS0mQPioaVL48fP5bbt29LZGSkNr9ixYpSp04dAZBhHtDpI2GyOnpK9+7fv4+2bdsif/78KFCgAFq1aoUrV65o84cOHQpFUTB27FgASac7mJpaJvWVa2hoKKpVq4asWbOiR48eaN++PQIDA5ElSxYMHToUAPD69Wt8+eWXUBQF69ev16Z9LG7evInKlSvDxsYG/v7+UBQFFhYW6Nq1q7bM6tWrkSVLFvTo0QNA+kgDOXbsGAIDA+Hm5oaiRYvC3t4eiqKgTZs2WrrS4cOHkSVLFnzxxReIjIw0cYn/3ebNm1G+fHkEBQWhVKlSWLp0KR4+fAgAuHr1KgoUKABfX1+EhYUBSJ+/weT65ZdfkD17djg4OMDc3BwBAQGYMmWKNv/48ePQ6XTo3r07Hj9+DCB9nI+ZzYMHD9CoUSNYWVnB1dUViqIgb968+P333xEbGwsA6NOnD+zs7DB16lQTlzbB4MGD4e3tDQ8PD3h6emLatGnab+XkyZMIDg5G7ty5tfITmRqDcgLwv5u5+v/9+/fDx8cH+fPnR48ePVC3bl04ODigYMGC2LBhAwDgxo0b2kXt9u3bRuunB4ZlUf+9fPlyWFtbY8aMGXj+/DkAIDw8HBUqVIC1tTW2b98OICHAK1SoEHLnzq1tIzPe6NVg1dCXX34JNzc3rFy5EteuXcODBw/QpEkTKIqC7777DkDCDfrzzz+Hoig4e/YsgLT9fJIqd8uWLREQEIBdu3bhyZMneP78OXr16gVFUTBq1Ci8fPkScXFx6Nu3L7JkyYKdO3emWXnfh/r56fV6vHz5EkOGDIGFhQXq16+PFi1aoFChQjA3N0fv3r21HPIffvgBjo6OGDRokCmLniIM23qo1q1bB1dXV7Ru3RrLli3D4sWLUapUKSiKgqlTp2q/4a5du8LR0RGbNm0yRdEznTe/h+joaLRp0wY+Pj6YO3cuDh8+jAULFiAgIADe3t5YuXIlgISKnJw5c6JkyZK4du0agLS5J7y5j8ePH6NBgwZwdXXFoEGDMGzYMFSrVi1Rm5JZs2ZBURRMnDgRQOa8xlPGwqCckrxoduzYEb6+vjh48KB2odq7dy8cHR1RokQJhISEAACmTZsGZ2dnrcY0PTCs1Y6KikK9evW0WvAuXbrAyclJm//q1SsAwJ49exAQEIBSpUpp83744QfodDqMHz8+0XYzujdvPkuXLsXTp08REhICBwcHfP7550bzb968iUaNGsHc3Bw3b94EAOzevRvZs2c3avCZ2gzP1SdPnmDChAk4d+4c7ty5A0VRtO9ZFRoaik8//dSo4e7Vq1fh5+eHRo0aabVmpmR4TOo5dubMGXh7e6Nz5864e/cuACA2NhatWrWCjY0NRo0aBSAh+KhatSq8vb1x+vRpAEk/sGQkERERWqPNunXrwtfXF5cvX9bmq+ditmzZtAerkJAQZM2aFY0bN9YqCBhgJZ/6+f7111+ws7ND//79jeZfvXoVWbJkQaVKlbTrwpQpU2Bra4uvvvoqTcpo+D0/ffoUAPD777/D3Nwc48aN0x5gX716hW7dukFRFMybNw8AcP36dTRo0ACOjo64f/9+mpSX6F0YlH/EDG/e58+fR1BQEHbv3o3IyEhkz57dqEZBDRZGjx4Ne3t7DBs2DEBCDUrdunXh6emJw4cPJ9quKT169AjdunVDQEAARowYAQBo0KABvL29cf36dQDGAVG3bt1gb2+P33//HUDCzb9GjRpQFAUPHjxI+wNIJYY3sZCQEJQvXx6KomDfvn04ceIEzM3Ntc8rPj5eW37Tpk3IkiULunTpAgB48eIFvv/+eyiKotWUpZU7d+7Az88P+fPnx++//46zZ8/C0tJSS2swfB199OhRODk5YfDgwQASel+YOXMmzM3NMWfOHJM9bL35O/nxxx9Ru3ZtAMDMmTOhKAqOHj0K4H/n6f3791G8eHHkzp1bSyVbuXIl3N3d032vF0l5s0Jg2rRp8PT0xNixYxEfHw8/Pz98+umnAIwfis+dO4csWbKge/fuWm35pEmTYGVlhfnz56erN3YZheH5eOvWLVSqVAkuLi64cOECNm3aZJTOp9frte9jxIgR0Ol0mD17trZ+2bJlERgYqPXMlRoPSIbnw7Nnz1C3bl20a9cO0dHRGDduHBRF0c4NtfLl0qVLqFChAjw8PLR1161bB0dHR3Tv3j3Vykr0vtjQ8yOm0+kkPj5erl+/Lp06dRJzc3N5/PixmJuby9OnT8XBwUFev34tALSeOHr37i3u7u5y4MABCQkJEVtbW+nRo4fEx8fLoEGDtO2mJbzRyj40NFTy588v48aNk0OHDsm4cePkyy+/FBGRevXqSXh4uPzzzz8iktDDiNpQtVGjRhIVFSUvXrwQERF/f39p0aKFNGnSRJDwAJumx5XS1M9IURT5+++/Zfbs2bJs2TIREdm4caPWo4elpaXcvXtXIiMjxczMTGusValSJcmfP7+cOHFCIiIixMbGRho0aCD+/v7adlKaXq83aiy2Z88e6dmzp8yaNUsKFiwoCxculFq1aomNjY04OzvLwYMH5fXr10aNzHLmzCkFChSQjRs3ikhC7wtNmjSR3Llzm3QkWvV3Eh8fL1OmTJGpU6eKra2tPHv2zGjYeL1er30PHh4e8tlnn8mNGzdk165dIpLQY1CtWrVk7dq1snr1apMdz4dQf0vqdeX06dMSFhYm69evl6ZNm0qNGjXk/v374unpKZcuXRKRhAaf6nr58uWTFi1ayJYtW8TKykpERHr06CHZsmWT8ePHy507d0xwVBmTeq7pdDqJiIiQ48ePy7FjxyQ6Olpmz54tbm5ukj17dlEURWtwrJ6TIiKDBw8Wa2truXTpknYt/frrr+XatWvy66+/ikjqNKJUGwCHhYXJpEmT5Pbt21K4cGHR6XRy//59sbS0lL/++sto2Xz58knTpk0lPDxcFi1aJCIiVapUkXbt2sncuXPl5MmTbPBJpmXCBwJKY2/mWG/btg0BAQEYPnw4ypQpg7/++kurYWzcuDHy5s2r1TQA/6t97NWrF3x8fIxe93322Wf4+uuvjWpW04Lhvi5fvoxZs2bh7NmzqFChAhRFQbVq1YyWP3/+PHLmzIl69eolqv3esmULFEXRcuaB9FPr/z7e53N/+vQp7ty5g9GjR0NRFOTPn1/Lp1R17NgR2bJlw6FDh7Rp6rnTrFkz5M2b1+hcunDhQrLK/T61mufPn0dkZCQGDBgAMzMzeHl5YcaMGUbLNG7cGC4uLlpesVo7BgDly5dHyZIljfrzNsXbD8PavZiYGLRt2xZdu3ZFo0aNMHToUC2d5tixY7CyssK0adO05dVz8enTp8iSJQuGDx+uzdu3bx/KlSuHffv2pc2BJIPh963X67Fs2TKt0WD58uWNvpeOHTvCyckJmzdvBmD8+Q0bNgyKouDMmTPatD179mDr1q2pfxCZgOH1Ij4+Hg8fPkT9+vWhKAoKFy5sVPN97do1FCtWDHny5DFaJz4+HnFxcQgKCkLNmjWNtr9q1aoUf2NhuL2nT58iICAALVq0QEBAABYuXKjdr/bu3QtFUTB9+nTtvqVeD86ePQtFUTB58mRtW3v27MHnn3+eLtLZ6OPGoPwj8GawtmXLFvz+++/49ddf4ebmpjWEU8XFxWHevHlQFAU//vhjou116tQJdnZ2CA0N1aalRev12NhYLe3kTXv27IGZmRkaN26M06dPY8eOHbCzs0NQUJCW/w4kBEI//fQTzMzMMHjwYK23huvXr6NRo0bIkyePlpdoKD0H528bMOfNG+KCBQvg5eWFb7/9Fvfv30exYsWMUk/UgOf48eNwdHREixYtjHrbuXDhAnx8fNCsWTPExcUl++HLsHzq5/vmQ51er8fIkSOhKApGjx6NyMhIBAYGQlEULd81OjoaAHDlyhWYm5ujUqVKuHHjhrb+sWPH4OLigi+++CJZ5U2ON49p2bJl+PPPP9GxY0coioLs2bPj0aNH2jI3btxA9erVtYdfw/WvXbsGOzs79OrVK02P4X29ed6pfxsG1K9fv0br1q3x9ddf4/79+yhatCgcHBwwbtw4AAkpRgBw8OBBODs7o2bNmomuMW3atIG3t7fRdYj+R6/XJxkUv/kbW7duHfz8/HDkyBEsW7YMefPmRZYsWXDkyBGj9caOHQtFUbQ2Nqrz58/D3t4ePXv2BGD8MPyh3rzOquU0PI7z589j0KBBuHHjBj777DMoioKgoCCj43v+/DkqVKiAwoUL49SpU0bbPHPmDBRFwZw5c/5zOYlSC4PyTOjBgwd48uQJAONg4PXr1xg1ahQURcGYMWNw5coVfPvtt1AUBcOGDcOrV6+05W/evIkmTZrAyckJv/zyC4CEC+aRI0eQM2dOtG/fPk2PafLkyXBwcMD48eONbs5LlizBt99+i06dOqFHjx64ePEiAOD58+fo3LkznJ2dsWfPHqNtRUZGag1+cuTIgaZNm6JMmTKwtrbGlClT3hrkpjdv3nS3bt2Ktm3b4vz580bLnTlzBmFhYahVqxa6dOmCY8eOAUjo1lBRFAwcOFALglRTpkyBoigoU6YMNm/ejBUrVqB169ZwdnbGtm3bkl1uw893ypQpKFCgQKJaqq1btyI0NBTBwcEYNWoUjh8/DgCYO3cuFEVBhw4dtGXVgG/s2LGwtbVFUFAQJk2ahJ9++gkVK1ZEtmzZEgUZqenevXvavw2P9c6dOwgMDNTaYOzfvx9FixZF1qxZ8ddffxltY+XKlbC3t8enn36qPYxGRERg1KhRsLe3N3qToTLlw+Ob3+v8+fOTDNBiYmLQtWtX5M6dG/369QPwv/NNbcdieF737dsXFhYWaN++PW7evImbN29ixYoV8PDw0NYnY4bnwcmTJzFjxgzExMQYLfPkyROcOHECZcqUQe3atbXrRPfu3WFmZqa9cVC3dePGDTRp0gRmZmYYOXIkzp49iwMHDqB9+/bw8vJK1lsaw+87Ojpau0a96dSpU7Czs0OtWrVw5MgR3LhxA25ubnByctKu/aqdO3dCp9OhQYMGuHr1KoCEB9p27drBx8cHd+7c+c/lJUotDMozkZiYGLRv3x4eHh5ao0u1Qc7IkSMxc+ZMlC1bFtOmTdN6dDh58iTy5cuHKlWqGNXUAQkXwKJFi0JRFJQrVw6tW7dGrly54Ovrm2avyXfu3IlcuXLB3t4evXr1Mqq5BYCcOXMiW7Zs8PT01BrFGZY/a9as6NChQ6JjAxKCu7p166Js2bKoW7duohqV9Mzwpnvjxg3UrFkTiqLg008/1XpBAIAVK1ZoqQHBwcFGwWJ8fDwqV66MgICAJG+CI0aMQPbs2aEoCuzs7ODr64tff/01WeU2vPnu3LkTuXPnhp2dHfr162f0RsOw3I0bN0700FC2bFn4+flh//79AP5XOxcXF4fly5cjICAANjY2cHFxQZkyZYxSHFLTkSNHkCNHDnz++efagzEAnDhxAhMmTMC3336LGjVqYNeuXYiNjUVMTAzGjRsHMzMz/Pzzz0bbevLkCSZPngxFUeDv7482bdqgWbNmsLW1Rbt27fD06dN08/D45veaN29eKIpi9AAXGhqKoKAgDBkyBGXKlMHy5cu1N1WRkZGoVq0a7O3tcfLkSQD/qy1/+PAh+vfvD3Nzc1haWiJPnjywsrJClSpVtJ5WKIHh9xAZGYn27dtDURTUr19f+6yBhK4Avby80Lx5c5QpU0br7QZISP3w8/ND9erVE23/+vXraNasGRRFga2tLbJmzQpHR0csWLDgP5X3zfP3hx9+gLW1NXr27ImoqCht+uLFi9GqVSsMGDAAn332Ga5cuaI9iI8ZMwaKomDu3LmJtj9hwgSYmZnB2toaZcuWRdGiRWFpaYkffvgh3fx2iAwxKM8kfvrpJ9jZ2SFHjhyYMGGCNsAIkFDrrQY4FSpUMKoxiYmJ0W78bwYFAHD37l307dsX5cqVQ9GiRdG5c2eji3tquX37tpbf2KRJE+zdu9coH1i9IG/duhVWVlZwdnbW5qvzXr16heHDh8PKygqbNm3SLsKGAa1erzc6HjVPMiOIj49H7969oSgKSpQogTVr1iTq1uvGjRsoVaoUHBwc8M033wBI+HzUz+LQoUNQFAX9+/fX0nbUADc+Ph6hoaH4448/tB5pUsLt27e1h4imTZtiz549Rt8tkNBLQvPmzaEoCr788kttulq2bdu2IUuWLGjXrp32fRumRzx58gR37txJs2DcsC/36tWrY9euXUblad26NRRFQa5cuTBz5kyjdS9duoSSJUuiePHiiIiISLTtxYsXo169eggKCkJQUFCSwYepGAY2N2/eRK1atbQa7927dxt9r9evX0fFihW1NzAq9Ttdu3Yt3Nzc8Mknn2jz3gz2R48ejV69eiX74TCzGzNmDKysrJA7d27MmzcPt27dMpr/559/wtLSEoqiYPny5QD+d118+fIlRowYYdTbiuE1MzY2Ftu3b8fMmTMxc+ZMLX3sQxl+t1u2bIG/vz8cHR0xYMAAXL582ejcatSoEZydneHo6IiNGzcabScyMhK5c+dG+fLlta4zDbe9b98+dO3aFY0bN0bLli0TvZEiSk8YlGdwu3fvRmBgoFbbaNhYE/hfoPLNN99AURStyzXD1Af1ZvlmCoHhhe3ly5dJ1janhpMnT0JRFHh4eODnn382esBQGd4k2rZtC0VRMH/+fADG5b558yby5cuHqlWr/uvryvScN/6mJUuWwMnJCZ6enpg6dSquXbuW6GFCvanNnj0biqIkaoil6tixI1xcXNKkgdykSZOgKAqKFy+OtWvXvjMfeNmyZciaNetbu/pr2bKlNsgRYLqBq0aMGAFzc3MUKFAAixcvNqrxV8v0/PlzZM2aVet6EvhfO4y4uDgsXboUOp3OaKTKN48nLCwsyTz89GDw4MEwMzND8eLFsXLlyiT7fNbr9Vi/fj08PDwQHByc5Cikbdq0gbOzsxZ0Z6axAdLC7t274evrC2dnZwwdOhTnz59/a4632lZD7SoU+N938ddffyE4OBiBgYGJKjFS0vXr11G7dm2YmZnhk08+wYEDB4w6F1D3feLECeTNmxeWlpbaoESGx7Vy5UooioIJEyZob1je/H28+baNKD1iUJ5B6fV6TJ8+HYqioGrVqti3b1+i2kZD8fHx8PDwgK+vL86dOwfgfxctvV6PNWvWwNLSEqNHj06T8r9LVFQUihQpgsKFC2sDohgaOXIkRo8erdXQXLp0Ca6urqhWrRrCw8MBGB/bkiVLYG5ujsmTJ2ea4ZQ7d+6sNX5MytGjR7U8ypiYGNSvXx/W1tZausebfRK7uLigdevWqZ5n+emnn77zPFu5cqWWO37//n106NAB5ubm2jlrOOrjpUuX4OnpiVq1amlBYFq+kg4PD0fBggW1Nhl///33Oxu5qaMHGo6+qQZB9+7dQ5MmTeDl5WWUSgAkPqb0FIwDwHfffQdFUdCiRQtERkYmGfyoPaqEhITg888/R5YsWYxSptRjOnr0KHLmzImKFStqnyXTDN7u0aNH2ud979491KlTB4qi4NdffzUKblXqNQFI+E58fX1RoUIFbbrhA9LcuXOh0+m03plS8qE3Li4O169fh6IosLe3x88//2yU7qUyvF6rtfdff/11kuWpVq0acuXK9dZ8dKKMgEF5BrZz5044ODigR48eiQLy8PBwtG3b1ij3e+HChVAUBSNHjtSmGQ5K0r59e7i4uKSL13tq94QjRozQyrhhwwYEBgbCwsIC48ePN2qYOnjwYNjb2yfZW8zDhw9RvHhxNG/eHC9evEjT40hp6mcRExMDZ2dnVKlSxSiIO3/+PCpXrgxFUfD7779rgc2WLVvg6emJ+vXra8saBjvqIECrV69OlXKrQdeVK1dQpEgRFCpUyOgNyObNm+Hv7w9PT0/s2LFDm75r1y54eXmhQYMGRttTP4f+/fsbvYJPS0+fPsVnn32WZGNiICG1qlu3bkafc+7cuZEnTx4tb9owsPj999/h6uqKbt26pbvAOynqcR0/fhwlS5ZEQEBAomU2btyIggULYsCAAdq0gwcPwtPTE61bt07yIXnIkCGJrlOU2OrVq+Hg4KClc+j1emzYsAGOjo6JRrYNCQnRUqjUB1wAmDNnDhRFwYwZM7RzTv3/nTt30KhRIyiKkiJdBW7atAlNmzY1mlarVi1kzZoVv/32m3YMahmGDRuGLl26aO2fQkNDUapUKfj7+2sNnw1/J3/88QcsLCzQqVOnNEmxJEoNDMozsBcvXuCLL76Ao6OjUVAwfvx4ODs7w83NDdu3bzdap0iRIggICNB6ojAMCnbs2AFLS8t0k7PaqFEj+Pn5YfHixahVqxbMzMzw6aef4uzZs0Z5z0BCgJQnTx4ULVpUa4Vv+Oo7M/U/q96I1DclU6dOxdOnT9GzZ0+YmZmhSpUq2Lp1a6LeFrp06QJHR0csW7bMaDtAQnrFTz/9lCa1kuPHj4ejoyOGDx+OkJAQVK9eHWZmZmjfvj1Onz5tVK4XL15o/VFv2bIFgHHqVVRUVKIc07R09uxZuLq6on379lp61/Xr19G4cWOYmZmhVq1aePz4sXZM27Ztg6IoGDBggFH6CpCQB9+hQwe4urpmiCG/Db+nSZMmQafTaV2r/vXXX6hRowZ0Oh3atm2LS5cuGX1nI0aMgJmZmVFDUHV7//zzD6pXr240XgAl9ujRI+h0Onz66adaGtiDBw/QpUsX2Nvb49q1a4iLi8OIESNgY2ODoKAgLF++3KgG/fXr1yhZsiQKFSqkNXQ3vAb88ssv+OKLL1KkMqNfv36wsLDA0qVLtWn//PMPFEVBy5YttYqlNWvWIGfOnLCzs8P48ePx4sUL7dyZP38+XFxc8Pnnnye5j1atWqFmzZpJdmtLlBEwKM/gLl26BG9vb7Rs2RLLli1D7ty54ejoiBEjRuD27dvaBVYNUNWGfT179tTSP9SbYXR0dJp3E5VUjaB6AT5//jxsbW2hKApKly6dZINAw23MmzcPjo6O7+wmLSPUQL7pzX6FDR+k8ubNCwcHB7i4uCAgIABz585NlIOvHvOZM2cQGBhoNIhOan4eb25bfZB68uQJqlSpAnt7eyiKgsqVK+O3334z6m0B+F9wcObMGRQtWhRFihRJ1N93ejBixAhYWVlh9erVGDx4MCwtLVG8eHFs2rQJkZGRiZavVasWXF1dtYD0zS4T33yYygjURp729vZaylHZsmWxffv2RN8rkNDnfVBQEMqWLWuUtpBevtP0Tr2ejxs3DjqdzijQPXToEDw9PVGyZEnkyJEDWbNmxdixY3Hv3r0kP9+dO3dCURR88803WsCubj8lUlbUbVy+fBlVq1ZF/vz5jYJ8tcvLUaNGoWrVqjA3N0enTp1w/vx5rRxquWNjY9GgQQO4ublpb4ENK18y+ptQIgblGcTbGjzFxcVpffwqioJ27drh3LlzSb4WVi9sn3zyCbJmzYp169alapn/jeEF/48//sDhw4dx7949o7xctZb0zS633nbzDgoKgqOjY7pIwUkJhoFtUgHe/v37td5X3uynN6n+1keNGgWdToeBAwemToGRuP909fWzoZ9//llr8Pe2YNyQmo+tDi5jajt27MD8+fOh1+tx584dFC9eHIqiwMvLC9OmTUuyAav6XV6/fh1mZmZo27atlmv9ZvCTERo4btiwAd27d9dymlesWAF/f39YWlpi1apViWr7Dc/H169fY+HChbCwsDBq3Ervx/D67u/vj/Lly2u//2fPnmHcuHFaT0AXLlx466A8qrZt28LJyUl7G5XS1P3PmjUL7u7uWk9QQMK5oDaCLlOmDA4fPpyo8sXwmrJ9+3b4+flpnRYQZSYMytO5N2/WBw8exIULF4wuWrdu3ULVqlWRI0eORMGo4frqv8PCwrS+a9/VODQtXLx4EeXKlYO9vT3s7e3h6+trVNP98OFDeHl5oXbt2toojYY3mFevXuGff/7R+uY+cuQIDhw4kKbHkNqePn2K7t27o0KFCqhTpw7GjBljFLQ1aNAATk5ORnmZb5436uvcO3fuoEaNGlqPJalJTWHIkycP8ufPjy+//NKon/RWrVrB2dlZG0LdsI0AkBB4qLWoly9fRs+ePXHw4MFUL/f7CAgI0HqzefnyJVatWqXVNr7ZRZzhMannbvfu3aEoiskfjN/H2wbTyp07t1GO8PPnz9GxY0eYmZlpv0H1PEzqTc+tW7dQunRp1KxZkz1jvKc3B4N79uwZ1q5dC3Nzc62dDZCQVlWqVCkEBwcn6vLVkHoduXjxIuzs7FI8ZejNa1FoaChat24NLy8v/P3339r0xYsXa7+fN9dPitrjVlqNl0GUVhiUZxDr1q2Dv78/nJ2dYWFhgXLlymmvLOPi4vDrr7/CzMzMqEsow4vh8+fPcf36de0CvWLFikR916a1CxcuIF++fKhYsSIWLFiAdevWoUePHtDpdBg+fLgWkKld+k2dOtVoOPY7d+5g+vTpKFiwYKrW/JqCejPatm0bPD09kStXLrRt2xalSpWCpaUl2rRpozXwVPuhb968eaIa2oiICKxfvx5du3bVBlpJza4D1W0vXboUjo6OKF++PPr374927drB2toajRo10voO379/P3LkyIEKFSokerNz584dTJkyBRUrVky1sv6bd72d6ty5MwIDA7UA/OHDh6hbty5y586NCxcuAEgcUDx48EB7sHz16hWmTZuWiqVPPXq9Hk+fPoWXl5fWnZ76u9y7dy/8/f1RunTpJNe9efMm1q1bp33fpr4GZVSrVq2Cv78/ChQogBo1akBRFJQqVUp7aH316hUWLVoEMzMzTJ8+PdG5GBkZiZ07d2LlypVaW4ikemtJKadOnULdunXRp08flCxZEvb29ujUqZPRMoUKFTLqPcXw9xcbG4uLFy9qlU5XrlzBn3/+mWrlJTIVBuUZwNKlS7WRKZctW4YlS5agYsWKsLKy0rq4e/bsGZo3bw4fHx+tSzkg4WJ2/vx59O3bF0FBQVqtZHrwzTffwNPTE/v379du0iEhIcidOzfc3Ny0i+7r169RqlQprYvEqKgobNiwQRuIpG3bthk2lzA8PFwL7JJKOapTpw6qVKmC48ePa4HP0qVLtW741BrwgQMHQlEULFq0CEBC+4BDhw7h008/1brNfPDgQYrn7CaVk/7ixQuULFkSn3zyCS5duqRNVx+uDG/GAwcOhIODgxagPnjwABs3bkTFihWh0+nQsGFDvHr1yqQDOqkPOoaf3eDBg+Hj42P0Zmrfvn2wsbHBgAEDjFKNnjx5gm3btqF69er47LPPtJQVVUbIo/7xxx/Rv39/ra/02NhYuLu7Y8SIEQCMz4OhQ4canYvx8fF49OgRNm7cqPUMtHv3bpMcR2awefNmODg44JNPPsGaNWuwceNGtGjRQhtoS63MuH37Nho2bAgfHx/tgTw2NhanT59Gv379oCiK1hA5Nand7VapUgUDBgzAp59+Cjs7O5iZmWHv3r3acgcOHNDaO6nX87i4ONy8eRM//fQTcuTIgSZNmmTIdkFE74tBeToXHR2NsmXLonbt2lo3UEBCPqeiKKhbt66Wk3vs2DHY29ujd+/eePz4MUJDQzFjxgwEBATA2to6zfNx/210zLJly6J9+/ba3wsXLoSHhwf8/Pzw008/GY1uuHv3buh0OrRv3x6tW7eGTqdDmTJljLr3yigjcaquX7+O3r17Y+zYsdq0f/75Rxt8Rr1JqWkpQELtcnBwMKytrTFnzhwt/Uiv18PV1RVlypTB1q1b8f3338PJyQk+Pj5pniKxbds22NvbY9euXdq0hQsXwsfHB4GBgVi5cqV2zt64cQPFihVD3rx5sXHjRnTp0gVmZmaJvltTiImJQalSpVC0aNFEPVPs3bsXiqIYBeUvXrxAz5494eTkhIMHD+LFixc4efIkunbtCmtra+TNmzdRv/vpLSBPKlXl6dOnqFSpEmxtbZE9e3YMHToUW7ZsQdasWTFr1ixtPfX3d+HCBZQoUQLe3t548uQJTp06hXbt2kGn06F8+fJpNspqRvfm96B+vr169YKTk5P2RgZISKH69NNP4eTkhE2bNmnTt27dCnt7e3z55Ze4fv06pkyZAl9fX9jb22PGjBmpfgyxsbGoX78+cuXKhbNnz2rT165dC3d3d1SoUMFo+aZNm2rHEB0djXXr1qFs2bJQFAVdunRhmhNlegzK04F3Pfnfu3cPdnZ22oX2+fPnWg8P5cuXx/r167XeGmJjY7X+ur/77jutJvmzzz5LsgeE1GR4QwkJCcGlS5eMRgR9/PgxypYti44dO2L37t0IDg6GlZUVvvjiC1y+fDnJHijUfnZz5MiBtWvXatP/LfhPr6KiolCyZEn4+fnh4MGD6NGjB6ysrPDTTz8BSAjAbW1tcfPmTTx58gTNmjWDoiho0KABDh8+rNWSq3mkS5YsgaIosLW1haWlpdY9XWp69OgRWrdubdRP+Pbt26EoCp4/f46TJ0+iWLFisLKyQu/evXHp0iXtzYD6nU2bNg3W1tZaQ0nD7zatvC1VZdasWShSpAh8fX2xa9cu7bf6559/wtXVFT/88IPR8hcvXoS3tzfq1q2Lb7/9Fp6ennBxccHChQu1ZdJbIK4y/AzeHAQpPj4eFy5c0HLhfX19oSgK5s2bpy1jeB2bMWMGHBwcUKxYMWTLls1k32tGlNT9wPCcqVChAooXL56oS82IiAhkyZIFzZo102rGHz58iC+//BIWFhbIkycPFEVBjx493jnI1Yd61/VXbRPUuXPnRPOGDRsGBwcHzJkzR5sWEhICCwsLVKxYUetWtGLFirh8+XKKlZcoPWNQbmKGF9u//voLd+7cMbooHzlyBI6Ojli5ciUWLFgADw8PeHp6YsqUKQgJCUkUTFy9ehXZs2eHoiioUKFCmtc2GpYnLi4OX3zxBezs7ODg4IAcOXJg7ty5WlCmvnK1tLRE/fr1cfDgQaNGchEREUZdff3999+YPn260f4y6qtM9Sa2ZcsW2Nvbw9LSEn5+fpg+fbp2Q/3999/h6emJatWqwcbGBoUKFcLq1asTNdwy/AyKFCmCFi1aJPuVdHx8vPZdqOdoUjfes2fPIjAwEEWLFtXKcfDgQbi6uiJ//vwwMzNDnTp1cOjQIaPv9unTp1rj3GfPnqFevXrayIFpQX1AfPOYdu/ejT///FMbGRYAzp07h2LFiiF79uza4FRqsNG/f3/ExcUZDXoyefJkrTekN9s6mLpXlaS+V0OvXr3C0KFD0aZNG/Tq1cvobYfqzJkz2gA/WbJkQZ8+fXD+/HmjZUJDQ1GpUiXY2Njgu+++S5Vjyey2b9+O77//HkePHjXqd7tHjx7w8PAwGvFUPY8HDx4MOzs7o7EmDhw4gICAAFSuXFkboj6lGJ5Dt2/fxqVLl4xSt0JCQuDk5ITWrVtr5536G7h8+TL8/f0RGBhotM7XX38NRVEQGBho0jEIiEyBQXk6cO/ePW1wHBcXF7Rr187oIuzt7Q1ra2vY2triyy+/xMWLF41qOk6dOqXlaMbExOCXX34xee74nTt3sHz5cgQGBmLo0KEYO3YsSpUqBUVR0LVrVwAJAZ2iKKhSpYpRag6QEIB/+umnyJEjR5LDL5s6uPmv3uyNol69elAUBXZ2dkmORlqgQAEtV/T58+dGr29v3bqF/PnzY8mSJdq0lMitP3/+PDw9PbFq1Spt2ttepQMJNd2urq4YPXo0gIQbcZUqVWBjY4PJkycDMK55PX78OPLmzYsVK1Zo32NafZ87d+6Eq6tropv9ihUr4O3tDVdXV1hYWKBYsWJGx3/r1i20adMGiqJg7NixiImJQZs2bbSGqIafx61bt7R+oVXp4XxN6ns1tG3bNnh5ecHb2xsVKlRA1qxZYWtri0GDBgEwPgdmzpyJnDlzolatWvDy8oKdnR06dOiAmzdvag0G//jjjyS78aTEDB+sHz58iIYNG0JRFNjY2MDe3h5t27bV5qttM9T0k9evX2vn308//aS9TVNTrqKiolJ8/Ik336p07doVVlZWsLe3R9GiRXH58mVtmdq1ayMwMNCogkg9XvXt35u9rqjtEYg+NgzK01hSN+dBgwahSpUqGDNmDNq2bQsLCwv07NlTa2A2ceJEKIqC4cOHG3VhqNfrsX//fhQvXhzdu3dPs2N4k2GPKK9evUKJEiVQtmxZ5M6dG0OHDtVSUSIiItChQwcoiqI9NHTv3h3m5ubajf/evXvYt28f2rZtixw5cmDGjBmJPrP0+vr/fd27d08bafWPP/7ArFmzkC1bNtSoUUMLqtXPbN26dVr/82pg+/r1a/z555/o2bMncufObTQqYkp48eIFcuTIgZo1axr1jnHu3Dm0a9dO68pMLU9ISAgaNmwIX19fbfk5c+bAyckJdevW1b6vp0+fYu/evWjZsiUKFSqk9bKQFtQytGvXDn5+fkbn1IIFC+Dm5obPP/8cq1evxuLFi1G7dm1kyZIF27Zt044zMjISX331FVxcXNC0aVN8+eWX8PHx0YLvpM5Lw1p0U3vb9wokpL5Vq1YN1atXxx9//IHY2FjEx8ejf//+UBRF6ypPPZYZM2bA3t4eERERCA8PR58+fbS+ptUHMfowUVFROHjwIA4cOIDixYtj5cqVOH78OPr06QNFUTBp0iQACT2PlC1bFo6Ojon6gu/Tpw9y5swJRVEwfvz4VE/re/DgAWbNmoUCBQrg+++/x8CBA+Hv74/g4GCtof6WLVtgbm5u1KOWqlGjRsibNy90Oh3bGhCBQbnJHD16FDdu3MDDhw9RoEABrFmzBgC0YZHNzc21wUlCQ0NRuHBh5MqVS6vlun79On799VdUqlQJRYsWNXn/zWFhYVptd9++faEoCvz8/LTGmmoQdPLkSRQpUgSFCxcGkHCTV2uFnJ2dERgYiGzZssHT0xO//PKLaQ4mFUVFRWlpO4Yjbw4YMABOTk7ajdfwZtq5c2fY2tqiSpUqGDRoEAYOHIgCBQrAzc0Ns2fPTpVybtq0CYqiYObMmVru6rRp04xqToH/BWmrV6+Gr68vOnbsCCCh4Zk6lHqBAgXQuHFjfPLJJ8iWLRsCAgKwdevWVCn3u7x69Qre3t5afquaylGiRAnUrVtXS6cBEhrIKYqCFi1aJOotZdy4cfDz84OiKMiVK9dbu2ZLL8G4IcPv1fDthdoOQO3NCUioWS9RogQURcHs2bONanOHDBkCd3d3o3M4PDwcv//+e9ocSAb3ZtpddHS0VmFRokQJo0b5oaGh6NKlC7JkyaJ1g7p8+XK4urqiRIkS2LhxI65cuYJly5ahYMGCuHLlChYvXpyov/zkeLMBcFRUFIoVK4aGDRuicOHCGD9+vHadOHXqFMzMzNC9e3c8efIEcXFxaNu2LWxtbTF+/HgACT0Sbdq0CYULF8Y333yDTp06aWl7RB8zBuVpwPACrAaliqLA2toa5cuXR4sWLbT5er0e0dHRKFy4MIoVK6a98jtz5oxWA+Lt7Y3ChQvD1tYWZcqUSZTPmVqmTJmSZE7i1atX4e7ublRbHxAQAEVRsGPHDgAwGi75+++/N+pV5OnTp9i/fz+GDBmCvn37YurUqUbbz4iNOIG3l3v37t1wd3dH//79tWn37t1D3rx5UaRIEe3hRq0tf/ToEZYuXYp8+fLBx8cH+fLlQ+vWrZMcNTIl1atXz6jHkNevX6Np06bIkycPDh06BOB/teXR0dHo1KkTnJ2dtQfEqKgobNmyBQ0aNEDp0qVRvnz5NM0vnjt3rtGN/syZM1AUxajR5b1792Bubq51zfbkyRN88cUXsLCwQKVKlbBt2zYt2FC/z9evX+PIkSNaY0c1dSw9t28wLFu9evWQO3duo94wFi1ahGzZsuHZs2eIiorSBgGqWrUq9uzZo7VRUH/H/fr1g7+/f5Lbp/d39+5dLdhdv349ihQpAltb20S5/H/88Qd8fX3RunVrAAlvPTZt2gR3d3coigI3NzeYmZmhQYMGRqmPKc3wmtazZ08oioK8efNq09TrwaBBg2Btba31/PTs2TMUK1ZMW75KlSpwd3c3yjUnIgblaSYiIgK3bt1Cv3790KhRI0yaNAk9evSAmZkZnJ2dteBBvblt2bIFiqJgxIgRWsrKrVu3sHDhQgwYMAA9evRIs67uFi1ahKxZs8LKygqHDx9OclQ4Dw8PbSARIOEGoygKevfurV2o1eBmz549UBQF8+fPB2Bco/jmiHUZ3alTp7SaVvVzePbsmdaDjmGXetOmTYODg4PRiKYvX77UgvNnz54hMjLSKFc5Nf3111+wsLDAkCFDtKBs27ZtyJkzJ1q1aqUtp35Pmzdv1kaKfVNMTEya3nwPHz6s5eKrn/usWbNgZmaGq1evassdOXIEHh4eWLhwIebMmQMXFxdkz54ds2bNQlhY2Dt7wti8eTN8fHyMHq7SI8Pf1J07d3Du3DktnUBt8Lp06VJYWVnh888/h729PXLlyoVFixbh4cOHRg8jqnLlyqFUqVImHxE4IxswYAB8fHy0NK6nT59i8ODBUBRF66lGPf9evHiBcePGJRrF8uzZs1iyZAkGDx6MLVu2pGp5x48fbzTGwIMHD5A7d25YWVlp+euG54irqyvq1KmjpbvdunULs2bNQr169RAcHIyRI0emanmJMiIG5WngzJkzyJIlC8qVKwdfX18cOXJEm6fmbCaVhtC4cWP4+PiYbCjh06dPo1ixYrCxsUHXrl2TrJE3bLBTvHhxo3nlypWDm5tbokanx48fN7rxvOltw3qnR++qxb9y5QocHR1RqVIlAMbBkdpriWEAGx8fj8qVK8PHxwebNm1CSEgIZs6cia+//jrVyqrX698ZePbp0wcuLi7Ys2ePNq9bt27Inj07Vq5cCeB/D1svXryAp6cnHBwctAcuwzckaenly5fo3bs3HBwccOLECQBA/fr1UbRoUaNBmtRBcGxsbGBnZ4evvvrKqJEakNB7hRp0xMfHG/VGkz9/fu0NUXo+Z0+fPo3SpUsjb968aNCggVZjqdbIRkdHa2/i+vfvj5CQEKPP6c6dO6hevbr2mx0zZgwOHz5skmPJaN6sXFB/h0ePHtUaOaqNY48dO4YCBQokupYCCdeTEiVKoGLFiqn6ZiKp68SLFy/QvHlzODg4GN2/5s+fr422rB6net6oXbTOnDnTaFuxsbHsb5zoLRiUp6B3BWjly5eHnZ2d1opevYHr9Xq4u7ujdOnS2uiH6nbOnz8POzs7tGzZMlFea2p6+PAhWrVqpXXrtm/fPkRGRiImJga9evVK1DczkDCgRUBAgFEtpNq7Sr58+XDp0iVERUXhypUraNKkCbJly6blR2Y0U6ZM0foSB95eox8TE4MRI0ZAURQtPUK9YcXGxmL69OlGjV4BYMeOHfDz84OtrS3y5csHRVEwdOjQFCv7kydPtAFfDIOuZ8+e4erVq0YDNgEJtWEeHh5o3bq1VkN/7tw5BAUFoXz58loNelxcHCZMmABfX1/ky5cPZcqUMXlKw7lz55A9e3Y0adIEoaGhcHd3R58+fbT56vc2YcIEKIqCr7/+2ijPOi4uDvv370dAQMBbG1KXLl0a1atXT90DSaaLFy/Cz88PJUqUwNy5c7Fu3To0bdoU5ubm+PTTT7XGgurnYBhgvX79GufOnUOXLl1QuHBhLW2JPpzhb0u9xrdt2xZubm5aLr5hl5pq/92GDelXrFhhNC8tqOfC/v37UahQIdSoUUOb9/LlS5QoUQKFChXS2lYYPpyWLFkSOXPmNBplmojejkF5CniztlGt9TB05MgR6HQ65M6dW0tHUGsL1BoFw9d56kW7a9euKFeuXKJgKbWMHDkSFhYWyJ8/v9ZdlVrTcfHiRdSpU0cLFA1b/i9evBg2Njbaq0r18+jdu7fWIC5XrlwoXbo0LCwskgzsM4LIyEjY29ujQIECiQa0+OOPP3DixAlERERo39+ZM2cQHByM/Pnza8up865du4ZixYohODjYaHCnXbt2oW/fvujWrVuiriKT4+XLl2jZsiUURcHdu3e16d9//z28vLzg5uYGFxcXzJ8/36gB38yZM2Fubo7Vq1dr3+v06dPh7u6Opk2bYvfu3Vi6dCnKlSuHtWvX4tSpU+ki9Sg+Ph5jx46FoigYMmQILCwsMGHCBKMHYiDh9xoQEICcOXNi9erVABK+m1WrVqFMmTIIDg42qh0EEh5o1q9fDwcHB4wcOTJdtHt4s6Ze/Xv69OkwMzPDr7/+qs179OiR9ltfvXo19Ho9Xr58iSpVqsDKygodO3bEokWL8MMPP6BEiRJwc3PDzJkz0/XbgPTq5cuXqFChAtq3b58oN//Ro0ewtLRE27ZttTYiV69eRY0aNeDu7q7lh6vn17179zB06NBUHX/i+fPn6Nu3rzYgmOFvediwYXB2dsayZcu0aTt37tQeatV7n/rAf+DAAbi4uBi1XyCit2NQ/h+otWlvdncWERGBHj16oHbt2mjfvn2inO/PP/8ctra22qt9wxt5cHAw/P39tZoodV5avuarXLkyLC0t8d133+H27du4fPkyqlSpAgcHB6NRGAcOHAhnZ2fUrl1bCyYvXrwInU6npTSon9Hz589ha2uLEiVKYOrUqViyZIlRbXpGcPnyZe0tBpDQ3Z+9vb2WS3zz5k2UL18eZmZmsLCwQN68eTFhwgRt+Xnz5kFRlCRrqGvWrJlkzVdKBz/q9oYMGQJ7e3ucPHlSu/l7enqib9++GDNmDJo3bw5nZ2cMGzZMWzc2NhbFixdHuXLltO/u+fPnmDZtGqysrKDT6aAoCurUqWM0amt6EBISgqpVq2pvfezt7VG8eHGMHTsWJ06c0HKiz5w5g/z580NRFGTLlg2FChWCtbU1KleujCtXriTa7q1bt1C9enUUKVLE6AHHFP5tRNvu3bvD09Mz0Xfz9OlTFC5cGOXLl9e+1/v376Nfv35wcnKCvb29Njqp+rBN76Y+tL754Ne/f384Ojoa9U6jXiO///57mJubGwW6P//8s1FPR6n10JfUds+fP49s2bKhatWqiUYNvnjxIipUqICCBQsajbrcqlUreHh4aI2eU7PMRJkZg/IPtGDBApQtWzbRsPXz589H1qxZkTNnTlSrVg2enp6wt7fHtGnTtGVu3rypDcyh3sjVoFvNs27VqlWSNe1p4f79+9i3b59Rg7zFixfD3NwcvXr10qbp9XosWLAAWbJkQa1atXD06FE8fPgQ2bJlM8p/VmtYxo8fD0VRjD4LwwEv0rMHDx4gT5488PLyMipv+fLlkSNHDuzduxfff/89ypcvj2nTpmHRokVaLwNqYH7v3j00bNgQjo6Oic6bBg0aIEeOHFAUBWFhYcku79s+U3W6Wqu1fft2zJo1C76+vti4caNRwFa+fHnkzJnTKID4/ffftT6oDQcoOnfuHH7++WeTd8n5LitXroSjoyNatWqFb775BiVLltSCdE9PT3Tq1AmHDx/G9evXsWDBAowZMwa9evUyajhnmEsOJDyopORbjH/zPqlAV65cwdSpU3Ho0CFt0J74+Hjt93fjxo1E25o/fz7MzMwwdepUo+tOWFgYbt26xeHN3/Dm4F+qN7+fN3/nT58+hZ+fHxo2bKilgRmu4+3tjapVq+LixYsAEq7FaheJhhUCKXkc77r+jhkzBj4+Pto1zPB4p0+fDmdnZ6M3u1evXoWDg4OWKkZE/w2D8g/Ut29fWFhYaP3IxsXFYdu2bQgKCkLPnj2NRlHr2LEjbG1tjWqZRo8eDTs7O6MLmnpxrl+/Plq0aJEuuohSL9gRERHaSIZv1nDv2LEDuXPnRvbs2XH58mXkz58fXbp0AZD4LULOnDlRoECBJPMO07PXr19j/vz56NmzJ2JjY7Uaox07dsDR0RE1a9aEi4uL1s88kHCDat26Nezs7LQ3H9u2bYO7uztatmyJ+/fv4/79+5g+fTpKlCiBdevWYeTIkUb5zMktM5D0Z3zw4EHY2tqiX79+2LNnD+bNm6fNu3Pnjpbe4uzsjAYNGhiV6ZNPPkGePHlw8uTJFClnWomMjESzZs3g6+uLv//+G3Fxcbh27RqmTp2K+vXrw9HREYqiwMzMDE2aNEm0vqlz41WxsbHad2tYpvj4eAwYMABmZmawsbGBtbU12rZtq6VKrF69GlmyZDFqkKr+vtVeksqXL5+mgzllNIafmfq34f9VP//8M6pXr44aNWqge/fuRo1h1TTFJUuWaN+fWikzZMgQWFtb48cff9R+c1u3bkX37t3/89snw3PEsLyGZT537hy+++47zJo1y6iW+8GDB6hYsSKCgoK0N0VquUJDQxEQEAB3d3ejQai+/PJLBAcHp2n7J6LMhkH5ezDMGQ8NDUXZsmWRI0cOrc/umTNnok6dOkYNF/ft26c11GvTpo02/dWrVyhSpIhRgKpemNNDHi6Q+Ebz+++/w83NzaiBj7rM4cOHUaFCBXh4eMDDwwO5cuUyWlc9JrWLx8GDBxulb6Rn6jEafi+G/27Tpg1sbGxQokSJRLVOx44dg5+fH5o1awYgId1jzpw5MDMzg4eHB4oVKwZra2uMGDEiRT+PlStXwt3dHTt27NBeLxuWLSYmBq6urvj000+NhkCfNWsWbG1tERQUhG3btqFz586ws7MzGu76ypUrUBQFPXv2NHp1nREcOHAAjo6O6NGjR6Lv8+bNm1iyZAmGDRuG8PBwbV56epNz+vRpODg4aOlhqr179+LRo0cIDg7GhAkTcODAAfTr1w8ODg7aQE5PnjxB7dq14ejomCg3fu7cuciTJ0+G/V7TgmFwe+LECXh7eyfqb//Ro0do06YNrK2tUb16dZQrVw62trZwcHDAr7/+qgW05cuXR7FixRLVfvfv3x9mZmbInz+/1ij8v1ZcGJ63jx8/xpAhQ4x6TwIS7jm9evWCubk5smfPrr056tatmxZoL1y4EL6+vvjiiy+09dQyNWzYEBYWFlq/6QAyzHWdKD1jUP4WUVFRWL58udENXL04z58/Hy4uLkZ9tqoXqxcvXqB169ZQFAWNGjXSGjoapgKsWbMG1tbWRoMGpUfqMb169QqDBg2CoijaK33DC3B0dDQaNmwIe3t7eHh4vLVRT5kyZaAoSoapaTX87mNiYtC4cWMMGDBAm3b9+nW4urrC3Nwcd+7cMVrn+fPn6Nevn9E8AFi3bh06d+6MGjVqGDW8Syn79+9HYGAgXF1dMXbsWKN56vnbpEkTFCpUSJu+a9cuZM+eHT179tQeNP/8808oioKyZcsiJCREW3bhwoUZMqXh1atX6NevH6ysrLTA9G1BT3oKxlV3795FoUKFUKpUKa0RrjoIV+HChdG5c2ft4f7ly5fa71UdoXPjxo3w9/eHn58fduzYgRs3bmDPnj0oWbIkFi5ciLVr16ZKmkRGZngePHr0SHuLVLly5UQj0m7atAlOTk6YMGGCloZ2+PBhFC1aFDly5MDOnTsBJDT4NzMzQ9++fbUUl2PHjqFu3boYPXo07O3t//MbizfP5++//x6Wlpbw9/fHsWPHjObPnTsX9vb2GD16NP766y9cuXIFPXr0gIWFhTYGgV6vR/PmzRONwHvw4EF4eHigUqVKsLe3z3BthIjSMwblSTh79qz2Srthw4bayJOqV69eoWHDhnBzc9NqIOLi4nD//n1Ur14duXLlwsqVKxEdHY2jR4/CysoKlStXNsrFrVq1KoYOHZouAwBD6oX8zz//RKFChZAzZ06jeWqgFxISojVYejNFRV3mxo0b2LBhQxqWPvkiIyO177hNmzbIkiWLUS7xqFGjoCgKvv32W22aeryzZ89O9ECWFsLDw7W8dsMBYoCE76R3796wsbHRRuocMGAAnJycjEa/3LJlC9zc3ODr64uePXumaflTy5UrV5A3b17UqFHjre020vPvcfXq1dDpdJg4cSKAhAewUqVKwcbGJlFPGWfPnkVQUJDRw9evv/6KwMBArTFrlixZULhwYS2PmZI2fvx4WFhYIF++fJg3b16Sg3c1btwYHh4eRg+wwP8ebjt06KD9DgcNGgRzc3NUrVoVX331FcqUKYOyZcsCSJnzb8OGDfD19YWjoyOGDRuGc+fOGVUwvHz5EuXKlUNQUJDRdL1ej3bt2kGn02kVBgcOHEDRokWRM2dObN26FevXr0fLli3Rv39/nD9/3qjCgYiSj0F5EtSeFRRFgZeXF3Q6HUaNGmWUK7d161Z4e3ujVq1a2rT169fDxsYGCxcu1F4D6/V6ZMuWTWskp8qIr4mnTJkCRVEwfvx4AInTbS5dugQ3Nzej48xoDGuT9u7dC3d3d/j5+SEmJgZnzpyBn5+f0Svb6Oho5M+fHwEBAdogNapp06ZBURSjUTvTyuXLl9GlSxctfcqw8dWCBQugKIr2sNGpUycEBARoDxt///03WrRogQ4dOqB+/fpYsGBBmpc/Nej1eowePRqKoqRql3Kp5dmzZ2jWrBk8PDy0VDn1oXD48OEAYNQGYNasWYl69rl9+zamT5+OXr16YerUqWl7ABlQx44dtQb4ISEhia7bhv2NqxUW6nVRnde5c2d4eHgYXQeGDRuGAgUKwMXFBTVq1EiRRsPR0dHam9mOHTviwIEDiRqcAgkNzz09PbXUOuB/583hw4fh7++PypUra/M2btyIAgUKaCkuOXPmTJQGRUQpg0H5W6xevRp2dnbo06cPOnToAFtbWxQvXtyoMUz37t1hb2+vdXE4Z84cKIqidbP24sULjB07FkWKFEHlypUTjSaYUag3l1u3bqFOnTqws7PDkydPABgHsWFhYXB1dcXIkSMzTENOIHFDviNHjmgDprRo0QK///47Xr9+jRcvXmgDrBjmaC5btgyKoqBo0aK4dOkSbt26hf379yNv3ryoWLGi9lmltbi4OPTp0wc6nQ4VK1bUArnLly/D3Nxc63/+119/hbu7O4oUKYJ+/fqhevXqyJYtGy5evJhuGjmmlMjISKM+2DOaAwcOwMnJSWu0+c8//6BcuXJwcnLS2gioAdatW7dQr149uLu7J+o1JCP9PtPCm5+H+nnt3bsXLi4uaNq0aaKG2H/88Qe2b98OANogYWrDbsOG7ps3b4aiKFoKi+rx48dJdrf5X0VGRqJHjx4wMzNL8o2kYaVSwYIFUb16dW38C8Pjb9KkCXx9fY0eXMPDw7Fhw4a3jsJMRCmDQflbREZGomnTpsiRIwcuXLiALVu2wN/fH1myZMFXX32FmJgY3L59G4GBgShSpAiePn2KI0eOwN7eHm3atMGmTZswf/58BAUFYcKECUYD7WRkagA6evRoo+kvX77Enj174OTklGhY5fTm9evX2Lt3b6K0Er1ej/DwcHh7e8Pe3h5FihRJVPt97do1lC5dGiVLljS6SderVw+KosDHxwdBQUHInj07vL29sWPHjjQ5prd59eoVZsyYATs7OxQrVgy7du1CeHg4PDw88Mknn2jLTZs2DWXKlIGfnx/KlSuX6XviyKgPG4a58WoAuGDBAlhbW6NHjx4AjI9tw4YNUBQFnTt3Nkl5MxLDwNTwLWCnTp3g7OyMhQsXAkjo2rZ58+ZQFEUbBG3nzp3Ili0b6tSpk2i7v/zyCxRFSdRINyWp5T1+/DgKFCiA4OBgbd6rV6/w3XffwcLCQusLfezYsbC0tDS6PqkVRj/88AOsra2TDNiJKHUxKH+HAwcOwN7eXuuj+/Lly/jyyy+hKAqCgoJw4cIFjBo1Ch4eHhgxYgSA/6UsWFtbw8LCAm3atDFZv+MpSb0wP3v2DH/88Uei+VevXoVOp0PevHlNkq7xIV6+fIlWrVohf/78Wu3R4MGDMWrUKAAJjaAcHR3h7e2tPUwZvpJeuXIldDod5s6dq23zyJEjcHFxQZEiRbBr165EA0eZguHNdOPGjXB3d4ebmxv27duHwMBANGzYUKtdVUd0VBt6Uvp15coV5MqVCzVr1oRer8fjx4/RokULmJuba7WbaoAVERGBwYMHp2pAmBmEh4fjxx9/1Lq0NXTr1i14e3ujfPny+OKLL2BtbY2iRYtizZo12uA6QMLoy4qi4Mcff9SmhYaGok2bNsiRI0eifPPklnfv3r2JxjbQ6/X46aefoCgKFi1ahM2bN8PX1xdubm4YN26cdj07c+YM8uXLhwoVKhi1KYiIiEClSpWQK1cuPHnyhAE5URpjUP4OhrVShv3NLly4EAULFkTWrFnRsmVLrYW92ivFyZMnsWHDhnQfnCbXm42SfvrpJxOV5P2ptYirVq2Cvb09qlWrBhcXF3h4eGD27NkAElICmjZtChcXF63RKvC/IDcsLAwtWrSAl5eXVpsEJIxq9/nnn6fpKKwfYseOHQgODoarqyvc3Nzg5eWllTU9N3AkY3q9XmvfoTbw3LZtGzw8PIxqahlQvb9r166hVKlSKFmyJID/tRlSU07GjBmjjQg7bdo0hIaGJkoFunz5Mho3bgxFUVCvXj0MHDgQLVu2hJWVFYYPH45Xr16l2Heyd+9e+Pv7Y8aMGQASxhj49ttvcfPmTdy9e1cbKdjCwgLdu3fH+fPnjWr/X716hVWrVsHc3BzFixfHunXrsG3bNgwbNgxOTk5aY2IiSlsMyv+FWitVp04dLVccSOjyrm3btvD09NQawDRs2NCEJTWd9BzQPX78GGfPnsXLly+Nbohnz56Fvb09zMzMUL9+fVy7ds3o+924cSPs7OwwaNCgJI9v9+7dcHV1NerDN722FzA8bnVQEPWcfTM9hzIG9XvMly8fHj16hPj4ePTv3x+KomD16tWmLl6GtGjRIiiKgoIFC8LCwgI1a9bUum999eqV1gvJu1K7QkND0b9/f/j7+8Pf3x8FChTAqlWrUqW8BQsWROHChdG2bVsoioKKFStqOepLly6Fl5cXmjdvDsD4GmD47zlz5mgjCjs4OMDOzk5760tEaY9B+b8wrJVSXwGrF7Xo6GgcOHAAOXLkgIWFBb7//nuj+WRacXFxWL9+PYKCgrQb6cOHD7Fx40aMHz8eJUqUgKOjI+rUqaMF3mptUkREBD7//HPY29trXQcC//tuIyMj0bVrV3h6epqsIeeHUt8SXLlyBePHj8eBAwdMXCJKjvXr18Pa2lpr3/HHH3+gWbNmGWYcgPRC/e0PGzYMiqLAxsYGCxYs0BoEq7/5NWvWwM7ODt26dfvXbb5+/TrVUsHU8nbv3h3m5uZaf+O3bt3SKgYePHiADh06wNzcXOtH/G2VJ3fv3sXu3buxevVqjsZJZGIMyt+DWitVoEABowab6kXu8uXLGbpHh4zuXQ9Bp0+fhre3Nxo3boyvvvpKGxAnLCwMsbGx6NKlC1xcXLQedAxvXEeOHIG3tzdatGiRZEpKaGhouhmFlT4+0dHRaNWqFRRFYVuADxAfH5/omhEbG4upU6eiY8eOsLKywldffWW0vKpevXrw9PTExo0bE81LavmUoNfrE/WeExUVhTJlysDDwwOurq7aoG6Gy+zatQteXl6oX79+ipaHiFIPg/L3pNZKjRs3TpvGGvH05ciRI1qOt+GNsU2bNtDpdLC3t8eIESNw8eJFLcj+559/kDNnTpQuXVrry1vtVSU6Ohrff/89LCwskrzpEZnasWPH8O233/Lh8D0Z9kxz69YtPHjwQGvsDCT04965c2fY2NjgwoUL2nT18z1x4gScnZ3RoEEDbb3UTN97Mw/c0JMnT3Dy5ElYWFjgs88+0yqM1GN88eKFVvuvDoDHaxdR+sag/D0Z1kpxWOH0JTY2Fn369IGiKKhbt67R24w///wTRYoUgb29PQIDA41qFNWb6ejRo+Hg4GA0KqfqypUryJcvH3LmzGnU0wIRZUwPHjxAixYtkC1bNri5uaFQoUJGed87duyAp6enlo8NJASzakDbu3dvuLq6Ytq0ackuy/sGyT/88AMaNGiAtm3bYtasWUYpcwMHDoSVlRVWrFihTVMD89OnT6No0aIoUqQIA3KiDIBB+QdgrVT69Pz5czRp0gQFChSAg4MDatSogW3btgFIeM17+fJlLFy4EObm5vjxxx+170+9cUVHR6NkyZLIkyeP1qXc3bt3te7Rli9frvVyQETp39SpU/H3338nmn758mXkzZsX+fLlw/DhwzFs2DAEBgbCwsJC6xI1MjIS3377LRRF0a4jwP+uF/fv34etrS1KlCiBR48epUh5z549q42++WaD9CJFiiBr1qwoX748fH19oSgKqlSpghs3bgBI6KbW09MTVapUSbLCaPbs2VAUxegtLxGlTwzKKUNTb2A9e/ZEzZo1cfToUVSrVg1Zs2bFgQMHtDSV8PBwtGzZEtmyZcP58+e19dUAffXq1ciaNSuKFy+OWbNmoWrVqvDz89NufESUMRw4cACKoqBbt26JBokaN24crK2tsXHjRq1R5N9//41atWpBURStG9tTp06haNGiCA4ORlRUFJ4/f479+/drYzRs2bIF9+7dS5Hyqt0tJtWXfK9eveDj44Nt27YhMjIS8fHxmDRpEmxtbdGxY0fcvXsXADB//nwoioLJkydr6964cQORkZEIDQ1F9+7dcfDgwRQpLxGlHgbllKGpQfnWrVthYWGB+/fv4+HDh6hbty6yZ8+OsWPHasvu379fG6I8Ojo60baGDRsGf39/2Nvbo0yZMjhz5kxaHQYRfaA3c7nVa8GTJ08wcOBA2NraaqOeAgkP4PXq1UOePHmMpgEJKSteXl6oVauWNn3BggUwNzdH2bJl0a5dO63b2/+aBvJmY031/w8ePICVlRUaNWqkBdl6vR7379+HnZ0d2rRpY7Sd2NhYDBs2DJaWlkZpN6VKlYKnpyfGjRuHWbNmoXDhwhg/fvx/KisRmQaDcsoUdu7ciezZs2Pt2rUAEhpFdejQAfb29hg1apRWyzRgwABYW1tj79692rpqA8+4uDjcvn07yVH9iCj9iYmJ0X6/hkH6qVOntPEl1BGV9Xo9SpcujXz58uHOnTsA/peS8vLlS3Tp0gVWVlbaw3hkZCSmTp2KQoUKITAwMFUGR1MfCkaNGgVFUTB9+nRtXnh4ODw9PdG3b1+tjKqwsDDkyJEDrVq10o7h/Pnz8PHxgZ2dHaytrdGqVSu8ePEixctMRKmHQTllCpGRkXByctJG5QQScsW//PJLmJmZ4bPPPsOzZ89w48YN5M+fH2XKlMG6deuwbNkylChRAr///jsbQhFlAOrvdOzYsbC3t0flypUTPUi/fPlSG1/i559/1gL2kSNHwtraGtu3b9eWVXs1WbhwIRRFMXpgBxJGy3yz55P/YteuXShbtixWrlypvakz3K6vry+KFi2Ks2fPAkjoCSY4OBgFCxbUlleP48WLF2jXrh2cnJyMgvWbN29i+/btuH79erLLS0Rpz0yIMji9Xi8ODg5SvHhx2bp1q4iIPHr0SEaPHi3z588XPz8/2bRpk7Ro0UJevHghkydPlj/++EM6dOggnTt3FicnJylTpowoimLiIyGif6P+Tnfv3i1RUVFy4MABadSokXzzzTcCQERErKyspE6dOlKhQgUZO3ashIeHi4hI69atxcHBQRYtWiS3b98WERELCwsREW1dBwcHo799fX21Zf6rkydPymeffSbHjh2TXr16SceOHeXu3btG2508ebKcPXtW1q5dK3FxcZI9e3YpX7683LlzRxYtWiQiIvHx8SIiYmNjI/Hx8eLo6Kgdm4iIv7+/1K5dWwICApJVXiIyDQbllOGZmZlJXFyc+Pr6SmRkpPz4448SHBwss2bNkr59+8q6detkwYIFcvToUenatatYWVnJpUuX5Mcff5Tjx4/Lzp07tRsxEaVvamC6aNEiURRFSpUqJX5+fjJ9+nSpVq2a9mCeK1cu6dWrl1y7dk3mz58vIiIBAQEyePBgWbdunfz444/y6tUrefXqlfz5558yd+5cqVy5shQuXFhEJEUf0osVKyYtWrQQNzc3cXR0lCtXrkilSpVk1qxZ2jLNmzeXsmXLypIlS2TPnj0iItKvXz9xd3eXH3/8Uc6ePasF8efOnZPjx49L6dKlxdfXN8XKSUQmZuKaeqJkU1/pTpo0CYqiIEuWLGjRogUOHDiAZ8+eacutX78etra2KFCgQJIjdBJRxqDmUX/55ZdwdnbGTz/9hO3bt8PHxweWlpYYP348wsLCEBcXh5YtW8LFxQUXL17U1m/dujUURYG7uzsqVKiAgIAAuLq6aiN1piT1+nT8+HGUKlUKvr6+OHfuHMqWLQtFUdClSxccO3YMAHDmzBkoioLPPvtMGyV68eLF8PHxga+vL/r27YsxY8agSpUqcHFx4aBARJkMg3LKNA4dOgRnZ2cMHz4cr169SnKkPfaoQpTxvPlbNhxrwNnZGWXKlEFERASuX7+Orl27QqfToWjRovjrr7+wbds2+Pv749NPP9UaPkZFRWH9+vVo06YN6tati169euHx48cpVt63Bcnjx4+HoiiYM2cOAODHH3+EjY0NPDw8tC4RBw0aBHNzcyxbtkzb1r59+1C2bFk4OTkhW7ZsKFOmDE6ePJli5SWi9IFBOWUaJ06cgJeXF4YOHapNYw0SUcal1+uN+ho3HMlSpfbRPXbsWO33PmfOHAQGBiJbtmzo1KkT+vXrB0tLS+zbty/RNSEmJiZFy/yuweVu376NGjVqwMHBQes1ZuvWrahXrx4URUHLli1x8eJFWFtbo27dukYjEMfExODRo0e4cOFCipaXiNIPBfj/1ixEmUBAQICULl1alixZIjqdTszM2GyCKL17/PixmJmZiZOTk+j1+kS/2zt37sjQoUPl7t274uTkJNWqVZM+ffpo84sUKSJRUVGyePFiqVixooiIhIWFyaBBg+TXX38VJycnefDggdStW1eWLl0qWbNmTZFyx8XFibm5eZLzZsyYIc+ePRNPT09p1aqVWFtbi6Iosnz5cundu7fUr19ffv75ZwEgiqLI8OHDZcaMGVK4cGH5559/BIB88cUX8tVXX4mIJPm5EFHmwl84ZSolS5aUjRs3ioWFBW9gRBnAwYMHJXfu3DJ79mwREe13q9frRURk+fLlUrhwYTl37pz4+PhISEiI9O3bVwYNGiQhISEiIjJ9+nS5efOmrFmzRqKjo0VExNPTUxYuXCg///yzuLi4iIjItm3b5P79+8kq74EDB6Ro0aIiImJubi5v1mtt3rxZvL295fvvv5cZM2ZI586dpUqVKrJ582YREWnYsKE0adJEVq1aJceOHdMalI4aNUr27t0rFhYW8vjxYwkNDZU1a9bI9evXjT4XIsrETFpPT5TCxowZg969eyM+Pp6pK0QZwKtXr1CgQAFUrFgR586dA2DcH3dwcDA++eQT/PXXX9o66tD0P/30E2JjYwEAzZo1g4ODA9avX59oH9evX8ecOXMQEhKS7PKq/ZmPGDECAIzSa/7++28ULVoUTZs2xcGDB3Hr1i2cP38euXLlQt68eXH58mUACYOd+fn5oWzZstq66nYePXqESZMmoWTJkloDUCL6ODAop0wlqcadRJQ+qYHo6tWr4eTkhCFDhhjNX7VqFczNzfHnn39q09avX4/cuXPD1dUVa9eu1QbWCQ0NhU6nQ5MmTbR87Xfld7+vyMhILFy4UAvoHzx4gM6dO8Pc3Bz3798H8L9BgH788UfY2toaNcIMDw9HQEAA7O3t8fPPP2vThwwZAkVRsHTpUqNtENHHi+/DKFPhK16ijEOn04mISMuWLaVs2bKyceNG2b17tzY/Pj5esmTJIvnz55d//vlHqlWrJs2bN5fSpUvLjh07pG7dumJraytxcXHi6ekpw4YNk40bN8rPP/8sIvLWfO9/A4OUlFWrVkmXLl3kwIEDAkDc3d2lTZs24ubmJr169TLaz6lTp6RixYoSHBwsz549k969e4uXl5e4u7vLmjVrpG7dulpaTps2baRkyZLy7bffSlxcnNFAQmBTL6KPEiMYIiJKU4ZBpzoY0MiRIyUiIkJWrlwpz58/FxGR2NhYsbKykvr160tgYKDExsbK1q1bZebMmVKsWDGxs7OTmJgYLSgeMWKEuLq6io2NzX8ul16v1/K8V69eLSdPnpQmTZpIxYoVtellypSRLl26yPr16+XIkSPadC8vL7lw4YLMnz9fcuTIIZs3b5Zp06bJxo0bpWbNmuLk5CSvX78WEZFChQpJ06ZN5e7du7Jq1SqjcnB0YaKPlEnr6YmI6KO1Y8cOo7979uwJb29vLFmyBEBCF4iurq7Q6XQYPXo0YmJitBxyAPjrr7/g4+ODffv2adPUdJYPZZgbbji4T79+/fD06VMACfne6sBjp0+fRlBQEIoUKaKtt3DhQri4uGg555cuXTJKoblx4wa+/vprnD9/HkBCagvHTiAiFWvKiYgoze3du1dq164t33zzjTbtu+++E0tLS1mzZo38888/4uTkJIMGDRK9Xi8PHz4Ua2trsbS0lJiYGDl8+LBMnDhRnJycxNbWVtuG4b8/hE6nk+joaOncubMUKVJELCwsZMOGDfLNN9+Io6OjLFiwQAoXLixnzpwREZGiRYtK165d5dy5czJv3jwRESlRooQULVpUbG1tpVOnTpIvXz4xNzeXuLg4uXDhggwcOFD27dsn9vb2IiLi5uYmQUFB//ETJKJMx9RPBURElDm8ePECEyZM0HpK0ev1iRpfq3+HhYWhffv2sLW1RUREhDZ/0qRJcHFxwbhx47RpDRs2hKWlJRo3boxBgwahb9++yJkzJ7Jnz461a9emSNnPnj0LR0dHZMmSBbNnz8adO3eMenA6e/YszM3NMXDgQG30zxs3bqBevXpwdnbWjmvZsmXw8PBA4cKFsXz5chw8eBDz5s1DxYoV4e3tjcWLF7NnKCJKEoNyIiJKEfv27YOiKBg5cqRRmgkA7N27F4DxKLs7d+6Eu7s7WrdubbRsyZIlUaJECa1LwFu3bmHGjBnIkSMHsmfPjvz586NHjx6IiopKsbLfunULpUuXRsGCBbXeW96cP2XKFNja2mLnzp3a9HXr1sHW1hZ9+vQBAERFRWHv3r3ImzcvFEWBk5MT7O3tUaFCBS1thYgoKQzKiYgo2dRgu3HjxsiePTv2798PICFX+7vvvoOiKNo0tfu/6OhoDB06FIqi4Pjx49q2Nm/eDGdnZ/Tt29eopv3Fixd49OgRwsPDU+UYVq9eDZ1Oh8mTJ2vTjh07hiJFisDZ2RkhISHInj07mjVrhrt37wIAIiIi0L17dyiKghs3bmjrhYaG4vDhw/jtt99w5MiRVCkvEWUuzCknIqJkU3tR+emnn+Tu3buycuVKiYiIEJ1OJ2XLlpXChQvL2LFjRUTEwsJC9Hq92NraSrNmzaRAgQLSs2dPbVsNGjSQatWqyY4dO2Tr1q0ikjDCp42NjTg7O4ubm1uqHEPdunWlcePGMmnSJDl27Jh8+umnUrZsWfHx8ZE1a9aIl5eXDB8+XDZs2CD79++XuLg4yZo1q7Ru3VqyZ88uX375pYgk9OLi6ekp5cqVk3r16knZsmVTpbxElLkwKCciomSJj4/XuiX08/OT4cOHy4oVK7Q+x2vWrCkNGjSQkydPytKlS7V1RBIaTLZr105Onz4ta9eu1bb53XffyZUrV2TdunUSGxurjUGQmt0F2tvbS58+fUSv10u5cuXk7Nmz8vPPP8vy5culRo0aIiLSsWNHKVmypEyfPl2uX78uIiKlSpWSbt26yW+//Sbbt29nl4ZE9J8wKCciomRRBwFasGCBDB06VLy9vSUmJkY2bNgg165dE5GEAYIKFy4sEyZMkJiYGLGwsJDY2FhRFEXKly8vIiJfffWVts2CBQvKmjVrZNq0aWJlZZVmx1K6dGnp0KGDiCT0nd66dWtxdnYWkYQacDMzMxk3bpycPHlSNm/eLNHR0WJlZSW1a9eWzp07i6+vb5qVlYgyFwblRESULKGhoVKzZk3p27evnDlzRrZu3SrOzs6ybds22blzpwCQggULyieffCIREREyZswYEREt2I6JiZF8+fLJ7du3jdJYPvnkE3F0dEzTY7G0tJTPPvtMAgICZOHChfLy5UttnloDXrlyZfn0009l8uTJ8ueff4pIQo3//PnzpWDBgmlaXiLKPBiUExFRsuzZs0cOHz4sY8eOlWXLlsnGjRvll19+EU9PT1m4cKEWuH7yySdSu3ZtmTx5shw/flxevnwpV65ckSlTpkjVqlXlm2++kUqVKpn4aETy/F979xYSVbvHcfw7y/E1tYsS00gqCcbsbFnRwSKoqIjKCrFCI7BgtCPSyZiUiqKgoKKDJZGVCoadEJHUJNQkuymRoBszCjsIonholJw170U07fZus/fL3rXemt/ncvHMrOd/9+Ph//xXTAxbt26lvLyc27dvf3eNy+Wip6eHrq6un7w7Efld2bzef/jesYiIyHd4PB5fmwp8buWw2Wx4vV7WrVtHXV0djY2NhIeH+9ZUVlayZMkSDhw4QFZWFqGhodTV1bF3714eP37MzJkz6e7upre3l/LycsaNG2dFad/V1tZGUlIS7e3tPHjwgMjIyH9Z093d7fsQkIjI/0on5SIiwr87nzFNE9M0fYG8s7MT+NrKYbPZCAoKwm63ExwczMDAgO//Fi9ezIoVK7h27Rq1tbUAJCQkUFZWhtPpJDQ0lLi4OB49evS3CuQAERER7Nq1i+bmZq5evfrdNQrkIvL/pFAuIuLnTNP0hewvodo0TQAMw8AwDJ48ecLq1atZu3YtKSkpFBUV+X4/bdo03rx5Q1NTE3a7HY/H4wv5a9asobW1lXv37vH+/XsAhg4dyrlz5ygtLaWwsJCoqKifWe5/bcmSJSQmJnLgwAHfhVURkR9FoVxExM8ZhkFHRwc7duzg0KFDvmemaeLxeDh+/Djz5s3j7du3GIZBVVUVKSkpnDhxArfbTUJCApGRkezevRv4fHr+ZYRhX18fhmFw69YtSktLv3lnSEjIzy/2LwgJCWHnzp24XC7GjBlj9XZE5DenUC4i4ufcbjeTJ0/m3LlzlJWV+S5mGoZBa2srubm5JCcnU1RURGVlJXV1daSmppKdnU1RURHTp08nLS2N+vp6jh8/Tn9/PwCvXr2ioqICl8uFzWbznb7/SmbNmsXhw4d9c9hFRH4UXfQUEfFjpmliGAZJSUncunWL6OhoFi9ezKVLlwA4deoUe/bsoaysjGXLlvl+19fXR2xsLEOHDuXevXvY7XZcLhf5+fnMnTuXKVOm8Pr1axoaGqitrSU6Opo//vjDqjJFRP72dFIuIuIHvnxB858ZhsGnT5+IjY1l4cKFhIeHc//+fe7evQt8vsxos9lwOBy+/zFNk0GDBnH06FEaGxtpaGhgxIgRXLx4kf3799PV1cWdO3doaWnh2rVrxMTEKJCLiPwHCuUiIn4gICCA3t5ecnNzaWlp8T03TZPAwEBCQ0Pp6ekhOzubzs5O8vPzcbvdjB49Gq/XS3FxMfC1X9zj8ZCQkEBwcLCv3SUoKIhjx47x9OlTqquraWpqYunSpZbUKyLyq1EoFxHxA42NjURFRZGRkcH27dupqKgAvo5CTExMpKGhgUmTJuF0OqmsrKSkpIQ5c+YQHx/PlStXfBc9BwYGCAgI8E1mcbvdvvd8aYcZO3asJXWKiPyqFMpFRPzAkCFDmD59OgAdHR0kJydTU1PjG4Ho8XiYOHEilZWV7N27l/DwcM6fP4/H42Hbtm28e/eOHTt2YJomdrudrq4ubt68CfDNafiXqSsiIvLX6Dq5iIgfGD16NFu2bKG+vp4FCxYwbNgwnE4nqampZGVl4XA4GBgYoLm5mbCwMDIyMsjJyeHChQtkZWXx/PlzTp48ybx585g6dSoAhYWFLFq0iPj4eIurExH59Wn6ioiIn+ju7mbTpk00NjZSUlLC9evXycvLY9++fbhcLjIzM6murubZs2d4vV5mzJhBZ2cnJSUlxMXFkZuby+nTp+nv78cwDDZu3EhOTo7VZYmI/BYUykVE/EhNTQ3Lly8nPT2d7Oxsbty4wcGDB1m1ahXDhw/n4cOH5OXlMX78eO7cuUNKSgobNmzg8uXL2Gw2+vv7+fDhA2FhYQwePNjqckREfhtqXxER8SOzZ89m8+bNnD17lpUrV5Keno5pmhQUFHD16tVvvly5evVqli1bRlFREfPnzyc1NZWgoCBGjRplYQUiIr8n3cgREfEjgYGBOJ1ORo4cyZEjR/B4PDidTvLy8nA4HISFhREYGOhbv3//fqKjoxXERUR+MLWviIj4Ga/Xy5kzZ8jMzKSgoID169djs9l4+fIlbrebCRMm+NbZbDaLdysi4h8UykVE/FBbWxtJSUm0t7dTVVXF8OHDrd6SiIhfU/uKiIgfioiIYNeuXTQ3N5Ofn2/1dkRE/J5OykVE/NTHjx9JS0ujuLiYFy9eEBMTY/WWRET8lqaviIj4qZCQEHbu3InD4fhm6oqIiPx8OikXEREREbGYespFRERERCymUC4iIiIiYjGFchERERERiymUi4iIiIhYTKFcRERERMRiCuUiIiIiIhZTKBcRERERsZhCuYiIiIiIxRTKRUREREQsplAuIiIiImIxhXIREREREYsplIuIiIiIWOxPaqfpz+PUC3EAAAAASUVORK5CYII="
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "s: pd.Series = error_counts\n",
    "top_num: int = 8\n",
    "\n",
    "top_ename = list(s[:8].index)\n",
    "df_filtered = df_errors[df_errors[\"ename\"].isin(top_ename)].set_index(\"ename\").loc[top_ename].reset_index()\n",
    "\n",
    "df_grouped = (\n",
    "    df_filtered.groupby([\"ename\", \"location\"]).size().unstack().apply(lambda x: x / x.sum(), axis=1).loc[top_ename]\n",
    ")\n",
    "\n",
    "\n",
    "fig, ax = plt.subplots(figsize=(6, 3))\n",
    "df_grouped.plot(\n",
    "    kind=\"bar\", stacked=True, ax=ax, edgecolor=\"k\", lw=2, color=sns.color_palette(\"Set3\", 12)[1:3], legend=False\n",
    ")\n",
    "\n",
    "leg = ax.legend(loc=(1.05, 0.8), fontsize=15, framealpha=1)\n",
    "label_params = {\"fontsize\": 15, \"labelpad\": 15}\n",
    "ax.set_ylabel(\"Ratio\", **label_params)\n",
    "ax.set_xlabel(\"\")\n",
    "\n",
    "tick_params = {\n",
    "    \"direction\": \"in\",\n",
    "    \"which\": \"both\",\n",
    "    \"length\": 6,\n",
    "    \"width\": 1.5,\n",
    "    \"colors\": \"black\",\n",
    "    \"labelsize\": 12,\n",
    "}\n",
    "\n",
    "ax.tick_params(**tick_params)\n",
    "ax.xaxis.set_ticks_position(\"both\")\n",
    "ax.yaxis.set_ticks_position(\"both\")\n",
    "ax.yaxis.set_major_locator(ticker.MaxNLocator(4))\n",
    "\n",
    "for side in [\"bottom\", \"top\", \"left\", \"right\"]:\n",
    "    ax.spines[side].set_linewidth(1.5)\n",
    "\n",
    "ax.set_yticks([0, 0.5, 1])\n",
    "plt.xticks(rotation=30)\n",
    "\n",
    "figures_path = Path(\"figures/\")\n",
    "fig.savefig(\n",
    "    figures_path / \"external_internal_ratio.pdf\",\n",
    "    bbox_inches=\"tight\",\n",
    "    dpi=300,\n",
    ")\n",
    "plt.show()"
   ],
   "metadata": {
    "collapsed": false,
    "ExecuteTime": {
     "end_time": "2024-03-18T23:07:40.842781Z",
     "start_time": "2024-03-18T23:07:40.682690Z"
    }
   },
   "id": "2fce96a1f8fc74e4",
   "execution_count": 15
  },
  {
   "cell_type": "code",
   "outputs": [],
   "source": [],
   "metadata": {
    "collapsed": false,
    "ExecuteTime": {
     "end_time": "2024-03-18T23:07:40.845350Z",
     "start_time": "2024-03-18T23:07:40.843838Z"
    }
   },
   "id": "8c13303b0eb273bf",
   "execution_count": 15
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 2
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython2",
   "version": "2.7.6"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}