sec2.1-failure_recovery_modeling.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "import numpy as np\n",
    "import pandas as pd\n",
    "import matplotlib.pyplot as plt\n",
    "import matplotlib.dates as mdates\n",
    "from util.plotting import *\n",
    "from datetime import date, datetime, time, timedelta\n",
    "import json\n",
    "import seaborn as sns\n",
    "from matplotlib.colors import ListedColormap, BoundaryNorm\n",
    "import warnings\n",
    "warnings.filterwarnings('ignore')\n",
    "pd.set_option('display.max_columns', None)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "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>incident_id</th>\n",
       "      <th>Incident_Title</th>\n",
       "      <th>incident_impact_level</th>\n",
       "      <th>Incident_color</th>\n",
       "      <th>provider</th>\n",
       "      <th>Playground</th>\n",
       "      <th>API</th>\n",
       "      <th>Labs</th>\n",
       "      <th>ChatGPT</th>\n",
       "      <th>api.anthropic.com</th>\n",
       "      <th>claude.ai</th>\n",
       "      <th>console.anthropic.com</th>\n",
       "      <th>Character.AI</th>\n",
       "      <th>investigating_flag</th>\n",
       "      <th>investigating_timestamp</th>\n",
       "      <th>investigating_description</th>\n",
       "      <th>identified_flag</th>\n",
       "      <th>identified_timestamp</th>\n",
       "      <th>identified_description</th>\n",
       "      <th>monitoring_flag</th>\n",
       "      <th>monitoring_timestamp</th>\n",
       "      <th>monitoring_description</th>\n",
       "      <th>resolved_flag</th>\n",
       "      <th>resolved_timestamp</th>\n",
       "      <th>resolved_description</th>\n",
       "      <th>postmortem_flag</th>\n",
       "      <th>postmortem_timestamp</th>\n",
       "      <th>postmortem_description</th>\n",
       "      <th>start_timestamp</th>\n",
       "      <th>close_timestamp</th>\n",
       "      <th>time_span</th>\n",
       "      <th>over_one_day</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>qv89mzdjvtrs</td>\n",
       "      <td>Elevated Errors</td>\n",
       "      <td>1</td>\n",
       "      <td>#f1c40f</td>\n",
       "      <td>anthropic</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>2024-02-26 17:09:00+00:00</td>\n",
       "      <td>We're experiencing an elevated level of API er...</td>\n",
       "      <td>1</td>\n",
       "      <td>2024-02-26 17:37:00+00:00</td>\n",
       "      <td>The issue has been identified and a fix is bei...</td>\n",
       "      <td>1</td>\n",
       "      <td>2024-02-26 17:41:00+00:00</td>\n",
       "      <td>A fix has been implemented and we are monitori...</td>\n",
       "      <td>1</td>\n",
       "      <td>2024-02-26 18:21:00+00:00</td>\n",
       "      <td>This incident has been resolved.</td>\n",
       "      <td>0</td>\n",
       "      <td>NaT</td>\n",
       "      <td>NaN</td>\n",
       "      <td>2024-02-26 17:09:00+00:00</td>\n",
       "      <td>2024-02-26 18:21:00+00:00</td>\n",
       "      <td>0 days 01:12:00</td>\n",
       "      <td>False</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>tmsczhzhjd63</td>\n",
       "      <td>Elevated error rate on claude.ai</td>\n",
       "      <td>0</td>\n",
       "      <td>#333333</td>\n",
       "      <td>anthropic</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>NaT</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0</td>\n",
       "      <td>NaT</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0</td>\n",
       "      <td>NaT</td>\n",
       "      <td>NaN</td>\n",
       "      <td>1</td>\n",
       "      <td>2024-02-16 20:35:00+00:00</td>\n",
       "      <td>We experienced an elevated level of errors on ...</td>\n",
       "      <td>0</td>\n",
       "      <td>NaT</td>\n",
       "      <td>NaN</td>\n",
       "      <td>2024-02-16 20:35:00+00:00</td>\n",
       "      <td>2024-02-16 20:35:00+00:00</td>\n",
       "      <td>0 days 00:00:00</td>\n",
       "      <td>False</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>t6v85cs8j6jb</td>\n",
       "      <td>Elevated error rate on Claude-2.0, Claude-2.1</td>\n",
       "      <td>0</td>\n",
       "      <td>#333333</td>\n",
       "      <td>anthropic</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>2024-02-13 03:59:00+00:00</td>\n",
       "      <td>We are currently investigating this issue.</td>\n",
       "      <td>1</td>\n",
       "      <td>2024-02-13 04:21:00+00:00</td>\n",
       "      <td>The issue has been identified and a fix is bei...</td>\n",
       "      <td>1</td>\n",
       "      <td>2024-02-13 04:30:00+00:00</td>\n",
       "      <td>A fix has been implemented and we are monitori...</td>\n",
       "      <td>1</td>\n",
       "      <td>2024-02-13 04:36:00+00:00</td>\n",
       "      <td>This incident has been resolved.</td>\n",
       "      <td>0</td>\n",
       "      <td>NaT</td>\n",
       "      <td>NaN</td>\n",
       "      <td>2024-02-13 03:59:00+00:00</td>\n",
       "      <td>2024-02-13 04:36:00+00:00</td>\n",
       "      <td>0 days 00:37:00</td>\n",
       "      <td>False</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>x5f4s8nvkyqq</td>\n",
       "      <td>Elevated error rate on Claude-2.0, Claude-2.1</td>\n",
       "      <td>0</td>\n",
       "      <td>#333333</td>\n",
       "      <td>anthropic</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>2024-02-13 00:21:00+00:00</td>\n",
       "      <td>We are currently investigating this issue.</td>\n",
       "      <td>1</td>\n",
       "      <td>2024-02-13 00:50:00+00:00</td>\n",
       "      <td>The issue has been identified and a fix is bei...</td>\n",
       "      <td>1</td>\n",
       "      <td>2024-02-13 00:59:00+00:00</td>\n",
       "      <td>A fix has been implemented and we are monitori...</td>\n",
       "      <td>1</td>\n",
       "      <td>2024-02-13 01:26:00+00:00</td>\n",
       "      <td>This incident has been resolved.</td>\n",
       "      <td>0</td>\n",
       "      <td>NaT</td>\n",
       "      <td>NaN</td>\n",
       "      <td>2024-02-13 00:21:00+00:00</td>\n",
       "      <td>2024-02-13 01:26:00+00:00</td>\n",
       "      <td>0 days 01:05:00</td>\n",
       "      <td>False</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>gr7nb0dwyydw</td>\n",
       "      <td>elevated error rates on claude-instant-1.2</td>\n",
       "      <td>0</td>\n",
       "      <td>#333333</td>\n",
       "      <td>anthropic</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>2024-02-05 23:07:00+00:00</td>\n",
       "      <td>We are currently investigating this issue.</td>\n",
       "      <td>0</td>\n",
       "      <td>NaT</td>\n",
       "      <td>NaN</td>\n",
       "      <td>1</td>\n",
       "      <td>2024-02-05 23:58:00+00:00</td>\n",
       "      <td>A fix has been implemented and we are monitori...</td>\n",
       "      <td>1</td>\n",
       "      <td>2024-02-06 01:55:00+00:00</td>\n",
       "      <td>This incident has been resolved.</td>\n",
       "      <td>0</td>\n",
       "      <td>NaT</td>\n",
       "      <td>NaN</td>\n",
       "      <td>2024-02-05 23:07:00+00:00</td>\n",
       "      <td>2024-02-06 01:55:00+00:00</td>\n",
       "      <td>0 days 02:48:00</td>\n",
       "      <td>False</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "    incident_id                                 Incident_Title  \\\n",
       "0  qv89mzdjvtrs                                Elevated Errors   \n",
       "1  tmsczhzhjd63               Elevated error rate on claude.ai   \n",
       "2  t6v85cs8j6jb  Elevated error rate on Claude-2.0, Claude-2.1   \n",
       "3  x5f4s8nvkyqq  Elevated error rate on Claude-2.0, Claude-2.1   \n",
       "4  gr7nb0dwyydw     elevated error rates on claude-instant-1.2   \n",
       "\n",
       "   incident_impact_level Incident_color   provider  Playground  API  Labs  \\\n",
       "0                      1        #f1c40f  anthropic           0    0     0   \n",
       "1                      0        #333333  anthropic           0    0     0   \n",
       "2                      0        #333333  anthropic           0    0     0   \n",
       "3                      0        #333333  anthropic           0    0     0   \n",
       "4                      0        #333333  anthropic           0    0     0   \n",
       "\n",
       "   ChatGPT  api.anthropic.com  claude.ai  console.anthropic.com  Character.AI  \\\n",
       "0        0                  1          1                      1             0   \n",
       "1        0                  0          0                      0             0   \n",
       "2        0                  1          1                      1             0   \n",
       "3        0                  1          1                      1             0   \n",
       "4        0                  1          1                      1             0   \n",
       "\n",
       "   investigating_flag   investigating_timestamp  \\\n",
       "0                   1 2024-02-26 17:09:00+00:00   \n",
       "1                   0                       NaT   \n",
       "2                   1 2024-02-13 03:59:00+00:00   \n",
       "3                   1 2024-02-13 00:21:00+00:00   \n",
       "4                   1 2024-02-05 23:07:00+00:00   \n",
       "\n",
       "                           investigating_description  identified_flag  \\\n",
       "0  We're experiencing an elevated level of API er...                1   \n",
       "1                                                NaN                0   \n",
       "2         We are currently investigating this issue.                1   \n",
       "3         We are currently investigating this issue.                1   \n",
       "4         We are currently investigating this issue.                0   \n",
       "\n",
       "       identified_timestamp  \\\n",
       "0 2024-02-26 17:37:00+00:00   \n",
       "1                       NaT   \n",
       "2 2024-02-13 04:21:00+00:00   \n",
       "3 2024-02-13 00:50:00+00:00   \n",
       "4                       NaT   \n",
       "\n",
       "                              identified_description  monitoring_flag  \\\n",
       "0  The issue has been identified and a fix is bei...                1   \n",
       "1                                                NaN                0   \n",
       "2  The issue has been identified and a fix is bei...                1   \n",
       "3  The issue has been identified and a fix is bei...                1   \n",
       "4                                                NaN                1   \n",
       "\n",
       "       monitoring_timestamp  \\\n",
       "0 2024-02-26 17:41:00+00:00   \n",
       "1                       NaT   \n",
       "2 2024-02-13 04:30:00+00:00   \n",
       "3 2024-02-13 00:59:00+00:00   \n",
       "4 2024-02-05 23:58:00+00:00   \n",
       "\n",
       "                              monitoring_description  resolved_flag  \\\n",
       "0  A fix has been implemented and we are monitori...              1   \n",
       "1                                                NaN              1   \n",
       "2  A fix has been implemented and we are monitori...              1   \n",
       "3  A fix has been implemented and we are monitori...              1   \n",
       "4  A fix has been implemented and we are monitori...              1   \n",
       "\n",
       "         resolved_timestamp  \\\n",
       "0 2024-02-26 18:21:00+00:00   \n",
       "1 2024-02-16 20:35:00+00:00   \n",
       "2 2024-02-13 04:36:00+00:00   \n",
       "3 2024-02-13 01:26:00+00:00   \n",
       "4 2024-02-06 01:55:00+00:00   \n",
       "\n",
       "                                resolved_description  postmortem_flag  \\\n",
       "0                   This incident has been resolved.                0   \n",
       "1  We experienced an elevated level of errors on ...                0   \n",
       "2                   This incident has been resolved.                0   \n",
       "3                   This incident has been resolved.                0   \n",
       "4                   This incident has been resolved.                0   \n",
       "\n",
       "  postmortem_timestamp postmortem_description            start_timestamp  \\\n",
       "0                  NaT                    NaN  2024-02-26 17:09:00+00:00   \n",
       "1                  NaT                    NaN  2024-02-16 20:35:00+00:00   \n",
       "2                  NaT                    NaN  2024-02-13 03:59:00+00:00   \n",
       "3                  NaT                    NaN  2024-02-13 00:21:00+00:00   \n",
       "4                  NaT                    NaN  2024-02-05 23:07:00+00:00   \n",
       "\n",
       "             close_timestamp        time_span  over_one_day  \n",
       "0  2024-02-26 18:21:00+00:00  0 days 01:12:00         False  \n",
       "1  2024-02-16 20:35:00+00:00  0 days 00:00:00         False  \n",
       "2  2024-02-13 04:36:00+00:00  0 days 00:37:00         False  \n",
       "3  2024-02-13 01:26:00+00:00  0 days 01:05:00         False  \n",
       "4  2024-02-06 01:55:00+00:00  0 days 02:48:00         False  "
      ]
     },
     "execution_count": 2,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# load incident data\n",
    "path = f'data/clean/incident/2024-08-31/incident_stages.csv'\n",
    "df = pd.read_csv(path)\n",
    "\n",
    "status_lst = ['investigating', 'identified', 'monitoring', 'resolved', 'postmortem']\n",
    "for status in status_lst:\n",
    "    df[f'{status}_timestamp'] = pd.to_datetime(df[f'{status}_timestamp'])\n",
    "    \n",
    "df.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[2, 3, 0, 0, 0]\n"
     ]
    },
    {
     "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>incident_id</th>\n",
       "      <th>Incident_Title</th>\n",
       "      <th>incident_impact_level</th>\n",
       "      <th>Incident_color</th>\n",
       "      <th>provider</th>\n",
       "      <th>Playground</th>\n",
       "      <th>API</th>\n",
       "      <th>Labs</th>\n",
       "      <th>ChatGPT</th>\n",
       "      <th>api.anthropic.com</th>\n",
       "      <th>claude.ai</th>\n",
       "      <th>console.anthropic.com</th>\n",
       "      <th>Character.AI</th>\n",
       "      <th>investigating_flag</th>\n",
       "      <th>investigating_timestamp</th>\n",
       "      <th>investigating_description</th>\n",
       "      <th>identified_flag</th>\n",
       "      <th>identified_timestamp</th>\n",
       "      <th>identified_description</th>\n",
       "      <th>monitoring_flag</th>\n",
       "      <th>monitoring_timestamp</th>\n",
       "      <th>monitoring_description</th>\n",
       "      <th>resolved_flag</th>\n",
       "      <th>resolved_timestamp</th>\n",
       "      <th>resolved_description</th>\n",
       "      <th>postmortem_flag</th>\n",
       "      <th>postmortem_timestamp</th>\n",
       "      <th>postmortem_description</th>\n",
       "      <th>start_timestamp</th>\n",
       "      <th>close_timestamp</th>\n",
       "      <th>time_span</th>\n",
       "      <th>over_one_day</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>qv89mzdjvtrs</td>\n",
       "      <td>Elevated Errors</td>\n",
       "      <td>1</td>\n",
       "      <td>#f1c40f</td>\n",
       "      <td>anthropic</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>2024-02-26 17:09:00+00:00</td>\n",
       "      <td>We're experiencing an elevated level of API er...</td>\n",
       "      <td>1</td>\n",
       "      <td>2024-02-26 17:37:00+00:00</td>\n",
       "      <td>The issue has been identified and a fix is bei...</td>\n",
       "      <td>1</td>\n",
       "      <td>2024-02-26 17:41:00+00:00</td>\n",
       "      <td>A fix has been implemented and we are monitori...</td>\n",
       "      <td>1</td>\n",
       "      <td>2024-02-26 18:21:00+00:00</td>\n",
       "      <td>This incident has been resolved.</td>\n",
       "      <td>0</td>\n",
       "      <td>NaT</td>\n",
       "      <td>NaN</td>\n",
       "      <td>2024-02-26 17:09:00+00:00</td>\n",
       "      <td>2024-02-26 18:21:00+00:00</td>\n",
       "      <td>0 days 01:12:00</td>\n",
       "      <td>False</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>tmsczhzhjd63</td>\n",
       "      <td>Elevated error rate on claude.ai</td>\n",
       "      <td>0</td>\n",
       "      <td>#333333</td>\n",
       "      <td>anthropic</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>NaT</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0</td>\n",
       "      <td>NaT</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0</td>\n",
       "      <td>NaT</td>\n",
       "      <td>NaN</td>\n",
       "      <td>1</td>\n",
       "      <td>2024-02-16 20:35:00+00:00</td>\n",
       "      <td>We experienced an elevated level of errors on ...</td>\n",
       "      <td>0</td>\n",
       "      <td>NaT</td>\n",
       "      <td>NaN</td>\n",
       "      <td>2024-02-16 20:35:00+00:00</td>\n",
       "      <td>2024-02-16 20:35:00+00:00</td>\n",
       "      <td>0 days 00:00:00</td>\n",
       "      <td>False</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>t6v85cs8j6jb</td>\n",
       "      <td>Elevated error rate on Claude-2.0, Claude-2.1</td>\n",
       "      <td>0</td>\n",
       "      <td>#333333</td>\n",
       "      <td>anthropic</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>2024-02-13 03:59:00+00:00</td>\n",
       "      <td>We are currently investigating this issue.</td>\n",
       "      <td>1</td>\n",
       "      <td>2024-02-13 04:21:00+00:00</td>\n",
       "      <td>The issue has been identified and a fix is bei...</td>\n",
       "      <td>1</td>\n",
       "      <td>2024-02-13 04:30:00+00:00</td>\n",
       "      <td>A fix has been implemented and we are monitori...</td>\n",
       "      <td>1</td>\n",
       "      <td>2024-02-13 04:36:00+00:00</td>\n",
       "      <td>This incident has been resolved.</td>\n",
       "      <td>0</td>\n",
       "      <td>NaT</td>\n",
       "      <td>NaN</td>\n",
       "      <td>2024-02-13 03:59:00+00:00</td>\n",
       "      <td>2024-02-13 04:36:00+00:00</td>\n",
       "      <td>0 days 00:37:00</td>\n",
       "      <td>False</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>x5f4s8nvkyqq</td>\n",
       "      <td>Elevated error rate on Claude-2.0, Claude-2.1</td>\n",
       "      <td>0</td>\n",
       "      <td>#333333</td>\n",
       "      <td>anthropic</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>2024-02-13 00:21:00+00:00</td>\n",
       "      <td>We are currently investigating this issue.</td>\n",
       "      <td>1</td>\n",
       "      <td>2024-02-13 00:50:00+00:00</td>\n",
       "      <td>The issue has been identified and a fix is bei...</td>\n",
       "      <td>1</td>\n",
       "      <td>2024-02-13 00:59:00+00:00</td>\n",
       "      <td>A fix has been implemented and we are monitori...</td>\n",
       "      <td>1</td>\n",
       "      <td>2024-02-13 01:26:00+00:00</td>\n",
       "      <td>This incident has been resolved.</td>\n",
       "      <td>0</td>\n",
       "      <td>NaT</td>\n",
       "      <td>NaN</td>\n",
       "      <td>2024-02-13 00:21:00+00:00</td>\n",
       "      <td>2024-02-13 01:26:00+00:00</td>\n",
       "      <td>0 days 01:05:00</td>\n",
       "      <td>False</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>gr7nb0dwyydw</td>\n",
       "      <td>elevated error rates on claude-instant-1.2</td>\n",
       "      <td>0</td>\n",
       "      <td>#333333</td>\n",
       "      <td>anthropic</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>2024-02-05 23:07:00+00:00</td>\n",
       "      <td>We are currently investigating this issue.</td>\n",
       "      <td>0</td>\n",
       "      <td>NaT</td>\n",
       "      <td>NaN</td>\n",
       "      <td>1</td>\n",
       "      <td>2024-02-05 23:58:00+00:00</td>\n",
       "      <td>A fix has been implemented and we are monitori...</td>\n",
       "      <td>1</td>\n",
       "      <td>2024-02-06 01:55:00+00:00</td>\n",
       "      <td>This incident has been resolved.</td>\n",
       "      <td>0</td>\n",
       "      <td>NaT</td>\n",
       "      <td>NaN</td>\n",
       "      <td>2024-02-05 23:07:00+00:00</td>\n",
       "      <td>2024-02-06 01:55:00+00:00</td>\n",
       "      <td>0 days 02:48:00</td>\n",
       "      <td>False</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>532</th>\n",
       "      <td>qz0h6b7jw268</td>\n",
       "      <td>Outage on some Ada and Babbage models</td>\n",
       "      <td>2</td>\n",
       "      <td>#e86c09</td>\n",
       "      <td>openai</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>NaT</td>\n",
       "      <td>NaN</td>\n",
       "      <td>1</td>\n",
       "      <td>2022-07-07 06:07:00+00:00</td>\n",
       "      <td>We have identified an issue affecting some Ada...</td>\n",
       "      <td>1</td>\n",
       "      <td>2022-07-07 06:56:00+00:00</td>\n",
       "      <td>We’ve moved affected models to other regions, ...</td>\n",
       "      <td>1</td>\n",
       "      <td>2022-07-07 08:12:00+00:00</td>\n",
       "      <td>All models have recovered and are operational....</td>\n",
       "      <td>1</td>\n",
       "      <td>2022-07-12 01:03:00+00:00</td>\n",
       "      <td>Summary:\\nOn July 6, 2022 22:17 PDT (July 7 05...</td>\n",
       "      <td>2022-07-07 06:07:00+00:00</td>\n",
       "      <td>2022-07-07 08:12:00+00:00</td>\n",
       "      <td>0 days 02:05:00</td>\n",
       "      <td>False</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>533</th>\n",
       "      <td>7h9pjy5d1lz5</td>\n",
       "      <td>Cluster Outage</td>\n",
       "      <td>2</td>\n",
       "      <td>#e86c09</td>\n",
       "      <td>openai</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>2022-06-24 02:02:00+00:00</td>\n",
       "      <td>One of our clusters has had an outage affectin...</td>\n",
       "      <td>1</td>\n",
       "      <td>2022-06-24 04:14:00+00:00</td>\n",
       "      <td>One of our clusters has suffered a major commu...</td>\n",
       "      <td>1</td>\n",
       "      <td>2022-06-24 06:03:00+00:00</td>\n",
       "      <td>We believe we have found a stable arrangement ...</td>\n",
       "      <td>1</td>\n",
       "      <td>2022-06-24 15:25:00+00:00</td>\n",
       "      <td>All models are operational.  Thank you for you...</td>\n",
       "      <td>1</td>\n",
       "      <td>2022-06-28 03:20:00+00:00</td>\n",
       "      <td>Summary:\\nOn June 23, 2022 18:56 PDT (June 24 ...</td>\n",
       "      <td>2022-06-24 02:02:00+00:00</td>\n",
       "      <td>2022-06-24 15:25:00+00:00</td>\n",
       "      <td>0 days 13:23:00</td>\n",
       "      <td>False</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>534</th>\n",
       "      <td>jzk2gzhfc34g</td>\n",
       "      <td>Logged in websites unavailable</td>\n",
       "      <td>2</td>\n",
       "      <td>#e86c09</td>\n",
       "      <td>openai</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>2022-06-21 07:00:00+00:00</td>\n",
       "      <td>Investigating reports of Playground not being ...</td>\n",
       "      <td>1</td>\n",
       "      <td>2022-06-21 07:01:00+00:00</td>\n",
       "      <td>Playground &amp; DALL-E websites are currently una...</td>\n",
       "      <td>1</td>\n",
       "      <td>2022-06-21 07:01:00+00:00</td>\n",
       "      <td>Playground &amp; DALL-E websites are currently una...</td>\n",
       "      <td>1</td>\n",
       "      <td>2022-06-21 07:38:00+00:00</td>\n",
       "      <td>A fix has been implemented by our service prov...</td>\n",
       "      <td>0</td>\n",
       "      <td>NaT</td>\n",
       "      <td>NaN</td>\n",
       "      <td>2022-06-21 07:00:00+00:00</td>\n",
       "      <td>2022-06-21 07:38:00+00:00</td>\n",
       "      <td>0 days 00:38:00</td>\n",
       "      <td>False</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>535</th>\n",
       "      <td>2bxby42r6wv4</td>\n",
       "      <td>Base babbage model currently down</td>\n",
       "      <td>2</td>\n",
       "      <td>#e86c09</td>\n",
       "      <td>openai</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>2022-06-20 16:34:00+00:00</td>\n",
       "      <td>Base Babbage models are currently down. This a...</td>\n",
       "      <td>0</td>\n",
       "      <td>NaT</td>\n",
       "      <td>NaN</td>\n",
       "      <td>1</td>\n",
       "      <td>2022-06-20 16:52:00+00:00</td>\n",
       "      <td>The babbage models appear to have recovered, w...</td>\n",
       "      <td>1</td>\n",
       "      <td>2022-06-20 17:06:00+00:00</td>\n",
       "      <td>Serving of babbage models is now healthy. The ...</td>\n",
       "      <td>0</td>\n",
       "      <td>NaT</td>\n",
       "      <td>NaN</td>\n",
       "      <td>2022-06-20 16:34:00+00:00</td>\n",
       "      <td>2022-06-20 17:06:00+00:00</td>\n",
       "      <td>0 days 00:32:00</td>\n",
       "      <td>False</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>536</th>\n",
       "      <td>4cmnntq2t89y</td>\n",
       "      <td>Degradation in serving fine tuned models.</td>\n",
       "      <td>2</td>\n",
       "      <td>#e86c09</td>\n",
       "      <td>openai</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>2022-06-03 21:15:00+00:00</td>\n",
       "      <td>A recent deploy has degraded serving of fine t...</td>\n",
       "      <td>0</td>\n",
       "      <td>NaT</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0</td>\n",
       "      <td>NaT</td>\n",
       "      <td>NaN</td>\n",
       "      <td>1</td>\n",
       "      <td>2022-06-03 22:21:00+00:00</td>\n",
       "      <td>Service should be back to normal after all rol...</td>\n",
       "      <td>0</td>\n",
       "      <td>NaT</td>\n",
       "      <td>NaN</td>\n",
       "      <td>2022-06-03 21:15:00+00:00</td>\n",
       "      <td>2022-06-03 22:21:00+00:00</td>\n",
       "      <td>0 days 01:06:00</td>\n",
       "      <td>False</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>537 rows × 32 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "      incident_id                                 Incident_Title  \\\n",
       "0    qv89mzdjvtrs                                Elevated Errors   \n",
       "1    tmsczhzhjd63               Elevated error rate on claude.ai   \n",
       "2    t6v85cs8j6jb  Elevated error rate on Claude-2.0, Claude-2.1   \n",
       "3    x5f4s8nvkyqq  Elevated error rate on Claude-2.0, Claude-2.1   \n",
       "4    gr7nb0dwyydw     elevated error rates on claude-instant-1.2   \n",
       "..            ...                                            ...   \n",
       "532  qz0h6b7jw268          Outage on some Ada and Babbage models   \n",
       "533  7h9pjy5d1lz5                                 Cluster Outage   \n",
       "534  jzk2gzhfc34g                 Logged in websites unavailable   \n",
       "535  2bxby42r6wv4              Base babbage model currently down   \n",
       "536  4cmnntq2t89y      Degradation in serving fine tuned models.   \n",
       "\n",
       "     incident_impact_level Incident_color   provider  Playground  API  Labs  \\\n",
       "0                        1        #f1c40f  anthropic           0    0     0   \n",
       "1                        0        #333333  anthropic           0    0     0   \n",
       "2                        0        #333333  anthropic           0    0     0   \n",
       "3                        0        #333333  anthropic           0    0     0   \n",
       "4                        0        #333333  anthropic           0    0     0   \n",
       "..                     ...            ...        ...         ...  ...   ...   \n",
       "532                      2        #e86c09     openai           0    1     0   \n",
       "533                      2        #e86c09     openai           0    1     0   \n",
       "534                      2        #e86c09     openai           1    0     0   \n",
       "535                      2        #e86c09     openai           1    1     0   \n",
       "536                      2        #e86c09     openai           1    1     0   \n",
       "\n",
       "     ChatGPT  api.anthropic.com  claude.ai  console.anthropic.com  \\\n",
       "0          0                  1          1                      1   \n",
       "1          0                  0          0                      0   \n",
       "2          0                  1          1                      1   \n",
       "3          0                  1          1                      1   \n",
       "4          0                  1          1                      1   \n",
       "..       ...                ...        ...                    ...   \n",
       "532        0                  0          0                      0   \n",
       "533        0                  0          0                      0   \n",
       "534        0                  0          0                      0   \n",
       "535        0                  0          0                      0   \n",
       "536        0                  0          0                      0   \n",
       "\n",
       "     Character.AI  investigating_flag   investigating_timestamp  \\\n",
       "0               0                   1 2024-02-26 17:09:00+00:00   \n",
       "1               0                   0                       NaT   \n",
       "2               0                   1 2024-02-13 03:59:00+00:00   \n",
       "3               0                   1 2024-02-13 00:21:00+00:00   \n",
       "4               0                   1 2024-02-05 23:07:00+00:00   \n",
       "..            ...                 ...                       ...   \n",
       "532             0                   0                       NaT   \n",
       "533             0                   1 2022-06-24 02:02:00+00:00   \n",
       "534             0                   1 2022-06-21 07:00:00+00:00   \n",
       "535             0                   1 2022-06-20 16:34:00+00:00   \n",
       "536             0                   1 2022-06-03 21:15:00+00:00   \n",
       "\n",
       "                             investigating_description  identified_flag  \\\n",
       "0    We're experiencing an elevated level of API er...                1   \n",
       "1                                                  NaN                0   \n",
       "2           We are currently investigating this issue.                1   \n",
       "3           We are currently investigating this issue.                1   \n",
       "4           We are currently investigating this issue.                0   \n",
       "..                                                 ...              ...   \n",
       "532                                                NaN                1   \n",
       "533  One of our clusters has had an outage affectin...                1   \n",
       "534  Investigating reports of Playground not being ...                1   \n",
       "535  Base Babbage models are currently down. This a...                0   \n",
       "536  A recent deploy has degraded serving of fine t...                0   \n",
       "\n",
       "         identified_timestamp  \\\n",
       "0   2024-02-26 17:37:00+00:00   \n",
       "1                         NaT   \n",
       "2   2024-02-13 04:21:00+00:00   \n",
       "3   2024-02-13 00:50:00+00:00   \n",
       "4                         NaT   \n",
       "..                        ...   \n",
       "532 2022-07-07 06:07:00+00:00   \n",
       "533 2022-06-24 04:14:00+00:00   \n",
       "534 2022-06-21 07:01:00+00:00   \n",
       "535                       NaT   \n",
       "536                       NaT   \n",
       "\n",
       "                                identified_description  monitoring_flag  \\\n",
       "0    The issue has been identified and a fix is bei...                1   \n",
       "1                                                  NaN                0   \n",
       "2    The issue has been identified and a fix is bei...                1   \n",
       "3    The issue has been identified and a fix is bei...                1   \n",
       "4                                                  NaN                1   \n",
       "..                                                 ...              ...   \n",
       "532  We have identified an issue affecting some Ada...                1   \n",
       "533  One of our clusters has suffered a major commu...                1   \n",
       "534  Playground & DALL-E websites are currently una...                1   \n",
       "535                                                NaN                1   \n",
       "536                                                NaN                0   \n",
       "\n",
       "         monitoring_timestamp  \\\n",
       "0   2024-02-26 17:41:00+00:00   \n",
       "1                         NaT   \n",
       "2   2024-02-13 04:30:00+00:00   \n",
       "3   2024-02-13 00:59:00+00:00   \n",
       "4   2024-02-05 23:58:00+00:00   \n",
       "..                        ...   \n",
       "532 2022-07-07 06:56:00+00:00   \n",
       "533 2022-06-24 06:03:00+00:00   \n",
       "534 2022-06-21 07:01:00+00:00   \n",
       "535 2022-06-20 16:52:00+00:00   \n",
       "536                       NaT   \n",
       "\n",
       "                                monitoring_description  resolved_flag  \\\n",
       "0    A fix has been implemented and we are monitori...              1   \n",
       "1                                                  NaN              1   \n",
       "2    A fix has been implemented and we are monitori...              1   \n",
       "3    A fix has been implemented and we are monitori...              1   \n",
       "4    A fix has been implemented and we are monitori...              1   \n",
       "..                                                 ...            ...   \n",
       "532  We’ve moved affected models to other regions, ...              1   \n",
       "533  We believe we have found a stable arrangement ...              1   \n",
       "534  Playground & DALL-E websites are currently una...              1   \n",
       "535  The babbage models appear to have recovered, w...              1   \n",
       "536                                                NaN              1   \n",
       "\n",
       "           resolved_timestamp  \\\n",
       "0   2024-02-26 18:21:00+00:00   \n",
       "1   2024-02-16 20:35:00+00:00   \n",
       "2   2024-02-13 04:36:00+00:00   \n",
       "3   2024-02-13 01:26:00+00:00   \n",
       "4   2024-02-06 01:55:00+00:00   \n",
       "..                        ...   \n",
       "532 2022-07-07 08:12:00+00:00   \n",
       "533 2022-06-24 15:25:00+00:00   \n",
       "534 2022-06-21 07:38:00+00:00   \n",
       "535 2022-06-20 17:06:00+00:00   \n",
       "536 2022-06-03 22:21:00+00:00   \n",
       "\n",
       "                                  resolved_description  postmortem_flag  \\\n",
       "0                     This incident has been resolved.                0   \n",
       "1    We experienced an elevated level of errors on ...                0   \n",
       "2                     This incident has been resolved.                0   \n",
       "3                     This incident has been resolved.                0   \n",
       "4                     This incident has been resolved.                0   \n",
       "..                                                 ...              ...   \n",
       "532  All models have recovered and are operational....                1   \n",
       "533  All models are operational.  Thank you for you...                1   \n",
       "534  A fix has been implemented by our service prov...                0   \n",
       "535  Serving of babbage models is now healthy. The ...                0   \n",
       "536  Service should be back to normal after all rol...                0   \n",
       "\n",
       "         postmortem_timestamp  \\\n",
       "0                         NaT   \n",
       "1                         NaT   \n",
       "2                         NaT   \n",
       "3                         NaT   \n",
       "4                         NaT   \n",
       "..                        ...   \n",
       "532 2022-07-12 01:03:00+00:00   \n",
       "533 2022-06-28 03:20:00+00:00   \n",
       "534                       NaT   \n",
       "535                       NaT   \n",
       "536                       NaT   \n",
       "\n",
       "                                postmortem_description  \\\n",
       "0                                                  NaN   \n",
       "1                                                  NaN   \n",
       "2                                                  NaN   \n",
       "3                                                  NaN   \n",
       "4                                                  NaN   \n",
       "..                                                 ...   \n",
       "532  Summary:\\nOn July 6, 2022 22:17 PDT (July 7 05...   \n",
       "533  Summary:\\nOn June 23, 2022 18:56 PDT (June 24 ...   \n",
       "534                                                NaN   \n",
       "535                                                NaN   \n",
       "536                                                NaN   \n",
       "\n",
       "               start_timestamp            close_timestamp        time_span  \\\n",
       "0    2024-02-26 17:09:00+00:00  2024-02-26 18:21:00+00:00  0 days 01:12:00   \n",
       "1    2024-02-16 20:35:00+00:00  2024-02-16 20:35:00+00:00  0 days 00:00:00   \n",
       "2    2024-02-13 03:59:00+00:00  2024-02-13 04:36:00+00:00  0 days 00:37:00   \n",
       "3    2024-02-13 00:21:00+00:00  2024-02-13 01:26:00+00:00  0 days 01:05:00   \n",
       "4    2024-02-05 23:07:00+00:00  2024-02-06 01:55:00+00:00  0 days 02:48:00   \n",
       "..                         ...                        ...              ...   \n",
       "532  2022-07-07 06:07:00+00:00  2022-07-07 08:12:00+00:00  0 days 02:05:00   \n",
       "533  2022-06-24 02:02:00+00:00  2022-06-24 15:25:00+00:00  0 days 13:23:00   \n",
       "534  2022-06-21 07:00:00+00:00  2022-06-21 07:38:00+00:00  0 days 00:38:00   \n",
       "535  2022-06-20 16:34:00+00:00  2022-06-20 17:06:00+00:00  0 days 00:32:00   \n",
       "536  2022-06-03 21:15:00+00:00  2022-06-03 22:21:00+00:00  0 days 01:06:00   \n",
       "\n",
       "     over_one_day  \n",
       "0           False  \n",
       "1           False  \n",
       "2           False  \n",
       "3           False  \n",
       "4           False  \n",
       "..            ...  \n",
       "532         False  \n",
       "533         False  \n",
       "534         False  \n",
       "535         False  \n",
       "536         False  \n",
       "\n",
       "[537 rows x 32 columns]"
      ]
     },
     "execution_count": 3,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# corner case filter\n",
    "# check order of timestamp\n",
    "count_lst = 5 * [0]\n",
    "corner_cases = pd.DataFrame()\n",
    "\n",
    "for index, record in df.iterrows():\n",
    "    if record['investigating_timestamp'] > record['identified_timestamp']:\n",
    "        count_lst[0] += 1\n",
    "        corner_cases = pd.concat([corner_cases, record], axis=0)\n",
    "    if record['identified_timestamp'] > record['monitoring_timestamp']:\n",
    "        count_lst[1] += 1\n",
    "        corner_cases = pd.concat([corner_cases, record], axis=0)\n",
    "    if record['monitoring_timestamp'] > record['resolved_timestamp']:\n",
    "        count_lst[2] += 1\n",
    "        corner_cases = pd.concat([corner_cases, record], axis=0)\n",
    "    if record['resolved_timestamp'] > record['postmortem_timestamp']:\n",
    "        count_lst[3] += 1\n",
    "        corner_cases = pd.concat([corner_cases, record], axis=0)\n",
    "print(count_lst)\n",
    "\n",
    "corner_cases = corner_cases.T\n",
    "# display(corner_cases)\n",
    "\n",
    "# filter out df corner cases\n",
    "df = df.drop(corner_cases.index)\n",
    "df = df.reset_index(drop=True)\n",
    "df"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "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>incident_id</th>\n",
       "      <th>Incident_Title</th>\n",
       "      <th>incident_impact_level</th>\n",
       "      <th>Incident_color</th>\n",
       "      <th>provider</th>\n",
       "      <th>Playground</th>\n",
       "      <th>API</th>\n",
       "      <th>Labs</th>\n",
       "      <th>ChatGPT</th>\n",
       "      <th>api.anthropic.com</th>\n",
       "      <th>claude.ai</th>\n",
       "      <th>console.anthropic.com</th>\n",
       "      <th>Character.AI</th>\n",
       "      <th>investigating_flag</th>\n",
       "      <th>investigating_timestamp</th>\n",
       "      <th>investigating_description</th>\n",
       "      <th>identified_flag</th>\n",
       "      <th>identified_timestamp</th>\n",
       "      <th>identified_description</th>\n",
       "      <th>monitoring_flag</th>\n",
       "      <th>monitoring_timestamp</th>\n",
       "      <th>monitoring_description</th>\n",
       "      <th>resolved_flag</th>\n",
       "      <th>resolved_timestamp</th>\n",
       "      <th>resolved_description</th>\n",
       "      <th>postmortem_flag</th>\n",
       "      <th>postmortem_timestamp</th>\n",
       "      <th>postmortem_description</th>\n",
       "      <th>start_timestamp</th>\n",
       "      <th>close_timestamp</th>\n",
       "      <th>time_span</th>\n",
       "      <th>over_one_day</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>286</th>\n",
       "      <td>w20mcckg1748</td>\n",
       "      <td>Elevated errors in ChatGPT</td>\n",
       "      <td>2</td>\n",
       "      <td>#e86c09</td>\n",
       "      <td>openai</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>2024-04-10 17:56:00+00:00</td>\n",
       "      <td>We are currently investigating elevated errors...</td>\n",
       "      <td>1</td>\n",
       "      <td>2024-04-10 20:39:00+00:00</td>\n",
       "      <td>We have identified the source of the elevated ...</td>\n",
       "      <td>1</td>\n",
       "      <td>2024-04-10 20:49:00+00:00</td>\n",
       "      <td>Errors are continuing to resolve as a result o...</td>\n",
       "      <td>1</td>\n",
       "      <td>2024-04-10 20:56:00+00:00</td>\n",
       "      <td>At this time our deployed fix has mitigated th...</td>\n",
       "      <td>1</td>\n",
       "      <td>2024-04-18 00:01:00+00:00</td>\n",
       "      <td>On April 10th between 11:00am - 1:43pm PST, a ...</td>\n",
       "      <td>2024-04-10 17:56:00+00:00</td>\n",
       "      <td>2024-04-10 20:56:00+00:00</td>\n",
       "      <td>0 days 03:00:00</td>\n",
       "      <td>False</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "      incident_id              Incident_Title  incident_impact_level  \\\n",
       "286  w20mcckg1748  Elevated errors in ChatGPT                      2   \n",
       "\n",
       "    Incident_color provider  Playground  API  Labs  ChatGPT  \\\n",
       "286        #e86c09   openai           0    0     0        1   \n",
       "\n",
       "     api.anthropic.com  claude.ai  console.anthropic.com  Character.AI  \\\n",
       "286                  0          0                      0             0   \n",
       "\n",
       "     investigating_flag   investigating_timestamp  \\\n",
       "286                   1 2024-04-10 17:56:00+00:00   \n",
       "\n",
       "                             investigating_description  identified_flag  \\\n",
       "286  We are currently investigating elevated errors...                1   \n",
       "\n",
       "         identified_timestamp  \\\n",
       "286 2024-04-10 20:39:00+00:00   \n",
       "\n",
       "                                identified_description  monitoring_flag  \\\n",
       "286  We have identified the source of the elevated ...                1   \n",
       "\n",
       "         monitoring_timestamp  \\\n",
       "286 2024-04-10 20:49:00+00:00   \n",
       "\n",
       "                                monitoring_description  resolved_flag  \\\n",
       "286  Errors are continuing to resolve as a result o...              1   \n",
       "\n",
       "           resolved_timestamp  \\\n",
       "286 2024-04-10 20:56:00+00:00   \n",
       "\n",
       "                                  resolved_description  postmortem_flag  \\\n",
       "286  At this time our deployed fix has mitigated th...                1   \n",
       "\n",
       "         postmortem_timestamp  \\\n",
       "286 2024-04-18 00:01:00+00:00   \n",
       "\n",
       "                                postmortem_description  \\\n",
       "286  On April 10th between 11:00am - 1:43pm PST, a ...   \n",
       "\n",
       "               start_timestamp            close_timestamp        time_span  \\\n",
       "286  2024-04-10 17:56:00+00:00  2024-04-10 20:56:00+00:00  0 days 03:00:00   \n",
       "\n",
       "     over_one_day  \n",
       "286         False  "
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "The timestamp of each status is:                                      x\n",
      "investigating 2024-04-10 17:56:00+00:00\n",
      "identified    2024-04-10 20:39:00+00:00\n",
      "monitoring    2024-04-10 20:49:00+00:00\n",
      "resolved      2024-04-10 20:56:00+00:00\n"
     ]
    },
    {
     "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>x</th>\n",
       "      <th>y</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>investigating</th>\n",
       "      <td>2024-04-10 17:56:00+00:00</td>\n",
       "      <td>303.066667</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>identified</th>\n",
       "      <td>2024-04-10 20:39:00+00:00</td>\n",
       "      <td>1260.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>monitoring</th>\n",
       "      <td>2024-04-10 20:49:00+00:00</td>\n",
       "      <td>510.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>resolved</th>\n",
       "      <td>2024-04-10 20:56:00+00:00</td>\n",
       "      <td>245.266667</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                                      x            y\n",
       "investigating 2024-04-10 17:56:00+00:00   303.066667\n",
       "identified    2024-04-10 20:39:00+00:00  1260.000000\n",
       "monitoring    2024-04-10 20:49:00+00:00   510.000000\n",
       "resolved      2024-04-10 20:56:00+00:00   245.266667"
      ]
     },
     "execution_count": 4,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# filter all flag == 1\n",
    "df_incident = df[(df['investigating_flag'] == 1) & (df['identified_flag'] == 1) & (df['monitoring_flag'] == 1) & (df['resolved_flag'] == 1) & (df['postmortem_flag'] == 1)]\n",
    "# display(df_incident)\n",
    "\n",
    "# select an incident example\n",
    "df_example = df[df['incident_id'] == 'w20mcckg1748']\n",
    "display(df_example)\n",
    "\n",
    "# select timestamp columns\n",
    "timestamp_columns = [col for col in df.columns if 'timestamp' in col]\n",
    "df_example_tmp = df_example[timestamp_columns]\n",
    "for col in timestamp_columns:\n",
    "    df_example_tmp[col] = pd.to_datetime(df_example_tmp[col])\n",
    "    \n",
    "df_example_tmp = df_example_tmp.T[:-3]\n",
    "df_example_tmp.index = ['investigating', 'identified', 'monitoring', 'resolved']\n",
    "df_example_tmp.columns = ['x']\n",
    "\n",
    "# add estimate_value, for plotting\n",
    "# Here we estimate the number of reports of each status manually using MA(Moving Average) method. \n",
    "# Because user reports was collected every 15 minutes interval, but the incident status was updated every 1 minutes.\n",
    "# As the result, the timestamp of each status is:\n",
    "print(f'The timestamp of each status is:{df_example_tmp}') \n",
    "\n",
    "estimate_value = [294+((328-294)/15)*4, \n",
    "                1410-((1410-285)/15)*2, \n",
    "                1410-((1410-285)/15)*12, \n",
    "                285-((285-136)/15)*4,\n",
    "                ]\n",
    "df_example_tmp['y'] = estimate_value\n",
    "df_example_tmp\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "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>Incident ID</th>\n",
       "      <th>S_1</th>\n",
       "      <th>S_2</th>\n",
       "      <th>S_3</th>\n",
       "      <th>S_4</th>\n",
       "      <th>S_5</th>\n",
       "      <th>P_I</th>\n",
       "      <th>P_R</th>\n",
       "      <th>P_C</th>\n",
       "      <th>P_L</th>\n",
       "      <th>Time To Resolve</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>286</th>\n",
       "      <td>w20mcckg1748</td>\n",
       "      <td>2024-04-10 17:56:00+00:00</td>\n",
       "      <td>2024-04-10 20:39:00+00:00</td>\n",
       "      <td>2024-04-10 20:49:00+00:00</td>\n",
       "      <td>2024-04-10 20:56:00+00:00</td>\n",
       "      <td>2024-04-18 00:01:00+00:00</td>\n",
       "      <td>2.72</td>\n",
       "      <td>0.17</td>\n",
       "      <td>0.12</td>\n",
       "      <td>171.08</td>\n",
       "      <td>3.0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "      Incident ID                       S_1                       S_2  \\\n",
       "286  w20mcckg1748 2024-04-10 17:56:00+00:00 2024-04-10 20:39:00+00:00   \n",
       "\n",
       "                          S_3                       S_4  \\\n",
       "286 2024-04-10 20:49:00+00:00 2024-04-10 20:56:00+00:00   \n",
       "\n",
       "                          S_5   P_I   P_R   P_C     P_L  Time To Resolve  \n",
       "286 2024-04-18 00:01:00+00:00  2.72  0.17  0.12  171.08              3.0  "
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\\begin{tabular}{llllllrrrrr}\n",
      "\\toprule\n",
      "Incident ID & S_1 & S_2 & S_3 & S_4 & S_5 & P_I & P_R & P_C & P_L & Time To Resolve \\\\\n",
      "\\midrule\n",
      "w20mcckg1748 & 2024-04-10 17:56:00+00:00 & 2024-04-10 20:39:00+00:00 & 2024-04-10 20:49:00+00:00 & 2024-04-10 20:56:00+00:00 & 2024-04-18 00:01:00+00:00 & 2.720000 & 0.170000 & 0.120000 & 171.080000 & 3.000000 \\\\\n",
      "\\bottomrule\n",
      "\\end{tabular}\n",
      "\n"
     ]
    }
   ],
   "source": [
    "# table. parameter value for the selected incident\n",
    "df_example_param = df_example[['incident_id', 'investigating_timestamp', 'identified_timestamp', 'monitoring_timestamp', 'resolved_timestamp', 'postmortem_timestamp']]\n",
    "# rename columns\n",
    "df_example_param.columns = ['Incident ID', 'S_1', 'S_2', 'S_3', 'S_4', 'S_5']\n",
    "df_example_param['P_I'] = df_example_param['S_2'] - df_example_param['S_1']\n",
    "df_example_param['P_R'] = df_example_param['S_3'] - df_example_param['S_2']\n",
    "df_example_param['P_C'] = df_example_param['S_4'] - df_example_param['S_3']\n",
    "df_example_param['P_L'] = df_example_param['S_5'] - df_example_param['S_4']\n",
    "# transform to hours\n",
    "df_example_param['P_I'] = df_example_param['P_I'].dt.total_seconds() / 3600\n",
    "df_example_param['P_R'] = df_example_param['P_R'].dt.total_seconds() / 3600\n",
    "df_example_param['P_C'] = df_example_param['P_C'].dt.total_seconds() / 3600\n",
    "df_example_param['P_L'] = df_example_param['P_L'].dt.total_seconds() / 3600\n",
    "\n",
    "df_example_param['Time To Resolve'] = df_example_param['S_4'] - df_example_param['S_1']\n",
    "df_example_param['Time To Resolve'] = df_example_param['Time To Resolve'].dt.total_seconds() / 3600\n",
    "\n",
    "# 2 digits\n",
    "df_example_param = df_example_param.round(2)\n",
    "\n",
    "display(df_example_param)\n",
    "\n",
    "# latex\n",
    "print(df_example_param.to_latex(index=False))\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "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>x</th>\n",
       "      <th>y</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>2024-04-10 17:07:20+00:00</td>\n",
       "      <td>74</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>2024-04-10 17:22:20+00:00</td>\n",
       "      <td>87</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>2024-04-10 17:37:20+00:00</td>\n",
       "      <td>303</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>2024-04-10 17:52:20+00:00</td>\n",
       "      <td>294</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>2024-04-10 18:07:20+00:00</td>\n",
       "      <td>328</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>2024-04-10 18:22:20+00:00</td>\n",
       "      <td>1213</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>2024-04-10 18:37:20+00:00</td>\n",
       "      <td>4201</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>2024-04-10 18:52:20+00:00</td>\n",
       "      <td>2984</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8</th>\n",
       "      <td>2024-04-10 19:07:20+00:00</td>\n",
       "      <td>3432</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9</th>\n",
       "      <td>2024-04-10 19:22:20+00:00</td>\n",
       "      <td>4428</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>10</th>\n",
       "      <td>2024-04-10 19:37:20+00:00</td>\n",
       "      <td>3021</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>11</th>\n",
       "      <td>2024-04-10 19:52:20+00:00</td>\n",
       "      <td>2910</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>12</th>\n",
       "      <td>2024-04-10 20:07:20+00:00</td>\n",
       "      <td>2691</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>13</th>\n",
       "      <td>2024-04-10 20:22:20+00:00</td>\n",
       "      <td>2308</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>14</th>\n",
       "      <td>2024-04-10 20:37:20+00:00</td>\n",
       "      <td>1410</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>15</th>\n",
       "      <td>2024-04-10 20:52:20+00:00</td>\n",
       "      <td>285</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>16</th>\n",
       "      <td>2024-04-10 21:07:20+00:00</td>\n",
       "      <td>136</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>17</th>\n",
       "      <td>2024-04-10 21:22:20+00:00</td>\n",
       "      <td>130</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>18</th>\n",
       "      <td>2024-04-10 21:37:20+00:00</td>\n",
       "      <td>94</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>19</th>\n",
       "      <td>2024-04-10 21:52:20+00:00</td>\n",
       "      <td>84</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>20</th>\n",
       "      <td>2024-04-10 22:07:20+00:00</td>\n",
       "      <td>78</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>21</th>\n",
       "      <td>2024-04-10 22:22:20+00:00</td>\n",
       "      <td>59</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>22</th>\n",
       "      <td>2024-04-10 22:37:20+00:00</td>\n",
       "      <td>60</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>23</th>\n",
       "      <td>2024-04-10 22:52:20+00:00</td>\n",
       "      <td>50</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>24</th>\n",
       "      <td>2024-04-10 23:07:20+00:00</td>\n",
       "      <td>68</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>25</th>\n",
       "      <td>2024-04-10 23:22:20+00:00</td>\n",
       "      <td>79</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>26</th>\n",
       "      <td>2024-04-10 23:37:20+00:00</td>\n",
       "      <td>75</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                           x     y\n",
       "0  2024-04-10 17:07:20+00:00    74\n",
       "1  2024-04-10 17:22:20+00:00    87\n",
       "2  2024-04-10 17:37:20+00:00   303\n",
       "3  2024-04-10 17:52:20+00:00   294\n",
       "4  2024-04-10 18:07:20+00:00   328\n",
       "5  2024-04-10 18:22:20+00:00  1213\n",
       "6  2024-04-10 18:37:20+00:00  4201\n",
       "7  2024-04-10 18:52:20+00:00  2984\n",
       "8  2024-04-10 19:07:20+00:00  3432\n",
       "9  2024-04-10 19:22:20+00:00  4428\n",
       "10 2024-04-10 19:37:20+00:00  3021\n",
       "11 2024-04-10 19:52:20+00:00  2910\n",
       "12 2024-04-10 20:07:20+00:00  2691\n",
       "13 2024-04-10 20:22:20+00:00  2308\n",
       "14 2024-04-10 20:37:20+00:00  1410\n",
       "15 2024-04-10 20:52:20+00:00   285\n",
       "16 2024-04-10 21:07:20+00:00   136\n",
       "17 2024-04-10 21:22:20+00:00   130\n",
       "18 2024-04-10 21:37:20+00:00    94\n",
       "19 2024-04-10 21:52:20+00:00    84\n",
       "20 2024-04-10 22:07:20+00:00    78\n",
       "21 2024-04-10 22:22:20+00:00    59\n",
       "22 2024-04-10 22:37:20+00:00    60\n",
       "23 2024-04-10 22:52:20+00:00    50\n",
       "24 2024-04-10 23:07:20+00:00    68\n",
       "25 2024-04-10 23:22:20+00:00    79\n",
       "26 2024-04-10 23:37:20+00:00    75"
      ]
     },
     "execution_count": 6,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# user reports data at April 10, 2024, source from: https://web.archive.org/web/20240410235249/https://downdetector.com/status/openai/\n",
    "data = [\n",
    "                { 'x': '2024-04-09T23:52:20+00:00', 'y': 18 },\n",
    "              \n",
    "                { 'x': '2024-04-10T00:07:20+00:00', 'y': 12 },\n",
    "              \n",
    "                { 'x': '2024-04-10T00:22:20+00:00', 'y': 18 },\n",
    "              \n",
    "                { 'x': '2024-04-10T00:37:20+00:00', 'y': 13 },\n",
    "              \n",
    "                { 'x': '2024-04-10T00:52:20+00:00', 'y': 12 },\n",
    "              \n",
    "                { 'x': '2024-04-10T01:07:20+00:00', 'y': 15 },\n",
    "              \n",
    "                { 'x': '2024-04-10T01:22:20+00:00', 'y': 13 },\n",
    "              \n",
    "                { 'x': '2024-04-10T01:37:20+00:00', 'y': 11 },\n",
    "              \n",
    "                { 'x': '2024-04-10T01:52:20+00:00', 'y': 9 },\n",
    "              \n",
    "                { 'x': '2024-04-10T02:07:20+00:00', 'y': 10 },\n",
    "              \n",
    "                { 'x': '2024-04-10T02:22:20+00:00', 'y': 10 },\n",
    "              \n",
    "                { 'x': '2024-04-10T02:37:20+00:00', 'y': 19 },\n",
    "              \n",
    "                { 'x': '2024-04-10T02:52:20+00:00', 'y': 16 },\n",
    "              \n",
    "                { 'x': '2024-04-10T03:07:20+00:00', 'y': 14 },\n",
    "              \n",
    "                { 'x': '2024-04-10T03:22:20+00:00', 'y': 8 },\n",
    "              \n",
    "                { 'x': '2024-04-10T03:37:20+00:00', 'y': 12 },\n",
    "              \n",
    "                { 'x': '2024-04-10T03:52:20+00:00', 'y': 7 },\n",
    "              \n",
    "                { 'x': '2024-04-10T04:07:20+00:00', 'y': 7 },\n",
    "              \n",
    "                { 'x': '2024-04-10T04:22:20+00:00', 'y': 2 },\n",
    "              \n",
    "                { 'x': '2024-04-10T04:37:20+00:00', 'y': 4 },\n",
    "              \n",
    "                { 'x': '2024-04-10T04:52:20+00:00', 'y': 4 },\n",
    "              \n",
    "                { 'x': '2024-04-10T05:07:20+00:00', 'y': 4 },\n",
    "              \n",
    "                { 'x': '2024-04-10T05:22:20+00:00', 'y': 2 },\n",
    "              \n",
    "                { 'x': '2024-04-10T05:37:20+00:00', 'y': 8 },\n",
    "              \n",
    "                { 'x': '2024-04-10T05:52:20+00:00', 'y': 3 },\n",
    "              \n",
    "                { 'x': '2024-04-10T06:07:20+00:00', 'y': 2 },\n",
    "              \n",
    "                { 'x': '2024-04-10T06:22:20+00:00', 'y': 2 },\n",
    "              \n",
    "                { 'x': '2024-04-10T06:37:20+00:00', 'y': 0 },\n",
    "              \n",
    "                { 'x': '2024-04-10T06:52:20+00:00', 'y': 1 },\n",
    "              \n",
    "                { 'x': '2024-04-10T07:07:20+00:00', 'y': 3 },\n",
    "              \n",
    "                { 'x': '2024-04-10T07:22:20+00:00', 'y': 2 },\n",
    "              \n",
    "                { 'x': '2024-04-10T07:37:20+00:00', 'y': 3 },\n",
    "              \n",
    "                { 'x': '2024-04-10T07:52:20+00:00', 'y': 2 },\n",
    "              \n",
    "                { 'x': '2024-04-10T08:07:20+00:00', 'y': 0 },\n",
    "              \n",
    "                { 'x': '2024-04-10T08:22:20+00:00', 'y': 1 },\n",
    "              \n",
    "                { 'x': '2024-04-10T08:37:20+00:00', 'y': 4 },\n",
    "              \n",
    "                { 'x': '2024-04-10T08:52:20+00:00', 'y': 2 },\n",
    "              \n",
    "                { 'x': '2024-04-10T09:07:20+00:00', 'y': 3 },\n",
    "              \n",
    "                { 'x': '2024-04-10T09:22:20+00:00', 'y': 1 },\n",
    "              \n",
    "                { 'x': '2024-04-10T09:37:20+00:00', 'y': 0 },\n",
    "              \n",
    "                { 'x': '2024-04-10T09:52:20+00:00', 'y': 1 },\n",
    "              \n",
    "                { 'x': '2024-04-10T10:07:20+00:00', 'y': 0 },\n",
    "              \n",
    "                { 'x': '2024-04-10T10:22:20+00:00', 'y': 1 },\n",
    "              \n",
    "                { 'x': '2024-04-10T10:37:20+00:00', 'y': 0 },\n",
    "              \n",
    "                { 'x': '2024-04-10T10:52:20+00:00', 'y': 1 },\n",
    "              \n",
    "                { 'x': '2024-04-10T11:07:20+00:00', 'y': 1 },\n",
    "              \n",
    "                { 'x': '2024-04-10T11:22:20+00:00', 'y': 0 },\n",
    "              \n",
    "                { 'x': '2024-04-10T11:37:20+00:00', 'y': 1 },\n",
    "              \n",
    "                { 'x': '2024-04-10T11:52:20+00:00', 'y': 4 },\n",
    "              \n",
    "                { 'x': '2024-04-10T12:07:20+00:00', 'y': 2 },\n",
    "              \n",
    "                { 'x': '2024-04-10T12:22:20+00:00', 'y': 3 },\n",
    "              \n",
    "                { 'x': '2024-04-10T12:37:20+00:00', 'y': 3 },\n",
    "              \n",
    "                { 'x': '2024-04-10T12:52:20+00:00', 'y': 11 },\n",
    "              \n",
    "                { 'x': '2024-04-10T13:07:20+00:00', 'y': 5 },\n",
    "              \n",
    "                { 'x': '2024-04-10T13:22:20+00:00', 'y': 6 },\n",
    "              \n",
    "                { 'x': '2024-04-10T13:37:20+00:00', 'y': 5 },\n",
    "              \n",
    "                { 'x': '2024-04-10T13:52:20+00:00', 'y': 8 },\n",
    "              \n",
    "                { 'x': '2024-04-10T14:07:20+00:00', 'y': 4 },\n",
    "              \n",
    "                { 'x': '2024-04-10T14:22:20+00:00', 'y': 17 },\n",
    "              \n",
    "                { 'x': '2024-04-10T14:37:20+00:00', 'y': 12 },\n",
    "              \n",
    "                { 'x': '2024-04-10T14:52:20+00:00', 'y': 3 },\n",
    "              \n",
    "                { 'x': '2024-04-10T15:07:20+00:00', 'y': 13 },\n",
    "              \n",
    "                { 'x': '2024-04-10T15:22:20+00:00', 'y': 4 },\n",
    "              \n",
    "                { 'x': '2024-04-10T15:37:20+00:00', 'y': 11 },\n",
    "              \n",
    "                { 'x': '2024-04-10T15:52:20+00:00', 'y': 11 },\n",
    "              \n",
    "                { 'x': '2024-04-10T16:07:20+00:00', 'y': 7 },\n",
    "              \n",
    "                { 'x': '2024-04-10T16:22:20+00:00', 'y': 14 },\n",
    "              \n",
    "                { 'x': '2024-04-10T16:37:20+00:00', 'y': 12 },\n",
    "              \n",
    "                { 'x': '2024-04-10T16:52:20+00:00', 'y': 9 },\n",
    "              \n",
    "                { 'x': '2024-04-10T17:07:20+00:00', 'y': 74 },\n",
    "              \n",
    "                { 'x': '2024-04-10T17:22:20+00:00', 'y': 87 },\n",
    "              \n",
    "                { 'x': '2024-04-10T17:37:20+00:00', 'y': 303 },\n",
    "              \n",
    "                { 'x': '2024-04-10T17:52:20+00:00', 'y': 294 },\n",
    "              \n",
    "                { 'x': '2024-04-10T18:07:20+00:00', 'y': 328 },\n",
    "              \n",
    "                { 'x': '2024-04-10T18:22:20+00:00', 'y': 1213 },\n",
    "              \n",
    "                { 'x': '2024-04-10T18:37:20+00:00', 'y': 4201 },\n",
    "              \n",
    "                { 'x': '2024-04-10T18:52:20+00:00', 'y': 2984 },\n",
    "              \n",
    "                { 'x': '2024-04-10T19:07:20+00:00', 'y': 3432 },\n",
    "              \n",
    "                { 'x': '2024-04-10T19:22:20+00:00', 'y': 4428 },\n",
    "              \n",
    "                { 'x': '2024-04-10T19:37:20+00:00', 'y': 3021 },\n",
    "              \n",
    "                { 'x': '2024-04-10T19:52:20+00:00', 'y': 2910 },\n",
    "              \n",
    "                { 'x': '2024-04-10T20:07:20+00:00', 'y': 2691 },\n",
    "              \n",
    "                { 'x': '2024-04-10T20:22:20+00:00', 'y': 2308 },\n",
    "              \n",
    "                { 'x': '2024-04-10T20:37:20+00:00', 'y': 1410 },\n",
    "              \n",
    "                { 'x': '2024-04-10T20:52:20+00:00', 'y': 285 },\n",
    "              \n",
    "                { 'x': '2024-04-10T21:07:20+00:00', 'y': 136 },\n",
    "              \n",
    "                { 'x': '2024-04-10T21:22:20+00:00', 'y': 130 },\n",
    "              \n",
    "                { 'x': '2024-04-10T21:37:20+00:00', 'y': 94 },\n",
    "              \n",
    "                { 'x': '2024-04-10T21:52:20+00:00', 'y': 84 },\n",
    "              \n",
    "                { 'x': '2024-04-10T22:07:20+00:00', 'y': 78 },\n",
    "              \n",
    "                { 'x': '2024-04-10T22:22:20+00:00', 'y': 59 },\n",
    "              \n",
    "                { 'x': '2024-04-10T22:37:20+00:00', 'y': 60 },\n",
    "              \n",
    "                { 'x': '2024-04-10T22:52:20+00:00', 'y': 50 },\n",
    "              \n",
    "                { 'x': '2024-04-10T23:07:20+00:00', 'y': 68 },\n",
    "              \n",
    "                { 'x': '2024-04-10T23:22:20+00:00', 'y': 79 },\n",
    "              \n",
    "                { 'x': '2024-04-10T23:37:20+00:00', 'y': 75 },\n",
    "              \n",
    "]\n",
    "\n",
    "df_report = pd.DataFrame(data)\n",
    "# select timestamp after 2024-04-10T16:00:00+00:00\n",
    "df_report['x'] = pd.to_datetime(df_report['x'], utc=True)\n",
    "df_report = df_report[df_report['x'] >= '2024-04-10T17:00:00+00:00']\n",
    "df_report = df_report.reset_index(drop=True)\n",
    "df_report"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "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>x</th>\n",
       "      <th>y</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>real_start_point</th>\n",
       "      <td>2024-04-10 17:22:20+00:00</td>\n",
       "      <td>87.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>investigating</th>\n",
       "      <td>2024-04-10 17:56:00+00:00</td>\n",
       "      <td>303.066667</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>identified</th>\n",
       "      <td>2024-04-10 20:39:00+00:00</td>\n",
       "      <td>1260.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>monitoring</th>\n",
       "      <td>2024-04-10 20:49:00+00:00</td>\n",
       "      <td>510.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>resolved</th>\n",
       "      <td>2024-04-10 20:56:00+00:00</td>\n",
       "      <td>245.266667</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>postmortem</th>\n",
       "      <td>2024-04-10 22:22:20+00:00</td>\n",
       "      <td>59.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>next_investigating</th>\n",
       "      <td>2024-04-10 22:37:20+00:00</td>\n",
       "      <td>300.000000</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                                           x            y\n",
       "real_start_point   2024-04-10 17:22:20+00:00    87.000000\n",
       "investigating      2024-04-10 17:56:00+00:00   303.066667\n",
       "identified         2024-04-10 20:39:00+00:00  1260.000000\n",
       "monitoring         2024-04-10 20:49:00+00:00   510.000000\n",
       "resolved           2024-04-10 20:56:00+00:00   245.266667\n",
       "postmortem         2024-04-10 22:22:20+00:00    59.000000\n",
       "next_investigating 2024-04-10 22:37:20+00:00   300.000000"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "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>x</th>\n",
       "      <th>y</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>2024-04-10 17:07:20+00:00</td>\n",
       "      <td>74.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>real_start_point</th>\n",
       "      <td>2024-04-10 17:22:20+00:00</td>\n",
       "      <td>87.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>2024-04-10 17:22:20+00:00</td>\n",
       "      <td>87.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>2024-04-10 17:37:20+00:00</td>\n",
       "      <td>303.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>2024-04-10 17:52:20+00:00</td>\n",
       "      <td>294.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>investigating</th>\n",
       "      <td>2024-04-10 17:56:00+00:00</td>\n",
       "      <td>303.066667</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>2024-04-10 18:07:20+00:00</td>\n",
       "      <td>328.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>2024-04-10 18:22:20+00:00</td>\n",
       "      <td>1213.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>2024-04-10 18:37:20+00:00</td>\n",
       "      <td>4201.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>2024-04-10 18:52:20+00:00</td>\n",
       "      <td>2984.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8</th>\n",
       "      <td>2024-04-10 19:07:20+00:00</td>\n",
       "      <td>3432.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9</th>\n",
       "      <td>2024-04-10 19:22:20+00:00</td>\n",
       "      <td>4428.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>10</th>\n",
       "      <td>2024-04-10 19:37:20+00:00</td>\n",
       "      <td>3021.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>11</th>\n",
       "      <td>2024-04-10 19:52:20+00:00</td>\n",
       "      <td>2910.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>12</th>\n",
       "      <td>2024-04-10 20:07:20+00:00</td>\n",
       "      <td>2691.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>13</th>\n",
       "      <td>2024-04-10 20:22:20+00:00</td>\n",
       "      <td>2308.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>14</th>\n",
       "      <td>2024-04-10 20:37:20+00:00</td>\n",
       "      <td>1410.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>identified</th>\n",
       "      <td>2024-04-10 20:39:00+00:00</td>\n",
       "      <td>1260.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>monitoring</th>\n",
       "      <td>2024-04-10 20:49:00+00:00</td>\n",
       "      <td>510.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>15</th>\n",
       "      <td>2024-04-10 20:52:20+00:00</td>\n",
       "      <td>285.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>resolved</th>\n",
       "      <td>2024-04-10 20:56:00+00:00</td>\n",
       "      <td>245.266667</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>16</th>\n",
       "      <td>2024-04-10 21:07:20+00:00</td>\n",
       "      <td>136.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>17</th>\n",
       "      <td>2024-04-10 21:22:20+00:00</td>\n",
       "      <td>130.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>18</th>\n",
       "      <td>2024-04-10 21:37:20+00:00</td>\n",
       "      <td>94.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>19</th>\n",
       "      <td>2024-04-10 21:52:20+00:00</td>\n",
       "      <td>84.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>20</th>\n",
       "      <td>2024-04-10 22:07:20+00:00</td>\n",
       "      <td>78.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>21</th>\n",
       "      <td>2024-04-10 22:22:20+00:00</td>\n",
       "      <td>59.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>postmortem</th>\n",
       "      <td>2024-04-10 22:22:20+00:00</td>\n",
       "      <td>59.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>22</th>\n",
       "      <td>2024-04-10 22:37:20+00:00</td>\n",
       "      <td>60.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>next_investigating</th>\n",
       "      <td>2024-04-10 22:37:20+00:00</td>\n",
       "      <td>300.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>23</th>\n",
       "      <td>2024-04-10 22:52:20+00:00</td>\n",
       "      <td>50.000000</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                                           x            y\n",
       "0                  2024-04-10 17:07:20+00:00    74.000000\n",
       "real_start_point   2024-04-10 17:22:20+00:00    87.000000\n",
       "1                  2024-04-10 17:22:20+00:00    87.000000\n",
       "2                  2024-04-10 17:37:20+00:00   303.000000\n",
       "3                  2024-04-10 17:52:20+00:00   294.000000\n",
       "investigating      2024-04-10 17:56:00+00:00   303.066667\n",
       "4                  2024-04-10 18:07:20+00:00   328.000000\n",
       "5                  2024-04-10 18:22:20+00:00  1213.000000\n",
       "6                  2024-04-10 18:37:20+00:00  4201.000000\n",
       "7                  2024-04-10 18:52:20+00:00  2984.000000\n",
       "8                  2024-04-10 19:07:20+00:00  3432.000000\n",
       "9                  2024-04-10 19:22:20+00:00  4428.000000\n",
       "10                 2024-04-10 19:37:20+00:00  3021.000000\n",
       "11                 2024-04-10 19:52:20+00:00  2910.000000\n",
       "12                 2024-04-10 20:07:20+00:00  2691.000000\n",
       "13                 2024-04-10 20:22:20+00:00  2308.000000\n",
       "14                 2024-04-10 20:37:20+00:00  1410.000000\n",
       "identified         2024-04-10 20:39:00+00:00  1260.000000\n",
       "monitoring         2024-04-10 20:49:00+00:00   510.000000\n",
       "15                 2024-04-10 20:52:20+00:00   285.000000\n",
       "resolved           2024-04-10 20:56:00+00:00   245.266667\n",
       "16                 2024-04-10 21:07:20+00:00   136.000000\n",
       "17                 2024-04-10 21:22:20+00:00   130.000000\n",
       "18                 2024-04-10 21:37:20+00:00    94.000000\n",
       "19                 2024-04-10 21:52:20+00:00    84.000000\n",
       "20                 2024-04-10 22:07:20+00:00    78.000000\n",
       "21                 2024-04-10 22:22:20+00:00    59.000000\n",
       "postmortem         2024-04-10 22:22:20+00:00    59.000000\n",
       "22                 2024-04-10 22:37:20+00:00    60.000000\n",
       "next_investigating 2024-04-10 22:37:20+00:00   300.000000\n",
       "23                 2024-04-10 22:52:20+00:00    50.000000"
      ]
     },
     "execution_count": 7,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_example_tmp.loc['postmortem'] = [df_report['x'].iloc[-6], df_report['y'].iloc[-6]]  # set a postmortem status\n",
    "df_example_tmp.loc['next_investigating'] = [df_report['x'].iloc[-5] + timedelta(minutes=0), 300]  # set a next_investigating status\n",
    "df_example_tmp.loc['real_start_point'] = [df_report['x'].iloc[1] + timedelta(minutes=0), df_report['y'].iloc[1]]  # set a real start point\n",
    "df_example_tmp = df_example_tmp.sort_values(by='x')\n",
    "display(df_example_tmp)\n",
    "\n",
    "df_test = pd.concat([df_example_tmp, df_report], axis=0)\n",
    "df_test = df_test.sort_values(by='x')\n",
    "df_test = df_test[df_test['x'] <= '2024-04-10T23:00:00+00:00']\n",
    "df_test"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 1000x600 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "set_half_column_figure_font(height_multiplier=1.0)\n",
    "fig, axes = plt.subplots(nrows=1, ncols=1, figsize=(10, 6))\n",
    "fig.subplots_adjust(wspace=0.0)\n",
    "\n",
    "axes.plot(df_test['x'], df_test['y'], color='black', zorder=1, markersize=5)\n",
    "# Fill the area under the line\n",
    "axes.fill_between(df_test['x'], df_test['y'], color='black', alpha=0.1)\n",
    "\n",
    "axes.scatter(df_example_tmp['x'], df_example_tmp['y'], color='red', zorder=5)\n",
    "# show the incident status\n",
    "axes.text(df_example_tmp['x'].iloc[0], df_example_tmp['y'].iloc[0]-50, r'\\textbf{$S_0$}', color='red', fontsize=16, ha='right', va='top')\n",
    "axes.text(df_example_tmp['x'].iloc[1], df_example_tmp['y'].iloc[1]-50, r'\\textbf{$S_1$}', color='red', fontsize=16, ha='right', va='top')\n",
    "axes.text(df_example_tmp['x'].iloc[2], df_example_tmp['y'].iloc[2]-50, r'\\textbf{$S_2$}', color='red', fontsize=16, ha='right', va='top')\n",
    "axes.text(df_example_tmp['x'].iloc[3], df_example_tmp['y'].iloc[3]-50, r'\\textbf{$S_3$}', color='red', fontsize=16, ha='right', va='top')\n",
    "axes.text(df_example_tmp['x'].iloc[4], df_example_tmp['y'].iloc[4]-50, r'\\textbf{$S_4$}', color='red', fontsize=16, ha='right', va='top')\n",
    "axes.text(df_example_tmp['x'].iloc[5], df_example_tmp['y'].iloc[5]-50, r'\\textbf{$S_5$}', color='red', fontsize=16, ha='right', va='top')\n",
    "axes.text(df_example_tmp['x'].iloc[6], df_example_tmp['y'].iloc[6]-50, r'\\textbf{$S_1$}', color='red', fontsize=16, ha='right', va='top')\n",
    "\n",
    "for x in df_example_tmp['x'][1:]:\n",
    "    axes.axvline(x=x, color='red', linestyle='--', linewidth=1)\n",
    "\n",
    "axes.plot([df_example_tmp['x'].iloc[1], df_example_tmp['x'].iloc[2]], [df_example_tmp['y'].iloc[1], df_example_tmp['y'].iloc[1]], color='blue', linestyle='-', linewidth=2, label='P1')\n",
    "axes.plot([df_example_tmp['x'].iloc[2], df_example_tmp['x'].iloc[3]], [df_example_tmp['y'].iloc[2], df_example_tmp['y'].iloc[2]], color='blue', linestyle='-', linewidth=2, label='P2')\n",
    "axes.plot([df_example_tmp['x'].iloc[3], df_example_tmp['x'].iloc[4]], [df_example_tmp['y'].iloc[3], df_example_tmp['y'].iloc[3]], color='blue', linestyle='-', linewidth=2, label='P3')\n",
    "axes.plot([df_example_tmp['x'].iloc[4], df_example_tmp['x'].iloc[5]], [df_example_tmp['y'].iloc[4], df_example_tmp['y'].iloc[4]], color='blue', linestyle='-', linewidth=2, label='P4')\n",
    "axes.plot([df_example_tmp['x'].iloc[1], df_example_tmp['x'].iloc[4]], [3000, 3000], color='blue', marker='o', linestyle='--', linewidth=2, label='MTTR')\n",
    "axes.plot([df_example_tmp['x'].iloc[4], df_example_tmp['x'].iloc[6]], [3500, 3500], color='blue', marker='o', linestyle='--', linewidth=2, label='MTBF')\n",
    "\n",
    "axes.text(df_example_tmp['x'].iloc[1] + (df_example_tmp['x'].iloc[2] - df_example_tmp['x'].iloc[1]) / 2, df_example_tmp['y'].iloc[1]+50, r'\\textbf{$P_I$}', color='blue', fontsize=16, ha='center', va='bottom')\n",
    "axes.text(df_example_tmp['x'].iloc[2] + (df_example_tmp['x'].iloc[3] - df_example_tmp['x'].iloc[2]) / 2, df_example_tmp['y'].iloc[2]+50, r'\\textbf{$P_R$}', color='blue', fontsize=16, ha='center', va='bottom')\n",
    "axes.text(df_example_tmp['x'].iloc[3] + (df_example_tmp['x'].iloc[4] - df_example_tmp['x'].iloc[3]) / 2, df_example_tmp['y'].iloc[3]+50, r'\\textbf{$P_C$}', color='blue', fontsize=16, ha='center', va='bottom')\n",
    "axes.text(df_example_tmp['x'].iloc[4] + (df_example_tmp['x'].iloc[5] - df_example_tmp['x'].iloc[4]) / 2, df_example_tmp['y'].iloc[4]+50, r'\\textbf{$P_L$}', color='blue', fontsize=16, ha='center', va='bottom')\n",
    "axes.text(df_example_tmp['x'].iloc[1] + (df_example_tmp['x'].iloc[4] - df_example_tmp['x'].iloc[1]) / 2, 3000+50, r'\\textbf{$Time\\ To\\ Resolve$}', color='blue', fontsize=16, ha='center', va='bottom')\n",
    "axes.text(df_example_tmp['x'].iloc[4] + (df_example_tmp['x'].iloc[6] - df_example_tmp['x'].iloc[4]) / 2, 3500+50, r'\\textbf{$Time\\ Between\\ Failures$}', color='blue', fontsize=16, ha='center', va='bottom')\n",
    "\n",
    "axes.set_xlim(df_test['x'].iloc[0], df_test['x'].iloc[-1])\n",
    "axes.set_ylim(0, 4500)\n",
    "axes.set_yticklabels(axes.get_yticks(), fontsize=18)\n",
    "axes.set_xticklabels(axes.get_xticklabels(), fontsize=15)\n",
    "axes.xaxis.set_major_formatter(mdates.DateFormatter('%Y-%m-%d %H:%M'))\n",
    "axes.set_ylabel(r'\\textbf{\\# User Reports}', fontsize=18)\n",
    "axes.get_yaxis().set_major_formatter(plt.FuncFormatter(lambda x, loc: \"{:,}\".format(int(x))))\n",
    "plt.gca().xaxis.set_major_formatter(mdates.DateFormatter('%Y-%m-%d %H:%M'))\n",
    "fig.autofmt_xdate()\n",
    "\n",
    "# save as pdf\n",
    "plt.savefig('plot/sec2-failure_modeling/fig-F-R-modeling.pdf', format='pdf', bbox_inches='tight')\n",
    "plt.show()\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3",
   "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": "3.12.5"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}