{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Callback api\n",
    "\n",
    "Вконтакт умеет самостоятельно высылать все события происходящие в группе посредством `callback api`\n",
    "\n",
    "- https://vk.com/dev/callback_api\n",
    "- https://vk.com/dev/groups_events\n",
    "\n",
    "В данном документе собран пример обработки этих данных\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Описание выгрузки\n",
    "\n",
    "- events_statistics.csv - посуточная статистика, по всем событиям\n",
    "- events_moderator.csv - посуточная статистика по действия модератора/группы\n",
    "- events_users.cs - посуточная статистика по действия мпользователей\n",
    "- moderation_delete_stat.csv - данные по удалению комментариев\n",
    "- moderation_delete_meantime.csv - среднее время до удаления\n",
    "- moderation_response_stat.csv - данные по ответам на комментарии модератором / группой\n",
    "- moderation_response_meantime.csv - среднее время до отевта модератором / группой"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Загрузка базы событий"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "import matplotlib as plt\n",
    "import pandas as pd\n",
    "import numpy as np"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [],
   "source": [
    "# очистка статистки\n",
    "!rm fanta/*"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [],
   "source": [
    "moderator_id = 13503331\n",
    "group_id = -24539231\n",
    "\n",
    "events_txt = {'message_new': 'входящее сообщение', 'message_reply': 'новое исходящее сообщение', 'message_edit': 'редактирование сообщения', 'message_allow': 'подписка на сообщения от сообщества', 'message_deny': 'новый запрет сообщений от сообщества', 'photo_new': 'добавление фотографии', 'photo_comment_new': 'добавление комментария к фотографии', 'photo_comment_edit': 'редактирование комментария к фотографии', 'photo_comment_restore': 'восстановление комментария к фотографии', 'photo_comment_delete': 'удаление комментария к фотографии', 'audio_new': 'добавление аудио', 'video_new': 'добавление видео', 'video_comment_new': 'комментарий к видео', 'video_comment_edit': 'редактирование комментария к видео', 'video_comment_restore': 'восстановление комментария к видео', 'video_comment_delete': 'удаление комментария к видео', 'wall_post_new': 'запись на стене', 'wall_repost': 'репост записи из сообщества', 'wall_reply_new': 'добавление комментария на стене', 'wall_reply_edit': 'редактирование комментария на стене', 'wall_reply_restore': 'восстановление комментария на стене', 'wall_reply_delete': 'удаление комментария на стене', 'board_post_new': 'создание комментария в обсуждении', 'board_post_edit': 'редактирование комментария', 'board_post_restore': 'восстановление комментария', 'board_post_delete': 'удаление комментария в обсуждении', 'market_comment_new': 'новый комментарий к товару', 'market_comment_edit': 'редактирование комментария к товару', 'market_comment_restore': 'восстановление комментария к товару', 'market_comment_delete': 'удаление комментария к товару', 'group_leave': 'удаление участника из сообщества', 'group_join': 'добавление участника или заявки на вступление в сообщество', 'user_block': 'добавление пользователя в чёрный список', 'user_unblock': 'удаление пользователя из чёрного списка', 'poll_vote_new': 'добавление голоса в публичном опросе', 'group_officers_edit': 'редактирование списка руководителей'}"
   ]
  },
  {
   "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>id</th>\n",
       "      <th>group_id</th>\n",
       "      <th>type</th>\n",
       "      <th>date</th>\n",
       "      <th>object</th>\n",
       "      <th>from_id</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>181741210</td>\n",
       "      <td>wall_reply_new</td>\n",
       "      <td>2019-09-28 16:26:46.508471</td>\n",
       "      <td>{'id': 39, 'from_id': 542380738, 'date': 15696...</td>\n",
       "      <td>542380738</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>1</td>\n",
       "      <td>2</td>\n",
       "      <td>181741210</td>\n",
       "      <td>wall_reply_new</td>\n",
       "      <td>2019-09-28 16:31:52.345462</td>\n",
       "      <td>{'id': 34, 'from_id': 542380738, 'date': 15696...</td>\n",
       "      <td>542380738</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>2</td>\n",
       "      <td>3</td>\n",
       "      <td>24539231</td>\n",
       "      <td>message_new</td>\n",
       "      <td>2019-09-28 18:00:01.024108</td>\n",
       "      <td>{'id': 494346, 'date': 1569693600, 'out': 0, '...</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>3</td>\n",
       "      <td>4</td>\n",
       "      <td>24539231</td>\n",
       "      <td>message_new</td>\n",
       "      <td>2019-09-28 18:01:39.427268</td>\n",
       "      <td>{'id': 494347, 'date': 1569693699, 'out': 0, '...</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>4</td>\n",
       "      <td>5</td>\n",
       "      <td>24539231</td>\n",
       "      <td>wall_reply_new</td>\n",
       "      <td>2019-09-28 18:02:10.980445</td>\n",
       "      <td>{'id': 263514, 'from_id': 64237711, 'date': 15...</td>\n",
       "      <td>64237711</td>\n",
       "    </tr>\n",
       "    <tr>\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",
       "      <td>75590</td>\n",
       "      <td>75591</td>\n",
       "      <td>24539231</td>\n",
       "      <td>group_join</td>\n",
       "      <td>2019-11-07 14:41:49.294970</td>\n",
       "      <td>{'user_id': 552953951, 'join_type': 'join'}</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>75591</td>\n",
       "      <td>75592</td>\n",
       "      <td>24539231</td>\n",
       "      <td>group_join</td>\n",
       "      <td>2019-11-07 14:42:17.045110</td>\n",
       "      <td>{'user_id': 226261179, 'join_type': 'join'}</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>75592</td>\n",
       "      <td>75593</td>\n",
       "      <td>24539231</td>\n",
       "      <td>group_leave</td>\n",
       "      <td>2019-11-07 14:42:22.021598</td>\n",
       "      <td>{'user_id': 226261179, 'self': 1}</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>75593</td>\n",
       "      <td>75594</td>\n",
       "      <td>24539231</td>\n",
       "      <td>message_new</td>\n",
       "      <td>2019-11-07 14:42:33.620959</td>\n",
       "      <td>{'id': 527408, 'date': 1573137753, 'out': 0, '...</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>75594</td>\n",
       "      <td>75595</td>\n",
       "      <td>24539231</td>\n",
       "      <td>group_join</td>\n",
       "      <td>2019-11-07 14:43:11.935461</td>\n",
       "      <td>{'user_id': 416842775, 'join_type': 'join'}</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>75595 rows × 6 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "          id   group_id            type                       date  \\\n",
       "0          1  181741210  wall_reply_new 2019-09-28 16:26:46.508471   \n",
       "1          2  181741210  wall_reply_new 2019-09-28 16:31:52.345462   \n",
       "2          3   24539231     message_new 2019-09-28 18:00:01.024108   \n",
       "3          4   24539231     message_new 2019-09-28 18:01:39.427268   \n",
       "4          5   24539231  wall_reply_new 2019-09-28 18:02:10.980445   \n",
       "...      ...        ...             ...                        ...   \n",
       "75590  75591   24539231      group_join 2019-11-07 14:41:49.294970   \n",
       "75591  75592   24539231      group_join 2019-11-07 14:42:17.045110   \n",
       "75592  75593   24539231     group_leave 2019-11-07 14:42:22.021598   \n",
       "75593  75594   24539231     message_new 2019-11-07 14:42:33.620959   \n",
       "75594  75595   24539231      group_join 2019-11-07 14:43:11.935461   \n",
       "\n",
       "                                                  object    from_id  \n",
       "0      {'id': 39, 'from_id': 542380738, 'date': 15696...  542380738  \n",
       "1      {'id': 34, 'from_id': 542380738, 'date': 15696...  542380738  \n",
       "2      {'id': 494346, 'date': 1569693600, 'out': 0, '...        NaN  \n",
       "3      {'id': 494347, 'date': 1569693699, 'out': 0, '...        NaN  \n",
       "4      {'id': 263514, 'from_id': 64237711, 'date': 15...   64237711  \n",
       "...                                                  ...        ...  \n",
       "75590        {'user_id': 552953951, 'join_type': 'join'}        NaN  \n",
       "75591        {'user_id': 226261179, 'join_type': 'join'}        NaN  \n",
       "75592                  {'user_id': 226261179, 'self': 1}        NaN  \n",
       "75593  {'id': 527408, 'date': 1573137753, 'out': 0, '...        NaN  \n",
       "75594        {'user_id': 416842775, 'join_type': 'join'}        NaN  \n",
       "\n",
       "[75595 rows x 6 columns]"
      ]
     },
     "execution_count": 4,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "import sqlite3\n",
    "from pandas.io.json  import json_normalize\n",
    "import json\n",
    "\n",
    "# забираем базу\n",
    "db = sqlite3.connect('file:/home/v.seregin/lab/fanta/database.sqlite?mode=ro', uri=True)\n",
    "df = pd.read_sql_query('select * from event', db)\n",
    "df['date'] = pd.to_datetime(df['date'])\n",
    "df['object'] = df['object'].apply(json.loads)\n",
    "df['from_id'] = df['object'].apply(lambda x: x.get('from_id',np.nan)).astype(pd.Int64Dtype())\n",
    "df"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Общая постуточная статистика по событиям"
   ]
  },
  {
   "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></th>\n",
       "      <th>Общая посуточная статистика пользователи + модераторы</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>date</th>\n",
       "      <th>type</th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <td rowspan=\"5\" valign=\"top\">2019-09-28</td>\n",
       "      <td>group_join</td>\n",
       "      <td>28</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>group_leave</td>\n",
       "      <td>21</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>message_allow</td>\n",
       "      <td>4</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>message_deny</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>message_new</td>\n",
       "      <td>21</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td rowspan=\"5\" valign=\"top\">2019-11-07</td>\n",
       "      <td>message_new</td>\n",
       "      <td>156</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>message_reply</td>\n",
       "      <td>282</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>photo_comment_new</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>wall_reply_delete</td>\n",
       "      <td>4</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>wall_reply_new</td>\n",
       "      <td>7</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>496 rows × 1 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "                              Общая посуточная статистика пользователи + модераторы\n",
       "date       type                                                                    \n",
       "2019-09-28 group_join                                                        28    \n",
       "           group_leave                                                       21    \n",
       "           message_allow                                                      4    \n",
       "           message_deny                                                       1    \n",
       "           message_new                                                       21    \n",
       "...                                                                         ...    \n",
       "2019-11-07 message_new                                                      156    \n",
       "           message_reply                                                    282    \n",
       "           photo_comment_new                                                  1    \n",
       "           wall_reply_delete                                                  4    \n",
       "           wall_reply_new                                                     7    \n",
       "\n",
       "[496 rows x 1 columns]"
      ]
     },
     "execution_count": 5,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_stat = df[['type','id']].groupby([df.date.dt.date, 'type']).agg('count')\n",
    "df_stat.columns = [\"Общая посуточная статистика пользователи + модераторы\"]\n",
    "df_stat.to_csv(\"fanta/events_statistics.csv\")\n",
    "df_stat"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Постуточная статистика по событиям от модератора"
   ]
  },
  {
   "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></th>\n",
       "      <th>Ко-во событий от имени модератора / группы</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>date</th>\n",
       "      <th>type</th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <td>2019-09-29</td>\n",
       "      <td>wall_reply_new</td>\n",
       "      <td>8</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>2019-09-30</td>\n",
       "      <td>wall_reply_new</td>\n",
       "      <td>3</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>2019-10-01</td>\n",
       "      <td>wall_reply_new</td>\n",
       "      <td>8</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>2019-10-02</td>\n",
       "      <td>wall_reply_new</td>\n",
       "      <td>13</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td rowspan=\"2\" valign=\"top\">2019-10-03</td>\n",
       "      <td>message_reply</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>wall_reply_new</td>\n",
       "      <td>9</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td rowspan=\"3\" valign=\"top\">2019-10-04</td>\n",
       "      <td>wall_reply_edit</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>wall_reply_new</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>wall_reply_restore</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>2019-10-05</td>\n",
       "      <td>wall_reply_new</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                               Ко-во событий от имени модератора / группы\n",
       "date       type                                                          \n",
       "2019-09-29 wall_reply_new                                               8\n",
       "2019-09-30 wall_reply_new                                               3\n",
       "2019-10-01 wall_reply_new                                               8\n",
       "2019-10-02 wall_reply_new                                              13\n",
       "2019-10-03 message_reply                                                2\n",
       "           wall_reply_new                                               9\n",
       "2019-10-04 wall_reply_edit                                              1\n",
       "           wall_reply_new                                               2\n",
       "           wall_reply_restore                                           1\n",
       "2019-10-05 wall_reply_new                                               2"
      ]
     },
     "execution_count": 6,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_stat = df[(df[\"from_id\"]==moderator_id) | (df[\"from_id\"]==group_id)][['type','id']].groupby([df.date.dt.date, 'type']).agg('count')\n",
    "df_stat.columns = [\"Ко-во событий от имени модератора / группы\"]\n",
    "df_stat.to_csv(\"fanta/events_moderator.csv\")\n",
    "df_stat.head(10)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Постуточная статистика по событиям от прочих участников"
   ]
  },
  {
   "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></th>\n",
       "      <th>Ко-во событий</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>date</th>\n",
       "      <th>type</th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <td rowspan=\"8\" valign=\"top\">2019-09-28</td>\n",
       "      <td>group_join</td>\n",
       "      <td>28</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>group_leave</td>\n",
       "      <td>21</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>message_allow</td>\n",
       "      <td>4</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>message_deny</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>message_new</td>\n",
       "      <td>21</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>message_reply</td>\n",
       "      <td>4</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>wall_reply_delete</td>\n",
       "      <td>4</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>wall_reply_new</td>\n",
       "      <td>7</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td rowspan=\"12\" valign=\"top\">2019-09-29</td>\n",
       "      <td>group_join</td>\n",
       "      <td>708</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>group_leave</td>\n",
       "      <td>198</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>message_allow</td>\n",
       "      <td>7</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>message_deny</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>message_edit</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>message_new</td>\n",
       "      <td>139</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>message_reply</td>\n",
       "      <td>16</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>photo_comment_new</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>user_unblock</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>wall_reply_delete</td>\n",
       "      <td>8</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>wall_reply_edit</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>wall_reply_new</td>\n",
       "      <td>27</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                              Ко-во событий\n",
       "date       type                            \n",
       "2019-09-28 group_join                    28\n",
       "           group_leave                   21\n",
       "           message_allow                  4\n",
       "           message_deny                   1\n",
       "           message_new                   21\n",
       "           message_reply                  4\n",
       "           wall_reply_delete              4\n",
       "           wall_reply_new                 7\n",
       "2019-09-29 group_join                   708\n",
       "           group_leave                  198\n",
       "           message_allow                  7\n",
       "           message_deny                   1\n",
       "           message_edit                   1\n",
       "           message_new                  139\n",
       "           message_reply                 16\n",
       "           photo_comment_new              1\n",
       "           user_unblock                   1\n",
       "           wall_reply_delete              8\n",
       "           wall_reply_edit                1\n",
       "           wall_reply_new                27"
      ]
     },
     "execution_count": 7,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_stat = df[(df[\"from_id\"]!=group_id) & (df[\"from_id\"]!=moderator_id)][['type','id']].groupby([df.date.dt.date, 'type']).agg('count')\n",
    "df_stat.columns = [\"Ко-во событий\"]\n",
    "df_stat.to_csv(\"fanta/events_users.csv\")\n",
    "df_stat.head(20)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/home/v.seregin/lab/jupyter/venv/lib/python3.6/site-packages/pandas/plotting/_matplotlib/core.py:338: RuntimeWarning: More than 20 figures have been opened. Figures created through the pyplot interface (`matplotlib.pyplot.figure`) are retained until explicitly closed and may consume too much memory. (To control this warning, see the rcParam `figure.max_open_warning`).\n",
      "  fig = self.plt.figure(figsize=self.figsize)\n"
     ]
    }
   ],
   "source": [
    "for event in df_stat.index.get_level_values(1).unique().values:\n",
    "    df_event = df_stat.xs(event, level=1)\n",
    "    idx = pd.date_range(df_stat.index[0][0], df_stat.index[-1][0])\n",
    "    df_event = df_event.reindex(idx, fill_value=0)\n",
    "    df_event.columns = (f\"{events_txt.get(event, event)} ({event})\",)\n",
    "    df_event.plot(figsize=(20,10)).get_figure().savefig(f\"fanta/{event}.png\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Обработка событий"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "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>id</th>\n",
       "      <th>group_id</th>\n",
       "      <th>type</th>\n",
       "      <th>date</th>\n",
       "      <th>object</th>\n",
       "      <th>from_id</th>\n",
       "      <th>reply_id</th>\n",
       "      <th>deleter_id</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>181741210</td>\n",
       "      <td>wall_reply_new</td>\n",
       "      <td>2019-09-28 16:26:46.508471</td>\n",
       "      <td>{'id': 39, 'from_id': 542380738, 'date': 15696...</td>\n",
       "      <td>542380738</td>\n",
       "      <td>39</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>1</td>\n",
       "      <td>2</td>\n",
       "      <td>181741210</td>\n",
       "      <td>wall_reply_new</td>\n",
       "      <td>2019-09-28 16:31:52.345462</td>\n",
       "      <td>{'id': 34, 'from_id': 542380738, 'date': 15696...</td>\n",
       "      <td>542380738</td>\n",
       "      <td>34</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>2</td>\n",
       "      <td>3</td>\n",
       "      <td>24539231</td>\n",
       "      <td>message_new</td>\n",
       "      <td>2019-09-28 18:00:01.024108</td>\n",
       "      <td>{'id': 494346, 'date': 1569693600, 'out': 0, '...</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>3</td>\n",
       "      <td>4</td>\n",
       "      <td>24539231</td>\n",
       "      <td>message_new</td>\n",
       "      <td>2019-09-28 18:01:39.427268</td>\n",
       "      <td>{'id': 494347, 'date': 1569693699, 'out': 0, '...</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>4</td>\n",
       "      <td>5</td>\n",
       "      <td>24539231</td>\n",
       "      <td>wall_reply_new</td>\n",
       "      <td>2019-09-28 18:02:10.980445</td>\n",
       "      <td>{'id': 263514, 'from_id': 64237711, 'date': 15...</td>\n",
       "      <td>64237711</td>\n",
       "      <td>263514</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\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",
       "      <td>75590</td>\n",
       "      <td>75591</td>\n",
       "      <td>24539231</td>\n",
       "      <td>group_join</td>\n",
       "      <td>2019-11-07 14:41:49.294970</td>\n",
       "      <td>{'user_id': 552953951, 'join_type': 'join'}</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>75591</td>\n",
       "      <td>75592</td>\n",
       "      <td>24539231</td>\n",
       "      <td>group_join</td>\n",
       "      <td>2019-11-07 14:42:17.045110</td>\n",
       "      <td>{'user_id': 226261179, 'join_type': 'join'}</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>75592</td>\n",
       "      <td>75593</td>\n",
       "      <td>24539231</td>\n",
       "      <td>group_leave</td>\n",
       "      <td>2019-11-07 14:42:22.021598</td>\n",
       "      <td>{'user_id': 226261179, 'self': 1}</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>75593</td>\n",
       "      <td>75594</td>\n",
       "      <td>24539231</td>\n",
       "      <td>message_new</td>\n",
       "      <td>2019-11-07 14:42:33.620959</td>\n",
       "      <td>{'id': 527408, 'date': 1573137753, 'out': 0, '...</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>75594</td>\n",
       "      <td>75595</td>\n",
       "      <td>24539231</td>\n",
       "      <td>group_join</td>\n",
       "      <td>2019-11-07 14:43:11.935461</td>\n",
       "      <td>{'user_id': 416842775, 'join_type': 'join'}</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>75595 rows × 8 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "          id   group_id            type                       date  \\\n",
       "0          1  181741210  wall_reply_new 2019-09-28 16:26:46.508471   \n",
       "1          2  181741210  wall_reply_new 2019-09-28 16:31:52.345462   \n",
       "2          3   24539231     message_new 2019-09-28 18:00:01.024108   \n",
       "3          4   24539231     message_new 2019-09-28 18:01:39.427268   \n",
       "4          5   24539231  wall_reply_new 2019-09-28 18:02:10.980445   \n",
       "...      ...        ...             ...                        ...   \n",
       "75590  75591   24539231      group_join 2019-11-07 14:41:49.294970   \n",
       "75591  75592   24539231      group_join 2019-11-07 14:42:17.045110   \n",
       "75592  75593   24539231     group_leave 2019-11-07 14:42:22.021598   \n",
       "75593  75594   24539231     message_new 2019-11-07 14:42:33.620959   \n",
       "75594  75595   24539231      group_join 2019-11-07 14:43:11.935461   \n",
       "\n",
       "                                                  object    from_id  reply_id  \\\n",
       "0      {'id': 39, 'from_id': 542380738, 'date': 15696...  542380738        39   \n",
       "1      {'id': 34, 'from_id': 542380738, 'date': 15696...  542380738        34   \n",
       "2      {'id': 494346, 'date': 1569693600, 'out': 0, '...        NaN       NaN   \n",
       "3      {'id': 494347, 'date': 1569693699, 'out': 0, '...        NaN       NaN   \n",
       "4      {'id': 263514, 'from_id': 64237711, 'date': 15...   64237711    263514   \n",
       "...                                                  ...        ...       ...   \n",
       "75590        {'user_id': 552953951, 'join_type': 'join'}        NaN       NaN   \n",
       "75591        {'user_id': 226261179, 'join_type': 'join'}        NaN       NaN   \n",
       "75592                  {'user_id': 226261179, 'self': 1}        NaN       NaN   \n",
       "75593  {'id': 527408, 'date': 1573137753, 'out': 0, '...        NaN       NaN   \n",
       "75594        {'user_id': 416842775, 'join_type': 'join'}        NaN       NaN   \n",
       "\n",
       "       deleter_id  \n",
       "0             NaN  \n",
       "1             NaN  \n",
       "2             NaN  \n",
       "3             NaN  \n",
       "4             NaN  \n",
       "...           ...  \n",
       "75590         NaN  \n",
       "75591         NaN  \n",
       "75592         NaN  \n",
       "75593         NaN  \n",
       "75594         NaN  \n",
       "\n",
       "[75595 rows x 8 columns]"
      ]
     },
     "execution_count": 9,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# id ответа\n",
    "df['reply_id'] = df[df.type.isin(['wall_reply_new', 'wall_reply_delete'])]['object'].apply(lambda x: x['id']).astype(pd.Int64Dtype())\n",
    "df['deleter_id'] = df[df.type == 'wall_reply_delete']['object'].apply(lambda x: x['deleter_id']).astype(pd.Int64Dtype())\n",
    "df\n",
    "# df['from_id'] = df['object'].apply(lambda x: int(x.get('from_id', 0)))\n",
    "# df['user_id'] = df['object'].apply(lambda x: int(x.get('user_id', 0)))\n",
    "# df['deleter_id'] = df['object'].apply(lambda x: int(x.get('deleter_id', 0)))\n",
    "# df['post_id'] = df['object'].apply(lambda x: int(x.get('post_id', 0)))\n",
    "# df['comment_id'] = df['object'].apply(lambda x: x.get('id',0))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Расчет времени до удаления коммента"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "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>reply_id</th>\n",
       "      <th>date_x</th>\n",
       "      <th>deleter_id</th>\n",
       "      <th>date_y</th>\n",
       "      <th>response_time</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <td>0</td>\n",
       "      <td>263515</td>\n",
       "      <td>2019-09-28 19:35:40.612116</td>\n",
       "      <td>13503331</td>\n",
       "      <td>2019-09-28 22:24:44.686827</td>\n",
       "      <td>2.817778</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>1</td>\n",
       "      <td>263516</td>\n",
       "      <td>2019-09-28 19:35:52.326412</td>\n",
       "      <td>13503331</td>\n",
       "      <td>2019-09-28 22:24:43.482553</td>\n",
       "      <td>2.814167</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>2</td>\n",
       "      <td>263516</td>\n",
       "      <td>2019-09-28 19:35:52.326412</td>\n",
       "      <td>13503331</td>\n",
       "      <td>2019-09-28 22:24:44.604662</td>\n",
       "      <td>2.814444</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>3</td>\n",
       "      <td>263517</td>\n",
       "      <td>2019-09-28 19:36:00.270410</td>\n",
       "      <td>13503331</td>\n",
       "      <td>2019-09-28 22:24:44.548332</td>\n",
       "      <td>2.812222</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>4</td>\n",
       "      <td>263519</td>\n",
       "      <td>2019-09-29 04:21:43.783550</td>\n",
       "      <td>13503331</td>\n",
       "      <td>2019-09-29 07:15:33.252939</td>\n",
       "      <td>2.896944</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>440</td>\n",
       "      <td>265792</td>\n",
       "      <td>2019-11-06 10:17:16.799365</td>\n",
       "      <td>100</td>\n",
       "      <td>2019-11-06 10:22:50.098677</td>\n",
       "      <td>0.092500</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>441</td>\n",
       "      <td>265793</td>\n",
       "      <td>2019-11-06 10:17:20.684167</td>\n",
       "      <td>100</td>\n",
       "      <td>2019-11-06 10:22:50.207598</td>\n",
       "      <td>0.091389</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>442</td>\n",
       "      <td>265824</td>\n",
       "      <td>2019-11-06 20:17:43.343814</td>\n",
       "      <td>100</td>\n",
       "      <td>2019-11-07 09:00:27.109002</td>\n",
       "      <td>12.711944</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>443</td>\n",
       "      <td>265829</td>\n",
       "      <td>2019-11-07 04:40:26.048848</td>\n",
       "      <td>100</td>\n",
       "      <td>2019-11-07 04:40:40.887074</td>\n",
       "      <td>0.003889</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>444</td>\n",
       "      <td>265830</td>\n",
       "      <td>2019-11-07 10:54:40.131366</td>\n",
       "      <td>13503331</td>\n",
       "      <td>2019-11-07 11:17:15.972805</td>\n",
       "      <td>0.376389</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>445 rows × 5 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "     reply_id                     date_x  deleter_id  \\\n",
       "0      263515 2019-09-28 19:35:40.612116    13503331   \n",
       "1      263516 2019-09-28 19:35:52.326412    13503331   \n",
       "2      263516 2019-09-28 19:35:52.326412    13503331   \n",
       "3      263517 2019-09-28 19:36:00.270410    13503331   \n",
       "4      263519 2019-09-29 04:21:43.783550    13503331   \n",
       "..        ...                        ...         ...   \n",
       "440    265792 2019-11-06 10:17:16.799365         100   \n",
       "441    265793 2019-11-06 10:17:20.684167         100   \n",
       "442    265824 2019-11-06 20:17:43.343814         100   \n",
       "443    265829 2019-11-07 04:40:26.048848         100   \n",
       "444    265830 2019-11-07 10:54:40.131366    13503331   \n",
       "\n",
       "                        date_y  response_time  \n",
       "0   2019-09-28 22:24:44.686827       2.817778  \n",
       "1   2019-09-28 22:24:43.482553       2.814167  \n",
       "2   2019-09-28 22:24:44.604662       2.814444  \n",
       "3   2019-09-28 22:24:44.548332       2.812222  \n",
       "4   2019-09-29 07:15:33.252939       2.896944  \n",
       "..                         ...            ...  \n",
       "440 2019-11-06 10:22:50.098677       0.092500  \n",
       "441 2019-11-06 10:22:50.207598       0.091389  \n",
       "442 2019-11-07 09:00:27.109002      12.711944  \n",
       "443 2019-11-07 04:40:40.887074       0.003889  \n",
       "444 2019-11-07 11:17:15.972805       0.376389  \n",
       "\n",
       "[445 rows x 5 columns]"
      ]
     },
     "execution_count": 10,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_new = df[df.type=='wall_reply_new'][['reply_id','date']]\n",
    "df_del = df[df.type=='wall_reply_delete'][['reply_id','deleter_id','date']]\n",
    "df_date = pd.merge(df_new, df_del, on='reply_id')\n",
    "df_date['response_time'] = (df_date.date_y - df_date.date_x).dt.seconds/60.0/60.0\n",
    "df_date.to_csv(\"fanta/moderation_delete_stat.csv\", index=None)\n",
    "df_date"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Среднее суточное время до удаления"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "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 tr th {\n",
       "        text-align: left;\n",
       "    }\n",
       "\n",
       "    .dataframe thead tr:last-of-type th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr>\n",
       "      <th></th>\n",
       "      <th colspan=\"4\" halign=\"left\">response_time</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th></th>\n",
       "      <th>count</th>\n",
       "      <th>mean</th>\n",
       "      <th>max</th>\n",
       "      <th>min</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>date_x</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <td>2019-09-28</td>\n",
       "      <td>4</td>\n",
       "      <td>2.814653</td>\n",
       "      <td>2.817778</td>\n",
       "      <td>2.812222</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>2019-09-29</td>\n",
       "      <td>8</td>\n",
       "      <td>0.624653</td>\n",
       "      <td>2.896944</td>\n",
       "      <td>0.001944</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>2019-09-30</td>\n",
       "      <td>5</td>\n",
       "      <td>3.786722</td>\n",
       "      <td>16.581944</td>\n",
       "      <td>0.003611</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>2019-10-01</td>\n",
       "      <td>3</td>\n",
       "      <td>0.021944</td>\n",
       "      <td>0.053611</td>\n",
       "      <td>0.003333</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>2019-10-02</td>\n",
       "      <td>8</td>\n",
       "      <td>1.056806</td>\n",
       "      <td>5.731667</td>\n",
       "      <td>0.001389</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>2019-10-03</td>\n",
       "      <td>10</td>\n",
       "      <td>0.806972</td>\n",
       "      <td>4.398611</td>\n",
       "      <td>0.001111</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>2019-10-04</td>\n",
       "      <td>11</td>\n",
       "      <td>0.610379</td>\n",
       "      <td>4.882500</td>\n",
       "      <td>0.000556</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>2019-10-05</td>\n",
       "      <td>4</td>\n",
       "      <td>0.329861</td>\n",
       "      <td>1.104722</td>\n",
       "      <td>0.001667</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>2019-10-06</td>\n",
       "      <td>11</td>\n",
       "      <td>2.345530</td>\n",
       "      <td>4.278611</td>\n",
       "      <td>0.001389</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>2019-10-07</td>\n",
       "      <td>9</td>\n",
       "      <td>6.458796</td>\n",
       "      <td>23.520833</td>\n",
       "      <td>0.003056</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>2019-10-08</td>\n",
       "      <td>16</td>\n",
       "      <td>2.372778</td>\n",
       "      <td>3.566111</td>\n",
       "      <td>0.001389</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>2019-10-09</td>\n",
       "      <td>7</td>\n",
       "      <td>0.251190</td>\n",
       "      <td>0.718333</td>\n",
       "      <td>0.000556</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>2019-10-10</td>\n",
       "      <td>9</td>\n",
       "      <td>1.505556</td>\n",
       "      <td>2.775000</td>\n",
       "      <td>0.202222</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>2019-10-11</td>\n",
       "      <td>6</td>\n",
       "      <td>9.813565</td>\n",
       "      <td>13.176111</td>\n",
       "      <td>1.819722</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>2019-10-12</td>\n",
       "      <td>7</td>\n",
       "      <td>1.625794</td>\n",
       "      <td>9.906944</td>\n",
       "      <td>0.000833</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>2019-10-13</td>\n",
       "      <td>6</td>\n",
       "      <td>0.611343</td>\n",
       "      <td>2.751389</td>\n",
       "      <td>0.003611</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>2019-10-14</td>\n",
       "      <td>1</td>\n",
       "      <td>0.585000</td>\n",
       "      <td>0.585000</td>\n",
       "      <td>0.585000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>2019-10-15</td>\n",
       "      <td>9</td>\n",
       "      <td>0.485000</td>\n",
       "      <td>1.286667</td>\n",
       "      <td>0.105833</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>2019-10-16</td>\n",
       "      <td>12</td>\n",
       "      <td>0.892269</td>\n",
       "      <td>2.657778</td>\n",
       "      <td>0.026389</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>2019-10-17</td>\n",
       "      <td>3</td>\n",
       "      <td>1.171667</td>\n",
       "      <td>2.129722</td>\n",
       "      <td>0.003889</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>2019-10-18</td>\n",
       "      <td>7</td>\n",
       "      <td>2.234722</td>\n",
       "      <td>11.442500</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>2019-10-19</td>\n",
       "      <td>17</td>\n",
       "      <td>3.442925</td>\n",
       "      <td>6.890000</td>\n",
       "      <td>0.001111</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>2019-10-20</td>\n",
       "      <td>11</td>\n",
       "      <td>2.098788</td>\n",
       "      <td>9.535278</td>\n",
       "      <td>0.001667</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>2019-10-21</td>\n",
       "      <td>7</td>\n",
       "      <td>4.186587</td>\n",
       "      <td>23.978889</td>\n",
       "      <td>0.124444</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>2019-10-22</td>\n",
       "      <td>1</td>\n",
       "      <td>22.062222</td>\n",
       "      <td>22.062222</td>\n",
       "      <td>22.062222</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>2019-10-23</td>\n",
       "      <td>21</td>\n",
       "      <td>2.598201</td>\n",
       "      <td>21.864444</td>\n",
       "      <td>0.001389</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>2019-10-24</td>\n",
       "      <td>9</td>\n",
       "      <td>8.162191</td>\n",
       "      <td>22.707778</td>\n",
       "      <td>0.071667</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>2019-10-25</td>\n",
       "      <td>11</td>\n",
       "      <td>4.081313</td>\n",
       "      <td>19.943889</td>\n",
       "      <td>0.007222</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>2019-10-26</td>\n",
       "      <td>70</td>\n",
       "      <td>2.495238</td>\n",
       "      <td>19.128056</td>\n",
       "      <td>0.000833</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>2019-10-27</td>\n",
       "      <td>17</td>\n",
       "      <td>0.758758</td>\n",
       "      <td>3.751389</td>\n",
       "      <td>0.006944</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>2019-10-28</td>\n",
       "      <td>8</td>\n",
       "      <td>1.146389</td>\n",
       "      <td>3.415278</td>\n",
       "      <td>0.030278</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>2019-10-29</td>\n",
       "      <td>11</td>\n",
       "      <td>1.283081</td>\n",
       "      <td>9.551389</td>\n",
       "      <td>0.000833</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>2019-10-30</td>\n",
       "      <td>26</td>\n",
       "      <td>1.488045</td>\n",
       "      <td>4.518889</td>\n",
       "      <td>0.000556</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>2019-10-31</td>\n",
       "      <td>14</td>\n",
       "      <td>1.688790</td>\n",
       "      <td>7.053889</td>\n",
       "      <td>0.010556</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>2019-11-01</td>\n",
       "      <td>19</td>\n",
       "      <td>1.264591</td>\n",
       "      <td>11.523611</td>\n",
       "      <td>0.019167</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>2019-11-02</td>\n",
       "      <td>10</td>\n",
       "      <td>5.820556</td>\n",
       "      <td>22.933056</td>\n",
       "      <td>0.086389</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>2019-11-03</td>\n",
       "      <td>13</td>\n",
       "      <td>1.615662</td>\n",
       "      <td>2.866389</td>\n",
       "      <td>0.278056</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>2019-11-04</td>\n",
       "      <td>9</td>\n",
       "      <td>1.565494</td>\n",
       "      <td>4.267778</td>\n",
       "      <td>0.058056</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>2019-11-05</td>\n",
       "      <td>8</td>\n",
       "      <td>2.084757</td>\n",
       "      <td>4.098611</td>\n",
       "      <td>0.011111</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>2019-11-06</td>\n",
       "      <td>5</td>\n",
       "      <td>2.598778</td>\n",
       "      <td>12.711944</td>\n",
       "      <td>0.004722</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>2019-11-07</td>\n",
       "      <td>2</td>\n",
       "      <td>0.190139</td>\n",
       "      <td>0.376389</td>\n",
       "      <td>0.003889</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "           response_time                                 \n",
       "                   count       mean        max        min\n",
       "date_x                                                   \n",
       "2019-09-28             4   2.814653   2.817778   2.812222\n",
       "2019-09-29             8   0.624653   2.896944   0.001944\n",
       "2019-09-30             5   3.786722  16.581944   0.003611\n",
       "2019-10-01             3   0.021944   0.053611   0.003333\n",
       "2019-10-02             8   1.056806   5.731667   0.001389\n",
       "2019-10-03            10   0.806972   4.398611   0.001111\n",
       "2019-10-04            11   0.610379   4.882500   0.000556\n",
       "2019-10-05             4   0.329861   1.104722   0.001667\n",
       "2019-10-06            11   2.345530   4.278611   0.001389\n",
       "2019-10-07             9   6.458796  23.520833   0.003056\n",
       "2019-10-08            16   2.372778   3.566111   0.001389\n",
       "2019-10-09             7   0.251190   0.718333   0.000556\n",
       "2019-10-10             9   1.505556   2.775000   0.202222\n",
       "2019-10-11             6   9.813565  13.176111   1.819722\n",
       "2019-10-12             7   1.625794   9.906944   0.000833\n",
       "2019-10-13             6   0.611343   2.751389   0.003611\n",
       "2019-10-14             1   0.585000   0.585000   0.585000\n",
       "2019-10-15             9   0.485000   1.286667   0.105833\n",
       "2019-10-16            12   0.892269   2.657778   0.026389\n",
       "2019-10-17             3   1.171667   2.129722   0.003889\n",
       "2019-10-18             7   2.234722  11.442500   0.000000\n",
       "2019-10-19            17   3.442925   6.890000   0.001111\n",
       "2019-10-20            11   2.098788   9.535278   0.001667\n",
       "2019-10-21             7   4.186587  23.978889   0.124444\n",
       "2019-10-22             1  22.062222  22.062222  22.062222\n",
       "2019-10-23            21   2.598201  21.864444   0.001389\n",
       "2019-10-24             9   8.162191  22.707778   0.071667\n",
       "2019-10-25            11   4.081313  19.943889   0.007222\n",
       "2019-10-26            70   2.495238  19.128056   0.000833\n",
       "2019-10-27            17   0.758758   3.751389   0.006944\n",
       "2019-10-28             8   1.146389   3.415278   0.030278\n",
       "2019-10-29            11   1.283081   9.551389   0.000833\n",
       "2019-10-30            26   1.488045   4.518889   0.000556\n",
       "2019-10-31            14   1.688790   7.053889   0.010556\n",
       "2019-11-01            19   1.264591  11.523611   0.019167\n",
       "2019-11-02            10   5.820556  22.933056   0.086389\n",
       "2019-11-03            13   1.615662   2.866389   0.278056\n",
       "2019-11-04             9   1.565494   4.267778   0.058056\n",
       "2019-11-05             8   2.084757   4.098611   0.011111\n",
       "2019-11-06             5   2.598778  12.711944   0.004722\n",
       "2019-11-07             2   0.190139   0.376389   0.003889"
      ]
     },
     "execution_count": 11,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "mean = (df_date[['date_x','response_time']]\n",
    " .groupby(df_date.date_x.dt.date)\n",
    " .agg(['count','mean', 'max','min']))\n",
    "mean.to_csv(\"fanta/moderation_delete_meantime.csv\")\n",
    "mean"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.axes._subplots.AxesSubplot at 0x7faa73cda4e0>"
      ]
     },
     "execution_count": 12,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1440x720 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Заполняем нулями пробелы\n",
    "idx = pd.date_range(mean.index[0], mean.index[-1])\n",
    "mean = mean.reindex(idx, fill_value=0)\n",
    "mean[(\"response_time\",\"mean\")].plot(figsize=(20,10)).get_figure().savefig(f\"fanta/moderation_delete_meantime.png\")\n",
    "mean[(\"response_time\",\"mean\")].plot(figsize=(20,10))\n",
    "# df_stat[df_stat.index.get_level_values(1)==\"wall_reply_new\"].plot(figsize=(20,10))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Расчет времени до ответа на комментарий"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "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>reply_id</th>\n",
       "      <th>date_x</th>\n",
       "      <th>from_id</th>\n",
       "      <th>date_y</th>\n",
       "      <th>response_time</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <td>0</td>\n",
       "      <td>263524</td>\n",
       "      <td>2019-09-29 08:45:14.612536</td>\n",
       "      <td>-24539231</td>\n",
       "      <td>2019-09-29 10:25:08</td>\n",
       "      <td>1.664722</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>1</td>\n",
       "      <td>263527</td>\n",
       "      <td>2019-09-29 08:51:49.795747</td>\n",
       "      <td>-24539231</td>\n",
       "      <td>2019-09-29 10:26:03</td>\n",
       "      <td>1.570278</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>2</td>\n",
       "      <td>263528</td>\n",
       "      <td>2019-09-29 09:24:53.139219</td>\n",
       "      <td>-24539231</td>\n",
       "      <td>2019-09-29 10:23:35</td>\n",
       "      <td>0.978056</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>3</td>\n",
       "      <td>263534</td>\n",
       "      <td>2019-09-29 10:54:59.718837</td>\n",
       "      <td>-24539231</td>\n",
       "      <td>2019-09-29 12:02:26</td>\n",
       "      <td>1.123889</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>4</td>\n",
       "      <td>263539</td>\n",
       "      <td>2019-09-29 13:48:31.923808</td>\n",
       "      <td>-24539231</td>\n",
       "      <td>2019-09-29 15:24:17</td>\n",
       "      <td>1.595833</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>329</td>\n",
       "      <td>265814</td>\n",
       "      <td>2019-11-06 17:13:17.665533</td>\n",
       "      <td>-24539231</td>\n",
       "      <td>2019-11-06 19:30:54</td>\n",
       "      <td>2.293333</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>330</td>\n",
       "      <td>265816</td>\n",
       "      <td>2019-11-06 17:50:55.760159</td>\n",
       "      <td>-24539231</td>\n",
       "      <td>2019-11-06 18:29:43</td>\n",
       "      <td>0.646389</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>331</td>\n",
       "      <td>265818</td>\n",
       "      <td>2019-11-06 18:25:38.003400</td>\n",
       "      <td>-24539231</td>\n",
       "      <td>2019-11-06 18:29:12</td>\n",
       "      <td>0.059167</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>332</td>\n",
       "      <td>265825</td>\n",
       "      <td>2019-11-06 20:36:41.177975</td>\n",
       "      <td>-24539231</td>\n",
       "      <td>2019-11-06 20:50:36</td>\n",
       "      <td>0.231667</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>333</td>\n",
       "      <td>265834</td>\n",
       "      <td>2019-11-07 13:20:14.999928</td>\n",
       "      <td>-24539231</td>\n",
       "      <td>2019-11-07 14:09:07</td>\n",
       "      <td>0.814444</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>334 rows × 5 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "     reply_id                     date_x   from_id              date_y  \\\n",
       "0      263524 2019-09-29 08:45:14.612536 -24539231 2019-09-29 10:25:08   \n",
       "1      263527 2019-09-29 08:51:49.795747 -24539231 2019-09-29 10:26:03   \n",
       "2      263528 2019-09-29 09:24:53.139219 -24539231 2019-09-29 10:23:35   \n",
       "3      263534 2019-09-29 10:54:59.718837 -24539231 2019-09-29 12:02:26   \n",
       "4      263539 2019-09-29 13:48:31.923808 -24539231 2019-09-29 15:24:17   \n",
       "..        ...                        ...       ...                 ...   \n",
       "329    265814 2019-11-06 17:13:17.665533 -24539231 2019-11-06 19:30:54   \n",
       "330    265816 2019-11-06 17:50:55.760159 -24539231 2019-11-06 18:29:43   \n",
       "331    265818 2019-11-06 18:25:38.003400 -24539231 2019-11-06 18:29:12   \n",
       "332    265825 2019-11-06 20:36:41.177975 -24539231 2019-11-06 20:50:36   \n",
       "333    265834 2019-11-07 13:20:14.999928 -24539231 2019-11-07 14:09:07   \n",
       "\n",
       "     response_time  \n",
       "0         1.664722  \n",
       "1         1.570278  \n",
       "2         0.978056  \n",
       "3         1.123889  \n",
       "4         1.595833  \n",
       "..             ...  \n",
       "329       2.293333  \n",
       "330       0.646389  \n",
       "331       0.059167  \n",
       "332       0.231667  \n",
       "333       0.814444  \n",
       "\n",
       "[334 rows x 5 columns]"
      ]
     },
     "execution_count": 13,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Извлекам отправителя сообщения\n",
    "df[\"from_id\"] = df[\"object\"].apply(lambda x: x.get('from_id',0))\n",
    "# Фильтруем только сообщения от модератора, забираем дату и id сообщения на которое дан ответ \n",
    "df_rep = df[df.from_id == group_id][\"object\"].apply(pd.Series)[[\"from_id\",\"date\",\"reply_to_comment\"]]\n",
    "#  преаброзуем unixtime в datetime\n",
    "df_rep[\"date\"] = pd.to_datetime(df_rep[\"date\"], unit='s')\n",
    "# приводим индекс ответа к к reply_id \n",
    "df_rep.rename(columns={\"reply_to_comment\":\"reply_id\"}, inplace=True)\n",
    "# Ищем сообщения с указанным в ответе reply_id\n",
    "df_date = pd.merge(df_new, df_rep, on='reply_id')\n",
    "# Расчитываем diff от момента создания до момоента ответа\n",
    "df_date['response_time'] = (df_date.date_y - df_date.date_x).dt.seconds/60.0/60.0\n",
    "# Сохраняем\n",
    "df_date.to_csv(\"fanta/moderation_response_stat.csv\", index=None)\n",
    "# выводим\n",
    "df_date\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "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 tr th {\n",
       "        text-align: left;\n",
       "    }\n",
       "\n",
       "    .dataframe thead tr:last-of-type th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr>\n",
       "      <th></th>\n",
       "      <th colspan=\"4\" halign=\"left\">response_time</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th></th>\n",
       "      <th>count</th>\n",
       "      <th>mean</th>\n",
       "      <th>max</th>\n",
       "      <th>min</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>date_x</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <td>2019-09-29</td>\n",
       "      <td>9</td>\n",
       "      <td>2.910309</td>\n",
       "      <td>17.188611</td>\n",
       "      <td>0.596111</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>2019-09-30</td>\n",
       "      <td>4</td>\n",
       "      <td>7.928750</td>\n",
       "      <td>21.863611</td>\n",
       "      <td>1.720556</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>2019-10-01</td>\n",
       "      <td>8</td>\n",
       "      <td>3.775625</td>\n",
       "      <td>10.535556</td>\n",
       "      <td>0.540556</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>2019-10-02</td>\n",
       "      <td>10</td>\n",
       "      <td>0.731167</td>\n",
       "      <td>3.130556</td>\n",
       "      <td>0.079167</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>2019-10-03</td>\n",
       "      <td>9</td>\n",
       "      <td>2.282716</td>\n",
       "      <td>4.780278</td>\n",
       "      <td>0.258056</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>2019-10-04</td>\n",
       "      <td>4</td>\n",
       "      <td>1.195833</td>\n",
       "      <td>2.793333</td>\n",
       "      <td>0.663333</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>2019-10-05</td>\n",
       "      <td>3</td>\n",
       "      <td>4.555556</td>\n",
       "      <td>10.355833</td>\n",
       "      <td>1.148333</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>2019-10-06</td>\n",
       "      <td>3</td>\n",
       "      <td>0.960185</td>\n",
       "      <td>1.357500</td>\n",
       "      <td>0.611667</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>2019-10-07</td>\n",
       "      <td>12</td>\n",
       "      <td>2.043449</td>\n",
       "      <td>11.253889</td>\n",
       "      <td>0.099444</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>2019-10-08</td>\n",
       "      <td>8</td>\n",
       "      <td>5.706424</td>\n",
       "      <td>18.895833</td>\n",
       "      <td>0.137778</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>2019-10-09</td>\n",
       "      <td>10</td>\n",
       "      <td>4.489083</td>\n",
       "      <td>22.314444</td>\n",
       "      <td>0.030833</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>2019-10-10</td>\n",
       "      <td>5</td>\n",
       "      <td>1.711944</td>\n",
       "      <td>4.049444</td>\n",
       "      <td>0.316389</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>2019-10-11</td>\n",
       "      <td>5</td>\n",
       "      <td>15.215444</td>\n",
       "      <td>22.857222</td>\n",
       "      <td>11.139444</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>2019-10-12</td>\n",
       "      <td>14</td>\n",
       "      <td>1.230159</td>\n",
       "      <td>2.660000</td>\n",
       "      <td>0.278889</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>2019-10-13</td>\n",
       "      <td>14</td>\n",
       "      <td>2.123790</td>\n",
       "      <td>11.101389</td>\n",
       "      <td>0.361667</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>2019-10-14</td>\n",
       "      <td>4</td>\n",
       "      <td>4.590694</td>\n",
       "      <td>14.281944</td>\n",
       "      <td>0.051389</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>2019-10-15</td>\n",
       "      <td>4</td>\n",
       "      <td>0.550972</td>\n",
       "      <td>1.141944</td>\n",
       "      <td>0.197500</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>2019-10-16</td>\n",
       "      <td>3</td>\n",
       "      <td>1.246296</td>\n",
       "      <td>3.079444</td>\n",
       "      <td>0.286111</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>2019-10-17</td>\n",
       "      <td>6</td>\n",
       "      <td>3.675417</td>\n",
       "      <td>9.498333</td>\n",
       "      <td>1.882500</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>2019-10-18</td>\n",
       "      <td>3</td>\n",
       "      <td>1.045741</td>\n",
       "      <td>1.967778</td>\n",
       "      <td>0.480278</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>2019-10-19</td>\n",
       "      <td>4</td>\n",
       "      <td>4.090694</td>\n",
       "      <td>5.430833</td>\n",
       "      <td>1.689167</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>2019-10-20</td>\n",
       "      <td>3</td>\n",
       "      <td>5.992685</td>\n",
       "      <td>12.615278</td>\n",
       "      <td>1.010278</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>2019-10-21</td>\n",
       "      <td>1</td>\n",
       "      <td>22.217222</td>\n",
       "      <td>22.217222</td>\n",
       "      <td>22.217222</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>2019-10-22</td>\n",
       "      <td>2</td>\n",
       "      <td>2.897083</td>\n",
       "      <td>3.923333</td>\n",
       "      <td>1.870833</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>2019-10-23</td>\n",
       "      <td>4</td>\n",
       "      <td>9.747222</td>\n",
       "      <td>14.165833</td>\n",
       "      <td>0.670278</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>2019-10-24</td>\n",
       "      <td>4</td>\n",
       "      <td>5.723125</td>\n",
       "      <td>17.348889</td>\n",
       "      <td>0.565000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>2019-10-25</td>\n",
       "      <td>14</td>\n",
       "      <td>4.470496</td>\n",
       "      <td>8.549167</td>\n",
       "      <td>0.311944</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>2019-10-26</td>\n",
       "      <td>26</td>\n",
       "      <td>1.671667</td>\n",
       "      <td>5.871944</td>\n",
       "      <td>0.078333</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>2019-10-27</td>\n",
       "      <td>25</td>\n",
       "      <td>0.678956</td>\n",
       "      <td>2.994444</td>\n",
       "      <td>0.016944</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>2019-10-28</td>\n",
       "      <td>20</td>\n",
       "      <td>2.093875</td>\n",
       "      <td>8.240278</td>\n",
       "      <td>0.138611</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>2019-10-29</td>\n",
       "      <td>17</td>\n",
       "      <td>2.917941</td>\n",
       "      <td>10.130833</td>\n",
       "      <td>0.085556</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>2019-10-30</td>\n",
       "      <td>12</td>\n",
       "      <td>1.352755</td>\n",
       "      <td>6.969444</td>\n",
       "      <td>0.147500</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>2019-10-31</td>\n",
       "      <td>18</td>\n",
       "      <td>0.914306</td>\n",
       "      <td>4.248056</td>\n",
       "      <td>0.025000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>2019-11-01</td>\n",
       "      <td>5</td>\n",
       "      <td>1.802500</td>\n",
       "      <td>2.660278</td>\n",
       "      <td>0.990556</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>2019-11-02</td>\n",
       "      <td>8</td>\n",
       "      <td>2.955972</td>\n",
       "      <td>8.075833</td>\n",
       "      <td>1.562500</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>2019-11-03</td>\n",
       "      <td>5</td>\n",
       "      <td>1.096833</td>\n",
       "      <td>1.841389</td>\n",
       "      <td>0.359167</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>2019-11-04</td>\n",
       "      <td>10</td>\n",
       "      <td>2.193056</td>\n",
       "      <td>5.026944</td>\n",
       "      <td>0.135000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>2019-11-05</td>\n",
       "      <td>12</td>\n",
       "      <td>1.447778</td>\n",
       "      <td>4.158611</td>\n",
       "      <td>0.016389</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>2019-11-06</td>\n",
       "      <td>5</td>\n",
       "      <td>1.007056</td>\n",
       "      <td>2.293333</td>\n",
       "      <td>0.059167</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>2019-11-07</td>\n",
       "      <td>1</td>\n",
       "      <td>0.814444</td>\n",
       "      <td>0.814444</td>\n",
       "      <td>0.814444</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "           response_time                                 \n",
       "                   count       mean        max        min\n",
       "date_x                                                   \n",
       "2019-09-29             9   2.910309  17.188611   0.596111\n",
       "2019-09-30             4   7.928750  21.863611   1.720556\n",
       "2019-10-01             8   3.775625  10.535556   0.540556\n",
       "2019-10-02            10   0.731167   3.130556   0.079167\n",
       "2019-10-03             9   2.282716   4.780278   0.258056\n",
       "2019-10-04             4   1.195833   2.793333   0.663333\n",
       "2019-10-05             3   4.555556  10.355833   1.148333\n",
       "2019-10-06             3   0.960185   1.357500   0.611667\n",
       "2019-10-07            12   2.043449  11.253889   0.099444\n",
       "2019-10-08             8   5.706424  18.895833   0.137778\n",
       "2019-10-09            10   4.489083  22.314444   0.030833\n",
       "2019-10-10             5   1.711944   4.049444   0.316389\n",
       "2019-10-11             5  15.215444  22.857222  11.139444\n",
       "2019-10-12            14   1.230159   2.660000   0.278889\n",
       "2019-10-13            14   2.123790  11.101389   0.361667\n",
       "2019-10-14             4   4.590694  14.281944   0.051389\n",
       "2019-10-15             4   0.550972   1.141944   0.197500\n",
       "2019-10-16             3   1.246296   3.079444   0.286111\n",
       "2019-10-17             6   3.675417   9.498333   1.882500\n",
       "2019-10-18             3   1.045741   1.967778   0.480278\n",
       "2019-10-19             4   4.090694   5.430833   1.689167\n",
       "2019-10-20             3   5.992685  12.615278   1.010278\n",
       "2019-10-21             1  22.217222  22.217222  22.217222\n",
       "2019-10-22             2   2.897083   3.923333   1.870833\n",
       "2019-10-23             4   9.747222  14.165833   0.670278\n",
       "2019-10-24             4   5.723125  17.348889   0.565000\n",
       "2019-10-25            14   4.470496   8.549167   0.311944\n",
       "2019-10-26            26   1.671667   5.871944   0.078333\n",
       "2019-10-27            25   0.678956   2.994444   0.016944\n",
       "2019-10-28            20   2.093875   8.240278   0.138611\n",
       "2019-10-29            17   2.917941  10.130833   0.085556\n",
       "2019-10-30            12   1.352755   6.969444   0.147500\n",
       "2019-10-31            18   0.914306   4.248056   0.025000\n",
       "2019-11-01             5   1.802500   2.660278   0.990556\n",
       "2019-11-02             8   2.955972   8.075833   1.562500\n",
       "2019-11-03             5   1.096833   1.841389   0.359167\n",
       "2019-11-04            10   2.193056   5.026944   0.135000\n",
       "2019-11-05            12   1.447778   4.158611   0.016389\n",
       "2019-11-06             5   1.007056   2.293333   0.059167\n",
       "2019-11-07             1   0.814444   0.814444   0.814444"
      ]
     },
     "execution_count": 14,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "mean = (df_date[['date_x','response_time']]\n",
    " .groupby(df_date.date_x.dt.date)\n",
    " .agg(['count','mean', 'max','min']))\n",
    "mean.to_csv(\"fanta/moderation_response_meantime.csv\")\n",
    "mean"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.axes._subplots.AxesSubplot at 0x7faa7a95f710>"
      ]
     },
     "execution_count": 15,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1440x720 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Заполняем нулями пробелы\n",
    "idx = pd.date_range(mean.index[0], mean.index[-1])\n",
    "mean = mean.reindex(idx, fill_value=0)\n",
    "mean[(\"response_time\",\"mean\")].plot(figsize=(20,10)).get_figure().savefig(f\"fanta/moderation_response_meantime.png\")\n",
    "mean[(\"response_time\",\"mean\")].plot(figsize=(20,10))\n",
    "# df_stat[df_stat.index.get_level_values(1)==\"wall_reply_new\"].plot(figsize=(20,10))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Архивация отчета"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "bash: line 3: syntax error near unexpected token `('\n",
      "bash: line 3: `zip -9 -y -r -q fanta.(date +%F).zip fanta/'\n"
     ]
    },
    {
     "ename": "CalledProcessError",
     "evalue": "Command 'b'cp fanta.ipynb fanta/\\nrm fanta.zip\\nzip -9 -y -r -q fanta.(date +%F).zip fanta/\\n'' returned non-zero exit status 2.",
     "output_type": "error",
     "traceback": [
      "\u001b[0;31m---------------------------------------------------------------------------\u001b[0m",
      "\u001b[0;31mCalledProcessError\u001b[0m                        Traceback (most recent call last)",
      "\u001b[0;32m<ipython-input-17-9916e9896fc8>\u001b[0m in \u001b[0;36m<module>\u001b[0;34m\u001b[0m\n\u001b[0;32m----> 1\u001b[0;31m \u001b[0mget_ipython\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mrun_cell_magic\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m'bash'\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0;34m''\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0;34m'cp fanta.ipynb fanta/\\nrm fanta.zip\\nzip -9 -y -r -q fanta.(date +%F).zip fanta/\\n'\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m",
      "\u001b[0;32m~/lab/jupyter/venv/lib/python3.6/site-packages/IPython/core/interactiveshell.py\u001b[0m in \u001b[0;36mrun_cell_magic\u001b[0;34m(self, magic_name, line, cell)\u001b[0m\n\u001b[1;32m   2357\u001b[0m             \u001b[0;32mwith\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mbuiltin_trap\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m   2358\u001b[0m                 \u001b[0margs\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0;34m(\u001b[0m\u001b[0mmagic_arg_s\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mcell\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m-> 2359\u001b[0;31m                 \u001b[0mresult\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mfn\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m*\u001b[0m\u001b[0margs\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0;34m**\u001b[0m\u001b[0mkwargs\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m   2360\u001b[0m             \u001b[0;32mreturn\u001b[0m \u001b[0mresult\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m   2361\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n",
      "\u001b[0;32m~/lab/jupyter/venv/lib/python3.6/site-packages/IPython/core/magics/script.py\u001b[0m in \u001b[0;36mnamed_script_magic\u001b[0;34m(line, cell)\u001b[0m\n\u001b[1;32m    140\u001b[0m             \u001b[0;32melse\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    141\u001b[0m                 \u001b[0mline\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mscript\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 142\u001b[0;31m             \u001b[0;32mreturn\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mshebang\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mline\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mcell\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m    143\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    144\u001b[0m         \u001b[0;31m# write a basic docstring:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
      "\u001b[0;32m</home/v.seregin/lab/jupyter/venv/lib/python3.6/site-packages/decorator.py:decorator-gen-110>\u001b[0m in \u001b[0;36mshebang\u001b[0;34m(self, line, cell)\u001b[0m\n",
      "\u001b[0;32m~/lab/jupyter/venv/lib/python3.6/site-packages/IPython/core/magic.py\u001b[0m in \u001b[0;36m<lambda>\u001b[0;34m(f, *a, **k)\u001b[0m\n\u001b[1;32m    185\u001b[0m     \u001b[0;31m# but it's overkill for just that one bit of state.\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    186\u001b[0m     \u001b[0;32mdef\u001b[0m \u001b[0mmagic_deco\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0marg\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 187\u001b[0;31m         \u001b[0mcall\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0;32mlambda\u001b[0m \u001b[0mf\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0;34m*\u001b[0m\u001b[0ma\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0;34m**\u001b[0m\u001b[0mk\u001b[0m\u001b[0;34m:\u001b[0m \u001b[0mf\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m*\u001b[0m\u001b[0ma\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0;34m**\u001b[0m\u001b[0mk\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m    188\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    189\u001b[0m         \u001b[0;32mif\u001b[0m \u001b[0mcallable\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0marg\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
      "\u001b[0;32m~/lab/jupyter/venv/lib/python3.6/site-packages/IPython/core/magics/script.py\u001b[0m in \u001b[0;36mshebang\u001b[0;34m(self, line, cell)\u001b[0m\n\u001b[1;32m    243\u001b[0m             \u001b[0msys\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mstderr\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mflush\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    244\u001b[0m         \u001b[0;32mif\u001b[0m \u001b[0margs\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mraise_error\u001b[0m \u001b[0;32mand\u001b[0m \u001b[0mp\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mreturncode\u001b[0m\u001b[0;34m!=\u001b[0m\u001b[0;36m0\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 245\u001b[0;31m             \u001b[0;32mraise\u001b[0m \u001b[0mCalledProcessError\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mp\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mreturncode\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mcell\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0moutput\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0mout\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mstderr\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0merr\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m    246\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    247\u001b[0m     \u001b[0;32mdef\u001b[0m \u001b[0m_run_script\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mself\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mp\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mcell\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mto_close\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
      "\u001b[0;31mCalledProcessError\u001b[0m: Command 'b'cp fanta.ipynb fanta/\\nrm fanta.zip\\nzip -9 -y -r -q fanta.(date +%F).zip fanta/\\n'' returned non-zero exit status 2."
     ]
    }
   ],
   "source": [
    "%%bash\n",
    "cp fanta.ipynb fanta/\n",
    "rm fanta.zip\n",
    "zip -9 -y -r -q fanta.(date +%F).zip fanta/"
   ]
  },
  {
   "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": "3.6.8"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 4
}