{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "import sys\n",
    "sys.path.insert(0, '/Users/joel/src/thundr/tracker')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "No buildsha.py defined. sha will be 'NOT_SET'\n",
      "Not using secrets manager\n",
      "use_my_whitelist_cache False\n"
     ]
    }
   ],
   "source": [
    "from tracker.unicorn import mysql"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [],
   "source": [
    "import pandas as pd, numpy as np"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [],
   "source": [
    "from importlib import reload"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<module 'tracker.unicorn.mysql' from '/Users/joel/src/thundr/tracker/tracker/unicorn/mysql.py'>"
      ]
     },
     "execution_count": 5,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "reload(mysql)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 34,
   "metadata": {},
   "outputs": [],
   "source": [
    "reasons_listeners = mysql.sql_to_dataframe('''select * from before_i_go_listeners where ts > '2021-01-31' ''', columns=['ts', 'listeners'], index_column='ts', type_map={'ts': np.datetime64})"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 71,
   "metadata": {},
   "outputs": [],
   "source": [
    "before_i_go_listeners = mysql.sql_to_dataframe('''select * from before_i_go_listeners where ts < '2020-07-31' ''', columns=['ts', 'listeners'], index_column='ts', type_map={'ts': np.datetime64})"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {},
   "outputs": [],
   "source": [
    "def to_day(dt):\n",
    "    return dt.date()\n",
    "\n",
    "def to_minutes(dt):\n",
    "    \"\"\"Return floating point number of hours through the day in `datetime` dt.\"\"\"\n",
    "    return dt.hour*60 + dt.minute + dt.second / 60"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 72,
   "metadata": {},
   "outputs": [],
   "source": [
    "listeners1Min = before_i_go_listeners.resample('1Min').mean()\n",
    "listeners1Min['dday'] = listeners1Min.index.map(to_day)\n",
    "listeners1Min['tday'] = listeners1Min.index.map(to_minutes).astype(int)\n",
    "avg_listeners_daily = listeners1Min.pivot(index='tday', columns='dday', values='listeners').mean()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 73,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "dday\n",
       "2020-04-16      5.077377\n",
       "2020-04-17     18.198834\n",
       "2020-04-18      9.405115\n",
       "2020-04-19      9.062851\n",
       "2020-04-20     13.771029\n",
       "                 ...    \n",
       "2020-07-18     94.510199\n",
       "2020-07-19     93.824698\n",
       "2020-07-20     93.987562\n",
       "2020-07-21    100.145638\n",
       "2020-07-22     94.202037\n",
       "Length: 98, dtype: float64"
      ]
     },
     "execution_count": 73,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "avg_listeners_daily"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 74,
   "metadata": {},
   "outputs": [],
   "source": [
    "daily_streams = mysql.sql_to_dataframe('''select * from before_i_go_daily_streams where as_of <= '2020-07-31' ''', columns=['as_of', 'listeners'], index_column='as_of')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 75,
   "metadata": {},
   "outputs": [],
   "source": [
    "df1 = avg_listeners_daily.to_frame('listeners')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 76,
   "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>listeners</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>dday</th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>2020-04-16</th>\n",
       "      <td>5.077377</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2020-04-17</th>\n",
       "      <td>18.198834</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2020-04-18</th>\n",
       "      <td>9.405115</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2020-04-19</th>\n",
       "      <td>9.062851</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2020-04-20</th>\n",
       "      <td>13.771029</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2020-07-18</th>\n",
       "      <td>94.510199</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2020-07-19</th>\n",
       "      <td>93.824698</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2020-07-20</th>\n",
       "      <td>93.987562</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2020-07-21</th>\n",
       "      <td>100.145638</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2020-07-22</th>\n",
       "      <td>94.202037</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>98 rows × 1 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "             listeners\n",
       "dday                  \n",
       "2020-04-16    5.077377\n",
       "2020-04-17   18.198834\n",
       "2020-04-18    9.405115\n",
       "2020-04-19    9.062851\n",
       "2020-04-20   13.771029\n",
       "...                ...\n",
       "2020-07-18   94.510199\n",
       "2020-07-19   93.824698\n",
       "2020-07-20   93.987562\n",
       "2020-07-21  100.145638\n",
       "2020-07-22   94.202037\n",
       "\n",
       "[98 rows x 1 columns]"
      ]
     },
     "execution_count": 76,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df1"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 82,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<AxesSubplot:xlabel='as_of'>"
      ]
     },
     "execution_count": 82,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 864x360 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "daily_streams.diff(1).plot(figsize=(12,5))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 77,
   "metadata": {},
   "outputs": [],
   "source": [
    "merged = pd.merge(avg_listeners_daily.to_frame('listeners'), daily_streams, left_index=True, right_index=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 78,
   "metadata": {},
   "outputs": [],
   "source": [
    "box = merged.diff(1).set_index('listeners').sort_index()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 79,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<AxesSubplot:xlabel='listeners'>"
      ]
     },
     "execution_count": 79,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "box.plot()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "daily_streams = mysql.sql_to_dataframe('''select * from before_i_go_daily_streams where as_of <= '2020-07-31' ''', columns=['as_of', 'listeners'], index_column='as_of')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 84,
   "metadata": {},
   "outputs": [],
   "source": [
    "daily_streams = mysql.sql_to_dataframe('''select * from break_my_heart_daily_streams''', columns=['as_of', 'listeners'], index_column='as_of')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 85,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<AxesSubplot:xlabel='as_of'>"
      ]
     },
     "execution_count": 85,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "daily_streams.plot()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 86,
   "metadata": {},
   "outputs": [
    {
     "ename": "ModuleNotFoundError",
     "evalue": "No module named 'unicapi'",
     "output_type": "error",
     "traceback": [
      "\u001b[0;31m-------------------------------------------------------------------\u001b[0m",
      "\u001b[0;31mModuleNotFoundError\u001b[0m               Traceback (most recent call last)",
      "\u001b[0;32m<ipython-input-86-00d54f131e03>\u001b[0m in \u001b[0;36m<module>\u001b[0;34m\u001b[0m\n\u001b[0;32m----> 1\u001b[0;31m \u001b[0;32mfrom\u001b[0m \u001b[0municapi\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mapi\u001b[0m \u001b[0;32mimport\u001b[0m \u001b[0mtracks\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m",
      "\u001b[0;31mModuleNotFoundError\u001b[0m: No module named 'unicapi'"
     ]
    }
   ],
   "source": [
    "from unicapi.api import tracks"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 50,
   "metadata": {},
   "outputs": [],
   "source": [
    "def create_slope_dataframe(df, colname):\n",
    "    # df should be indexed on time\n",
    "    slopes = (df[colname].diff() / df.index.to_frame(name='time').diff().time.dt.seconds)\n",
    "    slopes = slopes.to_frame(colname)\n",
    "    slopes[colname] = slopes[colname] * 3600\n",
    "    slopes['ts'] = (slopes.index.astype(int) / 1e6).astype(int)\n",
    "    return slopes\n",
    "\n",
    "dfyt = mysql.sql_to_dataframe(\n",
    "        f\"\"\"select ts, views1 + views2 + views3 as value from before_i_go_youtubes\"\"\",\n",
    "        columns=['ts', 'value'],\n",
    "        index_column='ts'\n",
    "    )\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 55,
   "metadata": {},
   "outputs": [],
   "source": [
    "dfytm = create_slope_dataframe(dfyt, 'value')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 67,
   "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>value</th>\n",
       "      <th>ts</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>ts</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>2020-04-17 04:07:06</th>\n",
       "      <td>NaN</td>\n",
       "      <td>1587096426000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2020-04-17 04:27:06</th>\n",
       "      <td>6.0</td>\n",
       "      <td>1587097626000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2020-04-17 04:47:06</th>\n",
       "      <td>0.0</td>\n",
       "      <td>1587098826000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2020-04-17 05:07:06</th>\n",
       "      <td>3.0</td>\n",
       "      <td>1587100026000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2020-04-17 05:27:06</th>\n",
       "      <td>6.0</td>\n",
       "      <td>1587101226000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2020-05-01 14:50:30</th>\n",
       "      <td>93.0</td>\n",
       "      <td>1588344630000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2020-05-01 15:10:30</th>\n",
       "      <td>822.0</td>\n",
       "      <td>1588345830000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2020-05-01 15:30:30</th>\n",
       "      <td>270.0</td>\n",
       "      <td>1588347030000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2020-05-01 15:50:30</th>\n",
       "      <td>726.0</td>\n",
       "      <td>1588348230000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2020-05-01 16:10:30</th>\n",
       "      <td>297.0</td>\n",
       "      <td>1588349430000</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>988 rows × 2 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "                     value             ts\n",
       "ts                                       \n",
       "2020-04-17 04:07:06    NaN  1587096426000\n",
       "2020-04-17 04:27:06    6.0  1587097626000\n",
       "2020-04-17 04:47:06    0.0  1587098826000\n",
       "2020-04-17 05:07:06    3.0  1587100026000\n",
       "2020-04-17 05:27:06    6.0  1587101226000\n",
       "...                    ...            ...\n",
       "2020-05-01 14:50:30   93.0  1588344630000\n",
       "2020-05-01 15:10:30  822.0  1588345830000\n",
       "2020-05-01 15:30:30  270.0  1588347030000\n",
       "2020-05-01 15:50:30  726.0  1588348230000\n",
       "2020-05-01 16:10:30  297.0  1588349430000\n",
       "\n",
       "[988 rows x 2 columns]"
      ]
     },
     "execution_count": 67,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "dfytm"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 68,
   "metadata": {},
   "outputs": [],
   "source": [
    "df9 = mysql.sql_to_dataframe('select * from before_i_go_daily_streams')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 72,
   "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>as_of</th>\n",
       "      <th>streams</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>as_of</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>2020-03-21</th>\n",
       "      <td>2020-03-21</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2020-03-22</th>\n",
       "      <td>2020-03-22</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2020-03-23</th>\n",
       "      <td>2020-03-23</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2020-03-24</th>\n",
       "      <td>2020-03-24</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2020-03-25</th>\n",
       "      <td>2020-03-25</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2020-03-26</th>\n",
       "      <td>2020-03-26</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2020-03-27</th>\n",
       "      <td>2020-03-27</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2020-03-28</th>\n",
       "      <td>2020-03-28</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2020-03-29</th>\n",
       "      <td>2020-03-29</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2020-03-30</th>\n",
       "      <td>2020-03-30</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2020-03-31</th>\n",
       "      <td>2020-03-31</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2020-04-01</th>\n",
       "      <td>2020-04-01</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2020-04-02</th>\n",
       "      <td>2020-04-02</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2020-04-03</th>\n",
       "      <td>2020-04-03</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2020-04-04</th>\n",
       "      <td>2020-04-04</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2020-04-05</th>\n",
       "      <td>2020-04-05</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2020-04-06</th>\n",
       "      <td>2020-04-06</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2020-04-07</th>\n",
       "      <td>2020-04-07</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2020-04-08</th>\n",
       "      <td>2020-04-08</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2020-04-09</th>\n",
       "      <td>2020-04-09</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2020-04-10</th>\n",
       "      <td>2020-04-10</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2020-04-11</th>\n",
       "      <td>2020-04-11</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2020-04-12</th>\n",
       "      <td>2020-04-12</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2020-04-13</th>\n",
       "      <td>2020-04-13</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2020-04-14</th>\n",
       "      <td>2020-04-14</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2020-04-15</th>\n",
       "      <td>2020-04-15</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2020-04-16</th>\n",
       "      <td>2020-04-16</td>\n",
       "      <td>607</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2020-04-17</th>\n",
       "      <td>2020-04-17</td>\n",
       "      <td>6492</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2020-04-18</th>\n",
       "      <td>2020-04-18</td>\n",
       "      <td>3337</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2020-04-19</th>\n",
       "      <td>2020-04-19</td>\n",
       "      <td>3232</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2020-04-20</th>\n",
       "      <td>2020-04-20</td>\n",
       "      <td>4937</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2020-04-21</th>\n",
       "      <td>2020-04-21</td>\n",
       "      <td>3994</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2020-04-22</th>\n",
       "      <td>2020-04-22</td>\n",
       "      <td>3687</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2020-04-23</th>\n",
       "      <td>2020-04-23</td>\n",
       "      <td>4285</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2020-04-24</th>\n",
       "      <td>2020-04-24</td>\n",
       "      <td>3733</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2020-04-25</th>\n",
       "      <td>2020-04-25</td>\n",
       "      <td>4502</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2020-04-26</th>\n",
       "      <td>2020-04-26</td>\n",
       "      <td>14196</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2020-04-27</th>\n",
       "      <td>2020-04-27</td>\n",
       "      <td>17451</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2020-04-28</th>\n",
       "      <td>2020-04-28</td>\n",
       "      <td>16737</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2020-04-29</th>\n",
       "      <td>2020-04-29</td>\n",
       "      <td>16726</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2020-04-30</th>\n",
       "      <td>2020-04-30</td>\n",
       "      <td>22647</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2020-05-01</th>\n",
       "      <td>2020-05-01</td>\n",
       "      <td>40902</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2020-05-02</th>\n",
       "      <td>2020-05-02</td>\n",
       "      <td>39542</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2020-05-03</th>\n",
       "      <td>2020-05-03</td>\n",
       "      <td>38058</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                 as_of  streams\n",
       "as_of                          \n",
       "2020-03-21  2020-03-21        0\n",
       "2020-03-22  2020-03-22        0\n",
       "2020-03-23  2020-03-23        0\n",
       "2020-03-24  2020-03-24        0\n",
       "2020-03-25  2020-03-25        0\n",
       "2020-03-26  2020-03-26        0\n",
       "2020-03-27  2020-03-27        0\n",
       "2020-03-28  2020-03-28        0\n",
       "2020-03-29  2020-03-29        0\n",
       "2020-03-30  2020-03-30        0\n",
       "2020-03-31  2020-03-31        0\n",
       "2020-04-01  2020-04-01        0\n",
       "2020-04-02  2020-04-02        0\n",
       "2020-04-03  2020-04-03        0\n",
       "2020-04-04  2020-04-04        0\n",
       "2020-04-05  2020-04-05        0\n",
       "2020-04-06  2020-04-06        0\n",
       "2020-04-07  2020-04-07        0\n",
       "2020-04-08  2020-04-08        0\n",
       "2020-04-09  2020-04-09        0\n",
       "2020-04-10  2020-04-10        0\n",
       "2020-04-11  2020-04-11        0\n",
       "2020-04-12  2020-04-12        0\n",
       "2020-04-13  2020-04-13        0\n",
       "2020-04-14  2020-04-14        0\n",
       "2020-04-15  2020-04-15        0\n",
       "2020-04-16  2020-04-16      607\n",
       "2020-04-17  2020-04-17     6492\n",
       "2020-04-18  2020-04-18     3337\n",
       "2020-04-19  2020-04-19     3232\n",
       "2020-04-20  2020-04-20     4937\n",
       "2020-04-21  2020-04-21     3994\n",
       "2020-04-22  2020-04-22     3687\n",
       "2020-04-23  2020-04-23     4285\n",
       "2020-04-24  2020-04-24     3733\n",
       "2020-04-25  2020-04-25     4502\n",
       "2020-04-26  2020-04-26    14196\n",
       "2020-04-27  2020-04-27    17451\n",
       "2020-04-28  2020-04-28    16737\n",
       "2020-04-29  2020-04-29    16726\n",
       "2020-04-30  2020-04-30    22647\n",
       "2020-05-01  2020-05-01    40902\n",
       "2020-05-02  2020-05-02    39542\n",
       "2020-05-03  2020-05-03    38058"
      ]
     },
     "execution_count": 72,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df9.set_index(df9.as_of.astype(np.datetime64))"
   ]
  },
  {
   "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": "3.7.1"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}