plot-throughput-difference-audio-video-timeseries.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "import sqlite3\n",
    "import pandas as pd\n",
    "import filterbyip as fbi\n",
    "import clusterbycolumn as cbc\n",
    "import requests\n",
    "import asinfo\n",
    "\n",
    "import matplotlib.dates as dates\n",
    "from datetime import datetime as dt"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Populating the interactive namespace from numpy and matplotlib\n"
     ]
    }
   ],
   "source": [
    "import warnings\n",
    "warnings.filterwarnings('ignore')\n",
    "\n",
    "%pylab inline"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<style>body {\n",
       "    margin: 0;\n",
       "    font-family: Helvetica;\n",
       "}\n",
       "table.dataframe {\n",
       "    border-collapse: collapse;\n",
       "    border: none;\n",
       "}\n",
       "table.dataframe tr {\n",
       "    border: none;\n",
       "}\n",
       "table.dataframe td, table.dataframe th {\n",
       "    margin: 0;\n",
       "    border: 1px solid white;\n",
       "    padding-left: 0.25em;\n",
       "    padding-right: 0.25em;\n",
       "}\n",
       "table.dataframe th:not(:empty) {\n",
       "    background-color: #fec;\n",
       "    text-align: left;\n",
       "    font-weight: normal;\n",
       "}\n",
       "table.dataframe tr:nth-child(2) th:empty {\n",
       "    border-left: none;\n",
       "    border-right: 1px dashed #888;\n",
       "}\n",
       "table.dataframe td {\n",
       "    border: 2px solid #ccf;\n",
       "    background-color: #f4f4ff;\n",
       "}\n",
       "h3 {\n",
       "    color: white;\n",
       "    background-color: black;\n",
       "    padding: 0.5em;\n",
       "}\n",
       "\n",
       ".CodeMirror {\n",
       "    font-family: 'Monaco for Powerline';\n",
       "}\n",
       "\n",
       "div.cell.code_cell {\n",
       "    font-family: 'Monaco for Powerline';\n",
       "}\n",
       "\n",
       "div.output_area pre {\n",
       "    font-family: 'Monaco for Powerline';\n",
       "}\n",
       "</style>"
      ],
      "text/plain": [
       "<IPython.core.display.HTML object>"
      ]
     },
     "execution_count": 3,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "from IPython.core.display import HTML\n",
    "css = open('style-table.css').read() + open('style-notebook.css').read()\n",
    "HTML('<style>{}</style>'.format(css))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "def asn_from_probe_name(probe_name):\n",
    "    base = 'http://leone-collector.eecs.jacobs-university.de/msmpoint'\n",
    "    msm_metadata = requests.get('%s/%s'%(base, probe_name))\n",
    "    asn4 = str(msm_metadata.json()['asn4'])\n",
    "    asn6 = str(msm_metadata.json()['asn6'])\n",
    "    return asn4,asn6"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "dailydb_location = '/data/vbajpai/datasets/youtube-data-collection-python3/youtube.db'"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "query = '''select   msmpoint.name as probe\n",
    "                  , dtime\n",
    "                  , video_address\n",
    "                  , video_download_bytes_sec\n",
    "                  , audio_download_bytes_sec\n",
    "           from youtube, msmpoint\n",
    "           where youtube.unit_id = msmpoint.skunitid\n",
    "           and msmpoint.skunitid <> ''\n",
    "           and successes = 1\n",
    "           and failures = 0;\n",
    "        '''"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "con = sqlite3.connect(dailydb_location)\n",
    "df = pd.read_sql(query, con)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "probe                       4422465\n",
       "dtime                       4422465\n",
       "video_address               4422465\n",
       "video_download_bytes_sec    4422465\n",
       "audio_download_bytes_sec    4422465\n",
       "dtype: int64"
      ]
     },
     "execution_count": 8,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.count()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style>\n",
       "    .dataframe thead tr:only-child th {\n",
       "        text-align: right;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: left;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>probe</th>\n",
       "      <th>dtime</th>\n",
       "      <th>video_address</th>\n",
       "      <th>video_download_bytes_sec</th>\n",
       "      <th>audio_download_bytes_sec</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>samsbox81</td>\n",
       "      <td>2016-09-05 16:29:41</td>\n",
       "      <td>193.212.4.15</td>\n",
       "      <td>2006335</td>\n",
       "      <td>1242187</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>samsbox81</td>\n",
       "      <td>2016-09-05 16:30:23</td>\n",
       "      <td>2001:4600:9:307::f</td>\n",
       "      <td>2032809</td>\n",
       "      <td>1146949</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>samsbox81</td>\n",
       "      <td>2016-09-05 17:26:17</td>\n",
       "      <td>193.212.4.15</td>\n",
       "      <td>1946785</td>\n",
       "      <td>1214358</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>samsbox81</td>\n",
       "      <td>2016-09-05 17:26:59</td>\n",
       "      <td>2001:4600:9:307::f</td>\n",
       "      <td>2008980</td>\n",
       "      <td>889966</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>samsbox81</td>\n",
       "      <td>2016-09-05 18:25:53</td>\n",
       "      <td>193.212.4.15</td>\n",
       "      <td>1927047</td>\n",
       "      <td>1181569</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "       probe                dtime       video_address  \\\n",
       "0  samsbox81  2016-09-05 16:29:41        193.212.4.15   \n",
       "1  samsbox81  2016-09-05 16:30:23  2001:4600:9:307::f   \n",
       "2  samsbox81  2016-09-05 17:26:17        193.212.4.15   \n",
       "3  samsbox81  2016-09-05 17:26:59  2001:4600:9:307::f   \n",
       "4  samsbox81  2016-09-05 18:25:53        193.212.4.15   \n",
       "\n",
       "   video_download_bytes_sec  audio_download_bytes_sec  \n",
       "0                   2006335                   1242187  \n",
       "1                   2032809                   1146949  \n",
       "2                   1946785                   1214358  \n",
       "3                   2008980                    889966  \n",
       "4                   1927047                   1181569  "
      ]
     },
     "execution_count": 9,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "def round_time_to_nearest_hour(dtime):\n",
    "    import datetime;    \n",
    "    d = datetime.datetime.strptime(dtime, \"%Y-%m-%d %H:%M:%S\")\n",
    "    d = d.replace(minute=0, second=0)\n",
    "    dtime = d.strftime('%Y-%m-%d %H:%M:%S')\n",
    "    return dtime"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "df['dtime'] = df['dtime'].apply(round_time_to_nearest_hour)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style>\n",
       "    .dataframe thead tr:only-child th {\n",
       "        text-align: right;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: left;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>probe</th>\n",
       "      <th>dtime</th>\n",
       "      <th>video_address</th>\n",
       "      <th>video_download_bytes_sec</th>\n",
       "      <th>audio_download_bytes_sec</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>samsbox81</td>\n",
       "      <td>2016-09-05 16:00:00</td>\n",
       "      <td>193.212.4.15</td>\n",
       "      <td>2006335</td>\n",
       "      <td>1242187</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>samsbox81</td>\n",
       "      <td>2016-09-05 16:00:00</td>\n",
       "      <td>2001:4600:9:307::f</td>\n",
       "      <td>2032809</td>\n",
       "      <td>1146949</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>samsbox81</td>\n",
       "      <td>2016-09-05 17:00:00</td>\n",
       "      <td>193.212.4.15</td>\n",
       "      <td>1946785</td>\n",
       "      <td>1214358</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>samsbox81</td>\n",
       "      <td>2016-09-05 17:00:00</td>\n",
       "      <td>2001:4600:9:307::f</td>\n",
       "      <td>2008980</td>\n",
       "      <td>889966</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>samsbox81</td>\n",
       "      <td>2016-09-05 18:00:00</td>\n",
       "      <td>193.212.4.15</td>\n",
       "      <td>1927047</td>\n",
       "      <td>1181569</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "       probe                dtime       video_address  \\\n",
       "0  samsbox81  2016-09-05 16:00:00        193.212.4.15   \n",
       "1  samsbox81  2016-09-05 16:00:00  2001:4600:9:307::f   \n",
       "2  samsbox81  2016-09-05 17:00:00        193.212.4.15   \n",
       "3  samsbox81  2016-09-05 17:00:00  2001:4600:9:307::f   \n",
       "4  samsbox81  2016-09-05 18:00:00        193.212.4.15   \n",
       "\n",
       "   video_download_bytes_sec  audio_download_bytes_sec  \n",
       "0                   2006335                   1242187  \n",
       "1                   2032809                   1146949  \n",
       "2                   1946785                   1214358  \n",
       "3                   2008980                    889966  \n",
       "4                   1927047                   1181569  "
      ]
     },
     "execution_count": 12,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "num_probes = len(set(df['probe']))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "probe                       4422465\n",
       "dtime                       4422465\n",
       "video_address               4422465\n",
       "video_download_bytes_sec    4422465\n",
       "audio_download_bytes_sec    4422465\n",
       "dtype: int64"
      ]
     },
     "execution_count": 14,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.count()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "df = df.dropna()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "probe                       4422465\n",
       "dtime                       4422465\n",
       "video_address               4422465\n",
       "video_download_bytes_sec    4422465\n",
       "audio_download_bytes_sec    4422465\n",
       "dtype: int64"
      ]
     },
     "execution_count": 16,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.count()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "df['v_MB_sec'] = df['video_download_bytes_sec'].div(1024).div(1024)\n",
    "df['a_MB_sec'] = df['audio_download_bytes_sec'].div(1024).div(1024)\n",
    "\n",
    "del df['video_download_bytes_sec']\n",
    "del df['audio_download_bytes_sec']"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "def filter_by_ip(df, column):\n",
    "  \"\"\"reads a DataFrame, calls fbi.filter_v{4,6} to get DataFrames with rows\n",
    "     containing only v4, v6 endpoints. Returns the two DataFrame as a tuple\"\"\"\n",
    "\n",
    "  dfv6 = fbi.filter_v6(df, column)\n",
    "  dfv4 = fbi.filter_v4(df, column)\n",
    "  return dfv4, dfv6"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "dfv4, dfv6 = filter_by_ip(df, column = 'video_address')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "dfv4['name'] = dfv4['probe'].apply(lambda name: \"#%02d\"%int(name[7:]))\n",
    "dfv6['name'] = dfv6['probe'].apply(lambda name: \"#%02d\"%int(name[7:]))\n",
    "del dfv4['probe']; del dfv6['probe']"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style>\n",
       "    .dataframe thead tr:only-child th {\n",
       "        text-align: right;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: left;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>dtime</th>\n",
       "      <th>video_address</th>\n",
       "      <th>v_MB_sec</th>\n",
       "      <th>a_MB_sec</th>\n",
       "      <th>name</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>2016-09-05 16:00:00</td>\n",
       "      <td>193.212.4.15</td>\n",
       "      <td>1.913390</td>\n",
       "      <td>1.184642</td>\n",
       "      <td>#81</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>2016-09-05 17:00:00</td>\n",
       "      <td>193.212.4.15</td>\n",
       "      <td>1.856599</td>\n",
       "      <td>1.158102</td>\n",
       "      <td>#81</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>2016-09-05 18:00:00</td>\n",
       "      <td>193.212.4.15</td>\n",
       "      <td>1.837775</td>\n",
       "      <td>1.126832</td>\n",
       "      <td>#81</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>2016-09-05 19:00:00</td>\n",
       "      <td>193.212.4.15</td>\n",
       "      <td>1.906452</td>\n",
       "      <td>1.091392</td>\n",
       "      <td>#81</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8</th>\n",
       "      <td>2016-09-05 20:00:00</td>\n",
       "      <td>193.212.4.15</td>\n",
       "      <td>1.896476</td>\n",
       "      <td>0.974868</td>\n",
       "      <td>#81</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                 dtime video_address  v_MB_sec  a_MB_sec name\n",
       "0  2016-09-05 16:00:00  193.212.4.15  1.913390  1.184642  #81\n",
       "2  2016-09-05 17:00:00  193.212.4.15  1.856599  1.158102  #81\n",
       "4  2016-09-05 18:00:00  193.212.4.15  1.837775  1.126832  #81\n",
       "6  2016-09-05 19:00:00  193.212.4.15  1.906452  1.091392  #81\n",
       "8  2016-09-05 20:00:00  193.212.4.15  1.896476  0.974868  #81"
      ]
     },
     "execution_count": 21,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "dfv4.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style>\n",
       "    .dataframe thead tr:only-child th {\n",
       "        text-align: right;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: left;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>dtime</th>\n",
       "      <th>video_address</th>\n",
       "      <th>v_MB_sec</th>\n",
       "      <th>a_MB_sec</th>\n",
       "      <th>name</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>2016-09-05 16:00:00</td>\n",
       "      <td>2001:4600:9:307::f</td>\n",
       "      <td>1.938638</td>\n",
       "      <td>1.093816</td>\n",
       "      <td>#81</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>2016-09-05 17:00:00</td>\n",
       "      <td>2001:4600:9:307::f</td>\n",
       "      <td>1.915913</td>\n",
       "      <td>0.848738</td>\n",
       "      <td>#81</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>2016-09-05 18:00:00</td>\n",
       "      <td>2001:4600:9:307::f</td>\n",
       "      <td>1.821363</td>\n",
       "      <td>1.107177</td>\n",
       "      <td>#81</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>2016-09-05 19:00:00</td>\n",
       "      <td>2001:4600:9:307::f</td>\n",
       "      <td>1.834697</td>\n",
       "      <td>1.188632</td>\n",
       "      <td>#81</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9</th>\n",
       "      <td>2016-09-05 20:00:00</td>\n",
       "      <td>2001:4600:9:307::f</td>\n",
       "      <td>1.909348</td>\n",
       "      <td>0.801767</td>\n",
       "      <td>#81</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                 dtime       video_address  v_MB_sec  a_MB_sec name\n",
       "1  2016-09-05 16:00:00  2001:4600:9:307::f  1.938638  1.093816  #81\n",
       "3  2016-09-05 17:00:00  2001:4600:9:307::f  1.915913  0.848738  #81\n",
       "5  2016-09-05 18:00:00  2001:4600:9:307::f  1.821363  1.107177  #81\n",
       "7  2016-09-05 19:00:00  2001:4600:9:307::f  1.834697  1.188632  #81\n",
       "9  2016-09-05 20:00:00  2001:4600:9:307::f  1.909348  0.801767  #81"
      ]
     },
     "execution_count": 22,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "dfv6.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "df = pd.merge(dfv4, dfv6, on=[\"name\", \"dtime\"])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "dtime              6581885\n",
       "video_address_x    6581885\n",
       "v_MB_sec_x         6581885\n",
       "a_MB_sec_x         6581885\n",
       "name               6581885\n",
       "video_address_y    6581885\n",
       "v_MB_sec_y         6581885\n",
       "a_MB_sec_y         6581885\n",
       "dtype: int64"
      ]
     },
     "execution_count": 24,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.count()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style>\n",
       "    .dataframe thead tr:only-child th {\n",
       "        text-align: right;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: left;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>dtime</th>\n",
       "      <th>video_address_x</th>\n",
       "      <th>v_MB_sec_x</th>\n",
       "      <th>a_MB_sec_x</th>\n",
       "      <th>name</th>\n",
       "      <th>video_address_y</th>\n",
       "      <th>v_MB_sec_y</th>\n",
       "      <th>a_MB_sec_y</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>2016-09-05 16:00:00</td>\n",
       "      <td>193.212.4.15</td>\n",
       "      <td>1.913390</td>\n",
       "      <td>1.184642</td>\n",
       "      <td>#81</td>\n",
       "      <td>2001:4600:9:307::f</td>\n",
       "      <td>1.938638</td>\n",
       "      <td>1.093816</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>2016-09-05 17:00:00</td>\n",
       "      <td>193.212.4.15</td>\n",
       "      <td>1.856599</td>\n",
       "      <td>1.158102</td>\n",
       "      <td>#81</td>\n",
       "      <td>2001:4600:9:307::f</td>\n",
       "      <td>1.915913</td>\n",
       "      <td>0.848738</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>2016-09-05 18:00:00</td>\n",
       "      <td>193.212.4.15</td>\n",
       "      <td>1.837775</td>\n",
       "      <td>1.126832</td>\n",
       "      <td>#81</td>\n",
       "      <td>2001:4600:9:307::f</td>\n",
       "      <td>1.821363</td>\n",
       "      <td>1.107177</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>2016-09-05 19:00:00</td>\n",
       "      <td>193.212.4.15</td>\n",
       "      <td>1.906452</td>\n",
       "      <td>1.091392</td>\n",
       "      <td>#81</td>\n",
       "      <td>2001:4600:9:307::f</td>\n",
       "      <td>1.834697</td>\n",
       "      <td>1.188632</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>2016-09-05 20:00:00</td>\n",
       "      <td>193.212.4.15</td>\n",
       "      <td>1.896476</td>\n",
       "      <td>0.974868</td>\n",
       "      <td>#81</td>\n",
       "      <td>2001:4600:9:307::f</td>\n",
       "      <td>1.909348</td>\n",
       "      <td>0.801767</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                 dtime video_address_x  v_MB_sec_x  a_MB_sec_x name  \\\n",
       "0  2016-09-05 16:00:00    193.212.4.15    1.913390    1.184642  #81   \n",
       "1  2016-09-05 17:00:00    193.212.4.15    1.856599    1.158102  #81   \n",
       "2  2016-09-05 18:00:00    193.212.4.15    1.837775    1.126832  #81   \n",
       "3  2016-09-05 19:00:00    193.212.4.15    1.906452    1.091392  #81   \n",
       "4  2016-09-05 20:00:00    193.212.4.15    1.896476    0.974868  #81   \n",
       "\n",
       "      video_address_y  v_MB_sec_y  a_MB_sec_y  \n",
       "0  2001:4600:9:307::f    1.938638    1.093816  \n",
       "1  2001:4600:9:307::f    1.915913    0.848738  \n",
       "2  2001:4600:9:307::f    1.821363    1.107177  \n",
       "3  2001:4600:9:307::f    1.834697    1.188632  \n",
       "4  2001:4600:9:307::f    1.909348    0.801767  "
      ]
     },
     "execution_count": 25,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "df.rename(columns={  'video_address_x'      : 'video_ep_v4'\n",
    "                   , 'v_MB_sec_x'           : 'v_MB_sec_v4'\n",
    "                   , 'a_MB_sec_x'           : 'a_MB_sec_v4'                   \n",
    "                   , 'video_address_y'      : 'video_ep_v6'\n",
    "                   , 'v_MB_sec_y'           : 'v_MB_sec_v6'\n",
    "                   , 'a_MB_sec_y'           : 'a_MB_sec_v6'                                      \n",
    "                  } ,  inplace=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "del df['video_ep_v4']; del df['video_ep_v6']"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style>\n",
       "    .dataframe thead tr:only-child th {\n",
       "        text-align: right;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: left;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>dtime</th>\n",
       "      <th>v_MB_sec_v4</th>\n",
       "      <th>a_MB_sec_v4</th>\n",
       "      <th>name</th>\n",
       "      <th>v_MB_sec_v6</th>\n",
       "      <th>a_MB_sec_v6</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>2016-09-05 16:00:00</td>\n",
       "      <td>1.913390</td>\n",
       "      <td>1.184642</td>\n",
       "      <td>#81</td>\n",
       "      <td>1.938638</td>\n",
       "      <td>1.093816</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>2016-09-05 17:00:00</td>\n",
       "      <td>1.856599</td>\n",
       "      <td>1.158102</td>\n",
       "      <td>#81</td>\n",
       "      <td>1.915913</td>\n",
       "      <td>0.848738</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>2016-09-05 18:00:00</td>\n",
       "      <td>1.837775</td>\n",
       "      <td>1.126832</td>\n",
       "      <td>#81</td>\n",
       "      <td>1.821363</td>\n",
       "      <td>1.107177</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>2016-09-05 19:00:00</td>\n",
       "      <td>1.906452</td>\n",
       "      <td>1.091392</td>\n",
       "      <td>#81</td>\n",
       "      <td>1.834697</td>\n",
       "      <td>1.188632</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>2016-09-05 20:00:00</td>\n",
       "      <td>1.896476</td>\n",
       "      <td>0.974868</td>\n",
       "      <td>#81</td>\n",
       "      <td>1.909348</td>\n",
       "      <td>0.801767</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                 dtime  v_MB_sec_v4  a_MB_sec_v4 name  v_MB_sec_v6  \\\n",
       "0  2016-09-05 16:00:00     1.913390     1.184642  #81     1.938638   \n",
       "1  2016-09-05 17:00:00     1.856599     1.158102  #81     1.915913   \n",
       "2  2016-09-05 18:00:00     1.837775     1.126832  #81     1.821363   \n",
       "3  2016-09-05 19:00:00     1.906452     1.091392  #81     1.834697   \n",
       "4  2016-09-05 20:00:00     1.896476     0.974868  #81     1.909348   \n",
       "\n",
       "   a_MB_sec_v6  \n",
       "0     1.093816  \n",
       "1     0.848738  \n",
       "2     1.107177  \n",
       "3     1.188632  \n",
       "4     0.801767  "
      ]
     },
     "execution_count": 28,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style>\n",
       "    .dataframe thead tr:only-child th {\n",
       "        text-align: right;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: left;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>dtime</th>\n",
       "      <th>v_MB_sec_v4</th>\n",
       "      <th>a_MB_sec_v4</th>\n",
       "      <th>name</th>\n",
       "      <th>v_MB_sec_v6</th>\n",
       "      <th>a_MB_sec_v6</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>6581880</th>\n",
       "      <td>2017-06-25 17:00:00</td>\n",
       "      <td>2.747395</td>\n",
       "      <td>1.138997</td>\n",
       "      <td>#91</td>\n",
       "      <td>2.712654</td>\n",
       "      <td>1.05301</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6581881</th>\n",
       "      <td>2017-06-25 17:00:00</td>\n",
       "      <td>2.747395</td>\n",
       "      <td>1.138997</td>\n",
       "      <td>#91</td>\n",
       "      <td>2.712654</td>\n",
       "      <td>1.05301</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6581882</th>\n",
       "      <td>2017-06-25 17:00:00</td>\n",
       "      <td>2.747395</td>\n",
       "      <td>1.138997</td>\n",
       "      <td>#91</td>\n",
       "      <td>2.712654</td>\n",
       "      <td>1.05301</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6581883</th>\n",
       "      <td>2017-06-25 17:00:00</td>\n",
       "      <td>2.747395</td>\n",
       "      <td>1.138997</td>\n",
       "      <td>#91</td>\n",
       "      <td>2.712654</td>\n",
       "      <td>1.05301</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6581884</th>\n",
       "      <td>2017-06-25 17:00:00</td>\n",
       "      <td>2.747395</td>\n",
       "      <td>1.138997</td>\n",
       "      <td>#91</td>\n",
       "      <td>2.712654</td>\n",
       "      <td>1.05301</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                       dtime  v_MB_sec_v4  a_MB_sec_v4 name  v_MB_sec_v6  \\\n",
       "6581880  2017-06-25 17:00:00     2.747395     1.138997  #91     2.712654   \n",
       "6581881  2017-06-25 17:00:00     2.747395     1.138997  #91     2.712654   \n",
       "6581882  2017-06-25 17:00:00     2.747395     1.138997  #91     2.712654   \n",
       "6581883  2017-06-25 17:00:00     2.747395     1.138997  #91     2.712654   \n",
       "6581884  2017-06-25 17:00:00     2.747395     1.138997  #91     2.712654   \n",
       "\n",
       "         a_MB_sec_v6  \n",
       "6581880      1.05301  \n",
       "6581881      1.05301  \n",
       "6581882      1.05301  \n",
       "6581883      1.05301  \n",
       "6581884      1.05301  "
      ]
     },
     "execution_count": 29,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.tail()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "dtime          6581885\n",
       "v_MB_sec_v4    6581885\n",
       "a_MB_sec_v4    6581885\n",
       "name           6581885\n",
       "v_MB_sec_v6    6581885\n",
       "a_MB_sec_v6    6581885\n",
       "dtype: int64"
      ]
     },
     "execution_count": 30,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.count()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "df['v4v6_v_MB_sec'] = df.v_MB_sec_v4.map(str) + ',' + df.v_MB_sec_v6.map(str)\n",
    "df['v4v6_a_MB_sec'] = df.a_MB_sec_v4.map(str) + ',' + df.a_MB_sec_v6.map(str)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "dtime            6581885\n",
       "v_MB_sec_v4      6581885\n",
       "a_MB_sec_v4      6581885\n",
       "name             6581885\n",
       "v_MB_sec_v6      6581885\n",
       "a_MB_sec_v6      6581885\n",
       "v4v6_v_MB_sec    6581885\n",
       "v4v6_a_MB_sec    6581885\n",
       "dtype: int64"
      ]
     },
     "execution_count": 32,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.count()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Calculate Absolute Difference"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 33,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "def get_difference(v4v6_throughput):\n",
    "    \n",
    "    v4, v6 = v4v6_throughput.split(',')\n",
    "    try: v4 = float(v4); v6 = float(v6)\n",
    "    except Exception as e: print(e, v4v6_throughput)\n",
    "    \n",
    "    difference = (v6 - v4) \n",
    "    return float('%0.2f'%difference)\n",
    "\n",
    "df['v_MB_sec_a'] = df['v4v6_v_MB_sec'].apply(get_difference)\n",
    "df['a_MB_sec_a'] = df['v4v6_a_MB_sec'].apply(get_difference)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Calculate Relative Difference"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 34,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "def get_slowness(v4v6_throughput):\n",
    "    \n",
    "    v4, v6 = v4v6_throughput.split(',')\n",
    "    try: v4 = float(v4); v6 = float(v6)\n",
    "    except Exception as e: print(e, v4v6_throughput)\n",
    "    \n",
    "    try: slowness = (v6 - v4) / v4\n",
    "    except Exception as e: return 0\n",
    "    return float('%0.2f'%slowness)\n",
    "\n",
    "df['v_MB_sec_r'] = df['v4v6_v_MB_sec'].apply(get_slowness)\n",
    "df['a_MB_sec_r'] = df['v4v6_a_MB_sec'].apply(get_slowness)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 35,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "del df['v4v6_a_MB_sec']; del df['v4v6_v_MB_sec']"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 36,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "def round_time_to_day(dtime):\n",
    "    import datetime;    \n",
    "    d = datetime.datetime.strptime(dtime, \"%Y-%m-%d %H:%M:%S\")\n",
    "    d = d.replace(hour=0, minute=0, second=0)\n",
    "    dtime = d.strftime('%Y-%m-%d')\n",
    "    return dtime\n",
    "\n",
    "df['dtime'] = df['dtime'].apply(round_time_to_day)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 37,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "dtime          6581885\n",
       "v_MB_sec_v4    6581885\n",
       "a_MB_sec_v4    6581885\n",
       "name           6581885\n",
       "v_MB_sec_v6    6581885\n",
       "a_MB_sec_v6    6581885\n",
       "v_MB_sec_a     6581885\n",
       "a_MB_sec_a     6581885\n",
       "v_MB_sec_r     6581885\n",
       "a_MB_sec_r     6581885\n",
       "dtype: int64"
      ]
     },
     "execution_count": 37,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.count()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 38,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "targets = ['v_MB_sec_v4', 'a_MB_sec_v4', \n",
    "           'v_MB_sec_v6', 'a_MB_sec_v6',\n",
    "           \n",
    "           'v_MB_sec_a', 'a_MB_sec_a',\n",
    "           'v_MB_sec_r', 'a_MB_sec_r' ,          \n",
    "          ]\n",
    "\n",
    "def get_throughput_across_probes(df):\n",
    "    dfgroupby  = cbc.group(df, column=['dtime'])\n",
    "    df_agg     = cbc.aggregate(dfgroupby, targets, ['median'])\n",
    "    df_merged  = cbc.merge(df, df_agg, column=['dtime'])\n",
    "    df         = df_merged.drop(targets + ['name'], axis=1)\n",
    "    df         = df.drop_duplicates()\n",
    "    df.columns = ['dtime'] + targets\n",
    "    return df\n",
    "\n",
    "df = get_throughput_across_probes(df)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 39,
   "metadata": {},
   "outputs": [
    {
     "ename": "AttributeError",
     "evalue": "'DataFrame' object has no attribute 'sort'",
     "output_type": "error",
     "traceback": [
      "\u001b[0;31m---------------------------------------------------------------------------\u001b[0m",
      "\u001b[0;31mAttributeError\u001b[0m                            Traceback (most recent call last)",
      "\u001b[0;32m<ipython-input-39-6cbabfe4ed8a>\u001b[0m in \u001b[0;36m<module>\u001b[0;34m()\u001b[0m\n\u001b[1;32m      1\u001b[0m \u001b[0mdf\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0;34m'dtime'\u001b[0m\u001b[0;34m]\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mdf\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0;34m'dtime'\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mapply\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;32mlambda\u001b[0m \u001b[0md\u001b[0m\u001b[0;34m:\u001b[0m \u001b[0mdt\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mstrptime\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0md\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0;34m'%Y-%m-%d'\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m----> 2\u001b[0;31m \u001b[0mdf\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0msort\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mcolumns\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0;34m'dtime'\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0minplace\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0;32mTrue\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m",
      "\u001b[0;32m~/.virtualenvs/youtube-data-analysis-python3/lib/python3.5/site-packages/pandas/core/generic.py\u001b[0m in \u001b[0;36m__getattr__\u001b[0;34m(self, name)\u001b[0m\n\u001b[1;32m   3079\u001b[0m             \u001b[0;32mif\u001b[0m \u001b[0mname\u001b[0m \u001b[0;32min\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0m_info_axis\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m   3080\u001b[0m                 \u001b[0;32mreturn\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0mname\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m-> 3081\u001b[0;31m             \u001b[0;32mreturn\u001b[0m \u001b[0mobject\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0m__getattribute__\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mself\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mname\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m   3082\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m   3083\u001b[0m     \u001b[0;32mdef\u001b[0m \u001b[0m__setattr__\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mself\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mname\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mvalue\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
      "\u001b[0;31mAttributeError\u001b[0m: 'DataFrame' object has no attribute 'sort'"
     ]
    }
   ],
   "source": [
    "df['dtime'] = df['dtime'].apply(lambda d: dt.strptime(d, '%Y-%m-%d'))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 40,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "df.sort_values(by = 'dtime', inplace = True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 41,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style>\n",
       "    .dataframe thead tr:only-child th {\n",
       "        text-align: right;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: left;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>dtime</th>\n",
       "      <th>v_MB_sec_v4</th>\n",
       "      <th>a_MB_sec_v4</th>\n",
       "      <th>v_MB_sec_v6</th>\n",
       "      <th>a_MB_sec_v6</th>\n",
       "      <th>v_MB_sec_a</th>\n",
       "      <th>a_MB_sec_a</th>\n",
       "      <th>v_MB_sec_r</th>\n",
       "      <th>a_MB_sec_r</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>1592615</th>\n",
       "      <td>2014-08-21</td>\n",
       "      <td>0.923781</td>\n",
       "      <td>0.058216</td>\n",
       "      <td>0.023630</td>\n",
       "      <td>0.022566</td>\n",
       "      <td>-0.88</td>\n",
       "      <td>-0.04</td>\n",
       "      <td>-0.97</td>\n",
       "      <td>-0.615</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1592625</th>\n",
       "      <td>2014-08-22</td>\n",
       "      <td>0.923466</td>\n",
       "      <td>0.058382</td>\n",
       "      <td>0.021913</td>\n",
       "      <td>0.021980</td>\n",
       "      <td>-0.88</td>\n",
       "      <td>-0.04</td>\n",
       "      <td>-0.98</td>\n",
       "      <td>-0.620</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1592648</th>\n",
       "      <td>2014-08-23</td>\n",
       "      <td>0.924054</td>\n",
       "      <td>0.058375</td>\n",
       "      <td>0.021990</td>\n",
       "      <td>0.021621</td>\n",
       "      <td>-0.88</td>\n",
       "      <td>-0.04</td>\n",
       "      <td>-0.98</td>\n",
       "      <td>-0.630</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1592671</th>\n",
       "      <td>2014-08-24</td>\n",
       "      <td>0.924189</td>\n",
       "      <td>0.058300</td>\n",
       "      <td>0.024550</td>\n",
       "      <td>0.025263</td>\n",
       "      <td>-0.88</td>\n",
       "      <td>-0.03</td>\n",
       "      <td>-0.97</td>\n",
       "      <td>-0.570</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1592694</th>\n",
       "      <td>2014-08-25</td>\n",
       "      <td>0.822330</td>\n",
       "      <td>0.058096</td>\n",
       "      <td>0.026024</td>\n",
       "      <td>0.025701</td>\n",
       "      <td>-0.78</td>\n",
       "      <td>-0.03</td>\n",
       "      <td>-0.97</td>\n",
       "      <td>-0.565</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "             dtime  v_MB_sec_v4  a_MB_sec_v4  v_MB_sec_v6  a_MB_sec_v6  \\\n",
       "1592615 2014-08-21     0.923781     0.058216     0.023630     0.022566   \n",
       "1592625 2014-08-22     0.923466     0.058382     0.021913     0.021980   \n",
       "1592648 2014-08-23     0.924054     0.058375     0.021990     0.021621   \n",
       "1592671 2014-08-24     0.924189     0.058300     0.024550     0.025263   \n",
       "1592694 2014-08-25     0.822330     0.058096     0.026024     0.025701   \n",
       "\n",
       "         v_MB_sec_a  a_MB_sec_a  v_MB_sec_r  a_MB_sec_r  \n",
       "1592615       -0.88       -0.04       -0.97      -0.615  \n",
       "1592625       -0.88       -0.04       -0.98      -0.620  \n",
       "1592648       -0.88       -0.04       -0.98      -0.630  \n",
       "1592671       -0.88       -0.03       -0.97      -0.570  \n",
       "1592694       -0.78       -0.03       -0.97      -0.565  "
      ]
     },
     "execution_count": 41,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 42,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style>\n",
       "    .dataframe thead tr:only-child th {\n",
       "        text-align: right;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: left;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>dtime</th>\n",
       "      <th>v_MB_sec_v4</th>\n",
       "      <th>a_MB_sec_v4</th>\n",
       "      <th>v_MB_sec_v6</th>\n",
       "      <th>a_MB_sec_v6</th>\n",
       "      <th>v_MB_sec_a</th>\n",
       "      <th>a_MB_sec_a</th>\n",
       "      <th>v_MB_sec_r</th>\n",
       "      <th>a_MB_sec_r</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>98576</th>\n",
       "      <td>2017-06-23</td>\n",
       "      <td>0.572040</td>\n",
       "      <td>0.756555</td>\n",
       "      <td>0.571938</td>\n",
       "      <td>0.724823</td>\n",
       "      <td>-0.00</td>\n",
       "      <td>-0.02</td>\n",
       "      <td>-0.00</td>\n",
       "      <td>-0.03</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>99176</th>\n",
       "      <td>2017-06-24</td>\n",
       "      <td>0.686526</td>\n",
       "      <td>0.745583</td>\n",
       "      <td>0.682394</td>\n",
       "      <td>0.686921</td>\n",
       "      <td>-0.00</td>\n",
       "      <td>-0.02</td>\n",
       "      <td>0.00</td>\n",
       "      <td>-0.03</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>99560</th>\n",
       "      <td>2017-06-25</td>\n",
       "      <td>1.531445</td>\n",
       "      <td>0.604259</td>\n",
       "      <td>1.496117</td>\n",
       "      <td>0.552988</td>\n",
       "      <td>-0.02</td>\n",
       "      <td>-0.03</td>\n",
       "      <td>-0.01</td>\n",
       "      <td>-0.04</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>99758</th>\n",
       "      <td>2017-06-26</td>\n",
       "      <td>2.026135</td>\n",
       "      <td>0.593721</td>\n",
       "      <td>2.069180</td>\n",
       "      <td>0.524764</td>\n",
       "      <td>-0.03</td>\n",
       "      <td>-0.03</td>\n",
       "      <td>-0.02</td>\n",
       "      <td>-0.05</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>99854</th>\n",
       "      <td>2017-06-27</td>\n",
       "      <td>1.193176</td>\n",
       "      <td>0.571957</td>\n",
       "      <td>1.172191</td>\n",
       "      <td>0.512903</td>\n",
       "      <td>-0.01</td>\n",
       "      <td>-0.02</td>\n",
       "      <td>-0.01</td>\n",
       "      <td>-0.04</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "           dtime  v_MB_sec_v4  a_MB_sec_v4  v_MB_sec_v6  a_MB_sec_v6  \\\n",
       "98576 2017-06-23     0.572040     0.756555     0.571938     0.724823   \n",
       "99176 2017-06-24     0.686526     0.745583     0.682394     0.686921   \n",
       "99560 2017-06-25     1.531445     0.604259     1.496117     0.552988   \n",
       "99758 2017-06-26     2.026135     0.593721     2.069180     0.524764   \n",
       "99854 2017-06-27     1.193176     0.571957     1.172191     0.512903   \n",
       "\n",
       "       v_MB_sec_a  a_MB_sec_a  v_MB_sec_r  a_MB_sec_r  \n",
       "98576       -0.00       -0.02       -0.00       -0.03  \n",
       "99176       -0.00       -0.02        0.00       -0.03  \n",
       "99560       -0.02       -0.03       -0.01       -0.04  \n",
       "99758       -0.03       -0.03       -0.02       -0.05  \n",
       "99854       -0.01       -0.02       -0.01       -0.04  "
      ]
     },
     "execution_count": 42,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.tail()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 43,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "dtime          1040\n",
       "v_MB_sec_v4    1040\n",
       "a_MB_sec_v4    1040\n",
       "v_MB_sec_v6    1040\n",
       "a_MB_sec_v6    1040\n",
       "v_MB_sec_a     1040\n",
       "a_MB_sec_a     1040\n",
       "v_MB_sec_r     1040\n",
       "a_MB_sec_r     1040\n",
       "dtype: int64"
      ]
     },
     "execution_count": 43,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.count()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Hardcoded dates"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 44,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "from dateutil import rrule\n",
    "a=dt.strptime(str('2014-08-01'), '%Y-%m-%d')\n",
    "b=dt.strptime(str('2017-07-01'), '%Y-%m-%d')\n",
    "dateranges = rrule.rrule(rrule.DAILY, dtstart=a).between(a, b, inc=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 46,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "df_ = pd.DataFrame(dateranges, columns=['dtime']).sort_values(by = 'dtime').reset_index(drop = True)\n",
    "df = pd.merge(df_, df, how='outer')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 49,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAa0AAACPCAYAAABJRYS+AAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nB8yIgPQYd+zKAAAAAElFTkSuQmCC\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7efd1b66f978>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "fig = plt.figure(figsize = (6, 1.3))\n",
    "plt.rcParams['font.size'] = 12\n",
    "ax = plt.gca()\n",
    "\n",
    "sl = df.plot(  x='dtime'\n",
    "             , y=['v_MB_sec_a', 'a_MB_sec_a']\n",
    "             , color=['Cyan', 'Magenta']\n",
    "             , kind='line'\n",
    "             , linewidth = 1\n",
    "             , ax = ax\n",
    "            )\n",
    "\n",
    "ax.grid(False)\n",
    "ax.set_xlabel('', fontsize=22)\n",
    "ax.set_ylabel('∆tp (MB/s)', fontsize=12)\n",
    "ax.legend(['Video', 'Audio'], fontsize=12, loc='best', ncol = 2)\n",
    "ax.set_yscale('linear')\n",
    "ax.set_ylim([-1, 0.2])\n",
    "\n",
    "yticks = np.arange(-1,0.3,0.3)\n",
    "ax.set_yticks(yticks)\n",
    "\n",
    "ax.spines['right'].set_color('none')\n",
    "ax.spines['top'].set_color('none')\n",
    "ax.yaxis.set_ticks_position('left')\n",
    "ax.xaxis.set_ticks_position('bottom')\n",
    "ax.spines['bottom'].set_position(('axes', -0.05))\n",
    "ax.spines['left'].set_position(('axes', -0.05))\n",
    "\n",
    "from datetime import datetime as dt\n",
    "ax.axvline(x=dt.strptime('2014-08-21', \"%Y-%m-%d\"), linewidth=1.0, ymax = 0.95, ymin = 0, color='black', ls = 'dotted')\n",
    "ax.axvline(x=dt.strptime('2014-12-10', \"%Y-%m-%d\"), linewidth=1.0, ymax = 0.95, ymin = 0, color='black', ls = 'dotted')\n",
    "ax.axvline(x=dt.strptime('2015-01-22', \"%Y-%m-%d\"), linewidth=1.0, ymax = 0.95, ymin = 0, color='black', ls = 'dotted')\n",
    "ax.axvline(x=dt.strptime('2016-09-05', \"%Y-%m-%d\"), linewidth=1.0, ymax = 0.95, ymin = 0, color='black', ls = 'dotted')\n",
    "\n",
    "plt.title('Throughput', fontsize=12, y=1.05)\n",
    "plt.savefig('youtube-audio-video-throughput-timeseries.pdf', format='pdf', bbox_inches='tight')"
   ]
  }
 ],
 "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": "3.5.3"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 1
}