{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {
    "collapsed": false
   },
   "source": [
    "###### Future question for Dr. Soto\n",
    "\n",
    "Should I concatinate the dataframes on integer indices or Date-Time indices? I dont think it will make a difference.. not sure which one is easier"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 777,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
   ],
   "source": [
    "import pandas as pd\n",
    "import numpy as np"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 778,
   "metadata": {
    "collapsed": false
   },
   "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>Timestamp</th>\n",
       "      <th>Temperature</th>\n",
       "      <th>Humidity</th>\n",
       "      <th>CO2</th>\n",
       "      <th>Noise</th>\n",
       "      <th>Pressure</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>count</th>\n",
       "      <td>9.014300e+04</td>\n",
       "      <td>90143.000000</td>\n",
       "      <td>90143.000000</td>\n",
       "      <td>90137.000000</td>\n",
       "      <td>90132.000000</td>\n",
       "      <td>90143.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>mean</th>\n",
       "      <td>1.469613e+09</td>\n",
       "      <td>22.766724</td>\n",
       "      <td>51.291637</td>\n",
       "      <td>550.204799</td>\n",
       "      <td>38.964730</td>\n",
       "      <td>1011.348933</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>std</th>\n",
       "      <td>7.900773e+06</td>\n",
       "      <td>1.592268</td>\n",
       "      <td>6.929620</td>\n",
       "      <td>318.321732</td>\n",
       "      <td>7.100703</td>\n",
       "      <td>4.217541</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>min</th>\n",
       "      <td>1.455917e+09</td>\n",
       "      <td>17.900000</td>\n",
       "      <td>27.000000</td>\n",
       "      <td>201.000000</td>\n",
       "      <td>35.000000</td>\n",
       "      <td>995.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>25%</th>\n",
       "      <td>1.462765e+09</td>\n",
       "      <td>21.700000</td>\n",
       "      <td>49.000000</td>\n",
       "      <td>354.000000</td>\n",
       "      <td>36.000000</td>\n",
       "      <td>1008.300000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>50%</th>\n",
       "      <td>1.469657e+09</td>\n",
       "      <td>22.900000</td>\n",
       "      <td>52.000000</td>\n",
       "      <td>416.000000</td>\n",
       "      <td>36.000000</td>\n",
       "      <td>1011.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>75%</th>\n",
       "      <td>1.476459e+09</td>\n",
       "      <td>23.800000</td>\n",
       "      <td>55.000000</td>\n",
       "      <td>639.000000</td>\n",
       "      <td>38.000000</td>\n",
       "      <td>1014.100000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>max</th>\n",
       "      <td>1.483257e+09</td>\n",
       "      <td>28.500000</td>\n",
       "      <td>76.000000</td>\n",
       "      <td>2777.000000</td>\n",
       "      <td>79.000000</td>\n",
       "      <td>1027.500000</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ]
     },
     "execution_count": 778,
     "metadata": {
     },
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df1 = pd.read_csv('NetAtmo_2016.csv', parse_dates = True,)\n",
    "df1.describe()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 779,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
   ],
   "source": [
    "new_index1 = pd.Series(range(1,90144))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 780,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
   ],
   "source": [
    "df1['Numbered_index'] = new_index1"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 781,
   "metadata": {
    "collapsed": false
   },
   "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>Timestamp</th>\n",
       "      <th>Timezone : America/Los_Angeles</th>\n",
       "      <th>Temperature</th>\n",
       "      <th>Humidity</th>\n",
       "      <th>CO2</th>\n",
       "      <th>Noise</th>\n",
       "      <th>Pressure</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Numbered_index</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>1455917199</td>\n",
       "      <td>2/19/16 13:26</td>\n",
       "      <td>18.8</td>\n",
       "      <td>76</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>1015.7</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>1455917255</td>\n",
       "      <td>2/19/16 13:27</td>\n",
       "      <td>19.2</td>\n",
       "      <td>75</td>\n",
       "      <td>718.0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>1015.7</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>1455917257</td>\n",
       "      <td>2/19/16 13:27</td>\n",
       "      <td>19.9</td>\n",
       "      <td>73</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>1015.7</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>1455917513</td>\n",
       "      <td>2/19/16 13:31</td>\n",
       "      <td>20.3</td>\n",
       "      <td>73</td>\n",
       "      <td>337.0</td>\n",
       "      <td>44.0</td>\n",
       "      <td>1015.8</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>1455917814</td>\n",
       "      <td>2/19/16 13:36</td>\n",
       "      <td>21.2</td>\n",
       "      <td>70</td>\n",
       "      <td>332.0</td>\n",
       "      <td>47.0</td>\n",
       "      <td>1015.7</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ]
     },
     "execution_count": 781,
     "metadata": {
     },
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df1.set_index('Numbered_index', inplace = True)\n",
    "df1.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 782,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
   ],
   "source": [
    "df1.drop(df1.columns[[0,2,3,5,6]], axis =1, inplace = True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 783,
   "metadata": {
    "collapsed": false
   },
   "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>Timezone : America/Los_Angeles</th>\n",
       "      <th>CO2</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Numbered_index</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>2/19/16 13:26</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ]
     },
     "execution_count": 783,
     "metadata": {
     },
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df1.head(1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 784,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
   ],
   "source": [
    "df2 = pd.read_csv('NetAtmo_2017.csv', parse_dates = True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 785,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
   ],
   "source": [
    "new_index2 = pd.Series(range(90144, 100992))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 786,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
   ],
   "source": [
    "df2['numbered_index'] = new_index2"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 787,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
   ],
   "source": [
    "df2.set_index('numbered_index', inplace = True)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 788,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
   ],
   "source": [
    "df2.drop(df2.columns[[0,2,3,5,6]], axis =1, inplace = True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 789,
   "metadata": {
    "collapsed": false
   },
   "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>Time</th>\n",
       "      <th>CO2</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>numbered_index</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>90144</th>\n",
       "      <td>1/1/17 0:00</td>\n",
       "      <td>482</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>90145</th>\n",
       "      <td>1/1/17 0:05</td>\n",
       "      <td>491</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>90146</th>\n",
       "      <td>1/1/17 0:11</td>\n",
       "      <td>480</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>90147</th>\n",
       "      <td>1/1/17 0:16</td>\n",
       "      <td>486</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>90148</th>\n",
       "      <td>1/1/17 0:21</td>\n",
       "      <td>490</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ]
     },
     "execution_count": 789,
     "metadata": {
     },
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df2.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 790,
   "metadata": {
    "collapsed": false
   },
   "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>Timezone : America/Los_Angeles</th>\n",
       "      <th>CO2</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Numbered_index</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>2/19/16 13:26</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>2/19/16 13:27</td>\n",
       "      <td>718.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>2/19/16 13:27</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>2/19/16 13:31</td>\n",
       "      <td>337.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>2/19/16 13:36</td>\n",
       "      <td>332.0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ]
     },
     "execution_count": 790,
     "metadata": {
     },
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df1.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 791,
   "metadata": {
    "collapsed": false
   },
   "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>Time</th>\n",
       "      <th>CO2</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Numbered_index</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>2/19/16 13:26</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>2/19/16 13:27</td>\n",
       "      <td>718.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>2/19/16 13:27</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>2/19/16 13:31</td>\n",
       "      <td>337.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>2/19/16 13:36</td>\n",
       "      <td>332.0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ]
     },
     "execution_count": 791,
     "metadata": {
     },
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df1 = df1.rename(columns = {'Timezone : America/Los_Angeles':'Time'})\n",
    "df1.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 792,
   "metadata": {
    "collapsed": false
   },
   "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>Time</th>\n",
       "      <th>CO2</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>numbered_index</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>100987</th>\n",
       "      <td>2/12/17 18:27</td>\n",
       "      <td>484</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>100988</th>\n",
       "      <td>2/12/17 18:32</td>\n",
       "      <td>486</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>100989</th>\n",
       "      <td>2/12/17 18:37</td>\n",
       "      <td>469</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>100990</th>\n",
       "      <td>2/12/17 18:42</td>\n",
       "      <td>485</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>100991</th>\n",
       "      <td>2/12/17 18:47</td>\n",
       "      <td>480</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ]
     },
     "execution_count": 792,
     "metadata": {
     },
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df2.tail()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 793,
   "metadata": {
    "collapsed": false
   },
   "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>Time</th>\n",
       "      <th>CO2</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>2/19/16 13:26</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>2/19/16 13:27</td>\n",
       "      <td>718.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>2/19/16 13:27</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>2/19/16 13:31</td>\n",
       "      <td>337.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>2/19/16 13:36</td>\n",
       "      <td>332.0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ]
     },
     "execution_count": 793,
     "metadata": {
     },
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df3 = pd.concat([df1,df2])\n",
    "df3.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 794,
   "metadata": {
    "collapsed": false
   },
   "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>Time</th>\n",
       "      <th>CO2</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>100987</th>\n",
       "      <td>2/12/17 18:27</td>\n",
       "      <td>484.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>100988</th>\n",
       "      <td>2/12/17 18:32</td>\n",
       "      <td>486.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>100989</th>\n",
       "      <td>2/12/17 18:37</td>\n",
       "      <td>469.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>100990</th>\n",
       "      <td>2/12/17 18:42</td>\n",
       "      <td>485.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>100991</th>\n",
       "      <td>2/12/17 18:47</td>\n",
       "      <td>480.0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ]
     },
     "execution_count": 794,
     "metadata": {
     },
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df3.tail()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 795,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.axes._subplots.AxesSubplot at 0x7fc30f81eb00>"
      ]
     },
     "execution_count": 795,
     "metadata": {
     },
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "ca3b9873cafef467138c03a5f781afff01615a03"
     },
     "execution_count": 795,
     "metadata": {
     },
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df3.plot()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 796,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
   ],
   "source": [
    "#df3.set_index('Time', inplace = True)\n",
    "#df3.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 797,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
   ],
   "source": [
    "#df3.plot()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 798,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
   ],
   "source": [
    "#df3.plot.hist()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 799,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Time     object\n",
       "CO2     float64\n",
       "dtype: object"
      ]
     },
     "execution_count": 799,
     "metadata": {
     },
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df3.dtypes"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 800,
   "metadata": {
    "collapsed": false
   },
   "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>Time</th>\n",
       "      <th>CO2</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>2/19/16 13:26</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>2/19/16 13:27</td>\n",
       "      <td>718.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>2/19/16 13:27</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>2/19/16 13:31</td>\n",
       "      <td>337.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>2/19/16 13:36</td>\n",
       "      <td>332.0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ]
     },
     "execution_count": 800,
     "metadata": {
     },
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df3.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 801,
   "metadata": {
    "collapsed": false
   },
   "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>Time</th>\n",
       "      <th>CO2</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>False</td>\n",
       "      <td>True</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>False</td>\n",
       "      <td>True</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ]
     },
     "execution_count": 801,
     "metadata": {
     },
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df3.isnull().head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 0,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
   ],
   "source": [
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 0,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
   ],
   "source": [
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 802,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
   ],
   "source": [
    "df3['Time'] = pd.to_datetime(df3.Time)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 803,
   "metadata": {
    "collapsed": false
   },
   "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>Time</th>\n",
       "      <th>CO2</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>2016-02-19 13:26:00</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>2016-02-19 13:27:00</td>\n",
       "      <td>718.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>2016-02-19 13:27:00</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>2016-02-19 13:31:00</td>\n",
       "      <td>337.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>2016-02-19 13:36:00</td>\n",
       "      <td>332.0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ]
     },
     "execution_count": 803,
     "metadata": {
     },
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df3.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "collapsed": false
   },
   "source": [
    "https://pandas.pydata.org/pandas-docs/stable/api.html#datetimelike-properties"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 804,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "1    Friday\n",
       "2    Friday\n",
       "3    Friday\n",
       "4    Friday\n",
       "5    Friday\n",
       "Name: Time, dtype: object"
      ]
     },
     "execution_count": 804,
     "metadata": {
     },
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df3.Time.dt.weekday_name.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 805,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
   ],
   "source": [
    "#isolating the seonc day\n",
    "Firstday = pd.to_datetime('2/20/2016 23:59:59')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 806,
   "metadata": {
    "collapsed": false
   },
   "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>Time</th>\n",
       "      <th>CO2</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>411</th>\n",
       "      <td>2016-02-20 23:39:00</td>\n",
       "      <td>400.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>412</th>\n",
       "      <td>2016-02-20 23:44:00</td>\n",
       "      <td>419.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>413</th>\n",
       "      <td>2016-02-20 23:49:00</td>\n",
       "      <td>407.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>414</th>\n",
       "      <td>2016-02-20 23:54:00</td>\n",
       "      <td>417.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>415</th>\n",
       "      <td>2016-02-20 23:59:00</td>\n",
       "      <td>417.0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ]
     },
     "execution_count": 806,
     "metadata": {
     },
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df3.loc[df3.Time <= Firstday, :].tail()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 807,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Timedelta('359 days 05:21:00')"
      ]
     },
     "execution_count": 807,
     "metadata": {
     },
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#almost a full year of data!\n",
    "(df3.Time.max() - df3.Time.min())"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 808,
   "metadata": {
    "collapsed": false
   },
   "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>Time</th>\n",
       "      <th>CO2</th>\n",
       "      <th>Day</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>2016-02-19 13:26:00</td>\n",
       "      <td>NaN</td>\n",
       "      <td>Friday</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>2016-02-19 13:27:00</td>\n",
       "      <td>718.0</td>\n",
       "      <td>Friday</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>2016-02-19 13:27:00</td>\n",
       "      <td>NaN</td>\n",
       "      <td>Friday</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>2016-02-19 13:31:00</td>\n",
       "      <td>337.0</td>\n",
       "      <td>Friday</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>2016-02-19 13:36:00</td>\n",
       "      <td>332.0</td>\n",
       "      <td>Friday</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ]
     },
     "execution_count": 808,
     "metadata": {
     },
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df3['Day'] = df3.Time.dt.weekday_name\n",
    "df3.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 809,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Saturday     14598\n",
       "Tuesday      14589\n",
       "Sunday       14539\n",
       "Monday       14488\n",
       "Wednesday    14472\n",
       "Friday       14438\n",
       "Thursday     13867\n",
       "Name: Day, dtype: int64"
      ]
     },
     "execution_count": 809,
     "metadata": {
     },
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# so many questions\n",
    "df3.Day.value_counts()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 810,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.axes._subplots.AxesSubplot at 0x7fc30f0ed1d0>"
      ]
     },
     "execution_count": 810,
     "metadata": {
     },
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "4d2d598854af700a098269dbb982bfd299de63df"
     },
     "execution_count": 810,
     "metadata": {
     },
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df3.Day.value_counts().plot()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "collapsed": false
   },
   "source": [
    "Switching to df2 because it is still note recognized by datetime"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 811,
   "metadata": {
    "collapsed": false
   },
   "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>Time</th>\n",
       "      <th>CO2</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>numbered_index</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>90144</th>\n",
       "      <td>1/1/17 0:00</td>\n",
       "      <td>482</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>90145</th>\n",
       "      <td>1/1/17 0:05</td>\n",
       "      <td>491</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>90146</th>\n",
       "      <td>1/1/17 0:11</td>\n",
       "      <td>480</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>90147</th>\n",
       "      <td>1/1/17 0:16</td>\n",
       "      <td>486</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>90148</th>\n",
       "      <td>1/1/17 0:21</td>\n",
       "      <td>490</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ]
     },
     "execution_count": 811,
     "metadata": {
     },
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df2.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 812,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
   ],
   "source": [
    "#df3['Time2'].head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 813,
   "metadata": {
    "collapsed": false
   },
   "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>Time</th>\n",
       "      <th>CO2</th>\n",
       "      <th>Day</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>2016-02-19 13:26:00</td>\n",
       "      <td>NaN</td>\n",
       "      <td>Friday</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>2016-02-19 13:27:00</td>\n",
       "      <td>718.0</td>\n",
       "      <td>Friday</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>2016-02-19 13:27:00</td>\n",
       "      <td>NaN</td>\n",
       "      <td>Friday</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>2016-02-19 13:31:00</td>\n",
       "      <td>337.0</td>\n",
       "      <td>Friday</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>2016-02-19 13:36:00</td>\n",
       "      <td>332.0</td>\n",
       "      <td>Friday</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ]
     },
     "execution_count": 813,
     "metadata": {
     },
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df3.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 814,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
   ],
   "source": [
    "df3['Time2'] = df3.Time.shift(-1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 815,
   "metadata": {
    "collapsed": false
   },
   "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>Time</th>\n",
       "      <th>CO2</th>\n",
       "      <th>Day</th>\n",
       "      <th>Time2</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>2016-02-19 13:26:00</td>\n",
       "      <td>NaN</td>\n",
       "      <td>Friday</td>\n",
       "      <td>2016-02-19 13:27:00</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>2016-02-19 13:27:00</td>\n",
       "      <td>718.0</td>\n",
       "      <td>Friday</td>\n",
       "      <td>2016-02-19 13:27:00</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>2016-02-19 13:27:00</td>\n",
       "      <td>NaN</td>\n",
       "      <td>Friday</td>\n",
       "      <td>2016-02-19 13:31:00</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>2016-02-19 13:31:00</td>\n",
       "      <td>337.0</td>\n",
       "      <td>Friday</td>\n",
       "      <td>2016-02-19 13:36:00</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>2016-02-19 13:36:00</td>\n",
       "      <td>332.0</td>\n",
       "      <td>Friday</td>\n",
       "      <td>2016-02-19 13:41:00</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ]
     },
     "execution_count": 815,
     "metadata": {
     },
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df3.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 816,
   "metadata": {
    "collapsed": false
   },
   "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>Time</th>\n",
       "      <th>CO2</th>\n",
       "      <th>Day</th>\n",
       "      <th>Time2</th>\n",
       "      <th>TimeDel</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>2016-02-19 13:26:00</td>\n",
       "      <td>NaN</td>\n",
       "      <td>Friday</td>\n",
       "      <td>2016-02-19 13:27:00</td>\n",
       "      <td>00:01:00</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>2016-02-19 13:27:00</td>\n",
       "      <td>718.0</td>\n",
       "      <td>Friday</td>\n",
       "      <td>2016-02-19 13:27:00</td>\n",
       "      <td>00:00:00</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>2016-02-19 13:27:00</td>\n",
       "      <td>NaN</td>\n",
       "      <td>Friday</td>\n",
       "      <td>2016-02-19 13:31:00</td>\n",
       "      <td>00:04:00</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>2016-02-19 13:31:00</td>\n",
       "      <td>337.0</td>\n",
       "      <td>Friday</td>\n",
       "      <td>2016-02-19 13:36:00</td>\n",
       "      <td>00:05:00</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>2016-02-19 13:36:00</td>\n",
       "      <td>332.0</td>\n",
       "      <td>Friday</td>\n",
       "      <td>2016-02-19 13:41:00</td>\n",
       "      <td>00:05:00</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ]
     },
     "execution_count": 816,
     "metadata": {
     },
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df3['TimeDel'] = df3.Time2 - df3.Time\n",
    "df3.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 817,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "1     60.0\n",
       "2      0.0\n",
       "3    240.0\n",
       "4    300.0\n",
       "5    300.0\n",
       "Name: TimeDel, dtype: float64"
      ]
     },
     "execution_count": 817,
     "metadata": {
     },
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df3.TimeDel.dt.seconds.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 818,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Time        datetime64[ns]\n",
       "CO2                float64\n",
       "Day                 object\n",
       "Time2       datetime64[ns]\n",
       "TimeDel    timedelta64[ns]\n",
       "dtype: object"
      ]
     },
     "execution_count": 818,
     "metadata": {
     },
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df3.dtypes"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 819,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
   ],
   "source": [
    "df3['TimeDel'] = df3.TimeDel /  np.timedelta64(1, 's')\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 820,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Time       datetime64[ns]\n",
       "CO2               float64\n",
       "Day                object\n",
       "Time2      datetime64[ns]\n",
       "TimeDel           float64\n",
       "dtype: object"
      ]
     },
     "execution_count": 820,
     "metadata": {
     },
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df3.dtypes"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 821,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
   ],
   "source": [
    "df3['CO2_over_TimeDiff'] = (df3.CO2 / df3.TimeDel)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 822,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Time                 0\n",
       "CO2                  6\n",
       "Day                  0\n",
       "Time2                1\n",
       "TimeDel              1\n",
       "CO2_over_TimeDiff    7\n",
       "dtype: int64"
      ]
     },
     "execution_count": 822,
     "metadata": {
     },
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# number of \"not a number\" in each column\n",
    "df3.isnull().sum()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 823,
   "metadata": {
    "collapsed": false
   },
   "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>Time</th>\n",
       "      <th>CO2</th>\n",
       "      <th>Day</th>\n",
       "      <th>Time2</th>\n",
       "      <th>TimeDel</th>\n",
       "      <th>CO2_over_TimeDiff</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>2016-02-19 13:26:00</td>\n",
       "      <td>NaN</td>\n",
       "      <td>Friday</td>\n",
       "      <td>2016-02-19 13:27:00</td>\n",
       "      <td>60.0</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>2016-02-19 13:27:00</td>\n",
       "      <td>NaN</td>\n",
       "      <td>Friday</td>\n",
       "      <td>2016-02-19 13:31:00</td>\n",
       "      <td>240.0</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2911</th>\n",
       "      <td>2016-02-29 17:03:00</td>\n",
       "      <td>NaN</td>\n",
       "      <td>Monday</td>\n",
       "      <td>2016-02-29 17:05:00</td>\n",
       "      <td>120.0</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>32931</th>\n",
       "      <td>2016-06-14 05:09:00</td>\n",
       "      <td>NaN</td>\n",
       "      <td>Tuesday</td>\n",
       "      <td>2016-06-14 05:10:00</td>\n",
       "      <td>60.0</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>48678</th>\n",
       "      <td>2016-08-09 05:21:00</td>\n",
       "      <td>NaN</td>\n",
       "      <td>Tuesday</td>\n",
       "      <td>2016-08-09 05:22:00</td>\n",
       "      <td>60.0</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>72565</th>\n",
       "      <td>2016-10-31 15:40:00</td>\n",
       "      <td>NaN</td>\n",
       "      <td>Monday</td>\n",
       "      <td>2016-10-31 15:44:00</td>\n",
       "      <td>240.0</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ]
     },
     "execution_count": 823,
     "metadata": {
     },
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df3[df3.CO2.isnull()]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 824,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(100991, 6)"
      ]
     },
     "execution_count": 824,
     "metadata": {
     },
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df3.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 825,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
   ],
   "source": [
    "# dropping rows that have \"any\" missing values\n",
    "df3.dropna(how='any', inplace = True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 826,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(100984, 6)"
      ]
     },
     "execution_count": 826,
     "metadata": {
     },
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df3.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 827,
   "metadata": {
    "collapsed": false
   },
   "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>Time</th>\n",
       "      <th>CO2</th>\n",
       "      <th>Day</th>\n",
       "      <th>Time2</th>\n",
       "      <th>TimeDel</th>\n",
       "      <th>CO2_over_TimeDiff</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>2016-02-19 13:27:00</td>\n",
       "      <td>718.0</td>\n",
       "      <td>Friday</td>\n",
       "      <td>2016-02-19 13:27:00</td>\n",
       "      <td>0.0</td>\n",
       "      <td>inf</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>2016-02-19 13:31:00</td>\n",
       "      <td>337.0</td>\n",
       "      <td>Friday</td>\n",
       "      <td>2016-02-19 13:36:00</td>\n",
       "      <td>300.0</td>\n",
       "      <td>1.123333</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>2016-02-19 13:36:00</td>\n",
       "      <td>332.0</td>\n",
       "      <td>Friday</td>\n",
       "      <td>2016-02-19 13:41:00</td>\n",
       "      <td>300.0</td>\n",
       "      <td>1.106667</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>2016-02-19 13:41:00</td>\n",
       "      <td>328.0</td>\n",
       "      <td>Friday</td>\n",
       "      <td>2016-02-19 13:46:00</td>\n",
       "      <td>300.0</td>\n",
       "      <td>1.093333</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>2016-02-19 13:46:00</td>\n",
       "      <td>307.0</td>\n",
       "      <td>Friday</td>\n",
       "      <td>2016-02-19 13:51:00</td>\n",
       "      <td>300.0</td>\n",
       "      <td>1.023333</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ]
     },
     "execution_count": 827,
     "metadata": {
     },
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df3.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 828,
   "metadata": {
    "collapsed": false
   },
   "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>CO2</th>\n",
       "      <th>TimeDel</th>\n",
       "      <th>CO2_over_TimeDiff</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>count</th>\n",
       "      <td>100984.000000</td>\n",
       "      <td>100984.000000</td>\n",
       "      <td>1.009840e+05</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>mean</th>\n",
       "      <td>547.647548</td>\n",
       "      <td>307.336608</td>\n",
       "      <td>inf</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>std</th>\n",
       "      <td>304.512740</td>\n",
       "      <td>1369.027580</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>min</th>\n",
       "      <td>201.000000</td>\n",
       "      <td>-3300.000000</td>\n",
       "      <td>-3.500000e-01</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>25%</th>\n",
       "      <td>362.000000</td>\n",
       "      <td>300.000000</td>\n",
       "      <td>1.200000e+00</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>50%</th>\n",
       "      <td>438.000000</td>\n",
       "      <td>300.000000</td>\n",
       "      <td>1.453333e+00</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>75%</th>\n",
       "      <td>615.000000</td>\n",
       "      <td>300.000000</td>\n",
       "      <td>2.040000e+00</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>max</th>\n",
       "      <td>2777.000000</td>\n",
       "      <td>424320.000000</td>\n",
       "      <td>inf</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ]
     },
     "execution_count": 828,
     "metadata": {
     },
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df3.describe()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 829,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "2         inf\n",
       "4    1.123333\n",
       "5    1.106667\n",
       "6    1.093333\n",
       "7    1.023333\n",
       "Name: CO2_over_TimeDiff, dtype: float64"
      ]
     },
     "execution_count": 829,
     "metadata": {
     },
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df3.CO2_over_TimeDiff.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 0,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
   ],
   "source": [
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3 (Anaconda)",
   "language": "python",
   "name": "anaconda3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.5.4"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 0
}