/gist:137e0bd2a813decb1bfc Secret

## gistfile1.txt
{
 "metadata": {
  "name": ""
 },
 "nbformat": 3,
 "nbformat_minor": 0,
 "worksheets": [
  {
   "cells": [
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "import pandas as pd\n",
      "import json\n",
      "import numpy\n",
      "from IPython.display import display, HTML"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [],
     "prompt_number": 90
    },
    {
     "cell_type": "markdown",
     "metadata": {},
     "source": [
      "First I load some data that I downloaded from pressureNet (using [this script](https://github.com/NatalieBlack/pressure)).  I had downloaded the data (which is all readings from the Toronto area from June 1 - Sept 11, 2013) in smaller batches, and so have to load the JSON files as separate dataframes and concatenate them."
     ]
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "df0 = pd.read_json('file://localhost/home/natalie/pressurenet/data/June-01-2013:00:00:00_June-30-2013:00:00:00_43-44_-80--79.json', convert_dates=['daterecorded'])\n",
      "df1 = pd.read_json('file://localhost/home/natalie/pressurenet/data/July-01-2013:00:00:00_July-31-2013:00:00:00_43-44_-80--79.json', convert_dates=['daterecorded'])\n",
      "df2 = pd.read_json('file://localhost/home/natalie/pressurenet/data/August-01-2013:00:00:00_August-31-2013:00:00:00_43-44_-80--79.json', convert_dates=['daterecorded'])\n",
      "df3 = pd.read_json('file://localhost/home/natalie/pressurenet/data/September-01-2013:00:00:00_September-07-2013:00:00:00_43-44_-80--79.json', convert_dates=['daterecorded'])\n",
      "df4 = pd.read_json('file://localhost/home/natalie/pressurenet/data/September-07-2013:00:00:00_September-11-2013:00:00:00_43-44_-80--79.json', convert_dates=['daterecorded'])\n",
      "pn = pd.concat([df0, df1, df2, df3, df4])\n",
      "pn[:5]"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [
      {
       "html": [
        "<div style=\"max-height:1000px;max-width:1500px;overflow:auto;\">\n",
        "<table border=\"1\" class=\"dataframe\">\n",
        "  <thead>\n",
        "    <tr style=\"text-align: right;\">\n",
        "      <th></th>\n",
        "      <th>altitude</th>\n",
        "      <th>client_key</th>\n",
        "      <th>daterecorded</th>\n",
        "      <th>latitude</th>\n",
        "      <th>location_accuracy</th>\n",
        "      <th>longitude</th>\n",
        "      <th>observation_type</th>\n",
        "      <th>observation_unit</th>\n",
        "      <th>provider</th>\n",
        "      <th>reading</th>\n",
        "      <th>reading_accuracy</th>\n",
        "      <th>sharing</th>\n",
        "      <th>tzoffset</th>\n",
        "      <th>user_id</th>\n",
        "    </tr>\n",
        "  </thead>\n",
        "  <tbody>\n",
        "    <tr>\n",
        "      <th>0</th>\n",
        "      <td> 0</td>\n",
        "      <td> ca.cumulonimbus.barometernetwork</td>\n",
        "      <td>2013-06-01 05:00:04.680000</td>\n",
        "      <td> 43.736296</td>\n",
        "      <td> 43.867</td>\n",
        "      <td>-79.417653</td>\n",
        "      <td> </td>\n",
        "      <td> </td>\n",
        "      <td> </td>\n",
        "      <td> 989.306274</td>\n",
        "      <td> 0</td>\n",
        "      <td>                          Public</td>\n",
        "      <td>-14400000</td>\n",
        "      <td>   e1551a39d3b6ec399f36acbb17beac</td>\n",
        "    </tr>\n",
        "    <tr>\n",
        "      <th>1</th>\n",
        "      <td> 0</td>\n",
        "      <td> ca.cumulonimbus.barometernetwork</td>\n",
        "      <td>2013-06-01 05:00:54.448000</td>\n",
        "      <td> 43.694750</td>\n",
        "      <td> 20.000</td>\n",
        "      <td>-79.291521</td>\n",
        "      <td> </td>\n",
        "      <td> </td>\n",
        "      <td> </td>\n",
        "      <td> 987.429993</td>\n",
        "      <td> 0</td>\n",
        "      <td> Us, Researchers and Forecasters</td>\n",
        "      <td>-14400000</td>\n",
        "      <td>   334bcef2adeb9b7c39ac26d495167f</td>\n",
        "    </tr>\n",
        "    <tr>\n",
        "      <th>2</th>\n",
        "      <td> 0</td>\n",
        "      <td> ca.cumulonimbus.barometernetwork</td>\n",
        "      <td>2013-06-01 05:01:04.736000</td>\n",
        "      <td> 43.736296</td>\n",
        "      <td> 43.867</td>\n",
        "      <td>-79.417653</td>\n",
        "      <td> </td>\n",
        "      <td> </td>\n",
        "      <td> </td>\n",
        "      <td> 989.026428</td>\n",
        "      <td> 0</td>\n",
        "      <td>                          Public</td>\n",
        "      <td>-14400000</td>\n",
        "      <td>   e1551a39d3b6ec399f36acbb17beac</td>\n",
        "    </tr>\n",
        "    <tr>\n",
        "      <th>3</th>\n",
        "      <td> 0</td>\n",
        "      <td> ca.cumulonimbus.barometernetwork</td>\n",
        "      <td>2013-06-01 05:01:18.016000</td>\n",
        "      <td> 43.231558</td>\n",
        "      <td> 42.000</td>\n",
        "      <td>-79.853251</td>\n",
        "      <td> </td>\n",
        "      <td> </td>\n",
        "      <td> </td>\n",
        "      <td> 988.316772</td>\n",
        "      <td> 0</td>\n",
        "      <td> Us, Researchers and Forecasters</td>\n",
        "      <td>-14400000</td>\n",
        "      <td> 859545c4ccb81384621a7b4b1baf3538</td>\n",
        "    </tr>\n",
        "    <tr>\n",
        "      <th>4</th>\n",
        "      <td> 0</td>\n",
        "      <td> ca.cumulonimbus.barometernetwork</td>\n",
        "      <td>2013-06-01 05:02:04.814000</td>\n",
        "      <td> 43.736296</td>\n",
        "      <td> 43.867</td>\n",
        "      <td>-79.417653</td>\n",
        "      <td> </td>\n",
        "      <td> </td>\n",
        "      <td> </td>\n",
        "      <td> 989.396240</td>\n",
        "      <td> 0</td>\n",
        "      <td>                          Public</td>\n",
        "      <td>-14400000</td>\n",
        "      <td>   e1551a39d3b6ec399f36acbb17beac</td>\n",
        "    </tr>\n",
        "  </tbody>\n",
        "</table>\n",
        "<p>5 rows \u00d7 14 columns</p>\n",
        "</div>"
       ],
       "metadata": {},
       "output_type": "pyout",
       "prompt_number": 91,
       "text": [
        "   altitude                        client_key               daterecorded  \\\n",
        "0         0  ca.cumulonimbus.barometernetwork 2013-06-01 05:00:04.680000   \n",
        "1         0  ca.cumulonimbus.barometernetwork 2013-06-01 05:00:54.448000   \n",
        "2         0  ca.cumulonimbus.barometernetwork 2013-06-01 05:01:04.736000   \n",
        "3         0  ca.cumulonimbus.barometernetwork 2013-06-01 05:01:18.016000   \n",
        "4         0  ca.cumulonimbus.barometernetwork 2013-06-01 05:02:04.814000   \n",
        "\n",
        "    latitude  location_accuracy  longitude observation_type observation_unit  \\\n",
        "0  43.736296             43.867 -79.417653                                     \n",
        "1  43.694750             20.000 -79.291521                                     \n",
        "2  43.736296             43.867 -79.417653                                     \n",
        "3  43.231558             42.000 -79.853251                                     \n",
        "4  43.736296             43.867 -79.417653                                     \n",
        "\n",
        "  provider     reading  reading_accuracy                          sharing  \\\n",
        "0           989.306274                 0                           Public   \n",
        "1           987.429993                 0  Us, Researchers and Forecasters   \n",
        "2           989.026428                 0                           Public   \n",
        "3           988.316772                 0  Us, Researchers and Forecasters   \n",
        "4           989.396240                 0                           Public   \n",
        "\n",
        "   tzoffset                           user_id  \n",
        "0 -14400000    e1551a39d3b6ec399f36acbb17beac  \n",
        "1 -14400000    334bcef2adeb9b7c39ac26d495167f  \n",
        "2 -14400000    e1551a39d3b6ec399f36acbb17beac  \n",
        "3 -14400000  859545c4ccb81384621a7b4b1baf3538  \n",
        "4 -14400000    e1551a39d3b6ec399f36acbb17beac  \n",
        "\n",
        "[5 rows x 14 columns]"
       ]
      }
     ],
     "prompt_number": 91
    },
    {
     "cell_type": "markdown",
     "metadata": {},
     "source": [
      "To get an initial look at the data I decided to plot the atmospheric pressure (the \"reading\" field) against the date (\"daterecorded\")."
     ]
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "pn.plot(x='daterecorded', y='reading', figsize=(15, 6))"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [
      {
       "metadata": {},
       "output_type": "pyout",
       "prompt_number": 92,
       "text": [
        "<matplotlib.axes.AxesSubplot at 0x314a0f90>"
       ]
      },
      {
       "metadata": {},
       "output_type": "display_data",
       "png": "iVBORw0KGgoAAAANSUhEUgAAA28AAAF6CAYAAACOS2D9AAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\n",
       "text": [
        "<matplotlib.figure.Figure at 0x6f491950>"
       ]
      }
     ],
     "prompt_number": 92
    },
    {
     "cell_type": "markdown",
     "metadata": {},
     "source": [
      "It looks like there are some outlying data points where the pressure is close to zero.  I don't know a lot about the atmosphere, but I don't think those were accurate readings.  There also seems to be a few periods of time for which there are far fewer recordings.  There isn't much I can do about that issue, but for the first issue I can try to remove statistical outliers."
     ]
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "pn = pn[(pn['reading'] > 900) & (pn['reading'] < 1100)]"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [],
     "prompt_number": 361
    },
    {
     "cell_type": "markdown",
     "metadata": {},
     "source": [
      "I settled for filtering out rows with a reading of less than 900 or more than 1100.  The resulting plot looks better:"
     ]
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "pn.plot(x='daterecorded', y='reading', figsize=(15, 6))"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [
      {
       "metadata": {},
       "output_type": "pyout",
       "prompt_number": 362,
       "text": [
        "<matplotlib.axes.AxesSubplot at 0x70290250>"
       ]
      },
      {
       "metadata": {},
       "output_type": "display_data",
       "png": "iVBORw0KGgoAAAANSUhEUgAAA28AAAF6CAYAAACOS2D9AAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\n3oQQQgghhBCiDPh/nU3bjTw2raMAAAAASUVORK5CYII=\n",
       "text": [
        "<matplotlib.figure.Figure at 0x70290ed0>"
       ]
      }
     ],
     "prompt_number": 362
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "heat = pd.read_json('file://localhost/home/natalie/torontodata/heat_alerts_list.json', convert_dates=['date'])\n",
      "print heat.code.unique()\n",
      "print heat.text.unique()\n",
      "heat[:5]"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [
      {
       "output_type": "stream",
       "stream": "stdout",
       "text": [
        "[u'HAE' u'HA' u'EHAE' u'HAU' u'EHAD' u'EHA']\n",
        "[u\"Toronto's Medical Officer of Health has extended the Heat Alert\"\n",
        " u\"Toronto's Medical Officer of Health has issued a Heat Alert\"\n",
        " u\"Toronto's Medical Officer of Health has extended the Extreme Heat Alert\"\n",
        " u\"Toronto's Medical Officer of Health has upgraded the Heat Alert to an Extreme Heat Alert\"\n",
        " u\"Toronto's Medical Officer of Health has downgraded the Extreme Heat Alert to a Heat Alert\"\n",
        " u\"Toronto's Medical Officer of Health has issued an Extreme Heat Alert\"]\n"
       ]
      },
      {
       "html": [
        "<div style=\"max-height:1000px;max-width:1500px;overflow:auto;\">\n",
        "<table border=\"1\" class=\"dataframe\">\n",
        "  <thead>\n",
        "    <tr style=\"text-align: right;\">\n",
        "      <th></th>\n",
        "      <th>code</th>\n",
        "      <th>date</th>\n",
        "      <th>id</th>\n",
        "      <th>text</th>\n",
        "    </tr>\n",
        "  </thead>\n",
        "  <tbody>\n",
        "    <tr>\n",
        "      <th>0</th>\n",
        "      <td>  HAE</td>\n",
        "      <td>2013-09-11</td>\n",
        "      <td> 185</td>\n",
        "      <td> Toronto's Medical Officer of Health has extend...</td>\n",
        "    </tr>\n",
        "    <tr>\n",
        "      <th>1</th>\n",
        "      <td>   HA</td>\n",
        "      <td>2013-09-10</td>\n",
        "      <td> 184</td>\n",
        "      <td> Toronto's Medical Officer of Health has issued...</td>\n",
        "    </tr>\n",
        "    <tr>\n",
        "      <th>2</th>\n",
        "      <td> EHAE</td>\n",
        "      <td>2013-07-19</td>\n",
        "      <td> 182</td>\n",
        "      <td> Toronto's Medical Officer of Health has extend...</td>\n",
        "    </tr>\n",
        "    <tr>\n",
        "      <th>3</th>\n",
        "      <td> EHAE</td>\n",
        "      <td>2013-07-18</td>\n",
        "      <td> 181</td>\n",
        "      <td> Toronto's Medical Officer of Health has extend...</td>\n",
        "    </tr>\n",
        "    <tr>\n",
        "      <th>4</th>\n",
        "      <td> EHAE</td>\n",
        "      <td>2013-07-17</td>\n",
        "      <td> 180</td>\n",
        "      <td> Toronto's Medical Officer of Health has extend...</td>\n",
        "    </tr>\n",
        "  </tbody>\n",
        "</table>\n",
        "<p>5 rows \u00d7 4 columns</p>\n",
        "</div>"
       ],
       "metadata": {},
       "output_type": "pyout",
       "prompt_number": 95,
       "text": [
        "   code       date   id                                               text\n",
        "0   HAE 2013-09-11  185  Toronto's Medical Officer of Health has extend...\n",
        "1    HA 2013-09-10  184  Toronto's Medical Officer of Health has issued...\n",
        "2  EHAE 2013-07-19  182  Toronto's Medical Officer of Health has extend...\n",
        "3  EHAE 2013-07-18  181  Toronto's Medical Officer of Health has extend...\n",
        "4  EHAE 2013-07-17  180  Toronto's Medical Officer of Health has extend...\n",
        "\n",
        "[5 rows x 4 columns]"
       ]
      }
     ],
     "prompt_number": 95
    },
    {
     "cell_type": "markdown",
     "metadata": {},
     "source": [
      "I load the list of heat warnings, which I downloaded from Toronto's open data portal [here](http://www1.toronto.ca/wps/portal/contentonly?vgnextoid=b95f614ba4353410VgnVCM10000071d60f89RCRD&vgnextchannel=1a66e03bb8d1e310VgnVCM10000071d60f89RCRD), and take a look at the data.\n",
      "\n",
      "Next I want to join the two dataframes on date, but my pressureNet data has datetimes and the heat data has just date.  I decide to add a column to the pressureNet dataframe with just the date, in the same format as the heat dataframe's date column."
     ]
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "heat.dtypes"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [
      {
       "metadata": {},
       "output_type": "pyout",
       "prompt_number": 96,
       "text": [
        "code            object\n",
        "date    datetime64[ns]\n",
        "id               int64\n",
        "text            object\n",
        "dtype: object"
       ]
      }
     ],
     "prompt_number": 96
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "import arrow\n",
      "pn[\"date\"] = pn.apply(lambda x: str(arrow.get(\"T\".join(str(x['daterecorded']).split(\" \"))).date()), axis=1).astype(datetime64)"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [],
     "prompt_number": 97
    },
    {
     "cell_type": "markdown",
     "metadata": {},
     "source": [
      "To do this I use the arrow library.  I build an arrow object out of the datetime string from each row from the pressureNet dataframe, get just the date part of that arrow object, cast it to a string, and then to a datetime64 dtype.  I don't think this was the nicest way to accomplish my goal, but it worked:"
     ]
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "pn[:1]"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [
      {
       "html": [
        "<div style=\"max-height:1000px;max-width:1500px;overflow:auto;\">\n",
        "<table border=\"1\" class=\"dataframe\">\n",
        "  <thead>\n",
        "    <tr style=\"text-align: right;\">\n",
        "      <th></th>\n",
        "      <th>altitude</th>\n",
        "      <th>client_key</th>\n",
        "      <th>daterecorded</th>\n",
        "      <th>latitude</th>\n",
        "      <th>location_accuracy</th>\n",
        "      <th>longitude</th>\n",
        "      <th>observation_type</th>\n",
        "      <th>observation_unit</th>\n",
        "      <th>provider</th>\n",
        "      <th>reading</th>\n",
        "      <th>reading_accuracy</th>\n",
        "      <th>sharing</th>\n",
        "      <th>tzoffset</th>\n",
        "      <th>user_id</th>\n",
        "      <th>date</th>\n",
        "    </tr>\n",
        "  </thead>\n",
        "  <tbody>\n",
        "    <tr>\n",
        "      <th>0</th>\n",
        "      <td> 0</td>\n",
        "      <td> ca.cumulonimbus.barometernetwork</td>\n",
        "      <td>2013-06-01 05:00:04.680000</td>\n",
        "      <td> 43.736296</td>\n",
        "      <td> 43.867</td>\n",
        "      <td>-79.417653</td>\n",
        "      <td> </td>\n",
        "      <td> </td>\n",
        "      <td> </td>\n",
        "      <td> 989.306274</td>\n",
        "      <td> 0</td>\n",
        "      <td> Public</td>\n",
        "      <td>-14400000</td>\n",
        "      <td> e1551a39d3b6ec399f36acbb17beac</td>\n",
        "      <td>2013-06-01</td>\n",
        "    </tr>\n",
        "  </tbody>\n",
        "</table>\n",
        "<p>1 rows \u00d7 15 columns</p>\n",
        "</div>"
       ],
       "metadata": {},
       "output_type": "pyout",
       "prompt_number": 98,
       "text": [
        "   altitude                        client_key               daterecorded  \\\n",
        "0         0  ca.cumulonimbus.barometernetwork 2013-06-01 05:00:04.680000   \n",
        "\n",
        "    latitude  location_accuracy  longitude observation_type observation_unit  \\\n",
        "0  43.736296             43.867 -79.417653                                     \n",
        "\n",
        "  provider     reading  reading_accuracy sharing  tzoffset  \\\n",
        "0           989.306274                 0  Public -14400000   \n",
        "\n",
        "                          user_id       date  \n",
        "0  e1551a39d3b6ec399f36acbb17beac 2013-06-01  \n",
        "\n",
        "[1 rows x 15 columns]"
       ]
      }
     ],
     "prompt_number": 98
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "tor = pd.merge(pn, heat, on='date', suffixes=('_p','_h'), how='left')\n",
      "tor[:5]"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [
      {
       "html": [
        "<div style=\"max-height:1000px;max-width:1500px;overflow:auto;\">\n",
        "<table border=\"1\" class=\"dataframe\">\n",
        "  <thead>\n",
        "    <tr style=\"text-align: right;\">\n",
        "      <th></th>\n",
        "      <th>altitude</th>\n",
        "      <th>client_key</th>\n",
        "      <th>daterecorded</th>\n",
        "      <th>latitude</th>\n",
        "      <th>location_accuracy</th>\n",
        "      <th>longitude</th>\n",
        "      <th>observation_type</th>\n",
        "      <th>observation_unit</th>\n",
        "      <th>provider</th>\n",
        "      <th>reading</th>\n",
        "      <th>reading_accuracy</th>\n",
        "      <th>sharing</th>\n",
        "      <th>tzoffset</th>\n",
        "      <th>user_id</th>\n",
        "      <th>date</th>\n",
        "      <th>code</th>\n",
        "      <th>id</th>\n",
        "      <th>text</th>\n",
        "    </tr>\n",
        "  </thead>\n",
        "  <tbody>\n",
        "    <tr>\n",
        "      <th>0</th>\n",
        "      <td> 0</td>\n",
        "      <td> ca.cumulonimbus.barometernetwork</td>\n",
        "      <td>2013-06-01 05:00:04.680000</td>\n",
        "      <td> 43.736296</td>\n",
        "      <td> 43.867</td>\n",
        "      <td>-79.417653</td>\n",
        "      <td> </td>\n",
        "      <td> </td>\n",
        "      <td> </td>\n",
        "      <td> 989.306274</td>\n",
        "      <td> 0</td>\n",
        "      <td>                          Public</td>\n",
        "      <td>-14400000</td>\n",
        "      <td>   e1551a39d3b6ec399f36acbb17beac</td>\n",
        "      <td>2013-06-01</td>\n",
        "      <td> NaN</td>\n",
        "      <td>NaN</td>\n",
        "      <td> NaN</td>\n",
        "    </tr>\n",
        "    <tr>\n",
        "      <th>1</th>\n",
        "      <td> 0</td>\n",
        "      <td> ca.cumulonimbus.barometernetwork</td>\n",
        "      <td>2013-06-01 05:00:54.448000</td>\n",
        "      <td> 43.694750</td>\n",
        "      <td> 20.000</td>\n",
        "      <td>-79.291521</td>\n",
        "      <td> </td>\n",
        "      <td> </td>\n",
        "      <td> </td>\n",
        "      <td> 987.429993</td>\n",
        "      <td> 0</td>\n",
        "      <td> Us, Researchers and Forecasters</td>\n",
        "      <td>-14400000</td>\n",
        "      <td>   334bcef2adeb9b7c39ac26d495167f</td>\n",
        "      <td>2013-06-01</td>\n",
        "      <td> NaN</td>\n",
        "      <td>NaN</td>\n",
        "      <td> NaN</td>\n",
        "    </tr>\n",
        "    <tr>\n",
        "      <th>2</th>\n",
        "      <td> 0</td>\n",
        "      <td> ca.cumulonimbus.barometernetwork</td>\n",
        "      <td>2013-06-01 05:01:04.736000</td>\n",
        "      <td> 43.736296</td>\n",
        "      <td> 43.867</td>\n",
        "      <td>-79.417653</td>\n",
        "      <td> </td>\n",
        "      <td> </td>\n",
        "      <td> </td>\n",
        "      <td> 989.026428</td>\n",
        "      <td> 0</td>\n",
        "      <td>                          Public</td>\n",
        "      <td>-14400000</td>\n",
        "      <td>   e1551a39d3b6ec399f36acbb17beac</td>\n",
        "      <td>2013-06-01</td>\n",
        "      <td> NaN</td>\n",
        "      <td>NaN</td>\n",
        "      <td> NaN</td>\n",
        "    </tr>\n",
        "    <tr>\n",
        "      <th>3</th>\n",
        "      <td> 0</td>\n",
        "      <td> ca.cumulonimbus.barometernetwork</td>\n",
        "      <td>2013-06-01 05:01:18.016000</td>\n",
        "      <td> 43.231558</td>\n",
        "      <td> 42.000</td>\n",
        "      <td>-79.853251</td>\n",
        "      <td> </td>\n",
        "      <td> </td>\n",
        "      <td> </td>\n",
        "      <td> 988.316772</td>\n",
        "      <td> 0</td>\n",
        "      <td> Us, Researchers and Forecasters</td>\n",
        "      <td>-14400000</td>\n",
        "      <td> 859545c4ccb81384621a7b4b1baf3538</td>\n",
        "      <td>2013-06-01</td>\n",
        "      <td> NaN</td>\n",
        "      <td>NaN</td>\n",
        "      <td> NaN</td>\n",
        "    </tr>\n",
        "    <tr>\n",
        "      <th>4</th>\n",
        "      <td> 0</td>\n",
        "      <td> ca.cumulonimbus.barometernetwork</td>\n",
        "      <td>2013-06-01 05:02:04.814000</td>\n",
        "      <td> 43.736296</td>\n",
        "      <td> 43.867</td>\n",
        "      <td>-79.417653</td>\n",
        "      <td> </td>\n",
        "      <td> </td>\n",
        "      <td> </td>\n",
        "      <td> 989.396240</td>\n",
        "      <td> 0</td>\n",
        "      <td>                          Public</td>\n",
        "      <td>-14400000</td>\n",
        "      <td>   e1551a39d3b6ec399f36acbb17beac</td>\n",
        "      <td>2013-06-01</td>\n",
        "      <td> NaN</td>\n",
        "      <td>NaN</td>\n",
        "      <td> NaN</td>\n",
        "    </tr>\n",
        "  </tbody>\n",
        "</table>\n",
        "<p>5 rows \u00d7 18 columns</p>\n",
        "</div>"
       ],
       "metadata": {},
       "output_type": "pyout",
       "prompt_number": 99,
       "text": [
        "   altitude                        client_key               daterecorded  \\\n",
        "0         0  ca.cumulonimbus.barometernetwork 2013-06-01 05:00:04.680000   \n",
        "1         0  ca.cumulonimbus.barometernetwork 2013-06-01 05:00:54.448000   \n",
        "2         0  ca.cumulonimbus.barometernetwork 2013-06-01 05:01:04.736000   \n",
        "3         0  ca.cumulonimbus.barometernetwork 2013-06-01 05:01:18.016000   \n",
        "4         0  ca.cumulonimbus.barometernetwork 2013-06-01 05:02:04.814000   \n",
        "\n",
        "    latitude  location_accuracy  longitude observation_type observation_unit  \\\n",
        "0  43.736296             43.867 -79.417653                                     \n",
        "1  43.694750             20.000 -79.291521                                     \n",
        "2  43.736296             43.867 -79.417653                                     \n",
        "3  43.231558             42.000 -79.853251                                     \n",
        "4  43.736296             43.867 -79.417653                                     \n",
        "\n",
        "  provider     reading  reading_accuracy                          sharing  \\\n",
        "0           989.306274                 0                           Public   \n",
        "1           987.429993                 0  Us, Researchers and Forecasters   \n",
        "2           989.026428                 0                           Public   \n",
        "3           988.316772                 0  Us, Researchers and Forecasters   \n",
        "4           989.396240                 0                           Public   \n",
        "\n",
        "   tzoffset                           user_id       date code  id text  \n",
        "0 -14400000    e1551a39d3b6ec399f36acbb17beac 2013-06-01  NaN NaN  NaN  \n",
        "1 -14400000    334bcef2adeb9b7c39ac26d495167f 2013-06-01  NaN NaN  NaN  \n",
        "2 -14400000    e1551a39d3b6ec399f36acbb17beac 2013-06-01  NaN NaN  NaN  \n",
        "3 -14400000  859545c4ccb81384621a7b4b1baf3538 2013-06-01  NaN NaN  NaN  \n",
        "4 -14400000    e1551a39d3b6ec399f36acbb17beac 2013-06-01  NaN NaN  NaN  \n",
        "\n",
        "[5 rows x 18 columns]"
       ]
      }
     ],
     "prompt_number": 99
    },
    {
     "cell_type": "markdown",
     "metadata": {},
     "source": [
      "I perform the merge.  From the first five rows there aren't any rows with heat warning data, but there were far more rows in the pressureNet dataframe than the heat one.  I check to see if the 'text' field is always NaN:"
     ]
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "tor.text.unique()"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [
      {
       "metadata": {},
       "output_type": "pyout",
       "prompt_number": 100,
       "text": [
        "array([nan, u\"Toronto's Medical Officer of Health has issued a Heat Alert\",\n",
        "       u\"Toronto's Medical Officer of Health has upgraded the Heat Alert to an Extreme Heat Alert\",\n",
        "       u\"Toronto's Medical Officer of Health has extended the Extreme Heat Alert\",\n",
        "       u\"Toronto's Medical Officer of Health has downgraded the Extreme Heat Alert to a Heat Alert\",\n",
        "       u\"Toronto's Medical Officer of Health has extended the Heat Alert\"], dtype=object)"
       ]
      }
     ],
     "prompt_number": 100
    },
    {
     "cell_type": "markdown",
     "metadata": {},
     "source": [
      "And luckily it isn't!\n",
      "\n",
      "Next I want to be able to measure the correlation between the text/code (there is a one-to-one mapping here, so I just go with the code field) from the heat warning data and the readings from the pressureNet data.  Pandas can only calculate correlation on numeric fields, and my codes are strings, so I decide to add a column of numeric codes with a one-to-one mapping with the existing string codes."
     ]
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "import numpy as np\n",
      "unique_vals, tor['code_no'] = np.unique(tor.code, return_inverse=True)\n",
      "tor['code_no'].unique()"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [
      {
       "metadata": {},
       "output_type": "pyout",
       "prompt_number": 101,
       "text": [
        "array([0, 3, 5, 2, 1, 4])"
       ]
      }
     ],
     "prompt_number": 101
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "tor[tor['code_no'] == 1][:1]['text']"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [
      {
       "metadata": {},
       "output_type": "pyout",
       "prompt_number": 375,
       "text": [
        "144016    Toronto's Medical Officer of Health has downgr...\n",
        "Name: text, dtype: object"
       ]
      }
     ],
     "prompt_number": 375
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "tor[tor['code_no'] == 2][:1]['text']"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [
      {
       "metadata": {},
       "output_type": "pyout",
       "prompt_number": 376,
       "text": [
        "138462    Toronto's Medical Officer of Health has extend...\n",
        "Name: text, dtype: object"
       ]
      }
     ],
     "prompt_number": 376
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "tor[tor['code_no'] == 3][:1]['text']"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [
      {
       "metadata": {},
       "output_type": "pyout",
       "prompt_number": 377,
       "text": [
        "126790    Toronto's Medical Officer of Health has issued...\n",
        "Name: text, dtype: object"
       ]
      }
     ],
     "prompt_number": 377
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "tor[tor['code_no'] == 4][:1]['text']"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [
      {
       "metadata": {},
       "output_type": "pyout",
       "prompt_number": 378,
       "text": [
        "573641    Toronto's Medical Officer of Health has extend...\n",
        "Name: text, dtype: object"
       ]
      }
     ],
     "prompt_number": 378
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "tor[tor['code_no'] == 5][:1]['text']"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [
      {
       "metadata": {},
       "output_type": "pyout",
       "prompt_number": 379,
       "text": [
        "132377    Toronto's Medical Officer of Health has upgrad...\n",
        "Name: text, dtype: object"
       ]
      }
     ],
     "prompt_number": 379
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "tor.corr()"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [
      {
       "html": [
        "<div style=\"max-height:1000px;max-width:1500px;overflow:auto;\">\n",
        "<table border=\"1\" class=\"dataframe\">\n",
        "  <thead>\n",
        "    <tr style=\"text-align: right;\">\n",
        "      <th></th>\n",
        "      <th>altitude</th>\n",
        "      <th>latitude</th>\n",
        "      <th>location_accuracy</th>\n",
        "      <th>longitude</th>\n",
        "      <th>reading</th>\n",
        "      <th>reading_accuracy</th>\n",
        "      <th>tzoffset</th>\n",
        "      <th>id</th>\n",
        "      <th>code_no</th>\n",
        "    </tr>\n",
        "  </thead>\n",
        "  <tbody>\n",
        "    <tr>\n",
        "      <th>altitude</th>\n",
        "      <td> 1.000000</td>\n",
        "      <td> 0.007098</td>\n",
        "      <td>-0.021511</td>\n",
        "      <td> 0.018626</td>\n",
        "      <td> 0.016783</td>\n",
        "      <td>NaN</td>\n",
        "      <td> 0.151741</td>\n",
        "      <td>      NaN</td>\n",
        "      <td>-0.012730</td>\n",
        "    </tr>\n",
        "    <tr>\n",
        "      <th>latitude</th>\n",
        "      <td> 0.007098</td>\n",
        "      <td> 1.000000</td>\n",
        "      <td>-0.187085</td>\n",
        "      <td> 0.560309</td>\n",
        "      <td>-0.244301</td>\n",
        "      <td>NaN</td>\n",
        "      <td> 0.037490</td>\n",
        "      <td> 0.008536</td>\n",
        "      <td> 0.004088</td>\n",
        "    </tr>\n",
        "    <tr>\n",
        "      <th>location_accuracy</th>\n",
        "      <td>-0.021511</td>\n",
        "      <td>-0.187085</td>\n",
        "      <td> 1.000000</td>\n",
        "      <td>-0.168497</td>\n",
        "      <td> 0.043525</td>\n",
        "      <td>NaN</td>\n",
        "      <td>-0.034505</td>\n",
        "      <td>-0.006546</td>\n",
        "      <td>-0.003210</td>\n",
        "    </tr>\n",
        "    <tr>\n",
        "      <th>longitude</th>\n",
        "      <td> 0.018626</td>\n",
        "      <td> 0.560309</td>\n",
        "      <td>-0.168497</td>\n",
        "      <td> 1.000000</td>\n",
        "      <td> 0.003294</td>\n",
        "      <td>NaN</td>\n",
        "      <td> 0.131755</td>\n",
        "      <td>-0.011353</td>\n",
        "      <td> 0.001112</td>\n",
        "    </tr>\n",
        "    <tr>\n",
        "      <th>reading</th>\n",
        "      <td> 0.016783</td>\n",
        "      <td>-0.244301</td>\n",
        "      <td> 0.043525</td>\n",
        "      <td> 0.003294</td>\n",
        "      <td> 1.000000</td>\n",
        "      <td>NaN</td>\n",
        "      <td>-0.009200</td>\n",
        "      <td> 0.031602</td>\n",
        "      <td> 0.092197</td>\n",
        "    </tr>\n",
        "    <tr>\n",
        "      <th>reading_accuracy</th>\n",
        "      <td>      NaN</td>\n",
        "      <td>      NaN</td>\n",
        "      <td>      NaN</td>\n",
        "      <td>      NaN</td>\n",
        "      <td>      NaN</td>\n",
        "      <td>NaN</td>\n",
        "      <td>      NaN</td>\n",
        "      <td>      NaN</td>\n",
        "      <td>      NaN</td>\n",
        "    </tr>\n",
        "    <tr>\n",
        "      <th>tzoffset</th>\n",
        "      <td> 0.151741</td>\n",
        "      <td> 0.037490</td>\n",
        "      <td>-0.034505</td>\n",
        "      <td> 0.131755</td>\n",
        "      <td>-0.009200</td>\n",
        "      <td>NaN</td>\n",
        "      <td> 1.000000</td>\n",
        "      <td>-0.123591</td>\n",
        "      <td> 0.000231</td>\n",
        "    </tr>\n",
        "    <tr>\n",
        "      <th>id</th>\n",
        "      <td>      NaN</td>\n",
        "      <td> 0.008536</td>\n",
        "      <td>-0.006546</td>\n",
        "      <td>-0.011353</td>\n",
        "      <td> 0.031602</td>\n",
        "      <td>NaN</td>\n",
        "      <td>-0.123591</td>\n",
        "      <td> 1.000000</td>\n",
        "      <td>-0.162111</td>\n",
        "    </tr>\n",
        "    <tr>\n",
        "      <th>code_no</th>\n",
        "      <td>-0.012730</td>\n",
        "      <td> 0.004088</td>\n",
        "      <td>-0.003210</td>\n",
        "      <td> 0.001112</td>\n",
        "      <td> 0.092197</td>\n",
        "      <td>NaN</td>\n",
        "      <td> 0.000231</td>\n",
        "      <td>-0.162111</td>\n",
        "      <td> 1.000000</td>\n",
        "    </tr>\n",
        "  </tbody>\n",
        "</table>\n",
        "<p>9 rows \u00d7 9 columns</p>\n",
        "</div>"
       ],
       "metadata": {},
       "output_type": "pyout",
       "prompt_number": 102,
       "text": [
        "                   altitude  latitude  location_accuracy  longitude   reading  \\\n",
        "altitude           1.000000  0.007098          -0.021511   0.018626  0.016783   \n",
        "latitude           0.007098  1.000000          -0.187085   0.560309 -0.244301   \n",
        "location_accuracy -0.021511 -0.187085           1.000000  -0.168497  0.043525   \n",
        "longitude          0.018626  0.560309          -0.168497   1.000000  0.003294   \n",
        "reading            0.016783 -0.244301           0.043525   0.003294  1.000000   \n",
        "reading_accuracy        NaN       NaN                NaN        NaN       NaN   \n",
        "tzoffset           0.151741  0.037490          -0.034505   0.131755 -0.009200   \n",
        "id                      NaN  0.008536          -0.006546  -0.011353  0.031602   \n",
        "code_no           -0.012730  0.004088          -0.003210   0.001112  0.092197   \n",
        "\n",
        "                   reading_accuracy  tzoffset        id   code_no  \n",
        "altitude                        NaN  0.151741       NaN -0.012730  \n",
        "latitude                        NaN  0.037490  0.008536  0.004088  \n",
        "location_accuracy               NaN -0.034505 -0.006546 -0.003210  \n",
        "longitude                       NaN  0.131755 -0.011353  0.001112  \n",
        "reading                         NaN -0.009200  0.031602  0.092197  \n",
        "reading_accuracy                NaN       NaN       NaN       NaN  \n",
        "tzoffset                        NaN  1.000000 -0.123591  0.000231  \n",
        "id                              NaN -0.123591  1.000000 -0.162111  \n",
        "code_no                         NaN  0.000231 -0.162111  1.000000  \n",
        "\n",
        "[9 rows x 9 columns]"
       ]
      }
     ],
     "prompt_number": 102
    },
    {
     "cell_type": "markdown",
     "metadata": {},
     "source": [
      "The built-in correlation method shows a low correlation between the code_no field and the reading field, which is not what I was hoping I'd find."
     ]
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "warnings = tor[tor['code_no'].isin([1, 2, 3, 4, 5])]\n",
      "warnings2 = tor[tor['code_no'] == 2]\n",
      "warnings1 = tor[tor['code_no'] == 1]\n",
      "warnings3 = tor[tor['code_no'] == 3]\n",
      "warnings4 = tor[tor['code_no'] == 4]\n",
      "warnings5 = tor[tor['code_no'] == 5]\n",
      "nonwarnings = tor[tor['code_no'] == 0]"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [],
     "prompt_number": 365
    },
    {
     "cell_type": "markdown",
     "metadata": {},
     "source": [
      "I isolate the days with any kind of heat warning as well as with each specific type of warning/warning-related message."
     ]
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "tor.describe()"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [
      {
       "html": [
        "<div style=\"max-height:1000px;max-width:1500px;overflow:auto;\">\n",
        "<table border=\"1\" class=\"dataframe\">\n",
        "  <thead>\n",
        "    <tr style=\"text-align: right;\">\n",
        "      <th></th>\n",
        "      <th>altitude</th>\n",
        "      <th>latitude</th>\n",
        "      <th>location_accuracy</th>\n",
        "      <th>longitude</th>\n",
        "      <th>reading</th>\n",
        "      <th>reading_accuracy</th>\n",
        "      <th>tzoffset</th>\n",
        "      <th>id</th>\n",
        "      <th>code_no</th>\n",
        "    </tr>\n",
        "  </thead>\n",
        "  <tbody>\n",
        "    <tr>\n",
        "      <th>count</th>\n",
        "      <td> 574590.000000</td>\n",
        "      <td> 574590.000000</td>\n",
        "      <td> 574590.000000</td>\n",
        "      <td> 574590.000000</td>\n",
        "      <td> 574590.000000</td>\n",
        "      <td> 574590</td>\n",
        "      <td>   574590.000000</td>\n",
        "      <td> 59764.000000</td>\n",
        "      <td> 574590.000000</td>\n",
        "    </tr>\n",
        "    <tr>\n",
        "      <th>mean</th>\n",
        "      <td>      0.352973</td>\n",
        "      <td>     43.640132</td>\n",
        "      <td>    298.515366</td>\n",
        "      <td>    -79.477884</td>\n",
        "      <td>    995.170459</td>\n",
        "      <td>      0</td>\n",
        "      <td>-13416276.301363</td>\n",
        "      <td>   178.327036</td>\n",
        "      <td>      0.294419</td>\n",
        "    </tr>\n",
        "    <tr>\n",
        "      <th>std</th>\n",
        "      <td>      8.555421</td>\n",
        "      <td>      0.180799</td>\n",
        "      <td>    514.602824</td>\n",
        "      <td>      0.184457</td>\n",
        "      <td>      7.652223</td>\n",
        "      <td>      0</td>\n",
        "      <td>  3647785.368813</td>\n",
        "      <td>     3.488297</td>\n",
        "      <td>      0.954214</td>\n",
        "    </tr>\n",
        "    <tr>\n",
        "      <th>min</th>\n",
        "      <td>    -59.100006</td>\n",
        "      <td>     43.000120</td>\n",
        "      <td>      1.000000</td>\n",
        "      <td>    -79.999870</td>\n",
        "      <td>    910.229736</td>\n",
        "      <td>      0</td>\n",
        "      <td>-28800000.000000</td>\n",
        "      <td>   173.000000</td>\n",
        "      <td>      0.000000</td>\n",
        "    </tr>\n",
        "    <tr>\n",
        "      <th>25%</th>\n",
        "      <td>      0.000000</td>\n",
        "      <td>     43.646621</td>\n",
        "      <td>     29.407000</td>\n",
        "      <td>    -79.569317</td>\n",
        "      <td>    989.965881</td>\n",
        "      <td>      0</td>\n",
        "      <td>-14400000.000000</td>\n",
        "      <td>   175.000000</td>\n",
        "      <td>      0.000000</td>\n",
        "    </tr>\n",
        "    <tr>\n",
        "      <th>50%</th>\n",
        "      <td>      0.000000</td>\n",
        "      <td>     43.685568</td>\n",
        "      <td>     40.500000</td>\n",
        "      <td>    -79.417533</td>\n",
        "      <td>    995.190002</td>\n",
        "      <td>      0</td>\n",
        "      <td>-14400000.000000</td>\n",
        "      <td>   179.000000</td>\n",
        "      <td>      0.000000</td>\n",
        "    </tr>\n",
        "    <tr>\n",
        "      <th>75%</th>\n",
        "      <td>      0.000000</td>\n",
        "      <td>     43.736305</td>\n",
        "      <td>    143.839500</td>\n",
        "      <td>    -79.379235</td>\n",
        "      <td>   1000.540100</td>\n",
        "      <td>      0</td>\n",
        "      <td>-14400000.000000</td>\n",
        "      <td>   181.000000</td>\n",
        "      <td>      0.000000</td>\n",
        "    </tr>\n",
        "    <tr>\n",
        "      <th>max</th>\n",
        "      <td>   1000.199951</td>\n",
        "      <td>     43.999995</td>\n",
        "      <td>   5000.000000</td>\n",
        "      <td>    -79.000225</td>\n",
        "      <td>   1344.660034</td>\n",
        "      <td>      0</td>\n",
        "      <td>  7200000.000000</td>\n",
        "      <td>   185.000000</td>\n",
        "      <td>      5.000000</td>\n",
        "    </tr>\n",
        "  </tbody>\n",
        "</table>\n",
        "<p>8 rows \u00d7 9 columns</p>\n",
        "</div>"
       ],
       "metadata": {},
       "output_type": "pyout",
       "prompt_number": 104,
       "text": [
        "            altitude       latitude  location_accuracy      longitude  \\\n",
        "count  574590.000000  574590.000000      574590.000000  574590.000000   \n",
        "mean        0.352973      43.640132         298.515366     -79.477884   \n",
        "std         8.555421       0.180799         514.602824       0.184457   \n",
        "min       -59.100006      43.000120           1.000000     -79.999870   \n",
        "25%         0.000000      43.646621          29.407000     -79.569317   \n",
        "50%         0.000000      43.685568          40.500000     -79.417533   \n",
        "75%         0.000000      43.736305         143.839500     -79.379235   \n",
        "max      1000.199951      43.999995        5000.000000     -79.000225   \n",
        "\n",
        "             reading  reading_accuracy         tzoffset            id  \\\n",
        "count  574590.000000            574590    574590.000000  59764.000000   \n",
        "mean      995.170459                 0 -13416276.301363    178.327036   \n",
        "std         7.652223                 0   3647785.368813      3.488297   \n",
        "min       910.229736                 0 -28800000.000000    173.000000   \n",
        "25%       989.965881                 0 -14400000.000000    175.000000   \n",
        "50%       995.190002                 0 -14400000.000000    179.000000   \n",
        "75%      1000.540100                 0 -14400000.000000    181.000000   \n",
        "max      1344.660034                 0   7200000.000000    185.000000   \n",
        "\n",
        "             code_no  \n",
        "count  574590.000000  \n",
        "mean        0.294419  \n",
        "std         0.954214  \n",
        "min         0.000000  \n",
        "25%         0.000000  \n",
        "50%         0.000000  \n",
        "75%         0.000000  \n",
        "max         5.000000  \n",
        "\n",
        "[8 rows x 9 columns]"
       ]
      }
     ],
     "prompt_number": 104
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "nonwarnings.describe()"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [
      {
       "html": [
        "<div style=\"max-height:1000px;max-width:1500px;overflow:auto;\">\n",
        "<table border=\"1\" class=\"dataframe\">\n",
        "  <thead>\n",
        "    <tr style=\"text-align: right;\">\n",
        "      <th></th>\n",
        "      <th>altitude</th>\n",
        "      <th>latitude</th>\n",
        "      <th>location_accuracy</th>\n",
        "      <th>longitude</th>\n",
        "      <th>reading</th>\n",
        "      <th>reading_accuracy</th>\n",
        "      <th>tzoffset</th>\n",
        "      <th>id</th>\n",
        "      <th>code_no</th>\n",
        "    </tr>\n",
        "  </thead>\n",
        "  <tbody>\n",
        "    <tr>\n",
        "      <th>count</th>\n",
        "      <td> 514826.000000</td>\n",
        "      <td> 514826.000000</td>\n",
        "      <td> 514826.000000</td>\n",
        "      <td> 514826.000000</td>\n",
        "      <td> 514826.000000</td>\n",
        "      <td> 514826</td>\n",
        "      <td>   514826.000000</td>\n",
        "      <td>  0</td>\n",
        "      <td> 514826</td>\n",
        "    </tr>\n",
        "    <tr>\n",
        "      <th>mean</th>\n",
        "      <td>      0.393948</td>\n",
        "      <td>     43.639945</td>\n",
        "      <td>    299.194298</td>\n",
        "      <td>    -79.477633</td>\n",
        "      <td>    995.040635</td>\n",
        "      <td>      0</td>\n",
        "      <td>-13408116.528691</td>\n",
        "      <td>NaN</td>\n",
        "      <td>      0</td>\n",
        "    </tr>\n",
        "    <tr>\n",
        "      <th>std</th>\n",
        "      <td>      9.037479</td>\n",
        "      <td>      0.181041</td>\n",
        "      <td>    514.424446</td>\n",
        "      <td>      0.184219</td>\n",
        "      <td>      7.655763</td>\n",
        "      <td>      0</td>\n",
        "      <td>  3662129.542021</td>\n",
        "      <td>NaN</td>\n",
        "      <td>      0</td>\n",
        "    </tr>\n",
        "    <tr>\n",
        "      <th>min</th>\n",
        "      <td>    -59.100006</td>\n",
        "      <td>     43.000120</td>\n",
        "      <td>      1.000000</td>\n",
        "      <td>    -79.999870</td>\n",
        "      <td>    910.229736</td>\n",
        "      <td>      0</td>\n",
        "      <td>-28800000.000000</td>\n",
        "      <td>NaN</td>\n",
        "      <td>      0</td>\n",
        "    </tr>\n",
        "    <tr>\n",
        "      <th>25%</th>\n",
        "      <td>      0.000000</td>\n",
        "      <td>     43.646545</td>\n",
        "      <td>     29.502250</td>\n",
        "      <td>    -79.569317</td>\n",
        "      <td>    989.890015</td>\n",
        "      <td>      0</td>\n",
        "      <td>-14400000.000000</td>\n",
        "      <td>NaN</td>\n",
        "      <td>      0</td>\n",
        "    </tr>\n",
        "    <tr>\n",
        "      <th>50%</th>\n",
        "      <td>      0.000000</td>\n",
        "      <td>     43.684683</td>\n",
        "      <td>     40.689000</td>\n",
        "      <td>    -79.417533</td>\n",
        "      <td>    995.053101</td>\n",
        "      <td>      0</td>\n",
        "      <td>-14400000.000000</td>\n",
        "      <td>NaN</td>\n",
        "      <td>      0</td>\n",
        "    </tr>\n",
        "    <tr>\n",
        "      <th>75%</th>\n",
        "      <td>      0.000000</td>\n",
        "      <td>     43.736305</td>\n",
        "      <td>    147.578000</td>\n",
        "      <td>    -79.379184</td>\n",
        "      <td>   1000.350220</td>\n",
        "      <td>      0</td>\n",
        "      <td>-14400000.000000</td>\n",
        "      <td>NaN</td>\n",
        "      <td>      0</td>\n",
        "    </tr>\n",
        "    <tr>\n",
        "      <th>max</th>\n",
        "      <td>   1000.199951</td>\n",
        "      <td>     43.999995</td>\n",
        "      <td>   5000.000000</td>\n",
        "      <td>    -79.000415</td>\n",
        "      <td>   1344.660034</td>\n",
        "      <td>      0</td>\n",
        "      <td>  7200000.000000</td>\n",
        "      <td>NaN</td>\n",
        "      <td>      0</td>\n",
        "    </tr>\n",
        "  </tbody>\n",
        "</table>\n",
        "<p>8 rows \u00d7 9 columns</p>\n",
        "</div>"
       ],
       "metadata": {},
       "output_type": "pyout",
       "prompt_number": 105,
       "text": [
        "            altitude       latitude  location_accuracy      longitude  \\\n",
        "count  514826.000000  514826.000000      514826.000000  514826.000000   \n",
        "mean        0.393948      43.639945         299.194298     -79.477633   \n",
        "std         9.037479       0.181041         514.424446       0.184219   \n",
        "min       -59.100006      43.000120           1.000000     -79.999870   \n",
        "25%         0.000000      43.646545          29.502250     -79.569317   \n",
        "50%         0.000000      43.684683          40.689000     -79.417533   \n",
        "75%         0.000000      43.736305         147.578000     -79.379184   \n",
        "max      1000.199951      43.999995        5000.000000     -79.000415   \n",
        "\n",
        "             reading  reading_accuracy         tzoffset  id  code_no  \n",
        "count  514826.000000            514826    514826.000000   0   514826  \n",
        "mean      995.040635                 0 -13408116.528691 NaN        0  \n",
        "std         7.655763                 0   3662129.542021 NaN        0  \n",
        "min       910.229736                 0 -28800000.000000 NaN        0  \n",
        "25%       989.890015                 0 -14400000.000000 NaN        0  \n",
        "50%       995.053101                 0 -14400000.000000 NaN        0  \n",
        "75%      1000.350220                 0 -14400000.000000 NaN        0  \n",
        "max      1344.660034                 0   7200000.000000 NaN        0  \n",
        "\n",
        "[8 rows x 9 columns]"
       ]
      }
     ],
     "prompt_number": 105
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "warnings.describe()"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [
      {
       "html": [
        "<div style=\"max-height:1000px;max-width:1500px;overflow:auto;\">\n",
        "<table border=\"1\" class=\"dataframe\">\n",
        "  <thead>\n",
        "    <tr style=\"text-align: right;\">\n",
        "      <th></th>\n",
        "      <th>altitude</th>\n",
        "      <th>latitude</th>\n",
        "      <th>location_accuracy</th>\n",
        "      <th>longitude</th>\n",
        "      <th>reading</th>\n",
        "      <th>reading_accuracy</th>\n",
        "      <th>tzoffset</th>\n",
        "      <th>id</th>\n",
        "      <th>code_no</th>\n",
        "    </tr>\n",
        "  </thead>\n",
        "  <tbody>\n",
        "    <tr>\n",
        "      <th>count</th>\n",
        "      <td> 29521</td>\n",
        "      <td> 29521.000000</td>\n",
        "      <td> 29521.000000</td>\n",
        "      <td> 29521.000000</td>\n",
        "      <td> 29521.000000</td>\n",
        "      <td> 29521</td>\n",
        "      <td>    29521.000000</td>\n",
        "      <td> 29521.000000</td>\n",
        "      <td> 29521.000000</td>\n",
        "    </tr>\n",
        "    <tr>\n",
        "      <th>mean</th>\n",
        "      <td>     0</td>\n",
        "      <td>    43.643843</td>\n",
        "      <td>   291.787895</td>\n",
        "      <td>   -79.485729</td>\n",
        "      <td>   993.677119</td>\n",
        "      <td>     0</td>\n",
        "      <td>-13734656.685072</td>\n",
        "      <td>   178.916229</td>\n",
        "      <td>     1.807120</td>\n",
        "    </tr>\n",
        "    <tr>\n",
        "      <th>std</th>\n",
        "      <td>     0</td>\n",
        "      <td>     0.179121</td>\n",
        "      <td>   514.789195</td>\n",
        "      <td>     0.188699</td>\n",
        "      <td>     7.315454</td>\n",
        "      <td>     0</td>\n",
        "      <td>  3023152.265227</td>\n",
        "      <td>     2.769632</td>\n",
        "      <td>     0.394566</td>\n",
        "    </tr>\n",
        "    <tr>\n",
        "      <th>min</th>\n",
        "      <td>     0</td>\n",
        "      <td>    43.015414</td>\n",
        "      <td>     4.737000</td>\n",
        "      <td>   -79.995023</td>\n",
        "      <td>   958.433228</td>\n",
        "      <td>     0</td>\n",
        "      <td>-18000000.000000</td>\n",
        "      <td>   175.000000</td>\n",
        "      <td>     1.000000</td>\n",
        "    </tr>\n",
        "    <tr>\n",
        "      <th>25%</th>\n",
        "      <td>     0</td>\n",
        "      <td>    43.646741</td>\n",
        "      <td>    27.095000</td>\n",
        "      <td>   -79.570170</td>\n",
        "      <td>   987.667175</td>\n",
        "      <td>     0</td>\n",
        "      <td>-14400000.000000</td>\n",
        "      <td>   176.000000</td>\n",
        "      <td>     2.000000</td>\n",
        "    </tr>\n",
        "    <tr>\n",
        "      <th>50%</th>\n",
        "      <td>     0</td>\n",
        "      <td>    43.695414</td>\n",
        "      <td>    37.416000</td>\n",
        "      <td>   -79.417536</td>\n",
        "      <td>   994.070007</td>\n",
        "      <td>     0</td>\n",
        "      <td>-14400000.000000</td>\n",
        "      <td>   180.000000</td>\n",
        "      <td>     2.000000</td>\n",
        "    </tr>\n",
        "    <tr>\n",
        "      <th>75%</th>\n",
        "      <td>     0</td>\n",
        "      <td>    43.736307</td>\n",
        "      <td>   111.656000</td>\n",
        "      <td>   -79.380830</td>\n",
        "      <td>   998.940979</td>\n",
        "      <td>     0</td>\n",
        "      <td>-14400000.000000</td>\n",
        "      <td>   181.000000</td>\n",
        "      <td>     2.000000</td>\n",
        "    </tr>\n",
        "    <tr>\n",
        "      <th>max</th>\n",
        "      <td>     0</td>\n",
        "      <td>    43.998816</td>\n",
        "      <td>  3443.000000</td>\n",
        "      <td>   -79.000505</td>\n",
        "      <td>  1012.913269</td>\n",
        "      <td>     0</td>\n",
        "      <td>        0.000000</td>\n",
        "      <td>   182.000000</td>\n",
        "      <td>     2.000000</td>\n",
        "    </tr>\n",
        "  </tbody>\n",
        "</table>\n",
        "<p>8 rows \u00d7 9 columns</p>\n",
        "</div>"
       ],
       "metadata": {},
       "output_type": "pyout",
       "prompt_number": 106,
       "text": [
        "       altitude      latitude  location_accuracy     longitude       reading  \\\n",
        "count     29521  29521.000000       29521.000000  29521.000000  29521.000000   \n",
        "mean          0     43.643843         291.787895    -79.485729    993.677119   \n",
        "std           0      0.179121         514.789195      0.188699      7.315454   \n",
        "min           0     43.015414           4.737000    -79.995023    958.433228   \n",
        "25%           0     43.646741          27.095000    -79.570170    987.667175   \n",
        "50%           0     43.695414          37.416000    -79.417536    994.070007   \n",
        "75%           0     43.736307         111.656000    -79.380830    998.940979   \n",
        "max           0     43.998816        3443.000000    -79.000505   1012.913269   \n",
        "\n",
        "       reading_accuracy         tzoffset            id       code_no  \n",
        "count             29521     29521.000000  29521.000000  29521.000000  \n",
        "mean                  0 -13734656.685072    178.916229      1.807120  \n",
        "std                   0   3023152.265227      2.769632      0.394566  \n",
        "min                   0 -18000000.000000    175.000000      1.000000  \n",
        "25%                   0 -14400000.000000    176.000000      2.000000  \n",
        "50%                   0 -14400000.000000    180.000000      2.000000  \n",
        "75%                   0 -14400000.000000    181.000000      2.000000  \n",
        "max                   0         0.000000    182.000000      2.000000  \n",
        "\n",
        "[8 rows x 9 columns]"
       ]
      }
     ],
     "prompt_number": 106
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "print len(warnings1)\n",
      "warnings1.describe()"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [
      {
       "output_type": "stream",
       "stream": "stdout",
       "text": [
        "5694\n"
       ]
      },
      {
       "html": [
        "<div style=\"max-height:1000px;max-width:1500px;overflow:auto;\">\n",
        "<table border=\"1\" class=\"dataframe\">\n",
        "  <thead>\n",
        "    <tr style=\"text-align: right;\">\n",
        "      <th></th>\n",
        "      <th>altitude</th>\n",
        "      <th>latitude</th>\n",
        "      <th>location_accuracy</th>\n",
        "      <th>longitude</th>\n",
        "      <th>reading</th>\n",
        "      <th>reading_accuracy</th>\n",
        "      <th>tzoffset</th>\n",
        "      <th>id</th>\n",
        "      <th>code_no</th>\n",
        "    </tr>\n",
        "  </thead>\n",
        "  <tbody>\n",
        "    <tr>\n",
        "      <th>count</th>\n",
        "      <td> 5694</td>\n",
        "      <td> 5694.000000</td>\n",
        "      <td> 5694.000000</td>\n",
        "      <td> 5694.000000</td>\n",
        "      <td> 5694.000000</td>\n",
        "      <td> 5694</td>\n",
        "      <td>     5694.000000</td>\n",
        "      <td> 5694</td>\n",
        "      <td> 5694</td>\n",
        "    </tr>\n",
        "    <tr>\n",
        "      <th>mean</th>\n",
        "      <td>    0</td>\n",
        "      <td>   43.632695</td>\n",
        "      <td>  264.226575</td>\n",
        "      <td>  -79.497286</td>\n",
        "      <td>  988.579502</td>\n",
        "      <td>    0</td>\n",
        "      <td>-13047629.083246</td>\n",
        "      <td>  176</td>\n",
        "      <td>    1</td>\n",
        "    </tr>\n",
        "    <tr>\n",
        "      <th>std</th>\n",
        "      <td>    0</td>\n",
        "      <td>    0.180082</td>\n",
        "      <td>  491.716751</td>\n",
        "      <td>    0.185150</td>\n",
        "      <td>    5.397092</td>\n",
        "      <td>    0</td>\n",
        "      <td>  4202346.483297</td>\n",
        "      <td>    0</td>\n",
        "      <td>    0</td>\n",
        "    </tr>\n",
        "    <tr>\n",
        "      <th>min</th>\n",
        "      <td>    0</td>\n",
        "      <td>   43.028590</td>\n",
        "      <td>    4.737000</td>\n",
        "      <td>  -79.979364</td>\n",
        "      <td>  967.937988</td>\n",
        "      <td>    0</td>\n",
        "      <td>-18000000.000000</td>\n",
        "      <td>  176</td>\n",
        "      <td>    1</td>\n",
        "    </tr>\n",
        "    <tr>\n",
        "      <th>25%</th>\n",
        "      <td>    0</td>\n",
        "      <td>   43.646365</td>\n",
        "      <td>   31.245250</td>\n",
        "      <td>  -79.624901</td>\n",
        "      <td>  985.508362</td>\n",
        "      <td>    0</td>\n",
        "      <td>-14400000.000000</td>\n",
        "      <td>  176</td>\n",
        "      <td>    1</td>\n",
        "    </tr>\n",
        "    <tr>\n",
        "      <th>50%</th>\n",
        "      <td>    0</td>\n",
        "      <td>   43.668456</td>\n",
        "      <td>   43.500000</td>\n",
        "      <td>  -79.417529</td>\n",
        "      <td>  986.789978</td>\n",
        "      <td>    0</td>\n",
        "      <td>-14400000.000000</td>\n",
        "      <td>  176</td>\n",
        "      <td>    1</td>\n",
        "    </tr>\n",
        "    <tr>\n",
        "      <th>75%</th>\n",
        "      <td>    0</td>\n",
        "      <td>   43.736316</td>\n",
        "      <td>   86.079500</td>\n",
        "      <td>  -79.379235</td>\n",
        "      <td>  993.769958</td>\n",
        "      <td>    0</td>\n",
        "      <td>-14400000.000000</td>\n",
        "      <td>  176</td>\n",
        "      <td>    1</td>\n",
        "    </tr>\n",
        "    <tr>\n",
        "      <th>max</th>\n",
        "      <td>    0</td>\n",
        "      <td>   43.997440</td>\n",
        "      <td> 2615.000000</td>\n",
        "      <td>  -79.003154</td>\n",
        "      <td> 1000.289978</td>\n",
        "      <td>    0</td>\n",
        "      <td>        0.000000</td>\n",
        "      <td>  176</td>\n",
        "      <td>    1</td>\n",
        "    </tr>\n",
        "  </tbody>\n",
        "</table>\n",
        "<p>8 rows \u00d7 9 columns</p>\n",
        "</div>"
       ],
       "metadata": {},
       "output_type": "pyout",
       "prompt_number": 366,
       "text": [
        "       altitude     latitude  location_accuracy    longitude      reading  \\\n",
        "count      5694  5694.000000        5694.000000  5694.000000  5694.000000   \n",
        "mean          0    43.632695         264.226575   -79.497286   988.579502   \n",
        "std           0     0.180082         491.716751     0.185150     5.397092   \n",
        "min           0    43.028590           4.737000   -79.979364   967.937988   \n",
        "25%           0    43.646365          31.245250   -79.624901   985.508362   \n",
        "50%           0    43.668456          43.500000   -79.417529   986.789978   \n",
        "75%           0    43.736316          86.079500   -79.379235   993.769958   \n",
        "max           0    43.997440        2615.000000   -79.003154  1000.289978   \n",
        "\n",
        "       reading_accuracy         tzoffset    id  code_no  \n",
        "count              5694      5694.000000  5694     5694  \n",
        "mean                  0 -13047629.083246   176        1  \n",
        "std                   0   4202346.483297     0        0  \n",
        "min                   0 -18000000.000000   176        1  \n",
        "25%                   0 -14400000.000000   176        1  \n",
        "50%                   0 -14400000.000000   176        1  \n",
        "75%                   0 -14400000.000000   176        1  \n",
        "max                   0         0.000000   176        1  \n",
        "\n",
        "[8 rows x 9 columns]"
       ]
      }
     ],
     "prompt_number": 366
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "print len(warnings2)\n",
      "warnings2.describe()"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [
      {
       "output_type": "stream",
       "stream": "stdout",
       "text": [
        "23827\n"
       ]
      },
      {
       "html": [
        "<div style=\"max-height:1000px;max-width:1500px;overflow:auto;\">\n",
        "<table border=\"1\" class=\"dataframe\">\n",
        "  <thead>\n",
        "    <tr style=\"text-align: right;\">\n",
        "      <th></th>\n",
        "      <th>altitude</th>\n",
        "      <th>latitude</th>\n",
        "      <th>location_accuracy</th>\n",
        "      <th>longitude</th>\n",
        "      <th>reading</th>\n",
        "      <th>reading_accuracy</th>\n",
        "      <th>tzoffset</th>\n",
        "      <th>id</th>\n",
        "      <th>code_no</th>\n",
        "    </tr>\n",
        "  </thead>\n",
        "  <tbody>\n",
        "    <tr>\n",
        "      <th>count</th>\n",
        "      <td> 23827</td>\n",
        "      <td> 23827.000000</td>\n",
        "      <td> 23827.000000</td>\n",
        "      <td> 23827.000000</td>\n",
        "      <td> 23827.000000</td>\n",
        "      <td> 23827</td>\n",
        "      <td>    23827.000000</td>\n",
        "      <td> 23827.000000</td>\n",
        "      <td> 23827</td>\n",
        "    </tr>\n",
        "    <tr>\n",
        "      <th>mean</th>\n",
        "      <td>     0</td>\n",
        "      <td>    43.646507</td>\n",
        "      <td>   298.374295</td>\n",
        "      <td>   -79.482968</td>\n",
        "      <td>   994.895309</td>\n",
        "      <td>     0</td>\n",
        "      <td>-13898837.453309</td>\n",
        "      <td>   179.613128</td>\n",
        "      <td>     2</td>\n",
        "    </tr>\n",
        "    <tr>\n",
        "      <th>std</th>\n",
        "      <td>     0</td>\n",
        "      <td>     0.178791</td>\n",
        "      <td>   519.945122</td>\n",
        "      <td>     0.189437</td>\n",
        "      <td>     7.186866</td>\n",
        "      <td>     0</td>\n",
        "      <td>  2638984.140779</td>\n",
        "      <td>     2.643099</td>\n",
        "      <td>     0</td>\n",
        "    </tr>\n",
        "    <tr>\n",
        "      <th>min</th>\n",
        "      <td>     0</td>\n",
        "      <td>    43.015414</td>\n",
        "      <td>     4.737000</td>\n",
        "      <td>   -79.995023</td>\n",
        "      <td>   958.433228</td>\n",
        "      <td>     0</td>\n",
        "      <td>-14400000.000000</td>\n",
        "      <td>   175.000000</td>\n",
        "      <td>     2</td>\n",
        "    </tr>\n",
        "    <tr>\n",
        "      <th>25%</th>\n",
        "      <td>     0</td>\n",
        "      <td>    43.646784</td>\n",
        "      <td>    27.000000</td>\n",
        "      <td>   -79.569345</td>\n",
        "      <td>   989.320007</td>\n",
        "      <td>     0</td>\n",
        "      <td>-14400000.000000</td>\n",
        "      <td>   180.000000</td>\n",
        "      <td>     2</td>\n",
        "    </tr>\n",
        "    <tr>\n",
        "      <th>50%</th>\n",
        "      <td>     0</td>\n",
        "      <td>    43.699368</td>\n",
        "      <td>    36.026000</td>\n",
        "      <td>   -79.417536</td>\n",
        "      <td>   995.592773</td>\n",
        "      <td>     0</td>\n",
        "      <td>-14400000.000000</td>\n",
        "      <td>   181.000000</td>\n",
        "      <td>     2</td>\n",
        "    </tr>\n",
        "    <tr>\n",
        "      <th>75%</th>\n",
        "      <td>     0</td>\n",
        "      <td>    43.736303</td>\n",
        "      <td>   128.397000</td>\n",
        "      <td>   -79.381984</td>\n",
        "      <td>   999.920197</td>\n",
        "      <td>     0</td>\n",
        "      <td>-14400000.000000</td>\n",
        "      <td>   182.000000</td>\n",
        "      <td>     2</td>\n",
        "    </tr>\n",
        "    <tr>\n",
        "      <th>max</th>\n",
        "      <td>     0</td>\n",
        "      <td>    43.998816</td>\n",
        "      <td>  3443.000000</td>\n",
        "      <td>   -79.000505</td>\n",
        "      <td>  1012.913269</td>\n",
        "      <td>     0</td>\n",
        "      <td>        0.000000</td>\n",
        "      <td>   182.000000</td>\n",
        "      <td>     2</td>\n",
        "    </tr>\n",
        "  </tbody>\n",
        "</table>\n",
        "<p>8 rows \u00d7 9 columns</p>\n",
        "</div>"
       ],
       "metadata": {},
       "output_type": "pyout",
       "prompt_number": 367,
       "text": [
        "       altitude      latitude  location_accuracy     longitude       reading  \\\n",
        "count     23827  23827.000000       23827.000000  23827.000000  23827.000000   \n",
        "mean          0     43.646507         298.374295    -79.482968    994.895309   \n",
        "std           0      0.178791         519.945122      0.189437      7.186866   \n",
        "min           0     43.015414           4.737000    -79.995023    958.433228   \n",
        "25%           0     43.646784          27.000000    -79.569345    989.320007   \n",
        "50%           0     43.699368          36.026000    -79.417536    995.592773   \n",
        "75%           0     43.736303         128.397000    -79.381984    999.920197   \n",
        "max           0     43.998816        3443.000000    -79.000505   1012.913269   \n",
        "\n",
        "       reading_accuracy         tzoffset            id  code_no  \n",
        "count             23827     23827.000000  23827.000000    23827  \n",
        "mean                  0 -13898837.453309    179.613128        2  \n",
        "std                   0   2638984.140779      2.643099        0  \n",
        "min                   0 -14400000.000000    175.000000        2  \n",
        "25%                   0 -14400000.000000    180.000000        2  \n",
        "50%                   0 -14400000.000000    181.000000        2  \n",
        "75%                   0 -14400000.000000    182.000000        2  \n",
        "max                   0         0.000000    182.000000        2  \n",
        "\n",
        "[8 rows x 9 columns]"
       ]
      }
     ],
     "prompt_number": 367
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "print len(warnings3)\n",
      "\n",
      "warnings3.describe()"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [
      {
       "output_type": "stream",
       "stream": "stdout",
       "text": [
        "17222\n"
       ]
      },
      {
       "html": [
        "<div style=\"max-height:1000px;max-width:1500px;overflow:auto;\">\n",
        "<table border=\"1\" class=\"dataframe\">\n",
        "  <thead>\n",
        "    <tr style=\"text-align: right;\">\n",
        "      <th></th>\n",
        "      <th>altitude</th>\n",
        "      <th>latitude</th>\n",
        "      <th>location_accuracy</th>\n",
        "      <th>longitude</th>\n",
        "      <th>reading</th>\n",
        "      <th>reading_accuracy</th>\n",
        "      <th>tzoffset</th>\n",
        "      <th>id</th>\n",
        "      <th>code_no</th>\n",
        "    </tr>\n",
        "  </thead>\n",
        "  <tbody>\n",
        "    <tr>\n",
        "      <th>count</th>\n",
        "      <td> 17222</td>\n",
        "      <td> 17222.000000</td>\n",
        "      <td> 17222.000000</td>\n",
        "      <td> 17222.000000</td>\n",
        "      <td> 17222.000000</td>\n",
        "      <td> 17222</td>\n",
        "      <td>    17222.000000</td>\n",
        "      <td> 17222.000000</td>\n",
        "      <td> 17222</td>\n",
        "    </tr>\n",
        "    <tr>\n",
        "      <th>mean</th>\n",
        "      <td>     0</td>\n",
        "      <td>    43.635359</td>\n",
        "      <td>   299.504357</td>\n",
        "      <td>   -79.476144</td>\n",
        "      <td>   998.575333</td>\n",
        "      <td>     0</td>\n",
        "      <td>-13268284.752061</td>\n",
        "      <td>   178.244281</td>\n",
        "      <td>     3</td>\n",
        "    </tr>\n",
        "    <tr>\n",
        "      <th>std</th>\n",
        "      <td>     0</td>\n",
        "      <td>     0.180330</td>\n",
        "      <td>   520.672903</td>\n",
        "      <td>     0.183222</td>\n",
        "      <td>     7.017629</td>\n",
        "      <td>     0</td>\n",
        "      <td>  3901858.699926</td>\n",
        "      <td>     4.387450</td>\n",
        "      <td>     0</td>\n",
        "    </tr>\n",
        "    <tr>\n",
        "      <th>min</th>\n",
        "      <td>     0</td>\n",
        "      <td>    43.003593</td>\n",
        "      <td>     1.250000</td>\n",
        "      <td>   -79.992424</td>\n",
        "      <td>   974.284546</td>\n",
        "      <td>     0</td>\n",
        "      <td>-21600000.000000</td>\n",
        "      <td>   173.000000</td>\n",
        "      <td>     3</td>\n",
        "    </tr>\n",
        "    <tr>\n",
        "      <th>25%</th>\n",
        "      <td>     0</td>\n",
        "      <td>    43.646312</td>\n",
        "      <td>    29.740000</td>\n",
        "      <td>   -79.557855</td>\n",
        "      <td>   993.064209</td>\n",
        "      <td>     0</td>\n",
        "      <td>-14400000.000000</td>\n",
        "      <td>   173.000000</td>\n",
        "      <td>     3</td>\n",
        "    </tr>\n",
        "    <tr>\n",
        "      <th>50%</th>\n",
        "      <td>     0</td>\n",
        "      <td>    43.675566</td>\n",
        "      <td>    37.500000</td>\n",
        "      <td>   -79.417526</td>\n",
        "      <td>   999.139984</td>\n",
        "      <td>     0</td>\n",
        "      <td>-14400000.000000</td>\n",
        "      <td>   178.000000</td>\n",
        "      <td>     3</td>\n",
        "    </tr>\n",
        "    <tr>\n",
        "      <th>75%</th>\n",
        "      <td>     0</td>\n",
        "      <td>    43.736284</td>\n",
        "      <td>   121.500000</td>\n",
        "      <td>   -79.382363</td>\n",
        "      <td>  1003.039978</td>\n",
        "      <td>     0</td>\n",
        "      <td>-14400000.000000</td>\n",
        "      <td>   184.000000</td>\n",
        "      <td>     3</td>\n",
        "    </tr>\n",
        "    <tr>\n",
        "      <th>max</th>\n",
        "      <td>     0</td>\n",
        "      <td>    43.999589</td>\n",
        "      <td>  3521.000000</td>\n",
        "      <td>   -79.001623</td>\n",
        "      <td>  1157.923584</td>\n",
        "      <td>     0</td>\n",
        "      <td>        0.000000</td>\n",
        "      <td>   184.000000</td>\n",
        "      <td>     3</td>\n",
        "    </tr>\n",
        "  </tbody>\n",
        "</table>\n",
        "<p>8 rows \u00d7 9 columns</p>\n",
        "</div>"
       ],
       "metadata": {},
       "output_type": "pyout",
       "prompt_number": 368,
       "text": [
        "       altitude      latitude  location_accuracy     longitude       reading  \\\n",
        "count     17222  17222.000000       17222.000000  17222.000000  17222.000000   \n",
        "mean          0     43.635359         299.504357    -79.476144    998.575333   \n",
        "std           0      0.180330         520.672903      0.183222      7.017629   \n",
        "min           0     43.003593           1.250000    -79.992424    974.284546   \n",
        "25%           0     43.646312          29.740000    -79.557855    993.064209   \n",
        "50%           0     43.675566          37.500000    -79.417526    999.139984   \n",
        "75%           0     43.736284         121.500000    -79.382363   1003.039978   \n",
        "max           0     43.999589        3521.000000    -79.001623   1157.923584   \n",
        "\n",
        "       reading_accuracy         tzoffset            id  code_no  \n",
        "count             17222     17222.000000  17222.000000    17222  \n",
        "mean                  0 -13268284.752061    178.244281        3  \n",
        "std                   0   3901858.699926      4.387450        0  \n",
        "min                   0 -21600000.000000    173.000000        3  \n",
        "25%                   0 -14400000.000000    173.000000        3  \n",
        "50%                   0 -14400000.000000    178.000000        3  \n",
        "75%                   0 -14400000.000000    184.000000        3  \n",
        "max                   0         0.000000    184.000000        3  \n",
        "\n",
        "[8 rows x 9 columns]"
       ]
      }
     ],
     "prompt_number": 368
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "print len(warnings4)\n",
      "\n",
      "warnings4.describe()"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [
      {
       "output_type": "stream",
       "stream": "stdout",
       "text": [
        "949\n"
       ]
      },
      {
       "html": [
        "<div style=\"max-height:1000px;max-width:1500px;overflow:auto;\">\n",
        "<table border=\"1\" class=\"dataframe\">\n",
        "  <thead>\n",
        "    <tr style=\"text-align: right;\">\n",
        "      <th></th>\n",
        "      <th>altitude</th>\n",
        "      <th>latitude</th>\n",
        "      <th>location_accuracy</th>\n",
        "      <th>longitude</th>\n",
        "      <th>reading</th>\n",
        "      <th>reading_accuracy</th>\n",
        "      <th>tzoffset</th>\n",
        "      <th>id</th>\n",
        "      <th>code_no</th>\n",
        "    </tr>\n",
        "  </thead>\n",
        "  <tbody>\n",
        "    <tr>\n",
        "      <th>count</th>\n",
        "      <td> 949</td>\n",
        "      <td> 949.000000</td>\n",
        "      <td>  949.000000</td>\n",
        "      <td> 949.000000</td>\n",
        "      <td>  949.000000</td>\n",
        "      <td> 949</td>\n",
        "      <td>      949.000000</td>\n",
        "      <td> 949</td>\n",
        "      <td> 949</td>\n",
        "    </tr>\n",
        "    <tr>\n",
        "      <th>mean</th>\n",
        "      <td>   0</td>\n",
        "      <td>  43.591688</td>\n",
        "      <td>  209.036393</td>\n",
        "      <td> -79.493094</td>\n",
        "      <td>  995.035589</td>\n",
        "      <td>   0</td>\n",
        "      <td>-13906849.315068</td>\n",
        "      <td> 185</td>\n",
        "      <td>   4</td>\n",
        "    </tr>\n",
        "    <tr>\n",
        "      <th>std</th>\n",
        "      <td>   0</td>\n",
        "      <td>   0.214829</td>\n",
        "      <td>  395.285907</td>\n",
        "      <td>   0.206161</td>\n",
        "      <td>    5.332409</td>\n",
        "      <td>   0</td>\n",
        "      <td>  2830858.713231</td>\n",
        "      <td>   0</td>\n",
        "      <td>   0</td>\n",
        "    </tr>\n",
        "    <tr>\n",
        "      <th>min</th>\n",
        "      <td>   0</td>\n",
        "      <td>  43.127080</td>\n",
        "      <td>   20.000000</td>\n",
        "      <td> -79.990436</td>\n",
        "      <td>  974.619995</td>\n",
        "      <td>   0</td>\n",
        "      <td>-21600000.000000</td>\n",
        "      <td> 185</td>\n",
        "      <td>   4</td>\n",
        "    </tr>\n",
        "    <tr>\n",
        "      <th>25%</th>\n",
        "      <td>   0</td>\n",
        "      <td>  43.558485</td>\n",
        "      <td>   29.740000</td>\n",
        "      <td> -79.635275</td>\n",
        "      <td>  990.559998</td>\n",
        "      <td>   0</td>\n",
        "      <td>-14400000.000000</td>\n",
        "      <td> 185</td>\n",
        "      <td>   4</td>\n",
        "    </tr>\n",
        "    <tr>\n",
        "      <th>50%</th>\n",
        "      <td>   0</td>\n",
        "      <td>  43.656342</td>\n",
        "      <td>   39.817000</td>\n",
        "      <td> -79.413306</td>\n",
        "      <td>  995.729980</td>\n",
        "      <td>   0</td>\n",
        "      <td>-14400000.000000</td>\n",
        "      <td> 185</td>\n",
        "      <td>   4</td>\n",
        "    </tr>\n",
        "    <tr>\n",
        "      <th>75%</th>\n",
        "      <td>   0</td>\n",
        "      <td>  43.739011</td>\n",
        "      <td>   76.227000</td>\n",
        "      <td> -79.379711</td>\n",
        "      <td>  999.800476</td>\n",
        "      <td>   0</td>\n",
        "      <td>-14400000.000000</td>\n",
        "      <td> 185</td>\n",
        "      <td>   4</td>\n",
        "    </tr>\n",
        "    <tr>\n",
        "      <th>max</th>\n",
        "      <td>   0</td>\n",
        "      <td>  43.855108</td>\n",
        "      <td> 2055.000000</td>\n",
        "      <td> -79.026341</td>\n",
        "      <td> 1003.940002</td>\n",
        "      <td>   0</td>\n",
        "      <td>        0.000000</td>\n",
        "      <td> 185</td>\n",
        "      <td>   4</td>\n",
        "    </tr>\n",
        "  </tbody>\n",
        "</table>\n",
        "<p>8 rows \u00d7 9 columns</p>\n",
        "</div>"
       ],
       "metadata": {},
       "output_type": "pyout",
       "prompt_number": 369,
       "text": [
        "       altitude    latitude  location_accuracy   longitude      reading  \\\n",
        "count       949  949.000000         949.000000  949.000000   949.000000   \n",
        "mean          0   43.591688         209.036393  -79.493094   995.035589   \n",
        "std           0    0.214829         395.285907    0.206161     5.332409   \n",
        "min           0   43.127080          20.000000  -79.990436   974.619995   \n",
        "25%           0   43.558485          29.740000  -79.635275   990.559998   \n",
        "50%           0   43.656342          39.817000  -79.413306   995.729980   \n",
        "75%           0   43.739011          76.227000  -79.379711   999.800476   \n",
        "max           0   43.855108        2055.000000  -79.026341  1003.940002   \n",
        "\n",
        "       reading_accuracy         tzoffset   id  code_no  \n",
        "count               949       949.000000  949      949  \n",
        "mean                  0 -13906849.315068  185        4  \n",
        "std                   0   2830858.713231    0        0  \n",
        "min                   0 -21600000.000000  185        4  \n",
        "25%                   0 -14400000.000000  185        4  \n",
        "50%                   0 -14400000.000000  185        4  \n",
        "75%                   0 -14400000.000000  185        4  \n",
        "max                   0         0.000000  185        4  \n",
        "\n",
        "[8 rows x 9 columns]"
       ]
      }
     ],
     "prompt_number": 369
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "print len(warnings5)\n",
      "\n",
      "warnings5.describe()"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [
      {
       "output_type": "stream",
       "stream": "stdout",
       "text": [
        "12072\n"
       ]
      },
      {
       "html": [
        "<div style=\"max-height:1000px;max-width:1500px;overflow:auto;\">\n",
        "<table border=\"1\" class=\"dataframe\">\n",
        "  <thead>\n",
        "    <tr style=\"text-align: right;\">\n",
        "      <th></th>\n",
        "      <th>altitude</th>\n",
        "      <th>latitude</th>\n",
        "      <th>location_accuracy</th>\n",
        "      <th>longitude</th>\n",
        "      <th>reading</th>\n",
        "      <th>reading_accuracy</th>\n",
        "      <th>tzoffset</th>\n",
        "      <th>id</th>\n",
        "      <th>code_no</th>\n",
        "    </tr>\n",
        "  </thead>\n",
        "  <tbody>\n",
        "    <tr>\n",
        "      <th>count</th>\n",
        "      <td> 12072</td>\n",
        "      <td> 12072.000000</td>\n",
        "      <td> 12072.000000</td>\n",
        "      <td> 12072.000000</td>\n",
        "      <td> 12072.000000</td>\n",
        "      <td> 12072</td>\n",
        "      <td>    12072.000000</td>\n",
        "      <td> 12072.000000</td>\n",
        "      <td> 12072</td>\n",
        "    </tr>\n",
        "    <tr>\n",
        "      <th>mean</th>\n",
        "      <td>     0</td>\n",
        "      <td>    43.649647</td>\n",
        "      <td>   291.636069</td>\n",
        "      <td>   -79.470711</td>\n",
        "      <td>   999.512005</td>\n",
        "      <td>     0</td>\n",
        "      <td>-13158250.497018</td>\n",
        "      <td>   176.479705</td>\n",
        "      <td>     5</td>\n",
        "    </tr>\n",
        "    <tr>\n",
        "      <th>std</th>\n",
        "      <td>     0</td>\n",
        "      <td>     0.171120</td>\n",
        "      <td>   520.591584</td>\n",
        "      <td>     0.183464</td>\n",
        "      <td>     6.532511</td>\n",
        "      <td>     0</td>\n",
        "      <td>  4042351.371450</td>\n",
        "      <td>     2.500021</td>\n",
        "      <td>     0</td>\n",
        "    </tr>\n",
        "    <tr>\n",
        "      <th>min</th>\n",
        "      <td>     0</td>\n",
        "      <td>    43.060856</td>\n",
        "      <td>    20.000000</td>\n",
        "      <td>   -79.995097</td>\n",
        "      <td>   970.759949</td>\n",
        "      <td>     0</td>\n",
        "      <td>-14400000.000000</td>\n",
        "      <td>   174.000000</td>\n",
        "      <td>     5</td>\n",
        "    </tr>\n",
        "    <tr>\n",
        "      <th>25%</th>\n",
        "      <td>     0</td>\n",
        "      <td>    43.650078</td>\n",
        "      <td>    31.740750</td>\n",
        "      <td>   -79.548248</td>\n",
        "      <td>   993.863770</td>\n",
        "      <td>     0</td>\n",
        "      <td>-14400000.000000</td>\n",
        "      <td>   174.000000</td>\n",
        "      <td>     5</td>\n",
        "    </tr>\n",
        "    <tr>\n",
        "      <th>50%</th>\n",
        "      <td>     0</td>\n",
        "      <td>    43.694753</td>\n",
        "      <td>    39.698500</td>\n",
        "      <td>   -79.417518</td>\n",
        "      <td>  1000.555023</td>\n",
        "      <td>     0</td>\n",
        "      <td>-14400000.000000</td>\n",
        "      <td>   174.000000</td>\n",
        "      <td>     5</td>\n",
        "    </tr>\n",
        "    <tr>\n",
        "      <th>75%</th>\n",
        "      <td>     0</td>\n",
        "      <td>    43.736302</td>\n",
        "      <td>   107.008500</td>\n",
        "      <td>   -79.379235</td>\n",
        "      <td>  1003.679993</td>\n",
        "      <td>     0</td>\n",
        "      <td>-14400000.000000</td>\n",
        "      <td>   179.000000</td>\n",
        "      <td>     5</td>\n",
        "    </tr>\n",
        "    <tr>\n",
        "      <th>max</th>\n",
        "      <td>     0</td>\n",
        "      <td>    43.991877</td>\n",
        "      <td>  3188.000000</td>\n",
        "      <td>   -79.000225</td>\n",
        "      <td>  1014.539978</td>\n",
        "      <td>     0</td>\n",
        "      <td>        0.000000</td>\n",
        "      <td>   179.000000</td>\n",
        "      <td>     5</td>\n",
        "    </tr>\n",
        "  </tbody>\n",
        "</table>\n",
        "<p>8 rows \u00d7 9 columns</p>\n",
        "</div>"
       ],
       "metadata": {},
       "output_type": "pyout",
       "prompt_number": 370,
       "text": [
        "       altitude      latitude  location_accuracy     longitude       reading  \\\n",
        "count     12072  12072.000000       12072.000000  12072.000000  12072.000000   \n",
        "mean          0     43.649647         291.636069    -79.470711    999.512005   \n",
        "std           0      0.171120         520.591584      0.183464      6.532511   \n",
        "min           0     43.060856          20.000000    -79.995097    970.759949   \n",
        "25%           0     43.650078          31.740750    -79.548248    993.863770   \n",
        "50%           0     43.694753          39.698500    -79.417518   1000.555023   \n",
        "75%           0     43.736302         107.008500    -79.379235   1003.679993   \n",
        "max           0     43.991877        3188.000000    -79.000225   1014.539978   \n",
        "\n",
        "       reading_accuracy         tzoffset            id  code_no  \n",
        "count             12072     12072.000000  12072.000000    12072  \n",
        "mean                  0 -13158250.497018    176.479705        5  \n",
        "std                   0   4042351.371450      2.500021        0  \n",
        "min                   0 -14400000.000000    174.000000        5  \n",
        "25%                   0 -14400000.000000    174.000000        5  \n",
        "50%                   0 -14400000.000000    174.000000        5  \n",
        "75%                   0 -14400000.000000    179.000000        5  \n",
        "max                   0         0.000000    179.000000        5  \n",
        "\n",
        "[8 rows x 9 columns]"
       ]
      }
     ],
     "prompt_number": 370
    },
    {
     "cell_type": "markdown",
     "metadata": {},
     "source": [
      "It seems that the mean air pressure for heat warning days is 993.677119, the overall mean air pressure for the 11 day window I'm looking at is 995.170459, and the mean for the days without heat warnings is 995.040635.  The mean air pressure for code 1 warnings (extreme heat warnings downgraded to standard heat warnings) is as low as 988.579502."
     ]
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "import scipy.stats\n",
      "scipy.stats.ttest_ind(warnings['reading'], nonwarnings['reading'])"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [
      {
       "metadata": {},
       "output_type": "pyout",
       "prompt_number": 109,
       "text": [
        "(array(-29.830170950049535), 2.2845887961730417e-195)"
       ]
      }
     ],
     "prompt_number": 109
    },
    {
     "cell_type": "markdown",
     "metadata": {},
     "source": [
      "It was suggested to me that I use a t-test for the warnings and non-warnings data sets instead of just looking at their means.  I think a t-test looks at the distribution of values in each data set.  After all, it's possible for two very different distributions to results in the same mean value.\n",
      "\n",
      "The results of this ttest_ind method are the t-statistic the the p-value.  A quick google search makes me believe that a t-statistic greater than 2.0 or less than -2.0 means a significant difference of means, as does a p-value smaller than 0.05.  My data seems to pass that test!"
     ]
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "warnings.hist('reading')"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [
      {
       "metadata": {},
       "output_type": "pyout",
       "prompt_number": 110,
       "text": [
        "array([[<matplotlib.axes.AxesSubplot object at 0xc68719d0>]], dtype=object)"
       ]
      },
      {
       "metadata": {},
       "output_type": "display_data",
       "png": "iVBORw0KGgoAAAANSUhEUgAAAYUAAAEICAYAAACwDehOAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\n",
       "text": [
        "<matplotlib.figure.Figure at 0xb4d31a10>"
       ]
      }
     ],
     "prompt_number": 110
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "nonwarnings.hist('reading')"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [
      {
       "metadata": {},
       "output_type": "pyout",
       "prompt_number": 111,
       "text": [
        "array([[<matplotlib.axes.AxesSubplot object at 0xbb3ceed0>]], dtype=object)"
       ]
      },
      {
       "metadata": {},
       "output_type": "display_data",
       "png": "iVBORw0KGgoAAAANSUhEUgAAAZEAAAEICAYAAACeSMncAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\ndjr+7QAAAABJRU5ErkJggg==\n",
       "text": [
        "<matplotlib.figure.Figure at 0x3063f6d0>"
       ]
      }
     ],
     "prompt_number": 111
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "tor.hist('reading')"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [
      {
       "metadata": {},
       "output_type": "pyout",
       "prompt_number": 112,
       "text": [
        "array([[<matplotlib.axes.AxesSubplot object at 0x71534bd0>]], dtype=object)"
       ]
      },
      {
       "metadata": {},
       "output_type": "display_data",
       "png": "iVBORw0KGgoAAAANSUhEUgAAAZEAAAEICAYAAACeSMncAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nBL/MSAAAAABJRU5ErkJggg==\n",
       "text": [
        "<matplotlib.figure.Figure at 0xb7cf9050>"
       ]
      }
     ],
     "prompt_number": 112
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "warnings[\"time\"] = warnings.apply(lambda x: arrow.get(\"T\".join(str(x['daterecorded']).split(\" \"))).time(), axis=1)\n",
      "nonwarnings[\"time\"] = nonwarnings.apply(lambda x: arrow.get(\"T\".join(str(x['daterecorded']).split(\" \"))).time(), axis=1)\n",
      "tor[\"time\"] = tor.apply(lambda x: arrow.get(\"T\".join(str(x['daterecorded']).split(\" \"))).time(), axis=1)"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [],
     "prompt_number": 113
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "starttime = datetime.time(13,0)\n",
      "endtime = datetime.time(23,59)"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [],
     "prompt_number": 355
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "warnings_day = warnings[(warnings['time'] > starttime) & (warnings['time'] < endtime)]\n",
      "nonwarnings_day = nonwarnings[(nonwarnings['time'] > starttime) & (nonwarnings['time'] < endtime)]\n",
      "days = tor[(tor['time'] > starttime) & (tor['time'] < endtime)]"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [],
     "prompt_number": 356
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "scipy.stats.ttest_ind(warnings_day['reading'], nonwarnings_day['reading'])"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [
      {
       "metadata": {},
       "output_type": "pyout",
       "prompt_number": 357,
       "text": [
        "(array(-42.752879538465386), 0.0)"
       ]
      }
     ],
     "prompt_number": 357
    },
    {
     "cell_type": "markdown",
     "metadata": {},
     "source": [
      "With a bit of experimentation I found that limiting my data to readings from 1:00pm onward maximized the difference as reported by a ttest.  I expected that the hours 11:00am - 5:00pm might be more statistically telling, since that is generally the hottest time of the day.  But perhaps air pressure and temperature are not that closely correlated."
     ]
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "warnings_day.describe()"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [
      {
       "html": [
        "<div style=\"max-height:1000px;max-width:1500px;overflow:auto;\">\n",
        "<table border=\"1\" class=\"dataframe\">\n",
        "  <thead>\n",
        "    <tr style=\"text-align: right;\">\n",
        "      <th></th>\n",
        "      <th>altitude</th>\n",
        "      <th>latitude</th>\n",
        "      <th>location_accuracy</th>\n",
        "      <th>longitude</th>\n",
        "      <th>reading</th>\n",
        "      <th>reading_accuracy</th>\n",
        "      <th>tzoffset</th>\n",
        "      <th>id</th>\n",
        "      <th>code_no</th>\n",
        "    </tr>\n",
        "  </thead>\n",
        "  <tbody>\n",
        "    <tr>\n",
        "      <th>count</th>\n",
        "      <td> 14423</td>\n",
        "      <td> 14423.000000</td>\n",
        "      <td> 14423.000000</td>\n",
        "      <td> 14423.000000</td>\n",
        "      <td> 14423.000000</td>\n",
        "      <td> 14423</td>\n",
        "      <td>    14423.000000</td>\n",
        "      <td> 14423.000000</td>\n",
        "      <td> 14423.000000</td>\n",
        "    </tr>\n",
        "    <tr>\n",
        "      <th>mean</th>\n",
        "      <td>     0</td>\n",
        "      <td>    43.643207</td>\n",
        "      <td>   327.061149</td>\n",
        "      <td>   -79.481421</td>\n",
        "      <td>   992.680759</td>\n",
        "      <td>     0</td>\n",
        "      <td>-13492699.161062</td>\n",
        "      <td>   178.903349</td>\n",
        "      <td>     1.802954</td>\n",
        "    </tr>\n",
        "    <tr>\n",
        "      <th>std</th>\n",
        "      <td>     0</td>\n",
        "      <td>     0.175201</td>\n",
        "      <td>   532.516588</td>\n",
        "      <td>     0.183891</td>\n",
        "      <td>     7.252837</td>\n",
        "      <td>     0</td>\n",
        "      <td>  3499611.298638</td>\n",
        "      <td>     2.781350</td>\n",
        "      <td>     0.397781</td>\n",
        "    </tr>\n",
        "    <tr>\n",
        "      <th>min</th>\n",
        "      <td>     0</td>\n",
        "      <td>    43.075259</td>\n",
        "      <td>     4.737000</td>\n",
        "      <td>   -79.981781</td>\n",
        "      <td>   958.433228</td>\n",
        "      <td>     0</td>\n",
        "      <td>-18000000.000000</td>\n",
        "      <td>   175.000000</td>\n",
        "      <td>     1.000000</td>\n",
        "    </tr>\n",
        "    <tr>\n",
        "      <th>25%</th>\n",
        "      <td>     0</td>\n",
        "      <td>    43.646658</td>\n",
        "      <td>    28.000000</td>\n",
        "      <td>   -79.569886</td>\n",
        "      <td>   986.769958</td>\n",
        "      <td>     0</td>\n",
        "      <td>-14400000.000000</td>\n",
        "      <td>   176.000000</td>\n",
        "      <td>     2.000000</td>\n",
        "    </tr>\n",
        "    <tr>\n",
        "      <th>50%</th>\n",
        "      <td>     0</td>\n",
        "      <td>    43.687811</td>\n",
        "      <td>    40.500000</td>\n",
        "      <td>   -79.417532</td>\n",
        "      <td>   992.849976</td>\n",
        "      <td>     0</td>\n",
        "      <td>-14400000.000000</td>\n",
        "      <td>   180.000000</td>\n",
        "      <td>     2.000000</td>\n",
        "    </tr>\n",
        "    <tr>\n",
        "      <th>75%</th>\n",
        "      <td>     0</td>\n",
        "      <td>    43.736302</td>\n",
        "      <td>   201.616000</td>\n",
        "      <td>   -79.379235</td>\n",
        "      <td>   997.729980</td>\n",
        "      <td>     0</td>\n",
        "      <td>-14400000.000000</td>\n",
        "      <td>   181.000000</td>\n",
        "      <td>     2.000000</td>\n",
        "    </tr>\n",
        "    <tr>\n",
        "      <th>max</th>\n",
        "      <td>     0</td>\n",
        "      <td>    43.997440</td>\n",
        "      <td>  3443.000000</td>\n",
        "      <td>   -79.001904</td>\n",
        "      <td>  1012.199951</td>\n",
        "      <td>     0</td>\n",
        "      <td>        0.000000</td>\n",
        "      <td>   182.000000</td>\n",
        "      <td>     2.000000</td>\n",
        "    </tr>\n",
        "  </tbody>\n",
        "</table>\n",
        "<p>8 rows \u00d7 9 columns</p>\n",
        "</div>"
       ],
       "metadata": {},
       "output_type": "pyout",
       "prompt_number": 359,
       "text": [
        "       altitude      latitude  location_accuracy     longitude       reading  \\\n",
        "count     14423  14423.000000       14423.000000  14423.000000  14423.000000   \n",
        "mean          0     43.643207         327.061149    -79.481421    992.680759   \n",
        "std           0      0.175201         532.516588      0.183891      7.252837   \n",
        "min           0     43.075259           4.737000    -79.981781    958.433228   \n",
        "25%           0     43.646658          28.000000    -79.569886    986.769958   \n",
        "50%           0     43.687811          40.500000    -79.417532    992.849976   \n",
        "75%           0     43.736302         201.616000    -79.379235    997.729980   \n",
        "max           0     43.997440        3443.000000    -79.001904   1012.199951   \n",
        "\n",
        "       reading_accuracy         tzoffset            id       code_no  \n",
        "count             14423     14423.000000  14423.000000  14423.000000  \n",
        "mean                  0 -13492699.161062    178.903349      1.802954  \n",
        "std                   0   3499611.298638      2.781350      0.397781  \n",
        "min                   0 -18000000.000000    175.000000      1.000000  \n",
        "25%                   0 -14400000.000000    176.000000      2.000000  \n",
        "50%                   0 -14400000.000000    180.000000      2.000000  \n",
        "75%                   0 -14400000.000000    181.000000      2.000000  \n",
        "max                   0         0.000000    182.000000      2.000000  \n",
        "\n",
        "[8 rows x 9 columns]"
       ]
      }
     ],
     "prompt_number": 359
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "nonwarnings_day.describe()"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [
      {
       "html": [
        "<div style=\"max-height:1000px;max-width:1500px;overflow:auto;\">\n",
        "<table border=\"1\" class=\"dataframe\">\n",
        "  <thead>\n",
        "    <tr style=\"text-align: right;\">\n",
        "      <th></th>\n",
        "      <th>altitude</th>\n",
        "      <th>latitude</th>\n",
        "      <th>location_accuracy</th>\n",
        "      <th>longitude</th>\n",
        "      <th>reading</th>\n",
        "      <th>reading_accuracy</th>\n",
        "      <th>tzoffset</th>\n",
        "      <th>id</th>\n",
        "      <th>code_no</th>\n",
        "    </tr>\n",
        "  </thead>\n",
        "  <tbody>\n",
        "    <tr>\n",
        "      <th>count</th>\n",
        "      <td> 323236.000000</td>\n",
        "      <td> 323236.000000</td>\n",
        "      <td> 323236.000000</td>\n",
        "      <td> 323236.000000</td>\n",
        "      <td> 323236.000000</td>\n",
        "      <td> 323236</td>\n",
        "      <td>   323236.000000</td>\n",
        "      <td>  0</td>\n",
        "      <td> 323236</td>\n",
        "    </tr>\n",
        "    <tr>\n",
        "      <th>mean</th>\n",
        "      <td>      0.362758</td>\n",
        "      <td>     43.639612</td>\n",
        "      <td>    309.844850</td>\n",
        "      <td>    -79.477183</td>\n",
        "      <td>    995.380575</td>\n",
        "      <td>      0</td>\n",
        "      <td>-13442954.373894</td>\n",
        "      <td>NaN</td>\n",
        "      <td>      0</td>\n",
        "    </tr>\n",
        "    <tr>\n",
        "      <th>std</th>\n",
        "      <td>      9.457962</td>\n",
        "      <td>      0.179994</td>\n",
        "      <td>    522.518671</td>\n",
        "      <td>      0.184118</td>\n",
        "      <td>      7.747306</td>\n",
        "      <td>      0</td>\n",
        "      <td>  3600175.892965</td>\n",
        "      <td>NaN</td>\n",
        "      <td>      0</td>\n",
        "    </tr>\n",
        "    <tr>\n",
        "      <th>min</th>\n",
        "      <td>    -59.100006</td>\n",
        "      <td>     43.000120</td>\n",
        "      <td>      1.000000</td>\n",
        "      <td>    -79.999870</td>\n",
        "      <td>    910.229736</td>\n",
        "      <td>      0</td>\n",
        "      <td>-28800000.000000</td>\n",
        "      <td>NaN</td>\n",
        "      <td>      0</td>\n",
        "    </tr>\n",
        "    <tr>\n",
        "      <th>25%</th>\n",
        "      <td>      0.000000</td>\n",
        "      <td>     43.646387</td>\n",
        "      <td>     30.000000</td>\n",
        "      <td>    -79.568931</td>\n",
        "      <td>    990.165771</td>\n",
        "      <td>      0</td>\n",
        "      <td>-14400000.000000</td>\n",
        "      <td>NaN</td>\n",
        "      <td>      0</td>\n",
        "    </tr>\n",
        "    <tr>\n",
        "      <th>50%</th>\n",
        "      <td>      0.000000</td>\n",
        "      <td>     43.682701</td>\n",
        "      <td>     41.907000</td>\n",
        "      <td>    -79.417531</td>\n",
        "      <td>    995.459961</td>\n",
        "      <td>      0</td>\n",
        "      <td>-14400000.000000</td>\n",
        "      <td>NaN</td>\n",
        "      <td>      0</td>\n",
        "    </tr>\n",
        "    <tr>\n",
        "      <th>75%</th>\n",
        "      <td>      0.000000</td>\n",
        "      <td>     43.736308</td>\n",
        "      <td>    159.068000</td>\n",
        "      <td>    -79.379235</td>\n",
        "      <td>   1000.859924</td>\n",
        "      <td>      0</td>\n",
        "      <td>-14400000.000000</td>\n",
        "      <td>NaN</td>\n",
        "      <td>      0</td>\n",
        "    </tr>\n",
        "    <tr>\n",
        "      <th>max</th>\n",
        "      <td>   1000.199951</td>\n",
        "      <td>     43.999977</td>\n",
        "      <td>   4908.000000</td>\n",
        "      <td>    -79.000415</td>\n",
        "      <td>   1145.450439</td>\n",
        "      <td>      0</td>\n",
        "      <td>  7200000.000000</td>\n",
        "      <td>NaN</td>\n",
        "      <td>      0</td>\n",
        "    </tr>\n",
        "  </tbody>\n",
        "</table>\n",
        "<p>8 rows \u00d7 9 columns</p>\n",
        "</div>"
       ],
       "metadata": {},
       "output_type": "pyout",
       "prompt_number": 312,
       "text": [
        "            altitude       latitude  location_accuracy      longitude  \\\n",
        "count  323236.000000  323236.000000      323236.000000  323236.000000   \n",
        "mean        0.362758      43.639612         309.844850     -79.477183   \n",
        "std         9.457962       0.179994         522.518671       0.184118   \n",
        "min       -59.100006      43.000120           1.000000     -79.999870   \n",
        "25%         0.000000      43.646387          30.000000     -79.568931   \n",
        "50%         0.000000      43.682701          41.907000     -79.417531   \n",
        "75%         0.000000      43.736308         159.068000     -79.379235   \n",
        "max      1000.199951      43.999977        4908.000000     -79.000415   \n",
        "\n",
        "             reading  reading_accuracy         tzoffset  id  code_no  \n",
        "count  323236.000000            323236    323236.000000   0   323236  \n",
        "mean      995.380575                 0 -13442954.373894 NaN        0  \n",
        "std         7.747306                 0   3600175.892965 NaN        0  \n",
        "min       910.229736                 0 -28800000.000000 NaN        0  \n",
        "25%       990.165771                 0 -14400000.000000 NaN        0  \n",
        "50%       995.459961                 0 -14400000.000000 NaN        0  \n",
        "75%      1000.859924                 0 -14400000.000000 NaN        0  \n",
        "max      1145.450439                 0   7200000.000000 NaN        0  \n",
        "\n",
        "[8 rows x 9 columns]"
       ]
      }
     ],
     "prompt_number": 312
    },
    {
     "cell_type": "markdown",
     "metadata": {},
     "source": [
      "The mean reading for warning days during the hours of 1pm to midnight is just 992.680759, compared to 995.380575 for the nonwarning days."
     ]
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "warnings_day.hist('reading')"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [
      {
       "metadata": {},
       "output_type": "pyout",
       "prompt_number": 313,
       "text": [
        "array([[<matplotlib.axes.AxesSubplot object at 0x4a559ad0>]], dtype=object)"
       ]
      },
      {
       "metadata": {},
       "output_type": "display_data",
       "png": "iVBORw0KGgoAAAANSUhEUgAAAYUAAAEICAYAAACwDehOAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz\nAAALEgAACxIB0t1+/AAAIABJREFUeJzt3X9Q1HX+B/AnCp3jKYGdLrarrcECEiCYbt7dmBQu/rgi\nfw3JlULqTQOnR13XqM3V2fQ9obpqzM6xu0FjcArtl9ike+jIJ60bMbGuLky2CxKWdSsRpaJIfH//\nMD6IqMDu57Pv/cDzMcPI58Puvp+f9bP74vN+fT5LiBBCgIiICMAQ2QGIiCh4sCgQEZGKRYGIiFQs\nCkREpGJRICIiFYsCERGpWBSI+mHIkCH4/PPPAQB5eXn4v//7P8mJiLQVwusUiPpuyJAh+Oyzz3Dj\njTfKjkKkCx4p0IB27tw52RGIDIVFgQYcq9WKp556CsnJyRg5ciTee+89/OpXv0JkZCRSUlLwzjvv\nqLfdunUrEhISEB4ejujoaPzjH//o9lhPP/00rr/+elgsFmzZsqXbz3Jzc/Hoo48CABRFgcViwbPP\nPguTyYTrr78eL730knrbU6dO4c4778S1114Lu92OP//5z5g+fbp+TwKRj0JlByDSQ1lZGfbs2YOQ\nkBAkJydj27ZtmD17Nvbt24eFCxfi+PHjuO6662AymfD2229jwoQJOHDgAObMmYOpU6ciNTUVTqcT\nzzzzDPbv3w+r1YoVK1Z0GyMkJAQhISHqstfrxdmzZ9HU1ISKigosWrQI8+fPx7XXXovf//73GDly\nJLxeL+rq6jBr1ixYrdYAPytEveORAg04ISEh+MMf/gCz2YzS0lLMnTsXs2fPBgDMnDkTU6ZMwdtv\nvw0AmDt3LiZMmAAAuPXWW5GRkYGDBw8CAHbs2IFly5YhISEBw4cPx+OPP95jrItbcmFhYXjssccw\ndOhQzJkzByNGjMDx48fR0dGBN954A48//jiGDRuGiRMnIicnB2znUTBiUaABady4cQCAL774Aq++\n+ioiIyPVr/feew8nT54EAOzZswfTpk3Dddddh8jISOzevRunTp0CAHg8HvVxAGD8+PFXHfO6667D\nkCFdL6nhw4fjm2++wVdffYVz5851eyyLxaLZthJpidNHNCB1TuuMHz8eS5Ys6dErAIAffvgBCxcu\nxLZt23DXXXdh6NChmD9/vvob/NixY3HixAn19hd/f+k4VzN69GiEhoaioaEBNpsNANDQ0ODTdhHp\njUcKNKDde++9eOutt1BRUYGOjg58//33UBQFbrcb7e3taG9vxy9+8QsMGTIEe/bsQUVFhXrfrKws\nvPTSSzh27Bi+++67HtNHQog+TQENHToUCxYswLp169DW1oZPP/0UpaWlfSooRIHGokADmsViQXl5\nOdavX48xY8Zg/PjxeOaZZyCEwMiRI/H8888jKysLo0aNwiuvvIK77rpLve/s2bPxwAMP4Pbbb0ds\nbCzS09O7vZFf2mi+2pv8Cy+8gDNnziAqKgo5OTnIzs7GNddco89GE/mBF68RSbB69Wp8+eWX2Lp1\nq+woRN306UihpaUFixYtwsSJE5GQkICqqio0NzfD4XAgNjYWGRkZaGlpUW9fWFgIm82G+Pj4bofj\n1dXVSEpKgs1mQ0FBgfZbQxSkjh8/jo8++ghCCBw+fBhbtmzB/PnzZcci6qFPRaGgoABz587FsWPH\n8NFHHyE+Ph5FRUVwOByora1Feno6ioqKAAA1NTXYvn07ampq4HQ6kZ+fr8675uXlobi4GC6XCy6X\nC06nU78tIwoira2tWLhwIUaMGIHFixfjT3/6EzIzM2XHIuqh1+mjM2fOIDU1Vf0QsE7x8fF45513\nYDKZcPLkSaSlpeHTTz9FYWEhhgwZgtWrVwO4MC+7bt063HDDDbj99ttx7NgxABcuLlIUBZs3b9Zp\n04iIqL96PVKoq6vD6NGjcd9992Hy5Mn43e9+h2+//RZerxcmkwkAYDKZ4PV6AQBNTU3dzsG2WCxw\nu9091pvNZrjdbq23h4iI/NDrdQrnzp3D0aNH8cILL2Dq1Kl44IEH1KmiTpeeheEPnqZHRNR/Wp0z\n1OuRgsVigcViwdSpUwEAixYtwtGjRxEVFaVeFerxeDBmzBgAF44ALr4wp7GxERaLBWazGY2Njd3W\nm83my47Zef630b7+8pe/SM/A/PJzML/xvoycXQhtTyDttShERUVh3LhxqK2tBQDs27cPN910E+68\n806UlJQAAEpKSjBv3jwAQGZmJsrKytDe3o66ujq4XC7Y7XZERUUhPDwcVVVVEEKgtLRUvc9AUV9f\nLzuCX5hfLuaXx8jZtdanj7nYuHEj7rnnHrS3tyM6Ohpbt25FR0cHsrKyUFxcDKvVih07dgAAEhIS\nkJWVhYSEBISGhmLTpk3qlNCmTZuQm5uLtra2bh9SRkREwSHoLl4LCQnR/HAoUBRFQVpamuwYPmN+\nuZhfHiNnB7R932RRICIyOC3fN/nZRxpSFEV2BL8wv1zML4+Rs2uNRYGIiFScPiIiMjhOHxERkS5Y\nFDRk9HlJ5peL+eUxcnatsSgQEZGKPQUiIoNjT4GIiHTBoqAho89LMr9czC+PkbNrjUWBiIhU7CkQ\nERkcewpERKQLFgUNGX1ekvnlYn55jJxdaywKRESkYk+BiMjg2FMgIiJdsChoyOjzkswvF/PLY+Ts\nWmNRICIiFXsKREQGx54CERHpgkVBQ0afl2R+uZhfHiNn1xqLAhERqdhTICIyOPYUiIhIFywKGjL6\nvCTzy8X88hg5u9ZYFIiISMWeAhGRwQW8p2C1WpGcnIzU1FTY7XYAQHNzMxwOB2JjY5GRkYGWlhb1\n9oWFhbDZbIiPj0dFRYW6vrq6GklJSbDZbCgoKNBkA4gCKTx8FEJCQnT/Cg8fJXtTaZDqU1EICQmB\noij44IMPcPjwYQBAUVERHA4HamtrkZ6ejqKiIgBATU0Ntm/fjpqaGjidTuTn56sVLC8vD8XFxXC5\nXHC5XHA6nTptlhxGn5dk/t61tp4GIHT6qlS/vzCOsRh5/zFydq31uadw6aHJrl27kJOTAwDIycnB\nzp07AQDl5eXIzs5GWFgYrFYrYmJiUFVVBY/Hg9bWVvVIY+nSpep9iIgoOPT5SGHmzJmYMmUK/vnP\nfwIAvF4vTCYTAMBkMsHr9QIAmpqaYLFY1PtaLBa43e4e681mM9xut2YbEgzS0tJkR/AL88uWJjuA\nX4z8/Bs5u9ZC+3Kj9957D2PHjsVXX30Fh8OB+Pj4bj/vnAfVSm5uLqxWKwAgIiICKSkp6n9a52Ee\nl7ksY/kCBV1v4MpP/2q9DF3yc3lgLHd+X19fD82Jflq3bp3429/+JuLi4oTH4xFCCNHU1CTi4uKE\nEEIUFhaKwsJC9fazZs0Shw4dEh6PR8THx6vrX375ZXH//ff3eHwfIgWNyspK2RH8wvy9AyAAodNX\n5UXfG+91YOT9x8jZhdB2f+l1+ui7775Da2srAODbb79FRUUFkpKSkJmZiZKSEgBASUkJ5s2bBwDI\nzMxEWVkZ2tvbUVdXB5fLBbvdjqioKISHh6OqqgpCCJSWlqr3ISKi4NDrdQp1dXWYP38+AODcuXO4\n5557sHbtWjQ3NyMrKwsnTpyA1WrFjh07EBERAQBYv349tmzZgtDQUGzYsAGzZs0CcOGU1NzcXLS1\ntWHu3Ll4/vnnewbidQoUxC5MkwZi/+TrgPpOy/dNXrxG1A8sChSM+IF4QeriJpARMb9siuwAfjHy\n82/k7FpjUSAiIhWnj4j6gdNHFIw4fURERLpgUdCQ0eclmV82RXYAvxj5+Tdydq2xKBARkYo9BaJ+\nYE+BghF7CkREpAsWBQ0ZfV6S+WVTZAfwi5GffyNn1xqLAhERqdhTIOoH9hQoGLGnQEREumBR0JDR\n5yWZXzZFdgC/GPn5N3J2rbEoEBGRij0Fon4YaD2F8PBRaG09rfs4I0dG4uzZZt3HGaz49xSIJBlo\nRWGgbc9gxUZzkDL6vCTzy6bIDuAnRXYAnxl/39EOiwIREak4fUTUDwNtumWgbc9gxekjIiLSBYuC\nhow+L8n8simyA/hJkR3AZ8bfd7TDokBERCr2FIj6YaDNwQ+07Rms2FMgIiJdsChoyOjzkswvm3LR\n96EICQnR/Uu//MZi/H1HO6GyAxDR5ZxDoKZ1iC7GngJRPwRyDn6gjcPXtX7YUyAiIl2wKGjI6POS\nzC+bIjuAnxTZAXxm/H1HO30qCh0dHUhNTcWdd94JAGhubobD4UBsbCwyMjLQ0tKi3rawsBA2mw3x\n8fGoqKhQ11dXVyMpKQk2mw0FBQUabwYREWmhT0Vhw4YNSEhIUM9WKCoqgsPhQG1tLdLT01FUVAQA\nqKmpwfbt21FTUwOn04n8/Hx1nisvLw/FxcVwuVxwuVxwOp06bZI8aWlpsiP4hfllS5MdwE9psgP4\nzPj7jnZ6LQqNjY3YvXs3VqxYob7B79q1Czk5OQCAnJwc7Ny5EwBQXl6O7OxshIWFwWq1IiYmBlVV\nVfB4PGhtbYXdbgcALF26VL0PEREFj15PSX3wwQfx9NNP4+zZs+o6r9cLk8kEADCZTPB6vQCApqYm\nTJs2Tb2dxWKB2+1GWFgYLBaLut5sNsPtdl9xzNzcXFitVgBAREQEUlJS1EreOfcXjMsXz0sGQx7m\n1368CxR0/VbcOaYWy135u4+l1eNfbrlznRaP1/n95X7+01IQ7S8XL3euC5Y8fcmrKArq6+uhOXEV\nb731lsjPzxdCCFFZWSnuuOMOIYQQERER3W4XGRkphBBi5cqVYtu2ber65cuXi9dee00cOXJEzJw5\nU11/4MAB9bEu1UukoFZZWSk7gl+Yv3cABCB0+qq86Hs9xxE6jVN51XGCmdH3fS2f36seKfz73//G\nrl27sHv3bnz//fc4e/YslixZApPJhJMnTyIqKgoejwdjxowBcOEIoKGhQb1/Y2MjLBYLzGYzGhsb\nu603m82aFzjZjD4vyfyypckO4Kc02QF8Zvx9RztX7SmsX78eDQ0NqKurQ1lZGW6//XaUlpYiMzMT\nJSUlAICSkhLMmzcPAJCZmYmysjK0t7ejrq4OLpcLdrsdUVFRCA8PR1VVFYQQKC0tVe9DRETBo1/X\nKXSefbRmzRrs3bsXsbGx2L9/P9asWQMASEhIQFZWFhISEjBnzhxs2rRJvc+mTZuwYsUK2Gw2xMTE\nYPbs2RpvinxGP9eZ+WVTZAfwkyI7gM+Mv+9op8+ffTRjxgzMmDEDADBq1Cjs27fvsrd75JFH8Mgj\nj/RYf/PNN+Pjjz/2MSYREQUCP/uIqB/42Ue+j8PXtX742UdERKQLFgUNGX1ekvllU2QH8JMiO4DP\njL/vaIdFgYiIVOwpEPUDewq+j8PXtX7YUyAiIl2wKGjI6POSzC+bIjuAnxTZAXxm/H1HOywKRESk\nYk+BqB/YU/B9HL6u9cOeAhER6YJFQUNGn5dkftkU2QH8pMgO4DPj7zvaYVEgIiIVewpE/cCegu/j\n8HWtH/YUiIhIFywKGjL6vCTzy6bIDuAnRXYAnxl/39EOiwIREanYUyDqB/YUfB+Hr2v9sKdARES6\nYFHQkNHnJZlfNkV2AD8psgP4zPj7jnZYFIiISMWeAlE/sKfg+zh8XeuHPQUiItIFi4KGjD4vyfyy\nKbID+EmRHcBnxt93tMOiQEREKvYUiPqBPQXfx+HrWj/sKRARkS5YFDRk9HlJ5pdNkR3AT4rsAD4z\n/r6jHRYFIiJSXbUofP/997jllluQkpKChIQErF27FgDQ3NwMh8OB2NhYZGRkoKWlRb1PYWEhbDYb\n4uPjUVFRoa6vrq5GUlISbDYbCgoKdNocudLS0mRH8Avzy5YmO4Cf0mQH8Jnx9x3tXLUoDBs2DJWV\nlfjwww/x0UcfobKyEu+++y6KiorgcDhQW1uL9PR0FBUVAQBqamqwfft21NTUwOl0Ij8/X21+5OXl\nobi4GC6XCy6XC06nU/+tIyKiful1+mj48OEAgPb2dnR0dCAyMhK7du1CTk4OACAnJwc7d+4EAJSX\nlyM7OxthYWGwWq2IiYlBVVUVPB4PWltbYbfbAQBLly5V7zOQGH1ekvllU2QH8JMiO4DPjL/vaCe0\ntxucP38ekydPxv/+9z/k5eXhpptugtfrhclkAgCYTCZ4vV4AQFNTE6ZNm6be12KxwO12IywsDBaL\nRV1vNpvhdruvOGZubi6sVisAICIiAikpKerhXed/Hpe5LGP5AgVdUyXKT/9qvYxefq7Vcuc6vR6/\nc/mnpSD7/7y0GARLnr7kVRQF9fX10Fqfr1M4c+YMZs2ahcLCQixYsACnT59WfzZq1Cg0Nzdj1apV\nmDZtGu655x4AwIoVKzBnzhxYrVasWbMGe/fuBQAcPHgQTz31FN56662egXidAgUxXqfg+zh8XetH\nynUK1157LX7zm9+guroaJpMJJ0+eBAB4PB6MGTMGwIUjgIaGBvU+jY2NsFgsMJvNaGxs7LbebDZr\nsgFERKSdqxaFr7/+Wj2zqK2tDXv37kVqaioyMzNRUlICACgpKcG8efMAAJmZmSgrK0N7ezvq6urg\ncrlgt9sRFRWF8PBwVFVVQQiB0tJS9T4DidHnJZlfNkV2AD8psgP4zPj7jnau2lPweDzIycnB+fPn\ncf78eSxZsgTp6elITU1FVlYWiouLYbVasWPHDgBAQkICsrKykJCQgNDQUGzatOmnw21g06ZNyM3N\nRVtbG+bOnYvZs2frv3VERNQv/Owjon5gT8H3cfi61g8/+4iIiHTBoqAho89LMr9siuwAflJkB/CZ\n8fcd7bAoEBGRij0Fon5gT8H3cfi61g97CkREpAsWBQ0ZfV6S+WVTZAfwkyI7gM+Mv+9oh0WBiIhU\n7CkQ9QN7Cr6Pw9e1fthTICIiXbAoaMjo85LML5siO4CfFNkBfGb8fUc7LApERKRiT4GoH9hT8H0c\nvq71w54CERHpgkVBQ0afl2R+2RTZAfykyA7gM+PvO9rp9W80ExnF3Ll3oq3tG9kxiAyNPQUaMAIz\n3z/w5vrZUzA+9hSIiEgXLAoaMvq8pNHzG58iO4CfFNkBfMZ9vwuLAhERqdhToAGDPYXgHoeva/2w\np0BERLpgUdCQ0ecljZ7f+BTZAfykyA7gM+77XVgUiIhIxZ4CDRjsKQT3OHxd64c9BSIi0gWLgoaM\nPi9p9PzGp8gO4CdFdgCfcd/vwqJAREQq9hRowGBPIbjH4etaPwHtKTQ0NOC2227DTTfdhMTERDz/\n/PMAgObmZjgcDsTGxiIjIwMtLS3qfQoLC2Gz2RAfH4+Kigp1fXV1NZKSkmCz2VBQUKDJBhARkXZ6\nLQphYWF47rnn8Mknn+DQoUP4+9//jmPHjqGoqAgOhwO1tbVIT09HUVERAKCmpgbbt29HTU0NnE4n\n8vPz1QqWl5eH4uJiuFwuuFwuOJ1OfbcuwIw+L2n0/ManyA7gJ0V2AJ9x3+/Sa1GIiopCSkoKAGDE\niBGYOHEi3G43du3ahZycHABATk4Odu7cCQAoLy9HdnY2wsLCYLVaERMTg6qqKng8HrS2tsJutwMA\nli5dqt6HiIiCQ7/+yE59fT0++OAD3HLLLfB6vTCZTAAAk8kEr9cLAGhqasK0adPU+1gsFrjdboSF\nhcFisajrzWYz3G73ZcfJzc2F1WoFAERERCAlJQVpaWkAuip6MC6npaUFVZ7Blv8C5ad/9VruXKfH\n46ddtIzL/FyP5c51Wjxe2lV+/tNSkO0vRl3u/L6+vh6aE33U2toqJk+eLN58800hhBARERHdfh4Z\nGSmEEGLlypVi27Zt6vrly5eL1157TRw5ckTMnDlTXX/gwAFxxx139BinH5GIugEgAKHzVyDGGJjj\nkH60fH77dErqjz/+iIULF2LJkiWYN28egAtHBydPngQAeDwejBkzBsCFI4CGhgb1vo2NjbBYLDCb\nzWhsbOy23mw2+1/VgojR5yWNnt/4FNkB/KTIDuAz7vtdei0KQggsX74cCQkJeOCBB9T1mZmZKCkp\nAQCUlJSoxSIzMxNlZWVob29HXV0dXC4X7HY7oqKiEB4ejqqqKgghUFpaqt6HiIiCQ6/XKbz77ru4\n9dZbkZyc/NN54BdOObXb7cjKysKJEydgtVqxY8cOREREAADWr1+PLVu2IDQ0FBs2bMCsWbMAXDgl\nNTc3F21tbZg7d656emu3QLxOgXzE6xSCexy+rvWj5fsmL16jAYNFIbjH4etaP/xAvCBl9HlJo+c3\nPkV2AD8psgP4jPt+FxYFIiJScfqIBgxOHwX3OHxd64fTR0REpAsWBQ0ZfV7S6PmNT5EdwE+K7AA+\n477fpV8fc0FE5JtQ9ZR2vYwcGYmzZ5t1HWMwYE+BBgz2FAb7OIP3vYM9BSIi0gWLgoaMPi9p9PzG\np8gO4CdFdgCfcd/vwqJAREQq9hRowGBPYbCPM3jfO9hTICIiXbAoaMjo85JGz298iuwAflJkB/AZ\n9/0uLApERKRiT4EGDPYUBvs4g/e9gz0FIiLSBYuChow+L2n0/ManyA7gJ0V2AJ9x3+/CokBERCr2\nFGjAYE9hsI8zeN872FMgIiJdsChoyOjzkkbPb3yK7AB+UmQH8Bn3/S4sCkREpGJPgQYM9hQG+ziD\n972DPQUiItIFi4KGjD4vafT8xqfIDuAnRXYAn3Hf78KiQEREKvYUaMBgT2GwjzN43zvYUyAiIl30\nWhSWLVsGk8mEpKQkdV1zczMcDgdiY2ORkZGBlpYW9WeFhYWw2WyIj49HRUWFur66uhpJSUmw2Wwo\nKCjQeDOCg9HnJY2e3/gU2QH8pMgO4DPu+116LQr33XcfnE5nt3VFRUVwOByora1Feno6ioqKAAA1\nNTXYvn07ampq4HQ6kZ+frx7S5OXlobi4GC6XCy6Xq8djEhGRfL0WhenTpyMyMrLbul27diEnJwcA\nkJOTg507dwIAysvLkZ2djbCwMFitVsTExKCqqgoejwetra2w2+0AgKVLl6r3GUjS0tJkR/CL0fMb\nX5rsAH5Kkx3AZ9z3u/jUU/B6vTCZTAAAk8kEr9cLAGhqaoLFYlFvZ7FY4Ha7e6w3m81wu93+5CYi\nIh2E+vsAISEhP531oZ3c3FxYrVYAQEREBFJSUtRK3jn3F4zLF89LBkOewZb/gs5t0Gu5c50ej9/5\n/cW0fPzLLXeu0+LxOr+/3M9xybIW4/Vc9nX/6Vwne//tT15FUVBfXw/NiT6oq6sTiYmJ6nJcXJzw\neDxCCCGamppEXFycEEKIwsJCUVhYqN5u1qxZ4tChQ8Lj8Yj4+Hh1/csvvyzuv//+y47Vx0hBqbKy\nUnYEvxg9PwABCJ2/9ByjMkDj6LU9lQEa58pj+Gog7Pta8Wn6KDMzEyUlJQCAkpISzJs3T11fVlaG\n9vZ21NXVweVywW63IyoqCuHh4aiqqoIQAqWlpep9BhKjz0saPb/xpckO4Kc02QF8xn3/Ir1VjcWL\nF4uxY8eKsLAwYbFYxJYtW8SpU6dEenq6sNlswuFwiNOnT6u3/+tf/yqio6NFXFyccDqd6vojR46I\nxMREER0dLVatWnXF8foQieiyEKDfRo33G/xgGWfwvndoue28ollDiqIY+jcOo+c3/hXNCrp+2zbi\nlcYKrny0ENxXNA+EfV+r901e0UxERCoeKdCAYfwjBY7j7xiD9b2DRwpERKQLFgUNGf3zU4ye3/gU\n2QH8pMgO4DPu+11YFIiISMWeAg0Y7CkM9nEG73sHewpERKQLFgUNGX1e0uj5jU+RHcBPiuwAPuO+\n38XvD8QjIgoOoZp/OOfljBwZibNnm3UfRxb2FGjAYE9hsI8TuG0Jtvco9hSIiEgXLAoaMvq8pNHz\nG58iO4CfFNkB/KDIDhA0WBSIiEjFngINGOwpDPZx2FPQAo8UiIhIxaKgIaPPyRs9v/EpsgP4SZEd\nwA+K7ABBg0WBiIhU7CnQgMGewmAfhz0FLfBIgYiIVCwKGjL6nLzR8xufIjuAnxTZAfygyA4QNFgU\niIhIxZ4CDRjsKQz2cdhT0AKPFIiISMWioCGjz8nrlT88fBRCQkJ0/zI+RXYAPymyA/hBkR0gaPDv\nKZDuWltPI3BTFETkD/YUSHeBmesHBtq8NccJxjEujBNs71HsKRARkS5YFDTEngL5R5EdwE+K7AB+\nUGQHCBoBLwpOpxPx8fGw2Wx48sknAz28rj788EPZEfxi9PzGZ/Tn38j5jZxdWwEtCh0dHVi5ciWc\nTidqamrwyiuv4NixY4GMoKuWlhbZEfrl0rOCHnzwQZ4VJJWx9p+ejJzfyNm1FdCicPjwYcTExMBq\ntSIsLAyLFy9GeXl5ICPQRbrOCur8+ssly1p9EZFRBLQouN1ujBs3Tl22WCxwu92BjOCXv/9901V/\nI3788cc1+c3aYrEE5Nz+nuoD/ZRSN/WyA/ipXnYAP9TLDhA0AnqdQl+nEgb7lENgC+Wlz3VJgMbR\nSyDG0XOMi59/Iz5nV9t/gv3/pu/7/kB+jwpoUTCbzWhoaFCXGxoaYLFYut0m2M7/JSIaTAI6fTRl\nyhS4XC7U19ejvb0d27dvR2ZmZiAjEBHRVQT0SCE0NBQvvPACZs2ahY6ODixfvhwTJ04MZAQiIrqK\ngF+nUFtbi2uuuQbDhg3D8OHDAQDr1q2DxWJBamoqUlNTsWfPHvX2hYWFsNlsiI+PR0VFRaDjdrNh\nwwYkJSUhMTERGzZsUNdv3LgREydORGJiIlavXq2uD6bswOXz33333erzPmHCBKSmpqq3N0L+w4cP\nw263IzU1FVOnTsX777+v3t4I+f/zn//gl7/8JZKTk5GZmYnW1lb19rLzL1u2DCaTCUlJSeq65uZm\nOBwOxMbGIiMjo9tp2FfKW11djaSkJNhsNhQUFARl/ubmZtx2220YOXIkVq1a1e1xZOTvT/a9e/di\nypQpSE5OxpQpU1BZWelfdhFAH3/8sUhMTBRtbW3i3LlzYubMmeKzzz4T69atE88880yP23/yySdi\n0qRJor29XdTV1Yno6GjR0dERyMiqK2Xfv3+/mDlzpmhvbxdCCPHll18GXfar5b/YQw89JJ544gkh\nhHHyz5iD6CFoAAAFl0lEQVQxQzidTiGEELt37xZpaWmGyj9lyhRx4MABIYQQW7ZsEY8++mjQ5D9w\n4IA4evSoSExMVNc9/PDD4sknnxRCCFFUVCRWr159xbznz58XQggxdepUUVVVJYQQYs6cOWLPnj1B\nl//bb78V7777rti8ebNYuXJlt8eRkb8/2T/44APh8XiEEEL897//FWaz2a/sAT1S+PTTT3HLLbdg\n2LBhGDp0KGbMmIE33nijszj1uH15eTmys7MRFhYGq9WKmJgYHD58OJCRVVfKvnnzZqxduxZhYWEA\ngNGjRwdd9qvl7ySEwI4dO5CdnQ3AOPmvv/56nDlzBsCFiwfNZrNh8r/++utwuVyYPn06AGDmzJl4\n/fXXgyb/9OnTERkZ2W3drl27kJOTAwDIycnBzp07r5i3qqoKHo8Hra2tsNvtAIClS5eq9wmm/MOH\nD8evf/1r/OxnP+t2e1n5+5M9JSUFUVFRAICEhAS0tbXhxx9/9Dl7QItCYmIiDh48iObmZnz33XfY\nvXu3ejbSxo0bMWnSJCxfvlw9LGpqaup2dpLM6xqulL22thYHDhzAtGnTkJaWhiNHjgRddqBn/rff\nfhuNjY3qzw8ePAiTyYTo6GgAxslfVFSEP/7xjxg/fjwefvhhFBYWAgj+/Lt370ZjYyMSExPVCzhf\nffVV9fUQbPk7eb1emEwmAIDJZILX6wVw5byXrjebzVK340r5O116qqnb7Q6a/L1lB4DXX38dN998\nM8LCwnzOHtCiEB8fj9WrVyMjIwNz5sxBSkoKhg4divz8fNTV1eHDDz/E2LFj8dBDD13xMWSdH3yl\n7OfOncPp06dx6NAhPP3008jKyrriY8g8t/nS/KmpqRgypOu//5VXXsFvf/vbqz5GMOZfvnw5Nm7c\niBMnTuC5557DsmXLrvgYwZS/c/8pLi7Gpk2bMGXKFHzzzTe45pprrvgYwXZuvNE/xsTI+S+X/ZNP\nPsGaNWvw4osv+vXYAW80L1u2DEeOHME777yDiIgIxMXFYfTo0epGrlixQj1MvvS6hsbGRnV6QIaL\ns0dGRiI2NhYWiwULFiwAAEydOhVDhgzB119/HXTZgcs/9wBw7tw5vPnmm7j77rvV2wZ7/s7nv6qq\nCvPnzwcALFq0KGj3HeDyz39cXBz+9a9/4ciRI1i8eLF6pBaM+YELv6GePHkSwIWplTFjxgC4fF6L\nxQKz2dztiFT2dlwp/5UEU/6rZW9sbMSCBQtQWlqKCRMmAPAju0Z9kT7zer1CCCG++OILER8fL86c\nOSOamprUnz/77LMiOztbCNHVvPrhhx/E559/Lm688Ua1eSXD5bJv3rxZPPbYY0IIIY4fPy7GjRsX\nlNmFuHx+IYTYs2eP2qDtZIT8LS0tIjU1VSiKIoQQYt++fWLKlClCCGPkP3PmjHpiQkdHh1iyZInY\nunWrECJ48tfV1fVodhYVFQkhhCgsLOzRaL5cXrvdLg4dOiTOnz8f0EZzf/J32rp1a49Gs6z8fc1+\n+vRpkZycLN58880ej+FL9oAXhenTp4uEhAQxadIksX//fiGEEEuWLBFJSUkiOTlZ3HXXXeLkyZPq\n7f/617+K6OhoERcXp55lIsvlsre3t4t7771XJCYmismTJ4vKykr19sGUXYjL5xdCiNzcXPHiiy/2\nuL0R8r///vvCbreLSZMmiWnTpomjR4+qtzdC/g0bNojY2FgRGxsr1q5d2+32svMvXrxYjB07VoSF\nhQmLxSK2bNkiTp06JdLT04XNZhMOh0OcPn2617xHjhwRiYmJIjo6WqxatSpo899www1i1KhRYsSI\nEcJisYhjx45Jy9+f7E888YT4+c9/LlJSUtSvr776yufsQffnOImISB7+5TUiIlKxKBARkYpFgYiI\nVCwKRESkYlEgIiIViwIREan+H4IZap9MfYboAAAAAElFTkSuQmCC\n",
       "text": [
        "<matplotlib.figure.Figure at 0x4af65d10>"
       ]
      }
     ],
     "prompt_number": 313
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "nonwarnings_day.hist('reading')"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [
      {
       "metadata": {},
       "output_type": "pyout",
       "prompt_number": 314,
       "text": [
        "array([[<matplotlib.axes.AxesSubplot object at 0x4b016810>]], dtype=object)"
       ]
      },
      {
       "metadata": {},
       "output_type": "display_data",
       "png": "iVBORw0KGgoAAAANSUhEUgAAAZEAAAEICAYAAACeSMncAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\n2zhMAAAAAElFTkSuQmCC\n",
       "text": [
        "<matplotlib.figure.Figure at 0x4a3ad590>"
       ]
      }
     ],
     "prompt_number": 314
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "days.hist('reading')"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [
      {
       "metadata": {},
       "output_type": "pyout",
       "prompt_number": 315,
       "text": [
        "array([[<matplotlib.axes.AxesSubplot object at 0x4b20bc90>]], dtype=object)"
       ]
      },
      {
       "metadata": {},
       "output_type": "display_data",
       "png": "iVBORw0KGgoAAAANSUhEUgAAAZEAAAEICAYAAACeSMncAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nTkSuQmCC\n",
       "text": [
        "<matplotlib.figure.Figure at 0x7b117a10>"
       ]
      }
     ],
     "prompt_number": 315
    },
    {
     "cell_type": "markdown",
     "metadata": {},
     "source": [
      "My conclusion is that there was some truth to my hypothesis: the distribution of atmospheric pressure readings on days with heat warnings is different from the distribution on days without heat warnings.  Although I still don't understand the exact relationship between air pressure and temperature.\n",
      "\n",
      "If I were to return to this problem in the future I would look further into trends for each of the different warning codes , recalling that code 1 (extreme heat warning downgraded to standard heat warning) days had the lowest mean reading of all the data sets."
     ]
    }
   ],
   "metadata": {}
  }
 ]
}