twitter-under-the-weather.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# How does the weather influence my Social Media behavior?\n",
    "Mental well-being might well be tied to the weather, e.g. you might be feeling blue after days of rain without a break. Similarly, your social media behaviour might be tied to the weather. E.g. you might post less if you're outside enjoying the sun. Or you might post more up-beat posts when the weather's nice. \n",
    "\n",
    "This notebook brings together the data from your Twitter archive and your GPS data from Moves to correlate your Twitter and Weather data. \n",
    "\n",
    "For this notebook to work you need to have both the [Twitter Archive Analyser](https://www.openhumans.org/activity/twitter-archive-analyzer/) and the [Moves connection](https://www.openhumans.org/activity/moves-connection/) added to your *Open Humans* account and already need to have this data available in your account (i.e. you need to have uploaded your Twitter archive on the Twitter archive analyser site and need to have connected your Moves account to your Open Humans account).\n",
    "\n",
    "The weather data comes from *Dark Sky* and their API. You will need a `DARKSKY_KEY` for their API to use this notebook. You can [create one for free on their website](https://darksky.net/dev). Dark Sky allows you to make 1,000 calls per day for free. This notebook makes at worst 1 call for each day that is in your `DATARANGE`. So the 1,000 calls can get you ~ 2 1/2 years of local weather data. Once you have made an account you can enter you `DARKSKY_KEY` in the cell below. \n",
    "\n",
    "The `DATARANGE_START` and `DATARANGE_END` give the boundaries for which your analysis should be run (in `YYYY-MM-DD` format). Choose the values so that you have some *Moves* data between those boundaries and also that you're not requesting too much data from *Dark Sky* (or pay for requesting more data from *dark sky*, after the first 1,000 calls each day of weather data will be USD 0.0001).\n",
    "\n",
    "You might consider paying for additional API calls to *Dark Sky* as it's rather cheap: Making 10,000 calls costs US $1.00 and you can [set up daily usage limits](https://darksky.net/dev/account/edit) to make sure you don't accumulate too much costs."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "DARKSKY_KEY = 'ENTER_YOUR_KEY_HERE'\n",
    "DATARANGE_START = \"2016-06-01\"\n",
    "DATARANGE_END = \"2018-05-08\""
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Let's load our dependencies for getting & analysing the *Moves* and *Twitter Archive* data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "import os\n",
    "import json\n",
    "import requests\n",
    "from datetime import datetime\n",
    "from collections import defaultdict\n",
    "import pandas as pd\n",
    "import numpy as np\n",
    "import seaborn as sns\n",
    "import matplotlib.pyplot as plt\n",
    "import tempfile\n",
    "import zipfile\n",
    "import pytz\n",
    "import io\n",
    "import sys\n",
    "from textblob import TextBlob\n",
    "import emoji\n",
    "from ohapi import api\n",
    "\n",
    "# sets the axis label sizes for seaborn\n",
    "rc={'font.size': 14, 'axes.labelsize': 14, 'legend.fontsize': 14.0, \n",
    "    'axes.titlesize': 14, 'xtick.labelsize': 14, 'ytick.labelsize': 14}\n",
    "sns.set(rc=rc)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Processing the Twitter archive data\n",
    "All the functions are already provided by http://twarxiv.org and we just copied the functions from there. It will basically take a zipped archive from a URL and return a nicely formatted `pandas` dataframe. Let's just take these as given for now:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [],
   "source": [
    "# THIS CODE BELOW IS COPIED FROM TWARXIV.ORG AS IT ALREADY DOES EXACTLY WHAT WE WANT FOR READING IN THE DATA\n",
    "\n",
    "# READ JSON FILES FROM TWITTER ARCHIVE!\n",
    "\n",
    "def check_hashtag(single_tweet):\n",
    "    '''check whether tweet has any hashtags'''\n",
    "    return len(single_tweet['entities']['hashtags']) > 0\n",
    "\n",
    "\n",
    "def check_media(single_tweet):\n",
    "    '''check whether tweet has any media attached'''\n",
    "    return len(single_tweet['entities']['media']) > 0\n",
    "\n",
    "\n",
    "def check_url(single_tweet):\n",
    "    '''check whether tweet has any urls attached'''\n",
    "    return len(single_tweet['entities']['urls']) > 0\n",
    "\n",
    "\n",
    "def check_retweet(single_tweet):\n",
    "    '''\n",
    "    check whether tweet is a RT. If yes:\n",
    "    return name & user name of the RT'd user.\n",
    "    otherwise just return nones\n",
    "    '''\n",
    "    if 'retweeted_status' in single_tweet.keys():\n",
    "        return (single_tweet['retweeted_status']['user']['screen_name'],\n",
    "                single_tweet['retweeted_status']['user']['name'])\n",
    "    else:\n",
    "        return (None, None)\n",
    "\n",
    "\n",
    "def check_coordinates(single_tweet):\n",
    "    '''\n",
    "    check whether tweet has coordinates attached.\n",
    "    if yes return the coordinates\n",
    "    otherwise just return nones\n",
    "    '''\n",
    "    if 'coordinates' in single_tweet['geo'].keys():\n",
    "        return (single_tweet['geo']['coordinates'][0],\n",
    "                single_tweet['geo']['coordinates'][1])\n",
    "    else:\n",
    "        return (None, None)\n",
    "\n",
    "\n",
    "def check_reply_to(single_tweet):\n",
    "    '''\n",
    "    check whether tweet is a reply. If yes:\n",
    "    return name & user name of the user that's replied to.\n",
    "    otherwise just return nones\n",
    "    '''\n",
    "    if 'in_reply_to_screen_name' in single_tweet.keys():\n",
    "        name = None\n",
    "        for user in single_tweet['entities']['user_mentions']:\n",
    "            if user['screen_name'] == single_tweet['in_reply_to_screen_name']:\n",
    "                name = user['name']\n",
    "                break\n",
    "        return (single_tweet['in_reply_to_screen_name'], name)\n",
    "    else:\n",
    "        return (None, None)\n",
    "\n",
    "\n",
    "def create_dataframe(tweets):\n",
    "    '''\n",
    "    create a pandas dataframe from our tweet jsons\n",
    "    '''\n",
    "\n",
    "    # initalize empty lists\n",
    "    utc_time = []\n",
    "    longitude = []\n",
    "    latitude = []\n",
    "    hashtag = []\n",
    "    media = []\n",
    "    url = []\n",
    "    retweet_user_name = []\n",
    "    retweet_name = []\n",
    "    reply_user_name = []\n",
    "    reply_name = []\n",
    "    text = []\n",
    "    # iterate over all tweets and extract data\n",
    "    for single_tweet in tweets:\n",
    "        utc_time.append(datetime.strptime(single_tweet['created_at'],\n",
    "                                                   '%Y-%m-%d %H:%M:%S %z'))\n",
    "        coordinates = check_coordinates(single_tweet)\n",
    "        latitude.append(coordinates[0])\n",
    "        longitude.append(coordinates[1])\n",
    "        hashtag.append(check_hashtag(single_tweet))\n",
    "        media.append(check_media(single_tweet))\n",
    "        url.append(check_url(single_tweet))\n",
    "        retweet = check_retweet(single_tweet)\n",
    "        retweet_user_name.append(retweet[0])\n",
    "        retweet_name.append(retweet[1])\n",
    "        reply = check_reply_to(single_tweet)\n",
    "        reply_user_name.append(reply[0])\n",
    "        reply_name.append(reply[1])\n",
    "        text.append(single_tweet['text'])\n",
    "    # convert the whole shebang into a pandas dataframe\n",
    "    dataframe = pd.DataFrame(data={\n",
    "                            'utc_time': utc_time,\n",
    "                            'latitude': latitude,\n",
    "                            'longitude': longitude,\n",
    "                            'hashtag': hashtag,\n",
    "                            'media': media,\n",
    "                            'url': url,\n",
    "                            'retweet_user_name': retweet_user_name,\n",
    "                            'retweet_name': retweet_name,\n",
    "                            'reply_user_name': reply_user_name,\n",
    "                            'reply_name': reply_name,\n",
    "                            'text': text\n",
    "    })\n",
    "    return dataframe\n",
    "\n",
    "\n",
    "def read_files(zip_url):\n",
    "    tf = tempfile.NamedTemporaryFile()\n",
    "    print('downloading files')\n",
    "    tf.write(requests.get(zip_url).content)\n",
    "    tf.flush()\n",
    "    zf = zipfile.ZipFile(tf.name)\n",
    "    print('reading index')\n",
    "    with zf.open('data/js/tweet_index.js', 'r') as f:\n",
    "        f = io.TextIOWrapper(f)\n",
    "        d = f.readlines()[1:]\n",
    "        d = \"[{\" + \"\".join(d)\n",
    "        json_files = json.loads(d)\n",
    "    data_frames = []\n",
    "    print('iterate over individual files')\n",
    "    for single_file in json_files:\n",
    "        print('read ' + single_file['file_name'])\n",
    "        with zf.open(single_file['file_name']) as f:\n",
    "            f = io.TextIOWrapper(f)\n",
    "            d = f.readlines()[1:]\n",
    "            d = \"\".join(d)\n",
    "            tweets = json.loads(d)\n",
    "            df_tweets = create_dataframe(tweets)\n",
    "            data_frames.append(df_tweets)\n",
    "    return data_frames\n",
    "\n",
    "\n",
    "def create_main_dataframe(zip_url='http://ruleofthirds.de/test_archive.zip'):\n",
    "    print('reading files')\n",
    "    dataframes = read_files(zip_url)\n",
    "    print('concatenating...')\n",
    "    dataframe = pd.concat(dataframes)\n",
    "    dataframe = dataframe.sort_values('utc_time', ascending=False)\n",
    "    dataframe = dataframe.set_index('utc_time')\n",
    "    dataframe = dataframe.replace(to_replace={\n",
    "                                    'url': {False: None},\n",
    "                                    'hashtag': {False: None},\n",
    "                                    'media': {False: None}\n",
    "                                    })\n",
    "    return dataframe\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Processing the Moves data\n",
    "This is copied more or less verbatim from the `moves-analysis.ipynb` notebook that's a Personal Data Notebook example. Let's take this code for given right now too: "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [],
   "source": [
    "weather_cache = {}\n",
    "\n",
    "def longest_daily_location(daily_segments):\n",
    "    \"\"\"\n",
    "    takes a daily segment log of Moves and returns the \n",
    "    lat/long for the location where most time was spent. \n",
    "    Can be misleading for days w/ lots of travel etc. \n",
    "    But the most quick/dirty solution for now.\n",
    "    \"\"\"\n",
    "    places_of_day = []\n",
    "    for i in daily_segments:\n",
    "        if i['type'] == 'place':\n",
    "            place_location = i['place']['location']\n",
    "            start_time = datetime.strptime(i['startTime'],'%Y%m%dT%H%M%S%z')\n",
    "            end_time = datetime.strptime(i['endTime'],'%Y%m%dT%H%M%S%z')\n",
    "            duration = end_time - start_time\n",
    "            places_of_day.append([place_location,duration])\n",
    "    places_of_day.sort(key=lambda tup: tup[-1],reverse=True)\n",
    "    return places_of_day[0][0]\n",
    "\n",
    "def weather_for_location_day(location, date):\n",
    "    \"\"\"\n",
    "    take a location as returned by longest_daily_location \n",
    "    and a date of YYYYMMDD to get the weather from the \n",
    "    Dark Sky API. Does not do any filtering yet\n",
    "    \"\"\"\n",
    "    if DARKSKY_KEY == 'ENTER_YOUR_KEY_HERE':\n",
    "        raise ValueError(\"Oops, you didn't add your Dark Sky API key! \"\n",
    "              \"Scroll back up to the first step in this notebook. :)\")\n",
    "    date = '{}-{}-{}T12:00:00'.format(\n",
    "    date[:4],\n",
    "    date[4:6],\n",
    "    date[6:8])\n",
    "\n",
    "    if date not in weather_cache:\n",
    "        url = 'https://api.darksky.net/forecast/{}/{},{},{}?units=si'.format(DARKSKY_KEY,\n",
    "            location['lat'],\n",
    "            location['lon'],\n",
    "            date\n",
    "        )\n",
    "        response = requests.get(url)\n",
    "        weather_cache[date] = response.json()\n",
    "\n",
    "    return weather_cache[date]"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Importing *Moves* and *Twitter Data*\n",
    "We can now start importing our data from the two data sources. The *Twitter* data will already come in a nicely packaged data frame. For the *Moves* data we will have to make our weather API calls afterwards. "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "reading files\n",
      "downloading files\n",
      "reading index\n",
      "iterate over individual files\n",
      "read data/js/tweets/2018_05.js\n",
      "read data/js/tweets/2018_04.js\n",
      "read data/js/tweets/2018_03.js\n",
      "read data/js/tweets/2018_02.js\n",
      "read data/js/tweets/2018_01.js\n",
      "read data/js/tweets/2017_12.js\n",
      "read data/js/tweets/2017_11.js\n",
      "read data/js/tweets/2017_10.js\n",
      "read data/js/tweets/2017_09.js\n",
      "read data/js/tweets/2017_08.js\n",
      "read data/js/tweets/2017_07.js\n",
      "read data/js/tweets/2017_06.js\n",
      "read data/js/tweets/2017_05.js\n",
      "read data/js/tweets/2017_04.js\n",
      "read data/js/tweets/2017_03.js\n",
      "read data/js/tweets/2017_02.js\n",
      "read data/js/tweets/2017_01.js\n",
      "read data/js/tweets/2016_12.js\n",
      "read data/js/tweets/2016_11.js\n",
      "read data/js/tweets/2016_10.js\n",
      "read data/js/tweets/2016_09.js\n",
      "read data/js/tweets/2016_08.js\n",
      "read data/js/tweets/2016_07.js\n",
      "read data/js/tweets/2016_06.js\n",
      "read data/js/tweets/2016_05.js\n",
      "read data/js/tweets/2016_04.js\n",
      "read data/js/tweets/2016_03.js\n",
      "read data/js/tweets/2016_02.js\n",
      "read data/js/tweets/2016_01.js\n",
      "read data/js/tweets/2015_12.js\n",
      "read data/js/tweets/2015_11.js\n",
      "read data/js/tweets/2015_10.js\n",
      "read data/js/tweets/2015_09.js\n",
      "read data/js/tweets/2015_08.js\n",
      "read data/js/tweets/2015_07.js\n",
      "read data/js/tweets/2015_06.js\n",
      "read data/js/tweets/2015_05.js\n",
      "read data/js/tweets/2015_04.js\n",
      "read data/js/tweets/2015_03.js\n",
      "read data/js/tweets/2015_02.js\n",
      "read data/js/tweets/2015_01.js\n",
      "read data/js/tweets/2014_12.js\n",
      "read data/js/tweets/2014_11.js\n",
      "read data/js/tweets/2014_10.js\n",
      "read data/js/tweets/2014_09.js\n",
      "read data/js/tweets/2014_08.js\n",
      "read data/js/tweets/2014_07.js\n",
      "read data/js/tweets/2014_06.js\n",
      "read data/js/tweets/2014_05.js\n",
      "read data/js/tweets/2014_04.js\n",
      "read data/js/tweets/2014_03.js\n",
      "read data/js/tweets/2014_02.js\n",
      "read data/js/tweets/2014_01.js\n",
      "read data/js/tweets/2013_12.js\n",
      "read data/js/tweets/2013_11.js\n",
      "read data/js/tweets/2013_10.js\n",
      "read data/js/tweets/2013_09.js\n",
      "read data/js/tweets/2013_08.js\n",
      "read data/js/tweets/2013_07.js\n",
      "read data/js/tweets/2013_06.js\n",
      "read data/js/tweets/2013_05.js\n",
      "read data/js/tweets/2013_04.js\n",
      "read data/js/tweets/2013_03.js\n",
      "read data/js/tweets/2013_02.js\n",
      "read data/js/tweets/2013_01.js\n",
      "read data/js/tweets/2012_12.js\n",
      "read data/js/tweets/2012_11.js\n",
      "read data/js/tweets/2012_10.js\n",
      "read data/js/tweets/2012_09.js\n",
      "read data/js/tweets/2012_08.js\n",
      "read data/js/tweets/2012_07.js\n",
      "read data/js/tweets/2012_06.js\n",
      "read data/js/tweets/2012_05.js\n",
      "read data/js/tweets/2012_04.js\n",
      "read data/js/tweets/2012_03.js\n",
      "read data/js/tweets/2012_02.js\n",
      "read data/js/tweets/2012_01.js\n",
      "read data/js/tweets/2011_12.js\n",
      "read data/js/tweets/2011_11.js\n",
      "read data/js/tweets/2011_10.js\n",
      "read data/js/tweets/2011_09.js\n",
      "read data/js/tweets/2011_08.js\n",
      "read data/js/tweets/2011_07.js\n",
      "read data/js/tweets/2011_06.js\n",
      "read data/js/tweets/2011_05.js\n",
      "read data/js/tweets/2011_04.js\n",
      "read data/js/tweets/2011_03.js\n",
      "read data/js/tweets/2011_02.js\n",
      "read data/js/tweets/2011_01.js\n",
      "read data/js/tweets/2010_12.js\n",
      "read data/js/tweets/2010_11.js\n",
      "read data/js/tweets/2010_10.js\n",
      "read data/js/tweets/2010_09.js\n",
      "read data/js/tweets/2010_08.js\n",
      "read data/js/tweets/2010_07.js\n",
      "read data/js/tweets/2010_06.js\n",
      "read data/js/tweets/2010_05.js\n",
      "read data/js/tweets/2010_04.js\n",
      "read data/js/tweets/2010_03.js\n",
      "read data/js/tweets/2010_02.js\n",
      "read data/js/tweets/2010_01.js\n",
      "read data/js/tweets/2009_12.js\n",
      "read data/js/tweets/2009_11.js\n",
      "read data/js/tweets/2009_10.js\n",
      "read data/js/tweets/2009_09.js\n",
      "read data/js/tweets/2009_08.js\n",
      "read data/js/tweets/2009_07.js\n",
      "read data/js/tweets/2009_06.js\n",
      "read data/js/tweets/2009_05.js\n",
      "read data/js/tweets/2009_04.js\n",
      "read data/js/tweets/2009_03.js\n",
      "read data/js/tweets/2009_02.js\n",
      "read data/js/tweets/2009_01.js\n",
      "read data/js/tweets/2008_12.js\n",
      "read data/js/tweets/2008_11.js\n",
      "read data/js/tweets/2008_10.js\n",
      "read data/js/tweets/2008_09.js\n",
      "read data/js/tweets/2008_08.js\n",
      "read data/js/tweets/2008_07.js\n",
      "read data/js/tweets/2008_06.js\n",
      "read data/js/tweets/2008_05.js\n",
      "read data/js/tweets/2008_04.js\n",
      "concatenating...\n"
     ]
    }
   ],
   "source": [
    "user = api.exchange_oauth2_member(os.environ.get('OH_ACCESS_TOKEN'))\n",
    "has_twitter = False\n",
    "has_moves = False\n",
    "\n",
    "# get our download URLs\n",
    "for entry in user['data']:\n",
    "    if entry['source'] == \"direct-sharing-70\":\n",
    "        twitter_data_url = entry['download_url']\n",
    "        has_twitter = True\n",
    "    if entry['source'] == \"direct-sharing-138\":\n",
    "        moves_data_url = entry['download_url']\n",
    "        has_moves = True\n",
    "\n",
    "if not has_twitter:\n",
    "    print(\"YOU NEED TO HAVE SOME TWITTER DATA IN YOUR ACCOUNT TO USE THIS NOTEBOOK\")\n",
    "    print(\"GO TO http://twarxiv.org TO UPLOAD IT\")\n",
    "\n",
    "if not has_twitter:\n",
    "    print(\"YOU NEED TO HAVE SOME MOVES DATA IN YOUR ACCOUNT TO USE THIS NOTEBOOK\")\n",
    "    print(\"GO TO https://ohmovessource.herokuapp.com/ TO UPLOAD IT\")\n",
    "    \n",
    "# read the twitter data\n",
    "twitter_data = create_main_dataframe(zip_url=twitter_data_url)\n",
    "\n",
    "# read the moves data\n",
    "moves_data_raw = requests.get(moves_data_url).content\n",
    "moves_data = json.loads(moves_data_raw)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Let's quickly look at the Twitter dataframe to see whether the data was loaded correctly:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>hashtag</th>\n",
       "      <th>latitude</th>\n",
       "      <th>longitude</th>\n",
       "      <th>media</th>\n",
       "      <th>reply_name</th>\n",
       "      <th>reply_user_name</th>\n",
       "      <th>retweet_name</th>\n",
       "      <th>retweet_user_name</th>\n",
       "      <th>text</th>\n",
       "      <th>url</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>utc_time</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>2018-05-11 19:58:25+00:00</th>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>Mama Hörnchen ♿️</td>\n",
       "      <td>MamsellChaos</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "      <td>@MamsellChaos oh, and https://t.co/c2N4g6BcUT ...</td>\n",
       "      <td>1.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2018-05-11 19:38:09+00:00</th>\n",
       "      <td>1.0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "      <td>The collection of personal data analysis noteb...</td>\n",
       "      <td>1.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2018-05-11 19:19:56+00:00</th>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>Mama Hörnchen ♿️</td>\n",
       "      <td>MamsellChaos</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "      <td>@MamsellChaos und ich update dich sobald der i...</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2018-05-11 19:19:39+00:00</th>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>Mama Hörnchen ♿️</td>\n",
       "      <td>MamsellChaos</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "      <td>@MamsellChaos fuer R kann ich https://t.co/n2H...</td>\n",
       "      <td>1.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2018-05-11 19:10:14+00:00</th>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>Mama Hörnchen ♿️</td>\n",
       "      <td>MamsellChaos</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "      <td>@MamsellChaos Ah, der Fitbit Data Import funkt...</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                           hashtag latitude longitude  media  \\\n",
       "utc_time                                                       \n",
       "2018-05-11 19:58:25+00:00      NaN      NaN       NaN    NaN   \n",
       "2018-05-11 19:38:09+00:00      1.0      NaN       NaN    NaN   \n",
       "2018-05-11 19:19:56+00:00      NaN      NaN       NaN    NaN   \n",
       "2018-05-11 19:19:39+00:00      NaN      NaN       NaN    NaN   \n",
       "2018-05-11 19:10:14+00:00      NaN      NaN       NaN    NaN   \n",
       "\n",
       "                                 reply_name reply_user_name retweet_name  \\\n",
       "utc_time                                                                   \n",
       "2018-05-11 19:58:25+00:00  Mama Hörnchen ♿️    MamsellChaos         None   \n",
       "2018-05-11 19:38:09+00:00              None            None         None   \n",
       "2018-05-11 19:19:56+00:00  Mama Hörnchen ♿️    MamsellChaos         None   \n",
       "2018-05-11 19:19:39+00:00  Mama Hörnchen ♿️    MamsellChaos         None   \n",
       "2018-05-11 19:10:14+00:00  Mama Hörnchen ♿️    MamsellChaos         None   \n",
       "\n",
       "                          retweet_user_name  \\\n",
       "utc_time                                      \n",
       "2018-05-11 19:58:25+00:00              None   \n",
       "2018-05-11 19:38:09+00:00              None   \n",
       "2018-05-11 19:19:56+00:00              None   \n",
       "2018-05-11 19:19:39+00:00              None   \n",
       "2018-05-11 19:10:14+00:00              None   \n",
       "\n",
       "                                                                        text  \\\n",
       "utc_time                                                                       \n",
       "2018-05-11 19:58:25+00:00  @MamsellChaos oh, and https://t.co/c2N4g6BcUT ...   \n",
       "2018-05-11 19:38:09+00:00  The collection of personal data analysis noteb...   \n",
       "2018-05-11 19:19:56+00:00  @MamsellChaos und ich update dich sobald der i...   \n",
       "2018-05-11 19:19:39+00:00  @MamsellChaos fuer R kann ich https://t.co/n2H...   \n",
       "2018-05-11 19:10:14+00:00  @MamsellChaos Ah, der Fitbit Data Import funkt...   \n",
       "\n",
       "                           url  \n",
       "utc_time                        \n",
       "2018-05-11 19:58:25+00:00  1.0  \n",
       "2018-05-11 19:38:09+00:00  1.0  \n",
       "2018-05-11 19:19:56+00:00  NaN  \n",
       "2018-05-11 19:19:39+00:00  1.0  \n",
       "2018-05-11 19:10:14+00:00  NaN  "
      ]
     },
     "execution_count": 5,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "twitter_data.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "If this looks good we can now go ahead and import the weather data from the Dark Sky's API and convert the data into a data frame too:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [],
   "source": [
    "\n",
    "weather_variables  = ['precipType','precipIntensity','temperatureHigh',\n",
    "                      'temperatureLow','precipIntensityMax','precipAccumulation',\n",
    "                      'apparentTemperatureHigh', 'apparentTemperatureLow']\n",
    "\n",
    "moves_processed_data = defaultdict(list)\n",
    "\n",
    "for datapoint in moves_data:\n",
    "    # we need to have observed segments for that day. If moves wasn't running we ignore the day\n",
    "    if datapoint['segments'] != None:\n",
    "        # did we stay in a place that day and did we walk that day?\n",
    "        has_places = False\n",
    "        walked = False\n",
    "        for i in datapoint['segments']:\n",
    "            if i['type'] == 'place':\n",
    "                    # yes, we were in one place w/o moving around too much, we can keep this day\n",
    "                    has_places = True\n",
    "        # is this day in our date range of interest and has data?\n",
    "        if datapoint['summary'] != None and has_places and datetime.strptime(datapoint['date'],\"%Y%m%d\") > datetime.strptime(DATARANGE_START,\"%Y-%m-%d\"):\n",
    "            moves_processed_data['date'].append(datapoint['date'])\n",
    "            for activity in datapoint['summary']:\n",
    "                if activity['activity'] == 'walking':\n",
    "                    moves_processed_data['steps'].append(activity['steps'])\n",
    "                    moves_processed_data['distance'].append(activity['distance'])\n",
    "                    walked = True\n",
    "            # whops, we have moves data for this day but moves didn't register you walking & your step count is zero\n",
    "            if not walked:\n",
    "                moves_processed_data['steps'].append(0)\n",
    "                moves_processed_data['distance'].append(0)    \n",
    "            # grab the location where we spent most time as baseline geo location for the day\n",
    "            location = longest_daily_location(datapoint['segments'])\n",
    "            moves_processed_data['lat'].append(location['lat'])\n",
    "            moves_processed_data['lon'].append(location['lon'])\n",
    "            # now we can grab the weather for our baseline location\n",
    "            weather = weather_for_location_day(location,datapoint['date'])\n",
    "            for variable in weather_variables:\n",
    "                if variable in weather['daily']['data'][0].keys():\n",
    "                    moves_processed_data[variable].append(weather['daily']['data'][0][variable])\n",
    "                else:\n",
    "                    moves_processed_data[variable].append(0)\n",
    "            if datetime.strptime(datapoint['date'],\"%Y%m%d\") > datetime.strptime(DATARANGE_END,\"%Y-%m-%d\"):\n",
    "                break\n",
    "\n",
    "moves_dataframe = pd.DataFrame(data={\n",
    "    'date': moves_processed_data['date'],\n",
    "    'steps': moves_processed_data['steps'],\n",
    "    'distance': moves_processed_data['distance'],\n",
    "    'latitude': moves_processed_data['lat'],\n",
    "    'longitude': moves_processed_data['lon'],\n",
    "    'precipType': moves_processed_data['precipType'],\n",
    "    'precipIntensity': moves_processed_data['precipIntensity'],\n",
    "    'precipIntensityMax': moves_processed_data['precipIntensityMax'],\n",
    "    'precipAccumulation': moves_processed_data['precipAccumulation'],\n",
    "    'temperatureHigh': moves_processed_data['temperatureHigh'],\n",
    "    'temperatureLow': moves_processed_data['temperatureLow'],\n",
    "    'apparentTemperatureHigh': moves_processed_data['apparentTemperatureHigh'],\n",
    "    'apparentTemperatureLow': moves_processed_data['apparentTemperatureLow']\n",
    "\n",
    "})"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Let's check whether the *Moves* weather look up worked out:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>apparentTemperatureHigh</th>\n",
       "      <th>apparentTemperatureLow</th>\n",
       "      <th>date</th>\n",
       "      <th>distance</th>\n",
       "      <th>latitude</th>\n",
       "      <th>longitude</th>\n",
       "      <th>precipAccumulation</th>\n",
       "      <th>precipIntensity</th>\n",
       "      <th>precipIntensityMax</th>\n",
       "      <th>precipType</th>\n",
       "      <th>steps</th>\n",
       "      <th>temperatureHigh</th>\n",
       "      <th>temperatureLow</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>18.01</td>\n",
       "      <td>10.72</td>\n",
       "      <td>20160617</td>\n",
       "      <td>1759.0</td>\n",
       "      <td>50.060944</td>\n",
       "      <td>8.818717</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.2515</td>\n",
       "      <td>3.5230</td>\n",
       "      <td>rain</td>\n",
       "      <td>992</td>\n",
       "      <td>18.01</td>\n",
       "      <td>10.72</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>18.11</td>\n",
       "      <td>11.41</td>\n",
       "      <td>20160618</td>\n",
       "      <td>5413.0</td>\n",
       "      <td>50.060944</td>\n",
       "      <td>8.818717</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.1499</td>\n",
       "      <td>1.8009</td>\n",
       "      <td>rain</td>\n",
       "      <td>6730</td>\n",
       "      <td>18.11</td>\n",
       "      <td>11.41</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>17.45</td>\n",
       "      <td>7.46</td>\n",
       "      <td>20160619</td>\n",
       "      <td>6471.0</td>\n",
       "      <td>50.060944</td>\n",
       "      <td>8.818717</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0076</td>\n",
       "      <td>0.1575</td>\n",
       "      <td>rain</td>\n",
       "      <td>6395</td>\n",
       "      <td>17.45</td>\n",
       "      <td>7.83</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>21.37</td>\n",
       "      <td>13.95</td>\n",
       "      <td>20160620</td>\n",
       "      <td>4613.0</td>\n",
       "      <td>50.060944</td>\n",
       "      <td>8.818717</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.2235</td>\n",
       "      <td>2.7127</td>\n",
       "      <td>rain</td>\n",
       "      <td>4377</td>\n",
       "      <td>21.37</td>\n",
       "      <td>13.95</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>21.62</td>\n",
       "      <td>17.37</td>\n",
       "      <td>20160621</td>\n",
       "      <td>2238.0</td>\n",
       "      <td>50.114025</td>\n",
       "      <td>8.753432</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.2235</td>\n",
       "      <td>2.3520</td>\n",
       "      <td>rain</td>\n",
       "      <td>2673</td>\n",
       "      <td>21.62</td>\n",
       "      <td>17.10</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   apparentTemperatureHigh  apparentTemperatureLow      date  distance  \\\n",
       "0                    18.01                   10.72  20160617    1759.0   \n",
       "1                    18.11                   11.41  20160618    5413.0   \n",
       "2                    17.45                    7.46  20160619    6471.0   \n",
       "3                    21.37                   13.95  20160620    4613.0   \n",
       "4                    21.62                   17.37  20160621    2238.0   \n",
       "\n",
       "    latitude  longitude  precipAccumulation  precipIntensity  \\\n",
       "0  50.060944   8.818717                 0.0           0.2515   \n",
       "1  50.060944   8.818717                 0.0           0.1499   \n",
       "2  50.060944   8.818717                 0.0           0.0076   \n",
       "3  50.060944   8.818717                 0.0           0.2235   \n",
       "4  50.114025   8.753432                 0.0           0.2235   \n",
       "\n",
       "   precipIntensityMax precipType  steps  temperatureHigh  temperatureLow  \n",
       "0              3.5230       rain    992            18.01           10.72  \n",
       "1              1.8009       rain   6730            18.11           11.41  \n",
       "2              0.1575       rain   6395            17.45            7.83  \n",
       "3              2.7127       rain   4377            21.37           13.95  \n",
       "4              2.3520       rain   2673            21.62           17.10  "
      ]
     },
     "execution_count": 8,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "moves_dataframe.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "If that looks good we can now do a quick sentiment prediction for our *Twitter* data:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [],
   "source": [
    "polarity = []\n",
    "subjectivity = []\n",
    "twitter_data['blob'] = twitter_data['text'].apply(TextBlob)\n",
    "for entry in twitter_data['blob']:\n",
    "    polarity.append(entry.sentiment.polarity)\n",
    "    subjectivity.append(entry.sentiment.subjectivity)\n",
    "twitter_data['polarity'] = polarity\n",
    "twitter_data['subjectivity'] = subjectivity\n",
    "twitter_data = twitter_data.replace(0, np.nan)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Let's now grab the mean sentiment per day and the number of tweets per day and store all of that in `twitter_mean`:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [],
   "source": [
    "twitter_data['date'] = twitter_data.index.date\n",
    "tweets_per_day = twitter_data['date'].value_counts()\n",
    "twitter_mean = twitter_data.groupby(twitter_data.index.date).mean()\n",
    "twitter_mean = twitter_mean.join(tweets_per_day)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>hashtag</th>\n",
       "      <th>latitude</th>\n",
       "      <th>longitude</th>\n",
       "      <th>media</th>\n",
       "      <th>url</th>\n",
       "      <th>polarity</th>\n",
       "      <th>subjectivity</th>\n",
       "      <th>date</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>2008-04-02</th>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0.422222</td>\n",
       "      <td>0.644444</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2008-04-03</th>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0.016667</td>\n",
       "      <td>0.916667</td>\n",
       "      <td>12</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2008-04-04</th>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>-0.333333</td>\n",
       "      <td>0.666667</td>\n",
       "      <td>9</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2008-04-05</th>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0.500000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>6</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2008-04-06</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>3</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "            hashtag  latitude  longitude  media  url  polarity  subjectivity  \\\n",
       "2008-04-02      NaN       NaN        NaN    NaN  NaN  0.422222      0.644444   \n",
       "2008-04-03      NaN       NaN        NaN    NaN  NaN  0.016667      0.916667   \n",
       "2008-04-04      NaN       NaN        NaN    NaN  NaN -0.333333      0.666667   \n",
       "2008-04-05      NaN       NaN        NaN    NaN  NaN  0.500000      1.000000   \n",
       "2008-04-06      NaN       NaN        NaN    NaN  NaN       NaN           NaN   \n",
       "\n",
       "            date  \n",
       "2008-04-02     2  \n",
       "2008-04-03    12  \n",
       "2008-04-04     9  \n",
       "2008-04-05     6  \n",
       "2008-04-06     3  "
      ]
     },
     "execution_count": 11,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "twitter_mean.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Now we need to set the index for the *Moves* data correctly so that we can join our two dataframes on it. To do that we need to correctly parse the `date` column."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [],
   "source": [
    "moves_dataframe['date'] = moves_dataframe['date'].apply(lambda x: datetime.strptime(x,\"%Y%m%d\"))\n",
    "moves_dataframe.index = moves_dataframe['date']\n",
    "twitter_moves = twitter_mean.join(moves_dataframe,lsuffix='_twitter',rsuffix='_moves',how='right')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Polarity/Subjectivity in relation to felt temperature\n",
    "And with that we are ready to plot how our `polarity` and `subjectivity` is dependent e.g. on the `apparentTemperatureHigh`:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<seaborn.axisgrid.FacetGrid at 0x7fd21d57ad68>"
      ]
     },
     "execution_count": 13,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 864x576 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "sns.lmplot(x=\"apparentTemperatureHigh\", y=\"polarity\", data=twitter_moves,lowess=True,size=8,aspect=1.5)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<seaborn.axisgrid.FacetGrid at 0x7fd21cce24a8>"
      ]
     },
     "execution_count": 14,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 864x576 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "sns.lmplot(x=\"apparentTemperatureHigh\", y=\"subjectivity\", data=twitter_moves,lowess=True,size=8,aspect=1.5)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "There seems to be very little interaction between the perceived temperature and the polarity/subjectivity. So, maybe it's more a function of how wet it's outside?\n",
    "\n",
    "# Tears in the rain: Polarity/Subjectivity in relation to percipitation\n",
    "Let's just plot `polarity` and `subjectivity` in relation to the presence/absence of any percipitation outside:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.axes._subplots.AxesSubplot at 0x7fd21cc62438>"
      ]
     },
     "execution_count": 15,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAZEAAAEOCAYAAABIESrBAAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADl0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uIDIuMi4yLCBodHRwOi8vbWF0cGxvdGxpYi5vcmcvhp/UCwAAIABJREFUeJzt3Xd4HNW9//H3zFatuq1muRc4tjGWGxiMCS2mhJYACSmkkNwEEm5yU256ckn7cRMu6TdAQsolhN472MFAwBiwLRcw5uDei5rVVto28/tjVtJKlmxprdWuVt/X8/jRaHZW+xWs9jPnzJlzDNu2EUIIIZJhprsAIYQQw5eEiBBCiKRJiAghhEiahIgQQoikSYgIIYRImjvdBQy1mppmGY4mhBADVFqab/S2X1oiQgghkiYhIoQQImkSIkIIIZImISKEECJpEiJCCCGSJiEihBAiaRIiQgghkiYhIoQQImkSIkKIrBONRmloaEh3GSOChIgQIuv86lc/5xvfuIHVq99IdylZT0JECJF13n33HQB27dqZ5kqyn4SIECKrWJbVud3c3JTGSkYGCRGRtBdfXMa6dWvSXYYQ3Rw+3HUtRK6LpN6Im8VXDI6GhgbuuutvAPz1r/ekuRohutTX1yVs16axkpFBWiIiKa2tLekuQYhe1dQc6tw+VHMI25bVH1JJQkQkJRhs7dyORMJprESI7g4ePACA4TYIh0I0Nh5Oc0XZTUJEJCWxJdLSIq0SkTkOHNgHgGdMbvz7/eksJ+tJiIikNDU19botRLrt378Pw2XgKcuJf783zRVlNwkRkZSmpsZet4VIJ8uyOHBgP2a+B3e+B3BCRaSOhIhISmI/s/Q5i0xRX19HJBLBlefBzPcCEiKpJiEiktLY2JiwLSEiMkPHRXVXngfTY2L6XBw6dDDNVWW3jLpPRCl1MnAvkKe1nnSU464CfgBMBbYBP9JaPzokRQqg+53AclewyBQdw3vNXE/8q5u6ulpisRgulyudpWWtjGmJKKU+AjwHbD7GcbOBfwA/AkqBHwL3KKVmpbpG0UVGZ4lM1HG3uivHOT82c9xYliUnOimUMSEC5AOnAy8c47gvAEu11o9prdu11k/En/NvqS5QdGlrawPD+UMNhdrTXI0Qjo6uVcPvtDrM+NfE7lcxuDImRLTWf9Fa7+rHofOB6h77qoFTBr8q0ZdIJILh8nRuC5EJ2traADA9zkebEf/a3t6WtpqyXUZdE+mn0UDPWdXqgZL+PLm4OIDbLX2jx8swAAwA3G6T0tL8tNYjhCPmfHHHQ8TlfPX5DHmPpshwDJHeGEC/JshpaAimuJSRwTRdYIcAsCyDmprmNFckBIRCPVrFznkOjY1t8h49Tn2FcMZ0Zw1ADUe2OkbF94sh4vV6sWOhzm0hMoFpxj/SOiZdjH8xDCM9BY0AwzFEVgMLeuw7FXg9DbWMWIFAbsJ2II2VCNHF6/UBYEft+FdngSqfz5e2mrLdsOjOUkq9C1yvtX4J+CNQrZT6EPAMcBlwJnBD+iocefLy8nvdFiKdcnKc+bI6wqPjq9/vT1tN2S5jQkQppYGJgAtwK6U6xo2q+L88AK31O0qpq3HuE7kTeA+4Qmu9ZciLHsEKCgoStgvTWIkQXTpayHbECQ8r/jU3Ny9tNWW7jAkRrbU6ysPdOjS11o8Dj6e2InE0+fkFCdvSEhGZoaNVbIdi3b5KiKTOcLwmIjKAdGeJTNRxQmOFY/GvFm63W7qzUkhCRCQlNze3120h0qkjROxQ/JpIKEZefr6MzkohCRGRlMTRLjLyRWSKjm4rKxLvzopY5OVKSzmVJEREUjrH4xO/8VCIDNB5YT1sYds2dsSSIegpJiEikmJZVue2bfdrsgAhUq7j2ocdtSB+r4jfn5POkrKehIhISnt7e8K2TG4nMkNH16ods7FjzomOzKiQWhIiIimJa4g0N8ucRCIzdHatWjZ2vLEsE66mloSISEpt7aGEbZm2TGSG3kZhGYZ8zKWS/NcVSTlwYH/n9v79+9JYiRBdLCs+Fbxh0JEnsVgsfQWNABIiYsAsy2Lnzh0YHmckzK5dO9JbkBBxkUgUAMM0wDTi+2TRtFSSEBEDtnfvHsLhEK5AGaavkO3btxGNRtNdlhC0tTnrBRkeU1Y1HCISImLAtN4EgDtQiitQSiQSZseObWmuSghobXUGfBheE8M0MNxG5z6RGhIiYsDeffcdAFyBMlyBsm77hEinw4edlbNNvzMiy/C5aGjouZq2GEwSImJALMti06aNGJ5cTG8erkA5AO+883aaKxMC6urqADBznAnKzYCb5uYmwuFwOsvKahIiYkB2795JW1sQdzw8TLcP01fE1q2biUTkD1Wk1/79ewFw5Xm6fT1wQEYQpoqEiBiQjushrtzSzn2uQBmRSITt2+W6iEivPXt2gwGu/HiIFHi79ouUkBARA7Jtm7OApCunpHOfK1ASf2xrWmoSApyhvDt2bMNV4MVwOR9t7mJnGpStWzens7SsJiEiBmTnzp0YLi+Gp2ulOJd/VPyx7ekqSwi2b99KNBrFPbprASp3kQ/DZXS2oMXgkxAR/RaNRqmpOYjpLeg2vYThyQXDJf3OIq3WrasGwFveNfW7YRq4S3PYt28vhw4dTFdpWU1CRPRbfX0dlmVheLuvV20YBqY3j0OHDvXxTCFSy7Is1qx5E8Nt4inrvhSut9IJldWr30hHaVlPQkT0W+cYfPeR6zMYbj9tbUEZSinSQutN1NQcwlsZ6Lwe0sFbmYvhMnjllZdk7ZsUkBAR/dZ5N7DryOVwO/bJ3cEiHZYvXwaAb1LBEY+ZXheeylwOHjwg9zOlgISI6Le2tvgcRKbniMeM+L7OY4QYInv37qG6ehWuIh/u0Uee4ADkTCsE4KmnHhvK0kYECRHRb6FQCADDdB/5YHxfxzFCDJUnnngE27YJTC/qHPDR+lYdrW/VdR7jLvbhKc9B601s2rQxXaVmJQkR0W/hcDwgzCNXijMMV/djhBgC7733LqtWve6ExJiuUVnhva2E97Z2OzYw0xmKfu+9d2FZ1pDWmc0kRES/dayrbpge2g+uo/3guq4HXZ5uxwiRatFolLvv/j8AAlWje13VMJG72IdvQh579uzihReWpry+kUJCRPRbS4uzlrrh8hJt3kW0eVfnY4bp7XaMEKn2zDNPsHv3LnwT8/GM8h/7CUBg1ihMr4uHH75P7hsZJBIiot9qapz7QMz4ioaJTG9ut2OESKXt27fy5JOPYua4Ccwe3e/nmX43garRhMNh/vzn22QxtUEgISL6JRaLsW3bVgx3DobLe8TjptcZ/bJly3tDXZoYYYLBVm699bfEYjFy55dgegb2MeYdl4t3bC5btrzHY489lKIqRw4JEdEv7733Li0tzbjzKnt93PTkYPqLeffdTTQ3Nw1xdWKksCyLP/3pVurqasmZXoS3LHDsJ/VgGAa580px5Xp45pknWLPmzRRUOnJIiIhjsiyLhx++DwB34aQ+j/MUTsayYjz22MNDVJkYaR544B42bFiLpyyHnBnFSf8c02OSt7AMw21yxx23snPnjsErcoSREBHH9OyzT7Ft21bc+eNxB0r7PM5TPBXTm89LL/2Tt95aP4QVipHgn/98nqVLn8GV7yFvYfkxR2Mdi7vIR96CUsLhML/5zc1yPS9JEiLiqJYte5aHH74Pw52Dr3zuUY81DBe+MacCJr///S/ZuPGtoSlSZL3XXnuFe+65E9PvIv/0igFfB+mLtzKXwOzRNDYe5pZbbqKx8fCg/NyRREJE9KqtrY2//vVP3HvvXRjuHAITz8X0HLv/2R0oxT9uMdGYxW9+czNPP/243Ngljsvrr6/gr3/9I4bHJP+MMZ1L3g6WnGmF5KgiamoOcfPNP+ucaFT0j4SIOMKmTRu58cbv8OqrL2H6ighMOAfTm9/v57vzxpAz/iwsw8vDD9/Pz3/+k861r4UYiFdeeYk77rgV2wX5Z1TgLjxyZOBgyJlZjH9aIfv37+PnP/8JdXW1KXmdbGSMtKmRa2qaR9YvPABbt27h0UcfiM90auAdPQNv6UmdU5okatnyBAB50y7r8+fZ0RDtB1YTbd6NYRgsWnQml112BaWlZan6FUSWsG2bp556nEcffQDT63ICpLj3yRV70/CccyNs8YUTBvSabe800KYPU1RUzFe/+i0mTJg44NqzVWlpfq8XoSRERjjbttF6E88//zTr168FwJVbga90Nq6cUX0+rz8h0vHzYy37CNVswAo1YpomZ555Nu9//4WMHTtu8H4RkTWi0Sh33vlnVqz4F2bATf6iCtwFA2uBJBMiHdreO0zw7Xp8Ph/XX/8VqqqOfi1wpBgWIaKUGg/8AVgEtAGPA1/TWkd6HHcR8DTQcwWkc7XWrx3tNSREHG1tbbz++qu88MIy9u3bA4ArpwRv2WzcgWO3FPobIh1s2ybatItw7dtYYWdqFKVmcN55FzB37nxcriNbO2Lkqaur5bbbfse2bVtwFfsoOL0c09/LrNHHcDwhAhDa20Lr6hqw4PLLr+SSSz6IaY7s3v++QmTg/3dS6xFgIzANKAQeBX4KfKfHcaOAt7XWs4e2vOHNtm02b9asXPkqb7zxmjNZomHgLpiAp/gEXDklxz1ssi+GYeApnIi7YDzR5n1EGjaj9Sa03kRRUTFnnHEmp59+JpWVY1Py+iLzbdiwlj/dcSvB1la84/PIm1uC4U7PB7dvbB6uHDfNbx7iscce4r333uULX7iBgoLCtNSTyTKmJaKUWgC8AZRrrWvj+64C/gSUaK2thGP/HfiI1vp9A32dkdgSOXhwP6+99iorV75KbW0NAIY7B0/RVOfejl6Wuz2WgbZEehMLNRJp2EK0cQe25TQ2J06czKJFi1m4cJH8wY4QbW1BHnjgHl5+eTmGaRCoGo1vUv5xndAcb0ukgxWO0bK6hsiBIHn5+Xzqk59lwYKFx/Uzh6vh0BKZD+zuCJC4aqAYmApsTtg/CihTSr0EVAG7gZu11v841osUFwdwu7O/6+TAgQOsWLGCV199lS1btgDOYlLuwkl4CifhCpRhGOltnrt8hbgq5mOXVRFt2UekcQc7d+5g587t3H//3cyePZvFixdz2mmnUVgogZKNqqur+f3vf09tbS2uAi95C0pxF/X/AnqqmV4X+aeX076lkdZ3Grj11t9yxhlncN1111FcnPwd89kkk0JkNNBzgHZ9/GsJ3UPkMLAd+B7wDnA5cLdS6qDWetnRXqShITg41WagQ4cOsmbNm6xa9QY7dmyL7zVw5VbgKZyEO39c76sSpplhuvEUTMBTMAEr2k60aSeRxp2sW7eOdevWceuttzJjxkksWLCQuXMXUFBw5DraYng5dOggDz54D2vWrALDIGd6ETnTizHM1HSnHg/DMMg5oQhvRYCW6lpWrFhBdXU1l112BeeddwFud+b9TaVCaWnvw/wzqTvre8CHtdZzE/YV4wTJ6Vrr14/x/PsBS2v9saMdl03dWbZts3fvbtasWUV19Sp27+5Y38PAlVuOu2A8nrxxGO7BP7MbjO6sY7HCLUSb9xBp2oXV7pxPGIbBiSdOZ968U5g3bwGjR5ek7PXF4GtrC/L000/w/NJniEWjuEf7yK0qGfTWx2B1Z/Vk2zahbU0ENzVghy3Kyyu4+uprqKqam7LriZliOHRn1eC0OBKNSnjsWHYApw1mQZnIsiy2bdtCdfVq1qx5s2u+H8PElTsGd8G4lAXHUDO9eXhHT8c7enpXoDTv7rwgf++9f2fixMnMn38K8+adwpgxlVn/hzxctbe3s3z5Up599klaW1sxc9zkzS3DOy53WP0/MwwD/9RCvOPyCG5q4OD2A/zud7cwbdoJfPCDH2bGjJOG1e8zGDIpRFYDY5VSY7TW++P7TgUOAdsSD1RK/QewS2v9aMLuGT2PyxaRSIRNmzaydu1q1q5dQ1NTIxC/xpE/Hnf+ONx5lRiuwZ0OIpN0C5RIG9GWPUSb97Bzl3MN5ZFHHqC8fAzz5i1g7tz5TJkybcQPycwEoVA7L774T5559klampsxPCY5M4vJmVaYtpFXg8H0ucibU4J/cgHBd+rZsmUzt9xyEyeeOJ0PfvAqlJoxYsIkY7qzAJRSrwFbgC/jXCN5ErhPa/1TpdS7wPVa65eUUt8GvgJcgjMk+Crg7zjdXquO9hrDpTurrS3Ihg3rqK5ezYYN6wiF4uubu3y48ypx54/DlVuetmscQ9Gd1R92LEy0ZS/R5r3EWg9gW85KdQUFhcydO595805hxoyTRky/daZobGzkhReeZ/nypQSDQQyPiX9aIf6pBZje1A9sSVV3Vl+iDSGCmxqIHHCuuU6ePJWLLrqEefNOyZqTmeHQnQVOGNwKbAVagfuBm+KPKSAvvv0/QADnPpLRwHvA5ccKkEzX2NjIunVrqK5exTvvbCQWcz4QDU8unlEKd/7Y+L0c2fGmHAyGy4uncDKewsnYVpRY60EizXtobtnHyy8v5+WXl+P351BVNZd58xZw8slz8Pv7tx63GLh9+/aydOkzvPbaK0SjUUyvi5zpRfinFQ5JeKSLu9hHwaIKIvXttOnDbN++lVtv/S0lpWVccP5FLF58Fj5fdr7vMqolMhQyrSVSX1/HmjWrWLPmTTZv1nT8/zB9RU43Vf44TF9hxjWNM6Ul0hfbtoi11RJt3kO0eS92pBUAt9vNrFmzWbBgIXPmzCMQOHK9eDEwsViM9eurWb58WXzeNXDlevCfUIhvQl5auq2GuiXSU6w5TNuWRsK7WrBjNjk5OZxxxlmce+4SKirGpKWm4zUspj0ZCpkQInV1tZ1Dcbdu7Rq57MopjQfHWExv3lF+Qvpleogksm0bK3Q4Hih7sELONSWXy8VJJ50cD5T55OVl9n/zTNPU1MQrr7zEiy8uo76+DgB3id+58FwZSOuJT7pDpIPVHqN9exOh7U1Y7TEAZs2azbnnLmH27LnDqqtLQiQuXSHS0tLMqlVvsHLlq2zZ8l58r4ErUIq7wLk4nsyd4+kynEKkJyvURKR5N9Gm3VghZxEi03Qxe3YVp522mDlz5uH1pmbK8eGuY+qcl176J6tWv0ksGsVwm3jH5+KfUpiyqdoHKlNCpINt2YT3tdK+tYlonXN9c9So0Zx11nm8731nU1hYlOYKj01CJG4oQyQSibB+fTUrV77Khg3riMWcMxFXoDwhOIZnP2nLliewbZv8Ey5PdynHxQo3E2naQ7RpZ2eg+P05LFiwkNNPPwOlZgyrs8VUCQaDrFz5Ci+++ELXhJ35HnyTC/BNzB+0lQYHS6aFSKLo4RDt25oI72nFjlqYpsm8eadwzjnvZ/r0mRnXdd1BQiRuKEKkrq6Wl176Jy//60Vamp0Za01fkXPXeMGEfq0QmMli7YcJbn8esDG8+eSMPQOXP/PPpI4lFmok2riDSNNO7IgzyqasrIJzz13C4sXvG5HXT/bs2cXy5ct4beWrhEMhMA28lQH8kwtwl/gz9gOv4bld2LbNqIsydz0QK2IR3t1C+7YmYk3OhOQVFZWce+4SFi06k0Agsz4nJETiUhkiWm9i6dJnWbduDbZtY7i8uAunOHNVZcGHbIeWrU9jx6dzBzC9+eROvTiNFQ0u27aJtdUQObyNaNMusC28Xi+nn76Y88+/iDFjsnum4Wg0SnX1Kl54YSmbN2sAzIAb/+R8p9WRxNTsQynaGKZx+R6wwczzkL+wPGO62Xpj2zbR+njrZG8rWDZen49Fpy/m3HPPZ9y48ekuEZAQ6ZSKENm1aycPPXQfb7+9HgDTX4y3+ATcBRMycq6q42FF22jd/PgR+3NPuHxYXdPpLysaInJ4G5HDW7AjrRiGwZlnns3ll19JcXHfi3YNR+3t7fzrXy+ydOkznRfKPWU5+KcU4BmT3gvlA9GwdDdWS9cSRGaeh+LzM+OD+Fis9hihnU20b2/GCjpD/E8+uYoLL7wk7V1dEiJxgxkira0t3HPP33n99RXYto0rUIavdDZmzuhh8wc3UFa4hdatTx2xP3fqJRk/oux42LZFtHkv4Zq3sMJNeDxeLrzwYi699EPD/kbGpqZGli17juUvLqMtGMRwGfgm5uOfWoArP3PP4HtjtUdpeGbXEfuLPzAh41tQiWzbJrI/SNuWRqK1zoX4SZOmcNFFlzB//qlpuU4nIRI3WCGyZ88ufve7X1FbewjTV4SvrApXbkXWhkeHkRoiHWzbItK4nXDN29jRNpSawRe/+B/DcmbhcDjM0qXP8PTTjxMKhTC9LnxTC/BPKcD0Dc8bA2OtEQ4/v/uI/UUXjMeVOzynBYrUt9P+XiPhfc69ThMmTOTqq69hxoyThrSO4w4RpdQ3gLu11gcGs7ChNhghovUmfv3rmwmHQ3hHz8RbOmvE3EU+0kOkgx2L0L7/DaLNexg1ajTf/e6Nw2ZGYdu2ef31FTz08H001Ndj+lz4pxfhn5SP4Rre7+NsDJEOsZYIwU0NhHe3ADBnzjw+8pFPDNnNi32FyEDeMZ8CdiullimlPqWUGnlDVXDWJr/jjlsJR8L4x56Br2z2iAkQ0cVwefCPPQNvyUnU19dx551/Zji06tvb27nttt9xxx23cvhwA/4Tiyg8fzw5UwuHfYBkO1eeh/xTyig8ZyzuEj/r1lVz443f4fXXX0trXf1+12itq3Bmyl0G3AAcUErdo5T6gFJqxLz7nnnmcerr6/COmoGnYHhcrBOpYRgG3pJZuHLH8PbbG6iuzuyp2w4ePMDPfvZDVq9+A/doP4VLxpE7a1TG3eMhjs5d7KPgzDHknVpG1I7xpz/9L/fdd1fnfWhDbUDvHq31Fq31zVrrhcBMYD3wD2C/UuoXSqnsHvsIaP0uYOAtmZnuUkQGMAwDX4nTN+28NzJTKNTOLbfcxL59e/FPLaDgzDHDvntnJDMMA9+4PArPqcSV72Hp0md54olH0lLLgE9BlFKGUur9wI+BbwGNwB9xZtPdoJS6aHBLzCw1NYcwPDlZN3R3oLxeL5WVlTI9CM59MgA1NQfTXEnfHn/8EerqavGfUEhuVUlGLkMrBs6V76Xg7LGYOW6eeeYJ9u3bO+Q19DtElFJVSqlbgN3AI4ABXKm1nqy1/i+t9b/hXDf5Q2pKzQxFRcXY0XZsOz1Nx0zg9Xq5/vrr+eMf/8j1118/4oPEis8QnKn3jUQiEZYtexYz4CYwozjd5QyJkXSSY3pMArNHE4vFWLr0maF//QEcWw1UAd8FKrTW12qtX0o8QGv9NBAavPIyz+TJU8C2iAVr011K2pSUlLBkyRIAlixZQknJ8BiVlCqxoLNE8aRJU9JcSe8MwyAWi2Hmuof1aoL9NRJPcjruyLesoR/cMZB31I+01ku01ndprYMdO5VSAaXUtzq+11rPGNQKM8xpp50BQKRhS5orSZ/a2lqWLVsGwLJly6itHbmBats2kcNbcHs8zJu3IN3l9MrtdpOTk4PVFMEKZ38LeiSe5HTMDJyfnz/kr33Mjn2llBvwAd+Nd2f17EydjnN95ObBLy/znHCCYty4CezZs5tY++GsmhOrv8LhMLfffjsPPfQQtbW1hMNhRuol2mjTDqxwC4sWn0Ve3tD/AffXkiUX8cQTj9C6rpa8U8qy+qbYjpOcJUuWdJ7kBChPd1kpEwtGaN1Qh9fr5X3vO3vIX78/LZEbgCacIGkBmnv8WwWsTlWBmcYwDK666mrAJnSweljcG5AK4XCYffv2EQ6H011K2tixCKFDG3B7PFx22RXpLueoLr30Q0yZMo3wnlZa19Rgx6x0l5QyHSc51113HbfffntWv0ejh0M0vbIfO2Lx8Y9/mvLyoV818ZghorX+LVAORIFze/l3evzriHHyyXOYPXsuseAhIvWZO6xTpI5t27TvfxM72sYHLrqUkpLSdJd0VC6Xixtu+CqTJk0mtKuFxpf3EWuNHPuJw1S2n+TYtk37zmaaXt6H1Rrlssuu4Mwzz05LLQOZ9qRSa70vxfWk3GDNndXU1MiNN36XxsZGcsadiTu/cjB+bMaTaU8codqNhGveYtq0E/nWt34wbCZhjETC3H33nfzrXy9iuAxyphfhn1aE4cqO7q1snvakQ6wlQuv6WiIH28jJyeHzn/8Sc+bMT/nr9jXtyVHf+Uqpf2itr4l/+xulVJ/Haq0/knx5w09BQSFf/OJX+OUvf07b3lfxVy7CUzAu3WWJFLNtm3DtRsK1b1NcPIrrr//ysAkQAI/Hy2c+83lOPHE6993/D1o2NhDa2UJg9mi8FZm1CJLozo5atOnDtG1uBMtmxoyT+NSnPkd5eUVa6zrWuz+YsN2aykKGoxNPnM7Xv/5tfvObm2nfuwIrNBNvyUkyl1aWsmNh2g+sJtq0i5KSUr71rR8watTodJeVlEWLzqSqai6PPfYQy5cvo/m1A7hL/QROGoVn1PBcsjlb2TGb9h1NtL97GCsUo7h4FB/96DUsWLAwIwZIDKQ76yyt9cspriflUrEo1bZtW7j11t9SX1+H6R+Fv3IhLl/hYL9MRhip3VnRlv2d10CmTJnGl770H8M2QHpyFlW7l7ff3gCAZ0yAwEmjcBcMv/srsqk7y7ZtwrtaCG5qwApG8fp8XHD+B7jookvx+4c+6AdjKvgGnJsMh/XNhKlaHjcYDHLvvX9nxYp/AQbuwkn4Sk7Kug/WkRYisbY6QjUbiLUexDRNLr/8Sj7wgctwuYbnehtHo/UmHn74PrZs2QyAd3wegelFw2phqmwIEdu2Ce9tpW1TA7HmCC63m3PPWcLFF19GQUH6Tk4HI0RuAE4CbgV24YzW6pR4A2ImS+Ua6wDr16/loYfuZe/ePWCYeAqn4B11IqZv+C1a1JuRsDyubdtYbXWE6zYRbXHmIjrppJO56qqPMXHipPQWl2K2bbN+/Voee+xBdu3aCQb4xueRM6N4WHwID+cQsW2b8L6gEx5NYUzTZPHis7j00g9lxFo1gxEibYAr/u8IWuthcWqW6hABsCyLN99cyWOPPcShQ86kfK7ccjxF03Dnjx3210xatj6NHW7u/N705pM79eI0VjQ4bCtCpHEnkYYtWKHDAEybdiJXXnk1SmX1RAxHsCyLtWs2a/f4AAAgAElEQVRX89hjD8VPiMA3uYCAKsLMydyBBMMxRGzbJnKojeDGBmKHQxiGwaJFZ3LppR+irCxzbpIcjBA562iPD5frJUMRIh2i0Shr1rzJiy/+k/fec+4nMdw5eAon4S6YOGzvdo+1Hya4/XnAxvTm4x97xrD9XWzbJhasIdq0k2jTLmwrgmmazJu3gHPOWcL06TMz4uJlujgnRK/z2GMPcujQQWf99amF5JxYiOnNvPPG4RYikfp2ghvridY405accsppfPCDVzFmTObdMpDSNdaVUndprT953D9oCAxliCTau3cPL764jNdee4X2ducNY/oKcRdMwFMwoXM68eGiZcsT2LZN/gmXp7uUAbNtG6u9nkjTLic4om2AM0Pz2Wefx5lnnkNx8ciY7ba/otEor776Mo8/8TCNhw9jel3kzCjCN7kgo6aVHy4hEgtGCb5dR3iPM+j15JOruOKKqzO6u3QwWiIG8FlgAc4UKB0qgYVa62HxV5euEOkQCoXYsGEtb7yxkg0b1hKNOpeWTH8x7ryxuPPHYfoKM/7st2XLEwDkTbsszZX0j21bxNpqiTbvJda8FyvirFOdkxNgwYJTWbhwEdOnz8Q0h3dXY6qFw2H++c/nePKpxwi1t+PK9xA4OXPuMcn0EOm416N9SyN2zGbSpMlcffU1w6K7NKmbDXv4NfBxYCXwAeBJnKnhG4CrjrfAkcLn83HKKadxyimnEQwGqa5exZtvruSddzYSrm0gXPs2picPV/5Y3PljceWUDPtrKOliW1FirQeJNO8h1rIPO+YMLPT5/FSdejoLFy5i1qzZeDzp/3AZLrxeLx/4wGUsXnwWjz32EC+/vJzm1w7grQwQmF2CK5De6yV93Xmf7jvybdsmsj9I6/o6rLYohUVFXHXlRzn99MXD/sRlIC2RvcBirfV2pVSb1jpHKeUCfg+8obW+M5WFDpZ0t0T6Egy2smHDetatW82GDes6u7wMlxdX7hjceZW48yowXL5j/KShkaktESvSSrRlH9GW/cRaD0J88bCCgkLmzVvAnDnzmTHjJAmOQbJnzy7uuutvbN6sMdxmfBqVwrR2cTUs3Y3V0jUvmJnnofj88WmrJ9YaoXV9HZEDQVwuFxdeeAmXXHI5Pt/wuqlzMLqzGrXWhfHtFqBIax1VSpUAq7TWkwet2hTK1BBJFIlE0PodqqtXs379Ohoa6uKPGLhySnDlVeLOr8T0FqSt2ytTQsS2Lay2eqIte4m27O8cVQUwduw4qqrmMXfuAiZPnjLsz/gylW3brFjxL+6//25aW1twFXnJm1+KuzA9JzzRxjCNy/eA7QRI/sLyzkWbhpJt27Rva6JtYwN21GL69Jl88pPXMmbM2CGvZTAMRnfWu0qp64A7gB3AlcD9QC7O+upikHg8HmbNqmLWrCps22bPnt2sX1/N+vVr2bZtC7G2GsI16zE8ufEWSiWuQBmGmXmjZVLBjoWJth4g2rKPWMv+zm4qt9vNSSdXUVU1j9mz52T8zLrZwjAMFi8+izlz5nHfff/gtddeoXH5PnJUITnTi4e8VeIu9GLmuLFtO20tkFhLhJY1NUTr2gkEAnzs059i0aIzM/5aZzIG0hJ5P87a6mOBq4HbgC3AGOBJGZ01NJqamnjrrXVs2LCOt95aT3u7M7LIMN24civioTIWw53as8ChbolYkVaizXucC+NtNRB/3xYWFlFVNZeqqrnMnDlr2HURZKMNG9Zx551/pqGhHlehl7wFZUPeEmh4bhcAxRdOGNLXtW2b0LYmgm/XY8ds5s8/hWuuuZbCwuE5BD7RoAzxVUr5tdbt8e3zgFOAbcDDWuthse7mcA+RRNFolM2bNevXV7NuXXXnjY1g4AqU4s4fjzt/LKZn8EfOpDpEbNvGCjc5wdG0ByvU0PnY5MlTqKqaR1XVXCZMmJSVZ3fDXVtbkHvv/QevvvoSmAaBmcX4Txi6UYfpCJFYMEprdQ2RQ20EcnP51Cc/x6mnnjZkr59qKb1PZDjJphDpaf/+faxdu5rq6lVs27a1c7/pH4WncCLugomY7sE5U09ViFihJiKNO4g278aK3xVvmi5mzjyJefNOYc6ceRQVDYvR5AJYt24N//d/d9DU1IS7xE/e/NIhGWo71CES2t1C67pa7IjF7Nlz+MxnPp9179OkQkQpVQP060NXa12WXGndXm888AdgEdAGPA58TWt9xBJsSqmrgB8AU3FaQz/SWj96rNfI5hBJVF9fFw+U1Wi9CcuyAANXboVzx3z+WAwz+eGYgxkiVrSdaNMuIo07sNrrAWco6cknVzFv3ilUVc0lEMg97tcR6dHc3MTf//4X1qxZheE2ya0ajXdCXkpbJUMVIlY4Ruu6WsJ7WvH6fHz06ms466xzs7J1nOyF9f9MQS1H8wiwEZgGFAKPAj8FvpN4kFJqNvAP4KPAc8D5wP1KqVO01m8PacUZatSo0Zx33gWcd94FNDU18sYbK1m58lV27NhGrHU/hunGXTjZmRwyTXfLx4K1hOs10ZY9YNsYhsGsWbM5/fTFzJu3QK5vZIn8/AK+9KWv8tprr/CPu/+PljU1ePcHyZ1bgukbvoNBIofaaFlTg9UWZcqUaXz+819K+wJR6ZAx3VlKqQXAG0C51ro2vu8q4E9AidbaSjj2f4EJWuvLEvY9BWzRWn/1aK8zUloifdm/fy8rV65gxYp/0dDgnPW788biGXWiM8Krn2dQybZEbNsi2ryHcJ3GaneGLo8dN54zF5/FwoWLsuICpOhbTc0h/vzn29i8WWP6XeTOL8VbPvjX7FLZErFjFsGNDbRvacQ0TS677AouvvjyrFweINFxD/FVSrlxuo+uACYAXpzRWX/TWv96EGqcD+zuCJC4aqAYp8tqc49jn+/x/GrgvEGoI6uNGTOWK674CJdddgXV1atYuvQZtm3bSrRlL65AGb7yubj8qenLjbbsI3RwLVa4GcMwmDNnPueffxFKzcjK5r84UmlpGd/+9g959tmnePTRB2hecQD/1AICs0ZhuDL/Pp5oY5iWVYeINYUpL6/gC1+4gcmTp6a7rLQaSKf4LcCHgNuBd+P7ZgBfV0q5tNa3HGcto3GmUElUH/9aQvcQ6evYY066X1wcwO3O7jOG/rr44vO5+OLzeffdd7n//vtZvXo1we3P4ymagrd09qBdhI+FmggdXOt0oxkGF1xwAVdccQWVlZk3U6kYGp/5zCdYvHght/zyl+zduofIoTbyTinDXZQZMzL0ZNs27VsanRsHLZsLL7yQz33uc2lZYTDTDCRELgYu1FpvStj3aLwb6UGckBlsHaen/emCMvpzXEPDsFg7a0iNHj2WL33p67z99gbuu+8f7Nu3jWjLfnLGLcaVc3z3kUYadxI68Ca2FWPmzFl89KOfZNw45wawmprmYzxbZLPCwnJ+8P2f8uCD97B8+TKaXtpHzknFzrQpGdQytdqitKxxhu7m5efz2WuvY86ceTQ3R2huPmLMT9YqLe392ulAQqQUp/uqp3eAwVg5pYYjWxKjEh7rz7E9jxMDMGvWbH784//mueee4pFHHiC48wX8FQvwFE0Z8M+ybZvQoXVE6jU+v5/PffbfmT//1Iz6cBDp5/P5uOaaa5k9ew5/+csfaX6rnsjBNvIWlGL607/4VXh/K63VtVihGLNnz+Haa6+jsDB9S9RmooF0Qm4EvtjL/usBPQi1rAbGKqXGJOw7FTiEM4S357ELeuw7FXh9EOoY0VwuFxdffDlf//p3CARyaN//JpGmPQP+OeGat4jUayoqKvmvH/6UBQsWSoCIPs2ePZef/OQXnHxyFZFDbTS+sJfwofT1GtiWTev6WppXHsSMGXz845/mP/7jmxIgvRjItCenA8uAvcAmnK6jmTjriVyhtV52vMUopV7Dae18Gee6x5PAfVrrnyql3gWu11q/pJSaiXMh/WPAM8BlwJ3AbK11b62lTiN9dNZA7Nq1g5tu+jGRaIycie/vtnrh0UZnRRp30r5vJSUlZfzwhz8hPz871pcXqWfbNsuWPceDD95DLBYjRxWRM2Pg828dz+isWGuE5jcOETscoqKiki996SuMGze006dkor5GZ/W7JaK1XglMxrkZsOOD+PfA5MEIkLirgAJgK/Ai8DRwU/wxBeTFa3kHZ/6u/8Lpwvo2TpAdNUDEwEyYMInrrrsB24oSOriG/pxw2LEIoUNr8Xp9fO1r35QAEQNiGAbnn38R3/vejykpLaNNH6Z5xQGs0NDMqhQ+EKRx+V5ih0Occcb7uPHGn0mAHMNAWiJjgL8AS4CO4U0RnJv9vqC1PtjXczOJtEQG7re//R/Wr19LzvizcOc5vY19tURCtRsJ17zF5ZdfyeWXXznktYrsEQwG+ctfbmPt2jWYAbczpXtx/0ZvDbQlYts2bfowbe804Ha7+eQnP8uZZ56dbOlZ6bhbIsADOOHxAWA6zvDey3CWyn3weAsUmasjDCKHe16a6s62bSKHt+Hz+bnggouHojSRxQKBADfc8DU+9KEPY7fFaPrXPsL7Wgf9deyYTcvqGtreaWDUqNF873s/kgAZgIEMf5gPVGitmxL2aaXUmzjXSUSWmjhxMhUVlRw4uA/bivY555bV3oAdaWXe6Ytl/LwYFKZpcumlH2L8+AncfvvvaX79IIHZo8mZNjgXuK1wjObXDxKtbWfq1BP48pe/QUGBdMEOxEBaIluJX5PowRd/TGQpwzCoqpoLdoxYW12fx8WCTo/m7Nlzhqo0MULMmTOfb3/7vygoKCC4oY7guz3vNR44KxSj6ZX9RGvbmT//VL75ze9LgCRhIC2RG4G7lVK3Ae/hdG1NBa4DfhkfMQV0XvgWWeTEExXPP/80sWAN7txy3PlH9jXHgrXxY6cPdXliBJg8eQrf//5PuPnmn1H3Ti3YEJiR3BQ9VihG06v7iTWGOfvs87jmmmtl+eQkDSREHop/PauXx87BGbHVcde4zCuSZTrmB7LanTNAf/mRrY1Yez1FRcUUF4864jEhBkPH3Fu/+MVPqdtUi+lz4Z8ysNaDHbNoXnmAWGOYc855P5/4xGckQI7DQP7LTT7GvykJX0WWKSoqpqCgkFh7790IVrQdO9rGxImThrYwMeKUlJTyzW9+n7y8PFrX1xI+2P+bEm3bdtY+rw9x2mlnSAtkEPS7JaK13pnKQkTmmzhxEm+9tR47GjpiDfeOFsqECZPSUJkYacrKyvnyl7/Bzf/z/2hdVYP7/WP7NU1KaHsz4T2tTJt2Atde+3mZRWEQSASLfpswYSIAsdCRrZFY+2EAxo+fOKQ1iZHrhBMUH/nwxztXF0y85807Nhfv2O6rYcZaIwTfricQCPDFL34Vj8c71CVnJQkR0W/jx08CwIoHRiIr1NESkRARQ+e8885HqRmE9wWJHGzr3J978mhyT+4+A3XrhjrsqMUnPvEZiouza/3zdJIQEf02frwzIisW6iVE2g/j9/spKSkd6rLECGaaJh//+KcBCL7T0OfUPNGGEJH9QaZNO4HTTjtjKEvMehIiot/KyyvweDxHtERsK4YVbmbs2PFykVIMufHjJ7BgwanEDoeI1rb3ekzbZuc9e/nlV8l1kEEmf/Gi30zTpLJyLFa4GdvuXPIeK9wM2J2LTQkx1M45ZwkAod0tRzxmRywi+4OUlVcwc+asoS4t60mIiAGprBwHdgw70jWHkRV2ZsIZM2ZsusoSI5xSMygsLCK8L3hEl1b4UBA7ZnPawkXSCkkBCRExIBUVziy+TuvDYYU6QkTWTBfpYZomM2achB2OYbV0X7I2Wud0cUkrJDUkRMSA9Boi8e3y8oq01CQEwNSpJwAQrQ912x+tD2GaJpMmyX3QqSAhIgakIyh6hojL5ZaRWSKtOk5wYq3dWyJWa5TRJSV4vXJfSCpIiIgBKSvrHiK2bWNHWigrK5ORWSKtOk5irGC0c58ds7BCMUpGywlOqshfvRgQv99PUVFxV4jEwtixcOdZoBDpkp+fD4AV7ho5aMe3Ox4Tg09CRAxYRcUY7EgQ24p2jswqL5cQEenl9+cAYEcThp9HnO2cnEBaahoJJETEgHWMwrLCzTIyS2QM0zRxud1gJQzxjW+73Z40VZX9JETEgFVWOveDWKFGrHBjt31CpJPLNLETQyS+6XLJR12qyH9ZMWCVleMA5/6QjpaIhIjIHHJD4VCSEBED1q0lEmqiuHiU9DmLjGDbdvcMiW9bltXr8eL4SYiIASsoKCQQyCXWXocdDUorRGSMaDSKYSakSHw7FoulqaLsJyEiBswwDMaMqcSOOtNJVFTIRXWRfpZlOS2RhE+1jkCJRKJ9PEscLwkRkZTEKU5kuhORCcLhMABG4kX0eIhEo+F0lDQiSIiIpCROcSLTnYhMEA47c2YZrq7uLMPtbIdCoV6fI46fhIhIyqhRo3vdFiJd2tud7tVuIRJvlUiIpI6EiEhKUVFRr9tCpEtHSwR3YncWYEAo1PuKh+L4SYiIpOTl5fe6LUS6dLZE3AktEcPAcJvSEkkhCRGRlNzc3M5tmb1XZIKOoDB63J1uuAwJkRSSv36RFJ/Pn+4ShOimc3SWu8cd626TUFhCJFUkRERSZIEfkWkikXiImN1DxHAZRMIyxDdVJEREUlwud7pLEKKbzrvSe4QIhiF3rKeQhIhIisvlSncJQnTTGRRGj5aIATGZOytlJEREUgxDZkoVmaVrgIfdbb8NmPJ+TRkJEZEUCRGRaUwz3jru2eiwbGk5p1DGdGwrpYqAW4FzcMLtn8AXtdZNvRw7A3gH6Dnk4lqt9b2prlXI1Noi8/j9PqD78rgd3/v9ub09RQyCjAkR4A6gAJiD0wL9O/An4KO9HDsKaNZaFwxdeSKRbUuIiMzSsaaNHekRIhGLnIKcdJQ0ImREd5ZSqgy4Aviu1vqg1voQ8APgKqVUSS9PKQYOD2WNortoVEa7iMxSUFAIgBXqem/alo0dtigslKl5UiVTWiJzcVof6xP2rcdZl2wusKzH8aMAj1LqaeA0oAan1fJrrbWNSLmOMflCZIqOOdystq61Q6x2J1AkRFInU0JkNE73VOcphNY6opRqBnpribQBGrgZeA04C3gIaMEJkz4VFwdwu+Ui2/GyrGDndmmpzJ0lMkE+eXl5BFu7LpVarREAJkwYK+/TFBmyEFFKXQI82cfDm/vYb9BzvB6gtX4QeDBh1z+VUn8EruUYIdLQEDzaw6Kf9u6t7dyuqWlOYyVCdCktLWP7zu3Ylo1hGsTiIZKbWyTv0+PUVwgPWYhorZ/CCYUjKKWWAM8ppTxa60h8nwfIw+mq6o8d9H4RXqRAMNjauR0KhfD5fGmsRgjHmDFj2b59G1YwiivPQ6zZCZHKyrFprix7ZcSFdWAtTotjXsK+BUAMqO55sFLqE0qpf+uxewawLWUVim5aWrrO6lpbW9JYiRBdxoxxwiLWFO72dcyYyrTVlO0yIkS01rXAA8DPlFJlSqkK4CbgLq11A4BS6gWl1CfiT4kCv1FKnaOUciulzgc+C/whHfWPRM3NXSHS2NiYxkqE6DJ27DgAop0hEqGoqJjc3Lx0lpXVMiJE4q4H9gMbcFofW4EvJzw+FWdoL1rr+4FvA3/EGer7K+ArWuuHhrLgkayx8XCv20Kk07hx4wGnBWJFLKy2aGewiNTIlNFZxO9M/9RRHp/U4/s/IC2PtEkMjqYmaYmIzDBq1Gi8Xi+x5gixZqc1UlkpIZJKmdQSEcNI95aIhIjIDKZpMmZMJVZLhFiTc1FdroekloSISEriNZHm5iOmNxMibcrKKrBjNtG6NgDKyyvSXFF2kxARSUkckSWjs0QmKS0tAyBS097te5EaEiIiKaFQCExP17YQGaK4uBgAK+hMf1JUVJzOcrKehIhISiQSxoiHSFjWrxYZJHGerNzcPNzujBk/lJUkRMRxk/WpRCYJBLrWDsnNlXVEUk1CRCTF5XZjW053gcslZ3oic+TkdK0d4vfLOiKpJiEikuL3+cFyurH8fn+aqxGii9vt6dz2eDxHOVIMBgkRkZTEbgLpMhCZJHE9dVlbPfUkRERS8vMLet0WIt1sW9alG0oSIiIpHUuR9twWIpNYlnXsg8RxkRARSUkcRilLj4pMEoslLI8rIZJyEiIiKYmtj8JCaYmIzBGLWQnb0aMcKQaDhIhISmJwSHeWyCSJwRGLxdJYycggISKSkrjIT15e72svC5EOiV1Yia0SkRoSIiIpiSEi66uLTJI4Osu2JURSTUJEJCXxrmBD5j0RGaR7iMhw31ST+SpEUkaPLuHUU0+jokIW/BGZJfGkxjTlPDnVJEREUkzT5Prrv5LuMoQ4QmKISCs59SSmhRBZJXG+rMR5tERqSIgIIbJKYoh4PNLZkmoSIkKIrOL1+nrdFqkhISKEyCo+n7/XbZEaEiJCiKzi9/sStiVEUk1CRAiRVRK7sCREUk9CRAiRVRKH9cryuKknISKEyFrSEkk9CREhRNaSlkjqSYgIIbKWTA6aehIiQois5fV6011C1pMQEUJkLZk7K/UkRIQQQiRNQkQIkXXmzTsFgJKS0jRXkv2MkbZoS01N88j6hYUYgdra2ti/fy+TJ0+VLq1BUlqa3+t/SAkRIYQQx9RXiEh3lhBCiKRJiAghhEhaRq3YopSaANwNLNZaH7UjUyl1FvALYCawD/iN1vr21FcphBCiQ8a0ROKh8Dqwux/HVgBPAncC5cBngV8opS5MaZFCCCG6yaSWSAlwITAB+Ngxjr0G2KG1vi3+/WtKqbuA64HnUleiEEKIRBkTIlrrh6GzS+tY5gPVPfZVAx861hOLiwO43a6BFyiEEOIIGRMiAzQa2NhjXz1Oa+aoGhqCKSlICCGyWWlpfq/7hyxElFKX4FzH6M2PtdY/Os6XMIBj3gPS11hnIYQQAzdkIaK1fgrng34w1HBkq2NUfL8QQoghkjGjswZoNbCgx75TcUZ3CSGEGCLDJkSUUi8opT4R//YfwBil1A1KKb9S6mzgE8Dv01agEEKMQBlzYV0ptRR4H/FgU0q1xx86X2v9L2AqUAygta5RSl0M/A/w38Au4Ivx44QQQgyRETcBoxBCiMEzbLqzhBBCZB4JESGEEEmTEBHHTSn1A6XUinTXIYQYenJNRAiRNZRS44E/AIuANuBx4Gta60haC8ti0hIRQmSTR3CmQJoGLMYJk5+mtaIsJy0RcQSllA18I/7vDq31j5RSHwW+D0wGGoDbtNY3xY//EXCJ1npB/J6dJ4CrgN8B44CXgY9prZuG+ncRI4dSagHwBlCuta6N77sK+BNQorW20llftpKWiOjLlTizJf9YKTUJ5wbP72it84ArgBuVUkv6eG4uzs2fpwEnAQuBz6S6YDHizQd2dwRIXDXO/WVT01NS9pMQEX15QGt9QGtta613AKVa66cBtNarAM2RU890MIFfaa0Pa613Am8CM4aiaDGijcZpJSeqj3895gzfIjkZc8e6yDg7e3x/vVLqc8BYnIk0vYDvKM/fnrAdBHIGtzwh+qVj0lfpt08RaYmIvkQ7NuLh8X3gS0C+1toPrD/G86X/WQy1vmb37nhMpICEiOiPU4EVWuulWuuoUqoAZ/SLEJlkNTBWKTUmYd+pwCFgW3pKyn7SnSX6YxtwvlJqNE631C3AbpyuLSEygtZ6rVLqdeAXSqkv41wj+QHwv1pr6c5KEQkR0R+3A2fhXCfZC3wdKAX+oJQ6BMiNXCJTXAXcCmwFWoH7gZvSWlGWk/tEhBBCJE2uiQghhEiahIgQQoikSYgIIYRImoSIEEKIpEmICCGESJqEiBBCiKRJiAgxhJRSWil1fbrrEGKwyH0iQmQgpdQPcO62hq4JL8N0TST4ea31XemoTYhEEiJCZDil1CzgLWByfFp+ITKGTHsiRrz4olvbcabM+DHOAkbVwEeAE4DngP8E/h/wQa31i0qpLwJfBibhzCP2Na31M/GfFwB+Ff95AE8C/661blVK7QBu0Vr/r1LqJWAlMBG4HDgMfF1rfX8/ag4AB4BPa60fTdj/RLyeVcCN8d/nZziz2d4PXKe1DseP7fN3EKK/5JqIEF2+AlwEjMGZd+kv8f0uYFZ8/0tKqQ/ifDB/BsjHWUb4UaVUx8JbNwFVwExAAdOBX/TxmtcB9+GsvvcT4G6l1DEnttRaB4EHgU927FNK5QFLcFahBKjAWWdcAXPij309fuyxfgch+kVCRIgut2utd2utD+PMVPx+IIDTYr9Nax2Mzwb7eeBvWus3tdYxrfVTwPPAp5RSBvBp4Jda60Na6zrg34DH+njNVVrrJ7TWYa31H4GDwCX9rPdO4ANKqeL49xcDe7XWK+Pf+4Efaq1btdbvAX/HafFwtN+hn68tBCDdWUIk0gnbO3FaIOUJ33eYijM1/r8n7DOBRpzpx4uAHZ0/VOuNwMZ+vCbALqCyn/W+AuwDPgz8CbiSrlYIQLPWen/C9zsTfvbRfgch+k1CRIguroTtnsuqRhMeawN+oLU+oosqvuYK9L+V7+rxvUE/l3LVWttKqb8DH1dK3YXTFfe9fv7sPn8HIQZCurOE6DI1YXsiTnAc6OW4LcDsxB1KqQlKKTPefXUY5zpEx2Mnx5cYPtZrAkwA9gyg5r8DZ+Bc29iotd6S8FhAKVWR8P3EhJ/d5+8wgNcWQloiQiS4Tin1MtCOcwH6+fh2T7cBzyqlHgCexlmC9SmcbqUXgL8B31RKvYhzgf53wLt0XahPtFApdQHwIk4QlMZ/Vr9orbcppVYC/03XfSUdQsAPlVLfBMbjXIT/TT9/ByH6Rc46hOhyF7AM2A/k4Vx8PoLWejnwVeDXQDPwV+CbWuuOD9/v4FyveBt4D2f48Lf6eM17gGuBeuBHwMe01r21fo7mTiAXZ5RXomacocrvAWuBpcBv+/k7CNEvcrOhGPES7hM5WWv99hC+7kvAaq31f77JwdAAAABsSURBVB7nz/khMFtr/eGEfZ/BuR+l5LiKFOIYpDtLiGFMKXUq8DXg/HTXIkYmCREhhiml1HM4NzV+S2u9Ot31iJFJurOEEEIkTS6sCyGESJqEiBBCiKRJiAghhEiahIgQQoikSYgIIYRI2v8HuFhetzVJkOEAAAAASUVORK5CYII=\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "sns.violinplot(x='precipType',\n",
    "               y='polarity',\n",
    "               data=twitter_moves[twitter_moves['precipType'] != 'snow'])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.axes._subplots.AxesSubplot at 0x7fd21cbda160>"
      ]
     },
     "execution_count": 16,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "sns.violinplot(x='precipType',\n",
    "               y='subjectivity',\n",
    "               data=twitter_moves[twitter_moves['precipType'] != 'snow'])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "the p-value for a subjectivity-difference between rain & shine is 0.827896273529459\n",
      "--\n",
      "the p-value for a polarity-difference between rain & shine is 0.9882673041937862\n"
     ]
    }
   ],
   "source": [
    "from scipy.stats import mannwhitneyu\n",
    "\n",
    "twitter_moves\n",
    "\n",
    "rain = twitter_moves[twitter_moves['precipType'] == 'rain']\n",
    "clear = twitter_moves[twitter_moves['precipType'] == 0]\n",
    "polarity_results = mannwhitneyu(rain['polarity'], clear['polarity'],alternative='two-sided')\n",
    "subjectivity_results  = mannwhitneyu(rain['subjectivity'], clear['subjectivity'],alternative='two-sided')\n",
    "print('the p-value for a subjectivity-difference between rain & shine is {}'.format(subjectivity_results.pvalue))\n",
    "print('--')\n",
    "print('the p-value for a polarity-difference between rain & shine is {}'.format(polarity_results.pvalue))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Neither the plots themselves nor the `mannwhitneyu` statistical test offers any evidence for the weather influencing the `polarity` or `subjectivity` of tweets in this case. ☔️✨☀️🌈\n",
    "\n",
    "## Tweets per Day & the Weather\n",
    "\n",
    "Let's now check whether the weather impacts how many tweets per day are being sent. Given that my own Twitter frequency fluctuates a lot over the years I want to limit myself to the year 2016 for the analysis:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {},
   "outputs": [],
   "source": [
    "tweet_year = \"2016\"\n",
    "# you can get these values from Google Maps by clicking on a point and copy&pasting the values \n",
    "# from the URL/bottom information box\n",
    "\n",
    "target_latitude = 50.114462 # If you're south of the equator this one should start with \"-\"\n",
    "target_longitude = 8.746019 # If you're west of the prime meridian this should also start with a \"-\"\n",
    "km_distance = 25 # we don't want to be too strict in filtering. "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<seaborn.axisgrid.FacetGrid at 0x7fd21cbb7860>"
      ]
     },
     "execution_count": 19,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 864x576 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# earth's radius in km = ~6371\n",
    "earth_radius = 6371\n",
    "\n",
    "# latitude boundaries\n",
    "maxlat = target_latitude + np.rad2deg(km_distance / earth_radius)\n",
    "minlat = target_latitude - np.rad2deg(km_distance / earth_radius)\n",
    "\n",
    "#longitude boundaries (longitude gets smaller when latitude increases)\n",
    "maxlng = target_longitude + np.rad2deg(km_distance / earth_radius / np.cos(np.deg2rad(target_latitude)))\n",
    "minlng = target_longitude - np.rad2deg(km_distance / earth_radius / np.cos(np.deg2rad(target_latitude)))\n",
    "\n",
    "twitter_moves['tweets_per_day'] = twitter_moves['date_twitter'] \n",
    "twitter_moves_subset = twitter_moves[(twitter_moves.index.year == int(tweet_year)) & \n",
    "                                    (twitter_moves['longitude_moves'] >= minlng) & \n",
    "                                    (twitter_moves['longitude_moves'] < maxlng) &\n",
    "                                    (twitter_moves['latitude_moves'] < maxlat) & \n",
    "                                    (twitter_moves['latitude_moves'] > minlat)]\n",
    "\n",
    "\n",
    "sns.lmplot(x=\"temperatureLow\", y=\"tweets_per_day\", data=twitter_moves_subset,lowess=True,size=8,aspect=1.5)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "There seems to be somewhat of a positive correlation between the temperature and the number of tweets per day. But how does it correlate to the presence of precipitation?"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.axes._subplots.AxesSubplot at 0x7fd21a49a4a8>"
      ]
     },
     "execution_count": 21,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAYUAAAEOCAYAAABmVAtTAAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADl0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uIDIuMi4yLCBodHRwOi8vbWF0cGxvdGxpYi5vcmcvhp/UCwAAIABJREFUeJzs3XecVPW5+PHPOdNne6MjIOhXoygqimCPXaJGk5iY5Ca5MbmWxCQ3yS/JTc+9N8mNSYyxK9bYwI4IgqAoiqDSVEC+0tsusL1On/P748wuxQVmYWfO7Ozzfr32tTtnz5x5YGf3Od/2fA3LshBCCCEATKcDEEIIkTskKQghhOgiSUEIIUQXSQpCCCG6SFIQQgjRxe10AIertrZVpk8JIUQPVVUVGd0dl5aCEEKILpIUhBBCdJGkIIQQooskBSGEEF0kKQghhOgiSUEIIUQXSQpCCCG6ZH2dglLqp8CPgDJgGXC91nq1Umos8E/gZKABeBT4b621rEMQQogsyWpLQSl1PXA9cAkwEHgL+KVSKgDMBBYBw4DLgetS5woh+rlEIkFdXS1S6j/zst1S+DnwS631ytTjXwIopb4IFAC/01rHgVVKqduBG4B7sxyjECLH3H33bSxfvpTJk6/gC1/4itPh5LWsJQWl1FBgFFCglPoIu0XwNnAjcArwUSohdFoG/EUp5ddah7MVpxAi96xfvw6A6urtDkeS/7LZUhiW+vxV4DIgCjwGPAV8DDTuc34DdvdWGVCzv4uWlQVxu129HqwQIocYdrdRNBqmqqrI4WDyWzaTQmfxpb9qrbcCKKX+C1gCrD3A+QfsRGxs7Oi1AIUQuceyLNrb2gFoamqhtrbV4Yjyw/6SazaTwo7U54Y9jm1KfR4MFO5zfjmQ4NMtCCFEPxIOh0kkEgC0trU4HE3+y+bso21ALXDSHsdGpT4/ApyglPLu8b3TgOVa60h2whNC5KKWluaur9taW0kmkw5Gk/+ylhRSg8h3A79SSh2rlCoD/og9FfVFoB74g1KqQCl1IvAD4PZsxSeEyE17JoVkMkl7e5uD0eS/bK9o/iPwAvaso21AE/CNVGtgMnAqsB14FrhVa/1YluMTQuSYpqa9e5Cbm5sciqR/yOo6Ba11DPhh6mPf730MXJDNeIQQua+x0U4KrhIvieYojY2NDBt2hMNR5S+pfSSEyGmNjfbcFHe5H4CGhnonw8l7khSEEDmtMwl4KnwA1NfXORlO3pOkIITIafX1dWAa0lLIEkkKQoicVldfi+l3YQbtIVBpKWSWJAUhRM6KxaK0NDdjFrgxTAPT75KkkGGSFIQQOau+3u4qcgXsVoIZdNPQUC8L2DJIkoIQImd1tgo6u47MoJtkMvmptQui90hSEELkrM5B5a6kEJBxhUyTpCCEyFmdaxQ6k0Hn584FbaL3SVIQQuSszm4i09+ZFFx7HRe9T5KCECJnNTfbxfBMvyv12U4OexbJE71LkoIQIme1traAAYbX/lNl+uzk0NIi+ypkiiQFIUTO6uhox/CYGIa9EaPhMbuOi8yQpCCEyFmhUAjDvfvPVGdSCIVkG95MkaQghMhZkUgEw210PTZMA0yDaDTqYFT5TZKCECJnxeMxMI29jhmmQSwWcyii/CdJQQiRsyzL6hpP6GIAWE6E0y9IUhBC5DRr37//kg8ySpKCECJnuVzuT2UFK2nZx0VGSFIQQuQsr9eLldidFCzLgqSF1+t1MKr8JklBCJGzAoEAxHeXybbi1u7jIiMkKQghclYwWEAymrRbCIAVSwAQCASdDCuvSVIQQuSsoqIiSFqQaiFYkeTu4yIjJCkIIXJWUVExAMmI3UJIhhOp4yWOxZTvJCkIIXJWaWkpAMlwfK/PncdF75OkIITIWaWl5QAkQ6mWQqgzKZQ5FlO+k6QghMhZ5eUVwO5k0JkcOo+L3idJQXSJRMKEw2GnwxCiS1dS6OhMCvG9joveJ0lBAFBXV8v3v/9dvv/971Bdvd3pcIQAoKKiEoBEKikkOuIUFhbh8/mcDCuvSVIQANTUVJNIJEgmk1RXb3M6HCEAKCgowOfzkQzFsSwLKxSnokJaCZkkSUEA0NbW2vV1a2vrAc4UInsMw6CiopJkRxwrmsRKWJSXVzodVl6TpCCA3RukQ2pfXCFyRHl5BVYsSaLV3lhHWgqZJUlBANDc3NT1dVNTo4ORCLG3sjJ7Wmq8IbLXY5EZkhQEAI2NDd1+LYTTOtckxJsiez0WmSFFyQUA9fV1gAGGSX19vdPhCNGlc/VyvCmaeixJIZMkKQgAamt3YXiCGKaburpd3W+DKIQDiovtOkfJtthej0VmSPeRIBIJ09LSjOktxPQWEQ6HaWmRwWaRGwoL966IKhVSM0uSgmDHjh0AmN4iTG8hADt31jgZkhBdCgoK93ocDBY4FEn/IElBUFNTDYDpLcb02qWKZVWzyBXB4O4NdTweL2639Hpnkvzviq4VzKavGMN0p45JUhC5Yc+SFn6/lLfINEkKgu3btwJg+kowDPdex4Rwmtfr6/ZrkRmSFARbt27BcPkw3fZm6IangG3btsgMJJETXC5X19fSdZR5aY0pKKXkJ5GnOjo6qKurxfTt3snK5SultbV1r1XOQjjFNM2umxPTlGHQTEv3f3inUuoepdSkjEYjsm7bti0AmP7dSaHz6y1bNjkRkhDCQekmhe8ChcBMpdR6pdR/K6WOzmBcIks2b94EgMu/e5Womfp6y5bNToQkxF4sy8KyLACSyaTD0eS/tLqFtNbPA88rpTzAhcDVwDtKqQ3Av4AntNZSRa0P6mwNmHskBZckBZFDYrFYt1+LzOhRB53WOqa1ngX8D3ALcBTwF2CzUupWpVQgAzGKDNqyZTMYLkzv7lWihjuI4fJ2tSKEcFI0Gu36OhaLHuBM0RvSTgpKqTKl1A1KqbeB9cClwI+BgcAJwLHAAxmJUmREPB6nunpbairq7reCYRiYvlJqa3cSCnU4GKEQEA6Hur4OhUIHOFP0hrS6j5RSLwKXANuwu4u+rrXetMcpbUqprwBpr3hSSv0D+JHW2kg9Pge71fEZoBq4TWt9b7rXEz23ffs2EokEnqJPV500/WUkOnaxdesWjj76GAeiE8K2ZyKIx+PEYlE8Hq+DEeW3dFsKdcAFWusxWuv/3jMhKKXKALTWzdgD0gellBoH/NsejwcBM4BHsVse3wb+opS6JM34xCHobjyhk4wriFyx51ax9uM2hyLpH9IdaP5Od8eVUkOAlUB56rynDnYtpZQJ3AvcCvwxdfjrwCat9T2px+8opR4DbgBmpxOj6LmtW+0/+K49pqN26ly30HmOEE7Zd3vYlpYW2X0tg9LtPjoaeBAYD+zbblvRw9e8HugAnmR3UjgFWLbPecuAqw52sbKyIG6362CniW7s2GH39u25cK2T6SsGw6S6eitVVVKqWDgnFrPHtVxlPhKNESAi78kMSnel8j3AZuAfwFPAl7ATxNnY01PTopQaCPwOOGefb1UAq/Y51gBUHuyajY0yEHooLMtiw4aNGN6iriJ4ezIME9NbwubNm9mxo2mvUgNCZNO2bXZpd3epnRQ2bdrOiBHK4aj6vv0l1nTHFE4Gvp1ar5DUWr+ktf4tcDtwWw/iuBWYorXWaZxrAFYPri16oKGhno6OdlzdtBI6mf5S4vE4O3bI3grCOfX1tQB4Kv2px3VOhpP30k0KEaDzVrFDKdV5B/8ycEU6F1BKnQ+cxu4uoz3V8ulWQXnquMiAzrECs5vxhE6dYw1bt27JSkxCdKe2thbDZeAu96Ue73I4ovyWblJ4A5iRWpz2HnCbUmoCcBOQ7lSArwNDgW1KqTpSYwipr1did0ft6TRgcZrXFj3U+Yf+gC0FGWwWDrMsi127dmAWuDGDbjANdu3a6XRYeS3dMYWbsFcwx4GfYrcQvgq0Av+R5jV+DPxmj8fDgEXAOOxWyM+VUt/DHtA+HfgacFma1xY91DnV9MAthc5pqZuyEZIQn9La2kIoFMJTGsQwDFwFbnbsrJGy7hmU7pTUBqBzWuoqpdSR2OsJarXWiTSv0Qh01UfqLMettd6WejwZ+CvwZ2ALcKPWekGa/w7RQ1u2bMZw+TDcwf2eY7i8GJ6grFUQjuncKtZVZE96NAs9hGo6aGlppqRk/zc04tDtNykopdK6S1dKkaqH1COpBXDGHo8XAlKaOws6Ojqord2Jq2DgQe+2XL4yWlu309jYSFnZpxe5CZFJnVvFuos89udiL7GaDqqrt0tSyJADtRRe3uexxR5/xFOSQDtQ0ptBiczavHkjsHe57P0xA+XQtp1NmzZQVnZKpkMTYi/bttnbwrpK7JaCq9ibOr6FY489zrG48tl+B5q11mbnB/ZahKnAWOzFa8XYA8PPAf+ejUBF79m0aQMApr/ioOe6/OV7PUeIbLKr+IIr1VLoTA4yIy5z0h1o/htwsta6c715G7BMKXUjsAR4PhPBiczYsGE9AK7AwUsFmKmksHHj+ozGJMS+kskkW7dtwVXkxXDZ96+uQg+Gy5Cy7hmU7pTUCuyd1/blJ1X3SPQNlmWxfv1aDLf/gIPMnUy3D9NTyPr162TXK5FVNTXVRCMR3KW7K+sYpoGrxMv27Vv32mdB9J50WwqvAPOUUvdjzwwCGI5dFfXVTAQmMqO+vo6mpkbcRcPSntJnBisJNW+ipqaaoUOHZThCIWydrVN3mW+v4+4yH+GGCFu2bGLMGNkVuLel21L4DvAi9pqER7DrH/0IWABcl5HIREasXWtXGHEFDlpWqkvnuZ3PFSIb1q9fC4C73L/X8c7Hnd8XvSvddQodwC9TH/ullPqL1vrnvRGYyIyupBCsSvs5ned+8snHnHvu+RmJS4h9ffLJGgy32TW43Mld4U99X3PxxZOdCC2v9WiP5jTc3MvXE73s449XY5jubjfW2R/TW4zh8rFmzcdYltQoFJnX3NxETU017nIfhrl3N6craJe80PpjGefKgN5OCrLuPIc1Njawc2cNZqBqrz2ZD8YwDFzBATQ1NUrFVJEVH39sV9L3DAh0+31PVYCOjnaZhZQBvZ0U5DYyh61evRIAd8HAHj/XlXpO5zWEyKQPP1wO7E4K7R/V0/5Rfdf3PQPt4x991NM9vsTB9HZSEDls1aoPAXAVDOrxc92p56xa9VGvxiTEvuLxOB98uAIz6O4aT4hubye6vb3rHM/AIBiwfPkSp8LMW5IU+olkMsnKlR9huAOYvp5XJTG9hZjeIlavXkksFstAhELYVq78kFBHB97Bwf1OmzY9Jp4BATZv3sTOndKl2ZskKfQTGzeup62tFXfB4EMuOewqHEw0GuGTT9b0cnRC7LZ48dsA+IZ3t152t87vv/PO2xmPqT+RgeZ+YsWKZQC4ioYc8jXchUP2upYQva25uYmlS9/HVeTBtc+itX15hxRgeEwWLJhPPB7PUoT5L62koJT6RprXm3IYsYgMWr58CRiurrGBQ+EKDsAwPSxfvkSmpoqMmD9/HolEAv/okoO2aA23ie+IQpqbm1iy5L0sRZj/0m0p/F0pdeC2HKC1/sFhxiMyoKZmO9XV23EXDMIw979eMbxzBeGd+5/NYRgmrsIhNDTUs2nTxkyEKvqxjo525s6bjel14TvioH9uAPCPLgEDZrz8gqxZ6CXp1j76LfCIUuoR7NpHe7XVtNarezku0Yvef/9dANzFww94Xrw1VdZq4Lj9nuMuHk68ZTPvv7+YUaOO7LUYhZg9eyahjg6Cx5VjuNO7X3UVevANL6Rmy3beffcdJk48M8NR5r90Wwp3Ye+p8BKwAlgJfLTHZ5GjLMti8eKFdtdR4dDDvp67YDCG6eHddxfJnZnoNbt27WT27Jcx/W78o4t79NzAsWUYpsHTzzxJOBzOUIT9R7pJYVQ3H0fu8VnkqE2bNrJjRw3uwiEYLs9hX88wXbiKhtHYWC+zkESvsCyLJ598lHg8TnBs+q2ETq4CD/6jS2huauLFF5/JUJT9R1r/+1rrzVrrzdi7ro3Z4/GW1GeRo95++00APKWjeu2anhL7Wm+99UavXVP0X2+99QYffrgCz4AA3mEFh3SNwNGluAo9zJ07G60/7uUI+5d0Zx8NUEq9A6wBZqaODQfWKqVUBuMThyESCbNo0dsY7sAhrWLeH1ewCtNTyJIl79Le3tZr1xX9T3X1dp586l8YHpOCk6sOeQ2N4TYpGF+FhcX9U+6itbXl4E8S3Uq3nXY/sBoYwO76RtuAJ4HbMxCX6AWLF79DOBzCU3pkjwrgHYxhGHjKRhOLxVi4cEGvXVf0L6FQB3feeSvRSISCkypxBdOd99I9T7mfwLFlNDY0cN99d8qY1yFK9y/FZ4Efaq3rSSUFrbUF/Ak4NUOxicNgWRZz584Gw8BTOrrXr+8pORLDdDFv3hz55RM9lkgkuO++u9ixowb/mBJ8w9KbgnowAVWKZ1CQ1atXMnXq471yzf4m3aTQCnQ3SjkAWcWck1au/JDq6m24i47A9Bx8L+aeMtw+3MUjqaurZdmy93v9+iJ/WZbFE088wocfLsczIEDw+N7b5t0wDArHV+Eq8jJv3mxeffWVXrt2f5FuUngdeEgp9RkApVSlUup84Dnsaaoix8ycOR0Ab8UxGXuNzmu/PHO6rHAWaZs+/TneeOM1XCVeCicM/NQmOofL9LooOmMgpt/FtGmPs2iR1EbqiXSTws2pc1cCfmAXMBt7nEFWMeeYNWtW88kna3AVDMbVgx3Wesr0FuEuGs6WzZv44IPlGXsdkT9eeWUGL730PGaBh+JJgzA9manJ6Qp6KJo0CNwGDz54L0uXShmMdKU7JbVJa/157O6iicA4oFxr/U2tdXMmAxQ9Y1kWzz//NAC+quMz/nre1Gu88MIzMrYgDmjOnJk888xTmAE3xWcOwgwc3sDywbhLfRSdMQjLhHvvvYOlS6WbMx1pp2ml1GDgKuAS4PPAF5VS6e/+LrLigw+WsW7dJ7gLh+IKVGT89Vy+EtzFI9i6dTPvvbc4468n+qZXXpnBtGlPYPrthOAqOPyFlOnwlPspmjSQpGFxzz3/ZMmSd7Pyun1ZuusUrsGuefQ74GLgUuB/gW1KqasyF57oiXg8ztNPPwUYeAeckLXX9VWNBcPkueemEotFs/a6IvdZlsVLLz2/u4Vw9mBcRd6sxuCpDOzVYpAxhgNLu0oq8D2t9TCt9SSt9USt9VDssYY7Mhee6In58+exY0c1ntLRuA5hd7VDZXoL8ZQdRX19ncz2EF0sy+KZZ57ixRefxSxIJYTC7LQQ9uWp8FN05iAsFzzwwD288cZrjsTRF6SbFMqAh7o5/jCQvb8+Yr9aWpp58cVnMFyern7+bPJVHofh8jFjxgs0NNQf/AkiryWTSR577CFmz34ZV6GH4rOGZK3LaH885X6KzxqM4TX5178eZM6cmY7Gk6vSTQrTsbuN9nUu8HKvRSMO2dNPP0koFMJbORbT7c/66xsuL74BJxKNRnnqqX9l/fVF7kgkEjz44L1d006Lzx582KuVe4u71EfRWYMx/W6mTXuC6dOfk+nU+0j3J7UZeFQp9T6gARcwGpgAPKmUuqXzRK31z3o9SnFAH3+8infeeQvTX4anbIxjcbhLRuFq2sDSpe+zYsVSxo07xbFYhDPi8Tj33XcnS5e+h7vcR9GkQZhel9Nh7cVd7KX4nMG0vF3D9OnPEYlE+NKXrj3kukv5Jt2WwkTsfRP8wInA8UAA+DD19ampj/EZiFEcQCQS4ZFHpgAG/kGn9mqNo54yDAPf4FPBMPnXvx6io6PDsVhE9sViUe666x92Qqj0U3zG4JxLCJ1cBR6Kzx6Cq9DD7Nkv8+ST/5IWQ0paLQWt9XnpnKeU+svhhSN66vnnn6a2dhee8mNwBXqvXMChcvlK8FZ8hqa6lTz99BN861vfdTokkQWxWIy77rqtqwR20ekDe7wvQra5UrOhWt7ewWuvzcGyknzta9/q9y2G3v6p3dzL1xMHoPXHzJs3G9NblJWFaunyVh6L6StlwYL5fPjh/vd8Fvlhr4QwMEDRxNxPCJ1Mv5viswbjKvHy+utzpcVA7yeF/p1is6ijo4MpD9yDZYF/yAQMMzcG8gAMw4V/yAQwTB566D6pbZ/HkskkU6bc1VXcruj0gRiuvpEQOpk+F8VnDsZV7OW11+bw/PPTnA7JUb390+vfKTZLLMvi8ccfoqG+Dm/lZ3AFKp0O6VNc/jK8VWNpaWnm4Yen9Pu7r3xkWRaPPvoAS5bYYwhFE/teQuhkJ4ZBuAo9zJz5Eq+8MsPpkBzTN3+C/dzbb7/J4sXvYPor8FYe53Q4++UtV7iCA1ixYinz5s1xOhzRy6ZPf4633noDV6mPoomD+mxC6GT63RSdORgz4OaZZ57qtyuf+/ZPsR/aunULjz/+CIbLS2DoREdnGx2MYZj4h0zEcPt4+uknWL9+ndMhiV7y9ttvZqXaaba5gm6KzhiE4bG7Pvvjfs/58ZPsJzo62rnrrn8Qi0XxDT4N09s7u1VlkukJ4B88kUQiwd1330ZLi4wv9HXr1n3Co48+gOF1UTzJ3rcgn7iLvRRNGEjCSnLnnf+grq7W6ZCySgaa+4hkMsm9997Brl078VYci6domNMhpc1dOAhv1VgaGxu4++7biMfjTockDlFjYwN33nkriWSCotMGZL24XbZ4BgQoOLGC9vY2br/970QiYadDyppDTgpKqe52b5lyGLGIA5g69TFWrvwQV8FgvFVjnQ6nx7wVn8FdNJxPPlnDv/71kAw890HxeJx77vknLS0tBMdW4BkQcDqkjPKPKsY3qoht27b0q/dsuqWzT1BKLd7j8dNAvVJql1JqQudxrbXswpYBr776CvPmzcH0leT8OML+GIaBf8gETH8Zb7/9BjNmvOB0SKKHnntuGuvWrcU7rAD/6GKnw8mKghMqcZX5WLTobd566w2nw8mKdP+63IG9/SZKqSuBC7GL4d2S+hAZsmjR20yd+hiGO0Bg+NkYrr7bXDdMN4HhZ2N6CnjxxWelfHEf8sEHy5kzZyauQg+FJ1X1m1W/hsugaMIADK+Lx594hO3btzkdUsalmxTGYW+qA/aua9O01guA24Ds7ebSzyxfvoQHH7zXnmk0/BxMT4HTIR020x0gMPwcDLePxx57iMWLFzodkjiIxsZGHnzwHgzToPC0ARh5MtMoXa6gh8KTK4nHYtx9zz+JRCJOh5RR6f50o4BXKeXCLqHdWS7b04NriB5Ytux97rrrNixMAsPOxuUvdTqkXmP6igkMPxdMN1Om3M3ixe84HZLYj84Vy21tbQTGluMu9TkdkiO8Q+wus5rq7Uyd+pjT4WRUurURFgDPAjHsVctzUgniN8DydF9MKTUCuBU4J3Wd+cCPtNbVSqmxwD+Bk4EG4FHgv7XW/WN0Zw+LFy/kgQfuwcLEP/xsXMHcW7F8uFz+MgLDzyW09Q2mTLmLaDTC2WenVXdRZNHMmdNZs2Y1nsFB/Ef2j3GE/QkeX06sLsybb77OZz5zPKeeerrTIWVEunf5NwJbgTBwhdY6BhQCVwHf78HrzQBCwJHAcUAFcL9SKgDMBBYBw4DLgeuA63tw7bzw2mtzuH/K3ViGC//wc3AHBzgdUsa4AhUEhp8HppdHHpnCrFkv9ZsZHn3BmjWr7a00A24KT+k/4wj7Y7hMik4bgOE2eejh+9m5c4fTIWVEuknhc1rr67XWX9ZaLwXQWjcDpwCXpXMBpVQpsAT4mda6RWu9C3sK69nAZKAA+J3Wuk1rvQq4HbihZ/+cviuZTDJ16uM88cSjGC4fgSM+iztY5XRYGecKlBMY8VlMT5Bnn53KY489RCKRcDqsfq+pqZF7770DCyg8bUDO7ouQba4iLwXjKoiEw9x99215Ob5wwO4jpZQb8AF3KqWe4tOL0xTwB9KYgaS1bgK+vc/h4cB27OTykdZ6z1VNy4C/KKX8Wuu8XjnS0dHBAw/czYoVyzC9xfYMnT6wWrm3uHwlBEZcQGjrAt544zV27drJDTf8gMLC/vN/kEs6S2G3tDQTHFuOpyL727vmMt8RRcTqwmzdtIVHH32A7373prxqRR1sTOF72GMAAG37OeeQRgmVUgr4NXbX1LlA4z6nNGC3ZMqAmv1dp6wsiNvdd+9itm7dyp///Ee2b9+OKziQwLAz+vS000NleoIER5xPqHoRq1ev5H//+Bt+/atfMWrUKKdD61csy+KOO+5g/frUeoQxJU6H1CWXuhYLTqwk0RJl8eKFHHvs0Vx99dVOh9RrDpgUtNb/VEo9gX03f1E3p4TowUBzJ6XUKcAs4O9a6yeVUud2c1pn6j3gO6Gxse9u+bho0ds8+uiDRKMRPOXH4BtwQp9cmNZbDJeHwLCziNatZNfOVfz4Jz/h61/7FmeddW5e3YnlspkzpzN37lxcJV4KT86NcYR4c5RkKA4WNL66laIJA3GXOHvjZK9fGEjz/GoefvhhAoFixo+fcPAn5pCqqqJujx909pHWuk4pNUprXQ2glPKkBpoPiVLqYuBp4Bda63tSh2uBY/c5tRxI8OkWRJ8XDod56ql/8dZbb2CYHvxDJ+EpPsLpsHKCYRj4qsbi8pcTrnmXRx6Zwpo1q/j6179NMBh0Ory89s47b/Hcc9MwA26KJw3Kmd3TWt/d2XVrmGyL0fruTsouGu5sUIAZcFM0aSAtC2q4//67KC4u4eijj3E6rMOW7k99p1Lqj0qpaqAVQClVrJR6RCmVdsdvqiTGNOAbeyQEsAegT1BK7Zn+TwOWa63zaiRn/fq1/P73/8Vbb72B6SsjOOoiSQjdcBcNJTjqYsxABYsXv8Nvf/vzflnGOFuWLn2fhx66D8NjUjRpEGYgN3byS4bjJNv2vgdNtsVIhnOjqKK71EfhhAHEEwn+cdstbNy4wemQDlu6SeEW7NIW32N3d44JVGKvaj6o1KD1g9gzjKbv8+1ZQD3wB6VUgVLqROAH2DOQ8kIsFuO556bx5z//oavSaXDkBZje7ptwAkxPAcER5+OtPI6GhgZuueV/mTr1sbyc8eGkDz9czr333YFlQtEZgxzvmtmTlei+93h/x53gHRik8NQqIpEwt976f2w7DGBwAAAgAElEQVTZstnpkA6Lkc7gjVJqF3Cy1nqbUqpDax1MHa8CPtRaD07jGmdhL4Lr7jdaAUHsGkvjsbuT7tda//Vg162tbc2dd8d+bNy4ngcfvI/q6m2YngJ8gyfgLsi99Qdt6+x1AkVHXel0KJ+S6KgjXPMuyWgrAwYM4rrrrueoo5TTYfV5K1Ys5a67/0nSSlA0cVDOVT5NtMdomrP1U8dLLx6Oq8DjQET7F97USvuyWoIFBfy/n/6KESNGOh3SAVVVFXU7YJRuG9ENdLdSIwKkdaurtX6Lg++3cEGa8fQJkUiYF154lrlzX8GyLDylY/ANPBHDzK03M0Ai3IQVCwEWbetnEhh6Rk6V1nAFKwmOuphI7Ufs2qX585//wHnnXcgXv/hlAgEZazgUS5a8y3333UkSKycTQl/jH1kEBrQvreWWv/4vP/7PnzN69FFOh9Vj6XYfLQV+tueB1FjC34F3ezuofLBy5Qf85jc/59VXZ2F4CggccR7+weNzMiEAhLYvpLNn0Iq2Et6ee4XqDNONf+BJBEdcgOktZv78ufz61z9j+fKlTofW5yxYMJ977rmdpGFRNGmgJIRe4h9RROH4KkKhDv761z+yatVHTofUY+l2Hx2PXTrbgz2OsAYYib1+4MrUCmRH5Fr3UXNzE1OnPs67774DGHgrFN7K4zHM3Bi4604yHqJ97b7DPFBw1JWY7tz8Y2ElE0TrVxOt/xisJCeffCpf/eo3KC+vcDq0nGZZFq+8MoNnn52K6XXZYwhluVvkri91H+0pWt1O23u7MA2T7373e5x2Wu7VSdpf91FaSQEgVZ/oc9h1i0LAOmCO1trRmgS5khSSySRvvvk6zz77FKFQCNNfgX/weFz+7jaoyy3JaBvt61/+1PGC0Z/L+ZXViUgzkZolJEK1+Hw+rrrqGs4//yJcrr67oDFTOkupzJs3255OecYg3MW5M6jcnb6aFABitSFaF+2EhMW1136DCy642OmQ9nLYSQFAKXUUcITW+rXUY8PpKqa5kBS2brWXu2/YsA7D9OCtOgFP2eg+sxCtLycFsO9+Y80biO76ACsR5YgRI/nmN77DqFFHOh1azojFYjz44L28994iXEVeis4YhCuYu63XTn05KQDEmyK0LtxBMpJg8uQruPrqL+fEgkA4zKSglBoAvAhMAGJaa79Sajh26evJWmvdm8H2hJNJIRIJM33687z66iySySTu4iPwDTgJ05ObXS7709eTQqdkPExk1wrizZswDIPzz7+Iq676Ur8fiA6FOrjjjn+wZs0q3BU+iiYO6jMF7vp6UgD739C6cAeJthhnnHE23/zmd3C7nU/Ihzv76H5gNXZJ68796LYBT2KvJcitdlEWrFz5AY8++iD19XWYngICQ8fjLjzozFyRQabbT2DI6cRLRhHZsYR58+awZMl7/Nu/fZuTTjrF6fAc0dLS3DV33jM4aJd+dvWNFmy+cBV4KD5nCC3v7GDhwgW0trZy000/xOvNza67dN8dnwV+qLWuJzVFJdVt9Cfg1AzFlpPa2lqZMuVubr31L9TX19uL0I68VBJCDnEXDCQ46hK8lcfR1NzMHXf8nXvuuZ3m5manQ8uqurpa/vTn37Nly2Z8I4somjBQEoJDTJ+LkrMG4xkQ4MMPl/P3v/+Zjo52p8PqVrrvkFbsmUf7GsDB1x7kjRUrlvLrX/+MRYvexvTbJSp8A07M6ZlF/ZVhuvBVje0qlfH++4v59a//H++/v9jp0LJi584ae/X8zp0Eji6l4KRKDLPf/KrmJMNtlxDxDitg7VrNLbf8L21trU6H9SnpJoXXgYeUUp8BUEpVKqXOB54DXspUcLkiFArx4IP3cvvtf6eltRVv1YkER17YJ2YW9XcuXwnBERfgG3gyHaEQ99xzO/fee3vO3qX1hpqaav7v//6HxsYGgseXEzy+PGcGN/s7wzQoPHUAvpFFbNmymVtu+SMtLS1Oh7WXdJPCzalzVwJ+YCf2uoXV2DWK8tamTRv4wx9+ycKFC1Ktg4vxVR7bZ2YWCbvyqrf8aIIj7VbDe+8t5re/+y/WrfvE6dB6XW3tLm655X9pbm4iOLacwNG5sypd2AzDoOCkSnyjitm2bQt/+9ufcuomJd2/bCGt9eexu4smAicB5Vrrb6a25cxLCxbM549//F2qgN0xBEdegMuXO5uOiJ4xfcW7C+zV1/N///ffXSVI8kFTUyN//dsfdyeEoyQh5CrDMCgYV4FvVBHbtm3httv+mjOFHtNNCk1KqXnAd4Ak9taZudcZ1ksSiQSPPfYQjzwyhSRuAsPPwTdgHIbRN6bxif0zDBNf1VgCR5yHZXp56qnHeOih+4jFDnmLkJwQiUS47Z9/pa62lsAxpZIQ+gA7MVTiHVbAunWfMGXK3SSTSafDSjspnA68AIxLfa5XSj2vlLopta1m3ojFotx1123Mnz8P01dKcOSFMrMoD7kLBhAceRGmv5yFCxdw++1/IxLpm1uBW5bFww/fz5bNm/CNKCJwrIx19RWGYVA4fgDuSj/Llr3PjBkvOB1SeklBa/2B1vourfVXtNbDsMtbvwrchD2ukBdisRj//OffWLFiKa7gQIIjzu9Ti7dEz9j7Qn8WV+EQVq36iL///f9ypgnfE/Pnz+O99xbhLvdRMK5SBpX7GMO0t/Y0g26mT3/O8SJ6aY+WKqXKlFKXK6X+AjyBvUahFvhzpoLLpmQyyQMP3MPq1StxFw4lMPxsDFffWDEpDp1hugkMOxN38RGsW/cJ9957O4mEo+W8emTHjhqmTXvcLm43YSCGSxJCX2T67J8fhsGDD95Le3ubc7Gkc5JSaiX2lplfBbYC3weqtNbnaa1/ncH4suaVV2bw/vuLcQUq8Q+dhGH2v/EDr9fLkCFDcnalZaYYhol/yARcBQP54IPlvPjis06HlBbLsnj00QeIxWIEx1XkzBaamdAf3pvuMh+BY0tpamrk2WenOhZHui2FbUAhMBwYAgxMPc4LW7Zs4oUXnsFwBwgMO6vfJoQbbriB++67jxtuuCGvf/m6YxguAkPPxPQUMGvWS31iuury5UvQ+mM8g4L4huXNr+On9Kf3ZuDoUlxFHhYsmM+2bVsciSHdMYVLgMHAj7D3Ur4eWKeUWqGUuiOD8WXFs89OI5lM4h98GoY7d2vLZ1JlZSUXXnghABdeeCGVlZUOR5R9hsuDb8gELMvi6aefzOmpqpZl8fzzz4ABBWPLnQ4no/rTe9MwDYJjK7AsixdffM6RGNIeU9BaJ7XWS7D3Ub4FuBXwATdmKLas2LJlMytXfoArOKBfzzKqq6tj7ty5AMydO5e6ujqHI3KGOzgAV+EQ1q37JKdbC6tWfUR19Ta8wwpxFeXvnTP0v/emZ2AAV6mX5cuXUFu7K+uvn1YnpFLqCuCM1Mcp2OMKrwO/T33usz74YBkAnrIxDkfirGg0yr333suzzz5LXV0d0Wi022JX/YG3bAyhtmpWrFjGUUfl5ozrBQvsX7vA6GKHI8m87t6b+VwM3TAMAmNKaFtSy4IF8/nCF76c1ddPt6VwJ/Y4whTgKK310VrrG4BXgNcyFVw2dN4NuoMDHY7EedFolOrqaqLRqNOhOMoVtOs8rl+/1ulQuhWNRvnwwxWYhR5cObyVZm/qb+9N75ACDJfBsmXvZ/21D9hSUEqNx95YZyDwHlAAXLHHerXR2Ntz9lnt7e1gmP12LEF8mmG6MVweR6cFHsjatZpoNIp/RImsSchThtvEPSBATU019fV1VFRkbxzlYN1HBdgb6HiA/9fN90NAn56S6nK5wLKwkol+OetIfJqVej/k6j7PmzZtBMBd7nc4EpFJnnI/sZoONm3amDtJQWv9JvCmUmqm1npylmLKqsGDh7B2rSYZacYVyO9ZHCI9VqwNrASDBg1xOpRuVVfbmx+6S/J7gLm/c6V+vtXV2znllOztZZbulNS8TAgAxx03FoB4qzNzgkXuibXY74XO90au6dyYxfTnZktG9A7TZ/98s70RT7/fFGDcuJMJBguINW3Aive9ujeid1mJGLHGtXi9PsaPP83pcLrVVZ9JSlrkNcNt/3yzXaix3ycFj8fLFVdchZWIEqn9wOlwhMMidSux4mEuu+xyAoHcnPgYCAQAsOK5u7hOHL5kzC6jne33Yb9PCgCf/exFDB06nFjTBmLNm50ORzgk3rqdWINmwICBXHxx7vaYFhfbGz0lO+IORyIyKRmyf77FxdldiyJJAXC73dx00w/x+fxEdrxHvCP7qwiFsxKhesLVi3F7PNx004/w+XJ3ivKoUaMBiDf0zf0fRHriDXY3YefPO1skKaQMHjyEG2+8GdOA8NYFJDryeym92C0RaiC09U2w4vzHd2/iiCNGOB3SAR199DEARGs6HI5EZIplWcRqOnC73Ywcmd2lYJIU9nDCCSdx/fU3Y5AktGU+sZZtTockMizeVkNoy+uQjPHtb1/P+PETnA7poIYMGcqoUaOJ7egg0dG3txEV3YvXhkm0xTj11NPx+7O7HkWSwj7Gjz+NH/zgJ3g8bsLb3yZStyqnq2X2iv0t2svjxXyWZRGtX0No6wJcLoMbb/whZ5xxttNhpe388y8CoGNlg8ORiN5mWRbtK+sBe7wz2yQpdOOEE07iF7/4DWVl5URrPyK0dQHJeP7235ruAIa3aO9j3iJMd8ChiDLLSkQJb19IZNcKiouL+dn/+1XOTj/dn9NPP4MjjxxDdFu7dCPlmfC6ZhJNUSZOPJPRo7NfqFOSwn6MHHkkv//9nzj++BNItNfQsXE2sdb87U4KDD0DsOdFm94i/EPPcDagDIm31dC+4RXirdtQ6lj+8Ic/M2bM0U6H1WOmafKNb1yHy+2mbcku4i39o1Bcvovu7KBjZQOFhYV8+ctfcyQGSQoHUFRUzI9+9DOuueZruIgT3vY2oe2L8rLV4PKXYngC4A5QMHoyLn+p0yH1KiseIVz9LqGtb2JaUa666kv89Ke/pKSk7/47jzhiBN/+9//AiiVpfWcHifb8G1/Y357T+bgXdbwhTNu7u3C73Nx880+6ph5nW/5u6tpLTNPkkksmc8IJ43jgwXvYtHEDifYavFUn4CkdnXdVKvPt32NZFvHmjUR2fYCViDB8+BFcd92NOT/DKF0TJ55JfX0dzz//NC1vVlM0aRDu0tydTttTpt+NWegh2bY74ZmFHkx/fv3pitZ00PbeLkhafOf6Gx3dx8Po64OotbWtWfsHJJNJ5s+fx3PPTSMcDmH6yvANPAl3wYBshZBRbeteAqBwzBUOR9I7Eh11hHcuJxmux+v18fnPf5ELLrgYtzu//qAAzJ07m6lTHwOXQcEplfiG5s+ezfHmKM2vbwPLTghFEwbmTTFAy7IIr2+h46N6PG4PN9xwMyedND4rr11VVdTtHaAkhUPQ1NTIM888xaJFbwPgLhqGr+oETF/f3gUrX5JCMtpGpPZD4qnCduPHT+ArX/k65eUVDkeWWUuWvMeUKXcRi8XwjSyi4IQKDHd+9BA3zt6CZVmUX5ofLTyAZDhO29JaYjtDFBYV8YObf5LV8S1JChmwYcM6pk59jHXr1gIGntIj8VYeh+nJzZo5B9PXk0IyHiZat4pY03qwkowceSRf+crXuxZ79QfV1du577472bp1M65CDwXjKvEM6PuzyBpn2wm+7JIjHI7k8FmWRXRrGx0fNZCMJDj++BO47robsj6+JUkhQyzLYunS93n++Wns2FEDhgtP2Ri8FcdiuvvWJih9NSlY8QjRhjXEGtdiJeNUVg3g6qu+xGmnTcQ08+NOuSdisRjPP/80r746C8uy8A4tIDi2Alew73ab5UtSiDdHaf+gjnhdGLfHwxe/8BUuuOBiR96nkhQyLJFI8M47b/Hii8/R2FiPYbpxl47BW3FMn0kOfS0pWIkI0XqdSgYxiotLuPLKqznrrPPyctygpzZv3shjjz3Mhg3rMNwm/jHF+I8qxfT0vUTZ15NCMhSnQzcR2dgCFowbdwrXXvtvVFU5Nx4pSSFLYrEYb731Bi+//CJNTY2p5DAab/kxmJ7cbsb3laSQjIeJNWhijeu6ksFll13Bueeej9ebHwOQvSWZTLJw4QKefXYqra0tmD4X/qNL8B9ZjOHqO8mhryaFZDRB6JNmIuubsRIWVVUDuPbabzBu3MlOhyZJIdtisSgLFsxn1qwZNDY2gGHiKR1ttxw8BU6H161cTwrJWAfRBk28cR2WlaC4uIRLL/0c5557QU5XNc0F4XCYuXNn88orM+yZcwE3/qNK8I8s6hOD0X0tKSQjCcLrmwmvb8GKJSkpLeXKK77AmWeekzOtWEkKDonFYixcuICZM6dTX18HhoG7eCS+imNzbrZSriaFZLSVaP0aYs0bwUpSVlbOZZddzllnnSctgx5qa2vjlVdm8Nprc4hGo5heF74xxfiPLMb05m6tq/aP7FpABWNzewZZIhQnvLaZyMYWrIRFYWEhl112BZ/97EU5916VpOCweDzOu+++w8yZL7FjRzVgT2X1VnwGV6Dc4ehsuZYUEuEmovWribdsBeym9+TJVzJx4pl4PB6nw+vTWltbmDdvDvNem0OoowPDbeIbVYR/dEmfHpB2SrwlSnhtE5Gt7ZC0KC0t49JLP8fZZ382Z1uxkhRyRDKZZPnyJbz88nQ2b94IgKtgoJ0cggMcXVGcK0kh3lFLtP5jEm128hw6dDif+9yVjB8/AZcrd+9m+6JQqIP58+fx6quv0NLSDAZ4hxUSGFOCuyw3/5jlCsuyiNWGCK9tJrYzBMCAgYO47NLL+8SNiySFHGNZFqtXr2TmzOmsWbMaADNQga/iM7gKhziSHJxMCpZlkWjfQbR+NYmOWgDGjDmayZOv5IQTxuVd+Y1cE4vFWLx4Ia++Oovt2+3Cj+5KP4ExJXgGB+X/fw9WIklkazvh9c0kmu1ChEcdpbjkksmceOLJfWYatCSFHLZ+/TpmzZrO8uVLATB9JXgrPoO7eDiGkb03mBNJwbIs4q3biNavJhluBGDs2BOZPPnKfrXoLFdYlsWqVR8xZ85MVq36CAAz6MY/pgTfiKI+OZ21tyTDccIbWohsbCUZSWCaJuPHn8bFF0/O+paZvaFPJAWl1HDgLmASEAKmA/+ptd5v+cd8SAqdtm3byqxZL/Hee4tIJpOY3iK8FcfiLhmZleSQzaRgWUniLVuI1q0mGW3BMAxOOeU0Jk++khEjRmb89cXBbd++jddem8PChQuIxWL2uMOIQnvcoTC3u0Z6U7wpQnhdM5Ft9nhBMBjk3HMv4LOfvbBPl07pK0nhfWAV8COgBHgBeFVr/Yv9PSefkkKnXbt2MmvWSyxcuIBEIoHpKcBbeVzGk0N45woA/APHZew1dieDVSSjrZimycSJZ3LZZVcwePCQjL2uOHRtba28+ebrvPbaqzQ12a05z+CgPe5Q6c/LrqXOPZJD65qJ19ml8gcNGsKFF17CpEln4vP1jQWpB5LzSUEpNR54Fxiota5LHfsicD9QqbVOdve8fEwKnRoa6pk16yXefHM+iUQcw1OALwvJIRMsyyLesnmvZHDmmecwefKVjq7qFOmLx+MsWfIec+fOYuPGDQC4Sr0EjirFO7QAw+z7ycGKJ4lsbiW0roVkan+K444by0UXXcZxx43tM+MF6egLSeF64L+01iP3OHYksB44Wmu9trvn5XNS6NTQUM8rr8zgjTdeJ5GIY3qL8VaNxV00LOfv0izLItFWTaT2Q5KRZkzT5KyzzmXy5CuprKxyOjxxCCzLYv36tcyZM4tly97HsqyucYe+shhuX52LzSIbWklGE7jdbiZNOosLL7yUoUOHOR1eRvSFpPBL4Eta65P2OFYKNAKTtNaLuntePJ6w3O7+MU2xtraWadOmMXfuXHvMwV+Ob+A43MHcvNNOdNQR3rWCZKgOwzA477zzuPbaaxk0aJDToYleUlNTw/Tp05k7d+7uxXCji/GPzu3FcJ0SoTjhT5qIbGrtWmw2efJkJk+eTFlZmdPhZVqfTAplQAMwUWu9uLvn9YeWwr527qzhhRee4b337P8Sd+FQfAPHYXqLHI7Mtu9+BiedNJ6rr74mb++4hL0Y7vXX5zJv3mza29sxPCb+I4vxjynB9OVecki0xwh90kRkcxskLcrKy7n0kss566xz8mK8IB19oaXwXeC3WuvhexwbDawDxmit13f3vP6YFDqtX7+OadMeZ926T8Aw8ZYfg7fyMximMytSrWSCaP3HxOo/xrISjBo1mi9/+WsytbQfCYVCvPHGPGbPnklra4tdnfWoEgJHleREt1IyHCe0ponwplZI7l4lP2nSWTlTkyhb+kJSOAlYCgzVWtekjl0L3AYM0lp3G2h/TgrQuZ/Dezz11ON2yW5PAf5Bp+AuzO5Mnnj7DiI7lpCMtlFSUso113yVCRMm5dXAnEhfJBLhzTdf5+WZL9LW2mpXZz2mFP+oYkcGpK1YktDaJsLrWrDiSSqrBvD5K7/AhAmT+u0q+ZxPCgBKqXewWwY3AxXADGCq1vp/9vec/p4UOkUiYWbMeJE5c2aSSCRwl4zEP/AkDFdmSxVYiRiRXcuJNW3ANE0uuOBirrzyCwQCfXP3OdG7QqEQr746i9mzXyYSieAq8lBwYvZ2g7Msi+i2djo+qicZTlBcXMwVV3yBs8+WPTf6SlIYAtwNnAm0A9OwZyQl9vccSQp727ZtCw8+eB+bN2/EcAfwDzkdd8HAjLxWoqOOcPUikrF2hg4dznXXXc/IkUdm5LVE39bS0syLLz7Lm2++bu8GN6yAgrEVmIHM/WGOt6R2Oau1dzm77NLLueSSz+H3948xg4PpE0nhUEhS+LREIsGsWTOYPv1Zkskk3srj8FYe12trGyzLIlr/MdHajzAMmDz5Sq644up+f+clDm7Tpg089tjDbNy4HsPromBcBb5hhb36GpZlEV7fQmhlA1bS4sQTT+Laa7/BgAGZuTnqqyQp9EPr1n3CvffdSUN9Ha7CIQSGTMRwHV55AisZJ1z9LvHWrZSUlHHDDd9HqWN7KWLRHySTSd544zWmTXucWCyG94hCCk6s7JW6SolQnPaltcR2hSgsKuJb3/wOJ598ai9EnX8kKfRT7e1t3HPP7axevRLTV0Jg+LmHvC1oMh4mtHUByXADRx2l+N73/pPi4tzaKEj0HTU11UyZchebNm3EVeShaNIgXAWHftMSawjTtngnyXCCE088iW996z8oKSnpxYjziySFfiyRSPDkk/9i/vy5mN5CAkec1+MtQZOxEKEt80lGWzjjjLP55je/I91F4rDF43GefvoJ5s2bg+lzUXj6QDwVPe/zj2xro31pLSThmmu+ykUXXZbzq/2dJkmhn7MsixdffJYZM17A9BQQGHkBpju9FoMVj9Cx+TWS0RYuuuhSvvzlr8svnOhVb7zxGo8//jCWYVF0xiA8lem3ZiObW2lbWovP7+fGG27mhBNOOviTxH6Tgkwi7ycMw+Cqq77EFVdcTTLWTmjLAqzkfiuSd7GSCTq2LUglhMskIYiMOPfc8/n+93+MiUnrOzuJ1YfTel5kWxtty2oJBIP84ue/lYTQCyQp9DNXXmnP0U5GGgnXLOFgLcXIzmUkQ/VMmDCJa675qiQEkTHjxp3MDTf8ACMJbYt3kgjFD3h+vDFC25Ja/D4/P/3Jf8k+HL1EkkI/YxgGX//6v3PkkWOIt2wm3rxxv+fGWrYSa1rP8OFH8O///h+yOllk3CmnnMpXvvJvJCMJ2t7diZXs/qYlGU3Q+u5ODAtuvPGHfXLns1wlv+X9kNvt5oYbbsbn8xHZtYJk/NNNdSsRJbJzaercH+D1eh2IVPRH559/EaeddjrxBnvHs+50rGwg2RHn8suvYuzYE7McYX6TpNBPVVZWcdVV12AlokRrV37q+9H61VjxMJdffpXsiCayym7NfpuCgkJCa5roWNVAx+rdH+0f1RPZ1MrQocO4/PKrnA4370hS6MfOP/8iqqoGEGveQDLW3nU8GQ8Ta1xHaWkZl1wy2cEIRX9VWFjIl750LVY8SUg3EVqz+yO8trmrG7S/FrPLJJlo3o+5XC4+97nP8/DD9xNrXIdvgN0MjzWtx0rGufTSy/F4pNtIOOOss87liCNGEIlEPvW9wsIi2Z8jQyQp9HMTJkxi6tTHCTdvxFs1FjCINW3A6/VyxhlnOx2e6McMw5ACiw6Q7qN+zuv1cuqpE7DiYSK7PiBa+yFWrJ2TTx5PMCjlr4XobyQpCE47bSIAsQZNtP5jAE499XQnQxJCOETKXAgA1q7VdHR0AOD3+zn66GNkoZoQeWx/ZS5kTEEAcNRRyukQhBA5QLqPhBBCdJGkIIQQooskBSGEEF0kKQghhOgiSUEIIUQXSQpCCCG6SFIQQgjRpc8vXhNCCNF7pKUghBCiiyQFIYQQXSQpCCGE6CJJQQghRBdJCkIIIbpIUhBCCNFFkoIQQogukhTEpyilfq2UWuh0HEKI7JPFa0KInKaUGg7cBUwCQsB04D+11jFHA8tT0lIQQuS654EGYAxwJnZy+B9HI8pj0lLoB5RSFvCT1McUrfXvlVJfAX4FjAIagXu01n9Knf974HNa6/FKqXOBl4AvArcDw4A3gWu11i3Z/reI/kUpNR54Fxiota5LHfsicD9QqbVOOhlfPpKWQv/xBeAU4A9KqZHA48AvtNaFwNXA75RSF+7nuQXA14DTgeOACcC3Mh2wENjv2a2dCSFlGVAGjHYmpPwmSaH/eFprvUNrbWmtNwFVWuuZAFrr9wENjN/Pc03gVq11k9Z6M/AecGw2ghb9XgV2S3ZPDanPlVmOpV9wOx2AyJrN+zy+QSl1HTAUMAAv4DvA8zfu8XUHEOjd8IRIm5H6LH3fGSAthf4j3vlFKhn8CrgJKNJa+4EPDvJ86bsVTqjl0y2C8j2+J3qZJIX+6XY5cekAAARuSURBVDRgodb6Va11XClVjD2zQ4hcswQYqpQavMex04BdwAZnQspv0n3UP20ALlJKVWB3A/0N2IrdlSREztBaL1dKLQb+opS6GXuM4dfAnVpr6T7KAEkK/dO9wDnY4wzbgR8DVcBdSqldgCwKErnki8DdwHqgHZgG/MnRiPKYrFMQQgjRRcYUhBBCdJGkIIQQooskBSGEEF0kKQghhOgiSUEIIUQXSQpCCCG6SFIQ4jAopbRS6gan4xCit8g6BSGyQCn1a+yVuLC7AGGU3UXdvqu1fsyJ2ITYkyQFIbJMKXU88BEwKlXGXIicIWUuRN5JbSK0Ebs8wh+wN2NZBlwDHAXMBn4K/BH4vNZ6vlLqRuBmYCR2Haj/1FrPSl0vCNyauh7ADOD7Wut2pdQm4G9a6zuVUm8Ai4ARwJVAE/BjrfW0NGIOAjuAb2qtX9jj+EupeN6H/9/enYTqGMVxHP9yNzIURYbCwuIvmVYsbA0ZFkoWlKnobiTkSnIjiSxk2GCBXJFhQSJTpiwUysYtv5tknoroZoxYnOd979Plvfe5KfT6fVbPOc/bGTbPec85z3P+rM36s4F0UugRoFbSl+y3FftgVpT3FKyaLQEmA/1JZ+bsyfJrgOFZ/pWImE560M4HepDClh6PiFIgoY3AKGAYEMBQYHOFOmuBw6TIYOuBgxHR7kGDkj4Ax4A5pbyI6A5MIEXJA+hHilEcwOjs3vLst+31wawQDwpWzXZJeizpLekk2PFAV9IMeaekD9lJm4uAfZJuSPom6RRwDpgbEZ2AecAWSa8kvQYWAicq1HlT0klJXyTtBl4C0wq2dz8wJSJ6ZempwFNJ17N0F6Be0ntJTUADaUZCW30oWLcZ4OUjq27KXT8kzRD65tIlQ0hHiS/O5XUG3pGOau4JPCgXKjUCjQXqBHgEDCjY3mvAM2AmKTD9DFpmCQDNkp7n0g9zZbfVB7PCPChYNavJXbcO4fg1d+8jsEbST0tCWcwJKD6rrmmV7kTBsJGSvkdEAzA7Ig6Qlr5WFyy7Yh/MOsLLR1bNhuSuB5MGghe/+N09YGQ+IyIGRUTnbLnoLWkdv3RvRBbStL06AQYBTzrQ5gZgHGlvoFHSvdy9rhHRL5cenCu7Yh86ULeZZwpW1Woj4irwibQhey67bm0ncCYijgKnSeEeT5GWcS4C+4C6iLhM2rDeAdylZeM6b2xETAIukx7sfbKyCpF0PyKuA5to+a6h5DNQHxF1wEDSpvS2gn0wK8T/IqyaHQAuAM+B7qTN2J9IugQsBbYCzcBeoE5S6WG6irTefwdoIr3uurJCnYeABcAbYB0wS9KvZidt2Q90I73FlNdMerW2CbgNnAe2F+yDWSH+eM2qTu47hRGS7vzBeq8AtySt+M1y6oGRkmbm8uaTvofo/VuNNGuHl4/M/iERMQZYBkz8222x/5MHBbN/REScJX0kt1LSrb/dHvs/efnIzMzKvNFsZmZlHhTMzKzMg4KZmZV5UDAzszIPCmZmVvYDL+KQJ0pq9EwAAAAASUVORK5CYII=\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "sns.violinplot(x='precipType',\n",
    "               y='tweets_per_day',\n",
    "               data=twitter_moves_subset[twitter_moves_subset['precipType'] != 'snow'])"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "While there is a slight trend it overall seems that Twitter isn't just for rainy days after all but is also happily used on sunny days. ☀️"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.6.5"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}