wrangle_act.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 67,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Requirement already satisfied: tweepy in c:\\users\\marc-aurelien\\anaconda3\\lib\\site-packages (3.9.0)\n",
      "Requirement already satisfied: six>=1.10.0 in c:\\users\\marc-aurelien\\anaconda3\\lib\\site-packages (from tweepy) (1.15.0)\n",
      "Requirement already satisfied: requests[socks]>=2.11.1 in c:\\users\\marc-aurelien\\anaconda3\\lib\\site-packages (from tweepy) (2.24.0)\n",
      "Requirement already satisfied: requests-oauthlib>=0.7.0 in c:\\users\\marc-aurelien\\anaconda3\\lib\\site-packages (from tweepy) (1.3.0)\n",
      "Requirement already satisfied: chardet<4,>=3.0.2 in c:\\users\\marc-aurelien\\anaconda3\\lib\\site-packages (from requests[socks]>=2.11.1->tweepy) (3.0.4)\n",
      "Requirement already satisfied: idna<3,>=2.5 in c:\\users\\marc-aurelien\\anaconda3\\lib\\site-packages (from requests[socks]>=2.11.1->tweepy) (2.10)\n",
      "Requirement already satisfied: urllib3!=1.25.0,!=1.25.1,<1.26,>=1.21.1 in c:\\users\\marc-aurelien\\anaconda3\\lib\\site-packages (from requests[socks]>=2.11.1->tweepy) (1.25.9)\n",
      "Requirement already satisfied: certifi>=2017.4.17 in c:\\users\\marc-aurelien\\anaconda3\\lib\\site-packages (from requests[socks]>=2.11.1->tweepy) (2020.6.20)\n",
      "Requirement already satisfied: PySocks!=1.5.7,>=1.5.6; extra == \"socks\" in c:\\users\\marc-aurelien\\anaconda3\\lib\\site-packages (from requests[socks]>=2.11.1->tweepy) (1.7.1)\n",
      "Requirement already satisfied: oauthlib>=3.0.0 in c:\\users\\marc-aurelien\\anaconda3\\lib\\site-packages (from requests-oauthlib>=0.7.0->tweepy) (3.1.0)\n"
     ]
    }
   ],
   "source": [
    "!pip install tweepy"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 121,
   "metadata": {},
   "outputs": [],
   "source": [
    "#We need to get the necessary library to compute our data wrangling steps\n",
    "import pandas as pd\n",
    "import numpy as np\n",
    "import tweepy\n",
    "import json\n",
    "import requests\n",
    "import matplotlib.pyplot as plt\n",
    "%matplotlib inline\n",
    "import matplotlib.dates as mdates\n",
    "import seaborn as sns\n",
    "import datetime\n",
    "\n",
    "# Style and size of graph\n",
    "\n",
    "sns.set_style('darkgrid')\n",
    "sns.set(rc={'figure.figsize':(10,8)})"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Gathering data #"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 122,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>tweet_id</th>\n",
       "      <th>in_reply_to_status_id</th>\n",
       "      <th>in_reply_to_user_id</th>\n",
       "      <th>timestamp</th>\n",
       "      <th>source</th>\n",
       "      <th>text</th>\n",
       "      <th>retweeted_status_id</th>\n",
       "      <th>retweeted_status_user_id</th>\n",
       "      <th>retweeted_status_timestamp</th>\n",
       "      <th>expanded_urls</th>\n",
       "      <th>rating_numerator</th>\n",
       "      <th>rating_denominator</th>\n",
       "      <th>name</th>\n",
       "      <th>doggo</th>\n",
       "      <th>floofer</th>\n",
       "      <th>pupper</th>\n",
       "      <th>puppo</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>892420643555336193</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>2017-08-01 16:23:56 +0000</td>\n",
       "      <td>&lt;a href=\"http://twitter.com/download/iphone\" r...</td>\n",
       "      <td>This is Phineas. He's a mystical boy. Only eve...</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>https://twitter.com/dog_rates/status/892420643...</td>\n",
       "      <td>13</td>\n",
       "      <td>10</td>\n",
       "      <td>Phineas</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>892177421306343426</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>2017-08-01 00:17:27 +0000</td>\n",
       "      <td>&lt;a href=\"http://twitter.com/download/iphone\" r...</td>\n",
       "      <td>This is Tilly. She's just checking pup on you....</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>https://twitter.com/dog_rates/status/892177421...</td>\n",
       "      <td>13</td>\n",
       "      <td>10</td>\n",
       "      <td>Tilly</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>891815181378084864</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>2017-07-31 00:18:03 +0000</td>\n",
       "      <td>&lt;a href=\"http://twitter.com/download/iphone\" r...</td>\n",
       "      <td>This is Archie. He is a rare Norwegian Pouncin...</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>https://twitter.com/dog_rates/status/891815181...</td>\n",
       "      <td>12</td>\n",
       "      <td>10</td>\n",
       "      <td>Archie</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>891689557279858688</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>2017-07-30 15:58:51 +0000</td>\n",
       "      <td>&lt;a href=\"http://twitter.com/download/iphone\" r...</td>\n",
       "      <td>This is Darla. She commenced a snooze mid meal...</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>https://twitter.com/dog_rates/status/891689557...</td>\n",
       "      <td>13</td>\n",
       "      <td>10</td>\n",
       "      <td>Darla</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>891327558926688256</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>2017-07-29 16:00:24 +0000</td>\n",
       "      <td>&lt;a href=\"http://twitter.com/download/iphone\" r...</td>\n",
       "      <td>This is Franklin. He would like you to stop ca...</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>https://twitter.com/dog_rates/status/891327558...</td>\n",
       "      <td>12</td>\n",
       "      <td>10</td>\n",
       "      <td>Franklin</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "             tweet_id  in_reply_to_status_id  in_reply_to_user_id  \\\n",
       "0  892420643555336193                    NaN                  NaN   \n",
       "1  892177421306343426                    NaN                  NaN   \n",
       "2  891815181378084864                    NaN                  NaN   \n",
       "3  891689557279858688                    NaN                  NaN   \n",
       "4  891327558926688256                    NaN                  NaN   \n",
       "\n",
       "                   timestamp  \\\n",
       "0  2017-08-01 16:23:56 +0000   \n",
       "1  2017-08-01 00:17:27 +0000   \n",
       "2  2017-07-31 00:18:03 +0000   \n",
       "3  2017-07-30 15:58:51 +0000   \n",
       "4  2017-07-29 16:00:24 +0000   \n",
       "\n",
       "                                              source  \\\n",
       "0  <a href=\"http://twitter.com/download/iphone\" r...   \n",
       "1  <a href=\"http://twitter.com/download/iphone\" r...   \n",
       "2  <a href=\"http://twitter.com/download/iphone\" r...   \n",
       "3  <a href=\"http://twitter.com/download/iphone\" r...   \n",
       "4  <a href=\"http://twitter.com/download/iphone\" r...   \n",
       "\n",
       "                                                text  retweeted_status_id  \\\n",
       "0  This is Phineas. He's a mystical boy. Only eve...                  NaN   \n",
       "1  This is Tilly. She's just checking pup on you....                  NaN   \n",
       "2  This is Archie. He is a rare Norwegian Pouncin...                  NaN   \n",
       "3  This is Darla. She commenced a snooze mid meal...                  NaN   \n",
       "4  This is Franklin. He would like you to stop ca...                  NaN   \n",
       "\n",
       "   retweeted_status_user_id retweeted_status_timestamp  \\\n",
       "0                       NaN                        NaN   \n",
       "1                       NaN                        NaN   \n",
       "2                       NaN                        NaN   \n",
       "3                       NaN                        NaN   \n",
       "4                       NaN                        NaN   \n",
       "\n",
       "                                       expanded_urls  rating_numerator  \\\n",
       "0  https://twitter.com/dog_rates/status/892420643...                13   \n",
       "1  https://twitter.com/dog_rates/status/892177421...                13   \n",
       "2  https://twitter.com/dog_rates/status/891815181...                12   \n",
       "3  https://twitter.com/dog_rates/status/891689557...                13   \n",
       "4  https://twitter.com/dog_rates/status/891327558...                12   \n",
       "\n",
       "   rating_denominator      name doggo floofer pupper puppo  \n",
       "0                  10   Phineas  None    None   None  None  \n",
       "1                  10     Tilly  None    None   None  None  \n",
       "2                  10    Archie  None    None   None  None  \n",
       "3                  10     Darla  None    None   None  None  \n",
       "4                  10  Franklin  None    None   None  None  "
      ]
     },
     "execution_count": 122,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#Getting the original csv file\n",
    "df_twitter = pd.read_csv('twitter-archive-enhanced.csv')\n",
    "df_twitter.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 123,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>tweet_id</th>\n",
       "      <th>jpg_url</th>\n",
       "      <th>img_num</th>\n",
       "      <th>p1</th>\n",
       "      <th>p1_conf</th>\n",
       "      <th>p1_dog</th>\n",
       "      <th>p2</th>\n",
       "      <th>p2_conf</th>\n",
       "      <th>p2_dog</th>\n",
       "      <th>p3</th>\n",
       "      <th>p3_conf</th>\n",
       "      <th>p3_dog</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>666020888022790149</td>\n",
       "      <td>https://pbs.twimg.com/media/CT4udn0WwAA0aMy.jpg</td>\n",
       "      <td>1</td>\n",
       "      <td>Welsh_springer_spaniel</td>\n",
       "      <td>0.465074</td>\n",
       "      <td>True</td>\n",
       "      <td>collie</td>\n",
       "      <td>0.156665</td>\n",
       "      <td>True</td>\n",
       "      <td>Shetland_sheepdog</td>\n",
       "      <td>0.061428</td>\n",
       "      <td>True</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>666029285002620928</td>\n",
       "      <td>https://pbs.twimg.com/media/CT42GRgUYAA5iDo.jpg</td>\n",
       "      <td>1</td>\n",
       "      <td>redbone</td>\n",
       "      <td>0.506826</td>\n",
       "      <td>True</td>\n",
       "      <td>miniature_pinscher</td>\n",
       "      <td>0.074192</td>\n",
       "      <td>True</td>\n",
       "      <td>Rhodesian_ridgeback</td>\n",
       "      <td>0.072010</td>\n",
       "      <td>True</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>666033412701032449</td>\n",
       "      <td>https://pbs.twimg.com/media/CT4521TWwAEvMyu.jpg</td>\n",
       "      <td>1</td>\n",
       "      <td>German_shepherd</td>\n",
       "      <td>0.596461</td>\n",
       "      <td>True</td>\n",
       "      <td>malinois</td>\n",
       "      <td>0.138584</td>\n",
       "      <td>True</td>\n",
       "      <td>bloodhound</td>\n",
       "      <td>0.116197</td>\n",
       "      <td>True</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>666044226329800704</td>\n",
       "      <td>https://pbs.twimg.com/media/CT5Dr8HUEAA-lEu.jpg</td>\n",
       "      <td>1</td>\n",
       "      <td>Rhodesian_ridgeback</td>\n",
       "      <td>0.408143</td>\n",
       "      <td>True</td>\n",
       "      <td>redbone</td>\n",
       "      <td>0.360687</td>\n",
       "      <td>True</td>\n",
       "      <td>miniature_pinscher</td>\n",
       "      <td>0.222752</td>\n",
       "      <td>True</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>666049248165822465</td>\n",
       "      <td>https://pbs.twimg.com/media/CT5IQmsXIAAKY4A.jpg</td>\n",
       "      <td>1</td>\n",
       "      <td>miniature_pinscher</td>\n",
       "      <td>0.560311</td>\n",
       "      <td>True</td>\n",
       "      <td>Rottweiler</td>\n",
       "      <td>0.243682</td>\n",
       "      <td>True</td>\n",
       "      <td>Doberman</td>\n",
       "      <td>0.154629</td>\n",
       "      <td>True</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "             tweet_id                                          jpg_url  \\\n",
       "0  666020888022790149  https://pbs.twimg.com/media/CT4udn0WwAA0aMy.jpg   \n",
       "1  666029285002620928  https://pbs.twimg.com/media/CT42GRgUYAA5iDo.jpg   \n",
       "2  666033412701032449  https://pbs.twimg.com/media/CT4521TWwAEvMyu.jpg   \n",
       "3  666044226329800704  https://pbs.twimg.com/media/CT5Dr8HUEAA-lEu.jpg   \n",
       "4  666049248165822465  https://pbs.twimg.com/media/CT5IQmsXIAAKY4A.jpg   \n",
       "\n",
       "   img_num                      p1   p1_conf  p1_dog                  p2  \\\n",
       "0        1  Welsh_springer_spaniel  0.465074    True              collie   \n",
       "1        1                 redbone  0.506826    True  miniature_pinscher   \n",
       "2        1         German_shepherd  0.596461    True            malinois   \n",
       "3        1     Rhodesian_ridgeback  0.408143    True             redbone   \n",
       "4        1      miniature_pinscher  0.560311    True          Rottweiler   \n",
       "\n",
       "    p2_conf  p2_dog                   p3   p3_conf  p3_dog  \n",
       "0  0.156665    True    Shetland_sheepdog  0.061428    True  \n",
       "1  0.074192    True  Rhodesian_ridgeback  0.072010    True  \n",
       "2  0.138584    True           bloodhound  0.116197    True  \n",
       "3  0.360687    True   miniature_pinscher  0.222752    True  \n",
       "4  0.243682    True             Doberman  0.154629    True  "
      ]
     },
     "execution_count": 123,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Download tweet image predictions TSV using the Requests library and write it to image_predictions.tsv\n",
    "url = 'https://d17h27t6h515a5.cloudfront.net/topher/2017/August/599fd2ad_image-predictions/image-predictions.tsv'\n",
    "response = requests.get(url)\n",
    "with open('image_predictions.tsv', mode='wb') as file:\n",
    "    file.write(response.content)\n",
    "    \n",
    "# Import the tweet image predictions TSV file into a DataFrame\n",
    "df_image = pd.read_csv('image_predictions.tsv', sep='\\t')\n",
    "df_image.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 124,
   "metadata": {},
   "outputs": [],
   "source": [
    "#Documentation about twitter API:https://developer.twitter.com/en/docs/twitter-api/v1/data-dictionary/overview/tweet-object\n",
    "consumer_key = 'YOUR KEY'\n",
    "consumer_secret = 'YOUR KEY'\n",
    "access_token = 'YOUR KEY'\n",
    "access_secret = 'YOUR KEY'\n",
    "\n",
    "auth = tweepy.OAuthHandler(consumer_key, consumer_secret)\n",
    "auth.set_access_token(access_token, access_secret)\n",
    "\n",
    "api = tweepy.API(auth_handler=auth, wait_on_rate_limit=True, wait_on_rate_limit_notify=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 125,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Use Twitter API to collect our data \n",
    "#tweet_ids = list(df_twitter['tweet_id'])\n",
    "\n",
    "#tweet_data = []\n",
    "#tweet_id_success = []\n",
    "#tweet_id_missing = []\n",
    "#for tweet_id in tweet_ids:\n",
    "#    try:\n",
    "#        data = api.get_status(tweet_id, \n",
    "#                              tweet_mode='extended',\n",
    " #                             wait_on_rate_limit = True,\n",
    "  #                            wait_on_rate_limit_notify = True)\n",
    "   #     tweet_data.append(data._json)\n",
    "    #    tweet_id_success.append(tweet_id)\n",
    "    #except:\n",
    "     #   tweet_id_missing.append(tweet_id)\n",
    "      #  print(tweet_id)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 126,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Write tweet data to json file\n",
    "#with open('tweet_json.txt', mode = 'w') as file:\n",
    " #   json_content = tweet_data\n",
    "  #  json.dump(json_content, file)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 127,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>created_at</th>\n",
       "      <th>id</th>\n",
       "      <th>id_str</th>\n",
       "      <th>full_text</th>\n",
       "      <th>truncated</th>\n",
       "      <th>display_text_range</th>\n",
       "      <th>entities</th>\n",
       "      <th>extended_entities</th>\n",
       "      <th>source</th>\n",
       "      <th>in_reply_to_status_id</th>\n",
       "      <th>...</th>\n",
       "      <th>favorited</th>\n",
       "      <th>retweeted</th>\n",
       "      <th>possibly_sensitive</th>\n",
       "      <th>possibly_sensitive_appealable</th>\n",
       "      <th>lang</th>\n",
       "      <th>retweeted_status</th>\n",
       "      <th>quoted_status_id</th>\n",
       "      <th>quoted_status_id_str</th>\n",
       "      <th>quoted_status_permalink</th>\n",
       "      <th>quoted_status</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>2017-08-01 16:23:56+00:00</td>\n",
       "      <td>892420643555336193</td>\n",
       "      <td>892420643555336192</td>\n",
       "      <td>This is Phineas. He's a mystical boy. Only eve...</td>\n",
       "      <td>False</td>\n",
       "      <td>[0, 85]</td>\n",
       "      <td>{'hashtags': [], 'symbols': [], 'user_mentions...</td>\n",
       "      <td>{'media': [{'id': 892420639486877696, 'id_str'...</td>\n",
       "      <td>&lt;a href=\"http://twitter.com/download/iphone\" r...</td>\n",
       "      <td>NaN</td>\n",
       "      <td>...</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>en</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>2017-08-01 00:17:27+00:00</td>\n",
       "      <td>892177421306343426</td>\n",
       "      <td>892177421306343424</td>\n",
       "      <td>This is Tilly. She's just checking pup on you....</td>\n",
       "      <td>False</td>\n",
       "      <td>[0, 138]</td>\n",
       "      <td>{'hashtags': [], 'symbols': [], 'user_mentions...</td>\n",
       "      <td>{'media': [{'id': 892177413194625024, 'id_str'...</td>\n",
       "      <td>&lt;a href=\"http://twitter.com/download/iphone\" r...</td>\n",
       "      <td>NaN</td>\n",
       "      <td>...</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>en</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>2017-07-31 00:18:03+00:00</td>\n",
       "      <td>891815181378084864</td>\n",
       "      <td>891815181378084864</td>\n",
       "      <td>This is Archie. He is a rare Norwegian Pouncin...</td>\n",
       "      <td>False</td>\n",
       "      <td>[0, 121]</td>\n",
       "      <td>{'hashtags': [], 'symbols': [], 'user_mentions...</td>\n",
       "      <td>{'media': [{'id': 891815175371796480, 'id_str'...</td>\n",
       "      <td>&lt;a href=\"http://twitter.com/download/iphone\" r...</td>\n",
       "      <td>NaN</td>\n",
       "      <td>...</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>en</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>2017-07-30 15:58:51+00:00</td>\n",
       "      <td>891689557279858688</td>\n",
       "      <td>891689557279858688</td>\n",
       "      <td>This is Darla. She commenced a snooze mid meal...</td>\n",
       "      <td>False</td>\n",
       "      <td>[0, 79]</td>\n",
       "      <td>{'hashtags': [], 'symbols': [], 'user_mentions...</td>\n",
       "      <td>{'media': [{'id': 891689552724799489, 'id_str'...</td>\n",
       "      <td>&lt;a href=\"http://twitter.com/download/iphone\" r...</td>\n",
       "      <td>NaN</td>\n",
       "      <td>...</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>en</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>2017-07-29 16:00:24+00:00</td>\n",
       "      <td>891327558926688256</td>\n",
       "      <td>891327558926688256</td>\n",
       "      <td>This is Franklin. He would like you to stop ca...</td>\n",
       "      <td>False</td>\n",
       "      <td>[0, 138]</td>\n",
       "      <td>{'hashtags': [{'text': 'BarkWeek', 'indices': ...</td>\n",
       "      <td>{'media': [{'id': 891327551943041024, 'id_str'...</td>\n",
       "      <td>&lt;a href=\"http://twitter.com/download/iphone\" r...</td>\n",
       "      <td>NaN</td>\n",
       "      <td>...</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>en</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>5 rows × 32 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "                 created_at                  id              id_str  \\\n",
       "0 2017-08-01 16:23:56+00:00  892420643555336193  892420643555336192   \n",
       "1 2017-08-01 00:17:27+00:00  892177421306343426  892177421306343424   \n",
       "2 2017-07-31 00:18:03+00:00  891815181378084864  891815181378084864   \n",
       "3 2017-07-30 15:58:51+00:00  891689557279858688  891689557279858688   \n",
       "4 2017-07-29 16:00:24+00:00  891327558926688256  891327558926688256   \n",
       "\n",
       "                                           full_text  truncated  \\\n",
       "0  This is Phineas. He's a mystical boy. Only eve...      False   \n",
       "1  This is Tilly. She's just checking pup on you....      False   \n",
       "2  This is Archie. He is a rare Norwegian Pouncin...      False   \n",
       "3  This is Darla. She commenced a snooze mid meal...      False   \n",
       "4  This is Franklin. He would like you to stop ca...      False   \n",
       "\n",
       "  display_text_range                                           entities  \\\n",
       "0            [0, 85]  {'hashtags': [], 'symbols': [], 'user_mentions...   \n",
       "1           [0, 138]  {'hashtags': [], 'symbols': [], 'user_mentions...   \n",
       "2           [0, 121]  {'hashtags': [], 'symbols': [], 'user_mentions...   \n",
       "3            [0, 79]  {'hashtags': [], 'symbols': [], 'user_mentions...   \n",
       "4           [0, 138]  {'hashtags': [{'text': 'BarkWeek', 'indices': ...   \n",
       "\n",
       "                                   extended_entities  \\\n",
       "0  {'media': [{'id': 892420639486877696, 'id_str'...   \n",
       "1  {'media': [{'id': 892177413194625024, 'id_str'...   \n",
       "2  {'media': [{'id': 891815175371796480, 'id_str'...   \n",
       "3  {'media': [{'id': 891689552724799489, 'id_str'...   \n",
       "4  {'media': [{'id': 891327551943041024, 'id_str'...   \n",
       "\n",
       "                                              source  in_reply_to_status_id  \\\n",
       "0  <a href=\"http://twitter.com/download/iphone\" r...                    NaN   \n",
       "1  <a href=\"http://twitter.com/download/iphone\" r...                    NaN   \n",
       "2  <a href=\"http://twitter.com/download/iphone\" r...                    NaN   \n",
       "3  <a href=\"http://twitter.com/download/iphone\" r...                    NaN   \n",
       "4  <a href=\"http://twitter.com/download/iphone\" r...                    NaN   \n",
       "\n",
       "   ...  favorited  retweeted  possibly_sensitive  \\\n",
       "0  ...      False      False                 0.0   \n",
       "1  ...      False      False                 0.0   \n",
       "2  ...      False      False                 0.0   \n",
       "3  ...      False      False                 0.0   \n",
       "4  ...      False      False                 0.0   \n",
       "\n",
       "  possibly_sensitive_appealable lang  retweeted_status  quoted_status_id  \\\n",
       "0                           0.0   en               NaN               NaN   \n",
       "1                           0.0   en               NaN               NaN   \n",
       "2                           0.0   en               NaN               NaN   \n",
       "3                           0.0   en               NaN               NaN   \n",
       "4                           0.0   en               NaN               NaN   \n",
       "\n",
       "  quoted_status_id_str  quoted_status_permalink  quoted_status  \n",
       "0                  NaN                      NaN            NaN  \n",
       "1                  NaN                      NaN            NaN  \n",
       "2                  NaN                      NaN            NaN  \n",
       "3                  NaN                      NaN            NaN  \n",
       "4                  NaN                      NaN            NaN  \n",
       "\n",
       "[5 rows x 32 columns]"
      ]
     },
     "execution_count": 127,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_api = pd.read_json('tweet_json.txt')\n",
    "df_api.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 128,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>id</th>\n",
       "      <th>favorite_count</th>\n",
       "      <th>retweet_count</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>892420643555336193</td>\n",
       "      <td>35686</td>\n",
       "      <td>7547</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>892177421306343426</td>\n",
       "      <td>30869</td>\n",
       "      <td>5595</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>891815181378084864</td>\n",
       "      <td>23206</td>\n",
       "      <td>3707</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>891689557279858688</td>\n",
       "      <td>38988</td>\n",
       "      <td>7728</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>891327558926688256</td>\n",
       "      <td>37239</td>\n",
       "      <td>8329</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                   id  favorite_count  retweet_count\n",
       "0  892420643555336193           35686           7547\n",
       "1  892177421306343426           30869           5595\n",
       "2  891815181378084864           23206           3707\n",
       "3  891689557279858688           38988           7728\n",
       "4  891327558926688256           37239           8329"
      ]
     },
     "execution_count": 128,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#Setting our columns of interest for the API data\n",
    "df_api = df_api[['id', 'favorite_count', 'retweet_count']]\n",
    "df_api.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 129,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(2331, 3)"
      ]
     },
     "execution_count": 129,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_api.shape"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Assess #"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 130,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "<class 'pandas.core.frame.DataFrame'>\n",
      "RangeIndex: 2356 entries, 0 to 2355\n",
      "Data columns (total 17 columns):\n",
      " #   Column                      Non-Null Count  Dtype  \n",
      "---  ------                      --------------  -----  \n",
      " 0   tweet_id                    2356 non-null   int64  \n",
      " 1   in_reply_to_status_id       78 non-null     float64\n",
      " 2   in_reply_to_user_id         78 non-null     float64\n",
      " 3   timestamp                   2356 non-null   object \n",
      " 4   source                      2356 non-null   object \n",
      " 5   text                        2356 non-null   object \n",
      " 6   retweeted_status_id         181 non-null    float64\n",
      " 7   retweeted_status_user_id    181 non-null    float64\n",
      " 8   retweeted_status_timestamp  181 non-null    object \n",
      " 9   expanded_urls               2297 non-null   object \n",
      " 10  rating_numerator            2356 non-null   int64  \n",
      " 11  rating_denominator          2356 non-null   int64  \n",
      " 12  name                        2356 non-null   object \n",
      " 13  doggo                       2356 non-null   object \n",
      " 14  floofer                     2356 non-null   object \n",
      " 15  pupper                      2356 non-null   object \n",
      " 16  puppo                       2356 non-null   object \n",
      "dtypes: float64(4), int64(3), object(10)\n",
      "memory usage: 313.0+ KB\n"
     ]
    }
   ],
   "source": [
    "df_twitter.info()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**df_twitter (Twitter archive) columns:**\n",
    "\n",
    "- tweet_id: the unique identifier for each tweet\n",
    "- in_reply_to_status_id: if the represented Tweet is a reply, this field will contain the integer representation of the original Tweet’s ID\n",
    "- in_reply_to_user_id: if the represented Tweet is a reply, this field will contain the integer representation of the original Tweet’s author ID\n",
    "- timestamp: time when this Tweet was created\n",
    "- source: utility used to post the Tweet, as an HTML-formatted string. e.g. Twitter for Android, Twitter for iPhone, Twitter Web Client\n",
    "- text: actual UTF-8 text of the status update\n",
    "- retweeted_status_id: if the represented Tweet is a retweet, this field will contain the integer representation of the original Tweet’s ID\n",
    "- retweeted_status_user_id: if the represented Tweet is a retweet, this field will contain the integer representation of the original Tweet’s author ID\n",
    "- retweeted_status_timestamp: time of retweet\n",
    "- expanded_urls: tweet URL\n",
    "- rating_numerator: numerator of the rating of a dog. Note: ratings almost always greater than 10\n",
    "- rating_denominator: denominator of the rating of a dog. Note: ratings almost always have a denominator of 10\n",
    "- name: name of the dog\n",
    "- doggo: one of the 4 dog \"stage\"\n",
    "- floofer: one of the 4 dog \"stage\"\n",
    "- pupper: one of the 4 dog \"stage\"\n",
    "- puppo: one of the 4 dog \"stage\""
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 131,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(2356, 17)"
      ]
     },
     "execution_count": 131,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_twitter.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 132,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>tweet_id</th>\n",
       "      <th>in_reply_to_status_id</th>\n",
       "      <th>in_reply_to_user_id</th>\n",
       "      <th>timestamp</th>\n",
       "      <th>source</th>\n",
       "      <th>text</th>\n",
       "      <th>retweeted_status_id</th>\n",
       "      <th>retweeted_status_user_id</th>\n",
       "      <th>retweeted_status_timestamp</th>\n",
       "      <th>expanded_urls</th>\n",
       "      <th>rating_numerator</th>\n",
       "      <th>rating_denominator</th>\n",
       "      <th>name</th>\n",
       "      <th>doggo</th>\n",
       "      <th>floofer</th>\n",
       "      <th>pupper</th>\n",
       "      <th>puppo</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>41</th>\n",
       "      <td>884441805382717440</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>2017-07-10 15:58:53 +0000</td>\n",
       "      <td>&lt;a href=\"http://twitter.com/download/iphone\" r...</td>\n",
       "      <td>I present to you, Pup in Hat. Pup in Hat is gr...</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>https://twitter.com/dog_rates/status/884441805...</td>\n",
       "      <td>14</td>\n",
       "      <td>10</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>902</th>\n",
       "      <td>758467244762497024</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>2016-07-28 01:00:57 +0000</td>\n",
       "      <td>&lt;a href=\"http://twitter.com/download/iphone\" r...</td>\n",
       "      <td>Why does this never happen at my front door......</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>https://twitter.com/dog_rates/status/758467244...</td>\n",
       "      <td>165</td>\n",
       "      <td>150</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1870</th>\n",
       "      <td>675149409102012420</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>2015-12-11 03:05:37 +0000</td>\n",
       "      <td>&lt;a href=\"http://twitter.com/download/iphone\" r...</td>\n",
       "      <td>holy shit 12/10 https://t.co/p6O8X93bTQ</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>https://twitter.com/dog_rates/status/675149409...</td>\n",
       "      <td>12</td>\n",
       "      <td>10</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>766</th>\n",
       "      <td>777684233540206592</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>2016-09-19 01:42:24 +0000</td>\n",
       "      <td>&lt;a href=\"http://twitter.com/download/iphone\" r...</td>\n",
       "      <td>\"Yep... just as I suspected. You're not flossi...</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>https://twitter.com/dog_rates/status/777684233...</td>\n",
       "      <td>12</td>\n",
       "      <td>10</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2266</th>\n",
       "      <td>667530908589760512</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>2015-11-20 02:32:25 +0000</td>\n",
       "      <td>&lt;a href=\"http://twitter.com\" rel=\"nofollow\"&gt;Tw...</td>\n",
       "      <td>Meet Naphaniel. He doesn't necessarily enjoy h...</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>https://twitter.com/dog_rates/status/667530908...</td>\n",
       "      <td>10</td>\n",
       "      <td>10</td>\n",
       "      <td>Naphaniel</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                tweet_id  in_reply_to_status_id  in_reply_to_user_id  \\\n",
       "41    884441805382717440                    NaN                  NaN   \n",
       "902   758467244762497024                    NaN                  NaN   \n",
       "1870  675149409102012420                    NaN                  NaN   \n",
       "766   777684233540206592                    NaN                  NaN   \n",
       "2266  667530908589760512                    NaN                  NaN   \n",
       "\n",
       "                      timestamp  \\\n",
       "41    2017-07-10 15:58:53 +0000   \n",
       "902   2016-07-28 01:00:57 +0000   \n",
       "1870  2015-12-11 03:05:37 +0000   \n",
       "766   2016-09-19 01:42:24 +0000   \n",
       "2266  2015-11-20 02:32:25 +0000   \n",
       "\n",
       "                                                 source  \\\n",
       "41    <a href=\"http://twitter.com/download/iphone\" r...   \n",
       "902   <a href=\"http://twitter.com/download/iphone\" r...   \n",
       "1870  <a href=\"http://twitter.com/download/iphone\" r...   \n",
       "766   <a href=\"http://twitter.com/download/iphone\" r...   \n",
       "2266  <a href=\"http://twitter.com\" rel=\"nofollow\">Tw...   \n",
       "\n",
       "                                                   text  retweeted_status_id  \\\n",
       "41    I present to you, Pup in Hat. Pup in Hat is gr...                  NaN   \n",
       "902   Why does this never happen at my front door......                  NaN   \n",
       "1870            holy shit 12/10 https://t.co/p6O8X93bTQ                  NaN   \n",
       "766   \"Yep... just as I suspected. You're not flossi...                  NaN   \n",
       "2266  Meet Naphaniel. He doesn't necessarily enjoy h...                  NaN   \n",
       "\n",
       "      retweeted_status_user_id retweeted_status_timestamp  \\\n",
       "41                         NaN                        NaN   \n",
       "902                        NaN                        NaN   \n",
       "1870                       NaN                        NaN   \n",
       "766                        NaN                        NaN   \n",
       "2266                       NaN                        NaN   \n",
       "\n",
       "                                          expanded_urls  rating_numerator  \\\n",
       "41    https://twitter.com/dog_rates/status/884441805...                14   \n",
       "902   https://twitter.com/dog_rates/status/758467244...               165   \n",
       "1870  https://twitter.com/dog_rates/status/675149409...                12   \n",
       "766   https://twitter.com/dog_rates/status/777684233...                12   \n",
       "2266  https://twitter.com/dog_rates/status/667530908...                10   \n",
       "\n",
       "      rating_denominator       name doggo floofer pupper puppo  \n",
       "41                    10       None  None    None   None  None  \n",
       "902                  150       None  None    None   None  None  \n",
       "1870                  10       None  None    None   None  None  \n",
       "766                   10       None  None    None   None  None  \n",
       "2266                  10  Naphaniel  None    None   None  None  "
      ]
     },
     "execution_count": 132,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_twitter.sample(5)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 133,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "0\n",
      "0\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "(None, None)"
      ]
     },
     "execution_count": 133,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#Checking the numbers of duplicated rows\n",
    "print(sum(df_twitter.duplicated())),print(sum(df_twitter.tweet_id.duplicated()))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 134,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "tweet_id                      2356\n",
       "in_reply_to_status_id           77\n",
       "in_reply_to_user_id             31\n",
       "timestamp                     2356\n",
       "source                           4\n",
       "text                          2356\n",
       "retweeted_status_id            181\n",
       "retweeted_status_user_id        25\n",
       "retweeted_status_timestamp     181\n",
       "expanded_urls                 2218\n",
       "rating_numerator                40\n",
       "rating_denominator              18\n",
       "name                           957\n",
       "doggo                            2\n",
       "floofer                          2\n",
       "pupper                           2\n",
       "puppo                            2\n",
       "dtype: int64"
      ]
     },
     "execution_count": 134,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#Checking the number of unique values\n",
    "df_twitter.nunique()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 135,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "10     2333\n",
       "11        3\n",
       "50        3\n",
       "80        2\n",
       "20        2\n",
       "2         1\n",
       "16        1\n",
       "40        1\n",
       "70        1\n",
       "15        1\n",
       "90        1\n",
       "110       1\n",
       "120       1\n",
       "130       1\n",
       "150       1\n",
       "170       1\n",
       "7         1\n",
       "0         1\n",
       "Name: rating_denominator, dtype: int64"
      ]
     },
     "execution_count": 135,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#Let's look if there are any denominator value that are not 10\n",
    "df_twitter.rating_denominator.value_counts()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 136,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0         1\n",
       "2         1\n",
       "7         1\n",
       "10     2333\n",
       "11        3\n",
       "15        1\n",
       "16        1\n",
       "20        2\n",
       "40        1\n",
       "50        3\n",
       "70        1\n",
       "80        2\n",
       "90        1\n",
       "110       1\n",
       "120       1\n",
       "130       1\n",
       "150       1\n",
       "170       1\n",
       "Name: rating_denominator, dtype: int64"
      ]
     },
     "execution_count": 136,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Sort by rating_denominator values\n",
    "df_twitter.rating_denominator.value_counts().sort_index()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 137,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "181"
      ]
     },
     "execution_count": 137,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# How many of the observation are retweet. This data will need to be dropped later\n",
    "len(df_twitter[df_twitter.retweeted_status_id.isnull() == False])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 138,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(2075, 281)"
      ]
     },
     "execution_count": 138,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#Let's check how many values in df_twitter.tweet_id don't exist in df_image.tweet_id\n",
    "len(df_twitter[df_twitter.tweet_id.isin(df_image.tweet_id)]),len(df_twitter[~df_twitter.tweet_id.isin(df_image.tweet_id)])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 139,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "12"
      ]
     },
     "execution_count": 139,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#let's check if some entries have more than one dog stage\n",
    "len(df_twitter[(df_twitter['doggo'] == 'doggo' ) & (df_twitter['pupper'] == 'pupper')])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 140,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0"
      ]
     },
     "execution_count": 140,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "len(df_twitter[(df_twitter['puppo'] != 'None' ) & (df_twitter['floofer'] != 'None')])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 141,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "1"
      ]
     },
     "execution_count": 141,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "len(df_twitter[(df_twitter['doggo'] != 'None' ) & (df_twitter['floofer'] != 'None')])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 142,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "1"
      ]
     },
     "execution_count": 142,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "len(df_twitter[(df_twitter['doggo'] == 'doggo' ) & (df_twitter['puppo'] == 'puppo')])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 143,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0"
      ]
     },
     "execution_count": 143,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "len(df_twitter[(df_twitter['pupper'] != 'None' ) & (df_twitter['puppo'] == 'puppo')])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 144,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0"
      ]
     },
     "execution_count": 144,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "len(df_twitter[(df_twitter['pupper'] != 'None' ) & (df_twitter['floofer'] != 'None')])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 145,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(2075, 12)"
      ]
     },
     "execution_count": 145,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_image.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 146,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "<class 'pandas.core.frame.DataFrame'>\n",
      "RangeIndex: 2075 entries, 0 to 2074\n",
      "Data columns (total 12 columns):\n",
      " #   Column    Non-Null Count  Dtype  \n",
      "---  ------    --------------  -----  \n",
      " 0   tweet_id  2075 non-null   int64  \n",
      " 1   jpg_url   2075 non-null   object \n",
      " 2   img_num   2075 non-null   int64  \n",
      " 3   p1        2075 non-null   object \n",
      " 4   p1_conf   2075 non-null   float64\n",
      " 5   p1_dog    2075 non-null   bool   \n",
      " 6   p2        2075 non-null   object \n",
      " 7   p2_conf   2075 non-null   float64\n",
      " 8   p2_dog    2075 non-null   bool   \n",
      " 9   p3        2075 non-null   object \n",
      " 10  p3_conf   2075 non-null   float64\n",
      " 11  p3_dog    2075 non-null   bool   \n",
      "dtypes: bool(3), float64(3), int64(2), object(4)\n",
      "memory usage: 152.1+ KB\n"
     ]
    }
   ],
   "source": [
    "df_image.info()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 147,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "tweet_id    2075\n",
       "jpg_url     2009\n",
       "img_num        4\n",
       "p1           378\n",
       "p1_conf     2006\n",
       "p1_dog         2\n",
       "p2           405\n",
       "p2_conf     2004\n",
       "p2_dog         2\n",
       "p3           408\n",
       "p3_conf     2006\n",
       "p3_dog         2\n",
       "dtype: int64"
      ]
     },
     "execution_count": 147,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_image.nunique()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 148,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "id                2331\n",
       "favorite_count    1988\n",
       "retweet_count     1682\n",
       "dtype: int64"
      ]
     },
     "execution_count": 148,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_api.nunique()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**Quality:**\n",
    "- contains retweets and therefore, duplicates\n",
    "- many tweet_id(s) of df_twitter table are missing in df_img (image predictions) table\n",
    "- unnecessary html tags in source column in place of utility name e.g. <a href=\"\"http://twitter.com/download/iphone\"\"  rel=\"\"nofollow\"\">Twitter for iPhone</a>\n",
    "- rating_numerator column has values less than 10 as well as some very large numbers (e.g. 1176). We are going to prorate them.\n",
    "- rating_denominator column has values other than 10\n",
    "- some records have more than one dog stage\n",
    "- We need to fix the date format\n",
    "- erroneous dog names starting with lowercase characters (e.g. a, an, actually, by)\n",
    "\n",
    "**Tidy:**\n",
    "\n",
    "- We need to merge, join, concat our 3 datasets\n",
    "- We need to reorganize the prediction and confidence columns to have something clearer"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Clean #"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 149,
   "metadata": {},
   "outputs": [],
   "source": [
    "#Let's create some copy of our original datasets\n",
    "twitter_copy = df_twitter.copy()\n",
    "img_copy = df_image.copy()\n",
    "api_copy = df_api.copy()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**1 - Define:** We are going to drop the retweeted lines in our df_twitter dataframe (181 lines)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**Code:**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 150,
   "metadata": {},
   "outputs": [],
   "source": [
    "twitter_copy = twitter_copy[twitter_copy['retweeted_status_user_id'].isnull()]"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**Test:**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 151,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0"
      ]
     },
     "execution_count": 151,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "len(twitter_copy[twitter_copy['retweeted_status_user_id'].isnull() == False])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 152,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(2175, 17)"
      ]
     },
     "execution_count": 152,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "twitter_copy.shape"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**Let's now remove the retweet columns since they are now empty:**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 153,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>tweet_id</th>\n",
       "      <th>timestamp</th>\n",
       "      <th>source</th>\n",
       "      <th>text</th>\n",
       "      <th>expanded_urls</th>\n",
       "      <th>rating_numerator</th>\n",
       "      <th>rating_denominator</th>\n",
       "      <th>name</th>\n",
       "      <th>doggo</th>\n",
       "      <th>floofer</th>\n",
       "      <th>pupper</th>\n",
       "      <th>puppo</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>892420643555336193</td>\n",
       "      <td>2017-08-01 16:23:56 +0000</td>\n",
       "      <td>&lt;a href=\"http://twitter.com/download/iphone\" r...</td>\n",
       "      <td>This is Phineas. He's a mystical boy. Only eve...</td>\n",
       "      <td>https://twitter.com/dog_rates/status/892420643...</td>\n",
       "      <td>13</td>\n",
       "      <td>10</td>\n",
       "      <td>Phineas</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "             tweet_id                  timestamp  \\\n",
       "0  892420643555336193  2017-08-01 16:23:56 +0000   \n",
       "\n",
       "                                              source  \\\n",
       "0  <a href=\"http://twitter.com/download/iphone\" r...   \n",
       "\n",
       "                                                text  \\\n",
       "0  This is Phineas. He's a mystical boy. Only eve...   \n",
       "\n",
       "                                       expanded_urls  rating_numerator  \\\n",
       "0  https://twitter.com/dog_rates/status/892420643...                13   \n",
       "\n",
       "   rating_denominator     name doggo floofer pupper puppo  \n",
       "0                  10  Phineas  None    None   None  None  "
      ]
     },
     "execution_count": 153,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "twitter_copy.drop(['retweeted_status_id', 'retweeted_status_user_id', 'retweeted_status_timestamp', 'in_reply_to_status_id', 'in_reply_to_user_id'], axis=1, inplace=True)\n",
    "twitter_copy.head(1)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**2 - Define:** \n",
    "We need to keep in df_twitter the tweet_id existing in df_image. We can drop the other ones"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**Code:**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 154,
   "metadata": {},
   "outputs": [],
   "source": [
    "twitter_copy = twitter_copy[twitter_copy.tweet_id.isin(img_copy.tweet_id)]"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**Test:**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 155,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(1994, 12)"
      ]
     },
     "execution_count": 155,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "twitter_copy.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 156,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>tweet_id</th>\n",
       "      <th>timestamp</th>\n",
       "      <th>source</th>\n",
       "      <th>text</th>\n",
       "      <th>expanded_urls</th>\n",
       "      <th>rating_numerator</th>\n",
       "      <th>rating_denominator</th>\n",
       "      <th>name</th>\n",
       "      <th>doggo</th>\n",
       "      <th>floofer</th>\n",
       "      <th>pupper</th>\n",
       "      <th>puppo</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>892420643555336193</td>\n",
       "      <td>2017-08-01 16:23:56 +0000</td>\n",
       "      <td>&lt;a href=\"http://twitter.com/download/iphone\" r...</td>\n",
       "      <td>This is Phineas. He's a mystical boy. Only eve...</td>\n",
       "      <td>https://twitter.com/dog_rates/status/892420643...</td>\n",
       "      <td>13</td>\n",
       "      <td>10</td>\n",
       "      <td>Phineas</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>892177421306343426</td>\n",
       "      <td>2017-08-01 00:17:27 +0000</td>\n",
       "      <td>&lt;a href=\"http://twitter.com/download/iphone\" r...</td>\n",
       "      <td>This is Tilly. She's just checking pup on you....</td>\n",
       "      <td>https://twitter.com/dog_rates/status/892177421...</td>\n",
       "      <td>13</td>\n",
       "      <td>10</td>\n",
       "      <td>Tilly</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>891815181378084864</td>\n",
       "      <td>2017-07-31 00:18:03 +0000</td>\n",
       "      <td>&lt;a href=\"http://twitter.com/download/iphone\" r...</td>\n",
       "      <td>This is Archie. He is a rare Norwegian Pouncin...</td>\n",
       "      <td>https://twitter.com/dog_rates/status/891815181...</td>\n",
       "      <td>12</td>\n",
       "      <td>10</td>\n",
       "      <td>Archie</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "             tweet_id                  timestamp  \\\n",
       "0  892420643555336193  2017-08-01 16:23:56 +0000   \n",
       "1  892177421306343426  2017-08-01 00:17:27 +0000   \n",
       "2  891815181378084864  2017-07-31 00:18:03 +0000   \n",
       "\n",
       "                                              source  \\\n",
       "0  <a href=\"http://twitter.com/download/iphone\" r...   \n",
       "1  <a href=\"http://twitter.com/download/iphone\" r...   \n",
       "2  <a href=\"http://twitter.com/download/iphone\" r...   \n",
       "\n",
       "                                                text  \\\n",
       "0  This is Phineas. He's a mystical boy. Only eve...   \n",
       "1  This is Tilly. She's just checking pup on you....   \n",
       "2  This is Archie. He is a rare Norwegian Pouncin...   \n",
       "\n",
       "                                       expanded_urls  rating_numerator  \\\n",
       "0  https://twitter.com/dog_rates/status/892420643...                13   \n",
       "1  https://twitter.com/dog_rates/status/892177421...                13   \n",
       "2  https://twitter.com/dog_rates/status/891815181...                12   \n",
       "\n",
       "   rating_denominator     name doggo floofer pupper puppo  \n",
       "0                  10  Phineas  None    None   None  None  \n",
       "1                  10    Tilly  None    None   None  None  \n",
       "2                  10   Archie  None    None   None  None  "
      ]
     },
     "execution_count": 156,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "twitter_copy.head(3)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**3 - Define:**\n",
    "Let's change the data format from object to date time format, and then let's separate both the date part and time part of this"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**Code:**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 157,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Creating a function to change object to datetime\n",
    "def object_to_datetime(str):\n",
    "    return pd.to_datetime(str)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 158,
   "metadata": {},
   "outputs": [],
   "source": [
    "twitter_copy['timestamp'] = twitter_copy['timestamp'].apply(object_to_datetime)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 159,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "<class 'pandas.core.frame.DataFrame'>\n",
      "Int64Index: 1994 entries, 0 to 2355\n",
      "Data columns (total 12 columns):\n",
      " #   Column              Non-Null Count  Dtype              \n",
      "---  ------              --------------  -----              \n",
      " 0   tweet_id            1994 non-null   int64              \n",
      " 1   timestamp           1994 non-null   datetime64[ns, UTC]\n",
      " 2   source              1994 non-null   object             \n",
      " 3   text                1994 non-null   object             \n",
      " 4   expanded_urls       1994 non-null   object             \n",
      " 5   rating_numerator    1994 non-null   int64              \n",
      " 6   rating_denominator  1994 non-null   int64              \n",
      " 7   name                1994 non-null   object             \n",
      " 8   doggo               1994 non-null   object             \n",
      " 9   floofer             1994 non-null   object             \n",
      " 10  pupper              1994 non-null   object             \n",
      " 11  puppo               1994 non-null   object             \n",
      "dtypes: datetime64[ns, UTC](1), int64(3), object(8)\n",
      "memory usage: 202.5+ KB\n"
     ]
    }
   ],
   "source": [
    "twitter_copy.info()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 160,
   "metadata": {},
   "outputs": [],
   "source": [
    "# extracting date from timestamp\n",
    "twitter_copy['Date'] = [datetime.datetime.date(d) for d in twitter_copy['timestamp']] \n",
    "\n",
    "# extracting time from timestamp\n",
    "twitter_copy['Time'] = [datetime.datetime.time(d) for d in twitter_copy['timestamp']] "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 161,
   "metadata": {},
   "outputs": [],
   "source": [
    "twitter_copy.drop(['Time'], axis=1, inplace=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 162,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>tweet_id</th>\n",
       "      <th>timestamp</th>\n",
       "      <th>source</th>\n",
       "      <th>text</th>\n",
       "      <th>expanded_urls</th>\n",
       "      <th>rating_numerator</th>\n",
       "      <th>rating_denominator</th>\n",
       "      <th>name</th>\n",
       "      <th>doggo</th>\n",
       "      <th>floofer</th>\n",
       "      <th>pupper</th>\n",
       "      <th>puppo</th>\n",
       "      <th>Date</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>1112</th>\n",
       "      <td>733460102733135873</td>\n",
       "      <td>2016-05-20 00:51:30+00:00</td>\n",
       "      <td>&lt;a href=\"http://twitter.com/download/iphone\" r...</td>\n",
       "      <td>This is Hermione. Her face is as old as time. ...</td>\n",
       "      <td>https://twitter.com/dog_rates/status/733460102...</td>\n",
       "      <td>11</td>\n",
       "      <td>10</td>\n",
       "      <td>Hermione</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "      <td>2016-05-20</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1852</th>\n",
       "      <td>675707330206547968</td>\n",
       "      <td>2015-12-12 16:02:36+00:00</td>\n",
       "      <td>&lt;a href=\"http://twitter.com/download/iphone\" r...</td>\n",
       "      <td>We've got ourselves a battle here. Watch out R...</td>\n",
       "      <td>https://twitter.com/dog_rates/status/675707330...</td>\n",
       "      <td>11</td>\n",
       "      <td>10</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "      <td>2015-12-12</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1975</th>\n",
       "      <td>672997845381865473</td>\n",
       "      <td>2015-12-05 04:36:04+00:00</td>\n",
       "      <td>&lt;a href=\"http://twitter.com/download/iphone\" r...</td>\n",
       "      <td>Say hello to Kenzie. She is a fluff ball. 12/1...</td>\n",
       "      <td>https://twitter.com/dog_rates/status/672997845...</td>\n",
       "      <td>12</td>\n",
       "      <td>10</td>\n",
       "      <td>Kenzie</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "      <td>2015-12-05</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                tweet_id                 timestamp  \\\n",
       "1112  733460102733135873 2016-05-20 00:51:30+00:00   \n",
       "1852  675707330206547968 2015-12-12 16:02:36+00:00   \n",
       "1975  672997845381865473 2015-12-05 04:36:04+00:00   \n",
       "\n",
       "                                                 source  \\\n",
       "1112  <a href=\"http://twitter.com/download/iphone\" r...   \n",
       "1852  <a href=\"http://twitter.com/download/iphone\" r...   \n",
       "1975  <a href=\"http://twitter.com/download/iphone\" r...   \n",
       "\n",
       "                                                   text  \\\n",
       "1112  This is Hermione. Her face is as old as time. ...   \n",
       "1852  We've got ourselves a battle here. Watch out R...   \n",
       "1975  Say hello to Kenzie. She is a fluff ball. 12/1...   \n",
       "\n",
       "                                          expanded_urls  rating_numerator  \\\n",
       "1112  https://twitter.com/dog_rates/status/733460102...                11   \n",
       "1852  https://twitter.com/dog_rates/status/675707330...                11   \n",
       "1975  https://twitter.com/dog_rates/status/672997845...                12   \n",
       "\n",
       "      rating_denominator      name doggo floofer pupper puppo        Date  \n",
       "1112                  10  Hermione  None    None   None  None  2016-05-20  \n",
       "1852                  10      None  None    None   None  None  2015-12-12  \n",
       "1975                  10    Kenzie  None    None   None  None  2015-12-05  "
      ]
     },
     "execution_count": 162,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "twitter_copy.sample(3)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**Test:**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 163,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "<class 'pandas.core.frame.DataFrame'>\n",
      "Int64Index: 1994 entries, 0 to 2355\n",
      "Data columns (total 13 columns):\n",
      " #   Column              Non-Null Count  Dtype              \n",
      "---  ------              --------------  -----              \n",
      " 0   tweet_id            1994 non-null   int64              \n",
      " 1   timestamp           1994 non-null   datetime64[ns, UTC]\n",
      " 2   source              1994 non-null   object             \n",
      " 3   text                1994 non-null   object             \n",
      " 4   expanded_urls       1994 non-null   object             \n",
      " 5   rating_numerator    1994 non-null   int64              \n",
      " 6   rating_denominator  1994 non-null   int64              \n",
      " 7   name                1994 non-null   object             \n",
      " 8   doggo               1994 non-null   object             \n",
      " 9   floofer             1994 non-null   object             \n",
      " 10  pupper              1994 non-null   object             \n",
      " 11  puppo               1994 non-null   object             \n",
      " 12  Date                1994 non-null   object             \n",
      "dtypes: datetime64[ns, UTC](1), int64(3), object(9)\n",
      "memory usage: 218.1+ KB\n"
     ]
    }
   ],
   "source": [
    "twitter_copy.info()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 164,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>tweet_id</th>\n",
       "      <th>timestamp</th>\n",
       "      <th>source</th>\n",
       "      <th>text</th>\n",
       "      <th>expanded_urls</th>\n",
       "      <th>rating_numerator</th>\n",
       "      <th>rating_denominator</th>\n",
       "      <th>name</th>\n",
       "      <th>doggo</th>\n",
       "      <th>floofer</th>\n",
       "      <th>pupper</th>\n",
       "      <th>puppo</th>\n",
       "      <th>Date</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>892420643555336193</td>\n",
       "      <td>2017-08-01 16:23:56+00:00</td>\n",
       "      <td>&lt;a href=\"http://twitter.com/download/iphone\" r...</td>\n",
       "      <td>This is Phineas. He's a mystical boy. Only eve...</td>\n",
       "      <td>https://twitter.com/dog_rates/status/892420643...</td>\n",
       "      <td>13</td>\n",
       "      <td>10</td>\n",
       "      <td>Phineas</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "      <td>2017-08-01</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>892177421306343426</td>\n",
       "      <td>2017-08-01 00:17:27+00:00</td>\n",
       "      <td>&lt;a href=\"http://twitter.com/download/iphone\" r...</td>\n",
       "      <td>This is Tilly. She's just checking pup on you....</td>\n",
       "      <td>https://twitter.com/dog_rates/status/892177421...</td>\n",
       "      <td>13</td>\n",
       "      <td>10</td>\n",
       "      <td>Tilly</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "      <td>2017-08-01</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>891815181378084864</td>\n",
       "      <td>2017-07-31 00:18:03+00:00</td>\n",
       "      <td>&lt;a href=\"http://twitter.com/download/iphone\" r...</td>\n",
       "      <td>This is Archie. He is a rare Norwegian Pouncin...</td>\n",
       "      <td>https://twitter.com/dog_rates/status/891815181...</td>\n",
       "      <td>12</td>\n",
       "      <td>10</td>\n",
       "      <td>Archie</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "      <td>2017-07-31</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>891689557279858688</td>\n",
       "      <td>2017-07-30 15:58:51+00:00</td>\n",
       "      <td>&lt;a href=\"http://twitter.com/download/iphone\" r...</td>\n",
       "      <td>This is Darla. She commenced a snooze mid meal...</td>\n",
       "      <td>https://twitter.com/dog_rates/status/891689557...</td>\n",
       "      <td>13</td>\n",
       "      <td>10</td>\n",
       "      <td>Darla</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "      <td>2017-07-30</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>891327558926688256</td>\n",
       "      <td>2017-07-29 16:00:24+00:00</td>\n",
       "      <td>&lt;a href=\"http://twitter.com/download/iphone\" r...</td>\n",
       "      <td>This is Franklin. He would like you to stop ca...</td>\n",
       "      <td>https://twitter.com/dog_rates/status/891327558...</td>\n",
       "      <td>12</td>\n",
       "      <td>10</td>\n",
       "      <td>Franklin</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "      <td>2017-07-29</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "             tweet_id                 timestamp  \\\n",
       "0  892420643555336193 2017-08-01 16:23:56+00:00   \n",
       "1  892177421306343426 2017-08-01 00:17:27+00:00   \n",
       "2  891815181378084864 2017-07-31 00:18:03+00:00   \n",
       "3  891689557279858688 2017-07-30 15:58:51+00:00   \n",
       "4  891327558926688256 2017-07-29 16:00:24+00:00   \n",
       "\n",
       "                                              source  \\\n",
       "0  <a href=\"http://twitter.com/download/iphone\" r...   \n",
       "1  <a href=\"http://twitter.com/download/iphone\" r...   \n",
       "2  <a href=\"http://twitter.com/download/iphone\" r...   \n",
       "3  <a href=\"http://twitter.com/download/iphone\" r...   \n",
       "4  <a href=\"http://twitter.com/download/iphone\" r...   \n",
       "\n",
       "                                                text  \\\n",
       "0  This is Phineas. He's a mystical boy. Only eve...   \n",
       "1  This is Tilly. She's just checking pup on you....   \n",
       "2  This is Archie. He is a rare Norwegian Pouncin...   \n",
       "3  This is Darla. She commenced a snooze mid meal...   \n",
       "4  This is Franklin. He would like you to stop ca...   \n",
       "\n",
       "                                       expanded_urls  rating_numerator  \\\n",
       "0  https://twitter.com/dog_rates/status/892420643...                13   \n",
       "1  https://twitter.com/dog_rates/status/892177421...                13   \n",
       "2  https://twitter.com/dog_rates/status/891815181...                12   \n",
       "3  https://twitter.com/dog_rates/status/891689557...                13   \n",
       "4  https://twitter.com/dog_rates/status/891327558...                12   \n",
       "\n",
       "   rating_denominator      name doggo floofer pupper puppo        Date  \n",
       "0                  10   Phineas  None    None   None  None  2017-08-01  \n",
       "1                  10     Tilly  None    None   None  None  2017-08-01  \n",
       "2                  10    Archie  None    None   None  None  2017-07-31  \n",
       "3                  10     Darla  None    None   None  None  2017-07-30  \n",
       "4                  10  Franklin  None    None   None  None  2017-07-29  "
      ]
     },
     "execution_count": 164,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "twitter_copy.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**4 - Define:**\n",
    "Some entries have multiple dog stages which needs to be cleaned before melting these 4 columns into one."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**Code:**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 165,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "9\n",
      "1\n",
      "1\n"
     ]
    }
   ],
   "source": [
    "print(len(twitter_copy[(twitter_copy['pupper'] != 'None' ) & (twitter_copy['doggo'] != 'None')]))\n",
    "print(len(twitter_copy[(twitter_copy['puppo'] != 'None' ) & (twitter_copy['doggo'] != 'None')]))\n",
    "print(len(twitter_copy[(twitter_copy['doggo'] != 'None' ) & (twitter_copy['floofer'] != 'None')]))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 166,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "<ipython-input-166-54957a2c1471>:2: SettingWithCopyWarning: \n",
      "A value is trying to be set on a copy of a slice from a DataFrame\n",
      "\n",
      "See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy\n",
      "  twitter_copy['pupper'][(twitter_copy['doggo'] == 'doggo' ) & (twitter_copy['pupper'] == 'pupper')] = 'None'\n"
     ]
    }
   ],
   "source": [
    "#I arbitrarily decide to change and replace the 9 entries of doggo and pupper to just doggo\n",
    "twitter_copy['pupper'][(twitter_copy['doggo'] == 'doggo' ) & (twitter_copy['pupper'] == 'pupper')] = 'None'"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 167,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[\"At first I thought this was a shy doggo, but it's actually a Rare Canadian Floofer Owl. Amateurs would confuse the two. 11/10 only send dogs https://t.co/TXdT3tmuYk\"]"
      ]
     },
     "execution_count": 167,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "list(twitter_copy[(twitter_copy['doggo'] != 'None' ) & (twitter_copy['floofer'] != 'None')]['text'])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 168,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "<ipython-input-168-f79215b414fa>:1: SettingWithCopyWarning: \n",
      "A value is trying to be set on a copy of a slice from a DataFrame\n",
      "\n",
      "See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy\n",
      "  twitter_copy['doggo'][(twitter_copy['doggo'] != 'None' ) & (twitter_copy['floofer'] != 'None')] = 'None'\n"
     ]
    }
   ],
   "source": [
    "twitter_copy['doggo'][(twitter_copy['doggo'] != 'None' ) & (twitter_copy['floofer'] != 'None')] = 'None'"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 169,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[\"Here's a puppo participating in the #ScienceMarch. Cleverly disguising her own doggo agenda. 13/10 would keep the planet habitable for https://t.co/cMhq16isel\"]"
      ]
     },
     "execution_count": 169,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "list(twitter_copy[(twitter_copy['doggo'] != 'None' ) & (twitter_copy['puppo'] != 'None')]['text'])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 170,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "<ipython-input-170-c1703200899c>:1: SettingWithCopyWarning: \n",
      "A value is trying to be set on a copy of a slice from a DataFrame\n",
      "\n",
      "See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy\n",
      "  twitter_copy['doggo'][(twitter_copy['doggo'] != 'None' ) & (twitter_copy['puppo'] != 'None')] = 'None'\n"
     ]
    }
   ],
   "source": [
    "twitter_copy['doggo'][(twitter_copy['doggo'] != 'None' ) & (twitter_copy['puppo'] != 'None')] = 'None'"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**Test:**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 171,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "0\n",
      "0\n",
      "0\n"
     ]
    }
   ],
   "source": [
    "print(len(twitter_copy[(twitter_copy['pupper'] != 'None' ) & (twitter_copy['doggo'] != 'None')]))\n",
    "print(len(twitter_copy[(twitter_copy['puppo'] != 'None' ) & (twitter_copy['doggo'] != 'None')]))\n",
    "print(len(twitter_copy[(twitter_copy['doggo'] != 'None' ) & (twitter_copy['floofer'] != 'None')]))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**5 - Define:** One column 'Stage' should be merging the column doggo, floofer, pupper and puppo"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**Code:**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 172,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>tweet_id</th>\n",
       "      <th>timestamp</th>\n",
       "      <th>source</th>\n",
       "      <th>text</th>\n",
       "      <th>expanded_urls</th>\n",
       "      <th>rating_numerator</th>\n",
       "      <th>rating_denominator</th>\n",
       "      <th>name</th>\n",
       "      <th>doggo</th>\n",
       "      <th>floofer</th>\n",
       "      <th>pupper</th>\n",
       "      <th>puppo</th>\n",
       "      <th>Date</th>\n",
       "      <th>stage</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>2234</th>\n",
       "      <td>668190681446379520</td>\n",
       "      <td>2015-11-21 22:14:07+00:00</td>\n",
       "      <td>&lt;a href=\"http://twitter.com/download/iphone\" r...</td>\n",
       "      <td>This is Skittles. I would kidnap Skittles. Pin...</td>\n",
       "      <td>https://twitter.com/dog_rates/status/668190681...</td>\n",
       "      <td>12</td>\n",
       "      <td>10</td>\n",
       "      <td>Skittles</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "      <td>2015-11-21</td>\n",
       "      <td>None</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2333</th>\n",
       "      <td>666337882303524864</td>\n",
       "      <td>2015-11-16 19:31:45+00:00</td>\n",
       "      <td>&lt;a href=\"http://twitter.com/download/iphone\" r...</td>\n",
       "      <td>This is an extremely rare horned Parthenon. No...</td>\n",
       "      <td>https://twitter.com/dog_rates/status/666337882...</td>\n",
       "      <td>9</td>\n",
       "      <td>10</td>\n",
       "      <td>an</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "      <td>2015-11-16</td>\n",
       "      <td>None</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1010</th>\n",
       "      <td>747461612269887489</td>\n",
       "      <td>2016-06-27 16:08:30+00:00</td>\n",
       "      <td>&lt;a href=\"http://twitter.com/download/iphone\" r...</td>\n",
       "      <td>Never seen a shark hold another shark like thi...</td>\n",
       "      <td>https://twitter.com/dog_rates/status/747461612...</td>\n",
       "      <td>10</td>\n",
       "      <td>10</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "      <td>2016-06-27</td>\n",
       "      <td>None</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1168</th>\n",
       "      <td>721001180231503872</td>\n",
       "      <td>2016-04-15 15:44:11+00:00</td>\n",
       "      <td>&lt;a href=\"http://twitter.com/download/iphone\" r...</td>\n",
       "      <td>This is Oliver. Bath time is upon him. His fea...</td>\n",
       "      <td>https://twitter.com/dog_rates/status/721001180...</td>\n",
       "      <td>11</td>\n",
       "      <td>10</td>\n",
       "      <td>Oliver</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "      <td>pupper</td>\n",
       "      <td>None</td>\n",
       "      <td>2016-04-15</td>\n",
       "      <td>pupper</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2233</th>\n",
       "      <td>668204964695683073</td>\n",
       "      <td>2015-11-21 23:10:52+00:00</td>\n",
       "      <td>&lt;a href=\"http://twitter.com/download/iphone\" r...</td>\n",
       "      <td>This is Ron. Ron's currently experiencing a br...</td>\n",
       "      <td>https://twitter.com/dog_rates/status/668204964...</td>\n",
       "      <td>8</td>\n",
       "      <td>10</td>\n",
       "      <td>Ron</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "      <td>2015-11-21</td>\n",
       "      <td>None</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                tweet_id                 timestamp  \\\n",
       "2234  668190681446379520 2015-11-21 22:14:07+00:00   \n",
       "2333  666337882303524864 2015-11-16 19:31:45+00:00   \n",
       "1010  747461612269887489 2016-06-27 16:08:30+00:00   \n",
       "1168  721001180231503872 2016-04-15 15:44:11+00:00   \n",
       "2233  668204964695683073 2015-11-21 23:10:52+00:00   \n",
       "\n",
       "                                                 source  \\\n",
       "2234  <a href=\"http://twitter.com/download/iphone\" r...   \n",
       "2333  <a href=\"http://twitter.com/download/iphone\" r...   \n",
       "1010  <a href=\"http://twitter.com/download/iphone\" r...   \n",
       "1168  <a href=\"http://twitter.com/download/iphone\" r...   \n",
       "2233  <a href=\"http://twitter.com/download/iphone\" r...   \n",
       "\n",
       "                                                   text  \\\n",
       "2234  This is Skittles. I would kidnap Skittles. Pin...   \n",
       "2333  This is an extremely rare horned Parthenon. No...   \n",
       "1010  Never seen a shark hold another shark like thi...   \n",
       "1168  This is Oliver. Bath time is upon him. His fea...   \n",
       "2233  This is Ron. Ron's currently experiencing a br...   \n",
       "\n",
       "                                          expanded_urls  rating_numerator  \\\n",
       "2234  https://twitter.com/dog_rates/status/668190681...                12   \n",
       "2333  https://twitter.com/dog_rates/status/666337882...                 9   \n",
       "1010  https://twitter.com/dog_rates/status/747461612...                10   \n",
       "1168  https://twitter.com/dog_rates/status/721001180...                11   \n",
       "2233  https://twitter.com/dog_rates/status/668204964...                 8   \n",
       "\n",
       "      rating_denominator      name doggo floofer  pupper puppo        Date  \\\n",
       "2234                  10  Skittles  None    None    None  None  2015-11-21   \n",
       "2333                  10        an  None    None    None  None  2015-11-16   \n",
       "1010                  10      None  None    None    None  None  2016-06-27   \n",
       "1168                  10    Oliver  None    None  pupper  None  2016-04-15   \n",
       "2233                  10       Ron  None    None    None  None  2015-11-21   \n",
       "\n",
       "       stage  \n",
       "2234    None  \n",
       "2333    None  \n",
       "1010    None  \n",
       "1168  pupper  \n",
       "2233    None  "
      ]
     },
     "execution_count": 172,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "twitter_copy['stage'] = twitter_copy[['doggo', 'floofer', 'pupper', 'puppo']].agg(''.join, axis=1)\n",
    "\n",
    "twitter_copy['stage'].replace('NoneNoneNoneNone', 'None', inplace=True)\n",
    "twitter_copy['stage'].replace(\"doggoNoneNoneNone\",\"doggo\", inplace=True)\n",
    "twitter_copy['stage'].replace(\"NoneflooferNoneNone\",\"floofer\", inplace=True)\n",
    "twitter_copy['stage'].replace(\"NoneNonepupperNone\",\"pupper\", inplace=True)\n",
    "twitter_copy['stage'].replace(\"NoneNoneNonepuppo\",\"puppo\", inplace=True)\n",
    "twitter_copy.sample(5)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**Test:**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 173,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>doggo</th>\n",
       "      <th>floofer</th>\n",
       "      <th>pupper</th>\n",
       "      <th>puppo</th>\n",
       "      <th>stage</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>1470</th>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>481</th>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>419</th>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>28</th>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1929</th>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>353</th>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>871</th>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1558</th>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>673</th>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>884</th>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "     doggo floofer pupper puppo stage\n",
       "1470  None    None   None  None  None\n",
       "481   None    None   None  None  None\n",
       "419   None    None   None  None  None\n",
       "28    None    None   None  None  None\n",
       "1929  None    None   None  None  None\n",
       "353   None    None   None  None  None\n",
       "871   None    None   None  None  None\n",
       "1558  None    None   None  None  None\n",
       "673   None    None   None  None  None\n",
       "884   None    None   None  None  None"
      ]
     },
     "execution_count": 173,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "twitter_copy[['doggo', 'floofer', 'pupper', 'puppo', 'stage']].sample(10)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 174,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>tweets</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>stage</th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>None</th>\n",
       "      <td>1688</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>doggo</th>\n",
       "      <td>72</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>floofer</th>\n",
       "      <td>8</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>pupper</th>\n",
       "      <td>203</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>puppo</th>\n",
       "      <td>23</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "         tweets\n",
       "stage          \n",
       "None       1688\n",
       "doggo        72\n",
       "floofer       8\n",
       "pupper      203\n",
       "puppo        23"
      ]
     },
     "execution_count": 174,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "twitter_copy.groupby(['stage']).agg(tweets=('tweet_id', 'count'))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**Code:**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 175,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>tweet_id</th>\n",
       "      <th>timestamp</th>\n",
       "      <th>source</th>\n",
       "      <th>text</th>\n",
       "      <th>expanded_urls</th>\n",
       "      <th>rating_numerator</th>\n",
       "      <th>rating_denominator</th>\n",
       "      <th>name</th>\n",
       "      <th>Date</th>\n",
       "      <th>stage</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>892420643555336193</td>\n",
       "      <td>2017-08-01 16:23:56+00:00</td>\n",
       "      <td>&lt;a href=\"http://twitter.com/download/iphone\" r...</td>\n",
       "      <td>This is Phineas. He's a mystical boy. Only eve...</td>\n",
       "      <td>https://twitter.com/dog_rates/status/892420643...</td>\n",
       "      <td>13</td>\n",
       "      <td>10</td>\n",
       "      <td>Phineas</td>\n",
       "      <td>2017-08-01</td>\n",
       "      <td>None</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "             tweet_id                 timestamp  \\\n",
       "0  892420643555336193 2017-08-01 16:23:56+00:00   \n",
       "\n",
       "                                              source  \\\n",
       "0  <a href=\"http://twitter.com/download/iphone\" r...   \n",
       "\n",
       "                                                text  \\\n",
       "0  This is Phineas. He's a mystical boy. Only eve...   \n",
       "\n",
       "                                       expanded_urls  rating_numerator  \\\n",
       "0  https://twitter.com/dog_rates/status/892420643...                13   \n",
       "\n",
       "   rating_denominator     name        Date stage  \n",
       "0                  10  Phineas  2017-08-01  None  "
      ]
     },
     "execution_count": 175,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "twitter_copy.drop(['doggo', 'pupper','puppo', 'floofer'], axis=1, inplace=True)\n",
    "twitter_copy.head(1)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**6 - Define:**\n",
    "We need to extract text from the Source column in order to better see on which device users accessed to Twitter for their post."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**Code:**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 176,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "<a href=\"http://twitter.com/download/iphone\" rel=\"nofollow\">Twitter for iPhone</a>\n",
      "<a href=\"http://twitter.com/download/iphone\" rel=\"nofollow\">Twitter for iPhone</a>\n",
      "<a href=\"http://twitter.com/download/iphone\" rel=\"nofollow\">Twitter for iPhone</a>\n"
     ]
    }
   ],
   "source": [
    "print(twitter_copy.iloc[0].source)\n",
    "print(twitter_copy.iloc[1].source)\n",
    "print(twitter_copy.iloc[2].source)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 177,
   "metadata": {},
   "outputs": [],
   "source": [
    "twitter_copy.source = twitter_copy.source.str.replace(r'<(?:a\\b[^>]*>|/a>)', '')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**Test:**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 178,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Twitter for iPhone    1955\n",
       "Twitter Web Client      28\n",
       "TweetDeck               11\n",
       "Name: source, dtype: int64"
      ]
     },
     "execution_count": 178,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "twitter_copy.source.value_counts()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**7 - Define:**\n",
    "We have to change the denominator which are not 10 in order to have consistency. We also need to prorate numerators"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 179,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "10     1976\n",
       "50        3\n",
       "80        2\n",
       "11        2\n",
       "170       1\n",
       "150       1\n",
       "130       1\n",
       "120       1\n",
       "110       1\n",
       "90        1\n",
       "70        1\n",
       "40        1\n",
       "20        1\n",
       "7         1\n",
       "2         1\n",
       "Name: rating_denominator, dtype: int64"
      ]
     },
     "execution_count": 179,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "twitter_copy.rating_denominator.value_counts()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 180,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "12      450\n",
       "10      419\n",
       "11      396\n",
       "13      261\n",
       "9       151\n",
       "8        95\n",
       "7        52\n",
       "14       35\n",
       "5        33\n",
       "6        32\n",
       "3        19\n",
       "4        16\n",
       "2         9\n",
       "1         5\n",
       "0         2\n",
       "420       1\n",
       "24        1\n",
       "1776      1\n",
       "27        1\n",
       "44        1\n",
       "45        1\n",
       "50        1\n",
       "60        1\n",
       "75        1\n",
       "80        1\n",
       "84        1\n",
       "88        1\n",
       "99        1\n",
       "121       1\n",
       "143       1\n",
       "144       1\n",
       "165       1\n",
       "204       1\n",
       "26        1\n",
       "Name: rating_numerator, dtype: int64"
      ]
     },
     "execution_count": 180,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "twitter_copy.rating_numerator.value_counts()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**Code:**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 181,
   "metadata": {},
   "outputs": [],
   "source": [
    "twitter_copy['rating_denominatorbis'] = 10\n",
    "twitter_copy['rating_numeratorbis'] = np.where(twitter_copy.rating_denominator == 0 , 10,\n",
    "         (np.where(twitter_copy.rating_denominator != 10, ((twitter_copy['rating_numerator']*10)/twitter_copy['rating_denominator']),twitter_copy.rating_numerator)))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 182,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "<class 'pandas.core.frame.DataFrame'>\n",
      "Int64Index: 1994 entries, 0 to 2355\n",
      "Data columns (total 12 columns):\n",
      " #   Column                 Non-Null Count  Dtype              \n",
      "---  ------                 --------------  -----              \n",
      " 0   tweet_id               1994 non-null   int64              \n",
      " 1   timestamp              1994 non-null   datetime64[ns, UTC]\n",
      " 2   source                 1994 non-null   object             \n",
      " 3   text                   1994 non-null   object             \n",
      " 4   expanded_urls          1994 non-null   object             \n",
      " 5   rating_numerator       1994 non-null   int64              \n",
      " 6   rating_denominator     1994 non-null   int64              \n",
      " 7   name                   1994 non-null   object             \n",
      " 8   Date                   1994 non-null   object             \n",
      " 9   stage                  1994 non-null   object             \n",
      " 10  rating_denominatorbis  1994 non-null   int64              \n",
      " 11  rating_numeratorbis    1994 non-null   float64            \n",
      "dtypes: datetime64[ns, UTC](1), float64(1), int64(4), object(6)\n",
      "memory usage: 282.5+ KB\n"
     ]
    }
   ],
   "source": [
    "twitter_copy.info()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 183,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0    13\n",
       "1    13\n",
       "2    12\n",
       "Name: rating_numeratorbis, dtype: object"
      ]
     },
     "execution_count": 183,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "twitter_copy['rating_numeratorbis'] = np.where(twitter_copy.rating_numeratorbis.astype(float) < 10, twitter_copy.rating_numeratorbis.astype(str).str.pad(2, fillchar='1'), twitter_copy.rating_numeratorbis)\n",
    "twitter_copy.rating_numeratorbis.head(3)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 184,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>tweet_id</th>\n",
       "      <th>timestamp</th>\n",
       "      <th>source</th>\n",
       "      <th>text</th>\n",
       "      <th>expanded_urls</th>\n",
       "      <th>rating_numerator</th>\n",
       "      <th>rating_denominator</th>\n",
       "      <th>name</th>\n",
       "      <th>Date</th>\n",
       "      <th>stage</th>\n",
       "      <th>rating_denominatorbis</th>\n",
       "      <th>rating_numeratorbis</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "Empty DataFrame\n",
       "Columns: [tweet_id, timestamp, source, text, expanded_urls, rating_numerator, rating_denominator, name, Date, stage, rating_denominatorbis, rating_numeratorbis]\n",
       "Index: []"
      ]
     },
     "execution_count": 184,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "twitter_copy[(twitter_copy['rating_numerator'] == 9)&(twitter_copy['rating_numeratorbis'] == '9')]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 185,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>tweet_id</th>\n",
       "      <th>timestamp</th>\n",
       "      <th>source</th>\n",
       "      <th>text</th>\n",
       "      <th>expanded_urls</th>\n",
       "      <th>rating_numerator</th>\n",
       "      <th>rating_denominator</th>\n",
       "      <th>name</th>\n",
       "      <th>Date</th>\n",
       "      <th>stage</th>\n",
       "      <th>rating_denominatorbis</th>\n",
       "      <th>rating_numeratorbis</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "Empty DataFrame\n",
       "Columns: [tweet_id, timestamp, source, text, expanded_urls, rating_numerator, rating_denominator, name, Date, stage, rating_denominatorbis, rating_numeratorbis]\n",
       "Index: []"
      ]
     },
     "execution_count": 185,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "twitter_copy[twitter_copy['rating_numeratorbis'] == '9']"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 186,
   "metadata": {},
   "outputs": [],
   "source": [
    "twitter_copy['rating_numeratorbis'] = twitter_copy.rating_numeratorbis.astype(float)\n",
    "twitter_copy['rating_numeratorbis'] = twitter_copy.rating_numeratorbis.astype(int)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 187,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "<class 'pandas.core.frame.DataFrame'>\n",
      "Int64Index: 1994 entries, 0 to 2355\n",
      "Data columns (total 12 columns):\n",
      " #   Column                 Non-Null Count  Dtype              \n",
      "---  ------                 --------------  -----              \n",
      " 0   tweet_id               1994 non-null   int64              \n",
      " 1   timestamp              1994 non-null   datetime64[ns, UTC]\n",
      " 2   source                 1994 non-null   object             \n",
      " 3   text                   1994 non-null   object             \n",
      " 4   expanded_urls          1994 non-null   object             \n",
      " 5   rating_numerator       1994 non-null   int64              \n",
      " 6   rating_denominator     1994 non-null   int64              \n",
      " 7   name                   1994 non-null   object             \n",
      " 8   Date                   1994 non-null   object             \n",
      " 9   stage                  1994 non-null   object             \n",
      " 10  rating_denominatorbis  1994 non-null   int64              \n",
      " 11  rating_numeratorbis    1994 non-null   int32              \n",
      "dtypes: datetime64[ns, UTC](1), int32(1), int64(4), object(6)\n",
      "memory usage: 274.7+ KB\n"
     ]
    }
   ],
   "source": [
    "twitter_copy.info()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 188,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>tweet_id</th>\n",
       "      <th>timestamp</th>\n",
       "      <th>source</th>\n",
       "      <th>text</th>\n",
       "      <th>expanded_urls</th>\n",
       "      <th>name</th>\n",
       "      <th>Date</th>\n",
       "      <th>stage</th>\n",
       "      <th>rating_denominatorbis</th>\n",
       "      <th>rating_numeratorbis</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>892420643555336193</td>\n",
       "      <td>2017-08-01 16:23:56+00:00</td>\n",
       "      <td>Twitter for iPhone</td>\n",
       "      <td>This is Phineas. He's a mystical boy. Only eve...</td>\n",
       "      <td>https://twitter.com/dog_rates/status/892420643...</td>\n",
       "      <td>Phineas</td>\n",
       "      <td>2017-08-01</td>\n",
       "      <td>None</td>\n",
       "      <td>10</td>\n",
       "      <td>13</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "             tweet_id                 timestamp              source  \\\n",
       "0  892420643555336193 2017-08-01 16:23:56+00:00  Twitter for iPhone   \n",
       "\n",
       "                                                text  \\\n",
       "0  This is Phineas. He's a mystical boy. Only eve...   \n",
       "\n",
       "                                       expanded_urls     name        Date  \\\n",
       "0  https://twitter.com/dog_rates/status/892420643...  Phineas  2017-08-01   \n",
       "\n",
       "  stage  rating_denominatorbis  rating_numeratorbis  \n",
       "0  None                     10                   13  "
      ]
     },
     "execution_count": 188,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "twitter_copy.drop(['rating_denominator', 'rating_numerator'], axis=1, inplace=True)\n",
    "twitter_copy.head(1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 189,
   "metadata": {},
   "outputs": [],
   "source": [
    "twitter_copy.rename(columns={'rating_denominatorbis': 'rating_denominator', 'rating_numeratorbis': 'rating_numerator'}, inplace=True)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**Test:**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 190,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(10    1994\n",
       " Name: rating_denominator, dtype: int64,\n",
       " 12      454\n",
       " 10      421\n",
       " 11      402\n",
       " 13      261\n",
       " 9       151\n",
       " 8        96\n",
       " 7        51\n",
       " 14       35\n",
       " 5        34\n",
       " 6        33\n",
       " 3        19\n",
       " 4        15\n",
       " 2        10\n",
       " 1         4\n",
       " 0         2\n",
       " 420       1\n",
       " 26        1\n",
       " 27        1\n",
       " 34        1\n",
       " 75        1\n",
       " 1776      1\n",
       " Name: rating_numerator, dtype: int64)"
      ]
     },
     "execution_count": 190,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "twitter_copy.rating_denominator.value_counts(),twitter_copy.rating_numerator.value_counts()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 191,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "<class 'pandas.core.frame.DataFrame'>\n",
      "Int64Index: 1994 entries, 0 to 2355\n",
      "Data columns (total 10 columns):\n",
      " #   Column              Non-Null Count  Dtype              \n",
      "---  ------              --------------  -----              \n",
      " 0   tweet_id            1994 non-null   int64              \n",
      " 1   timestamp           1994 non-null   datetime64[ns, UTC]\n",
      " 2   source              1994 non-null   object             \n",
      " 3   text                1994 non-null   object             \n",
      " 4   expanded_urls       1994 non-null   object             \n",
      " 5   name                1994 non-null   object             \n",
      " 6   Date                1994 non-null   object             \n",
      " 7   stage               1994 non-null   object             \n",
      " 8   rating_denominator  1994 non-null   int64              \n",
      " 9   rating_numerator    1994 non-null   int32              \n",
      "dtypes: datetime64[ns, UTC](1), int32(1), int64(2), object(6)\n",
      "memory usage: 163.6+ KB\n"
     ]
    }
   ],
   "source": [
    "twitter_copy.info()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**8 - Define:**\n",
    "We need to merge the 3 datasets in order to make our EDA (Data Viz & Insight) easier to compute."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**Code:**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 192,
   "metadata": {},
   "outputs": [],
   "source": [
    "merged_df = pd.merge(twitter_copy, img_copy, on='tweet_id')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 193,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>tweet_id</th>\n",
       "      <th>timestamp</th>\n",
       "      <th>source</th>\n",
       "      <th>text</th>\n",
       "      <th>expanded_urls</th>\n",
       "      <th>name</th>\n",
       "      <th>Date</th>\n",
       "      <th>stage</th>\n",
       "      <th>rating_denominator</th>\n",
       "      <th>rating_numerator</th>\n",
       "      <th>...</th>\n",
       "      <th>img_num</th>\n",
       "      <th>p1</th>\n",
       "      <th>p1_conf</th>\n",
       "      <th>p1_dog</th>\n",
       "      <th>p2</th>\n",
       "      <th>p2_conf</th>\n",
       "      <th>p2_dog</th>\n",
       "      <th>p3</th>\n",
       "      <th>p3_conf</th>\n",
       "      <th>p3_dog</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>892420643555336193</td>\n",
       "      <td>2017-08-01 16:23:56+00:00</td>\n",
       "      <td>Twitter for iPhone</td>\n",
       "      <td>This is Phineas. He's a mystical boy. Only eve...</td>\n",
       "      <td>https://twitter.com/dog_rates/status/892420643...</td>\n",
       "      <td>Phineas</td>\n",
       "      <td>2017-08-01</td>\n",
       "      <td>None</td>\n",
       "      <td>10</td>\n",
       "      <td>13</td>\n",
       "      <td>...</td>\n",
       "      <td>1</td>\n",
       "      <td>orange</td>\n",
       "      <td>0.097049</td>\n",
       "      <td>False</td>\n",
       "      <td>bagel</td>\n",
       "      <td>0.085851</td>\n",
       "      <td>False</td>\n",
       "      <td>banana</td>\n",
       "      <td>0.076110</td>\n",
       "      <td>False</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>892177421306343426</td>\n",
       "      <td>2017-08-01 00:17:27+00:00</td>\n",
       "      <td>Twitter for iPhone</td>\n",
       "      <td>This is Tilly. She's just checking pup on you....</td>\n",
       "      <td>https://twitter.com/dog_rates/status/892177421...</td>\n",
       "      <td>Tilly</td>\n",
       "      <td>2017-08-01</td>\n",
       "      <td>None</td>\n",
       "      <td>10</td>\n",
       "      <td>13</td>\n",
       "      <td>...</td>\n",
       "      <td>1</td>\n",
       "      <td>Chihuahua</td>\n",
       "      <td>0.323581</td>\n",
       "      <td>True</td>\n",
       "      <td>Pekinese</td>\n",
       "      <td>0.090647</td>\n",
       "      <td>True</td>\n",
       "      <td>papillon</td>\n",
       "      <td>0.068957</td>\n",
       "      <td>True</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>891815181378084864</td>\n",
       "      <td>2017-07-31 00:18:03+00:00</td>\n",
       "      <td>Twitter for iPhone</td>\n",
       "      <td>This is Archie. He is a rare Norwegian Pouncin...</td>\n",
       "      <td>https://twitter.com/dog_rates/status/891815181...</td>\n",
       "      <td>Archie</td>\n",
       "      <td>2017-07-31</td>\n",
       "      <td>None</td>\n",
       "      <td>10</td>\n",
       "      <td>12</td>\n",
       "      <td>...</td>\n",
       "      <td>1</td>\n",
       "      <td>Chihuahua</td>\n",
       "      <td>0.716012</td>\n",
       "      <td>True</td>\n",
       "      <td>malamute</td>\n",
       "      <td>0.078253</td>\n",
       "      <td>True</td>\n",
       "      <td>kelpie</td>\n",
       "      <td>0.031379</td>\n",
       "      <td>True</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>891689557279858688</td>\n",
       "      <td>2017-07-30 15:58:51+00:00</td>\n",
       "      <td>Twitter for iPhone</td>\n",
       "      <td>This is Darla. She commenced a snooze mid meal...</td>\n",
       "      <td>https://twitter.com/dog_rates/status/891689557...</td>\n",
       "      <td>Darla</td>\n",
       "      <td>2017-07-30</td>\n",
       "      <td>None</td>\n",
       "      <td>10</td>\n",
       "      <td>13</td>\n",
       "      <td>...</td>\n",
       "      <td>1</td>\n",
       "      <td>paper_towel</td>\n",
       "      <td>0.170278</td>\n",
       "      <td>False</td>\n",
       "      <td>Labrador_retriever</td>\n",
       "      <td>0.168086</td>\n",
       "      <td>True</td>\n",
       "      <td>spatula</td>\n",
       "      <td>0.040836</td>\n",
       "      <td>False</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>891327558926688256</td>\n",
       "      <td>2017-07-29 16:00:24+00:00</td>\n",
       "      <td>Twitter for iPhone</td>\n",
       "      <td>This is Franklin. He would like you to stop ca...</td>\n",
       "      <td>https://twitter.com/dog_rates/status/891327558...</td>\n",
       "      <td>Franklin</td>\n",
       "      <td>2017-07-29</td>\n",
       "      <td>None</td>\n",
       "      <td>10</td>\n",
       "      <td>12</td>\n",
       "      <td>...</td>\n",
       "      <td>2</td>\n",
       "      <td>basset</td>\n",
       "      <td>0.555712</td>\n",
       "      <td>True</td>\n",
       "      <td>English_springer</td>\n",
       "      <td>0.225770</td>\n",
       "      <td>True</td>\n",
       "      <td>German_short-haired_pointer</td>\n",
       "      <td>0.175219</td>\n",
       "      <td>True</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>5 rows × 21 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "             tweet_id                 timestamp              source  \\\n",
       "0  892420643555336193 2017-08-01 16:23:56+00:00  Twitter for iPhone   \n",
       "1  892177421306343426 2017-08-01 00:17:27+00:00  Twitter for iPhone   \n",
       "2  891815181378084864 2017-07-31 00:18:03+00:00  Twitter for iPhone   \n",
       "3  891689557279858688 2017-07-30 15:58:51+00:00  Twitter for iPhone   \n",
       "4  891327558926688256 2017-07-29 16:00:24+00:00  Twitter for iPhone   \n",
       "\n",
       "                                                text  \\\n",
       "0  This is Phineas. He's a mystical boy. Only eve...   \n",
       "1  This is Tilly. She's just checking pup on you....   \n",
       "2  This is Archie. He is a rare Norwegian Pouncin...   \n",
       "3  This is Darla. She commenced a snooze mid meal...   \n",
       "4  This is Franklin. He would like you to stop ca...   \n",
       "\n",
       "                                       expanded_urls      name        Date  \\\n",
       "0  https://twitter.com/dog_rates/status/892420643...   Phineas  2017-08-01   \n",
       "1  https://twitter.com/dog_rates/status/892177421...     Tilly  2017-08-01   \n",
       "2  https://twitter.com/dog_rates/status/891815181...    Archie  2017-07-31   \n",
       "3  https://twitter.com/dog_rates/status/891689557...     Darla  2017-07-30   \n",
       "4  https://twitter.com/dog_rates/status/891327558...  Franklin  2017-07-29   \n",
       "\n",
       "  stage  rating_denominator  rating_numerator  ... img_num           p1  \\\n",
       "0  None                  10                13  ...       1       orange   \n",
       "1  None                  10                13  ...       1    Chihuahua   \n",
       "2  None                  10                12  ...       1    Chihuahua   \n",
       "3  None                  10                13  ...       1  paper_towel   \n",
       "4  None                  10                12  ...       2       basset   \n",
       "\n",
       "    p1_conf  p1_dog                  p2   p2_conf  p2_dog  \\\n",
       "0  0.097049   False               bagel  0.085851   False   \n",
       "1  0.323581    True            Pekinese  0.090647    True   \n",
       "2  0.716012    True            malamute  0.078253    True   \n",
       "3  0.170278   False  Labrador_retriever  0.168086    True   \n",
       "4  0.555712    True    English_springer  0.225770    True   \n",
       "\n",
       "                            p3   p3_conf  p3_dog  \n",
       "0                       banana  0.076110   False  \n",
       "1                     papillon  0.068957    True  \n",
       "2                       kelpie  0.031379    True  \n",
       "3                      spatula  0.040836   False  \n",
       "4  German_short-haired_pointer  0.175219    True  \n",
       "\n",
       "[5 rows x 21 columns]"
      ]
     },
     "execution_count": 193,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "merged_df.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 194,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>tweet_id</th>\n",
       "      <th>timestamp</th>\n",
       "      <th>source</th>\n",
       "      <th>text</th>\n",
       "      <th>expanded_urls</th>\n",
       "      <th>name</th>\n",
       "      <th>Date</th>\n",
       "      <th>stage</th>\n",
       "      <th>rating_denominator</th>\n",
       "      <th>rating_numerator</th>\n",
       "      <th>...</th>\n",
       "      <th>p1_dog</th>\n",
       "      <th>p2</th>\n",
       "      <th>p2_conf</th>\n",
       "      <th>p2_dog</th>\n",
       "      <th>p3</th>\n",
       "      <th>p3_conf</th>\n",
       "      <th>p3_dog</th>\n",
       "      <th>id</th>\n",
       "      <th>favorite_count</th>\n",
       "      <th>retweet_count</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>892420643555336193</td>\n",
       "      <td>2017-08-01 16:23:56+00:00</td>\n",
       "      <td>Twitter for iPhone</td>\n",
       "      <td>This is Phineas. He's a mystical boy. Only eve...</td>\n",
       "      <td>https://twitter.com/dog_rates/status/892420643...</td>\n",
       "      <td>Phineas</td>\n",
       "      <td>2017-08-01</td>\n",
       "      <td>None</td>\n",
       "      <td>10</td>\n",
       "      <td>13</td>\n",
       "      <td>...</td>\n",
       "      <td>False</td>\n",
       "      <td>bagel</td>\n",
       "      <td>0.085851</td>\n",
       "      <td>False</td>\n",
       "      <td>banana</td>\n",
       "      <td>0.076110</td>\n",
       "      <td>False</td>\n",
       "      <td>892420643555336193</td>\n",
       "      <td>35686</td>\n",
       "      <td>7547</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>892177421306343426</td>\n",
       "      <td>2017-08-01 00:17:27+00:00</td>\n",
       "      <td>Twitter for iPhone</td>\n",
       "      <td>This is Tilly. She's just checking pup on you....</td>\n",
       "      <td>https://twitter.com/dog_rates/status/892177421...</td>\n",
       "      <td>Tilly</td>\n",
       "      <td>2017-08-01</td>\n",
       "      <td>None</td>\n",
       "      <td>10</td>\n",
       "      <td>13</td>\n",
       "      <td>...</td>\n",
       "      <td>True</td>\n",
       "      <td>Pekinese</td>\n",
       "      <td>0.090647</td>\n",
       "      <td>True</td>\n",
       "      <td>papillon</td>\n",
       "      <td>0.068957</td>\n",
       "      <td>True</td>\n",
       "      <td>892177421306343426</td>\n",
       "      <td>30869</td>\n",
       "      <td>5595</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>891815181378084864</td>\n",
       "      <td>2017-07-31 00:18:03+00:00</td>\n",
       "      <td>Twitter for iPhone</td>\n",
       "      <td>This is Archie. He is a rare Norwegian Pouncin...</td>\n",
       "      <td>https://twitter.com/dog_rates/status/891815181...</td>\n",
       "      <td>Archie</td>\n",
       "      <td>2017-07-31</td>\n",
       "      <td>None</td>\n",
       "      <td>10</td>\n",
       "      <td>12</td>\n",
       "      <td>...</td>\n",
       "      <td>True</td>\n",
       "      <td>malamute</td>\n",
       "      <td>0.078253</td>\n",
       "      <td>True</td>\n",
       "      <td>kelpie</td>\n",
       "      <td>0.031379</td>\n",
       "      <td>True</td>\n",
       "      <td>891815181378084864</td>\n",
       "      <td>23206</td>\n",
       "      <td>3707</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>891689557279858688</td>\n",
       "      <td>2017-07-30 15:58:51+00:00</td>\n",
       "      <td>Twitter for iPhone</td>\n",
       "      <td>This is Darla. She commenced a snooze mid meal...</td>\n",
       "      <td>https://twitter.com/dog_rates/status/891689557...</td>\n",
       "      <td>Darla</td>\n",
       "      <td>2017-07-30</td>\n",
       "      <td>None</td>\n",
       "      <td>10</td>\n",
       "      <td>13</td>\n",
       "      <td>...</td>\n",
       "      <td>False</td>\n",
       "      <td>Labrador_retriever</td>\n",
       "      <td>0.168086</td>\n",
       "      <td>True</td>\n",
       "      <td>spatula</td>\n",
       "      <td>0.040836</td>\n",
       "      <td>False</td>\n",
       "      <td>891689557279858688</td>\n",
       "      <td>38988</td>\n",
       "      <td>7728</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>891327558926688256</td>\n",
       "      <td>2017-07-29 16:00:24+00:00</td>\n",
       "      <td>Twitter for iPhone</td>\n",
       "      <td>This is Franklin. He would like you to stop ca...</td>\n",
       "      <td>https://twitter.com/dog_rates/status/891327558...</td>\n",
       "      <td>Franklin</td>\n",
       "      <td>2017-07-29</td>\n",
       "      <td>None</td>\n",
       "      <td>10</td>\n",
       "      <td>12</td>\n",
       "      <td>...</td>\n",
       "      <td>True</td>\n",
       "      <td>English_springer</td>\n",
       "      <td>0.225770</td>\n",
       "      <td>True</td>\n",
       "      <td>German_short-haired_pointer</td>\n",
       "      <td>0.175219</td>\n",
       "      <td>True</td>\n",
       "      <td>891327558926688256</td>\n",
       "      <td>37239</td>\n",
       "      <td>8329</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>5 rows × 24 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "             tweet_id                 timestamp              source  \\\n",
       "0  892420643555336193 2017-08-01 16:23:56+00:00  Twitter for iPhone   \n",
       "1  892177421306343426 2017-08-01 00:17:27+00:00  Twitter for iPhone   \n",
       "2  891815181378084864 2017-07-31 00:18:03+00:00  Twitter for iPhone   \n",
       "3  891689557279858688 2017-07-30 15:58:51+00:00  Twitter for iPhone   \n",
       "4  891327558926688256 2017-07-29 16:00:24+00:00  Twitter for iPhone   \n",
       "\n",
       "                                                text  \\\n",
       "0  This is Phineas. He's a mystical boy. Only eve...   \n",
       "1  This is Tilly. She's just checking pup on you....   \n",
       "2  This is Archie. He is a rare Norwegian Pouncin...   \n",
       "3  This is Darla. She commenced a snooze mid meal...   \n",
       "4  This is Franklin. He would like you to stop ca...   \n",
       "\n",
       "                                       expanded_urls      name        Date  \\\n",
       "0  https://twitter.com/dog_rates/status/892420643...   Phineas  2017-08-01   \n",
       "1  https://twitter.com/dog_rates/status/892177421...     Tilly  2017-08-01   \n",
       "2  https://twitter.com/dog_rates/status/891815181...    Archie  2017-07-31   \n",
       "3  https://twitter.com/dog_rates/status/891689557...     Darla  2017-07-30   \n",
       "4  https://twitter.com/dog_rates/status/891327558...  Franklin  2017-07-29   \n",
       "\n",
       "  stage  rating_denominator  rating_numerator  ... p1_dog                  p2  \\\n",
       "0  None                  10                13  ...  False               bagel   \n",
       "1  None                  10                13  ...   True            Pekinese   \n",
       "2  None                  10                12  ...   True            malamute   \n",
       "3  None                  10                13  ...  False  Labrador_retriever   \n",
       "4  None                  10                12  ...   True    English_springer   \n",
       "\n",
       "    p2_conf  p2_dog                           p3   p3_conf  p3_dog  \\\n",
       "0  0.085851   False                       banana  0.076110   False   \n",
       "1  0.090647    True                     papillon  0.068957    True   \n",
       "2  0.078253    True                       kelpie  0.031379    True   \n",
       "3  0.168086    True                      spatula  0.040836   False   \n",
       "4  0.225770    True  German_short-haired_pointer  0.175219    True   \n",
       "\n",
       "                   id favorite_count  retweet_count  \n",
       "0  892420643555336193          35686           7547  \n",
       "1  892177421306343426          30869           5595  \n",
       "2  891815181378084864          23206           3707  \n",
       "3  891689557279858688          38988           7728  \n",
       "4  891327558926688256          37239           8329  \n",
       "\n",
       "[5 rows x 24 columns]"
      ]
     },
     "execution_count": 194,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "merged_df = pd.merge(merged_df, api_copy, left_on='tweet_id', right_on='id')\n",
    "merged_df.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 195,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "<class 'pandas.core.frame.DataFrame'>\n",
      "Int64Index: 1987 entries, 0 to 1986\n",
      "Data columns (total 24 columns):\n",
      " #   Column              Non-Null Count  Dtype              \n",
      "---  ------              --------------  -----              \n",
      " 0   tweet_id            1987 non-null   int64              \n",
      " 1   timestamp           1987 non-null   datetime64[ns, UTC]\n",
      " 2   source              1987 non-null   object             \n",
      " 3   text                1987 non-null   object             \n",
      " 4   expanded_urls       1987 non-null   object             \n",
      " 5   name                1987 non-null   object             \n",
      " 6   Date                1987 non-null   object             \n",
      " 7   stage               1987 non-null   object             \n",
      " 8   rating_denominator  1987 non-null   int64              \n",
      " 9   rating_numerator    1987 non-null   int32              \n",
      " 10  jpg_url             1987 non-null   object             \n",
      " 11  img_num             1987 non-null   int64              \n",
      " 12  p1                  1987 non-null   object             \n",
      " 13  p1_conf             1987 non-null   float64            \n",
      " 14  p1_dog              1987 non-null   bool               \n",
      " 15  p2                  1987 non-null   object             \n",
      " 16  p2_conf             1987 non-null   float64            \n",
      " 17  p2_dog              1987 non-null   bool               \n",
      " 18  p3                  1987 non-null   object             \n",
      " 19  p3_conf             1987 non-null   float64            \n",
      " 20  p3_dog              1987 non-null   bool               \n",
      " 21  id                  1987 non-null   int64              \n",
      " 22  favorite_count      1987 non-null   int64              \n",
      " 23  retweet_count       1987 non-null   int64              \n",
      "dtypes: bool(3), datetime64[ns, UTC](1), float64(3), int32(1), int64(6), object(10)\n",
      "memory usage: 339.6+ KB\n"
     ]
    }
   ],
   "source": [
    "merged_df.info()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 196,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(1987, 24)"
      ]
     },
     "execution_count": 196,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "merged_df.shape"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**9 - Define:**\n",
    "Improve the readability of the predictions"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**Code:**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 197,
   "metadata": {},
   "outputs": [],
   "source": [
    "# store the true algo with it's level of confidence and if false put NaN as confidence\n",
    "prediction = []\n",
    "confidence = []\n",
    "\n",
    "def get_prediction_confidence(dataframe):\n",
    "    if dataframe['p1_dog'] == True:\n",
    "        prediction.append(dataframe['p1'])\n",
    "        confidence.append(dataframe['p1_conf'])\n",
    "    elif dataframe['p2_dog'] == True:\n",
    "        prediction.append(dataframe['p2'])\n",
    "        confidence.append(dataframe['p2_conf'])\n",
    "    elif dataframe['p3_dog'] == True:\n",
    "        prediction.append(dataframe['p3'])\n",
    "        confidence.append(dataframe['p3_conf'])\n",
    "    else:\n",
    "        prediction.append(np.nan)\n",
    "        confidence.append(0)\n",
    "\n",
    "merged_df.apply(get_prediction_confidence, axis=1)\n",
    "merged_df['breed'] = prediction\n",
    "merged_df['confidence'] = confidence"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 198,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>tweet_id</th>\n",
       "      <th>timestamp</th>\n",
       "      <th>source</th>\n",
       "      <th>text</th>\n",
       "      <th>expanded_urls</th>\n",
       "      <th>name</th>\n",
       "      <th>Date</th>\n",
       "      <th>stage</th>\n",
       "      <th>rating_denominator</th>\n",
       "      <th>rating_numerator</th>\n",
       "      <th>...</th>\n",
       "      <th>p2_conf</th>\n",
       "      <th>p2_dog</th>\n",
       "      <th>p3</th>\n",
       "      <th>p3_conf</th>\n",
       "      <th>p3_dog</th>\n",
       "      <th>id</th>\n",
       "      <th>favorite_count</th>\n",
       "      <th>retweet_count</th>\n",
       "      <th>breed</th>\n",
       "      <th>confidence</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>892420643555336193</td>\n",
       "      <td>2017-08-01 16:23:56+00:00</td>\n",
       "      <td>Twitter for iPhone</td>\n",
       "      <td>This is Phineas. He's a mystical boy. Only eve...</td>\n",
       "      <td>https://twitter.com/dog_rates/status/892420643...</td>\n",
       "      <td>Phineas</td>\n",
       "      <td>2017-08-01</td>\n",
       "      <td>None</td>\n",
       "      <td>10</td>\n",
       "      <td>13</td>\n",
       "      <td>...</td>\n",
       "      <td>0.085851</td>\n",
       "      <td>False</td>\n",
       "      <td>banana</td>\n",
       "      <td>0.076110</td>\n",
       "      <td>False</td>\n",
       "      <td>892420643555336193</td>\n",
       "      <td>35686</td>\n",
       "      <td>7547</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>892177421306343426</td>\n",
       "      <td>2017-08-01 00:17:27+00:00</td>\n",
       "      <td>Twitter for iPhone</td>\n",
       "      <td>This is Tilly. She's just checking pup on you....</td>\n",
       "      <td>https://twitter.com/dog_rates/status/892177421...</td>\n",
       "      <td>Tilly</td>\n",
       "      <td>2017-08-01</td>\n",
       "      <td>None</td>\n",
       "      <td>10</td>\n",
       "      <td>13</td>\n",
       "      <td>...</td>\n",
       "      <td>0.090647</td>\n",
       "      <td>True</td>\n",
       "      <td>papillon</td>\n",
       "      <td>0.068957</td>\n",
       "      <td>True</td>\n",
       "      <td>892177421306343426</td>\n",
       "      <td>30869</td>\n",
       "      <td>5595</td>\n",
       "      <td>Chihuahua</td>\n",
       "      <td>0.323581</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>891815181378084864</td>\n",
       "      <td>2017-07-31 00:18:03+00:00</td>\n",
       "      <td>Twitter for iPhone</td>\n",
       "      <td>This is Archie. He is a rare Norwegian Pouncin...</td>\n",
       "      <td>https://twitter.com/dog_rates/status/891815181...</td>\n",
       "      <td>Archie</td>\n",
       "      <td>2017-07-31</td>\n",
       "      <td>None</td>\n",
       "      <td>10</td>\n",
       "      <td>12</td>\n",
       "      <td>...</td>\n",
       "      <td>0.078253</td>\n",
       "      <td>True</td>\n",
       "      <td>kelpie</td>\n",
       "      <td>0.031379</td>\n",
       "      <td>True</td>\n",
       "      <td>891815181378084864</td>\n",
       "      <td>23206</td>\n",
       "      <td>3707</td>\n",
       "      <td>Chihuahua</td>\n",
       "      <td>0.716012</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>891689557279858688</td>\n",
       "      <td>2017-07-30 15:58:51+00:00</td>\n",
       "      <td>Twitter for iPhone</td>\n",
       "      <td>This is Darla. She commenced a snooze mid meal...</td>\n",
       "      <td>https://twitter.com/dog_rates/status/891689557...</td>\n",
       "      <td>Darla</td>\n",
       "      <td>2017-07-30</td>\n",
       "      <td>None</td>\n",
       "      <td>10</td>\n",
       "      <td>13</td>\n",
       "      <td>...</td>\n",
       "      <td>0.168086</td>\n",
       "      <td>True</td>\n",
       "      <td>spatula</td>\n",
       "      <td>0.040836</td>\n",
       "      <td>False</td>\n",
       "      <td>891689557279858688</td>\n",
       "      <td>38988</td>\n",
       "      <td>7728</td>\n",
       "      <td>Labrador_retriever</td>\n",
       "      <td>0.168086</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>891327558926688256</td>\n",
       "      <td>2017-07-29 16:00:24+00:00</td>\n",
       "      <td>Twitter for iPhone</td>\n",
       "      <td>This is Franklin. He would like you to stop ca...</td>\n",
       "      <td>https://twitter.com/dog_rates/status/891327558...</td>\n",
       "      <td>Franklin</td>\n",
       "      <td>2017-07-29</td>\n",
       "      <td>None</td>\n",
       "      <td>10</td>\n",
       "      <td>12</td>\n",
       "      <td>...</td>\n",
       "      <td>0.225770</td>\n",
       "      <td>True</td>\n",
       "      <td>German_short-haired_pointer</td>\n",
       "      <td>0.175219</td>\n",
       "      <td>True</td>\n",
       "      <td>891327558926688256</td>\n",
       "      <td>37239</td>\n",
       "      <td>8329</td>\n",
       "      <td>basset</td>\n",
       "      <td>0.555712</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>891087950875897856</td>\n",
       "      <td>2017-07-29 00:08:17+00:00</td>\n",
       "      <td>Twitter for iPhone</td>\n",
       "      <td>Here we have a majestic great white breaching ...</td>\n",
       "      <td>https://twitter.com/dog_rates/status/891087950...</td>\n",
       "      <td>None</td>\n",
       "      <td>2017-07-29</td>\n",
       "      <td>None</td>\n",
       "      <td>10</td>\n",
       "      <td>13</td>\n",
       "      <td>...</td>\n",
       "      <td>0.116317</td>\n",
       "      <td>True</td>\n",
       "      <td>Indian_elephant</td>\n",
       "      <td>0.076902</td>\n",
       "      <td>False</td>\n",
       "      <td>891087950875897856</td>\n",
       "      <td>18775</td>\n",
       "      <td>2795</td>\n",
       "      <td>Chesapeake_Bay_retriever</td>\n",
       "      <td>0.425595</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>890971913173991426</td>\n",
       "      <td>2017-07-28 16:27:12+00:00</td>\n",
       "      <td>Twitter for iPhone</td>\n",
       "      <td>Meet Jax. He enjoys ice cream so much he gets ...</td>\n",
       "      <td>https://gofundme.com/ydvmve-surgery-for-jax,ht...</td>\n",
       "      <td>Jax</td>\n",
       "      <td>2017-07-28</td>\n",
       "      <td>None</td>\n",
       "      <td>10</td>\n",
       "      <td>13</td>\n",
       "      <td>...</td>\n",
       "      <td>0.199287</td>\n",
       "      <td>True</td>\n",
       "      <td>ice_lolly</td>\n",
       "      <td>0.193548</td>\n",
       "      <td>False</td>\n",
       "      <td>890971913173991426</td>\n",
       "      <td>10907</td>\n",
       "      <td>1813</td>\n",
       "      <td>Appenzeller</td>\n",
       "      <td>0.341703</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>890729181411237888</td>\n",
       "      <td>2017-07-28 00:22:40+00:00</td>\n",
       "      <td>Twitter for iPhone</td>\n",
       "      <td>When you watch your owner call another dog a g...</td>\n",
       "      <td>https://twitter.com/dog_rates/status/890729181...</td>\n",
       "      <td>None</td>\n",
       "      <td>2017-07-28</td>\n",
       "      <td>None</td>\n",
       "      <td>10</td>\n",
       "      <td>13</td>\n",
       "      <td>...</td>\n",
       "      <td>0.178406</td>\n",
       "      <td>True</td>\n",
       "      <td>Pembroke</td>\n",
       "      <td>0.076507</td>\n",
       "      <td>True</td>\n",
       "      <td>890729181411237888</td>\n",
       "      <td>60147</td>\n",
       "      <td>16885</td>\n",
       "      <td>Pomeranian</td>\n",
       "      <td>0.566142</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8</th>\n",
       "      <td>890609185150312448</td>\n",
       "      <td>2017-07-27 16:25:51+00:00</td>\n",
       "      <td>Twitter for iPhone</td>\n",
       "      <td>This is Zoey. She doesn't want to be one of th...</td>\n",
       "      <td>https://twitter.com/dog_rates/status/890609185...</td>\n",
       "      <td>Zoey</td>\n",
       "      <td>2017-07-27</td>\n",
       "      <td>None</td>\n",
       "      <td>10</td>\n",
       "      <td>13</td>\n",
       "      <td>...</td>\n",
       "      <td>0.193054</td>\n",
       "      <td>True</td>\n",
       "      <td>Chesapeake_Bay_retriever</td>\n",
       "      <td>0.118184</td>\n",
       "      <td>True</td>\n",
       "      <td>890609185150312448</td>\n",
       "      <td>25827</td>\n",
       "      <td>3849</td>\n",
       "      <td>Irish_terrier</td>\n",
       "      <td>0.487574</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9</th>\n",
       "      <td>890240255349198849</td>\n",
       "      <td>2017-07-26 15:59:51+00:00</td>\n",
       "      <td>Twitter for iPhone</td>\n",
       "      <td>This is Cassie. She is a college pup. Studying...</td>\n",
       "      <td>https://twitter.com/dog_rates/status/890240255...</td>\n",
       "      <td>Cassie</td>\n",
       "      <td>2017-07-26</td>\n",
       "      <td>doggo</td>\n",
       "      <td>10</td>\n",
       "      <td>14</td>\n",
       "      <td>...</td>\n",
       "      <td>0.451038</td>\n",
       "      <td>True</td>\n",
       "      <td>Chihuahua</td>\n",
       "      <td>0.029248</td>\n",
       "      <td>True</td>\n",
       "      <td>890240255349198849</td>\n",
       "      <td>29495</td>\n",
       "      <td>6566</td>\n",
       "      <td>Pembroke</td>\n",
       "      <td>0.511319</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>10 rows × 26 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "             tweet_id                 timestamp              source  \\\n",
       "0  892420643555336193 2017-08-01 16:23:56+00:00  Twitter for iPhone   \n",
       "1  892177421306343426 2017-08-01 00:17:27+00:00  Twitter for iPhone   \n",
       "2  891815181378084864 2017-07-31 00:18:03+00:00  Twitter for iPhone   \n",
       "3  891689557279858688 2017-07-30 15:58:51+00:00  Twitter for iPhone   \n",
       "4  891327558926688256 2017-07-29 16:00:24+00:00  Twitter for iPhone   \n",
       "5  891087950875897856 2017-07-29 00:08:17+00:00  Twitter for iPhone   \n",
       "6  890971913173991426 2017-07-28 16:27:12+00:00  Twitter for iPhone   \n",
       "7  890729181411237888 2017-07-28 00:22:40+00:00  Twitter for iPhone   \n",
       "8  890609185150312448 2017-07-27 16:25:51+00:00  Twitter for iPhone   \n",
       "9  890240255349198849 2017-07-26 15:59:51+00:00  Twitter for iPhone   \n",
       "\n",
       "                                                text  \\\n",
       "0  This is Phineas. He's a mystical boy. Only eve...   \n",
       "1  This is Tilly. She's just checking pup on you....   \n",
       "2  This is Archie. He is a rare Norwegian Pouncin...   \n",
       "3  This is Darla. She commenced a snooze mid meal...   \n",
       "4  This is Franklin. He would like you to stop ca...   \n",
       "5  Here we have a majestic great white breaching ...   \n",
       "6  Meet Jax. He enjoys ice cream so much he gets ...   \n",
       "7  When you watch your owner call another dog a g...   \n",
       "8  This is Zoey. She doesn't want to be one of th...   \n",
       "9  This is Cassie. She is a college pup. Studying...   \n",
       "\n",
       "                                       expanded_urls      name        Date  \\\n",
       "0  https://twitter.com/dog_rates/status/892420643...   Phineas  2017-08-01   \n",
       "1  https://twitter.com/dog_rates/status/892177421...     Tilly  2017-08-01   \n",
       "2  https://twitter.com/dog_rates/status/891815181...    Archie  2017-07-31   \n",
       "3  https://twitter.com/dog_rates/status/891689557...     Darla  2017-07-30   \n",
       "4  https://twitter.com/dog_rates/status/891327558...  Franklin  2017-07-29   \n",
       "5  https://twitter.com/dog_rates/status/891087950...      None  2017-07-29   \n",
       "6  https://gofundme.com/ydvmve-surgery-for-jax,ht...       Jax  2017-07-28   \n",
       "7  https://twitter.com/dog_rates/status/890729181...      None  2017-07-28   \n",
       "8  https://twitter.com/dog_rates/status/890609185...      Zoey  2017-07-27   \n",
       "9  https://twitter.com/dog_rates/status/890240255...    Cassie  2017-07-26   \n",
       "\n",
       "   stage  rating_denominator  rating_numerator  ...   p2_conf  p2_dog  \\\n",
       "0   None                  10                13  ...  0.085851   False   \n",
       "1   None                  10                13  ...  0.090647    True   \n",
       "2   None                  10                12  ...  0.078253    True   \n",
       "3   None                  10                13  ...  0.168086    True   \n",
       "4   None                  10                12  ...  0.225770    True   \n",
       "5   None                  10                13  ...  0.116317    True   \n",
       "6   None                  10                13  ...  0.199287    True   \n",
       "7   None                  10                13  ...  0.178406    True   \n",
       "8   None                  10                13  ...  0.193054    True   \n",
       "9  doggo                  10                14  ...  0.451038    True   \n",
       "\n",
       "                            p3   p3_conf  p3_dog                  id  \\\n",
       "0                       banana  0.076110   False  892420643555336193   \n",
       "1                     papillon  0.068957    True  892177421306343426   \n",
       "2                       kelpie  0.031379    True  891815181378084864   \n",
       "3                      spatula  0.040836   False  891689557279858688   \n",
       "4  German_short-haired_pointer  0.175219    True  891327558926688256   \n",
       "5              Indian_elephant  0.076902   False  891087950875897856   \n",
       "6                    ice_lolly  0.193548   False  890971913173991426   \n",
       "7                     Pembroke  0.076507    True  890729181411237888   \n",
       "8     Chesapeake_Bay_retriever  0.118184    True  890609185150312448   \n",
       "9                    Chihuahua  0.029248    True  890240255349198849   \n",
       "\n",
       "   favorite_count  retweet_count                     breed  confidence  \n",
       "0           35686           7547                       NaN    0.000000  \n",
       "1           30869           5595                 Chihuahua    0.323581  \n",
       "2           23206           3707                 Chihuahua    0.716012  \n",
       "3           38988           7728        Labrador_retriever    0.168086  \n",
       "4           37239           8329                    basset    0.555712  \n",
       "5           18775           2795  Chesapeake_Bay_retriever    0.425595  \n",
       "6           10907           1813               Appenzeller    0.341703  \n",
       "7           60147          16885                Pomeranian    0.566142  \n",
       "8           25827           3849             Irish_terrier    0.487574  \n",
       "9           29495           6566                  Pembroke    0.511319  \n",
       "\n",
       "[10 rows x 26 columns]"
      ]
     },
     "execution_count": 198,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "merged_df.head(10)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 199,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "308"
      ]
     },
     "execution_count": 199,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "sum(merged_df.breed.isnull())"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**Test:**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 200,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "<class 'pandas.core.frame.DataFrame'>\n",
      "Int64Index: 1987 entries, 0 to 1986\n",
      "Data columns (total 26 columns):\n",
      " #   Column              Non-Null Count  Dtype              \n",
      "---  ------              --------------  -----              \n",
      " 0   tweet_id            1987 non-null   int64              \n",
      " 1   timestamp           1987 non-null   datetime64[ns, UTC]\n",
      " 2   source              1987 non-null   object             \n",
      " 3   text                1987 non-null   object             \n",
      " 4   expanded_urls       1987 non-null   object             \n",
      " 5   name                1987 non-null   object             \n",
      " 6   Date                1987 non-null   object             \n",
      " 7   stage               1987 non-null   object             \n",
      " 8   rating_denominator  1987 non-null   int64              \n",
      " 9   rating_numerator    1987 non-null   int32              \n",
      " 10  jpg_url             1987 non-null   object             \n",
      " 11  img_num             1987 non-null   int64              \n",
      " 12  p1                  1987 non-null   object             \n",
      " 13  p1_conf             1987 non-null   float64            \n",
      " 14  p1_dog              1987 non-null   bool               \n",
      " 15  p2                  1987 non-null   object             \n",
      " 16  p2_conf             1987 non-null   float64            \n",
      " 17  p2_dog              1987 non-null   bool               \n",
      " 18  p3                  1987 non-null   object             \n",
      " 19  p3_conf             1987 non-null   float64            \n",
      " 20  p3_dog              1987 non-null   bool               \n",
      " 21  id                  1987 non-null   int64              \n",
      " 22  favorite_count      1987 non-null   int64              \n",
      " 23  retweet_count       1987 non-null   int64              \n",
      " 24  breed               1679 non-null   object             \n",
      " 25  confidence          1987 non-null   float64            \n",
      "dtypes: bool(3), datetime64[ns, UTC](1), float64(4), int32(1), int64(6), object(11)\n",
      "memory usage: 370.6+ KB\n"
     ]
    }
   ],
   "source": [
    "merged_df.info()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**10 - Define:** Let's clean our dataset to have just useful columns for our analysis"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**Code:**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 201,
   "metadata": {},
   "outputs": [],
   "source": [
    "merged_df.drop(['p1', 'p1_conf', 'p1_dog', 'p2', 'p2_conf', 'p2_dog', 'p3', 'p3_conf', 'p3_dog', 'id', 'img_num', 'jpg_url'], axis=1, inplace=True)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**Test:**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 202,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "<class 'pandas.core.frame.DataFrame'>\n",
      "Int64Index: 1987 entries, 0 to 1986\n",
      "Data columns (total 14 columns):\n",
      " #   Column              Non-Null Count  Dtype              \n",
      "---  ------              --------------  -----              \n",
      " 0   tweet_id            1987 non-null   int64              \n",
      " 1   timestamp           1987 non-null   datetime64[ns, UTC]\n",
      " 2   source              1987 non-null   object             \n",
      " 3   text                1987 non-null   object             \n",
      " 4   expanded_urls       1987 non-null   object             \n",
      " 5   name                1987 non-null   object             \n",
      " 6   Date                1987 non-null   object             \n",
      " 7   stage               1987 non-null   object             \n",
      " 8   rating_denominator  1987 non-null   int64              \n",
      " 9   rating_numerator    1987 non-null   int32              \n",
      " 10  favorite_count      1987 non-null   int64              \n",
      " 11  retweet_count       1987 non-null   int64              \n",
      " 12  breed               1679 non-null   object             \n",
      " 13  confidence          1987 non-null   float64            \n",
      "dtypes: datetime64[ns, UTC](1), float64(1), int32(1), int64(4), object(7)\n",
      "memory usage: 225.1+ KB\n"
     ]
    }
   ],
   "source": [
    "merged_df.info()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**11 - Define:** erroneous dog names starting with lowercase characters (e.g. a, an, actually, by). We will replace all lower cases by None"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 203,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "<ipython-input-203-58ccd32c4e48>:1: SettingWithCopyWarning: \n",
      "A value is trying to be set on a copy of a slice from a DataFrame\n",
      "\n",
      "See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy\n",
      "  merged_df['name'][merged_df['name'].str.match('[a-z]+')] = 'None'\n"
     ]
    }
   ],
   "source": [
    "merged_df['name'][merged_df['name'].str.match('[a-z]+')] = 'None'"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 204,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "None    644\n",
       "Name: name, dtype: int64"
      ]
     },
     "execution_count": 204,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "merged_df.name[merged_df.name == 'None'].value_counts()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 205,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Zuzu     1\n",
       "Zooey    1\n",
       "Zoey     3\n",
       "Zoe      1\n",
       "Ziva     1\n",
       "        ..\n",
       "Aiden    1\n",
       "Adele    1\n",
       "Acro     1\n",
       "Ace      1\n",
       "Abby     2\n",
       "Name: name, Length: 912, dtype: int64"
      ]
     },
     "execution_count": 205,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Sort ascending by name to check if there are more names starting with a lowercase alphabet\n",
    "merged_df.name.value_counts().sort_index(ascending=False)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Storing the data #"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 206,
   "metadata": {},
   "outputs": [],
   "source": [
    "merged_df.to_csv('twitter_archive_master.csv', encoding='utf-8', index=False)\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# EDA and data driven Insights #"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**Plotting a basic line of the number of tweets over time will give us insights on how WeRateDogs is trending:**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 207,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "As we can see below, WeRateDogs has been losing momentum over time.\n",
      "It might be a collateral consequence of Twitter losing users:\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 720x576 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "print('As we can see below, WeRateDogs has been losing momentum over time.') \n",
    "print('It might be a collateral consequence of Twitter losing users:')\n",
    "merged_df.groupby([merged_df[\"timestamp\"].dt.year, merged_df[\"timestamp\"].dt.month]).agg(tweets=('tweet_id', 'count')).plot()\n",
    "plt.title('Number of tweets over time')\n",
    "plt.ylabel('Number of tweet')\n",
    "plt.xlabel('(Year, Month)')\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**How are retweet and favorite functions doing over time?**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 208,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "As visible retweet and favorite functions of Twitter dont seem to have followed a strong pattern:\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 720x576 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "print('As visible retweet and favorite functions of Twitter dont seem to have followed a strong pattern:')\n",
    "#plt.subplots(1, sharex=True)\n",
    "merged_df.favorite_count.groupby([merged_df[\"timestamp\"].dt.year, merged_df[\"timestamp\"].dt.month]).sum().plot()\n",
    "merged_df.retweet_count.groupby([merged_df[\"timestamp\"].dt.year, merged_df[\"timestamp\"].dt.month]).sum().plot().set_ylim(0, )\n",
    "plt.title('Retweeting and Favoriting trend over time')\n",
    "plt.ylabel('Count')\n",
    "plt.xlabel('(Year, Month)')\n",
    "plt.legend(('Retweets', 'Favorites'))\n",
    "plt.savefig('ret_fav')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**What is the most used device for Twitter?**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 209,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "We can clearly see that most people are using their cellphone device to access to their account:\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 792x360 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "print('We can clearly see that most people are using their cellphone device to access to their account:')\n",
    "merged_df.source.value_counts().plot(kind='barh',figsize=(11,5), title='Most used devices for Twitter').set_xlabel('Count of Tweet by devices');"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**Are there some stages that get more retweet and favorite counts than others?**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 210,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Doggo and Pupper seem to be getting most of the retweet and favorite:\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 864x432 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "print('Doggo and Pupper seem to be getting most of the retweet and favorite:')\n",
    "plt1 = merged_df[merged_df['stage'] != 'None'].groupby(['stage'])['retweet_count'].sum()\n",
    "plt2 = merged_df[merged_df['stage'] != 'None'].groupby(['stage'])['favorite_count'].sum()\n",
    "\n",
    "\n",
    "# Numbers of pairs of bars you want\n",
    "N = 4\n",
    "\n",
    "# Position of bars on x-axis\n",
    "ind = np.arange(N)\n",
    "\n",
    "# Figure size\n",
    "plt.figure(figsize=(12,6))\n",
    "\n",
    "# Width of a bar \n",
    "width = 0.3  \n",
    "\n",
    "# Plotting\n",
    "plt.bar(ind, plt1 , width, label='Retweet count by stage')\n",
    "plt.bar(ind + width, plt2, width, label='Favorite count by stage')\n",
    "\n",
    "plt.ylabel('Counts')\n",
    "plt.title('Retweet and Favorite counts by stage')\n",
    "\n",
    "# xticks()\n",
    "# First argument - A list of positions at which ticks should be placed\n",
    "# Second argument -  A list of labels to place at the given locations\n",
    "plt.xticks(ind + width / 2, ('doggo', 'floofer', 'pupper', 'puppo'))\n",
    "\n",
    "# Finding the best position for legends and putting it\n",
    "plt.legend(loc='best')\n",
    "plt.show()\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**What are the top 5 Breeds that get most of the favorite count?**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 211,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>breed</th>\n",
       "      <th>count</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>golden_retriever</td>\n",
       "      <td>1782411</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>Labrador_retriever</td>\n",
       "      <td>1115302</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>Pembroke</td>\n",
       "      <td>957293</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>Chihuahua</td>\n",
       "      <td>710061</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>French_bulldog</td>\n",
       "      <td>555613</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                breed    count\n",
       "0    golden_retriever  1782411\n",
       "1  Labrador_retriever  1115302\n",
       "2            Pembroke   957293\n",
       "3           Chihuahua   710061\n",
       "4      French_bulldog   555613"
      ]
     },
     "execution_count": 211,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "plt3 = merged_df.groupby(['breed'])['favorite_count'].agg(count=('favorite_count', 'sum')).sort_values(['count'], ascending = False).head(5)\n",
    "plt3.reset_index(inplace=True)\n",
    "plt3.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 212,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "We can now visually see the top 5 breeds in terms of favorite count:\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 720x576 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "#Let's plot this dataset\n",
    "print('We can now visually see the top 5 breeds in terms of favorite count:')\n",
    "g = sns.barplot(data=plt3,x ='count', y ='breed', color = \"b\")\n",
    "\n",
    "#plt.gca().invert_yaxis()\n",
    "\n",
    "# Graph features\n",
    "plt.title(\"Top 5 Breeds in Favorite Count\", fontsize = 14)\n",
    "plt.ylabel(\"Breed\", fontsize = 14)\n",
    "plt.xlabel(\"Count\", fontsize = 14)\n",
    "\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**What are the top 5 Breeds that get most of the retweet count?**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 213,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>breed</th>\n",
       "      <th>count</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>golden_retriever</td>\n",
       "      <td>511808</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>Labrador_retriever</td>\n",
       "      <td>342444</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>Pembroke</td>\n",
       "      <td>253945</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>Chihuahua</td>\n",
       "      <td>226709</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>Samoyed</td>\n",
       "      <td>167079</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                breed   count\n",
       "0    golden_retriever  511808\n",
       "1  Labrador_retriever  342444\n",
       "2            Pembroke  253945\n",
       "3           Chihuahua  226709\n",
       "4             Samoyed  167079"
      ]
     },
     "execution_count": 213,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "plt4 = merged_df.groupby(['breed'])['retweet_count'].agg(count=('retweet_count', 'sum')).sort_values(['count'], ascending = False).head(5)\n",
    "plt4.reset_index(inplace=True)\n",
    "plt4.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 214,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "We can now visually see the top 5 breeds in terms of retweet coun.\n",
      "It is actually fairly similar with the favorite count for the first 4 ones:\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 720x576 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "#Let's plot this dataset\n",
    "print('We can now visually see the top 5 breeds in terms of retweet coun.')\n",
    "print('It is actually fairly similar with the favorite count for the first 4 ones:')\n",
    "g = sns.barplot(data=plt4,x ='count', y ='breed', color = \"b\")\n",
    "\n",
    "#plt.gca().invert_yaxis()\n",
    "\n",
    "# Graph features\n",
    "plt.title(\"Top 5 Breeds in Retweet Count\", fontsize = 14)\n",
    "plt.ylabel(\"Breed\", fontsize = 14)\n",
    "plt.xlabel(\"Count\", fontsize = 14)\n",
    "\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**What are the most popular name given to dog which are shared on WeRateDogs?**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 215,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>name</th>\n",
       "      <th>count</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>Oliver</td>\n",
       "      <td>10</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>Charlie</td>\n",
       "      <td>10</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>Cooper</td>\n",
       "      <td>10</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>Tucker</td>\n",
       "      <td>9</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>Penny</td>\n",
       "      <td>9</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "      name  count\n",
       "0   Oliver     10\n",
       "1  Charlie     10\n",
       "2   Cooper     10\n",
       "3   Tucker      9\n",
       "4    Penny      9"
      ]
     },
     "execution_count": 215,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "plt5 = merged_df[(merged_df['name'] != 'None')&(merged_df['name'] != 'a')].groupby(['name'])['tweet_id'].agg(count=('tweet_id', 'count')).sort_values(['count'], ascending = False).head(10)\n",
    "plt5.reset_index(inplace=True)\n",
    "plt5.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 216,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "We can now visually see the top 10 names:\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 720x576 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "#Let's plot this dataset\n",
    "print('We can now visually see the top 10 names:')\n",
    "g = sns.barplot(data=plt5,x ='count', y ='name', color = \"b\")\n",
    "\n",
    "#plt.gca().invert_yaxis()\n",
    "\n",
    "# Graph features\n",
    "plt.title(\"Top 10 Name Count\", fontsize = 14)\n",
    "plt.ylabel(\"Name\", fontsize = 14)\n",
    "plt.xlabel(\"Count\", fontsize = 14)\n",
    "\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**Is there any correlation, relationship between tweets that will get retweet & favorite?**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 217,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "There is a strong relationship betweeb the retweets and favorites:\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 720x576 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Create a scatterplot between rating numerator and rating denominator to see their trend\n",
    "print('There is a strong relationship betweeb the retweets and favorites:')\n",
    "merged_df.plot(kind='scatter', x=\"retweet_count\", y=\"favorite_count\",alpha=0.5, color='r', s=100)\n",
    "plt.xlabel('retweets'), plt.ylabel('favorites'),\n",
    "plt.title(' Scatterplot of retweets vs favorites');"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 218,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "The correlation grid:\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "<style  type=\"text/css\" >\n",
       "    #T_ccd00beb_fe86_11ea_9efd_38d547a76cfdrow0_col0 {\n",
       "            background-color:  #b40426;\n",
       "            color:  #f1f1f1;\n",
       "        }    #T_ccd00beb_fe86_11ea_9efd_38d547a76cfdrow0_col1 {\n",
       "            background-color:  #3b4cc0;\n",
       "            color:  #f1f1f1;\n",
       "        }    #T_ccd00beb_fe86_11ea_9efd_38d547a76cfdrow1_col0 {\n",
       "            background-color:  #3b4cc0;\n",
       "            color:  #f1f1f1;\n",
       "        }    #T_ccd00beb_fe86_11ea_9efd_38d547a76cfdrow1_col1 {\n",
       "            background-color:  #b40426;\n",
       "            color:  #f1f1f1;\n",
       "        }</style><table id=\"T_ccd00beb_fe86_11ea_9efd_38d547a76cfd\" ><thead>    <tr>        <th class=\"blank level0\" ></th>        <th class=\"col_heading level0 col0\" >retweet_count</th>        <th class=\"col_heading level0 col1\" >favorite_count</th>    </tr></thead><tbody>\n",
       "                <tr>\n",
       "                        <th id=\"T_ccd00beb_fe86_11ea_9efd_38d547a76cfdlevel0_row0\" class=\"row_heading level0 row0\" >retweet_count</th>\n",
       "                        <td id=\"T_ccd00beb_fe86_11ea_9efd_38d547a76cfdrow0_col0\" class=\"data row0 col0\" >1.000000</td>\n",
       "                        <td id=\"T_ccd00beb_fe86_11ea_9efd_38d547a76cfdrow0_col1\" class=\"data row0 col1\" >0.928844</td>\n",
       "            </tr>\n",
       "            <tr>\n",
       "                        <th id=\"T_ccd00beb_fe86_11ea_9efd_38d547a76cfdlevel0_row1\" class=\"row_heading level0 row1\" >favorite_count</th>\n",
       "                        <td id=\"T_ccd00beb_fe86_11ea_9efd_38d547a76cfdrow1_col0\" class=\"data row1 col0\" >0.928844</td>\n",
       "                        <td id=\"T_ccd00beb_fe86_11ea_9efd_38d547a76cfdrow1_col1\" class=\"data row1 col1\" >1.000000</td>\n",
       "            </tr>\n",
       "    </tbody></table>"
      ],
      "text/plain": [
       "<pandas.io.formats.style.Styler at 0x242eaa70160>"
      ]
     },
     "execution_count": 218,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#Setting up the correlation grid to see the relation between retweets and favorites\n",
    "print('The correlation grid:')\n",
    "df_of_interest = merged_df[['retweet_count', 'favorite_count']]\n",
    "corr = df_of_interest.corr()\n",
    "corr.style.background_gradient(cmap='coolwarm')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**As expected after looking at the scatter plot, the correlation coefficient is very strong (0.9288)**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 92,
   "metadata": {},
   "outputs": [],
   "source": [
    "# regex to match fractions\n",
    "#pattern = \"\\s*(\\d+([.]\\d+)?([/]\\d+))\"\n",
    "\n",
    "# function which will match the above pattern and return an array of fractions, if any\n",
    "#def tokens(x):\n",
    " # return [m.group(1) for m in re.finditer(pattern, x)]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 93,
   "metadata": {},
   "outputs": [],
   "source": [
    "#Let's right a loop that will go over all the rows to change those for which the denominator is not 10:\n",
    "#for i, row in twitter_copy[twitter_copy.rating_denominator != 10].iterrows():\n",
    " #   d = row.rating_denominator\n",
    "  #  k = row.rating_numerator\n",
    "   # if d == 0:\n",
    "    #    #twitter_copy.rating_denominator.replace(d,10)\n",
    "     #   twitter_copy.set_value(i, 'rating_denominator', 10)\n",
    "    #elif d < 10 and k > 10:\n",
    "        #twitter_copy.rating_denominator.replace(d,10)\n",
    "     #   twitter_copy.set_value(i, 'rating_denominator', 10)\n",
    "    #else:\n",
    "     #   twitter_copy.rating_denominator.replace(d,10)\n",
    "      #  twitter_copy.rating_numerator.replace(k,(k*10)/d)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 94,
   "metadata": {},
   "outputs": [],
   "source": [
    "#Let's right a loop that will go over all the rows to change those for which the denominator is not 10:\n",
    "#for i, row in twitter_copy[twitter_copy.rating_denominator != 10].iterrows():\n",
    " #   d = row.rating_denominator\n",
    "  #  k = row.rating_numerator\n",
    "   # if d == 0:\n",
    "       # twitter_copy.rating_denominator.replace(d,10, inplace=True)\n",
    "    #    twitter_copy.rating_denominator.iloc[i] = 10\n",
    "    #elif d < 10 and k > 10:\n",
    "        #twitter_copy.rating_denominator.replace(d,10, inplace=True)\n",
    "     #   twitter_copy.rating_denominator.iloc[i] = 10\n",
    "    #else:\n",
    "     #  twitter_copy.rating_denominator.iloc[i] = 10\n",
    "      # twitter_copy.rating_numerator.iloc[i] = (k*10)/d\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 178,
   "metadata": {},
   "outputs": [],
   "source": [
    "#Let's write a function that will pick the first non-NaN value among the doggo, floofer, pupper and puppo columns\n",
    "#def func(x):\n",
    "#    if x.values[0] is None:\n",
    "#       return None \n",
    "#    else:        \n",
    "#       return twitter_copy.loc[x.name, x.values[0]]\n",
    "#twitter_copy['stage'] = pd.DataFrame(twitter_copy[['doggo', 'floofer', 'pupper', 'puppo']].apply(lambda x: x.first_valid_index(), axis=1)).apply(func,axis=1)\n",
    "         "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 117,
   "metadata": {},
   "outputs": [],
   "source": [
    "df_retweet_by_day = merged_df.groupby([merged_df[\"timestamp\"].dt.year, merged_df[\"timestamp\"].dt.month]).agg(retweets=('retweet_count', 'sum'))\n",
    "df_favorite_by_day = merged_df.groupby([merged_df[\"timestamp\"].dt.year, merged_df[\"timestamp\"].dt.month]).agg(favorites=('favorite_count', 'sum'))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 118,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 720x576 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt0 = df_favorite_by_day.join(df_retweet_by_day)\n",
    "\n",
    "plt0.plot();"
   ]
  }
 ],
 "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": "3.8.3"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}