wrangle_act.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Importing important libraries\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "import pandas as pd\n",
    "import numpy as np\n",
    "import os\n",
    "import requests\n",
    "from io import BytesIO\n",
    "import tweepy\n",
    "import re\n",
    "import json\n",
    "from timeit import default_timer as timer\n",
    "import matplotlib.pyplot as plt\n",
    "%matplotlib inline\n",
    "import seaborn as sns\n",
    "sns.set()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Gathring Data (1)\n",
    "Download the file manually then read into a dataframe with an appropriate name using the pandas library."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [],
   "source": [
    "df_1 = pd.read_csv('work_directory/twitter-archive-enhanced.csv')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Gathring Data (2)\n",
    "- Downloaded programmatically using the Requests library and the provided URL.\n",
    "- Store the data to anew file with name extracted from the URL.\n",
    "- read the data and assign it to a new data frame."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [],
   "source": [
    "folder_name ='work_directory'\n",
    "#Make directory if it doesn't already exist\n",
    "if not os.path.exists(folder_name):\n",
    "    os.makedirs(folder_name)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [],
   "source": [
    "url = 'https://d17h27t6h515a5.cloudfront.net/topher/2017/August/599fd2ad_image-predictions/image-predictions.tsv'\n",
    "r = requests.get(url)\n",
    "with open (os.path.join(folder_name, url.split('/')[-1]),mode = 'wb') as file:\n",
    "     file.write(r.content)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [],
   "source": [
    "df_2 = pd.read_csv('work_directory/image-predictions.tsv',sep = '\\s+')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Gather (3)\n",
    "- Using the tweet IDs in the WeRateDogs Twitter archive, query the Twitter API for each tweet's JSON data using Python's Tweepy library.\n",
    "- Store each tweet's entire set of JSON data in a file called tweet_json.txt file.\n",
    "- Each tweet's JSON data should be written to its own line and seprated by '\\n'.\n",
    "- Because of there may be some deleted tweets for which no status would be found then try-except statment should be used to avoid exceptions.\n",
    "- append such not found tweet_ids to an empty list to know how many tweets in archive data set was not found."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [],
   "source": [
    "consumer_key =    '********************************'\n",
    "\n",
    "consumer_secret = '********************************'\n",
    "\n",
    "access_token =    '********************************'\n",
    "\n",
    "access_secret =    '*******************************'\n",
    "\n",
    "\n",
    "auth = tweepy.OAuthHandler(consumer_key, consumer_secret)\n",
    "auth.set_access_token(access_token, access_secret)\n",
    "\n",
    "api = tweepy.API(auth, wait_on_rate_limit=True)\n",
    "tweet_ids = df_1.tweet_id.values\n",
    "len(tweet_ids)\n",
    "\n",
    "# Query Twitter's API for JSON data for each tweet ID in the Twitter archive\n",
    "count = 0\n",
    "fails_list = []\n",
    "start = timer()\n",
    "if not os.path.isfile('work_directory/tweet_json.txt'):\n",
    "    with open('work_directory/tweet_json.txt', 'w') as file:\n",
    "    # This loop will likely take 20-30 minutes to run because of Twitter's rate limit\n",
    "        for tweet_id in tweet_ids:\n",
    "            count += 1\n",
    "            print(str(count) + \": \" + str(tweet_id))\n",
    "            try:\n",
    "                tweet = api.get_status(tweet_id, tweet_mode='extended')\n",
    "                print(\"Success\")\n",
    "                json.dump(tweet._json, file) # Save each tweet's returned JSON as a new line in a .txt file\n",
    "                file.write('\\n')\n",
    "            except tweepy.TweepError as e:\n",
    "                print(\"Fail\")\n",
    "                fails_list.append(tweet_id)\n",
    "                pass\n",
    "end = timer()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0"
      ]
     },
     "execution_count": 7,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "len(fails_list)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Gather (3) continue\n",
    "- Read this tweet_json.txt file line by line. \n",
    "- convert string type to list of dictionaries data structur.\n",
    "- Extract data required as per project requirement\n",
    "- DataFrame contains tweet ID,retweet count, and favorite count."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "dict_keys(['created_at', 'id', 'id_str', 'full_text', 'truncated', 'display_text_range', 'entities', 'extended_entities', 'source', 'in_reply_to_status_id', 'in_reply_to_status_id_str', 'in_reply_to_user_id', 'in_reply_to_user_id_str', 'in_reply_to_screen_name', 'user', 'geo', 'coordinates', 'place', 'contributors', 'is_quote_status', 'retweet_count', 'favorite_count', 'favorited', 'retweeted', 'possibly_sensitive', 'possibly_sensitive_appealable', 'lang'])\n"
     ]
    }
   ],
   "source": [
    "df3_list = []\n",
    "with open ('work_directory/tweet_json.txt',\"r\") as j_file:\n",
    "    for l in j_file:\n",
    "        dic = {}\n",
    "        data=json.loads(l)\n",
    "        dic['id'] = data['id']\n",
    "        dic['favorite_count'] = data['favorite_count']\n",
    "        dic['retweet_count'] = data['retweet_count']\n",
    "        dic['followers_count'] = data['user']['followers_count']\n",
    "        df3_list.append(dic)\n",
    "print(data.keys())"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [],
   "source": [
    "df_3 = pd.DataFrame(df3_list,columns = ['id','retweet_count','favorite_count','followers_count',])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [],
   "source": [
    "df_3.to_csv('work_directory/tweet1.csv')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "2330"
      ]
     },
     "execution_count": 11,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_3 = pd.read_csv('work_directory/tweet1.csv')\n",
    "len(df_3)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Assesment"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "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>1773</th>\n",
       "      <td>678255464182861824</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>2015-12-19 16:47:58 +0000</td>\n",
       "      <td>&lt;a href=\"http://twitter.com/download/iphone\" r...</td>\n",
       "      <td>This is Jessifer. She is a Bismoth Teriyaki. F...</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>https://twitter.com/dog_rates/status/678255464...</td>\n",
       "      <td>9</td>\n",
       "      <td>10</td>\n",
       "      <td>Jessifer</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2323</th>\n",
       "      <td>666428276349472768</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>2015-11-17 01:30:57 +0000</td>\n",
       "      <td>&lt;a href=\"http://twitter.com/download/iphone\" r...</td>\n",
       "      <td>Here we have an Austrian Pulitzer. Collectors ...</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>https://twitter.com/dog_rates/status/666428276...</td>\n",
       "      <td>7</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>877</th>\n",
       "      <td>760893934457552897</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>2016-08-03 17:43:45 +0000</td>\n",
       "      <td>&lt;a href=\"http://twitter.com/download/iphone\" r...</td>\n",
       "      <td>This is Wishes. He has the day off. Daily stru...</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>https://twitter.com/dog_rates/status/760893934...</td>\n",
       "      <td>11</td>\n",
       "      <td>10</td>\n",
       "      <td>Wishes</td>\n",
       "      <td>doggo</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1156</th>\n",
       "      <td>724771698126512129</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>2016-04-26 01:26:53 +0000</td>\n",
       "      <td>&lt;a href=\"http://twitter.com/download/iphone\" r...</td>\n",
       "      <td>Nothin better than a doggo and a sunset. 11/10...</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>https://twitter.com/dog_rates/status/724771698...</td>\n",
       "      <td>11</td>\n",
       "      <td>10</td>\n",
       "      <td>None</td>\n",
       "      <td>doggo</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>278</th>\n",
       "      <td>840268004936019968</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>2017-03-10 18:27:58 +0000</td>\n",
       "      <td>&lt;a href=\"http://twitter.com/download/iphone\" r...</td>\n",
       "      <td>This is Monty. He makes instantly regrettable ...</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>https://twitter.com/dog_rates/status/840268004...</td>\n",
       "      <td>12</td>\n",
       "      <td>10</td>\n",
       "      <td>Monty</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1580</th>\n",
       "      <td>687127927494963200</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>2016-01-13 04:23:58 +0000</td>\n",
       "      <td>&lt;a href=\"http://twitter.com/download/iphone\" r...</td>\n",
       "      <td>Meet Gordon. He's an asshole. 9/10 would still...</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>https://twitter.com/dog_rates/status/687127927...</td>\n",
       "      <td>9</td>\n",
       "      <td>10</td>\n",
       "      <td>Gordon</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1313</th>\n",
       "      <td>706904523814649856</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>2016-03-07 18:09:06 +0000</td>\n",
       "      <td>&lt;a href=\"http://vine.co\" rel=\"nofollow\"&gt;Vine -...</td>\n",
       "      <td>Ever seen a dog pet another dog? Both 13/10 tr...</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>https://vine.co/v/iXQAm5Lrgrh</td>\n",
       "      <td>13</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>1393</th>\n",
       "      <td>700029284593901568</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>2016-02-17 18:49:22 +0000</td>\n",
       "      <td>&lt;a href=\"http://twitter.com/download/iphone\" r...</td>\n",
       "      <td>This is Coops. His ship is taking on water. So...</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>https://twitter.com/dog_rates/status/700029284...</td>\n",
       "      <td>10</td>\n",
       "      <td>10</td>\n",
       "      <td>Coops</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>304</th>\n",
       "      <td>836380477523124226</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>2017-02-28 01:00:19 +0000</td>\n",
       "      <td>&lt;a href=\"http://twitter.com/download/iphone\" r...</td>\n",
       "      <td>This is Ava. She just blasted off. Streamline ...</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>https://twitter.com/dog_rates/status/836380477...</td>\n",
       "      <td>12</td>\n",
       "      <td>10</td>\n",
       "      <td>Ava</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "      <td>pupper</td>\n",
       "      <td>None</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>853</th>\n",
       "      <td>765371061932261376</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>2016-08-16 02:14:15 +0000</td>\n",
       "      <td>&lt;a href=\"http://twitter.com/download/iphone\" r...</td>\n",
       "      <td>This is Louie. He's had a long day. Did a lot ...</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>https://twitter.com/dog_rates/status/765371061...</td>\n",
       "      <td>11</td>\n",
       "      <td>10</td>\n",
       "      <td>Louie</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "      <td>pupper</td>\n",
       "      <td>None</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>987</th>\n",
       "      <td>749036806121881602</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>2016-07-02 00:27:45 +0000</td>\n",
       "      <td>&lt;a href=\"http://twitter.com/download/iphone\" r...</td>\n",
       "      <td>This is Dietrich. He hops at random. Other dog...</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>https://twitter.com/dog_rates/status/749036806...</td>\n",
       "      <td>8</td>\n",
       "      <td>10</td>\n",
       "      <td>Dietrich</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1266</th>\n",
       "      <td>709852847387627521</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>2016-03-15 21:24:41 +0000</td>\n",
       "      <td>&lt;a href=\"http://twitter.com/download/iphone\" r...</td>\n",
       "      <td>*lets out a tiny whimper and then collapses* ....</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>https://twitter.com/dog_rates/status/709852847...</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>2324</th>\n",
       "      <td>666421158376562688</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>2015-11-17 01:02:40 +0000</td>\n",
       "      <td>&lt;a href=\"http://twitter.com/download/iphone\" r...</td>\n",
       "      <td>*internally screaming* 12/10 https://t.co/YMcr...</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>https://twitter.com/dog_rates/status/666421158...</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>1416</th>\n",
       "      <td>698635131305795584</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>2016-02-13 22:29:29 +0000</td>\n",
       "      <td>&lt;a href=\"http://twitter.com/download/iphone\" r...</td>\n",
       "      <td>Here we are witnessing five Guatemalan Birch F...</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>https://twitter.com/dog_rates/status/698635131...</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>329</th>\n",
       "      <td>833479644947025920</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>2017-02-20 00:53:27 +0000</td>\n",
       "      <td>&lt;a href=\"http://twitter.com/download/iphone\" r...</td>\n",
       "      <td>This is Poppy. She just arrived. 13/10 would s...</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>https://twitter.com/dog_rates/status/833479644...</td>\n",
       "      <td>13</td>\n",
       "      <td>10</td>\n",
       "      <td>Poppy</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>988</th>\n",
       "      <td>748977405889503236</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>2016-07-01 20:31:43 +0000</td>\n",
       "      <td>&lt;a href=\"http://twitter.com/download/iphone\" r...</td>\n",
       "      <td>What jokester sent in a pic without a dog in i...</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>https://twitter.com/dog_rates/status/748977405...</td>\n",
       "      <td>10</td>\n",
       "      <td>10</td>\n",
       "      <td>not</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>531</th>\n",
       "      <td>808106460588765185</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>2016-12-12 00:29:28 +0000</td>\n",
       "      <td>&lt;a href=\"http://twitter.com/download/iphone\" r...</td>\n",
       "      <td>Here we have Burke (pupper) and Dexter (doggo)...</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>https://twitter.com/dog_rates/status/808106460...</td>\n",
       "      <td>12</td>\n",
       "      <td>10</td>\n",
       "      <td>None</td>\n",
       "      <td>doggo</td>\n",
       "      <td>None</td>\n",
       "      <td>pupper</td>\n",
       "      <td>None</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>196</th>\n",
       "      <td>854732716440526848</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>2017-04-19 16:25:34 +0000</td>\n",
       "      <td>&lt;a href=\"http://twitter.com/download/iphone\" r...</td>\n",
       "      <td>This is Marlee. She fetched a flower and immed...</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>https://twitter.com/dog_rates/status/854732716...</td>\n",
       "      <td>12</td>\n",
       "      <td>10</td>\n",
       "      <td>Marlee</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2234</th>\n",
       "      <td>668190681446379520</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>2015-11-21 22:14:07 +0000</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>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</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",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>911</th>\n",
       "      <td>757597904299253760</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>2016-07-25 15:26:30 +0000</td>\n",
       "      <td>&lt;a href=\"http://twitter.com/download/iphone\" r...</td>\n",
       "      <td>RT @jon_hill987: @dog_rates There is a cunning...</td>\n",
       "      <td>7.575971e+17</td>\n",
       "      <td>280479778.0</td>\n",
       "      <td>2016-07-25 15:23:28 +0000</td>\n",
       "      <td>https://twitter.com/jon_hill987/status/7575971...</td>\n",
       "      <td>11</td>\n",
       "      <td>10</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "      <td>pupper</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",
       "1773  678255464182861824                    NaN                  NaN   \n",
       "2323  666428276349472768                    NaN                  NaN   \n",
       "877   760893934457552897                    NaN                  NaN   \n",
       "1156  724771698126512129                    NaN                  NaN   \n",
       "278   840268004936019968                    NaN                  NaN   \n",
       "1580  687127927494963200                    NaN                  NaN   \n",
       "1313  706904523814649856                    NaN                  NaN   \n",
       "1393  700029284593901568                    NaN                  NaN   \n",
       "304   836380477523124226                    NaN                  NaN   \n",
       "853   765371061932261376                    NaN                  NaN   \n",
       "987   749036806121881602                    NaN                  NaN   \n",
       "1266  709852847387627521                    NaN                  NaN   \n",
       "2324  666421158376562688                    NaN                  NaN   \n",
       "1416  698635131305795584                    NaN                  NaN   \n",
       "329   833479644947025920                    NaN                  NaN   \n",
       "988   748977405889503236                    NaN                  NaN   \n",
       "531   808106460588765185                    NaN                  NaN   \n",
       "196   854732716440526848                    NaN                  NaN   \n",
       "2234  668190681446379520                    NaN                  NaN   \n",
       "911   757597904299253760                    NaN                  NaN   \n",
       "\n",
       "                      timestamp  \\\n",
       "1773  2015-12-19 16:47:58 +0000   \n",
       "2323  2015-11-17 01:30:57 +0000   \n",
       "877   2016-08-03 17:43:45 +0000   \n",
       "1156  2016-04-26 01:26:53 +0000   \n",
       "278   2017-03-10 18:27:58 +0000   \n",
       "1580  2016-01-13 04:23:58 +0000   \n",
       "1313  2016-03-07 18:09:06 +0000   \n",
       "1393  2016-02-17 18:49:22 +0000   \n",
       "304   2017-02-28 01:00:19 +0000   \n",
       "853   2016-08-16 02:14:15 +0000   \n",
       "987   2016-07-02 00:27:45 +0000   \n",
       "1266  2016-03-15 21:24:41 +0000   \n",
       "2324  2015-11-17 01:02:40 +0000   \n",
       "1416  2016-02-13 22:29:29 +0000   \n",
       "329   2017-02-20 00:53:27 +0000   \n",
       "988   2016-07-01 20:31:43 +0000   \n",
       "531   2016-12-12 00:29:28 +0000   \n",
       "196   2017-04-19 16:25:34 +0000   \n",
       "2234  2015-11-21 22:14:07 +0000   \n",
       "911   2016-07-25 15:26:30 +0000   \n",
       "\n",
       "                                                 source  \\\n",
       "1773  <a href=\"http://twitter.com/download/iphone\" r...   \n",
       "2323  <a href=\"http://twitter.com/download/iphone\" r...   \n",
       "877   <a href=\"http://twitter.com/download/iphone\" r...   \n",
       "1156  <a href=\"http://twitter.com/download/iphone\" r...   \n",
       "278   <a href=\"http://twitter.com/download/iphone\" r...   \n",
       "1580  <a href=\"http://twitter.com/download/iphone\" r...   \n",
       "1313  <a href=\"http://vine.co\" rel=\"nofollow\">Vine -...   \n",
       "1393  <a href=\"http://twitter.com/download/iphone\" r...   \n",
       "304   <a href=\"http://twitter.com/download/iphone\" r...   \n",
       "853   <a href=\"http://twitter.com/download/iphone\" r...   \n",
       "987   <a href=\"http://twitter.com/download/iphone\" r...   \n",
       "1266  <a href=\"http://twitter.com/download/iphone\" r...   \n",
       "2324  <a href=\"http://twitter.com/download/iphone\" r...   \n",
       "1416  <a href=\"http://twitter.com/download/iphone\" r...   \n",
       "329   <a href=\"http://twitter.com/download/iphone\" r...   \n",
       "988   <a href=\"http://twitter.com/download/iphone\" r...   \n",
       "531   <a href=\"http://twitter.com/download/iphone\" r...   \n",
       "196   <a href=\"http://twitter.com/download/iphone\" r...   \n",
       "2234  <a href=\"http://twitter.com/download/iphone\" r...   \n",
       "911   <a href=\"http://twitter.com/download/iphone\" r...   \n",
       "\n",
       "                                                   text  retweeted_status_id  \\\n",
       "1773  This is Jessifer. She is a Bismoth Teriyaki. F...                  NaN   \n",
       "2323  Here we have an Austrian Pulitzer. Collectors ...                  NaN   \n",
       "877   This is Wishes. He has the day off. Daily stru...                  NaN   \n",
       "1156  Nothin better than a doggo and a sunset. 11/10...                  NaN   \n",
       "278   This is Monty. He makes instantly regrettable ...                  NaN   \n",
       "1580  Meet Gordon. He's an asshole. 9/10 would still...                  NaN   \n",
       "1313  Ever seen a dog pet another dog? Both 13/10 tr...                  NaN   \n",
       "1393  This is Coops. His ship is taking on water. So...                  NaN   \n",
       "304   This is Ava. She just blasted off. Streamline ...                  NaN   \n",
       "853   This is Louie. He's had a long day. Did a lot ...                  NaN   \n",
       "987   This is Dietrich. He hops at random. Other dog...                  NaN   \n",
       "1266  *lets out a tiny whimper and then collapses* ....                  NaN   \n",
       "2324  *internally screaming* 12/10 https://t.co/YMcr...                  NaN   \n",
       "1416  Here we are witnessing five Guatemalan Birch F...                  NaN   \n",
       "329   This is Poppy. She just arrived. 13/10 would s...                  NaN   \n",
       "988   What jokester sent in a pic without a dog in i...                  NaN   \n",
       "531   Here we have Burke (pupper) and Dexter (doggo)...                  NaN   \n",
       "196   This is Marlee. She fetched a flower and immed...                  NaN   \n",
       "2234  This is Skittles. I would kidnap Skittles. Pin...                  NaN   \n",
       "911   RT @jon_hill987: @dog_rates There is a cunning...         7.575971e+17   \n",
       "\n",
       "      retweeted_status_user_id retweeted_status_timestamp  \\\n",
       "1773                       NaN                        NaN   \n",
       "2323                       NaN                        NaN   \n",
       "877                        NaN                        NaN   \n",
       "1156                       NaN                        NaN   \n",
       "278                        NaN                        NaN   \n",
       "1580                       NaN                        NaN   \n",
       "1313                       NaN                        NaN   \n",
       "1393                       NaN                        NaN   \n",
       "304                        NaN                        NaN   \n",
       "853                        NaN                        NaN   \n",
       "987                        NaN                        NaN   \n",
       "1266                       NaN                        NaN   \n",
       "2324                       NaN                        NaN   \n",
       "1416                       NaN                        NaN   \n",
       "329                        NaN                        NaN   \n",
       "988                        NaN                        NaN   \n",
       "531                        NaN                        NaN   \n",
       "196                        NaN                        NaN   \n",
       "2234                       NaN                        NaN   \n",
       "911                280479778.0  2016-07-25 15:23:28 +0000   \n",
       "\n",
       "                                          expanded_urls  rating_numerator  \\\n",
       "1773  https://twitter.com/dog_rates/status/678255464...                 9   \n",
       "2323  https://twitter.com/dog_rates/status/666428276...                 7   \n",
       "877   https://twitter.com/dog_rates/status/760893934...                11   \n",
       "1156  https://twitter.com/dog_rates/status/724771698...                11   \n",
       "278   https://twitter.com/dog_rates/status/840268004...                12   \n",
       "1580  https://twitter.com/dog_rates/status/687127927...                 9   \n",
       "1313                      https://vine.co/v/iXQAm5Lrgrh                13   \n",
       "1393  https://twitter.com/dog_rates/status/700029284...                10   \n",
       "304   https://twitter.com/dog_rates/status/836380477...                12   \n",
       "853   https://twitter.com/dog_rates/status/765371061...                11   \n",
       "987   https://twitter.com/dog_rates/status/749036806...                 8   \n",
       "1266  https://twitter.com/dog_rates/status/709852847...                12   \n",
       "2324  https://twitter.com/dog_rates/status/666421158...                12   \n",
       "1416  https://twitter.com/dog_rates/status/698635131...                12   \n",
       "329   https://twitter.com/dog_rates/status/833479644...                13   \n",
       "988   https://twitter.com/dog_rates/status/748977405...                10   \n",
       "531   https://twitter.com/dog_rates/status/808106460...                12   \n",
       "196   https://twitter.com/dog_rates/status/854732716...                12   \n",
       "2234  https://twitter.com/dog_rates/status/668190681...                12   \n",
       "911   https://twitter.com/jon_hill987/status/7575971...                11   \n",
       "\n",
       "      rating_denominator      name  doggo floofer  pupper puppo  \n",
       "1773                  10  Jessifer   None    None    None  None  \n",
       "2323                  10      None   None    None    None  None  \n",
       "877                   10    Wishes  doggo    None    None  None  \n",
       "1156                  10      None  doggo    None    None  None  \n",
       "278                   10     Monty   None    None    None  None  \n",
       "1580                  10    Gordon   None    None    None  None  \n",
       "1313                  10      None   None    None    None  None  \n",
       "1393                  10     Coops   None    None    None  None  \n",
       "304                   10       Ava   None    None  pupper  None  \n",
       "853                   10     Louie   None    None  pupper  None  \n",
       "987                   10  Dietrich   None    None    None  None  \n",
       "1266                  10      None   None    None    None  None  \n",
       "2324                  10      None   None    None    None  None  \n",
       "1416                  10      None   None    None    None  None  \n",
       "329                   10     Poppy   None    None    None  None  \n",
       "988                   10       not   None    None    None  None  \n",
       "531                   10      None  doggo    None  pupper  None  \n",
       "196                   10    Marlee   None    None    None  None  \n",
       "2234                  10  Skittles   None    None    None  None  \n",
       "911                   10      None   None    None  pupper  None  "
      ]
     },
     "execution_count": 12,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_1.sample(20)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "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>retweeted_status_id</th>\n",
       "      <th>retweeted_status_user_id</th>\n",
       "      <th>rating_numerator</th>\n",
       "      <th>rating_denominator</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>count</th>\n",
       "      <td>2.356000e+03</td>\n",
       "      <td>7.800000e+01</td>\n",
       "      <td>7.800000e+01</td>\n",
       "      <td>1.810000e+02</td>\n",
       "      <td>1.810000e+02</td>\n",
       "      <td>2356.000000</td>\n",
       "      <td>2356.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>mean</th>\n",
       "      <td>7.427716e+17</td>\n",
       "      <td>7.455079e+17</td>\n",
       "      <td>2.014171e+16</td>\n",
       "      <td>7.720400e+17</td>\n",
       "      <td>1.241698e+16</td>\n",
       "      <td>13.126486</td>\n",
       "      <td>10.455433</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>std</th>\n",
       "      <td>6.856705e+16</td>\n",
       "      <td>7.582492e+16</td>\n",
       "      <td>1.252797e+17</td>\n",
       "      <td>6.236928e+16</td>\n",
       "      <td>9.599254e+16</td>\n",
       "      <td>45.876648</td>\n",
       "      <td>6.745237</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>min</th>\n",
       "      <td>6.660209e+17</td>\n",
       "      <td>6.658147e+17</td>\n",
       "      <td>1.185634e+07</td>\n",
       "      <td>6.661041e+17</td>\n",
       "      <td>7.832140e+05</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>25%</th>\n",
       "      <td>6.783989e+17</td>\n",
       "      <td>6.757419e+17</td>\n",
       "      <td>3.086374e+08</td>\n",
       "      <td>7.186315e+17</td>\n",
       "      <td>4.196984e+09</td>\n",
       "      <td>10.000000</td>\n",
       "      <td>10.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>50%</th>\n",
       "      <td>7.196279e+17</td>\n",
       "      <td>7.038708e+17</td>\n",
       "      <td>4.196984e+09</td>\n",
       "      <td>7.804657e+17</td>\n",
       "      <td>4.196984e+09</td>\n",
       "      <td>11.000000</td>\n",
       "      <td>10.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>75%</th>\n",
       "      <td>7.993373e+17</td>\n",
       "      <td>8.257804e+17</td>\n",
       "      <td>4.196984e+09</td>\n",
       "      <td>8.203146e+17</td>\n",
       "      <td>4.196984e+09</td>\n",
       "      <td>12.000000</td>\n",
       "      <td>10.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>max</th>\n",
       "      <td>8.924206e+17</td>\n",
       "      <td>8.862664e+17</td>\n",
       "      <td>8.405479e+17</td>\n",
       "      <td>8.874740e+17</td>\n",
       "      <td>7.874618e+17</td>\n",
       "      <td>1776.000000</td>\n",
       "      <td>170.000000</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "           tweet_id  in_reply_to_status_id  in_reply_to_user_id  \\\n",
       "count  2.356000e+03           7.800000e+01         7.800000e+01   \n",
       "mean   7.427716e+17           7.455079e+17         2.014171e+16   \n",
       "std    6.856705e+16           7.582492e+16         1.252797e+17   \n",
       "min    6.660209e+17           6.658147e+17         1.185634e+07   \n",
       "25%    6.783989e+17           6.757419e+17         3.086374e+08   \n",
       "50%    7.196279e+17           7.038708e+17         4.196984e+09   \n",
       "75%    7.993373e+17           8.257804e+17         4.196984e+09   \n",
       "max    8.924206e+17           8.862664e+17         8.405479e+17   \n",
       "\n",
       "       retweeted_status_id  retweeted_status_user_id  rating_numerator  \\\n",
       "count         1.810000e+02              1.810000e+02       2356.000000   \n",
       "mean          7.720400e+17              1.241698e+16         13.126486   \n",
       "std           6.236928e+16              9.599254e+16         45.876648   \n",
       "min           6.661041e+17              7.832140e+05          0.000000   \n",
       "25%           7.186315e+17              4.196984e+09         10.000000   \n",
       "50%           7.804657e+17              4.196984e+09         11.000000   \n",
       "75%           8.203146e+17              4.196984e+09         12.000000   \n",
       "max           8.874740e+17              7.874618e+17       1776.000000   \n",
       "\n",
       "       rating_denominator  \n",
       "count         2356.000000  \n",
       "mean            10.455433  \n",
       "std              6.745237  \n",
       "min              0.000000  \n",
       "25%             10.000000  \n",
       "50%             10.000000  \n",
       "75%             10.000000  \n",
       "max            170.000000  "
      ]
     },
     "execution_count": 13,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_1.describe()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "None        745\n",
       "a            55\n",
       "Charlie      12\n",
       "Oliver       11\n",
       "Lucy         11\n",
       "           ... \n",
       "Augie         1\n",
       "Fido          1\n",
       "Kona          1\n",
       "Jennifur      1\n",
       "Pepper        1\n",
       "Name: name, Length: 957, dtype: int64"
      ]
     },
     "execution_count": 14,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_1.name.value_counts()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "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_1.info()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "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>509</th>\n",
       "      <td>676101918813499392</td>\n",
       "      <td>https://pbs.twimg.com/media/CWH_FTgWIAAwOUy.jpg</td>\n",
       "      <td>1</td>\n",
       "      <td>Shih-Tzu</td>\n",
       "      <td>0.225848</td>\n",
       "      <td>True</td>\n",
       "      <td>Norfolk_terrier</td>\n",
       "      <td>0.186873</td>\n",
       "      <td>True</td>\n",
       "      <td>Irish_terrier</td>\n",
       "      <td>0.106987</td>\n",
       "      <td>True</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1156</th>\n",
       "      <td>733460102733135873</td>\n",
       "      <td>https://pbs.twimg.com/media/Ci3GDeyUoAAKOxn.jpg</td>\n",
       "      <td>1</td>\n",
       "      <td>chow</td>\n",
       "      <td>0.931275</td>\n",
       "      <td>True</td>\n",
       "      <td>beaver</td>\n",
       "      <td>0.028831</td>\n",
       "      <td>False</td>\n",
       "      <td>dhole</td>\n",
       "      <td>0.017379</td>\n",
       "      <td>False</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>810</th>\n",
       "      <td>692187005137076224</td>\n",
       "      <td>https://pbs.twimg.com/media/CZskaEIWIAUeTr5.jpg</td>\n",
       "      <td>2</td>\n",
       "      <td>Siberian_husky</td>\n",
       "      <td>0.810592</td>\n",
       "      <td>True</td>\n",
       "      <td>malamute</td>\n",
       "      <td>0.119745</td>\n",
       "      <td>True</td>\n",
       "      <td>Eskimo_dog</td>\n",
       "      <td>0.029265</td>\n",
       "      <td>True</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>369</th>\n",
       "      <td>672970152493887488</td>\n",
       "      <td>https://pbs.twimg.com/media/CVbeyGUU8AEq300.jpg</td>\n",
       "      <td>1</td>\n",
       "      <td>leaf_beetle</td>\n",
       "      <td>0.340154</td>\n",
       "      <td>False</td>\n",
       "      <td>rhinoceros_beetle</td>\n",
       "      <td>0.139698</td>\n",
       "      <td>False</td>\n",
       "      <td>crayfish</td>\n",
       "      <td>0.058034</td>\n",
       "      <td>False</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>355</th>\n",
       "      <td>672594978741354496</td>\n",
       "      <td>https://pbs.twimg.com/media/CVWJkJXWsAInlZl.jpg</td>\n",
       "      <td>1</td>\n",
       "      <td>Great_Pyrenees</td>\n",
       "      <td>0.755945</td>\n",
       "      <td>True</td>\n",
       "      <td>Old_English_sheepdog</td>\n",
       "      <td>0.082337</td>\n",
       "      <td>True</td>\n",
       "      <td>Afghan_hound</td>\n",
       "      <td>0.027037</td>\n",
       "      <td>True</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                tweet_id                                          jpg_url  \\\n",
       "509   676101918813499392  https://pbs.twimg.com/media/CWH_FTgWIAAwOUy.jpg   \n",
       "1156  733460102733135873  https://pbs.twimg.com/media/Ci3GDeyUoAAKOxn.jpg   \n",
       "810   692187005137076224  https://pbs.twimg.com/media/CZskaEIWIAUeTr5.jpg   \n",
       "369   672970152493887488  https://pbs.twimg.com/media/CVbeyGUU8AEq300.jpg   \n",
       "355   672594978741354496  https://pbs.twimg.com/media/CVWJkJXWsAInlZl.jpg   \n",
       "\n",
       "      img_num              p1   p1_conf  p1_dog                    p2  \\\n",
       "509         1        Shih-Tzu  0.225848    True       Norfolk_terrier   \n",
       "1156        1            chow  0.931275    True                beaver   \n",
       "810         2  Siberian_husky  0.810592    True              malamute   \n",
       "369         1     leaf_beetle  0.340154   False     rhinoceros_beetle   \n",
       "355         1  Great_Pyrenees  0.755945    True  Old_English_sheepdog   \n",
       "\n",
       "       p2_conf  p2_dog             p3   p3_conf  p3_dog  \n",
       "509   0.186873    True  Irish_terrier  0.106987    True  \n",
       "1156  0.028831   False          dhole  0.017379   False  \n",
       "810   0.119745    True     Eskimo_dog  0.029265    True  \n",
       "369   0.139698   False       crayfish  0.058034   False  \n",
       "355   0.082337    True   Afghan_hound  0.027037    True  "
      ]
     },
     "execution_count": 16,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_2.sample(5)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "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>img_num</th>\n",
       "      <th>p1_conf</th>\n",
       "      <th>p2_conf</th>\n",
       "      <th>p3_conf</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>count</th>\n",
       "      <td>2.075000e+03</td>\n",
       "      <td>2075.000000</td>\n",
       "      <td>2075.000000</td>\n",
       "      <td>2.075000e+03</td>\n",
       "      <td>2.075000e+03</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>mean</th>\n",
       "      <td>7.384514e+17</td>\n",
       "      <td>1.203855</td>\n",
       "      <td>0.594548</td>\n",
       "      <td>1.345886e-01</td>\n",
       "      <td>6.032417e-02</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>std</th>\n",
       "      <td>6.785203e+16</td>\n",
       "      <td>0.561875</td>\n",
       "      <td>0.271174</td>\n",
       "      <td>1.006657e-01</td>\n",
       "      <td>5.090593e-02</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>min</th>\n",
       "      <td>6.660209e+17</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.044333</td>\n",
       "      <td>1.011300e-08</td>\n",
       "      <td>1.740170e-10</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>25%</th>\n",
       "      <td>6.764835e+17</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.364412</td>\n",
       "      <td>5.388625e-02</td>\n",
       "      <td>1.622240e-02</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>50%</th>\n",
       "      <td>7.119988e+17</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.588230</td>\n",
       "      <td>1.181810e-01</td>\n",
       "      <td>4.944380e-02</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>75%</th>\n",
       "      <td>7.932034e+17</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.843855</td>\n",
       "      <td>1.955655e-01</td>\n",
       "      <td>9.180755e-02</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>max</th>\n",
       "      <td>8.924206e+17</td>\n",
       "      <td>4.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>4.880140e-01</td>\n",
       "      <td>2.734190e-01</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "           tweet_id      img_num      p1_conf       p2_conf       p3_conf\n",
       "count  2.075000e+03  2075.000000  2075.000000  2.075000e+03  2.075000e+03\n",
       "mean   7.384514e+17     1.203855     0.594548  1.345886e-01  6.032417e-02\n",
       "std    6.785203e+16     0.561875     0.271174  1.006657e-01  5.090593e-02\n",
       "min    6.660209e+17     1.000000     0.044333  1.011300e-08  1.740170e-10\n",
       "25%    6.764835e+17     1.000000     0.364412  5.388625e-02  1.622240e-02\n",
       "50%    7.119988e+17     1.000000     0.588230  1.181810e-01  4.944380e-02\n",
       "75%    7.932034e+17     1.000000     0.843855  1.955655e-01  9.180755e-02\n",
       "max    8.924206e+17     4.000000     1.000000  4.880140e-01  2.734190e-01"
      ]
     },
     "execution_count": 17,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_2.describe()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "True     1499\n",
       "False     576\n",
       "Name: p3_dog, dtype: int64"
      ]
     },
     "execution_count": 18,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_2.p3_dog.value_counts()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "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_2.info()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "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>Unnamed: 0</th>\n",
       "      <th>id</th>\n",
       "      <th>retweet_count</th>\n",
       "      <th>favorite_count</th>\n",
       "      <th>followers_count</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>1775</th>\n",
       "      <td>1775</td>\n",
       "      <td>676975532580409345</td>\n",
       "      <td>1016</td>\n",
       "      <td>2645</td>\n",
       "      <td>8941422</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>453</th>\n",
       "      <td>453</td>\n",
       "      <td>816829038950027264</td>\n",
       "      <td>4633</td>\n",
       "      <td>0</td>\n",
       "      <td>8941397</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>200</th>\n",
       "      <td>200</td>\n",
       "      <td>852311364735569921</td>\n",
       "      <td>9360</td>\n",
       "      <td>31617</td>\n",
       "      <td>8941394</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>548</th>\n",
       "      <td>548</td>\n",
       "      <td>802265048156610565</td>\n",
       "      <td>1363</td>\n",
       "      <td>6877</td>\n",
       "      <td>8941398</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1286</th>\n",
       "      <td>1286</td>\n",
       "      <td>707038192327901184</td>\n",
       "      <td>756</td>\n",
       "      <td>2116</td>\n",
       "      <td>8941417</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "      Unnamed: 0                  id  retweet_count  favorite_count  \\\n",
       "1775        1775  676975532580409345           1016            2645   \n",
       "453          453  816829038950027264           4633               0   \n",
       "200          200  852311364735569921           9360           31617   \n",
       "548          548  802265048156610565           1363            6877   \n",
       "1286        1286  707038192327901184            756            2116   \n",
       "\n",
       "      followers_count  \n",
       "1775          8941422  \n",
       "453           8941397  \n",
       "200           8941394  \n",
       "548           8941398  \n",
       "1286          8941417  "
      ]
     },
     "execution_count": 20,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_3.sample(5)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "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>Unnamed: 0</th>\n",
       "      <th>id</th>\n",
       "      <th>retweet_count</th>\n",
       "      <th>favorite_count</th>\n",
       "      <th>followers_count</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>count</th>\n",
       "      <td>2330.000000</td>\n",
       "      <td>2.330000e+03</td>\n",
       "      <td>2330.000000</td>\n",
       "      <td>2330.000000</td>\n",
       "      <td>2.330000e+03</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>mean</th>\n",
       "      <td>1164.500000</td>\n",
       "      <td>7.419381e+17</td>\n",
       "      <td>2610.340343</td>\n",
       "      <td>7355.005150</td>\n",
       "      <td>8.941412e+06</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>std</th>\n",
       "      <td>672.757386</td>\n",
       "      <td>6.823077e+16</td>\n",
       "      <td>4415.199850</td>\n",
       "      <td>11421.365768</td>\n",
       "      <td>1.428663e+01</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>min</th>\n",
       "      <td>0.000000</td>\n",
       "      <td>6.660209e+17</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>8.941392e+06</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>25%</th>\n",
       "      <td>582.250000</td>\n",
       "      <td>6.782926e+17</td>\n",
       "      <td>529.250000</td>\n",
       "      <td>1279.500000</td>\n",
       "      <td>8.941398e+06</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>50%</th>\n",
       "      <td>1164.500000</td>\n",
       "      <td>7.183508e+17</td>\n",
       "      <td>1219.500000</td>\n",
       "      <td>3191.500000</td>\n",
       "      <td>8.941415e+06</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>75%</th>\n",
       "      <td>1746.750000</td>\n",
       "      <td>7.986712e+17</td>\n",
       "      <td>3028.750000</td>\n",
       "      <td>8998.000000</td>\n",
       "      <td>8.941422e+06</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>max</th>\n",
       "      <td>2329.000000</td>\n",
       "      <td>8.924206e+17</td>\n",
       "      <td>74967.000000</td>\n",
       "      <td>151537.000000</td>\n",
       "      <td>8.941503e+06</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "        Unnamed: 0            id  retweet_count  favorite_count  \\\n",
       "count  2330.000000  2.330000e+03    2330.000000     2330.000000   \n",
       "mean   1164.500000  7.419381e+17    2610.340343     7355.005150   \n",
       "std     672.757386  6.823077e+16    4415.199850    11421.365768   \n",
       "min       0.000000  6.660209e+17       1.000000        0.000000   \n",
       "25%     582.250000  6.782926e+17     529.250000     1279.500000   \n",
       "50%    1164.500000  7.183508e+17    1219.500000     3191.500000   \n",
       "75%    1746.750000  7.986712e+17    3028.750000     8998.000000   \n",
       "max    2329.000000  8.924206e+17   74967.000000   151537.000000   \n",
       "\n",
       "       followers_count  \n",
       "count     2.330000e+03  \n",
       "mean      8.941412e+06  \n",
       "std       1.428663e+01  \n",
       "min       8.941392e+06  \n",
       "25%       8.941398e+06  \n",
       "50%       8.941415e+06  \n",
       "75%       8.941422e+06  \n",
       "max       8.941503e+06  "
      ]
     },
     "execution_count": 21,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_3.describe()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "8941429    220\n",
       "8941394    179\n",
       "8941406    178\n",
       "8941419    159\n",
       "8941422    142\n",
       "8941418    139\n",
       "8941393    125\n",
       "8941395    105\n",
       "8941397     85\n",
       "8941421     83\n",
       "8941416     82\n",
       "8941417     74\n",
       "8941408     73\n",
       "8941430     73\n",
       "8941404     70\n",
       "8941398     59\n",
       "8941400     48\n",
       "8941399     48\n",
       "8941427     42\n",
       "8941401     38\n",
       "8941409     35\n",
       "8941420     29\n",
       "8941415     28\n",
       "8941396     27\n",
       "8941425     25\n",
       "8941426     24\n",
       "8941414     22\n",
       "8941424     18\n",
       "8941413     17\n",
       "8941428     15\n",
       "8941403     13\n",
       "8941423     11\n",
       "8941392      6\n",
       "8941405      6\n",
       "8941402      6\n",
       "8941467      3\n",
       "8941407      3\n",
       "8941501      2\n",
       "8941431      2\n",
       "8941495      2\n",
       "8941503      2\n",
       "8941472      2\n",
       "8941480      2\n",
       "8941490      2\n",
       "8941469      1\n",
       "8941473      1\n",
       "8941491      1\n",
       "8941492      1\n",
       "8941468      1\n",
       "8941502      1\n",
       "Name: followers_count, dtype: int64"
      ]
     },
     "execution_count": 22,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_3.followers_count.value_counts()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Quality issues:\n",
    "### df_1 :\n",
    "- As per the project requirements; only original ratings (no retweets) that have images should be included\n",
    "- Columns of retweeted_status_id and its related data have entries which not part of our analysis.\n",
    "- Inaccurate data in name 'a' represented 55 times.\n",
    "- Inaccurate data in rating_nominator and df_denominator.\n",
    "- Null values reprsened as \"None\" in columns doggo, floofer, pupper, puppo.\n",
    "- Invalid dtype in timestamp represented as object type.\n",
    "- Column tweet_id represented as int64 across all data set.\n",
    "### df_2 :\n",
    "- Columns of predictions have lables represented as letters.\n",
    "- Column 'p1_dog' have 543 false prediction, p2_dog have 522 false prediction, p3_dog have 576 false prediction. \n",
    "- Column 'p1' have invalid data like(china_cabinet, shield, orange, walking_stick,...etc).\n",
    "- Column 'p2' have invalid data like(ice_lolly, Japanese_spaniel, china_cabinet, necklace,...etc).\n",
    "- Column 'p3' have invalid data like(kimono, cab, axolotl, passenger_car, tripod, grocery_store,...etc).\n",
    "## Tidness issues:\n",
    "### df_1:\n",
    "- Variable dog stages have four columns.\n",
    "### df_2:\n",
    "- Variable prediction have three columns.\n",
    "### df_3:\n",
    "- Table df3_clean have same observation unit that in df1_clean.\n",
    "- Column name id in stead of tweet_id."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Cleaning"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "befor cleaning copying DataFrames in pandas is done using `.copy` method"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(2356, 17)"
      ]
     },
     "execution_count": 23,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df1_clean = df_1.copy()\n",
    "df1_clean.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(2075, 12)"
      ]
     },
     "execution_count": 24,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df2_clean = df_2.copy()\n",
    "df2_clean.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(2330, 5)"
      ]
     },
     "execution_count": 25,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df3_clean = df_3.copy()\n",
    "df3_clean.shape"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Define\n",
    "##### df3_clean Column name 'id' in stead of 'tweet_id'."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "metadata": {},
   "outputs": [],
   "source": [
    "df3_clean.rename(columns={'id': 'tweet_id'},inplace =True)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Test"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "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>Unnamed: 0</th>\n",
       "      <th>tweet_id</th>\n",
       "      <th>retweet_count</th>\n",
       "      <th>favorite_count</th>\n",
       "      <th>followers_count</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>792</th>\n",
       "      <td>792</td>\n",
       "      <td>771136648247640064</td>\n",
       "      <td>2857</td>\n",
       "      <td>8987</td>\n",
       "      <td>8941404</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "     Unnamed: 0            tweet_id  retweet_count  favorite_count  \\\n",
       "792         792  771136648247640064           2857            8987   \n",
       "\n",
       "     followers_count  \n",
       "792          8941404  "
      ]
     },
     "execution_count": 27,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df3_clean.sample()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Define\n",
    "##### Null values reprsened as \"None\" in columns doggo, floofer, pupper, puppo.\n",
    "use replace function to replace it with\"\""
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "metadata": {},
   "outputs": [],
   "source": [
    "df1_clean.iloc[:,-4:] = df1_clean.iloc[:,-4:].replace('None','')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Test"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "(2356, 17)\n"
     ]
    },
    {
     "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>pupper</th>\n",
       "      <th>floofer</th>\n",
       "      <th>puppo</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>8</th>\n",
       "      <td></td>\n",
       "      <td></td>\n",
       "      <td></td>\n",
       "      <td></td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1154</th>\n",
       "      <td></td>\n",
       "      <td></td>\n",
       "      <td></td>\n",
       "      <td></td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1153</th>\n",
       "      <td></td>\n",
       "      <td></td>\n",
       "      <td></td>\n",
       "      <td></td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2276</th>\n",
       "      <td></td>\n",
       "      <td></td>\n",
       "      <td></td>\n",
       "      <td></td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>339</th>\n",
       "      <td>doggo</td>\n",
       "      <td></td>\n",
       "      <td></td>\n",
       "      <td></td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "      doggo pupper floofer puppo\n",
       "8                               \n",
       "1154                            \n",
       "1153                            \n",
       "2276                            \n",
       "339   doggo                     "
      ]
     },
     "execution_count": 29,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "print(df1_clean.shape)\n",
    "df1_clean.loc[:,['doggo','pupper','floofer','puppo']].sample(5)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Define\n",
    "##### Variable dog stages have four columns\n",
    "- creat a new column 'dog_stages' that contain all stages of every dog.\n",
    "-  drop the old columns"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "metadata": {},
   "outputs": [],
   "source": [
    "df1_clean['dog_stages'] = df1_clean.doggo + df1_clean.floofer + df1_clean.pupper + df1_clean.puppo"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "metadata": {},
   "outputs": [],
   "source": [
    "df1_clean.drop(columns =['doggo', 'floofer', 'pupper', 'puppo'] , inplace = True)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Test"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(2356, 14)"
      ]
     },
     "execution_count": 32,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df1_clean.shape"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Define\n",
    "##### Inaccurate data in name 'a' represented 55 times.\n",
    "Extract the correct name using REG expression"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 33,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "56       a\n",
       "393     Mo\n",
       "446     Bo\n",
       "449     Bo\n",
       "553     Bo\n",
       "        ..\n",
       "2349    an\n",
       "2350     a\n",
       "2352     a\n",
       "2353     a\n",
       "2354     a\n",
       "Name: name, Length: 79, dtype: object"
      ]
     },
     "execution_count": 33,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "short_name = df1_clean [df1_clean.name.str.len() < 3 ]\n",
    "short_name.name"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 34,
   "metadata": {},
   "outputs": [],
   "source": [
    "pattern_2 = re.compile(r'(?:name(?:d)?)\\s{1}(?:is\\s)?([A-Za-z]+)')\n",
    "for index, row in df1_clean.iterrows():  \n",
    "    try:\n",
    "        if row['name'] == \"a\":\n",
    "            c_name = re.findall(pattern_2, row['text'])[0]\n",
    "            df1_clean.loc[index,'name'] = df1_clean.loc[index,'name'].replace('a', c_name)\n",
    "        elif row['name'] == 'an':\n",
    "            c_name = re.findall(pattern_2, row['text'])[0]\n",
    "            df1_clean.loc[index,'name'] = df1_clean.loc[index,'name'].replace('an', c_name)\n",
    "        elif row['name'] == 'O':\n",
    "            c_name = re.findall(pattern_2, row['text'])[0]\n",
    "            df1_clean.loc[index,'name'] = df1_clean.loc[index,'name'].replace('O', c_name)\n",
    "        elif row['name'] == 'Al':\n",
    "            c_name = re.findall(pattern_2, row['text'])[0]\n",
    "            df1_clean.loc[index,'name'] = df1_clean.loc[index,'name'].replace('Al', c_name)\n",
    "        elif row['name'] == 'my':\n",
    "            c_name = re.findall(pattern_2, row['text'])[0]\n",
    "            df1_clean.loc[index,'name'] = df1_clean.loc[index,'name'].replace('my', c_name)\n",
    "        elif row['name'] == 'by':\n",
    "            c_name = re.findall(pattern_2, row['text'])[0]\n",
    "            df1_clean.loc[index,'name'] = df1_clean.loc[index,'name'].replace('by', c_name)\n",
    "        elif row['name'] == 'Ed':\n",
    "            c_name = re.findall(pattern_2, row['text'])[0]\n",
    "            df1_clean.loc[index,'name'] = df1_clean.loc[index,'name'].replace('Ed', c_name)\n",
    "        elif row['name'] == 'Jo':\n",
    "            c_name = re.findall(pattern_2, row['text'])[0]\n",
    "            df1_clean.loc[index,'name'] = df1_clean.loc[index,'name'].replace('Jo', c_name)\n",
    "        elif row['name'] == 'JD':\n",
    "            c_name = re.findall(pattern_2, row['text'])[0]\n",
    "            df1_clean.loc[index,'name'] = df1_clean.loc[index,'name'].replace('JD', c_name)\n",
    "        elif row['name'] == 'None':\n",
    "            c_name = re.findall(pattern_2, row['text'])[0]\n",
    "            df1_clean.loc[index,'name'] = df1_clean.loc[index,'name'].replace('None', c_name)\n",
    "        elif row['name'] == \"Mo\":\n",
    "            c_name = re.findall(pattern_2, row['text'])[0]\n",
    "            df1_clean.loc[index,'name'] = df1_clean.loc[index,'name'].replace('Mo', c_name)\n",
    "        elif row['name'] == 'Bo':\n",
    "            c_name = re.findall(pattern_2, row['text'])[0]\n",
    "            df1_clean.loc[index,'name'] = df1_clean.loc[index,'name'].replace('Bo', c_name)\n",
    "        elif row['name'] == 'to':\n",
    "            c_name = re.findall(pattern_2, row['text'])[0]\n",
    "            df1_clean.loc[index,'name'] = df1_clean.loc[index,'name'].replace('to', c_name)\n",
    "    except IndexError:\n",
    "        df1_clean.loc[index,'name'] = np.nan"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Test"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 35,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "149     to\n",
       "2086    of\n",
       "Name: name, dtype: object"
      ]
     },
     "execution_count": 35,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Test for short name\n",
    "short_name = df1_clean [df1_clean.name.str.len() < 3 ]\n",
    "short_name.name"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 36,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0"
      ]
     },
     "execution_count": 36,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Test for None \n",
    "(df1_clean.name=='None').sum()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Define"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "##### As per the project requirements; only original ratings (no retweets) that have images should be included"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 37,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "2075"
      ]
     },
     "execution_count": 37,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#Use the image_prediction table to guide the selection and removal of tweets without images in the archive table\n",
    "tweets_with_image = list(df2_clean.tweet_id.unique())\n",
    "len(tweets_with_image)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 38,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "True"
      ]
     },
     "execution_count": 38,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# confirming that all the tweets with images exist in the archive dataset\n",
    "len(tweets_with_image) == df1_clean.tweet_id.isin(tweets_with_image).sum()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 39,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Cleaning df1_clean\n",
    "df1_clean = df1_clean[df1_clean.tweet_id.isin(tweets_with_image)]"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Test"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 40,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "(2075, 14)\n",
      "(2075, 12)\n",
      "(2330, 5)\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "True"
      ]
     },
     "execution_count": 40,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "print(df1_clean.shape)\n",
    "print(df2_clean.shape)\n",
    "print(df3_clean.shape)\n",
    "df1_clean.shape[0] == df2_clean.shape[0]"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Define"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "##### Columns of retweeted_status_id and its related data have entries which not part of our analysis.\n",
    "###### Use its notnull data to drop retweets and replies then drop it all in df1_clean and df3_clean."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 41,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "81"
      ]
     },
     "execution_count": 41,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# extract the tweets that include data in the retweet_status_id.\n",
    "retweet_entries = df1_clean.retweeted_status_id.notnull()\n",
    "df1_clean[retweet_entries].shape[0]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 42,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Dropping the retweets from the archive data set\n",
    "df1_clean = df1_clean[~retweet_entries]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 43,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "23"
      ]
     },
     "execution_count": 43,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Extracting replies entries with the same method as the retweets\n",
    "reply_entries = df1_clean.in_reply_to_status_id.notnull()\n",
    "df1_clean[reply_entries].shape[0]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 44,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Dropping the replies from the df1_clean set\n",
    "df1_clean = df1_clean[~reply_entries]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 45,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "104"
      ]
     },
     "execution_count": 45,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Check the image_prediction table df2_clean for extra tweet ids not in the df1_clean\n",
    "np.logical_not(df2_clean.tweet_id.isin(list(df1_clean.tweet_id))).sum()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 46,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Dropping the retweets and replies ids from the image prediction dataframe df2_clean\n",
    "df2_clean = df2_clean[~np.logical_not(df2_clean.tweet_id.isin(list(df1_clean.tweet_id)))]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 47,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "366"
      ]
     },
     "execution_count": 47,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Check the df3_clean table for extra tweet ids not in the df1_clean\n",
    "np.logical_not(df3_clean.tweet_id.isin(list(df1_clean.tweet_id))).sum()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 48,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Dropping the retweets and replies ids from the dataframe df3_clean\n",
    "df3_clean = df3_clean[~np.logical_not(df3_clean.tweet_id.isin(list(df1_clean.tweet_id)))]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 49,
   "metadata": {},
   "outputs": [],
   "source": [
    "#Dropping columns in_reply_to_status_id, in_reply_to_user_id, retweeted_status_id, retweeted_status_user_id, retweeted_status_timestamp\n",
    "df1_clean.drop(['in_reply_to_status_id', 'in_reply_to_user_id', 'retweeted_status_id', 'retweeted_status_user_id', 'retweeted_status_timestamp'],axis=1,inplace=True)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Test"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 50,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "(1971, 9)\n",
      "(1971, 12)\n",
      "(1964, 5)\n"
     ]
    }
   ],
   "source": [
    "print(df1_clean.shape)\n",
    "print(df2_clean.shape)\n",
    "print(df3_clean.shape)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 51,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "True\n",
      "True\n",
      "True\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "<ipython-input-51-c0f698a2d89c>:1: UserWarning: Boolean Series key will be reindexed to match DataFrame index.\n",
      "  print((df1_clean[retweet_entries].shape[0])==0)\n",
      "<ipython-input-51-c0f698a2d89c>:2: UserWarning: Boolean Series key will be reindexed to match DataFrame index.\n",
      "  print((df1_clean[reply_entries].shape[0])==0)\n"
     ]
    }
   ],
   "source": [
    "print((df1_clean[retweet_entries].shape[0])==0)\n",
    "print((df1_clean[reply_entries].shape[0])==0)\n",
    "print(np.logical_not((df2_clean.tweet_id.isin(list(df1_clean.tweet_id))).sum())==0)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 52,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "RangeIndex(start=0, stop=1971, step=1)"
      ]
     },
     "execution_count": 52,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df1_clean = df1_clean.reset_index(drop=True)\n",
    "df1_clean.index"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 53,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "RangeIndex(start=0, stop=1971, step=1)"
      ]
     },
     "execution_count": 53,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df2_clean = df2_clean.reset_index(drop=True)\n",
    "df2_clean.index"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 54,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "RangeIndex(start=0, stop=1964, step=1)"
      ]
     },
     "execution_count": 54,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df3_clean = df3_clean.reset_index(drop=True)\n",
    "df3_clean.index"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 55,
   "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>dog_stages</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>372</th>\n",
       "      <td>813051746834595840</td>\n",
       "      <td>2016-12-25 16:00:16 +0000</td>\n",
       "      <td>&lt;a href=\"http://twitter.com/download/iphone\" r...</td>\n",
       "      <td>This is Bauer. He had nothing to do with the c...</td>\n",
       "      <td>https://twitter.com/dog_rates/status/813051746...</td>\n",
       "      <td>13</td>\n",
       "      <td>10</td>\n",
       "      <td>Bauer</td>\n",
       "      <td></td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>800</th>\n",
       "      <td>739844404073074688</td>\n",
       "      <td>2016-06-06 15:40:26 +0000</td>\n",
       "      <td>&lt;a href=\"http://twitter.com/download/iphone\" r...</td>\n",
       "      <td>This is Simba. He's the grand prize. The troph...</td>\n",
       "      <td>https://twitter.com/dog_rates/status/739844404...</td>\n",
       "      <td>12</td>\n",
       "      <td>10</td>\n",
       "      <td>Simba</td>\n",
       "      <td></td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "               tweet_id                  timestamp  \\\n",
       "372  813051746834595840  2016-12-25 16:00:16 +0000   \n",
       "800  739844404073074688  2016-06-06 15:40:26 +0000   \n",
       "\n",
       "                                                source  \\\n",
       "372  <a href=\"http://twitter.com/download/iphone\" r...   \n",
       "800  <a href=\"http://twitter.com/download/iphone\" r...   \n",
       "\n",
       "                                                  text  \\\n",
       "372  This is Bauer. He had nothing to do with the c...   \n",
       "800  This is Simba. He's the grand prize. The troph...   \n",
       "\n",
       "                                         expanded_urls  rating_numerator  \\\n",
       "372  https://twitter.com/dog_rates/status/813051746...                13   \n",
       "800  https://twitter.com/dog_rates/status/739844404...                12   \n",
       "\n",
       "     rating_denominator   name dog_stages  \n",
       "372                  10  Bauer             \n",
       "800                  10  Simba             "
      ]
     },
     "execution_count": 55,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df1_clean.sample(2)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 56,
   "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>1070</th>\n",
       "      <td>719332531645071360</td>\n",
       "      <td>https://pbs.twimg.com/media/CfuVGl3WEAEKb16.jpg</td>\n",
       "      <td>1</td>\n",
       "      <td>Dandie_Dinmont</td>\n",
       "      <td>0.224415</td>\n",
       "      <td>True</td>\n",
       "      <td>miniature_poodle</td>\n",
       "      <td>0.204882</td>\n",
       "      <td>True</td>\n",
       "      <td>Norfolk_terrier</td>\n",
       "      <td>0.090633</td>\n",
       "      <td>True</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>583</th>\n",
       "      <td>679475951516934144</td>\n",
       "      <td>https://pbs.twimg.com/media/CW37xZbUoAAUXe5.jpg</td>\n",
       "      <td>1</td>\n",
       "      <td>Maltese_dog</td>\n",
       "      <td>0.145742</td>\n",
       "      <td>True</td>\n",
       "      <td>toy_poodle</td>\n",
       "      <td>0.139407</td>\n",
       "      <td>True</td>\n",
       "      <td>West_Highland_white_terrier</td>\n",
       "      <td>0.108821</td>\n",
       "      <td>True</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                tweet_id                                          jpg_url  \\\n",
       "1070  719332531645071360  https://pbs.twimg.com/media/CfuVGl3WEAEKb16.jpg   \n",
       "583   679475951516934144  https://pbs.twimg.com/media/CW37xZbUoAAUXe5.jpg   \n",
       "\n",
       "      img_num              p1   p1_conf  p1_dog                p2   p2_conf  \\\n",
       "1070        1  Dandie_Dinmont  0.224415    True  miniature_poodle  0.204882   \n",
       "583         1     Maltese_dog  0.145742    True        toy_poodle  0.139407   \n",
       "\n",
       "      p2_dog                           p3   p3_conf  p3_dog  \n",
       "1070    True              Norfolk_terrier  0.090633    True  \n",
       "583     True  West_Highland_white_terrier  0.108821    True  "
      ]
     },
     "execution_count": 56,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df2_clean.sample(2)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 57,
   "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>Unnamed: 0</th>\n",
       "      <th>tweet_id</th>\n",
       "      <th>retweet_count</th>\n",
       "      <th>favorite_count</th>\n",
       "      <th>followers_count</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>53</th>\n",
       "      <td>59</td>\n",
       "      <td>880465832366813184</td>\n",
       "      <td>5478</td>\n",
       "      <td>26015</td>\n",
       "      <td>8941394</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>676</th>\n",
       "      <td>906</td>\n",
       "      <td>754482103782404096</td>\n",
       "      <td>1897</td>\n",
       "      <td>5154</td>\n",
       "      <td>8941406</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "     Unnamed: 0            tweet_id  retweet_count  favorite_count  \\\n",
       "53           59  880465832366813184           5478           26015   \n",
       "676         906  754482103782404096           1897            5154   \n",
       "\n",
       "     followers_count  \n",
       "53           8941394  \n",
       "676          8941406  "
      ]
     },
     "execution_count": 57,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df3_clean.sample(2)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Define"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "##### Invalid dtype in timestamp represented as object type."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 58,
   "metadata": {},
   "outputs": [],
   "source": [
    "#convert timestamp to date time type \n",
    "df1_clean['timestamp'] = pd.to_datetime(df1_clean['timestamp'])"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Test"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 59,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "datetime64[ns, UTC]"
      ]
     },
     "execution_count": 59,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df1_clean.timestamp.dtypes"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Define"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "##### Inaccurate data in rating_numerator and df_denominator.\n",
    "##### the denominator should always be 10 but some aren't in all cases So dividing by it or not doesn't mattar.\n",
    "##### we are going to drop denominator later and just use the rating_numerator as the rating value.\n",
    "##### some rating extracted incorrectly \n",
    "- extract it with REGS\n",
    "\n",
    "##### Most images with more than dog have denominator greater than 10 and numerator greater than 20.\n",
    "- Calculating the dogs counts in the tweets featuring many dogs by dividing denominator/10\n",
    "- Calculating the rating_numerator for images with more than one dog counts by dividing numerator/dogs counts\n",
    "\n",
    "##### investigate each outlier individually through its image found some images not a dog and some not found and those will be dropped:\n",
    "- Drop rows"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 60,
   "metadata": {},
   "outputs": [],
   "source": [
    "#try to extract correct rating with REGS\n",
    "df1_clean['rating_numerator'] = df1_clean.text.str.extract('(\\d+\\.?\\d?\\d?)\\/\\d{1,3}', expand = False).astype('float')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 61,
   "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>rating_numerator</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>320</th>\n",
       "      <td>820690176645140481</td>\n",
       "      <td>84.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>382</th>\n",
       "      <td>810984652412424192</td>\n",
       "      <td>24.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>658</th>\n",
       "      <td>758467244762497024</td>\n",
       "      <td>165.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>722</th>\n",
       "      <td>749981277374128128</td>\n",
       "      <td>1776.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>842</th>\n",
       "      <td>731156023742988288</td>\n",
       "      <td>204.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>918</th>\n",
       "      <td>716439118184652801</td>\n",
       "      <td>50.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>939</th>\n",
       "      <td>713900603437621249</td>\n",
       "      <td>99.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>963</th>\n",
       "      <td>710658690886586372</td>\n",
       "      <td>80.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>981</th>\n",
       "      <td>709198395643068416</td>\n",
       "      <td>45.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1045</th>\n",
       "      <td>704054845121142784</td>\n",
       "      <td>60.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1120</th>\n",
       "      <td>697463031882764288</td>\n",
       "      <td>44.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1288</th>\n",
       "      <td>684222868335505415</td>\n",
       "      <td>121.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1420</th>\n",
       "      <td>677716515794329600</td>\n",
       "      <td>144.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1478</th>\n",
       "      <td>675853064436391936</td>\n",
       "      <td>88.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1696</th>\n",
       "      <td>670842764863651840</td>\n",
       "      <td>420.0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                tweet_id  rating_numerator\n",
       "320   820690176645140481              84.0\n",
       "382   810984652412424192              24.0\n",
       "658   758467244762497024             165.0\n",
       "722   749981277374128128            1776.0\n",
       "842   731156023742988288             204.0\n",
       "918   716439118184652801              50.0\n",
       "939   713900603437621249              99.0\n",
       "963   710658690886586372              80.0\n",
       "981   709198395643068416              45.0\n",
       "1045  704054845121142784              60.0\n",
       "1120  697463031882764288              44.0\n",
       "1288  684222868335505415             121.0\n",
       "1420  677716515794329600             144.0\n",
       "1478  675853064436391936              88.0\n",
       "1696  670842764863651840             420.0"
      ]
     },
     "execution_count": 61,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#check for incorrect rating assume that over rated numerator more than 20\n",
    "inaccurat_rating = df1_clean[df1_clean.rating_numerator>20][['tweet_id','rating_numerator']]\n",
    "inaccurat_rating"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 62,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "320      7.0\n",
      "382      0.7\n",
      "658     15.0\n",
      "722      1.0\n",
      "842     17.0\n",
      "918      5.0\n",
      "939      9.0\n",
      "963      8.0\n",
      "981      5.0\n",
      "1045     5.0\n",
      "1120     4.0\n",
      "1288    11.0\n",
      "1420    12.0\n",
      "1478     8.0\n",
      "1696     1.0\n",
      "Name: rating_denominator, dtype: float64\n"
     ]
    }
   ],
   "source": [
    "# Calculating the dogs counts in the tweets featuring many dogs\n",
    "dogs_count = df1_clean.rating_denominator[df1_clean.rating_numerator > 20 ]/10\n",
    "print(dogs_count)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 63,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "C:\\Users\\ZENAB\\anaconda3\\lib\\site-packages\\numpy\\core\\_asarray.py:83: VisibleDeprecationWarning: Creating an ndarray from ragged nested sequences (which is a list-or-tuple of lists-or-tuples-or ndarrays with different lengths or shapes) is deprecated. If you meant to do this, you must specify 'dtype=object' when creating the ndarray\n",
      "  return array(a, dtype, copy=False, order=order)\n"
     ]
    }
   ],
   "source": [
    "df1_clean.loc[df1_clean.rating_numerator > 20, ['rating_numerator', 'rating_denominator']] = [inaccurat_rating.rating_numerator/dogs_count , 10]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 64,
   "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>rating_numerator</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>382</th>\n",
       "      <td>810984652412424192</td>\n",
       "      <td>34.285714</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>722</th>\n",
       "      <td>749981277374128128</td>\n",
       "      <td>1776.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1696</th>\n",
       "      <td>670842764863651840</td>\n",
       "      <td>420.000000</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                tweet_id  rating_numerator\n",
       "382   810984652412424192         34.285714\n",
       "722   749981277374128128       1776.000000\n",
       "1696  670842764863651840        420.000000"
      ]
     },
     "execution_count": 64,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#investigate incorrect rating and drop\n",
    "inaccurat_rating = df1_clean[df1_clean.rating_numerator>20][['tweet_id','rating_numerator']]\n",
    "inaccurat_rating"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 65,
   "metadata": {},
   "outputs": [],
   "source": [
    "df1_clean = df1_clean.drop([382,722, 1696])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 66,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Dropping the extra tweet ids from df2_clean not in the df1_clean\n",
    "df2_clean = df2_clean[~np.logical_not(df2_clean.tweet_id.isin(list(df1_clean.tweet_id)))]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 67,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Dropping the extra tweet ids from df3_clean not in the df1_clean\n",
    "df3_clean = df3_clean[~np.logical_not(df3_clean.tweet_id.isin(list(df1_clean.tweet_id)))]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 68,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "0\n",
      "0\n"
     ]
    }
   ],
   "source": [
    "print(np.logical_not(df3_clean.tweet_id.isin(list(df1_clean.tweet_id))).sum())\n",
    "print(np.logical_not(df3_clean.tweet_id.isin(list(df1_clean.tweet_id))).sum())"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 69,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "RangeIndex(start=0, stop=1968, step=1)"
      ]
     },
     "execution_count": 69,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df1_clean = df1_clean.reset_index(drop=True)\n",
    "df1_clean.index"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 70,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "RangeIndex(start=0, stop=1968, step=1)"
      ]
     },
     "execution_count": 70,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df2_clean = df2_clean.reset_index(drop=True)\n",
    "df2_clean.index"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 71,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "RangeIndex(start=0, stop=1961, step=1)"
      ]
     },
     "execution_count": 71,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df3_clean = df3_clean.reset_index(drop=True)\n",
    "df3_clean.index"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Test"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 72,
   "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>rating_numerator</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "Empty DataFrame\n",
       "Columns: [tweet_id, rating_numerator]\n",
       "Index: []"
      ]
     },
     "execution_count": 72,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df1_clean[df1_clean.rating_numerator>20][['tweet_id','rating_numerator']]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 73,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "(1968, 9)\n",
      "(1968, 12)\n",
      "(1961, 5)\n"
     ]
    }
   ],
   "source": [
    "print(df1_clean.shape)\n",
    "print(df2_clean.shape)\n",
    "print(df3_clean.shape)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Define"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "##### Column tweet_id represented as int64 across all data set"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 74,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Convret to str\n",
    "df1_clean.loc[:,['tweet_id']] = df1_clean.loc[:,['tweet_id']].astype(np.int64).astype(str)\n",
    "df2_clean.loc[:,['tweet_id']] = df2_clean.loc[:,['tweet_id']].astype(np.int64).astype(str)\n",
    "df3_clean.loc[:,['tweet_id']] = df3_clean.loc[:,['tweet_id']].astype(np.int64).astype(str)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Test"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 75,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "object\n",
      "object\n",
      "object\n"
     ]
    }
   ],
   "source": [
    "print(df1_clean.tweet_id.dtypes)\n",
    "print(df2_clean.tweet_id.dtypes)\n",
    "print(df3_clean.tweet_id.dtypes)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Define"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 76,
   "metadata": {},
   "outputs": [],
   "source": [
    "false_p1dog = df2_clean[df2_clean[\"p1_dog\"]== False]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 77,
   "metadata": {},
   "outputs": [],
   "source": [
    "false_p2dog = false_p1dog[false_p1dog['p2_dog']==False]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 78,
   "metadata": {},
   "outputs": [],
   "source": [
    "false_p3dog = false_p2dog[false_p2dog['p3_dog']==False]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 79,
   "metadata": {},
   "outputs": [],
   "source": [
    "df2_clean = df2_clean.drop(false_p3dog.index,axis = 0)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 80,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "RangeIndex(start=0, stop=1665, step=1)"
      ]
     },
     "execution_count": 80,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df2_clean = df2_clean.reset_index(drop=True)\n",
    "df2_clean.index"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Test"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 81,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(1665, 12)"
      ]
     },
     "execution_count": 81,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df2_clean.shape"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Define\n",
    "##### One variable 'prediction' have three columns.\n",
    "##### Columns of predictions have lables represented as letters.\n",
    "- Use wide to long method to correct table tiddness issue"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 82,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Rename columns of dataframe\n",
    "cols = ['tweet_id', 'jpg_url', 'img_num', \n",
    "       'prediction_1', 'confidence_1', 'breed_1',\n",
    "       'prediction_2', 'confidence_2', 'breed_2',\n",
    "       'prediction_3', 'confidence_3', 'breed_3']\n",
    "df2_clean.columns = cols"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 83,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Reshaping the dataframe\n",
    "df2_clean= pd.wide_to_long(df2_clean, stubnames=['prediction', 'confidence', 'breed'], \n",
    "    i=['tweet_id', 'jpg_url', 'img_num'], j='prediction_level', sep=\"_\").reset_index()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Test"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 84,
   "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>prediction_level</th>\n",
       "      <th>prediction</th>\n",
       "      <th>confidence</th>\n",
       "      <th>breed</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>1</td>\n",
       "      <td>Welsh_springer_spaniel</td>\n",
       "      <td>0.465074</td>\n",
       "      <td>True</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>666020888022790149</td>\n",
       "      <td>https://pbs.twimg.com/media/CT4udn0WwAA0aMy.jpg</td>\n",
       "      <td>1</td>\n",
       "      <td>2</td>\n",
       "      <td>collie</td>\n",
       "      <td>0.156665</td>\n",
       "      <td>True</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>666020888022790149</td>\n",
       "      <td>https://pbs.twimg.com/media/CT4udn0WwAA0aMy.jpg</td>\n",
       "      <td>1</td>\n",
       "      <td>3</td>\n",
       "      <td>Shetland_sheepdog</td>\n",
       "      <td>0.061428</td>\n",
       "      <td>True</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>666029285002620928</td>\n",
       "      <td>https://pbs.twimg.com/media/CT42GRgUYAA5iDo.jpg</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>redbone</td>\n",
       "      <td>0.506826</td>\n",
       "      <td>True</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>666029285002620928</td>\n",
       "      <td>https://pbs.twimg.com/media/CT42GRgUYAA5iDo.jpg</td>\n",
       "      <td>1</td>\n",
       "      <td>2</td>\n",
       "      <td>miniature_pinscher</td>\n",
       "      <td>0.074192</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  666020888022790149  https://pbs.twimg.com/media/CT4udn0WwAA0aMy.jpg   \n",
       "2  666020888022790149  https://pbs.twimg.com/media/CT4udn0WwAA0aMy.jpg   \n",
       "3  666029285002620928  https://pbs.twimg.com/media/CT42GRgUYAA5iDo.jpg   \n",
       "4  666029285002620928  https://pbs.twimg.com/media/CT42GRgUYAA5iDo.jpg   \n",
       "\n",
       "   img_num  prediction_level              prediction  confidence  breed  \n",
       "0        1                 1  Welsh_springer_spaniel    0.465074   True  \n",
       "1        1                 2                  collie    0.156665   True  \n",
       "2        1                 3       Shetland_sheepdog    0.061428   True  \n",
       "3        1                 1                 redbone    0.506826   True  \n",
       "4        1                 2      miniature_pinscher    0.074192   True  "
      ]
     },
     "execution_count": 84,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df2_clean.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Define\n",
    "##### Table df3_clean have same observation unit that in df1_clean.\n",
    "- use merge function to make them one table called Twitter_archive_master"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 85,
   "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>dog_stages</th>\n",
       "      <th>Unnamed: 0</th>\n",
       "      <th>retweet_count</th>\n",
       "      <th>favorite_count</th>\n",
       "      <th>followers_count</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>868</th>\n",
       "      <td>724405726123311104</td>\n",
       "      <td>2016-04-25 01:12:38+00:00</td>\n",
       "      <td>&lt;a href=\"http://twitter.com/download/iphone\" r...</td>\n",
       "      <td>This is Ashleigh. She's having Coachella withd...</td>\n",
       "      <td>https://twitter.com/dog_rates/status/724405726...</td>\n",
       "      <td>10.0</td>\n",
       "      <td>10</td>\n",
       "      <td>Ashleigh</td>\n",
       "      <td>pupper</td>\n",
       "      <td>1133</td>\n",
       "      <td>1592</td>\n",
       "      <td>5113</td>\n",
       "      <td>8941414</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "               tweet_id                 timestamp  \\\n",
       "868  724405726123311104 2016-04-25 01:12:38+00:00   \n",
       "\n",
       "                                                source  \\\n",
       "868  <a href=\"http://twitter.com/download/iphone\" r...   \n",
       "\n",
       "                                                  text  \\\n",
       "868  This is Ashleigh. She's having Coachella withd...   \n",
       "\n",
       "                                         expanded_urls  rating_numerator  \\\n",
       "868  https://twitter.com/dog_rates/status/724405726...              10.0   \n",
       "\n",
       "     rating_denominator      name dog_stages  Unnamed: 0  retweet_count  \\\n",
       "868                  10  Ashleigh     pupper        1133           1592   \n",
       "\n",
       "     favorite_count  followers_count  \n",
       "868            5113          8941414  "
      ]
     },
     "execution_count": 85,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "Twitter_archive_master = df1_clean.merge(df3_clean,how = 'inner',on = 'tweet_id')\n",
    "Twitter_archive_master.sample()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 86,
   "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>dog_stages</th>\n",
       "      <th>retweet_count</th>\n",
       "      <th>favorite_count</th>\n",
       "      <th>followers_count</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>1594</th>\n",
       "      <td>672980819271634944</td>\n",
       "      <td>2015-12-05 03:28:25+00:00</td>\n",
       "      <td>&lt;a href=\"http://twitter.com/download/iphone\" r...</td>\n",
       "      <td>Extraordinary dog here. Looks large. Just a he...</td>\n",
       "      <td>https://twitter.com/dog_rates/status/672980819...</td>\n",
       "      <td>5.0</td>\n",
       "      <td>10</td>\n",
       "      <td>NaN</td>\n",
       "      <td></td>\n",
       "      <td>938</td>\n",
       "      <td>1706</td>\n",
       "      <td>8941426</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1515</th>\n",
       "      <td>674739953134403584</td>\n",
       "      <td>2015-12-09 23:58:35+00:00</td>\n",
       "      <td>&lt;a href=\"http://twitter.com/download/iphone\" r...</td>\n",
       "      <td>\"🎶 DO YOU BELIEVE IN LIFE AFTER LOVE 🎶\"\\n11/10...</td>\n",
       "      <td>https://twitter.com/dog_rates/status/674739953...</td>\n",
       "      <td>11.0</td>\n",
       "      <td>10</td>\n",
       "      <td>NaN</td>\n",
       "      <td></td>\n",
       "      <td>359</td>\n",
       "      <td>1031</td>\n",
       "      <td>8941422</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1939</th>\n",
       "      <td>666293911632134144</td>\n",
       "      <td>2015-11-16 16:37:02+00:00</td>\n",
       "      <td>&lt;a href=\"http://twitter.com/download/iphone\" r...</td>\n",
       "      <td>This is a funny dog. Weird toes. Won't come do...</td>\n",
       "      <td>https://twitter.com/dog_rates/status/666293911...</td>\n",
       "      <td>3.0</td>\n",
       "      <td>10</td>\n",
       "      <td>NaN</td>\n",
       "      <td></td>\n",
       "      <td>309</td>\n",
       "      <td>450</td>\n",
       "      <td>8941430</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                tweet_id                 timestamp  \\\n",
       "1594  672980819271634944 2015-12-05 03:28:25+00:00   \n",
       "1515  674739953134403584 2015-12-09 23:58:35+00:00   \n",
       "1939  666293911632134144 2015-11-16 16:37:02+00:00   \n",
       "\n",
       "                                                 source  \\\n",
       "1594  <a href=\"http://twitter.com/download/iphone\" r...   \n",
       "1515  <a href=\"http://twitter.com/download/iphone\" r...   \n",
       "1939  <a href=\"http://twitter.com/download/iphone\" r...   \n",
       "\n",
       "                                                   text  \\\n",
       "1594  Extraordinary dog here. Looks large. Just a he...   \n",
       "1515  \"🎶 DO YOU BELIEVE IN LIFE AFTER LOVE 🎶\"\\n11/10...   \n",
       "1939  This is a funny dog. Weird toes. Won't come do...   \n",
       "\n",
       "                                          expanded_urls  rating_numerator  \\\n",
       "1594  https://twitter.com/dog_rates/status/672980819...               5.0   \n",
       "1515  https://twitter.com/dog_rates/status/674739953...              11.0   \n",
       "1939  https://twitter.com/dog_rates/status/666293911...               3.0   \n",
       "\n",
       "      rating_denominator name dog_stages  retweet_count  favorite_count  \\\n",
       "1594                  10  NaN                       938            1706   \n",
       "1515                  10  NaN                       359            1031   \n",
       "1939                  10  NaN                       309             450   \n",
       "\n",
       "      followers_count  \n",
       "1594          8941426  \n",
       "1515          8941422  \n",
       "1939          8941430  "
      ]
     },
     "execution_count": 86,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "Twitter_archive_master.drop(['Unnamed: 0'], axis = 1, inplace = True)\n",
    "Twitter_archive_master.sample(3)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Store Data"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "##### Export data as CSV file\n",
    "            - Twitter_archive_master\n",
    "            - Image_prediction"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 87,
   "metadata": {},
   "outputs": [],
   "source": [
    "Twitter_archive_master.to_csv('work_directory/Twitter_archive_master.csv')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 88,
   "metadata": {},
   "outputs": [],
   "source": [
    "df2_clean.to_csv('work_directory/Image_prediction.csv')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Data Analysis, Viz & Reporting"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 89,
   "metadata": {},
   "outputs": [],
   "source": [
    "df1 = pd.read_csv('work_directory/Twitter_archive_master.csv')\n",
    "df2= pd.read_csv('work_directory/Image_prediction.csv')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 90,
   "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>Unnamed: 0</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>dog_stages</th>\n",
       "      <th>retweet_count</th>\n",
       "      <th>favorite_count</th>\n",
       "      <th>followers_count</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>853</th>\n",
       "      <td>853</td>\n",
       "      <td>727685679342333952</td>\n",
       "      <td>2016-05-04 02:26:00+00:00</td>\n",
       "      <td>&lt;a href=\"http://twitter.com/download/iphone\" r...</td>\n",
       "      <td>This is Cilantro. She's a Fellation Gadzooks. ...</td>\n",
       "      <td>https://twitter.com/dog_rates/status/727685679...</td>\n",
       "      <td>12.0</td>\n",
       "      <td>10</td>\n",
       "      <td>Cilantro</td>\n",
       "      <td>NaN</td>\n",
       "      <td>610</td>\n",
       "      <td>2859</td>\n",
       "      <td>8941413</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "     Unnamed: 0            tweet_id                 timestamp  \\\n",
       "853         853  727685679342333952 2016-05-04 02:26:00+00:00   \n",
       "\n",
       "                                                source  \\\n",
       "853  <a href=\"http://twitter.com/download/iphone\" r...   \n",
       "\n",
       "                                                  text  \\\n",
       "853  This is Cilantro. She's a Fellation Gadzooks. ...   \n",
       "\n",
       "                                         expanded_urls  rating_numerator  \\\n",
       "853  https://twitter.com/dog_rates/status/727685679...              12.0   \n",
       "\n",
       "     rating_denominator      name dog_stages  retweet_count  favorite_count  \\\n",
       "853                  10  Cilantro        NaN            610            2859   \n",
       "\n",
       "     followers_count  \n",
       "853          8941413  "
      ]
     },
     "execution_count": 90,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df1['timestamp']=pd.to_datetime(df1['timestamp'])\n",
    "df1.sample()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 91,
   "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>Unnamed: 0</th>\n",
       "      <th>tweet_id</th>\n",
       "      <th>jpg_url</th>\n",
       "      <th>img_num</th>\n",
       "      <th>prediction_level</th>\n",
       "      <th>prediction</th>\n",
       "      <th>confidence</th>\n",
       "      <th>breed</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>459</th>\n",
       "      <td>459</td>\n",
       "      <td>669603084620980224</td>\n",
       "      <td>https://pbs.twimg.com/media/CUroc7QW4AATIff.jpg</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>Maltese_dog</td>\n",
       "      <td>0.659619</td>\n",
       "      <td>True</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "     Unnamed: 0            tweet_id  \\\n",
       "459         459  669603084620980224   \n",
       "\n",
       "                                             jpg_url  img_num  \\\n",
       "459  https://pbs.twimg.com/media/CUroc7QW4AATIff.jpg        1   \n",
       "\n",
       "     prediction_level   prediction  confidence  breed  \n",
       "459                 1  Maltese_dog    0.659619   True  "
      ]
     },
     "execution_count": 91,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df2.sample()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 92,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1440x432 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "stage =  df1.groupby(['dog_stages'],as_index = True)\n",
    "ind = np.arange(len(stage))  # the x locations for the groups\n",
    "width = 0.5       # the width of the bars\n",
    "labels =stage['rating_numerator'].count().index\n",
    "\n",
    "stage_rating_max = stage ['rating_numerator'].max()\n",
    "stage_rating_mean = stage['rating_numerator'].mean()\n",
    "stage_rating_min =stage['rating_numerator'].min()\n",
    "\n",
    "plt.figure(figsize=(20,6))\n",
    "\n",
    "plt.bar(ind+width/3,stage_rating_max,color= 'b',  width= width/3, alpha = .7);\n",
    "plt.bar(ind,stage_rating_mean,color= 'c', width= width/3, alpha = .7);\n",
    "plt.bar(ind-width/3,stage_rating_min,color= 'w', width= width/3, alpha = .7);\n",
    "\n",
    "\n",
    "plt.title('Rating distribution per Dog Stage',size = 40)\n",
    "plt.ylabel('Rating' ,color='b', size=22);\n",
    "plt.xticks(ticks= ind, labels=labels, fontsize=16);"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## The previous figure show \"rating distribution\" for each \"dog stage\", though we could not compar each stage to others because the \"rating count\" not involved in our calculation.\n",
    "\n",
    "## ie. the mean of \"doggopuppo\" the highest while the \"pupper\" is the lowest, then if we get each stage rating count relative to the total stages rating count into our calculation as a proportional factor we realize that the relative mean of \"doggopuppo\" the lowest while for the \"pupper\" is the highest.\n",
    "\n",
    "## Next figure will introduce proportional factor into calculation. \n",
    "## But befor move forward I would represent another sight. Have a look at \"doggopuppo\" and \"doggofloofer\" distribution there is no \"Variance\". The \"max\", \"min\", and \"mean\" all have the same value. that give us intuation about the data beneath bars, I guess its only one observation. let's check it out."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 93,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "dog_stages\n",
       "doggo            62\n",
       "doggofloofer      1\n",
       "doggopupper       8\n",
       "doggopuppo        1\n",
       "floofer           7\n",
       "pupper          201\n",
       "puppo            22\n",
       "Name: rating_numerator, dtype: int64"
      ]
     },
     "execution_count": 93,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "stage['rating_numerator'].count()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## That's what I am talking about. Both stages \"doggopuppo\" and \"doggofloofer\" only have one observation.\n",
    "\n",
    "----------------------------------------------------------------------------------------------------------------------------------------\n",
    "\n",
    "\n",
    "\n",
    "## Let's move forward and check \"Relative Mean Rating\"."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 94,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1440x432 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "stage_rating_count = stage.rating_numerator.count()\n",
    "\n",
    "total_rating_count =stage.rating_numerator.count().sum()\n",
    "\n",
    "proportion = stage_rating_count/total_rating_count\n",
    "\n",
    "stage_rating_proportion = stage_rating_mean * proportion\n",
    "\n",
    "plt.figure(figsize=(20,6))\n",
    "\n",
    "plt.bar(ind,stage_rating_proportion,color='c', width=width/3,alpha=.9);\n",
    "\n",
    "plt.title('Relative Mean Rating VSDog Stage',size = 40)\n",
    "plt.ylabel('Relative Mean Rating' ,color='r', size=22);\n",
    "plt.xticks(ticks= ind, labels=labels, fontsize=16);"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Here it is, the Relative Mean Rating which you can compare stages to each other.\n",
    "## That prove the point we mentioned early the relative mean of \"doggopuppo\" the lowest while for the \"pupper\" is the highest.\n",
    "\n",
    "--------------------------------------------------------------------------------------------------------------------------------------"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## In the next will examine the Favorite and Retweets for each stage but with `.sum` which will represent actual count of observation in another word no need to get the proportional ratio, henc it will represented directly."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 95,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "dog_stages\n",
       "doggo           1138255\n",
       "doggofloofer      15452\n",
       "doggopupper      106097\n",
       "doggopuppo        43418\n",
       "floofer           81983\n",
       "pupper          1312025\n",
       "puppo            448441\n",
       "Name: favorite_count, dtype: int64"
      ]
     },
     "execution_count": 95,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    " stage ['favorite_count'].sum()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 96,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1440x432 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "stage_total_favorite = stage ['favorite_count'].sum()\n",
    "\n",
    "stage_total_retweet = stage['retweet_count'].sum()\n",
    "\n",
    "plt.figure(figsize=(20,6))\n",
    "\n",
    "plt.bar(ind+width/6,stage_total_favorite,color='c', width=width/3,alpha=.7,label='favorite');\n",
    "plt.bar(ind-width/6,stage_total_retweet,color='w', width=width/3,alpha=.7,label='retweet');\n",
    "\n",
    "plt.title('Total Number of Favorites and Retweets per Dog Stage',size = 40)\n",
    "plt.ylabel('Favorite and Retweets Count ' ,color='b', size=22);\n",
    "plt.xticks(ticks= ind, labels=labels, fontsize=16);\n",
    "plt.legend(fontsize=20);"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Same data with diffrent visuelisation"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 97,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1584x576 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "fig, ax = plt.subplots()\n",
    "fig.set_size_inches(22, 8)\n",
    "ax.bar(labels, stage_total_favorite, width, label='Favorite')\n",
    "ax.bar(labels, stage_total_retweet, width, bottom=stage_total_favorite,label='Retweet',color='orange')\n",
    "\n",
    "plt.title('Total Number of Favorites and Retweets per Dog Stage',size = 40)\n",
    "plt.ylabel('Favorite and Retweets Count' ,color='r', size=22);\n",
    "plt.xticks(ticks= ind, labels=labels, fontsize=16);\n",
    "plt.legend(fontsize=20);\n",
    "\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "---------------------------------------------------------------------------------------------------------------------------------------\n",
    "\n",
    "\n",
    "## Now How is the interaction? is people get intersted with page posts and contents? and is it growth over time or not?\n",
    "## All these question could be answerd via the interaction over time which figure out the people engagement."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 98,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1440x432 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1440x432 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "df1['year']=df1['timestamp'].dt.year\n",
    "df1['month']=df1['timestamp'].dt.month\n",
    "time = df1.groupby(['year','month'],as_index = True)\n",
    "fav_time = time['favorite_count'].mean()\n",
    "ret_time = time['retweet_count'].mean()\n",
    "fol_time = time['followers_count'].mean()\n",
    "\n",
    "plt.figure(figsize=(20,6))\n",
    "\n",
    "fav_time.plot(kind='line',color='r',fontsize=22)\n",
    "ret_time.plot(kind='line',color='g',fontsize=22)\n",
    "\n",
    "\n",
    "plt.title(' Interaction overtime ',size = 40)\n",
    "plt.ylabel('Favorite and Retweet Counts ' , size=22);\n",
    "plt.xlabel('Year, Month ' , size=22);\n",
    "plt.legend(fontsize=20);\n",
    "\n",
    "plt.figure(figsize=(20,6))\n",
    "fol_time.plot(kind='line',color='g',fontsize=22)\n",
    "\n",
    "plt.ylabel('followers count' , size=22);\n",
    "plt.xlabel('Year, Month ' , size=22);\n",
    "plt.legend(fontsize=20);\n",
    "\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# people interaction \n",
    "## The Peopel interaction via favorits and retweets is growing as shown in the first figure. Notice that favorites always more than retweets that because less people have intersted in writing posts, while most of them like and go.\n",
    "## in the other hand we see the followers count over time descending!! is that wierd? what does it mean?\n",
    "## Don't confuse, it's concept of a new-follwer. Early all people was a new follower who can hit follow. Then after aperiod of time as the number of follwers increased the new followers will decreased. Make sense\n",
    "\n",
    "\n",
    "---------------------------------------------------------------------------------------------------------------------------------------"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Image Prediction \n",
    "## In the following figure we Examin the second data set which show the confident destribution for various prediction level.\n",
    "## clearly it show that the first prediction has the most confident level "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 99,
   "metadata": {},
   "outputs": [],
   "source": [
    "image = df2.groupby(['prediction_level'],as_index = True)\n",
    "ind2 = np.arange(len(image))  # the x locations for the groups\n",
    "width2 = 0.5     # the width of the bars\n",
    "labels2 =image['prediction_level'].max().index"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 100,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1440x432 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "max_conf=image.confidence.max()\n",
    "mean_conf=image.confidence.mean()\n",
    "min_conf=image.confidence.min()\n",
    "\n",
    "plt.figure(figsize=(20,6))\n",
    "\n",
    "plt.bar(ind2+width2/3,max_conf,color= 'b',  width= width2/3,alpha=.7);\n",
    "plt.bar(ind2,mean_conf,color='c', width=width2/3,alpha=.7);\n",
    "plt.bar(ind2-width2/3,min_conf,color='w', width=width2/3,alpha=.7);\n",
    "\n",
    "plt.title('Image Prediction Confident Destribution',color='b',size = 40)\n",
    "plt.ylabel('Confiden Level' ,color='b', size=22);\n",
    "plt.xticks(ticks= ind2, labels=labels2, fontsize=16);"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 101,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1440x432 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.figure(figsize=(20,6))\n",
    "df2.prediction.value_counts()[np.arange(10)].plot(kind = 'bar', color= 'y');\n",
    "xlabels = df2.prediction.value_counts()[np.arange(10)].index\n",
    "\n",
    "plt.title('Top Ten Dog Bread',color='b',size = 40)\n",
    "plt.ylabel('Dog Count' ,color='b', size=22);\n",
    "plt.xticks(ticks=np.arange(10), labels= xlabels, fontsize=18);"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Here the top 10 dog bread commonly shared over the page and people interact with"
   ]
  }
 ],
 "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": "3.8.5"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 4
}