eda_neural_network.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [],
   "source": [
    "import numpy as np\n",
    "import pandas as pd\n",
    "import matplotlib.pyplot as plt\n",
    "import re\n",
    "import ast\n",
    "# import nltk\n",
    "import glob\n",
    "import tensorflow as tf\n",
    "\n",
    "from sklearn.preprocessing import OneHotEncoder, StandardScaler\n",
    "from sklearn.metrics import accuracy_score, precision_score, recall_score, confusion_matrix\n",
    "from sklearn.model_selection import train_test_split, cross_val_score\n",
    "from sklearn.linear_model import LogisticRegression\n",
    "from sklearn.ensemble import RandomForestClassifier, GradientBoostingClassifier\n",
    "from sklearn.model_selection import GridSearchCV\n",
    "from sklearn.pipeline import Pipeline, make_pipeline\n",
    "\n",
    "# from nltk.sentiment import SentimentIntensityAnalyzer\n",
    "\n",
    "from tensorflow.keras.models import Sequential\n",
    "from tensorflow.keras.layers import Dense, Dropout, Activation\n",
    "from tensorflow.keras.optimizers import SGD\n",
    "from tensorflow.keras.metrics import Precision, Recall, Accuracy, AUC\n",
    " \n",
    "\n",
    "plt.style.use('ggplot')\n",
    "%matplotlib inline"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [],
   "source": [
    "pd.set_option('display.max_columns', 100)\n",
    "pd.set_option('display.max_colwidth', 100)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [],
   "source": [
    "# get data file names\n",
    "filenames = glob.glob('data' + \"/*.csv\")\n",
    "\n",
    "dfs = []\n",
    "for filename in filenames:\n",
    "    dfs.append(pd.read_csv(filename))\n",
    "\n",
    "# Concatenate all data into one DataFrame\n",
    "df = pd.concat(dfs, ignore_index=True)\n",
    "#link = 's3a://helen.assignment/master.csv'"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "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>backers_count</th>\n",
       "      <th>blurb</th>\n",
       "      <th>category</th>\n",
       "      <th>converted_pledged_amount</th>\n",
       "      <th>country</th>\n",
       "      <th>country_displayable_name</th>\n",
       "      <th>created_at</th>\n",
       "      <th>creator</th>\n",
       "      <th>currency</th>\n",
       "      <th>currency_symbol</th>\n",
       "      <th>currency_trailing_code</th>\n",
       "      <th>current_currency</th>\n",
       "      <th>deadline</th>\n",
       "      <th>disable_communication</th>\n",
       "      <th>friends</th>\n",
       "      <th>fx_rate</th>\n",
       "      <th>goal</th>\n",
       "      <th>id</th>\n",
       "      <th>is_backing</th>\n",
       "      <th>is_starrable</th>\n",
       "      <th>is_starred</th>\n",
       "      <th>launched_at</th>\n",
       "      <th>location</th>\n",
       "      <th>name</th>\n",
       "      <th>permissions</th>\n",
       "      <th>photo</th>\n",
       "      <th>pledged</th>\n",
       "      <th>profile</th>\n",
       "      <th>slug</th>\n",
       "      <th>source_url</th>\n",
       "      <th>spotlight</th>\n",
       "      <th>staff_pick</th>\n",
       "      <th>state</th>\n",
       "      <th>state_changed_at</th>\n",
       "      <th>static_usd_rate</th>\n",
       "      <th>urls</th>\n",
       "      <th>usd_pledged</th>\n",
       "      <th>usd_type</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>5</td>\n",
       "      <td>A Free to Win Online Mobile Strategy Game to end the corruption of the rich players always winni...</td>\n",
       "      <td>{\"id\":272,\"name\":\"Mobile Games\",\"slug\":\"games/mobile games\",\"position\":3,\"parent_id\":12,\"parent_...</td>\n",
       "      <td>83</td>\n",
       "      <td>CA</td>\n",
       "      <td>Canada</td>\n",
       "      <td>1506125024</td>\n",
       "      <td>{\"id\":1178629187,\"name\":\"Adam Tompkins\",\"is_registered\":null,\"chosen_currency\":null,\"is_superbac...</td>\n",
       "      <td>CAD</td>\n",
       "      <td>$</td>\n",
       "      <td>True</td>\n",
       "      <td>USD</td>\n",
       "      <td>1507560063</td>\n",
       "      <td>False</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0.70837</td>\n",
       "      <td>1000.0</td>\n",
       "      <td>751419376</td>\n",
       "      <td>NaN</td>\n",
       "      <td>False</td>\n",
       "      <td>NaN</td>\n",
       "      <td>1506350463</td>\n",
       "      <td>{\"id\":22664159,\"name\":\"Mississauga\",\"slug\":\"mississauga-ca\",\"short_name\":\"Mississauga, Canada\",\"...</td>\n",
       "      <td>Iron Age Kingdoms - First Free To Win Mobile Strategy Game</td>\n",
       "      <td>NaN</td>\n",
       "      <td>{\"key\":\"assets/018/444/378/81e59ccab10259506aab6ee340d25b73_original.png\",\"full\":\"https://ksr-ug...</td>\n",
       "      <td>105.0</td>\n",
       "      <td>{\"id\":3151330,\"project_id\":3151330,\"state\":\"inactive\",\"state_changed_at\":1506125024,\"name\":null,...</td>\n",
       "      <td>iron-age-kingdoms-first-free-to-win-mobile-strateg</td>\n",
       "      <td>https://www.kickstarter.com/discover/categories/games/mobile%20games</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>failed</td>\n",
       "      <td>1507560063</td>\n",
       "      <td>0.810764</td>\n",
       "      <td>{\"web\":{\"project\":\"https://www.kickstarter.com/projects/1178629187/iron-age-kingdoms-first-free-...</td>\n",
       "      <td>85.130189</td>\n",
       "      <td>domestic</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>19</td>\n",
       "      <td>Lizard is a dark comedy and modern western, that takes place in the Mojave Desert.</td>\n",
       "      <td>{\"id\":11,\"name\":\"Film &amp; Video\",\"slug\":\"film &amp; video\",\"position\":7,\"color\":16734574,\"urls\":{\"web\"...</td>\n",
       "      <td>2358</td>\n",
       "      <td>US</td>\n",
       "      <td>the United States</td>\n",
       "      <td>1561667598</td>\n",
       "      <td>{\"id\":1624125450,\"name\":\"Darrin Williams\",\"slug\":\"lizardshortfilm\",\"is_registered\":null,\"chosen_...</td>\n",
       "      <td>USD</td>\n",
       "      <td>$</td>\n",
       "      <td>True</td>\n",
       "      <td>USD</td>\n",
       "      <td>1569109188</td>\n",
       "      <td>False</td>\n",
       "      <td>NaN</td>\n",
       "      <td>1.00000</td>\n",
       "      <td>1100.0</td>\n",
       "      <td>289952460</td>\n",
       "      <td>NaN</td>\n",
       "      <td>False</td>\n",
       "      <td>NaN</td>\n",
       "      <td>1563925188</td>\n",
       "      <td>{\"id\":2452741,\"name\":\"Mojave\",\"slug\":\"mojave-ca\",\"short_name\":\"Mojave, CA\",\"displayable_name\":\"M...</td>\n",
       "      <td>\"Lizard\" a short film</td>\n",
       "      <td>NaN</td>\n",
       "      <td>{\"key\":\"assets/025/890/832/ff7ffe91d4b15641a58cb7787182e070_original.jpg\",\"full\":\"https://ksr-ug...</td>\n",
       "      <td>2358.0</td>\n",
       "      <td>{\"id\":3762384,\"project_id\":3762384,\"state\":\"active\",\"state_changed_at\":1569117788,\"name\":\"\"Lizar...</td>\n",
       "      <td>lizard-a-short-film</td>\n",
       "      <td>https://www.kickstarter.com/discover/categories/film%20&amp;%20video</td>\n",
       "      <td>True</td>\n",
       "      <td>False</td>\n",
       "      <td>successful</td>\n",
       "      <td>1569109188</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>{\"web\":{\"project\":\"https://www.kickstarter.com/projects/lizardshortfilm/lizard-a-short-film?ref=...</td>\n",
       "      <td>2358.000000</td>\n",
       "      <td>domestic</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>5</td>\n",
       "      <td>My friend and I are making an app that has never been made before. Making a football turn based ...</td>\n",
       "      <td>{\"id\":272,\"name\":\"Mobile Games\",\"slug\":\"games/mobile games\",\"position\":3,\"parent_id\":12,\"parent_...</td>\n",
       "      <td>172</td>\n",
       "      <td>US</td>\n",
       "      <td>the United States</td>\n",
       "      <td>1453500203</td>\n",
       "      <td>{\"id\":1843687675,\"name\":\"Mike Teorsky\",\"is_registered\":null,\"chosen_currency\":null,\"is_superback...</td>\n",
       "      <td>USD</td>\n",
       "      <td>$</td>\n",
       "      <td>True</td>\n",
       "      <td>USD</td>\n",
       "      <td>1458681077</td>\n",
       "      <td>False</td>\n",
       "      <td>NaN</td>\n",
       "      <td>1.00000</td>\n",
       "      <td>25000.0</td>\n",
       "      <td>37585826</td>\n",
       "      <td>NaN</td>\n",
       "      <td>False</td>\n",
       "      <td>NaN</td>\n",
       "      <td>1453500677</td>\n",
       "      <td>{\"id\":2473224,\"name\":\"Pittsburgh\",\"slug\":\"pittsburgh-pa\",\"short_name\":\"Pittsburgh, PA\",\"displaya...</td>\n",
       "      <td>Total Franchise Football 2016</td>\n",
       "      <td>NaN</td>\n",
       "      <td>{\"key\":\"assets/012/361/145/208a4baab227d49c5d23f46a8361cd9c_original.jpg\",\"full\":\"https://ksr-ug...</td>\n",
       "      <td>172.0</td>\n",
       "      <td>{\"id\":2335556,\"project_id\":2335556,\"state\":\"inactive\",\"state_changed_at\":1453500203,\"name\":null,...</td>\n",
       "      <td>total-franchise-football-2016</td>\n",
       "      <td>https://www.kickstarter.com/discover/categories/games/mobile%20games</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>failed</td>\n",
       "      <td>1458681110</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>{\"web\":{\"project\":\"https://www.kickstarter.com/projects/1843687675/total-franchise-football-2016...</td>\n",
       "      <td>172.000000</td>\n",
       "      <td>domestic</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>23</td>\n",
       "      <td>More than a bike trip, a state of mind. And a book.</td>\n",
       "      <td>{\"id\":48,\"name\":\"Nonfiction\",\"slug\":\"publishing/nonfiction\",\"position\":9,\"parent_id\":18,\"parent_...</td>\n",
       "      <td>501</td>\n",
       "      <td>US</td>\n",
       "      <td>the United States</td>\n",
       "      <td>1345516420</td>\n",
       "      <td>{\"id\":376746530,\"name\":\"John Farley\",\"is_registered\":null,\"chosen_currency\":null,\"is_superbacker...</td>\n",
       "      <td>USD</td>\n",
       "      <td>$</td>\n",
       "      <td>True</td>\n",
       "      <td>USD</td>\n",
       "      <td>1348540386</td>\n",
       "      <td>False</td>\n",
       "      <td>NaN</td>\n",
       "      <td>1.00000</td>\n",
       "      <td>500.0</td>\n",
       "      <td>1727301949</td>\n",
       "      <td>NaN</td>\n",
       "      <td>False</td>\n",
       "      <td>NaN</td>\n",
       "      <td>1347676386</td>\n",
       "      <td>{\"id\":12589335,\"name\":\"Brooklyn\",\"slug\":\"brooklyn-ny\",\"short_name\":\"Brooklyn, NY\",\"displayable_n...</td>\n",
       "      <td>The Trans-American Psychogeographic Literary Cycle Tour</td>\n",
       "      <td>NaN</td>\n",
       "      <td>{\"key\":\"assets/011/414/575/8e408be031b76a8b9f7d9915c6b0ef62_original.jpg\",\"full\":\"https://ksr-ug...</td>\n",
       "      <td>501.0</td>\n",
       "      <td>{\"id\":311225,\"project_id\":311225,\"state\":\"inactive\",\"state_changed_at\":1425915817,\"name\":null,\"b...</td>\n",
       "      <td>the-trans-american-psychogeographic-literary-cycle</td>\n",
       "      <td>https://www.kickstarter.com/discover/categories/publishing/nonfiction</td>\n",
       "      <td>True</td>\n",
       "      <td>False</td>\n",
       "      <td>successful</td>\n",
       "      <td>1348540386</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>{\"web\":{\"project\":\"https://www.kickstarter.com/projects/376746530/the-trans-american-psychogeogr...</td>\n",
       "      <td>501.000000</td>\n",
       "      <td>domestic</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>0</td>\n",
       "      <td>This is the birth of a revolutionary artist...Revery! A burst out into the world with GAME CHANG...</td>\n",
       "      <td>{\"id\":39,\"name\":\"Hip-Hop\",\"slug\":\"music/hip-hop\",\"position\":8,\"parent_id\":14,\"parent_name\":\"Musi...</td>\n",
       "      <td>0</td>\n",
       "      <td>US</td>\n",
       "      <td>the United States</td>\n",
       "      <td>1510267562</td>\n",
       "      <td>{\"id\":1163714172,\"name\":\"Revery (deleted)\",\"is_registered\":null,\"chosen_currency\":null,\"is_super...</td>\n",
       "      <td>USD</td>\n",
       "      <td>$</td>\n",
       "      <td>True</td>\n",
       "      <td>USD</td>\n",
       "      <td>1512960191</td>\n",
       "      <td>False</td>\n",
       "      <td>NaN</td>\n",
       "      <td>1.00000</td>\n",
       "      <td>2500.0</td>\n",
       "      <td>144196412</td>\n",
       "      <td>NaN</td>\n",
       "      <td>False</td>\n",
       "      <td>NaN</td>\n",
       "      <td>1510281791</td>\n",
       "      <td>{\"id\":2374418,\"name\":\"Canoga Park\",\"slug\":\"canoga-park-los-angeles-ca\",\"short_name\":\"Canoga Park...</td>\n",
       "      <td>The Golden Ticket to the Wonka Factory</td>\n",
       "      <td>NaN</td>\n",
       "      <td>{\"key\":\"assets/019/151/238/a73360e8eb4dc7140eb13f638ea8f11d_original.JPG\",\"full\":\"https://ksr-ug...</td>\n",
       "      <td>0.0</td>\n",
       "      <td>{\"id\":3214412,\"project_id\":3214412,\"state\":\"inactive\",\"state_changed_at\":1510267562,\"name\":null,...</td>\n",
       "      <td>the-golden-ticket-to-the-wonka-factory</td>\n",
       "      <td>https://www.kickstarter.com/discover/categories/music/hip-hop</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>failed</td>\n",
       "      <td>1512960191</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>{\"web\":{\"project\":\"https://www.kickstarter.com/projects/1163714172/the-golden-ticket-to-the-wonk...</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>domestic</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   backers_count  \\\n",
       "0              5   \n",
       "1             19   \n",
       "2              5   \n",
       "3             23   \n",
       "4              0   \n",
       "\n",
       "                                                                                                 blurb  \\\n",
       "0  A Free to Win Online Mobile Strategy Game to end the corruption of the rich players always winni...   \n",
       "1                   Lizard is a dark comedy and modern western, that takes place in the Mojave Desert.   \n",
       "2  My friend and I are making an app that has never been made before. Making a football turn based ...   \n",
       "3                                                  More than a bike trip, a state of mind. And a book.   \n",
       "4  This is the birth of a revolutionary artist...Revery! A burst out into the world with GAME CHANG...   \n",
       "\n",
       "                                                                                              category  \\\n",
       "0  {\"id\":272,\"name\":\"Mobile Games\",\"slug\":\"games/mobile games\",\"position\":3,\"parent_id\":12,\"parent_...   \n",
       "1  {\"id\":11,\"name\":\"Film & Video\",\"slug\":\"film & video\",\"position\":7,\"color\":16734574,\"urls\":{\"web\"...   \n",
       "2  {\"id\":272,\"name\":\"Mobile Games\",\"slug\":\"games/mobile games\",\"position\":3,\"parent_id\":12,\"parent_...   \n",
       "3  {\"id\":48,\"name\":\"Nonfiction\",\"slug\":\"publishing/nonfiction\",\"position\":9,\"parent_id\":18,\"parent_...   \n",
       "4  {\"id\":39,\"name\":\"Hip-Hop\",\"slug\":\"music/hip-hop\",\"position\":8,\"parent_id\":14,\"parent_name\":\"Musi...   \n",
       "\n",
       "   converted_pledged_amount country country_displayable_name  created_at  \\\n",
       "0                        83      CA                   Canada  1506125024   \n",
       "1                      2358      US        the United States  1561667598   \n",
       "2                       172      US        the United States  1453500203   \n",
       "3                       501      US        the United States  1345516420   \n",
       "4                         0      US        the United States  1510267562   \n",
       "\n",
       "                                                                                               creator  \\\n",
       "0  {\"id\":1178629187,\"name\":\"Adam Tompkins\",\"is_registered\":null,\"chosen_currency\":null,\"is_superbac...   \n",
       "1  {\"id\":1624125450,\"name\":\"Darrin Williams\",\"slug\":\"lizardshortfilm\",\"is_registered\":null,\"chosen_...   \n",
       "2  {\"id\":1843687675,\"name\":\"Mike Teorsky\",\"is_registered\":null,\"chosen_currency\":null,\"is_superback...   \n",
       "3  {\"id\":376746530,\"name\":\"John Farley\",\"is_registered\":null,\"chosen_currency\":null,\"is_superbacker...   \n",
       "4  {\"id\":1163714172,\"name\":\"Revery (deleted)\",\"is_registered\":null,\"chosen_currency\":null,\"is_super...   \n",
       "\n",
       "  currency currency_symbol  currency_trailing_code current_currency  \\\n",
       "0      CAD               $                    True              USD   \n",
       "1      USD               $                    True              USD   \n",
       "2      USD               $                    True              USD   \n",
       "3      USD               $                    True              USD   \n",
       "4      USD               $                    True              USD   \n",
       "\n",
       "     deadline  disable_communication friends  fx_rate     goal          id  \\\n",
       "0  1507560063                  False     NaN  0.70837   1000.0   751419376   \n",
       "1  1569109188                  False     NaN  1.00000   1100.0   289952460   \n",
       "2  1458681077                  False     NaN  1.00000  25000.0    37585826   \n",
       "3  1348540386                  False     NaN  1.00000    500.0  1727301949   \n",
       "4  1512960191                  False     NaN  1.00000   2500.0   144196412   \n",
       "\n",
       "  is_backing  is_starrable is_starred  launched_at  \\\n",
       "0        NaN         False        NaN   1506350463   \n",
       "1        NaN         False        NaN   1563925188   \n",
       "2        NaN         False        NaN   1453500677   \n",
       "3        NaN         False        NaN   1347676386   \n",
       "4        NaN         False        NaN   1510281791   \n",
       "\n",
       "                                                                                              location  \\\n",
       "0  {\"id\":22664159,\"name\":\"Mississauga\",\"slug\":\"mississauga-ca\",\"short_name\":\"Mississauga, Canada\",\"...   \n",
       "1  {\"id\":2452741,\"name\":\"Mojave\",\"slug\":\"mojave-ca\",\"short_name\":\"Mojave, CA\",\"displayable_name\":\"M...   \n",
       "2  {\"id\":2473224,\"name\":\"Pittsburgh\",\"slug\":\"pittsburgh-pa\",\"short_name\":\"Pittsburgh, PA\",\"displaya...   \n",
       "3  {\"id\":12589335,\"name\":\"Brooklyn\",\"slug\":\"brooklyn-ny\",\"short_name\":\"Brooklyn, NY\",\"displayable_n...   \n",
       "4  {\"id\":2374418,\"name\":\"Canoga Park\",\"slug\":\"canoga-park-los-angeles-ca\",\"short_name\":\"Canoga Park...   \n",
       "\n",
       "                                                         name permissions  \\\n",
       "0  Iron Age Kingdoms - First Free To Win Mobile Strategy Game         NaN   \n",
       "1                                       \"Lizard\" a short film         NaN   \n",
       "2                               Total Franchise Football 2016         NaN   \n",
       "3     The Trans-American Psychogeographic Literary Cycle Tour         NaN   \n",
       "4                      The Golden Ticket to the Wonka Factory         NaN   \n",
       "\n",
       "                                                                                                 photo  \\\n",
       "0  {\"key\":\"assets/018/444/378/81e59ccab10259506aab6ee340d25b73_original.png\",\"full\":\"https://ksr-ug...   \n",
       "1  {\"key\":\"assets/025/890/832/ff7ffe91d4b15641a58cb7787182e070_original.jpg\",\"full\":\"https://ksr-ug...   \n",
       "2  {\"key\":\"assets/012/361/145/208a4baab227d49c5d23f46a8361cd9c_original.jpg\",\"full\":\"https://ksr-ug...   \n",
       "3  {\"key\":\"assets/011/414/575/8e408be031b76a8b9f7d9915c6b0ef62_original.jpg\",\"full\":\"https://ksr-ug...   \n",
       "4  {\"key\":\"assets/019/151/238/a73360e8eb4dc7140eb13f638ea8f11d_original.JPG\",\"full\":\"https://ksr-ug...   \n",
       "\n",
       "   pledged  \\\n",
       "0    105.0   \n",
       "1   2358.0   \n",
       "2    172.0   \n",
       "3    501.0   \n",
       "4      0.0   \n",
       "\n",
       "                                                                                               profile  \\\n",
       "0  {\"id\":3151330,\"project_id\":3151330,\"state\":\"inactive\",\"state_changed_at\":1506125024,\"name\":null,...   \n",
       "1  {\"id\":3762384,\"project_id\":3762384,\"state\":\"active\",\"state_changed_at\":1569117788,\"name\":\"\"Lizar...   \n",
       "2  {\"id\":2335556,\"project_id\":2335556,\"state\":\"inactive\",\"state_changed_at\":1453500203,\"name\":null,...   \n",
       "3  {\"id\":311225,\"project_id\":311225,\"state\":\"inactive\",\"state_changed_at\":1425915817,\"name\":null,\"b...   \n",
       "4  {\"id\":3214412,\"project_id\":3214412,\"state\":\"inactive\",\"state_changed_at\":1510267562,\"name\":null,...   \n",
       "\n",
       "                                                 slug  \\\n",
       "0  iron-age-kingdoms-first-free-to-win-mobile-strateg   \n",
       "1                                 lizard-a-short-film   \n",
       "2                       total-franchise-football-2016   \n",
       "3  the-trans-american-psychogeographic-literary-cycle   \n",
       "4              the-golden-ticket-to-the-wonka-factory   \n",
       "\n",
       "                                                              source_url  \\\n",
       "0   https://www.kickstarter.com/discover/categories/games/mobile%20games   \n",
       "1       https://www.kickstarter.com/discover/categories/film%20&%20video   \n",
       "2   https://www.kickstarter.com/discover/categories/games/mobile%20games   \n",
       "3  https://www.kickstarter.com/discover/categories/publishing/nonfiction   \n",
       "4          https://www.kickstarter.com/discover/categories/music/hip-hop   \n",
       "\n",
       "   spotlight  staff_pick       state  state_changed_at  static_usd_rate  \\\n",
       "0      False       False      failed        1507560063         0.810764   \n",
       "1       True       False  successful        1569109188         1.000000   \n",
       "2      False       False      failed        1458681110         1.000000   \n",
       "3       True       False  successful        1348540386         1.000000   \n",
       "4      False       False      failed        1512960191         1.000000   \n",
       "\n",
       "                                                                                                  urls  \\\n",
       "0  {\"web\":{\"project\":\"https://www.kickstarter.com/projects/1178629187/iron-age-kingdoms-first-free-...   \n",
       "1  {\"web\":{\"project\":\"https://www.kickstarter.com/projects/lizardshortfilm/lizard-a-short-film?ref=...   \n",
       "2  {\"web\":{\"project\":\"https://www.kickstarter.com/projects/1843687675/total-franchise-football-2016...   \n",
       "3  {\"web\":{\"project\":\"https://www.kickstarter.com/projects/376746530/the-trans-american-psychogeogr...   \n",
       "4  {\"web\":{\"project\":\"https://www.kickstarter.com/projects/1163714172/the-golden-ticket-to-the-wonk...   \n",
       "\n",
       "   usd_pledged  usd_type  \n",
       "0    85.130189  domestic  \n",
       "1  2358.000000  domestic  \n",
       "2   172.000000  domestic  \n",
       "3   501.000000  domestic  \n",
       "4     0.000000  domestic  "
      ]
     },
     "execution_count": 6,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "<class 'pandas.core.frame.DataFrame'>\n",
      "RangeIndex: 409250 entries, 0 to 409249\n",
      "Data columns (total 38 columns):\n",
      " #   Column                    Non-Null Count   Dtype  \n",
      "---  ------                    --------------   -----  \n",
      " 0   backers_count             409250 non-null  int64  \n",
      " 1   blurb                     409236 non-null  object \n",
      " 2   category                  409250 non-null  object \n",
      " 3   converted_pledged_amount  409250 non-null  int64  \n",
      " 4   country                   409250 non-null  object \n",
      " 5   country_displayable_name  409250 non-null  object \n",
      " 6   created_at                409250 non-null  int64  \n",
      " 7   creator                   409250 non-null  object \n",
      " 8   currency                  409250 non-null  object \n",
      " 9   currency_symbol           409250 non-null  object \n",
      " 10  currency_trailing_code    409250 non-null  bool   \n",
      " 11  current_currency          409250 non-null  object \n",
      " 12  deadline                  409250 non-null  int64  \n",
      " 13  disable_communication     409250 non-null  bool   \n",
      " 14  friends                   312 non-null     object \n",
      " 15  fx_rate                   409250 non-null  float64\n",
      " 16  goal                      409250 non-null  float64\n",
      " 17  id                        409250 non-null  int64  \n",
      " 18  is_backing                312 non-null     object \n",
      " 19  is_starrable              409250 non-null  bool   \n",
      " 20  is_starred                312 non-null     object \n",
      " 21  launched_at               409250 non-null  int64  \n",
      " 22  location                  408872 non-null  object \n",
      " 23  name                      409250 non-null  object \n",
      " 24  permissions               312 non-null     object \n",
      " 25  photo                     409250 non-null  object \n",
      " 26  pledged                   409250 non-null  float64\n",
      " 27  profile                   409250 non-null  object \n",
      " 28  slug                      409250 non-null  object \n",
      " 29  source_url                409250 non-null  object \n",
      " 30  spotlight                 409250 non-null  bool   \n",
      " 31  staff_pick                409250 non-null  bool   \n",
      " 32  state                     409250 non-null  object \n",
      " 33  state_changed_at          409250 non-null  int64  \n",
      " 34  static_usd_rate           409250 non-null  float64\n",
      " 35  urls                      409250 non-null  object \n",
      " 36  usd_pledged               409250 non-null  float64\n",
      " 37  usd_type                  408890 non-null  object \n",
      "dtypes: bool(5), float64(5), int64(7), object(21)\n",
      "memory usage: 105.0+ MB\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "(409250, 38)"
      ]
     },
     "execution_count": 7,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.info()\n",
    "df.shape"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Data Cleaning"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [],
   "source": [
    "cols_to_convert = ['created_at', 'state_changed_at', 'deadline', 'launched_at']\n",
    "for c in cols_to_convert:\n",
    "    df[c] = pd.to_datetime(df[c], origin='unix', unit='s')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[nan '[]']\n",
      "[nan False]\n",
      "[nan False True]\n",
      "[nan '[]']\n",
      "['$' '€' '¥' '£' 'kr' 'Fr ']\n",
      "[ True False]\n",
      "['USD' 'EUR' 'CAD']\n",
      "['CAD' 'USD' 'EUR' 'JPY' 'AUD' 'GBP' 'SGD' 'DKK' 'NOK' 'MXN' 'HKD' 'SEK'\n",
      " 'NZD' 'CHF']\n",
      "['domestic' nan]\n"
     ]
    }
   ],
   "source": [
    "print(df['friends'].unique())\n",
    "print(df['is_backing'].unique())\n",
    "print(df['is_starred'].unique())\n",
    "print(df['permissions'].unique())\n",
    "print(df['currency_symbol'].unique())\n",
    "print(df['currency_trailing_code'].unique())\n",
    "print(df['current_currency'].unique())\n",
    "print(df['currency'].unique())\n",
    "print(df['usd_type'].unique())\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "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>backers_count</th>\n",
       "      <th>blurb</th>\n",
       "      <th>category</th>\n",
       "      <th>converted_pledged_amount</th>\n",
       "      <th>country</th>\n",
       "      <th>country_displayable_name</th>\n",
       "      <th>created_at</th>\n",
       "      <th>creator</th>\n",
       "      <th>currency</th>\n",
       "      <th>currency_symbol</th>\n",
       "      <th>currency_trailing_code</th>\n",
       "      <th>current_currency</th>\n",
       "      <th>deadline</th>\n",
       "      <th>disable_communication</th>\n",
       "      <th>friends</th>\n",
       "      <th>fx_rate</th>\n",
       "      <th>goal</th>\n",
       "      <th>id</th>\n",
       "      <th>is_backing</th>\n",
       "      <th>is_starrable</th>\n",
       "      <th>is_starred</th>\n",
       "      <th>launched_at</th>\n",
       "      <th>location</th>\n",
       "      <th>name</th>\n",
       "      <th>permissions</th>\n",
       "      <th>photo</th>\n",
       "      <th>pledged</th>\n",
       "      <th>profile</th>\n",
       "      <th>slug</th>\n",
       "      <th>source_url</th>\n",
       "      <th>spotlight</th>\n",
       "      <th>staff_pick</th>\n",
       "      <th>state</th>\n",
       "      <th>state_changed_at</th>\n",
       "      <th>static_usd_rate</th>\n",
       "      <th>urls</th>\n",
       "      <th>usd_pledged</th>\n",
       "      <th>usd_type</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>27934</th>\n",
       "      <td>128</td>\n",
       "      <td>Help us take our hand-dyed farm yarns on the road in a custom trailer complete with a dye statio...</td>\n",
       "      <td>{\"id\":348,\"name\":\"Knitting\",\"slug\":\"crafts/knitting\",\"position\":6,\"parent_id\":26,\"parent_name\":\"...</td>\n",
       "      <td>8995</td>\n",
       "      <td>US</td>\n",
       "      <td>the United States</td>\n",
       "      <td>2016-06-02 23:45:41</td>\n",
       "      <td>{\"id\":2004942297,\"name\":\"Kelly L. Corbett\",\"slug\":\"romneyridge\",\"is_registered\":null,\"chosen_cur...</td>\n",
       "      <td>USD</td>\n",
       "      <td>$</td>\n",
       "      <td>True</td>\n",
       "      <td>USD</td>\n",
       "      <td>2016-07-13 16:32:43</td>\n",
       "      <td>False</td>\n",
       "      <td>NaN</td>\n",
       "      <td>1.0</td>\n",
       "      <td>8000.0</td>\n",
       "      <td>1875733470</td>\n",
       "      <td>NaN</td>\n",
       "      <td>False</td>\n",
       "      <td>NaN</td>\n",
       "      <td>2016-06-13 16:32:43</td>\n",
       "      <td>{\"id\":2523917,\"name\":\"Woolwich\",\"slug\":\"woolwich-me\",\"short_name\":\"Woolwich, ME\",\"displayable_na...</td>\n",
       "      <td>Romney Ridge Yarns &amp; Wool Pop-Up Shop!</td>\n",
       "      <td>NaN</td>\n",
       "      <td>{\"key\":\"assets/012/618/886/97cc3d7948615efb7b5c5941c00d0aff_original.JPG\",\"full\":\"https://ksr-ug...</td>\n",
       "      <td>8995.0</td>\n",
       "      <td>{\"id\":2543599,\"project_id\":2543599,\"state\":\"active\",\"state_changed_at\":1468592118,\"name\":\"Romney...</td>\n",
       "      <td>romney-ridge-yarns-and-wool-pop-up-shop</td>\n",
       "      <td>https://www.kickstarter.com/discover/categories/crafts/knitting</td>\n",
       "      <td>True</td>\n",
       "      <td>False</td>\n",
       "      <td>successful</td>\n",
       "      <td>2016-07-13 16:32:43</td>\n",
       "      <td>1.0</td>\n",
       "      <td>{\"web\":{\"project\":\"https://www.kickstarter.com/projects/romneyridge/romney-ridge-yarns-and-wool-...</td>\n",
       "      <td>8995.0</td>\n",
       "      <td>domestic</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>162077</th>\n",
       "      <td>128</td>\n",
       "      <td>Help us take our hand-dyed farm yarns on the road in a custom trailer complete with a dye statio...</td>\n",
       "      <td>{\"id\":348,\"name\":\"Knitting\",\"slug\":\"crafts/knitting\",\"position\":6,\"parent_id\":26,\"parent_name\":\"...</td>\n",
       "      <td>8995</td>\n",
       "      <td>US</td>\n",
       "      <td>the United States</td>\n",
       "      <td>2016-06-02 23:45:41</td>\n",
       "      <td>{\"id\":2004942297,\"name\":\"Kelly L. Corbett\",\"slug\":\"romneyridge\",\"is_registered\":null,\"chosen_cur...</td>\n",
       "      <td>USD</td>\n",
       "      <td>$</td>\n",
       "      <td>True</td>\n",
       "      <td>USD</td>\n",
       "      <td>2016-07-13 16:32:43</td>\n",
       "      <td>False</td>\n",
       "      <td>NaN</td>\n",
       "      <td>1.0</td>\n",
       "      <td>8000.0</td>\n",
       "      <td>1875733470</td>\n",
       "      <td>NaN</td>\n",
       "      <td>False</td>\n",
       "      <td>NaN</td>\n",
       "      <td>2016-06-13 16:32:43</td>\n",
       "      <td>{\"id\":2523917,\"name\":\"Woolwich\",\"slug\":\"woolwich-me\",\"short_name\":\"Woolwich, ME\",\"displayable_na...</td>\n",
       "      <td>Romney Ridge Yarns &amp; Wool Pop-Up Shop!</td>\n",
       "      <td>NaN</td>\n",
       "      <td>{\"key\":\"assets/012/618/886/97cc3d7948615efb7b5c5941c00d0aff_original.JPG\",\"full\":\"https://ksr-ug...</td>\n",
       "      <td>8995.0</td>\n",
       "      <td>{\"id\":2543599,\"project_id\":2543599,\"state\":\"active\",\"state_changed_at\":1468592118,\"name\":\"Romney...</td>\n",
       "      <td>romney-ridge-yarns-and-wool-pop-up-shop</td>\n",
       "      <td>https://www.kickstarter.com/discover/categories/crafts/knitting</td>\n",
       "      <td>True</td>\n",
       "      <td>False</td>\n",
       "      <td>successful</td>\n",
       "      <td>2016-07-13 16:32:43</td>\n",
       "      <td>1.0</td>\n",
       "      <td>{\"web\":{\"project\":\"https://www.kickstarter.com/projects/romneyridge/romney-ridge-yarns-and-wool-...</td>\n",
       "      <td>8995.0</td>\n",
       "      <td>domestic</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>324259</th>\n",
       "      <td>128</td>\n",
       "      <td>Help us take our hand-dyed farm yarns on the road in a custom trailer complete with a dye statio...</td>\n",
       "      <td>{\"id\":348,\"name\":\"Knitting\",\"slug\":\"crafts/knitting\",\"position\":6,\"parent_id\":26,\"parent_name\":\"...</td>\n",
       "      <td>8995</td>\n",
       "      <td>US</td>\n",
       "      <td>the United States</td>\n",
       "      <td>2016-06-02 23:45:41</td>\n",
       "      <td>{\"id\":2004942297,\"name\":\"Kelly L. Corbett\",\"slug\":\"romneyridge\",\"is_registered\":null,\"chosen_cur...</td>\n",
       "      <td>USD</td>\n",
       "      <td>$</td>\n",
       "      <td>True</td>\n",
       "      <td>USD</td>\n",
       "      <td>2016-07-13 16:32:43</td>\n",
       "      <td>False</td>\n",
       "      <td>NaN</td>\n",
       "      <td>1.0</td>\n",
       "      <td>8000.0</td>\n",
       "      <td>1875733470</td>\n",
       "      <td>NaN</td>\n",
       "      <td>False</td>\n",
       "      <td>NaN</td>\n",
       "      <td>2016-06-13 16:32:43</td>\n",
       "      <td>{\"id\":2523917,\"name\":\"Woolwich\",\"slug\":\"woolwich-me\",\"short_name\":\"Woolwich, ME\",\"displayable_na...</td>\n",
       "      <td>Romney Ridge Yarns &amp; Wool Pop-Up Shop!</td>\n",
       "      <td>NaN</td>\n",
       "      <td>{\"key\":\"assets/012/618/886/97cc3d7948615efb7b5c5941c00d0aff_original.JPG\",\"full\":\"https://ksr-ug...</td>\n",
       "      <td>8995.0</td>\n",
       "      <td>{\"id\":2543599,\"project_id\":2543599,\"state\":\"active\",\"state_changed_at\":1468592118,\"name\":\"Romney...</td>\n",
       "      <td>romney-ridge-yarns-and-wool-pop-up-shop</td>\n",
       "      <td>https://www.kickstarter.com/discover/categories/crafts</td>\n",
       "      <td>True</td>\n",
       "      <td>False</td>\n",
       "      <td>successful</td>\n",
       "      <td>2016-07-13 16:32:43</td>\n",
       "      <td>1.0</td>\n",
       "      <td>{\"web\":{\"project\":\"https://www.kickstarter.com/projects/romneyridge/romney-ridge-yarns-and-wool-...</td>\n",
       "      <td>8995.0</td>\n",
       "      <td>domestic</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>394741</th>\n",
       "      <td>128</td>\n",
       "      <td>Help us take our hand-dyed farm yarns on the road in a custom trailer complete with a dye statio...</td>\n",
       "      <td>{\"id\":348,\"name\":\"Knitting\",\"slug\":\"crafts/knitting\",\"position\":6,\"parent_id\":26,\"parent_name\":\"...</td>\n",
       "      <td>8995</td>\n",
       "      <td>US</td>\n",
       "      <td>the United States</td>\n",
       "      <td>2016-06-02 23:45:41</td>\n",
       "      <td>{\"id\":2004942297,\"name\":\"Kelly L. Corbett\",\"slug\":\"romneyridge\",\"is_registered\":null,\"chosen_cur...</td>\n",
       "      <td>USD</td>\n",
       "      <td>$</td>\n",
       "      <td>True</td>\n",
       "      <td>USD</td>\n",
       "      <td>2016-07-13 16:32:43</td>\n",
       "      <td>False</td>\n",
       "      <td>NaN</td>\n",
       "      <td>1.0</td>\n",
       "      <td>8000.0</td>\n",
       "      <td>1875733470</td>\n",
       "      <td>NaN</td>\n",
       "      <td>False</td>\n",
       "      <td>NaN</td>\n",
       "      <td>2016-06-13 16:32:43</td>\n",
       "      <td>{\"id\":2523917,\"name\":\"Woolwich\",\"slug\":\"woolwich-me\",\"short_name\":\"Woolwich, ME\",\"displayable_na...</td>\n",
       "      <td>Romney Ridge Yarns &amp; Wool Pop-Up Shop!</td>\n",
       "      <td>NaN</td>\n",
       "      <td>{\"key\":\"assets/012/618/886/97cc3d7948615efb7b5c5941c00d0aff_original.JPG\",\"full\":\"https://ksr-ug...</td>\n",
       "      <td>8995.0</td>\n",
       "      <td>{\"id\":2543599,\"project_id\":2543599,\"state\":\"active\",\"state_changed_at\":1468592118,\"name\":\"Romney...</td>\n",
       "      <td>romney-ridge-yarns-and-wool-pop-up-shop</td>\n",
       "      <td>https://www.kickstarter.com/discover/categories/crafts</td>\n",
       "      <td>True</td>\n",
       "      <td>False</td>\n",
       "      <td>successful</td>\n",
       "      <td>2016-07-13 16:32:43</td>\n",
       "      <td>1.0</td>\n",
       "      <td>{\"web\":{\"project\":\"https://www.kickstarter.com/projects/romneyridge/romney-ridge-yarns-and-wool-...</td>\n",
       "      <td>8995.0</td>\n",
       "      <td>domestic</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "        backers_count  \\\n",
       "27934             128   \n",
       "162077            128   \n",
       "324259            128   \n",
       "394741            128   \n",
       "\n",
       "                                                                                                      blurb  \\\n",
       "27934   Help us take our hand-dyed farm yarns on the road in a custom trailer complete with a dye statio...   \n",
       "162077  Help us take our hand-dyed farm yarns on the road in a custom trailer complete with a dye statio...   \n",
       "324259  Help us take our hand-dyed farm yarns on the road in a custom trailer complete with a dye statio...   \n",
       "394741  Help us take our hand-dyed farm yarns on the road in a custom trailer complete with a dye statio...   \n",
       "\n",
       "                                                                                                   category  \\\n",
       "27934   {\"id\":348,\"name\":\"Knitting\",\"slug\":\"crafts/knitting\",\"position\":6,\"parent_id\":26,\"parent_name\":\"...   \n",
       "162077  {\"id\":348,\"name\":\"Knitting\",\"slug\":\"crafts/knitting\",\"position\":6,\"parent_id\":26,\"parent_name\":\"...   \n",
       "324259  {\"id\":348,\"name\":\"Knitting\",\"slug\":\"crafts/knitting\",\"position\":6,\"parent_id\":26,\"parent_name\":\"...   \n",
       "394741  {\"id\":348,\"name\":\"Knitting\",\"slug\":\"crafts/knitting\",\"position\":6,\"parent_id\":26,\"parent_name\":\"...   \n",
       "\n",
       "        converted_pledged_amount country country_displayable_name  \\\n",
       "27934                       8995      US        the United States   \n",
       "162077                      8995      US        the United States   \n",
       "324259                      8995      US        the United States   \n",
       "394741                      8995      US        the United States   \n",
       "\n",
       "                created_at  \\\n",
       "27934  2016-06-02 23:45:41   \n",
       "162077 2016-06-02 23:45:41   \n",
       "324259 2016-06-02 23:45:41   \n",
       "394741 2016-06-02 23:45:41   \n",
       "\n",
       "                                                                                                    creator  \\\n",
       "27934   {\"id\":2004942297,\"name\":\"Kelly L. Corbett\",\"slug\":\"romneyridge\",\"is_registered\":null,\"chosen_cur...   \n",
       "162077  {\"id\":2004942297,\"name\":\"Kelly L. Corbett\",\"slug\":\"romneyridge\",\"is_registered\":null,\"chosen_cur...   \n",
       "324259  {\"id\":2004942297,\"name\":\"Kelly L. Corbett\",\"slug\":\"romneyridge\",\"is_registered\":null,\"chosen_cur...   \n",
       "394741  {\"id\":2004942297,\"name\":\"Kelly L. Corbett\",\"slug\":\"romneyridge\",\"is_registered\":null,\"chosen_cur...   \n",
       "\n",
       "       currency currency_symbol  currency_trailing_code current_currency  \\\n",
       "27934       USD               $                    True              USD   \n",
       "162077      USD               $                    True              USD   \n",
       "324259      USD               $                    True              USD   \n",
       "394741      USD               $                    True              USD   \n",
       "\n",
       "                  deadline  disable_communication friends  fx_rate    goal  \\\n",
       "27934  2016-07-13 16:32:43                  False     NaN      1.0  8000.0   \n",
       "162077 2016-07-13 16:32:43                  False     NaN      1.0  8000.0   \n",
       "324259 2016-07-13 16:32:43                  False     NaN      1.0  8000.0   \n",
       "394741 2016-07-13 16:32:43                  False     NaN      1.0  8000.0   \n",
       "\n",
       "                id is_backing  is_starrable is_starred         launched_at  \\\n",
       "27934   1875733470        NaN         False        NaN 2016-06-13 16:32:43   \n",
       "162077  1875733470        NaN         False        NaN 2016-06-13 16:32:43   \n",
       "324259  1875733470        NaN         False        NaN 2016-06-13 16:32:43   \n",
       "394741  1875733470        NaN         False        NaN 2016-06-13 16:32:43   \n",
       "\n",
       "                                                                                                   location  \\\n",
       "27934   {\"id\":2523917,\"name\":\"Woolwich\",\"slug\":\"woolwich-me\",\"short_name\":\"Woolwich, ME\",\"displayable_na...   \n",
       "162077  {\"id\":2523917,\"name\":\"Woolwich\",\"slug\":\"woolwich-me\",\"short_name\":\"Woolwich, ME\",\"displayable_na...   \n",
       "324259  {\"id\":2523917,\"name\":\"Woolwich\",\"slug\":\"woolwich-me\",\"short_name\":\"Woolwich, ME\",\"displayable_na...   \n",
       "394741  {\"id\":2523917,\"name\":\"Woolwich\",\"slug\":\"woolwich-me\",\"short_name\":\"Woolwich, ME\",\"displayable_na...   \n",
       "\n",
       "                                          name permissions  \\\n",
       "27934   Romney Ridge Yarns & Wool Pop-Up Shop!         NaN   \n",
       "162077  Romney Ridge Yarns & Wool Pop-Up Shop!         NaN   \n",
       "324259  Romney Ridge Yarns & Wool Pop-Up Shop!         NaN   \n",
       "394741  Romney Ridge Yarns & Wool Pop-Up Shop!         NaN   \n",
       "\n",
       "                                                                                                      photo  \\\n",
       "27934   {\"key\":\"assets/012/618/886/97cc3d7948615efb7b5c5941c00d0aff_original.JPG\",\"full\":\"https://ksr-ug...   \n",
       "162077  {\"key\":\"assets/012/618/886/97cc3d7948615efb7b5c5941c00d0aff_original.JPG\",\"full\":\"https://ksr-ug...   \n",
       "324259  {\"key\":\"assets/012/618/886/97cc3d7948615efb7b5c5941c00d0aff_original.JPG\",\"full\":\"https://ksr-ug...   \n",
       "394741  {\"key\":\"assets/012/618/886/97cc3d7948615efb7b5c5941c00d0aff_original.JPG\",\"full\":\"https://ksr-ug...   \n",
       "\n",
       "        pledged  \\\n",
       "27934    8995.0   \n",
       "162077   8995.0   \n",
       "324259   8995.0   \n",
       "394741   8995.0   \n",
       "\n",
       "                                                                                                    profile  \\\n",
       "27934   {\"id\":2543599,\"project_id\":2543599,\"state\":\"active\",\"state_changed_at\":1468592118,\"name\":\"Romney...   \n",
       "162077  {\"id\":2543599,\"project_id\":2543599,\"state\":\"active\",\"state_changed_at\":1468592118,\"name\":\"Romney...   \n",
       "324259  {\"id\":2543599,\"project_id\":2543599,\"state\":\"active\",\"state_changed_at\":1468592118,\"name\":\"Romney...   \n",
       "394741  {\"id\":2543599,\"project_id\":2543599,\"state\":\"active\",\"state_changed_at\":1468592118,\"name\":\"Romney...   \n",
       "\n",
       "                                           slug  \\\n",
       "27934   romney-ridge-yarns-and-wool-pop-up-shop   \n",
       "162077  romney-ridge-yarns-and-wool-pop-up-shop   \n",
       "324259  romney-ridge-yarns-and-wool-pop-up-shop   \n",
       "394741  romney-ridge-yarns-and-wool-pop-up-shop   \n",
       "\n",
       "                                                             source_url  \\\n",
       "27934   https://www.kickstarter.com/discover/categories/crafts/knitting   \n",
       "162077  https://www.kickstarter.com/discover/categories/crafts/knitting   \n",
       "324259           https://www.kickstarter.com/discover/categories/crafts   \n",
       "394741           https://www.kickstarter.com/discover/categories/crafts   \n",
       "\n",
       "        spotlight  staff_pick       state    state_changed_at  \\\n",
       "27934        True       False  successful 2016-07-13 16:32:43   \n",
       "162077       True       False  successful 2016-07-13 16:32:43   \n",
       "324259       True       False  successful 2016-07-13 16:32:43   \n",
       "394741       True       False  successful 2016-07-13 16:32:43   \n",
       "\n",
       "        static_usd_rate  \\\n",
       "27934               1.0   \n",
       "162077              1.0   \n",
       "324259              1.0   \n",
       "394741              1.0   \n",
       "\n",
       "                                                                                                       urls  \\\n",
       "27934   {\"web\":{\"project\":\"https://www.kickstarter.com/projects/romneyridge/romney-ridge-yarns-and-wool-...   \n",
       "162077  {\"web\":{\"project\":\"https://www.kickstarter.com/projects/romneyridge/romney-ridge-yarns-and-wool-...   \n",
       "324259  {\"web\":{\"project\":\"https://www.kickstarter.com/projects/romneyridge/romney-ridge-yarns-and-wool-...   \n",
       "394741  {\"web\":{\"project\":\"https://www.kickstarter.com/projects/romneyridge/romney-ridge-yarns-and-wool-...   \n",
       "\n",
       "        usd_pledged  usd_type  \n",
       "27934        8995.0  domestic  \n",
       "162077       8995.0  domestic  \n",
       "324259       8995.0  domestic  \n",
       "394741       8995.0  domestic  "
      ]
     },
     "execution_count": 10,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df[df['id'] == 1875733470]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "2048996672    4\n",
       "373804148     4\n",
       "1460555984    4\n",
       "949167493     4\n",
       "865816103     4\n",
       "             ..\n",
       "60795511      2\n",
       "42880109      2\n",
       "1890468972    2\n",
       "405175403     2\n",
       "1329596417    2\n",
       "Name: id, Length: 180753, dtype: int64"
      ]
     },
     "execution_count": 11,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df['id'].value_counts()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(180753, 38)"
      ]
     },
     "execution_count": 12,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#remove all duplicate rows\n",
    "df.drop_duplicates(subset='id', keep='first', inplace=True)\n",
    "df.shape"
   ]
  },
  {
   "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>backers_count</th>\n",
       "      <th>blurb</th>\n",
       "      <th>category</th>\n",
       "      <th>converted_pledged_amount</th>\n",
       "      <th>country</th>\n",
       "      <th>country_displayable_name</th>\n",
       "      <th>created_at</th>\n",
       "      <th>creator</th>\n",
       "      <th>currency</th>\n",
       "      <th>currency_symbol</th>\n",
       "      <th>currency_trailing_code</th>\n",
       "      <th>current_currency</th>\n",
       "      <th>deadline</th>\n",
       "      <th>disable_communication</th>\n",
       "      <th>friends</th>\n",
       "      <th>fx_rate</th>\n",
       "      <th>goal</th>\n",
       "      <th>id</th>\n",
       "      <th>is_backing</th>\n",
       "      <th>is_starrable</th>\n",
       "      <th>is_starred</th>\n",
       "      <th>launched_at</th>\n",
       "      <th>location</th>\n",
       "      <th>name</th>\n",
       "      <th>permissions</th>\n",
       "      <th>photo</th>\n",
       "      <th>pledged</th>\n",
       "      <th>profile</th>\n",
       "      <th>slug</th>\n",
       "      <th>source_url</th>\n",
       "      <th>spotlight</th>\n",
       "      <th>staff_pick</th>\n",
       "      <th>state</th>\n",
       "      <th>state_changed_at</th>\n",
       "      <th>static_usd_rate</th>\n",
       "      <th>urls</th>\n",
       "      <th>usd_pledged</th>\n",
       "      <th>usd_type</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>122464</th>\n",
       "      <td>100</td>\n",
       "      <td>New folk music &amp; song show by Derby trio Moirai. A 100 year old tale of fake news, family values...</td>\n",
       "      <td>{\"id\":37,\"name\":\"Country &amp; Folk\",\"slug\":\"music/country &amp; folk\",\"position\":5,\"parent_id\":14,\"pare...</td>\n",
       "      <td>3852</td>\n",
       "      <td>GB</td>\n",
       "      <td>the United Kingdom</td>\n",
       "      <td>2019-08-28 09:55:39</td>\n",
       "      <td>{\"id\":612927420,\"name\":\"Moirai\",\"slug\":\"alicewheeldonstory\",\"is_registered\":null,\"chosen_currenc...</td>\n",
       "      <td>GBP</td>\n",
       "      <td>£</td>\n",
       "      <td>False</td>\n",
       "      <td>USD</td>\n",
       "      <td>2019-10-16 17:00:11</td>\n",
       "      <td>False</td>\n",
       "      <td>NaN</td>\n",
       "      <td>1.24678</td>\n",
       "      <td>2500.0</td>\n",
       "      <td>1103759530</td>\n",
       "      <td>NaN</td>\n",
       "      <td>False</td>\n",
       "      <td>NaN</td>\n",
       "      <td>2019-09-16 17:00:11</td>\n",
       "      <td>{\"id\":18114,\"name\":\"Derby\",\"slug\":\"derby-gb\",\"short_name\":\"Derby, UK\",\"displayable_name\":\"Derby,...</td>\n",
       "      <td>Framed: The Alice Wheeldon Story - help us spread the word!</td>\n",
       "      <td>NaN</td>\n",
       "      <td>{\"key\":\"assets/026/358/508/1b804b34c2c5dbe3174260f0c019bbb1_original.png\",\"full\":\"https://ksr-ug...</td>\n",
       "      <td>3021.0</td>\n",
       "      <td>{\"id\":3802244,\"project_id\":3802244,\"state\":\"inactive\",\"state_changed_at\":1566986139,\"name\":null,...</td>\n",
       "      <td>framed-the-alice-wheeldon-story-help-us-spread-the-word</td>\n",
       "      <td>https://www.kickstarter.com/discover/categories/music/country%20&amp;%20folk</td>\n",
       "      <td>True</td>\n",
       "      <td>False</td>\n",
       "      <td>successful</td>\n",
       "      <td>2019-10-16 17:00:11</td>\n",
       "      <td>1.250602</td>\n",
       "      <td>{\"web\":{\"project\":\"https://www.kickstarter.com/projects/alicewheeldonstory/framed-the-alice-whee...</td>\n",
       "      <td>3778.068189</td>\n",
       "      <td>domestic</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "        backers_count  \\\n",
       "122464            100   \n",
       "\n",
       "                                                                                                      blurb  \\\n",
       "122464  New folk music & song show by Derby trio Moirai. A 100 year old tale of fake news, family values...   \n",
       "\n",
       "                                                                                                   category  \\\n",
       "122464  {\"id\":37,\"name\":\"Country & Folk\",\"slug\":\"music/country & folk\",\"position\":5,\"parent_id\":14,\"pare...   \n",
       "\n",
       "        converted_pledged_amount country country_displayable_name  \\\n",
       "122464                      3852      GB       the United Kingdom   \n",
       "\n",
       "                created_at  \\\n",
       "122464 2019-08-28 09:55:39   \n",
       "\n",
       "                                                                                                    creator  \\\n",
       "122464  {\"id\":612927420,\"name\":\"Moirai\",\"slug\":\"alicewheeldonstory\",\"is_registered\":null,\"chosen_currenc...   \n",
       "\n",
       "       currency currency_symbol  currency_trailing_code current_currency  \\\n",
       "122464      GBP               £                   False              USD   \n",
       "\n",
       "                  deadline  disable_communication friends  fx_rate    goal  \\\n",
       "122464 2019-10-16 17:00:11                  False     NaN  1.24678  2500.0   \n",
       "\n",
       "                id is_backing  is_starrable is_starred         launched_at  \\\n",
       "122464  1103759530        NaN         False        NaN 2019-09-16 17:00:11   \n",
       "\n",
       "                                                                                                   location  \\\n",
       "122464  {\"id\":18114,\"name\":\"Derby\",\"slug\":\"derby-gb\",\"short_name\":\"Derby, UK\",\"displayable_name\":\"Derby,...   \n",
       "\n",
       "                                                               name  \\\n",
       "122464  Framed: The Alice Wheeldon Story - help us spread the word!   \n",
       "\n",
       "       permissions  \\\n",
       "122464         NaN   \n",
       "\n",
       "                                                                                                      photo  \\\n",
       "122464  {\"key\":\"assets/026/358/508/1b804b34c2c5dbe3174260f0c019bbb1_original.png\",\"full\":\"https://ksr-ug...   \n",
       "\n",
       "        pledged  \\\n",
       "122464   3021.0   \n",
       "\n",
       "                                                                                                    profile  \\\n",
       "122464  {\"id\":3802244,\"project_id\":3802244,\"state\":\"inactive\",\"state_changed_at\":1566986139,\"name\":null,...   \n",
       "\n",
       "                                                           slug  \\\n",
       "122464  framed-the-alice-wheeldon-story-help-us-spread-the-word   \n",
       "\n",
       "                                                                      source_url  \\\n",
       "122464  https://www.kickstarter.com/discover/categories/music/country%20&%20folk   \n",
       "\n",
       "        spotlight  staff_pick       state    state_changed_at  \\\n",
       "122464       True       False  successful 2019-10-16 17:00:11   \n",
       "\n",
       "        static_usd_rate  \\\n",
       "122464         1.250602   \n",
       "\n",
       "                                                                                                       urls  \\\n",
       "122464  {\"web\":{\"project\":\"https://www.kickstarter.com/projects/alicewheeldonstory/framed-the-alice-whee...   \n",
       "\n",
       "        usd_pledged  usd_type  \n",
       "122464  3778.068189  domestic  "
      ]
     },
     "execution_count": 13,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df[df['id'] == 1103759530]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "successful    97906\n",
       "failed        71720\n",
       "canceled       8593\n",
       "live           2534\n",
       "Name: state, dtype: int64"
      ]
     },
     "execution_count": 14,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df['state'].value_counts()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Text(0, 0.5, 'Count')"
      ]
     },
     "execution_count": 15,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAWQAAAFVCAYAAADYPZUoAAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADh0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uMy4yLjEsIGh0dHA6Ly9tYXRwbG90bGliLm9yZy+j8jraAAAgAElEQVR4nO3deZxcRb338U+ZwRhEImSEm5mgAYkiIAJBRL0iEoKACyj4I24EjOQRQfC6gvg8KKAPXFFkuaABJEHU8AMXorIajIIiiIgsQSGSAJOJhElCCIuBhHP/qBpy0ulZOumZrkl/369Xv7q7ztJ1qqe/U119llAUBSIi0ngvaXQFREQkUiCLiGRCgSwikgkFsohIJhTIIiKZUCCLiGRCgSyyEQsh7BNCKEIIRza6LtI3BbL0KYSwXQhhWgjh7yGEZ0IIy0IIc0MIM0II76qY92shhEPq8JqfbXSIlMKsfPt3COGhEMKlIYQ3NLBurwshXJDek6dDCM+GEB5I79ObG1WvUv3Gpr+FXRtdl6GkpdEVkLyFEPYAfgc8D1wG3AeMAF4HvA9YAfy2tMgpwAzgFxv40p8FFgDTN3A99fAT4Jr0eASwC/BJ4NAQwhuLonh4MCsTQpgCXAj8O9XtLmAV8T05FDg6hLBTURRzB7NeFcYS/xYWEOsn/aBAlr6cAmwK7FYUxVofrBDCccB/NKRWg+vOoiguLxeEEB4EzgE+CJxdrxcKIbyiKIoVvUzfD5gGzAXeXRRFZ8X0k4DP1Ks+teqr/tI7DVlIX8YBSyrDGKAoihe6AyF9Re0+Dn9y+Wt+9/whhMNDCLNCCI+EEFaGELpCCL8IIexSXm9a5jXAOyuGC8Z2Tw8hTK+sTwjhyDRtn1LZliGEs0MI/0zDDUtCCH8JIXxxA9ulOwifq6jDp0MIN4QQFoYQngshLAohXN5d98rtDCFMDyFMCCHcEkJ4CvhlH697JhCAwyvDGKAoilVFUZxdrXccQjgqhHBfavuHQwhfqjLP/iGEK9KwzLMhhCfS9ryzyrxzQggL0pDWVSGEpcCTaaip+1vTpaX3b04f29b01EOWvvwTeH0I4YNFUfysl/keBz4O/BC4mdiLq3QcsDRN+xfwWmAq8IcQwu5FUTyY5vs4sdfZBXyj4jVqdSWwN/B94G/E3v4OwD7At/q5jk1DCK3p8Qhg51SvLuCnFfN+AfgTcC5xW3cmDm/sm4Y3llTMvwdxmOEi4lBPj0II2wK7Azevx3DEp4CtgUuAJ4CPAWeGEDqKovhxab4jgS2Jw1MdQHuq/+wQwruKori5Yr2bEYe0/gCcDGwF/B74JvAV4nvdvcxjNda5+RRFoZtuPd6AtxJ7gQXwAPAD4BjgDT3MXwDTe5j28iplbwBWAhdUlC8A5tTyGsQwKYB90vOR6fkF1dbTj23fJy1f7XYfsEM/t3FCWuZLVbajAPbrZ33el+Y/dz22oRN4Zal8U+I/uFv7Uf+tif98rqkon5PWfXovr3tko/+Gh9JNQxbSq6IobgXGE3tvI4GjgAuAuSGEm0MI29WwrqcBQrR56nU+DvwDeEvdKw/PEsP+LdWGDGowDZiYbu8Dvgy0AteEEF5TnrG0jS8JIYxM2/g3YDnVt/FvRVH8pp/12DzdP1n7JnBpURRPlOr5DLEnP648U3f9AUIIm4UQRgGrgdvo+T06az3qI1UokKVPRVHcUxTFkUVRbE389Xwy8WvofwJXhxBe2p/1hBB2CyH8irhnxnJiGD8OvBHYYgDq/Rxxb42dgflp/PS8EMKEGlf1YFEUv0m3XxVF8d/A+4FtiWO6Lwoh7JvGSp8mDg10b+NIqm/jAzXUozuIX1Fj/QEeqlK2BBhVLgghvDaEMDOEsIz4PnUR638Q1ev/eDnoZcNoDFlqUsRdvC4LIXSPFb8d2BO4pbflQgivJo4tPgmcRuwVP038Wvtd4ljkhlrn77koiu+FEK4G3gO8EzgMOC6EcEVRFJPW94WKorgthLAc2Le7LO3/ewMwDzgRmE/spRfATKp3gJ6p4WXvTfe7rUeVV/c1QwhhM+J79HLie3IPMZRfAE6itK0ltdRf+qBAlvVSFEURQriNGMjt/VjkA8TQfX9RFOX9lklfi1dWvkQv61pK/OGpUtXhk6IoFgEXAxeHEIYRf3j8cAjh20VR/Lkfde9JCzC89PwjwDDgwKIo5ncXhhBeTh2+ARRFMT+E8Ffg7SGEHYqi+PuGrrPCBKAN+ERRFJeWJ4QQTq9xXbryxXrQkIX0KoQwMYSwzj/uEMIIYP/0tPyL/1NUD8vuHlqoWM/RVN+Xuaf1QPya/9YQwqal9WxBHN8ur3vT8jwARVGsBu5OT3taf59CCBOJPcm/lIqrbiNxb4N6fda+nO5nhhDWabcQwrAQj3LccT3W3dN7tD+1j/E/le7Xu42bkXrI0pezgVEhhFnEr7DPANsQe4OvAy4riuKe0vx/AvYLIXwZeITYmZ4JXJuW/WEI4XxgGbF3fRBx17rKv8U/AVNCCKcB9xO/Nv8y/eh0PnA5cFMaOnklcDTwMGuH++uA34UQfk78ur+MuFfHMcThhMpduHqyewjhY+nxcGCn9HrPA18tzfdz4L+IP/ZNI+6dMpF4ZF9XP1+rV0VR3BhCmEo8Uu8fIYTykXrbE3ehey1x3LxWtxB3R/x2+hG0A9iVuBviPcSx/v6aSxzu+HQI4RniePrioihuWo96NY9G7+ahW943Yi/4f4h7CnQRP/hLiDv+fwJ4ScX844jjqE+SdusqTdub+KFfQfyA/poYHHOABRXr2Yq4j+9SYhgXwNjS9C8SA3glMbA/wbq7vY0i/kO5K73es8Tx3e8Co/ux7fvAOru7rQYWAz8D3lxlmUOIveanU3vNBF5Nld346GUXwX7U7fXEUH6A+I/u38Rx+e8Tj6qs3IYjq6xjevn9SWW7ANcR/3mtSO/NO3qYd533rWL6QcCdqW5F5fbrtu4tpIYTEZEG0xiyiEgmFMgiIplQIIuIZEKBLCKSCQWyiEgmtB9yz7T7iYgMlMqDhwAFcq86O9c5/3c2Wltb6eqqy7EGQ57aIlI7rJFzW7S1tfU4TUMWIiKZUCCLiGRi0IYszOwHwHuBxe6+cyrbEriCeI7dBYC5+7I07SRgCvFQ1ePd/fpUPp54GOcI4pWAT3D3wsyGEy87M554aO/h7r4gLTOZNeccON3de71UjohIIwxmD3k6cEBF2YnAbHcfB8xOzzGzHYFJxJO4HABcYGbD0jIXEq/DNi7dutc5BVjm7tsTz19wZlrXlsQrJ7+FeN7eU8ys7idDFxHZUIMWyO7+e+KJYsoOZs2FHWcQT8zSXT7T3Ve6+3ziCWH2NLPRwObufqu7F8Qe8SFV1nUVMMHMAvBu4EZ3X5p63zey7j8GEZGGa/ReFlu7+yIAd19kZlul8nbi6Re7dV/99vn0uLK8e5lH07pWmdly4tm+XiyvssxazGwqsfeNu9Pa2lpttiy0tLRkXb/BpLaI1A5rDNW2aHQg96TaPnpFL+Xru8xa3H0aay5fX+S62wzkvVvPYFNbRGqHNXJui5x3e3ssDUOQ7hen8g7iSdC7jSFexrwjPa4sX2sZM2shXlRyaS/rEhHJSqMDeRbxCsak+6tL5ZPMbLiZbUv88e72NLyxwsz2SuPDR1Qs072uw4Cb0jjz9cD+ZrZF+jFv/1QmIpKVQQtkM/sJcCvwejPrMLMpwBnARDN7kHipmzMA3P0+wImXgbkOONbdu6/3dQzxgpXziJf+uTaVXwKMMrN5wOdIe2y4+1LiVY7/nG6npjIRkazoiiE9K3To9NCgtojUDmvk3BZpDFnnshho7RdV3XljyFt49MJGV0GkKTR6DFlERBIFsohIJhTIIiKZUCCLiGRCgSwikgkFsohIJhTIIiKZUCCLiGRCgSwikgkFsohIJhTIIiKZUCCLiGRCgSwikgkFsohIJhTIIiKZUCCLiGRCgSwikgkFsohIJhTIIiKZUCCLiGRCgSwikgkFsohIJhTIIiKZUCCLiGRCgSwikgkFsohIJhTIIiKZUCCLiGRCgSwikgkFsohIJhTIIiKZUCCLiGRCgSwikgkFsohIJhTIIiKZUCCLiGRCgSwikgkFsohIJhTIIiKZUCCLiGRCgSwikgkFsohIJloaXQEAM/sv4JNAAdwDHAVsClwBjAUWAObuy9L8JwFTgNXA8e5+fSofD0wHRgDXACe4e2Fmw4HLgPHAEuBwd18wOFsnItI/De8hm1k7cDywh7vvDAwDJgEnArPdfRwwOz3HzHZM03cCDgAuMLNhaXUXAlOBcel2QCqfAixz9+2Bs4EzB2HTRERq0vBATlqAEWbWQuwZdwIHAzPS9BnAIenxwcBMd1/p7vOBecCeZjYa2Nzdb3X3gtgjLi/Tva6rgAlmFgZ6o0REatHwQHb3hcBZwCPAImC5u98AbO3ui9I8i4Ct0iLtwKOlVXSksvb0uLJ8rWXcfRWwHBg1ENsjIrK+Gj6GbGZbEHuw2wJPAFea2cd6WaRaz7bopby3ZSrrMpU45IG709ra2ks1mkfu7dDS0pJ9HQeD2mGNodoWDQ9kYD9gvrs/DmBmPwPeBjxmZqPdfVEajlic5u8AtiktP4Y4xNGRHleWl5fpSMMiI4GllRVx92nAtPS06OrqqsPmDX25t0Nra2v2dRwMaoc1cm6Ltra2HqflEMiPAHuZ2abAs8AE4A7gaWAycEa6vzrNPwv4sZl9B2gj/nh3u7uvNrMVZrYXcBtwBHBeaZnJwK3AYcBNaZxZRCQbOYwh30b8oe1O4i5vLyH2Us8AJprZg8DE9Bx3vw9wYC5wHXCsu69OqzsGuJj4Q98/gWtT+SXAKDObB3yOtMeGiEhOQlGoo9iDorOzs++5Stovau97piFo4dELG12FXuX89XQwqR3WyLkt0pBF1b28Gt5DFhGRSIEsIpIJBbKISCYUyCIimVAgi4hkQoEsIpIJBbKISCYUyCIimVAgi4hkQoEsIpIJBbKISCYUyCIimVAgi4hkQoEsIpIJBbKISCYUyCIimVAgi4hkIodr6slGaGO8ekruV06RoU89ZBGRTCiQRUQyoUAWEcmEAllEJBMKZBGRTCiQRUQyoUAWEcmEAllEJBMKZBGRTCiQRUQyoUAWEcmEAllEJBMKZBGRTCiQRUQyoUAWEcmEAllEJBMKZBGRTCiQRUQyoUAWEcmEAllEJBMKZBGRTCiQRUQyoUAWEcmEAllEJBMKZBGRTCiQRUQy0dLoCgCY2SuBi4GdgQL4BPAP4ApgLLAAMHdfluY/CZgCrAaOd/frU/l4YDowArgGOMHdCzMbDlwGjAeWAIe7+4LB2ToRkf7JpYd8DnCdu+8AvAm4HzgRmO3u44DZ6TlmtiMwCdgJOAC4wMyGpfVcCEwFxqXbAal8CrDM3bcHzgbOHIyNEhGpRcMD2cw2B/YGLgFw9+fc/QngYGBGmm0GcEh6fDAw091Xuvt8YB6wp5mNBjZ391vdvSD2iMvLdK/rKmCCmYUB3jQRkZo0PJCB7YDHgUvN7K9mdrGZvRzY2t0XAaT7rdL87cCjpeU7Ull7elxZvtYy7r4KWA6MGpjNERFZPzmMIbcAuwOfcffbzOwc0vBED6r1bIteyntbZi1mNpU45IG709ra2lu9m4baIcq9HVpaWrKv42AZqm2RQyB3AB3uflt6fhUxkB8zs9HuvigNRywuzb9NafkxQGcqH1OlvLxMh5m1ACOBpZUVcfdpwLT0tOjq6trQbdsoqB2i3NuhtbU1+zoOlpzboq2trcdpDR+ycPd/AY+a2etT0QRgLjALmJzKJgNXp8ezgElmNtzMtiX+eHd7GtZYYWZ7pfHhIyqW6V7XYcBNaZxZRCQbOfSQAT4D/MjMXgo8BBxF/GfhZjYFeAT4EIC732dmTgztVcCx7r46recY1uz2dm26QfzB8IdmNo/YM540GBslIlKLUBT96yia2d7AH9OPYuXyFuBt7v77AahfIxWdnZ19z1XSflF73zMNQQuPXljzMhtjW6xPOwymnL+mD7ac2yINWVTdy6uWIYvfAltWKR+ZpomIyAaoJZADVfZMIO4+9nR9qiMi0rz6HEM2s1npYQFcbmYrS5OHEQ93/uMA1E1EpKn050e9Jek+AMuAZ0vTngNuAS6qc71ERJpOn4Hs7kcBmNkC4Cx31/CEiMgA6Pdub+7+9YGsiIhIs+t3IJvZlsA3iAdubEXFD4Luvnl9qyYi0lxqOTDkEmA34qHFnVTf40JERNZTLYE8AZhYOueEiIjUUS37IS8GnhqoioiINLtaAvlk4FQz22ygKiMi0sxqGbL4KvH6dovN7GHg+fJEd9+ljvUSEWk6tQTyVQNWCxER0X7IIiK5aPgJ6kVEJKrlwJAV9LLvsQ4MERHZMLWMIR9X8XwT4oEihxKP4BMRkQ1QyxjyjGrlZnYn8aCR8+pVKRGRZlSPMeTfAu+rw3pERJpaPQJ5EpDnxatERIaQWn7Uu4e1f9QLwNbE6+wdU+d6iYg0nQ05MOQF4HFgjrv/vX5VEhFpTjowREQkE7X0kAEws32BHYnDF/e5+5x6V0pEpBnVMobcDvwcGE88QT1Am5ndAXzA3Tt7XFhERPpUy14W5wKrge3dfRt33wYYl8rOHYjKiYg0k1oCeSJwrLvP7y5w94eA49M0ERHZAPXYD/mFOqxDRKTp1RLIs4FzzWyb7gIzezVwTpomIiIboJa9LI4HrgYeMrPuq063A3enaSIisgFq2Q/5UWB3M5sI7EA8Um+uu/9moConItJM+gxkMzsQuBB4k7svd/cbgRvTtJFmtgCY6u43DGhNRUQ2cv0ZQz4O+Ja7L6+ckMrOBE6od8VERJpNfwJ5F6C3YYmbgDfVpzoiIs2rP4H8Knrfta0ARtWnOiIizas/gdxB7CX3ZBdgYX2qIyLSvPoTyL8GTjOzEZUTzGxT4NQ0j4iIbID+7Pb2DeAw4EEzOw/oPvfxG4g/+AXgmwNTPRGR5tFnD9ndFwNvIx4A8k3iGd9+Tgzqu4G3u/tjA1lJEZFm0K8DQ9z9YeAgM9sC2J7YK37Q3ZcNZOVERJpJTSeoTwH85wGqi4hIU6vH2d5ERKQOFMgiIplQIIuIZKLmi5wOFDMbBtwBLHT395rZlsAVwFhgAWDdPyKa2UnAFOLlo4539+tT+XhgOjACuAY4wd0LMxsOXEa8HuAS4HB3XzBoGyci0g859ZBPAO4vPT8RmO3u44gnwD8RwMx2BCYBOwEHABekMId4VrqpxGv9jUvTIYb3MnffHjibeEIkEZGsZBHIZjYGeA9wcan4YGBGejwDOKRUPtPdV6br+80D9jSz0cDm7n6ruxfEHvEhVdZ1FTDBzMKAbZCIyHrIIpCB7wJfYu2TGG3t7osA0v1WqbwdeLQ0X0cqa0+PK8vXWsbdVwHL0QmRRCQzDR9DNrP3Aovd/S9mtk8/FqnWsy16Ke9tmcq6TCUOeeDutLa29qM6Gz+1Q5R7O7S0tGRfx8EyVNui4YEMvB14v5kdBLwM2NzMLgceM7PR7r4oDUcsTvN3ANuUlh8DdKbyMVXKy8t0mFkLMBJYWlkRd58GTEtPi66urnps35Cndohyb4fW1tbs6zhYcm6Ltra2Hqc1fMjC3U9y9zHuPpb4Y91N7v4xYBYwOc02mXiBVVL5JDMbbmbbEn+8uz0Na6wws73S+PARFct0r+uw9Brr9JBFRBqp4YHcizOAiWb2IDAxPcfd7wMcmAtcBxzr7qvTMscQfxicB/wTuDaVXwKMMrN5wOdIe2yIiOQkFIU6ij0oOjs7+56rpP2i9r5nGoIWHl379Qc2xrZYn3YYTDl/TR9sObdFGrKoupdXzj1kEZGmokAWEcmEAllEJBMKZBGRTCiQRUQyoUAWEcmEAllEJBMKZBGRTCiQRUQyoUAWEcmEAllEJBMKZBGRTCiQRUQyoUAWEcmEAllEJBMKZBGRTCiQRUQyoUAWEcmEAllEJBMKZBGRTCiQRUQyoUAWEcmEAllEJBMKZBGRTCiQRUQyoUAWEcmEAllEJBMKZBGRTCiQRUQyoUAWEcmEAllEJBMKZBGRTCiQRUQyoUAWEcmEAllEJBMKZBGRTCiQRUQyoUAWEcmEAllEJBMKZBGRTCiQRUQyoUAWEcmEAllEJBMKZBGRTLQ0ugJmtg1wGfAfwAvANHc/x8y2BK4AxgILAHP3ZWmZk4ApwGrgeHe/PpWPB6YDI4BrgBPcvTCz4ek1xgNLgMPdfcEgbaKISL/k0ENeBXze3d8A7AUca2Y7AicCs919HDA7PSdNmwTsBBwAXGBmw9K6LgSmAuPS7YBUPgVY5u7bA2cDZw7GhomI1KLhgezui9z9zvR4BXA/0A4cDMxIs80ADkmPDwZmuvtKd58PzAP2NLPRwObufqu7F8QecXmZ7nVdBUwwszDAmyYiUpOGB3KZmY0FdgNuA7Z290UQQxvYKs3WDjxaWqwjlbWnx5Xlay3j7quA5cCoAdkIEZH11PAx5G5mthnwU+Cz7v6kmfU0a7WebdFLeW/LVNZhKnHIA3entbW1r2o3BbVDlHs7tLS0ZF/HwTJU2yKLQDazTYhh/CN3/1kqfszMRrv7ojQcsTiVdwDblBYfA3Sm8jFVysvLdJhZCzASWFpZD3efBkxLT4uurq4N3raNgdohyr0dWltbs6/jYMm5Ldra2nqc1vAhizSWewlwv7t/pzRpFjA5PZ4MXF0qn2Rmw81sW+KPd7enYY0VZrZXWucRFct0r+sw4KY0ziwiko0ceshvBz4O3GNmd6WyrwBnAG5mU4BHgA8BuPt9ZubAXOIeGse6++q03DGs2e3t2nSDGPg/NLN5xJ7xpIHeKBGRWoWiUEexB0VnZ2ffc5W0X9Te90xD0MKjF9a8zMbYFuvTDoMp56/pgy3ntkhDFlX38mr4kIWIiEQKZBGRTCiQRUQyoUAWEcmEAllEJBMKZBGRTCiQRUQyoUAWEcmEAllEJBMKZBGRTCiQRUQyoUAWEcmEAllEJBMKZBGRTCiQRUQyoUAWEcmEAllEJBMKZBGRTCiQRUQyoUAWEcmEAllEJBMKZBGRTCiQRUQyoUAWEcmEAllEJBMtja6AyMau/aL2Rleh7hYevbDRVdgoqYcsIpIJBbKISCYUyCIimVAgi4hkQoEsIpIJBbKISCYUyCIimVAgi4hkQoEsIpIJBbKISCYUyCIimVAgi4hkQoEsIpIJBbKISCYUyCIimdD5kEVkUGyM54WG+p4bWj1kEZFMKJBFRDLRVEMWZnYAcA4wDLjY3c9ocJVERF7UND1kMxsG/A9wILAj8GEz27GxtRIRWaNpAhnYE5jn7g+5+3PATODgBtdJRORFzRTI7cCjpecdqUxEJAvNNIYcqpQV5SdmNhWYCuDutLW11fQCxSlF3zM1CbXFGmqLSO3Qt2bqIXcA25SejwE6yzO4+zR338Pd9yAGeLY3M/tLo+uQy01toXYYgm1RVTP1kP8MjDOzbYGFwCTgI42tkojIGk3TQ3b3VcBxwPXA/bHI72tsrURE1mimHjLufg1wTaPrUSfTGl2BjKgtIrXDGkOyLUJRaKBdRCQHTTNkISKSOwVyEzGzd5jZfWZ2l5mN6GW+OWa2x2DWrfTax5vZ/Wb2ox6m72Fm56bHR5rZ+TWuf4GZtdajrkPJxtZWZvZUum8zs6saXZ96aaoxZOGjwFnufmmjK9KLTwMHuvv8ahPd/Q7gjsGtkuTK3TuBwxpdj3pRINeZmb0ccOJ+zsOA04AzgT3cvSv1PM9y933MbDPgPGAP4kEqX3f3n6aTIH0zLd/l7hPSes8D3kh8377m7leb2U7ApcBLid94DiXuX11Zh1cABrzbzPYDLgK+4O7vTfU+H7jD3acPbAv1zMy+B2wHzDKzy4mHto8AngWOcvd/mNk+lOpdWvZVwPeAV6eiz7r7H8xsFPAT4FXA7fSyD+hgMrMjgC8Q3/e7ie/XV4nv4xLgo+7+mJl9jbhN26X777r7udXW4e4f76kdKl57SLVVb8xsLPArd9/ZzG4DPtG995SZzQE+D/ydKp+dxtS4dxqyqL8DgE53f5O77wxc18u8/xdY7u5vdPddgJvSh+Ui4FB3fxPwoTTvycBN7v5m4F3At1JIfwo4x913JQZ7R7U6uPvFwCzgi+7+0bpvdR24+6eI/0zeBVwI7O3uuwH/j/gPqjfnAGen9jkUuDiVnwLcktYzizUh1DDpn+jJwL7pPT4BuAXYK9VzJvCl0iI7AO8mno/lFDPbpId1QM/tUDZk2qpGM4mdDsxsNNDm7n+h589OdhTI9XcPsJ+ZnWlm73D35b3Mux/xDHQAuPsyYC/g991f2d19aZq8P3Cimd0FzAFeRvzA3Ap8xcy+DLzG3Z+tsQ65GglcaWb3AmcDO/Ux/37A+al9ZgGbm9krgL2BywHc/dfAsoGrcr/tC1zl7l3w4ns8BrjezO4Bvsja2/trd1+Z5l8MbN3DOqDndigbSm1VC2dNB8aAK9Pjnj472dGQRZ25+wNmNh44CPj/ZnYDsIo1//xeVpo9UHE+jR7KussPdfd/VJTfn76qvYf4gf6ku99UWQd3P7ViuXKdKuuVg9OA37r7B9LX0jl9zP8S4K3pH9KLzAyqt2cjVXuPzwO+4+6z0rDM10rTVpYeryZ+bnv6O+mtHfozT25t1W/uvtDMlpjZLsDhwP9Jk3r67GRHPeQ6M7M24Bl3vxw4C9gdWACMT7McWpr9BuLRg93LbkHs8b4zHeKNmW2ZJl8PfMbMQirfLd1vBzyUxhVnAbv0UIdKDwM7mtlwMxsJTNjQba+zkcRD3AGO7Mf8lW25a3r4e+KPmZjZgcAW9aviepsNWBqz7X6Py9s7eT3XAT23Q9lQaqtadQ/3jHT3e1JZ1c9OjhTI9fdG4Pb09ehk4HTg68A5ZnYzsYfT7XRgCzO718z+BrzL3R8nnnHuZ6nsijTvacAmwN3pa/xpqfxw4N70ejsAl/VQh7W4+6PEr3h3Az8C/lqvBqiT/yb27v9A/GGyL8cDe5jZ3WY2lzi2DrHt9zazO4lfXR8ZkNrWIP3o9A3gd+k9/g6xR3xl+hvpWs91QM/tUCRVvNYAAAMWSURBVDZk2mo9XEU8T42Xynr67GRHR+qJiGRCPWQRkUwokEVEMqFAFhHJhAJZRCQTCmQRkUwokEVEMqEj9aTppPOFfJ14JONo4AngXuAMd7/RzBYA57v7WTWudw5wr7sf19e8ItUokKUZ/RTYFJgCzAO2At4JjGpkpUR0YIg0FTN7JfGkORPd/TdVps8hhvOL3D2kQ5TPB95BDO6HKJ1b2syms+4hz9sCY4HfAq/qPhFQOjfHfODN7n6HmW0CfJt4Xt9RxBMI/cjdT9zwLZahRGPI0myeSrf3m1m1Eyp9kHgK01OJwxmjU/nLgDuB9xLPxHYO8H0z6z4HyAnE85BcWlru0X7W6XjgA8RDfscRD4fP/kQ4Un8aspCm4u6rzOxI4jmnp5rZX4E/AFe6+23uvtTMVgMr3P1fpeUWAt8qrWqame0LfBiY7e7Lzew54kmdXlyu4ixrPXkN8ABws7sXxHNI/HGDNlSGJAWyNJ10VZZfE4cf3ko8of/nzexkd696InwzGwacSOy9tgPDiVf3mFOHKk0HbgQeSKdrvQa41t1fqMO6ZQjRkIU0JXf/t7vf6O6nuvvbgEuAr5nZS3tY5AvEywF9i3iq0l2BXxBDuTfdoVq+HNImFXW5kzjW/BXiZ3IGcKOZ6fPZZNRDFonmEj8PLwOeY91Tfv4n8Et3/yFAOrfu64i7zHWrttzj6X506fE65yh29xXEK1xcmX4g/BOwPXEoQ5qEAlmaStpb4krgB8RzQa8gXovwS8Sx4CfTfsjvSBda7b500gPA4Wb2n8TzFX+GuBdF+TzSC4A9014UTwFLibvVPUrsfZ9I7Al/taJOnwMWAXcBzwMfAZ4k/rgoTURfiaTZPEXsfZ4A/A64j3gB1R8Tx4chXlR1G+CfrOnVnk68EvO1xCtrPE08sX/ZWcRe8ty03Kvd/Xni3hPbAX8jHpDylYrlVhCvo3c7cU+OXYED3f2ZDd5aGVK0H7KISCbUQxYRyYQCWUQkEwpkEZFMKJBFRDKhQBYRyYQCWUQkEwpkEZFMKJBFRDKhQBYRycT/AjLJTMwxcV3SAAAAAElFTkSuQmCC\n",
      "text/plain": [
       "<Figure size 360x360 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "fig, ax = plt.subplots(figsize=(5,5))\n",
    "\n",
    "x = df['state'].value_counts()\n",
    "ax.bar(x.index, height=x.values, width=0.8, color='green')\n",
    "ax.set_title('Status Bar Chart', fontsize=18)\n",
    "ax.set_xlabel(\"Status\", fontsize=14)\n",
    "ax.set_ylabel(\"Count\", fontsize=14)\n",
    "# plt.savefig('status_bar_chart')"
   ]
  },
  {
   "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>state</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>spotlight</th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>False</th>\n",
       "      <td>82847</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>True</th>\n",
       "      <td>97906</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "           state\n",
       "spotlight       \n",
       "False      82847\n",
       "True       97906"
      ]
     },
     "execution_count": 16,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#spotlight column is data leakage. Needs removal \n",
    "df[['state', 'spotlight']].groupby('spotlight').count()"
   ]
  },
  {
   "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>state</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>staff_pick</th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>False</th>\n",
       "      <td>157840</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>True</th>\n",
       "      <td>22913</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "             state\n",
       "staff_pick        \n",
       "False       157840\n",
       "True         22913"
      ]
     },
     "execution_count": 17,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df[['state', 'staff_pick']].groupby('staff_pick').count()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(172160, 38)"
      ]
     },
     "execution_count": 18,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# remove entries that are cancelled\n",
    "df1 = df[df['state'].isin(['successful', 'failed', 'live'])]\n",
    "df1.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "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>backers_count</th>\n",
       "      <th>blurb</th>\n",
       "      <th>category</th>\n",
       "      <th>country_displayable_name</th>\n",
       "      <th>deadline</th>\n",
       "      <th>goal</th>\n",
       "      <th>launched_at</th>\n",
       "      <th>name</th>\n",
       "      <th>pledged</th>\n",
       "      <th>slug</th>\n",
       "      <th>staff_pick</th>\n",
       "      <th>state</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>5</td>\n",
       "      <td>A Free to Win Online Mobile Strategy Game to end the corruption of the rich players always winni...</td>\n",
       "      <td>{\"id\":272,\"name\":\"Mobile Games\",\"slug\":\"games/mobile games\",\"position\":3,\"parent_id\":12,\"parent_...</td>\n",
       "      <td>Canada</td>\n",
       "      <td>2017-10-09 14:41:03</td>\n",
       "      <td>1000.0</td>\n",
       "      <td>2017-09-25 14:41:03</td>\n",
       "      <td>Iron Age Kingdoms - First Free To Win Mobile Strategy Game</td>\n",
       "      <td>105.0</td>\n",
       "      <td>iron-age-kingdoms-first-free-to-win-mobile-strateg</td>\n",
       "      <td>False</td>\n",
       "      <td>failed</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>19</td>\n",
       "      <td>Lizard is a dark comedy and modern western, that takes place in the Mojave Desert.</td>\n",
       "      <td>{\"id\":11,\"name\":\"Film &amp; Video\",\"slug\":\"film &amp; video\",\"position\":7,\"color\":16734574,\"urls\":{\"web\"...</td>\n",
       "      <td>the United States</td>\n",
       "      <td>2019-09-21 23:39:48</td>\n",
       "      <td>1100.0</td>\n",
       "      <td>2019-07-23 23:39:48</td>\n",
       "      <td>\"Lizard\" a short film</td>\n",
       "      <td>2358.0</td>\n",
       "      <td>lizard-a-short-film</td>\n",
       "      <td>False</td>\n",
       "      <td>successful</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>5</td>\n",
       "      <td>My friend and I are making an app that has never been made before. Making a football turn based ...</td>\n",
       "      <td>{\"id\":272,\"name\":\"Mobile Games\",\"slug\":\"games/mobile games\",\"position\":3,\"parent_id\":12,\"parent_...</td>\n",
       "      <td>the United States</td>\n",
       "      <td>2016-03-22 21:11:17</td>\n",
       "      <td>25000.0</td>\n",
       "      <td>2016-01-22 22:11:17</td>\n",
       "      <td>Total Franchise Football 2016</td>\n",
       "      <td>172.0</td>\n",
       "      <td>total-franchise-football-2016</td>\n",
       "      <td>False</td>\n",
       "      <td>failed</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>23</td>\n",
       "      <td>More than a bike trip, a state of mind. And a book.</td>\n",
       "      <td>{\"id\":48,\"name\":\"Nonfiction\",\"slug\":\"publishing/nonfiction\",\"position\":9,\"parent_id\":18,\"parent_...</td>\n",
       "      <td>the United States</td>\n",
       "      <td>2012-09-25 02:33:06</td>\n",
       "      <td>500.0</td>\n",
       "      <td>2012-09-15 02:33:06</td>\n",
       "      <td>The Trans-American Psychogeographic Literary Cycle Tour</td>\n",
       "      <td>501.0</td>\n",
       "      <td>the-trans-american-psychogeographic-literary-cycle</td>\n",
       "      <td>False</td>\n",
       "      <td>successful</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>0</td>\n",
       "      <td>This is the birth of a revolutionary artist...Revery! A burst out into the world with GAME CHANG...</td>\n",
       "      <td>{\"id\":39,\"name\":\"Hip-Hop\",\"slug\":\"music/hip-hop\",\"position\":8,\"parent_id\":14,\"parent_name\":\"Musi...</td>\n",
       "      <td>the United States</td>\n",
       "      <td>2017-12-11 02:43:11</td>\n",
       "      <td>2500.0</td>\n",
       "      <td>2017-11-10 02:43:11</td>\n",
       "      <td>The Golden Ticket to the Wonka Factory</td>\n",
       "      <td>0.0</td>\n",
       "      <td>the-golden-ticket-to-the-wonka-factory</td>\n",
       "      <td>False</td>\n",
       "      <td>failed</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>338763</th>\n",
       "      <td>48</td>\n",
       "      <td>The bee all end all Kickstarter for crazy bitches. \\nWe're raising funds for bee hives!</td>\n",
       "      <td>{\"id\":10,\"name\":\"Food\",\"slug\":\"food\",\"position\":8,\"color\":16725570,\"urls\":{\"web\":{\"discover\":\"ht...</td>\n",
       "      <td>the United States</td>\n",
       "      <td>2019-03-29 22:35:40</td>\n",
       "      <td>550.0</td>\n",
       "      <td>2019-02-27 23:35:40</td>\n",
       "      <td>Bitches Bee Crazy Hives</td>\n",
       "      <td>2145.0</td>\n",
       "      <td>bitches-bee-crazy-hives</td>\n",
       "      <td>False</td>\n",
       "      <td>successful</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>338764</th>\n",
       "      <td>179</td>\n",
       "      <td>Detective Tori Jones and her partner Officer Rashida Thompson are back!  Kicking crime's ass and...</td>\n",
       "      <td>{\"id\":33,\"name\":\"Webseries\",\"slug\":\"film &amp; video/webseries\",\"position\":19,\"parent_id\":11,\"parent...</td>\n",
       "      <td>the United States</td>\n",
       "      <td>2012-05-19 04:00:00</td>\n",
       "      <td>30000.0</td>\n",
       "      <td>2012-04-15 19:31:32</td>\n",
       "      <td>Lesbian Cops: The Movie - Season 2</td>\n",
       "      <td>30045.5</td>\n",
       "      <td>lesbian-cops-the-movie-season-2-0</td>\n",
       "      <td>False</td>\n",
       "      <td>successful</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>338765</th>\n",
       "      <td>18</td>\n",
       "      <td>Help me produce the play I have written for my senior project!</td>\n",
       "      <td>{\"id\":285,\"name\":\"Plays\",\"slug\":\"theater/plays\",\"position\":6,\"parent_id\":17,\"parent_name\":\"Theat...</td>\n",
       "      <td>the United States</td>\n",
       "      <td>2014-09-30 16:00:00</td>\n",
       "      <td>500.0</td>\n",
       "      <td>2014-08-31 14:03:20</td>\n",
       "      <td>Generations (Senior Project)</td>\n",
       "      <td>606.0</td>\n",
       "      <td>generations-senior-project</td>\n",
       "      <td>False</td>\n",
       "      <td>successful</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>338766</th>\n",
       "      <td>0</td>\n",
       "      <td>This documentary asks if the twenty-four hour news cycle has altered people's opinions of one an...</td>\n",
       "      <td>{\"id\":21,\"name\":\"Digital Art\",\"slug\":\"art/digital art\",\"position\":3,\"parent_id\":1,\"parent_name\":...</td>\n",
       "      <td>the United States</td>\n",
       "      <td>2013-06-07 18:39:19</td>\n",
       "      <td>10000.0</td>\n",
       "      <td>2013-05-08 18:39:19</td>\n",
       "      <td>US: A Story of our United States</td>\n",
       "      <td>0.0</td>\n",
       "      <td>us-a-story-of-our-united-states</td>\n",
       "      <td>False</td>\n",
       "      <td>failed</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>338767</th>\n",
       "      <td>1</td>\n",
       "      <td>a short film created by children ages 8-10. the premise of the movie is based on 3 kids who try ...</td>\n",
       "      <td>{\"id\":330,\"name\":\"Family\",\"slug\":\"film &amp; video/family\",\"position\":7,\"parent_id\":11,\"parent_name\"...</td>\n",
       "      <td>the United States</td>\n",
       "      <td>2014-08-06 21:04:08</td>\n",
       "      <td>3500.0</td>\n",
       "      <td>2014-07-07 21:04:08</td>\n",
       "      <td>Lights, Camera, and Behind the Action!</td>\n",
       "      <td>25.0</td>\n",
       "      <td>lights-camera-and-behind-the-action</td>\n",
       "      <td>False</td>\n",
       "      <td>failed</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>172160 rows × 12 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "        backers_count  \\\n",
       "0                   5   \n",
       "1                  19   \n",
       "2                   5   \n",
       "3                  23   \n",
       "4                   0   \n",
       "...               ...   \n",
       "338763             48   \n",
       "338764            179   \n",
       "338765             18   \n",
       "338766              0   \n",
       "338767              1   \n",
       "\n",
       "                                                                                                      blurb  \\\n",
       "0       A Free to Win Online Mobile Strategy Game to end the corruption of the rich players always winni...   \n",
       "1                        Lizard is a dark comedy and modern western, that takes place in the Mojave Desert.   \n",
       "2       My friend and I are making an app that has never been made before. Making a football turn based ...   \n",
       "3                                                       More than a bike trip, a state of mind. And a book.   \n",
       "4       This is the birth of a revolutionary artist...Revery! A burst out into the world with GAME CHANG...   \n",
       "...                                                                                                     ...   \n",
       "338763              The bee all end all Kickstarter for crazy bitches. \\nWe're raising funds for bee hives!   \n",
       "338764  Detective Tori Jones and her partner Officer Rashida Thompson are back!  Kicking crime's ass and...   \n",
       "338765                                       Help me produce the play I have written for my senior project!   \n",
       "338766  This documentary asks if the twenty-four hour news cycle has altered people's opinions of one an...   \n",
       "338767  a short film created by children ages 8-10. the premise of the movie is based on 3 kids who try ...   \n",
       "\n",
       "                                                                                                   category  \\\n",
       "0       {\"id\":272,\"name\":\"Mobile Games\",\"slug\":\"games/mobile games\",\"position\":3,\"parent_id\":12,\"parent_...   \n",
       "1       {\"id\":11,\"name\":\"Film & Video\",\"slug\":\"film & video\",\"position\":7,\"color\":16734574,\"urls\":{\"web\"...   \n",
       "2       {\"id\":272,\"name\":\"Mobile Games\",\"slug\":\"games/mobile games\",\"position\":3,\"parent_id\":12,\"parent_...   \n",
       "3       {\"id\":48,\"name\":\"Nonfiction\",\"slug\":\"publishing/nonfiction\",\"position\":9,\"parent_id\":18,\"parent_...   \n",
       "4       {\"id\":39,\"name\":\"Hip-Hop\",\"slug\":\"music/hip-hop\",\"position\":8,\"parent_id\":14,\"parent_name\":\"Musi...   \n",
       "...                                                                                                     ...   \n",
       "338763  {\"id\":10,\"name\":\"Food\",\"slug\":\"food\",\"position\":8,\"color\":16725570,\"urls\":{\"web\":{\"discover\":\"ht...   \n",
       "338764  {\"id\":33,\"name\":\"Webseries\",\"slug\":\"film & video/webseries\",\"position\":19,\"parent_id\":11,\"parent...   \n",
       "338765  {\"id\":285,\"name\":\"Plays\",\"slug\":\"theater/plays\",\"position\":6,\"parent_id\":17,\"parent_name\":\"Theat...   \n",
       "338766  {\"id\":21,\"name\":\"Digital Art\",\"slug\":\"art/digital art\",\"position\":3,\"parent_id\":1,\"parent_name\":...   \n",
       "338767  {\"id\":330,\"name\":\"Family\",\"slug\":\"film & video/family\",\"position\":7,\"parent_id\":11,\"parent_name\"...   \n",
       "\n",
       "       country_displayable_name            deadline     goal  \\\n",
       "0                        Canada 2017-10-09 14:41:03   1000.0   \n",
       "1             the United States 2019-09-21 23:39:48   1100.0   \n",
       "2             the United States 2016-03-22 21:11:17  25000.0   \n",
       "3             the United States 2012-09-25 02:33:06    500.0   \n",
       "4             the United States 2017-12-11 02:43:11   2500.0   \n",
       "...                         ...                 ...      ...   \n",
       "338763        the United States 2019-03-29 22:35:40    550.0   \n",
       "338764        the United States 2012-05-19 04:00:00  30000.0   \n",
       "338765        the United States 2014-09-30 16:00:00    500.0   \n",
       "338766        the United States 2013-06-07 18:39:19  10000.0   \n",
       "338767        the United States 2014-08-06 21:04:08   3500.0   \n",
       "\n",
       "               launched_at  \\\n",
       "0      2017-09-25 14:41:03   \n",
       "1      2019-07-23 23:39:48   \n",
       "2      2016-01-22 22:11:17   \n",
       "3      2012-09-15 02:33:06   \n",
       "4      2017-11-10 02:43:11   \n",
       "...                    ...   \n",
       "338763 2019-02-27 23:35:40   \n",
       "338764 2012-04-15 19:31:32   \n",
       "338765 2014-08-31 14:03:20   \n",
       "338766 2013-05-08 18:39:19   \n",
       "338767 2014-07-07 21:04:08   \n",
       "\n",
       "                                                              name  pledged  \\\n",
       "0       Iron Age Kingdoms - First Free To Win Mobile Strategy Game    105.0   \n",
       "1                                            \"Lizard\" a short film   2358.0   \n",
       "2                                    Total Franchise Football 2016    172.0   \n",
       "3          The Trans-American Psychogeographic Literary Cycle Tour    501.0   \n",
       "4                           The Golden Ticket to the Wonka Factory      0.0   \n",
       "...                                                            ...      ...   \n",
       "338763                                     Bitches Bee Crazy Hives   2145.0   \n",
       "338764                          Lesbian Cops: The Movie - Season 2  30045.5   \n",
       "338765                                Generations (Senior Project)    606.0   \n",
       "338766                            US: A Story of our United States      0.0   \n",
       "338767                      Lights, Camera, and Behind the Action!     25.0   \n",
       "\n",
       "                                                      slug  staff_pick  \\\n",
       "0       iron-age-kingdoms-first-free-to-win-mobile-strateg       False   \n",
       "1                                      lizard-a-short-film       False   \n",
       "2                            total-franchise-football-2016       False   \n",
       "3       the-trans-american-psychogeographic-literary-cycle       False   \n",
       "4                   the-golden-ticket-to-the-wonka-factory       False   \n",
       "...                                                    ...         ...   \n",
       "338763                             bitches-bee-crazy-hives       False   \n",
       "338764                   lesbian-cops-the-movie-season-2-0       False   \n",
       "338765                          generations-senior-project       False   \n",
       "338766                     us-a-story-of-our-united-states       False   \n",
       "338767                 lights-camera-and-behind-the-action       False   \n",
       "\n",
       "             state  \n",
       "0           failed  \n",
       "1       successful  \n",
       "2           failed  \n",
       "3       successful  \n",
       "4           failed  \n",
       "...            ...  \n",
       "338763  successful  \n",
       "338764  successful  \n",
       "338765  successful  \n",
       "338766      failed  \n",
       "338767      failed  \n",
       "\n",
       "[172160 rows x 12 columns]"
      ]
     },
     "execution_count": 19,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "cols_to_remove = ['friends', 'is_backing', 'is_starred', 'permissions', \n",
    "                   'currency_symbol', 'currency_trailing_code', 'spotlight',\n",
    "                  'current_currency', 'currency','usd_type', 'source_url',\n",
    "                  'country', 'disable_communication', 'fx_rate', 'photo',\n",
    "                  'static_usd_rate', 'urls', 'created_at', 'is_starrable',\n",
    "                 'location', 'profile', 'creator', 'id',\n",
    "                 'usd_pledged', 'converted_pledged_amount', 'state_changed_at']\n",
    "df2 = df1.drop(cols_to_remove, axis=1)\n",
    "df2"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "(172160, 12)\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "successful    97906\n",
       "failed        71720\n",
       "live           2534\n",
       "Name: state, dtype: int64"
      ]
     },
     "execution_count": 20,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "print(df2.shape)\n",
    "df2['state'].value_counts()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {},
   "outputs": [],
   "source": [
    "df2['launched_year'] = df2['launched_at'].dt.year\n",
    "df2['launched_month'] = df2['launched_at'].dt.month\n",
    "df2['campaign_length'] = df2['deadline'] - df2['launched_at']\n",
    "df2['campaign_length'] = df2['campaign_length'].dt.round('d').dt.days"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "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>backers_count</th>\n",
       "      <th>blurb</th>\n",
       "      <th>category</th>\n",
       "      <th>country_displayable_name</th>\n",
       "      <th>deadline</th>\n",
       "      <th>goal</th>\n",
       "      <th>launched_at</th>\n",
       "      <th>name</th>\n",
       "      <th>pledged</th>\n",
       "      <th>slug</th>\n",
       "      <th>staff_pick</th>\n",
       "      <th>state</th>\n",
       "      <th>launched_year</th>\n",
       "      <th>launched_month</th>\n",
       "      <th>campaign_length</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>5</td>\n",
       "      <td>A Free to Win Online Mobile Strategy Game to end the corruption of the rich players always winni...</td>\n",
       "      <td>{\"id\":272,\"name\":\"Mobile Games\",\"slug\":\"games/mobile games\",\"position\":3,\"parent_id\":12,\"parent_...</td>\n",
       "      <td>Canada</td>\n",
       "      <td>2017-10-09 14:41:03</td>\n",
       "      <td>1000.0</td>\n",
       "      <td>2017-09-25 14:41:03</td>\n",
       "      <td>Iron Age Kingdoms - First Free To Win Mobile Strategy Game</td>\n",
       "      <td>105.0</td>\n",
       "      <td>iron-age-kingdoms-first-free-to-win-mobile-strateg</td>\n",
       "      <td>False</td>\n",
       "      <td>failed</td>\n",
       "      <td>2017</td>\n",
       "      <td>9</td>\n",
       "      <td>14</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>19</td>\n",
       "      <td>Lizard is a dark comedy and modern western, that takes place in the Mojave Desert.</td>\n",
       "      <td>{\"id\":11,\"name\":\"Film &amp; Video\",\"slug\":\"film &amp; video\",\"position\":7,\"color\":16734574,\"urls\":{\"web\"...</td>\n",
       "      <td>the United States</td>\n",
       "      <td>2019-09-21 23:39:48</td>\n",
       "      <td>1100.0</td>\n",
       "      <td>2019-07-23 23:39:48</td>\n",
       "      <td>\"Lizard\" a short film</td>\n",
       "      <td>2358.0</td>\n",
       "      <td>lizard-a-short-film</td>\n",
       "      <td>False</td>\n",
       "      <td>successful</td>\n",
       "      <td>2019</td>\n",
       "      <td>7</td>\n",
       "      <td>60</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>5</td>\n",
       "      <td>My friend and I are making an app that has never been made before. Making a football turn based ...</td>\n",
       "      <td>{\"id\":272,\"name\":\"Mobile Games\",\"slug\":\"games/mobile games\",\"position\":3,\"parent_id\":12,\"parent_...</td>\n",
       "      <td>the United States</td>\n",
       "      <td>2016-03-22 21:11:17</td>\n",
       "      <td>25000.0</td>\n",
       "      <td>2016-01-22 22:11:17</td>\n",
       "      <td>Total Franchise Football 2016</td>\n",
       "      <td>172.0</td>\n",
       "      <td>total-franchise-football-2016</td>\n",
       "      <td>False</td>\n",
       "      <td>failed</td>\n",
       "      <td>2016</td>\n",
       "      <td>1</td>\n",
       "      <td>60</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>23</td>\n",
       "      <td>More than a bike trip, a state of mind. And a book.</td>\n",
       "      <td>{\"id\":48,\"name\":\"Nonfiction\",\"slug\":\"publishing/nonfiction\",\"position\":9,\"parent_id\":18,\"parent_...</td>\n",
       "      <td>the United States</td>\n",
       "      <td>2012-09-25 02:33:06</td>\n",
       "      <td>500.0</td>\n",
       "      <td>2012-09-15 02:33:06</td>\n",
       "      <td>The Trans-American Psychogeographic Literary Cycle Tour</td>\n",
       "      <td>501.0</td>\n",
       "      <td>the-trans-american-psychogeographic-literary-cycle</td>\n",
       "      <td>False</td>\n",
       "      <td>successful</td>\n",
       "      <td>2012</td>\n",
       "      <td>9</td>\n",
       "      <td>10</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>0</td>\n",
       "      <td>This is the birth of a revolutionary artist...Revery! A burst out into the world with GAME CHANG...</td>\n",
       "      <td>{\"id\":39,\"name\":\"Hip-Hop\",\"slug\":\"music/hip-hop\",\"position\":8,\"parent_id\":14,\"parent_name\":\"Musi...</td>\n",
       "      <td>the United States</td>\n",
       "      <td>2017-12-11 02:43:11</td>\n",
       "      <td>2500.0</td>\n",
       "      <td>2017-11-10 02:43:11</td>\n",
       "      <td>The Golden Ticket to the Wonka Factory</td>\n",
       "      <td>0.0</td>\n",
       "      <td>the-golden-ticket-to-the-wonka-factory</td>\n",
       "      <td>False</td>\n",
       "      <td>failed</td>\n",
       "      <td>2017</td>\n",
       "      <td>11</td>\n",
       "      <td>31</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>338763</th>\n",
       "      <td>48</td>\n",
       "      <td>The bee all end all Kickstarter for crazy bitches. \\nWe're raising funds for bee hives!</td>\n",
       "      <td>{\"id\":10,\"name\":\"Food\",\"slug\":\"food\",\"position\":8,\"color\":16725570,\"urls\":{\"web\":{\"discover\":\"ht...</td>\n",
       "      <td>the United States</td>\n",
       "      <td>2019-03-29 22:35:40</td>\n",
       "      <td>550.0</td>\n",
       "      <td>2019-02-27 23:35:40</td>\n",
       "      <td>Bitches Bee Crazy Hives</td>\n",
       "      <td>2145.0</td>\n",
       "      <td>bitches-bee-crazy-hives</td>\n",
       "      <td>False</td>\n",
       "      <td>successful</td>\n",
       "      <td>2019</td>\n",
       "      <td>2</td>\n",
       "      <td>30</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>338764</th>\n",
       "      <td>179</td>\n",
       "      <td>Detective Tori Jones and her partner Officer Rashida Thompson are back!  Kicking crime's ass and...</td>\n",
       "      <td>{\"id\":33,\"name\":\"Webseries\",\"slug\":\"film &amp; video/webseries\",\"position\":19,\"parent_id\":11,\"parent...</td>\n",
       "      <td>the United States</td>\n",
       "      <td>2012-05-19 04:00:00</td>\n",
       "      <td>30000.0</td>\n",
       "      <td>2012-04-15 19:31:32</td>\n",
       "      <td>Lesbian Cops: The Movie - Season 2</td>\n",
       "      <td>30045.5</td>\n",
       "      <td>lesbian-cops-the-movie-season-2-0</td>\n",
       "      <td>False</td>\n",
       "      <td>successful</td>\n",
       "      <td>2012</td>\n",
       "      <td>4</td>\n",
       "      <td>33</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>338765</th>\n",
       "      <td>18</td>\n",
       "      <td>Help me produce the play I have written for my senior project!</td>\n",
       "      <td>{\"id\":285,\"name\":\"Plays\",\"slug\":\"theater/plays\",\"position\":6,\"parent_id\":17,\"parent_name\":\"Theat...</td>\n",
       "      <td>the United States</td>\n",
       "      <td>2014-09-30 16:00:00</td>\n",
       "      <td>500.0</td>\n",
       "      <td>2014-08-31 14:03:20</td>\n",
       "      <td>Generations (Senior Project)</td>\n",
       "      <td>606.0</td>\n",
       "      <td>generations-senior-project</td>\n",
       "      <td>False</td>\n",
       "      <td>successful</td>\n",
       "      <td>2014</td>\n",
       "      <td>8</td>\n",
       "      <td>30</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>338766</th>\n",
       "      <td>0</td>\n",
       "      <td>This documentary asks if the twenty-four hour news cycle has altered people's opinions of one an...</td>\n",
       "      <td>{\"id\":21,\"name\":\"Digital Art\",\"slug\":\"art/digital art\",\"position\":3,\"parent_id\":1,\"parent_name\":...</td>\n",
       "      <td>the United States</td>\n",
       "      <td>2013-06-07 18:39:19</td>\n",
       "      <td>10000.0</td>\n",
       "      <td>2013-05-08 18:39:19</td>\n",
       "      <td>US: A Story of our United States</td>\n",
       "      <td>0.0</td>\n",
       "      <td>us-a-story-of-our-united-states</td>\n",
       "      <td>False</td>\n",
       "      <td>failed</td>\n",
       "      <td>2013</td>\n",
       "      <td>5</td>\n",
       "      <td>30</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>338767</th>\n",
       "      <td>1</td>\n",
       "      <td>a short film created by children ages 8-10. the premise of the movie is based on 3 kids who try ...</td>\n",
       "      <td>{\"id\":330,\"name\":\"Family\",\"slug\":\"film &amp; video/family\",\"position\":7,\"parent_id\":11,\"parent_name\"...</td>\n",
       "      <td>the United States</td>\n",
       "      <td>2014-08-06 21:04:08</td>\n",
       "      <td>3500.0</td>\n",
       "      <td>2014-07-07 21:04:08</td>\n",
       "      <td>Lights, Camera, and Behind the Action!</td>\n",
       "      <td>25.0</td>\n",
       "      <td>lights-camera-and-behind-the-action</td>\n",
       "      <td>False</td>\n",
       "      <td>failed</td>\n",
       "      <td>2014</td>\n",
       "      <td>7</td>\n",
       "      <td>30</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>172160 rows × 15 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "        backers_count  \\\n",
       "0                   5   \n",
       "1                  19   \n",
       "2                   5   \n",
       "3                  23   \n",
       "4                   0   \n",
       "...               ...   \n",
       "338763             48   \n",
       "338764            179   \n",
       "338765             18   \n",
       "338766              0   \n",
       "338767              1   \n",
       "\n",
       "                                                                                                      blurb  \\\n",
       "0       A Free to Win Online Mobile Strategy Game to end the corruption of the rich players always winni...   \n",
       "1                        Lizard is a dark comedy and modern western, that takes place in the Mojave Desert.   \n",
       "2       My friend and I are making an app that has never been made before. Making a football turn based ...   \n",
       "3                                                       More than a bike trip, a state of mind. And a book.   \n",
       "4       This is the birth of a revolutionary artist...Revery! A burst out into the world with GAME CHANG...   \n",
       "...                                                                                                     ...   \n",
       "338763              The bee all end all Kickstarter for crazy bitches. \\nWe're raising funds for bee hives!   \n",
       "338764  Detective Tori Jones and her partner Officer Rashida Thompson are back!  Kicking crime's ass and...   \n",
       "338765                                       Help me produce the play I have written for my senior project!   \n",
       "338766  This documentary asks if the twenty-four hour news cycle has altered people's opinions of one an...   \n",
       "338767  a short film created by children ages 8-10. the premise of the movie is based on 3 kids who try ...   \n",
       "\n",
       "                                                                                                   category  \\\n",
       "0       {\"id\":272,\"name\":\"Mobile Games\",\"slug\":\"games/mobile games\",\"position\":3,\"parent_id\":12,\"parent_...   \n",
       "1       {\"id\":11,\"name\":\"Film & Video\",\"slug\":\"film & video\",\"position\":7,\"color\":16734574,\"urls\":{\"web\"...   \n",
       "2       {\"id\":272,\"name\":\"Mobile Games\",\"slug\":\"games/mobile games\",\"position\":3,\"parent_id\":12,\"parent_...   \n",
       "3       {\"id\":48,\"name\":\"Nonfiction\",\"slug\":\"publishing/nonfiction\",\"position\":9,\"parent_id\":18,\"parent_...   \n",
       "4       {\"id\":39,\"name\":\"Hip-Hop\",\"slug\":\"music/hip-hop\",\"position\":8,\"parent_id\":14,\"parent_name\":\"Musi...   \n",
       "...                                                                                                     ...   \n",
       "338763  {\"id\":10,\"name\":\"Food\",\"slug\":\"food\",\"position\":8,\"color\":16725570,\"urls\":{\"web\":{\"discover\":\"ht...   \n",
       "338764  {\"id\":33,\"name\":\"Webseries\",\"slug\":\"film & video/webseries\",\"position\":19,\"parent_id\":11,\"parent...   \n",
       "338765  {\"id\":285,\"name\":\"Plays\",\"slug\":\"theater/plays\",\"position\":6,\"parent_id\":17,\"parent_name\":\"Theat...   \n",
       "338766  {\"id\":21,\"name\":\"Digital Art\",\"slug\":\"art/digital art\",\"position\":3,\"parent_id\":1,\"parent_name\":...   \n",
       "338767  {\"id\":330,\"name\":\"Family\",\"slug\":\"film & video/family\",\"position\":7,\"parent_id\":11,\"parent_name\"...   \n",
       "\n",
       "       country_displayable_name            deadline     goal  \\\n",
       "0                        Canada 2017-10-09 14:41:03   1000.0   \n",
       "1             the United States 2019-09-21 23:39:48   1100.0   \n",
       "2             the United States 2016-03-22 21:11:17  25000.0   \n",
       "3             the United States 2012-09-25 02:33:06    500.0   \n",
       "4             the United States 2017-12-11 02:43:11   2500.0   \n",
       "...                         ...                 ...      ...   \n",
       "338763        the United States 2019-03-29 22:35:40    550.0   \n",
       "338764        the United States 2012-05-19 04:00:00  30000.0   \n",
       "338765        the United States 2014-09-30 16:00:00    500.0   \n",
       "338766        the United States 2013-06-07 18:39:19  10000.0   \n",
       "338767        the United States 2014-08-06 21:04:08   3500.0   \n",
       "\n",
       "               launched_at  \\\n",
       "0      2017-09-25 14:41:03   \n",
       "1      2019-07-23 23:39:48   \n",
       "2      2016-01-22 22:11:17   \n",
       "3      2012-09-15 02:33:06   \n",
       "4      2017-11-10 02:43:11   \n",
       "...                    ...   \n",
       "338763 2019-02-27 23:35:40   \n",
       "338764 2012-04-15 19:31:32   \n",
       "338765 2014-08-31 14:03:20   \n",
       "338766 2013-05-08 18:39:19   \n",
       "338767 2014-07-07 21:04:08   \n",
       "\n",
       "                                                              name  pledged  \\\n",
       "0       Iron Age Kingdoms - First Free To Win Mobile Strategy Game    105.0   \n",
       "1                                            \"Lizard\" a short film   2358.0   \n",
       "2                                    Total Franchise Football 2016    172.0   \n",
       "3          The Trans-American Psychogeographic Literary Cycle Tour    501.0   \n",
       "4                           The Golden Ticket to the Wonka Factory      0.0   \n",
       "...                                                            ...      ...   \n",
       "338763                                     Bitches Bee Crazy Hives   2145.0   \n",
       "338764                          Lesbian Cops: The Movie - Season 2  30045.5   \n",
       "338765                                Generations (Senior Project)    606.0   \n",
       "338766                            US: A Story of our United States      0.0   \n",
       "338767                      Lights, Camera, and Behind the Action!     25.0   \n",
       "\n",
       "                                                      slug  staff_pick  \\\n",
       "0       iron-age-kingdoms-first-free-to-win-mobile-strateg       False   \n",
       "1                                      lizard-a-short-film       False   \n",
       "2                            total-franchise-football-2016       False   \n",
       "3       the-trans-american-psychogeographic-literary-cycle       False   \n",
       "4                   the-golden-ticket-to-the-wonka-factory       False   \n",
       "...                                                    ...         ...   \n",
       "338763                             bitches-bee-crazy-hives       False   \n",
       "338764                   lesbian-cops-the-movie-season-2-0       False   \n",
       "338765                          generations-senior-project       False   \n",
       "338766                     us-a-story-of-our-united-states       False   \n",
       "338767                 lights-camera-and-behind-the-action       False   \n",
       "\n",
       "             state  launched_year  launched_month  campaign_length  \n",
       "0           failed           2017               9               14  \n",
       "1       successful           2019               7               60  \n",
       "2           failed           2016               1               60  \n",
       "3       successful           2012               9               10  \n",
       "4           failed           2017              11               31  \n",
       "...            ...            ...             ...              ...  \n",
       "338763  successful           2019               2               30  \n",
       "338764  successful           2012               4               33  \n",
       "338765  successful           2014               8               30  \n",
       "338766      failed           2013               5               30  \n",
       "338767      failed           2014               7               30  \n",
       "\n",
       "[172160 rows x 15 columns]"
      ]
     },
     "execution_count": 22,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df2"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## EDA"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "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>backers_count</th>\n",
       "      <th>goal</th>\n",
       "      <th>pledged</th>\n",
       "      <th>launched_year</th>\n",
       "      <th>launched_month</th>\n",
       "      <th>campaign_length</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>count</th>\n",
       "      <td>172160.000000</td>\n",
       "      <td>1.721600e+05</td>\n",
       "      <td>1.721600e+05</td>\n",
       "      <td>172160.000000</td>\n",
       "      <td>172160.000000</td>\n",
       "      <td>172160.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>mean</th>\n",
       "      <td>146.600674</td>\n",
       "      <td>5.144890e+04</td>\n",
       "      <td>2.174109e+04</td>\n",
       "      <td>2016.126208</td>\n",
       "      <td>6.321068</td>\n",
       "      <td>32.935792</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>std</th>\n",
       "      <td>922.705482</td>\n",
       "      <td>1.208034e+06</td>\n",
       "      <td>7.633733e+05</td>\n",
       "      <td>2.271547</td>\n",
       "      <td>3.372272</td>\n",
       "      <td>12.109187</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>min</th>\n",
       "      <td>0.000000</td>\n",
       "      <td>1.000000e-02</td>\n",
       "      <td>0.000000e+00</td>\n",
       "      <td>2009.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>25%</th>\n",
       "      <td>4.000000</td>\n",
       "      <td>1.500000e+03</td>\n",
       "      <td>1.100000e+02</td>\n",
       "      <td>2015.000000</td>\n",
       "      <td>3.000000</td>\n",
       "      <td>30.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>50%</th>\n",
       "      <td>27.000000</td>\n",
       "      <td>5.000000e+03</td>\n",
       "      <td>1.554000e+03</td>\n",
       "      <td>2016.000000</td>\n",
       "      <td>6.000000</td>\n",
       "      <td>30.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>75%</th>\n",
       "      <td>90.000000</td>\n",
       "      <td>1.500000e+04</td>\n",
       "      <td>7.000000e+03</td>\n",
       "      <td>2018.000000</td>\n",
       "      <td>9.000000</td>\n",
       "      <td>35.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>max</th>\n",
       "      <td>105857.000000</td>\n",
       "      <td>1.000000e+08</td>\n",
       "      <td>2.353205e+08</td>\n",
       "      <td>2020.000000</td>\n",
       "      <td>12.000000</td>\n",
       "      <td>98.000000</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "       backers_count          goal       pledged  launched_year  \\\n",
       "count  172160.000000  1.721600e+05  1.721600e+05  172160.000000   \n",
       "mean      146.600674  5.144890e+04  2.174109e+04    2016.126208   \n",
       "std       922.705482  1.208034e+06  7.633733e+05       2.271547   \n",
       "min         0.000000  1.000000e-02  0.000000e+00    2009.000000   \n",
       "25%         4.000000  1.500000e+03  1.100000e+02    2015.000000   \n",
       "50%        27.000000  5.000000e+03  1.554000e+03    2016.000000   \n",
       "75%        90.000000  1.500000e+04  7.000000e+03    2018.000000   \n",
       "max    105857.000000  1.000000e+08  2.353205e+08    2020.000000   \n",
       "\n",
       "       launched_month  campaign_length  \n",
       "count   172160.000000    172160.000000  \n",
       "mean         6.321068        32.935792  \n",
       "std          3.372272        12.109187  \n",
       "min          1.000000         1.000000  \n",
       "25%          3.000000        30.000000  \n",
       "50%          6.000000        30.000000  \n",
       "75%          9.000000        35.000000  \n",
       "max         12.000000        98.000000  "
      ]
     },
     "execution_count": 23,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df2.describe()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "The longest Kickstarter campaign length is: 98\n",
      "The shortest Kickstarter campaign length is: 1\n"
     ]
    }
   ],
   "source": [
    "max_days = df2.campaign_length.value_counts().index.max()\n",
    "min_days = df2.campaign_length.value_counts().index.min()\n",
    "print('The longest Kickstarter campaign length is: {}'.format(max_days))\n",
    "print('The shortest Kickstarter campaign length is: {}'.format(min_days))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "There are 57600 Kickstarter projects with longer than 30 days length\n",
      "There are 114560 Kickstarter projects with less than 30 days length\n"
     ]
    }
   ],
   "source": [
    "more_than_30 = df2[df2['campaign_length']>30].shape[0]\n",
    "less_than_30 = df2[df2['campaign_length']<=30].shape[0]\n",
    "print('There are {} Kickstarter projects with longer than 30 days length'.format(more_than_30))\n",
    "print('There are {} Kickstarter projects with less than 30 days length'.format(less_than_30))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "There are 35 Kickstarter projects with longer than 90 days length\n",
      "There are 172125 Kickstarter projects with less than 90 days length\n"
     ]
    }
   ],
   "source": [
    "more_than_90 = df2[df2['campaign_length']>90].shape[0]\n",
    "less_than_90 = df2[df2['campaign_length']<=90].shape[0]\n",
    "print('There are {} Kickstarter projects with longer than 90 days length'.format(more_than_90))\n",
    "print('There are {} Kickstarter projects with less than 90 days length'.format(less_than_90))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(172125, 15)"
      ]
     },
     "execution_count": 27,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# remove entries that are longer than 90 days in length\n",
    "df2 = df2[df2['campaign_length']<=90]\n",
    "df2.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "successful    97880\n",
       "failed        71711\n",
       "live           2534\n",
       "Name: state, dtype: int64"
      ]
     },
     "execution_count": 28,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df2['state'].value_counts()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "2015    30521\n",
       "2019    24933\n",
       "2016    23821\n",
       "2017    23117\n",
       "2014    21288\n",
       "2018    20549\n",
       "2013     8684\n",
       "2020     8135\n",
       "2012     6418\n",
       "2011     3305\n",
       "2010     1166\n",
       "2009      188\n",
       "Name: launched_year, dtype: int64"
      ]
     },
     "execution_count": 29,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df2['launched_year'].value_counts()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1152x432 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.figure(figsize=(16,6))\n",
    "df.set_index('launched_at').category.resample('MS').count().plot()\n",
    "plt.xlim('2009-01-01', '2020-05-01') # Limiting to whole months\n",
    "plt.xlabel('Launch date', fontsize=16)\n",
    "plt.ylabel('Number of projects', fontsize=16)\n",
    "plt.title('Number of projects launched on Kickstarter, 2009-2020', fontsize=18)\n",
    "plt.show()\n",
    "# plt.savefig('monthly_recap')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "#df1['country_displayable_name'].value_counts()\n",
    "fig, ax = plt.subplots()\n",
    "\n",
    "pie_data = df2['country_displayable_name'].value_counts()[:5]\n",
    "colors = ['lightcoral', 'orange', 'green', 'lightskyblue', 'red']\n",
    "explode = (0.05,0,0,0,0)\n",
    "ax.pie(pie_data.values, explode=explode, labels=pie_data.index, \n",
    "       colors=colors, autopct='%1.1f%%', shadow=True)\n",
    "ax.set_title('Distribution of Top 5 Countries')\n",
    "\n",
    "plt.axis('equal')\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 33,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "'{\"id\":272,\"name\":\"Mobile Games\",\"slug\":\"games/mobile games\",\"position\":3,\"parent_id\":12,\"parent_name\":\"Games\",\"color\":51627,\"urls\":{\"web\":{\"discover\":\"http://www.kickstarter.com/discover/categories/games/mobile%20games\"}}}'"
      ]
     },
     "execution_count": 33,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df2['category'][0]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 34,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "'{\"id\":11,\"name\":\"Film & Video\",\"slug\":\"film & video\",\"position\":7,\"color\":16734574,\"urls\":{\"web\":{\"discover\":\"http://www.kickstarter.com/discover/categories/film%20&%20video\"}}}'"
      ]
     },
     "execution_count": 34,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df2['category'][1]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 35,
   "metadata": {},
   "outputs": [],
   "source": [
    "# To extract category type from category column\n",
    "def extract(cat):\n",
    "    x = ast.literal_eval(re.search('({.+})', cat).group(0))\n",
    "    if 'parent_name' not in x.keys():\n",
    "        return x['slug']\n",
    "    else:\n",
    "        return x['parent_name'].lower()\n",
    "   "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 36,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "'film & video'"
      ]
     },
     "execution_count": 36,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "extract(df2['category'][1])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 37,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/opt/conda/lib/python3.7/site-packages/ipykernel_launcher.py:1: SettingWithCopyWarning: \n",
      "A value is trying to be set on a copy of a slice from a DataFrame.\n",
      "Try using .loc[row_indexer,col_indexer] = value instead\n",
      "\n",
      "See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy\n",
      "  \"\"\"Entry point for launching an IPython kernel.\n"
     ]
    }
   ],
   "source": [
    "df2['category_type'] = df2['category'].map(lambda x: extract(x))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 38,
   "metadata": {},
   "outputs": [],
   "source": [
    "df2 = df2[df2['state'].isin(['successful', 'failed'])]\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 39,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Text(0.5, 1.0, 'Category Type Count')"
      ]
     },
     "execution_count": 39,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 576x432 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "fig, ax = plt.subplots(figsize=(8,6))\n",
    "x = df2.category_type.value_counts()\n",
    "ax.bar(x.index, height=x.values)\n",
    "ax.set_xticklabels(x.index, rotation=90)\n",
    "ax.set_ylabel('Count')\n",
    "ax.set_xlabel('Category Type')\n",
    "ax.set_title('Category Type Count', size=16)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 40,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "film & video    25067\n",
       "music           21122\n",
       "technology      18524\n",
       "publishing      18022\n",
       "art             16811\n",
       "food            14216\n",
       "games           11075\n",
       "fashion          8666\n",
       "comics           6816\n",
       "design           6234\n",
       "crafts           5882\n",
       "theater          5141\n",
       "photography      4992\n",
       "journalism       3950\n",
       "dance            3073\n",
       "Name: category_type, dtype: int64"
      ]
     },
     "execution_count": 40,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df2.category_type.value_counts()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 41,
   "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></th>\n",
       "      <th>pledged</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>launched_month</th>\n",
       "      <th>state</th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th rowspan=\"2\" valign=\"top\">1</th>\n",
       "      <th>failed</th>\n",
       "      <td>1476.448930</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>successful</th>\n",
       "      <td>31577.697567</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th rowspan=\"2\" valign=\"top\">2</th>\n",
       "      <th>failed</th>\n",
       "      <td>2157.471983</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>successful</th>\n",
       "      <td>55499.218732</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th rowspan=\"2\" valign=\"top\">3</th>\n",
       "      <th>failed</th>\n",
       "      <td>2012.464565</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>successful</th>\n",
       "      <td>39138.135382</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th rowspan=\"2\" valign=\"top\">4</th>\n",
       "      <th>failed</th>\n",
       "      <td>2122.505285</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>successful</th>\n",
       "      <td>24897.800510</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th rowspan=\"2\" valign=\"top\">5</th>\n",
       "      <th>failed</th>\n",
       "      <td>2241.477163</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>successful</th>\n",
       "      <td>33573.515058</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th rowspan=\"2\" valign=\"top\">6</th>\n",
       "      <th>failed</th>\n",
       "      <td>1630.195719</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>successful</th>\n",
       "      <td>36670.002050</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th rowspan=\"2\" valign=\"top\">7</th>\n",
       "      <th>failed</th>\n",
       "      <td>1474.726397</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>successful</th>\n",
       "      <td>28447.707852</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th rowspan=\"2\" valign=\"top\">8</th>\n",
       "      <th>failed</th>\n",
       "      <td>1359.479155</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>successful</th>\n",
       "      <td>26616.067393</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th rowspan=\"2\" valign=\"top\">9</th>\n",
       "      <th>failed</th>\n",
       "      <td>1400.694982</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>successful</th>\n",
       "      <td>35100.851661</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th rowspan=\"2\" valign=\"top\">10</th>\n",
       "      <th>failed</th>\n",
       "      <td>2981.827108</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>successful</th>\n",
       "      <td>49943.560606</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th rowspan=\"2\" valign=\"top\">11</th>\n",
       "      <th>failed</th>\n",
       "      <td>2102.879505</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>successful</th>\n",
       "      <td>31990.368814</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th rowspan=\"2\" valign=\"top\">12</th>\n",
       "      <th>failed</th>\n",
       "      <td>1405.164850</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>successful</th>\n",
       "      <td>35838.808749</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                                pledged\n",
       "launched_month state                   \n",
       "1              failed       1476.448930\n",
       "               successful  31577.697567\n",
       "2              failed       2157.471983\n",
       "               successful  55499.218732\n",
       "3              failed       2012.464565\n",
       "               successful  39138.135382\n",
       "4              failed       2122.505285\n",
       "               successful  24897.800510\n",
       "5              failed       2241.477163\n",
       "               successful  33573.515058\n",
       "6              failed       1630.195719\n",
       "               successful  36670.002050\n",
       "7              failed       1474.726397\n",
       "               successful  28447.707852\n",
       "8              failed       1359.479155\n",
       "               successful  26616.067393\n",
       "9              failed       1400.694982\n",
       "               successful  35100.851661\n",
       "10             failed       2981.827108\n",
       "               successful  49943.560606\n",
       "11             failed       2102.879505\n",
       "               successful  31990.368814\n",
       "12             failed       1405.164850\n",
       "               successful  35838.808749"
      ]
     },
     "execution_count": 41,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "pledged_by_month = df2[['launched_month', 'state', 'pledged']].groupby(['launched_month', 'state']).mean()\n",
    "pledged_by_month"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 42,
   "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>pledged</th>\n",
       "      <th>backers_count</th>\n",
       "      <th>avg</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>category_type</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>art</th>\n",
       "      <td>9.690949e+07</td>\n",
       "      <td>1004471</td>\n",
       "      <td>96.478141</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>comics</th>\n",
       "      <td>9.015795e+07</td>\n",
       "      <td>1357605</td>\n",
       "      <td>66.409558</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>crafts</th>\n",
       "      <td>3.500879e+07</td>\n",
       "      <td>242521</td>\n",
       "      <td>144.353637</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>dance</th>\n",
       "      <td>1.445048e+07</td>\n",
       "      <td>160403</td>\n",
       "      <td>90.088565</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>design</th>\n",
       "      <td>5.157686e+08</td>\n",
       "      <td>2004470</td>\n",
       "      <td>257.309213</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>fashion</th>\n",
       "      <td>1.523490e+08</td>\n",
       "      <td>1098838</td>\n",
       "      <td>138.645591</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>film &amp; video</th>\n",
       "      <td>4.107300e+08</td>\n",
       "      <td>2828913</td>\n",
       "      <td>145.190046</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>food</th>\n",
       "      <td>1.133548e+08</td>\n",
       "      <td>932256</td>\n",
       "      <td>121.591913</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>games</th>\n",
       "      <td>8.759638e+08</td>\n",
       "      <td>5496631</td>\n",
       "      <td>159.363769</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>journalism</th>\n",
       "      <td>2.367019e+07</td>\n",
       "      <td>240949</td>\n",
       "      <td>98.237326</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>music</th>\n",
       "      <td>1.423550e+08</td>\n",
       "      <td>1521355</td>\n",
       "      <td>93.571205</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>photography</th>\n",
       "      <td>3.756104e+07</td>\n",
       "      <td>319097</td>\n",
       "      <td>117.710405</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>publishing</th>\n",
       "      <td>1.555082e+08</td>\n",
       "      <td>2158082</td>\n",
       "      <td>72.058512</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>technology</th>\n",
       "      <td>9.802297e+08</td>\n",
       "      <td>5153994</td>\n",
       "      <td>190.188367</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>theater</th>\n",
       "      <td>2.727492e+07</td>\n",
       "      <td>265604</td>\n",
       "      <td>102.690152</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                    pledged  backers_count         avg\n",
       "category_type                                         \n",
       "art            9.690949e+07        1004471   96.478141\n",
       "comics         9.015795e+07        1357605   66.409558\n",
       "crafts         3.500879e+07         242521  144.353637\n",
       "dance          1.445048e+07         160403   90.088565\n",
       "design         5.157686e+08        2004470  257.309213\n",
       "fashion        1.523490e+08        1098838  138.645591\n",
       "film & video   4.107300e+08        2828913  145.190046\n",
       "food           1.133548e+08         932256  121.591913\n",
       "games          8.759638e+08        5496631  159.363769\n",
       "journalism     2.367019e+07         240949   98.237326\n",
       "music          1.423550e+08        1521355   93.571205\n",
       "photography    3.756104e+07         319097  117.710405\n",
       "publishing     1.555082e+08        2158082   72.058512\n",
       "technology     9.802297e+08        5153994  190.188367\n",
       "theater        2.727492e+07         265604  102.690152"
      ]
     },
     "execution_count": 42,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "total_pledged = df2[['category_type', 'state', 'pledged', 'backers_count']].groupby(['category_type']).sum()\n",
    "total_pledged['avg'] = total_pledged['pledged']/total_pledged['backers_count']\n",
    "total_pledged"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 43,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "148.12443094019318\n"
     ]
    }
   ],
   "source": [
    "a = sum(df2.pledged)\n",
    "b = sum(df2.backers_count)\n",
    "print(a/b)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": 44,
   "metadata": {},
   "outputs": [],
   "source": [
    "def word_count(blurb):\n",
    "    count = 0\n",
    "    if type(blurb) == str:\n",
    "        count = len(blurb.split())\n",
    "    else:\n",
    "        count = 0\n",
    "    return count"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 45,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/opt/conda/lib/python3.7/site-packages/ipykernel_launcher.py:1: SettingWithCopyWarning: \n",
      "A value is trying to be set on a copy of a slice from a DataFrame.\n",
      "Try using .loc[row_indexer,col_indexer] = value instead\n",
      "\n",
      "See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy\n",
      "  \"\"\"Entry point for launching an IPython kernel.\n"
     ]
    }
   ],
   "source": [
    "df2['blurb_length'] = df2['blurb'].map(lambda x: word_count(x))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 46,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Index(['the United States', 'the United Kingdom', 'Canada', 'Australia',\n",
       "       'Germany', 'France', 'Mexico', 'Italy', 'Spain', 'the Netherlands'],\n",
       "      dtype='object')"
      ]
     },
     "execution_count": 46,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df2['country_displayable_name'].value_counts().index[:10]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 47,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "the United States     119325\n",
       "the United Kingdom     18833\n",
       "Canada                  7861\n",
       "Australia               3964\n",
       "Germany                 2909\n",
       "France                  2312\n",
       "Mexico                  2195\n",
       "Italy                   2002\n",
       "Spain                   1737\n",
       "the Netherlands         1447\n",
       "Sweden                  1173\n",
       "Hong Kong                955\n",
       "New Zealand              773\n",
       "Denmark                  728\n",
       "Singapore                585\n",
       "Switzerland              565\n",
       "Ireland                  524\n",
       "Belgium                  473\n",
       "Austria                  405\n",
       "Norway                   394\n",
       "Japan                    379\n",
       "Luxembourg                52\n",
       "Name: country_displayable_name, dtype: int64"
      ]
     },
     "execution_count": 47,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df2['country_displayable_name'].value_counts()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 48,
   "metadata": {},
   "outputs": [],
   "source": [
    "def country_cat(x):\n",
    "    country_list = ['the United States', 'the United Kingdom ', 'Canada', 'Australia',\n",
    "                   'Germany', 'France', 'Mexico', 'Italy', 'Spain', 'the Netherlands']\n",
    "    for country in x:\n",
    "        if country in country_list:\n",
    "            return country\n",
    "        else:\n",
    "            return 'other'"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 49,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/opt/conda/lib/python3.7/site-packages/ipykernel_launcher.py:1: SettingWithCopyWarning: \n",
      "A value is trying to be set on a copy of a slice from a DataFrame.\n",
      "Try using .loc[row_indexer,col_indexer] = value instead\n",
      "\n",
      "See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy\n",
      "  \"\"\"Entry point for launching an IPython kernel.\n"
     ]
    }
   ],
   "source": [
    "df2['country_name'] = df['country_displayable_name'].map(lambda x: country_cat([x]))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 50,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(169591, 18)"
      ]
     },
     "execution_count": 50,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df2.shape"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## One Hot Encode!\n",
    "\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 51,
   "metadata": {},
   "outputs": [],
   "source": [
    "# one hot encoding!!\n",
    "df2_ohe = pd.get_dummies(df2[['category_type', 'country_name', 'staff_pick']])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 52,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Index(['backers_count', 'blurb', 'category', 'country_displayable_name',\n",
       "       'deadline', 'goal', 'launched_at', 'name', 'pledged', 'slug',\n",
       "       'staff_pick', 'state', 'launched_year', 'launched_month',\n",
       "       'campaign_length', 'category_type', 'blurb_length', 'country_name'],\n",
       "      dtype='object')"
      ]
     },
     "execution_count": 52,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df2.columns"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 53,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(169591, 44)"
      ]
     },
     "execution_count": 53,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data = pd.merge(df2, df2_ohe, left_index=True, right_index=True)\n",
    "data.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 54,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "(169591, 44)\n",
      "(0, 44)\n"
     ]
    }
   ],
   "source": [
    "good_data = data[data['state'].isin(['successful', 'failed'])]\n",
    "live = data[data['state'].isin(['live'])]\n",
    "print(good_data.shape)\n",
    "print(live.shape)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 55,
   "metadata": {},
   "outputs": [],
   "source": [
    "good_data['state'] = good_data['state'].replace(to_replace=['successful', 'failed'], value=[1,0])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 56,
   "metadata": {},
   "outputs": [],
   "source": [
    "model_data = good_data.copy()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 57,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Index(['backers_count', 'blurb', 'category', 'country_displayable_name',\n",
       "       'deadline', 'goal', 'launched_at', 'name', 'pledged', 'slug',\n",
       "       'staff_pick_x', 'state', 'launched_year', 'launched_month',\n",
       "       'campaign_length', 'category_type', 'blurb_length', 'country_name',\n",
       "       'staff_pick_y', 'category_type_art', 'category_type_comics',\n",
       "       'category_type_crafts', 'category_type_dance', 'category_type_design',\n",
       "       'category_type_fashion', 'category_type_film & video',\n",
       "       'category_type_food', 'category_type_games', 'category_type_journalism',\n",
       "       'category_type_music', 'category_type_photography',\n",
       "       'category_type_publishing', 'category_type_technology',\n",
       "       'category_type_theater', 'country_name_Australia',\n",
       "       'country_name_Canada', 'country_name_France', 'country_name_Germany',\n",
       "       'country_name_Italy', 'country_name_Mexico', 'country_name_Spain',\n",
       "       'country_name_other', 'country_name_the Netherlands',\n",
       "       'country_name_the United States'],\n",
       "      dtype='object')"
      ]
     },
     "execution_count": 57,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "model_data.columns"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 58,
   "metadata": {},
   "outputs": [],
   "source": [
    "X = model_data.drop(['backers_count','blurb', 'category', 'country_displayable_name', 'deadline', \n",
    "                     'launched_at', 'name', 'staff_pick_y','pledged', 'slug', 'state', 'category_type', 'country_name'],\n",
    "                   axis=1)\n",
    "y = model_data['state']\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 59,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(169591, 31)"
      ]
     },
     "execution_count": 59,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "X.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 60,
   "metadata": {},
   "outputs": [],
   "source": [
    "X = X*1"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Modeling"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 61,
   "metadata": {},
   "outputs": [],
   "source": [
    "X_train, X_test, y_train, y_test = train_test_split(X, y)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 62,
   "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>goal</th>\n",
       "      <th>staff_pick_x</th>\n",
       "      <th>launched_year</th>\n",
       "      <th>launched_month</th>\n",
       "      <th>campaign_length</th>\n",
       "      <th>blurb_length</th>\n",
       "      <th>category_type_art</th>\n",
       "      <th>category_type_comics</th>\n",
       "      <th>category_type_crafts</th>\n",
       "      <th>category_type_dance</th>\n",
       "      <th>category_type_design</th>\n",
       "      <th>category_type_fashion</th>\n",
       "      <th>category_type_film &amp; video</th>\n",
       "      <th>category_type_food</th>\n",
       "      <th>category_type_games</th>\n",
       "      <th>category_type_journalism</th>\n",
       "      <th>category_type_music</th>\n",
       "      <th>category_type_photography</th>\n",
       "      <th>category_type_publishing</th>\n",
       "      <th>category_type_technology</th>\n",
       "      <th>category_type_theater</th>\n",
       "      <th>country_name_Australia</th>\n",
       "      <th>country_name_Canada</th>\n",
       "      <th>country_name_France</th>\n",
       "      <th>country_name_Germany</th>\n",
       "      <th>country_name_Italy</th>\n",
       "      <th>country_name_Mexico</th>\n",
       "      <th>country_name_Spain</th>\n",
       "      <th>country_name_other</th>\n",
       "      <th>country_name_the Netherlands</th>\n",
       "      <th>country_name_the United States</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>45158</th>\n",
       "      <td>1500.0</td>\n",
       "      <td>0</td>\n",
       "      <td>2017</td>\n",
       "      <td>11</td>\n",
       "      <td>30</td>\n",
       "      <td>7</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>293978</th>\n",
       "      <td>3000.0</td>\n",
       "      <td>0</td>\n",
       "      <td>2016</td>\n",
       "      <td>11</td>\n",
       "      <td>30</td>\n",
       "      <td>20</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6918</th>\n",
       "      <td>1000.0</td>\n",
       "      <td>0</td>\n",
       "      <td>2013</td>\n",
       "      <td>11</td>\n",
       "      <td>30</td>\n",
       "      <td>22</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>43504</th>\n",
       "      <td>1800.0</td>\n",
       "      <td>0</td>\n",
       "      <td>2014</td>\n",
       "      <td>12</td>\n",
       "      <td>55</td>\n",
       "      <td>23</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>300413</th>\n",
       "      <td>500.0</td>\n",
       "      <td>0</td>\n",
       "      <td>2011</td>\n",
       "      <td>5</td>\n",
       "      <td>20</td>\n",
       "      <td>21</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>34793</th>\n",
       "      <td>200.0</td>\n",
       "      <td>1</td>\n",
       "      <td>2014</td>\n",
       "      <td>6</td>\n",
       "      <td>23</td>\n",
       "      <td>15</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>293345</th>\n",
       "      <td>7500.0</td>\n",
       "      <td>0</td>\n",
       "      <td>2013</td>\n",
       "      <td>7</td>\n",
       "      <td>30</td>\n",
       "      <td>21</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>27761</th>\n",
       "      <td>5000.0</td>\n",
       "      <td>0</td>\n",
       "      <td>2015</td>\n",
       "      <td>4</td>\n",
       "      <td>30</td>\n",
       "      <td>20</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>120414</th>\n",
       "      <td>350.0</td>\n",
       "      <td>0</td>\n",
       "      <td>2019</td>\n",
       "      <td>6</td>\n",
       "      <td>30</td>\n",
       "      <td>13</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>304007</th>\n",
       "      <td>10000.0</td>\n",
       "      <td>0</td>\n",
       "      <td>2014</td>\n",
       "      <td>9</td>\n",
       "      <td>47</td>\n",
       "      <td>16</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>127193 rows × 31 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "           goal  staff_pick_x  launched_year  launched_month  campaign_length  \\\n",
       "45158    1500.0             0           2017              11               30   \n",
       "293978   3000.0             0           2016              11               30   \n",
       "6918     1000.0             0           2013              11               30   \n",
       "43504    1800.0             0           2014              12               55   \n",
       "300413    500.0             0           2011               5               20   \n",
       "...         ...           ...            ...             ...              ...   \n",
       "34793     200.0             1           2014               6               23   \n",
       "293345   7500.0             0           2013               7               30   \n",
       "27761    5000.0             0           2015               4               30   \n",
       "120414    350.0             0           2019               6               30   \n",
       "304007  10000.0             0           2014               9               47   \n",
       "\n",
       "        blurb_length  category_type_art  category_type_comics  \\\n",
       "45158              7                  0                     0   \n",
       "293978            20                  0                     0   \n",
       "6918              22                  0                     0   \n",
       "43504             23                  0                     0   \n",
       "300413            21                  0                     0   \n",
       "...              ...                ...                   ...   \n",
       "34793             15                  0                     0   \n",
       "293345            21                  0                     0   \n",
       "27761             20                  0                     0   \n",
       "120414            13                  0                     0   \n",
       "304007            16                  0                     0   \n",
       "\n",
       "        category_type_crafts  category_type_dance  category_type_design  \\\n",
       "45158                      0                    0                     0   \n",
       "293978                     0                    0                     0   \n",
       "6918                       0                    0                     0   \n",
       "43504                      0                    0                     0   \n",
       "300413                     0                    0                     0   \n",
       "...                      ...                  ...                   ...   \n",
       "34793                      0                    0                     0   \n",
       "293345                     0                    0                     0   \n",
       "27761                      0                    0                     0   \n",
       "120414                     0                    0                     0   \n",
       "304007                     0                    0                     0   \n",
       "\n",
       "        category_type_fashion  category_type_film & video  category_type_food  \\\n",
       "45158                       0                           0                   0   \n",
       "293978                      0                           1                   0   \n",
       "6918                        0                           0                   0   \n",
       "43504                       0                           1                   0   \n",
       "300413                      0                           0                   0   \n",
       "...                       ...                         ...                 ...   \n",
       "34793                       0                           0                   0   \n",
       "293345                      0                           0                   0   \n",
       "27761                       0                           0                   1   \n",
       "120414                      1                           0                   0   \n",
       "304007                      0                           0                   0   \n",
       "\n",
       "        category_type_games  category_type_journalism  category_type_music  \\\n",
       "45158                     0                         0                    1   \n",
       "293978                    0                         0                    0   \n",
       "6918                      0                         0                    0   \n",
       "43504                     0                         0                    0   \n",
       "300413                    0                         0                    1   \n",
       "...                     ...                       ...                  ...   \n",
       "34793                     0                         0                    0   \n",
       "293345                    0                         0                    1   \n",
       "27761                     0                         0                    0   \n",
       "120414                    0                         0                    0   \n",
       "304007                    0                         0                    0   \n",
       "\n",
       "        category_type_photography  category_type_publishing  \\\n",
       "45158                           0                         0   \n",
       "293978                          0                         0   \n",
       "6918                            0                         1   \n",
       "43504                           0                         0   \n",
       "300413                          0                         0   \n",
       "...                           ...                       ...   \n",
       "34793                           0                         0   \n",
       "293345                          0                         0   \n",
       "27761                           0                         0   \n",
       "120414                          0                         0   \n",
       "304007                          0                         0   \n",
       "\n",
       "        category_type_technology  category_type_theater  \\\n",
       "45158                          0                      0   \n",
       "293978                         0                      0   \n",
       "6918                           0                      0   \n",
       "43504                          0                      0   \n",
       "300413                         0                      0   \n",
       "...                          ...                    ...   \n",
       "34793                          1                      0   \n",
       "293345                         0                      0   \n",
       "27761                          0                      0   \n",
       "120414                         0                      0   \n",
       "304007                         1                      0   \n",
       "\n",
       "        country_name_Australia  country_name_Canada  country_name_France  \\\n",
       "45158                        0                    0                    0   \n",
       "293978                       0                    0                    0   \n",
       "6918                         0                    0                    0   \n",
       "43504                        0                    0                    0   \n",
       "300413                       0                    0                    0   \n",
       "...                        ...                  ...                  ...   \n",
       "34793                        0                    0                    0   \n",
       "293345                       0                    0                    0   \n",
       "27761                        0                    0                    0   \n",
       "120414                       0                    0                    0   \n",
       "304007                       0                    0                    0   \n",
       "\n",
       "        country_name_Germany  country_name_Italy  country_name_Mexico  \\\n",
       "45158                      1                   0                    0   \n",
       "293978                     0                   0                    0   \n",
       "6918                       0                   0                    0   \n",
       "43504                      0                   0                    0   \n",
       "300413                     0                   0                    0   \n",
       "...                      ...                 ...                  ...   \n",
       "34793                      0                   0                    0   \n",
       "293345                     0                   0                    0   \n",
       "27761                      0                   0                    0   \n",
       "120414                     0                   0                    0   \n",
       "304007                     0                   0                    0   \n",
       "\n",
       "        country_name_Spain  country_name_other  country_name_the Netherlands  \\\n",
       "45158                    0                   0                             0   \n",
       "293978                   0                   0                             0   \n",
       "6918                     0                   0                             0   \n",
       "43504                    0                   0                             0   \n",
       "300413                   0                   0                             0   \n",
       "...                    ...                 ...                           ...   \n",
       "34793                    0                   0                             0   \n",
       "293345                   0                   0                             0   \n",
       "27761                    0                   0                             0   \n",
       "120414                   0                   0                             0   \n",
       "304007                   0                   0                             0   \n",
       "\n",
       "        country_name_the United States  \n",
       "45158                                0  \n",
       "293978                               1  \n",
       "6918                                 1  \n",
       "43504                                1  \n",
       "300413                               1  \n",
       "...                                ...  \n",
       "34793                                1  \n",
       "293345                               1  \n",
       "27761                                1  \n",
       "120414                               1  \n",
       "304007                               1  \n",
       "\n",
       "[127193 rows x 31 columns]"
      ]
     },
     "execution_count": 62,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "X_train"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 63,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Model, Accuracy, Precision, Recall\n",
      "    Random Forest: (0.7357894240294354, 0.7588246816846229, 0.7933237763669875)\n",
      "    Logistic Regression: (0.6277654606349357, 0.6216018255063808, 0.9037476960884702)\n",
      "    Gradient Boost: (0.7423227510731638, 0.7484712296470935, 0.832193323776367)\n"
     ]
    }
   ],
   "source": [
    "def get_scores(classifier, X_train, X_test, y_train, y_test, **kwargs):\n",
    "    model = classifier(**kwargs)\n",
    "    model.fit(X_train, y_train)\n",
    "    y_predict = model.predict(X_test)\n",
    "    return model.score(X_test, y_test), precision_score(y_test, y_predict), recall_score(y_test, y_predict)\n",
    "print(\"Model, Accuracy, Precision, Recall\")\n",
    "print(\"    Random Forest:\", get_scores(RandomForestClassifier, X_train, X_test, y_train, y_test))\n",
    "print(\"    Logistic Regression:\", get_scores(LogisticRegression, X_train, X_test, y_train, y_test))\n",
    "print(\"    Gradient Boost:\", get_scores(GradientBoostingClassifier, X_train, X_test, y_train, y_test))\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 64,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/opt/conda/lib/python3.7/site-packages/sklearn/linear_model/_logistic.py:940: ConvergenceWarning: lbfgs failed to converge (status=1):\n",
      "STOP: TOTAL NO. of ITERATIONS REACHED LIMIT.\n",
      "\n",
      "Increase the number of iterations (max_iter) or scale the data as shown in:\n",
      "    https://scikit-learn.org/stable/modules/preprocessing.html\n",
      "Please also refer to the documentation for alternative solver options:\n",
      "    https://scikit-learn.org/stable/modules/linear_model.html#logistic-regression\n",
      "  extra_warning_msg=_LOGISTIC_SOLVER_CONVERGENCE_MSG)\n",
      "/opt/conda/lib/python3.7/site-packages/sklearn/linear_model/_logistic.py:940: ConvergenceWarning: lbfgs failed to converge (status=1):\n",
      "STOP: TOTAL NO. of ITERATIONS REACHED LIMIT.\n",
      "\n",
      "Increase the number of iterations (max_iter) or scale the data as shown in:\n",
      "    https://scikit-learn.org/stable/modules/preprocessing.html\n",
      "Please also refer to the documentation for alternative solver options:\n",
      "    https://scikit-learn.org/stable/modules/linear_model.html#logistic-regression\n",
      "  extra_warning_msg=_LOGISTIC_SOLVER_CONVERGENCE_MSG)\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[[22065  2350]\n",
      " [13432  4551]]\n",
      "Cross val score: 0.6243582454656187\n",
      "ROC_AUC score: 0.6326978890112288\n"
     ]
    }
   ],
   "source": [
    "lg = LogisticRegression()\n",
    "lg.fit(X_train, y_train)\n",
    "y_predict = lg.predict(X_test)\n",
    "cv_score = cross_val_score(lg, X_train, y_train)\n",
    "score = cross_val_score(lg, X_train, y_train, scoring = 'roc_auc')\n",
    "tn, fp, fn, tp = confusion_matrix(y_test, y_predict).ravel()\n",
    "print(np.array([[tp, fn], [fp, tn]]))\n",
    "print('Cross val score: {}'.format(np.mean(cv_score)))\n",
    "print('ROC_AUC score: {}'.format (np.mean(score)))\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "rf = RandomForestClassifier()\n",
    "rf.fit(X_train, y_train)\n",
    "y_predict = rf.predict(X_test)\n",
    "cv_score = cross_val_score(rf, X_train, y_train)\n",
    "score = cross_val_score(rf, X_train, y_train, scoring = 'roc_auc')\n",
    "tn, fp, fn, tp = confusion_matrix(y_test, y_predict).ravel()\n",
    "print(np.array([[tp, fn], [fp, tn]]))\n",
    "print('Cross val score: {}'.format(np.mean(cv_score)))\n",
    "print('ROC_AUC score: {}'.format (np.mean(score)))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "gb = GradientBoostingClassifier()\n",
    "gb.fit(X_train, y_train)\n",
    "y_predict = gb.predict(X_test)\n",
    "cv_score = cross_val_score(gb, X_train, y_train)\n",
    "score = cross_val_score(gb, X_train, y_train, scoring = 'roc_auc')\n",
    "tn, fp, fn, tp = confusion_matrix(y_test, y_predict).ravel()\n",
    "print(np.array([[tp, fn], [fp, tn]]))\n",
    "print('Cross val score: {}'.format(np.mean(cv_score)))\n",
    "print('ROC_AUC score: {}'.format (np.mean(score)))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 172,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Top five: ['goal', 'staff_pick_x', 'launched_year', 'campaign_length', 'category_type_food', 'category_type_technology', 'category_type_design', 'category_type_journalism', 'category_type_crafts', 'category_type_art']\n"
     ]
    }
   ],
   "source": [
    "rf_feature_importances = np.argsort(gb.feature_importances_)\n",
    "print(\"Top five:\", list(X.columns[rf_feature_importances][-1:-11:-1]))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 137,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Top five: ['staff_pick_x', 'launched_year', 'category_type_publishing', 'category_type_comics', 'category_type_music', 'country_name_the United States', 'category_type_games', 'category_type_fashion', 'country_name_other', 'category_type_dance']\n"
     ]
    }
   ],
   "source": [
    "lg = LogisticRegression()\n",
    "lg.fit(X,y)\n",
    "importance = lg.coef_[0]\n",
    "\n",
    "# for idx, elem in enumerate(importance):\n",
    "#     print('Feature: %0d, Score: %0.5f' % (idx, elem))\n",
    "    \n",
    "lg_feature_importance = np.argsort(importance)\n",
    "print(\"Top five:\", list(X.columns[lg_feature_importance][-1:-11:-1]))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 125,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Feature ranking:\n",
      "1. goal (0.291412)\n",
      "2. launched_year (0.165930)\n",
      "3. staff_pick_y (0.125546)\n",
      "4. staff_pick_x (0.116686)\n",
      "5. campaign_length (0.099623)\n",
      "6. category_type_crafts (0.018910)\n",
      "7. category_type_art (0.012902)\n",
      "8. category_type_comics (0.010053)\n",
      "9. blurb_length (0.007473)\n",
      "10. launched_month (0.000188)\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "Text(0.5, 1.0, 'Feature Importances for Gradient Boost Classifier')"
      ]
     },
     "execution_count": 125,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "\n",
    "n = 10 # top 10 features\n",
    "\n",
    "\n",
    "importances = gb.feature_importances_[:n]\n",
    "\n",
    "indices = np.argsort(importances)[::-1]\n",
    "features = list(X.columns[indices])\n",
    "\n",
    "# Print the feature ranking\n",
    "print(\"Feature ranking:\")\n",
    "\n",
    "for f in range(n):\n",
    "    print(\"%d. %s (%f)\" % (f + 1, features[f], importances[indices[f]]))\n",
    "\n",
    "# Plot the feature importances\n",
    "fig, ax = plt.subplots()\n",
    "\n",
    "ax.bar(range(10), importances[indices], color='lightcoral', align=\"center\")\n",
    "ax.set_xticks(range(10))\n",
    "ax.set_xticklabels(features, rotation = 90, fontsize=12)\n",
    "ax.set_xlim([-1, 10])\n",
    "ax.set_xlabel(\"Importance\", fontsize=16)\n",
    "ax.set_title(\"Feature Importances for Gradient Boost Classifier\", fontsize=16)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Neural Network"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 103,
   "metadata": {},
   "outputs": [],
   "source": [
    "n_samples, n_feats = X_train.shape\n",
    "\n",
    "tf_model = Sequential() \n",
    "\n",
    "denselayer = Dense(units=1,\n",
    "                input_dim=n_feats,\n",
    "                kernel_initializer='uniform',\n",
    "                activation='sigmoid')\n",
    "hidden_layer = Dense(units=200,\n",
    "                kernel_initializer='uniform',\n",
    "                activation='relu')\n",
    "hidden_layer1 = Dense(units=100,\n",
    "                kernel_initializer='uniform',\n",
    "                activation='sigmoid')\n",
    "out_layer = Dense(units=1,\n",
    "                 kernel_initializer='uniform',\n",
    "                activation='sigmoid')\n",
    "tf_model.add(denselayer)\n",
    "tf_model.add(hidden_layer)\n",
    "tf_model.add(hidden_layer1)\n",
    "tf_model.add(out_layer)\n",
    "tf_model.compile(loss='binary_crossentropy', optimizer='adam', \n",
    "                 metrics=[Accuracy()])\n",
    "\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 104,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Train on 127193 samples\n",
      "Epoch 1/10\n",
      "127193/127193 [==============================] - 15s 122us/sample - loss: 33.6918 - accuracy: 0.2489\n",
      "Epoch 2/10\n",
      "127193/127193 [==============================] - 15s 121us/sample - loss: 21.3995 - accuracy: 0.1912\n",
      "Epoch 3/10\n",
      "127193/127193 [==============================] - 17s 133us/sample - loss: 24.0078 - accuracy: 0.1960\n",
      "Epoch 4/10\n",
      "127193/127193 [==============================] - 18s 139us/sample - loss: 17.8051 - accuracy: 0.2009\n",
      "Epoch 5/10\n",
      "127193/127193 [==============================] - 18s 138us/sample - loss: 38.3142 - accuracy: 0.2630\n",
      "Epoch 6/10\n",
      "127193/127193 [==============================] - 18s 143us/sample - loss: 24.2508 - accuracy: 0.1995\n",
      "Epoch 7/10\n",
      "127193/127193 [==============================] - 18s 141us/sample - loss: 23.2316 - accuracy: 0.2118\n",
      "Epoch 8/10\n",
      "127193/127193 [==============================] - 18s 142us/sample - loss: 31.4619 - accuracy: 0.2044\n",
      "Epoch 9/10\n",
      "127193/127193 [==============================] - 19s 151us/sample - loss: 24.9912 - accuracy: 0.2036\n",
      "Epoch 10/10\n",
      "127193/127193 [==============================] - 25s 194us/sample - loss: 30.1951 - accuracy: 0.2397\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "<tensorflow.python.keras.callbacks.History at 0x7fa8c89314d0>"
      ]
     },
     "execution_count": 104,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "model.fit(X_train, y_train, epochs=10)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 95,
   "metadata": {},
   "outputs": [],
   "source": [
    "n_samples, n_feats = X_train.shape\n",
    "\n",
    "model = Sequential() # sequence of layers\n",
    "\n",
    "denselayer = Dense(units=1,\n",
    "                input_dim=n_feats,\n",
    "                kernel_initializer='uniform',\n",
    "                activation='sigmoid')\n",
    "\n",
    "model.add(denselayer)\n",
    "model.compile(loss='binary_crossentropy', optimizer='adam', \n",
    "                 metrics=[Accuracy()])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 96,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Train on 127193 samples\n",
      "Epoch 1/20\n",
      "127193/127193 [==============================] - 24s 186us/sample - loss: 32.5718 - accuracy: 0.1868\n",
      "Epoch 2/20\n",
      "127193/127193 [==============================] - 26s 206us/sample - loss: 34.1207 - accuracy: 0.1768\n",
      "Epoch 3/20\n",
      "127193/127193 [==============================] - 28s 217us/sample - loss: 35.6821 - accuracy: 0.1969\n",
      "Epoch 4/20\n",
      "127193/127193 [==============================] - 25s 200us/sample - loss: 13.9899 - accuracy: 0.1027\n",
      "Epoch 5/20\n",
      "127193/127193 [==============================] - 25s 193us/sample - loss: 37.8606 - accuracy: 0.2211\n",
      "Epoch 6/20\n",
      "127193/127193 [==============================] - 24s 188us/sample - loss: 38.9339 - accuracy: 0.2881\n",
      "Epoch 7/20\n",
      "127193/127193 [==============================] - 24s 191us/sample - loss: 24.9688 - accuracy: 0.2213\n",
      "Epoch 8/20\n",
      "127193/127193 [==============================] - 25s 194us/sample - loss: 44.2456 - accuracy: 0.3001\n",
      "Epoch 9/20\n",
      "127193/127193 [==============================] - 26s 206us/sample - loss: 34.9283 - accuracy: 0.2163\n",
      "Epoch 10/20\n",
      "127193/127193 [==============================] - 24s 190us/sample - loss: 19.9765 - accuracy: 0.1934\n",
      "Epoch 11/20\n",
      "127193/127193 [==============================] - 25s 197us/sample - loss: 26.4955 - accuracy: 0.2260\n",
      "Epoch 12/20\n",
      "127193/127193 [==============================] - 23s 179us/sample - loss: 41.0121 - accuracy: 0.2712\n",
      "Epoch 13/20\n",
      "127193/127193 [==============================] - 23s 178us/sample - loss: 29.5323 - accuracy: 0.2643\n",
      "Epoch 14/20\n",
      "127193/127193 [==============================] - 23s 181us/sample - loss: 25.6145 - accuracy: 0.1943\n",
      "Epoch 15/20\n",
      "127193/127193 [==============================] - 24s 191us/sample - loss: 26.2651 - accuracy: 0.2065\n",
      "Epoch 16/20\n",
      "127193/127193 [==============================] - 23s 181us/sample - loss: 7.9128 - accuracy: 0.1580\n",
      "Epoch 17/20\n",
      "127193/127193 [==============================] - 25s 200us/sample - loss: 18.4265 - accuracy: 0.1990\n",
      "Epoch 18/20\n",
      "127193/127193 [==============================] - 25s 199us/sample - loss: 37.2528 - accuracy: 0.2338\n",
      "Epoch 19/20\n",
      "127193/127193 [==============================] - 23s 180us/sample - loss: 29.9580 - accuracy: 0.2286\n",
      "Epoch 20/20\n",
      "127193/127193 [==============================] - 23s 181us/sample - loss: 26.8920 - accuracy: 0.2263\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "<tensorflow.python.keras.callbacks.History at 0x7fa8c94a9a90>"
      ]
     },
     "execution_count": 96,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "model.fit(X_train, y_train, epochs=20)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [
    {
     "ename": "NameError",
     "evalue": "name 'LogisticRegression' is not defined",
     "output_type": "error",
     "traceback": [
      "\u001b[0;31m---------------------------------------------------------------------------\u001b[0m",
      "\u001b[0;31mNameError\u001b[0m                                 Traceback (most recent call last)",
      "\u001b[0;32m<ipython-input-2-b6ea9c1392c3>\u001b[0m in \u001b[0;36m<module>\u001b[0;34m\u001b[0m\n\u001b[0;32m----> 1\u001b[0;31m \u001b[0mlg\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mLogisticRegression\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m",
      "\u001b[0;31mNameError\u001b[0m: name 'LogisticRegression' is not defined"
     ]
    }
   ],
   "source": [
    "lg = LogisticRegression()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [
    {
     "ename": "NameError",
     "evalue": "name 'lg' is not defined",
     "output_type": "error",
     "traceback": [
      "\u001b[0;31m---------------------------------------------------------------------------\u001b[0m",
      "\u001b[0;31mNameError\u001b[0m                                 Traceback (most recent call last)",
      "\u001b[0;32m<ipython-input-1-1192c00b0714>\u001b[0m in \u001b[0;36m<module>\u001b[0;34m\u001b[0m\n\u001b[0;32m----> 1\u001b[0;31m \u001b[0mprobabilities_lg\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mlg\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mpredict_proba\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mX_test\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;36m1\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m      2\u001b[0m \u001b[0mprobabilities_rf\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mrf\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mpredict_proba\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mX_test\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;36m1\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m      3\u001b[0m \u001b[0mprobabilities_gb\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mgb\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mpredict_proba\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mX_test\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;36m1\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m      4\u001b[0m \u001b[0mprobabilities_nn\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mgb\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mpredict_proba\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mX_test\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mflatten\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m      5\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n",
      "\u001b[0;31mNameError\u001b[0m: name 'lg' is not defined"
     ]
    }
   ],
   "source": [
    "probabilities_lg = lg.predict_proba(X_test)[:,1]\n",
    "probabilities_rf = rf.predict_proba(X_test)[:,1]\n",
    "probabilities_gb = gb.predict_proba(X_test)[:,1]\n",
    "probabilities_nn = gb.predict_proba(X_test).flatten()\n",
    "\n",
    "\n",
    "def roc_curve(probabilities, labels):\n",
    "    df = pd.DataFrame({'prob': probabilities, 'y': labels})\n",
    "    df.sort_values('prob', inplace=True)\n",
    "    actual_p = df.y.sum()\n",
    "    actual_n = df.shape[0] - df.y.sum()\n",
    "    df['tn'] = (df.y == 0).cumsum()\n",
    "    df['fn'] = df.y.cumsum()\n",
    "    df['fp'] = actual_n - df.tn\n",
    "    df['tp'] = actual_p - df.fn\n",
    "    df['fpr'] = df.fp/(df.fp + df.tn)\n",
    "    df['tpr'] = df.tp/(df.tp + df.fn)\n",
    "    df['precision'] = df.tp/(df.tp + df.fp)\n",
    "    return list(df.tpr), list(df.fpr), list(df.precision)\n",
    "\n",
    "tpr_lg, fpr_lg, thresholds_lg = roc_curve(probabilities_lg, y_test)\n",
    "tpr_rf, fpr_rf, thresholds_rf = roc_curve(probabilities_rf, y_test)\n",
    "tpr_gb, fpr_gb, thresholds_gb = roc_curve(probabilities_gb, y_test)\n",
    "tpr_nn, fpr_nn, thresholds_nn = roc_curve(probabilities_nn, y_test)\n",
    "\n",
    "fig, ax = plt.subplots(figsize=(10,6))\n",
    "ax.plot(fpr_lg, tpr_lg, label='Logistic Regression')\n",
    "ax.plot(fpr_rf, tpr_rf, label='Random Forest')\n",
    "ax.plot(fpr_gb, tpr_gb, label='Gradient Boost')\n",
    "ax.plot(fpr_nn, tpr_nn, label='Neural Network')\n",
    "\n",
    "ax.set_xlabel('False Positive Rate (1-Specificity)', fontsize=16)\n",
    "ax.set_ylabel('True Positive Rate (Sensitivity, Recall)', fontsize=16)\n",
    "ax.set_title('ROC Plot', fontsize=20)\n",
    "ax.legend()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## NLP"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 195,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[nltk_data] Downloading package vader_lexicon to\n",
      "[nltk_data]     /home/ec2-user/nltk_data...\n",
      "[nltk_data]   Package vader_lexicon is already up-to-date!\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "True"
      ]
     },
     "execution_count": 195,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "nltk.download('vader_lexicon')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 150,
   "metadata": {},
   "outputs": [],
   "source": [
    "# drop two rows of blurb with Nan, and Japanese\n",
    "df_nlp = df2.drop(axis=0, index=[14, 55244])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 197,
   "metadata": {},
   "outputs": [],
   "source": [
    "vader_analyzer = SentimentIntensityAnalyzer()\n",
    "\n",
    "def vader_neg(words):\n",
    "    d = vader_analyzer.polarity_scores(words)\n",
    "    return d['neg']\n",
    "def vader_neu(words):\n",
    "    d = vader_analyzer.polarity_scores(words)\n",
    "    return d['neu']\n",
    "def vader_pos(words):\n",
    "    d = vader_analyzer.polarity_scores(words)\n",
    "    return d['pos']\n",
    "def vader_com(words):\n",
    "    d = vader_analyzer.polarity_scores(words)\n",
    "    return d['compound']\n",
    "\n",
    "df_nlp['blurb_neg'] = df_nlp['blurb'].map(lambda x: vader_neg(x))\n",
    "df_nlp['blurb_neu'] = df_nlp['blurb'].map(lambda x: vader_neu(x))\n",
    "df_nlp['blurb_pos'] = df_nlp['blurb'].map(lambda x: vader_pos(x))\n",
    "df_nlp['blurb_compound'] = df_nlp['blurb'].map(lambda x: vader_com(x))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 205,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Index(['backers_count', 'blurb', 'category', 'country_displayable_name',\n",
       "       'deadline', 'goal', 'launched_at', 'name', 'pledged', 'slug',\n",
       "       'staff_pick', 'state', 'launched_year', 'launched_month',\n",
       "       'campaign_length', 'category_type', 'blurb_length', 'country_name',\n",
       "       'blurb_neg', 'blurb_neu', 'blurb_pos', 'blurb_compound'],\n",
       "      dtype='object')"
      ]
     },
     "execution_count": 205,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_nlp.columns"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 206,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Index(['backers_count', 'blurb', 'category', 'country_displayable_name',\n",
       "       'deadline', 'goal', 'launched_at', 'name', 'pledged', 'slug',\n",
       "       'staff_pick_x', 'state', 'launched_year', 'launched_month',\n",
       "       'campaign_length', 'category_type', 'blurb_length', 'country_name',\n",
       "       'staff_pick_y', 'category_type_art', 'category_type_comics',\n",
       "       'category_type_crafts', 'category_type_dance', 'category_type_design',\n",
       "       'category_type_fashion', 'category_type_film & video',\n",
       "       'category_type_food', 'category_type_games', 'category_type_journalism',\n",
       "       'category_type_music', 'category_type_photography',\n",
       "       'category_type_publishing', 'category_type_technology',\n",
       "       'category_type_theater', 'country_name_Australia',\n",
       "       'country_name_Canada', 'country_name_France', 'country_name_Germany',\n",
       "       'country_name_Italy', 'country_name_Mexico', 'country_name_Spain',\n",
       "       'country_name_other', 'country_name_the Netherlands',\n",
       "       'country_name_the United States', 'name_length'],\n",
       "      dtype='object')"
      ]
     },
     "execution_count": 206,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "model_data.columns"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 217,
   "metadata": {},
   "outputs": [],
   "source": [
    "# merging nlp tables with model table\n",
    "df_nlp_extract = df_nlp.loc[:, 'blurb_neg': 'blurb_compound']\n",
    "model_data_nlp = pd.merge(model_data, df_nlp_extract, left_index=True, right_index=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 219,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Index(['backers_count', 'blurb', 'category', 'country_displayable_name',\n",
       "       'deadline', 'goal', 'launched_at', 'name', 'pledged', 'slug',\n",
       "       'staff_pick_x', 'state', 'launched_year', 'launched_month',\n",
       "       'campaign_length', 'category_type', 'blurb_length', 'country_name',\n",
       "       'staff_pick_y', 'category_type_art', 'category_type_comics',\n",
       "       'category_type_crafts', 'category_type_dance', 'category_type_design',\n",
       "       'category_type_fashion', 'category_type_film & video',\n",
       "       'category_type_food', 'category_type_games', 'category_type_journalism',\n",
       "       'category_type_music', 'category_type_photography',\n",
       "       'category_type_publishing', 'category_type_technology',\n",
       "       'category_type_theater', 'country_name_Australia',\n",
       "       'country_name_Canada', 'country_name_France', 'country_name_Germany',\n",
       "       'country_name_Italy', 'country_name_Mexico', 'country_name_Spain',\n",
       "       'country_name_other', 'country_name_the Netherlands',\n",
       "       'country_name_the United States', 'name_length', 'blurb_neg',\n",
       "       'blurb_neu', 'blurb_pos', 'blurb_compound'],\n",
       "      dtype='object')"
      ]
     },
     "execution_count": 219,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "model_data_nlp.columns"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 241,
   "metadata": {},
   "outputs": [],
   "source": [
    "X = model_data_nlp.drop(['backers_count','blurb', 'category', 'country_displayable_name', 'deadline', \n",
    "                     'launched_at', 'name', 'pledged', 'slug', 'state', 'category_type', 'country_name', 'name_length',\n",
    "                    ],\n",
    "                   axis=1)\n",
    "y = model_data_nlp['state']"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 242,
   "metadata": {},
   "outputs": [],
   "source": [
    "X_train, X_test, y_train, y_test = train_test_split(X, y)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 243,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Model, Accuracy, Precision, Recall\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/home/ec2-user/anaconda3/envs/python3/lib/python3.6/site-packages/sklearn/ensemble/forest.py:246: FutureWarning: The default value of n_estimators will change from 10 in version 0.20 to 100 in 0.22.\n",
      "  \"10 in version 0.20 to 100 in 0.22.\", FutureWarning)\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "    Random Forest: (0.718689560828341, 0.772984953753037, 0.733141979301423)\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/home/ec2-user/anaconda3/envs/python3/lib/python3.6/site-packages/sklearn/linear_model/logistic.py:433: FutureWarning: Default solver will be changed to 'lbfgs' in 0.22. Specify a solver to silence this warning.\n",
      "  FutureWarning)\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "    Logistic Regression: (0.6221991603377518, 0.6104718941659486, 0.9738033635187581)\n",
      "    Gradient Boost: (0.7424406811642058, 0.7559087204563977, 0.8249110608020699)\n"
     ]
    }
   ],
   "source": [
    "def get_scores(classifier, X_train, X_test, y_train, y_test, **kwargs):\n",
    "    model = classifier(**kwargs)\n",
    "    model.fit(X_train, y_train)\n",
    "    y_predict = model.predict(X_test)\n",
    "    return model.score(X_test, y_test), precision_score(y_test, y_predict), recall_score(y_test, y_predict)\n",
    "print(\"Model, Accuracy, Precision, Recall\")\n",
    "print(\"    Random Forest:\", get_scores(RandomForestClassifier, X_train, X_test, y_train, y_test))\n",
    "print(\"    Logistic Regression:\", get_scores(LogisticRegression, X_train, X_test, y_train, y_test))\n",
    "print(\"    Gradient Boost:\", get_scores(GradientBoostingClassifier, X_train, X_test, y_train, y_test))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 244,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/home/ec2-user/anaconda3/envs/python3/lib/python3.6/site-packages/sklearn/model_selection/_split.py:2053: FutureWarning: You should specify a value for 'cv' instead of relying on the default value. The default value will change from 3 to 5 in version 0.22.\n",
      "  warnings.warn(CV_WARNING, FutureWarning)\n",
      "/home/ec2-user/anaconda3/envs/python3/lib/python3.6/site-packages/sklearn/model_selection/_split.py:2053: FutureWarning: You should specify a value for 'cv' instead of relying on the default value. The default value will change from 3 to 5 in version 0.22.\n",
      "  warnings.warn(CV_WARNING, FutureWarning)\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[[20405  4331]\n",
      " [ 6589 11073]]\n",
      "Cross val score: 0.7386450120162378\n",
      "ROC_AUC score: 0.8125653700448027\n"
     ]
    }
   ],
   "source": [
    "gb = GradientBoostingClassifier()\n",
    "gb.fit(X_train, y_train)\n",
    "y_predict = gb.predict(X_test)\n",
    "cv_score = cross_val_score(gb, X_train, y_train)\n",
    "score = cross_val_score(gb, X_train, y_train, scoring = 'roc_auc')\n",
    "tn, fp, fn, tp = confusion_matrix(y_test, y_predict).ravel()\n",
    "print(np.array([[tp, fn], [fp, tn]]))\n",
    "print('Cross val score: {}'.format(np.mean(cv_score)))\n",
    "print('ROC_AUC score: {}'.format (np.mean(score)))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "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": "3.7.6"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 4
}