Similar products.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Similar products"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "% matplotlib inline\n",
    "import pickle\n",
    "\n",
    "import numpy as np\n",
    "import pandas as pd\n",
    "from sklearn.neighbors import NearestNeighbors\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Load video games review dataframe"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [],
   "source": [
    "with open('pickle/video_games.pkl', 'rb') as input:\n",
    "    video_games = pickle.load(input)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style>\n",
       "    .dataframe thead tr:only-child th {\n",
       "        text-align: right;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: left;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\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>overall</th>\n",
       "      <th>reviewText</th>\n",
       "      <th>reviewerName</th>\n",
       "      <th>summary</th>\n",
       "      <th>date</th>\n",
       "      <th>helpfulRate</th>\n",
       "      <th>helpfulCount</th>\n",
       "      <th>rank</th>\n",
       "      <th>description</th>\n",
       "      <th>price</th>\n",
       "      <th>title</th>\n",
       "      <th>brand</th>\n",
       "      <th>reviewLength</th>\n",
       "      <th>summaryLength</th>\n",
       "      <th>reviewerNameLength</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>asin</th>\n",
       "      <th>reviewerID</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th rowspan=\"5\" valign=\"top\">0700099867</th>\n",
       "      <th>A2HD75EMZR8QLN</th>\n",
       "      <td>1</td>\n",
       "      <td>Installing the game was a struggle (because of...</td>\n",
       "      <td>123</td>\n",
       "      <td>Pay to unlock content? I don't think so.</td>\n",
       "      <td>2012-07-09</td>\n",
       "      <td>0.666667</td>\n",
       "      <td>12</td>\n",
       "      <td>6629</td>\n",
       "      <td>Dirt 3 is a popular rally racing game for Play...</td>\n",
       "      <td>246.63</td>\n",
       "      <td></td>\n",
       "      <td></td>\n",
       "      <td>779</td>\n",
       "      <td>40</td>\n",
       "      <td>3.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>A3UR8NLLY1ZHCX</th>\n",
       "      <td>4</td>\n",
       "      <td>If you like rally cars get this game you will ...</td>\n",
       "      <td>Alejandro Henao \"Electronic Junky\"</td>\n",
       "      <td>Good rally game</td>\n",
       "      <td>2013-06-30</td>\n",
       "      <td>-1.000000</td>\n",
       "      <td>0</td>\n",
       "      <td>6629</td>\n",
       "      <td>Dirt 3 is a popular rally racing game for Play...</td>\n",
       "      <td>246.63</td>\n",
       "      <td></td>\n",
       "      <td></td>\n",
       "      <td>304</td>\n",
       "      <td>15</td>\n",
       "      <td>34.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>A1INA0F5CWW3J4</th>\n",
       "      <td>1</td>\n",
       "      <td>1st shipment received a book instead of the ga...</td>\n",
       "      <td>Amazon Shopper \"Mr.Repsol\"</td>\n",
       "      <td>Wrong key</td>\n",
       "      <td>2014-06-28</td>\n",
       "      <td>-1.000000</td>\n",
       "      <td>0</td>\n",
       "      <td>6629</td>\n",
       "      <td>Dirt 3 is a popular rally racing game for Play...</td>\n",
       "      <td>246.63</td>\n",
       "      <td></td>\n",
       "      <td></td>\n",
       "      <td>282</td>\n",
       "      <td>9</td>\n",
       "      <td>26.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>A1DLMTOTHQ4AST</th>\n",
       "      <td>3</td>\n",
       "      <td>I got this version instead of the PS3 version,...</td>\n",
       "      <td>ampgreen</td>\n",
       "      <td>awesome game, if it did not crash frequently !!</td>\n",
       "      <td>2011-09-14</td>\n",
       "      <td>0.700000</td>\n",
       "      <td>10</td>\n",
       "      <td>6629</td>\n",
       "      <td>Dirt 3 is a popular rally racing game for Play...</td>\n",
       "      <td>246.63</td>\n",
       "      <td></td>\n",
       "      <td></td>\n",
       "      <td>3489</td>\n",
       "      <td>47</td>\n",
       "      <td>8.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>A361M14PU2GUEG</th>\n",
       "      <td>4</td>\n",
       "      <td>I had Dirt 2 on Xbox 360 and it was an okay ga...</td>\n",
       "      <td>Angry Ryan \"Ryan A. Forrest\"</td>\n",
       "      <td>DIRT 3</td>\n",
       "      <td>2011-06-14</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>2</td>\n",
       "      <td>6629</td>\n",
       "      <td>Dirt 3 is a popular rally racing game for Play...</td>\n",
       "      <td>246.63</td>\n",
       "      <td></td>\n",
       "      <td></td>\n",
       "      <td>294</td>\n",
       "      <td>6</td>\n",
       "      <td>28.0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                           overall  \\\n",
       "asin       reviewerID                \n",
       "0700099867 A2HD75EMZR8QLN        1   \n",
       "           A3UR8NLLY1ZHCX        4   \n",
       "           A1INA0F5CWW3J4        1   \n",
       "           A1DLMTOTHQ4AST        3   \n",
       "           A361M14PU2GUEG        4   \n",
       "\n",
       "                                                                  reviewText  \\\n",
       "asin       reviewerID                                                          \n",
       "0700099867 A2HD75EMZR8QLN  Installing the game was a struggle (because of...   \n",
       "           A3UR8NLLY1ZHCX  If you like rally cars get this game you will ...   \n",
       "           A1INA0F5CWW3J4  1st shipment received a book instead of the ga...   \n",
       "           A1DLMTOTHQ4AST  I got this version instead of the PS3 version,...   \n",
       "           A361M14PU2GUEG  I had Dirt 2 on Xbox 360 and it was an okay ga...   \n",
       "\n",
       "                                                 reviewerName  \\\n",
       "asin       reviewerID                                           \n",
       "0700099867 A2HD75EMZR8QLN                                 123   \n",
       "           A3UR8NLLY1ZHCX  Alejandro Henao \"Electronic Junky\"   \n",
       "           A1INA0F5CWW3J4          Amazon Shopper \"Mr.Repsol\"   \n",
       "           A1DLMTOTHQ4AST                            ampgreen   \n",
       "           A361M14PU2GUEG        Angry Ryan \"Ryan A. Forrest\"   \n",
       "\n",
       "                                                                   summary  \\\n",
       "asin       reviewerID                                                        \n",
       "0700099867 A2HD75EMZR8QLN         Pay to unlock content? I don't think so.   \n",
       "           A3UR8NLLY1ZHCX                                  Good rally game   \n",
       "           A1INA0F5CWW3J4                                        Wrong key   \n",
       "           A1DLMTOTHQ4AST  awesome game, if it did not crash frequently !!   \n",
       "           A361M14PU2GUEG                                           DIRT 3   \n",
       "\n",
       "                                date  helpfulRate  helpfulCount  rank  \\\n",
       "asin       reviewerID                                                   \n",
       "0700099867 A2HD75EMZR8QLN 2012-07-09     0.666667            12  6629   \n",
       "           A3UR8NLLY1ZHCX 2013-06-30    -1.000000             0  6629   \n",
       "           A1INA0F5CWW3J4 2014-06-28    -1.000000             0  6629   \n",
       "           A1DLMTOTHQ4AST 2011-09-14     0.700000            10  6629   \n",
       "           A361M14PU2GUEG 2011-06-14     1.000000             2  6629   \n",
       "\n",
       "                                                                 description  \\\n",
       "asin       reviewerID                                                          \n",
       "0700099867 A2HD75EMZR8QLN  Dirt 3 is a popular rally racing game for Play...   \n",
       "           A3UR8NLLY1ZHCX  Dirt 3 is a popular rally racing game for Play...   \n",
       "           A1INA0F5CWW3J4  Dirt 3 is a popular rally racing game for Play...   \n",
       "           A1DLMTOTHQ4AST  Dirt 3 is a popular rally racing game for Play...   \n",
       "           A361M14PU2GUEG  Dirt 3 is a popular rally racing game for Play...   \n",
       "\n",
       "                            price title brand  reviewLength  summaryLength  \\\n",
       "asin       reviewerID                                                        \n",
       "0700099867 A2HD75EMZR8QLN  246.63                       779             40   \n",
       "           A3UR8NLLY1ZHCX  246.63                       304             15   \n",
       "           A1INA0F5CWW3J4  246.63                       282              9   \n",
       "           A1DLMTOTHQ4AST  246.63                      3489             47   \n",
       "           A361M14PU2GUEG  246.63                       294              6   \n",
       "\n",
       "                           reviewerNameLength  \n",
       "asin       reviewerID                          \n",
       "0700099867 A2HD75EMZR8QLN                 3.0  \n",
       "           A3UR8NLLY1ZHCX                34.0  \n",
       "           A1INA0F5CWW3J4                26.0  \n",
       "           A1DLMTOTHQ4AST                 8.0  \n",
       "           A361M14PU2GUEG                28.0  "
      ]
     },
     "execution_count": 3,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "video_games.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Load products metadata"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style>\n",
       "    .dataframe thead tr:only-child th {\n",
       "        text-align: right;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: left;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>asin</th>\n",
       "      <th>brand</th>\n",
       "      <th>categories</th>\n",
       "      <th>description</th>\n",
       "      <th>imUrl</th>\n",
       "      <th>price</th>\n",
       "      <th>related</th>\n",
       "      <th>salesRank</th>\n",
       "      <th>title</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>0078764343</td>\n",
       "      <td>NaN</td>\n",
       "      <td>[[Video Games, Xbox 360, Games]]</td>\n",
       "      <td>Brand new sealed!</td>\n",
       "      <td>http://ecx.images-amazon.com/images/I/513h6dPb...</td>\n",
       "      <td>37.98</td>\n",
       "      <td>{'also_bought': ['B000TI836G', 'B003Q53VZC', '...</td>\n",
       "      <td>{'Video Games': 28655}</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>043933702X</td>\n",
       "      <td>NaN</td>\n",
       "      <td>[[Video Games, PC, Games]]</td>\n",
       "      <td>In Stock NOW.  Eligible for FREE Super Saving ...</td>\n",
       "      <td>http://ecx.images-amazon.com/images/I/61KKRndV...</td>\n",
       "      <td>23.50</td>\n",
       "      <td>{'also_viewed': ['B000067NP1', '0439573947', '...</td>\n",
       "      <td>{'Video Games': 44080}</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>0439339987</td>\n",
       "      <td>NaN</td>\n",
       "      <td>[[Video Games, PC, Games]]</td>\n",
       "      <td>Grandma Groupers kelp seeds are missing and wi...</td>\n",
       "      <td>http://ecx.images-amazon.com/images/I/416QZg89...</td>\n",
       "      <td>8.95</td>\n",
       "      <td>{'also_bought': ['B000314VVU', 'B000PXUOTE', '...</td>\n",
       "      <td>{'Video Games': 49836}</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>0439342260</td>\n",
       "      <td>NaN</td>\n",
       "      <td>[[Video Games, PC, Games]]</td>\n",
       "      <td>This software is BRAND NEW. Packaging may diff...</td>\n",
       "      <td>http://ecx.images-amazon.com/images/I/61Wvu-Uj...</td>\n",
       "      <td>NaN</td>\n",
       "      <td>{'also_viewed': ['043934302X']}</td>\n",
       "      <td>{'Video Games': 49156}</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>0439339960</td>\n",
       "      <td>NaN</td>\n",
       "      <td>[[Video Games, PC, Games]]</td>\n",
       "      <td>a scholastic clubs fairs cd rom game</td>\n",
       "      <td>http://ecx.images-amazon.com/images/I/51k3oRCF...</td>\n",
       "      <td>NaN</td>\n",
       "      <td>{'also_viewed': ['B00028D7TG']}</td>\n",
       "      <td>{'Video Games': 52262}</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "         asin brand                        categories  \\\n",
       "0  0078764343   NaN  [[Video Games, Xbox 360, Games]]   \n",
       "1  043933702X   NaN        [[Video Games, PC, Games]]   \n",
       "2  0439339987   NaN        [[Video Games, PC, Games]]   \n",
       "3  0439342260   NaN        [[Video Games, PC, Games]]   \n",
       "4  0439339960   NaN        [[Video Games, PC, Games]]   \n",
       "\n",
       "                                         description  \\\n",
       "0                                  Brand new sealed!   \n",
       "1  In Stock NOW.  Eligible for FREE Super Saving ...   \n",
       "2  Grandma Groupers kelp seeds are missing and wi...   \n",
       "3  This software is BRAND NEW. Packaging may diff...   \n",
       "4               a scholastic clubs fairs cd rom game   \n",
       "\n",
       "                                               imUrl  price  \\\n",
       "0  http://ecx.images-amazon.com/images/I/513h6dPb...  37.98   \n",
       "1  http://ecx.images-amazon.com/images/I/61KKRndV...  23.50   \n",
       "2  http://ecx.images-amazon.com/images/I/416QZg89...   8.95   \n",
       "3  http://ecx.images-amazon.com/images/I/61Wvu-Uj...    NaN   \n",
       "4  http://ecx.images-amazon.com/images/I/51k3oRCF...    NaN   \n",
       "\n",
       "                                             related               salesRank  \\\n",
       "0  {'also_bought': ['B000TI836G', 'B003Q53VZC', '...  {'Video Games': 28655}   \n",
       "1  {'also_viewed': ['B000067NP1', '0439573947', '...  {'Video Games': 44080}   \n",
       "2  {'also_bought': ['B000314VVU', 'B000PXUOTE', '...  {'Video Games': 49836}   \n",
       "3                    {'also_viewed': ['043934302X']}  {'Video Games': 49156}   \n",
       "4                    {'also_viewed': ['B00028D7TG']}  {'Video Games': 52262}   \n",
       "\n",
       "  title  \n",
       "0   NaN  \n",
       "1   NaN  \n",
       "2   NaN  \n",
       "3   NaN  \n",
       "4   NaN  "
      ]
     },
     "execution_count": 4,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "video_games_meta = pd.read_json('./data/video-games.meta.parsed.json')\n",
    "video_games_meta.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Playstation\n",
    "We try to find similar product to the playstation aka others console. We can first try to use the related column"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style>\n",
       "    .dataframe thead tr:only-child th {\n",
       "        text-align: right;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: left;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>asin</th>\n",
       "      <th>brand</th>\n",
       "      <th>categories</th>\n",
       "      <th>description</th>\n",
       "      <th>imUrl</th>\n",
       "      <th>price</th>\n",
       "      <th>related</th>\n",
       "      <th>salesRank</th>\n",
       "      <th>title</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>45933</th>\n",
       "      <td>B00BGA9WK2</td>\n",
       "      <td>NaN</td>\n",
       "      <td>[[Electronics, Television &amp; Video, Blu-ray Pla...</td>\n",
       "      <td></td>\n",
       "      <td>http://ecx.images-amazon.com/images/I/41omR-LT...</td>\n",
       "      <td>28.12</td>\n",
       "      <td>{'also_bought': ['B00BGA9X9W', 'B00CXCCI8A', '...</td>\n",
       "      <td>{'Video Games': 13}</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "             asin brand                                         categories  \\\n",
       "45933  B00BGA9WK2   NaN  [[Electronics, Television & Video, Blu-ray Pla...   \n",
       "\n",
       "      description                                              imUrl  price  \\\n",
       "45933              http://ecx.images-amazon.com/images/I/41omR-LT...  28.12   \n",
       "\n",
       "                                                 related            salesRank  \\\n",
       "45933  {'also_bought': ['B00BGA9X9W', 'B00CXCCI8A', '...  {'Video Games': 13}   \n",
       "\n",
       "      title  \n",
       "45933   NaN  "
      ]
     },
     "execution_count": 5,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "playstation = video_games_meta[video_games_meta.asin == 'B00BGA9WK2']\n",
    "playstation"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Related column\n",
    "##### Buy after viewing"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style>\n",
       "    .dataframe thead tr:only-child th {\n",
       "        text-align: right;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: left;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>asin</th>\n",
       "      <th>brand</th>\n",
       "      <th>categories</th>\n",
       "      <th>description</th>\n",
       "      <th>imUrl</th>\n",
       "      <th>price</th>\n",
       "      <th>related</th>\n",
       "      <th>salesRank</th>\n",
       "      <th>title</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>45932</th>\n",
       "      <td>B00BGA9X9W</td>\n",
       "      <td>NaN</td>\n",
       "      <td>[[Video Games, PlayStation 4, Accessories, Con...</td>\n",
       "      <td></td>\n",
       "      <td>http://ecx.images-amazon.com/images/I/41ih-c4l...</td>\n",
       "      <td>59.99</td>\n",
       "      <td>{'also_bought': ['B00BGA9WK2', 'B00CXCCI8A', '...</td>\n",
       "      <td>{'Video Games': 15}</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>47181</th>\n",
       "      <td>B00CXCCH1I</td>\n",
       "      <td>NaN</td>\n",
       "      <td>[[Video Games, PlayStation 4, Games]]</td>\n",
       "      <td>With its first new major sports league partner...</td>\n",
       "      <td>http://ecx.images-amazon.com/images/I/510xZgGR...</td>\n",
       "      <td>59.99</td>\n",
       "      <td>{'also_bought': ['B00BI83EVU', 'B00BGA9X9W', '...</td>\n",
       "      <td>{'Video Games': 102}</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "             asin brand                                         categories  \\\n",
       "45932  B00BGA9X9W   NaN  [[Video Games, PlayStation 4, Accessories, Con...   \n",
       "47181  B00CXCCH1I   NaN              [[Video Games, PlayStation 4, Games]]   \n",
       "\n",
       "                                             description  \\\n",
       "45932                                                      \n",
       "47181  With its first new major sports league partner...   \n",
       "\n",
       "                                                   imUrl  price  \\\n",
       "45932  http://ecx.images-amazon.com/images/I/41ih-c4l...  59.99   \n",
       "47181  http://ecx.images-amazon.com/images/I/510xZgGR...  59.99   \n",
       "\n",
       "                                                 related  \\\n",
       "45932  {'also_bought': ['B00BGA9WK2', 'B00CXCCI8A', '...   \n",
       "47181  {'also_bought': ['B00BI83EVU', 'B00BGA9X9W', '...   \n",
       "\n",
       "                  salesRank title  \n",
       "45932   {'Video Games': 15}   NaN  \n",
       "47181  {'Video Games': 102}   NaN  "
      ]
     },
     "execution_count": 6,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "buy_after_viewing_asin = playstation.related.values[0]['buy_after_viewing']\n",
    "buy_after_viewing = video_games_meta[video_games_meta.asin.isin(buy_after_viewing_asin)]\n",
    "buy_after_viewing"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "##### Also bought"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style>\n",
       "    .dataframe thead tr:only-child th {\n",
       "        text-align: right;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: left;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>asin</th>\n",
       "      <th>brand</th>\n",
       "      <th>categories</th>\n",
       "      <th>description</th>\n",
       "      <th>imUrl</th>\n",
       "      <th>price</th>\n",
       "      <th>related</th>\n",
       "      <th>salesRank</th>\n",
       "      <th>title</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>33598</th>\n",
       "      <td>B003NSLGW2</td>\n",
       "      <td>NaN</td>\n",
       "      <td>[[Video Games, More Systems, PlayStation 2, Ha...</td>\n",
       "      <td>Grip- iT is the newest gaming accessory that e...</td>\n",
       "      <td>http://ecx.images-amazon.com/images/I/41QPWHza...</td>\n",
       "      <td>4.99</td>\n",
       "      <td>{'also_bought': ['B00GME9XWE', 'B00HHUSNPU', '...</td>\n",
       "      <td>{'Video Games': 57}</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>44201</th>\n",
       "      <td>B009GZX49Y</td>\n",
       "      <td>NaN</td>\n",
       "      <td>[[Video Games, PlayStation 3, Games]]</td>\n",
       "      <td>Experience the series that redefined the actio...</td>\n",
       "      <td>http://ecx.images-amazon.com/images/I/61IJHp-q...</td>\n",
       "      <td>29.99</td>\n",
       "      <td>{'also_bought': ['B000P46NMA', 'B0050SXFKE', '...</td>\n",
       "      <td>{'Video Games': 681}</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>45932</th>\n",
       "      <td>B00BGA9X9W</td>\n",
       "      <td>NaN</td>\n",
       "      <td>[[Video Games, PlayStation 4, Accessories, Con...</td>\n",
       "      <td></td>\n",
       "      <td>http://ecx.images-amazon.com/images/I/41ih-c4l...</td>\n",
       "      <td>59.99</td>\n",
       "      <td>{'also_bought': ['B00BGA9WK2', 'B00CXCCI8A', '...</td>\n",
       "      <td>{'Video Games': 15}</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>45934</th>\n",
       "      <td>B00BGAA3S2</td>\n",
       "      <td>NaN</td>\n",
       "      <td>[[Toys &amp; Games, Electronics for Kids], [Video ...</td>\n",
       "      <td></td>\n",
       "      <td>http://ecx.images-amazon.com/images/I/31vC6fcF...</td>\n",
       "      <td>59.99</td>\n",
       "      <td>{'also_bought': ['B00D8S4GRY', 'B00J3Y56WG', '...</td>\n",
       "      <td>{'Video Games': 46}</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>45935</th>\n",
       "      <td>B00BGAA29M</td>\n",
       "      <td>NaN</td>\n",
       "      <td>[[Video Games, Kids &amp; Family, PlayStation 4, G...</td>\n",
       "      <td></td>\n",
       "      <td>http://ecx.images-amazon.com/images/I/51BqJurZ...</td>\n",
       "      <td>59.99</td>\n",
       "      <td>{'also_bought': ['B00DUARBTA', 'B00H5V9S6E', '...</td>\n",
       "      <td>{'Video Games': 478}</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "             asin brand                                         categories  \\\n",
       "33598  B003NSLGW2   NaN  [[Video Games, More Systems, PlayStation 2, Ha...   \n",
       "44201  B009GZX49Y   NaN              [[Video Games, PlayStation 3, Games]]   \n",
       "45932  B00BGA9X9W   NaN  [[Video Games, PlayStation 4, Accessories, Con...   \n",
       "45934  B00BGAA3S2   NaN  [[Toys & Games, Electronics for Kids], [Video ...   \n",
       "45935  B00BGAA29M   NaN  [[Video Games, Kids & Family, PlayStation 4, G...   \n",
       "\n",
       "                                             description  \\\n",
       "33598  Grip- iT is the newest gaming accessory that e...   \n",
       "44201  Experience the series that redefined the actio...   \n",
       "45932                                                      \n",
       "45934                                                      \n",
       "45935                                                      \n",
       "\n",
       "                                                   imUrl  price  \\\n",
       "33598  http://ecx.images-amazon.com/images/I/41QPWHza...   4.99   \n",
       "44201  http://ecx.images-amazon.com/images/I/61IJHp-q...  29.99   \n",
       "45932  http://ecx.images-amazon.com/images/I/41ih-c4l...  59.99   \n",
       "45934  http://ecx.images-amazon.com/images/I/31vC6fcF...  59.99   \n",
       "45935  http://ecx.images-amazon.com/images/I/51BqJurZ...  59.99   \n",
       "\n",
       "                                                 related  \\\n",
       "33598  {'also_bought': ['B00GME9XWE', 'B00HHUSNPU', '...   \n",
       "44201  {'also_bought': ['B000P46NMA', 'B0050SXFKE', '...   \n",
       "45932  {'also_bought': ['B00BGA9WK2', 'B00CXCCI8A', '...   \n",
       "45934  {'also_bought': ['B00D8S4GRY', 'B00J3Y56WG', '...   \n",
       "45935  {'also_bought': ['B00DUARBTA', 'B00H5V9S6E', '...   \n",
       "\n",
       "                  salesRank title  \n",
       "33598   {'Video Games': 57}   NaN  \n",
       "44201  {'Video Games': 681}   NaN  \n",
       "45932   {'Video Games': 15}   NaN  \n",
       "45934   {'Video Games': 46}   NaN  \n",
       "45935  {'Video Games': 478}   NaN  "
      ]
     },
     "execution_count": 7,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "also_bought_asin = playstation.related.values[0]['also_bought']\n",
    "also_bought = video_games_meta[video_games_meta.asin.isin(also_bought_asin)]\n",
    "also_bought.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "##### Bought together"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style>\n",
       "    .dataframe thead tr:only-child th {\n",
       "        text-align: right;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: left;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>asin</th>\n",
       "      <th>brand</th>\n",
       "      <th>categories</th>\n",
       "      <th>description</th>\n",
       "      <th>imUrl</th>\n",
       "      <th>price</th>\n",
       "      <th>related</th>\n",
       "      <th>salesRank</th>\n",
       "      <th>title</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>45932</th>\n",
       "      <td>B00BGA9X9W</td>\n",
       "      <td>NaN</td>\n",
       "      <td>[[Video Games, PlayStation 4, Accessories, Con...</td>\n",
       "      <td></td>\n",
       "      <td>http://ecx.images-amazon.com/images/I/41ih-c4l...</td>\n",
       "      <td>59.99</td>\n",
       "      <td>{'also_bought': ['B00BGA9WK2', 'B00CXCCI8A', '...</td>\n",
       "      <td>{'Video Games': 15}</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>47179</th>\n",
       "      <td>B00CXCCI8A</td>\n",
       "      <td>NaN</td>\n",
       "      <td>[[Video Games, Kids &amp; Family, PlayStation 4, G...</td>\n",
       "      <td>Step onto the field and immerse yourself in th...</td>\n",
       "      <td>http://ecx.images-amazon.com/images/I/51yQq3Dn...</td>\n",
       "      <td>59.99</td>\n",
       "      <td>{'also_bought': ['B00BGA9X9W', 'B00BGA9WK2', '...</td>\n",
       "      <td>{'Video Games': 51}</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "             asin brand                                         categories  \\\n",
       "45932  B00BGA9X9W   NaN  [[Video Games, PlayStation 4, Accessories, Con...   \n",
       "47179  B00CXCCI8A   NaN  [[Video Games, Kids & Family, PlayStation 4, G...   \n",
       "\n",
       "                                             description  \\\n",
       "45932                                                      \n",
       "47179  Step onto the field and immerse yourself in th...   \n",
       "\n",
       "                                                   imUrl  price  \\\n",
       "45932  http://ecx.images-amazon.com/images/I/41ih-c4l...  59.99   \n",
       "47179  http://ecx.images-amazon.com/images/I/51yQq3Dn...  59.99   \n",
       "\n",
       "                                                 related            salesRank  \\\n",
       "45932  {'also_bought': ['B00BGA9WK2', 'B00CXCCI8A', '...  {'Video Games': 15}   \n",
       "47179  {'also_bought': ['B00BGA9X9W', 'B00BGA9WK2', '...  {'Video Games': 51}   \n",
       "\n",
       "      title  \n",
       "45932   NaN  \n",
       "47179   NaN  "
      ]
     },
     "execution_count": 8,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "bought_together_asin = playstation.related.values[0]['bought_together']\n",
    "bought_together = video_games_meta[video_games_meta.asin.isin(bought_together_asin)]\n",
    "bought_together.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Seems like most of the related products are games for Playstation or accessories.\n",
    "\n",
    "#### NearestNeighbors\n",
    "We can try to use a NearestNeighbors method to find similar product. However, there is only the price column that is a number.  \n",
    "\n",
    "##### Categories\n",
    "We transform the categories to a one hot encoding format.  \n",
    "An Amazon categories is represented by a list of categories, sub categories, subsub categories ...\n",
    "\n",
    "For example `['Electronics', 'Television & Video', 'Blu-ray Players & Recorders']` or `['Video Games', 'PlayStation 4', 'Consoles']`. The leftest category is the most specific.\n",
    "\n",
    "We can encode a list `['Electronics', 'Television & Video', 'Blu-ray Players & Recorders']` as a category, but we can also consider the subcategory as a category to. So we will have `'Electronics'`, `'Television & Video'`and `'Blu-ray Players & Recorders'`. We will use this representation as it allows more combinations and makes more easier the neighbor search. \n",
    "\n",
    "For example `CategoryA -> SubCategoryZ -> SubsubCategoryA` is very similar to `CategoryA -> SubCategoryY -> SubsubCategoryA`. Only the `Subcategory` makes it different."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "['Legal',\n",
       " 'Home Office Furniture',\n",
       " 'Board Games',\n",
       " 'Personal Finance',\n",
       " 'Hand Tools',\n",
       " 'Waterskiing & Towsports',\n",
       " 'R&B',\n",
       " 'Database',\n",
       " 'Gaming Mice',\n",
       " 'TV Remote Controls']"
      ]
     },
     "execution_count": 9,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "flat_list_categories = [item for sublist in video_games_meta.categories.values for item in sublist] # Flatten the list of categories\n",
    "flat_list_categories = [item for sublist in flat_list_categories for item in sublist] # Flatten the categories, subcategories etc..\n",
    "flat_list_categories = list(set(flat_list_categories)) # Remove duplicates\n",
    "flat_list_categories[:10]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "We get 468 number of categories\n"
     ]
    }
   ],
   "source": [
    "print(\"We get\", len(flat_list_categories), \"number of categories\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "OneHotEncoder(categorical_features='all', dtype=<class 'numpy.float64'>,\n",
       "       handle_unknown='error', n_values='auto', sparse=True)"
      ]
     },
     "execution_count": 11,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "from sklearn.preprocessing import OneHotEncoder\n",
    "from sklearn import preprocessing\n",
    "label_encoder = preprocessing.LabelEncoder()\n",
    "\n",
    "# Encode category string to number\n",
    "label_encoder.fit(flat_list_categories)\n",
    "categories = label_encoder.transform(flat_list_categories)\n",
    "\n",
    "# Encode to one hot\n",
    "on_hot_encoder = OneHotEncoder()\n",
    "on_hot_encoder.fit([[cat] for cat in categories])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([ 0.,  0.,  0.,  0.,  0.,  0.,  0.,  0.,  0.,  0.,  0.,  0.,  0.,\n",
       "        0.,  0.,  0.,  0.,  0.,  0.,  0.,  0.,  0.,  0.,  0.,  0.,  0.,\n",
       "        0.,  0.,  0.,  0.,  0.,  0.,  0.,  0.,  0.,  0.,  0.,  0.,  0.,\n",
       "        0.,  0.,  0.,  0.,  0.,  0.,  0.,  0.,  0.,  0.,  0.,  0.,  0.,\n",
       "        0.,  0.,  0.,  0.,  0.,  0.,  0.,  0.,  0.,  0.,  0.,  0.,  0.,\n",
       "        0.,  0.,  0.,  0.,  0.,  0.,  0.,  0.,  0.,  0.,  0.,  0.,  0.,\n",
       "        0.,  0.,  0.,  0.,  0.,  0.,  0.,  0.,  0.,  0.,  0.,  0.,  0.,\n",
       "        0.,  0.,  0.,  0.,  0.,  0.,  0.,  0.,  0.,  0.,  0.,  0.,  0.,\n",
       "        0.,  0.,  0.,  0.,  0.,  0.,  0.,  0.,  0.,  0.,  0.,  0.,  0.,\n",
       "        0.,  0.,  0.,  0.,  0.,  0.,  0.,  0.,  0.,  0.,  0.,  0.,  0.,\n",
       "        0.,  0.,  0.,  0.,  0.,  0.,  0.,  0.,  0.,  0.,  0.,  0.,  0.,\n",
       "        0.,  0.,  0.,  0.,  0.,  0.,  0.,  0.,  0.,  0.,  0.,  0.,  0.,\n",
       "        0.,  0.,  0.,  0.,  0.,  0.,  0.,  0.,  0.,  0.,  0.,  0.,  0.,\n",
       "        0.,  0.,  0.,  0.,  0.,  0.,  0.,  0.,  0.,  0.,  0.,  0.,  0.,\n",
       "        0.,  0.,  0.,  0.,  0.,  0.,  0.,  0.,  0.,  0.,  0.,  0.,  0.,\n",
       "        0.,  0.,  0.,  0.,  0.,  0.,  0.,  0.,  0.,  0.,  0.,  0.,  0.,\n",
       "        0.,  0.,  0.,  0.,  0.,  0.,  0.,  0.,  0.,  0.,  0.,  0.,  0.,\n",
       "        0.,  0.,  1.,  0.,  0.,  0.,  0.,  0.,  0.,  0.,  0.,  0.,  0.,\n",
       "        0.,  0.,  0.,  0.,  0.,  0.,  0.,  0.,  0.,  0.,  0.,  0.,  0.,\n",
       "        0.,  0.,  0.,  0.,  0.,  0.,  0.,  0.,  0.,  0.,  0.,  0.,  0.,\n",
       "        0.,  0.,  0.,  0.,  0.,  0.,  0.,  0.,  0.,  0.,  0.,  0.,  0.,\n",
       "        0.,  0.,  0.,  0.,  0.,  0.,  0.,  0.,  0.,  0.,  0.,  0.,  0.,\n",
       "        0.,  0.,  0.,  0.,  0.,  0.,  0.,  0.,  0.,  0.,  0.,  0.,  0.,\n",
       "        0.,  0.,  0.,  0.,  0.,  0.,  0.,  0.,  0.,  0.,  0.,  0.,  0.,\n",
       "        0.,  0.,  0.,  0.,  0.,  0.,  0.,  0.,  0.,  0.,  0.,  0.,  0.,\n",
       "        0.,  0.,  0.,  0.,  0.,  0.,  0.,  0.,  0.,  0.,  0.,  0.,  0.,\n",
       "        0.,  0.,  0.,  0.,  0.,  0.,  0.,  0.,  0.,  0.,  0.,  0.,  0.,\n",
       "        0.,  0.,  0.,  0.,  0.,  0.,  0.,  0.,  0.,  0.,  0.,  0.,  0.,\n",
       "        0.,  0.,  0.,  0.,  0.,  0.,  0.,  0.,  0.,  0.,  0.,  0.,  0.,\n",
       "        0.,  0.,  0.,  0.,  0.,  0.,  0.,  0.,  0.,  0.,  0.,  0.,  0.,\n",
       "        0.,  0.,  0.,  0.,  0.,  0.,  0.,  0.,  0.,  0.,  0.,  0.,  0.,\n",
       "        0.,  0.,  0.,  0.,  0.,  0.,  0.,  0.,  0.,  0.,  0.,  0.,  0.,\n",
       "        0.,  0.,  0.,  0.,  0.,  0.,  0.,  0.,  0.,  0.,  0.,  0.,  0.,\n",
       "        0.,  0.,  0.,  0.,  0.,  0.,  0.,  0.,  0.,  0.,  0.,  0.,  0.,\n",
       "        0.,  0.,  0.,  0.,  0.,  0.,  0.,  0.,  0.,  0.,  0.,  0.,  0.,\n",
       "        0.,  0.,  0.,  0.,  0.,  0.,  0.,  0.,  0.,  0.,  0.,  0.,  0.])"
      ]
     },
     "execution_count": 12,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# The one hot encoding for the 223 category\n",
    "on_hot = np.asarray(on_hot_encoder.transform(223).todense()).reshape(-1)\n",
    "on_hot"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "0 / 43905\n",
      "4000 / 43905\n",
      "5000 / 43905\n",
      "6000 / 43905\n",
      "7000 / 43905\n",
      "8000 / 43905\n",
      "10000 / 43905\n",
      "11000 / 43905\n",
      "12000 / 43905\n",
      "14000 / 43905\n",
      "16000 / 43905\n",
      "18000 / 43905\n",
      "20000 / 43905\n",
      "22000 / 43905\n",
      "23000 / 43905\n",
      "24000 / 43905\n",
      "25000 / 43905\n",
      "26000 / 43905\n",
      "28000 / 43905\n",
      "29000 / 43905\n",
      "30000 / 43905\n",
      "31000 / 43905\n",
      "32000 / 43905\n",
      "33000 / 43905\n",
      "34000 / 43905\n",
      "35000 / 43905\n",
      "36000 / 43905\n",
      "37000 / 43905\n",
      "39000 / 43905\n",
      "40000 / 43905\n",
      "41000 / 43905\n",
      "42000 / 43905\n",
      "43000 / 43905\n",
      "44000 / 43905\n",
      "45000 / 43905\n",
      "46000 / 43905\n",
      "47000 / 43905\n",
      "48000 / 43905\n",
      "49000 / 43905\n",
      "50000 / 43905\n"
     ]
    }
   ],
   "source": [
    "def get_features_from_product_metadata(product):\n",
    "    categories_labels = []\n",
    "    for category_list in product.categories:\n",
    "        categories_labels = np.concatenate([categories_labels, label_encoder.transform(category_list)])\n",
    "    \n",
    "    categories_one_hot = np.zeros(468) # 468 categories\n",
    "    for cat_label in categories_labels:\n",
    "        categories_one_hot = categories_one_hot + on_hot_encoder.transform(cat_label)\n",
    "        \n",
    "    features = list(np.asarray(categories_one_hot[0])[0]) # Categories features\n",
    "    features.append(product.price) # Add price feature\n",
    "    return features\n",
    "\n",
    "# Create a list of products with onehot category and price features\n",
    "\n",
    "video_games_meta_dropped_price_nan = video_games_meta[np.isfinite(video_games_meta['price'])]\n",
    "\n",
    "X = []\n",
    "X_asin = []\n",
    "for i, product in video_games_meta_dropped_price_nan.iterrows():\n",
    "    features = get_features_from_product_metadata(product)\n",
    "    \n",
    "    X.append(features)\n",
    "    X_asin.append(product.asin)\n",
    "    \n",
    "    if i % 1000 == 0:\n",
    "        print(i,\"/\",len(video_games_meta_dropped_price_nan))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "We now can use these products representation to find the nearest neighbors for a given product"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Return the indices of neighbors products\n",
    "def get_nearest_neighbors_product(product, X, X_asin, n_neighbors=2):\n",
    "    nbrs = NearestNeighbors(n_neighbors=n_neighbors, algorithm='ball_tree').fit(X)\n",
    "    distances, indices = nbrs.kneighbors([get_features_from_product_metadata(product.squeeze())])\n",
    "    \n",
    "    asins = [X_asin[i] for i in indices[0]]\n",
    "    return asins"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {},
   "outputs": [],
   "source": [
    "asins = get_nearest_neighbors_product(playstation, X, X_asin)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style>\n",
       "    .dataframe thead tr:only-child th {\n",
       "        text-align: right;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: left;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>asin</th>\n",
       "      <th>brand</th>\n",
       "      <th>categories</th>\n",
       "      <th>description</th>\n",
       "      <th>imUrl</th>\n",
       "      <th>price</th>\n",
       "      <th>related</th>\n",
       "      <th>salesRank</th>\n",
       "      <th>title</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>14094</th>\n",
       "      <td>B000930DOG</td>\n",
       "      <td>NaN</td>\n",
       "      <td>[[Video Games]]</td>\n",
       "      <td>The year is 1963. Your name is Bond, James Bon...</td>\n",
       "      <td>http://ecx.images-amazon.com/images/I/515YPQDX...</td>\n",
       "      <td>27.95</td>\n",
       "      <td>{'also_bought': ['B00009X3V2', 'B00005V3EX', '...</td>\n",
       "      <td>{'Video Games': 22785}</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "             asin brand       categories  \\\n",
       "14094  B000930DOG   NaN  [[Video Games]]   \n",
       "\n",
       "                                             description  \\\n",
       "14094  The year is 1963. Your name is Bond, James Bon...   \n",
       "\n",
       "                                                   imUrl  price  \\\n",
       "14094  http://ecx.images-amazon.com/images/I/515YPQDX...  27.95   \n",
       "\n",
       "                                                 related  \\\n",
       "14094  {'also_bought': ['B00009X3V2', 'B00005V3EX', '...   \n",
       "\n",
       "                    salesRank title  \n",
       "14094  {'Video Games': 22785}   NaN  "
      ]
     },
     "execution_count": 21,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "video_games_meta[video_games_meta.asin == asins[1]]"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "The similar product obtained for the playstation console is a James bond game. We would expect to get a similar console. This is due to the playstation price of $28.12, which is probably a mistake.\n",
    "\n",
    "We try it on a Xbox game :"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "metadata": {},
   "outputs": [],
   "source": [
    "xbox_game = video_games_meta.iloc[0]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style>\n",
       "    .dataframe thead tr:only-child th {\n",
       "        text-align: right;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: left;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>asin</th>\n",
       "      <th>brand</th>\n",
       "      <th>categories</th>\n",
       "      <th>description</th>\n",
       "      <th>imUrl</th>\n",
       "      <th>price</th>\n",
       "      <th>related</th>\n",
       "      <th>salesRank</th>\n",
       "      <th>title</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>0078764343</td>\n",
       "      <td>NaN</td>\n",
       "      <td>[[Video Games, Xbox 360, Games]]</td>\n",
       "      <td>Brand new sealed!</td>\n",
       "      <td>http://ecx.images-amazon.com/images/I/513h6dPb...</td>\n",
       "      <td>37.98</td>\n",
       "      <td>{'also_bought': ['B000TI836G', 'B003Q53VZC', '...</td>\n",
       "      <td>{'Video Games': 28655}</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>24292</th>\n",
       "      <td>B00175GC4U</td>\n",
       "      <td>NaN</td>\n",
       "      <td>[[Video Games, Xbox 360, Games]]</td>\n",
       "      <td>The Number One fighting game franchise in the ...</td>\n",
       "      <td>http://ecx.images-amazon.com/images/I/51yZJeFK...</td>\n",
       "      <td>37.99</td>\n",
       "      <td>{'also_bought': ['B000FW8FK0', 'B002E1SBVS', '...</td>\n",
       "      <td>{'Video Games': 12736}</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "             asin brand                        categories  \\\n",
       "0      0078764343   NaN  [[Video Games, Xbox 360, Games]]   \n",
       "24292  B00175GC4U   NaN  [[Video Games, Xbox 360, Games]]   \n",
       "\n",
       "                                             description  \\\n",
       "0                                      Brand new sealed!   \n",
       "24292  The Number One fighting game franchise in the ...   \n",
       "\n",
       "                                                   imUrl  price  \\\n",
       "0      http://ecx.images-amazon.com/images/I/513h6dPb...  37.98   \n",
       "24292  http://ecx.images-amazon.com/images/I/51yZJeFK...  37.99   \n",
       "\n",
       "                                                 related  \\\n",
       "0      {'also_bought': ['B000TI836G', 'B003Q53VZC', '...   \n",
       "24292  {'also_bought': ['B000FW8FK0', 'B002E1SBVS', '...   \n",
       "\n",
       "                    salesRank title  \n",
       "0      {'Video Games': 28655}   NaN  \n",
       "24292  {'Video Games': 12736}   NaN  "
      ]
     },
     "execution_count": 25,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "asins = get_nearest_neighbors_product(xbox_game, X, X_asin)\n",
    "video_games_meta[video_games_meta.asin.isin(asins)]"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "We get a similar product, which stay in the same category \"Xbox360 games\"."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Get enough products to reach 10k reviews\n",
    "We start from one xbox game and get 1000 nearest neighbors"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 33,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style>\n",
       "    .dataframe thead tr:only-child th {\n",
       "        text-align: right;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: left;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>asin</th>\n",
       "      <th>brand</th>\n",
       "      <th>categories</th>\n",
       "      <th>description</th>\n",
       "      <th>imUrl</th>\n",
       "      <th>price</th>\n",
       "      <th>related</th>\n",
       "      <th>salesRank</th>\n",
       "      <th>title</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>0078764343</td>\n",
       "      <td>NaN</td>\n",
       "      <td>[[Video Games, Xbox 360, Games]]</td>\n",
       "      <td>Brand new sealed!</td>\n",
       "      <td>http://ecx.images-amazon.com/images/I/513h6dPb...</td>\n",
       "      <td>37.98</td>\n",
       "      <td>{'also_bought': ['B000TI836G', 'B003Q53VZC', '...</td>\n",
       "      <td>{'Video Games': 28655}</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>19</th>\n",
       "      <td>0545076412</td>\n",
       "      <td>NaN</td>\n",
       "      <td>[[Video Games, PC, Games]]</td>\n",
       "      <td>PROTECT, BEAUTIFY, BUILD, GROW, LIVE, AND ENTE...</td>\n",
       "      <td>http://ecx.images-amazon.com/images/I/41CQQPbp...</td>\n",
       "      <td>39.99</td>\n",
       "      <td>{'also_viewed': ['B000FHE6EY'], 'buy_after_vie...</td>\n",
       "      <td>{'Video Games': 55546}</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>24</th>\n",
       "      <td>0700026657</td>\n",
       "      <td>NaN</td>\n",
       "      <td>[[Video Games, PC, Games]]</td>\n",
       "      <td>Anno 2070, the newest version of the award-win...</td>\n",
       "      <td>http://ecx.images-amazon.com/images/I/51R06L8c...</td>\n",
       "      <td>39.99</td>\n",
       "      <td>{'also_bought': ['B005OM6CGU', 'B001AWMCMQ', '...</td>\n",
       "      <td>{'Video Games': 7621}</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>142</th>\n",
       "      <td>9882107753</td>\n",
       "      <td>NaN</td>\n",
       "      <td>[[Video Games, Xbox 360, Accessories]]</td>\n",
       "      <td>Professional and exclusively design for your X...</td>\n",
       "      <td>http://ecx.images-amazon.com/images/I/4119cGkv...</td>\n",
       "      <td>35.99</td>\n",
       "      <td>{'also_viewed': ['B00GJDJP3K'], 'buy_after_vie...</td>\n",
       "      <td>{'Video Games': 88073}</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>152</th>\n",
       "      <td>9944860212</td>\n",
       "      <td>NaN</td>\n",
       "      <td>[[Video Games, PC, Games]]</td>\n",
       "      <td>This software is BRAND NEW. Packaging may diff...</td>\n",
       "      <td>http://ecx.images-amazon.com/images/I/51P3ar3k...</td>\n",
       "      <td>39.99</td>\n",
       "      <td>{'also_viewed': ['B0007II11K', 'B00006C2HA', '...</td>\n",
       "      <td>{'Video Games': 53215}</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "           asin brand                              categories  \\\n",
       "0    0078764343   NaN        [[Video Games, Xbox 360, Games]]   \n",
       "19   0545076412   NaN              [[Video Games, PC, Games]]   \n",
       "24   0700026657   NaN              [[Video Games, PC, Games]]   \n",
       "142  9882107753   NaN  [[Video Games, Xbox 360, Accessories]]   \n",
       "152  9944860212   NaN              [[Video Games, PC, Games]]   \n",
       "\n",
       "                                           description  \\\n",
       "0                                    Brand new sealed!   \n",
       "19   PROTECT, BEAUTIFY, BUILD, GROW, LIVE, AND ENTE...   \n",
       "24   Anno 2070, the newest version of the award-win...   \n",
       "142  Professional and exclusively design for your X...   \n",
       "152  This software is BRAND NEW. Packaging may diff...   \n",
       "\n",
       "                                                 imUrl  price  \\\n",
       "0    http://ecx.images-amazon.com/images/I/513h6dPb...  37.98   \n",
       "19   http://ecx.images-amazon.com/images/I/41CQQPbp...  39.99   \n",
       "24   http://ecx.images-amazon.com/images/I/51R06L8c...  39.99   \n",
       "142  http://ecx.images-amazon.com/images/I/4119cGkv...  35.99   \n",
       "152  http://ecx.images-amazon.com/images/I/51P3ar3k...  39.99   \n",
       "\n",
       "                                               related  \\\n",
       "0    {'also_bought': ['B000TI836G', 'B003Q53VZC', '...   \n",
       "19   {'also_viewed': ['B000FHE6EY'], 'buy_after_vie...   \n",
       "24   {'also_bought': ['B005OM6CGU', 'B001AWMCMQ', '...   \n",
       "142  {'also_viewed': ['B00GJDJP3K'], 'buy_after_vie...   \n",
       "152  {'also_viewed': ['B0007II11K', 'B00006C2HA', '...   \n",
       "\n",
       "                  salesRank title  \n",
       "0    {'Video Games': 28655}   NaN  \n",
       "19   {'Video Games': 55546}   NaN  \n",
       "24    {'Video Games': 7621}   NaN  \n",
       "142  {'Video Games': 88073}   NaN  \n",
       "152  {'Video Games': 53215}   NaN  "
      ]
     },
     "execution_count": 33,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "asins = get_nearest_neighbors_product(xbox_game, X, X_asin, 1000)\n",
    "video_games_meta[video_games_meta.asin.isin(asins)].head()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Then we concatenate the reviews of the 1000 xbox video games"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 39,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Reviews count: 10610\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style>\n",
       "    .dataframe thead tr:only-child th {\n",
       "        text-align: right;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: left;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\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>overall</th>\n",
       "      <th>reviewText</th>\n",
       "      <th>reviewerName</th>\n",
       "      <th>summary</th>\n",
       "      <th>date</th>\n",
       "      <th>helpfulRate</th>\n",
       "      <th>helpfulCount</th>\n",
       "      <th>rank</th>\n",
       "      <th>description</th>\n",
       "      <th>price</th>\n",
       "      <th>title</th>\n",
       "      <th>brand</th>\n",
       "      <th>reviewLength</th>\n",
       "      <th>summaryLength</th>\n",
       "      <th>reviewerNameLength</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>asin</th>\n",
       "      <th>reviewerID</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th rowspan=\"5\" valign=\"top\">B000006RGQ</th>\n",
       "      <th>A3DABGGV3YAE2V</th>\n",
       "      <td>4</td>\n",
       "      <td>i would recommend getting this game for many r...</td>\n",
       "      <td>C. Glass</td>\n",
       "      <td>FIGHTING FORCE</td>\n",
       "      <td>2011-06-02</td>\n",
       "      <td>-1.0</td>\n",
       "      <td>0</td>\n",
       "      <td>18320</td>\n",
       "      <td>Fighting Force, from the same developers that ...</td>\n",
       "      <td>38.24</td>\n",
       "      <td></td>\n",
       "      <td></td>\n",
       "      <td>308</td>\n",
       "      <td>14</td>\n",
       "      <td>8.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>A1LBGXE29GG1IR</th>\n",
       "      <td>3</td>\n",
       "      <td>Throughout the years there have been dozens of...</td>\n",
       "      <td>Coffee Addicted Writer's Reviews \"Billy\"</td>\n",
       "      <td>Fighting Force</td>\n",
       "      <td>2012-10-14</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1</td>\n",
       "      <td>18320</td>\n",
       "      <td>Fighting Force, from the same developers that ...</td>\n",
       "      <td>38.24</td>\n",
       "      <td></td>\n",
       "      <td></td>\n",
       "      <td>208</td>\n",
       "      <td>14</td>\n",
       "      <td>40.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>A11LS12ZU93SV6</th>\n",
       "      <td>2</td>\n",
       "      <td>You know you're in trouble when other terrible...</td>\n",
       "      <td>DeRosset Myers \"DeRo\"</td>\n",
       "      <td>Pretty awful</td>\n",
       "      <td>2006-04-13</td>\n",
       "      <td>0.6</td>\n",
       "      <td>5</td>\n",
       "      <td>18320</td>\n",
       "      <td>Fighting Force, from the same developers that ...</td>\n",
       "      <td>38.24</td>\n",
       "      <td></td>\n",
       "      <td></td>\n",
       "      <td>1582</td>\n",
       "      <td>12</td>\n",
       "      <td>21.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>ASHK9LB9MU5ZI</th>\n",
       "      <td>5</td>\n",
       "      <td>This game is very fun. I have only played a de...</td>\n",
       "      <td>\"isams\"</td>\n",
       "      <td>Fighting fun!</td>\n",
       "      <td>2001-09-04</td>\n",
       "      <td>-1.0</td>\n",
       "      <td>0</td>\n",
       "      <td>18320</td>\n",
       "      <td>Fighting Force, from the same developers that ...</td>\n",
       "      <td>38.24</td>\n",
       "      <td></td>\n",
       "      <td></td>\n",
       "      <td>712</td>\n",
       "      <td>13</td>\n",
       "      <td>7.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>A2VRO53CM5EP28</th>\n",
       "      <td>5</td>\n",
       "      <td>I played this game when i was a kid and forgot...</td>\n",
       "      <td>ItsMe</td>\n",
       "      <td>Fun game</td>\n",
       "      <td>2013-01-13</td>\n",
       "      <td>-1.0</td>\n",
       "      <td>0</td>\n",
       "      <td>18320</td>\n",
       "      <td>Fighting Force, from the same developers that ...</td>\n",
       "      <td>38.24</td>\n",
       "      <td></td>\n",
       "      <td></td>\n",
       "      <td>129</td>\n",
       "      <td>8</td>\n",
       "      <td>5.0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                           overall  \\\n",
       "asin       reviewerID                \n",
       "B000006RGQ A3DABGGV3YAE2V        4   \n",
       "           A1LBGXE29GG1IR        3   \n",
       "           A11LS12ZU93SV6        2   \n",
       "           ASHK9LB9MU5ZI         5   \n",
       "           A2VRO53CM5EP28        5   \n",
       "\n",
       "                                                                  reviewText  \\\n",
       "asin       reviewerID                                                          \n",
       "B000006RGQ A3DABGGV3YAE2V  i would recommend getting this game for many r...   \n",
       "           A1LBGXE29GG1IR  Throughout the years there have been dozens of...   \n",
       "           A11LS12ZU93SV6  You know you're in trouble when other terrible...   \n",
       "           ASHK9LB9MU5ZI   This game is very fun. I have only played a de...   \n",
       "           A2VRO53CM5EP28  I played this game when i was a kid and forgot...   \n",
       "\n",
       "                                                       reviewerName  \\\n",
       "asin       reviewerID                                                 \n",
       "B000006RGQ A3DABGGV3YAE2V                                  C. Glass   \n",
       "           A1LBGXE29GG1IR  Coffee Addicted Writer's Reviews \"Billy\"   \n",
       "           A11LS12ZU93SV6                     DeRosset Myers \"DeRo\"   \n",
       "           ASHK9LB9MU5ZI                                    \"isams\"   \n",
       "           A2VRO53CM5EP28                                     ItsMe   \n",
       "\n",
       "                                  summary       date  helpfulRate  \\\n",
       "asin       reviewerID                                               \n",
       "B000006RGQ A3DABGGV3YAE2V  FIGHTING FORCE 2011-06-02         -1.0   \n",
       "           A1LBGXE29GG1IR  Fighting Force 2012-10-14          1.0   \n",
       "           A11LS12ZU93SV6    Pretty awful 2006-04-13          0.6   \n",
       "           ASHK9LB9MU5ZI    Fighting fun! 2001-09-04         -1.0   \n",
       "           A2VRO53CM5EP28        Fun game 2013-01-13         -1.0   \n",
       "\n",
       "                           helpfulCount   rank  \\\n",
       "asin       reviewerID                            \n",
       "B000006RGQ A3DABGGV3YAE2V             0  18320   \n",
       "           A1LBGXE29GG1IR             1  18320   \n",
       "           A11LS12ZU93SV6             5  18320   \n",
       "           ASHK9LB9MU5ZI              0  18320   \n",
       "           A2VRO53CM5EP28             0  18320   \n",
       "\n",
       "                                                                 description  \\\n",
       "asin       reviewerID                                                          \n",
       "B000006RGQ A3DABGGV3YAE2V  Fighting Force, from the same developers that ...   \n",
       "           A1LBGXE29GG1IR  Fighting Force, from the same developers that ...   \n",
       "           A11LS12ZU93SV6  Fighting Force, from the same developers that ...   \n",
       "           ASHK9LB9MU5ZI   Fighting Force, from the same developers that ...   \n",
       "           A2VRO53CM5EP28  Fighting Force, from the same developers that ...   \n",
       "\n",
       "                           price title brand  reviewLength  summaryLength  \\\n",
       "asin       reviewerID                                                       \n",
       "B000006RGQ A3DABGGV3YAE2V  38.24                       308             14   \n",
       "           A1LBGXE29GG1IR  38.24                       208             14   \n",
       "           A11LS12ZU93SV6  38.24                      1582             12   \n",
       "           ASHK9LB9MU5ZI   38.24                       712             13   \n",
       "           A2VRO53CM5EP28  38.24                       129              8   \n",
       "\n",
       "                           reviewerNameLength  \n",
       "asin       reviewerID                          \n",
       "B000006RGQ A3DABGGV3YAE2V                 8.0  \n",
       "           A1LBGXE29GG1IR                40.0  \n",
       "           A11LS12ZU93SV6                21.0  \n",
       "           ASHK9LB9MU5ZI                  7.0  \n",
       "           A2VRO53CM5EP28                 5.0  "
      ]
     },
     "execution_count": 39,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "xbox_games_reviews = video_games.iloc[video_games.index.get_level_values('asin').isin(asins)]\n",
    "print(\"Reviews count:\", len(xbox_games_reviews))\n",
    "xbox_games_reviews.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python [conda root]",
   "language": "python",
   "name": "conda-root-py"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.6.1"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 1
}