unsupervised_learning-Part_1.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Unsupervised Learning - section 1"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "- We are going to discuss about value of unsupervised learning to find hidden structure in the data\n",
    "- Try hands-on with market basket analysis, dimensionality reduction and clustering\n",
    "- Interpret the results and improve the model based on the results"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Unsupervised Learning is used to discover a better representation of your features\n",
    "- Goal: To learn structure hidden in your features without training labels to simplify your input variables and make them more useful\n",
    "- Application: Reduce the number of features for visualization and more effective supervised learning and identify subgroups"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Market Basket Analysis (MBA)\n",
    "- We will see how market basket analysis identifies which combination of products are often purchased together (antecedent and consequent).\n",
    "- How to find and evaluate association rules\n",
    "        - For evaluating association rules, there are 3 key metrics: Support, Confidence and Lift\n",
    "        - Support (Prevalence) : How frequent are itemsets, antecedent or consequent purchased together.\n",
    "        - Confidence (Predictability): Given the purchase of the antecedent, how likely is the purchase of the consequent\n",
    "        - Lift (Interest) : How much more likely is the relationship an association rule than we wuld expect by chance\n",
    "        \n",
    "- How to act on the resulting association rules\n",
    "- how can you apply assocoation rules beyond MBA"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Statistical Significance\n",
    "- fisher's exact p-value of association: allows you to assess the rules you have discovered are likely to generalize to the new data as opposed to the results of randomness of your sample\n",
    "- Small sample size may result in unreliable rules\n",
    "- Fisher's contigency table"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {},
   "outputs": [],
   "source": [
    "import numpy as np\n",
    "import pandas as pd\n",
    "import matplotlib.pyplot as plt\n",
    "import seaborn as sns\n",
    "%matplotlib inline\n",
    "from itertools import combinations\n",
    "from jupyterthemes import jtplot\n",
    "jtplot.style(theme='chesterish', context='talk', fscale=1.8, spines=False, gridlines='--', ticks=True, grid=False, figsize=(20, 12))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Preparing the data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 49,
   "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>0</th>\n",
       "      <th>1</th>\n",
       "      <th>2</th>\n",
       "      <th>3</th>\n",
       "      <th>4</th>\n",
       "      <th>5</th>\n",
       "      <th>6</th>\n",
       "      <th>7</th>\n",
       "      <th>8</th>\n",
       "      <th>9</th>\n",
       "      <th>10</th>\n",
       "      <th>11</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>citrus fruit</td>\n",
       "      <td>semi-finished bread</td>\n",
       "      <td>margarine</td>\n",
       "      <td>ready soups</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>tropical fruit</td>\n",
       "      <td>yogurt</td>\n",
       "      <td>coffee</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>whole milk</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>pip fruit</td>\n",
       "      <td>yogurt</td>\n",
       "      <td>cream cheese</td>\n",
       "      <td>meat spreads</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>other vegetables</td>\n",
       "      <td>whole milk</td>\n",
       "      <td>condensed milk</td>\n",
       "      <td>long life bakery product</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>whole milk</td>\n",
       "      <td>butter</td>\n",
       "      <td>yogurt</td>\n",
       "      <td>rice</td>\n",
       "      <td>abrasive cleaner</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>rolls/buns</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>other vegetables</td>\n",
       "      <td>UHT-milk</td>\n",
       "      <td>rolls/buns</td>\n",
       "      <td>bottled beer</td>\n",
       "      <td>liquor (appetizer)</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8</th>\n",
       "      <td>potted plants</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9</th>\n",
       "      <td>whole milk</td>\n",
       "      <td>cereals</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>10</th>\n",
       "      <td>tropical fruit</td>\n",
       "      <td>other vegetables</td>\n",
       "      <td>white bread</td>\n",
       "      <td>bottled water</td>\n",
       "      <td>chocolate</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>11</th>\n",
       "      <td>citrus fruit</td>\n",
       "      <td>tropical fruit</td>\n",
       "      <td>whole milk</td>\n",
       "      <td>butter</td>\n",
       "      <td>curd</td>\n",
       "      <td>yogurt</td>\n",
       "      <td>flour</td>\n",
       "      <td>bottled water</td>\n",
       "      <td>dishes</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                  0                    1               2   \\\n",
       "0       citrus fruit  semi-finished bread       margarine   \n",
       "1     tropical fruit               yogurt          coffee   \n",
       "2         whole milk                  NaN             NaN   \n",
       "3          pip fruit               yogurt    cream cheese   \n",
       "4   other vegetables           whole milk  condensed milk   \n",
       "5         whole milk               butter          yogurt   \n",
       "6         rolls/buns                  NaN             NaN   \n",
       "7   other vegetables             UHT-milk      rolls/buns   \n",
       "8      potted plants                  NaN             NaN   \n",
       "9         whole milk              cereals             NaN   \n",
       "10    tropical fruit     other vegetables     white bread   \n",
       "11      citrus fruit       tropical fruit      whole milk   \n",
       "\n",
       "                          3                   4       5      6   \\\n",
       "0                ready soups                 NaN     NaN    NaN   \n",
       "1                        NaN                 NaN     NaN    NaN   \n",
       "2                        NaN                 NaN     NaN    NaN   \n",
       "3               meat spreads                 NaN     NaN    NaN   \n",
       "4   long life bakery product                 NaN     NaN    NaN   \n",
       "5                       rice    abrasive cleaner     NaN    NaN   \n",
       "6                        NaN                 NaN     NaN    NaN   \n",
       "7               bottled beer  liquor (appetizer)     NaN    NaN   \n",
       "8                        NaN                 NaN     NaN    NaN   \n",
       "9                        NaN                 NaN     NaN    NaN   \n",
       "10             bottled water           chocolate     NaN    NaN   \n",
       "11                    butter                curd  yogurt  flour   \n",
       "\n",
       "               7       8    9    10   11  \n",
       "0             NaN     NaN  NaN  NaN  NaN  \n",
       "1             NaN     NaN  NaN  NaN  NaN  \n",
       "2             NaN     NaN  NaN  NaN  NaN  \n",
       "3             NaN     NaN  NaN  NaN  NaN  \n",
       "4             NaN     NaN  NaN  NaN  NaN  \n",
       "5             NaN     NaN  NaN  NaN  NaN  \n",
       "6             NaN     NaN  NaN  NaN  NaN  \n",
       "7             NaN     NaN  NaN  NaN  NaN  \n",
       "8             NaN     NaN  NaN  NaN  NaN  \n",
       "9             NaN     NaN  NaN  NaN  NaN  \n",
       "10            NaN     NaN  NaN  NaN  NaN  \n",
       "11  bottled water  dishes  NaN  NaN  NaN  "
      ]
     },
     "execution_count": 49,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df = pd.read_csv('.\\\\data\\\\grocery_transactions.csv', header = None)\n",
    "df.iloc[:12, :12]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 50,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "9835"
      ]
     },
     "execution_count": 50,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "len(df)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 51,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "<class 'pandas.core.frame.DataFrame'>\n",
      "RangeIndex: 9835 entries, 0 to 9834\n",
      "Data columns (total 32 columns):\n",
      "0     9835 non-null object\n",
      "1     7676 non-null object\n",
      "2     6033 non-null object\n",
      "3     4734 non-null object\n",
      "4     3729 non-null object\n",
      "5     2874 non-null object\n",
      "6     2229 non-null object\n",
      "7     1684 non-null object\n",
      "8     1246 non-null object\n",
      "9     896 non-null object\n",
      "10    650 non-null object\n",
      "11    468 non-null object\n",
      "12    351 non-null object\n",
      "13    273 non-null object\n",
      "14    196 non-null object\n",
      "15    141 non-null object\n",
      "16    95 non-null object\n",
      "17    66 non-null object\n",
      "18    52 non-null object\n",
      "19    38 non-null object\n",
      "20    29 non-null object\n",
      "21    18 non-null object\n",
      "22    14 non-null object\n",
      "23    8 non-null object\n",
      "24    7 non-null object\n",
      "25    7 non-null object\n",
      "26    6 non-null object\n",
      "27    5 non-null object\n",
      "28    4 non-null object\n",
      "29    1 non-null object\n",
      "30    1 non-null object\n",
      "31    1 non-null object\n",
      "dtypes: object(32)\n",
      "memory usage: 2.4+ MB\n"
     ]
    }
   ],
   "source": [
    "df.info()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Plot basket size"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 52,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x1923b376128>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "df.count(axis = 1).value_counts().plot.bar(title = 'basket_size')\n",
    "plt.xlabel('No. of items in basket')\n",
    "plt.ylabel('No. of Transactions')\n",
    "plt.tight_layout();"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Reshape the data to create transaction product matrix"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 53,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "trx_id  basket_id\n",
       "0       0                   citrus fruit\n",
       "        1            semi-finished bread\n",
       "        2                      margarine\n",
       "        3                    ready soups\n",
       "1       0                 tropical fruit\n",
       "        1                         yogurt\n",
       "        2                         coffee\n",
       "2       0                     whole milk\n",
       "3       0                      pip fruit\n",
       "        1                         yogurt\n",
       "dtype: object"
      ]
     },
     "execution_count": 53,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_stacked = df.stack()\n",
    "df_stacked.index.names = ['trx_id', 'basket_id']\n",
    "df_stacked.head(10)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 54,
   "metadata": {},
   "outputs": [],
   "source": [
    "df_stacked.reset_index('basket_id', drop= True, inplace = True)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Checking item frequencies"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 55,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "whole milk          2513\n",
       "other vegetables    1903\n",
       "rolls/buns          1809\n",
       "soda                1715\n",
       "yogurt              1372\n",
       "dtype: int64"
      ]
     },
     "execution_count": 55,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_stacked.value_counts().head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 56,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "bags                     4\n",
       "kitchen utensil          4\n",
       "preservation products    2\n",
       "sound storage medium     1\n",
       "baby food                1\n",
       "dtype: int64"
      ]
     },
     "execution_count": 56,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_stacked.value_counts().tail()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 57,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "169"
      ]
     },
     "execution_count": 57,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_stacked.nunique()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 58,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "<class 'pandas.core.frame.DataFrame'>\n",
      "Int64Index: 43367 entries, 0 to 9834\n",
      "Columns: 169 entries, Instant food products to zwieback\n",
      "dtypes: uint8(169)\n",
      "memory usage: 7.3 MB\n"
     ]
    }
   ],
   "source": [
    "# Transaction-product matrix\n",
    "# pd.get_dummies() will convert the categorical variables into dummy/indicator variables \n",
    "items = pd.get_dummies(df_stacked)\n",
    "items.info()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 59,
   "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>Instant food products</th>\n",
       "      <th>UHT-milk</th>\n",
       "      <th>abrasive cleaner</th>\n",
       "      <th>artif. sweetener</th>\n",
       "      <th>baby cosmetics</th>\n",
       "      <th>baby food</th>\n",
       "      <th>bags</th>\n",
       "      <th>baking powder</th>\n",
       "      <th>bathroom cleaner</th>\n",
       "      <th>beef</th>\n",
       "      <th>...</th>\n",
       "      <th>turkey</th>\n",
       "      <th>vinegar</th>\n",
       "      <th>waffles</th>\n",
       "      <th>whipped/sour cream</th>\n",
       "      <th>whisky</th>\n",
       "      <th>white bread</th>\n",
       "      <th>white wine</th>\n",
       "      <th>whole milk</th>\n",
       "      <th>yogurt</th>\n",
       "      <th>zwieback</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>trx_id</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",
       "      <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>0</th>\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>...</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",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>0</th>\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>...</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",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>0</th>\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>...</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",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>0</th>\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>...</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",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\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>...</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",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\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>...</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",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\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>...</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",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\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>...</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",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\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>...</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",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\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>...</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",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>10 rows × 169 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "        Instant food products  UHT-milk  abrasive cleaner  artif. sweetener  \\\n",
       "trx_id                                                                        \n",
       "0                           0         0                 0                 0   \n",
       "0                           0         0                 0                 0   \n",
       "0                           0         0                 0                 0   \n",
       "0                           0         0                 0                 0   \n",
       "1                           0         0                 0                 0   \n",
       "1                           0         0                 0                 0   \n",
       "1                           0         0                 0                 0   \n",
       "2                           0         0                 0                 0   \n",
       "3                           0         0                 0                 0   \n",
       "3                           0         0                 0                 0   \n",
       "\n",
       "        baby cosmetics  baby food  bags  baking powder  bathroom cleaner  \\\n",
       "trx_id                                                                     \n",
       "0                    0          0     0              0                 0   \n",
       "0                    0          0     0              0                 0   \n",
       "0                    0          0     0              0                 0   \n",
       "0                    0          0     0              0                 0   \n",
       "1                    0          0     0              0                 0   \n",
       "1                    0          0     0              0                 0   \n",
       "1                    0          0     0              0                 0   \n",
       "2                    0          0     0              0                 0   \n",
       "3                    0          0     0              0                 0   \n",
       "3                    0          0     0              0                 0   \n",
       "\n",
       "        beef    ...     turkey  vinegar  waffles  whipped/sour cream  whisky  \\\n",
       "trx_id          ...                                                            \n",
       "0          0    ...          0        0        0                   0       0   \n",
       "0          0    ...          0        0        0                   0       0   \n",
       "0          0    ...          0        0        0                   0       0   \n",
       "0          0    ...          0        0        0                   0       0   \n",
       "1          0    ...          0        0        0                   0       0   \n",
       "1          0    ...          0        0        0                   0       0   \n",
       "1          0    ...          0        0        0                   0       0   \n",
       "2          0    ...          0        0        0                   0       0   \n",
       "3          0    ...          0        0        0                   0       0   \n",
       "3          0    ...          0        0        0                   0       0   \n",
       "\n",
       "        white bread  white wine  whole milk  yogurt  zwieback  \n",
       "trx_id                                                         \n",
       "0                 0           0           0       0         0  \n",
       "0                 0           0           0       0         0  \n",
       "0                 0           0           0       0         0  \n",
       "0                 0           0           0       0         0  \n",
       "1                 0           0           0       0         0  \n",
       "1                 0           0           0       1         0  \n",
       "1                 0           0           0       0         0  \n",
       "2                 0           0           1       0         0  \n",
       "3                 0           0           0       0         0  \n",
       "3                 0           0           0       1         0  \n",
       "\n",
       "[10 rows x 169 columns]"
      ]
     },
     "execution_count": 59,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "items.head(10)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Sum indicator variables by transaction"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 60,
   "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>Instant food products</th>\n",
       "      <th>UHT-milk</th>\n",
       "      <th>abrasive cleaner</th>\n",
       "      <th>artif. sweetener</th>\n",
       "      <th>baby cosmetics</th>\n",
       "      <th>baby food</th>\n",
       "      <th>bags</th>\n",
       "      <th>baking powder</th>\n",
       "      <th>bathroom cleaner</th>\n",
       "      <th>beef</th>\n",
       "      <th>...</th>\n",
       "      <th>turkey</th>\n",
       "      <th>vinegar</th>\n",
       "      <th>waffles</th>\n",
       "      <th>whipped/sour cream</th>\n",
       "      <th>whisky</th>\n",
       "      <th>white bread</th>\n",
       "      <th>white wine</th>\n",
       "      <th>whole milk</th>\n",
       "      <th>yogurt</th>\n",
       "      <th>zwieback</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>trx_id</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",
       "      <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>0</th>\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>...</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",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\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>...</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",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\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>...</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",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\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>...</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",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\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>...</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",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\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>...</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>1</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\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>...</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",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\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>...</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",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8</th>\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>...</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",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9</th>\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>...</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",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>10 rows × 169 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "        Instant food products  UHT-milk  abrasive cleaner  artif. sweetener  \\\n",
       "trx_id                                                                        \n",
       "0                           0         0                 0                 0   \n",
       "1                           0         0                 0                 0   \n",
       "2                           0         0                 0                 0   \n",
       "3                           0         0                 0                 0   \n",
       "4                           0         0                 0                 0   \n",
       "5                           0         0                 1                 0   \n",
       "6                           0         0                 0                 0   \n",
       "7                           0         1                 0                 0   \n",
       "8                           0         0                 0                 0   \n",
       "9                           0         0                 0                 0   \n",
       "\n",
       "        baby cosmetics  baby food  bags  baking powder  bathroom cleaner  \\\n",
       "trx_id                                                                     \n",
       "0                    0          0     0              0                 0   \n",
       "1                    0          0     0              0                 0   \n",
       "2                    0          0     0              0                 0   \n",
       "3                    0          0     0              0                 0   \n",
       "4                    0          0     0              0                 0   \n",
       "5                    0          0     0              0                 0   \n",
       "6                    0          0     0              0                 0   \n",
       "7                    0          0     0              0                 0   \n",
       "8                    0          0     0              0                 0   \n",
       "9                    0          0     0              0                 0   \n",
       "\n",
       "        beef    ...     turkey  vinegar  waffles  whipped/sour cream  whisky  \\\n",
       "trx_id          ...                                                            \n",
       "0          0    ...          0        0        0                   0       0   \n",
       "1          0    ...          0        0        0                   0       0   \n",
       "2          0    ...          0        0        0                   0       0   \n",
       "3          0    ...          0        0        0                   0       0   \n",
       "4          0    ...          0        0        0                   0       0   \n",
       "5          0    ...          0        0        0                   0       0   \n",
       "6          0    ...          0        0        0                   0       0   \n",
       "7          0    ...          0        0        0                   0       0   \n",
       "8          0    ...          0        0        0                   0       0   \n",
       "9          0    ...          0        0        0                   0       0   \n",
       "\n",
       "        white bread  white wine  whole milk  yogurt  zwieback  \n",
       "trx_id                                                         \n",
       "0                 0           0           0       0         0  \n",
       "1                 0           0           0       1         0  \n",
       "2                 0           0           1       0         0  \n",
       "3                 0           0           0       1         0  \n",
       "4                 0           0           1       0         0  \n",
       "5                 0           0           1       1         0  \n",
       "6                 0           0           0       0         0  \n",
       "7                 0           0           0       0         0  \n",
       "8                 0           0           0       0         0  \n",
       "9                 0           0           1       0         0  \n",
       "\n",
       "[10 rows x 169 columns]"
      ]
     },
     "execution_count": 60,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "items = items.groupby(level = 'trx_id').sum()\n",
    "items.head(10)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Visualization for exploring items"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 61,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x1923acd80f0>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "support = items.sum().div(len(items)).sort_values(ascending = False)\n",
    "ax = support.plot.bar(title = 'Item support')\n",
    "ax.locator_params(nbins = 30, axis = 'x')\n",
    "plt.xticks(rotation = 60)\n",
    "plt.xlabel('Items')\n",
    "plt.ylabel('Support')\n",
    "plt.tight_layout();"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Unsupervised Learning - section 2"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {},
   "outputs": [],
   "source": [
    "### Writing it into a function\n",
    "def retrieve_transaction_data():\n",
    "    # Load grocery transaction into  dataframe\n",
    "    df = pd.read_csv('.\\\\data\\\\grocery_transactions.csv', header = None)\n",
    "    df = df.stack().reset_index(-1, drop = True)\n",
    "    df.index.names = ['trx_id']\n",
    "    df = pd.get_dummies(df, prefix = '', prefix_sep = '')\n",
    "    return df.groupby(level = 'trx_id').sum()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {},
   "outputs": [],
   "source": [
    "data = retrieve_transaction_data()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Create itemset candidate"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "0    Instant food products\n",
      "1                 UHT-milk\n",
      "2         abrasive cleaner\n",
      "3         artif. sweetener\n",
      "4           baby cosmetics\n",
      "dtype: object\n",
      "(9835, 169)\n"
     ]
    }
   ],
   "source": [
    "data = retrieve_transaction_data()\n",
    "item_id = pd.Series(dict(enumerate(data.columns)))\n",
    "\n",
    "print(item_id.head())\n",
    "transactions = data.values\n",
    "print(transactions.shape)\n",
    "\n",
    "min_support = 0.01\n",
    "item_length = 1\n",
    "candidates = list(zip(item_id.index))\n",
    "print(\"Candidates :\", candidates)\n",
    "candidates_tested = 0\n",
    "item_sets = pd.DataFrame(columns = ['support', 'length'])"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Candidate generation"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Length 1:        169 [(0,), (1,), (2,), (3,), (4,)]\n",
      "Length 2:     14,196 [(0, 1), (0, 2), (0, 3), (0, 4), (0, 5)]\n",
      "Length 3:    790,244 [(0, 1, 2), (0, 1, 3), (0, 1, 4), (0, 1, 5), (0, 1, 6)]\n",
      "Length 4: 32,795,126 [(0, 1, 2, 3), (0, 1, 2, 4), (0, 1, 2, 5), (0, 1, 2, 6), (0, 1, 2, 7)]\n"
     ]
    }
   ],
   "source": [
    "for i in range(1, 5):\n",
    "    remaining_items = np.unique([item for t in candidates for item in t])\n",
    "    new_candidates = list(combinations(remaining_items, r=i))\n",
    "    print('Length {}: {:>10,.0f}'.format(i, len(new_candidates)), \n",
    "          new_candidates[:5])"
   ]
  }
 ],
 "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": "3.5.4"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}