{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 52,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "import re\n",
    "import glob\n",
    "import numpy as np\n",
    "import pandas as pd\n",
    "import seaborn as sns\n",
    "import matplotlib.pyplot as plt\n",
    "%matplotlib inline\n",
    "\n",
    "pd.options.display.max_rows = 10\n",
    "plt.rcParams['figure.figsize'] = (12, 6)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "We'll use the same dataset of beer reviews."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>abv</th>\n",
       "      <th>beer_id</th>\n",
       "      <th>brewer_id</th>\n",
       "      <th>beer_name</th>\n",
       "      <th>beer_style</th>\n",
       "      <th>review_appearance</th>\n",
       "      <th>review_aroma</th>\n",
       "      <th>review_overall</th>\n",
       "      <th>review_palate</th>\n",
       "      <th>profile_name</th>\n",
       "      <th>review_taste</th>\n",
       "      <th>text</th>\n",
       "      <th>time</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>7.0</td>\n",
       "      <td>2511</td>\n",
       "      <td>287</td>\n",
       "      <td>Bell's Cherry Stout</td>\n",
       "      <td>American Stout</td>\n",
       "      <td>4.5</td>\n",
       "      <td>4</td>\n",
       "      <td>4.5</td>\n",
       "      <td>4.0</td>\n",
       "      <td>blaheath</td>\n",
       "      <td>4.5</td>\n",
       "      <td>Batch 8144\\tPitch black in color with a 1/2 f...</td>\n",
       "      <td>2009-10-05 21:31:48</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>5.7</td>\n",
       "      <td>19736</td>\n",
       "      <td>9790</td>\n",
       "      <td>Duck-Rabbit Porter</td>\n",
       "      <td>American Porter</td>\n",
       "      <td>4.5</td>\n",
       "      <td>4</td>\n",
       "      <td>4.5</td>\n",
       "      <td>4.0</td>\n",
       "      <td>GJ40</td>\n",
       "      <td>4.0</td>\n",
       "      <td>Sampled from a 12oz bottle in a standard pint...</td>\n",
       "      <td>2009-10-05 21:32:09</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>4.8</td>\n",
       "      <td>11098</td>\n",
       "      <td>3182</td>\n",
       "      <td>Fürstenberg Premium Pilsener</td>\n",
       "      <td>German Pilsener</td>\n",
       "      <td>4.0</td>\n",
       "      <td>3</td>\n",
       "      <td>3.0</td>\n",
       "      <td>3.0</td>\n",
       "      <td>biegaman</td>\n",
       "      <td>3.5</td>\n",
       "      <td>Haystack yellow with an energetic group of bu...</td>\n",
       "      <td>2009-10-05 21:32:13</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>9.5</td>\n",
       "      <td>28577</td>\n",
       "      <td>3818</td>\n",
       "      <td>Unearthly (Imperial India Pale Ale)</td>\n",
       "      <td>American Double / Imperial IPA</td>\n",
       "      <td>4.0</td>\n",
       "      <td>4</td>\n",
       "      <td>4.0</td>\n",
       "      <td>4.0</td>\n",
       "      <td>nick76</td>\n",
       "      <td>4.0</td>\n",
       "      <td>The aroma has pine, wood, citrus, caramel, an...</td>\n",
       "      <td>2009-10-05 21:32:37</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>5.8</td>\n",
       "      <td>398</td>\n",
       "      <td>119</td>\n",
       "      <td>Wolaver's Pale Ale</td>\n",
       "      <td>American Pale Ale (APA)</td>\n",
       "      <td>4.0</td>\n",
       "      <td>3</td>\n",
       "      <td>4.0</td>\n",
       "      <td>3.5</td>\n",
       "      <td>champ103</td>\n",
       "      <td>3.0</td>\n",
       "      <td>A: Pours a slightly hazy golden/orange color....</td>\n",
       "      <td>2009-10-05 21:33:14</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   abv  beer_id  brewer_id                            beer_name  \\\n",
       "0  7.0     2511        287                  Bell's Cherry Stout   \n",
       "1  5.7    19736       9790                   Duck-Rabbit Porter   \n",
       "2  4.8    11098       3182         Fürstenberg Premium Pilsener   \n",
       "3  9.5    28577       3818  Unearthly (Imperial India Pale Ale)   \n",
       "4  5.8      398        119                   Wolaver's Pale Ale   \n",
       "\n",
       "                       beer_style  review_appearance  review_aroma  \\\n",
       "0                  American Stout                4.5             4   \n",
       "1                 American Porter                4.5             4   \n",
       "2                 German Pilsener                4.0             3   \n",
       "3  American Double / Imperial IPA                4.0             4   \n",
       "4         American Pale Ale (APA)                4.0             3   \n",
       "\n",
       "   review_overall  review_palate profile_name  review_taste  \\\n",
       "0             4.5            4.0     blaheath           4.5   \n",
       "1             4.5            4.0         GJ40           4.0   \n",
       "2             3.0            3.0     biegaman           3.5   \n",
       "3             4.0            4.0       nick76           4.0   \n",
       "4             4.0            3.5     champ103           3.0   \n",
       "\n",
       "                                                text                time  \n",
       "0   Batch 8144\\tPitch black in color with a 1/2 f... 2009-10-05 21:31:48  \n",
       "1   Sampled from a 12oz bottle in a standard pint... 2009-10-05 21:32:09  \n",
       "2   Haystack yellow with an energetic group of bu... 2009-10-05 21:32:13  \n",
       "3   The aroma has pine, wood, citrus, caramel, an... 2009-10-05 21:32:37  \n",
       "4   A: Pours a slightly hazy golden/orange color.... 2009-10-05 21:33:14  "
      ]
     },
     "execution_count": 2,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df = pd.read_csv('data/beer_subset.csv.gz', parse_dates=['time'], compression='gzip')\n",
    "review_cols = ['review_appearance', 'review_aroma', 'review_overall',\n",
    "               'review_palate', 'review_taste']\n",
    "df.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "13014"
      ]
     },
     "execution_count": 3,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.beer_id.nunique()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "5990"
      ]
     },
     "execution_count": 4,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.profile_name.nunique()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAtoAAAF6CAYAAAAnGv9AAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\n298NLs8otAAAAABJRU5ErkJggg==\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x1102f3b38>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "df.beer_id.value_counts().plot(kind='hist', bins=10, color='k', log=True,\n",
    "                               title='Log reviews per beer');"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAt0AAAF6CAYAAADFxuQ5AAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nTkSuQmCC\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x10ae6aeb8>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "ax = df.brewer_id.value_counts().plot(kind='hist', bins=15, color='k', log=True,\n",
    "                                      title='Log reviews per brewer');"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAtoAAAF6CAYAAAAnGv9AAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x10995bcc0>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "df.profile_name.value_counts().plot(kind='hist', bins=15, color='k', log=True,\n",
    "                                    title='Log reviews per person');"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAs4AAAFxCAYAAABjmC4RAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\n/D9JEsaKiIwVaQAAAABJRU5ErkJggg==\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x10b092588>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "df.review_overall.value_counts().sort_index().plot(kind='bar', width=.8, rot=0);"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAVYAAAJbCAYAAAC/2bzAAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nsIqIOPb/AaXpIgDIkhjCAAAAAElFTkSuQmCC\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x10a1bdbe0>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "fig, ax = plt.subplots(figsize=(5, 10))\n",
    "sns.countplot(hue='kind', y='stars', data=(df[review_cols]\n",
    "                                           .stack()\n",
    "                                           .reset_index(level=1)\n",
    "                                           .rename(columns={'level_1': 'kind',\n",
    "                                                            0: 'stars',})),\n",
    "              ax=ax, order=np.arange(0, 5.5, .5));"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Groupby"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Groupby is a fundamental operation to pandas and data analysis.\n",
    "\n",
    "The components of a groupby operation are to\n",
    "\n",
    "1. Split a table into groups\n",
    "2. Apply a function to each groups\n",
    "3. Combine the results\n",
    "\n",
    "In pandas the first step looks like\n",
    "\n",
    "```python\n",
    "df.groupby( grouper )\n",
    "```\n",
    "\n",
    "`grouper` can be many things\n",
    "\n",
    "- Series (or string indicating a column in `df`)\n",
    "- function (to be applied on the index)\n",
    "- dict : groups by *values*\n",
    "- `levels=[]`, names of levels in a MultiIndex"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<pandas.core.groupby.DataFrameGroupBy object at 0x109d6aef0>"
      ]
     },
     "execution_count": 10,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "gr = df.groupby('beer_style')\n",
    "gr"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Haven't really done anything yet. Just some book-keeping to figure out which **keys** go with which rows. Keys are the things we've grouped by (each `beer_style` in this case).\n",
    "\n",
    "The last two steps, apply and combine, are just:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>abv</th>\n",
       "      <th>beer_id</th>\n",
       "      <th>brewer_id</th>\n",
       "      <th>review_appearance</th>\n",
       "      <th>review_aroma</th>\n",
       "      <th>review_overall</th>\n",
       "      <th>review_palate</th>\n",
       "      <th>review_taste</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>beer_style</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>Altbier</th>\n",
       "      <td>5.824369</td>\n",
       "      <td>27180.877301</td>\n",
       "      <td>4672.791411</td>\n",
       "      <td>3.819018</td>\n",
       "      <td>3.599693</td>\n",
       "      <td>3.774540</td>\n",
       "      <td>3.684049</td>\n",
       "      <td>3.696319</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>American Adjunct Lager</th>\n",
       "      <td>4.889320</td>\n",
       "      <td>9202.093117</td>\n",
       "      <td>1588.377193</td>\n",
       "      <td>2.802632</td>\n",
       "      <td>2.501012</td>\n",
       "      <td>3.088731</td>\n",
       "      <td>2.743252</td>\n",
       "      <td>2.705128</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>American Amber / Red Ale</th>\n",
       "      <td>6.224573</td>\n",
       "      <td>23036.901055</td>\n",
       "      <td>3450.877410</td>\n",
       "      <td>3.853947</td>\n",
       "      <td>3.675518</td>\n",
       "      <td>3.864496</td>\n",
       "      <td>3.712805</td>\n",
       "      <td>3.759913</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>American Amber / Red Lager</th>\n",
       "      <td>4.973025</td>\n",
       "      <td>25205.314563</td>\n",
       "      <td>7096.400000</td>\n",
       "      <td>3.566019</td>\n",
       "      <td>3.243689</td>\n",
       "      <td>3.600971</td>\n",
       "      <td>3.358252</td>\n",
       "      <td>3.408738</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>American Barleywine</th>\n",
       "      <td>10.726263</td>\n",
       "      <td>23309.185865</td>\n",
       "      <td>3049.765721</td>\n",
       "      <td>4.001948</td>\n",
       "      <td>4.000556</td>\n",
       "      <td>3.893434</td>\n",
       "      <td>3.971619</td>\n",
       "      <td>4.040345</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Vienna Lager</th>\n",
       "      <td>5.186376</td>\n",
       "      <td>14034.549348</td>\n",
       "      <td>2368.122905</td>\n",
       "      <td>3.766294</td>\n",
       "      <td>3.459963</td>\n",
       "      <td>3.837989</td>\n",
       "      <td>3.620112</td>\n",
       "      <td>3.634078</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Weizenbock</th>\n",
       "      <td>7.889529</td>\n",
       "      <td>27997.852252</td>\n",
       "      <td>1618.282883</td>\n",
       "      <td>4.027928</td>\n",
       "      <td>4.041441</td>\n",
       "      <td>4.084685</td>\n",
       "      <td>4.068468</td>\n",
       "      <td>4.111712</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Wheatwine</th>\n",
       "      <td>11.072160</td>\n",
       "      <td>46881.403930</td>\n",
       "      <td>2000.454148</td>\n",
       "      <td>3.945415</td>\n",
       "      <td>4.056769</td>\n",
       "      <td>3.813319</td>\n",
       "      <td>4.004367</td>\n",
       "      <td>4.066594</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Winter Warmer</th>\n",
       "      <td>6.573012</td>\n",
       "      <td>22849.504034</td>\n",
       "      <td>2661.158994</td>\n",
       "      <td>3.849075</td>\n",
       "      <td>3.716659</td>\n",
       "      <td>3.759848</td>\n",
       "      <td>3.694115</td>\n",
       "      <td>3.734457</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Witbier</th>\n",
       "      <td>5.902119</td>\n",
       "      <td>28802.044147</td>\n",
       "      <td>3107.098328</td>\n",
       "      <td>3.647826</td>\n",
       "      <td>3.631773</td>\n",
       "      <td>3.755853</td>\n",
       "      <td>3.598328</td>\n",
       "      <td>3.631438</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>104 rows × 8 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "                                  abv       beer_id    brewer_id  \\\n",
       "beer_style                                                         \n",
       "Altbier                      5.824369  27180.877301  4672.791411   \n",
       "American Adjunct Lager       4.889320   9202.093117  1588.377193   \n",
       "American Amber / Red Ale     6.224573  23036.901055  3450.877410   \n",
       "American Amber / Red Lager   4.973025  25205.314563  7096.400000   \n",
       "American Barleywine         10.726263  23309.185865  3049.765721   \n",
       "...                               ...           ...          ...   \n",
       "Vienna Lager                 5.186376  14034.549348  2368.122905   \n",
       "Weizenbock                   7.889529  27997.852252  1618.282883   \n",
       "Wheatwine                   11.072160  46881.403930  2000.454148   \n",
       "Winter Warmer                6.573012  22849.504034  2661.158994   \n",
       "Witbier                      5.902119  28802.044147  3107.098328   \n",
       "\n",
       "                            review_appearance  review_aroma  review_overall  \\\n",
       "beer_style                                                                    \n",
       "Altbier                              3.819018      3.599693        3.774540   \n",
       "American Adjunct Lager               2.802632      2.501012        3.088731   \n",
       "American Amber / Red Ale             3.853947      3.675518        3.864496   \n",
       "American Amber / Red Lager           3.566019      3.243689        3.600971   \n",
       "American Barleywine                  4.001948      4.000556        3.893434   \n",
       "...                                       ...           ...             ...   \n",
       "Vienna Lager                         3.766294      3.459963        3.837989   \n",
       "Weizenbock                           4.027928      4.041441        4.084685   \n",
       "Wheatwine                            3.945415      4.056769        3.813319   \n",
       "Winter Warmer                        3.849075      3.716659        3.759848   \n",
       "Witbier                              3.647826      3.631773        3.755853   \n",
       "\n",
       "                            review_palate  review_taste  \n",
       "beer_style                                               \n",
       "Altbier                          3.684049      3.696319  \n",
       "American Adjunct Lager           2.743252      2.705128  \n",
       "American Amber / Red Ale         3.712805      3.759913  \n",
       "American Amber / Red Lager       3.358252      3.408738  \n",
       "American Barleywine              3.971619      4.040345  \n",
       "...                                   ...           ...  \n",
       "Vienna Lager                     3.620112      3.634078  \n",
       "Weizenbock                       4.068468      4.111712  \n",
       "Wheatwine                        4.004367      4.066594  \n",
       "Winter Warmer                    3.694115      3.734457  \n",
       "Witbier                          3.598328      3.631438  \n",
       "\n",
       "[104 rows x 8 columns]"
      ]
     },
     "execution_count": 11,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "gr.agg('mean')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "This says apply the `mean` function to each column. Non-numeric columns (nusiance columns) are excluded. We can also select a subset of columns to perform the aggregation on."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>review_appearance</th>\n",
       "      <th>review_aroma</th>\n",
       "      <th>review_overall</th>\n",
       "      <th>review_palate</th>\n",
       "      <th>review_taste</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>beer_style</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>Altbier</th>\n",
       "      <td>3.819018</td>\n",
       "      <td>3.599693</td>\n",
       "      <td>3.774540</td>\n",
       "      <td>3.684049</td>\n",
       "      <td>3.696319</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>American Adjunct Lager</th>\n",
       "      <td>2.802632</td>\n",
       "      <td>2.501012</td>\n",
       "      <td>3.088731</td>\n",
       "      <td>2.743252</td>\n",
       "      <td>2.705128</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>American Amber / Red Ale</th>\n",
       "      <td>3.853947</td>\n",
       "      <td>3.675518</td>\n",
       "      <td>3.864496</td>\n",
       "      <td>3.712805</td>\n",
       "      <td>3.759913</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>American Amber / Red Lager</th>\n",
       "      <td>3.566019</td>\n",
       "      <td>3.243689</td>\n",
       "      <td>3.600971</td>\n",
       "      <td>3.358252</td>\n",
       "      <td>3.408738</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>American Barleywine</th>\n",
       "      <td>4.001948</td>\n",
       "      <td>4.000556</td>\n",
       "      <td>3.893434</td>\n",
       "      <td>3.971619</td>\n",
       "      <td>4.040345</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Vienna Lager</th>\n",
       "      <td>3.766294</td>\n",
       "      <td>3.459963</td>\n",
       "      <td>3.837989</td>\n",
       "      <td>3.620112</td>\n",
       "      <td>3.634078</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Weizenbock</th>\n",
       "      <td>4.027928</td>\n",
       "      <td>4.041441</td>\n",
       "      <td>4.084685</td>\n",
       "      <td>4.068468</td>\n",
       "      <td>4.111712</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Wheatwine</th>\n",
       "      <td>3.945415</td>\n",
       "      <td>4.056769</td>\n",
       "      <td>3.813319</td>\n",
       "      <td>4.004367</td>\n",
       "      <td>4.066594</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Winter Warmer</th>\n",
       "      <td>3.849075</td>\n",
       "      <td>3.716659</td>\n",
       "      <td>3.759848</td>\n",
       "      <td>3.694115</td>\n",
       "      <td>3.734457</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Witbier</th>\n",
       "      <td>3.647826</td>\n",
       "      <td>3.631773</td>\n",
       "      <td>3.755853</td>\n",
       "      <td>3.598328</td>\n",
       "      <td>3.631438</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>104 rows × 5 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "                            review_appearance  review_aroma  review_overall  \\\n",
       "beer_style                                                                    \n",
       "Altbier                              3.819018      3.599693        3.774540   \n",
       "American Adjunct Lager               2.802632      2.501012        3.088731   \n",
       "American Amber / Red Ale             3.853947      3.675518        3.864496   \n",
       "American Amber / Red Lager           3.566019      3.243689        3.600971   \n",
       "American Barleywine                  4.001948      4.000556        3.893434   \n",
       "...                                       ...           ...             ...   \n",
       "Vienna Lager                         3.766294      3.459963        3.837989   \n",
       "Weizenbock                           4.027928      4.041441        4.084685   \n",
       "Wheatwine                            3.945415      4.056769        3.813319   \n",
       "Winter Warmer                        3.849075      3.716659        3.759848   \n",
       "Witbier                              3.647826      3.631773        3.755853   \n",
       "\n",
       "                            review_palate  review_taste  \n",
       "beer_style                                               \n",
       "Altbier                          3.684049      3.696319  \n",
       "American Adjunct Lager           2.743252      2.705128  \n",
       "American Amber / Red Ale         3.712805      3.759913  \n",
       "American Amber / Red Lager       3.358252      3.408738  \n",
       "American Barleywine              3.971619      4.040345  \n",
       "...                                   ...           ...  \n",
       "Vienna Lager                     3.620112      3.634078  \n",
       "Weizenbock                       4.068468      4.111712  \n",
       "Wheatwine                        4.004367      4.066594  \n",
       "Winter Warmer                    3.694115      3.734457  \n",
       "Witbier                          3.598328      3.631438  \n",
       "\n",
       "[104 rows x 5 columns]"
      ]
     },
     "execution_count": 12,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "gr[review_cols].agg('mean')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "`.` attribute lookup works as well."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "beer_style\n",
       "Altbier                        5.824369\n",
       "American Adjunct Lager         4.889320\n",
       "American Amber / Red Ale       6.224573\n",
       "American Amber / Red Lager     4.973025\n",
       "American Barleywine           10.726263\n",
       "                                ...    \n",
       "Vienna Lager                   5.186376\n",
       "Weizenbock                     7.889529\n",
       "Wheatwine                     11.072160\n",
       "Winter Warmer                  6.573012\n",
       "Witbier                        5.902119\n",
       "Name: abv, dtype: float64"
      ]
     },
     "execution_count": 13,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "gr.abv.agg('mean')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Certain operations are attached directly to the `GroupBy` object, letting you bypass the `.agg` part"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "beer_style\n",
       "Altbier                        5.824369\n",
       "American Adjunct Lager         4.889320\n",
       "American Amber / Red Ale       6.224573\n",
       "American Amber / Red Lager     4.973025\n",
       "American Barleywine           10.726263\n",
       "                                ...    \n",
       "Vienna Lager                   5.186376\n",
       "Weizenbock                     7.889529\n",
       "Wheatwine                     11.072160\n",
       "Winter Warmer                  6.573012\n",
       "Witbier                        5.902119\n",
       "Name: abv, dtype: float64"
      ]
     },
     "execution_count": 14,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "gr.abv.mean()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Exercise: Find the `beer_style`s with the greatest variance in `abv`.\n",
    "\n",
    "- hint: `.std` calculates the standard deviation, and is available on `GroupBy` objects like `gr.abv`.\n",
    "- hint: use `.order` to sort a Series"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 87,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "# your code goes here"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 89,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "beer_style\n",
       "American Strong Ale                 3.492789\n",
       "American Double / Imperial Stout    3.006386\n",
       "Eisbock                             2.723260\n",
       "Fruit / Vegetable Beer              2.647306\n",
       "Flanders Oud Bruin                  2.527305\n",
       "                                      ...   \n",
       "Kristalweizen                       0.318685\n",
       "Czech Pilsener                      0.316302\n",
       "Roggenbier                          0.281707\n",
       "Low Alcohol Beer                    0.219064\n",
       "Happoshu                            0.000000\n",
       "Name: abv, dtype: float64"
      ]
     },
     "execution_count": 89,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "%load -r 15:17 solutions_groupby.py"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Now we'll run the gamut on a bunch of grouper / apply combinations.\n",
    "Keep sight of the target though: split, apply, combine.\n",
    "\n",
    "Single grouper, multiple aggregtaions:"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Multiple Aggregations on one column"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>mean</th>\n",
       "      <th>std</th>\n",
       "      <th>count</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>beer_style</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>Altbier</th>\n",
       "      <td>3.599693</td>\n",
       "      <td>0.568426</td>\n",
       "      <td>326</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>American Adjunct Lager</th>\n",
       "      <td>2.501012</td>\n",
       "      <td>0.720936</td>\n",
       "      <td>1482</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>American Amber / Red Ale</th>\n",
       "      <td>3.675518</td>\n",
       "      <td>0.626926</td>\n",
       "      <td>2749</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>American Amber / Red Lager</th>\n",
       "      <td>3.243689</td>\n",
       "      <td>0.613034</td>\n",
       "      <td>515</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>American Barleywine</th>\n",
       "      <td>4.000556</td>\n",
       "      <td>0.514814</td>\n",
       "      <td>1797</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                                mean       std  count\n",
       "beer_style                                           \n",
       "Altbier                     3.599693  0.568426    326\n",
       "American Adjunct Lager      2.501012  0.720936   1482\n",
       "American Amber / Red Ale    3.675518  0.626926   2749\n",
       "American Amber / Red Lager  3.243689  0.613034    515\n",
       "American Barleywine         4.000556  0.514814   1797"
      ]
     },
     "execution_count": 15,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "gr['review_aroma'].agg([np.mean, np.std, 'count']).head()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Single Aggregation on multiple columns"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>review_appearance</th>\n",
       "      <th>review_aroma</th>\n",
       "      <th>review_overall</th>\n",
       "      <th>review_palate</th>\n",
       "      <th>review_taste</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>beer_style</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>Altbier</th>\n",
       "      <td>3.819018</td>\n",
       "      <td>3.599693</td>\n",
       "      <td>3.774540</td>\n",
       "      <td>3.684049</td>\n",
       "      <td>3.696319</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>American Adjunct Lager</th>\n",
       "      <td>2.802632</td>\n",
       "      <td>2.501012</td>\n",
       "      <td>3.088731</td>\n",
       "      <td>2.743252</td>\n",
       "      <td>2.705128</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>American Amber / Red Ale</th>\n",
       "      <td>3.853947</td>\n",
       "      <td>3.675518</td>\n",
       "      <td>3.864496</td>\n",
       "      <td>3.712805</td>\n",
       "      <td>3.759913</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>American Amber / Red Lager</th>\n",
       "      <td>3.566019</td>\n",
       "      <td>3.243689</td>\n",
       "      <td>3.600971</td>\n",
       "      <td>3.358252</td>\n",
       "      <td>3.408738</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>American Barleywine</th>\n",
       "      <td>4.001948</td>\n",
       "      <td>4.000556</td>\n",
       "      <td>3.893434</td>\n",
       "      <td>3.971619</td>\n",
       "      <td>4.040345</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Vienna Lager</th>\n",
       "      <td>3.766294</td>\n",
       "      <td>3.459963</td>\n",
       "      <td>3.837989</td>\n",
       "      <td>3.620112</td>\n",
       "      <td>3.634078</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Weizenbock</th>\n",
       "      <td>4.027928</td>\n",
       "      <td>4.041441</td>\n",
       "      <td>4.084685</td>\n",
       "      <td>4.068468</td>\n",
       "      <td>4.111712</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Wheatwine</th>\n",
       "      <td>3.945415</td>\n",
       "      <td>4.056769</td>\n",
       "      <td>3.813319</td>\n",
       "      <td>4.004367</td>\n",
       "      <td>4.066594</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Winter Warmer</th>\n",
       "      <td>3.849075</td>\n",
       "      <td>3.716659</td>\n",
       "      <td>3.759848</td>\n",
       "      <td>3.694115</td>\n",
       "      <td>3.734457</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Witbier</th>\n",
       "      <td>3.647826</td>\n",
       "      <td>3.631773</td>\n",
       "      <td>3.755853</td>\n",
       "      <td>3.598328</td>\n",
       "      <td>3.631438</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>104 rows × 5 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "                            review_appearance  review_aroma  review_overall  \\\n",
       "beer_style                                                                    \n",
       "Altbier                              3.819018      3.599693        3.774540   \n",
       "American Adjunct Lager               2.802632      2.501012        3.088731   \n",
       "American Amber / Red Ale             3.853947      3.675518        3.864496   \n",
       "American Amber / Red Lager           3.566019      3.243689        3.600971   \n",
       "American Barleywine                  4.001948      4.000556        3.893434   \n",
       "...                                       ...           ...             ...   \n",
       "Vienna Lager                         3.766294      3.459963        3.837989   \n",
       "Weizenbock                           4.027928      4.041441        4.084685   \n",
       "Wheatwine                            3.945415      4.056769        3.813319   \n",
       "Winter Warmer                        3.849075      3.716659        3.759848   \n",
       "Witbier                              3.647826      3.631773        3.755853   \n",
       "\n",
       "                            review_palate  review_taste  \n",
       "beer_style                                               \n",
       "Altbier                          3.684049      3.696319  \n",
       "American Adjunct Lager           2.743252      2.705128  \n",
       "American Amber / Red Ale         3.712805      3.759913  \n",
       "American Amber / Red Lager       3.358252      3.408738  \n",
       "American Barleywine              3.971619      4.040345  \n",
       "...                                   ...           ...  \n",
       "Vienna Lager                     3.620112      3.634078  \n",
       "Weizenbock                       4.068468      4.111712  \n",
       "Wheatwine                        4.004367      4.066594  \n",
       "Winter Warmer                    3.694115      3.734457  \n",
       "Witbier                          3.598328      3.631438  \n",
       "\n",
       "[104 rows x 5 columns]"
      ]
     },
     "execution_count": 16,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "gr[review_cols].mean()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "collapsed": true
   },
   "source": [
    "Multiple aggregations on multiple columns"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr>\n",
       "      <th></th>\n",
       "      <th colspan=\"3\" halign=\"left\">review_appearance</th>\n",
       "      <th colspan=\"3\" halign=\"left\">review_aroma</th>\n",
       "      <th colspan=\"3\" halign=\"left\">review_overall</th>\n",
       "      <th colspan=\"3\" halign=\"left\">review_palate</th>\n",
       "      <th colspan=\"3\" halign=\"left\">review_taste</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th></th>\n",
       "      <th>mean</th>\n",
       "      <th>count</th>\n",
       "      <th>std</th>\n",
       "      <th>mean</th>\n",
       "      <th>count</th>\n",
       "      <th>std</th>\n",
       "      <th>mean</th>\n",
       "      <th>count</th>\n",
       "      <th>std</th>\n",
       "      <th>mean</th>\n",
       "      <th>count</th>\n",
       "      <th>std</th>\n",
       "      <th>mean</th>\n",
       "      <th>count</th>\n",
       "      <th>std</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>beer_style</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>Altbier</th>\n",
       "      <td>3.819018</td>\n",
       "      <td>326</td>\n",
       "      <td>0.559731</td>\n",
       "      <td>3.599693</td>\n",
       "      <td>326</td>\n",
       "      <td>0.568426</td>\n",
       "      <td>3.774540</td>\n",
       "      <td>326</td>\n",
       "      <td>0.676367</td>\n",
       "      <td>3.684049</td>\n",
       "      <td>326</td>\n",
       "      <td>0.601171</td>\n",
       "      <td>3.696319</td>\n",
       "      <td>326</td>\n",
       "      <td>0.624957</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>American Adjunct Lager</th>\n",
       "      <td>2.802632</td>\n",
       "      <td>1482</td>\n",
       "      <td>0.750881</td>\n",
       "      <td>2.501012</td>\n",
       "      <td>1482</td>\n",
       "      <td>0.720936</td>\n",
       "      <td>3.088731</td>\n",
       "      <td>1482</td>\n",
       "      <td>0.961824</td>\n",
       "      <td>2.743252</td>\n",
       "      <td>1482</td>\n",
       "      <td>0.777189</td>\n",
       "      <td>2.705128</td>\n",
       "      <td>1482</td>\n",
       "      <td>0.813102</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>American Amber / Red Ale</th>\n",
       "      <td>3.853947</td>\n",
       "      <td>2749</td>\n",
       "      <td>0.532465</td>\n",
       "      <td>3.675518</td>\n",
       "      <td>2749</td>\n",
       "      <td>0.626926</td>\n",
       "      <td>3.864496</td>\n",
       "      <td>2749</td>\n",
       "      <td>0.649929</td>\n",
       "      <td>3.712805</td>\n",
       "      <td>2749</td>\n",
       "      <td>0.592133</td>\n",
       "      <td>3.759913</td>\n",
       "      <td>2749</td>\n",
       "      <td>0.651262</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>American Amber / Red Lager</th>\n",
       "      <td>3.566019</td>\n",
       "      <td>515</td>\n",
       "      <td>0.601159</td>\n",
       "      <td>3.243689</td>\n",
       "      <td>515</td>\n",
       "      <td>0.613034</td>\n",
       "      <td>3.600971</td>\n",
       "      <td>515</td>\n",
       "      <td>0.768101</td>\n",
       "      <td>3.358252</td>\n",
       "      <td>515</td>\n",
       "      <td>0.624799</td>\n",
       "      <td>3.408738</td>\n",
       "      <td>515</td>\n",
       "      <td>0.699792</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>American Barleywine</th>\n",
       "      <td>4.001948</td>\n",
       "      <td>1797</td>\n",
       "      <td>0.495381</td>\n",
       "      <td>4.000556</td>\n",
       "      <td>1797</td>\n",
       "      <td>0.514814</td>\n",
       "      <td>3.893434</td>\n",
       "      <td>1797</td>\n",
       "      <td>0.594070</td>\n",
       "      <td>3.971619</td>\n",
       "      <td>1797</td>\n",
       "      <td>0.538368</td>\n",
       "      <td>4.040345</td>\n",
       "      <td>1797</td>\n",
       "      <td>0.555808</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Vienna Lager</th>\n",
       "      <td>3.766294</td>\n",
       "      <td>537</td>\n",
       "      <td>0.584182</td>\n",
       "      <td>3.459963</td>\n",
       "      <td>537</td>\n",
       "      <td>0.620528</td>\n",
       "      <td>3.837989</td>\n",
       "      <td>537</td>\n",
       "      <td>0.702348</td>\n",
       "      <td>3.620112</td>\n",
       "      <td>537</td>\n",
       "      <td>0.619194</td>\n",
       "      <td>3.634078</td>\n",
       "      <td>537</td>\n",
       "      <td>0.667539</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Weizenbock</th>\n",
       "      <td>4.027928</td>\n",
       "      <td>555</td>\n",
       "      <td>0.510392</td>\n",
       "      <td>4.041441</td>\n",
       "      <td>555</td>\n",
       "      <td>0.481699</td>\n",
       "      <td>4.084685</td>\n",
       "      <td>555</td>\n",
       "      <td>0.586429</td>\n",
       "      <td>4.068468</td>\n",
       "      <td>555</td>\n",
       "      <td>0.511419</td>\n",
       "      <td>4.111712</td>\n",
       "      <td>555</td>\n",
       "      <td>0.518735</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Wheatwine</th>\n",
       "      <td>3.945415</td>\n",
       "      <td>458</td>\n",
       "      <td>0.493138</td>\n",
       "      <td>4.056769</td>\n",
       "      <td>458</td>\n",
       "      <td>0.548569</td>\n",
       "      <td>3.813319</td>\n",
       "      <td>458</td>\n",
       "      <td>0.650347</td>\n",
       "      <td>4.004367</td>\n",
       "      <td>458</td>\n",
       "      <td>0.600859</td>\n",
       "      <td>4.066594</td>\n",
       "      <td>458</td>\n",
       "      <td>0.612987</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Winter Warmer</th>\n",
       "      <td>3.849075</td>\n",
       "      <td>2107</td>\n",
       "      <td>0.493310</td>\n",
       "      <td>3.716659</td>\n",
       "      <td>2107</td>\n",
       "      <td>0.571915</td>\n",
       "      <td>3.759848</td>\n",
       "      <td>2107</td>\n",
       "      <td>0.631923</td>\n",
       "      <td>3.694115</td>\n",
       "      <td>2107</td>\n",
       "      <td>0.576365</td>\n",
       "      <td>3.734457</td>\n",
       "      <td>2107</td>\n",
       "      <td>0.627473</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Witbier</th>\n",
       "      <td>3.647826</td>\n",
       "      <td>1495</td>\n",
       "      <td>0.591286</td>\n",
       "      <td>3.631773</td>\n",
       "      <td>1495</td>\n",
       "      <td>0.599275</td>\n",
       "      <td>3.755853</td>\n",
       "      <td>1495</td>\n",
       "      <td>0.717391</td>\n",
       "      <td>3.598328</td>\n",
       "      <td>1495</td>\n",
       "      <td>0.622560</td>\n",
       "      <td>3.631438</td>\n",
       "      <td>1495</td>\n",
       "      <td>0.674364</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>104 rows × 15 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "                           review_appearance                 review_aroma  \\\n",
       "                                        mean count       std         mean   \n",
       "beer_style                                                                  \n",
       "Altbier                             3.819018   326  0.559731     3.599693   \n",
       "American Adjunct Lager              2.802632  1482  0.750881     2.501012   \n",
       "American Amber / Red Ale            3.853947  2749  0.532465     3.675518   \n",
       "American Amber / Red Lager          3.566019   515  0.601159     3.243689   \n",
       "American Barleywine                 4.001948  1797  0.495381     4.000556   \n",
       "...                                      ...   ...       ...          ...   \n",
       "Vienna Lager                        3.766294   537  0.584182     3.459963   \n",
       "Weizenbock                          4.027928   555  0.510392     4.041441   \n",
       "Wheatwine                           3.945415   458  0.493138     4.056769   \n",
       "Winter Warmer                       3.849075  2107  0.493310     3.716659   \n",
       "Witbier                             3.647826  1495  0.591286     3.631773   \n",
       "\n",
       "                                           review_overall                  \\\n",
       "                           count       std           mean count       std   \n",
       "beer_style                                                                  \n",
       "Altbier                      326  0.568426       3.774540   326  0.676367   \n",
       "American Adjunct Lager      1482  0.720936       3.088731  1482  0.961824   \n",
       "American Amber / Red Ale    2749  0.626926       3.864496  2749  0.649929   \n",
       "American Amber / Red Lager   515  0.613034       3.600971   515  0.768101   \n",
       "American Barleywine         1797  0.514814       3.893434  1797  0.594070   \n",
       "...                          ...       ...            ...   ...       ...   \n",
       "Vienna Lager                 537  0.620528       3.837989   537  0.702348   \n",
       "Weizenbock                   555  0.481699       4.084685   555  0.586429   \n",
       "Wheatwine                    458  0.548569       3.813319   458  0.650347   \n",
       "Winter Warmer               2107  0.571915       3.759848  2107  0.631923   \n",
       "Witbier                     1495  0.599275       3.755853  1495  0.717391   \n",
       "\n",
       "                           review_palate                 review_taste        \\\n",
       "                                    mean count       std         mean count   \n",
       "beer_style                                                                    \n",
       "Altbier                         3.684049   326  0.601171     3.696319   326   \n",
       "American Adjunct Lager          2.743252  1482  0.777189     2.705128  1482   \n",
       "American Amber / Red Ale        3.712805  2749  0.592133     3.759913  2749   \n",
       "American Amber / Red Lager      3.358252   515  0.624799     3.408738   515   \n",
       "American Barleywine             3.971619  1797  0.538368     4.040345  1797   \n",
       "...                                  ...   ...       ...          ...   ...   \n",
       "Vienna Lager                    3.620112   537  0.619194     3.634078   537   \n",
       "Weizenbock                      4.068468   555  0.511419     4.111712   555   \n",
       "Wheatwine                       4.004367   458  0.600859     4.066594   458   \n",
       "Winter Warmer                   3.694115  2107  0.576365     3.734457  2107   \n",
       "Witbier                         3.598328  1495  0.622560     3.631438  1495   \n",
       "\n",
       "                                      \n",
       "                                 std  \n",
       "beer_style                            \n",
       "Altbier                     0.624957  \n",
       "American Adjunct Lager      0.813102  \n",
       "American Amber / Red Ale    0.651262  \n",
       "American Amber / Red Lager  0.699792  \n",
       "American Barleywine         0.555808  \n",
       "...                              ...  \n",
       "Vienna Lager                0.667539  \n",
       "Weizenbock                  0.518735  \n",
       "Wheatwine                   0.612987  \n",
       "Winter Warmer               0.627473  \n",
       "Witbier                     0.674364  \n",
       "\n",
       "[104 rows x 15 columns]"
      ]
     },
     "execution_count": 17,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "gr[review_cols].agg(['mean', 'count', 'std'])"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Hierarchical Indexes in the columns can be awkward to work with, so I'll usually\n",
    "move a level to the Index with `.stack`."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th>mean</th>\n",
       "      <th>count</th>\n",
       "      <th>std</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>beer_style</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\">Altbier</th>\n",
       "      <th>review_appearance</th>\n",
       "      <td>3.819018</td>\n",
       "      <td>326</td>\n",
       "      <td>0.559731</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>review_aroma</th>\n",
       "      <td>3.599693</td>\n",
       "      <td>326</td>\n",
       "      <td>0.568426</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>review_overall</th>\n",
       "      <td>3.774540</td>\n",
       "      <td>326</td>\n",
       "      <td>0.676367</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>review_palate</th>\n",
       "      <td>3.684049</td>\n",
       "      <td>326</td>\n",
       "      <td>0.601171</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>review_taste</th>\n",
       "      <td>3.696319</td>\n",
       "      <td>326</td>\n",
       "      <td>0.624957</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th rowspan=\"5\" valign=\"top\">Witbier</th>\n",
       "      <th>review_appearance</th>\n",
       "      <td>3.647826</td>\n",
       "      <td>1495</td>\n",
       "      <td>0.591286</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>review_aroma</th>\n",
       "      <td>3.631773</td>\n",
       "      <td>1495</td>\n",
       "      <td>0.599275</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>review_overall</th>\n",
       "      <td>3.755853</td>\n",
       "      <td>1495</td>\n",
       "      <td>0.717391</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>review_palate</th>\n",
       "      <td>3.598328</td>\n",
       "      <td>1495</td>\n",
       "      <td>0.622560</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>review_taste</th>\n",
       "      <td>3.631438</td>\n",
       "      <td>1495</td>\n",
       "      <td>0.674364</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>520 rows × 3 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "                                  mean  count       std\n",
       "beer_style                                             \n",
       "Altbier    review_appearance  3.819018    326  0.559731\n",
       "           review_aroma       3.599693    326  0.568426\n",
       "           review_overall     3.774540    326  0.676367\n",
       "           review_palate      3.684049    326  0.601171\n",
       "           review_taste       3.696319    326  0.624957\n",
       "...                                ...    ...       ...\n",
       "Witbier    review_appearance  3.647826   1495  0.591286\n",
       "           review_aroma       3.631773   1495  0.599275\n",
       "           review_overall     3.755853   1495  0.717391\n",
       "           review_palate      3.598328   1495  0.622560\n",
       "           review_taste       3.631438   1495  0.674364\n",
       "\n",
       "[520 rows x 3 columns]"
      ]
     },
     "execution_count": 18,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "gr[review_cols].agg(['mean', 'count', 'std']).stack(level=0)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "You can group by **levels** of a MultiIndex."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th>mean</th>\n",
       "      <th>count</th>\n",
       "      <th>std</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>beer_style</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\">Altbier</th>\n",
       "      <th>review_appearance</th>\n",
       "      <td>3.819018</td>\n",
       "      <td>326</td>\n",
       "      <td>0.559731</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>review_aroma</th>\n",
       "      <td>3.599693</td>\n",
       "      <td>326</td>\n",
       "      <td>0.568426</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>review_overall</th>\n",
       "      <td>3.774540</td>\n",
       "      <td>326</td>\n",
       "      <td>0.676367</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>review_palate</th>\n",
       "      <td>3.684049</td>\n",
       "      <td>326</td>\n",
       "      <td>0.601171</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>review_taste</th>\n",
       "      <td>3.696319</td>\n",
       "      <td>326</td>\n",
       "      <td>0.624957</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                                  mean  count       std\n",
       "beer_style                                             \n",
       "Altbier    review_appearance  3.819018    326  0.559731\n",
       "           review_aroma       3.599693    326  0.568426\n",
       "           review_overall     3.774540    326  0.676367\n",
       "           review_palate      3.684049    326  0.601171\n",
       "           review_taste       3.696319    326  0.624957"
      ]
     },
     "execution_count": 19,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "multi = gr[review_cols].agg(['mean', 'count', 'std']).stack(level=0)\n",
    "multi.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>min</th>\n",
       "      <th>max</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>beer_style</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>Altbier</th>\n",
       "      <td>3.599693</td>\n",
       "      <td>3.819018</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>American Adjunct Lager</th>\n",
       "      <td>2.501012</td>\n",
       "      <td>3.088731</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>American Amber / Red Ale</th>\n",
       "      <td>3.675518</td>\n",
       "      <td>3.864496</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>American Amber / Red Lager</th>\n",
       "      <td>3.243689</td>\n",
       "      <td>3.600971</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>American Barleywine</th>\n",
       "      <td>3.893434</td>\n",
       "      <td>4.040345</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Vienna Lager</th>\n",
       "      <td>3.459963</td>\n",
       "      <td>3.837989</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Weizenbock</th>\n",
       "      <td>4.027928</td>\n",
       "      <td>4.111712</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Wheatwine</th>\n",
       "      <td>3.813319</td>\n",
       "      <td>4.066594</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Winter Warmer</th>\n",
       "      <td>3.694115</td>\n",
       "      <td>3.849075</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Witbier</th>\n",
       "      <td>3.598328</td>\n",
       "      <td>3.755853</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>104 rows × 2 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "                                 min       max\n",
       "beer_style                                    \n",
       "Altbier                     3.599693  3.819018\n",
       "American Adjunct Lager      2.501012  3.088731\n",
       "American Amber / Red Ale    3.675518  3.864496\n",
       "American Amber / Red Lager  3.243689  3.600971\n",
       "American Barleywine         3.893434  4.040345\n",
       "...                              ...       ...\n",
       "Vienna Lager                3.459963  3.837989\n",
       "Weizenbock                  4.027928  4.111712\n",
       "Wheatwine                   3.813319  4.066594\n",
       "Winter Warmer               3.694115  3.849075\n",
       "Witbier                     3.598328  3.755853\n",
       "\n",
       "[104 rows x 2 columns]"
      ]
     },
     "execution_count": 20,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "multi.groupby(level='beer_style')['mean'].agg(['min', 'max'])"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Group by **multiple** columns"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "brewer_id  beer_style                    \n",
       "1          Czech Pilsener                    3.943662\n",
       "3          American Adjunct Lager            3.750000\n",
       "           American Amber / Red Ale          3.750000\n",
       "           American Brown Ale                3.324324\n",
       "           American Double / Imperial IPA    4.333333\n",
       "                                               ...   \n",
       "24926      Belgian Strong Dark Ale           4.000000\n",
       "           Fruit / Vegetable Beer            4.500000\n",
       "24964      American Malt Liquor              3.600000\n",
       "25680      Euro Pale Lager                   3.500000\n",
       "27039      American Double / Imperial IPA    4.875000\n",
       "Name: review_overall, dtype: float64"
      ]
     },
     "execution_count": 21,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.groupby(['brewer_id', 'beer_style']).review_overall.mean()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th>review_appearance</th>\n",
       "      <th>review_aroma</th>\n",
       "      <th>review_overall</th>\n",
       "      <th>review_palate</th>\n",
       "      <th>review_taste</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>brewer_id</th>\n",
       "      <th>beer_style</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>1</th>\n",
       "      <th>Czech Pilsener</th>\n",
       "      <td>3.605634</td>\n",
       "      <td>3.514085</td>\n",
       "      <td>3.943662</td>\n",
       "      <td>3.633803</td>\n",
       "      <td>3.718310</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th rowspan=\"4\" valign=\"top\">3</th>\n",
       "      <th>American Adjunct Lager</th>\n",
       "      <td>3.250000</td>\n",
       "      <td>2.500000</td>\n",
       "      <td>3.750000</td>\n",
       "      <td>3.500000</td>\n",
       "      <td>2.750000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>American Amber / Red Ale</th>\n",
       "      <td>3.666667</td>\n",
       "      <td>3.472222</td>\n",
       "      <td>3.750000</td>\n",
       "      <td>3.486111</td>\n",
       "      <td>3.583333</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>American Brown Ale</th>\n",
       "      <td>3.513514</td>\n",
       "      <td>3.270270</td>\n",
       "      <td>3.324324</td>\n",
       "      <td>3.081081</td>\n",
       "      <td>3.175676</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>American Double / Imperial IPA</th>\n",
       "      <td>3.833333</td>\n",
       "      <td>3.833333</td>\n",
       "      <td>4.333333</td>\n",
       "      <td>3.833333</td>\n",
       "      <td>3.833333</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th rowspan=\"2\" valign=\"top\">24926</th>\n",
       "      <th>Belgian Strong Dark Ale</th>\n",
       "      <td>3.500000</td>\n",
       "      <td>3.500000</td>\n",
       "      <td>4.000000</td>\n",
       "      <td>4.000000</td>\n",
       "      <td>4.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Fruit / Vegetable Beer</th>\n",
       "      <td>4.500000</td>\n",
       "      <td>4.500000</td>\n",
       "      <td>4.500000</td>\n",
       "      <td>4.500000</td>\n",
       "      <td>4.500000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>24964</th>\n",
       "      <th>American Malt Liquor</th>\n",
       "      <td>3.300000</td>\n",
       "      <td>2.800000</td>\n",
       "      <td>3.600000</td>\n",
       "      <td>3.200000</td>\n",
       "      <td>3.400000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>25680</th>\n",
       "      <th>Euro Pale Lager</th>\n",
       "      <td>4.000000</td>\n",
       "      <td>3.500000</td>\n",
       "      <td>3.500000</td>\n",
       "      <td>3.500000</td>\n",
       "      <td>3.500000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>27039</th>\n",
       "      <th>American Double / Imperial IPA</th>\n",
       "      <td>4.625000</td>\n",
       "      <td>4.750000</td>\n",
       "      <td>4.875000</td>\n",
       "      <td>4.875000</td>\n",
       "      <td>4.875000</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>10432 rows × 5 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "                                          review_appearance  review_aroma  \\\n",
       "brewer_id beer_style                                                        \n",
       "1         Czech Pilsener                           3.605634      3.514085   \n",
       "3         American Adjunct Lager                   3.250000      2.500000   \n",
       "          American Amber / Red Ale                 3.666667      3.472222   \n",
       "          American Brown Ale                       3.513514      3.270270   \n",
       "          American Double / Imperial IPA           3.833333      3.833333   \n",
       "...                                                     ...           ...   \n",
       "24926     Belgian Strong Dark Ale                  3.500000      3.500000   \n",
       "          Fruit / Vegetable Beer                   4.500000      4.500000   \n",
       "24964     American Malt Liquor                     3.300000      2.800000   \n",
       "25680     Euro Pale Lager                          4.000000      3.500000   \n",
       "27039     American Double / Imperial IPA           4.625000      4.750000   \n",
       "\n",
       "                                          review_overall  review_palate  \\\n",
       "brewer_id beer_style                                                      \n",
       "1         Czech Pilsener                        3.943662       3.633803   \n",
       "3         American Adjunct Lager                3.750000       3.500000   \n",
       "          American Amber / Red Ale              3.750000       3.486111   \n",
       "          American Brown Ale                    3.324324       3.081081   \n",
       "          American Double / Imperial IPA        4.333333       3.833333   \n",
       "...                                                  ...            ...   \n",
       "24926     Belgian Strong Dark Ale               4.000000       4.000000   \n",
       "          Fruit / Vegetable Beer                4.500000       4.500000   \n",
       "24964     American Malt Liquor                  3.600000       3.200000   \n",
       "25680     Euro Pale Lager                       3.500000       3.500000   \n",
       "27039     American Double / Imperial IPA        4.875000       4.875000   \n",
       "\n",
       "                                          review_taste  \n",
       "brewer_id beer_style                                    \n",
       "1         Czech Pilsener                      3.718310  \n",
       "3         American Adjunct Lager              2.750000  \n",
       "          American Amber / Red Ale            3.583333  \n",
       "          American Brown Ale                  3.175676  \n",
       "          American Double / Imperial IPA      3.833333  \n",
       "...                                                ...  \n",
       "24926     Belgian Strong Dark Ale             4.000000  \n",
       "          Fruit / Vegetable Beer              4.500000  \n",
       "24964     American Malt Liquor                3.400000  \n",
       "25680     Euro Pale Lager                     3.500000  \n",
       "27039     American Double / Imperial IPA      4.875000  \n",
       "\n",
       "[10432 rows x 5 columns]"
      ]
     },
     "execution_count": 22,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.groupby(['brewer_id', 'beer_style'])[review_cols].mean()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Exercise: Plot the relationship between review length (the `text` column) and average `review_overall`.\n",
    "\n",
    "Hint: Break the problem into pieces:\n",
    "\n",
    "- Find the **len**gth of each reivew (remember the `df.text.str` namespace?)\n",
    "- Group by that Series of review lengths\n",
    "- I used `style='k.'` in the plot "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 93,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "# Your code goes here"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 162,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.axes._subplots.AxesSubplot at 0x119ad3048>"
      ]
     },
     "execution_count": 162,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAlAAAAFICAYAAACfj8AuAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nAABJRU5ErkJggg==\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x119a60ef0>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "%load -r 1:5 solutions_groupby.py"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Bonus exercise:\n",
    "\n",
    "- Try grouping by the number of words.\n",
    "- Try grouping by the number of sentances.\n",
    "\n",
    "Remember that `str.count` accepts a regular expression.\n",
    "\n",
    "Don't worry too much about these, especially if you don't remember the syntax\n",
    "for regular expressions (I never can). Just jump to the next exercise."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": 99,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.axes._subplots.AxesSubplot at 0x111991d68>"
      ]
     },
     "execution_count": 99,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAsAAAAGACAYAAACjn9QeAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x114c9c978>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "%load -r 18:20 solutions_groupby.py"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": 101,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.axes._subplots.AxesSubplot at 0x1100e7b38>"
      ]
     },
     "execution_count": 101,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAsAAAAGACAYAAACjn9QeAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nICoEYAAAAESFAAwAAICoEIABAAAQlf8P5MVGdolQjQEAAAAASUVORK5CYII=\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x10fa7c048>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "%load -r 21:26 solutions_groupby.py"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 40,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "# Your code goes here"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Exercise: Which **brewer** (`brewer_id`) has the largest gap between the min and max `review_overall` for two of their beers.\n",
    "\n",
    "Hint: You'll need to do this in two steps.\n",
    "\n",
    "1. Find the average `review_overall` by brewer and beername.\n",
    "2. Find the difference between the max and min by brewer (rembember `.groupby(level=)`)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 43,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "# Your code goes here. You've got this!"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 105,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "brewer_id\n",
       "29      4\n",
       "506     4\n",
       "478     4\n",
       "435     4\n",
       "688     4\n",
       "       ..\n",
       "6266    0\n",
       "884     0\n",
       "882     0\n",
       "6272    0\n",
       "4410    0\n",
       "dtype: float64"
      ]
     },
     "execution_count": 105,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "%load -r 6:13 solutions_groupby.py"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 106,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "# Show for those with counts > 20ish"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Create our own \"kind\" of beer, which aggregates `style`."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0                    american stout\n",
       "1                   american porter\n",
       "2                   german pilsener\n",
       "3    american double / imperial ipa\n",
       "4           american pale ale (apa)\n",
       "Name: beer_style, dtype: object"
      ]
     },
     "execution_count": 23,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "style = df.beer_style.str.lower()\n",
    "style.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "kinds = ['ipa', 'apa', 'amber ale', 'rye', 'scotch', 'stout', 'barleywine', 'porter', 'brown ale', 'lager', 'pilsner',\n",
    "         'tripel', 'biter', 'farmhouse', 'malt liquour', 'rice']"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "'(?P<ipa>ipa)|(?P<apa>apa)|(?P<amber_ale>amber ale)|(?P<rye>rye)|(?P<scotch>scotch)|(?P<stout>stout)|(?P<barleywine>barleywine)|(?P<porter>porter)|(?P<brown_ale>brown ale)|(?P<lager>lager)|(?P<pilsner>pilsner)|(?P<tripel>tripel)|(?P<biter>biter)|(?P<farmhouse>farmhouse)|(?P<malt_liquour>malt liquour)|(?P<rice>rice)'"
      ]
     },
     "execution_count": 25,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "expr = '|'.join(['(?P<{name}>{pat})'.format(pat=kind, name=kind.replace(' ', '_')) for kind in kinds])\n",
    "expr"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0     stout\n",
       "1    porter\n",
       "2     other\n",
       "3       ipa\n",
       "4       apa\n",
       "dtype: object"
      ]
     },
     "execution_count": 26,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "beer_kind = (style.replace({'india pale ale': 'ipa',\n",
    "                            'american pale ale': 'apa'})\n",
    "            .str.extract(expr).fillna('').sum(1)\n",
    "            .str.lower().replace('', 'other'))\n",
    "beer_kind.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "brewer_id           \n",
       "1          other        3.943662\n",
       "3          apa          3.347826\n",
       "           brown ale    3.582418\n",
       "           ipa          3.662791\n",
       "           lager        3.772727\n",
       "                          ...   \n",
       "23980      porter       3.681818\n",
       "24926      other        4.166667\n",
       "24964      other        3.600000\n",
       "25680      lager        3.500000\n",
       "27039      ipa          4.875000\n",
       "Name: review_overall, dtype: float64"
      ]
     },
     "execution_count": 27,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.groupby(['brewer_id', beer_kind]).review_overall.mean()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>apa</th>\n",
       "      <th>barleywine</th>\n",
       "      <th>brown ale</th>\n",
       "      <th>farmhouse</th>\n",
       "      <th>ipa</th>\n",
       "      <th>lager</th>\n",
       "      <th>other</th>\n",
       "      <th>pilsner</th>\n",
       "      <th>porter</th>\n",
       "      <th>rye</th>\n",
       "      <th>scotch</th>\n",
       "      <th>stout</th>\n",
       "      <th>tripel</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>brewer_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",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\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>3</td>\n",
       "      <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>1</td>\n",
       "      <td>0</td>\n",
       "      <td>3</td>\n",
       "      <td>0</td>\n",
       "      <td>2</td>\n",
       "      <td>4</td>\n",
       "      <td>11</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>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>18</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>2</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>3</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8</th>\n",
       "      <td>2</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>3</td>\n",
       "      <td>0</td>\n",
       "      <td>8</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>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>23980</th>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>4</td>\n",
       "      <td>2</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",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>24926</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>3</td>\n",
       "      <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>24964</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>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>25680</th>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>27039</th>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>2</td>\n",
       "      <td>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",
       "  </tbody>\n",
       "</table>\n",
       "<p>2333 rows × 13 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "           apa  barleywine  brown ale  farmhouse  ipa  lager  other  pilsner  \\\n",
       "brewer_id                                                                      \n",
       "1            0           0          0          0    0      0      3        0   \n",
       "3            1           0          3          0    2      4     11        0   \n",
       "4            0           0          0          1    1      0     18        0   \n",
       "6            0           0          0          0    0      0      3        0   \n",
       "8            2           1          0          0    3      0      8        0   \n",
       "...        ...         ...        ...        ...  ...    ...    ...      ...   \n",
       "23980        1           0          0          0    0      4      2        0   \n",
       "24926        0           0          0          0    0      0      3        0   \n",
       "24964        0           0          0          0    0      0      1        0   \n",
       "25680        0           0          0          0    0      1      0        0   \n",
       "27039        0           0          0          0    2      0      0        0   \n",
       "\n",
       "           porter  rye  scotch  stout  tripel  \n",
       "brewer_id                                      \n",
       "1               0    0       0      0       0  \n",
       "3               0    0       0      1       0  \n",
       "4               0    0       1      0       2  \n",
       "6               0    0       0      0       1  \n",
       "8               0    0       0      1       0  \n",
       "...           ...  ...     ...    ...     ...  \n",
       "23980           1    0       0      0       0  \n",
       "24926           0    0       0      0       0  \n",
       "24964           0    0       0      0       0  \n",
       "25680           0    0       0      0       0  \n",
       "27039           0    0       0      0       0  \n",
       "\n",
       "[2333 rows x 13 columns]"
      ]
     },
     "execution_count": 28,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.groupby(['brewer_id', beer_kind]).beer_id.nunique().unstack(1).fillna(0)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Exercise: Which Brewers have the most different `kinds` of beer?\n",
    "\n",
    "Hint: we used `df.profile_name.nunique()` to find the number of different profile names.\n",
    "What are we grouping, and what is our grouper?"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": 33,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "brewer_id\n",
       "1        1\n",
       "3        6\n",
       "4        5\n",
       "6        2\n",
       "8        5\n",
       "        ..\n",
       "23980    4\n",
       "24926    1\n",
       "24964    1\n",
       "25680    1\n",
       "27039    1\n",
       "dtype: int64"
      ]
     },
     "execution_count": 33,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# %load -r 27:29 solutions_groupby.py"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Exercise: Which kinds of beer have the most brewers?"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": 121,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "apa           512\n",
       "barleywine    250\n",
       "brown ale     304\n",
       "farmhouse     136\n",
       "ipa           738\n",
       "             ... \n",
       "porter        474\n",
       "rye            80\n",
       "scotch        137\n",
       "stout         695\n",
       "tripel        165\n",
       "dtype: int64"
      ]
     },
     "execution_count": 121,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "%load -r 30:32 solutions_groupby.py"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "We've seen a lot of permutations among number of groupers, number of columns to aggregate, and number of aggregators.\n",
    "In fact, the `.agg`, which returns one row per group, is just one kind of way to combine the results. The three ways are\n",
    "\n",
    "- `agg`: one row per results\n",
    "- `transform`: identicaly shaped output as input\n",
    "- `apply`: anything goes\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Transform"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Combined Series / DataFrame is the same shape as the input. For example, say you want to standardize the reviews by subtracting the mean."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "def de_mean(reviews):\n",
    "    s = reviews - reviews.mean()\n",
    "    return s"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0        0.63927\n",
       "1        0.63927\n",
       "2       -0.86073\n",
       "3        0.13927\n",
       "4        0.13927\n",
       "          ...   \n",
       "99995    0.63927\n",
       "99996    0.13927\n",
       "99997   -1.36073\n",
       "99998    0.13927\n",
       "99999    0.63927\n",
       "Name: review_overall, dtype: float64"
      ]
     },
     "execution_count": 30,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "de_mean(df.review_overall)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "We can do this at the *person* level with `groupby` and `transform`."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>abv</th>\n",
       "      <th>beer_id</th>\n",
       "      <th>brewer_id</th>\n",
       "      <th>review_appearance</th>\n",
       "      <th>review_aroma</th>\n",
       "      <th>review_overall</th>\n",
       "      <th>review_palate</th>\n",
       "      <th>review_taste</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>-2.763000</td>\n",
       "      <td>-14386.800000</td>\n",
       "      <td>-250.800000</td>\n",
       "      <td>0.350000</td>\n",
       "      <td>-0.050000</td>\n",
       "      <td>0.500000</td>\n",
       "      <td>-0.100000</td>\n",
       "      <td>0.450000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>-1.622833</td>\n",
       "      <td>-8941.000000</td>\n",
       "      <td>4594.447761</td>\n",
       "      <td>0.462687</td>\n",
       "      <td>0.082090</td>\n",
       "      <td>0.395522</td>\n",
       "      <td>0.044776</td>\n",
       "      <td>-0.014925</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>-2.420000</td>\n",
       "      <td>-11859.162500</td>\n",
       "      <td>450.287500</td>\n",
       "      <td>0.025000</td>\n",
       "      <td>-0.968750</td>\n",
       "      <td>-0.875000</td>\n",
       "      <td>-0.631250</td>\n",
       "      <td>-0.350000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>1.673333</td>\n",
       "      <td>-6783.687500</td>\n",
       "      <td>-1250.687500</td>\n",
       "      <td>0.250000</td>\n",
       "      <td>0.281250</td>\n",
       "      <td>0.093750</td>\n",
       "      <td>0.375000</td>\n",
       "      <td>0.312500</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>-1.996861</td>\n",
       "      <td>-25015.457746</td>\n",
       "      <td>-2934.570423</td>\n",
       "      <td>0.242958</td>\n",
       "      <td>-0.774648</td>\n",
       "      <td>0.295775</td>\n",
       "      <td>-0.242958</td>\n",
       "      <td>-0.683099</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>99995</th>\n",
       "      <td>-0.536528</td>\n",
       "      <td>-12955.893333</td>\n",
       "      <td>2269.973333</td>\n",
       "      <td>0.040000</td>\n",
       "      <td>0.580000</td>\n",
       "      <td>0.466667</td>\n",
       "      <td>0.520000</td>\n",
       "      <td>0.520000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>99996</th>\n",
       "      <td>0.118182</td>\n",
       "      <td>-8688.272727</td>\n",
       "      <td>-3685.636364</td>\n",
       "      <td>0.090909</td>\n",
       "      <td>0.045455</td>\n",
       "      <td>0.090909</td>\n",
       "      <td>-0.181818</td>\n",
       "      <td>0.090909</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>99997</th>\n",
       "      <td>-0.500000</td>\n",
       "      <td>-822.000000</td>\n",
       "      <td>-201.500000</td>\n",
       "      <td>0.375000</td>\n",
       "      <td>-0.500000</td>\n",
       "      <td>-1.250000</td>\n",
       "      <td>0.250000</td>\n",
       "      <td>-0.375000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>99998</th>\n",
       "      <td>-0.060876</td>\n",
       "      <td>-40408.033766</td>\n",
       "      <td>-6994.153247</td>\n",
       "      <td>-0.720779</td>\n",
       "      <td>0.331169</td>\n",
       "      <td>0.410390</td>\n",
       "      <td>-0.049351</td>\n",
       "      <td>0.468831</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>99999</th>\n",
       "      <td>-1.517449</td>\n",
       "      <td>-4006.112840</td>\n",
       "      <td>-3498.871595</td>\n",
       "      <td>0.025292</td>\n",
       "      <td>0.492218</td>\n",
       "      <td>0.575875</td>\n",
       "      <td>0.145914</td>\n",
       "      <td>0.577821</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>100000 rows × 8 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "            abv       beer_id    brewer_id  review_appearance  review_aroma  \\\n",
       "0     -2.763000 -14386.800000  -250.800000           0.350000     -0.050000   \n",
       "1     -1.622833  -8941.000000  4594.447761           0.462687      0.082090   \n",
       "2     -2.420000 -11859.162500   450.287500           0.025000     -0.968750   \n",
       "3      1.673333  -6783.687500 -1250.687500           0.250000      0.281250   \n",
       "4     -1.996861 -25015.457746 -2934.570423           0.242958     -0.774648   \n",
       "...         ...           ...          ...                ...           ...   \n",
       "99995 -0.536528 -12955.893333  2269.973333           0.040000      0.580000   \n",
       "99996  0.118182  -8688.272727 -3685.636364           0.090909      0.045455   \n",
       "99997 -0.500000   -822.000000  -201.500000           0.375000     -0.500000   \n",
       "99998 -0.060876 -40408.033766 -6994.153247          -0.720779      0.331169   \n",
       "99999 -1.517449  -4006.112840 -3498.871595           0.025292      0.492218   \n",
       "\n",
       "       review_overall  review_palate  review_taste  \n",
       "0            0.500000      -0.100000      0.450000  \n",
       "1            0.395522       0.044776     -0.014925  \n",
       "2           -0.875000      -0.631250     -0.350000  \n",
       "3            0.093750       0.375000      0.312500  \n",
       "4            0.295775      -0.242958     -0.683099  \n",
       "...               ...            ...           ...  \n",
       "99995        0.466667       0.520000      0.520000  \n",
       "99996        0.090909      -0.181818      0.090909  \n",
       "99997       -1.250000       0.250000     -0.375000  \n",
       "99998        0.410390      -0.049351      0.468831  \n",
       "99999        0.575875       0.145914      0.577821  \n",
       "\n",
       "[100000 rows x 8 columns]"
      ]
     },
     "execution_count": 31,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.groupby('profile_name').transform(de_mean)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Apply\n",
    "\n",
    "So there's `gr.agg`. and `gr.transform`, and finally `gr.apply`. We're going to skip apply for now. I have an example in a later notebook. "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Resample\n",
    "\n",
    "Resample is a special kind of groupby operation for when you have a `DatetimeIndex`."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "2009-10-05 21:31:48    1\n",
       "2009-10-05 21:32:09    1\n",
       "2009-10-05 21:32:13    1\n",
       "2009-10-05 21:32:37    1\n",
       "2009-10-05 21:33:14    1\n",
       "                      ..\n",
       "2010-03-07 01:29:31    1\n",
       "2010-03-07 01:30:35    1\n",
       "2010-03-07 01:32:46    1\n",
       "2010-03-07 01:33:29    1\n",
       "2010-03-07 01:34:05    1\n",
       "dtype: int64"
      ]
     },
     "execution_count": 32,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "review_times = df.time.value_counts().sort_index()\n",
    "review_times"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 33,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "DatetimeIndex(['2009-10-05 21:31:48', '2009-10-05 21:32:09',\n",
       "               '2009-10-05 21:32:13', '2009-10-05 21:32:37',\n",
       "               '2009-10-05 21:33:14', '2009-10-05 21:33:48',\n",
       "               '2009-10-05 21:34:24', '2009-10-05 21:34:29',\n",
       "               '2009-10-05 21:34:31', '2009-10-05 21:35:09', \n",
       "               ...\n",
       "               '2010-03-07 01:22:40', '2010-03-07 01:25:09',\n",
       "               '2010-03-07 01:25:50', '2010-03-07 01:27:21',\n",
       "               '2010-03-07 01:28:05', '2010-03-07 01:29:31',\n",
       "               '2010-03-07 01:30:35', '2010-03-07 01:32:46',\n",
       "               '2010-03-07 01:33:29', '2010-03-07 01:34:05'],\n",
       "              dtype='datetime64[ns]', length=99307, freq=None, tz=None)"
      ]
     },
     "execution_count": 33,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "review_times.index"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "The number of reviews within a given second isn't that interesting."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 34,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.axes._subplots.AxesSubplot at 0x1039bbb38>"
      ]
     },
     "execution_count": 34,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAsAAAAFlCAYAAADlFl4vAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\niIiIKFYYABMRERFRrPx/GgGJsKStW7MAAAAASUVORK5CYII=\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x10a1bbf60>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "review_times.plot()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Right now the frequency is way to high to be meaningful. `resample` lets you adjust the frequency."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 35,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.axes._subplots.AxesSubplot at 0x10af00cf8>"
      ]
     },
     "execution_count": 35,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAsoAAAF8CAYAAADW7Z21AAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nAAz8B3GYW0Tr7E5uAAAAAElFTkSuQmCC\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x10b026d30>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "review_times.resample(\"D\").plot()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "We've essentially grouped by day here (but syntax of `.resample('d')` is much nicer than what we'd have to do to use the `.groupby(grouper)` spelling). By default the aggregation function is `mean`, i.e. take the average of all the values that fall on that day. You can also sum."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 36,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.axes._subplots.AxesSubplot at 0x10af61128>"
      ]
     },
     "execution_count": 36,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAscAAAF8CAYAAAAjExYFAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nAElFTkSuQmCC\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x10adbe630>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "review_times.resample('D', how='sum').plot()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "The `freq` you pass in to `resample` is pretty flexible."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 37,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "2009-10-05 21:31:48    3326\n",
       "2009-10-10 21:31:48    3172\n",
       "2009-10-15 21:31:48    3203\n",
       "2009-10-20 21:31:48    1838\n",
       "2009-10-25 21:31:48    2697\n",
       "                       ... \n",
       "2010-02-12 21:31:48    3736\n",
       "2010-02-17 21:31:48    3577\n",
       "2010-02-22 21:31:48    3402\n",
       "2010-02-27 21:31:48    3292\n",
       "2010-03-04 21:31:48    1529\n",
       "Freq: 5D, dtype: int64"
      ]
     },
     "execution_count": 37,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "review_times.resample('5D', how='sum')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 38,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.axes._subplots.AxesSubplot at 0x10991e048>"
      ]
     },
     "execution_count": 38,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAsIAAAF8CAYAAADFOt1BAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nERERkUT6P0U76kTYHCJlAAAAAElFTkSuQmCC\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x109d6c320>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "c = review_times.resample('3H', how='sum')\n",
    "ax = c.plot(alpha=.5)\n",
    "pd.rolling_mean(c, window=8).plot(ax=ax)\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Exercise: Plot the number of distinct brewers reviewed per day\n",
    "\n",
    "- Hint: The documentation for `resample` is being worked on, but the `how` in `.resample` is *really* flexible. Try the first thing that comes to mind. What function do you want to apply to get the number of unique values?\n",
    "- Hint2: (Sorry this is harder than I thought). `resample` needs a `DatetimeIndex`. We have datetimes in `time`. If only there was a way to take a column and **set** it as the **index**..."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "%load -r 33:36 solutions_groupby.py"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Aside: Beer Recommender\n",
    "\n",
    "See [Harvard CS109](https://github.com/cs109/content) for a more complete example (with chocolate instead of beer).\n",
    "\n",
    "One place where transform comes in handy is as a preprocessing step for any kind of recommender. In some sense, raw score I assign a beer is less important the the score relative to *my* mean."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 40,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>beer_id</th>\n",
       "      <th>profile_name</th>\n",
       "      <th>review_overall</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>2511</td>\n",
       "      <td>blaheath</td>\n",
       "      <td>4.5</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>19736</td>\n",
       "      <td>GJ40</td>\n",
       "      <td>4.5</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>11098</td>\n",
       "      <td>biegaman</td>\n",
       "      <td>3.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>28577</td>\n",
       "      <td>nick76</td>\n",
       "      <td>4.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>398</td>\n",
       "      <td>champ103</td>\n",
       "      <td>4.0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   beer_id profile_name  review_overall\n",
       "0     2511     blaheath             4.5\n",
       "1    19736         GJ40             4.5\n",
       "2    11098     biegaman             3.0\n",
       "3    28577       nick76             4.0\n",
       "4      398     champ103             4.0"
      ]
     },
     "execution_count": 40,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "deduped = df[['beer_id', 'profile_name', 'review_overall']].drop_duplicates()\n",
    "deduped.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 41,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "user_counts = deduped.profile_name.value_counts()\n",
    "top_users = user_counts[user_counts > user_counts.quantile(.75)].index"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 42,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "beer_counts = deduped.beer_id.value_counts()\n",
    "top_beers = beer_counts[beer_counts > beer_counts.quantile(.9)].index"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 43,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "top = deduped.query('beer_id in @top_beers and profile_name in @top_users')\n",
    "user_means = top.groupby('profile_name').review_overall.mean()\n",
    "beer_means = top.groupby('beer_id').review_overall.mean()\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 44,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.text.Text at 0x10fbf7f60>"
      ]
     },
     "execution_count": 44,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAA6UAAAEZCAYAAABvtoNhAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nuZiRqROzAAAAAElFTkSuQmCC\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x109de39e8>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "fig, axes = plt.subplots(figsize=(16, 4), ncols=2, sharey=True, sharex=True)\n",
    "\n",
    "sns.distplot(user_means, kde=False, ax=axes[0], color='k', norm_hist=True, hist_kws={'alpha': 1})\n",
    "sns.distplot(beer_means, kde=False, ax=axes[1], color='k', norm_hist=True, hist_kws={'alpha': 1})\n",
    "axes[0].set_title(\"User Averages\")\n",
    "axes[1].set_title(\"Beer Averages\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 45,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "beer_id  profile_name  \n",
       "5        Ek0nomik          3.0\n",
       "         HalfFull          4.5\n",
       "         Jesstyr           4.0\n",
       "         JordonHoltzman    4.0\n",
       "         KTCamm            4.0\n",
       "Name: review_overall, dtype: float64"
      ]
     },
     "execution_count": 45,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "s = top.set_index(['beer_id', 'profile_name']).review_overall.sort_index()\n",
    "s.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "###  `de_mean` the scores in `s`"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 46,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "beer_id  profile_name  \n",
       "5        Ek0nomik         -0.857143\n",
       "         HalfFull          0.789474\n",
       "         Jesstyr           0.233333\n",
       "         JordonHoltzman    0.222222\n",
       "         KTCamm            0.166667\n",
       "Name: review_overall, dtype: float64"
      ]
     },
     "execution_count": 46,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "standardized = s.groupby(level='profile_name').transform(de_mean)\n",
    "standardized.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 47,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "from scipy.stats import pearsonr"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 48,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "def pearson_sim(reviews_1, reviews_2, reg=2):\n",
    "    \"\"\"\n",
    "    (regularized) Pearson correlation coefficient between sets\n",
    "    of reviews for two beers, made by a common subset\n",
    "    of reviewers.\n",
    "    \n",
    "    `reviews_1` and `reviews_2` should be have the same index,\n",
    "    the `profile_name`s of people who reviewed both beers.\n",
    "    \"\"\"\n",
    "    n_common = len(reviews_1)\n",
    "    if n_common == 0:\n",
    "        similarity = 0\n",
    "    else:\n",
    "        rho = pearsonr(reviews_1, reviews_2)[0]\n",
    "        similarity = (n_common * rho) / (n_common + reg)  # regularization if few reviews\n",
    "    return similarity, n_common"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 49,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "def beer_similarity(standardized, beer_1, beer_2, simfunc=pearson_sim, **simfunc_kwargs):\n",
    "    \"\"\"\n",
    "    Compute the similarity between two beers.\n",
    "    \"\"\"\n",
    "    # get common subset...\n",
    "    reviewers_1 = standardized.loc[beer_1].index\n",
    "    reviewers_2 = standardized.loc[beer_2].index\n",
    "    common_idx = reviewers_1 & reviewers_2  # set intersection\n",
    "\n",
    "    # slice the Multiindex, unstack to be N x 2\n",
    "    common_reviews = standardized.loc[[beer_1, beer_2], common_idx].unstack('beer_id')\n",
    "    # ... review similairty for subset\n",
    "    rho, n_common = simfunc(common_reviews[beer_1], common_reviews[beer_2], **simfunc_kwargs)\n",
    "    return rho, n_common"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 50,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "1266"
      ]
     },
     "execution_count": 50,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "beer_ids = s.index.levels[0]\n",
    "len(beer_ids)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 51,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(0.58989530654977695, 3)"
      ]
     },
     "execution_count": 51,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "beer_similarity(standardized, beer_ids[0], beer_ids[10])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 509,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "CPU times: user 51min 33s, sys: 50.4 s, total: 52min 24s\n",
      "Wall time: 53min 7s\n"
     ]
    }
   ],
   "source": [
    "%%time\n",
    "sims = []\n",
    "\n",
    "for i, beer_1 in enumerate(beer_ids):\n",
    "    for j, beer_2 in enumerate(beer_ids):\n",
    "        if j >= i:\n",
    "            continue\n",
    "        sim, n_common = beer_similarity(s, beer_1, beer_2)\n",
    "        sims.append((beer_1, beer_2, sim, n_common))\n",
    "        print((i, j), end='\\r')\n",
    "        \n",
    "sim = pd.DataFrame(sims, columns=['beer_1', 'beer_2', 'score', 'n_common'])\n",
    "sim.to_csv('beer_subset_similarity.csv', index=False)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 53,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>beer_1</th>\n",
       "      <th>beer_2</th>\n",
       "      <th>score</th>\n",
       "      <th>n_common</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>6</td>\n",
       "      <td>5</td>\n",
       "      <td>0.306201</td>\n",
       "      <td>11</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>7</td>\n",
       "      <td>5</td>\n",
       "      <td>0.385503</td>\n",
       "      <td>6</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>7</td>\n",
       "      <td>6</td>\n",
       "      <td>0.553703</td>\n",
       "      <td>8</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>10</td>\n",
       "      <td>5</td>\n",
       "      <td>0.500000</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>10</td>\n",
       "      <td>6</td>\n",
       "      <td>NaN</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   beer_1  beer_2     score  n_common\n",
       "0       6       5  0.306201        11\n",
       "1       7       5  0.385503         6\n",
       "2       7       6  0.553703         8\n",
       "3      10       5  0.500000         2\n",
       "4      10       6       NaN         1"
      ]
     },
     "execution_count": 53,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "sim = pd.read_csv('beer_subset_similarity.csv.gz')\n",
    "sim.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 61,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.axes._subplots.AxesSubplot at 0x10b7ea0f0>"
      ]
     },
     "execution_count": 61,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAsYAAAFxCAYAAABwT27lAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x10efd8e48>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "sns.kdeplot(sim[sim.score != 0].dropna().score)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 62,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "sim = sim.set_index(['beer_1', 'beer_2']).score"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 63,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "beer_2\n",
       "1632     0.684993\n",
       "2948     0.666667\n",
       "5441     0.666667\n",
       "16520    0.656113\n",
       "1445     0.618115\n",
       "Name: score, dtype: float64"
      ]
     },
     "execution_count": 63,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "sim.loc[21690].nlargest(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": "3.4.3"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 0
}