chezou/pandas_redshift.ipynb

## pandas_redshift.ipynb
{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# PandasとRedshiftのデモ\n",
    "\n",
    "`ipython-sql`と`redshift_sqlalchemy`を使ったRedshiftのデータ操作のデモ。  \n",
    "`%`で始まるのはマジックコメントで、これを使うと`%%sql`でSQLを記述しその結果をpandasのDataFrameとして受け取れる。"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/Users/michiaki-ariga/.pyenv/versions/anaconda-2.1.0/envs/analysis/lib/python3.4/site-packages/IPython/config.py:13: ShimWarning: The `IPython.config` package has been deprecated. You should import from traitlets.config instead.\n",
      "  \"You should import from traitlets.config instead.\", ShimWarning)\n",
      "/Users/michiaki-ariga/.pyenv/versions/anaconda-2.1.0/envs/analysis/lib/python3.4/site-packages/IPython/utils/traitlets.py:5: UserWarning: IPython.utils.traitlets has moved to a top-level traitlets package.\n",
      "  warn(\"IPython.utils.traitlets has moved to a top-level traitlets package.\")\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "'Connected: redshift_user@development'"
      ]
     },
     "execution_count": 1,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "%matplotlib inline\n",
    "%config SqlMagic.autopandas = True\n",
    "%config SqlMagic.feedback = False\n",
    "%load_ext sql\n",
    "\n",
    "import os\n",
    "import matplotlib\n",
    "import redshift_sqlalchemy\n",
    "matplotlib.style.use('ggplot')\n",
    "\n",
    "conn_url = os.environ['JUPYTER_REDSHIFT_CONN_URL']\n",
    "%sql $conn_url"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## %%sqlによるSQLの記述\n",
    "\n",
    "今回は全体のデータを使いたかったのでwhere句を入れてないが、ログデータなど大規模なデータはDataFrameに全てロードしようとすると、メモリに載り切らないので注意\n",
    "\n",
    "今回は、age(年齢)、balance(預金)、education(学歴)、既婚歴(marital)を見てみる"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {
    "collapsed": false,
    "scrolled": true
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>age</th>\n",
       "      <th>balance</th>\n",
       "      <th>education</th>\n",
       "      <th>marital</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>25</td>\n",
       "      <td>50</td>\n",
       "      <td>secondary</td>\n",
       "      <td>married</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>25</td>\n",
       "      <td>-7</td>\n",
       "      <td>secondary</td>\n",
       "      <td>married</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>25</td>\n",
       "      <td>122</td>\n",
       "      <td>secondary</td>\n",
       "      <td>single</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>25</td>\n",
       "      <td>1348</td>\n",
       "      <td>tertiary</td>\n",
       "      <td>single</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>25</td>\n",
       "      <td>0</td>\n",
       "      <td>secondary</td>\n",
       "      <td>single</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>25</td>\n",
       "      <td>37</td>\n",
       "      <td>tertiary</td>\n",
       "      <td>married</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>25</td>\n",
       "      <td>131</td>\n",
       "      <td>secondary</td>\n",
       "      <td>single</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>25</td>\n",
       "      <td>55</td>\n",
       "      <td>secondary</td>\n",
       "      <td>married</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8</th>\n",
       "      <td>25</td>\n",
       "      <td>787</td>\n",
       "      <td>secondary</td>\n",
       "      <td>single</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9</th>\n",
       "      <td>25</td>\n",
       "      <td>61</td>\n",
       "      <td>secondary</td>\n",
       "      <td>single</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>10</th>\n",
       "      <td>25</td>\n",
       "      <td>10</td>\n",
       "      <td>secondary</td>\n",
       "      <td>married</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>11</th>\n",
       "      <td>25</td>\n",
       "      <td>96</td>\n",
       "      <td>secondary</td>\n",
       "      <td>single</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>12</th>\n",
       "      <td>25</td>\n",
       "      <td>784</td>\n",
       "      <td>tertiary</td>\n",
       "      <td>married</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>13</th>\n",
       "      <td>25</td>\n",
       "      <td>333</td>\n",
       "      <td>tertiary</td>\n",
       "      <td>married</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>14</th>\n",
       "      <td>25</td>\n",
       "      <td>249</td>\n",
       "      <td>secondary</td>\n",
       "      <td>single</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>15</th>\n",
       "      <td>25</td>\n",
       "      <td>1135</td>\n",
       "      <td>secondary</td>\n",
       "      <td>single</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>16</th>\n",
       "      <td>25</td>\n",
       "      <td>-454</td>\n",
       "      <td>secondary</td>\n",
       "      <td>married</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>17</th>\n",
       "      <td>25</td>\n",
       "      <td>641</td>\n",
       "      <td>tertiary</td>\n",
       "      <td>single</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>18</th>\n",
       "      <td>25</td>\n",
       "      <td>-439</td>\n",
       "      <td>secondary</td>\n",
       "      <td>single</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>19</th>\n",
       "      <td>25</td>\n",
       "      <td>40</td>\n",
       "      <td>tertiary</td>\n",
       "      <td>single</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>20</th>\n",
       "      <td>25</td>\n",
       "      <td>276</td>\n",
       "      <td>secondary</td>\n",
       "      <td>single</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>21</th>\n",
       "      <td>25</td>\n",
       "      <td>-76</td>\n",
       "      <td>secondary</td>\n",
       "      <td>married</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>22</th>\n",
       "      <td>25</td>\n",
       "      <td>1</td>\n",
       "      <td>secondary</td>\n",
       "      <td>single</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>23</th>\n",
       "      <td>25</td>\n",
       "      <td>17</td>\n",
       "      <td>primary</td>\n",
       "      <td>single</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>24</th>\n",
       "      <td>25</td>\n",
       "      <td>335</td>\n",
       "      <td>secondary</td>\n",
       "      <td>single</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>25</th>\n",
       "      <td>25</td>\n",
       "      <td>768</td>\n",
       "      <td>secondary</td>\n",
       "      <td>single</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>26</th>\n",
       "      <td>25</td>\n",
       "      <td>382</td>\n",
       "      <td>secondary</td>\n",
       "      <td>single</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>27</th>\n",
       "      <td>25</td>\n",
       "      <td>-250</td>\n",
       "      <td>secondary</td>\n",
       "      <td>single</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>28</th>\n",
       "      <td>25</td>\n",
       "      <td>1270</td>\n",
       "      <td>tertiary</td>\n",
       "      <td>single</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>29</th>\n",
       "      <td>25</td>\n",
       "      <td>54</td>\n",
       "      <td>secondary</td>\n",
       "      <td>single</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>45181</th>\n",
       "      <td>78</td>\n",
       "      <td>158</td>\n",
       "      <td>primary</td>\n",
       "      <td>married</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>45182</th>\n",
       "      <td>78</td>\n",
       "      <td>2735</td>\n",
       "      <td>primary</td>\n",
       "      <td>married</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>45183</th>\n",
       "      <td>78</td>\n",
       "      <td>499</td>\n",
       "      <td>secondary</td>\n",
       "      <td>married</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>45184</th>\n",
       "      <td>78</td>\n",
       "      <td>2628</td>\n",
       "      <td>unknown</td>\n",
       "      <td>divorced</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>45185</th>\n",
       "      <td>78</td>\n",
       "      <td>2235</td>\n",
       "      <td>unknown</td>\n",
       "      <td>married</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>45186</th>\n",
       "      <td>78</td>\n",
       "      <td>240</td>\n",
       "      <td>primary</td>\n",
       "      <td>married</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>45187</th>\n",
       "      <td>78</td>\n",
       "      <td>8556</td>\n",
       "      <td>primary</td>\n",
       "      <td>married</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>45188</th>\n",
       "      <td>78</td>\n",
       "      <td>1016</td>\n",
       "      <td>tertiary</td>\n",
       "      <td>divorced</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>45189</th>\n",
       "      <td>78</td>\n",
       "      <td>2787</td>\n",
       "      <td>primary</td>\n",
       "      <td>divorced</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>45190</th>\n",
       "      <td>78</td>\n",
       "      <td>220</td>\n",
       "      <td>secondary</td>\n",
       "      <td>married</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>45191</th>\n",
       "      <td>78</td>\n",
       "      <td>631</td>\n",
       "      <td>secondary</td>\n",
       "      <td>divorced</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>45192</th>\n",
       "      <td>78</td>\n",
       "      <td>301</td>\n",
       "      <td>primary</td>\n",
       "      <td>married</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>45193</th>\n",
       "      <td>78</td>\n",
       "      <td>0</td>\n",
       "      <td>unknown</td>\n",
       "      <td>divorced</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>45194</th>\n",
       "      <td>78</td>\n",
       "      <td>4807</td>\n",
       "      <td>unknown</td>\n",
       "      <td>married</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>45195</th>\n",
       "      <td>78</td>\n",
       "      <td>229</td>\n",
       "      <td>primary</td>\n",
       "      <td>divorced</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>45196</th>\n",
       "      <td>78</td>\n",
       "      <td>55</td>\n",
       "      <td>primary</td>\n",
       "      <td>divorced</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>45197</th>\n",
       "      <td>78</td>\n",
       "      <td>226</td>\n",
       "      <td>tertiary</td>\n",
       "      <td>married</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>45198</th>\n",
       "      <td>78</td>\n",
       "      <td>1204</td>\n",
       "      <td>tertiary</td>\n",
       "      <td>married</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>45199</th>\n",
       "      <td>78</td>\n",
       "      <td>4917</td>\n",
       "      <td>primary</td>\n",
       "      <td>divorced</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>45200</th>\n",
       "      <td>78</td>\n",
       "      <td>4807</td>\n",
       "      <td>unknown</td>\n",
       "      <td>married</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>45201</th>\n",
       "      <td>78</td>\n",
       "      <td>3219</td>\n",
       "      <td>secondary</td>\n",
       "      <td>divorced</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>45202</th>\n",
       "      <td>78</td>\n",
       "      <td>1389</td>\n",
       "      <td>primary</td>\n",
       "      <td>divorced</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>45203</th>\n",
       "      <td>78</td>\n",
       "      <td>832</td>\n",
       "      <td>primary</td>\n",
       "      <td>married</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>45204</th>\n",
       "      <td>78</td>\n",
       "      <td>3417</td>\n",
       "      <td>secondary</td>\n",
       "      <td>divorced</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>45205</th>\n",
       "      <td>78</td>\n",
       "      <td>3208</td>\n",
       "      <td>unknown</td>\n",
       "      <td>married</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>45206</th>\n",
       "      <td>78</td>\n",
       "      <td>680</td>\n",
       "      <td>primary</td>\n",
       "      <td>married</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>45207</th>\n",
       "      <td>78</td>\n",
       "      <td>38</td>\n",
       "      <td>unknown</td>\n",
       "      <td>married</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>45208</th>\n",
       "      <td>78</td>\n",
       "      <td>527</td>\n",
       "      <td>primary</td>\n",
       "      <td>divorced</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>45209</th>\n",
       "      <td>78</td>\n",
       "      <td>1780</td>\n",
       "      <td>unknown</td>\n",
       "      <td>married</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>45210</th>\n",
       "      <td>78</td>\n",
       "      <td>14204</td>\n",
       "      <td>primary</td>\n",
       "      <td>married</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>45211 rows × 4 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "       age  balance  education   marital\n",
       "0       25       50  secondary   married\n",
       "1       25       -7  secondary   married\n",
       "2       25      122  secondary    single\n",
       "3       25     1348   tertiary    single\n",
       "4       25        0  secondary    single\n",
       "5       25       37   tertiary   married\n",
       "6       25      131  secondary    single\n",
       "7       25       55  secondary   married\n",
       "8       25      787  secondary    single\n",
       "9       25       61  secondary    single\n",
       "10      25       10  secondary   married\n",
       "11      25       96  secondary    single\n",
       "12      25      784   tertiary   married\n",
       "13      25      333   tertiary   married\n",
       "14      25      249  secondary    single\n",
       "15      25     1135  secondary    single\n",
       "16      25     -454  secondary   married\n",
       "17      25      641   tertiary    single\n",
       "18      25     -439  secondary    single\n",
       "19      25       40   tertiary    single\n",
       "20      25      276  secondary    single\n",
       "21      25      -76  secondary   married\n",
       "22      25        1  secondary    single\n",
       "23      25       17    primary    single\n",
       "24      25      335  secondary    single\n",
       "25      25      768  secondary    single\n",
       "26      25      382  secondary    single\n",
       "27      25     -250  secondary    single\n",
       "28      25     1270   tertiary    single\n",
       "29      25       54  secondary    single\n",
       "...    ...      ...        ...       ...\n",
       "45181   78      158    primary   married\n",
       "45182   78     2735    primary   married\n",
       "45183   78      499  secondary   married\n",
       "45184   78     2628    unknown  divorced\n",
       "45185   78     2235    unknown   married\n",
       "45186   78      240    primary   married\n",
       "45187   78     8556    primary   married\n",
       "45188   78     1016   tertiary  divorced\n",
       "45189   78     2787    primary  divorced\n",
       "45190   78      220  secondary   married\n",
       "45191   78      631  secondary  divorced\n",
       "45192   78      301    primary   married\n",
       "45193   78        0    unknown  divorced\n",
       "45194   78     4807    unknown   married\n",
       "45195   78      229    primary  divorced\n",
       "45196   78       55    primary  divorced\n",
       "45197   78      226   tertiary   married\n",
       "45198   78     1204   tertiary   married\n",
       "45199   78     4917    primary  divorced\n",
       "45200   78     4807    unknown   married\n",
       "45201   78     3219  secondary  divorced\n",
       "45202   78     1389    primary  divorced\n",
       "45203   78      832    primary   married\n",
       "45204   78     3417  secondary  divorced\n",
       "45205   78     3208    unknown   married\n",
       "45206   78      680    primary   married\n",
       "45207   78       38    unknown   married\n",
       "45208   78      527    primary  divorced\n",
       "45209   78     1780    unknown   married\n",
       "45210   78    14204    primary   married\n",
       "\n",
       "[45211 rows x 4 columns]"
      ]
     },
     "execution_count": 2,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "%%sql\n",
    "\n",
    "select age, balance, education, marital\n",
    "from tmp.bank\n",
    ";\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## 要約統計量を見る\n",
    "\n",
    "`df.describe()`で平均、標準偏差などの統計量を概観できる"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "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>age</th>\n",
       "      <th>balance</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>count</th>\n",
       "      <td>45211.000000</td>\n",
       "      <td>45211.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>mean</th>\n",
       "      <td>40.936210</td>\n",
       "      <td>1362.272058</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>std</th>\n",
       "      <td>10.618762</td>\n",
       "      <td>3044.765829</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>min</th>\n",
       "      <td>18.000000</td>\n",
       "      <td>-8019.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>25%</th>\n",
       "      <td>33.000000</td>\n",
       "      <td>72.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>50%</th>\n",
       "      <td>39.000000</td>\n",
       "      <td>448.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>75%</th>\n",
       "      <td>48.000000</td>\n",
       "      <td>1428.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>max</th>\n",
       "      <td>95.000000</td>\n",
       "      <td>102127.000000</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                age        balance\n",
       "count  45211.000000   45211.000000\n",
       "mean      40.936210    1362.272058\n",
       "std       10.618762    3044.765829\n",
       "min       18.000000   -8019.000000\n",
       "25%       33.000000      72.000000\n",
       "50%       39.000000     448.000000\n",
       "75%       48.000000    1428.000000\n",
       "max       95.000000  102127.000000"
      ]
     },
     "execution_count": 3,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df = _\n",
    "df.describe()\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## 箱ひげ図を見る\n",
    "\n",
    "既婚歴に応じた年齢の分布を箱ひげ図で描画する。  \n",
    "赤い線が中央値、青い箱の上下が75%点、25%点。+で表現しているのが外れ値。"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.axes._subplots.AxesSubplot at 0x10b542320>"
      ]
     },
     "execution_count": 4,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAX4AAAEeCAYAAACNLn6mAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nYII=\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x10b57b5c0>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "df[['age', 'marital']].boxplot(by='marital')\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "既婚歴と預金の箱ひげ図。  \n",
    "外れ値は省略(`sym=''`)している。"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.axes._subplots.AxesSubplot at 0x10a557eb8>"
      ]
     },
     "execution_count": 5,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAY0AAAEeCAYAAAB2VUk8AAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nIiISjaFBRESiMTSIiEi0/wczrzqWX+2coAAAAABJRU5ErkJggg==\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x10a503be0>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "df[['balance', 'marital']].boxplot(by='marital', sym='')\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## 散布図の例\n",
    "\n",
    "seabornというグラフ描画ライブラリを使って散布図行列を描画している。  \n",
    "学歴毎の年齢と預金の散布図。"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/Users/michiaki-ariga/.pyenv/versions/anaconda-2.1.0/envs/analysis/lib/python3.4/site-packages/matplotlib/__init__.py:872: UserWarning: axes.color_cycle is deprecated and replaced with axes.prop_cycle; please use the latter.\n",
      "  warnings.warn(self.msg_depr % (key, alt_key))\n",
      "/Users/michiaki-ariga/.pyenv/versions/anaconda-2.1.0/envs/analysis/lib/python3.4/site-packages/matplotlib/__init__.py:892: UserWarning: axes.color_cycle is deprecated and replaced with axes.prop_cycle; please use the latter.\n",
      "  warnings.warn(self.msg_depr % (key, alt_key))\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "<seaborn.axisgrid.PairGrid at 0x10243bf60>"
      ]
     },
     "execution_count": 6,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAcMAAAFhCAYAAAAMUsE7AAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x10b0f8ba8>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "import seaborn as sns\n",
    "sns.set()\n",
    "sns.pairplot(df, hue='education')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.4.3"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 0
}