{
 "cells": [
  {
   "cell_type": "markdown",
   "id": "dbfef547",
   "metadata": {},
   "source": [
    "<img src = \"DB.jpeg\">"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 67,
   "id": "99298c2e",
   "metadata": {},
   "outputs": [],
   "source": [
    "# 패키지 로딩하기\n",
    "import pandas  as pd\n",
    "import seaborn as sns\n",
    "import numpy   as np # 수학, 선형대수학"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "id": "dbdc21a4",
   "metadata": {},
   "outputs": [],
   "source": [
    "# 데이터 읽어오기 : 예제 데이터 : diamonds\n",
    "# data = pd.read_excel(io         = \"directory/filename.xlsx\",\n",
    "#                      sheet_name = index or \"sheet_name\",\n",
    "#                      header     = 0)\n",
    "diamonds = sns.load_dataset(\"diamonds\")"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "dccfc100",
   "metadata": {},
   "source": [
    "### 데이터 전처리 : Data Pre-Processing"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "55236610",
   "metadata": {},
   "source": [
    "#### 1. 데이터 보기\n",
    "- 전체 보기 : data\n",
    "- 일부 보기 : head(), tail()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "id": "eed2fb13",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>carat</th>\n",
       "      <th>cut</th>\n",
       "      <th>color</th>\n",
       "      <th>clarity</th>\n",
       "      <th>depth</th>\n",
       "      <th>table</th>\n",
       "      <th>price</th>\n",
       "      <th>x</th>\n",
       "      <th>y</th>\n",
       "      <th>z</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>0.23</td>\n",
       "      <td>Ideal</td>\n",
       "      <td>E</td>\n",
       "      <td>SI2</td>\n",
       "      <td>61.5</td>\n",
       "      <td>55.0</td>\n",
       "      <td>326</td>\n",
       "      <td>3.95</td>\n",
       "      <td>3.98</td>\n",
       "      <td>2.43</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>0.21</td>\n",
       "      <td>Premium</td>\n",
       "      <td>E</td>\n",
       "      <td>SI1</td>\n",
       "      <td>59.8</td>\n",
       "      <td>61.0</td>\n",
       "      <td>326</td>\n",
       "      <td>3.89</td>\n",
       "      <td>3.84</td>\n",
       "      <td>2.31</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>0.23</td>\n",
       "      <td>Good</td>\n",
       "      <td>E</td>\n",
       "      <td>VS1</td>\n",
       "      <td>56.9</td>\n",
       "      <td>65.0</td>\n",
       "      <td>327</td>\n",
       "      <td>4.05</td>\n",
       "      <td>4.07</td>\n",
       "      <td>2.31</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>0.29</td>\n",
       "      <td>Premium</td>\n",
       "      <td>I</td>\n",
       "      <td>VS2</td>\n",
       "      <td>62.4</td>\n",
       "      <td>58.0</td>\n",
       "      <td>334</td>\n",
       "      <td>4.20</td>\n",
       "      <td>4.23</td>\n",
       "      <td>2.63</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>0.31</td>\n",
       "      <td>Good</td>\n",
       "      <td>J</td>\n",
       "      <td>SI2</td>\n",
       "      <td>63.3</td>\n",
       "      <td>58.0</td>\n",
       "      <td>335</td>\n",
       "      <td>4.34</td>\n",
       "      <td>4.35</td>\n",
       "      <td>2.75</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",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>53935</th>\n",
       "      <td>0.72</td>\n",
       "      <td>Ideal</td>\n",
       "      <td>D</td>\n",
       "      <td>SI1</td>\n",
       "      <td>60.8</td>\n",
       "      <td>57.0</td>\n",
       "      <td>2757</td>\n",
       "      <td>5.75</td>\n",
       "      <td>5.76</td>\n",
       "      <td>3.50</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>53936</th>\n",
       "      <td>0.72</td>\n",
       "      <td>Good</td>\n",
       "      <td>D</td>\n",
       "      <td>SI1</td>\n",
       "      <td>63.1</td>\n",
       "      <td>55.0</td>\n",
       "      <td>2757</td>\n",
       "      <td>5.69</td>\n",
       "      <td>5.75</td>\n",
       "      <td>3.61</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>53937</th>\n",
       "      <td>0.70</td>\n",
       "      <td>Very Good</td>\n",
       "      <td>D</td>\n",
       "      <td>SI1</td>\n",
       "      <td>62.8</td>\n",
       "      <td>60.0</td>\n",
       "      <td>2757</td>\n",
       "      <td>5.66</td>\n",
       "      <td>5.68</td>\n",
       "      <td>3.56</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>53938</th>\n",
       "      <td>0.86</td>\n",
       "      <td>Premium</td>\n",
       "      <td>H</td>\n",
       "      <td>SI2</td>\n",
       "      <td>61.0</td>\n",
       "      <td>58.0</td>\n",
       "      <td>2757</td>\n",
       "      <td>6.15</td>\n",
       "      <td>6.12</td>\n",
       "      <td>3.74</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>53939</th>\n",
       "      <td>0.75</td>\n",
       "      <td>Ideal</td>\n",
       "      <td>D</td>\n",
       "      <td>SI2</td>\n",
       "      <td>62.2</td>\n",
       "      <td>55.0</td>\n",
       "      <td>2757</td>\n",
       "      <td>5.83</td>\n",
       "      <td>5.87</td>\n",
       "      <td>3.64</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>53940 rows × 10 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "       carat        cut color clarity  depth  table  price     x     y     z\n",
       "0       0.23      Ideal     E     SI2   61.5   55.0    326  3.95  3.98  2.43\n",
       "1       0.21    Premium     E     SI1   59.8   61.0    326  3.89  3.84  2.31\n",
       "2       0.23       Good     E     VS1   56.9   65.0    327  4.05  4.07  2.31\n",
       "3       0.29    Premium     I     VS2   62.4   58.0    334  4.20  4.23  2.63\n",
       "4       0.31       Good     J     SI2   63.3   58.0    335  4.34  4.35  2.75\n",
       "...      ...        ...   ...     ...    ...    ...    ...   ...   ...   ...\n",
       "53935   0.72      Ideal     D     SI1   60.8   57.0   2757  5.75  5.76  3.50\n",
       "53936   0.72       Good     D     SI1   63.1   55.0   2757  5.69  5.75  3.61\n",
       "53937   0.70  Very Good     D     SI1   62.8   60.0   2757  5.66  5.68  3.56\n",
       "53938   0.86    Premium     H     SI2   61.0   58.0   2757  6.15  6.12  3.74\n",
       "53939   0.75      Ideal     D     SI2   62.2   55.0   2757  5.83  5.87  3.64\n",
       "\n",
       "[53940 rows x 10 columns]"
      ]
     },
     "execution_count": 4,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# (1) 전체 보기 : data\n",
    "diamonds"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "id": "4213a957",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>carat</th>\n",
       "      <th>cut</th>\n",
       "      <th>color</th>\n",
       "      <th>clarity</th>\n",
       "      <th>depth</th>\n",
       "      <th>table</th>\n",
       "      <th>price</th>\n",
       "      <th>x</th>\n",
       "      <th>y</th>\n",
       "      <th>z</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>0.23</td>\n",
       "      <td>Ideal</td>\n",
       "      <td>E</td>\n",
       "      <td>SI2</td>\n",
       "      <td>61.5</td>\n",
       "      <td>55.0</td>\n",
       "      <td>326</td>\n",
       "      <td>3.95</td>\n",
       "      <td>3.98</td>\n",
       "      <td>2.43</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>0.21</td>\n",
       "      <td>Premium</td>\n",
       "      <td>E</td>\n",
       "      <td>SI1</td>\n",
       "      <td>59.8</td>\n",
       "      <td>61.0</td>\n",
       "      <td>326</td>\n",
       "      <td>3.89</td>\n",
       "      <td>3.84</td>\n",
       "      <td>2.31</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>0.23</td>\n",
       "      <td>Good</td>\n",
       "      <td>E</td>\n",
       "      <td>VS1</td>\n",
       "      <td>56.9</td>\n",
       "      <td>65.0</td>\n",
       "      <td>327</td>\n",
       "      <td>4.05</td>\n",
       "      <td>4.07</td>\n",
       "      <td>2.31</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>0.29</td>\n",
       "      <td>Premium</td>\n",
       "      <td>I</td>\n",
       "      <td>VS2</td>\n",
       "      <td>62.4</td>\n",
       "      <td>58.0</td>\n",
       "      <td>334</td>\n",
       "      <td>4.20</td>\n",
       "      <td>4.23</td>\n",
       "      <td>2.63</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>0.31</td>\n",
       "      <td>Good</td>\n",
       "      <td>J</td>\n",
       "      <td>SI2</td>\n",
       "      <td>63.3</td>\n",
       "      <td>58.0</td>\n",
       "      <td>335</td>\n",
       "      <td>4.34</td>\n",
       "      <td>4.35</td>\n",
       "      <td>2.75</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   carat      cut color clarity  depth  table  price     x     y     z\n",
       "0   0.23    Ideal     E     SI2   61.5   55.0    326  3.95  3.98  2.43\n",
       "1   0.21  Premium     E     SI1   59.8   61.0    326  3.89  3.84  2.31\n",
       "2   0.23     Good     E     VS1   56.9   65.0    327  4.05  4.07  2.31\n",
       "3   0.29  Premium     I     VS2   62.4   58.0    334  4.20  4.23  2.63\n",
       "4   0.31     Good     J     SI2   63.3   58.0    335  4.34  4.35  2.75"
      ]
     },
     "execution_count": 5,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# (2) 일부 보기 : data.head(n = 5) or data.tail(n = 5)\n",
    "diamonds.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "id": "2283c0b3",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>carat</th>\n",
       "      <th>cut</th>\n",
       "      <th>color</th>\n",
       "      <th>clarity</th>\n",
       "      <th>depth</th>\n",
       "      <th>table</th>\n",
       "      <th>price</th>\n",
       "      <th>x</th>\n",
       "      <th>y</th>\n",
       "      <th>z</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>0.23</td>\n",
       "      <td>Ideal</td>\n",
       "      <td>E</td>\n",
       "      <td>SI2</td>\n",
       "      <td>61.5</td>\n",
       "      <td>55.0</td>\n",
       "      <td>326</td>\n",
       "      <td>3.95</td>\n",
       "      <td>3.98</td>\n",
       "      <td>2.43</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>0.21</td>\n",
       "      <td>Premium</td>\n",
       "      <td>E</td>\n",
       "      <td>SI1</td>\n",
       "      <td>59.8</td>\n",
       "      <td>61.0</td>\n",
       "      <td>326</td>\n",
       "      <td>3.89</td>\n",
       "      <td>3.84</td>\n",
       "      <td>2.31</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>0.23</td>\n",
       "      <td>Good</td>\n",
       "      <td>E</td>\n",
       "      <td>VS1</td>\n",
       "      <td>56.9</td>\n",
       "      <td>65.0</td>\n",
       "      <td>327</td>\n",
       "      <td>4.05</td>\n",
       "      <td>4.07</td>\n",
       "      <td>2.31</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>0.29</td>\n",
       "      <td>Premium</td>\n",
       "      <td>I</td>\n",
       "      <td>VS2</td>\n",
       "      <td>62.4</td>\n",
       "      <td>58.0</td>\n",
       "      <td>334</td>\n",
       "      <td>4.20</td>\n",
       "      <td>4.23</td>\n",
       "      <td>2.63</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>0.31</td>\n",
       "      <td>Good</td>\n",
       "      <td>J</td>\n",
       "      <td>SI2</td>\n",
       "      <td>63.3</td>\n",
       "      <td>58.0</td>\n",
       "      <td>335</td>\n",
       "      <td>4.34</td>\n",
       "      <td>4.35</td>\n",
       "      <td>2.75</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>0.24</td>\n",
       "      <td>Very Good</td>\n",
       "      <td>J</td>\n",
       "      <td>VVS2</td>\n",
       "      <td>62.8</td>\n",
       "      <td>57.0</td>\n",
       "      <td>336</td>\n",
       "      <td>3.94</td>\n",
       "      <td>3.96</td>\n",
       "      <td>2.48</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>0.24</td>\n",
       "      <td>Very Good</td>\n",
       "      <td>I</td>\n",
       "      <td>VVS1</td>\n",
       "      <td>62.3</td>\n",
       "      <td>57.0</td>\n",
       "      <td>336</td>\n",
       "      <td>3.95</td>\n",
       "      <td>3.98</td>\n",
       "      <td>2.47</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>0.26</td>\n",
       "      <td>Very Good</td>\n",
       "      <td>H</td>\n",
       "      <td>SI1</td>\n",
       "      <td>61.9</td>\n",
       "      <td>55.0</td>\n",
       "      <td>337</td>\n",
       "      <td>4.07</td>\n",
       "      <td>4.11</td>\n",
       "      <td>2.53</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8</th>\n",
       "      <td>0.22</td>\n",
       "      <td>Fair</td>\n",
       "      <td>E</td>\n",
       "      <td>VS2</td>\n",
       "      <td>65.1</td>\n",
       "      <td>61.0</td>\n",
       "      <td>337</td>\n",
       "      <td>3.87</td>\n",
       "      <td>3.78</td>\n",
       "      <td>2.49</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9</th>\n",
       "      <td>0.23</td>\n",
       "      <td>Very Good</td>\n",
       "      <td>H</td>\n",
       "      <td>VS1</td>\n",
       "      <td>59.4</td>\n",
       "      <td>61.0</td>\n",
       "      <td>338</td>\n",
       "      <td>4.00</td>\n",
       "      <td>4.05</td>\n",
       "      <td>2.39</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   carat        cut color clarity  depth  table  price     x     y     z\n",
       "0   0.23      Ideal     E     SI2   61.5   55.0    326  3.95  3.98  2.43\n",
       "1   0.21    Premium     E     SI1   59.8   61.0    326  3.89  3.84  2.31\n",
       "2   0.23       Good     E     VS1   56.9   65.0    327  4.05  4.07  2.31\n",
       "3   0.29    Premium     I     VS2   62.4   58.0    334  4.20  4.23  2.63\n",
       "4   0.31       Good     J     SI2   63.3   58.0    335  4.34  4.35  2.75\n",
       "5   0.24  Very Good     J    VVS2   62.8   57.0    336  3.94  3.96  2.48\n",
       "6   0.24  Very Good     I    VVS1   62.3   57.0    336  3.95  3.98  2.47\n",
       "7   0.26  Very Good     H     SI1   61.9   55.0    337  4.07  4.11  2.53\n",
       "8   0.22       Fair     E     VS2   65.1   61.0    337  3.87  3.78  2.49\n",
       "9   0.23  Very Good     H     VS1   59.4   61.0    338  4.00  4.05  2.39"
      ]
     },
     "execution_count": 6,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "diamonds.head(n = 10)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "id": "11b3ce23",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>carat</th>\n",
       "      <th>cut</th>\n",
       "      <th>color</th>\n",
       "      <th>clarity</th>\n",
       "      <th>depth</th>\n",
       "      <th>table</th>\n",
       "      <th>price</th>\n",
       "      <th>x</th>\n",
       "      <th>y</th>\n",
       "      <th>z</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>53935</th>\n",
       "      <td>0.72</td>\n",
       "      <td>Ideal</td>\n",
       "      <td>D</td>\n",
       "      <td>SI1</td>\n",
       "      <td>60.8</td>\n",
       "      <td>57.0</td>\n",
       "      <td>2757</td>\n",
       "      <td>5.75</td>\n",
       "      <td>5.76</td>\n",
       "      <td>3.50</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>53936</th>\n",
       "      <td>0.72</td>\n",
       "      <td>Good</td>\n",
       "      <td>D</td>\n",
       "      <td>SI1</td>\n",
       "      <td>63.1</td>\n",
       "      <td>55.0</td>\n",
       "      <td>2757</td>\n",
       "      <td>5.69</td>\n",
       "      <td>5.75</td>\n",
       "      <td>3.61</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>53937</th>\n",
       "      <td>0.70</td>\n",
       "      <td>Very Good</td>\n",
       "      <td>D</td>\n",
       "      <td>SI1</td>\n",
       "      <td>62.8</td>\n",
       "      <td>60.0</td>\n",
       "      <td>2757</td>\n",
       "      <td>5.66</td>\n",
       "      <td>5.68</td>\n",
       "      <td>3.56</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>53938</th>\n",
       "      <td>0.86</td>\n",
       "      <td>Premium</td>\n",
       "      <td>H</td>\n",
       "      <td>SI2</td>\n",
       "      <td>61.0</td>\n",
       "      <td>58.0</td>\n",
       "      <td>2757</td>\n",
       "      <td>6.15</td>\n",
       "      <td>6.12</td>\n",
       "      <td>3.74</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>53939</th>\n",
       "      <td>0.75</td>\n",
       "      <td>Ideal</td>\n",
       "      <td>D</td>\n",
       "      <td>SI2</td>\n",
       "      <td>62.2</td>\n",
       "      <td>55.0</td>\n",
       "      <td>2757</td>\n",
       "      <td>5.83</td>\n",
       "      <td>5.87</td>\n",
       "      <td>3.64</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "       carat        cut color clarity  depth  table  price     x     y     z\n",
       "53935   0.72      Ideal     D     SI1   60.8   57.0   2757  5.75  5.76  3.50\n",
       "53936   0.72       Good     D     SI1   63.1   55.0   2757  5.69  5.75  3.61\n",
       "53937   0.70  Very Good     D     SI1   62.8   60.0   2757  5.66  5.68  3.56\n",
       "53938   0.86    Premium     H     SI2   61.0   58.0   2757  6.15  6.12  3.74\n",
       "53939   0.75      Ideal     D     SI2   62.2   55.0   2757  5.83  5.87  3.64"
      ]
     },
     "execution_count": 7,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "diamonds.tail()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "id": "63d6b57e",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>carat</th>\n",
       "      <th>cut</th>\n",
       "      <th>color</th>\n",
       "      <th>clarity</th>\n",
       "      <th>depth</th>\n",
       "      <th>table</th>\n",
       "      <th>price</th>\n",
       "      <th>x</th>\n",
       "      <th>y</th>\n",
       "      <th>z</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>53930</th>\n",
       "      <td>0.71</td>\n",
       "      <td>Premium</td>\n",
       "      <td>E</td>\n",
       "      <td>SI1</td>\n",
       "      <td>60.5</td>\n",
       "      <td>55.0</td>\n",
       "      <td>2756</td>\n",
       "      <td>5.79</td>\n",
       "      <td>5.74</td>\n",
       "      <td>3.49</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>53931</th>\n",
       "      <td>0.71</td>\n",
       "      <td>Premium</td>\n",
       "      <td>F</td>\n",
       "      <td>SI1</td>\n",
       "      <td>59.8</td>\n",
       "      <td>62.0</td>\n",
       "      <td>2756</td>\n",
       "      <td>5.74</td>\n",
       "      <td>5.73</td>\n",
       "      <td>3.43</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>53932</th>\n",
       "      <td>0.70</td>\n",
       "      <td>Very Good</td>\n",
       "      <td>E</td>\n",
       "      <td>VS2</td>\n",
       "      <td>60.5</td>\n",
       "      <td>59.0</td>\n",
       "      <td>2757</td>\n",
       "      <td>5.71</td>\n",
       "      <td>5.76</td>\n",
       "      <td>3.47</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>53933</th>\n",
       "      <td>0.70</td>\n",
       "      <td>Very Good</td>\n",
       "      <td>E</td>\n",
       "      <td>VS2</td>\n",
       "      <td>61.2</td>\n",
       "      <td>59.0</td>\n",
       "      <td>2757</td>\n",
       "      <td>5.69</td>\n",
       "      <td>5.72</td>\n",
       "      <td>3.49</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>53934</th>\n",
       "      <td>0.72</td>\n",
       "      <td>Premium</td>\n",
       "      <td>D</td>\n",
       "      <td>SI1</td>\n",
       "      <td>62.7</td>\n",
       "      <td>59.0</td>\n",
       "      <td>2757</td>\n",
       "      <td>5.69</td>\n",
       "      <td>5.73</td>\n",
       "      <td>3.58</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>53935</th>\n",
       "      <td>0.72</td>\n",
       "      <td>Ideal</td>\n",
       "      <td>D</td>\n",
       "      <td>SI1</td>\n",
       "      <td>60.8</td>\n",
       "      <td>57.0</td>\n",
       "      <td>2757</td>\n",
       "      <td>5.75</td>\n",
       "      <td>5.76</td>\n",
       "      <td>3.50</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>53936</th>\n",
       "      <td>0.72</td>\n",
       "      <td>Good</td>\n",
       "      <td>D</td>\n",
       "      <td>SI1</td>\n",
       "      <td>63.1</td>\n",
       "      <td>55.0</td>\n",
       "      <td>2757</td>\n",
       "      <td>5.69</td>\n",
       "      <td>5.75</td>\n",
       "      <td>3.61</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>53937</th>\n",
       "      <td>0.70</td>\n",
       "      <td>Very Good</td>\n",
       "      <td>D</td>\n",
       "      <td>SI1</td>\n",
       "      <td>62.8</td>\n",
       "      <td>60.0</td>\n",
       "      <td>2757</td>\n",
       "      <td>5.66</td>\n",
       "      <td>5.68</td>\n",
       "      <td>3.56</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>53938</th>\n",
       "      <td>0.86</td>\n",
       "      <td>Premium</td>\n",
       "      <td>H</td>\n",
       "      <td>SI2</td>\n",
       "      <td>61.0</td>\n",
       "      <td>58.0</td>\n",
       "      <td>2757</td>\n",
       "      <td>6.15</td>\n",
       "      <td>6.12</td>\n",
       "      <td>3.74</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>53939</th>\n",
       "      <td>0.75</td>\n",
       "      <td>Ideal</td>\n",
       "      <td>D</td>\n",
       "      <td>SI2</td>\n",
       "      <td>62.2</td>\n",
       "      <td>55.0</td>\n",
       "      <td>2757</td>\n",
       "      <td>5.83</td>\n",
       "      <td>5.87</td>\n",
       "      <td>3.64</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "       carat        cut color clarity  depth  table  price     x     y     z\n",
       "53930   0.71    Premium     E     SI1   60.5   55.0   2756  5.79  5.74  3.49\n",
       "53931   0.71    Premium     F     SI1   59.8   62.0   2756  5.74  5.73  3.43\n",
       "53932   0.70  Very Good     E     VS2   60.5   59.0   2757  5.71  5.76  3.47\n",
       "53933   0.70  Very Good     E     VS2   61.2   59.0   2757  5.69  5.72  3.49\n",
       "53934   0.72    Premium     D     SI1   62.7   59.0   2757  5.69  5.73  3.58\n",
       "53935   0.72      Ideal     D     SI1   60.8   57.0   2757  5.75  5.76  3.50\n",
       "53936   0.72       Good     D     SI1   63.1   55.0   2757  5.69  5.75  3.61\n",
       "53937   0.70  Very Good     D     SI1   62.8   60.0   2757  5.66  5.68  3.56\n",
       "53938   0.86    Premium     H     SI2   61.0   58.0   2757  6.15  6.12  3.74\n",
       "53939   0.75      Ideal     D     SI2   62.2   55.0   2757  5.83  5.87  3.64"
      ]
     },
     "execution_count": 8,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "diamonds.tail(n = 10)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "d27a64eb",
   "metadata": {},
   "source": [
    "#### 2. 데이터의 정보(Information)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "id": "14678143",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "<class 'pandas.core.frame.DataFrame'>\n",
      "RangeIndex: 53940 entries, 0 to 53939\n",
      "Data columns (total 10 columns):\n",
      " #   Column   Non-Null Count  Dtype   \n",
      "---  ------   --------------  -----   \n",
      " 0   carat    53940 non-null  float64 \n",
      " 1   cut      53940 non-null  category\n",
      " 2   color    53940 non-null  category\n",
      " 3   clarity  53940 non-null  category\n",
      " 4   depth    53940 non-null  float64 \n",
      " 5   table    53940 non-null  float64 \n",
      " 6   price    53940 non-null  int64   \n",
      " 7   x        53940 non-null  float64 \n",
      " 8   y        53940 non-null  float64 \n",
      " 9   z        53940 non-null  float64 \n",
      "dtypes: category(3), float64(6), int64(1)\n",
      "memory usage: 3.0 MB\n"
     ]
    }
   ],
   "source": [
    "# data.info()\n",
    "diamonds.info()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "id": "e9a01b04",
   "metadata": {},
   "outputs": [],
   "source": [
    "# <class 'pandas.core.frame.DataFrame'>\n",
    "# DataFrame : 행과 열로 구성됨, 2차원 구조"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "id": "9b1b420f",
   "metadata": {},
   "outputs": [],
   "source": [
    "# RangeIndex: 53940 entries, 0 to 53939\n",
    "# Index : 행\n",
    "# 행    : 0 ~ 53939 : 53,940개"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "id": "4098f9e2",
   "metadata": {},
   "outputs": [],
   "source": [
    "# Data columns (total 10 columns)\n",
    "# 열 \n",
    "# 열의 개수 : 10개"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "id": "4a9982d6",
   "metadata": {},
   "outputs": [],
   "source": [
    "# Null     : Missing Value : 값이 없는 것 : 결측치, 결측값\n",
    "# Non-Null : 값이 있는 것\n",
    "\n",
    "# 참고 : Missing Value 탐지, 처리\n",
    "# 처리 : 삭제(행, 열), 대체(Imputation) : 다른 값으로 바꾸는 것"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "c4caa359",
   "metadata": {},
   "source": [
    "#### 3. 데이터의 속성"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "id": "e291f1d3",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "53940"
      ]
     },
     "execution_count": 19,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# (1) 데이터의 행과 열, 차원(Dimension)\n",
    "# data.shape\n",
    "diamonds.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "id": "39142f92",
   "metadata": {},
   "outputs": [],
   "source": [
    "# (53940, 10)\n",
    "# 53490 : 행의 개수\n",
    "# 10    : 열의 개수\n",
    "# 숫자가 2개 출력 : 2차원"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "id": "538c22ba",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Index(['carat', 'cut', 'color', 'clarity', 'depth', 'table', 'price', 'x', 'y',\n",
       "       'z'],\n",
       "      dtype='object')"
      ]
     },
     "execution_count": 20,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# (2) 열의 이름 : data.columns\n",
    "diamonds.columns"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "id": "9f31c4fd",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "carat       float64\n",
       "cut        category\n",
       "color      category\n",
       "clarity    category\n",
       "depth       float64\n",
       "table       float64\n",
       "price         int64\n",
       "x           float64\n",
       "y           float64\n",
       "z           float64\n",
       "dtype: object"
      ]
     },
     "execution_count": 21,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# (3) 데이터의 유형 : data.dtypes\n",
    "diamonds.dtypes"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "ef4e108e",
   "metadata": {},
   "source": [
    "#### 4. 슬라이싱(Slicing)\n",
    "- 열\n",
    "- 행\n",
    "- 행과 열"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "id": "e43cead8",
   "metadata": {},
   "outputs": [],
   "source": [
    "# data.loc[행, 열]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "id": "05403707",
   "metadata": {},
   "outputs": [],
   "source": [
    "# (1) 열"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "id": "fb766569",
   "metadata": {},
   "outputs": [],
   "source": [
    "# i. 하나의 열 : data[\"열의 이름\"] or data.열의이름"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "id": "f0506d85",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0        0.23\n",
       "1        0.21\n",
       "2        0.23\n",
       "3        0.29\n",
       "4        0.31\n",
       "         ... \n",
       "53935    0.72\n",
       "53936    0.72\n",
       "53937    0.70\n",
       "53938    0.86\n",
       "53939    0.75\n",
       "Name: carat, Length: 53940, dtype: float64"
      ]
     },
     "execution_count": 25,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "diamonds[\"carat\"]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "id": "6612a98e",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0        0.23\n",
       "1        0.21\n",
       "2        0.23\n",
       "3        0.29\n",
       "4        0.31\n",
       "         ... \n",
       "53935    0.72\n",
       "53936    0.72\n",
       "53937    0.70\n",
       "53938    0.86\n",
       "53939    0.75\n",
       "Name: carat, Length: 53940, dtype: float64"
      ]
     },
     "execution_count": 26,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "diamonds.carat"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "id": "f90e9358",
   "metadata": {},
   "outputs": [],
   "source": [
    "# ii. 2개 이상의 열\n",
    "# data[[\"열의이름1\", \"열의이름2\", ...]]\n",
    "# data.loc[: , [\"열의이름1\", \"열의이름2\", ...]]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "id": "4e8e0a6e",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>carat</th>\n",
       "      <th>price</th>\n",
       "      <th>x</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>0.23</td>\n",
       "      <td>326</td>\n",
       "      <td>3.95</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>0.21</td>\n",
       "      <td>326</td>\n",
       "      <td>3.89</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>0.23</td>\n",
       "      <td>327</td>\n",
       "      <td>4.05</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>0.29</td>\n",
       "      <td>334</td>\n",
       "      <td>4.20</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>0.31</td>\n",
       "      <td>335</td>\n",
       "      <td>4.34</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>53935</th>\n",
       "      <td>0.72</td>\n",
       "      <td>2757</td>\n",
       "      <td>5.75</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>53936</th>\n",
       "      <td>0.72</td>\n",
       "      <td>2757</td>\n",
       "      <td>5.69</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>53937</th>\n",
       "      <td>0.70</td>\n",
       "      <td>2757</td>\n",
       "      <td>5.66</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>53938</th>\n",
       "      <td>0.86</td>\n",
       "      <td>2757</td>\n",
       "      <td>6.15</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>53939</th>\n",
       "      <td>0.75</td>\n",
       "      <td>2757</td>\n",
       "      <td>5.83</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>53940 rows × 3 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "       carat  price     x\n",
       "0       0.23    326  3.95\n",
       "1       0.21    326  3.89\n",
       "2       0.23    327  4.05\n",
       "3       0.29    334  4.20\n",
       "4       0.31    335  4.34\n",
       "...      ...    ...   ...\n",
       "53935   0.72   2757  5.75\n",
       "53936   0.72   2757  5.69\n",
       "53937   0.70   2757  5.66\n",
       "53938   0.86   2757  6.15\n",
       "53939   0.75   2757  5.83\n",
       "\n",
       "[53940 rows x 3 columns]"
      ]
     },
     "execution_count": 28,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# \"carat\", \"price\", \"x\"\n",
    "diamonds[[\"carat\", \"price\", \"x\"]]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "id": "7536f13a",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>carat</th>\n",
       "      <th>price</th>\n",
       "      <th>x</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>0.23</td>\n",
       "      <td>326</td>\n",
       "      <td>3.95</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>0.21</td>\n",
       "      <td>326</td>\n",
       "      <td>3.89</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>0.23</td>\n",
       "      <td>327</td>\n",
       "      <td>4.05</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>0.29</td>\n",
       "      <td>334</td>\n",
       "      <td>4.20</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>0.31</td>\n",
       "      <td>335</td>\n",
       "      <td>4.34</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>53935</th>\n",
       "      <td>0.72</td>\n",
       "      <td>2757</td>\n",
       "      <td>5.75</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>53936</th>\n",
       "      <td>0.72</td>\n",
       "      <td>2757</td>\n",
       "      <td>5.69</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>53937</th>\n",
       "      <td>0.70</td>\n",
       "      <td>2757</td>\n",
       "      <td>5.66</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>53938</th>\n",
       "      <td>0.86</td>\n",
       "      <td>2757</td>\n",
       "      <td>6.15</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>53939</th>\n",
       "      <td>0.75</td>\n",
       "      <td>2757</td>\n",
       "      <td>5.83</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>53940 rows × 3 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "       carat  price     x\n",
       "0       0.23    326  3.95\n",
       "1       0.21    326  3.89\n",
       "2       0.23    327  4.05\n",
       "3       0.29    334  4.20\n",
       "4       0.31    335  4.34\n",
       "...      ...    ...   ...\n",
       "53935   0.72   2757  5.75\n",
       "53936   0.72   2757  5.69\n",
       "53937   0.70   2757  5.66\n",
       "53938   0.86   2757  6.15\n",
       "53939   0.75   2757  5.83\n",
       "\n",
       "[53940 rows x 3 columns]"
      ]
     },
     "execution_count": 29,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "diamonds.loc[:, [\"carat\", \"price\", \"x\"]]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "id": "b93e9b2e",
   "metadata": {},
   "outputs": [],
   "source": [
    "# iii. 2개 이상의 열 : 열의 이름에 규칙(패턴)이 있는 경우\n",
    "# data.filter(regex = \"pattern\")\n",
    "\n",
    "# 참고 \n",
    "# regex : regular expression : 정규 표현식"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "id": "dfaa5682",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>carat</th>\n",
       "      <th>cut</th>\n",
       "      <th>color</th>\n",
       "      <th>clarity</th>\n",
       "      <th>price</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>0.23</td>\n",
       "      <td>Ideal</td>\n",
       "      <td>E</td>\n",
       "      <td>SI2</td>\n",
       "      <td>326</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>0.21</td>\n",
       "      <td>Premium</td>\n",
       "      <td>E</td>\n",
       "      <td>SI1</td>\n",
       "      <td>326</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>0.23</td>\n",
       "      <td>Good</td>\n",
       "      <td>E</td>\n",
       "      <td>VS1</td>\n",
       "      <td>327</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>0.29</td>\n",
       "      <td>Premium</td>\n",
       "      <td>I</td>\n",
       "      <td>VS2</td>\n",
       "      <td>334</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>0.31</td>\n",
       "      <td>Good</td>\n",
       "      <td>J</td>\n",
       "      <td>SI2</td>\n",
       "      <td>335</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>53935</th>\n",
       "      <td>0.72</td>\n",
       "      <td>Ideal</td>\n",
       "      <td>D</td>\n",
       "      <td>SI1</td>\n",
       "      <td>2757</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>53936</th>\n",
       "      <td>0.72</td>\n",
       "      <td>Good</td>\n",
       "      <td>D</td>\n",
       "      <td>SI1</td>\n",
       "      <td>2757</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>53937</th>\n",
       "      <td>0.70</td>\n",
       "      <td>Very Good</td>\n",
       "      <td>D</td>\n",
       "      <td>SI1</td>\n",
       "      <td>2757</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>53938</th>\n",
       "      <td>0.86</td>\n",
       "      <td>Premium</td>\n",
       "      <td>H</td>\n",
       "      <td>SI2</td>\n",
       "      <td>2757</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>53939</th>\n",
       "      <td>0.75</td>\n",
       "      <td>Ideal</td>\n",
       "      <td>D</td>\n",
       "      <td>SI2</td>\n",
       "      <td>2757</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>53940 rows × 5 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "       carat        cut color clarity  price\n",
       "0       0.23      Ideal     E     SI2    326\n",
       "1       0.21    Premium     E     SI1    326\n",
       "2       0.23       Good     E     VS1    327\n",
       "3       0.29    Premium     I     VS2    334\n",
       "4       0.31       Good     J     SI2    335\n",
       "...      ...        ...   ...     ...    ...\n",
       "53935   0.72      Ideal     D     SI1   2757\n",
       "53936   0.72       Good     D     SI1   2757\n",
       "53937   0.70  Very Good     D     SI1   2757\n",
       "53938   0.86    Premium     H     SI2   2757\n",
       "53939   0.75      Ideal     D     SI2   2757\n",
       "\n",
       "[53940 rows x 5 columns]"
      ]
     },
     "execution_count": 31,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# 포함 : 특정 문자를 포함하고 있는 열\n",
    "# 열의 이름 중에서 \"c\"라는 문자를 포함하고 있는 열\n",
    "diamonds.filter(regex = \"c\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "id": "8b0f0972",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>price</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>326</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>326</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>327</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>334</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>335</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>53935</th>\n",
       "      <td>2757</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>53936</th>\n",
       "      <td>2757</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>53937</th>\n",
       "      <td>2757</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>53938</th>\n",
       "      <td>2757</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>53939</th>\n",
       "      <td>2757</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>53940 rows × 1 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "       price\n",
       "0        326\n",
       "1        326\n",
       "2        327\n",
       "3        334\n",
       "4        335\n",
       "...      ...\n",
       "53935   2757\n",
       "53936   2757\n",
       "53937   2757\n",
       "53938   2757\n",
       "53939   2757\n",
       "\n",
       "[53940 rows x 1 columns]"
      ]
     },
     "execution_count": 32,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# 열의 이름 중에 \"ce\" 문자를 포함하는 열\n",
    "diamonds.filter(regex = \"ce\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 33,
   "id": "f7ebb21f",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>carat</th>\n",
       "      <th>cut</th>\n",
       "      <th>color</th>\n",
       "      <th>clarity</th>\n",
       "      <th>depth</th>\n",
       "      <th>table</th>\n",
       "      <th>price</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>0.23</td>\n",
       "      <td>Ideal</td>\n",
       "      <td>E</td>\n",
       "      <td>SI2</td>\n",
       "      <td>61.5</td>\n",
       "      <td>55.0</td>\n",
       "      <td>326</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>0.21</td>\n",
       "      <td>Premium</td>\n",
       "      <td>E</td>\n",
       "      <td>SI1</td>\n",
       "      <td>59.8</td>\n",
       "      <td>61.0</td>\n",
       "      <td>326</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>0.23</td>\n",
       "      <td>Good</td>\n",
       "      <td>E</td>\n",
       "      <td>VS1</td>\n",
       "      <td>56.9</td>\n",
       "      <td>65.0</td>\n",
       "      <td>327</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>0.29</td>\n",
       "      <td>Premium</td>\n",
       "      <td>I</td>\n",
       "      <td>VS2</td>\n",
       "      <td>62.4</td>\n",
       "      <td>58.0</td>\n",
       "      <td>334</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>0.31</td>\n",
       "      <td>Good</td>\n",
       "      <td>J</td>\n",
       "      <td>SI2</td>\n",
       "      <td>63.3</td>\n",
       "      <td>58.0</td>\n",
       "      <td>335</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",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>53935</th>\n",
       "      <td>0.72</td>\n",
       "      <td>Ideal</td>\n",
       "      <td>D</td>\n",
       "      <td>SI1</td>\n",
       "      <td>60.8</td>\n",
       "      <td>57.0</td>\n",
       "      <td>2757</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>53936</th>\n",
       "      <td>0.72</td>\n",
       "      <td>Good</td>\n",
       "      <td>D</td>\n",
       "      <td>SI1</td>\n",
       "      <td>63.1</td>\n",
       "      <td>55.0</td>\n",
       "      <td>2757</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>53937</th>\n",
       "      <td>0.70</td>\n",
       "      <td>Very Good</td>\n",
       "      <td>D</td>\n",
       "      <td>SI1</td>\n",
       "      <td>62.8</td>\n",
       "      <td>60.0</td>\n",
       "      <td>2757</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>53938</th>\n",
       "      <td>0.86</td>\n",
       "      <td>Premium</td>\n",
       "      <td>H</td>\n",
       "      <td>SI2</td>\n",
       "      <td>61.0</td>\n",
       "      <td>58.0</td>\n",
       "      <td>2757</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>53939</th>\n",
       "      <td>0.75</td>\n",
       "      <td>Ideal</td>\n",
       "      <td>D</td>\n",
       "      <td>SI2</td>\n",
       "      <td>62.2</td>\n",
       "      <td>55.0</td>\n",
       "      <td>2757</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>53940 rows × 7 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "       carat        cut color clarity  depth  table  price\n",
       "0       0.23      Ideal     E     SI2   61.5   55.0    326\n",
       "1       0.21    Premium     E     SI1   59.8   61.0    326\n",
       "2       0.23       Good     E     VS1   56.9   65.0    327\n",
       "3       0.29    Premium     I     VS2   62.4   58.0    334\n",
       "4       0.31       Good     J     SI2   63.3   58.0    335\n",
       "...      ...        ...   ...     ...    ...    ...    ...\n",
       "53935   0.72      Ideal     D     SI1   60.8   57.0   2757\n",
       "53936   0.72       Good     D     SI1   63.1   55.0   2757\n",
       "53937   0.70  Very Good     D     SI1   62.8   60.0   2757\n",
       "53938   0.86    Premium     H     SI2   61.0   58.0   2757\n",
       "53939   0.75      Ideal     D     SI2   62.2   55.0   2757\n",
       "\n",
       "[53940 rows x 7 columns]"
      ]
     },
     "execution_count": 33,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# 열의 이름 중에 \"c\" 또는 \"e\"라는 문자를 포함하는 열\n",
    "diamonds.filter(regex = \"[ce]\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 34,
   "id": "98ae83af",
   "metadata": {},
   "outputs": [],
   "source": [
    "# 시작 : 특정 문자로 시작하는 열의 이름\n",
    "# data.filter(regex = \"^pattern\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 35,
   "id": "4e740c74",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>carat</th>\n",
       "      <th>cut</th>\n",
       "      <th>color</th>\n",
       "      <th>clarity</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>0.23</td>\n",
       "      <td>Ideal</td>\n",
       "      <td>E</td>\n",
       "      <td>SI2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>0.21</td>\n",
       "      <td>Premium</td>\n",
       "      <td>E</td>\n",
       "      <td>SI1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>0.23</td>\n",
       "      <td>Good</td>\n",
       "      <td>E</td>\n",
       "      <td>VS1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>0.29</td>\n",
       "      <td>Premium</td>\n",
       "      <td>I</td>\n",
       "      <td>VS2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>0.31</td>\n",
       "      <td>Good</td>\n",
       "      <td>J</td>\n",
       "      <td>SI2</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>53935</th>\n",
       "      <td>0.72</td>\n",
       "      <td>Ideal</td>\n",
       "      <td>D</td>\n",
       "      <td>SI1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>53936</th>\n",
       "      <td>0.72</td>\n",
       "      <td>Good</td>\n",
       "      <td>D</td>\n",
       "      <td>SI1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>53937</th>\n",
       "      <td>0.70</td>\n",
       "      <td>Very Good</td>\n",
       "      <td>D</td>\n",
       "      <td>SI1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>53938</th>\n",
       "      <td>0.86</td>\n",
       "      <td>Premium</td>\n",
       "      <td>H</td>\n",
       "      <td>SI2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>53939</th>\n",
       "      <td>0.75</td>\n",
       "      <td>Ideal</td>\n",
       "      <td>D</td>\n",
       "      <td>SI2</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>53940 rows × 4 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "       carat        cut color clarity\n",
       "0       0.23      Ideal     E     SI2\n",
       "1       0.21    Premium     E     SI1\n",
       "2       0.23       Good     E     VS1\n",
       "3       0.29    Premium     I     VS2\n",
       "4       0.31       Good     J     SI2\n",
       "...      ...        ...   ...     ...\n",
       "53935   0.72      Ideal     D     SI1\n",
       "53936   0.72       Good     D     SI1\n",
       "53937   0.70  Very Good     D     SI1\n",
       "53938   0.86    Premium     H     SI2\n",
       "53939   0.75      Ideal     D     SI2\n",
       "\n",
       "[53940 rows x 4 columns]"
      ]
     },
     "execution_count": 35,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# \"c\"라는 문자로 시작하는 열\n",
    "diamonds.filter(regex = \"^c\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 36,
   "id": "2a1e6a14",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>price</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>326</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>326</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>327</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>334</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>335</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>53935</th>\n",
       "      <td>2757</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>53936</th>\n",
       "      <td>2757</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>53937</th>\n",
       "      <td>2757</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>53938</th>\n",
       "      <td>2757</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>53939</th>\n",
       "      <td>2757</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>53940 rows × 1 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "       price\n",
       "0        326\n",
       "1        326\n",
       "2        327\n",
       "3        334\n",
       "4        335\n",
       "...      ...\n",
       "53935   2757\n",
       "53936   2757\n",
       "53937   2757\n",
       "53938   2757\n",
       "53939   2757\n",
       "\n",
       "[53940 rows x 1 columns]"
      ]
     },
     "execution_count": 36,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# \"pr\"로 시작하는 열\n",
    "diamonds.filter(regex = \"^pr\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 37,
   "id": "c461b72a",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>carat</th>\n",
       "      <th>cut</th>\n",
       "      <th>color</th>\n",
       "      <th>clarity</th>\n",
       "      <th>price</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>0.23</td>\n",
       "      <td>Ideal</td>\n",
       "      <td>E</td>\n",
       "      <td>SI2</td>\n",
       "      <td>326</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>0.21</td>\n",
       "      <td>Premium</td>\n",
       "      <td>E</td>\n",
       "      <td>SI1</td>\n",
       "      <td>326</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>0.23</td>\n",
       "      <td>Good</td>\n",
       "      <td>E</td>\n",
       "      <td>VS1</td>\n",
       "      <td>327</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>0.29</td>\n",
       "      <td>Premium</td>\n",
       "      <td>I</td>\n",
       "      <td>VS2</td>\n",
       "      <td>334</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>0.31</td>\n",
       "      <td>Good</td>\n",
       "      <td>J</td>\n",
       "      <td>SI2</td>\n",
       "      <td>335</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>53935</th>\n",
       "      <td>0.72</td>\n",
       "      <td>Ideal</td>\n",
       "      <td>D</td>\n",
       "      <td>SI1</td>\n",
       "      <td>2757</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>53936</th>\n",
       "      <td>0.72</td>\n",
       "      <td>Good</td>\n",
       "      <td>D</td>\n",
       "      <td>SI1</td>\n",
       "      <td>2757</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>53937</th>\n",
       "      <td>0.70</td>\n",
       "      <td>Very Good</td>\n",
       "      <td>D</td>\n",
       "      <td>SI1</td>\n",
       "      <td>2757</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>53938</th>\n",
       "      <td>0.86</td>\n",
       "      <td>Premium</td>\n",
       "      <td>H</td>\n",
       "      <td>SI2</td>\n",
       "      <td>2757</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>53939</th>\n",
       "      <td>0.75</td>\n",
       "      <td>Ideal</td>\n",
       "      <td>D</td>\n",
       "      <td>SI2</td>\n",
       "      <td>2757</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>53940 rows × 5 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "       carat        cut color clarity  price\n",
       "0       0.23      Ideal     E     SI2    326\n",
       "1       0.21    Premium     E     SI1    326\n",
       "2       0.23       Good     E     VS1    327\n",
       "3       0.29    Premium     I     VS2    334\n",
       "4       0.31       Good     J     SI2    335\n",
       "...      ...        ...   ...     ...    ...\n",
       "53935   0.72      Ideal     D     SI1   2757\n",
       "53936   0.72       Good     D     SI1   2757\n",
       "53937   0.70  Very Good     D     SI1   2757\n",
       "53938   0.86    Premium     H     SI2   2757\n",
       "53939   0.75      Ideal     D     SI2   2757\n",
       "\n",
       "[53940 rows x 5 columns]"
      ]
     },
     "execution_count": 37,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# \"c\" 또는 \"p\"로 시작하는 열\n",
    "diamonds.filter(regex = \"^[cp]\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 38,
   "id": "507b5ed9",
   "metadata": {},
   "outputs": [],
   "source": [
    "# 특정한 문자로 끝나는 열 \n",
    "# data.filter(regex = \"pattern$\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 39,
   "id": "c8f34c82",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>table</th>\n",
       "      <th>price</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>55.0</td>\n",
       "      <td>326</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>61.0</td>\n",
       "      <td>326</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>65.0</td>\n",
       "      <td>327</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>58.0</td>\n",
       "      <td>334</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>58.0</td>\n",
       "      <td>335</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>53935</th>\n",
       "      <td>57.0</td>\n",
       "      <td>2757</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>53936</th>\n",
       "      <td>55.0</td>\n",
       "      <td>2757</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>53937</th>\n",
       "      <td>60.0</td>\n",
       "      <td>2757</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>53938</th>\n",
       "      <td>58.0</td>\n",
       "      <td>2757</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>53939</th>\n",
       "      <td>55.0</td>\n",
       "      <td>2757</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>53940 rows × 2 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "       table  price\n",
       "0       55.0    326\n",
       "1       61.0    326\n",
       "2       65.0    327\n",
       "3       58.0    334\n",
       "4       58.0    335\n",
       "...      ...    ...\n",
       "53935   57.0   2757\n",
       "53936   55.0   2757\n",
       "53937   60.0   2757\n",
       "53938   58.0   2757\n",
       "53939   55.0   2757\n",
       "\n",
       "[53940 rows x 2 columns]"
      ]
     },
     "execution_count": 39,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# \"e\"로 끝나는 열\n",
    "diamonds.filter(regex = \"e$\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 40,
   "id": "7902983a",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>table</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>55.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>61.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>65.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>58.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>58.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>53935</th>\n",
       "      <td>57.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>53936</th>\n",
       "      <td>55.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>53937</th>\n",
       "      <td>60.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>53938</th>\n",
       "      <td>58.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>53939</th>\n",
       "      <td>55.0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>53940 rows × 1 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "       table\n",
       "0       55.0\n",
       "1       61.0\n",
       "2       65.0\n",
       "3       58.0\n",
       "4       58.0\n",
       "...      ...\n",
       "53935   57.0\n",
       "53936   55.0\n",
       "53937   60.0\n",
       "53938   58.0\n",
       "53939   55.0\n",
       "\n",
       "[53940 rows x 1 columns]"
      ]
     },
     "execution_count": 40,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# \"le\"로 끝나는 열\n",
    "diamonds.filter(regex = \"le$\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 41,
   "id": "53804f28",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>carat</th>\n",
       "      <th>cut</th>\n",
       "      <th>table</th>\n",
       "      <th>price</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>0.23</td>\n",
       "      <td>Ideal</td>\n",
       "      <td>55.0</td>\n",
       "      <td>326</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>0.21</td>\n",
       "      <td>Premium</td>\n",
       "      <td>61.0</td>\n",
       "      <td>326</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>0.23</td>\n",
       "      <td>Good</td>\n",
       "      <td>65.0</td>\n",
       "      <td>327</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>0.29</td>\n",
       "      <td>Premium</td>\n",
       "      <td>58.0</td>\n",
       "      <td>334</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>0.31</td>\n",
       "      <td>Good</td>\n",
       "      <td>58.0</td>\n",
       "      <td>335</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>53935</th>\n",
       "      <td>0.72</td>\n",
       "      <td>Ideal</td>\n",
       "      <td>57.0</td>\n",
       "      <td>2757</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>53936</th>\n",
       "      <td>0.72</td>\n",
       "      <td>Good</td>\n",
       "      <td>55.0</td>\n",
       "      <td>2757</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>53937</th>\n",
       "      <td>0.70</td>\n",
       "      <td>Very Good</td>\n",
       "      <td>60.0</td>\n",
       "      <td>2757</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>53938</th>\n",
       "      <td>0.86</td>\n",
       "      <td>Premium</td>\n",
       "      <td>58.0</td>\n",
       "      <td>2757</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>53939</th>\n",
       "      <td>0.75</td>\n",
       "      <td>Ideal</td>\n",
       "      <td>55.0</td>\n",
       "      <td>2757</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>53940 rows × 4 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "       carat        cut  table  price\n",
       "0       0.23      Ideal   55.0    326\n",
       "1       0.21    Premium   61.0    326\n",
       "2       0.23       Good   65.0    327\n",
       "3       0.29    Premium   58.0    334\n",
       "4       0.31       Good   58.0    335\n",
       "...      ...        ...    ...    ...\n",
       "53935   0.72      Ideal   57.0   2757\n",
       "53936   0.72       Good   55.0   2757\n",
       "53937   0.70  Very Good   60.0   2757\n",
       "53938   0.86    Premium   58.0   2757\n",
       "53939   0.75      Ideal   55.0   2757\n",
       "\n",
       "[53940 rows x 4 columns]"
      ]
     },
     "execution_count": 41,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# \"t\" 또는 \"e\"로 끝나는 열\n",
    "diamonds.filter(regex = \"[te]$\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 43,
   "id": "0ad7277c",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>cut</th>\n",
       "      <th>color</th>\n",
       "      <th>clarity</th>\n",
       "      <th>depth</th>\n",
       "      <th>table</th>\n",
       "      <th>price</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>Ideal</td>\n",
       "      <td>E</td>\n",
       "      <td>SI2</td>\n",
       "      <td>61.5</td>\n",
       "      <td>55.0</td>\n",
       "      <td>326</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>Premium</td>\n",
       "      <td>E</td>\n",
       "      <td>SI1</td>\n",
       "      <td>59.8</td>\n",
       "      <td>61.0</td>\n",
       "      <td>326</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>Good</td>\n",
       "      <td>E</td>\n",
       "      <td>VS1</td>\n",
       "      <td>56.9</td>\n",
       "      <td>65.0</td>\n",
       "      <td>327</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>Premium</td>\n",
       "      <td>I</td>\n",
       "      <td>VS2</td>\n",
       "      <td>62.4</td>\n",
       "      <td>58.0</td>\n",
       "      <td>334</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>Good</td>\n",
       "      <td>J</td>\n",
       "      <td>SI2</td>\n",
       "      <td>63.3</td>\n",
       "      <td>58.0</td>\n",
       "      <td>335</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",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>53935</th>\n",
       "      <td>Ideal</td>\n",
       "      <td>D</td>\n",
       "      <td>SI1</td>\n",
       "      <td>60.8</td>\n",
       "      <td>57.0</td>\n",
       "      <td>2757</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>53936</th>\n",
       "      <td>Good</td>\n",
       "      <td>D</td>\n",
       "      <td>SI1</td>\n",
       "      <td>63.1</td>\n",
       "      <td>55.0</td>\n",
       "      <td>2757</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>53937</th>\n",
       "      <td>Very Good</td>\n",
       "      <td>D</td>\n",
       "      <td>SI1</td>\n",
       "      <td>62.8</td>\n",
       "      <td>60.0</td>\n",
       "      <td>2757</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>53938</th>\n",
       "      <td>Premium</td>\n",
       "      <td>H</td>\n",
       "      <td>SI2</td>\n",
       "      <td>61.0</td>\n",
       "      <td>58.0</td>\n",
       "      <td>2757</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>53939</th>\n",
       "      <td>Ideal</td>\n",
       "      <td>D</td>\n",
       "      <td>SI2</td>\n",
       "      <td>62.2</td>\n",
       "      <td>55.0</td>\n",
       "      <td>2757</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>53940 rows × 6 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "             cut color clarity  depth  table  price\n",
       "0          Ideal     E     SI2   61.5   55.0    326\n",
       "1        Premium     E     SI1   59.8   61.0    326\n",
       "2           Good     E     VS1   56.9   65.0    327\n",
       "3        Premium     I     VS2   62.4   58.0    334\n",
       "4           Good     J     SI2   63.3   58.0    335\n",
       "...          ...   ...     ...    ...    ...    ...\n",
       "53935      Ideal     D     SI1   60.8   57.0   2757\n",
       "53936       Good     D     SI1   63.1   55.0   2757\n",
       "53937  Very Good     D     SI1   62.8   60.0   2757\n",
       "53938    Premium     H     SI2   61.0   58.0   2757\n",
       "53939      Ideal     D     SI2   62.2   55.0   2757\n",
       "\n",
       "[53940 rows x 6 columns]"
      ]
     },
     "execution_count": 43,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# 연달아 있는 열을 슬라이싱\n",
    "# \"cut\" ~ \"price\"\n",
    "# data.loc[: , \"start\":\"end\"]\n",
    "diamonds.loc[:, \"cut\":\"price\"]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 44,
   "id": "737426e5",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0            Ideal\n",
       "1          Premium\n",
       "2             Good\n",
       "3          Premium\n",
       "4             Good\n",
       "           ...    \n",
       "53935        Ideal\n",
       "53936         Good\n",
       "53937    Very Good\n",
       "53938      Premium\n",
       "53939        Ideal\n",
       "Name: cut, Length: 53940, dtype: category\n",
       "Categories (5, object): ['Ideal', 'Premium', 'Very Good', 'Good', 'Fair']"
      ]
     },
     "execution_count": 44,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# 참고 : data.iloc[: , index]\n",
    "diamonds.iloc[:, 1]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 45,
   "id": "1d7f070d",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>carat</th>\n",
       "      <th>cut</th>\n",
       "      <th>x</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>0.23</td>\n",
       "      <td>Ideal</td>\n",
       "      <td>3.95</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>0.21</td>\n",
       "      <td>Premium</td>\n",
       "      <td>3.89</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>0.23</td>\n",
       "      <td>Good</td>\n",
       "      <td>4.05</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>0.29</td>\n",
       "      <td>Premium</td>\n",
       "      <td>4.20</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>0.31</td>\n",
       "      <td>Good</td>\n",
       "      <td>4.34</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>53935</th>\n",
       "      <td>0.72</td>\n",
       "      <td>Ideal</td>\n",
       "      <td>5.75</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>53936</th>\n",
       "      <td>0.72</td>\n",
       "      <td>Good</td>\n",
       "      <td>5.69</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>53937</th>\n",
       "      <td>0.70</td>\n",
       "      <td>Very Good</td>\n",
       "      <td>5.66</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>53938</th>\n",
       "      <td>0.86</td>\n",
       "      <td>Premium</td>\n",
       "      <td>6.15</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>53939</th>\n",
       "      <td>0.75</td>\n",
       "      <td>Ideal</td>\n",
       "      <td>5.83</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>53940 rows × 3 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "       carat        cut     x\n",
       "0       0.23      Ideal  3.95\n",
       "1       0.21    Premium  3.89\n",
       "2       0.23       Good  4.05\n",
       "3       0.29    Premium  4.20\n",
       "4       0.31       Good  4.34\n",
       "...      ...        ...   ...\n",
       "53935   0.72      Ideal  5.75\n",
       "53936   0.72       Good  5.69\n",
       "53937   0.70  Very Good  5.66\n",
       "53938   0.86    Premium  6.15\n",
       "53939   0.75      Ideal  5.83\n",
       "\n",
       "[53940 rows x 3 columns]"
      ]
     },
     "execution_count": 45,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "diamonds.iloc[:, [0, 1, 7]]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 46,
   "id": "9d73b309",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>cut</th>\n",
       "      <th>color</th>\n",
       "      <th>clarity</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>Ideal</td>\n",
       "      <td>E</td>\n",
       "      <td>SI2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>Premium</td>\n",
       "      <td>E</td>\n",
       "      <td>SI1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>Good</td>\n",
       "      <td>E</td>\n",
       "      <td>VS1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>Premium</td>\n",
       "      <td>I</td>\n",
       "      <td>VS2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>Good</td>\n",
       "      <td>J</td>\n",
       "      <td>SI2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>53935</th>\n",
       "      <td>Ideal</td>\n",
       "      <td>D</td>\n",
       "      <td>SI1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>53936</th>\n",
       "      <td>Good</td>\n",
       "      <td>D</td>\n",
       "      <td>SI1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>53937</th>\n",
       "      <td>Very Good</td>\n",
       "      <td>D</td>\n",
       "      <td>SI1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>53938</th>\n",
       "      <td>Premium</td>\n",
       "      <td>H</td>\n",
       "      <td>SI2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>53939</th>\n",
       "      <td>Ideal</td>\n",
       "      <td>D</td>\n",
       "      <td>SI2</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>53940 rows × 3 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "             cut color clarity\n",
       "0          Ideal     E     SI2\n",
       "1        Premium     E     SI1\n",
       "2           Good     E     VS1\n",
       "3        Premium     I     VS2\n",
       "4           Good     J     SI2\n",
       "...          ...   ...     ...\n",
       "53935      Ideal     D     SI1\n",
       "53936       Good     D     SI1\n",
       "53937  Very Good     D     SI1\n",
       "53938    Premium     H     SI2\n",
       "53939      Ideal     D     SI2\n",
       "\n",
       "[53940 rows x 3 columns]"
      ]
     },
     "execution_count": 46,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "diamonds.iloc[:, 1:4]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 47,
   "id": "e2e993e2",
   "metadata": {},
   "outputs": [],
   "source": [
    "# (2) 행\n",
    "# data.loc[조건, :]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 48,
   "id": "20971079",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>carat</th>\n",
       "      <th>cut</th>\n",
       "      <th>color</th>\n",
       "      <th>clarity</th>\n",
       "      <th>depth</th>\n",
       "      <th>table</th>\n",
       "      <th>price</th>\n",
       "      <th>x</th>\n",
       "      <th>y</th>\n",
       "      <th>z</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>0.23</td>\n",
       "      <td>Ideal</td>\n",
       "      <td>E</td>\n",
       "      <td>SI2</td>\n",
       "      <td>61.5</td>\n",
       "      <td>55.0</td>\n",
       "      <td>326</td>\n",
       "      <td>3.95</td>\n",
       "      <td>3.98</td>\n",
       "      <td>2.43</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>0.21</td>\n",
       "      <td>Premium</td>\n",
       "      <td>E</td>\n",
       "      <td>SI1</td>\n",
       "      <td>59.8</td>\n",
       "      <td>61.0</td>\n",
       "      <td>326</td>\n",
       "      <td>3.89</td>\n",
       "      <td>3.84</td>\n",
       "      <td>2.31</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>0.23</td>\n",
       "      <td>Good</td>\n",
       "      <td>E</td>\n",
       "      <td>VS1</td>\n",
       "      <td>56.9</td>\n",
       "      <td>65.0</td>\n",
       "      <td>327</td>\n",
       "      <td>4.05</td>\n",
       "      <td>4.07</td>\n",
       "      <td>2.31</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>0.29</td>\n",
       "      <td>Premium</td>\n",
       "      <td>I</td>\n",
       "      <td>VS2</td>\n",
       "      <td>62.4</td>\n",
       "      <td>58.0</td>\n",
       "      <td>334</td>\n",
       "      <td>4.20</td>\n",
       "      <td>4.23</td>\n",
       "      <td>2.63</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>0.31</td>\n",
       "      <td>Good</td>\n",
       "      <td>J</td>\n",
       "      <td>SI2</td>\n",
       "      <td>63.3</td>\n",
       "      <td>58.0</td>\n",
       "      <td>335</td>\n",
       "      <td>4.34</td>\n",
       "      <td>4.35</td>\n",
       "      <td>2.75</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",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>53935</th>\n",
       "      <td>0.72</td>\n",
       "      <td>Ideal</td>\n",
       "      <td>D</td>\n",
       "      <td>SI1</td>\n",
       "      <td>60.8</td>\n",
       "      <td>57.0</td>\n",
       "      <td>2757</td>\n",
       "      <td>5.75</td>\n",
       "      <td>5.76</td>\n",
       "      <td>3.50</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>53936</th>\n",
       "      <td>0.72</td>\n",
       "      <td>Good</td>\n",
       "      <td>D</td>\n",
       "      <td>SI1</td>\n",
       "      <td>63.1</td>\n",
       "      <td>55.0</td>\n",
       "      <td>2757</td>\n",
       "      <td>5.69</td>\n",
       "      <td>5.75</td>\n",
       "      <td>3.61</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>53937</th>\n",
       "      <td>0.70</td>\n",
       "      <td>Very Good</td>\n",
       "      <td>D</td>\n",
       "      <td>SI1</td>\n",
       "      <td>62.8</td>\n",
       "      <td>60.0</td>\n",
       "      <td>2757</td>\n",
       "      <td>5.66</td>\n",
       "      <td>5.68</td>\n",
       "      <td>3.56</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>53938</th>\n",
       "      <td>0.86</td>\n",
       "      <td>Premium</td>\n",
       "      <td>H</td>\n",
       "      <td>SI2</td>\n",
       "      <td>61.0</td>\n",
       "      <td>58.0</td>\n",
       "      <td>2757</td>\n",
       "      <td>6.15</td>\n",
       "      <td>6.12</td>\n",
       "      <td>3.74</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>53939</th>\n",
       "      <td>0.75</td>\n",
       "      <td>Ideal</td>\n",
       "      <td>D</td>\n",
       "      <td>SI2</td>\n",
       "      <td>62.2</td>\n",
       "      <td>55.0</td>\n",
       "      <td>2757</td>\n",
       "      <td>5.83</td>\n",
       "      <td>5.87</td>\n",
       "      <td>3.64</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>52051 rows × 10 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "       carat        cut color clarity  depth  table  price     x     y     z\n",
       "0       0.23      Ideal     E     SI2   61.5   55.0    326  3.95  3.98  2.43\n",
       "1       0.21    Premium     E     SI1   59.8   61.0    326  3.89  3.84  2.31\n",
       "2       0.23       Good     E     VS1   56.9   65.0    327  4.05  4.07  2.31\n",
       "3       0.29    Premium     I     VS2   62.4   58.0    334  4.20  4.23  2.63\n",
       "4       0.31       Good     J     SI2   63.3   58.0    335  4.34  4.35  2.75\n",
       "...      ...        ...   ...     ...    ...    ...    ...   ...   ...   ...\n",
       "53935   0.72      Ideal     D     SI1   60.8   57.0   2757  5.75  5.76  3.50\n",
       "53936   0.72       Good     D     SI1   63.1   55.0   2757  5.69  5.75  3.61\n",
       "53937   0.70  Very Good     D     SI1   62.8   60.0   2757  5.66  5.68  3.56\n",
       "53938   0.86    Premium     H     SI2   61.0   58.0   2757  6.15  6.12  3.74\n",
       "53939   0.75      Ideal     D     SI2   62.2   55.0   2757  5.83  5.87  3.64\n",
       "\n",
       "[52051 rows x 10 columns]"
      ]
     },
     "execution_count": 48,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# carat이 2이하인 데이터\n",
    "diamonds.loc[diamonds[\"carat\"] <= 2, :]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 50,
   "id": "b1119e30",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>carat</th>\n",
       "      <th>cut</th>\n",
       "      <th>color</th>\n",
       "      <th>clarity</th>\n",
       "      <th>depth</th>\n",
       "      <th>table</th>\n",
       "      <th>price</th>\n",
       "      <th>x</th>\n",
       "      <th>y</th>\n",
       "      <th>z</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>0.24</td>\n",
       "      <td>Very Good</td>\n",
       "      <td>J</td>\n",
       "      <td>VVS2</td>\n",
       "      <td>62.8</td>\n",
       "      <td>57.0</td>\n",
       "      <td>336</td>\n",
       "      <td>3.94</td>\n",
       "      <td>3.96</td>\n",
       "      <td>2.48</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>0.24</td>\n",
       "      <td>Very Good</td>\n",
       "      <td>I</td>\n",
       "      <td>VVS1</td>\n",
       "      <td>62.3</td>\n",
       "      <td>57.0</td>\n",
       "      <td>336</td>\n",
       "      <td>3.95</td>\n",
       "      <td>3.98</td>\n",
       "      <td>2.47</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>0.26</td>\n",
       "      <td>Very Good</td>\n",
       "      <td>H</td>\n",
       "      <td>SI1</td>\n",
       "      <td>61.9</td>\n",
       "      <td>55.0</td>\n",
       "      <td>337</td>\n",
       "      <td>4.07</td>\n",
       "      <td>4.11</td>\n",
       "      <td>2.53</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9</th>\n",
       "      <td>0.23</td>\n",
       "      <td>Very Good</td>\n",
       "      <td>H</td>\n",
       "      <td>VS1</td>\n",
       "      <td>59.4</td>\n",
       "      <td>61.0</td>\n",
       "      <td>338</td>\n",
       "      <td>4.00</td>\n",
       "      <td>4.05</td>\n",
       "      <td>2.39</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>19</th>\n",
       "      <td>0.30</td>\n",
       "      <td>Very Good</td>\n",
       "      <td>J</td>\n",
       "      <td>SI1</td>\n",
       "      <td>62.7</td>\n",
       "      <td>59.0</td>\n",
       "      <td>351</td>\n",
       "      <td>4.21</td>\n",
       "      <td>4.27</td>\n",
       "      <td>2.66</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",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>53921</th>\n",
       "      <td>0.70</td>\n",
       "      <td>Very Good</td>\n",
       "      <td>E</td>\n",
       "      <td>VS2</td>\n",
       "      <td>62.8</td>\n",
       "      <td>60.0</td>\n",
       "      <td>2755</td>\n",
       "      <td>5.59</td>\n",
       "      <td>5.65</td>\n",
       "      <td>3.53</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>53922</th>\n",
       "      <td>0.70</td>\n",
       "      <td>Very Good</td>\n",
       "      <td>D</td>\n",
       "      <td>VS1</td>\n",
       "      <td>63.1</td>\n",
       "      <td>59.0</td>\n",
       "      <td>2755</td>\n",
       "      <td>5.67</td>\n",
       "      <td>5.58</td>\n",
       "      <td>3.55</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>53932</th>\n",
       "      <td>0.70</td>\n",
       "      <td>Very Good</td>\n",
       "      <td>E</td>\n",
       "      <td>VS2</td>\n",
       "      <td>60.5</td>\n",
       "      <td>59.0</td>\n",
       "      <td>2757</td>\n",
       "      <td>5.71</td>\n",
       "      <td>5.76</td>\n",
       "      <td>3.47</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>53933</th>\n",
       "      <td>0.70</td>\n",
       "      <td>Very Good</td>\n",
       "      <td>E</td>\n",
       "      <td>VS2</td>\n",
       "      <td>61.2</td>\n",
       "      <td>59.0</td>\n",
       "      <td>2757</td>\n",
       "      <td>5.69</td>\n",
       "      <td>5.72</td>\n",
       "      <td>3.49</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>53937</th>\n",
       "      <td>0.70</td>\n",
       "      <td>Very Good</td>\n",
       "      <td>D</td>\n",
       "      <td>SI1</td>\n",
       "      <td>62.8</td>\n",
       "      <td>60.0</td>\n",
       "      <td>2757</td>\n",
       "      <td>5.66</td>\n",
       "      <td>5.68</td>\n",
       "      <td>3.56</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>12082 rows × 10 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "       carat        cut color clarity  depth  table  price     x     y     z\n",
       "5       0.24  Very Good     J    VVS2   62.8   57.0    336  3.94  3.96  2.48\n",
       "6       0.24  Very Good     I    VVS1   62.3   57.0    336  3.95  3.98  2.47\n",
       "7       0.26  Very Good     H     SI1   61.9   55.0    337  4.07  4.11  2.53\n",
       "9       0.23  Very Good     H     VS1   59.4   61.0    338  4.00  4.05  2.39\n",
       "19      0.30  Very Good     J     SI1   62.7   59.0    351  4.21  4.27  2.66\n",
       "...      ...        ...   ...     ...    ...    ...    ...   ...   ...   ...\n",
       "53921   0.70  Very Good     E     VS2   62.8   60.0   2755  5.59  5.65  3.53\n",
       "53922   0.70  Very Good     D     VS1   63.1   59.0   2755  5.67  5.58  3.55\n",
       "53932   0.70  Very Good     E     VS2   60.5   59.0   2757  5.71  5.76  3.47\n",
       "53933   0.70  Very Good     E     VS2   61.2   59.0   2757  5.69  5.72  3.49\n",
       "53937   0.70  Very Good     D     SI1   62.8   60.0   2757  5.66  5.68  3.56\n",
       "\n",
       "[12082 rows x 10 columns]"
      ]
     },
     "execution_count": 50,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# cut이 \"Very Good\"인 데이터\n",
    "diamonds.loc[diamonds.cut == \"Very Good\", :]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 51,
   "id": "185cebe9",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>carat</th>\n",
       "      <th>cut</th>\n",
       "      <th>color</th>\n",
       "      <th>clarity</th>\n",
       "      <th>depth</th>\n",
       "      <th>table</th>\n",
       "      <th>price</th>\n",
       "      <th>x</th>\n",
       "      <th>y</th>\n",
       "      <th>z</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>0.24</td>\n",
       "      <td>Very Good</td>\n",
       "      <td>J</td>\n",
       "      <td>VVS2</td>\n",
       "      <td>62.8</td>\n",
       "      <td>57.0</td>\n",
       "      <td>336</td>\n",
       "      <td>3.94</td>\n",
       "      <td>3.96</td>\n",
       "      <td>2.48</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>0.24</td>\n",
       "      <td>Very Good</td>\n",
       "      <td>I</td>\n",
       "      <td>VVS1</td>\n",
       "      <td>62.3</td>\n",
       "      <td>57.0</td>\n",
       "      <td>336</td>\n",
       "      <td>3.95</td>\n",
       "      <td>3.98</td>\n",
       "      <td>2.47</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>0.26</td>\n",
       "      <td>Very Good</td>\n",
       "      <td>H</td>\n",
       "      <td>SI1</td>\n",
       "      <td>61.9</td>\n",
       "      <td>55.0</td>\n",
       "      <td>337</td>\n",
       "      <td>4.07</td>\n",
       "      <td>4.11</td>\n",
       "      <td>2.53</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9</th>\n",
       "      <td>0.23</td>\n",
       "      <td>Very Good</td>\n",
       "      <td>H</td>\n",
       "      <td>VS1</td>\n",
       "      <td>59.4</td>\n",
       "      <td>61.0</td>\n",
       "      <td>338</td>\n",
       "      <td>4.00</td>\n",
       "      <td>4.05</td>\n",
       "      <td>2.39</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>19</th>\n",
       "      <td>0.30</td>\n",
       "      <td>Very Good</td>\n",
       "      <td>J</td>\n",
       "      <td>SI1</td>\n",
       "      <td>62.7</td>\n",
       "      <td>59.0</td>\n",
       "      <td>351</td>\n",
       "      <td>4.21</td>\n",
       "      <td>4.27</td>\n",
       "      <td>2.66</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",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>53921</th>\n",
       "      <td>0.70</td>\n",
       "      <td>Very Good</td>\n",
       "      <td>E</td>\n",
       "      <td>VS2</td>\n",
       "      <td>62.8</td>\n",
       "      <td>60.0</td>\n",
       "      <td>2755</td>\n",
       "      <td>5.59</td>\n",
       "      <td>5.65</td>\n",
       "      <td>3.53</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>53922</th>\n",
       "      <td>0.70</td>\n",
       "      <td>Very Good</td>\n",
       "      <td>D</td>\n",
       "      <td>VS1</td>\n",
       "      <td>63.1</td>\n",
       "      <td>59.0</td>\n",
       "      <td>2755</td>\n",
       "      <td>5.67</td>\n",
       "      <td>5.58</td>\n",
       "      <td>3.55</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>53932</th>\n",
       "      <td>0.70</td>\n",
       "      <td>Very Good</td>\n",
       "      <td>E</td>\n",
       "      <td>VS2</td>\n",
       "      <td>60.5</td>\n",
       "      <td>59.0</td>\n",
       "      <td>2757</td>\n",
       "      <td>5.71</td>\n",
       "      <td>5.76</td>\n",
       "      <td>3.47</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>53933</th>\n",
       "      <td>0.70</td>\n",
       "      <td>Very Good</td>\n",
       "      <td>E</td>\n",
       "      <td>VS2</td>\n",
       "      <td>61.2</td>\n",
       "      <td>59.0</td>\n",
       "      <td>2757</td>\n",
       "      <td>5.69</td>\n",
       "      <td>5.72</td>\n",
       "      <td>3.49</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>53937</th>\n",
       "      <td>0.70</td>\n",
       "      <td>Very Good</td>\n",
       "      <td>D</td>\n",
       "      <td>SI1</td>\n",
       "      <td>62.8</td>\n",
       "      <td>60.0</td>\n",
       "      <td>2757</td>\n",
       "      <td>5.66</td>\n",
       "      <td>5.68</td>\n",
       "      <td>3.56</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>11694 rows × 10 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "       carat        cut color clarity  depth  table  price     x     y     z\n",
       "5       0.24  Very Good     J    VVS2   62.8   57.0    336  3.94  3.96  2.48\n",
       "6       0.24  Very Good     I    VVS1   62.3   57.0    336  3.95  3.98  2.47\n",
       "7       0.26  Very Good     H     SI1   61.9   55.0    337  4.07  4.11  2.53\n",
       "9       0.23  Very Good     H     VS1   59.4   61.0    338  4.00  4.05  2.39\n",
       "19      0.30  Very Good     J     SI1   62.7   59.0    351  4.21  4.27  2.66\n",
       "...      ...        ...   ...     ...    ...    ...    ...   ...   ...   ...\n",
       "53921   0.70  Very Good     E     VS2   62.8   60.0   2755  5.59  5.65  3.53\n",
       "53922   0.70  Very Good     D     VS1   63.1   59.0   2755  5.67  5.58  3.55\n",
       "53932   0.70  Very Good     E     VS2   60.5   59.0   2757  5.71  5.76  3.47\n",
       "53933   0.70  Very Good     E     VS2   61.2   59.0   2757  5.69  5.72  3.49\n",
       "53937   0.70  Very Good     D     SI1   62.8   60.0   2757  5.66  5.68  3.56\n",
       "\n",
       "[11694 rows x 10 columns]"
      ]
     },
     "execution_count": 51,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# carat이 2이하이고, cut은 \"Very Good\"인 데이터\n",
    "diamonds.loc[(diamonds.carat <= 2) & (diamonds.cut == \"Very Good\"), :]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 52,
   "id": "f64edb58",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>carat</th>\n",
       "      <th>cut</th>\n",
       "      <th>color</th>\n",
       "      <th>clarity</th>\n",
       "      <th>depth</th>\n",
       "      <th>table</th>\n",
       "      <th>price</th>\n",
       "      <th>x</th>\n",
       "      <th>y</th>\n",
       "      <th>z</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>0.23</td>\n",
       "      <td>Ideal</td>\n",
       "      <td>E</td>\n",
       "      <td>SI2</td>\n",
       "      <td>61.5</td>\n",
       "      <td>55.0</td>\n",
       "      <td>326</td>\n",
       "      <td>3.95</td>\n",
       "      <td>3.98</td>\n",
       "      <td>2.43</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>0.21</td>\n",
       "      <td>Premium</td>\n",
       "      <td>E</td>\n",
       "      <td>SI1</td>\n",
       "      <td>59.8</td>\n",
       "      <td>61.0</td>\n",
       "      <td>326</td>\n",
       "      <td>3.89</td>\n",
       "      <td>3.84</td>\n",
       "      <td>2.31</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>0.23</td>\n",
       "      <td>Good</td>\n",
       "      <td>E</td>\n",
       "      <td>VS1</td>\n",
       "      <td>56.9</td>\n",
       "      <td>65.0</td>\n",
       "      <td>327</td>\n",
       "      <td>4.05</td>\n",
       "      <td>4.07</td>\n",
       "      <td>2.31</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>0.29</td>\n",
       "      <td>Premium</td>\n",
       "      <td>I</td>\n",
       "      <td>VS2</td>\n",
       "      <td>62.4</td>\n",
       "      <td>58.0</td>\n",
       "      <td>334</td>\n",
       "      <td>4.20</td>\n",
       "      <td>4.23</td>\n",
       "      <td>2.63</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>0.31</td>\n",
       "      <td>Good</td>\n",
       "      <td>J</td>\n",
       "      <td>SI2</td>\n",
       "      <td>63.3</td>\n",
       "      <td>58.0</td>\n",
       "      <td>335</td>\n",
       "      <td>4.34</td>\n",
       "      <td>4.35</td>\n",
       "      <td>2.75</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",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>53935</th>\n",
       "      <td>0.72</td>\n",
       "      <td>Ideal</td>\n",
       "      <td>D</td>\n",
       "      <td>SI1</td>\n",
       "      <td>60.8</td>\n",
       "      <td>57.0</td>\n",
       "      <td>2757</td>\n",
       "      <td>5.75</td>\n",
       "      <td>5.76</td>\n",
       "      <td>3.50</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>53936</th>\n",
       "      <td>0.72</td>\n",
       "      <td>Good</td>\n",
       "      <td>D</td>\n",
       "      <td>SI1</td>\n",
       "      <td>63.1</td>\n",
       "      <td>55.0</td>\n",
       "      <td>2757</td>\n",
       "      <td>5.69</td>\n",
       "      <td>5.75</td>\n",
       "      <td>3.61</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>53937</th>\n",
       "      <td>0.70</td>\n",
       "      <td>Very Good</td>\n",
       "      <td>D</td>\n",
       "      <td>SI1</td>\n",
       "      <td>62.8</td>\n",
       "      <td>60.0</td>\n",
       "      <td>2757</td>\n",
       "      <td>5.66</td>\n",
       "      <td>5.68</td>\n",
       "      <td>3.56</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>53938</th>\n",
       "      <td>0.86</td>\n",
       "      <td>Premium</td>\n",
       "      <td>H</td>\n",
       "      <td>SI2</td>\n",
       "      <td>61.0</td>\n",
       "      <td>58.0</td>\n",
       "      <td>2757</td>\n",
       "      <td>6.15</td>\n",
       "      <td>6.12</td>\n",
       "      <td>3.74</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>53939</th>\n",
       "      <td>0.75</td>\n",
       "      <td>Ideal</td>\n",
       "      <td>D</td>\n",
       "      <td>SI2</td>\n",
       "      <td>62.2</td>\n",
       "      <td>55.0</td>\n",
       "      <td>2757</td>\n",
       "      <td>5.83</td>\n",
       "      <td>5.87</td>\n",
       "      <td>3.64</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>52439 rows × 10 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "       carat        cut color clarity  depth  table  price     x     y     z\n",
       "0       0.23      Ideal     E     SI2   61.5   55.0    326  3.95  3.98  2.43\n",
       "1       0.21    Premium     E     SI1   59.8   61.0    326  3.89  3.84  2.31\n",
       "2       0.23       Good     E     VS1   56.9   65.0    327  4.05  4.07  2.31\n",
       "3       0.29    Premium     I     VS2   62.4   58.0    334  4.20  4.23  2.63\n",
       "4       0.31       Good     J     SI2   63.3   58.0    335  4.34  4.35  2.75\n",
       "...      ...        ...   ...     ...    ...    ...    ...   ...   ...   ...\n",
       "53935   0.72      Ideal     D     SI1   60.8   57.0   2757  5.75  5.76  3.50\n",
       "53936   0.72       Good     D     SI1   63.1   55.0   2757  5.69  5.75  3.61\n",
       "53937   0.70  Very Good     D     SI1   62.8   60.0   2757  5.66  5.68  3.56\n",
       "53938   0.86    Premium     H     SI2   61.0   58.0   2757  6.15  6.12  3.74\n",
       "53939   0.75      Ideal     D     SI2   62.2   55.0   2757  5.83  5.87  3.64\n",
       "\n",
       "[52439 rows x 10 columns]"
      ]
     },
     "execution_count": 52,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# carat이 2이하이거나 또는 cut이 \"Very Good\"인 데이터\n",
    "diamonds.loc[(diamonds.carat <= 2) | (diamonds.cut == \"Very Good\"), :]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 55,
   "id": "5e3ae42e",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>x</th>\n",
       "      <th>y</th>\n",
       "      <th>z</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>3.95</td>\n",
       "      <td>3.98</td>\n",
       "      <td>2.43</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>3.89</td>\n",
       "      <td>3.84</td>\n",
       "      <td>2.31</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>4.05</td>\n",
       "      <td>4.07</td>\n",
       "      <td>2.31</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>4.20</td>\n",
       "      <td>4.23</td>\n",
       "      <td>2.63</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>4.34</td>\n",
       "      <td>4.35</td>\n",
       "      <td>2.75</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>53935</th>\n",
       "      <td>5.75</td>\n",
       "      <td>5.76</td>\n",
       "      <td>3.50</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>53936</th>\n",
       "      <td>5.69</td>\n",
       "      <td>5.75</td>\n",
       "      <td>3.61</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>53937</th>\n",
       "      <td>5.66</td>\n",
       "      <td>5.68</td>\n",
       "      <td>3.56</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>53938</th>\n",
       "      <td>6.15</td>\n",
       "      <td>6.12</td>\n",
       "      <td>3.74</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>53939</th>\n",
       "      <td>5.83</td>\n",
       "      <td>5.87</td>\n",
       "      <td>3.64</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>52439 rows × 3 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "          x     y     z\n",
       "0      3.95  3.98  2.43\n",
       "1      3.89  3.84  2.31\n",
       "2      4.05  4.07  2.31\n",
       "3      4.20  4.23  2.63\n",
       "4      4.34  4.35  2.75\n",
       "...     ...   ...   ...\n",
       "53935  5.75  5.76  3.50\n",
       "53936  5.69  5.75  3.61\n",
       "53937  5.66  5.68  3.56\n",
       "53938  6.15  6.12  3.74\n",
       "53939  5.83  5.87  3.64\n",
       "\n",
       "[52439 rows x 3 columns]"
      ]
     },
     "execution_count": 55,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# (3) 행과 열\n",
    "# carat이 2이하이거나 또는 cut이 \"Very Good\"인 데이터의 \"x\", \"y\", \"z\"\n",
    "diamonds.loc[(diamonds.carat <= 2) | (diamonds.cut == \"Very Good\"), \"x\":\"z\"]"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "983b72ce",
   "metadata": {},
   "source": [
    "#### 5. 새로운 열(변수) 만들기\n",
    "- 연산\n",
    "- 변환 : 로그, 루트, 역수 : Cox-Box Transformation\n",
    "- 범주 줄이기\n",
    "- 범주화"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 56,
   "id": "035f4543",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>carat</th>\n",
       "      <th>cut</th>\n",
       "      <th>color</th>\n",
       "      <th>clarity</th>\n",
       "      <th>depth</th>\n",
       "      <th>table</th>\n",
       "      <th>price</th>\n",
       "      <th>x</th>\n",
       "      <th>y</th>\n",
       "      <th>z</th>\n",
       "      <th>price_per_carat</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>0.23</td>\n",
       "      <td>Ideal</td>\n",
       "      <td>E</td>\n",
       "      <td>SI2</td>\n",
       "      <td>61.5</td>\n",
       "      <td>55.0</td>\n",
       "      <td>326</td>\n",
       "      <td>3.95</td>\n",
       "      <td>3.98</td>\n",
       "      <td>2.43</td>\n",
       "      <td>1417.391304</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>0.21</td>\n",
       "      <td>Premium</td>\n",
       "      <td>E</td>\n",
       "      <td>SI1</td>\n",
       "      <td>59.8</td>\n",
       "      <td>61.0</td>\n",
       "      <td>326</td>\n",
       "      <td>3.89</td>\n",
       "      <td>3.84</td>\n",
       "      <td>2.31</td>\n",
       "      <td>1552.380952</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>0.23</td>\n",
       "      <td>Good</td>\n",
       "      <td>E</td>\n",
       "      <td>VS1</td>\n",
       "      <td>56.9</td>\n",
       "      <td>65.0</td>\n",
       "      <td>327</td>\n",
       "      <td>4.05</td>\n",
       "      <td>4.07</td>\n",
       "      <td>2.31</td>\n",
       "      <td>1421.739130</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>0.29</td>\n",
       "      <td>Premium</td>\n",
       "      <td>I</td>\n",
       "      <td>VS2</td>\n",
       "      <td>62.4</td>\n",
       "      <td>58.0</td>\n",
       "      <td>334</td>\n",
       "      <td>4.20</td>\n",
       "      <td>4.23</td>\n",
       "      <td>2.63</td>\n",
       "      <td>1151.724138</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>0.31</td>\n",
       "      <td>Good</td>\n",
       "      <td>J</td>\n",
       "      <td>SI2</td>\n",
       "      <td>63.3</td>\n",
       "      <td>58.0</td>\n",
       "      <td>335</td>\n",
       "      <td>4.34</td>\n",
       "      <td>4.35</td>\n",
       "      <td>2.75</td>\n",
       "      <td>1080.645161</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   carat      cut color clarity  depth  table  price     x     y     z  \\\n",
       "0   0.23    Ideal     E     SI2   61.5   55.0    326  3.95  3.98  2.43   \n",
       "1   0.21  Premium     E     SI1   59.8   61.0    326  3.89  3.84  2.31   \n",
       "2   0.23     Good     E     VS1   56.9   65.0    327  4.05  4.07  2.31   \n",
       "3   0.29  Premium     I     VS2   62.4   58.0    334  4.20  4.23  2.63   \n",
       "4   0.31     Good     J     SI2   63.3   58.0    335  4.34  4.35  2.75   \n",
       "\n",
       "   price_per_carat  \n",
       "0      1417.391304  \n",
       "1      1552.380952  \n",
       "2      1421.739130  \n",
       "3      1151.724138  \n",
       "4      1080.645161  "
      ]
     },
     "execution_count": 56,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# (1) 연산 : price_per_carat : price/carat\n",
    "# data[\"새로운 열의 이름\"] = 연산\n",
    "diamonds[\"price_per_carat\"] = diamonds.price/diamonds.carat\n",
    "diamonds.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 66,
   "id": "9f2536dc",
   "metadata": {},
   "outputs": [],
   "source": [
    "# (2) Box-Cox 변환(Transformation) : 로그, 루트, 역수\n",
    "# 참고\n",
    "# 변환을 하는 이유는 주로 대칭을 확보하기 위해서"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 68,
   "id": "b7c7754a",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>carat</th>\n",
       "      <th>cut</th>\n",
       "      <th>color</th>\n",
       "      <th>clarity</th>\n",
       "      <th>depth</th>\n",
       "      <th>table</th>\n",
       "      <th>price</th>\n",
       "      <th>x</th>\n",
       "      <th>y</th>\n",
       "      <th>z</th>\n",
       "      <th>price_per_carat</th>\n",
       "      <th>price_log10</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>0.23</td>\n",
       "      <td>Ideal</td>\n",
       "      <td>E</td>\n",
       "      <td>SI2</td>\n",
       "      <td>61.5</td>\n",
       "      <td>55.0</td>\n",
       "      <td>326</td>\n",
       "      <td>3.95</td>\n",
       "      <td>3.98</td>\n",
       "      <td>2.43</td>\n",
       "      <td>1417.391304</td>\n",
       "      <td>2.513218</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>0.21</td>\n",
       "      <td>Premium</td>\n",
       "      <td>E</td>\n",
       "      <td>SI1</td>\n",
       "      <td>59.8</td>\n",
       "      <td>61.0</td>\n",
       "      <td>326</td>\n",
       "      <td>3.89</td>\n",
       "      <td>3.84</td>\n",
       "      <td>2.31</td>\n",
       "      <td>1552.380952</td>\n",
       "      <td>2.513218</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>0.23</td>\n",
       "      <td>Good</td>\n",
       "      <td>E</td>\n",
       "      <td>VS1</td>\n",
       "      <td>56.9</td>\n",
       "      <td>65.0</td>\n",
       "      <td>327</td>\n",
       "      <td>4.05</td>\n",
       "      <td>4.07</td>\n",
       "      <td>2.31</td>\n",
       "      <td>1421.739130</td>\n",
       "      <td>2.514548</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>0.29</td>\n",
       "      <td>Premium</td>\n",
       "      <td>I</td>\n",
       "      <td>VS2</td>\n",
       "      <td>62.4</td>\n",
       "      <td>58.0</td>\n",
       "      <td>334</td>\n",
       "      <td>4.20</td>\n",
       "      <td>4.23</td>\n",
       "      <td>2.63</td>\n",
       "      <td>1151.724138</td>\n",
       "      <td>2.523746</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>0.31</td>\n",
       "      <td>Good</td>\n",
       "      <td>J</td>\n",
       "      <td>SI2</td>\n",
       "      <td>63.3</td>\n",
       "      <td>58.0</td>\n",
       "      <td>335</td>\n",
       "      <td>4.34</td>\n",
       "      <td>4.35</td>\n",
       "      <td>2.75</td>\n",
       "      <td>1080.645161</td>\n",
       "      <td>2.525045</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   carat      cut color clarity  depth  table  price     x     y     z  \\\n",
       "0   0.23    Ideal     E     SI2   61.5   55.0    326  3.95  3.98  2.43   \n",
       "1   0.21  Premium     E     SI1   59.8   61.0    326  3.89  3.84  2.31   \n",
       "2   0.23     Good     E     VS1   56.9   65.0    327  4.05  4.07  2.31   \n",
       "3   0.29  Premium     I     VS2   62.4   58.0    334  4.20  4.23  2.63   \n",
       "4   0.31     Good     J     SI2   63.3   58.0    335  4.34  4.35  2.75   \n",
       "\n",
       "   price_per_carat  price_log10  \n",
       "0      1417.391304     2.513218  \n",
       "1      1552.380952     2.513218  \n",
       "2      1421.739130     2.514548  \n",
       "3      1151.724138     2.523746  \n",
       "4      1080.645161     2.525045  "
      ]
     },
     "execution_count": 68,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# i. 로그(Log) 변환\n",
    "# data[\"새로운 열의 이름\"] = numpy.log10(data[\"열의 이름\"])\n",
    "diamonds[\"price_log10\"] = np.log10(diamonds.price)\n",
    "diamonds.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 69,
   "id": "b1fb3070",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>carat</th>\n",
       "      <th>cut</th>\n",
       "      <th>color</th>\n",
       "      <th>clarity</th>\n",
       "      <th>depth</th>\n",
       "      <th>table</th>\n",
       "      <th>price</th>\n",
       "      <th>x</th>\n",
       "      <th>y</th>\n",
       "      <th>z</th>\n",
       "      <th>price_per_carat</th>\n",
       "      <th>price_log10</th>\n",
       "      <th>price_root</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>0.23</td>\n",
       "      <td>Ideal</td>\n",
       "      <td>E</td>\n",
       "      <td>SI2</td>\n",
       "      <td>61.5</td>\n",
       "      <td>55.0</td>\n",
       "      <td>326</td>\n",
       "      <td>3.95</td>\n",
       "      <td>3.98</td>\n",
       "      <td>2.43</td>\n",
       "      <td>1417.391304</td>\n",
       "      <td>2.513218</td>\n",
       "      <td>18.055470</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>0.21</td>\n",
       "      <td>Premium</td>\n",
       "      <td>E</td>\n",
       "      <td>SI1</td>\n",
       "      <td>59.8</td>\n",
       "      <td>61.0</td>\n",
       "      <td>326</td>\n",
       "      <td>3.89</td>\n",
       "      <td>3.84</td>\n",
       "      <td>2.31</td>\n",
       "      <td>1552.380952</td>\n",
       "      <td>2.513218</td>\n",
       "      <td>18.055470</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>0.23</td>\n",
       "      <td>Good</td>\n",
       "      <td>E</td>\n",
       "      <td>VS1</td>\n",
       "      <td>56.9</td>\n",
       "      <td>65.0</td>\n",
       "      <td>327</td>\n",
       "      <td>4.05</td>\n",
       "      <td>4.07</td>\n",
       "      <td>2.31</td>\n",
       "      <td>1421.739130</td>\n",
       "      <td>2.514548</td>\n",
       "      <td>18.083141</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>0.29</td>\n",
       "      <td>Premium</td>\n",
       "      <td>I</td>\n",
       "      <td>VS2</td>\n",
       "      <td>62.4</td>\n",
       "      <td>58.0</td>\n",
       "      <td>334</td>\n",
       "      <td>4.20</td>\n",
       "      <td>4.23</td>\n",
       "      <td>2.63</td>\n",
       "      <td>1151.724138</td>\n",
       "      <td>2.523746</td>\n",
       "      <td>18.275667</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>0.31</td>\n",
       "      <td>Good</td>\n",
       "      <td>J</td>\n",
       "      <td>SI2</td>\n",
       "      <td>63.3</td>\n",
       "      <td>58.0</td>\n",
       "      <td>335</td>\n",
       "      <td>4.34</td>\n",
       "      <td>4.35</td>\n",
       "      <td>2.75</td>\n",
       "      <td>1080.645161</td>\n",
       "      <td>2.525045</td>\n",
       "      <td>18.303005</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   carat      cut color clarity  depth  table  price     x     y     z  \\\n",
       "0   0.23    Ideal     E     SI2   61.5   55.0    326  3.95  3.98  2.43   \n",
       "1   0.21  Premium     E     SI1   59.8   61.0    326  3.89  3.84  2.31   \n",
       "2   0.23     Good     E     VS1   56.9   65.0    327  4.05  4.07  2.31   \n",
       "3   0.29  Premium     I     VS2   62.4   58.0    334  4.20  4.23  2.63   \n",
       "4   0.31     Good     J     SI2   63.3   58.0    335  4.34  4.35  2.75   \n",
       "\n",
       "   price_per_carat  price_log10  price_root  \n",
       "0      1417.391304     2.513218   18.055470  \n",
       "1      1552.380952     2.513218   18.055470  \n",
       "2      1421.739130     2.514548   18.083141  \n",
       "3      1151.724138     2.523746   18.275667  \n",
       "4      1080.645161     2.525045   18.303005  "
      ]
     },
     "execution_count": 69,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# ii. 루트(Root) 변환\n",
    "# data[\"새로운 열의 이름\"] = np.sqrt(data[\"열의이름\"])\n",
    "# 참고 : sqrt : square root\n",
    "\n",
    "diamonds[\"price_root\"] = np.sqrt(diamonds.price)\n",
    "diamonds.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 71,
   "id": "45d2d814",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>carat</th>\n",
       "      <th>cut</th>\n",
       "      <th>color</th>\n",
       "      <th>clarity</th>\n",
       "      <th>depth</th>\n",
       "      <th>table</th>\n",
       "      <th>price</th>\n",
       "      <th>x</th>\n",
       "      <th>y</th>\n",
       "      <th>z</th>\n",
       "      <th>price_per_carat</th>\n",
       "      <th>price_log10</th>\n",
       "      <th>price_root</th>\n",
       "      <th>price_inverse</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>0.23</td>\n",
       "      <td>Ideal</td>\n",
       "      <td>E</td>\n",
       "      <td>SI2</td>\n",
       "      <td>61.5</td>\n",
       "      <td>55.0</td>\n",
       "      <td>326</td>\n",
       "      <td>3.95</td>\n",
       "      <td>3.98</td>\n",
       "      <td>2.43</td>\n",
       "      <td>1417.391304</td>\n",
       "      <td>2.513218</td>\n",
       "      <td>18.055470</td>\n",
       "      <td>0.003067</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>0.21</td>\n",
       "      <td>Premium</td>\n",
       "      <td>E</td>\n",
       "      <td>SI1</td>\n",
       "      <td>59.8</td>\n",
       "      <td>61.0</td>\n",
       "      <td>326</td>\n",
       "      <td>3.89</td>\n",
       "      <td>3.84</td>\n",
       "      <td>2.31</td>\n",
       "      <td>1552.380952</td>\n",
       "      <td>2.513218</td>\n",
       "      <td>18.055470</td>\n",
       "      <td>0.003067</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>0.23</td>\n",
       "      <td>Good</td>\n",
       "      <td>E</td>\n",
       "      <td>VS1</td>\n",
       "      <td>56.9</td>\n",
       "      <td>65.0</td>\n",
       "      <td>327</td>\n",
       "      <td>4.05</td>\n",
       "      <td>4.07</td>\n",
       "      <td>2.31</td>\n",
       "      <td>1421.739130</td>\n",
       "      <td>2.514548</td>\n",
       "      <td>18.083141</td>\n",
       "      <td>0.003058</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>0.29</td>\n",
       "      <td>Premium</td>\n",
       "      <td>I</td>\n",
       "      <td>VS2</td>\n",
       "      <td>62.4</td>\n",
       "      <td>58.0</td>\n",
       "      <td>334</td>\n",
       "      <td>4.20</td>\n",
       "      <td>4.23</td>\n",
       "      <td>2.63</td>\n",
       "      <td>1151.724138</td>\n",
       "      <td>2.523746</td>\n",
       "      <td>18.275667</td>\n",
       "      <td>0.002994</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>0.31</td>\n",
       "      <td>Good</td>\n",
       "      <td>J</td>\n",
       "      <td>SI2</td>\n",
       "      <td>63.3</td>\n",
       "      <td>58.0</td>\n",
       "      <td>335</td>\n",
       "      <td>4.34</td>\n",
       "      <td>4.35</td>\n",
       "      <td>2.75</td>\n",
       "      <td>1080.645161</td>\n",
       "      <td>2.525045</td>\n",
       "      <td>18.303005</td>\n",
       "      <td>0.002985</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   carat      cut color clarity  depth  table  price     x     y     z  \\\n",
       "0   0.23    Ideal     E     SI2   61.5   55.0    326  3.95  3.98  2.43   \n",
       "1   0.21  Premium     E     SI1   59.8   61.0    326  3.89  3.84  2.31   \n",
       "2   0.23     Good     E     VS1   56.9   65.0    327  4.05  4.07  2.31   \n",
       "3   0.29  Premium     I     VS2   62.4   58.0    334  4.20  4.23  2.63   \n",
       "4   0.31     Good     J     SI2   63.3   58.0    335  4.34  4.35  2.75   \n",
       "\n",
       "   price_per_carat  price_log10  price_root  price_inverse  \n",
       "0      1417.391304     2.513218   18.055470       0.003067  \n",
       "1      1552.380952     2.513218   18.055470       0.003067  \n",
       "2      1421.739130     2.514548   18.083141       0.003058  \n",
       "3      1151.724138     2.523746   18.275667       0.002994  \n",
       "4      1080.645161     2.525045   18.303005       0.002985  "
      ]
     },
     "execution_count": 71,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# iii. 역수(Inverse) 변환\n",
    "# data[\"새로운 열의 이름\"] = 1/data[\"열의이름\"]\n",
    "diamonds[\"price_inverse\"] = 1 / diamonds.price\n",
    "diamonds.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 73,
   "id": "e64460c8",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>carat</th>\n",
       "      <th>cut</th>\n",
       "      <th>color</th>\n",
       "      <th>clarity</th>\n",
       "      <th>depth</th>\n",
       "      <th>table</th>\n",
       "      <th>price</th>\n",
       "      <th>x</th>\n",
       "      <th>y</th>\n",
       "      <th>z</th>\n",
       "      <th>price_per_carat</th>\n",
       "      <th>price_log10</th>\n",
       "      <th>price_root</th>\n",
       "      <th>price_inverse</th>\n",
       "      <th>cut2</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>0.23</td>\n",
       "      <td>Ideal</td>\n",
       "      <td>E</td>\n",
       "      <td>SI2</td>\n",
       "      <td>61.5</td>\n",
       "      <td>55.0</td>\n",
       "      <td>326</td>\n",
       "      <td>3.95</td>\n",
       "      <td>3.98</td>\n",
       "      <td>2.43</td>\n",
       "      <td>1417.391304</td>\n",
       "      <td>2.513218</td>\n",
       "      <td>18.055470</td>\n",
       "      <td>0.003067</td>\n",
       "      <td>Ideal</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>0.21</td>\n",
       "      <td>Premium</td>\n",
       "      <td>E</td>\n",
       "      <td>SI1</td>\n",
       "      <td>59.8</td>\n",
       "      <td>61.0</td>\n",
       "      <td>326</td>\n",
       "      <td>3.89</td>\n",
       "      <td>3.84</td>\n",
       "      <td>2.31</td>\n",
       "      <td>1552.380952</td>\n",
       "      <td>2.513218</td>\n",
       "      <td>18.055470</td>\n",
       "      <td>0.003067</td>\n",
       "      <td>NonIdeal</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>0.23</td>\n",
       "      <td>Good</td>\n",
       "      <td>E</td>\n",
       "      <td>VS1</td>\n",
       "      <td>56.9</td>\n",
       "      <td>65.0</td>\n",
       "      <td>327</td>\n",
       "      <td>4.05</td>\n",
       "      <td>4.07</td>\n",
       "      <td>2.31</td>\n",
       "      <td>1421.739130</td>\n",
       "      <td>2.514548</td>\n",
       "      <td>18.083141</td>\n",
       "      <td>0.003058</td>\n",
       "      <td>NonIdeal</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>0.29</td>\n",
       "      <td>Premium</td>\n",
       "      <td>I</td>\n",
       "      <td>VS2</td>\n",
       "      <td>62.4</td>\n",
       "      <td>58.0</td>\n",
       "      <td>334</td>\n",
       "      <td>4.20</td>\n",
       "      <td>4.23</td>\n",
       "      <td>2.63</td>\n",
       "      <td>1151.724138</td>\n",
       "      <td>2.523746</td>\n",
       "      <td>18.275667</td>\n",
       "      <td>0.002994</td>\n",
       "      <td>NonIdeal</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>0.31</td>\n",
       "      <td>Good</td>\n",
       "      <td>J</td>\n",
       "      <td>SI2</td>\n",
       "      <td>63.3</td>\n",
       "      <td>58.0</td>\n",
       "      <td>335</td>\n",
       "      <td>4.34</td>\n",
       "      <td>4.35</td>\n",
       "      <td>2.75</td>\n",
       "      <td>1080.645161</td>\n",
       "      <td>2.525045</td>\n",
       "      <td>18.303005</td>\n",
       "      <td>0.002985</td>\n",
       "      <td>NonIdeal</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   carat      cut color clarity  depth  table  price     x     y     z  \\\n",
       "0   0.23    Ideal     E     SI2   61.5   55.0    326  3.95  3.98  2.43   \n",
       "1   0.21  Premium     E     SI1   59.8   61.0    326  3.89  3.84  2.31   \n",
       "2   0.23     Good     E     VS1   56.9   65.0    327  4.05  4.07  2.31   \n",
       "3   0.29  Premium     I     VS2   62.4   58.0    334  4.20  4.23  2.63   \n",
       "4   0.31     Good     J     SI2   63.3   58.0    335  4.34  4.35  2.75   \n",
       "\n",
       "   price_per_carat  price_log10  price_root  price_inverse      cut2  \n",
       "0      1417.391304     2.513218   18.055470       0.003067     Ideal  \n",
       "1      1552.380952     2.513218   18.055470       0.003067  NonIdeal  \n",
       "2      1421.739130     2.514548   18.083141       0.003058  NonIdeal  \n",
       "3      1151.724138     2.523746   18.275667       0.002994  NonIdeal  \n",
       "4      1080.645161     2.525045   18.303005       0.002985  NonIdeal  "
      ]
     },
     "execution_count": 73,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# (3) 범주(Category) 줄이기\n",
    "# cut : 다이아몬드의 품질, 5개의 범주\n",
    "# \"Fair\", \"Good\", \"Very Good\", \"Premium\", \"Ideal\"\n",
    "\n",
    "# cut2 : \"Ideal\", \"NonIdeal\"\n",
    "\n",
    "# data[\"새로운 열의 이름\"] = data[\"열의이름\"].apply(lambda x : \"Ideal\" if x == \"Ideal\" else \"NonIdeal\")\n",
    "diamonds[\"cut2\"] = diamonds[\"cut\"].apply(lambda x : \"Ideal\" if x == \"Ideal\" else \"NonIdeal\")\n",
    "diamonds.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 74,
   "id": "c23058f9",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>carat</th>\n",
       "      <th>cut</th>\n",
       "      <th>color</th>\n",
       "      <th>clarity</th>\n",
       "      <th>depth</th>\n",
       "      <th>table</th>\n",
       "      <th>price</th>\n",
       "      <th>x</th>\n",
       "      <th>y</th>\n",
       "      <th>z</th>\n",
       "      <th>price_per_carat</th>\n",
       "      <th>price_log10</th>\n",
       "      <th>price_root</th>\n",
       "      <th>price_inverse</th>\n",
       "      <th>cut2</th>\n",
       "      <th>cut3</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>0.23</td>\n",
       "      <td>Ideal</td>\n",
       "      <td>E</td>\n",
       "      <td>SI2</td>\n",
       "      <td>61.5</td>\n",
       "      <td>55.0</td>\n",
       "      <td>326</td>\n",
       "      <td>3.95</td>\n",
       "      <td>3.98</td>\n",
       "      <td>2.43</td>\n",
       "      <td>1417.391304</td>\n",
       "      <td>2.513218</td>\n",
       "      <td>18.055470</td>\n",
       "      <td>0.003067</td>\n",
       "      <td>Ideal</td>\n",
       "      <td>Ideal</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>0.21</td>\n",
       "      <td>Premium</td>\n",
       "      <td>E</td>\n",
       "      <td>SI1</td>\n",
       "      <td>59.8</td>\n",
       "      <td>61.0</td>\n",
       "      <td>326</td>\n",
       "      <td>3.89</td>\n",
       "      <td>3.84</td>\n",
       "      <td>2.31</td>\n",
       "      <td>1552.380952</td>\n",
       "      <td>2.513218</td>\n",
       "      <td>18.055470</td>\n",
       "      <td>0.003067</td>\n",
       "      <td>NonIdeal</td>\n",
       "      <td>Premium</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>0.23</td>\n",
       "      <td>Good</td>\n",
       "      <td>E</td>\n",
       "      <td>VS1</td>\n",
       "      <td>56.9</td>\n",
       "      <td>65.0</td>\n",
       "      <td>327</td>\n",
       "      <td>4.05</td>\n",
       "      <td>4.07</td>\n",
       "      <td>2.31</td>\n",
       "      <td>1421.739130</td>\n",
       "      <td>2.514548</td>\n",
       "      <td>18.083141</td>\n",
       "      <td>0.003058</td>\n",
       "      <td>NonIdeal</td>\n",
       "      <td>ETC</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>0.29</td>\n",
       "      <td>Premium</td>\n",
       "      <td>I</td>\n",
       "      <td>VS2</td>\n",
       "      <td>62.4</td>\n",
       "      <td>58.0</td>\n",
       "      <td>334</td>\n",
       "      <td>4.20</td>\n",
       "      <td>4.23</td>\n",
       "      <td>2.63</td>\n",
       "      <td>1151.724138</td>\n",
       "      <td>2.523746</td>\n",
       "      <td>18.275667</td>\n",
       "      <td>0.002994</td>\n",
       "      <td>NonIdeal</td>\n",
       "      <td>Premium</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>0.31</td>\n",
       "      <td>Good</td>\n",
       "      <td>J</td>\n",
       "      <td>SI2</td>\n",
       "      <td>63.3</td>\n",
       "      <td>58.0</td>\n",
       "      <td>335</td>\n",
       "      <td>4.34</td>\n",
       "      <td>4.35</td>\n",
       "      <td>2.75</td>\n",
       "      <td>1080.645161</td>\n",
       "      <td>2.525045</td>\n",
       "      <td>18.303005</td>\n",
       "      <td>0.002985</td>\n",
       "      <td>NonIdeal</td>\n",
       "      <td>ETC</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   carat      cut color clarity  depth  table  price     x     y     z  \\\n",
       "0   0.23    Ideal     E     SI2   61.5   55.0    326  3.95  3.98  2.43   \n",
       "1   0.21  Premium     E     SI1   59.8   61.0    326  3.89  3.84  2.31   \n",
       "2   0.23     Good     E     VS1   56.9   65.0    327  4.05  4.07  2.31   \n",
       "3   0.29  Premium     I     VS2   62.4   58.0    334  4.20  4.23  2.63   \n",
       "4   0.31     Good     J     SI2   63.3   58.0    335  4.34  4.35  2.75   \n",
       "\n",
       "   price_per_carat  price_log10  price_root  price_inverse      cut2     cut3  \n",
       "0      1417.391304     2.513218   18.055470       0.003067     Ideal    Ideal  \n",
       "1      1552.380952     2.513218   18.055470       0.003067  NonIdeal  Premium  \n",
       "2      1421.739130     2.514548   18.083141       0.003058  NonIdeal      ETC  \n",
       "3      1151.724138     2.523746   18.275667       0.002994  NonIdeal  Premium  \n",
       "4      1080.645161     2.525045   18.303005       0.002985  NonIdeal      ETC  "
      ]
     },
     "execution_count": 74,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# cut3 : \"Ideal\", \"Premium\", \"ETC\"\n",
    "diamonds[\"cut3\"] = diamonds[\"cut\"].apply(lambda x : \"Ideal\" if x == \"Ideal\" else \"Premium\" if x == \"Premium\" else \"ETC\")\n",
    "diamonds.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 78,
   "id": "2585c8fd",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>carat</th>\n",
       "      <th>cut</th>\n",
       "      <th>color</th>\n",
       "      <th>clarity</th>\n",
       "      <th>depth</th>\n",
       "      <th>table</th>\n",
       "      <th>price</th>\n",
       "      <th>x</th>\n",
       "      <th>y</th>\n",
       "      <th>z</th>\n",
       "      <th>price_per_carat</th>\n",
       "      <th>price_log10</th>\n",
       "      <th>price_root</th>\n",
       "      <th>price_inverse</th>\n",
       "      <th>cut2</th>\n",
       "      <th>cut3</th>\n",
       "      <th>price_group</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>0.23</td>\n",
       "      <td>Ideal</td>\n",
       "      <td>E</td>\n",
       "      <td>SI2</td>\n",
       "      <td>61.5</td>\n",
       "      <td>55.0</td>\n",
       "      <td>326</td>\n",
       "      <td>3.95</td>\n",
       "      <td>3.98</td>\n",
       "      <td>2.43</td>\n",
       "      <td>1417.391304</td>\n",
       "      <td>2.513218</td>\n",
       "      <td>18.055470</td>\n",
       "      <td>0.003067</td>\n",
       "      <td>Ideal</td>\n",
       "      <td>Ideal</td>\n",
       "      <td>Very Cheap</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>0.21</td>\n",
       "      <td>Premium</td>\n",
       "      <td>E</td>\n",
       "      <td>SI1</td>\n",
       "      <td>59.8</td>\n",
       "      <td>61.0</td>\n",
       "      <td>326</td>\n",
       "      <td>3.89</td>\n",
       "      <td>3.84</td>\n",
       "      <td>2.31</td>\n",
       "      <td>1552.380952</td>\n",
       "      <td>2.513218</td>\n",
       "      <td>18.055470</td>\n",
       "      <td>0.003067</td>\n",
       "      <td>NonIdeal</td>\n",
       "      <td>Premium</td>\n",
       "      <td>Very Cheap</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>0.23</td>\n",
       "      <td>Good</td>\n",
       "      <td>E</td>\n",
       "      <td>VS1</td>\n",
       "      <td>56.9</td>\n",
       "      <td>65.0</td>\n",
       "      <td>327</td>\n",
       "      <td>4.05</td>\n",
       "      <td>4.07</td>\n",
       "      <td>2.31</td>\n",
       "      <td>1421.739130</td>\n",
       "      <td>2.514548</td>\n",
       "      <td>18.083141</td>\n",
       "      <td>0.003058</td>\n",
       "      <td>NonIdeal</td>\n",
       "      <td>ETC</td>\n",
       "      <td>Very Cheap</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>0.29</td>\n",
       "      <td>Premium</td>\n",
       "      <td>I</td>\n",
       "      <td>VS2</td>\n",
       "      <td>62.4</td>\n",
       "      <td>58.0</td>\n",
       "      <td>334</td>\n",
       "      <td>4.20</td>\n",
       "      <td>4.23</td>\n",
       "      <td>2.63</td>\n",
       "      <td>1151.724138</td>\n",
       "      <td>2.523746</td>\n",
       "      <td>18.275667</td>\n",
       "      <td>0.002994</td>\n",
       "      <td>NonIdeal</td>\n",
       "      <td>Premium</td>\n",
       "      <td>Very Cheap</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>0.31</td>\n",
       "      <td>Good</td>\n",
       "      <td>J</td>\n",
       "      <td>SI2</td>\n",
       "      <td>63.3</td>\n",
       "      <td>58.0</td>\n",
       "      <td>335</td>\n",
       "      <td>4.34</td>\n",
       "      <td>4.35</td>\n",
       "      <td>2.75</td>\n",
       "      <td>1080.645161</td>\n",
       "      <td>2.525045</td>\n",
       "      <td>18.303005</td>\n",
       "      <td>0.002985</td>\n",
       "      <td>NonIdeal</td>\n",
       "      <td>ETC</td>\n",
       "      <td>Very Cheap</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   carat      cut color clarity  depth  table  price     x     y     z  \\\n",
       "0   0.23    Ideal     E     SI2   61.5   55.0    326  3.95  3.98  2.43   \n",
       "1   0.21  Premium     E     SI1   59.8   61.0    326  3.89  3.84  2.31   \n",
       "2   0.23     Good     E     VS1   56.9   65.0    327  4.05  4.07  2.31   \n",
       "3   0.29  Premium     I     VS2   62.4   58.0    334  4.20  4.23  2.63   \n",
       "4   0.31     Good     J     SI2   63.3   58.0    335  4.34  4.35  2.75   \n",
       "\n",
       "   price_per_carat  price_log10  price_root  price_inverse      cut2     cut3  \\\n",
       "0      1417.391304     2.513218   18.055470       0.003067     Ideal    Ideal   \n",
       "1      1552.380952     2.513218   18.055470       0.003067  NonIdeal  Premium   \n",
       "2      1421.739130     2.514548   18.083141       0.003058  NonIdeal      ETC   \n",
       "3      1151.724138     2.523746   18.275667       0.002994  NonIdeal  Premium   \n",
       "4      1080.645161     2.525045   18.303005       0.002985  NonIdeal      ETC   \n",
       "\n",
       "  price_group  \n",
       "0  Very Cheap  \n",
       "1  Very Cheap  \n",
       "2  Very Cheap  \n",
       "3  Very Cheap  \n",
       "4  Very Cheap  "
      ]
     },
     "execution_count": 78,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# (4) 범주화 : 수치형(숫자) --> 범주형\n",
    "# price(수치형) --> price_group(범주형)\n",
    "# 0 이상     ~ 5000 미만  --> \"Very Cheap\"\n",
    "# 5000 이상 ~ 10000 미만  --> \"Cheap\"\n",
    "# 10000 이상 ~ 15000 미만 --> \"Expensive\"\n",
    "# 15000 이상 ~ 20000 미만 --> \"Very Expensive\"\n",
    "\n",
    "# data[\"새로운 열의 이름\"] = pandas.cut(data[\"열의이름\"],\n",
    "#                                       bins   = 구간의 정보,\n",
    "#                                       right  = True or False,\n",
    "#                                       labels = 범주의 이름)\n",
    "\n",
    "diamonds[\"price_group\"] = pd.cut(diamonds.price,\n",
    "                                 bins   = [0, 5000, 10000, 15000, 20000],\n",
    "                                 right  = False,\n",
    "                                 labels = [\"Very Cheap\", \"Cheap\", \"Expensive\", \"Very Expensive\"])\n",
    "\n",
    "diamonds.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 77,
   "id": "eb9444a8",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([    0,  5000, 10000, 15000, 20000])"
      ]
     },
     "execution_count": 77,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# 규칙이 있는 2개 이상의 숫자를 만들 때\n",
    "# numpy.arange(start = 숫자, stop = 숫자, step = 증가의 폭)\n",
    "np.arange(start = 0, stop = 25000, step = 5000)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "2fb04c49",
   "metadata": {},
   "source": [
    "#### 6. 열 삭제하기\n",
    "- 하나의 열\n",
    "- 2개 이상의 열"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 79,
   "id": "bdcfc650",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Index(['carat', 'cut', 'color', 'clarity', 'depth', 'table', 'price', 'x', 'y',\n",
       "       'z', 'price_per_carat', 'price_log10', 'price_root', 'price_inverse',\n",
       "       'cut2', 'cut3', 'price_group'],\n",
       "      dtype='object')"
      ]
     },
     "execution_count": 79,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "diamonds.columns"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 80,
   "id": "9f4905c9",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>carat</th>\n",
       "      <th>cut</th>\n",
       "      <th>color</th>\n",
       "      <th>clarity</th>\n",
       "      <th>depth</th>\n",
       "      <th>table</th>\n",
       "      <th>price</th>\n",
       "      <th>x</th>\n",
       "      <th>y</th>\n",
       "      <th>z</th>\n",
       "      <th>price_log10</th>\n",
       "      <th>price_root</th>\n",
       "      <th>price_inverse</th>\n",
       "      <th>cut2</th>\n",
       "      <th>cut3</th>\n",
       "      <th>price_group</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>0.23</td>\n",
       "      <td>Ideal</td>\n",
       "      <td>E</td>\n",
       "      <td>SI2</td>\n",
       "      <td>61.5</td>\n",
       "      <td>55.0</td>\n",
       "      <td>326</td>\n",
       "      <td>3.95</td>\n",
       "      <td>3.98</td>\n",
       "      <td>2.43</td>\n",
       "      <td>2.513218</td>\n",
       "      <td>18.055470</td>\n",
       "      <td>0.003067</td>\n",
       "      <td>Ideal</td>\n",
       "      <td>Ideal</td>\n",
       "      <td>Very Cheap</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>0.21</td>\n",
       "      <td>Premium</td>\n",
       "      <td>E</td>\n",
       "      <td>SI1</td>\n",
       "      <td>59.8</td>\n",
       "      <td>61.0</td>\n",
       "      <td>326</td>\n",
       "      <td>3.89</td>\n",
       "      <td>3.84</td>\n",
       "      <td>2.31</td>\n",
       "      <td>2.513218</td>\n",
       "      <td>18.055470</td>\n",
       "      <td>0.003067</td>\n",
       "      <td>NonIdeal</td>\n",
       "      <td>Premium</td>\n",
       "      <td>Very Cheap</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>0.23</td>\n",
       "      <td>Good</td>\n",
       "      <td>E</td>\n",
       "      <td>VS1</td>\n",
       "      <td>56.9</td>\n",
       "      <td>65.0</td>\n",
       "      <td>327</td>\n",
       "      <td>4.05</td>\n",
       "      <td>4.07</td>\n",
       "      <td>2.31</td>\n",
       "      <td>2.514548</td>\n",
       "      <td>18.083141</td>\n",
       "      <td>0.003058</td>\n",
       "      <td>NonIdeal</td>\n",
       "      <td>ETC</td>\n",
       "      <td>Very Cheap</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>0.29</td>\n",
       "      <td>Premium</td>\n",
       "      <td>I</td>\n",
       "      <td>VS2</td>\n",
       "      <td>62.4</td>\n",
       "      <td>58.0</td>\n",
       "      <td>334</td>\n",
       "      <td>4.20</td>\n",
       "      <td>4.23</td>\n",
       "      <td>2.63</td>\n",
       "      <td>2.523746</td>\n",
       "      <td>18.275667</td>\n",
       "      <td>0.002994</td>\n",
       "      <td>NonIdeal</td>\n",
       "      <td>Premium</td>\n",
       "      <td>Very Cheap</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>0.31</td>\n",
       "      <td>Good</td>\n",
       "      <td>J</td>\n",
       "      <td>SI2</td>\n",
       "      <td>63.3</td>\n",
       "      <td>58.0</td>\n",
       "      <td>335</td>\n",
       "      <td>4.34</td>\n",
       "      <td>4.35</td>\n",
       "      <td>2.75</td>\n",
       "      <td>2.525045</td>\n",
       "      <td>18.303005</td>\n",
       "      <td>0.002985</td>\n",
       "      <td>NonIdeal</td>\n",
       "      <td>ETC</td>\n",
       "      <td>Very Cheap</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   carat      cut color clarity  depth  table  price     x     y     z  \\\n",
       "0   0.23    Ideal     E     SI2   61.5   55.0    326  3.95  3.98  2.43   \n",
       "1   0.21  Premium     E     SI1   59.8   61.0    326  3.89  3.84  2.31   \n",
       "2   0.23     Good     E     VS1   56.9   65.0    327  4.05  4.07  2.31   \n",
       "3   0.29  Premium     I     VS2   62.4   58.0    334  4.20  4.23  2.63   \n",
       "4   0.31     Good     J     SI2   63.3   58.0    335  4.34  4.35  2.75   \n",
       "\n",
       "   price_log10  price_root  price_inverse      cut2     cut3 price_group  \n",
       "0     2.513218   18.055470       0.003067     Ideal    Ideal  Very Cheap  \n",
       "1     2.513218   18.055470       0.003067  NonIdeal  Premium  Very Cheap  \n",
       "2     2.514548   18.083141       0.003058  NonIdeal      ETC  Very Cheap  \n",
       "3     2.523746   18.275667       0.002994  NonIdeal  Premium  Very Cheap  \n",
       "4     2.525045   18.303005       0.002985  NonIdeal      ETC  Very Cheap  "
      ]
     },
     "execution_count": 80,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# (1) 하나의 열 삭제하기 : del data[\"열의 이름\"]\n",
    "del diamonds[\"price_per_carat\"]\n",
    "diamonds.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 81,
   "id": "81526269",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>carat</th>\n",
       "      <th>cut</th>\n",
       "      <th>color</th>\n",
       "      <th>clarity</th>\n",
       "      <th>depth</th>\n",
       "      <th>table</th>\n",
       "      <th>price</th>\n",
       "      <th>x</th>\n",
       "      <th>y</th>\n",
       "      <th>z</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>0.23</td>\n",
       "      <td>Ideal</td>\n",
       "      <td>E</td>\n",
       "      <td>SI2</td>\n",
       "      <td>61.5</td>\n",
       "      <td>55.0</td>\n",
       "      <td>326</td>\n",
       "      <td>3.95</td>\n",
       "      <td>3.98</td>\n",
       "      <td>2.43</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>0.21</td>\n",
       "      <td>Premium</td>\n",
       "      <td>E</td>\n",
       "      <td>SI1</td>\n",
       "      <td>59.8</td>\n",
       "      <td>61.0</td>\n",
       "      <td>326</td>\n",
       "      <td>3.89</td>\n",
       "      <td>3.84</td>\n",
       "      <td>2.31</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>0.23</td>\n",
       "      <td>Good</td>\n",
       "      <td>E</td>\n",
       "      <td>VS1</td>\n",
       "      <td>56.9</td>\n",
       "      <td>65.0</td>\n",
       "      <td>327</td>\n",
       "      <td>4.05</td>\n",
       "      <td>4.07</td>\n",
       "      <td>2.31</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>0.29</td>\n",
       "      <td>Premium</td>\n",
       "      <td>I</td>\n",
       "      <td>VS2</td>\n",
       "      <td>62.4</td>\n",
       "      <td>58.0</td>\n",
       "      <td>334</td>\n",
       "      <td>4.20</td>\n",
       "      <td>4.23</td>\n",
       "      <td>2.63</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>0.31</td>\n",
       "      <td>Good</td>\n",
       "      <td>J</td>\n",
       "      <td>SI2</td>\n",
       "      <td>63.3</td>\n",
       "      <td>58.0</td>\n",
       "      <td>335</td>\n",
       "      <td>4.34</td>\n",
       "      <td>4.35</td>\n",
       "      <td>2.75</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   carat      cut color clarity  depth  table  price     x     y     z\n",
       "0   0.23    Ideal     E     SI2   61.5   55.0    326  3.95  3.98  2.43\n",
       "1   0.21  Premium     E     SI1   59.8   61.0    326  3.89  3.84  2.31\n",
       "2   0.23     Good     E     VS1   56.9   65.0    327  4.05  4.07  2.31\n",
       "3   0.29  Premium     I     VS2   62.4   58.0    334  4.20  4.23  2.63\n",
       "4   0.31     Good     J     SI2   63.3   58.0    335  4.34  4.35  2.75"
      ]
     },
     "execution_count": 81,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# (2) 2개 이상의 열 삭제하기\n",
    "# data.drop(columns = [\"열의 이름1\", ...], inplace = True or False)\n",
    "# inplace = True : 해당 열을 삭제했을 때에 data에도 그 결과가 반영되도록 함\n",
    "\n",
    "diamonds.drop(columns = [\"price_log10\", \"price_root\", \"price_inverse\", \"cut2\", \"cut3\", \"price_group\"],\n",
    "              inplace = True)\n",
    "diamonds.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "432d4dc9",
   "metadata": {},
   "source": [
    "#### 7. 정렬하기\n",
    "- 리스트 정렬하기\n",
    "- 데이터프레임 정렬하기"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 82,
   "id": "061b1a5b",
   "metadata": {},
   "outputs": [],
   "source": [
    "# (1) list 정렬하기\n",
    "# list.sort(reverse = True or False)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 83,
   "id": "d1251b5f",
   "metadata": {},
   "outputs": [],
   "source": [
    "walk = [30, 40, 20, 30, 30, 60]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 87,
   "id": "72f38e29",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[20, 30, 30, 30, 40, 60]"
      ]
     },
     "execution_count": 87,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# 오름차순\n",
    "walk.sort(reverse = False)\n",
    "walk"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 88,
   "id": "7cddd7cb",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[60, 40, 30, 30, 30, 20]"
      ]
     },
     "execution_count": 88,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# 내림차순\n",
    "walk.sort(reverse = True)\n",
    "walk"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 89,
   "id": "ed5de263",
   "metadata": {},
   "outputs": [],
   "source": [
    "# (2) 데이터 프레임\n",
    "# data.sort_values(by = 정렬한 열의 이름, ascending = True or False)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 90,
   "id": "a32dbd89",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>carat</th>\n",
       "      <th>cut</th>\n",
       "      <th>color</th>\n",
       "      <th>clarity</th>\n",
       "      <th>depth</th>\n",
       "      <th>table</th>\n",
       "      <th>price</th>\n",
       "      <th>x</th>\n",
       "      <th>y</th>\n",
       "      <th>z</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>31593</th>\n",
       "      <td>0.20</td>\n",
       "      <td>Premium</td>\n",
       "      <td>E</td>\n",
       "      <td>VS2</td>\n",
       "      <td>61.1</td>\n",
       "      <td>59.0</td>\n",
       "      <td>367</td>\n",
       "      <td>3.81</td>\n",
       "      <td>3.78</td>\n",
       "      <td>2.32</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>31597</th>\n",
       "      <td>0.20</td>\n",
       "      <td>Ideal</td>\n",
       "      <td>D</td>\n",
       "      <td>VS2</td>\n",
       "      <td>61.5</td>\n",
       "      <td>57.0</td>\n",
       "      <td>367</td>\n",
       "      <td>3.81</td>\n",
       "      <td>3.77</td>\n",
       "      <td>2.33</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>31596</th>\n",
       "      <td>0.20</td>\n",
       "      <td>Premium</td>\n",
       "      <td>F</td>\n",
       "      <td>VS2</td>\n",
       "      <td>62.6</td>\n",
       "      <td>59.0</td>\n",
       "      <td>367</td>\n",
       "      <td>3.73</td>\n",
       "      <td>3.71</td>\n",
       "      <td>2.33</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>31595</th>\n",
       "      <td>0.20</td>\n",
       "      <td>Ideal</td>\n",
       "      <td>E</td>\n",
       "      <td>VS2</td>\n",
       "      <td>59.7</td>\n",
       "      <td>55.0</td>\n",
       "      <td>367</td>\n",
       "      <td>3.86</td>\n",
       "      <td>3.84</td>\n",
       "      <td>2.30</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>31594</th>\n",
       "      <td>0.20</td>\n",
       "      <td>Premium</td>\n",
       "      <td>E</td>\n",
       "      <td>VS2</td>\n",
       "      <td>59.7</td>\n",
       "      <td>62.0</td>\n",
       "      <td>367</td>\n",
       "      <td>3.84</td>\n",
       "      <td>3.80</td>\n",
       "      <td>2.28</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",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>25999</th>\n",
       "      <td>4.01</td>\n",
       "      <td>Premium</td>\n",
       "      <td>J</td>\n",
       "      <td>I1</td>\n",
       "      <td>62.5</td>\n",
       "      <td>62.0</td>\n",
       "      <td>15223</td>\n",
       "      <td>10.02</td>\n",
       "      <td>9.94</td>\n",
       "      <td>6.24</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>25998</th>\n",
       "      <td>4.01</td>\n",
       "      <td>Premium</td>\n",
       "      <td>I</td>\n",
       "      <td>I1</td>\n",
       "      <td>61.0</td>\n",
       "      <td>61.0</td>\n",
       "      <td>15223</td>\n",
       "      <td>10.14</td>\n",
       "      <td>10.10</td>\n",
       "      <td>6.17</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>27130</th>\n",
       "      <td>4.13</td>\n",
       "      <td>Fair</td>\n",
       "      <td>H</td>\n",
       "      <td>I1</td>\n",
       "      <td>64.8</td>\n",
       "      <td>61.0</td>\n",
       "      <td>17329</td>\n",
       "      <td>10.00</td>\n",
       "      <td>9.85</td>\n",
       "      <td>6.43</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>27630</th>\n",
       "      <td>4.50</td>\n",
       "      <td>Fair</td>\n",
       "      <td>J</td>\n",
       "      <td>I1</td>\n",
       "      <td>65.8</td>\n",
       "      <td>58.0</td>\n",
       "      <td>18531</td>\n",
       "      <td>10.23</td>\n",
       "      <td>10.16</td>\n",
       "      <td>6.72</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>27415</th>\n",
       "      <td>5.01</td>\n",
       "      <td>Fair</td>\n",
       "      <td>J</td>\n",
       "      <td>I1</td>\n",
       "      <td>65.5</td>\n",
       "      <td>59.0</td>\n",
       "      <td>18018</td>\n",
       "      <td>10.74</td>\n",
       "      <td>10.54</td>\n",
       "      <td>6.98</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>53940 rows × 10 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "       carat      cut color clarity  depth  table  price      x      y     z\n",
       "31593   0.20  Premium     E     VS2   61.1   59.0    367   3.81   3.78  2.32\n",
       "31597   0.20    Ideal     D     VS2   61.5   57.0    367   3.81   3.77  2.33\n",
       "31596   0.20  Premium     F     VS2   62.6   59.0    367   3.73   3.71  2.33\n",
       "31595   0.20    Ideal     E     VS2   59.7   55.0    367   3.86   3.84  2.30\n",
       "31594   0.20  Premium     E     VS2   59.7   62.0    367   3.84   3.80  2.28\n",
       "...      ...      ...   ...     ...    ...    ...    ...    ...    ...   ...\n",
       "25999   4.01  Premium     J      I1   62.5   62.0  15223  10.02   9.94  6.24\n",
       "25998   4.01  Premium     I      I1   61.0   61.0  15223  10.14  10.10  6.17\n",
       "27130   4.13     Fair     H      I1   64.8   61.0  17329  10.00   9.85  6.43\n",
       "27630   4.50     Fair     J      I1   65.8   58.0  18531  10.23  10.16  6.72\n",
       "27415   5.01     Fair     J      I1   65.5   59.0  18018  10.74  10.54  6.98\n",
       "\n",
       "[53940 rows x 10 columns]"
      ]
     },
     "execution_count": 90,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# i. carat : 오름차순\n",
    "diamonds.sort_values(by = \"carat\", ascending = True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 91,
   "id": "f5966fb2",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>carat</th>\n",
       "      <th>cut</th>\n",
       "      <th>color</th>\n",
       "      <th>clarity</th>\n",
       "      <th>depth</th>\n",
       "      <th>table</th>\n",
       "      <th>price</th>\n",
       "      <th>x</th>\n",
       "      <th>y</th>\n",
       "      <th>z</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>27415</th>\n",
       "      <td>5.01</td>\n",
       "      <td>Fair</td>\n",
       "      <td>J</td>\n",
       "      <td>I1</td>\n",
       "      <td>65.5</td>\n",
       "      <td>59.0</td>\n",
       "      <td>18018</td>\n",
       "      <td>10.74</td>\n",
       "      <td>10.54</td>\n",
       "      <td>6.98</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>27630</th>\n",
       "      <td>4.50</td>\n",
       "      <td>Fair</td>\n",
       "      <td>J</td>\n",
       "      <td>I1</td>\n",
       "      <td>65.8</td>\n",
       "      <td>58.0</td>\n",
       "      <td>18531</td>\n",
       "      <td>10.23</td>\n",
       "      <td>10.16</td>\n",
       "      <td>6.72</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>27130</th>\n",
       "      <td>4.13</td>\n",
       "      <td>Fair</td>\n",
       "      <td>H</td>\n",
       "      <td>I1</td>\n",
       "      <td>64.8</td>\n",
       "      <td>61.0</td>\n",
       "      <td>17329</td>\n",
       "      <td>10.00</td>\n",
       "      <td>9.85</td>\n",
       "      <td>6.43</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>25999</th>\n",
       "      <td>4.01</td>\n",
       "      <td>Premium</td>\n",
       "      <td>J</td>\n",
       "      <td>I1</td>\n",
       "      <td>62.5</td>\n",
       "      <td>62.0</td>\n",
       "      <td>15223</td>\n",
       "      <td>10.02</td>\n",
       "      <td>9.94</td>\n",
       "      <td>6.24</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>25998</th>\n",
       "      <td>4.01</td>\n",
       "      <td>Premium</td>\n",
       "      <td>I</td>\n",
       "      <td>I1</td>\n",
       "      <td>61.0</td>\n",
       "      <td>61.0</td>\n",
       "      <td>15223</td>\n",
       "      <td>10.14</td>\n",
       "      <td>10.10</td>\n",
       "      <td>6.17</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",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>31592</th>\n",
       "      <td>0.20</td>\n",
       "      <td>Premium</td>\n",
       "      <td>E</td>\n",
       "      <td>VS2</td>\n",
       "      <td>59.0</td>\n",
       "      <td>60.0</td>\n",
       "      <td>367</td>\n",
       "      <td>3.81</td>\n",
       "      <td>3.78</td>\n",
       "      <td>2.24</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>31591</th>\n",
       "      <td>0.20</td>\n",
       "      <td>Premium</td>\n",
       "      <td>E</td>\n",
       "      <td>VS2</td>\n",
       "      <td>59.8</td>\n",
       "      <td>62.0</td>\n",
       "      <td>367</td>\n",
       "      <td>3.79</td>\n",
       "      <td>3.77</td>\n",
       "      <td>2.26</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>31601</th>\n",
       "      <td>0.20</td>\n",
       "      <td>Premium</td>\n",
       "      <td>D</td>\n",
       "      <td>VS2</td>\n",
       "      <td>61.7</td>\n",
       "      <td>60.0</td>\n",
       "      <td>367</td>\n",
       "      <td>3.77</td>\n",
       "      <td>3.72</td>\n",
       "      <td>2.31</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>14</th>\n",
       "      <td>0.20</td>\n",
       "      <td>Premium</td>\n",
       "      <td>E</td>\n",
       "      <td>SI2</td>\n",
       "      <td>60.2</td>\n",
       "      <td>62.0</td>\n",
       "      <td>345</td>\n",
       "      <td>3.79</td>\n",
       "      <td>3.75</td>\n",
       "      <td>2.27</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>31596</th>\n",
       "      <td>0.20</td>\n",
       "      <td>Premium</td>\n",
       "      <td>F</td>\n",
       "      <td>VS2</td>\n",
       "      <td>62.6</td>\n",
       "      <td>59.0</td>\n",
       "      <td>367</td>\n",
       "      <td>3.73</td>\n",
       "      <td>3.71</td>\n",
       "      <td>2.33</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>53940 rows × 10 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "       carat      cut color clarity  depth  table  price      x      y     z\n",
       "27415   5.01     Fair     J      I1   65.5   59.0  18018  10.74  10.54  6.98\n",
       "27630   4.50     Fair     J      I1   65.8   58.0  18531  10.23  10.16  6.72\n",
       "27130   4.13     Fair     H      I1   64.8   61.0  17329  10.00   9.85  6.43\n",
       "25999   4.01  Premium     J      I1   62.5   62.0  15223  10.02   9.94  6.24\n",
       "25998   4.01  Premium     I      I1   61.0   61.0  15223  10.14  10.10  6.17\n",
       "...      ...      ...   ...     ...    ...    ...    ...    ...    ...   ...\n",
       "31592   0.20  Premium     E     VS2   59.0   60.0    367   3.81   3.78  2.24\n",
       "31591   0.20  Premium     E     VS2   59.8   62.0    367   3.79   3.77  2.26\n",
       "31601   0.20  Premium     D     VS2   61.7   60.0    367   3.77   3.72  2.31\n",
       "14      0.20  Premium     E     SI2   60.2   62.0    345   3.79   3.75  2.27\n",
       "31596   0.20  Premium     F     VS2   62.6   59.0    367   3.73   3.71  2.33\n",
       "\n",
       "[53940 rows x 10 columns]"
      ]
     },
     "execution_count": 91,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# ii. carat : 내림차순\n",
    "diamonds.sort_values(by = \"carat\", ascending = False)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 92,
   "id": "35d5fe69",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>carat</th>\n",
       "      <th>cut</th>\n",
       "      <th>color</th>\n",
       "      <th>clarity</th>\n",
       "      <th>depth</th>\n",
       "      <th>table</th>\n",
       "      <th>price</th>\n",
       "      <th>x</th>\n",
       "      <th>y</th>\n",
       "      <th>z</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>0.23</td>\n",
       "      <td>Ideal</td>\n",
       "      <td>E</td>\n",
       "      <td>SI2</td>\n",
       "      <td>61.5</td>\n",
       "      <td>55.0</td>\n",
       "      <td>326</td>\n",
       "      <td>3.95</td>\n",
       "      <td>3.98</td>\n",
       "      <td>2.43</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>30779</th>\n",
       "      <td>0.33</td>\n",
       "      <td>Ideal</td>\n",
       "      <td>E</td>\n",
       "      <td>VVS2</td>\n",
       "      <td>61.0</td>\n",
       "      <td>54.0</td>\n",
       "      <td>741</td>\n",
       "      <td>4.49</td>\n",
       "      <td>4.52</td>\n",
       "      <td>2.75</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>30777</th>\n",
       "      <td>0.30</td>\n",
       "      <td>Ideal</td>\n",
       "      <td>D</td>\n",
       "      <td>VVS2</td>\n",
       "      <td>59.5</td>\n",
       "      <td>56.0</td>\n",
       "      <td>741</td>\n",
       "      <td>4.38</td>\n",
       "      <td>4.42</td>\n",
       "      <td>2.62</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>30776</th>\n",
       "      <td>0.40</td>\n",
       "      <td>Ideal</td>\n",
       "      <td>G</td>\n",
       "      <td>SI1</td>\n",
       "      <td>62.5</td>\n",
       "      <td>54.0</td>\n",
       "      <td>741</td>\n",
       "      <td>4.74</td>\n",
       "      <td>4.77</td>\n",
       "      <td>2.97</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>30775</th>\n",
       "      <td>0.40</td>\n",
       "      <td>Ideal</td>\n",
       "      <td>G</td>\n",
       "      <td>SI1</td>\n",
       "      <td>62.3</td>\n",
       "      <td>54.0</td>\n",
       "      <td>741</td>\n",
       "      <td>4.74</td>\n",
       "      <td>4.77</td>\n",
       "      <td>2.96</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",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8197</th>\n",
       "      <td>1.20</td>\n",
       "      <td>Fair</td>\n",
       "      <td>H</td>\n",
       "      <td>SI2</td>\n",
       "      <td>64.8</td>\n",
       "      <td>58.0</td>\n",
       "      <td>4368</td>\n",
       "      <td>6.61</td>\n",
       "      <td>6.56</td>\n",
       "      <td>4.27</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>14634</th>\n",
       "      <td>1.00</td>\n",
       "      <td>Fair</td>\n",
       "      <td>G</td>\n",
       "      <td>VS1</td>\n",
       "      <td>64.4</td>\n",
       "      <td>56.0</td>\n",
       "      <td>5908</td>\n",
       "      <td>6.31</td>\n",
       "      <td>6.26</td>\n",
       "      <td>4.05</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8200</th>\n",
       "      <td>1.00</td>\n",
       "      <td>Fair</td>\n",
       "      <td>H</td>\n",
       "      <td>VS1</td>\n",
       "      <td>65.4</td>\n",
       "      <td>61.0</td>\n",
       "      <td>4368</td>\n",
       "      <td>6.22</td>\n",
       "      <td>6.17</td>\n",
       "      <td>4.05</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8186</th>\n",
       "      <td>1.50</td>\n",
       "      <td>Fair</td>\n",
       "      <td>I</td>\n",
       "      <td>I1</td>\n",
       "      <td>71.3</td>\n",
       "      <td>58.0</td>\n",
       "      <td>4368</td>\n",
       "      <td>6.85</td>\n",
       "      <td>6.81</td>\n",
       "      <td>4.87</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3605</th>\n",
       "      <td>1.00</td>\n",
       "      <td>Fair</td>\n",
       "      <td>J</td>\n",
       "      <td>SI1</td>\n",
       "      <td>64.9</td>\n",
       "      <td>58.0</td>\n",
       "      <td>3427</td>\n",
       "      <td>6.29</td>\n",
       "      <td>6.23</td>\n",
       "      <td>4.06</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>53940 rows × 10 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "       carat    cut color clarity  depth  table  price     x     y     z\n",
       "0       0.23  Ideal     E     SI2   61.5   55.0    326  3.95  3.98  2.43\n",
       "30779   0.33  Ideal     E    VVS2   61.0   54.0    741  4.49  4.52  2.75\n",
       "30777   0.30  Ideal     D    VVS2   59.5   56.0    741  4.38  4.42  2.62\n",
       "30776   0.40  Ideal     G     SI1   62.5   54.0    741  4.74  4.77  2.97\n",
       "30775   0.40  Ideal     G     SI1   62.3   54.0    741  4.74  4.77  2.96\n",
       "...      ...    ...   ...     ...    ...    ...    ...   ...   ...   ...\n",
       "8197    1.20   Fair     H     SI2   64.8   58.0   4368  6.61  6.56  4.27\n",
       "14634   1.00   Fair     G     VS1   64.4   56.0   5908  6.31  6.26  4.05\n",
       "8200    1.00   Fair     H     VS1   65.4   61.0   4368  6.22  6.17  4.05\n",
       "8186    1.50   Fair     I      I1   71.3   58.0   4368  6.85  6.81  4.87\n",
       "3605    1.00   Fair     J     SI1   64.9   58.0   3427  6.29  6.23  4.06\n",
       "\n",
       "[53940 rows x 10 columns]"
      ]
     },
     "execution_count": 92,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# iii. cut : 문자형 : 오름차순\n",
    "# \"Fair\", \"Good\", \"Very Good\", \"Premium\", \"Ideal\"\n",
    "# 품질 순서로 정렬\n",
    "diamonds.sort_values(by = \"cut\", ascending = True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 93,
   "id": "106b9abd",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>carat</th>\n",
       "      <th>cut</th>\n",
       "      <th>color</th>\n",
       "      <th>clarity</th>\n",
       "      <th>depth</th>\n",
       "      <th>table</th>\n",
       "      <th>price</th>\n",
       "      <th>x</th>\n",
       "      <th>y</th>\n",
       "      <th>z</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>12005</th>\n",
       "      <td>1.01</td>\n",
       "      <td>Fair</td>\n",
       "      <td>D</td>\n",
       "      <td>SI1</td>\n",
       "      <td>64.4</td>\n",
       "      <td>61.0</td>\n",
       "      <td>5147</td>\n",
       "      <td>6.17</td>\n",
       "      <td>6.13</td>\n",
       "      <td>3.96</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2322</th>\n",
       "      <td>1.01</td>\n",
       "      <td>Fair</td>\n",
       "      <td>H</td>\n",
       "      <td>SI2</td>\n",
       "      <td>64.5</td>\n",
       "      <td>54.0</td>\n",
       "      <td>3167</td>\n",
       "      <td>6.26</td>\n",
       "      <td>6.15</td>\n",
       "      <td>4.00</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>39713</th>\n",
       "      <td>0.41</td>\n",
       "      <td>Fair</td>\n",
       "      <td>D</td>\n",
       "      <td>VVS2</td>\n",
       "      <td>64.5</td>\n",
       "      <td>57.0</td>\n",
       "      <td>1089</td>\n",
       "      <td>4.68</td>\n",
       "      <td>4.65</td>\n",
       "      <td>3.01</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>29444</th>\n",
       "      <td>0.50</td>\n",
       "      <td>Fair</td>\n",
       "      <td>G</td>\n",
       "      <td>I1</td>\n",
       "      <td>64.6</td>\n",
       "      <td>60.0</td>\n",
       "      <td>701</td>\n",
       "      <td>5.04</td>\n",
       "      <td>4.96</td>\n",
       "      <td>3.23</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2336</th>\n",
       "      <td>1.09</td>\n",
       "      <td>Fair</td>\n",
       "      <td>F</td>\n",
       "      <td>SI2</td>\n",
       "      <td>66.3</td>\n",
       "      <td>56.0</td>\n",
       "      <td>3170</td>\n",
       "      <td>6.47</td>\n",
       "      <td>6.39</td>\n",
       "      <td>4.27</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",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>29316</th>\n",
       "      <td>0.31</td>\n",
       "      <td>Ideal</td>\n",
       "      <td>H</td>\n",
       "      <td>VS1</td>\n",
       "      <td>62.6</td>\n",
       "      <td>55.0</td>\n",
       "      <td>698</td>\n",
       "      <td>4.38</td>\n",
       "      <td>4.34</td>\n",
       "      <td>2.73</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>29315</th>\n",
       "      <td>0.31</td>\n",
       "      <td>Ideal</td>\n",
       "      <td>H</td>\n",
       "      <td>VS1</td>\n",
       "      <td>61.1</td>\n",
       "      <td>57.0</td>\n",
       "      <td>698</td>\n",
       "      <td>4.39</td>\n",
       "      <td>4.35</td>\n",
       "      <td>2.67</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>29314</th>\n",
       "      <td>0.31</td>\n",
       "      <td>Ideal</td>\n",
       "      <td>H</td>\n",
       "      <td>VS1</td>\n",
       "      <td>62.0</td>\n",
       "      <td>54.0</td>\n",
       "      <td>698</td>\n",
       "      <td>4.39</td>\n",
       "      <td>4.35</td>\n",
       "      <td>2.71</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>29309</th>\n",
       "      <td>0.34</td>\n",
       "      <td>Ideal</td>\n",
       "      <td>F</td>\n",
       "      <td>SI2</td>\n",
       "      <td>61.9</td>\n",
       "      <td>57.0</td>\n",
       "      <td>438</td>\n",
       "      <td>4.50</td>\n",
       "      <td>4.52</td>\n",
       "      <td>2.79</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>53939</th>\n",
       "      <td>0.75</td>\n",
       "      <td>Ideal</td>\n",
       "      <td>D</td>\n",
       "      <td>SI2</td>\n",
       "      <td>62.2</td>\n",
       "      <td>55.0</td>\n",
       "      <td>2757</td>\n",
       "      <td>5.83</td>\n",
       "      <td>5.87</td>\n",
       "      <td>3.64</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>53940 rows × 10 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "       carat    cut color clarity  depth  table  price     x     y     z\n",
       "12005   1.01   Fair     D     SI1   64.4   61.0   5147  6.17  6.13  3.96\n",
       "2322    1.01   Fair     H     SI2   64.5   54.0   3167  6.26  6.15  4.00\n",
       "39713   0.41   Fair     D    VVS2   64.5   57.0   1089  4.68  4.65  3.01\n",
       "29444   0.50   Fair     G      I1   64.6   60.0    701  5.04  4.96  3.23\n",
       "2336    1.09   Fair     F     SI2   66.3   56.0   3170  6.47  6.39  4.27\n",
       "...      ...    ...   ...     ...    ...    ...    ...   ...   ...   ...\n",
       "29316   0.31  Ideal     H     VS1   62.6   55.0    698  4.38  4.34  2.73\n",
       "29315   0.31  Ideal     H     VS1   61.1   57.0    698  4.39  4.35  2.67\n",
       "29314   0.31  Ideal     H     VS1   62.0   54.0    698  4.39  4.35  2.71\n",
       "29309   0.34  Ideal     F     SI2   61.9   57.0    438  4.50  4.52  2.79\n",
       "53939   0.75  Ideal     D     SI2   62.2   55.0   2757  5.83  5.87  3.64\n",
       "\n",
       "[53940 rows x 10 columns]"
      ]
     },
     "execution_count": 93,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# iv. cut : 문자형 : 내림차순\n",
    "# \"Fair\", \"Good\", \"Very Good\", \"Premium\", \"Ideal\"\n",
    "# 품질 순서로 정렬\n",
    "diamonds.sort_values(by = \"cut\", ascending = False)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 94,
   "id": "655fb204",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>carat</th>\n",
       "      <th>cut</th>\n",
       "      <th>color</th>\n",
       "      <th>clarity</th>\n",
       "      <th>depth</th>\n",
       "      <th>table</th>\n",
       "      <th>price</th>\n",
       "      <th>x</th>\n",
       "      <th>y</th>\n",
       "      <th>z</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>31595</th>\n",
       "      <td>0.20</td>\n",
       "      <td>Ideal</td>\n",
       "      <td>E</td>\n",
       "      <td>VS2</td>\n",
       "      <td>59.7</td>\n",
       "      <td>55.0</td>\n",
       "      <td>367</td>\n",
       "      <td>3.86</td>\n",
       "      <td>3.84</td>\n",
       "      <td>2.30</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>31597</th>\n",
       "      <td>0.20</td>\n",
       "      <td>Ideal</td>\n",
       "      <td>D</td>\n",
       "      <td>VS2</td>\n",
       "      <td>61.5</td>\n",
       "      <td>57.0</td>\n",
       "      <td>367</td>\n",
       "      <td>3.81</td>\n",
       "      <td>3.77</td>\n",
       "      <td>2.33</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>31599</th>\n",
       "      <td>0.20</td>\n",
       "      <td>Ideal</td>\n",
       "      <td>E</td>\n",
       "      <td>VS2</td>\n",
       "      <td>62.2</td>\n",
       "      <td>57.0</td>\n",
       "      <td>367</td>\n",
       "      <td>3.76</td>\n",
       "      <td>3.73</td>\n",
       "      <td>2.33</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>0.23</td>\n",
       "      <td>Ideal</td>\n",
       "      <td>E</td>\n",
       "      <td>SI2</td>\n",
       "      <td>61.5</td>\n",
       "      <td>55.0</td>\n",
       "      <td>326</td>\n",
       "      <td>3.95</td>\n",
       "      <td>3.98</td>\n",
       "      <td>2.43</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>11</th>\n",
       "      <td>0.23</td>\n",
       "      <td>Ideal</td>\n",
       "      <td>J</td>\n",
       "      <td>VS1</td>\n",
       "      <td>62.8</td>\n",
       "      <td>56.0</td>\n",
       "      <td>340</td>\n",
       "      <td>3.93</td>\n",
       "      <td>3.90</td>\n",
       "      <td>2.46</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",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>26431</th>\n",
       "      <td>3.40</td>\n",
       "      <td>Fair</td>\n",
       "      <td>D</td>\n",
       "      <td>I1</td>\n",
       "      <td>66.8</td>\n",
       "      <td>52.0</td>\n",
       "      <td>15964</td>\n",
       "      <td>9.42</td>\n",
       "      <td>9.34</td>\n",
       "      <td>6.27</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>23644</th>\n",
       "      <td>3.65</td>\n",
       "      <td>Fair</td>\n",
       "      <td>H</td>\n",
       "      <td>I1</td>\n",
       "      <td>67.1</td>\n",
       "      <td>53.0</td>\n",
       "      <td>11668</td>\n",
       "      <td>9.53</td>\n",
       "      <td>9.48</td>\n",
       "      <td>6.38</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>27130</th>\n",
       "      <td>4.13</td>\n",
       "      <td>Fair</td>\n",
       "      <td>H</td>\n",
       "      <td>I1</td>\n",
       "      <td>64.8</td>\n",
       "      <td>61.0</td>\n",
       "      <td>17329</td>\n",
       "      <td>10.00</td>\n",
       "      <td>9.85</td>\n",
       "      <td>6.43</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>27630</th>\n",
       "      <td>4.50</td>\n",
       "      <td>Fair</td>\n",
       "      <td>J</td>\n",
       "      <td>I1</td>\n",
       "      <td>65.8</td>\n",
       "      <td>58.0</td>\n",
       "      <td>18531</td>\n",
       "      <td>10.23</td>\n",
       "      <td>10.16</td>\n",
       "      <td>6.72</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>27415</th>\n",
       "      <td>5.01</td>\n",
       "      <td>Fair</td>\n",
       "      <td>J</td>\n",
       "      <td>I1</td>\n",
       "      <td>65.5</td>\n",
       "      <td>59.0</td>\n",
       "      <td>18018</td>\n",
       "      <td>10.74</td>\n",
       "      <td>10.54</td>\n",
       "      <td>6.98</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>53940 rows × 10 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "       carat    cut color clarity  depth  table  price      x      y     z\n",
       "31595   0.20  Ideal     E     VS2   59.7   55.0    367   3.86   3.84  2.30\n",
       "31597   0.20  Ideal     D     VS2   61.5   57.0    367   3.81   3.77  2.33\n",
       "31599   0.20  Ideal     E     VS2   62.2   57.0    367   3.76   3.73  2.33\n",
       "0       0.23  Ideal     E     SI2   61.5   55.0    326   3.95   3.98  2.43\n",
       "11      0.23  Ideal     J     VS1   62.8   56.0    340   3.93   3.90  2.46\n",
       "...      ...    ...   ...     ...    ...    ...    ...    ...    ...   ...\n",
       "26431   3.40   Fair     D      I1   66.8   52.0  15964   9.42   9.34  6.27\n",
       "23644   3.65   Fair     H      I1   67.1   53.0  11668   9.53   9.48  6.38\n",
       "27130   4.13   Fair     H      I1   64.8   61.0  17329  10.00   9.85  6.43\n",
       "27630   4.50   Fair     J      I1   65.8   58.0  18531  10.23  10.16  6.72\n",
       "27415   5.01   Fair     J      I1   65.5   59.0  18018  10.74  10.54  6.98\n",
       "\n",
       "[53940 rows x 10 columns]"
      ]
     },
     "execution_count": 94,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# v. cut : 오름차순, carat : 오름차순\n",
    "diamonds.sort_values(by = [\"cut\", \"carat\"], ascending = True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 95,
   "id": "9ac26e52",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>carat</th>\n",
       "      <th>cut</th>\n",
       "      <th>color</th>\n",
       "      <th>clarity</th>\n",
       "      <th>depth</th>\n",
       "      <th>table</th>\n",
       "      <th>price</th>\n",
       "      <th>x</th>\n",
       "      <th>y</th>\n",
       "      <th>z</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>27415</th>\n",
       "      <td>5.01</td>\n",
       "      <td>Fair</td>\n",
       "      <td>J</td>\n",
       "      <td>I1</td>\n",
       "      <td>65.5</td>\n",
       "      <td>59.0</td>\n",
       "      <td>18018</td>\n",
       "      <td>10.74</td>\n",
       "      <td>10.54</td>\n",
       "      <td>6.98</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>27630</th>\n",
       "      <td>4.50</td>\n",
       "      <td>Fair</td>\n",
       "      <td>J</td>\n",
       "      <td>I1</td>\n",
       "      <td>65.8</td>\n",
       "      <td>58.0</td>\n",
       "      <td>18531</td>\n",
       "      <td>10.23</td>\n",
       "      <td>10.16</td>\n",
       "      <td>6.72</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>27130</th>\n",
       "      <td>4.13</td>\n",
       "      <td>Fair</td>\n",
       "      <td>H</td>\n",
       "      <td>I1</td>\n",
       "      <td>64.8</td>\n",
       "      <td>61.0</td>\n",
       "      <td>17329</td>\n",
       "      <td>10.00</td>\n",
       "      <td>9.85</td>\n",
       "      <td>6.43</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>23644</th>\n",
       "      <td>3.65</td>\n",
       "      <td>Fair</td>\n",
       "      <td>H</td>\n",
       "      <td>I1</td>\n",
       "      <td>67.1</td>\n",
       "      <td>53.0</td>\n",
       "      <td>11668</td>\n",
       "      <td>9.53</td>\n",
       "      <td>9.48</td>\n",
       "      <td>6.38</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>26431</th>\n",
       "      <td>3.40</td>\n",
       "      <td>Fair</td>\n",
       "      <td>D</td>\n",
       "      <td>I1</td>\n",
       "      <td>66.8</td>\n",
       "      <td>52.0</td>\n",
       "      <td>15964</td>\n",
       "      <td>9.42</td>\n",
       "      <td>9.34</td>\n",
       "      <td>6.27</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",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>52303</th>\n",
       "      <td>0.23</td>\n",
       "      <td>Ideal</td>\n",
       "      <td>F</td>\n",
       "      <td>VS1</td>\n",
       "      <td>61.4</td>\n",
       "      <td>55.0</td>\n",
       "      <td>548</td>\n",
       "      <td>3.96</td>\n",
       "      <td>3.99</td>\n",
       "      <td>2.44</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>52304</th>\n",
       "      <td>0.23</td>\n",
       "      <td>Ideal</td>\n",
       "      <td>F</td>\n",
       "      <td>VS1</td>\n",
       "      <td>61.1</td>\n",
       "      <td>56.0</td>\n",
       "      <td>548</td>\n",
       "      <td>3.97</td>\n",
       "      <td>3.99</td>\n",
       "      <td>2.43</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>31595</th>\n",
       "      <td>0.20</td>\n",
       "      <td>Ideal</td>\n",
       "      <td>E</td>\n",
       "      <td>VS2</td>\n",
       "      <td>59.7</td>\n",
       "      <td>55.0</td>\n",
       "      <td>367</td>\n",
       "      <td>3.86</td>\n",
       "      <td>3.84</td>\n",
       "      <td>2.30</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>31597</th>\n",
       "      <td>0.20</td>\n",
       "      <td>Ideal</td>\n",
       "      <td>D</td>\n",
       "      <td>VS2</td>\n",
       "      <td>61.5</td>\n",
       "      <td>57.0</td>\n",
       "      <td>367</td>\n",
       "      <td>3.81</td>\n",
       "      <td>3.77</td>\n",
       "      <td>2.33</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>31599</th>\n",
       "      <td>0.20</td>\n",
       "      <td>Ideal</td>\n",
       "      <td>E</td>\n",
       "      <td>VS2</td>\n",
       "      <td>62.2</td>\n",
       "      <td>57.0</td>\n",
       "      <td>367</td>\n",
       "      <td>3.76</td>\n",
       "      <td>3.73</td>\n",
       "      <td>2.33</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>53940 rows × 10 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "       carat    cut color clarity  depth  table  price      x      y     z\n",
       "27415   5.01   Fair     J      I1   65.5   59.0  18018  10.74  10.54  6.98\n",
       "27630   4.50   Fair     J      I1   65.8   58.0  18531  10.23  10.16  6.72\n",
       "27130   4.13   Fair     H      I1   64.8   61.0  17329  10.00   9.85  6.43\n",
       "23644   3.65   Fair     H      I1   67.1   53.0  11668   9.53   9.48  6.38\n",
       "26431   3.40   Fair     D      I1   66.8   52.0  15964   9.42   9.34  6.27\n",
       "...      ...    ...   ...     ...    ...    ...    ...    ...    ...   ...\n",
       "52303   0.23  Ideal     F     VS1   61.4   55.0    548   3.96   3.99  2.44\n",
       "52304   0.23  Ideal     F     VS1   61.1   56.0    548   3.97   3.99  2.43\n",
       "31595   0.20  Ideal     E     VS2   59.7   55.0    367   3.86   3.84  2.30\n",
       "31597   0.20  Ideal     D     VS2   61.5   57.0    367   3.81   3.77  2.33\n",
       "31599   0.20  Ideal     E     VS2   62.2   57.0    367   3.76   3.73  2.33\n",
       "\n",
       "[53940 rows x 10 columns]"
      ]
     },
     "execution_count": 95,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# vi. cut : 내림차순, carat : 내림차순\n",
    "diamonds.sort_values(by = [\"cut\", \"carat\"], ascending = False)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 96,
   "id": "b8079ecb",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>carat</th>\n",
       "      <th>cut</th>\n",
       "      <th>color</th>\n",
       "      <th>clarity</th>\n",
       "      <th>depth</th>\n",
       "      <th>table</th>\n",
       "      <th>price</th>\n",
       "      <th>x</th>\n",
       "      <th>y</th>\n",
       "      <th>z</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>24328</th>\n",
       "      <td>3.50</td>\n",
       "      <td>Ideal</td>\n",
       "      <td>H</td>\n",
       "      <td>I1</td>\n",
       "      <td>62.8</td>\n",
       "      <td>57.0</td>\n",
       "      <td>12587</td>\n",
       "      <td>9.65</td>\n",
       "      <td>9.59</td>\n",
       "      <td>6.03</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>24297</th>\n",
       "      <td>3.22</td>\n",
       "      <td>Ideal</td>\n",
       "      <td>I</td>\n",
       "      <td>I1</td>\n",
       "      <td>62.6</td>\n",
       "      <td>55.0</td>\n",
       "      <td>12545</td>\n",
       "      <td>9.49</td>\n",
       "      <td>9.42</td>\n",
       "      <td>5.92</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>26467</th>\n",
       "      <td>3.01</td>\n",
       "      <td>Ideal</td>\n",
       "      <td>J</td>\n",
       "      <td>SI2</td>\n",
       "      <td>61.7</td>\n",
       "      <td>58.0</td>\n",
       "      <td>16037</td>\n",
       "      <td>9.25</td>\n",
       "      <td>9.20</td>\n",
       "      <td>5.69</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>26744</th>\n",
       "      <td>3.01</td>\n",
       "      <td>Ideal</td>\n",
       "      <td>J</td>\n",
       "      <td>I1</td>\n",
       "      <td>65.4</td>\n",
       "      <td>60.0</td>\n",
       "      <td>16538</td>\n",
       "      <td>8.99</td>\n",
       "      <td>8.93</td>\n",
       "      <td>5.86</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>24784</th>\n",
       "      <td>2.75</td>\n",
       "      <td>Ideal</td>\n",
       "      <td>D</td>\n",
       "      <td>I1</td>\n",
       "      <td>60.9</td>\n",
       "      <td>57.0</td>\n",
       "      <td>13156</td>\n",
       "      <td>9.04</td>\n",
       "      <td>8.98</td>\n",
       "      <td>5.49</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",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2711</th>\n",
       "      <td>0.25</td>\n",
       "      <td>Fair</td>\n",
       "      <td>D</td>\n",
       "      <td>VS1</td>\n",
       "      <td>61.2</td>\n",
       "      <td>55.0</td>\n",
       "      <td>563</td>\n",
       "      <td>4.09</td>\n",
       "      <td>4.11</td>\n",
       "      <td>2.51</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>28270</th>\n",
       "      <td>0.25</td>\n",
       "      <td>Fair</td>\n",
       "      <td>E</td>\n",
       "      <td>VS1</td>\n",
       "      <td>55.2</td>\n",
       "      <td>64.0</td>\n",
       "      <td>361</td>\n",
       "      <td>4.21</td>\n",
       "      <td>4.23</td>\n",
       "      <td>2.33</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>38153</th>\n",
       "      <td>0.25</td>\n",
       "      <td>Fair</td>\n",
       "      <td>F</td>\n",
       "      <td>SI2</td>\n",
       "      <td>54.4</td>\n",
       "      <td>64.0</td>\n",
       "      <td>1013</td>\n",
       "      <td>4.30</td>\n",
       "      <td>4.23</td>\n",
       "      <td>2.32</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>31611</th>\n",
       "      <td>0.23</td>\n",
       "      <td>Fair</td>\n",
       "      <td>G</td>\n",
       "      <td>VVS2</td>\n",
       "      <td>61.4</td>\n",
       "      <td>66.0</td>\n",
       "      <td>369</td>\n",
       "      <td>3.87</td>\n",
       "      <td>3.91</td>\n",
       "      <td>2.39</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8</th>\n",
       "      <td>0.22</td>\n",
       "      <td>Fair</td>\n",
       "      <td>E</td>\n",
       "      <td>VS2</td>\n",
       "      <td>65.1</td>\n",
       "      <td>61.0</td>\n",
       "      <td>337</td>\n",
       "      <td>3.87</td>\n",
       "      <td>3.78</td>\n",
       "      <td>2.49</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>53940 rows × 10 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "       carat    cut color clarity  depth  table  price     x     y     z\n",
       "24328   3.50  Ideal     H      I1   62.8   57.0  12587  9.65  9.59  6.03\n",
       "24297   3.22  Ideal     I      I1   62.6   55.0  12545  9.49  9.42  5.92\n",
       "26467   3.01  Ideal     J     SI2   61.7   58.0  16037  9.25  9.20  5.69\n",
       "26744   3.01  Ideal     J      I1   65.4   60.0  16538  8.99  8.93  5.86\n",
       "24784   2.75  Ideal     D      I1   60.9   57.0  13156  9.04  8.98  5.49\n",
       "...      ...    ...   ...     ...    ...    ...    ...   ...   ...   ...\n",
       "2711    0.25   Fair     D     VS1   61.2   55.0    563  4.09  4.11  2.51\n",
       "28270   0.25   Fair     E     VS1   55.2   64.0    361  4.21  4.23  2.33\n",
       "38153   0.25   Fair     F     SI2   54.4   64.0   1013  4.30  4.23  2.32\n",
       "31611   0.23   Fair     G    VVS2   61.4   66.0    369  3.87  3.91  2.39\n",
       "8       0.22   Fair     E     VS2   65.1   61.0    337  3.87  3.78  2.49\n",
       "\n",
       "[53940 rows x 10 columns]"
      ]
     },
     "execution_count": 96,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# vii. cut : 오름차순, carat : 내림차순\n",
    "diamonds.sort_values(by = [\"cut\", \"carat\"], ascending = [True, False])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 97,
   "id": "3fa4e8cb",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>carat</th>\n",
       "      <th>cut</th>\n",
       "      <th>color</th>\n",
       "      <th>clarity</th>\n",
       "      <th>depth</th>\n",
       "      <th>table</th>\n",
       "      <th>price</th>\n",
       "      <th>x</th>\n",
       "      <th>y</th>\n",
       "      <th>z</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>8</th>\n",
       "      <td>0.22</td>\n",
       "      <td>Fair</td>\n",
       "      <td>E</td>\n",
       "      <td>VS2</td>\n",
       "      <td>65.1</td>\n",
       "      <td>61.0</td>\n",
       "      <td>337</td>\n",
       "      <td>3.87</td>\n",
       "      <td>3.78</td>\n",
       "      <td>2.49</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>31611</th>\n",
       "      <td>0.23</td>\n",
       "      <td>Fair</td>\n",
       "      <td>G</td>\n",
       "      <td>VVS2</td>\n",
       "      <td>61.4</td>\n",
       "      <td>66.0</td>\n",
       "      <td>369</td>\n",
       "      <td>3.87</td>\n",
       "      <td>3.91</td>\n",
       "      <td>2.39</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2711</th>\n",
       "      <td>0.25</td>\n",
       "      <td>Fair</td>\n",
       "      <td>D</td>\n",
       "      <td>VS1</td>\n",
       "      <td>61.2</td>\n",
       "      <td>55.0</td>\n",
       "      <td>563</td>\n",
       "      <td>4.09</td>\n",
       "      <td>4.11</td>\n",
       "      <td>2.51</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>28270</th>\n",
       "      <td>0.25</td>\n",
       "      <td>Fair</td>\n",
       "      <td>E</td>\n",
       "      <td>VS1</td>\n",
       "      <td>55.2</td>\n",
       "      <td>64.0</td>\n",
       "      <td>361</td>\n",
       "      <td>4.21</td>\n",
       "      <td>4.23</td>\n",
       "      <td>2.33</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>38153</th>\n",
       "      <td>0.25</td>\n",
       "      <td>Fair</td>\n",
       "      <td>F</td>\n",
       "      <td>SI2</td>\n",
       "      <td>54.4</td>\n",
       "      <td>64.0</td>\n",
       "      <td>1013</td>\n",
       "      <td>4.30</td>\n",
       "      <td>4.23</td>\n",
       "      <td>2.32</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",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>24784</th>\n",
       "      <td>2.75</td>\n",
       "      <td>Ideal</td>\n",
       "      <td>D</td>\n",
       "      <td>I1</td>\n",
       "      <td>60.9</td>\n",
       "      <td>57.0</td>\n",
       "      <td>13156</td>\n",
       "      <td>9.04</td>\n",
       "      <td>8.98</td>\n",
       "      <td>5.49</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>26467</th>\n",
       "      <td>3.01</td>\n",
       "      <td>Ideal</td>\n",
       "      <td>J</td>\n",
       "      <td>SI2</td>\n",
       "      <td>61.7</td>\n",
       "      <td>58.0</td>\n",
       "      <td>16037</td>\n",
       "      <td>9.25</td>\n",
       "      <td>9.20</td>\n",
       "      <td>5.69</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>26744</th>\n",
       "      <td>3.01</td>\n",
       "      <td>Ideal</td>\n",
       "      <td>J</td>\n",
       "      <td>I1</td>\n",
       "      <td>65.4</td>\n",
       "      <td>60.0</td>\n",
       "      <td>16538</td>\n",
       "      <td>8.99</td>\n",
       "      <td>8.93</td>\n",
       "      <td>5.86</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>24297</th>\n",
       "      <td>3.22</td>\n",
       "      <td>Ideal</td>\n",
       "      <td>I</td>\n",
       "      <td>I1</td>\n",
       "      <td>62.6</td>\n",
       "      <td>55.0</td>\n",
       "      <td>12545</td>\n",
       "      <td>9.49</td>\n",
       "      <td>9.42</td>\n",
       "      <td>5.92</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>24328</th>\n",
       "      <td>3.50</td>\n",
       "      <td>Ideal</td>\n",
       "      <td>H</td>\n",
       "      <td>I1</td>\n",
       "      <td>62.8</td>\n",
       "      <td>57.0</td>\n",
       "      <td>12587</td>\n",
       "      <td>9.65</td>\n",
       "      <td>9.59</td>\n",
       "      <td>6.03</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>53940 rows × 10 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "       carat    cut color clarity  depth  table  price     x     y     z\n",
       "8       0.22   Fair     E     VS2   65.1   61.0    337  3.87  3.78  2.49\n",
       "31611   0.23   Fair     G    VVS2   61.4   66.0    369  3.87  3.91  2.39\n",
       "2711    0.25   Fair     D     VS1   61.2   55.0    563  4.09  4.11  2.51\n",
       "28270   0.25   Fair     E     VS1   55.2   64.0    361  4.21  4.23  2.33\n",
       "38153   0.25   Fair     F     SI2   54.4   64.0   1013  4.30  4.23  2.32\n",
       "...      ...    ...   ...     ...    ...    ...    ...   ...   ...   ...\n",
       "24784   2.75  Ideal     D      I1   60.9   57.0  13156  9.04  8.98  5.49\n",
       "26467   3.01  Ideal     J     SI2   61.7   58.0  16037  9.25  9.20  5.69\n",
       "26744   3.01  Ideal     J      I1   65.4   60.0  16538  8.99  8.93  5.86\n",
       "24297   3.22  Ideal     I      I1   62.6   55.0  12545  9.49  9.42  5.92\n",
       "24328   3.50  Ideal     H      I1   62.8   57.0  12587  9.65  9.59  6.03\n",
       "\n",
       "[53940 rows x 10 columns]"
      ]
     },
     "execution_count": 97,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# viii. cut : 내림차순, carat : 오름차순\n",
    "diamonds.sort_values(by = [\"cut\", \"carat\"], ascending = [False, True])"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "e7f454cb",
   "metadata": {},
   "source": [
    "#### 8. 데이터 합치기\n",
    "- 위/아래     : pandas.concat()\n",
    "- 왼쪽/오른쪽 : pandas.merge() "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 98,
   "id": "ce6822ca",
   "metadata": {},
   "outputs": [],
   "source": [
    "# (1) 위/아래\n",
    "# 2개 이상의 데이터가 열의 이름도 같고, 열의 순서도 같을 때에\n",
    "# 예를 들면, 2022년 매출 데이터, 2023년 매출 데이터\n",
    "\n",
    "df1 = pd.DataFrame({\"Year\"   : [2022, 2022, 2022, 2022],\n",
    "                    \"Item\"   : [\"A\", \"B\", \"C\", \"D\"],\n",
    "                    \"income\" : [1000, 2000, 3000, 4000]})\n",
    "\n",
    "df2 = pd.DataFrame({\"Year\"   : [2023, 2023, 2023, 2023],\n",
    "                    \"Item\"   : [\"A\", \"B\", \"C\", \"D\"],\n",
    "                    \"income\" : [500, 1000, 500, 1000]})"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 104,
   "id": "f0b31b98",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>Year</th>\n",
       "      <th>Item</th>\n",
       "      <th>income</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>2022</td>\n",
       "      <td>A</td>\n",
       "      <td>1000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>2022</td>\n",
       "      <td>B</td>\n",
       "      <td>2000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>2022</td>\n",
       "      <td>C</td>\n",
       "      <td>3000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>2022</td>\n",
       "      <td>D</td>\n",
       "      <td>4000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>2023</td>\n",
       "      <td>A</td>\n",
       "      <td>500</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>2023</td>\n",
       "      <td>B</td>\n",
       "      <td>1000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>2023</td>\n",
       "      <td>C</td>\n",
       "      <td>500</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>2023</td>\n",
       "      <td>D</td>\n",
       "      <td>1000</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   Year Item  income\n",
       "0  2022    A    1000\n",
       "1  2022    B    2000\n",
       "2  2022    C    3000\n",
       "3  2022    D    4000\n",
       "4  2023    A     500\n",
       "5  2023    B    1000\n",
       "6  2023    C     500\n",
       "7  2023    D    1000"
      ]
     },
     "execution_count": 104,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# data = pandas.concat(objs = [data1, data2, ...], axis = 0)\n",
    "\n",
    "# concat : concatenate : 연결하다\n",
    "# axis : 축\n",
    "# axis = 0 : 행\n",
    "# axis = 1 : 열\n",
    "\n",
    "df3 = pd.concat(objs = [df1, df2], axis = 0)\n",
    "df3.reset_index(drop = True)\n",
    "\n",
    "# reset_index() : 행의 이름을 수정\n",
    "# drop = True   : 기존의 행의 이름을 열로 저장하지 않고 삭제함"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 105,
   "id": "339b3d3b",
   "metadata": {},
   "outputs": [],
   "source": [
    "# (2) 왼쪽/오른쪽\n",
    "# 2개의 데이터만 합칠 수 있음\n",
    "# 각각 데이터에 일부의 정보를 저장하고 있는 경우\n",
    "\n",
    "# 예를들면, df4 : 고객정보, df5 : 거래정보\n",
    "\n",
    "# data = pandas.merge(data1, \n",
    "#                     data2, \n",
    "#                     on  = \"기본키 or 주요키 or Primary key\",\n",
    "#                     how = 결합 방법)\n",
    "\n",
    "# how : \"inner\", \"outer\", \"left\", \"right\""
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 106,
   "id": "69e53c68",
   "metadata": {},
   "outputs": [],
   "source": [
    "df4 = pd.DataFrame({\"id\"  : [1, 2, 4, 5],\n",
    "                    \"age\" : [10, 20, 40, 50]})\n",
    "\n",
    "df5 = pd.DataFrame({\"id\"  : [1, 3, 4, 6],\n",
    "                    \"bt\"  : [\"a\", \"b\", \"ab\", \"o\"]})"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 107,
   "id": "a7df303f",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>id</th>\n",
       "      <th>age</th>\n",
       "      <th>bt</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>1</td>\n",
       "      <td>10</td>\n",
       "      <td>a</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>4</td>\n",
       "      <td>40</td>\n",
       "      <td>ab</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   id  age  bt\n",
       "0   1   10   a\n",
       "1   4   40  ab"
      ]
     },
     "execution_count": 107,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# i. inner join\n",
    "# 2개의 데이터에 동시에 있는 primary key만 합침\n",
    "# pandas.merge(data1, data2, on = \"primary key\", how = \"inner\")\n",
    "\n",
    "df6 = pd.merge(df4, df5, on = \"id\", how = \"inner\")\n",
    "df6"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 111,
   "id": "c05663ae",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>id</th>\n",
       "      <th>age</th>\n",
       "      <th>bt</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>1</td>\n",
       "      <td>10.0</td>\n",
       "      <td>a</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>2</td>\n",
       "      <td>20.0</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>3</td>\n",
       "      <td>NaN</td>\n",
       "      <td>b</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>4</td>\n",
       "      <td>40.0</td>\n",
       "      <td>ab</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>5</td>\n",
       "      <td>50.0</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>6</td>\n",
       "      <td>NaN</td>\n",
       "      <td>o</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   id   age   bt\n",
       "0   1  10.0    a\n",
       "1   2  20.0  NaN\n",
       "2   3   NaN    b\n",
       "3   4  40.0   ab\n",
       "4   5  50.0  NaN\n",
       "5   6   NaN    o"
      ]
     },
     "execution_count": 111,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# ii. Outer Join : Full Join\n",
    "# 2개의 데이터에 primary key의 합집합\n",
    "# pandas.merge(data1, data2, on = \"primary key\", how = \"outer\")\n",
    "\n",
    "df7 = pd.merge(df4, df5, on = \"id\", how = \"outer\")\n",
    "df7 = df7.sort_values(by = \"id\", ascending = True)\n",
    "df7.reset_index(drop = True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 113,
   "id": "a3644841",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>id</th>\n",
       "      <th>age</th>\n",
       "      <th>bt</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>1</td>\n",
       "      <td>10</td>\n",
       "      <td>a</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>2</td>\n",
       "      <td>20</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>4</td>\n",
       "      <td>40</td>\n",
       "      <td>ab</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>5</td>\n",
       "      <td>50</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   id  age   bt\n",
       "0   1   10    a\n",
       "1   2   20  NaN\n",
       "2   4   40   ab\n",
       "3   5   50  NaN"
      ]
     },
     "execution_count": 113,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# iii. Outer Join : Left Join\n",
    "# merge()에 먼저 들어가는 데이터      : left\n",
    "# merge()에 두 번째로 들어가는 데이터 : rigth\n",
    "# merge()에 먼저 들어가는 데이터의 primary key만 합침\n",
    "\n",
    "# pandas.merge(data1, data2, on = \"primary key\", how = \"left\")\n",
    "\n",
    "df8 = pd.merge(df4, df5, on = \"id\", how = \"left\")\n",
    "df8"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 114,
   "id": "661f4acf",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>id</th>\n",
       "      <th>age</th>\n",
       "      <th>bt</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>1</td>\n",
       "      <td>10.0</td>\n",
       "      <td>a</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>3</td>\n",
       "      <td>NaN</td>\n",
       "      <td>b</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>4</td>\n",
       "      <td>40.0</td>\n",
       "      <td>ab</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>6</td>\n",
       "      <td>NaN</td>\n",
       "      <td>o</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   id   age  bt\n",
       "0   1  10.0   a\n",
       "1   3   NaN   b\n",
       "2   4  40.0  ab\n",
       "3   6   NaN   o"
      ]
     },
     "execution_count": 114,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# iv. Outer Join : Right Join\n",
    "# merge()에 먼저 들어가는 데이터      : left\n",
    "# merge()에 두 번째로 들어가는 데이터 : rigth\n",
    "# merge()에 두 번째로 들어가는 데이터의 primary key만 합침\n",
    "\n",
    "# pandas.merge(data1, data2, on = \"primary key\", how = \"right\")\n",
    "\n",
    "df9 = pd.merge(df4, df5, on = \"id\", how = \"right\")\n",
    "df9"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "255a10d7",
   "metadata": {},
   "outputs": [],
   "source": [
    "# 참고\n",
    "# 만약에 primary key가 2개 이상\n",
    "# pandas.merge(data1, data2, \n",
    "#              on  = [\"primary key1\", \"primary key2\"], \n",
    "#              how = \"outer\")"
   ]
  }
 ],
 "metadata": {
  "hide_input": false,
  "kernelspec": {
   "display_name": "Python 3 (ipykernel)",
   "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": "3.11.5"
  },
  "varInspector": {
   "cols": {
    "lenName": 16,
    "lenType": 16,
    "lenVar": 40
   },
   "kernels_config": {
    "python": {
     "delete_cmd_postfix": "",
     "delete_cmd_prefix": "del ",
     "library": "var_list.py",
     "varRefreshCmd": "print(var_dic_list())"
    },
    "r": {
     "delete_cmd_postfix": ") ",
     "delete_cmd_prefix": "rm(",
     "library": "var_list.r",
     "varRefreshCmd": "cat(var_dic_list()) "
    }
   },
   "types_to_exclude": [
    "module",
    "function",
    "builtin_function_or_method",
    "instance",
    "_Feature"
   ],
   "window_display": false
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}