Posted Updated  HS Python27 minutes read (About 4116 words)

Pandas 입문 2

groupby 집중 연습

라이브러리 불러오기

  • pandas 라이브러리 불러오고, supermarket_sales.csv 파일 불러오기
  • 참고로 미얀마 자료. 얀곤이 옛 수도, 네피도는 현 수도
Read more
Posted Updated  HS Python29 minutes read (About 4299 words)

Pandas 입문 1

판다스

라이브러리 불러오기

1
2
import pandas as pd
print(pd.__version__)
1.3.5

테스트

데이터 프레임

1
2
3
4
5
6
temp_dic = {"col1" : [1, 2, 3], 
            "col2" : [3, 4, 5]} # 먼저 딕셔너리를 만든다.
          
df = pd.DataFrame(temp_dic) # 판다스는 객체가 시리즈와 데이터프레임으로 나뉜다.
print(type(df))
print(df)
<class 'pandas.core.frame.DataFrame'>
   col1  col2
0     1     3
1     2     4
2     3     5

시리즈

1
2
3
4
temp_dic = {'a':1, 'b':2, 'c':3} # 인덱스는 숫자나 문자나 모두 가능하다.
ser = pd.Series(temp_dic)
print(ser)
print(type(ser))
a    1
b    2
c    3
dtype: int64
<class 'pandas.core.series.Series'>
  • 언뜻 보기에는 같아보인다.
  • 그러나, 다른 클래스고, 메서드도 다르다. 조심해야 한다.

데이터 불러오기

구글 드라이브 연동

Lemonade2016.csv 파일

1
2
from google.colab import drive
drive.mount('/content/drive')
Mounted at /content/drive

1
2
3
DATA_PATH = '/content/drive/MyDrive/Colab Notebooks/data/Lemonade2016.csv'
juice = pd.read_csv(DATA_PATH) # 객체를 대문자로 썼는데, 개발자들이 좋아하는 방법이다. 눈에 띄는 걸 좋아한다. 
juice

Date Location Lemon Orange Temperature Leaflets Price
0 7/1/2016 Park 97 67 70 90.0 0.25
1 7/2/2016 Park 98 67 72 90.0 0.25
2 7/3/2016 Park 110 77 71 104.0 0.25
3 7/4/2016 Beach 134 99 76 98.0 0.25
4 7/5/2016 Beach 159 118 78 135.0 0.25
5 7/6/2016 Beach 103 69 82 90.0 0.25
6 7/6/2016 Beach 103 69 82 90.0 0.25
7 7/7/2016 Beach 143 101 81 135.0 0.25
8 NaN Beach 123 86 82 113.0 0.25
9 7/9/2016 Beach 134 95 80 126.0 0.25
10 7/10/2016 Beach 140 98 82 131.0 0.25
11 7/11/2016 Beach 162 120 83 135.0 0.25
12 7/12/2016 Beach 130 95 84 99.0 0.25
13 7/13/2016 Beach 109 75 77 99.0 0.25
14 7/14/2016 Beach 122 85 78 113.0 0.25
15 7/15/2016 Beach 98 62 75 108.0 0.50
16 7/16/2016 Beach 81 50 74 90.0 0.50
17 7/17/2016 Beach 115 76 77 126.0 0.50
18 7/18/2016 Park 131 92 81 122.0 0.50
19 7/19/2016 Park 122 85 78 113.0 0.50
20 7/20/2016 Park 71 42 70 NaN 0.50
21 7/21/2016 Park 83 50 77 90.0 0.50
22 7/22/2016 Park 112 75 80 108.0 0.50
23 7/23/2016 Park 120 82 81 117.0 0.50
24 7/24/2016 Park 121 82 82 117.0 0.50
25 7/25/2016 Park 156 113 84 135.0 0.50
26 7/26/2016 Park 176 129 83 158.0 0.35
27 7/27/2016 Park 104 68 80 99.0 0.35
28 7/28/2016 Park 96 63 82 90.0 0.35
29 7/29/2016 Park 100 66 81 95.0 0.35
30 7/30/2016 Beach 88 57 82 81.0 0.35
31 7/31/2016 Beach 76 47 82 68.0 0.35 

  <script>
    const buttonEl =
      document.querySelector('#df-377049de-a05e-4731-8fe4-a3adc97f89dd button.colab-df-convert');
    buttonEl.style.display =
      google.colab.kernel.accessAllowed ? 'block' : 'none';

    async function convertToInteractive(key) {
      const element = document.querySelector('#df-377049de-a05e-4731-8fe4-a3adc97f89dd');
      const dataTable =
        await google.colab.kernel.invokeFunction('convertToInteractive',
                                                 [key], {});
      if (!dataTable) return;

      const docLinkHtml = 'Like what you see? Visit the ' +
        '<a target="_blank" href=https://colab.research.google.com/notebooks/data_table.ipynb>data table notebook</a>'
        + ' to learn more about interactive tables.';
      element.innerHTML = '';
      dataTable['output_type'] = 'display_data';
      await google.colab.output.renderOutput(dataTable, element);
      const docLink = document.createElement('div');
      docLink.innerHTML = docLinkHtml;
      element.appendChild(docLink);
    }
  </script>
</div>
  • 데이터를 불러왔다. 데이터프레임 메서드를 하나씩 해볼 것이다.
  • 첫번째 파악해야 하는 것
    • 데이터 구조
1
juice.info()
<class 'pandas.core.frame.DataFrame'>
RangeIndex: 32 entries, 0 to 31
Data columns (total 7 columns):
 #   Column       Non-Null Count  Dtype  
---  ------       --------------  -----  
 0   Date         31 non-null     object 
 1   Location     32 non-null     object 
 2   Lemon        32 non-null     int64  
 3   Orange       32 non-null     int64  
 4   Temperature  32 non-null     int64  
 5   Leaflets     31 non-null     float64
 6   Price        32 non-null     float64
dtypes: float64(2), int64(3), object(2)
memory usage: 1.9+ KB
  • 결측치(NaN)가 있다면 Non-Null Count의 수가 다르다. Date, Leaflets는 결측치가 하나씩 있기 때문에 다른 칼럼에 비해 수가 하나 모자르다. 31.
  • .head() 함수는 상위값
  • .tail() 함수는 하위값
1
juice.head()

Date Location Lemon Orange Temperature Leaflets Price
0 7/1/2016 Park 97 67 70 90.0 0.25
1 7/2/2016 Park 98 67 72 90.0 0.25
2 7/3/2016 Park 110 77 71 104.0 0.25
3 7/4/2016 Beach 134 99 76 98.0 0.25
4 7/5/2016 Beach 159 118 78 135.0 0.25 

  <script>
    const buttonEl =
      document.querySelector('#df-85ef7723-5da0-4047-9470-852c38d80b3d button.colab-df-convert');
    buttonEl.style.display =
      google.colab.kernel.accessAllowed ? 'block' : 'none';

    async function convertToInteractive(key) {
      const element = document.querySelector('#df-85ef7723-5da0-4047-9470-852c38d80b3d');
      const dataTable =
        await google.colab.kernel.invokeFunction('convertToInteractive',
                                                 [key], {});
      if (!dataTable) return;

      const docLinkHtml = 'Like what you see? Visit the ' +
        '<a target="_blank" href=https://colab.research.google.com/notebooks/data_table.ipynb>data table notebook</a>'
        + ' to learn more about interactive tables.';
      element.innerHTML = '';
      dataTable['output_type'] = 'display_data';
      await google.colab.output.renderOutput(dataTable, element);
      const docLink = document.createElement('div');
      docLink.innerHTML = docLinkHtml;
      element.appendChild(docLink);
    }
  </script>
</div>

1
juice.tail()

Date Location Lemon Orange Temperature Leaflets Price
27 7/27/2016 Park 104 68 80 99.0 0.35
28 7/28/2016 Park 96 63 82 90.0 0.35
29 7/29/2016 Park 100 66 81 95.0 0.35
30 7/30/2016 Beach 88 57 82 81.0 0.35
31 7/31/2016 Beach 76 47 82 68.0 0.35 

  <script>
    const buttonEl =
      document.querySelector('#df-2d82ba80-103f-4ffb-86a3-eb0b3ce806a0 button.colab-df-convert');
    buttonEl.style.display =
      google.colab.kernel.accessAllowed ? 'block' : 'none';

    async function convertToInteractive(key) {
      const element = document.querySelector('#df-2d82ba80-103f-4ffb-86a3-eb0b3ce806a0');
      const dataTable =
        await google.colab.kernel.invokeFunction('convertToInteractive',
                                                 [key], {});
      if (!dataTable) return;

      const docLinkHtml = 'Like what you see? Visit the ' +
        '<a target="_blank" href=https://colab.research.google.com/notebooks/data_table.ipynb>data table notebook</a>'
        + ' to learn more about interactive tables.';
      element.innerHTML = '';
      dataTable['output_type'] = 'display_data';
      await google.colab.output.renderOutput(dataTable, element);
      const docLink = document.createElement('div');
      docLink.innerHTML = docLinkHtml;
      element.appendChild(docLink);
    }
  </script>
</div>
  • describe() 함수
  • 기술통계량 확인해주는 함수
1
2
juice.describe()
# type(juice.describe()) -> dataframe

Lemon Orange Temperature Leaflets Price
count 32.000000 32.000000 32.000000 31.000000 32.000000
mean 116.156250 80.000000 78.968750 108.548387 0.354687
std 25.823357 21.863211 4.067847 20.117718 0.113137
min 71.000000 42.000000 70.000000 68.000000 0.250000
25% 98.000000 66.750000 77.000000 90.000000 0.250000
50% 113.500000 76.500000 80.500000 108.000000 0.350000
75% 131.750000 95.000000 82.000000 124.000000 0.500000
max 176.000000 129.000000 84.000000 158.000000 0.500000 

  <script>
    const buttonEl =
      document.querySelector('#df-617326f6-852f-40cd-8bdb-9e8c2ac9098d button.colab-df-convert');
    buttonEl.style.display =
      google.colab.kernel.accessAllowed ? 'block' : 'none';

    async function convertToInteractive(key) {
      const element = document.querySelector('#df-617326f6-852f-40cd-8bdb-9e8c2ac9098d');
      const dataTable =
        await google.colab.kernel.invokeFunction('convertToInteractive',
                                                 [key], {});
      if (!dataTable) return;

      const docLinkHtml = 'Like what you see? Visit the ' +
        '<a target="_blank" href=https://colab.research.google.com/notebooks/data_table.ipynb>data table notebook</a>'
        + ' to learn more about interactive tables.';
      element.innerHTML = '';
      dataTable['output_type'] = 'display_data';
      await google.colab.output.renderOutput(dataTable, element);
      const docLink = document.createElement('div');
      docLink.innerHTML = docLinkHtml;
      element.appendChild(docLink);
    }
  </script>
</div>
  • Location 칼럼은 문자라서 describe함수가 적용되지 않는다.
  • value_counts()
1
2
print(juice['Location'].value_counts())
print(type(juice['Location'].value_counts())) # 얘는 시리즈네?
Beach    17
Park     15
Name: Location, dtype: int64
<class 'pandas.core.series.Series'>

데이터 다뤄보기

  • 행과 열을 만져보자.
  • 열 추가(칼럼 추가)
1
2
juice['sold'] = 0 # 새로운 컬럼 추가
print(juice.head(3))
       Date Location  Lemon  Orange  Temperature  Leaflets  Price  sold
0  7/1/2016     Park     97      67           70      90.0   0.25     0
1  7/2/2016     Park     98      67           72      90.0   0.25     0
2  7/3/2016     Park    110      77           71     104.0   0.25     0

1
2
juice['sold'] = juice['Lemon'] + juice['Orange']
print(juice.head(3))
       Date Location  Lemon  Orange  Temperature  Leaflets  Price  sold
0  7/1/2016     Park     97      67           70      90.0   0.25   164
1  7/2/2016     Park     98      67           72      90.0   0.25   165
2  7/3/2016     Park    110      77           71     104.0   0.25   187
  • 퀴즈
    • 매출액 = 가격 * 판매량
    • Revenue
1
2
juice['Revenue'] = juice['Price'] * juice['sold']
print(juice.head(3))
       Date Location  Lemon  Orange  Temperature  Leaflets  Price  sold  \
0  7/1/2016     Park     97      67           70      90.0   0.25   164   
1  7/2/2016     Park     98      67           72      90.0   0.25   165   
2  7/3/2016     Park    110      77           71     104.0   0.25   187   

   Revenue  
0    41.00  
1    41.25  
2    46.75
  • 행과 열 제거
  • drop(axis=0, 1)
    • axis를 0으로 설정 시, 행(=index) 방향으로 drop() 실행
    • axis를 1로 설정 시, 열 방향으로 drop 수행함.
1
2
juice_column_drop = juice.drop('sold', axis = 1) # 열 방향, 'sold'열 하나가 통째로 삭제
print(juice_column_drop.head(3))
       Date Location  Lemon  Orange  Temperature  Leaflets  Price  Revenue
0  7/1/2016     Park     97      67           70      90.0   0.25    41.00
1  7/2/2016     Park     98      67           72      90.0   0.25    41.25
2  7/3/2016     Park    110      77           71     104.0   0.25    46.75

1
2
juice_row_drop = juice.drop(0, axis = 0) # 행 방향, 인덱스 0이 통째로 삭제 
print(juice_row_drop.head(3))
       Date Location  Lemon  Orange  Temperature  Leaflets  Price  sold  \
1  7/2/2016     Park     98      67           72      90.0   0.25   165   
2  7/3/2016     Park    110      77           71     104.0   0.25   187   
3  7/4/2016    Beach    134      99           76      98.0   0.25   233   

   Revenue  
1    41.25  
2    46.75  
3    58.25

데이터 인덱싱

1
juice[0:5]

Date Location Lemon Orange Temperature Leaflets Price sold Revenue
0 7/1/2016 Park 97 67 70 90.0 0.25 164 41.00
1 7/2/2016 Park 98 67 72 90.0 0.25 165 41.25
2 7/3/2016 Park 110 77 71 104.0 0.25 187 46.75
3 7/4/2016 Beach 134 99 76 98.0 0.25 233 58.25
4 7/5/2016 Beach 159 118 78 135.0 0.25 277 69.25 

  <script>
    const buttonEl =
      document.querySelector('#df-3ff8756e-ccea-4443-937c-96dd4ef951a2 button.colab-df-convert');
    buttonEl.style.display =
      google.colab.kernel.accessAllowed ? 'block' : 'none';

    async function convertToInteractive(key) {
      const element = document.querySelector('#df-3ff8756e-ccea-4443-937c-96dd4ef951a2');
      const dataTable =
        await google.colab.kernel.invokeFunction('convertToInteractive',
                                                 [key], {});
      if (!dataTable) return;

      const docLinkHtml = 'Like what you see? Visit the ' +
        '<a target="_blank" href=https://colab.research.google.com/notebooks/data_table.ipynb>data table notebook</a>'
        + ' to learn more about interactive tables.';
      element.innerHTML = '';
      dataTable['output_type'] = 'display_data';
      await google.colab.output.renderOutput(dataTable, element);
      const docLink = document.createElement('div');
      docLink.innerHTML = docLinkHtml;
      element.appendChild(docLink);
    }
  </script>
</div>

boolean 값을 활용한 데이터추출

1
juice['Location'] == "Beach"
0     False
1     False
2     False
3      True
4      True
5      True
6      True
7      True
8      True
9      True
10     True
11     True
12     True
13     True
14     True
15     True
16     True
17     True
18    False
19    False
20    False
21    False
22    False
23    False
24    False
25    False
26    False
27    False
28    False
29    False
30     True
31     True
Name: Location, dtype: bool

1
2
3
# Location이 Beach인 경우
# juice['Location'].value_counts()
juice[juice['Location'] == "Beach"]

Date Location Lemon Orange Temperature Leaflets Price sold Revenue
3 7/4/2016 Beach 134 99 76 98.0 0.25 233 58.25
4 7/5/2016 Beach 159 118 78 135.0 0.25 277 69.25
5 7/6/2016 Beach 103 69 82 90.0 0.25 172 43.00
6 7/6/2016 Beach 103 69 82 90.0 0.25 172 43.00
7 7/7/2016 Beach 143 101 81 135.0 0.25 244 61.00
8 NaN Beach 123 86 82 113.0 0.25 209 52.25
9 7/9/2016 Beach 134 95 80 126.0 0.25 229 57.25
10 7/10/2016 Beach 140 98 82 131.0 0.25 238 59.50
11 7/11/2016 Beach 162 120 83 135.0 0.25 282 70.50
12 7/12/2016 Beach 130 95 84 99.0 0.25 225 56.25
13 7/13/2016 Beach 109 75 77 99.0 0.25 184 46.00
14 7/14/2016 Beach 122 85 78 113.0 0.25 207 51.75
15 7/15/2016 Beach 98 62 75 108.0 0.50 160 80.00
16 7/16/2016 Beach 81 50 74 90.0 0.50 131 65.50
17 7/17/2016 Beach 115 76 77 126.0 0.50 191 95.50
30 7/30/2016 Beach 88 57 82 81.0 0.35 145 50.75
31 7/31/2016 Beach 76 47 82 68.0 0.35 123 43.05 

  <script>
    const buttonEl =
      document.querySelector('#df-f434477b-a70d-4d15-a445-7a178f78849e button.colab-df-convert');
    buttonEl.style.display =
      google.colab.kernel.accessAllowed ? 'block' : 'none';

    async function convertToInteractive(key) {
      const element = document.querySelector('#df-f434477b-a70d-4d15-a445-7a178f78849e');
      const dataTable =
        await google.colab.kernel.invokeFunction('convertToInteractive',
                                                 [key], {});
      if (!dataTable) return;

      const docLinkHtml = 'Like what you see? Visit the ' +
        '<a target="_blank" href=https://colab.research.google.com/notebooks/data_table.ipynb>data table notebook</a>'
        + ' to learn more about interactive tables.';
      element.innerHTML = '';
      dataTable['output_type'] = 'display_data';
      await google.colab.output.renderOutput(dataTable, element);
      const docLink = document.createElement('div');
      docLink.innerHTML = docLinkHtml;
      element.appendChild(docLink);
    }
  </script>
</div>

iloc vs loc

  • 차이를 확인한다.
1
juice.iloc[:, 0:2] # 전체 데이터를 가져와라 그리고 0번부터 1번(n-1) 칼럼을 가져와라

Date Location
0 7/1/2016 Park
1 7/2/2016 Park
2 7/3/2016 Park
3 7/4/2016 Beach
4 7/5/2016 Beach
5 7/6/2016 Beach
6 7/6/2016 Beach
7 7/7/2016 Beach
8 NaN Beach
9 7/9/2016 Beach
10 7/10/2016 Beach
11 7/11/2016 Beach
12 7/12/2016 Beach
13 7/13/2016 Beach
14 7/14/2016 Beach
15 7/15/2016 Beach
16 7/16/2016 Beach
17 7/17/2016 Beach
18 7/18/2016 Park
19 7/19/2016 Park
20 7/20/2016 Park
21 7/21/2016 Park
22 7/22/2016 Park
23 7/23/2016 Park
24 7/24/2016 Park
25 7/25/2016 Park
26 7/26/2016 Park
27 7/27/2016 Park
28 7/28/2016 Park
29 7/29/2016 Park
30 7/30/2016 Beach
31 7/31/2016 Beach 

  <script>
    const buttonEl =
      document.querySelector('#df-d28556da-097b-4c4b-90f3-3548273e5785 button.colab-df-convert');
    buttonEl.style.display =
      google.colab.kernel.accessAllowed ? 'block' : 'none';

    async function convertToInteractive(key) {
      const element = document.querySelector('#df-d28556da-097b-4c4b-90f3-3548273e5785');
      const dataTable =
        await google.colab.kernel.invokeFunction('convertToInteractive',
                                                 [key], {});
      if (!dataTable) return;

      const docLinkHtml = 'Like what you see? Visit the ' +
        '<a target="_blank" href=https://colab.research.google.com/notebooks/data_table.ipynb>data table notebook</a>'
        + ' to learn more about interactive tables.';
      element.innerHTML = '';
      dataTable['output_type'] = 'display_data';
      await google.colab.output.renderOutput(dataTable, element);
      const docLink = document.createElement('div');
      docLink.innerHTML = docLinkHtml;
      element.appendChild(docLink);
    }
  </script>
</div>

1
2
3
%%time

juice.iloc[0:3, 0:2] #인덱스 0~2번, 0~1번 칼럼
CPU times: user 735 µs, sys: 0 ns, total: 735 µs
Wall time: 843 µs

Date Location
0 7/1/2016 Park
1 7/2/2016 Park
2 7/3/2016 Park 

  <script>
    const buttonEl =
      document.querySelector('#df-3313e9ec-a014-4d12-9209-3fc4ffa27eab button.colab-df-convert');
    buttonEl.style.display =
      google.colab.kernel.accessAllowed ? 'block' : 'none';

    async function convertToInteractive(key) {
      const element = document.querySelector('#df-3313e9ec-a014-4d12-9209-3fc4ffa27eab');
      const dataTable =
        await google.colab.kernel.invokeFunction('convertToInteractive',
                                                 [key], {});
      if (!dataTable) return;

      const docLinkHtml = 'Like what you see? Visit the ' +
        '<a target="_blank" href=https://colab.research.google.com/notebooks/data_table.ipynb>data table notebook</a>'
        + ' to learn more about interactive tables.';
      element.innerHTML = '';
      dataTable['output_type'] = 'display_data';
      await google.colab.output.renderOutput(dataTable, element);
      const docLink = document.createElement('div');
      docLink.innerHTML = docLinkHtml;
      element.appendChild(docLink);
    }
  </script>
</div>

loc

  • 라벨 기반
1
2
3
4
%%time

juice.loc[0:2, ['Date', 'Location']] # 인덱스 라벨과 칼럼 라벨 # 인덱싱에서 n-1개념과 다르다.
# juice.loc[인덱스라벨, [칼럼 라벨]]
CPU times: user 2.58 ms, sys: 0 ns, total: 2.58 ms
Wall time: 6.81 ms

Date Location
0 7/1/2016 Park
1 7/2/2016 Park
2 7/3/2016 Park 

  <script>
    const buttonEl =
      document.querySelector('#df-69c0c626-2e18-4cf9-b0b9-4afab44368fd button.colab-df-convert');
    buttonEl.style.display =
      google.colab.kernel.accessAllowed ? 'block' : 'none';

    async function convertToInteractive(key) {
      const element = document.querySelector('#df-69c0c626-2e18-4cf9-b0b9-4afab44368fd');
      const dataTable =
        await google.colab.kernel.invokeFunction('convertToInteractive',
                                                 [key], {});
      if (!dataTable) return;

      const docLinkHtml = 'Like what you see? Visit the ' +
        '<a target="_blank" href=https://colab.research.google.com/notebooks/data_table.ipynb>data table notebook</a>'
        + ' to learn more about interactive tables.';
      element.innerHTML = '';
      dataTable['output_type'] = 'display_data';
      await google.colab.output.renderOutput(dataTable, element);
      const docLink = document.createElement('div');
      docLink.innerHTML = docLinkHtml;
      element.appendChild(docLink);
    }
  </script>
</div>
  • iloc가 속도가 빠른 것을 %%time으로 확인 할 수 있다.

컬럼명 별도 추출

  • loc만 할 수 있는 기능
1
juice.loc[juice['Leaflets'] >= 100, ['Date', 'Location']]

Date Location
2 7/3/2016 Park
4 7/5/2016 Beach
7 7/7/2016 Beach
8 NaN Beach
9 7/9/2016 Beach
10 7/10/2016 Beach
11 7/11/2016 Beach
14 7/14/2016 Beach
15 7/15/2016 Beach
17 7/17/2016 Beach
18 7/18/2016 Park
19 7/19/2016 Park
22 7/22/2016 Park
23 7/23/2016 Park
24 7/24/2016 Park
25 7/25/2016 Park
26 7/26/2016 Park 

  <script>
    const buttonEl =
      document.querySelector('#df-cce733ce-3a3e-407b-a638-9ea77dfa236f button.colab-df-convert');
    buttonEl.style.display =
      google.colab.kernel.accessAllowed ? 'block' : 'none';

    async function convertToInteractive(key) {
      const element = document.querySelector('#df-cce733ce-3a3e-407b-a638-9ea77dfa236f');
      const dataTable =
        await google.colab.kernel.invokeFunction('convertToInteractive',
                                                 [key], {});
      if (!dataTable) return;

      const docLinkHtml = 'Like what you see? Visit the ' +
        '<a target="_blank" href=https://colab.research.google.com/notebooks/data_table.ipynb>data table notebook</a>'
        + ' to learn more about interactive tables.';
      element.innerHTML = '';
      dataTable['output_type'] = 'display_data';
      await google.colab.output.renderOutput(dataTable, element);
      const docLink = document.createElement('div');
      docLink.innerHTML = docLinkHtml;
      element.appendChild(docLink);
    }
  </script>
</div>
  • iloc는 위가 안 된다.

정렬

  • sort_values()
1
juice.sort_values(by=['Revenue'], ascending = False).head() # 내림차순으로

Date Location Lemon Orange Temperature Leaflets Price sold Revenue
25 7/25/2016 Park 156 113 84 135.0 0.50 269 134.50
18 7/18/2016 Park 131 92 81 122.0 0.50 223 111.50
26 7/26/2016 Park 176 129 83 158.0 0.35 305 106.75
19 7/19/2016 Park 122 85 78 113.0 0.50 207 103.50
24 7/24/2016 Park 121 82 82 117.0 0.50 203 101.50 

  <script>
    const buttonEl =
      document.querySelector('#df-fdaed60a-bf98-46fe-a701-24e216acb756 button.colab-df-convert');
    buttonEl.style.display =
      google.colab.kernel.accessAllowed ? 'block' : 'none';

    async function convertToInteractive(key) {
      const element = document.querySelector('#df-fdaed60a-bf98-46fe-a701-24e216acb756');
      const dataTable =
        await google.colab.kernel.invokeFunction('convertToInteractive',
                                                 [key], {});
      if (!dataTable) return;

      const docLinkHtml = 'Like what you see? Visit the ' +
        '<a target="_blank" href=https://colab.research.google.com/notebooks/data_table.ipynb>data table notebook</a>'
        + ' to learn more about interactive tables.';
      element.innerHTML = '';
      dataTable['output_type'] = 'display_data';
      await google.colab.output.renderOutput(dataTable, element);
      const docLink = document.createElement('div');
      docLink.innerHTML = docLinkHtml;
      element.appendChild(docLink);
    }
  </script>
</div>

1
2
3
4
juice2 = juice.sort_values(by=['Price', 'Temperature'], ascending = [False, True]).reset_index(drop=True)
juice2
# 가격과 온도에 따라서 정렬을 해주다보니 인덱스 번호가 뒤죽박죽이 되어버렸다. 인덱스번호를 리셋 해주고 새로운 데이터셋으로 만들어준 것이다.
# 그리고 juice2라는 새로운 객체에 저장해준 것.

Date Location Lemon Orange Temperature Leaflets Price sold Revenue
0 7/20/2016 Park 71 42 70 NaN 0.50 113 56.50
1 7/16/2016 Beach 81 50 74 90.0 0.50 131 65.50
2 7/15/2016 Beach 98 62 75 108.0 0.50 160 80.00
3 7/17/2016 Beach 115 76 77 126.0 0.50 191 95.50
4 7/21/2016 Park 83 50 77 90.0 0.50 133 66.50
5 7/19/2016 Park 122 85 78 113.0 0.50 207 103.50
6 7/22/2016 Park 112 75 80 108.0 0.50 187 93.50
7 7/18/2016 Park 131 92 81 122.0 0.50 223 111.50
8 7/23/2016 Park 120 82 81 117.0 0.50 202 101.00
9 7/24/2016 Park 121 82 82 117.0 0.50 203 101.50
10 7/25/2016 Park 156 113 84 135.0 0.50 269 134.50
11 7/27/2016 Park 104 68 80 99.0 0.35 172 60.20
12 7/29/2016 Park 100 66 81 95.0 0.35 166 58.10
13 7/28/2016 Park 96 63 82 90.0 0.35 159 55.65
14 7/30/2016 Beach 88 57 82 81.0 0.35 145 50.75
15 7/31/2016 Beach 76 47 82 68.0 0.35 123 43.05
16 7/26/2016 Park 176 129 83 158.0 0.35 305 106.75
17 7/1/2016 Park 97 67 70 90.0 0.25 164 41.00
18 7/3/2016 Park 110 77 71 104.0 0.25 187 46.75
19 7/2/2016 Park 98 67 72 90.0 0.25 165 41.25
20 7/4/2016 Beach 134 99 76 98.0 0.25 233 58.25
21 7/13/2016 Beach 109 75 77 99.0 0.25 184 46.00
22 7/5/2016 Beach 159 118 78 135.0 0.25 277 69.25
23 7/14/2016 Beach 122 85 78 113.0 0.25 207 51.75
24 7/9/2016 Beach 134 95 80 126.0 0.25 229 57.25
25 7/7/2016 Beach 143 101 81 135.0 0.25 244 61.00
26 7/6/2016 Beach 103 69 82 90.0 0.25 172 43.00
27 7/6/2016 Beach 103 69 82 90.0 0.25 172 43.00
28 NaN Beach 123 86 82 113.0 0.25 209 52.25
29 7/10/2016 Beach 140 98 82 131.0 0.25 238 59.50
30 7/11/2016 Beach 162 120 83 135.0 0.25 282 70.50
31 7/12/2016 Beach 130 95 84 99.0 0.25 225 56.25 

  <script>
    const buttonEl =
      document.querySelector('#df-374e41cd-177f-4906-a3ce-e839f106f7a7 button.colab-df-convert');
    buttonEl.style.display =
      google.colab.kernel.accessAllowed ? 'block' : 'none';

    async function convertToInteractive(key) {
      const element = document.querySelector('#df-374e41cd-177f-4906-a3ce-e839f106f7a7');
      const dataTable =
        await google.colab.kernel.invokeFunction('convertToInteractive',
                                                 [key], {});
      if (!dataTable) return;

      const docLinkHtml = 'Like what you see? Visit the ' +
        '<a target="_blank" href=https://colab.research.google.com/notebooks/data_table.ipynb>data table notebook</a>'
        + ' to learn more about interactive tables.';
      element.innerHTML = '';
      dataTable['output_type'] = 'display_data';
      await google.colab.output.renderOutput(dataTable, element);
      const docLink = document.createElement('div');
      docLink.innerHTML = docLinkHtml;
      element.appendChild(docLink);
    }
  </script>
</div>

Groupby()

  • 피벗 테이블을 만드는 것과 똑같음
  • 요약하려고
1
juice.groupby(by = 'Location').count()

Date Lemon Orange Temperature Leaflets Price sold Revenue
Location
Beach 16 17 17 17 17 17 17 17
Park 15 15 15 15 14 15 15 15 

  <script>
    const buttonEl =
      document.querySelector('#df-9d46cfb2-4ae7-44c5-a9de-d88d60478500 button.colab-df-convert');
    buttonEl.style.display =
      google.colab.kernel.accessAllowed ? 'block' : 'none';

    async function convertToInteractive(key) {
      const element = document.querySelector('#df-9d46cfb2-4ae7-44c5-a9de-d88d60478500');
      const dataTable =
        await google.colab.kernel.invokeFunction('convertToInteractive',
                                                 [key], {});
      if (!dataTable) return;

      const docLinkHtml = 'Like what you see? Visit the ' +
        '<a target="_blank" href=https://colab.research.google.com/notebooks/data_table.ipynb>data table notebook</a>'
        + ' to learn more about interactive tables.';
      element.innerHTML = '';
      dataTable['output_type'] = 'display_data';
      await google.colab.output.renderOutput(dataTable, element);
      const docLink = document.createElement('div');
      docLink.innerHTML = docLinkHtml;
      element.appendChild(docLink);
    }
  </script>
</div>

1
2
import numpy as np
juice.groupby(['Location'])[['Revenue', 'Lemon']].agg([max, min, sum, np.mean])

Revenue Lemon
max min sum mean max min sum mean
Location
Beach 95.5 43.0 1002.8 58.988235 162 76 2020 118.823529
Park 134.5 41.0 1178.2 78.546667 176 71 1697 113.133333 

  <script>
    const buttonEl =
      document.querySelector('#df-d175d024-02cf-46a7-a0ef-d792b2802f03 button.colab-df-convert');
    buttonEl.style.display =
      google.colab.kernel.accessAllowed ? 'block' : 'none';

    async function convertToInteractive(key) {
      const element = document.querySelector('#df-d175d024-02cf-46a7-a0ef-d792b2802f03');
      const dataTable =
        await google.colab.kernel.invokeFunction('convertToInteractive',
                                                 [key], {});
      if (!dataTable) return;

      const docLinkHtml = 'Like what you see? Visit the ' +
        '<a target="_blank" href=https://colab.research.google.com/notebooks/data_table.ipynb>data table notebook</a>'
        + ' to learn more about interactive tables.';
      element.innerHTML = '';
      dataTable['output_type'] = 'display_data';
      await google.colab.output.renderOutput(dataTable, element);
      const docLink = document.createElement('div');
      docLink.innerHTML = docLinkHtml;
      element.appendChild(docLink);
    }
  </script>
</div>
Follow

Links

Categories

Recents

Archives

Subscribe for updates

You need to set client_id and slot_id to show this AD unit. Please set it in _config.yml.

follow.it

×