In [1]:

import pandas as pd
import numpy as np
import os

1. Pandas 데이터 프레임 기초¶

1.1 데이터 프레임 읽어 오기:¶

In [2]:

os.chdir(r"C:\Users\Gram\Desktop\myPyCode\01 Python 분석 기초 - 실습\data")

In [3]:

df = pd.read_csv('data_studentlist_en.csv', header='infer',encoding = 'latin1')

In [4]:

df.shape

Out[4]:

(17, 8)

In [5]:

df.size

Out[5]:

In [6]:

df.ndim

Out[6]:

In [7]:

type(df)

Out[7]:

pandas.core.frame.DataFrame

1.2 데이터 프레임 요약해보기:¶

In [8]:

df.info()         #데이터프레임의 구조 보여줌

<class 'pandas.core.frame.DataFrame'>
RangeIndex: 17 entries, 0 to 16
Data columns (total 8 columns):
name         17 non-null object
gender       17 non-null object
age          17 non-null int64
grade        17 non-null int64
absence      17 non-null object
bloodtype    17 non-null object
height       17 non-null float64
weight       17 non-null float64
dtypes: float64(2), int64(2), object(4)
memory usage: 1.2+ KB

In [9]:

df.describe()     #기술통계 요약 보여줌

Out[9]:

	age	grade	height	weight
count	17.000000	17.000000	17.000000	17.000000
mean	22.352941	2.294118	170.035294	61.064706
std	1.169464	1.104802	7.849756	11.851685
min	20.000000	1.000000	155.200000	45.200000
25%	22.000000	1.000000	165.300000	53.000000
50%	22.000000	2.000000	169.200000	61.300000
75%	23.000000	3.000000	176.100000	68.200000
max	24.000000	4.000000	182.100000	85.700000

1.3 데이터 프레임의 상단과 하단보기:¶

In [10]:

df.head(3)   #열이름 제외하고 상단에서 3행

Out[10]:

	name	gender	age	grade	absence	bloodtype	height	weight
0	Jared Diamond	M	23	3	Y	O	165.3	68.2
1	Sarah O'Donnel	F	22	2	N	AB	170.1	53.0
2	Brian Martin	M	24	4	N	B	175.0	80.1

In [11]:

df.tail(3)   #열이름 제외하고 하단에서 3행

Out[11]:

	name	gender	age	grade	absence	bloodtype	height	weight
14	Linda Carter	F	22	2	N	B	158.2	45.2
15	Richard Swayze	M	24	4	Y	B	168.6	70.2
16	Andrew Daley	M	21	1	N	A	169.2	62.2

1.4 데이터 프레임의 열 이름(헤더 이름) 알아보고 변경 후 header 변수에 저장:¶

In [12]:

df.columns    #열 이름(헤더 이름)보기

Out[12]:

Index(['name', 'gender', 'age', 'grade', 'absence', 'bloodtype', 'height',
       'weight'],
      dtype='object')

In [13]:

df.columns = ['NAME', 'GENDER' , 'AGE', 'GRADE', 'ABSENCE', 'BLOODTYPE', 'HEIGHT', 'WEIGHT']
df.head(3)     #열 이름(헤더 이름) 변경   
               #새로운 열 이름 리스트를 대입함

Out[13]:

	NAME	GENDER	AGE	GRADE	ABSENCE	BLOODTYPE	HEIGHT	WEIGHT
0	Jared Diamond	M	23	3	Y	O	165.3	68.2
1	Sarah O'Donnel	F	22	2	N	AB	170.1	53.0
2	Brian Martin	M	24	4	N	B	175.0	80.1

In [14]:

header = df.columns   #열로 헤더이름 가져오기

In [15]:

df.index

Out[15]:

RangeIndex(start=0, stop=17, step=1)

2. Pandas 데이터 프레임 슬라이싱¶

In [16]:

df.NAME           # NAME 열 뽑기

Out[16]:

0        Jared Diamond
1       Sarah O'Donnel
2         Brian Martin
3         David Hassel
4      Clara Rodriquez
5     Jennifer Lorentz
6          Susan Clark
7      Margareth Jones
8         John Bertsch
9       Jake Timmerman
10       Joshua Connor
11        John Matsuda
12        Eddy Johnson
13    Rebecah Anderson
14        Linda Carter
15      Richard Swayze
16        Andrew Daley
Name: NAME, dtype: object

In [17]:

type(df.NAME)     # 출력된 것은 시리즈이다.

Out[17]:

pandas.core.series.Series

df[" "]해도 이름 가져올 수 있음: 데이터프레임에서 시리즈 하나 가져옴
df[[""]]: 칼럼이 하나있는 데이터프레임 = > 칼럼을 하나 가져오되 시리즈로 가져오라
이름여러개 넣으면 데이터프레임일부가져옴

In [18]:

df['NAME']         #데이터프레임에서 시리즈 하나 가져옴

Out[18]:

0        Jared Diamond
1       Sarah O'Donnel
2         Brian Martin
3         David Hassel
4      Clara Rodriquez
5     Jennifer Lorentz
6          Susan Clark
7      Margareth Jones
8         John Bertsch
9       Jake Timmerman
10       Joshua Connor
11        John Matsuda
12        Eddy Johnson
13    Rebecah Anderson
14        Linda Carter
15      Richard Swayze
16        Andrew Daley
Name: NAME, dtype: object

In [19]:

df.loc[:,'NAME']   #콤마 왼쪽은 행, 오른쪽은 열 

Out[19]:

0        Jared Diamond
1       Sarah O'Donnel
2         Brian Martin
3         David Hassel
4      Clara Rodriquez
5     Jennifer Lorentz
6          Susan Clark
7      Margareth Jones
8         John Bertsch
9       Jake Timmerman
10       Joshua Connor
11        John Matsuda
12        Eddy Johnson
13    Rebecah Anderson
14        Linda Carter
15      Richard Swayze
16        Andrew Daley
Name: NAME, dtype: object

In [20]:

df.loc[:,['NAME','GENDER']]  #칼럼 'NAME'과 'GENDER' 뽑음
                             #콤마 왼쪽은 행, 오른쪽은 열   

Out[20]:

	NAME	GENDER
0	Jared Diamond	M
1	Sarah O'Donnel	F
2	Brian Martin	M
3	David Hassel	M
4	Clara Rodriquez	F
5	Jennifer Lorentz	F
6	Susan Clark	F
7	Margareth Jones	F
8	John Bertsch	M
9	Jake Timmerman	M
10	Joshua Connor	M
11	John Matsuda	M
12	Eddy Johnson	M
13	Rebecah Anderson	F
14	Linda Carter	F
15	Richard Swayze	M
16	Andrew Daley	M

In [21]:

df.loc[2]   #2번째 행을 보여줌

Out[21]:

NAME         Brian Martin
GENDER                  M
AGE                    24
GRADE                   4
ABSENCE                 N
BLOODTYPE               B
HEIGHT                175
WEIGHT               80.1
Name: 2, dtype: object

In [22]:

df.loc[2:4] #2~4번째 행을 보여줌

Out[22]:

	NAME	GENDER	AGE	GRADE	ABSENCE	BLOODTYPE	HEIGHT	WEIGHT
2	Brian Martin	M	24	4	N	B	175.0	80.1
3	David Hassel	M	23	3	N	AB	182.1	85.7
4	Clara Rodriquez	F	20	1	Y	A	168.0	49.5

In [23]:

df.drop(columns=['NAME','GENDER'])   #칼럼 'NAME'과 'GENDER'를 제외한 나머지
                                     #영구적으로 제외하고 싶으면 inplace=True해서 다시 대입
                                 

Out[23]:

	AGE	GRADE	ABSENCE	BLOODTYPE	HEIGHT	WEIGHT
0	23	3	Y	O	165.3	68.2
1	22	2	N	AB	170.1	53.0
2	24	4	N	B	175.0	80.1
3	23	3	N	AB	182.1	85.7
4	20	1	Y	A	168.0	49.5
5	21	2	N	O	162.0	52.0
6	22	1	N	O	155.2	45.3
7	23	1	N	A	176.9	55.0
8	23	3	N	B	178.5	64.2
9	22	2	N	B	176.1	61.3
10	24	4	Y	O	167.1	62.0
11	22	2	N	AB	180.0	75.8
12	21	1	N	A	162.2	55.3
13	23	3	N	O	176.1	53.1
14	22	2	N	B	158.2	45.2
15	24	4	Y	B	168.6	70.2
16	21	1	N	A	169.2	62.2

In [24]:

df.loc[:, (header!='NAME') & (header!='GENDER')]  #칼럼 'NAME'과 'GENDER'를 제외한 나머지 
                                                  #영구적으로 제외되지 않음

Out[24]:

	AGE	GRADE	ABSENCE	BLOODTYPE	HEIGHT	WEIGHT
0	23	3	Y	O	165.3	68.2
1	22	2	N	AB	170.1	53.0
2	24	4	N	B	175.0	80.1
3	23	3	N	AB	182.1	85.7
4	20	1	Y	A	168.0	49.5
5	21	2	N	O	162.0	52.0
6	22	1	N	O	155.2	45.3
7	23	1	N	A	176.9	55.0
8	23	3	N	B	178.5	64.2
9	22	2	N	B	176.1	61.3
10	24	4	Y	O	167.1	62.0
11	22	2	N	AB	180.0	75.8
12	21	1	N	A	162.2	55.3
13	23	3	N	O	176.1	53.1
14	22	2	N	B	158.2	45.2
15	24	4	Y	B	168.6	70.2
16	21	1	N	A	169.2	62.2

In [25]:

df.loc[:,(header =='NAME') | (header == 'GENDER')]  #칼럼 'NAME'과 'GENDER'를 뽑아서 보여줌

Out[25]:

	NAME	GENDER
0	Jared Diamond	M
1	Sarah O'Donnel	F
2	Brian Martin	M
3	David Hassel	M
4	Clara Rodriquez	F
5	Jennifer Lorentz	F
6	Susan Clark	F
7	Margareth Jones	F
8	John Bertsch	M
9	Jake Timmerman	M
10	Joshua Connor	M
11	John Matsuda	M
12	Eddy Johnson	M
13	Rebecah Anderson	F
14	Linda Carter	F
15	Richard Swayze	M
16	Andrew Daley	M

3. Pandas 데이터 프레임 조건부 슬라이싱:¶

특정 행을 가져오고 싶을 때, 조건은 칼럼에 걸기

In [26]:

df[df.GENDER == 'M']     #성별이 M인 경우

Out[26]:

	NAME	GENDER	AGE	GRADE	ABSENCE	BLOODTYPE	HEIGHT	WEIGHT
0	Jared Diamond	M	23	3	Y	O	165.3	68.2
2	Brian Martin	M	24	4	N	B	175.0	80.1
3	David Hassel	M	23	3	N	AB	182.1	85.7
8	John Bertsch	M	23	3	N	B	178.5	64.2
9	Jake Timmerman	M	22	2	N	B	176.1	61.3
10	Joshua Connor	M	24	4	Y	O	167.1	62.0
11	John Matsuda	M	22	2	N	AB	180.0	75.8
12	Eddy Johnson	M	21	1	N	A	162.2	55.3
15	Richard Swayze	M	24	4	Y	B	168.6	70.2
16	Andrew Daley	M	21	1	N	A	169.2	62.2

In [27]:

df[-(df.GENDER == 'M')]  #성별이 M이 아닌 경우

Out[27]:

	NAME	GENDER	AGE	GRADE	ABSENCE	BLOODTYPE	HEIGHT	WEIGHT
1	Sarah O'Donnel	F	22	2	N	AB	170.1	53.0
4	Clara Rodriquez	F	20	1	Y	A	168.0	49.5
5	Jennifer Lorentz	F	21	2	N	O	162.0	52.0
6	Susan Clark	F	22	1	N	O	155.2	45.3
7	Margareth Jones	F	23	1	N	A	176.9	55.0
13	Rebecah Anderson	F	23	3	N	O	176.1	53.1
14	Linda Carter	F	22	2	N	B	158.2	45.2

In [28]:

df[df.HEIGHT > 170]      #신장이 170이상인 경우

Out[28]:

	NAME	GENDER	AGE	GRADE	ABSENCE	BLOODTYPE	HEIGHT	WEIGHT
1	Sarah O'Donnel	F	22	2	N	AB	170.1	53.0
2	Brian Martin	M	24	4	N	B	175.0	80.1
3	David Hassel	M	23	3	N	AB	182.1	85.7
7	Margareth Jones	F	23	1	N	A	176.9	55.0
8	John Bertsch	M	23	3	N	B	178.5	64.2
9	Jake Timmerman	M	22	2	N	B	176.1	61.3
11	John Matsuda	M	22	2	N	AB	180.0	75.8
13	Rebecah Anderson	F	23	3	N	O	176.1	53.1

In [29]:

df[ (df.WEIGHT > 70) & (df.WEIGHT < 80)]  #AND 조건

Out[29]:

	NAME	GENDER	AGE	GRADE	ABSENCE	BLOODTYPE	HEIGHT	WEIGHT
11	John Matsuda	M	22	2	N	AB	180.0	75.8
15	Richard Swayze	M	24	4	Y	B	168.6	70.2

In [30]:

df[ (df.GRADE == 1) | (df.GRADE == 4 )] #OR 조건

Out[30]:

	NAME	GENDER	AGE	GRADE	ABSENCE	BLOODTYPE	HEIGHT	WEIGHT
2	Brian Martin	M	24	4	N	B	175.0	80.1
4	Clara Rodriquez	F	20	1	Y	A	168.0	49.5
6	Susan Clark	F	22	1	N	O	155.2	45.3
7	Margareth Jones	F	23	1	N	A	176.9	55.0
10	Joshua Connor	M	24	4	Y	O	167.1	62.0
12	Eddy Johnson	M	21	1	N	A	162.2	55.3
15	Richard Swayze	M	24	4	Y	B	168.6	70.2
16	Andrew Daley	M	21	1	N	A	169.2	62.2

In [31]:

df[ (df.GENDER == 'M') & ((df.HEIGHT < 160) | (df.HEIGHT > 180))] #AND와 OR 조건

Out[31]:

	NAME	GENDER	AGE	GRADE	ABSENCE	BLOODTYPE	HEIGHT	WEIGHT
3	David Hassel	M	23	3	N	AB	182.1	85.7

새로운 컬럼 추가:¶

In [32]:

df['BMI'] = 10000*df['WEIGHT']/df['HEIGHT']**2

In [33]:

df.head(5)

Out[33]:

	NAME	GENDER	AGE	GRADE	ABSENCE	BLOODTYPE	HEIGHT	WEIGHT	BMI
0	Jared Diamond	M	23	3	Y	O	165.3	68.2	24.959660
1	Sarah O'Donnel	F	22	2	N	AB	170.1	53.0	18.317544
2	Brian Martin	M	24	4	N	B	175.0	80.1	26.155102
3	David Hassel	M	23	3	N	AB	182.1	85.7	25.844071
4	Clara Rodriquez	F	20	1	Y	A	168.0	49.5	17.538265

In [34]:

df.drop('BMI',axis=1)
df.head(5)

Out[34]:

	NAME	GENDER	AGE	GRADE	ABSENCE	BLOODTYPE	HEIGHT	WEIGHT	BMI
0	Jared Diamond	M	23	3	Y	O	165.3	68.2	24.959660
1	Sarah O'Donnel	F	22	2	N	AB	170.1	53.0	18.317544
2	Brian Martin	M	24	4	N	B	175.0	80.1	26.155102
3	David Hassel	M	23	3	N	AB	182.1	85.7	25.844071
4	Clara Rodriquez	F	20	1	Y	A	168.0	49.5	17.538265

In [35]:

df.drop('BMI',axis=1,inplace=True)
df.head(5)

Out[35]:

	NAME	GENDER	AGE	GRADE	ABSENCE	BLOODTYPE	HEIGHT	WEIGHT
0	Jared Diamond	M	23	3	Y	O	165.3	68.2
1	Sarah O'Donnel	F	22	2	N	AB	170.1	53.0
2	Brian Martin	M	24	4	N	B	175.0	80.1
3	David Hassel	M	23	3	N	AB	182.1	85.7
4	Clara Rodriquez	F	20	1	Y	A	168.0	49.5

파일로 출력:¶

In [36]:

df2 = df.drop(columns=['GRADE','ABSENCE'])
df2.to_csv('data_mine.csv',index=False)

In [37]:

df3 = pd.read_csv('data_mine.csv',encoding='latin1',header='infer')

In [38]:

df3.head(3)

Out[38]:

	NAME	GENDER	AGE	BLOODTYPE	HEIGHT	WEIGHT
0	Jared Diamond	M	23	O	165.3	68.2
1	Sarah O'Donnel	F	22	AB	170.1	53.0
2	Brian Martin	M	24	B	175.0	80.1

엑셀 파일로 읽어오고 저장하기:¶

In [39]:

dfx = pd.read_excel('data_studentlist_en.xlsx', sheet_name='Sheet1')

In [40]:

dfx.head(5)

Out[40]:

	name	gender	age	grade	absence	bloodtype	height	weight
0	Jared Diamond	M	23	3	Y	O	165.3	68.2
1	Sarah O'Donnel	F	22	2	N	AB	170.1	53.0
2	Brian Martin	M	24	4	N	B	175.0	80.1
3	David Hassel	M	23	3	N	AB	182.1	85.7
4	Clara Rodriquez	F	20	1	Y	A	168.0	49.5

In [41]:

dfx.to_excel('data_studentlist_en2.xlsx',sheet_name='NewSheet',index=False)

4. Pandas 데이터 프레임 신규 생성: 딕셔너리 사용¶

자료를 불러오는게 아니라 내가 직접 생성하고 싶을 때

In [42]:

data = { 'NAME' : ['Jake', 'Jennifer', 'Paul', 'Andrew'], 'AGE': [24,21,25,19], 'GENDER':['M','F','M','M']}
df4 = pd.DataFrame(data)
df4                       #리스트가 칼럼의 값으로 들어감
                          #칼럼의 길이(리스트의 길이)가 같아야 오류안남

Out[42]:

	NAME	AGE	GENDER
0	Jake	24	M
1	Jennifer	21	F
2	Paul	25	M
3	Andrew	19	M

In [43]:

df5 = pd.DataFrame(np.random.rand(10,5), columns=['A','B','C','D','E'])  #랜덤 데이터프레임
df5.head(3)     #numpy배열로 데이터프레임 객체만듬
                #10개의 행과 5개의 열이므로, 칼럼도 5개 잘 맞음

Out[43]:

	A	B	C	D	E
0	0.390709	0.019426	0.114121	0.797595	0.051237
1	0.489058	0.025332	0.429168	0.987400	0.408086
2	0.855244	0.558176	0.436755	0.738553	0.914112

5.SQL¶

5.1 좌우 완벽하게 일치하는 경우: merge¶

In [44]:

df_left = df.loc[:,['NAME','AGE','GENDER','GRADE','ABSENCE']]
df_left

Out[44]:

	NAME	AGE	GENDER	GRADE	ABSENCE
0	Jared Diamond	23	M	3	Y
1	Sarah O'Donnel	22	F	2	N
2	Brian Martin	24	M	4	N
3	David Hassel	23	M	3	N
4	Clara Rodriquez	20	F	1	Y
5	Jennifer Lorentz	21	F	2	N
6	Susan Clark	22	F	1	N
7	Margareth Jones	23	F	1	N
8	John Bertsch	23	M	3	N
9	Jake Timmerman	22	M	2	N
10	Joshua Connor	24	M	4	Y
11	John Matsuda	22	M	2	N
12	Eddy Johnson	21	M	1	N
13	Rebecah Anderson	23	F	3	N
14	Linda Carter	22	F	2	N
15	Richard Swayze	24	M	4	Y
16	Andrew Daley	21	M	1	N

In [45]:

df_right = df.loc[:,['NAME','BLOODTYPE','WEIGHT','HEIGHT']]
df_right

Out[45]:

	NAME	BLOODTYPE	WEIGHT	HEIGHT
0	Jared Diamond	O	68.2	165.3
1	Sarah O'Donnel	AB	53.0	170.1
2	Brian Martin	B	80.1	175.0
3	David Hassel	AB	85.7	182.1
4	Clara Rodriquez	A	49.5	168.0
5	Jennifer Lorentz	O	52.0	162.0
6	Susan Clark	O	45.3	155.2
7	Margareth Jones	A	55.0	176.9
8	John Bertsch	B	64.2	178.5
9	Jake Timmerman	B	61.3	176.1
10	Joshua Connor	O	62.0	167.1
11	John Matsuda	AB	75.8	180.0
12	Eddy Johnson	A	55.3	162.2
13	Rebecah Anderson	O	53.1	176.1
14	Linda Carter	B	45.2	158.2
15	Richard Swayze	B	70.2	168.6
16	Andrew Daley	A	62.2	169.2

In [46]:

pd.merge(df_left,df_right,on='NAME')

Out[46]:

	NAME	AGE	GENDER	GRADE	ABSENCE	BLOODTYPE	WEIGHT	HEIGHT
0	Jared Diamond	23	M	3	Y	O	68.2	165.3
1	Sarah O'Donnel	22	F	2	N	AB	53.0	170.1
2	Brian Martin	24	M	4	N	B	80.1	175.0
3	David Hassel	23	M	3	N	AB	85.7	182.1
4	Clara Rodriquez	20	F	1	Y	A	49.5	168.0
5	Jennifer Lorentz	21	F	2	N	O	52.0	162.0
6	Susan Clark	22	F	1	N	O	45.3	155.2
7	Margareth Jones	23	F	1	N	A	55.0	176.9
8	John Bertsch	23	M	3	N	B	64.2	178.5
9	Jake Timmerman	22	M	2	N	B	61.3	176.1
10	Joshua Connor	24	M	4	Y	O	62.0	167.1
11	John Matsuda	22	M	2	N	AB	75.8	180.0
12	Eddy Johnson	21	M	1	N	A	55.3	162.2
13	Rebecah Anderson	23	F	3	N	O	53.1	176.1
14	Linda Carter	22	F	2	N	B	45.2	158.2
15	Richard Swayze	24	M	4	Y	B	70.2	168.6
16	Andrew Daley	21	M	1	N	A	62.2	169.2

5.2 좌우 완벽하게 일치하지 않는 경우: join¶

In [47]:

df_left_small = df_left.loc[:10,]
df_left_small

Out[47]:

	NAME	AGE	GENDER	GRADE	ABSENCE
0	Jared Diamond	23	M	3	Y
1	Sarah O'Donnel	22	F	2	N
2	Brian Martin	24	M	4	N
3	David Hassel	23	M	3	N
4	Clara Rodriquez	20	F	1	Y
5	Jennifer Lorentz	21	F	2	N
6	Susan Clark	22	F	1	N
7	Margareth Jones	23	F	1	N
8	John Bertsch	23	M	3	N
9	Jake Timmerman	22	M	2	N
10	Joshua Connor	24	M	4	Y

In [48]:

df_right_small = df_right.loc[7:,]
df_right_small

Out[48]:

	NAME	BLOODTYPE	WEIGHT	HEIGHT
7	Margareth Jones	A	55.0	176.9
8	John Bertsch	B	64.2	178.5
9	Jake Timmerman	B	61.3	176.1
10	Joshua Connor	O	62.0	167.1
11	John Matsuda	AB	75.8	180.0
12	Eddy Johnson	A	55.3	162.2
13	Rebecah Anderson	O	53.1	176.1
14	Linda Carter	B	45.2	158.2
15	Richard Swayze	B	70.2	168.6
16	Andrew Daley	A	62.2	169.2

1. 양쪽 온전하게 일치하는 경우 데이터 프레임 결합하기 (Inner Join): 교집합

In [49]:

pd.merge(df_left_small,df_right_small,left_on='NAME', right_on = 'NAME', how='inner')

Out[49]:

	NAME	AGE	GENDER	GRADE	ABSENCE	BLOODTYPE	WEIGHT	HEIGHT
0	Margareth Jones	23	F	1	N	A	55.0	176.9
1	John Bertsch	23	M	3	N	B	64.2	178.5
2	Jake Timmerman	22	M	2	N	B	61.3	176.1
3	Joshua Connor	24	M	4	Y	O	62.0	167.1

2. 왼쪽 기준으로 데이터 프레임 결합하기 (Left Join): 왼쪽에서 다 선택, 오른쪽은 공통값만

In [50]:

pd.merge(df_left_small,df_right_small,left_on='NAME', right_on = 'NAME', how='left')

Out[50]:

	NAME	AGE	GENDER	GRADE	ABSENCE	BLOODTYPE	WEIGHT	HEIGHT
0	Jared Diamond	23	M	3	Y	NaN	NaN	NaN
1	Sarah O'Donnel	22	F	2	N	NaN	NaN	NaN
2	Brian Martin	24	M	4	N	NaN	NaN	NaN
3	David Hassel	23	M	3	N	NaN	NaN	NaN
4	Clara Rodriquez	20	F	1	Y	NaN	NaN	NaN
5	Jennifer Lorentz	21	F	2	N	NaN	NaN	NaN
6	Susan Clark	22	F	1	N	NaN	NaN	NaN
7	Margareth Jones	23	F	1	N	A	55.0	176.9
8	John Bertsch	23	M	3	N	B	64.2	178.5
9	Jake Timmerman	22	M	2	N	B	61.3	176.1
10	Joshua Connor	24	M	4	Y	O	62.0	167.1

3. 오른쪽 기준으로 데이터 프레임 결합하기 (Right Join): 오른쪽에서 다 선택, 왼쪽은 공통값만

In [51]:

pd.merge(df_left_small,df_right_small,left_on='NAME', right_on = 'NAME', how='right')

Out[51]:

	NAME	AGE	GENDER	GRADE	ABSENCE	BLOODTYPE	WEIGHT	HEIGHT
0	Margareth Jones	23.0	F	1.0	N	A	55.0	176.9
1	John Bertsch	23.0	M	3.0	N	B	64.2	178.5
2	Jake Timmerman	22.0	M	2.0	N	B	61.3	176.1
3	Joshua Connor	24.0	M	4.0	Y	O	62.0	167.1
4	John Matsuda	NaN	NaN	NaN	NaN	AB	75.8	180.0
5	Eddy Johnson	NaN	NaN	NaN	NaN	A	55.3	162.2
6	Rebecah Anderson	NaN	NaN	NaN	NaN	O	53.1	176.1
7	Linda Carter	NaN	NaN	NaN	NaN	B	45.2	158.2
8	Richard Swayze	NaN	NaN	NaN	NaN	B	70.2	168.6
9	Andrew Daley	NaN	NaN	NaN	NaN	A	62.2	169.2

4. 좌우 결측치 상관없이 데이터 프레임 결합하기 (Outer Join):

In [52]:

pd.merge(df_left_small,df_right_small,left_on='NAME', right_on = 'NAME', how='outer')

Out[52]:

	NAME	AGE	GENDER	GRADE	ABSENCE	BLOODTYPE	WEIGHT	HEIGHT
0	Jared Diamond	23.0	M	3.0	Y	NaN	NaN	NaN
1	Sarah O'Donnel	22.0	F	2.0	N	NaN	NaN	NaN
2	Brian Martin	24.0	M	4.0	N	NaN	NaN	NaN
3	David Hassel	23.0	M	3.0	N	NaN	NaN	NaN
4	Clara Rodriquez	20.0	F	1.0	Y	NaN	NaN	NaN
5	Jennifer Lorentz	21.0	F	2.0	N	NaN	NaN	NaN
6	Susan Clark	22.0	F	1.0	N	NaN	NaN	NaN
7	Margareth Jones	23.0	F	1.0	N	A	55.0	176.9
8	John Bertsch	23.0	M	3.0	N	B	64.2	178.5
9	Jake Timmerman	22.0	M	2.0	N	B	61.3	176.1
10	Joshua Connor	24.0	M	4.0	Y	O	62.0	167.1
11	John Matsuda	NaN	NaN	NaN	NaN	AB	75.8	180.0
12	Eddy Johnson	NaN	NaN	NaN	NaN	A	55.3	162.2
13	Rebecah Anderson	NaN	NaN	NaN	NaN	O	53.1	176.1
14	Linda Carter	NaN	NaN	NaN	NaN	B	45.2	158.2
15	Richard Swayze	NaN	NaN	NaN	NaN	B	70.2	168.6
16	Andrew Daley	NaN	NaN	NaN	NaN	A	62.2	169.2

6. concat: 데이터 프레임 단순 결합하기¶

6.1 세로 방향으로 결합¶

In [53]:

# columns로 매칭.  
pd.concat([df_left_small,df_right_small],sort=True,ignore_index=True)       

Out[53]:

	ABSENCE	AGE	BLOODTYPE	GENDER	GRADE	HEIGHT	NAME	WEIGHT
0	Y	23.0	NaN	M	3.0	NaN	Jared Diamond	NaN
1	N	22.0	NaN	F	2.0	NaN	Sarah O'Donnel	NaN
2	N	24.0	NaN	M	4.0	NaN	Brian Martin	NaN
3	N	23.0	NaN	M	3.0	NaN	David Hassel	NaN
4	Y	20.0	NaN	F	1.0	NaN	Clara Rodriquez	NaN
5	N	21.0	NaN	F	2.0	NaN	Jennifer Lorentz	NaN
6	N	22.0	NaN	F	1.0	NaN	Susan Clark	NaN
7	N	23.0	NaN	F	1.0	NaN	Margareth Jones	NaN
8	N	23.0	NaN	M	3.0	NaN	John Bertsch	NaN
9	N	22.0	NaN	M	2.0	NaN	Jake Timmerman	NaN
10	Y	24.0	NaN	M	4.0	NaN	Joshua Connor	NaN
11	NaN	NaN	A	NaN	NaN	176.9	Margareth Jones	55.0
12	NaN	NaN	B	NaN	NaN	178.5	John Bertsch	64.2
13	NaN	NaN	B	NaN	NaN	176.1	Jake Timmerman	61.3
14	NaN	NaN	O	NaN	NaN	167.1	Joshua Connor	62.0
15	NaN	NaN	AB	NaN	NaN	180.0	John Matsuda	75.8
16	NaN	NaN	A	NaN	NaN	162.2	Eddy Johnson	55.3
17	NaN	NaN	O	NaN	NaN	176.1	Rebecah Anderson	53.1
18	NaN	NaN	B	NaN	NaN	158.2	Linda Carter	45.2
19	NaN	NaN	B	NaN	NaN	168.6	Richard Swayze	70.2
20	NaN	NaN	A	NaN	NaN	169.2	Andrew Daley	62.2

6.2 가로 방향으로 결합¶

In [54]:

# index로 매칭.
pd.concat([df_left_small,df_right_small],axis=1,sort=True,ignore_index=False)         

Out[54]:

	NAME	AGE	GENDER	GRADE	ABSENCE	NAME	BLOODTYPE	WEIGHT	HEIGHT
0	Jared Diamond	23.0	M	3.0	Y	NaN	NaN	NaN	NaN
1	Sarah O'Donnel	22.0	F	2.0	N	NaN	NaN	NaN	NaN
2	Brian Martin	24.0	M	4.0	N	NaN	NaN	NaN	NaN
3	David Hassel	23.0	M	3.0	N	NaN	NaN	NaN	NaN
4	Clara Rodriquez	20.0	F	1.0	Y	NaN	NaN	NaN	NaN
5	Jennifer Lorentz	21.0	F	2.0	N	NaN	NaN	NaN	NaN
6	Susan Clark	22.0	F	1.0	N	NaN	NaN	NaN	NaN
7	Margareth Jones	23.0	F	1.0	N	Margareth Jones	A	55.0	176.9
8	John Bertsch	23.0	M	3.0	N	John Bertsch	B	64.2	178.5
9	Jake Timmerman	22.0	M	2.0	N	Jake Timmerman	B	61.3	176.1
10	Joshua Connor	24.0	M	4.0	Y	Joshua Connor	O	62.0	167.1
11	NaN	NaN	NaN	NaN	NaN	John Matsuda	AB	75.8	180.0
12	NaN	NaN	NaN	NaN	NaN	Eddy Johnson	A	55.3	162.2
13	NaN	NaN	NaN	NaN	NaN	Rebecah Anderson	O	53.1	176.1
14	NaN	NaN	NaN	NaN	NaN	Linda Carter	B	45.2	158.2
15	NaN	NaN	NaN	NaN	NaN	Richard Swayze	B	70.2	168.6
16	NaN	NaN	NaN	NaN	NaN	Andrew Daley	A	62.2	169.2

In [ ]: