새싹/TIL

[핀테커스] 230913 pandas merge & metplot 시각화

jykim23 2023. 9. 13. 18:12

import pandas as pd

customer_master = pd.read_csv('https://raw.githubusercontent.com/jin0choi1216/dataset/main/table_data_anlaysis/shopping_mall/customer_master.csv')
item_master = pd.read_csv('https://raw.githubusercontent.com/jin0choi1216/dataset/main/table_data_anlaysis/shopping_mall/item_master.csv')
transaction_1 =pd.read_csv('https://raw.githubusercontent.com/jin0choi1216/dataset/main/table_data_anlaysis/shopping_mall/transaction_1.csv')
transaction_2 =pd.read_csv('https://raw.githubusercontent.com/jin0choi1216/dataset/main/table_data_anlaysis/shopping_mall/transaction_2.csv')
transaction_detail_1 =pd.read_csv('https://raw.githubusercontent.com/jin0choi1216/dataset/main/table_data_anlaysis/shopping_mall/transaction_detail_1.csv')
transaction_detail_2 =pd.read_csv('https://raw.githubusercontent.com/jin0choi1216/dataset/main/table_data_anlaysis/shopping_mall/transaction_detail_2.csv')

데이터 결합

인덱스 설정 필요

In [ ]:

transaction = pd.concat([transaction_1, transaction_2,], ignore_index=True) # 인덱스 
transaction_detail = pd.concat([transaction_detail_1, transaction_detail_2,], ignore_index=True)

In [ ]:

transaction

Out[ ]:

transaction_idpricepayment_datecustomer_id01234...67816782678367846785

T0000000113	210000	2019-02-01 01:36:57	PL563502
T0000000114	50000	2019-02-01 01:37:23	HD678019
T0000000115	120000	2019-02-01 02:34:19	HD298120
T0000000116	210000	2019-02-01 02:47:23	IK452215
T0000000117	170000	2019-02-01 04:33:46	PL542865
...	...	...	...
T0000006894	180000	2019-07-31 21:20:44	HI400734
T0000006895	85000	2019-07-31 21:52:48	AS339451
T0000006896	100000	2019-07-31 23:35:25	OA027325
T0000006897	85000	2019-07-31 23:39:35	TS624738
T0000006898	85000	2019-07-31 23:41:38	AS834214

6786 rows × 4 columns

In [ ]:

transaction_detail

Out[ ]:

detail_idtransaction_iditem_idquantity01234...71397140714171427143

0	T0000000113	S005	1
1	T0000000114	S001	1
2	T0000000115	S003	1
3	T0000000116	S005	1
4	T0000000117	S002	2
...	...	...	...
7139	T0000006894	S004	1
7140	T0000006895	S002	1
7141	T0000006896	S001	2
7142	T0000006897	S002	1
7143	T0000006898	S002	1

7144 rows × 4 columns

In [ ]:

merge1 = pd.merge(left = transaction_detail, right = transaction, 
                on = 'transaction_id', how = 'left')
merge1

Out[ ]:

detail_idtransaction_iditem_idquantitypricepayment_datecustomer_id01234...71397140714171427143

0	T0000000113	S005	1	210000	2019-02-01 01:36:57	PL563502
1	T0000000114	S001	1	50000	2019-02-01 01:37:23	HD678019
2	T0000000115	S003	1	120000	2019-02-01 02:34:19	HD298120
3	T0000000116	S005	1	210000	2019-02-01 02:47:23	IK452215
4	T0000000117	S002	2	170000	2019-02-01 04:33:46	PL542865
...	...	...	...	...	...	...
7139	T0000006894	S004	1	180000	2019-07-31 21:20:44	HI400734
7140	T0000006895	S002	1	85000	2019-07-31 21:52:48	AS339451
7141	T0000006896	S001	2	100000	2019-07-31 23:35:25	OA027325
7142	T0000006897	S002	1	85000	2019-07-31 23:39:35	TS624738
7143	T0000006898	S002	1	85000	2019-07-31 23:41:38	AS834214

7144 rows × 7 columns

각 테이블과 컬럼의 특성을 고려하면 좋지 못한 사례이다. price(거래 영수증의 총가격)와 quantity(제품의 묶음) 과 함께 해석하다보면 혼선이 발생한다. price 제거하고 merge 한다.

In [ ]:

merge1 = pd.merge(left = transaction_detail, 
                  right = transaction.drop('price', axis = 1), 
                on = 'transaction_id', how = 'left')
merge1

Out[ ]:

detail_idtransaction_iditem_idquantitypayment_datecustomer_id01234...71397140714171427143

0	T0000000113	S005	1	2019-02-01 01:36:57	PL563502
1	T0000000114	S001	1	2019-02-01 01:37:23	HD678019
2	T0000000115	S003	1	2019-02-01 02:34:19	HD298120
3	T0000000116	S005	1	2019-02-01 02:47:23	IK452215
4	T0000000117	S002	2	2019-02-01 04:33:46	PL542865
...	...	...	...	...	...
7139	T0000006894	S004	1	2019-07-31 21:20:44	HI400734
7140	T0000006895	S002	1	2019-07-31 21:52:48	AS339451
7141	T0000006896	S001	2	2019-07-31 23:35:25	OA027325
7142	T0000006897	S002	1	2019-07-31 23:39:35	TS624738
7143	T0000006898	S002	1	2019-07-31 23:41:38	AS834214

7144 rows × 6 columns

In [ ]:

# 위에서 생성한 merge1데이터프레임과 customer_master, item_master를 merge해보세요.
merge2 = pd.merge(left = merge1, right = item_master,
                  how = 'left',
                  on = 'item_id')
merge2

Out[ ]:

detail_idtransaction_iditem_idquantitypayment_datecustomer_iditem_nameitem_price01234...71397140714171427143

0	T0000000113	S005	1	2019-02-01 01:36:57	PL563502	PC-E	210000
1	T0000000114	S001	1	2019-02-01 01:37:23	HD678019	PC-A	50000
2	T0000000115	S003	1	2019-02-01 02:34:19	HD298120	PC-C	120000
3	T0000000116	S005	1	2019-02-01 02:47:23	IK452215	PC-E	210000
4	T0000000117	S002	2	2019-02-01 04:33:46	PL542865	PC-B	85000
...	...	...	...	...	...	...	...
7139	T0000006894	S004	1	2019-07-31 21:20:44	HI400734	PC-D	180000
7140	T0000006895	S002	1	2019-07-31 21:52:48	AS339451	PC-B	85000
7141	T0000006896	S001	2	2019-07-31 23:35:25	OA027325	PC-A	50000
7142	T0000006897	S002	1	2019-07-31 23:39:35	TS624738	PC-B	85000
7143	T0000006898	S002	1	2019-07-31 23:41:38	AS834214	PC-B	85000

7144 rows × 8 columns

In [ ]:

merge3 = pd.merge(left = merge2, right = customer_master, 
                  how = 'left', 
                  on = 'customer_id')
merge3

Out[ ]:

detail_idtransaction_iditem_idquantitypayment_datecustomer_iditem_nameitem_pricecustomer_nameregistration_dateemailgenderagebirthpref01234...71397140714171427143

T0000000113

S005

2019-02-01 01:36:57

PL563502

PC-E

210000

김태경

2019-01-07 14:34

imoto_yoshimasa@example.com

1989-07-15

대전광역시

T0000000114

S001

2019-02-01 01:37:23

HD678019

PC-A

50000

김영웅

2019-01-27 18:00

mifune_rokurou@example.com

1945-11-29

서울특별시

T0000000115

S003

2019-02-01 02:34:19

HD298120

PC-C

120000

김강현

2019-01-11 8:16

yamane_kogan@example.com

1977-05-17

광주광역시

T0000000116

S005

2019-02-01 02:47:23

IK452215

PC-E

210000

김주한

2019-01-10 5:07

ikeda_natsumi@example.com

1972-03-17

인천광역시

T0000000117

S002

2019-02-01 04:33:46

PL542865

PC-B

85000

김영빈

2019-01-25 6:46

kurita_kenichi@example.com

1944-12-17

광주광역시

...

7139

T0000006894

S004

2019-07-31 21:20:44

HI400734

PC-D

180000

김윤성

2019-01-04 13:24

shishido_akira@example.com

1955-01-13

대구광역시

7140

T0000006895

S002

2019-07-31 21:52:48

AS339451

PC-B

85000

김무경

2019-02-11 19:34

aihara_miki@example.com

1945-02-03

대구광역시

7141

T0000006896

S001

2019-07-31 23:35:25

OA027325

PC-A

50000

박준석

2019-04-17 9:23

matsuda_saki@example.com

1979-05-25

서울특별시

7142

T0000006897

S002

2019-07-31 23:39:35

TS624738

PC-B

85000

이가빈

2019-02-20 18:15

shinndou_masatoshi@example.com

1963-02-21

인천광역시

7143

T0000006898

S002

2019-07-31 23:41:38

AS834214

PC-B

85000

이승채

2019-04-07 3:20

tahara_yuuko@example.com

1944-12-18

대전광역시

7144 rows × 15 columns

In [ ]:

merge3.isnull().sum()

Out[ ]:

detail_id            0
transaction_id       0
item_id              0
quantity             0
payment_date         0
customer_id          0
item_name            0
item_price           0
customer_name        0
registration_date    0
email                0
gender               0
age                  0
birth                0
pref                 0
dtype: int64

In [ ]:

merge3['price'] = merge3['quantity'] * merge3['item_price']

In [ ]:

print(merge3['price'].sum())
print(transaction['price'].sum())

971135000
971135000

In [ ]:

merge3.info()

<class 'pandas.core.frame.DataFrame'>
RangeIndex: 7144 entries, 0 to 7143
Data columns (total 16 columns):
 #   Column             Non-Null Count  Dtype 
---  ------             --------------  ----- 
 0   detail_id          7144 non-null   int64 
 1   transaction_id     7144 non-null   object
 2   item_id            7144 non-null   object
 3   quantity           7144 non-null   int64 
 4   payment_date       7144 non-null   object
 5   customer_id        7144 non-null   object
 6   item_name          7144 non-null   object
 7   item_price         7144 non-null   int64 
 8   customer_name      7144 non-null   object
 9   registration_date  7144 non-null   object
 10  email              7144 non-null   object
 11  gender             7144 non-null   object
 12  age                7144 non-null   int64 
 13  birth              7144 non-null   object
 14  pref               7144 non-null   object
 15  price              7144 non-null   int64 
dtypes: int64(5), object(11)
memory usage: 893.1+ KB

In [ ]:

merge3['birth'] = pd.to_datetime(merge3['birth'])
merge3['payment_date'] = pd.to_datetime(merge3['payment_date'])
merge3.head()

Out[ ]:

detail_idtransaction_iditem_idquantitypayment_datecustomer_iditem_nameitem_pricecustomer_nameregistration_dateemailgenderagebirthprefprice01234

T0000000113

S005

2019-02-01 01:36:57

PL563502

PC-E

210000

김태경

2019-01-07 14:34

imoto_yoshimasa@example.com

1989-07-15

대전광역시

210000

T0000000114

S001

2019-02-01 01:37:23

HD678019

PC-A

50000

김영웅

2019-01-27 18:00

mifune_rokurou@example.com

1945-11-29

서울특별시

50000

T0000000115

S003

2019-02-01 02:34:19

HD298120

PC-C

120000

김강현

2019-01-11 8:16

yamane_kogan@example.com

1977-05-17

광주광역시

120000

T0000000116

S005

2019-02-01 02:47:23

IK452215

PC-E

210000

김주한

2019-01-10 5:07

ikeda_natsumi@example.com

1972-03-17

인천광역시

210000

T0000000117

S002

2019-02-01 04:33:46

PL542865

PC-B

85000

김영빈

2019-01-25 6:46

kurita_kenichi@example.com

1944-12-17

광주광역시

170000

실습

In [ ]:

# 1. 성별에 따른 매출 총합을 비교하고 어떤 성별이 매출이 더 높은지 구해보세요.
merge3.groupby(by = ['gender'])['price'].sum() # 방법1
merge3[['gender','price']].groupby('gender').sum() # 방법2

Out[ ]:

pricegenderFM

481045000

490090000

In [ ]:

# 2. birth 칼럼을 활용하여 고객의 태어난 년도 별 매출을 구해보세요.
merge3['year'] = pd.to_datetime(merge3['birth']).dt.year
merge3.groupby(by='year')['price'].sum() # 방법1
merge3[['year', 'price']].groupby(by='year').sum() # 방법2
merge3.pivot_table(index='gender',  values='price', aggfunc='sum') # pivot_table 사용
pd.pivot_table(data = merge3,index='gender',  values='price', aggfunc='sum') # pivot_table 사용

Out[ ]:

pricegenderFM

481045000

490090000

In [ ]:

merge3.head(2)

Out[ ]:

detail_idtransaction_iditem_idquantitypayment_datecustomer_iditem_nameitem_pricecustomer_nameregistration_dateemailgenderagebirthprefpriceyear01

0	T0000000113	S005	1	2019-02-01 01:36:57	PL563502	PC-E	210000	김태경	2019-01-07 14:34	imoto_yoshimasa@example.com	M	30	1989-07-15	대전광역시	210000	1989
1	T0000000114	S001	1	2019-02-01 01:37:23	HD678019	PC-A	50000	김영웅	2019-01-27 18:00	mifune_rokurou@example.com	M	73	1945-11-29	서울특별시	50000	1945

In [ ]:

# payment_date칼럼을 활용하여 구매년월(payment_month) 칼럼을 생성하고 
# 구매년월, 아이템 이름에 따른 매출 총합과 수량 총합을 구해보세요.
# ex) 구매년월 : 201903


# pd.to_datetime() 사용 안한 이유는 이미 birth 타입이 datetime 이다. 그러므로 바로 .dt 사용 가능
merge3['payment_month'] = merge3['birth'].dt.strftime('%Y%m')

subset1 = merge3[['payment_month', 'item_name', 'price', 'quantity']]
subset1.groupby(['payment_month', 'item_name']).sum()

Out[ ]:

pricequantitypayment_monthitem_name193808PC-E193809PC-APC-DPC-E193810PC-A......199906PC-E199907PC-APC-BPC-CPC-E

210000	1
100000	2
180000	1
210000	1
200000	4
...	...
630000	3
150000	3
170000	2
120000	1
210000	1

2763 rows × 2 columns

데이터 시각화

시각화 갤러리

https://matplotlib.org/stable/gallery/index.html
https://seaborn.pydata.org/examples/index.html
https://plotly.com/python/

matplotlib 개념

https://matplotlib.org/stable/api/_as_gen/matplotlib.pyplot.figure.html
Figure vs Axes vs Axis
Figure 도화지
Axes 그래프
Axis 축

In [ ]:

!pip3 install matplot

Requirement already satisfied: matplot in /Users/kimjuyoung/anaconda3/envs/ai2/lib/python3.8/site-packages (0.1.9)
Requirement already satisfied: pyloco>=0.0.134 in /Users/kimjuyoung/anaconda3/envs/ai2/lib/python3.8/site-packages (from matplot) (0.0.139)
Requirement already satisfied: matplotlib>=3.1.1 in /Users/kimjuyoung/anaconda3/envs/ai2/lib/python3.8/site-packages (from matplot) (3.7.3)
Requirement already satisfied: contourpy>=1.0.1 in /Users/kimjuyoung/anaconda3/envs/ai2/lib/python3.8/site-packages (from matplotlib>=3.1.1->matplot) (1.1.0)
Requirement already satisfied: cycler>=0.10 in /Users/kimjuyoung/anaconda3/envs/ai2/lib/python3.8/site-packages (from matplotlib>=3.1.1->matplot) (0.11.0)
Requirement already satisfied: fonttools>=4.22.0 in /Users/kimjuyoung/anaconda3/envs/ai2/lib/python3.8/site-packages (from matplotlib>=3.1.1->matplot) (4.42.1)
Requirement already satisfied: kiwisolver>=1.0.1 in /Users/kimjuyoung/anaconda3/envs/ai2/lib/python3.8/site-packages (from matplotlib>=3.1.1->matplot) (1.4.5)
Requirement already satisfied: numpy<2,>=1.20 in /Users/kimjuyoung/anaconda3/envs/ai2/lib/python3.8/site-packages (from matplotlib>=3.1.1->matplot) (1.24.4)
Requirement already satisfied: packaging>=20.0 in /Users/kimjuyoung/anaconda3/envs/ai2/lib/python3.8/site-packages (from matplotlib>=3.1.1->matplot) (23.1)
Requirement already satisfied: pillow>=6.2.0 in /Users/kimjuyoung/anaconda3/envs/ai2/lib/python3.8/site-packages (from matplotlib>=3.1.1->matplot) (10.0.0)
Requirement already satisfied: pyparsing>=2.3.1 in /Users/kimjuyoung/anaconda3/envs/ai2/lib/python3.8/site-packages (from matplotlib>=3.1.1->matplot) (3.1.1)
Requirement already satisfied: python-dateutil>=2.7 in /Users/kimjuyoung/anaconda3/envs/ai2/lib/python3.8/site-packages (from matplotlib>=3.1.1->matplot) (2.8.2)
Requirement already satisfied: importlib-resources>=3.2.0 in /Users/kimjuyoung/anaconda3/envs/ai2/lib/python3.8/site-packages (from matplotlib>=3.1.1->matplot) (6.0.1)
Requirement already satisfied: ushlex in /Users/kimjuyoung/anaconda3/envs/ai2/lib/python3.8/site-packages (from pyloco>=0.0.134->matplot) (0.99.1)
Requirement already satisfied: websocket-client in /Users/kimjuyoung/anaconda3/envs/ai2/lib/python3.8/site-packages (from pyloco>=0.0.134->matplot) (1.6.3)
Requirement already satisfied: twine in /Users/kimjuyoung/anaconda3/envs/ai2/lib/python3.8/site-packages (from pyloco>=0.0.134->matplot) (4.0.2)
Requirement already satisfied: typing in /Users/kimjuyoung/anaconda3/envs/ai2/lib/python3.8/site-packages (from pyloco>=0.0.134->matplot) (3.7.4.3)
Requirement already satisfied: SimpleWebSocketServer in /Users/kimjuyoung/anaconda3/envs/ai2/lib/python3.8/site-packages (from pyloco>=0.0.134->matplot) (0.1.2)
Requirement already satisfied: zipp>=3.1.0 in /Users/kimjuyoung/anaconda3/envs/ai2/lib/python3.8/site-packages (from importlib-resources>=3.2.0->matplotlib>=3.1.1->matplot) (3.16.2)
Requirement already satisfied: six>=1.5 in /Users/kimjuyoung/anaconda3/envs/ai2/lib/python3.8/site-packages (from python-dateutil>=2.7->matplotlib>=3.1.1->matplot) (1.16.0)
Requirement already satisfied: pkginfo>=1.8.1 in /Users/kimjuyoung/anaconda3/envs/ai2/lib/python3.8/site-packages (from twine->pyloco>=0.0.134->matplot) (1.9.6)
Requirement already satisfied: readme-renderer>=35.0 in /Users/kimjuyoung/anaconda3/envs/ai2/lib/python3.8/site-packages (from twine->pyloco>=0.0.134->matplot) (42.0)
Requirement already satisfied: requests>=2.20 in /Users/kimjuyoung/anaconda3/envs/ai2/lib/python3.8/site-packages (from twine->pyloco>=0.0.134->matplot) (2.31.0)
Requirement already satisfied: requests-toolbelt!=0.9.0,>=0.8.0 in /Users/kimjuyoung/anaconda3/envs/ai2/lib/python3.8/site-packages (from twine->pyloco>=0.0.134->matplot) (1.0.0)
Requirement already satisfied: urllib3>=1.26.0 in /Users/kimjuyoung/anaconda3/envs/ai2/lib/python3.8/site-packages (from twine->pyloco>=0.0.134->matplot) (2.0.4)
Requirement already satisfied: importlib-metadata>=3.6 in /Users/kimjuyoung/anaconda3/envs/ai2/lib/python3.8/site-packages (from twine->pyloco>=0.0.134->matplot) (6.8.0)
Requirement already satisfied: keyring>=15.1 in /Users/kimjuyoung/anaconda3/envs/ai2/lib/python3.8/site-packages (from twine->pyloco>=0.0.134->matplot) (24.2.0)
Requirement already satisfied: rfc3986>=1.4.0 in /Users/kimjuyoung/anaconda3/envs/ai2/lib/python3.8/site-packages (from twine->pyloco>=0.0.134->matplot) (2.0.0)
Requirement already satisfied: rich>=12.0.0 in /Users/kimjuyoung/anaconda3/envs/ai2/lib/python3.8/site-packages (from twine->pyloco>=0.0.134->matplot) (13.5.2)
Requirement already satisfied: jaraco.classes in /Users/kimjuyoung/anaconda3/envs/ai2/lib/python3.8/site-packages (from keyring>=15.1->twine->pyloco>=0.0.134->matplot) (3.3.0)
Requirement already satisfied: nh3>=0.2.14 in /Users/kimjuyoung/anaconda3/envs/ai2/lib/python3.8/site-packages (from readme-renderer>=35.0->twine->pyloco>=0.0.134->matplot) (0.2.14)
Requirement already satisfied: docutils>=0.13.1 in /Users/kimjuyoung/anaconda3/envs/ai2/lib/python3.8/site-packages (from readme-renderer>=35.0->twine->pyloco>=0.0.134->matplot) (0.20.1)
Requirement already satisfied: Pygments>=2.5.1 in /Users/kimjuyoung/anaconda3/envs/ai2/lib/python3.8/site-packages (from readme-renderer>=35.0->twine->pyloco>=0.0.134->matplot) (2.16.1)
Requirement already satisfied: charset-normalizer<4,>=2 in /Users/kimjuyoung/anaconda3/envs/ai2/lib/python3.8/site-packages (from requests>=2.20->twine->pyloco>=0.0.134->matplot) (3.2.0)
Requirement already satisfied: idna<4,>=2.5 in /Users/kimjuyoung/anaconda3/envs/ai2/lib/python3.8/site-packages (from requests>=2.20->twine->pyloco>=0.0.134->matplot) (3.4)
Requirement already satisfied: certifi>=2017.4.17 in /Users/kimjuyoung/anaconda3/envs/ai2/lib/python3.8/site-packages (from requests>=2.20->twine->pyloco>=0.0.134->matplot) (2023.7.22)
Requirement already satisfied: markdown-it-py>=2.2.0 in /Users/kimjuyoung/anaconda3/envs/ai2/lib/python3.8/site-packages (from rich>=12.0.0->twine->pyloco>=0.0.134->matplot) (3.0.0)
Requirement already satisfied: typing-extensions<5.0,>=4.0.0 in /Users/kimjuyoung/anaconda3/envs/ai2/lib/python3.8/site-packages (from rich>=12.0.0->twine->pyloco>=0.0.134->matplot) (4.7.1)
Requirement already satisfied: mdurl~=0.1 in /Users/kimjuyoung/anaconda3/envs/ai2/lib/python3.8/site-packages (from markdown-it-py>=2.2.0->rich>=12.0.0->twine->pyloco>=0.0.134->matplot) (0.1.2)
Requirement already satisfied: more-itertools in /Users/kimjuyoung/anaconda3/envs/ai2/lib/python3.8/site-packages (from jaraco.classes->keyring>=15.1->twine->pyloco>=0.0.134->matplot) (10.1.0)

In [ ]:

import matplotlib.pyplot as plt
import numpy as np

x = np.arange(100, 300, 10)
y = np.random.randn(20)

plt.plot(x, y)

Out[ ]:

[<matplotlib.lines.Line2D at 0x15680e3d0>]

In [ ]:

plt.figure(figsize=(10, 1))
plt.plot(x, y)

Out[ ]:

[<matplotlib.lines.Line2D at 0x15687bd90>]

In [ ]:

plt.xlabel('X-axis')
plt.ylabel('Y-axis')
plt.title('Title')

Out[ ]:

Text(0.5, 1.0, 'Title')

In [ ]:

plt.plot(x,y, color='blue', linestyle = '--', marker = 'o')

Out[ ]:

[<matplotlib.lines.Line2D at 0x15691ef70>]

여러개의 그래프 그리기

In [ ]:

fig = plt.figure(figsize = (16, 10))

axes1 = fig.add_subplot(2, 2, 1) # row, column, index
axes1.plot(x, y, color = 'r', linestyle = '--', marker = 's')

axes2 = fig.add_subplot(2, 2, 2) # row, column, index
axes2.hist(y)

axes3 = fig.add_subplot(2, 2, 3) # row, column
axes3.scatter(x,y)

Out[ ]:

<matplotlib.collections.PathCollection at 0x156a27a90>

In [ ]:

fig, axes = plt.subplots(2,2,figsize = (15,10))
axes[0,0].plot(x, y)
axes[0,1].hist(y)
axes[1,0].scatter(x,y)
axes[1,1].plot(x, y,'r--s')

Out[ ]:

[<matplotlib.lines.Line2D at 0x156be0730>]

하나의 figure 안에 두개의 그래프 그리기

In [ ]:

x1 = np.arange(100, 300, 10)
y1 = np.random.randn(20)

x2 = np.arange(200, 400, 10)
y2 = np.random.randn(20)

plt.plot(x1, y1)#, label='x1')
plt.plot(x2, y2)#, lable='x2')


plt.legend(loc = 1)

No artists with labels found to put in legend.  Note that artists whose label start with an underscore are ignored when legend() is called with no argument.

Out[ ]:

<matplotlib.legend.Legend at 0x156cef5e0>

여러 그래프의 label, title 다루기

In [ ]:

fig = plt.figure(figsize = (16,10))

axes1 = fig.add_subplot(2,2,1) # 2x2 사이즈 만들어서 첫번째
axes1.plot(x, y, color = 'r', linestyle = '--', marker = 's')
axes1.set_title('axes1_title')

axes2 = fig.add_subplot(2,2,2) # 2x2 사이즈 만들어서 두번째
axes2.hist(y)
axes2.set_title('axes2_title')

axes3 = fig.add_subplot(2,2,3) # 2x2 사이즈 만들어서 세번째
axes3.scatter(x,y)
axes3.set_title('axes3_title')

fig.suptitle('Figure Title')

Out[ ]:

Text(0.5, 0.98, 'Figure Title')

실습

20230913_과제(댓글로 제출)

자동차 제조사에 따른 자동차 종류의 개수
자동차 배기량과 고속도로 연비 시각화
자동차 구동방식 별 도시연비 비교 시각화

In [ ]:

import pandas as pd
import matplotlib.pyplot as plt
import numpy as np

mpg = pd.read_csv(
    'https://raw.githubusercontent.com/jin0choi1216/dataset/main/mpg.csv')

In [ ]:

# 1. 자동차 제조사에 따른 자동차 종류의 개수
model_count_by_manu = mpg[['manufacturer','model']].\
    drop_duplicates().\
    groupby('manufacturer').count()

def random_color(lens: pd.DataFrame):
    return [{np.random.random(), np.random.random(), np.random.random()} \
            for i in range(len(lens))]

x = range(model_count_by_manu.size)
count = model_count_by_manu['model']
manufacturer = model_count_by_manu.index

plt.bar(x, count, color=random_color(model_count_by_manu))
plt.xticks(x, manufacturer)

plt.show()

In [ ]:

mpg2 = mpg[['displ', 'hwy']]

In [ ]:

# 2. 자동차 배기량(displ)과 고속도로 연비(hwy) 시각화
import numpy as np
import matplotlib.pyplot as plt
displ = mpg['displ']
hwy = mpg['hwy']

plt.scatter(displ, hwy)
plt.show()

In [ ]:

# 3. 자동차 구동방식(drv) 별 도시연비 (city)비교 시각화
mpg_drop = mpg.drop(mpg.query('drv != "r" and drv != "f"').index)
#mpg_drop
mpg_city_by_drv = mpg_drop[['drv', 'cty']].groupby(['drv']).mean()

x = range(mpg_city_by_drv.size)

drv = mpg_city_by_drv['cty']
values = mpg_city_by_drv.index

plt.bar(x, drv, color=random_color(mpg_city_by_drv))
plt.xticks(x, values)

plt.show()

저작자표시 비영리 변경금지

'새싹 > TIL' 카테고리의 다른 글

[핀테커스] 230915 미니프로젝트 (0)	2023.09.15
[핀테커스] 230914 데이터 시각화 (3)	2023.09.14
[핀테커스] 230912 pandas & 시계열데이터다루기 (0)	2023.09.12
[핀테커스] 230911 데이터 사이언스 라이브러리 pandas 실습 (0)	2023.09.11
[핀테커스] 230908 postgresql 실습 (0)	2023.09.08

현재글[핀테커스] 230913 pandas merge & metplot 시각화

For Engineering

[핀테커스] 230913 pandas merge & metplot 시각화

데이터 결합

실습

데이터 시각화

시각화 갤러리

matplotlib 개념

여러개의 그래프 그리기

하나의 figure 안에 두개의 그래프 그리기

여러 그래프의 label, title 다루기

실습

'새싹 > TIL' 카테고리의 다른 글

'새싹/TIL'의 다른글

티스토리툴바

[핀테커스] 230913 pandas merge & metplot 시각화

데이터 결합

실습

데이터 시각화

시각화 갤러리

matplotlib 개념

여러개의 그래프 그리기

하나의 figure 안에 두개의 그래프 그리기

여러 그래프의 label, title 다루기

실습

'새싹 > TIL' 카테고리의 다른 글

'새싹/TIL'의 다른글

관련글

티스토리툴바