04wk-14: msno를 이용한 결측치 시각화

1. 강의영상

2. Import

import numpy as np
import pandas as pd 
import missingno as msno

3. 데이터

df = pd.read_csv("https://raw.githubusercontent.com/guebin/MP2023/main/posts/msno.csv")
df

	A	B	C	D	E
0	0.383420	1.385096	NaN	-0.545132	-0.732395
1	1.084175	0.080613	-0.770527	-0.272143	-0.749881
2	1.142778	1.258419	NaN	-0.072007	-0.440757
3	0.307894	0.521400	0.446974	0.329530	-1.457388
4	0.237787	0.132401	-0.516630	0.177995	0.416182
...	...	...	...	...	...
995	0.041092	-1.308165	1.085820	1.136210	NaN
996	-1.286358	1.547987	NaN	-0.174334	-0.579486
997	0.710257	1.764058	NaN	-0.353928	NaN
998	-1.908729	-0.804691	NaN	NaN	-0.066739
999	0.650026	2.206549	NaN	-0.919945	NaN

1000 rows × 5 columns

4. 결측치 숫자 파악

A. df.info()

df.info()

<class 'pandas.core.frame.DataFrame'>
RangeIndex: 1000 entries, 0 to 999
Data columns (total 5 columns):
 #   Column  Non-Null Count  Dtype  
---  ------  --------------  -----  
 0   A       668 non-null    float64
 1   B       656 non-null    float64
 2   C       608 non-null    float64
 3   D       668 non-null    float64
 4   E       660 non-null    float64
dtypes: float64(5)
memory usage: 39.2 KB

B. msno.bar()

msno.bar(df)

5. 패턴파악 및 시각화

A. msno.matrix()

msno.matrix(df)

B. msno.heatmap()

msno.heatmap(df)

C. msno.dendrogram()

msno.dendrogram(df)