판다스 데이터 훓어보기 describe, info, dtype

판다스 데이터 훓어보기 describe, info, dtype¶

In [1]:

import pandas as pd
import numpy as np
df = pd.DataFrame({"score1" : [99, 65, 79, 94, 87],
                  "score2" : [79, 85, 78, 93, 97],
                  "score3" : [59, np.nan, 39, np.nan, np.nan],
                  "weight" : [5.43, 0.12, 10.44, 9.33, 4.22],
                  "diff" : [-2.1, 5, 2, -5.4, -3.3],
                  "place":["A", "B", "A", "A", "B"]})
df

Out[1]:

	score1	score2	score3	weight	diff	place
0	99	79	59.0	5.43	-2.1	A
1	65	85	NaN	0.12	5.0	B
2	79	78	39.0	10.44	2.0	A
3	94	93	NaN	9.33	-5.4	A
4	87	97	NaN	4.22	-3.3	B

describe¶

In [2]:

# 컬럼별 카운트, 평균, 표준편차, 최소값, 4분위 수, 최대값을 보여줍니다.
# 문자로 구성된 컬럼은 무시 합니다.
df.describe()

Out[2]:

	score1	score2	score3	weight	diff
count	5.00000	5.000000	2.000000	5.000000	5.000000
mean	84.80000	86.400000	49.000000	5.908000	-0.760000
std	13.38656	8.414274	14.142136	4.148128	4.200357
min	65.00000	78.000000	39.000000	0.120000	-5.400000
25%	79.00000	79.000000	44.000000	4.220000	-3.300000
50%	87.00000	85.000000	49.000000	5.430000	-2.100000
75%	94.00000	93.000000	54.000000	9.330000	2.000000
max	99.00000	97.000000	59.000000	10.440000	5.000000

info¶

In [3]:

# 컬럼 이름, 결측치를 제외한 값 개수, 타입을 보여줍니다.
df.info()

<class 'pandas.core.frame.DataFrame'>
RangeIndex: 5 entries, 0 to 4
Data columns (total 6 columns):
 #   Column  Non-Null Count  Dtype  
---  ------  --------------  -----  
 0   score1  5 non-null      int64  
 1   score2  5 non-null      int64  
 2   score3  2 non-null      float64
 3   weight  5 non-null      float64
 4   diff    5 non-null      float64
 5   place   5 non-null      object 
dtypes: float64(3), int64(2), object(1)
memory usage: 368.0+ bytes

dtypes¶

In [4]:

df.dtypes

Out[4]:

score1      int64
score2      int64
score3    float64
weight    float64
diff      float64
place      object
dtype: object