결측값 처리

실습 데이터 생성

import pandas as pd
import numpy as np

# 실습 데이터 생성
df = pd.DataFrame(data = np.arange(18).reshape(6,3),
				  index = ['a','b','c','d','e','f'],
                  columns=['col1','col2','col3'])
                  
df['col4'] = pd.Series(data = [1.7, 1.2, 2.4], 
                       index = ['a','e','c'])
df.loc['c'] = None
print('Sample Data')
print(df)

Sample Data col1 col2 col3 col4

a 0.0 1.0 2.0 1.7

b 3.0 4.0 5.0 NaN

c NaN NaN NaN NaN

d 9.0 10.0 11.0 NaN

e 12.0 13.0 14.0 1.2

f 15.0 16.0 17.0 NaN

df.info()

<class 'pandas.core.frame.DataFrame'>

Index: 6 entries, a to f

Data columns (total 4 columns):

Column Non-Null Count Dtype

---

0 col1 5 non-null float64

1 col2 5 non-null float64

2 col3 5 non-null float64

3 col4 2 non-null float64

dtypes: float64(4)

memory usage: 412.0+ bytes

df.describe()

 

 

dropna() - 결측값 삭제

df.isnull()

df.dropna()

df.dropna(axis=1)

df.dropna(how='all')

df.dropna(how='all', axis = 1)

 

 

fillna() - 결측값 대체

• DataFrame.fillna(value=None, ***, method=None, axis=None, inplace=False, limit=None, downcast=None)

df.fillna(1)

replace_set = {'col2':0,'col4':'100'}
print(df.fillna(replace_set))

col1 col2 col3 col4

a 0.0 1.0 2.0 1.7

b 3.0 4.0 5.0 100

c NaN 0.0 NaN 100

d 9.0 10.0 11.0 100

e 12.0 13.0 14.0 1.2

f 15.0 16.0 17.0 100

replace_set = {'col1':df['col1'].mean()}
df.fillna(replace_set, inplace=True)

replace_set = {'col1':df['col1'].max()}
df.fillna(replace_set)

 

앞에 값으로 대체

df.fillna(method='ffill')

 

뒤에 값으로 대체

df.fillna(method='bfill')

특정 값으로 대체

# 특정 값으로 대체
df.fillna(value=100, inplace=True)

# 0으로 대체
df.fillna(value=0, inplace=True)