Data_analysis
14 Jul 2020

Series NaN handling

Pandas의 결측치

Pandas에서는 null값을 missing data 혹은 missing이라고 부릅니다. missing과 null은 번갈아가며 쓸 수 있지만 pandas의 표준대로 말하자면 missing이 더 잘 쓰입니다. 우리나라 말로는 결측치라고 합니다.

import numpy as np
import pandas as pd
from pandas import Series, DataFrame
import matplotlib.pyplot as plt

test= Series([None, None, 1, 2, 3, 4, 5, 6, None])
print(test)
'''
0    NaN
1    NaN
2    1.0
3    2.0
4    3.0
5    4.0
6    5.0
7    6.0
8    NaN
dtype: float64
'''


isnull()

import numpy as np
import pandas as pd
from pandas import Series, DataFrame
import matplotlib.pyplot as plt

test= Series([None, None, 1, 2, 3, 4, 5, 6, None])
print(test.isnull())
'''
0     True
1     True
2    False
3    False
4    False
5    False
6    False
7    False
8     True
dtype: bool      --> NaN은 True, 그 외는 False
'''
.
.


notnull()

import numpy as np
import pandas as pd
from pandas import Series, DataFrame
import matplotlib.pyplot as plt

test= Series([None, None, 1, 2, 3, 4, 5, 6, None])
print(test.notnull())
'''
0    False
1    False
2     True
3     True
4     True
5     True
6     True
7     True
8    False         -->Nan은 False, 그 외는 True
'''
.
.


NaN값의 개수 구하기

import numpy as np
import pandas as pd
from pandas import Series, DataFrame
import matplotlib.pyplot as plt

test= Series([None, None, 1, 2, 3, 4, 5, 6, None])

test.isnull().sum()
'''
3
'''


NaN이 아닌 데이터 조회하기

import numpy as np
import pandas as pd
from pandas import Series, DataFrame
import matplotlib.pyplot as plt

test= Series([None, None, 1, 2, 3, 4, 5, 6, None])

print(result[result.notnull()])
'''
2   -3.0
3   -6.0
4   -5.0
5    3.0
6    0.0
7    3.0
dtype: float64
'''
Tags:
0 comments