파이썬 '판다스 데이터 분석' - 복사 , DataFrame.copy()

 

DataFrame 복사에 대해 테스트

Make a copy of this object’s indices and data.

DataFrame.copy(deep=True)

 

df2 = df

df3 = df[:]

df3 = df.copy() # deep = True (default)

 

---

 

df2 = df

# 복사 테스트 #
df2 = df
df2.tail()

<결과>
394	44.0	4.0	97.0	52.00	2130.0	24.6	82.0	2.0	vw pickup
395	32.0	4.0	135.0	84.00	2295.0	11.6	82.0	1.0	dodge rampage
396	28.0	4.0	120.0	79.00	2625.0	18.6	82.0	1.0	ford ranger
397	31.0	4.0	119.0	82.00	2720.0	19.4	82.0	1.0	chevy s-10
398	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN

 

drop

df2.drop(398,inplace=True)
print(df.tail())
df2.tail()

<결과>
      mpg  cylinders  displacement horsepower  weight  acceleration  \
393  27.0        4.0         140.0      86.00  2790.0          15.6   
394  44.0        4.0          97.0      52.00  2130.0          24.6   
395  32.0        4.0         135.0      84.00  2295.0          11.6   
396  28.0        4.0         120.0      79.00  2625.0          18.6   
397  31.0        4.0         119.0      82.00  2720.0          19.4   

     model year  origin             name  
393        82.0     1.0  ford mustang gl  
394        82.0     2.0        vw pickup  
395        82.0     1.0    dodge rampage  
396        82.0     1.0      ford ranger  
397        82.0     1.0       chevy s-10  

<결과>
	mpg	cylinders	displacement	horsepower	weight	acceleration	model year	origin	name
393	27.0	4.0	140.0	86.00	2790.0	15.6	82.0	1.0	ford mustang gl
394	44.0	4.0	97.0	52.00	2130.0	24.6	82.0	2.0	vw pickup
395	32.0	4.0	135.0	84.00	2295.0	11.6	82.0	1.0	dodge rampage
396	28.0	4.0	120.0	79.00	2625.0	18.6	82.0	1.0	ford ranger
397	31.0	4.0	119.0	82.00	2720.0	19.4	82.0	1.0	chevy s-10

df2에 원본의 주소를 가지고 있기때문에

df2가 df의 398 인덱스를 가진 행이 드랍된 결과를 반환

 

df3 = df[ : ]

df3 = df[:]
df3.drop(398,inplace=True)
print(df.tail())
df3.tail()

<결과>

경고문이 발생한다.

<결과>
	mpg	cylinders	displacement	horsepower	weight	acceleration	model year	origin	name
393	27.0	4.0	140.0	86.00	2790.0	15.6	82.0	1.0	ford mustang gl
394	44.0	4.0	97.0	52.00	2130.0	24.6	82.0	2.0	vw pickup
395	32.0	4.0	135.0	84.00	2295.0	11.6	82.0	1.0	dodge rampage
396	28.0	4.0	120.0	79.00	2625.0	18.6	82.0	1.0	ford ranger
397	31.0	4.0	119.0	82.00	2720.0	19.4	82.0	1.0	chevy s-10

복사를 한 df3의 398번째의 행만 drop

 

 

df3 = df.copy()

df3 = df.copy()
df.drop(398,inplace=True)
print(df.tail())
df3.tail()

<결과>
      mpg  cylinders  displacement horsepower  weight  acceleration  \
393  27.0        4.0         140.0      86.00  2790.0          15.6   
394  44.0        4.0          97.0      52.00  2130.0          24.6   
395  32.0        4.0         135.0      84.00  2295.0          11.6   
396  28.0        4.0         120.0      79.00  2625.0          18.6   
397  31.0        4.0         119.0      82.00  2720.0          19.4   

     model year  origin             name  
393        82.0     1.0  ford mustang gl  
394        82.0     2.0        vw pickup  
395        82.0     1.0    dodge rampage  
396        82.0     1.0      ford ranger  
397        82.0     1.0       chevy s-10  

<결과>
mpg	cylinders	displacement	horsepower	weight	acceleration	model year	origin	name
394	44.0	4.0	97.0	52.00	2130.0	24.6	82.0	2.0	vw pickup
395	32.0	4.0	135.0	84.00	2295.0	11.6	82.0	1.0	dodge rampage
396	28.0	4.0	120.0	79.00	2625.0	18.6	82.0	1.0	ford ranger
397	31.0	4.0	119.0	82.00	2720.0	19.4	82.0	1.0	chevy s-10
398	NaN	NaN	NaN	None	NaN	NaN	NaN	NaN	None

경고문없이 사본만 깔끔하게 drop 되는 것을 볼 수 있다.

 

데이터프레임을 복사할때 copy() 를 사용하는 것을 권장한다.