dataframe的增删改查

#dataframe的增删改查之增
import numpy as np
import pandas as pd

df = pd.DataFrame(np.random.randn(3,4),index=list("abc"),columns=list("ABCD"))
df

	A	B	C	D
a	-1.191323	-0.774396	0.234996	-0.565204
b	0.042025	-0.003400	-1.517832	-0.729930
c	0.459724	0.341114	-0.070776	0.515991

#增加列
df["E"] = df.A + df.B  
df

	A	B	C	D	E
a	1.241630	0.384139	-1.225968	-1.160028	1.625769
b	-0.736530	-0.148695	-1.622824	-0.145651	-0.885224
c	0.370135	-1.130262	0.032585	-0.416405	-0.760128

df.insert(1,"insert",6) #insert(loc, column, value, allow_duplicates=False)
df

	A	insert	B	C	D	E
a	1.241630	6	0.384139	-1.225968	-1.160028	1.625769
b	-0.736530	6	-0.148695	-1.622824	-0.145651	-0.885224
c	0.370135	6	-1.130262	0.032585	-0.416405	-0.760128

df.assign(F = df["A"] / df["B"]) # DataFrame has an assign() method that allows you to easily create new columns that are potentially derived from existing columns.
#注意此时的F并未正真的添加到df中

	A	insert	B	C	D	E	F
a	1.241630	6	0.384139	-1.225968	-1.160028	1.625769	3.232243
b	-0.736530	6	-0.148695	-1.622824	-0.145651	-0.885224	4.953305
c	0.370135	6	-1.130262	0.032585	-0.416405	-0.760128	-0.327477

#增加行
#merge / concatenate / join 等函数实现
#其他章节研究一下

#删除列
del df["A"]    #我本来想删除E列，报错了，E列实际上可能没保存在df的内存地址中
df

	B	C	D	E
a	0.224567	-1.049519	0.157046	-1.692286
b	1.359846	1.605034	0.213492	1.141026
c	0.206875	-0.037353	-0.092492	-0.673894

df.pop("B")  #看来删除实际上改动了原来的dataframe
df

	C	D	E
a	-1.049519	0.157046	-1.692286
b	1.605034	0.213492	1.141026
c	-0.037353	-0.092492	-0.673894

#删除行
del df.a

---------------------------------------------------------------------------

AttributeError                            Traceback (most recent call last)

<ipython-input-38-5c6c11f8ca66> in <module>
      1 #删除行
----> 2 del df.a


AttributeError: a

#删除行
del df.loc["c",:]
df

---------------------------------------------------------------------------

AttributeError                            Traceback (most recent call last)

<ipython-input-39-87bdf8cd1515> in <module>
      1 #删除行
----> 2 del df.loc["c",:]
      3 df


AttributeError: __delitem__

#改动列
df.C = "bar"
df

	C	D	E
a	bar	0.157046	-1.692286
b	bar	0.213492	1.141026
c	bar	-0.092492	-0.673894

#改动行
df.c = "barr"  #可以发现重新赋值没用
df

	C	D	E
a	bar	0.157046	-1.692286
b	bar	0.213492	1.141026
c	bar	-0.092492	-0.673894

df.c  #看来是把df.c当成变量了

'barr'

#改动行
df.loc["a",:] = "barr"  #改动应该都是先选择后赋值
df                      #所以我应该先学查找的啊

	C	D	E
a	barr	barr	barr
b	bar	0.213492	1.14103
c	bar	-0.0924917	-0.673894

#重头戏之查找
df

	A	insert	B	C	D	E
a	1.241630	6	0.384139	-1.225968	-1.160028	1.625769
b	-0.736530	6	-0.148695	-1.622824	-0.145651	-0.885224
c	0.370135	6	-1.130262	0.032585	-0.416405	-0.760128

df["A"]  #该方式选择的列，返回series

a    1.241630
b   -0.736530
c    0.370135
Name: A, dtype: float64

df.loc["a"]  #该方式以行的label选择行，返回的是series

A         1.241630
insert    6.000000
B         0.384139
C        -1.225968
D        -1.160028
E         1.625769
Name: a, dtype: float64

df.iloc[0]   #该方式以行的数字索引选择行，返回的是series

A         1.241630
insert    6.000000
B         0.384139
C        -1.225968
D        -1.160028
E         1.625769
Name: a, dtype: float64

df[0:2]   #以切片的方式选择行，返回dataframe

	A	insert	B	C	D	E
a	1.24163	6	0.384139	-1.225968	-1.160028	1.625769
b	-0.73653	6	-0.148695	-1.622824	-0.145651	-0.885224

df[0:-1]

	A	insert	B	C	D	E
a	1.24163	6	0.384139	-1.225968	-1.160028	1.625769
b	-0.73653	6	-0.148695	-1.622824	-0.145651	-0.885224

#选则具体的单个元素
df.loc["a","A"]

1.2416298344644054

df.iloc[0,0]

1.2416298344644054

df.loc[["a","c"],["A","C"]]

	A	C
a	1.241630	-1.225968
c	0.370135	0.032585

df.iloc[[0,2],[0,3]]

	A	C
a	1.241630	-1.225968
c	0.370135	0.032585

df.iloc[[0,2]] #不连续选个行的方法

	A	insert	B	C	D	E
a	1.241630	6	0.384139	-1.225968	-1.160028	1.625769
c	0.370135	6	-1.130262	0.032585	-0.416405	-0.760128

df.loc[["a","c"]]

	A	insert	B	C	D	E
a	1.241630	6	0.384139	-1.225968	-1.160028	1.625769
c	0.370135	6	-1.130262	0.032585	-0.416405	-0.760128

df[[0,2],["A","C"]]

---------------------------------------------------------------------------

TypeError                                 Traceback (most recent call last)

<ipython-input-81-970a7eaa38bd> in <module>
----> 1 df[[0,2],["A","C"]]


~/miniconda3/lib/python3.7/site-packages/pandas/core/frame.py in __getitem__(self, key)
   2798             if self.columns.nlevels > 1:
   2799                 return self._getitem_multilevel(key)
-> 2800             indexer = self.columns.get_loc(key)
   2801             if is_integer(indexer):
   2802                 indexer = [indexer]


~/miniconda3/lib/python3.7/site-packages/pandas/core/indexes/base.py in get_loc(self, key, method, tolerance)
   2644                 )
   2645             try:
-> 2646                 return self._engine.get_loc(key)
   2647             except KeyError:
   2648                 return self._engine.get_loc(self._maybe_cast_indexer(key))


pandas/_libs/index.pyx in pandas._libs.index.IndexEngine.get_loc()


pandas/_libs/index.pyx in pandas._libs.index.IndexEngine.get_loc()


TypeError: '([0, 2], ['A', 'C'])' is an invalid key

#中间开了小差，继续
df

	A	B	C	D
a	-1.191323	-0.774396	0.234996	-0.565204
b	0.042025	-0.003400	-1.517832	-0.729930
c	0.459724	0.341114	-0.070776	0.515991

df[df["A"] > 0]   #if then语句 ，根据A列数值选择行

	A	B	C	D
b	0.042025	-0.003400	-1.517832	-0.729930
c	0.459724	0.341114	-0.070776	0.515991

df[(df.A > 0) & (df.D > 0)]  # &  |

	A	B	C	D
c	0.459724	0.341114	-0.070776	0.515991

df[df > 0]

	A	B	C	D
a	NaN	NaN	0.234996	NaN
b	0.042025	NaN	NaN	NaN
c	0.459724	0.341114	NaN	0.515991

df["E"] = ["AA","BB","CC"]
df

	A	B	C	D	E
a	-1.191323	-0.774396	0.234996	-0.565204	AA
b	0.042025	-0.003400	-1.517832	-0.729930	BB
c	0.459724	0.341114	-0.070776	0.515991	CC

df[df.E.isin(["AA","CC"])]   #siin函数制造bool值然后进行选择

	A	B	C	D	E
a	-1.191323	-0.774396	0.234996	-0.565204	AA
c	0.459724	0.341114	-0.070776	0.515991	CC