Python与数据科学实战课程——第三章Pandas：多级index

import numpy as np
import pandas as pd
from pandas import Series,DataFrame

创建多级index的series

s1 = Series(np.random.randn(6))
s1

0 0.001150
1 -0.019820
2 0.922708
3 0.656468
4 -0.719109
5 0.957127
dtype: float64

s1 = Series(np.random.randn(6),index=[["1","1","1","2","2","2"],["a","b","c","a","b","c"]])
s1    #此时s1就有两级index   ,是一个二维结构，因此可以向dataframe转换

1  a    0.925230
   b    0.766318
   c    0.687740
2  a   -0.622509
   b   -0.593884
   c   -0.119193
dtype: float64

s1["1"]

a 0.925230
b 0.766318
c 0.687740
dtype: float64

type(s1["1"])   #可以看到其实这个series是由s1["1"]，s1["2"]这两个series组成

pandas.core.series.Series

s1["1"]["a"]

0.92522961872552423

s1[:,"a"]

1 0.925230
2 -0.622509
dtype: float64

多级index的series与dataframe的互换

df1 = s1.unstack()
df1

	a	b	c
1	0.925230	0.766318	0.687740
2	-0.622509	-0.593884	-0.119193

df2 = DataFrame([s1["1"],s1["2"]])
df2

	a	b	c
0	0.925230	0.766318	0.687740
1	-0.622509	-0.593884	-0.119193

s2 = df1.unstack()
s2

a  1    0.925230
   2   -0.622509
b  1    0.766318
   2   -0.593884
c  1    0.687740
   2   -0.119193
dtype: float64

s2 = df1.T.unstack()
s2

1  a    0.925230
   b    0.766318
   c    0.687740
2  a   -0.622509
   b   -0.593884
   c   -0.119193
dtype: float64

创建多级index的dataframe

df = DataFrame(np.arange(16).reshape(4,4))
df

	0	1	2	3
0	0	1	2	3
1	4	5	6	7
2	8	9	10	11
3	12	13	14	15

df = DataFrame(np.arange(16).reshape(4,4),index=[["a","a","b","b"],["1","2","1","2"]])
df

		0	1	2	3
a	1	0	1	2	3
	2	4	5	6	7
b	1	8	9	10	11
	2	12	13	14	15

df = DataFrame(np.arange(16).reshape(4,4),index=[["a","a","b","b"],["1","2","1","2"]],columns=[["BJ","BJ","SH","GZ"],[8,9,8,8]])
df

		BJ		SH	GZ
		8	9	8	8
a	1	0	1	2	3
	2	4	5	6	7
b	1	8	9	10	11
	2	12	13	14	15

df["BJ"]

		8	9
a	1	0	1
	2	4	5
b	1	8	9
	2	12	13

type(df["BJ"])

pandas.core.frame.DataFrame

df["BJ"][8]

a  1     0
   2     4
b  1     8
   2    12
Name: 8, dtype: int32

df["BJ"][8]["a"]

1 0
2 4
Name: 8, dtype: int32

df["BJ"][8]["a"]["1"]         #多级index的dataframe的引用顺序  一级列名->二级列名->一级index名->二级index名

0

df["BJ"][8]["a"][1]       #多级index的dataframe的引用顺序  一级列名->二级列名->一级index名->二级index

4