Pandas - 数据重塑 Pt.1

这篇随笔主要介绍如何利用 Pandas 进行层次化索引、合并数据和重塑数据
Pt.1 部分主要介绍利用 Pandas 进行层次化索引
Pt.2 部分详细介绍利用 Pandas 合并数据
Pt.3 部分主要介绍利用 Pandas 合并数据和重塑数据
Pt.4 部分主要介绍利用 Pandas 重塑数据

import pandas as pd
import numpy as np

层次化索引

Series 的层次化索引

data = pd.Series(np.random.randn(6),
                 index=[['a', 'a', 'b', 'b', 'c', 'c'],
                        [1, 2, 1, 3, 2, 3]])
data

a  1    0.971509
   2    1.042921
b  1    0.316566
   3    1.730938
c  2   -0.457496
   3    1.183176
dtype: float64

data.index

MultiIndex([('a', 1),
            ('a', 2),
            ('b', 1),
            ('b', 3),
            ('c', 2),
            ('c', 3)],
           )

切片

data['b']

1    0.316566
3    1.730938
dtype: float64

data['b':'c']

b  1    0.316566
   3    1.730938
c  2   -0.457496
   3    1.183176
dtype: float64

data.loc[['b', 'a']]

b  1    0.316566
   3    1.730938
a  1    0.971509
   2    1.042921
dtype: float64

在“内层”中进行切片

data.loc[:, 2]

a    1.042921
c   -0.457496
dtype: float64

series.unstack( ), frame.stack( ) : 数据重塑, Series → DataFrame, DataFrame → Series

frame = data.unstack()
frame

	1	2	3
a	0.971509	1.042921	NaN
b	0.316566	NaN	1.730938
c	NaN	-0.457496	1.183176

frame.stack()

a  1    0.971509
   2    1.042921
b  1    0.316566
   3    1.730938
c  2   -0.457496
   3    1.183176
dtype: float64

DataFrame 的层次化索引

frame = pd.DataFrame(np.arange(12).reshape((4, 3)),
                     index=[['a', 'a', 'b', 'b'], 
                            [1, 2, 1, 2]],
                     columns=[['Ohio', 'Ohio', 'Colorado'],
                              ['Green', 'Red', 'Green']])
frame

		Ohio		Colorado
		Green	Red	Green
a	1	0	1	2
	2	3	4	5
b	1	6	7	8
	2	9	10	11

names 属性

frame.index.names = ['key1', 'key2']
frame.columns.names = ['state', 'color']
frame

	state	Ohio		Colorado
	color	Green	Red	Green
key1	key2
a	1	0	1	2
	2	3	4	5
b	1	6	7	8
	2	9	10	11

切片方式

frame['Ohio'], frame['Ohio']['Green']

(color      Green  Red
 key1 key2            
 a    1         0    1
      2         3    4
 b    1         6    7
      2         9   10,
 key1  key2
 a     1       0
       2       3
 b     1       6
       2       9
 Name: Green, dtype: int32)

frame.set_index( keys, drop ), frame.reset_index( level ) : 将 DataFrame 的列转为索引

frame2 = frame.reset_index()
frame2

state	key1	key2	Ohio		Colorado
color			Green	Red	Green
0	a	1	0	1	2
1	a	2	3	4	5
2	b	1	6	7	8
3	b	2	9	10	11

frame2.set_index(['key1', 'key2'])

	state	Ohio		Colorado
	color	Green	Red	Green
key1	key2
a	1	0	1	2
	2	3	4	5
b	1	6	7	8
	2	9	10	11

frame2.set_index(['key1', 'key2'], drop=False)

	state	key1	key2	Ohio		Colorado
	color			Green	Red	Green
key1	key2
a	1	a	1	0	1	2
	2	a	2	3	4	5
b	1	b	1	6	7	8
	2	b	2	9	10	11

frame.swaplevel( level1, level2 ), frame.sort_index( level ) : 重排与分级排序

frame

	state	Ohio		Colorado
	color	Green	Red	Green
key1	key2
a	1	0	1	2
	2	3	4	5
b	1	6	7	8
	2	9	10	11

frame.sort_index(level=1)

	state	Ohio		Colorado
	color	Green	Red	Green
key1	key2
a	1	0	1	2
b	1	6	7	8
a	2	3	4	5
b	2	9	10	11

frame.swaplevel('key1', 'key2')

	state	Ohio		Colorado
	color	Green	Red	Green
key2	key1
1	a	0	1	2
2	a	3	4	5
1	b	6	7	8
2	b	9	10	11

frame.swaplevel('key1', 'key2').sort_index(level='key2')

	state	Ohio		Colorado
	color	Green	Red	Green
key2	key1
1	a	0	1	2
	b	6	7	8
2	a	3	4	5
	b	9	10	11

frame.groupby( axis, level ).sum( ) : 汇总统计

frame.groupby(level='key2').sum()

state	Ohio		Colorado
color	Green	Red	Green
key2
1	6	8	10
2	12	14	16

frame.groupby(axis=1, level='color').sum()

	color	Green	Red
key1	key2
a	1	2	1
	2	8	4
b	1	14	7
	2	20	10