沧海拾珠

Pandas中unstack和stack用法

发表于 | 分类于 |

1. DataFrame.unstack(level = -1 , fill_value = None)，unstack 本身意为不堆叠，这里我理解为将数据结构展开。

level值可以改为其他值，如0。fill_value可以指定空白值为多少，如fill_value = 0。此外还可以将需要unstack的元素传入作为参数，如DataFrame.unstack(‘Medal’)。

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
>>> s
one  a   1.0
     b   2.0
two  a   3.0
     b   4.0
>>> s.unstack()
     a   b
one  1.0  2.0
two  3.0  4.0
>>> s.unstack(level=0)
   one  two
a  1.0   3.0
b  2.0   4.0

当dataFrame用到分组groupby的时候，数据会堆叠在一起，如：

1
2
3
4
5
6
7
8
9
10
11
df.groupby(['Edition','Medal']).size()
thlete                   Medal 
AABYE, Edgar              Gold      1
AALTONEN, Arvo Ossian     Bronze    2
AALTONEN, Paavo Johannes  Bronze    2
                          Gold      3
AAMODT, Ragnhild          Gold      1
AANING, Alf Lied          Silver    1
AARDENBURG, Willemien     Bronze    1
AARDEWIJN, Pepijn         Silver    1

使用unstack()

1
2
3
4
5
6
7
8
9
10
11
df.groupby(['Edition','Medal']).size().unstack()
Medal	             Bronze	Gold	Silver
Athlete			
AABYE, Edgar	      NaN	1.0	NaN
AALTONEN, Arvo Ossian 2.0	NaN	NaN
AALTONEN, Paavo Johannes2.0	3.0	NaN
AAMODT, Ragnhild	NaN	1.0	NaN
AANING, Alf Lied	NaN	NaN	1.0
AARDENBURG, Willemien	1.0	NaN	NaN
AARDEWIJN, Pepijn	NaN	NaN	1.0

应用unstack()之后medal数据分为单独的三列，之后方便作图。

2. DataFrame.stack(level = -1, dropna = True)，暂时还没用到这个功能。

1
2
3
4
5
6
7
8
9
>>> s
     a   b
one  1.  2.
two  3.  4.
>>> s.stack()
one a    1
    b    2
two a    3
    b    4