python - Pandas: boxplot of one column based on another column

Question

Say I have a dataframe like the following:

my_dataframe:

   Age  Group
0  31   A
1  24   A
2  25   A
3  36   A
4  50   NaN
5  27   A
6  49   A
7  24   A
8  63   A
9  25   A
10  65  A
11  67  A
12  59  A
13 NaN  B
14  30  B
15  19  B
16  57  B
17  62  B
18  30  B
19  50  B
20  42  B
21  45  C
22  59  C
23  28  C
24  37  C
25  29  C

I would like to boxplot the age of each Group (A,B,C). Note that I have some NaN values in the dataframe. How can I do this in Pandas?

Answer 1

误读了第一次，所以给出了直方图的答案……在下面。对于箱线图，代码是：

bp = df.boxplot(by='Group')

suptitle('Bla Bla')

更改或删除自动生成的顶部标题。

可能是一种更优雅的方式，但以下适用于直方图：

df[df.Group =='A'].Age.hist()
df[df.Group =='B'].Age.hist()
df[df.Group =='C'].Age.hist()

http://pandas.pydata.org/pandas-docs/dev/visualization.html也有一些花哨的语法来做到这一点。但由于只有 3 组，简单的解决方案可能就足够了。