Seaborn plot 与 pandas 系列对象

Question

I have 4 pandas Series objects, they are different in size and also they have different indexes. I want to create a barplot or boxplot to show how median values of these Series differ.

eg one of my Series is:

13. 0.912
14. 1.4324
15. 2.3910
16. 1.4324
17. 5.2331...

another:

10. 2.1231
11. 3.4244
12. 4.123...

I can't set seaborn.boxplot or seaborn.barplot to visualize something like this:

Answer 1

Use concat with DataFrame.stack and Series.reset_index for DataFrame and then plot:

s1 = pd.Series([1,2,3])
s2 = pd.Series([20,1,3,6,90], index=list('abcde'))
s3 = pd.Series([4,5,2.6], index=list('ABC'))
s4 = pd.Series([7,20.8], index=list('XY'))

df = (pd.concat([s1, s2, s3, s4], axis=1, keys=('a','b','c','d'))
        .stack()
        .rename_axis(('a','b'))
        .reset_index(name='c'))
print (df)
    a  b     c
0   0  a   1.0
1   1  a   2.0
2   2  a   3.0
3   A  c   4.0
4   B  c   5.0
5   C  c   2.6
6   X  d   7.0
7   Y  d  20.8
8   a  b  20.0
9   b  b   1.0
10  c  b   3.0
11  d  b   6.0
12  e  b  90.0

sns.barplot(data=df, x='b', y='c')

Similar idea with DataFrame.melt and remove missing values by DataFrame.dropna :

s1 = pd.Series([1,2,3])
s2 = pd.Series([20,1,3,6,90], index=list('abcde'))
s3 = pd.Series([4,5,2.6], index=list('ABC'))
s4 = pd.Series([7,20.8], index=list('XY'))


df = pd.concat([s1, s2, s3, s4], axis=1, keys=('a','b','c','d')).melt().dropna()
print (df)
   variable  value
0         a    1.0
1         a    2.0
2         a    3.0
21        b   20.0
22        b    1.0
23        b    3.0
24        b    6.0
25        b   90.0
29        c    4.0
30        c    5.0
31        c    2.6
45        d    7.0
46        d   20.8

sns.barplot(data=df, x='variable', y='value')