Can someone help me define best way of visualizing 3 columns in pandas? I tried using stacked bar plot and searched for other solutions on SO, but nothing worked. Any help is appreciated. Here is a dummy pandas dataframe:
Name hour var
Nem 0 2
Kiz 4 1
Hue 5 2
Kiz 0 3
Nem 7 7
UPDATE: is that what you want?
(df.pivot_table(index='Name', columns='hour', values='var',
aggfunc='sum', fill_value=0)
.plot.bar(stacked=True)
)
Explanation:
In [55]: (df.pivot_table(index='Name', columns='hour', values='var',
....: aggfunc='sum', fill_value=0)
....: )
Out[55]:
hour 0 3 4 5 7
Name
Hue 0 6 0 2 0
Kiz 3 0 1 0 0
Nem 2 5 0 0 7
OLD answer:
you can use seaborn module for that:
import seaborn as sns
sns.barplot(x='Name', y='var', hue='hour', data=df, saturation=0.8)
data:
In [20]: df
Out[20]:
Name hour var
0 Nem 0 2
1 Nem 3 5
2 Kiz 4 1
3 Hue 5 2
4 Kiz 0 3
5 Nem 7 7
6 Hue 3 6