pandas plot dataframe barplot with colors by category

Question

I would like to use pandas to plot a barplot with diffrent colors for category in column.

Here is a simple example: (index is variable)

df:
         value   group
variable               
a             10      1
b              9      1
c              8      1
d              7      2
f              6      2
g              5      3
h              4      3

I would like to make a barplot with coloring on group. I would also like to specify the colors. In my original dataset I have many goups. Could someone help me with this?

Solution

Just pass a color parameter to the plot function with a list of colors:

df['group'].plot(kind='bar', color=['r', 'g', 'b', 'r', 'g', 'b', 'r'])

If you want to plot the value as bars and you also want the group to determine the color of the bar, use:

colors = {1: 'r', 2: 'b', 3: 'g'}
df['value'].plot(kind='bar', color=[colors[i] for i in df['group']])

You can also use something like:

list(df['group'].map(colors))

Instead of the list comprehension.

Comments

[FaCoffee]

How do you do this when you have several bar plots but not all of them have the same number of categories, although the categories across the different plots are the same?