Pandas plot 在同一个条形图上具有不同变量的几个 df

Question

I have 4 Dataframes with different location: Indonesia, Singapore, Malaysia and Total each of them containing the percentage of the 5 top revenue-generating products. I have plotted them separately.

I want to combine them together on one plot where X-axis shows different locations and top-revenue-generating products for each location.

I have printed data frames and as you can see they have different products in them.

print(Ind_top_cat, Sin_top_cat, Mal_top_cat, Tot_top_cat)

Category     Amt   
M020P     0.144131
MH        0.099439
ML        0.055052
PB        0.050057
PPDR      0.048315                

Category     Amt   
ML        0.480781
M015      0.073034
PPDR      0.035412
M025      0.033418
M020      0.031836                

Category     Amt   
TN        0.343650
PPDR      0.190773
NMCN      0.118425
M015      0.047539
NN        0.038140                

Category     Amt      
M020P     0.158575
MH        0.092012
ML        0.064179
PPDR      0.050803
PB        0.044301

Thanks to joelostblom I was able to construct a plot, however, there are still some issues. enter image description here

all_countries = pd.concat([Ind_top_cat, Sin_top_cat, Mal_top_cat, Tot_top_cat])
all_countries['Category'] = all_countries.index
sns.barplot(x='Country', y='Amt',hue = 'Category',data=all_countries)

Is there any way I can put legend values on the x-axis (no need to colour categories on I want to instead colour countries), and put data values on top of bars. Also, bars are not centred and have no idea how to solve it.

Answer 1

You could create a new column in each dataframe with the country name, eg

Ind_top_cat['Country'] = 'Indonesia'
Sin_top_cat['Country'] = 'Singapore'

The you can create one big dataframe by concatenating the country dataframes together:

all_countries = pd.concat([Ind_top_cat, Sin_top_cat])

And finally, you can use a high level plotting library such as seaborn to assign one column to the x-axis location and one to the color of the bars:

import seaborn as sns

sns.barplot(x='Country', y='Amt', color='Category', data=all_countries)

You can scroll down to the second example on this page to get an idea what such a plot would look like (also pasted below):