CODEWITHSUNDEEP
pythonseaborn
AIMCognitiveGurgaon BigData
AIMCognitiveGurgaon BigData

Reputation: 133

plot percent bar in seaborn from dataframe

I have a dataframe which looks like :

|---------------------|------------------|------------------|
|      Gender         |    Male          |    Female        |
|---------------------|------------------|------------------|
|      Scenario       |                  |                  |
|---------------------|------------------|------------------|
|           A         |       79         |      217         |
|---------------------|------------------|------------------|
|           B         |       59         |      408         |
|---------------------|------------------|------------------|
|           C         |       420        |      330         |
|---------------------|------------------|------------------|
|           D         |       208        |      1330        |

I would like to plot a seaborn percent bar graph, it should be a single bar which should represent the percent of female i.e. for Scenario A the bar should show 217/(217+79) and so on. How can i achieve this. I am able to plot stacked bar or side by side bar from the following code.

stacked = my_dataframe.stack().reset_index().rename(columns={0:'value'})
# plot grouped bar chart
sns.barplot(x=stacked.Scenario, y=stacked.value, hue=stacked.Gender)
plt.show()

Upvotes: 0

Views: 481

Answers (1)

Stael
Stael

Reputation: 2689

not my most elegant code, but this works:

import pandas as pd


stacked = pd.DataFrame({'Scenario': list('ABCD'), 'Male': [79, 59, 420, 208], 'Female': [217, 408, 330, 1330]})

pd.DataFrame(stacked.apply(lambda x: {'Scenario': x.Scenario, 'Male_pct': x.Male / (x.Male + x.Female), 'Female_pct': x.Female / (x.Male + x.Female)}, axis=1,).tolist())

then just plot it how you were plotting already.

Upvotes: 1

Related Questions