Create a single horizontal stacked bar chart from pandas data

Question

I’m looking to create a single horizontal stacked bar chart as an alternative to a pie chart to show percentages.

I’m pretty sure plot.barh will do the job (it creates multiple stacked bars in another piece of code I have). But here it doesn’t stack the bars. How do I fix it?

import pandas as pd
import matplotlib.pyplot as plt

data=pd.DataFrame(data={"R1":["Yes","Yes","Yes","No","No"]})
freq = data["R1"].value_counts(normalize=True)*100
fig,ax = plt.subplots()
freq.plot.barh(ax=ax,stacked=True)

Asked By: CharlesW

||

Source

Answer 1

Stacking means putting several series of one dataframe on top of each other, so their values cumulate.
You have a single series only, with two values, which consequently are plotted in one bar series.

You could make a DataFrame out of your series, transform it so that Yes and No are two columns with one value each and then plot it the way you tried:

freq.to_frame().T.plot.barh(stacked=True)

to examine the difference:

freq

# Yes    60.0
# No     40.0
# Name: R1, dtype: float64

freq.to_frame().T

#      Yes    No
# R1  60.0  40.0

Answered By: SpghttCd

Create a single horizontal stacked bar chart from pandas data

Question:

Answers: