Format y-axis as percentage in a seaborn.objects plot

Question

I found a suggestion to use ticker.PercentFormatter. This changed the decimal numbers on the y-axis to percentages like I want, but the style formatting (grid, background, etc.) is lost, and the legend is pushed outside the plot area and cropped out.

Why does the plot change, and how do I fix the script so the plot retains my legend and styling?

import seaborn as sns
import seaborn.objects as so
import matplotlib.ticker as ticker
import matplotlib.pyplot as plt

market_data = {
    'Year': ['2001','2001','2002', '2002', '2003', '2003'],
    'Company': ['Shony','Ponysonic','Shony','Ponysonic','Shony','Ponysonic'],
    'Market Share': [0.15, 0.35, 0.25, 0.4, 0.2, 0.45]
}

market_chart = (
    so.Plot(data=market_data, x="Year", y="Market Share",color="Company")
    .add(so.Area(), so.Stack())
)

market_chart.show()  # looks good, but y-axis not in percent format

fig, ax = plt.subplots()
market_chart.on(ax).plot()
ax.yaxis.set_major_formatter(ticker.PercentFormatter(1))
plt.show() # display with percent y-axis. Lost theme styling.

Silikazi · Accepted Answer · 2025-11-19 01:06:52Z

1

The plot changes because calling .on(ax).plot() disables Seaborne theme system, which is what applied the background grid, facecolor, legend layout, etc. The fix is to apply the percent formatter inside Seaborne theming context, after the plot is created, and without manually creating a new fig, ax.

import seaborn as sns
import seaborn.objects as so
import matplotlib.ticker as ticker
import matplotlib.pyplot as plt

market_data = {
    'Year': ['2001','2001','2002', '2002', '2003', '2003'],
    'Company': ['Shony','Ponysonic','Shony','Ponysonic','Shony','Ponysonic'],
    'Market Share': [0.15, 0.35, 0.25, 0.4, 0.2, 0.45]
}
market_chart = (
    so.Plot(data=market_data, x="Year", y="Market Share", color="Company")
    .add(so.Area(), so.Stack())
    .label(y="{value:.0%}")        # format ticks without Matplotlib
)

market_chart.show()

edited Nov 19 at 1:06

answered Nov 17 at 9:33

Silikazi

4803 silver badges13 bronze badges

Sign up to request clarification or add additional context in comments.

5 Comments

timelessbeing Nov 17 at 22:45

It didn't work for me. " ax = fig.axes[0] ^^^^^^^^ AttributeError: 'NoneType' object has no attribute 'axes'"

Silikazi Nov 18 at 5:30

I updated the code adding fig = plt.gcf() try again. The error occur because market_chart.show() is returning None

timelessbeing Nov 18 at 10:04

different error now: " ax = fig.axes[0] ~~~~~~~~^^^ IndexError: list index out of range"

Silikazi Nov 19 at 1:03

I have updated the answer without using matplotlib assuming you all you want is percentage and your theme.

timelessbeing Nov 19 at 5:27

doesn't work. y-axis values are decimal not percent.

Collectives™ on Stack Overflow

Format y-axis as percentage in a seaborn.objects plot

1 Answer 1

5 Comments

Your Answer

Hot Network Questions

Collectives™ on Stack Overflow

1 Answer 1

5 Comments

Your Answer

Sign up or log in

Post as a guest

Related