Barplot with continuous bars but broken y-axis

Question

I'm trying to do a barplot with python's matplotlib that has a broken y-axis to represent an outlier. But due to aesthetics, I need the bar of the outlier to be continuous, not to be broken. Any idea on how I can do this would be really appreciated. Reducing the gap between subplots is not an option.

I've learned how to do broken-axis plots using matplotlib docs but I cannot find anything like I need to do

This is an example code

# %% create a broken-axis barplot
import matplotlib.pyplot as plt
import numpy as np

# create fake data
x = np.arange(1, 6)
y = np.exp(x)

# add outlier
y[2] = 15000

# bar plot with broken axis
f, (ax, ax2) = plt.subplots(2,1, sharex=True, facecolor='w')

ax.bar(x, y)
ax2.bar(x, y)

# set limits
ax2.set_ylim(0, 100)
ax.set_ylim(10000, 20000)

# hide the spines between ax and ax2
ax.spines['bottom'].set_visible(False)
ax2.spines['top'].set_visible(False)

# hide ticks
ax.xaxis.tick_top()

plt.savefig('broken-axis-barplot.png', dpi=100, bbox_inches='tight')

Basically bar 3 needs to be filled up between the two chunks

Answer 1

In your code, you have the ax which has data till 100 (the lower limit). To extend the bar outside the plot area, you can add , clip_on=False while plotting that bar, which will extend the bar outside the area of the first plot. So, basically, changing that line to...

ax.bar(x, y, clip_on=False)

...gives you the below plot