Plotting seaborn histplot bar_label with condition

Question

I want to plot a seaborn histogram with labels to show the values of each bar. I only want to show the non-zero values, but I'm not sure how to do it. My MWE is

import numpy as np
import matplotlib.pyplot as plt
import seaborn as sns

xlist = 900+200*np.random.randn(50,1)

fig, ax = plt.subplots()
y = sns.histplot(data=xlist, element="bars", bins=20, stat='count', legend=False)
y.set(xlabel='total time (ms)')
y.bar_label(y.containers[0])
## y.bar_label(y.containers[0][y.containers[0]!=0])
plt.show()

The graph looks like

and I want to remove all the 0 labels.

possibly related https://stackoverflow.com/questions/67112449/bar-labels-in-matplotlib-seaborn — matec, Jan 11 '22 at 21:48
not really related... I just want to get rid of all the `0` labels — Medulla Oblongata, Jan 11 '22 at 21:51
It means you can iterate over `y.containers` and use `ax.text` to manually label the bars yourself. — Quang Hoang, Jan 11 '22 at 21:53

Corralien · Accepted Answer · 2022-01-11T22:10:39.333

4

Update

A best version suggested by @BigBen:

labels = [str(v) if v else '' for v in y.containers[0].datavalues]
y.bar_label(y.containers[0], labels=labels)

Try:

labels = []
for p in y.patches:
    h = p.get_height()
    labels.append(str(h) if h else '')

y.bar_label(y.containers[0], labels=labels)

edited Jan 11 '22 at 22:10

answered Jan 11 '22 at 21:59

Corralien

109,409
8
28
52

Plotting seaborn histplot bar_label with condition

1 Answers1