如何在seaborn条形图上添加点？

Question

How can I put points onto a seaborn bar plot? The following code produces the bar plot.

import matplotlib.pyplot as plt
import seaborn
import matplotlib
matplotlib.use('TkAgg')
seaborn.set_context('talk')

data_df = pandas.DataFrame([3, 1, 2, 4], index=['a', 'b', 'c', 'd']).transpose()
points_df = pandas.DataFrame([3.5, 0.5, 1.75, 4.25], index=['a', 'b', 'c', 'd']).transpose()
plt.figure()
seaborn.barplot(data=data_df)

plt.show()

How can I now add the data in points_df as red points on this graph?

Answer 1

Is this what you are looking for?

data_df = pd.DataFrame([3, 1, 2, 4], index=['a', 'b', 'c', 'd']).transpose()
points_df = pd.DataFrame([3.5, 0.5, 1.75, 4.25], index=['a', 'b', 'c', 'd']).transpose()

plt.figure()
sns.barplot(data=data_df)
sns.scatterplot(data=points_df.T, legend=False, zorder=10)

for some reason, I had to change the zorder of the points, otherwise some would be hidden behind the bars

Answer 2

You can use the pointplot from seaborn for plotting the red points using join=False and using zorder to bring the dots to the front. The key here is to define an axis instance ax and pass it to both the plots.

import matplotlib.pyplot as plt
import seaborn as sns
import pandas as pd
import matplotlib
matplotlib.use('TkAgg')
seaborn.set_context('talk')

fig, ax = plt.subplots()

data_df = pd.DataFrame([3, 1, 2, 4], index=['a', 'b', 'c', 'd']).transpose()
points_df = pd.DataFrame([3.5, 0.5, 1.75, 4.25], index=['a', 'b', 'c', 'd']).transpose()
sns.barplot(data=data_df, ax=ax, zorder=0)
sns.pointplot(data=points_df, join=False, color='red', ax=ax, zorder=1)