Seaborn figure-level vs axes-level in Matplotlib - Practice Questions

Challenge - 5 Problems

❓ Predict Output

intermediate

Output of figure-level Seaborn plot

What does the following code output when run?

Matplotlib

import seaborn as sns
import matplotlib.pyplot as plt

iris = sns.load_dataset('iris')

p = sns.catplot(x='species', y='sepal_length', data=iris)
print(type(p))

A <class 'seaborn.axisgrid.PairGrid'>

B <class 'matplotlib.axes._subplots.AxesSubplot'>

C <class 'matplotlib.figure.Figure'>

D <class 'seaborn.axisgrid.FacetGrid'>

💡 Hint

Figure-level plots return a FacetGrid object.

Seaborn figure-level functions like catplot return a FacetGrid object that manages the entire figure and axes internally.

❓ Predict Output

intermediate

Output type of axes-level Seaborn plot

What is the type of object returned by this code?

Matplotlib

import seaborn as sns
import matplotlib.pyplot as plt

iris = sns.load_dataset('iris')

ax = sns.scatterplot(x='sepal_length', y='sepal_width', data=iris)
print(type(ax))

A <class 'seaborn.axisgrid.PairGrid'>

B <class 'matplotlib.axes._subplots.AxesSubplot'>

C <class 'matplotlib.figure.Figure'>

D <class 'seaborn.axisgrid.FacetGrid'>

💡 Hint

Axes-level functions return a matplotlib Axes object.

Seaborn axes-level functions like scatterplot return a matplotlib Axes object that you can further customize.

❓ data_output

advanced

Number of subplots in figure-level Seaborn plot with col parameter

How many subplots are created by this code?

Matplotlib

import seaborn as sns

iris = sns.load_dataset('iris')
g = sns.catplot(x='sepal_length', y='sepal_width', col='species', data=iris)
print(len(g.axes.flat))

A 2

B 1

C 3

D 0

💡 Hint

The col parameter creates one subplot per unique value in the column.

The 'species' column has 3 unique values, so catplot creates 3 subplots arranged in columns.

🔧 Debug

advanced

Why does this axes-level plot not show multiple subplots?

This code tries to create multiple subplots by species but only shows one plot. Why?

Matplotlib

import seaborn as sns
import matplotlib.pyplot as plt

iris = sns.load_dataset('iris')
fig, axes = plt.subplots(3, 1)

for ax, species in zip(axes, iris['species'].unique()):
    sns.scatterplot(x='sepal_length', y='sepal_width', data=iris[iris['species'] == species], ax=ax)

plt.show()

A The code works correctly and shows 3 subplots.

B The axes array is 1D but should be flattened before use.

C The sns.scatterplot ignores the ax parameter in this context.

D The plt.subplots call creates only one Axes, so loop fails.

💡 Hint

Check how plt.subplots returns axes when rows > 1 and cols = 1.

plt.subplots(3,1) returns a 1D array of 3 Axes, so the loop assigns each scatterplot to a different subplot correctly.

🚀 Application

expert

Choosing figure-level vs axes-level for complex multi-plot visualization

You want to create a grid of scatterplots showing relationships between multiple variables grouped by a categorical variable. Which approach is best?

A Use Seaborn's pairplot (figure-level) with the 'hue' parameter to create the grid automatically.

B Manually create matplotlib subplots and call sns.scatterplot (axes-level) on each subplot for each variable pair.

C Use sns.scatterplot (axes-level) once and rely on matplotlib's subplot adjustments.

D Use plt.plot directly without Seaborn for better control.

💡 Hint

Figure-level functions handle complex grids and grouping internally.

Seaborn's pairplot is a figure-level function designed to create grids of plots with grouping and coloring easily, reducing manual subplot management.