Learn Creating Swarm Plots

The swarmplot is very similar to the stripplot, but it has a major advantage: points do not overlap.

In a stripplot, we use "jitter" to randomly scatter points so they don't sit on top of each other, but overlaps can still happen. The swarmplot uses a specific algorithm to adjust the points along the categorical axis so that they form a clear, non-overlapping shape that resembles the distribution of the data (similar to a violin plot).

Key Parameters

dodge=True: when using a hue variable (e.g., separating smokers vs. non-smokers), this parameter separates the groups into distinct "swarms" side-by-side, rather than mixing them;
linewidth: adds a border around each point, making them distinct even if they are small.

Example

Here is how dodge changes the visualization. Notice how the blue and orange points are separated.


              1234567891011121314151617
            
import seaborn as sns
import matplotlib.pyplot as plt

# Load dataset
df = sns.load_dataset('tips')

# Create a swarmplot
sns.swarmplot(
    data=df,
    x='day',
    y='total_bill',
    hue='sex',
    dodge=True,   # Separate Male/Female into side-by-side swarms
    size=4
)

plt.show()

Task

Swipe to start coding

Visualize the tips distribution using a swarmplot with distinct styling.

Set the style to 'whitegrid'. Pass a dictionary to set the 'axes.facecolor' to 'seashell'.
Create a swarmplot using the tips dataset (df):
- Map 'day' to x, 'total_bill' to y, and 'sex' to hue.
- Set the point size to 2 to fit more points without crashing into each other.
- Add a border to the points using linewidth=1.
- Separate the categories (male/female) side-by-side by setting dodge=True.
- Use the 'rocket' palette.
Display the plot.

Solution

Everything was clear?

Thanks for your feedback!

Section 1. Chapter 10

single

Ask AI

Ask anything or try one of the suggested questions to begin our chat

Swipe to show menu