Seaborn Bar Plot Tick Label Customization in Python

In this tutorial, you’ll learn how to customize tick labels in Seaborn bar plot.

We’ll cover basic label adjustments, formatting date-time labels, and more.

 

 

Differentiating between Axes and Figures

Axes

In Seaborn, an Axes refers to an individual plot or graph. This is the element where the actual data visualization takes place. You can think of it as the canvas on which your data is painted.

Figures

On the other hand, a Figure in Seaborn acts like a container for one or more Axes. It’s the overall window or page on which everything is drawn. You can have multiple Axes (plots) in a single Figure.

You want to create a bar plot showing a specific metric across different categories.

First, we import the necessary libraries and prepare sample data:

import seaborn as sns
import matplotlib.pyplot as plt
import pandas as pd
data = {
    'Category': ['Plan A', 'Plan B', 'Plan C', 'Plan D'],
    'Metric': [200, 150, 300, 250]
}
df = pd.DataFrame(data)

Now, let’s create a bar plot:

sns.barplot(x='Category', y='Metric', data=df)
plt.show()

Output:

This bar plot is an Axes object. It’s contained within a Figure, which is created by Seaborn/matplotlib.

 

Accessing and Modifying X-axis and Y-axis Tick Labels

Accessing Tick Labels

The first step in customization is accessing the tick labels. In Seaborn, this is done through the Axes object.

Continuing from the previous example:

ax = sns.barplot(x='Category', y='Metric', data=df)
x_labels = ax.get_xticklabels()
y_labels = ax.get_yticklabels()
print("X-axis labels:", x_labels)
print("Y-axis labels:", y_labels)

Output:

X-axis labels: [Text(0, 0, 'Plan A'), Text(1, 0, 'Plan B'), Text(2, 0, 'Plan C'), Text(3, 0, 'Plan D')]
Y-axis labels: [Text(0, 0.0, '0'), Text(0, 50.0, '50'), Text(0, 100.0, '100'), Text(0, 150.0, '150'), Text(0, 200.0, '200'), Text(0, 250.0, '250'), Text(0, 300.0, '300'), Text(0, 350.0, '350')]

The get_xticklabels() and get_yticklabels() methods retrieve the current labels as Text objects.

 

Formatting Tick Labels

Enhancing the visual appeal and clarity of your Seaborn bar plots involves more than just changing the text of the tick labels.

You can also format these labels by adjusting their font size, rotation, and color.

This level of customization is vital for creating plots that are not only informative but also engaging.

Adjust Font Size

You can set the font size using fontsize property:

ax.set_xticklabels(ax.get_xticks(), fontsize=11)
ax.set_yticklabels(ax.get_yticks(), fontsize=14)
plt.show()

Output:

Rotate Labels

You can rotate the labels using the rotation property:

ax.set_xticklabels(new_x_labels, rotation=30)
plt.show()

Output:

Change Label Color

You can change the label color using the color property:

ax.set_xticklabels(ax.get_xticks(), color="blue")
ax.set_yticklabels(ax.get_yticks(), color="green")
plt.show()

Output:

 

Using Custom Functions to Format Labels

Let’s create a function to format our x-axis labels. In this example, we’ll use a function that abbreviates longer words in our dataset’s category names.

def abbreviate_label(label):
    if len(label.get_text()) > 4:
        return label.get_text()[:4] + '.'
    else:
        return label.get_text()
ax.set_xticklabels([abbreviate_label(label) for label in ax.get_xticklabels()])
plt.show()

Output:

You can also use lambda functions to format y-axis tick labels based on the value.

import seaborn as sns
import matplotlib.pyplot as plt
import pandas as pd
data = {
    'Category': ['Plan A', 'Plan B', 'Plan C', 'Plan D'],
    'Metric': [2000, 1500, 3000, 2500]
}
df = pd.DataFrame(data)
ax = sns.barplot(x='Category', y='Metric', data=df)
ax.set_yticklabels(['{:.0f}k'.format(y/1000) for y in ax.get_yticks()])
plt.show()

Output:

 

Conditional Formatting Based on Data Values

We can make the label red if the value is above a certain threshold, indicating higher importance:

def format_ytick_label(label, threshold=200):
    value = int(label.get_text())
    color = 'red' if value > threshold else 'black'
    return {'color': color, 'weight': 'bold' if color == 'red' else 'normal'}
formatted_labels = [format_ytick_label(label) for label in ax.get_yticklabels()]
for i, label in enumerate(ax.get_yticklabels()):
    label.set_color(formatted_labels[i]['color'])
    label.set_weight(formatted_labels[i]['weight'])
plt.show()

Output:

 

Format Date-Time Values on Axes

Let’s begin by importing the necessary libraries and creating sample date-time data.

import seaborn as sns
import matplotlib.pyplot as plt
import pandas as pd
import datetime
data = {
    'Date': [datetime.datetime(2023, 1, 10), datetime.datetime(2023, 1, 11), 
             datetime.datetime(2023, 1, 12), datetime.datetime(2023, 1, 13)],
    'Metric': [200, 220, 180, 250]
}
df = pd.DataFrame(data)

When plotting, Seaborn will automatically handle the date-time formatting, but it might not be in the most readable format.

sns.barplot(x='Date', y='Metric', data=df)
plt.show()

Output:

To format these date-time labels, we can use strftime from datetime module:

import datetime
df['Date'] = pd.to_datetime(df['Date'])  # Convert Date column to datetime data type
df['Date'] = df['Date'].dt.strftime('%Y-%m-%d')
sns.barplot(x='Date', y='Metric', data=df)
plt.xticks(rotation=30)
plt.show()

Output: