Customize Seaborn Heatmap x-axis and y-axis Tick Labels

In this tutorial, you’ll learn how to customize Seaborn heatmap tick labels, covering aspects like setting custom labels, adjusting label rotation and font size, automatic label rotation, and even hiding labels.

 

 

Set Custom Labels

You can use set_xticklabels and set_yticklabels methods to set custom labels for both axes.

First, import the necessary libraries and prepare your data:

import seaborn as sns
import matplotlib.pyplot as plt
import pandas as pd
import numpy as np
np.random.seed(0)
data = np.random.rand(4, 6)
metrics = ['Metric A', 'Metric B', 'Metric C', 'Metric D']
days = ['Monday', 'Tuesday', 'Wednesday', 'Thursday', 'Friday', 'Saturday']
df = pd.DataFrame(data, index=metrics, columns=days)

Now, let’s create a basic Seaborn heatmap:

plt.figure(figsize=(10, 4))
sns.heatmap(df, annot=True)
plt.show()

Output:

Next, customize the tick labels:

plt.figure(figsize=(10, 4))
ax = sns.heatmap(df, annot=True)
ax.set_xticklabels(['Mon', 'Tue', 'Wed', 'Thu', 'Fri', 'Sat'], rotation=45)
ax.set_yticklabels(['A', 'B', 'C', 'D'], rotation=0)
plt.show()

Output:

 

Rotate Labels

Rotating these labels can help avoid overlap and make each label distinct.

First, let’s update our dataset to include more days:

extended_days = days + ['Sunday', 'Extra Day 1', 'Extra Day 2']
extended_data = np.random.rand(4, len(extended_days))
df_extended = pd.DataFrame(extended_data, index=metrics, columns=extended_days)

Now, create a heatmap with these extended labels:

plt.figure(figsize=(12, 4))
sns.heatmap(df_extended, annot=True)
plt.show()

Output:

To rotate the labels, you can set rotation property to the rotation angel you want:

plt.figure(figsize=(12, 4))
ax = sns.heatmap(df_extended, annot=True)
ax.set_xticklabels(ax.get_xticklabels(), rotation=90)
ax.set_yticklabels(ax.get_yticklabels(), rotation=0)
plt.show()

Output:

 

 

Adjust Label Font Size

You can increase the font size of the labels using the fontsize property:

plt.figure(figsize=(12, 4))
ax = sns.heatmap(df_extended, annot=True)
plt.xticks(fontsize=12, rotation=90)
plt.yticks(fontsize=12, rotation=0)
plt.show()

Output:

 

Automatically Rotating Labels

You can apply automatic rotation using labelrotation property:

plt.figure(figsize=(12, 4))
ax = sns.heatmap(df_extended, annot=True)
ax.tick_params(axis='x', labelrotation=45)
ax.tick_params(axis='y', labelrotation=45)
plt.show()

Output:

 

Set Tick Label Alignment

You can adjust the alignment of the labels using ha property for horizontal alignment and va property for vertical alignment:

plt.figure(figsize=(12, 4))
ax = sns.heatmap(df_extended, annot=True)
ax.set_xticklabels(ax.get_xticklabels(), ha='center')  # Horizontal alignment
ax.set_yticklabels(ax.get_yticklabels(), va='center')  # Vertical alignment
plt.show()

Output:

 

Move x-axis Labels to Top

You can move the x-axis labels to the top using tick_top() function:

plt.figure(figsize=(12, 4))
ax = sns.heatmap(df_extended, annot=True)
ax.xaxis.tick_top()  # Move x-axis labels to top
ax.set_xticklabels(ax.get_xticklabels(), rotation=45, ha='left')
plt.show()

Output:

 

Order y-axis Labels

You can reorder the y-axis by reindexing the DataFrame with the new order:

new_order = ['Metric D', 'Metric C', 'Metric B', 'Metric A']  # Define new order
reordered_df = df_extended.reindex(new_order)
plt.figure(figsize=(12, 4))
sns.heatmap(reordered_df, annot=True)
plt.show()

Output:

 

Reverse Axis

To reverse the y-axis, you can use invert_ yaxis function:

# Reversing the y-axis
plt.figure(figsize=(12, 4))
ax = sns.heatmap(df_extended, annot=True)
ax.invert_yaxis()
plt.show()

Output:

The output heatmap is similar to the above ordered one.

 

Hide Labels

You can hide labels by sending an empty array to xticks or yticks:

plt.figure(figsize=(12, 4))
ax = sns.heatmap(df_extended, annot=True)
plt.xticks([])
plt.yticks([])
plt.show()

Output: