How to create a Triangle Correlation Heatmap in seaborn - Python?

Creating a Triangle Correlation Heatmap in Seaborn in Python

When exploring data relationships, correlation heatmaps are great—but they can display mirrored, redundant information. A triangle heatmap—showing either the upper or lower half—provides a cleaner view. In this guide, we’ll demonstrate how to build and customize a triangle correlation heatmap using Seaborn.

Why Use a Triangle Heatmap?

Standard correlation heatmaps repeat values symmetrically across the diagonal. Masking one triangle (upper or lower) declutters the plot, focuses attention on unique pairs, and simplifies interpretation—ideal for presentations and exploratory data analysis.

Step 1: Prepare the Correlation Matrix

import pandas as pd
import seaborn as sns
import numpy as np
import matplotlib.pyplot as plt

# Load dataset and compute correlation
df = sns.load_dataset("iris")
corr = df.corr()

This computes Pearson correlations among numeric columns of the 'iris' dataset.

Step 2: Generate a Mask for the Triangle

# Generate a boolean mask for the upper triangle
mask = np.triu(np.ones_like(corr, dtype=bool))

The mask sets the upper triangle to True, hiding it from the plot.

Step 3: Plot the Triangle Heatmap

plt.figure(figsize=(8, 6))
sns.heatmap(corr, mask=mask, annot=True, cmap="coolwarm", 
            fmt=".2f", linewidths=0.5, vmin=-1, vmax=1)

plt.title("Triangle Correlation Heatmap (Lower Triangle Only)")
plt.show()

This renders only the lower triangular half with annotations and diverging colors, and adds grid lines between cells.

Step 4: Customize the Appearance

You can enhance the plot further by adjusting styling:

plt.figure(figsize=(9, 7))
sns.set_style("white")

sns.heatmap(corr, mask=mask, annot=True, cmap="RdBu_r",
            fmt=".2f", linewidths=1, cbar_kws={"shrink":0.5})

plt.title("Styled Triangle Correlation Heatmap")
plt.show()

• cmap="RdBu_r": provides a bold diverging palette.
• linewidths=1: increases cell separation.
• cbar_kws={"shrink":0.5}: resizes the color bar.

Step 5: Interpret the Output

Read the heatmap by noting the intensity and sign of colors:

• Dark red/blue: strong positive or negative correlation.
• Light or near-neutral shades: weak relationships.
• Only unique pairs are displayed, making it easier to find linear dependencies or select non-redundant features.

Conclusion

Triangle correlation heatmaps using Seaborn provide a polished and efficient way to explore variable relationships by removing redundancy and enhancing clarity. With easy masking, styling, and annotation options, you can create intuitive visualizations that support better analytical decision-making.