Heatmaps and Pairplots for Correlation Analysis

Correlation Visualizations

Heatmaps and pairplots help you discover relationships among many variables.

Correlation matrix

corr = df.corr()
print(corr)

Heatmap with seaborn

sns.heatmap(corr, annot=True, fmt='.2f', cmap='coolwarm', linewidths=0.5)
plt.title('Correlation Heatmap')
plt.show()

Pairplot (scatter matrix)

sns.pairplot(df, diag_kind='kde', hue='target')
plt.show()

Annotating the heatmap with significance stars (optional)

import scipy.stats as stats
p_vals = df.corr(method=lambda x, y: stats.pearsonr(x, y)[1])
mask = np.triu(np.ones_like(corr, dtype=bool))
sns.heatmap(corr, mask=mask, cmap='viridis', annot=True, fmt='.2f')
plt.show()

Tips

• Use mask to hide the upper triangle (redundant).
• For large datasets, consider sns.clustermap to cluster similar variables.