Visualisation

Now that we have a sense of the numbers, let’s visualise the relationships.

Pairplot (The “Big Picture”)

A Pairplot allows us to see the relationship between every variable at once. This helps us spot clusters and potential correlations.

# hue="species" colors the dots so we can spot clusters
# corner=True removes the duplicate charts to keep it clean
sns.pairplot(df_clean, hue="species", corner=True, height=1.5)
plt.show()
Figure 1: Pairplot of Penguin Dimensions

Correlation Heatmap

Finally, let’s quantify how strongly these variables are related.

  • 1.0 = Perfect positive correlation (move together)

  • -1.0 = Perfect negative correlation (move opposite)

  • 0 = No relationship

This matrix is crucial for deciding which variables to include in a regression model.

# Calculate the correlation matrix
corr_matrix = df_clean.corr(numeric_only=True)

# Plot the heatmap
plt.figure(figsize=(6, 5))
sns.heatmap(corr_matrix, annot=True, cmap="coolwarm", vmin=-1, vmax=1, fmt=".2f")
plt.title("Correlation Matrix")
plt.show()
Figure 2: Correlation Heatmap of Numeric Variables