Function cluster_visualization

nav/clustering.py:78–116

(regional_features, X_scaled, feature_cols)

        st.metric("Largest", f"{largest_cluster_size} regions")

def cluster_visualization(regional_features, X_scaled, feature_cols):
    col1, col2 = st.columns(2)
    
    with col1:
        if len(X_scaled) > 1:
            pca = PCA(n_components=2)
            X_pca = pca.fit_transform(X_scaled)
            
            pca_df = pd.DataFrame({
                'PC1': X_pca[:, 0],
                'PC2': X_pca[:, 1],
                'Region': regional_features['REGIONAL_OFFICE_NAME'],
                'Cluster': regional_features['Cluster'],
                'Total_Cost': regional_features['Total_Cost']
            })
            
            fig_pca = px.scatter(
                pca_df,
                x='PC1',
                y='PC2',
                color='Cluster',
                size='Total_Cost',
                hover_name='Region',
                title="Regional Clusters",
                size_max=40
            )
            fig_pca.update_layout(height=400)
            st.plotly_chart(fig_pca, use_container_width=True)
    
    with col2:
        cluster_summary = regional_features.groupby('Cluster')[feature_cols].mean().round(2)
        
        display_summary = cluster_summary.copy()
        display_summary['Total_Cost'] = display_summary['Total_Cost'].apply(lambda x: f"\u00a3{x:,.0f}")
        display_summary['Mean_Cost'] = display_summary['Mean_Cost'].apply(lambda x: f"\u00a3{x:,.0f}")
        display_summary['Cost_Per_Record'] = display_summary['Cost_Per_Record'].apply(lambda x: f"\u00a3{x:,.0f}")
        display_summary.columns = ['Total', 'Mean', 'Var', 'Per Record']
        
        st.dataframe(display_summary, use_container_width=True)

def geographic_distribution(regional_features):
    map_data = []