Update app.py

app.py

@@ -213,16 +213,68 @@ if show_regression and selected_feature in df.columns and selected_target == 'Me
 
 st.pyplot(fig)
 
+# Simple Linear Regression
+X_single = df[['MedInc']]
+y = df['MedHouseVal']
+
+X_train_single, X_test_single, y_train_single, y_test_single = train_test_split(X_single, y, test_size=0.2, random_state=42)
+
+model_single = LinearRegression()
+model_single.fit(X_train_single, y_train_single)
+
+y_pred_single = model_single.predict(X_test_single)
+
+r2_single = r2_score(y_test_single, y_pred_single)
+
+# Plot the regression line for simple linear regression
+fig, ax = plt.subplots()
+ax.scatter(X_test_single, y_test_single, color='blue', alpha=0.3, label='Actual')
+ax.plot(X_test_single, y_pred_single, color='red', linewidth=2, label='Predicted')
+ax.set_title('Simple Linear Regression: MedInc vs MedHouseVal')
+ax.set_xlabel('MedInc')
+ax.set_ylabel('MedHouseVal')
+ax.legend()
+st.pyplot(fig)
+
+# Multilinear Regression
+X_multi = df[['MedInc', 'AveRooms', 'HouseAge', 'AveOccup']]
+y = df['MedHouseVal']
+
+X_train_multi, X_test_multi, y_train_multi, y_test_multi = train_test_split(X_multi, y, test_size=0.2, random_state=42)
+
+model_multi = LinearRegression()
+model_multi.fit(X_train_multi, y_train_multi)
+
+y_pred_multi = model_multi.predict(X_test_multi)
+
+r2_multi = r2_score(y_test_multi, y_pred_multi)
+
+# Add checkbox for multilinear regression plot
+show_multilinear_plot = st.checkbox('Show Multilinear Regression Plot')
+
+if show_multilinear_plot:
+    fig, ax = plt.subplots()
+    ax.scatter(y_test_multi, y_pred_multi, color='blue', alpha=0.3)
+    ax.plot([y_test_multi.min(), y_test_multi.max()], [y_test_multi.min(), y_test_multi.max()], 'k--', lw=2, color='green')
+    ax.set_title('Multilinear Regression: Actual vs. Predicted MedHouseVal')
+    ax.set_xlabel('Actual MedHouseVal')
+    ax.set_ylabel('Predicted MedHouseVal')
+    st.pyplot(fig)
+
+# Compare R-squared values
+st.subheader('R-squared Comparison')
+st.write(f"Simple Linear Regression R-squared: {r2_single:.4f}")
+st.write(f"Multilinear Regression R-squared: {r2_multi:.4f}")
+
 # Prediction
 st.subheader('Predict Median House Value')
 
-# Input values for prediction
 input_values = {}
-for feature in X.columns:
+for feature in X_multi.columns:
     input_values[feature] = st.number_input(f'Enter {feature}', value=float(df[feature].mean()))
 
 if st.button('Predict'):
     input_data = np.array([list(input_values.values())])
-    prediction = model.predict(input_data)
+    prediction = model_multi.predict(input_data)
     st.write(f'Predicted Median House Value: {prediction[0]}')