Update app.py

app.py

@@ -396,6 +396,152 @@ elif st.session_state.current_page == "EDA":
     if st.button("🔙 Go Back to Model Report"):
         switch_page("Model Report") 
         
+
+# Model Building
+elif st.session_state.current_page == "Model Building":
+    st.markdown("<h2 style='text-align: center;'>Model Building</h2>", unsafe_allow_html=True)
+
+    # Introduction
+    st.markdown("""
+        <h5>📌 Introduction</h5>  
+        In this section, we explore different **Ensemble Learning** techniques to improve model performance.  
+        We implemented three ensemble models:  
+        - 🏆 <b>Voting Regressor</b>  
+        - 🎯 <b>Bagging Regressor</b>  
+        - 🌲 <b>Random Forest Regressor</b>  
+    """, unsafe_allow_html=True)
+    st.markdown("<br>", unsafe_allow_html=True)
+
+    # Voting Regressor
+    st.markdown("""
+        <h5>1️⃣ Voting Regressor</h5>  
+        🔹 **Concept:** Combines multiple models (**KNN & Decision Tree**) and takes the **average prediction**.  
+        🔹 **Why Voting Regressor?**  
+        - ✅ Works well when models have different strengths.  
+        - ✅ Reduces variance while maintaining interpretability.  
+    """, unsafe_allow_html=True)
+    st.markdown("<br>", unsafe_allow_html=True)
+
+    # Bagging Regressor
+    st.markdown("""
+        <h5>2️⃣ Bagging Regressor</h5>  
+        🔹 **Concept:** Uses **bootstrap sampling** to train multiple models on different subsets of data.  
+        🔹 **Why Bagging Regressor?**  
+        - ✅ Reduces overfitting by averaging multiple models.  
+        - ✅ Works best with **high-variance models** like Decision Tree.  
+    """, unsafe_allow_html=True)
+    st.markdown("<br>", unsafe_allow_html=True)
+
+    # Random Forest Regressor
+    st.markdown("""
+        <h5>3️⃣ Random Forest Regressor</h5>  
+        🔹 **Concept:**  
+        - Uses **multiple Decision Trees**, trained on different feature subsets.  
+        - The final prediction is the **average of all tree predictions**.  
+        🔹 **Why Random Forest?**  
+        - ✅ Handles **non-linearity** well.  
+        - ✅ Less prone to overfitting compared to a single Decision Tree.  
+    """, unsafe_allow_html=True)
+    st.markdown("<br>", unsafe_allow_html=True)
+
+    st.markdown("""
+        <h5>⚖️ Combining High & Low Variance Models</h5>  
+        A crucial step to improve ensemble performance is **choosing models with different variance levels:**  
+        - **Voting Regressor:** Uses a combination of **high-variance (Decision Tree, KNN with small K)** and **low-variance (KNN with large K, Decision Tree with depth constraint)** models.  
+        - **Bagging & Random Forest:** Use **only high-variance models** (Decision Trees with deep splits) to maximize variance reduction.  
+
+        This technique helps create a **balanced ensemble**, preventing excessive overfitting or underfitting! ✅  
+    """, unsafe_allow_html=True)
+    
+    st.markdown("<br>", unsafe_allow_html=True)
+
+    # Hyperparameter Tuning
+    st.markdown("""
+        <h5>⚡ Hyperparameter Tuning using Optuna</h5>  
+        We optimized hyperparameters for **KNN, Decision Tree, Bagging Regressor, and Random Forest** using **Optuna**.  
+        Below are the **optimized parameters** for each model:  
+
+        ### **🔹 K-Nearest Neighbors (KNN)**
+        - `n_neighbors`  
+        - `p` (Distance metric)  
+        - `weights` 
+        - `algorithm`  
+
+        ### **🔹 Decision Tree**
+        - `max_depth`  
+        - `min_samples_split`  
+        - `min_samples_leaf` 
+        - `max_features`  
+        - `min_impurity_decrease` 
+
+        ### **🔹 Bagging Regressor**
+        - `n_estimators`: 10 to 50  
+        - `max_samples`: 0.7 to 0.9  
+
+        ### **🔹 Random Forest**
+        - `n_estimators`: 10 to 50  
+        - `max_samples`: 0.7 to 0.9  
+    """, unsafe_allow_html=True)
+    st.markdown("<br>", unsafe_allow_html=True)
+
+    # Model Performance Insights
+    st.markdown("""
+        <h5>📊 Model Performance Insights</h5>  
+    st.markdown("<br>", unsafe_allow_html=True)
+
+    # Model Performance Table
+    st.markdown("""
+        <style>
+            table {
+                width: 100%;
+                border-collapse: collapse;
+                text-align: center;
+            }
+            th, td {
+                padding: 10px;
+                border-bottom: 1px solid #ddd;
+            }
+        </style>
+        <table>
+            <tr>
+                <th>Ensemble</th>
+                <th>Training Score</th>
+                <th>Test Score</th>
+                <th>Generalized Score</th>
+            </tr>
+            <tr>
+                <td>Voting Ensemble</td>
+                <td>95.8027%</td>
+                <td>92.1368%</td>
+                <td>92.89%</td>
+            </tr>
+            <tr>
+                <td>Bagging Ensemble</td>
+                <td>98.6861%</td>
+                <td>95.0407%</td>
+                <td>95.45%</td>
+            </tr>
+            <tr>
+                <td>Random Forest</td>
+                <td>97.9244%</td>
+                <td>97.9244%</td>
+                <td><b>94.71%</b></td>
+            </tr>
+        </table>
+    """, unsafe_allow_html=True)
+    st.markdown("<br>", unsafe_allow_html=True)
+
+    # Choosing the Best Model
+    st.markdown("""
+        <h5>🏆 Choosing the Best Model</h5>  
+        - We checked for **overfitting** (high training accuracy, low test accuracy).  
+        - We avoided **underfitting** (low training and test accuracy).  
+        - The best model had a **balanced performance across training and test data**.  
+
+        ✅ **Final Choice: Bagging Ensemble** due to its strong generalization ability! 🚀  
+    """, unsafe_allow_html=True)
+
+        
 # Hands-on Model Page
 elif st.session_state.current_page == "Hands-on Model":
     st.title("Hands-on Model")