Update app.py

app.py

@@ -273,7 +273,7 @@ with tab1:
     st.markdown("""
     To optimize performance, we applied a "Two-Stream" strategy:
     1.  **Stream 1 (Linear Models):** Linear models (like Linear Regression) were trained on a feature set pruned using VIF to avoid multicollinearity.
-    2.  **Stream 2 (Tree-based Models):** More complex models (like Random Forest, Gradient Boosting) were trained on a comprehensive set of 157 features to capture non-linear relationships.
+    2.  **Stream 2 (Tree-based Models):** More complex models (like Random Forest, Gradient Boosting) were trained on a comprehensive set of 156 features to capture non-linear relationships.
     
     Our Champion Model is a **Stacking** model from Stream 2, which demonstrated superior performance.
     """)