Update app.py

app.py

@@ -138,7 +138,7 @@ def batch_predict(app_choice, file_obj):
 
 # --- 5. AUTONOMOUS DESIGN LOGIC ---
     
-def generate_and_screen(num_to_gen, app_mode):
+def generate_and_screen(num_to_gen, app_mode, toggle_view):
     candidates = []
     generated_fps = []
     max_attempts = num_to_gen * 15 
@@ -196,9 +196,9 @@ def generate_and_screen(num_to_gen, app_mode):
                 "_s": s, "_t": t 
             })
             
-    df = pd.DataFrame(candidates).sort_values(by="_s", ascending=False)
+    df = pd.DataFrame(candidates)
     
-# 1. Pareto Frontier Logic
+    # PARETO CALCULATION
     df_sorted = df.sort_values(by="_s")
     px_pts, py_pts, cur_max = [], [], -1000
     pareto_indices = []
@@ -207,15 +207,19 @@ def generate_and_screen(num_to_gen, app_mode):
             px_pts.append(row["_s"]); py_pts.append(row["_t"]); cur_max = row["_t"]
             pareto_indices.append(i)
     
-    # Toggle filtering: If Multi-objective chosen, only show Pareto optimal rows
+    # APPLY THE TOGGLE logic (Point 1 from Professor)
     if toggle_view == "Multi-objective Pareto optimal":
         df = df.loc[pareto_indices]
 
-    fig = px.scatter(df_sorted, x="_s", y="_t", color="Classification", color_discrete_map={"Excellent": "#2ecc71", "Review": "#f1c40f", "Bad": "#e74c3c"},
-                     title=f"Pareto: {app_mode}", labels={"_s": "Strength (GPa)", "_t": "Tg (°C)"})
+    # Create the Plot
+    fig = px.scatter(df_sorted, x="_s", y="_t", color="Classification", 
+                     color_discrete_map={"Excellent": "#2ecc71", "Review": "#f1c40f", "Bad": "#e74c3c"},
+                     title=f"Pareto Analysis: {app_mode}", labels={"_s": "Strength (GPa)", "_t": "Tg (°C)"})
     fig.add_trace(go.Scatter(x=px_pts, y=py_pts, mode='lines', name='Pareto Frontier', line=dict(color='black', dash='dash')))
 
+    # Final sort for the table
     df = df.sort_values(by="_s", ascending=False)
+    
     output_path = tempfile.NamedTemporaryFile(delete=False, suffix=".csv").name
     df.to_csv(output_path, index=False)
     return df.drop(columns=["_s", "_t"]), output_path, fig
@@ -223,13 +227,19 @@ def generate_and_screen(num_to_gen, app_mode):
 def calculate_weighted_rank(df, app_mode):
     if df is None or df.empty: return None, None
     def clean(x): return float(str(x).split(' ')[1])
-    s_n, t_n, d_n = (df['Strength (GPa)'].apply(clean)-0.01)/0.14, (df['Tg (°C)'].apply(clean)-50)/250, df['Degradability'].apply(clean)
-    sa_p = np.exp(-0.4 * (df['SA Score'].apply(clean) - 1))
     
-    # Point 5: Weight adjustment per mode
-    if app_mode == "Aerospace Mode": df['Rank Score'] = (s_n * 0.5 + t_n * 0.3 + sa_p * 0.2).round(4)
-    elif app_mode == "Battery Mode": df['Rank Score'] = (t_n * 0.5 + s_n * 0.2 + sa_p * 0.3).round(4)
-    else: df['Rank Score'] = (d_n * 0.6 + s_n * 0.1 + sa_p * 0.3).round(4)
+    # Standardize normalization for all properties
+    s_norm = (df['Strength (GPa)'].apply(clean) - 0.01) / 0.14
+    t_norm = (df['Tg (°C)'].apply(clean) - 50) / 250
+    d_norm = df['Degradability'].apply(clean) # Degradability is usually 0-1, so no norm needed
+    sa_penalty = np.exp(-0.4 * (df['SA Score'].apply(clean) - 1))
+    
+    if app_mode == "Aerospace Mode":
+        df['Rank Score'] = (s_norm * 0.5 + t_norm * 0.3 + sa_penalty * 0.2).round(4)
+    elif app_mode == "Battery Mode":
+        df['Rank Score'] = (t_norm * 0.5 + s_norm * 0.2 + sa_penalty * 0.3).round(4)
+    else: # Biomedical Mode
+        df['Rank Score'] = (d_norm * 0.6 + s_norm * 0.1 + sa_penalty * 0.3).round(4)
 
     df = df.sort_values(by="Rank Score", ascending=False)
     output_path = tempfile.NamedTemporaryFile(delete=False, suffix=".csv").name