Update app.py

app.py

@@ -359,6 +359,7 @@ def predict_inverse(antimicrobial_activity_target, substrate, microorganism, num
     for column in targets:
         df_columns.remove(column)
     constrained_columns = ['Substrate', 'Washing_cycles', 'Microorganism ']
+    not_constrained_columns = [column for column in df_columns if column not in constrained_columns]
     ### Add input constraints
     for c in test_data_columns:
         if c.startswith('Substrate'):
@@ -373,14 +374,15 @@ def predict_inverse(antimicrobial_activity_target, substrate, microorganism, num
                 domain.append({'name': str(c), 'type': 'categorical', 'domain': (0.0, 0.0)})
         elif c == 'Washing_cycles':
             domain.append({'name': str(c), 'type': 'categorical', 'domain': (int(num_washing_cycles), int(num_washing_cycles))})
-    for df_column in df_columns:
-        if df_column.startswith('Substrate') or df_column.startswith('Washing_cycles') or df_column.startswith('Microorganism'):
-            continue
-        if df_column in numerical_columns:
-            domain.append({'name': str(df_column), 'type': 'continuous', 'domain': (0.,1.)})
+        elif c in numerical_columns:
+            domain.append({'name': str(c), 'type': 'continuous', 'domain': (0.,1.)})
         else:
-            domain.append({'name': str(df_column), 'type': 'categorical', 'domain': (0,1),
-                           'dimensionality': dimensionality_dict[df_column]})
+            seen_not_constrained = []
+            for not_c in not_constrained_columns:
+                if c.startswith(not_c) and not_c not in seen_not_constrained:
+                    domain.append({'name': str(not_c), 'type': 'categorical', 'domain': (0,1),
+                                   'dimensionality': dimensionality_dict[not_c]})
+                    seen_not_constrained.append(not_c)
     print("Domain is ", domain)
     print(len(domain))