Update app.py

app.py

@@ -81,50 +81,92 @@ os.makedirs(app.config['MODEL_FOLDER'], exist_ok=True)
 
 # Prediction analysis models loaded from Hugging Face.
 
+
+#classsification model on the task
 src_path = hf_hub_download(
     repo_id="WebashalarForML/Diamond_model_",
-    filename="models_list/mkble/DecisionTree_best_pipeline_mkble_0_to_0.99_al.pkl",
+    filename="CLASS_DUMMY/LR_best_2_clas_shp_change.pkl",
     cache_dir=MODEL_FOLDER
 )
-dst_path = os.path.join(MODEL_FOLDER, "DecisionTree_best_pipeline_mkble_0_to_0.99_al.pkl")
+dst_path = os.path.join(MODEL_FOLDER, "LR_best_2_clas_shp_change.pkl")
 shutil.copy(src_path, dst_path)
-makable_model_0 = load(dst_path)
+shape_change = load(dst_path)
+
 
 src_path = hf_hub_download(
     repo_id="WebashalarForML/Diamond_model_",
-    filename="models_list/mkble/DecisionTree_best_pipeline_mkble_1_to_1.49.pkl",
+    filename="CLASS_DUMMY/LR_best_2_class_blk(M)_change.pkl",
     cache_dir=MODEL_FOLDER
 )
-dst_path = os.path.join(MODEL_FOLDER, "DecisionTree_best_pipeline_mkble_1_to_1.49.pkl")
+dst_path = os.path.join(MODEL_FOLDER, "LR_best_2_class_blk(M)_change.pkl")
 shutil.copy(src_path, dst_path)
-makable_model_1 = load(dst_path)
-
+blk_change = load(dst_path)
+'''
 # Prediction analysis models loaded from Hugging Face.
 src_path = hf_hub_download(
     repo_id="WebashalarForML/Diamond_model_",
-    filename="models_list/mkble/DecisionTree_best_pipeline_mkble_1.50_to_1.99.pkl",
+    filename="CLASS_DUMMY/LR_best_2_class_wht(M)_change.pkl",
     cache_dir=MODEL_FOLDER
 )
-dst_path = os.path.join(MODEL_FOLDER, "DecisionTree_best_pipeline_mkble_1.50_to_1.99.pkl")
+dst_path = os.path.join(MODEL_FOLDER, "LR_best_2_class_wht(M)_change.pkl")
 shutil.copy(src_path, dst_path)
-makable_model_2 = load(dst_path)
+wht_change = load(dst_path)
+
+
+src_path = hf_hub_download(
+    repo_id="WebashalarForML/Diamond_model_",
+    filename="CLASS_DUMMY/LR_best_2_class_pav(M)_change.pkl",
+    cache_dir=MODEL_FOLDER
+)
+dst_path = os.path.join(MODEL_FOLDER, "LR_best_2_class_pav(M)_change.pkl")
+shutil.copy(src_path, dst_path)
+pav_change = load(dst_path)
+
+src_path = hf_hub_download(
+    repo_id="WebashalarForML/Diamond_model_",
+    filename="CLASS_DUMMY/LR_best_2_class_open(M)_change.pkl",
+    cache_dir=MODEL_FOLDER
+)
+dst_path = os.path.join(MODEL_FOLDER, "LR_best_2_class_open(M)_change.pkl")
+shutil.copy(src_path, dst_path)
+open_change = load(dst_path)
+
+# other param class change
+src_path = hf_hub_download(
+    repo_id="WebashalarForML/Diamond_model_",
+    filename="CLASS_DUMMY/LR_best_2_class_col_change.pkl",
+    cache_dir=MODEL_FOLDER
+)
+dst_path = os.path.join(MODEL_FOLDER, "LR_best_2_class_col_change.pkl")
+shutil.copy(src_path, dst_path)
+col_change = load(dst_path)
 
-#classsification model on the task
 src_path = hf_hub_download(
     repo_id="WebashalarForML/Diamond_model_",
-    filename="models_list/classification/3_pipeline.pkl",
+    filename="CLASS_DUMMY/LR_best_2_class_qua_change.pkl",
     cache_dir=MODEL_FOLDER
 )
-dst_path = os.path.join(MODEL_FOLDER, "3_pipeline.pkl")
+dst_path = os.path.join(MODEL_FOLDER, "LR_best_2_class_qua_change.pkl")
 shutil.copy(src_path, dst_path)
-mkble_amt_class_model = load(dst_path)
+qua_change = load(dst_path)
+
+src_path = hf_hub_download(
+    repo_id="WebashalarForML/Diamond_model_",
+    filename="CLASS_DUMMY/LR_best_2_class_cut_change.pkl",
+    cache_dir=MODEL_FOLDER
+)
+dst_path = os.path.join(MODEL_FOLDER, "LR_best_2_class_cut_change.pkl")
+shutil.copy(src_path, dst_path)
+cut_change = load(dst_path)
+
+'''
 
 #print("makable_model type:", type(makable_model))
 #print("grade_model type:", type(grade_model))
 #print("bygrade_model type:", type(bygrade_model))
 #print("gia_model type:", type(gia_model))
 print("================================")
-print("mkble_amt_class_model type:", type(mkble_amt_class_model))
+#print("mkble_amt_class_model type:", type(mkble_amt_class_model))
 
 # List of label encoder names.
 encoder_list = [
@@ -256,56 +298,52 @@ def process_dataframe(df):
         # -------------------------
         try:
         
-            # for model 0 to 0.99 
-            df_pred_0 = df_pred[(df_pred[['EngCts']] > 0.00).all(axis=1) & (df_pred[['EngCts']] < 0.99).all(axis=1)] 
-            df_pred_0['change_in_amt_mkble'] = pd.DataFrame(mkble_amt_class_model.predict(df_pred_0), columns=["pred_change_in_eng_to_mkble"])
+            # for model BLK CODE
+            df_pred_0 = df_pred.copy()
+            df_pred_0['Change_Blk_Eng_to_Mkbl_value'] = pd.DataFrame(blk_change.predict(df_pred), columns=["Change_Blk_Eng_to_Mkbl_value"])
             print(df_pred_0.columns)
-            df_pred_0 = df_pred_0[['Tag', 'EngCts', 'EngShp', 'EngQua', 'EngCol', 'EngCut', 'EngPol',
-                               'EngSym', 'EngFlo', 'EngNts', 'EngMikly', 'EngBlk', 'EngWht', 'EngOpen',
-                               'EngPav', 'EngAmt', 
-                               'change_in_amt_mkble'
-                              ]]
-            df_pred_0['Makable_Predicted'] = pd.DataFrame(np.expm1(makable_model_0.predict(df_pred_0)), columns=["Predicted"])
+
+            # for model WHT CODE
+            df_pred_0['Change_Wht_Eng_to_Mkbl_value'] = pd.DataFrame(shape_change.predict(df_pred), columns=["Change_Wht_Eng_to_Mkbl_value"])
+            print(df_pred_0.columns)
+            '''
+            # for model PAV CODE (need change)
+            df_pred_0['Change_Pav_Eng_to_Mkbl_value'] = pd.DataFrame(mkble_amt_class_model.predict(df_pred), columns=["Change_Pav_Eng_to_Mkbl_value"])
             print(df_pred_0.columns)
 
-            # for model 1 to 1.49
-
-            df_pred_1 = df_pred[(df_pred[['EngCts']] > 0.99).all(axis=1) & (df_pred[['EngCts']] < 1.50).all(axis=1)] 
-            df_pred_1['change_in_amt_mkble'] = pd.DataFrame(mkble_amt_class_model.predict(df_pred_1), columns=["pred_change_in_eng_to_mkble"])
-            print(df_pred_1.columns)
-            df_pred_1 = df_pred_1[['Tag', 'EngCts', 'EngShp', 'EngQua', 'EngCol', 'EngCut', 'EngPol',
-                               'EngSym', 'EngFlo', 'EngNts', 'EngMikly', 'EngBlk', 'EngWht', 'EngOpen',
-                               'EngPav', 'EngAmt', 
-                               'change_in_amt_mkble'
-                              ]]
-            df_pred_1['Makable_Predicted'] = pd.DataFrame(np.expm1(makable_model_1.predict(df_pred_1)), columns=["Predicted"])
-            print(df_pred_1.columns)
-
-            # for model 1.50 to 1.99
-
-            df_pred_2 = df_pred[(df_pred[['EngCts']] > 1.49).all(axis=1) & (df_pred[['EngCts']] < 2.00).all(axis=1)] 
-            df_pred_2['change_in_amt_mkble'] = pd.DataFrame(mkble_amt_class_model.predict(df_pred_2), columns=["pred_change_in_eng_to_mkble"])
-            print(df_pred_2.columns)
-            df_pred_2 = df_pred_2[['Tag', 'EngCts', 'EngShp', 'EngQua', 'EngCol', 'EngCut', 'EngPol',
-                               'EngSym', 'EngFlo', 'EngNts', 'EngMikly', 'EngBlk', 'EngWht', 'EngOpen',
-                               'EngPav', 'EngAmt', 
-                               'change_in_amt_mkble'
-                              ]]
-            df_pred_2['Makable_Predicted'] = pd.DataFrame(np.expm1(makable_model_2.predict(df_pred_2)), columns=["Predicted"])
-            print(df_pred_2.columns)
+            # for model OPEN CODE  (need change)
+            df_pred_0['Change_Open_Eng_to_Mkbl_value'] = pd.DataFrame(mkble_amt_class_model.predict(df_pred), columns=["Change_Open_Eng_to_Mkbl_value"])
+            print(df_pred_0.columns)
 
-            # Concatenate the DataFrames row-wise
-            df_pred_main = pd.concat([df_pred_0, df_pred_1, df_pred_2], ignore_index=True)
+            # for model SHP CODE (need change)
+            df_pred_0['Change_shape_value'] = pd.DataFrame(mkble_amt_class_model.predict(df_pred), columns=["Change_shape_value"])
+            print(df_pred_0.columns)
+
+            # for model COL CODE (need change)
+            df_pred_0['Change_color_value'] = pd.DataFrame(mkble_amt_class_model.predict(df_pred), columns=["Change_color_value"])
+            print(df_pred_0.columns)
 
-            # df_pred_main = pd.concat([df_pred_0, df_pred_1, df_pred_2])
-            print("df_pred_main------------->", df_pred_main.columns)
-            df_pred_main['Makable_Diff'] = df_pred_main['EngAmt'] - df_pred_main['Makable_Predicted']
-            print("df_pred_main------------->", df_pred_main.columns)
+            # for model CUT CODE (need change)
+            df_pred_0['Change_cut_value'] = pd.DataFrame(mkble_amt_class_model.predict(df_pred), columns=["Change_cut_value"])
+            print(df_pred_0.columns)
 
-            # Fill NaN or missing values in prediction columns
-            # df_pred_main[['change_in_amt_mkble', 'Makable_Predicted', 'Makable_Diff']] = df_pred_main[['change_in_amt_mkble', 'Makable_Predicted', 'Makable_Diff']].fillna("-")
-            
-            for col in ['Tag', 'EngShp', 'EngQua', 'EngCol', 'EngCut', 'EngPol', 'EngSym', 'EngFlo', 'EngNts', 'EngMikly','EngBlk', 'EngWht', 'EngOpen', 'EngPav']:
+            # for model QUA CODE (need change)
+            df_pred_0['Change_quality_value'] = pd.DataFrame(mkble_amt_class_model.predict(df_pred), columns=["Change_quality_value"])
+            print(df_pred_0.columns)            
+            '''
+            # Concatenate the DataFrames row-wise
+            #df_pred_main = pd.concat([df_pred_0, df_pred_1, df_pred_0], ignore_index=True)
+            df_pred_main = df_pred_0.copy()
+    
+            for col in ['Tag', 'EngShp', 'EngQua', 'EngCol', 'EngCut', 'EngPol', 'EngSym', 'EngFlo',
+                        'EngNts', 'EngMikly', 'EngLab','EngBlk', 'EngWht', 'EngOpen','EngPav',
+                        'Change_cts_value', 'Change_shape_value', 'Change_quality_value', 'Change_color_value',
+                        'Change_cut_value', 'Change_Blk_Eng_to_Mkbl_value', 'Change_Wht_Eng_to_Mkbl_value',
+                        'Change_Open_Eng_to_Mkbl_value', 'Change_Pav_Eng_to_Mkbl_value', 'Change_Blk_Eng_to_Grd_value',
+                        'Change_Wht_Eng_to_Grd_value', 'Change_Open_Eng_to_Grd_value', 'Change_Pav_Eng_to_Grd_value',
+                        'Change_Blk_Eng_to_ByGrd_value', 'Change_Wht_Eng_to_ByGrd_value', 'Change_Open_Eng_to_ByGrd_value',
+                        'Change_Pav_Eng_to_ByGrd_value', 'Change_Blk_Eng_to_Gia_value', 'Change_Wht_Eng_to_Gia_value',
+                        'Change_Open_Eng_to_Gia_value', 'Change_Pav_Eng_to_Gia_value']:
                 try:
                     df_pred_main[col] = loaded_label_encoder[col].inverse_transform(df_pred_main[col].astype(int))
                 except ValueError as e:
@@ -318,7 +356,6 @@ def process_dataframe(df):
         print("EngWht", df_pred_main['EngWht'].unique())
         print("EngOpen", df_pred_main['EngOpen'].unique())
         print("EngPav", df_pred_main['EngPav'].unique())
-        
            
         # Final return with full data for pagination.
         df_pred_main['EngBlk'] = df_pred_main['EngBlk'].fillna("-")
@@ -361,24 +398,8 @@ def report_view():
     
     df_page = df.iloc[start_idx:end_idx]
 
-    # # -------------------------------------V CHnages --------------------------------------
-    # # 👇 Inject arrows based on Makable_Diff
-    # def add_arrow(row):
-    #     try:
-    #         pred = float(row['Makable_Predicted'])
-    #         diff = float(row['Makable_Diff'])
-    #         if np.isnan(pred) or np.isnan(diff):
-    #             return '-'
-    #         arrow = '↑' if diff > 0 else '↓'
-    #         return f"{pred:.3f} {arrow}"
-    #     except:
-    #         return row['Makable_Predicted']
-    
-    # df_page['Makable_Predicted'] = df_page.apply(add_arrow, axis=1)
-    # # -------------------------------------V CHnages --------------------------------------
-
-
-    # -------------------------------------V Colored CHnages --------------------------------------
+     # -------------------------------------V Colored CHnages --------------------------------------
+    '''
     def add_colored_arrow(row):
         try:
             pred = float(row['Makable_Predicted'])
@@ -390,7 +411,7 @@ def report_view():
             else:
                 return f'{pred:.3f} <span style="color:red;">↓</span>'
         except:
-            return row['Makable_Predicted']
+            return row['Makable_Predicted'] or []
 
     df_page['Makable_Predicted'] = df_page.apply(add_colored_arrow, axis=1)
     # -------------------------------------V Colored CHnages --------------------------------------
@@ -428,14 +449,15 @@ def report_view():
     print("------------------------------------------------------------------------------------------------")    
     print("df_page['Makable_Diff']", df_page['Makable_Diff'])
     print("------------------------------------------------------------------------------------------------")
-    
+    '''
     # ------------------ V changes --------------------------------------
     # Prepare chart data (convert to JSON for JS)
     chart_data = {
-        'EngCts': df_page['EngCts'].tolist(),
-        'Makable_Predicted': df_page['Makable_Predicted'].tolist(),
-        'Makable_Diff': df_page['Makable_Diff'].tolist()
+        'EngCts': [],
+        'Makable_Predicted': [],
+        'Makable_Diff': []
     }
+    
     # ------------------ V changes --------------------------------------
     
     return render_template('output.html',