(type, values_of_this_column, lsh_result, folder_path, model)
| 348 | |
| 349 | |
| 350 | def cal_precise_unionablity(type, values_of_this_column, lsh_result, folder_path, model): |
| 351 | #计算values_of_this_column与候选列的精确可并行分数 |
| 352 | #后面加上u_NL的分数 |
| 353 | col_dict = {} |
| 354 | for r in lsh_result: |
| 355 | file_name, column_name = r.split(maxsplit=1) |
| 356 | file_path = os.path.join(folder_path, file_name) |
| 357 | df_query = pd.read_csv(file_path) |
| 358 | if type == 'Uset' or type == 'Usem': |
| 359 | unionablity = u_set_or_sem(type, values_of_this_column, df_query[column_name].to_list()) |
| 360 | else: |
| 361 | unionablity = u_nl(values_of_this_column, df_query[column_name], model) |
| 362 | if file_name in col_dict: |
| 363 | if unionablity > col_dict[file_name]: |
| 364 | col_dict[file_name] = unionablity |
| 365 | else: |
| 366 | col_dict[file_name] = unionablity |
| 367 | |
| 368 | return col_dict |
| 369 | |
| 370 | |
| 371 | #save_dict_to_json(u_set_distribution(u_set_Lsh()), 'result/uset_lsh.json') |
no test coverage detected