| import json | |
| ## 移除非AC的正解代码 | |
| upudate_file = "/home/luoxianzhen/yang/eval/ALLmode_results/tcb-Sample_Right_code-rank5-all.json" | |
| correct_code_result = json.load(open(upudate_file, "r", encoding="utf-8")) | |
| tcb_bench = json.load(open("/home/luoxianzhen/yang/data/Ours/TestcaseBench-v28.json", "r", encoding="utf-8")) | |
| print(len(tcb_bench)) | |
| update_count = 0 | |
| for item in tcb_bench: | |
| tcb_id = item['tcb_id'] | |
| # print(tcb_id) | |
| code_status = correct_code_result[tcb_id]['codes'] | |
| for c_code in code_status: | |
| remove_list = [] | |
| if c_code['status'] != ['AC']: | |
| print(f"del {tcb_id} - {c_code['code_id']}") | |
| remove_list.append(c_code['code_id']) | |
| remove_list.sort(reverse=True) | |
| # 删除对应索引的元素 | |
| for i in remove_list: | |
| del item['solutions'][i] | |
| print(f"删除了 {update_count} 个样本") | |
| json.dump(tcb_bench, open(f"/home/luoxianzhen/yang/data/Ours/TestcaseBench-v29.json", "w", encoding="utf-8"), indent=4, ensure_ascii=False) | |
Xet Storage Details
- Size:
- 1.03 kB
- Xet hash:
- 253da65687cf3c95256443ecb3624a8536478fdcb1cc0e5e6a5e35f8d1b95166
·
Xet efficiently stores files, intelligently splitting them into unique chunks and accelerating uploads and downloads. More info.