| from prompt import test_gen_prompt, predo_prompt, input_template, System_prompt | |
| from config import cfg | |
| import sys | |
| sys.path.append("/home/luoxianzhen/yang/methods/utils") | |
| from dataset_all import get_datasets_by_name, get_ours | |
| def get_datasets_prompt(): | |
| al_dataset = get_datasets_by_name(cfg.dataset_name) | |
| print(len(al_dataset)) | |
| prompt_list = [] | |
| for item in al_dataset: | |
| tcb_id = item["tcb_id"] | |
| random_generator = { | |
| "tcb_id": tcb_id, | |
| "type": "input", | |
| "system_prompt": test_gen_prompt, | |
| "user_prompt": input_template.replace("[[Question]]", item["query_en"]) | |
| } | |
| prompt_list += [random_generator, ] * 10 | |
| oracle_generator = { | |
| "tcb_id": tcb_id, | |
| "type": "answer", | |
| "system_prompt": System_prompt, | |
| "user_prompt": predo_prompt.replace("{question}", item['query_en']) | |
| } | |
| prompt_list += [oracle_generator, ] * 20 | |
| return prompt_list | |
| import json | |
| data = get_datasets_prompt() | |
| json.dump(data, open(f"/home/luoxianzhen/yang/data/prompt_saves/|qwen3-235b-a22b|predo|.json", "w", encoding="utf-8"), indent=4, ensure_ascii=False) | |
Xet Storage Details
- Size:
- 1.2 kB
- Xet hash:
- 31c69cd1cc2debd02f9a865ecb19f75231d93bf4d38ea47733235b52f1d45a59
·
Xet efficiently stores files, intelligently splitting them into unique chunks and accelerating uploads and downloads. More info.