OpenCompass/opencompass/datasets/tnews.py

82 lines
2.5 KiB
Python
Raw Normal View History

2023-07-04 21:34:55 +08:00
import json
from datasets import Dataset, load_dataset
from opencompass.registry import LOAD_DATASET
[Feature] Support ModelScope datasets (#1289) * add ceval, gsm8k modelscope surpport * update race, mmlu, arc, cmmlu, commonsenseqa, humaneval and unittest * update bbh, flores, obqa, siqa, storycloze, summedits, winogrande, xsum datasets * format file * format file * update dataset format * support ms_dataset * udpate dataset for modelscope support * merge myl_dev and update test_ms_dataset * udpate dataset for modelscope support * update readme * update eval_api_zhipu_v2 * remove unused code * add get_data_path function * update readme * remove tydiqa japanese subset * add ceval, gsm8k modelscope surpport * update race, mmlu, arc, cmmlu, commonsenseqa, humaneval and unittest * update bbh, flores, obqa, siqa, storycloze, summedits, winogrande, xsum datasets * format file * format file * update dataset format * support ms_dataset * udpate dataset for modelscope support * merge myl_dev and update test_ms_dataset * update readme * udpate dataset for modelscope support * update eval_api_zhipu_v2 * remove unused code * add get_data_path function * remove tydiqa japanese subset * update util * remove .DS_Store * fix md format * move util into package * update docs/get_started.md * restore eval_api_zhipu_v2.py, add environment setting * Update dataset * Update * Update * Update * Update --------- Co-authored-by: Yun lin <yunlin@U-Q9X2K4QV-1904.local> Co-authored-by: Yunnglin <mao.looper@qq.com> Co-authored-by: Yun lin <yunlin@laptop.local> Co-authored-by: Yunnglin <maoyl@smail.nju.edu.cn> Co-authored-by: zhangsongyang <zhangsongyang@pjlab.org.cn>
2024-07-29 13:48:32 +08:00
from opencompass.utils import get_data_path
2023-07-04 21:34:55 +08:00
from .base import BaseDataset
@LOAD_DATASET.register_module()
class TNewsDataset(BaseDataset):
@staticmethod
def load(**kwargs):
tnews_targets = {
'news_agriculture': '农业新闻',
'news_travel': '旅游新闻',
'news_game': '游戏新闻',
'news_tech': '科技类别公司新闻',
'news_sports': '体育类别新闻',
'news_edu': '初升高教育新闻',
'news_entertainment': '娱乐圈新闻',
'news_finance': '投资资讯',
'news_military': '军事类别常识',
'news_car': '车辆新闻',
'news_house': '楼市新闻',
'news_world': '环球不含中国类别新闻',
'news_culture': '书籍文化历史类别新闻',
'news_story': '故事类别新闻',
'news_stock': '股票市场类别新闻',
}
[Feature] Support ModelScope datasets (#1289) * add ceval, gsm8k modelscope surpport * update race, mmlu, arc, cmmlu, commonsenseqa, humaneval and unittest * update bbh, flores, obqa, siqa, storycloze, summedits, winogrande, xsum datasets * format file * format file * update dataset format * support ms_dataset * udpate dataset for modelscope support * merge myl_dev and update test_ms_dataset * udpate dataset for modelscope support * update readme * update eval_api_zhipu_v2 * remove unused code * add get_data_path function * update readme * remove tydiqa japanese subset * add ceval, gsm8k modelscope surpport * update race, mmlu, arc, cmmlu, commonsenseqa, humaneval and unittest * update bbh, flores, obqa, siqa, storycloze, summedits, winogrande, xsum datasets * format file * format file * update dataset format * support ms_dataset * udpate dataset for modelscope support * merge myl_dev and update test_ms_dataset * update readme * udpate dataset for modelscope support * update eval_api_zhipu_v2 * remove unused code * add get_data_path function * remove tydiqa japanese subset * update util * remove .DS_Store * fix md format * move util into package * update docs/get_started.md * restore eval_api_zhipu_v2.py, add environment setting * Update dataset * Update * Update * Update * Update --------- Co-authored-by: Yun lin <yunlin@U-Q9X2K4QV-1904.local> Co-authored-by: Yunnglin <mao.looper@qq.com> Co-authored-by: Yun lin <yunlin@laptop.local> Co-authored-by: Yunnglin <maoyl@smail.nju.edu.cn> Co-authored-by: zhangsongyang <zhangsongyang@pjlab.org.cn>
2024-07-29 13:48:32 +08:00
if 'data_files' in kwargs:
kwargs['data_files'] = get_data_path(kwargs['data_files'],
local_mode=True)
2023-07-04 21:34:55 +08:00
dataset = load_dataset(**kwargs)
def preprocess(example):
label_desc = example['label_desc']
label_desc2 = tnews_targets[label_desc]
example['label_desc2'] = label_desc2
return example
dataset = dataset.map(preprocess)
return dataset
@LOAD_DATASET.register_module()
[Feature] Support ModelScope datasets (#1289) * add ceval, gsm8k modelscope surpport * update race, mmlu, arc, cmmlu, commonsenseqa, humaneval and unittest * update bbh, flores, obqa, siqa, storycloze, summedits, winogrande, xsum datasets * format file * format file * update dataset format * support ms_dataset * udpate dataset for modelscope support * merge myl_dev and update test_ms_dataset * udpate dataset for modelscope support * update readme * update eval_api_zhipu_v2 * remove unused code * add get_data_path function * update readme * remove tydiqa japanese subset * add ceval, gsm8k modelscope surpport * update race, mmlu, arc, cmmlu, commonsenseqa, humaneval and unittest * update bbh, flores, obqa, siqa, storycloze, summedits, winogrande, xsum datasets * format file * format file * update dataset format * support ms_dataset * udpate dataset for modelscope support * merge myl_dev and update test_ms_dataset * update readme * udpate dataset for modelscope support * update eval_api_zhipu_v2 * remove unused code * add get_data_path function * remove tydiqa japanese subset * update util * remove .DS_Store * fix md format * move util into package * update docs/get_started.md * restore eval_api_zhipu_v2.py, add environment setting * Update dataset * Update * Update * Update * Update --------- Co-authored-by: Yun lin <yunlin@U-Q9X2K4QV-1904.local> Co-authored-by: Yunnglin <mao.looper@qq.com> Co-authored-by: Yun lin <yunlin@laptop.local> Co-authored-by: Yunnglin <maoyl@smail.nju.edu.cn> Co-authored-by: zhangsongyang <zhangsongyang@pjlab.org.cn>
2024-07-29 13:48:32 +08:00
class TNewsDatasetV2(BaseDataset):
2023-07-04 21:34:55 +08:00
@staticmethod
def load(path):
tnews_targets = {
'news_agriculture': 'A',
'news_travel': 'B',
'news_game': 'C',
'news_tech': 'D',
'news_sports': 'E',
'news_edu': 'F',
'news_entertainment': 'G',
'news_finance': 'H',
'news_military': 'I',
'news_car': 'J',
'news_house': 'K',
'news_world': 'L',
'news_culture': 'M',
'news_story': 'N',
'news_stock': 'O',
}
[Feature] Support ModelScope datasets (#1289) * add ceval, gsm8k modelscope surpport * update race, mmlu, arc, cmmlu, commonsenseqa, humaneval and unittest * update bbh, flores, obqa, siqa, storycloze, summedits, winogrande, xsum datasets * format file * format file * update dataset format * support ms_dataset * udpate dataset for modelscope support * merge myl_dev and update test_ms_dataset * udpate dataset for modelscope support * update readme * update eval_api_zhipu_v2 * remove unused code * add get_data_path function * update readme * remove tydiqa japanese subset * add ceval, gsm8k modelscope surpport * update race, mmlu, arc, cmmlu, commonsenseqa, humaneval and unittest * update bbh, flores, obqa, siqa, storycloze, summedits, winogrande, xsum datasets * format file * format file * update dataset format * support ms_dataset * udpate dataset for modelscope support * merge myl_dev and update test_ms_dataset * update readme * udpate dataset for modelscope support * update eval_api_zhipu_v2 * remove unused code * add get_data_path function * remove tydiqa japanese subset * update util * remove .DS_Store * fix md format * move util into package * update docs/get_started.md * restore eval_api_zhipu_v2.py, add environment setting * Update dataset * Update * Update * Update * Update --------- Co-authored-by: Yun lin <yunlin@U-Q9X2K4QV-1904.local> Co-authored-by: Yunnglin <mao.looper@qq.com> Co-authored-by: Yun lin <yunlin@laptop.local> Co-authored-by: Yunnglin <maoyl@smail.nju.edu.cn> Co-authored-by: zhangsongyang <zhangsongyang@pjlab.org.cn>
2024-07-29 13:48:32 +08:00
path = get_data_path(path, local_mode=True)
2023-07-04 21:34:55 +08:00
data = []
2023-10-27 20:31:22 +08:00
with open(path, 'r', encoding='utf-8') as f:
2023-07-04 21:34:55 +08:00
for line in f:
line = json.loads(line)
item = {
'sentence': line['sentence'],
'label_desc2': tnews_targets[line['label_desc']],
}
data.append(item)
return Dataset.from_list(data)