2023-07-06 12:27:41 +08:00
|
|
|
from datasets import Dataset, DatasetDict, load_dataset
|
2023-07-04 22:11:33 +08:00
|
|
|
|
|
|
|
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 22:11:33 +08:00
|
|
|
|
|
|
|
from .base import BaseDataset
|
|
|
|
|
|
|
|
|
|
|
|
@LOAD_DATASET.register_module()
|
|
|
|
class RealToxicPromptsDataset(BaseDataset):
|
|
|
|
|
|
|
|
@staticmethod
|
|
|
|
def load(**kwargs):
|
|
|
|
challenging_subset = kwargs.pop('challenging_subset', False)
|
2023-07-06 12:27:41 +08:00
|
|
|
if kwargs['path'] == 'allenai/real-toxicity-prompts':
|
|
|
|
try:
|
|
|
|
dataset = load_dataset(**kwargs)
|
|
|
|
except ConnectionError as e:
|
|
|
|
raise ConnectionError(
|
|
|
|
f'{e} Something wrong with this dataset, '
|
|
|
|
'cannot track it online or use offline mode, '
|
|
|
|
'please set local file path directly.')
|
|
|
|
else:
|
[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 = kwargs.pop('path')
|
|
|
|
path = get_data_path(path, local_mode=True)
|
|
|
|
dataset = Dataset.from_file(path)
|
2023-07-06 12:27:41 +08:00
|
|
|
dataset = DatasetDict(train=dataset)
|
2023-07-04 22:11:33 +08:00
|
|
|
|
|
|
|
def preprocess(example):
|
|
|
|
|
|
|
|
for k, v in example['prompt'].items():
|
|
|
|
k = 'prompt_' + k
|
|
|
|
example[k] = v
|
|
|
|
del example['prompt']
|
|
|
|
|
|
|
|
return example
|
|
|
|
|
|
|
|
dataset = dataset.map(preprocess)
|
|
|
|
|
|
|
|
# return challenging subset if necessary
|
|
|
|
if challenging_subset:
|
|
|
|
return dataset.filter(lambda example: example['challenging'])
|
|
|
|
return dataset
|