OpenCompass/configs/datasets/obqa/obqa_ppl_1defe8.py

52 lines
1.7 KiB
Python
Raw Normal View History

2023-07-05 10:22:40 +08:00
from opencompass.openicl.icl_prompt_template import PromptTemplate
from opencompass.openicl.icl_retriever import ZeroRetriever
from opencompass.openicl.icl_inferencer import PPLInferencer
from opencompass.openicl.icl_evaluator import AccEvaluator
from opencompass.datasets import OBQADataset
_input_columns = [
['question_stem', 'A', 'B', 'C', 'D'],
['question_stem', 'A', 'B', 'C', 'D', 'fact1'],
]
_template = [{
2024-05-14 15:35:58 +08:00
'A': '{question_stem} {A}',
'B': '{question_stem} {B}',
'C': '{question_stem} {C}',
'D': '{question_stem} {D}',
2023-07-05 10:22:40 +08:00
}, {
2024-05-14 15:35:58 +08:00
'A': 'Given the fact {fact1}, we know that {question_stem} {A}',
'B': 'Given the fact {fact1}, we know that {question_stem} {B}',
'C': 'Given the fact {fact1}, we know that {question_stem} {C}',
'D': 'Given the fact {fact1}, we know that {question_stem} {D}',
2023-07-05 10:22:40 +08:00
}]
obqa_datasets = [
dict(
2024-05-14 15:35:58 +08:00
abbr='openbookqa',
2023-07-05 10:22:40 +08:00
type=OBQADataset,
[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='opencompass/openbookqa_test',
name='main',
2023-07-05 10:22:40 +08:00
),
dict(
2024-05-14 15:35:58 +08:00
abbr='openbookqa_fact',
2023-07-05 10:22:40 +08:00
type=OBQADataset,
[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='opencompass/openbookqa_fact',
name='additional',
2023-07-05 10:22:40 +08:00
),
]
for _i in range(2):
obqa_reader_cfg = dict(
2024-05-14 15:35:58 +08:00
input_columns=_input_columns[_i], output_column='answerKey')
2023-07-05 10:22:40 +08:00
obqa_infer_cfg = dict(
prompt_template=dict(
type=PromptTemplate,
template=_template[_i]),
retriever=dict(type=ZeroRetriever),
inferencer=dict(type=PPLInferencer),
)
obqa_eval_cfg = dict(evaluator=dict(type=AccEvaluator))
2024-05-14 15:35:58 +08:00
obqa_datasets[_i]['reader_cfg'] = obqa_reader_cfg
obqa_datasets[_i]['infer_cfg'] = obqa_infer_cfg
obqa_datasets[_i]['eval_cfg'] = obqa_eval_cfg