mirror of
https://github.com/open-compass/opencompass.git
synced 2025-05-30 16:03:24 +08:00
24 lines
1.1 KiB
Python
24 lines
1.1 KiB
Python
|
summarizer = dict(
|
||
|
|
dataset_abbrs=[
|
||
|
|
'######## MathBench Accuracy ########', # category
|
||
|
|
['mathbench-college-single_choice_cn', 'acc_1'],
|
||
|
|
['mathbench-college-single_choice_en', 'acc_1'],
|
||
|
|
['mathbench-high-single_choice_cn', 'acc_1'],
|
||
|
|
['mathbench-high-single_choice_en', 'acc_1'],
|
||
|
|
['mathbench-middle-single_choice_cn', 'acc_1'],
|
||
|
|
['mathbench-middle-single_choice_en', 'acc_1'],
|
||
|
|
['mathbench-primary-cloze_cn', 'accuracy'],
|
||
|
|
['mathbench-primary-cloze_en', 'accuracy'],
|
||
|
|
['mathbench-calculate-cloze_en', 'accuracy'],
|
||
|
|
'######## MathBench CircularEval ########', # category
|
||
|
|
['mathbench-college-single_choice_cn', 'perf_4'],
|
||
|
|
['mathbench-college-single_choice_en', 'perf_4'],
|
||
|
|
['mathbench-high-single_choice_cn', 'perf_4'],
|
||
|
|
['mathbench-high-single_choice_en', 'perf_4'],
|
||
|
|
['mathbench-middle-single_choice_cn', 'perf_4'],
|
||
|
|
['mathbench-middle-single_choice_en', 'perf_4'],
|
||
|
|
],
|
||
|
|
summary_groups=sum(
|
||
|
|
[v for k, v in locals().items() if k.endswith("_summary_groups")], [])
|
||
|
|
)
|