31 lines
901 B
Python
31 lines
901 B
Python
# Copyright (c) Alibaba, Inc. and its affiliates.
|
|
from evalscope.config import TaskConfig
|
|
from evalscope.run import run_task
|
|
|
|
|
|
def run_swift_eval():
|
|
|
|
task_cfg = TaskConfig(
|
|
model='qwen/Qwen2-0.5B-Instruct',
|
|
datasets=['general_mcq', 'general_qa'], # 数据格式，选择题格式固定为 'ceval'
|
|
dataset_args={
|
|
'general_mcq': {
|
|
'local_path': 'custom_eval/text/mcq', # 自定义数据集路径
|
|
'subset_list': [
|
|
'example' # 评测数据集名称
|
|
]
|
|
},
|
|
'general_qa': {
|
|
'local_path': 'custom_eval/text/qa', # 自定义数据集路径
|
|
'subset_list': [
|
|
'example' # 评测数据集名称
|
|
]
|
|
}
|
|
},
|
|
)
|
|
run_task(task_cfg=task_cfg)
|
|
|
|
|
|
if __name__ == '__main__':
|
|
run_swift_eval()
|