jiangzhs/OpenCompass
1
0
Fork 0
mirror of https://github.com/open-compass/opencompass.git synced 2025-05-30 16:03:24 +08:00
Code Issues Actions 2 Packages Projects Releases Wiki Activity

hle biomed test

Browse Source
This commit is contained in:
Flaick 2025-04-25 10:09:33 +00:00
parent e8bc8c1e8c
commit 12597edea6
2 changed files with 17 additions and 3 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

6
opencompass/configs/datasets/HLE/hle_llmverify_gen_6ff468.py
View File

@ -3,7 +3,7 @@ from opencompass.openicl.icl_retriever import ZeroRetriever
from opencompass.openicl.icl_inferencer import GenInferencer
from opencompass.evaluator import GenericLLMEvaluator
from opencompass.datasets import generic_llmjudge_postprocess
from opencompass.datasets import HLEDataset
from opencompass.datasets import HLEDataset_Bio_Med
# ----------------------------- Detailed Config -----------------------------
@ -68,7 +68,7 @@ math_eval_cfg = dict(
]),
),
dataset_cfg=dict(
type=HLEDataset,
type=HLEDataset_Bio_Med,
path='cais/hle',
reader_cfg=math_reader_cfg,
),
@ -81,7 +81,7 @@ math_eval_cfg = dict(
hle_datasets = [
dict(
type=HLEDataset,
type=HLEDataset_Bio_Med,
abbr='hle_llmjudge',
path='cais/hle',
reader_cfg=math_reader_cfg,

14
opencompass/datasets/hle.py
View File

@ -15,3 +15,17 @@ class HLEDataset(BaseDataset):
dataset['test'] = dataset['test'].rename_column('question', 'problem')
dataset['train'] = dataset['test']
return dataset
@LOAD_DATASET.register_module()
class HLEDataset_Bio_Med(BaseDataset):
@staticmethod
def load(path: str):
dataset = load_dataset(path)
dataset['test'] = dataset['test'].filter(lambda x: x['image'] == '')
dataset['test'] = dataset['test'].filter(
lambda x: x['category'] == 'Biology/Medicine')
dataset['test'] = dataset['test'].rename_column('question', 'problem')
dataset['train'] = dataset['test']
return dataset