Initial commit

opencompass/configs/datasets/ClinicBench/ClinicBench_llmjudge_gen.py

+from mmengine.config import read_base
+
+with read_base():
+    from .ClinicBench_llmjudge_gen_d09668 import ClinicBench_datasets
\ No newline at end of file

opencompass/configs/datasets/ClinicBench/ClinicBench_llmjudge_gen_d09668.py

+from mmengine.config import read_base
+from opencompass.openicl.icl_prompt_template import PromptTemplate
+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.ClinicBench import ClinicBenchDataset
+
+
+QUERY_TEMPLATE = """
+Answer the following multiple choice question. The last line of your response should be of the following format: 'ANSWER: $LETTER' (without quotes) where LETTER is one of Options(e.g. one of ABCDEFGHIJKLMNOP). Think step by step before answering.
+
+Question:\n
+{question}
+
+Options:\n
+{choices}
+
+""".strip()
+
+GRADER_TEMPLATE = """
+    Please as a grading expert, judge whether the final answers given by the candidates below are consistent with the standard answers, that is, whether the candidates answered correctly. 
+    
+    Here are some evaluation criteria:
+    1. Please refer to the given standard answer. You don't need to re-generate the answer to the question because the standard answer has been given. You only need to judge whether the candidate's answer is consistent with the standard answer according to the form of the question. Don't try to answer the original question. You can assume that the standard answer is definitely correct.
+    2. Because the candidate's answer may be different from the standard answer in the form of expression, before making a judgment, please understand the question and the standard answer first, and then judge whether the candidate's answer is correct, but be careful not to try to answer the original question.
+    3. Some answers may contain multiple items, such as multiple-choice questions, multiple-select questions, fill-in-the-blank questions, etc. As long as the answer is the same as the standard answer, it is enough. For multiple-select questions and multiple-blank fill-in-the-blank questions, the candidate needs to answer all the corresponding options or blanks correctly to be considered correct.
+    4. Some answers may be expressed in different ways, such as some answers may be a mathematical expression, some answers may be a textual description, as long as the meaning expressed is the same. And some formulas are expressed in different ways, but they are equivalent and correct.
+
+    Please judge whether the following answers are consistent with the standard answer based on the above criteria. Grade the predicted answer of this new question as one of:
+    A: CORRECT 
+    B: INCORRECT
+    Just return the letters "A" or "B", with no text around it.
+
+    Here is your task. Simply reply with either CORRECT, INCORRECT. Don't apologize or correct yourself if there was a mistake; we are just trying to grade the answer.
+
+    <Original Question Begin>: {question}\n {choices} \n<Original Question End>\n\n
+    <Gold Target Begin>: \n{label}\n<Gold Target End>\n\n
+    <Predicted Answer Begin>: \n{prediction}\n<Predicted End>\n\n
+    Judging the correctness of candidates' answers:
+""".strip()
+
+ClinicBench_datasets = []
+
+ClinicBench_reader_cfg = dict(
+    input_columns=['question', 'choices'],
+    output_column='label',
+)
+
+ClinicBench_infer_cfg = dict(
+    prompt_template=dict(
+        type=PromptTemplate,
+        template=dict(
+            round=[
+                dict(role='HUMAN', prompt=QUERY_TEMPLATE),
+            ],
+        ),
+    ),
+    retriever=dict(type=ZeroRetriever),
+    inferencer=dict(type=GenInferencer),
+)
+
+ClinicBench_eval_cfg = dict(
+    evaluator=dict(
+        type=GenericLLMEvaluator,
+        prompt_template=dict(
+            type=PromptTemplate,
+            template=dict(
+                begin=[
+                    dict(
+                        role='SYSTEM',
+                        fallback_role='HUMAN',
+                        prompt="You are a helpful assistant who evaluates the correctness and quality of models' outputs.",
+                    )
+                ],
+                round=[
+                    dict(role='HUMAN', prompt=GRADER_TEMPLATE),
+                ],
+            ),
+        ),
+        dataset_cfg=dict(
+            type=ClinicBenchDataset,
+            path='xuxuxuxuxu/Pharmacology-QA',
+            reader_cfg=ClinicBench_reader_cfg,
+        ),
+        judge_cfg=dict(),
+        dict_postprocessor=dict(type=generic_llmjudge_postprocess),
+    ),
+)
+
+ClinicBench_datasets.append(
+    dict(
+        abbr=f'ClinicBench',
+        type=ClinicBenchDataset,
+        path='xuxuxuxuxu/Pharmacology-QA',
+        reader_cfg=ClinicBench_reader_cfg,
+        infer_cfg=ClinicBench_infer_cfg,
+        eval_cfg=ClinicBench_eval_cfg,
+    )
+)

opencompass/configs/datasets/Earth_Silver/Earth_Silver_gen.py

+from opencompass.datasets import Earth_Silver_MCQDataset
+from opencompass.openicl.icl_inferencer import GenInferencer
+from opencompass.openicl.icl_prompt_template import PromptTemplate
+from opencompass.openicl.icl_retriever import ZeroRetriever
+from opencompass.openicl.icl_evaluator import AccEvaluator
+
+
+SYSTEM_PROMPT = 'You are a helpful assistant for answering earth science multiple-choice questions.\n\n'
+
+
+ZERO_SHOT_PROMPT = 'Q: {question}\nPlease select the correct answer from the options above and output only the corresponding letter (A, B, C, or D) without any explanation or additional text.\n'
+
+
+reader_cfg = dict(
+    input_columns=['question'],  
+    output_column='answer', 
+)
+
+
+infer_cfg = dict(
+    prompt_template=dict(
+        type=PromptTemplate,
+        template=dict(
+            begin=[  
+                dict(role='SYSTEM', fallback_role='HUMAN', prompt=SYSTEM_PROMPT),
+            ],
+            round=[  
+                dict(
+                    role='HUMAN',
+                    prompt=ZERO_SHOT_PROMPT,  
+                ),
+            ],
+        ),
+    ),
+    retriever=dict(type=ZeroRetriever),  
+    inferencer=dict(type=GenInferencer),  
+)
+
+
+eval_cfg = dict(
+    evaluator=dict(type=AccEvaluator),
+    pred_role='BOT',  
+)
+
+
+earth_silver_mcq_dataset = dict(
+    type=Earth_Silver_MCQDataset,  
+    abbr='earth_silver_mcq',  
+    path='ai-earth/Earth-Silver',  
+    prompt_mode='zero-shot', 
+    reader_cfg=reader_cfg,
+    infer_cfg=infer_cfg,
+    eval_cfg=eval_cfg,
+)
+
+
+earth_silver_mcq_datasets = [earth_silver_mcq_dataset]

opencompass/configs/datasets/Earth_Silver/Earth_Silver_llmjudge_gen.py

+from opencompass.datasets import Earth_Silver_MCQDataset, generic_llmjudge_postprocess
+from opencompass.openicl.icl_inferencer import GenInferencer
+from opencompass.openicl.icl_prompt_template import PromptTemplate
+from opencompass.openicl.icl_retriever import ZeroRetriever
+from opencompass.evaluator import GenericLLMEvaluator
+
+SYSTEM_PROMPT = 'You are a helpful medical assistant.\n\n' # Where to put this?
+ZERO_SHOT_PROMPT = 'Q: {question}\n Please select the correct answer from the options above and output only the corresponding letter (A, B, C, D, or E) without any explanation or additional text.\n'
+GRADER_TEMPLATE = """
+    Please as a grading expert, judge whether the final answers given by the candidates below are consistent with the standard answers, that is, whether the candidates answered correctly. 
+    
+    Here are some evaluation criteria:
+    1. Please refer to the given standard answer. You don't need to re-generate the answer to the question because the standard answer has been given. You only need to judge whether the candidate's answer is consistent with the standard answer according to the form of the question. Don't try to answer the original question. You can assume that the standard answer is definitely correct.
+    2. Because the candidate's answer may be different from the standard answer in the form of expression, before making a judgment, please understand the question and the standard answer first, and then judge whether the candidate's answer is correct, but be careful not to try to answer the original question.
+    3. Some answers may contain multiple items, such as multiple-choice questions, multiple-select questions, fill-in-the-blank questions, etc. As long as the answer is the same as the standard answer, it is enough. For multiple-select questions and multiple-blank fill-in-the-blank questions, the candidate needs to answer all the corresponding options or blanks correctly to be considered correct.
+    4. Some answers may be expressed in different ways, such as some answers may be a mathematical expression, some answers may be a textual description, as long as the meaning expressed is the same. And some formulas are expressed in different ways, but they are equivalent and correct.
+    Please judge whether the following answers are consistent with the standard answer based on the above criteria. Grade the predicted answer of this new question as one of:
+    A: CORRECT 
+    B: INCORRECT
+    Just return the letters "A" or "B", with no text around it.
+    Here is your task. Simply reply with either CORRECT, INCORRECT. Don't apologize or correct yourself if there was a mistake; we are just trying to grade the answer.
+    <Original Question Begin>: Q: {question}\nPlease select the correct answer from the options above and output only the corresponding letter (A, B, C, D, or E) without any explanation or additional text.\n\n<Original Question End>\n\n
+    <Gold Target Begin>: \n{answer}\n<Gold Target End>\n\n
+    <Predicted Answer Begin>: \n{prediction}\n<Predicted End>\n\n
+    Judging the correctness of candidates' answers:
+""".strip()
+
+
+# Reader configuration
+reader_cfg = dict(
+    input_columns=[
+        'question'
+    ],
+    output_column='answer',
+)
+
+# Inference configuration
+infer_cfg = dict(
+    prompt_template=dict(
+        type=PromptTemplate,
+        template=dict(
+            begin=[
+                dict(role='SYSTEM', fallback_role='HUMAN', prompt=SYSTEM_PROMPT),
+            ],
+            round=[
+                dict(
+                    role='HUMAN',
+                    prompt=ZERO_SHOT_PROMPT, # prompt mode: zero-shot
+                ),
+            ],
+        ),
+    ),
+    retriever=dict(type=ZeroRetriever),
+    inferencer=dict(type=GenInferencer),
+)
+
+# Evaluation configuration
+eval_cfg = dict(
+    evaluator=dict(
+        type=GenericLLMEvaluator,
+        prompt_template=dict(
+            type=PromptTemplate,
+            template=dict(
+                begin=[
+                    dict(
+                        role='SYSTEM',
+                        fallback_role='HUMAN',
+                        prompt="You are a helpful assistant who evaluates the correctness and quality of models' outputs.",
+                    )
+                ],
+                round=[
+                    dict(role='HUMAN', prompt=GRADER_TEMPLATE),
+                ],
+            ),
+        ),
+        dataset_cfg=dict(
+            type=Earth_Silver_MCQDataset,
+            path='ai-earth/Earth-Silver',
+            prompt_mode='zero-shot',
+            reader_cfg=reader_cfg,
+        ),
+        judge_cfg=dict(),
+        dict_postprocessor=dict(type=generic_llmjudge_postprocess),
+    ),
+)
+earth_silver_mcq_dataset = dict(
+    type=Earth_Silver_MCQDataset,
+    abbr='earth_silver_mcq',
+    path='ai-earth/Earth-Silver',
+    prompt_mode='zero-shot',
+    reader_cfg=reader_cfg,
+    infer_cfg=infer_cfg,
+    eval_cfg=eval_cfg,
+
+)
+
+earth_silver_mcq_datasets = [earth_silver_mcq_dataset]
\ No newline at end of file

opencompass/configs/datasets/Earth_Silver/Earth_Silver_llmjudge_gen_46140c.py

+from opencompass.datasets import Earth_Silver_MCQDataset, generic_llmjudge_postprocess
+from opencompass.openicl.icl_inferencer import GenInferencer
+from opencompass.openicl.icl_prompt_template import PromptTemplate
+from opencompass.openicl.icl_retriever import ZeroRetriever
+from opencompass.evaluator import GenericLLMEvaluator
+
+
+ZERO_SHOT_PROMPT = 'Q: {question}\n Please select the correct answer from the options above and output only the corresponding letter (A, B, C, D, or E) without any explanation or additional text.\n'
+
+GRADER_TEMPLATE = """
+    Please as a grading expert, judge whether the final answers given by the candidates below are consistent with the standard answers, that is, whether the candidates answered correctly. 
+
+    Here are some evaluation criteria:
+    1. Please refer to the given standard answer. You don't need to re-generate the answer to the question because the standard answer has been given. You only need to judge whether the candidate's answer is consistent with the standard answer according to the form of the question. Don't try to answer the original question. You can assume that the standard answer is definitely correct.
+    2. Because the candidate's answer may be different from the standard answer in the form of expression, before making a judgment, please understand the question and the standard answer first, and then judge whether the candidate's answer is correct, but be careful not to try to answer the original question.
+    3. Some answers may contain multiple items, such as multiple-choice questions, multiple-select questions, fill-in-the-blank questions, etc. As long as the answer is the same as the standard answer, it is enough. For multiple-select questions and multiple-blank fill-in-the-blank questions, the candidate needs to answer all the corresponding options or blanks correctly to be considered correct.
+    4. Some answers may be expressed in different ways, such as some answers may be a mathematical expression, some answers may be a textual description, as long as the meaning expressed is the same. And some formulas are expressed in different ways, but they are equivalent and correct.
+    Please judge whether the following answers are consistent with the standard answer based on the above criteria. Grade the predicted answer of this new question as one of:
+    A: CORRECT 
+    B: INCORRECT
+    Just return the letters "A" or "B", with no text around it.
+    Here is your task. Simply reply with either CORRECT, INCORRECT. Don't apologize or correct yourself if there was a mistake; we are just trying to grade the answer.
+    <Original Question Begin>: Q: {question}\nPlease select the correct answer from the options above and output only the corresponding letter (A, B, C, D, or E) without any explanation or additional text.\n\n<Original Question End>\n\n
+    <Gold Target Begin>: \n{answer}\n<Gold Target End>\n\n
+    <Predicted Answer Begin>: \n{prediction}\n<Predicted End>\n\n
+    Judging the correctness of candidates' answers:
+""".strip()
+
+# Reader configuration
+reader_cfg = dict(
+    input_columns=[
+        'question'
+    ],
+    output_column='answer',
+)
+
+# Inference configuration
+infer_cfg = dict(
+    prompt_template=dict(
+        type=PromptTemplate,
+        template=dict(
+            round=[
+                dict(
+                    role='HUMAN',
+                    prompt=ZERO_SHOT_PROMPT,  # prompt mode: zero-shot
+                ),
+            ],
+        ),
+    ),
+    retriever=dict(type=ZeroRetriever),
+    inferencer=dict(type=GenInferencer),
+)
+
+# Evaluation configuration
+eval_cfg = dict(
+    evaluator=dict(
+        type=GenericLLMEvaluator,
+        prompt_template=dict(
+            type=PromptTemplate,
+            template=dict(
+                begin=[
+                    dict(
+                        role='SYSTEM',
+                        fallback_role='HUMAN',
+                        prompt="You are a helpful assistant who evaluates the correctness and quality of models' outputs.",
+                    )
+                ],
+                round=[
+                    dict(role='HUMAN', prompt=GRADER_TEMPLATE),
+                ],
+            ),
+        ),
+        dataset_cfg=dict(
+            type=Earth_Silver_MCQDataset,
+            path='ai-earth/Earth-Silver',
+            prompt_mode='zero-shot',
+            reader_cfg=reader_cfg,
+        ),
+        judge_cfg=dict(),
+        dict_postprocessor=dict(type=generic_llmjudge_postprocess),
+    ),
+)
+earth_silver_mcq_dataset = dict(
+    type=Earth_Silver_MCQDataset,
+    abbr='earth_silver_mcq',
+    path='ai-earth/Earth-Silver',
+    prompt_mode='zero-shot',
+    reader_cfg=reader_cfg,
+    infer_cfg=infer_cfg,
+    eval_cfg=eval_cfg,
+
+)
+
+earth_silver_mcq_datasets = [earth_silver_mcq_dataset]
\ No newline at end of file

opencompass/configs/datasets/FewCLUE_bustm/FewCLUE_bustm_gen.py

+from mmengine.config import read_base
+
+with read_base():
+    from .FewCLUE_bustm_gen_634f41 import bustm_datasets  # noqa: F401, F403

opencompass/configs/datasets/FewCLUE_bustm/FewCLUE_bustm_gen_634f41.py

+from opencompass.openicl.icl_prompt_template import PromptTemplate
+from opencompass.openicl.icl_retriever import ZeroRetriever
+from opencompass.openicl.icl_inferencer import GenInferencer
+from opencompass.openicl.icl_evaluator import AccEvaluator
+from opencompass.datasets import AFQMCDatasetV2
+from opencompass.utils.text_postprocessors import first_capital_postprocess
+
+bustm_reader_cfg = dict(
+    input_columns=['sentence1', 'sentence2'],
+    output_column='label',
+    test_split='train')
+
+bustm_infer_cfg = dict(
+    prompt_template=dict(
+        type=PromptTemplate,
+        template=dict(round=[
+            dict(
+                role='HUMAN',
+                prompt=
+                '语句一：“{sentence1}”\n语句二：“{sentence2}”\n请判断语句一和语句二说的是否是一个意思？\nA. 无关\nB. 相关\n请从“A”，“B”中进行选择。\n答：',
+            ),
+        ]),
+    ),
+    retriever=dict(type=ZeroRetriever),
+    inferencer=dict(type=GenInferencer),
+)
+
+bustm_eval_cfg = dict(
+    evaluator=dict(type=AccEvaluator),
+    pred_role='BOT',
+    pred_postprocessor=dict(type=first_capital_postprocess),
+)
+
+bustm_datasets = [
+    dict(
+        abbr='bustm-dev',
+        type=AFQMCDatasetV2,  # bustm share the same format with AFQMC
+        path='./data/FewCLUE/bustm/dev_few_all.json',
+        local_mode=True,
+        reader_cfg=bustm_reader_cfg,
+        infer_cfg=bustm_infer_cfg,
+        eval_cfg=bustm_eval_cfg,
+    ),
+    dict(
+        abbr='bustm-test',
+        type=AFQMCDatasetV2,  # bustm share the same format with AFQMC
+        path='./data/FewCLUE/bustm/test_public.json',
+        local_mode=True,
+        reader_cfg=bustm_reader_cfg,
+        infer_cfg=bustm_infer_cfg,
+        eval_cfg=bustm_eval_cfg,
+    ),
+]

opencompass/configs/datasets/FewCLUE_bustm/FewCLUE_bustm_ppl.py

+from mmengine.config import read_base
+
+with read_base():
+    from .FewCLUE_bustm_ppl_e53034 import bustm_datasets  # noqa: F401, F403

opencompass/configs/datasets/FewCLUE_bustm/FewCLUE_bustm_ppl_4b16c0.py

+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 HFDataset
+
+bustm_reader_cfg = dict(
+    input_columns=['sentence1', 'sentence2'],
+    output_column='label',
+    test_split='train')
+
+bustm_infer_cfg = dict(
+    prompt_template=dict(
+        type=PromptTemplate,
+        template={
+            0:
+            dict(
+                begin=[
+                    dict(
+                        role='SYSTEM',
+                        fallback_role='HUMAN',
+                        prompt='请判断以下两句话说的是否是一个意思：')
+                ],
+                round=[
+                    dict(role='HUMAN', prompt='{sentence1}，{sentence2}'),
+                    dict(role='BOT', prompt='两句话说的毫不相关。')
+                ]),
+            1:
+            dict(
+                begin=[
+                    dict(
+                        role='SYSTEM',
+                        fallback_role='HUMAN',
+                        prompt='请判断以下两句话说的是否是一个意思：')
+                ],
+                round=[
+                    dict(role='HUMAN', prompt='{sentence1}，{sentence2}'),
+                    dict(role='BOT', prompt='两句话说是的一个意思。')
+                ]),
+        }),
+    retriever=dict(type=ZeroRetriever),
+    inferencer=dict(type=PPLInferencer))
+
+bustm_eval_cfg = dict(evaluator=dict(type=AccEvaluator))
+
+bustm_datasets = [
+    dict(
+        type=HFDataset,
+        abbr='bustm-dev',
+        path='json',
+        data_files='./data/FewCLUE/bustm/dev_few_all.json',
+        split='train',
+        reader_cfg=bustm_reader_cfg,
+        infer_cfg=bustm_infer_cfg,
+        eval_cfg=bustm_eval_cfg),
+    dict(
+        type=HFDataset,
+        abbr='bustm-test',
+        path='json',
+        data_files='./data/FewCLUE/bustm/test_public.json',
+        split='train',
+        reader_cfg=bustm_reader_cfg,
+        infer_cfg=bustm_infer_cfg,
+        eval_cfg=bustm_eval_cfg)
+]

opencompass/configs/datasets/FewCLUE_bustm/FewCLUE_bustm_ppl_9ef540.py

+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 HFDataset
+
+bustm_reader_cfg = dict(
+    input_columns=['sentence1', 'sentence2'],
+    output_column='label',
+    test_split='train')
+
+bustm_infer_cfg = dict(
+    prompt_template=dict(
+        type=PromptTemplate,
+        template={
+            0: '{sentence1}。\n{sentence2}。\n两句话说的毫不相关。',
+            1: '{sentence1}。\n{sentence2}。\n两句话说的一个意思。'
+        }),
+    retriever=dict(type=ZeroRetriever),
+    inferencer=dict(type=PPLInferencer))
+
+bustm_eval_cfg = dict(evaluator=dict(type=AccEvaluator))
+
+bustm_datasets = [
+    dict(
+        type=HFDataset,
+        abbr='bustm-dev',
+        path='json',
+        data_files='./data/FewCLUE/bustm/dev_few_all.json',
+        split='train',
+        reader_cfg=bustm_reader_cfg,
+        infer_cfg=bustm_infer_cfg,
+        eval_cfg=bustm_eval_cfg),
+    dict(
+        type=HFDataset,
+        abbr='bustm-test',
+        path='json',
+        data_files='./data/FewCLUE/bustm/test_public.json',
+        split='train',
+        reader_cfg=bustm_reader_cfg,
+        infer_cfg=bustm_infer_cfg,
+        eval_cfg=bustm_eval_cfg)
+]

opencompass/configs/datasets/FewCLUE_bustm/FewCLUE_bustm_ppl_e53034.py

+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 HFDataset
+
+bustm_reader_cfg = dict(
+    input_columns=['sentence1', 'sentence2'],
+    output_column='label',
+    test_split='train')
+
+bustm_infer_cfg = dict(
+    prompt_template=dict(
+        type=PromptTemplate,
+        template={
+            0:
+            dict(round=[
+                dict(
+                    role='HUMAN',
+                    prompt=
+                    '语句一：“{sentence1}”\n语句二：“{sentence2}”\n请判断语句一和语句二说的是否是一个意思？'
+                ),
+                dict(role='BOT', prompt='两句话说的毫不相关。')
+            ]),
+            1:
+            dict(round=[
+                dict(
+                    role='HUMAN',
+                    prompt=
+                    '语句一：“{sentence1}”\n语句二：“{sentence2}”\n请判断语句一和语句二说的是否是一个意思？'
+                ),
+                dict(role='BOT', prompt='两句话说是的一个意思。')
+            ]),
+        }),
+    retriever=dict(type=ZeroRetriever),
+    inferencer=dict(type=PPLInferencer))
+
+bustm_eval_cfg = dict(evaluator=dict(type=AccEvaluator))
+
+bustm_datasets = [
+    dict(
+        type=HFDataset,
+        abbr='bustm-dev',
+        path='json',
+        data_files='./data/FewCLUE/bustm/dev_few_all.json',
+        split='train',
+        reader_cfg=bustm_reader_cfg,
+        infer_cfg=bustm_infer_cfg,
+        eval_cfg=bustm_eval_cfg),
+    dict(
+        type=HFDataset,
+        abbr='bustm-test',
+        path='json',
+        data_files='./data/FewCLUE/bustm/test_public.json',
+        split='train',
+        reader_cfg=bustm_reader_cfg,
+        infer_cfg=bustm_infer_cfg,
+        eval_cfg=bustm_eval_cfg)
+]

opencompass/configs/datasets/FewCLUE_chid/FewCLUE_chid_gen.py

+from mmengine.config import read_base
+
+with read_base():
+    from .FewCLUE_chid_gen_0a29a2 import chid_datasets  # noqa: F401, F403

opencompass/configs/datasets/FewCLUE_chid/FewCLUE_chid_gen_0a29a2.py

+from opencompass.openicl.icl_prompt_template import PromptTemplate
+from opencompass.openicl.icl_retriever import ZeroRetriever
+from opencompass.openicl.icl_inferencer import GenInferencer
+from opencompass.openicl.icl_evaluator import AccEvaluator
+from opencompass.datasets import CHIDDatasetV2
+from opencompass.utils.text_postprocessors import first_capital_postprocess
+
+chid_reader_cfg = dict(
+    input_columns=['content','A','B','C','D','E','F','G'],
+    output_column='answer',
+)
+
+chid_infer_cfg = dict(
+    prompt_template=dict(
+        type=PromptTemplate,
+        template=dict(
+            round=[
+                dict(
+                    role='HUMAN',
+                    prompt=
+                    '{content}\n请选择______处所填的词\nA. {A}\nB. {B}\nC. {C}\nD. {D}\nE. {E}\nF. {F}\nG. {G}\n请从”A“，”B“，”C“，”D“，”E“，”F“，”G“中进行选择。答：',
+                ),
+            ])),
+    retriever=dict(type=ZeroRetriever),
+    inferencer=dict(type=GenInferencer),
+)
+
+chid_eval_cfg = dict(
+    evaluator=dict(type=AccEvaluator),
+    pred_role='BOT',
+    pred_postprocessor=dict(type=first_capital_postprocess),
+)
+
+chid_datasets = [
+    dict(
+        abbr='chid-dev',
+        type=CHIDDatasetV2,
+        path='./data/FewCLUE/chid/dev_few_all.json',
+        reader_cfg=chid_reader_cfg,
+        infer_cfg=chid_infer_cfg,
+        eval_cfg=chid_eval_cfg,
+    ),
+    dict(
+        abbr='chid-test',
+        type=CHIDDatasetV2,
+        path='./data/FewCLUE/chid/test_public.json',
+        reader_cfg=chid_reader_cfg,
+        infer_cfg=chid_infer_cfg,
+        eval_cfg=chid_eval_cfg,
+    ),
+]

opencompass/configs/datasets/FewCLUE_chid/FewCLUE_chid_ppl.py

+from mmengine.config import read_base
+
+with read_base():
+    from .FewCLUE_chid_ppl_8f2872 import chid_datasets  # noqa: F401, F403

opencompass/configs/datasets/FewCLUE_chid/FewCLUE_chid_ppl_8f2872.py

+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 CHIDDataset
+
+chid_reader_cfg = dict(
+    input_columns=[f'content{i}' for i in range(7)], output_column='answer')
+
+chid_infer_cfg = dict(
+    prompt_template=dict(
+        type=PromptTemplate,
+        template={
+            i: dict(
+                round=[
+                    dict(role='HUMAN', prompt=f'以下句子是否通顺？\n{{content{i}}}'),
+                    dict(role='BOT', prompt='这个句子是通顺的。'),
+                ], )
+            for i in range(7)
+        }),
+    retriever=dict(type=ZeroRetriever),
+    inferencer=dict(type=PPLInferencer))
+
+chid_eval_cfg = dict(evaluator=dict(type=AccEvaluator))
+
+chid_datasets = [
+    dict(
+        type=CHIDDataset,
+        path='json',
+        abbr='chid-dev',
+        data_files='./data/FewCLUE/chid/dev_few_all.json',
+        split='train',
+        reader_cfg=chid_reader_cfg,
+        infer_cfg=chid_infer_cfg,
+        eval_cfg=chid_eval_cfg),
+    dict(
+        type=CHIDDataset,
+        path='json',
+        abbr='chid-test',
+        data_files='./data/FewCLUE/chid/test_public.json',
+        split='train',
+        reader_cfg=chid_reader_cfg,
+        infer_cfg=chid_infer_cfg,
+        eval_cfg=chid_eval_cfg),
+]

opencompass/configs/datasets/FewCLUE_chid/FewCLUE_chid_ppl_acccb5.py

+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 CHIDDataset
+
+chid_reader_cfg = dict(
+    input_columns=[f'content{i}' for i in range(7)], output_column='answer')
+
+chid_infer_cfg = dict(
+    prompt_template=dict(
+        type=PromptTemplate,
+        template={i: f'以下句子是否通顺？\n{{content{i}}}\n这个句子是通顺的。'
+                  for i in range(7)}),
+    retriever=dict(type=ZeroRetriever),
+    inferencer=dict(type=PPLInferencer))
+
+chid_eval_cfg = dict(evaluator=dict(type=AccEvaluator))
+
+chid_datasets = [
+    dict(
+        type=CHIDDataset,
+        path='json',
+        abbr='chid-dev',
+        data_files='./data/FewCLUE/chid/dev_few_all.json',
+        split='train',
+        reader_cfg=chid_reader_cfg,
+        infer_cfg=chid_infer_cfg,
+        eval_cfg=chid_eval_cfg),
+    dict(
+        type=CHIDDataset,
+        path='json',
+        abbr='chid-test',
+        data_files='./data/FewCLUE/chid/test_public.json',
+        split='train',
+        reader_cfg=chid_reader_cfg,
+        infer_cfg=chid_infer_cfg,
+        eval_cfg=chid_eval_cfg),
+]

opencompass/configs/datasets/FewCLUE_cluewsc/FewCLUE_cluewsc_gen.py

+from mmengine.config import read_base
+
+with read_base():
+    from .FewCLUE_cluewsc_gen_c68933 import cluewsc_datasets  # noqa: F401, F403

opencompass/configs/datasets/FewCLUE_cluewsc/FewCLUE_cluewsc_gen_c68933.py

+from opencompass.openicl.icl_prompt_template import PromptTemplate
+from opencompass.openicl.icl_retriever import ZeroRetriever
+from opencompass.openicl.icl_inferencer import GenInferencer
+from opencompass.openicl.icl_evaluator import AccEvaluator
+from opencompass.datasets import CluewscDatasetV2
+from opencompass.utils.text_postprocessors import first_capital_postprocess
+
+cluewsc_reader_cfg = dict(
+    input_columns=['span1', 'span2', 'text', 'new_text'],
+    output_column='label',
+)
+
+cluewsc_infer_cfg = dict(
+    prompt_template=dict(
+        type=PromptTemplate,
+        template=dict(round=[
+            dict(
+                role='HUMAN',
+                prompt=
+                '{text}\n此处，“{span2}”是否指代“{span1}“？\nA. 是\nB. 否\n请从”A“，”B“中进行选择。\n答：',
+            ),
+        ]),
+    ),
+    retriever=dict(type=ZeroRetriever),
+    inferencer=dict(type=GenInferencer),
+)
+
+cluewsc_eval_cfg = dict(
+    evaluator=dict(type=AccEvaluator),
+    pred_role='BOT',
+    pred_postprocessor=dict(type=first_capital_postprocess),
+)
+
+cluewsc_datasets = [
+    dict(
+        abbr='cluewsc-dev',
+        type=CluewscDatasetV2,
+        path='./data/FewCLUE/cluewsc/dev_few_all.json',
+        reader_cfg=cluewsc_reader_cfg,
+        infer_cfg=cluewsc_infer_cfg,
+        eval_cfg=cluewsc_eval_cfg,
+    ),
+    dict(
+        abbr='cluewsc-test',
+        type=CluewscDatasetV2,
+        path='./data/FewCLUE/cluewsc/test_public.json',
+        reader_cfg=cluewsc_reader_cfg,
+        infer_cfg=cluewsc_infer_cfg,
+        eval_cfg=cluewsc_eval_cfg,
+    ),
+]

opencompass/configs/datasets/FewCLUE_cluewsc/FewCLUE_cluewsc_ppl.py

+from mmengine.config import read_base
+
+with read_base():
+    from .FewCLUE_cluewsc_ppl_868415 import cluewsc_datasets  # noqa: F401, F403

opencompass/configs/datasets/FewCLUE_cluewsc/FewCLUE_cluewsc_ppl_12e4e0.py

+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 CluewscDataset
+
+cluewsc_reader_cfg = dict(
+    input_columns=['span1', 'span2', 'text', 'new_text'],
+    output_column='answer')
+
+cluewsc_infer_cfg = dict(
+    prompt_template=dict(
+        type=PromptTemplate,
+        template={
+            0:
+            dict(round=[
+                dict(
+                    role='HUMAN',
+                    prompt=
+                    "{text}\nHere, is the pronoun \"{span2}\" used to mean \"{span1}\"?"
+                ),
+                dict(role='BOT', prompt='No.')
+            ]),
+            1:
+            dict(round=[
+                dict(
+                    role='HUMAN',
+                    prompt=
+                    "{text}\nHere, is the pronoun \"{span2}\" used to mean \"{span1}\"?"
+                ),
+                dict(role='BOT', prompt='Yes.')
+            ]),
+        }),
+    retriever=dict(type=ZeroRetriever),
+    inferencer=dict(type=PPLInferencer))
+
+cluewsc_eval_cfg = dict(evaluator=dict(type=AccEvaluator))
+
+cluewsc_datasets = [
+    dict(
+        type=CluewscDataset,
+        path='json',
+        abbr='cluewsc-dev',
+        data_files='./data/FewCLUE/cluewsc/dev_few_all.json',
+        split='train',
+        reader_cfg=cluewsc_reader_cfg,
+        infer_cfg=cluewsc_infer_cfg,
+        eval_cfg=cluewsc_eval_cfg),
+    dict(
+        type=CluewscDataset,
+        path='json',
+        abbr='cluewsc-test',
+        data_files='./data/FewCLUE/cluewsc/test_public.json',
+        split='train',
+        reader_cfg=cluewsc_reader_cfg,
+        infer_cfg=cluewsc_infer_cfg,
+        eval_cfg=cluewsc_eval_cfg),
+]