From 48ac21f37190d91fd95d3f80aa4cb13b1dbdcda1 Mon Sep 17 00:00:00 2001 From: marcry Date: Tue, 29 Apr 2025 09:16:25 +0000 Subject: [PATCH] support nejm ai benchmark --- .../nejm_ai_benchmark/nejmaibench_gen.py | 67 +++++ .../nejmaibench_llmjudge_gen.py | 116 ++++++++ opencompass/datasets/__init__.py | 1 + opencompass/datasets/nejmaibench.py | 251 ++++++++++++++++++ 4 files changed, 435 insertions(+) create mode 100644 opencompass/configs/datasets/nejm_ai_benchmark/nejmaibench_gen.py create mode 100644 opencompass/configs/datasets/nejm_ai_benchmark/nejmaibench_llmjudge_gen.py create mode 100644 opencompass/datasets/nejmaibench.py diff --git a/opencompass/configs/datasets/nejm_ai_benchmark/nejmaibench_gen.py b/opencompass/configs/datasets/nejm_ai_benchmark/nejmaibench_gen.py new file mode 100644 index 00000000..bd1e741b --- /dev/null +++ b/opencompass/configs/datasets/nejm_ai_benchmark/nejmaibench_gen.py @@ -0,0 +1,67 @@ +from opencompass.datasets import nejmaibenchDataset, nejmaibenchEvaluator +from opencompass.openicl.icl_inferencer import GenInferencer +from opencompass.openicl.icl_prompt_template import PromptTemplate +from opencompass.openicl.icl_retriever import ZeroRetriever + +import os + +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' + +# 将相对于当前文件的相对路径转换为绝对路径 +def to_abs_path(relative_path: str) -> str: + # 当前脚本所在目录 + base_dir = os.path.dirname(os.path.abspath(__file__)) + # 拼接并规范化绝对路径 + abs_path = os.path.abspath(os.path.join(base_dir, relative_path)) + return abs_path + +# Reader configuration +reader_cfg = dict( + input_columns=[ + 'question', + 'options', + 'Subject', + 'prompt_mode', + + ], + output_column='label', +) + +# 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=nejmaibenchEvaluator), + pred_role='BOT', +) +nejmaibench_dataset = dict( + type=nejmaibenchDataset, + abbr='nejmaibench', + path=to_abs_path('data/NEJM_All_Questions_And_Answers.csv'), + prompt_mode='zero-shot', + reader_cfg=reader_cfg, + infer_cfg=infer_cfg, + eval_cfg=eval_cfg, + +) + +nejmaibench_datasets = [nejmaibench_dataset] diff --git a/opencompass/configs/datasets/nejm_ai_benchmark/nejmaibench_llmjudge_gen.py b/opencompass/configs/datasets/nejm_ai_benchmark/nejmaibench_llmjudge_gen.py new file mode 100644 index 00000000..38bd2e8e --- /dev/null +++ b/opencompass/configs/datasets/nejm_ai_benchmark/nejmaibench_llmjudge_gen.py @@ -0,0 +1,116 @@ +from opencompass.datasets import nejmaibenchDataset, nejmaibench_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 +import os + +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. + + : 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\n\n + : \n{label}\n\n\n + : \n{prediction}\n\n\n + Judging the correctness of candidates' answers: +""".strip() + + +# 将相对于当前文件的相对路径转换为绝对路径 +def to_abs_path(relative_path: str) -> str: + # 当前脚本所在目录 + base_dir = os.path.dirname(os.path.abspath(__file__)) + # 拼接并规范化绝对路径 + abs_path = os.path.abspath(os.path.join(base_dir, relative_path)) + return abs_path + +# Reader configuration +reader_cfg = dict( + input_columns=[ + 'question', + 'options', + 'Subject', + 'prompt_mode', + + ], + output_column='label', +) + + +# 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=nejmaibenchDataset, + path=to_abs_path('data/NEJM_All_Questions_And_Answers.csv'), + prompt_mode='zero-shot', + reader_cfg=reader_cfg, + ), + judge_cfg=dict(), + dict_postprocessor=dict(type=nejmaibench_llmjudge_postprocess), + ), +) + + +nejmaibench_dataset = dict( + type=nejmaibenchDataset, + abbr='nejmaibench', + path=to_abs_path('data/NEJM_All_Questions_And_Answers.csv'), + prompt_mode='zero-shot', + reader_cfg=reader_cfg, + infer_cfg=infer_cfg, + eval_cfg=eval_cfg, + +) + +nejmaibench_datasets = [nejmaibench_dataset] diff --git a/opencompass/datasets/__init__.py b/opencompass/datasets/__init__.py index b00162d1..c15bc1fb 100644 --- a/opencompass/datasets/__init__.py +++ b/opencompass/datasets/__init__.py @@ -108,6 +108,7 @@ from .musr import * # noqa: F401, F403 from .narrativeqa import * # noqa: F401, F403 from .natural_question import * # noqa: F401, F403 from .natural_question_cn import * # noqa: F401, F403 +from .nejmaibench import * # noqa: F401, F403 from .NPHardEval import * # noqa: F401, F403 from .obqa import * # noqa: F401, F403 from .olymmath import * # noqa: F401, F403 diff --git a/opencompass/datasets/nejmaibench.py b/opencompass/datasets/nejmaibench.py new file mode 100644 index 00000000..63882f76 --- /dev/null +++ b/opencompass/datasets/nejmaibench.py @@ -0,0 +1,251 @@ +import re + +import pandas as pd +from datasets import Dataset + +from opencompass.openicl import BaseEvaluator +from opencompass.registry import LOAD_DATASET, TEXT_POSTPROCESSORS +from opencompass.utils import get_logger + +from .base import BaseDataset + + +def _parse(item, prompt_mode): + # 1. 从 Choices 字符串里按行拆分出每个选项 + raw_choices = item.get('Choices', '') + # 去掉首尾空白并按行分割，过滤掉空行 + lines = [ + line.strip() for line in raw_choices.strip().splitlines() + if line.strip() + ] + + # 2. 用正则去掉行首的 "A. "/"B. " 等前缀，只保留选项内容 + options_list = [re.sub(r'^[A-Z]\.\s*', '', line) for line in lines] + + # 3. 写回 item + item['options'] = options_list + + # 4. 重建带标号的选项字符串 + options_str = '\n'.join(f'{chr(65 + i)}. {opt}' + for i, opt in enumerate(options_list)) + + # 5. 构造 question、label、prompt_mode、start、end + item['question'] = f"{item['Question']}\n{options_str}" + item['label'] = item['Answer'] + item['prompt_mode'] = prompt_mode + item['start'] = chr(65) + item['end'] = chr(65 + len(options_list) - 1) + return item + + +@LOAD_DATASET.register_module() +class nejmaibenchDataset(BaseDataset): + + @staticmethod + def load(path: str, prompt_mode: str = 'zero-shot', **kwargs): + # 读取 CSV 文件为 DataFrame，并将 NaN 转为空字符串 + df = pd.read_csv(path, encoding='utf-8') + df = df.fillna('') + + # 转换为字典列表 + data_list = df.to_dict(orient='records') + + # 将数据列表包装为 Dataset + dataset = Dataset.from_list(data_list) + + # 根据提示模式进行解析 + if prompt_mode == 'zero-shot': + dataset = dataset.map(lambda item: _parse(item, prompt_mode)) + elif prompt_mode == 'few-shot': + pass # TODO: Implement few-shot prompt handling + return dataset + + +class nejmaibenchEvaluator(BaseEvaluator): + + def score(self, predictions, references, test_set): + method = test_set['prompt_mode'][0] + + if len(predictions) != len(references): + return {'error': 'preds and refrs have different length'} + correct = 0 + count = 0 + details = [] + for idx, (i, j) in enumerate(zip(predictions, references)): + i = answer_cleansing(method, i, test_set['options'][idx], + test_set['label'][idx]) + detail = { + 'pred': i, + 'answer': j, + 'correct': False, + 'Subject': test_set['Subject'][idx], + } + count += 1 + if i == j: + correct += 1 + detail['correct'] = True + details.append(detail) + result = {'accuracy': 100 * correct / count, 'details': details} + return result + + +@TEXT_POSTPROCESSORS.register_module() +def answer_cleansing( + method: str, + prediction: str, + options: list, + label: str, +) -> str: + + # Clean up unwanted phrases in the prediction + for unwanted_phrase in [ + 'I understand', + 'A through J', + 'A through E', + 'A through D', + ]: + prediction = prediction.replace(unwanted_phrase, '') + + options_num = len(options) + options = [chr(65 + i) for i in range(options_num)] + options_str = r'\b(' + '|'.join(options) + r')\b' + prediction = re.findall(options_str, prediction) + + if len(prediction) == 0: + prediction = [] + return prediction + else: + # If there is a "label" and its length is 1, + # process prediction accordingly + if len(label) == 1: + if method == 'few-shot': + answer_flag = True if len(prediction) > 1 else False + # choose the first or last element based on the answer_flag + if answer_flag: + prediction = [prediction[0]] + else: + prediction = [prediction[-1]] + elif method == 'zero-shot': + # choose the first element in list + prediction = [prediction[0]] + else: + raise ValueError('Method is not properly defined ...') + + # Remove trailing period if it exists + if prediction[0] and prediction[0].endswith('.'): + prediction[0] = prediction[0][:-1] + + return prediction[0] + + +def _generic_llmjudge_postprocess(judgement: str): + match = re.search(r'(A|B)', judgement) + grade_letter = (match.group(0) if match else 'B' + ) # Default to "INCORRECT" if no match + return grade_letter + + +def nejmaibench_llmjudge_postprocess( + output: dict, + output_path: str, + dataset: Dataset, +) -> dict: + # Get the original dataset + original_dataset = dataset.reader.dataset['test'] + + judged_answers = [] + original_responses = [] + references = [] + details = [] + + # Initialize statistics dictionaries + stats = {'Subject': {}} + + total_correct = 0 + total_count = 0 + + # Process each sample + for k, v in output.items(): + idx = int(k) # Convert key to integer for indexing + original_responses.append(v['prediction']) + + processed_judge = _generic_llmjudge_postprocess(v['prediction']) + + # Get category information from the dataset + sample = original_dataset[idx] + subject = sample.get('Subject', 'unknown') + + # Initialize category stats if not exists + for level, key in [ + ('Subject', subject), + ]: + if key not in stats[level]: + stats[level][key] = {'correct': 0, 'total': 0} + + # Record the judgment + if processed_judge is not None: + judged_answers.append(processed_judge) + try: + gold = v['gold'] + references.append(gold) + except KeyError: + get_logger().warning( + f'No gold answer for {k}, use empty string as reference!') + gold = '' + references.append('') + + # Check if the answer is correct (A means correct) + is_correct = processed_judge == 'A' + total_count += 1 + + if is_correct: + total_correct += 1 + # Update category stats + for level, key in [ + ('Subject', subject), + ]: + stats[level][key]['correct'] += 1 + + # Update category totals + for level, key in [ + ('Subject', subject), + ]: + stats[level][key]['total'] += 1 + # Add to details + details.append({ + 'id': k, + 'question': sample['question'], + 'options': sample['options'], + 'origin_prompt': v['origin_prompt'], + 'llm_judge': processed_judge, + 'gold': gold, + 'is_correct': is_correct, + 'Subject': subject, + }) + + # Calculate overall accuracy with two decimal places + overall_accuracy = (round( + (total_correct / total_count * 100), 2) if total_count > 0 else 0.00) + + # Initialize results dictionary + results = { + 'accuracy': overall_accuracy, + 'total_correct': total_correct, + 'total_count': total_count, + 'details': details, + } + + # Calculate accuracy for each category and flatten into results + for level in stats: + for key, value in stats[level].items(): + if value['total'] > 0: + # Calculate accuracy with two decimal places + accuracy = round((value['correct'] / value['total'] * 100), 2) + + # Create a flattened key for the category + flat_key = f'nejmaibench-{key}' + + # Add to results + results[flat_key] = accuracy + + return results