【python】读取xlsx文件，并测METEOR指标

import pandas as pd
import jieba
from nltk.translate.meteor_score import meteor_score

def preprocess_text(text):
    # 使用 jieba 进行中文分词
    words = jieba.lcut(text)
    return words

def calculate_meteor(reference, candidate):
    # 预处理文本
    reference = preprocess_text(reference)
    candidate = preprocess_text(candidate)


    print("-------------------")
    print(reference)
    print("-------------------")
    print(candidate)
    print("-------------------")

    # 计算METEOR
    score = meteor_score([reference], candidate)
    return score

def main():
    # 读取xlsx文件
    file_path = '路径'
    df = pd.read_excel(file_path)

    # 打印列名以进行调试
    print("Columns in the dataframe:", df.columns)

    # 假设数据有一列 'reference' 和五列 'candidate1', 'candidate2', 'candidate3', 'candidate4', 'candidate5'
    references = df['参考答案']
    candidate_columns = ['one', 'two', 'three', 'four', 'five']

    # 计算每个reference与每个candidate的METEOR指标
    for candidate in candidate_columns:
        if candidate not in df.columns:
            print(f"Column {candidate} not found in the dataframe.")
            continue

        meteor_scores = []
        for ref, cand in zip(references, df[candidate]):
            if pd.isna(ref) or pd.isna(cand):
                meteor_scores.append(0)
                continue

            meteor_score_value = calculate_meteor(ref, cand)
            meteor_scores.append(meteor_score_value)

        # 将METEOR指标添加到DataFrame中
        df[f'METEOR_{candidate}'] = meteor_scores

    # 保存结果到新的xlsx文件
    output_path = '路径'
    df.to_excel(output_path, index=False)

    print("Saved METEOR scores to:", output_path)

if __name__ == "__main__":
    main()