nlp大模型-实习笔记1

需求：Excel转json/jsonl

代码：

# 本代码功能：把Excel变成格式规范的json/jsonl文件
import pandas as pd
import json

path = r'/Users/mac/Desktop/test_le/m.xlsx'
data = pd.read_excel(path)

output_file = 'm.jsonl'


with open(output_file,'w',encoding='utf-8') as f:        
    for index,row in data.iterrows():
        qa_pairs = row['faq']
        # 提取这一列当中'question'的键值
        Q = row['question']
        # 变成json类型
        qa_json = json.loads(qa_pairs)
        # 修剪qa格式，输出键'q'及对应的值
        trimmed_qa = [{"q":item["questions"]} for item in qa_json]
        # 把键'Q'和'q'合并
        final_data = {"Q":Q},{"questions":trimmed_qa}
        #将数据保存为JSONL文件
        data_str = json.dumps(final_data,ensure_ascii=False)
        f.write(data_str + '\n')