基于儿科问答语料，利用lora微调Qwen2.5-0.5B-instruct模型。这里上传的模型是经过微调20个epoch后的lora模型，下载后可与Qwen2.5-0.5B-instruct进行merge.
Based on pediatric question-and-answer corpus, fine-tune the Qwen2.5-0.5B-instruct model using lora. The model uploaded here is a lora model after 20 epochs of fine-tuning and can be downloaded to merge with Qwen2.5-0.5B-instruct.
