zz三种方法实现监督微调:LLaMA Factory, trl 和 unsloth

https://www.luochang.ink/posts/sft_note/

三种方法实现监督微调:LLaMA Factory, trl 和 unsloth

 

下载模型的方法

https://github.com/luochang212/sft-note/blob/main/model/download_qwen.py

# USAGE: python download_qwen.py

from modelscope import snapshot_download

model_dir = snapshot_download('Qwen/Qwen2.5-7B-Instruct', cache_dir='./')
# model_dir = snapshot_download('Qwen/Qwen2.5-0.5B-Instruct', cache_dir='./')

print(f"model_dir: {model_dir}")

 

posted @ 2025-12-30 15:41  blcblc  阅读(7)  评论(0)    收藏  举报