QwenLM / Qwen

The official repo of Qwen (通义千问) chat & pretrained large language model proposed by Alibaba Cloud.
Apache License 2.0
12.47k stars 1.01k forks source link

如何修改模型的结构 #1290

Closed zzc-1024 closed 1 week ago

zzc-1024 commented 2 weeks ago

我想要通过修改模型的结构,实现让模型在保留原来的lm_head和权重的基础上,新增一个用于标注的lm_head,但是我发现transformers并没有提供修改模型结构的相关接口,请问有没有什么办法可以修改模型的结构(使用torch或者tf实现均可)。

jklj077 commented 1 week ago

I don't quite understand. In the modeling code, transformers modules are PyTorch nn.Modules, and you can just modify the code as you wish.