在经过Ptuning微调之后,无论是通过ptuning/README.md中给的代码还是直接使用仓库中ptuning/web_demo.sh(代码中的模型和检查点路径都已修改正确),向程序发送提问均无输出,有一条警告消息(Some weights of ChatGLMForConditionalGeneration were not initialized from the model checkpoint at module and are newly initialized: ['transformer.prefix_encoder.embedding.weight']
You should probably TRAIN this model on a down-stream task to be able to use it for predictions and inference.)
Is there an existing issue for this?
Current Behavior
在经过Ptuning微调之后,无论是通过ptuning/README.md中给的代码还是直接使用仓库中ptuning/web_demo.sh(代码中的模型和检查点路径都已修改正确),向程序发送提问均无输出,有一条警告消息(Some weights of ChatGLMForConditionalGeneration were not initialized from the model checkpoint at module and are newly initialized: ['transformer.prefix_encoder.embedding.weight'] You should probably TRAIN this model on a down-stream task to be able to use it for predictions and inference.)
![图片](https://github.com/THUDM/ChatGLM2-6B/assets/79252396/cd158e7f-c9e8-41b1-9366-e467f067ef76)
Expected Behavior
No response
Steps To Reproduce
train.sh中设置per_device_train_batch_size=4、gradient_accumulation_steps=4、其余均不修改
Environment
Anything else?
训练集是在ptuning/README.md中给的生成广告词的训练集前加入了一些我们的个人信息和给模型的自我认知(很少几条,如图),测试集直接用了ptuning/README.md生成广告词的dev.json![图片](https://github.com/THUDM/ChatGLM2-6B/assets/79252396/4bc9101c-8a09-4f9f-ad8c-74605527d411)