peft使用prefix tuning时报错

[Lyn0222]

在使用mindnlp.peft进行prefix tuning的时候报错：

Traceback (most recent call last):
  File "E:\git\biogpt\finetune.py", line 144, in <module>
    loss = grad_fn(**batch)
  File "D:\install\anaconda\envs\biogpt\lib\site-packages\mindnlp\core\autograd\function.py", line 35, in value_and_grad_f
    values = fn_(*args, **kwargs)
  File "E:\git\biogpt\finetune.py", line 131, in forward_fn
    outputs = model(**batch)
  File "D:\install\anaconda\envs\biogpt\lib\site-packages\mindnlp\core\nn\modules\module.py", line 391, in _wrapped_call_impl
    return self._call_impl(*args, **kwargs)
  File "D:\install\anaconda\envs\biogpt\lib\site-packages\mindnlp\core\nn\modules\module.py", line 405, in _call_impl
    return forward_call(*args, **kwargs)
  File "D:\install\anaconda\envs\biogpt\lib\site-packages\mindnlp\peft\peft_model.py", line 557, in forward
    prompts = prompts.to(inputs_embeds.dtype)
AttributeError: 'tuple' object has no attribute 'to'

• 硬件环境

Windows+cpu

• 软件环境: -- MindSpore version (e.g., 1.7.0.Bxxx) :2.5.0 -- Python version (e.g., Python 3.7.5) :3.10 -- OS platform and distribution (e.g., Linux Ubuntu 16.04):windows -- GCC/Compiler version (if compiled from source):

原因在于get_prompt函数输出为一个元组，根本上是以下代码导致：

if peft_config.num_transformer_submodules == 2:
                past_key_values = ops.cat([past_key_values, past_key_values], dim=2)
            past_key_values = past_key_values.permute([2, 0, 3, 1, 4]).split(
                peft_config.num_transformer_submodules * 2
            )

进行split后输出的为元组，没有to方法导致报错。 修改建议： 增加类型转换即可解决

Screenshots/ 日志 / 截图 (Mandatory / 必填) If applicable, add screenshots to help explain your problem.

Additional context / 备注 (Optional / 选填) Add any other context about the problem here.