[BUG/Help] 在ptuning中的main.py中，第130行model = model.half()，采用cpu加载，造成超内存报错

[zhuzongjian1]

Is there an existing issue for this?

• [X] I have searched the existing issues

Current Behavior

进行chatglm2 ptuning时，sh train.sh超内存

Expected Behavior

No response

Steps To Reproduce

1、放入官方数据集 2、进入ptuning文件夹，vim train.sh 3、 增加RUNNING_STEP=400；修改model_path 4、sh train.sh

Environment

- OS:ubuntu1804
- Python:3.10
- Transformers:4.30.2
- PyTorch:2.1.0
- CUDA Support (`python -c "import torch; print(torch.cuda.is_available())"`) :True

Anything else?

No response

[zhuzongjian1]

改成model = model.half().cuda()