Open zhuzongjian1 opened 11 months ago
进行chatglm2 ptuning时,sh train.sh超内存
No response
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
- 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
改成model = model.half().cuda()
Is there an existing issue for this?
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
Anything else?
No response