[BUG] 用 llama-factory 微调，运行示例脚本报错

[vdogmcgee]

是否已有关于该错误的issue或讨论？ | Is there an existing issue / discussion for this?

• 我已经搜索过已有的issues和讨论 | I have searched the existing issues / discussions

该问题是否在FAQ中有解答？ | Is there an existing answer for this in FAQ?

• 我已经搜索过FAQ | I have searched FAQ

当前行为 | Current Behavior

环境配置好之后，运行 ：
CUDA_VISIBLE_DEVICES=0 llamafactory-cli train configs/minicpmo_2_6_lora_sft.yaml

报错：
File "python3.10/site-packages/transformers/tokenization_utils_base.py", line 803, in
self.data = {k: v.to(device=device) for k, v in self.data.items() if v is not None}
AttributeError: 'list' object has no attribute 'to'

我打印了self.data，是一个字典，其中一些 value 是 tensor，一些 value 是被 list 包裹起来的 tensor ， 直接 to.('cuda') 就会报这个错误

期望行为 | Expected Behavior

正常运行

复现方法 | Steps To Reproduce

照着官方给的 llama-factory 运行

运行环境 | Environment

- OS: ubuntu 22.04
- Python: 3.10
- Transformers: 4.44.2
- PyTorch: 2.3.0
- CUDA (`python -c 'import torch; print(torch.version.cuda)'`): 12.1

备注 | Anything else?

No response