modelscope / modelscope/ms-swift
torch.npu.mem_get_info 出现 got multiple values for argument 'device' 错误
Nobody has claimed this yet.
- Dominant language
- Python
- Stars
- 15.7k
- Forks
- 1.7k
- Avg merge
- 1d 16h
- Merged PRs (30d)
- 136
Description
Checklist / 检查清单
- I have searched existing issues, and this is a new bug report. / 我已经搜索过现有的 issues，确认这是一个新的 bug report。
Bug Description / Bug 描述
infer时，使用partial绑定device导致出现 got multiple values for argument 'device' 错误。
How to Reproduce / 如何复现
复现环境：quay.io/ascend/vllm-ascend:v0.16.0rc1
复现脚本：
swift infer --val_dataset "$INFER_DATASET"
--infer_backend vllm
--vllm_gpu_memory_utilization 0.8
--vllm_max_num_seqs $VLLM_GLOBAL_BATCH_SIZE
--model Qwen/Qwen3-VL-2B
--torch_dtype bfloat16
--write_batch_size $VLLM_GLOBAL_BATCH_SIZE
--max_new_tokens 2048
--enable_thinking false
--temperature 0.0
--result_path "$RESULT_PATH"
Additional Information / 补充信息
暂时可通过包装一层解决。
def _mem_get_info_wrapper(device=None):
return original_mem_get_info(device=vllm_device)
Contributor guide
First steps
- Read the whole issue, then the project's contributing guide.
- Comment on the issue to say you are picking it up — it saves two people doing the same work.
- Fork the repository and make your change on a branch.
- Open a pull request that references the issue number.
Research direction
Start at swift/infer_engine/utils.py around line 369 and inspect how partial binds the device argument for torch.npu.mem_get_info. Reproduce with the provided swift infer command in the listed vllm-ascend environment, then verify that inference no longer raises the duplicate device-argument error and that the existing device selection still works.
Written by the indexing model from the issue text.
Assessment
- Tech stack
- python, pytorch
- Domain
- backend, machine-learning
- Issue type
- Bug
- Difficulty
- 3/5
- Estimated time
- 1-2 days
- Activity status
- Quiet
- Clarity
- Clearly specified
- Newbie friendliness
- 48/100