export模型oom - Githubissues

hiyouga / LLaMA-Factory

Unified Efficient Fine-Tuning of 100+ LLMs (ACL 2024)

https://arxiv.org/abs/2403.13372

Apache License 2.0

34.32k stars 4.23k forks source link

Closed yangyang6666 closed 1 hour ago

yangyang6666 commented 4 hours ago

使用NPU训练完成，导出时报错。

No response

No response

yangyang6666 commented 3 hours ago

hiyouga commented 1 hour ago

export_device: cpu