Ya Wen

icon indicating an email [email protected]

icon company @https://github.com/NLPIR-team，@https://github.com/nusnlp icon location lIT+USTC+CAS+BAIDU+AISPEECH+ALIBABA+Dream icon location Chinese Academy of Sciences; Research: Natural Language Understanding & Computer Vision & ASR; My Thesis Title: Multimodal Machine Translation; AI Architect

Results 1 comments of Ya Wen

deepspeed报错

原因在于，这一块进程管理没有做好：NPROC_PER_NODE 和 --nproc_per_node 的差别，多卡运行的时候，请在运行脚本设置NPROC_PER_NODE，不要在swift sft 运行入参传 --device_map， --nproc_per_node ，因为这个会被transformer框架进程干预。 The reason is that the process management in this area is not done well: the difference between NPROC_PER_NODE and --nproc_per_node. When...