DDP training question · pytorch/examples#1096

(2 评论) (0 反应) (1 负责人)Python (21,634 star) (9,429 fork)batch import

distributedhelp wanted

描述

Hi, I'm using the tutorial https://github.com/pytorch/tutorials/blob/master/intermediate_source/ddp_tutorial.rst for DDP train,using 4 gpus in myself code, reference Basic Use Case. But when I finished the modification, it was stuck during run the demo,meanwhile,video memory has been occupied.Could you help me?

技术栈: python
领域: machine learningbackend
议题类型: bug
难度: 5
预计时间: 1-3 hours
活动状态: stale
清晰度: mostly clear
前置要求: PyTorch basicsDistributed training conceptsCUDA environment
新手友好度: 20
研究方向: The user is stuck while running DDP training with 4 GPUs based on a PyTorch tutorial. The issue may arise from version incompatibility or misconfiguration. The assignee should check the referenced tutorial file and provide guidance. More details about the error logs are needed.