Trainbow

Results 12 comments of Trainbow

gpu number无法使用

> @Trainbow Is it convenient to post the yaml file for creating the test task? By the way, can it be successfully scheduled using the default scheduler of k8s? I...

ms2.2.11跑单机训练mindyolo模型报错：Failed to call bind, url: 127.0.0.1:2229 Address already in use

通过k8s的msjob启动任务会注入MS的环境变量 执行bash run.sh的命令 会报错显示端口已经被绑定占用 在脚本中加入unset来清除MS_的环境变量 然后再次运行k8s下发msjob的单机8卡任务，就可以正常hccl组网