Socket error Event: 32 Error: 10053.
Closed this issue · 1 comments
Littlechickencub commented
Connection closing...Socket close. Connection closed by foreign host. Disconnected from remote host(Cy_lap) at 12:03:51.我用的Xshell 7 连的实验室的服务器,使用places2数据集训练LaMa,但是经常会报这个错误,就是训着训着突然就报这个错误,今天是在第一轮保存模型到5380的时候又报了这个错误,上次是有这个提醒torch.multiprocessing.spawn.ProcessExitedException: process 1 terminated with signal SIGKILL,我就改小了num_workers为8,batch_size为8。我的设备型号是 GTX1070 4个,每个GPU显存都是8GB,可用内存104G,磁盘可用内存483G,CPU个数2个,核心数12,processor:48,型号Intel(R) Xeon(R) CPU E5-2678 v3 @ 2.50GHz,请问怎么改?感谢
DQiaole commented
你好,我们没有遇到过这个问题。