docker运行pytorch报错：RuntimeError: DataLoader worker (pid 493) is killed by signal: Bus error. Details are lost due to multiprocessing. Rerunning with num_workers=0 may give better error trace.

Posted on | Edited on | In , | Views:
Symbols count in article: 375 | Reading time ≈ 1 mins.

使用docker运行Pytorch的时候如果报如下错误：

1
RuntimeError: DataLoader worker (pid 493) is killed by signal: Bus error. Details are lost due to multiprocessing. Rerunning with num_workers=0 may give better error trace.


应该是docker容器的共享内存空间不够导致的。先把docker当前该保存的状态保存好后，使用exit退出docker，重新docker run运行相应的镜像并加上--shm-size参数，例如--shm-size 10G代表使用10G的共享内存。

参考

https://github.com/pytorch/pytorch/issues/2244