RuntimeError: unable to write to file ＜/torch_2818186_3541370554＞ ERROR: Unexpected bus error encoun

在容器中，训练yolov5时，报如上问题。pytorch将共享内存的临时文件保存在了/torch_xxx文件中，即容器中的根目录下。容器磁盘空间不足导致该问题的发生。目前可以通过以下代码暂时关闭pytorch的shared memory功能来规避：

将此代码贴到你启动代码的最前面，顺利解决：

import sys
import torch
from torch.utils.data import dataloader
from torch.multiprocessing import reductions
from multiprocessing.reduction import ForkingPickler

default_collate_func = dataloader.default_collate


def default_collate_override(batch):
  dataloader._use_shared_memory = False
  return default_collate_func(batch)

setattr(dataloader, 'default_collate', default_collate_override)

for t in torch._storage_classes:
  if sys.version_info[0] == 2:
    if t in ForkingPickler.dispatch:
        del ForkingPickler.dispatch[t]
  else:
    if t in ForkingPickler._extra_reducers:
        del ForkingPickler._extra_reducers[t]