完整信息：
-- Process 1 terminated with the following error:
Traceback (most recent call last):
File "/home/lzk/anaconda3/lib/python3.7/site-packages/torch/multiprocessing/spawn.py", line 19, in _wrap
fn(i, *args)
File "/home/lzk/IJCAI2021/GraphWriter-DGL/train.py", line 278, in main
train_loss = train_one_epoch(model, train_dataloader, optimizer, args, epoch, rank, device, writer)
File "/home/lzk/IJCAI2021/GraphWriter-DGL/train.py", line 63, in train_one_epoch
for _i , batch in enumerate(dataloader):
File "/home/lzk/anaconda3/lib/python3.7/site-packages/torch/utils/data/dataloader.py", line 278, in __iter__
return _MultiProcessingDataLoaderIter(self)
File "/home/lzk/anaconda3/lib/python3.7/site-packages/torch/utils/data/dataloader.py", line 682, in __init__
w.start()
File "/home/lzk/anaconda3/lib/python3.7/multiprocessing/process.py", line 112, in start
self._popen = self._Popen(self)
File "/home/lzk/anaconda3/lib/python3.7/multiprocessing/context.py", line 223, in _Popen
return _default_context.get_context().Process._Popen(process_obj)
File "/home/lzk/anaconda3/lib/python3.7/multiprocessing/context.py", line 284, in _Popen
return Popen(process_obj)
File "/home/lzk/anaconda3/lib/python3.7/multiprocessing/popen_spawn_posix.py", line 32, in __init__
super().__init__(process_obj)
File "/home/lzk/anaconda3/lib/python3.7/multiprocessing/popen_fork.py", line 20, in __init__
self._launch(process_obj)
File "/home/lzk/anaconda3/lib/python3.7/multiprocessing/popen_spawn_posix.py", line 47, in _launch
reduction.dump(process_obj, fp)
File "/home/lzk/anaconda3/lib/python3.7/multiprocessing/reduction.py", line 60, in dump
ForkingPickler(file, protocol).dump(obj)
File "/home/lzk/anaconda3/lib/python3.7/site-packages/torch/multiprocessing/reductions.py", line 314, in reduce_storage
metadata = storage._share_filename_()
RuntimeError: unable to open shared memory object </torch_30603_1696564530> in read-write mode用指令ulimit -a来查看当前用户的各项limit限制：
我的问题出现在dataloader，查了一番博客，主要有2种解决方式：
- 增大空间：ulimit -SHn 51200；
- 限制number_workers为0(但是杀敌一千，自损八百，这么做以后我的epoch时间慢了一倍)
- 换成小数据集就不会有这样的错误，但同样治标不治本