cannot allocate memory

[luhc15]

the problem arise when run after 1epoch, but works well in first train epoch as well as the followed 1st validate epoch. batch_size is 8 and work_number is 2。why this happened ?
File "/home/luhongchao/anaconda2/lib/python2.7/site-packages/torch/utils/data/dataload
er.py", line 301, in iter
return DataLoaderIter(self)
File "/home/luhongchao/anaconda2/lib/python2.7/site-packages/torch/utils/data/dataload
er.py", line 158, in init
w.start()
File "/home/luhongchao/anaconda2/lib/python2.7/multiprocessing/process.py", line 130,
in start
self._popen = Popen(self)
File "/home/luhongchao/anaconda2/lib/python2.7/multiprocessing/forking.py", line 121,
in init
self.pid = os.fork()
OSError: [Errno 12] Cannot allocate memory

[zou3519]

It sounds like you don't have enough RAM in your machine.

[Lmy0217]

@luhc15 @zou3519 Do you have enough swap space? I got this error, then I created swap(reference this) and fixed it.

[glenn-jocher]

Recommend this. Setting swap space equal to RAM space and swappiness=0 worked for me.
https://www.digitalocean.com/community/tutorials/how-to-add-swap-space-on-ubuntu-16-04