ocr pytorch训练异常

[0/300][12160/51244] Loss: 162.309189

Traceback (most recent call last):
  File "crnn_main_v2.py", line 186, in <module>
    main(crnn, train_loader, val_loader, criterion, optimizer)
  File "crnn_main_v2.py", line 116, in main
    train(crnn, train_loader, criterion, Iteration)
  File "crnn_main_v2.py", line 101, in train
    cost.backward()
  File "/usr/local/lib/python3.6/site-packages/torch/tensor.py", line 166, in backward
    torch.autograd.backward(self, gradient, retain_graph, create_graph)
  File "/usr/local/lib/python3.6/site-packages/torch/autograd/__init__.py", line 99, in backward
    allow_unreachable=True)  # allow_unreachable flag
  File "/usr/local/lib/python3.6/site-packages/torch/utils/data/_utils/signal_handling.py", line 66, in handler
    _error_if_any_worker_fails()
RuntimeError: DataLoader worker (pid 26226) is killed by signal: Killed