an error in training phase
Jillian2017 opened this issue · 0 comments
while I'm running the train.py, an error appears. the detail information is attached.
the 1 epoch
Variable containing:
0.2366
[torch.cuda.FloatTensor of size 1 (GPU 0)]
Traceback (most recent call last):
File "train.py", line 163, in
loss.backward(create_graph=True)
File "/usr/lib64/python2.7/site-packages/torch/autograd/variable.py", line 167, in backward
torch.autograd.backward(self, gradient, retain_graph, create_graph, retain_variables)
File "/usr/lib64/python2.7/site-packages/torch/autograd/init.py", line 99, in backward
variables, grad_variables, retain_graph)
RuntimeError: function SignBackward returned an incorrect number of gradients (expected 2, got 1)
Does anyone meet the problem before?