I have tried to run the test code, but it still raised a runtime error, as
0 images have been proceeded
test.py:128: UserWarning: volatile was removed and now has no effect. Use with torch.no_grad(): instead.
prediction = net(Variable(torch.from_numpy(image_scale), volatile=True).cuda())
Traceback (most recent call last):
File "test.py", line 208, in
main()
File "test.py", line 199, in main
output = predict(model, image.numpy(), (np.asscalar(ori_size[0]), np.asscalar(ori_size[1])), is_mirror=args.is_mirror, scales=[1])
File "test.py", line 128, in predict
prediction = net(Variable(torch.from_numpy(image_scale), volatile=True).cuda())
File "/root/anaconda3/envs/pytorch/lib/python3.6/site-packages/torch/nn/modules/module.py", line 491, in call
result = self.forward(*input, *kwargs)
File "/xxx/CE2P/models.py", line 278, in forward
x = self.bn1(x)
File "/root/anaconda3/envs/pytorch/lib/python3.6/site-packages/torch/nn/modules/module.py", line 491, in call
result = self.forward(input, **kwargs)
File "/xxx/CE2P/modules/bn.py", line 184, in forward
self.activation, self.slope)
File "/xxx/CE2P/modules/functions.py", line 218, in forward
x, x, ctx.eps)
File "/xxx/CE2P/modules/functions.py", line 16, in _check
raise RuntimeError("CUDA Error encountered in {}".format(fn))
RuntimeError: CUDA Error encountered in <function CompiledLib.bn_forward_cuda at 0x7faf37cc8f28>
The configuration is: pytorch 0.4.0, CUDA 8.0, GPU P100.
I am sorry for my fault that I forget to change the CUDA_GENCODE. When I changed it to CUDA_GENCODE=-gencode=arch=compute_60,code=sm_60, it works.
Thanks,
Hi,
I have tried to run the test code, but it still raised a runtime error, as
The configuration is: pytorch 0.4.0, CUDA 8.0, GPU P100.
Thanks,