Cuda Illegal Memory Access

[sabrinazuraimi]

Sorry for always opening up issues here.. I tried looking around online, and I think the illegal memory access might have to do with the dataset that I'm using?(SUNRGBD dataset) I can't think of any other reason for this error so would be much appreciated if you could help.

CustomDatasetDataLoader dataset [sunrgbd_dataset] was created

training images = 5285

model [BaseModel] was created create web directory ./checkpoints/test_depthconv/web... error in depthconv_col2im: an illegal memory access was encountered THCudaCheck FAIL file=/opt/conda/conda-bld/pytorch_1513368888240/work/torch/lib/THC/generic/THCTensorMath.cu line=26 error=77 : an illegal memory access was encountered Traceback (most recent call last): File "train.py", line 55, in model.backward(total_steps, opt.nepochs dataset.len() opt.batchSize + 1) File "/disks/local/Documents/DepthAwareCNN-master/models/Deeplab.py", line 107, in backward self.loss.backward() File "/home/muser/.pyenv/versions/anaconda3-5.0.1/lib/python3.6/site-packages/torch/autograd/variable.py", line 167, in backward torch.autograd.backward(self, gradient, retain_graph, create_graph, retain_variables) File "/home/muser/.pyenv/versions/anaconda3-5.0.1/lib/python3.6/site-packages/torch/autograd/init.py", line 99, in backward variables, grad_variables, retain_graph) File "/disks/local/Documents/DepthAwareCNN-master/models/ops/depthconv/functions/depthconv.py", line 91, in backward gradweight = weight.new(*weight.size()).zero() RuntimeError: cuda runtime error (77) : an illegal memory access was encountered at /opt/conda/conda-bld/pytorch_1513368888240/work/torch/lib/THC/generic/THCTensorMath.cu:26

When I tried running it again today, the error no longer appeared(did not change anything with the code..)

[garrett-cn]

I have the same problem.Have you solved it?