RuntimeError: cuda runtime error (59) : device-side assert triggered at /pytorch/aten/src/THC/generic/THCTensorCopy.cpp:70

When I run the segmentation codes , i got this error!

It seems that there is a problem in calculating the loss function！The loss value cannot be calculated.

I run this codes in pytorch0.4.1 and python3 and opencv 3.2.0

Does your machine has GPU with CUDa?

yes, it's cuda 8.0, cudnn5.1 and gpu is 1050ti!

You might want to upgrade it and try. I was using Cuda 9.0 and cudnn 7

OK I will try i, thanks!

Thanks, I have solved this problem. It turned out that I didn't convert 255 of the input label to 19, not because of the problem of the CUDA version.
There are 255 label in the voc12 data set. How can I handle this data set to use your network properly?

You can ignore it in the loss function.

nn.CrossEntropy(ignore_index=255)

If I set nClasses = 21，is 255 label also ignored in IOUEval.py ?

The easiest way is to do something like this:

ESPNetv2/segmentation/DataSet.py

Line 40 in df58a62

if 255 in np.unique(label):

For cityscapes, I map 255 to 19. For VOC, you can map 255 to 0

I really appreciate your answer, it's help me a lot. I'll try the approach you suggested.
Thanks !

When you test ESPNetv2 on camvid dataset, did you augment the dataset and did you ignore the background classes when training ？

We didn’t train on CamVid dataset.

sorry，I was mistaken.