about ERFNet ， test_erfnet.py

cardwing / Codes-for-Lane-Detection

Learning Lightweight Lane Detection CNNs by Self Attention Distillation (ICCV 2019)

MIT License

1.04k stars 333 forks source link

about ERFNet ， test_erfnet.py #155

Closed noobliang closed 5 years ago

noobliang commented 5 years ago

why I run the test_erfnet.py ,it has no response，and no displaying error , , I check it ，its stuck in model = torch.nn.DataParallel(model, device_ids=range(args.gpus)).cuda() .., , ,

cardwing commented 5 years ago

What is your video memory? It may be caused by insufficient video memory.

noobliang commented 5 years ago

What is your video memory? It may be caused by insufficient video memory.

I had set gpus 0, imagesize 256*512,batchsize 1,,, but I found it just use my cpu, a core is 100% ,,,,the using rate of gpu is 0% ... I dont know why..

cardwing commented 5 years ago

You may need to check if pytorch-GPU is correctly installed.

noobliang commented 5 years ago

You may need to check if pytorch-GPU is correctly installed.

but if torch.cuda.is_available(): is ok, 　　thanks for your reply ,, let me ponder it then..