Closed crenaudineau1 closed 3 years ago
Hello,
I am on Windows, python 3.8.8, pytorch 1.8.0 and cuda 10.2.
I need to fix num workers to 0 in dataloader (customdataset) for a working code but it is very slow. If I let num_workers > 0 I have a error.
I have update test.py with :
if name == 'main': test()
But the problem persists. Someone have fix this issue on Windows ? How ?
Thanks a lot.
I've never tried this code (actually non of deep learning codes..) in Windows. Please let me know if you find a solution!
Thanks
Hello,
I am on Windows, python 3.8.8, pytorch 1.8.0 and cuda 10.2.
I need to fix num workers to 0 in dataloader (customdataset) for a working code but it is very slow. If I let num_workers > 0 I have a error.
I have update test.py with :
if name == 'main': test()
But the problem persists. Someone have fix this issue on Windows ? How ?
Thanks a lot.