search

uvipen / Super-mario-bros-A3C-pytorch

Asynchronous Advantage Actor-Critic (A3C) algorithm for Super Mario Bros
MIT License
1.04k stars 228 forks source link

RuntimeError: cuda runtime error (801) #18

Open arkseal opened 3 years ago

arkseal commented 3 years ago

Hi, when I try to run this code using CUDA I always am getting this error. It works fine on CPU and other programs work fine using CUDA. I tried to search online but I didn't find anything of use. Here's the entire Traceback: THCudaCheck FAIL file=..\torch/csrc/generic/StorageSharing.cpp line=253 error=801 : operation not supported Traceback (most recent call last): File ".\train.py", line 84, in train(opt) File ".\train.py", line 73, in train process.start() File "C:\Users\Aarush\AppData\Local\Programs\Python\Python38\lib\multiprocessing\process.py", line 121, in start self._popen = self._Popen(self) File "C:\Users\Aarush\AppData\Local\Programs\Python\Python38\lib\multiprocessing\context.py", line 327, in _Popen return Popen(process_obj) File "C:\Users\Aarush\AppData\Local\Programs\Python\Python38\lib\multiprocessing\popen_spawn_win32.py", line 93, in init reduction.dump(process_obj, to_child) File "C:\Users\Aarush\AppData\Local\Programs\Python\Python38\lib\multiprocessing\reduction.py", line 60, in dump ForkingPickler(file, protocol).dump(obj) File "C:\Users\Aarush\Documents\Visual Studio Code\Maching_Learning\machine-learning-env38\lib\site-packages\torch\multiprocessing\reductions.py", line 240, in reduce_tensor event_sync_required) = storage._sharecuda() RuntimeError: cuda runtime error (801) : operation not supported at ..\torch/csrc/generic/StorageSharing.cpp:253 (machine-learning-env38) PS C:\Users\Aarush\Documents\Visual Studio Code\Maching_Learning\Super-mario-bros-A3C-pytorch> Traceback (most recent call last): File "", line 1, in File "C:\Users\Aarush\AppData\Local\Programs\Python\Python38\lib\multiprocessing\spawn.py", line 116, in spawn_main exitcode = _main(fd, parent_sentinel) File "C:\Users\Aarush\AppData\Local\Programs\Python\Python38\lib\multiprocessing\spawn.py", line 126, in _main self = reduction.pickle.load(from_parent) EOFError: Ran out of input

qf-123 commented 2 years ago

the same problem😟

arkseal commented 2 years ago

I think it worked when I used Linux, try using wsl if you want to stay on Windows