AssertionError: Torch not compiled with CUDA enabled

[astatonn]

Hi, I'm trying to run DragGAN but I got this message error:

To create a public link, set share=Trueinlaunch(). Traceback (most recent call last): File "C:\Users\Lucas Lima\miniconda3\envs\draggan\lib\site-packages\gradio\routes.py", line 406, in run_predict event_data=event_data, File "C:\Users\Lucas Lima\miniconda3\envs\draggan\lib\site-packages\gradio\blocks.py", line 1300, in process_api fn_index, inputs, iterator, request, event_id, event_data File "C:\Users\Lucas Lima\miniconda3\envs\draggan\lib\site-packages\gradio\blocks.py", line 1037, in call_function utils.async_iteration, iterator, limiter=self.limiter File "C:\Users\Lucas Lima\miniconda3\envs\draggan\lib\site-packages\anyio\to_thread.py", line 32, in run_sync func, *args, cancellable=cancellable, limiter=limiter File "C:\Users\Lucas Lima\miniconda3\envs\draggan\lib\site-packages\anyio\_backends\_asyncio.py", line 937, in run_sync_in_worker_thread return await future File "C:\Users\Lucas Lima\miniconda3\envs\draggan\lib\site-packages\anyio\_backends\_asyncio.py", line 867, in run result = context.run(func, *args) File "C:\Users\Lucas Lima\miniconda3\envs\draggan\lib\site-packages\gradio\utils.py", line 488, in async_iteration return next(iterator) File "C:\Users\Lucas Lima\miniconda3\envs\draggan\lib\site-packages\draggan\web.py", line 135, in on_drag max_iters=max_iters): File "C:\Users\Lucas Lima\miniconda3\envs\draggan\lib\site-packages\draggan\api.py", line 191, in drag_gan for qi in neighbor(int(pi[0]), int(pi[1]), r1): File "C:\Users\Lucas Lima\miniconda3\envs\draggan\lib\site-packages\draggan\api.py", line 171, in neighbor points.append(torch.tensor([i, j]).float().cuda()) File "C:\Users\Lucas Lima\miniconda3\envs\draggan\lib\site-packages\torch\cuda\__init__.py", line 221, in _lazy_init raise AssertionError("Torch not compiled with CUDA enabled") AssertionError: Torch not compiled with CUDA enabled

I have AMD Radeon RX 590 Does anyone knows what about it?

[Zeqiang-Lai]

PyTorch only supports AMD GPU on Linux. Check this out if you are using linux. https://pytorch.org/get-started/locally/

Otherwise, you could use CPU mode

python -m draggan.web --device cpu

[f-izzat]

@Zeqiang-Lai I still get the same error as the original post even when running wih CPU mode

[astatonn]

@Zeqiang-Lai I still get error even running with cpu mode.

[Cmiiii]

Hello, did you fix this?

[ElvisWangTech]

I pulled the latest code and found that line 207 of draggan.py is also hardcoded。 device = torch.device("cuda")

[ShifterXD]

did he fix the bag wish cuda?