尽管拥有所有必要的库和导入，但出现“NotImplementedError：无法使用 CUDA 后端的参数运行 'torchvision::nms'”

Question

完整错误：

NotImplementedError: Could not run 'torchvision::nms' with arguments from the 'CUDA' backend. This could be because the operator doesn't exist for this backend, or was omitted during the selective/custom build process (if using custom build). If you are a Facebook employee using PyTorch on mobile, please visit https://fburl.com/ptmfixes for possible resolutions. 'torchvision::nms' is only available for these backends: [CPU, QuantizedCPU, BackendSelect, Python, FuncTorchDynamicLayerBackMode, Functionalize, Named, Conjugate, Negative, ZeroTensor, ADInplaceOrView, AutogradOther, AutogradCPU, AutogradCUDA, AutogradXLA, AutogradMPS, AutogradXPU, AutogradHPU, AutogradLazy, Tracer, AutocastCPU, AutocastCUDA, FuncTorchBatched, FuncTorchVmapMode, Batched, VmapMode, FuncTorchGradWrapper, PythonTLSSnapshot, FuncTorchDynamicLayerFrontMode, PythonDispatcher].

当我尝试在 Windows 11 机器上训练 YOLOv8 模型时，我得到了这个信息，在第一个 epoch 中一切正常，然后就会发生这种情况。

---

我也在第一个纪元结束后立即收到此错误，但我认为它不相关。

Error executing job with overrides: ['task=detect', 'mode=train', 'model=yolov8n.pt', 'data=custom.yaml', 'epochs=300', 'imgsz=160', 'workers=8', 'batch=4']

我试图利用 CUDA GPU 训练 YOLOv8 图像检测模型。

Answer 1

当您安装了 Torch 和 Torchaudio for CUDA 但未安装 Torchvision for CUDA 时，会出现此错误。

卸载Torch和Torchvision，我使用pip：

pip uninstall torch torchvision

然后到这里使用漂亮的界面安装两者的正确版本。我得到以下命令：

pip install torch torchvision torchaudio --extra-index-url https://download.pytorch.org/whl/cu116