解决问题：AssertionError: Torch not compiled with CUDA enabled

错误警报

Traceback (most recent call last):
  File "C:/Users/Administrator/Desktop/PubLayNet-maskrcnn/PubLayNet/maskrcnn/infer.py", line 117, in <module>
    main()
  File "C:/Users/Administrator/Desktop/PubLayNet-maskrcnn/PubLayNet/maskrcnn/infer.py", line 91, in main
    prediction = model([image.cuda()])
  File "E:\Anaconda3\lib\site-packages\torch\cuda\__init__.py", line 178, in _lazy_init
    _check_driver()
  File "E:\Anaconda3\lib\site-packages\torch\cuda\__init__.py", line 92, in _check_driver
    raise AssertionError("Torch not compiled with CUDA enabled")
AssertionError: Torch not compiled with CUDA enabled

错误原因

此错误是由于下载的torch没有cuda,在运行时就会出错，经过查阅，在程序最开始的地方加上：
device = torch.device(“cuda” if torch.cuda.is_available() else “cpu”)
代码其余地方出现.cuda()的地方改成.to(device)就可以在无gpu的环境中运行了。

举例： 

batch_data=Variable(torch.from_numpy(batch_data.copy()).type(torch.FloatTensor)).cuda()

改成

batch_data=Variable(torch.from_numpy(batch_data.copy()).type(torch.FloatTensor)).to(device)

原文链接：https://blog.csdn.net/m0_37690102/article/details/108364458