Question

即使我成功安装了torchvision和cuda，也遇到此错误：正在VMWare（Ubuntu 18.04）中运行该程序。

python3 ./decode/train_dataloader.py --img_data ./datasets/image_fmri --output ./tmp/feat_data
Traceback (most recent call last):
  File "./decode/train_dataloader.py", line 26, in <module>
    net = torchvision.models.vgg19_bn(pretrained=True).cuda()
  File "/home/home/.local/lib/python3.6/site-packages/torch/nn/modules/module.py", line 260, in cuda
    return self._apply(lambda t: t.cuda(device))
  File "/home/home/.local/lib/python3.6/site-packages/torch/nn/modules/module.py", line 187, in _apply
    module._apply(fn)
  File "/home/home/.local/lib/python3.6/site-packages/torch/nn/modules/module.py", line 187, in _apply
    module._apply(fn)
  File "/home/home/.local/lib/python3.6/site-packages/torch/nn/modules/module.py", line 193, in _apply
    param.data = fn(param.data)
  File "/home/home/.local/lib/python3.6/site-packages/torch/nn/modules/module.py", line 260, in <lambda>
    return self._apply(lambda t: t.cuda(device))
  File "/home/home/.local/lib/python3.6/site-packages/torch/cuda/__init__.py", line 161, in _lazy_init
    _check_driver()
  File "/home/home/.local/lib/python3.6/site-packages/torch/cuda/__init__.py", line 75, in _check_driver
    raise AssertionError("Torch not compiled with CUDA enabled")
AssertionError: Torch not compiled with CUDA enabled

Answer 1

问题已解决：出现错误是因为程序代码在虚拟机（VMWare； OS：Ubuntu 18.04； Native System OS：Windows 10）内部运行。丢失（因为无法在虚拟机中安装图形驱动程序）。

此页面是在Ubuntu中安装cuda的良好指南：https://www.pugetsystems.com/labs/hpc/How-to-install-CUDA-9-2-on-Ubuntu-18-04-1184/

在Ubuntu 18.04（在VMWare内部运行）中运行程序时出现此错误：AssertionError：未在启用CUDA的情况下编译Torch

1 个答案: