我遇到了同样的问题。
尝试按照以下方式安装Torch:
# http://pytorch.org/
from os import path
from wheel.pep425tags import get_abbr_impl, get_impl_ver, get_abi_tag
platform = '{}{}-{}'.format(get_abbr_impl(), get_impl_ver(), get_abi_tag())
accelerator = 'cu80' #'cu80' if path.exists('/opt/bin/nvidia-smi') else 'cpu'
print('Platform:', platform, 'Accelerator:', accelerator)
!pip install --upgrade --force-reinstall -q http://download.pytorch.org/whl/{accelerator}/torch-0.4.0-{platform}-linux_x86_64.whl torchvision
import torch
print('Torch', torch.__version__, 'CUDA', torch.version.cuda)
print('Device:', torch.device('cuda:0'))
一些流传的代码片段使用平台: cp36-cp36m 加速器: cu80 Torch 0.4.0 CUDA 8.0.61
设备: cuda:0
torch-0.3.0.post4-{platform}-linux_x86_64.whl
,这会导致相同的错误,因为 device
是 Torch 4 的特性。如果您已经安装了错误的版本,则可能需要执行 !pip uninstall torch
。!pip3 install http://download.pytorch.org/whl/cu92/torch-0.4.1-cp36-cp36m-linux_x86_64.whl
!pip3 install torchvision
torch.cuda.get_device_name(0)
我在Google Colab中的结果是Tesla K80。
import torch
if torch.cuda.is_available():
device = torch.device("cuda")
else:
device = torch.device("cpu")