重新安装 Windows 10、Microsoft Visual C++ 和 CUDA 11.6 后,我安装了 Python 3.10 库。要安装 Cupy,我使用以下命令pip install cupy-cuda116
:pip install cupy
和pip install cupy-10.5.0+cuda116-cp310-cp310-win_amd64.whl
. 一个错误:
Collecting cupy-cuda116
Using cached cupy_cuda116-10.6.0-cp310-cp310-win_amd64.whl (62.4 MB)
Requirement already satisfied: numpy<1.25,>=1.18 in c:\python310\lib\site-packages (from cupy-cuda116) (1.22.4+mkl)
Collecting fastrlock>=0.5
Using cached fastrlock-0.8.tar.gz (68 kB)
Preparing metadata (setup.py) ... done
Using legacy 'setup.py install' for fastrlock, since package 'wheel' is not installed.
Installing collected packages: fastrlock, cupy-cuda116
Running setup.py install for fastrlock ... error
error: subprocess-exited-with-error
× Running setup.py install for fastrlock did not run successfully.
│ exit code: 1
╰─> [16 lines of output]
building without Cython
running install
C:\Python310\lib\site-packages\setuptools\command\install.py:34: SetuptoolsDeprecationWarning: setup.py install is deprecated. Use build and pip and other standards-based tools.
warnings.warn(
running build
running build_py
creating build
creating build\lib.win-amd64-cpython-310
creating build\lib.win-amd64-cpython-310\fastrlock
copying fastrlock\__init__.py -> build\lib.win-amd64-cpython-310\fastrlock
copying fastrlock\rlock.pxd -> build\lib.win-amd64-cpython-310\fastrlock
copying fastrlock\__init__.pxd -> build\lib.win-amd64-cpython-310\fastrlock
copying fastrlock\_lock.pxi -> build\lib.win-amd64-cpython-310\fastrlock
running build_ext
building 'fastrlock.rlock' extension
error: Microsoft Visual C++ 14.0 or greater is required. Get it with "Microsoft C++ Build Tools": https://visualstudio.microsoft.com/visual-cpp-build-tools/
[end of output]
note: This error originates from a subprocess, and is likely not a problem with pip.
error: legacy-install-failure
× Encountered error while trying to install package.
╰─> fastrlock
note: This is an issue with the package mentioned above, not pip.
hint: See above for output from the failure.
安装“Visual Studio”或在“Visual Studio Installer”中,选中“桌面开发”,它将安装大约 10 GB。安装 CUDA,可以从 Nvidia 网站下载。之后使用命令在命令行上安装“Cupy”
pip install cupy-cuda116
(其中“116”是 CUDA 版本 11.6)。在 Python 中,检查是否安装了 Cupyimport cupy