在nvidia ngc平台下载docker image，根据实验要求在环境里安装opencv。因为当前实验tensorflow要求版本较低，我下载了比较久远的版本镜像。如果大家是自己复现代码，还是使用尽可能新的环境，则可以避免大部分环境问题。
通过pip安装opencv，指令如下：
pip install opencv-python
安装过程的输出：
Collecting opencv-python
Downloading https://files.pythonhosted.org/packages/77/f5/49f034f8d109efcf9b7e98fbc051878b83b2f02a1c73f92bbd37f317288e/opencv-python-4.4.0.42.tar.gz (88.9MB)
100% |████████████████████████████████| 88.9MB 919kB/s
Installing build dependencies ... done
Getting requirements to build wheel ... done
Preparing wheel metadata ... done
Requirement already satisfied: numpy>=1.13.1 in /usr/local/lib/python3.5/dist-packages (from opencv-python) (1.14.5)
Building wheels for collected packages: opencv-python
Building wheel for opencv-python (PEP 517) ...
PEP 517信息出现后，过了很长时间，出现报错信息：
Configuring Project
Working directory:
/tmp/pip-install-zujf20md/opencv-python/_skbuild/linux-x86_64-3.6/cmake-build
Command:
cmake /tmp/pip-install-zujf20md/opencv-python/opencv -G 'Unix Makefiles' -DCMAKE_INSTALL_PREFIX:PATH=/tmp/pip-install-zujf20md/opencv-python/_skbuild/linux-x86_64-3.6/cmake-install -DPYTHON_EXECUTABLE:FILEPATH=/usr/bin/python -DPYTHON_VERSION_STRING:STRING=3.6.8 -DPYTHON_INCLUDE_DIR:PATH=/usr/include/python3.6m -DPYTHON_LIBRARY:FILEPATH=/usr/lib/x86_64-linux-gnu/libpython3.6m.so -DSKBUILD:INTERNAL=TRUE -DCMAKE_MODULE_PATH:PATH=/tmp/pip-build-env-tr73ewqw/overlay/lib/python3.6/site-packages/skbuild/resources/cmake -DPYTHON3_EXECUTABLE=/usr/bin/python -DPYTHON3_INCLUDE_DIR=/usr/include/python3.6m -DPYTHON3_LIBRARY=/usr/lib/x86_64-linux-gnu/libpython3.6m.so -DBUILD_opencv_python3=ON -DBUILD_opencv_python2=OFF -DBUILD_opencv_java=OFF -DOPENCV_PYTHON3_INSTALL_PATH=python -DINSTALL_CREATE_DISTRIB=ON -DBUILD_opencv_apps=OFF -DBUILD_opencv_freetype=OFF -DBUILD_SHARED_LIBS=OFF -DBUILD_TESTS=OFF -DBUILD_PERF_TESTS=OFF -DBUILD_DOCS=OFF -DCMAKE_BUILD_TYPE:STRING=Release
Copying files from CMake output
Traceback (most recent call last):
File "/usr/local/lib/python3.6/dist-packages/pip/_vendor/pep517/_in_process.py", line 207, in <module>
main()
File "/usr/local/lib/python3.6/dist-packages/pip/_vendor/pep517/_in_process.py", line 197, in main
json_out['return_val'] = hook(**hook_input['kwargs'])
File "/usr/local/lib/python3.6/dist-packages/pip/_vendor/pep517/_in_process.py", line 141, in build_wheel
metadata_directory)
File "/usr/local/lib/python3.6/dist-packages/setuptools/build_meta.py", line 209, in build_wheel
wheel_directory, config_settings)
File "/usr/local/lib/python3.6/dist-packages/setuptools/build_meta.py", line 194, in _build_with_temp_dir
self.run_setup()
File "/usr/local/lib/python3.6/dist-packages/setuptools/build_meta.py", line 237, in run_setup
self).run_setup(setup_script=setup_script)
File "/usr/local/lib/python3.6/dist-packages/setuptools/build_meta.py", line 142, in run_setup
exec(compile(code, __file__, 'exec'), locals())
File "setup.py", line 496, in <module>
main()
File "setup.py", line 289, in main
cmake_source_dir=cmake_source_dir,
File "/tmp/pip-build-env-tr73ewqw/overlay/lib/python3.6/site-packages/skbuild/setuptools_wrap.py", line 629, in setup
cmake_source_dir, skbuild_kw['cmake_install_dir'])
File "setup.py", line 381, in _classify_installed_files_override
with open('scripts/__init__.py', 'r') as custom_init:
FileNotFoundError: [Errno 2] No such file or directory: 'scripts/__init__.py'
----------------------------------------
ERROR: Failed building wheel for opencv-python
Running setup.py clean for opencv-python
Failed to build opencv-python
ERROR: Could not build wheels for opencv-python which use PEP 517 and cannot be installed directly
升级pip工具包：
pip install --upgrade pip setuptools wheel
重试安装：
pip install opencv-python
问题解决啦~！
Package | 解决 Could not build wheels for opencv-python which use PEP 517 and cannot be installed
背景在nvidia ngc平台下载docker image，根据实验要求在环境里安装opencv。因为当前实验tensorflow要求版本较低，我下载了比较久远的版本镜像。如果大家是自己复现代码，还是使用尽可能新的环境，则可以避免大部分环境问题。问题通过pip安装opencv，指令如下：pip install opencv-python安装过程的输出：Collecting opencv-python Downloading https://files.pythonhosted.org/pa
安装opencv时报错：Could not build wheels for opencv-python which use PEP 517 and cannot be installed direc
一、问题背景
最近写一批程序，需要用到opencv里面的函数，但是导入cv2的时候，新电脑提示没有安装，所以首先安装opencv的库，直接pip install opencv-python，不过报错了。
然后从镜像站中安装opencv，直接pip install -i https://mirrors.aliyun.com/pypi/simple opencv-python，不过仍然报错，报错原因和上一次一样：Could not build wheels for openc...
pip install --upgrade -r requirements.txt -i https://mirror.baidu.com/pypi/simple
Looking in indexes: https://mirror.baidu.com/pypi/simple
Collecting prettytable
Downloading https://mirror.baidu.com/pypi/packages/5f/ab/64371af206988d7b15c8112c9c277b8eb4618
Could not build wheels for opencv-python which use PEP 517 and cannot be installed directly
Could not build wheels for opencv-python which use PEP 517 and cannot be installed directly
直接运用conda安装：
随后完成cv2的安装。
pip install opencv-python时遇到如下问题：ERROR-Could not build wheels for opencv-python which use PEP 517 and cannot be installed directly
解决方案：
pip install opencv-python --no-use-pep517 ×
pip install opencv-python --no-cache-dir √
版本信息：
Ubuntu：20.04
python：3.
Package | 解决 Could not build wheels for opencv-python which use PEP 517 and cannot be installed
35359
Package | 解决 ImportError: libGL.so.1: cannot open shared object file: No such file or directory
26470
Package | 解决 ImportError: libGL.so.1: cannot open shared object file: No such file or directory
山居秋暝LS:
Package | 解决 module ‘cv2‘ has no attribute ‘gapi_wip_gst_GStreamerPipeline‘
一月σண:
Matlab | mex混编多个cpp文件
weixin_42674476:
Package | 解决 Windows10 Anaconda环境 CUDA11.5 pytorch运行报错 No CUDA runtime is found
1LOVESJohnny:
