进入虚拟环境使用pip命令安装opencv报错

一、问题描述

pip install opencv-python==3.4.1.15，报错   raise ValueError("check_hostname requires server_hostname")
ValueError: check_hostname requires server_hostname

pip install -i https://pypi.tuna.tsinghua.edu.cn/simple opencv-python==3.4.1.15，依然报错

分析：问题出现在了没有关代理，关闭代理试试看就可以了

conda install -i https://pypi.tuna.tsinghua.edu.cn/simple opencv-python==3.4.1.15报错

conda install opencv-python==3.4.1.15，依然报错

conda install opencv-python，依然报错

pip install -i https://pypi.tuna.tsinghua.edu.cn/simple opencv-python出现报错如下：

Building wheels for collected packages: opencv-python
  Building wheel for opencv-python (PEP 517) ... error

 ERROR: Failed building wheel 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

pip install --upgrade pip setuptools wheel，出现报错ERROR: Could not install packages due to an OSError: [WinError 5] 拒绝访问。: 'd:\\users\\anaconda3\\envs\\pytorch\\scripts\\pip.exe'
Consider using the `--user` option or check the permissions.

分析：python -m ensurepip修复pip，再使用python -m pip install --upgrade pip升级pip

python -m ensurepip

python -m pip install --upgrade pip

pip install -i https://pypi.tuna.tsinghua.edu.cn/simple opencv-python依然报错

  ERROR: Failed building wheel for opencv-python
Failed to build opencv-python
ERROR: Could not build wheels for opencv-python, which is required to install pyproject.toml-based projects

pip install opencv-python，还是依然报一样的错

分析：pip install opencv-python安装的是最新版本，python3.6版本不支持，所以找一个python3.6支持的版本即可，比如opencv-python版本4.3.0.38

pip install -i https://pypi.douban.com/simple/ pip install opencv-python==4.3.0.38

这次opencv成功安装了

WARNING: Ignoring invalid distribution -ip (d:\users\anaconda3\envs\pytorch\lib\site-packages)
WARNING: Ignoring invalid distribution -ip (d:\users\anaconda3\envs\pytorch\lib\site-packages)

这个可以找到指定文件夹里面删去以~开头文件夹即可

二、分析与总结

未关闭代理报错

关闭代理之后发现pip缺失报错还需要修复pip，修复之后重新升级pip

未指定版本导致安装的是最新版本，与python3.6不匹配也报错

最终指定opencv-python版本情况下安装成功

接着由于包重名还需要删去指定路径里面~开头文件夹