ModuleNotFoundError: No module named 'mmcv._ext'报错，不知道是否是版本不匹配的原因。换了多次pytorch和cudatoolkit版本才最终装好

1、nvcc检查cuda版本

nvidia-smi

+-----------------------------------------------------------------------------+
| NVIDIA-SMI 456.71       Driver Version: 456.71       CUDA Version: 11.1     |
|-------------------------------+----------------------+----------------------+

2\  conda新建环境，使用的是python3.7 然后安装pytorch

pip install torch==1.8.0+cu111 torchvision==0.9.0+cu111 torchaudio==0.8.0 -f https://download.pytorch.org/whl/torch_stable.html

3、mmcv安装，

git clone https://github.com/open-mmlab/mmcv.git
cd mmcv
pip install -r requirements.txt
set MMCV_WITH_OPS=1
pip install -e .

​

pip list 检查是否安装有mmcv

4、mmdetection安装

git clone https://github.com/open-mmlab/mmdetection.git
cd mmdetection
pip install -r requirements.txt
python setup.py build_ext --inplace

python setup.py build_ext --inplace 报错如下：

\lib\site-packages\torch\utils\cpp_extension.py:369: UserWarning: Attempted to use ninja as the BuildExtension backend but we could not find ninja.. Falling back to using the slow distutils backend.
  warnings.warn(msg.format('we could not find ninja.'))

修改对应env下Lib\site-packages\torch\utils\cpp_extension.py 

self.use_ninja = kwargs.get('use_ninja', True) 改为 

self.use_ninja = kwargs.get('use_ninja', False)

再次  

python setup.py build_ext --inplace

， 正确的是显示 running build_ext，然后

pip install -e .