mmcv安装记录

由于官方提供的win版本有限，需要自己源码编译mmcv，特此记录。

1 准备

提前下载好mmcv-master.zip或者gi clone下来：

GitHub - open-mmlab/mmcv: OpenMMLab Computer Vision FoundationOpenMMLab Computer Vision Foundation. Contribute to open-mmlab/mmcv development by creating an account on GitHub.https://github.com/open-mmlab/mmcv

还有提前下载好mmflow-master.zip

安装：

Visual Studio Community 2019

Anaconda3

cuda11.1

2 安装python环境

conda create --name mmcv python=3.7  # 3.6, 3.7, 3.8 should work too as tested
conda activate mmcv  # make sure to activate environment before any operation

 torch-1.8.0+cu111-cp37-cp37m-win_amd64.whl

https://download.pytorch.org/whl/cu111/torch-1.8.0%2Bcu111-cp37-cp37m-win_amd64.whlhttps://download.pytorch.org/whl/cu111/torch-1.8.0%2Bcu111-cp37-cp37m-win_amd64.whltorchvision-0.9.0+cu111-cp37-cp37m-win_amd64.whl

https://download.pytorch.org/whl/cu111/torchvision-0.9.0%2Bcu111-cp37-cp37m-win_amd64.whlhttps://download.pytorch.org/whl/cu111/torchvision-0.9.0%2Bcu111-cp37-cp37m-win_amd64.whl

pip install  torch-1.8.0+cu111-cp37-cp37m-win_amd64.whl

pip install torchvision-0.9.0+cu111-cp37-cp37m-win_amd64.whl

cd mmcv

pip install -r requirements.txt

3 编译mmcv

严格按照官网要求设置

（0）Set up MSVC compiler

测试是否配置成功：

  

（1）cuda

安装了cuda，环境变量自己会有

（2） CUDA target arch

查询自己的gpu算力，我的是3090，算力是8.6

（3）ops和cpu cores

 然后开始编译

conda activate mmcv
# change directory
cd mmcv
# build
python setup.py build_ext # if success, cl will be launched to compile ops

编译成功：

如何没有打印很多信息，则是失败了，重新确认一遍流程。

安装mmcv：

# install
python setup.py develop
# check
pip list

 安装成功：

 

发现，mmcv和mmcv-full都有。

---

注：

如果输入conda list，报错

CondaError: Expected exactly one `egg-info` directory in 'D:\code\PycharmProjects\mmcv'

 则删除conda环境中的文件mmcv-full.egg-link