1.创建环境 conda create -n mmdetection python=3.7
2.安装pytorch=1.6(稳定)pip install torch1.6.0+cu101 torchvision0.7.0+cu101 -f https://download.pytorch.org/whl/torch_stable.html
3.pip install mmcv-full -f https://download.openmmlab.com/mmcv/dist/cu101/torch1.6.0/index.html
(这方法以前管用,现在不太行)
现在使用:安装mmcv:进入https://openmmlab.oss-accelerate.aliyuncs.com/mmcv/dist/index.html,根据python,cuda,torch等版本选择合适的mmcv-full版本,
这边下载的是mmcv_full-1.2.7+torch1.6.0+cu101-cp37-cp37m-manylinux1_x86_64.whl版本
pip install mmcv_full-1.2.7+torch1.6.0+cu101-cp37-cp37m-manylinux1_x86_64.whl
4.安装mmdetection
git clone https://github.com/open-mmlab/mmdetection.git
cd mmdetection
pip install -r requirements/build.txt(两个都要安装好像)
pip install -v -e . (注意有-e后面有一小点,别漏)
从零开始安装mmdetection(供自己学习用)
最新推荐文章于 2024-04-11 16:44:59 发布