jetson-inference教程(国内镜像):
(1)安装git和cmake, 并检出库到本地:
sudo apt-get install git cmake
git clone GitHub - dusty-nv/jetson-inference: Hello AI World guide to deploying deep-learning inference networks and deep vision primitives with TensorRT and NVIDIA Jetson.
(2)更新submodule
cd jetson-inference
git submodule update --init
(3)切换到国内镜像:
sed -in-place -e 's@https://nvidia.box.com/shared/static@https://bbs.gpuworld.cn/mirror@g' CMakePreBuild.sh
(修改文件目录:jetson-inference/CMakePreBuild.sh)
文件里面:https://nvidia.box.com/shared/static 替换成:https://bbs.gpuworld.cn/mirror
(4)编译jetson-inference
mkdir build
cd build
cmake ..
(耐心等待。现在你应当不会遭遇到无法下载模型的问题了)
因版本更新后 CMakePreBuild.sh 无相应连接;
进行连接批量替换即可
find -type f -exec sed -i -e "s/https:\/\/nvidia.box.com\/shared\/static/http:\/\/bbs.gpuworld.cn\/mirror/g" {} \;