TX1环境搭建

1,刷裸机

sudo ./flash.sh -jetson-tx1 mmcblk0p1

2,安装cuda

从之前主机的目录下拷贝cuda-repo-l4t-8-0-local_8.0.84-1_arm64.deb

执行:

sudo dpkg -i cuda-repo-l4t-8-0-local_8.0.84-1_arm64.deb

sudo apt-get update

sudo apt-get install cuda-toolkit-8-0

echo "PATH=/usr/local/cuda-8.0/bin:$PATH" >> ~/.bashrc

source ~/.bashrc

OK 安装完毕

3,安装cudnn

同上拷贝libcudnn*的三个deb包

sudo dpkg -i libcudnn6_6.0.21-1+cuda8.0_arm64.deb

另外两个同上

4,安装opencv(注:此处的opencv已经添加了Gstreamer选项编译,且版本必须是3.0及以上)

安装opencv的依赖库
找一个装好opencv的tx1板卡直接把
/usr/local/lib 
/usr/local/include 

下面对应的opencv文件拷过来

5,安装模型推理工具Tensorrt

前面同上,拷贝过来
sudo dpkg -i nv-gie-repo-ubuntu1604-ga-cuda8.0-trt2.1-20170614_1-1_arm64.deb
sudo apt-get update 
sudo apt-get install tensorrt-2.1.2
出现OK字样说明安装成功

6,编译视频编解码demo

同上,复制Tegra_Multimedia_API_R28.1.0_aarch64.tbz2 tar -jxvf Tegra_Multimedia_API_R28.1.0_aarch64.tbz2 编译即可

转载于:https://www.cnblogs.com/mahxn0/p/8431495.html

你可能感兴趣的:(TX1环境搭建)