Jetson Xavier安装ROS+CUDA+PyTorch

• ROS web: http://wiki.ros.org/cn/melodic/Installation/Ubuntu

sudo sh -c '. /etc/lsb-release && echo "deb http://mirrors.tuna.tsinghua.edu.cn/ros/ubuntu/ `lsb_release -cs` main" > /etc/apt/sources.list.d/ros-latest.list'

sudo apt-key adv --keyserver 'hkp://keyserver.ubuntu.com:80' --recv-key C1CF6E31E6BADE8868B172B4F42ED6FBAB17C654

sudo apt update

sudo apt install ros-melodic-desktop-full

sudo rosdep init # Errors occur: command not found
sudo apt install rospack-tools

sudo rosdep init # Errors occur: Website may be down
sudo chmod a+rw /etc/hosts
gedit /etc/hosts # Add: 151.101.84.133      raw.githubusercontent.com

sudo rosdep init
rosdep update # Errors occur: unable to process source, doesn't matter

echo "source /opt/ros/melodic/setup.bash" >> ~/.bashrc
source ~/.bashrc

# Check architecture
uname -m

# Check version of Linux
cat /etc/*release

# Check JetPack info
sudo pip3 install jetson-stats
sudo jtop

# In another pc, download installation packages using 'NVIDIA SDK Manager'

# Visit https://developer.nvidia.com and sign in
# Click 'JetPack'(in Popular SDKs) -> 'FOR ANY JETSON DEVELOPER KIT'(in Installing JetPack) -> 'Download NVIDIA SDK Manager' and get more information by following 'Install Jetson Software with SDK Manager'

# install SDK Manager
sudo dpkg -i XXX.deb
sdkmanager

# Download packages that Xavier needs
#   In STEP 01: Jetson -> Host Machine -> Target Hardware Jetson AGX Xavier modules -> Linux JetPack 4.4 -> CONTINUE
#   In STEP 02: I accept ... -> Download now. Install later. -> CONTINUE
#   In STEP 03: Download packages
#   In STEP 04: Finish

# Next, copy three packages from download folder into Xavier, these packages are:
# 'cuda-repo-l4t-10-2-local-10.2.89_1.0-1_arm64.deb'
# 'libcudnn8_8.0.0.180-1+cuda10.2_arm64.deb'
# 'libcudnn8-dev_8.0.0.180-1+cuda10.2_arm64.deb'

# In Xavier, run installation commands

sudo dpkg -i cuda-repo-l4t-10-2-local-10.2.89_1.0-1_arm64.deb
sudo dpkg -i libcudnn8_8.0.0.180-1+cuda10.2_arm64.deb
sudo dpkg -i libcudnn8-dev_8.0.0.180-1+cuda10.2_arm64.deb

sudo apt update
sudo apt install cuda-toolkit-10-2

• Check Jetson info: http://www.gpus.cn/gpus_list_page_techno_support_content?id=39
• Download PyTorch installation package: https://elinux.org/Jetson_Zoo#PyTorch_.28Caffe2.29
• PyTorch web: https://github.com/pytorch/pytorch
• TorchVision web: https://github.com/pytorch/vision

# Download PyTorch package according to Jetson info, for example, JetPack 4.4 -> PyTorch v1.6.0 -> TorchVision v0.7.0

sudo apt-get install libopenblas-base libopenmpi-dev
sudo apt-get install python3-pip
pip3 install Cython
pip3 install numpy torch-1.6.0-cp36-cp36m-linux_aarch64.whl

git clone --branch v0.7.0 https://github.com/pytorch/vision.git torchvision
cd torchvision
pip3 install setuptools
sudo python3 setup.py install

python3
import torch # no error appears
print(torch.__version__) # get 1.6.0
print(torch.cuda.is_available()) # get Ture
import torchvision # no error appears

sudo pip3 uninstall torch
sudo pip3 uninstall torchvision