- Jetpack 4.3
- SD card version is r32.3.1
- Other lib versions (e.g. cuda, cuDNN, tensorrt...) showed in here
- Japanese languages
- ROS and so on
-
Flash Jetbot images to SD card
-
Using
jetbot_image_v0p4p0.zip
for this example -
Update as you do eveytime for Ubuntu
$ sudo apt update $ sudo apt upgrade
-
Change power mode
- Default of pre-build images are 5W mode(2 Core), so follwings change the fast mode
$ sudo nvpmodel -m 0 # Change to MAXN(4 Core) mode, 1 is 5W mode $ sudo jetson_clocks # Maximise performance
- Check performance
$ sudo nvpmodel -q $ sudo jetson_clocks --show
- Reference here
-
Tensorrt test
- Though the jetbot demo., check tensorrt can work well
- Demo is here
- Dwonload model
ssd_mobilenet_v2_coco.engine
(v0.4) from here - Call Python interpreter
- Check whether
ObjectDetector
go
>> from jetbot import ObjectDetector >> model = ObjectDetector('ssd_mobilenet_v2_coco.engine') >> from jetbot import Camera >> camera = Camera.instance(width=300, height=300) >> detections = model(camera.value) >> print(detections)
- Dwonload model
-
Japanize keyboard
- Follwing may not be optimal or clever.
- In short, both
Settings
andLanguage support
should be set correctly - If you rebooted and japanize-setting were ignored, Open
Settings -> Region and language
and remove all languages except日本語(mozc)
- jetop
$ sudo apt install python-pip
$ sudo -H pip install jetson-stats
- Set path(ver.CUDA10.0)
$ echo "source /opt/ros/melodic/setup.bash" >> ~/.bashrc
$ echo "export PATH=/usr/local/cuda-10.0/bin:${PATH}" >> ~/.bashrc
$ echo "export LD_LIBRARY_PATH=/usr/local/cuda-10.0/lib64:${LD_LIBRARY_PATH}" >> ~/.bashrc
$ source ~/.bashrc
- ROS
- Use repo. https://github.com/karaage0703/jetson-nano-tools
- Instruction is here
- Show Jetpack version
$ cat /etc/nv_tegra_release #
- http://www.neko.ne.jp/~freewing/raspberry_pi/nvidia_jetson_nano_2020_jetpack_44/
- https://jetbot.org/master/software_setup/sd_card.html
- https://github.com/NVIDIA-AI-IOT/jetbot/wiki/software-setup
- http://www.neko.ne.jp/~freewing/raspberry_pi/nvidia_jetson_nano_sample_application/
- https://github.com/NVIDIA-AI-IOT/jetbot/blob/master/notebooks/object_following/live_demo.ipynb
- https://github.com/NVIDIA-AI-IOT/jetbot/wiki/examples
- https://denor.jp/jetsonnano%E3%81%AE%E5%88%9D%E6%9C%9F%E8%A8%AD%E5%AE%9A%E3%81%A8%E6%97%A5%E6%9C%AC%E8%AA%9E%E5%85%A5%E5%8A%9B%E8%A8%AD%E5%AE%9A
- https://toyo-interest.com/jetson-nano/jetson-nano%E6%97%A5%E6%9C%AC%E8%AA%9E%E5%85%A5%E5%8A%9B%E3%81%A7%E3%81%8D%E3%82%8B%E3%82%88%E3%81%86%E3%81%AB%E3%81%99%E3%82%8B/
- https://qiita.com/MuAuan/items/9b2b63d12e094ad2e739