jetson-stats is a package for monitoring and control your NVIDIA Jetson [Orin, Xavier, Nano, TX] series.
jetson-stats is a powerful tool to analyze your board, you can use with a stand alone application with jtop
or import in your python script, the main features are:
- Decode hardware, architecture, L4T and NVIDIA Jetpack
- Monitoring, CPU, GPU, Memory, Engines, fan
- Control NVP model, fan speed, jetson_clocks
- Importable in a python script
- Dockerizable in a container
- Do not need super user
- Tested on many different hardware configurations
- Works with all NVIDIA Jetpack
jetson-stats can be installed with pip, but need superuser:
sudo pip3 install -U jetson-stats
Don't forget to logout/login or reboot your board
π That's it! π
Start jtop it's pretty simple just write jtop
!
jtop
A simple interface will appear on your terminal, more capabilities are documented at jtop page.
You can use jtop such a python library to integrate in your software
from jtop import jtop
with jtop() as jetson:
# jetson.ok() will provide the proper update frequency
while jetson.ok():
# Read tegra stats
print(jetson.stats)
You can also use jtop with your virualenv!
More information available at advanced usage page.
You can run directly in Docker jtop, you need only to:
- Install jetson-stats on your host
- Install jetson-stats on your container as well
- Pass to your container
/run/jtop.sock:/run/jtop.sock
You can try running this command
docker run --rm -it -v /run/jtop.sock:/run/jtop.sock rbonghi/jetson_stats:latest
More information available at docker documentation page.
If your company benefits from this library, please consider π sponsoring its development.
jetson-stats has usage and reference documentation at https://rnext.it/jetson_stats, there is also a π troubleshooting page.
jetson-stats has a community Discord channel for asking questions and collaborating with other contributors. Drop by and say hello π