jetrag

jetrag

CONTAINERS IMAGES RUN BUILD

• jetrag (temporary name for Ollama-LlamaIndex-based, Streamlit-enabled container)

Starting jetrag container

jetson-containers run $(autotag jetrag)

This will start the ollama server and enter into a bash terminal.

Starting "Jetson Copilot" app inside the container

First, create a directory on the host side to store Jetson related documents. The data directory is mounted on the container.

cd jetson-containers
mkdir -p ./data/documents/jetson

Once in the container:

streamlit run /opt/jetrag/app.py

Or you can start the container with additional arguments:

jetson-containers run $(autotag jetrag) bash -c '/start_ollama && streamlit run app.py'

This will start the ollama server and streamlit app for "Jetson Copilot", an AI assistant to answer any questions based on documents provided in /data/documents/jetson directory.

It should show something like this:

  You can now view your Streamlit app in your browser.

  Network URL: http://10.110.50.241:8501
  External URL: http://216.228.112.22:8501

Accessing "Jetson Copilot" app

From your browser, open the above Network URL (http://10.110.50.241:8501).

CONTAINERS

jetrag
Requires	L4T ['>=34.1.0']
Dependencies	build-essential cuda:12.2 cudnn python numpy cmake onnx pytorch ollama
Dockerfile	Dockerfile
Images	dustynv/jetrag:r35.4.1 (2024-05-23, 5.9GB) dustynv/jetrag:r36.3.0 (2024-05-23, 5.9GB)

CONTAINER IMAGES

Repository/Tag	Date	Arch	Size
dustynv/jetrag:r35.4.1	2024-05-23	arm64	5.9GB
dustynv/jetrag:r36.3.0	2024-05-23	arm64	5.9GB

_{Container images are compatible with other minor versions of JetPack/L4T:
    • L4T R32.7 containers can run on other versions of L4T R32.7 (JetPack 4.6+)
    • L4T R35.x containers can run on other versions of L4T R35.x (JetPack 5.1+)}

RUN CONTAINER

To start the container, you can use jetson-containers run and autotag, or manually put together a docker run command:

# automatically pull or build a compatible container image
jetson-containers run $(autotag jetrag)

# or explicitly specify one of the container images above
jetson-containers run dustynv/jetrag:r35.4.1

# or if using 'docker run' (specify image and mounts/ect)
sudo docker run --runtime nvidia -it --rm --network=host dustynv/jetrag:r35.4.1

^{jetson-containers run forwards arguments to docker run with some defaults added (like --runtime nvidia, mounts a /data cache, and detects devices)
autotag finds a container image that's compatible with your version of JetPack/L4T - either locally, pulled from a registry, or by building it.}

To mount your own directories into the container, use the -v or --volume flags:

jetson-containers run -v /path/on/host:/path/in/container $(autotag jetrag)

To launch the container running a command, as opposed to an interactive shell:

jetson-containers run $(autotag jetrag) my_app --abc xyz

You can pass any options to it that you would to docker run, and it'll print out the full command that it constructs before executing it.

BUILD CONTAINER

If you use autotag as shown above, it'll ask to build the container for you if needed. To manually build it, first do the system setup, then run:

jetson-containers build jetrag

The dependencies from above will be built into the container, and it'll be tested during. Run it with --help for build options.