l4t-ml

l4t-ml

CONTAINERS IMAGES RUN BUILD

CONTAINERS

l4t-ml
Builds
Requires	L4T ['>=32.6']
Dependencies	build-essential cuda cudnn python numpy cmake onnx pytorch:2.2 torchvision torchaudio tensorrt protobuf:cpp tensorflow2 opencv pycuda cupy onnxruntime numba gstreamer rust jupyterlab
Dockerfile	Dockerfile
Images	dustynv/l4t-ml:r32.7.1 (2023-11-13, 2.4GB) dustynv/l4t-ml:r35.2.1 (2024-01-04, 7.1GB) dustynv/l4t-ml:r35.3.1 (2023-12-11, 7.0GB) dustynv/l4t-ml:r35.4.1 (2024-03-07, 7.1GB) dustynv/l4t-ml:r36.2.0 (2024-03-07, 8.9GB)

CONTAINER IMAGES

Repository/Tag	Date	Arch	Size
dustynv/l4t-ml:r32.7.1	2023-11-13	arm64	2.4GB
dustynv/l4t-ml:r35.2.1	2024-01-04	arm64	7.1GB
dustynv/l4t-ml:r35.3.1	2023-12-11	arm64	7.0GB
dustynv/l4t-ml:r35.4.1	2024-03-07	arm64	7.1GB
dustynv/l4t-ml:r36.2.0	2024-03-07	arm64	8.9GB

_{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 l4t-ml)

# or explicitly specify one of the container images above
jetson-containers run dustynv/l4t-ml:r36.2.0

# or if using 'docker run' (specify image and mounts/ect)
sudo docker run --runtime nvidia -it --rm --network=host dustynv/l4t-ml:r36.2.0

^{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 l4t-ml)

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

jetson-containers run $(autotag l4t-ml) 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 l4t-ml

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