Skip to content

Latest commit

 

History

History
 
 

l4t-tensorflow

Folders and files

NameName
Last commit message
Last commit date

parent directory

..
README.md
README.md
 
 
l4t-tensorflow.json
l4t-tensorflow.json
 
 

README.md

l4t-tensorflow

CONTAINERS IMAGES RUN BUILD

CONTAINERS
l4t-tensorflow:tf1
   Builds l4t-tensorflow-tf1_jp46 l4t-tensorflow-tf1_jp51
   Requires L4T >=32.6
   Dependencies build-essential cuda cudnn python tensorrt numpy protobuf:cpp tensorflow opencv pycuda
   Images dustynv/l4t-tensorflow:tf1-r32.7.1 (2023-12-06, 0.9GB)
dustynv/l4t-tensorflow:tf1-r35.2.1 (2023-12-06, 5.5GB)
dustynv/l4t-tensorflow:tf1-r35.3.1 (2023-08-29, 5.6GB)
dustynv/l4t-tensorflow:tf1-r35.4.1 (2023-10-07, 5.5GB)
l4t-tensorflow:tf2
   Builds l4t-tensorflow-tf2_jp46 l4t-tensorflow-tf2_jp60 l4t-tensorflow-tf2_jp51
   Requires L4T >=32.6
   Dependencies build-essential cuda cudnn python tensorrt numpy protobuf:cpp tensorflow2 opencv pycuda
   Images dustynv/l4t-tensorflow:tf2-r32.7.1 (2023-12-06, 1.0GB)
dustynv/l4t-tensorflow:tf2-r35.2.1 (2023-09-07, 5.7GB)
dustynv/l4t-tensorflow:tf2-r35.3.1 (2023-12-06, 5.7GB)
dustynv/l4t-tensorflow:tf2-r35.4.1 (2023-10-07, 5.7GB)
dustynv/l4t-tensorflow:tf2-r36.2.0 (2023-12-06, 7.3GB)
CONTAINER IMAGES
Repository/Tag Date Arch Size
  dustynv/l4t-tensorflow:tf1-r32.7.1 2023-12-06 arm64 0.9GB
  dustynv/l4t-tensorflow:tf1-r35.2.1 2023-12-06 arm64 5.5GB
  dustynv/l4t-tensorflow:tf1-r35.3.1 2023-08-29 arm64 5.6GB
  dustynv/l4t-tensorflow:tf1-r35.4.1 2023-10-07 arm64 5.5GB
  dustynv/l4t-tensorflow:tf2-r32.7.1 2023-12-06 arm64 1.0GB
  dustynv/l4t-tensorflow:tf2-r35.2.1 2023-09-07 arm64 5.7GB
  dustynv/l4t-tensorflow:tf2-r35.3.1 2023-12-06 arm64 5.7GB
  dustynv/l4t-tensorflow:tf2-r35.4.1 2023-10-07 arm64 5.7GB
  dustynv/l4t-tensorflow:tf2-r36.2.0 2023-12-06 arm64 7.3GB

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 the run.sh/autotag helpers or manually put together a docker run command:

# automatically pull or build a compatible container image
./run.sh $(./autotag l4t-tensorflow)

# or explicitly specify one of the container images above
./run.sh dustynv/l4t-tensorflow:tf2-r32.7.1

# or if using 'docker run' (specify image and mounts/ect)
sudo docker run --runtime nvidia -it --rm --network=host dustynv/l4t-tensorflow:tf2-r32.7.1

run.sh 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:

./run.sh -v /path/on/host:/path/in/container $(./autotag l4t-tensorflow)

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

./run.sh $(./autotag l4t-tensorflow) my_app --abc xyz

You can pass any options to run.sh 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:

./build.sh l4t-tensorflow

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