search

dusty-nv / jetson-inference

Hello AI World guide to deploying deep-learning inference networks and deep vision primitives with TensorRT and NVIDIA Jetson.
https://developer.nvidia.com/embedded/twodaystoademo
MIT License
7.78k stars 2.97k forks source link

Hello World on Jetson Orin Nano - docker/run.sh: line 198: [: =: unary operator expected #1718

Open moh-C opened 1 year ago

moh-C commented 1 year ago

Hey! I have been having really a bad time with Jetson Orin Nano since the compatibility issues are just endless. I started S1E1 from NVIDIA Developer channel on YouTube which did not work at all (since they are for JetPack 4.6 as opposed to 5.1 which is for Orin). NVIDIA forum suggested I start watching the S3E1 from that very same channel wherein you start giving tutorials which would be compatible with Orin. When I try to run the docker/run.sh, I get the following error:

docker/tag.sh: line 4: docker/containers/scripts/l4t_version.sh: No such file or directory docker/tag.sh: line 6: [: =: unary operator expected docker/tag.sh: line 36: [: =: unary operator expected localuser:root being added to access control list CONTAINER_IMAGE: dustynv/jetson-inference: DATA_VOLUME: --volume /home/mohammad/GitHub/jetson-inference/data:/jetson-inference/data --volume /home/mohammad/GitHub/jetson-inference/python/training/classification/data:/jetson-inference/python/training/classification/data --volume /home/mohammad/GitHub/jetson-inference/python/training/classification/models:/jetson-inference/python/training/classification/models --volume /home/mohammad/GitHub/jetson-inference/python/training/detection/ssd/data:/jetson-inference/python/training/detection/ssd/data --volume /home/mohammad/GitHub/jetson-inference/python/training/detection/ssd/models:/jetson-inference/python/training/detection/ssd/models --volume /home/mohammad/GitHub/jetson-inference/python/www/recognizer/data:/jetson-inference/python/www/recognizer/data V4L2_DEVICES: --device /dev/video0 --device /dev/video1 DISPLAY_DEVICE: -e DISPLAY=:1 -v /tmp/.X11-unix/:/tmp/.X11-unix docker/run.sh: line 198: [: =: unary operator expected docker/run.sh: line 213: [: =: unary operator expected

I have also cloned the latest version. My camera is also plugged in. But I really do not get why should I get "[:=: unary operator expected" error. I would really appreciate it if you told me what should I do for it to be compatible with jetpack 5.1. Or if this repo is just for Jetson Nano (so JetPack 5.x should be excluded)? How and where should I start for Jetson Orin?!

dusty-nv commented 1 year ago

Hi @moh-C, sorry for the confusion, and yes jetson-inference is fully compatible with JetPack 5 and Jetson Orin. When you cloned the repo, did you use --recursive? If not, do git submodule update --init

BTW I hope that https://github.com/dusty-nv/jetson-containers is also useful to you

yasmineLalabouali commented 9 months ago

I'm having the same errors even after running git submodule update --init, is there any way to solve it ?

topherbuckley commented 7 months ago

Are you running it from the root of the repo? e.g. cd docker && ./run does not work for some reason. I'd guess it has to do with relative paths and expecting to be run from the root directory.

andanders commented 3 months ago

@moh-C I had the same issue. As @topherbuckley pointed out, you must run run.sh from the parent dir. I.e. docker/run.sh.