Problem running docker container: nvidia:tlt-streamanalytics (DeepStream 5)

Accelerated Computing Intelligent Video Analytics TAO Toolkit
1

Hi;

I’m having issues running the docker container for the Transfer Learning Toolkit on my Jetson nano (A02), according to the instruction found here.

I can pull the container no problem. However, in order to work I did have to modify the command to include “sudo”:

$ sudo docker pull nvcr.io/nvidia/tlt-streamanalytics:v2.0_py3

Otherwise I get “permission denied” errors. I’ve remoted into the nano using ssh, btw.

Next I try to run the following command, just like the instructions says:

$ docker run --gpus all -it -v "/path/to/dir/on/host":"/path/to/dir/in/docker" \ -p 8888:8888 nvcr.io/nvidia/tlt-streamanalytics:v2.0_py3 /bin/bash

And I get this error:

standard_init_linux.go:211: exec user process caused "exec format error"

I re-tried the command using “sudo” and I get the same error.

Can I please get some help trouble-shooting?

My nano is running JP4.4, updated.

Thanks!

1 Like
3

Please note that the tlt docker should run on host PC instead of Jetson devices.
See tlt user guide for more details.

https://docs.nvidia.com/metropolis/TLT/tlt-getting-started-guide/index.html#requirements

2 Likes
5

Hello,

Can I run it on jetson?
Is there any possibility to run it on jetson?
Is there any possibility to run it on jetson?
If nano doesn’t work, can it be possible in xavier nx, tx2?

Thank you.

6

The training is expected to run with your host PC. See Integrating TAO Models into DeepStream — TAO Toolkit 3.22.05 documentation
After training, the etlt model or trt engine can be deployed in the xavier, nx, tx2, nano, etc.