System Setup Jetson TX2

Robotics & Edge Computing Jetson & Embedded Systems Jetson TX2
1

Hello everybody,
I want to setup TX2 on Ubuntu with this document: GitHub - dusty-nv/jetson-inference: Hello AI World guide to deploying deep-learning inference networks and deep vision primitives with TensorRT and NVIDIA Jetson.
But My program failed when I install nvidia-docker2. I’m tried by something on google but not success.
So Can you have me a exactly document for setup Jetson TX2?
Thank you very much!

2

Hi letrung096, the Docker portion of the tutorial is for installing in the cloud or a host PC, for setting up the training machine. Your Jetson TX2 gets JetPack installed to it if you haven’t already done so.

The neural network training is done on a server or PC that has a larger GPU and larger disk, and then the network model is deployed to the Jetson which runs inferencing with TensorRT.

3

Hello Nvidia’s supporter. I’m thank you for your response.
I installed Jetpack but I still not can install nvidia-docker2. So I can’t install Digit server. Wish response soon.
Thank you!

4

Hi,

Please use DIGITs on the cloud or host PC(x86 based environment).
It cannot be used on the Jetson platform.

Thanks.

5

Hi,
Maybe I found problem.
I tried to download nvidia-docker from https://nvidia.github.io/nvidia-docker/$distribution/nvidia-docker.list follow code:
curl -s -L https://nvidia.github.io/nvidia-docker/$distribution/nvidia-docker.list |
sudo tee /etc/apt/sources.list.d/nvidia-docker.list
sudo apt-get update
[GitHub - NVIDIA/nvidia-docker: Build and run Docker containers leveraging NVIDIA GPUs]
but in reality, this link can not access :((
why I can’t access nvidia-docker2 for jetson tx2?

6

I understood. Because I use Jetson TX2, this use arm so not support by DIGITs.
Can you give me difference way to setup DIGITs for Jetson TX2?
thank you very much!

7

Hi,

DIGITs cannot be used on the TX2 due to some required tool is not supported(ex.nvidia-smi) .

In general, it’s NOT recommended to use Jetson for training. You may face issues on the limited storage and bandwidth.
If this is essential for you, you can try to setup Caffe/TensorFlow on Jetson directly.

Thanks.