How can we run a project created on jetson nano on another computer?

Greetings to all. Me and my teammate developed a project for the Alpr system on jetson xavier. We used cuda for the project. We will use this project at the entrance to the parking lot. We have installed everything on the Xavier computer. How can we run it on another computer?

Hi @ozlematiz1637 ,

You can deploy the project to different jetson modules. You have said that you would like run the project on another computer. What kind of computer do you want to use?

Regards.

for any computer. Regardless of operating system. I don’t know how to do it. We can’t sell the Xavier machine :D. We need to set up the project on different computers as a backup. We will present to more than one place

Sevgiler

I think the first requirement would be to have a CUDA capable GPU, although I’m sure you understand this requirement. The only thing I can think of next is having a way to install all the dependencies your program has then finding the right compiler. This is just my thought.

it could be. I did not come across a previous project being run elsewhere. or There are few examples

Hi,
With some checking, we think it should be:

If your sample is a pure CUDA sample, if you re-compile it on x86 PC with NVIDIA GPU, it should be run fine on the x86 PC. We would suggest use same CUDA version.

If your sample depends on prebuilt libs in

/usr/lib/aarch64-linux-gnu/tegra

It is not able to be run on x86 PC. Can only be run on Jetson platforms.

Why could you not sell the Jetson computer?

Just clone the sdcard, and sell a ready-to-use kit including the Jetson, with whatever enclosure you can find.

You might even be able to make a nice margin/mark-up on those systems :-)

If you can’t use the Jetson, then you need some small PC of another kind with an NVIDIA graphics card, which ends up being somewhat hard to build at a low cost.

you are absolutely right. I’m confused about this. As I thought

Understood. thank you for the information