Cheap way to run DIGITS?

I am looking at using a Jetson Nano for our FIRST Robotics Competition team over the summer; planning to use it to recognize and locate game pieces.

I looked at the tutorials on Github (https://github.com/dusty-nv/jetson-inference/blob/master/docs/imagenet-training.md), and the part on doing retraining requires a DIGITS server.

If I am not in a rush, what is a reasonable DIGITS setup for retraining?

It appears that dropping a couple of hundred for a 6GB GTX 1060 will work?

…or am I better off giving my money to AWS?

Since you didn’t mention the details of retraining jobs, like dataset size, network complexity, and other parts, I can only guess a general approach from my personal experience.

I personally prefer the approach to have one dedicate machine with GPU for your Robotics team. It makes data transfer easier. Also learning how to use AWS requires additional time. I don’t know the pricing part of AWS.

Enjoy your FIRST Robotics Competition.