Running Karting Tutorial Model without ML-Agents package

The ML-Agents package is not compatible with all platforms, so what I'm looking to do, using the karting tutorial as a base to learn, is to train a model on that, but then 'play back' the model on a project without the ML agents package installed.

I believe from what I've read, that the barracuda package is all that's needed for this, but the karting project seems to really depend on the ML agents and training and inferencing are all combined into the same code.

since I'm going to be poking around with things I barely understand, is it possible to run this brain without the ML agents, or am I on a long path to learning that I cant actually do that?

this relates to ML-Agents 1.0.8

 

> ML-Agents package is not compatible with all platforms
Which platforms are not supported? ML-Agents should run on all platforms Barracuda supports. If you want to run a model without any ML-Agents code, you will need to remove all ML-Agents code and then manually use the Barracuda library to query the neural network model. I do not know how much wrk with will be and it is not a supported use case of ML-Agents.

 

it does not run on GameCore, you just have to add the package and you can no longer run a project. i've got a thread going in the relevant forum though, thanks.