Is there a way to continue ml agents training once built?

Hello,
I would like to have a Unity WIN64 Build with agents that continue training when build and run.
With my current setup, the ml agent training is only possible in the Unity editor before building the scene.
Is there a way to have agents connecting to a parallel running Python ML-Agent trainer once built?

ml-agents: 0.27.0
Unity 2021.1.0b10

Thanks!

Found it: ml-agents/docs/Learning-Environment-Executable.md at release_19_docs · Unity-Technologies/ml-agents · GitHub

2 Likes