MLAdapter
A framework for training and utilizing machine learning agents in games. Creates an RPC interface through which an external process can query game state and control in-game actors. Once trained, agents can be run in-engine via neural networks loaded from ONNX models.
Status has not changed across the versions tracked.
