ray.rllib.core.rl_module.rl_module.RLModule._forward_train#

RLModule._forward_train(batch: Dict[str, Any], **kwargs) Dict[str, Any][source]#

Forward-pass used before the loss computation (training).

Override this method only, if you need specific behavior and outputs for your loss computations. If you have only one generic behavior for all phases of training and evaluation, override self._forward() instead.

By default, this calls the generic self._forward() method.