ray.rllib.core.rl_module.rl_module.RLModule.forward_train#

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

DO NOT OVERRIDE! Forward-pass during training called from the learner.

This method should not be overridden. Override the self._forward_train() method instead.

Parameters:

  • batch – The input batch. This input batch should comply with input_specs_train().

  • **kwargs – Additional keyword arguments.

Returns:

The output of the forward pass. This output should comply with the output_specs_train().