def extra_action_out(self, input_dict, state_batches, model_out): if extra_action_out_fn: return extra_action_out_fn(self, input_dict, state_batches, model_out) else: return TorchPolicy.extra_action_out(self, input_dict, state_batches, model_out)
def extra_action_out(self, input_dict, state_batches, model, action_dist): with torch.no_grad(): if extra_action_out_fn: stats_dict = extra_action_out_fn( self, input_dict, state_batches, model, action_dist) else: stats_dict = TorchPolicy.extra_action_out( self, input_dict, state_batches, model, action_dist) return convert_to_non_torch_type(stats_dict)
def extra_action_out(self, model_out): if extra_action_out_fn: return extra_action_out_fn(self, model_out) else: return TorchPolicy.extra_action_out(self, model_out)