980fe014b6215730ac4fdfa451b067e6fb44e622,tensorforce/agents/dpg.py,DeterministicPolicyGradient,__init__,#DeterministicPolicyGradient#Any#Any#Any#Any#Any#Any#Any#Any#Any#Any#Any#Any#Any#Any#Any#Any#Any#Any#Any#Any#Any#Any#Any#Any#Any#Any#,129

Before Change

            entropy_regularization=entropy_regularization, **kwargs
        )

        action_spec = next(iter(self.actions_spec.values()))
        if len(self.actions_spec) > 1 or action_spec.type != "float" or \
                (action_spec.shape != () and action_spec.shape != (1,)):
            raise TensorforceError.value(
                name="DeterministicPolicyGradient", argument="actions", value=actions,
                hint="contains more than a single float action"

After Change

        // Config, saver, summarizer, recorder
        config=None, saver=None, summarizer=None, recorder=None,
        // Deprecated
        estimate_terminal=None, critic_network=None, **kwargs
    ):
        raise TensorforceError(message="Temporarily broken.")
        if estimate_terminal is not None:
            raise TensorforceError.deprecated(
                name="DPG", argument="estimate_terminal", replacement="predict_terminal_values"
            )
        if critic_network is not None:
            raise TensorforceError.deprecated(
                name="DPG", argument="critic_network", replacement="critic"
            )

        self.spec = OrderedDict(
            agent="dpg",
            states=states, actions=actions, memory=memory, batch_size=batch_size,
            max_episode_timesteps=max_episode_timesteps,
            network=network, use_beta_distribution=use_beta_distribution,
            update_frequency=update_frequency, start_updating=start_updating,
            learning_rate=learning_rate,
            horizon=horizon, discount=discount, predict_terminal_values=predict_terminal_values,
            critic=critic, critic_optimizer=critic_optimizer,
            preprocessing=preprocessing,
            exploration=exploration, variable_noise=variable_noise,
            l2_regularization=l2_regularization, entropy_regularization=entropy_regularization,
            parallel_interactions=parallel_interactions,
            config=config, saver=saver, summarizer=summarizer, recorder=recorder
        )

        policy = dict(

Instances