sac-halfcheetahbulletenv-v0:
    env: HalfCheetahBulletEnv-v0
    run: SAC
    # Minimum reward and total ts (in given time_total_s) to pass this test.
    pass_criteria:
        episode_reward_mean: 400.0
        timesteps_total: 200000
    stop:
        time_total_s: 1800
    config:
        horizon: 1000
        soft_horizon: false
        q_model_config:
            fcnet_activation: relu
            fcnet_hiddens: [256, 256]
        policy_model_config:
            fcnet_activation: relu
            fcnet_hiddens: [256, 256]
        tau: 0.005
        target_entropy: auto
        no_done_at_end: false
        n_step: 3
        rollout_fragment_length: 1
        train_batch_size: 256
        target_network_update_freq: 1
        min_sample_timesteps_per_iteration: 1000
        replay_buffer_config:
          learning_starts: 10000
          type: MultiAgentPrioritizedReplayBuffer
        optimization:
            actor_learning_rate: 0.0003
            critic_learning_rate: 0.0003
            entropy_learning_rate: 0.0003
        num_workers: 0
        num_gpus: 1
        metrics_smoothing_episodes: 5