hopper_bc:
    env: Hopper-v2
    run: CQL
    config:
        # SAC Configs
        input: d4rl.hopper-random-v1
        framework: torch
        horizon: 1000
        soft_horizon: false
        Q_model:
          fcnet_activation: relu
          fcnet_hiddens: [256, 256, 256]
        policy_model:
          fcnet_activation: relu
          fcnet_hiddens: [256, 256, 256]
        tau: 0.005
        target_entropy: auto
        no_done_at_end: false
        n_step: 1
        rollout_fragment_length: 1
        prioritized_replay: false
        train_batch_size: 256
        target_network_update_freq: 1
        timesteps_per_iteration: 1000
        learning_starts: 10000
        optimization:
          actor_learning_rate: 0.0003
          critic_learning_rate: 0.0003
          entropy_learning_rate: 0.0003
        num_workers: 0
        num_gpus: 1
        clip_actions: false
        normalize_actions: true
        evaluation_interval: 1
        metrics_smoothing_episodes: 5
        # CQL Configs
        min_q_weight: 5.0
        bc_iters: 10000000
        temperature: 1.0
        num_actions: 10
        lagrangian: False
        evaluation_config:
            input: sampler