c52dfd3b934f3fbdd6746e96917b552d4b193cf0,examples/mujoco_all_sac_real_nvp.py,,run_experiment,#Any#,110

Before Change

        qf=qf,
        vf=vf,

        policy_lr=policy_config["learning_rate"],
        lr=variant["lr"],
        scale_reward=variant["scale_reward"],
        discount=variant["discount"],

After Change

        from rllab.envs.mujoco.swimmer_env import SwimmerEnv
        env = normalize(SwimmerEnv())
    elif variant["env_name"] == "multi-direction-swimmer":
        env = normalize(MultiDirectionSwimmerEnv())
    else:
        env = normalize(GymEnv(variant["env_name"]))

    pool = SimpleReplayBuffer(

Instances