['--alg', 'sac', '--env', 'HalfCheetah-v2', '--learn', 'brac', '--traj', 'expert', '--seed', '0', '--data', '100000']
9663.50944703817
episode: 0 training return: tensor(-5996.4912, device='cuda:0')
episode: 1 training return: tensor(-4287.2061, device='cuda:0')
