!!python/object/apply:collections.OrderedDict
- - - batch_size
    - 64
  - - gamma
    - 0.99
  - - learning_rate
    - 0.0001
  - - n_timesteps
    - 1000000.0
  - - noise_std
    - 0.3
  - - noise_type
    - ornstein-uhlenbeck
  - - policy
    - MlpPolicy
  - - policy_kwargs
    - dict(net_arch=dict(pi=[300, 200], qf=[400, 300]))